From 92ac798dab92ec46f5f24e3307be337ac3964b49 Mon Sep 17 00:00:00 2001
From: "Documenter.jl" <documenter@juliadocs.github.io>
Date: Mon, 6 Nov 2023 21:40:05 +0000
Subject: [PATCH] build based on ff869cf

---
 previews/PR3562/.documenter-siteinfo.json     |    2 +-
 previews/PR3562/JuMP.pdf                      |  Bin 4689518 -> 4689215 bytes
 .../PR3562/api/JuMP.Containers/index.html     |   22 +-
 previews/PR3562/api/JuMP/index.html           |  340 +-
 .../algebraic_modeling_languages/index.html   |    2 +-
 previews/PR3562/changelog/index.html          |    2 +-
 .../PR3562/developers/checklists/index.html   |    2 +-
 .../PR3562/developers/contributing/index.html |    2 +-
 .../custom_solver_binaries/index.html         |    2 +-
 .../PR3562/developers/extensions/index.html   |    2 +-
 previews/PR3562/developers/roadmap/index.html |    2 +-
 previews/PR3562/developers/style/index.html   |    2 +-
 .../extensions/DimensionalData/index.html     |    2 +-
 .../PR3562/extensions/introduction/index.html |    2 +-
 previews/PR3562/index.html                    |    2 +-
 previews/PR3562/installation/index.html       |    2 +-
 previews/PR3562/manual/callbacks/index.html   |    2 +-
 previews/PR3562/manual/complex/index.html     |    2 +-
 previews/PR3562/manual/constraints/index.html |    2 +-
 previews/PR3562/manual/containers/index.html  |    2 +-
 previews/PR3562/manual/expressions/index.html |    2 +-
 previews/PR3562/manual/models/index.html      |    2 +-
 previews/PR3562/manual/nlp/index.html         |    2 +-
 previews/PR3562/manual/nonlinear/index.html   |    2 +-
 previews/PR3562/manual/objective/index.html   |    2 +-
 previews/PR3562/manual/solutions/index.html   |    2 +-
 previews/PR3562/manual/variables/index.html   |    2 +-
 .../PR3562/moi/background/duality/index.html  |    2 +-
 .../infeasibility_certificates/index.html     |    2 +-
 .../moi/background/motivation/index.html      |    2 +-
 .../background/naming_conventions/index.html  |    2 +-
 previews/PR3562/moi/changelog/index.html      |    2 +-
 .../moi/developer/checklists/index.html       |    2 +-
 previews/PR3562/moi/index.html                |    2 +-
 .../PR3562/moi/manual/constraints/index.html  |    2 +-
 previews/PR3562/moi/manual/models/index.html  |    2 +-
 .../PR3562/moi/manual/modification/index.html |    2 +-
 .../PR3562/moi/manual/solutions/index.html    |    2 +-
 .../moi/manual/standard_form/index.html       |    2 +-
 .../PR3562/moi/manual/variables/index.html    |    2 +-
 .../PR3562/moi/reference/callbacks/index.html |    2 +-
 .../moi/reference/constraints/index.html      |   12 +-
 .../PR3562/moi/reference/errors/index.html    |    2 +-
 .../PR3562/moi/reference/models/index.html    |   26 +-
 .../moi/reference/modification/index.html     |    2 +-
 .../PR3562/moi/reference/nonlinear/index.html |    2 +-
 .../moi/reference/standard_form/index.html    |    2 +-
 .../PR3562/moi/reference/variables/index.html |    2 +-
 previews/PR3562/moi/release_notes/index.html  |    2 +-
 .../submodules/Benchmarks/overview/index.html |    2 +-
 .../Benchmarks/reference/index.html           |    2 +-
 .../Bridges/list_of_bridges/index.html        |    2 +-
 .../submodules/Bridges/overview/index.html    |    2 +-
 .../submodules/Bridges/reference/index.html   |    2 +-
 .../FileFormats/overview/index.html           |    2 +-
 .../FileFormats/reference/index.html          |    2 +-
 .../submodules/Nonlinear/overview/index.html  |    2 +-
 .../submodules/Nonlinear/reference/index.html |    2 +-
 .../moi/submodules/Test/overview/index.html   |    2 +-
 .../moi/submodules/Test/reference/index.html  |    2 +-
 .../submodules/Utilities/overview/index.html  |    2 +-
 .../submodules/Utilities/reference/index.html |    4 +-
 .../tutorials/bridging_constraint/index.html  |    2 +-
 .../PR3562/moi/tutorials/example/index.html   |    2 +-
 .../moi/tutorials/implementing/index.html     |    2 +-
 .../PR3562/moi/tutorials/latency/index.html   |    2 +-
 .../manipulating_expressions/index.html       |    2 +-
 .../moi/tutorials/mathprogbase/index.html     |    2 +-
 previews/PR3562/packages/Alpine/index.html    |    2 +-
 .../PR3562/packages/AmplNLWriter/index.html   |    2 +-
 previews/PR3562/packages/BARON/index.html     |    2 +-
 .../PR3562/packages/BilevelJuMP/index.html    |    2 +-
 previews/PR3562/packages/CDCS/index.html      |    2 +-
 previews/PR3562/packages/CDDLib/index.html    |    2 +-
 previews/PR3562/packages/COPT/index.html      |    2 +-
 previews/PR3562/packages/COSMO/index.html     |    2 +-
 previews/PR3562/packages/CPLEX/index.html     |    2 +-
 previews/PR3562/packages/CSDP/index.html      |    2 +-
 previews/PR3562/packages/Cbc/index.html       |    2 +-
 previews/PR3562/packages/Clarabel/index.html  |    2 +-
 previews/PR3562/packages/Clp/index.html       |    2 +-
 previews/PR3562/packages/DAQP/index.html      |    2 +-
 previews/PR3562/packages/DiffOpt/index.html   |    2 +-
 .../PR3562/packages/Dualization/index.html    |    2 +-
 previews/PR3562/packages/ECOS/index.html      |    2 +-
 previews/PR3562/packages/GAMS/index.html      |    2 +-
 previews/PR3562/packages/GLPK/index.html      |    2 +-
 previews/PR3562/packages/Gurobi/index.html    |    2 +-
 previews/PR3562/packages/HiGHS/index.html     |    2 +-
 previews/PR3562/packages/Hypatia/index.html   |    2 +-
 .../PR3562/packages/InfiniteOpt/index.html    |    2 +-
 previews/PR3562/packages/Ipopt/index.html     |    2 +-
 previews/PR3562/packages/Juniper/index.html   |    2 +-
 previews/PR3562/packages/KNITRO/index.html    |    2 +-
 previews/PR3562/packages/Loraine/index.html   |    2 +-
 previews/PR3562/packages/MadNLP/index.html    |    2 +-
 previews/PR3562/packages/MiniZinc/index.html  |    2 +-
 .../PR3562/packages/MosekTools/index.html     |    2 +-
 .../MultiObjectiveAlgorithms/index.html       |    2 +-
 .../PR3562/packages/NEOSServer/index.html     |    2 +-
 previews/PR3562/packages/NLopt/index.html     |    2 +-
 previews/PR3562/packages/OSQP/index.html      |    2 +-
 .../PR3562/packages/PATHSolver/index.html     |    2 +-
 previews/PR3562/packages/Pajarito/index.html  |    2 +-
 .../ParametricOptInterface/index.html         |    2 +-
 previews/PR3562/packages/Pavito/index.html    |    2 +-
 previews/PR3562/packages/Plasmo/index.html    |    2 +-
 previews/PR3562/packages/PolyJuMP/index.html  |    2 +-
 previews/PR3562/packages/ProxSDP/index.html   |    2 +-
 previews/PR3562/packages/SCIP/index.html      |    2 +-
 previews/PR3562/packages/SCS/index.html       |    2 +-
 previews/PR3562/packages/SDDP/index.html      |    2 +-
 previews/PR3562/packages/SDPA/index.html      |    2 +-
 previews/PR3562/packages/SDPNAL/index.html    |    2 +-
 previews/PR3562/packages/SDPT3/index.html     |    2 +-
 previews/PR3562/packages/SeDuMi/index.html    |    2 +-
 .../PR3562/packages/SumOfSquares/index.html   |    2 +-
 previews/PR3562/packages/Tulip/index.html     |    2 +-
 previews/PR3562/packages/Xpress/index.html    |    2 +-
 previews/PR3562/packages/solvers/index.html   |    2 +-
 previews/PR3562/release_notes/index.html      |    2 +-
 previews/PR3562/search_index.js               |    2 +-
 previews/PR3562/should_i_use/index.html       |    2 +-
 .../benders_decomposition/index.html          |    2 +-
 .../{ccf7f2c9.svg => 74fb9305.svg}            | 1002 ++--
 .../{50ad43f9.svg => 80b5cf75.svg}            |  670 +--
 .../index.html                                |   12 +-
 .../algorithms/parallelism/index.html         |    2 +-
 .../{21585044.svg => 6096df34.svg}            |  214 +-
 .../{88a7f5c2.svg => 6e144cc3.svg}            |  214 +-
 .../tsp_lazy_constraints/index.html           |    6 +-
 .../optimal_power_flow/index.html             |   22 +-
 .../{2588df61.svg => 0215129c.svg}            |   56 +-
 .../{472a82a6.svg => d9e67996.svg}            |  120 +-
 .../{e09dc019.svg => e02e8c47.svg}            |  124 +-
 .../applications/power_systems/index.html     |   12 +-
 .../two_stage_stochastic/3b3c353b.svg         |   80 +
 .../two_stage_stochastic/5becafc9.svg         |  106 +
 .../two_stage_stochastic/60f511d7.svg         |  100 -
 .../two_stage_stochastic/7add9a3e.svg         |   77 -
 .../two_stage_stochastic/7f7c697a.svg         |   73 -
 .../two_stage_stochastic/8f63bbf6.svg         |   96 -
 .../two_stage_stochastic/a4f8868e.svg         |   94 +
 .../two_stage_stochastic/bea383e2.svg         |   76 +
 .../two_stage_stochastic/index.html           |   60 +-
 .../tutorials/applications/web_app/index.html |    2 +-
 .../conic/arbitrary_precision/index.html      |    4 +-
 .../tutorials/conic/dualization/index.html    |   66 +-
 .../{053884ea.svg => 618bca67.svg}            | 4872 ++++++++---------
 .../{b2c746af.svg => f9ac62e0.svg}            | 4870 ++++++++--------
 .../tutorials/conic/ellipse_approx/index.html |   16 +-
 .../conic/experiment_design/index.html        |   54 +-
 .../tutorials/conic/introduction/index.html   |    2 +-
 .../conic/logistic_regression/index.html      |   24 +-
 .../{1521bd4d.svg => 8380bd2b.svg}            |   76 +-
 .../{1b95f9f7.svg => ca1157bb.svg}            |   70 +-
 .../tutorials/conic/min_ellipse/index.html    |   10 +-
 .../conic/quantum_discrimination/index.html   |   12 +-
 .../{c9b2b8b7.svg => 4350a7ce.svg}            |   70 +-
 .../conic/simple_examples/index.html          |   10 +-
 .../tutorials/conic/start_values/index.html   |   16 +-
 .../conic/tips_and_tricks/index.html          |   30 +-
 .../getting_started/debugging/index.html      |    2 +-
 .../index.html                                |    2 +-
 .../getting_started_with_JuMP/index.html      |    2 +-
 .../{1e380636.svg => 6cdf6131.svg}            |   94 +-
 .../{054004dc.svg => 76abfa9b.svg}            |  118 +-
 .../{c55aa536.svg => c9b9ec62.svg}            |   88 +-
 .../index.html                                |   10 +-
 .../getting_started_with_julia/index.html     |    2 +-
 .../index.html                                |    2 +-
 .../getting_started/introduction/index.html   |    2 +-
 .../performance_tips/index.html               |    2 +-
 .../tutorials/linear/callbacks/index.html     |    2 +-
 .../tutorials/linear/cannery/index.html       |    4 +-
 .../linear/constraint_programming/index.html  |    2 +-
 .../PR3562/tutorials/linear/diet/index.html   |    8 +-
 .../{528bd41e.svg => 492bb32b.svg}            |  108 +-
 .../{0d89ccc9.svg => 96afa50f.svg}            |  108 +-
 .../{d825a2fa.svg => ca63c86d.svg}            |   94 +-
 .../{7f2d6f44.svg => de2fb0ef.svg}            |   84 +-
 .../linear/facility_location/index.html       |    8 +-
 .../{9faf22a9.svg => 22aa7ce9.svg}            |  296 +-
 .../{79a4ee80.svg => 57fc90bf.svg}            |  292 +-
 .../{cc63b3c1.svg => f8d71086.svg}            |  110 +-
 .../linear/factory_schedule/index.html        |    6 +-
 .../tutorials/linear/finance/index.html       |    2 +-
 .../linear/geographic_clustering/index.html   |    2 +-
 .../tutorials/linear/introduction/index.html  |    2 +-
 .../tutorials/linear/knapsack/index.html      |    4 +-
 .../linear/lp_sensitivity/index.html          |    6 +-
 .../tutorials/linear/mip_duality/index.html   |    2 +-
 .../PR3562/tutorials/linear/multi/index.html  |    6 +-
 .../linear/multi_commodity_network/index.html |    4 +-
 .../multi_objective_examples/index.html       |    8 +-
 .../{1f70cf5d.svg => ba5f521f.svg}            |   78 +-
 .../{717b7729.svg => c5b0bcc4.svg}            |   80 +-
 .../multi_objective_knapsack/index.html       |    8 +-
 .../tutorials/linear/n-queens/index.html      |    2 +-
 .../tutorials/linear/network_flows/index.html |    2 +-
 .../PR3562/tutorials/linear/sudoku/index.html |    2 +-
 .../tutorials/linear/tips_and_tricks.jl       |   31 +-
 .../linear/tips_and_tricks/index.html         |    4 +-
 .../PR3562/tutorials/linear/transp/index.html |    2 +-
 .../nonlinear/complementarity/index.html      |    4 +-
 .../nonlinear/introduction/index.html         |    2 +-
 .../nonlinear/nested_problems/index.html      |   24 +-
 .../portfolio/{a315d69e.svg => e1860e4a.svg}  | 1126 ++--
 .../tutorials/nonlinear/portfolio/index.html  |   10 +-
 .../nonlinear/querying_hessians/index.html    |    2 +-
 .../{5e56f1a3.svg => cf89a871.svg}            |  206 +-
 .../nonlinear/rocket_control/index.html       |    4 +-
 .../nonlinear/simple_examples/index.html      |    4 +-
 .../{97a644df.svg => b724747b.svg}            |  148 +-
 .../{07f114d3.svg => b9835bf8.svg}            |   66 +-
 .../{d9b9da98.svg => e697ac28.svg}            |  294 +-
 .../index.html                                |    6 +-
 .../nonlinear/tips_and_tricks/index.html      |    2 +-
 .../user_defined_hessians/index.html          |    6 +-
 219 files changed, 8826 insertions(+), 8829 deletions(-)
 rename previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/{ccf7f2c9.svg => 74fb9305.svg} (69%)
 rename previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/{50ad43f9.svg => 80b5cf75.svg} (72%)
 rename previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/{21585044.svg => 6096df34.svg} (75%)
 rename previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/{88a7f5c2.svg => 6e144cc3.svg} (75%)
 rename previews/PR3562/tutorials/applications/power_systems/{2588df61.svg => 0215129c.svg} (85%)
 rename previews/PR3562/tutorials/applications/power_systems/{472a82a6.svg => d9e67996.svg} (86%)
 rename previews/PR3562/tutorials/applications/power_systems/{e09dc019.svg => e02e8c47.svg} (86%)
 create mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/3b3c353b.svg
 create mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/5becafc9.svg
 delete mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/60f511d7.svg
 delete mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/7add9a3e.svg
 delete mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/7f7c697a.svg
 delete mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/8f63bbf6.svg
 create mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/a4f8868e.svg
 create mode 100644 previews/PR3562/tutorials/applications/two_stage_stochastic/bea383e2.svg
 rename previews/PR3562/tutorials/conic/ellipse_approx/{053884ea.svg => 618bca67.svg} (58%)
 rename previews/PR3562/tutorials/conic/ellipse_approx/{b2c746af.svg => f9ac62e0.svg} (57%)
 rename previews/PR3562/tutorials/conic/min_ellipse/{1521bd4d.svg => 8380bd2b.svg} (88%)
 rename previews/PR3562/tutorials/conic/min_ellipse/{1b95f9f7.svg => ca1157bb.svg} (88%)
 rename previews/PR3562/tutorials/conic/simple_examples/{c9b2b8b7.svg => 4350a7ce.svg} (79%)
 rename previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/{1e380636.svg => 6cdf6131.svg} (84%)
 rename previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/{054004dc.svg => 76abfa9b.svg} (82%)
 rename previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/{c55aa536.svg => c9b9ec62.svg} (85%)
 rename previews/PR3562/tutorials/linear/facility_location/{528bd41e.svg => 492bb32b.svg} (79%)
 rename previews/PR3562/tutorials/linear/facility_location/{0d89ccc9.svg => 96afa50f.svg} (79%)
 rename previews/PR3562/tutorials/linear/facility_location/{d825a2fa.svg => ca63c86d.svg} (81%)
 rename previews/PR3562/tutorials/linear/facility_location/{7f2d6f44.svg => de2fb0ef.svg} (80%)
 rename previews/PR3562/tutorials/linear/factory_schedule/{9faf22a9.svg => 22aa7ce9.svg} (79%)
 rename previews/PR3562/tutorials/linear/factory_schedule/{79a4ee80.svg => 57fc90bf.svg} (78%)
 rename previews/PR3562/tutorials/linear/factory_schedule/{cc63b3c1.svg => f8d71086.svg} (86%)
 rename previews/PR3562/tutorials/linear/multi_objective_knapsack/{1f70cf5d.svg => ba5f521f.svg} (87%)
 rename previews/PR3562/tutorials/linear/multi_objective_knapsack/{717b7729.svg => c5b0bcc4.svg} (83%)
 rename previews/PR3562/tutorials/nonlinear/portfolio/{a315d69e.svg => e1860e4a.svg} (79%)
 rename previews/PR3562/tutorials/nonlinear/rocket_control/{5e56f1a3.svg => cf89a871.svg} (89%)
 rename previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/{97a644df.svg => b724747b.svg} (88%)
 rename previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/{07f114d3.svg => b9835bf8.svg} (87%)
 rename previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/{d9b9da98.svg => e697ac28.svg} (88%)

diff --git a/previews/PR3562/.documenter-siteinfo.json b/previews/PR3562/.documenter-siteinfo.json
index 993afc8f4f7..dcab2cc0eda 100644
--- a/previews/PR3562/.documenter-siteinfo.json
+++ b/previews/PR3562/.documenter-siteinfo.json
@@ -1 +1 @@
-{"documenter":{"julia_version":"1.9.3","generation_timestamp":"2023-11-06T21:11:47","documenter_version":"1.1.1"}}
\ No newline at end of file
+{"documenter":{"julia_version":"1.9.3","generation_timestamp":"2023-11-06T21:16:31","documenter_version":"1.1.1"}}
\ No newline at end of file
diff --git a/previews/PR3562/JuMP.pdf b/previews/PR3562/JuMP.pdf
index cd957d92fefe623859533b38b97fd58c7221011d..9a406aebcbbe01ac0fc82e40865e0ed4d9e924e4 100644
GIT binary patch
delta 2101209
zcmV)qK$^eqoSp)|nw|o%UMH6ot1c(Ag(r*wf2H>UOK9R&<N~Z|1aW-j8YYyuIVu*^
zi)ES>(1IDHu+(Ksn9jhP-Z4a#^EmM)PL~xgAXQAfBefkVGcAbQk>ZkYrkY4WLTd20
zn$P+JgJ1MuM#E{@m^3Nj<(@gumJ~{_Ryr-29<wGq*5a`iXf?$S&tn!pvyXMffN<H?
ze<+MV3FbjsF~ds(T3=n(F6)p~L@1~x*HL=3N)1j&HbX@@5mYvzk_oEX(^HvFF|R-(
zq2|=6X+!CaHdHa;4#qwxZMUJ&P$G~uQBnyd+H(G_W|^FYU%kmBGlI;*s(;T+FfsEN
zt^D`ZRj-_9C+IP&-r#~oo2gQo2-6lQf1RlE7PwofL=clGH#;$>kF#Na^4u*3ySPLO
z3Pho)K1t{9hIa@}15c-YR7y$$&{iq4;K>oAJRt8!?qJbvX`zZBs_NohJbpyTb`UKD
z0=g|)LuIN@QY7T`1bFKFF&kEG;usQpel=1j<cY7BEso<ZF$M%q>?<wIPASA7e^YoN
zn(CSP9)bm^E`~VspoCA(+!$ggX4v6)jIA{~YiKi0f)Z5N<Y`>+6hcuh(s1?$6kN~K
z1l1}A>No!RRSTz3xC@7G#*Aswh-2X**6u8NofIpiY@%5{dPQ5NaUi*c&e|WdcgyGx
zOqyn*Sr752JVLMi)$n4_Vxfv{f74$8ky4ePxT{veX$Om*O8X^alP74eg5GgMoiP=&
zGRjj`q!E?3%>r3hX^N}l$>+SiKf!3e=KAMD8USxwgWI$zPVBkW7xMFJ(fDb2emSFF
ztG@wbo%Q6vO&%dAXPdJS#9Yw{NIG%Ba%!)O$Q7YVm!o5ho_+pfUMMs2e}6Ne?IR{c
zOjzk-BAsW1E;tN-2eJz8HdlOZh|#7Ru&=kz6$ZW^wF)goi87`_3sa58YOwCkC>sYL
z21GS{2jlAkDj4Imo1sEYG(ZKFOi+oV8-9_1GG4QSUh}xRa<9`Wq39leZKkc7z@Z`9
zsA_5*kOyBchsy`h#_fQCe<WJhs-+TOcm&8R#NLlCXg4R?16W)pGoe~K&#T3Y<rMmj
ze}>{k^vCx44Y1Cj5w(Wtgjh?JV$qL&%Sxo%cm*%m1RYF%akzr;iTW6=(fKQr&l9{N
zH;ZK+&5yQ-Py0w>^1&Ce#UiN|)p=}tFxuL|R90Wrk6o>SzQB5Wf8+!>Cq~sOWxiK|
zbvc2qDbJ!&Y9GKDBb3(~l{J0%azh`jkmAipG+ZQui4)KK;<PNDg7Kc>ajA4@wX&?7
zQb**`TBUZ$O$lQNmAXPHYX)H%6e>qXf8IpjSWaye;kur5hhvK#ZTI^ogMcwnL6jOk
zc-}O2F{Z2U)FAKUf0fhdSbRSpPkSG4`{E`P5{9)REt@<<3fsGqpXEXZ1#Y>v<R<YP
z60Um}?=G|!j1w+o^&vVsx2Tc6tE}f*vOq{dI8CxJ#B4zx2AIw39kK)EaCmpG%C+=>
zKt?zu$c9I!2g(6E)z6n6*rzQ@2t&9L6<U}_a_2Y#wxS?ye^+*yosCjE#N1r$$YV4!
zI3dM~uk_iK)9yg+1Jw3>ld$S2viAL<P7Qn_CSuQJAm2V67%6Ab`FIvbWS{0c!H|e-
zXMv&453IK!$jnY08{v{r-lQ*jvg<{dfoOeX@{S$_zZXXNG5?RUH0<NKQbB;%vXr9z
zyr?MpZ2pmRe}arv19feAW^eZ7z1i)C6gP~3sud~a0(kI2b3YU_v#05khI9?&rIU*t
zFiPLnC5>omIra|ETAZz%d`ITq<!yAf^JbWov0_rIRf_xb^D$OVxN5~Qv!jED>+SXo
zq#-S0W$bOWwt`ztt7g?H`HI-dxJ!Y>l+Hm}%T-DTBTRdnO=gQTS9Ib3>AXiJczl5<
z$KP<3reX&D?tgEZ<Zd<JldUeKfACRuPdZJTD5$KJ{ZuENFj<0-ZFiP>vEp8PWmN!g
zbP=wQ@4J`FDRuH6^LHV4ablT4`x>hN!m3y#?wd#R+M(El$1(@x)@A_bVw5xJynKa=
zu}>)^(}L)%NBz{wD`mMN2-^j<cKTEyK>&Fue(CF?EQ{GvCiGH4+v>Vqe|~~a&%OGj
z_t7WCh~X7bD$jho7LyoYh0H_ExssRfz0V31V|A%{)}(;UO5FBuXJ%-K$y&k3g=rWZ
z)4)}G<Lj#?VmXNPSy#QgiO4_`nwHnGvJe+tfK^Yu>wOuE?Q-eW+<q}?C;3@q+3b?3
zWkM2~wWg1CMbg#2tmtDDf6Og*+Fdgoi#c>AzQS-Eb_pXg4y4v7xhxb3HlJzk3GXQa
zY=#l#DoKmc`+ffkj`rEyjWJGG)>uB)_M&w_B))}BzO5CZ-|7m-*{v-jhp~-06biL-
zxz~1rTTk<sAED!O?@~9qxZYjnp(bdQ6fN^mBjFHNovEgKKI|miJ}nnC3qpG<)h~Q)
zgOCwnEv4mUXMg-51^7x3o|;j8o`xxBS&PPhzj4KoU0yCs`?UNowfy7z$IE{Q!_6O;
zQK&W)vm!CR2>~;g;Wh*&lLtEze+&fJN;*J}LEnO0lT$8v|3fH=6e%v=#I~Ib51y87
zSt^b`|5dCi(e&`wfgXN&JAJ-=`w(zYL?cuD@cHL~**B>Dmr?to#DijtNWl)De;<DT
zPAQj^3T_|FK9qf~!?uvaJ7`I{-48}e(F}il{`yAU$tdBdH9t@S_Mruef5Siao&43`
zetBE15S%D$m{dUQIlL7N>_gZGo3@8{u0cEF5C_gC1FT_ScSAL85ATAh=~bAvrhSpx
zhaCtgx9`WsSZ@Fqihz`v0c?1e0t#Dr^SjuOck?HNo~|)d4ujKj?e+jBIiYs15@5Ij
zCZvPrW1@Qj#{%b(vhSd9f5?sQxjg{N^Jpcf?kKzpS~CRu(LzcXFKd_FXcwHEX4$vk
z=pdT6SHzUtORuj?fayig05c@*{x~jJ&hJ!E)P`t^nW-I}J3o;?1p6Tpn#>-$0Fes%
zBnBp@%Z>paRNna&;JMKWCPs8TUP-AMK*lBEtgKWzK&A$U^t+QCe;k|)t(0*lPHf{7
zwG$%Yx6f0SSkb9?CCYF=Wef<HWhpb8Pz$t|FTIW3eSk${F`I(2_0{n*W{DXci&w%-
z?#GOBBB(54b_|#u2Y}dPUO!!oLoB{Lv&!dY2lg06_|OJgMrFozSibmpLemwm;eAGJ
zKVo(YNCdJXVyA!@e=dmmj0C7#@B$NZ8nZK{%cn1+CKVA1G9x=0kBIoyez=q-!j$!E
zEW`ETrjMr;uyJ91nFqMNj!x|$a+&iqv5XM_F(5K?`elEDyNp$Aj>&gJCM5wVE7+3#
zc;}o=30>}ync}ja2jQWc^AoC`Szt)0fb2vakQY8)qGe`3e`W{-w5(zV0W%-k$7y`I
z$zp6=Wf?kdh|wlHZNKENnI0Q&2JG7h>T^{q1WxRpF>h1Yr)K>WX24m#AjJK%w4o2E
zs&a;05vp{0{AnO~SF*TTe)HcPK?y3{j$~iq%bXke`4}KErt|;Y7yz5pCy;59q+=kj
ze8q95FHOi0e^qYi3tDaN`j+`wa$;G><8hN@Y&S`!l1N!ru&Dgy%3f>DbPS#z7gDa$
zuggp&=5z=i57qw3F&H3<6`6{vVP;VBte;<Xin+b$d4^o-I5~U2<gb!B#wL?u(DpBu
z6-#e>LDaU^8Wrj@qU=V<8Oxw>6OF#7WF}NzcZ{`Be;^b{VoOwPRh4LgHl8&)8}g5Z
z%yB<5iV|f^5gBtunC>s^9IC4*ou^#_FW)P>{F{mK1d?&h0OPcW%v|Sk<gy@oDDiTJ
zM}V<G8q59*e#FTvHt9tJ)I<XmVG5Nij@$gMzp%^Y-V=Sh*<NZ!cT3Nm?a3yyJpf6x
zE-O{2e_8P{FB&ABWjo<Y%;+%uFW?gP>WZ>*jd^Pa4GP9++_6@IuHy7I$8yJ!(L6GW
zgJx&eYCgHTy=2HgqT7>Ay3c@fVpLg~wzVi9dao`|r5AH<(AMXb$es=v%TC+1=9N<j
z(#ggCHZ)+2P!3hxlA4r`6hUp;&RS&FQ)4;;f6sS^`%|C77(%74B8EZH%%If8RMm`I
z>f9ij=xe8kwG=2PkXzN+ct&AeHpznutq5bHN|@aE%s5b5vJ6WX+04$E3fsLAzUw<n
zP3u@ZA`Bb%Bc+fqtc#Y~hGD)=LxXQzfoWB<NlogP{_z1<1F)nZoYu8U2e878O4CWd
ze|>0H?2*uhD4rddi>raFbzg8{2xyZYd4}Bez)g8vKKJCFSHWb^9ks7kLDeLaA&?Oc
zbrm!>gl=)RVH3W^6{IALi+U)yzq@Mve5r)6-fvvaRxnPuklC<0KG(!*jgYdU!cATO
zjsYp*nr$G!5j)sRalXrJq$lj^^2&}@f8E`2bEz3paO;bb#WQg2hU8Z$5~_<Gk?BeN
z_r^4Di4+WpNS12Ayd@Hr>gPPqDco@bR}K4Y(k3;e)BW6}jcU@XE~p@oEZ2Z}Kcw$1
zTKm%-PQC}@r__(bzVF!@P$uSeNFEVT?(Y!ch7nL@)!MECGTltNxoNAM;fumUe=xVH
zS{*V4lkR55b|}84`A9bjEV(i^-YYAX+S$R~NVQTEJ5XP(TDnP1cDjBj%c`aCXx?t6
zPl;Lm@_(H^?XR?QL?V-w7_i(N8N0rEP<~_AwhPx&GdeWS4cB#(8Y>4uWU>YW=F5Ei
zpPIGPA^$kd+PX=9KNpNhQ`Vzsf7y=PiV(>`ry|6S&aPUF0KMqf_6%t;f}JTwXifyz
z*}%JEwM^jYCg}&xEKg9@v(Q30wIrdA7x4)93Dg$ypf+Ygm7SnFy8oi%Y><$!JSbgO
z!KB^t?2RQMty-Hedgm9lX@)uHawK|w_L!L3fqBI2aeu|18MN=QDq?UKe_|KfER&h$
z+~efetr^3bHMQKnTM0Kcy(6+8D$^v}#k3$gSrG$^O_kU7?d~ZBEiIL)O`Vo!l<7^A
zlowNs7+%Foe$8D7$yqxhuUx7(&2B<qXo!(@F$>L>*Olp;bcrubm#JwTh{prAKRw1k
z6IzeJLeJ&eFCpE-<*-nue;KnI%R>GWm32(-_)Qiu&0v{G$JT+m4AlPo1QU`_tSgq;
z#Pn->;%WKWApi3y>HWnsJ0@CTQn#uY8AZ6x&F-+=n}an?vJZ>{smszazilPBw%Cn9
zng&}_U}^F`a)fv~@Ax*Rbx<C!Qu~+CV2l%%ETIttcFF*Yb!PHEf6V$PX1ue)VY4(#
zbVqvpvF4(cjH0u%N3<H`vnVAjSvw=9Y7A7lW}6uTwO#2aYVChP_JjZ;rOL+ITOwML
zq&4B9EKis(bsz#JYr=CaEejs<=oT&C!c}hP3huPtLpcAw4}On-ILz<p(<>FdN=DV?
z>R!l>O){lQ60S@YJ2H4x$nY#=$*4MH&xK3^5xlNsQ8qqlc9}tgxO2W$^yGhIXKPsp
z<#B=h(FO8bcj5p3{Q34TBu!R6m!a-H6_edV6qi5pJSUSFL=AuDa4Bv;!+@k250GQf
zw;<Q#luQ2q5Q-utsz}R=XuE^q!QHlHOGW#!SSw8re;?@K=Ue~z_U$C#pox)8^TX#~
z2j&h?_m@%kq>=~C7*T>9KL0uV@j)q9lnU-{%-ytmY?k{%9pAwiZQbjnJ^uOm+ZzoR
zqJ&Fh_<<5|HzR+fIQ;95<j?;8^KF(uaH5@1N&|7{j}kC&H|cIH{C8L7obiuQ07}6L
z<esH4xSJjyGtoPNC7Yg;{<-73V8*=z+&g4AD%EsUy0-&_SF`Y(S`8<P$en2JlgmV8
z$9E;5a8KTRRQ~ek6NZiV+_%8G@#~+yi9J9mPN;j;IiP>^3>1%eIH06(NSqSD%!0>=
zi^nDaCWl&p8A}Y6naB;WJKup|lKWwjlrVl?JDCM08BpL0NFE$k3l`~Mu{sl}!Sdkq
z2}5V7j_)2Ybu(DsPSR{G#7Y?A4<(?vrH^TU@o^oN{iLNlGA1CT0y;BRMRX+Bgldyr
zB+e6a$8dkn*cEZ!a78#<&l?ry4Y)aDV4M*g-0BTs^<GwOb7blx9q^I!kDCY0B`1(e
z((r!T`H(g?`?WCx!WA!4C1<L{NL0|ikmy*yl#h0!s!f{GL}<vwYiK-Uk?;3oML7{v
z6|qtaR%}xK5eC1=4^eqVD{(|iKL9NxVngB4Xt{syjfvf8NlS^4tcaFg*1-XJu)~C!
z*I*H|?+I?*jM)U;gj&>d!+D$lhy{^})xQ20I53G_c^_<-Y)YvO5q7QD^lVd(aUK|v
z`5yF_T9!!1N6=K8_9VVQ9Ez%LPfDc-z<M6(%$}ShM0#G~VZL`(U7MiPw?B+cpA3uF
z_<w(%PuAb<lVy-eAYfDxGI<qb@@mubUf@DV5{udZc}z&MuLBlv;*ONKQrB_CuCY67
zaJ81SVf8qy^nR=aB~r^GS2}lx&BTfKVTNcxdN%qr{Bi|Bkxx6F*^`*Q%9FYo@y4Pz
z(CT<B6`*2<pb$%B$}A3x2`qRAB`{@@wa$M$fv;Ihp3#{iTx$-k7l{;3<1qj{?q3i{
z<d0N9i7KL0SwGn)=qIU?Jf1qg_8CqDH<`FiCsj4mt3-go0?}nRuv>_tfs$x-+Ylfb
zR$&vi+xhbe1|(9W<`wycN_06*pk-NM^^e30Wr%%M3rvsw70l)m#;qn!iDlw9boPJP
z=XO6%ni6eo5vSn>oSwJ_Vv!p{wb#H3SV=>SvsjC<P#;LnD&|V&v0bC%jcR#RVmu{1
za{ZB-_=|}KddBk3Wiz;dco+?}Ixb*WJ&ypOh%swDW|rENw5&<5%b6yd{$d^hUmy-;
zsGdgv0mO>++%bi^lu}l$Of#i~n(TkP{{y|$rqjkG8un{x&WY8wNZOK2;2~JTKGG#R
z5ZH%_q!uFe6+{Y6K&ecC2E<nZY^YM_42%)SMG>%dR>9XcL)66EfL){VLJ<KLpaJpP
zB7&hxu1Z)-s4_*Yl8P03^uh$IaVl+6r3#hYd4Y=t!)&AKP|YGPIxbl322X!Y$7QGz
zUJ*tU#zYaVB)8B9Jh~Ye?^H0+FMx3Nu?maCz{{Ozc?&hoC(Kt!&kRwr*{9yjDB$7#
z+9`$MF5mZ(Ya+BHOs=(+T}qtP(vRU2N&2R8d3^FXjQVbgP=au}o*A}S-gIF&Nz_M%
zPA1>Y6j%RYE0T?+Xgc&2q3nO5iu)lXBOFYTsHB#u3tDYh>)IIRG3DrE%Aw6PYZ_2;
z+LiOoiW4kogQULGF)p&AN<Fg>mT+5EREg6e5=UZKRM1!|5Uj%cG;{QAu^0`Sui)E8
z%=Y`X!mYppHHF*U3DHTl<*BHh?m!D&x}oru(AwY8E0`n#wTIix2@!wt2=I~QHF#o7
z#gZhQW<)%CUG#|Evp<}}B4$@+a|uKuHAEf<P=x>q70y^L>!zftDOf+V_MuB#j|>Sj
z5vzLjltr1_eNGsxU}1&u?-Xu{pk3I!0*l+=dK_H)OXj#`1aw(DbxD3|vaftlLKT_{
z^Byss>^Cq>dE0q$G#7vO9Kd2Vh@KOBGgCU>L$<$gfNM)6D{HPk??Ye-P-$8}YB8Oc
zJSF7<L-KC5s_5mZx5!W4UDyikVzC=6Rq?8WmeNG3vff%sou*QOnC+bGqp2-wd*>5p
zA5j6>MVg-Z`Vp;&ShpWBDTzW|`nFv5M1x;=N;KEJztdhnF=T%~u{aHd*JVE=Wc$OW
z90XCQUq6KmR-1=SVbNfN{lK<YT=3>5J+;_g{{N}8rI<(3?u1lOAX?gNI1Q@Xpn`|=
z<#WM^LVfvNIca5=8ahT3z@phM@^HkQ5uKOZm0uY*91F-^NP%Hd<A)-vr@%O~M57M>
zu27kGl&i|rIFElI4DGaPK0EVKUQ+Eyls0Q#KMVH<{<xcc8g=(~c}6T|yN0DG54z~X
zc&MWKT@|#Tt^T4nr1RG#-P}6Ox{Zc$^2j9VlXXE&26d2B9+m$XZ`Bh^eZ#CL!uB)Q
z=?V~*2$0za{7R>}4l>M2cdV?FmTVUf`ezL5)!Q00=puhCLem|u*G2nd$jk_0Q0IP^
zckY|HXD80^w9d&~<uv0<HAMa&PBT=ACo|26<wefc6?c9`NY@;Nwjxt|Q}<Ji5fZ4e
z@G7=Ub+?8|ODt;l@AEbxyxDdiT@?>c+<sdoWks-*v;{>dmhBz9J+C*N7%Z5XeXiP0
z3<E>x+FpO?a~~#+g04P|kl^dD?!dF_&B@nZ(8jjoQ(zrQ+vZMf3v_?blL<vA>PYYN
zE}|z9HQTmvuk*yExDbiL18K0`Zg=kA>6URIW%Z;U0Fzfi>5h1(2mFSD7=txTprJLq
zHc$!)<q9~lsnekZl+Mk|*j`50N#!MUqJ8b?6as(cb8sYWFutNz0Snbocph3|`=$zo
z#_qc&4E3D%=RKD@(c1m{Ey0jN!Zt*zLUmV3@g;A?7{bv;0K<}lxn{r3eD@ZEUfpfF
zB}jre)cUCvtq4b*=)+BWCqHfbQ6y=bM~|shlvSeR3Y_wTB33dlRn#lX@D8-D)76`L
z8c7q}@T#q*<@#p<!~Y+jZ~p<05x|$xvM>{~H%&eWf9JK(=eE(i6o@S%n~umvp0&|F
z2R1urd<Ivud8R+T%L5lJ%U1XD$ws3|kKRAB(Llbj>9jNWhLB*@Njvt@z^+#?(jYaB
zZ^j%LX<#o`aMCISk7m|^lLq#k4kq=D8Dv2o+Yg;Ikk|cIn#XiUUK(xYy<S?Oi^vY=
z^?{cLfA*b@R=xytf+*TJ?9Z5KU$BpB_-O@G-VXgVYE8O=qXv~VCNu|*8tHdhEZEH3
zOd&fy<<L_jy<EXndkQO$Ts6|~^jWZ-ck7odl5hMz<Ewp1U$%Q|irlsFO<@SoXzRw?
z72LIsq327h2ksi_>wbR?8QpwA_rPBx`$olsf8HR12R)YGpzn0xu#tYF<c$&ecJ+*~
z5a;9s1+;bJH%eZ&$ak1$+xhY`1aK^yE*k}}X{>zz@`=1F-&<Ld@AM|Xmh0zKrt!pb
z-E#TTXyi-yZZsk`yS~SGdfDgO+4}z&ju)HmSBB%o%ljV<N4>av({SWF;}^TCxPevN
zf50kkjg`;OxHV>OjhS0x=GK_GHD+#&nIoGcn<JYen<H08u8v$CxjJ%n<m$-Pk*gzD
zN3M?C9Jx7ibL8g8&5@fUH%D%c+#I<%vYWQj9l1MlcjWHK-I2Q^yVWah^@>});#RM?
z)hllGid((nR<B&F-o1un9g)`0k>QAiPNi3H9I=z}e*+ylj-cOZyA{wOYcYnea_l&Q
z-S%baRc+E{q-vO{&pD2t({#Op-^e&d+OglLTYY98EzfK*&llki{YHJt)Bg)0vli@^
zu^c}Xlhaogm&qYJCzB9Z5`UnUJqB_N<`(3doN~$kA3{;wNvRj9otZ=j**)6Gl3s1E
zilV1KPxSQT<?{aW^-aJ@5sggo)B7JMW)D#NFQay+#FJu-NWo6;f1Q4NrIbrb1-Bcs
zo3elF^jpaD7toS&dmb7}VLbo-{>uyXgHgg!Ykr~x?4|{Z)8F<;yMOoBA1`qhf)iyY
zCKV7nMwkTyyN!0UBQrZH*P#965YL?1-(V*O_B2$}@ADVI)U+3-U(>m0FGW884{y_C
z{qJxchBNKeOiou^>?Y<BxE}%Kyt%90-3pO|7~Ckrbt?ute~|(@zwqv(_@BF9yHDup
z%`*Gk8M+tmp7j=3Nq<hLJ*x*R6=5ap(9^)z;iB$_%k+1=(%mz{PWfxpJi>Go-qzt&
ziF3ef3%t~Jcp)W>_pO#r@bbWztdxTpo~4@-4Ip6R!ZWz302l`$;bDUF)xqYuW9Z=3
zg^W~0C}^g5K;|j;tDrU94Xv+lsMp-I6q>k0{yk2k2#W}%>3{3&p|4u68H{v3GgzPE
z>>->*FEiNqn?GytHfpkO_G~(KL5Exss&q5i>9NR%G;T3^2^z>H;Y^i}#cLoW4V1yu
zM&tWHoZudt9RR-hqtYAi(T792uZcD$u=8)GdT7pbOuO92v@;-F@-DD21*9;?0#2LM
zrXW=25ra4Jn13Pzw_o)Era2K@H&Z?Uw)qO-tZRgo-VQ5j-xZZzSbe5ErxN3U+7@fI
z9ah%;5P_@<EA?bpxd0WCwL+TZHFMy)2BmuH2b8uz$!v#`(nOfPh0;Yth0!hf7k_EM
z6CkjVqi{ovHqDR+1U`r(tiUyF2bYutpl_$l4hb_iXMYsa71|%oL-fSQtR>9#6c6zH
z3}{`pLkod`?%S;)xj=I^^fgOnOC}|{K?^X}$?`&yJhL-CWS`t4>KbU#7KXiaIYSJt
zuoB^b%A>AO?ZO@-aAJ@25Oqn$3g)+;b99quZ1EBaU(7fIs5yjyLl?1|6tFo|(Kb-y
z-{$crVt+@7h`NU6=^{@EAX2JY(t}Fa6-I&*RM^GQm~a#V^9d2dW5tHD(BQAo%fta1
zqK#@Mc-WnvM=Pk=V(gIvGEEi;E11b*&e?<={ymG(hr(Bh%tD#q8PBt~qaK?+tjdzc
zF<~)ULb*y+RdYQ6vMZz>il<xwLK8Gfie3Q%41Xhvr#=Du(uED#(KtxK<b0;RvCe?r
z@t6!!x`0o=g20HNtRBtDr5%d*F06nuM9DI4gjiVNK0X9l(X!FIFJe6uL+Sy7D6o4m
z1Z)bUS(#lyQw*sBhOU}YN~mt`_X*VCFcb@+%wvda5Qb>sC%Xv4C4@1Y=i>=TUb03x
zv46N$&jug{L^Fc}5&r_hyt5g6h%xT(Y`7*Sz2(W#7)JqdOu0LJ_x2Y(s)Bn_!3ptz
zEt6&Ww92}cs~yo9m9uej!{c(VIU{C$1}!Z(CT9emiyypi#j-{SG6;@0Rr%ds$Q+sH
z_UHTiRQ8H<t0p|e)`Z=i8%1IzEFVo<>3^T5ehF_l=Fwx3RYTE+b<0C#H$7B#{h`{I
z$orYN&A=HzBj*3m9%a_$F(}p05k?SdSQk2&Q-jL0%`U|*2^Wl<bW!_q;bBV%dfhSW
z6)@xk+Swf*=%?M!$45i7w0rw8AthmkCqFi)NIq!_e4t78r0)3heyqqsR((}9Vmp-0
z?Cz@s0^y=;y^_gSRiR!;nw)vn4W+*3W9os14NZ3IK4vn)T$hGIwh7cFa3`%9Xj*xe
zMA!XmZ<A78B7d*yN;SaX5sdrm3e!ZWQdd-~D}_kjE#58`=bRv8&0vQY{HUDv;Jv`>
z<NkC_f2btDO8wEL&(RrC7b3Vh+sSD=$jq|7u2t2r!^5?qYg)w(5Rj!->AO7;W(^c|
zzcVa7B)0UBa9r?1hG>e-ln>3=))KJ=T<h91MGjaspMRRPh9Cf9^hvCa;b%x=$B_K4
zYD+y#5NO<ATNEW&7FJES`M;5C274=4K@o(`qGGj{cU;gE&Wf@?aBm`vxoBeqvtgW&
zwJe7g-IBGbr?7DL8k^FUxEhk4c)HP%Iw~U+#ky8hHyu1!_b&r+kVHZ``__fz0a+&j
zja|Gf5PxavwQNSYQU1_ZSIz3c_p^P-qu{!t8&k)SE_LHlL|~47)fq+qmbD0|tFcx+
zKoDjB3}#9$RN3BjY4s#Z<d<LGClzN%sGUYN)`9QO;a|*CSV3mcp36#BaApQMr+$%U
ztWjO&4>CBTtWr%2RoSZ(t77Yo71h{LO0~PCV1HT=UA4W9$x@6ZK^6)tsC!-8s}`dk
zD2SwYw-`(@V#*e)v9%b5WSPIbPr9%;q8jti+mDy6nO*w#7}^&uv5T%$r3&k}i>BeC
zoE=EERYI*E>;TS3Ce(~4)b=mpFwi6`-^=o=0_x1Dc}-P|&_m%8U|%=C63$fv1a;ZJ
zc7MZ!Bs5E)<>HqB;+hT9t=~hL)Mo}9vt+(43^#Xrf)@{60=2OJ?=Ate%i1)?fv_%o
zu=MRT=XJcbEwbhA;*SKY8iGYf0Xzr{3Pho)gg9*K{XiPG1BuN_+JjPB<3AZ!a7#Nd
z@r}`tRBIMHcv@T<)R-ONmyl$AEfaegT7T*A22@oN9k{v=SC?JZ5!v@on9T@!xKg{C
zs2e{<lGTi2RG?z};2POoi*f+!7B=Ft%a?7mwy&UXoph0A#3IO;R+7c;M%9g-?*BC2
zT8B$D(*q=X<`m;Dt(8g=uKFe^t>@=&<+nyc6Uh8Ev4s^ViXf``{zqE**<9b47!TC&
znOOh-_xG270Kf8tP?xbBKNORcU>29JC@v<G&|t%V{MLtU_^VG%{lbT;6|g^Ki~`~R
zYBqJh$>9=|MH?=`A0{3z1M@!ZrP+no`&TVF?D<cd)Nm%eCXvsOjz^W;4evk-AA{d)
zd{^z`P=-M>E+#4WuSR2Kk3M~t+IZ(ZZRLHiE#M*nQgIb78sQQN2p(VvP>`tu;cqx1
zUJTqgzVe42kIsOFd?=58(-DEY0n-*@VyggVM!?*jW0r~;D^&w5GSCI!B>6;)i$D#K
z2Oclwv5XntlSE=Jf4v_xiJ(k9pV@ggrc$A_sWa<S-zBf-vLZ->;dye|20fWjpoOU-
z=0?Qy(AbBDUDX@*+}2OKw_fUF*f*)z&cLigNe-;z#k3j{zb}7b%3pMYx?C5=UVDQk
zIfISP!dT_P7;t&>+hHh{5f_`4=BiLG0~{00hs}jDu?6ksfA<Y7gtFqP#*e1M&<}*;
z{i>!Z12o2;P1D>4-U2jZ08v&!8%489&1CRfFaZ}D&Zup&##sWGT^f_pIOQZ=N!%-{
zDETtwR2wi~>B}4w2<Cpo2mx(9mrZ1|9MFkCh*7`bVT#eY=}aJIh4YbyG}ZocI_+!!
zeYKx%)-G3Te?NLHt7nXvvGil;F+&>jvl(ckl*rtenlevAx>|1p=F#%K@&$c>7C=n%
z>)O^Zuih(~0gYLNZ*2d8ntw4hZ&3ZEg~9+ZTR7&$xDAYFYc3C)IGH327*o$)n;Pb2
zpa#g}n@s&?CKEaYmQ~8W#4j)6H8>u%<?JVR)BQRIf35#b^=xMA)pFILT;a@tvu+Wv
zea(Ng`>JLzice(jMvDcob5*q5LLz5;2FE4s+jGl*X$8z-Om@*mpNs4&h>Ss+ET-4M
z*}kdoH}hMNELg-UOov%LOl!ji-{n1ZToa4@O4n5eYG3?wbX{hX)i8(#tgYv)<A|9Z
zZHVdPf4%4hUlNf5gk6mPu+^ZA_aWkrn|pq!5uBbUIvy9on)Gmt=8uiZUX_*E<|_k)
zOW~W8@im_hy|i#VznOUXyf3<xqlSrj`j|?2R?lzO`^q9%<NERXO5VhPK`B8ftYSqo
ztSDLY(R|^dTU;|<lF?TNXs|q!)0aVWcY~&#e*<A`6|~XmQ*O^^$WjXS$F$j%PbEgZ
ze8r0?12<qE)r`T(+Qn_wE@;ajrLIE9pH<CkLZ=MWpm=<tgL&EiCQD^C6G*D(w~6<6
zULHHq)cxrDeW0M0ff^dmR8TW?e3RZ;KxZ^4QO|D^SCGBBcDqbhpsh-nbP9|gtsJ%K
zGwNs83z@q9Cj0Q3%GM-7;XL)(acNCnyXPj01JvL?4-0GZlAE|#H9ZWuqPxqGIJro(
zlb&TR1WI?CI+w8=KNW-JWryWu0f*&f0=MO50|>sCf-)^9mlPxfFax$IL6=7)1R#G-
zS~EBRVG)LppTloorBs$a7y4)^5$WUTyQiZ$y&`zJ0_Y6D>D6iA>(6=Joxpz0*I(%f
zyvFx9K5OF@j=x9B=H&2eqploX>6~T?aOP*6e*gI8MP@Oi2$sL{VgBp=<K;gubD^yC
zLU}#>@$-cjG=e~{NLJxU)W82V{CI!)$DHaIG7kiqwAPpjWFQGgJy;tA@=}7#Nct!#
zH3_3h82v3VFErg&|NQ;eFgx?>_r^C`Vi4w(m7Ks(uhv@lFuxcVY|x2I(x$#<O<h<Q
zkFyuZM!cUtk*r{qy<kMpxZsa?GsL8Ei64e79^VqbhEH50#RcHT3nVQSz->~#gDI)}
z`<f(<=OL}J^kxl}Yev925*2U&Au|G#Gh+k_C)x>(t}2M-2}D&V{v#LkU7Nzq4Fv{4
zN}N->$?vz*tChUl4D0Vb?NWM7UQx_RewVi;1SbK#m)|7>FMl&6#g}hEt~n<qwa^CU
z&re)WIKJiH6gZ*4EO43FTC_sU7H{(>TL~jo_GFY*=ilt{-yilmDxh#-p(m+V@mIkM
zsl(#x%5~&OKFkGzVW<{#&$9^i07ZquxCPGgZB;HrC8S@VU@nhKqREiL!a@}+0?k8I
zd7qmFKCrGO<bPeVFw$FYAG9#G8)ny{IM5MB$JH!BTY?tVJWBwj1nQb56cP;_AD`#x
z`;w;dAi!Iu`P=Vu_>&fYp^MoQ{x@Fwvpz=-{~rFyc}Rya*)<b(P6VQIhp-%!2$UEk
zvW<kn5Vyv|BsOwp$(dldPnJ{~%$)AsTnmIEcS`<4pns}HpmYs*Zz&*fxIi2Sk5waS
z0!&PyLbF?j4mQTp?mY+^L;00;<NUA0s6~FIM=?9bx3Mpeep`*R7QVYi-gCZ>@26-6
zZyO=qD|!b}Z3HMd2vY6Iz13s|@1}Ms_z@f#N}yG^@@FGJyCvx$gj#@A2O)%%vQBdP
zPHtGDZ-2=78_ce%));M#^Q~*nFnj@5<*Jcs@P{i#&i|6_ZUlQbC>^jVm@A0~fvqTa
z4z)Jdcn^oP4kIrRzhm|cy6s?8r@HlXxn4@9Qi$nn`8t<c0I7hCYkov7U(e<L*F-%o
zF*&YNLfS0RyX0FSq#lCJ+^ue)to3dAIInHAf`8rG>R16Np|n283Md0Ll%|ZVpy!yd
zs43jDilnlMVLWaXI4a*wGAzs*iyZvFHN0CB9SuuThBpUkm})oU@OCGsM(f$i)IXw6
z9W}<EX~c6`m1-+~yXCx(ADK?f;|{@3Q>947!lYs?W_lTAMdvMDsEW01)vH<JT~p7D
z`F}Oly%BpH^+9DmzSTbJCOz!OtD^^(lE!27;FYkWj$hu}F#?1IzL2q%3v6$7EPI&k
zmaT)CO<CbFm^l!(^)_*a3i-W)SMlj3AGb5->|-CP_|@w?uZlat*E1_Usph9^ccF*)
zZe)6h8==)_YTZXEY^&~>s#1H3`*{~(!+&H(ql#VPk!8e4c`lO})A41sQhRWEq9sPt
zocBen2xvb&J<&1<qgo@YBSh>y_8R$Xtn}bThvY4qvpSy)gUVYZ4Xv~-$!rDOm*GP_
z(>7M(L>bT87n*Gd3B=euTb%v1wqck2Il9KdoIf*mrhZ;PwLwT}k{j7lM01@rfPdGK
zQJVISTs*l~x+Sz?lBNnS(Md%v>zOH%Y8(>s_iO$=F>oDLur6mhnC#sEsI}1Us0~2L
zxo}U?>wIsdlzz+h*wi8QURYe<py(KIUw@R0MW8d!?~fu-f$!aCV#dQHHgd-9R2`H)
zptQn(9Ssq3fWM!7eLmURFV%gtRDYlX&OWabTljr8-O8^gYdxqPsXWz%!QW8ss5VJ~
zX`AA9Gu**}8};VtKm_3|c^ryJ7HiH>+8f0$MjLmv)mT~z^;Lk=t5%>VW?$x0U9@(K
z)q#tf_dy*5m!&$ab+Jc9f*B79ZVv_4&8tAy%y6yPrF+!Ismp#4I^c2}AAj*6a198j
zU@~^`Z5SRK59dYeJC9gKW0hq|w(7CB4_x+3-9eLE{!Sl+rk3$oTpaK0_ZF2Rl9z<d
z+6zuM93k9#-$>`oXk!;<zvvxERf^hA4{}eYTlwV{v^l72q`(W`>f9FPlx@=lowYc5
zuf$Ir*3Bfk(2J?^-O#~934iWU%Z($v{kU}SmMQQbv*kDy_pR{7Hb-cCq!$X%!Za4B
zJdlb_XPcYOhGRiGuS={@?FCVE8)4oJ$9<R~q_pA?m?2QPsM@Q78N6-qY9HUG9fexR
z(^i$=rd0{r(({jy@)~S|bie2wNPSE_jzQELs$u=(!{_pfW^oU^LVtGZrx~G111(@5
zfYVQGM7`(V8#Bk<NOS;pDg8JGVP}MtVaIsiOml8-L?4gvZYVks+mw?W1F_XYnt3G?
z2e<kry63!#Yn2RURI!69ZbWsr^c_@n5dMj?_sZ{+=~jL*r6q^tJfM|;Hq#$f_UHC4
zt$x2<(vC{HjdlNic{VyYsgzC}vs(}Zj3(oY5~;fQahC5V8J}&=i}Jcx@Nybj;gUCy
z%l~|0^SRu^<`*X72HFbrmo+Q|BwWNG5H=#YTjCCqG9?nnASnZF@%BGFX4J1YqbY~^
zG?MaxP`O*;`$&eAFB}47@Ip-o<E{F?*XuJH!meK5ZecnYDIq^K%YKl+wn0n);qT{8
zTSwXR<NpA0(Nvcat~3+3$1DW8{FgaQJtvnilL!QV9Jdj^&#&m*Y#>YV{Qw4o%#3yf
z_$Fs>h7W-pIR|YJD~J636v-l4WOvW>K4Af)?q-wq{Hj>2UOxRk@#)j<^sxJIQF?OB
z8{y>i@O2XKg2VrU!=N>j6M|VKriaJrmp$jw!tVq>&AC$Wb8zqBY7Tcwc$gtzM2N$k
z(<07)KjM3L5O$CA&u|s-U0mbhs7*6myib(P!Kr<tE*;EtPDcs?R*Tc&*N2~W9MuqT
z736yVu=`^-Dq;d>#!O!xcX3=R!IYPi4@_G<egEh5b@#avD1<rl@ecm~v_R21%e-`v
z<S)PS={I=!6HHdYPyb6(A1fO^eVhIc3~mvB_KYXO3Tz&(5$2O*!3QMUNEj6u*Wu|T
zwn=7!W#PKYd;+sLAxF9ta$W%BxV7L93ORt9G-`rz7OXFjiy+Vd0s|m130x5nc}xfR
z0boqLzvTDh{0Oe}3=jlz9MH=Rkhq77z~w4WkM}xW;hl=pB-|HPm%NT&4tJIdL~5je
zU?QIHUsKR1Wfs=pQjg2y6kw-KCX>b`KA(v~CX1EcRYeoYC0(!U1&{bv#`yE;DaK!Q
zob(Jp0q$g+hvK)Qa#9ynYfRSPMkdh&Ne4f^5jq=I1WX0-2kV8(qNVgxt|z=|9K&Cv
z7}qsvo<-H-d>oroUDWZsYCf$QH?NO>`U3ew73mh35J;x5J0dCbmVm;i>$td%sRh(5
z*;y&ZL*3C{<oeT^>o-&NIHr)=G0%N5n51Mv$+t#RNja~QqB2SS3W(D0j40T+gi}vO
z)aZLe*DIh&<2B@Y+tJhvnfR*LFytF1bd@89=(pq^&Y2YY+3p#P`IPWX;(WMNXei8B
zaL)L#-q9v>`{oaS{SlYw<s}l9*nY`he%!>RZ7*AykQEzwkkq=B30<<03(0L;xzHsW
zxsagV%7rf3$c12YD;K&Xb6G0hyF*Yp7QCQ-mw7A&BY&9nX-ZRfiMj#uX&7(^|A3{l
z+KI^8SZ}BAF!95Ge*gNnFaHn>mi;$Qg>xyKfCR|)m>-P=^TO<J_~@AyHmFIk0DH0q
zEHEWG;Ha#t17D_{mJXO3Pe#ARQ8@B?EhSl;4B%j5Wk7~dgh;Rs7Ixq0U=gZuNE%fs
z;!~u8Yk%V-tS_tLWBk$QG8I<+YubICcAwFv##-y5&%IOzuGV=7ngyFeltDxaYS%KP
zU~|xr(Th=ngB@yRl<HdtB?G%?uoY}G4z`Kt7UWiSE+a(mG6b{=g)&sG<M@gl23=vK
zLeww=NmWC%o74mlXx(h#cK_uO*YzEgLg!9Mm4A}XodIY_D<Z9QgooU;)HTBCWi~B8
zW9_GKf9?YdQq>v(VE1_<;$*$<nqr?f8v~-%L2|xrj4|Gc5#WY9Q;o3FIx9LT*raP!
z7`tVu9xg?#Kiny)ur|`CR&Wn*NEvQyd6O&Y`lREX-nY^tme8fjPFL6H3U!jDHpc=b
zHGgDgihu){wQ5jg6skWGE?4hyj#+4^#5zoW%!U62Wmz_=vYy2d7-Weu$eAIN!gchV
z`I>50jnXv(meQ1rRHszat~(<gWtuZkM@J4k$E${}h(Uan4oX>Vv99#ZdZmrE-5c4;
zn;b&AUFWM_ksi{ucBA0QJhF6mbv>-*9)EF%KkOQtBZ&cXkV#fwC?sX@<WQx}eP&wz
zEaj0Fu}gWR<%FC^{yp{6aHqc`TYbrX9w;~nj)k7N_E2*KS0f*#Hz$Op_Iu(mUz~K+
z(v`L~4ECo=`*<*ZPa=6OCGl}3Ki*gY6*H(gtTF-pVpEVv^vwmmYil^*X<vKrwtu2;
z6BzA~nQZgEjux_V48gkAlulF_a%q)Fhx&9x9%l?!t0+2cNd^sZL$*wE6#Qn%6!Bd>
z(U#{Cy?Onm<l_7dB=^(GTlq1si$`sxDu;#Gj9lY^Q#nmjU(*7;Q9CfM#uL#zS@c}-
zHp7>Jz|ppgWM6DX{F;us{r*EOQ-3==eG)9Q6lzC0jY+edXhGiE3UXauSW5I_E=9El
zM_$ywg&MI`MYXK{Iu~6k&w{KZV>#+K8}u6n)iTkUb;ate5u>uhsC*LM((D<c<TXUa
zZ*^LQef};78ZM$jLn+#Nru~*0_7&em5|8$de6=Cf$o+r_#cL*M)au<@?0+Qs)VE2u
zkEmPqTv#j%(x@lZk}adX?GTkoLL$=<<&~HdGQE`Lvy?T{%Xy}+RL&XYjd$~T(Y$iG
zZ&GwR%&-&CDv6iG`+~Pjp;)3~zECQ$o#ZI;ie)^&A;_qjC8(Di^-_A8;ZfZ7-R=F*
z8#rd`Q4(%$tO^X~%;YcCbbqckXe77l<cxZ@x?{B}QWC0#C(BlB8^5@>5?8KPMYEX8
z(v#dr=8l*ZST$6<CTg4_1~HY^^V5*?@_O8Z@`ryuRIGFh5^Y<^Bw9H*q~{CWf{U=O
zD=KSh5!EzJ$#ZUGi;u&f2(-FTz{ycTiKgne$`iU~HX9u2z)St?{(tL31hvdxQ(E!d
zQCmGMYr~vZaJ1npgyQhy#UA>e9>k9FwQ<Cm4N!@Gk1emoH$Q(`hnUeYXJ9zSXW9dk
zGmG;={o4+Mt?+o&UA$mDklc9RE3nlJa0LgyqZvAeEpRZ{<en6l1n$K!i%QOL24%EU
zZY$2Z+0hxwl+BTRgnw15Tlw?w$2rqdyrbbehJP0PNw1mq?`MUM-`k*1S~$d6vk`s$
z>~-{&qM5gE#x-M~`(wy65h+z*UWa$|MaQrrvYZ5>deV|Abp?%nKWuZ)L?XB^@b`pb
z$MB7?$OSWX)47SoZZ;@|yFHetk{<9lDO-78Z#)x!!UZOsdw)L%^bG$j8Btg$dusjq
z9%kqmz6^&+xd8i&<aVjd7D~aj?*0PNB2Hj3Fm>tJlWtS#e<YA;Qt0~H)2rV(Huv-5
zS6#09FD0!*LvqU1E==}%IStlz^)I&s(PUO!dwLq~8`jV9aXkfT!Mw1ehVs(FbxEd`
zkgZf9Qn(Kz)qm0neu=!W9|s|4^18MK)?TM9ZT>4~#sR#d18+UQHDBqgp9oYuxyCQQ
zqu!mh*3hOoM6?tA>e+Xo(bifx=A>V-^@!x4LBP6)Nc<)ON@8;)8!?WEp@oDa*aR*{
z%ZKgqc5GW9-Jx-t=kviKNMFv{iB}KtPO{)ewfL<?7JqPNheb6E-Mk@+(k``4bAVp`
zD5flvRy|EdrDKvE6}z5dTC327n1$8(06X=*y6`H$;2EUXQOM&Bz;Y&L)~DT$+_S9W
zanft8pqay?10^;TaPH6nBfvR#kz7{SoUj-T9%^abzEfUn=jZS*Um$BrNljBvx1L$6
zhqhrel{c<OQa=&2!t}+>B)=!szapEWBOm(QjAunWZ7K~N9n<m9WJ8<Fm7DQS?q?qX
z$;ZI|1HeB>(U-wj0u#5@GX+3fm!FL;CV#zK%aZFh4&C=x=m#!CkOUu9U9ReOw{KN;
zGt;|NR+&jQa~G3Mem~&*A<2?udls@Ri2{KG5ClMqPk&5&`u_6%_43n2s>v{mV8rzG
z`vmlW(=Twksc|v@OiP%)eoVi-axOF-?{p~6rKH2qe@kaPeUo6R1<)11>6=mD+kao^
zIMZ{8?ZSJdZJ>a32zv4so=EqaT5&owwbUlkrF%d(;qCP6*Uv9Ji2!tl=KT5f^1qi+
z2ojD7fzxk4Ui<?kfk`VS(1IB^{rR8i_m_VZ3Yjuz*8ij5?;T0YIFZy^)7f~Ezx>Ll
zf78RCsj+l+`YSa3;hS^j&`tlH{(nXEc?-G)0-bA(*ntj?87+voW70p0pmTyqeMaCF
zfvE^8MgLJpn2_L8qzFX-tBZ$??)hsK&uD$ylh=nw8fAKX4wOSP{~E`LY$JDvCQ8S+
za(9Tn0VyJ*{|}>E_<b<l!mm;ci2j330GL2Y#0^u@6chM3ebZc|5f~H;NeJoZbWX}v
z*0-0zkOdZhZzLl^lj(3BM;|}$w2#BMY~U>=0|!qUa#Ay8Y{uJrrhrN}Mml8Y1B`qX
zcu1it%QbFKM^`8?xkV4ONKQ~xH!00dm=D9T0kaTHNbw|?1u_IzrjrETK6sS{-N75+
z5(!3~VDt{qrkUv+?@8DR>0!|~AoYXV$R|MxilHulhU?+*hWnA`1KkXK*SQB*U4PI4
zJs~@_9l~MOC6vLX*|#}7dH$KM)lP&DBe8)nCnBJpgfN$k;J6)P-zs=poBj3w1LDI_
zT!$EK3`(8?F*?Dtk_BR{(M*}x=J39yr;>bw0M)}P6RdIrR=E$><Kl0?YAg}a5082t
zO}FrWt7BmDl&FXF%#6T!)Fa>0ND85FF#z3<FHAwOZ$%TvKD{26X9FjU0mXBM3;J_)
z`fa4wJNY)ww_VJqtM{VF?i=N8{p1?_ksm<xE|_%fle@S0B0Y)&XlJ}~9FPs5RSYSf
z1GKWta?=8x;q+-#@K4_0-it||do%33mq=!Rstd-=EU**B!_sb`C_@zS91ta?;i|;e
zBEy0;#*T2^4I;k*$fE*p;3#~o@T4RKg=W^4=-G$AHTPe4il=X=z#TKU!{Thg#F>VT
zI8ZQWGub89ZX~hAS*jHgS#_Okt{?G%@K{0GQC?HlQruG=xrvn>ugzE@>xKePX(pI|
zq~}lt66CyJt3&sxeQFT~*+~Seuu0(*5_sY1^nJKmES*s=OvYrsFP$+bbuOI^RQDJ(
z!H<flYo<jJe=cmy?W(Aqm{A}_B+JjAe@w%je}1nrQEx!QXy{e&p`w^t8^yGNv9<j4
ze5Ss+?*#o1PkncHsPCUHjxQo$eb_O7N5JJcr{if2N5K_S2y-0o=1+j|+FVi)bgb=B
zz$*I!DcqZZke$;FJyeR0ISMjHiXV$A7BmO=e1B+NOzRH#lwd}|5`N0<Q!Ua&&u7@o
zv-T9uy=T*;gEz&x{n$ya8KRjD4>+t13FrlzX(Oj*c2O%*kn52%==hQYyJZ!BKw1cH
zd~jtHmZG({t^y?QNCsq_IWk#XpCA2On!9`E$D2v)<jhaI_?+}lQMBP?2^2xckgik$
zY))iqWD{c^>H1PkP8IPp76Ef|7;+dBD%!;-Y=uCF1>V4sJEq;k%dFmM&=B_C+qlu8
zZi;%Cr|Ys30cNCplB`^?TAbT|n~ff4X+u<+*Nsun5mh>%hh;G~`_D~T5A(DoD|tVQ
z_c*DWN$m5>o-A7Zupl1oWEnc6<x&tY71@%mMS{0e2*1nvQO;a_1^A7Ea}|Sg|Ig-_
zFIR~5Y@N=lcq@&Nl7usLm0q1B2Fj<9MATe5rv}*7(+2)(;sNtKd|vH;PQeIW^Glz$
zbXU4u<-SF%h{wAqm-`5@e@Ta;bMNSmrI=xYNbjASq`0CmTRiWjFITGkyQ-_O^TF$D
zrXTP-4R9&V0B=B$zY|t(+W@sHDWA%BbXsAERDwLVsq2~CCS#UpWNdUnBPWPW`@o%w
z2o85^Zwu%pu9Ch}k8i!TRPOPuf4&)bxTFa`Pz3p&Q0?U@mMQbPcIl~I2UlG$c^tkM
z?wde*wuaszubX5$2N{0Mcm%!03nyROHTh)=;N+`RiQ)%{lR3_^|E(qbcIZSkVosOy
za=vbd{Cr>p4(iaf$#uP_LDepA*$5KAu}XqgObRtd-#!Hd+^8h%$;5)6e_qM8Q?)@D
z%WPeI-r;w*k~C0`ftpdqSv@}0F+NjoDk)7@e;2&4sD2RhD&yq@uhq5rTn|2b6Z82J
z%?m<%pvxrC5pu;P-&{078RpG_I(*EsY6_EGR*b+%wYzhz9HN0aig1QrmgD+;s!0Ww
z5tYw`z#Y6&6sGMA5x>_XmrRxgAAb(9JM`UCf0n-cULK+E0hI^pyQlmpefLza>02Rl
zb&b3P1F~@<X_~|Hf#PV=NZ-q&^=^E1U5~xoizQ92;AXs!IOb_UDoQBIT%0OpGqi=G
zv=V1E!5a{jnx;4yDXo>8m130T42?@v3n^|+AwnqmEe`mm^hadU6p4w|?tf0pY8Dz~
zDA;9a${Jf6v8I*f<>@80Z=#`eXT?3`r)veA!a;~Ob5LX29X54cJMvO^@U7Lj!Ax91
z6~<C0C3R$3a)J_eba$2z<bpw6Fcvb0Sk1}>R>wRVK^}xXm8({;sE0!Z;9QO$W#Fqu
zE2L%0&iF-G98%tqLR|-5$bTP>PF4fuwIP<!9a=AKq4vn;?bd#0y8kcr?TB*A4A*0i
zgPjd)E+czOTekiAqngu-4+BfKYP+yh<t!pErJZ~83a0bEq@2d2ZXXt)9TT!06*>PZ
zI{;LUck5#1On-6xt|zw2ZQV~A)<%viuj}rf#i^v<p5DsRl>cM_aDQkV8~|@+6Ic`C
zd-Q{r2_rfMZEtg2UN((0{6KMAOEPV<_S)jbRX-?;5$IPKW=y>u*QMWqg>FW1MVp#h
zX*;$KqTZo2WtfC+RF%yiP`6Q+s#>Y&U0v@UsO}hEk4Zf8V(i%e9Xd*jup`A49VHX#
z8{_){TT7<VQ-aI-G=FLYLopNmdW663^?lrlwbryTe*}l}1dmU_gsdgcgMrXMrQ6`i
zo3i!72C+QD&A%kb@}5e;Y)RxB6rR;Lh<qh(vdHKd-;kpDs_q<JqW^S6T#l;&6s5)b
zUN!VoaPjMyJ&->J?=(^e(XPO)DXH=x4R0L(I%ftmmrNP62!BvsM7g9{I(rl&qw2y>
z5V*Uuked&^aJ=bbgmY9x-2}*S8LhO3X6)_KrCC?|Os255SEZOiDQ!keSaoTR1{l~U
zquX~9z}@_Ut_SuCq_NmDbn$>~WC<V(4AC9jMV3kr1C0kTm$I{w@I&{I<)onR<c=sk
z2e1RdkE=mb-)(rze<Qv`S6?&XBc_{pZAp8>Z8M$%K_C(Y#+Gcpb9J+#WMFy;xT!YM
zcX5^TRW+s}Yukn5<vU8g<+EaH)iGH8)$B8@2JGXtOrbP$L#JXn#$Mm#7emDrzIX5S
z>zwDx8rpFF{CfEtkDuO`u^0*zx8#`xxe<TaWgOE9i+_*te1&Iiu6yZpqHyWm+quRj
za>taEe?)+91#zRw8sCt?_vq|s8#f$|NyQoRB2qc^*fl{~M~}|W>e^{ubqf2QZch69
z-Y7)3(nNye*s&cweR7RCBJg?;Q-d-^({TRAubsQh8<>6AO3lkR2Mxg$11(nG4XA(q
z!jDG{BAl7p?jBhhI;T2`!KL4>sg70w`R&oh8m8omlC5atkU6TON4lSfU@x0o`V=Eg
z7kMs;P9ql|w&I^aXE)N&=^KN{S9JPqXUHlaGr&waW7D$bX7*;o9wmh?BvT$So%;+i
zOOO=BxN`2Bk7crC!p`3<)q`jbU*CVjGyz^E2O2#@Z=xIf_quIhwX>2baIZw@|L~++
z-3X!oa&dg7^&nNK2M5S1CWI|5TXqAAvbFV&;=xicpOi&`(^R*V6n<T)cp3F86$LN1
zfRFUH$M7y8WOx)49f(#4@#!}H9>7}qMp1i4-6rV#G9J;@v0O)X-I<uwd(D6I)1p&t
z52ve{2QaRhdYG>pML(Cb!jJBes!xh@-KkREBi6&%Sjc1_lkygS*yziKa~d=GPDj~e
zm}Z*qE`29(LWijzrC$MMAF9x=huZ-!jmK!|J`HqZyo(ZkM;VE{Rn@#w={bK2?G*F4
z#g_L`84iz{i(<c#`vtv7H*$Y6Nj9>z{O)!#Y~z$}+<csLUlc(fUOP1(2yWS}D<PTv
zac*S$x!K3@Q)uOm32nLpPPG?<nyWnLK~3a9cVTX%Rc)xr`LonpILm-nCoAh+gm520
z4SYUt(VOPjIqj!B^~3a1(gUhH(QGe4DYN4~^h9~pHm~v1cx*HSCpv%q$O*$sLMSJC
zoUgCR%EY?2NV6U2wI8NcBa=-tB)d6VT!v1TXsaV%xsI0Oda{Gj<DEW*&Hh?}kypZW
zS%ijZDzV2^muW5{lnm!%pO;Y;$P$=BtaqSlHe4<$iGelOv5vW}0Kt^2?$ma-!LH?Q
z1qyv)y(u{<Bi~%Zq#l3R0C}s1NrE0G*GBc5(Pt&-Aqp`^uLX15jI&#72>$0&3x{NO
z{RxaekLa)Xq}=bV43JziL~|npq}KxhD*@a<+ith=>WyN53$otht9SOwaX~WMt;1Ju
zWME|rsNBrhKurb@LJp`?(<Lqpwz6J`z%-0BZxywZ7Qp769!`IWwA3Tu8}D^eI^~q<
z7;xok6V;M{ZlT)8SiJ4h5mn#%9W3)HwllY~Y`C<zOOfN0EQ~Dk=RR8d!)tG;@_0b=
zbo~fUNG6;cKwUe56BJC*42$}U;;uOqj1^cpT9?gj#dCkAL_?|Tm%f$4xQtvX`|=&t
z_@+H2YDR2(S8`l;TPobU83mo<?Dit6bwQ21*y<me+xmxl7!iQHhkhHG2##qnicG~5
z$oVB>CP8)Do-g93`?G3t9@rM~=@P=akZ*61iuM$6Ua3d(;Qc>4<j<GayZ-}x?;QxY
z@vjA=9G3w{0VkKc%mp=nfg=C}4kb^2p5*D{>GkvJAMYAZRyYEyPM<$cz&9j+g5-;a
zP8L8I4X4kar(a*BRECdNK2S=n`LO)m@tMxg8XS)R{snM;wg_STJ0DlR4{2QdUmY7I
z5Fd)~T;d(=U-Br)hvAXNCb^gobN=n~mnWG`06s%=e*1j-?^ADo1Su!L;pz9EPvHh?
z5Zb8|5D3S%{q^tDkEefD5k<5VE_~z9kC99#B+E3$@Yx2Lzy2mqfAGy;c(C^F^uIXt
zXBaMpCx80S>0c~gwx~N`)TJ>*Bh(pfMVf_sp~J0&y8L#2Hd1B$kx?MV6B&EF=5x3f
z&Mu4$$P<qw#ufR036RToujv{=ktg|C5{3Wef4KY)TjWTSaF^GFdjrY{Ytmjw^k2Yb
z1#cq3-ye;)7uF~i?pTs3cn80Xs06%ZLKojh$TpZqbfc$2fE0#@<r}9rDG(Ippa?*#
z90F<x5^Xoayc>=cm=OiA{9|B7D^S~-Ofq=e;6(@K?}K+wHw-KAYAwLSLGWrL(7Niy
zA^L+Ce@pR=rpW?+lq-CB2>F5V9Y)oS;Eu@`@|n{HP(ELic=^WPb-2F~;@wcJK&+Hd
zS{($jB0=Eqhd7Lvz0L(Ae+Ad9i`*35OTMc)Pbk1~p15Yn{YKPwTepImWFdeDp(eFp
zb5~O9CU;c}86o@VaS35JKDTivmx`%J$u0UmT5i?Xrefrf%90${^IQr~DocII2`c!<
zTMaBTPFtzQOrywbd#?NKS;5JqYT}T1Vnc;hwF1{|H9cNCZBr+@e^Fg+hA89e`I$ST
zaRCzpCIrwW2TBMYn~v~DzLl<TEL!#hvH~=vy5bn1DV$9Ubb*s}&ildPUaL!9$8I=U
z6;yUpGn1}1rvLrcuAqpibT|e@bi&D@vM#vL(8wgbzKr6=(qq51D=6w%9UKFqw!-01
zKSPfBQC~gHi05ZQfAC|J<o0E6w?%6}fgrn5`zhgF5j&^Zog&sW%WO=bUv7&n42|pr
z_Y}<r0t>wps%h;nX9`7kc%T|=_~xgTd7`Hh(NccJF}h3mxx7k}mMIrb{x3;-pvF+?
zNR6FiLmMM7W%wmq6oQb-^tQ;$%D=e5auWn1RZ;atSFngSe}+2xwPb^O)#j^i8y0Q@
z{ge^DrjZ;uDOOm`;5dBx=zh%-dl(1>XM18#b220>yj~}@`l#f;2nrwk?XN#i-Iedz
zy}X>y88Xg+mbjj({AhJ18nKj%nMbi5{M0%rRGP(0hGkvVMLiJ($7io*Y>Bq1NLCmw
z9D50+xiHB$e+HFP@l%1w36xBw5GTeP<Wj4)F4U55rR021-{tEJI2ceU9}%3<mG2+J
zj24Bl(9a0aQOaes6P`=JUO&k3A#<Dap7r3CQ3}W(QIFA;Mh_)XAWig;R9jYA3rU*Q
zO)27k;zYvxGNton^psVCl-h`%vPM{IC+NKfKWOOWe=oRp2|?1leCyUTa~^N?n$H58
zwhuHY0e-rD0O@?^_Cd~EidQ`6_bh_9*L?&lzO?G!DPW^k(g=m>n4T7;XUD9hTwTCq
zrrk=3d+CedA<QcC=1a?!ce3J+Q*O;}9hdC|-+nA^W2`+1<@7NavlOlcE5teL$G#$b
ze=Bi!e*@Z$ox^@JR^V|cw8DenalA?bv@RI4Y~4nSYhYLlFVh@<H$uBDSwYBZp$#2`
z5U&TJj%W*|wlg&k@SNTHui#=7FNFsYa;&g0z7le^k~U2TpZjsVJ^-WLwya=8Ah;So
z2qWSeLQ`8I#_I4}IuIX#%WnNwa6vAkoI3~?f9~+CZ92Q5=8I?h09<zKzk-WOYxXg?
zXeAujQA}MkE^jlK_>9T5;bONg{MS6rZNzi8-79!PT%V7DQvny+u5BJ;dKz<LG&F@I
zXBN3acDoE(0T}adyWzNO1bHtQYZ#Lt_od?cZV}tMi6B<Hzl%~+P6;gsO3jW)qg#{|
ze}wIG<$4J#L{=aQj5Gbe1|lmg0GUvF9}2I5szP6Xb*Z2w`bN}W<Ky{C^zpNTm)6fy
z%U!hSrPPF6D2!`6K9X%?N)5`mI_sq)2xFv95Yn1!X4L~UG&MKj4NIT%b^>n&u7h~C
zy~<qX&|-Nta03$uGF-q~yurzuo_(!ce}+kXY6i^}7IU7fto&}_1k8kqObJDi3$Y<E
zgqWtx-%PH4|5$tkN@?M^DDP}jrrA%)Z#5NWYuHFl&#C3m>R2j!U>#ex@GQXy&=Z8?
zdsM0-;u0Sn5jI0WB)7BeCI^)5&aB-nPPhq<{#C*(wPEyzXE=D4YF}i;@PgObf2x;e
ze=0L>d)Z>8kh72Od7MB)db-UP-&`dV4HZawJFJpR5AYtkQX5tfdda+|&eD~M>Foz^
z^1u64b78A_V)PP|>cdM1ZUWIwf6sj79FIA$oS_1HD<wwq$0p^vfr!7uHZ4OQ)Ei;f
z;qim&N9(uy=jBD5y=#K$w}pAVf8fHbAKmlrm7TsrkKjn$3T@N_y#P6}pKy2qsrp{t
za}LX3c%!)A4=918Lz&a`G&lB<?)KZDEp~G(1ytVo7>}LgsA+!M5l?K5fUpL)b1gCJ
zdH3LMy<$Ag#o2W(pO(UN;XKN)%5qoZI4d|tY^_-YF1#sMp*N~c<sG{re{%9rLLtsO
zI4>Q1Vb$NFePfLS52B;Scw1u#FHy1^$VpLOthSt{B|ggPkWN^p56QfJ`^b*B#qg4Q
zr|_<>aWcOUT?*jh3$eGg%l5PGAn6|aZz*pWP05HDD`<GnI}_ux9%D4)@*wCOaR##e
z>`V2k9oMVRKx%a-z>X4De>Im<Q&-HTWplaI_7Ag{1B2K}y6LCf{8`h>z|Zq!GepqD
zCe*^Hp_MSpHlY-c^L`b>v75c6({m}=(COM6yTKr?_QKw8F~}|(q`2L8^tdeMxe)KB
zpQFFgdb7cf@UjvN55_Iq5qp70#g-Qed_Vh?tv<&fY8h)ygd5D_e;VDI;qmxovv~Of
z`!I*&I+m-qZu&Wf8Od*OGf%=b9r$I@jIgS$z8`mo6_PW+c=9<q7>Q`z2&cmw)ri+c
zBOLe0=3$S*$8FK^<P*Y%B2}<R+0JKZ3$ev323;2o;_+HO>Zx#Tp{(sD-^wiUst!g@
zh||=$SMqfHcF2aB2JQb;h^@<)@vj9GmuS`n7MJ5^2qu@L)&(qokXmv9fd>!-KqNo?
zb>gR=UoM|7Kiz~p8D<q2JbivU0bStq37k%9bus`<3pjm#KmGc~IchpCbV$yHpu^B_
zOW*4KRe+^Cfc^ozzZwbt{u>=vx(@Na@Lah!P*8LrT{(v<!o8)tI31eq)F#rE=`ioV
zeg5*olL#Zf+DJx!TxEVef0Fo>#Dmg#TpE%~q9s4xg`)q`=Bwr{-D~o_{+hp6k>u)H
zC=gwR`3WTw4U)`DEuFs_{>SUQ(DsxdjeheKBjw6Lpzfbd^wJp}&xU7g96#`hksf@B
zs4%BJF#hZ3;oryS%m1fvNyR|w)9>G3`~@k135zGt%n5ye`uoq*x0nBnR9oY*K*qzh
zR;n`|;ABzqq!loUn3dsS@k|FHOoq(sf{si49RxZjVI+|&AAkRK8qR!e)USa#ZpIM_
z8ICY-Fk~oXTH_Nx!h<zTiiy;)FutRSW$k2q85Nujr6}42^ShTPM8}+&j51(y7e!&2
zv{R&om(+iM&q|AQH>&X=etUn_9Cz3^zs}irW`E(kvv@R)DFT8!hSuBrufOrrA9U##
zA}5?Z{S$8d?r$brB7^?l>Az(BD?SpPWmb~H0ckRs*31LnULNm9=)^wF_6%zza|#S|
znr15;10sPd=yzm!BD8WepV2%@YMR_MZ#ZWPL~*o#I!U6ERNl#*lgqa6uc)mD0bSBD
zCLrh$aK)?70pOW8>&%Y;tZ}c&d^tg4Ed%9>`O2B`6L?lKWyFKPvx*sI#^J*fCDX#B
zit~?4#<^Ks6k<EsT>5R&x?9)BiM&QL<g&=CmgP0bG(2pI&~%eH4j<%9D*-LSuR^J?
z#=9hc7hQsLl5IsMEs1#5p<b3rbEfsfbZ8kGGY%i*2}^ki9dytdLK{X<D1e}O&_R;c
zYwj8&hajyk`dRX9Kix<u%ak)D)J)W-&@iwiYV&bcUf8^O1=wwhOJt6SqQCNjfH;37
zVdUtG17BNFqvDnstIBfoC7N`I=>!2F*P<|glx-9i?;WGCL?K0C^<SnZ-=ZQCt>AJi
ztrV4&xUi8ixUKx0#|PdTWIPwQLMO*^rj@9o=vRU?(se;9)k+h1H}&btOXgZ9u8X6U
zg|4n^rD|QVthB;89iwwP9isC{GazqPHA@39sbSPCKW{umz68gIqLOo18$h+?%(?M@
zh83uMMl`{85Vdm*+599e85aAQ=}NGE?&fF#y`Jgj?<S~iO3-qi&Ynk)kc}}V1likP
zTaKFNlS#$#OOoZv^@K!;B)TL*x-R8uJ#I}jN|MogFtg&pnyXk>tzxt`Vwxl0dy^80
zk^#ptDIp6qWL`g-lt>Ev5z0~|O`}kM6*Dr4V-#0R?MPBGjEH<8A))6%k9UiYG%kf8
zoI&~s<927hPE(tKlGjsj2L9Z9xK6<xxC!SJuiwiH`l-+L951c8X%5BQT2ou*P${PP
zRGYka2wCS^P=KgfhY(y3&guVtJxfPP^kkGJJuAaB1*^l|c{TLzEhNLxgoO5gex96S
zPEJNyNr;Q42EOL_;C?{mKX~G!JM+~?D2~s~!km!KX{6_CX&zHJ$&7>vh28Tw=biBO
zkXbIkF@Y0)7Seu!g0cy<j&jL&Cw8nT(IJPoCG)<8>;3|VU|`(H#bD<YX`AQ<pjbzD
z=IgWwGEx-x$uL@SDp3qn!>r<eV>H)VLm0;<zjK+l9r3Z{KN>%IhZT9R03*w}tT!4F
zEi7T7?C%Z%EpLh?cMNYVvsO&Tzs;!k4ys<gQctM5d8xe1nka2yB1nn}ge{U|iVbt<
zoR&HDB~2%}qW*;u1m%au)tGg?Kqav^pW@E5Npz#Xmw@WFQQHFghUR;JHem8gir``l
zn7pU7*5!yPlc<`sVLVH}r!<8l-%AVZO3Iwa$z_)<!j_!5CoGr+y`GCBbHw;q&QiiL
znw0mG*vXkCywp~ij1;6sx~pL)gxZv_nu3IyIDN^bPtjVItF@l!lAzp&DP8p&CH}X_
zOH~!AGmZk*VW~FIGUSARoP8MG!taCW7JeyzaKb%FyS^5LW<k5A7E+XLf{ZQA9p-og
z35r(_0lW%?V}??Fp(w7=#W4$&>pH3UeT!*q6Ypb9V*M3K(R^q3sWiKd)={Z8!1Ck~
zo`e>8Xu(~#TbVY+$`S^%Rk$`{nVMIIt~!sr2kV-BSRK!!O8c~bb|=t>5!pZ*{Wm@d
zX}?w#?<;s?mZ>Hq&s{3&n!P|?wyAQB$IfnCxvIXhA-e_UVWBs`1jB%fCxMBc5_HM@
zM_r+0s^Nujc_!jo@QIM@G+4arYrcVOQ@PGnflYW1!>|D_*CFM9_~Q3qx|v_3*cz;U
z&f@3yL8TGX#Se9VTP06gY!qLo+j@AN>E$MY<rRhNxQ9|=>J9rij>`t#HpT-_nSN*u
zQ%!|XG~2FuPvy3?LQoFbFA;uu@QpY=lkqb&H0Ai&YEzrfHJ%f&6WYTNY(Q&bgz+3`
zjb^GT;hDN;zYVFhVK?QDRrlI-ND(2C>Nlaw%BJ3^1bv`?aZT4`?!4Nmio=j?s6xl^
z<vFU*%5QKME6;nEDW#X~uAS?AiEH|+%H|TcLsXeIU;XzbZN^^W)^k{3KJxbSPKXaf
zu>r9Pam#ZMR+8yjMA?t9RDIkYcR)vkOGnW3=nxh=5k88<2Ea1LGS5L+3Z`ojDQ80=
zijj-xp7TzBz)Ut&1Eig(9Tsl`HgrsiCk5*$kQv-?!n>>$7yQAPABCQa+>7CI*Gc?+
z$XE}<vjIEAQ0vig=cDNset8U4o)dTaQZ2`sqRTmNZO6^-rD{WyZSO3}9G7PuCvY+D
zc*=AF=a`j!t6rC~mO0gy&utfV>pIK0?^Uu~&#$L{*K~-qo+fYaHy&fF+CHF<1F`|M
zjZwvOfVO@YMYcd^I78P;Zo9e^lf3q3I4$(dmvi>Iq2tnSplD*`@EjD4VGe5yu9}^c
z+&0kF$e)hlVQDu|)G<1E4vJbcr^`N?PeW=QW~~6LB)A?XzCR2;F47iMoN4zXdq($C
z*tbi6=Bdi{@M>w7Y}}69@iGpeVG`8j<rdp^X_fA*g*M0fqD#;?mhP|BlH0x)QKqlQ
zOkZcoFU(|)r%kD;MG-~QqFA=!Ke|_SZi&y*LOOK$tu-h5tqLV^M0KO3Q7FX>hw&*d
z4fY)9S{(>&P9Kw3iiSqdxh_1r-G%2SeS8gn-Q15yR~^*0l&uD0iWq8rS7aKO)Pq=h
zH;dO5HH0edI}5gqB6Pl6XlhuOg)T4fi2yy{-OR=Hoh}r^%H6DObES;#ttmQ7TWgN#
zsw}^BG*@`rZuX2%C|f5<QeT8MIOUnQ4jj|m406n~x2I9%#~Oc1d33RD7km|Z&{3FJ
zl-h`{u3PZ9w%UN9u~vq4q`^CNnO|UZUM<diXu*Y{w=dZSL90~Xr?J?lYo@+w@I=Zr
zW$Nln<v6!e%XHi2o1>6Rx%bxYGIbON-677l`_mHpFF0YHml3Ws6Spei1;HGbG1x9A
zw-WvZ{s@2d0##?zz{H|1jW19yY1s|P2GG>=ieqpFPC%I&=Y*y|WOExI5zoiTO;t}H
zA6%T-mg$Z{dndb&p$)|07!2KE4_auotcoGBoQ^4b-McaDBp$mbyN;m=%-|3VtqYIY
z3!ynxyt0QX1=toM+P{ywdp;X531YQ)r08?6guW3X7fRLZw5s>)(D?pe@J?uBm+`L!
z6o<G62Di8e2f*EzP;M9|mk>-l1ec=a2P*^gjHZ{+<p&{upMuj(je`+_X(fh_@59eu
zIhUHQD_y|3Qgj*m+tN3_JS$;ogrHwSygVBae*8OKH+nAOxbR*X8z>-MlAc_`6Xm|9
zQJgMKBejWi>7LM@{>v{PKR)p!f}n3veZPG?{r=Pop@n3i?eOpKPyPWY!IYJQHVnZG
zfBbj&_358~g+ivxne~6s=eIeMMl&moC;9U)eE1JN{E-GLXNNz-(C=j$KKyU^7frB5
z*fJgnD+1iD5$1zr&bmZ)B!S8?4(>RKotz2FEWkQh1ZEuebZdRHg7l4Rjo9cL>6t_|
z(0%4;sgk}aT5$zYkAU<DN#7u7SiFxm{zlRw^v~aa=(nSbo&mjsIzHF@#j7K~ZjjXR
zH&F}U>*E7S>30Ceq5eHVx(xI;%19;*m5D>osfbtRIy}3i@fb+LTf8k_<11k+sK_C_
zM#1(x2QP_?Bz43h+<uSvWI>R)1%E$r=!FbpOybTY&P~hZt%bRezKsHgX|j+!k$+E9
z-y+9<{#}GfvV%ChL%;cZ*z-kKq-C`ANP`&Xu9zlF^XY0mG4Vkvp=!0rYV#J63!-8A
zUDmsZS*wV*fFs*ep1qCoXkw>I^s~{z#aWL|8AvYA`tq!K&{MA<Dz3Ir8<$^s7j1da
z0*D&q;S<-@H3488%aWvJhRoGl^af!M<T*?^pq06V^z;@H<9W7Q9wk4q)Xm%;R2*js
z{w2sC*Ql9L`5pP1)jtV!8FhUWz>k8wvGl0vVMjpqTbC&42S0!O@@qKGeQYTTA1%do
z3nV2IN`BS?&25ROmT_;ZyaxggU*3USJ#)sO?L8=<Ys})`|M5e51?FF&xCnLo%HMw2
zHOsis_fAs1v2%_}U-wQ@zp`_dY7zI&Qoj;snG2zsShNIZ=cA~N=nN@}*Pn_TkCNCK
zDic|~PIf6&?qdk@9SI9sWONmm?&t?>f7iF}*K87oN;3hnPr?*!2ZWr2bySOfae9TJ
z|DeDA_-*L!{FGF~@%h|>hdEFKQ%Yym@FqC{Uhm+=3`LS0q;nmW{PqqO7x^P(6iwRm
z7d_oB-oO#I6A67d5ta{A&=@>n;ukz)`6KY#-1+I-xPk^4W#K(_A-YUFkG9N7W}M=b
z(7wbCz*CMCJhbdpEseP)UbP{ngc?M_wz*UDnpX)GQ^pk6tafknd6^OdE!7|cujEMS
z%^1~$d2AdNMqgu4SF~pTDoENXU`@}%!YW9R%)o@>?Ev-+XO>n$iM^bWfu^gsKk5g+
z0)HpEE@Kx#$ok*XZ@UlS^GIwUY+@wx7=*20R*K^g1Ge$EKMZ<2#HXRyfLO=y<S`H%
znpVp`t!Y}Brtn_UQ#F&;<U@CGJ7GI5-Ue)_LrwEBu%Tm)s}0k-OJC)JN$3-uyQ}MI
zd^T{0%~R=9apk?~7JikFMa*O3N~sKU<A0%fRX@2@XGU+!b%QEx-dSBaFV6-}GDZ`R
z*-w%lN<Di3(K^pMld?@zt;bzPvGX(Skv4!4w`v1j({+jWLst>cO?o?lJ`KnQ&?3YY
zkHJ}JdT45#Gn(O?IG<%-i$>mBEu1?woGZ%&3$+)>vjT76$Yb>I7$7+^nrPCYhJV#D
za8vF+4W!x$;%RBuQ4}r)2oFI~IAGPWqo|83YqeAa;aZgV{(9<pi8he3zC%KN=Je+<
ziG2rNo~&LEvF_L?Cj6=V7#oe%lX5t#+P>A97BE}4bgdmuo;||#Jv4H1f@>wY&cCE(
zDzqkHsX698vZ@jHNb8i4NUK>k(ti}-1fU4LuMsd(v(V@hK7ts{ox1f$3kr5kKMTjO
z5$)Hpmf3ft(9SSiPhKpY#E|-sOSjB*WIkI0tRm;oyYzlofZRvZ^KmKh!)2a-M;_k6
z<h|Q{8awjh31D!(d(L;%xSj~g&$H({S;#{4>>bcft(nkr%G-RaR9f6@_J7T%CGc?t
z?cMVNNLqJ8R-&hHpra~9?fq(T#h_{)bH`f;^Xjyey1u8$4sJ&Wn~>WB&58EK*pS#!
zU20$S9a3f@J1>oUeaWV>Rb-UYveR_2O|ibSssn~eG2v0pw<Rts0;T(Ag7#?0aOOV{
znzlsQ>UQ(4$_;RLF)NB<D}N=$WUY^x73KYtnn;Z5^Zcb2z@EST+cLHNO;}s^8|lAd
zjw(5D+WUN>whL{5KpszPHBpq?oE31A;n>i(?zD>v9d_PMFl}u&8OP%KmRvW4ihTFl
z=()`~Pn<0395-Txo7AGa$jw|Fkx$vwSk4UXg!XtJuCpOLZ>L|db7_i8rYQXFc)dm=
zaRRfvp{JPT5{H@qcdUcgQ7S7t`sv@bTA~M-CXeIe>=~v>wIX`!8mK5~QEnpk?G%Qz
z_LOMiQN05nK?pm?!~Nnv3QiH-W?%32M1787$W5MV7w*3)gZe*+BL@JN!B+wkw|4Lc
zZCsP#s3w=I<Q)pP6HN%b1Ahkx^5F@;`~3TR`XWC{2dFb>IS$AU&?d$e&%xO^VWpko
zT+j@8YBc+SZ<or*Ys<oA!DhWI1D8$^k1loxL!Bar=YXiKuvD`bEU@S?i)l;Uhe7O&
zJdUq+2SrN3!E;a~BgW(;WD8?q<z5|C6ZjS-*<VXNzM?&(q|l(g(tr7RSVUj5D~MJ1
z!>g@k*SxFN(Q_oeTUqUR+&8tXm?*Hb@?+{M?$uDzs8l?z-&sh(>1?4C?B8sRvdYn$
zluK)2lOu}&BHzBx{#xxzDib!|E&7uD)a%lgoJs9D8@!F02IaI+vWAxtKA%$T=;m@|
zGL{zAq?-Po7eg$r{(t!E&tdk--@lLRVOwT6P?^mLFl*K3(FTNP_c0r-FG;tk<1KI3
z0bM)@<J??2%2i!>4gr3t%QBRHR1ZX8bLu-WLDa(HV$VWI^2B{Z4nC~4y*fG&Y-0QN
zt(G1ath6K)(H1MqVK0G%zCO+X@&afv(-s@Gigs)^x~%Ffoqq}|+gbKlIn(JxR(4@i
zS=o+WW911LAzMSfdUdm$JR?4~Z-JF8@ng+gXZd^1Nhv1`)!=t5>lcSq3j9$6HK~GJ
zx1EjNZ1Ftk)O<+@(<h)VYCMr&aaCy9(|DY_Dn4qC!;LSyDlh^`Pb(p<?_Kk|2i#OX
z>g{Zsh9Nt(c7Jru5N~ZgMqXNT7o0dVNJ%Zw)mSDTnQQ6RjKRF5v8!{`@mVK4>iLWz
zu*HoP7NgXD;|n^V<Qzi7LWhkbEM1KA=JTHu2p}I+a&|6#hQ&h?Ioa@jq3>7%USFl=
zRE0Bv*|+4aq#peaKAA6hTV)!0HiS54<sXkTNMU?UZht1Jn2b4llwlJ9^U<{)Rhr&e
zdwDy+X#NDc^<hU~>_VT?WY^QWDXgc@2XsXA!dus=BPtT87JH@Dc<QKBfs0)^pi_2@
zojT<paIVuvSQiY{oPpVjqhtFzd57<M$IWit8|M?i4f{{P8t&D|IH_Rgl9JCmNhT7T
z1<A@1M}M(nv2dNX>_y3{hgcM9hThaoeZyMBZ#I^Pd8sPSQ%4<9YjFS+5+V-z?mH2P
zp#@4c3133VH$r_Dlp!`?q2aAIz^3kMC(ZbjoV=0+wRp@pd>e--qx2mK>|P)@kqafu
z2E;n7-G+|Wjc6Xh?%u+|3eT<uQF4e-37_jntA7;cZ%B)0;T}$^noqBDLoHJH<1GcF
zr6G@j%C3Htc^KYMs_|)wV$sKdQ;v~J=q>PsrZkXRhqpi_y7e)e)Z3`AO9U_{qtZ^l
ziwYQ5hLy1H);irZG#zgA)~}(Xb!=ejyU2hy)0!`sO=Q3;zQI|@u>B-8pT54ZF)b)V
zX@ATHWVHVRU-}SnB&;8~Vz&`-AW)i<5oxF2MTDH{p_EBRzao;1Tt-E8pu!H^jc+!A
z+eXEN{_9yhQ93sLX46xSNvkz3KCpbA+7yU&EFTj^>g(LG_faw&cDU+xz;C0(soZ|o
z_Yv-JG%}lhZ>a^5PB38tCg}9%NB&9<Pk-v`xUMvZcoyw_2Xf@%+n@!x_if}tltgf@
zg`Q1+lHV!-_n?GMmQ#4&mxUz9+mU0JZKOB|Qcj<YR6G6IzDy$Z^U8JDEz}%rjl8;U
zqZR6q>SSZy&VG_HM`KJst>D|_otkA$V1EDU#^GCt$b)BFDK<JIK#zqgmxTtKz<(yQ
z{7Xcs6_~iFc~O!|$+4z4`2gh#+Frra1S(SVKNzJzVWt($Qk85nsanId6R5}UQD$mX
zcHY-^rK;*m6VR-5JNN;+D&~$h&hXuMi77o^ebSV3Kse@NyFk3%X|?AKi0@+EHnM3i
zr1LYOr2zI7Z8=o&oMg?kwUhELv_5IkSd!lEkuKa#?W@i<iZWMX9e2SM<#^RzEmqw?
zwM8`IEWlc5zRdYBmKy6S&cr6iqT9fESycERw}au`m%&#86StOD2naQok+CromzYvD
zDSy>iOOM+&5Wf3Y=$J*ZTn=aW0@4MtYr8;i4fdAyR5Ur|(430?`VJqm#gQUgE3Jbb
z<cCPk48P$#`04ITr@LQvPcOTNBU0y8u;5Mi^0@;Z(EJISUpm`)0Ob(6m#^KY_gWjr
z&l5kSH6lNre+T}ykM{_HX8`{Lc)a%nX@CBapJyJ2GB2+?&CL=lKZZv><PnVT>nzQW
z<C)GExm0D-;~y`-?sVOPR?(%O=}Z1QyzKto$qhM($`E#ce%+-3LQoh?=d|*Sz_-7;
z&%58G)RQjFOUj4U-M25Dc=NGB8Z|?zowK%q6g1;ZopVuP8Y+;|UTe!6BZi6?Du3dQ
z8W5Xwo;3h=kaT6m%^Ozya11c1m3S^%snsyPHWqExo;8!LmW0NFlQbVxG$U!6j|4B&
zmK{kxtu9jZ@$ug2bP(*m|7Vi+s?qYX(9cDave=zJ@*bNv9|u<SNZO!~T~w{Gdx6?z
zcU|-qIz>L*+E=aC<STLo-GrD;zkhx88=aA~e%fcjE5gxUW_(usK?$b*Bk!YmGw_Eo
zpvc9M2AZ_T7^}P)+0%P(iC!dS(2VfXBd?rRkW-OQjMROosD^?`#^z~oHW%0w*vp(S
zg$788P>YkYKx#zg;|MeO4#F`gO(TCey)N3=Zx-U&yV+-bL>J8aXg?JE@PAMeH|c(!
zKJ>p%<909}{Ap%d=k#2%uWuiYHq4bU%jbWPtI$7!PF3cV_81ppvygmJYqQErb<sLb
zg!TkRs&N!kT}v`jaMDvwqP5c58onvOa@tJ??NkR-_~p@1XTrP<fs0ib*`PmMSY=cs
zlB+mr^j7x)fb}s2qcS7gS$`8vP8U6LRFjcuc63#`Dr2cERFt0UsuONmI-JHUJ_B4E
zq|+{__bT@jsk)#Z&TW03JC%ajg(GA+sLiR+q)sLs-)x;OXbp<y!_k>8@*y>j+Vt*K
z)UJ{C?w`A|gu|3mlgH3j1dECnP~@02WdZaH{G<+UWK6(T$4_^&W`B8JLd(qZmd{da
zm+6(*NhJXlu*r>*-?`J9-wmfd7sYOR0H%+*(ZGtN$Ed{T91KPW<p5e;|IV6cfb(m6
zI^O+!cW=DEWF&W;kw#Y5)3~^&MeCKfAbVQiQiQkZX{`y002cH~mwF%}aonjauFT+^
z0`|R-EUNPPh!gj1Ab)4$RL7J%1$N(|+O)p!q{y|F&(3Z8PN8+>9YILD;*O9zf>(E+
zLf`13xI|*<P^HBE3W=OPIG0L^cDjG7eXTqN>-skeq^!4R-K9YGexTa>=I?TwsD8oP
zu#g*;`qJg$?D;6y`@A-m{&R%kl6_dkom)qqivpZ)eQlq)T7SATitm*p>2^Z;264`j
z=Wib7jncJI-U$0aL;!Jk`2cT71W+!uxM6`45>YQNXxx4Qjci+S%>-^F978N5X6_Bs
zvm(3-ju^R})O1nudXFI&@EF{UV@qh=h`JF{uZXgZ$PDgJ&1h3=xeP~&9X^t}qF#HW
ztT7{)4%5}OGJiB;_A3I=1it@_y|bSsWk$YStQNIVsJ8?VQ_R2!K$vcg5hC9!MgZo_
z{=6!WV>%kff2Xw9vweX=8jKIJOTS`2O*G|Mv;o<Krp?9cn1Z3StC#{bPN9oSn1V*f
zbxgq$vkFtt*mWIK&?vc@DQH$*#S{pSEno^7T~{#$&3}r@6zHp%0%7HAnF8ROg$>~c
zG6kqDpP{f}BPAqm$P@_47nlMu`$kMbs(t9;#%+$!y7CSoq`f&)AoOj(6iA8dnF1+s
zo+*&0U>#Fn9ODG$f6o-~?UJ)Dz7bPkFe-|4Bc{NBSJwMm$9b!DZJf8#eh^t;wBzy}
z-jFPSKw!#mSm7FZD~N(d&IU9=BX2{J06Z#Ba08M6_<|b@-Z-j+){RI4A@$ZIfl$8@
zNg$=JCkZ4>^CW?o{d$r>s+}hZ#O!jles#u6=Po!}cdZ7!^jdwzby|M&rR6{K-Yk~^
z*%uR+@{$M@myKi#Dt}r_ZzQ=5zW1-FqkU*u)LXlT5x;tX+-BfYCUaN}V&^auWD^9*
z&kyyWUR7OPZ5v(#?e0>Er1%i|u~d9|nfUZ+clWfrI!HBH<}_F_Jv~o=H#q(Wj)O)|
z7JwNE)6>uC>ow=X;O7oMigPLPWAV4cZ+)|u;BW@u6~N8jDt`#`5BRyq`;g{Ex-vI%
z(D)H}=QiGv{yNX%_%S%s*u+<5-umX-)1P-dt8jJDZgNa}Yvj`Uja3X+N$Z!meuZ36
zC|9^H2;VrSaRnMLv`ypjx69+<JFG;7b!d+>a4lM=nc=qzzYm$SiR*iO+Ko%mMlh?V
z?>~3HPq_Y1Q-4%q$C>m_Pye0bFax2LpExrB*5J37_&5CvCtm&Y<@rzFe*=bO{|;06
zq?pj)-SjhEde4-#Mod4_HE>`~Xm2K99(7kZBAFH5Kq0Lo->01tmVwl+9sUu9k+D-k
zwDb~~bT&>UDRr!@Brq+bB=bi*NK30iqGn|!;VN*+w143e>4_>_Mpc_@nu=6^pLWmF
z?sK*+dccM$jwi<6;NOyMfhn*OTKm$FlME~fNp|rTL*dA1t8W<;%s8VYYUr57u$EJi
zR;`~$8og-<c@YO1qVIA1h#dx9aim15#*GA%$B;BG1S(|fCJWc!-+yMh7o_^sd?!A8
zv;<SyOn-Z$(J&Bs#bkq<bt&28=`@>?P2Pyghm9Dw6-lD)-)^&k35xtYHA_mm?W=O1
zw;P0c0UPW=tD4(ETXc?vVuvG1msG&bp>#>QA<3${oR~r)0v!@jOl3FwkY-~Fqtq2s
zS()W%CefPfX2iCm(a!ot1dfplPpKBe$hojB<A2_fI;NwA(JxhVAqCSIf}<upC(o==
zNc>u+Yw~uy=W>G0$yOy@+r&?jBxWCN>?kReASGmoBIKj*i;kH)aSrwgsXIvJ2bv`w
z(w$g@WTAC8V1sS$@~TBzWwm0oIC(3r_v+wa$=6B%18tWq6wBd#re2N|C2<h^NFDsD
z?|*Z7(`kk3Kr4&-R&Ohc6S|5e?0z=SlcsqG4zUWDwlVk93aa*dVfCAR9z2QP4@r;2
zF;JHokI5(T`j#lr_dEiO+Z%1KX*KG)#&?VXD*QZK%;kJz%j~>%^;ltnx-IJ%_GKRB
z#a>wwaSQQdO8nf!n`Cm6*;rze#1*FZ4SyEVWlmNg1$7zC0iUl_zHe7maLdj`=%Q5#
zFA9aY>O)@9ZO>OGcO070oTNf<!b46e7%?xg0*7SYU13W9eYOy79W#iFFn@*{s1p+$
zZyUr%Si%1`J?}n`Tyvtp{2Y?Bj+vUL)YXlA$w1h1+-ckxtkGRgHcTl*Z%$zqD1ULT
z@-FN1zyM9-&YyGCh6&^}aC~EK+#3$f9yeUfv&j(y9j~QL*;=fnR2Nez{m0bL%D8u5
zf$xTfLo0=utpIkRFjTc_Yt}2EyU_Y1bf;NmH>WFAer=Yew#(8@9nRS+WubOP;SOmV
z=N9AYU~PaVMj-BOEGex)Xl*)6UVnVOOe6@&7fC!8Bh{1qxZH!k_!a*fd-0#kRgqp`
zybSc;F4roU^cHskjQ%bON#$f1tx44>);H05gsK>ME6*|6V!X9>SDQL7Mc%HP`#?Lb
z;vvKbU6Q1TMXck{u17S;3Gu#5fX5sG5_?%KI-VE*Bj4#o5`{SiWKSY4F@M3^i|19Z
zfkE)V2THy!%!JtLb-s8(VM3G;J0i?sh{%C*Ju1Oz%<{wf&WK9veK5%n^xsxRVo_nj
z_7uNliX*)XMFC(omnChvROVw4<T8=B{;J04OV3Tq?3DMhm`H@bEyXtHTaHQDb%ZB;
zZ#A>hcNE)P?}6E#?klW&-hX>+A9|8sS7l(F;NW0~<m>HVBfC;5fpI4a*ocoR1cD)E
z(f3=x2H}eeHfZDo*Z^#$FpnzCrxH{F8xOTnp<*In8`w~s8QgXQM*^v(9Xe!ypu#%o
zUtE4S(zJLX6#%+#;4JuJn+8@zBr(jEvvuGA?i(Ho{9|6zX_KC`r+>yKMY=QWiB_}7
z9>UWULXF+Xmkl=Wr3!fx7A#xNA<UX_3s%xg);kas)OW4@R`KsK`(c4*%J`Z6Qg5;)
z{)ALaNHKC4pTlkgS3&Eu<5XZ3K(IWnAU;n<2_TmrD}=~_*}D1p`$S@cED`98^7XoW
ztU%n&A~);{PGDZksek^9Yv@MWy-NE)Ii@fbQx>=zVty&5oqEEKoi?*#^#S|Py-=SX
z8ricRs-Nl|WO%qIRM$ILeyn#`S(UJ`1;fW#wfhI^7n00P)DiXHH4g#qDyJnZ+sRth
zfFHrPsTHRgs~t#^a()M*q+!PU&L1bRo{|{n#`Gj-U_GU=$A28@*E<lUEi>AWQIk3f
zdJ4qHKBM@i>qk~Rw%9XBOzP73rKe`8U#OnT9&_E@h3@23XJNttgz6-p!j?ESK=zJg
z8@R+O?}X_oPoPWe2OYOP$tHLyK%#S9Pq`j#0x&?upxGJGCc<gXNdIkBzfcGRwC_m%
z3R00EwMeBT3xCIG9o1jfb<?2&4UYchbx2rp&5ZgZk+9U*bm;d)!UM;5M8X5-6G(WN
z{Xs}r3f!62elQXqq`nae58OT!36J(W9|;c}W=QyGQGXr!4bM>A%x2(L(I;W%(bBI#
zzuYjb?M!|^-*O!R3(2&dd;4t&SOPF#kC@*D0ShZ||9`g*jgNa=J=UR0p3tt$=LM>L
zn(q?g$TP<C*@TiqfcrAX-avdaCY}w+3=dR#S7sAj7CID{9xgk7LIvS@Q6V3_*$PI)
zbycPEN~+KZsgXIM8d?u(2$$8!lM2R;3{)A{Czzk^D^i*ezC1myoOniFA|tv^1|4i=
zIV3gRUw^{%Y2_`FOq7;R7;mJe)55A{;s;e`=v?OLugsj`{ZYSkbl0xCV$XbvWfl>u
zP=NTXAXe%d7LNYPbMNTGxXG6};s$rt%-(Wv`Mh43_B1tnKngv)sSxL@s&7f22jWwF
zE9-;UM#62Ye?7A$x-zecwLC)mF<qVRVe`y1Ykw}OzQ+-DQO!YXS*_BI&9k&dmkseA
zPsd}L5c!BX$)CqUYoUCiUEZOfog;-M8@jSadhqRA{i0rCITAur7WJ6fBpy+wsBgNY
zi0aQ(X7K@T*rAzf{%%W&*Fx2L7`lv`u4E&N(jr-!4!rZjtH#=|Lf%#~rAFKx&Z4z2
zAAbV8!Fh(o`yGMz!0{b{_rUoC@E&G=5bzEMHA?so2Hu0zHv;d0+ZN!B_I*4bv}Uk9
z$jh*%+UFdO*ldZJ7wC1AS8)u_Q{0TBSq9V@{Etz<$yy@6sYbc|4G<TUW~w4>lOK1w
zrVQ}5RFbo#Z(fVC&1)@>`QA!lk-ENC7k?_l%&_OEp&20105q9~`f6JYrPMhn{&w!6
zyU=3BUB3@^5ZErr8F#b5{q3p6hrD1uIJ5LZLt^R9a43d!j+N5YE+-ggnUbR>AJo{V
zeLNWPjJ{q8W)%z!LfaFXr0^X=mMY2{_O<oQ))!?$JoS5Go_Hyq+a%M+VvNUh$2x`n
zuf8_+)>k%11B5Ihe0^^)!?>_mz+tc{aqvc^^l+e6s~(!QD#a9+N29X$8=Crsx8ME`
zJ}Oe|m%&#869P9mmx1a%6PKD}9w>j+8p)0<HS9iLQ9tlPoH&~_YH7OP3=%h<=7u;S
z;KTv`|A3P?sVvpi+c3bPddcG5j_s6OeqZGB{nPRN>AN?Lixb{}Q_K5r3($n5PmnZd
z&Ef!r)v&z(vHbKbrLy!K=|L&ArpM85Pv7S7MT4gmK>q*^UmQYQ|3=S=<{^KtOZ)2D
z1i{dwXyzr&=<v*|Bt4c^T9@G>!?uT?-+y?L%)&^obts6cLa#m@Kr8xT2!-(xv7>)T
zY#;$NP)JGhgwqg1c;f$%SK(WJ%&D%{yx+zr5%k@ghb=)WsRV$Y2(y{(_zlPqwpF2&
z7Bn_OKrGxC5*_+XH^$$`$VY#rHSny>+$GMc3KDN|-|InR{O_}PMX*uA@$38H-|PF+
zpR?RG3h-|E<&UR0ff|JNYEd*B%<|W-%WqFVj!;>JQb^j8Ka5=d`h6KjzOkhSN#c>#
z8q--SNRk;Y)__2zs(8Leg*Ob6x)gM_BZYN@fs_LgS3=y45MoyAj1_-rtu`Z8rCHEy
zKBfrWuqmB*+-+O2cLMdqi;N1D87ewfC-EiU4qvQ{erArWwjwiGIqF6dN855#Dv03x
zB*#Q?N9{S8C{%*neuU7QdgeyIGQAiHD<6s8Zsgl48)dzqXM_$q#v6LfrbLw|P3nmz
zlHASseGs#mRrG!ny(WL}*HV7~dL7Li)LH!7VNu7};U+VUT9BvsFdarO>qnxa8)@jM
z?nVxPR#zpzMkCDMQOOBIW1c)?+8Nc(*gS)hXS~9)CLRST+PWOubQ7ypb*StX$<YEk
zo6TcUKBEyD!l?qsbyeUWBC$ZjK;Wz_w%S@*%hsy=mZeoCMbLjk)Y%Gxm(F1;QK%4V
zoQV-FP!Kmsyet@u2b_sC)F@%ugKc}P)qO0eBs7kZXgFs2bhH(CY1Ki&bHO5?+Y#P_
zt0K@9Ecu`<BMI7(M?--I<v>gG>3EGXuc5pg+4xqe&j%UUhTp<T!uOK3qD|;z)Jb5H
z+}Xo)mDqHcG9G^`pVx$;K+cIrfmN{p&#aI=^O|KBEoQU<6DXJ<Rwpj0N~;B?YfV?&
z8pBJv(DIy<fU8a{=CbbS6R0BFgjv_*W5Q(^S6UZ}*6J*~dQ3-5v9-(tR!oL>xz_FL
z>?O<3t4>|=w@;-X)510ljM6U&W(hbsElhcEPV9_dyfA;k#dtaUwy`L@4()lFJfPvC
zT0W+{nXRs~`r>nWg>7VH%V?r8)HM|DG8?CYuuSn`dMPLwv;wi70Wo50<u0+HRu&Ff
zwW|_WLyoK{+p$viRyI^oUdMrA4w#l~1Z6y#kt!-a2iLVO8bdjqO2A`SoEaOlD1j{R
zycky5l{A0MSUKd>kat`u?`Y+8FuQ9fi<dZ4xo;g+t?pBIbX?N+*IpNKF43w&AIdia
z6cqvJeVaNa&T81Hw1M9`RKQ(D(HzK%Hrp^#0_s``Rn9kqUJajqL$A9*qlV$OmYi(6
zKbSZ)++1hrU@4EE>MT%J7-c`D&@uqRJAFf;b*X=d8X(T4&du2}SS!G)5weooWfxV}
z*1@=1Z<SofoH@g5TX5})^pZT$c@aoapIf*NSYzS3AGgj(j-aRs+BF61eMO+6I>v?(
zA9qkz`P>502&X13z}E*L=0bXexLu(s0ER$$zlxbktH}!FL)95lvE1~EVkr`1X-(8U
zr&_+YBUmqg`b>fw>m_~RT+<~D3AuQhWOxf7Mt~$e8!};6Pn5pGd>lIEFQ6vi=3{_s
z`nE7Zw;RwuAc%k#3iW3Pksj*@b-xCvpW6Kz<#>DdYk<%-9?ZktuK{{;?3gI&-R{>Y
z(Yw1}BYNCLU%Fq9qc7dB5x?2T&)u(4sgHNRkT_F+9{-o!FLQhMt6Evt{lalnbt?J5
z?w7W0pw?}<F7c|2JFe7C$zB^sfdgdQwn{7G$(C9u8N-b8T%#z@a^(Qe8#ga$MXgKo
zdRgb!dMMLz;#u9C#%`^}R-Cmy=_&5;)OygiyyTG`Uax5s=9Q&FF1))IYO38bM$_Rx
zQ;UIr+^jRTK{{Hi_w7_{#SiR;jkLD0R;X&HDRf<3_XSndn+mywwcfY6tk;aIv_M(2
zMUuJ2mirc^=djbbV?9($exRkd!LM$545f>+)&s*`Xs&_T7B=epDIcs_r6>(1P_#Bw
zeX&xHa{E39G@Y^(B)Y}vTKdXbzCX&1-wZo{NHu}ibbitK5l{Jq<>k!|yUzqIvQ8>d
z`ppzecdv<o5?yCTlv@e!S+p*Y7~KeAXxW^_AFMzs<j&!@7x62~do@jse6L+-Y>lI9
z-kc`0<-UIDuGX%qZm#Ct?6@u$TCD-rm9;IH@i&H9Ag|6l@2vfN-<f#V2X|AG>5685
z)-499C_}b@_Zeir`C5Y<c5M_%mke@%aHBz<!+*U&j(mT<LE`_`AjcT3tPv{sD}!{u
zI=uwc;>Ld_YR5t>czvgnbepesl7816QG0-JtCO6;Z?1Qef$#rT)W(0xNrs!)`yJ-e
z9bc3)u8pkt=37dqd~<8~(%e|qtGLpC!KsNgayp4sR=SimOzCkg8oEoW>S|o}D&xJ1
zr8YfYm(q6{{dT!NgWv?!tXn;gSfZ*^`{LdCv~;~><9o~Pma`MejVMd5srtL5@s<pl
zd%kR{u6ZRO@98h%H*14fV|NBG>$+<c3%iPiS+UjBbm=EtP;jB9+%1Cd-WW`O^mFgG
zzB152;M=K5%_&&br2C}}(cSI(`f6f($qIfVsYHghBx(F~74!ZpmPmsGb>_Gq_hb;p
zxnWbOP;R}I10B-aIse|<k6^4Y8b9z-Zh%PgObs68Uv9sRGe9m2BS<ChzGI<WgwFaA
zAUpuj#+Pd+IX-Z)1hHf5!vyhZ{);6tVPQ~D5Vv1Jya}?WEI_)Cq&4=7?WFSIEzNsc
zuQAOtZoO8MeJ2vW_V4AT|J!Zi`+AW;kplrXGPUVVrAyIoY)kFuO`9#N4~d^9<|cOJ
z%$)WksHI#%>T%a-GyF|q^KW!1$d_RhC=&rPm(iLz6#+4q!CM3<mw5;+3x5G(9)}#h
z0AqpOwKqVz1nz=VLGG$6a{s@`@FnpfX&;hZ$m<o!Z-(>YaCZ3cZQ#RiPp9{%FBge}
zVOD_=!~53(Xn@mS;Pg_f!2mEVVR-*O{Pn`Q(Da<>LC&S5$I$<lKGoq_f~6Ti-vAEJ
z2Eot2&~v77i1Wg_ac)3R^nVB%IffDGUeYX1kEWU0L|jbR=J5CX?@v5!fitK5zy72B
z_V)ht-&4CGWfVhhhJU_4`2m!`q!l!bOmjQ@_;>jF^hXOj663Tc{!x7R@oi|{{F!9x
zILVZ2tyCpb;3Q}qw02DTp(s;)<y_IukYt4<E2NzYsCN0~Vqq2$Ie%$KwTS9`&|JKb
z#E%h%c05df0^ije;(LTtz=VMoQt*nQX|`j=cJxmh=o@+ix`ia!afC+occ5Cyz_<=o
zHHb<M)1QD?y(tm1GL>bLtRy3n8564<vqI@prSFOclsP<W?p;H=G)w!p^pF!MnyTJN
z+Agm4@4Pwqad6)FIe(|=99}o@tr&2U1R93YjC^C1G05%6QZpo@$jCAwEAPPD(S>3P
zT!+2`-+)Z%k7AokiZ9{=#I2lQ8s%ityYj6p#ea(<x>?|RE?izaTNGNR^?0^;Bgd0>
z3X1y5_%|{|#N;NyiH`A|(s`yo9dhjSE*55cr%~f5My&9tgMaYur^G>q&r@2-CszyK
zy*onYBvKslbRLg<1ou@bil^^^3p1`zAu8v6B_6_l5>Fz3*JZFMb`?WuIte<NN+X%D
zvXkj4t(w<_v=qd^wyT01An?CR)A`WnI-2KIn#dJ%Qq9uTqr6KAhd_A%9rr^sbUA}Z
z>5g)$ba;3UI)8=|I|Nm?x-;cNfiW}Etj@KiEh>Q&0ri7o9o>WCsNs6)P+#;e;@wdk
zYnDSgJnzkyfMB$AZN4O}qln=?C0|ldh5Q}zr3O*L{b>q12H5eom?jyfDZsWnxlF!}
z$0-8xQ*CR{6eDOlKPNWt&r%ex&bDI{y|1h3nvT%=Q-4V8sG|na`+4x(99u8<>r86~
zTC~VZIx}=5phUQtEjvD<zBug?-V#qc$dl~L=zH!6BsD|uAP4Y-0@doxpEt<7rYdBL
zz8KH9OgY5cV7m*Fv&@MOEm-xO^zPTB0O2D|Ljd^0_X7cj3Fue`d>oPTR3kb?!e!Cm
zgtQE;X@6IwBm6GfsF|CQPfz<URDah~u0vJxyV1xfP3(-|@fe0AA6uM<r~QK%XlAYZ
zY^STKZgsjEZN=%*Jye~yX(gm{P42g+Ql#~UBtBUw4b_cGY0&nMgbji2dUj9T0*2ak
zh-+~RnUJDcAvuXlzC94T1TzR7{SA>gMZ*^Hb$@<0zW&u|%-1iG-@lm@1g{J8E>uIA
zJ>5>$MKsQ)8$9vf+2G?Y@9&y)Z{4@#C0D&G)``Pfw_6EJYQea}+F)aC2Kjb$G9-|y
zLs$EtQ=9^$J9NDdx^R>ZnGW6TgASHl%y#H@A9T973g<d>w+p&(nn{EX{WycZ{p=HT
z2Y*2!Kl$uysD`soByW7JJvdZZ-17OC%I*!~^`u1RYTPNm728u4p(Pz{_<^~KKwt{>
zXX~q`>h-l$p$t*=JXIH!p4Xf6RbSg8U(H*xsQ9>->dlG-^L43}V2=8^>U(1iA=O(*
z6iivO;VPAD1`QGkV7YD(bH0|gj<(?Cw|`LI@qr|m$peW7I0{Q0n|vY15ut9_%pQYT
z#lr@b`VyJwl;98Wc<|q(O5qiN9msh*d%nJdu1D^y>=s2VvTqGT=!CE(o%Oy+A)-ZG
zP_?7$nP^oV^4SzgJD(v_(!#b;@)4h@k#6yshIZ>S=MV9jn&>s3sb+te&(u6U=zlW}
z$c{eKB7cm})KvEHnFjHS&(yQme1`iUdfbBD1|#7Q_>q@`#1?(VkM5EMr{uz!7IoqF
zn0H7km?_fU&8C!tL<M};8J}WC4Ip)FC-R^EsufG~+qq4t7FfD^tEm`et6%i4rzMg}
zXIm70-rR)(hSDcjC)lT)HuJaiUw?bWP098+J~chqZiUJ#M69%5UtwlfwAZ~J5f>3%
z$#t<pbu*78wy3x*uM;cW;g$WWhoq!#l2jF9{JCSuPc$+sapBTDBa>B%Zh=qpkR%tR
zHe+|{%DA4j%LPfzjN2>Y?#lSmZn3l*n%V4lRu*a;nbjx70&CQK0jIbMH-CGdrDPXJ
z+~UYFj*Ppt0>}N8tuUt;PeYrEiOrm2Yl=;m%2c|`Ejr%rVjMoe%M^-lX>9X>D~mPu
z8I@Im(Womo_9B~^Yi8%#*|)BGJF6}f%t77emsSJ;pSv*r<$^F!B4La#b(OUOy4P!{
z9Qx;9e~B#}zN27niC`qEdVk4JdC5AJ!pv@^QZBiXN+F$Fsgz41m22H(1mqYB;eP|{
z+hu@DCZ<m1QPULUed6{Ko0Z1~Gqf*_<(`W+GLzc1FgKS4gVhTyVqsQn<SI&KV<~*S
zT_>?`fsK`^k<r|B7CN8<Q?{<0kCpO^pB1{Bjzuq3SQlz1cIu7Gb$_6$OD$5!{bzO0
znZG%5X-OecX{&vO%5jkrlev-V#}{Ch)Ld>)6jGjJDD?WiPAwj7dTt!!)>V*Wchd$=
zEs{Zsn$s-mV-iRax}w-j+qf77nagjsX!gqui(yTpB3y5@v$SK%?&=*^LR#y#Gn5`r
zOF^E*>W`KPlu*;PXJ|&g!3S52nBtQZ%Vvx>nQ@#|nsIzn&OO!*e_=P5%4f?0s3}&e
ziTY^#ShZ+|Z_^nVTL|+$g1wIMLz)8!TeS97{ik7?sT=>&LV-nq=tPuzj7H|g)bpC(
zLv!CwpKtF^{{vh32%wiCTnQAHA$B1X0Wp`bG(0GmUW674f9M>-<qY3V)`0EA3lwO9
z7P%EYEt+5pY=Q2n{r4S8e2pYa3hMTt$dZQh;_#bCWcSv|?$hq{viopRx^u!CaM-=P
zc0eB_{RK%ct?3+qunM}DuiY<?Qlh2nM3<IQDY_i}_H;Lgdj*~-fSv#j_fA8kztDB2
z?=aBy>k4hpf5Fg&^yS!pQQ<LBNxCdiS~u|3iMK!e`ttowCKEnfjPJZK!C5t@zja!W
zmRbKOO%oSG+zj0{>1NV9?MRxAvJ45=M4`!A{hWRu(@NXRU-O$aMp#|^7X748NuJI2
z8#idt-!W-&PVz31%0()3!05}V-pzYksIw%W-{>%Kf7h4c>Lnc$4~Ui(eVE+`u#aQg
z$^TD-|3$EI3*zn5=;`@o_vdcx5aUq<(B0>+yZAvX5Xxg`rEuiIZ@+b~yPw8f(;iD9
zX^AlL5B>V~){Wl$oI*KB+7fBW_iRf*(mv_VDkUg@^Ol%oP~!$3SRQ7%DNBmOm>WI~
z9Q&|Uf5}DbYOm^yNFPi1_PWYpqjJC)w@q&?=^&diM@&H(k6-+)?Co?X=m6<&sXONw
z!KLh;A3`;9MV*2@g{?N}XzAP8CZzx!ZfP6m7@@HoR##x>7P=cB1O&LD?jw#p$m+B&
z8QSC=Y-|K^ql3^23xA(@+a#xjt#g7vo$d_Se<KvOl=@Ovm~){@elPr8qKbAvf`8I@
z0sala8=a+gH+#S05RJWD-+Lv?bQNNw+?)#fU73@=oPG=-rJo}Hcrd+?niF)JKg!#j
z6+%)5SQr!`s05hMX|ILS*~pVL^VAA^i09OFNK3}DNt#QhS?mG@2u1@M(<q|?1zXE!
zfB(bXu*A99CDsc|5zm%<OiRd1b4whBvg0dWmxT~17~o6bZqbx@UKDr9{^9N(tlfVX
zZ;svU@XhVNE~<`52mpLr_HTDf;ezOdrlUC8n?6iKi3(c}icUwbg`{iIVYSc*Tcd+f
z|BL8o`0RgNi4H45P<+W+bT~p<Zrf^je;~OS_>#5opj{dHWG+0)Qwonimcs)KA?!$z
zweX-C$NKT+@bEA#>Uvh*9&;%=3pP5_t)lWkmC+c=8<Y9vr(7EM2<|o^wWfLwu+~F4
zXJo4^!`|E=gj#Ksg1yD{l_nv8K=S8%Wy40<iy$Al8r6Eac<|gDZe&f$v3<kxf1)4K
z25VnNt->{ch&!OvR)ApC5=~S4C2!PNS@zK~kOVELJZ{D9|IP<Xbp#nsd0qvDAWY0M
zk~#x5RHysjmk$<^qV{8>ia{yij~(4~dF|11?3iskR7);Uq?L_G(fVN7AO)IE)2ux;
z9H$2~Zx^<j_aTUrB6J)+icm6?e`a7Ud^q)g9zLzO{a=gWLzW6fpOm*k!%(yBzdWo2
z5Y>06CuZF=fM(PFO5g+`J>Q!BmzBUVRJ6ERLpYGx3Wz=6KvUdTpu!u(KdlmIEjJ_Y
z$8t7NCI}0)@Xq!AQ$_Yt&4(tKuH*?rRlHLZuI-4&*~g}@xGpDA7RL_jf3nMlC)q^M
zIX@rnWBxEtOx{!jS9{emz8WOj^u?wlUo{TpDHnMr>D!K`Tn>R@fYtJyFH463qzPee
z@HV3p`(pTDt(W_pDa>Qj(H>J)u6~7yM{Dkk`#gyjO+ahv6d`NXcH=N(-Ev-45C~Ar
zR$ye*62rk^w&!$E#SgA+e+Srp76>0|!ns;3<53+qzN-R+ewGX{>qMktO4b_coNFaE
z49!toA;k~lO>;`{BAcp+G?$ktkG8g*dcoDxwFE8*E_UXP)}?{Ck{=L-0pB4tjmBAy
zM)N%fVi4LZCgkJYJvz72d7#!Jc((3gxv4Nf5XsH)%OVUlH!83de=E7pq5#0rM#J)K
zA}(@n>uiEk_y6UXZY~e;MaWu6pe%?&1)DS^o1NnO%{g%OAmPG6Qq<*n{2xw|zjBHe
z<3`4~VM^{lFCQEL;iyJJ+T`Hf#=?r?$Ku!*GaWD6`s5RiZtt(3;t>ZIxRJ~C69yw_
z1IG&I$ihivO(SGbe|~FkKnP*fXdhO31HuR=CqCsR-ZVDizYC2Y`O!6#{yGP6wO=l?
z-#j1od|FU*a?i)J&U&IllfQc{VcYZ<!3tAn`2|j(sW{S%oYf6>G<CmH6()!S?**Ez
zd7e}E-YsEpV!7QnTBd3wdITPGci>X3SSjZxGLv({p^CL!N)bg|gCeDZy=1Bk!2{^M
zqs+gSBI0y-^Z0=gH{;3o&aPN{d(}@2^kKqjiYl2?{?EvzPGOpKGoAKW)`F|qv46fl
zzwG`2e@K;=@kKuq0x~$4P?HxG0XLUX)CMPiT1#)_HWI%3SLkSCVTMIk@sV*1*ll}&
z011#;9~MY*N-~2ib`G0eAen!kB1KXprIys)2BzK966?YGiuID@@M)04`>V&-tFK-(
z4puk<tA^K)1Mm&We?js^LxTknM#J#>dHC^8N@e)-$RCtaYyMdNzvFMZzt!M)1@JF_
zfcsmE5Z1r&=fL-2To>ue+Q@<UqxjB4yrcbHS|#~oc%`u;UrfCH{m-x8UdgnDuU^Ql
z>2E9w)-vlKxCVWKURpDE47|}xNvs30F8ogt%Ryd&ER%SLcu4ml%HFb>Arc0Q&4P`P
zZH6xIn{<$MuhiYh7b(eeL4IC^CTReF;flRW>k5}i>w7SFM>G$=ZJ`fWP5wnQy)wth
z;U#f@Nc(35@u48s*pX3rN|pf{yzoXDtSzGy3dKMfFhV(>Efb{&d70Yi!3b})qn0{c
zWV?bC8b-Tb|Ml+6O(JrOB1sB%n!zX%mh?#UubWjlq=YqgFcJkZ#F6CpcmK|RL~2ov
zF+v8=tF2xCvM$H+KCEF96xQ#RV?FP~w{rekm*YqvZ~cfIYZv8sE7%cQIsaHSwPwGU
zpiVKh1|>9grd~IIR+<g1Tg5)EOHo$vzMK7T2+>M_ZXNad%eoNFyI1-YDN6IM2?laj
zSZzj<|9$=cuXlMgF+xhhK944UHnShdlJHLL(k=OOquYj`S%|Jlj*W(Dr=K}3y!5S{
zZ*Fdtwb!t&q4xr%X(``jlEwI=2?RLZlDKtGAR{>Qv?RZ;OEF4#uq}yNmty%GsE*{j
zNK&?ojxl824G3lT0YaSgkhv1a3<t9SL&tyzZ(BJ(aW38nOTkQ?c3(Ds!V(DbuGMbC
zDLIl^Z(1okA+JexLSB>XguEu%33*NO-OVj&-q9@>d%cd6_9(R1r-($Ug_Ei!`F`2U
z&<GBKCwUp7aH+)FkuJ5sPE6dk+434To8{orNZat}kTRuhg{d*beB4NaytUw>HKp8)
zO49m%XKPY4!$Q4&unxX|bOK$c*r#=W1Gczb)@$SF)FeBPPEE4&=(OnUIyyDUIY(!|
z$Z~XQlJ9T!v;hQOIMlh=jfM1MD)PDb5A9Jby~*{(xRH0DA9F)u!ribgv_|1q<S<<c
z1+*Wj#-?FxRWjwsC6xZTjE~n$rjOk6>PEMERnZS?*|ELA)T%sx);h!%;!ts)Xby>Q
zTwC@NcIvB!CGk(K&Eq}XO`@VQp+7m)Mu;6sVkcktj+MFRftqP=>Pd;(c(`oxaeX%V
zsANS2^1hC$#TCLgAKUB-TfmIn8tWH!-GifQaN15TgEMxSflf8u<Ps?3-3-b|c?^`P
zOH57@_br?L5(<}pEJckSZ<adeR$S;X=@@NsxCac|h1__rH~Ywyg}24bTBj{6wb*0h
zZ>v@rcQyCO35-%H9h-ah(9r`Y+>RZnf>(XS`dNTYU3|6&E!ib`_{i;{;cY6<vB!J;
z?0ciH*GLxMapO#3oH|4;JO3DwEZx84@EOK3;p=_(>-qJ6>c6vNf!qffKm79fDqKem
zLbGML_3~T_|MA=K@#=@3X+62(+L2o?W2h+<03@S_18<ILOaW?nzoXvBq~*^apN8(r
zhxe1}n)3h)?zkrh&FYoayvk_mSNHalte1Q|U6)X?bK2^4F=<+6mvypEcs0{W9+ogH
zw4a!A`d&nTQ@eo13{Gy_EkK4dbI6<-t0>O#{}&vDzsak>^2eX$@GIZ^mKSR-&Hju_
zKNm4__|Ndq;4d4DUF>dAA{!Ie7|Vg{6Xz4#i3MpN2P7x6*ryrSCjyby*|Lj!>iMpo
z2`>D(y4Rk|G_BUpMDXS`wZ!DB&*P*o#-}uu4Dl#`&ThkG_aUZd`B=Hpw8zF-vh~j2
z=9c;Rl#sau*ppf8^X0%+kv+$k`$My1Zm-F54L{B8=Y{8LvQviJ4BK;&4;xfY7@Wyo
zojA7%IOms#CR4%NWE+rP@IjczCyt^(>-i=?$Ja=Xe_~i}u6)>)lXMvTu`1b~Y#Ad%
z=GRAm@~x3j)tlqm<w`Evc`@E+*&K+CEKfqt$$i|=XztECPD(a%PvWQ;8g`<NW6WOE
zaoq-BwqsE2kM2Tiuh*w}RTmxGr$*b@NDyXvE|Q)zorO)y=`FBnIfA<wmpGTPFvmq5
z_3RjX5<VB&S{WxClpIL;8I&1o>f{55eD!pHB#^@l&g67Rl+_$74QJ+Ft1dH#)Ec~1
z9vUr=u1Z>;fejkr(o>8Uo0W4=@M>_1osF8FLuNGN>R3U%(dki|xMQFhqJ9xke?FqE
zoGnB<t(LHYU5LVEhyn^ewMP&=BzvWvJe{yS(J;F*i;`MMnx5+Vv!hqX>6nF6Wa?Ib
zm&r}`#LB|(e760#l410hqE+_NJdwhB2DNH2u#zjJa@x;{GRa~crwxmBtLdql^Qg*d
zvhm5<|D?>;l1ann%7^eqhL>wK5Rey2>B(ElTMX@c9Y-puypxX08|arh`x)mcAD@s#
zkLsUdNhbyH{5(xMHW59?r`71xtuEDn;05-}$e<H(h<+Zv;qi{2F?qUNy=<nA^BcpP
zMlzk0OW@@ri(qCPkHBkD7z_RCw8P%W+6pacib7l&KaVDvPCteiaYnC@rc9lTWAVvZ
zf!tK~t6!>jkJ6)t5UW(0Kj1^UPC0X)tf#{8SP^>3N9K8<I`Z5YcB{*jJH0=D7Gh|y
zWL;Zk%bsCNEfq`CGaZ+0sY9IQO0w6ns*0j6Pvh8esph>*+u(xA?bX(4!7FFl9+tID
z8lS2YDiEk<Qo0^<T20*X%zWZ4a$t^;igOK?*=u<+V!JT0EoqirfjEoVP0mY0j?H&L
zo0QR%Ke~GD@K6Rf&A7)dKW`}uDy8q|*Q>t)j@_u2aU&%Y12Q-=mjV4Y6_=430VbD!
zy$B3{L=`1c4`49B#F-S`TNk-4_Ec<p8Z4&&|4>qvACm37#_hqeV^N>Tj}%2(H$NKP
zeBK?-yH6*JjaPxdo96tz0Upr&3z}a#YCM2)7Mk;0^Ussk#_@IF3$?bEFVD|`@ASO4
z5Lg2EAHeh8BZ&Noua?K5lS}gk*@QrR86J6mNh3BsWhu><V@c;bT+G-n&;OqP+Uab9
zR?*3$-0FX?=iU3RHVA`P-n!=7+fD}1g0jIh5REcEHb4JszVE))uv5MQFkjj^qJl5b
zOd>YUMP+5E<4Y_==788jVy?(_+AHK~!_UtjP4(vQ&!ghjC@`)@W{#&kjM9$nfm)h>
z*l#EoW<B^qW|NqGeMm?Hf?=XK7#sy<+58@Xy^f|s(Wm7Z=oMN&%xK3M@VY#MwM8HL
z8N!rkh}3C^r~BC=%UXvT+2|^2yydBoe-dKxW{T)QMeS=uP2FxtI=7ik=)ElCi~eCg
z6vR2&3ZsS;K^z#>*@ME*>*&Gc-9sLKlZXrV^X7w!hUyh(yxoBqBpR?rgyVADM$R)K
z)Y2(t+iY>qc39^NP?5VvmjKCz95yCBPFJzu7Hk!O*T^b_)2<Pd!ZJsFhSF9!ZHCBe
z7o~{vp+ox{@>qp@C686o7xNgd<}wOG7|wE8T#Y0KR^HGom&MgwMiiBI?l<IrvI=>F
zTvj2bTo$vwdB^p6EwS8Q$Y-dKV1~HDDh3s_|AYCgT5*FNS|MK9qP4}6dNA+N-tzT)
zMwagw{$N6@R@{QG67d>emBlYP654UEuW@CZ3gNc1QqYbM#uX~ANbA);#FJIPm1l(r
zaha!{)lzzm6Cf4sEtI5fPC1N!^4JDpHKd0^kq@wCz(`GA;;qD#0j~^D%|f4${W@)|
zb@cEkZz*WoZiN)IhsSv<HQ9i;5)kMzbEWi3MS&5OhPt9yZlU@+_hxlx#+yL#W{}S8
zeWyGZfigZkq=v1V{Jvx6(@V^E9{lv>EjGvhrFqL@UuSmh`?dCy#Zp&)N8wTzEGKa^
zV^%zN89eo<mU~o`4mD*JolmS>r>4-EbOwFhl4NB|b=A+cE*vaZ#nw!_aNdkYI&2IR
z)F7R;m$%sW*fU+zc+%5!AzD-?cXrbbl{Hv5(~c_;HsqtdG;n!~wkYQpb-t$t?W)QO
zx!P8PG@{}ZTM65l8d7$DcArJWt@0s5p6vrAay|sB71v7S$Z3sB<61w{h2RiodV1jH
zQhB-P2W}0}7sX1N;k)9OrlsK|sKt6(Vx><6&Uck4D4iK7-b@+DnsC)mFzz=oCr*-Z
zJE^Tt<2gR_AMbIj!B(|BJ@1`{T<qJ|hrL%W2l#kpkR4nB9xr}>SSW?XMbZ|H#+l5Y
zjOcdgW*%`8H&tBrZ{(yB>;@+tKIEiwxec6D0bX`eDNK4eS|2#kH?yQa^l!7ofI=5*
zOKuN0?lrU`-TfrJiZ=`4#vS?CUpiU9Jau1{hradDFN=#`=K48Fp#fn?bCd%(6^Dat
z$lnX{8Yvk{LEH`!^G!C~&4T=L4K%5RegVrVyY`nMTnQ8cGBB3`q#kOg{>ca}mwo{W
z440k%E(ZfLG%}YFU_vK<Tg!6nHV)nUEA#`GAxQ95<#LtVc28AylX#cPDw)h??qaf;
z-ye`7)g-7#t0mus+}#odkcYs*L$VzHJ;>o7PnXZ9pRO7QC%gfthR?qSpbL`zf~1qy
z3=Tk84a4X6;kRcgm8IiC2TG|m9ge<x`ZO=!G<X^T^b6qSn?s0y<F9nQ(RCQdrTeC_
zNrIt6(UsS9MTh4+O44Czq;;cQ%(5>pzkmMxB=Zua2sVC=AN<$H=hJ_m_=2pKLP<6J
z@%<@YKn+5BHDC}P^ziq;hQFTv$)!%ZtfO-@41|=!-*G^5=9dhq2ZB^uYfJ+vNWvKh
zYlG17QihaPiUg~FArK6~Wa4*%)eta9i^jKn-ZcxKq#7rz(OjyeO(FqFDv~rL=#Vx;
zUSGah8Mh|tjAx?!cy-i4-_<yWbdE%rkjLb6`Xl}x=Mon!g^7O(Ak7&9<fm(pb7&}w
zLP9t1akXrVF^VgHl9uL7(lU>^a#lF!xE8?O8YYRJwwfG&f=hg#-5uqPpo5Q&J40KW
zcSAkrUH9Xaw0D<+bhBsg@xH^@Yolp4w!=PTD>Thomarw?7ut39jPJ+i=e+R#t8GIE
zqBgWA+J+8BggY&RwqJnhErYyJ7Vo-WGlCKUe8(*^9uKbvl@Zp%212*dK?L35DQVZ>
zBd<uAVgV+9>u%0ex+k_YUTC7>BFz?$QCPl3yG=}KGgD<&ODy_Ud?&G?FZ?>)8(G#*
zUKp)k7e1WMMmehX)P@_UyYV)oS#D+fupLY~EYJzm?(E3XO5KheElOj=Sl@&5qJ{~=
znZW0Kdel(DD8;oLi5eviF^Nl2BgdF_w2^HrIL@<wQVA{f7o(}c#FHnYseu?q!JTMo
zrG;^(Llf#4ob0-bxr)c!9dmRjtn%F)uN@;#cX&rk*YA4qp7%W8nol3#5N!oIUvo(E
zbdxps%qcG0!N_p){6Sen@Lb|Hf5g<ja8reI*p0D@QO|Z0czxVW6$rGUleplfG2pol
z0JA@Td@Jn*de`awnAeL|gnr#Tg+#l59)<W?$pFh3lmwT}<uV#-W2|#E)mHsUYZ4l;
z$kjRAMT08AcdR9o!CuN4_nfhS9Fj1HtSUsMJ1~DnBXvdARzi_SUGrRQCFiY#<7r+x
zCDf_}&g5v<96aST*Vk%JmvV2cwBy-W8^`Z|+Rh=lUGtJ<kXdV9^D4R;%zDjxM?GsC
z<q;uFTc5Lo36A3JnOzczEpf!?I-Lef(Q~qa5z#um$$Js4G<OGt(<2S0Kgs5sGc_?H
z5)4d-%w36YA<-&s0i)Q-b+op)3<^S-(4?977;r*?;R>&H4!X|9xkjH+-s1X_(u4_r
z+N!Xa+i9Rrs6OW{i5S;o*FB*K>i~)+6$v*6_DP+m?=vDbg%SEBk-z;ehd=4!&onqj
zmj9)p-zPP4_|NcbG}R5ko{TjLrmd!&YeldeD1gxBk792kj1GcWV84hvXGZypqH1kb
zEZKfAT<)qn0u@twApWM@Zz`CMTJhL_&!dAl2z2hlk(mLRB}8_2c})3?<4pct#^wZ~
z0;RF83GRn+*}~fj3f6}#VXufgji1Cy{fO;zCrg%L{8}ufjV0)Au{6A-*psXx=MKdV
zCc@4fi?0-u(D68nzT_S1W<M+r#+~1a@@UCs&6GXpCYIOHJw{TMSuGXfmz(Z?F@C$T
zwD?xIF9rpea)Kxko~v@%*D|MQT=e7ED#-EP=V0|(^iNkzKI3vE_qP^89<k7Dtz=+E
z`mQWwlGEOw^Nia{uxbaynN@koYr(kMrNk`VE+G=Lgl?BsTa+VvQ1G631XnEqgC-rw
zbpSk;@lYNsRTrjWUy)=VO?Z%hI3b{}aY!wY>_>G2y<~b*5Fhi*y?Hs2sX(w?^FkpB
zw9KFkA3e=4=fWkAN`88IwD@9509KP8Hag`iUHI8Q6;zZ;!M$_I<<s0;O4)MVKL_gt
z9o-z}d!?n&;-f!qy~NUYOrd&r_aS$7)(vJ~&xMR1`Wa(Gv-ABI^2u(0Dph6J`IjR5
z^;kDRew2T;nz#I`9sNcA)hgNYuXgm={&nA6ihs?_<?XnCu{mt`7aM)Lf2|tKL;l6r
zcej5rxL5p3rr>s~r}z6yY`|L0B`x_&oFIjJHab52FFll;3+<i<20`^Zm%Lz@rD4g}
zZ`&SD;gRUOxu3vSqJtHGlxlPtfSpXcyTcorsqZ+=_PbQ>a45is=8-YxZ0@6a1ZAsS
zRX^n3h0mDA1I7$Kwy=>)Txifkd=TqBA3gQyOqD<TWrQtdka`O)tc#WES@$DwH44u_
zs}Z;bRx`MC6{kkf0!Bs&6%+4#*ENR*Q=;j9FtSfE$sHQo52+@9i55g4-(nI|#iXj*
z>{gks!qiIPxMW7uRN-i-a+CQ{k@bQ(lfSeQpe&cX#9!J{83TOuDu3ZJhe=OXXFCoH
zi`Jt8*{O3KhlNEBQ{D;#{s(dx(|CecGKm{*$yXfd25-F;V^;lBIYZTN_<<zGG%iK(
zf<-$zR{tff+n1q#*N-$GP6heg)#1nUOyb7ldA?ib;rc@bpSq0e`u8QMWQXS76_w<Z
zujsYn+Pru2_)8obX;{!)+uqmD;_EBjCY<qBgr(#Ty3IKIt%&@N8{eP<Z$)@5K;OOP
z`E3Kf<+ryYW(mrouy@h<^u+H0Y9bvDT3v^ToN#fWl?P#eo#^VZOIT3qV-fB_qr8=s
zNc*EA#A76`BnmXZXL10m+;m_9mv}HY9XJr?#_~(hp`vK19)+$vEk`#!F@G4kb|xo4
z*9wI7hrn#~5?mUDQ}_jGJCl4u+Rh|Cn6@)13u(JjQaEc^+J1!li)5gH?9n|Kn##U$
zDJRS?mjIoA<Wpl)X{#_T2VzsPH!ksg_03c^`t<i7OA`FWXl9!}5zY9XIqEynRPjyo
z;5sznIDW;>DQbG}Yy2?W=P8s5T5Y(5{g2c({#$~*>;shV2Nlk&<YUrk6_>nJ$>-+{
zY_wtXn5*71&4MKpaCIS8Hz5la{rjz@V6WdOoy+%siW*}CEuzu$wyEcBSr2a%6g{*r
zwCuX|ceKFGnH-aRB#$g!*Gk;-I<_)X#cw;NG)vfPEo9JVM*V_GEBQ#&FPO9vcSe1s
zOp5kQ3dbe;u2NLyoEr2CjR8=hFs1#|zc;wY;VJ0^83Uz!8<Mx)JD_N#s`o7Q%FBHG
z^W*ac>Hk3_r<HL669Y0hGnYZ|EE57TH<w{~7$=w2{0N<YqRf3h{`&I$K}!hb{>{r_
z>$;6jp|;EKuZPbE<ss(x^?`?r^5+AGl?PGo{PM$hub`K&|33U!<=RuR^;C>Kb=ESh
z6qcYUdS~}wwJxX-V-G8`ZL;ef(|?AIyo{$i2alcs!2}1!Q4%jrgU(foN6Cm(ndiKY
z5o92<rxOByaDEhVT;gd;8IjCNVwXK5-FuejPg0yj3L<sO{+xi;8pY4mewyKwY#yIX
z$oAfV-Ks%?UGgwvm!2U>);Z5*BuO1jqH>PdOFT}Zc2xr(@h4@bNv$ZPS&AjmIB)$C
zyozq<W<I6Sc$X2Jhzm<bd=^Ab1)U*g*UROZWR@I%Z9$|(U1o_TBk>xGAYAZ$bXB78
zCrVtGK4y%WBUhX-S^?FfguZaF<ub1n@DeSVQgSk9L-|L&f5ij8VjvP9CCY+m)!k-6
zM(?dlX!(@&Sw}0HZXE%8DsqL-p+t3$Wjo%Fe!7nLARhUBl=x{y@^oyrLqn_F0rs<{
zjWSSwYn^rCby-K}HO}HuM(k2K+ZX#fv$352xEOC$G$=ElH0?`s1@T;5M(LrF2&TT(
zGI-wQ(#pvm!>jcnRn3_ts_2ZvzN#4r${RoU^6szqvqESLDo3<+e_gIuuGds{8%9kj
zo#)O4F7?)q&iy9N@jm8M^fu6_LdwUK38T<|j^S|X_l!hPA>e4_7i~ziyx-GG9NOsU
z_kFT&ZX<;%^e|0mMiJLh<L1)ap~}20OtZrIH`Z0Z-hG<W#RB&s)Q9oEP{k_~Y;)@K
zoF<%f&t0S;!_9;n<G96>pj_<2et;v9>r6|aFyvl?VGL7Pe7eecc&OC+hd1<A+`Dam
zIIJ6DE7_$nk&-Vkg<Y5&FARnWBB=mj4g9BhGPw6N!7SjB8`QX2wEOAy+{>b0AUI{o
zK2{klJ7ie}8T-eqFsZUeXXKwdyfP|qK^TpEHD{L9$TQ(+huMWw=PVEdJu1yfZ-J+=
zesu2B+z79AurylvcW%W*r6Jz1vJU`%a!m}QkudiQdE0`GUw4d<<HEa9F>@=96uC`e
zet2jR1D{e1wnaSDwyOmvr-HT1^HL$D*SzH%kXP2Y!P1@7fJ=?f0tr0nR__#V;H4Qg
zZo?T1cx_sJ9~if|v*KuY10~$JKQEAmu_f+RHqJEE=nG2aTh2Zrjj|(|cFt&jZzz=T
zyZ+yI9q-Ul_S^A#MqST#-Fa@t0R*m?b_3$PLLnbjj2$`93e-OEn2)?+3;Y9FM&#Wf
z__t&k(JORU#%}cTsY$Y6x0_|og8OoATz%ts!&&OiH;$|G&W$rpW5nwJ-j|Eep9T7p
zTGOvwGqRPtTc4}`0~A~V9tvfbBU}j*0x~w20i+&jx9I^15K{p$mvOl*6PMohH7AzZ
zIs}(ERSBPeAeX)V>*2REolJPP18RZ_&Nv+EZwx9!zHwxf1E2Lr(H%t^zHGTX+d11f
z71(+zspCUx96A;G3i0{3tM3<y#!pHltwhVlIis{U9ntR>i(*7Mb}VAFINNq4AFn=T
zA~Pw+Ii)2sL(9(JFUw)AEIT0q1!k)p4vHz1a=tBp%kiX;?3^6tNjZU)#!AjFi>|f;
zP7US|S6eX3QswGp2Uugp(Y93VaaoFC6|l5_MTj*DY>CtJ@5@50Sa0khQjA4K1dWrd
zh}daK{(b)6^KFQ>R*?zY5KZD{31)ORY*H=5d8pb7%osXk8lxnn+fe4M3ObZxe!aLf
z%;&IwtY8QVwJRyFGRb7|u`x2<mLz@|2jr9#x{~DMvJ^po04_<)vJ{rEn2O}}HV|l(
zNbpt=_=d#eOun*Y5d#!R5u|+ndlds#5Jo{Pfo)P4rI`Ct%348_!ilDmvM-#|ttS|T
z!IX0Tl%}>E8ZFgESK579E(B5rk*L&erQF1Sb}kjVkg`&4z$vnPkQAd8+bW*hQl=o;
zmNEs&r<IgJ1yqr&rHm9m$0f;|mZT3jdMSFMyR=!<+d3tQGf}uYFFCxq^IL_Ln5&CR
zaBd}pM$(=mjEB-zKpeZj=2DpH>gD4~KNr)2FDdnEiPaox%a)|XbfH>5SPs7S3Vo%2
z*oT#L5s&k<UMn~rl5E8BkYrowucEW1^cN(rFD}V;7~-|mk+9le4-ttHBd<+K@+Ofi
zuN{Qat|9bmN4T~mcaqVa+VvqW7bMFF%5otageKV+aMyjt{9Kk3lDpnT3KJ=0$v|Ix
zx=VDs%oVZY++M!pKA_>WM8s1(dg@z${Y-q;(X<?+E-s&K&p@(8e#fZAZGYA!-nP4!
zwhZNwDg4k%mMctZyVr%&S)JrS_pI}gowcyDq3~s8Gn6*_Yv6e5c2C=C?Fv70|7q88
z>b=j`!xG!$!h+xoGcH;mLY5oXL<}XLHolbk$5Wg<7HQ+tHpZN^6rZ*pg?5R5ozwuN
zyzLiadvi_xMZp`A2JzyoSC+U(lOLR=n_hp;p6Tw;Evj0qjvH^hy{bOlKb(CTsfj4_
zk;dlz=d<_)ir=hkN=_QHWZc((n#Z%ZHP^IKYgwG-8R_oF*La|K@-DqO@E|Fyoui`J
z0LX?#*kAyK;IgB%1m5Wg95VxdL4n3_vmf!3Sq#&--@+$32BTK_N2_e{<lT@{P*yr*
z#mVStxw6Z7@l|y?CRvd<w}<P2K>>~|0xa}6h$C4k&VI!4<r6q^J1Ev1XJX_eEu|Y*
zr(?0`Vjg3yCBfi6d21bFqxh4TIW$b*X`y7c>F?LEg{-K5NkmV8R>5c(4Oyks69TJX
z4F2i3DjA*tRwY9mz1Q0#x8_+1Hv$CgFNuejAzTR+m%$PO6_=8D7bbtL8Ov_tHt_DR
z&<}*mAvt^ifq__w7u{R8xh?k8cK6bt$X@#UouNpIq8_&F9AsJ~HSY&!D7&AX>^>h(
zuZK@(Z8|4Bf>Yh=j}B--(l1DwwAeWSVKsEGZ{4>?DV3$?M30eDYkD00_cY@1L4&6q
zKyLuY2WNoypXupo9fp5>@w>7&LO^;HtvtmQ9UjvzNspzS)(voJ4g!zI@2_7DGO>`y
zm)Ao}x(-$tsk=Yl4(|uyAjI|Zho<w#?+0QlPMS2*;-}x<0J`UY55Jdv<G+mSAtQUJ
zutl+=-x4dyjmA-|&r+gKOlxA=Fw2GnN79i<(}o0T{O}$}5-5L?@Qq0$?^Qe>#dArm
z8kI0KR%%5&DpnaoFHk+J%JHnpN}JId*CU;T-?#+fJg7+o%Od09bMr<jd!H#yDHM*D
zqjpFi8ie+$13-b+b-(`Zey|#*U~6;sKzvJU5hr{D$s0{a7AG8ODe=8ZwV}_-&##|d
zGxK${0MM2wgcg6Yk`ps6yNwY(w43#e44OC%p2mu0-{Ak`54*_=<iFmx7y*{a3n+ro
zGy2qEgvHQg8lRIf@<bmY&IArYIAe;yZTf;Csr24g=AF!%xyL1`^hp9_3sQjYK*%J4
z$xJPQOo{Zu`O)N=pO8!)6BYE|V5Vk8fkBWGTWG8Fo{xVIR<c_mN5Pld^*6=WM!{VW
zla3pR&x`yLKLUblce9>ISQURN$&gnhjqq}_rc|I<pzH@}s)2TercQThs=@aGnrb0#
zYpU6egO>Jkqd;+R{WxY6IH3TqWkvzzr!ZsYdb3kVB%mCAB4v}ACA0IEQjjZ#Vi|{e
z91bqg;5dI+oigV5c24G4LRbDK(o`Unu!qT%8REFt4@Q@9#Nc@5kWs!@iFZ3Q_nREd
zkzq++J6Ob<uwbGgI-42gOWZ9w8aiQ7w@@!T+Oh<z#86AD&)l<f@_W*=lv%5p`^gM(
z|LGj$(wRV6BP7h|9=P3Y=W8bq<*abdwOH=)Bu{^O)i8s&;z=F_J?-w8!iy(m8^v_O
zjJi8zN#7Jev>_z~eI=Ee&qZb?rAT4FA2(a%Oh9B+?XheZP&%?}XD%pm+--zae&!T+
zhA1-GV8TL(vME36Dpd8!jr-c+b~^b@;RUq_;iIF=1n8B)WmO^*!$0lMj@k?3T}#x{
z?LL2PECI4Dz!jH9Du!e$XJhuWkt5&)5UEkZd9{Qo;b*Rp8z=~x%006{!3ytMT7lR~
zJ}R`o-@S|`*3{g+xB!aB%~bG|EZygPe!tUbC8X1nL+?JUqsM5WgKr^Tb{Q?Jvs7^p
z6K*~{HU!+^9{D)fKwLXK1`*s$0NCH|=M8_}mu0-!i5t94Vm14#c;%EprCP`vep1qY
zL(6HwQI>isml{Y^X;3{_4(2)q_;_p$xpBnpqG+(Ex|I|=38~psOuV$WZfhF>a(Nos
zwvnRMfOmI|fX;AIjZKUT?T~nE*=}IWFjyq?V|lH|Qn#Fn&&e}$x+1N{h!TII`AmOi
zzRtQ%4HAhfogEV!VYwO(d1kO;W`quLDL6=6w!T#CwHVKJL9-p&QW>y-eJa{x=~<q$
zX>XV{&<NGIY)n7_)X#d5ic%FoHl_?=3}-%~^Rb^i1F0!p$!MS7zRT_}TKqzrIZXc-
zx4soLviqm|qiQv;0rpX{z@0h;TY!IMN70znKY$%Upbmnf`+g8tQpTM>tG2<C`3G#_
zZh9n87N;yq%8<C(pWI=^@m>Ma9Ro*3vM`d01Xn#K-EH`^rqf=2o_Ic)-6)fZaTRIb
zES5{+<~q~dWe9sgTxt6R7PrB!*?#L}SKO?4C0sed>2MrwFZwYJXEP`fJH>xk^ThWq
zu1#6-CI;*D=5M-jp`*4QQ;42q#X3)NW-Lu1o8}T~ZW#?Zl&a2$9QZgVz<5#Phm}%l
z`5I=)!5P09npOi1LtBbPk@1%m9Q*c|R)>~N*|E~hsbH8S=n-F-)@DU5u7WfDzFblj
zeydWFDH|z@be;;GgOXxcnJFvD7z!()RZ`Od$%Q4`j7*_}a(4+LX44J~RXU0)ohgdu
z9;Uj@{2zcE^GBB<TnQ8cGdPz4q#kOw6k`dKE&(~00UsDBe_2~^8^;cQ_pg}8whyy&
zA?H#Q2GZoXC%`#q5a$4G4a1P-jibhv0(p(3|NeX&DV8=XS*uNyrVq9`vm6eG<i{b$
zF1o0Ua?vH_ll;vp$=DlZos$VtHjvB&<y`<lPLVI4yrN*`lUJMz4=D&d0+CeWEHjXo
zg_(@XUU<l=e;h@QXJw6zG8M2Hm5)hTzhKLtqO4@7Lbl9d1#y6UcF@Yp-VFVsD-^!U
z5ESIaEPHTNHkE@hB&TvTw#w#mwt%6<DL`T9Y@zVWvBe&8YM09#E9apuC<ygYbe7YR
zl_LiH3Qi=KV-*{P<*_ga$(d!hXqC4tAEF194fcoxf1?&-Mg|o{`v6yfQRcGpY(+}K
z+F%OhKowg>@P#CN2~tuNO|wxxN6`vO`ds!%L6KcZiJ(ZnrUWXB)lycp&>+E`3k`5+
zA~8wK;Nu|n12b`voLNycg%6Su4%j0@vZw(H(UHUk$$-itB|&jE3n>{CHpiS%?AQxM
zip9Scf5v5^C{q>`Gy|0s2XRNTSRlaz3SY%Gpq%35${a#4g(5B@7!AuDt6;=|2~idh
z2SOzkoERg)cfql!g1v(`FjYcSU^5&eA?!mig@Tkoatv$~LMn5RlDudDoP_KOg|Ca_
z0#ebGwE#x6h<s2CWO&&FI6{obz8-wumah(!e*{SaGs0Kquq6Voe9fSW_V7ELDCb3N
z!Tg1Sl$=EafCt93M?d{kJv&1nGLZeU`tA2WfU}@ekd3XH-rwAuA0Kbzir`|!K5K%{
zc5_m)8V2jZ>4MRm);U%6bapeFpS~N7YvpBus$Ptmrk+ogTMYm8q50*s88x-Ci;0uj
zf3(3KC&&=iz!S2|i6Gm;1!VvLSVY)jfc=nH{9*vt0QP1%U>Ovt$H&!+`D}bzH)qw0
zXD8LG`a`ok%d5ZM)zIm>hN{!1o;G)qwqg<Svc8+$&&T!MB9qC~f9tEs=;zsoGub2{
zHUWlnte<10B!;qG^<+Ao6+z?{+aH~ke^e?5EU9!!l+&F*s8N<xr}vkQjQ%s3zE#!F
zv-wp$FWZ^(>fhD#>gkzX3|`jb=FCQ~okTo)ZJ>{%psipbI30{XubwD~KdpY5y_!|e
z)OXN{5U+s|?->z{<naCR@tQ4$C;?<?hI@z!k$~(4rAWq+v`P6ZDPbLy5T^$ne<b;t
ztR<aVui1AEftv#P&*7F}NjE0Ml@1c%!0^Wa2O2k^YK`F|0+6b8MmE~bNjYi`p$8}K
zA~^Qoplac$g)l27wW$NDPvC~D%9$l_PC^-k<p62Tn2)FoatjYaC3R6rd!TY<WeR<=
zCUt$mDQVArIK2TUY<mE=ZBKtKe}`_%g)N8f>jL&27qA`T0(LvlwZ`y(F5E8$wNVdF
zOWOA0MER)$46>WkdaDf`#G}1j^@)31a6=&owjI~wzP=8&>+3wk*TM9GTZ?N6H^Oyd
z`fz$qI(6N+iPMAA_2G1xw3LV62!yvgyq%MW=htXD4i9Na)f2<2>IrdKf9yoJq1f4c
ztHV0VE^l=rt0PXzQe2|?We2aI1LSD5_4)j$RV7qU9>kBE$!v=I{(mo@%g=Xjn&#cz
zk5%>l{d;|VKc8Jr^muk#UCqY#w>ZYAuxD+ndHwo!G_9`c(amIfeer(Myt%lXTwmN?
zjAzrkW<Jtyn%f)H3P+z7e=NICgMM3+v_MHS>IBP9qdicJ@U@A)fx{Fm<Q=(rjo|^R
zjsaEf!)bUURJ)vnzqs=cfWPg<&LjBi1GyGV_r+i8;xDEl2{-0G2{&%=P0rfee^kcX
z+sNE8a`HoxTsEsD7vtwQat`$cqst6VE-3U+2JK9cXN#oCcKesRf4U?E?Ev#u<FYjQ
zcKEI_tQr@zBPsUaBxEqxhtZ)IiYEV!jY~UZw$@nk9A+#J38qY)V9IV#Fr`&6W%~tQ
z1q$plPDz?jwA%=0M4S`-IMKFP`a#=lTCtRUpG?x?T4s`n{wL)goM^$dap`YM^LzR{
znNABNCF*XU%k8-Ce{*>;UHLpOs-(1NllQKY92q1H1*6CEwu(N96`EBt1B!}k6bbj2
zrJ#5{udn8lxAoh(tcC|*a{cGP;>RVgvkglv$>NXOU(4pcnawAoo4X1XO+A`duSes_
z&7}G3;$}83g-=UCM4h}+KbjHO86Mg_<T-ir{;GE&VL)#Ae{4b|gZTLYo4WxU`v>fr
zl~Rvn(B^vVcG<+q!Reg($YcU%{$?JVAE^#kuV4Fm&W4WXEDiCT<sAU9-19AS6DMWh
z+WUjMqWuurdQ@yZH}><(w##8(C%Nu@;Or15IoAhXZbPaH=%|uE_zl|5LBIWa`6seE
zmOQ_e|4g~9e<2PA*zD<}M(pM&Hz-HBv?J18+7Q1;C5Ne04^n~nPSq*@TpVC5->153
zb1fgylG8+~4<`~jaXNTzXqWe`+|a7^Tj^kxH9m>w@()`VJ|m#Xc72qG_=wE*fc|2p
zJXmJkH?*nKohR}Dkw9+0gUm*KLYrt~Xv@XBn^T^L<HI}F{{;cbt<55rAzTR*0X3Ha
z{WcQ<G&h$qY!@e&fSC!Ke}ymq-2b@9q=FQ|#-H(v|GV8U|6RC37L@?i{QL6~uNw_Q
zd(|kq8L;{Ft@(cWlQW&7<$=(W)*=>Yfg}h`V{H&R-s;gpLduYCh7ds8|L(pC3G<%5
z=|8&k+x5B$Lu`kSUq2dl<*%=!T8^%O<vS}`pfpl@5fpOjK|aD5e|1`97mV^=H_N0z
zMda8f<P>uvDR4qTC6XlQF!Ults4&9gw~JzwAWwW$7Gp)ICGTXRB?%_^zFh|E1+7X3
zGcSWv0+p($Z!b?)s^?@R*Bi%dmqtq^wA3pOvx#GLeyAljy;D=<Q&O1KRmZTaM2Rqy
zy77T3zR(Lz2Fmf(e|}zkl{>G+S4sVw;;Y<bExt<Xv*YVAS#o^M$kHB<FDAoEd@<_N
z<Lg6**%e=0zlY<C!9B*;e4$(G39M2=X*H<kMh}mk-9=smvCVb9i@$5<V06VjCt5YS
z9%ddfB4AFo^~$bL3k!yqc8yk8s}I+941w5>xkqo$B|v$Oe<Wd?mU2ln9vOg9>x6KJ
zRCH6jT@Dijm`bCCpap_e2Vd!L``kK&OUkh^ugPnfy|X6X3-ZOPJF-LAS5G)6YzS4p
zw{?ZcPgqL{X-K@u6%FR1upO+de!;j3XBnTw_#qR+;>whv1relU3=u|?cA?Vt*Uam@
zQ3AE)o=dfae<GrAe(2$&fjX;2P;R5uI*VS%Dz&2=<od^UJdpgrJ7FjtF^=%|F)Sr@
z=cjGb*p9uy5?;m#uhbe|W}3sI>2N<@<&J0IRdQUztEBG2WoQbPanhO+uYSm{BKmJT
zeaZ-$0r_r+Fwz2C;2f*Tas$VVBjXrr^E~oO8ey=~e@gD?2BjBH+7q?OE#sUvw|Q+;
z5W!XYM{k%kbyY>BJ|f65d$sdal4`X-wf^aN1YkMmrTV8W^%20SKdgUdu?Ctz^+36*
z28P8bGcHh5QrEz+SOX0;!~nmc29}IZsDUNJoi(szvV<Z!0r;6UknMOvPG=lfayr)-
z%XxSSe<vEzAbe)+V>>Q&!x@8;THnU;aob5i|E!5MN$wswx$J6x&UMttImbCJ*jRZG
zQbT1e+<nfWf)|uiR`%aE2wQ>ps786n<r=+CQbDp^suPdon=kJ{&dqe%5LwWCx0q9>
z-aO31^OK>b2(y(D%-eDCsH>-8vEtm0!|?`#e=Lem0qA`N0y{U^Wb)Wp>4!~DA+1C0
z#L(Bzr39r>DE(PYC);f?IrR-Cm(46|p^Q+I>F)zAy5=Gs7Up{=q~F|tGD!}UsdOyK
z7%pY7)MIiWaLmt3&;d0gfP10qGD_R;&^H?Up(`ah0lJbQYvC*CO6r9s+r;xow5-WK
zf0CUtXYEN1?RO^6Y);PQ1o=|KUU5D(d*(wlw#$L+nY(PxaiPtN#RoN~#y)dCSqbOV
zqTHtaW;9dNdLo(uMKJn7G|6wN2Du7NIF4WTa~lMWa#Q?Fcq~$=S+N>9hwU$yQ_I_y
z9x^n*<9<f?K;4ZOsNl@GJGM1ean4I~e|LKG#;B3WF*j?c=~5df&=#}8^vpSfVTzEM
zPUPc+FyE5oeJJ5rsf&Hu)qUFKefV6T_faUUG1F7cqcd2Vsu<@ZvuX~zlH*#~G0kvt
zY+?J_y@Wkbqvr9<!;Xwe$@$1>N5-V&xbe6{5k^i`a&cTz@Jda@r&j%x$5c?lE7{=y
zt81xX969&EllqA`)OzVxIIq2?Qy3xu{4L$wOF#a7yI=kX|BZC;mmypU6qn(r0Th>F
zp9#bPJ(nk-34njjxZbk|jYi1e3gc1#c;dTk(W{|Z<9UR9XUDU0k?P1&)qQwmeiV;}
zDT-9C-Vj%9Cu6b-K?~Y1QfOnnrO^fn%Xd=ReC6}1N<byNvZI7hkaTmNqT2A1<ZA7*
zp6Y^-K8&Z5hx{<V9z`Gv9r-Po9~qe|>l+Jxjy06@<|}`qmiHwY4-maK*_tnV23;ZB
zi&ryy(VpZE`d01^tu1RqX@qmC->visip@^3M&*P9zkzjf$=56!oKQ4wSvJ1OvNKg{
z*?Fd=J@#ImB#K%(7iuiJP<&I_Y1Z4TKja;z-5=v;@^f8mH?E|}YFMgb!OVNb=#KZx
zAV5RK2snSH@JTU_r18?<kYynrg4u{w38}oQ84+ZO(rk`?_3gaI!fLC&VBSWB2HAC|
zWUg5!iAyYKS3dY$v)UZ3V7)!lWt)68J_oyRWG&p;7Mc7Wv!*ri?wI>3N9TG*v5LaK
zZ+Lh8dF$<)JZ2vnH4MQCaMgEm#%PeZ%^D>-*hYU#dB)7*OVURe<2icb5*zgVYW-D%
zTamx440c6)=U7)cGVit5n$4>2PJC;m!psf&s=q0slmRqijnC(fqZ(x>>F--S@8{4O
zX?*u~E)k_->l&N0D}qD5j`;#QZB?09Eis}cg2={S{m6ad^zoq0zUZ@BM(C}|$nCs@
zT(5s5%|8Bc*3{k#s?r9f=>KIH$xZ#rbTPihDb6iDgoA_qMoBkMX-1(y7e;=!OwB5R
zO8u1Lthkw<8Rnp((1gkBT7FDiRP#MHDYKRpweaKOUIZI~h{MO>jSBsjTfspIV^yx{
zsfyQ1Xb*&%@WCwK|6IPs+Py!uG!P<|{K9|4ANrS!sGIrBeNvE=8KtEN-DO54DP7_M
z+6zYqm$<o6pYo+#HQk=3Eitaf)8LX%Og)iii0-p}m*<F&uCace8{ft^lBa1io(ODw
zvA2u=^;Es}Y^u+^Fu_?pt38zWLb}>AX}daI$>OnXYU}hASEwRV`uI4{(v5EJ)jog9
z24TH!JK=hFgV-z^H_q2aJ>O8OrWQK4ak9FLf2DSRHA}bI4@2IwUAGJldxP?r6>FB2
zH<du2JKE|+u055JeYS6~n&%)KgV}q75!uP5v+6FS5V@c-cTvVr@~|<aeEBMuKWOn2
znXIF};@?F2OdTV&`~m6z-^*_lU`l`AKQSTzEgZFh3Q;IZP{oL4U+9yVubE<O`GB&(
za0*=3(hDer<Wys@RK<ouxbx5!mT`(QW$6_(Pr0C5s#uiRRC|^PWOCw}BSt|$8U3>m
zlU9Iti-H&v*0!vC8mO%gjOLbSL#zHfsK#tz$D3j>N#A2Uf0Ho)?X<4b5i@^fZl~lh
z9YF4z^Z3<Pifv>8D3QT?kd<%$#7BGzP<Dne*f4Qua54rEUD0K_wiT<z@j1`Ker*~?
zcjsHMJ)zDP*9~hC$tzr2MQ-{UtR~+Q+PZs67|i?7_2B|b>&y)qO5avezQv?%#@cp`
zPd=@|if?XD$uw`MoOti%=b3M*Y?;SUoVH+a6Z)zxP<&D*EKZ34rN2K(Gbh#{9NAJw
zOo5~(fcE3#nxvZaGilvj^Qqz{U9lXxt6^)iB3<i}#K~iqEL``d`r5opWUKYsj+5?_
z-&~$}zw8n+pE)$AYow=%>(;02{{S;R$vl@KTnQ8cH87U}q#kLv45kURPL~3XI1`s^
zyf!G8`}7H?e*ho%{@dr@-gs7_{H^<ziU|`g)@cU2%lp1@gJ#vyN-_tE<I!oy26T`<
z?ndkQ4L<jB#mMQ3_U#oN>Nh||-{$*5-Qo<;)Yy18UiHm5W8P&%OM4`f&nwlDFrPnQ
zM9^{ky+}*@v}LdL$~~_bKS;KKI|3jE(OkPc@7_vEe=GCp6Zcj3XeYUGhJMt&sw_j*
z(M6WYdg%#)w5N#Ob;h^FEv@UQ{4y)u@fb0JDYAr!+xN-#3!7WXxR=G|I^?yJcRFIl
z?AvxiY-yP?V%9(JSx)$Z%~t)dc&nC7+i=ud(dZOc=JuZUoUBtYu|<=lo6oqra>o;&
z$yZF#f7rFShZgH>QCH{a$pX!{S>6H5WxP$1)M8%jnJo}~R!o-Qeu;X$vJyLxm5?9g
z3}=OAK~_JAxyTe*bCl|7xB7dzI>lBy#>n`sM!U!eRj*7^9g<qS3&*={$0ej~qfL&>
z6oYn-lPvW*$Fgc-s@~72Ppjg6&^y9b8Me`&f6W8FVBo$rS;ppw!19ZaOcJl{O4={*
zidPiU>fXfKBFpA+7{OI*bIF3WHR^RU%NKvC9*KOf^{KSu`?~)x^9Isy#jE8<6Q6$5
zQ;j#Qan$R(nS|Hv@;Tuow|OkKbZ>({hD|YNnim7`k+;in3RqXpHHkCLd}N5OIML4l
ze^xRTHDra2&i1jcvUN;5-O)L(G_77&YMpt@D2Mi)G<0(qYB0LU&R!DygihK`z(kjV
zvqjZbO}D7Pr~v8cXyJlKuUTi^!K2KdA68yqx|LR=($8}iv_`VO{Pk0wqAr#Nr@X;a
zEVJ6f6wkC$)0BVbDH{c5h$Kc8NbFR2f8g3eo6OZX>+GN_rhPPK++dQFQ`@@B&Gpo(
z<1abC2cpuNc|ZbzI3ZGOxrt2i1j(5a`lu(M(I<OCBm2vLcRazsc3_I|%!_c$69hBf
zE;Oj-a;qh2jzch^Tq#SN!a&mk!L63YsGu1$J!i&vjM|f?=bSk%j+!2gUNgO)f0Ash
z68$q(Voy!U1q84|tAF`l*EuBj%!F{!Au!495$BLlooTVqaW;^9-ayho<*yCi`Hqif
zKGH54NY&E&MvGd+OS1t=0Tk9aiMD@pS~K^?cWoz;73FhzUQ%dVeD}CVL}ST|)`nSL
zi=)(Xi>~XD8OtUr?L5rUmtyw5e{Hy_y(P_wZCk8KCn&O2;p|#Q@&3%Y2w$7MzW0P8
zma^__o0v+&LZ}`S8`3>Y@r<byUd>?!XLH=z<;r9)ggs=ccEug0mk<cX3YY?M8T?bp
ztsufJQ?9nJsj1#U1kb2#P*X?hYtua5v(Md26m1mr%fbOwdTUjlKUC%Qe^K|ILGy~V
z+Cx1k=0pSns)~5)K;jIJap@^``5<DGyP!Sd81Nm!S1{D;@XXefkj&bJSxal6`$dV8
zv7!oikA2=z3b3$G!52w9N~g}c-kB)})s9*{Ly%In?bS!q(YLPZ>qyejOjpaPOdgYa
z`uwdpDbGpbk<ap`yQS&6f6b%P-;W|u(-X^S#kSy5=oGYXTOb5uDI)y3GYc+4sYGVQ
zWY(Zk((z+H3CMdb=e6k@?WWUtsg^~`tvNfnAq+Lb?TQd@Ppwo_cGgC7K0A<7uAEuS
zmYZ?byUC<qo)_=45P~lE7(e=T4CG)d6q+)-HB|dHrr`MJ7xL}<f9KnOmZvG2DXB02
z{P7mAYbBWS5>gM$;o<V<zb?PO{hdl}pFMbSiIHgy?jZz+o$J&?8K>IfqIPv9S6I}r
zPfBPIQ~&w%j|;i-YeO(bF$F9}R)Ruem5!_cy&N5_7RN4d;g*ZT`}AZZ{$JUr_ZJIh
z2jvx6af2?Uh-X`if9=WavTiK_OpeFzd23M5a|Zq)fT5<Xp(`zP#Yhs8xiJ+wb(kL)
z@D^iWi4n?u)6Yljai7qD#ea$p=*q9LLJ54$@g(qrHiGV;KEkM{L--NY0fTP3x`aT2
z5LNLpTbL_v_4EpJ%>;zy#X{m^+pTqA0=j%MBc@RG{e_Tne}J7@tgm0!sDvGn_%R^S
zGGUjf`*nOA%omo9*8F*FfZl>M2W<0{)Bw`FUm~6#mkjt)@y^&N&yPpX@J=G}fOp~<
zF~^K}Ct3dl?<5jW@J=Gp_PrC2e#JYpwWSZxb@>ymO(pJIo6LyC)PS|=`s3E7634Ag
zCH8X>8J(@Yf3T#R4`EX)FwBsJ^`(sk;CvkCQR_CJ1Z=UMf-2NgDFaoA)P@DEP!+P@
za`mbZBJqhT<g4W(mxZPAny$*+j*aLkHT37G+B$yz*DY#^WPV%sD50V`kC)`9)Ma&b
zE*Z2Q2kWv;?~9LSLZVHf{cyQ6xeJj-QVdN<Wb`Y^e>$fPB8h+BNE;elODuGLkwm>>
z!iA$cd&e|%qbKMbrjVrTu2V@V=BduFQAzHtj;Sn#V^WwCRF={(Cl}yBxQ|qfGCrpD
zZmCUveUCCeq=g)(CQC+FTFJ)1mBoI!E%`4V&eWr%=C!bVthB5Lrdaim6@54>R<Sf;
z`mu^fe~-i}mb4eEMCZX+J*IPs)p4CmtdeziVwLDT7ORI;uCY3-a*frl%Fo5AW>PI2
zXL>G|ySt4HWJHY7bK*uCC^pT~hc=eBP)LX~;zl~jjNFTxrzCnPtd!^isy17#_;5sV
zb&{8=Skei35Y>3D9t*E=R23B8#w~nfkWy4^fAy4KMOF<_ii%m=EHrN@B|lzHk3-E<
zv|<Cq*IBHBgl%=}g;dbs5l%sZ_9zAGJIE-9^eu!ktZ(5Hyy^~}V138f<cPYJOh(nM
zT+&nbB`U!Jk2A>`p;CziZ^HC>WEjgrBfU_0iAA=tJU}96_-5shBY;*483nX52oiJ_
ze}SCk@#!_Zu*ewZ1VVxwPOU%FQ9buD^&xYQf{bU<$+I{JOC!IPT96PPJ;Fj{UEZf5
zqVsi(hmPr7h-h5r!bDx2pL@!i>552f6Us)8lp_5Sh$8wLv&;Bqjaz!r1~C*y8AeoU
z$($BvBAQ971L#)s0yC^XDJ-S4%sIE%f5%4gP-tZwP#>6C`1%Or1<$w+T|hgOCjwLQ
zthiV4FjyVLRH^tQzca$V(g~I{%6g@_A=;e4QZBy0wseBIZPMjcrt-pBm#4I2N}Sd!
zVQE_ry9!U?(IY%X*4^PLqVph69n-n+)VR)tr@A`7#8X7#6Ffx_9povp57XzVe_=!`
zPo0No=BW;#J3MttSXQ1Q`*3}p8V5A<6c%&_Pjvu2z*DEhec`D?m{y(|#<cPjmUI?R
zo#y(bi=tWa9MU5mwQSDrufMR72{HDoM(<HcexHSmHvHvxS^Dd~M2bY>embv|MznhI
zKrz3+M)$FCqF87RkA5Zl6goQ%f6;PO*a37Sofk$UfWDZ{%i~?LkM)Y9vf4PHna!}E
zk#t_p3p;?S{eFso`r<Kg=Bi_u#^37%eFCQSxDZPkP3Pq(@HE%2+)0CYypbsfgT$@S
zD+TlKJ;*xaFC7VKpl30!zUr1=+!FD3tjdyC-RP(wnGwGI1$6JteekE_F7KtSF9>{T
z(GRY>FU-M5&ad~U-{It+PA|9dF1~elvi_?O`uF?i+y4Qwss=QdAzTR*0yi<0P736I
z^dEou{QaFwOHll+s_&1V@BV$)FT!XeP}1qIU+=;VLM70_2~hCzPJjIU^y}T93WZEr
z3KxFy-%o2Ktq~4vAo<fT^7J>p`8^Lt?@s@TL%$X_^7PNskIZ0?uoE&8Mpn2V5tb(q
z-uXm!CZW*_N&b8i2PHEgj2lx`Qh42elZ`NQieV;=(Yk>d!wWPrZ7n<#PcTyq8Knb|
z_+!)mK7Hr^DS=I=Z<~Ib@R>kTt)~G-(bT0&+R>Eu!gzI+4bLTU5VlN~JX`|X-an#M
zwL3YXy*0QNJPZIk3Dm(uv#Ya_B=H|&D{8hdu_d)l_tN!njo#S(DL8)|!8Rs;GmVTh
zVVbQcTki!jBS{dLB1_~Q0#OJ!Q6ft8Uh>s3SDx}#=7+l+7Ri+>KCdCsaEq~V)iba1
zD{H`}j|*BUck@hBuK29EmR8VPkTP&q51CiI3Sm|m{gPyIlOATuMJq)kaplm+tykAJ
zUoq4AwJcW}?ngA&$h5A#hC-!(qkxPa3EG1oTT)u9KN~fYG^UaTUXoPFDp9J8&orag
zVN8On+${->CDMhYsI$CbCQd8QtLFO|iTV+ORAHUzY8aC-nj{Fe0;iIuqBl{sim;-m
z3Ax~0qLKM!oEIN7JzZI&LQh(vkB_LcG<Nn3a6vO+J<;NCvqf89D)MH3u1}K%gLSLb
zZ+m~5pZ%B4ombg^rWD8OS^oHYru<g(t94seW72wI*2)$lxH>UXXGEZVRAzFN46i6W
z)v9jpJSPE6cJM6D6jaMqN+!)=*v+Hl4CNA}lX;H@>S^|l$}36UDg`~2(A0zjw;2v=
zYV9g>zzOprRA1`QS5vEhC_qjItp?UtzrLn9thqA-0y<&roFv~KCn=(%q||2M<hecq
zLrrSq^a`pWaR0-*X@<3{Gf)c)uATB-n!>WIO(7$IsGaibn2YA3j#F(jMZFs?g?#{|
zZKr%Y=3*pQxvl>0Va$cAOruQC<-IcI{zLe((^vWL+v)4;SVP`_2raRt`sLlIC;`b8
zy2IY~Ihlg(Q)3lWTpfrq>QY!*N#$u=mo~er?nuZaHy0ymbXAPZ6oW6O|4fx&-lo2g
zRO}0>&?p|+tfESNg&NkOdFUAw;nbPrIa`t24dE@irV}B<-5}Tb8m&-v%Fp-z7$T5^
z2z!(6y@}0d?|){00}Qc9P79^F$v=)o^ue!aUBnG8UUSt8tBC*CkeTLs*%>%TA6jzn
z(s>u_K@D9&E7k)pTtPo&Xik+6OlfcE-8rCw;lAD;(Wp|$7sC|In#E3|>)sCB4^qqI
zX9+ujhkGi<WIOAJp<Wb`umQr=l!rRe+qTkIt)u2zGS_Z@bw9A)3%Jc*z_r!YXrW-d
zKwe0*&@s?}3u1J%Zi9{R>6DO+$)RqRr9{+6UE9ugK;g5hdo8!<4BTe9HEVS{$xA1!
zn`^w>6=79jW{D8j`keM^8Fw^bBU0ZzP83U5k#$n_J8s|JmNP0f%e()5$95zDZLg88
zFdOkiPM*?#cT96|XH=|U3>DW3!3tJ@KT?lih)U41NMUn_6L7ec)UX=I2ISblP2no9
zQBg8ZXJ|NA7~jb8`&bUcZHVvI{9$NI_H_2cg)MQ3(x((@a<edDFUOu1+u64wVke)S
zFdE0}sy_~U3JA657fP`qm@%J<5RHxBs5LKAa@B)>$%hSJU-Hw)*QV%U=cthGOmRt8
zErlCItEZ}wQ)`V6l|8J|#3UZ=zL8xvNx6s5`pz=FHm%n((pAwSM~En2TKz`!BkP3o
zB~IhhqjtHC1)b!H*0tGI<xY<ZFh@`UlwwNEebb!MM7RC=#3yFbIIhT$cRx(KTao1Q
zc^9OAYJ-au<M^elKyJcnt=#{2?9H*$1D;|9NjQ=tnFF&nz3!Ww2D-Opqp><B$I_il
zvYsv0qlapT$(ET^sGD}rwAqy^+Z2tB>a15JESne4g-27X3^g9=1n)YH|28g55!eh%
z^}TFvD-y>~Pi0$Cn`-0BHv1#CcEg<d+mbAQg3}bIdQVZ&xW3!-)TZppU0uIV((SGI
zZgURWutQJ1q_cyvT(3##YPC=XLe4yNBH;Yy%Z6u=AROOVdK^}To-1?B4GfZCS-J$~
zRv9|aPU7MLhBV1Z{xYsTU7&L!Zd@E;i;MOA4jzQHzMb;jPB9^M%*C}Pc8cOquyB=s
z%Tpwm6L_T6Lq&pY$7Lt1D6ZBGt#jEZ-^d6gb3)JDWP^64ni6=^Ok6Vhurw~?g5Ml5
zlU2g96QTpb9c|Z(uG@*ZP(S6v4i_t|#&$|Z14q(T{7A0dR67NBAg4qB*vu$RDw`tc
z23wVTDI8xy?s~}kBUk_~RzOi>r9CWvNcI}0jtp=w&5>3Z<8f!lg?nKd?h<f75!$tr
z)R+&V(9rR5J4LnUu;JS&zm4T0_R@aJcyAQ7=3t=PDdW9S?)(^JW@Gsqg*kKKj;Wb4
z-W%maA(ZI^+i63>0}Zm{@?otF8SjRo=Y)x!eZP#+%8tjbfQDNqcL=P&-s3obKn2wp
zcL=N@)Z2jXnhQ*ubvq`LBvdz1?gV@<R0s?8lpTvH#&|E@j$0&0s%Ly*6Y;P;?;ypk
z9=20=e4TvUrn^9WCU`GytZW1EyfUfJ(jT{?ektD=@jf=*&6E|y+qv!-Bfd4n^Z*K1
z7vX6P{`%DYxB8@#Vp<_^cTXyRDRP3i+DMNob!3k`X2U5&^HPzMPgRT6XKS!|^ggYG
zm{j@m$Cp!g<<Fny#b_%`NTe+{klmA8`B27itb%?temtKqM?F%er=)zt_dJT+M1fvV
zG?!C!+`-|(67gXseKA$G9fv4;YpMzY!g`|5mlNdDl(-OH6VtDP1ArZWjw`$#xZNI2
zhBs&-(Qm<fWq130Xk&a>s^hc?ZN;^e?MLIs>)_o|Q?$<8k}x3$bvlvLVZV{;LMQW_
zJKRD89n9v@q1wfpOOc`331O~>W=GL(z%rbp)9sf<++C8|G4WXx!&$uEQPezGbKV^+
z3ocw1Cua??gPF4^h7%xvR7cIlJr{p&zAls?P~r{E75aXs;D*4CJw6=O6ap7(c2!eb
zF_+tT1@qH%o99vIM;apGe9vGj)aJn4j>eUJEBQP<!yZHnPo0O~cz!A<Ik)AQc0U#n
z43D2`wU7_yW7_3L{IpdnUy^LbvJeVS4E-cqbEQ^YGEUg7Qf7XCwg?YxTrYGRYJXZ#
zlh5Ph$2k(}m&?LN(s0N3nrN7;&{KT9f}#^vngzv;a%M_f-QYG3LG{SlHKykX329F+
zn?9eEytdDK&YQ<W*Xk%9H$PUVP4hbECxa(i`sI9pAh$)%E+l)p+?-^%&NMy6u(c^)
zH*%(r&kc1mj<r^Q!NQf5We`Y*KiSI76_)dxH!nvG9&n8qp4MBQ_SCX(wu!)9sD%ZU
z8_0)-8(FoaJZFY8thFhB4pc3dD*{Hren;{33W`&6+*9nE)H6IjTaA$j^C^7r_=Bf!
zqyC|zdPdM+`1jQL#ONchMpUj<JNeE;e6CD;DzDeemnzhMp7L!GACAeaUufZ)6zN@j
z0jZ~WIpTAb+*9lkf4Q!6)x4r>@`rGc>)r*0t8st`LOMEtgH72&9ZYsXVRv_<_O~j_
z9Jd^Wa!D?+@U|$!&!Vcr6Q!_&*)lpb=CD}(EpsM0?vToxVfmRBvd3~}rUEAFDj$gM
z6!y_1d$II?WwqmqMXXmfr*T!cU4bkmcvKIoLac8&AA0*%X*P$)TfTj(tP)C@+rNFQ
zj1?Z-HE(X7P2!+rnk!S||39zahIc7I+j(OqZOVq%Z}Hfxw3o}bw{B!zU1g9nDIA0#
zQ?*Xr;!WDoDbWgJ$qfgY#Ssy!AI;S#MC?^ULTO_nR$gO0dmD!npyKe_QQp`m$&Erc
zIv0?`Xw-i*PX7m~6f2j}XfzW8GB`Jv5qu330y#LBp~)mDf4Pn%xe@jIE7FvVp?Cx@
zU{FA2cJ~15Y1Q6ht(8`WGy=4aeqX^e$;hm#u40Z+UDJ8QK|V&j2(pJy|D5>r?Pd9R
z`OCY4$ug&fmD9)XlfWAse+rHljh?I!%qTH^e4l>#%DFW7S?~j#D}^76|91GTFJF{!
z_=LdUgt&aMe;~s1@Az5qzDUo-_v*7zg2s=;J9E6F{MY;x$B)4$jZJd#n(*Os`Ss(c
z7oJ@Z_$`|6w~v?qzECH$=S=JAx9>0EIw--Elauod3`~FgWBUE&uT?-1%$W=S!k^y`
zBpPd&lQxL_<ySua9dG`G4=eAc|HVhYS2cY4_w+YZe{fH*V>}90INtu2U_MFaozG%7
z5n6f1!9OnIla?9DjJqvWaOO1?8&T#2qKq4Z)6k*}(IvG)*_wG2o>3O(8Rs7mG4P(g
zb^cxB4Q(%93=jBHK0jr`Rs0~5G(M`|5Jf({9j*bS`%yA}<KtYsCL}dpcj&3(`K;r2
z##0Msf2BBkKs~cYqHxVP_jN$S;=zFv>XR-%ZQC86fk*RxjL!g!>!)3Q+O#_^OV1ov
zKgeaJ82937pP#nvj?cm|2XMB}nrUv)Xvu_<)#tF1)q{RNZQmWgLw0cb2l<V$Qdl+Q
z_tW;>_FH?y{OxQwfN{>56j1$s(C6oEyXUiFe_pwfME$Uc&vBC=*4`F-jhj#s6sJeK
z37L|V+#C@^9WssIQn$;~6G9POz66+lP3$ga?~se<l%r!##@Phqldz^79<ISVvOWl$
zFJ1{5JiM7Re3YWo8-WYJ9C2`u$7+m}(%52V;H^ckHieHy2H|=p5iKiL2^*f8v+eag
zf0#FJQtgAyUa(a|_JmlE^c3T;3$vt+Gx%*hm(w3%k`M3v{H;#A1i_pz)Wg8MqZWH?
zQ<x-pjehlfI!m>PQ3M{5ip+l4UI)wzxwE&7`J6|lVQy*sY@VSwZ?2B0$?ruUFH#iT
zxbhB}E68|>QD=u2g;fFfYVUV9TG!A9e_S(bMgO)?i!40$aI8UCWLxgaSjPo;3)z?-
zl+ss`4WB}Gda6Zsy3#&7H@JoDQJAH*U9a&Cku{Y!^`^b9hxCSS-w!ok!YBrgZnlGM
z<te#rm4W6}5@gHlURo1lf+Sa;uoofQm&*c<zq@7%ZU-0emI0T$DF})j<N#5@e;kI}
zsXVyg%QKozTzb=867ZLc;$O>BzDQYMzUF}9iq!u7G~xB0N&3atqoA67#hmN08k1NX
zH@-^k82aD&`ltW=`PT3tmhlxjI@N01Ov&c&+Q3KNv`BG}#iQ#j`qxd1R9I<viJCNT
zgIZ&WO2fl(xFF`RSb-M&ug&kTe_cGM)9L!D$ag9b;vJQ*thsuf4=2`cSBs3O({xbN
zc}Q)R<4SIdgSmsQ(+t4gogA@SV=1<lirkb4Ova+-)i@S;ZIi28Ecf}DCOUN<CM%-+
z8&ZzOIo(nDj!M?X2nL(CHFlYrZ^{x)nsaD~_AG7oFPUdUPaUD2VU!GYf27?vac!*@
zn2fY0;)*MC8P^+}!g1V&zUY^^o`J#VaeCjfww@1hpqlY}yVUGl`IBk7GTuavi(lW!
zZ-+g;n0jH?<O8g5g(^)xBDK<K&I54?Kk#3X#dKGG&P>?z<3a_QxP}P;nDe#ga*gU`
zrB}-yWzcXzrODSMF}6iZf9Z4I?7KE2<?t0TC92hRqT_n~l-Ynx4_QbrP}&CKkXmw;
zi1Hd<PV+{FYDt{wPA^gB<rK=9@NadZ?Lsmx<(ank5^F)wavYtS?Q!XrWTDF(9WBJF
z`!*v|R^!cD6-s|my3)yADCq*!#&|ZI#K4go!MrcT+--&yn*3f;e|z|yqR|VscOu2N
zT_xwPM@FGzpQ60M^8lDG3m)X8nnkujxt*8qWJKi}*0QTfRm&%A6N$xhk=S7b(2@q?
zP*8e(I}qRO2Bl@9G}cIPk4z@JA*{agBQDDBN>h`TB}1f3eps&wOez9)wmcN<9<VKd
z(K)KE%nr59YmC3@e;M2Ol5I1-#4o9nZQYey!Av3qa4jj2OuE{u?C>Wm7rh_I@;LX>
z!ro57Y|l4ia2WOh@hhySTCeXWeaaV!!5FblgogS;y=9ZFLfb_C5l!x{bZ`yaU{Yae
z$y)h`iBw{%mh%qT+VH%$6g~|Pl}>mz+~vCS;1O&}?wI9<e_A4z+yu0_U+bRNpnypw
zcN%Qd37&JiF+|(>c}>PK{OmOG=OtRnwA|Z1XZuC=LFKmTN;EwV2+5f;rUis7>BJO3
zNZ6K3x4~*!aKw#%7z2kC49w0dVY2H>{@CX=fMGtYssjz}0J{v10biV8xL|gg2Ec)Z
zH-2^!A%Dsef700(vpf3m3_v*2A#OW>n1c>{ZdWBZM9{1WN<*;%7-&Ilhz_=^co_g3
zQVi>M8f>#Ox@1jD+@5$oRsch2?KJW6=tCqHdqvJRVYJqRIQK9Vq<0L!bx>GZ>e;ji
zL<@z`R#3|+P#`u`C_r}74MX8b!`cAO=Llhkp+Gf_e?x&vT$|lBlWkuQ7R=*NaF&7d
z9gk@D<=)FKOw*rg??VR<bwc9M%u1kk)ivz$$2_k=hjT|e1*SpA+T;y49ObE|k-XuV
z-jB^pwqAWPe;6FrFs+WT5t(qy?3XTMgfo1v=Y)yQgboHU27?kMdIhE-DK>wWF-0Xv
z^omZ1e-ZnkfhLfuhHa{0BpA)K6&(~zY0^`6+2)Q?_BEx)2{|OQOJY?;M=222VRlJ1
zv^m5!3=M|aCDk;}E~&(o6ld#M%u;G#M#wwhXu32#$2_a#NL4m5reeaB<S|f=Vor8;
zr@<T$SvOXW=HiiR8YxE&v(_Gt4m32gqz^bHe`tVWrq3VJ_Pbf>#KGwl6Aki~>aS4H
z;}cX^3eiBJFdv><e~hhoX7z3?htfyRM?~V9Z7sH!--kOBQb}gEeW5Gi;dN_}mrnnA
zp&_@bh$Xd&v|eG;kGHC*2GgvqLM@iDZbZrcMALYyib^bG0!YusAJC!4L4hgUs=i$X
zf8=1w`a{ak92<K56$-^kHy#^^2GeNKc^Kz0>QG_>(R4y=AQIOG#&gq@8cz{|$Gp_S
zptSTYIS2yVh8etX$6@+LRcLIVQGWJzb?~sKl%GU_IcQD{Hb@H%B$~#{Pa<$>5=Y@a
z%NgJhJ{;ia+f=XZDU#*{sK{q%)`nZSe-0deyo;H^LG0RrLyv-kDriH8X=nxvgM(@s
z1_u?mGzr8v{hQ&ULpaBr_IFrCyVWVksY&xAijdq@Al7vdF=NzIhd&8{od!cfbP3bz
z@Qe~g0}u&+oZ$Bo3(?R`aa!8Ib{HqtGEX->N)FNGj|=riJsP?pQ(yc9=Gj@%f1zP;
zZz}W7UOw$*`uaB+qpY*nGtN1Cc}oL=wdq(4Z+Hk}m<MsV_f^iEzHcW-%So693L?ER
z-k%_*bQ{_0AQn=8f@&~Ld$=sK`5-?(F?&SQcz=RQT#mTc0g@j_1cp)Xx}KHv-Eq4z
z^K*MOVdmUsrm##3O)WYT7$BL&e?g@hXt$!|W;;HOL=zuwk5Y;2L1C><2^F9jm;0ub
z+YTMY?2!+iH6enh3gm=M4`M@wLMaNyx5=pn+d!cVC!Df3N;Qq|dr*l>vt^^QFYXj5
zU~=WAYg|c}&Vg?Gouz|${w7BOLvKX~hIwp?3K)n6({T2=`M7Z~5KSk5e}PDZJVKuA
z;ptPjq*P4s{?HxY{)>usWBj10EzW^MZ$k$Mcgl_^)nLf>vK>F6QQ$x{od5?Sac%!-
zyxGoOj}Or1Bdh|cnt<B=3MFAaT)KOb3x?%PJ?#3#c}|78q@j`mpxOh@dY@UU1Z`s@
zpGtUt+!kbC!pI|AKKtiWJ6CM1Bxl&h+Jre6ug0cpcI18C6YMB;_^}=L#-k9W{_P{A
z<-_luGhCom&3fYxH0bx+$IJf#MexMammypU6qf<g0TY+m7z)vUS<b(H`}svy6|Q}M
z{`T_U7g9!RDWcUU=e3YvPyhV>^2@LC^e-Iz`6Pvj;q<>e^?RDGaIQW5=k$-4zaJr_
zEr=)sE^0YKfCvF<Y!Ml|o?oqW5-(fS$LjixKmY;*P^RVSX$)c05fTD@`1}fBY`s)x
zB>)BmRSY6Xhq5St5x^ZOs38g~(~U*L9h7VoE~+P?1Wx$0s53(QKq#4z6yd7Ni;SqN
zH{oDd8>bG2RkZT@@O9KsVTk%8s2yo6qwqfBstBhB(pbS(fQC4DVT^wg;>HQB%)Ec{
zKuEwW$@_e|vYwSmaAu6q4dFBT+<O~VdCIFnBTw>Q5a0BFjh7V~tuo2i@kqSQm+4X(
zUb4Dny3H>8s;t)Or>w)iPBpX=e%6f821gS_bg|)R4{$VAn4q79qZBSLI_$6C&l#Q!
znj^4M#L6bO6qK|7CbS(v8Yp@qFe%<8Ra^z(Qa?V!scDE~6e^T9Lum*~sGm!nYU^!&
zA4=^7&TmkEx&tV+5ul%Ck{|`3_e}dIIN(zJG>1(n048X=9j5ewpko*WLk881tk@-i
z=IbTGbgADkf<{BCW|Xwz+A_-qG-ZXf-TwRmXb1@Sr`cZ+O3&M;<eN>e8dpv;CB|{z
z#j6H^0abt@%%Ex-I%cFiK^m*U`4O{~GJCbnSg&1w_A!%2C}|pI;W}nEVq{hRc|IGY
zo%c^XqZwMwtl+z+T0RZ~kml^n8+omkZ_L6%{kYL<3CxvwZP*6Dz>Xx=SQ%P)n*p^^
z2?Z^Es(1is@WQ$$8lm+t?@xdx%LATmFGILXH0~2xA!;!L2V3AAOD?{;soM{CmPN+p
zYPhq1J_;4P6)p#$B@!py^Q_QIKksj9h)XnY%o^N@iL`VW$(b>01P&u*+;!iXJxd3$
zb><73wvtot!|1#~J7_Ro4q!x<=j^knqXA^=YTkbb#>%0yJ1;YQE$KRAIDC#=xN{g+
z6YSfRV#1cgkR>kMjT_6`KxnPdp<Cj50HMWyf-64D5<v^&-MoJW3~DT-O2h3Y5Ep@i
zW;+_(b$L#vBoena7uhuWJ(E5VK4827BaH4gm-aUYpc%v&o@RGmAlH~>*s;4$L<5WM
zZEX&ub8n0ouPtj1@GZ<FUG>c!qGu<1bJ=m(3ZGGU?s|alp#HGP%Y1mE&8b;}9O)Q;
zWyw))pR!;z#vvm`!w8vS1{qn&5pvt1S=_=b&Ns%pflFZlh#Z8&bx>wlq-4Q7&Ds>o
zpk^6#1qghnCHXh>UP5-=b#~p-Zoyv$><py8PFKinYaPOYyfD|LaCq4VQljg)(e4h4
z42v`^_-FNF4B8zQ(UT0xNpq}orNGU97Fctwjf@k%Z=5)m!o!#l_kgtf7U@__+$lVO
z6&}PO_njD`6{ySr=0WiRNYUc6Xdeayg`sZ}VJ8H}v(GIA%S_aDvjn#(=W1xAoXb&l
zHTN!KN?xYy;_I|^o02U{MpO;%XBY@ZSUZ~PL74)`<)V9<sTvX0EP{63t#hn@M5Vy?
zxXEIrE(qwv3b8<(akN|O<W)&*8DV3cc1~0~WpTyND`Dekod;D3xPYbqv*1IruGB0-
zYbQXpM{1b*nsdU~-l`a;9!@a&ZR%gG1w$i!Yt;mnbEvH~g~?5@-JygPztm*~HQ;wp
zpn!D(wR)cY1wD&^zMLD))DBF4z2}P6a<*X@DM)X#H%17!g5)*6NI`I?;V|<ebKM%j
z3x^Dbqj^xL0DyQ}gIgh6)UyP-8eBF$d)CxvrAGq{I+=1d5+AIyIdqhar~a0mFo7*}
z7gkK0dRw>-9eR^DqahyDCy>%C!1Ee|sl71CpmSe_^<W?&LsiO{Pa1E3MB+HZw`_Z=
zFy)eE+ns^BN}L+rey2NAA=C!Tyl#E=xLN`HRv|vG-H*06%aE)JY7*h?t(`J2CG%+F
z4Bk#_i}$R2ZzaxdYbM*fG>f|SwC<;O8;+hrn$&*skt>&-srB$s1qhX)WOW1a5M`Fn
zm4_52dh3ZNRxNm_y=zl{V;_W+403#2IBU!)f}Z=mPgPFH&W|%v6-ZH&o(q%}T{jrM
z(%S2!GN0yB<y3k`bdT!ezG!_>ZT}p6s7;rnde$SiN?2&hlAB>KVsyUguyDDzNI34N
zj`5g4k~S28WErvHdbaZgj&#XF!<{+fm*+C{1ct4B-=o)J8(27hB%-GsaN9`LEHaqE
zHvaOrzusj|PGYWYT?B|sfg&(a!LffJZQe=`oN_8ABO(8OFOQq+D2KCMAu`n9EF*V5
znXa>w+eMOTyH7LaqR}-|E~B(VoQ~VK8J266W6-A(NbrnO>_}f<reEVTUOEOP1!mY3
zbJGB%_5@NO6Y%7J;Pl78PrqT9omS;<NnnNUg>()>GCyQoHQkRTVt!Uw54l=XdgyiP
z|NQa$X}a?DOP3SNf8)%&l9of<Y_qXG3utfiz6AHD%62c;jvyz256g)4N{YhMu9e_!
zWDHn`n+-N{EXNvxsFCZva`^l>g5ydPf*)FuP##yT9gU5DzNCqA$YuI-ck{R=<4U(|
zVVupQ?{z!)xNIJ;C^zq>)V^#Ug!5`6+g%~2*~&-c(;Ati-kgHcyYi(+4BBk4Yu2|H
z4itVX_PrZ!w8IMDZ^BLDXjr(`a6>C}+YEWTfPe2S89*G?f3yj`qjGCmEXeb+c$d1-
zH_{>^U!7ln7)o7tOa>oc;!erX)QmMZXmX;|^SXx-TM3PeYHv2>H!$AnSRHlm$5*TA
zN)5Kr;CfiE8+bX7VFP~XBtN>uou-S+D)*4_&OVZl_PAI`jbPt9>FH$~u#EYv>_KaO
zWrvD~fz80cP#hzg)Z!#A1H@EXy-zOzh02`DD}8}~HKH%=zVUUzdl;owAdFV?3rUsq
z9<G3eCz|-Yf?RIgg%!V_Z_27`QXxld*&}+XerU!3tR~Xg4#W0JqSZxtmZmx(QZ2(c
zj^Z<q530>NI;1jVJ|sPm-YeQ%rMYEv<vTQDZwJ{a`$3!MSRGU5@oKEiX~e$Od1=(|
zqt0u8LDzGy9p9g5h9)&egVs`LdK797tq6#$=UN|Cfu$!Zt#Bxn<DsEpVAPv!Z&@dn
zWFW&v*mqvzw71L3)v{hr5Zm>lMIm6bUV>R3N|L)lTS*?RSX^>9m>Go9PFU@ZD;9}l
z2Cu35PbN&mm9KYIH7b-o6&-}7&%~&2Z#Hv(Rplb`u7>SRH?I_y(L_~W({fS`??#s@
zx{vC&={|dSpXK)@7cGAw^{#*A))+mMvc20c*T+F2xYJv}%FK_0_QGnBo;!He^p1d8
zXWj^y6`zBESuv_3sTEbP7}n{gMwM%$n%|v*NWJb@pR!8MG_3)c&V{|gw^GZkt!e6i
z4AYC;A2}7{kgfI7Ww|Qj<}{9xXd$zo$-?>}1Ip9~>6lOFJ{M5F2KU-gQF?3Z+j&is
zIXMs5&R)2;cDfFxSy4}1KYSWk-kE4y<S<+B-j&C4dE})hGv*ES_LyaMo*E_Vro4Sf
za->gS+|6JrQ%!5rKeCp(!<Db8SsvwoVc3>NfcHbkF?8?VX?ylI{FQ_NF+cN}SN`+H
z?T@w%QhP992|H={1I;6onmAg`EAHoy-R6V=5CQzW!rVZsDK%<gRT${cosd7!0}3J{
zQ5(0s)}E?uz)3#la0vz;(sxZ}9dx+k?u}XP!xogV{q6dXSlq*~ja%M;ZIbVQBe2;Y
zhRt4u4Vk*z3j^}f-=Oal;(<OM^%?m_*PyDn@~J!wC7H5ZA3riJ15)}?3bikVR$gc|
zN&)a-4>OAB!?>${I1;6I_#yalIlHd}==>fnpZbUb;6#igY@i-jz^0zvk^EQ6pzZ$&
z{XcO%=%fl|Ze(+Ga%Ev{3T1AW?_duF0x~$45h?}~12Zu<m%)hxCzn?x3J!l7XuK+u
zsgkrKSNW7Udvp1e#9OXp5Ah}W`DqYufFebXGzX4E5NtI1>n8*ZKYTmz!;ep=uTMW*
zBpwX23XC{>{p|qsfYV>#)Tq_L05C1#@b&xQmlw{3rt3r(axNuZhQ3?+RIkqxEcF1o
z19*Kl2)_S?t}{J{&~JJ#_nm(zD7pkaIZjWcdx^a`U7C7o6X;SKIJ!=+zkdDciN`5$
z=CuFnKl;13uTTGZ>L;YZF@f^%_wP^s0ZL%fii76NkOcnt=i#@fzxGsXaaka7ajlhV
z#3hJ2y|1-n(mx8~LYlgQa3M`!(PWV(t4z69JWPlok;;Gm_;%=8e*S;4%*7xhvGk(}
z-iRjmwvvHy18)Q)l28>1mHY3PcyEHR`Q`Ola}Q$kggEBs%^V8_OsJj({x-|7P>vZj
zGEBOsyzulDhPSgNxnPpZfv+#Ke323}3t|&r`@|)cq!IM3d}#=(y$<lir&q?~9LUYn
z!ug=&$}&lok;XCG6<B|2vO4G~bqc(RA|kV{CkSdkK@WkmJ;f^!072lm>3b`+!?vjf
z^sVC>%bYi?4iXINnEa22-`^{YKF36rSoKe@UN^id@N=cfG$Ln;hd<MFnwEMxNoUY7
zlqOF{d%Y7c!T(QsXCoUXYov(`>c|kr9)4mrz3-nWk?-O}FiwATn#A;e>Va>8YSwoi
zdsQx(;#aFSoTu6GINzp)=KI6N@x{PE8jr>n#VDhy6bRZ;uZ9|Bm{xL*Vk<`val}iz
z#4_PD@!zy8@t=Z6JDNy}V{YTwDduqCx$-`y(yn(%+k0q*2Q3s+;IcUs-CD>UvT1!I
zO67Ali%^ErA_0GW79k{~geheai9#9APQzLHK10$8MqbR5{N-1E_y;}wi8^~9^xvuL
z_dE<g{QK}XpQfx3wj?1UY{^GZFj*kX4}v*zpMh*4p`>FR-F6Z;a%K#3L}QUGLF!ab
zcPSMEpK3_rnLcffhCZM4ULm<WfHp$rbgXil<Yj_JmPvm!+K|Pwdmd}@QKazhUC3=E
zaig+Q25OhdUQ^cIDsCu36aq6+W+;=Jqe4P}xgk`9UW%8*Y;wT<jgx<BQ?9OaUr*^A
z-l;=GdJo&4B2iJttTxwX<<nb4(a(w>m*1icgM4NG)|6~svK(|t;u@tkxyL!(*(!L(
zXBat{3~PVjKnhBhWzT>}ar9V*leM6*<dhz?oE@vh=}LzbF|Z%)I)p70CQdF}xak0@
z1Qs~VEp=orX;7^s&=qiU>LDOS8q;RYNr?qjHC4P$+*w}@Py7<+kU01f^O&U2=LBf(
z?V>-sM7i2jReLnh)pyBQVhfLb;v75$i0S3lY8rnvd8#1{qN6%V?k}ce6jw9|=%VUo
zU$IO<S%dAVT`QI)MPZ?b-j_Nml>Aae1Tbp9$8OR~+|H-<8S*geAs2l|F6YRFSUWej
zcUbUR?k5!GVtkmFFF)klY<-DNGeK|6)UoqXIu<jjE!s=S{kP)~bJdQE4u?ED&KWo}
z@LXiA0mxKw@FwHE54UNp<VJ2alDWz`Zd2HJD7TVqkus7Sxs_fY<f>@DJc?N|Na^5K
zX1#HtUKMTP_725PG8e0uz_RS);)u7IXIwbxiQIv`xPUTH9xB~iK)K4k*5p+NApq`|
zKq?9xe+nu*4A*y*rU2pN{%yjf42%g-qcGbr5z_-C(F<GFylSi3UDyjwRpVIKqhOY-
zWLIHN&Wbq@_b=>G8fFB(r##$D;zrJdW`;f=QrM#;V^N~4@zO;pxO3@yGP~_bTUZv@
zpAgaYRxgglpRCNZiW>ZZsMaFdD(2eDockXZfBcoR^p*99{o-GN6%<wd`p&Ql%LFA*
zK)kz>_7aMc1ApQ6q;LBQV=GW3e&I%3-bulKy-o7FCA1=WPEi`oy$#A|<|KL=1Y~i}
zq-t=EpT{JP_aK56^asd(w~Ur$FP$XkZyzIj>8L6)<p(Q_m$!a9i9jYFcM5*aHwSrb
ze};r`FdN{ja8Mg?*e|aY94rMQ;U0v8rL-MP6I7DWRf|K6H{LXYh6coJ+N02046I?I
z7p>qbCR=P{zwA~pF~Hp2-4d0wl+sH>S)j6wY3Lx;1c#0Y6>G_G$h=)7H>(P%CBOEu
zZfmy)SCG*_r9M2!S~R8J&NUIkR5(xwe@t6}j#K{A7?3WprAEx9N2OU;wCy#!+8E?#
zz}t%0Zqcqlrb1%-7|47%j*Wk_QL?VUgH`A^BRuC8Umfktv^HiE&Mg6*Fm8a{ACopg
zwp+9-kjap>J_a(XI{hx0J};Betwj_tTG_d^vSaj(#}*tJiI~>PPLYr*bWLh*e_>2L
zKQ4CdmhlR9BBZ^K!Hz0=6k^CH<h;F|M{`m0_U3Xz(m;C33XlmTGc+xvk%JGyZnunA
zu!E2`KL$I&8CJi*m@1{vtuW7HdU%eWwM#r~1gUCHXsDL%d4#i$njeSVZW*s&$3v?9
z80@&4er>Al&P2(n(e93|HDQ&se;f-UN&I$11am^##JTh6qOk^lIa`6d!nj-9%P<NT
z(({MFD4b=gUVu*K*Sb$tMZ^v-H<kpt3VoHW>Csq|WuC_rC^mnL3GSEi3U)T+?~lRG
z8m6on>b7o4BR5^QARL?3*sR9pRJotVu$oqh-9mkLSNq#9>lF;ml+Zs0e?p^~7Iuh$
zsr;_$J3D5cDS){a5SNO?F#_7r4X0)teQvbj4}x#6m{-u#A-8|X%8Ag5X<f?mW0d}}
zwLTmpof$DoF-)CfbY{W2tz+A^Z95&?PCB;DFSc!)9ox1#wrzEMbN0UD-t(*0s!^-P
zJL*@hsyUxo4{K@X=R3?kEUChFiqo=zF}e-BVfR|QTPmXejkSr!qH^A?!csqPInlzz
z#hK!x0-UXr#9!*-;*tumb4eYQ7M0#s&g!kgoWodN*bX?fi>M=tAf5z7gNltu2NrzL
z64XeEn#(ePcN*(_<&4}To~&5A49&!XRSwU39pOk6aSnP->#5H@Px^tYAUAxEYL=l*
zam{-y*}1A{T+h&hge}3yndfqecN@Rj7Em}a^eb8+$>r0tRe=M3$ighny&9Lzi(3r_
z{b(Z8w^+RBeKu?Q*&1xNQM3qxf^1QYtYZ0in6s*;mu><Pl}J$+mC!h9$chit^A+T~
zBOzLPY%PI4v@@M|n*AAJ4SUR7@BE?=S(!r|ek#kb?4!$mamIzJ8+MqEj&OCBDJN_+
z(Bc5Rs~1&{`oDnR=Z-Gyd90K33s;&5leqQUG2=MPV}I8JGLCX&vZOX!v?h~MuKgf&
z$~_FYRzbdoYN>eZ)ixE;W+~qq<56q@+CV4ncZ|7ZW27`9GKc${d6S4bIr@2)kL@a`
zFy|_$DtT@aZr3VzSnn0jROc?4!knL<>ea0p75CKLAGLs)&Z%J+J609=6Wjqmnu7X0
zgf~|FJ@_02v*P;CoNDT0P%)U=>;$R<xrcIjpW<#^RK!#L$`=>zbT;3}Axu?suJahv
zq$!4@Lo0R~@s83|C4uXrs!x;ET-6AdWBC&Qrm2)nQhQt0*K8?qb_asl_@RWdt7%6A
zfxZtYG$TO!)Nr{&op^ChRpsr~2{nH26%_hzlT0INw^JYt6=mA`vwaDw>cXkKS}}7<
zW$;OOyDwYcV=-&NKQoYKo)IQs*D2n=?=QAici(Qt%5ZuQYoJ0jXGE8ZfYQ<Vt_W$M
z`zF!3g?QMKZ1=3<&lz~d7ZkQiD_rw+GtU`&`gMS!Nm7IRXf_3J<y!GxMk0Jx-q#Cv
z4>4jUoT&CvPeBF}R|*)6D9_#5i{R(QLn7N8H$~IgFmoMwqrB2AE=HN8#nEK|i}O*q
z=I&xTw<C}?gr5`omckz{cE!y7_6w~H2M6rlbFNSxMiLMVtkc+#oiwf6s%fv-AR|S`
zG#@|*FUCy{f3NQ0WSU=QIj0YO5yO}maBQ@pGukH(hKQCKeG};Y5pc7aIvgm?{Hp^x
z{I{+R6Bpb~ideK6s#zH=l(O}VP~~@>;tx&BmuLKd1!9~#;+)Q0^~hBtq&Q0ZdC}lb
zRnm!L8G%aA3ZM_U&5dYlk1#AXep!8Q-f1I_!4d5DHd)9CQaR*9avm>yZy-fox0(i3
zX7yRR*w)p%e^;JMRDsePDF6}XKAj*LipF-X^!4$oh1Ga0&&E`ISB1F&D#uJ|IcUMf
z-g&Gw23=$&S;vLZKim}4Y+5Q9i!E5;^(MF*%GLN^1wJ=h+B`GNK3H(k>}Ogl3rq=s
z=-ijYp-DC&wV}_oa*ula@twdMFE{Gn%nO0Hvefc$Q$a0^D{?d?w5ozMw26EllcJeo
z{0-t7t<a&MgD?Qp%+3t$6pOmPgk$pNVv={dyR2YkCv5adqm?rMu7Mxzhi4dh)A?EQ
z#k41@vwjhA<BmaH?7sHELmFbJyt@pTk?a@hY+>`sM`TPPPZAEcBVgi=g#T@0>6BlT
z04BOJzZ&$~ze<wXb5>x8qbB3J^_YCK<J;T7#U#Xh1}>OiQ~s5?i+yfSRa@T6a3!W+
zB}g?QUJr4%-GD^c=v9=#Y(|@wIwd7HM8M@E6d})iP+B^MNUiKRjWiue5w-x>#t+WM
zVvtff45v!K5hnPCw$)JZnjudpZ~$e$YNpI1Jr>RYfS)wXvv<i8M+>EtE{0=>v-5V4
zZ8yq+@vP|1euJcKK5=xt)Vg1!_Fta2t4S8=Tb$`DW4%J)>2X8CwLMST+46R^mzJ@t
zwPi|xYAt|ZA&CB4eOllq7?lH{-(ka&I7m@pW(^>=?KAn<TcM*HVs1R=2`)D*%UZP8
zn~E-l*ues*mT=L+gM<a~wyG#7m;)XzLp<lIa*_IH_3>|nukkKv--7Jdpa(L8u9bO?
zWvfDn(Vi(ltdl6%(5ciy-l?u2UZi8e4fSthrTHnbviBUEW`T%V*$@D=Jac!(4U?|G
z_YbGRjLLjzu~=S@YqZ$VPnt6DDOhNVM0K8kw<Fu?g_~M1^glX_TYaVac5vh>uES;a
z$~u`yP}INHv}H0$I4mkf7~8AYit6iD!duwr*!bT_T8%o@=$Z=|ED|2EAQQ{#(q5gi
zGx{(SL4S?CJT7hw&!7T?I8dy5z8Y5}_k-zzYrMY~PkQNEiY(4!yDU#ykKDiv-IL{b
zqr~%zO^l2{4cR)6);6Nxgohc{gYOF#&-E^qftFz0>Z4)_Ca|-n?Mg;H%iV{V#J3?a
z@wYrH1c}f<E#4ZZ@Ip+?VyrVBl7d%H`|>9pjiFq*A=*YB+Uo$u-Zo(lTt@*E!84~3
zBCW$ECEM`x3F0*VhEmkCnob6RbH2gO9JFTt*oRnjFM|kS(cQaKS?FLE9d*Dev2$|@
z$}O~|7tp%P?9cS4fFKNmz5Iw}t<f<C^n*QTfE?=O2+w((_E%E0BOM7l&$au@i^zzl
zccKACxvewkGB|*OWIR99sK7!6)|ndWKYw97-*_=;Iyv2F6Ld)Dt*xXq;oc{-Gr`MS
zFz%0jpIQ6@(OJ<ZNGxBGedDVAxjh!2@{8;16f%|{!ZkRKApt~xnby={h5#kb<u{@d
zhnqe_fl3l8n1cxn5-95UqaF=7RX-s4Uo7hvjrvQ-tzZJ0j{Y!<j?VoP<#Ntz^$)NM
z_K?%G`SUzsVE%S9##Nlnek}kgsiNB_WdeOA99ggEh+1YM^M*@8hw|LwWu)(`gUoh{
z`me^Co|Tb|Sdg@mX7-@N7R?w`(-c9(u(X{{xd6X_f6>ikxHy=3O3R}CJW@4Ny_l6K
z(G1%y&f)`*m{8wfI<Ub`RNUMPL}7xSvQw5IhiPn;nK+tw@GpnH(Kl)IvtY;;Pm6f;
zZuPD@-E>D;(ieE?*>t#-Z|L5l`bP^!zpquEg$Gw~Wq>CKgPtP9@@yZkA8IR=N?v0~
zCS#pSRiTz}c6=14c%PXn5^mQgXIwEASH-4m`^yB72Rg0@pQ~7I1d^g#e=ibBL~mBo
z-X*h_mpT217KV8F)SYheZ)4J+tmuk<X5J%s!bX=Fa@>VrL6bE54YW@`7Ov4OD;&0Z
zL84)qI;B57wZ$&m<E?)Bcy}^^N!$uDALEN)TRGJ4EBy)U^zt~fRd%3Z>}ilmY(-rW
zslHjj(Dp03;9RMe8mMU=?3FUU-UPaDL<Wzg0<;d6?S%3ov}m%C7j7Cd24#&nGwp?b
zEuz!H0@$>yNuwj$ZKM8$d9Ym8WRd@Vs-vLWUt96^jUG5kjC51GsNqau+-7>T5+2pE
z#h4F{HLPHKd%})|hpz>><?m)Z!!RDr=BPCQ2v3(QecEkFYS5k=2Ypb(eLCr-bh=U@
z`;w}8sddFitRm)F^8gh$uHVtE3Zo>Kl8y2WzX%W;J4Mr6Hc>pm`oTSPAbpk(t~|EK
zkkl5JWmS7Y%b%TAZ@G1^^wV{Wp#xmefw!C`=vfiM?OHqPo|)YGc(mVe)4G{BI#-y0
z2HhLWx<US?-UQB1?H)uMx6+vY)cs8k5|+27k^}u0(2Z~bi%xj$)#jLV4v4gx65S87
zw5)%F2k#F;Ec^Ki+j;ioc3e33e@w4eNDTo&wMR<f*QvKdsTbNF8#{h2d*b3Np!$2O
zswqAr)D74xC72il?isAp9sGqEP_6cW89$*+&B<8TO}%PtA}B*|FHGMY+Q`ERIkuxX
zFox@H=r@dEclY4bkzo2%v+!qFFp)DpadVU+t<|~Ql3v~-0T?HbJxjlL{9Ydg#@@mK
zKhT$*g3~%t-jDmb$sIl@O=8)RedwSP&jxLhDaJ(Z3zW44FdED&TgHf(?AH_kG0qvr
zIL$cStDUI(WP((GF`_wA&?8dzLFg!xbIb5Jv{J5<q9;L36uerTui%h#rA}&#ngIBo
z5kEVk&qy=lB-t>{8E2xZ=aAEFk&bV%5(fP7LG*59MTH!4gVCE9-|Rz<6hahkYr?lD
z2gx{mHr1BwDr&<cq%*6T6lyx4Yl2u0<k9G_B>0Bl5Q}GSN;{mn*O|_Pmek_o5D%By
z!cyYz%Ry^_wZ<3?y`0lb)bxen^@vEg!*ys)-)ES^u_^4JmM6qpQA<M$1%Q-_LEzOE
zO=9I~$3)(?omC0a{j&N5O?`^KW(CM2(jS62x+r9sSLe-1;dMkvHl}lc1PKPqA;slB
zSIE(Gbz4dQj824sOv6+AriWin?}$Cf*jauI*7rih0KIQ0sv0P`Zh?Qq0DDequbjsm
zF7?0lm(8e1$VL%h6!0toFOmKhv>`t5Q)LHS&#?{*`Pd|_0y34m4USTvtcW*PQUK43
z)4+}`DIlFnmlExomaiP}YcE%z6&gIe1mjgPW_XPOeRXV-UP{-!Vi!T55B>6DYT>VJ
zdc;h=`39`ddF@}rAm21qrrWh5BhGs#xwi`yM_f;>Qm3u4sL9m15*G7CJ$mfGN3QJb
z8tt~@vE`AGis?yDf%gRP6PsK_NZdI3g(l@>@Vuf{Fb1`T1{+&|j-VMhDiJG4<!UpJ
z_Tr}YU-o4pVXI<?mYA4ib`B2Z)rk@D2^?O-QDBchaJ(kJ!1B=CsNiM!pm=VXPWuM)
zK@o!x_~8EYVw?|(jyil+jp4FuI2RBm44DhUbBi_?Q09-o2f-u6Vt7Py+NUD)zq`Q!
zF1r|W1z<_a{67=^jLm~`*(C&ZL94euP}Eyq4Tpp=|H*9CGBzd+sj;ysuCB*);X^>c
z5Dsf1K=jcbBj$z;V<7;+e`7kG3lk0#fa7^IoGbV#$rsMWVN|tVX)6%H7t7^zFkdte
zgy)g^zYX!+o;wQqsn}eB0st8Q4eh7>e~lODX}llmVkYFzY~U`Z3Sc~s|Er!*NNMwp
zWl6RFUk|Da9~^=@QP^*ha8jM6PAQv>oR^F&`7sH%lMQ1aTsyJE-+`C+sl}3Chlm%v
zR1H7gQD%bp=raw)!_`p(SAZ7=A0`3M9JUXa>1J8kuqlb>4l^0nFPaNT&=XWO@Wlc3
zZr`^XX`X5`6kn@96yHePN>3+CsGZ`O>Xpqtk$-zXOuX8aiPvaS^8b#LP|i>$%X4YT
zc-%sF%eOX<m=w-=WwgvM0=7R+jHhPI+p_f8(K6ZYCFvzR`FrRQ?f}p7z%n6Iu`2;|
zIX7yn4XC+O_#W!gX&Kq|ZWZ!s1e0k=O~GyRwfn+nM2RE-vd&~26wJ`S4;g6ozjsX}
z^);wkk32{ZvOoIlC3)9yo;LSfr`ukFm8unX>c|Wh@fLCOKq6JVj9J-47YEd)={n03
zB^hW=GBcPRX)YPZ$^kKCCbC%gGy7dw4<8Q2ch+wU{ki&BB&5?b`u{e~;^i1AM#s>f
z8%Hou87v(fH7|teD1{dM!^NBZ+!Z)^=sf=#aTNBhi>?&vj`O@&wx9~Mlvn_P%W7;4
zFV50>#+e)uc}>T>fE`}%n1jf3`bimy8M&fv#z$_I5;X}^fB~=!z3}s-GRX?GYOB=q
z{yBM(h!I$-O`hgEGLO4=BHD;1ePT?Str~_pY5g8>V`WMWS&#}phsk``y}mol@<cvk
zL5YN5MxMo>yvZEmkLOhx4QV0n?yzwq=Yl#woapd?jA^uWNS~)Fu&t5Xn!&WE?i)u}
zc!ma6twfhq|1o2&9R3}q1nrSobEweOvjza&HbSg9G@<}gJ+j#SUI>V>)D~f(#sRMZ
zGwQ89Eix_ZCxMAZfzD7~X!cZJaE<0!%49x|hYaoLJI(h9>OIk3cQhPG?2898i{3Jx
zCli7YnbL9*#K*Ig1<{aMD|<K~DIBw7=u#?{fzzlhLk0wX#QdesHu<@c$Tb_;Deg5X
zvpb%h%zh;nt<=j7_)X<9y*6vVw0JU!$qrSlS);f0*NdY2`Gs<<BUN);3v=qWqu?B*
z)=-z`A*db?Z=ij_CZG)ux7a|b{Pq3pq4g$snU;W7nC0|V>sPauKw1%caMo|yRQ1}N
zhA$V|mTJHuUrG0xk01SPRmV6wkkzm9ZyBDvR^a&iVwaxK_0Y%s7s%S%=+@R2elCX)
zbLm=ZDuRwd)=ktElx(S8DNH|(l1101SFhohckmLQz>$hT5E%DsRpA1Ku~izXA%}KN
z6`oZZ#+&9#eb@JYHGrU53&Cln9FO*+8Nog0A}T;-ld1sT38NnGnsQH^r5dj)>m=a$
zsUwB7m7}baN+<FLQw5pwzH0Lj%oZ27R60*H^9<n<wIm}kvQ^4CywLTG`E-os6t9h$
zbxYyHvV>X<M7T`K87g<9SW0qOXcY9N?CzLKc>CCROUklKKy4vB^*q<Hqa|t-rM59S
zvj-6IO16b}cY!m#wY2s^C<_`l-cn;BKu57L!~FOK7K#=iYDbBO-_|b#{7NG?q8-Vq
zq4WMd3q19Ugc70nP!~#5FO_}f)%+_fCgx=qhGFn`x<=#(?v}+u4k!OU+JNBvQRecJ
zV85G@0AB?Ay}XG3$jybz++Af!fZMu5z&HTmm)F9YqBMKmw3v8n!OWouhHje>n=)>l
zbw9O6*XmT(hX$6Z!pNi<lK~^yLkVL#A`luZ(GmDjJ(^AD8zkOZE=CGR*imG=n7-3=
z#8(yhA9cv?6mf|e($UYRBmSB$s)+-9rPpfFT;Y(5A%nC!_>wWYkSn$*pV!nmhiRk)
z8yXJJ*VinHf^X0jk#E3%&Ix|F;)V#UXleXvf|P05qA-O3;8@_-x5*Vk1CMnO^X)l}
zH;Zqt@3;HBG&j)^$&s4QkB1FdS1M>(9!KE8F19p5zwVFc`<GO!j7d!!P*32I*CRw6
zCd*gq`oF*p$+$&N!6khOvoIRF8O60j%9|D95L70?UW1Tp#qrZPT2=9`>na;2iy?)n
zvR*67hfqiWlqgl(!f>J6xE01wRJx%}AzY`bK6Saqj(-gE`R-?(K5<2(35Gd*wGq_1
zvagg(rVICrZTg02qClNbJ=53nn*B{kR~DbkwNUHmj+xqvg4m@}u(hXQh_fe5iJMF{
zw!CHc30c|->K(yNXo;3pyflL+E0o{|@=wXolj6*P7Dp}JUgfEZSlE1mT*Km1>{|qx
zbP4@LUF$_+L>250KF%5=@wUeHGde43zN9RIm<AT__A~Y$8SCL$z{}7$cTwvRQFQVY
zSKisgMiqr~3<hjWSGTxMsp_r;%X@#rR*@c#&I4D2>N#mOm19%hhE*wAlQSfIO3Klc
zI)fHKtrlhJ6(2gF*1vi&*-DpJYp{#734~-7FRl6^zGr75ykW56<Soa<+_H=<@e3k-
z@OK#RL1sZO-~P;V8_nZa;-DWy9ao;A+E|kgDvK?-N@;bFQE2XKk&FshuoFE}K}pv}
zz#kesQ*xYF+ejQGUDUQfRv5-}Y$+QX%kE(SI}~G`TJLO)QCN8V8S1utd4zaBif@XC
z!Ldf=V@B}w=;`G_mgkf9VZEVwZCf|FYcdFaGcvvXKu4&O_I{6gyKE@o(w`DCn`842
zw67)c*b=wgS6~(MoEIf2;r&tDL-(~3ymt0VpP<UBc61M_2K}bS=0{6;SH$<5VreOW
zOPq?R`U8H(g_1FoU*;HDC^nPtE93Y3MI~C1C(yjy$zrN2h;)g?BWYvwO5vW5Swe(@
z$4M&oU=kE04TTMNt2#(QSgai$E1K&ReWT(qnJ6%G_Cgp?T1NGY`u9>5-c2%lRr=XZ
zts32RZ=im6nxoj6C#BngzJ#zkVTlAl{g|EF5BZh&Ddr<5{2$z?c}R4hL{aXNvoJVW
z`piZ9#c3Akd0`WW57wyF#`}9j-Hn{+H9m!-FsV$1tLu|i>B5J=bo|jq?}(D<Ngd|0
zUR+FOqMWOxUi`wJHx%x8bunpA#DJEu8_IK*X&GyJUFbw5Gmn(Lhi$I?a;<#87!N)?
zb#Z^2))+(a9KoRC-C4Sb^Jq0QZ`TkhF|%Y)Rd6a#pU-|ZlXF^!jtocvN_p6dAspy^
zh{{CZY+Vov(nbx>sER2)#ju*<u+r#aZgj8{#3G+N(>Qp*e&BUU$$jt>IUVnWg50FP
zvhtGS9;&4l!)O!Q4n`zrDmFb}W8OFsc2>%%pXWG4AS)|5k>^F$PI<lZP~NU^ljXK$
z<LLe5HSGv5NnWEQWIo|?A=$(h2L@FJl-cOyrAhzb0R_23)3RD;c{z(tku^)Q9l4`b
z&w56gwL}>m))yV6E%|$^NJg7x_9$`iAVhQul1*bP*TY<kkXC{xR6z#NoUn#0HtZ13
z0;6(a?#&oQ{~KptW`>qK#*94Y2FH<0xe{DB*OzBB5VL3?f(74tJGacsLbIFxtd|C6
zQK*++<phKobw3?BM+iG(p!==b*|wYxoGNeE(M@|=w%c1%*P4S%1;%$_o8mG|RZOvb
zNRD?Su-1fm+sGJUda)lcTCEQ0{e0_$KSNiba~+83oVnL%-aQ;?mnY>a0{+*Q0#cD;
ziOz`L`sz=Gr?)WK26xG*#pL?_`(b*lSY0w7w5q_|L|fdg!s=SFDT*uT!)8n4(Ef5D
zatubR?*W%aiKsouHIYVEYiCg#r3nSKh_#uv6ASg5)k<@ftkOTg>fM-jwEoAMGE<Ie
z92%QFALD4E1>RAqcZN&}uwcPkHsd!f#rwHnPPX)gb7d}d^g}yS&fpE}S#PqeX#VR1
z<HtSY<p(lkQv4Su+^72SUK+sQOSst|et-eZx#SnkKsE9)Cff>@iC6<EG(Ep4<{MW;
z4lU>`+4&2N7T6DkKlR$G`$yGnQXZRizFV_$h(>Dcl44}tak9(5RWe4;k_UOpAImOv
z({>bbiDpKy{RF{zr{efxW2X$^-y!#+)K#Gnc>gN1OM*fVIV9J<M!iQ^-W0fduI;|4
zhnAlFSph;XR0D3NgN1jz!~-#)Q~H=h9PdIHbW4t!t!D-pcF36CIQ=oOqs_q-sCS@h
zq;>Q=HZIV|`>vFOVW&k?#KWxl3gMZpLLkgA9qr*wl4P4NcYp+65i*`^tmJ955qMs`
z6Sbq>4EEUo+0DyK-I?@(Z)LGaZZbo<7LZu!*90+bi%Gh7Hw?r_dDO{pv8f}70q@Dr
z6IH-%HADfnA*Zu?)aBwC_U29b<aUB`ibc6#)<n|e22Dx>71$#tws;+Rre@)@nVc5X
zbAyu#6f6JI3eE1S=H2<IN<6>L^<HZvb$TZ<NMd}eVx2D-N2-8yXs{Gt6T4nO+4%7m
z9WOAYO#O)J1R~cBukG?f*Alhj<OBmUIno7^wUGeK4Px||N-6jW(10rDS||&IgOnJw
z%@CPCq&tl)TS+5y`ED`}!G<b-uPbPa#KX5FXyQoR(i?HZE3iIugVH0Vti-2gh@RDR
z2zpNzFOu!vq`w7}p}l`iEj}HY4n3mPZ`v*t75UmBPyYEQZT#uT&pWjR|JNkCVOCg<
zW@-*FqMV*3Y<rxjOf}c8UN^9G_deE#L%vm{sw0~Bk3T4}HWa?%$_=k%aC1{gYDnk<
zFZom@+Fj<8Px@#%xb~hTtDz!7AS4fH&KAPtp&6Pol%V1=-UZ!qNH#zJ@7f&Hu~IUE
zvok0;LD8eLB8XHDc*zEwG<1Z&w6v15Hs2>e%pXe}%EYwndblxJQ;~C79{rolI#?DR
zb{*`T^6`|3as?TLaPZuGe)__fW2-?Qb8u26QJMxd$SWTSpLw&kvTLCEZ)l@MEIOl_
z)MR||usVjvN~1z_Y@<CW2JXRVS=4;eS?W_J7Jm+@_y{X;Z*Sqrot!<B%4q&^av>$a
zVl)_CA~JUw`n?;SKkmT_6ck-FRi!dF-{zm|*mY$Yvxv`)+Ux<|hLFg}pF_MwkrmKM
z{$s((TyL4&QQ<Wp9jH(vPkDb>5mNGQ^?N$YF~!gDON<mFZU2hL+FyU#GuOb{Dktde
zxZQ6+H+9e{F3>_Z-=v;#Mh3Gtg`kW8vhD~X`RO<}K~w0w!`)6OrEl0nhqbhysk+<V
z#}&HWc*qN^<Sc7tM4~AYnQS)UnuKCMi^7Ricu_Tz7mSo98{@8f43pLH79g{L$-X^4
zrfJHkicd^7wQl4hz*x>Qfzy~RvflH-!Ul;fI&#fr7UtO*n*ob_O&#tT&0Ptg<#8jf
z(onlnc1@`VlUv4CLswNM&|lH9@=jNqcsW1Js*K6Yz2<|OR>G(SHKx`Mb&hxJd-PDU
zXmP734Yhmwrkf&GS*gDso^wqx%~|!m(3@9mFVq1P-V)J-y|yAmMF2li|IwV8vZ`2!
z3(H@}AhL6{jwUg)(V^kBL!-m92A@dr-PmZ+QQ=O?i^M{BXzD~=_W2?#6(jgg>(>B5
zP4hnn{t*VTq!Ag=QKg|P{9J80!mX6u81%Fsc*26z6}AiXZhk)4ZGw4Bqq$G+2<(JF
z=7XuoGYkMXZZh<K{PF2}7bO-h{$Xn1etpi(yF`=98N=z@=D%*e_`LE8d_X|b-lTpH
z732*%%yd0Z3Z@E@_-;#YK_~NYieJc|qhMh|M0l4D6p^_)`gpPF5^w=fCU**bfXE~F
z4{!iLg&<UxF;I6ehN|&^K8Xk-F=Ps8^uGSMS)l@dVHkV@>;Z;J9P>!$wwFgLG7Uqh
z+1BvLi9z{2KfPaz2f?bOBMA64d<VO|LW#UkmeapPKGjY!{0uVe_AvdSi@!T?LY`@x
zMfOtz8j6*KWn7jC2-pCS^fCvC^benKj%_2}G+U5O=0aC<9M)MrFjl2OR@qYQ<Yo3s
zI=_n8PiWQl&6F;~HK{4OJbpHHrX3*$)ETPxF{&a~r6H6lnpHsD2@nd8xZ9Lv@mq@O
zWPKS691{T=XQ;y-j2%_9^Sz!cybEx+t>?`p5z0eEMqo^~lIeg*FwZFw8$Ql-8(u1k
z81aL7QNmj0=HKS?*n<VWx_I_&VnuURgLr>0h}yJNd668*NOn_*6%#G7!)ZfFX0V#-
z7sxTyhR14#uq<=faN`;^8Tm&n+SzfF(&JmfNQr(!&0zUmrAThl)!ULa$xx4EwP9S5
zv*yE~xP9JP`2hg$4`c|fvhUx-p7^J&L5-jtu98DtatoFCsW1rA3osYOYKdVi<Sjxi
zxyI7wMLN#5IJ6a){zfXy8tB}2d}2BzVVHz#6oRhY#7o1@BWxJ=C&^*wd^KK+4Pz+`
z4{mGMzs;17pE=UCW~!@0LUYk@+S|ers-IlTTIRyZH+TWT{m?Q?nPdB5$`)wBk}WM2
zM6w1JNFp7;5C2S|w&=$3gPRP0Z{ZRb^EZ}~VGPJ3v2Ztq2a3;N7mDGdx>;M3Fr%oe
z0(bOKkpabemdk*j!BrWn2)-SH*bSeXWnBLTr#|z=t?pwUl}SKXeS)dpPEe5bsBt7=
zQdN-+nXLe{Rv3d8jdPT&aGgOo-sgR}W!t5mAjs2{F4}`8HB1H2GPPIzwu-`@H@Tzm
z&C9|H!CQjbBzw;gb|)iD!=W+mTZmy__qp-mb!m<CfaJz*aJUnnlN!vXWv{wT!Xq_9
z)_V-xyp69~Uab(XZee%RJwam<a?x!aAyT;$byx(nin=t?xig_q3j(eL#;h0VKkk#Y
zd*tc|0z7HIG-{aT>RGI3n+e;LL+miNILLPKS^dg>B~0;>bedgpCaoSscYMb7yOGFK
z0{IZ}5di<Xnzy@yD?T)H69N%uv9<PFZkZE(6njKG`yi#5EjUR+ZNQSto#*uDg>v;V
z;1vh#$*WgadlYo5OOq17mivEgpTVD#HY!Rt?B1Vq1Lq9<$_?k-zsOGJ=2AxgG~VUm
zR?d_l&noRc8vzOwUTqLzINk54U!ClTLQIBTwK#y&k2!IAB^ToLy}t?rX#wlUg9!dH
z=JpkDuSb^w!FNTypDQG=s6}N)oO39m?@t0=T~AGdb?;<Y1kJ|Tatg@=vyda(+JBg-
zTx6eQ;7{oI)7XCks~P9fYahk9H(NC>eBBp1if#MY8&8Eg*F<v2&MsA#iRe;uuLG-#
z)@&^~#Kr55#HK^5*mz^C*m<`sM+tDdL%NDyDx9;#Rh}`$hf*v`YpXE5C0Si~1zP|@
zE&{{6C$S7e&WbtiOTp;=;^Mr^aT-c|Nql84>TfN~cv7P<f_(Ei4Qj%t&J)?HE@nUx
zFe25JixZ8UczQVf3Tp9EF=x<x=EN95R0=G78SWN82L5|V1$LhePek9i6|{zjb#rYU
zPH#kqV{;QfhAI)zm)h#Mm}=M@mRW$QREC>;q1-hm%>Y!_eWsLt2ai}tU?6RldqQtU
zD5EB!{M<jfn*7WOIgiN$ukHISC6POv<=ej(xUc7IlTn8lBBk4z9kgAC4<-9ZVjP7{
zUJjyuBAl}nITr36-FYOe^QYFN8D)n}3TN7GvJ65Q^Cbom!Bw3Zh1Y0RHC+G!v(;+N
z{W-gGk3@88W2{qU;;JT#7%_&d1QipayEK8n$sXvwvoUFK$A`Za;jR3KuswRSQ;x(*
z)pt$&9ZRa<m>juSg;ucL4?~HF2n{NPTJK8_!m}fR1G;n!j+8&zE5zeH1N*Uy`E?It
zxFRbJb*z}u?T!l%tgn=%-P!>QFz~Wz56$R#(yQ<S)QD&GOUDfeG*gJ@Q4$4+6pz3A
zi3;iOGL0a3gI5_A%@hTalDTHM!jy`bSek{&JLBz^>&fcOHWa%%bjl6x)jWNk)eJA&
z77J^W&Bmk~Z?`LiGtid<bW2jlbdFv4&gBTE(fLxflCVt?IdCS>+#&$uCcaM*%l5M&
znN6t+CB<GR#7A3lsm|&uu-Pic$z>vEY^95bW6~s=H_8KPTOum=LrOJ69ZN;9mfQ@i
zd6wDU=nSxWi7-3I4MRKB+MCSws@5e97W>0|OqRBbMk&yOIXTLObD>}|#*z=F%^xFI
zA;mq|n0f-&<l>7B*X;luRBb;=g1?IurX~fQCvVntf-mbWPx2C0w-?4LUAZt}EvcVN
z87rvD6_NPIT{Qgk50ie=1=K96fEn%A8FdmW=cF+0V{30_E=har6LQmhHQZK4E0Tra
z(3!zwu;J+GpVam09uq4Gav5GuKfAf7k@W_0v}6~1)_0C_oh*P;mUT-?R2|FN2Fu+h
z`%Old4BSGQ_HkN{eRF!SHS6A^bqb4alhYYjQgh-*A%`-qUV(6ZRL8@J#Ae;BZm<2e
zJC8j1=JEWj0{PoA8RW}51E*V81iV<650;;t0aYqd2Ai2rf#$wS-I?!xfYq6v6TzF<
z9-?2=bkX%pYdhfT^tH=!x_&oI%qouc7bXQIeoNQs5oZ}33xl}To#eui`=7INqr5A<
z+<s>G24`d+y4AV^TrO7WI5boBiCYHGf3;I^+akEIc-hys;RYAi5ojQst=Bmn3*;<M
z!{LeZoR$_>`xt3}^_wgGr8Q*zVU73*PL(){NVW=GtQ|l(;i3w-kNQwqr>(w%ubDMw
z+<g)Rk6+hlV&lol*KAu<7~AC78r`^K(>bN@+uNQJmh<=G;IbDMWd^fVbFRl~VkX>8
z(tXNbE91T-Qn~j$)u>mgGX@Oq3@dksYI!d7H-0@HYKA4`kn~^Es+$eyx`}x#ebMaQ
zXkQ{E8c_hlZJM%oYDq|VF^~ReMY4xHj-%3Q^Yqv_r>K66$5jj(?=CS^c6QA~x<c|2
z1gAs`?3HG@4{bZPYp1MVT>Zi7&5`(`GEI}l{gmLb>rsSP?GS=mRA;psf;Zcg)_Gya
zOVNI5#))}mr6!CeimuUmT90PsnB_VXe<dC8v|R!9>=ZIp@+|?+*-zYC^vmRoWVlLk
z6w^$eG4^U!p<;Vwa+wg{kww3(CoK{7!g_JK8$a!qLQ^#t67N}Ff*X}yQt)OGhvG4^
z8dh59?Rm81yab1ytO%PWvkU9UgVwpOr3alCE%wqL8?t4g_@8!wWTWDdf3HFL2Ju|(
z;=cm;IdiUnKaqHt&#=hQyT4J}`B|O47pTpcNl`bMY!|&=>=sp|Ptx~G^X@^#$T@TN
z$q($(7sUHtMGB;LW-Aapd*b)~Vf(F!>#4!eV*|Rb%*U@Bh{+bsR_7ng6g3iA=X!$N
z7^y<k6~Y+@_EniD`efCpn!r;o)u2rKjyM4Bzcb(xt<66>rt#MUCGZ#=ITdUj@E_Y{
zJOI4iB`)7}bzh%LgQeO!0p%pzn2m?}H7AwE@Sv8M*2Wlk=9jleijc`xuWzmryyVUc
zbPOqMmJhsR4s4)XSNM*5S>A||BO%}L-V$f~aOlqnuWFO;dG2hB%yxA<Ja0N?3BLfH
zPkDp{<ePwXS$T9G&I;4cxtmt(MSs|Kw*dopRhBNUMdxQrTvXYuz(_WkHrrgLqMIj9
zdd1;7<ceEC?WxI&A?o^Am~T4@tn&Um{omT54O6bxl3;!7?{U0P#jPRh5k4!7`PMd+
zKHc45Uf}Gv6fN#W^W>|!6@-NQQNk47nyc~2FV<I*GlsJLI^BmYcL>eF=VoLQ1~?e6
zZy!!C1_>^O)&Kc>0Oe+8P6J+pL`};HAf!yg)`h76+}@?N654rWip<t=^o3VVF)AYx
z`2~!K`y}U5{s|axuhT||<)&w6W7X*EI}Kg9%DYj<WVbkuZdi4zpUN+0NF**_J(4+Z
z@Mi=Xr0x~cERg@eW}PIYDg9Yg)Qcd?nd)?0*i9acb(RR**_sF?6~oPeA<5g3s9}?O
z6Ez_~PHrHk>f7ae{C=N`rPxk<GPo=nHDLPk0Ph_o;YUgHo#C}rvq#%(WU)p5Pnr~K
z#S5};K{IqyALLI{ah|jRjKd$GgnVAe)lEdFUbPz4k%UiI)V(THrDY~?m82B*cPZT5
zd$0pmC#HUAB*JaPf2}v2_KfzN-`O6JKwgx9jnByy+;7kyU{*>HT5;nHT<@ptzo%nF
z{<U?jM09LME1N5Wcwb^_92f{?4D;9IHZUvoua9TX)7%fucILhGd->9$p_824;xnXj
zJ?kf@*R~9lIYw<$X%Y4V{a;F!1M0F=UYxbHMVI(LTBV6Mo=R=j0z@(Y$4rbmQfe~*
zwUmTDXl(L=Bbue*!*(NSxJKkAQ_Q&}M`K%DQF;jJSV@u&uOio|lI#F??C+N0SQ0$a
z?AS_PaiJ!xe0B+#@}g67`#PjVW3qD6*QsM?ykWY6*h2naBByMN3j|aJpw!~1Jn_F-
zejcp)@i=kO-gm+2S!(nmD2>OMEyVy)nbc=Qkz2^}G8UwLXW~oLI3LM@ODE|9_lrHN
zl!HrfB)@ZuivQUjkpH1dVd82T__SzFZrQccx#|BUMNxcBO&^e0Q*E2m(8#~M1V35l
zR%$L!wz2<&Ixf8wQ?=ZD{^qDM=Xu<g?8g%!5iK*0N5$o?OPSt+nw>TUMa~M4=Q#e(
zMj6(n`J)EiFQ`H73?FfbfOZx!_c<nGmZoH*^A4k)qcUjiq&7S6>zQ^WNc)3yQ6^z^
z3Y3+P;{75UZkt(!20)i7TkX~8Jgwd@cKs4!s!}CI)+xxtYsz!1<^8wwaoJk(avJ4Q
z0|ZN&ot!m?WQ7z>8J|fhI70`piQJZhzAMRVz@?Q7P`cgbX60i%yp+_oI+3$96R3-O
z$qdU7xVfJb1K0NB8Fj-RE~daAR(Oyu`9br%FYEv1-3$Yp&R&$5fdGHt@Z2^*62;(R
z#ZJ^hNqw9MGJdKqCxaiX@cuwex5}NWAfvyKPaX=r5D}y;iCyX{CTRzJpm%Fo>1<F>
ze6T$xYrRe^%C4ktlccSW{J2xqFpVZvuqQHJb<_-Hbz1~l3A7Zv=TB5DjnA=v%}m&H
zWJ)r87o4dYg^BJS)+wzZ6y#HgH^O`LQ{C7d3bdb>>)&AgjUs$^Qc+F?E;0F}dl^+6
z=F*9T#4?v~806Ym9$y5Qu)n%+D~hukBY;OLQ+eS-C%f8I*2@TSN0x5VDejkeq?c7|
zHP`l+q4??z<bvEg$jcFT!H6FfIb@%2_|R2!eMa;+UbGzGc+V9Ffbu&+D&Pbg`UM?!
zNd-Nke@9&F*OLXXnjE^BNjWhyz{p=#{9fUnYCvkA$jNca8s7)F{G^~B?2dmj*KYPB
z^c1}by5D}%)w$o|4`1N>C^&5OyVsP>YxpmX$ddQ%WIhdVPgX_6n>$s2h01N$y6)aO
z5z`n{W}GIcD=?+h)lu;RTVTXaYR!|_#pXf=F{(jv)`vQh3tT}Cr-2J8uzNE<&GfWk
z!o)@*tDRQD{Y0}X@3u%0nMz}+z#()vdD?MD$Uc>%?gHlqYgmpKwqLHsrG5d0Ut?qV
zrGx9VfEA+XLHfilt%W#&SeS@J9)tBs8<K=S+DwUX8<r+B8G;c1{c)}=lV?Ln!s=_1
zTxrJJWkj*ZIGcE7@~$ZgeaXVb-8`7%anV2HP?+OUai9T62pz}J#CPll9a05m<CC+o
zwH?C`6?$*?>K+de|E^l!_lhjmnPLfi82>906s;FtO=KKKddzpW4t^!dXk<vhm&$jF
zUYMVh${xpFp$lx2O+gzqRB(Gr4>v#?KXab=s`ACoQkag}(;wC$J!HA)zO<dzRw^g0
zO&l#Dn^6S_lfY8Fl<%spq$$A5L8_l(ooU+32a-^f)Cn*bk3Cp&Vq{ayn7qu$Nq82p
zWxLyTEOTeJZR1M8<XS!w^f+9~-8_g#nw1ChZsSs)QNgrHzcW%eQkV~f9^UG0Cfv6d
z*LEj3?*WP8>k6f?<P<vX%4FdkbL^!<bT86q1djd0y0!oUar+WxBqJ{oAZAWzDGm|G
zs=90U2X8_SnS6`y_u`k-w~8KaH~42=KZ+NNNpA)INu5OL+>3nt_1<j-<l_E&jm#f(
zADB(VY(Dq;*!B8wn{@^i1WkTD|LMdkC9mFt{HFi>`5yj*jXZm#3)L?(P%-8k^SRSI
z7ohVG8QEJc(1|(Xp^!*HUDN%S5Z<pCHzND}Xu)#6ZU{1J-(O1Sw3R*jKu4MWwJl$H
z%5o_y5<pHD@mt0zCMQ84AC6CUQpch1Uc!z(G%MVyUPq)*L->;At@y-Fz$ow2V|Px9
z=ct@QvmGqIcqWVM;SE}MmhBy(K$1bO?Eo|dc=0l4a^oF_RXuU_eJE=bWzA10sc<YI
zZUYM<#AFS@ji(~pO)c$muUktB%(fq;_G<aiJa!~y1`7-#h2U2YE|kG>kNYvq@`SpJ
zm5O~z#t$P-T}`1eZnNjZ6}h}ZI<nF@G}%TX8yy5%47-{Ixr-yb?bTt*BojNw&VasQ
zT5a(VR=`tuD?q!q*32i-=zzM~Wqt2?a<sQ@=b~_A=*`5;Zs7Y@<W@-0s+IaY=e0C(
zMU$m@hpR?pdT8#+7uPUVNJVa)Q*m)gd1ncr6QM{!zlNP(X0Im&wZ_+cB$1ff_<^$=
zuaY^1c-a17#smM&#oH!yJrh)t9`NY$7`t4Z5bxr%kT{%(e&Shssbg0uSBH>!zM#pJ
z?E`DKeJRasGJpt<Y^-HB4vXaV<l6a<e#07q<T^p*i|zTwjZi_)ko&b=viYUTiqFis
z;3&WJm&{+(O%^<QYJw|b=wmqcD1U6UdK56Q8f?nnR`Jj2ITRARCagK|RRGo=DzlEW
zWY;}*0&yI29XneFT{o9h;IPnuJSl^l<u(WgYDtG0B!TTM*c#ZEhrs~<KMHH`t$vhC
zc}PS;-q8CV7|pWZ73xRoJVh{rDxELLu0m{Og18wS;)V|#yy`Dmm>5kqF_QCD(QCt3
zi*M_{pd*w^&+h-&98cH}umeV2F7UDbY~i@!pbHLF2<s$Ha<~>3HDx-_X)pg#&N>Wl
zNJ8uenX3oDoBJg0Meks2x<k2NHi#{4ofOySpG$IsLsvr3pLS@fxf!sh4YX`|nYKFj
zM=XS5*R>{ht-27D==SP&c+-^ysX;XK748(cIHe9F5(@a}5+<hFM*vh*mnB?6yO5f?
zdtA$>!fcSf;6Yo;eKu}!k~<DqXk_d}nGXiLmL)~fC89J1Ahl7N!3`k$@NWL-wHgYU
zG=P?^56-;yS;$RKV5xN?z>802AnMIH3HPw%GB(y#)i9RwMxK0x;HQ0q4|RaR6hw5`
zN>rf73uM4|BTX<}bpQ-|Z9h(+O|;X^R%y0UxTlbh4Wr8jV3<bKN(JI{LSD|^F9C>c
zOn!}%mMS1aGod`o%+sH0R5CbNuo$hq2E*}qB&fhhi4yBj?Vm4t-7d)#ff#<LB<0|O
z;*;Gw#5Nrtc7x>eJQQ7bOE~qplhos&uj0TxXe9vwa0&6&MYP13R4ClDnUknr4H+$1
z0FhFa%;!9LS&)ClTuW>4OiA5iKbR=D2f~pRzO?=U<{w@xH*;G5KOD3)9Xnylw0vn!
zurxXgm?Ge{);S9pOK4D$XW+DiXyX4F$dVRt&X1aA>I5?bstmE&%Ha&700V~qO=<Ff
zx8tn;-Hrour78V|pakg1*ps#+_q=PIS`dZ~YTVUVLB*R``|Cnz{X-&JTNDV?`;EAE
z7@wE7pEZ1KHu;Qa(%z|MQlD#3FY|F*2P){}+14fC(?jk1XDZ;&{q5U@@GfX90DLqH
z0zn<hESan8%kkf6g0tnVbAQ~iF7=1z$Ic#SwgB~>LvSB0u{<EJ8pH4dUQok_VWCY@
zV{`-?NDw72`vCPQ&hvlJvk4=mo_D!iVXq2;Jzq}&C-go&0a<tU)n;R+dxE!B34j<r
z;dH&QcHo=9$(=D+A)Eq7Ip3<HnKOBbl{o$rEwKU_!nV};brYBb4VvUR^_71$wk7_#
z0%I%-ow)#mTtM4Y&RADhb}@G;c0Zf`pz~ja_`Td~c&yTr&BUC?dxSNqx>!=V{YC*;
zQrn)YA7k@Z4@Sn))2;No2y6(Y#J>QFs5bYdD}eJemjM<uCG~`KJUh1A80~k<IJ;Hz
zqRISk0&IE`Ejm$VYWH2)07r=4n3MZk%JsXqB*Gp(G(Zg1Yz8>5Rq<PdGzU@cWQTGZ
ze#nU<<#2~xPEPK{KiuYABf0rR-KAtkCps6yXBxS`LA#E1urdpomc}qUQkklV!m&#`
zPA*)~-m-7ynS$=?N1mxISpqDo1YM>gadfwB(N{Ti^pV7pgep)nvn7JmTpTO3mqT(w
zf=NvQ3;<~Id5Knf!=hRxa9dh(lZgkZaKqO$mB+g#-PG_?qnQ%9^iMBhU(3KG#*bC>
z(88sYD^&w#)8MNvhB(5a@;2JY(rHFZ6JN6|+rN^Q5P2>9Cjn7h_g&e{G?*GF5|wN`
zGI%9b4Y`X1i5g4Xm;$ORFKqh3SW3sTjwKa*iU8Pck7YTSErOzx4EapY=S=KG0hI5v
zZzcmjlD}dt;c5qb$C9IcIz`L-d>*rX=T?N=kpb6E@X_tmt5Tb4f0EJ!w10hWCmHlX
zQ&ewEO}lHWB?nkz;JO-y7l${eQK_>wO|^GNMNTuW6DX+F+2%1Hm!4kBLGkTfOgcQ>
zzX5J<LR9uQr#!v?oyCRS3eo;Rj3JK{_1|t!7{d$*zc^+xNpa1y!-|o=b^**m@w|d2
z7AM@1dAOiLLyR7}pna^@)|%xcx`<_Y_VU#;r9UT7BWn;6(j@EV-f8An>5A;Y;`*=Y
zyEg|q#+hnNGgPB&S}3yM=ES5Xs9ng;-vEktmGtWBx77C{)``pC#IzybDO#*M1iA_!
zVb4^&8!p!hneKnMob*lFgLPkmp=|jZF|5Lw*5sF@3-5c}-E#s-{LoLYC)9key(i+q
z?k$p@)=!<^CLeo3Lc}zkxc^PHVGEbq8X9!Q1&KjVSqV<mj?s!xX>qfGf_Ym$^8tXU
zqi3gZIA?Hg<5GQw*zJ=o^NwMDDF|R3lSnBe6je2`u3o$Pow>PHC|ptnB>p|fZdF4n
zf%jWsns1P1dGKGE1CFcs2_8E@!3c;L|D~8NuG?|WvISi}lw};2!l)-XhAxGJo1iSb
zq&RQnsXeFDp^XJ%gDc|6gXMI5D-EcX4f9%;&I-+yqukfC;+Z36rx|U}IF)0~koHvD
z0{W-NtQ04kcIzuNTI2&gqBE98a!#@nnzE6Ctz^o&^|;pejUyZ~<gYm3;)Yrw?n9RH
zE(xA<oTWY|b%LUbnfy?BVkfrYjA@cSLZ&lD>sA18ro`1x7)?t8Yc6f(GXscP=M=AN
z=7NUDS}66`-x@){&i*YO?5vz>P**yn43aR(u#rer)9KSjih31@bzMrla8vxKUEa;|
zqr9BDp@N}&maBXru;F(v$axDZa@8N5j`0vnu@=9BwJ}gQVA;{>8N_GRi($!jpHg6%
zfc_opqU)w!myLZ?89Eq}$O34TpyI?6P3G?Sn|YJ$ak#7BhH`eUB-ydxIz0|!=7gYq
zAw-HuP*4H~+#JAngtlHt#NzjHMV-_~6Cu$Rq?iytR<=fy2}^<8i;`HU!1`DbVnRZM
zU+IJLuQi$384c2#i-NkkvS7f!y>JB%GXb<wnI1(Gb8bS39saP>3Kf7<Y%~rN*Q>BZ
z5glQV3^lH6JAebBiHw|hQn4NkI#NT}O7I68iaXcGvNNpOWqLzze17Kff;z57gVb!^
z4)LA5<D#d||MxoF{y&9;d>*W0`gK@xX3aGSl;6q-*Ub#VrgOJxC}g2^m}NAUh&oVG
zFDE7qf|MKaYj!>Ji#~v&X<roZpO{LQx;Q}vau{i|tV^=Dch$mt2e)`yj`s5=dE*Gv
zRs}1|<@LTi_oLqZ%R%Z}{5DsoXYQodp?hQ*m4;D#=VUBUB-&V`@qE*c^hw|{1Djgv
zUt&`+GFt~3crgQt6Qg;@x#RH8{nH{KQ|7F3y}p7c3eaew38a89oz&o?O8@8U>ummF
zJX0|f^RYE|{`&^csTj>%)N>LG^gi5Uo;Y<LRhlfyQ{1s@{3i+X;UrFvt+NkKZc>Vy
z#fc5@)k8B6K#D?~;LecMcCt2W^amt4|Kb(B+addudinf^dX7J9?cCI#uLnN-MgA{e
ztBiiB?q8&+slouis~w^YElQ;+D`{ePj5}Ee1*skZU*(Vbx4sc=g2w!3du=$-I@H)i
zKEY{zoI&2!U+0|!SJ=U7OcfsV5^>v_5@g-+u3wx65>^ld+bywwi7|HlxXjhjFK@R1
zz257M1em{bcV#`?)1HW=xU5rWm^qI};bV-TTu8@b#d84WC@=Do$2ck5r7m+aza<uQ
zhh50|YfC>w7A}q=fJ*7-T))OkayJ6Njz9ah5L#3;A+ebqI8W%`DvGPxxQ$QC7Z5@l
z8zW&+uq2Nq4Lql_S(6VhYHeYzZgCF+R@^sAg{qkU(Ld=WOiLZlTS1NI+VYfA^{?Ph
z4pqlO8)yR}>fZzg(Xa~j;Fj`#v&klZRvKE-#o%cjGxCaKlfau^p5&|^X5jx@cf_Pc
zI-zP9$rU59aCaf(wI<SaN>skwC>nCuR4Kk)P(|9V!G^0Y=kqn#6(0+0!AI!;_q1no
zq{H&;g@Hg#x7IdNz5})Q>4{aRv*;*WRSJ4$o<aUe5bp!(o04oOf(m--!o?EJCb~L!
zQOFCff2sU}m%-S2AtP^{G6<EXW+LOgCuz@D^<9~i9~b4GGj?+*sZ@h+b~8um;&VMP
zOs-`2JRpv%`b58yMuQ0qPATFy7&uh1D;dN7jiYg5a3!bI#2FHOH5-Du^QnV#RO%y#
z{AL7zJes1+hPK4y-KlLiE*=+RkFFg(=j?43Njs~`eS;`^Hs8KKlki>ydL$#QX<d+O
zR-c9*>W2C~zA>6}o%e}Qk}^>x=nAn=t8QKLf7p7*=uU#IdwAlBZBA_4wryK8vC|Vz
z?EGS5Vw)3AY}?7icJj}C?tQ<#-@2;nRCU!_y}Hji`|R!ARIY2JSPr4bbfPWe#xafm
zGD*O*O_FJ}*yeZdGgY@yJkQw+Rf?T(ct}V)%zhB1Vgy7!eB)h(tE4)sBbz`VG#=e(
zJP}<fUtvV#DKbQ4gT3n7pa{~S>CMP<lOcJ8GCS;)t7Wn&Fr#?r`HIf$keoF}rTwJv
z71<joo}u&89bkN<xPWgJArF22jmWd7xXAturP10Xbf9P?q)8r3ps7x*wRwf&TdO5r
zKQn5#MGR0g7k)A%6i?<{vMUXTE6z0rv3f|Bf17Mnr`5|F^jm<4T$rk>D^|uXRf43h
zci$@mf_@9j`(P27OGV}a$GeBQ*0ChSM?fua&P|&P0bTzDa$dLQ)&1ysTSFRDYj@Dn
z#Q0$!S;+v;0$cZsq7|(V?P5kajk#ev;G82)&>YzPp1y=bi9!Kep_m-@?hq0FomC89
z|3~PQ9V1G+`4l>UWO|aT=cra~{!iZ@4Q<%f@g$;~ymemB&3VyrzqGJg4T={8NG14e
z&ELK8LDN5V_6Y|wR@jz%+9%IGfoK;;V!n>&Y|l8U`RACf&i*JpE-DK0Ye&es$zu|^
zc?8rzqs(;vfk1ie1p=z&?}-<qo3*+TUo%eEf*gFlKt~8XkNh`ySE<EEIl{W=8|KH1
zvyoBHdrjl|k$a0o!uyOmHMgweqP$nbql2p3(tOZj?03eeMs$Rp2Z@j|gT~?V-p(zg
z9cj&Rq=ehXmJ@7`Q`&v6yQ=ex2s40c{hNklu<_;9xS1j+A8O@tpanDrMXQ3h-q3eQ
z-Vd1SMWL4ea~rUD*jY)KNxqxf!twJn$y+#DxmuI3ae!2Wc&Jc235K2Nny;#=x27*h
zK693GAc+V#b^tSz-bGj2>-qUa=6^1MWX^x&D4d?3&u(YaK3#&g9jH$^3edm>lF;2~
zJsOhK*H?=E(RwJ<!R}qRzC+UJ$eWR+uOhn?SHl#7D(~_I+=kFnco|ze0s&^hz|+$S
zvQlPKPa4;@-6Iar>F4+SFx&DCA`q5LvD?=NfI}9BEzabp31j~`djSZ&J>P;*ZRHrL
z;`G;WD*2pcf@3h)2iY)*D4ehKL{ppy23;S6HvK=&*w>z}e1cD(XwA=3MP0wvN!Ed(
zsgpr8$aOd3NY{7T>PTF+01%&eAg|;W@H(VZyy@ff6lz=yictg9BDz0|Osf5j;Pv|)
zJZG>~PcgiUEO<~71t`{uL*Q$$B532vINJ$=eG}}TCp4sjWqBAN<FcJ~G<MU$x}%8D
z1FDFGBa?(M^2bfbywo_R^=Wmz#X@zD%T&Q?co?(QDAbvz4QRgJ*Rgdvr@>?nYQ7|Q
zUPosAGn#3sWOK@YII6Z<$(`02X7}e541^g4M*;-X3V4lMV_PE|FCGsS6k`0=!O|t<
z@G_X5R&-~3m_jJ>)fCLMSisg(c}nms2`1d<;temw+M{w(Kwl)q7i?5pV5Xq7gBPjC
zs>nc(3d&5oLtwQI#H+#hM@uKL)w#8?+|n_m_@HGuxmtSdI`;wfj5%@Tw9ljetFNjS
z!UVX}^GZLH3_SaKXe~6d?ty1@uxD1N5Kk*{sSA7K%Ct=)tZ|^w55L~vG4T3(G}2uh
zFOCy^Cq?is^vf5kElTDm3$qk+=s9k>-`pl~G8PF564N9eYf;xa<9%7M<qFQ=HugU<
zmw#AMP7$c?-IjB3-L7?#2Mk+ox^l;f6fFbu=6lPSLy#ef&RtCD!B$i0WBQ*T&v2g~
zP|hp~7O~_l*3y`_1R>yk?7JI@XdIVS(YZ$6Uo~F)JXHtiu9p9z*Ud;nd^)d#mxyFr
ze-AI-UqXt1Cw^vN7o~hfSt<n-S>y|D`)NKNXN7DM8tYG_qyHpdEI!?UtZdFaUPpm8
za@x$&wzPhY%M+hj99f84yZ5>-u1Z|%dYA8mtbOAx&FoN!_iGp@sd3J0N&W_#C@an|
zRYpg#C-6MRl1Bp}DVfmFHzb2Patm+_Q5eb<8U-dZ$Mj)s|B76vpmC;!OKJyDXR3F4
zXW3?ha|Oc=af=-d1|enb$Y_x&&ocnYPjCzm3R7xNt1ZVi)chwciYiAQAqM}~P+y1B
zv1srl$VtxKJXP7m&S6Ih<YU<Bc}~rI)q{kgqd%<khZGooCzk89V>q0@P{e6wAr&oo
z;!<K{*s*lZ@s_Riy_fBr_WOmLtJx`6S54F4lZYmrxnSHnpi^D`$Q#-T{?Y~%YpG#n
zoi)%SC_8?kG2^q|@_mNl%i(M_xRNC}hS=S5PeX;Y6hqhidXdGTzLaJqrm{d7Dw-Ui
z&t46$Yky5eR>E9C2oej$G%v4Di`8+|Z-ggRH?~(eHp7#`#Jx4y(KqnvD9Xs(S+Mw!
zBcEDhqd3cy9`xOBj9u=hi?;-<z+jS075~=$47%m=@AP+35+Psk#G++UC!5wU0L2Zc
zJmaOhUH#5fH^(ZM+Darl`*w(@WN!15wxl@H2pxL5r;`gaBR@sl;;YXXu(Qz4r@$Z`
zsw}ElgIb+z)sVv~29_ur!=`SgX%~{KJ}?M}?P**U+8eaJ(jF;43swf~)$^gONPhD9
ztfs5=6J8D?v;`LslzzWg)7N&r^wu}{<z*}P=Zl3OjzBz?9A)w*8O5l+Q-l}n8SVUU
zDhP)b?dqhngDiLrX`DsFovg(%Gg^))A`Xd8!b3S^oufRwHmT-c9-q>*>&dGWbjgh#
zg@e8n9>b(E`^=?{WEf?@!s!Z&G2Ty0&o!2@QO;rvXVFXJ!-U*n7?uT_(8W$^o=JB8
zKrEs@zf0sKssRc$eZPEunwYfoCgLP2ici}8;wZnRr?1-K8-^F$#5%Ww%h<Ol7w`@D
zqmYO4o_tkJ<}W<P@AQ-{4AAa&DuhoVQoq$y<4oX*nG+3JPD)S!Vr$BJ+IzZFEIT5z
zX|NHnS!Ikp6Lxh~7Kb~k8D*$>HDZcyTe|*mA<eY|zH{H26NX?iyHv-Jp=bFVk<M#{
z|8zPZ5@BxB4hOT7%@t9@O5qj_O?w>TREr<#SE-Dr+Tn;j;z$01b3+SRSkG$Jw(c=P
z#bRA<ToTN^=?wV|h>)>S-7?r(YxI)5hjY7m=w>1?$23bydp{*9?Cs2R_&Ezi7NGHB
zeU<#%EU9>ypc#;p|CBi%j#<31q2z4Z9~Gd^LpMQEcHZ8S{R#W`;`En21|Q4uKPc(m
z?j!zDbA5iB!+oo#6PEMt_G~BJ&h@2WIQnP&U=1`VX9|m(Oi7szveK7zh40&17)Z3F
zFH0LHGX3=Kt4Ae^c<@inZ$J8yK;!XnYA7fMkBuV(@&!`^>RDQ&$af&-1UPOW<(Z&s
z8AIs8Ds#g;Q%N3vQuC|hAr98EjD6~a253c`>t;?Ff8*t}4VSLDsp>W|j<E5#l$|$n
zZ(Y<mH4R_Xlsi#RaI2=4Ga`V=O@5g`N4e52D$bT4?kS_kis+a&;#j?7f|g_YWR5*-
z>dtZ3D-{X<7HX3GNNRSG&2A5nu^DbCbDP+rcVwEB4w^Z}&6wN~(S8kA$BfB1ppgc9
zDyep<X0h_tJTjOJzXu;yJJ`k)<#eNRLPj?UYJrQ6XUh@SM@fAU!M4$d;|MyBu$J<U
z&`fV6W7Tj|+S9YXIeT&|4UP%&Kzn8u;rtqnYs;PROdv`}6#K3xx~c)tbltLZX+*ds
z%%}7;XTO$FT{?Vn0;8J-^^k@%B2)g5KD4h0$1zAax17vUJ@Mu%@63s5m{n;l@P(#y
zZJuxKoQG%yfrY^~4txorS&~X9?L;{rC4x|QqV^)FpQEN;$s(BDFePW1)&HclSXfkR
zf3Nw_5!noFnh$K+XiEeZquw5i)C`}oizXYfl3(!dicVrSRuPoTxqegGw%N|wWV~dU
z!9y+{c1fz?1vsM{o}0vyt#XRCsp$!u?kd`bjFOQwUk#^!4aLx4XFxWK9t^!%a)PMs
z4q7b+Bqd(7N4z;V75W{XN#M)S-CgQ`5;-7VFCaqMBci4tU$X&gX^gVH6AK$+evZ1A
zt&VNv_yllcO}~q2f)o+^Gt-Cfe66&(^@EB{jE^SrWALRqwHCd%GBVtD|9n5IZFPO)
zC=+U1Y&yYibFb{LGd1*>j#+eWe9|?Q#`(9o8^cO$-#%CtXj<GC6@)r7?T!xd3tOJn
z4l6nh^#kOOP7@&LR4SCWY5ucfRoDBxIK`vOh`BMJMa8!5R+ena$ALA6NSIK-7w^7q
zp;A;zs4q&1cXRX*pUnT}&<E}(PYvi>GMo-p=$O#;ri)cA@m9`d9?$Cd<s|2!jV&lH
z8BSK1T-s73DsCt3w2}|BzNDsKFgJ*FN<&;Td1}+j+iI(0^}xRRL>XmO*R-gxJG^dM
zs4@1_yGKsd9P-AR;rWj(2rmWB8u+=o?v$UD?R6sgwX<kM0EPGuyW5-J7Zu(n;xwy_
zpa{Zrlo8kk#oX}lI-*}{FuAS%?4Pg@h&gH@;$JXHCbE6(M9OW0>eDCB1W><R$R3KL
z#|~mcS4W&Svl5Nu^V@klDP+5S1z>G9$%uCo7<qF+AxDKFZOk)=eJdGo`z(c$_3MX5
zrJBx0RU!UEvBE^T8v_{#N9)y}>~VJ}L{%HDSufnhtuoF(Nr->q*Ghc>v4Y@eAX-d~
zL53hW2Mn0t<DO*y;pN`#%cuLhV<2p9$@~9)yuo=mKw#yJm>|PeC`zDFZ`ApO6K(y4
z?hyi81R`21c~qmWD!J^`a<E-plP%HyfH6VAAHXBoAfZV}z-}N%v1Whvn)>;4JJes2
zW3(Y;<liQf`5!LZR6!u;r{Rxh_)NUUbc8RnXrJ;gv(?p^@CuWB0)JbRZgHL+kJUQ;
z5{*k?rkH*<e~13_&<#wx6g@CM!0$S53X4pW9gS0T#GiBA@7!AdD;<{yt!(Y&G;az5
zUfWKvnLKCR&#~`|Lx-a|OmBfG6fU#fK*ytHQcodfVq7}d@8jwZ#qRFc0xDg(QE-_^
zyGnI3264;v!cQXG?z9jIy4{GBP;&CZ0W#R2P4JgDVT-iFVPHqN+nv0WMw{?wXdb&4
zrD|l|P1LKvgfzI?m74!pCQ9x1bf1A>B3~>DvHYx$*Sn#)#@Cxe{?4rw+^UzSrw2)z
zRxTV$=?a173lE+hf7|_!%fBkKq6BZQF1|9DqG0HKNk<3ANwzo!vn5k0(`?L)L{^j5
zmQlEm@GBcb8GxI-iIT8;oo}kjwYH@Qf6s4>1RA#AqWl3|G&s#I{rGKu9bR0$Du_n=
zTr(;-1JmEFk^Z#*v?NCid!$l!(4{T=8-IYCAAk(8B^IW8D@$H|JKaU>Mi>;Gn|kb{
z>M3p}H`(IvV3|T<S7TGkJ)RV73a8%$72*23z)m#EY!GNarvJczOxDqd#VN$iEBQyZ
zC3_mJn|b%#!Uk{r+d$|8)hgw0Fu~qVC$86yHof>GV<=I~F3Ycm*<M*3j|5z<TO<U-
z*mm$mc^bAcKYabXU{cv~Q{|h``U(Cs5AsrS3P%B)gds|DPa87Lc((lbMUBIR1!c-G
z1)Z!P_`*Q(ZGC6)x3)=}>^SYZS@zkSt29k&%bZaHkL0A<jN=9UfdsgHlGQQv^r5$6
z1J3RX&dOYvYMyfj=?M2aqr&uLlqLxNK+n|Gh0FoH{2HGk4o@gh6fc&mWNZ{&`-C@g
z&Zd<eQsF`c|8Y>RS%WF&-E;%}W3h9fYWvvQNk7mTi}i=X!By!)Kas2Swu#%WPHz$f
zbJbL9&f-#L-`I=trLLjst~v#d9fx}aIzva|i&8nixuC_42~P`oX&jk#ptTHFpKT{2
zk{;0`_nEy?*4PPtT85uNBRg{n&y2ZHP3p=6bzsApckE6IOL6*cK#=NB)TK$-8UmhR
zvk4$d&L@`0m2{}UGKMqx?G?2Dbm%urom;?*Ho{V{CVZSAKCXqs97w~Wt@yKOaYSs=
zD%(CIv<3U$-Ujn|4<D(LaN=?c-~GLn<Wzb95f=4;Dy)c?fX$mda#1O&t5T=SD7PSv
zgBVMreK<`o$PwBLzrP{uLPJGYCWG9=iV!Gj-jv;*deM|xw9@{ePN-}7w7dgtZ5$){
zpiUI#uJc5=4}q20zoF_WI!;optBNA*lW!S^_`Pro-Br4zj$3u=uTL-mqh)+*4+MdY
zyVLiYKwm}23rn1pS~zD$gr93`hj#^ntXimC6WV*+fzeQTG4IM>c!XK70=;9Luxf$H
z-K~^7DqiSd!uEK98TlF!Ys*R>=L(E2--6td^XaHL5<{+XZg40zZlo^3Ni|^>kduSx
z$e+10#jx<10yka=BZ|(HHjrUIu5CT_dMsVnVoz5m5<eEKl20}HpG0OKq(yoj$D6@B
zf{jHlUvd~xP(uagR!URQTg!=BB`QG3ij>bO^_}0kZOaKLW=tb3BB;7d3nrTo7j0N8
zt<IGf1=Fd~>l|%hdFy%7@jcEV4~(GiPi=o?B1{4E6TNHo#G=Zw?4{2fWPyl7+C>Xt
z%|V>idG7kr0fT`=Q-wZk1nkZNNBc~3%e()&#TJRZ@C4t!KZAtM+QpU<Q~>YZ1e+d)
zUSG7ac!_aycvuQTvN)&Ae#MCv;|&|m;RM;Nf5(nl)`6QyvZRVKx$0c$vcZu$A`SaP
z`Uy!Ioq1Kf25eif2jseA!M2pdW6_68yAQjdk*^Liq^WX@V6Y`b-lfWDmYw|z%6rmg
zU~Jll16jVP<d~7p>~t&EV*u~TQ-f#|CPS84`gJb^rDtltVD*i;N(P(ZV!KC;j0Ke(
z&wzESE0u_CpkUP}%c<5-$H@7*^W{*$C-qL)Dt;=))fgv~$TWIx35weOGqp&3BXQxY
zTNv}zk>&A-xHB_cZ!JYfx1=RgYa{K}K#BrOa>V^KJSI^kwIHNZ1ke}5I(LkZ$pV8~
zXEONja45$ZAFRS*FlYY-g9j~B(NDC@m@8{8$#P`HrBq2<uKl~G|37q1mT!AxPKV6l
zS((!*`qsYA|HO8ZlSraWWG(t>)_iqjj2Sr(cahZ69gp|OxQ6B0AT-ilJ9ypbAke+|
zrqCRXHh7q#`BWDYWPo>xPq8;bcVw9nkgLoXalIz~HlYy>R>4P^Sy6#HpL0a4(nHhK
z64l}CbqkplMXLClgtKOZP58J(mma&3hf=<yoMq#W{im20aX?qxLp)ZHIh_ej;JMd?
zW@r=#d5nNZ`tMBSew$5`ty-OoisfnJm95J0%k2+cLyudM&ss_82jwyzSyL-_in<zw
zd+L6QF?z^Fxt~dv`D<#7LrcHezr${re+}&S$)}y)C#?fB{FCZ=y>1_Zmf8P>N{BR5
ze&9s=?>*V28MZsLY@*~IT$y~&P5KvyzyerTN_i|{yWgc2OFrS5drpyhU6B3u5J3)m
zf|{5tSmj@poNR#3;v`Cf;WT&svAfs-u5mD0X81OvJmd_omzJJHE%mrphyBdN)1;{C
zf^>lkFn<PC7}G7T;a4;Ymq@HY8GlQcK4z_?4e>9dym!piK~>Bva=$`@I;G`Kip7O0
z+T$0}8=rWCS*Bv3pBT}!jXM7w`7n)o*O&v|TL0YY621b?H@fS7%Jb)<p9~xf5s&3J
zAv10J@GjLMEWICf)%ypsZNqi0u=Y)Z@@E3eJ<`+SotA{F*@5Sr+3~%-lpLykYgAIy
zaJtoQN%@=iZbrW&#J?xIP+@7H4R-|jpg+-$7NJ4KUiO6mGSnsUL14K;2qSkB+8<JC
zPd<$%2Rwk|`1g9^(*?y{f#lG#RGRB*ISU42H#>t7-KBr+Xhmh7zBDPgzHL#J6}fYX
zn+G0o1p}f7uvV1KWE3zTM5-Z{?FYLj^mXbT7G<7hd%L=2Ti=RJSR)d~cOK+YUaL+A
z19*I8oy(zCYYZAs_J7!yslAiz=)oRq^vb=mkNTqo?R8lk_q`3=${1gKSk3x@0^pF)
zu~~UI|9^s`tp67r1tB5A!Gpf$!{q>rH=9Bqx8(V7rp_xLw?=34FNd?;Il3Y25wMV%
z{%>c)LL^D(y4>*@75%fZKxUV~Tk9Ez87x*>(nnvnzpoI7hkrPwTc84MgJv0PDTcij
zugw|dZLDP?EoY=@Jz9oLZQ7A4&EsrX^7@AIO^?MYzHgQf!a-ka31VLtpCw>HWo{;(
z#H=Z)j8L%T6~UCW%;|_mpj>t~4U>E^9~SU2UL|CIYECKYvbS_X)LGd%e%$xKPWBi_
zd7idJyq9ylpXPicKzv{r$~Y>0bar~EVC@|Zv9FUG(k+)%MbaCZCN`875fl+6Fr)4k
z>puv?_NSsi*71CjOF8ARI2HIMgoQbwyS|2)9({w80f@?c5Q7@uuW_00Z&Y;XqT$H(
zBt-ejrV=N;3WJ*8m~MDiH(xele{D?8^lM)hkYucskMcur%2|7j3t;*q6g1zsJ*F(9
zqgM8~u^i;5<aK_3Nnz<6Un6ZD;Pp~xY#j0Ay?;cl1tq2Gy1r5lJOLE+sQ;=Ug4kB3
zwmy}9{y7VV;#^v?=s&*WMaQ4NxU(`?h=^u^jW}TxX9w-PivAQ_EnRjpQ9W7ut_#y4
zW`zwiOA$;!aD%yTWW!4#wts~nM!5#d+5j2!`1iUs%RmWofA(2c(klH3HnyW7m;cJ_
zxh1V>Ppf%>O`%sW;~KDk4sThNOs^D@qw%Rrbe1$@Qn{=Vqm@5N;7jymXkY}dK}Rfb
zvVaT>pG!d7-TO5?{Imkskh!94Q~spKS4{Wq$}Q)$t`HN}))Z~_%aqTyb(Ba`@E|<&
z1P9WVKi2LkMbFw+kuc(0sBji=+-vXsO`vd<<?lOA%RHK~4ke&4d4k(nAlibPgBhVL
zZ{iv5S30+dxI<xC$xV|)U9X(kyI{qnq#n_aR(#D`rhAe!o4}}H8Yr8Qq-3?+Y;nrs
zF#%S*z2{j8sbKMwrj(Mz;A!_M$0+~Tzlfja8*+_beR1=d%`_DkNQD4giLRYxdyIwv
zQo-jZ909981d5;H=vsy|(y6`5<fkJuT+Q$vWDXG?)|`*@I?!fEY)}-KCT6U*2)R?F
zrq7xxi5!UmdEtN6(*-`Lba0itouM5Da(e2L8D551+*}&zUL>LA65Wb9a+HoC8R@d;
zTBH5LwU!Ez@=X;Nf8SK+@&0v<TO<2NvTZ&fOxU<LzBdZ+g)=^1Dkg-hm}0e|tA(;x
zD~Q%u=vNMHrKc4g4R!DBhhNjf{xU4E3yV#Hlo82Sc3-fyM0aw~)Gh}7n3rKH&7$&(
ztFF;u23Jkf2Q_`W8t*=5;B(ou=?Jfx&`mWxNOx8r-reWkO%xWqNj#UYWM9-t{1g&g
zA&C`nznujl9d)Xk<>Q9J@u(aW_Ky#ZhG^8&hj|vE!;E4vmh|wP(A7m<76vnNM^wi<
zn<Yd>3Tuo{>nwKtY%^^tO1g~-RLnhSz@rmu@5NUxX;bQjkQJR_D4hfy+Iz96tk%O3
zs@Y995WU_<vk=ND|A@yXrE?c>loy>x*03NhkTL_3PJdT~8Ln@B1V|{A|8D5(+@tNF
z<fpHs6l<F{hYD(~b;tY*43Q+Xj2MCG=)|{_vZoJsnAKSH(Jc&pgBo#gq!>+p7c+!n
zTTeATof9=JP=XJ4U}QjWvU-!sZk~ySdpp{r{H%_6Xu;dhR=XoV!;|$$C+&e;nWF}F
z)9nF%PRQ)FXsCmz=0`9}NFqThDGQU)%K$olFn(Xy#F~t+Ox)?*R14IX`%qC{?^qOR
zD;4s%&e4S}yQt>*bH8uHSRy8G&K~y+VzqO)<>D`l3=1LN?1v!Kd&<G*@;zRwPtA4X
zod$lcD3X{nuvX#B)P-oe$TR65snTo3s|)~8-x2jmUA`-dS}1w+j$9bX_M!F5F**#p
zCSV-W*oWcU4F26zMFmOL5oHqu*<=S8`xpnT%DB`6E`3`Oei>!`MwP}tBjHudHb&p&
z*AV|zDafvHBR3biJHw|jH6<LRTXL88V52G{8%3QUE1t(3I^qXR#Zh89yfs~(wfqLw
z`&pGBG&)M1tkkOquzvqWh{v(*D!2dRF|S(Wea&&P8a{L3B4JLZ(vFoKLtz|bbwg<I
z-HY6vph;)F$zcH!m<-E+t83Muw%%}_#E@g3#SoDQ-*nMT(89lePOI)xo=RoBOYCuw
z*r3aiGEX5TZP2hf8x2k-MNZC6S+oH#q;9Y{j7EN8*ZessGg+GE@u0nnqphHy$_Ap;
zaxC9Z>nUb6XgF*6ZBkc%Sg6?SL7-*Qf}`mpO|olL>9nW9n`Wr|X}?v6Ja;Z(fIg&Z
z?rJr2p^7kV4YJa5KSVQJ%hP{WVR_!U+3xt&9+ePL#@I=adRDe-SgBWeydwud<2mw>
zOc%;%$PMSs8yWtZsp!Z|Ui@yRa70I-UA71lDQeV)V?jhi%s(JyM^s8o8}9%mPEl7H
zNR3!qODq0|)Ac^StVVQM6YE|j)Dg^3$00umf4vv8DB&_rPc}Z<kw${EHKaT&yl^?e
zL}jW_x%R7?WRHX6l?}FSlFz3mLfLqu4ff|755a>KLO9P$-97rU<hu*5zH4?<iLrdN
zq-y>d$uqOBSPE#Nrg)JaFwC<rUwWbLi|7<x*00r4eTyLcdFbyu-BxQF#!Pwlp8@>p
zwOR?QwD+>+dIN^xpTp_mhdx`bdIK9cKjc2U3Srs&0miJX%n?A2BQ=m<s`z1d#<{jw
z{_xF{l+-OJ+&XDU#<FgDLA9<j{Mv5Am$|qsr!=22T(c`D-Y`<sScJ(ir$p<3Z{y&l
zIdr6TfvD^2pPeW;{fM>+tj4_at_CUh(p`e3TVAcWVlRcIC8uR2YP%^pm4vzA_>?k^
zxTO_7&_D9OL~_8Bfy6kxdjY!f?RC){ZA=?Is{^;-R6Q0ChmD_EWebX+?<2dfna{mC
zd{wKl?>&1aodJ(6Hu8ncK=^O%7M63x9)hf^qzQh<Yb)^C9OK=8ieAO!M0<ALVej!M
z!{s4`8MRw4yDbr|Ea)S}B|jh^XNDG7$#*14Q~n~GxX1u<s+-0yeU`<s4%{aJyx^<P
z^+j{@8P7`u1u>Uj?2R2|WjBmnZlyb&<4}DwiM}8@lQIf;Q+oJWZ-tC5k=4X|1$M#`
z9GA<_4^bX)4jomtETGH{H2K+&V}khm*F#$&g}LT5YbrA#N62a_OOg$aFcbNdJlDad
zj*LR+W)XqLcLcFg-xUGy3@sxIb?leQ*y#6lEW&{Cv^8R|CBgP-D;F>+V;5vy_kxZb
zWfhk!GGr&NS!%?l$aN#>xD)MgDSGmwy+Rp}zK#pWVwyzt&vyTO?80LYspO1SjZ{m<
zPn*PK1N&1mGA%Ez^J2?(Ru;k>|Fb&6rPYnbZXoA)u5;6j*X=|=zT}){Opeczx3Qn7
zc-FX@kCrX#Z7KZNen?T?ip$-S7xA9AU0&Ww$kmjWu<<PQ`1JEe$sfbXzO?^dnmH5v
z*TU7_aqx4TyE$*x^4aGVm#gKc#r*#)b{*L)d0%SLXOEBkaDo3=#2>q^MVl4Z6Z5iU
zoqewOpNXw*NK?`#98bqwIotXEH*L*b+%Y?WWyo=EcE2;8`@Z_kg(a`s#Y3v%x;OG{
zsSn&Ja!i^cLv7fxYYo$X=Hp|@f?i#Jb-BkcuE+nJ1=>g1M(V%o#H@JSyWW3-4T@(~
zUReM8KsGgtk*{N)y`P=1H+(*yPf0d=oJuhCzbz!5u*@#UW~C~^yzu_lVwu04?X!z8
zO%dYkBL_*}iXbyIt%Lw<@_1SIRPuR56R*VNfJVyr<#uh{j?TA_%EU=L_od)m@g6kg
zuSk&-V!!>1BFWcXavjlCcbhje<a=o!-ZGdF1xU^M;Ui$)tdMFZ6|d-<K<qSb_7}KS
zqC><eH`&C7<fK?_Z(a)GMd@ciKEOfAjveW>71vtt5R`i*ZsV;(S7ld?^0d5q1e{rS
zZ3^YQ!>A*7f5yG_VKG50K2Q|!?EfEJg7p)4Z4E~a=&m}hN~8HrYTfRR^|7}<Z$y#^
zQ!bkatq(vtL8lt@F`K73v8PA`<)dG0UX&JFN9H-m+*%;)sTP*Cmn&&x(Q=j+aEQVZ
zW?}82!)scDpwM`2uo7zaNIc0hDZ6%qUA&U814!XT`g=yQA!&Avl$B8e_w`cK|JG{;
z?`dxWd?J!CDS;$%I(2c^4g3ofoTN26EOL^{)>7osf}K4A`n2Ptk_cquY&Qw!^^h3~
zhFr$r_$&e+N7trc4+a}n{g1hxU)Wuxb#~57vE!~EeVKOu91KuLjTlo<Anfa)iH!Uh
zAc^AaxH1o}UGCTmc1of^r1{P&!sHprQKOa!I4Qy1TtT)RhRE<<3D8xT!lOksVAK=A
zmTROm)P%+>jliEETn1LT{X}UFtjWHGmxfj~3{0wFv49m*%Z}O79;hKtCb<el<3DE?
z4#Nm$6O>coF2{kXi=*G1#W@M;#rqKhA7w{6hL-UC%6YinYXw$G``6EAstmxK=6d2&
z%~)_qn^okIz#R7DF-o0P(lZ#B2{mMrQD(_nIR^2|oCpPx2|@n!o(}=&)IOBi8Y$3R
zCV0UdUckyLky7+pC(uwim?IZ0e}atJWz1NcLUaT_Ohm;=n37ziu%A;8vU_5I>^wib
z1Ns`)rWdP6S{}T+$FUHFAo~ZgQw~TP_=`W08{8l6fZ};XF^TU%|7p30-O<lY>h}ek
z33{}KeG>d|i-{?cpx5(bXJ77<-|Il5^HY~ICD7$Hky=nOy|@Bfv>60++izTOBo<@L
zZFp)iq=T3uJ$@!(kV0@pdhpnQnRN-H#Z4pyME2W6<n|AgYYh;06Tayvdc6W+Y^Ev9
z>rBKawryIRF9So(7ZH>D77)?HGEWCv^K>}6zVDk}u(OJ?+8x#-67}X~){BqojOCHu
zx{P|FJ9;0Cu~vyj<)eNfa^hV41<1@@gVNSu=^eC_ukJ$V__5EsLLU;+Kj?oA4ESy^
zz8T}JgN+*Bc$}@2Z&&T!w=)8UOjsIx@j7`mIm_t`llGlx;P5b7mU|}okBw!qqrBC*
zND%-AaRgUO&jOP1bi?wZri?igw}S3u((l5EF?-#F{2=UsjRBRun>V{AZU#f6aX|;m
z`TQk1|Bu^%kN4X~n+Xs=7m=uQbGIX{SYE#)XF#-AKYQJ9F#Y2T3J3rufI*~%t(Ap|
z&vLp6kI+DNizzY$2gdZWY=whSQ$f9?#I{FPJ$bCztCWC)qMa3#4ur9qJ#)qFBMRA(
za}**iHl2AYD4Go&YSmxCb>(Wzng*+>=-&(|z-gGstc!CH#VSc!l{0Y2b5}*x=n{ws
z-oD~;Mua#&<e&c$-2jlUr&k@Wr-uaLy&fRx`=hb6*6zXgg=y|;UNNkHN#O-aDX&7B
zDU~C#e2=5qzLM#{{w`oV?TY5XeZ~|6<l&PX#wess8KRV{=$Gyrz5W9qt}e~^GG?q5
z3@~vG1Y`tmZYV5Tke8jXqMep`Lwe-=+m1@u(APp{Pf`uD(gp;0#oGVX*5?KNb5hl$
zy4!m_vG?ZAGNLy*w0D}RXM^!;$)fdFKdS46qRLtTOf=WXba!j!BiTdM^EX1(y>%IU
zHF0H@?kp7?DsM%EyhBb*UX)J>XHkxr;f`9qi7#UC?op%7PzhQ*_+c0ORh_Fma|m3(
z{iuWZnMefjgB8FheS)o7C9EPI`9{<-{BNEq1=+(DC*)sEI)xD2yuw(zqp_yzDc1a!
zPViaG-=cZ7Z9HdRlm;K-INmN1;+`RrR^+XM5}#YJ1#y`t?b<Vl3PjeW$Jghd6H`^l
zOUqLt#5L3z7aD%vj~0f0M(^4mPaWS~Fi-3%%W!Z~V&MT!JKN|XWvlEagChyg98pJ$
zbcUA$!|KjKk@f@lX=bBA{*B9TCt33KRX_GI(8+UXg^q)^&yd-g`RZcbCutTZnfSzn
zJZBEr4p^qz9cD^O8f}#`<0%DY?L2&N9%}U;<Ujt@9QL?z-4S?ZOuD;;65X=+YIZN<
z+Gw0r|C0o?KGV1jl1&$o=PF$MGLzLznigbn?9aHG(aC8B(@XPXBIDNohE2T$sj*h<
zN$?y(17aF-hrXrrJw7kg$FD;8c{s0dOpV7~y5XT8S$+a^7fEnuC431muP7;T`MaUn
z&_<i7P^=3(t%_OOVczymeY3^7&N=njjYf?g@0kE84L9AW$fp!5^zPoo>6+OomVIn5
z^0U-EUvj?aCRXl->~GR_ia6#L!sB(*qsEe%T_E@;KX+cfL+#<AOJ5D&3dOQM3$~2x
zdzX28mxJX+-7K$!_`-3l^q3m5&X9agMO*LJ#An0V#nF{jxs~5Ym+7?b`(Mgo2#n`(
zkZOSQbchv?oHL(&lYb98rw1)X_of37BHo7~VyhFI#x=-K7JM(tWGzPUV_jMu)q+|B
zdRJ<D$l#_wUWc}p=c)YBZ_V0Hts!)nFo{=r!!m7Rzr$8dZOH7-^!&0&OOzztt-?Cg
zZfM9euB6e2vQwSWzM+p5^1kwlmE8+X{u=|7i90gDIcQ7dML*cIQg1o205P-o`WV|9
z1vy5u^SCQ+27LIu*6kCF2jizTBNrG}n<IU*1AD$aY1v44EX|m$?cdNdwoupn;T05Q
zimjWb$j!)szFRFq-7~ybf9XH|7*IX{;p_S$(y*J(7O$#1vhaw(>@M10o`cwlOCcbg
z=usF(h(sK2gqpo`fhICE$dud#g5mqLLtf-I2y=p*+-hV!MFez2&?Z(*To(CsNIQc(
zdGgQNFgto|%(zhN_1G|O4cY~M(8wNDHod@=3GV5B)vI1>ygS(qYZDY%m#rX`_%-L|
zNuIQ_l+!C*@kK<wlRu3<#_H$QP93HazP}pJt!v+?AYEnKPQV954owI6|B}t0YN<+}
zjBgRmaI}D2EEyHbXJObaF>_zJr!<+IW30f)2IrMSV+naXC@Ozh?%iL|Iv1?jh^<})
z$#qrOPg!DKOX6nxr6=rg-cw(IAc=_Hg}j0wW)nI9ET&}Zh@D;gTdImFcNc;Y5f2F3
zL45WT{%Lu&x~-7Wh=#-l`3_sFqsP$u3bBH;YYL=e?~kng61GMoTjkDwHA^3uMzF-=
z+s;=qj#+Z(A`+_fR2A7`D0_Hs@iw@V#bK3q5_o;DgM-lhM12wznks*PygIyexN9Xb
zH=odd3^}|(E94lH!6)hV-R~pDc?}*OY2VJ_zAK{d3+oQutm)w}9`c0LWNzt(9D^C|
zxB+0@LrL~=7vKGV;d@w;vz@b+ZH=iR<FAjww%~Gw{&r$(0g*%5n-Xmej~Dl4Bekul
zxwJ^EHh<!SoA%PKb_@=GF!e`Q!DyfD>*-fNT;YH1kscy%<^V(K#Z(!EWhrw~4Hm)^
zQT*!MIsQi>7aFl_it$$a);jTks`U}z8V3pu7^IwWEr^uJ!1k3Tn@l!>OPdo3mRaF&
z&u+S9>f#00zpv=^L_M2!@6dP^ab5ko^ge%QP(wMCm)=y1V;-YRr1>q@N#VOiUKrq*
zdO%4$gDIv2XDysG6ni1Icyn@75p*wql&BGWn#w&+IV%74v&e^;h(hf8YR$yNz8Nsz
z@>$B3W+M|yIvYl#zeHckuJ|=}i<HuezcZMHd*$*wJIZoqT)kSf^XKQ6^&K3EbWmN&
zUX8Y8p}U<yqeG#D&YiSlp#8dk9fgaR0em|AgH|byJ*u#>Paw1*Q#4E7a=y2|&E(Kw
z{3@hJyl*04<Q`ul7Ez8WKQfwARUD|={QXFTes$GhNSY-!!3}2ec~HF4mw~GXk+b{_
zsf6OL{phPd2pZIuIrC1V`J{E|^XHysul)y>>KjOrpGJ>Y0uvi#(g%rZ%^w_-m%ZmB
z+D<qG@3Cw~4wIyP4H@(<#wJEP95{GWo~B7fqw_bH8#LE|H94Puaep~Eq5&kZA<iSl
zWos_GJT*e}lRzGW2+u@V54^qVAHEu}uE#&H3`Be)d2?%wcu+RmT%GD_C*lThv6CIa
zN$D~pvPL~0aLIWLXv%c4%f#N)+a0-hzBYXzk6-ipX8z4JU`hy7axSKk@vD3zAMyx6
zI%mY<(2_T8DK{oy<b7Xn%m*YH3JQ7TBoE%7m1Q0qj^pd^{rrCaz#YPBU!HcfEIvk)
z!<Z}xr+{Vq25+N|HZn#dRfPU)jxC3-!%Za}<MxsQ4a83|hxcM(!;n^7cblqD)2=zj
z@_^5pJi*Yi7fr}h50^kHKx#az27%WXeC&b$D3is8#Xyvfu!#NLCl_E#3F)>q>cX8q
zH;bOl2ZG&n`;Cdb2Jx%XO2l~0h=*}$boejg?CY}732+X_?`@`)PWytWLtvDCDIlW|
zBV{wON;hdW142e6C3OWlSM<a@qwKaLNR`c>>TgNP=5nxZS1GBRTMOAb0wdN$7qZoK
z`Y?V3j6$TFignd==)f4-5?3e7<K-wk|IzV9nd-am4*W^Sg16ym$_4!vtG+H9T~12I
zDn`YA;*f%b?TAYtS-}%M3U5xrY+Bj7{p_t^t7%oW!X1mrCY8O88L=}iJ0O#WX6pl=
z+I@ZW+(Z(IpGE%kiYhXzzwPevt5$Z(<xkSd$Fm^Hh7!-19eU8Xgu>0)QT}}TGw=uX
z1L_W`86?un_MakK&P0)1`sR-gIBz=8P6ZPh1S=;G=o>s3C8*)(Gt&dRrt^39B1fqE
z(leNe>x_4(;S*Un{k_6e+M@U)%jJyJJ$wmsTaze<6cNwn4*(|mZ6EGo&QsJa`a)vK
zC)!HV&QkpVH!FtupYM4noQafrX0VrD5qV`k6{U$L?MF)P9qi9{O8L;5!sOssYqamE
z-6ZmMFVjXNB?jPQAvM4uRfo3U@dKei`@QSb=D(Cuh6zWl$r#<s4Lh)#PvU)l-^%yr
zN8hIlIR-q302#-2zz@j7mW87v%uQj7lzm?W+Jp?>b<7x|2VgPyxv00}x1gjPl(XU!
zWKI&c`o57wkP=X$RsPU<zo`ChaJL7lhFG5=3CJHr@HL0T)dgq{d=csKSueq4LygFI
z26wtd-2<<rCyr!f6nUw?sP$So|A0R$D0EF?8P<}5ZNqgI5KFML7thGf$88+=F{2rc
za&0EWCh*L9`0p(kPo){Tz|wFSdiQ0|>xy6AQ%@}MySOFhVbEW(nOmR2;4VE}6<He3
zFF5v{4`3>~PEx=vO5)7SC%3j;1%9DL{>LurGeMrT+(ajcgm+KMchEx)tHC%If5x&u
zSD=1najH{QgZ#&e+$<1&TtL;W{N>trgzBeQZLU(OI}O1)ca(w;)@96gI`rtB<1;yz
zyxpqlgo$IU$k;Sfh!q+h<C=*x8u@Hf{9V+rjBxknQ3XJMW;?V1$8&;jzr%SFS0t%a
zq0qO7^U)BMVnw<;j92}z%ujXP_xHe{daXgP#tqwHhCOY(Vm!9ivF52oD~AE0L~Uwj
zwmOhTi6v+sNhyB5i+4Ke`Kto2ftb*ehBVyu%zXQ)E2hpOEj{uNOvzg>hv=u;DsVP8
z2t@32Crn_0<q+>I%LP+0--IK5`X!>>UX^c>-4ZsoaTxBTO`*8NE4JlAFe9_?`b$QI
zQiFXW=53;V-O7Hh!h$xDRU+a+$&z{FD1?38V9S5*k=dgiPbo&z;{3uK_0eIF%6@*L
z4K+VCh;~{hvM7sSD(Z|!B96@VIpv$fUo6Gr1#-^uvmG9JbQI{uoEXwgH5Bk8d+SQm
z6>*>)oJ{z_x|J}FMwkOi`SV}TJT~6&jozbr`nNuyrY^kjH{O=tWcQ!pOviE|OpOg!
zDEd3ZP%3hK8X({p{G@w53YMgCvxm%DyMAHKFPz|An92pxk~XlF;%-q6PblA;^UL77
z03L(;D@-;e#Cjs<v3|F2t+}ACi+{=KAzNY|mfZK!0IL{sj&DdoOK1ir&-Oy^ymd3P
zW&t1xf{ZZP8i^8^rGgp9iTZ65{)f6Q2`GF3A6Hv~)}^w>5z7f0t;`!oXejYOY{&|r
zl=X!;v&=08E62AMa*+*mN4)jgPGYzPEWp}|_S$~Pa!?R{b7^C~UHfT6!XP?S<&0n1
z>rTj3K+4%DM`<f^0Fpsefp_j6v)QURovc6n0q>hxJi=5Fmq<5}c4OoSmj&JiUT)xE
zr};ac-2d@-s6||XiKWH7V~1EMraqaY+I7i?o|j^u039y-G@0Ajo&Yc2f2KGE=tx*Q
zO)kzG53I`a@S<oQrCT01Jm!~q@B$4QVV$mBT+4cwXsH_3<V)fakDfcrF6xBkVaHH;
z@Dx2^x|@Woo+VARXC<`G(qINI9VCuG@fpkzc9{)O*@WJW3dmeJVyyoAvX4>h_fMKo
z8JhsVcuiL9s1)PQ;;)AT&oY7w0O##JQ9pHa<j2-M&Ze8JwbyDX<e_j^B)sJB+h!PC
zEK+_N)=pZ<i(HUtYA3r~y`O&BWzRAFc)W%bcV`a9%m&T5ugrxAezQ7m1xFuM^!IBS
zj!rmxuo2NK-IP6wHGvPR1K7uVUe`{jSN5onjmlp^g7aHA&0<BJ6>=hYpdtc6n(NC>
zV%o$PZk;2IU`*6dK~8})s`~;$b0=mZKr#iJZzx<ZyU*x4W9~TdDv?%jb(5KzIu?U?
z#<1p@!FITjI};DJ_4ld{vLU9&abSow4&y2vLcH7JbbvjY=_4J7@*4NEBtsn0dcw$)
zhSOUflcxyZwX}S^S>lljAj($erOR0;%&OcpNKdsyuhoh#ccZt5gRuK3e%#af>yT4`
zOk%Sv+HOJ*sX|b3x^;l1YP6~KMK^RVord9`BpTvHN7rYo;b`4}ao*<&iLlXab$MnS
zfBp9oODmS}VX1T>7NUIf2iRMB$bs@UEvi9*>WfTV42=~&p5;BD)l-2s{*Isab^30~
zjp^ASy>B<r;Z{}^94zdJP<Jd^7B@h(rk)nl%+9ek{C&LwkD<|Yku(t!Ph;sL=+IN+
z3qH71@Xi27J)OX~SZR6OXW6jEM~Ahv<X@TR2zJ`pR4oKp9T^+k{K!|U+`5B}l;2od
zOg8iV$!#rVF`p&YGJJ$UFkCiEv`++<V@x<nkiXQ#uOy@8kQSW!>S>MMd)pjYR>lTC
zqrbZuOxVfTPzK-Ht<j~b@4>K06^D64jnDRrTKPhM!cd0t={5eL+CCsNGW(ZhytWWN
zr9!na6dy6hL(E7Ik!N?rK$MH@5=)9F5t>t}9`?;x>^4vp=%ox9tK|#6^5JK(roRp2
z`Cy=wX51{zkoc0hT}ub~+EMov*QA`I^KB)~+5CRyJQxPgcT{;$5-^o=%hCJOY3Os-
z?51k`RN;$SXgY%M$B#neYv0^{&&H`RXsLy+bC^FeG>LS*?{A*^(2Hrox)P0ll_ge2
zXa_VI((l1aAkfP2ubG%(+b_A!f-q#SkOIVhfwzivYED(lQok>gxYuw)X*g}~uL4Gt
z*1p0yh<Dj~2tSs?hUD1sNx?3q4%16GlTT*HcLm>*oO8Dh!cM~R>i}LOy-EyFq^5V%
z5JgwMOmVl%J9S6e2F&jGt!>Cces<y03QG-h!BdkoFeYkbRj`Mo*3k=Fgk#jT>Ku%o
ztrLW#qeD3`bw$sFX$e2B1*R~aL{LW0;S9Ia7H<$1+t|xU(<jk=SQB~vaGamThvV-K
zZQ5o{*EK&ytG%<H`%BqZ?R=$O-Wm9@`ggrgUwfO;B6dwMynW91q0L?<%CY4j#Y9tD
zXw7pL@G{k+OdN2Sqs(>ac5;24qgxXP$di1~xN48ISevV)MfSy=jnX({&sOwQN(V12
z?>iuc+`&h%QiKLB!pGRnvnf-8y$cYKZfqr>J0=5(h;WFbxTYib>Hi)EOVGzR{%}x-
zRMT`~Nu$kN%9W(k(8DV=Dmyt*D9n5L$*z3};0Fz>6w}wQsHTu)n`J37SBAAl)7M!D
z`i{HtSWA-)szCT&`Ps}$Aw6pvKMJ~a%;M+6A#7HsxZ_76vzet06;BFf_nM9BxC1Cp
zP0{ruAv(UOu@dwuxrg>!B+eRbq>{6Iwb2=2vFk}=^ON0R>HL%Ya2Af{F0RfNCJu0*
zF)i?~pF%&i_YfE$^F(ZFP~Z(5GXTqpS3S)a>H)ZQ_yC^oo{#&87FMYm#^MiGl@Q)N
z$4o!kN;Ry$q)HwMc6k^nenPPq*$fg~#M==3nr}-dV!x(Het84u7Q%z|f5kTt5D4RD
z?9dv7hKpbnQOfQ`cP(L$m@Tp?<cw!r53>63dVjt;4k)4c@(Fu<dcL}x<mz)p9r-#a
z@$m8h0~nDqH<0qXDY^;0pLPkn>QbAJWa@lGzjwt->kft&E~E=amptv$UA*Upfvj%f
zC;_2}Tpv&JKjcP7q@T62BB+-MKQ$&c`GjoGw~@NPQp?oP;pd>Ff^}mHNqah1-<Ch9
zmMITxU#nd%-ewf<f1mdh(ljDw@=8(;*_5V*Bix-?2iuOYPBOiHBbla_OS(PFzbZ)j
zWvUWf$ZnTkd;BaUUXR<0IW0Z>^N5%X&jv^pR}(Ktklx5w)JGXr^Nu)_q7jbiM>P&&
zDU6&lC9B8)b%WE9Lcp|7GrXS5H(%00qEp{*K=%ZTl2)aHgPOsJlTK)CayN0REg+6+
zKP5~}cd3i;_uIS#=(wLKaQUSTgjqRo>LA4~3Cwa5?m`dSiR9l;w{fG0E+%8mD}h^W
zz<h@Idy4!s`OYPVQK6IyU`{rX@gHq;bksPQ?Xh@N0*)zM#gJkCr?-23cXOeQ5VBp0
z8OC|2T`-5mUpoi$hEgiaIE@5>`Obkh;S@;YEx&_Ra<1C&B{sT<o2xycWYB}6%;Fj_
zruB60$;bz|$eOw9Kle%TM`5lVKrJ30)d`aM{9y){{*yUw4$^Lq@5i&^*;ba<kIiw$
z>ZJ-Uv5^*wDK0do4s-n@cinmv^!2(y@Q}5nf@8LQ>M&D&KBkB9L|%f1ir@{!I>eJ(
zha(48_FK?{03F+bI2E3<8o~`U5u6lfdwRoqa2*jhrNH|q)n6E>|MdJ2c&LD9+xvUY
zYIBFdlPo&O48RfgJrCqd&UqbC;O3E579etF4ew#5tL;lW7}kS~QqM|rd)Rbc*QptN
zRL&JxePvn_;e2wBcm1yIcut_rW&hBhEsW(&=sW~}@}!ep(aOvb*7w?@8$=j+Iy~?2
zVGPmJ-=~-c6*gd<R;-W!akxMAA`!z2=NrQ3uw0zi17+n%1Ka73#>w$`*aA6Y1t=xr
z%(KsOqo$glYR~v9D>V2b`qvFqTav|h2X8{C>of4lPU+Pqto5DE(4cS~^+fs8^n<I|
z`Culye(op~j1@Z--uMyff>ch5XD$LctmtSQ5>yfNvtg__cajbOiTIn6!+x0#h<Vyx
z1AYp+IS9eNt+oLgKc#Ql-rWO7Qt5uq)$o~HHpAd4weDbNj}yWaJ8pjlNY^0+2JDs<
zjU<Q~EkZfZgVBmEf3pf^k0xy76!fv(wGA(yWAb|ncEDs**-OxFpzmwIta&B(-YWbF
zcFm+=%S!D&Yi2J20t?a@9Sk)cO%~g&lWKaHU?p=xJtfoEHI?U*;3F<WFO3@&Gawrr
ziiB}v^7ImQJ{?Z$IGZpS2}Nsy&fWXv=2_`~2qoWA163{B^Y6*EHV}cpekl*ZZtS0s
z64RmXbLOytv4yVa68mC-G$J*Llxjp3sXJ_t7qPe@%#q*#HYxQ~aV~06F^SX#F?)XD
zdMg&R0kW1O<@8@%36zyqvp`94352wew(AO4-Nn4J10<y^MO=J=<%z#-jO)lq3v|g_
zSDiMVL5TPA^fN+zHvA!4GSX){#2gmY#Km%)!biMq(nK2j4J-*`;q!XaUh!WSpv*s>
zSk?DmpM8aZ<;@2|;|~FKX?%Q*Rf?{<L`X=ndJid!@1KGS-a|`4ytLBq8Fn5fhZ%WI
zctL}b!wv_JylA~oZ9O6a=zD*^6cY}HtNlL!k3ew0yHY}(cNiN}&+V&PLCacl76)}c
zgcA&u7}X_nXN^~>#jG-4EXeZyn$_&RH7uu5`e-q|^%HN6=x%>TVFg`dnB!`9E3!A;
z!mp+XQ|e=^h%gk%qnuk2q3cW_IiE`ERV(sps5)v*UR6!5<IxJ<s(qVrS;1Sykm4Ba
zQ;rEUbOo@AFnEqW)-(+@j8#pz>L8bn>gJnq;h<MaqJk=l;oujVMwmClu>!MHjF>wJ
zW@#Bh>7$DTUK@YB(!#i79^euQMx9_JXwg>I%|C-iNH;Hi1yYP5!7-4cVW`Z{IR2Io
zB($EfGV$IXXRNkExF#B4+`>KS+07){GE#x`<T}~ba`Ls3%6ww5wA*5qa6ml;95NgX
z?e<dBQdcV0Q%WsG>{v@d8){V(_V1~sNCK45n^3%1ObLILm&xG!YbgM}@I33kks;K=
zZ7ZdanIKlh{-qR9eqYIZboFKujhvAzfgcnKf^<Zw9OelYl~<=VZ9Z)_WKPk*qZ{i~
zE^DaQ7@6dnMfS77wwn;|4aL|MyS#nnD~O_EAa@V;d_|I!*eml8TnB%ZeH*kNP~Fbs
z%G!lwVC{cFYZr#e%G~sopT1@7n|WMYJ1}r}b$4$j(P-8L{G<~~lwU)2GCLhP%gLQk
zw(cv+sPQ5m7gB@9hbO~fp#0v`=vVH+ntVsf@@&z1<e7rqa5Xovqgd`VgrR9hR~_@$
zOa0kcCZ$|q(v__^DKniPOEqaWo4FM9Mn3wOWwc)7Q1C2e=w&?2>6*39HRYgpvwtXx
zr(^O~csfVsE}ROpmrsSsA#6%5)cmAwbHgQw(%2;QG*eF8bX<BhW<~JBX)Kw_=($BU
zR58;0{n?1BH<&dweUfSV52y3PhnK-u0uz@}5it~(n_n*}moSqE1ed@?FAIN%FJNI{
z?Og}Rt?Ax^p4ujt=FnXF_dBFUlsJ<7REi#Cucgs956+WZ-F@qH_i=Z4-+ee)?1G98
zg6ZDBc0dm_eS)T&-gN;$c?;e9_wLKH*2dFypbNFOmaag5NBVaACkv4%fPMk&p8`Ur
zKhbrh=g`xY_bP2caC8}Z@=|}E*!0X)nl4Y2KJ>VB4~R3k{crD|cA8btD*60p{*u4m
z-gkfRq(a_Xr9<p~|Gvu)(1Nnjbl_F8zWecK_jUJ+M4g%|0+~yD?_6Up&}2g-^rUQl
z)SJsCt&s#rGJ_-akuFV2IMU9M;*J!@OhYDeWJ)wWjhoQG+y2SxY|MZBt336LmQh1*
z4uvF0{rK^%6L&s;td=ka6#67(HCh<<7R00y*;F7xY!EB8MP;m!)cVMpESVnY8z$e&
zOAAh+3vy|VRpiCP(hdltp0Tit^!Ea!bt)BS`SP3Y{-9@{h}P1re+uRI%0+j7b-!lU
zxqutVc`~k1A(?Z)x-);oIx*}B;cQamG`B-sDH(Z3wDvqKd3$ma_iT88GR_jXQ($T~
z0djjFXNHt>23?k2U=*-f=&JhNGG5RfqKb$Qt;($sSNb8rdUW?VqICGOOfGg&xfHxz
zO8PDW#dD2JoAwZ%IZF+>hqGF__sZY2g!n=!8#8suZ8>AT_E~=mJ?FS1=f#*Qb80V`
zmN9*KlHK4^X>ydwT}V~R$P%zr__nw1x#wd%GM`aSrp)C^@1IJIxkhB;9aEMBJv`MD
zQ{d!&ri1!cN$Gn7Q(i1YM#uu{6H$jEuAZj(9<PX+ezK@|m8b(HhD=ep0V9R_{J?w~
zsSxWMi>p%Q5Ym4`U!`Wg;HXV8SnIAHJd|)!7<|l(mAKWqZ&|Q-xnWt{V9~+bL%%py
zn`eG<U`emyP%M^;o3UMysS35*_bv#`F;v_sHLr?!9y<YKzjxysmbv=Ty5ss)s=yru
z>sBM7cQ3GqDs-4Cj03B<Xodj_{g^Ns_v+sp7N`bkQEY#}@XQCAI|oKFgW}7sK$)0Y
zi@_DpO$`0Qvc(>o`c9#P?H!(Jvs>FX1XjbXm#MgsHKm6g-bHqd5;l6Y1ap63U5^d-
zk#o9`c;ifuW4~_9z5V3|PMRI1xT|5VK_v1jkFd`*seP1QG{C(ydca9lwm2yPhlW>L
zSgXagH-Ue8y)g}33`ttC*QK`e@*H8&QTd%Un-Hifdw6(|aM~$N)qT*US*e>_exP0v
zjF+gQ<*(GJY36g1+DdQnx69$FvCu&xQ`9WV;k^QHX14H?3=zg8os*e9w0=SU5UG2>
zglcpP`}TjG%#d1o1R?IqDjAo1YC)UDt%H=#99)0?(9F2GLySM1l3wycs=bQ3aS5Jy
z@!Ju6&6N#xku><vBz#=Dm=uM`TyqCEkFaaPQ<Ya%OA7$rJtJG~m80QQ0KP7+t<6~g
zq~9)k8T;l?WNJ6S{>%lZmgQ4%kdkk8X=oPXC0Yd$t_hd{xYl_V^1lt3R^AH{(-LnT
zG6H}7&X|!hZWc6i<$p%hv|P3b8-aKsZd&S}5IE$@i-8l2GR9QK2X5AIY((*Nd)ahB
zd1od1Y{xEGdKLJiAU1bC&r7R7%v#%(3aH(3P-()ZK1d?1Bo=A_Ox|>0`y98mPm8{V
zeOkn9AG4D1F4;#Qt=Pv=`>4l^A0{2hskVQ;x7o&)Z6bmEkZt-VP9SRD$iFlgZ~>U@
z`K()z_lqUJ6${e%N4<GXa9em6Td+mm!h$VgpL3R5VHztC{OpYt8<<GLh@WMlW8J&C
z+XgN2+5j==*HUwHe%rvnao8f2jbmIrVSo@-%9UR5zhmuVP-t<He;PNe*J57qKhA$w
zjK@6$;$y<>Ead)rUy-SRqK|j>Bh{k>$_DaTPqu+M@}><5V4?<kW`zeDk{SsDak(L>
zm8qrPupwEBf14aFK^m*<9UBJy>IkIO4Fgm$&DdVCXO<OS9mR{neaAJ?dHW%PIu~cm
zmAVccx}cs+5wq2ZsK#T?Ep}1Zsat>GwM`WE*Os{7U7c>%PYwgJ{gKxopWVORZc>d3
z9Y_grC0j-nw3ia<+qqH_HQ*%IN6xjK3I?xKgAc)5;(TJ@`xApZO^IHmB&`GP+=MGF
zITb_m4)Gnuu)4l5$9f&?XHH>}<EHh^2CH1EbH{KRg5kRQHlTQ;ytUJ4#_50A&xunN
zpR5d<<o9gFa&II!u4Pv4`vmp1sTEyYOK`E4^eJ|soD9avhid`1%wW79Es)nz*iejG
z>a+X6oWJtU)#b%31{iAXJuy0jzEB6x&NG4<wG7qkgnvl#VS7bN?#^zntVljq<_)GU
zo9mXFxyS=km!-%nCpRr(ndu&9R<1q9f}K}pwA{nze^rq9=l-ure!so%{ss30_LuRG
zJQD#kmoZ2}6ag`p!DKurmzo0$oB<BEHv|jF1AiU|Htr!>`frds->J&H8}*7d+4&ga
z|E*0gEa!7nR`4pCrjJ#ucgwVbl}NBb9)vZ7VMP_I-iY;XnO3kO3dterF<5zfKu%Sx
zW+T?SWm>_?ix<c~2rKYT4t2%qHe$V7rWLGG3gCUzW3WmR%CrNZ--z{YnU=AJb2Oi7
zAAf@tJn}ZqR;;IOSnro<1*=Vo=YPhkBSzputX8heYP1pW-EytqH7PYb2CwC_0B<Y0
zEWQ!%-Eytq)hRtZ2Cw$QPzQHon7UuA6}T#8h{wQHP8f?F-i@K^ez{igQpysK!An+H
z=<R#li1%)}*6@N*WPav&@NN>@4n%=0E`KN9TPLFF$t#*hyTJ+VZCGFP3oai@QM%P9
z^0DBQ>zArlA4~C!YV}z1fx`tUgbHAYPap9xJ8O;AjFsz4Bhc1(M{-<HNTml_y)sMp
zk@19cY-g8N9B-AsaCVM#F4v1Uz^88B=Ji6=`wY5uj4G92pYO!dSL`ce_2?EM^?%mN
zvM+ZE=kYDZ=za*MHRWwk4||0+*0~R$OfK8R3yFDY*~bds9-i72=A^J4%GX$~>&oR;
zDr)=LwM_<Gl~t*YuT2u4#p{UdBt{8BHFlDCA+3@@F*`$7(zWfjMW0hWOD^9fg#Ci$
zrk+_+BRwTH5t$no{*&s{zkFE*<rSJ{roVQ{(CN=@xVS;Qb*ug#JU4omVe~)~1Tr}}
zGnWCR9%{Fe3Jdl^mvL_k69G1t!94~jm&8pAn}0wvgE{^B_2UaqBK&Ywc5n=Z;ChaK
zZIBU{ng7ULqGnI8^pSf;NGv2ai5ISDW+b+e9!vC7hhJ^{Rhc-_zo>|9q=)$7N7JRp
z=gQR-{8L5DBldI9XB`CEQiN851sS6t9(a6DbLkAIpWSmB?!=u^afF{2uYQ#Aqt1zw
z@_!zl6EBeNDvdBnGE-&<9H)8x=yOyDqX*IsoJpmvrn`1r41bf7=AZoZ?IlYn5SRts
zOXy7}%wpzw@@fpoUmJrcRE=ZYx}GYT=H2{SWDdlH-;DV=A#+U@f*C)PG!*n^3!*x`
z7i2sM?>63&;t2mvrz=eMVqwl+q$RnXoPWyQptWRe$Q3gN+P6Wzi{cY=SR)u_gk8XT
zV#yL2sz(@da5zC;{1zb(;NobUkzRzbP=-XAC+Cf@=cis!SD?11sI+C*&1!Aln-3Vo
zr+eEOH6j@~T10s3TzLU+#B=RP#h2|3SNy86L0dE!f~;lA87*gH6WzIyMWq>5On)$3
z(kwZWV3sZ@%iH>A`h31ACMs*0Hs&EQv68ywyZT+vqgphOz2rJZf+|i{S{>y+fqv^{
zC)__$lycg;n!kPPzI}eZ{55ecB`pIjhTp!w_y;I~Nh>HO5QaGX{m1a<%YS;R$(K3f
z+y_7ZhdzZ!*!=QSsuvt5m2j<<YJXG$Jj|iA_JK32WGP*g>>YyGl$s<e^r)jIJ|e|X
z8^{FFHrty#E5)R<d6v2m$Ofa#KE9FCg;YI7k&M#UIR3NLkK-LFm`e9Z-dl=tCYH&~
z5Eg5=;0)*5?M_g6PmW=lLzU$;7irp17UG;I)WzRQTvC+HfRcAsgE~~ZU4JX{wTVwv
zN3j5Mrk7I#-xa~oYG#(JIs~zqhWsda>)$D<@be@w$1f;jXYxgSBJA2co-(*pLYKMt
z`x-^5t6XEpoWm*hqSU#xzohb>p7Kk6s_J-oZn(fy%$p){O2P4)mm}DiN)a}jp@~<d
zkIB*iYFU7|1DUSGPiSelS$}YdU_ELkuU<;MHqksaCPtOU%G-LdB{Mp4xPMet?<L6q
ziDMx%zOtGhs$7zAj+sZOinCzVW9d}Un^p{|LPgV5C(|wh(-^E|X}X8yIwiaV<(rLK
zTe8G7DoA^GJ&zoipn%tD#7BWzD`Kuk3c8=KiIogBQAgFs2z9xiXMatR%}sa~GGUf=
z14$#7E;<D-pXFN#LW2@ZN4fM(H&!GP4O0U9iUdG0rOkRt63UH*WR=n%QpbU%@Kn4u
z)Uh>W=o4!G698_0sesLuM@v}nS*(CkCAFCCSUF)LMiqW(<3N^xx?eP-)#;=m|F#&5
zr=q#k=J5A9amy@ZF@J_+135q`Et+3ykuxWf><&a@OD^ZTs%BQJy&#}tgiB9tk>6|1
z|K4r^Z&#R8%n|xM_b}1gEmQ@!Auo4SLC&u=fbf>vi+qPFz%fPl2dSc^wxWug!MBou
zas3P(Ot+~Mn6`F8@w(`Qg8StJ=-&wDq;8f-fMEjEj`?q*Dt{bvW-8Tu>JSnsA&H6J
z?yx8Bt7O&K({=$oYg3fgd}q^9+;(rt$MPip8p{070~(6X)-tdJWmac9`e3XnYV3sn
zTu`HBXwB^0>X5q$YAY3=s1p_(LpjRe?oqNEL5W;!%zYC@OXezJLQ&02V-=%o2~CQI
z_c(}jhMHepJAX*7dgDT3>Ks(Jz-Shngs3UmQ+;IMO<}T|VZjS>j_gJswS<%#`?>k5
z2hPo{4trof_nqMx1v3Z}s%~B))Xcew{0sG+qz=DYqny-tYkgnt;5LJ6RbL(mr5%et
zKns``T_|S6M5xJP%?&K1AldV<Fn?B5H~&r$7uVtT^M5P3rkuQ&U+Iki#36n!%O+(=
z3t*Q_sx0Q4s#;XzBDf#BFJB+kX)>#p%_WuQ_|T4?Up{Qt4x`mSis_VJc5-d0G@EEb
zT&u0#OSN!PZ`{aniO)YTLM-T7V@Tpz*Bh1-^5A-FIGKSey2h6BHb$|RYS}5bq<IG|
z%%8dvb${vSKINU0rZb&u^?~kC-DlSls9r$VuO#@8V5tdDD+!svm<DW^&C=(Y)d>NN
zQmLo;%ddR+ogV(^S6!&U{U;3lUgqJ$|AwFaQpE;g%eW^Dbmwx7FdxWEi1A2vB%!2Z
zoaDTj#7@rqULE2(Suzc$db+F2F@DiVhIl9C*na_klsdn*D>v@CMz@8J>9W7?4e}=4
zdKeJ=U)ZIHYu<+(d|ct!5%#?>(=_G%I(g5i+r~$V)f`suy4eV)8MpdjbThw?rd#->
za1O#e$IOMLq6Kil%uTnSuJ>u~IL8}E7-XRBqmiKAQd9IbhM`iEoqt`sW7*-WPrAhJ
z>3<akU=smKQ?LcsVX-#g(vk`8o&=ZHl)~Br3(4_SwLUW4@F^i=-oP?;qaljweBLF-
zM<46Nbtk%qQP@D&pAb?{0@tsONLx20#TIqDhnVcqjeEC#Cn`c$pY2G$%qFkul*SgA
zM}^+N5-A$37cX~!iO7^Gb*DGaX%3$_c7K09svOnJPzu+E05p}~&%r%YgN7PYr>c2x
zv^b2_h7M#bcpjb>A56FKOX`(IJ}E5<!L){HS`>O_JJQt0Ih**lsQXMWSiowk;W{1>
zC{4$feH_MR18*KHh3EJv;cmv`EwVlYz{dk_>zPApXj6xG(Q;H<$MoS(4uqrJHh;E&
zA+?C{Xty*wVLlAUI?Ui=8S)&MIcKKY3r^tegBNW$$WUcwNS%p@@CFzOT4}3n?VrIu
zNRNxY0jZ6p%2OZ(OLvR)df3)stOT<gd%}2_==4g4YFmgu)7c%U&hb=NSSa}%zw_&D
z<M;Mr$06KMjfsWMb5vt2vqIgj8h=#N)czRN@N_)v(ds!2<A!>4ES;XC9_?4Ibl-4=
ziVGK2laB+*nTAW`ro-)o?XY+ouqny3coJ$#Gso>l2;XJuVj&9k9oTJFyA%4O5N#lj
zv2J^G5&LMmg<l!#o~IPCAdxvYv$RvM5PmK%X;dBk&V`@j;%vYqV%_o_kAD&vHt%yB
zTI>y*#ZKE)@Otz!?vJ8c-j+(?j4!0~xF)%icb9Y$Gq{)>cfB`(J`BhP&^%T%&jA|z
zIZ{yrJ^fF`aGEol<kJ7lUF!1;GXV>CqIg)^brfB!JDvj~7mR>x{t+PWTIC(mW=G;M
zh&v<sVPV%Xw6QpN3Wg3Cj5deSd>d1r3142}RT%#5t;>u%jqR{N8#p1e#@@3#cPoW`
z`;kOduTx#xj-2d!cK81SfZ23EmjQ1M69F)nvF#lc0Wg;VtO6&0wOl!mEVmKP_gBmx
zSc=7)z{3JN^L7j<v-~nak!>ll0xR*~ckyf<eX@Ik-kIrU7mHt=ERsuaf86NpZ*TV>
zZ$CY`xM^Y}Xm<PY`wj2`#lJxDrjj=eKop1Dk3Vm}exsBryzY1jN;$_%<9CCf^5dO@
z!4`nO06yNe0N4J1iPr<4LulvU%WXXci5J5spZ*EA-(o9@m%^4x2fBC<?vDP)Z$Ey1
zqj3n7SmnODkM`HskGKE6wF6R|5=PnWpMSo&2Z96PhTViE2Keo-f8Tz8`+G|@_eJp)
zBHbVS_SYY`=FTq>QcDq1s+5!!q(Bj8aZ}0??j8k5g{4%NPFV{EIq7L&mI}jhpsZRd
zkMQ;m78m*B3Za0WGV%ru=)#dpe|#_1onKb?N2Eb?D1bQJL#8M>;ns8%zgL%ySqnLT
zM@`ZFrq3uz&2gv@|9AwdJ0${25l~$(jzL7>T~TL>(tl&lylDDH<t#=xCIVJ>^q%QG
z$9sv0(2?i;Qe(^#t4-Snz&uA>5XSov-ebSYCPim5^jqCo(R#)W5vCg}%aYk;l}v)`
zXBjZ3CQjhrqpGN=UalJ(@3Ht@MD;TT;eM}bAD2p63n>BImvmYSPXRBN#99j`e}R;y
z&)H-~YDRGIZ8C$-Qt6b-Bmz@+#`k9M`%HR7InkcrufNgTzwp`5*w%&D|M`}G=1%DC
zKevClTxo+YipJAL<SbhwOmB=>V<Xv#gyc3z_b7>doJnIb!{zH_QJ_|~bbYS3kmZ1j
z2mFTo63bx9a<(F}vmW53ZM$EBfA-#Q&VQu0Q2|>jw0<I@Z4}T{8kRsGM`r_dZBR$W
zi%>V{buk}GsLNfbi_o&c>~Ro%HH*F)Kz<mZE#xIpMxL?6!zA|k9!FLfHR60)C+C$!
znwvq7p%tH-*}B1i$<CR@pbvCUwubXkVPX=NQU%{q_=#M4Dq?6;p_TZIe@O+Mf&qrR
z*gB;ix%X#m6%=M`rjD->ZL`w5nYoBl3K{0n-$L%FV>tHQ(a>+`%e7M9+IRfOU2^gD
z!DS9(xvljhW)QPn<TgHW^Zn(19A#%*-YlGX+|yk@5n80m&6MVN)E%aoqVFq&*m#HQ
z4^TF;?HYSfOkl}bDA&2XBJhZp&e<zIwNW%4g3pS58Ot5{ju)LxzZ?LrHyj9fs6IcJ
ziChaI8WD9kCAM+GntluM@h-hyxJr!I!=&vretwUa#9Rw00{l#u?_3K(e|;8dWGO*4
zo9arOWGm1uRp(vg<227n`-D}*nCy8NkC=Uy_89pr{hId=^AjG$F>PuIr@6?{k{`!3
zTno`=$spW3aiiw9nr(zEUuZ!j(`AV?RkMy%u^2~DnU32090oWez*O6se;w~PR{a`E
zU77U&MIi`A{+8hR0$)o}e~(j))AE+$=Ylqd@><(^0Br4MW<A(;d?bbw&^_96oT3Fx
z?#zW0Q#o;c)KhCggoF#wX26^=UpSCh^=K*HPf!e1(Cbt?)ic*A=GKCtYrF*PPyuZT
zsqukGOSH5XIwqzJ!E3|n4^hOZt+n3pb-6W8ODXTbjuXNztY8epf52OX6*(SO8fqJ1
zrEYyDtkjfF=tBxC4MBT8tTYra2rCUmxgJ)U)+fSBLrI(uE7yrHg_VZl{d`zi<*^GZ
zr_oMfr6FX_hm}>e844@cQA}ZFAJ{3ZoCiCGmCI~2hn4G~NnzzOTghQ%RmHlnQny|T
zE0PgKZJuDlac={be{jWHN=qRnuCulMs-mzE7db<SD^?AUc;4ts>s}TDF=%`rRNJ~;
zAxww~^jtF6@cUK&@!(ifUP3SizPd4=+k9|e9t3yucUHk~rY#P+LisE!lbhQds9R_9
zYo3r7$*Mb*+C)xJeKhA)wZNvzX{tw+m4XFO!ZwnI+9JQze?sI)K9j;%-QSrI>lV)i
z?<js>-ICF?_RG@G;ShG&VAAUG&cURB*AGU}Qdq30C7w|^YS5pGO69go{PVkqV$N7o
z+*UhIi;+Cb6{EEF?d#F^;rdYyxix35yiOOSASIEsLr31H>UUi0n&a^JRD@GTIOPX8
z<RD$yZHYs#e=Fl!!7WB6kfm1Son8<}t)o*C<&KcmePs>R9h&h~?A|+){f2zD9R)pM
z!`>-01TSz3EyZ=G(6sKILaUDQo<JP)g)!OM$z@Eon%0#n9@)6UiMG8e(o=#=8(h{U
zIT;z61F%kUPZ*X__^GZoteWah!}>mMSj{NT8&*y5mnmQicz-^YRSM0~a<imTX;nXa
z<hV*J&9rvDFTEX=X;iD4%_Wg0&sVo*aqp|&48od?4)@8UO?mBEri~<c?d957(o4ec
zkKy&)JH<%0D_&+(h82&3GZw3-y32+#c~`shY(wfna6+rff`9Q@=uoj^Bvl(ypz#dE
zxxak0orj(7ZGX{Wc)u(rZ?fAh%kb0kvF)<XWI1Ek@0aD+0dR3<m)>C#`#57(5zz8A
zJ2#oNn5#_XoV-_)muh8`<4cwdQ@d=_F#+7+8C*Uvw`Gd&IJ7bEdBQ%gq6--}fR5XI
zAPLj9v6!>%2o65q>4Mq*ov7my-_RuIL_u9fH}iWj-G9ul7O159MVhZQ#t6Bf`ML<y
z{=CFa$qjS7frR4e+`*MdSfUmBV}(hR$(b^rKO++LzdxJz@h%vM$|<Bp;!U`QhqVEh
zWQ2<sp(Q0z#SXAyYo{z~j{WA%YS!25SAuTRLslI5eFTr`ydQh+1a}mK4QvJYwUpO@
zD+E!+K7R;WsG_R31V?qW#o@^5*g&zHwJ51};cS{DZ-O~I>J2nGX1a{M2u+UD@nX46
z&V}{Pu&Z6S)0_k#AHcBPVbsa)=kT7%N`2O;;sinNl)*4o8!~{916>{#FQ%LM#ZR}6
z@S3oQpT(($DJ(KIv%9!(W0NMW%RV2yye}R{On+q)6%RE&cIwA4F6(%M4Vl7gY?NZR
z(~=x7H}un#PH%+{uAq(`&P4OSe^4QQxV1h1?;+yRXCGVG^zS0L-UstI92+p3kcPYj
zW-ykRVns57w+~)n{K<nNG(()MumG=z@fps_<Z$ht!A?krhrR)+4!O!}Ak~K0Y5jul
zIDf+xVeSu4zoVj>8e(_0&rYZK+RN8YzNX*w84vJ4I8U_NUhJQXjYGI08Wl2~*N8@G
zd{G@2jp+Z+#6_a{+~0n;Xoex&5RD9J(`!T{6|qIFSx)7)YRXWHC7tZybCWH;7uub$
z4iA3=R<}AJ;qo-=YPy+U9x|bqq*=fvp?_1AxTrpqb*-=ES$8q(+*z9#AI}C(EMzdR
zF&8EXOlNEtXD?Mb&Qv+3nw%bHlJ8(>9nABgSQ*{gqqd)Xl7BDG!+31qOhXFu8k~Xq
z{hS5R3C?OVc+VQy9&4RBtU6cKNPB8GQ5+uaI*K+VEw2HQGQudYPPnI)mbB7Wl}FKs
zL8Qf&wGAML2fU7>2^q#qaJ1+nMb6b-vRxiTUamt*0k+0zCq~1A*}zB;tMr*ArK1$~
zEutY+t;`kw_M!d!Kkno@hnE3w4if?}GnXNP2osl%WDClFLwNc9+pkYDPC<%b`Idj`
z|31Dw{pYEkkVPq!RNWsxp7@1E5a?B>gYbaeU;pa<eENG$HIF5!i!l6$?*97TRZqS~
zkp@X3mDU<lh!iBrS$Eb3f&6L^sSZ*a5`Zzq#D2MZn5Cj<9)wj*<u>%*!Qvu+Xh{^%
zQYPM@61rx8a_LV$O7Z0D3jZW&(2W#8A08o7G$=6mhT?lcacu2`HCVScq=uRl)R2N3
z`qiYkhLl?qGTKIj_&p5IC8;8H>7Q#i^yNNsz@b6Qc;eCK(J+p9D=G5-@*a_NNK=LQ
zr<XwSWGVq`1=wa3W{{-t#Y#3s)88pJFPZM7nUxWL%}GFT4ZVBn$GlG^qFv;9KWa(^
z5uC5b0GQ_pwNTj3@GXvOLsE2QhH;B0D_XD63+<b^s*KDot5gPfoMnNrnlRw+QB`bI
zFCHeuXDZ(7sD6HU-m2OQ<H=3nV~R+tzlTvl?`Q(hNJGw|g29MTx9xQ}UB`7<gA?Fp
zbUDv|^uD3Jnh|9WQ&Cv3Oznh5w`Q5WEMcN;M=TRWkggtu@icF{?GZfp)(w?Q?UCfJ
zz|><b*`orX$;4`oF0e;sgj3C2kJ@9h>P3k)hIU+H)(g7!O3W!?lxii$nvg6RF0^Ez
z^-Vz|#|2AfYrwj3VagW-7%2dvKtq%3Poq?SN;@CfWrn{KrSW$*#xu$bV_Y5MIeIE>
zit!SKaXjA-=jnbXY?Mgu!Z69-ewW=p>E*A~nB$KB_Dz4}VPyC3?jIaPZV>h)Atp?=
z;oTZx*(nixh-60+27{pR@i>WnoEaa;68$<^f;5<#ZXX0Mgnz&bC3YsDkT41Vwn7zu
z{NUJsYQn(b4ZCOAE)8?ewWol&x6XItGn_7|)8})&ine;S*EZeB2>px(TN~ji^gIE7
z9-$57$x<9S?9V~oD?tt|BX4#ge=Cq@ZFxP7{5G0LwB-MT{Aq+XkhkRS;FExNAY{mH
zAM7sR^`IgZ`1T<9=RmsFMO+8|S&%k=us4?TK%WG@<<K}}@ZE0kLj%6P!14>h{{-+i
z6>vBBr%~EKUmM|r`Xux@=`&is-wl52_VE+l{-v(`h*Iqa|1?M&;G-4+@sq%(y%_Bm
zGx}i{`lj*wub*i4k0O6Mkl2m<X@s_r2NAS>WczWL#J(LDlXd#kj;r3<^-5}gd`J7g
z;4IFabLe11-i@f300OxubNBKEHJ?!MhF!m<WPWUwHu5gF!Xfd{zkQoNO-Y=7w0rD3
zsK?eRjdo6e)QNv|11CL?e1_7W#c<xxJ`d@Jo&y=Vxu;pcbKR%slGodh7u0#L&TnNp
zx-c&aVuq8rjs5tnhYS;C>O-x6SmjxI!wAzc+06s+!oW||L`wQKwaYs`b|1IQAImGG
z_dIHkfO!G;JoN7Qaxukz=a&42PnQ5vKTg8*v@u?O$a!!JQjFNXI2O1?O-QN2gmwI1
z>qMK+wt94v&dB)tKGB0)<P*}Ej1$^69XZHjcovhZjV@RtK!-w?6KD5-ky1a7(o=2m
z>0B-yc*JZoJ!P4h*T}KW?R0FTCv)_DCMUI+(~toh9W`la=Lf_Vb#v%5If@bl2YPl0
z#(LW)7^NEWQL}HRsWC|*R4pgA<tbL`b&{EyB5UUpXvPamMp{v2TQ;}HId+n^dkwT0
z1(tw;2*<5*B7-r`=E@a+i2@p>Njc1;hpV81a8lQaH&iH`x9g{?tC7TzIMiBkmNI!c
zc{nKwoD`)6pd?h~L=O>+rwAp%KETEKRDPNYiN?F)Ts<9u9vU8nrNFwTHlO4LR05Q(
zTjR&<!H8A_5U59_9S@W%O5o0mTWgAc8-jk}q<1(MZYPqW#Od9CtO#VFfC^`O9u`Jq
zy(9_d;bZTb;sXci@v!Pb9?5TMunoo6dBCFBX;vZxvjea*@~Ws;%5z3uHN}NPyeNxy
zc5|wSp5k+~SE^n`a>HsaWgtzx;=rz}0R^?vxeYbAL^TqB<%pG|YWU--i!rNlfWHN=
z6PE@wv=*B&gji^QMWCu;`7o!KkCN8<;!(6+SUM0<C=C}Z9TlZYW6l^sMNqFBK{JuP
zO9v_?lvkJ9Z%Oikim95Sd)yUP6dzwY)+sh(2Z}fqe%8`aQomrSttq~)FCES7W>rtp
zM#Ab#Rj(wuWi?kakfvU7u&xHm2clzs;?hx#xW%925+!$k`O=}R6WaTYr9&Ajj2d^o
zaC#qAQCeC$(yXg-?IaTt>)eKgY*ILO#H5BI$ZC`&T}`-^t?MGOBr$Eocx2tyeV$(5
zO2SL>8;r1Qi*v=NZ_6`4O-nr#viD<&Yr-O@4EB||N@3kl^9H#ZsCe?V-T5jGI*clk
z7_k^@%$`ku6Br9J8On$&YK#2d^MS)#4qFZZ7C>0CtZKYRmH;lE)YwpbUoFY4TK8p%
z_Y=i_ijyzbO(s<j?wm}l9-NsBAwL02O;Vqas2ok#1}s!=JFUOA#8nix)ehxkCim$%
zRjTvj{@Dc{PX#Hhgws%S#%XoAfCB9^jBKQ1-g51KZaQI`<M8-y1Emru@c@SihLeFy
z9QL~MfL2Hl4)s(ka-~0{W1Hxd^t2*)^}B+K>JEogN%y3K$`)9Tzp(se3#_=kpRVJt
z$9-R|<F8s5NFLF7D+2qm3Q74;PFI;AoC$j+X|<5h;(lH)AxPnzxmYhNsy|&XtMNBh
zhML=d(HFtP+EUB+8f~fPx2`Sq{2tI2$6v15tF@)1x?_F&n0$kFKzdROZJ?RO19%Lf
zO-1m5L_}8LisFU>m#t6ivC@YS^|%j+Cc5gHqF!~?RqHW8Q;SZb`jG7ahu#Rf_R5;}
z+^q$i^bRRHSr_*hnUM8CBb<wTCi3nO{Z27|pq9df)@CzE`gUF_eU3Y{_W6*FqVvN0
zR@U=Jrs$CzpljwgztIItw2fcf=&qGxq{50#Z#UU!pYcs4iyZ%|rO&LQ4=8)|-$wH{
z(<inG@x!Q^hlTOurlKByrw~=FLE>5pzDIk)EN%ZO&pv*$S)SncwyaXLDEbhVR4P?}
z@#JMTmR}e@R>bVvAfClEs#VG65~OK4D_g7hEnF^{nie*_jUH{>89vH%a`pSTj;_99
z+DBX`*{c-zs9ix&Hf31xC>%HDz<$et^qzL-$Kh!JG$QySH`n*PPjqNfA(rxWX`$&k
zh}ko9t4ANc6|U2Er$4Qu_!U)C;6rYIf?UP9FY7>}K<BRivJN<9=EEbUt%phM<BXJ9
zeNwS?P{NF7(lS?9<5a0aF3OB{3}d@&(J_$Q;TJd>m@AOP9Ur~V`<<6udzy6F4vC`k
z7C(@Lu@E%wa(65ZeYDdoW&3Nl&%(W~L=;0Qcind&iP?+E=FeQ6K=>q`ceG0vUeI_q
zS)Th+?0txx=C*Nt)$jiR<NjKgK_w6r0Wg=54*?XH9T7Dtm#uFLoPPuA87^<Xe*W~r
z(-eTtP@O+MU;cRMCxI53L^u5B`-^{o5}34N0H>JNdie8y!*4JDDkL&#&aD4M|9@;~
zTE>Ycy+~){Y5wvnAO1@Zf1<|H+2OCy^t*4)nNx1~|L||3&ui2z5b9iOr5e=1F{1?$
zcTD<65p|Cd5_*czSATRyLRFc!SIs@f=<ka7L_vIl0?>o@?i&gJ%S70!IW<rz=@=)7
zj!$a?E~Oa#J_#;miIOgGnFJTXMt>Y{5g0@dPP+DhL545@YsMA<N?oukD0)FHVt<Q5
z>V;L@ccOb3g$;BO8AJg;0bMi<r77t0N*Hy-3@`p(5lEM?_kS4kj9tL@n%jwtx=7;e
zMxGJg`_Mc*>kTxeWZ>W=G^J)}Y(~=ysGQ(1QElH)#ufpkB;f!VOW|50fTq&?IlOJJ
z3)ICmw&6G5-A*1H#%hBHLNFo4)99A*KACPAFZoGf<g;alh$>0Mk*Q+dcriyJsb6Gx
zWF}T-Y!sw*)_-SuK>${L4cE~V^V?4GI1I}M+?<F2o}{8&G6JL+Jeg3rC3oTt-hIn;
zNDZy;@ETe|ARYaOP$@+6Kxf|~#*j(w_1cHwSch5J7>Aq!vvPuIlqIxG;O&DKE$G%Y
zLxe8Cs1u9?t+bW9_RnA+q{pYe0jY_x$~ll)iq<U_$$#~Dc*%W_I*q#0{KnCQD;({E
ze?o$4TZlhW2u@IgoF2Xl{xM#2A_lkp?UC<f&0z>PSfgX0a}H|^XGW;US%Yfo*`LE2
za{_A)L%6{j6{DteSfl(|_t{nmws4Uxm-JN5Wm|GDv^!xvJpBz=Q85MX=~33R>6Y=z
z5D1+TWKSt+m_<8{vg9QxpqGn}<*d6(I`vMr9-YnxO2Wr4C*?0nXlAV|%CeZd6t&@X
z&X;qHZ1X6Pd<Dx}Umj~nbogY@rR!TfhqQ+$_1TN_VV4tg3tN8{s+}+$pUwtM)-WSC
z%&uV0X0kIwdz8f1$a0<9Ftr{%<=Qpx2v3XG6d?Pa=+s|LJC=yd8Zjx5lL?}TE*Ho<
znx!XB?=a>6(XT&$AG$j~C;4)GUcLwog##kV*;p6DT+dQXxqnegg$WbrspK-n_7E4x
zCpBRKQ_7)mQjdR)GxX&0ztBr_=jWC26Lr7|3RzVi=%G+lG6+sj@x5C<$*m336a;mx
z=Zirx$({><f_LY&Wr**ja?yla(lQGrmQZz}rK0Gsd9JV<NmGUNFCfs|sS(hcfu8ij
zRW?Q;N*U`jrBcn+EJidZ2F^_A-AjFSnyJMx@8_Bjj{JX>?fU?j9}&fmzL?pM=$iV~
zE(<!D(a>*mXGQ7-Q6y{BQx=6YyR=d$A6l;fwwVM+QJYpL=0xo+cp#tHe$*V9DP}&(
z6-jnkI+&orXSpIB1#W%YAjA21T$T|G12ZGbWv2HH>Gf}-w=f-r1<mwlUE(HreHgQ1
zH(N>zn1O$HK(2561cfR_pryQ^P+;hN(0%JCC=?iE+DzeiOrgN~@(NDIH)_<hj`y$$
zB2g2db|NMOFw|~}m`_czD#G(rQE1EP*eO=RXmV$!Z#^pGyzql&Zmj>$yQRn7QFh2J
z^{@aat3<=2_P(gbU?mv>H=*)`Kqt?hlldCU@{fN8Vo8xZBDYR~zc5S<G;uTE!@ePA
zRh&+A9OtM;A%3~ms&&AtE6QZB@6qRsV#%{7|J%M%jUlQPMa~CQ$Y*kLG8&v8NmjVF
z^=NYBb*e~F2qdM9Zd{E>VO&lX-w)gJj_#66No<}O#Zh09tZzpV`ZQo$xC=$+D*<*g
zquYO)la!YDH?AZ+%RD95*HQ1xP<lUrqiTxz&Z-Nv%n{|j2Q%hrZ7-|i-g1#$FEhIK
zrMbAM4RdpOf7)CGM}`VEcrv!O%<w4(pXVB+k^W&(r7kSnnWIZB7;v*DzacnCJ5ru8
z{#0X-l#jqz;$Im5e3@g{Mni8}rNO?AqBMU~q~#CpGfbr!3dhX)4vM9k(0Un6`b{NP
zeWeWcFo~Tb^ZPHvNy|KvQ{V)n@y4xcKwerMJ)U2RKt9d9reWMz^2`><FKU##s=8b4
z*Q)NieWlXlh^=!%_$?aG-{cxOA-oM8H@Sk^#R<#{*aff43EZ&FQq+UoIF8x|H!OdU
zfp;fygS>_t_t1jg%?)h0VNc-3Ufd7kvw^$cx}i>jouX7eZmtc=zYliZz;3<-`?#Q`
zpJkE`gSNp7P3};`SG?B4B=$LKPgZI1jK!NEH_8r<p%4_&&3u<3gb$`m{$EziF>vVB
zb1SdT+nCRWLg|OU4h-D-@I%m<yJj>?eJaI=-HqX!jItrmR68VYV~R?#i*J>8%9Quw
zcJUN^^$V>wZgEOIq>53=+S1n_6n1d8oOcT%0tAwXTZWebZw?azFf^CJ5&{zeGnY{a
zEhm=}dJ8Ro&=RS6ziBglpUm{-aDF>H-<+S~oDia&-o8yh9}NA1p~HYv1ek+^>Fs^`
z<HZ;oXq{<!W1OQE={?aCk53K~K>+Olj!)4;fxpqZ(08bCjHj?n0clzKa^f#8zX+6}
z6$moKN|(;yT-Nc=w_guNR2ZMHZ-;*lBo(XSA}Qj3(u^1<2W}?p>FfL9`ImRV>H7EK
zH$~gAghiY&&@<y~LQ+fQW3=@2q@5?(k+efRT0Tof!>HB}>95BpZ$0aM^yTS&!J!p}
z4(<J|F1$1vNH#|+r@2t471f5J^7c`nUXRWcthBPz=vRwz4#uiB=SsI|04oiOTo`oc
z8((yPKO=woFY-72-69u5pl~Vj8n$cE^iD&U=MZz#(2Y_Qy`cDx_{7w7iw|Y0uouw_
znD>WSl+7HhA7wG82UvEF$&FYT7fG43?23eJBk@A92|up_+0Jdl%AHm1rd5^pNpK5|
z^F9CEV_(I$q2nq3hF6Mxqn3*jQ;g41BRFh-^)ZSD@Awu&zWRV859N95I9x_umk2s?
zkgtR22{|cQ<)o;m#o7M69cVVrk({h<DojXSfERFQD`6JvNmWGNK2qdXC8$uV^hx*b
z^k83_gPOy8+*JsKDiLcP6sqAS!B}{sxa78c<>GF*j9rf4TKpg*l`QdLiI73KmUFLv
z%Ag37)gYH%_-dBbZlU<em=|ulm88i(#{7!^W&i-rF)VrdFdwusiuTeg92>V*222z6
ztsH~f>dKJPY@>E%*R(&k>PnQ%VsGZ2YM+uS|Kxmn>GB1-P}Hy0ULxnIzB7dBEZ%CU
zZ@I}fM}ca=bMKSg3fS}vc}@{k&(jfqQ|nGS4zcDk%WOp6rt__Q$zv1-g?|tel@F#R
zMO9<QXRnn4;@iot)$?XO+O@3|r{r1{{7JdBes{fHr<2_3c&)Swucbg?;u6JL2>WhQ
z@$xba=p@if6q@c5VQtFT$|s=NB<pC>N;?~)4IET?e9@&8BX8Y0Hg+S+=NX-U9e6xy
z$VfwQ9_4C=tEt@8&YI5lT2Fb9bqXilq3*aes4==US%#n9A+R0bbf9w7WatO1RdokF
zJ@qtKY4pL+sjyO{J39S<u##dK0(X-F(x;PRZ#X6@8D6xQnmLCu1((N{VaL~nyZD1N
zn-t1bp?vTTW|!<lZ9Zk3e*86meJeMIHPv<vp*sehVfmrAAK$0p<afDiWJ7^wLO@&v
z8W@T`PnQ;)oz?ojQGNQ^{`r<*lbQ=8IE>VWz^o>EEcQd0t)E?KSO1U9Yid^NcDW^%
zce)vFxK?TJca_pSo!@FI-uN!2TZVfRv=DX^9m8Xr=qk*1trEEXBH5sS3}undiJ!o`
zn*ggs((FXi)Mx9GffW{<(z5Lw<?3xNN->Y$TjlaW1}28%e72IIThndj1QIGnzNQuB
zOSJ`$a3Ao($Go<hObj$Z>CDFD(sJ)8eI;;MxK&3CQh>Es!#ER!ey-!)D6+sjaH`<y
ztP;A3Sax|SeO&%*4P738(y!9cILq!b!I2RHq`?Sfd~~msskzkwmY?N}Mv7qgFCBj4
z=Ih%VE0A^9^`Lr!EobCPw~4iKK|3|)L7oNsycncP_CYLdc|TBfS*~n!AK7jHU}+S0
z15)j|abw7Wehk@m0C+z`zLVqfa18O}wGpZRgJxr1=z1!imOpYO)JJ=AT9YYI?8x%E
zE2o7pCx@Su(~tLDQzdL&{Qi%=e*pq>JC~u%6cht6GdGtp77bpP(0>aI0yH_d-hT@?
zfdMp^aeoV6w~rPLasrn@f*=!@1Lh4Wmm~xXp?~1aaC!UB*KaR8FM;xFtYXMj%dggN
zBJiSceuY%D4Sxq7Y-m-58)SrQ<V4xr_cP7Jya|(qU%!%X*RPlVjVwsbBzLF({{0dj
zC<#nDaT55{>(f7foc?_I7p2;6hy&aZZj9F5h5*N1QYWJY(@K^F>aakUn2yX}!}rX%
z!hfOl$3K6aNXzey_!P%&)|9&7>%vx^ff{&`g8;GNoD@b~(iVrmaa(uu=QZ;<6PXbe
zZH+(Vfq}OJf7q^NR)9;|@c8>9@vB2aQm;Yk3Yd1Pm3kZzD2e?2l2#nsAw3JLU*DlR
z)4C$uJC-K=cKUHO4>Rky#;Wc<_tTko{C`{NW)0>!^~YHlfTrpXhIwgjt3T?X#Ws*9
zU_5I?*Yf*AM?^crq+e{7ytmk!N=~TV+|tjtABKlb&_T_wVXob(F1TV!!?23}DDI*N
zBFLE#1T{g>G68y^`0+eN0ZbV`P+T|mLTF~ixHVP$>SMpym0;+zjug8FCiIbF*MFL6
zu8tJD_RQ*s#ct5I;!#``y9i>x?iahmgDnpgyYp-`0yARwl39PK*u7(1Jt}s|1;!JD
z*wKPDq8Hh)1<%9rH(V#Yi$=d`yr_N?vj2ET-{E*LZ|~W+#Ba)E;|jG!XFv2_xT+(g
z$*lcUjuvcv-ukIZJ+?wNyf&M0wSTNQwwV-o8VIB0`<ss8T{7-lD;6FwDg(M={(ff^
z9)v+m(U*L7ODpVVmz;Xw_e))F?ooV`ww;JC`RTdqFLLNo0+90}2=(3yGHgTie3wSB
zW_>IAXOW-f+c`hyWup@(LH(pRdBbB5+lNm@p)kC(OH>}$b}nH`4N;0xrGM}QvI@DZ
z9;vnP9KV%NOZmLM5(4G$43Fp!CFE-(r#ho+TU{GI)y$w(>olM4ILt~}cXSD~fUcxU
zPuJPXQ%PG2_eEPukh?b!Q!03ANdqw#l=fa6#)Hw);YW$Xx|R}$d44kv1JA?~emA?+
znz_wrE+kBN?kJTKQQc>8lYc*nh7>0qD963tYt%RSFY#kP%Y-8WEztqEf#SymMaztq
z1H}uWh+#~Ytj-j>mX@N$Ilxeg+*C)kXh8|q++r<*rAz&yHB3&Y=VfBwi$-zB#FEVI
z)k~$&Weh|<=GC%nFQI&}S{lQ^3__Su%d+K!;)GgSg?S$g6erZuS%0R`nG=eOS{An!
zT+J=DEYV0b`q`)tyR(C@q)f_NCWY?V3QR~-rfC_COqBCslBaH-<~Ssuqw3>oi)=^c
ze0a%)3#AN_s3s1nAy(q8@pNlK6e~WGw7RxM`E8!pxtSJwh^UPj#ooI9%z*CLxwOx^
zAk4Iy(z@{_R!>3PjeiXqHa%*YE8*hEQ_)wg+F-Mtp_xxC!4*)=y^p_85n7Fh*S1*8
zV*&d*=7Qn4agI`b5OXohk&GiM2GixJAAQCe5~*p^T~b8C_fF3uL)-*npmLq&Pog)L
z0tU**x)vz~<8F|`glmx@Qj>!|Ot=<_nTYmsptzBG2PIc_(0|AMsGbbYU7O*&@F}cA
z24@cbj>Kzkbkx$S&DG3DR>~-*l&kvXsz2Bp!_izqq5Krun|qX>fjbnQ4!Zy=GI^$y
z&7G6M70xUf>Mdvhq-O?Q3ZZ!iPyhwY2}MU0PcED#U`c9eU59XP7g$oMHu3lt)sh>h
zc)W}@#N+3_-G6Ai88R)=gaY-N-Xzvs+^mg%Y)!6kg*=Pbj*g-djo3(j@e`txyaLm|
z9*Sc!J`+3#8vDkH=P}tqmtZl|Pkq)95fy&R8hVWQ#E6Kj;1TyQrlpsqC`y$zZ~Lub
zBekW>&$|~(?aMT5h1cg13(13#6cn*Q=W5i?fh=T*2!9n!LKc+Dr8ibHhBifAs}mfT
zB4$*8%rcWA8ekqh4y8DeBIY!fQk+399P{{)P;8KkAt8hgr)cBKkQ+%|qs-@1$$6hF
z(c6bpG=+9hBG<#v4odlZ&<--o0cZ!MIFEKvit}g(p?FV~$0U0;!ImK&6!HLEV>5;*
zJwnD8bAP*vG5}w}5(&K>IqDP1aR~aA(FZ5Tgz9{9oP5)_C&}eKCgFc~lAKamN@X5@
zmBC<+>>ZpV=lqk*GLs|c9F<a>$&oz+E(VGlIdaZysaigfBV&{h!$5LuCUfLqUWL0O
z@ujz8Zq$Ohn*AnF34CDYO<)^n5VLn#H+vfa#(&Q>ff&PUbu>Oms2++Bt{cRIO<>Re
zvnG&Gxg2H!DdF!if&DB;m_S1D024?k-Z6nhEhkJMIW`BGz?%{uOyI_94oFD^c6Z+a
z6H4xJrQZV+3XG=xQX1{wY-cYz>?PO6DAl4ks?^d%6vgM(Sn=MWEvl=|NuzF7x~x)(
zEq@Q%+GfrUg{pJ6$wH<)CJUv%%I_FuQ;e*0J(1$=E2xh9S}3+YYaJQh6K+ZknjNXF
zV3M^5Q%4qWp@YLrtEjcX%bU(M<W3mLywzP%d8ywq@H)y_+oc#TqH;zw^<A}U%(Soc
za-X6DhI*z2pRTit7SC)cR0*Yi**}KeiGM54<=!Y&h#0pmJ*)2MJQp-*scGkpJ7U(5
zj?|}<zp1Y{o0pYc^;M$V*S{t7JUW!&fGQ!FQikpf@v2+LTS<?Ro*PDg=F#QSK&f}L
zf+DEUt*T(EMm(}Bs~V;S<D$Fau_~%gO?QK>z$LRBhB|#|nR4o<psg4UGA+y`+J8!^
z99JMVQ2v$zF*3^m1!9!qe1RCHXr`gy19{9CAEkJF;h4a{TN^tErBWit6^>EL-vgzR
zSq^|wDaCmxl~SBg*ZfxCKpvfBU<W-oQeCs6amNgI&uVKqr-aG@$bZ>(Liu2IEn@Cx
znNin_s#A&+s*vFfN^t`D&$|ZR#(#IFW1Zur13Jul6<QxC))6&SY^%`LUx1-mdte^1
zW1g*7008sSawgQo55kh~i4p$ZR#n1kiC$#3h=Pye?u+mu=5$5obtkm0HoC;q`n<e%
zk2zHqPjlVvk{|-;Fss1L*b1)u=I4q4ivTe-wh|6P`NDCq+)%3g+(*mVGJk_n%vy2#
zLUdg7>S*+x4u<R06eHBSlehVf8p3=5k#MxrY?m8-bX|I$KjRpF2L48~y~$g?>o-t%
z;?|&M_@R`DDJC}pN#}%|@1Hu~-J{f8y;FE6!Pd7O+qP}nww+9D+mnB6+qP|I;!JGY
znApjiz4!B-d<T8i)!o(As}H)cR;~M2J5%I>jaC{phua1Wzp##Q|4h`%4Flun*9JQ!
zOMTgwsk^&1>fR;PX0!46XA?jS97yWHOWKWi?auH|Z{vz`_9*%i#k%R#IovoITcbIa
zK{GAjn86STvT&065*+%L&<+oZ;p71kEbe<~*K-x8-D11bA)uzlXmEK~!rkomKbSi@
z9U$K3$S31U*#vEC`UX9$HT@AZ(TqJH0gz^-V>4<zy^OY)J6o6E=<9$E41xJRN93*V
zqjGxiTrkglh6WUp9dh^Q8x&cajMQAH4~3(;W&%+~NeFc%EdPn(ZVs&k$DYlF<O}!j
zInq?`T98juj}2aiN&Pt1;ZGYFC0r~I_|-pB@{vczx&{}WIu1OWF}0iy3#ImImFwbO
zDwm6uI%_(QFUv3i=2%)qm43j__7Sk7j&`m~ZZ?xw23I=VRBosj7&t?Sj=@x88`)TJ
zX-v&z*?N&^SD#(8>g?w+$43}X>sk2^hw~b=bvpwLTs}4S*9af!Ifpy{nU#Xd+CruQ
z1H)mY8P`#wrrC@^Qh~JSq-`O>(E~EH5(SkPf5L{QWj9GVy?l7vgaUnD*@@%8iCNoy
z@<QmTUx6iqSfyZwg<)r(qYuwTEe!K6`H(e_xw&HB7H{y1d3>hUzz-1r9;k_@u69nj
zRrs6TAG=EI#GxLzD!ppid-<L;Qgfx~O|0A1AcDcZ&hnlG32X@ilQ}8a$N=b*ypv&M
zjON#<6@+BjO2;a1AsrNd*(;Xl=ui3oEN4+WT|0igl5Fem5Poe$1TZMs<$itMGHK>R
z9yDcT$3YOu3HB-M5PT)qUucgDH5r@08L`qDT2(qm0>kl)7FEv=N?W@&9X`x5_C6vL
z3DW`m{PM)c4QPKEGy8YxXJhl%y~fMI>9;=BwLM=*d5H~$tD*9f3U5E85hKHq0#L#;
zSxRjqVKc+IvB`8x!?i5oi&n%;1G|g^hrt`7QT2%4i~=u!Sq8vKVt-BuMxhyy?zsjY
zAI88|L`?%Ph*<`Rv;2iY6MnA84S`|$hdBjaz-L0&hK+$~ie3J<66_xq1xJ4Dj9s*o
zkZiWqpIYGu7<!2DKa~JuKPP6{yWT7;1CK}<yWZh)jsaL<&;&AUb4TFJ@Thtu2&j6<
z^*<#+D0D5A31z!j^bnY4a>LKnpMzwdYrutx86##3`Nz5+3;)~de;PIV|23*L0-ZZc
zrUW+tZirplmgzrb&n`sK_9L?EqKjCs>H~{4lF<Xs7mccGH#T)~Di2fnXIb=eY8lkj
z8uky{v|T{>!68QYsFVA}S;ARO*w_a~xL5Y>K!#d2#gE~<1CK!eXS^pCWqWS4oFl&&
zMZ1tM0cE=oU#=kdIi)J_fN3AuWJ4Y{Hm?WLv0&?S7j=Q3uwWnhdlhTl?inY&whY~s
zFYV8KeK?6#Uk~Jni-C}lB|>io!(r)5{PjH}`_$}@?fnF5I3sBI6`2x=j3iUOO>P<p
zDjXAl+=4HTd=i@kbp7Rldfj`$P@seHKFjB_Z1;1sFKy~%FIOf$x4pJv0d)hq^LK9F
zFYDuX;MU)o4xm&l9q^E{EG38Rs`Yx3`@O5vM0&k2OoWqdJntisLun})y?&uNpg~TY
zAc2XP;PxVqmf0RCZYNj{yYh&T@p#3LX0mvIwhoech9I_@OX%Q)7&#y7cZEtFgfd#P
z$M2I6?~=J%KetOXI6DCx0o#~#GG=?Jj;_+iMIzaT<nG6exaj#lr>ur)w3E4FC;2i<
z+iluT1TSTao*B>+6KIIY0so+ul)AwrnahID+%gv*7fwdVHoCB<)s36!c04$K6?Y>*
zwqi|G!K?&_->+b5%4xSEw<@7+FVuO3{4JcGon)^Y%U+M7atKy7bEDpg(pT=W1_cuB
z5Tcr_4gV!KmZG~xLwohmXep!~r)3+G=XOl#9)#Prv6I@w&ujNQ#*30peXJ{QJcG<C
z4hFss>y*uP@4w3jy|m5(6rf%n;s}KRY5|0u`1|oBA03dm&hU>3p!Ki!c?jjS_kt=S
zPEuXSxAkpInC%j?ys5$HSByj34o@#fK9;Q8+RMK=!_Ce_ZpJKKxg|X|35o#TO>TTN
zt%A?`GoVPwdn+D<Ei^{ZuFpop9e3zM*)+6`w>0C+2k$N{ek{zfG(KI#W^o(9llhZY
z%=f393w72mN1NM;8SIul9dZ94-JcN7C$*UX7^etqt(Bp94(D+kz?QquiMxr`9RplP
zR2@);Z}7$~dc??S81#imJ{H-f{J~b64){KKp}WrHS65CYLdz&qswR*4-d2o~rH0vB
z^2Y@Ila%blauEeGoc^Nr&LuP;nYWN~Y+Hy1<xFL|;-nkx?W&{p;*I<hmx{NBUSR(%
zek~v?gE}BJ9^1QJHn7_G(Pgsv25ZwBHF}cm>FB$gM0s&z;x3UGSC<Ka;*=~>KhQqx
zgBY4;E=nX&fh3mll&sJ%aN5g3X4(Xr2f8pVKWU6GDIc|mitLnZ-!l-PKvHSZj&~(k
zL;K7UfS)Yng<l3zrA9=n8i2pvnubhMzknJJszxO=V7X095XLIp<kMtzv%D!j+T6|!
zh;%#F*`{7nQ{N=(9%Pv8S{t;VH){X%{zT^;$<&$LM0iqov`bOnPk&2p?2@^rILgp_
ztRJMu0$R2Cm`89%S91i&+}0~lFKatzksWC4d?mxxKvy0}XSLKFpiI2nTgh-8fp8PQ
z(=F6qMU^PV1~kcegu2nJ)XGY46m^D(C>uChxOPdS4;3Ab(5Pqo5oSVHyw-&Mb`cQC
z*0WWh#TnA1L+I)`jh{39MF!&@*>|#9_>Lb?pQoDg2`_nG*oF)IoJNNUhh-f*j)33o
zc@f>FU#LH(d;bX@zQz@If)aiXDPLFz`)+X(NliVui~|hnk{4*17HD4DCNAUOtQhbD
zJZ|-G;3Zywh#MY(I-h|sF?)LBRiP?LXXlfN1JQ-q!CQPgLj1C??gsH26geZU@Bd5C
zF-oUEO<ThJ>4N-k{9yF^VL-?XN19k|D9_m16n;-WEOPWLE?6=C{*3%IvV+X}TPlD=
z9Th*<7Km>7C~Ea+)Us7~O;Dps08+qd{avWU)BGV$kC>4x3D#Qv1o_xKw_e{|!&cVP
zGT(hFOvY{#B}}LEC2G68ZPW-zZcXyLyy|pXYrPkmj|Mz6|B8#r8jz7SX)xV+PqAp1
zZG3G6=$sJ;yS&yLeCm%^cQRF{3E2XX)Y0k0&z)Zwxjo<dS6>5a1dfa}rLY`;=BQDj
z*}~*VIJU!GStu!8C?jIanYWtaNggsn*&<MGvRYUfhUw9yjon4Uj2qLAS<$b1xCOQN
z*)$yHn*a-A#Lg;7T;Fv?j$bq}CWe<##RwM2woEYCg=UW@UEb0{3Oxh|&{UpmtaH+Z
z`55MbVQ|5_l6=Xcmz2L;qeyP0Fv{p3`c-_{?M(W4@J{$U(@%r1p94?9(Xaw$$?|3l
zn%^jL6pd|7lmNU7qF?wmew8zl8P4NQNNnTGK>+saOp_Y8KS+>mNwAH`%~1{>%r54c
z{gK`08rx>YE`6E?0)fP|)1dWZdq-t%I3|gIGu_r9=dIlv1F4VL9;-r42@Asf07RGX
zOfe|bYe~zpc|59Gl+cL*^s$#b%+Uzh$KUeN{>_*t1YDhRa_(?!Ja!0UN#~U83f8<8
z8vyAo`N}M{%uz?LYV^>7K$kBGIMATBWb9OA<%CJ(8k)B&xzH$+8m45#c-TD&IR1d=
z?4NsfHsR&r_lF@<Chc6yFrpIipJyCUYr_2>N4%lvk!%|z#G0U!tjG-TEh6!3@uRrJ
zWVGU9wKWVt;lsgnj5|+*BV34k`&zVcfMG!cM&9j#S!rMl6Z2IJ)u<@F)4^U%{l$AM
zhS-+D*@u!5J1m(wOne15IlB^#q_{|~&;MEUr}w2DUX&EH#E}3y6}ovinilZe7}+^}
z+|lXhQ#u(1?1z*ftXY>9j^gagOSHKy6kA8U;4nvZOe_a@{bsDr{xTfnD<UQ7cIr?5
z;vve-6Fbpsv~(1>;b~2|pL1{5-Yt%6gWjr$81)Hs)jz)+eE9<(W?u;5<>E`o{O~%5
zioYWiJg{Pq6|=B*@pv^MI2u}E_un5ZEf*0EK8+UxUhRJ#UJkalSzI{7H(<85AGhxu
z2ry?_06#El+L<^dWg0&=d?E00+b%Y|3<!Wkg*ouThNKS7*m&3hVqv^Sf8a3wo5n2k
z={^fh^KZXwYPn)y4W8Y3)?L=d;sHOhhQ#Y=7iwD~mfb<Mf1ZFS%S5W(eCQGbC%2Y~
z#fRF00x;v2)Z*zARWs~w)b#yxWuer?Us3)>57yEfh(S%Jh-E_j0S?j^<0%q0`TYPf
zN3}|+FR$u9>`mS>^w?^L_`~>(*vLKXl$m0j!K21p<8Zc|-p@>7FfD@seNtR*`G%op
zb6R0!Q50kwmG6g?01kOAt4$iv8f|MV%f<*>VVj*IxAA1VE9JuvDm^{zvjT--nC1fI
zi<VXR44;qczAc;%g3~JLmI8+s6hDAUN{o5{$H6j_r^ED|Is^X%kH%@y)V`eoe14F*
z3a<_*)MQ}ej1ACq^<)Aa)@U+o$DN&e4p61&ORRXg4on!NB_S4(KuK%5AV+_fCbT{@
zq~jDDq5~rq)o@%Gd(2^`73nEHv*FUac}#KqT+s<NS%`^BdhGs&U!)3?oe!W0@%}jC
zGx{>fu&b#g(5~Pxqg6TA>SnrGJ|NGnb*{d|$HAs|i(^DcxIi=bilTB^lzD%}T9xT0
zpp4kmQ%<-7Y9gbe=Z>DlaM#63TI6a%-D0UBk1VP2^^hXSnBgG-i}mpxSZ`T_)yz7k
zj%M2Y?3ihuJt@7EGOcCoiV^^abK-_r?+ZLh!S7vzR^OgG4V_5WHjx*1^fcCoHc>uq
z`sDgaqJMItc;~i$bU7kvMg;sR<-4fU&ylx#Fd@84r+_nBt(<(0n*AqNv`2D%BMF=Y
zV`*QgaI9P$D%W2pdEaTV5G}6=LmR%=b}Elppvwa>SJ+`!PHWa_Y8a5|RNx&E&v;VO
zK<C8YaYxr>yj_|c#?xakRlUvRsSI;AF5$c1Kq?kw5s%HgF~!*yIk{*y=Wl<P=qi6t
zkokIGdfuq5fZkTCg{?LbncX)gKsXIVi~#O`AzVHf`krSf+SbDzdyqwZ1>*biD~{nw
zG|U-e;++s0zaQcBJPA+`r1T3K-=V7@$l3{u2WkK{KS+fS0Y7=)<&>!EB;Ta#wZX`<
z7B|CiZuO;{bhu6RHi)JhUl1w;^<NOqIk(?z`Ir4m14AQ>pHf(E03i&?WXNo#077%G
z11Skz?EB?mK<-R?0#YIF_g|Bk_W~`_gN0(D0*O~JP*|m=Zd?HB;*`S}47&>J-Y7gb
zN2U!&Nqjy87;Fg(o&wfUhLNO~+aOhvxQNjx1eydhI$sIdqDwouuZJ8%joG04GP-+i
zUC?_7e^A{$otP4l?+~9scAZM^5Sm1+s1!A=yQG--=Mb{;oS(XwTC?C~2Ftb{CbF5U
z7s!r^YIS6|iYj2%pNh427^Op&QG}UkU;D9sp<h|4a`Ms9hk~l6_lB?zZAH#3JZ|pw
zeSzDN@-a3;#2wM-cYfx`085GIPtgEbVwo4q{{z8dorR!wEbbnV7GNrx`G-7Rpy;Yn
z;jZYSQWLJpE$uWJ?c|~$V>fe~C@ei3eF5SFDpaoJ3ZPP?#ye$JqZ4Vr6t&!zXqaW%
zBKSt&4?WJ+-gtNJ9iMF=#Po9?%29etnMd1~UH8`M5P)?eGk$JnVM*2KCMZbQJEwrO
z#EHr(;CIrBWtY$U_L#Bo1&GYa$IkO{J1LuDWi~~<?X8ImdXTt$bq8p36}7fSAR|-~
zFThXb13*Ji&V1OIRjPtZM>DZxj=CvIi#E-Q*?rt9=7{xT@f`a@!&H2Eg_t?Te>`A*
z$Qt$i_`iNz^D1X%f3I<RsvU~$?7lAC`>G<W@<0D2)=k4N-A9{kYh5-nV@3QG$l`)V
z(V?QyWO9F4gE!#2;o%OFr8p><$`E$E-a$;R1|WpTxEnZK^})eiH!ZvUx6w>}wv=mI
zAYYmU4}Ys)!STHtB#H8b(J59nT@E=RLZ=EHHqUhYdeC{N1Xfy{pFB<FH30$qv)*sm
z*ESbJRtW}r^@zBeccI?HbE$NHkbM-3Of6A8_o?J8C>v0D*`QP8kLRD#)VA*kUDt#Z
zw$--Xs)Y=rC4hke_BR{pNV7xBz1+IG@6FbN3X=TJ4QK(@{uMimi=v%166?hvN+rls
z9s44$v9fLPE#1g6Tl}=@o>;xhmg2x8pUDTd#vD2oH0*|v#8SWO>%a^aXJn)~ueb{O
z4ZI(Jy#9Y%IXKgt(izawexTTtX&WT)IUquUY2Kvp8-UR;L*m!@WLZrH#iKht2F4?l
z$&z*GJtJ*SCb8*g48mYyk^>r1E(s}kql*4@v9E5_!J$HXAu7BkJwK#i6_#sfU#)=C
zZtw)0!8O}40XW3T&H!R9b7Vm%(MVAu@sWRgU<WLV3Y?46%#GfZp<!lQ{j-t^RRTY^
zNexX$0FKfAxRr;i{))7@PCtP>v!2)j=gf`^?NMB&!))mO=sl})UGMNHom*d%rtOeD
zryskTT`F&hv1l%+@~D}Q8qF%RCdxt`Gjp&AelRAYtQLfXK?MzRTsh2mq>eb&n_OS*
z#AELNd$Us5<$(#&VyLwC{F&2Yi>?j_Kx1u$^*ZQZdATkWNX{O8$^2ikC-pFyg`A6y
zO&zz@v;y8pS8?q)=MIc`HJJl1h6JT|8w1CzFfFz-HNSUf>Zv9#u~Cw!l_H~JJ~!)Y
z<&KhaUrl$0f!RaeUa6)fxJN2q;*(KlBMFUb;o^m|>z(%alm2A1D3M%{kj08vz)krz
zGtN3)h**xs-kK#JR<4t8#1x7Aj9{Oi+}79+^3p&!*|Lr&QY<>Q7J!_We`p}PB)Hec
z_)qwVF%npjRN3QEOv#ExdQFs?jMcH8JWztw?BTslY*xy9qIY=OfubRal^zS`nVBDX
zF_L4*^mlX)qJwoit2_RQgQ)!xU{xcczJZuiWwjzuOr5AZYe(IkgyRRpbm*ZB#M+A1
zSMZ3fn<5~U*daL8qX?j8zS{B)j$P7Avg7tV!*kKX>!wENbvqe=&pLE8f$HC7G=_q@
z?s7Wyr1=PLqRDCPw{brU=#0I1TdSZc!%enDnwHU(x43Fg=Rf?Xa71kdX!<Wy>3zfq
zo%MKpC|YqC(S2JDONB)xCzhsv<F#cBFC4rN`d(G+nEcI3?j>W%KWoSWCZ6XnQ`Bx0
z)+4!RY)G$Er9gX<YzJ=hax*;lD=z5O0?~|dD&=c7f(_VEbUm#WBx+~rng$biedSua
zY}4FRCV$+bi(<;QaWY`+W}f`-NHgZFdxmn#QWL+SjMKz@iKu+(&SmQf{a+L(vg*IA
z<kj<Ku{tSJUNRgrOmRf#VM)Lyu54bg9NDlT*!nt@jVdYyk)TCpJp#3!4O~kn)Y5T5
zUsc7tP0jb!&4?2vw&uMEO*lGx+R+@!-3sPY1|coAF6i#;b+LdmLQ@>0WiM$Dg*;S?
zx%Ud=4YCW{5(?G1grs!O(K#^d#W*e)@dNXAR(tzf)9qL{?Y#nTNn@JBT$f^P+ulUD
z{KG6QC-sXuZ&xPWNB%%xsT(~@^$QKhebbD=iZz>IFWGHe7_GP$60(1Lg%1o;H!@;S
zc={7f!=+hEGc^E8CB>Q?1Fy!$iLsIjzL|>kp6Mbn=s?NwZyi1p=EIwHi7#)+yH1+F
zcJWr_?fNq?iQ=Q7JVMM^uI@lZ{E@e^Vlxo(u0IPQSq`U)CuryDs*k|uCB74~a#}&d
zBHws(G{xMN)zy&G7YygOeTl=b((1=ehRA)(3gx?L6<C1gwu<KD8M-XJncE$?^k`LJ
z*ZM}F+ijWrGG!(u=5k-JE+{V{#^-k6#;>W=Tx0D4Yw(nEOtT*n7fu#<>$v7Sm7%In
zq4x>w=sI*@qzG#tWu1+ol_>A7HE#gx^`zes!1EG<jnu#xeD9?tHR!NEdy1wb)o@f?
z6IP-k!U=F%5mA<5{E{pp5#GACy`S;H<U-Nb4qC?->X(3bFVJhUP*g-^+<71feY>L4
zM|hKJ9_Xj|L!0VwtZ-pw2h5`^a$)EW{8Lvr9hGTK>3~*?eyPOQ8C!T5TC%=3=LM$m
zo+`WujiupyJ+93ns<V|pN^!Vp4KPXE*OaWdXa?Y)?YN;WWPS8{q><eomo?H9Q^i!+
zbInRdhi)nBUlEfO6}Dy3VcO32zec8P_2+_?)?8`IUAJ4zP}3kC1qvcx&7k?nS?S0;
zyp+J7B0K)`8_MW3imFgT9Zot1Aq$0N?7=B;_6nj8oR=%AwMz(mz#E))`bD^y%sD-k
z&;igdP&O`g6wb=y$So`%G5cQB!zfehg6`BJ7gp!aJf8GDk}dYTOhu_kJ|Y(G&tBIE
z>6j32kv$VgU1X}+hec)wU{T#~s<!MfdNnI(8GF%Gf|Xj1XuTynlHu2Q%EowAXL6iY
z2C{xtp$^gAV-dDlLbTtl4TzkK?Yh0bm;;EvS&TLk1xq8iUvZ?hR1EOls8tl`BcJe<
z(roo*^JPgdtwDZ&d?Ev71jkEm6Dl`zvPl3VIzP=t`>vs1eQ=hQ*ic0Ua`Mx=wH!d8
zd0AcP>QpIQk^dGx+lu!UxID~Sn)hImG`4{QrSB<oB1}W=J>1ReaIS@OND=k_#|wDV
zH-=)8^IxY71qxlKAI|To<&aX#jel3n$=gASt1g+{!)Up-dc|3B-Ab$EVV8df80E^3
zs%idGt@idUWnMdaKC+$tY$<x<wDf>s*|IsNgACEbcjM=-+4WN`<J2tAA;o#(v%OYt
zt-Wm(fDCUzkgW2(m5aRq)JmtH!K8sWAfWswDWQQG6@-~BZ7z<D67W}B*5QETKf-{g
zh{2P4Q~!YJ;K&qITphNKTUS*#wq%0}ea?1n;z_I`n$afRX9=(-79I6JxDdq=cOdyc
znteRietSybWZ%I}!!hna#AE7w1*VEm27xMJ7<qc(_xdibYhfd@NZE*CE9CY}b@z$>
z69K-1H3XbMk$g5q1A@B45*V(knj=6bneL7SdItV6t6E`eP}ShRPT)XMI=<I2=DG7^
z%}y&acJ=*u%UDRKk-ClDL5Zo?_&=UH-~Zjyw(s56Z85rSL{dgKO}qs3)7brX6I<ma
zWsQ&_t8Bs=?dG)lD+&6`EfmfPm+$&!Y_=U!+nmGApi~k^6F`NS+GBeHQDt~wlD%#)
z{_+Hdtc{#$>FFyB1=E8~MTN-q@nZ6V@yT?eV`jBhSjc&>@T0fabjR((V0oL}n(7rw
zJiD)eGn-&MGwg=ec*e(iIn)%Eq!+zPEIIGY9o;pf<fJUvybp8AEjh}mP&QWkA+2Ig
zo2U7JsR-((4**)HD{wYd!}1-ybsD*qx=J>)tzgzIuSNw91S|i_s*yWNO{y=iZg`{_
zQzy@wVFT6(jXI&dQ2W4MQMjP=9>;g+u#R9EByu88s|yxG{p{{*;u%RL8;w(o(18$$
zxeK;#zV@xdX{3Srz~#s2w#G`{OHD7N3A-w+`9tC43AoV0A01|j2$<wv!4@lnzjNMP
zB^=swPOn$i2)Bnz;$D$8YW$2Q17KZN>jXmnc>mniDQc{@+fnFa01m;aBfXrn^Q#^~
z`Fe>E0nVMr8;S7aHD*JZGRsjoizqJxi%M8d_wi;@2w~a?ap^HH$;t7Pse!xUtO9C0
z3_qKP0i!7lZ8KJrb;=rh*&5gPS5Q~e!7rk{g3qWGw;s>|xEJ7}#lHRD_4-WuKnhP2
z-UUa_UET!;9}JHoq~H%C?u0f0p*#?snR}dS7|S&E$M9=eT*viW&cz`_H|p`t>pD&R
zQh}1-46e?Y)?_!u5~aFc%=IwAoDYTh2EvXO0Pi_Vl|S{T!&eoXTx#n(JSm)dZDn_>
zzc92TG9~cgVt7?|A51I>+%|G;T?%bn!L_8K=z2)n7VXj*%*N&#)tU$xM0iHZgL@bL
zy;eE{nV`*YjWZG<n4lSN<eZ|1H%SECwn=PeUwq>_x%}D@+06b?oZw1BEQ#Ko>wVEf
z0zfOGCuE4;#cjn-oI1^w@sH9XVUzN7PBX%WCx^!t2;Pe71{-)@KxGFKlxc4Bun2=U
zTdZRVS_z-ZcWc8FWa-uK{jzohrTc_f{*f!4@zeKMsIdNMPaL=^W%q=EE?r(qXRNEb
zu`t*~k@xt#$l|V3`86v0u(*7FNRva44)809SYeU_G-Q1?sz#||-tt}GQg{Oiq-zug
zr{AGb2#=eO^Y`INs6OvYeo4nobtFsazK3dwX4KjNWdZf9a}=i<7GO<F1^a_+W@;tY
za?SlZd1Ve`)q%baQ+>*b&@rq7G6Zk=-3Fz-^WMoa;0QAk_Teiye7l;koZ@Z$4(Liw
z?@YRQnn$zJjW@up_1>uy3K3&#{pv9csP2A{0mrDC-PRvHQK(KZemM77)}XY{>G&15
ziO!ylNfS8#%C`7cuxMOh3Sz{)Y`l=t(j4Dayi8bTQ&-YN+ap(H@OR=&X7(2C2*tLx
zz?i`8OD--&(I;orImOSD)Y5;736P<v4<-u(?`EJL&0;#iPX!m7u|#R5pR1@xPcPo)
zl5mt(=l$|a-w$Q~g?%?q>`iC*bZ=&FWxTH|70ZXac1%lQER21`n~4T$eC%l~kXdRH
zyS0f+k(JOnQrny+f!J}<jXn7f;wu+e5mqdETDA|A6T_q92jCD(#Rb_BOLJ~OaRHZz
zcb_4H91TC{c&_wETe^*b-DXQ0g*`e**C$GqAeA7b10&QwF*&-liQvp`y(!zh!Rtc5
zX`U0PH*O{S&pM~C1TJ64mIWs6ig>B{7bXaXciLQ)Fe(@;6DLa&b0TdU(GQLe_kY-e
zb$R#^Ky>ccSzDcfzNZTCDa_L2=fzPUV-HHEta%Q$-}mc2p%k#37GAV5ko(KU%--mv
zAZ&<S2I~c2?%6_gp5<EFT7E_7Ze{1?-_HmT0s(6g_8n3n_~|&E?K}hs1t9;{53&ys
zI?<tB@<)iIwhwPN8T=8P3F7(<dU(eyX!fTLFkd?$2R+*E>Os{GoK3RW@f*c)0)_lH
zucI?&XdAGH)&Tr&v8m8Q9%we39L&8?RML9@Z}l#OqX?7ZKeF9m^Z8YfY~Mqb9`yDh
z7<-_wF{Ph8b<xhiu9f}6U;X$Rj-XuYY;)n1z>tX*;+T}*96`L*I}rA+;<R-O;>WfW
zfG(nSeH7iqBd9>K=5w4iJoSIEj)a=Q)Ee%NawN_@Xs*=j-zC{k-$*JYNT(uA;swf)
z!)W3PkzcgmS;oxbbIHX<&T`N2;!5+QiZXx7zTmxbCFaR}?{M$%Yz=7{)S8Vfk+j7K
zAtz)D3tfCM$r{fhBTmxEKK>O&6S^rR0+OG*edj5>Y{{!4zDua_Y!`cya#_8Js`H3{
zP_R|*cwv1XV2lE<U)rU3<V>)gE3=Y~0a4ni<BIXVu-+}VR)yAM0^_GVm1xvE^mQpy
zz|)6+)ALYS2(7NoMrMk!VeWrmuxpipv6K(W-HTqqfU+#wj7xIQnv5POS#Mruz)F{}
z&EIu0%_%I)9SEm^^0?+GEm@|i{9gm!=HsX7t*W8-7;cEKkR1|i>B<dO=n><YL1q_x
zk@aH-O>RQ1$9I1wuRjD6;<xmN$arUHxu(b9tJ1lI^n*-%eG>OIdXwxYj#{$J>g>vj
zDi!8BrPdt<Yvu#9(`Fy$&9h5#0DCwiM*u(biHWX1m&{5W%qvrkqw3!C`Wm~FE`)R?
zwx#prV6OTLrbp8^(WeR`Js4NL#BY4;`LWo-owP3uwr6g4fm9nQS~a{`+~B`0Jj!^L
zpCEh)=sBvBqL;N~H7u*xoaOXM)^vsW?<VfX)qiarVwWMuq+Irwb@#2D1Bw*dZnc^t
zc3mQ+7g0k~7A6tcn)<RyOmO9KJ>Tuy0SpUSJKGg{&i%Qe*%b0-O`qwlf0-gC5fpVg
zBNH3*W#-TRa5_|M@8PO;&>6Y%7kE{19xJFmL<w<y=~&$=pLW`QsGsK3{Gx<@A@hOq
z8QfxOwQLk1)rRq`!Sg7X2hdI5@{Eq`#iG_K*S&o<X6vOYm;I%8WE_yFkzLD7f_?1n
z$(qMg*^{f8>QtUm@yxthpWbqHw^!HV(6~pYH((zS87yht!*WE^#ZS6oJD`PG!f2SW
zkIYUvVy`M(^W8rscdcSlZ3~b$16N+2>XH)78bBjT;jGnhz;>%W1au^KG1VaL(z+`j
zn+`Kz&<kTvOc9`TXx5U*_36ja-XEr#`KRm;_Z-QUTn^rpp}3?DpR4=GIyEg2+Z0V&
zJUcX?&;{IaoxI5~{g9Bz)B|}d$%PL$!w$Zp{6bc{JMQ_9++!KbZjyCrr0LL63;X+0
z`^ieHhJgyC9ZxIV00TC`mb#bn8Okiq3V(CQ-i2N`I>^bH$bPcBl7QbR{J-fR<{vQD
zllgT^v*(Pd+0m1ctf+XoG<q?EAc39ysuH<2f<bj7>Si;E2W?-A)q)G|^ix^SY@kIV
z!e1`0pJ$emp`mcgj-pW&V-lH56<vB%fG#xLvXJzakvsC20L%w`;76)pi5Prp6?1`o
zIbkv5*sBbBco&pr4ks}XUVqYi`s{pbrl31(B_Y39xDt{2+ve3BC@0&62m@C>>PC@z
zRuJu{Q`Pq4heV>}pxJGH0keYC@>W;1{AsTbL6f_0gmV!fZL+RN9VmWAf2GB8bt$Wq
zFdbb~mgr5d00ce6bokY$M+kE=Xm<ZWxW!XybnuDwa`(Q?Hdctt@Kqno_-)j#p&t@=
zwVefwIo8Z}pCYcA?H(wQ_`F_4t~L_v&rwVh4jdWhaJ>j+D$PWaKq+M_K|6_hiY0HX
zXs5mn3SkBgdf<)F5u?p<0?IL~{Gr9JdYC8;{)s!W0oZ-Fk3pJcTpXniHpB|dRQBan
zNT8DDhuoqY3QA@^n1UkPL5ztnAa3Gk?<$x1y9u&SyH1sV<!wH<<_w*JCN!7dK7Y=u
zSxBuGi_H#-KAOVsz2;U_01s?|v~>V|YCXG@Ybg0@dW=lfHH)_Es#fP2{FC1{xphl3
zXTB>p1K<^h<1ENpHkUPo>3Pz1)yJhDPJxG<W45@om2nU8Cb;}`{uyXllU&SVv__9p
zhenrvBj!{^T3efdlCLzh9R0=8nxsp<8{q@vMH(jpUF1J%Q6e7Z-#X17F44)A4Oo}A
zNdKrEj=2IAwzDLe+GdrbZdIefd*hm2?A{&~2ZX@2DLCuh99&a46`3YxzYopKzQBtj
z*^Mun9>!PoYqDgXz6NK})VmFc_q>mYZgj%B{lctx{}QGSwr##dr0oqTBO=+Cwy;rB
zT7_2+PF~;T%vIrH*a)58O=nW%!3dIq4z#zmjGG7(-%!e^Magx|2e<mf(yxTiZ<_Y-
z1x&?w;xAwD=%WQI4HD%AV~TT2?AvNbd-!hJdE&3wz%t6zH+NbLt;+AW^3NxU>%=4S
zy||RPo&{O3=NO2yhir&sEA=g>+-_O8`{xgqVKArzl^!jm(c|XSo;Ru2KDZRM+rP38
zPdJZ|aJWq5%%h>nW0JOg+?|Ea{)=%)2UITc9EUd-5KU-lU(c9k=v3^t*ofS2PNB)#
zoybJSkeKl?J6&fwzh27lxDZv>DztzFGqfDbkGs3Co6IxJ9_6^p%2Nu7-fk{FmE%-a
zXlrr~-B|_yh?-=UT4sq9S0ryVPgauLL%Yh+9eneRKD|Mg8#A`J#db94z8Gz-0<san
z#wL2>Fn+TSA4xnH^tvW%hj-cZ_LDChDJ}ABy4m*vkAG$63fgT)u~hxTV{ADbv>P>K
zPwlPYh*x<{Jxu^HoVe2!`2}3h7|{}hh^C@U5(4i?x}wI?l~;8qEQ^R!o-(gn9ltfB
zuDjlH6{Dvre&tu2s^ZEls#TY)1TY>b!M}thl;q;Y4a$#o$>MiKc{z_^RCi_UVO-Y4
zD={`dhZ1rB^kS74r<#go{;JI>;F+yqW^uI3*inX_{u`XSI{|05L|>Ct+4#^-pe_62
z9aWA0&gnX)6N(GpsL*-v$-J|ow!e4PFno`<mEykt4QdMV+xP!u1=GAp$zgwx^&DJj
z3AHecX;ziklz=6j9S2-a6u)PKfzj<DCvLJZL4=S1wh9;E>sg{{;2zd!p@z0vYwe39
z&5HM(9ZtO(I<G6O#H*2KL(eh%#Gwp_P}X8wgzD%<Zmc8vFG&>+Q{rrAJ*XnRSP`({
zOCd6qg7-u9a26hllxW{bZF0<~eiJnjMsF=MT0UwT7J&5fSaEowuUJwwya=~A1DZhH
z*hwP-EsRF0g0?c9K@)IFQZ*79xxrh1T2M3+w|oM#iFcCe@ns5mRx&F|0lYrdzYY?w
z5dl1oU;&NPap(}Esb)x+edknxKz-wS5hGg|Gl2}CseIEwt&#np$B>B$uf!!&l@V*O
zy@pX1e85{!8hN;I$$VJdT%i?RsjTpEvN?isud(qQad|ag1C(LTUF-xDVx$?y>3Wym
zTmPuSvg0nS!idO<fMY@7-8jsV)T<tuqecVq8q&jXY4u*C@kAs1ZcqdL*?m3}Kb6L%
zOESBEZ2#R-(!N+dorDw3FNB2>y7hbiXzJyE!T|6>5pgBDjYnv;(G#_Lh1B7&-IyVi
zRfB)SIO%MLHX;o$0kDMT5xq(vlT^B(rXma^L$1R8pc56V(m-&MsImr1bBWc!LlT;D
zV2t6@5iGkTbK=;*{gMnE!w>#J(ZgWKATZ)I6r>k}>c*kK&`i;R4S}(yk;B81<>8Q<
z<N)1jSV^fI`{9&qOmv_J=ulvA@lr$6H8Qp56m=O8Y+)~yQBZVsibkIufI$y;BZ<E<
zVP)R;C!weQBJ{gIr6ojV+5PyTCx9@!(tArTb8YYFv{*v{qWt``-0oscZLRk8!0r91
z9>1lfu=XH7qgCqw=!qCvT7w&D#<ahW4N%XD!bt_c3#D>YiwuUPq_=`J)6#6&&p%uy
z6LbhmZw|XZ#U#Gs@1dg;9e^{SQM$cTw>^`YG}Osox!uQIF0i~}*EW29K$ynP;ih3y
zx6qo~s)>?5ab1b{sLq;~v-pG2PO1egmmSO@{(Vef1C7UP;vadzmI#~-weQgX3vg2Z
z3vA6`=#7A0XXPfov-&n8`1`9yAc6RXK?=mMU6Xp(pGF^`&goQ+yRFB$RCU%&jY_U_
zdS0H5ksp^9g;9F$_|=iJEI6{@qW(EMrEsC0v5BV%sM}Vriz~PB?SG1O)Ul*3AkmGn
zG|QRWFW-)bxQnaOglQtwaL9_22RusrmYkDXjmt&$L{i>kzGi%Xr)`Cm-H3Ru&A8$a
zy%8lA|LfQi#@!=}!|;Kpq;+5M?x>7Ugpp)`b<{sOZhgQooj4DKM12Vo-VGAIc-4;g
z1v~f*CP+Z5CFcXC?EVrr)$cTqv4;(r>MpPYoElCBv<daX)dAXpmS(u74S-QpAocSh
z`FkgLASlw?R@{RqO1!EwDkaTAMP`9?L>)Y)FPW64QqR8LQ75$r8k-z$8|)|qo?S4_
zryt3{v!5Sm6quCEaG}+$3CH%Je=*i*mOUDeSsR7beB4bffaDpb2ix}`%q#TPA^H%R
zWPjnv4Ts0@{!5{33Q?>b9T2K7n61sKG9joe*f!HdpBw^p<3L?ydaWy2rh-XVq&ltW
zs7H@Pm?Lo<c{yp5O(6BJ`B67dqr~3}>0etrPmsv>`WH0BFJ7PJxxUnV&w!>4Zed|~
zQ%H<5_(V?1rN5cOR#r}JWBgst_QJ)v?XYkj8-z4Ui-f}4mf6nrJ%D2>H_~}@JiGGT
zmXd=Wb~g38)TF%T19KR|kBcGR*Z1p{)}HP+UG<!p4TF=sM`;)3PnIf0i-Y?&Uy33l
zj#cSz4q&|}d&x8SKVU?xYzvcT)K(Q@Z7EGTZGUa~GWFDxZkq}IRz|%R;{&WCk46p7
z6(1i@_nlP?uIMT_B>=GNzRj6Hzh}5+m2_UkcoKgn%k+J?t`Tq~F+NtuJ6u@t<K|s5
zSpSpav==){>=cM7Ua=@TJA1h=&NbK>qo`GUSTgJzaZH+ilW&h_2D9PZ6`RNhUEq}U
zZsX^>WJ+Gm5s5g;QWE*So4*AT78mVLzrqYmBm<;4J7tM_kq)?*9rOZ>NXJ^OQG;BQ
zr`J1C`F-32ZOv0q7W9DT*t)r%W&QHEjZ&}(J(cUqU+BPfb+=OVfNqE7Nr=A)E}n#}
z!iMs|V;9CGU!0Q#DT7rp-js^G)1(Bep^Sg<vpt3k{0B3Hl1ddYb-#(m2TS*flxk0@
z<?RD1dxBZ-x(PttW420T*MAq9Y?hsONcx<d?D7_V@3yfKhKyKQ`e^3P6XYpXmC8Kw
z8U`+1D>PQjRM}JrQZnRU>WiBhO$&tInwqAuGD1Lh-tyAhCado%fLwE~8ro_6vo;;%
zxhot+&T*HI`^)oNs9oDgh`8vtSIq34r>o$BN>%8t^b8<xY74X29y9i?_Gq6JMv}nX
z$JSk$w=Y=ty9R00sNJn^VY$z_<YC{v#}iC<!)@DaWxy6CL!%vUU~)qBpPhpb0{PYx
znpgYb1C%1PQ+|6tFEs-nQvM)?107>mc=V7Z!4|`-l3qt#<mYd-=fQRxoI__i2eT-;
zG0^Uetx3RSAcpgk^z5=TLSp`z!8dQ+#o3!(5)i|iydL)(F9AY%chN_mUUZpJCIP3+
zLXz(D-iwla)8H;$3NNg~g3EA^{6ff`F5G)KQF<7RoG#VBk-DzYtm!&cgOlRkY3bj>
zx%(!3IXx?Pq3c~dO)CNWlP-_?v)e0e7_Pe)lz&YhDsy5+7M>Q4s=2db7#KztntZ^v
z-XtO8b^LD8@hXXR6OAh(^<39=2)Yl*l77K$7tTJ90A&vF4;)m8$5=}Fk(Xs0S}<z3
zBx8Q&Q?CXfaf{=Q?<0+CZzsz=yT1t*Pn-PyUxc3d2c#ZBgq{|NCqa?+VhEoDaCGPc
ze7p^!LT7Vd0p99v>uv}czrC=b=f^U3jo%JVq5#CmSUSwoqjb9uD}rT$245T^^cGg4
zGnUv)(`9`hf!E`4e^g-qsMNNo;L8S}<{cHcn4B7rr=pZhI59*BiBH9^yy(8|{1LF?
z4TNPI#5#n-q1>C5s0}Zff7wC?!1Rez$<06ADH0X<+!b*TkboYHpb8;?Z~@IFuxRrJ
zou&jcs!+F^stmM?BB#83GKqr(r=1=^p%|tEu?M}_Wnclzonwm@#Ya#z{tRjP-Kz}J
z$v4u$D=*W6&o}66#LqJ?P83|GBjcbVLbVwx<`iIqnrr3u!rFE!h6#&+nc*E4YB#Mq
zTy-)HM+Xkmcp7ch_h82_Zx`S3<XNfGVYvh`fnitIYHTu{q1nI5Q3kAjUCRQjc4p9p
z_VT*tV=;2S<8x&wdZQfzD&ZDjB+GFrzQkRijuWV$s@3>d*p+7a*LIl(%y*|e-T%BW
zT$`I*0E4w%Z+$+t@52lOqSKkN4To3pTt=qF=BbJh1%=_rFkX8Eo<ol_@N#|XHfy^m
zko9-?%1CF1FM3NTYBH^&+2jhTImq{&1|WYT<gYgl$UN8|ogbPCLVwAaN%G=FVs`dw
zIK$FN?eV}t3na&A1A%lxib;mfg@#KST1&!VAK?E!8;fZUGMPLAu;yaO`B%6LEufYr
zHypD|c3`DJ%4SFBf<CT11U`fGzht83*3SBj&Ssq9^1k>#x60sn{pq!#_RghfsSRy`
zj#1ZQ+@Nj23uc+`Dn^AjBQjd1sC(d3)g~&Cugxbw562T6G31mQ6WNDP!n>_&r<DI-
z0MhxeJDX{I6p(uWWh5@>3|0+N=GMOMQiN|$5@|(_qCcRokb*MmoxO7AsUJ~h7qtW(
zT#-yJ8=D7>?XS}yk9l95HvCNXz36LcHrzn6%7ZjWD}k5|gugPZUv$j0GV;yb^v6x*
zS4wo~;6R=z<MS%r4P_<&eMp&Qc3+x})C_nj{ZTkk5m3}B$)W@to(H2LDZGi9M=f4>
zsE`{mV?yF6i73YN<}Ob9`<rx`$6;`^JU(s9S8=F&`8?*X^20ROspLztIx#RlDRj!I
z{tsx4B%U#FcB6bQ5uGYA*RVb-2$Xb*zQlTwMpSE{tv;!Vzruxz0G{Xz1@$18InqhG
zZIOdpIAEF-bQ_f-1ww9IY<iKu%7>4b64@c&X7KH?^ipiRkQE1W_+}_J$qtrINR^tt
zAp<tc70nVx<nZsK(vfeooA6oqKk7D3zv4~m;{B#FK|gIxY~u0!!^>h7FY%j3QAzq6
zjZ<PPnLvT;LX?xBP^fKdofld*IGr;vR2h|VUVw&-xjM=~x8ei{CuP_i)BS>_VbhYZ
zVlB4!fbkBVa-hc|sg059gA}qnZ+3%q(-~&MeU93lX-;AAJ}c@=?$aB8n@G9c)UHan
z#&|y}E6Qu<a&^HAec+xKx@%#u^}o9w&%xTesY(WJaA0M)E%iKXv6d?aA*7Y=uG4~I
zd%#wuPGMn~_ht&3?LA*bMnCp4Z(~Vxw2TJxwN5x>&jy?Nh;T_0!l6KBZpRi5yDNLh
z<!K|C$?wH$cysg_ycUw$HG6)KYZq)vR@U;;C!W=uGQBh-I{OVBeHZiao2;!@UY;8i
zs=p{LYb-W}*_I)Hjz@sJN)ljHRN!zi4*)8RCy95_HrdfHJLPdjX*kzI5jHtEW$8&=
zLk?w>N9-0Q&tx6uDvV%Zi{#5O&y7oMm#*#8uI6*EavS`Xrz}EUWu!jVh6zWU+0PZo
za>=()w;Qk6Qs-!KNPmxoEVQmRG7-~Z551<Su1s74HhBn2KA^9f5+70fAfv&(^8w#0
z6+(cA+XA7!Z(!B{g57B;!}t+vg3AnVJW_gW^=ncI+U0!uq8X}*M<-yMVi9pZN4<JH
zx#vo(KLXoZ+r~JC&lg0(G7wB6yEnXDV4$q_){%XBDvNYyzo~CBQ=r{yY@2H1tB&1}
zurXaa69Y1E_+j3M{*@n!eU8sY<^z0qvI^|eh>C|rRueFJy_iW<u=!Cp*ZAbAOeF?w
zk3rP-hRW{Qz$TLNtzVt`fOg+YxkPj(C&84(KX7K$i%{@JNh#s@4_m;Z+s}QphR*jx
zrXfdASRwrIV51Gg1!h@4&XFGOhlWv%ulKd2DJ5c$_Vzm|kEC{1F~f(MvjO-K2uhH@
z-;=>A`WMx5V=3{H4VUW)CNdV;Y}62~q~s^9;isHfYZ+EFr_0E+a|fX-YfM)NyPJ=O
ze}70nZUWieM=F$Y%j@yy9wU|w+KHbOh2RZ14Im`Wpr(=qJ*psb@OR242}i3fH{apt
z&K*P+s^&>^(zEKtaZL;m-T+GPZ`YTg$dNS<AA#`M{OAf4_qFpa^N2I#XI$SJ!7K-Y
zQ<@;6jASxh(#V5n<t8chQjtK$I{@d6e~gekj3vyXT@^d&gu&@pTfNop<0nz%FP^5z
zBks{|91}rREpu5l9@hfUoulQunsZ02vNZO2Rga}P!TXJ{&=9~+2mm!FPU{6rT%~)`
zn8N*$@$$)DuQ~sz8KzOW-{OZ-p{ByE?I2ptE6pD@Pf!wR1)6s|7WUIq1@mcH{TZ?c
zvyBV>w4HFCYmHwQnd8+?urH3#&FL<7I^Z`k&V!Sx&Euiakc&Txqoj|HK%8$~5kpY}
z!y+C@>Y8g{xjNMXtd-0)oSW-Sa8PUJCM@ydU?1BOU_OsuZ?b39z=uR^fLJ_SDuyIn
zI;^%39P8>Qd+Ghr#ISeeYrcZgTkA5PJ_1vB8jm@j-N0%0J9(A8H?k1D>h)&si0)70
z(2dkpx;bp;RI9uCU~MHe(N!_?x<Ag8^01MdzJ5sNT&ZOgq>??c7vQt)ig?wTGRw|r
zZ3{j22yVRwS^PTS9}3!gi>gd&b5`9t&Zr#)s-M@INPYTtg-0!Xke@pNfOI{i%|&qj
zM7x-onSUlYbf{_e*6^bs1cYszHt;(TVCu4H`~RKzfHE`x?314YNBhYUrK3!<afZ(W
zfJ>df-W`0dKxRlQc~*|!KYs4*AkrhJ*38bk0^VPd46tf~pB57^i0a-C2)Aw!0oX)n
zV=6B%usa56LOis59Z+}gh-;ttjDTTNeBb9hw#O5H3HKk;Hw>7Me@{GgJ;cxuhT&QQ
zlz?eHDb%vC6l!cxyR$`V@Z*M33R(a_w_f^hcVMQpO$IG{kT<eTzd14xUsVQ6Pdw#&
z1sJcAx4SHFvNIA#d6uao+Mg6_$a;!d>=v8HcbJ1V(-7rWDRP{yfLmMxxYd4qrsy=I
zZ9ISp#85tb^iubs6Q)JiC|{uVxooO^eS{icF9(1=lNpjS-Q2vo8fqPb)jl79px1(6
zBI6+=wf2cjsYbfvzK8~5w4hxPl2*jIsM(bNq>W9(YgBZ%qC~C*2LxRiK^L~e_OIyf
z>TZ1ERGs_bA8Kw|bkw`H6UWAsArb}nXq|{FJ4m5mA9#ltvlXe7YFArsVPqzIZbE<_
zQ80wa-NoEa-ywfyd&XVLPpTTg7xz_M$q!Q1o`|Mp>{p|N6#D@Ky4Nd7mcVwg#j(M@
zU8-4PXk~ky<n<_a1mzL;TyRHsq3=(ctVH@PilFQW)4$~oh998~wYSqf2t%YPM*zPS
z;O$E``2O|p{C=cRU~*mB2PdObZ%}piQ_K9&k1LzB33;N+FJk?VNTMMiv2x_@#6W=B
z#v?v~q`agyVUDBG^2Je)G9LMXJzaEc*LL-sbqtP$YKEQVS1Mg3XJ)LiD(SoO(`b||
zmofBXqR4z>8YS!urgh2~gmt5^m;+%9d!;W+eA=94B8p_0YVyXDiA)Xl8o7AD_AJBi
z4=twF^i%PLGm}uiVW>1fLNb&@wlhy5>oEv%O@r66rE!xZNCFS%{1wB*$Ym0=f6^9?
z;O#}kChX&Z8HA!}Vex3f;UWCgpY0{fXS{oz3c{v1#G!TW=BcrLZXLm-bRzC)UzrOI
z@aqJ6>J=1*Ja-N3soVzI;mo$inPMO(2YWnhzh7Boi+oI?n^VXD^H$avJiX4@>*zem
z7G~!p4tUJ4#R705)}||06YBoAzwpD+e~1s+{;AOgn!-UOKX|yv3X#&y!9mE-EZF!M
z&B`j1t*b4aRfl%CwCFaYq6ZT;^^;5#Co+msU|HKJSdGUgqS<U~oB(SS*!pL0-&5R3
zH#lz3xa%elT+)C7tSHrH>4U`*WGZ_i2(PzR)gRmmdk)0iyDk&n(7X+xsXYhyNJZD|
zWQ9`2Sg&*bE)ugMf7NAl2AC~uu|7haqi1+0e&5Pz25#@__^!E4vHXHwGFh6QeshOn
zY-Exu65n{QqdpfEf=-n_)-i@!gJ-EoUD~j^bDl3XLs}OC*g}ZEx=GiC;s!(wqbz1N
zEjltk&Id29vupW@_BLf4Sjqp>EYNSssfS+%53OWo`(rwRu6e^zqlbJhQ2YQNoC7Bm
zAXE(hw{Tmu3vc)RB>~WfvBMw5`FX0kBnstLYZ>ZA9dfEE2UCuwEuqB=J+F-j(Snp|
zjxX8`*L~FwsC81HO5*Cx6jS40>zbNj<X2QB={mN~+~hEs1wN8-1P#>&_FiSDI;|qv
z5YeaZO=+z_vNflq$@QpytxT;Aow;X2Mbg!(5nP#Hx&%kw>MN<@enWGS<-KJJr}Z$0
zW3l{7&#mQP?kHsj9*~U9%-T%JHm$7%lgF1Zm@g9qbgTu6MT#sN<5a1IFo#%Nmcu>u
zE^rGdz9LArV8B2x=5S90mVi3KR<WT=7R4=S&ii^h@Us0s0Fyv$zsLB%?t^GX)G=o8
zl}aC^`Sjj)Xoixj;6I3Fg@n%A({et&w;h`CBze=()L{@;y6(6`b+=Oy_MMMm%j%`{
ziyT0i&3-bEZf<`lm!F2d{9b-X3=SBxaO~2HpAh0G2I$BS4Nsc__iXgr*VP(Pkx56N
zEvBJ2ztP6qrfp2gSE}5Y`XH}vF!RvDyL=+glO_Op`#82^66Rq}sg3(NX13em&2xV9
zG^~_kVIP|qdERPhDtqDa45X|v!yacel<r=Bzkg~|Ff4!acz{z=Z3b)=ep}(^jZ=u6
z!p*c$bE*}Jo`;JrG%*jUzt2tGlH|*LK9Xq~8@Xd2!>`CHznRD?r}&Mm^w>JykkzD>
zr0RmgHN47LN+-5&GP}0)*pga?z9O};6dLL0*-`3bN82q%ghT6F>F}gY-X~EpXxQzJ
zld&3}q6L37ip=EP>ue3Blt<|Hvn~))cPstPRa9(N=wPd^ZB-qesB!wLrGeb^K8D#y
zArYsV$$F_sd+u+|ME!~PT;gwSk8dx;Galb!Tr0r>!^7&?{&tGZsOKw1B}pZWc5|PN
zCr#SEZ?!e<l&;q_{Mwx$;}^#HX?JhiPFDmN1cZOsLfh!-9hR6!#_d9jp(JuAnrTG`
zT8M9}YFfI3&5F{QHYd*E33^~>RNd4f?3;cqRpvaze1D;Z_~izO@Mg$dT8PUXV=^S1
znQ{G0PnKgjeMI2-;zNu2Wun56UcL(g(7nk`N4748I4hi+v4NHw=&P4Qd=PUxzQrRr
z091eBULK)_c153G-G#ZuXSdWGWG<>9C+&=i8@Ovx57~;jl$rUZ?$xAF64n{aI+bNL
zz<vG>mlB@K=x{oo^S`mRxEjCGDUPItn&YrvkBwj?Qw_wK2fRTOaw^hb(9NEAG-1%^
z8j3u3DpgLw8fYsAIm#*#q?_SB5k;L$*ARbOymlF{Lwf0Eoa--xoVNgDg@t9__dKzI
zdW9Xl=3|Z5!3LfuQRRmFekG>YJ34aTuSDD$!Y%jx@(scU;@W+`948w=9cp1j5ggv8
zmR2yk**4E7FX<4{&Ip&Ui-c>^g58YrX&RoW%Oz*!D8~Rcy+gI(LUp_=W<K!bDz$&6
zh}Tc9CRZ|Fwl7(pa5UL<)O%Pfu$&hmYFs`0erH^8d--1s?)6^_C>}K5|A5PTp)B3f
z7d{H%x_!BLg+PDI>6VA`gz6?p?7#GF`qpEo|6Qrx(Bsm4s+#@Ps2fMCNVannaGC=<
zjD@RLHyvmmg>apIyIA?~J<b8z=l=x&pQE|omjQ1M6PKU}Hx!p6d<`pqNpmbU48G@A
z^dES}w!BuQQuSs?sNx3XMsZ@h@D6a{?<2cyw=MTFy#zRPCgXO?+AK?-rup#uz=v<1
zZtqWDy@47mb6QwAy#F=`yutCW;5caXV1-}?#PI%Q`0<%@Y4Ez?r8oz`%i{kIKlRHe
z5DsSu{3gW9C#yu5|AyCp9q)^AUVN_1jTAIq67PJDcfdcVSsX8eGmRbT;+VLH>*c5S
zZ=ZM);j1_82FJ9w2IlxTRx#wV=3nD#8hO#klMejjrI4~V<M;SK-c#{=Ty0BBMr&qN
z(NfTgLM<sb^2(%kP7a(Yyc}p4{zC5i(eHyiPRy<(74t@QEAga%y`UL!SiVV{^K!}{
zr<8Q{KvB6!1ubfy^E%nSk&84(*)nPakNKPGIIia<E#?uC=AqO|r?rbW1uDuHPS&N)
zs|U{B5+y#aC7vhUjj|NUM0rk?RT~|&e8g`)%np2xG{XOp_q?i!`(^YKZ-ODh$2a8L
z>-*E6Pt@&@W_0d<pT9hX8wvyi^e7K#BtQK1>+svtca&<q^`onCgH~-^O`tn!bTt5m
z4pofJlV~=pEitY5F#U;+-Hf197s_CeK|v`?-Z){ItrT;fPKJ&R9Jq~9Dd;!Ek?=pp
zj@|5ejCIb@6|;3HdtNiIpyMz2GE_dt)92UOUvXfJVV%N%r9^Akes|%gOyBR7=OiBu
z((5^`q)5q-BxeE!!FbX@Gg50HbAnM{a90cjTYkJ?K+n*Kl-eS{2OP=)6V}re1T_2{
z@&Q8eZP867S4?R^pU}YvUxTi8OC_aX%5+kZ_4iTq%-dcnb(|t7CP8)-zdk|HG39JW
z(avjKNrowZ2A!?=er;=kC-EMG8%!y1^U$U5l?$Wwc-ULvkWX8#9*GCN=X23lTEndE
z)FAKWcx(^s`4}#pXZ=1t6A?Jx#zW<zh|p47i3pzz&*$+4$}KUQNV8M9+%d~dC&h8?
zYBU!{o<_SeX)~Q0s?KBlvW7J<@Qza4)^Ipiqcv}Tc~qt|C}&)eM!0)FC#!Waa#XUG
z98gU8TJ$P4<!#z7$ZMIw)R|CyK;GApu6Ql7-m3T!ok{c2t)g7b$l3-@(VK9T^2dk#
zE5%kqVaoOQ$2et}$NY~_T*z!vM5Kn=U0MV)>}C~$n}|;M6Gs&Y&p?XKMP6rx2Z_0L
z=TCKi&ZRz2wMn}t7Bf3-ERz;WZm^Lfq8n~yb=?XYQ){8h3Mn>+CxXEu)mw_A_a?G-
zGxuj<SYu<API*|$n^ChqxK}%S(&St%RA~H;6sFJls)$RbgrQZD1ZIWERZ*reIh2db
zIsAWyBxdFwd?5MbPki_VZ+?rDL!kXTPW@7Umf^!6!w;eC+92$3-y>o4O^eCn3SmCr
z+E@pTWG50DJi|J3KZyr93x}MS>txaSyry(lOUn>haCZr;bee)d*b%@GK!bDyewsTW
zSPMNnYGy|(+F@rE7n|g79I0sbunM+RtC{0)%IJGZJm}|0)^K&i&*_GWUa*eMDT9$J
z0MzOM^EqV#E@dZns-g?Zrpn2>h0|d|#Wc7N#pyKGCEY@gz;ydW-GrC%fDIRa;}cDa
zcCS*?aD?0&N-$mts520DV*>AL`BmYj2K^G91<)_ePhgvR2=q&{I?%5L{IXs$6j(-7
z;i=K0o%m>u=InG>*lGv#XkG>cI{;KQ5NjUsH8ffQlKp{Rft^~V(KurPl2u6`vbIM9
zwQk6ej{Z&)63YW!_Pc)powFf-UBq)Qqe8QoJh@OJ&EN9O%W>o}g3Rx0sYd+06=5Un
zNG+LgmOj+r)J+b!X8twq0OK|ub9ABmDtCZ~u%r^j#j0JKt-J(;&nVZo1#2?zD*GM}
z0?hdd`?pp06$fT8Cu<+7+@p$Q8J5}$ume|^r*r_jz+r>=%UOgy$jw22e*nTxD6b>z
zWb(g&u#*)x5O&JnafBTlTEm+E0m2T}Fd^(kcp9Ph0SG%;;R1x6Qrt$^$rBeK?9>Y9
z5q2uod4!!(Jddzbit7kFnSC5#-(m2Cu+yclMcAqF4<YPy=beZMMl+$W>bVhW|2o1>
zr8$PMQz=epS1Ts`Rs9QpLhU2sZM%Xn=i@D*_&~gEpU?;5ET)*46m2iPFVtd*EjuGD
zg|Qr%iODXABRQNB*V^goDL4}LPI$V4kk((<N@@xIk2*;JSim@OaVLpTJ=96Me?(s?
z=L0Q&d-_VL3q&ZK=_~1gJ$p)Vsjrlebz-+uVA!!HQz(<bb$p+HXflOTO4-hPUXN1@
z+fSP-c{F0IYFBwIvn^*}yIpp&69$-hgZ#)&SZE-Exug@8X5Q<W@|vER6qZS$DYr{I
zC)X?O?CW}FQX5?0(w<pJvqS&k@}8Nrz#Ko-Guut#LC%cCAn-vwGigjX4C|hmp6i*-
z_rU(!PTFzBY$zOm&mX5xb<%c|c+mTitkKtX($@Q0QW~c1RBcUwy%f(SJ}vjGu&Uvb
zQrYUM;Tfuwp-+`xfCD_xDjH|SbENP}Py8@TQaoIg6>0z))NCgp!NOpx=0^NII-$^L
zEUil95vk`e^8H3mEt`25bI>yHmPS4|PsZAF>~;hVY*<Nuajb=f*=Z$(6_~<ZMXgln
zDDqOYk^3Rx9S%HlQlCcEeVS1(+kKifX*4&nvhSryp#}@Y9CXwea1!RYxfg_DX38+K
z{iHL`)v0YpJ4$C$^DN!>l?P@tc)|qIe4VtH&;OFWq{<`ePC(mX&d;Vfc?YC*AFEq3
zzl}_L`b0v1GfXLoiZ)ybq+}Yprvl5RQ&_r%MIfDpHqX^-6{Pz(2{0PNRBl$Onoou+
z>E|?qgax9M!oyM@c`{3A%2iL_$txjdwx=n%WX4Pm{J1d$x;}8zQMhk}U0_;z`qXmY
zt|XOPKgSRj1B`*`#}O|==5{c_#kTgKz;x11Z!&y;;QLT?CGtayxbiZtw<w1V)p$pJ
z?tQ1?95ZdlrFWjn{kIyBS$MbTKGj+|3Z=X0tSTqV$dM;^c)2st<<l9PPiB3ae~)Ue
zDyp2SvphJoow};WAEmxeP374)ux6z>-t-mRM+loXeX*RuD0sH#i$^#Xe!htpGY)#7
z=SwJmcL?hVP=D+b(*=C;3OVUt_zs9&R7m>OD4`0k2Y!D#{-TsnI)Ri3XPE+ZCB@Vi
zp%h^Of#*zk+s)puo7_oRU-t-dah+~5pubL$0{R!|CIR>tN|F%P3$?8=W8)9&CgHIc
z=*BpP*Oj_SZPyDaI?eEUR5xS8mP{_d{9HH*r=NIS`~x#}x|b0O3ljn}F_)1K0Tq{D
z%rPgIR)!6rf71Nf|MvFFPG%7Z-_(3RzwQ3q^+GsO!niPed*AT`qe<wX2Iqw}e)#L(
z!}s0SN}*7eBJeN#e_bQ#oe0V^$zOky!@uzG7Yx=h4F5|*-|I4R_&NLo3!EbygiM4r
z6dq3zmV**8#7uS~VRaPJ#LG!+<jg5CD)Bf~k|Nrke{3?dKrl1uoV7DEgLx^dLEDN5
z;@QjyW)&m8VE|6DA+b3an1#g?3sW#KDhorj&oCE^G%z%)rZE&pDzH~cPov2+(#_1I
z>g`$q&*;sP@QOZKaaLIALEH-Hsw&R#=}V{Nj+8XZE`6WR5lv_5tYWT~;EZOB6yB*%
z(u^@ef7+@Ty-6|VLd9Gyz!}A;D1y|Vq!_J*jC9Im^zE9tT7om0p&)`WpQIV31Zj2Q
zVY*Z^S4(hCGZtao-4r-3C9yG9f~-4D_E@a!gYpK%VuKR4MiL8HVFCjVkCt&cLQsJq
z+{1ChVh62j;)}^KhWmXTZ7R&cSmKxz#@NH(e>S)0b8O6`3Uyx1CR@q&xDZ8IFA$7+
zQ6BT8b@`PuB7}wfX~il(bI0<BEa6M#GV5T}Rp~LYwm`=`&Ys)7qB8S=wO3b3dOInh
z_6W7)R+vp}*`m~w62pLm2Y~B{#E}CgR3m?M$v*M*+O<{zqXFd*O%E77!XU7428<om
ze=TkXC&Ze;+K4$dG_lzvW&2`InQ4P*_1|AVhOX!DuM^+efkEmiE7@V?JQ|S^N4OiT
z7i^&8afJJ<PX_4x8=(AR_9Bvq5c(B~*4c}|f<`a+V>{E?NDle5@%K}<YwkECiVKAs
z_&4dQ!d;%hlvHjXnsZF!GnU@2Y2}bLf8absB|Q+Rj~bZuv>GU!XgF(_Y}p{4?bS@E
zcI5ypH!uopVo&WTzb^$vGh)`yyw1gY2%3VTUq~=FvCAmp_+nQ&ie#>*h&(SY2|nFG
z&<jO&LD8*li&BI%GxnpgpS%UxWFi)kypponWZE9A{pBUfHSeLc5end>-+(Ddf3-Xv
z^Fhiu5tll;#iSLM386LVjX`zQv^n%4A?&DbG3_itQl*unQj-L;o>LkwCO%)g31?@u
zn+<SIAeAz>#Bs%`Qx=)mLdi!wrvm7XJ}lN>gw?Fjwf@6#FG`N>pg7Sp?l_tKX$QJk
zKw3ebW`W$~)ILnk@S?Z~K(Aa+fB9J8DN%GjI!W=^oRPJ#zo`yo0R44zJ*=<}@FiSV
zhcb*wqWkzdlyN~g8E#1@E+(;&Gj<Useoi3@u1iGSc>41&KNqN0VgAP!sIHg%j7AX@
zT>PAz+L+NNEpzH^B45ke8Hu$`Q2uj7X5J}YE%Md8osn4?;r%C7+l=Bxe^#n(+eE&a
zwR0lFI|hA|6T6heM!yZRHkc2qi6H-t!YIXiu&jxI3{omnw<qN7Vaq)1@7TlT%bgN%
z)2Bc-?te}h{;71lF6Kzk1Qc|s92`oGpwts=n7hiF{k)V)LaN@FcAj~doNvkcic3xI
zcdEODDYHwXAF?gtfex%|f9rxv09>jOMyx(*(U>6-$n_HPC0AO-uk*yJHcwMx-=j7U
zON04zR0~MCh2LOAB(>nzH&q*&ho(_~e!b#TV11aXKgl+x`r3T2%;_3XfZS0p<(_Hl
zn6(*zG!KwAn@##roy?pK5<0KK5nV4V*=RvwVaP0Qx~EL|H*a|Qe_e_R(}+?;8(T{X
z=Z--dRgcazN7ZM3LM6v?uF()raM6_@>LyQX=B-br5{sGdT`GXf=8zR`baPC7R~FTw
zm<nZs+qTAb+S?r2V!PpRk#Nnbk&i`#GZQbyUkyL;o_VX5VAtFd^I{LBrJlxlDMi$K
zvi5{Ek`^keRdc+=f2W4YNK@kMj-PlFr!6U7BKhRWr<S#KtKxlCRrS~I`f99lN|-SO
zw`r3ODQi4gjonfXNbz&MB2Y3@0C?ei%?gf%mNV6Xl3#YzIvcpPicxt+Sn)KkV#wY)
zxAASjueLqKR1kg{-{MfuAICScMsMRs@;#@hZR>q9Re2(}e}_@_CP_et0j>fj`R9Ov
zuR|KWO9;*a!liCO${H^3IU>&iI%@o}T6SOl^W*#9zWtK~ru~<P!mg2k67b*q*o?q)
zo-dbwj%|{}OM){JQEF$<qY=!aqdIl!`8Mo~@`5y17N+i>+zrA3P_dn#j+9SB%PLb}
z<%EgW-m>P;f3~x*HZ77QtDJC)spPrb#QZXgn@QBBU4~-T@5Ao<u=|>IYn}2uB*>9)
zbog7-EeS+@O|5=QSI}JOQM%|Qdf3;I-d5h)$uEpJqf`3hfv34lnpc~&pL-g)OBZMr
z3Z<)j_x(ro(8-FPsgf#eqyfH0k1w@Q*6yg`>G!wye>`tq!76m_pb25D8}5|m`e^dQ
zCFfX61M=`Px~o%<AteX6c_P!EvyKRs{5mu&jk4WM!hPMY5D}HyppPCpPI}<$7ru8a
zdN`5}6%mk&TIG(WLs4%D-Q%dxTwcxQ=CD2bFa@jQgO^T+wtDo#i{9Bi2NVDg+4I`G
zbj2&Qe^+qRy+f1q#0CVc18~KMD^R~ud7qG}^4o)>WU64L)T$Naf>v`)E?w0@M}}w_
zP0QS{Czmh4dVDr&QY|%kEOFs7E+L6n(N<IDKbp8`iy&e9*$F|GI=U^etMYH!9j~u2
z)RuXf?4=nOXRjOdmPyeG+tk||_@}(RdH-jre_W<`4kq&68=2k>eb`%nU$`;%E$VJ=
znB=@3%Q!oa&b*uFIb6+rbf+-)%Bru5??-BL``M?5Mdbwu`a&jO7!{(2S=L;}xb|A4
z_GW|6Et8-7lOlxm%|-6dZ&(x<wF$}lMy~QU2;E<a-U}mkyuvEzUBVSfJa#}s?sL(*
ze{#YYfHmX~;Qu2KrFF;<T~F~wKbdb3%Gq^Ntz8<fD(HCRY=YEk<SbEe44i)q^HCY*
z-ZeP4gnW3d=$&4x{1GXk70v2mX1o$iZrp5t<7#uYF=|L++c+}5+~vDdn;6Nvo(n{M
zLcB<7r`NqqFG_u#?1Goa&QvA3Tghf`2LBH*(33is0dEcy0x>j~QIi)Fmjo0TD3?2s
z4Wxg73c)=6^8Ukv%tMeO*!VTx_^-G3hu<IgfRyzD(Rcs(`Vb!&4MKa>85*ZS*ZuLo
z?$?JOIoY8zNo@qi5Bm1UZyhW7Da$lSlBu*7v63lB613^84MN99y-ZDz5=k~lGBPAz
zB$*nL8LC)TjrQ}!!y;u91eiJHNR?*GkUW1$d=g21=F*oNmA}VlU+8+xWw!K8SLOyW
z)_aN*nizw=SQ2UZniw1yQ6y#3B#T~Xl3ivbbsmYk-1(&+mM~Fp<s-?nDj|6elu;8!
z&nYWe323NoK?Lb|Z9U_oA4OEYsH#yR)!Ws)DQQS~fcFNl$eWU0IB93RseXcn>6d@`
z{v}T&4$_vwI${EJ2ZJ6~B~!VT88J>6?FPBUo<g2}?0bse4Hgw?$kht4Gip(WxFp-l
z0a?;8g5y1aW@dj4#Vl_|@zT^DrN|lBz+2$U^`lWi1m}13eHB%s1o_t))#xH2y5Qwe
zA(Epqv$-pCL<7$5OkvikX~|FDHywYQvPRGh&2((=6}z06a#lDuoB3OGURz;E`X}jp
zP3;~!U-xi^&NGTTbiQtZug7(Mv+qMX&#3Oyc{X<^>O9l2gF4UnyFup_jTl0u^D66j
zUP&sD5!!n1Xb!+~iYt2m$x6tpY_hSJQmP(%fYCw=U<3{eVGETcQ@u<OjJtmxL^v-3
z&1>D^Oe?ohUK8T^=(`LbQ5d<8n2HH}Plq#+Ni9gOXJW|W=4#4ms#~*w`FAXT%XONZ
zB>WhQqXwlWn47pf;rY(B#2Dd2hd~Kz^han(KoO`X*5iq4QUD|Cp<wMOKc*vX<lhU&
zQ<&z>h*Wxk;GlR1Or^JeM!A25x3gif+ItiRr^ey?Nt<g(AAC%eWN<h5(o`85gHt%J
zW#>tyro1JUy7tFN<+Ujl48qwlHq#>-Dv@34HCbwwkkkFDlRlD!s*(j4%X*1^L@k?V
zW;i(x%q~loRU@r7OY2Dfr6x6*c_t3i9O9JcQnF^Ac~$0PaBa(~dtHB(RLhjLR6<KV
z>-BQlxO7ZFfLa{a>;1+>*hXtV>ov|JF4~QbifZQaUsh-1g_mQHW=CsC{xq&d_JE-q
zf?B|-w+!8D0dB_2E<*<frBrJSNyXuDW9=Lrz!F9)xbg;slclZ+hnNAdrcf&ETd~jy
zOM1#EoHcJ5fHCSMhh~4^lY=kTBIO@rUh+Ty1S;kY8NpUI2%J!WGlp%S22Bancwzn>
z%d_glq-tU@Ct*{W%}H6)-?dLF6N}?x75a>8A(g7W_GD6yHg#E>Sx%*)vcmgtuVtfP
zuDTa;5LumJaP&H7D9h5wux_rEN^JG3a2ZAQT#{<MM#Abo>!E+mnvLUvCC!T43IYf@
z^D&%fbU;GV)9mgTQCw$(#k*AmPcp#)+*ZWbEU=hB)!-Hb%o&K9V9Nlt5lSo0?rHW{
zj4#<?*$XNACMogU@1HE%9EmbGxVO#Ky?mg}-C}EGIAPBFX|`rU%<rgWV`JE8J4@3J
zy8DSX1Rx}uY1V(H2&I^J!p3eHe&mT{RL-(zg%rJ*8ILOlt-UaQHmX}J8X@N67{QiB
z=cFHJ_&z3`enL;8G}m!66BbR{I3;AjJDW7?Wyz$~5JN?TShoD+bJDW%<ZT3~_&hai
zT4+i<7CC-NR#8e$cNVbeED9a?lJ_N--3CEJ0(&#@oacWvXqG|SY6BL=B-D|FUnYz+
z)z003w;6>qk|8<jVU(c}BZZh6@-seePgNECz$hy?zS&S*9vhiCPBEhlMp3@qydg5<
zu~)lj%MeyK$BhmoH$RP5r}^Rod38o{mNC6zJ0r(&oTBcH9t+;R<N$Z|<igawG**lt
z90uN7N<M!bSggpqpi-P#!!DGkv1{P9k8C#kAaEwK8~5ExR*F+K%dkvZfjCnwp1COs
zs64Ie_YNspFETItNp0LD@OwTlf0*MB-t*2q!Qv`d>z7)^hed0@L<V5QC9-Z5taf8A
z$qAnkyOq$iqSpX=gWH+bO{VX&n<!^!+}?f58}WbX4yK1mnATNnH%smEWouh$!`48`
zvkqfi)S%m851G9-aAIbHQ9J>ylsq4!%vNB2fP(XJ0$M5NT~odVtrR3sQv`dQVpiu>
zQC&wc`x3_Dez2m+o5zEi=YbVQ@dU8KDc%>Xu-<M2E1cTfDp+B)kAoFP=p?YhsoGVr
z!iRspyufL1gi-p=V1@PafnbIAd>vTft<M81Gl{Gl1*_edOQ(ggcp_Y36gL11tNnLE
zmVJy5lTp|Hhr<?a$@j@qvo-m6mL}n%aT}DIAgEA@Np4?lk(!cEQ<J=|o1ACGNzTo;
zR47S#JW_GORtsZwW3PqPKB`8lxxe14(OZ9cM`5-ve#(KiN*!l&M}o0xjBPhg_ouCn
zm(^YSSr6bNBDvFgR6J$XiNwZx3Oy7q65f^n{d-RK6#P;TYv)?$QP9O3zqC(lVbQ+p
zmp>h6S@XV5JPVXh<s6bk*?qf060gT45!1|iriv@ahhnDL*ja<ymyKe1(g_D+P>6q{
zU&gXi2glct{F;~wmLQEfx`q@g1W>#-AOA|&@+)*ecbME4-_Zj?8<v&{_KqG!491%B
zie8~GirA;_Jbhny?axGi79z=?f05mP>ERDlIlfW$XR7*j=tg$`@BSU%0o)+$N$8mN
zI-P<w!m?8$5Zvg=jwDbALCeeaBzAvtM%wDES|>|d5|GnP?=ZI1DxyYBJVWArvbTEA
zqn-KPPw4Msi{bSTJ<+>SD3x{%;yx^w4S9nVo`TPl<sE^yBc{ns1eok4>^_+RcR87{
zQ`U!NvMy@~<ke^?UN;#sdGMU>6*?{hEj(j7XE_cpk(gz0GQVWD7t7*X?je8lm&%U1
z;JSs^Yx2hX6p$%4v#F;r0kyuj7nSF+95B&k6W`?aZ~a-pq*ZEg^^q)4G0;7pM~zp`
zJI&4)WYeyZs~5`A<aq(E+AKJCJpH+h%^}UllN^nix|G4Ke7+b`Yl?Lodaf*nw0AnF
ze|`6DZeo_3C~9KsCf|Hp<Ar|>!@9u$)(wlmR#5gni$=_oDli|oXqdAWjpK6JkhhCx
zh24D?4RDs^j{mep<G4&VWNj5eZGR`-wiFz0yv={-DEU-<KP;<_IY6SXZb8e}QrH(C
z0D9rok??@WAxT%r;d_dZoz+W!ssc(S(nVK%eabUtY>XYXBA?mZ`m28ykcOtLf-6yv
zG`JM^J%N2}#eSt$dC_DFZ~DJ#ApnrSVa4r})sLLxq0f!%T1gUK`xh?O2(1i{4r6Q{
z@I(H<#e=}{K*Wjv-@*uQ#OlcOW{B0NIpT4j6dt#e@{LrZlVbF2IAW^d<fuISe=bwf
ztzDCRf^k}PI9lL)F8Un&>ER=c6&I%Qj#t0QJFU8!>fm?M<-Y)oR>9qu0dEcy0x>n0
z0jn+(mw-q*D3>;y4LJos0wC}KmvoyAX@8;o=ksr`KRxlV1kRlHU;Rh^+so_If1dIM
z5kDrZZT|7)$v*%kn6k1_LNY5&^XI>s-=F@TvrYA71?h`xjX2VmB=)32tsPVTQKv5@
zy4f-yvi`@Pe>B;hpMM-Csx-|=<eVhW*1drv0JAP<D707Th-tAHv`mo(!U!NkB7Yei
z$dF_W4rDgGbiey&;J;rx8YJ%g{`^UEF*nlDCm9;ZMbL(4db7HuvJ9-r=bEXjrv}X&
zQDxMK-CU5|6lk9O+PN&hrK^~yYGoDL1H(X>UIr1O>_)a4yKi?5R5WXwbDmkN(1JN`
za%xq~qg#pJfPTCBO(|ct+kpcE6o0uiA*SuXEi+sfbMt<og0I|yulUa&f`4GIZO*iz
zNeI;l#>3o3cYdDmXWCj|Y_B&pcwuYXTkuHiv;uZUubK&6aPvCEDcuhK<NaliW`nNA
z@&xP-2$F)GD;fSUZ7N}y6qr+c$0uDS%acKcD2^FPB17R)BZ2PJ>N)$s(tie+uo<I7
zG}Fd1KoUjHwZd4s3rn8LQ*RA^>LuLg+YRm%GY*P5PuoXv$rM*Pb?0o^9Jm*T{ztUN
z=nl={^WE{O{HCO0XfT%oE?8!qmFKdg0uttmxx(%E{J=56Wy#rl$L5VPJAVqP9z<kP
z4s1o`g9&lUInLPHaHdS*8GkC>aT{b?MH|r+vhRn`hH}Wt^}sCH0G~K`G2%!B5x`K7
zqPz@IJ>(J4Nse*zk)G5ILT3!uV~0<%ZBn_$yRtT-OoF0A(L{005KX>_83z;Sy*?v8
zJC=Bat86A@Ybfzp2=keq5lP}6&lsN@NPJq7xeAWN5o4LPDrY!j<$t*GN5nngoq@Al
zF>>TyA$suC_KNw~-+$@6xuEeLZMMT+A5lQ%Gbm!zew+8;HCto48+8#TY9!`bL?@Z1
zwWLmOLrIYaC0V%JY~>l?XvVpXP`4R7OxbF7CeJM#5+P{mFgyl72&7OtPFHrK6--<B
z3Vst{ZZ?54-9`!qx_?+=gr{8quQOuM6(t-aF^#)LR@h!5M+}^CpDP=Ea4^MOYV1~K
zukAJi$qW>8ohu_lIiFGUYs3yNT9Y1j*de{MWLgxSB4g~ERjo6+DyvEjj9^bWy|sOm
zSB*0}3mr3hNKw%m3C_TPO?vjnB5)>guE`RjX-0i|MRj1~T7Sn73I%RNAOTqATuy)>
z0fiNu&&41<B^5S3f<csI+!kt`10ewgX4HMh^hhm8+XKv$`YJcBLc?_Fz?0$l5I83;
zOpxq+V~O2H0B)Jq6K3r;mQUkix>W*HfN5K-dM5S&;Y~g?H`UNW>$)*PJtB$5Ms;;t
z`=edN=*}o(S%2ny_#I2}VM0!0gBUC?NnE+c%4<Vp3?mAV(9X0~H>Pusca~yJ=;q{W
zF?A`#Sw>!4jJ|GJRZ>mY+aMNLg$<^6bJ#H%@wD$K3Ovwv5?jP6lU#sdEqjno@SCxO
zS)}8e&rr-~l&5w#!ZWT#Ts@Gz0<Ikk*qFVfbj%rBaDO?Tz2Y00g7c}_D+06?EpM<J
zBCW3QhPke|50Y5;@(I~XYRl-FQeQ3Xq;iC2Quy`x=__`H^Zh1{V~E_YtLZCtfU}YO
zxI@RVB{j3w?J$V&!I82=&f#%@J-#!_BpxtKd<~RwTxjgZ_m4<u)~U3POuHL<qWg$p
zk*YT|jDKSS7&BB|X@;>5xrY)>9(ohiM~soMkYI`>QW9}S%#}_jH)MO;j-_tRC<>TV
zvfy$;Mu|;<ob$;UB_x2cltU>cm^|Y+lTu>0U`%~RN->h5lvDbvPASOfD*W;hDW$7Q
z$AS)Jlu&18M(}5egg>veoK41~_Pk{a$T$;!pMQm=@g-lb>+@i8eC3wUY5z+<@Q_s4
zmA#W^25{cv<y^H8sjOlg;Zdb#Xl-Ru+ENoj;5UObxSUrQ1_Yi<Hk+btKKeRNQacL2
z7zYcDET*-bdd8L%iMYe`EK+;;@3y<CopGFL6vw=pQOCRrn4C2-wXb-0&OB5ubrbK6
zk$)-9#y3*kW{?pen62&QNoP5Xm4WOn)X;j1FP6;V7ihF9CX89(ymm|zlq1A5D)}L(
zuT@!4R}b%Q10;;`ciKnVmu15{|08Y^!^3#0bAA%T!I*l<PvpuQ9eyodMV$6dapZ4>
zaU@m1IGki!ITY9)Lz~;M`%jSof#Ok3(0}Uj8s>AE4m06^=4kBUH=YM}V_XX-3<P3y
z+%<RsMkpcnZ%x6OrO@wB{`j-hJGW-!ViW_7BQBxa*;-g78j@lDiHG-9TDqNwnzKLJ
zH(9tvrw@vsIs-1~QzW00hs+7dD+&1-bLmqgc}F;79ytsux2-r!-zUzkgWupS&wpQk
z<IO+m;ZH>Dz3sm(>6cy(Z~oo<;t#Q}Ft%g_&zN=q=q1LyA#_Jky<<DDKnga#IlY-h
z<;)a$nl;O8$!i?ue3uUD3!i-zF{iC+F#W@Z|I2HI@U82cgc8J`#3u>j7CwEW>l*Gu
zxQF4p33M?y3<?~1VsDz=I$XAmw11k^D3|$yqEb%hd^gxg%J%$Sw9zykkBCfXVHA>-
zx7xCy>m^8V-6ccdiS@?B@jlZfg!I(9OskR@l3d~|wzIjs4~P5OiaGF#e_znf#)ZnJ
zMI>G5^5%D)8~DGl&{;#M?UV92QU21OJ&eyWK9|$a^~!UTJs;<9%G;(<WPd$ClOG=R
zUcETA%Zljh%JbXG^RVMqX`q4zSJna*s)Px3PSLmuno9-EHJkI4={TWKXA9F|s=9g4
zTJ<ihPz$E6vOwc1Xs#7=zOC<wZtZOdm{3}Km_)g>rad=0v^#%2K$OwJM~0~LHqofj
z3D3`W-&3U`!=TN>FHk;6H-GY9|Mp!N0@Fh%QllOZ`O9}zQWBr5m68srlad;#Qp#vZ
zCuJ%(PNv-h>2W#j8f4O`$3P-!EJOR=5z&4Xu9!x3uK!R(B(-5$$`3^JrVx=t-JZ#M
ztB43IXmkZ4`YLE=0UoGzL?i*3;V_AaumW9t6|^HO(2m%Pb`&rJ_<#P;PP$2J5kfko
z5{D%IR4XMNQYj@dQmvG9$X2woKlLrARN3LY$Dkdln6dhsfIAW1*x<pJp%m>Q5N43a
z4*v^d)jqrRxNIL`I{$wx<O%<o;{)=%Dagb9Us2kRCeYh!0+B4{xO+EEpu><y7{}au
zL_EnJtwjmxkUA`qx_?upl+ln%DT9q_g(M?3;++F|Z#kdp4(UAv?+9&~bK;x9JC}fC
zVml|cd|deM0`|EDy7qn@-r29m75m7}`yYyT1d=fI(c*bqct-%4)W8Scd4qUI`G3KK
zA9$x%h(som{|oYN08hF}Yf(Zvq!NlG{!}X^9a1SJF;cCRbbrWJ$g@B7EvHn`LA}R7
z9sx`^^G!gW*vg5}#)fA)Y5&CgZiootQVDr>>v7pWx^w@>A|9@oRQ3b$yeY)vB>i+B
zfafH@W6uUf^MJ<+Hq;l3Xle1Knf%5e(Ie0J_&>?g;{73_WD{3{h;&39QW@N>QbaPK
zPC{axN)hRZjSom@cSbDEduN~j0}UwLhL-_v4if?~HJ9Ns1{9b5`xz&fsHP360ezQJ
zrwyuqU>(r)oS*Y`%I}_YJ4`yN1US<Rnjj>E8)Uen8?A*l`Z8ZI!W-N2J)k4>#L<pA
zVtKoKu{KC{@LY6cvL;3?vIa0|7ImUVo3z5uBVFbapS|QBLb`Y@AxRxgqL;5@PqNbQ
zDXpzY<<d$l99j3Vr%>67PdeAx#^Q)<TII}tEAmx&G$GA|M-T864bf%upgXmSmpr9`
znwMlEAj?4IOEI_cJ(l-+{k`3J@0#~)grnND7upRLoB-pB*2boOXK2qmU7vQ|pLgEp
z6}L+R9W=Nz7w8ZrtcY`sCUnp|chH=gF|Vo4SOLy2REM?bwk>Pdy0AkmSewcPn$SUi
z^V~7!^ZtrRbB``yO=ubgs!)V>fA0`i{=S1)dI$ClagHtr8Xnu3uQ-qIUx5OJ2+pBz
z-7io)s2B2I|Mp#O0_#mQHL>$1fBCMfm&)dPd#Rgr_EIUSx0kv}>SZl9u7}<0!{g<!
z%aPHUcY;I;1lm6YA}Z+TobNs(AX}k-Ae?+?q;s8!=xPyORt?%@`tOT~q!A>pZ;0qm
zg@`a{VZr=q5fOIK*bYSWyP%y5@WAXNBJ`A;dssz8*nuv-Cfezl@_A>>yJ&~jn{XQc
z$Y{r2g?8#i+KCWmP5N+1ZBM<u)J-~jsg%^)OWouk+PNY0T@I<3aNfJ19mF7ioE_l~
zv~UM2LnZ42y+SjCL=%)L0X0G&-v>*spX;6sjQ}4m^VK5w!gg(F2U+ZmtdtCm&=<Lu
zDxz%owq?YqZ8B>wFgR=0ima%tuoG@NWsg(5(s@Ec{2RC`3S1M#jFqC}@e{>_pR#kg
zT;qj6&Qn8;+fZ?h);&fOQ8qh&;po_8@}bGZ3KeD#S>wD^<0T|ie)Y^Naf|bM`oM;}
zKDTjnI}1apY~aC0Wz+3b^WSR4Sd(=rdfilvHD!>}jTPgD?Pyq%+6d%7^y{}@hvv!W
zY!pGRy8yCDgaI59njSNEj!jVxDrYcrdDJd()42dm3YMJxmUgpa+0|QrqRU#_R$7XJ
zIpR%`n<P$l2^NH%A~#OqmGg(lKxSP6%dz~4;+;MD{AprNtC-j9oVilX7D6F*L}VIb
z%vd0~Q<>1m>G(*z5z2*jB(Odb&mjvM;w{0^9BmkPXpA~OHe<=URm(}sr+J_~z!5-@
z!)+KQ{I)*hG8xdqnb8%0j!Aq@*Q1#|XJxSIJty;wG0oOV-)hS+IScd!)*Tj?Hl8Am
zYscD%YorD%0{MrA8XE)WmBE{25fhHH*rr-H1#(l)O{tvB;h_{ZUawz3bIu!cRaS{0
zT}y2i<47~%EkB0WybY*dj_W}JfkV~ywz0Z3K@KPFVr*4P5N$?(=KIGfEy&WA!a5p+
zGJ_iZ`s~T)6?@j?%+(ZH&18kel}3xiF`%x?)e2=7++rnRO}3N$#LVdm$!1y2`)j<_
zA&vy(!;<00byX=Rw8EB}4L_%T1=Qk%IBrpr3>0x~tt9BGLai&yk<)gn<=UaFsxHaY
zN1Y$1$dFR#!wL_7Mz{crF|X^#NG*&sEp<uLXJgohLR&b<Xd}>J+e^ejED7wwJO-$<
zuox?xZyz6{DXEv77kI2k6Px2Fml`qpyq6o6qG4yE!h++5t|{|2ZebawUU3KdW)wuZ
zc-!~uO=!ae6wCR*CfEQD9CFd(1`k3ZlJJh3xp|%4dJZIiGt3^gT|LPglx;cgdylc!
zbt+GFuTNQpMK}2qBdtJmi&<>gm_j!EmipfQgi)^XZ7)at2?HWrjP)M=B(L-Z=ll8-
zj!t?Zg<JfIBTn8dIBw`Vdw_;|gud&*)=)wejK4!U>@Wak-vXu`!NOZzulE9D+d0;&
z(k}c(Pmc?KDLP6;++8aQ8JISlqo+D==wVG)n<GiPXt{C_n_zC@s?t6z4JRd_9P6eH
z$**WO8Sl>fzo*e?CBU{5Tr2JF)`XIvkX^8<UGE|^9M6;s6mGb0WskW;(opwkZ5hIj
zYmD=ASd4Wmo6K2#Lp>ITNHtlR8Z(5bpIp)`@(w_M%Q)7<cw(VTtJ;Df^s^k^-Xdy~
z$Bku0Hznt_3_m4AM=T^)3>uC*^#x<dQeakJh)0Oe*IVeE<5q=`LS10x%&7~=VzG|(
z1J?`Mt{i!>ldgelvICURQtOV&uG!gRnA=pU{_dOA2vDKgH?3jV0ZD-L!dlZQ3SiHy
zd`n$_PGu~;-a0VJ3qfn4sttHk$+7G$zZN;=i8+&UiTEc@Z1&{yin+~uVVmmNBdV$j
zdV3$G$`dNJrTv_0^Kc%x;)tUvu%1JCatY<;+Ce)f*N47KUeUBZ)lR6&<oac87$Yr*
zh3g>JXd#KJrM^807@->P15XDTpn#)3F^yb*Bt;wR+O3SJ=<1xpSll{iALgedaovZ-
zWj-z9Tih6bgW-HT)Ga?|Af8-osC)dF-hmNmYij1G?uKG#70Omg+6P(@)3N5Af^^z{
z+6XN7GeAwg$3R$y_({r(tn@%elJ_`Chr)zwt-LS+6yWslyPdhd72Nf8l0u4K!L@sT
z%}B1!`=cKx<?U+<ax%DUjxSt3u)#Y6LEdy=<1mZ9k*PrGGrrGh94+ZyCI4S7_|6X@
z#}DuLV>_P`g7{%wK8(u0b++92Oy|8dKfUK?aT6t3HJ|g)*)vAA)%Se5^StuaQr}Ug
z=k!Xwk2Y1H;4u{|{!g?SS+M3l=DWRrLYr5MeMg<n3c2^e!PV+?6hHMhZ7#KWwbXaC
zX)zvxdLM0C!W~tq=$kr6bzUv@J#}(2gSUkA!z}tx43~TSWwv)bgsC*fQwPQJ=7&?(
zPF#lRhzNthg2W@>^wXi8@*7>@aygsmI<u)L6ZX@0efK@HxpFXPYdj2%XqN46c;xd=
zD<ibDZ7Xa}fVOgevgYxCtl#y7sHIS8!{UG_y(B#JHO_~6N`)o7!|~9D@lHCmwIV>T
zX=$nt`l>|MkzijqIqJxk0N>%z+nn)EO?poJ&PU$no$Z`>YeshDz}u?%Hg{Sb{{!7j
zz!{eTZw?azF*cXsG6oX?IhPUm1Sgj%u?;c-ZI@iJ4L<|j!he^Zu?;VO1{aw*{`&FL
z3&kl=;<W$j|MkD#KVJU*(oaYzO*nVc-@d>28&WVLET04y#8^H3{*URmm%sL6^I+P*
zH6|uTD4l-y4Vo+8(59AwO{vyOH8$m-0+?BAM})tHHicO!N2!EaE10D`$M4dA2m49e
z%s5NanlL>`qu?FrNEO(BR2ttqN}-o1g-Gb0|AZ+tSH6`IKmuXT^dm6QdEgd_mKdrB
zi5?|%c9xtoBDL##sMGDU;f^?CH;jFgRLKcIeUR$W%6v4;&q^txs2qHkd;}Aw9#52T
zg)pnH!1!M-j?Q6>Ye~RE1I7ST3iMHJ^eiEj8%>NEJWtO_zpG(?FkoI;A|RL^g!xw5
zGcezd@CeLz5;tJJ?V-;5F@JUJU6}8rZp3`&y9Z*v!`M#DclzCdd1;6grolW%jDJOx
zYOi0!e~5<#+#B!#)4jwM+`nq->#wh(A;wPWDi27?^BQy(nn*wlju2we_x~DuNgj0$
zv3g0d>PNpBo6J~$;=r-phyyMI02GZ-2ojqsE4X0RTT2deaz%fo2b*shaj+T(Y%^Dn
zpln26<6-WvZ9-$tbgksz?Ik}b{Ozy5W1B@yjvgNVvIkM<qOglWm*f8)C{hv&xrs@l
z^k=}tDdME)27Dk&IF&>y)(zli?+cSRUI&mxm!V~-AP{DMS&+i1%ZRob12;vBgnYw#
z=?fKvSwWZ?bnzV9Gjh+n`TPFb*h*z=KHn$1wyotGMxH>Jq`1)W_q@H<U>cAt<?YjP
zmOy}PoQ|NJ2r35mno4E66fMeeq5;IIlPIhwRw%Sdm?uueh*(|1pD>Z;%1kEhM)Apl
z&j%q^hmakA6xx{P>M)O3ygNi*&?=Z$)q18YYmsL-QJfEY-&U{t-lr7|oo}`+EL@3Y
zqmF7GBusflht_G_J8BsM$-;8n${b8S*Sc-lNg1NSADx6pr6(-4opdUfNiHUnuE)J8
ziW$P0>ZLBFC?wh|I`~u*HZX9TLcHO9zv;p5cdKT9#8I#U5In<G@0E~UECwmtPg4*)
z$`PYW(N4U;+*00@w8T<V7UGz;GpX89+9J8_Brlb=l<eBpTgZ^AiC@asLs`LcQW2oc
zrUQ$fPw)bWptNPF_BgAwlCN2%ZT&e`IX^VRaZRjs_hh(M!XtcvbgGlg7JT7+HnD>q
z_eV;9&Tg$_&h{_mVR>P=9A+l>OB9lD!F!4H{qF6CLC1(yU)Pyi*e=5C3T4@eIw*Ft
za166X;6RCP*eXbMvlS>}=!I}RW9XW)tV(=F>572Lnoln5z(D?2t{W^6rX5>LiM%z!
z7D9qUfrWC}HQuv{YaLO&x$fM~O3c0!Di@Z20h%b)L#i(d33spra)v#N_T$R98zs<w
ziGVR>H2L;8;*Q&j{m2`Md&&@9?^6rz9b`RiF%A}44~MSxyf3Bmn~Ztg7amUqZn%3X
z-$`8Hs43-itwZQrOcNBnY_@I*?$n$sMn+YZ$Td2nYt6f@BEBh~ZCL+%&(-ss#nRn>
z^h5#`pi0&py6hv~MpqIHT8R&!D`{!fN$yH;|9qm?;{kL91HF@c4_!ej&`D%^zeI%e
zs;O=h7_&;R;&!$op|Vn^9!OU@i4Wi_y~O7dmTt5=8A~Va3T5e9@8>N2u#X`vz0`Hq
z(i?K2)WLry^pj~zH_9i{mVV^hXiMLJ`*E~ou*iB)bgdVBsbDl?9!XnTiPvaL+xmY>
zTQ(`4ozb@bf2S?N5zg$Dt-n2wJ|(wcT(@^qF-c*Dh#Dkm_cPcGx1F+$&!yTZQ+BLI
z00kSD=ic{jE~gAoE4u_EVTrV+ZOgB0$MMzY#&MD@S|0a2xIx_D9;hEemW3>T$EuKe
zr%1+A4Sw5yF`^%UY1B$UCGltXp0sK1pZq}G7FH6;#a@6XDH_NJZTH?!2yKc_L(}s)
zULM7SI_t1~okA@33EIR|Vv2IgY=Z<nRSuau@yOsmtNza(sq}GSVQ$@#O3))14v$nO
z5yclajA!Bd%rQ*?Oa%}JGJpAhl}>+$n?FHg;q3IE(DZvAhfe>Re)eZJuL!o_9S{up
z$^uVV5lkme93;&swun%WMEwKfn?-EY3{r)4a$PLI$*ErM+L=@4Yo7p}fi09+dCGHP
z{FatC)npskWHaub?-e^jvT{N_$&S#GbMYy5eA{Tp-Sb_yBlkqz=;zpfkvkwHT7HTh
z<u*I+pYIhr0&#{UekEN1T?c;F$-Tjz*1~$QEXe0qQi)6Jt~B5&+^wWmz2gWjemGXM
zv{IZof9Ul%MFyXk1FzSUslK_YK-qYnlp`L3MmoN8AXDvRsExx~8R%j>+h%+!@-GZf
z3cbPycMecFR-kms2T==u$pBt!z<I#mq^{RbkDX7@_4<^?`S`i`U8Qj3Rv<BqwfxJ)
zKQ03>8)bX2NI+sesXzba2kau}yKv~D+b)0lp(Rz<0&9N&tOu4Bggq%eX&}7T{E^c9
z-q~KUq2h$eXIT-}(y%`KLnZm$^Sxq68Tr+7><Ei^`8i-z8_l?Xd%9Q52+L;X?qcUR
zi`cdx;ldEN&Q_NeB>l=KcpTEKgW%wPO*p=G=O5v?>UAP49q}c&<Evk(r#fy+y-vGn
z`E3IH?Mt4J_D~8;KJas{dg|0FlpI(3<ZqXXO(Zm_+)EVX`BT1$IVg%nY5h$@W*lKW
zGs0XFlV;&9z`VA9H;Cc--Jy$bWqhl*g!I%eoI18e^Q@aG@q;7S$|BYNGl3hzdXFn@
zb$C04^0<RAd#8D7#fW9PpCP|g9-Gn0J3PE6a+%*@0%;UL+(}LQQTOO;CDdX*=$mZ*
zhL)+wNwl9i6lNrCwrI*XVU9<hb~$4Ua~OOj5icPr4tNoN7Um7{J<PkY))LwzRm2$9
z_GaS5SKmAHyLbi)*}Sa>Bx~@u7K=#cn~=<1*e5zhtDdSoqMD6od47cGozyo<&g0qq
zwo7#%JrxndmG#++NS_2ndb%=P{r2OcZ=hHaBYAIu`n<o|oU_{Z0hCB~LU!M6zs<W_
z_}zRrDzyH8f==xso?W3U_}V&DeG7BgGBwEDhbkq<Q7V~blGv)a7n6m2KMHk7quR=i
z`6UWF4@&8Xqqcr;A+I=06Np)#9MYm(P20K3mPJho{;(^5u9J0}3gSXBLOS=(X<N*0
zdLA8(q4o;ZGK5pKVp|-{VJH=}*hH1|9!pzERTe#e9<SKyC`axSgRT9Q7RU2ese?M&
zBhUC!k+6i9w!HB?(o(r)cFrqJ$*nf!S-iKDtb~76!L%<;HvUuvxj_pT7x-wgTPO(I
zL_t_XL0G0BciUthC+W50T%t^)X?l24ZK9My7fNCFmy+cKwVHXj;;*r<^N&lJhP5{j
zUZJ9YIpezYfED~^#vc9=S9zH<`jYEn_w;M_5e}9fY*9y}oq1H`TbjP~aP3m;r3`V(
z8t6fbDnk^#uDl>yNeitll<%~~a&Di2crQndq9zsk-WpjLr{z~dI+mQA5xF*qM{FB(
z0}V2<tg?AL*|N!=ZET`%odSke&YG#pBph9TX8WL-cPapbuL93#0)(=Cm3tL;ZHGD7
z3CSzi#b0Z%!xkH0C#35cPwd;l4mHq&7Ik5gd3~Ic+#AX^@wiq48&m5U;AR>*dF90O
z%89S8oWx=C5vh7UGf>96spRBeOHdZFdU3LMua-m<qPqp<r5>!Hys9fBl#w0fdVT{8
zoYVjJkC*=f7id7&mjQ1M69O?emvL_k6qj4bHz$`6y$zv%grk4??dz8p8bxUNRaz3{
zD)akJDVRu2Twb}Aco*<1z=b;w0p|uwsYEU%N+WSJC^u5zQz_Fs(gsR^l1M}Wba=<V
zkN4<j;EQlDhEJ48yo>-b{e_$^@Ba9X*c-NnJ*S3P%3E3FXW(4A8CB$z#s6;wM|S=W
zw>Tc+cqci3zj(pH4XMD~QU-wxxmyiGC^CZM7ROJ;-8XQI0E9q$zh1*+f#D2#2G+x=
zB6Kt|WugEkkJmG$AZ~u*1eA{c<&{OsY1GV<ghVy*l_}mzPCV}>$q=Es%`-%P&4p<<
z`kMHRs-QhBhvGsN6(<xIJ))Iomw_h{=tbm$L3mW6VtNGSe~d~TPZO7ERC?m2Vo_1U
z2&Up2nNCs<Q_k-tg|J>p{Ubk~W;OPj-ZtOX=E`HadzyKZ9-%FT7hDp?ZHpM!1?Tw5
zZCo80r81G~hSn-Ylol;jbWY(dFq0u*wGwNNPYEELi}aRA2A8x&?9QF3FhvZ{79$!B
z3@G+AFd|yme}UI})n6CCqMwoKTcIOwd2SKsag;wUakbot*{0ZgbZu5SN8vucH{U+K
zUjAGd04Pm>sp+3TU)%#J7!kM*!XRF9`s?4*?=OFEsYXvk3B^|kxIg%xccImtA2E7a
zittmV04n$y^>cs)qI2-d2+k<f-p_02l9*7)H*1JBe~~2s)MvWOv5`bWp2f48TqNIc
z+8b<3ijf;yiae^Hi?{B7p`ORP7DQYAOx|mXeTR^p@w!^|%oT86%Vm~8R%Z5=!dfuE
z*)->IW+lwTyJ~jD!7Q_}eika3oIaEWTIMRwFiV+7jZ-p7Y-^f~0YR(Xw@q(%8|#)d
zfmvdxf69gzgI3C~WXhGKQ?B&K>s)o`N0HwpBdjTNx}FV6NsMg=iv=5a<GN!Kcd%zM
zO*W>TF~*t&d+1556bt@%Ty8xZkN<p+s_kyvB5$}#lrS-|Y_8RvlDD)dn8g`6r^aH=
zq^gLMGl30gEkSr&Y7R$O!xz<DQLnt$aa0@5f7xQ;@1e@6BoZBBLzRhDd0jHDagADe
z$Jaqxv_RC(=d43RZD$?sM&Ag``fWQygVV@*wiG`%4LPk<^9J1RmWDI#+*p6=yNGg3
z3HV|Wm|7QxTVnM>{6|#DC}PF4G^rwXz@XoU4Z5D{(M3o)QWKTC=G+MTz!GTm<waT;
zf0G@LdGy;=xxj9(0|+9lX({#>LeM#Z7_jpUNj)#-V3ff+h)fgCpnYS&!)F6(Wd0b}
z1|bO-d?5{Ndw9u-mTV!-fR6!2_yQKT;o87rUTzoDzCWgp3J7JqRL5LuiM1lX&5Rjj
zQky0(pH)&51)`<eS4lUgQ=muGv7VFJe+^X(ZoM8nq=}43Wf!78poz)dH-cQ(1xb3_
zNHU;>NuL{4cra@M+#1k@f_X0?wrxupg4Y7o2NWD6!7TaN3SLt^sNnFJg4c7hw_OFV
z2S1|Vji^sh@J5iGz#$BgMlYlpQF4smOI#jS^5~P-B-&PT%?WV6K=uJ8mx53$f1j=7
zHPwSkuF@SosCOw@XlDP|$dO}0P>&wbkmOKnN{=R+N$(n}JKB*PXCwcuZW^SO7aSD>
zvfQ1L0F^uIxG@9z0jB5LjG9Zuxd~JBwNkgqKchuE0YZ1Fp8b+L7f6nM<gJ38D;T~n
zlUrEhhF^=ZSkm!wW?rxcsAzDZf7^7ZjF}xnO({EuwdxwfPgZ=c7VnfK0*$0CtP7M+
zrQ%&8FhgSeYzF^dQWS1yW__*x*Wc*$FMRk54tD#I|N5an^E7n&WBP}y4t59|1Oy2i
zgJ_yP!gRtAhx_3`b|wMB5-RQeBo1<>H34b&$)eF%XzA`&K;0HVAkVJPe><0p-jfFv
zrM+ddNQT1lxXq&EEK;Q!8LVNcI%p|H;CW;-yU!(?xkVdx#}>|6TQI~jyRo)Kj{?em
zL~q@OIqo1KC8itv*+^iXq|oC??Yvj2U&Gupgw+4nWisLYzRAhTD>_D5X??2NbkQ1?
zs)LpgL^F93S^^(ZRyBiTe^7*?gi|%^d1%1LFuJZ^3A<Sjd2u@D779e}Ob!1$i0&{7
z9dtSFOc9%N(B%r5=Vh}eE2Ao#+F2oE_b?no$r*0ZOpkX~J$BJ~>!CR;bq7s!M|wYG
zKY%8~4ZE55GS0L==BVV6yasI6+G-n~kZpCjVpEjDK}d&1??6gJf7E_@lzTGW#4k6N
z7;Vl8bKDBTlv;*49?dD8f67PaiqlGk!`j?jBzk@eu^T&RDkt8K>Bl%O9lUjj{HK^4
zH=J-fpA`~w7iYGSUMLWX)v>^xk;o@6Rfq>1;>k=C_IEy3@cE>s&<9~2hNA<s@@d66
zC~MSiY04udV2+|Ie~mxp_J2Zm6on4B5VDYS&{dT1xwR?!ov^Y(AZJhEPkTyg>g>=(
zT^0+@Iu)M7<*Okd@%zCrG7d!tVi{7Ab0C(M2ys8cx$crruC5_03WUWD!r~yp!$@=x
z79l-32Vr4|m^*T5TjN{kaueRS_*ATHYj{4Jdl0r^@j9^ie}lFsCE%QEA{p;^)<gQr
zH#|bS(RsQ<7{;fAI}54QIk+=L<lJ^RkT-NQvZGs@ak)Bp(U1x~7cXuI*GL<PEIJF<
z6}6I4txU%Y9;{Z3i?a`tJ5J+rnKNKwRuT}ZXuxd41WTw!g2lKv9hgi=N=|{v2m)Hq
zr#E(BdQwase;20%lk$nhDL#s6P82(mbwABF4U5x32|{vkiZUri6y)qV9HQBYdK>P^
z9Rs_^J&lXhfl7vi;1sBU5?uiW5BTw;mNzU=2PhFD_$fdk@9s2E3Dru;_}Pj7xJVtS
zc!=K5uFyT3ZswP{Fbd~X=or()%)e198c}b>FUdZ(e>*Tm1L9ztYFwTUPBg^PbD|Gb
z#3)(*mai`8=Ll5Rq)k<yRP%m?VlIR(rLF|Ib28q5ZtkpN+>YE<obtiy&p06aK-&<A
zPr=#!tc61LlgZ*YVEN~z*Q}7YS_zk(ul4@c(e2!@tosm}5T#Fn&`4tCe5wVC!#3$e
ze|2DBMi|0_etcNgeF(LS-lqVl1+gY4ny%hD_ont*(IA<v5u(R4s6lb|F;bdn#vVC1
zb1#KsM}_>adilsnp~aD;=JS6SlwRML0dEcy0Wp`6u`v_@HJ5>=4JUuqTT725sS&>S
zuSnm#l7}H6c%;-Ssj6psq`i6Fx7D6(?_N9y@5O)L0R$wFRAyFH_gX&GR5c0lCgLN=
zk<)LJoW4C>-kv_cYMh+#2ArDSew~0GNctBf4O%lf0AV#uZ{Me%pQTimJ{S6+lv>lr
z(Qi-R=JZK}rx`$h0i1t6IfO9(l|C=@9MZhFuFQ=T41E+mImah@e&$({K9*)$m*~<w
zph1{Vzr6kUBw2-CUh7bhRJFWF)C^>vhNLmX&-gInXas2lNe}d49(^2-%_p#N#mVU)
z!Iv>-r3ILIvgm~|hB;rP8PC>AI5KrRJK7O1L53`qQKwJAf=GXQ<Ci2I+C@4u*R$gj
zl}6wh^aayc)T8LN(2m-%MIz-q|5EHUv*ip^L6+q;tx#&)8fq3^s-xCuW{<34*cw5S
zb9x#qL29L$m~_NNM_YTgXmvpIY|Cgfws^DxvgJ8O_AEd<hj<;eKr?$I9*nMHK(bV?
zWtLt%);^7zHCBI$BSAOc&oeoiUenXUud)hj|M}JUi9&gHR*&+B@w6Y8Ft=4WvV)p{
z=D%snebN|3F}y-Ed&4s<v)@i%#u{M_7XH9g&cy&6&a^o;>1jvX6?%H|yt8T+PR{)*
zW9x03Odri314Jm}y_%#z`gm>h^gD&jdi1An#Mi@+uT_6SdEN7k?j$>+V(TCB1kYdj
z;S{6MHBo)d5Xu6XZzWL9sA9H(K=}IB{q^<j>EEj)Z^#=nar*oBr|<wZ2u;QTOQUxB
z{h!mXPk-&H)+H}mO<EiBZ)G(g%aU*gQK(Y!#sSE3qnwt1qMMqLpQ?S1lIp1Gh$}fx
z3OEbEUXp(gaUaF-C8dhE;kgV3R`O-!oFeIOB<J!ybQ6*qM(N&7mbMbQ?50d1Qzjoo
zcAkC@JP%rgeLy6C{zXpzpoc%wWIg-oKXK~!!bVR2n*J8zV27}$NC<?<KQnF-mXi{*
zpPB4J!q6@csrzOU$0S2zgrDbavId2lO;7ho%maUA7TTCn3rViJIZr15H_XsqDZb|z
zW-)RV*Pl?ei3HF$QI6rKGUBq;(yLYYlz~7l#90C)vXY9JmB1y`uv{g}N13h*lvb<f
z?O8$WEP4<bZ01P`jn|t?{w2Sbu^0a3oy=ni))uZbT|rrSNdN}|5BdS;1g~YP6bs<+
zLb-nlXGQ)kxmB@&qETLI!xYiHE3EcJXs8t<QBq^PpRoDDr&ISm_y8Nruk`scFe~1}
zXKXXYg?gd6xS38P8xDudrY(?9G<xVx4+0kCbR{xeW&C20Xr?fFVfUPwnd4^%##<9d
z_02&2^QQV(uexf2TXukb;B$TFI7#K!7F2(djXK}?S($IR@$HVvxZ3ngZ55?Wq>gf_
z-Z{(tFB53my|CC}<RC+M$d^O?^-T<Qvbdg*4)3OvvUv2)os`F~B;jka=y!;C+t|f+
zc6lin@AE4Xj1jYQbeQYV0olGOEr0&Y54;4&OB5}!?UFzJaEMEl@s4t-mmK6$ZT5eo
zT<RqUxl|E*luNxNaTx)`25D#L`2a@T!JdRy;So@LD}`gnfI!s&+;a@*&l?ANps46L
z4Z1;AuIfxFc+O1(&X{z%iqD`q?Bn|#@QtQ(wCg2%9JU`j&1)+RV9z~f`a!o+F5M~a
z+~xXIbsM=?6{WGey;eP2;YwF}ueX2J&W?TP)IDXbsVpevc6**>7I|DfFBLP<siFW_
zyST#gF+23{(t3)6dc3KyoQ}s{S9Ef=-t_v!V~9)V{9JJ>=$4<b-9y@kFX?k%h2N{~
zn=LWaVNb1m6695Aq6K;HTq?P$&TIKTXDh~DFgNwsLz26i9AiAX^i&&+k)waaQ+{-2
zFPfX)dy?34XEyfVH+t`JcIiN)ZDrz-(tXZ3q>ilcOrJ8d)XHs5!c2Aod0|y9N0yPg
zoqlfSo0K}@hBi7kiYr4icUMty*Hz7Fr!@7m9=RD^xJF7jq1m#&HjV5W^mgoIr-WCv
zkH`J=c!#x_fvB2F>yBoiX7+zD18HRr-#z5nwE^l(qz0KtE37gRbHhUE#!Tc15v}$5
z|H?#$00a7!4k7STUh#bYcOVlb>(-fw?b>FdVMwA9gaJ!TR@+4R`MKL78~0vhi^07Y
z)8MSbb5=a+tYf<xm$J_Fl%u@AA##%hb?R9p7^NPbU!@*3n?dSfJ-dI@b5-9#QsKGV
zNu{#qPUh%Qqy{VO0wN6)ahy5ldAJQZ-^HgXqi6BLsPB`&>#_}&LY2&hP=X0-lcdYl
zh5zCNS|jE2Qay&(f_&X?t!Dt5H+|w)@_J5B&}dV594fc-++1~Ul-jup`mIEUh{6yI
zX`QpQ5VkXj1>w<ki<^H-R@U^oe>bHIHwxbOp|ffc3>sTG<|GQcV47yd%l6f!B7Eks
zr%++}=7!oIbzyVEzUjg;vwMHT4B$zcYpPS&nPo^Jw`F8awxX1>vegm;xq%|rEul>t
zYC5HAy`s$eY|3P51^9iLs%GmknUJ)FfWJG+<D8|u6NpO_dOm+?Q-E7J*H(x<>L%L+
zcJD2(GP7{Xe9$egvR066+t<95(Y~@`^3mH+1%{AguMNS;(8##!GE`|pKGg4j8L9}H
zpf`{7eQqXkP%;I=D7#ITJUpA8?tx2CrBHalt9kbBHtZ?)U#qTi(~afW+UfG$ul$Zt
z!BY`*54rnR5(j@HoXBeJ|DZk}P`2~S+cGaL+%2WZRk+OI=)CJpUm(-GBL&i#qXp8H
zRJ>hGl(QoA6I=DnWwTBw4?_BGO|1SA?T=MXNV=ertJMu0c@;b2d|S9#Py;CO*QILE
zLXyZ;rmlf<#{R`Q>4dQj7Nryx`#2e=h@B=Fh?#ax=aGM0Rg*j~%lZ9r$>)$cTrasj
z1al3giS_QNQEs~n{18fjGlXRqWj9|`+<pOZQ&q5@(#(oqOy<~MrLJEr*xnSBj+{{|
zUHFW{Z}E*T2AYFp{TlAI*Eu19vQMaVx$>=F<BWmrx)whlzwZTHfS$17N4-`D1_?_Z
z0$k~Bi}!y3T_CkEetu8X1%whpzvkUh7bwzDJ@g9N%_I&=CTFs}`54p%pyPLsx<2sL
zmSb@1ey#VcX~Qk>Lw>%M#KF*(;n45?Ah<=aM7i?DdrW?QDJ&6QX*I$cTKDiOjU4Qg
z^6dWq2P9tkcBbdC39t;69MGLY{L0v7J#iF0ez0Kf&=r7IbzRf#8!uYB9uP&94gsP_
z8S}+N{0kPt`KtZx3gBao1;iv8M?g%+O|Q234(Actuqd<|gOzLDtB(RCN{0VIg`G(6
zgGUSql+cm>KseI8l)~SXJO2yGE&Csr0dEcy0Wp_hs6rE$D$xzif8Jhy|M>YyrX~FN
zPHu35yfu1>e`8TFmzjU<UkUk%JOeRBypG>7d}m(K@r-D>2bu7TpOC(DPqOLhr{8pq
z{OsXqh=jojM~Wk#l{PPl?j_A9r?~@1aV|Hu$#)(>su@w!pwLg}m4q?*?YQ1o6=X85
zYLjcFxDi#)wS3zif2CzVf^6a7I%0{u9#WATh&ao{U5#zhA?;1np|NM=+ir8<t*~dg
z_MRL4jPrtBK$Kiw3(e#eT5VZjS+7a!gw)@q^<HX9*yuqX6Xo~b;S#WWo8(!zqtq|H
zZ!{uM^7XHVZ`zWKS(1z@dNb2*)WSHEvC9?}M<nF*C_~J)e;?(%YK1KSPotd+Z-S|Z
z$4}j_&mT`;XRM${f$R(Z_2ntNKn+4WHGsAP9S(o}XZYjkm!52LCQ?WqBPjggZ-4zc
zbS?i(KGRFqBxzV>8e0d+nuLQ<Ae3_j4WCeeB;h%plQ(m)NEy8VoKfDx5LJ4Y`~s!l
zfdhFNbuo;Sf7gWd{_<DX9y|scmSGIE@`b}Zd-hIre9dHqA>C7ymsSH2h4TZk5l>z|
zB8^FdU0NH4MdUwmZA}ZWt<W?^fKeS$B^$O&XWn3Vi`wL%M>~4*rByg55Q+<m?mvj^
zEt~8p(#EvHME3j=JqeGxpK)c4O0rE6%=o*cMViFde_?)C)bBX|%6Y+-*0Zmc+4oMc
z72Xc--R$b1Q842m`?%tMPEZVLr`(L9>?rPA^hD7K=jJL=c85i5wdeQ{O+RWyzb!jq
zQoZ|%f^fH)s^u0iJ>^wfb;2v!=i{)!vRu*YLMZ!L{p}34opH}#+upsw9_ROA2HUZ}
zW3acvfBQItZ8?}3>>2Y)2HPR9V_wb*HnD!xiZ*ftOC02OdoQNF7tQoL!j>acQ{w7V
z5|vVbxBaZE8tH_U<~|+|n<YQUQkf*_2!3jb#)*9TT+`&&D($-g6i^|g>=;<Xgrhk$
zGj@-&m++{U=eiivzvh%Lu6#2TP2^-EX{RJKe?k3hB)l9FfaZJv+I--BzT*}t+luq7
z5!p|<8-!qm0_wR~l>!N?u<yMVH|x(602pc%pU?AWl^~`183C91vqo5JXPC-W@Rj1g
z*&PP|TIs_B!Ed?P1HbM4J-}~g1lk6_!*l`slmJmEnE5GPc|dBF(n`ia*y1^{izUjI
zf3%d?0zaX8E@#fjQIVuRb5>ed4(w9moGDE6xj3EU?+YZ2<$%itlE3{fhkx_SpSiR4
zZunp9`lU=GhyM=0hRl42Fh>TFFa|6(=*odTF<Zk(b)rDp$JpX5g@b$%qs95`CRvuJ
zr=D)CE39O~tWt_IVWjZV-=F<wM;_cgf4)_^<SJ?=_u8pnaIaXAe5-1tm)y!eEx(sj
zB>*qe7E=slg0<1kBCnf~-IEwwl~;GE(peG*jfcsS*S8o?Z>0wc{m6dXd#nrn_0Cb3
zAX5P^N8D73dpn9!jmzcdsPJ>1JSPj#vqYEeIxgotO51+at#w4r35(_opsynuf8n*+
zE+cN5^61Mor*q_|2=%gQuBVg5R4(jhS<8x3p{?iX0ma2)+&6DZI^Zr<F9scoZny18
z+9-A;`1n0Z6KgH6@D5LMn#4iQG)vgJ9kL)ixSsCu-ANM@P&n0TXLZ=LpRgLHJt4e^
zYC<`^Q0K~?X~aY^o)E(LoG307f0|~z0OqsH+34d{of$u>I$Nm{sB@Y5-YVNL)CpzT
zK7z5C2V}s@K+zIVccyvsnPjz8bKhqI|IU19&ZVia$cMqmHZ8ohA!TSy1vx&crStcH
z{E&v=@({&COgrRnKeQqLPUv)p`R}+D_QKBJ<?zmuIOyY<Ec)*hIcW);e|DGBx3*QG
zb^Ev)IQu0|k2(rF2i76e42P}VsBmE2AF?9lpuDtP%4L9f$cyn!xj>*wlvZ69h!$f<
z8yaVbj;Zu5*9~?K=am<VGs$zSRY6dk^sKM8iBt>!Sqj(Ow~9S8->$k|kq+XI98Q<<
z*Cwv5u4JT^StYHE#A(Y&e|>d-RNky09qvyn$*rx_Cdn&X9p3X(`B$sz>jLt=V|K++
zlp;GT7VujRA@2>3mGx}D>P#paLH0MC^|IfSwx4k=R8B=6Wt<>A74c_a&w*yq-Fmpy
z%?7Y}5Pbu0F;?4F`REJ2-*o~G`8n*@{bH&sDo3smaQ-H=nn5!1e|bNTS_#RCM9<^(
z@D&Fg3btrx6zaTN?W=#h8n#-wkD}X5#PX93eP-sNni)%iU!R2Ijl#ta)z|0%S~``t
z@bPe1u~n<R(hb`oqciTtdHi$c<9T9aYe(j7Uj-A;M#!p~+@&Ev@m|@PJVQz4RuYN)
z>O__Rq;$W>9!v#zf2i6F=s{m_Ign37&qe8N6p&-!qokTpdXZGn%65YZ*L_ToA#X5t
zv#3T8O4Fvqa9<lgO%Os3DFECF&D+$jaLLUvQJz+oL|3OuSGLY_QO{w_+I+@`VSAZ;
zF>pKBuG`op3bOC!Q8v`(T~K@6>tdtgZFM1sey59J;J3T1e@m4;f(L&_s#_8PBeeF>
zOKF1cAgyH}YXmzKYw$rDftq&!=;x;k!w5i4rq19SYi5N8ZuVNX7Xh9PBiRk+ox{wM
zZ=S|SQS4{ROu>2Gq^oV+AVC>QuiBgLSZoo>a3#e_|5zxy`9ny)10^IxTu^UI?GhO`
zq&bwfzEGZ;e=W%ZrLB4=*mBPxxwka6O;oPSP&zAcszaL%B%O|7@57Mfz@nR;J_R7v
zSXLR`Os#R~jjsU>swKVZnv6D-A*!1RaG}{+pL^F+IitI&nqtP0Smc*k@Vgo)Z2xxv
z5DBI0Gmmm*yY}i>=Ud?)1l;5QK>#>ranr>Z{q{;@e~#McxmEt!X6%{!?x1bq<R)_~
zQ;_gpY67`OcbGSEu#eC9?$*50AV=)jr<Fzu%dmZq%i91bP@8XXdAp~45Y4@;3(qH(
zp%UY*>p6l?vm0(x6~+m4wjZGmu0hjOYkuHmbC6oF&wt=$^N!;3Wpl6CYr6+uHgAdU
zTsGh0L?CO$*8MC8g2;b^d$=v(ool#zTpU>m&)Z(vy4+T3NoAm#+$#6uQk$8fvP)Sd
zM!^O?wMo&$<>LCbQ}k+k{6B*UIG6E<MH2!uF_-WbI~A8x+YQbE&bK1m4YmmbBTnF#
z(W^ld12Zx(mvNOd6_+xhGbVrSTg#Fww+-FTSLg?ZCJ69VH8oWu&6TUXi!(cySBdRS
z?jqjg_X9qgkM5Q<(sd;+FM6b5lOO<szyV0f!*2(9`2Okm@$}sp>0pJ^VAbK{*8}hi
z$$vp|qtOQoAPmCc<Llw47b%tD=g1F9iO7%T?~Xt9>obDm9>D(rygpZ3g3$lW&xzkd
z>KDh=zLA3FNAa7t_y+w;?j`v#+|$@Zm-_)X!s+$rj~||77C{Pc!q@OI{`ckM>EBPI
zK*peu)*b%w^(nj{1c6Q+zzC1#@cS=^U!Ru(=M5tYB_Ray@cQTX-wu~A=M5qt9@D4h
zb<nr7#-jvyJI=;(N(HO@DNsSRBYQzRY+(9*mvQF}8Gj~)?_ZcN=2xCypN$Nr9$F)Z
zkfpGR>o_8>dsMp0w%`MWEX|s^ZGHA;jb6yWuChXr31<D%&+_n3e)$6v3`X#u*!64Z
z4#KlY|2q7gZG3O@lNJuxe2}dgvFZSlG{IuXy*?|DDm42nWUwliULhzQUg=k`K{sIq
z!)8`Wu74^s5K~xQ9rG(&);wuVW7g;?595hI$6;mu$U_<8vZ9N^U0x*ps^WsVEY3m-
zI#S&9vO%uw<4jJ?<ifi4E2UWbGOZL7+FL1-C=}a*%uBVh+z`A0K5n@cJ3j+Xko9G`
zu{GkwaRogEd6i$(9p&b(z-?gbs$A%@P+2rRc1lBO;EHbA5L}qhXt==B_L8x-&AztB
zCEHUs8-R1HZfOK>lj@IXg@Ll%QVEUzLCdXGWObN7*fR6*tX$MMP>#3Zf$+(%C~o+*
zm(l1AAAgwP#9N*-Ezzh)?MZ;GYSi0mWKSiuqsOYTpFV0)_gd83TGSV{s3lfaQEp}p
z8Zfdno7`;1he!JesH&3v@Vxb4YWN(nlQu>e)_92rDtf;R=3;!y@h`BoMDgTqwF*7$
zrjM_0Coi8y!24RPybSm>@~5=l2=Hl!Ws*(mB7fkkmU$cK^MZljILjvLv%+Ycc=}eT
zj|6ELXInviH&F$~F`!4PlGec*J7_X~lK2||NyD?iq~9$dX`J9^?0p-+`QHMWTC#7n
z`mrEWONQr0{RJ`~2{OY<S!?bCndH1M3LXnG$@8?*{RJ||B@5anr5+42*<fOvx&ma9
zwSUoJJro@c^Z(z1Ok!{DyuSuzlF<SIAFRiI{y!DW{O+vXAsW9lYc)txUN~E8kg39$
zrD9tO%;nN#bfYpmahF43<5iF%QnlIryV4SafNLU*seQu^_O&0?c9|(37pDy}`Hn6S
zy9E9+5-y{00V3gT3E?^cyGun`l1|9FrxdV=bIOcE5wfTt0recyC^@+dw%?b5>J2vm
zqnEkr4MqV`mg@~G0mqjt>kS`&it?p(As5_7;`7|>x`GIGrHyUUJvo>{Be<Va8k3e*
z9|vc-BW=mBMyNYz$`{_!2$1WxV4Z+kRftN|#nL91rsf@#*;y(P1;!UfW$X6Zy;ib`
z+B2qB%ML4vd>z+{+S2fY>elyb`L?xu;4UuT_J%T&@wI+PP;gDY@>0fsuY0YKU=}TO
zc~Pg0B3F}(RSi}1zN5y2D&VWB#uKoc*ZVadDP+j>b@(H50ibL7G1q~-+^Yj2T48X>
zVmD%AMLZi-B{a-9jMUV7q8QIaF{|h@+ZTCOMi|>yO$vC*#-Z(Xd=p1U)FVZe4KSST
zhVN*0DiIL{I-$KaxCVNE80reY97f>;;v(;D8h*YCOocj)##+F|B?LcH-t5U(^>NvR
zHO{!uORmpe#^)`V{%f~TZYgLhLYz9Pg|l)2K&dWLBE)Sx#?9KemJB!9acIZeMG88E
z13!()o~G6Y`yd(4i)QWJBg;Cd2s>h$C*CkXHP}I(%{H-!n%ZZ7U4!g!i{{D-w!BaK
zU<n7|yF9NtVWkbDM%5L_n=ln^iaINx6mW=C6X5cH&`@xzBW*jfnPriTR^_dfB1lzT
z%PCt<4!xOv#xa>*TZL>=tf;}Vf0nTW3YY_Y)#o;tb%KCK{bsMawZqY~H<ErDwS@J(
zIVl{m(llD{&q?5ag|=oK6)L!7)d(T5uk2~l?uR~z4y*5~c4@($-8h@@T^^U#Z`Edr
zgi*@@)?WVCLL}VqD9@<;Y>~-3wLmQP>L6C6*nx+owZZj8%4|2>14l}kcIEOK#H##7
zTiv%&V>zJB`fTPd5Gc+WYtWaoumEFvib-ZwbaiZM)yFD-af&&+_8khc;6~~|`#kAz
z5<N?K%6U%G;<Q@(oH!Ma*C(3$oNxxoabS&;bPC|}p<~J9R;@Xo&29N}WX|lZo&%H?
zz({2n4Y-d-xHoDu%~>>?c2E-&HWC5Fe#_Dlk>6gmd+CZ@Lo`6LD@Cg*)oJTLiOwl|
zg0ZN!C#8~q9M27mTuOamloy_xhEaXqy-$IWsaNLKc@c$TJUfi)i-MEn>sApimQ4~=
zjPcgkr^-Hu`dvikqIOY*dw-Lft~@l1jzybtjGa%ii297MS^Q`U{>Grz)K*s^d^XRN
zAGX1u1dgtP+8)zoUUovARKogYYq<r4&~Oe#rVpckOa9z~ubk(b<`A#w*fsE0cR93a
z3Q>gvJ7&qtoF&6<Dg>LAk7=b{17*7a-V{J@y+BI>6;0PEcz<bi|9uK@Qh@W<pplBr
z&F%Lp@=b|hwyk>=#t1NK?2TXMlOh`~mhe8WX==Q3P>Xt>z5vn_9r*l|<Q>80BxqFu
zmdSvB8AXe?5>&fwkGBixweW<G6R=!M83b%k6ZeHxD#RMMDQvk;)1Lf9mtFRrS(4dX
z_;w7(IMIEaB0RFQqt7b>Ih8>L+pB&duX}(}&{>XW@HnMlFt<@9?xHFv+*gOmCQdVO
zEKWc>1`g`cUZAJ?2shAEjOV#{g+Fi%rt~C#<s%$I-M<f$HUHRP6!iv=fKHIJ+v0vJ
zmr!=_!&HF`fKblMe#sH&5U_Wa`;YJh$T^jLq%J%NL;H4dg+~~4vf>g7vLPF<g%o5X
z*L#H%faSgne!^J<w{38ww1sVo!|Hq2+FZ_$3*-!r>^e+V?;TDOa(2vaxMo1>Rr{NN
zB+jQM1LZA=(+{bJ&N#7=gxx&CxCP>{p*c6sxN&W?0-}MxwKL6Af5ml6;hYT;n(hLs
zTds+gW2$z^V>WnPTb$-Psj)078#L4dsIH|m$;^pFLAK_l3a=>+@BYATpt#BN`Tj#Q
zh9ZQM3^h}J>!bUn1GBvCyb>DRZP`v6$UL&L<n6{7{HXQA#9OwQW$ruQ0c1*WA7($7
z-S7<}0q>Ua4LeqOCxoSZpdavwUo@07;Nr|T5zW<tfoQpdin}<3$C~SoPCo4dz!FC)
z=%%_y-0V?Xa(8V1;&VH4?;B7d0gJY4+WYNKMwA}h^)){O`SS7fU+%B&K$ih;4if<~
zmmzi`6aq6gmq8f<CYRyy4V{0w4)Xoy&+lG%S^{TI`&0kZ{`c|u^4CkdAmol|RKqV{
zU;GUvfk`U{Fq}yZ!{5IRzrTFnk{u`G^bAA)htB@~V`#4YkT7+eFy&e!R+s`Opw*zY
zW76M>m@2{r2_I#um3?!yFiS;9ropyU_$5Bgv>{$V4luG<T_F}be$an1Nej&~Cg>H(
z8Aap^{W-}QNdk%_pf2gUzosjmN6H$Q<EEv|NppxupsY+>xgrhW*Tnleyo=a6WKPL;
zMulmXQ|9|lj7eF@Aha?}Nt3CeOsJfa)I4>OW>Il1{x?Wd$1$|FSIxaG1RQNdz$4*#
zk{^*T#SbGOgk#RwmS=zE63=+o%7xO*5q>R|%W?h?X_)FMcT(X@d2;zrn5DY%!wvqm
zWpuQ&fZ9C6NYMy}qSFQVGOPK*Xl9IQM|=M_TIhzktz+0h#y{q#c9xO2JEc8so>p?r
z5W7W?7SXXAND1Vy1oWawK|^-Vy)#FtnIaX>pnepJc|RXCA%%Z5Y^fbs87Rj<sg~ja
zr2~}Y#%xD%Ug-c0#WdOB<J}N1UC^w{*}zqHvJF?)4!uz)xn+jy_6c<=!k(Q%O9&|<
z{_ox80bYexjP$AF73TTOZlfOUU3K8=yu);n(a}=m3vR<nW?ZL1AxK_Q=#UdeA-J47
zd^&p(<Z{71%3gnJYLBp&n&KnurJ=ZDFAY9VvzG=rShtt@iT~1G8U$DDr5+vZr3mPI
z?4_o-W-kp++wA2~!kLq2Pz?3aT)rZ5B$ekxkc%>v$DE*kNvUv3=L1g8Ut6$|WKlIm
zexCI-=mgWG>4hFFB>CNwmNuAHjW)=q4aU+Fv{f@=-1vXgX?6XCQp~ym(J`a-e`!pd
zq~uIH^O7-f#S}uvr%O<r6No15HuByq2y#1b2#nBcCR?&IKMH#Yx7yj3ERi%BM91HJ
z&YLByEVJDF9}=!-UXyS=I+&kI{9V!>64O4?K1SNftv53-i979eX7GIEPBl62c9v~B
z)lpFzD#w3!s(GW*QGV!7RZ79Q?NB(71W8IBrL-Nzd09(J>(F(YAz|L>O8Ql{2~lnM
zL7|zoZfz7$S|qEY3f5(%#X?aQ?p$>a=ogCL+47&H1WR+<m;VffRh`R!nZzUo8|Jfg
z-qTc;tcHY3G=KV;55LmQ?`W`(p8tfQU&nd)@aKQ<qpvTn5T>2HCu~R&DdgWE%m=|7
zAs)$2Bv86g4A@NKPR=O*QxV2xvLtSYE!|KbUC4&OM<z);Q%W$^@TQa?r{%E@JeCaT
zs<<;K`6tPPr6(w;q_bFn7%81rMyqn}a--m@)T4)-0IXDT6wPi7S7bxG8--62ji%hI
zC~JSE+^f$o$D(HlxpxAxm+=)m6z!=UJQ@&R%``-u(v+fLu(?7!>hza;wXe~&?mnLK
znc$ViHbu5oaqPxzMIMr%lj2G8kQB}8jdP79VQmeJxZI1yZk|`L5VY|#@<~|u*sY<%
z0t+lSOgf@>dDK!1_aM6)ffZ!IZ>`mnkOhCu+_Wj4)v^-V$HXgZ_-1r>qp*T5r~N+S
zlhEa{Y<92S?nf67qPrV~+vrLc6QHM{E1jZrPd(a0?&o<03mY?z=U`!JcRZDnW;LzE
zLT`(p`*~l%M8~A#IhYvfZx*i)cl-AJe6Fk?V~~H&`n6!t*FGBiSn@vi;j)|m6<mK*
z4EWE%1%ZLGE4FeBm)-oY;38wVehw~*Gni@~g}qEz;6@yF^SgqBh=KJvI0(mt+)^Q~
z(<N2ZfO1h3FLwgAdtobpL5!8p0Sqh?Mr=Y1ABNcOm8~HL%sD(#vf51IzABw3Yi=hf
zOI5nI92kU)cCZYc(Q>YDQW#_)>KcDm+V_WoNv;K4gUDW`vTq<yAEDJC{Fj7(jx~aG
zW#xTA>c(5^=@h6bEg9btdP2^(WSX~JxX2Bfp_Czu_l3C>_oW(@84cPyUY&mjnD2J;
z_=N;YYAW}|yDF8$nLz8M!;}}YL>bHB-PNh+-a#NVIqMDr&tfGNF?BtT?~Q+K*X_83
zv$rixc2iFh*65Oe=DCx!7dGTW>j!;Xg71*^T={}q=$0ierA_^}mv}8z?iac8ooPZe
z#PiF6t!=b|cEHxSwvO&A4rIDFyJ&B^#`uOnNX5uame4J$A&*@fw_h1_GYH85^>z(-
z=F177e3`Vl1648uu<zXjWhQ^jSa|%q39vf6K04s&H<P%NGjhFJs^=8I!C2;m*lwHN
z)Z!sAd+Vfg*HAErTu+}-IiX8+%kf`qE?De`W({)R{&6VkZzgf4{u5cH9vAg*M^>Q4
z6mLhGxJX{@t^5zm26dgJo+R$297l*m!G9p(&*el(7fA&qH)%%7xp9AhC+BICZjF@7
z1S3&w+~CA#ie7n~8n0(pe14bwZE92jUcb>DoAAR+kxVDP`+XOXXE+tj8a#eTBAuT{
zp5D>CWnN{q`jMB@)ZU>!MjTW<2Tm<oi4t~l!a3f`M62ny%;Kl;r>3x_Vm|&V45#Go
zEK(~^&e%`>4zzLzS8jhI7oS;A1q<pZ`nl3982MWWvOy2VXR{Z&7TR=~J{0lUov>kK
z!Yzl!*hAZ_UAx$>@fx0JR*U5evrAgVCC|a%iq~jn<d0?*X;#lp#jk6TA8Cw81+9)}
z*J89CbtuoXuC=>zDoxHgx~ekzE5|Iz!zmST#b*18gP+JD$xMHZM_I$Nw2KYKrXJW3
zFtMJ0q5BoA|1zNAwdZ3;vsUb?t>`YbYJu?BRzt(rIxE}rg*?+qW)<`;ecN$Nt)ex-
zg+=79oqF(H%b0qfck2kT^)XU$g-_P|q?ek~>cw>J=Lh=HEHKK@nhEkJZ>7$yeLR)E
zz-eHg*9<F|-?uL2+3fVO-H5VSsqt*kp7cRpJ?WD6Xib}B(F{Id1coLTk8k*Df61x-
z3*sHr?3V#=4if@0F_+=MB@>r==_x3eZTk(Nf7*X}`?%AY1&}*T?k{h<e|B>wSflbT
zbiaSw#Rfy5$YTf2DInW@|FiqL`?Zi5NNeTell;D3plO{|9$lpQ`HSxUkj9VFnS$^B
zO<lj0adh{0_nQn*BkYw<gh>z&w+QPFRq!E`9Z4_+rA=5*V&TjzD(i1k1+4<l$tFGX
ze;|6Mtz|yZGtn<(656bS5YKuRrEf%USTxD%s7XooG+gm5%3I@dKhaB;79+hlzLWn(
z^psxuFO43)&@aPN($taCQ>C|=jt$c<IDL_{a)RH9CKk!GLRgUugK`KPiH2Ee9g1io
z)B>G}Y=4l;_i(ctfQbTNTnI?>%qxf(f32LwM<c?DwG-k-X&wNHddFhH_^AhjBQTf&
zirRF{dI?!CebMO;OSNyNYiI5JG(4d`(O6%!c9}SuVXKG(gYud-5(kmeq4=nCECp-t
zpy09;i{?33umG>TF&kllC_W9(u;2+68b%$lO%`>kg)PW7BTzv$`ex4?AuHw(e;nP^
zvMI8U@mJpI)#x^(a2s6?R)yOU&mHJ$tq{y&y=_OA??bm8g$la94}dm7mwgadUOd)1
zErgj@S%_0JKNXx@AB}8-lk*CLU4;|hiBmH_6`X8}QMQ4}Iz{9k?Qol?TtR~S@L(G#
z*eY`2rrnxCVb{fD?Q^eSV)`&)e;Z5;D^o&{5}LMXMDsK&>{IW{w_%@*GPV|uUUOy3
zaA}@@1sCkS@HV(0$sH?8<vv`R=U>4^_x^JmTmTgty0BhGEO;dj&GW6{U=><>Pi<;7
ziDl)uNLJgXa@@S`x1cI&oXl{{IAe9QuK)~{%@{S95dXRNLuFuSW8Pz`e;>u%T<7Ql
z`#J%Kp{L7*lo)4BzNPVd9%>B7Tz3#!`gmo`+&57)V{I+ZwvSb^qMmV$o|F}l&A7%e
zsp;Cj&WoIDzP_)4qzW&NHDGdiq|5jcqatw~vyn{5`0{mRVyu?RQn}&+AM#R>a4xEi
z++Jt@F4l?-kpctrS}{s%f8e21Jfpf~t&8Ha=|lz+E|dT>F-4i3LodQ~&vi91j9NeU
z?~@2!uD1z;`j8kr8$s5Y0XGd{r?5n!B!zQhI2+sJyUIwHnO;<YCrJhv&I<725F9{T
z{oTX`Z}9Q;cM~@Y*xX~|R+CsZGYXKDaFZ%81y<nczQrAvU4O-Gf1PBM>^|A5tV&Vc
z=!*KGovL1YM%4~e1%B^U(C!|^SWRNtT8Lzw{=b|bF;BMyOKIB;##!9S;C6Iw=RMdQ
z^YyN0ut>$AQ!bnz#v=K|brn}S*reE=oM=|2lW!0C-Jwt~G$KXqX3-_)?Z&6xskFKd
zLaUrP+hw#WjGK(jf1ZP4ksItayPj<K>Wzp;*miV50r*GSb__vT1CO`u7_2f(_blPn
zBo@w0JfsO`8{3Z2KKPP(o#;L+&nnvISmRw1AbnR>v6{p}MHI4ho9qJ(D;mk#iU*cZ
z8Ei<#tC+D#=D>LYv%4G<yK1&jOzL1BBE^?tcHJ<tKR4;of51NPNe>3u^_k3I+Ia{b
ziy<mw9VZUJ7w2#-yLrk+%v7?QftnxOke_cuBl{>MWumM}*2MQH^|QC1Uaw|B6r2+K
zdn%+9yFj0zWK^c7Q?5zeh;>BCZ%&C_@T7KPxsMNH#jsUu&Uh&rHE|mgXB{QMheiCc
zCgmjwIUS(Yf3HID%HUZ_Mm{;ldmSr^*S5eli7KPJ@h3rk^5)a)IB!A)Xy=pxW+#`1
z-d5OwdGF}ZYx@sZQa}U=(Ac)fbq$wH$q>#)yT&Cu=|%J2_i(9=vca94!r)UIv|U1L
zWdVUb3-zQmj8%elK@T_~g7o;H2dq~>e2;rwO=989e?%$3*=>Rzu%fY+e)bb(?HE>M
zW2{1O8_7oe0&BC)k+SxgR}dkNUUr*J01$5$UGh(OrmP*eiagLr5Vg@s06?eMyGL(v
zx6iqP1&A}=Xx|P%0u%H<_K~t?04lg@Ou^YUz-ozwZ20j{l(nO98{GgB*Kc$PAsFT0
z$3ITiE)2nKWPKm|Y=W#O<@n*JZJTFYITG3j8k+#&P`Q9x9sOyZ`B~aAyZ;9z1a7#O
z0dEcy128!^mvOTsB$gN~3IQ;e0SGWCe_3sBBexO$?q9)=Nk4Eo!<Qk8z`!L}rvX~v
zaJ~X<4I|L`wuyR*WJv4c^6&c$SCXZyFM9RC`CyB?J0yoQ&&)it>s=w^3_CVH8}AC6
z?DW?gV%&2%8%n`=a)c|VgJc5!Ad^ktIw&T_qK~{u#b00$jOZhB(MZZVa3+&Zf5*QP
z^bxVagep`p65WK7Au56qJ3}7p;T}}+RG&hZ5CrQJW`taqi-w|VN{o@zEj$^@RQCuN
zg{V)lTami5H%v}-Vlm8tbd)g?JpvcQo(X$+n8hd6fP>-SbQO=MvCarsB?Vp2M~p;w
z!(liF)zpg`OLgg&#{;Lb)k*={f3mAvKnvDOQXRdhQo3fSCJXo%)zYMEM!lqnHTy$G
z^;o|J7R_E2jgTu@*Nf^TVeApb1nOtNUv(Nqk5N%Y;o9J%?tv{sNZczKLQ%bwGa*%-
z6Pb`%S4mj8U?V2nDp|b&BSpt5CS)SjsS-`(>Pf5_eYLkhF4W_~*|GPZe>h;L`VZ!h
zDXLtCoK(=ktWk2n9&lYORG*Sbt{wv?O)jfnV7a93#Jyw^S8o8h$)`FhCIzSK6^Btp
zAEu>{bp{}xQYhF17D?ju90gBBbPpURWknyu)~V>qtmt;t%q1Bi((~e|(jy3AAwdD4
zK_xLM+%yW*;|xUOWa=vfe+?odsu(6@#v+W6kivB;7A_SYa9ws%pOS`9RYfq;V=z#j
zSyf?#MD{sfeDh87`Udtc2<x}aKmYY_#8$Eyh7jyAeV9ycFE5{z5=g;ox!^bNAql8i
zg~M)1;KC}zA0<um-E1<OU*C`JTH|$sY2J*wuANT}_m_Y9)O~;5e~r4<P=DiUHtn#-
zl}Av!9O^DtKm+2__ZL79X$02`_!{Ao_2Tlfc{88gUANs$^XBzc^RE5WEj51k>wSxz
ze`=x7v}>o`LK9L|p?TXbW)JhbcG2fAz4iC@!+7-V?9+{2rruw-m_Nr%&6d3)ucp&k
zzfkJ8zqqOCQjby7e<cRD>)oq#bN%qX)2qLYryou8?QH&`o!4F5ZS&XWhvvH*>M!25
zcij!Lm@QdDwpb5uAwFyY{6fsO#B$rbGFbk)`F{3p*1R@fA<dKw*C2NkB!QH?|N8QB
z!vx1E5v+nw-7E-DBsE(!nUkne10@I%;*cps5Ohf#g2t`2e*ko}#;l+N#1HYhDX_v;
z$gl(euv-#|*ZN4$j=~M#Z9(!-6cP<lSoTNZqb<no$OX@Chl7tUgQFvd9ua`-)&ZzQ
z;w*6|P=%wjjH0NJ_o(q43s5+Oe0E6rPEgl^%fy7zzJ+*7I}pDiu);AohRw*bUlhj(
zZ`8tNYE4R}e|pp63AmH<vV0Qt37w36P6GgNI0gWoHUJR%46I~uIXDv}3Q};2!A4jU
zwcC1DQo%MkKY#@IV_`1=$Y5Vk(CIvxSilm&j%x;HCd;)Ph>gQMoD!072K5CqDW0=;
z=TM<H14yV<VhciPBS<sYA&LVHBO~OJO7h|oC>}?!f9|cw;J62Y*_P_VBW0W!%x<?N
z7_CM5*h3Oz5@FBJ5pr9c<y`qQxIy^%d`$QR{}?_#E~Cb$)j*yIpQoXMvlv=J1<jOI
z&I+!g=I88erSU)vGXL$x7V?@`>(%eN@oWkR{p0Nq`uEk(U3b6ubJP4Z?tXrFZ|`Qm
zH2;~me;?-KkL|~~&W1N&e*51Ekah7}eD9&BQVZ=Pmx=}g3it>`yN7NzACD%BW-^|(
zqj_^TnoQo0?mjL)quYV8hbF-~BxQEdg5$xWWVNwJcti1FRa&%FL(9d|R?zvfqa67m
zxev~R?a3aJh%CDmm@k`ug<%Z};TZlE_DW^pe~Gl?BUr_=SZvEkTjm3sp|GsKS2!>(
z<YZgMA(stDaoOlTkI*jwr_UMZV0*-YJ)%F#9$_DBZAmTreDBb`mpvp=IxHCMvm1IZ
z;GI+!)6nW^AFET6?XsBiV_A%4Sq+n04d|4zNRn(Qaa|TECrg}PURhizx+;qYD>^8P
zf0O$k7wvR0?#BNccfX#gHbw?DhPLR?96C1-e(GG$PjiwXIS%5%i9L8Z2r46>NVaNZ
z3Oj0K8io+Fv*P4s<f@YrxeQ;8i905QBtC{v;?vtOcG~(Rlk{C9x?}G^ZnP~^E)?i+
zKMU=XS!MX1DFwuX_&?0s&OB@uBp>6_f4tEw$f`#WIYWz`(9VK6Mzqu!K4$zz@P{a9
zv12~Wx5sONOp^3N2o*G#=v%<f9HHq}8ZGfDErN><=#D-|b|Lz0wk^BZwmk8AzM#7;
z<;!-W=VCp1_*yT={gL))n4I>B&7AgqIq-y})(6SYZs)^!_@PEGeqi)M#~8ive-5~A
z2&}&mfjfHPKYrn3H|$_aIU-By*TZ#|EUPS8WHrdAa7W}D8;I82F%VKz2rrL2t`uEy
z$Ac9eaK~T9_dkAk7){Pb9DRuOH9IFX;ZDDui}kQE3sf)Vb1zl?1iWv(<7emgNLJfY
z%Qq|fkztUmo|5#={pL;Pi>4gBUi^-=5ls3``-os7UI!ERA6rjkk+IaS7$qCd*H8Q&
z*T8_c;={1Sb0RJ7@TZ`e2qgu-?fd-uIdfCs@%v7;N#d7%>vKY89Db~0ibpxKJUX-g
zSf@9u>bL&~V?REZk!}bR0y8p~5PS_4ms-avCx5kD%WmYh65aP#=m*SUkyU&EjYixa
z+dy`c>s^pVFv%o~@x@KD$nW3cQ<PdQspZxH8nr00San#emuT7j(aG-haDF>{dePWf
z;Rvki-oACfFC_mDk{b<m7C;yc-P?Qj^HEA=_&W22QfkeY<=>9)bb8R>xCihP!0Evv
zgn#}od_D1dNd4lt+BZ@lz7)TCjBm6*=3bI7!##~nbh#g3xSxJ``|%*N2*7t}?$2+B
zzYb#|j6~tQ?Y_Pr!VA<Ov{RiW0pPm7fA78>zElc@vJ@`-;{RU<B%P3fq%nMtB!B)z
zcE9n<AGx#kuKPE3eGlEG@Z`Ji-G7+AY=1#_z@STGhz6ho36hqHd!fUt1UfyQ9*k5+
z=`7@55_E#0VR9fkX)h%D5AeAlKQVaH?Ib@rreuoz7*yjqEkDYD00b;##UiMMF4sz2
zHuJxL3!(s~zYi{G1!{YN3*8Zy&HOLmqO|}IJK>^@K<g&<OOFNJh{I-n7jRHYD1WVX
z!a?yJeAN#&&dA1>*k}{Wj;s$9+V~mI6MIS`<q?p3^D{lIL~OHU3y84?QdaMT7@#0?
z6JT^Vz%~oE02tqu!+E+&cvr|yuy!knCC-W;W)2<AkuAPijtGyvb6i_;ofG7((Nm!d
zSq@#QvD7ZDE4sKkpI*G29ITwMEPvK;F#5dAtOTfriV~C~EIYcaEH3pS->XFE8-635
zqSn0R?k4jxAA=ETRurs~$V`}Zj$9`Ek}lT02d(t6*I82eGL5D_hF*Bg44)$^rDc-s
z0nd4cFqTvuM<*3!bGdJzajx8Ce$l2D&}UWN$I*|O%P`Y(nFTZS=}nziC4ae7OB@}R
zK(<KK(e4EHm=_LHueh_bG_vnWnX0WVPlKnt#BsS*o^ObUtej>fs`OD^SZ2)QGe)Di
zAZo+K>Lc%n-lzxe%M4+*!P>~gkMcTWnq&xT99rW{SbvpOqJ&WhCotohDP#?cvQqmN
z%j~{Vo3?~?)_(nSOYU2@%zxtr{edQ)9^g2ojVaqeps+LS35-7`wQMGKMOk|_Nz>M;
znr)C<#xR?H!V$!E41UgYnY6UIH2{l1bidn*UUZt>z)>kD6+v-U5Uf>B#qw-1QPL%)
zDtxQ5QGA)bQP;F3dYO~EQsjc+6&UQdCf)rVEcr$2Pp`$RUK%VXg#kO}bYFk#Sv<lu
zE2^C3NZ(H}6&DLK6NM?2+Te4Bh4YYr1K(apPm{SOIxkEL9NSz|@mo(N*Bnt@6R$A(
zgdIsEysCaFGOHWrWxjvT=8aL4=iW3GEeH_Gr=gbR8PF4R<=T*FprfkjcfXH{&M(X~
zE%d4c9#IL>`T|rfEWoV^d?tTsv|z5c#3J+I7n0`*+(*K%n@OzX44v@u?zt&Pv1}HQ
z6emo5xCN=K6ZA!!T;h32VOT60J*74sw_3V1H8!?)kzvp4D=Mc<t8OJKSMt^}$#!U#
zAm`nr;6uuAD~Xl*&t&a9*|o9woQ<5AShAwfcq;?jHElj@ACYlKRTF=@YT#3-np6R0
z*Z&uN+@R&B*Pdm~tTMF`RaUQI2{QLa_I{eiz3=D!VqjZh?0JLBn*TgwbeA#?ZIv)o
zGY)mGu~~_1Q{HkV3}8<SoUe?iOf1G#K1d2{#{?2;9KWE-fo%RY>&i>78t-Hi5<`FV
zC5kH|*8VZ#Ya+7<uC0I6M!|J%g`5&$O?sN(x`xuO4}Ie@Mo#P~iK=EUQ-~v?^e|4`
zuuVUDMQ9~YT&atZHBO`)U_a$hyb0wSN+rHHpR_E=9@7Ik8_-Whvux2hPV$3rOf}NV
z-Urlpy$#P&vc$%GOqe5cVPYBTuZ|V|wToJRh^CF$w#8-vn*V?Cwy*{*1iMGZnP3eL
zpZ@p9w;#U#3WA6I6S}gIVh;yw7~j()5o@fJmd#>%cI^2aG7IcOw84%=1GW;j4y+Ad
zy90742%4knzV&-(^#UC#)eMfN(Q?N+LDN)t2o!5;neThsSh(t^h%!|+!c(A<$A(89
zUq*$;aJA2+E02Hrvpan24qq~=WQ?)F<v7KLkQdlG#3@ukAXp78vZYD0=TyKmWmLRn
zrMS}C%3B-RW;lZ!k%Ur?1pZZLt1_3CM()zYv<ijNRPN*UD>#f~<(A&A3F6!Ujj$M9
zFJyER!u{{B?|IxXI4*P^fJ0y8=#}>+>#m8kN=a<eBBy`urFRL^%BIqJiw^yIySYXL
zc)BlL!%mXxp%?7SYA`r{B*GP7jQwo`f~4qh3J6!6N|6Cr3ski{3QwA%1v;e++V+JA
zO*rKV`ZC(xGH0s=fgbaA-`F9Q!;b4SXKR_MF;CH3ndotfJr16Ps6i@v(bz(zZelB%
zSl>Z@_t<~-jy~)q%Q0Aok<Yys_9&5c+K%PyInTH*<l025Kt<z(k$ADvPpO!eC4Cx|
zITJcpz3<S&7Cq0qqK%%e6x?z}$fP)-6j=r<kfNAeMe8d`u0}n1jZ;g`;38hSFYWd{
z+A?bF;TmL1i8}Sf`fE^&7|d04#$AmR=8RfzKBj++T5q6|QTuQFe_hfeyNBR~0Jr2j
zJtsM|dFW%C6}gbIAi(Ki;drPA`&DcV?<sXDwG9r~Y^&HPjsaMv`&Vo@RuR(ak0j|g
zlUT`_LN*e#PsK)YG=!iApOw;Qh+IY)ogc_0mD|25hY|9*mSfz2(&IvJZ0Cl{6A!CO
zJKKLTTv!{>0*u*dZP-2_)xE1rMU2u{hWa7O7G$Wca4aFRnezoiq!Ka7+2c#SVk3<n
z2i8oLFEH7%mCs?=<c(0i-BhjqyjK4%&({p!*K;Oq!4Zp=uP;7s$8CN+z_T*pPG_{h
zqaeL>M(gJDYTJ2Uz`}7*=yp1n#a>pBoj-q<6{56)h1q0NTP@sz>}CWOkPU|wcRHQ}
zp#TVOR>@jm@8huAV`FQ<-3r12w&5V8-RaaE7!Zd1YOUQ4E`1Q(%_uCOizxuw2i;J@
z-hL9V^(ZS*+APfiN;-ui`=G?khl<hW19%&7+APlkPAUZ```{FhrK!7n+s!;LU=e?g
zRchJ^3*dZE>4j6!FyU?2VX~R`1x#cL686DF8qSgN9)B^gozHpseRAcyh#$_ZNZRx7
z1vYa(4~k1Jco$F{3Xf*bKNi@`{X8f(c+OoQv08X{*S{0k$m_xhZX3@vedGZ5jTBa%
z-wxjc)Lo8@jPCyjoL%x+3T19&b99%e(J~AHGnWD19u${i6b{9I!255<FFRe1pjF`b
z7_0ez@5kMLyZM2bi~`gB^|_1fXhB(TIwPHdx}Se{KX!l2jZV?>AZTf3q7$@03xwD)
z2W4YdMvLTJBnKfu2?Yf?LayJW_VFJnFS$LAMVH%AP}8sJoM$VLS5~%FR){kw&cHki
z<BD8X#9YVz*j8D8>GGjLYM9iQG1aIr_%l$g68iL@_7=vQ^JzIR&lgM@QrbW?`SqLb
zzDws98Ej>hUuo#)u#E2h?f#6nRx#}5-lAbp-r8G+b!SxYp_m;_kPS*>SZ`wC&f*n2
zylqy-1e)8u$xKj&tQbcHYz99L6My9FqlSvV`pl!`oo4iZ-{|Y`mP_<nOuX7hH>X0S
zNH=?heo;C`>g}IcCxOuM$wUXoLc<ZG6S58HH^WqcKQhnLUBp;xVxccc&7NC);oRx2
z%jNmP#8xrE+^>OEuza^zl89w6{+g5}xw1t>4~1@&sfjd^yvQ@5Cvp+hyXB1JfZ9G%
z&B(GMvX57PGqP4$zKgW0O)T7*F$y(rqh=)0fdlr`NqK=S1!BESO4|2WIXeV};O@GW
zwI&u4QOp|fkC2iD$s>AJfRmuJcTwq63Ir1mj*&y!^hZn#-E;Bfp*LM-p~U%GB!0<$
zzE+AlP*lYf3s-jQNX4ePg*>zDh-hzOyh<t*Ml~XT*C(?n=Dp<Dc0LC~Q8L=+K0QXw
zK2j!J;uyH3VD;1%k(CLsQErP849K|40B5<rlCz?u+{A;>CsC)72Jz_I<)#KnNjZyQ
z&M6p!*g82!@gfKt$T`NyqwkXQ)g~71OiMnF8_7A!BGwLaZu{3H=(EHVfwz6eh`rBn
zCe`47N1Q#;zUP@Z<e2(>$Tt0nr9)*jPS=L%3`#Q=g`0K^Z{Lj#sI+(6+>H%RSvS1W
zrtIFA46MAyB4-`U$k3>`RYcl%UO~iLdCB*WMuZe;C&pPqGoQ%WsvLD-_1jYZ&DeI<
zPlx_#`aadOe0f6Dj$7qEoRPQBu+e=;{6-po0@FE`g0*)r;Ib5p<~>)iU~w$i2nx<0
zV&+iL1PTo!MYu^)rzC6vwiSR1uB1h<%tpXUmMQzc$+0Q0kBL{_>DAyiqfkLt?3-XW
zLKh`LT+Z5daOqBPn?b06YjXax4Y-ntvVn)oWAmOXSU}D<w!y+^#cXdfZHfi244=(^
zd#_-kbH=d^CORmmhva)mC0aPqywBU`Hz7y(P0nwER}QJ-AJ^1p5iaffui)Z`-(5Dq
z#XIFZSA69@T$=Y^!Nuihy$vpom7jbEIb4V*v=WEr{Z??`6j!%_fk}Bj`Fg)Rw@Gi%
zbowPQeNB5^2w3aEDtJ*2mD>PDsDch8s{o^i0oHu73S#IL-}h_kY7@)8(M7Xp?L&Sv
zo7MjVb;hX!mjQ1M69O_cmr;`!6Sr#_4#f!!lplyczJ2;X-}l`dmjQ1M69F=pVVEWq
zmt-Rj&ws+_DxClM@wX>QB-QUmhc-glcyb;DVoP+qN;~XotBn>O88NQTR02{Bbe7OI
z1bvbM@Dq$Rr@s)+`t&yj*-Bx(TOpr~#BV^pSR~ORg&<veg0xxW6!`Qcan>Rh)|Rz^
za)6(C{f-W;w9pdf(=voW-<m<gv*10QlcbPf=YO}VtjupO^tY?n05rT5K}88cf|!1{
z6<UG>vF&;-@8P}B3$g6&r{hkYG$XHy70+|lk92~i-I+h0Fw-16zi`hVbFb1+p@fF1
zt56z*M3>82P<Op_if~<$#G*+~3yE4$B_N3-zVo5-7G6-hZR04>G^_}OQ?eMFEF<d-
zaepkseIY7=&T_>XEhL-K@5Dz<tAExL8x@WF>5V7NO!A}i7@ue-%)2rLu4`N2c6&o|
zNRZ%Cyud_#*3Nq3O*+7^X;5a)SlW{{b-><@yW?#V*h(9%T<LCago8eij#7IsRDG5~
zpCZkzK5JP0D4!Jq%#d~U50WSh%o0hG0DsAc>8e$ZEQjP>t@Frx87pF3&du;$h23lu
zs=X(hvXR-oBYF9{>Z<=MtEyF3Px0DL38aIs<h|ZE-mzf!N+-MwXX^&(h=+l%C;8En
zm76mcny;N9Y;{!yp`q!juC#C+RpoDfl2QDm67#d&sGB{P3O9JJ<h4K>uE`{SI)8Kc
ztNZ3LrqEl4k-ZEMIz^^p*tI!JiTj#{eTxpR%jAkR+Zcq=w_5YKnp-ns;HEg4BQMj#
zA)+XIRr4z+-Oiuq{mpYazBXH1AZV5#JB}rf!?dSHhD(Rq-*%a}$p+~C5)B%KO(W|m
z%?7hX8wAHAaC?fD?Z~oQ>@*|YMSmmpAdQ*c8~D7lMH`$jE;LV9v4{n>_Ie`&dV1iK
zGS@dbl0>V(jQyD;77A|)tg1)JMhN<0F4*=QSl+n5a+~%zrl*GRu3U2!M|HDRcFU$y
zDQBV6xU~A3Nzb)c&)u(PPFzOI@6eJTuTa|a&-0C~)G<2;)ogsot(4!|qJLj9C~oOP
z_M}=Krinjft>mYEdAH#d1!}Z(yUNc3h-jKlmbxz0T^PfWJIM?fv%0aQ5-n86bZAb5
z!so@OO|Da}W!UKYuLN~s+6zneiVJ_Ki92&*EupMVX$UTz*(IgI7swzfjCH}`w~JDj
zv}*f?rB)i02+GYA>KZcr9)G7K0C>xEZ4)OSvDk<+)Ebz+gKt1(P<p|)^8AGbSY6XG
zuCU#WbG5X*<ZxWcCs_gMOHey~F#+k0W~F=J){(Y>v}lfM+^x*3ZW_JgH7GB<4hOtm
zG#SrO9v-2|t!Q$cS3E7L=H7y<n}YyPi&zrJ0?C2RZj-+(OW8vlOMkl_mJIgrn-W5O
zsR4aX{YGAnI#*MN%A2aU8y8AJhYHuGkIiEwY~IMGqDzXAzJ8tl`uXwnpG6S>eLtq>
zr+@wVlwN>Q!e9aBWR3`b{eJr6>F+bt+S(FQCR0s6@V9KY!;{~rqZcKrxpdCj!Ko^&
z<}92bXkjrQtx@N|xqs22WO6GsgBc6HnxN^o0x@^%46*1s+E;P*O1Pm8taVe*?{2~x
z(6t#m!qXQ9%XC<ez#HZ2Ks^%k5u)LhT6G)pari?Nu6M5<f&U26xSm%wC*`JG&pnPF
z6VPeHBk36-8qacczjF}{E@odr*yx?vr9KAhzlh(v`Y)$fW`C%+NCPa<pjp=qn)3_2
zFERy$TaqyokkzhRI<#=GxOvvq8(h7=T4^1GnAMp(mYef_h-=~U+bNGlPc7q2x3On{
z=Of;;USH*Zq53-i=LEx~ND<Y1eoN-=3_L1$p(D(JM+N-;{{$Xmy@<9O-|Z|pHo_<r
ztdKvcJV197`G4GWa3jKymocBy$W62l15U(fkK6hrBuE^pKQiNnCquK<2sHW`vX=Wc
zky9dK);X;<o9l%{#bJoM9X~D6(P5=Hop{jnN*7RZS!~+zW0h^+dbxANaLhL_S~P~c
z-Z22p?Ek%sG`&m2@9tda6tsGelwu98nr07AxPU4q+J6e^dkJJ%*O@2m_S5n~IVY4`
zpdI3iNwt%T{N{JCS0I!MkHUV0b_4d8?_n=XFO_3odo6F9awi^B;8DS`0G}XUr@+Xb
zE0?D_@4MfibQIBt83P65efB*NF&Ba07l?ZoqO)x-92oTLj<c=BZ~!wjP+zCUM&S|C
zOpX0IM1L0qeioVk{5nMMux?v9p5_uv`g|R7G{VP)pYm;=D@ZMH<yM>Tm!(lQ#o8P7
z$Y>+1x2qoAe{QmJ^FS3!DU=LwMTPomENRWE9$#P}76&#ylo9|Dz~4$q%8R5vpk#Kc
zI%?4e6t0A530H&a0v}UM4f@j(>5e=paU#w@)ql*|J7NnZkrPC1SR2+YbzzSqfRNR^
zn>TPRix|5bnVP@wI=FOY&g%Jw6y^@660E?_LaIE>Gc}f>n5i#2-f2k7mEo*B+LfMN
z3OT8)&+pT-SVqLO-)jd3VjawbwPOVO2<;dl-m4ws?$YjL23A|AhnlccgRC82)dU9u
z!++C4)+d@kp>Ez$r3*#ssnXDvG=l1_Jl3cC@Si!*1Pu7*N-O1lO&EbbLK8-a_iDnp
z`+D|DS`pbDav)dSG@@w2=o?D9igouFyabYN7F2UWqn4y*@f72(rpEe9Rk2YC+JeiZ
zqhCleX5xS@*ksM#4rVz>RUrs7b?3gXs(&NUN2uxu@m^IOcOR<Y>v#1{D*<k$A3oK$
zCJN&bo_nw;Mxc)n#R&0UQH;9}MR9qQC?@*%qWD%vzKTz~d>>XRvuqNi=1=PGQ|H#B
z#)<|)7#f6SD(R*+BSBFyRXYE}F5}m(3R34VD1CfYP@CngL2p@OUe;ALDLEKkxqo&&
zpp4B8m8#%&9FSwv6q=9S^|8Xh2mMVRR<1kN?rOaTuza9TzmAt-{kp(h&N}yLR6AH(
z)+uyD+j${Pp*4hc{?*XTPVJk~+-Sd=&aGEhY_go1r)O#~HM!f<{`Z0IMJ%fpSeJ8T
zn+a@!icR6Qn_NFIX1SU-3(oxVyno(t0KZzdEF<w-KE)_(kBR1-|4)Oa%9|kD4WAoF
z60`<BU8U`GwV!9g3RV;U%NsK*N|^yGw@Je)K9g9^f$8S4Nz{&$xrEg;yg^T0yS3_5
zS}%QuOV>HGOy>rNOP4~&syd4H{}o}J#hV8^2zsBAtg!0D^`+r!n1?mv?0@FCr{mbh
zB|Mw13@v)i`Vc%iEz9qFvhpOWjGmA1#{It_F#C?~c^yselUhZ*u$?8zN_%16!&i7s
zkU&Nom&?Z3W<M;078=gCxPmxblCq0!1_#DL$aZDg`K0*TX3Wt~SY+5?;42@&js70G
zuT!@ZySKvi0_TWswcs|+iZG3nQKmZ*cg2xRm(s`Q?)a?+eyfGwec~x}2z_*5H%)ZE
z8LtT^egFJ;`Y%8VJK~oCZw?ayGMAx}AQS>NIg{b2Czqfu4yS)mp}*JPK7V=PX-JfQ
zV--UxGyQ<{Yed&bv(~RN(4g>|!tdfeB25sqVS?+zf!k}OHH;Bt7;&IXDT3L5PA%eH
z9JR9Himy=cvz?$Jm$dDVfTWOfnh!Yq5h-1godnLn0;Qbmji-QlHyLk`u(}00GVOta
z$U{5_VghxMVv&ESRe9{8wbG-jykPkL8KugVfCy>K_hG#1QIO2gZ{$wmb-32I3jEV=
z1OL=*Ohu7<B?)a(Tb21;(HO^)X-%VfB>9?6NoX#Rr|zI6G<QrmIg!LNhi3IwDJGO3
zTkb0NqCN&oi+#W#*JvMV)Ne*H`Ke03MbToHrI32D`I&#{Csc34{E@HZeuFJ2sdAfp
zuVupY--q_)?mc}Ly(eJa1ZNOVKa+1CpD+KJFiLXItaaV*KVHHO1%d%`UrEW7Fx_AO
z>b}4HgTf|v!x_hCm=6E&(_epf<jS|?Wj#kpb7QowysY3Te$^Q%m_}ttvQmDesfKdC
z@^?QyoS1(_Rvx84MGmO9Y>gv12Md)+){oTlgIX{C;j{c+>zhP2q5>Nni?)NdW@)Iv
ziTp1zsJilP<g>vb=LJQQyc+Q#M6?0tDPU9<$j5y6NQ2VeGlNQy*>`Im6hS&aC;}){
zJD>&z-c1nYJZlX4G;V3@B<~3WonuD3!a$Yu3sQenQO6mYsjZREFf@;-MvNZEZL5g2
z%=qGGqg5qu*T?Sm5!T3LgcIuP)-a)?x89I}MNeCz$zl2=#wW`S?NNwX`lOJ-XHxR!
zf>}7f=j5q48qSPG8U%Er_}J>ow;7=r!))<-681#%dG1;sqd*N97^&wNoQ1s|6_!GR
z<cfbOEl7vO59jPg9M<;bNYT3!t(kQt(HCd@Qy+RL;u!~px}#x&_*gcIEVlZIpl$GR
z+#s!EUg|4H<bVk~<@)>lXq;sVCL9e~*b-*FW?oIW$v>J$gI2(Vw;TMKXgKLV7$raM
zjD%%U=pDLo2vvIKF(<rG;>payNy;nDj39qPUscAlYe&xpM9^1EE~|9?M94N<OwJ2t
zuyUgPpAbM+dxH4BprlEq>(-4XS0W}sJ7IDus;;tUVk4ycMvKd~v8ft|P?f}ZCl07H
zwJZQE=b-4FYty@&d(~)H1TTX4zMjq1ty^7vwqk<zal$VsOuobhe<p#G^gmX-_EdkV
zBzst?dtTSL{cOa(N*U8h&=loN@+nY;N>|yZOgUUiZ>y@_$62S<G^R%6x4PC!27oTh
zJF{Gfl>C%Ds6297+HvKO=kpdGupaq1lsqO)SMWkwOZlaYV?s;}|4a<)wXX)|>p524
zxTWAvab{PqITs?0Z`pqZNi;ZU!4!Xrhk%|4HWITSNqMsrY*Kh2jBJQApqK<nfkgW-
zz%k_}il$G*EW<oT;Z;VL#+XWKPMJ7}xs69wGIU%Q^2lP@X*YOO8as4>Fwx!yjdh-B
zb)CW%mx8fZ=%@IU8hOj9)U7X=1xxwgJ~E0!AY|N{bHAPaW`iTGZ=;tvCb@q*+wShN
z4Uk?w&Km(yAGK};guA@W6g)8N)y8OYYv7Urhe_Xy+|6{2XlqZvAe5kd!uk06LdI82
zdlE1(2CaUoofps-l2z%Z$d^u_0U9}Yv7s$TbO1u}t0vaHP%=PBIA~is9iI|63|PWf
z1t87vQVy#?WAKFwvSB4LjMsm2vDKSglix=szj)8@C>Ju~zWyMn^BJ)cSj-9CkP!#7
z(!;xrHpi4JKZ17ZO$Ex&NQ!u-I!zS@MSNf9KH-l@Ze^C4L7TzbdP3+FLkZ#qi2^}f
zDrXb>N$ZBZEi;n#yFRtH$jP?cG1VUHG&Ce76!3i3mD3I?`1Q1dim87g-AM`l&VEyd
zPg~!5FIQ^nlj%;MD_>^=Ipf%81JXN&5q@d|P1z2F1rS1RjAm%<8OVvoE;Dx~+c`j6
z%1cOZ7(mXraKHk>IhY#B{D_0)!$g%Af;Kqp4crAm8;obl`ip}$L7flUpdF_3;&{*o
zWtleeqM%LCE(qEL@q&M#O%d0FHs#NH&<4W{hx*0oIfmAVR|M@EaVu!ot<MB)((gvl
zUgG4rpuI{%OVB2O7Y1!2_>G`V#B?HP@9Q@QZ6c-<)33%cJT6rDJj1b1;q&apd4-n_
zi$JHZRJNRJT&`^;lh`Hol1ZG_HV6?D_I4W<1A<#twk9z_@UVZAv7F9R!Xok+NzD%|
z&9qQ9EaWLtGo$FMwn;02m~N=vH&%?rV79a_YoP~@+9-!NW-u%d%gdzmV%YLt%R-+y
z#eb25)s?w0Rrk^pSQdc~vN?d9=$BqHYwS3*<h65p^b&7w@wSrFrSH052KU8qF_-Z~
zYNDp-F@5>U+M0j1wElq5%lEyY&9m3^>DyH!j?xuNx#qZdXb_PT%yBoXVtjZLuqF;L
z;#u;ji90F}^7O4`cuO`L<#Cqx=)&Bv17iCfd?M73-q5&yM~GD1K1l{Okw;}fp1z$K
zCY8l&-pO=@9cUlQigh1Imt>HK?w73?k;*X-!x}X`>{fr#*`3fA1lzsRT^$lECcx0T
zE+wM~!?@N@*ugQsrA6RbjLx%rX0WF6LeVWVLLgg_<g{fbBMdxru{(#Kr;x%)4EF)@
z*WY;e2j2Sy+lIAR|BWqwlx}$U@9v*rOU?$aL#86G^-OuW2FyDwfl$$U*a1R=k8@pz
zA$C%Rg@S(uScgR|@S5VrEk#ReBbY?jBP@F7VH%}u<GYVz{P0bFo!uIc?qt6a4Z~EM
zky4AeSwf_}HFos5-z1m%i_a%BhzgqH^;_wQY~SmzE^RSP&zQW8ZA@vt8#j)+bcTKt
zX-`eu+#&LoRkDd}5L{&dMJ1YLx1kf5%WJDH!!3XGUW{Al#bVoF7<-NmD~}m8jAyL!
zI<xH7dO4SkjK1i^Xp`!APf7n)yy(?^ti!QQ+Q?fnm%J7CUTkLHJZQG5_jJdMBJ+LN
zGAYK#J0^QrT?f^?O}D6Rs${b$Hbe|qt!eWl5d(&dp%i?&h(}K3ZxJ<$SUF+79uR9Y
z`!|2YDzUuw_DN#JM9Vw<{}OBXX@SW4w?ROCTp5d?RJONa19y&Qv-Y_67}>2I*{$9w
z;ATN=2v`87<?i}Waxre9mt%eJo}`Z)ljvdYAY?;PpX~4>14$PbHC0YEb;l0`2yx{R
zE3$p+$bHVrvO1cb>ewoVb?N#(9lcb#VUvH3#^S|mCPs5C+dDn8uxFFaGGX+{u^^|Q
z@y;7nis^4lB+(!D_03FL=e@%lMqX@LHq4+y=c-?sILrGLE{}d=BFV{(eyd`)M6zY*
z<BqzSjqB1eeUkeW%do(N7tghfHaQ%$jAkZHCLzfAfBUwg!wO|?WOH<KWnpa!Wp0-a
zZw>|mGB=lTZwnIwHZYeV3jrpVcR3ENfA!^;_n)44Y7)>Ls{7aXr~f>)jS^ZhFfjf4
z<H;wW1SYMR5Rh5K>5qR+zdilEh{&Qjv;G@>ep|@2jFU{QHQkMu`SUM)`X?p+M46?t
z(|<$OA3i&0PPysd(?3W)uTii-C~&QnTA)DM7fO<F$D~h+D0r<a(vhM&k`D0lfAU3h
zuTMp~RfGpAha-KU|E_$hc#dVz<zFEUWBw?Gu5*;C&Lx*l@A8s$N>!=EIXyQq4Pfi_
z;wyVI`Hf)a++~zyz9P~U3=4FXbc~w!0VFyID`E||2O+F1NzywbobdT8knSRk7WDH2
zs19qtgDN5elD{visF<*pgd2v^e`Hj*;{gxhQP6*#<>09>Cdu$`snEWpl-HI+o>d{m
zrg9nc0nmn(+W}2V1`c)xO(5f-Nw{Wc>|YDo9e6OqQY83v8L#7@99DS;N+Fn#;-LN~
z6jR_nWB>beKpR$W2Q+WBjNDCbsqfBmOMN*B0X)cbbIAy~1<A||Dmkw<e_LkHp0}8U
ziBTcD^Up$7Q`6Kx7KtV#2`twu$$=lX9@mk0$WPHn@GRoQIuIOvH7@ZAg~Ioev?!_4
z_qjKueiHb4O24n)ej`_D!u%@gHrqHnjNN@sC>wq1Lu`n0f@zeQ6VzXG(;<6Lw)Sk7
zm*)rnntrQ{K==hWa><N9e=iqRL;7qyg-`HE;lyR&b945ol;!GyVQn0`4kb)<!jGYZ
zrAXRhp#*+vO1yr661b^SxGyD!q3cjWM-ToON*K<JC}BOQPf3YfmD2--F&9=>Lrj+)
z(9PxJ<7x!j$dF+qI}A}VTsVdy$}ywMv;gQ&$q@VKppiV8l1*)Ae}~x44*3W>hEeRW
z17ieo3_H*=%ggivm`}-$Q_AW_#*YDp95sFnBiUhyM5graq14V|OKpV@cHo#uN(s%Z
zE#bPnIn@bQQ=MWFb){L799~sH{Mt-pB9__6gJBJHGz&3g*qsPF%PsZg(VyLw2|{xQ
z%8+nm&Xi@R5R&Q`e>9f~8ykbq5ZrPhG3T>0+BfoV!+E%U_w`0$dvN*==|SkcWhiqI
zvODSU@dcZ=*;n6SbZ%Zc4cez`!<0je-Nc8+RBzbChjH9Gd@#|)JBANdGOkL{DNH@-
zS*CXC38j=8de1qr=iE+_-e1{iy*v4_tHrAdI(K_R79VN8f6*ftN3O#N9o@QP7-0nC
zPSp$g{Yma(bKPuF10z<f?Ef_*hLP(qLK$Z9prvQjz(CGR&o1Gbdp@R7;1VaP)0lC!
zNKqCxws7PKPPB*}$B73BF$`RX5Ewn>V+f(h2{Yzuh~Q~nKRHvuEw}SJSWTLBGqK5u
zP>q%M_j6^%e{J>5rSMXwa%L_UT$>z0o@X-bCO$T1e#16C4CB_}gH+7PgXlnY7hr3j
zGW(9jQe0|(B5(GRa<SD&rAx&7U4OG0A9m+M1nvk*4DewTw+<ad44?LnCHIzF>I*S!
z*}c9y%PsZg(XBqFzj4hZ!pJPhmPO{&BVA>?b!(xEf0ehXTwd0-w`Nvez8EX~;<eyq
zLd9<E1H;O^50Z4z(>(+v=~UP=dMrq<)exI<%og`g1!-KF9gu8vK92#(B10iZK-zQ5
z<PcCo;22QD3he-8qEmPbP)0J;Y6z$U_PCA!M)LU*u#F+YxMDkispz&H1DF;}as63(
zmUv4kf4Ul8&v~!p;>{c3Oy1`}{~K0&2Tru4`GX=psemaopRNDx4{MV9cqbj!eg{=)
z*g;{NM9!48LyKu?+NSnD0>-av`WI=+KH#a_93S^kHm=K!s<H!+h|c&i5DCY$a2pPN
z=*m?T^|s(;H>_7f!(l3sk7@Z4O>5bZznT~`e-CJSSSKA_L-eWljso|VTk6ZBdwC2u
zxM2q7gLnRr-6<h!^13zT`50MsCSJ40<1F<&8ykjzt5`2^#$9ha2iVAnVNKlEuyEcl
zJSff<PBWuQ2v_eO2s+G$ZDpVeTbgw>?U~lq)=W&82@~n)B&K7K{@FRo@#uu~OPqW6
ze@KgsJ#kz=9lhJ=-W{WNtC(>mDvMjgTkhW1vDDBgIV%pMLVY(8Dk<A0MW+EVpFx#d
zl{~DBj*?Aux{gt@L1txH_*EM=0-#v2((|c=%5LatvJ?gu;y13ik1MF7cx{<d9T#0N
zXS3Lqsy!;ASF?OD*)n<G&CG|Dc_X+zf2|xiLN%m%I-}*<*=`|;*z*7=Q|D8sxidzO
z(Vx1u&$CaDUtfeyeqDKXO2eawi-8i&bLAH{#;Gwuo({c}=mL;|TAYx@TvMG-pZjGQ
z|6Y~Oo-Eg_axP_OmufB>JkaW_Rvp>IwJo31<r7`4$@TZQ)z)SC!M!UrZC2Hae_^)e
zExCnrz3(9e4MJf<)Ln$2i%@y9-&p1O5y#VbcHeKjO7%+7!`#lM-m+R9#ChcqctHLN
zX)c)Huxgm~^*ci=@Crrs5{}B%?oi5`TUy7TX{*b%dql<Ol4(~deogoEJLb62c72zL
zyy913N`1sL<y%>l&r|nt6IU8if1M7)3*EjozgGVVK_im_+do18a`2VV`Hv79sbwVN
zRE8|$6u$lO`_w%7o(3Vu*9{d7f(j=xBn;*~pPhAy`mY{tAO3U<@LICHH7q34$LHUk
z?rERkyyu!UMk5nw$L1;JTqTCS(fK3jV)5krw^@*+=!t5+6eRefZ4U`Ve=_y1kq7DE
zUHj(=#acA2QkDrbn}Ry%&rvCVys1hC+to-1Sg(sROq1)<<Swyq-oJ?+pU3NBB?IHG
z>Vjz@SI_ZpQ!i>e183TjxA=jENhmmA<-O7%^>D2-mzPo-wp?j&nOn7JLfR<{w_D*!
zI`10{{Usgtj3#BA3roqje_En}Id0nA@~R%OC1LlstPS(S^%LYA8uId+H{Rf?0O3c&
zwtDycf@HyNDO@O0k7`LDB#CNX?Xui+Q%?D|O3C{Vw4s&~6`+jKiznY#h63`laNZ6k
z<hu2#wuG$%s2S)hX*cVer@AD6&A-HR69B3)V5Mbf%~iX`2V8-ge>so-QsNvAty9Jy
z<s#p{)yM~s;uQT?`?NlY)KXd>DS+`#*3=w6hKB%#+O<-9ct{NYjnKAxSME_fN7q$*
zKhRdAnITDSTk(Edtz+KTYMMGuJx7t_yd_M?O}iRa*=MMQpc{e7zBe9@1qxT)ZJQLp
zTPa7c&}%_1zpOdU3mKH=PVV#n00O~ommxP86qiARHxvUhHZ_;Q!U!gR#T&Vj?6xuA
zuiy_XgvO1kS6j8ShpLoW-b|&)wv_k<R^s1xgKUsQ^9IT1fWsLM*p0pq0J8bBk<HIf
zkFTenKDBAQ2o}6*zW!(c|B(0z5+6F-cmUxvG+*DF-@Zwy99|E+jFejA<?(yKyM1}q
z5O4%Uyx`^88^B*0@KW7>U47#9M&HBBbNau9)4uD!5k%X)p<dg6bj+h9UJgeZqiw>2
z;{cz-<K_3SU!Ek9Bt>-TYr64YA74-ZdEx{bBoGla|M>ot{xBMZ4ysWuh!~;y>tD?u
zPk-ZJ$NUR`{FBaETky{q5n|(96gvIWvCl#`Sz&Ng|MAzK4SVt)>(9eHtrj{44ha+^
z`v61-B1%7jXzL+*d;*R`l<$`bMGj9Y)X&i<fi7dG3MmU>qCpytTbHsH>6@^|Ks_Ng
z3UA|v`1;Y#2niFdm2a1VMGh!`fp?suk;kVfrGV%;)`cl58HAUPjn?wDjy>73O5A6J
ztw0zz;o-zqw=L_Z&nV~Oh*D8RA0~5-owKwhrln}4jKgT=CQvdDPs}eH5QjVjy90b(
zoIgfmye51;;oQ9vR;dZZ1%Hzbi`fd)>0yY%PFS`SS7Hz-Za%squxP1&gqC_!Sr0%2
zDFRHGxZt!kpjT4Bjk&^WYo&$tR~oWbm<?{RrXQ@8kYOS~(uCf<CXkp^JL84XeruF=
z!l0dJ%}Tbd$>Y9t-nhNL<2|Gwoo7zZ+oZ{N&r=e$gaNYv3$e$LpxbPzgz1xXIxCmg
z$SI(YmVw2Sxumkhf)hl4Z;jPJP%b6%^5xk{)&4QB?K%h(reHqvdp33<sGVvcW;}98
zSSB8fu<mZ!3nWQ4t=IwuPnZv;IOAt5aQq?ml$d(%Xg9#^RmTnAP5SPt9ae!AV8LH$
z;Ugcsj}gn_lUg8!cgwHDl5`g!xs}!vJUge<At5nvn>q}&;RGvxSS-;QzB6e}umUd*
zaVdu-?O^22OVUOq9F}SKN|Iq|O_7+Muq~M(P9`3aFh}!)?2w_|FhtJrO+GBk4i(K|
z+`CzJC6*++)ca|>M<Fp7C_b8mo8qAvF}I~KWKrxFCc`_J>n1L|)S_9BV)RRz`Vgj<
z8SRlxs;}QP1@}&WgN}Dbx;l>m1@0c4??eK94NyEO07ij8IYgE76*kpL`^O+;;BT)7
zc9MIr=57<Cf_$A@ye{Fi><s&zF6XDMj(i?ZhB0&0dNKzk>yzQ6Fvd@KU^=Q2LEz?P
zn-w@UL7cl-flHuQC~yh!r~;R7Umew{^}(>5iRxX4lCyC^!7EZ{qa6JM?2qd5bz$?m
zd_fdzvHlv{Gg8>7CyI2&_E0tDl>V{p@q!&qlzKImpchYmov}|7M36&G>UycZQI}>&
z4m|;hm!wDzXaN$J4M`4i0mherNe<h8ni@cOd3yFn<6r3+m2V?z3%%+2SWWN-BQfh!
zF5bjQr-VJaur;&7cOA3o@;jMc&72LUn;<mY#B&Q6hy764t}M#6ya2juy3FYAh=+-P
z54s!r9*0z5laICZZm<z6UnKV?K4L!ObUe}~n=FY<_-jZnvku1*-#z_)8|ig_`u?_)
z$JNZ(V6wt2FshrFtfFvzM89U~s&y7-#L!4Wh)kigP~J!wZsbHB2b8;-K^u(6I^1Y|
z8Qgxp7sKu6o3;S2XCWsP1yddG`s(<{)$KcVpqln|o~Cu_l!@HJQx%;EQjgP|X@%d1
z-E2vA6dE4nrm_ojECgrwwll|nsi$^>i=j(H?%`svA{a9lZ>Oz-bKKm?#cFCdxahmY
z;vOz~BSNs_qci`HD!q>#4~h#pc6fT0pZifD&&lQ6I9b{$t7@KZ$jSxbWIR*29R{(l
zR!v~dI@qOJl^4aWTG`Nq#6qX%PGl(r6+_>c(1yArhSuP0xh9*}@7j`oKXp=4QfI=!
zHM;JN+|tU&bb4H}U+djEPo@C<VI0nrnSAyu&7fPg=}v|XZHs<H4W)3HQWz+GQGLQz
zP@89D@)I|CUv+Dvgx`#gejy1uB=xW^xU{?zrRthRv;P#$OD(i_oQKGvjz6<apMI{f
zq-dZbowVy(S|fVxa5{j0wf77-VHNOM^s9{r&)+#C^z`^0VCHXvSwDWar{UHb)Iz6^
zx@QnUz-X~E5<7tTS2gMQEg$~tR^~Y5t<IbwjCyrt2?MHt(X!`cIkWnVA%ncI+?+en
zY_y(RR(*6wK3j9@muIW@!Z3_Wn?kyME_`|$tHW&(yuff5h0^1H15&vY+)9lcSC(Y%
z^=~XS*KBG48-((zSn3_s@a-%N8E<F&^YU{7+yvU@W`<UG&&=LnH4pl#*_Y6E%)WjP
z7N0qs+TxS9Rxzlc7dCfvhc`LBC*+XUz=kEvvvMe*?U2KJy&TxgFOowEbxRJ6-V<^_
zh&q<rXSehF`OO@E-T7+<aa$J2XSZ^H6bp1G?Hr6S+FiW$ib37ut*W<Paa<qk8c~j}
z@iVmTYA0UWB*pPo*j0ConhQ*xGb`Fb&7iWgntTT}C;f!_NiEE<aC?>dl~C_hKZfgl
zUH!_@HF&f7F}xe<$MC*YtKU9yX)D(SXtT<dkd}_1(_SWjj<>?v=v38Cdg=JidA#}D
zFQ@nO+(u=vsqmRV)vITJIqmgk9+j|g_5R)x=&eT1P_5f|IrxN$mw=X3o-CYUtv2v|
zq>_Czr0><{%}70`?fHx)1G<nZlA$xY=Zi}OZ0((Um@zk8-IKOir_Q!s*!|<{>3=Aj
z=4c9KZe(+p25$}y0x~$4!T1^!0Wp_Bnm{O*p-v8=f9NORy*M9!{rKU9qez?u8^6Yn
z@z<A+m%m>|0gaZ-sOh(FFY&!rf+;U2p&Zj*PyhTeeSP_Hq}syg1;EFxHKv140(C&t
z+Q3w7%J3PG!T{qMU@!oK0r>PKf9Izx2y+T168PtzIBWOjr|+u_R<OX`I15Jzb)4xP
z9#|l4e{r@rbBON_ua?IpNWV1l;d5{C9$RIi)xk?MOF#`)csBEe#+lXm;gQR;^HUNb
z({PE`$Y|`s&+$nt&7d}(DOjH;L*363`7|dITJjaxga@^P4L4G$uw8@WbX#^E>AGjj
z-p`U`KW~?Ppt3Ie`L^uJGVMIkQ!bA*N#Rgde+@&*pCoJdrj=D|=EF!73|enwtjj0~
z#*tw6rXv^_5FIH>h08LqI)dgjP!{9pGScLoLJJ+?E4xhMJ4rLtc9UjJyK27KXs2z)
zPI|-OnTg7ib}X8a;2GC5xTB{?KEP(%pFBr=ch2G{i|z4U$R0?#W$loL84-5Yj-Kcq
z8rsoQ+^QY@=x+P-x~Z;f!ivq^t_fB%4&siN)=&;ee_>q=A<whWj)*8>V!$;JDGD9M
zzHH`9fdx{^j|;$u72QBEYKV5Ds2T$-08;qn)g*o8ih*O?Sz&vkduVJ=ajVMqqi=^l
zpHVz1n6`8Iw`ra<V4V0&=1;0#F?|D)kLzA>#Dlc=E1wCHS*`X}zJcft$~RElsC>id
z+wtcPf66z?Ps$e+FD}aw-yLr%-#}$2<r|Uwyz-6GURORISC@lu6E7>@y#eaqT>`(A
zXdCb>77}(^OM_ZKAx;dG#pYSdEP^vw`3cA|Cri4UW{4qnuV+__0z^f7=DF*roNy(1
z4&PF^Mh;q|^Hy0OOFXtzlMwJJn$N|Whk~$Me+M@!dPvC>B=uPIAQBZYpXT%MeW8vb
z9D@%a`OB|-`W@PSfW9!(^l$3<txUtGzowsKQDucLqy~wuhRq{y5$2O*0qjdATS*ud
zQg*SK#GRahb%p0PSy=L5M!L5OM?x@Z6cC4f8A{zvVHlrl4XJEHd3<HDTk9hH!(85$
ze_q4(HeiCEBE1fVc~xT{ixznn&K5M11-uWD$2ng?1Wb_eW^Y8I)ptM~T&SzQs??nL
zmr_4DFT8u7*Jp&j6Q_seS-}a6sFn6fI61|FsD1$KrPE7T(fg=9f>*qn{@zv0)SZ|;
zEZYiZCNOK=lQ6Sj{F)+lhgpM7jDy}uf2YJ__&hAf3O?Xc9rsVd#|Rdh%JRABhuAFt
z!+fsDuN-sFGCevb?X|f4{t59PWphP(nVscxe361^u&FQabg*(g^~B#NY;*Wd<L?t@
z3I^NkozOpw(F*n=yXxm)FCfor3Y!ZjE&n@?56ioPBhL>0IXJ>3!it#%Tf%Y7e?|lT
zJ~g<Vnzj?-8o#?cF??9oH4FvwUVkQ6*i7QS>?o48{-N|}OvBQ08$Si6rQ~+$&|X>1
zq%96TAIT>be~sxkp0uX}B7@c|Pz8YwkK#$Y$1<`}jg6z@ruU`ernLTRcM6o2=_If%
zE}f8yrF#S=b9a@~*dtq1x-npmf5=4{r;kluj+uYq)2CdWX22>@IXuI<3k1Df`M5g9
zdK;@yfy-PUSZrx)AK)szu%m>MP<6{Z#l=bf<Z|U~1auP=ZlFr_v(NAgSw#?X#xG%h
zL}V_g9h&fSbO?FM9uE<<I4;R}km>BeQV;hwTPuzFlC0NEUz7*AgPP0Tf1!0v+cY=K
zInj?6>O@+;TmaPqA{6PbV4XJJEzOoDoqJu=Y<-tz=U%O5*PTdxkNu%<lIW5xd+x1m
zJpxWM(MVHgFZ&^nFzxtw;+Cv~BHfKG={Al*shZs%zl%F}KYW@X+jwe&eR)+O-u_uS
zu}(<kR;r+BlqkJTzY2wdf1Yi0)f48eRcMf{Be71`*-%Yr7e0`#sX7NOy3C&YWt}X?
z_sCPH5y7^naKxSM5BlTpc=HNkfVN^Jh<Dp+1@Z6u<KXV%K$-n2kG?-?#TAypFgxBx
zX^=3rzM1&2#X8t#mGT?2WM9&HgwdXB4_xX7JwM!jIp47S)H3K5fAF1cUE#=V)z{+9
z%G#Z9)^o1yQf`3jw}`0oObUD5uDJW{$^wQxjbhWYRR~RGT9&IL*uQ95oa@UW-OMA;
zK^`wAjM!T>J#I!G?h(t7CAg_=rB}nBKG9p<?gZT;&{8z~(|%VQ&%m(_%UX*5wmrBY
z(-hgUZ>Z9;J?Of@e<+SYM>|t1Yavc7X@TzCH*OK%sW7SQ5!Kjrj=gL}NkN#*Tgbju
zmz99c*F7DygtvBQwEI5qSBLdq@;GTd>2gcCbsD{u#!Cxm6&9}!Ko}lRI0nI6@9x1l
zIjfdDmxLTa$QrlO854nWuIEnGt0rGWR_P0VKbEDt$74d6f1hHKJ>q3wmo=%0y&lbV
zJM%K^>)vX1G9HRT403M1n=7P+ZDPlBDlcI}-M*T8&*5H5q^_JQ;z^wduupBR{9VZy
z*3SLn-4bXTsLA(CG2LvY0@^1|LWyS&$#8SJxEQV6$F~N{rOZfCc2+Xq(JC1YA69$^
zc^ouqzo$Gsf5mujtWZ<U1&1AY2Rjopw;g%!DMsBf6#t?>RCiqbl9V$XTWfI;jUk}W
zEEh8=70Jb-q_XkKo!$>BW0<J_Xs`!VJw>G&Gv@YHQLS0B-TU$0Q&e^#OmCJoE%A=>
zC<Sqe+EbpM;wVp)<?8w@+4nV99pViiQ*V(TG;ib1e`eP_Si^D<K)5g-P$z^4C5oQB
zq=2*h${}+T*?BRh?3Bc`C8O5yW;%4iKeU&gmU^bLx1y(vcfE_JP(FnP%?w2fvJGv*
zTxArM|5#_#t<h;^5f~ulQHm1vCB8|b#3hDM>ArGGO#NHE60A6liI-&RTm0vN3$?1_
z!K%0BTxh0AVT@ahldWc|#(j~zI-(JD4SB3-H3F9CwS?6r%Bq2J%`K&0NCGI1E0JEB
z-{QS!6m{Ssl+e3`=OtE?wKghvlh=tR8>&%Vt9qChkN5WQ6C$|(0Mf|uNS6U`4if@0
zIg>Hj69P3kmqFASCzrff4z7RH^Y?GxJn(#_9)DwjA(jPyb$iwUOF;P2hu<{90mKSk
z!>4|WchaV;3pi-JDZI2h!!9lIPn%DHa!TBTIBS_#g%j`fd>fvWb)YlE<+;hw@I0-b
zrMTu3OkIKJf^W}Tuxs&ccQ>~-1qb{#!1~ml9iiHdo12WUevW)27>a+g*JcW&I|LZd
zk#ljCt(d6%^qw)vl;=oA#nm<6LS_!x^^JQ@fzMReEv*s{qO7cB#7M@SEHct^tw4#`
z#Jk4Zh=fU!c%88>5|qDfsMf5T@HCyRrBE-TtS?-co~OEVe3ri=%|$+M_52NP<P}41
z9L1>cAC~j7luqw3pf!J5k0F-;m;hfb5fq7yJclZ+y5N~9P<T%?v(P5e%u#&^auaBp
zBv2L`5pd?bk;gzQs&9)%7fJ^z!*!*3GOt*6H@u8eAc+RbtZ(F-r2Rxg?}R3m8(BUf
z-8bE_)v&ss=snyM9TshrJr5&2hz#1rGQ`^oatE=xk>2G9Cw6~jC>csJ;i2lUX?Wxo
zyOH%*uXIS}J&heNSj=Qm0F$O#3{~u*>Q0ROg+_v;*Tvo!>~0xbTc1MtZEdBOHI*9g
zUl9?{ypaZ44=$x+oEw7c<G-HLma)^9N;8Fzy6b^$4Djn>LXxQIYf**5PCYPo^4gN4
zik7GZB2GqEMrD5j72y!ntYuhL8T9a^sX<Tb+!*wXi|HMXE&^&7n8{QhW(RsDW~8F*
z_|GOgz*}ax`@h&RlwK~GLdod#h59Ap?-qN$WAoiFxB2ft2I8yMuZ!_1$+Hv@%9_eY
zC)d>z8)dHrifwFE@pjvud+p=w)bICNoFOE8oy7)agL!|eoF*~=ig_-OXNFlVsMr`Y
z-hyI-;zG83hopJh1f&&n&e2`TTzq<a`1ivaAz;?q%U`}eeEU7W{1qR5bKy*RclmEz
z`t^ch%h4?J%ReuFdib2OjoB?7b6#Jx$DdXlu~ncw8hz0V^Tc3l2TWOL4R*C7*-!&F
z9s7XW3HyKYlS#0Yp7z3$&xdUGi?dmtxE7zR(sl{y!n0ttK>BK<Dy(}8d3}dA6c8L0
zr_oa#v~H$78MPhl$*bh*uj~>UBzHZ0vIWM+>~Vd@E#sY|xddx{eI(6L46I}zT(zV_
zj8Hr6lHZlTboTL)Juh_))aSR%@_YTJi>kVBT|IxpWlFv*mK#hFk8D}(kd6Kz8`Jh(
zBWPP$Di<Uqd2K>TLM$Foaxon*7tNt``YSx+E;E-0CnYn^x|Q{p>bR(@s1Zc=++MX|
zIc0$8$vqPW!x^AGYSO+EvTJJKmvpx`{XL7vF7NjGn(wL$ePhavYbJ@3$oICb8wRt4
z`89vxyEks|MjTmIoa}0BlPcf4vVcA3CaZ+xq%vVZ${{@O<#US+lIco(RWD3D=Zd_x
zW&4tot{KyWaEVMUiT3m9xK}R%uNOvjS=B<~L2Tq{q!w>@8(~uw$}EJ7_;gupcrHGU
zD_$ZvR(a7OT2zZegvx@|4$PbEdPiF$%M5=vMYY<ymP{M$iXS4|!Md9mq+)elx$7iy
zoYqU`H_9{skM3U7h&k2?w`tJrNb=Z9Q_U!Ibs%~w-dBHyyd~<AJXa?VF5C2-TUo@>
z)mo-7`?sPy6g9D<NPn$KloAc!i#!(7L}j(!-YofW+-;^N0l;9usD`y#YZ`Y@D;j^H
zo@4E(0%C2W(Q=~E>C~8w-^odkDxxISOwQ3R{<#|R=Fa&H-_+SkRk($1Hd93uKIXk3
z^mzPfGkRAjBE|(00@FSf$sT`r+JquGLoZCB{Ql{kFsAObcDPevDy!(m4^QFlk764n
z?hw-UAU9!Y&a|eo@hG>UXpNr5oP&Q7z682OZ9_4@Jm@i$amu2M#}v!-a$Y@2>jR18
z^@+(Mi_<!Q-HN!2({ESRzg!z^REaJnmL8X(9xGr%xy4S+GvRG}fI=;~t`*<c6o@Pp
zYgIcn;V+oc($%L))3+o+Si`jN1oq?4o01T=a7<dsmLE6O6{BIMF?$E}yjFh*s;-lU
zoHI;;?TJEz4t1sGxrj(|v9N2JDxpkucA_MJvT?t)$XNm*;`CHg4$n~6SPqIn(IwvN
zp50h9oWoyeCQss8Sjw<zbxr98Si!VZ1Cv~z-|KC*3vR$Mz|!0GyFO;rn#8gcsZDW2
zD}y500^JiW%!&>79RR67R==7PrBp$G4#ha6C>4rTMHkUN&?o@yGe-*n(l_JgNTPpU
zk=*ZQIb11JIYIE>ajH^zq(`_iFelX!Qp(`rtnaC`WvbH+_Dpr6V3FT^>vMvkz%k7g
zrNH)eqALZtx=vfSoOo2D83vvN*3UZ*RWpzj){Yqs4%G`YX3zwRRha9jGh(2BtYco<
zQ&5~_5`aS42@afNMyw^L=}u&>w|b?Q)?*X8z7~0jx-|3$M0@tid)l?-y`;-ag<)tV
ztcWz*EVQkfd+(&<tFpKo8##&Mg*>sIsS@x0vBY?8=yQ?t3ZpKC?%L;UcZu=T<`#nz
zkOP_}Hl9F1_l$WY+44bbMxO$IS?{?aMB9rwa0{}9ZrO=BykuS@W|lKh@`xA<kC!o?
zLfNsI9G4n(1`0+qOm<iK-L{p*1WAJHrA)OUa+0J2rPZ`@w~1{@6WfWtiM>efSvfDK
z(Z*L67#Nxt4~)BxWS^e46b3?@#7D#AU5*O=@zkm=pz4Np%?XtPVOem0smC>{>^QDa
zfN3g7e@dYMeZ7!yMxjtvnX&qeLV>dA>(59!nqdaPD<0RUkShu|0W#(^cn025R&OWt
zpb>Jf=O4DA7>ifYr=-P(frNA7Qns9s7fh|(*#6Q!rdN!cV042-img5-=4e)QZOloJ
zznpnoh&<XibxMjfnpmNK;fxe%ZkPji3JTg-(RuQn5Jn4)dB5t-vcA!(V;%)raiqo$
zMn%#3H8@hw)cBK&Lw-`AB?ufrrBM7(m@Gu_#stW^O){9A$jS9q-%qsl#F!vE(zBS&
z9&D+*RDwz05oOyvN`+bl_6!~4FcF9`9cNH51`bh83uQYLs0_w`T+JD`>L)Gzo?Bnl
zP7|?Ya-|k6T-&M#+Pgq{r!(#wGar2|kJQ9TpW{Gta4B23adYo@XMGr~OVn3Y6Mr?B
z5+$9haPr68h?Ty@Pr56u!>4ZwH|j`N+(}0$a#!VJrlPc0+AHfvztYM2w#+kzYGTAm
z@xEtsYd9^3okdxHyZH(7&J<~0zd1#UoF!D4%7i2`A?u`W*HDc=KX%R58^O3rPsOn9
z+@){yv~H_7>f|yUVVWjY)8q<b4;)Hvs`~oOvl7{;Rm7pIWu&ogUCz>*7p9j@km-cW
ztgcC?d!Qlm=oEdVLY2m4>nVDO%h4<#W1$uij%T$^<0vM7$=t-r2`Ae;r;feo!xU(8
z>z@76W3bhfDQ|QXwrS}l&e?!VZbrX<6?$mV!&DjmixEqm+%|YIfd?DDnc_=~A{_3L
z7|BVQ!^Pih#(ok*UWYD6VgzBJZ{tj$*!X^nG#p(|8W4N@^R{Edpj25E%Y&fwk)RYa
zJS2dmTXvFv80kY7MYP1*bJ7@oqJ)qgJ3bm&E)cxLoWXc~sI-OACt#Y`0@<mbp?nE9
z1L9^KP|bnz{<qsLJmE1XRs^pPn@mv(Fv{>}qy^nwY_Fb>me4F!AkRpP3hApE3QOA-
zb><Z6pjlAz6PV+UP%t8ReA&i4uvpLJd{*}r%yMymMxZeFR47Piq!tq&E94m<n56<Y
z6v&sIsxQ)(Q`Pv6tL+pw)O)4x;Qa9AP|z!dV%nNnq9v|PDlo~9PzL}{XiwV5SwP4Q
zQ4K_RaY`J+Io!{Q9_QU8sdOTDyxYOi^Bw;_oW(oYW#M50qoe3fmnB#`2}5??SCi}~
z(Lo1)%qi6$zwGpU&?5#zO0k*b4#Mkdfd+GYAjBZr?!H~t73MbY#H1&6cWH%)@@7(8
zU!lo1Jexp?%x`nME*{UTN^mteMYauWL+(VzDw~Pn68Z{Gd%SgoGVkBToffoLjd?S?
z%bIdGa_Y<Sc38uWWe9c1c^9G$H+<{rM$~D4t<?7>qjt@h0beLrmE5l8I?FCa-c4+R
zZZi2vGW;czF4N@I6xi96Xj2jB0`FDFzP@Fd-e*yHip_3JQ?@gKWvG9tSk$a;JQqvx
z*pX{GX}TA<hnWDh{z4fv5%O&89$t3BNL1vPN)E*CA+EGsML$vA=h%zgy@{cyfTqiT
zx)Zy58%JG*)QtN>1!a^nhJ2t~c4Bvhj?sHcy}kA~ZqWdA%TD`SSjLsdEt)Cqjt>1P
zZ77tTKu8+A{{+Z1O2!zCa(^d!SC)C`@cQtyH_ia~Xnd97n^8(*N)>L8@ryTE#-4Yg
zQw+2n1@if)L(v%;5~Nu3Ul06(=uBgO3hc;9kDqtKEE8hS4)(7cWi>UOYM@@m15C3x
zY&N6ju8rBIFJ*7bk77zvLhb0AlJQkDvgOC^$Owfr#5x^uCy|kp3}k)dO!`zWsV=4+
z1HZTb!9e$V!+P|7=oaketCsD+^w15I0rK+m)7!)U0KZKF(h6m6WOH<KWnpcX3vUhy
zmmhjB7MFY?FeiVlT1$`Qwh_M1uh7wZXfBJaB8$MLf$sLs0=_w!g8|2<_<;b~L-rE?
z{1hcp&6nCUBONrFmPl58Up+{csHZ<BJ$-q)zC3+=wm5mk1l~+9zfT}9wEP7v4;@V&
zKsgK3%h&1WPg)x%*Htdm+FH51{LgZy^T`5B3kY(-`Q(2Q<d+4x%yKu+a^2GZd_KkZ
z^Eh)~|0#_~OAm9I-`J2_X}O%VM6?v;P`dE^%gYZ>IzjsQ3^WB5oN-v<5#^L2o*bb9
zV=^8TEGW|O`+1a)&&~{cTc^mls`qz)en?&9xeDGgRb9|1Y3WGhTDSB#e|RVaSp~#e
z$d6gd*i3(hz0!duXIMGyGc0n^2y*wcF_sZL5B1bRA&FE3rIVgE>&ZhsH7XPJw5+sW
zT;sc+au+)mf-11^1Wqs{X#f83STBLAsqXSOc+8;D$~0`;%{11`kZ2a9i&p@RW^U|C
z)(CD}OZlfbX~yHn)-{S|K}9byk)C^fetG)u(_Me3C_q@ztSIpGudh!(|DvZ~<>e2P
zR#qPUZ*KZ^Lgjtz_Mhn=Pd_g4yhTkO)2INri2wH3gH2>!dWee|ew7DXtH@(5<j==s
z3M~@Xbr&I%9+OEzrZ`2_i}YV8Cv1|NNJ1v~*Xbt6!B;(Ap_*4@ZPh6&Ciy2?)Qc+)
zQZ0Wut3d8K)k^FlE-{RFev}2)5m5ZW@j5~{&#96m!I30c8SGOUFQz@tPq3CGPh)CE
zN!0}T*E|MS<JkF^$!}4TF+Rs}<+u2?ewLQY+e;~J9>wJ+&}jrIf5oTsxUuc$TI!<R
z{AM0))%lY24(ZLR|EwJ^8qQcQrG^bwo!Wmup<C>CTeS!aWAjt)v<Njbq;tPG_03Ph
zou{nX!Yt&e&E7B3Gg`m1w&WoOyXO3)X(gExC)YT|x><u|;(esWG8qFcEmuu&olCU1
zeVlLO?Mw5k%4Atp8Bs}8N+OHT(SovUCZny)jXZt(ZTelZS2Jq4Mqv=Q*UnLupk#lB
zlUTq>?4%&9Za=$oNQ$B~UH^RhV``rK{mWffW0eDLM>D!wU#zW#e2^7E+qw#+7=p=N
zG10s@%A0lbsGKDFMp`yUtDBdFgrWE7rmwZD$*jndYcK?{RLCRQ&u>Q{sYzx(WLXZ~
z^_Q@Nv{ad0EG}1Eh3P#kH9|Q1u3QDz6kXxnmy&7@pnrKD76;xHiUT|)j{Tu1J2NtV
zGk0ZJyq~Z}QcIXShOkBo6%za*gaIgW+!5xAV8LBbR>B%dK7}Eqq#rN8pRh(sCd{Q&
zv!5`nmD8aj%oZVeOaIe;!Wv0jWXemDasXkClx%p+Tc^E*Sq{q4aB|FxqRZ}l-w4Tz
zhini$WPg>~2%$p2^$@1S@Gd|_6&4X<kD;oOQm!NRPh2ZkjT9`%IwaFu=asjni^EH^
zV7cs$!&Xw6>0P?Q?l>Ib$&t>75JpaUk6otcVoKSWAv0l(q%v{Yyv*BA*Z@zt7PjN<
zeA<LtB*m}>GwQxFlv`S%kKCT^dMa-;Bf<9@#eWdlt}wK%<HsID%cq;9>6>&QFS*OU
zRtE<vXy0MKq)$iweLApJ;KOTmV6V_(r-NHQlFc8^Zez=(QkFGV9DiLr>D#E@mOXIe
zURB_XmtH&>2IYw!eM*ql!l1M%z9npZbaD#BFosE4gx=ieleL;F_RlB7-gji2yN@3R
zS%0ymS>sr)f1oj7ki879pFeukP8w)92(_dNFMo}UT$I@FCFNXOOUuF&cnE*hgp666
z+zYc!vu%z9pV8(m96bU^L9vdk$@0n;S!$p_;#VS6qd$t@xKyD|{J2JVg5b<{b?+o8
z7q{zj9b_TY`TL5JwjoOr*1;SMxe-6E^?y~6`y$;6d0hYKAWP2n=1#NQhAcZ^u!mXM
zOYZQU`0a|_2^`{fsqG{YsVhs5;hOkq>`t-YhHF3pA8p}A{J7q>T6kP2d#i<^`aYBj
z2t&wP@`yP)jvMjgs#)dKeYvZPWB(Ykj??IUD;+i=L-0yG=U}I9#BbNWDyQy?R)3vS
zUxsjjW%a7>6koRC#;F?Iv2a`Q+m)+LxGup(ZMs^~7_xUN;GN>@He_#=b#O4`M*Mc|
zX@xv48g-D}T14(3J5(VQfwT=-*1)KbwsIqWyLwbvxi9;ykgIx416dZ%U}WiUm9{m_
zb{lqF;}9I1zFHN=MW8BU_vM?;%73>(>PnE~`75qB)7vXAnU56)cQE8eg>f-wI7!{p
za+kVpW4I>B2D7euZ?C$*kRm)fmIo?~%Q3_Bb+4nWR;~&lW7r_FeQo9Kl^1|j#B{Kg
z2P$+$m3m(8D<PegoAO(8Z_k6os-#~l8SzvM@_9aWLf2*M6VK)>=TP$Lp;zU2X*~0@
zoT*7?fjAv#SX-2*iP96ho^4r9`{Z*&a6b%ztCU%$w2bI6jh;?EE$<+=n$P)|d}d}t
z_TVH_LUOCd;-rVmr*(MaBvMm*D0cn-2ali9(U$>l4i%RXbuJW_)Kx$yf4y5xleD-E
zzR$1V4~*BcEPqtJ)m7a+?`+lHlFV&uPn+ys@*c96{Qbzr24kQBcasAHLkrn{B+HK_
zklml1?0$Z`eZBqkiMq4G5m?oI{m}vcAo(vyzG<kl0Ky=2Uq8BEze%YKA2&X<l!$y-
zzIJ@3%R7Q&3GjTt<=txFfBz8pQ2kkb@^Mes?(!afk9y*>_FK|O{S)dMe?gz6Bp-$)
zwMGi<P(Cidef{z#(~zX_CR~P7{onVmxBtA=12hJOw66R8$6NS8BM5Y=10y_|?yrA!
zf4u#z)>^{nfZ>zI5KZ6%$(U-^?gfUQdiaPttQjt1*vR$&6T?BKe;<U<{P^(auRpu$
z&fk9?`>(ZdzHUIt2^xbip+N6yY0cfxY!U7IL*7z1jOX@bXjD<(jmaJE#!sJKE)Lvk
zVJy@w1?y>@22Opeber^b4NCR}0p8Xc<t_C;^qaN8_shGHP>hP3X#rMgHa*+<F|vm=
z5($zsX8^ruD@9&df0|iV-j9#OIKjuPU3NH7)o9u+t{e!S(=W6*N+XCKS{%P0MgtWJ
zvb14^V;VVx;7s91r7&VhvPUy2oZNW;Eus?A%^W925C~u#?)oa^A)4qJ<k0T)sJ#HG
zb~-(V07bJ&CqIvy*+7cxkE}9E8v!tzZ0xt=gj)*zz-K7Ve~U|tb|DtxY-!<epupv|
zC94=gcCkIj*9*%P8ewJB`Ri}8`zQbS3k!$f{@*D2V`xWq|L*<~7OxG(j@=<B=IH_S
zNU`jc@Vw^qYDX1>Uhu3uu3~4;l;Ut;9;@Z`)>Ezb;Q|dD*Ld<OV#mkPJ$qK+1B)?v
zMOltze0=A1e<7@$%xp-FGLZxEnaR@Nc*PJW&N6`4JjxC9c%a0~=Znz`D;kFj;IQF}
zaKQ(r6(*eT)!9bA_6|-)3RjBiBo-(g;|A9tetBo==W5M59YMt9$9AIJjK&7aOi~QG
zSEI~Kz;FnaD7-QQ${7b!;E+5HT!hTl==uax7PmdRf5^ZtiTc8Bw=rmMqT7tZV|2-R
z4y67ZA12Qs$=F37*6I41&eTVR?7vG+c}wT$!z03S0p*=O+dldYRIL-d={pBh$IL~#
z)xeNvc%JRiKbHeT_Cuc@xz2+_7I#UTDZ<TtC&;Z}Y+!6GPhmJGE|7)T_c|`*P<ky~
zk(T2Ef90@|#A1w?$A#?JRBGP|Y%>BI$dVCGp5?xVskvYEUP|EJ5^C<T_m<0>Ej>Cu
zV#-HsAvQ#moj^AOvH`U28PYi}9Cis~dTGW8TtXWY+8j|qNnA3DONTE)$_jzNf;g0q
z?p)^D5!pZ*dp32BOQRDp$G76s*D7w};Gk4Te}Wt^>}V9;b*Y+~U?FA>MB-)?Hqcc)
zi#i8gWqD7BtLP3;6?2=GeOVU<u*v;z>qNp<+K)txW<)lShMrNKgESc4xan1-Zw0H!
z@VVvKT=VDNNf#qV4k%RE&+kOJ8I28;WlyoriOZ6N@@A1*Tnm(a&li@cg(()uy!CMF
zfAKhL&8uHe@%ACyjKpJvweR`WDF|z?1)^PC`L5Nr`GF5-5`9m>7N^moW-YfQmLtVZ
zgxitWK-h)zGdZWD8un$`v<qiw$+Feyyd5<*O~k5)$rfsAI^?=tuBq`=tdGT<6ZQ#U
zZTehOytcW57Hb|V6_v7{ke#|`T|^fJf8Xjd)TRFovYwdXGDY_4U{y^kBdo)ds%jch
zNb+M<?RfQ!XCWG_HLaD;LZY8-SGVg&Q-tCKW|O2eU<NT?N0T*F`q&vY&PENe+*h8e
z#<^KvW$LT$og?&jy&Cz-a++0wl-fZ0kj7MSs_|GUr#W89m`+yGp=9`ch95P1f2p*!
zoAXYLTw{Wt?Ze}>uTSu!4;0lu6KoHw*x573l<uVbJ~}T*kBKPLQ;#?mK9l>)v7|<J
z$oczeoQ+YHUYql89{2{Hme&&r=kRZ|j-A^t@)jym$5^C`T$MMekIdC~?pxZ&w}G2s
zN~TXrDWS2#;38~Tq>Zu2+36y$e~(3W<67;*cW$iPhqnP+sNFg|2{w+U;eC=dnrsR@
zTvc(h#m?hiY%gy6bMNhA+rlg;d8*oaSjE2eC#Z$TYF|iwnQaO?=KnG~DG2SH44V{;
zeUeWF+2L7JO4IUdC|TE&G*|KJ0$@*~xypxq!I0`e-Ar>UO&c_~ZucZwe{1;;n7Xw!
zTASW0tE5Y3*rK%0(>*19j-aoid&($bHN2kgh1bZ4@R<m6SjEnsX=V^?!c+L3A`s50
zlJEWh-{oi@;RbRTIK)XQY^C`@g{!5QTfs(FQhr$0!-{N^iJ5M_B(95ML9SWuOsuw}
zv4OG*l;foIwNjn~k6z`0f9vq!lDIGGEBPG{yN-4)6x@o;;m$g4J1QG!GcXq1Noe!>
z$J2Nf?Pc#<MSB^(MYMe}V{StgXCKndh-@LvOO1Le0dQEw&P9;b626RblIueJwPCw8
zr<`Cz0cl9QtB;;3#P16*d|wcWp)Yp)vX_YYf|3KpJZl?q3|ld!e`_M)y)f!|-U!+X
z9E*`TC#-X}`1nh;fQCB=wdOS**?*c)3FE9AJ!K1@goLDagcD;5=5Y5?&a-N9HA70m
zYn=6OXn)Li#`8y_lr&Jt8Q=k)u_j4oB%xZR|L;Y@DJAD4VO>Hc6IEk*sN6#s-W;FY
znGpTd6Gd@Tm<@!>fBPg&xi>2Y<ZX?fX<O~@XC_yHNHx5gKiF^D-V?9p4>qneeza$u
z(`P8PMzfrCmRe|nT3XBF`SVylo*5RNOQEwit7ZlUTpmo|EOus|XWSGR9yAj-Rmm+s
zs`}PF7g`>E*;HxVc2&&N8}eQCZTv!|!9Z09u1^>Tgf@PrfAQn7ai-D@$pVE=XlM7f
zn?94Oq(cOF#xW!cZEfio!xkULv_p(#dNd%}73bg`0uzf$>y9{PQu;t9>08?eSlN=(
z_Tlrn$%|SjR5Pu*eC2TE)3UHhp25*>F3-FT+>5O>Rl37i4dvgoJXB3RF=tb}uN1CK
zojzhrWm_UoP;#%z3h)x|nkwx}I4gmm&tO$UUuH@U1QkW;4ayb2!+K1T>IN)3ZOOyp
zW1nd@Pl)@-&GJkia>ar4-|MC?`K^fwMO&ii^_h17{{#OoN0*U_7Zd_BFqaX04HK7-
zfEOu$Tgj5+I1at%SLg?lO@aWIh>VP|UFC}CThHXJqffnErVi7W`TYS`QHxWwY}G+2
zTclv}5x@gLrX2nr<nZm~{PFVBRpVfVBV}Ru_-jzyAo;J7{Ls)~l@dm);p6-8>zkBd
z_&W22QfkeY<<E}q^!}=q;~vWMrQTmHD*mQ_6<?6<aOLZgKfCwW&>rQ)ckNr2Na+nY
z#TM;LFUgnTo`{;l?@+$X`)?mVzsNi##dp=*zka;@_hmjQbW(VuhTp%xga*_~Xa|Gx
zUP!{>pFfAcUVf<}ifAcZ_{4v|t;jUi2nROE{Pj0E{K1Vsb7$?{@W0sgd;J|b{Ac)o
zH~+wvU?*f0ta%0eXM*Jb!aJYEjv`3&gwWqFqH|;hgmIrsRg5&v#YVkziuFz!Lp0Gl
z);A!{vWb_-6}^*R-d~M`pd+k-#QK4(O~ks!Ybl9BYtScE`vpxSo%+VRO?uMIXrdbx
ztG$!jz>1Wxpq?hT4)@7&>v&NLYv7=N#YJR8yRo+HQ4rIe!rGe7b=y7Q4Isb^C-t*|
z@Nuw4YaxC0uEF<3eKu+CQO9l3ZOob<QBkP{LE%L06?Ve4ds-WCDJLA_L2v;GrJ=%Q
zi}~}#dW4q*Oi`csJpiT(gtV?;1-sFCTr&xoL_U$a!ZVhLkQ6nxhi5)fAoMzanB53%
zARmax(Lu<w<tnQy^0HA>lkmh-cvh9w=XDAA%Ytjf{#^P&S&@Z1k89C~=I&X4jwX2{
z{O5H30Gd(?Hr(m}X`R-JrbSm^^5ybn$GfCHtoF}o11p!f`q|v}Y`Jy3mV}i47;Yo$
z1bDL~H^_DpJyiBi*!ZH+b|yA|;nsXvkSEW{Qf?E5G=UHJe(BVY-MDPvZ6e(tVoGGJ
z(7MKnA%lpZTrwpD$8_Ct|JtaT|1UlrlLdxg5-9UPu}qygAg46C)Bo>AV*_MLp~NvL
zlM&kbngwR;`p`wE%#TBNHw+u_VhTNuffq^O7}f;S4?6rMyLi*M06%em6uzuMUJq_z
zP~9{l3~Z|MJtXJBE7)`e@$xHw*5P|6!n={!Kv<^`;~0cB3N&g%fxCc*8^F(F$cS;k
z-8I;s)3pot-S}+a4k;LUI(t7^ZXK`mOtf?mdn*uxdL=bIW&REN@%45})?_4|iR=FP
ze2x?P6o?#SPf`hTHG*A#Mz7VTY(vIw61$ku$~Z!KWjmSDM5-hfBgw{dNyZdj<fhct
z3G{wIHh^{^1UV>G;5<qgsI`(nlaXI(jB}<Hun2aehRbAlJ>?C7VsG(a|BSbgM4|Da
zQs><w`l`D@tomH+t?KT)jNDOlM-F1Fo7OpU)|`s$V^MP%>ZVeEdSR#Vq;i{CuZ^TP
zVh=z0g{)ZVNqp0cb{j)HQ(G^orRMEyXaP>XeVct-FUf-u3Sd!^=ZsaGQjwoY?Xe<W
zP(l_5aji0-EA*WPjr>~u^UvSI?8(2rjpJsC3d~%xQnTHN(wpOF9^$Nbbc5$6L0g+%
z*N%opCmhfmJ`US|$+>+S8Yn>k(DOGmbl7Iu@*baC`Q0KqN9Oo$58Bbtp20HJlgBVT
zm(il3x<Zm;bvJo*?Wf^r2}9^>N~NL}!)V%(H*L)RyUS?%XTE``wSuYa*@%)LX?f7W
zDr>1F#t0!kIL+b)*976097ovZbH*$7KXu}@dtMuO8Fqtz2gPjKc!4;&Dw!cr8G=)W
zz~#^9{8{JV;GFLzk2%HAE5;lp8tW`H?VjZZctirEI!M{b33RmtQjsay^(8;EO|V*!
zbGf~Je(r>7_nfw{V(+ftAt!vVh(6v4Vi}?z6l!Tu!l0QZGBqSRO@c0yg0AAB(Uulz
zTkkUop;Y;QR*GC<>qK5BLDf>0tC`KJtXMR2`6R|Aowvv*C`zb!X@fHk;-J{+v{YPr
z*%HF3ab)#vKjs`U-zZ}?^N^Io$8<JhXAlTpv%1hhL7)!1E~ZA+WX6a%5^$L!Y)<*^
zV9`3^Ih2}>0OWeG$k`R@<#-oCYY8Cl9uFwXmb(~#k33DrbC*OQVPP?5`DJy6pizK^
zd2fyP&jdkORK34TArxBx`;-VA&<ipP-389%l-d`8gOx%$_xuPPn13J2vhNhoIVz?)
zkLsA{8_07uGtxJ$6vS1&V-o<iapJjY?=0HxpVJmrj4wSz-*=1XR9Y5m=||Bw04LP0
z()W~qPHlZsNzJ{-y{KUFh01G{-?i^eDAU$jNTRxV&Ez4{Qy$rMnuAD>V@JVE(?xnT
zpKhG${ba{aIJ%$x@-4E<)HomKEwZEgDS}PL%EfpdA-hbUbBQ-dCsX5mhEHJ~_F+QH
zr?O61KL-b}jxv@*?MGbJy%IXdMMfBR5br2|<{fYRqDE0t#<i<N|LsMYr+gpp?4Q#X
zR$3UZA6mqFMRckyi$(uD-XS+QLEaj@#yd)JqoejbT$@ydoFNoMdUljGMo!U9VbA9C
zb9qNC)#_`{wc3BUe_5^W2X)zEQ68#LV{K!y-yrMgT1NEEZ<Iqcsb`sWVbp7FxxuJ^
zkw;a-ztWcUQ<k*5p6oo8Vrgjg9cQJoj4~rkh1L9SduDWwWF3v0om<TBewrbc6konI
z1zj`J>NuY&Z!yBfE{yDsV5Hr|JL<(;51iw)tZwSY>IddCJP2>&;PSUzwXPdOu|>~o
z7~y|Gye%j4Rp~=|+@EaCq6HAGZmKYU=~v4lMO8SrzZwl~?03rju8k38$nywnGOKZJ
z?IFwjKLfS3zGX&v;~266UTXjIv`5*a%`!GBDp|%QZODz^#<`ZAg7W6-NM&<$^t`&@
z)(9AnbybMWRC~$-7ySZ{SrhGsCW5EUSURZ)My=MCwL6*v-29nqag^@>Z~4q?Q~H@%
z?JGZRTvv6+b-$(dDz!bFY9M&uXdrm=)@-Ay@^%4^+Ua8pL2>6{+NNEb>+}F<nsaOR
z={I}mA<yJ~dRY7`6ABHZ8`uD_DV3I^2j}|Krj}GvbzNtyru)c+`}3WHo&N*-LHi<?
z0dEczm(g1`6qn<H4$J|tx3Gc^kqLi=CG)wGnOw+}+~D%|n$KZ|Y%p4+bmuvY7K)kD
zJp<kBE9m}}d{W3;uY}Xq$b17tTy`??FtHm<#IrQ&Gy4J$gV?tf62OXeuuEGZ<DSc&
zudFrm+|GXPD)zo;KYY{Iq!#pHHnk3#mo0BXUPaNR`a-4XUZ~vrDFuW!CEQ8V_nxV-
zvHEcgV8IZ+C+&%+*Cv)HN#^~?vGjhveXWg^{AtGc8ffN~>pq5d%}#g2_3Pvn))vlf
zg_jYRasQdlXkq^ckzK`AmjQ1M6aq3amoeTx6qo;FCnlGW69xpAU5yT)0rHprjSgdf
zfqY5<gnoMY^<!Xn{`~EBK<7o!$T`rmlGg<7odC1oH);glN4_JUXgBkrr58?Gj+ty^
zjSsDKlraN_^T<XLlN!Y&&k4bF08RQN8@A|$Aq?dJ*HM^Bh>)?+J7N|zj@Z)S{P;al
zAOh;TP648*+{9bP2pJQ+Z&wwP4F=DDF*8+3;SBv$<(Gu9x--XUl?c*tj50cxoyqYn
z^G|#$`5E(keX%k+ZZc4vjl<8N^oUvFx_X@z&Mo<RUzfD80-UcTJ+lCq^N@78!%66l
zvYgUF@;ltF#;CqE+|CH)0*5YE32rHy5?C|}RAG|fWgSX^K-<xqmC-_J%MdqzoLSMS
z48_M9I2jUm=IC$dt?}_STH-B4CRt_seCYRK_WhDSs~e{5|AbBV@o~yq!h>V7IN;(Y
z1`Xtzh#ubXJwq`eW<2(T_v(;+-3wwgSN_qVHz3qXfaX$cI2VwM*roZ+eXHR$(g>2t
zGu)=e{K#)==(#;7K3vXSSR817HHu=;!h$(of;8hadMXk<R5>@?{4VB)F=IdWyV2*=
zOizgqpE0@Z<%bR;P-JC1Hp(Dvya4YwbmTPKiWD*u!T2IG?dlHE{_-B=T2ZJ`2}%td
zd8qv>v(&0Awe^CoC5y+|%gK4=@TfMsY?FT<1o;>vqF{GS(7^~7mKfK6gMtoTkjOdo
zy0wj-Py%79bJTc0*>b6v+>{kmO2wl$ww#4Z4yW-m&XO29XE&U?Y7Sp@0h@@=YuNt2
zgh4nA{0J>yDu!A^_wpsuAGMMT>>Hh_SOe$rFn>l`f#}vw-_e<jqCQD4q;SBSVn2)y
z7Z}&Q$wZ?z9QX;&)ab>3Fj}O&<@R%@e#{ArhGRd}k6EGYg5T7SJ;;cyPAa$69AhKs
zfEQj){T5#O%BjZQLryiuC?|b47*2J6qn~d{vE+M3uOUT4@8F-%`7_KNI)8@oPG3-Z
zVE&7S473&APkrZ9Il3}Q*e`<`a}OER7-b$UZkTZJ`nP38AhcP3$fhMhg(BR_aXEAU
z>#d;gGTazP5jDXha?L}+8eu#FzU3j^<>r(vVJRoH;#{?4C4s7U^+;G)W?J8Nj!HpT
zqQa;hnF8gx4#UWX)eE`S&<&YD03nsGqL9zCyqr|Q?T+OP`w<!BSk5rECSf@W@zdd~
z?{@jKoMp>wc-WqQlmY2NBCtu>k=ky;?6i<!Z|ynhlVRz=U!ET<Fn6sz6$GJRA%z3A
zXUPN{W5Z&K#n#Y$iqGY2sA!9F3(5CZN$r>gW$o0MT8o)DHw+81pB>_~#xUu?<NS1X
z=KdWc=?1m_<XKBk$Dn5&<3bxqIm)@7le%KnHRjf&u2^h;4P9#ZbY))3)NbFndu;Aa
zYup_RdUM^Oo^j&C4-5Cz!w!3!b?h12aN6?j*oHyxJa=i%Iqk;aJcfIv8|Tg%kh%Ji
zdsYJ!)u|NX+}L$)@*;sj%S+$XW&%}dw^oN07LQU}!9um(RN_s6rW9+oUKw`VdymCj
zMZ3B69?P_U3f9=#VIxmH_TU<Y(@Z9<EqESU*YbPI#coMpW;O}LTVdQ-0|42)0ab0`
zX3K<Vzw5Ke{Tiq#wnAv`QW$g!kLRZ>1hO7#dd-Ev*xE7ZYKXakHfZ5sAsP9`N~Xgr
z2F~uj6~p4t-Bt`M;_Sa-;P!WX#lSJ%Uomix>pc~J-JWP-xHE3)nB`heIDs1Vg~?u;
zGua3qAEBCcaMkAB<fGczxL`^H+k;IG;f(5$zBR<nkmnd%T}YG;7!HlIB$C4Hwb{=b
z4+6A#!c>uAZY9AyEo0DAS+#w7%P^(sEooTe_!u7>3~i!NbhD&j-tt3&?hu7Co?4?|
z$0>AwdI0j$9&qI#&-L0~2^wbvSbWkJgJJJ5oo5(}Y}(I*O+~VOI2Po%8xphBS4#V&
zeFnqcLt-_?;t?s-hgn*1dB!=Vfm?gA#zQ(o#Y4GsF9*kGXl&p-nK<)sn@`%z!MTLx
zC%-)0#qeJ=EPx6f1ownJ1jF2655X`d?cZ&GhW6|rio6-E^_ggDIlX`I^q?V}R);4%
z2gNWycn*p|BjaQHgM&q);{eQ8Pr>c#Qs;fj^r+!wR?}_oQ}Sw^oxPF6O_Y@J9J)k}
z9m5E&A-p?`;B<EkLGt5_obLK4%EU;$FbsK_Yx3Yy$iz|u>?ajK`W6bYg91QO*&xF#
zLJ;x)DoniA=~McQ|H5%x5#_1Ljr3#i23XZdX&cx6g@V6dKTiJxbowh3mjQ1M6qiwQ
z3KN&B{0J$RP?Zj#e|i4n{>w>Mllb{|KW$Ah-l(Ac^ymBO>q$k(v@S0=&6~fT&`g@-
zCoX>Z`5n^q`p@ZCk!x>!*9%|v(%tq6n2j+iYk}F?nT&J@t@kql>kxK_KzKyL6A%Gs
zJwoM+6wa(5`nY}-Sr^M0pb_p&;$SZzqDwsz&GbMxK2p?xe^XJXt)hDQU{^r%!giOB
zqW*fbVSSvaEi>y^o`DT>Vrt;k+d2xHR^EBg=fa>3ET+PO40KC*qHH}wq(YHgPo@2_
zN*Y3lwF-RR3}l6$jbV3%>LOFmuUflib7Y(8%T}FmpL^@{S{epFrkzG9K&xP3L@3g(
zgyib29r-}3x9*k>^a2EYo1h<;0dEcz0x>w30i+%_m){2)G`D7&4yCb|f$BXI0Xdfe
zdp#$YMobT^0S}k+Ob=lKG@Z9sO%H4e1U2#rcb5Tg4if`1GBB5st_m&zGM7O&8Yh3%
zT5V4oNfQ3fujr3^UwXQ`s^3<tNZDlBi*zpta?;)*gp3)$8{^^ZLCNZL_uKcWwg;Hl
zfU$wwAVsN}sqU(-m#3;`OwN)rikxL-iG)bX1s1|9G7f}`imec2mB(Y;A>}y=!79Wo
zgjdPRa+g)+xC3URa*~~0P}aDz9x{KdmkkK_92bf}r2vXOG{{1E@cAN|BB=R_q--R4
zN?F<DGeodaHYd@Qt#a0gM(mUmB|t^x$m7yx5Xmc18_z{ch;|&jP<ZIeN^d~1tn}E;
zXA23kJdkL@3@VDO0KROjP-3CERBURzJoUI|qiF1*EE$pJ!NO5i$WcN$MZkZi$=V&U
zE~O$ifH&o`F?Pzu8y*jG6h(u0!Ir04v70LpK#_>W?o1-bMN-T-yJfhPIe8o-8pYm~
zr&tk0kb`eH29b+XaV)Na-kgId52t!lWI(t#TvmXx#oJKQ%O&oCf=eJHImATKJ|Q}S
zm;F*wKDojdRWJsYLktB&A|iifP(^GuD%cVsaAvS2`ZB2i;|k?LWdjk}sWA6n#&*FK
zKI{^4QWEA6z3hS@LR6M}PDSIaP^@AxAN&N0%To&~6*ol{5ve<%vWg|z$r?~{3^D;!
zF84^qAc+AVs~F49*fd5ePqA~Bcpz*51!S!=Dw(ni4wx+C9E5;OB}RYXL6t-eP`Jls
zEe@F26^a!CU8K3BQuGqX$g@~fk&H{?d;F=Cvxq<%WRt_r4?k2Nf4i!yx06ZJRtMK7
ztz7+OH2I{eADiiEJv}tYdUI6$x%yDOJGAqQ{d(9QA_cX$0QsQ_PGx7cNJKnqo83|M
zR$=>t>gVQTQ@vN;A;N#y<b+tZCc`n_@}J(|Z!0C(dyP!$bU*-V*zR5(JOV^ugFWyC
zsZQobh4+U*vVg7kt6%>4FCbCS327s)n_Q2_M}<pO@0xKlJ-8YSYbAKNud2O4+t$;G
za`VfdzO+9dw1c)*c79{GnY7q{S3u1mZbiZ_(!mNC;Ns1jYHxqq3=it|u-bdSTYap*
zv`2T$vNt%dA$HQ%lXfOBnaURX^{ly`4(r(*Ipx-0>(kNT$L32hS16$c6YLG9SP8HJ
zFbG#M^K6dNMf}cT0cugU0QG|KLL6=79e{uaet8$PN6Z5WbPybXvqD^Qjt$uv<WEH4
zq7G07IBXvwHadR@N0wODf<nM9J;!m@C}gjs1sl>N$iC15*or_6_iJH*$N=hK2^gI`
zuru^R)y11o#kTz@{h9#<53vkThW(|;y=**JyC`1P0;kFa@CG$HZ4&9^mm-~L{YbyH
z1@SkK*Bq8k%e=;Eou__Yd%n$Uii>on)Dd|>u;~cf5pI7)%CC-XI91s$!-`4KYar87
zM#E&Cj%YDBIn5~_?eub`dV90|upKoM1pPnuKghrDF532L_I*{=6a9JgX>=vW*3I<1
zlB?=(&15{9)Pw2q)wDUE4lXZ8lXK+6&X%c3IRlQ6ipFYGU>t+A^k!D0!Yw7{tS(w>
z($7%6Y%+g3lVB1O6!jaf)0PD&BP{pQ!aeDL<|niE(H5`eAhGOp;Cn@f=M;@&7f=iy
zW>6_W0eV2Pe?G~<mj93qTH2y$fy&_l(bNN9_K$$?9fl5k10dpYE;L0N?a>3A-WeT$
zISkeL!98y{ZAMhNbvV&JVNhB_Pm&tG4kN}!EmMEN^I&bSvh;u#?S;*D-Wu&<ZgDb1
z%r624I&0?_QF=um7f+;l_1zkW4cuZ4dTwWE!4{)zy$~8ij<cHpe6ULZ@AD=AAN(!A
zbJw_e-w~lBI_#exam!YYv&KFcp)?z@PS#mQDnYQEAIO~W3`q;w2_s@~5l{u+5&#gD
zHc)?*3FHCcd4=%D3mQtJmSF-}!4e5Lqmh}GjK=Wt^m=waYA>!&^su?C&ZhP0bo8nI
zG?mqec?7xsb-b#Kt)XWkozX;gin2{Qx(yU<uiIui8jNSvq${%3Y;<`wu8-?41Nncp
zxso0cBX4{F_GQ!--^r^C_>$~%H3$36te$_IAR9j&pIuMN+FNx6nu2+x1zhG@AaWoW
zHLbzPRwfNzgA)xZT8<ZKKzj(;>&mxi#KZE)8N&L=b;lMDOfAo5f;$J@sv(bV>D19|
za_VT;iM<}zqIhyZH(<Y+fi1>mi}_6|Tr#&6F7ifG36YZy#aR?6q#D~;UIRE_rT2gC
z>A3Ux^Jjg2J#9`#1;|gE;q@hy9~AbiP1PY~_3z=;@$h)4FWSp-AB19fq;HI2MGSZV
zWq9!z=w;2FfOu*Qq~$R-Z8XN_JFllwH=-uj5nbPt(njI2sPTj@T86%^#uHi%ru4_F
z@tvYwjeoMDt7`mJGi}eB@u=BYJ@kLl7~RlEM?@*gx`$cICy?Z|jcKDab9yxUt=8u-
z+welyr`Oeg{`cS6;Nr4At>j4vAy=EJ+r6rADC*^N4TfDppYJg9J4dgfuB3VL>lB2v
zV(dy7(15a!b2^VPKJprzzymF>eZLie$-D$G*=<smGrdAH(i<xum`i&zW|@CJ!>Umf
znPseB8hM{=!{vr9Z`9D`xK1SJ74M>B(9ym~)b(El^z7d=`CY^IOx}0X?ZE$)o9=71
zpXdzv^!t@F<V)Yvogp7FA8aq~GTdZ;lDZPVmFy>zmW>rQGS)hg+vi$X42`8Fe$A%*
zsYbF(5Gfy;A2$*$K7e*Tn0$Y{?s_nq=grd_t&I5@eQj{8S$_H8v$jD#+Osz7pX&v{
zPiJSN;RvTVKAXZ2qk7saxny4h?6&{`XO~@;*LT_NL7n+OeCyg~Y9zNbHInn2L~EBU
z60O|_ZF0QLd|-*Rm7o4B<^#)4$3%a;`Jhv@n-89>=&JcZ-lx^=@eoM9nl|H}jwRF{
z8E5Y@+@L4wK0NoDuf7nEzx@5=S6B;6K3vKV)9C)T2^@&ecMk<DEyF8q1h0HM;-b_n
z0(SnhjHBNHiUT*75w0{712Z!;m*Fx76PF5M57YrUx8Gq8?g;}0PvDo)XfzW8Gcz@p
zA%X}Km-<ddD3>th1O$KNHV(blSL6>gO%Mc^h>nV=Ejz-u@$}8%lV|M9^kICN-yd+5
zNJ=TCO6h}g5h;*-#6|#;mp?D^^6lyM`_oTX)E6rpfmN69e_VhYB!7bBn}#kHKp2F}
z_aB#Eze=eLU$1;=DG~Xy{M+$2J%2%P+ygvc@chMU;C~2wsq}xXu6(`a-|qQKXpeN_
zv-WGINa+dn5?j!hUXm}vJ+(#(<50fdo`3uP^OMX=0{({C_s93A|33ALGGqiwefj;z
zQ)ti#0-d@j6ujKaU;n=R@$`>Mp-`5>g<t&nZHc5c!YLa_{`#A|{EHiZ=FaHd<$tm3
zk2;LJ{O9sd9$<fquoE&8M&1B_i?F;X;hoQ9ClZ8SNbT<@v2$id3FB^41t~oBWFy--
zV78OS5H)Pa?4}5rHVMzfGu!Ex=PyPo{v}_-Kfajh2=X_xfo8_iCiDtyrI`t-^k9}H
zn|hE6_X;g~$#+BRYq<-$w#!UEwx_2x|CX8MX6Q6*R&jqn3eX1b<c0C@DUP_8#7-e#
zvf8U7LeNXw4%K9o6-ukZBbawdyP}H1UeR0T0-g8sEpbHnudhY%#9h{jXJXHWU-&YW
zW@x2n6}imS=id@_!az0ClM|OldV#iW=uA}2e5}+za<$I1GmX&3$~AGHH*qWY>s$A4
zQR$IVf|-B2S9+dx19L=HdL7k00?W)m`eGyr-h(QzHp6d|$kJ<`wC2a^$6tS5x;wvo
zo92y5fHFO0cn?>j^Cj~#KQ03eMTu&2hpGN_^>T6kn)@R4qoB_lRyIS$ZJ&?2iij}+
zjOr=MYgH@uek`mLUeTkma8_s>_EJ>_+_ip2N`Qa2!}{OuUbSL{XXU^FN_G}M_Xbu6
zFkqCpk8gZrF_ibuLo31V%z*yNJSdp3zVkGN`-eB01s6AlzmK95d<_)!@`9`qtULx0
zm&Z2Mk>G0>TP=^x>dE?^CDjMckTjMDVSFI)cw+eZow*ogWksG5)w(hi?$yAHa|E)&
zu+M+eQH_izC$Ax@lWS1~(K(*;LzGXFNmAn~nmaedR!J>1K|g+)_1!$OD%M%Nx>10R
zbD;vESvU0*-<LT-k+5pmyJ+?09M#eE@PCzjgSiqu#=V&zdk&JTY6+zgM7DpT%!>SB
zy@2fhxPk3Hp%#!Wu5`oVR@4F_AEM9=Y&U;b3kO8gA<u!%7R{EbOx{TC6@X+PlWg)h
z$`&mP@8XjN*yh{Hl;;R`@l87FCkr+ZMi0Fk<r9KkQST@&3pQ{sFYSQgB-o3H6MHk#
zt^1pkculI|pwJiZs~hu%(1Qm)`MLgyI5U@LW{*1rMHRQyKjRP-Roqe3w^ebs2M2#V
zE7U@v+E_{NhxPlgpy-bxn_Zm~k+nuxYxg4SST7=s$TnPjg6RMXir~a`B6xuccCHf@
z3bM4g&bLD{imWTi<1qfRTZ!5Wki$M#WaV*WtrTeWnaEl#jB<OCWnJYQo*<jZr$O|v
zQ#R3)81zR7)SV=mnsttGa3Zc+dEtN2?8KGN#eAw=W!TP06=w#$P}&bi-~(=v;uD*P
zv)!VW>fe82X~%!miKQLoj1x;oapT0&_TGPDA$jJBrQ@K{dAUnDlv7QI#ENnVTnwu3
zeS#}gk1RLmaUaEetm!DOsM}(8I!->IZi`9k#$7RBL4rpCGD#i5$5M#6#_fM@ht;)L
z!qXI1hpc8^O0YVnO={>nw&5&fo`dZ4J`csy5S$+sO%oc2CRlFD)A{$uKsN^M#~Hek
z3{ta%kgZ{RU7yIxG^jgrV*;Rv_`N}Q&4(*(-H>VIR!^S}x-)jd!W3h?xfL?0oi!c*
zwrIgLd0}aaT5P}+rP<^*nP-2XlU)j#n0EAM_&Q8r8H-(6UB%#P-$M=F3&|@ul-1R|
zQAp<etgcq<PHg1{Z|CstCb4s7fb8M;$E0;N7~Z0K#+UEKxZ#|!d`asT0|~WX%thdw
z)%T@hkFRP2EvZ7V|17d~y62KDvt@Tanu~|*EqcvoI=`~Fc1o;9<zRpI?c4DgZy@2A
z2&%J@U}b3(6GuL=RRc<yowHtyEN-jsXIkSFTC@V)aD8z;Hqkn~stvRZAMPza2ra9G
zF|B)?i#n<SgQc{5jfX7rSHW)59oJl0WKu;@tVas>-jd-8d>Y%YDgPVln$%AHavY%z
z<jD!0lJ*Zk-hi-jVpM-QW7JY2kv*wdSq)|U<@{h$e7k&SQJc@as4X-PuX+PbZTZ|u
z4?>gJ8PikVnU<|u;WR3~tjYJ?%pnU$J=%%Y;Wcewg~?Q(ZQ9PJTj$HCZNdkYqamNO
zoTAaTDKkH><np4KE}CTLw(C4EZ=;(ALxvI{y}t<UI4&D_Lo$E&kC7z^{&&s9g}TsE
z<X4Uq`nsk6S-o`zpeS{nL@+p$in1XnqKVlE<U&>4YXA<zu>rG;zQkitmPYtCk5UTY
zw$arN%-@IZQ50^2t9)`j9)hk45%DSb8<wpD55KpSWs#?SqbYpZgS;Qy!l1g&B@C5-
zd}00_%qtsf$YGb6a}OK~dmj>PhTZ4O<$Euez;h2Ee@bO%iYcxg$z~^#o#)o$>)F6b
zCr9Kl;zUZo6iyty*LfX9wZ*RSUCbn9?4f*N`Dm>~)Kidkl}-98eq(x&o3c*Z3G`t=
zHh{+DfjkCi4Faa9s-jtxw7kpE1i<QB3U12rE!C(G>MEeU)5acO@fMO=IQywk=iMas
zMRx;Pe{-AciLWatsf1ROwertaeQo0#@zsHA4wcuJ6WVM>dBq@ZyIsz&*av1bj25Vd
zw+KKPLCW0DXzWQDM<7?q;c`xbb&3F2f!ui4XIII7$i3--atrm8tE^e}_obHvW|@wZ
ziYqewk+(8SdI+slb8<&Mm-sv<hq*?b3e(#me|;`GSA2I*)kEYbiWGA}9vCmwl!~J1
z)hQ~rbLVE{;)IqUc6)IExl+d05?%oe917s`T(_)l7JVV9W<ABv5Q9DQHx9h}QW&zO
z)Fq+stI14No;}6)n*}0WZ#-H*$#=jExU7s_wyEizOa8}Fn+&<O0%w`f(Ure@()l%>
ze~$7RyObriy^9u@CFT2xhjIgI%fhdD@>Db9_ah!G8}!tNf)9(S>?nRjip`y07UZF&
z7^9qnHwZItFqTKr%>tR8=JF16XY@4$SKhK~-UYSCfoiD_Kz%|#7T~)1&pFgg9J={(
zRjQoosI5Q-+!<>W?IZ(KS-qvcVSp+OJ?&E~)!iUN*@QjQOl9Z6D?u>T?{%EBWtuV(
zM!=kDnDU#BbFynN22lx<n{{0=N-I8C_su@u{{hPm*PxdHZw?ffVVEWpmp^t7&VSg`
zU$4LKbls9x(WS5HHvfA6y8F7DH|TUwCgA?>FT3;tEhrld?^}7^_P2lbpLQQ-sXcrF
z7(VSB(Ey)e>5QmzQQ7pWz(>3bWEhAyNDLj;JMm_aH;FHNF`Ve?b5q_C!=@Q_x|BJ1
zIVJSq{psMeIhFxQa27&U$gLowkbiY*AKyOjrzd}WXrLH$D)JsDwT+%I_LiqY{m2rB
zC%r@>MH**`nu=937zv1L-Eh*|7yUI2t2T)-D2;I?F+o&zyfOj>WqDU`Yy{H3*a)NA
zZ()Q{VlqMstE4Zva@_+xfa1ViNq7J{dWF`%Iq2~hgFdSL7NCz3Gw4zDEPoqKn{E+s
z{l4c<5+4}BI)y&GnGu4_`Nc*U)qV>jj1sdE5&)OF!B8W>KZt;+Xf5cPVc5jp@X`=v
zq`;G)%D9G%<BviF<dq{kOP?TWqn%UWR=ViRkFV?b3AkAjY=Z?JmGiJ#_s3lVf>q$-
zv+ITw8Ms<^BSJtk>EV%pihsI%Ru$YT_i|?%F_Q=rNwHHfhxocvo)=E9+m}o&%G|Fz
z_OkpC>z!Y|`|S&leEByKjXLh*YHJ5s>#!`A2?&LKP!zqhTl~#9rZ>KAFmbOGoGG2M
z_c5DFoK0Lmzp?4}fbLz6XxKIsIc%c}5)zj?@dMT<&+~`DnLB1(Y=2#yY9`e6H&i_-
zB^)|0d6Fp*KTXLe^Mxzei|ma~CO==#k2OWzYA(xRFK;nQ#uLd@Qv`-K#?>*<p@y<f
zd6w8OnZ!{OQ_iE$ZGO$F%gWp9P+dnfZ5%1D-D{KQRSoc5BpcH-xc1{S%2-ogm6&%F
zy8J{rRu0x_L0newV}AsRL9JKOG#=JQ+s>*`cbi#$UNuU;u;fE|vc0RkcBR`ZPc$aw
z8$at}rN#-#xl}3J{%nxR4cCuJ3ZuE{(Bl2@wClH#t@vb%$0(k0tvkl5bG1+L+0%h<
z;@4QMhO=)Jq{~PtYiX7GgkuAaW0mi59zCJyc3nI?&pa=jVSo0_s$&;Rb>(iyqD$Rj
z^|3Sp^kjQ4I#A_Er8^j;mY4^3ET~BgI+wgD5!@U}zIk35d2N!{wwMqGz3uthF&|r+
zyA3<otV7{a?N?_N>n06K`QafpaQ(65g5$f3+LSuE`KGUz%n^Y{OuAWSZ0L`j*(5Yy
zDm(&hb+xK0b$>1qwt11WxzJ4uHkav2OP1|xg7kDOq|#syRWZB(!)xZ0L6;>cgv90+
z@%At+SeIIJS{bVK5)91y#ql8Nryr&W<g)5Lt|XopQkQiAkw9+0^KC!W1)HRz$H@%2
zWLt!J4X_D0zCvQ%DTwup@b9rhy1T2{fgz&O3rkw){Dvr&BJ_XgE+6p3`O{5elg>Qx
zfW}HeaxM8c$L~{9D-_sD!bXXQuEggg8Ct5&q$nyh4B-|j0>|Y78YwDARt4Fl?6I6c
z<qGCIJ|dxkVP*oghbk4jjp+GAd4lrIx}_SU<bw7FFH6<W8r=F3Bra9Y&;|{zZPZhJ
z-D7(ST*&Sca8-Z1R;Fw^n=ZQxY9)}*a@^2Nd+y8_rn(YpD8Z07rAnTEI$Ih96_Uq}
zJ4`2)jc0D?l$<4FdfQHkJ}MpJES1yVPwN{{MKCLU1@m)7M$Q#b@bhz>gV3=|>O;zn
zAvo9yAE#Sxk#)4|eGh2ePp#b6aL+i*=8!iJ%MSu4WD|exV4V}eBhO~u7|7i9Y<$YI
z%c$P7%M&wkoQ~b)9o_nP-O49<CUqTv!~;rcS>_u_qfoPgbW*}&TK+KcgpEg1pSf!0
zF&%#2$qZbg;?if&_#wF`@I&&)<<yt-y2F~|ugLbtMc>q6b*q{7zAL*qpEo4i2y8QZ
zZXDKmwUdAU{k$sDdmh~8zY*7C4V2M-WrdB}Qr3P@TQ*~hNCmU9a_SB-Lg#@+dbihw
zi}(G6A96R(qT|5y2Z~ZXjHyI)Ev$#OIF91#VL&ReWTh;cB_X!4p=)o-NV^z6OYhqo
zy|iNEN~>9v4@HHVrb*=_Ak6*V_M+6!9ZozXtK@%q*vG?Jx_c}5cM^*fQ~$La`tqe~
zsv{say-e<r{;zM|&F*BB!O5JpU%o$wa&{gkSzCJoT3oz8%xGyjZ_LU;l3$WVhOiyJ
zbZ)!lfWa!gklBa4FxN=4%Hy4lZ$n*$zB-%k2)dWNI{C?~*fkfDH_y0z8EI>r*7LF_
zT>mwZwb)^mvvvY(Iuf}*nb<vWp1j~CwXR?+P3vXAgnQib7c?|bX|aBHJU9HCu$=1t
z2N14HmvL1$69O|cm!YKzYqvFi52cEip^_jI0yHw00m>IAmlg>St$zYFitc`S``wMq
zEPeC#X3c`rMnqE6`^TF<CT#q6O1p2QFfrKae^Xi(%HY3A3h&6{5yDs(-1G?>pZ({<
zZ-0INP6>ehm$%A^gtjV%>2dyOtq9VGIQ^bK2cazRPD(^jO_Rf_jo45CmxcG!4QUVT
zHdL#>^J)|nNb3sND1Tvs$HoOgwNz-}CoZc6-|zKeq3VJn@wAY^pZF;mmOlc%K=DU@
z)t}2&Kz*EUew%LIWT{%`oX3_Bw6S=1J`in5RS8lahE_?oEEzANkK~F}qL<e0tk(Q4
zi&{XH1YzkQJS>LwlCdp3mz9d|vV>=qCzPe~zFyyv!-`j|EPqC7g^hr7!o|oYs<L!v
z319yG{xQRS0MxhD`}~|+1uG}(rd#JBW#UIA_qJivz4Pw8IeRyIGr>cxNj_C@Z249T
z4`gqqik6wF+Y483h9!Y`V=WCxqENLV0b>P4yNF;X!W9<C<Gm5VeHH_U8_&VUV%3KH
zj^Jc1@0ddvA%8g0L)m$ObLqSCF92l7JDs5LG{iwkF0b?VhMzP)gV*`VjFI<}UucQ*
z3e#X+#>h`l3od-3xO^%cg*i~dLg9>%BztZ!*048jtp0Ejt%0aO-C9)kj;NqxLcX~(
zF0*DAa&=))W$|SJB&dfbb}#6H5#IO}I?$hzguTy_sedIF=A>gI@8Bd;@n7T<E?Sw%
z(DPG^TvaChWi2_Ymrt$`optak!@SJSvNEe|gqixpeilQ~alx4B$qAHsD%2%>FSETB
zo?S_o^D_#Zem&c0N|gM+hvkX21m+c)pZ986d6`#KR^Gf?Brtz+*_tR@U$bwvv6W=g
zw}saQK7aE}K51<)eXj($6Jsv%VvaHYp4x4>2)M1PfPu2dNhhQXS5cvnV3Mg{dEjiL
z#RFH;Syr+Smq?S`=iy3bZuyj?rp&#rtY(ou*+!FB*eG5sH(9UF`o36jrQyYWH#pJ$
zPPtr=aoNZF)$63eXAj;_NVraOKDkI#YeJ5kK!24WXOcjl{Nw2i;q*2H6ps?jI_Sd8
z$?3_*r-w*T+C(2(kiQ&bAq}&l7Ua(iqU<(E6oy*DAPmPM;#U#G4nk*yHm*hZGwRr<
zx7I%S`j|g$A;gFpPr*F(s1&uglLgV*n1ar@F+tW)JwX~|d5~rSX=*vbng6)C+NOlp
z$$wve`bAFv!gt?IsS-ya&O09`D^x(qh_dU)n>Vww?qd$ogIJ9R#2NLcV-+%bqZ|HS
z%q<Ojg3}!pIuLKxGKC5cae5Lm+BMEmvk9nkS@kVWlE-?C1S4}T)U!h}jYJW<QvJAS
zjB@bruqIRMr?=X7t=mE6@GFDzPSKckbAP1`)uVnl!B*&oth<I9vu;B|Y3f=xT8`lD
zm$hz<F3PG?)?M-U8P?s(bkw>(4`+x%2D`=y>%KK3LD4FhjYM^-!l>CGd9@&aj@e8N
zve=A0Of5CY&nOs<X|r6(04NG?Gtlb|OT#uYFf|G8Wjr!9G`lorYWbNI*%7mqKYt$c
zh-_wA?dN#(fU+xLemJ(T0H$;caw2*n4r;B}WDX`>g67ei`yiYhPP&9l)PPhjbZlY`
zR4*)fp`D*7Sg1S)hwpde5^6`GNFN!CLV1j;WH2!CGd*gNBGSiB<>XNPwG3{>h#m33
zXo6bH)b=dRqhg{fP0+C>(0K;5oPTRv3Rz6+Ctn>i84>)W>vFl5K#{S7+ch2J)-5Ql
zxpRkt5jE7329IT9c<d)lYdjlC?Xs_HJ;PazpXropt>@EWs4+b0yLPHhs4*x+6l<eq
z^1}v`ye~=c3o==rSu^?Pu_94Eqm5nZQ60007_3&tI+ilYBHhZtkcQ%#XMe>OeViYz
zKTAlfq2`cwNjlt3A%r;|o@QFqeh34}3Jv6eevA_u8b1-jh+DBZ7Q#TT=73VcoTzh9
zC2=>_U|hWW5`I)Fgd3@S-${?~CnUHUWGg*xtBGs=VBK1|o!DwndZb}@3)P9OMgdp6
zf?CJc9h4@f%C<3<osgwW7=N)AG#$pf(U_GUS}X6k5;q2p=0DR|jOBBawXH0@%??Hz
zz|YzaMlmAAnkzHtRQ?8)S<*?z^mb69GQ-+#T-nBnbOcq<XW}H~QT~TD&ZbGNt3Q&U
z&!A+b9K-)Ou8B*~t3mL18X81xBQlfYu?n|FH3$}wtH|gymro#n>wk0?h<KbVYAhvL
z9M3`t@6s+toF<3WtXD0wUVk-x^yPdctOox+Po>pmu078|7pVJcHtsiPmAamOf6k0`
zdj?w06Uu~oV@Jr}yxf@B9=jQ;cHane(ezHx45$iO(>2I(YhWS@9xgt%ux#6t;mVb5
zoqI}`yYP31SV*$_+J7@x2%W8j*ELu^9hx91D`Hq%U^%BCzioA8V!m%-i+xZz%gg$<
z7yVb$cWNyCxLT5sWJD=~d{su09}judh69aiK$c_0%J}mU<a`uR7?=0?5Hj8^f{?E|
zrAS~RS`xp0dP^EHunPLy)N(Pd(f2phDsykL5OZ*-4iHS`#DDjVNLAk7IHkP59CuNl
z8`6AMXx7}ckz6A5VGE*uu=Z*a-cd^h^`bgIDa0SBW<J$h&WHz86OF*$mQtuj2dV4a
zxTl&R$XpT3l75{Dm6v1eSr@eposc>jDuGsLMQ1B?B`k_@F5IG0S4B)m=oM56$u#N<
zC-~683FTWNA%Ac5Z!B+Rx~>qTxgvJcw6D+73Pa^YH6NjWrW+{qfclD)wMNh}qM9YZ
z`?e|VsG|a@GwVZj_+QlWs^*{$OG;nP7a5}K3l@PUwB)icSnuj)V_!F(J*lOfY^9%X
z>I&t{1S6@_SxrOp^PwUOtP1jR>%dvh$LS}_s0&m{m4DTWuMV(Xj?wp*YbBRwnCkd=
zNrCQ?AoW5EmtymNdqw+pDV9TH!(m-eX4pbsD26s4gO?-l44aR@_b3=S0?)9NV+&pG
ztQ6x;v@e|NY#`|%yf&X#+0gHoXO<nsNs{?jN9Q{(fM&;7joRG_(!M?t*qys5lx48{
z24ppN;eW;#?Ilm}RWZ`7xl7rH#xf#E9<<W1_MqBMdjF&^J(D7`LvCF3apQs>$0ETu
z9KAnokBk!`2?2x|-O6pR2_w{|s>Sl-(_70Wk<rT$QP{-il#o%UDYvB|rW}{RD|85M
zV3d}^dmEX72>EhU!`$&O&Y<Uue$yy+XReS3#eWu}8oBP^v07-eg6fQtEXv81g+K2g
z!D0w4h?ZU+ciE|tinNw^;~3~mWc116sI<u6kGWe7vg2;Z%sf8sx|?CWhA-=Ge0@aG
zDR-+m8gw@!MYOA_<^Ieg$k1XLb~ji+&@tg<CmQlMO2tHjkNu6D8uhoDJ2mEStySD=
z?|+~44=o4<LIe4H%;{<nsUsp<$04*(+OI{@Is{mEkeWhUqE&lZ+i{2*q;r%B<5aGN
zK%>yVr^NH!2I^VIvQg`kZ9zUATN**<i8i06jw6;vR+zA22bSj1r1sfbET<LU8Oy05
z2Q8-tQ6#;yCLU*pNWw_7)|ouMeZwOo?SEFHQSwj&r1<<kwi;0Wcr>N8a4xP@&{{Sg
zC*7n_bxZG8ws2_K4aksXHxNH`La7#BTYw<6g&b^t?RY{IkWP!`tB#FwDrzA<!bY8y
z>d8NcqG3beSoN_I+zFd*AjfRF0g*n~PB8Sa2U@9WBN%FI34DyT4EK#5vzi)e%ztY3
z@N7paWG92ZUj><t>>Iu8oE#z^5~&8!i}|-#o_ilUzLa4VoNUF|Pshh2s94FUHD|pm
zLry1St${k1ft?qq#v6!EYVhdWrQSxe<Eeb;PP!HNr|#tvKCNl6{BlfKuqbs_B1;1*
zDMvlECUC57**4arK4DL#9Ve}!hJX5m24bvFXh58@Q6DoDI5UJRj*i{jM4#Hn25L!d
z$a$N`A%HP#4o=9+gjDttq7f+yRY1oT2_JfTvo@hNwCQx2R-r5PYB=P8equ>+ZC$R$
z%CTrzBN|Et9N+kMP6PS8lNi~lV+#UB8Xvj=$H$d4xL5FMI6KxN(PKV$T7M)<b{rSU
z;&?TPgH~kdI0dC2KCBF+>lAa9*g$oiVy+RDLyu?luu~w$Mbko=j#u!AR<E_(x`mpH
zF&<W#IFuFGCme%>bdYnVuB@?S1JEXPFAYl}oo}JJl#2lG9Qv}c##6K|#=1eP^8P~W
zwiQ7;h8@c~SYg)Ee>U#$@PF0K7}Gv7y&HK}4PD_W&#Wo$R&U%R2fGHcy_<&-cEim@
z4fee2D#Ss$27%S9t_`y^8mQyj?4+=DgC%wBU+JfKtPyj$X=siQ`?a5oy&B|s2TPm-
zZw(?ly`BlMr7QbiyRqZq&hRun4o$SAMc)r~7?loZPD}zDEMtX!4S(^jTjlfcvZ1SY
z5fk-{YLR@^y;?54K1CPuis8Ic({)^HqTWd7^$nK4cVeAOL&F9nozmW#l%Sm&iFv5-
zT0-BA#k_IFmZEse;C2cJ#Nd!@bXc)`VVKyH$2Dt^>xld*!uVQ+J*8RWt<dU}W)0tt
z)mxhFM`wLX0udm}o_~@6N_)i>cP)Y5pi3jj`ZE$BsnXABb|tVKF(p#BWGJ7~#|Uj4
zr$Rg<G~!?IT6IQf{D(H8TLsW#Z4%!#;J=u1LTK2^1;R<8nV7EBHA3qYi*RVOj-u-<
zBj*svgUYdE6TU$Q-6{TQuRC+NTjLl@F%`gZ2c!8XAI7y*^?x#qLxyQN?`0S?uR)(7
z%gS`8u_3FMU~KyBMUAAFUq}Sbt`$T#2=OU8v2%=b?*YNZ#W?rrw@k6+UUz3risBjW
z`8>#mUGBFj3qe>1AgzRsV+SOB$IY~O_*#$ix)+H{cRg;KOY&|!X6{$Un(GR1u(z%t
z(3xo!I+)79seh5jp>8GmBGwdg23mw6<)WY^ZWC*ezHOIyQ$rQ;g-&Bi5^T{~jED?Y
zt)QNDe>O|etq~cOe?iQD_`?laRzwpa-KkM&E1Ho}X-ntmsI-O}8<o}&gQL<KL`97Z
z>FJOsu5_f;S4{P45I}Ji*5K<5hV=Z~3I$76ibe-RsDDFrvcd|g(^MNR+*-5Oxn3HY
z_)1ARU;Lpv7qaMf5S2~2eDdp%@6|ZUR7ggtWB1s^rU7vcpDsKEJLn8Gtz$FH4O+)W
zm^I*7l&vAO(!uF4c4S=x4xd=pKs`In2i8c8(b!=#MPF}i17x~2dDM2L-t}!Jg_P{(
zbE<s9W?RJMYKUTTF3Rq;2Z@|q!}6_%adUHg0_#UQg<@_kodcn_1{jXKHPBGttpUd3
zZViF3qr9d#6n0k(jYZuS+fdN;T6k+u#@rgn4u#xn$(26~R{eRf{{vWrc9+331rq}^
zGdGvvG6oX?IhR2<8Yh>}84#?0zx$s*{;-o(B>H{-aW{(M7UQGvDAK3b-S0aA+C%Qp
z|H2PM{`*cLOCsXG{K;uUUEcluRm<i1pSzdV`=!nGNRvIX**4UyAqXAyVw@5IbW?NO
zdyT&a)n!l-7FBHRUgx$z>ME+n2i2pV_m4)Kr=lo>E%r$Lj4RYh;TBbYYh8Je6x(ZW
zRc7j=ryk~5=l10>XX=rbn#MGtJZX&F6UVftVOeIgGTI9pf#gkQl;%of8>MfLjlHro
zzDr=9Yj0QEHSe7@XKRl`(QF>m(O6jvzV*#I;crRytb<b>KaJ!Eo)07i=b!UB%9uwD
z<JzPVZRIWVAgdL0J|oG0R%dW^lu?q)`Ay}y^3V8ob)G7vIS&$1|6}KS_#V2QW_$5%
z#w1V7cJS`(^b$pM+L<+&G0Gy$glrEFe|}d;yk8|z5w{AAB!B(x(b({yv-MZ<E5PQg
z05i^ryvk#hJt-ntjBc7tE8(ey^i=lHrfgb+h&cJc=<B9z8obbd!Oya7&SIi)65gUg
zZM~2)$-C`Y5T9Z&Bgq$<B`gjlg<wY#Cp;kRDkGFJg=Bkev^T;kSX-NnM(1k5hqOlc
zBs(?!`p(;u2z`wJsoo+HfECihx<u=$MsFo#h^xj5dQ;_q`U4k6((kutb5SUmc9ZgE
zQ|S=AfXU`)L&SZ5aRs>jRdpN%V)t8mLTnHc66aB9TlJ(jnJ~%;E5ij7ry5KE%w9s1
zf6V*y)UmmsSlhv_qNS>6&3aOMMoztooV~S(e1TgMgbU4&5jGbTtuq{|w|(A53$&Do
zl7m;mcvqD6>y3R3!XT6lBxqdP`-c>$-9SPeT}DDq{=^r5eS7~%T;yr)zofa1e)#W=
zJt2jRex2>4scm!hL15r_o$D*6dPRzu(9_If@TE=*gTd?q7EQE<Xb(SOEPCJCaH=Zg
zEf7Rjk|8k4CkizZ_-&g}a00!^I4`b?NfZBs(eiyuGjd%gw5udm3Wu0IsYl+xMH>*@
z^hl+)7akpdcYyefdVwG_uuOUx7upJsaXuI&YzQQY*7q$mr+V8$Gs<sRXg+URXypL8
zHuAoOUg0|4*nkd%#HMd!8$1r_)j09@+YwvET0nOaqgu-u!0IOjN8RJ*<N;DEwm`*X
ziuir|qDzcl-=^_+#<iY<bh>+TkUnRQ*d1kc#s8Fluatva<*0ID-dAqOzceX|+rG@7
zWagmdA*UM2ZpTPQqKCk!k8A$n4}t?&_5BjN23%high2z#B<GvHKV$H?zA3r`C-2cp
z_VN2im1+eksm%&D-3@0Yj02QFvA7J$0*Piim*GH8eAa>hkm!=2yCPa7Hb+k4Pc5~U
zEVVFylz7vUFf+|*c3M*Kd+%!{B-dMHUD7i8sT-%MMwN^pk6j&MR%c1uDOf($uSfcl
z$8qf~kH)qYX|a4Ivo2%uMd?L{+kw}10wsmkJ_BWl=B|)P2TVH5NK$bv+>VGa<aQX~
zbL#seBJH(6zc$W|h^+1iiKIj)0;_Cmdn+P;0>E8*o$0p)L>9wPcSVTX@k0P%eVFCg
zj2|LqNO4-RoAHB@*n3u(-G&)NCUpF>n(lL3Mq|pS;yB-ZZc7_4jBI{V8`@*OBpGFu
zj>S=Bv@#5j+MaW5pKt6rTlbbOE~4ytPp+=u5%oRPeuQ`v0+8(_V4YCw$Q(%eeICkx
zh67h&%`JaJQIceFSMkeV;#Zi8UXgi)0cGDPE3M68MdmgCpuq}bYm#_XALmjLFV1eg
z%-GiwclW9l<d#_NhU@5j;RT4!a9>k2nEP!ft6`Ap;fI?ACq3%nR#nS+;hh^vZpY$?
zz_<w~A}hD2@|Y1M0OC~JK@NIgp<<eU2t>4D405iphT(cp_JTKcBSpuoCB)H!a!I66
z;oMuuqtJh1YQm8R>SNjNFMH@7EK@PTZfqg=VQ)h%mFVbM!I;X(ItB_`9rX*fgk1mf
z-Rd|dn85bIb4S8DkL>00{H4!TcSn62g57HWa;>yI@>O@~aS!t8Vs2F>lKn`3A>r!U
z{NE_uMapv5t~5I0(!sDw55H|YMX<S;J}s=uCm(*wvY1FU$7!OSdNx(n5Gj0SRXNMR
z*L9V*$m+t}d>njr7z~zxY%RabslZ&#s}Oe>`xh9Gg9`}lw1|J}FdHfB@C>l~mkof9
zn*cp<uE_PrRhZ`icF956Dbjg=aD>!TPN|3I0mi6-_|&&Ywg#PCuWkVdahb=s65bbQ
z6Wck_@Ww!*a!F@wC7WBd2lNUNnw7RQtM~hSBwTD1RxCwFbD8;KtL5vMtP)yZHv~7E
z&@R)qa|3rNf>qmY2NPt4_&relvF!~o2$jywYABdx?%JB>j;*QQH;kNrVuVV6s4evf
zMn}HHl@r&GP>Y%kKYBcJ)HEG!`9dA}c>l-J5l5jP+>!B366ns+SU)IbmUWY~HXn?U
zNnS={j%~@<<?i4>8*a?Ba~$kh$tPTo_>Awxq#=IPJYQ(O%Cm#pFl^{mfNk6Sd8{am
zRf?RwQ0CI*(>1D%*WtQ<#{C&&>~#5|$I0W0@wx|x-Bxe$T+{v%+J;z&fO8M(X^Qy=
zaK9tgSLwh{UsOeDueH7B*?Zt%-JOSDw<nmOo;y|9G;{SDL5n1KH7E>k7tqY(x1L+A
z?dFw>A>2)>ew~FfUA97H{qnUDfn6XhQh;^A;jfEQFaQ3sT>0{UW4~eo7-<{KPo|aN
zW!h}3p&$g=qVX*wTf4YoKqH1<Ebml!%fhX0!WOh&TuicyGXk{9_mUQz_`{NCM<qbS
zyyMR$H6u;BSLA$>+@w+jDr5KVUtd1|?b9zxAi;kUg<K3~S}KO+wY;LM2$%!d<!gEO
zn7Mcz7bT(u^hlV0F-h<6JJGd<Ps=Xd(ovAg^sV)4YK_7KqctN2mCle;$2xv*X&j^h
zTPscXl{XSoeTFPi<08GKQ7UbEMS7x2uj%3M)D@|IU3Q<B-AmJE1pJrm$4j@SOeHV^
z2E`THn$Vwv@qx4vy-Z42Z)JkEMvYmWGk~H6;ibaA+R#mZSZZMcQxlFwBTy6l&h1we
zFv(g}QjF4KHmnwljH!HkO;xoP)%L$Xy;iy}NcDmJ(oHxGn6tX_hs1D&Zo*wjT5EUe
z-BZjG+tq-jo%X9%v!j{UuKrDfA`34IW3kBk*oA$$TR}<*Yt0=BnvtX{#$)b<=CXZ5
zkb1``d}wNa(bb#j=$_b7P|qOQF%@gk^hSQXRhR}JOXD_-Ry;c`@n%jA9rOE+*8yPT
z@?hmTut=?@VVzS-jY^v62N^5WPPNXply&7Y(3KK;^!aAbsAPG#K@^2ndLk+u_f_E)
z3qvv=TWjNQjQ-}6pt>F_scW69r(8o+QjI70oaGaLi)#f`EvK*@3S(y$7JBN}f~gXQ
z5TnCUS*1O(FumSmWVz;zb5nDw;Uk4eg-c3F;*`Y~ZlD*GFSF7%2G6hy9cATbc1lY{
z*i)mchkA~k54TfZH)pt1T!LQaz6emcWG(nBZ?Yc`@>@=<B)oZ1vMzpwnR3NKZr*TH
zfFu}QTX^)pbjq=PdrTRbkBHeM<B1ssq+Q>eJf}+@R=<ho)ECj4^Spi8Oa+ZQ_A1Ns
zdD(D|!s~agN0P=}#`G0S(YyDO?S_x)-eAo|b$3%-)NlF!`PTdY0ICAQtd{|A4io`5
zmob?JEQf<75Ql>$5r=~%5{H8&6SspU6nAu&^3Ep{0y#C8!JIcIe_Y*;+qx0H->1+E
zEW`gFV6mv}tON8e#oiUYI0%~bA}M-KFZ%i&QbTGumhD}OPH+~`TB4|r<j0xea7al`
z-zGVIdf2}{d^j3A1rePJdV2jjDYzlwzmkyX-4v7(-l*yI$Mp9nDYb`V4~Lb~7&ro)
zNBDKmk48nvp%NVGfB7+31y@GFq32(HgyVqc@%b^|FF57d@PvqNenRic9g}J<35SQA
z)*6?~fqcs6<M|)2e|?bHBq@^5r}>9py}UmB{=gfkd=g2=>C2CYe8U<gOw^N8Qpgyl
zKmIv=efZ2_TVWEO5ib9Ovp>E~Eah*=agxBX^xnCSV<;m!e>-`og2}gLjvesnpk)@g
zf@(;+>st8X7}YA8@lbQc(e%d|vG8YRq&HOiYMosa2FSotN?oA@2gNw(JqH&qysh3=
zf`mDasIm19B+N!cM05+1WG&u1V2L@eAJ;w`gf(Gg`+CIEzKUiRG@)Issl&Q=DNy&5
zx@sq!N)=gxf0lTupJpA(ToumT%$YsryIJ>KLKfZJzH-+gyK!Oj9m0N<Rb_=+gvQgZ
zdpn_3&^s`PS|V(&PW@ME*`-katQ9da94%R^-+|{0NN{9;#ZVbER)!oZ|FXi~(fP@V
zWDP59?)%caySW!j&*t-eAs8h??y;Y<P#?zkvorz<e;U=%3~eWavp4j1642uuZBu^-
z;ho`#rK25oFzA#5_EosKqk%V}QsAX5T3+<l;}MDex-~0V<)cVIn(r!&BThsVHMign
zlHrmT9>jKxWH2-*Z?ypohzSK(a9->UrpGIhWZ+0jua_I(dF4;Yql)l;%fZly>5wQm
zz6=7ye`FLPInEA6HYz6q^8UdK<WB_cRt7Ii2L3#NfuOqT5?QQeHg2unB{km-R5h}I
zm!TVlJP)?n7&aB@G14-?SA|+QSX8K7g^JWt5-QQY;#PGWuShP5;75(6J(`inO(vGr
zZGeeoh|BniN1Y60Q@bC}wT&XmFp6u2wwKQOf7`k4(ME9HBQ|dLSA3TNJR6vk**J>l
z9t|dz!Ecc19&H5AJ!10{4@9O9_vc)D5n~#~d531?%}zP1)&Q@ZAvQ|0hZgX9hw+Pu
zB7d08`Rv7wr(VH8Jj@m9z>L3{_zq_T?;T)~>O>^_PEswcNI{L_mZ8~gqhpUY!jkle
ze`*6_^-CFI?I}kT#jH$gw)dP(R)8NosOLCZ?m2gJ&oQ(CJ;x9`zoy!MhX(2nTt_3o
z`547DLvx$7grSY#x<~A|-l2aJ@56J@LMAtgXNKms$#jo4g6AG_fHmTXwdWEJ&2iC`
zGF;1iKE4;BHqZy7dXb^!Uc7x4z|cnYe`1f=aqgJgv?sie4;lkCLWNPYuT!*5R;ouE
z!F!Kbc&=6|p{1()XNzUW^KFvQYSb+36m65W>d{8<-6M{$Rt%9&G!CTqYP@1MYf{so
zJM^fILc39Oyi>GIma9h_(W5<L=et_`-4jc;ClkFpSuhpEsCgd83T@%=2JR`=e+cF&
z-Uv<V@TeXc{I$d)2G);pb8OlqS2hEj#-?i=`mijo^oy&smu={C^5yD*l4hb6R*#Z3
zj+V!HO_tf;9}HlKdb^A;#J2F3ZeMT%sng1FWVYaTB>fq1MI_X<0KUK`5A|84+ZnEJ
zsPSt9CKzG;sQE^ZX0+Ze1%3U_f33FF9%ik<vC^X6*JaL}F8A-R6Rw7_QdmriFUGRm
zbsg*$DPu^BlpTo$*<F*gj)n2dw#IA7K5FjMqZ#S9OH^OKYZAp;b0>-FOSvXdeeNw1
z#gICQ>NW)YGHG&3I6G=a$I!|e>_%zo5l4(tJz^WB;+T=DM_+optw+~Qe_)>U{NXKI
zK`xG5<0ECpLiRXn?%AW6>=tg7jJ|%?><nwoo$O3s$_Br?*&>E$&sZ+-j~r)tljUMa
zi{(may-qf>mHjG)*tLDP_oI%TN4iVnXIMC$gx8~H+&!961IEj~e%IufwdPK8+?O)1
z5MGwN{)sJ;$B;T--+d$-e}E&Ev!69yswn)h;gW*s<8iwzb+x-DN_|uABuQN%*Bq}d
z^A;KEf&67dq;NZ#O(&bx0@>5~@-vN=3?isei{?FAe*Ivxp!M~;CTXlScM`R}lxwoq
z=iVZ03~7-zM<nj$WNYNF7z}mtBDhgY`V6gT%Qm@Uhy&bxhS)9Qf0)hgK0_z&epm)7
zVr4vei^iO6=Up8At>ID~tcb?0gi8~7o9TuyDPSEhY`R5^9P}+V6T@?LAG~djqM`2+
z$+bGwjmHj`lC5p25%bn<!@D|#%6E+<I%ZMNV(p3gCKGiT%A0c*-=35A(K~NM45Pw;
z9IXa}!|i(4BaYC!e;%<d8SJ;~T#x)G3I=PZMN*@@M~2pdEgPhuM;svqJz}>27`ID8
zkNzhPwrC4{xlzHE9&HJ>Y><q;epmcG-kSUPd#scJ!4{4fFHU(`?hPI!NBUL4mK*#G
z4HCZMX5<&kT^)2?9jezSTWw(Wv>ELMWAgHG{|1zO!8^L%f2C?$Qgr8vFK$>Q%6<oB
zmPyDsYVyX=oVs;8IRhH7pPV6<5V~2}w()+sQZWeS&H3eiM{LUFw<}$JA9GLrz-wZn
zPsN1j<osG%E8)ZsKL3K;UA7vi;Dwf>HV<>OH#3f-j6<)BylnR_TO37@p4Z|;T7!W{
z75*!TUCN4;e`Sc$k-3FmVm;M%wl?4#M7m2~x4_hw*ORYWcq@#y`J)!UPmu4^R5HW!
zph0f>-&B&J%J7#P;cWl}VVsXXx9j1<f4+VF%a`AjfXDuuv&zYW{wUDDAM<5&Fb54F
z2jB5J3MF&!K99!6+bLy1N_qDaKG+hzOb=Frs+pX<f2DrUsY%#qwI&aj?{Ngk*a{#?
zwGJ7vrPlJk)5Kb<rHIm|R*ig>QSjM(3D4K3@?}2gb<GM-{cC#oIz4<wR-N}gK$aM_
z$ri_lXzd!SN>~+)qRiGLpSv-HP+3U<#NkXTZEs7|0=m-9o8baDk!EV8wKmrj((alB
zSQVirf7M@}euf9uR!CV4xo`=LffwGE@2)6hcV>7w|MCN$`vy;aBEK`vqtXU^A`d4P
zG$QDXRX(=@Ud`GkZ7_SyO+IN;TpYc;Bb<$@vGHjR0N~kf$&(E;rXEQcAy&n$cWY>(
zObJr+NU}=T)~qQpt0*+<^F*3G^Y^$A`1#GGe=8-Nvu_}s`=1Nh!NEj0Hm=WnKBzVU
z%wW)Mur}sdt0(eq^h@`dn@3kZ1C%)jky_y#Q_pZd-#x=Q+>Lc$nFqsja*`)_I%!__
zg0siL+x6j{QWkiDU_Z}(EM_tEI5eK)&e$Zp3B3I3G|cU2@96JI$nSt_Rk3kFoCHiU
ze<XypI2DC=M}t_tJ{5`Yj2MaxzUzH4OE{AS>F7J$>-{vR=MP9(ZA$3y+1RJW`I$cy
zMcXxhcB=XNv+GgwIh410CoaL78djWh&J*>9HcLsp!l{d^VNBwl%{gk{VE%}F?it^Q
z(|{nBUR)Yw_AS6j>fPr&XKz%z&gojOe}K;$&9WgA=yrr|tu;2tV}q(^SsI2B)=Z<e
z+$+`EUa0VV_kw{$kMUGJO%hS@isuljuDiQ;7xfDaKDYqv`P$>UNL^`uUFi+P?ZjTk
zrjhSW%+J;EbDjub9#_dIxE5uwhPfjpqV4$GJ*P}2_SiWhOfn+raE^$0l?a<Ze;ova
z!X+lMEyfz+>bJ!iGv2Ry7wD(DC_^nc8i}P*%#}|~kk(`rha75%{i0em&PL<@*%+fT
zCT=|1$2l;zf%%Js=QV_eZ5QR44Na&a0#wiTU9ICX9L#1TkQEk6+G0?ad~HC<_5V13
z7mv;l+7NPRPqc4_QR6>9Sq6ZWBYEjtYoEgzP4q4}CyN`f=9?^j#_XH?yZmp{82tS|
ztS%bzmjQ1M6c#u(ATS_rVrmLJJTFXTZfA68ATu^Mmk~b&Czlka6cB%G*?vH7^Wa;M
zLomop4l_aC%OSs?B3az3+1gq%AWLGC#oDV#7R8qz3txVFc>R3%@}czNm^Z@7<@5U@
z-~osKf<vP<ixYxbC6>=$%io?jmlj^H@Y0+s1uqAmJ-nOeM<qP;5CLA|`O#?s-;{ut
z%<u98UI+Z_pC9A%1Sfxubx#P<<Pq{EJyD_b;_$N2Q)^}7JkU=7Ho`ss{qwI697*CV
z*!VHN(Qogc4}UyR0s$Ng%eP+-u~{p@l$VQCmI+~&KmWOWfA~s))~Td{HpBD3KVyfX
z<#*(;7r<d|tuX_K1qZh4#oEABe3UuNfkw+^<OcB70IxJqX@FfLKyZ3#Y&S0Uj9fGz
z1^pVIl(8Cktq(Xp_ODR2$t|%*g8EXPL?}kqZ@_S^<H#~P&E#`Sf@x}BnfoU^4)|U4
zvl7@Vd=1i=unHA^FWD8}jAsf2f_xh>gLiopmw%@e7JrVT)66<Sd-GSE_q5`ysMU^X
z<<?pfRU&DIN182i+o1d$${Xy-9Ad{JOX$p{Sivbgl0|Awi&BRVVsBT_0@kISpVLA$
zH_>1^+E!X}nx+UN`HPn+4J-VlVMc_C-{E?xam4_YHKoyy8q-neiZvuE-m*NC&r{V#
zP&(Q*3x5Vm%G09%g?H7iqt8uNP_3c*0zKIFEv(W6$*fk@DpxpPZh4?B3S20Gyk&V>
zGaqPYHDlfD#6fU;e<P0MCAL2gT#`+4)!q)-zb09@;(7B_6DeHE3B@H-Tv6g6+Riwk
zxMjw;6)VmlIct@;zgN{<LrYjwR9;icv~Vn@D}Q7ZKXkTg9ZV+FA{a*;hL(w~0y)gd
z8i%Dqo!6agHO9}68daw{XcTyTPa!6So7-^4ne*aJ3eH0d;QZwy@M$jLV{};2P-Fc|
z@e9OvpO(~S_97vDiZU>PIa?!~P#4dzvhx9=cgO8~oIa_lMM0sE602o4TJ?Up`|%2t
zZGSbB(AN`Mghe2}%?S;VfP259G`e_xabS|xGk8&GEie_%RpWM;S_}ays<ICGDVQ9N
z6*a*VI`dxmIZZ`*v}c}|q((`3orH?w4A6E-NQ1jjW(D!>s49+~uN9CKOxqQ-PCy@B
zPWEz2NF@u@uKC6-VJLXlH7nCy-WYSI0e{l#75TGZ0q7yK2!6egFe9fBfogF$nw66g
zHD+Al6b49}zN(n;&>Q0ORq771XCkaQdA+*J7=_J>FB|GExo3`B+I!IV3+68ReikRC
z3@kW*XI+(6Gl*mtXqI%687Vh4>jL}thQVtEX|VsgVz}1!FVgkmAEag8ho*1n;(z&M
z59ShA8yGlH0)5kxjAPa)ip#OEA9J{Cf=VzfXtkz#hn1n-5{O$?hBiqky)vqJ$a>WZ
zTGOhsx+~E|+fKU<v#>8@TN%XON7DdzJn{(OwYqoU878?!+!+C$5q5?Fd!kMch?>ed
z%ZOEA>H#V)Hli3>#6ZTL05&r427mEu1b8BR4W0K!uVWHf1+ROmdyQS^YG(~ybM3q$
za*cpm;0kSXFD0(8!xpyc!rJSwK~)mnkA7?F(&QFYB)D>np@iMH(iouKWTinM9$#sU
zo%gOZMrh+wFzpH*+75FQ&lM{S9inc^J3D$Z%MCm|g?L!)b5K`LM(ptQBY&h*D~z$X
z{P@`Apefz^Jv4=Q+(nb#JAL$doGecIzPehx^!b`q-1L@aUGAqTBzxTH3Owi2Vd#gu
z--#D78+^agL2J-=ZuGr<!#K+=d;@`aIt?Tv_V|VY(n;Si^o}PYyL<zo`v$&&Ks@0a
z#@>5<!&#iHzTvLAT7AP&4u8`P-7T%Ue2qvUjg<q%H@p|0aCM4y(Rpg|65s`S7-}W7
z7xoEDK}6bd&eKOBK58j6#KCtwty1rKO)B21#&n%=&FTg~d|W2?hi2n3X*upw1Nq|u
z`KSW~7X|rc3Ed6l%iRO!lq1A_S!&8VDZs5HHSg$<-d9>^_1q)0bbrpj!+RHBOeiX{
zS66GQl!iIEqMnbZ?6*;&T_=Wk>Ip^-v80|el-eRa?$~Uf0xD*Or%>WatSE{*E1i1(
zT-9Mo<C#}JR$~3JK*!&f)EfY(0NKj_mekCIG&;o7Tp^r|U={pD(jUJ3=g0TIeESDD
zvE|2qV^_Xt5R(Xg`G1waMqTn=IlcTW4bVeL9+s$N%IX}}m4Pm0Y-{<pJZLah_~QHK
zpRqYG?KPxl-S|{W=2F)(OW#&diBM~Ene=DhSa55zB-$z}@mqv~W8+sCUq;2RiPr9W
z=?bI%wLE-Z9=;-Xjg>BT@jL?#7XD`3g|I1H3>J;UrAcVRjDK^W`^cH#APj6Z_Y~Jg
z2?<L)*w-=!d!gf@J<itzfGr*6r3qja2o))Or|lb@Ak;!j+ks{##5c)+n5eA+8JwBi
zK7ad#1AoA<-@5k<=rP)kHg<W`vBXk{2rHEf8b2+_#O|ecafGGZC!_+qXmVwynXwTd
zFy+^!Ln#yY*cVav^?a5HX27AG<_HZKjT6#EYnPQ18<(A}6d`}Btd0-+q@~SEE?dM=
z{n>afh05>+Ig$yj=rUuFrISl08+0tTiZVk7yP{PF3keS)6_W!TIA@ZAJeh0CqtgW~
zwdiKW?U;Hsy;EJWnAEm#(3DV?TS@ug$5C3rp@UVfW*Rd)(I;!yCU(ad7ZYk>6~!VQ
zeOTNCD{?o_Qa*q61-Ymnhe*A7I##;d4yjVv(g%#NqSLWfo}2Xux!J)hoG;aBn1Xj{
zI6eg|%i=wct=M&1iG7JwH%kD!{?+RceX(HjrY1mT5uo2%QCZRwNlOpAmM%@gR+1*Q
z)>gt+T1)X$vE#&VpKTt;;hu9EYmm=Ub{ieok2mXxo1cGebn16#GDZ7&R6+DMa_C{V
zZ{mrRxjKq_^;5MBul6_${kODci@RG=m3^^x?3?X66%YHSOX#&$GS^@`ylqU&vS|z)
zv9f3?azrCH@ri1lDRVMDQJpNXUq?%P4*03{nF98$*urPzVwQEW+uD;?ODOcg6VWa1
zbAe-<7oUF|+Trsso9i@E(^AO-yV6o3QKh$0*?el$D`On`B4wVnF0=9O-Hc_N%Bcz0
zke=0;xm4q22yJ_*CV6BZaxND<#@2-O+;-wq+gV7Q0prT+v*J&k&y19rVYAW|j0~|l
z#M3{geF9VX%~9vjS7y947QEN!oxNNwhp*5M6vBV{Y6gxaz5{WrT<S6pzm5?xUd4lT
z!y_#D+N?YE3PjdMuIZmjRBG1h+e?MKet%ZxiKZ@t>g3c#jX`BdAb1@&GpU>m^+B<T
zAdh<O=WVIi%wZ*xYht!e-kJe5mFz-Za{(EvkLHMVQOA}D%GyuYS!pGn?-M*qWj#Hc
zxr2X(^PwrhRwBkNEL!|55kt4q?HHIot<4UN+^;y5lm%FZ3P)x=k$8!=Lqn4<bv*ep
zj_E*WLq{f>GmxRzQ_EK_uDEXF$xWEG6RzQ<x~<NvVk*};>n<T_yzUc{);DPk!W><X
zwqvT8@$WcvRE~WxQ0pXPrZF<;bZ?+f#U_8r0<5+PA9dpMDSL0@BF%YB4k`87#CoGf
zV&Bl+G;2$%ltQJOgvOlV{IqdvEb2Eu2@FWx%RsAWiDJfZLfd;65?)4%M-NlI&g<x-
zv~)E5S8kAn;YyZXGNs#oo$cgkTSZ>MIA6&GA&Ky`aor_p1L&MmPu-0#*FVwcSd)Jm
zlC?Buut~L<tY!EmTqihI;EgSXuS11-!YbhN;1y5#fAlWOAr2XP)oDhXw??Z&!CXzK
z+`+}E#Bo!kNc>P@hAZ4CMKw0xsU`8dNF=L5?`3La-Qs79u$31%wh;>;F&Fijj&$dx
z<j10s5`l3^Sz_vizb@c-R2+MJo5X);X8D@Q@8lAsou?(|XTL|%sbUJU%SKz`Q4Y}F
zIx!XX$^lAfvAzX1+4Spn*QT3(b^k8!`TaZnKX#wGU8d$RJSZlep*N|8hj+REB6UL-
z&kG*A1A{YeS;rgQQclX!WI!?1s-~J99hwk}72`=Lv6a3jy$!c<PcohpN!x$UkSiA5
zsn3lsGGWC6nR(+}85~90jH7t+{?2K+2bp|tI}r{|J+-Yiec8EuK4~YjOMbKSFR47!
z-q9I~$tAz!&pHE>1`LR=B#LKsE4ho9i<u3%@4W&=pl`_*4~Qu)y=)UE_mM4DkdWDL
zDO;=&Av0VfTRhDrkhWyY736<rc1yN=U(qRneoD3kmqE4>ll#hcoc$iM9cS1h+Y!=+
zY(<dqIQANUeeDet^w(tO6aJdwIr!_eX1b@pCbQqcUy~Vj_-g`bOYV@rz#?!i_p|S9
z(_)_y>gJ6LC7F<x;;_Sq5J)F{2&H}WA>z(Hgv@>iA3|o>;X??dEnR<#HfqCGeAsy@
zfj~bcTY}3VTXFCBKxV&(Y{wb)$aaLZ<JVw>_OQefXRSd(w+^}W+CwM@OFnu1+N13C
z_P-QT*8;BSjJ$Dd8L_@cY#C?R6I(_|TmC>=utj}+PBD{1jq!Jr!-(}g<S@>#M-C&T
zEjb9otTVTc@dMP8F@Ar<VJh1eb8jx&;+ID(?%|im8TR<)5z-F7929eA@>3h~=tfX-
zdM^d_@<Vy4EPCTd_;P(ieR>#cff5^3w+}ekWXD;LRJOEI{aSrd0)=izS~m+NUNRS2
z237ZiF3)1nYDR!tKk|hs&)7NIOur-5?55w#n6HB8YovL$JI_0lYHPZ6x5gnF@?<vG
zTCqBwve*gcRamjuUCrVV;0YFwod*`X-7I!f7Q4H%cw-xyyuW`w{2zR+8u*t1Zw?fd
zu_Y4%Gc%U~XbLBP#a+ve?6?ix_gB;pT#eoj;G%)9>Z$><OVGO@s~|J0S!Diy5z3Zq
zlk!8hG@V7=?sm~MNgYxYMeFk0?>Bz?_3i%k_RCA@n`7PxCvRWB-2{BW;a|bwqBS=s
z1hY!qzW%uV^C#!h!siYj&AC$Waqzo`pXU9C5*{c-fRA{8|KYTNzm$NF9Dm9ed>-j{
z|NbL>Kkymubw3kFGu|OThHq5JR2)7QD7985J_qOoNF)6FzrOzV#*<2%1si{jKj>f2
zueblcQ3b7LO6uFc|9FenwGvEud9%VW8^rC;|J;6i`#a^jG@ln>K5nfs$b5o>P4&&%
zz*PKdG@k~4^$n<Rz-$`KW<Uu8nrTp2zvl1bcUs5qN9C9Xw$nd<{(d7@{_*RqmxC+9
z(kDxf%rsMb7NjYtH{>)~PMyAcjMoNQds<ehJk#FMrvrS}WR+SnKvf~@s@<;X!0+z^
zYM-Z7QA#rIivbsURTZTyQ_53n{(Apmx%`=w;ma$3>IqlxBUFsd-$ZGrt1jL17@k=`
z(Rii+zXXcI&t!&ZlEF05MBsnMC)3$W!#hC{KluoXm>NcH94a*q;V{H5$!5pD`F+%3
z_+Do?&ckg6x?yM+CJu=~!oh6NZ}3v30~1_QWjBzoi~^Vxh6J>pey(xMisG@F#LyEA
ztr)|9DB<|Iu86fvh)|4ibiKzma<yYxxrsWB`Y=3cbcf%Q<4@PajOY=`7Z9wOb%IK;
zF-#MJF?B@e<XJB)+kn{Yc_w|IJZ?ZWPXWtK2#zF!vE~2}5*E`<=Up3_pEJ;bGsDM!
z1_PlhO9SO0l7<-=*sQY6UM9o%tU(^TI%{NqEusU<$ve$LK9}f+YnehvkBr)?=}K~a
z?UpoybtanmDfyWL51*tmh|WyF$ye<hGsdAJj$<8|NT{1&JXsE1S+ktb%*?S-CAIH3
z;^kqQO}CD~=-x&VrX`RPisp!SA14_7d43M(o;hx*9{h4dk-{-43}M(1qNEf|+dvtA
zu16>dKz5Q23!aX}DiV@y<<gq*uJaZRin%~V;GfrHG?z?qRakVCnk3NEO^^XYT1%zK
zA>snixnkf&y^)T&9z#8bgI1^$O*&v5bL;_j!tns@nED*tXN$JMos47$yi=<tz`K3G
z6FII=ipIm*8P){lq<Ei^X|f-fDHtDrPnoTg4zPrAH=TWvaIKhj?1-SaguCjK$%Cjy
z4L+HCf^cl|$-sj11;-7aRGLX8DRsFb1YdMUpSPIRjEi!#p6k*$$&4(d8J>N9*(1GS
z@F||XVTlwOs3p=WFJdZIfhKHy449Lz*D@ZF)b@g}n82LvouLt~Cw#@V`wG#2sDrOy
zjy-&Za6Ir8OugbOrp;H#XdU<prXHT%;49?W2fjiqaoSfdloZC|M4gRty<p8o=POJq
zuFs{fw2|@(9NqmNY@}nMQx~J7#brOyS~43#;U{KQbFD24D^aH*Dr(O>cfF$`=6V<v
z6@N8n^Hh`eY@TrJkj+z$``J8yp)RudXthhx&=|;c3;tUXDaTGsy+p}{0;YtVnk!LQ
zWWA<5#ZV$c+X7v2%Zzb_403l}QOaPnZO=}WV_y5-eu%jqXD2OLN%ljcNe4f~9DDd7
z;dtPOn0n!dr)`wBzUcE+tVOTvfu8Ycm=R2&*``l>1|LdY=B5`b0iMBs>bFvs$qO;H
zx{yU=53PHyRLnSvuEk7rEv9G<++3{;Ip-n>YpcX4SZ6^-RgCWSEM>~i2yWN26y+d7
z`ZD@7L`P|XW9S_nG1rRd*da?HnskVcm}8IVNH`uuM@-!#I+EvijE-c)527QcE~{K^
z(XmRFDLP{6mgrdb+#DT$5p}odSSLe{js#(yq9YmI4$+a|_8>YE8#F{m7}=nD2S>!T
zVn4>T$40D2htP-^_K1v_;Xz<T#N}R$GThmQ)f*n5=l<di5#x1J8}_(n!HM4B!(1z>
zsvqs>Bbs!8ALiHte}v-!05SDS{ovyQ`b69>qk^n2nWt%&2x8iQCQxUYbB=Ot2;fsO
zf@o#-ixC8**EbloSB#sL4MvqfkGBq|kQ?10EIVDs9vIXa=hM|kN?=~7_eST=o$;^f
zx7=+M(#J1cS(M#Kf-km|Khjgpq9i*!eYpv39*c&r$dyl0Pg;pwX40vDM{o5V$rL2w
z!ZkJ-%iMbPXx?If)lM=VdT+53t`!AR+1A}7WbApS%JAlhL^w9ZxG`!`j$1Kq++QNp
zD;*DwT@7h1KnH9?suY^2RCaFWn`)(;=b(_Gh_0-3R$#7*Z?spZ^#zfk<4&r!0>19O
zlS;TIJ55y5G4JFJq6!nvExU?WoM!b-6<}7L>pYSxDK9&Jl!%SLS`)4<9yQqkb8PS^
z0TvD16&%;GD`HQtrFNq0`NFCa{KMcHN}I=txatd@a^6TWCLJ*~PBAX+^$8!le|4eZ
zhV=p@*jiEQq3<Vm`d{$;6t)r5O~+4ZClR614Ef!TIy1<)L0L#l0wAPXwp}hCn9-Fe
znPTGJ=zw8=*TpL09~fG7Shs6#nbN)2PK0Y!YO&D#a%w@8>5y7rjy+Nf!to%rz|>`G
z(IMF7Rud}O28`1>g-X^Q>~firg6DI0E0j<-2D|*~73R7X?eeQvM26-jnBl#eN;9Ol
zV0O*07VMg$hJu+UzS`I~Y7nk<3zC)|>@jLkW!j^E2I1HtYEX{*QG-x7ZHJ6=DSiG%
zi^N!%SsBOH-Q2XMp<>A0;s`av#YsNR+jS<$L5xt(wZUuq1u@et9prO2DpB>@7Jzvp
z1asYtjAINVBJ_+vUNbFwZ#WRH8|y#i6u7y$u>MmPrU0lwSHD_rm+QY+w`yGfVXi&a
ze}rR)^&jQ9f4}}C)CcQ71!2c%)alo_KbqHng!94rZ$%EnFTW>Bxouc1_My5!sc}k3
zyzTr9StZIe`(f&0fy3L^U1>*&%C)}#G{HneVDr9HrYPbzav)5%!r|npScKN<Kzb*h
zaNR;mdZU<9-}@{g(he5kzBNOqG6A9Pu!zY08&r<{e}ttpN#le)B`kI3$O#KE$F>s}
z(;;MjR`%G)SHee{&hsi_x;3qG$D9xsrd1ih+D0y?RbiW_Rm`==v`RR3m{uvr{b`j@
zA55zel`g1;6CwnxriHYWQ;Xi<Ay%1!qudR<g(lY~=4^AqZ7bZ4!I@CE7N$dd4b)f4
z0N%qye<&nwE(njo8*|;P2&b%o$WX?$YvFzj;Fab`tpJ3&wU;gx#42SNOU4uxq&D3l
zN6M8fD%O6EBq90qO8*60DTMPuD!nF0TgOJ;N}-(hw^9i8ak2h7l2fInPQKiLwJYCw
z&rX$K%yrYVQ;tJqXieZ!xuwnyx!gji|AM`qe|*YPrCekE<3Oy+y&h>5IH&2*<Rdk>
zm+>o2KAY_@BDLa}d=QXM=V+y7p5g?7W4QhD3ViRNJ4iag73LkXkdN#^Pue^#FNZS1
z<+IuyHsy<jH7^Rzk0SrA0oJDcIcdR9A&p>{LNMX@xtp^S>ZD;|WeHZB<IjafY&I%l
zf3$V3+qmc^5B<D8DIrFJx98)SAx(5ocx<WyVTRn&bLH~t$lS}`wc)SjqAe^IY#w#p
zJyR#!(C63Bp%X|1#B?F$(qTZtDrZjl{tAcXsd~dHWM;iPPs_#WTdzw;x1}-|5d!#@
zi$bTy<@NB-z!Y6s+1epoxmlb|xIB$ce=H7Ld(9`ladRDGqc_FONLW>3OjV09;i!6C
zuJ=SHp&|K@i+G_H$&XkZ^|omK90l7*_r_%v;d<t<vukUIl=M+;_+|vmYZ!9JB*&v^
z4MX*R?yX@kv3#o0<PBovJ$)+1{_8=Q8h5EauD@K6aApmY+VsX5=6bx-J$DJVf8mU1
zQ-L#d$1F!9&WQf|I77r;aYn>Az!^3LYTj(enbAyo+Z$(?Yeh$e?SnI-O$E--9kU#b
zI3xP+;|vjZ#TgOf0B6J$lA5ClXBskJ>3ic0b3Jo%VjIqgHWfI-cT93L;*99Ok26Hv
z6=y_@1Dqjh+)UMuGbNd_p*PMjf7czHoi2UlY#cpdyQn=jS`m4AAez{1ZH;k4tBm?O
zt<tvonpTbeLB?(0KM-+O{~%%<_y;lCvRS9yKjOBC=<Obe>G5nYI?eW2mx>^~ykVB2
z5o1{SeTXe-yCMt=aR4x62DEEU@ztRQBS2;&mPcm|N5<Ou#PZsjOezpke~#Z=lgS|%
zOVL`CN$n$qD>vjng~)9YpUJ%*$;Hz9r|m_V5SR{fr6yA?#rM}_yk|zK&M^%${kDn>
z(d9U%;Tt;DV~Fz4hPTUyxQF?5IMgQZzY#(AU)ey!U7~O7n+mE@inNhe0E@ieeDPqH
z@nDtRft$||bxZ{^${Y2oPnKN6IhbyGmr60yZ>wMtU5@<{-O#IkMU-#$OU$pMUlKw0
z{Sp!P^vgOal3y0UBCkm|`Q`4wQNP5{S+Dmk{BrHTb)!F@UvK{d!`n^fmjQ1M6qn#K
z1{IeW!W7VdfeGV1|MvOw6Hg*>=1sT_U-Z9^&!_)BQ2~_{%qy6F|M3(aC<#nDF$rZE
zw{rUPU(+8?f2U;2V8&t_$O#zd)1RS3)ACEwu>;gGH%9A5$4W8hCu2O5;Ym_4N~%yW
zjY`m{Djn`o%^C$(s0j6z+HE6vA?k6$oTF4ZP8t$_h*T9)Q#fe~CxqXgzZs5GAvumC
zKd|*3FFw;|3W=7FC|}d^%QCJ|hI!Num4=6Fl~bWTvzp4k@bPwz4~LpZgQfC+5HmSK
z(KD?mL6E`D@p7jLqo_=4+p+q-sxApSm=NhLKICj352#Ek&a{{J@}WEfE>;Xo;!Hnk
zM$$Te=A0Z%U+L{ElO8$~64Z?F?=4)3pRwCgAJq;AGFSU(OLgd@T>EQ$lt^=jkG5|V
zJ|6T@BLCw)+7i6dM+vI~A8mSH>ZAAZk$to?u;3eyno;r5-RbLmwCUaQagC4SlC7k@
z+DBWeJ3czI&4a#l_{eNGEw#-G&8$#l1j%7v-4lgp-gf+4fd)ixJsKR6DkH{yQW?e2
zO;@FQ5L?K+J&r9c!7H(ausXn&ruU`TavvWVTRO*>*Il><TXv_f!xmy}H6Pbti!@B3
z_g#!FE!A<14j0AfM4CI+|K^RC7Q_@DfBFVo6r=A>U+1H(2{nAI`RI7b`RV5QYF;fc
zwYMleF0dMW<}bTba8^YvVvGo*EFKioZGzxtK!{fW3QfkeavmXm#4($e8ilMvet--T
zz(TqiY2si3KhnS|AyT}zIe%DiDLr{vFe|pbWlSp>M_@iF4wZ(jB8>d9CQ39Fe_e}`
zTXrT34S`86sc0>5Zt1>wXZ!N&5lQi0P&q3zynm9Uli1QR^AaZNa5y!zZ7vzkMQz(r
z6ni>wL!e|1F;siOjFLmjLT5TUQEpoHJSLB1QJG@N@l7_bQ2$dp;ZtqQZ*$HN+TcB#
zXs8UvY0cBl)AxIAvXMsiL^EH&e=DPa*+@<xLnl{QG);pAdMMF)pr9pUr=t|dg@Rfq
zbE~LIYf)BSjJ)GtqK&D7>oZ;^ev|ehV&4|)2+uaT%C0oF7#biEP8JIhX|CsS3RtC(
zETX($5IFdtF7MV)RAmv1t;CijaRAi$etm|r&V<1=v8*+bf~@Aau}MCwfB!;#xe)($
zTvQ&_X{Ia*zN}7J`oPO;=R0DV>Fh`9R(hy{5}8W2@C>*Y_!UN1UN%dtTAaBow;K~R
zFViiJygJ19__Atjt`bD3Xc*=hzTYvM2ZC9RaSQL{-HFnTV|dXQZfPgQ%sVJ?|0T4&
z&B6qmh|fctpZ9TuJ#Kqxe|D4)sk~UXy<ksKkuSB~)#|Qn>3*cH{8jmF`)zQ+b)n9l
zUn^YHIeuq@k%^BwytvKx{Pl3ikXLhM1>;`Vft^^@Pt@Fv7-%xY1q)}aFIdS-*rwOb
z2P_kyJ1%bR>X3OgfKn`_U|r!Q%v|u-dse*UQohI+A>1W^ywaT@e;6<2J9xu>QEElf
z$f=!oA5|4v!bh2%BxAJDJUixijwNv}oT5B*g}LJyB%ug;BZEY^RF?LR8QSf<oojqz
zZEfdPY}zFt$42Vzt=nqUqNGPFUSfr8C~F(HfZtGkukH*>r<oMp#;{}-Anyxj0?UNB
zIGV%zmAz)JQK>SJfBf|~KK&CP{*0YNg!jML^+y?oPyd<z5jLI&gdK+0fiTAG7-a7d
z=96F^7r>e9LPDe35Im2QxQ{a#B01FkE?JyGuPNQVjdKVdEs?lFh6*@;1!Tq}1~!$E
zkOI60lTD_z2;LY_R_S@!2Re0*S#BSh?ZYJQGw?uGb9ctTf2k+M>DO-WY>LF}NXZg$
zgwrE&Ef#5lZDkHG;pB)T(Lrj_DF`%637|sl(FqwTHRBaR<2q(@-fZ_N2+Y9&D1}s9
z*?vnP8q$|SdNUuQ(Ty7Nv7AQixkWz<_`WH6rWXDa<(=$dxxY%jRi)dT<E^Dc$xD}a
z$|6g`2Ayj(e{at=`sMo?ypNCj7#;1Og1Sb}LQ5K)_8AvbqSbCxZqhm>K`<rCi7M@u
zj5=BBJ(Vw9laX_v)H)w9t81m?z?2bn8gwE?0bC0WuDFESMv?*;(SR_tw@RiIgT3lL
z1c376I8VkfK<Z=^kn22Id>=ZV=4h_e3e=K0<Sh0`fB4BUdhxa!V<YzDiYX1WH)^Bo
zBJiqOenLY+DkfyFA$5N9MLMQjFX8)TRZpP=v#Qg|W^HoJ`<7qj+?&dBKJLz2X#4u6
zxd)Y$ulg}YJjkuAv(pUWA-5`)yB_wO^(`{$mcOSxi!B~St{SdpnZkw*J=G%AW=<Lk
zAlm8re|XY^UJGFd4aX;xH@r62X-V&x#F`o7%2Jv2FXja@crkpeOQkI{qAc;{?U&y2
zk4P4%V(?6;KXBxu*hNb*)<;AY3aqbiOU+j^uR6ZAW?UI&jB5&cZ%aMq%#7zbY$<no
zIgs^Y?(8whCKO|JXWuhZ+7j$1CLMCV_-(nze-$m)%lxt~Y*ly?22oDdGiiaLrRAPv
z(p@j1kme@z+^Wjy%Qf(m7VCE4E%eED0+_FR*hZfQTdGaZ^-9lirn6PId~B}%bF4sm
zM=mdjMACt?J9$Y5&U#*?5P7e@seE07Pr_-27ajMJBWXTR>f}C%PZFbDZ9Bfo?%EX?
ze<TC;!BF%Ad(r`#!FNpSx$Hmi&<cj?V&@Lfs1OROPSslw!Uc0n0>6eKJ5`(0;Y&N}
z06@Z4q_grb2N1%?06<9Zivi?LUJ`)xyu^*@F9#4Z?gaosZQ&e%P<?y>fQ&>x4j`22
z1R(oG)x+SS*ve4r*a40+4{8fL$`(`de^q5FCf`nD2N`vc`fTyIpZb77zgd3zGDv+u
zyLhJa`qYO|9;7~`_bBxVeD0(^ph7{;eo^W}s9%u!5Q-;LA3|ko#%1_vqW^-_hfuyC
z^&u4FiN$@-+(>=;bO*$8?n9aV^d}WnC^zfr4+xx7F@hh^IPtI<y7;9q9*y?3e}X1i
z;pXd!Gzc9{Xv)DTjNyt2F8=y3rsZQ4#<ab!4rA`)C5JJTm*YS9NtM@@(&kTNQdO0X
z`{iL%JKhyx6S2LMVH45ASA<Oi(NBa;gz0>+r-O!aaI*!6d~oML5EXvop=ED2t1yN$
zB-C#GIBx5u?YeSy>I7<~1)&{qPXJm@phc>`9J5+B29VWuz8JCY<0JuA$4UIX#h2qp
zJM0DUL9E~$dJtuN0eB4LK8_uP<y1A^@W&I`RxlH3=rDW7rs(&_=hOcIJW{l=mjPlI
z6c#WsATS_rVrmLJJTFXTZfA68AT>ETlTjZgm-5UMnNHko{~UjIwx(nuNuK7RT|FPW
zbJr%w?15r3Z>KIFkVo**8FT{d#C-lW@7-$)d&^%$<<E*R-|>f5NhKqaz~`gFA8Q(<
zJg{cS%AM_px7f`TY5@ai1IU-5(J~VRGc!3jmjR?6EVqcy6!-F%(IG4p0Wp_BX&)zl
zTFH*vI1s)2EA#`+u*l*D1P!>I_5gF6pl^dY4U(L4nEd}?iV|syMbVZiIR|->s8?jI
zuPDjk#~_D44sRcaPiKvT6W)MR!^igl=z^qAkaW_T!2t-XVfgqt{QX%<W$AdM1EtiO
z4oAN|eVf-84W33opaWiC90L8*K!=)tzSWtIQ~K>+U*h!<XS~;aPKai@LmkH}I^<E3
z4of2uwTW_Q9MEjw?e(9JKMyi3Ns3_O&-lgvy?-43JMaanAO$)*{QL6|FQ5jYy&AL+
z!ssyk`Zjz&{Ka7}!X(`zO#Gl<zkUqt%-5vjAW6s4T4M?wgCxb{U~LdOUYc|gjHE$B
zIyIyxq!k*{F_PYliszTh!xbEVQ&D;xa~h>w&=jeP+`}3#G$~+0)ABq+o9-Z0#+B!f
zQ6a?{Qb;w&0#zzmtnf10IAM)mn1eI;xNJvfz;fC0%d%|eMNo#1#%Lp2yNRy(_V^rv
z5dry%BrLC=Y%d_VY7URfRniM5tF?n#mJ3cOsMr2R&Scx3f=-}zGryaEMlbtbhwoS`
zz*cgY?>y!aD69jH2ow!)Ooxs~w}guYO9{;9x-3~BdW{WBs+hgX!CQk^X%tGUJBzxE
zMkS~&RHHkb7_A7^Y9<pwHdQ*XU%}1_=iJ)Wc_ZkkoWIY-RnWo(<Y)-}s|j9EajA4>
zXF5ABC%_!m+ET6bv#@r5Rww_#zCbpb(u(KEts-DS1nKx_N-cdwAccI1Cf}E3do93)
zMcJl5`7_xK&ypz6R5LYZ(mT3|4#n?Sl<>I8Cdz<_)oztqlnKI_WwqGHMLQ*oQft-X
zK@gk#O={6f3*#!iYzRml!>aNDF#Ujpq_tmZI%3ko<8m~)$JG;m2^y@loR0My!f&!1
zQk}F`?`S!UwG=8vgJp;mMyVMv39&fybs@znBMfOK>(6Fh9($mXahj<Uh`KSS)Fse0
zIz4s?XfwBpC4JN$&U%PA`j|0ON0I-ze9z@fNC<Bkn~rlHG4i+L67DqSYRiQYK3zGE
zXr<QfH&xA!Dr*scP#9adm7f-9vI<u=Qn~FY0e2{TxttWhtp+4e6YhOIi#4ozgEn)2
zDzzZ96@5>%rdy}NgkpSL1YTyZCqEgzPF=lp{Ex}k-e=b!dCU3gF&+A5vuF{Gr0~|9
z)ilpUT)FqrCE?Vy>yE3U)#&$*9$5;j4M*HSX(1;4Pyv2_*)IfBThAmkoMz3`n~s&(
zAYaRdCNQ;?x7E!z5%@~w7>bLI5y_IZ_sd-ekDl@rlK^m&N3+o}k&G!Hlbid_P>cNq
zuANH`wQ4Tv9DCDKSK#E<n9pHDFF3W&9B{wD$vZ7f!V>~VsUU)@RK6i_@;1P2bzet&
zfxdFTF>uI#-0;2}Jm`g&*GBk!<%V2GJ%LlAS361f?OG*~5;4v%*Nx1l5+_;?i>X0m
zD6Pbz=^OpL;*9vNzh0)soFP|wahh(0^f4k9XTB~Zof0t;S5?>d%w%Hq4|5J&bA)_#
z<U#X};S|wtcCCCaOuL!6C@Z|ba`MyiC?JdJCcQC#Ga9oatx0<rrSKN3$sA~Trgun<
zL`oX4C#fkBcOW$k@Cl@*l<-Va!`89`sky1{iqx=`wn+^J6cW@ENez>5ht#mqJCmA~
zb4jG8o=fZ+w~`t@hs~sh1MW|1n1m;i8ork8q~^A|IjP|*y-jM6QZjY%Pa-u8eFsv*
z5IaeK4MS-q7N>*^<_V;RA@4wH7~)83ii)2J%`(+_l)hwNV^Wvt3(_1x+dK4y2nvw{
z*W@ql&BVta-X@)u5?iDt6{ObuZnGBDf%1>%tfl2~$_t`~*A2GE*QS*KWjRu-OvL?q
zN@r~JUL`B*dEH!cgfGm+CaR6I3^ntMo@Y#dsqipi8=i$L5NJcS*tBa7z7a?%#j|;w
ziyiQ7W>7X31ijJ>%F-z7A2mbgLuO5#09Dc*?U0|6wd~-Qp9iG}SI@LVe&)2!4w}-A
zcDP`N%eK_5vQD%|KdY$EvG*wHbG+VyKF4S!S4lQTct$xdk@qO)CE{4lZ$3?f6r-_!
z=*M>=b@X+W-sN-k?=hDB6wkK`ea;4KKFft(lQrVIi?<8CcA7G!d9aXbtH_D$K2L9&
zKB5f>gKpa<N*f(15WIh!;xyx2Q`2EpC?Xxxb6J%hHugt4LvOwmGsQ8V%Q5xhr>RQx
zyrGzSUWUw`a`T>-jZU`cc~g$d#hWi~$8fo)wjY2Ow%kb0ruNrI+L+z>t^4)TQqZQi
z6Q?&&OF%F;xB1G!EnP?_99*-qP0IMvmyy#x8AHAEPA?nX<EbfcGo;2_n|$+>%RhHN
zU4Br0s-9f_shQzbu)-Oh;Ttcfv+wth!~dgDqF9#!Viyz^F)<)8Aa7!73Oqb7Ol59o
zbZ8(rFgKG?A0~gbSIutQFbux?Df9wSMTw*&APA78&9L28<TmUytUVP&`~HVfvSeGf
zv;JiV#j;F&BtQO0G3rqn^`$$UyPK0k<#izNR-GRd$Oj|8U}TX=c>v80)cHyM*cxM5
z#z6)e<D3jH^B_;!Z5#xVfG7iY8;>Aw4rJK=X-_haWgdTaoAf>7<h_1d5UJl`_xX#9
zRm#X<kwkQq^oSgYHE`VhJb&v<aj8D<&fVM<OQ2%}_3Np7?lgl3&JvTUe5wG?9mkj#
z#l`yZ=IfKS>h@3feUKYxMn_K5^q9Xr57rp95z{p~?zI_(4zvqafn`mW)yr@7kYHQ5
zqNKuXs#kxTiCa0I0VJM=nW%{;7zr{e3tpog;z>d)N?b@7CvnsqGMmFLD@r6WtYM98
zhh4%wN=(rAaY#>;EWmvq3yrdU_<4C$vz7bN4<N6Z++2fX&@`%pG>fms&!vrIB*kJ=
z@+R35-fv3~ZloOEXiJpJ2w}w*XN663EaSAwC`5nKWZZvbbf)1IMrYDjX7r7=6h@b}
zIK!_q8c-ABz~~jM$EZONXULbVC#PKubLq^aSdTo~lOvlVEO~;5-Yd9U_sTm>fht<w
zEz$%%3hn#%gqG?T5AXNV%Zl{BnI>!6zWOf2prfTZBG5W-!D;YuE`2G=Y&A=hobr+D
zr0{={c<~>vnHGJ)K{^Q}HC3Kv#AaoR$gBlzj+pGLy*5RDh~@r5MUalTu<H^xauP86
zL6<N1zVh?|UP?~4$O~S)ldVvGGGeoGztV47M3YW^F6R-JYLPnvn|zDlo~=~J<QtDf
z!fyLpL(YSe16fMe=r~CDNteAX(p)5Ml}>+_tVTjQkS1E%eIXJk4461X;!x~Ox~rw9
zk7sF%{;6n`ENu}}G;F;UK&h^hNmP5+5S`TF#`CK5=yevY3mSu8=)1t8joJ?lAqD4}
zBsSpJ;IIL#6eeq@h4@MOP+PZ9b%8wy+Q;#xcY9s>TJ&r1*CJ*D>-qgEfs#-LTt6zz
zX`F+r^^dXKaQ{Isqm~^DOJ@9*n{>W_yYb#v{)epI+E@Oa`~`PxwU<F95EKJ6Ff*3{
zq#i7nh|d%~x4+F5WVV;#z$Fs_G?!s?1SfyJSxavtHw?b}SIi%XmXxR$5E!tXWYONb
z$gSvU(d{WHvj6{3>Osv&Gj<%W(ZggOk|^@y<D(py{+VF<x;wt^zPuQkyb2b*o?d@U
z$OFKC5%{CClSfp};Pm=7{rL!>9bZSj2*4P=JU<7%+x>&Vz%wxNh5H9j$ZrPu(&c}y
zU-&xJ=dgcB<5^B#>mRE`TUO{p9x+j70bh=15*eHJ;d#irquKxU`uh&5NpgRF-HoQ0
z=oEpOzQ66>cgkbT<L!Yz%i{MPo0&gSz4PFg-`=#Dp8nna(dNxhve$!b_ORG3!)(J8
z6=@A_3ufb-iaH9+1;lh`nAq^gf;4|IdX|B(8Ha>6v4WOoke4r(=UH3k5}QHKd49s^
zBVQY`3ZfS!@{Y&4s}uib`JvwRVlUQ&Uc{63P<3nTD18Z4+fdh0k(1uD1!adc>gv8}
z_Mv!K@qh^vC}Ul4e0LG_^phR1J@RdKizZY7-DqKo$h?c+)+xDwp0Zub7;Ar0eN+(Q
zD3zoCm-P*#BI=dCg888$^HSb3_P${+ehUpzSJ!15_c~=>j*TE|247K)mw;RGVglP8
z&D!n`t#i>vV3qgSuSRp!#U9pa3u{8JybfouUPZF@3i*B}lF4KQ!X8~h=UDBBl3i5`
znN_jNUtjw8m}NDDq13f`y&Zq@mW&?qwtP90w<6^v9&8evNNExzZ`z!kg-6M|W>J*P
zt@cA%+gUf~o@jt|RgJqgBh*#4%`|hAW#kTv%z~n0W(b$+xX$^vwxP3rj_6HqqgA?5
zl$a+DlZ08CR1?M2)X8ucF!H;Us-);sP3GuQM5f!<;f<;0eWjCjq^N&z3bTkoWdf&j
z;B=tr=ik$h#2vj%AuyFFN7?j;pK`hAzI>~^9f7k8oU^v8<p_BX9ksHN1e9|%wQgE7
zeJ#y6RvbrZ`l(3FoLu?$3bjAfIiy~rWo_2A=A<wUN6J-t6J$w)M15LFGqbdN=<az(
z@9(L5FVXlpD4X6}0hE7Rjlk~m&TO%~HoO2sle^5g)KfvKO6eWV;dzncbB9sU8yU0N
z*}9XKb~%6B*H|UG<+@I~^cQCFd|eNtkt0>>=5b0jgcfx?dj+JH$!N=s+TC<@u+6lO
zOAaBa+PaMilSHY)RM=EzcUbS8dF=5%>K^a9>LT84mt2!8`#gW|i#}NgGxP*i*DK28
zY-6*`%lcH|GY^GDPsohog)3a!d%-US)1qb1wRG9{bp{-(mfK9g>y>qL7Orr;*w>(;
z+=P!(V$}ran4RN-pY_bf4b98Zs=+7+H0~tPRC_Y^B2|Pl%4XT&NhDPAl1?w_5}Smh
z__dA0+<Vy5vZsHSiY3fjM-IzAvbVBZy*?+v92=rX@&}x5C)LRQab{%I$XHUm%&M6&
z&(zym!O?pat-gcTfe(Ju{3r8C_hr%PqbIKNIPR{YKh=IlVlWDQ93_tD<U4R4rb8>A
z1Fk~sO74paX<M=`SFQkj0mq*M>5bv^>rBpPhQMix)+v85t=GxTpq#U8JLg_yjrGb|
zvy%A#M3&a5$i;DYc>bm6(p)38S)qiqd1`Bzrc;}rJG)ITE|xFP;udbmr@%Gbjynz(
z&$;3i6U#W;9zVZt<W4h$m8E!l9!IMpnNcFz+kFRoCb2k5gy-#FH@K7H8ahIA`g|br
z(OsjxBQ$?4v@b%l3PKKDM7Jc{InYY*Nf>{|O_Joj^m75ng$><Wmy{4B$AnbmKCg#_
zaxZyE%$owRmv}B*B)Iz7*Qa_b;w!&tCN%3W`pw?%w_f^3e(Ps{x{%$0o;WQrw&PoF
z1G^Udda@M+28o+_(Gl(?58R=x0Az}3zlFCSjr|P}$(z5Ptp5%0AB>Cxmq86C6PJ9#
z6%LmRq#j?lVu}{$X8|*p0oex=meMvSmxS^bq<;jhF7W)@=g%)ZktFC_RNo(;FaLk(
ziPB0l$?^32_ZR;FrC`#+2_z$OPk;UQ^vBD;3WN-qGwXlR=eH@4Ml%b>gZ%Y3e)<nR
z{Fw$zXHS2Jq2J3g{Pe%mziEL-fGy(zup|MvCBXay%vqPf4j`0tjHBBQVkKr8n6}GM
z1%GEw^<abASwU*YwN|Q8J5o1Ll3*+52zXLEB&AZ|uk)+sUU4`nA(Dz9DW4MQn^Xcx
z!BG2AkmuzQq`Jz)5doISqtqqc^-r()E-6-i&J+CbJkCb?&2LflDlRqZR`|b{Q4jd9
z9Meu5(u}PjRvH6=MF)F~p|`pfE=q%DD1UTuQSKZZM36DL7H~lm>^Zm?<@E-mOcnVQ
zyhqvcyC3NtSe7!)pgZzS$SXwMQbB{_-83#}U4qQZ6Q1<&lJup_K{dclnyDic9LC(D
zZdR*Yn^}<Hyr-G9Ub~)}K&z<Xw-E8I!a--t7>ys%35V#!5uJ$TSA~U2v<1t?Jbx$N
zBO1N+jkG}^LtQqkn*O-S1pc|R&^2wMJT2Uz-?zM}1?;ots56MDQF+cKYa8AD7(v3w
znFuLRpvym-@bKJ|=LmKneGq9u#6%|;mk48BV$D4=IxZKV(LOxIBIPtwP@JpGh$hs)
zxf)T<i+KEfZ#0<&MHQ0~Mb&V%!+%!!4!lSF!agpuin^jKQSbr#qAq3iTUM@sK@kA`
zPUFJ}x)1q&p`%tl8lk<42_79d+GO0KR!*Cphpe_0%)HO8Amf%LGJ%^(t7pRzgHe{*
zdQj?MM)zzu0h&k<_`oHC=KKmi#=;mi6$%Q}gnhr=HIkjw42w(|nKh$mn|})0HsY9n
zQnd1t$J)xJ1Ri3`!(ra?_8g&jJ(0S1%pRk$V@_)?%An?Vrfe<bWd70iE|qL-<CX|t
z)4Luurob+^-epScN=g=1n3TSuI7QgTbv~drc^C>_bchG&>s$A4G6Gi;n9*hwfzwl`
zas=K%T@yW0M56FSl4E2S<A3OpeuIzfV*Kd6t`}E~DAlcs=#T)GkH7vrb$5RGHc+Gm
zbKLZ60`Jn2G}`bGt1MtRhZzn3*o++r@Z9T`8|6%Suwfw72sz$rOJ;J2A#s+;>MNMs
zuK9?`^B8^!liTHlT4r*G9G^0zElDj}#RTeSCbfzyrnP&1QmZ&H+J8!pq%NC5tiueQ
zpx6k$CaXn<%!+x7`7z0@cxrOHpJ5LKVJ%Zq9#N6aAl6xh2P-B1pzIcvBHbyQG9(eH
zOBMbn^$4Md%mFjyIIj{$K>^r0$KvTqDpQS5p@vQwp9Ebe1z@|Jx&+HpsY`(x0>tpI
z!9XTOHbeFr=}=v2l7H64G^bBRFeMS!X?z3EaL6h{Sd@$i?*;l&IudbD=|~0jmO}S?
z8MBz0mAQjtrUEiV$z-(=ujRKeGkN>)q+qQ}s`!fF^K!UHsYj;}@S4+6AiEY;_IUEk
zgIT~NLO&~J#esJ@Yi?5CkTk_#_sVg849}Z9yVyxdo%s-wSAUs(cA9)h9Mh9a&COPx
zv3T9dG}0V)fDz5okrh+Q-P7zT6^-lvD31A7o-J{c5MZE3kg=o-U78T7!Md^$_;??D
ziFn@^)Ps}sE&B*WNJBd{9**j1k7U}K!aF`}Rv0aKb66=gVJLdYb73XY?(O=)a)d_Q
zzx_!^a8M-|-+zuCZLQkce#M+~_2`j`Ig}&(`dUCL+Nt+y<G4*m_{beI;06P&A_W-S
zPO1Ln4EQ=FoGPNAypQ~bJPFD&A_tl~H3t$Vw6`)d$VU#gSi&N#OHniOa}*%=z-F->
zGK1@`tRPFTU9L;LwgHR+?f|~;21pb$;Y<&3CU4=$P=ASTa7kM>XG@((#mMY7!30H`
z+KX1mX-dE`XL`XzdBa;39whbJVY#4@(egc!2M9C}qU@eP?+xbjdCCas>IpqyhL#+?
z5RDx!7D|`xu!Xk55c66vQ`6bL0~Ez+ahbD$tT`R#_a-Ilu{ywSf2geBS^RFH9ptw<
z#_txFM}Pe8FrVkQI>v9UXor_Qzn4$jrg^SpNC-QKk9jV0lzMGk(j(2k?%t+=@=L`>
zw9mN|MIixY^4(~U!r=`QEXhTR5hiNcxlLelJEjE+nUh#+!MZca8BH)uE7?PRg7qDc
zxwI6Kn;xe*nIkZRq8~lT91h4lZ<HM-GcwmU<bOI-nB4a3TetT8*iKDJka4({BHEgL
zsjAnG>402Pj^~VO3Wy8GXF+rHP=g3Lp57kF8YG(${ptsJB&=k@xE|t6yjG}}l60l?
zd8h}84abM6$Y=!sM(~=0aa#@nH}lP(W>K=<LG9AFO&&@D6QFyDAggMRe(Z_b*=zbq
zLVvo~SF>!--_pO3SR5T0F?PROXi&9sfAUmZM<)zDZI%>p7z&xY(RHx3qufx8y%XPl
zLy}xGrAB!AkRTLtxbKl%HY!0%=IpaozJ>Y>m2V+#Rrz-ISe17;Mavwh2b7K+UOT#+
zLf+Cg>TM-$sQb?$dz3IGiF4U-ipV?Ec7Le1L<SPJu8*3zG1ka)9CXy^vMF~8bYO?i
zk>{AswMOR)FJ-fj#~5vMThL4o>6!>OGTKM@2<SC;ei@Z1+cmtG_I=kSU`?g<ip6rk
zq&-XtHKx^Pj^b@H=t%QNY!6ffCaapRe23`z>F)ScSs1KiBVO{B@pC2R5~Q`0uYY@E
zsw)0mFDR)+D4X|^np7RBH4|ba*4xpp0dIa3;6Jl#00>n>-%sxv04<p^=E#xoW)Lee
zQ?YaEOS%SphVFQYl7EaD!>k#LXvVoq-VEOEQLDgyE<NBrI;P>!D-xSQtTY7zi}D9`
z3V_gz^2oBoVLH?2>a7ZQ()qJ^>wlUeaa7l;W<ES^k0VRVu&&Sh<G-e<SCz(-gh|dJ
za{h*6&l|b|csZH9R_E8(@hUYYvE5eH8m+62h$(D!Q&k=My(8vkG`nup)VZS|okw#u
zbe(669@t+RwbVn<waL!3dizrw>B<qFyquM7de=*rgQ(BFOE_YtPvFaG`F}&XmAQ{7
z7NuoYH*=u!KdJ8}(7W@D+P_y6T~KpG4Zb5@FCS5_+1pnfV+$*ov9gZL;x!c+t&8|l
z*5&&iEns~>J}D$yg*vIIDBj}*lVUoTdZcJ7lqntxPjZWL)yos*lmORnZrt{hnwf|;
zG=$b^tGAMmc0b&>-^s4hDu2t9s+Bv{J5}G7-9Q@J>s(t^$uC6=HD2!((yZzrJv3Z5
z@1uF_)^HqUBcn_8K8woSfvQl0qqgqj(vgNa9CA6Ce6=i+xz$O*Ryp-=)+|Y+UZ?ik
z0}q;4NGfkz@nc$Aa=N8B<htpK4aJFi9_NgTAv^d>JN?k=BTarWvVW+m8!R3R`lTf(
zzxH;k-2XJA8-m-{oSI{}iETBpY{c@?ZkWXDxuoiryN2zsX&YQ8sAma*E5^-m7{1Up
znD{Tp2RVaTz=T;;h7HwQOScFIU2WD??cmnF1vY5F^oy?Gs`r{MRa=dJx4T=fJD#%H
znayhfyjAa4*`$SyNq>^=9+ibdN{-9<aQ{BugePo#e(1uR6+%7Ivnod-nZ@CSceA`(
zjG?S$xc4Gp2Bi;%`y2)KHA(e24&k8`<3Dn(wz;!Vp~(vwV8Y?LQ>#_wo7~aR0$v~Q
z?C9l+tyk@WF}CST+b(kmR4Q%Pca{oTsl%#W1dg*SANRI^t0=SEhf&3jv(;E5{aS{l
zDqhF1>GSwG=Hw%kjyX$(wf>tDjeh`DmW7ugHy9KFG?!6w3KRk|GM9l*EGL(Y{uZWx
z8}6sCub&@yUJ_^CgrDI?|Ml|v@b?2<pmLU3X}aHkJcJiY0+UX3-~|)bcR&B^zCHXx
z$qx1B0QJX>(R!pmf#gx8#&{;ftE4}LI#9S8je685y~34h)F;259u0?u5z41EGKIwJ
zpP%15^5oY~{qj88%{j_naH1_9rt0v2hi8hTCj*Ko2Lo>@MUB?fBFl;^QJ3^gokyH1
z72%n;6S;8xX@xuFk&dD94$4#Qq#cuJ+#7aMc9X5b-Ivk3?IyTmmK!S15xaq90yL$#
z*KW!%?<aD#*iCDEtQzIn!s)}+^P788S*E0|wnl4dm$?;7Ua~YW;=Z`sjbUJa<|&JG
zMc+WGJg=?yq9a~1<_90cHE+M-1>+gg?&mFlDdQ)+p<MI)rD8%(G{i4K3nFOYob4Vg
zp~JoE=cP?fYnZimBIk9YBhGL+*^PuwR7X#4_>^#(;Wg2#PAdt~L0vcbkmQ0%E-59$
zH}9}TzgLPmG?+9@VdyQ+dFc~>%bI&@%Z!$kLW5E2Mj2DWy)g}O?M4CXn8zWMi%%gt
zZby|idW0*sIk<uW<~cf<z@P@cMYmQlV}Vj=2nygN1B5Ju5?8bOouqZLf?1#vDNM8s
zifQxi2uCQv5M&beORLntCZhIY2@<SBkZuXfB^Zp@tVKre7&0f5Ffbl}-uA^exn#J+
zYM_$pj(fdEkZhE!`piKu7**#mzBPenxJoF_y9!30m%PV#L;6YplyJJ0Iy2BDGivlZ
z9GJBWi&sMf#V}=J7CdYu%b6AWKxiTq4umE`X=7-jD_F$lv*Af=&#d_mAp0PP_Eesw
zwHwFqqN9nF$3s>@9fZ$+O3ad=gS*gKcMMuuIp(;1i=ZX33Q+EzL2FEL-FJ=!u$=b@
z*INNBB|kFZj`2$v=A6H$t#YNN(a*em3&LDs7Jz#=LB_^}VlZ&ymIMrU3Jj&VvL$PX
z9nY#;a)0ZH@(+P}0m9%&rGd_PyYe?TBlc;i2H*t^weN!(tQXdQuQz#8L64Nm9pv+p
zS+Ay)CtkkA3d;qoVOraXRn%?{eVXCS!DP2Cgt?lO$liurgowPi!ycD|-)0bo55&SC
zt5%)zig8#++jIS7z&i}?^9^f-s0cJ|SPL{UK<QT(3Ozp;**W|@L(*VvEnOh_<ty+0
zz?YwKu=K9`Cl398QQ7eBukN?d5MLpTHBBIl!BOd3ba^M3$J~*q_7t@Au`1e2VWV6a
zK$M5uWYM5rQ@XcW{t(ivMB*4{23-=K+d#RY86uSC*)w>Kj^W+M9_|l>ANfGlO0)cH
zdXebs;*i1ilGx~?NH*Zq1{bA!VZ*7HLfr{b(Wkhc;`vm6f(X$>ixdO~3fzZ;0za?J
z5}3T{ByGyUSSkfmQ#2j3g;r^dXy^Lm5{1kE3hmUCd+^Omq7aVdlu)q~cCw<((LGcR
zSvhfxjc}3M^7(BX=}MElaw-k`D%-Mm$2eF<`M%j3*LN9mtfTtoP~LYF<w>KN#`2kH
zuSXv@RWEISDq+a%l4JHBCrYg4rJO`(K)@nR`)OiMF47^`gg&+1RCM1i8K|M9SxdMs
zRIH_2utUZO28T85wYi-`xRO(0dI~{CR|wdglqJS~>nEqkN;4yTTT+%f$ZGGOlqHsf
zAyOX*^1DfFlnkK__ivMxWk?{l-`_e_G^N>AT(sbSwGiJ@eIj$5Do2AQs>Wlfx!mq^
z`5!%=hZCImp+dZ$)3Y#FTneXh3NfsTwo;z0u>vR8bBa=566%fX6Q|NW3OjI))o52<
zBE@>u(6|CpnytCwB)H>2-W|JzLuuPdDX{jg*e%(voDjrtbsk{)a#R@_IcC(9TRu)6
z9(P`UO7<*Iu~sVYdEb#JxjuyS8miWm->jt-?So?>gT9US<I(c}K>5u|T2sDe;N+pm
zy_dwsWCd7rBL;gPcrILM<@Oa8P%K>0>ZRgY*T>IzxZ?UV;K8RR3z!z<vlt@fin(uO
zkInd17>TB#i)z4jix3wQ#3dX`n`|${6=+C*?j6NC(nW}R6yk<1of!0dZ*J(U2GiR$
z9h^1vToZUD+$Y~8>_74%6hY2|=dya`uzXN$P7u1z%^;s}CS5z)n<k@C`52%Zw4RRW
zCcJYWcR^RVKTT$}6c71;aUXvHB6u}re;J=X7K2KQu(Llqm3`JK;4iu?3>0W3N(%XZ
zPh7z0$xwmm0ERFL?$Wq#avDXyFKKt7kP@0%F<Bltn%vF*9;H~=c*@#)_D2O*_b<bO
z^M&U2fn>g$#5T#mdGuZRHgy<yjEf2m#`lSzmUy)W&Dz(trZa=CqK;T|H;HY|8_1$D
z-U*t4mQ43&!9pc-37g<j>Z67e@KTI_R-@;X7D9+{`iXoA9BKG9AeS4AiQ5pwW&SJG
zL7pf25rJZCn?iKA+H@Vrk?W8ox9PvIDb&tI0{=u^`KY9k(s^9H_i_-<&wYz{k!z98
zOx|3^t>F4?s~u;(;sq-FuaO?cBoJJh+0Zw$W2%@ya9bd}s4!Ze1Z3;SwoPe&PxJ9%
zUOqrsgxa{Xl>jQ=MjS1Kjlh@AR-+@nMc$aJC8F>--faMP8SjUJV!Fx;JyJ-#vK6lN
zOXbBzJ3@R8p3=>scTt-7y~<?WT(wmnz=7FRz>d;8zV-TW^#NEa@xfQP?}xt$=N0@-
zMjyoA6W*KfH<@(+fA8a;@i%20Kaamtetui{J?8O_=$kNpTkuWqxI6y^bo)&fmjPlI
z6qivPF%$tdm$A}BCzp)}7bt&7krV|<*2|mZAlofbARZq;fMCn+w@!9n-)`UEK0g3E
zC%jQkci(^Q6g`misU+RBrgKUO3#$A6qx<<?N^R-7(S=e1(B<g2r*Cul1j^G073flz
zPYxCR1&S`+f9nTb_x#&mKE>yWPI}k9XNsnOLtoPq2xXL{%hE_hFmZn#8mDM>WiP*c
z|LIL;kraKa=KJIO+aGUSC~K?$iQPYbyu}9yN`Tj$20`L>fBvic_3dw!LZK{$kH6^i
z>l{gEh1V{U{P`Ez{gWR4M1vvt?%!$Xk2;O)e((NH6RZ*TLMFmM1_;jx%T9~nLnb?r
zFc5^qu%E<6&a4*JKc|1HqzHzSO|bKdU?;6LW&}Hen>IkSjR-_M!wv~lhG1q0I!F*S
zWOyVf8J+Lx{yq{=@X=p}z+z<rnT)@51P%6JY2+FKq-yM_Xe$~sLD-;9*>W$5jqoM1
z81xbNazEJ+RIIiFl`Ynd0U=rPT7pQjVYIhP_FAL|P8g#&J$Zj!c`FDf1Ai;xr4K{K
z_Oqj}{M<N)34ynaLR;@%^Ee7l_|ZN=ml168I}x>myjUrii;kF0^CvX7Tt(Um+o2z!
zzGOTHn~Y0eU)f)?0T3x#1!s8xjQT{6B?2<CTfCR>Ww5gngqL;EAd!2{U6$*Sxkh@e
ze*F1c$L@UlIxv3@6tQViMwvW&9Q^ZWL_rR~2E&96x^e5*Bm>!SP(ingy+Es^_k2cL
z2WQy}6x<Ld{MGq0L}dQ{lAR&mxF?ES;U;L|Q4dfq?$d|H^!L4SE7d=X(wjN0+_MI(
zBT*G}O318%(VbQUg%huY_fQQ48LT)!Kf&4*ZWas}1Q~ykQ(NTsQ9`|;5%6)?er&$7
zT+|S{hhlGIPVviOiX^j8!A&TdSq*B50x2}I(I|K8Mk^fy05`0fc;7Cf+6bp7239aH
zj1Cm8Cg$nQu1jom1I!M%u(l{@{)wC?WOP~>t*0`^z%T(<eajm`*F@G6F6k(OldfBr
z%%P@)m85?+SJtlQE;**Q!6nOFI_GgZkJeEKY2<EO74(dGS#pJf9>)2A9+@>!<Y<B}
zx+&)e0*nq+TZAdc3m6DKC3|diRMr_BP>vU9@qZBx#?(&Yz?{Run3pviFwXxcIM~K<
z!NGX+SsaiBrL-IeX-O{9wl!8*UZxGC5`j@%$3lNh#S+$TBmJ*Mire7@UK=BfoXEc9
zf240mwH+Is^nNzg&=l9H;+b$_p05TooZQc$03x<rivqMmMO{|Z5*13#QdZUp5}dG9
zw==3QhzcicPQ0LZoTnw5qZSSoA3fMypRjq!Ef^;YHn(2THOtBUINsRYk8`<AiDTO&
zDsg{|p<|(#(NtRcvwJ_gO8|%DGm6iCLPyGjP%dza3ps9;UfJljrIK5z2_>#pagzz8
zA-0C9Gr!S0{C>5OaNcQ7l~umG0u*SWwB;0a(E{UjDG-g{USHsHr5pwz8J~4L((k42
z&h#qo;}|LScvKj%>fw0NStAEChMo6HXN7-rj*o70zymuq)z_K%j{6ZvD~y@&88d>k
z2PR`9;HghxKtW~fiy_abz5w!!;!eo3x2I4(5EDJsA7({T9`D%bvV!-%qI+M->*<lT
zwBL^4EMYFHmo^akLV>e(0Ny|pK<}L&tq4=DQI<65tkY#z<^Vdez~{tsM$WBHM{s|L
zt(*c!7;k<jU_bib@|Gob&^t)_Nc@fNe4A|XvPoi`hNo5BNZ~NBl}2LDByMS=E@Sj`
z?`dp;Bv4YgX+<n|NW{;NAm?`1L23aKCNhuBu(UT{q9}j+c2vS98{KX;h*_5QQ@hPV
z2`K)=Jup7{KG*fw(*Q;t1dx7S*Z_Y<s?faiuWtY&#SVL_|GSsMM!tYzDyYAv|BI1z
z724L#|CLXdezF;Ap0%Fxaeb}XAr`Ls6hQWp*ys{O)*$?sn#vG@K<nC*D#@HOZ!jQh
zuDaS-P3vy^S8D9m=5GQ7jyA(;+qMN<Pn&~IKAuosf%Vo|>I{v126NhX@9%%>a=*5@
z6E6keVKkc^afo@vzSlROx=*F#XOd8yI$K;ya!UF6VSGns)H?2U&Dhum%A<x>O2}E)
z_3|mE5iwHuru}L9M5Ql^6<j|u%SF8y+0?ft!?r_<vS;o2NrJ)_dZFyNq@JD)C?^11
zb4K7k>GEH>3Gg~JZTUX^7$|?n4-(|_1$|_q2jHXU8S8DVLxR-(Iz9WOO#@L0uPm1^
z9_322xOtwUQBtY>Sg1TH?ocpcR7+iPhF^~&P`b+bEjqtu%{&?xa!(wz1QFjjmn|dQ
z#9a_GUef9|!3%hjo8UW<d!(%;v?0~Mj_Tjh+X`;_X)7I-qteGO=z4#GWRODZy6&c<
z5z(Q&oICU4)he!Y01UQZ+ZSr|tJnH12k1w;(_BZjCvPejDO~rnUz@VRjp$<?Gl~fp
z3-9B&aMLU<xbVK(8_Tp)bvv9F>hY_Mk87I1xHLk^^Cx%PyqIvSX{luThrfXHVRw`x
zg>rLmJ?A0T?OF5JM$dn@sh*c3={(ky7usQaN~m&LEP5LjlNq$!vp3$Ga<VQ`e7Q_Z
z(Y@Ist$rUj2K)E?d=wa4A*d(*BPllF5O=~Q9agQF$uXl{Uh}iOaJO+Kof!+mYJ_%@
z8Z=r-qhzBUKT1B%V^mWFu@<?EpxPa(NpkgU1G^>z{bArCeVBi<-*99o%Q=r?uyHE{
zIF}3d(q2>P?O(ETSCnLF?DD!&I$gE4(2aMEP6n3LHQD~-GGOC6X#MCZz}WJNTLQca
zrl-L?KLjY_4kj(f@uTc+j_!QRIs5!pM%g1ujn*NbTS{P7wl>P)Kt@e}=_#g@<9T0h
z1KX=SlN)##4rPDb2e+#McH5SWLP}u3aEgoP5$6n%GG;GKa2DoLQ!OdwaP{Z_<g&Pi
zGnK<%DVKYg<<vNhm<=tP(qc-_g>^$RfurgJ+a#;iVxFN1adumXyN&{=?b8JHK2#ZP
z_#eXY)IGsm5Z{+H)z?xZZEdKEj=L#kZNhm!Ijn0%0DFJ>tgf<g8(~<5Hvy#V>hJF;
zkSswuK5HuC)N*Gb_h#8!2fE><N5Lp9og>c2LixN0<j%)Ix!fmueffYVr#LT3?wW98
zA3V;Wl|UO-gyFpxHu~sshTMPQ)YNfB=3^pU8grat?)Grld>p5kk1tI$o{y8l=|ifV
z*Dnjlk@|nbBol`?j-`B%<hk~Fb9sWL-Sd`q$Ts%@Wnu(<k^g`OVVv3MKM?aVc!kKI
zh4lDB*MXRLf!}C6pmfIsctI40*!)BHN=f3WR;gEmz~PRQQYe(Quk1J}3We5(*LR#0
zy=CO7e^c8{Vk2i-u5}cS_nZ_NV>obcHfG9wODJr$QRg;l(-y_Hwo^S4$+lVQzb^7)
z%5K;mi&<@Tq?Jb0aV4%8N5%if!&9)?OJbv^5LtLm_Jxh8xRvMXmO`@Tt=X))@&3zr
zqk(M{T$>}f87;$qaMkz1_*7foQcAP%|2WFugpB_HXuEQiaRL(rGc+_Ymmv}t6qDgP
zCYM_h7Y%<|tl|S$46xm^W5BmKdoz4W;=N=K@g={X;!_m$Y)NAv2TgZNWU<~aQX_BQ
zZ}Rr@%l+HSPwyIURya}?Zr}d6DSjaNuabPz(9J3(j8?aAKW@K$mJ$qKcfL?ct@*P2
z-SH>Ae$vYE2<7=wub(U`{-qUPkUrs^uSfpwUO#_@=MhevYd>d*lxDz}_(c0MO7dlR
zB%-FU4v$kdT(7@>`{hMulBDn^d<`G{udm-;{`1lkXp|R#`u2|>FW~`dCA5Q^k-{=Z
z{`}YNkC(soQ0rQfrw9sv_}ib~Z{3|=GmE`s7E5D@Ml4p6`HnYZywKrMVlgrUiCId_
zCS-qh6BCKdHiB?(MC?!RPR&9_FBJBW9WW1xG9-qKWfdCkhZXTF^O}W>U(#v<13l$W
zBy)6!$t){_M1iDwn$%mkIzbs<_yv@pj_&;W^CIU!L6AhZNYIi<N@iLh32|Ve_KD>3
zJF<axF%&Ea0_azi@2{VX1pY?%=UL@Q@HKy7l`7W=vpUN=!74Jr<Pk?#1proVR$oI-
zXy#N=tzc7MpT<;a!#1IMXbNXBRLFwVG+A3dziv_EF(i<kB0T|-tT3b}lmtJb!*v|8
zMPclONAhi+M@|c6eLtG+$K>PLM?1GyKAwa}eSv(~euVON0!ba_Vs#%KCFVH9>?(gc
zfX7t`KnXPd15^mK6X@(t6#_KEPboX7LO{e(3wNpzkQbKx6)I$$;A=0<)^Jp|SJ7M?
za7Q!EukE}byn^K7!8?+_9#$MJjUcicgggaNQ3!%N#iBtAvUHJHv<4yFURBf~HfeW(
zTsW~54()}mLX+kgZFVYHjoD}gYI}cG%pOBqTD!#TZtW8>dkpD{nB7xyUCbUs-iq1X
z==+Jdog?us8P#ZeT1Ft9u+mJ{mGkQ=D?m+j<EYaG5*8-<u@q>)3vb;OG;0)~r`Dy;
zhwFXPLl9gNPqgQeDcP<h7t6@vH`wBlc~#^?xxllzoTN$lxM$T#vC8rz<tu;xN(s4I
zg;!pDXGp{6Y&}j~YN3I?9_J8al`8dkfiSG+;&P6^&w%ut19G5ekiY#dZ~x?nzwlt~
z-R%H~KzF~tap;e_jJ*B(_V-ZmZvf^H9sqMrVA&*B0Lxog+h?!?2+}@g0Z)V2iJ5i6
zaaddh%OdsE!;KA#g|>oHT$pRd8H56V^L$$&RLkmfX)ZJ)vwns)2bxk-Z!|(OEc%k~
zhNt&3m;Fhp%;jJQc+c3_#_|R>j`Xn#=rAX@Bt&9eCpT5+NeDZQg`t{5w$dBrj_x9%
z<`jI$+OT?)+z~8NLs$BXaL@8IS=%0EUKtT3j!h5fk)l;;Kt0<bM0ETVdym9_M!;p@
zU(q0BmE}?oKcGQ4h8!)~;&V+xxF60xI=U`z5~jSRTV=CLIIBH7yM(Dc?N&@(LUpuD
z7|YI{RqIVcEkS6hCy2~930b;A>Y=@MKUHva&^DTc-Dqaj(Sl>=x}hXvU)GI2Qs@{Z
zytEVUohWoH-%(~t$u-4JZW8uh9DiFaHoDPAlm24xyJVDGcIQP@JwOLZoo@d1gwP3y
zslw?Y%F}g85AlKqWo{DgR3U?cSG!7u@(#>8=O>={ybk3iVNdh%ZXI|oWjt=`5J$J1
zC{Ttty7TL_bA*>a7Z)WLtM@#!-U(Sl%pz=iC+@(xl!K?wn(xz>dlwfUf6zIhl)ES}
zcF;CDaoy+>fw5nAJur5oPXxxEd*=gVPl-I;`RgIt%ff6gzx6QNjXoJ>_eng5*?ECa
zOu<^A$X{g&V`y8ZFdluz6!xuKOkq6wj45>7yTBBBN{Zs<x<EhO=-X;9qA>zKxVea&
zU;wI*Yc3)eM#|xp%|!%Yf1|V4Hy2U)*l9h{A$l6bPR#g|%*XT=kv|=}V;zbA$Nu7B
zW^T!db9IhzXSC;85IYTq!D<VCQHN1m&CaHVH~n~3I*Kg_IW!ds+LqxX+`-a{%xPv+
z5}BKjr2DpH9KYI@j3a2dv6PKsLIe<Uafm$;P-e}O4h96rhkjN`e~l9b@t0~;JL8M*
z`QF=*o@e|!L&%(XE9UhI>hoZvq0G<*`lfDLdnM7&+<&C9p~2NuJhqeiGV;;NSx*X(
zIu7!a_o4ws#suSRs3;Z3b;-^ua#ra(WWK$Z7Ifn7aud6jMk*CeQ!duI0k*I6CXgb|
zkfoC#7@do_5m$;7f4(GsM|`7W9buUuD3kRy@H+PxCp<vTl2q+2>J~FtZo}87Hp@qs
zIpLsh{MA@ql@pA(ntsJ`>-Ga|KL9<hTlS+t0N8$3DOeN6MW?f7CYqqdk0u<cx?T>S
zHXN#&8~G5|da!1xv0jvQTZ`?(mIm8eiD@J&vt4RcM!+Dge{>)_`hc&m@rZT8yh0n{
zu|#FE2C5DA5$##CiVi9{uUWzS54ml%ah2c!n87-mUj++@7_a%`G-oUPXUK(eov%V&
zmVAbVHUC}IMZ<BPLuE(ZiS39b{S8N=2u;6IOo|>O9yU=g(?NxYm3A0npK3W>;>VXd
z8d^Cd&(cX}f169lvCDGg?{L=(TYB`BIp_y3SjNPc5ct!`RZfxCT&z!IA|K0Mk27kN
z*w(=#@9T$TGg?~mtdFFtk$<k<ZF3Ig)^ZNz=AzU%B1FnJvkB66O2bArp@cR4qz$uJ
zMZ-IR9_ApniAUO7ta<dZvZIjJ_0XE6V+|KgEO?|@e_UMS5>(^<15UV(lh-0>)fniK
zMoA^O7^<3HGJ{}O7N-*CyHv6<SH{%mzTBnWE2eR)J}>DkxFG|}J}%Au(s-v5f0}rd
z=ISovOPEZdE&Jydi;hau`0xbR47j%*l<I39buN}U@Rpv(7m9S{vK&e`%=uBnjY6}X
zNM+zQe>AxPHU0{!KbVt$*x<El<pe~V91qE*ou-;DlQ}*G5!uir=SNRLfHZ6>wr`zS
z_6}ODAA?ve`%~u!j1|~4Kqt~T+-vM0gIwq}Vldfmu%1(jdwPqldOcko?~y)nx8lNv
zemN=H<5$<udLYL>G^$<cLm10QD1B449XGl&f76ZwG6I`aMP~<t<FX<q4ewDy=UJPY
zuN|~_Zg#=w%4vyHZ^Lx!bzSfy=X86BZztY!Nqzqw;?l|A^j1taI&9{hkuH*;k!)8T
z*t%?6SUV(JR)Qza8a?j<$}2J>?3Q>;q5s$sGQ8fmAIU`GbMIP~b)G+fbwd1{IML8+
ze=iCUjrnc-qGWU_e%ZkDen|&$TMe`v(yNBoN0j4!;$T5KD$rrSv}xv;pIY^!QC{G<
z%Big$6IwXqm0tRp!{jkN>s6#4Ue$5KKF&q}iQtyxxUhl(VPIS+<o+1BX~u~#u6Jrx
zM5%5Xul`z^AOW1J{r<M>!jZ}x_Zed{e<oXY$2Zi<DVm;9ZR+ZtdMmkuRGq@%Z8C#G
z!(^9CDJMKj_|P_)a?BTh{Vthe^Ys$W>~lN~Vkc%MWep#)O{R<&pqtYL*?H>L-2Bjp
z$tQm5xWx>1W4G!!Q5(z(I~{3b_7b=i-;3c^e7RnbKp!Mrv_h*T>DneUbG6&De=WJg
z7;kVw>-ehZ#hl=x*gz`cXxL<+rfHC)KssHf+Gar0EJjc`Qv4FvH`#TVtPOSn2qo<Y
zu?vVH(P&SmoX%)+t+`!;97yLCtnhAdAMg2}w3yVvj{n*v+<T;m9uze`e^38*Bm9dP
z*y~~rb7+I}Y+VZ72XS6;i8dVO1=O<ZZef>_9Ty~j;|_LlfHWW2il&urADX%2rD6U~
zULB@somVc#>Z?QBwNMML)(O};^Kmq;aqUD+18DFjrfq!Y#!};j2CG@GChAe;dvii<
zHdIKH4@2)fq;r@n>&!MO&VPsvSt$m0q?>KC(ieW;a{Ah+&g&OXj>+yq5($*Of3ZiM
zJ%9cZM_+oU*FVgPNg2|8{txY?Z-bWsViy#baZe``12Z!>li{c)mzN$Fu784d7xw(~
z+xIs-i6rP-RNpUeH~+es3#Alig3HriKi~Kp2!er?Cm|hk0#3jF{q)Ptp9+agnltNv
z(Z8P>nwD{*skNqW<7xi*Ge7-}ZhlV>gR`gqhDSg9$2oJ#o&I_HGtuX@7OWsGaIKYU
zv>+TaS`u-`z~4$*Ah{Kxj(>y;zMk(j_gE^@jWY47N>3<446V=}uxJ0E^vhjTT7XqY
zrAU<&<jKmU3OOkZOoX|@pn+C_V;tRAz%YPUDN>pJAb6D}N;<>q<wn8-pXuZA$iNa7
z%K2*ra+j5m6he0hRolQVF5V;PgUMo_(zlM}=~lGn%T9=|hGGCQB7YM^|6?FV!w}5C
zf_DKgSAb7wPq;(E9hM(YoxkM2cjA6EJ_FnVm~i5Qa0kuM*kalwuTlw7-i;t~LbYH|
z*OaIb-iO!Kat-j3l1Y&7jF+@b8ltV35-!7~$aUc*x-RkzQ)OcK&L(!XXamUT6qg1*
z1~P$6=>=ly|Ag4pl79^l<3vQ@2RS7NMsVyn&fm`&ua;*UCuL)1@?LZ^zwb@A@N42x
z|IT_PBUC_4Q?hI^>y;O)hsUHh?-egu!<aN2v@4Q^ZG4Inw_SR@UY-F?I-b`bV*ey4
zGqR+!P+NMNUzzHtDn~rOMDNHc0)L`n!tyrB6*dK$r|NT=0)I`&x_qEfJAu9$kO9yt
z9?%~HwDR92IQt|&R0nFx0;0hclYDB;FxL`O;E6TGBG(GJ?M@`GmUx6DFtmgGOfqLP
z*&b^LNsKtl^(HX6Ni~<Cd`5R(g9e=|dvXk=<hoa}D?#vLIb32r9;%CbZ}X`r4+_?k
zveEd`+NLC2!GFy6{P%Ou9LlPcSe6w2HSNJmEE}~N&>_>M)8IP`cu8C7++8Sjt*HX)
z({yykK7x8otd@27nAi~=+s@+09TG5MU;p%DvTC{k)~iivtLa-iI>CSSO7GtJ;ZrdK
ziGooLJ`>Pa+pF%Df>$VnUS|!CMAVSUk{z~0h4^}IEq^j6<eb|K_+x_QHR&2NcqYf|
zON`YC&iaf~XPM8v_=I0GQxqm!34v{4D?n)hcd(GUfojdrkSk`4Xch~&$ZV}$oKYA$
zt85?Vg$WxvBxETc<WqDJ$_;98l0{Efw=6n6^_ZFq;V=3X;l4!23+!XIS^t!e$$qb>
zvSu$?(|;_rplRr%|Ah~gzsy2KqHr)&q*XAPLN8o!FSS-#>PtDJqVd=wA2w8Nq72iS
zwr9Ugt@yp2YbP6Yx>EQ^sMzETrUbf$RcJ*AUP%klwfuM<dhTNmAsqXuAt0BP=qk6$
zzg{Ot&Nx-BY9l@)9j)0vQc39K6vNL#yNAdgn18SdPXx0u`JipwTzJcO8Vb1DJ>t4Z
zJ9>6B;JinGA?{YITb;G1;*ZDW+7#d7Dp?y?)oPUq$RbIU*d;fNrkbd~HZs;y$LCGe
z{E`VVNWXREuJsvQxC<&@quf({T)FKcT~&i-Fg5h@ahW;998ea#q$@pljdv>K{U8BF
z7=IMFvys(4x%n4ldP+gVnmSZQ$@DA`MhjV7DWNgmf2we=UDY+Tmr~WX;fh^^{(Ek>
z1@3iCCpg?wgG`%%WUjFM(g^{*;J{QCXUYy3c@MgMilG8c3UpDHd04CKu{IT2Sc1fm
zyXFqiB8Yh(F9<ea$2?pgYUq3Fpty4w3V#(pOqoR;+G?E^`X%_)M`}Q;d9oi=7YAsi
zv3b}%9B-QN-;Xyb0!|XzCi9x~5K}^_tgkQ_`Z@&^#B^|RwMqp+Yo@Jev4*Q8k(bK0
z^pxVHg0h==`8rXWcK7XcY1LZ5amJ^<?pGBTMgC;fdOiEiF`z`M>;Vf)30lq`{eN;U
znQ<vxbk*x~Iux7q^=bB3ciag|5Rmg}r=*)s<zeTHXenzs_a%e>hkgZvX|DW`N>a!B
zZ3wa{g=9yT>1lkqR$`@1SDPCC#Ia&dO+H78VL+U_#$&dJJS#(`k#5en`Xp8%zV6~1
zsmKB~rx=O@s8hV2b_?z8+8HPYOMk29g~w`DHjnmZCBj|dCrwnED!-}_FLh&iu0vnW
zE6BP#xax12lN{e&m1T;nx*NjoIisFa;VIO~k>;evT1!$=d<A!s6onk0a;8h1wPRkb
z;iQ=}U{$*sC`zkXn5W|M_P6hTze|f`cS|Ho<n5NXzojL{cK1<|84k1rnSYWo;@--;
zI-HhAtR;bBMRO}MRnuieIr3WFGV;wf@BVEEJjT0GjN>-%`H`A;xy`#3RQn&}-T4u~
zjQ8JLM!Y@|NlZmx#yEES?{EL}?F?$cobzmxXLDB*+Jf4Wfiz40c(-Yph7;7PS&}T%
zH_Q!gTc&S!7!t%XA!Bdk_J7L&9?LA+mc?z=0w<NC44YYrJ=IQ+kiI~-X4z*1xTAzX
zwv2q)W<&s;m}CuxoIva{ejvwJW~PyC0|MYlmi-9P!6-p*B2*r!5R%)B(6Qys^9Vwy
zm~t&ZWs|P`0R<F4zHK)7i6ZJYYg#60_G6?CbI{k_$|m$EICn^XQh$I((JBI4MbIs2
zVZ~JY7*X7(^L-Vg71MJ*BQWi%fySL<<jY53Me;nNv!e3uE=Zn7Ip`5wQ6SU7N>*|8
zkAOI2;Nd{<s-V0KxtPV_)gYTASiR~e$^r)AhVYTIy#~fashy@1gpPhXN%mO^f#BnR
z%i73LeCnuXtO-sciGM-VvCJnJT2LU>dyZ2;WX8;P`sVIlO1T3;Wqtqs2!I+GQian1
zI0#2{913HPB4}w9isJ}6po_K-VV^-F&m{TTj6sQ;45IK(F=J4weF-GRbVSt#gFbT}
zXrB($D~LC>?N5{iIXhd$2afBjBik;?Ub|f<k*P5YcC`<{hksQBzp9Ghbrm631`%oZ
z!i#y>=mO{NO5rC|ZM@V^kEw==SJo!J?;y=uO%ys7p?{lS;jY$py1z;K1?IV)<aL+X
z9qLd@U(-{h&;18Cy#`V0)|&e~ukM?lK=NguOH@>gXRMeXr8(LgdCQ0w43&DQ2LyCQ
z=GI~Q)oI(KbblLQC7P>X<ylc|K*e3Y5)=@=GSn?7Up7IZBAz*alD43P|2z<ekGbbx
z`;7OK#Bf#prz=puKQ;Z`dyP1)NPe?Mu><|o@9ac}iI36T3gdXS?a|KdR#2|tq?z08
z^wM+c;OlDovu(MGQZCcb9a=6hF6W#?Ly1}mvj^fJb$@;@=V+RFP^|0jX6a$`#J!nJ
zI=}N6COOm8A~^-c{L%|564&#Wc)v4UOn0_o+z#_b%KGS1gWSNp6_s3ap+tPF4KYoy
z8*!!(ADJ(}0&jqIyMV0%aR{(bk3y>1bphK}Z9Clz*t#ZaePXjQ{ON@HO&4k#ZmUP-
zxS2Ors(;TP1O=d4hJia!qRc4qor_&r>L_v1yalcFbXGPrmibHb4Leq}yV(o%yL_GT
z)^_Qwn02BRVc-Z<k_mvh2x(FiUT3pRIisPMu?g)v_RW7BIXE5PlNVC@PSwTo<+Is~
zCxE;Y{qj}+OQfk!SoxT*lo!%+Zin^EIX%<sbAO}gb>ZXHJ#R;#%JXHLy8IJAg?^T2
zCS~r_*R!e~-_<-U(<}ExHP3&oXxXxB-tCZHtCw9c$NduaF4Gwh!d`eo1O_LqbWG>}
z#gGE7TFgfD^rBSMK$$de%`ZBySuI}5anoDiO}5fbAsV_1q-;%pWj=B+pV<?PtLeu|
z)PIvyuKD7R&O?)q&cAHegrp5}^=8GeK_G!jyV*8_p#f0bp#{iH&RE5uCZ=Qvn;pKp
zyBAsyCo8~&Ct2+@m^}DdAyw=%1jHR>1ZLRN_uII7?SYoQ70d~eku%};bo7Bkc;#z9
z>{RGc#jj7!(Mclv{rvO<Fw6HfYXvZ>dVd-&M%{eN=uGG-f2~iZC=-Tei%}~dNep!a
z8b2yFZVQ_HL<P;LkFUaiTRZ@bq9d0M?4G%<wedAGLeI=~*u~dTl>PAdmb`%j9t%b@
zi}Hx@Rx)m}QNgT*AEtt)Z=EY0e9MfsYw)Z!@M%$NZllIJkP^fJJVPs<^3krM`+s}*
z8JIcZXDA#K!ffD)<cUTdF=_wqyCMzH7izb&ai0TKy=X_@-caP$&7x6UA^TC{WDMO?
z`?~gC(Kusl*tNGzyB2~yEoYQW(OwpUk?(5SJ0?ZD(e9qTrfn1+K`)HyY=hpYWsgit
zc@RB|Oj$gP9=Rk+(M4~qU#}>yv>G$+hV6PyHDAkJ>Y|qyv(op=+s*#~F=gCGmjPlI
z6csf!ATS_rVrmLJJTFXTZfA68ATu<V;h`ocf7M(|k0dz}zW1-F<BU*yY=^^-p;c0M
z^)%AnSVmlS_XO-*m;(zE@b3vf({?^9GiQNAbya2AVaLPaFCy&bmoFE7`Q5|I=Z6n(
zpe}}46h>Y?|9lbj1E;@&(?g9HBLveRE}y?%et6<sYI?oUOK}cBFGHU#z2o%}grye3
zf6+@^AB__94FtVpe3x(Zx~0$d`sjZTbV6VAlqh2CA)mu9;BqTYFHJ3#0`%ihJ7N5I
z*PlLr_rMcNoH^}3`nUeur_T?6d*~TN$(a^#`P0`2|APX-z{-n2#}t>BU;c9W`Qi6H
z*)kZXJ`DXI`t-|}OZVjaBxA=(#$0Q}e@4cFli<}wYsbL<6lJVPf=Dt&l9M8NBXOxn
zZZuPR)^Pjb&5B9N$}yovlu<`2iZp>VilR|e{_d$i$7hJc^YA^lrSF`yk1uf+B%#aj
zC$Qf<A@=i(q;$*~+s}wZkX{X6rI(V745{zMZ=rSJR+1{od^K7Iq@c_#Pu!Y3fBU;<
zEf9#J?I}w>Ia{FxbKLZ;VKk2@MZXphlH~onxnHW>GQ)L0;+!;v^!D3NuMD5ejHmb4
z!5HQQt_)LPYV!&`Of;YH{Z^Tgh`$9Bni9$_2|yd-l-8sutEzcYD=5zdG*e_=SwWDD
zRwZM;#m@@!Y-SXwnb0F@v-5qse?DsN-57ee_G!QR+VO<yT-qHJlZrcnuLlSkCIyZt
z;<SZH!IXA=Ym)pS<w|5IT%%l>l$ngQ@sn;|!G`z~x~IrrTJu(lSz<ych6~;#7{F9N
z4ouV5Q7!*qhc9y-&`G8hl#yO`8B_LE9ktGwhSr8@gub=n_m7*rUQbEvf3~^240}#+
zKSiU+%0y3*zfQ}>0uyGemwk&|I6_c%vtArBr_Rue<ccXA^ZaRPZ#c&kvm01e;LsYG
zk?IVV6UaEv6eTDo;Y?G^kY<}HN)$}uNtP%H(;D2Y7OfZ;r&ywq4CPpe(<n+>+6O?K
zWs24@R2vtcbij-|!xDi@e+C>jY8iYWAmmAE20>t1ij6#j1cnCB;Ms52aN^<vkc`{0
z^2~mUiefVdE0%a4MT)fenGOQ{&b-w>fZr;`Sr7vJHYx2!2LZ-1YtL{HU=2Z+%{==p
z@}#w*{3gk_x<o~IUF$PAR(`t`XYN#rfV$7&EH9SQ+V9YE1loFhf0_U|S`T4V7k=-7
z@`TZ8ug4n7E1gptDM4XU%%&2Y7ei<~g7XR1wN%&Rwb0^r4BND0&h01}+qA9sY@!*7
zbmQ{lNTlb4Ua*KpI&t52AQ6hc_*g6lfLg{j9Ww`jnzoEyBW`E#v(d~LGii#qL&#;3
zp%HR9Y-a_zoMvYQf8A`hvw~m_8(Kj(rx{vCIj7kffMx(AI}>lCJmW50($AA#)47q1
zJ0@j5^5cHF=^;6?l0ys`p!2hSs0e{@)9fZH?IL=}Ue4_vIdRcaK~weQ!yXhUH>vm`
z%Nd@h-$UwNXMV5ZX@TV(F;vGgpUp-~i;P~Q9&f3>qGMepe^h1svs+{bbo$*tbFsQ3
zb7p$I+%a40$SMPRS0ytTHhNnL4!74hat5oTxSAPEYD?+iQ2VgSOdAsosvA_+<a<Y*
zcewWy=C;$HOWkK>wypE_QVi;Vf>JRGr6yEgQlmL<ugp@Jt<3fdQ5xPS3vP-)n*-Mu
z+{jTIA#%Mfe^VI7vtc+(pC?FCM4+_vB!BpcU;a!#en*XcHuRsN>DMw0zx?&`2Vc-z
zA#6!Pp0H%ns3pSuLP(zE70C`H5FF#m-A`iU%*cE|;4)cKsl%RbC`isMDdF)Q5OGy!
zu#cAwYuv4#E32|qeMaY2D+N$$C7;r$xAf~veo8nzfB(*_YShc%gv(kP#M{_uQQU3`
zuV{|KW{~=jP2Edk<1!+$<fr#6qkg?^I9Z^SX4(};14&KfN*}@Ur)(fsp;JPGw(&<W
zZi9h<6;H#atUg`CiGg4YI#XAwv>Q9~JOghs@qEFDH}Wrz+?~{_=0++*KnNOc-u>%+
z>@qC?e~Fq=^IiY}KSWQPhsUeTd(f~!Uc*s|7%gXkU?V&+hsac1i?PLc^<Fguq(8la
zE0L^huM{gty(riTUUzU%%O}MrBO{}z*;H8~U6tHLMv03G7GN1V(NKE#??nX*1r#Dc
zKe#^?gU<A}ZggTrX1JjSmm93vV0L$5yDxp7e;W$Poragmy}YUdo#Z4_+wu78_0&wU
zR^0-%8PzR`+9&pC>$ykN%GX+r9_jhHI(hI`3!|E3o|LJddd9Q$$I}Bv0eaX+VpNeV
zJ0U}GA`9GRXNB2Wg+9|3DkU>^fUhF2%6up)omY5_E-gRB?99TUtH^_Z7daS0F8bel
ze;^7dEHZk8&nTXx0e+S9P$YKdr2Gcwp*qSwzS()6DF=@o#g+3+IbqrLnhCPO`b7e@
z9EK1?$@}M=q|niwJI$2htFHBb^9&ieVr5M~qqQ?+F=^R#IwNArYBGe%xH>VU?I<P)
zaz=Cqnn0++Zx`Kx(2P4gGrIE$GA-=+f6*NPFa<||_`MW1&WpALMPA~Y#CG5-5;(wa
zm*b`Xr`T-2gjOUk9C;Xd$RzG1v2hU)S!v-{ip`|;KjJtxi;53$CjgIvK_gVr86TG4
zb%Op_B}#9Vb2O^}+@E98$~1F&_s8W3OCS^4_G}E*L~n5gNy&t*Y(jgHm1O4ye?dle
zV2C;jWX>BLFBA)Xn}kwFDkYCRV6+c!0w-D5v>#|<9euU9WZQ!({OGwiZEKo21GZW5
z(s-v!zL(-mpKxnJ(-~PiUu3k+fTfE+y*)9CBRU{MRErQH%{%$N#xp|`+R;_q*4~Ls
z)l7y0-1~fNC-R%cx%-_5&0TZ%e{Mt?u{8IgNi}!5Ja=iy@!U6eLN-pciY_)jGm6(L
z?LAGGAIS=r2SQ(qGj|Ee3r9p8r4nW>+M5|glk4p?@!R*0LhH4$j0jb$W~>!y8zi;M
zksOw1DE&cTL(&67y^~u1{_!6l-#<F%hE;;o#wkPMrs@16m$*Dp0pTQdf75G4Tf#Cy
z7_8~?aXI&pJ-Cnz_1J^oQrHNxP62sm@+toUIq1CAxwFzo<xeG46+`=yg~*|Epte{Z
zR1^8TNPo@Q!W7gP&C=Gzz$(Qx5i*CIM(mYlTtHAR?hS$boYEBO*?QBeG_bN?`O`So
zFPK(ooS0DB?<n%63$9#se>cPLPM}MEquPe;rVJNypojLJ38U}@0@Y7=DmEmWno4_6
zD_?DHa-YO)BmAxAq8f9NGv*Ro)fhl^`P7`Ntg|`_T?KOzjREE?q7WOsV8bbYUg-l&
z&|ckk@@XGJca1fd$*2KI1%dEZNFKVUmf;Cmd;#}WdRbIGCPgeNe_%6+A&(e@Kj68n
zk}5OkpQP1RuwcVO$|<Ivy2?Y!CP?iqNm$Et1@cTAx|C}`U6R_5;4`vHsHb!nt2%&|
zeOL%e$PA>z$_%9A9om)Fb=rwf<ZFByZZE>J>JiBaxSpRigF%X1(Dc2$Cyz5I*Pd>b
zmo_nvoMxT08o9#ee-K5XqTA)6@&P>^_J}Nz#o(}^HzY;#xRtG>HqX^Tz-*hvR%`yU
z?L<oQWzcZBg4XCsEoeJskw#YLk%<}Z!Mdv{?xeD30!!HrAh~AU<m%B(<?2z_VMAy$
z-k(bEI^Q08u1PBO+fdCighL6w)ZDJcTGlbkp%>6-TSnbif3)<l#i~|(Q4q^GR)mnE
zndoxaEu3$Xx}RvM$|zoZKrPczDntzjNkKQd$hZpY(n6@!L~UZN!QzS23d(^d&e_7O
zryU>J&Gw3p(B=>_R%Hna$M%B-LH<S|H1t;l@2_(q7o#bu#0=DsF*%7a#Tr}&QF2}u
zZ~oRS2EMgIe;E|Wp<bB0cj@WL>mZt-p7`|Mt<)BtcFebox{`ZKBlIhe^9?I+nBB_C
zLlCs#*n12BK{JE0H-I_IRffq58a`K|$Pvc#{_>Vq45aGw)8rEjh+n0;B_uL%-{Q7^
zOWFG4kh#L8o_AEZ;w{W>zT$P_KCyOL7829Z&HBkwe_<k3#JW+WYIBBI;SK*CT)SAR
zxNJ>Y`5=8aVyvtM?75>d<&!aFEWNDODpT>ydOwrtYM`K)OTpkiWqSdAcQvr1yHXA8
zTK^AJ1G^b*s|I$R=0r}Bu`;H1MSCs)`vcxq^wj1q04rKNfTOYbeg(Ib4jd{kM$mDB
zhp&2fe-!5~W@l&7w%&VTK3((Ps2A)1*M&MmFX@W+HlFh}?w;qJmrCL_?`<sS|NU^B
z<$MjQA9v>DdiK!gu$-<rZR9z7^5Z-wPty@Pa>XfLn(7&ux!sU_n3D{JJ;{Z&#Afay
zWe?Vkyx)Pfgii2Zi)`$1UoF8E@6|Rw=)AURXC(Qq;~Gh>YZJfOb9MY4a$K9(&3@}J
zyV-4Zqw>T5PU18J%+76_LZU}HM;{5hZc|A5R}#k+lI~=7gbw)BbgimmqdyLH{E(Ox
z42M%U8_tMr8MF^A4LROjN4xrWik$xg=<g{~mjPlI6ah7tQ3(PQm+Y1qD3`Z67dZ!_
zsE-s$QRSESITu!c=xZMnM;&{>b9f?srs8-Rq*O|38VBhFjxL^`etG|C$J3Mqev0bz
z^?mp2uAh|AhDol5uOB;lprl~Z!2pusbcf&mK78AKE*vsv&K!Nizi(44tzix<vHbZL
zKKu<I{)EKRyWu~9^rH;JhabbAaeyVpj`6@)q5}Lm#(V&O=ABP$M;2Oo#+ARE#Y)Kx
zFyqd16`Xn9vkhkF1ezT;M(f7x(A+>v<gJ-U=F#jNkYV_z2gAu|I2xcr^H6AbTBJ`j
z2ZhF^3{e46DKtc#DbQ@SjT9QFEmP^|)JjkHxmBq0oG1R_dF-w5J3mCTs}O2Ttzf?u
zqy_A?XU2<vtE}Qu7AqG5vQ-|oEQEf=t#D5UEHlMU&_P6P1${%(J!$KRj)S|bZ%23J
zaI*D;%a&G4z4IOQS5C-%<`;UK$I&K6ey~s0V&WAnQbl`9yvcriu$@mw1A|912czn_
z0WMeI(9G=*fB9qVG7SefTH&ULqc`p(wtfOyMnzD6S%GHj$sq{m1VvZOv~#VX|2-=x
zdUpm6dO;~*DBzDVpae`wY?Fl<?}A2DL6by2#8q0a)Y=;)N}e=PJ=Obc85CrcPX59+
zqc)lg&vlW*J%zAEd;)Ktz&wU$gu{uw3XL&F76oOSI?_CH2yN_26dO#oM4qqoqx)m-
z7nFDuOepZtKZZG*LZxaG$Rh;0#8uEdmtH&<CV#syO@Hv)Z{LTm<x6rCo==ApbW>Ul
z+T;2#Hp9LU8tQyvHc%<ggz7nmKxHXg@i9!nK0&5P-ux#_qiOl_Hlk*+MVMY@@iHo)
zfMa~d19artw0wD+N5(qlk)#`iGOe%>Otn<hh?1OKMyn75>6mi1ACVSd3bmJG1)9=@
z6Mv{gzPi*fbux-^T!XzJhjzuHhjQq=v;$^&@l@!glE`6I94=RHHk_6(@JU&PMn8%9
zP)Ds(d<%ZVr<I@YAxF8SBB;EAAH4*b=N8Db8e55$Fi(nVJIz5(o&htI3=Ik@UoQt*
zVv>)h*m1+uZW3sPYe(vy8dben!RyA=GJoVo9kt?@k81?;+@7GvJD;(^6LmAuS8z;T
zr$hN=8XISsk|#(~z~5pI9=8-quG9-I8YaA*(aVX&8qVy^47Op`p?zYn4e>SRx@y#H
ztqt)t#@YgR16w^ejZLQ70=!_U4e_;xx<(_#PNy}poITAZ2-&pJKHOLU3K1n|*?%d*
zbScxCQBM*yj4Cu*mOm$Fw&OgNBd#j>C})_`H!b;SZkV@sDEVlPQn{;2J{PlCCmC++
z(HQPi@=@L}Ato~E|Hrb=W}p`EceohURb`)xS*%+O#MY>rlzmQE1E7lwOZV3SsosV_
zx(nPjvJcHn0J<vL0ih8Y@(~jMk$?Buefrn;Z-4swSHZB`zljvrKbo9C**?-OFey>V
zC|SH$%vk5lKv7B>H7LU{cSVBSu{3-gb_!jkkW>jNtv{&MW1@`0SDga_mI(6zRa2IM
zIF7ApydQOAk*W+4Cy~XVJK~c1(k=GSz37&%`aY9luYU}?Z^Q0$LRA~EM1SIVWb6(8
zn?hA!_@(W9g=}sD!*F9MR7S;HT(>i=^?XNCW}Lx@sKDX&LEyJCTA6chspg%VkXPoA
zn~G>%`wR}<T(M=$h?w?bkdgre>lX>#i17IR>qqL>2CK~HL0~9S+6)g$>yXJHui!Pc
zn=F7P@=TpGs6+C6zEtn}<$s0#YK{pG`FW^BigSAyC--?ZAn4S!)++$%SA7Kpj!UU`
z24bcXoad0Jb0t`&+MIp@$dxvqf<+Dq1k`h>vE@2#s-=y!y;9RV#E=pLNA7pF<{Nn}
zgrr1usH=40ne1s#Dt?DhO%8XdLS2eCP3^D>l>r4+arRmM^A*B4a(`Ps4sABqE(?lj
zgdp89eb-|ZqN@>18RV$?DT5r1D;nfKhn6e_j)y98tT;<h9Gx;RwORT6q}_Cl^tm9l
zuV(D(I&7hJDQVdkF8OpJ(oUT&j24jV&5>4Dm^Rg%S*zjc^QF^yV|k%RRp+gBMI5%)
zeI=k<b!JQ5_iVlxw|@-C3ay}&??Uk%Msruk%^UEJ^IT~TV>l*B9kDzSBNc)V3Fo4r
z4f{fY(aQBB&Q#p;F0z;JoP{JWQIcF^Q3yfNG4JD=mWh3wMCEEfRyAeiZ*+Brp_@~Z
zQ)pvNqT?1SmQ(nS_1KM-myB#XURN60vWM?EzHH{0H~Kxtxqq^NLPsik>J60|5I{?_
zSFE_IS_7b&g%bq}IX#(>2%f0t8Z$LKyb)2GDU2tf`P3u2>fjvDnrX{zg_-8DAe|Il
z?yG=PDs1|(=gc7~u@kf$y(A&xU3%N161Y9qx&n3t%(x!);0#{sGw3=oIa`Np87>1a
z`I!0D2Tz;427d_4K<G}0xq33iLgPE)E$cBaotBZfVlAfC)lOr3<usoVb;^02#sxRW
zo=meL-l`GY$7rt3Ph}+b%<5ZcB%LSTP9w%Kyl$uwYnZibHR3eGtJg^Hvi6&`Br%Q0
ztZv=9k-{^~BTuw+gQzbR$8aLoSqr`E;*_0|<8XSAnScEqt^^CB$8aw93hd^5*vtrE
znKdwZF6<_DkC13w_0CMn&FHkE9sNl>o(OP72_DDCxB&fjKzPqMPGwUK9x%^x<_<M@
z@POg$sv7*oELKY9aQ_7Nsln6Q=hgZAOGywWUO1qQX<UG!(9Ldh7V71jC$)f;<+wb=
z&G;5^Z-30U$cr(_@p~!PNXGq)Zf%wE#$wPqy=@kF0SM*L-QN79$p>c%8Zb*OZdvN8
zfS0@-mCvGWmsd&4%XIXrqEL<5=is#n*XC(0z$MWgI(;X&lwe{!o0U~PQCIo&RDCJM
z=V=4W*p1HPm|p!i@kO-og8^zgq`bqV=f{|X(0>mPV6B7MjL-t|fcYz!J0UMQ?l5Fs
zTWTuJDj$dzWA#Zdgg-C)o9cbi)H3PLt4iMznww|6fTp0={_UO6#81FPBRfDn9oDj?
z>9ieHrhZ3!{N#EiR-5OvfEACi`sT6i#(cB9!mGglF0oB`%cQiY!3|74jj7(Eu6)tB
zc2dsg&Y54xkk1ATZYUnCe^p>RE(>_O5bp1?l!P-(+VN)>vN~Sk;~~{~_b-j;{D1c8
zDy~3ECPAk-kXNQmF(CbrOs@3*o6%SR*~V1jJ}6r{m&ijG90`Mwg-T>b+*gN}?L!w_
zf3~Oj_fBLVuk+ioYFDyy^SXUL3pk0GjNE5E0cQqO{foyogDq>St}?nNf2n>YiCM~n
zaJ&r_!BNoydQQLRmdSCY1r(f{T)*wz4#)z~Jf<P{0h)V;&x+&HRjD)|Wj9r@@;+&I
zf>;+|r?$Nm4#e6(yLrY-NJ=KJd*%3IHx}zEyu?=c-|6SX7C9NyR66!2@8t9q{PISY
zJ&9!TKRK9+W{G+tn$>TEoxU+JzBBbd!4obvmjQ1M6ah4s(Y7=rmq6qK43|2tG$?=7
zT3c`1$Q6F~ub9W|%W&?O#Uem9X|@4cG>B7RTdyI=5}jx*lPXCyNsI1p?{|ixC7G6G
z(XrzIiF`&pb9k=boSBi<S}?)1){;rDh_Fl>O@w1QNDxLaqtZ8u8E5Gm!#w_ja4hIV
zcov)|LbE915s1j7@HF2-v~)>Q02O~DC?}ZIirxo-(x^au8dM-@)-dT>3QvP2KI@ng
z>0Q<{1rqOYhAEx+oMOs4QW77^MlkEhl+#HtQGu+4iYky{+6$(lNrD3?o2EiD;c12g
z3v|XSJW1Cy*GF<pnu}@1khLB|CuxSq&{;B)1h`1@JU+Z8iu@Q2Jq?-}p%Z@vw+ctD
z3Kk9)<k;Xr#Y7oqbaG1spK+u{g#WE2Dnw9`w2$EPl4ihW6Oxn&DIr-7HB1QN2r(4Q
z3JG<MP|Qj}b|4Z~QXCDYS%sH_^2{b<j4;d^3CrojTC&4P%ypCtMqsX^pc)9ZK^VM4
zfiJc*G$@@Y!<>kLsKA_}=pcVZoQkB1jLd1B9$HWoB7>xG){w6iraGM{3CeqVg=2_h
z;)A#l6NR}kDTVa#!jNPLg;yv}DuPOU(8p7>K}qJVO7T?8J8~&S)_Wh16awpmBoAn!
zY+}>QM?u!27=qCB6v-2$N_-KNqgNoo0}>1r-W@1R_|C9E#)Gmf*noemGE_I7vc*`)
z3rHh{QHYW>NtAVMj)j<<2Nk1|WDg#|vx!2fwdD$<kij$qT10sH=%=5u_g@xy_M#}t
zhGjpO)ike8sMN)2_UG(P_7Vfs<!SaVpEM^@X)a+il7Le(NGrJsj#7c^czTo_FV7m1
z_m^35k-fN*K51rUksW_$e}DIe{y+QJG>iJjEc^876Th6)d|b@=hw|UqfB*R3Wm)Cf
zq+HbLL-WP-{D<eykMQq6T}_lcAP>CeltoqvE>SW;aX9kfRef~=JkU_?UH0og{t1k8
zJOC~{a<o{^=chZkJh+gjN^(Mk{haBq-;~L+m*u>yju+!e&eDGbmc1P}O<omDwZlJs
zZhkp##!U`)r7vEWMT2*|Mqx++NTIj9Mlnb$LOTF5Nvg;+u!PU3i!=bF%BaL@fVvf8
zH2^%wC_6S_FeP9+Kq4zCS8Rb0+pnHjlD2w6K&!8w5RmGX0H2<8qLTnegXhn)w^cbg
z&YP3$?W@<>`}}`%bGnf`Z^s{U_@`*{qN!=6u}Q?cye^m3B(K|*iN5-4KAnw!E<dLP
zb0{ev)hXszm<cYqdR-S9W3EDjXxtFx4TdguRw-^#whs`y-Qn7b>Q?MF;&7-Nt?0Wk
z402=KQ!x+Yp$(@TYPyp>(`~%F=`EVoZG+Rg;v}^=q<Mepa8`uv2-gvQMYzOsdN7tp
z&n0>d#2mnghQ|?JqJE|3LJUC2yx^Z^7qbPK&CBXTMx*R~HD2WZDQDC1d{iyx`RKQ2
z)7klXPF|SdyZm3vF$t`H`(YT)D<hycT0-DEyrhEQB2cp!xewuRGd|3kkIOSYDKE40
zDxX%fi~N70qS?qXBo+UCK2Xn~c|_1VMX}a%uLdaEESs`|>*}m1i}|d`$5mESJ&fv)
z%cg;gs~pa)zKoh`{5xJ<R$m5djpmONQcGd5jG~)FQ7UTGHI`BNwUQ*2^XRx$f8cVD
zRzw+Ue>h+SEvZODDI|}|4N!EkJga#<8($_hr{#YvJ1fhJOd@(x`wU`z5Fujm4c7Z%
zO1Bt?wovf2Z-94|p|%dfA5mpOUMKb}U>gkG0;U2%hP?x)MDj~)93FA01Camvw=>Fr
zOnGxDsbjEe7sP`D!2a@9qz^r$cWwyMd$ED^;Rf*D3jEDkB-=q-qv#_9mIG%@Do$)8
zfhd2P_D39BvD54YgZSg@m-2m?y<*SmJY^7f+#m-w;^rS~@!jf3w%f$bjzm&3D8o&i
z;kxC!-gIttlj}b7&~@GSJ43GrMLVve?bV>A8asbZJ4GzloOa{bm~w;013e3m2X4FW
zyijk$gWuQRZZObF2e&hWt(V^_`s<umK!Se^Ru7SA+YRn!qTRCV*i5J0fTGf2kGuz<
z1l3EiAcl0!K|WUD-f2&GXsHF)_k+2MA40h12fI5-AAI9*FBioAE+WJsE{NiL+d#!W
z&_+GT^nDN&`@n-;7uS!_mwcr>F`}O*aY&vR@h$S?I)c*ib_~N2)QBE(&~fw-cRYUv
za_j=nLz6(tzSV=0Kt6>epd)Ib?~w%3_uL`v-Q983sTsBI%jsq!ovzTu#E_P4bt(a&
z=+wg1sRVW)99z<-cPh~-+MP-qtmsXr63uK;kH*Dx)Ks&{MZKkD%0G8IVsA2BA$IVb
z_Hp|;vHgch{=xSzw=9HPiJq>*?t_1JO7{|d)w3R0_ey<dO@dMV8dr9Z>Fyi0h=*FO
zd##btdI+-Vutv5vZQFa!6(Ku9-4E+5w-B+$I?JWAPZ8@{Mb}t&prSWn-LlTB(KJ8D
zXgd0s*Y#{%)H_r?^n8c*)ATz6<+-KpzqhDay5IL)#+}DIGID!$zBBmccrkx>^m~Zt
zxku-r9sahzfZ$r(cDNKa50~i9eg9qPHdhDuA)RLJ=cH2vw{;x(w3~U;@8k4c`Z&`y
zQk^f1_>Q{Q)fL~S)yCi5>NTi!hq(l5W_3?yjO_?_A1tFE4@4N*VVM<@J87OCmUW7D
zSaz_YH(^<k*9}l?QI%)&{PJmQp%}K>j()-<47W&AV-1grI!*)hy$wkG7LcqimiGY3
z7T4;!T{|w(&hO`xUUO^xH@ukNKKb4}klb};fBhVNeWH12f!co8Vf7GLvgRAY(ydns
zJ1-2rnh*ORXZ;`_-kR<I2eKdDmm#zW69O|dm%*V56PJSI9Vvg!T04^?HxAC{S9E9x
z%OyyHkE$tEX|y|A<;-_em$)|aE4dBMiT(2pk|3L8x71Q=baz3kUu+Qk2m&8K$?5k=
zPG6p0-kv_bYBM?E5uBRdew%<^Ncsgy4=qj(Kv)ga+qdbbXDOAX<An|*rPg#fy7qL&
z!v_tXT0o!!4j+G<0lL#bhstO5O2>(>{ozA=pX8)(-7`zX{DnHEH#(G7k`7BPjnNqU
zp>{yM!^`38+Ye8Yha^R?@iLyqyU%Y=|9Kh*Xss5~`RV6xPw|D(AhcH#SP`_H{`htJ
z?diu+Y73tSf=^nD*ue*qFf}AS2pwN#_>cx>NOALt4uXHp5S&PZFa!$*ex~@E25ks3
z^!d3Y*y+<OBYxu#5y99*o-zijS^}M+Ij^GtK^Wsk6qFWZ00o62VaD30?mS(uh}1!Z
zB=M)Oa{7hd{XlJX@YA2E<+s|4oc=xiGorj9*AvcBt`&~Hx*%9iN(3KR>?{IFC27L#
zBCga-q40l7UlvP-5pa}y(UL$Y)FeCzx-OPPYbcw)iA#Wv6Q6Bder0L!a*Suq*N(5_
zvm=|$JgCk%DqOqAwS|`wp4`tpGkL3sD{Ys=T8(>*cU&Y__*e`oIUAnZxbP9;A}@Y1
zxrmpv<k`zcCxE(f9<n%$+QM@So!tlG!D=>~=re!Pv*L1IOHbLcWXF``mUHY6s%&;o
zi1}^m85_Aa8P>;Ft0r;+R$0@IThVwm*;702@BaEef3gl04MHI3RjBZwk|U&!;nSKU
z+ztY0QJ@PC3IgbW&T|l0ICoas&}a4MkKd=^$v?i#dTS*>*-<jZ-8L4^tt|})#?~pI
zyu5!$^q{F1QZ3*!D9CH=iyml{Fb{|x(Q&BzqQ}i5uGCEVv??AFJ&?e%P4qx^1io3c
zd>>s8lIyWD5xCGZmE;JhDSjqe7TAatuS7S;7%Q9AxDj9Wk9-4Ba6()2-iVT`kdBm=
z!lJ5(#yIRbz5>VNV-tgsW3%OV$nQ`5ZF+xyC0@J7wSkwk<Y(-I@RC8;pgX(@F9AMl
z#v|~1$G8-JV;f!*uklK}c8_ZdFD(qXyL`#5BChnFBvyg{FYy=SGzDj0rdGQH*pNsH
z5e6rp*?UCKA#>W7h>2(V5=o7Nk>)tHMRQV%;(zdXCCy1aZPJ{qPo+6o>vgI_9Yue+
zJ|?AEi~2oqFP!WqLRUwc(NT;}Aq1pj=0RySEqPF&DW_Q+AV<bGq0!`#OLXSv12UJ`
zmvKgbTlRWn)Y81|n#I`b!kjD!wAEAmZ;9-FPH8B4Rx2qry$yIbrzA~<@iqL!NsVS`
zWdi)Pn^6vOm0V|()GFpjL+ks^kXe6enwDDL*h!L=_N1H}l{`a|6%kHnU~Ii0$v5<m
zG4h#Z|K(WM!+Jr-LXT6wR*D&@AW+R~<!&mk(NUQ8?@8rpO$pH9{kgmjl%FBo6Mt_O
zaiwM+MF{FKTwZ(1KPXxKzvS}!N4|lmLm{<(3^&y-CLk5iV?#-g46tt~XrX_ox+%Mr
zbD^vLO1O3pYXdH8gme!|eYB1F$BdV6Ql4Tqp3f~l*t9t{X;HWmtKDPT!U}}+c$XWw
zRm63kQ(JoD?Yq%<^ph7DTqIXZ8m|?lNic_gW@?FhA$yC!EGjbFW894K&n+{fb7=gE
zY>Jt-#_+9lLdSW5TH}kGL70DeU*VH2&F~Gctn6Kt+Mq7_=Ne)0Zb_S&u^wlg&F7KR
zM?RQ)sgh6K-3#|+=bo#ktH@A#^t_5r(Pz{cd(Te3_yg)v^DN{kA%3bow_JN>;x#|f
zU4Bw>6SbO~<jid?Yg6lde{J__o$D)#ztYyalsPh1h8g^+jbl73(msDzB<apP=Pk*?
z*9+6n*p?Ov$8fPM4IvQY>UH|=Ul+^LiiUwOj-t`B;GJ}+mJ-^h74LjU^R7xYTdq;+
z_PdI^)O&K_^wny17SEq%CeW>}-20Y|QcII^EbTb^eJPn()wkZ6f|cPBm+URD1-%wj
z9k0lm+ZndW*}p7o%u|2%Ds7jwCd_=3!*cFjo@Q&AH(z`iyI<z(<(qCk{H8V6R_%6A
zdUDyC>o;z_=fEB^wUOOeHIM8xtSFGHdSu({?;1;N<G>jNE?Y_aU7ZKJHaoS#aIT+z
zysK8=D8a3J9~A=KIQKl=&3008-(?vjW6iC)kkReALE~8MFp+<4so^!bYe;{S;9PK_
zGi8vV=fc|YfNA!9t#a7LGQgiztlcL=H7mNZ#dvCmX-Z6)hi8YGXZiKYwh{KeT{-Yt
z1Y^fslt_^yH+*jeMQNgwI2W^I6WqEoP)VPPV?rhbIgKhc7qu98Kc%tmLs~1Pg$y-$
z!<XHann4nLnt^|=<kANETt*oRv()YhN&Wa5<SI&B;Td{S>?BH|fqOcObs05~&ZaKD
zuz%2<phu0l>*Rc9d>aH31bA3fYIb1IOL`OteJ|^^gX<8mR)Ae1@R<>~Lsv%_!2*#x
zXH3O<RdAzJ%can)7|N_-(zknZNndt7q>NB%F5b@G_`-i**{ar;vnN0&j5cR$^x3b^
z{yH?#z^<@8@GVE}{#s&O!w`Mjw*RxrGA7hFr{oPv3|j1$l-y*046#ONWyTQVRkX2-
z5#*YAPElrKs^fnzuyK7C;l64)nLyrYDOsJH7_POSa_^~Km-x-+SnSS|m22;PD7&fl
zZLA1{=b(S&imw{0<--?hg-_&-4%inyl@DS$FB?U9EtR00W&Emc3f$omI6Qztp+m<{
zOPJiWhwQLN3G001rhsEX1nEY}=l*5HZhmOJhv7IQtb<XKG;u%b{GL<*VZA@o|BRLq
zKG5<s<R$I9g2lt{C7%O}BiE=$Xt(j+K`SLmKEi*MS}Ey;lMAhOa1{kkC`z{mNoS@`
z(Udvz(qq5xvYmL?=_s_c0?{p@yX+|-P{3ILK5pN!6~=#mIQIxE>?LnmVc&Y!lJz1G
zSsNssTdF>Dlym*w?@ohi|EO@5U~P9a3id!oF{f`_iAwya)L>zMCrlb>TI?=zIG%|T
z{S5?K%m}oX0b&;vmjM|K6aqIgmytOaCzoSa7pQ-$!`t((A3r_Gv?ZWVSbTo_c>3=X
z-;_00pwz=Ze>}wp8bP2}gF->u9sc<D@Y~bh3xo_=3LpQYzu)hHbXIufBFJBUmBYX2
z;ZHOegCG7k4*gM<k;8w6f6xMZfW43-U?c{FEx>Y6BKVNN&L9j1A$52hL?>oe3G26^
zf)szjaImA;c_6Wq)*4fZ9f_MVNU)6v1U!kICV?;{W`?9d6Vo*bgo&SidZtONbV%bM
z-S=ti`1$<B$|!J>gjc^zck@VN5(Jw^MOY?}uxTwCVMh3Il%i4QCEblrU-Mni<+)7o
z$LG`Rh>=VpyP$3*riE`u3B1EM6NC+LN}_)r1<|QQ1Xc%iM2Wbzc9=G+tZ+eR8N<Y-
zEE)3l{6#s;d?tgXZ2UIK6yR%-wfH3RJ4#BF%<=Pemb_`63A3~ryOgK+J;NAfB$PQ>
z?p=AFWyw6NQBr+{!KU0+>%1f!-!i_<E^hG}B{gYlLrJtKP<c+tO>K4E8au6gpND_i
zNTHo;R@MpHOs@r<A`T;7=6Q5gQIpfCFz?wUOlDe*R84S3O3D$pMxH<mPi8jZVv0Ov
z_ZIC7d~Icf+~s5sf*qBEKEAVWcik9<AUwFnjY$*=efDDvY*9o>bA*WitUy!0f9StI
zejnJKpOX#+dEZWx4jVm~ph!JUq5rnsTsrE1RS%$p0Np~qt6NEyLbB`4m_%0{_muyP
zTdnT={JoG?k&SfH^3~xzUM$c^Gr$Q2hR;xU@QoH_@-OD~am-Bgj16LtulvYZEhHeH
zl^jqZxgvQY{W2beMthyp&1`wTC-vv3fnI3uILdqENHnY`6?eR^5v%HuDd;A%ludwt
z;fAMQ4U7|(0wD(R0ucx2MTk_WA+L}OJPL53!6TffSvDc$^bHiVSpqXx;nTBH0~v{i
z5<bH?G05x2$t%3pK!J*eoFTAfpA-qZ4SdG7ebO%n?2{p0=nl*zg^rnfB!?r^>duDH
z6obZg6L*s0s!`d@yu0`vQ|y($k1RKTT`Wh+3XL;-#fetxnDs)W{F6L1r`5Lg!YF)F
zHJGdzbBc_tR(~mDUZdP;%<D;O)Br=N6c%d4;=;GbO=sjAi3x*78Q)#gsY$NUG;7Pp
z*Ue_9hNL9Zs|O%v_ri8!jzhFofNdx(?^`8sT3FXm(!3tyOLkIyeDRLlzrKQhFTseQ
zv2)K#YN1WD=PR6oGQz25M{A0sf(Xtxw0}E9`=Y#Ev{zaQr{MzHXZ2^x$`xA?gfrnY
zNKdh(MCc{GH-fIKNFQ)aemhAY{8+m=PkNM;O~^*<H5@h`D6KH26(V<EfwNI3g^FT-
zkx{RqF;mzWb++~iqh3S0!l-Y5?&OeBuOaRk^?LMKqu$4tY}C*5B^&jwJr|?S*1gcE
zv;FTFbw>M>M!k#n#i*aAeKzWB+2blv4F6>+p{y5l^@@kH!a4UD;*rgHmv|JDeTjHr
zOt3+9iU*QWr?lz<@jwUyRV!#G98b>f*iIfVKJ1l-bad9%JWv`sQu5G$UI{PlClpgi
zRE78c6N)Ju@=h_u>5B_0Q~j1<u~|`>3^SXtT~Sd-D!!BL+Q;l%MP;&Qd^dX)6*<=s
z^7&~_{dX%Wq!Gv$JXJwJiporAVD1%E3<@;1!Q_4kL6f)!{43Q{YK;5JDfOH+QqTgy
z&DFQpQ-L>wwo^@Eqvh>?aNym!ZWNN>`RF5or3DeBYp8c3;EzRlu+2_R1pMi6fK&N>
zGw9~Dx$Yc}?`%bOr=mC;d^Gk=1@k`fD9+%ncr=QmC};z9mEzRUb~08r`i$cAt(z66
z9(_h}81*htoXwpaQXEF*9mQd^xlnP~&bAe&9(-1Dy2PU>&Wd<{To7?nWJmt_vm<V4
zg<N!?tVP`SnlQSW711bRHD1}Qhz<o%Ms|6#BHF16aiX>CVGx~|Q7|nbd``C_+L&G2
z$1<4yjzrbfkqCc6N8)iR?s12d2v3R4gAh6u#TZIe;7@8ujKv7RVz6SL^vK?q+*_R*
z5w{xSfZ`Ab$YIHUm+e^;BX|dV*4#XoTiD_qmD>*WBs}J=JROhL3$~UST+C0FZF=+A
zQ5bKzQL;R#+UUkRDeW3luc@6c!@Z?D`MCd9=BP9$tBp*b?~BxJp-yQQ){TELAuemk
zISqfbaStfh3c6|BgV_Em)pq0Fra78i5@9W@pBoY7tmuh<Dn}zu23g6Kh*Jw1#0F!7
z#$@~APXb0l7Zx2T*#oWQ`HE{TO{u`o@!JxSWRr`Xw2qJ<HG>i-RCn&g3CaiI=A(O@
zn@)?d*Gmaoq3jfa*T5nb%{}TaiU`LNjYQXaIhF|>Ia^@>-;Y=0iH8sz2WU3WYU;s9
zjwz<FnXwdqPNgI{9%ehi&C?zx2dr9?C67`QcWHK>%>jeFZ4UV0d*#wx>pEV}%$Xfu
zEhOKfW_v&WQpt6dEQ^@cQ1M$7pG7X+pFZTWGlvKhez<+g$7YrFPn3aE&G2HnrQ-4$
z-A<TO>&7Ffr9D;4m|#(!ZkzoYds;(Fa>x@(MKNZ7Tzl287EUxXHN}y0|8>Dsbu_jg
z@YZ=%)FLdJqgQBvN`SI_^x94+LJV^s@aTG8j~kF0+D?hOaEsya_W1g#Q-ZHAqZwc6
zZAf;do^kAF31{p3>Ad|h>g5LAhI;F~2Bd7yFL@mk)l;|^9@en;IG-wj9HRrA&wOO-
zZ*V?;^`WpC$9ul(M(dNU#L6jQlzNb!xeu+O=ejINB`9KD%0jGOUn2`O>|R-57@d#>
zwtj~!)HBW<S=YzJr{!U4IC+ewF^)94C21=T$^FL602<Oi?#98>V-4Jmk7>|*-LpUD
zVqrZsO2-Wh;(l&~InRTd^v7!8DXe2ytR5bJO%-!C`bb8SvIh<cPUIs~sL6M7qH`*R
zn4Cl3R+`GJ2$W>8(MS1`oZt3tc<03hd`Z7oxsWd@q`r_Z$z*i$CAkjHmq$X{*Ta0o
zm`iHcsbBdQGYuavsEr^w4|Kf}g|RhO`Ma3Bt!D0E34gCRjn^tvBe*qtx3YpEX;O24
z!FO4P_3iJ+$8Hvzp2>tkDq+P_!+Q1S@&onT!bu!H2qRTO)x5NW?gFXmp>r@GtEnh0
zyZjss6y;RXojK5W7(^##K#p5~(px(KboIm9_v}rvDeq5Z`$+?#j9aY*@0?2=-_;Ja
zV%^7IcivU-y%=uBm%=!OpTpZgdB!4tQ>a<HXn-@*s}s7|b`;%hA~-T>4z8vGMFdt2
z9)y_#WXvrE$s~^AU0yY#?cxO<Fg?)<bS3fGbM2Ap@TT_OE+i1}b7(~l(pE25$xRoF
zfcxIrTXM_y<tmeHHW|0{z9>62t&VS*iuc|ocNOJxe14mL1{~j3r8>>ZVd^vqZis#U
zAH=sJhnE3j7ZsP`G6oZu*B1pTmr`OEssY}&{bCmr0e=!((WFRuTqz$n<FbIaQbK8U
zCtXrjScug!X-4Ik)GAx1Q-^GWc|w<Lx&wj~#>I-+tt7+mvfEIWT}V+)&5nXyiuNmE
zz8Q`Mm`MV?z7xz~gsm1!F?Q?FH38gThwjZNETAiq2$(yeE44sqst}X%lMp|$k#~&|
zh@Gj3Wq%QPLGTTt8alWUJqezoqHGTUEe9O=&OSvxQ6xhV%}SVWhT}TSHsnC#E->4u
zgu}4wx)iShUtIv75g)PFjJrO+I^6X-+;7Kc0e7E6l8^S`A5FK;*Lh*IzmE^c<rZGq
zb04nv?1f1_o@%Vi{yNz^DBhgoU$=5yZ{}wKCtsVxk^AVArF-=X=P_*6Dc^YH&_ytA
z)G~}ywwO`%zJd%?QlQ8N%d#=s=Yh$%kph{OK;I0=0?;PMA@>2=1QDdKhRd>(wPUc8
z1+I$Zu@}P{WY#b;yXG@<MQE?|w3lpU7e#-Pm56c%e&1y1+)}8^u62c%<eq0<vdQkz
zGZ5f-d!GE7t()Ljp`AIan>Z1*D`k@j)iqj%OKuHPcNiWN#FshDf$AK8$+AhNk#oDG
zm8T{do15gS;*rrxc%!BuC@nm|TWd#5(rTfcy+yRI>oNi;r1$Sp<~NgAIWsE3LAZZU
z0AbKn7+D=-4ZvrD@8sJ&|1LgG^Rwq?YTK`~%D3~igqIWWF01@j3M)ldTU7qP$S51B
zlBMX3GO2hq%w$Y;zStu>%@RP`pW2G?IF(Vgm`A4em<!X%>^4OorXNSopS-?dBb8_;
z3HfxwfFYt?QDn<T9xyEw4uiFphrfUQ@9%lYJlo1RG`CZ(Z#CZ9YkaHkWV|)H2Bm8q
zXeKwylQuZ4{`vjeFlqVmd7L-eSx|0D8DLWHFTD{foZ_T+@&t#EF#B2>VbmFX7=$?$
zZ!`NE*Ecxx@n&BeEu44nu@-M8v2teQJmF4xqedCP<*t%qS@A=o{AK3^+Dd<|j(v<#
za!mV_i)BfQ)fmPX=Rir*+MSZyIbT3jf$;Xj9VzA(nDA(2aq^=hHN59ms#(XJ2v+B5
z*{M2yXZih+$ENoy@w%C>1-!^$-3cu)LI&TVb(pC^9T_ddNZKYWaZNZv&nx$qZslqL
zD>767@3JFrC9%?U;@!c=C4+w`Xc5i3qM-3S?ocj>=<6nU-r}YO!&m`L0Pim4y;crD
z+I=GmQTDWRP5GEQO!jV%V<5xTs--YU884v51G=P)N8U`SiDa2XEtwt1Vr!Ggns!`G
zNnxgX5lXBTO04BN7PF}03ms@NtP2kBE=rYM{p_2@<tqayEN;ReZA*V#EU1LM6e&wQ
z%|-^wqa>nu^+Zb;@t)ZRBO<8jYzyFfhNbZxO`tlP=hrHx-KkR9%su?W*+LnXm1L97
zWJL$$8Uynw*Bw-s1kFe*yoYWu=!5D4-6_rnyznZ_<9sH<DhegTnG6%NffTALk>zJq
zij}noO()_nA6r0+7UX|EOInb6&5Tj?ogV{-v2w?~Y`=ZXBMwCXE>OH=Dni@8N8_^5
zI78zSh0}OUq|O^y-xsK%sFijYQ81@Q_e05x=2sg<L)p5&mbhjSPhV>1B&*Dzqni?n
zp0tsy{%->c;Pa5AaFYzbMaxgkKu!ZQEl;Xz03vDg=F_j=zW#sX%fCS2fInyihVVEi
zK!@>~f6_Nl;c?ve(h$*sc!{4EQBvJ21)5P(7RO`9mM_D`D3p9{L)-kFn$ficbS~}q
z<eWTptzT)|%2bGIZ7v`E-ZqxGjbBN;)v3@gqQYb8Cyp<p=x5@!`CYo=sQ(U|ufyhZ
z#WiaDN2buDqnv+%6$DxV4K1{#DbSIt<mD3}(c_ZOH@0lnMkUY)8;y(wfnecN{nm&&
zw>0dQCZts;l%{f@wqIa>!f6I13nxZ@5lA+pO67Q9P&2UO^Osj1*Qc9>&K8sw2E}^|
z(Fzq(J+MS4bKwAYFTIN`-S5%?xw=CEXFhrw0=1epA&P%f_&hXI%)D+#W<Re72#3H5
ze5Q`2!3Fg~<yz#mYEfaqE7YPZaZ+-swcgXY*|C&wvJa@0A7~DO`SsCOSKK#k$-k`4
zG`y+FsOvTlN+5KY-!e!XoR}vLLoQdp!aAFNZKiRY?Kp$#KhiiY8$Wr-=PQyA@cDen
zb(-ffL|K1dYXpp^HfUGQ9_cWVZEJdE-RkiyoDA{qX`bpaze$HnbsZaa>BI~bj;!8Q
zC~dPKsbd5EwOCAP+Dkf{aQbLEVskYZ!;wNuI~|=ufwHG-%g%3Q!L;+NB3*x#g&zC4
za%xoZl21A0miI4a4MEE{bkpN0LqDEq&8LZN{`P-<XTw7~8ww-TLsPxszzXEN$QXqV
zr`9vtYEToKmvd`Rll-k$w?5sH6}cR_kIL8yQ-gd-o&0Eux>ED8%LkgB#b#Hlv=-6t
zOqTIi=Rb~)<GF#7tyELfjh&hYlEF8Zqsvl&P8+PHo+>3K1hA?cjlO~$hD~0BtH@o;
zkDO=KM>$)rQBD|R)9$_<kG7Ws5;faM>fWU~7YjhBqDr|b#rLU7JxgHPvjP*NaRHz!
zetmSIvwpq!=<L>i8TwuvhHu9>e9eC$h^cU1%=EtB!);w{mHpK&X)%>}>=v*A=Krwz
z)3=uaViyzwI5C$2*#|p+TT5>wxedPiSM<??z#35$^+3D^JbrAD+b(<vkj*K`3<6{V
zWPkws^Fyg5m85!fbyXpUvfW)0MN%I>QmnGPe7nfY&v#!w?moYxzF6T1th#)By#U{k
z{1YU9Y3O1Bgh9A`e82qlQA%a_`ofo%5|J;<e>?t7PY(!=M}X&l3!WaV2L6P=ml}Sn
zcfMZI-|p!l+>dhNT>F?L(lA3k$2;irD9M-Mky;~#b$A@`%AlYA_VMRCnN0fp4s`Lt
z8)bBDk5W!p2Ny$1c(1cPnr+d%ntqK^KEE5h7~w5CYG+#k?qakNq<oxw|NEyj!_3Y=
z_M7jfs5HgW{1hvHq_ORkU-K&ZqJ?*W%`Pde1V~m{JDxJ^(*3{l6gB4vEtOERb;SMG
zteBmj;sj7P<u^%keIZanDr#r_zP=DxVZl!pf;4L$s-z@CRQFpezhp4-#W0mGaf|B>
zlEkx62}1?oWQG}JBaGJ!M5Be#*hsklSN6#)3OQT^S>H~7`Hk_t>Pd90s;73=w@+(;
z<&VN?#u8fzWQQsLTy;J=FTm(tw=27h9)-ltqV-)y&!NaT?iYep*J()tt-jH98rTn=
zcK_XmrWO|sn%o3UnJ$eCelK9MiD6u3I2At)FbkJ{8F)}L!x7|B6d1{8Ms1~pZvvBs
z8JangD9;{$KI@lSS{NO^AyYn?K|W8D#H30tt-!yvEu(Y_h1b;(U}6?py9*hJxQ_n)
z)9UUk?FBJlnkiRdrU-=wY-j!c$4?Jn49`aV1nxe5qyP%Du7dSK8=R;BpB|L8_dl&`
z;*H=t-CR~+HSU>ojfHmBkF2DdjbM=BOP)fUNU%+RAPcd`VAs0SZJpXtfk3Zr+znJD
z6lbb-H`W(IHpjrM&y~9W!#bN0$0Ua-S<PgVeDOU`akJ}bPNu-?S>{fiQpsP|SK%p^
zGXgO}9Vf1qGu@J66cnqc`~>UHGt8J%1QpCKC}%v+BAz9Mbe(cb$wmou>L$=Rb~N+X
zEuiCnOasAd^(xnr6U&ii-j=1Dfr4>zD^5Lwz~6+^P##%7htsvBk?Eflmu_SV|37Eu
z4m|y&BGbE7Mg>G+dEHu4QF&o;4MGJFwu@)lSIS31&woYIr*vPYMDpI;Rq1Z3gXH^q
z5@6#MhghR>MTzrTD<&(+5ZkJ?8B*7vZAwsol<$V@95khu!Gf3s>(TGw-qUJQ@LWC0
z&WiP2R}3%bCc7?QqRSqXHjh<qw(G?t6R{`NJF&3ByauTJKv^4|=u@6I$z7^;Vy-&U
zr{^^GCDj$##iLwPBn(vxBV0XJ^-Y4(VEqUw{an_FI@F(Xoilv1vCf$*#0a=JFaEH9
zL!FbMzOBkC_Cd18j#|{=F&r6Ywa{EFIBT;WnAd13GO}Y8Hu!V2uF^JnRi~5lFyGE=
z?vr&H`YP>bwF7BwOzQ;6Z^J%jkLP_%HDJi!;gB7E{oH+e`?&kh)Oe)@UH<<4E_|)|
zAQF-hQ@&*k|M<t{>)kKC%yfq5tiV%$1ce{`*N<<P?wemz>%z-RI5Hb*(hHa994UNW
z=4!!!DrB4;I69$i95|SjhK3b~UYdFxB*wM`CI?+t`>M`d92mkTkYyE06t;Vt-a4FP
zsF$c-Drq=MJ41Iqf19eca)R!y#7GBu_gu~aLwhJfJyRCtKt0bk&aoEhOxs$2iCEKH
zCIWe${utuUoT+~+5zR;@QJNX{LgmldzJdB?2&Jq`?&1V_%f;hNiG`jEGjJ114dkaI
z`O$cM9qNF5Ct9phGIEhzw-j|~QGin8icesMeEIaCH9d|3^8#}jV>n}n88-DD2UZT>
zm?{?zpX&TFRF5N1^F}2ph0~*d6KYA>B}rPH1BP1EVCbRB6gaM`4c?0!sPgx0=oM?`
zwDnbHaYIF<tt*v9y-Jjr`VUEB>R=@O9lJb?{gUoP=sv+?R=)_F?C{WxcG9<4nM>Bl
zB|n&wtiz(VS~R<!0T@?GX}HA`-&LgmE5*qo!fdGAp<h`(UzvqMV?SwsP)#ic_ed|X
zwfV*_Ls{*{Ft@HU_{`N;7sD7iXEx@#Bb{>^0#oET_e-Jqa>&^zlGx;1CbF@zWqvag
zD{-O?(v#MCy@a)hDH<!%sO?bUWFPrtD~#1}^qVT1mI<kx3`2#^R5kJYTJ@H24=@w0
z$(vO_ZuEMbZL7Mu(XZ5hhO;o%UZ}B9*<Qsq^{C3#<t^Nn=KC=X6rEkugY*@PWdD`(
zaZ}q~tI=;{wq(-=zDgDS%~h`=Ej*aM?scJ(SucBINxP7AQyND;=XpK$7v>WhM^vZ`
z@{bs6%{Nt?<?^<ZtmJGAfYza%-VMeX)(t1}#ai<Q^*Gxb=lT_YVLn8`6OY1)er5ft
z>k)pxj@#e&^RvT&B|2L_9n!gBuBsa2eI%aPG5Io|i_$8go5`jEUZ7NT%{R3NvL=#*
zRHg1wI1RK2Un)6qegnPWP3)yD>dPTxB%~~eZ}EY}t$WBa1$aBLpoQ3N5}mNp^b)&A
z!8b_+6@sxB_<BHppk)@OiA0*+7R9+mGe2~rc|X$@luTi(68rXWQWW|{{>-f~PVhCc
zbDlqxRf4#b?(KXz6dKBFrc7Ul#<r|CA0lMqfw_}<Jw)R1Qw10{yA}iI_MJ+I=(Ok{
zK8_|~)x;B8?pW6gZ{0)^>_-S{p-@dMq5TrI5>~@R;`=dwBopz8T5t7uiliPT2T<(z
zm<eZtxDk2-hN;GIoVORp37L%|i!e^$YX<K@F*>0+ZFlCs&9ACWEbQCCd+p6AYVD<U
z5OwxeIJ%vQJ7EU`H<9(KDK`lW@3A2C0uP3*M`8U|hkGZu((iF`AZXb~K&q+ARI0*W
z+lMVNV*PS|Wkx8_yI$h;kVQGNK^?NU18As$It7jtRXK-s66JdEx1;yQVP=3;YbF_`
z9lbXa3524FFgTu?I%5umqT#f~xbD&RcAPDGhx3KuIzSs<OhZk+u1EMx80|w+coD~e
zy5ju$9UMNsPmy%{x(WqZ+88~EN=T<qiXPNbkjIUG(SxBYpzRqphvTVKCx#9NS|zh}
z_d?k;p@AI~5SoOs8+4u_#&F8=TMe-?zAZU#$s~(~mZmSy2<b|{B=X|54#%EnJ0W3s
z)GN<+blv_;Pm}EAIgc41f0x%A+ve`IXFf`D@ZtEB={Yqg>-H^}^m4`6ip=-pnD?eB
z`dqbthu?-wUA>-Z=U>(=-7{15nrI@sG;FW?X!1@>o>jL($4ce;%Mmj#f0QRBkRNrm
zue^>r?tQhuS%GTuZUIOwjCB8(mkYp!M>pex`t8pXmLIGRt+`(Ewe3{vF3y>+6pRRU
zp@{N%N(+j;%y9vML|}ZNU-dV_Z*_Z#nTy<iPV!Q>nQ6&O+mGIn&mfSenOGG6hXf84
zClF2&)=wnktmM8?-Ot)l=Q8Fs!N&FXRPVQ0mb*PYcH}(5_yCMV6NJ%mgtd~dA*>yJ
z1YxbDGZD5)@&sYstOE!;-LitPn=C5`>#(zcFzpx*rsnCr$Qqt~bpn@wm0{S9uAex6
z<-ES*qz={z&-t&DzEBGm0JWp#`|hW48`sSc?ABdKHfV)LJKd;#;g~k76}oVKDVet&
zy~jbR?YeQ@%s!pzat9#z8!)}318ubevCfhc_$<Ss^Y($(LuRPn>q%ccCm&|s>GXbG
z?I?AlcXyRzM`;_Za<}L>ta5+4mjs7@`%V*B9qv2z66N7GbuZDL1|^i?b$z=LO1Y(U
z4wQ0RX?+LA(ag@c0Hq0qD<iaZCsTMU^csb?qmNK{w=S(ycxU<n3U4JX)fbtb&vkjG
z=1sQe_d1UIJipJeZLiaFisLRX_2;U~_dGrC4UxyQPABm^GAbnBZR0{-OHUbpyU}~|
znl-{&yLIC;G0*P&gpHUs3H-@y)Cyf<h0I3n=p$^jZQaa9o#_YIsFid^LD^e!1;_{f
z_aOs;%R@5bgMAv`;iG&O+t{MZyU|F*@~Y^Lc0sG#LeN3@-y;kC=p%>Oe3I+de_Lnz
z*6#G4`mR#FTle|sn|s?FSs1@59UWh6oOCyr-iyh1^?CoE<53tnJ|%L^zl@GvneAtH
zzu!LY{s$8yf|nsT7!;MTB@~zNFEc5ZxOf+)e~<Ue=eG}cI-6wVZU*<~x4VDuYNIS$
zg_@c_zum<Lghru<22hE)&G)~XFL%Fp5{hK4eEcQ<zD|jBR(Wty<fqTN`AZ&tkjWT)
z^Iw|!ty@Mn|1^Kd0&{}B(n&DN283IJbptB+kj3^Q7=zLj){9uEnFVG2ZK+19z_r-q
ze|Fx8+i7c=M{XzX28<$`RS@CX?MNJeO?xMCBNjJeaXpediA$0=42fHkI0W1FO3%`J
zmwT(Dv*|eFA<r`T&P7U@MdH9*X4rlmh9V1O7C+?C`1G-i8tR_wEZ?=7#kWa*>!;RT
z2gr_GwA-_-;LhwB2W5lV<xOivEDRtKe@g+j1W+BR3(}eet6b=^T5%n|Kym?JGE#D&
zt5FW+$-p&|>vc0QZ~0T}NYM?VQ=0dA`-4qb$nCZIYcgYTi2E1njAg6}XiLVbq%Mgz
z*$t+~YEnq{m}qO`44<>1BujPweE-^1BOhOTTfI|+SK3HSjT3yH9i%XMU(ePrf2-FO
zYm^CY;kqERh|LvddfVE@DKLDsHcqb%Mv-{Iwaq!}Y(O256d)|s0KFpbjzFEncg0Z`
zuFJ75OoIM26;W&00E1V-VI5n<%0k%22H43)89bg1u=NTAh#mQRwTOk9v3Rnz=g<Mk
zFp4En6(^Cr8$#lKxoAgrn9`3Me`MuYcAq7?hn&WspOaXs+$D4MIeDfeZ?|v;If`ub
za3aNuwG-h%ao|3@B0EoEc5ft=jfa-Ouvy(17?35l!$~km-m++qgkiI~b1;BHyDPv~
zN?4fEM5?jo2!3{p{ai0h-YagY@Cnq10x3`Yi%_3Rn;Gg;&E7$MmaSW;e@~6sKI&7i
zeVh7R+q|FpEVMpTp94_O)MugPE%liKeK++XBZn>rM^YcMMujt|53yHnAN5%+VxeZ_
zlsk+1NWdvMPX1i#v)#fO<j^Xw-8nU13|2afNaN0$zg^uK7=R_VPhx-sL779T&vtd^
zU;wGfyFPWU7O^m?iP+oje}4}5K__ON#(mI73QNBT_o=j*;Xc*u9o%Qxx`q4HnC;^}
z_1d?&&$Z3_xz9okGWR(E^~`-1qTX_!DbROwAJjnwXO84P$e_r;@!UranhJpSWX9DZ
z7HTG*0hZ6<K1fFK&*nbcEu29PoI-8SDf)<7@onzI>x%w%b!T9Ze`1Do5(Pv#-&o`$
zVc4wh91N(C{jLaKD`H_%6S3wtr8<ZFfDMYRJBj>&izF6*5%N=MGedr=**nP3vULmj
zsWIC}eip8s$<MXT`^itO!EN$$0P30iEL^uGKU1LZCO>ii$sNgkfF!jTb3FF}NpB8t
zPi9;#VxeZ>m9umXfA<mpRKi_;F8A4P;S6$;0=zz)HOekqNEt(iKDr+B7mji9$M{IM
zkfg^eKZ0yGVFns)CEQP<g&;2)>}lb&{r5%;E8J)cG@DJBg9fE^@8U?SMJ!ByBG%p(
zyQ+CfM0w%w&f-+_lCdt8uR%{{N&vnO_SCCHJ`tPpEA=5kf2KJveQ#3t9(OZ!->{aR
zwxw>asmhR7zW>q)@7u1Z)VV2M1*w;_mznPmL#^8$Dj;1bS#F_AdJhs_Y1zv&(}=-G
zQ}=Q8-B%r1tJ?Wmt=ct)f3;*U+nu87%5Dg9+!uQ*yOX`p_PWNf!Fk+op^MDQxq6Yr
zL8Aa3Zx;`se~oRp7*b=c8?e^_IH}hW+A+vuky!!r5R?{V+iyc^Z}YvYUBQ;$fgkr|
zMw71`D1Kk}U9)krDReD$1L<w6acn3~eElH@K;iHY6*qq=Z=|lx?)`2P;^n&0Y<2rr
zvU-TvFu-z)V9tubKE}7Am?lt4c;bRrqrwM&=uj5ZfA^ybhxHDX=S_!k$uu_|4rp>W
zBFgWxwA>rJwWZOLLb&R|wpyGWcAxHyp%j`$cg~1Cqy)42loop*z}8YhkMPt=jwxDc
zop!yII2uViAM~W&q|=9rdWG!4{8a7S53FZQHK&q-&+*tWj<|x86fTnNQgF&1#kd|L
zkZqNnf2$-`81;x+*=|$M?md0|`V^F*D~pVoUcVq!t-mZ)uhJ`@*%#^+SOq!kP2e()
z<o1h0cMwc2#=3R3q|a#Nmgi+pkXpv3Z@BLZ!y?Dan?Dv3HH;^b_N<=HEcO_k*$Lgk
zTIZ*(u*~Y~WSfkw1(Y2xUH?#OQ2zd7d+#W7e@fjMHhQP`+$9>EJ|kgY1S2FowV4Ni
zlJ~&JZ{}eLiqz0WpyV7h764&$D2@f0m%diiULSi#<ajcE0jn<i1T**a1`Fp6`>`WZ
zLixoj`rhK<X1l=#`;Hg(iC>a?UWaul`-72>H$cxH0EbI!+kM-Loas3)jwOotCNSk)
z7UIEwOfZ=p+Dh@M@><`edjAjXAlg)y0dEcy0yQv~0jn+(m$!HhDVMl_7peiCmt%n!
zTYo~=Shn5SLbCBflge2#jg28}jhRXQ^5J1sKSbffLI}A4UPb{ODqS_+IAM)mNWVnm
z`_yw%AlQWjau)pHOO5&_qX|+If4c|(x4ZCtH$Ic(+Vm*3M;C|U6umTxk{_p6D^>Ky
zkpCqQkFSkzw78)Ys9n~}kOvz-_Jh=qM1Stjuzqi>LlHe9eO0`+0-Z17qsq~&_tArr
z+3Z2&;KTHhhG}-E_8dok@1T{g<ZMVvYp(^kg)Y62$KhGmdxT~K0A+-MkULEQA_k7j
z{YGbBoJ#~)H`;;-6fxTVVKO@T9vhmB!fDdop%+OL5L~BTib56|Xj>=H9bb3MhkxXo
zY)kar$uAFbH4P?C>Q3u4P>@th>jKXz81*<_ONuhuVXu~P`Y7MFPIKNft_Q@<9Nqzi
zu}7{OASo3fW2L6mPi;uP8kkv%2})ShM@&#f`hagGUYno{Puu_T6BMi!U{$YPykIr0
zYv;HX5tF`Qym#;9oWGsIKr?Yl8-GUV)>60<lJk&zN*6sA(~61=?OGNE^bOVbYMO>V
zjw!ThnsvgU%UN2MFuA@@-Ff=GlKB}WlnPPgZ@<goPkQ(ZHP*oo|4mJQmTu(m@8O>@
z4crjyX)=&t3_>E@5-bNL0<B@A*hz%Z@zrp@h)&HkeUm*m#nMm))5=Y0?0>cC4nooD
zX^_Aw>5jsjFKy4eIAg~6FnL>STYUPB-*RK>1n=xlldUtSDJbqXcLPQWFcgxGhLM&G
zN-JnOQ<Nwb7Ze<~XLctlyS3jy1!6G7C!s>iKbo>4fi9@%$&1prtfHys^xAeJv)h6V
zWTX*F+9x4HonX*QWRmq~E`K}>|F^u}T<RpS6P?{w+(rij3We@D=wP4~LYD;zOhKo+
zEbK#PzZF~P#OUfDSxoL0(YJm^u@rdrte@>d-9bblRb9o&Ray*AugciW8Q9rB$!@PW
z05}c~P`{U#lk{Y|`<6m!{%84l$ah`+W@*>wPBR^Wm%0lVx_)5=lz)F%^a=CeU(q={
z6&48UR4%(Yf^*h^-(MdAZME>S8G)E^D(e{`qwc{JzcC)E|Ld5tZHXGh63*jOlb6G=
z=(mi~ofCZFZ<K*)Gz~AsBxl{*NL`xk;v&DhNOMZTu+Tv(LE`HICZ^%3zcW{paU9Uq
z7;E{p>f5KXE#GxKR)6W=453ghms5MiMS>}JR+2Q9R%1H#ffJt>gHQ|x0%J_klrpB`
zYH|11?^k_UIU@p<0WC76wB%H>Gv$obb+Vi}L&HT5E+*Gh^In-Go(li47_w@3EaxhI
z;<d<q$HBZ*D!w())&G2@mz?w*Lr7_v?tnf+nKBBhD!#mdZGY9b?u3eBx}CUyJcXhS
z-%g2(pZI~1`$wg7kX}w14%Xwe!5m)g=)Ay~@uYgoQ!GjFrQKofq--Mo8XNbzleY7G
zG*|Z8oXP;y^DIA}m)TwJ=Cpu-t>UFHx^ojmQ?LV6i|SemNB!R11G$-ljii7Mfkp*q
zY|?Dft<Si-2!C9i@{XpbYpgXBxJ~Ov_$g-18;MtWATNS4i~gTLAcGSMOs8H-fj}D;
zC~l5dWQ*ppgBdvOy->j{>7X>rSKn|>Xzh*|6O_Vg(@MNHCgYw$+x-(LL*6p!RLjRb
zBwXlqHPG}%nUUa=mlH?a)v>Tv=xSG|PFhoKxnoRXoPWVq@-^f0te&@)=jA3}_Ro~y
z@!FPFs*kDeT{4=P&KyCwEP==Y%ou?an_n3G-E5|!%FKG-##<j~{^340Q!C1hZgUs=
z*i5YytpfIu0`zVXotgm%5AHc^X55_uY5F6<wTyDHxUG_!WzQF@_b=Sn%$W;;5vtx@
z&eJD%?|&sp_oKKmIx<ckfAr`C;V9qH5-E_BtuOhW@q_pA=t!ybTqg><wY`mkqB-qD
zcn%6G&|2c%6%^icqTYwVeoZ$Jh}q88R>p1weB8tFTL{qnExbu58oRaLK*KtUbLvTG
zIEr01TtVYpQ^1`l?ACSz1w&RGzk$LiP}wq_Sbw_(!Vb6Ct@s8YXz3fD1W2T^roEE}
zk!F{m4xTB+Ja;0q+k!1*w2;m`l78GRqA%l8s3=Xg|0n)i0mb@>SW>*wS4aAo<T<@+
z12wxRtFvcrPBdwwzgEonrJ-p`E47lZ@ZFjT9i{b@i;?27P*Xgkw1%}DC1yr(Y*RG5
z%zvq@l9j#BwBc5fmNF_GGG&_SU@8gaxYyi_Q%fanWo_wb#?>6K9Kq<Mi40IxOKDQ0
z)Ev-UAhOJb<4H(oT=KhbbAo@3sVl7B#QrgRS&Guj$*BhjD{z`BibVkhRs?MyCal0q
zA*Tbm2OL~#ByJpBYFa<$GgA>P&vP2`)_(+jxWu(6w8A4RTrg7D<Lg_9$8<qGvcTLe
zqEoYYs@CY|)VDyoJDiT0#AEw%B1Mvu$a>SD9C%*cfu!pj^HRM3ioC-ZrA79ow;tVZ
z^#+>C2n<g;o&w4X=PMq5UlqDv#|_)#K~Zg<WINFORC;FnowetF{Wk2DS{V1Fqhuu@
zEyX2VxBLFmV!y7pF^Gpv!9NFrU__WWvb|O2ZVhi69*?4$@EpTEiZIbnM^u)(wY?33
zlj2G4PZ?P7eg?rorQ}YXZv$bSaPZKX+<PVTWi{~x^X=53w}@=N{|DB$I=q(wZw?ay
zHJ2fFArt~LIg{b2CYSJ%1O%7ni5H%KfK~?c^2ht<Cmva9>DLAsQCZTj2)iO0l>eJB
zAkn2vqF)h5NF1WgRPjjBCBdNSv0%hNq!4_>>c7W$zogajmoEO2<rS!q{r4UpUO=7`
zZo2q2f2-nZ6|G9-ClO&H1z~kaT-1DdcyW$Y$CW&MpgdC>_=>zJ6)~Pmn$)R(#OV1C
zeE4gI%ATrtH^Zu43*oodfYIyp<JH2HZ7L27G_<;f*OsS<W!l1(KxUjCW9jXQ(JL}3
zfBV!ty}dvEJ2o^_4C(&ok0<{EB`|442bwvlyPtn|-=2PJv6gNaSmjzPRk<N>q8vM|
zfEg5t$igIhVv)e}KR>^B%`abn3dazVX|OFua@k2Y62u+Jt%2UAL?SY&!wv6)%V9PP
z9|gQ^d7IK)c;?=AGbMy#MvTOiblET^gk#Q(L@@FPB8p$8tRdoCi=CR3FYJWQl4GbB
z;TaB#%KjOi2O`RlqV%*Qn@x@!p7iiE3O?8?p<i{){e5^xfzN%AMmryWK4;g6cB3wt
zD_N{%B3AVR`ne2Z4IeR>2CKWqz<TOeIO@jA8fLh8kQ*x}nX{PPXjXA_7ptP%?fO<S
zFs`LS@M+g~wDVirLf;U21V@GnT8uNHM{rA;(0gSTjKPPRQ#9K2?$pBep+^t)t>P9O
zo^AYmQxuCKY$;d(-J!UDPg);pWF)1L^>GrVkjeZKiiDbME39Yf?+J_6z)(7m^XnhH
z`-@(FCSd7o_g?`0k=NnfKi%)%zE==ivT#DIEEA;81;o4)q&pYU4k(m#jNErWij9<^
zVZw=JwB(XbwRjhO7>HI$G;^kMR0lFEi0NN2y-_;ox3!Q8yCKPcE3HPo^=Y(<yFkYB
z)~Cy92MCF^NgHKIo+sY=jL}~?{ta@fEG}!{u&7p4AT^Ux-dhEVDGJV>cr9{K;{6k_
zluUTKY?Q!ZN>?O6If*ptK@t#_X)P)Vs7eBAhXf8&x*`Ffcn|tP5&*wZ%LNJSv4O*s
zu1J7O+VRYTB)~O)rLTA=FE~u;vINixhJ@rH5<o|3zt#7Vz;Q}fBw&H5<^A^t30P!S
znES+m<CLyQz=R$1F#<4>vK)J#FmRaA75=qh%H5PG?MAUJH}q)b^6o#E8@5U4hF&5{
z#VAq9mBeF9#q)I8YhvLTzB=WVi}E`pW-{R|8XUGN!+ouP;o7cNK~Gi6hZOB~ro}>4
z8CevPl+dN5=FHVJv=q#ccx&j;gAJvLm)JS;8Isbz(>jruF7?>JI;uo;yG|&DE~S6r
zYxb`>{R_c+Kd5o#CQ_2GE;^@`teb|#Qn`jb$3s6db$fl1p4w7_rt=Dys_$7^!f8=X
zQac7PW^GY_Akz4hbEL#vnwm_O9y%PU(X3#RAQZ@)b0gff@y)twG)8IZW!_oWjIoAp
z96RuljPR0FnB6G#n1<$h*`gn{vUB^rtd=H?r)r(o>ODpJE~Ueywfmq^g{4y)r2N-`
z##Hs0vO^UNdZ{DU_7Q9urc!N)xwlp{J=9_dk($bXX}dLq)X=-MmIoj;<;mU5u7Q9W
zZ2_VPy#1GpTqD&&RK9HFx`KYZtwx_7z6n@DC$$>f3|mRj$&5PJ(kC*!+<J-uQtBSd
z7QXJ>mW{A%x0o=Qxqxi*Af?^Tr2qzKN0Pimn=nuhylNSm1W|UGQI`!q58?ac;<cbS
zaI(CA-Ehj8jBe@Ur{%AFdq<AA8aSEmm5=3I^>9q$VJrv#hL_s!;V`WyXVw!~q&um>
z8MkC<G5b)aJ4LS^zb<A2y@pB~c4gXA(H>_N0g`9jrS~jIDQj|mQZZ>^q)BkO%TXm&
zvYDUgEhAVnjy^dloUpP8vriMhZdM_rrt)`x)GgLL1iHmp1tny@Kd<~hJhtttU%qUb
zX9%mAb;ET1Qia}Kvk7oaYB|El7srwfw0z5{*A@mpULBw04mr~l3L5N%3)iXCjEt0F
zT8%iG$e=}0q253>C1EH>I*HrS7K(`uayK)=aMP5^&rxgVPPStdb20~7@w$JWm29Se
z!L(vCHFU6<*ZbK_js2+2)L4Seh?%PEE;iFpeZXcK3=gsyO~s8FItcE#&D5CIY^H`D
z>d2^F9d&daHldJa#qBY?o0LdDgWM<-e#uH72}U8~!$t|m<x(M96HBvdaT-btWlouZ
zhGhR*lxXKz)QXsP_8ki^l{rpPk`)erjgRwjRvSz!yw=dmIYFJ7#r$~}jm9!3RJ-cj
zDV;djZ=mNgR({?bIEL(4_ynRA*=2fi4~7XC5kH*p@u}7j%lu$;c;~ICdAfP54piy-
zil9l9P%$msv8lQsb5zR{qhnKbK?_J>Z)q0qMzM`DaGLzaJ!Yy7&e6^{oWdi2b9H{C
zt+JKuayaiQ^WVssgH8T)sQ)IM67}PEHzex`e(M-My;Y~;nA|zpHRx>N&2fO#D(T-=
zN5E9shu%KoKd!eGd$5Wjz+>h*U@g;bIA$Sxje>mn0G3n5WzTve`-jO~Vc+=EOyWW8
z8^errlk5wQ^-tTxaG2aRF#tn<EpLfeyHRWl(H<=&Zog=_1=3Q{l+%OQ)Iq?q#bTR{
zm|_Tq6Yls{v`<e(EAHfRsIqBYO_$ONWg{lSk7gZxJ0enqQ&@7#6P8<Lly3OOan15t
zf6cogS#<r8Z!$#=%GT2*aj|8tHpnXe&x3$Zo?eu#WmJ5d4=Z2OYM5kyjKjgLD9P9`
ziWDVX^4Nq*EqYtR54ym~hJS|5cG7$~)eZyUPob*W^UBq+Uo$$uTeP*$Rg=n091NC`
zwsLQSV{Iy5V~)=>`n1}qn@`jNN2<Hs9S_Dg7*`I)H_+`d(QqR5(23Ph1GR@%H=jhE
zORmK*lgZ^5#LDuzTV&En3qg(~hV{<WNKPVU&TjP8GLcJ_b*{o=JC!e&%8t}*?@&J0
zc$SOD3icP3j@6ngg<}PK|FW?b)pF5T!9Jv9Z0{7b{s)7I8JD5V6cm?nWIGj?@(nU5
zmyDAas(<U{xA#v6$s#GbtLpyo{qXmp70MbA#@p`qZ->|bM$x100EN||`~GM5<?wT%
zkSR;y<1hOAc8R33!fO{ve*R5%e^BEm>TH7V{z+ZGm1$)6SNAJTuteAknFt$NfG|f`
zc3K1<nCw6TO%M{ob`mQ&vszd`PgO|~uqB(!&VMU1J83QA$n416G#a9fA`o#lySUgP
z8Wx#~mC3Nk5Rq&G$TW~zk)FZ#rjnkK+s=KPQrE{s(@O6gkBehYy|KqJ@s1zqP8<Lk
zG3udzGIzSm*UVWoaa7}0D|0IznLfRRQ+%tVd^Z1zJS9Ff^FsWA%dPP#)6SfxcTvI*
z{eMQN)@)?_#Txc5+k!Qq5>Bgon?n%JY9fw;ETEVJEsiqT*TmoC7pVPN`jcI$h288f
zr~zc+_|9rTfzlCi5E6?PWHcPQ&a4sEgwoPv2Wj};H?7mmZV47NW1Mi#+~ml%l33}x
zi7YzU<Gb4+u_B^c2?9cKZN0n55iq)^?0>p|T>+f?SPAT|qO0NF(V6%hd6QcTFUmgi
zI6tC%WT%zfM|qoEO{S1$jE+Y#Cq7dub0oJldJ~x2*S`ORpQ|JLCw6(3of%}FQx9l)
zZRXa@r(>roGcTnFOO&}QJ$0`MLMm}nv}=Uk3bbXX>G0x``#leV1KH^t9qN_Pi+_65
z6K(WwOci`;IePlbay$V-42+H}APtGjs-xl{FS4#xl9)Gg!h9r`q7FwM770pw;I>hu
zDy@0ZuZbCk_*)D+ii*8(y+M>&^X9llt)E+EDbr}oa+m(uj2bEt9hA4nGFm5|{AI&%
zAZ$Uy1x(lhOu&`Bb-{Y;H4SC@w10m~Fb_64F1o!nPZyCb5{7juw8&Oj5sb~qwxPNv
z{*dcL=#YYD2xVK~iJ1iIR{nhd+BI)}c{|hr-B{C7(y{^Z+*^}W<VBm_+1;c|w>EXj
zp2*ryydCvmDYU)fQp@utU2i#00D*ty!adIusJxgD!uyUBNKgWdz9kB8CV#P#vlthb
zg8Sa{1feGaHL5;OMEKNO0b7Zz@pU<8T=aJRdB7~#oyqt;#1p`JguY)B^BRU=1S)t8
zHibY3V5tY+K=PB0RMd%FC-;8-7UcF?AaVT|xjhOGW+FFkkb5_O3v#<aD8}8g*Ef?`
z=iMS%W8Duz+yGYSFyydK7k@)NFZ7g(_9UNfsa!Ddt9&4d=3Pr6fg3x6ZfD)g1XmQS
zJVKxhjs{s7cbEY?5@Jc22|Ir%D9}^Z5Iu@7Uq^slWe_xgoDpEpa8<+HVq8|G30e@T
zR+aL&Fe*A!J6S?|)6wD@U|)Ce*a`4JIROuXcc6{1q^8Qk6TG^vihs@U4^7(S%}m$e
zpC#BAU2MbG^u)QD4{s7+Ybf(+aKeWtotue%5~Muc13wywcXdzvNZ3ResqPOT$&m@A
zZc5{9Cb5#Uh{6#b10W4WEC79906CXX-U1l+OTM5{i$Vqa7~o{A7vKwUN~y}|7>}IM
zG@RY^oZUR~@4$EDc9*W07bAc2_cH#S+E(gow_Hp5qRj>VCSS6Z#7g%`WFz?>15gQs
zh~UarFewAa!0j}o8g;-bTSq|VLbi_IR(OWr%;1^6N6HfV`mJ2VF=cF(1WUc^bgr9l
z*L(xJ%Co9K>TAvhGBxNkeLa_8`%Gf)UB05_RC@JIE~37Elsr|lo^yYTdTyT*zf+u=
z=RWuD{?f>-=>oTrXB-mom$HmTT0xt*0G@)!TPWYcd{|1BU&@vC{g;yUcD;71vp{(o
zf2v5MT$PEtz<1PK%@1i{?wDpk(MvPY9or6?`4a-tC~9g(%kzGut{tHqZ6AcyH6d82
z(k<tQVfALF1+3omZk&J0W*!`!y*0K}uj2F`)i=G<Hb>UYk<-C9V|Gn<^zNRRU85B8
z5j7;5&b|J&8j{T<R&r)g_y^UHXkuNBYe@b(kbSq{3;NVR8|Fdf2U>fft11jx*Bw*{
z)Kw_+*dWt0UzYo^t5-7Rpwwxl_I8W8pgv{D6_^L9k5HZ-S8soqN_(EQ<@hqJwXChW
z&#}VrVg0#hzj#Y(qcIL{@-JIStaPbF);at^I9^f6w;|Lxz7IN66y$P2$GG5OM*)W3
zP)6bJn(JhL+MFAz5qgXb_aZ4}=-!|4>QEa$4hOs$NvT*{=F7P{)tScn)AM$;67z$~
zw-m3T75=#0^iF?ES!FmQs=Hn(!|ku!|5z11s^n4OIPNY_(<$=m)v<1<XN=LXf6}Xx
z!m0OG)fr=jF;msSxOS~kp#h5Ua-zb4_hndIpy0$T00_Ph6$<-n9a>8BwGK`1ZFR1@
z0d_O%0>Ey1;}srd)|reGWBenTpV6C&AFXuUz^%&87ukPF0aZ<yAB<Ihm@#BEdAj*b
zV7)N%u4N$68ux{-`G_6qqrIsNWHX6%oCPbad{7lg2oy2?uPMKOP5JIco3{l|I|tw~
zCFPzRpBXX=o)LBqIOmMSH&>SLW_n&1vGCA8sD{Oj($A%?9I6WY{NFaGp}rid<DBBd
z%4ztAoFyMU?G|W3Pl_P8y~VL@rLfME6IBRzsd#Dc{{o@Q%x0GXZw?ayHJ4%NJQSA@
zoEOc1feYiEe*OC4h36?z{*6@(xg7aN-Zi3YL|6E>_W6!z{EFx@@S$;53MmzGtHYR5
z=655lNmqEPkTVT|LcJCG_)XxcCtJZilU7pR`1|r(s0l3veM{TIpANjZ%te@i8=vUR
z?R%P6dM9$M)0~vabIJ3Y-c4v5XDC54tv6185N&{2?iVH)1`Z7HbVX&@YI-J_p0d&O
z-cLFhWHl92D@?wDJO@)akN7jS7ZK@8{f4^cE6_m@Ya<?*;kog77+vN|pG4p@nxm{{
zp+@xuY0V7U2*#Zlr6Z6h<N<;_?1T_h9g!jbe=Uowm_{61pVD_cNf$oSVq)He^%QP@
z-;@77zh3^DF{P|$)|lb9Z!h72lE5VL2$nNxVEFq#!yhj{QrNs=Ipg>a)8P+(`uop;
zeDhOwRGy>NaAVN0YDWbeO-T(#dZx^1HOgmWhNO3$Ws_vmxQy?v4sIeV&p_Qoj#4gG
zAQEc)-lzYUHG#rDCwIaxt&RnX6BmVl5e{eXMjh*7>sdQ0NT&RW0&Bkcsd9r#G3P0u
z;J1CImP~p#qdspR)&TR|QY8L)dt5XV1?Grx(RCOLvvJ>Vk6Xlzjz=qBi+hX_BFBiV
zICKl!WnmbY*(g07lo_<t(j+Glor5kuIz`6#zs}DtyLc2rGb<Dw72qb1LU`tXty_{?
zE3;(qlNt97pBttpK&VzuxPMH+I)1-H8O79^fx5bI%nMD3pRS55+-A_8CZu9QF3D{!
zv#v7MIO47s&I;6C*Xd&P!Xn0*bsjf*D+*$=gS|e^X7)>Tgv%@_tWjBa8hH(qQ9x}T
zfm$L({J3qmPBO4FIhLHYn8eI~46eKV2IWgH`VB&Tg!yH|aY686uk{t<ASeg0m%Yvm
zd3`QYRN5ju+ZSc`LPqU6Bgd4R$**{!kr&8?$ZZpRH5{5ne~f_?-CD^ks9FQcuhT$T
z)C7S#rwxnRV`=RiMMSV(KoVRrrDr@WCoE_spLwRd4Z6jtU`1iWWY<Q2-5KI8n`%b|
zF`M-!RArdQvVx*&f_Q)hX<!ydD*7)qhTS|8bzTY!nW^kO0=NWa^t;ZwXCMQb5cqka
zl8Z0B-oa_hjFx@G$6Y`!6gAeYclFn0GlM8+7_N>d)`Yi<WxzXB_uJJtGAV?m)wsf_
zD|{)&3&Ny+7l(8B`39ta7R;Ck`OB|-_#GeqfW)C%{3nur8;9Y;e}|t#wSNcLp>z=#
zi)@8~Z2_1Mf_YpZ33h_eSeK)8w?nMt48xfgundcm_L}0xZbi?2fC)5nrGUAqKn=}I
z@1wuU@o!sX4>7s_PH3H$u2ShT>2qr4hIhhUqemA$#|$yi$98Ui6PZQsdeu)2FHd6(
zDh<eU_*R7Z+44!}UQEi%a`SO~T2^Dz)pMjbH7TvJ{?QzE>+Oy`pwGkba<n~!#u=_n
ziNjWfcW7+OP%<RV!X?Q$$PpuDXfB3^ADB$yGg;}0pI3@%_mu943Jfz`9xW=3Zr&oe
z58<*%YFP}gc#pe(1u6^iWSvuxX2G_t%j&Xi+qP}nw$=5QZT)52Ho9!1%SM-N^zF0H
zj{9;RS7b&;<a$^!=U8KYBUxtl&6SR#)lc$;oFa2T{u<?_9mBMAOiCNlH({*2&|-u$
z#T|yBoG@nk=UKA6_M$_A(iv2IUn1eR^wkb!y_#)8C7;mKz{_(#fu^A9Vjvy;YWMr!
z#jo@X!7$O!*K00`*(b#<mc*3sA|k96M@gI{B^<@+;4Fm}09wz=hjl0Mk&X(ar<Bnk
zTr6lXD-zLmUX=5%4v6>goV&#<16g{Jngy=zLO$FyiBE>}8gbS>Y<Nvql*Ttp8lqVo
zE&>ueQ6tUtk;+yPWz(Q>rzsg*`JeIzj-w@F*V|jffnJ<kFdkc)h^?m<US|2ko?T?c
z1>>(Ctp+zVfUe{*2d;ovhV<+-e^}%JvTfw4>Fc?Ow&+sfE1D1O>Wsi)+{xeTw&P9-
z#=Kqt>Lop1Z%nh$EE4@{1rIq|DU~<93fI$|k7`;pEz<YJ*Vp7J5UtjakATa-zUpE9
z0iGA*-L#^N5Z`(oH;A7RCosXb7hkPeS=Hnc14U_-fO%X5otAKZH2g6N`m(&Jnpu}n
zj_NYW3-l!GW{!rn0W89sB_H^tggAeBDXHg6`O|BwA3F~Iu26ZdO#u#%9g4vd=}=BT
zVM8xL_96?CW|D|KLnLh%7Vq+r*?+Nu`cfE)f~aL!BcD@yXV)Xhk?NjiKrg5JBsR+6
zAkI{z0h&`5jad#Xo5^IJ?(~a_lLBfT#v1}t3ztjJ+?Gcj?{C4Hv2sq(bjL-_a4wE_
z0_;Up;)9l@=#|50*hx`_cfe*2NQQ{|1*f3)WZPWFNtat4=y^uQ>+9S8aMsK`J5XF>
z6i*G@E;6)-wooTl#;&<&Q`5UaqgQGPnJcsp1K44y^_-qe`lyR9R0Y#dwwM^gnE&LI
zcWCKbE1#bF&s=g>NJ#!6jFjPkFAOkm#b|a!xyj_f`3t+xOk}S;rp2=+{C56y_!PIJ
zu~J=+=CLD05xI{m_vxuuzw_py@!ROqjaOp#O4=)44&_{`ruZn5!E}zd$I@_O{Z*+U
z9?%of8(;6r7%Y2N*O#8-v<tl~F~t4bUg00tQXC0IH|OGnR!j+kzcl`Vo9V+Xp7rT_
z9A<}fCu)!3j=EzCy<XBtX;%3OYsZFvw>CR(_|!8-Em#M#na3k5^Nk&%Fz+ic$g369
zkd=qyXLrB<xXMj>ad)cYa6FV|)w3>T7ogPS+_QBUDMP8!LHAePsgjFnc25yTni<HW
z_VQzOa>uNM9`70L0<9GNHQoU>sPU#xccu<EidtDw^3*q49M4-(WXuE>iDG#oA(M+M
zSx$k|mFe+dUcqe-mWGd`i?oU6Y_Lk*@0V8D2_+Ol-OKGZA#Ht4(;DKDv1|WaQUF7?
zyd1v5$su=&&B9BTd<?3BBs=AUOU(pwomU3Dh<EY?HajR=8&t`z6Dz|aL5fY3AJw0+
zWJbR0rhNST8HnVZH>gbd5?K=`Wj7DwDsg0>8ilBQ@vJ}4O94c_a*-aY4#8f*cja-$
z>ZbFG#yPQDpQdEdmV8+Os^_Pe9w2;M&cTvD^+Cjy1+TnE{wXLow9_oRw}n_smU7vO
z>sXYF5>s9X|627uDL)beD4dvXAS(BHWcE^;NKyQ(I-ktYzT7oLCge2xE$bH&ecAMy
zO)e^cllyA@+?uu0b);T>tafmKdw%FyGi1Qmz2EYX^?=iyQ$%q5yg#ELB!o*6xll~#
zXN-gRj-w<v1F_$d>x`7Q@EomUZM$2QzOflT2iiPx{TA;YT3-02HLJAEqpo<9Z^ZN$
zyg))O=f9p4R$$P%02<IlRFWDfY%K@@+%Cha1PuU2lw(~3Y8mT*anc6-C1}b?^2J>&
z>h#*Y*i<B)B&~=;;FIW{6@eXe8itd{T%;a%k&46?|I`M349Z!KMi$mVdcH1l)>*OJ
zaDNL5SU3jfVSlm06D&zZ5+gQ@b$n9hiDU&>jL#A??}=8`Z0(_Ps!{@=C(ZCHzM<-C
z6o*DLqTM4vfAUp5Exm3%>dKIkP^AvIA%4-l;QPqAyL%#qbrP(w6mqb1`P<afJc6GZ
zsj0Y$6kD(PaQZx513H));WF)ZXBn24$J{`K7QF3P$A-#m8VMnb+Y9AK)l|~F-^c62
zZxhecZY76TUe<_}EfMR?%D#%$US|*+O}@?$xxs%JI-y-?0&(Z>UlF*yGfhk-kVmUQ
zOdlEa9uVso>a5aMeipsIS}}0SA?<oYW8bt_Lubt<3Co@|Z${M_e*|<>eCHKEXzL;`
zY^$EK_>ligdCVQVo%~~W@Y0FN`Hs=rq~S#NfZaKu`f6pdb4<n`l*6el{qEpSXh(Mm
zO!ol-68=D_%Kosw*pWP8k>$EJ?oXWU_LrtANCwr3u8&4L&0538G108g?k$`pyKGPU
zk?n36g+im~pJQFi!Eh2Y@6C81_9R+d6-zChg<$YgyRAaYe7aOOZXEXN3(tnblx$5E
z+($@O=U#*P^e_`fCM>!G!j4zZmO93!)pzFsfKnNROeMYUa}{t-sb<*>FN0QCjZZY@
z(!0lus7HO@j*joFPEF*<#4T-dQrbnCoa10xY^Bnc6o5A&0;4wi7)@pwDxUdMC%;MW
zu8t;2*%o85ByzA_gA|D5PoRvZEe*K%tdb#48SKCsjeA>34L$3c@|;)(CQmgR$kcEE
zsJcu3a_)P)btvu~%TUia6-&i>!RU7FKk=tHw(Y67xItDcpRX^DV$css)-Vz-(bS|W
zj~b#$&T=*~xZ)-(V84VxL=|ecStwIPJ{RJzB<62b-WQA~u5V)pcs^Z!8pFZ<b(yxk
z6Rv3Cq@c%eSwIYx{S#epA-!N0;IQ-#ut_iWuME19xo~B2T=Xc&ESV4>1hUV(7^F+<
z@!LBb7}QV8@r>JYWOHrh<9GgR?IN@l^NEBy=z1fr=bcQ2&ayb^R9xHcGT38WNC<BS
z8(<Du+(wF{=V^8H-nsctQgY8XJo<%N-hQ|)OW(X)Lc{pimKq$<xWapE7Ir}birjl1
zkyA{zy@0)*q>}2BE#9_W85~k78dOYRKnC)L4tQWWCWr#z@Vn^}jAa{QVUiCRL$^z^
zU0vPd^j=#om_k{}(g)5COF0+0D^eesYmLE<I2=E-pd@bFPEiY}-wC-NR{kTELrJKV
zVs32xgk$@qN{-+=bKWIE&KL<`lGI(%UQ^^}Tzvyp;#gUqW=sAL%3RvX@y?N66@#oT
zX!6bcp*hPaI3?9OfxW0ZY&Yfy7A0riQI=%-9JQaD_|dFNSS<0PYc~Cncc5O4BaN%u
z!M!b0P77{)zrket72TpUY<~~6V>I{L>3u0l3O@IE{gwpbkPPIXsysaf#ILY6?_>KO
zg|WG(v>N@c&oAz9#zii(){OVE>&MeOA1hl5AKk{(x;*awdR-jw%W`7k$b&N6y^^zE
zJ}8POYO)`s^6I0br5bNV1FdZo(z;qZ)`p?aB}^L}9dzMnFB3A4Ijbi;))<3oS9*uv
zue~G(onrKGliQ}+rW6+e46%jWDbbxF(K23;ELF$OyFT8fgrzq2?(Nm+$a+q?@Osoz
zyC^ld+)cAjrL$LYU+<3XZ7)kBf>`MIR}wQPfz{>%1t05~6t&gbyp7V=K6+P_#Rks3
z`PoSe8o6%G2T-WctvsF`0Y+IF1JQf2r#5I*>=!Kw%@Tk5Kb&j;2V160pyI728&m(-
z;sGn|!oiVcwcq_4C<VialSJ+%wo*M?lcY;CTOYnFT=GGIpN7v4I9$aKs}3mbWWB_U
zzdCK0E|l?#QP3<EzCHdH<ocDb@)d*hTRH>&J=M&FRrd$<8|kSV6Bi9FEBf55i+fr}
z#7a>KF?xoIQ?vr+s|ubrd0BGD$}oo^U~^pA0(~;S3o@FQ3<JbLxMijjdCzroGyGrC
zj}LMBEHv}ZbPr1N5&qOW^ys@g<N|*(;rXm{@gx9$PFWZRw8mxp<HfS%<FO!pgUAlX
zPB7Z!g@lBJ&G&E)!I3?T`{bTsdd8E}$GnE4b9<B>6LZk#9c$sNf4j0-*3dpI5AsA@
z8J*}efPSE*{cP9SRQRf8ZIEqI6@C9gNG_WLHin3zL9lVKu>oo8u`WS-+p!w3h@qg*
z(=h^lNUYig+proy0VDH@y|nk&rL~c<pk=PN<Vj*uB^X0ESr(dj7CXId7@28ifi!z=
z=1{Q^q4*O$9{FDN4ic6&Ebr_}$vo}>w${NSEb7pj|Ff4PlclraIq@_gb$GNHZbG*E
zqNXUqbSv+8)ANiO{>Z^q;RCL~=^@BHzgCV8eXT30l1K}%+rq^)cQzaFEV)Z`bro0`
zWpl3CV~eWa_W~R>+oL}9qft|fo?*a|oo`G0mO*dO|4Vh;+N{5Vm7WDz@K?{xW%oDO
zq!AU;Ql8v;s>!9j2Z>Ac3>BR9gG~R)w<(`0e~SV`*vJ+Ip6+67WtJ@}!nXBwt3!F^
z=ysM;DeM`5bA6SRg>6s^ZQvq?8bJgH)9@ru1i6Tvf1m2i`~-PS7c7+CX`bVsDk!^N
zb-yJmdsY!w!txYwL{;KLCX{QxWBrPhin&P}I-Db~x^3SYY;E)#{Rkn`JdI(RGt4_~
z*13mDqHbfdE>0f@K_{d(72l?FPY4HstA@EO?rj0^=Pbd<k^{BLw;0_(=hHSv{uz0e
zr&C7^$gmqe@6D~vQ2wYwo9VYHchHU93WN!RW`@$0`j8XtoKV+!@_BL+_38Xpfcoon
zzH5#`_vF-d^!4OK!!qN1{<q+p03F4a;Z*$>Y%A6BFW3mQs=d?$UxJ+J`TFh|r?e-t
z)(dq2XR$g~RbS`5eEZdMmy;t-LHH<AJ3M{%pt6_bl2)+-9~e&jMDc=^NLlGK{CErg
zY#@AL2Po@RtO3OlM5e}SH?dfSV<7x`Te)yJ_6BUC4tp4id1UMd2^oK%mXRzs6prW>
zY$6H1Vjz5J2O<uu*zU8aW$qK<xN_A+V&Nv>-=WH4e2AGAY6Xdftb~EFPtyZBwXmTZ
zY{W~h+~BOC2T<cJY>%6c<26k?kT{|q;O66P0+6i3^NNAfMKrO*V&?oHIH}_rf9>{i
z{Cq?`z6`$?bf*dD@^GUk77;nn%P~!4UCIi%{deVK)tjN6_5oAUJgWB}L6+P0nwtRf
ziJ8IFzr7C3+?MIq(uVKNGfkykg$CwQQjXuY5c?D-J<hg_B*iZoUXJ|h<6J;1>Rp0W
zZG4V}tB-vrDrFhV8kQ}(t|TZrJO5hVX>a55p#;;YnG|i-C;uw+wc_66l$ZyHH@B{g
zwZpbjW^@Cu#|R5@RE3$;BULV(uvLIq-pf8kc)g=tJ*T7Ps}R<{esQ+Xd}ipVpkEu4
zO=MknvH(W?0lu#0O+-KH3ab9Mq&lCc5(;U11ZvsmzaN8bGAcPTbfbth$4=b2?KnJy
zHl+<QwPoVb1Up^gB;^j0F18FJ{rG*WAs4g5?PlCsl;0O+DM34`KN%PG44VNWV&N`)
z@4^W*X?I`VJCfv?B341HN#@Y}a@Y+O0|HljcoY-Xl);z6)=_M;XeH#en-DaxHf+2h
z>PK>aGwFDPvJ-xJn|?GP|LG40__>>YFc<bL3d&24>HU5|w@bzhB0=os$lpnRjX@ZD
z{QA0>0)PtA6%WB#IGVe;yIKHYoe+OuaImxf-#l~R01^{5u(k)Q6aYO6o1wE+-~adZ
zIE+jNOi<$C?e6fnpDC4^j(lm====2z)ybM7<Nn^1fNjIz`~COv9I#&t29niEFd(FV
zw_KQas>dDWpZWM<N@)LtiCG&G@MeVj-mEyp^tba9oKhR?5=SWz1BRF%e3~g4k_h<{
zWU%5_2jZqk&G58rIv|r9Ny8Zv9x{7*_J=Z`V=YVWMo^bbF-V&UXkhUMtTNj?Zk@>9
zdy){sx6@v@%WtQ2s<0p2lcPraU>dB<%l?75VV~}w3K#xDm301t&Wt<Fv>fVQ=W1`A
zz+1=w+<_c)E4F3kAp#Jt*au}C+*)XCDn~k*ow!Y~nJhM|V*u4c4aZ?wa!sOnYx2={
z1w$e|_=EmW)CpM@!r{o1R@49N9AvA1qFSJ*K$F$X;#joTT<pPXr^rGG)$kj4ae{1Q
z)dbc4z`xY5DH36c8heegqU5eFV}&^k@*2dvqxQn!hkE*xmVy=k&F5>k;%&#}ZT)j>
zS0edTb<h;>iwBQq8VbY_2BHaiT*J^Tu%pa>Xv-NluY9hsH2yq);W(Q`i^h;;Iy9JF
zG&WH-5f|-Np-bEsUYKMs##w%#aW)6+x94+e`AVQ!)<la)^CH?kbyRUUAIsQM7#p0`
zoeg_DGN!_`CN8Z6Y_OekLQ9VoKgW^#CyLaF{6E5*H%c@+kJ<FxJzmx=L5N%I^b5ae
zOe&jTqHPfCJZyUfKA8n`4o-D7Cv~c-nAdRSeU_Q7N<lxUCQSmvvNaYUYHqZK>s;v=
zandGH>RadlAw8^lv*^_VDn_YWGMdLtGQ-B48rAwFQfcc4XKEAs#tEIv)|%ro8~h|1
z;=_xT`%CTOm!k6>EK-uArRibFeIA<sA?=SCp3<bQ27D3tSb{a;N*8m8o%dj3F{PrI
z19u8J@f)Po3jasPQQsf=oX$RPbi88|eEk7Tca}Rt)N_}KvJ#M0lV%Ic*V2VKESB*&
zRp-$_mS0$P0HjQx8ID^T2OSWOl4T*z2bHW0mUA7!qOdaA=J8Fl$i#@($_|Usj+D1b
zabwwdcKR>{jwz*5^231XFh3&E7y<B;$!goVy2`$Yf{yL@0n)U>`MH+EgMT0rzRNXT
z$^vBaM9+1WTToRKxC!`)Sf8nPt81~Ya-sNj)^VL^WLQI@W(#VJkOy?zrsEcNu6FwY
zEIBBEjjnDb2GC)rm2GJCmUPZ@sqI&D2Q&32)!nrB2^H^~^$~EO_Oo+p-)esyp}ry6
zs@q0y4=0s(G*syz)6>SpINHh*{HohUzsP|3Fup`gkEBNKFRA+RFxJy3in2Ku)sn0n
z%F4&=B9YkHn&j|np)hqXr7RmVe$b5eJ-`D%(Se?)z!bgpX{KRB5b2|O*4f8<EA+h5
zsPulmS8<(*>i9UOai;6E(uihYdmA6WtGef#sC>9dDBzJdh%V)KP2AH$7^xuF^-XCJ
zDo+r|>?x^J+AgONZ@{8UTW^%@aXybq=1jpSL-VDjoiRbXt$c32Pb*K&ps}bl`ESbD
z5o3=NpWONbnWC^eNQsAjZlSknZd%<dSDU)K&RYRq(Yl%o9pvQ1y}roW;vgHrdzg$S
z7%z`jNt`mHWad8ps6?j>Zhv^lzny~fYmxR(O&ZQFBN7)qO{PV#)m!ArN;>DZk)+R9
z4SHzp{b8j~l`(b+x;kc^&AxhPn*1ohy|;bKeMVz+>9ck*?#;u&gF_kuPf??}uNYE2
z3Y(`=p0h4AQDQqiEl!}EP%<i`EE`3)P9d*ktI!^KMF2CcX2b9%Sr-YAdtRyWoIMu*
znwfFvn+a<23_<14BZR@xfnO%@v*Aj<)E0?);W>~td>=0Nx)nwK`QOZeANvOxY1^TW
zcU>PM90CT6@BN+rPlL+`n+Y*<j!J*g;CnzhmL<i16ckJj4wnD>G3N#bl`&8Q^!ZYb
z+E8Y1v|^jPT?s;^4mDbLsxmtbfN2wg(2O!k4m6Embd7*MJmJHMF=y$?fy?RSJ?Mhn
zZIE#$A*0vh1R_TxA^(TEJ=4sk!O5xL`HWyr&YyJLhV1XAVaAeCQ@@QEcl(_g5ds`v
zL1)tj6GnCs`=2hXDDc~m355;-PtA})e`$h<GW(AC0<Ir7djk+($Q1oo`;dzimB#IQ
ziv1l&?#vn}0v8L36n<9zCh8CP2wF<zexUq>H#!7383}!U3~`Gb^P_F)e}7L!ID@Ae
zPfo-<z9PQ>jsE()oeg=B*`LK8vdKpXy<YszXBtWqLW|24#&t7mxwfVUaHA43j(V33
z){2|OsfB$o)4@c(DP2}{2NPhZS7x9%il_Mt6(!3{vysAKa4YOG7s%yK6u|a&FBaKN
z#cGoAN*UrY%FwdA>J_Vl1c1+TO``UcJH@$H-0VjYYc!V5&4H&Hz1d{fho*0oo(|0a
zUE}F+6hw=+8AFd2ExYmq*nYt)K+djKz%GI^%Ai}TAAmMGPY|3cC+PE_v-7DkD$k=c
z>eD&xd6IIqRcwcv8RJmPnQ7@?mb;zkM|i`X`VK4hKe}8${Zab@lUnXW!}2~M_dW<?
z?3voSd(=NUermj?9{SZa4NLNTp%~!~2kkq|=20%?=suQ&0lr8CkV~I&Pj^TwQe|WC
zPr3qMtq|i4D-tN;9{7O}N#&hK*th_H&l<#ZtJy*_{=x8eRskMd7xk;wLlM@v3tKr1
z=}r%>>h;T~%p%ym0DMl?Y!7?vtg+FjPYdMaGZ5UgP%XCBIOCb|Z93=N7sItr%PfQ0
zweL5q6t|?DK4$tY0By~bIT)Snp414HCt}p^RkY<gDV@6*qsx`6ZPez*RH@86!Bc#l
zf!T~o96z<U&2zdqJ)yaCu1Vi-0=@#((G3pULDkHAcjg|{C-+#)i~~n9d2lb+?AuK3
zq;&it{A1-qP{5qVeB2^jB+|zwX^@rX(R@5PmbkM9mEXS-ATof8gkP412!{kOfRad^
zNWcG+r8bJ#H(pssA#NlS&I5&l`Lb>rA{=I_rp6|jv<VE7<ArkH+X&H7`U9$E<gU-k
z3N^(FSU(P=gegEV2j0WGEfnhT<ZcH1Oohu(cNN#4m#V}Ex1w4e*Wb7<DwX^_UG9E0
zSbiBx=E3#^&{w24r9H^{RV9OS_>y6JlSp!v$#FuDbH2binfSNX+&W_umUWp-Fu5=Y
z7uwuf_}l^}IaNH0c;{)ge0&k!Du#Z<xq)2f<qUv?x=GLk9na%9m)uL}5OSGm-YC|2
zZT|K!Z8pe0dlUJYB>ogBcMqMcgNzo9pV$vtIhkJxh!QWsr#Q%UVn|N=p&w_yl02Dq
zJ<(PJmWf8C%^PMz(9v8{E9ED;@#sY1F!ZUI*hYAl)PjJ{L+k*)=tSa>0=AQxOo`w&
zn@?dQumL6G$7e7I0e(pJz)NhRW9c_60g&DCDi{WZ#Y-$heeU4ykYf$CG;9q9EtcLa
zTk=x?5Vhz~W5b**Fu(lzT(x3UpC8O5TN!bE#y+N8cy!5p&-R+L0)|&`-TvyRSm#f(
z+1D+K@f3~up+Iv8QH~A8I|T89jssO9gK+&eIU70tqV-vc%|bj|Qx!K7o+QX;D-=n5
znYw6n^6^9h>uGk;_N^{j{oxqG7Jux(WzjqU@R46HAC;+d!ka0<9AFk8&D@b<jo_4u
zf*#b?^DCq3N<{T_CFAmQN`XIqPS+cd9sQZ&wR8}=QSWX32z5lEGV+NRn%UBKJQtvW
zW7P(b)X9Y!{wjl@h=LU=T!^Mg+G3e1ci$k1QFq%M&CjVts;!3g=>tR?K@W%d{xhBf
z>Ynm+T{Uk_HlS-Xr9WDCe+`$1M_KDm)?QA8TLcO4owpmZT$s@L^_by*rV4b6tGNud
z?9k%UhB}zaR-P(^55B^iWuqyCCr8`xCAEcxQIs~Mg)~=S)AOxGoSefBH&hf8EtI#7
zj?3j3SE|EwgaVk87I`7`Tun>W5K?jg$A37#9NVK56{+21bFwg}C5OpeXnFP!@8)My
zFH1gH7`q-@tT|iKo%r;9)^xy-tkt&JEp-YgzcjTt;&Pw68|h(i1v*=;;06Q7wbZ5x
zpYjmQGCKtp_pqs^33Q1fmu8EN<CipPI_MNHkfk)LYJ-9>(KfbT7>nYC<w$S=$7Uha
zJy(pHj%Wj@;|O1?Yi2sTk0+v|6;6|8kLR<@wmQ4K!xF{mPY%|-Y*G`vzl#ZyzG{Hr
z2cTTZaZU*xn%Cn{-USPYZ*y5KQC$Bt%hAL5Y%&Za=4mBA?o_@K)l;(rK@P=3?0~l9
z8C=WSZSi*P=QXV1<F$A80s2P3Q)P>qNp;+YY&WWzQ|WdrgFjz=23^uk7y9(WBI}xi
zggz}Gq(NR_?EoU9kfuW792(Dqn}17}?YUq+GSm&VGs1F$pRw5LypLm%+bdLbuid)u
zbj}*2OHsaOlnpln>_<|9wRtG52BYWc?F)R#z#}*Tq-eljvi?tSr+^qhTuAwwlKs2f
zS-8=+$D15C_&o@=+9gO0L0NYILWa>ia#4zW1BPgEYB>FuO=KKd2(w2cV$9`}@34Wu
z5b7ic+*s@azr2^C-FqA=McrfVAG5((+xNDa8@q>z9}XU-xuNLES9zOcSt9D&WLy52
zro|v)+AWEG5Z=qF<AzrN#&{d2xGX-XGBAOxugbdX_#n>#hc1_|X0R95X7To(bA0UY
zKiaw#y?BORs$*o7KFyrZm5Y*BS|)~z-whqEa$gI+ENur&S8P5Wd;$isM^_r}z^>xh
zJX_Cy*@`{Y<%~RS7b{_;mF2z_q=W%6(pM&dCJcF!5>nYPLkrD-7&iDh@lw<ouS_%i
zkbpFcwER=P#rum70eb-jfg_Ktf8mZiU(dJ+hoDy`>TrgtAXvV#TJGr?dRMjckxSze
z@9fyiZ2Rtt)N(-xF%cs(m9e=5NwPmBr>S8V%giDy>>Hu4ANEH&I@7teE$v-5*ABjE
z&GeCHxfc)HM`uL<O?o8B$TQ)(;W;A}VG-{+Y$i#Y#hfG5rq7C6kz=V)XtxvQAI-2f
z^xKo|4xG6L4-ofd?C*2PZx{k>v`)4h_K8z^)I+ZBD2}`=sd*Xyhy<FuL2}b_RMX^>
z3fhEKZ%;L&6|alg2)0SO46!ITx~ve(js@%nGv$I?!cQ-NVCJkyOElZJXCLur8Qvv%
zkCC|3tERo^OrHwF^`RaP9On&*c!noAjd9~^$y5q4Llk?nGFBnZ?jNA0VQ<*bx>|DK
zI;pjGNJ!y_!>fWVe@j*jPl}@%$<m6ow;#{97s*i6>yT3=MNf#|Gu@AS*DGIETgE4C
zV|b?)-F7Ac$uBDzEljPwd~B2S6~7&~n&9Mdt@_kH6Ra1;hH{pty56Mow#H+|3XE{A
z1$el}gdh38UU!x3W{LQr3RYuH3wEfByUWBWbx)-G26(5VQy-^S9<JOV5>>>Be<aap
z5np8bpXF+!gITv_0V?;wTODD&>I+4i2)AM7mg0i}kG63Q8S_O*N&mn%-!^Ftg~TMk
zVuytXcl4dDn<#F%_bpWkL|hlwuR)eV<aoI?juB3F>+;Wk8iX^maeCWLsXJ-pDXZ%F
zxH=G{LFEj^pJ})h%TxAN7bKMxtQ|WCUz1lmUS*9}!?cEWoxqJe?5fqzNPkAtVC;;M
z8grHcgiUe)3g4TYWh4$kCBh6i)uYO#-C?Zz(Z0m{%ciS23Sxh(%Q~-p&uD5patiwo
z4kRkFYYaJ`^rrJ|5OL2atr-t@yA(9j6pBV3V@Mu>6p8vA(Qs^l)Y<fZLs{*N^93wt
zYGh(~MF!O<u|wS$qS9TaxN{T7Ij!}fcHl4o1GrGMJ6B9LU9A!KyEFv5BV;&i>7~M-
zpFwgm=&)4jES#EkOfnU>l1L?yZ1A6yq@x-?=tfPXW`I=)9J2iSTidn6J=LFcwYIbi
zXR*$BWb7KrBPDShU!+Lw*7TjGi0?cBpPRHgsfp9Z4FZW2IOD9TIZ2v}ImqX*?lN0|
zN|_uX+W=WbH#jC)V(PAs7#2Dm<86{C3`7Qd-16?<w(C*&3d_P;1ufx}stsyXuw{uI
ze0`FfeN&6}iYEqc;}y-SNDc}2CKp+MsQ=#8lM}7ZW>q(!jK6&+5hY6JYsTg5!QZOB
zEPGuLL>Xgkmyq3^v6aOU{cYIEUv$v{RQSEVpEaK(_^$kFA%uGW(AWf*H1nlp8l<I^
z-#~xAdp>K%Xk`L`E>jlzvN;;)+n}x*bLQm9S=Lp1Y5LE$gteaM{}LWc+c2uR)8XPf
zX+Rs(_6Nxbs=JXWKSXEUiPP;WnKC)<45hc`X}J(BsYqwFphlNHBY|z3QYp^^;72lc
z#J#I->MqG?^q{83gm4H9=HDe^l*ZcK*75k$yFGH471Jqz634I4#rt5Fb3P2%iFkzA
zq^5}~hb%CeYevh+h^!=^jv-63XM@d%a->}BkFc6WYSqclopL29%)U3L&gZsGuH{{p
z*b5YjagJ^Eu-g!=m=(^%o9Q_KvRxDKXGs)7)UB;}*_halX6j-gvLIXi2y}=_&Vf9@
z44!LDMPCVq#?Ep}b*ZV;fj}M;$!{<BF3!rKR7-^g$9)c3U0x_%Me`cjg&J|$A@lY8
z{E><yvKRd~yQ)^DJ~2cDDs2`<B{0hj(J!MNVyeLyrmT?fmm#NiT$BVrKdiMd`inXX
zlHHAkgz{AYP-!Pn@Np4ec2hgQ(Cf3tynYC;VS1?jJG9+0L+oaYqzqno%x4jNwomcS
z2lKIDW;1maPINnHsgYPbf?0!OJXD*Oqx#Vc41BYRj@jvYN8~T!ANco|z(RHOm3?1a
zrgubKhmKarhwIv(XtWmqzZvgV!%}x|Ip}2rzz0PyxUPQ^qDrGv&5mckOf__sANH!@
z;gV4E#kQtR$e~;A63N(M@6!6%sc*SJ#F$@sZKTpu-rPUhl>kwrZ=KhM|E3Vc=bPDb
zNzo8{Rj0qQV%2%q@o@!@8#bm1R#EakUP#_wP^Z4mvH#UOfzVEf2;dwXT)@UY2sEGt
z86p)>bseh~@cz_ZpKs)SiE6bITEGAAece6X=VNV%gqj7sevQd>)TaGwGeu~JLId~f
z0SJJe&)nejMK`e0u{h9*5-RR!b)^64DvPOm*EGu-wZ9*J=Yw}XOewjsL6Ty(3Lj|H
zyV*qL4hE5zu8ciA(N2W2qs#2Q{=levdT+IyT_OV9&7D?5kED84uXJCXBOqoc37ofQ
z;m~0|ZGK(zO|>ivoWK0)(X4)QdNFbxqvs+x2M)4EK?z@Yb$<aD*_^P7%Zi&=VYN|i
z&*roN9_2Mga^dhze_&Aq_gO30{{fzGabYC@iuwD1jtJBsUG`+~P*rKJjp<U6a6ecp
z|6qVuOLhQ@;mg_gD-MdJeY4>C#M(gNY?gh|yyGQA6k4<&7z>JMTnF(?cv2<G-+n1c
z=<b%v+CMy$=6DNumuLi`qSsX8dM)@ijf^#zGr#cKCw{~e{!3lcq!8J2yKOw2H(q3Y
zWm<%BU#U*dxy)zhiA|)Hj2)YMS*GuCAwmIQnFXRQjU(8=%YRA;C*S=oVHp{R3i_2*
zaIFF|YN`~53vs9dvXX10*&+f4zW<O>Nv$c+3b2O^_?@WR%u<%O-=W$v^Bh2V6`>cx
z^^17AQ(|g_#xYT!4I#LQ@C|cuch0;%k)CG~<??|xt0V03L7Yegg0k3P`N9B#egp^*
z^95}Dq;bR!G$6J}hIhHGI0&1Uw1_#ccc7F*lc+MKQB$#H$bd3K&I&rS_<^z3NK^64
zm57_fS{WmL`#!zxZ!CxSR~-me{Jl?K?cI*jo`h^2=(Z#|Wd>&+`fSh|T5DNiUHXH}
ziHxVAmqmzoi%9^<c(4-RZM@ina#1mWm1hofv~`5{U-geQ>K5#7Taf7{57$^Uuiv87
z3NxVC45tN8cO^l`u>s!WDl|t|0!<KEDtY;f9-;xf6?quku~IId1dH@i@lw>+O_)i1
zf^i=hAd3NOsa1{%FV^K9Yzni;|E8uQy}$`RRwI8md;i0GB33i^HLO737t=X_y?>rC
z=M*d0JI{!E&nKWT^5auf_R!li&mqtX(|@jzIey=R8NIV54ZcLb!l+wEwPgLN+te0?
z0*EkDb8Ib~gnHex(8V0H(i__*$;YlcvO9~%a3%s>T#Leo%>a67+yZg;%lcxGDj1CR
zQcwi+Y9>HgIPEgJprBY)L1X~Xr?^lM=y?b6fm~vHkfetF!n&^#%|^v*9G7A~y?KJ_
z)EH_#YyK}vAV;UOb(C0D&Q+AM)KQg|^Nxmz4agoNkT&Th#zQbvobeu{f3qfqQePyV
zX_Jd4jx<>ZKkdI>epiXFHt8VH19YH*S~;Q$HsaJ!lwGQ{yEp2xt~CI>U>RqEN~m~!
zj+Fe(p{kX+s=KZ+x3cNJq3RKP<l-pW>&{&7YC&!CSxJ~Y{~2-*dKDgz)g;1qoZ`E?
zK~aIfG%L7~BK}fB<+MQR=-F1RhbCpYz51^^*WzZa%_vQ)>&PmK@nDUOOb#3D@da~I
z@iXZ3ZA0xx@my$&Eqwvth}$>c(Egx*WB851Y%#)D^{3WQ4*^|{P+*2-lC|8xDZWyL
zC_}n!w0LO};R;fSKwvend~iXU0^&}bKZ{g8_|Nl1C8K~6>Daauf_hp-a7*w_9d#}B
zXpWe5OtRYJt2Uphxi$u)rML?73&rkC<+6TR=0}Oi?Auj|%8Uk}C{yE6vWTx<xj(Jq
zqNgoVx!sE+=Y$k=x_U=lJ;-oQ+=FNqO(pDiNQCV#M0Ry>!UkoH4$8u2@x&%=4?j+C
zVu}jP4U(r$&64#avRY+*){)A1&P$-`KyYZuwhq$0`kFpPWfkog2=sXXe&q8(t+ir~
z(O?V~J&0{R4ptL@trBH<N+<65fV~#w7h;bj${ZS=#FiDOYL@b5?~m36%qurM8UwtW
zXqRfo(b`;7>&WuB%~?F67-orKR}a*)bkQ<a2J-q{u98~wEiotXdX;)E=f#)@C`0gP
zX6N9wfnTr(K|#Mu?ZBPZrv5RgS};D)s8zF&e{h_)X3lH?^3)w5&b?<52L`TJQEUN1
z=S5J$bWLqrPg(8vys{t<iffmdF`ckmHdplrACuv0t;MupLK1hKzNnJ6Bk+=vNA&QV
z0CZLUA1TC^2<~P-5&P?N>O>egbs?sW0Ea_Dcl&eN-1t|oH`<R<brDE^3BOSgH}OIE
zMy2+Q2W*i58A#U2+`Bo(RxrwWE?4;x-VGM|ERP}|4Hblh)YsQRMGXlUu*rCqx%8sj
zN2aS>QBr&uH>GrwI&FHh8xd{<q<L4;`7}JNUTvz`^rI)=*HbD!6T%`~Tgy8;9i6?3
z%nc`5E?SDk0%vh9?GG8j*-B#gar%BSNaQ9g0myBDc5+ciL(&e@ytzD(XpSoBFk>v8
zL(>&mxk9=WX~lPC{h4iEWUTN`s85Ll)^TIrZf=WHc5P25Q>Ht_3%GQ1U@5kuSNtjG
zOsYJ#l3##g9dAc#LB-*N^PG~NYROOK`x#W;#2(K}S`Nsdykg#qlofW#6SJlTjfkQH
zd{A!yPtfmn*EcB5r~pKGB#ZU6Q*aqtKF9hF$#h&R64%*0vE@TOntt0Lyyy_|X8PSC
z%xnRj@PhcEkS6=<ry%y@qvP)GzS(~_1<Ob|C;c6v8@LOsV{)QHVL|U7n;W*MJ7}0`
z;M<QIc&g*u-+ygnG`%AGvQD20>Ul#)&^H4(Xh#L|095H`OU@#{`R@@YVGamUk1r-L
zLcx_$i?W>;Br5)p9>{?$7hf0Z;D4AR%3LAa>XTZYx#((`_cP4wUPo#1nVdHW_F0+i
z+Zgam`6}I*{7ZPaG7v(p%pOd#6F-Mfkp)I*NGaVKI{6cDtZRMl#W-j?`UXdzpECm*
zbMTVvrV;&@tqzeSw&<h5Q8sDr|6K9=9-cgBO>`R_@Ruye!#1-z`O0<gI6n%uo&J#M
z_}r_h>J@@^1&hzeQb05J-i4apKPP3LEK@xh)j#H)v9~nAP~bLnTYeo)Hy3U%1>G+`
z`wMN;Z$Jzkyh=??HCvLN7w%1;jp+&~pXIPvvfj83()v5@9IuSdRG%4KdZr<;l6Cm=
zUdP6QHPI0f#}r2J%T@q6J0LT-F_K7&yBr8t>M?!RkFty5$6_DwmKas#q2x9C8>+WA
zn9lWdu8&l@<CYb(TRIl^<g~~7fiE}|C_jl0L;q8Sy<RT?zg`#;L#;x+|0WWk1K*gf
z+CwTo`|>5K7p)a&vhl&@`N}>V!mxzgre<)^D85F*tK?}F7k`9PPW-08FDTgFIKAQa
zf?dezbs?>r3_7lM_)0$<QWeu+OVmb|4|m-d=0VF?#!NaDr&g<#={b7MfPqC%!^+fU
zTKST$GSege&kq)rYo*w$M4?B^vj|%@qJe%!&t~YvX8y1_W4jmUwO``(wH!7kkn<jx
zaz+(>4IpVf_4@x~M;Vl8z!fZb8lcq?Rz9HVz0FVGxKw;&Nd?fa_h|?(<3o-tfAVob
ziqt^nypV<bRK?{aH1lxame3$28Tlp+Xg?(miLe$6_kRyRi9i&zZID~|q$#Zba+v>Q
zHF|<F`-Rc~!uJ#OOu(2OQ&k@08H9up)U^x0=x(A1l4w_oB9vTgE!i`P{c7GXDFx87
z<lcOMu18Gct-!a4zEmT!nn%=Sc;5LZFmy%L$X2mXz1Y96N<mGAUHzIPPL+$#$n1Jl
z4$h{2Upl`_o!9hdvS4qXT2#Ux8i;|#T1X6q^MXCF3YvZ_b>5Amd)wMCinxbQ$xg}#
z+=UpF;C8mN5qN4U*RwXMY}((JqIQ5ZJJwp1PW#|UN6O9|aQUB^#6t4*lGiEuE}?u9
zk<)B0?U>{1XxC$%2N$kU8VTxSVsSiNgm_O`({|$Z`YT1#CBW-D^ev$UJJx!ZUTjl1
zEsV2&g5l4o!FTovAUuL{h+b-e<HGZ%uxCWiETqp3;)us_2A4t9wPe%Fp#tC(ChVQA
zk#32?ahvS*Yr5_cC~Al@a$P)`DMWiVftA#<i8wpUD{k^*00Dv2N4kzH5_~=#vcO43
zd*CUpdZC6^ri#CrBT6FWI#Z|oNyv%G()oDm`F2gQo%kYR{Cr5M_fBwwjsREku*+?)
zgQt9*;#+q;kgTt?AM@YTrWas*=BqUk@~fo7^2`u^;nn;-Z7}JSWv%E#6z2Ia``S`w
ze?2u7#DayZnS|P%T@JoDZ)&T1ST`eLw?g$<j${iQ>~KV8OLW_1s8y@&k-*!Vr1u95
z;h}s_tKcX(RrPh2QnW&`W(rT7H3cYtd1lK!??PBJ&!Q=8!r0WQUNPr?mPQQY*7)PI
zS`*8iU)w|NvT1=%Udt&qah1o!#0F|<OySmSq&S-MJA=wdAN?UkDw6-TTyy;g^Vhz2
z`fmgZ_6=wC6G<E5Lghb;K+qqw=)j$OtZC4m_UZ?$40wnvp=+N1jbioV!GZ_M)q~;!
zYbq$<K$(HW34%00rFsS!z(B(9fIU}fL?c9zK>bC1q(7h{|11c&k_Fs)R(LuWvUcR1
zM!QxW$qYWn(z^Xy$04OUw&`p+KHN+txu+Geu7cbWf<Fc(yF?)@iZQ^5j>4p6Uvm;9
zS$?1t9mumDXwdVsRGH79A(tcrGg+W<QbOUPsZ9ecCEAFBwWkKt0kNb&3FqkOwT90u
zVU<J{dDI^3lv_z@QW77Dg>;r_tdK}0<fPah6gFxm)+UwoXxDW*JBOrnVV(tky&A5_
zmP_;Fd&$X3UnCs+TShS_<yQ{!r1TMH^_{p$N+Zire;hmLeC_JL95LHXu_lYAa+EM$
zrplbgo|z;eIiHC~0B~CfHR=WkCRbShq#t3GszTuj=oXe2sa&MV4w`wB-P#3Ideo{r
znE#T8USciw=VErS;U=5uQpA;`LQ#?Q7N9*L!3vQ0lx2Em5dAZc5uj0Us6mb~Fpr4`
zS*WSb_rYyrz#5pbLYkqEV|=win>)lpo2|oH%DJEYGnicW0Ym;1sVu)3e16~v!=19S
z`9aD}&<j=xmnq9Y>bYB12d0biD1Z6uKr%iIWRc+NFa^M{U`1o}WRs9#W4Jc{_(YGx
zfQyMQ>p+N!SJ#JilG1Y@$ct3;AN*yE-V<bs5%QYSm|&Wk8;DVr0VdRVFUlA=h`%25
zNi8N*6I+`q4_F}3gVe?}0u73iU!Vy|Mi??ytP@7WqkOw(%-FB`JUUa;=(R4t`ut<+
zs!JEa@GYrPPi3GrXEphp0Qgw^@?Tcy(p3?<t;+RK%Ga#lQ$yV}ZSaNEPkgEa1j14e
zDIJ4ekOz7C1X4s9)A95BY4+x|mrHqg>GuY_Rr%@x0B4GM0wJm^oAp>6%iPt4+cniz
z+oHkJOlJzc)=a16Pen*PtdDkIg_)nr0X&#dpnjy5e!(xN-e~&0nqSjQ33Bi$?ExRZ
z`}fJXBbF~+c?H%dg(=U;wOw>h8)n~N()HL@!*P(>^a*tt<D2w#+2jLF1`!?vVpXmi
z+%My@0dfhBe@p+}r@1Ykno+?;-!<LZ&=oT>h3+g&PeZ8n0qLnn-d?e)U#nl8={67}
zS2$_udd;^ui8PUqoeXl#yhSGYrib!IDNkJrc*B=@Lo{=#_Vj7{yi{jnwW9Jg$JIZ+
zi$v6YZF*;>k;}0?Hudt}B2fR$RxdQ2t@Kn5=<Gr09-4peU+NdW<Dk2$wsPeqGp<E{
zN_B4`L%(9|_S}&#i?LmuEt{#yo38$Au=-Fxb9uQv=G$mZ**p@S72Nj6>vD-Bd_}O8
z#;#=Q#x0)`H<%*<)J_-LZX<XBVax#<><I}HCcfz;*(90qj9^YD8koqA8K5elmGNBx
zc<=nVsmZj06w^h#SHIB|qEs)CJ{=F@n({*<8qrw{M#V~cCuB)#1`h^X_PX%^!EPeE
zo2S-{!Q%+L0IoBim7$poHG#u2ik7BECGi<a74e0m<MZH{#6g6#2;EYTlx2L`lLM4i
z=};3bHKKt+$SLp>ae^1AFh^2^=fjQ~Q0^DlgrMm+%<f9~Cp8&v7nAT#v3wYu8Vx7-
z98rU%>jxTw>?DG;RWq6FF-hLM-=7RZL6;phV^g+AOTt5iv-rB6%pum8lMd4Bq5g%7
z*zMqON$6n_pf+^DLUEv5diXa&MLr6fvY+1MdXZj->y0-BeQ4A0L<Jcx8FZ;MAhkiI
z8yQNL`%IqtIMfqqQnDq$OKT9^9xASmqL@Ke*xg`58egQK#3B8crf^RgzJL$=Ak3@}
z!I}NEnFgRetR%#RkJ`U9e<x1wdgyv%ap@}ib58R?cFEsu#eA$LYl}$vny6=kUEcw_
zb>kIAu#d@nvoPTbRIo38y!tU75IOx5V}S@}yR=@WWGd-*Q|&}a$QxT5M{k~%!AzmA
z<ArU(<w%QQL@m`r*@tg)c~VHnHsmek?|YCzD1xB0?VQx)+0R-Ytx{H=Y6R3b!W?Py
z;m-&RdalW>INrCmte1I)<1J%39^B^AEsqxpssb-Y8~zwHQJe0NIOpgkfUCDHxa@CX
zy{ahTc+pu{jsRC`{VPdAujfsS3*I3(b$9}J+P3YSS>ys#A~4zrvpuMyYq}v{zQAZM
zvx0*mT=n~kg3Sn&RIfiOX2hK{^QwByQV6oww+$a}4^}@pd5ltj@r5a28c@tFoqTma
z*o4(55cFw@ZTqw8<>VV$18}zCJ@cMnJ@NGAS}bZvZV7Y9N9?C}6{tqo#T5xd@cXV~
zj%|zQHU;CRHfR6PNAWdrx)z(^$I?Ql1mK4Xw!}G}T46-FsXDL3oc0HswjHu^Zm`kN
z&lg&EvY51}gx|qOk}dPc27qzzENs@Z=~LM*x;8#WA`BvUU*k|715D>T`MC*su@_xH
zY))7pxg%~0wqb@4rhTSuLwvT<@8`ot_eF;)q+31R2ES%>0pMNa-V&jFr9wmu=VlTM
zf2bvXHf`|m^0P0qeRaMq>yP8g#U@fU%DKJQ_5R6vIC+Wc>0(GE(EgqDvN6HX)ySI`
zf~ym)6<;Gc+<m2P0N}g23E&k+eE_wba)pV%9`iQ%J_D1m*t=~Q>;-DWQk_I<j?g6d
zJJpC+Rj9OY(c<?zsyDghUxX^u2}(!A4SjBQWEi8R^bd5NC-lha2Hz3p{Nc3G-mp))
z_GhG8>E@B~_cUrp!>c?Zsi$3KBEPPVkQ*{fYWdwL4VNuK$59JitKqX^1dx;NCcRIV
zt}ady_i}oT%tc(UMwXcp8q&_5O4F2HYvuu%`8`_w{tJyAn9B6ua)bxy;`9RzjEe)9
zB*;w#B!<N<1aw_q)aMKJ-?IOl&3P93a`?PCd^p}^Iy6n>D^pSE|MI>=LWhCUR2+1M
z(v=5%yS;xM*ORTYVwZbizY9HH%n6kh9y>+l$D;{#u0|3FwHn7~%h-+m5{mv*ALb+9
zp=PVG|Jfj_Lu0vJvcOt^V>LMJEm0$$63~wTZY`Dy0gx|(pKqN95i+Rh=tQ^308<kW
zQC2%`!3X<c!DZ^R35GTr=k&lXWJx?r`U(4RuDcPYrzA2%?NhiI@OjJDwC1*7o4I@T
zlzn0-e5WoVe5hw(D<RJ=3-~LZOFJUms(Sf`XITqodBy80w|CQ25h%RNtq0WNXlI{P
za5NA@0gyAZbH|f<8>H-w)yk+KLCe{ZN(k92OtK6d+bfV3Umc77EWr@%p$#_l#I(S~
zZ|WS*OBt@KKG!HtRL5dh8zr<sBgf%q{i91Nwy#MeH|@@>a;@u#6T+XY3?K8f&wCNW
z^(hhq6-W1;dX-Ae0DKJZFR(UW+$&1sJUh@k10?C9NNp^9j~vsfog~?dW6yDJ<hT<~
z;U0Y6ceD25)&An{?`jg>EopbLWVol-&oaxBD5wgV-udyZK{JKyV$9x33eQ;E3Vf>;
zrqa2Jaz=tR_*BJ&aOpIu>|s=#ut>)*kW_^$d$e~<TF=I$5@CU5!?bX|_Z`$_t{p4m
z05H^X#qu^F=9=q@kkn=Op24|^egubp8Z`(^Vd%P|)cOX$6-BX<-LXUwM0&46hB!N}
zV`2L;{a_jq&VZPqT}nvuiixem@w-m+vchR}lN4z`VdCM*SoN4yRBi72=V!(}QlYN`
z;k9fLl!YXXH|c>JRaC&Xs-kb+($7BA1Pp(yOU+JW&chves_-fTg09bTuAyaTY8bAi
zgeg8|Icp^@iT)LwdMlwvOnln8->S+~e;BhqOqVOznZrzDaVG#}d)&}*nW6-9;rsMo
zhQu-=iEJduU&mU4Um=9w^0@=+XF`P+39Rt>-yC1tQl)gEGKJtKs@By`C?_fAfOz33
zb2?ZEEc?N2a?aUUFeADdho;bqp(Mx1s`?CXl5w>#85}rTNW)e-pM1{E8PVmff0^um
zYqmMn(L`uBxqqwx++}toSQ~T8*GF>-S6U2fm5<+$r%YMJMo`tSy-f9;w-J8fHk2G0
z6QOqc$=ENml2#8~Ei|d*a+yz}1KtuPvG^cPG_0`>Q1kCuc_bTjwq)r9=03no9eWwr
zT*e}!TJV)h|Jq~zU0f8WY^3;55&3j$@i;F|S$<e-*{C57b6vABDwSRU@@jYu$W?nQ
zCHq)qA-0$wsmDWUr5KQ|?oKamw_vz*D+g~mYi(%=E!4=GL;=~)5?96nfI-RS))qXJ
zksIL<#8e6+#Xt#Yvp?|7HsTitY?vfbRX6#XlPw5j2ua5o3V5fj=Dc5Dy}I3|L91@j
zY|I_TPz^IAJBVF%4a+R17t##ePjddXDk-P8V_!KQf(MO<-saVQz(d05qakH`ZdLzX
zg{0N#P&T$`-=vV_>b*mFKrcprHdpddgt{w)=Jew}V-YHiVkRiPu@B2)k{3tFDJTo%
zsYAlG>UOqisfCNg5W5cp_zJ>7RZ7sQ0^FXzkT{JIYb+CEf=Q|?HJL_hEQ)qnP3GMm
zXe|62R}q~Qgrcu}a^pFts@HZ7F$9%KK}S3id4E%x<q?>zEwpJSpnWKI2k28x&L$h%
zTihronI@DMHXoe?6<@rSS(21~8@jH)f}(>uc4FF?NWKm>yb=u#gnx~NMY@Fe#UaMd
z7X9G`b_+*pIp}6VHNH}=vON1BFQdutaAtT9-EqM3{{WFdZoiQ1U00>^lE-*Yuueh=
za#rD$21DLySFjG`-Z?Ih57vJHd<sE55;@&XV&lvtSrC%+rv&RDHO*bN6~|+)5Ud|Z
z?tepaTQI*LpcUW=BjfbJz#DH;Tr@4VM#wd^X^LSY6u;7$4-w2%PKEGNRMe?-d#Ndt
zd3jBy^CzW<-z8<B&0o75%xvWTZZ55GpL});N9V|tuNIIuLoU9U*Fk^C<R<P*Y*Mr$
z8Ro7u>w`f~LB2GN?<;d5wn=SYIxd^0gj&3VwiFAtB2=+RiHwZzxuqMV1N=toe{V!>
zH+OHN<{b@d>~oBzbQHwbipiI;++i&D1GECXvCIkgB;c)NvME48J^Rp>pi`dhn~>q8
z+*-xD9%={&eBO&b>8^jCxv>sd6XAIpiJI96%x(s*0HX~%d=tyK7`ivrvUY>XXE!_}
zw9JQDDqpUpQ~RSJ+4Ak+=iYU%uoH<NTs(=Nn$z~=mP$AMmfkW&<U${$H)Iv{fpsdm
z5fI-9cEfKRrmMORwDLCC!?#XLF3qbs6Inp-R&tf@X(Q~r@mYVtT}4OzSlp2!7H5dK
zV^FKMxx@ykSfLgfWFuqx+)6p-m2BKHU^C{bP5@Nn@4%QPs#U^UV{HU$H&a)@g6Ld7
z2Q07*jA&~nInPx@-1L6ifF8`R21QZ5!mc*kqz!s{&AZr?LuH)PMhS2m5!=n#6~tt8
z$e)9lG)xKE=~sW2U)%Tt*c=mY%8N?&2QkB;+0Zvu#M3$2ImnG2+%sf*$S4a_bNFW*
zZzF)a8NC9ah`#xA_C=wY!fGHd?YZpamYu7rW&bZn#*H75WnN{5W1kHdyq0WL`X94B
zyyW{vR_|uX3afeahM%J=9GPk=!XjPoF-rDdr1<n_BYb~uX1$vJ{>zryUw*$sP<OBL
zHurqJ()+AW5mez6L->kNABj-+1GECXjn4JKz#~JFQh&YxO<!3lcDn>M3+FPst-B9O
z7mn24?oBJy`Li4qKNv_&&C^;3?4L@m7JbP2=+d3<&((}_#y`|C^!77Y`|()8SqEqO
zIr?oR)53p#>Gvc0y%(Pq*fF~0$Kno>p~Uc+pGH^DT4Z$y3^B>CHZd0)vc4ZVvMTzs
z<bIcKFz@EGx^F`_toi@+6{tvevp6oS27Yziw;SE;Mre0qu!60M4*ap$O3Nf!Zse@i
zflo28uC^|O*&>x<<~fPNWNfg|v)HOIITv(Pq&$C8Yrxg3(sB>Dsu58$W^6WsyBmZR
za3MPP&oS6Ed4K%l%V0lXuy+Hr0=$f&#&dut?KW1nsnk+?Gxccj7%mP)P_<Pzb}wax
zMgB`tc+%MeAq=y!euFojQCZA1V=i}^F%uz2J7edW;p(qSKL36+!2)A=S+>K|xGl95
z`A2_u$ZSBend4hS5f>?L!+bY3E12^bN*rAH_6NWSbFP`yYHKECNIARsd(N?I1+%US
z6lt9jt<zExaz<+*Z`L4bZzVMEKyo+7ZzJhqu<#V+^=G6xe(h@y>}$K>xsAGwp~6#8
zw~9HX+tY<BM|$&d7T)=m*;zd=TyNW$w|g>`x0&a^!F>+v3@~TEBKJoWW+yne5!WVs
zI{Lu5&wD9sJLyVPQOF~{mmBZ@1&u}5Lze+>4if@4FqZ+VE)xPbGM8a84JUtF%W~|v
z4c+@I`Uft<2l%RTbyd4<->TVV;@#A&l1w(ai_9j!ACLq|P*O{}a(89pg)Fy3fg}z9
z9`GSg|2oOj_otVSr@y=#IC&AQ^7{1g$4TJ}iGP*Eht8h7Qo<Q^`uK7B?O96g@Or@u
zq%;OEkN*d}+w(W00+vu0ywrdBn+HYZjmAsi|JU@pmdjt;@>XB61QVZ0vcxNu^h%l}
z6U(WXuJi97zdT85{<N*mgTNNGw&GXzdj1AV<1cu}U%FiIh8MQRVyocNN?^^_*IyQU
z2G~;wQwe~rmGAh}xK!>klyM|SH96@?;ANWRR$uC)ailHB%k>u5r&)i$y}m2S!y2ru
zGU>T4J+u>v<(D!q^d46FoJuDgtzmLysR!P9A9E{{hQhoB*&bxlR<eWq^iumh>aS{R
z5r>5R40ET247y^TW#4FPvAgxQrc%DzLbKKUE^By@;;EcvQplpHl%cxlO)8~ciw%qD
zk{PGl_sOrfkEj34J8FN7P*I)!{^KdV0HcJ#ek&)Uai>53dHUn&uT!bq&4WTWm(E!`
zxVe((`+(A;0M}fn!+>|ardE2tq>l7Fre|mM-1VKsVsZIW#AN-D1*bCdx}+cK$yT&-
z(g-6>?MCs$V&MhXYHfs%J?q>ytXoO6^l;0B(*2bR9t6;0bOL|%C;(Bl!UezT2donk
z4kDoNuGoFCt2$WWB22rg9f%-(?Q0cLMPo;DeF6VpP#<_<lu0sw`(2*?fiHi-!X}2(
ze{<0vtsZ&$_vvp*nHz!wsxt`&p|o8SEKgd*5J_woVbN<O#Ge<jvt`NUom&@+_(waH
zyJ|2>Xlqco6JURvtDu+M>B`=IKXLybk>mj#ox#o<?uP-|0A6dM(NXRTytnue8?CJX
zbWQ@E^A0ETSf7}dOQSY1uXN9EGyjnFzw~8_F?kRm9x@#VOG)iI7f5fx4KnIz94n96
zsr|!R+R%RT5D0hHK3JhcK;a<J^{Vz$3{IR8V=iewvsQmgnoty(iBXcyP+t<qG3^=0
zBVL-AQXORzM`c>}Y&?wgr7dG)so8{%P{)GEK*jgU<!0+Z3s8{$--+5`?QWnZB`P>R
zC|Dw|u-<<K%g+SM!vL)V4<QO;9)mkXEu0FCRT=V#<ZYHBUnndJY2i1ebGlksFS*y1
zm@+Tk!e4(*j$WHPIqqEGcnU}EM=2gpC~Obha**ct;c{4i8?g8wFn@m#E<p*0A6A~v
z#}~>kvLc+WiF?;e8?LBuL5dH4EsJr7^m(}B8wzzE^9TJPg?cF#v%hgb=Q^p0Q+5hh
z^1W*56U_mU2vbun8a8~j_a#|%5FF9lmZg{4b?tuxU&DOp$xXSJov0s1X9IOxX!?Co
zw?_ErQFs&{6i=5k5i@;)YC%JxXr^mBYuT|Gfp814nRa2Y0AO^7C$AMXy<OK%v<_=^
z11%`r`Y~uh5S~qS(Hx9@VK(kZ+BQMY%rAzb>ZEX(-x=pVm^}8aCF-dOrI+0n5O*SW
zShIf{h?&I4?Sl|AUIaZ%M?tmU!1%z8iCfSM)7G}nsJZfJ4l)abN2+Axq8U@{1~IK~
zCxC|)y#b&u)ci5SqIM#<mMe4WkPJA}vTJv>%KvI&%y>k#d8i7nqZ=_hRlS+YWbri~
zzSsLsRUcN#hN@Lz!;kS5Wkr;>MK?v)Ye|2S)3T7f4#pQ}(TkN&%S*j{jzW74^>8mY
zv?mK^eT<<>5E0y0g8GRB^*BK5z?+!3^_^k2!b*>#jfkP;ry~VuDNT10Q<KtR;2u-b
zK6ZGTOM>;%NC?x?UY>r+%&k3QZ(GXzI=<Bs78zuJ)mzWkGD3ZtS1}x0wb(%UC>nno
z82iGcAL8657$J51a_&z!_i=zWfOmzI-xqkJh1JQbw2*z)(Nu33(y{9TsqXGGWNmTA
zy<4#>6@!vWPo>qlf3`Bf$h>^3zZ(Zy&-!}37uJh;z5Ad&48aDpwy^id__R|3|Nrvo
z&-nCVfHr`K!ulTryp=*(&F9+M(hPqMxJ&88WQ`1VPDUQ@%7%(eVe5}k5jI}MmNw7!
zFf2|*w%t;GnyKc<^}@oDo5mT=XtW%rwE~01Ow*6HLh9Ef$~P`?isZsH$)&`@1&RAG
zKa9-==DOq<cU~8b!UAHhgV1JorcD@JR|XlfkmUuHVU=KUdUG5oXR;RNRd#<)cwK%1
z$-^4oKvI>o;W6H;ywKi$jclJ8*$xA=0lX}k#AAS$P8j9*7POM_jWb<+M!#)&kZT09
zU4H)>-0^;{>rcl_$RDCU9j!n{^W`_6@SDT<Y``8+d{Cd+BKXrH_U($)A&9V^``p?U
zIo*D_<1m1>Rs`8RoNwl)4Yhx4137KDUE7EXOWUuToHB`}-1CI`0ekQGhHZ%FQYV`N
zD{oU`S|sn?Fi)k~EPXB+DlNSK*6`Mo>(GL_Cd;-e*LBi#xM4-hl?1EBjq^IKO?flJ
z($jW)fTg1fU>S|A6LC~_aVsAkKgi!fJ$_Kiw;Bhzb>-f$)kx3lSl)k`u+cU%;?dib
zTO4Qo4THQDG#C25Z~!777x8(Pv~9g>S_x#eFMS25ZL19Y+uk}l*i!1QD@WI5oHb?x
zYzxzhj*ZvPj3U1`nKX3G<ARd{FuPo+V>msqdOCm~4WQbVqY$PCLdRKcT%**3&HY#9
zNQw<f@99`lvH<$Fq85MZleQ7lF>hCsDC64!BDJb9;}yp187g(_<!y9o+E1G~SIeww
z5nM!?p=Mfy?o~~1mRa+t4@)+SQK{KX*F#67jN00!F$!%ZjbjG6&Yx*Sn?3#Y`)y0B
z)|`m3>rqS60WwMi%?9$u^mjZ)_6@R`hLA(>vt~DqQaOQaV$y#u`F^*aYpq0#YOCjZ
z=WfAe4dh;%nVas?fB`9#TT%g<gM7ZbDVVf(NBeszM7LG0Y6|B%o%EB60B41=>b4?)
zjZI~IVq?FJoK2)BPHrE6(-H}9Um|EVO#HJ(y%sMN+lkP4iJy&>oaN*gIc0fku}&k?
zhIH&_Y~#bvmaTt=KckcN#F^q^59~vt+D|YRsWzM<3ebrK&%+Z|Camd)YA*fsL}CHY
z<-lwfY>4GSm>yKiPmDh+8J}4wquo9m{cA0;Qh3>>nN5XkAeU`s)3fscSAJr$otYcX
zL*}L4z-H3yNXwxcPGIK;ilLP_fG<o)nTmD6;oU`SY>s~gi*J7Ir+qcx5USj?hx&bm
z+=qfgo{D&1JVUH~%Fp6dD&hXA_6ER#R?pg7b$(%Vt8Ss~=M{&9il1Mlq(irzU}5>}
zD!@7+NqK2B^L`K7x*77M72d1y>vFP~mhqQ#4ay6z!wp?$A6v6*qlI_zre(t}%Pyt3
zWkXBnYT19wm~`eIy;Is%+?J0qH;EH>Yq0!#T|=6wwwvBWItA1duio~iqYdv@fa9pU
zhxezwB6&a@Q|)UX!n(KDG?AsaWTCbO?CY(8C+#xn7GC8K=gU}J1~(-1P5-^NwtL0h
zI^$hW!;_qaGiuBI$ZJoQC2MkVAe~1(dZS;<+srsv%vo7@zL}**C!%btBKpcF3ixEi
zh|oN0R7&1#m?F>e#6(-&csqB*FT<({_v?Cp`*`{vfq6xjA!`~G6*e(4ATS_rVrmLJ
zJTFXTZfA68ATyU?wLvMDDIXZ9e_H$IyN9nYcoLtpVCh%-=)b*sxcKFwSH^0=gcZ}f
z#|wI(m0-%tNqOk6r(gdzy}x+VuUdF<kT664;kREuOx?)aWX`~0PHwF+jX4F6Jgf~&
z(WBX%2BtGGk%nmuOr%={pWJ(~x|*^OT=c8Sm-C9zd}?0l1hIEzHl8UMe<N{;vVwyF
z8-vX=_+s2?a%&D}GX09mCP%SfPG;?xR_;s6+^#6^t6hW5wVqcc_to_5HjSHkT<fIU
znsdQfdsl^&EI3E5`pblguctU03#j1D^zSJSGa-!*Sk4OJEG?_>k4$#)<ewkj|LNWL
zf`RNenhJks(um-v$N6gre@r`P<@9m>F1TP`8b~Hsz-DNSDCT6qiOHnP$h+x6OUHyV
zz8?Na!&umBDQo(8)LfjJxh!Kv5?Zv3W}YAG!7SRsBx+V9p>MQGqNT5(&s6j^s#<-|
zQ$h9j>Eiu#@g~{USm|hr=kvnB;@_HW1=GSQQQJ2UdBubiQf0dYe+O^a7_Id!gOXWq
zwStD8IWF*9;kH;mjZ}K)A*_{G$V2(FAK$P;rz=K^s~Tp4m;zkVMn@INy0e9s-`_nZ
zx(}dwU472$T*DGf8#`UYi<#1z){3#z_|?>ursvuDMCQ{N=#0nHmN~7Rvu+Wvd^1%d
z&AMGL%Dp+=AS_66e}X--RL6Eem|~u`;&GUw({$qlv#vxYO3C^zMW>i<4P_davW;6J
zAkEM*UX@unANfpcpI0NUUNPF)+(=6EaYddYoP<x(v5aR&88t`nqtUmE=2A&!pj@vB
zPjNMAL{K3{aJTGseCCrrA?Xw4ZK8YTq!4{94?QOWEQlrKe?=6p4_R}aR?jPweJHvU
zSAL&ob$K`2$KMD?wz=!`PQNd4w`O@$7CPE%ErDDh?}~*YfI1}Vbx%RwBAfTrw67**
z#+avM?F&&~+k6wnE;-8`j^>4|X*z+OuR4c}Ad3Wx_Se#x&)##|NqoMX^_W*A>N4Zi
zyeIMdZKlB9fAa`QZ*Poyo~6;x7yphvV6&emi}@69T$!C@*RL8xm*wcrWg6whUWrp)
zjr_+J|G9||v&l1Lba|PXU0L*y7L7(KbCLqHQ<u@?@adaQ&)rZ3zrMNFFlQIg6|zgE
zLqihXw(F{SI~p)`b1jl{ikEAkX@K*yAi&4onS%2<f5H#PV+(v_|GPL}TBbElnnsE@
z@=M@PW0{n;U-2_z<F)0YI^&kh<7l#BpzT~*rLEOmn)BkEN`Ic}Ssd9PMU4r&h&Li!
z4oA}tWFP^IIT7(112E@&$cPtZ2+S*0!Gji~CSXD)ECUVPEu#(EU5E=>EexumEW`g8
z*0ngef7m`Rkta}3I1f;C9@a&Y2`*^c0@(G@|5L21Ewdqf5v*%~L=gQ>#JYKQjqwgK
zFPw;TD|TyVb5ZPcoD1Y^jQK2}t8wPBAtwS|?Qoo=FMwZDhlm0frSce7<(!U{Q(f{l
zw#kSQ>g7<F)>wc`9=Wee`;?4Dm%rl7wIr@we-5x2TFvk~OBeZN2w_vO&UsrTNHyzm
zy8wqM^Ieh9R={jt)nx9flB>YjtkGhl;|5f<c9x}-12d09BVCIL5K25(z6553lECGh
zM^~RzWq=MQg7a>lb6SdRB1j@%t)h?YOBL0T?nlT&MlBQKkO{h(%9~!TmJxV@G_QJ|
ze}Izp?3H6!IsDo&T#nRQj=avzp=)h4v%=TG!(#<!)mjgiq9+nlIX9*T#Od_ZKAJ$S
z(!SZ>3c8HTjl!1);4aym*SRFJsgyExahcY%k}7B`k+PaL9AdT$&6<~otDzWVB<_1L
z;<knYJ#}tKn^56=R|SLlsqHEWKNMF=f6XXg)YATCC@u0X(@ONxE>(OhZF@Y+=ppKs
zBr~g$c^p93iRQ0{%6J0|VOx^luZCQLj*XLsL=L$qY8E<s&JHV~nczo`HiyLI#IwU?
z++7N%^ciaF#izEaZc|U!*x_C)=xT#%1^*Vu><VR*ySzy%pzdBF)&g4l)=4k<e^6^J
z0FP|wwVwMCuB)n(W~%Mg=+q6GeSN4MP&V_*9cwps0PPXz#5wy<(I24foc%X&8HJjt
zvqTEReK=A9P`K(BmT=UarC}HXoCTt9*4hh0baCL1*q?H+lY6E%z=pW1M}H$IDO9k!
z6~5lebr+SWrX6aShU_aix0Ij;f3O@!0^^wLE%q(kbsTamsF2`}L>)`&6f>$PGGffi
zVSjq7d>ZS7fQW<Jl8n8Z0;2DYxD(wt%9hS@%{gx%>+A$6@|6U=5wWNr=r)bJuC&tB
z+VJa2ukVMVV%Kye-;8%&S%s|K`OC908|xWY-42TH#>RC9aaRqD+a!@Qe~hb<lmeJF
z()|AH>RMR4<5yZK<g>H<N^6Ozd6r)RbunyauliQhbef9i`UQXzd}2WY^YE~iV{Y@!
zXVo*U0XSdHOlB!lB2`&Gm0?SP#kazx%i(gCtLlv9K3x2KS#{7#;kBvFs$9Nmy(YS~
zVomTI;%7~EWgZuYW!9m#e@|oIUe8?gEfxt?tJSRUyp_x?uYNSW`mi*e+dgB4^J`Me
zU@QEc*3kLc{7tAky`;88(Ul7w4sTXU{rEJJR~T75&F#wFcq=g9?z9qCUHRhKD*Fb_
z(v6O`$Z^Cad5GXmB%kHtnPRxBdeg>7-n-l%6bFvcZgt}kAe=J|e^L<zN;crKfNN8U
zSFRL&o1|vL@e*MVO_%@4z1RQ{3h@+4L?Hi!A{2>0nq^{?z83S8H!ZP5wS!bAI75Q#
zN*B9MVwn%7B`JB%tdiiQkkk#y+ss8zNC;ji*#jKKsf3Y?i@7$x{CTGglT5$@l_4&_
zzZz=q0c~JaB{_2Hf83_He8{T!Aqlb=m#(cxwxA6%b`XtRV?;@}uqL#cd^JV|)EU8}
zn&O9{B1)K8w~!(sm{STX+Ict(6hiHyeSJ+TQkY9ga(jzlHR@znD_q`=_Xt_oz|aH`
ztx0?I<35%|;%>O?Ssbz4h?Zo}<RB3pmmh|@(C%Pc*Hu{uf0a~{xQ>aO3+Ka_m!g~G
zaE{%S5vYHB%k0M;1nW>)b|f!m`Nm}h;U<L$(@;L0QDFRV`P<{wDa|yhZN!4s)KQIJ
za!bx3xy|jDtXXo8TfK(OU4bTmkCD*1!CZMnBKj0edQ(Spffbw{M{|XW2n$MSm`DxH
z0;#DvxDscne|Y6OHst6Ic&-h`CsK1nC)sj&_Vnq~XP?F)9pOV`$%fVAxRAuAa^0Gv
zDULV7I`vjaPz8YIJ4SvhcW4t;ONssRuRA>8ZUM@O!1~j*l{qXjnQSOt?3CWYGt+s)
zUWlPUhO<iTQk6GnPH9I|q({juPl^mYAa1KxkWV|Ye-5w&#+NcM7sN@y1eVd*pN7j$
z5T>hNwPB%+#D^l%an}mAllX8*40M6FGalR@Nqo*n1XQn3z{!@s#Ab~^e6)SzbE~P$
znG~vJvU^&nuj~~9F~C`RS9<5LAmA~2omL|B8Y@mYE<_Rl;{*<Y((;t^68qf%C=JQZ
zfuRH|e*qR63ykpvWmt=YB5^9DSUoRkhM?;=rB6C9@y_0~a_pQJ1<+zbyWczK1<@+#
zZogCV<)_K#Bmjv#OFk!nF#Sp3CCgWH>A}ltjiC%(0}?aG1h>VL)mTJh;ys>l@MIZI
z+9*t=_w5=HcdYJgd{L(%ISujvit(sbMY#fbf82?X2gMHdy+EGQTJc>YXa*p@Y6{#9
zufT+%`z+V2zk@b~6--dH(vI2IiW@puOD~M~I*hvh<sCU&HEjOyoj}*a+`AIwK>F+U
zSTB-2sw+*VM6TuGL<c4+Nj9-D?Vn&q8Q@C7?U?aUzg6ouk?cy3XXZV=J{MBP8m0Pc
zf0gEa(1fQ&dQBm=LP|$0U!%*lJEk4?M}n}<uOR+&-?QF&t60!2b^1Gxn$w1c$J(vx
zwY<@|p&!CI-3@nK`CWgTA@5TqBfDxQVG{GQeT`$;S$-(hOS$dOeo@*~c_rI)(8g&6
zC|b)#^3t@~xMxra7_`cscdg$laeTddf80K-3b75qh8U>nRYrtDn>6|J`0!c)UL7W@
znNF{~T>kctcRLqyJ|f<l%Zr_>HIK?lHx%#3JObpRKkln2=vs28^1xoTT>h1=Xm%PY
z@fDrWa~t1*VwYk?SQC|wqF(nj<e8S>?LK_rW<tK_UbLRMg|1|5jMY4zj`O?rL4?>F
zqmA`8I<`_<W91hke)?$_y3l*0HSzeJ141sg-PFQk1A|jbm3q3BPs=e6Pdr*CAFi6e
zN?qRm`VTIuAdQy+Zw?azHZqs-7CRJ|lEE-1m&+^|I)4xVK@cBDb{{+0eRK2pdUOAx
zv2(&3<)C~0&?)*s(qAR1(VEUFC9GE6>!<GfucZV_&m%o3rPlN~`t0d#jt8wg^-vG=
zsN=z*O8qs^qv-P!f6x8mXw&}+50QcnUnk0v9`5Os)J&85sX));@rT!MZ)6g`meo2G
zB+(!%@qaUp`8oQAPsamGWNdCBqS64j<bEs{+SDp1JxF?dq4!hnWqdyl>5ao!<GBCu
z%^ZO_av!7n?~uQzu}pcCk;mgB|B`9IOPak*eIe3G@(<&v6~k1Dl8OmQQk!LrR9Khy
zFBZBWf(46H`)g~22~g<t_RC#AWZ7lNS^~VS&3~Ytd!rE#wi~~`slL5?z4_P7n#MSR
z4!a*e-S7{nmCzpOXB1ZY?(<LGhnw$evQx+Qilmg*8dEs7k|a8I)+m7hS;}l`bV*e&
zNou5F!=KL|yQ<~=#0|A22~|xQ<g(Ku3LomudL{-nJo0KtYssE2me)eL87Eq!i(+}w
zx__RjhgWIzVV3KuS66&LCqL$fr$}+8Xe5cO9u%EDpDCr)$CP-c<BI5KZI-n0lr&%+
zO;zbm37Irdv`jUSC4nG}#KAB+tqkntk5bTQG+CMz!v#UctEsKZ`*J*386ywd@!je6
z3lxF9ulXq~q&tnG5drm#;`7-QY0g3gH-DpO7c~g9gwd$FH>v(2>nbk{nrm6Nq_)Gl
z?lS2q>KX5*(hob4%6n9r38uBz?;VS?px9}RuBq;rg-v^#OuZi`vpT^afTD*DvsuYm
zI6R4}MgbaXs?|yudwYS@tJcB!HZs8q3;xPBs;F+;NWbg(sCV1~DY9VFE6XlvS$}nM
zm{-w4^SXZta%pSKR;wiH=TXdIZF9{QswikX$v<*F<<hQV#(DB*nD+6#KIGG7cJoG9
zn%6WkK#By<Wo8uT8rGwO(Lx7bQ`})QIr>B23T4cg$xKO<8`5sGvHix_+1@Z`46GK;
zj2u#^jA^bPRo{zPgos(s)3-F5oE)XuRWhulB+iQfrtUopPfLlQqPLp7morTnNF|pW
zFc>3$>^uL7T|W(MWcRo3&z#9_5cag(h%h*zC>v-HmK}&dsbx=gBw=(AbWAv%#7@bS
z6#$`4mYgq`nr<^&r!xfyNkyrGDmF5u*<<MO>CnlKP2Y9;On$03R6eG5lis<HKe9`Y
zE0SaDK-lhlLV9PRo_eO^lDUW?Cb3OF#kZk<TI61%m5b#v_|Msf-f|!<059_F6DjOm
z%S2V8I%h4{K_hy)@g3!s!SR`BUT{H2r^qq#bBQEIf0t5OHz_)xZ)~o-)66&6JgVGg
zkF9h4AQ>er!Ho$vGA-4V;CKL^Pmif;QHrAGqNLVbJz1Z8tgV<>m1$KnfSdy=@o)cs
zBDG4z4CfX{p%Gd_IqDvp1rXrAI8#*0pYpuM>P(YAo!T|IbS1Tscg?xd=^a3qqjJ^8
z(HW9jwJ%e)Kr4bGD;i7ba_@Jys?8-h@^)Kt`NQ2oS-ZhZb1p7@$qbr?(6j>t=Vppf
z?OGHc-!#;w;2bI*{;+npXe2v)vty2blM55Q)5QW_I&jd=<ayBx<k+EUn~GV_to8kn
zFwf<rL=wT(a;~rXbCT(}$4wUcn8hzE4}Bn^uojA|x)^X;S*=u;lrvFc6=G%~1+9=U
z&2sxP=5jF<!&m^dsb-go7OSobNZw5gf~$rmdC#A+ck9Ikp=sqgn!VIQ+2o^tpYIM@
z-v0X^KemI0QNl}Ga{FWEW)_E75E`Hy@}bSe((;p74%rS3P6^7S%Td3iRr_mvyK(g{
zGY*EV<!6jlbB0<fKA{zZh9!?{OJ2HA4td`+8n1VVM)Sg@gW2uh+kpir2C)xLxqNkZ
z0OtnzR%+v#B%t+Ns`A&i7O+o$Q7YYm=^6g~^W9c8zT)?#&CW^TLgga^la=7inFoYY
zhrK4|T|lS+r7>4T_bf@}G|BCLNp9Cx9g+jNdWl5UwakkO1n7#dY%@}{d#cJtAMf_$
zM%ozxevID_Id%=DleHZ{Y1)5jOROjupN}LU^QZB|)6Un(Swfgm<by+hhQg+R*M69)
z0MF(i(@;7bT1r@of5%iaz>2S3hO**jayEoTDV&E=4_gox1#m;IR;lh>2n#4scsYl#
zDDoJv7eiRu07urGW=?YjJOjO=B~3z=L|4HpDXcbC=X1b~9tA7u#KS53xo#Ft%R%xC
zz|LCQ;i2Yk+T2q$>T>RX32Cdi+a~u6v|P*`ILOP|+_6^D;GPP^mE+090wX1Xv3`~V
znwwxT4;ceI!(o!=mDE@Lv`2HQ-sd=Ks{B@zhLG%rusnAdqNQv-@*I!Ju2;!0EEkqz
zS%!r8T`oh$?$%`R>v&9dp37wbiuTU0%E0xvhh-V)yb;387Hb}VtSNOdZ*fJSqTIeB
zP<4L8D*_d#jVl7RT<2dA&?-=l`2TQ4KvP)s`dU{6Dr%cBk(Tx%>A5iDoxPIu6}2~q
zz`@&CiJHQzzQ=g_ymZV-vq$vhg=UseQQa|%37@DjH|DGn^Rg4bfGl=;_)0;jmh5>s
zWEm4LNmRWr%=UnP4swnTP8HP{K9_?L_#oy+THe(!a|6n~gSj@!UQs;>Wj{B!s(4n%
zx3&GT9yOx;I~?;Gw*azhlyaPM3m~~lN#J@l&EtRzT<?wmogHaWUh|GXMez;p2!KH`
zUY~k{I|6`ONWXPQAnjW1I|9?&1efCViu{XGmTThmAVXY#vRB9JLDE@YU4+-4Ok$s8
zd?qmFjqrLEDD%N(yxyf0J-6+B`~qG-5Y3%3@rMzIA8>9=M4l2BZqk@g65k1fpRd0y
zyMQq65{EgJ!oDTKR2i85-@xFC(jF(r^ZhgtoJ9V76c|$5mkR)Hr5RWmi?}6Hdkwfx
zF&t%a#wkI6^^B{rH}9YA;ykzTJ>WVwCH9pR1CG}hiE&!EP1Ty^!fq3&b9;e0h)1PZ
zsMrV7QKSi`3n&o#L(1(>U;eoP>JuTZ{f|sG&^i7e1hVdAz!o+K@v{45z&0{xt<c&{
z$9zcu$!~WDZEq7w^7{rjNBXGvwt=~LzQbb71};Z`>jaFN4tU>2lvHmVa!nyk8lpE)
zIlk>)DDbN`z61=~`00S((g3v#jcF8WI_p%saGaB3^HK(bcp6W$E=2^Aue5Gk1IE({
ztI+cO1~{mN53r<x6%Lm2r*f@rqqvmJJBk^_7{Cqa&(abt1pEE&fOtFcZiTnU4IB@J
zF7>x>h2jO`S~$c$43CzfS^MWPo9~PaA<h)o*Ped|o0kvsFS@+$G}dR*C|e+Rs?V=u
z&zF>Ir*?d3VvV$~P$pT;G}3^{*p*F#q6I^~<=A|W#x?oIIAE)x!yeaUtHBG4svKJt
zCi9!pwEqItiiaSV0dEcy0XCNri7ylaFgKTx{68vx+%^op&#&+?IYbdjJ%M0>%x*S7
zPHnokB&RkB3N(iX1zPmim$GG#ML);(v^~tuOZrHW6d&J?uD({fdV4xQonAhYsk{mn
zJXTL%D&Q|P{{_t-I$L=F<p`>$Z`FraTBGB~nIA@LB7S&&9r)Say&wp@0$%e2?p}BU
zc|9S2KY(A~r}yo;dbe%9#n&|g#a9_w^CPYFpebSUI*9zZzx(v`?xeH&FCT4~)7^Qa
zSi??t*3+`?mf<Z+wXBl~&9mim7}n3Qbn{qCS+b0?RyosIhWs74bO4L#h(5j2jUeM2
zK^-vIDyrzvnF82YWjzdl8Nst!*5uvZkM4GVTKQ;--OoO<4$qri4##&H<y<Ux{q=lS
zkk^VF4Fo|eX#8;AE4F~#g=Sv8;FZO4)8Ef$y%SN{h|-Aqv2#lDxJwSn<_s`~gFq^>
z0VK}3iJ^q&bc%2=?K@295B%<aHh$1LKGVR>@X*9I8zF^vdU`9~y?HwQev)Bmt*(QA
z{O50{^o1c%6xf)=;IaDtx%zVYQKI&*e|>!ISewF(pjmTv7m?yi+bTH`(OMba*8TbZ
zwGyAae|yoJ94iPC8Fg{BFL%PI5anVTsL^<aT+TS|dbI*6@B>Ml$wt-UxHL2{Pr9mg
zcFEZ!nX7(&$eG4=(J?SIB733BRZWC{W+cq-@0*kMb5<RXW7f!-QZb}fKo#)qYb;QF
z(zigX31@uJp>s?+Pi!({(!a}+w$*0BaZ$1JB-*6D-F>p>>wOowMujlYTo1zP8dTIq
z84D6M-FG&XhVgSFc!iwb<+*qGL-W^sl@IbqiMGNY54*Itf#pUhw8yM%z7@uQ^=?d@
zK7l1gF-XMa6Ht2#VIos+I+qU$3H#@&&yo8Io6l5x$Xw)GKXEp%oF$36;%sladUWlr
zK^>IWPRax`5!aZChg7gIr^%NsnM^C2{<P*SK6#&7x2VVJ4w9+yg`CEGqWK%ITxgt%
zLrv?JGPRob*Bi=4<!u}oYP>Xma0jC_3|zUs`D{v9e@zMleLq<`tmBm9sVf}|MM#^)
zvN4drEenC^B*crZshw(dY_3#4JWcP^dP8F8R?~o!@)uHYTgt_*?y`@yLcq_^dmeAK
zG}h{fRtrh-l+?CbEyAVbgj!tS*CUWz_9Mjv`QekUe&H|Q@#2){{;ZdOe(S{O>bL5r
z)XOaZ2fh;w3@K<~4p>(#o13c!_5xukD$OP0YKWDZrM$x(`YbF5UbGT-wqFH>DS>jD
zD!UDip(a?;^En4knk7U2EauPd^J@-Oo~4@CzWKb@?N6bdGXmuaxJ7aKJL(~ZUNBxP
zRfJsUgU!ZTa)MPs)2{G;xE5mNjbK>EgW`r8A+KH?FY^=uCqEv>E-m9@&WW$f2CL4c
zb<SKx4H%2uJKz>-U}6r=)IcA&8`CESZkxytD!23o?g<WptCwZxiW9Vziq7*K1P}Qb
zK+0U=AOg3_wzOI|<xdC-P9ge%AXwlAxTVt|54V9E0A+$3;2gn!4J4;Y;H}gE8CM7Q
zZKwf*QL+9vpaul;JT)K|-;f#*k`|}|sn>RDz(7#ehW`OIU^uXBJZcmbXltl}UD{K0
z%9(?dFTvJ~(VJ-vq4zPYMhUu-)DVZ>;flaTJ{-vk2*j1Ffazy!JRropQ67(G1qAB9
z!U_nP^Q?eW@K&sUfRM9+6%YsqumaLh^Q?djklV2W<ajJ~w`2u~yOf~qA}eqnumZg<
zdj~5Z<ga7}#IF6U0LGxA_D8Y;#FMw7<4FMmZkrA6nfk7VSg9FLeHiFAqyVuH>kCqV
zdSg6gG9cs;KJ*BJbABCrYufkr4vJ(q3hH~CP5%5m^19`J2K;(%V7twhe1Qt1*%c+%
zLae+O46CF4m$(6R98(a-xPkPagu0lACS#`e7p`*^M}Q_Mhjctg&<B3@#{x$HDv~R=
zeAH+cI0EDr66Fb2a|BKWW2MXsM-avbe$DAL*8e<D5m4{?VV+iqoam!C+~;-+RooDG
z?i$N2JdqH8>&7(#DF~W{Cl-#_dM@uZ7AYh-#~_<B<;8l-ue0O@^1bz3Fr)?M$qUpx
zPX>@yA0sceGX{}|n<k_5OBoQ4w889V3=r9Q7j~B5`qIyZG5z*LAnllk0Qux>X@LlC
zyoqQ}0=CX`ZR=B)l<QB!J%WQE(WdwgLIvmsvhyr|QMZM8>`%Y8nrPb6{HNm$NShLP
zp0g>}-jK2>1uZZ(LaXhB4MdiH`OVyaNZ3H&nDKZJ=ONJ65jJTSH5<mO^gD&<2PfFW
zd^D4-KopbhRfx^>D*+#7ITW!vUDa+vzCtu|Vew&Tp2@4n#YuSsa)-^`aO0+C4BQiN
zp35MA&`|+bqOAxxzfXIod0;RH(ozC(^N&?_o{Q~?(H`ZM5^aT7(hr2+Z00|Oyd4)7
zqE(RxYl%2<W7`x|u5NDZK7UY9MDAdkP@t~xOK*aySNH!heN8R`Wn7f<;c;KO&YYh_
zoggB4l?i@egB?EN#*n-}+D9eoIx2Tl`D|!^et8PLyV#Y0yQ*g{hyroqGPXfhuI?`5
zfcw4_wCjzR$y%w|L}B%)ETUXpr|b><dq|{?U9j_IH4lY@Snbt~P*(N>xOn2B#rlDQ
zS8(x^CI+6ar7O2{@sui)9=Kql-=|cY?I4tI$i)LVWiCw84Y+vFmaoSm74PS%_+d#T
z5Tw?o4!%%r>i;8czA6^C|2{|!YGd97!d^bVc{=?A60@l#mjQ1M69G1taBm9~0yZ?4
zVR;xQm)SoUp8}VUmwG@LlYbiKMUspNmoJJ>1`u(b%Dr>*EUkCt`KjDH9g_v>$K##T
z=6(U`!{~ee+ZSTz5usD}U7tG(90{wkx<L)kmDh$EY*a4Niwxvi6gzzaXeFBG_(bb`
z*<@^3i%QN|WylOs2@%Lbvlg1N(4s|5u?Qlz)7Dmua~l3kuE&;Qntwpw5)grmyHDT7
z_>U&p)rdw}1;ScjV>ZE_%kT90zTEfacY>q=iffz#FnKIA^}vDa;c&Tv#}$s;fGfer
zUHqIoT=I}s`}@M(Gj=WT6kAq`Yj>IbuhW4uhX_A6PR2SA<COSN>EO*@Z#xk0XF=F1
z>l%<BvAx?>La=`ku79Dvt`icgi2Z~R-QJmK-n{GS*;!yj01Jm`fb|BK9zG-UOwv3Z
z@8QoABkq;Qr*-j1qjd1|1n0d5W&{?cC77)-6#ISOG0@q=V2{a;OlGI`KpO?0LzR*3
zzqim(tazv}#!svisxpG|#Xh^$jpy_+B14=<7VdK0vL<ST870LqJZq7-WQaaea)X!S
znn5sPc<@v~R|WywRMB&&k|d*0LEUq=uR<7h1AlybpwmM{gr>>fN}{`pwnbBEd6ssV
zr(NV~qE&Io=;>7IB37}_7Hfl6FYLdQSiN}M7o~18zg<0zRi0~zboX2tK?65}xpo%@
z2_T|4su?4h-GxDrNV%w9Tux9X;znn!4UNsX>6)*XIBLH@E#^g$BP+yqF;Cd&#K*22
zZGXD!*+RG>KcQ5rl?hvDuY@P!;r=zepLw+(I+ZU8I%Q^O#uZbBNSMr}r5{SrQcd+n
zGm%DrAm%)_isFdS{B#ni?tGf@Fqit5p&~VDKe2z=J9;fK6vV5sqQSPv@5NDYmp@#u
zzeq@Zp{3XOz;zhmAHoxi@8`Wr>)dcRtADd|>zC;xy*v<kEq&D3cIe~uGJVwaUPB)Z
z#tnVct6xJOXp1PgN*`wKfTj3;T-J%@d=lGW+|~(?ju1x)r5yF-ByuY8;rz0_rasg7
z@m}{G+$`+3%lZx09X8nR>@?cdHEEKRpw(-4rC(G>4f(Xjx|n?f>PtV7!3%UmQGc1s
zoW^VEsK&ajBO}-AqxaE<){;jDbYQMsZBN(JC9e!h{{~YYLU~*)MFu_BP-0)EJ?x5g
zd~NO3Sa(E}4+-}(G$=|eO?L4DZz7X=tp?!>(#D|d2IGQ}^&z5OJ*16FU|p$FQgGei
zohDDN7HwqZqijfG{6-^<5jCr&D1RaupH?>g=?$~G1G;Y?qn>MMQ;TM;pI|;;+StO&
zQcIkUvDr+FBu^8jDZ}f{F1_#2VJe&XVm-Y=U=9-_4SUbux~5{hMxUzG+B!9Yn&Q#%
z?|N5|Eo8lppAZPxQlv43_4=c95%IF6Jf0EeDChaSs)?9;2eNC*lbBJny?^GACs{-s
z$FD!+Nolgc+n3DnU5#QVXOiA;yFX{jlhPTNzEsn;<iS`jSMG87EzMvimo?||8TJig
zK7UUevSN7N`V8=<7OzFI({O>7On9|P%=MEm(iWyTH#^y>$JmevDM|b?>>$V6$=y#w
zRG(5ZX}pFel!mC@8PfBkY=4tm77m<{-K`ES7kRHz<~KOM8<g}dO|9p5bfej{d<w{}
zo{)GgDH#u=+7?p7(te>G`)U3w<;KA)k4Ft-GqR8}E6CmC%khrxiBEKSp*5CqnQ`gl
z+$S2}uPt*T)Xa&*GLzEbn3*cEB^l04>&{IrqHyBTE6a8x3j;yjQGaeb9@HNJk4A|C
zwl4+4d90UqtY&nWTpbV3tlP2GY7}zv?;ti$kWIXOb_3O1)1W2SX}6q*{|WH|JaXWc
z)y5OpUTUAIk*tlB_ik|P-?s0KAn8C{w|vz)^fMz{{a9;n2bt5y@hFyz$FKY9YR6dC
zMP*BJXelg?f_JM{M-9R2#;SZGXa66qORgiA0dEcy0XCOmuO|}%GccFo=NBiJdrBBC
zCw4dIDIkZ@%=D1;W<4pr`@GY;?{4-VZ{EEk?Ys&WyxD#Hv;+P@m(WTW9Djcf{B9rb
z5dyD(Cw{@>y(h@)(eMlS^Rs*}>(XvJ{A!*O1dI<EvgVhp^q^5-+6<Q-fByLWjn3q2
z6`g#PxAxo1$D9A$w9HsXiYV;<_T@%?AOwZM>;SB?&hLKx$L`b3Us|wpSp{HOY3Ho1
zWCfZ9NIU1ELiuTs6%(y3bALxHBVrlR>p;CC-ZuXE_4BTI^5MIQaCA%|w2+Oi<S?!#
zQX#gBaX|(@*o%O(dfe$`1ytY{m{{$rWCgDX7vd4cTPDs5UC)EK();^?PAd<d6U2p}
zZBWSCDS~!+hd?s<`}4qQdJe;H!nIwK#&ebc+kjSpJ5V}Hpn5WwK!2fxgf~AbM)-*g
zl?M6~75tfOp=LvIQL*w`Xq)uDKHfVm$18Dl0rhNN<ioY7ub~}5y;$MP>DCw7C56!9
zTR4YPAX*g(7ZBgC^2_`vh`wDdR>s1*NFRnB5yVO#Cd)sc3Uad(6Ss9@kxxiyAs*I=
zc#EX}eR_JTZj27fYXsNgg>{#1Oc)&jHkXV{7(@Zfm(ffZ69F@q<xChlfAw0zXKYm7
z#+L4Z+ltezl+k#jtT8Qwl=w<Qe1@wCR`IKaxRH~Y5Y<whl8+Z*ryx)TnjzCK$&{$r
zmK5%JNF^sFJVMgr$y}f(<xE&`VOP08A5;hn`zbm&1uy5Q7-3<FFK2sMl$>J0$GVdx
z@ho<sJF(r34A=(AfoSCTe?Tss5|e)RyTIVJ^2W|F^z)FMd*}R7AWHyD*cX_Sc~q(Q
zTFC|c4zu_5nhu-_#<w+{LB*wODd`G?Ns%F;3;cNnB(gDBf%xOkdiQt!?R#D;>C}JI
z$}b}pz5CbhuTnA|fE(BY2wYO&Ob@W$aYn>7+yHh0VKFLA@pKSJ4N}J9G8+3}*)XDQ
z;dYlQP8b@0F5jQ?Z?ZpTvzeHxjAInn*?b+6m*$X67iscT5J%cBgJlt&W4!GlNvgGB
z7nF_O;aYcy0+$&ymGrq_u}@P8-^2w`+o7HlIpT6`tOumWMEj}KVWSE{z9NwVvV|o_
zym<Oy48KcEhA}u=U>d{kV4_FzGX?t%zCuuuld1-PTXLPLALiv7mCl&W{95~HlAp|O
zRI(YGJNPgqUFlh&D!JbRcWRD|@Nrh+(+rlMxSR(M8h1(nu+QVSiDEYdgyQ%I=;N=8
zkdUolT{mP*N&#wVnb+5@?U3=Qxzxy5I+0|ghbTIoWRUEMiu~tNXX?tDm}P1o2TAI&
zCGhT2m3?qz6EbW>^R$kGk@Ku+_B%F*=i5Ku<t<#MsOT}&nIIRxz|lt7)&t$Q|IRzS
zUqynPJ8EAe`L-ajt6K;aC+qc=Pkk)Ju1?!}%TFwXyRMg!PZ$(`7k~RlAG>$TlUZPg
zKAYl>Lc^vhh-h^Sr(8ZZTRxO6VO3+1;e?P(&c&hsbBkSekUbkKy2kRSt|SpmsxfLw
zc1?-RR*}V;u%9$w9})XPK&TP@-jx8FsL(fvM%B30Ehv$KTH3VkbHu^Ydf4V@)eKNf
z7Z#>#?A^Wb{`QxDzBn9bEwPdNAANkzu{i1-hmb^;d9x*-lCOPSf!&1hE4+2>kFDHN
zW99vmjL~z2N~<LP-X)Q=JcJWT<l;e*_ZuXxBVyM{Hq=`_-yM@;V5F3NEmMqpB%gz1
z9s1>$6c*|PV75N51NShiz<dre<Vwg)P*UK|4MMVW9KKY4TX9@FPXuN`DU>18TFE}^
z$Vd>hApH*$AJIOH5HdEY$9(CI<b)vcljkSWcR@=JSsMG2dOY^+hmS7QlIFm6&%WdK
zv2O|N#sV8_wG;08(I2@8q*kp1;kFPyDy!rw%t8_I`mAn?ISO3wus^Cu{H<@S5!t`R
zdduuotRu32+OjKydRIRRu-28=HXY!AbwaT&J<9^u7+dMtFv~5)*q`b~K{-n7k&Lan
z_fsCVd6Zz~DD{j2JftDw9NXb^1iV$qSyi>`w{N=UC&Nu_F+m><C-?}i-qN`94P-H-
ziC@cB-`tHR6?9mL{%P_#wC^XXd|<>!H+TULWS+=>VXG16w?kBgcLx$fR1RvA511tv
zbc#*9VRFjoi_<Q4noj$e%Ud&kJMC;EM-`_VmGsXxaz!1P_(qxs&Nh-!6dI9y?I&KG
z*+39PmeXviQ_poNvj0;+{sLQ(o-Arur*k_?@r~||$x=`m-<a2Mz)!yFAq9l=qnfrL
zV{lx59bS6V18rnD^aeLQ8p!>d9?jx6yy?+kI&jmYrR)5g9%Aq=*#Co@9*ACf<F0g9
zqJj42H$BjBJ?*Y_Po;r&%uSE8Z+Wzo!)3E?j+|tng_`=z(vuDw_Y{T+Y|Hlr`T(0E
z_;u%~4b<c3s2AMyXmD;_BN%Rav;@9hdoY`Shu#gIQwFodVRuRCC4<>KEN%t0F_?A4
z3I9Tshda^dja#)-*7^fVJP9I0VxcY|n)4w_b?@y0qFWycn^g6*hW32fX_b7Nu6Zx3
zfq3M-tn3A5$Hs3)PQ$5Ah;KL`Hk_qJoG!L(4SszV{}c?iR_?Dc4>l%pF-zB+2OAE5
zdvZ#yJ`fgXgDze&Mm!nBk(8lT1aC7BmMdj0R@p<E^+J9x^Imxm`Eywl^7FC=r{5<7
z8>4xd8kbkdFI%R_vng}PpZ9s+Jl{GnWj<S2hjhU8p@EC6<5Unw9tZ|&jJev7x!<dW
zC|Rd_({ZrnTm~1I-e?>gMN}=a`9zX`0CKaKQrlL%aq^Tjg#|)aciXNd3}Z=I^oN9D
z8k(*o97L&XLyLXj*v>hoVV*lk$Qr;sOqd5LCfVga@ANIj9Y3|}{lhdjZ$fy{#H{J=
z4mRD(u60W1z32SIj?!6}J{Qz5U7ab>c^U|ew+90APZ|32yck}YX3uU0q$hfRYwGQe
zA21Ako&IeRkhVaWED8tNOGYFAA2eVUj7QxP0m=2PUmM{VZ*t#RU|VSC_>aBb6<E$-
zLk0f!m;N}91BY@@lXRd%1~N0$aSP3AXJ2~Yfy-~ep}Tg(T><csD_2*e8Se^&^=3BL
zE1cH>1*UKsR19<U$%&;1&#ORxd6Ro}KlE=pXpVtE$=jCuV7LG55A-#YPo|MlztPN<
zriKkwtkcESz1NLc571?}RM;YG<C+I2ph3)vt=F4cck7BUM*hnQ7m65r^l?WVx&dqV
zjiQ*Cc17^>@kN!q7-TAqsM#}+<rO)OP$iaQYQdFAo+7r~om#_PU~=t$fyXuZi4<XI
zPxIVM4yQSL$#S+=>XdR<DX&<cUR@(XA{)w*<TuI|q$wT{TWn>nOFI2Cr7PFOM)upp
zfaSq7vtq5dVD?v$4LGsB`pc{eCP{CR*7&t%>dDee!M)apo~a?P8f76XU-IR$@%wu9
zTwFh&Sc}uW`(<cbUgm0lFSo_@a8}wfr_LMLoqMyWI=Rx3Qn@>BIHuo~&fI#-tCZuB
z&m)9NBsmADS`+5zPFQL7)_~K!05cD&^M7(Wd%jR3f%l7lz7w-7f-yF?ZRSrpUBS{y
zu90A@J2%;aj%pI-esB$N)|SC#*Tf8dpZSf`6b|9+K~W8;KPRd@*UpC}^fVoorU%nf
z?(v%AuG|NkfNpW%Kt=6m3*TVMRsOTfAN;qOJU*CFbNp;;<M=o4>|Oo`U2^BQmjQ1M
z69YCjHJ1hI9y*t*Rv4oJ{kKk67-R!~)OIk)QCp4K4fAld!qz83d8`wxf*VZaFo}Ju
zB9P_YUb8Ct<ri&6yOU$uTZFWk;+Qf>E`^;W2D#GH!LJ9YV&SXQg)-tv$cE(kCC87_
zbLi0~7jf#h%63Xg#>cW<DATFuOw;nFIyF2AD9hu{`0R_PvOGP|V<k^&Y0_PPeACJF
zEM{a%qwyRiZB+y*Jekretio6*3}wfJm}a?E1By%ldq9N0d<^xq!o3)9+{af0QDpE2
zp`l4O0n;7Uf33L>|GZyQHn?II7SbtO`4YaK$UY{fGMlQfui7asPcIdlR?V@g1hH<&
z4?kosVWft6E1hn8AoSW`5(dE3?h#B$f5pg81Pf(dVAx<!=O^tR8!I@E^VrRFvAs(@
zvorQrSwyl`jL2K`RjeCJnA;IA);)_I7339g_oWn_V=FV1R-iIHr)0pknB+<4V+kXw
z?8j9t3uOTv(8VLoalX}Zs20t2F-!{#Vi{2iJRg2nde~XMv<8fzmapI6uZjZLf2fex
zH(fs8udM`O;UK+lyZmW=6epSBqU|E?AB44Adh<vaJk9HEelQ6uK)%xs_V-nhBuVJl
zv|awQ$`}lWk?SV(ZG990NDOM7;vTFU$C{T&Q7m*d9skUQUbTkkv5<>2z4$tZ+At?|
z+vVGJ-i&0*c5A_{k75wxNi1l4e|{7T#&mB#YmQJl1KieG>AEKjQ>lW&4%iy-gzKgM
zWmS5`B}C~M`{sU6)D8g!wlHLs2p2-rV;uv-6EPD&^3&{N(;Whk7^0y}_oCdJ_A6A(
zPHh=42ck?zd`h*8Sa_8KbyvmLz{R=r))m@FZL5-P>0Nyt1II}_5us1Re=H$+f`ja|
zvyQfwa8?S>1JAMFQ&Cc1#g(@GLX*U7zS1m)P^6NCK;1DN08NDtio(hS$x=}s)U?gP
z#E6i*VsX@kwg8lba2bK5{~D8uPv#(5OuNIR3IA9ZWmYGQoI@HedQQH^goS&RD{bZT
z+WJIEDH!LjXLd4Co%)9Lf5-E*y;fg)T~XK(B`16(y?RWa*qcube*DxBQb|CE$3Yo|
zqq6qJfZwb2uPO>N9~&n|6g~rt$S-m$0*bQr%6A1<RV#kgFf>PC@ICb!;UCF$oMwir
z39lycGE|Uko~?p5qYMcN4!H1Ap+z_SC{cUJ+(VD`=rmJLp2m7yf6N@$JUy~*Ek;l8
zk`=_UI*S5UeNZM?MSN1M`aJsaD3x+$rZ3f7nyWyU)<X`Zif!lTfo87BM|*XBo)%jA
zMQh%bp$ad6;>M3^rgejZqj0NP8zDo{x~2BvydG>dk{SSnjsf*N8U=90Ze>>CYXMFJ
zNJ^lNhQ%7(%_j%If0ASjdLh;EIXj?r5RR+};A3&11@?eFJ=Www_K@ya*>We=$AQ|N
zX@~jd{sA(kA3I>Ar9{Ss&u9mQvo03&7kka;BrcQ!T`S#s#w>jJQbWB_1&^+{YAP@V
zS@|9*8NRimmO#A+sPAU*w?Ay4dVo_mnRULAK0uvPxFf4Ae`L2#ny<Fn6wHi-?Ub>q
z3#DPpqz~#sg~VFxgX>0F!NlUSo*xLQk?k~yoGq@L4Z#o~trkZG^T=X79TW_v$|OJJ
z5kke?S}`lw;1cS)Z`04Z&^mZQf7G&vsN^^^QtosVL`Wo&9SZ9iNw>#Qgb$?D3{%_$
zsl1CPhQ0zLe|=LzQ^UT9MM03R`(Q1#AREpg?5U7p&9uUAsmw`cdDm+r^u`3-^x_Z5
zH458c=J*7PNVqerVA$G_YkR_#3S{G9Q{SNuI8kPWV#_$YT~I}RMA-u;G-BAWJ!1GO
zd~WNM1=UAH?@gOv_1rc(8jPV%8mVbf=PrrTac4TLe+tCjVCOrWDN)$V*7n#8(*&?%
zm0j#L?{Y{2Ljz_DV;PLG^`#32C^h4v9jL57A*(Q5l7KKZImJ2G2~R082Sdy6CmU5}
zlB4pZO3sFTJ0*eSLz-3KIr>zeQfymt$!#ZNjLE_iMbqjNhLi^FO&F+27;*|VJl~Tr
z&~o~_e_WnlsOIt%dV{N{z2+jxq|h7oZ;H-b7AmY^3u6ds^+2WGPV>M@s+i)|UCM$D
z+S`y!10Fcr4*1&!@`M`}NG~XC+>><{Ou4p48(+S-XM$uyVG}<Jjx}f6KHK;aE0&aN
zd!08*<G>Zp^#cjJQ5xsW^HwRjG=BKd0050te`0cI0N}-Ncm1J(ypbprF@&!wzSr)!
z>IkiGxM~|;u~(m@0L(TWAu?C*RwWf6f$j<F8wjQyU>!lbx&f?fe1%>NvT1;#_Y)&;
z<mw2@8wfhMzYRs#7>}%f2fg^PKhdqv#<A^Za3%g76<v0yty*+<u%vB!V9)%-3j_pS
zf1R?rTtB_V9ds1?^A5%rwqOfs%W?f=$JW_y1i4(e*xovuS6@p}7c;wo*SsmW*M=Vp
zw}t1p2W#hcSr<;=OlHmGt%_e^GicsZ-Z9Ya1wCSjVP^sI@eM;9TYF};?^|EBKQryP
z>?K)Pb{09k5brUrqqwtW=eS{=yNSO0f1U1SF(U<+R@A#wy2Rjkv^4-7)Kgvx@=7u1
z`^Iu^BmHwuwM{&4d6=yd9omtWwj`HT)A6-&5TH7zF;O(ovLgBaJaAb!Q4}+h8&CKZ
z$kI~#-<|L)+yj;?|1PIViYIR=zQb{ntK>o~)a|*;TTUL8SkhA6GdovouFvSEV;w6b
zbo|rxPClcSPA!)kjzXrxl|7pa2ch!8$$orw^C_5eD66A<;HY5<s4aoz57TMhtCMjP
z+XI}`)j7J`k301vc$i~#o@{f$W~aDx9;q2PZLoOVIQr24B7OcZl#LM6mjQ1M69P6j
zm*Fx76PGSy7&`_N1MZ%&ftPP%7#@Fat3^uUOXSCH-F@kF_i}f5z5D$$VCPjJ<8Al)
zxikEM=D$Ysprg(kqa2v-^=tRtd#$bG=fDqY4Zx4*_kmyO@c~TW8Rm%}^Z4M=<oRIv
zG5r20|Cf1*n)*-eQ{n*eedetBk(u6T5T2Czy2l@0zq!+S`-Wf5<n`n0-M@eD+5<Rp
z3N&_q`+6rIAQ%O~b`}&Lt^4_>?(^NB2M&d^RzcqJ-<Ktp-l<@{u>9@^-TlZPe#4U?
zhVDOU>epcz-Tl4$3omeraZoxj1{QYQVyru>Vu+dT$buj$jq!38XYR~d<-#^sqgAAq
zZ8EW7*u=DRM3srLX;}j1O)7sf^K4?~AO^$_EIckhPx*=b95ao)4`M96kxb508O9)~
zYman7QWTYoW|s~xWpSq2GFy!H8nbQvBWaaF7ZgSBU@34o_ORgn5I9&{$l4)xclPG~
zJqHise~>kZu6KJGM-BmG`rwh~8yQbpJr%B;CWT=LR}!A^Fi>l@w;+GVQz%jbHol)l
zUbE3pFU4x-D_LY~#o0<ahq7aDy*D~<5%}7<t#!&nyBS%4Ql$0rAC{Pl^#XZRMlcTl
zjUp>lF;mfF;wQ&3AF?`}?IO#=vP7oaGQfkbBj*(PeqZmtz0V}PUllzBs!w7Sa!@6X
z2HLxx<jeg7(7T`RAIyKde`abnsig(MD?(c@_@}ZUEs`KREZcBJ@_Bm|0;M^)RwUoE
zK<+n%!8cGeSX2G-M?REu?|$80&})w7Fk8@UpUit@J=IjUEQE_B!ErEb_V{UAVNo2M
zg&})n(nRu)&EtT!tlN5#KWxel*sq;#_Vw%L>A^U;Yg(CE+TMQ}I}(M)fVTAt94yH-
z%E@Z|WqXTg70`T(Sg5e$AQKE!z79m`Y9|+!iBJpN`9%_^l}=7IA^D#Dx5y$cQ*ku0
zP*ZK295`T;6Jr2nvMu~8QB-#?Cv+@jlcvFTmO`Qr8g8O=<E-eRFkP|{AC_zv1`nA8
z8?&wLV3bF^XqA896+)<2vD<QFry>_x%h0AbtMbnG@3!`f+R1gX;yH^RjmB~iZKN_7
zm|G^{BQ`;MYF~?TVcX*T#;-_!Q7&kMGObg8x$iIa=u+0hFE7o1ixmNqQE^rfu;3l%
zUb!G>n4Xi&a_T^A9>5M|W4Vdtu1xo<&sY6(PDD_5<DY+&gj6G+UdC+$7cgSpHD{}v
zS}4y(RGb$!lVP?E9ymyml}j{t`L+Ukx}UAYrQhPVk}J%_BN-DxPCAY`Q<B#YIb)K6
zW8ydyG#gPI(o1q<iiI}+f9wyM#!dg39GaGdam)hXII5L_&M2J)P>oClP&UH34j+5a
ziySBBq^^G?d;A`$>}Ex89BXmLZFRrTj}J~ug-xQykC&@2a$O%nyPm;ncb9SGG7e$J
z@x#>|`OJL(_!&pPC_$@Hp#;2oHj93jb|qyYdn@hM+|JOh-?e*WE$fKW-A`)(@sgn^
z!R&g;Q<hIboQ!dG)NnoH6tYg}XQkPP$*C@|2Wx*6x;Q(Po<+lpxN=CcoB%ak8&4Q#
z2d=dIOa?o26y#&NwT^18r*#A$`*qAm8L>ICV3w=wT`Q75<y_TBDwv~Qx)}$p)jR6X
z(G<DMck7^`n!{LfUv5drX$fEL4ab$s%aQ<9GezyL@I@^bZyhXWF*E&|<+uaD@g~e{
zOdo$&uT_+sKj&l_8`!#_Q+VCJQ)eqsP}Yn$caEk)Au*yCONo3{6q(Ju{MPfx{%G1g
ztR*czaI!+}>{$d`N)^w-rX@$W;BJXcqX0q_Z;wq&(k_|z4qWqM7H95^%NA$s+kn$(
zi1X<<)qMh@GeBL(&*2lwP&BMOn4bb^%d3CSFm~wQkDPaL4SDQ%Q;sm*qrDYJp^cBG
zP1;W2_}8odgjwmRoP}Kme<_PIeVo~%#oq|X2M;Rfv1FrsE7(3pTR&cXR{-_`YN3JO
z0I+XYY5@E5&()XBg!vXKmB2-SeY2VWJz$T>*M{qW!yS5C1NLiRxJxKbMREdicb<PQ
zQm9EzK<?gh4Tv?#X~^9R1{`au0=fTnd%*~*-QsD)Jx1kuc15*~xC1+}38p6bw2dbk
zr@)JnV!}py-0XLon4aL;=%A+gaU1>xE?R`ra!_c`vJ9SsSKc*BIsxn>D3)5Ip%bV%
zM^iU5g#k4;E66mbzM&$L<o&t#eiwfqHxI`7qHd^9Tc4oIAvV^qZFLY>cQ&k=-HN6n
zI|ouym30+RU&HQPtVX^afJ-s3^W+fXaB+ZgGt_;#8&P-~&+=^yYYp7?*?uR7yb{DU
zkv}!irp`ulg0I7F7}o~mWy~AExDDrR7`K`HhA?ik(g_&1-PiRnuI20^YWshGfa6+c
z6MlIcq_5%j=18Al6fxkfc3*Ql13lV@LO1K(0EKQj?nj|pj<ktFw-UTOU|6neIBeZc
zG}cJ=MYj}9hW5@BjV0yn%(!*}^cMLOTfwKvOvTW1_(3-bO>y1$2~rK!a6H5Kb#}DF
zv?Si^(42+I*BEXK3pZS^f`xzMDAjcN9D1Pc7Em!-<<YdeyaX!dx{CBndV8Q^7dW>H
zEyp#en9CKicDDZR;_^DISi-rU?PkS4LF}Jvz}7~N`|Q}r9>T5dxaNA!1$DXROW?29
zO5g6&tC`ynWoxP{F2^CdHtW-lc+FWk*Oo2EwN_W(c4=o{agyTKo9lmG#IbM5v<%%!
z|0V~8diLbvnHjAPoCn1>?1Hdiv2k6|1S!#h#za_(+2S`ii5fO@#N4TpJF)k(D{Qg1
zoO;`9K`<LmJsBr4FP&|FJEuh3^LFK(@?3q_lczUW73r>S>Tp4>td6&A>KMnF6WH4~
zb&Qj5eBmC9`(hSn?u>s-!mHk-sbh%CS6_V%BQ+mN1H<KC6O?8}oQ(PW%LAXs4*3@d
z>9f?lm~=&!m_mMi%7kIU5GKEMq5e6y4RXD}PACGUH{f=e$fYdK^a^GR(ci2MvfZdK
z#&Kae0*J?i!S`_2O5;IVhaOtWoQbTiCrwftTB@N#;BC1D(nWvt_+Z|zfB!&C5rV6_
z1SywhKJvq=#6vE(P)3BR|F*&~ClpxbW-Xai0m57|ds;HS20GICQ{PN{KD#?3Ut<<0
zS`z*wSuK~WTxs%KHwtmuyQ)-$u2<10)HF)DVbhCP4-Ffe=7wQjFmS-Emg5uy8x%%k
zIfgh>*UXHyV7Y(ew=b&ZkuRoY^PZo_a*t?ePx+kXgwnK1Z75R|H66j6%%GpPtAu}U
zRb?rTpN}J1lI681JEaUu4YKu8qT1)MemYB-(1uC_Tdpm3BAH|(%@eN82BH8=OI0`j
zM%P|)>#P?yWI3&V1)v?j77I_c*Q24PKj_|1(%If;cou&)rh#c*qRRrSFqNzTi8d_w
zH;Ne$l(*0vkQUv3#ETP#JOCEwb!a0mWbD#Hwn(rbOi2QRsTDcZlB0>4mRvU}R}zB~
zzn_X3OsUdL=%3<(;|lA?9(S3|B@QDqol?QBCLHN7>>GR2%=*Vglk=2*dNnO&CZJn3
zwBoN>P}wflXRYh0Sq=pQ&F14Y&e#vVsM7TT!`#7>CveKKAGc-P&8mg+$~o450Xsih
zQkMa54if=3mk}xk6qinJ7|a3gw;OL5a|0OmDy1NJtZT(tP?bMxppc*23>fmWOW$sn
zrE(aZf8yGl_wnyuEq1h=StMgLYQ^+SYAG|E0)O|HH+>gcHRig+rnH<lUH+WA(45Y7
z(N=(KdwgCNLd6akXuJGGBYLwq8mf9|(%N3X{prnHu=cy()2I9H@5_QcQZ4Hzf?;`p
z-qVzawigybdbUgO0xj$Ev2>w=m1+;=n>4N@f5l;RW=bwj>n!CDMKd+q2dEHChOGMh
zQpAvW$~~a4?Gl8gOrm26YxrPht)|e@_^}<#9~b9JDWv?woZ`1dD=S*Q9GCJFm3+&B
zsE9zzX7u;m1?B}Ao0`^L9>=5Rys^4!jCc%g^;O@!BGW0VA3c{DQ&UV;HZ05ZlvbSG
ze;6++#6_{~v>YBP&`r<eyjv8aWlX_t-v0AaG{jBzH2UFYAZnl@6j)S!c{5Ym-X@cf
zY-;)nuOXAj*9|3^qhykns^K`!Ba;~NGMU60-;qpWl5UVmT(A9PlA@ihR{sBhOrpsV
z+J%$o7>0ItGKtDFa{RNDZ@^aQ?vLqie}XU;&QiWbyE_ra<efl-am3?97)M+t!ng%@
z6Jh6>rkrA)XqtSQxzaS52xG0QBfDLWuRTADK1e7r#BW3>Ddp<p%9GC68R{-d$v6Mq
zkvz&ryiDgU?2+}UGG+(|=%YetF~n;}limorCUbdcH@=TCit|5Jdp=N=g;XDMRA)Ix
z8pSPCbv}ERL$|1RC5@P@m822tb$~SD@{f~79C4X6;;r|SM$2+{kVYJEnKWXpkFG!v
z*EMstW4EYl3dsGA%|<YF>OF-^3!2Wy<C(c%E|-#Z7@dDZy)a%?<@@pS6}MWR`9h&`
zesvDC@TrBsRTbyz^7rS8xb^G#)%`}hH5r;}26xC!wnD6o3>^d5pT@5)v!F=+qtsaX
zUlPsNA9G)!H7Vzz;vENG;)*y#ff{(P!a-sf-nChLGq(3c47TAQ)26dvTyo9f%bnP(
zf4^^N<I8_Z>}gzYbW;amVwJsQ@9evYDHLap2(KAE>O5!gs`4%6LR_2U4{J2&nhbBI
zXKF~@bI`@Ca-e`QJxxz!`UdJnXM{_A)ZHRgGR7$?=q*9d{TM<J$~6b=$<VLkIfxm4
zMifVzp8A>sPa0vbRi?R1K@~4olivmhtZcpU`FMXPTd;hV2~IW79Nm6>KT`CZk?*U%
zzOnvN6f5nhn77<+H^NvMEdr%J)~kJ$pbg`vx9~V0%DGCL%w&97Y<ykfxs7rShP`cI
zAFEwOA1B$3>$eniYM1(DgEdEwc>-c~A;aG(M??ylGG{x$ZLfi^d*6fVTn9KO2#W>F
zL2Q4rz~E}dPOark<))=rPm>$#fIPwGxwt{yE45VpRYGu-u<-Wvpk;QSFx;8UsVPUi
z6{2|=^wzy==K0t2E@wk}t)$*jR)^Es(@MZ)OJ!%Dya`kL<*?;3llHn!H2A)^Ra!69
z&~^yr>ag+B-^X%)45ZB@v(-YO=2boIA1r@`_)&@(D8VRDcc7RV@-oHD8sCv(W|D4D
z%v`Vi6mv35<-FgHVg@u+rkaz8PKI`OiWxxD%-FNEZ_(~fGJ{k?yR&{P%FymkGBbH6
zkjxzMILXWrmq})B!QCYDd8R3>$BCxN?8i#e<a1=!`s80*GW3&wZONx~f4~M52K;}C
z+;|K4ji_;|IE@J>^DQmt3B>p}{^eypv~GvBfjcihlxzWx{PGfLJBPpkul@28l`zim
zsXlbFG1`6yfgWG-)G~1Gul?xK3u`Anb&NlE)Z}EJuOaWsoY#U@(&lxS1Dtsy`8Z+T
zz%8@oP1F5U`GUk9+**USOp@148-@*hZm-5aX)E|&K=x`FmjQ1M69G4uG5IPKmtT1p
zEPo`o4bJyhba)#aP4G?Eda7o6Z(XHwV(9`^Ns8=~6JKB@{`&?=kObM?GqVo|J&$e@
z009vA0JdEIwaDeK50|eGe|pho@gi98YWe#60{B4UCrG^M?BW50)3AK~zWnkerE+*(
z@G??rjhDyY0l)3(vxa~p;EWfXK6?W+UVkgR0DrIPe;t=++wxSMGX))=h_b{hjr1UO
zqG^e4IQ{nZhX+aGOA%fAnZD+KU%o#4=V4xqGa$gh@{jKi>4DK8bWn>9xPG(z@ooA2
z;qNom#>;uU2YLY^<?=@wFx>f()ENNPDV?)+P$$UD!?`GQdPM5Pl^Rr$MI~5Nk$+9U
zQKc3aY)}zqrTczZc%y2)a8}P$B^n3`WU2<K88nc2!5c59&rT+@SW=`gNT_|}MGbd;
z?7Wbcfao_Xkm;Jt@(QY;lAb4`ppy@6?A<0QE4<n$Q?7aRM&&jLjXY>nR<$l_&gzx)
z{ah3!%@qY6a4=pN?Hh+k#(PPMCV#4z!cnf-46ju_SdD84Gp7SRy^6BOJK$!@@_FCk
zd=OFDd30WB($%@FRK0Z9&AKzir4?x<jm!6Kb@kV~qXn@Q1Ov~m!&ob<+!(5wHXG)$
zXSmmk6}9lu+$&cMLYlZyKFXpm&D~zZUxQ7-XUiWi^=2p03gf4~)BD;aBY%aBn^}1W
z)P?OuTEK5$-@B0<2@4y{9LW`GAD~9^N}(?~WOf06?<5B>0)Hoxzx*bbf8xU*aB$jo
z|II_cH#Tzl_wrLp6*mZ@bu(dHA;fEh<$_@_L?YWsSRIRlyq3g$lGzxO?ypi!ob62Z
zYNksY6U{EQP$qDn215>m`+p5qO6e=5cTM-xRmW;J*3$iWN(ro_GWB#oG!LY!5~|-T
zV_?g?=8s)jgV$9vyjfqj%-}>s`1KA4c{Pdq_CzA<uaiC66Z7WFE@Q(IFerGZ_pm0S
z9|5XGii8YvmcCG!P8ib!LX{OV<tMH_$(bhI^=)iCnfO!!dVb<q?gVeJE8CZdd>AEv
zXf=gKy1Q0h>bcCzcv+WS(Roz~^6xK=mm*$`W|x+B0Y26tnc2GF63&fMl?1W-geMB=
z1Xwj|Dv6dYL!30jV5&jfsxSPmiP&YSXtj+BP<qa(dN=nNltaPBVw8rnA4Xx6l5W?|
z2uhVRZPvL)nA2zcl1UWT!DCuT1~s97x}0^i6oQroXMjF`A>*%gxow6fgA<6V_&?K^
z3{0JyqbK?oHmXT+w>zVjF){0Ar0NFi;F`X=OLxn>rlS1n^k<T!&8rZvIlE;pFt~K4
z?uIqK*9{e#Js_x^R@h*v-<K&F**qUQmGT&CF@R1~E4F)&yuo04R~vb&&d;QOnF_N_
z)t?`?FApbmo7<Cur+F()wld@|Iiqj<MF(6r0=dn)2W=gBn~_+oc^GGdzNofqb~iA1
zEb+pDt~Z|Sm9PMZo*djgo*c4fhB>I2dY^QfWq#>_bxb9VR~i}H+;A$EI^M6Fd9K<~
z_L_uA$x}&ML#Lp-&3B_oUt1i1M_QSl*x93yA6biv;Ou!^+V^_D(MD($XDc2>(Tm!N
zqW7&`Zllax;8u@9C}PKsN{abuN9IJk9{ZwE!hKLwMT-FKR!ra-Y;@q@dF%{4y=M<P
zhz;55%#Uk)88QsioYa!=mlL=u$vGo%L#*L>ua>*ri_<pX(@h2EOTxZ?(TkI*Mz!gq
z_Tq*+wHFuVwO*W2I9N9uALSupdRXshhl0aQZG>i|qT6EMlv!dd_N|%4F-QlE_sNar
zxkGs;if?Y+ynk(2RR{q8foq#6YR!D!H=Oh#vImdAH-4n-BSmd_anmC9hso~|Fn^@<
zoL-KcF1PeQOkQUNqP&lPlz615eGhS!(rz%iRn%&gdXIin3SF4gfwHe@WnwMRN_7xx
zcBprHF1icpYy5^CeDwZ;4^k^0^vuCtS)~Czt+L$NWN*t5lH5n+5OIBB=3a9xaWb4|
zrfA5nuA--&d1+)Q`EDPvrBSN#fi{PujWqVsR>s=_)jP{Uz-Oa>Tm94<1_iN)Hvc^Q
zBV3fPFPw)}T$eR+H-tJNRRd<!&Y}E<mFQOOq8=43Myr0v;;Cv;$zun^I%8ki>%=T!
zm3pl+8xHj4cW!c6%i1QT+0n*VZJWA>Hb}@cod@y&(mxwtaa60eGyXjDhLfj;Oq1^1
zEt6}3BhDM`jB9LvTvyI%Z%^@qt0Z4Kc(vI#H@LIVxW315l<RzV?HsgFvPTlgF(}hZ
zt&yg8dn#zt*QQj})!2F}emVSe9Diny;XBo`UQ0*PiZ|bk?VeP3XsvTM|MaYrlZK~B
z$A;Az;8}fL&xtlQ;89;=+i+tQ``xup*UtU5PL;gH78VJAFKmshDN4Y;pH-!kC9^1F
zMT}+}=_wO=zoaI4?w_$9LPEz%4YNPV$1=sK{VB*$>T$}SAFp%qxz{jEQT93;jdS$G
z=5I6w?Cum}uu#>rDpr5G%E*?k;=t(V$7}gpg&?$yi^AeaT3#3fh0E5_Jdji(fu#0<
zK*sp09EIqATXu8#5RVb~^eY<uhDxO{n3nmZpRXHEM+8Y5Ht_gadH?Az*ST3IoP%NR
zw@2=}8nvsy36O3S#vPc~ihwX@nDQ&)t?NrbF*L++mVCSio&lOUs0Q^9z`!*ffHuNl
z5yvStCq7_qwD4BV%zF$z=q>JM&CM~z4wfiS)C;?R`@%X5>u!85M>*G)u0j+VO)@O~
z+x2Y~(W&9a(lgkG2ESRmub=mf4VQMh0<1fdfR-}Hr$w^eX`qkzvzHTUD$5DS@pB#y
z2B(1Cyvt+7(hCrECYb{Z-oWWPDpLU3ze|ZY{ke?o*_k9?S#xyd4acsPK41BPzCJ-l
zbmf+R<>$pKU18|Dy|3G@gvRhRD~8vQXbx446}fl29ps=0ImjdAFkl1w#5%g?vHw^1
zTkF(<V?Dd#Htt4!MRRx+*`)@)4(lcvT|pcm_Z=y9>}>B67{X&~H+tS4>x@y<F52Af
zV{Wn1{}~w_zMGS3N(1wvb6c+;trX#5)3_ZSW5OngDpO|WriiSw@|#o7^kf*(-W#Kl
z`CSJ~o0q?W7#V*J<>IZLt=Hvk$^4Nw>Tu`2_a}3=)i%i03~Hx$n{G4f9SmF<=7-eD
z&~X4+dFT6rmFj_d>z!AR{X(@oy|V?+<x;de(8u9vda0B02tO5X0pK#crfGY|VecQl
zX_-}lQX^9ELt8Ph0|{Hg6!B-CTi&*@yE50OCcV>}LZyFh<7p?*zYRTgRjpNy_73LM
zM5yKAUVCGYf0pmU<*Nswqx|w*Im_qztgLGjF5Y5;n9jKZuc6yZBVnU@>SEAx&Rlth
zYF~QR=(IcRbm+(cX-Ti|`P1p|jXhD+3^jutzJSw4S@yz;*^{681@ARL0AYJ;-PS!o
zEHe|^8$C&H+Y5FuPFY%ecI3N}yFl~$$jq~HM^U_5SbJ;a$lII5>w#Q~y=d_MScCYg
zcLeWRvV7iV{B-ZSDt0#IZVH;dF~R*0*4pVHmjQ1M69F}sQ4lyJmmrHq3j#7XmqDB-
zCx6XaZExE+68_#_LEygZeOQX*kR0-`DA3(r_6l6FMbiWBE(iiQQCl~$71+vd?%&@t
zw3S$~-PnrTZGqI3M2f?CdFB~X6k<|FQQ)Mk3)0BS`HV&~%G)Rns{)fWf?~tUu~qD4
zP;!d#M<H94ZIT8lYfO5EmO;y7d^Bj;n17{(r;cTo?3E=~h6ZJQka==cHYAxkv$E`D
z$_%j*-j1<OITP?7#ygXOOaW2xK0LHmITwtyARmP?t+X%?T1mO!%2Tgg#5-{hV-w3$
zi+PNZ5f+jeU+&@3r1%Ui46{x~h_t@EMWzg{j1XzdBFh7oMLN77$xEZ?GcBItzkifO
zQirwV@;<WAZ@@Ytk@C(5x#!VJGM0k!<jTE>R+O;>i+Ce*c#M??XyGY)Cr=qI7FAJF
zJ{wsFWfcs`Iy|Xh%UiHOC_}`B4TCFVLr}qInJwa8H1ZZWL5N-?g<?aoa}02WENY6d
zL^NLB7f3~$EanKTAcc$#NyUYvA%CM4$~})(mbFE+ILOl&RZPCLFgBGr;E*wk5@W=h
zbCgxWP;4_jD=KD}%LufHL;^POM=l0{XNsqw1=r-FA&zj;DC=NWoQtL5?VQWn;MknA
zj7>=;I0xo{kcq@iA)89RXcE4kLYXZ^m11#c3`@z#ir}FsmDylq7#eS8Jb%d6mR45T
zl@-CoVf-=&tL%#o;EuVtRK&6}r!vB*oMjG%DXg(FF~g0mA+ZgcP=YH+6H#MeWCTW*
zmM=G1nu(wN^i%cjf7f;OVmfWw>iFh;D_8#<PcK#Vb2GcBXD0^nZ%(U!RIjREPVD01
zZGGOJ*g!fWfZ!ckJlMIUGk-iK#-N?$)9Qu7>yN8nn|Dq1QvEQ;pm8(Rtog6c@$YGq
z`>e%_h;)b&@mc3AzC_xF_=UM?!`z#v11u@KoX&Zp;eCi|Ev9kFXkd&A6gAQ5v!3wP
zi@TSXNg>9+-oBE*A3nA1_55#D^>N&Ox_Ph9n@ROyR$t7<m-XdLMt`^UY}D5J^A)6g
zHbBfoIs=?KIlWDq-Trd6H*GT;kFMs`c{3fKSMzo>YtMckUES34EyC?<KgybbbBKM|
z5#DJdoCok0t@h;C3<wuz4wjMxG3R6u^LLhfc|!6Iquy5k{?C7vL^*JDdUJJky0L>b
ztP{K@#H<sL>b|F{et&7Mn%VL7=)6{CfU4e%T9{?3+~V@@x9zXTEv%&M;>J-kZ866Y
zz}Xrgg3NLR-?Y}{#RW2?b$&7bktDugB&)uC{=9lKYtE1B_N03A@~C=O-?pbKihVQs
zSVN#`TTk1$WV<2)6gY2gX6JReWy-DB^~HGfb8}l19xcEvyMI%RpJAjV;<%Wu(?zx`
zx`@z4e4w#|>0xEHJw5EXbV)gDOV7J6ec_|?1;<#Z;-s()k-U(MWg~VC?ixk*)2BIm
z(Q#oT8VHPE)*_g5M(4PWl-5i7x1b{i>S6-_IRIpq*|5@13i4LlxLpclS%*Z+@MHWU
z1IP4qAqdIBgMZp|LiiZ91y9(j3WH=ul%WCeHfzaHcnX4BpR_^p?-1Muq$3l*KM3yx
z?I3(_K_6mxJHDQujiwi8?QDF0*@tmpg3$0U0_bXo4nS>&3xRBW1>opm<_enb3mkS0
zGljDc01nG{z#+9mL5H|NwngQzypx}f7mkn=!Arr=w0{vAYMhQxGvG;svu|ZLE^9j|
zIDGb63LVdgP9Pp=zZOF7Q$UpD{iEZeIiKrEGp{d8giM-?`f5D=_~UugT)xL&HJ)D7
zxB62%85kzOBrUZ~1pkAVI0M=Mc^UK%?BQF>yKg>li18?N-RcPNvsv<k__)FKIuKh6
zwx2-ooqvIKC^M~6AE|D?oXyvJEyCAJ2EhD}_p~G&pS~qSc)+0|usb`myb$Pd3o(ya
z6lj;-YyqX^+oCvz>BdN=v)ve^K_7Ve#LLBU3~DoHX!%46b|jx=xAG(|pLp@PFa(ht
zts&BKDSIN4TS277`WG7wPsSfgz9eZM@SDnRfq(X{<ZAC}Jx4Q$GzV12mQfcAJ&y;d
z@VX!&U=hI3*`S#vF?vlVZy-&^fgPY&55~gk351~+CeA&Q;8r@^MMy$j>2~O7QA)V4
zk_dZ_zpavZ5G=ir=y`NMBzj?V@6o+5+Sk2lJ>k-jB1Ja9c7Qcn2djvzl{4+o(<fw+
zw0}!a$4O_X=xdjXJ^bpKaXSERcZwcl^-X_|dHWjJuN&A8VpKlf;}>;|iaJI`2Ov^V
zixxdwAy86f07DYu8T!^P?FS*l*&JNQ*n1%;fzi`e87M~&t5bGtOa`j5IRKb4Kr^ud
zKQf36hX}F8E4V+q_1!YmoPiPue+cBrp?_%P9+G<d2$K2@K}oD-gZUel3J+qYA5E(P
z^aaw;%lWH7TK4yS*uX7@Z}7wr*Fr1!A)yt`eQ5cP8U)idY}asIgANgFi-hf?(*!+R
zR4=|16&4k(UH9}5Lqaq<m@APpDRMXjZScD6Z)esH#<GYuE9XK;AvSy?AzYBM#DCri
zK_s{nwh2Kf_+uu96ft|s(BJ`fT2MriU?0rl<wOKEUP55Vwmp<IR1-TClO*gh-Ax?4
z(GJB~H}T<%v+L%MrqpJCx&Cq5%qDm;t>^vijSzNW6TMzbjv&QZ_B-?h`-Qc?{%**2
z-q+^{t3b9L{Cd2v&lOiut&fDgx_@Xmm55;t?;;K1iqSukkybj~85TJ)qo8=aC&^|*
zujij?*q+6{K<Imw?3N;VUyJd&#h8X<D9-m8Ds~x)y9~u$hGKWP#1suk5Ttc3AWlWb
z&E;%z0K#Mdh9)Ni_)6|@<tznoE(xZE5G2Wx*U{L4L%D!x2q}b&ShMyJX@AM+EIHA7
z6svN$Y|A!Zk<G^qg}t$I0FZNrBql_|hmZnHkR7Xpx*pubW!Y5-3h4=~d&2=hFa}u>
zbAya}ImpjFGRl4zU}wLPk$M+*=n3YmVTeDGyAHwf-nN!}+Ch{y!1s2@@4?X%FeBwP
z%t*WgX2j)r=N)F``?I^4x_=G#6uL${AcvKD$t}*t<l_l1r!u_moyh=HTt=jN&>ILy
z2Y?D#P0Qy^;9y$3FIdgcjk8G;tsjwrYX>Kv$)Iw~EM*%XE=VnkSB5KgQ7UvgTEBmo
zru0aLUg>amK5rIZlzcCFcL;~-1@3Cx4G$<b{$W6|aRUO1%}NsPj(@&mck~@|w-u9@
z`I(LB8n$a}Iv#(#MnE!=?{rk4tSw~sus8}2)CzE~Jm8C(&o`56eNq2jy&kone!Fg8
zL0Pj8@Dly`N<LqCe^n1Wk6oM_d!zvQma98A_NeSUg7LdMH|_-O&W-mLbnV-p@#Old
zKC5p>@;e_apeF)+$$u$A)=9n@U6zHOGsv~uUz7Dz|1rKCUyIA@X7;g?tLnI(j4$dB
z<0%w+c0Fr8&PJ07s>Pwlj2_s_^4+KQ$(1iR%2%CWP=@JHLcbo}jwd&hv-6vFsP$3o
z16OTK+5xBTYHk)JO+#w#aS_iN4Yn96=RjIf-pY?3sLSP7FgA#LUB2xF1{nFSI0kqw
zeKi@8T-eYKB(dLXnSxN$$F(AHV^6YLz7Br8CsKe0neKZ2A15Vymoa=s69O|gmk}@}
z6qjC)7{!0rfBo&_mp7i+=ggaM8-D12?;me}zR}4@>lsdT`q%fj@IZlJ;N*l9neuM>
z>p#=)Z-1v~3o#2gCeok30zfnJk0ehgP@dcvts8j?jzU(G(T)KkPVxktm2zLd7%uTY
zjgr(Rj|BN?l&`ujVH7^sC^L-`)jm8)_W|;jpA~=TAZ3K$mpsD1<-0Jhpg`Ud9+Pl|
zrzja^3onRS`4;m!R(@Gx4%oCT@9E)eOyXj}!925q>5?F#RERD-UY>3(lw}HRUW+gz
zd9RYl63}|qN)?`6(zu?-!*e0#d4zR}GP=hZMv;Z#cl^E0=+Q8qW*Q7r)@7!-&KJSV
z@il*@!ymkhn41yA{L|g@O;wc|t;HAd_@tm|x<uiY)Q3i(JQJ$J92Y#Oa=<9pgxN}x
z<1<Lh&s3OD{uAffj4UK5xnq_aN?m*l7m=RvSzuDnSeuaz=}BRk6q-_r3Z$!YXXVL~
z3@$G!TYeTKd)*~fX$`a1Qi}W*-!8J3=mUSz#~Y!TF8)XQ!(tSufPkD3M7-g(uDD{(
zQ`o~sc4Dp3e!L}wiTPDv#gwKqT59F2aYd$42a+x6#7Q%<6jzoz2Zmm(td1!T^U^qc
zku0h>?CAbgcJ__?y{7-rT9@O}&rWI0JxiR8RV(A=oz8t<atKPR;W~SvbJbolqo9AR
zw>_<76w{`bQQ*&0GeWs@^G~>wq??^c=eb1}a;Y&i{(s-+Ho!6EI^Kh-N7YZ#(#IE1
zYR|0p52R9{jba>Q-3lKkypyz&X({*8POg(cg9)bol(}n%d9zb1=5BH}BsR==U(VK>
zCe8auerHxf-4lsIo~y;Q1qG5FXL5fMF_b0F_9h70GNXZFJgE!97$!`|-#`}xLMdS=
zG&<44bgg3^G0<5IoTdv|sG`up!i)YLpzSsDs+Zv9<>{dgR&GkT%-OoKhbeWK0rg6`
zACPzw%-Yl>GuKif0o6GfyVitMCykEJ&72l!NdEFtGXCh4lUYUsFvCntOT&MhEBP9T
z^HH+J^dL&Mz@_T?HqXD#XE2%ji71(3SW8)`K{ht0=>+;2Uo6ljpm)WgIiYegkd`>!
zt6YGN%kV8R#j)9ZGo_3bDg|X~2%ih~zQ|rF|1Y(6h}f>tr58Hoj@OJY-wD3VIY(|V
z3k?^|+uHcKpVv-;X<Iuf+0TE=;abWMY6SwVXg<b+iX7)fdc^8FUV!-20~m9bM;8IF
z!3?zs7DZkdyM*JDbeiEsDGb7+^uVlF9Y3-vMRZHoxyjh1ZToowkc_*IidffGQ0QvV
zqa2eFwJ_yF;#(@h*u;@Zm0V5Zk64@r2HsK3Ba1VFd2T5*bbKV~n$&;JHK$FCP<miu
zlrK4Zyb}I-8@Ph`4<fB{z_6fcLLLq0w?(fhwNs;yUz0}GFJtF6ohYeIGqMtj6qYC4
z*CY>8;-PBSN2N|tY9p?%+??1FY(swUIl6=%2E3o{oDq5?fFHsrWN5Ebgaa3lT51G3
zn*05<kA2}}Krsn&?`(gPM_!qY;>}_~GP-3Y@O-tV(CjfZH`#SC0=+Ynj$xW2c>#Ob
zcTvn;lD}tZhU{m{X{kK1__=JP7zf3kRp=uGg-*FY6|zQOV(pgOiua{clg52bYhz05
zY6BCEKUXRIzHAL7)jQtmiOB-tRHR)IlDE)~4|(gAIu0H^hI@Zln?82nLq*Dzq(cGt
z#-YpPz*AsqYF0mX>RFLmj60nAqFO!W4>hw=QMv-s8iu`BciuND!@TcjHSmpE<1%-=
zd_olfq4yxV^&F%qfE6XG1ci=00a=p=*#+MwJw1^=U?ha4uF?I|W^3tI_D?rm(Gurl
zZTpOhEc=HgXTyK)F|L4dEiA3Zy6u|-dch?F-tm$uD1O>5KzIg{NPo}~0SQc7O;&1<
z>T)c)Ra+))B+b$9Z>PDG0(c!yP2|vZeg~0Y+HMRG;4^#1ePE2iuxiJ6&JIfwS@#EW
z14b-to}Vt)2=-B9CiTV%bj}J&(@8Ch;q}>AV9G4|_2++Au*LLX1u5B9h>w^y5<`P|
z1=B4iAvYMG+s5RLPiaccH8AB>Y_?3PjPhQx&JC%}Dj6`B)2beaDNXO=l1>ZWA{+@h
zTR{sA@1G;+xez79M{-VuD3ufm`%%U<?E7;ycS?mQXQRYv-f|_-D>?q{bia-+r#PP!
zrk)<18@zvPlPZU8x>CoA-5jJTmgfr<2sYQV_u67Nw<)QZ&lkH83>z`+Q%VT#lXs3~
z?sTlXpun+nbszq`lW#FSh)gYT(30YI>|&27bwZchfg@ikJu|ZI?VUdXLohv<NDI6@
zk<<Iygi6SUx!T;@rhM%Xpm3oY%8?8fr{eNq*ZY5bUnX+kEM@9^w?smWgyoK~qkGNs
zVL@`#vJ1j3(!un&WmbqEJ-?O;GQQZo1k*&RcE6H0`<(CNNlSq_o$n~sxbrOu^p<B+
zC?B0wE(^O^E$NN;bXZb@=-$1Z<0Qwgo)gowa5p4ZV#^`P*;D)i>O(mHK9rmoOAgTJ
zor8ZDNe;YVMwy}HZbLp4nwK#ve7vk_D80cPO1|(I6TwI=?s?ABftiq@*?PiZ*zR2W
zu0i0hSp2SkI02|YUegaJB*dji-%Vr;m7a^_9Nw>>)RHOT0?J>1<I}(K;V%drDuMq+
z(C>u|pZ=Ks89vN-fY_mg5HV&QSRQZS<r9DA{1`1~tQCU>&oE{l#;}twa7<dUjTRTw
zYl?UCvy#w%M@>_hVrl=ngH6RNTx*$C_S^GLr76c4w~uJ8f4-4VrIbUDkvmsUz2t^e
zvGFS2MoY`OotE&Ul?w&Xb1HQd#ZDtaw0s-w$cU)5?)Vl=Vo}9;EdVdeJ&k#x=Wl<z
zmdBPr*wtf8X-N-$!6hs(Aq;gy!Y~hJ`va~WA7=@RG1<#y3nFY~DW3H_gyH!r+Ux4+
zD*7zGuS>DB&ekKL3k6KdS`sy)b=1~fZ+mG0Ft?k(NQ%59N%T~WjadnydOTkYa;v$J
zOu3LKasK4;3&j9_*z!xxjDS8|%d>waoMb{8ig#GTArxo|{i-vnXIfAxxhEVwa~dt3
zy`UJ~;H1QSu7X0Jn2D0iD)E5siH+}V(a>HVfmTCHIFg$g%rV7FQ!^~uF_gYOGjmwK
z>1U$oK?M7_!id8C4@%3dVPKoB{2CJ)Z(1VF{%q$o-w(vWW6$o1RXU1cpHdAP^c?&p
znGXH^KaPv0m!W((69O|gmtp8U6qgM|9x0danHZpdPp2Q>zJ1VH{aRP&NwG$quH1uo
zOdI*mI$9`WTh^M+@>HI9enQ;V@+ickTa<s>kyLatnB2ape!aXs{60~^`lwvM=I753
z@&_R(45kT2x!C;iOY`f)ceT)=F9NVHv~$*ud;yvrM2)kcJYbn0zOY4DU(%;Xr^VWA
z<b2706C+PT_igfD>TgqjnQK3qRiI<zNGs2y=F#ly=@E=3>G+yN==SMsUlWy&cG#Ew
z8ggF>;8yxVPyt=(%RYZZ2Wbki|4Q?H#SPvM$IXLbufoE(HaZm{4EyFuVn6js-_|oW
zSn=l2G46_E=!#=}=a~6;xwddQniXN^0vv^Zd@=)TXnFC`E8jz26qoi{yHwJ`xbWOl
z>!eE#(dnqpD>P7#s?V~J`TklgA$sUbnW^dKEG=ygXz9@Qkbi6r)BXv#(J!BSTy`#5
z0~~KuFInJd;3&Lz%~}%_K$kypq&oVnGI6Kw<Bv~G_2lCq*P&5hTrJn=`OL^`*rBa|
zTaR3yvm=Syb<Tk2_BXjfXqdJi`8Go)*S>7l->|ibb77?s;7I>%5~GSfOeBW=5-}(W
zej@Q{zeM6_iW9GKLYwoN;XH39PJ<oBww#7-Jkm5SD87z#pLfc8f22fC#(YgA(!8-l
zRMU03$Jj9_P4U2}LG6^aN16tCuj5pIb)-at%KBT4jfQhB9$;)lrx3%H1+_^ckqYFm
zOk|TpiwdnfQlf*rWv#idZ|%6%Bp_*Vuk8AN;@<szcGo1v+L7~kwQF)TFk>jYW|7d(
zHKgGJf1eQ<t)gf|<PSgU=D+;sx7-+GX#Px1KX=3E=6}uivMpZ`%qubpHVQR=tqInR
zQ8C0UwijVBrbWG3#EqH-o)6h|v5X|5t>vyl57|DDg%fZ7IKpBbWf)<y8xYngYWgXS
zVxu$G?pa;2mJew`-Q&ltB5n*_77LuIcZ6=eMsDEbL{@nN|0K^vhA8O#m*Kg@Sp=is
z#B*sk&m}NSSw_6i2eU#@Ni&3h_3`!&7`3uyfWAt5xXicr3`rbp7=Te3&-IoLt%cEg
z<p?(sR{MnsD;krwtG}19R-^7CtTSZv3VpqYu$st_3&Iy7tR_n7h7KpJmE0?YwQBt@
zAgq-}2N2d;BHc|`Yl-1*!n!`|im=v=_Y>9wO_K@hf$qzM^+1W4u+|cPFKdML0Apvu
zdR){L!g`!(3Sm7^VkWG$M7u^<?>BZKtOrgc6V~;p*+W>@O|yrv-j+rRVXaypLs%uF
zIg3XUR>K$qygQz-8i!0l!aYQ1vxplt;}Fp%ei_1QEUyu8bA+|WV;vcNt?~UBVQnF-
z`)75{T3X##ci1Z7#=xb2$f*MkqpLPFLD}dX&H;M^oN+^cEc<c_F4KsF-$KT4kU;w*
zQiAbr0Q6l(bRm=1OWT^{Xpj;$i0xCN@<ec}@p8<ZX}~)9fXGZL-ax6-2#IJUY?-iT
zM^FTT#_kv9Cs|vWK}SlQX&3}L<7lK!1l~WxNSk**y>GiPVq{Ez<}r-dPIM|5;yW_+
zZr9E`c)vD^Vg70&jYs+V{UbQ{?mu$#?XCOub$cs{u$N~PF@`y4jAKPUzh5)Hz*dQw
z%J{9pvY|Z9PJtNka;6wf;p4h~)+z^Lu%Lp@`u%@hKi<7GQmvolnank*4N`i|oHJ99
z6wZEFcj+rJ67<7=`iwQJA%RSvB2_c}ZB=b#4i(8v%(}9o<Tt>$R8Uri_lo)|qKPN8
zd8*cS)6?A(BZe3hoXRLZo9mLbaK(^eu$JV4=XpJt8aUtcQf&2TS7%$;5_t@2MayTI
zPkAD*mFOtn(%TD9CmV84(WdW(my>Dt7?4Mp_n`Fl3Fb?GhPiac(wx{dqdG0PDLB5n
zs7;HBX8#$&Z_O#5H^gdUCRgUt#pX;k1ardlw#7NG4n2smK{9N0rkADUnx9M5U2^>|
z6tmo#a80yT!fpF^=}_l?Qh&g+$$c;Vn#et{516!ahP6rEOUfh9Ax0^2TSj1&4aI*1
zwrm5J@>WrQ(O$wMg=?DMsSRN$>v=x>$TLvvi+-b;lf1)}W>fs3+kB9|KUcoW_UN;S
z;2SvSSfjS5zRb0N*jC-(!yqcdAR0GTk{l^ClQ$ncD37pSNMckXXy3|*&Lm*;<YR^g
zQLh!7z*t5d-RmM0vrizGa>f<DVFWB<NrpB+Q^_HJubvWHm@{`SB*~iT3QE<QDTSqe
zb)4YC(gJDSsH?)Rut-kQ0K~bp1Gq#xs%Z?7C25Q$EmqT*BF4J0;N=~KFc5ag2(glu
z;t&=&9SEm$ld0?1?c;mmH4a<H53+sH#Prnx*!JRyW_fm5n=oIy0j~_$1FT_;oB`3|
z_zN$8JLE#);nrYqW%X2>r&ZsS+S~!w$Thi*<F}Qxqm5CjU`wu9s5C)L2NL%iBiF2!
zhIjMBGibCCk2i*#G)<RD%*veSmGKOlJRM;R&uD+f80&$7%)^|3?t_g%nmLkN)@D93
zNKPP(KfoF!ClBZr$DdDHM5bkSQ)|>ms%4*l)2L|h8m`{I&y6Kr+2_WM%l=+A=EA{r
z#sLePbTsL7t>_MU)7NuApk`6q=NV9ekJX^nG&c#9&<&vyxDZ)o%0bEw{mJXoZg-6Z
zt$S<SHo@d#$9!^^j#x))&JK%u*AeI<{LJ&_G~-6tv!$@N;mkCNk%R*j%NT1cb^?A^
zZ*k3{;u(+)hq&W>4Oos;Vn*Vg=lz}m+_34`z4Uy)1*27w@R|jetYbJsd+p}^x&e=X
z5xz8(qF?r(4`$SOkju)6_6;s8D0A$xm7|#Y&b0F{l?;iOq0urE12Z==mx1a%6PF8W
zE+~J!TT64}HWI%3SMWG<FoFiU@kqH;CCPTGa?7r7*_<|;Ty_qd%l`c|K!60mmq^J|
zWm*ymG#Xzw`UOySe|EC_>FM(E^uxQx&I(6hRrm2n2mC<te~^6BP-g*z(a?Q->wbNa
zQW?H3e4&(D^JV$F<4-z0YjE5G&V0e?*&=_W_FC};{QVk!=XS9-^-t9~QqcY-QI>p#
zmKLNAG%3-&oPPWG`AH`BrSK+v4Ik~Vw~wd)J++h3)(Jq{{rlTfcz_y&cB<17__6E$
z`cL=A(=RP#okN&rO|Y%Ywr$(CZQHi}m2Decwr$(4>ax0wF7x*N&zYUc&i!P@FbA<B
z<6W_dLJJw2#ui>N1f$HghkX==py*TKmbsFXI<cI7TW(1TO7uFFaaB&7_yIut`F?M<
zz4<&UcZQhh)Le-#G6SDj*wPN7TPKe<ppCDK;8KAB!vT;=%6K2t+Uk;4BQ)|TC%&2s
zB3gx7QnRZW2SI``gC6)I0t?S0G<_pK`9vOUa_^LX3miIdw6X<HC~Smkb9h%mK~~x0
zzml&09RR>-&f97<f|9J1cmqOp50)VgNo>R;i&c#GFS`m5H8ih{hazk=J`kewN5t5H
zK~Me8H{XifhOl;?aZX8xnb5u`*?in1B!3QernNr`H$EG@*^j;Q%T&G#-P;Z=7pKf<
zurS*3A4Q&hf;i64KLU<W9tXP9ACjGo`R|t6YM(+A3?y5n4F;|ZcL6FecPU92YjMd`
ziAF;i7qhUNEpz7Ye)1@V!egi<t;iD+w50Yb2}07Pd+TQ8$#lcbOt#sgq_)Y8eCiAM
zaM@}*)Z{7L%4ATVL|IIPf?HJ+;LjXYXBNUP5`kGty2qYQ$nWem_n;51Lo&<pkX~N;
zv`T}t?O=vy=pZ#pLIC?qE51F%*Y4v%0#58W5o`BLyB5?xSHJDzy>5?TaeIR&tM<fK
z3XLv#vKD%J)yJUwD7LU#6;7a-wGe&xwt&&hs!EHORHP|@<c$eFA08U{jG5pEaWxX~
zNMoHn_K0tk;+YRNdM&&r{diFP*yBLy7DXUr0|W@(RuX#|zyJbhJ-cvyd!!Q0RIx6X
zT08%&;lT2o6xuI5`3aVfsVYyMh_x<33#+Uf4j;1Xk1nNc25IydDmnty9s(^^vj3#i
zZL%`Z(Tweg$UcRuqaW=aS<AA<@YWc3z_E`?a)pN<hMQLCzQh-a(aDU(ZHnHmMu53<
zCL6$B-fJ-Qq5@DTI|yg7T;f+^@xRvfmsW(QpTe0U#&n?1DUg0Qc1+M3D+zizv#YgZ
zuD>`bP-|Iv=|24bMMv}I{3(&foZTTQBL-T@!-k%biJ?A`0IU<d=CBvfxFnSH(?E(R
zGOx`zZ%U1GDWfBJ4K;oe3U6FXyi_;xTuz4SMv4d*Fog?kw(Nt~KTaoU?EC7Z6Zk}b
z0-iUe-c;sZEQ|U$0c$%(CV;!zgAy&~Bnu0(e3hc6DwCyqvNQmEws_~A5F$!WCnD=i
zU8Bg}S@yJVwDD3_7)5+*Z{lrP8v82}6+DXGoWJCRGff#nRz<a_{o}1!CNY~t(6Jcv
zP1e~9AUaA`{x#r&Ml4z3LX}!L4?ea%rR=}hP5uD%+{A264T<GRdr$IFiLn&xBDs>j
zJe+=b*=1X&o6<wrPlh$CBB41H@$Gz<gdXn38$IG4v)9tcCt=Kzx{S>U)6A7~1#JT}
zt>Jxv{{WYEP&2=Gi;u8gj)QQ#ZE6RQN9Q;O%w+Wtf|BxUP)AJmm-|TVJ~nnC#vb)g
zsJNHwSrWING*lCifeG1w&KiFwx#iNQg`Du(atQdY`#AabSJ<9vBx$Lb(x)xBM9=u3
z>2d}PxXj29C849lcV)C3Y^r$o7WDUQ;7+yY#Pl#S7daJ*?A~g#rKY?5QB!o(5gf7v
zkhLsEl*rX`#ff{rCAM{N!z}ulG=WZHh{<=v4tnGT&N$>6PWy!WbaonUWrVi>W_&PF
z?P?L+oct)9mAkFvb~-MkZ$ldnFmFH_lnc%>9I2$(c(lt?ZOWWDPa^A?WnjFyQmAm*
zpygLrONkqnO)9M8UY5~GXb{#e(AFpfw0r2iq`Eh|i<;mXU6WG3IOEzdqIF23DBKIs
z%=t|KAk2-}@z4Ut?3V_K3F?bQsh%2_NIq!eoZSBMDKEyu*DYPYrC6bu)%SSwa`Ifa
zo0*s?L&~~$RH_ON%I57OlDD{~^vvQjfqaJ^x;}%Y9CqXkC!7MIA|hL+AuyQ(U`T&6
zAe_B^%oI_HAb?RVd_Hk*e$62QPREGBKLB>`_E?f0KiA*b3R3n+9&~|1nZDTnl5h_h
zFOB#7E2&^F<WQS?2NX`1@IkN-_of==e4Q@a@&O{dN*jv0Fl<Hd8nDvI(-WgEqt4_n
zpfTe*&V~IM;PaG$g<L_j2<BbDl5w9GTshn4$&Rt0FKl9Qtm8hg0uk@(Jg`_$!%<c;
z3=$_S)D6%v(;04LLuKLal^+PnAX~MzkQD|}Gic&)Y(3~NK1#*hbblD4DPNA6hT*^k
zdn>#%A5eJ7A)o*hoOn@g?aOiRwcy+r-Nd6@M<Ns%l}Bj~b!}YGTNGqK!D&DzZqo3-
zeG{MhSMErxVP>GY3#eI=Lk}XC%)j(vWR`zD!z8<<e`6p+G?DEjMK-|m7vD{|D<rl{
z%=a1>qCwkABv^g)An%-O7yD9k48%<7z%4ayV8r1bcS>rTW6g^pq$SsS5}*NWcrBjK
z9;~!*DEo$mqBJ>1Wl8~VBo8}+rOd_yYK}$-ua*9rJpI_qQTdw1Jo8z>ELy9DTu&qr
zZO2d3Z<kw-sT9QSyxlF#?OX_uc`B7Yxzq1l6EF@tGeXuEq*Rxp5lxW}WnTJ(JU#oy
z2W3SmsZPpWnU`vHi3fMzQ-#hU6bszH=?jVpOZ07?#5@a^@O1&_pU^Y1#m2H-hEHOU
z@4?jKy=zZ@kR(cQNJA8x_jzZb#(L4yriKl_N2MrUYxAy;g@InXG;f-S%;KdBZ|;u9
zuo(JkRNa*g*H2Mob|B<<vBNquZZ-f~*B3TCTEKX`P0`So{yN4PPe~$OAr2NLmd{(;
zK;eFzc+s1{g%kjk-DLtOcJ8*0sv8i)=|cmdO(anX1FJ_)W)uYBdzacSJEoyvVQF_v
zJxMGjAmbf#-E<l|<S~8O41;SU(&t!*21?lCu+;7C(M67u4$vODQT%;#>U8}-m-Gk-
zY>=uIRjd-!#ZqI^2^5xNQ5LGM{QNJFu(GV2*#mwUl@kChA5)6yWphzOP6uS8QF;58
zqlw$7%fBVp1jV&7!+OfpGxh$7Z}M7kl)~w$=*rNgTK*e`s#|g2E(V@{QdE3$WH|}~
z^2dlZibqMl{sPLmPgF5{F)s&Ve&jI+<a2I(hL=;wRYCD`L9^S^N7G_kWiFAp2&|4o
z5}&efh~j`X8DI70RVRB(mY%tkmd16BbR8aBZU>c?YLIAG@8PIErrXPbFD6V*!HnVe
zG-P~*AcGH>=Rb2hesm{X=_a@lL&E2*+YoC|8Ex!+HM*UdnV@d*zDDPw7;#Pu0txtM
zzj3VZ%IEC{%+_v`yfzM7^589laMrtx5V$ejrrZs#mVd{6{&9-LnqrnIP$HNxso(_#
zVTZ3y+Vnu;eI1OVb1Gw|+n95rDDxZ6%JJmQE(I!WaT`^2P?CvR<HxpMEH=f5516TA
zbr@RIQgy%B*+`TuQ&h<PfiUb&M*I&fLFZ&;CSoFTG_ir<<71RFcd&G`B4W*88YQC2
z5URss2K3%HVrZb-lIju!a8C>u5XLhi3g!;t<=$1~3iSB#Q~3~`>r3Obcsa>H!(QAy
zT;H85%h;d0mt-WN1RBKm`M!^oBNp<_D$PHtC=%u>^AWQ3qWKGs?wK0|kA12MQo<n-
zU^ka&=ANg9MxtWho}d4nRzytR%plo#We=F?2MF}>{fCb<QuRDUFzyi?f`ZBfu0z(w
zDh=WXe5@Z_4jWpuq-By}1>MK>eLYVb<Gj&Mvq~kGw_rFYmM4>$xi)Y>MJ6=y$tR*n
z2S`0IlhKlS<0PYAaB<W;M>K0zXUZc@(J@Qblu(O4kljm@6=k~;aAsGE(_|W2f6f5@
zLNXkFmRHO~lE^3QZ!CyECQD7$<WFT%RHmqFKY~c5&uE!Ym^g)58I&<8C!WwxdSVg3
zY0N-{F3xAR;(=*c>XHY${k7A&OVR{RpIzIX!8Dc)XH!<ONn@msW2(ULvrczY4nCeI
z)ZXf#w>}C5)+D#b9x_iW-z!&eq!9$fRFAP+-+7>}Evlg7skMF@m#oVjfYmU~Y0i#3
zfpIXHF~~WJ<7PLDY=XDR2h1xH*2Jd6C?^o!E)-xYY-Q!Mq-=Kr7cVWw%|HSCz(puq
zhQKAd9ZYnfO@~DS+a0-UL!d$r;Ke~DtQH)paqot_JZtS@G0;j*HzhDcm&gG!;t~68
zoOu#KG%MG>-P=kdz=^H}O!l3urlE+iO`}`*$9Y-1D4WJ5l)50}k<<>gSfS)MzRS=T
ze!TvY!Ac%!0uIYBM5(Zj!h%XRM_)Al-O~L&*T_>7?ni|yAThQPywCcraRiIx%UEYy
z7!#^wOn8cFF_c~ri|<&rGhzTc(iL^(gxpRB1m~LQ^o*r+Rf6R*S>?kQRLS$B{oCsX
z>+4IC3^3uy_YHF&Y?o{eE}AB>_C7bu$Y0rw2FSw#62E(*r#sIy4z)!Gqy%04)q}P1
zpCXB6yq(g`cEc_M@Tu3Pw?X42h3msOj)&eIhj4ldOMJ(8!Wvm{OE3UL#zoy9_OKxs
zYXi93YoYGrZ8x)#6fiP<9HiOEB9HiuOWlObmQ<{#f6Kb*um_+9E&N=$7Ar!>!Wf>5
zE88R%4l$M5C0VaqglaX<9VM5WrYS0XpKXB-f;&Y%+l|1j!GHOReuQX6()L;h;^qs5
z2(_==(DDT(5HI$$Qepub0}vINR4QCt>o+>6ayejUcyTjOzmN}a;%sZ6j2oL|c_Blb
zurQQ7uk<d#do|8)kBiuJheVD%#kYv=aPIfooL&F?X32O8CPW+iG$oGKv}2`V$sH>~
zR^HP5t%UvVVBC#8crq1wA1$Wfa^a&#A={1%OKHlAi8&<4d=Cw{Krj`j@h#CV?KiC%
z9g(a^6{yo(#cI#6mmh|eZw$5Rk83S7>bYxTpDt`Iv#E5O+dH{`K1`lIj*?kKE#0-l
zul#`hT_As=MQsY#u44<>Sfwm_bnvx}rtgd7ectfitxWT9N2H8jjvozxH#e(KrDCwu
zN$IYOrK%hQ^S1-k4w5;}<M8kHO4>9SWPF9kiPW?VG8Q`Qks?hyHP?My^Mm{e)raDA
zOqv7V9e}XsaI%;j<9W*z{&VtRc;FKx5C@Q5=QZhBdU^;C;9ci+b}qflD>OLp(fsr8
zSh9kDy}I^Jz)x{q)>vTR<+%qmsiyP-Yrd!MZ1=oMumrH|uEEu=HP<ST!T${PiLX}&
z1Ja^{{vJQRDx#9k3A}UUBpn@U9QK~y4-RZaS-AeX?WI#X6dVKZIX&l{`}rPVxMR8$
zxqmU}&wVJ*Zhrn=H7UcWnwQWlANd8#kH>SS@4C9^K6p5RG(@=n$LH&h;bQrP!1^tQ
zl0P0V`Z8b#B+}ixQ(Fm%E$TyTcKHQ8)@&6;WaRa{neQ`&;9D%^Xz=A{X~i)un(y=D
zuHDQ10K`O*hiD}A+)K?j@<cdBh?%I^o(f6iqF^`nP-cz|hVW59hb23Zb=U35WO~ZP
z6amT`sj8H;-xFi5-v-?8n3tYn*Y<#yVPc*hKmrhi<vYc$!&_(5M&QtHb!+EhtdLG}
z1b$VpQAut1<8k!o1#U}RvT1oa>gjQ2Pq8Kh^^!DzXRX*guk+V9?UP$OALDtH3bCm{
z>!PE$bEzX~%iKPHpS{Pgf5u#d6JHRP#@8sgTw9P$*=jv47V(u^uRvph9^ADy+zu=h
z+7<w;^*J(CIQo0bo(A<h`|unyqzs0gQI!QLy}M}%8?-X2x00(Pjgr{rnIo^GvRz0S
zF1t;F*An%~zvf}h_9YNXdkTq|&KxJnY3)to?9*%Q4jd2XM8YkuQm2&njuMj;3hz(}
zVK5#(c6VPs6q|pymQs|=2do3BvqCRh@HGL?dYa}mGp-qzuXVD3IaV}0cT4#?jW$?$
zTr`M}YUkdHhllC7Ni!chBGXO3G+#Vg$`$hfGfqGZCtK3=HCeo>W|=eI?_bLEkC-*%
zVyf*Sn=K1t$T@Mfq=Z?L&-?hZza&(dUZp$znpe!{V@s13;>BQQ@E@vkwmoJ3a)$%%
zmd#qs@n!p>WvRv{mF9O!!=#=%(v1!JnyZS-TRi4y(Mw-6-oJROyBE`8I3Ci8l)22e
z*ni<oa?~|j$-L-yo5GwqW30N$=vMI@DHrA;te@f1-w!OcPe;S*@x@em?>ShKe-qtj
znD)n<p~<XG*oR{ojHBW)tgBZLN!J1NPnCWA_0dq<Gt|Yzw9Bp>Fytz<u5?Cyp_u5V
zJ<2(4w|danQ_H#Vwxkp3bGo;^%vNyr8@;;UV@#qhO{A}<^P0~yv~ly3+hOOkbCDCW
zAFgpMQ!VT4wE~6+diE`%qg0nAZMk=t{>`$86)t&^RG>;=uasLPU0$M1>Q@4=>a$(H
zY7K2;K>qW{mVt--?#At!OzOB0!qD-gQIFDZXOIbZ3E$K{LC%p3{~XO&UDaL?e>25q
z_eTsOH5$pq>#b@i>egHvy%Qn-<_79LT*+rPOmuetQ18;`2$5@!SvML^OUm;OZ~zxe
z_D>em(NtW#i&Z69uQ`*m7hePLEge}Y+DZ<?;i#HortwC>)|yJyHvf7;c?}_q$?vH)
z#lZb-ih1?7$68(#(R|8=Wdun}zIi^`^`XWjq$qT}ZJd_!5@U{OQN0|G8FJ%{A%W+)
zN6q`}Y2)6&w;z*rRk;hCfg{jFDXBbKw{1XQ2V{~W8!fw~T5cy>j*~89za0w!z1zR5
z8vsSPPnuCdXc}kY^X9Ccbd~|yfklkoTp~ow>W-&b0M||~M5WXW#-L-No59_IMG3G(
zubW3d8jAOkuk-1E(?l^4{D(&ww(<)V=}ogXFp4+9yU`l?gs8vQ>6_Z7-|I|Az9q1D
z3=)p-(bX<OaGBKS=TM_ze8t)qW2r>f6r;QNkmmrcjAoP?mi@I;Pl2lb_N@wxc^MM;
zknS4J=Dd%7Q%gHn$Cdj#MTiUZFLfDhHlNAKk`?TKHwqPyq<Ry}_nZU}>7m!VssUB6
zPmT46JO|VMbtoLx$EikSK3ZRUeB%It!v8G%>uI=WxV)jPW^eo#-y(U!b)A*!Es93D
z)F;}Mp6^Vq+9TBCO>?kp`W2t`_PcvJ;By{S-G=m~c5^@6LiAt?F0V;Hz6Ah1REW<-
z1LE$04x)_3E%GTR<KHhEo=kDQdNIT~p+y?&&Whd2NKaYW@I-rv(~aj)e9!f{Iq&cy
zT4yJHBMsIg@9x;&$JiYc;x((e?d{wa<2x>1U2?*_jj<o5e&7G=kxujDHlW7{7%gHD
z9;tw#N>GYb=<`eQR*=ee?+ajBkPcdPEjckChEXS)C_c?3^eG2E|ADyG>Dm9?HE$Iu
zKZxuF;0;uxJ!Y_r*mz$INFEo!7FWiyq_NRNDV*32EC@UXUt<H7c8>uQLbbh+fF9}p
zK6e)WNJnzS;i_~%3t~PO&_NWVNR%VS9%V@2_+H0`*k{D{0E7hoB_Rma<$!wKu<5Z3
z0n)AsX@yrw2zmW`G(G$SoXP}k_FwuB%E`)|k=TWWmJy66L6I@lg@p*HXm4l+oq3dF
zY-k7TBNreW<Pi$Q<M{b$SgDrzb)kPEy1l%ELY;*6mSn=-|K)tt4gh$4Jsd<tsBuo`
zLeM%G_U!=m@4qrSeJ=dT0XLEux}G*{FDT%`gfOgsRmqi&VvGUsei{k_()XRfdiG)S
z1$Du_d(d7%K0gR{Gn@g=Z3=tQ8B)UU<#Jg^w6QG%=7mv5K%K#U-|r`agkA;$`qhom
z2hmeHX=i$*?)d8#!<16G{P}_rX<<Kq%R2(XK9m^G9qman2np79`Tj~^tx8Gwj-}ak
z_<x4kepAn~n7E`P&xF`oe!b5t77S)4^!w|WS;4!`dB73RXDb5=Q{=4Hp~2G%i|q6x
z?EAy*F<Zfit@6Z5&EL}nr`q33V-rl!l@o8gt&c!BZ)zMhf7&{Xjf1G`wx|Ueqc!)H
z8l1#>hPWZ3<-WS8bPvqi2S{|AiF9?xjvYf{?~SOIV{zDyIc%)3rkgG72%HgHA{^xj
z>S4ycl!gVmLrMVfsD<I81?!c!Pf3e>gk<{oMNPvE)fwX5LP@{;+j;~7beXN^=peg#
zI<VDq+;x_Sm{P<mBnB&gfpA94^rywavAS=%`IS!@{?Y(;9I4m2zg!n5ma@(QUWPds
z|Myp#^$-jbgm`1eRs7QQG}fd8iTC@hW5*<+tr1)D91g&N&8f>#L5Y~*)De~gtmn45
zR{&rjG_TAoaqXigG#<@LHJhrE#gJhh^yj@?N0-Q_-e#q#^OAtlVYMgA8`Vs=bV_IH
zK1RYTXZJlZVlN)jb<SF2C7m%~HWJHb1&;}e_QrkOof%NV%{Wy?W0*yrmi_BRcQ?Xj
z_ubk7m;}H#&$X%s=pB;?GlvketEQ8p5h#=)Nwuz}ldd$aN>gV(nq90b7K9-Y{xr_0
zZGIco*ulrC16^%pw^i`EBnE7w`5_Ck)Z_1LC1XpoX)=CYS!Wv`J7G|I5U01Fa33;s
z%6jcGSAP*4P$PUX?}h8BWggD9+@%}$dW2&?`UALqCJ@YLz?zA_%Qpy>g$aU!T)1~}
zDHiK&=4h9|A_@_L99hk^vjatY9bVeeAzj$Si?%9hxsdoF_4ifdVo?uaZpFO|lGcej
z$$K)B)M6*=6wF{&D?pQ+G^3Wc5Q>L51Hc1;c%YUY-nXMUo8-7y%-JWTz~LrJ(Cf?L
zQ~|%vnsc9}=rf<nR3~LIEYTjCq?jUodSyv4zV-~fxs29OY<2I`xKb~B9|n+QYTvCf
zUTs6ZVcm6q4^20FD2|fplG@MqOtm&wBGw(|=Ss6_D<ceu_?SL$VWE}+w`PDEXQ(O&
z!3BEB!mh~f&gx82tBnB7rlzH$MqKRMTmYzX_2Xa!3jW<*xG)ec@r`<9Sf^X*&+MSQ
zpcu!_&*079<*k1Bs0AKtO$8Vf8(3E;1QvksDPLS16a-g_lZv~xLSlwY1eV`@?9jQ?
z&C?8IQWU|mwdHnVq9n&|tRBoE;R`__oNAM{R#aZ5@^Yr&@I|~5v5k@*I?M<b8GtLd
zz{7`$EEuUJn@(8|C}}YEp;b_n-wj7%DE*8C24J^v!3LCw^_il{Skuq1Okq|d2b#Rc
zwOkY~GWbdDD9PA1NYesv#oQLpO!VRAJykuDCEzKg`kd(=uG;JP)R#E%1eL}urB&5T
z5EUM4YgKikw>3QBw(9z!?^URh)Bx-*1Stw-ximP)MtE{UPM{x^p>ECFy*1fh1M!zt
zJL@+ys9TfgV((ykTsaai=LFFxI1ykuXLd~>*HSl}mqW-DDi+&4c9huBCM_yh@n&I{
zVrUAp5mqojY?SITxy=sT=#u33>B*+(kZD0&=!|t@e`PDy%LxvDCGqtI0su}znpjSi
z)PI(b7p<%d#8KE5{jdrt2g5zg4SJRw+FZXU_HiOjgaGkDJ(aDw+;aM*i)h$5myouz
zvg!lwv{G72Jr65L?)(xCMkCS^4O8@OTTu2(>^AvVpSUZ$iq~c%oS~BKO1t%Q9_pZ0
zI3UMB63EG8;OdZmQ~HqP5&)(zwhc<mI{Pvd)_=Oqnq|myox69Wb7+q<T)DKvHHI4!
zkGEq)mHPQLZhx9nP?gi3sLr)m`*_c{TZ0aFYek}0UU?KI=`W3Pwl>*ff2@bs{ij~w
zp|X%Ota4>t+Ji1%RyFzZoy4@IP6oiz8!V9u6#z*p)%69(a<=-B1wf@C2f^S!)=6xg
z{x|aCGHiOb^q8q~^^y%Ikb8)#iH8qVgD+`9sRdb+&-X30FR#OIXx@OJZ0gnPnKDe$
ze#htWY?wQGH((T|X(XGA(+|b}Np`yEHP@?ewgjDez)aZ#2X8%8nMB*8?d6Yj_1|5C
zl#}Mf<5U@r6aDL-LqOFsz7Nk=lpz+SGH=;TfLV@Q{v4df;${KRX}C||bem1nb(h_^
zfG!%uZzF-%z5dDW7e>PG^*tgQ3gq9P-|G=~oiOSYZm0)0QzUNRAN}o)gaIXL&8gf+
zd{{QGkU!BH+kOwK4tjTZ2f&;XoRJ_*rmj$vWJ93VTNXf^;Q)1%d|URPJ0Ue@ac$y#
zCH@nk*5BbLB9d%1g;vb)6^Alj?|r@K0pT<@Z836Z`KXS;ciY55B1(bGwhNX~3gj4-
zhuCUJCU2BeZEq#dU^*-_B2?ys<-SaCO!M^k^xf_y`eJ<MCK=<q?qNLiukok@$6a8W
zaJ~NR3`zVQc>s~l&GmVyEu9iBs4cakN}%cwE8x<(X~XhVmiZB^vd;|<)0b++GP4+!
zY~=!#8E+n~zFW&2=rCGEZlMaSl0w`4P4NyatfHDZ)iJpl*e`x6-`Sqzy$}}+*6;G#
zo@6Iv)*`xkRMw)Jdepy9sm=d*MRz|)X4I*7o`0VQ01vZa|0$gXazptyfeA$_Big^9
zm`vysYjBSs(ns5O=G*z3d}4#a25LtpqpH)C0tzbQ+*VUp-3fZr{8}2eIu;WETx$^l
zCKg0lF$XRJ@^W_QJaeK>D<OU$lv6STLDxikOoh<zXs2|gJm!VQBxvG@=AE_M2z8Wq
z;~IAWW@V_7u4a^41X0jFf^vV+rx5<SqzVi9t4DBQT$-%Rq5>@jE2X&~<-iw~F4R1q
zN{14g3N!I&TGOOOk1~f0Q*~gxMh{l9l@4O{QA7UIY?<GQSD~WEJ|zQmTd5~Alrjk>
z9V4X*{t3h1-@no~Pcl?@T)2&8jP6mH@JHJLsA%Z#Tz{w(a#Wxj%^s~JF0OkVTpIM(
zXLEl;-b!e^>1@scONv$W7DVcAYSV9l$GN5ce0A7%&-g&Bw^{w8>um%cD9~yP5vqB+
zlz5t-kaTgOR-f*0ZZmdKzqTf5(b`3|h)<5A>qG{f?`;WvaiPT*#;I|}?^c!lX40tv
zp`z$Ag{&w{mE@3bNQ8;GVE78&<^3<`xZuRhXDiLLKDHWeav4SaYi>E=M_wGM-avI~
zrEvNBj}ku|^lX&(Tw8HAe$#gcKn`<fx;YykzV~kql@4L^J8yl2e0y(f$?9^t=T1vo
zi|abnm~t-C-nq=nN>h-F%6!IU`&np!1!?;+IncV)W|3|~Vm8qGzEB6}U2p`-+B|0_
z{q(spNmvdTvcl3Eu1(mJBBXx>g4#2&4He)Bsb&+gVXG0Wg@P(rSF|)BOMmaw8*<JQ
zs<3~(VZ7>=Je9(Ndv$oSD-mL)r2?Sn#A4+1=YK%8?-IN0j&zq$NI(N(J)!<p{XKgu
ztLzx~MvA-m3#iVaMmrYdr6rG1=5)7(nLQZjr%Ovah#BtN7%9g~|0r27R)9@JCQq?%
zO!@1m&G*qV1>J*n*b^u_53`>S`cjLa?Tg(cI)p%r6IVmZ&hkZ4WjXP72avUPa%y~;
zJ9XOQ<&(9E;^Okv%J^cgIb|OJs<&wh#?N<|A7%0o#Gk5xU(CD@4ovPdTLJ@HggR~M
z1Ce?MK%hqRFoY~PXC-@VRI^r}Q>zM=8S(Iz+o0{Yo3_`2+iH#1wj9D0`{9uRi-X6z
z=y^UlVh-_<YeY@#LJRjT<XkI(e=&7MWFKK<t*CEWLp`2-uqHGPBxx4`or6{_k!H}>
zZ;AnX=b-Jh+&^W%n#XN{=2&;65k{PYH=mTEE!v2cY-J_aYb_ylEL{$TW)H+D(6YHp
zE;&kFey9o-fJFjh<=o45{aAv;2fQdCtfpp_>ekc1qq}z6gd)3?>1VU-Nay8av5gW9
zhpn>nALf~k^@h}EgHrkdY>TVt-L=-3gs_YV$4^dz@I@`Z6-77u+VLeyQF92j6r6GB
zBd}Q+OWjy_-awS!Syw9p{|NDnyL3RwX>u=Dr@{8PrEwQS<JU>n5wbS~Tv|@0cecn-
zFHJq)n>C{S<LL4xO}xewVhnN<jvRh?uq~wC@Xa$pr>T+CvoT5mLn-VC;PHRmfuZ_W
zfL#1SdoIlW`7)U1*vw5si&IyZD_m`C`xTyrDdls|k~f#}l3q^qCGgDc+qdF;Jyixn
z6t$@=5EUz;u~F@{kB;QU*1_PYafuwYJ*p2`SI(l1K^(Lc&qGDr=_4!f`UrCsW+tD~
z9JH*Q4%z<Oc#j1*LG7Xc3kh-B_gCT4V==d1@6pSS&;uX8d8@fecs~gbb>dwPO!Mu7
z9>s1<&3Ek5HTVJ>xAj>g9+<G^eZb-yTs?U@=!2p_X*ysy4{T8p?=a`eusN%34S8cG
zowrw`ukEyG`<*F-MQH6UgAcLYVe#peZ|h*PHOBtJlm|G=E~GT(S@up+CU5L+bG!(Q
zXNCIR7OpqJ-G2{v3#G~LT1l9+<z2@8ms0Kw<tOw?!IeHEHK$F-2bMv@<u!=uW@U{1
zSZC-h4^p!7O2Rt=1sfC7Nq!b7=c@B}UA|>>kcLjHbM;%ehsfkhr&S1a`kLumC7;0a
zk|iCa4VVi3Jp){UlmW{KWp(`}L|J!zwquEGQ7t(KJ%d_AQAWY}Iw8$k)2GZyknEt_
ztY$NJkbz8S{UrSLCu=^JC;FNyM0cK_P4ef@_sKu0EBeJQ@t@SqIKXB_17-dP{4EaP
zQUbJO9Z5Oi{<U_^5Jr;7tiVEnydpDhZDVwz^P}(ZgNr}^5X#s*=3{raxnlWB{Ca(t
zib~B5P;hx~<L$Z4pU0u3>;Wz>dm%(!2rV&8WV4z*1`1te>IKLF`lKr9_#u6iWNOcy
zA8*e0dIYTR#DIN$_Maw+$$V=60Q(3&cmm2|s%QDCpOZom7^9j?BB=4DX8EEJq-gdp
z4R8Bz%~TAdjo#9J#_HQ&cFr8w#<1uYBF=9U@D&Roih8q!w2f2q0=`(kHv9=U+xi}a
z`jXdhmfv8Mku+^X7Wi1ZVdhn`ziJwUe$ff`2Ph0r5bjmgjX~Q!PRLIVKH_re&;kak
zl1Mk=O^ZBQ2(F~)=aYQcHob}_<)bE$6r^6UqhZKe;<|P|>H5w-Bj4s?3TGye6)sPG
z|MC8GqS4O7>2I4N5T3*2f52=s^;|Neig}N(ZzeXL^dLReRF$<LNn81T1B<%o%tI3d
z67zcONA~E`sdU%CNPw-t#>{c*mH^0L*qn;}z!j*pUWExb57spb|AK!Xtq<(E_Cf#@
zzo0((%j*LH`>%orgmgYY-#N#auaIxsceJE%p&&o!w)?l81|j|B0bt+x%bEj%PBS+z
z8UGUr|Mt%E8S-iW3LMr*|NcfLMiY7$*LRI?MW)AldI3A|2TLH7;T90j#WM1N4N=9@
zONgCMV3bL6o*d4uRT8)laR@(BilI@ZLb;&^JgLSy68$G&!uLius^I>r@}CZZGn<DV
zvWP|-M`$9HS`=>UP0WwBeN=vw@t~>tsqAMKhL)Yem?|kTvxobcC^6#{kr5~e(WCzo
zR=EvLxyNhRNQ-HyF9cko?b#8rV4{XL`!9M|0Gqq<$ag@gQ%}T8FEKB^vejw%;I#K3
z_PPjKtl*`ziWFo_iyp<dK2!dv8nSz+uQW#O(|f8H@E%wd9KSuF<kWiKpUpLyc0S4c
zu)Te5M+}`mGI|<B-F$#tl+#^RCSmBZ98ogLIce0l*28ohRRf^i6ev`MZrvYdqHyGo
z(^K)4KP_!Q%K1NZs%I*R{LeEx*d-+X`40zc*(wICjwOGqYzPVC7Tdd*$;;X(l|B}O
zxd(v1G&vxj_ggPYkf@6M!ZBxZgv%ndH-8)p3jHYCS#B;(byxgzU)m<6jH|xZLmsDP
zXn<%AQJk^74i4yMI4Go^E_EpecLA8IBkg1UncWg2#K2qTU5C4xPxOQR+tx2tc@bx7
ztN(NW!Q;&_9S85OomVv}rR$wtMd-_MgVgurlrK&HM&j`XMU88^Dfif5YNWDtPp{xr
zNvb6=0R(RSH+t&y1E`>~u0-dQHFIx*7G<hJ-p3*jw*nxc<zh8<$A^IsF~l2gP+XZK
zs#To{R0j=ac-~#RF~3jATUZa})5+eUqw7Vcn{{ntKj;$E_6X1`(2e&qJy;j^6ml1Z
zf0NC$TD61Fxm14PYHFjt(<j>UYW9?qnrj)4ca3M`6<m}_N$74mRpLyj(65kIb&s~1
zeI$4UP5~&!KWRC%Y((Sr+n<!$b80syZAepiq%*!s_H{++Wfiw6g<8Zs9qEWzq1)3i
zsL`t>fl_1}TS6>^-s7Df^R233Zqomy*@>`%?2T=xrwJlT^}Kd>QJ4Ygw#Cy>X`ql!
zHyD(g+pshs=UyASZ_zl`>*s09*9|&2+>kl=!)Dz1v8zQZEXkV&$glbWi3>Oi&s<5X
z7N@g4zA%7CyB4$eKG1qn(T59w(clW9?=1th3nPdH!lrNxg)Li#DCRK7N9-&OMhY!s
ziS@y(w?&;9F9gXotjElhFM1<<qIkCY-})jOO9pTWflYhcJeImF8teb<HQ>l-WaFa(
z=t(EgwZr$#*Z(={8J-}iWY2<Q+Tpj~;N-nE8hECT0u>w)I+XnmsAwfyU80S}tID8>
z`qgnxTI$gGai(z77qA5gc)t*{u?$-_a(;Qsz`p~XtS5RaTjYf$f*ZW~>)1_bSSmKW
zp3g{5j%=7N$Q!Wk$Osm%JHTsr_iTa=IDQWG0|p3$-j?cL)h?_0C7@08TTCU_FDwO@
z2&HTf3iR}UeZFl|W^HB+<=ip%#O(<P2!5Se=f*Bl0eJ#GuBm@(f(7H<Bog}koAyE2
zZJ6JO8`YNVGLW1+X0;;?(?cZ6Yx>I1CzHHX-i{eoxHLZSydn&nnBVCf?3MZ(5G1f9
zmgeWn%mS5Qo?h<+`OkE5TTGbl!AGGKLYe9zO&O!lWYtQebgOxx)farLS*PWUm*F>b
zwDjJq{#n~^$t=J5RE_A>@+@m})Y3=2u;<%SdlsbH8C#ZUA^Tr}<(v)1Ju`#6r2_9-
zR1ohxN+z3B`W;nA_&fmuF1~WCfK{b4REg0oSx?teDdeI%N0^<@uaN>oWk|z3t&)89
zK#A5R3vM#&6tppMW5J&f)jx|_WyuDCjg48&0vo3Zg3~*DZwKu)S`Q8{dmE4Kc__PK
z4nJF8Y1eEzC9Tt<mf3N?3ZBa49R^`SdhPjsMjUc{YvDrUXKQXl&GGcE0K#N=5h-=k
z1#OVTq%AMQRAw&a=DM0Vp&D#kU1sHEs89IpnXNciPrW*FjZ``Q!zZ_^E`?j!zcKrx
z)|S1!?>Q1hxYfC!jwA>dC#Q7l7X~et^iy;pZq%Bdt@4=NE==OaS2PL!86JPAaMb~O
z?z1PLji0=Nbb8`FwTvuR0oH}@CrK**T-Tk5yj=zP9JV~;9Ys%n_dAv%`$*;)+i;)L
z&j?DSlAqfq3`yx`)Z^(U2kf=cp>Me8t^d5O3j%zIf1NSrltCVcxw9eggJXU1G}gyd
z?BWt<F452wh8Fnu>ZVp?OVfQIJZy!R6@c48fRNG0KKo}Bc&YEY0c_PII-Cm)wRNKN
z4VSkwu8rwo6kB+%rXF)q7G4%}x`fZd+?XyX7$nimWN#gLu%!|`w0csQ&%lwZa_Xgn
zCz!>Q_*HMsZ}1XB+mCsGgk@mmg}o$*R4{hik{ZdORJat(be11yj>EZHDO4Y13dhYW
z>*(EyYFvsR$Q;Jq0TSuE2F3iNQZdFSc1sjt4@j9NPBeqMn29RlL^11Z!e$TE*hov{
zY6C)xrGb<AUF^d+p?IP|apDX*4(}*Ji57d*VgyWOJd0Ca6IIP@xQKHjv<8J<{`S64
zvXiHoG(eC&9`QYVv=FuW^^APuXI#qDji#<jq9WR^*J9X(0A1+js4k1RINJSf*mneC
z8SLQDsaN7Mvf{=Pf@K&TpV6l(9|Ij=*KP6PsRO|{DQdRL4vVnXX}vZN<a+(BCszOO
z4Fsx@>AWmh%7{d1B5;9Y=^o2db&y%Q=vievXt<<bgyMY#=cQrE7FXzC>`^W<%rxy~
z_i6z!gu)HWfD2u9H4FDY))-jHmXv?tSh<qR<<Y6?!^nFiz;$qlfh}n!{*(`auMR*q
z$NI#gj;#Y>CmJ^GrW!-^pwV|84tc$#x=e?aQwX+RtLo-&ksWRf(YHChRc?({)fS9k
zUpz~*vrZ5fH0j9|Xwtw<bGS>b=S|Zx3UwfdR@!kb08V0`yi4-n#k(PPpQN}f(<|Na
z{o`jV9k;o2uGkWiMF9=?$14eX?nzDi*PL<aqKw<_0pA?kG<E}GIODu_<K$|1?WGrx
z@5bBIiFrXQTuN-!TGliYypbWgw(+b3%C-M20ai#QKs%(!;DcD1;#F0QLC8P@v@mZN
zw@u7~0AejWf4voAjh{mE+X>r(SQs;*Ry^OP%Q0J?wey8%h?qup^M?hBn!lFXp&7X+
zg3j4p2^f(m7g?s-&fT%8qHT*-n}LItQO@%Fx5U~;Gy;pU?KHVWK#UqNH#!Fl?Rqkt
zEsxr^Ua@1;1=3iLI`(p%O~}1=ZoBjjM<%${0DG6YZ`LUBy``7Qvxp2RvjZlSIG&AG
zc$8T<KJl6l%H$ytFZx`Bt(K>g@#lVfvIZRg-v79S1!Ek|i!_<xTlA?tL9#pjI`vPD
zaG{jCO4Uq%VB!&!_2wb!-PglbXY9!hl`dxOuPcSYO?A?54%J+j%EalKN6belYS3Zy
z0fP6Ab0Z?N>HW;k?H#d_MvlG%h7=`;+lSipcGM=%*DZ`kdnRAM5+;$OjVxiqD0xpN
z!Ij1ybSS5EZcbt2r~RB1>L|xYKos2=uJFqqe5pniTa?MUAQpth^Ogy+3opPflx7Ds
zYO~yxYW6?s{IK4nba9xImOeBP>8-?90ct32&8v%pf;Ma#y>-3ok1FoFYSCF`(r-Mn
zP%cD27O-0IKGH0azQzt<hjl{Y7h9EJ(?wG16hH{riS?W66(>TY+OLZPhkWxA3T=NF
zZB|<MWR`#<{N(w{YiTJay%b;$?MCmq?@>JaBX(*=Dfa;U_ZPW9;YpzNEn2_x0Yigm
zGB*6~=^l@}D#jk0I<_GZ(6&bN!lc;ip)9IBT|r+9D`hGPrnZZlB1NtFHi&&PJ<{B2
zx@e5fV7?#My}m7t+3#v^tI3i$^p1h6Rc=r1MR`-%5vB4|f4f`eU~WG<WM%1W#s{2T
zbJCT6!9k!IysEa8-ORms)2>gS0BZFnl?|^aSPU{B&Y(P8?--_Hf5k^%!I`<IXfcbk
zmtU>H!r9c|r76Xcv(8j4m(JrJXA=53ioqFY<Ych2C>RwGNsy2$)9+RW4@}9XntGN$
zNkDiS=eQKcPUnL{gl2fy$+HT$;IX~I1$AaKmSKqV$AUFnE#?Z*jv0~~0>n4OPn!08
zmVNY>wqV@sFSJ`DmkA7ZY^Mi^ZkFhwP0f+a!4AMce?MIuwWQ(AP}No_QjZ;&9y!qo
z-#E7C=Ht}s#fUcZiB1|@(K|M1p}H(&-Gwicx3->q8(AEkK5^pDbeKQmouPrw!gk&b
z(8l`?TvZ#w9cA11d)-|j1Gv>$zEv2MU}G+jo6r0pt?PWd`4}kwPHjGI4#zd0!HLMa
zQ)6zyR)3LN5dO-^^3By1=B8phJaWI~vQ)}oIgr-+fm%%w6HaD;<1A7tefo8Ac}{EU
zjZ}!Q*08u^o!2{Jd$_W=vL&FRAgu<vqSju<$oPvp@M%S3J>I4Y4UoQ)fGZM?s*&y7
zBkZE73<fu0Pv+GVLR8;YXo;p7TP0Q?9LVW~=5tW%TVf>>>XU0HgVrs_0dvIgvA`qr
zIeB?>E}z0O04L8dxLZBY-2wSvLuy?#<%d0?uy0264_JUv!ls_T`ysc{Ht+Xx7o5T`
zQ!APly7z(<AN}|213=pmx<}BqBnN>mU8zl1YKje-x(`sLBEhF3WKaahhi+)bp7J-X
z(xfWc9}3zUzovQC@e(H;O*3!$HEH}g3?CyWkv-huInvu?2aiT*RWmw=r9)M&Ype9_
z98|=mFBq*inNovF@Kp2Ex1j|KwQ^5<-%WSs%B=SX4SoLw4d5_gdHJxZx>W9i0O;Gt
zJ~;a=X^hWnj88yHXB*o3Okiye#1CU@Krzu4h)2%0R)AC3#jT4^U7HH+0*9^BuAbuJ
zR(y&GH{Jx7bpTej&8rJCYvpiUo8Q<gSFz~?LDOzD6k7-(Q|(2YwFf#jt9=r7bXwzr
z=Q$dTx80V708pt`d&-hb$;Q*qgNqfN*aihKN9L2*LCUciV4Yj2=8gvPmT|$|2U0(y
zSsL9(qqRl~^BSi*s;)Rd&OVMet~6!D1^E~H2&j*$HWwU}p<z#Rr#vjf>3end5drap
z1|_wp!7zvIc@xLhrsW}cPkl~z2pseQ!zKlFptD?v0Cvq;(VW1#HvO!<`|J?s7Uf(K
zrOH`shMYV$Jbj6m`@Y!k>?ZJsq-&g>AA&+%18{0Lm$awOdL7U&(~<(qK`w+1w2+R4
zo_>a47pVq0xE|Wt6jNw3MK4KW{saMpj-hfWj_2<D;=+;+w*x4^`F`JA8&50G(Mc0z
z^zvWT0cPNi*X~B8Ss-&Z&2HcfiJd8UQQu|0(t&B;Iwu)#i;Kt($_Q@{?@#?(l9o;d
zwidiRD$WD!^rB@*b=QrLqTzjL+OK-k_<ohQ1S`wBAALK^<{dUU%hH?Ry*RlZ5b`w*
z$!oqc=Ddu5=Ss!u2lT!c&?<qB!bbl*F?s1u0=`H|dQ2;kV)5$5y~@E2O)s(Mc;=kf
zs(e9R_a{99+8V0~V|Lc<M<8oM{R5xrmL5^Js(Z=8B7*+dd3O%M{e4RbB3XLG4NG4|
ze`R{C<98<&`GYvHni!61RXLurRG+v1RMS%5+-4^@taf6<Byk(6#_Bx&Q%m*Z>@4+-
z3g`h*oW?NXA_@UCfIo^68Uyo>+R_)^>dz6YQ_={Vs&C6y#bY@AvtstaD{n2j)(gP(
zeN_{VvF@Mz(fFn6H{yp!s{?9b8(4T`E^DML``1YHkCng>0#ERIw($Dfr${>ekRryf
zzm12JILYXbeFeX|m#b<tk~rDFv<KkJ5dl-ofx^EW{-O!(aXDkWEiOw><Ic^c@tG)%
z;bU=Y#o)Zbmopz>J9PStm25GDHMchW*+02@r}nRFIt%5>*71wFa`)^jDEP*efW}zJ
z-_3KGFD1M*pRE7XV&ta>tg&qKN9T$Uo~N(7SXNKvUY3&HVrpc_=HmF=5}pV&!YC>K
z)Ss>l7TKza5gP}-u|3HoQJMRgUXJ5KL){kIOp3Sv%&9btUt4=}-@^_xJA9X#cQ!PC
zV=6_-yk3Htes>J_(nb_Uyd=yk{(#~l8Rq^cKQTBtIRCHW<jg3l0;9~Bq{jXQ{E%Ts
zgS`ddG*SXC$hf5m#qWg}w9Va+ZQ{*!l71~gn4D;)B2a@BvQ617#?bqTUg_|P9xC0R
zO*Vcg9NdyqIjV{^gQi+aG$#v$Zrt2Rr5td*looi<eck&NTJgTEuL~PIqgcJ!X-U)D
zLZZW;azqlenc4Y5vN~n?nX%=Yt^>JbDOe2{Ax)Kx8gECCRFPP5>Qqpbo6wSfhdRJN
zd(z%^*P>r48S0Yey^71<3gz#vRQJV%lKj$W5^xQrK7c}R^@_e}6_<Ruk_i0yBSEJP
z96G_Dfh<Q_mzsv1Vcx=^=0@1H0hWxIP`tpE93{(2<JL0QG~j)kjKK_X=x=S?u!;__
zS3Q94>=&V(BNy+PCRM*}F1P-rmxLLEaoapcy*z!v=b9%F>wkNy<b-Cu6tgib_S6?4
zNBp+Ag?dt??|{hRo`f+7U8)L;{d?j=!o39g$ymLt8DW)YoifXpn~TcDtn6O?j_6({
zqeJv5!8};FeChMK$R~{`IAQ1Xl`I8N-<E{wh6u3(^qIa6qAV{+1hfp=Hv9JHV4Jd1
z5n#65@p^jZ7$?PpO*|OjM8aRJV;(y)B3_*FI~(crwqD;b?{)bqTfUC8jR&e0p0F_q
zKOHt+AwDx^D`g`?8CSJaxJpX`Fy@`iotol2aT_QbV{#g&G`QdeQ?L*=U1kB8o%}Zq
z<l#d`7|4b7?QnJf68EDOOW6kYW|@XS8K-jJy(BTY$euuLijMpZ@l#XG!s~&Rr=KiC
zl)|ZRE<ZQSaobCIM%~gzat=XHAAZb8WonLjoLv)~FiwU)MaYkn^_%d*fA-rnW$;o%
zJas+kizuF%Hc^43Z1>BuZaomdY^(5<W?u7EPVW5M!e+T(DD*(OydLQux<Rytmmcz$
zKm=2C3BG8>)vue%OmjYFBelqKANhvZn537B3}1PqL|xY&O?c!a_vjZJCZPE}XJ;@S
zTH`O+(^_C*nEN+mHfGg<I_oc652?f;a4IVN;Sn*bCD9znsX*zl2}VpnZ^sqpDRQ^V
zc&YaYmE#BV7!e)V79$>M6f5r0y7?k$mO0*QZtg0=6lYMtW=*ykYpjpA1_cc4J$3nD
z>*jj-dz(QCTMB9Q^ZeL}-T?lC+yKI7X`bTgTa*{=08EdPpZmhldDf0r^R1aVQHzxD
z%f)U_R+43G15;BrUrrK$S1u4;fNJ+Bo1Nr`RixcY-Z-_YN}49%Wy{wDfjeSJR|)=C
zn>HMn9ZXn+yWP9$(~3{N02XAN79zJFW~gVlsTs;Wn#66`gHO1#_EXdmT-Va8_j{rD
zssHrwsdtV*ig&t5DO1iY!cK-V*tun<#_!-@^+Xq0wckZyW<BkI%85A6kTQ4NQdWu`
zp8!6qR)P|7H4qYFh}kQ&cwh{x0j5vYF(y82jj={E$~UYzEBd)^@)@cOOg_Z9IV)15
z+7A+i9K5eANp5?W$A92rIXsvhbGD>4@Rq}pX>NcY97lwfOVIN48{~rO?QHbee2?&>
z$Z4tl69lI|D=0a@$iaZx($`}_!J9l8t|mc!FKUn@gu@|XzD}59xW(djQY2x)+}xp@
zO)u%p->uLn{i(r%eqxRJ5590qqX;g4@o*5)mJM=EcHDIvif1~}6NB+f&fdCwtlXn1
zU9@yZ)(D>71J-$%E2}*>(hD2!cOn5B@0^qqemXzv2XY|*QETh*dAN5g0L~vzmZB&Q
zo4L<Oc%G>aN}-vZ^pOBJ0-7fLd?2}LG>x$sz7&FMERDQ<iTh<b(9k;E8T}ip+G=s8
z$Wx0pL(J(sE%Zrwz>=d4ofWSKiuiZ=)nC?Pq(^gBTZubj0(Q{x)crzY4xI@rLvC8y
zFxgXKaVBqoH2qU7geP2MdxKCWO_f%6iA?EE5~ydHorF>5`Vc3}8fi;b9H{%okp*S{
zKZrqgtFb)IMUPd)PEU7Kpr)@*dFzBs@#?8!HTX%bzO)H()$XBY2Zeni`Auovt4h5q
zoKkRp*$88ZxwecZCd+154GQ0Omo|6!aU#;*z2_LPCVMgQ`<W7D>>B}$=a>g;h&L+=
z8RxHMH!Es=4xIRP!fdz;{@=}Gb3M!uY_%>YxnsDI$8U8MPNHsFr&19gvHkEH10qM#
zKHftPAkJ^R{m}FwsV5qZ*o~WqU&n0mcP3Z2;4?;0w@#*^0pPr5C>C3@>6?olrInQn
zij^OL%IyBUP=dx?O=Q>06t0ii%?-vFwfui(yoFd}X)5UCSQ|a%DDgwJ;_cE%p2vtx
zBo?MR(v*^v`@@Bn<`lA%b054KYi-q;J^JFaW|dP@qT4E4BmpBv6$HD|X6b|>x++@%
z|Hkz;2?WtF*4(`HI0&Ip%ds)8!G}LV?>I#O+(Mo{Kz8=A+TBCuef2rFIJk6E_K_N|
zwT`fnlk?)7$%z@+H`Ip<wUZj-AJE@}Y1uj2EFZyPq2PvpbPQdxrM5LMrz4+;{WrT^
z!&P=`=1Cn4c#(ln$Ei3|{*bn$^Rh~ol-D@+)|SWkFH>~OeRY_EO3Z%pNr6AQlD1R=
z$p0Szh(LG0!sq)&Kgi|iyp<#81HBz2040bXBLF$nWdd}|Si8Sc$(2<d%1~4ChtJO}
z&o8HFoS(5|atYn4??)Ia5HRcAum-{1t@dkYLPNhRd$an-Fg{?UXb<&e*#F}9yLUE%
z^KtjiFs>|3Fg?wn^O|=h5oLT%QEwlj=u%qJ-c5f(&3bSalsGQ!ZmxXYom6C)w|93M
z3sG&`j(W75PMOLP>k&pb;bj<Cia=O0V$eH8mdZ)I-`gfJA!zbuUhv;f#~$I4aJZ3i
zMCi7ongRHD%@M0$4DI&b*AX}$zij6*;*^ombfV^r&mDzuyGu%pQY#s7+_X6Ymw=H@
zBuIbUBjfi423Q7SG~1R!2oN*-Hp+YlHf%HONA?bqvNDWvv%Wh*?+!b$(82p&2|F>f
z_M*NifFJu_Gik_F2)BaLTTI<!<%p>x_Cxj>naVJ37wtLp<?!ZNOkHxsx6LyiBlMw!
z2E`|1|3?xU6vHr#Zfk!KpFYcH?A{&p7<YeZ=MRJaFBtK*fMCSpTR86LM|ho?&*i$g
zd6Ls<$0kWyhCD=Z<5f8q$e=st!^2RWvb!_GN%86U=JQ+Di?L!zXJmh3HH3TM8AfgT
zL+E$YGpzOfbrhDHX9uJEDE~vf&jS4#AVp`}Vrd@M(;Lj!mBBn;95tDLW=59-t7Csu
z8O}S42fF3LHjVx1#U_3+mK11znu88q(n0P0N=s){b;!hW!&*yCXAbV2M<px{;#U>_
z|IVW>pYJWVRwQAyhR9{%jtH$lIpnl;@r(x%^ND&ujVHlf?9QvNp4fiqQVNx{2*Ivi
zN>MK(2-6i46hpUq>va78Ob7O<9XWrEwm02%<1AM+v>sx2@V}hvK7Nwkg4u;Iip{V&
zy0AQ^OzvWXK37tVNl37!4C^aS{znUvtY_mTnxBy}wAsrN{Q|%0L(v5#j-;9{bnvq)
zBPL#1l$WA!GkGx)<_)fPHt&I2Ytm+=gH~#6t;|Jl*6OVaBIvk6>zO3y?PGt+FI^L4
zT)FSSboR^imxW{qPd+HMUtAZQdqfvX+o=s<0^fuf__ZVLs*Wq(-%z10#P^|R-EhV>
z`YL(a#e8#)u=xTWh)}B3&7QLT(wU_r;T|_Ica_EX?6x%Tfsy0d-DA^TT7k*aZt@d5
zJ+iYYaCgTG-DJ`Z4Eh02W3hkhl|=TKhbq{$t>I<@=MDHJ^-wx;Zk$O?7m}0ZKB{t8
zix~uY_1>4+(?a1hQn{*kPh~ycw1c&@&e#{;-#(0c?^Xk3-lrS07Q;K`%TZ8e94z1L
zaQ((|zDAj>-m@0^D{g?2<yw$0)~fe@t|QkQE#st6{f$WQ`+Y8B{q27^rS$}fGHR_R
zd5Yxg+gLi8tWR2u(C%bf8h6PN;A#Eg&LKJOpzKD2QoGhK$#FJ7eI4B4wHG$1_m1~r
z?58(hiuMy<`J9SY^whUAm)+5B=`JMVbKbKdTlTiUd1=m*-knX!{U-em?}kDbhI$mn
zQFRzTr>MW&X*Jxnw7h?fY;=+>NTnwa)v23KZ)%J*5Po;3d+uw@*upd^z&Q;xSPm2H
zVC@@4$VyUZurOP+jH`C6Z&T^kp18qQt=~EoAw(c<T~Rh;Z?A?q4Sl#xTEyj$$#fSg
z$;kt@;O==B+pfv<0wTvB;9`}B6-ErK0}4O1_sBo(1NqEJ{-=M7ug;YY72chvz^6c7
z>YEy{xyy%<?~`U=o8C)#!WQo8_jJ*gu6M|_Yq-#oVd&fQ4|^mcO+BH_#`jzT1~}q8
zQiy3mg=_ljyCn;T-Pi}}$r9eMDp3QSX(@79j4`|wW~y^nMGV7<U%YBfc9Nt91^2yr
zOswt#g%%`CB&8H<;d-UFZG|+h{qg(u{p~+qPf|~p0dEcy0yj68f$BXIm$(B!D3_8r
z8KD9Gw=_8!1OtDTdX(oByz&f*gyF{tp7&zf`O@&6_DKwK&BiGujL9gJzT8IVvjkL7
z6S`EB&GJ$k^%vV{h{|Z1FnUE4-`hsn={8d{k_?v`DMU~jK$0A{QefRMN{W+Ss)4(;
zXA)p`{#q|(tR93P&RVG<R*Tb!p@C(lS6V3;WjN!6l#_o}%5}qK>qUS>73N)Ma8-$c
z)7x0~wD%lh|6T30_@ra7l&rmW!`;+Ll3!W16NJG;AT^zf){Y$F2=gsLa;>d3hf24s
z97?;b4Xsw_@x86B6x5~2G2o{0ppbWPJ36_|<K~_dfzy$a>pShvq6O)Y2m{S&xBw%U
z+9;`xdklY&<2|zS+-6f)9n1c|#*i?RHiM1FDT3@4Aiq-g$9BYmMZi2adfyUYzFIyz
z4G9sHTfjBkD-P1ox0^hO>l;bdupgY|^|Pb%dVzWR^Rk~;*sew07F~0K&m4<ymXTbi
zO*{XVwVlQJ49vgbxu;!l*50XCXIrdngUuVEo-2PFo~*N#7L|o|@dMPR+<B#oO=;8V
zjv4cyti96Ciwx1u%cP;ZyK3n}#5Oa7k-MLVDKbg@$&uZC$B}(jYzxBo5V><6wIQ1u
zHEW}dI&Hnav5s3=P%c%0#Wj!rJt+c=3T+sic?PGvQ@n_QQ9(KHgj1eNIR9l>TBmGh
zV9I|775Qi`NKU|1AU=n9Npg&)5=LZiFbShQdj(A8@rrlC4QFVIQ4|C=B_x<_FQF+N
zxU!>^at%!l@mV#0yGB!qf>=Mx95fuGsjxS(a(sXQhst@Fn=PCe3EXUpTwmor-cB%>
zR*>S=lbnOFG$`Oi7gT@#aAXl8s+23EMfiU~jCC;N2?V&P+W6>K4Rmsj<uuiawPDr*
zDfdeV*fYd%1W4}b9ff&mh;CHH9-qQ3I|hu1{Mjko1w|>;J;|4AnJM#Ym6@&|5*_DR
z8U<Lo>Tu2FE?Ne6$oWQ_4%ehias-q~<rRnPs4PRpQ5Sohz{Yd2=8WN#^m;67w3mOq
zI^}#%GMBZ@YKzDzW>_~&<_u-c?hRK0S3n^o@8Bl@xOnd9iMImmch4P$tEyd6L%>lj
z#wa=n@Dg^x%L>Hm%OicgJB!e}_&^Ixuj8ilYtZSv(ksX_x<@E8m-%c~?C;^|t%J_K
z(QW(fbfBoO4;1IsG-I1pK5#3y<d=VZ5Qm8UQ{>@DZu9xKIE8g~mwe9;BG*Sg!+UPj
z(S9nU3a1Yr3#KAuxf+iaic{Y?7~{6Yt$Z?MGm?8J7~{@ET!ajtb}?R?ZMpJXK#-ik
zF>e$@A4zfo#~5QG*LOqdbpdw?j|EgQ3P+s4F@|ES7n?%qInkxzR+=|y@KS%uH8?gz
zXjK!#8jktWgG_oNg7JIcSYX*Uc^(xtH=edhNJB`?a-qzQJ2)mLoJ?g5CvknIeFCIX
z%T6y)jJnv9mTeV8>4USUL97#S^ARLHG~Djud=xVNot{Dnc_1%5_2n38F-byn!gKl*
zf>Mgdp67NOIq@MwxS>uDzn*`9LC#bdDD|;0r&Z8Sl6)(W;l1EtNcYu7iM!x|_cI{P
z!=m`&+u#K{L_pTmPC^#PAr1_k(T}f6Br0g{ONH)b0_>iQ_-+hUM=u<>p`R|&+n}g$
ze*fzbGvzRDt5^GueP5;w^HibMl10lNbyXhs^2o93+@y#8{y1<-zMg;mSo?j`bLOO%
zNlEF`f}>f-XfyEKEx?Ncc?#{&?13wU&9^mXnjVrObg#~3xeJFjt3)w-#O46r)mMQa
z!ew1&7+R!mz7y2$rkyp<5X<`5rF#%pyHuzA*r0+<bmJB>jc#~1u5qjItG#tp{XoO3
zk6I?Xd{-$$&8<Xmq!2dd3Lu8eos+`et_lZ3ZyaWOR%Uwl1k@a$$#J&dLe>p~Nxqu&
z&T3E*!Tb(o{r>d!@IPOi=9jTMEfWMYH#s$z0oex=0XLUHX&)z-HbNPn0ok|KLK&t4
ze@%fl;mSe~MTvx$6-r$GgwT5&{bvovrExH3?vR67NUuPejB-7cmGD9-O(fN2FYlY9
zL?FA&vb@_Acs80}?6g7_(jd>CxE@!PQSX&`6UNKr71O*F1@Ix%k0UEujGwMss5nb5
z7>^RAL<qV{U9<;Pk4%!hgy}3oi4Z3?e=9^2lJ*{uty9|DpS={$x!qD(!eO4cXpv`%
zv`nQ2f>NnUYCGeAl<Ectr0L_SI6ufwP9SYFijqga&f5@6Jro}a<gh&Zjqn|<s(WJo
zCg>}YlqeW*?9U@X$#L^i=4F|7csFm`Djc~mab;tw5&dR?&5}OSnoxRcnVu#)e};Iq
z5u*&0P0tu&P3iRPjBm#;$tqfhGmmL9@5>Yc=akx)ldn&3)EY*z&Zg%YpBbW3mHF8(
zh{{=M8W`Ct#BVr0F=|Kd6CzyaXNuj4W9Cs4uj?9{#u5O=FnFr??MRWsQ~l~8w0X+>
z<RQ9LCCS(s?76X8lmmq7_)&O)e-IC9NmnxECR{852O5oapX`cD=wjAu4^f41C89MS
z1&#&f%vF`8Rm?t|o@e#9gWe{nWSgQ;sINiGT4>%Ivvv1RInmUX(AdCfxaTHEFeuAk
z(f^DpiiX-K6X$r5&=Q}{?A4xlt&`y+@fP<*PdD%&8}e)AqH1MbrxDW-f3aY)gpPq5
zq^PqXO=wP((EZ*(Ok`Mg9TWxm>op>2CDEoPI6I|cT^DbqGkF!g_DRB?kZDXjyI^}3
zFf5H``=Y{II&|<){*>YEr|#SP=hOd2MlnX5w3pw0JoyJwFd{%aG$jU7vOoWG`Tgl1
zJycxqw+!+(s<l#0{>DJIf52c$UpOreGb>D1pukPOdKQr$NiX$J5a4D62`|%{FwMJ7
zIh~jJjTaEsO%@Xj{i)KFSiln4mWc}^G-1<>U~DoA6eEn*o`={9NapBDN0=O-dED;f
zYBWijdb=)QwRpdq&k#owsa=l?lMUwgK~%)>fx0ZyhafuQj2$4Ve<^plkc109aQA*0
z<dPF3*&}%)HF2PK59AF%;%YBzlp=}_M<Zz#B%a%E$-JLY6$y_eZ&W3RS8uD$v8lQ!
zL4>nC#FOfR++d|$>6(eqg9@cws84j*o)H>t3A_e~^f*KV?wRuq9P4Zi{=*Ej5Aen+
z0<T9>VjUTU?F_S>e<k3hcd|x6d?TpnXG?lFYlK%DHXaE?5cF-nHy1>iftezBxO1ut
z_Lf<JRd-){yrRB_XoyTZsOQvoZH8a%aTV{OICQswW>xDra7NXV^0W}oNT`K#6KsHZ
z!iQwF>>#c<k)(3%<m5x{7W=F~4`WYJqz>Xdaum!F!3So>e_$$#R~5Pt-W^bF@4u8D
zh3b@G2_GOcH_*9~kQqvZ8BUFMeb{UO7^ng7N+fL^?2!5{v$UpQ`UhZ8-=_(V;J^mQ
zh6KleM4Fu7WC)e3Hf(3%^8_R$mlg~@fc)iGdiggz{23<ufa9+)^+(x;UjBRem(QR!
z0D}h!fEm%6f87F1FPu2*64(kt2^XBu-5_>i2Et${zYP{d!KohZ%`DLeq5^;m2yw@1
zLz|~~75AzY-$AKT0LJPkRRYp}g~BW7`!G*6^P<RAel&trIDMEU8?5%xpcG%kYAu2%
z$h=O$>lk7euNmuk`#5qK^9_zjOAI)h2Xn+HHU^O9e}o!7HC!uELX|%jy=aldEm`Fh
z&&@7tldp&A-{7kNd&cz_@l`^?$i!WIRXg~qW*U8vuZQX1;49bQC(IY|RX`wKdl8-|
zbD0|}J-t0_Qurx<5s$5Jgr~XqZs#@Ly3LmgICA*8)mScN>0Gl-S5A1o;__0ieJ-_2
zDKxrDf4ZQU!J!V$Qn%u|>5j2f_if7PEH&Cv_1K3bf#PMS#2h}^Er}7t8hgs|-wR@A
ztOH=RITY*kLDh~6;@m;1NySjM&mB%h_q5;QeY0!1FP7Kkjw|jK*1LZiU(zmR`n0yl
zo~|YLrU4Ll*mE0Bv7E}{rbJwi6ss7^Yq}{Ff6T(bIqw@BWdrA~H}6+LZUM2q=~s1p
z*69uvUAqa9$_C*k;&4B*z_`Ix4^iH@(-mjk*Jq!#gSfFoOMBx2x18O2x>2_1dv3B6
zGuBRuInX}HzB*hsP6{vyrJA0C=5Cv6jA?h*$TfY(#-`>QW2v#rN1M9RUgN$$#BQ|L
zf5v7mYSrH4l9EJv$!vEx&^lV>C<BWI=|;E1*+v?(28`=hHPX1Sgl#dSWBT3wvs86!
z_F8M7rCwVz_Y1mboEs4TM|ICSXtxRy{S`-4M_i5-`5E`8-l1@&sK*T9Aiq7tGflJD
z*y|y#6!lWmuA43CxdUe)SoV=Upbp}(e?u3vTL1OlPq_bju8_{DBAh?EklsN(#)IM3
zZ%W-z67haqJyH*>!*s2`q$QzjpzA=THJ9)Eeb0BwZc}oke7}=7sWjiuK5!;Z5aZ;|
zeREyDn|)9oa!tP9TdmaOmG_+wkcV8|f1EtzrrS;~f;F}4f=xYq<^N=Z(u}~if8ReG
zokbEtK*%o!K~MLUFe+=Yi11#N4o8BU|7{AMz5i;{fH?`ucerUVn5xsw0{<?p>w`GR
zmK20`hnfWpB0y`upjqH;yP>C=1$TqkiJ8>I$S-OZFa?q>nr1<|a)-mSzVKR0{^j`{
zrT*gtZ7@ne2xY&h)XzNf=8H=Gf5$Q3;0Tw*LQwf7rBlWsSz-0c#WbE7%g;5cxzT!(
zkV|tapW~F`Y`55WnA{CELXK<o!z=p`qCw8bAn}0U8ac9WO2%r)(iif3F0B-lH(kh1
zb{-~si=FVqm{X?qUJyII3-c+3`9GCs4P+CPuU2Q9U&8X$A4k}}I$M9_M)2DIcwkyP
zY<?rVT9IY;j$gCqJNURR*xt(bGOez$$&4r+s&2y%9j0B3S+%YH+lEw^8!N$g{<{g*
z26p)!bmwno8U`_!k>D8=mjYii6qgt39w>k9TuYA}xedP0uc$vTvM7mq0E0oM+inAU
zbK=`#4}lXq2Rq1a4*T~*y&qLwRa4WAT)aSUO{*n}`bbeERWzTzO?>+JaQgi4?jqG>
zm_;yR`uueQ`hnB`fm5T#$pA1dVfy@I`t^x(q3Ls?kK$ZP`WX7#(i5K_C0ObKFZ6$b
z=SQO;^_PM^K!2b8_uMax#`vpv2^6GzN|ZT$d`|;hdYaS^XjX8~zkUAYfhYDqe0~_&
z)j~3-VEXdo;rEF;ew;$@2hOCkIQ=<=W(Jb}n>aIC86v?XYOAL|sN=i;eEa&BFMkI{
z!~W~L@<}m);Oz7xenn)K8|$X;>AQbrfHdy}%n{dW($tEMI^=<+;mh=(grV8dHm%=%
zt7FP4A)3Jwn6!pEmI<1s%!8*`TJwD0+s4dQog%EMu;IUXDrv0$O5^h=|J8rm@3||D
z`q%XEb$a-eP(`h^p)S@6iS*RQDz$_vFa?H`szEk4NyD6RN@rA@;fPmS+x35zM!`sw
ziog<<84k+JCCygWoLic8=O&_+DdeX79NVwpFszkYijpEjN8>cp&UqDugl<H*{{7`g
z8uyJxeQchQ&LfbPE7a4YHiWYBq{3E4ySSB8@6<W%WWRG2?MReiPvtq5_EtZhCi^rs
zyu?L4)>VDljRe^iM70Bn%+-Ga0uq^WSRlA!&GHhIp=U8>w0z~BAC*+}4LulSjL_y5
zMJ7j)IYLKIGI9t3s;iS2#kjy>68^F=NHef<7=)&9rQ_Ub{>+(9yh0SB(*76pzZk!Q
z$hW#A5&O1t8H<UcOEmCVcvJ1$_=S3D`IN?BH_jw3Fi~~s>3Mi5{F;A+n_cFp^pkaF
z&qd(sl4jt(m(0AX1_hI;#?>#++HtEBT692#Ri4w_S^YaYaXs;nq1{b}jI_{OJu6Xq
zOSaJZ*N^j^l9V#**>#z;$64JRWV?+WsrW-x36F|AKZ6ucMP1kRi%vPR#sjfUtrW~U
zIgSd<U-Aha1aGTgB$$7!I3o8^A7a12=}Q@Dn$2Zqt+tFFR428z%3t~hC33G8-Mfp0
zB?QPV@#cL*Ab2f0id6J`w4%I*tMmuAG&5YiLg*I3gu~5NW9WFa`o^nKAQ?A$K~b-H
zX~X`wa+K6#As!|Fp89Ud*L!T7y&ayc=9+PC%i3Qu%FzJG;C+AMOVX+MEn(vIJ{|&T
z*DRy#_nP2loQZ5$q>@Zh*hM0nz$ATMw0vz}3UYJM>Dde~)u^b`Dvv^0=3vHS&;lKw
z0MRS&%p_0G_xTW5<&)ffCTElQ(?u%PO;@09PN6WdNav49TSK-cqV3&?%O#VNKn;Ml
z=SrE4uY=^hNvVJ4Fz=wRIV95v@BKG#ra`ZQMLU6^n9at|6)}(}QB-h1ESlCZdvxS_
zwMR!@30a5M)WEXIHG^>t>Yx08u-0_iHzu2BBi^X;DgU~GvheG5V#eVv4mz}XhxYn`
zC@e4|%s6X7JEEYHAxtYB@Gr+-<`a0gHL6^SQD4LT%}9SL0SqYA8rBpzgvc3BLAzgP
z#0s{tB3F=N9=C)Q(TIl=StEZM(c|OrjST<qb9F=to4D{c+y4;bR54L#X6Zoq<Db*l
zhfgEbHd_$@;966p*IHh(H<peJG!&f4x?z*s7~mF3B&${}0<~<hpW*46(T=(kF>P#H
zL`eBiCjWmg9Nk?@<39=|n4tr0SjKRm*4~;qLTKnM_bl?<+lmaICXS3N@NH1yY7MRt
zs{mht>%62sWr2O%yQoM)Nru|Y6v=AZ-bItfu$I;t=Ckzof^zx{0S}(XufOr>AN1od
zQ%D_6e}=9<$~-<x@;9%l8-y(>t0zpJ3g|V$d=h^wW`jK0N&=;0<k-HR#7@qD#-Ldz
zOX}fpq`SHuf!tF`#5EH&s`Xa#HlLtNw<GueA(?hSl}@MUMec_Mv>|v7WIgPS1#gfV
ziD{0EcH-Q-EJwEUZc+3#Lv|N}o0N}mp8xVAy&37kX45AgVo$M<5@~33UVQOix6wb0
z(mH?o$^qlj-57o4QMW`~F-a~?KCVyEsUn-3&=;5tv(0W!c6kbB!siH5>Oxd!N2RC<
zZA}<8D4WK0a@C3VZd5uPZIw`!MLMZVTBM~Lw1YGNl>d95)Q+ROAvMbxxA(D?%1S1+
z{b?)TU@H&HXG84TF%WPkxf?-VVD-6@&om*=iAKL2M7R;#7mA97P_*4!Hwv}3<#HIn
z4Z--UF)i;TmlahRB!9KpK5R@gHF{83N_KOnYz||%Asb~V&{H>-4N7L5A*Lx%anQzj
zHK+~f>wM~Lpw#V$um^M><TQQNPDBr{d;?J#z3KaCJ28(^EJe<}zcr#=5_UQUec2KX
zIIHC2$%HCTr@CoF=aE*?_si>Oi6y?y)h|)DUMn@pkZW{bc7MwKuzWV;E}}DipPeKu
zLsM>rb9^3r?sP-5fRK&I^gG9!$a2MHuSsl%uIe;EHQs2OddN}}+jH%d)?rjPqy^FW
zzK^uPFb&w=ECi%ZjgFh%s~)5$=buK3&?MgZaADcm!%VLkizzY?NlK^P8sK(H>@cz$
z664YHzK_JXW`7!Wx4*Lf+QwhQn*OcV(n+e9^j*aJ)OaZ#(`h<3&4t_;!Xr~Q&R<a`
zR~2HMZ>Inc1HCRl>7omLmxEC{WZIeT3jEw;{dIbCMECxA&RaBIN7i|j0nevf^1LDM
z=F}&U&-U<=pZlOb4#@`8HaNw1*(%a1W}4*vj@yfw@qcRi`^)XxKNz8%dU|+~8`zoX
zwBAQe4S`rw_dnAI2hs<}1+*b}jK1}a1&<6l?HVnZ72Wl2G1?^}y38Zf?%eJ(2A6px
zy55^MqJB!r!Qi-+rsmg@(RkngN}<xC16hBn+_x7|B?b;xP`bHO+7k0&acqcLc~|;Q
z*$9citbc}|`}-~X`>=pE1TUjkep|sy!BBXuqJg|@aPQmUWfqXRcp8S3{D&M^l^xV|
zo5f$Xn{Ax*^PF0~oN2CX&XM@SymB+{o3n?J#XhLtDci#`*pRJ={`+lZOF=gUGSa`a
zuF7i_IjXMqt^o5Avl|^SrwTJR-CCHVZ(d+Fsejukenx7Ic+FgHMZEIW;>DG2d&PTL
z3>)Hw=-uDPW`kwUs-HIdEjIhGfHnlrW3X`_!TU*sZj-6yYBaf2OY!1xN3Hhmjl)Y>
zhe)^>Qrrb1;rx%L7Ca2E7%k?SG2cd;F%=<3JEP{g;p(#r-sLTABNo{On5X94#DQYx
z1%LAo-5tUp$%Q<ast5MK<1U#Wm(7OEZHOjr{8|F}+jPwNSF@nL-@27a89baNz|uZi
zzR>J$=^`zcXqk9cZqQoDoAnn*g|Asc^A&<T3iyT`O$-<Avv&<L66()$?G5MJaRF@z
zUdM3ZK7!YZLHBxS4CwnjX1%W|?;G(o?|-uY?1b;|GB@B;(IvkRd?=ZaKV$ee?VG~_
z+7i5`G5D4@bYD$k-%H7!tXU^}?WN@L?dKg;QUWB?HIEz;>87caDup-Cj}lcbmFQY=
zESErBE>}svl+uXUuPu#0{hE@LIR088MWof-xrAQ1M6~ePl_te8_Bqz&ozk67V}I-F
zD%J~Hb5GT+w<Jw%S|M+|QLOK=CY%=Da7{F7DR1PIr(zbZ#$uMxKl-SP<wAn{@|Y5{
ziuVzt@GmmRQh}LkH*3veTadND4d#nSRrAw*&1h5cc(q^%6jR!bJjzq|9%`v+TrUGs
z$fU3%t@ZcnmMT!!i==Al&sZ6|P=D>JXy2%@u65^Jo8>{y&T%WJGPDtl&+sG{_?+Jf
zhSh4(XeUzs{Tkv(ry#k@BNt)r^P(<-^%sNC{fh;C(Nr62*?CcS(yo_ToiH{rWLKD_
zY5;Pmya`Y-H`b&8oeNUGxwr{5iP7j$RrBIYx$ypaBcEdip-0MB>Wx5ZT7M3Vy~7c4
zx<yHkr7am<;ubyHs9LyV1wG*F$~fG<pa+Cv8suBtg;$f<$(cXEQQgOb1+U7gUPjP9
zav?WTY{cT}OP>#9GhBU_CHDy-IXyLR_S}XGczlH$$Z^56yLGA`95b4107nvK&Gnbm
zdCjKtu6Z96hgn{S!jXxV<}E%K)(#mLeNwyyBi;<Z7Kp<Puj61XGs?bYm%o<6KD$O#
zQMp@$1mp960Pr-IL2EG+0y8+5k)lHrm-i<qDt}zblG`>Az0X(h2aEuXy~;IJl00&i
zZ%O9n@+paZi4XB5zn{jwkP;|Lj4yMbDN$gPcn|1)-N2Ym-zGYJd^)~9{c`4F(!@xn
zVS4>KG5mqzzl`EVB`3`oQJhV$->1*-DFubc5f4Er=XhxRZ160f-*INJ1v}uup5JM~
zVt;!Mcrg5Ya^F)sf42<3!ogeM_I<QW@o+6QquklVcI@bEKYw}s^hD$9Q(~3-=uZ8&
zm)EC%p8C!x#R=7B`s4eP`$2F<xPgf&V!4`r{55@j`n9)OLuD93MU|4Wfr?SYPoQ;6
zxcezUB@h~k&<O01IENP}0&mIx{P;F?SATx{Sl$`%ospByDneUBkXv&g3{>BmB4`rP
zln9_e`0*6q2M+{ZAU*UH-)*Ve49~1OFTyG<Cz9J9Rw0Q3n_x{A3VFsq&fw<-rnn^5
z6ZQEEo&Ln%K4Dvqm;UuFzvrjW>F?=xN87-zLE@ZUAlkw;!*l|Cw2fwq3G5-E!hhaw
zV&$6wBZl+qW^trh+1tH3G#LTO(YVCnk!P`=Pi`ofI224A8t#lgNd5dyQ8$0Njjfcs
ziU)+jBwui*+HkkK<_4Afo#8cxXE(0chd8iGx0~UU0o~m50iteRYVl}9C6QF4aY2}Z
zjHqx^6OYr1;}980-+4SASK_s~Uw_x}lGYN-<W6`|PJr^Cf!Fqajqt*_0s0<aa<hqb
z-qV@Y*UesXpK~{TImRwLSOQVOH|`K%r8IbtO=HOW9T(U>v_8rQy5zSYevU!U2V@fi
zBaJrPJ?7cDyA&5M#moo9N-9x;upS}c1eMmdSSiuDODi59EF}Pn5_Q)Azkk)l$~VKf
zCWPK6SW0y5nis{TR=6q?-I9&%P4$}(i7Q>z=6((E!n2U(E+%g^v2yUDS&;S-u4+X#
zBQ)wDlcNyF$RdtO_h7!kmGdhH?+@^scHdO=66fSqJe6?}&7Y(xVG{Z$wZ5hj;PWKm
zQXc>)a%pCk57A9I%g3wl27gE;rc^T&RLs-jK<CK;>dFDSbM&)PUVTrzlNb>FQ|K8H
z^*LdAUdC8HgLl6sJ~^pOhmOj;=+IjU`M&7cldJY9%1YDbC(!H22!`qeBUF#=3j0AP
zJJpPO7r%km*H;k|+7iKLPot+ArNC#4O&FD`Ld+v|s8RAvg(Pt{`hPL~1u(gHI*opp
zlRK0ITFF*f9Q!)<UnJWhKL~fE;Fw#v7RQw14^U?)Qv~LX-FpeO<_#M3X@kDHK94kD
z1m^uyGx`Au5uPJ->Ps{F!WCBD(`8fD{3zV0Z1EI*Otx0AH}@hbFm7<X)HBTD@gXQ4
zS6gxY3|O%hBUN&g&VOnNtDN24Sc4HLb-30u0ffI=dACGF1c_p00;@=gj=FB>+LokY
za@Rp3HXb{SnV2gVzjkY}2e?501O4k?DE0uyP;7MHny=~Hsr3vAHn2qDxl-2GEKzX{
zrdd>D?;oL@nbXL0uAI?{YNfVGXO12Sw0*pE=9Uv`**${mW`7ea-^>7E9QKjU+(;ry
zQG`C~@I&Zv=6;UJbt%Aheg?d2MwotdeGw-g#f+S5L4;y+MVTYCtr962&nic3nm)NZ
zo#*kMB&yo%g<L#7q?S;r1h36hW1~^*S+kDO-;B}#eMJedN7o{SBw!30=R{Jdi~g~&
zAJGRHHHAs4CVwfMj2YrZQQd&9w3ov$W=Oeia&A!8l~S@rw~JCgHie<WMGVY|-&1YZ
zj}9Iyglk#bjP8InVbQJ0oqVMPA_aeQmiIWz%>)gIM}-Ae?L@p_M4BSDVy!WkGphv|
zC)64EHPCgrC@kcn_3V03ZZ#%mGk^nPFi$|+I}yWCPk)14%0mOf5eWtGzWN$K=h3O@
ze5E3+v}LiPSSax{s8(QCfQyxYZw6)nI8&Ie*_{Aqn0BCcU&z+lpx+~y3q>j>9C0$V
zoa99&9#N$+wCFl*>use#=_b}KwqWVKP9->sm=JYi;ga3MO5Qi~GvJ+iCI493C1!9|
zqiutkoPW;qEUnG(SQ8P80L2n}J8|3`%^MobCBkOlie><aN)ho3Ye<7<tt4wJsMomy
z8`SqMWP#O03sV=eJy6GLgzHw=b>#8+QIW{1a&NsmW#$c&z3N^W#VJIRj;jZNT}Rk<
zWCx6ypxF0e%qToJvn!3>dcTJ(^+peNigZwRb$@tZU!u+sq5P%7r8&oy7u@p5N_^z8
zR4v^0&s)jgW}FB7>7d<riHfeGWy+SjJR4G(S%s7-&9tNsQ|4Srjx^*&0$jE$<t$C+
z*v6Yr-te5xE1}*D$pC5<)bc)4MG3#j?uL5#%z0Jq-JZtM-aDbS19fvR2k6NlY4;hZ
zet&h~ErY!$gWXQhfOsCH^)8#aKmgvDGuT!G<izpZ<VHSiZ6bG@w3g_R7Z0wtMnUM;
zI}b^@<xADCWsgw%D^RhhAj?nX?dwGlldk&Zk9zm)i093G40wj1rT3ZOoDpl~+XTO7
zf^Q~hKs*b&{ISGy4C3HM6(f}A#N5d{d4HZ|WHOX8d|mh@Wmb6xE$Q#vjF)C<HQ=yp
zEv*p5%~Z*tS~SRqiC=l{vFe5Fnwgc{ZsuUXE%gfivDD%?1a#eaAJsYcPFZzvcNsoW
zP_Bv+izu6}*{Ptc=3}L^=9}r^Sh<DTx@6E(yK*PqW(uw|%WTm0ySQs+En&>Oxqs_B
z-1T;X2E?0?Y3xJ1F@#B5*HQD*TvSs_*<!awDppR7?Oho_q(eHf4@6oMpsd;ZR_|<C
zEHmf7c+q_!0(v*2=4Ig`SAjM)OH<Lpo9xn5Ta9fWlWL3frCC1cn5^D0#Y`36X~E(;
z&Np*2;9P}F;=$|U)|VofQUuC&lz&knC30|V>&<Q1FpB=g^|WS<sO3{E1W7ZqwNN&z
zSKh9bGjCvWGvWg#WylrwnZ45Oba<;Y+!fh26Eq-Rgj``C;)Ng_#QHz^$y;8s%k8yg
zd#uE6a~}uT@u0@{!H#pn)m!#{*QD7@(13VA;>=xtkaM$%bz5Q?C}Pb2^)3fFmJ?Gd
z$nCCsy}VDp);p$OOv<ee%{|d=WF6P+#TP{bH2=FmJma|d4=<bmmoZKm6#+4qac>I~
zmu_SkDu26@E4LBN_bcWH7DhK3uc|Fpak+O@rOftbDn+uT#5a%<{k|J-3^*JPhoqG_
zm>rS;8jaIWG_Z2|ZIaUuFYjM3fBRH4S>ZHToW6dYfEy(L1<5}eJy`%@6ii>=r=Q-W
zM8nTJKP076{8;|%_^l7G3LN)<BR_C>wFJ4pLVtdMKcB+)(l3tIsTGfrg7R;fvg9Z9
zv>;WW$^GCxJcpmZeteOceJQ*NAH!?>_U-HC_m^=o#wa1Jn|}HJ5*kQ>P!1<B!YecV
z@$c!^m%ooxd#D^RRMHr&XHbD;{D`T0p+ZxFiWnNr&=4<(J`>}Dcu4*8$8Xc{%Rhc-
zJbxqe%rY6MGPFGewYNe6XvW_1edL252Ao(d@|1a6S}yP#y7wZS+LHilMmR|egTBC7
zC^VtDIGy9q3ldUB7>Fc)`dLo@;>I7jv+{2GGj{!6r;*ctrhf)`T_NmPVu7#{gl>(n
zoKSe@GucK$D=#GZ+evKX3{j9>r^*`U^?yiqRV6{7R!mz9t8G;twj=Mu`|xUHu+Ly=
zK{0et5^g27QO!!Dzp*9yq%ay=9|)uQ9rOo2vI4c$&D5mgws(P7aN{*<fJX!8g*9w-
zK*BPm0vrX)<3}A{0h3RXv!Ayiv76;9P)I4X#z#Y;g_f3yE8&!{P{;ueC(YXqD1Uad
zybgu-R%oN{vGbcrY;$datTMF6wVnIW%YkoUVy(DIKPl>IBLxx`_0YaWnizH5?U66X
zvVro&$Y*#tF!m0SPoEwU<;ZDXJw7tZX^!n$s>eq;jU*K4yR`pS5*s-K#fgb|PMFiI
z%Z8{AbHPoK%NY;9b^fj)1hN|be1Gm`atOo}1|J#z_OJX3q7s>ieU8)DlGWSk8KTj<
zUwDG|XB^>_`yTV#=g%Q|_Z#ury{>h<$b%q`V^6`0I2E?+v;G3S_OI&-Ud{=+%az<p
zVVm;|R8hnKK8_L*M?8E|93>4L^=g_Gy-rBTKjt1PCg{a$;ZnTKu}@|iHh;$8;Oy$=
zZ`cstA0pXY`E5sJPzlS~?ye9?;KaI8?~9Pc28$B!QtMkuY~)N5Cot+c5t0~I-O4y2
zhOPQ1lYP{C_ER>$O@1FY`lj7$x`tI)7TqPqTS;uxTPDjX#3Q`ZhU%O$jyIHbn;NQd
z^q!rBIlQt2gDX)vfzic<@PDrDExVMGBE|N81b;*59Z^lBO0PCc>}PV4Rmx1CrKP4_
znCCkNF6+3$Go{2Vk2#fZZKgmas@)}26l(`3S_3HZxIT#0&vNf!RGYkHj*&sg%s^g-
zFA(<2(O8#Z^0H+wh6F5o<uXHoz|FwK?A86OrNn@e&xTYq%${iK#DC8KB-C%mCXV>l
zT^L9;$}<(yChmp{F2~G`_L<MXc@=FCe3;wFv_i0NxuB%_IF=<6eb%{hJ7h_mO*{)|
zm!jFKs8Mky?$0R~k<}%gL7$eBL#-~M*kzkBa@1p3U7!jn1E{SlOxD<m{rXCkwd{BO
zI777KEQH8sR-h)(YJYcW#YKG~uaBDgN6T-xxS|>}Onmb^VO$*F`w~;<EyiV+m5@^}
zBWoRV#+T`;mpy+rwS}_?Le+amSI!_J(kD>w)f|dxW^PRBZ99LL#-4lCC6Ai*=u}4)
zKiG{Z&ZQw!dt)-*%x&}2iJy#j*y=_pn$G*qX{f<*g!f0=)_=zDbuN1~0#sy^4thDU
z@ANh(+Cq#1)<x-;k43VEuko1VvZn22CiG4?s3o9U+lFDrED;2JYt~jR%=a$7^y*zS
z0t`BsB-vB!JW}~JYfm?8&^(%JO|ZS(qgh1x#WmRn*<g{bKA2LGE^@1FS+j!`C}#&N
z$ZN}*9QEg0)_*N3`jrzs)Gnkff<oq_IHQauyt7}sGM;jt^Ew!%f}{K|>S9V!;|Az+
zvGhlT!f<vu%rNS1k!Y2)DKhM&y5_IRBk!=)-7C4O`Ei@JSNWacM%xFII6AxS;w0vw
zEvbFX9}*zPmOw^z;51En%q_*mMutsksU{YjoL9r4Qh%VPx<z8=o~zmUP64w+Tg;q_
z)9f8(;%;$eg`biG_59XvXZ+;kc*awmXGm5%imt05%7I^V?tC(5zM60{Yndp4TnvL~
zNqAWmum*%qv2&>CoJV%(O9-`Kwip_FC)R=q`S~pMTR~E_9JGbX-YJd2BnOhS)oLpM
zOxxTXz<<OH4QoWUT|~N&4-{}@Tg0)C>Rk@B<^zy-vVb@;^*A4JILq$0`v8SrXjCI}
zC9%Dv0qaSs;mz-v5FVJL5_n;(8E2?v^&9VYAg3AmY;{(H;#OY9z12BXBrG);Oy^xU
zB``}>xdLAj{UU-%Z_{DLc3r#q$PKL#*4lBI7k`f2(1DN&#@_ofy9A6f`L2+;nRmtd
zV_+le%0SYVd2h9a&32l2rX8Wu>1<!SU85#xO&dm4NVVIC)oKk1g>+WLW)Bj!q`PXh
z;31aO9$u^k7bbvv%6vDI*vJ_hjF<SFVl6l+z|C)5JmiV0^f+>V4asel=lue@B6y>P
ze}8zD-i9j%%(U?WYqh5Hi8%s<upLYA&>b<xR*aC9Phn(CnoE?hoiob&+SC&lcCj6e
z^OSlU_`5;60>0+_*U^K)CoRx9Chi2<*#(k(O|ryao!xS&4%wChXUs5^n{%YbR_&p@
zt*QrfpEHzgm$3VG34z)*1=p3HyNz2+Lw|~VPd(kJ9p#{TtJHR*dqrwQg28)|uOurJ
zsQ>%Qdwk_?0bLQiQbBM}5`4J#rD<ODxv;<dsJAw^RHTyy#U0y6Z%@s-w%4Dl=aw^S
zW}{$s19(L+h=P;1CkX~MQ_@Sd`f^|ukuX>)Hw~CnSt_m%2Z6P|QR_`Fk|+i<&wsN!
zz-c4kyMeg^IAa6ko&-3MAhozRod^99!Pu)Hj?Eb6GHEw^TXimm7Ta%5M_MIN2E)3A
z7B2C1YQeFd#Lji$65WmW6!y#Kiri(&<R2?{&AUx+d7Kht+o4#w1*BSyaseTikeOS~
z{OIVC#Lm#AnFdgo5~&Nfm9EUSQGZ&yQC*joi`n;6WaV*oxu6t?-#Z1j^So{l-NGfE
zPK3~1U6l(JA~(|O#_D#2A4T-OS^OI>DDH)IU060`>+l?DIS^X=nPHjktoCz$d19>d
z61I!O5!G)w$Dl(wI1E3X3%K)ud!}4CS?O;Fj`M93-foz${Krbl+Mf~+tu+*8c0`ha
zsYhU^W?>~QG{b>@e-c%&*7KW~P>eYDiR6n9{4DQ{{|&Pjj<Z7D@cu7eteUi!k>D8=
zmtp8U6_<fBEhm5NTRW5Fwh_+vSMU!Uj0PI7s;#Y>nY+CzWwtj{DY7jkzQ9WK_tkiV
zAcy2|NIBVn%UQD71inT;(d5H#10TM>Tt8m^_AYTS%qlQq`1o}Iy20r$aJs0~U;vnw
zFns(NetzX#XgaQRAm>ujVd%G|vwHg?!4d^r=zzB`1|fgZOF;+F@0<TW(}ky1yelq&
zgLJQnGpEB-8sO6NBs!qC!tL#skDp$68a~h&n)A1hm;b%=6Hvl3usHnd$BVy#5}34N
z00qp_yng@B@axOp3yaK}GwXlR-|r)ymT}^#wWhQ2Jb(U$5C5i{KM}EXcK9QZ{_y0S
zIpv1`4*!27{=7xp0wK<|R%(Q}lFT?m%pH^dRzzGuI=&Jp33*OP!q>MinnOZ|zx)5h
z486y>(t8S!5yfVTBjl8HjH7#EmMiEf#pw6J=+QbAN)mHRoODKyfH>%QgG1T<e^_o5
z{B8lQ2p*9E@jqDbiYZQNgNC6rS@1|pUXgJ^GERR=WmQpCO{Bl>>B<$AsCBtgsF64+
zx`((-4VG(9e(|LKCFN3cSCPd-|M=JOUBS`+B#VhUK9BGICG}Pvy-73Vksx06>h2%s
zw~cb%EsqsBOUb~&NeV7CLt{S`{22w`Eua;_3&DgG2Mb;bri5osiVm&_US@>|h@g);
zQlx*NqiXVRkqWSAP$84?7?H_JrV`m@o?G&x9`RelP|q@ycitNw?}m3xWM+oUZj|kA
z8LY_G>z|Pa%a(#OLAD!^sp_+BD_zmThZW_5%Vi#A+mYQW%6ht1I&Oi?tge+uM4QQM
z6!VD8Y!vNo5v+)o6BEFb>@}B+r1)vCpRs?}y9Kl?cx7X%af;xT6HKEld5lhB3&yju
z@tV&|k}3rf5yKwkqK@E3-Pk{s6^KmCC(eP$QgThXX@;nc>TWTO^PyyhDaEiRqLjh(
zXf?@QYiDk~ooVA$!Amrmj9?Lq9Hp18F}|*MN-anJt>xHDvKiy8aj!VbahuF{%VvK?
z<~pVl$I2WXQ=|?AGBcv2_aVejvi2AR<Zivp3u4z9C*xAoU$d2jepo+t!pkOJdS->#
z2`^0PNLnM^YRiF=yEm?llH85>iX>G`7S7RL<(Sc*rG{rb+kOG92p(gyaE{>7GRsS;
zq2}{RMRa4zT~5$pWODo9a(ryWZufs2SFn>I8b1d*X_&>&?ENX7*(;wlu>(_PC2<tY
z+01rsWDl}f8(Avh`<neS+g#S?9pk>PzhTD%9p$98g6wxV@EWx^?p!FD%ZhY%$<?Yi
zFs~_e!{mB(e#+#6WZC^tQ?9AzmrHu_YPg~+4~muj+2t$h((||zKqtkV%vyg~O+bo%
z)_xR$lc{bWMOz5<*AnM>Jf)<>Ph-#7hZboz87Ffel@iKieASx>VM(#G1&KhJ@Oe*x
zY_`H77`h_~;s~!@0G!_j1;hN75PUBpN=t%CoUWtmB3kGa<7OfU3bmq5pi5ps==8n&
zZ+6!UwPVQDWY<eGCw1QR>bQSyvErnnibQG@<t=4XFf28!uwyP&Aw5?9{Qld}UHQlN
zm$t|ULid~jx(&ZfO^cSOI_j|9b#3b3bJN?jX(*kev+m&|hpUBqbk8tJ1O<~^?=fP_
zb{om~0@!El_QNbTj?8bps*{2+4;Eqx!h9y;d=^sndL(a(4JxU<-ywhbMuC`1t5sB*
zki*c~r$0sQR$TG^8Ll7)h8<l-2cNCb8WyB`wnkv3d|4}+d~|sm5IYCF0t44fpg3p4
z5IH5}!i_@F7L2$v_b?E<hr9-b49or*Eq|27M#DyIxxyo&vn4k-W27nn<}v>{!N*Ch
z$Pw2%%{1_N$iE31Gr52Fj_Y0j7QJCd9rkx+W6gN8O)O$S^t|sAmO=;Pj`V$!csOyq
z>yu=DjeII)d6>n<k&!(#hwqjp*Cz$gO=w*o@Fq^TIg3c0K&H4vH?B~`$og29xPRs=
zh>~Ho$8S<aUm)p0Xf~<TQgtRdg;Zb<w3l>t>9;S_x{Y}4p4WdeUSv1Oj_Gso5)>ap
z-P`#q@!CJHHM|7bD)Wpjd6>mE>*?8QZU3V3haj)2Tu}wdO9EcgWKqitDm5QUDN<@3
ziW8Ez#fkWz%{xtt=`&cmC??|=``_sA!Z_s2zS-B(_qCLA3#I7|rhii^7AzJ`v}H37
z9#As@A3Z|PIRbwyM-`G((U<``bf4K&8;SSZx{8FkYEz_pw6Yc7mupn<zl9|b#(=4w
z67N4=u__A{4byV9Vl^KyElL4268g};VQ(!)(An5YF~LWK6x%%;rwXAM7tn6A=Tg=-
z9Z7NJEPSC}-K!oiU00%4%FGxiwYNtDX0Kv&^wYZ&FbjX<d&ERfz0YW*LXuA1>-yob
zOLQTc*fw^lB$MROI%>IPoVOD2PnHEBt(mb=Tj0iVX&Fpxm>2%@nFgtu6JnCT0U1}C
z!LCe$q70_3IX=|@V;DO3OnC4xi;W}G<i<{F-U98&8GmlqJ+oc6TRtmdR}_ldqibV^
zY>=igD@%XsRBB_5EzC_wuIr-=4JEZRHvjTZhN$Abw&WKZ6xy3gJ5t-U;$sLs;?czj
zDYRP$e>X}i=p!;Mj?V8<F<~v84RfPija1pmZdWzmA3!qJ7z4rt8NtOb$w4Pj1B90t
zgsIdWTl;)u+nxt)_!q;qR6wv*Ky3yPqK+)yTg!hpERLK4ZWY>Ya94zub|lX+l~U7#
z@M$WaF_pUov?6#RnBgb&xdO#)&ecBGly8)-pr#_%_PVA$)0~=wz^WzJ?FW5L$KBmd
zzuaIOIAS%is&swAYaK@7w(+Ba{Q#~AhA=dEQm>IOR&iOnYwvbN>d;Z(BjF0novbl~
z*u;O}zULr;DW;!`<)^|`SIu_)XscZawNW#g&W*y^4dRM$IH$bzq_13oQ+{Gpo3=JB
z&N8_avjL@%LIfqy<J_igj^2pq?wKzmYGX8g%4&C#0qD6hz&PIOu}0hAEvg~vP;S0*
z>08QcJ)q}RR{Q&7yu{qNZZ$ALo2yQv#^!%}d@tLizFRuWLN_s#K1GYHV4&(Zuek?D
z)>7LHJn8eQvC@n`IlU*Y=?6|D@naK<%DxFUEs2$PZ4=dQOjkaH3UT)-k}|-+*|u02
zP`WmY+l=GsPI5EbD&0r_c{7jC6y$8ruX$FboIoTAT{laA<Ce`{SSO{Q?))d@Nl*cD
zmyzHZ6ag}qf$BXIm&bcND3>~T8KVJ!m)>|8q6C2+SVFgic^N<fe<iROP(ub&F4B|P
zL<TWM@42Ng%2N_AQX-Fw<3c?y+#WZRsRc5#QMSV}xGh^{V+Mb$Y^5bdJSiucMku1S
zwRmL>jIYB?2h4ebSyf~;ela^QFq=enj#tmE60e!dt%w(f#p{(tkBN6&3@hR_F~5I^
z%T|`^ZHqsA_70yte=eXE!RuIUJVx+_Gb2mcuTqJZ>uqJ1E5=;yyHpu&wT<DWtU#n<
zN%0tnlw(Ghe5Oh4Qi1KW$qZ9UV@+yQn7$bydp~1)KU1fyl9z-s`J*BhXe!LN#1R)f
zW|IW-kD6mI$%Q$pC0`ggZj<@2Y*u8Bv6?tm=IEHBIH%AuC(F&&W=fahs3?!Q44QVA
zu3A}k#>-ZDkY44r&^oImsVm*+gP*%Ym-2fV`+pO~(nSYe)o?1sIp2H!pa|!tt=dcz
z(oDL0{h`EhlYbju;R<pBm=s6%v4k(;NQ=oSC_V+Y(ifUqMZ9^u4T|9@uRy^i$=f_;
zI|e!3vF-WPaKwF&yP+7K@@*)<IVQE;W9fIZ*yh|kTYj7EW1=1?#W>)D!X9X1S5Dy6
zhkq(U)Uo$Rz7)$^##aNMrp1VkIBcI?j)-y6QpVu$$RG#glpUz!gB*~fHE4FJ{=F<V
z&W!w?u@CCIfuNU0*Fi3vJgJWcrSI_wPzipnE%HK_MI9Ub##g?AD7jB7-D5mHaLt4;
zv*35xaJ^OM;V#q0n>cU8Yj|C^@sdtZI)D8bxFqGlLfO9quHj{^pk*y(_HvIWxtql{
z=jqw%+ibT6(I8n8n`O!w4NJ#MrIL(`l0Qp)eFpT$LrYWtTsAh+o8V^7)4{iJXzsHF
z!nFgf5`bw@oS8qsisvefO65VNkm(Tu@_r$d&Ym`fBvol#%sA)F|2%C|gzW<U`hQFr
zdu}}2=rH4x5Zo`6^1a0KGw~_-L$}5sPg{)8lH9YR2-=NLD-!1ADpq+8m$)U7#uxEP
z-M&^A|5l}f8?AtI)hx#{cdQO4hVg9PDk<D2tP{uSWKoiMEahi&Q~8@H0DM4$zr5=G
zF_GpLj!wRcaI7fI3;OncSX<%fa!5)@@$!G0!nMq_)svZ4Nx`%hS6UO#0QGtR>h+{=
zX`3R<hgTa+YMOPTTv$(fGh0tQijBsP?vHJUw|OKtoz&%Fj_zfWLI~uS%(=)3+m7vY
z_^MFy6w*zlL5p)Q8*(iUSaXUVAmbPKy7Io#rs;`aoszc@Q8{f{ILREgehOkx^n8C=
z=A0^7zS-U&yiGbV>}$whIyoB~YAni1vnR=#QhT5@o4o!$L2GcyNA{x@yh>5`@k;`@
z?3*l=8b<W<Ey>;WDr>P=FqDu=HLJ6vW=ifaw1lOXZu++l=Kjh`I0h8ydQQvW<PKy<
z4Ijf5UB{{W_Y!w;M9wtTP(4T7!O^$oe;GFe1Y5b3UzY)I4ilFld^0DPV8}2Em%pwG
zEPop}68`RA!7l-N2bfC^IYaUg1W1!}FF=b1aSrI6VHC2hCTbihC$HPIzy6-#O19RD
zos|^3`CwyqcgPveJo7?QAvrS6gyhNioRx@%QYE|z`j>n%F+>pL4adsNhPi$!rt&3n
zCdHy}#H8Z)@Ftj&^bN7`PAVnFd&mO?X@5TIoWU6%A}bY*k9Af`pi-r>@e)P9s(!wX
z2%wl$L;+QTzRQN3sMgUN@~QGMs8T606h!4hdt$015-8T0A)HcBX;=Vdd;k?d#S9q{
zkFJi0nBGa{!=&EnJI17xbY{i^L$w>jLLuvV8GKyV#`wqGSBMl+BQj<d(RHwXaDS?4
zU4C7!;9sd$PtHVF%`Qr1)emY!(&-bH6(gwz5=`W3afv3XLqTN|b5t2{K+M_c0!=L4
z00t#hL@?}~ib(LRFO?F*R2QBSDA8qL1st3z06840!!aj^>hY}b2s$DYsH7tx4M@=u
zIfCLk0y3mZ%a9>ab%708vg!hlaeq-8rjm`g>Y}Vv5|vSm_@c8EF%n2+lmtptF0dfc
z;a$e`!oUJ3yc6|e!m>n?&R4vV?9|qUj8rpc;4(Q;8JIpfm6095Olqs_vH?aa#h_Hb
z4B*38mBKq#k7b9k41?iaFe&-^HG%@7Kp{%7{5?<!VS-6PVMHmF5(8~@7Jrx>qbik5
zCPi5WW^#pEMi}N$M_~Dg1l1m)m01y$p>F0<_8vbr&9mA0Y<~D*bUHQxQU1y_FGp=V
zo=**RkN^DGes|c8+OhH7i-Xy;ef+pN&|q}%2VOP@aIB{ox(9&OdqiG+fJr@L8TA3n
z_Qfy00GGZ%=lTHq;LCzd^nZb>yeChZm-E@_;kZ3&UOqo)UXMSv$4{O>_vZB<AI8x3
z%@~WAw&Q7gp~<Dw;nnzJb~!&CUvybYU;QvXn~c7leLPZ$0!P8K#~44yNQ7CTPMpsC
z{HSI-9re?H{)_x&G_g%D&(Ay2_hYsdOKrUByuNA~zUt=bbULdOEq{%^cT`h2=;JG)
zmB}>U&gN(1d7aN4H~(s0G|%vWg&afvX?sL2TZaJloUB7?i<vD0m0rXm2#@!g!^@La
zC;Wageb+p_n*Ow%%%;s@^Y2$L^zWOuZTsQkA5C*Mo7ma>P2-*QKBUI6ljuthe}kSY
z4og=KLou;g<c8PQxqlM4klHC(kD$75|MtS72mYkBDWt_06c*=}Ur=gY=ZKvELU9cs
zMB9V~h1<j8Cc~A1L$Lvdf4~5`%i1g1<z_}&HN8})$Trpve%;z-n{W*AIvlCuqw9FK
zic2BD9BYr!My*|eEvlyHXkGD}hvj%`f1kXYe9#ru_|&K7^?!Uan!Y(7|2&zF|93e$
zpR|8mhv_EjqLs5+6araBH$@)VgSG#wtN2pt0*2?%bTsejsW=X9NIL0pb!bOpXJO`h
z0DBMLLsLr+-`m&KP}U&5NLU*}I<8chzzaxsVRpPZM%W5UO^n$VZ8hNn?J*-~i(Y59
zQsZ1|O3M3wKYxl3`_6^3pSHwu@&K>yk$`FRVbQbIBoU*ts7RaZIhuc*{8lv`om@2U
zNA26j1s9{p4G`xCXku<}>iU^>waDED<xOE3Wet`wZUxJ74a-iI&Cy6~ov-V+5t|a9
zkr0q*$&tl|?n|Yt?`OX~dKMl%^%sBsJiUB>GM@K+9)EdA;_7i-_dzcBzpPBd_dor3
z@+(Y<@h{%%rX@wX4He)37M|S&4qo32u-^t8yaEn(Tc#>BnYx=RMavqcYy}a4%{nsJ
zPKioeE_p-;-j^F4G1F&ywG{>?hwdC3Yjg(=oAUvjq{%i6;q+v&9UZ_Ki-ru}I|T1&
z0ekstwSOP%?NE1U#U-twQg|CwCU=cWaf?bSeo4@h2umVv*Eb@aIX&u0RQtwA!8*hd
zXS^@F*Vqg1kb(A{Py5IOyq~wLv4mQh)p?j!`^D_KT`1x$r7a@t&7^&Md16mz@0(xd
z<FonX-T2*HM+17mn*H^BOFeV21)KAwqS0qt4}U;6RkXcqXK1$1FPhWYbaL99X}A67
zNV#rNt~c|iHTB&?)in5M_2A3H@geJ>WU_@>4_R$V>9~1nF#Tck-R$+Od2YU;phsWm
zK|D(?NSaw)xuX(Y2`eTAE2z<)3lfBZ4JrDKG|-GAGNd&2xVq9PL6Gax08R=Kt!I4L
z!+(SwvQb(;7)V172Gkn8gMsw9gMqYuFd!~ka&iafRS5}*ZdZnIBCkYA12|)0t0(<;
zaH3Ba6t0K@7RnILpvjW`08T1mi^vep<gf)E#EH$Z@Bl_HYEBx$2|R~hm4E2%Z2x4q
zcS0lRS^ct#BlaRM&(NYi<PG-2z4M;lxqsWH5mk6CqDmMZQP=go<m>vL^P-67@`)n8
zVNQ)`KJ~5zyV8W1f)A`-KQZ!pSj(vw<63xZAnSKe=Y%}&irz03)DX18;;n89*%ixD
zZj=6ZOQrNud5J?(d5>$EbHq}<mjvB|?H4-Q*E^^+IF&8nq_<pGa4O4ou`Jugu7A1Q
z4OgDfA=W#$>JQJzng}F`x*tyWjI7{`CFU#Xfswd?tlOj)@lnX95OYSqKnS-lpadcd
z-dXAp(Lp(s(`|GBe6w>8aC~zcYBvVz+P+3FLO1Uu^h)*a(7un3YSYy&t_N3F>)keC
zxhyh!NA+rW=O7F@*DjsV-RRYQjDOm#BOV;v>}AuQIWOw#t$p1p!_^Tux;6dD!q<Ca
zn%8ek^Pn5kHMD;A+VN{Z>-zOyCfWwAl&;ZAiz1S{4r<rhfqj$H{gbl=QVIPhv9`2m
zez7ec)EI0poe?CY4PZjTPWlrV>$M@&TPL|gdT`W>md{;}=1*OZ3icsFUkKZl@&Acm
zpCOl#;29K`>Q)*Im+Uh*370dB88Lrl&KShtNL&#Uj3ArriC9^Z=&jb;)sCYvw2t)o
zO5CU_bfFsT(yX*Jn$4~P62N<jJaEwI>qMt7cTcZ(pUwg%O^jrkPp`jD3|~<E%P2ln
za?*?uCD`=(ZTj&urCj0X2|qxo5ctvfzrk<$a4(p_66_H_?BQMmi{%B!55s@|Pwsar
z=WEOGmOpw4!hVjDDSli@&8ToPu^hwN3?6=Z{o#(r?teNfK3QTFS5WsI8zW#Qr8uF6
z`>ucZ`J=1(>C9xi1`uuetjcG~g?<RNRJ#^ugb6io=x>zE*p}AKS^^Q3^R_;)SY?|G
zB`lE=G8qVoM5qO5{3VeJKx%)$J$2xL!lz*I7eXxM19QI-Nh*;rIO7Lmty3(6B+)b0
zl%kerA_#`H`o^gP0hWCe@US%^P7RS}RZH5~D;av_{s$R*Ytl&PBXm6nyF#IqULlZ3
zZ7L79hFCW1*oS|A+*AX{j#l$3{Nugg=EI*hc@<m|$ydAv4&ZYbyJLS#>h<{ih`5_%
zVXSI}j7RV6o;U<D9xy{D%c$jS^Wb~9r4yhi5z;Kjg^#J;)(cRU7$!Clfjl~V#?qKs
z9D(n1E@46m2~p=G9=-<RC{MmGt69c1k=E9|PP4-!sUt2>+IG)6m@vtS#a9dI!kN1l
zXWGIj6?Hv!zks@5+|7S;>OScg6P=l5iKcqljz`e4s1+VBr&HRQVwpxrqYh}V9p?+!
zYP~>x1K(77Cx-5wa~@%&BIzcq^d9A^dNx3`grT)q-oG``=84MIxqd&U(!M0mbyJ<F
z`F<vtj8tu1^uoF*@2Y59r<f!PM7DFAR`}r_jnHy*Iz8pjp{{>-lEOW8Zc{$S+DU3I
z??&z=s9+mn`8euOMyTkjReDluNvM%oEoJnacduifxf-X~o{AD87s-+YZODC%dg85?
zY2jikD_2koBgx~ELH8Om?OX#EIG5gvVi|qi%J!lEz*0J5o=}0O+`EEYQxm8l%9yMP
zGZ}?Lskf~O+DCt*%~INu5#Ui`t^G89Kj+!h_0(vIm;Fw&hUvs^@*qCE^W%!qh@1Tm
zpYZ0oh$OX;JzHpx#`jTsqGa?OPY~NX#nDB}f>y<cj_AH!^yt36EPlPb-u-h$msCIi
zU;q5=&b<H(-UJ41Xk3kz({F#De!ctiQfeJtd-|wSQqF(q!wA(NsM0W^OLU!J^bqW#
z4j()fIFj*7lQB4xYH|k%06UTY{PuNPJo(F)i~?uejcL)%gk>CODXa;{_<B)V>>0$u
z6DN2yC0g?gPe&4k4LD9M{__aFVyor5sR<l_C=5-OnqW1~3rMPvL?qET5Z|7`{~Owc
z)0k1?Wd48n6P<p+mp@=(Va@a}U-VmUhfe>P{_1qCBiNv}Il+=(KC>a0PA(p?QEHLk
zx`cD|Y6(00#lZ#8+hUP@E0=ONXA>h_3KVV#haB>_Z00tf2X<3&-$!KB|H95p%QNl6
z1a*kV3>$61jfvM1A7ZDS<tZU`70d(+=-fW@p=EzuOK{&R2h=&Vb~STwa;k#JDrATu
zc@@wG7dMkpha1Jk_Q4+psRKUxZuG4;1|KBh7!XLRh$!aX8DQst(cnh(xs2hoNr6!b
z*dt~x!@+=o>ncrf7{(yNhX*G4w2ZQ@f~_1>gpf6QK2sPfTvWlRPf}{RUOPlcz}Q~a
zhS7iBW(}-cAd<(}O0bd$E#7VA8*Jq`K^@{TfMQ%7M7+U(mX#GZ5vRl|%RwBTAMP3V
zE3mx<A(`!;n{TdSoKH|MZSbp>0<!g{Gl_17>YW4)1KA-6<5H&dK?G?*z`a^8Kc$4(
zqoKHm`?&@uv6-V|2<v1K$SUG?xmyG@|2Kcn9`jRM%(u($VoQ}m<Ss?Zcak#<Vuu`Q
z3Ez&X2gs3vXsahs+^2(yoO2pE+gLH{Bj&@AeW(qOu7etc*!dXLBqLJjI?XMpd7V|s
zZSO!z)aE3tYR&V&0bLv(eFs(H1Jh&d1W+QiX{{a1W7Z|$zElF87hB$dngzH$s5E~@
zM!H^GSjx(V;(VPf-s;z>#CCZ)%|>!j?&D6{N9pOX&O=0e%v|D@C{}hS(jJc}YsC$K
zOl#iwbuuVUyiM2&+}Tcrub%?(dTl0=#`|sUBx@Me4qMDc!u%k%m?28l%l;myr%J>1
zJTF<0C)S^eL{_q%;_h3<Fv+Xx%Cdjp>`8jstpIi>VZ*?72%{l_J%%u<iKc8ey^{Kx
zD&Cdt?Q=+(e_XwVml!95&p5o8mlnyD860V{3Y@bt*`g53_I46D4D>dE!iEU<5Vt6-
zA{uxdp(S*j0{eL3RHj<WZ&bBWCW|+9RUiSLd|S*~SenhTj2EBG;W>TpgL;1)k`B~{
zr|pZ|&lO=L(WY!_viNxpt3d7~4?WG4&n$iZvW`0yRO(dQ-F*ayN7_M9huHNPWz|3o
zm+z}EZ&-yHC#XZb3i0)giI+rDoH|MnQv!o|AdoYo^E$7ex$pVH%Xyv0Z<ogD=#R-f
zrtmF4s@~>>q38KqPN5k_mREnJ_w{~VUP16Xb*kE#>+w9yM~7z!0rfEx+@1QN`h9}G
zWr7bA)FEE@*!;G<3+_+!31GxRPR?C~C(G$B9fRw~ur1}cHeOP&ohMK$25%Kh8(OFE
ze@~jc*r@1cersvANGn<WPF_dx@6gIa1b<szIS?Eu_L=%U@5C#V+**I0TgwvuB;OpI
zjG?NEJuZqBJL(}$#hxw~OO2|V@Oz5IpeWVV{0<Y*A(({-{}_YKDWUJ%VsEn7!~Ar3
zry<=qhIfjc*RzvYd4D;EQ>~P7vC$8Er^fJ5wjts+<P(QL<W9yj#_JP+vr|~j#b%kd
z^!0Kzf-lXCUM~)pc(s3=f1WLL$0g&&if6RUmnG7d*&f1z$*PZ!m47`oCGO&UoSP2k
zCZrNKKHOo61mxUkg6aO=)szMYO~fv32+uY+=dqG%ILWoF#@bn9QPZ^i@)b-DBi><B
zhh*WH*{c;{7T!JE8$8=MK^@{%$QF(vUP;1*st?hIe15)~Szmv4i08KGYT~HH$L`;_
z9v@=|I~gMKW3ZDzxPDLH-!y543F;8vQ>N|n%~w_5cbc06es3o{mrir9kBwixr?74q
z?K+Xn9CS&6<7Cj)9nC{7DRAAmL-yRf5pwk<22jM}pcmt~bc;c{slO~dcCjYzX^h{$
za49sV3Hy33ggbwZxZ-Qy{fmotFt9e!R{lQjU??Dh!&1-PTU!+Ux+@q=S;7Rn{S^#`
ze%FF{i;cQk#Lk|fSF;9=ahXgSbea6!Wxl~>juX@&9)QrBFAu~^7!19iVB&xO7RESA
z9rT58=pAzlg9$}Mbz%MwcMD@2-41J<6LwTm&bUNL@B4rMZ!(s{1a*i<j@VI2Ib%Se
zO3MEWw=negx`i<gWQQP(OxaPHKcj*`$^8FsZ()psxJ{0+go|U2omuXa-S2h<V{~ZS
zfN2*tA9QBiFe0V>{a(Kq9{V<=Dg>m5*auCCdcS%xe)VFQn^ktd`2G*zAixm{Wo~41
zbaG{3Z3>s1;28;*U=A}C0XUbDJ~$_r6_**Gf7CgCzy0>@mls*4ujpHK-|ydE{_`>~
z06Yi-;qdqGFR=lP62|L;3Mj0#hd=&#`2FRtl?0QP!pC3q`DLK#g(R9ZkiK1{`Ri};
z@DFPIg*uzy5C2VFzsK%U1Pq6NAO1%4WzTW1h~v^)#DU{zgbR*{2Vr7U<~R^P;36dw
ze^cO;@73qGPgX|$s(44dRfK_xY`er3O)W^A2iwMj=W)6pXY!R=Ww{5Yqxk(Uf5$ms
z{7qP3$fH1Szbm8Gv2;KRO%Re0x}U2XXdw#v{Uo%|6D7-Nk)UQyoib+<r(EU;V&>F@
zo>^WUiUbPkGFEmRj&wiwoMw96Sj-X=f5#`3PP-)_>_qHt&Tb$EAXL`bCm;q+fN_I<
zvHbcbzlnR$$-B;IBvdcuKK_2EEA@R#>SFVwCCCk!yIy%N!q5v-XaX_6Zzq6vGkOC+
zV}uIo!2psg(rAdd72tdY@Y1*w1vsi0o|~#EoqR}oYzl*mCbG-}GaL&bHZxkWe>Rss
zi9FlG3xDrq_1!GlV71ml8~r3*(I`-bYTZ=1$nwY%1<3Otd_G42lLhO>M8Dt0Bulhc
zSnpox1_z^DIrAi8RUojshh$(MawTK)IC6gu$(^|0jn4+|l87jHH14D^k~BveLE>GF
z)DQ=-wqVl|Q>@Q<@ZPbvtxuW*e<4cflh3)?o3zo!?ne&a>Z7hTIk7nW^Vq5-tcH=T
zsm-44r_3Gl)2QunxIFK~`EERJ<BUEB)KAfG3|d$)KNb7~3VuI88^F69n?DwKqXp_H
zRg@TesK6Umh{O?Fg6m_B$D(S=cTTUW2Gtlm#ulfkn`zy!bi+(}=!Vsef69&^toJOj
z(CzGRIlW_}R=c^K*xrr72DUaw@sGt;gP@htq@{4Wrr?PWet~hHdLI|uSq1Mjb@R(^
zW~G~}@9Z(QD;S$4cpDg>I*B>niSgZ7Y+#Hz(0`8M#vq*i8JT^^aNiBk2Jn#6j^_Xm
zUU*sae=NBkC*J7SQ%6qwOu6i}S2yln$_9%}$}673A>)LHnqUvnT}iZKNIBNB67O{5
zWh+YP+02}0Gt<YdS`HOV-=ShwE1zq2b)YEUdx|rV-5f)qcy3O6qL=@g85w_Nf&en}
zmJA!JUbd@f`2{VvB_LZBv<^sLXq{Go^gjUP-5lRQQstcCIm$~3X!x0re8{ccjn4+|
zGN%mB!Cis~u5Q$|tbQzMwyhi1yk>CMqRVu-VjOoC>h53XZT>;XVfj<|2O-A(`lsT2
zKymH|Xajifgfkm1Rg2)6?8<-Cew4)4%y@U%3A~;9@Q$;b2+v#2J2vA1L4a+W@!`$j
zD{m0KeKOW?xOOU<9yie~)B@w1R=sWDeJ7cHt8tN<v1Qx7jV<G#1YvbU((xOH_ohI)
zj^(F5KEa{X*>7ob|4JeI=%Td?ju>f)?`Gx*?}#z|M7v+9t>LO#Y=(crwP=M?1m2(P
z_u{&Os~qAPtoidbRKZE1)B@Esf0Z(fnTIzx3qx7BZc8mB2(&<1v)-NDj&%{|7NbAJ
zZTElXw)QB1)N)!wJgQ=KI*?NNIZ8YkHP$e#%d<Fh$n!WB1ia;xrEIw)@9etTx~792
z&+r_3CLQsd{?xHO+8uwj6+QCqb(pnsg<o|>i#=w3FCa(D>(XPS=uW1TZGmC3>1<El
z?TB}76f$lv$pRkRj!16VBT|J%uO@kH&!YUqcKf&OsVZ}pZNc3PA~2!OeYapHQOW{{
z{@?`$XU(b+mxer26+m<B#4Z8>zGl<)u5mcOA;{G?I3>s-&2N7g%|bZ~qNme5ay(Ol
zcvG2pEyjbeRUmY?eLf3@M%xL7M(hPhaeI~H_w3p7NSweeeQ$cS`VN+N(u2AeRsb)Q
z^>AmWgEp7eKfWnyzP7%wl|5nxuRBPPN+iP@+8Xa&Nhk2vC_!g`lKKqXXVLCfIu%2r
zXy+A%fqrih>wJIC)o(1s!chVha5y%gtKufUDRbglz<HUO7!}H<X0=)EDO;?3H>*#6
z%9gE-x@MWxs)DWiLG6InXXo`jLBR~Dq{7piqPGGavDCa4mgfn<ALj!jp16L#$p-*T
z5zF)WfGQqPg2kK8Y^43L%1Un`k(M$O=sg#9;s?>IfO~)4m~M5O*%VYk%p*6Hxb$+f
zfh#^OwZ{GYZ1L*}vtme^SLpZIXJ6q4wJx&^Eo3o+!7FcKR_4*YzyN#8WYm2Tlr{e}
zB?#a<ZuxdzDg`*xEIXyIEsgVe4b6zJ(s7-suwA{l58~XqSb!JKH$%0NBt@jOlWca_
zaGl~)#ioDlOr}jO!^j*d@ObEU9MXy2sMs>{QlE>nNEDe*jCgquJ5JjqXzXA4+B+ks
zW~WhS-K7ou^gKpAjA?Q|PWA5O3f*e~s&;yI{<;5}^UvIZI;+l|yXwyO@}JAaeWVry
zXdUx`mFPwV(T=|7gYD^X?59{1DQDHw3b9E;J;Z;BET_Kb1&2rHPSSNB`SsB+^E9L1
z^dIc5=u8JEz}2F@nF|9oYDVAf-}I69h2vm6E@Jp&$JhDrs$X@kO%`)Ci5iTL)e^Mc
z(zgv#AO48NmerQ8U#k9Ua*tlwd4>A2JaOI6DZ-ZuqL=A^1nk~}kUq>*i?n=bz7!`1
zr5Atmb<K$<&J?T9dRBP9n%8s=Pgk4ZYu++q1-FYSp4J`Aa@_;Bu9h(*3g^@~I)`^9
za9iUwC?}luS9raUWEZ!o>_U=L=|i%9@v&$wQ87nsZ>`GRqDUt9=z>+Rt7CL+^CjIr
z6;F8Sb44Cg1F&O1pG-y0>3^mRwdZQsUPpf(+q!ZxFpT2@-f9s~?YBPjJUcJb&sW~c
zw%fYT95t+CZylw&!^Xyxn9TWq_9{0ETBw9oZl<lnsDsEGAGb?OinQCw1+{Ck41scH
z{}8L($w|iey3OE%5GS=YdFZiI?5pYq>-$@^C)MKTjoWM!gH6?O`uZ~6t^ONJ1uuX6
z^50m7o+9v{u_mgu5HDpkwef>~CA&L3`C4k|L9Xxf$n*}asj(f{yyl$BORt{-v@~%w
zQ`4U_#q<X@-Ywgn_U4?GCC3EAVPZL}@xFNjIVXcN#q_s6SFYkj0UtMiauG`7Bj-O-
zjPoCo1Zc*l{y}3+sJ4oA%Azo~9u|MnY+C!O1C}!qe@2L>!1lYo#=WUry3w&rBRSXY
zGocy}#U5;{HF8ls*$7uz$1QWdGmpGt<xA^m<;&<bOWykLwB(u$3Q){8G2%{3Zpa>?
zTTsg-F<2Igq|oT&-kyNCMQG4u1z9}(`sI4`g<8;j7b{qcuZveVOPogKxe|Y=kvck~
z<~Lo#mW!79bypdumy{_2DR|$GK-PSvGMEnJ_9xL!eVp4W{H0RRT5ebQ2Z{WEKa-a1
z5G?*(Obj;mwm-XhCL!BRrL+L4`ahSh!A^Y+?IQ``@T5yb*l$ggwOkF-6@ZAkHFf?W
z$)x-DGEE!UcaY^y(;S}Ma3oPgQK;Y+Gr?w#mm0szp=bl?Ph}J3c-k!e!)=da+0TCM
zEaggzb#R6`=c8T;X%=9aO<jBl&y6-@3-Jj&H|Qw(US;|DhOeH#Ng@9SmZ=0emyzHZ
z6ah4s;lL#m0y#97K^X!jm#d-~o&k5aN~0M<0)Krry@@7JRL;4o2{{5Xxu>{`sSdGk
zGwj=%k*4n2QDUt(VW+IDAYRc*a%`Q@BKFZ#blGtD_Nn=GF>7IlmG{(UE#OqdK5J=E
zdmDA0+W<S@*s=SOI-Iq9nen7--o(@(Kcpe?!#U+{QZqrbQGuPH>C;u!D|A^N-hbpJ
zoPV`^8O;!oJ$`Iw0DUIUIz+E5PGknV9Wio-V{wAWD>nKfZ-$+{=lrx8LChO|T+NXI
zuME%wS7Zz*>!z%~-hoJ*PfS^-%lvSV7=37aTfLqe49W(-W^X9P8k4O_zRM2&Dyev%
zn;k_gj!d#0=8r7aS<6_gK#H%4;ZUiD=zlA;#E=j8cGOF{E_tFgl@#lE-8k_{miQQs
z+bSufk$sbUd?ke`;$02*1WBtQmPuwqRAk~l;jfKG1@MFMBu6;7*OK|m9m;7fMU^;2
zH|6{kP5xM_b9F>Lk+W92`4k#!&>RJ8a>5>R7-EJjhf(vjv=FhK(3y=&B|`t`eSacF
z+Oy2Shio`^>&YRHT-e;Qw~9BjYUdRvQ9CW$^RzKemQ5>`y;;7wb#v}w@E(V;8e*A`
zOtAJg>;to^3}t%{f;F%3B~C|87b1St?!!=-EJcso@wa2;Mbpkkr`s{M=c9)AwHZ}2
zDUq#8eC*|tFo!*4NF*tnxJHr_cz@E$r|>no4Qo~kXB!C~>md&{J-nQTB=2~3_+O(7
ztzoO12%FM7)S;GsPlKquX_Z7hDl|i-*InKEar!C<lk~6en^qEUs)uFu2oYDRhxMF-
znTy&1RjiaTix}h5P#BftOzrPR)$zPrZx^#q3z3bnFa4D=QF_be2i?1B2Y-fBjesr2
zJ$a}jHNgY8EV)iW{G|d)c|}*JrQ=XJlP97V*4KqnEeqhJue^SsY9L<pDL)l$sK2H%
z%#>YZ+=8;pNb*tPp>C6`*3Fg<kceGCY{<J~$9SJO7<Uce*1`dLHv00KYi;3RK~VsG
zsR>tIpD0KvL0{VPobqo=Uw?ta1`Dp8y-8n(-h2#un8>OhkprQzoXNYEUW3YWP_iip
zLQbIU4lk3F`!@mZiD*_sER-y*$L#bb3z4^fR0b+hH7oGq7kJynn_0FAilw`2L)Jnp
zR6&LXq(_99w;a}VViT-_oqQF$VTJy@D*Q4!piGIma-C+iAImY?mVZi|<^VGKR=6v!
z%xDh%3EJMpl|90p8FcJ>!Xw?8A>I=WkN0P6+$z9b8QN-yg_5zGOd7X~KQq>Gys;)U
zoON*9<(=Vy@q=@B?a*3?g-VI#t;~P!;8Nv3{e5mH7snxE)I%9*5`*s=jy1DisnpkS
ztm*xja11PK9z4>0m4C_vKHh!h#LUt?#<3b=p=4xvJ#HWO1$-PhKZ@BU7W4AmOO-l(
zPk*;cf5V8<=H%a`>5lRb&>-J^K7rvR7YaliwH*bfPEjGE-0Qfbd=CXOj=CKMw!(#x
zRJEUPpqz}!eFn&*bU%(DUgR1NtrAzM{D=^ngIE|D+nlT4zCO2eI>R<_-=9!cLo764
zhJ_gZ2UpK`aJZAJ2S7C>v$ulT^{2qS!R)zg*WTDlEPLL0Q7GHqS=0Xoyc=IamyzHZ
z6ah7t;W7pjmn<eXD3_3`8Jq!dmo2Lqqkkz1FralL^qvc6$=%?B3CaezN<K$H>{Lks
z%K^`*WUKo-oPtHG7;KeTbOJ0+fqgV?mJuUnZ$yR$qRfiy9^-5@XAW7S#?<w3o?Us$
zK7QzDZCMWmvWt7mt^pQf`yEfiUuKrRM3WwlrjZK_uJk7OnO?9T%C3m}rLMq_GJooe
z<DwpmE}9$h`>e_OrHAhew8#785aTK1Dnm)nEAC1=*U=$I*aJOs+51%XPr1Q~c7zZ0
zwA6Anl)a{4jMQdZo5&tmvF!sw!h~sMvGb|cfLF-e?%ZMLi+U(NT??cJB77{~A=qRp
z^7_>Lnhg=g8)a;0Lj+hA$kq_iK!5G6(mELf>Og7T{Sg<8Mm}%2Krv{;oe7{k_XdTu
zSLvI0SlQ#07022@)sM?6X*Cp;$iFZtrYQ4sVJQpf{v#c8H1hd^9RVYc8sxyUq#`(l
zVA|EJkE2DS98OuO<1>$&ykhSrxO=@UG&sugOdpAc(a4Yn6C>J$?<Qw3+JB75ICMBG
zCL{r^h5T{;V3MCWH#yi@d5U$&GZCnn$L)ifDzxaV)=zWRS+;-mVzCP{hL(p0g5HS{
ziYl|@Ynb8v>(6)DIXK^mi8}Hlo5CO&kYs0uQ6XX80%eO>`wSK_Hzhgl2+4ni;)=)3
zZW7x?*N}!456*89Rz~g&w|{)Sfkjde#PJrrjzw}?k@OLs5qil=8lK@gVI)}_HD~$e
z-FtLyi$1XuZrQ!T1;YwLnCHTJu3{i4zm%)^nvImW$;Hto2b-L4T;)a?h0nP`)@3w4
zrc{b!W=5v?hso&YPs`EK86_dCAv4~HYA~;XIXNYpZ%C2o_{6(cw14ZAG>k2Bi7^t2
zvptV7nul`8#GuuHj+R7#VM6oyfKFQEYOsYkP88iFmGiq&p^MN8Kl_oC_V9qaT-^W5
z5JQT*PnpLFs$hWUbEyWqsNH6A8<p8$)>DlgX#@v}rn=9Ej-(I=0zyqM8f*(u7OYhu
zj^Gq+@*RtK<J`xRw11TKF`;35A2XK{TqX?hR#HzKC96Y8V%D(?=xvzgNr4M3B>?lx
zl5oc~p}TcatD$sPRfGu@@lC5DY*5~BP!(a8E3><*DsmXaPRb;Mboe$^5tdM_wZASP
z$n)Gf=43Bp-#okntRbRRf0^rf$w0##;wjds%2w|y>73^EmVXH4+CW#8bPj{qX-@*y
z(BD<kVU3E-CGw79<$UBdwP10Huy2<n$z|HxmdaFa!CYAeD9X>-SLOmuPWX~vU)9HC
z6j>$rB~eZ!VJ2)lTY(@)%2+;Af!L82kR%<R;U5qM8%j2!U`uzeZ?}kmgmQmjx@80;
zwu-I2eFP-A_J5%B*Ep%eAa+s~^9TrU69VHE(`q-g_8K{&CyJJzi>28E^eJPFS|wW5
z!s^9CWNSa{B*q?MD`1^x%?8UcD59%$;UI*a+AmOHJq+G3USY6GN41$=>yhnB3nI@B
zZ<b|bp#^c7t?>{!#lwY5N!lBRl+>tL=$fT;*o?^VYkxDM;_p`vnh~yPM&uQ3GotKU
zsQjYUJDL#<?K~Xa+l&}+a?p%GHs_lW=vkd<MU3{{pb;^GcQy$?W8{YH|Br15w4m7Y
zunn<8&F5PXlq|o2xZQ$CPQI9*j2<pli<5NPws^rVbRSZyqFH>lPL2Y_h6(D0#z$#a
zG^<ZFK7WD=Zo+DY^W6sGR0VENaz+0_F0+ndq7KjJA?8-gI*FIXa+T72D?wXUS`2F^
z?$WxsfuJ2KgLq3pmqyVI1znm&EyRTmT##%#+)6GacQMkm(*Q|ZWsJLlTw+&8T<Xp<
z$L@@A6AoR`btJNq&P{!7Y&*KE+Kv>xxS5=6p?~d2vMR-~+ekCsupRHTGFhoC&RE%I
z-*LX-)W>9T)3qN)t6eRSSmN1@cC0tG0AlBdM?2O#XMkFn7_bIA)*E|(R_sxN5_-uZ
z%&@UHU6Kzl>X$B1E>{~tHKOC+@7@c9jRq*NQvnV!$w(&S4>3ucy=_Fd!6Z={GV4t0
zmwybHgsV+u5XY~1%%nk6d4YH*oAe>#Jt-R^Wu$y<jjp%5A?RG|n5S%B5Ey&pJf@6S
z&w@Ixv<HM8vo{@%2pfTvc1QSyas_a<;a8UA;fx*R63Xb`sBDK4&zv@U%68jXdhVoM
z8cFx{?|sJY?v)h%=@wNtV6VJ(x2(E>-G8cPoGZGe#wFu5g_gr0c2XuDPR#yoif&-M
zVh4NrPBbjFyRF_+(zv;T+CRVdbe`c`f<y&@uTrt2Aa)v+T(3efe^W6pmSqBjUgsXG
z`1>-^Y7r*>HL`at6#cNV%PY$h+_`<V(D0hL#eU{t^d~8`L>5M)rdZ2UQAT&ZR3Ib)
zT%&uVxD)a}OjL%@#pY@?Z#hP+#E14R2m1>&Gc1>Z3Jw$jIG2%cB@>sy`7J4zn`bNy
zmqD``kpb+t)3X_80e{-cz@;*sEE=8G-YLBosogPMBDHr_*_dQx`MJGI_uOuL3fsc5
z`gE=~Cb3zZt-6%!nOanOFQdp@EsHu;Rr2ps(=_*ZOkzz&BZ`!%x>`x7%@co}P|9iM
zt{Y+RAQa~eL3S3}|AhK42(^9UdxR3q3Hu)7`(_dA{0%SGTz?mP$=_%ln-vb9(wu2s
zHdB0g0`nN{JF@Nk6kU{noY6XHCGfG+ij<mZ8>!$Yn(*Y!Nb0}~-JS19_8_J7UIN`s
zDLKWG&_>LS(pbYArSe!ynaQ`Q#wnJ0S!B&cUOFA-h;FaVuVufJ(6E55<B9avsqA<*
zQrGNIBvnbG9DlJ^ngSQb0hi8aCn{Oy!QVN?@-SGFz@b6&QyAdc7pSM)+G#?r!z3})
zvFR&3O%Oe(Q~UL>@|zAUW3kG|M@M&QTdcv+x>OszeCbym$oqTw(g!3^zh|SlS;Weo
z8OM}1`}mL2V&GNY6^fH2&K9_5O>ve=Js(Q)buz2`J%4zgfkx|;$Mz9+kkgiFem{@C
z^!LtkOMjJNTDqNPnznK$%!QdYX$DQ&EA+~l+uY+05>hjLcOYDFx}|i?nJKcB$eOpw
zK==|ak!cwOP$0gMWQ5Mv0GVOy(Aqkx4p>5wB?3F4<@*j(fXU>opi*Zrk!MmqCE&D7
zud}F3h<``t*!3w!9DURo*KQ)4QRtwHk{N?Lp^N?_vmk4xqE3717|o(YN0*@Y5KrOJ
z%6v*Ci@SGaHub>VJm?OXAjtd@JAny?0WSe6&rgvKI*@P>E*^@Y@$XjRu(`h-9C%c_
zch^jJmYe%490UV-7tJJ`VqnF&W<p)N6LE8q)qhqpt&8h+a5;uJCL87AcF%QmeT(ti
z;m3AdI(S>Jto1(YL|DlXi_PaZ-GItz@(;*z>uldb!+ih7yu4&wcHvx6#ii%VW!~R0
z?eF0C`js$mhNA<sj;6pqFdM;8my-xg;H`sKTF^fa-pw#{;8oFf*au#XOjNoPmrlJB
ztAE{WuW)%1(uB|(Le<VqI-bH|PzCmK<Zg_g2zA7I^KUyG9hhY>BKAR9kuyC-G%k>M
z74~=m`_>TL1^TTJb&wa);<&dOzqj1dU(^gF?xe=iF}a|{#hCQ4XUOwq$EtEHR(hY?
zhtq)xqCv9HiUP}|f!T?D6G)v->WfacGk+UKPwR{<l?PfGb7fIZuuXNIkd8WL%V#an
zn*r$n&7&o<5768&1?5~8RTcV7=FU8*6~L}&c<$3Kd#*YEsf*&~(e_ZJ1t`>eLYFs-
zSQoAJV)=Eu?T%+8sEv<O*O7|(nZkndNv9cp0lIQtkrkX@X2()$rbQPRq315f{C^vT
zVL2T7GCLzV<+GG{Vv>$x0-w)@M>Zr=+%?0){DCprhG9^cNqhBdM{vPt>udyWJyAhV
zp%~SgMfr8G*x!4if>M$FQH<z?DU6+#)U5pb@oi{IetVtzEs%j9{ai9Ylf;;7YZ?x;
zwZkrayvHF4knOGT3_kjtrqWGkB!5T>+3k=BiF7v>-?Pu$EMi^Hz@Ids*vGCx+QjI&
z5Jg4(na|0j%_@3|5J!C?+}s?_Mv!{$E#DA_SRJg}KKc%-Rui9g%9<i`gn=sBxn$5S
z82q#3jhv=n({-#8GUPnb!c|SjD*HiI*$}Q0kzzk3A0#JXgy$&o)Z}HjB!9k66{*uh
za(;(yk}>3zXdzP7JSbUZmdNTlHu)WOE{84NP#pM5-fW)V4sVQNj_;IJ2F5XP;*2+{
z;YrIMRi(YO5ayXmGm;Z+RF!a3bEm%RDkn%Zc|w)QQ)P#b)mO(RK_}DDRrEAXo{*V(
zDJ!~V8rFF|u3~Wppca*Xr+)>rGF-HMetW#3ogd75^vlg6)@f8P)`0m*b}Z0fv@8lI
zsDvr;W-Mpx55Z}9$#155Hs_uc6F9m#Yf@`wX*?!fSy`#7g`JtOFR8}!lTeQ{)B4;D
zUlA?3WUy~HQ@ufH)J|3^0KG(A=lxtCVam)|7~gq5P|V8dW#qJX0tVq<?*x4@myzHZ
z6aqFkmjRFs6qjAQ88CnN1%^u!0Dr2wy2^G>Pu1R%%x!B=Np>%pL-vx_4=8~YOsOqd
zwzu{mOO{9w1VH?JqU_=KgFSqEIemZm=>zZ(O(GZl;rnj~#}6$3I?Fc$9inq4fIEEu
zaroucS|9j0@qyL?@Dces@tNMffJ@xM9r<u?Um`kb51tRl*XMuozqT8C)A-aM2MUm1
zg|g+Nw2aO|p~>wW_jdX1*Y7{SSYhAs*{So_?=Sy-=_e2<7_fNw=Z}~20KpkZ{@}sz
z)DD0A`|#V#-x`HV*_u?Y`1jiq$rw!Xv5@@bS9|yuKm3^+Lr#bP4NZTvVeH{Qhkx(@
zYlM@r17Toc=WBn2?ZKOzvXGrf5M*O9-%es9XTh70u2Xf^Wa`NdCYBtVm<@sG!o=9L
zJOR@tlbN`f7+Z>UP>iT3oR265z-H%?7#ME2PjAHnN4t^RSx6tH%US*hA#;`^$a7NU
z%Ky_y@+i$JiE5rGWKf=EMfxPyAlF&WaU_{!B+c1L{%3#nEU0soq4W^Lup>#3zhPW7
zO1~EirCH8i6|YzhWfQWy%cgH7v2htQS<HUVGVcAw21gov<k6ayBP)*B!Z8+y7|R`2
zC#CTTepEL}tX%Tt&iItL6`sd?rN4U@?dC8{l~3Qge~WIBGRron-YvRdGKA_DJE~ij
zGz**X5G;Qgh}b53F*RP_Y^lf-v*#t|Z}IL#-bCi%f!>|igZh{r@7-B-Eb@C6!mSiG
z^2O`GfqRZ~C-y4na&zt|j(v+GKg+dl2XnVsU13mNPSP|hCttlXD6fs`c^XMZ=}uj9
z?8w(35LLUZu2l$vF5J`!feJH8O&*lPKdEDwg-m~Zn^WJf+cldB#-YE<m~SPqarqKw
z69|8jS10x~8Jl^RbQn+Kp-TcgB*DabBYN}0b@Amix<M@d+?H1MRr+D==Ua@ZR9eIk
zCralj8sIe6)p$`U4=u)IS{CqZhCU>NEc<STbb9*|Y@y+8y@Ic@`n()$xAvnv9ZXd}
zk!633R>oeIE@0xD{m`YVz>HE=43p#aRYGSG?jW!<ifEX$h^R0E7-Z7YtBPS&SZ<4A
z7C0FwP*ch%Zmme@&EgG)vlx#X1vE1^_NqCzxXN}kQi~%O<D9Rm)j>4u6naWUF@?gA
ze4no>$ML$w5N8y2CL)-?P{t=EB5c7X#a4g5QdaAy2&}qM<ZkWfkKYg7oxgvZ_y%vB
z4?Sfz&dbuYw&v=SS?pSu1gNOGB|tZfPcS&M+4#+9W;_ITw88g1;Op&UbN+#DKKF1j
z?rjv4$)08u2Y<1jYIa=8%w=?3mQNldMz5ofo8{vz8YK5X`FQ7x%|BK=-jj)bwm5%A
z=e#H)4`@8hOi0}L9r-&tE)-z4`jJfVd0+mH6itly`1~CS4W)2T&T}(~jhrE{yy9~*
zctm!^j+&Lw7$s`F8Iu?tdu!l_EUm&Ap~jfV2d4|3Byr(X!Dn&Eeu`$LIBXGRCCA-A
z`xU9Orw<`NTB>BoMkdZqhr$Ku6~upypF0Q%6%wT(C5`=LjR#FA)v2q0P@0lLt8|@&
zs;D_C)Jke2QZ<mq%z{FyG(uH}I;!WbBztSwAt=DBR%SufM19SaRhd?W5T@GbsB#%-
z`OLZ<?bi2-9eHP>dy*M><RmO!m=P*7N(=RHP)e(kMkUc$wK16yC@VZF%rt*46su7c
zwMV7UJXP%{RVp7<xlKyfQDdZvY*Z$gCo?*$37pls=}Dbbo2blBV+=BJ+-XL8)xBax
zE^*Ml3rD?~#KwTjWXb<oI4Y#Tqct401d_3IYd9(mMNq|0rN9~jQH3FclzF9^KGuFz
zYxmUp#;H6Ui#%0@2e}ud#2kNn9CiPpO)VA2=TiL-V`1f@wa3=Zwh~gfZ?i+8+e%<v
z0=Bs<7PfC+YB72batxe3dUCUrjWa_BcNc6BIEA)Tg<Hj3CZO_z;ix$TsYxa&u0(*S
zD&v;WM_|UB{0y1$#hr_Up6{@G?YOek8HXP;nOfObP6y74fzWbRO~rqS&*r?R@L&q;
zq<TshE^Ki3%X}o(XrdV(=SvKtYKRL^MEa{~=#r}~8XCB(Bd=(vqke~mE-Brrp^lGF
zG=$z@YPsdG;6b5VR?hFD4ZNa}Pe$NHPd+{YR_Q3Oz$#tq2f!+`@mFD$uE!m)N<a4f
zuu4zyF|bNc^*XH5>u7&JtnxVd2(0p0@l;r)r@9WSbUp5eRXlOco*#*-_?!)1f048g
zRq;6(fBk*UT3&lIiH)2MP8Xkps`$i7bIEZ3s;J8T*{?`7qTv+sIpwgA(FBgNx*U$;
z#jz_egL?6LR#-*saEw}b9;esCSs^8)&V6yIv5F-O1E-hRN40-DDXCI7cUy@xa1&Jf
zTI<qOv>z3Uq9=b=9A;I4=Wb@TUv6A3Wh$$7v|HaRb`&^Zhwvmj;@BPtFYM^3>?kU6
z#ZGk4ff~8PztRL)pn^(*f?1-1&p_E#RBbgzcAmBwfU;`pz#mc|ASk;#D+8!~aYE;E
z*9;Zv2CADz?81M-k>=>!Eq1h9-)nZnL}I!t{@hApW5{K)Zr?_vMO=l_fn!dN0hVwT
z?>QH7Eg*r$71VySwmS_|eU&Xir{X6qY)yL;uD(vL<6APxwW7rfGUBPB9#M|-{6)Q|
z39o5=PE#-(ah(BA2_Or)zJ9M82O26AdfM^T!3i!ysm6bg#__1OI8Py4_;OPITG{&{
zn-HX5l)W5K5vAI1NCe6ZzP<sywzBH8SNlNDM4M)kyqXkT--nSh=cO~<6uiraVYBx8
zIJ05HHZ;C#DHyaqs3KM{$%<;3-sXIPT;&OO8(YMRu{C>ZV<y>vahfnPnO$#bE(4#Z
zJppT$?zVqhx@=PM#m1o0neag7F(@8MtZ+>D8p2oi4h?;V4fj!wj|omGp$Oj_Y@+eU
z)5J}Yk3s~mNPj{+1p~cmX`v*qxtHX1pRtglo~P|P;rfoC@)EMxcCWinR+hqV)ymq7
zc)CCagL~2z6&HEgLehJBE>V~rSjvcwbd>!kI0k>B;mFiCK2F6)!@k9;U;!wi3VVdi
z{Mf=2JC1?Pi#yEbPaoMX38X@K9VaW1NKffAQP(=V7xe>phO$d;#`Ch3Z{t;f!j57o
z-O5oQu4SKavzdVP2JCF4Parq2!4S>0vjQ%TrRm*r2F%7;{{-8B1ja#pf^9&GoUFMA
z+6I3X*vh3(ZR76k^r6&hv)wq6y}b?la3jY#a=<6xM#ae?-Gv+7OkyKvp2J0c5^m&t
zGBLF@-DBI1+ofWH7JU{?R;sQ2b6Am$LqkAMl8s}V^9}v+6)D3hsmHqbKBiQzyeuB2
zHI@X+o3uwMUnZa1(X{>ZUXiB-&cvRy^UHs+%TJ-9AdoN9;QEHMNhu8SsN23_qhxl^
z;JRe8H^DvWi^cM0!?w3@3u~;}Q}<+)j;~+2rtyP;;rm!C_hS1_c9ZmW&+@wPk~4ww
zbA*@BWDNR);<l&Zr0eP}xkVFnnD$iAT-ePmionm~3*bTUFbRE7LOH*;kD50rVYfWA
zt}7v0^Q46xfrHwVL@k!wO|)(*Ry(n_G<2JocFt$*lj-40zq`JezKz1xZ^R9`hwO>!
z&;JLjwZ_?(k>D8=0XLU{o;(znUB($ae|mbR4fqykZnjTJ>`UelUy{$yqeN|#T5AvH
zpk65Q)k5-Bu_RgiEM)QN?&10FaM7l4!ZSHlJpU*NHzYnuqS12Uh=kQtJiioQKT4@A
zt_NI3O097@{Py_f<Gm)29(2M*$9rce_16j);rBWIultpwxw@-PC4vqg6SBk=4|_V2
zI%1+9og?<~`}5~JmnX*=Ab-@o_!L`e9SyWM5rw)f-I3*8SC1(XLTZij&eiWy_N5xZ
zjR&~=SM_4}$DEow+(AAkeJNpG8eKTu@#(Is@aVDfk_a|BDt>*+{yjb4{WEl*VL`?g
zf4tnq8%C4RUKM0T(6;#XSMlTSw;VOigcR1HjdVdm%Hmh-&};dU7Jr*3EH-H^b8oSc
z#0oTpWfDfIW&-t`JI9xU39w^;Cb4WWuymPmY5FeZ<X7%87~Xu8CiGgqbR>8!$c0=2
znVvqY5CgWE_X9)~ug$1R8ey0QsHg3Sq~gryfN-Odqm!W&3#SH?!I%vFwA%4X8=<ti
zq2rbI!gx18eB6#`lz#yE4G?+UKL*=t=usor`-WMtGz;ct6{__D*Nio*5S<myxmz!F
zAH<MB$c{TbXyL+;A<kpO*^K>(k?r>=g|cH|<$yv#tX74ZLTiN9Okaxg_&vhP<OTRD
zLVo=&i$8JmGdk<wi+{_mFO3*k{9XJWIeG>dfF-a2>4qs_S$|-+`;cH;2-X2Qh4l~%
zHIrJ*fFfW7&QT{L3<<>=&`uC|wpqrwWE%_00G<BFkjn0OZ)FVjI;38<3d>arf$OPC
zU+NL%cO9#28aI&+=x$o_l>YZZ|1pYhqkj@v!=0+Us!~=;NNlbq4`_YeTeQtAZgltb
zByzySg9dLrjen8j9Kfwm*ipS`65~LJIcKciuHTH^p0MEcE7|Q?C^s;hSD3w7y&1Ee
zM;tSI*km=tG6IdTF7Ort9WwRxlzK!kk`=cponDYCvq4@BoH}2)BuYAUC`gT;5z0h(
zTmD53Eo5`)S^m}bHaG>)Kv#K-P>ZAurwgCES0S<!ZGQ{3*kQt2x_zkSn1?&PYjdoH
zSf-glCd6c3C)gUWw3nSMz0UI5Bgh<GZP#zkZYfyVJu|WzV%Z{$uzDKyb?D0AB%%Vn
z5?>isf^Cy{l_^xrTXf<pRi!|Yo7Qt#nIdgDPD&{#xLmm_bXK3Hv)&b*-Gb6E$FXX1
zLq;<cV1J;#J)<EA2zY+i{#gsLP&3JZx39!#5TdoIr(RFi>E<%^g$%gao;g1$0Y!Gr
z;97`<i&hdS!{34d7ZkX)MSAGfc+t3~pz^fV7G6!aj9Z=v)5LRqI!p`0bS8XHX;#DI
zKD=RNRwFo&b=}JBGP}nULq?*7?UbQTRA|#FJb%iJX++hc$}#I%zp7zo!uM3P6bWO)
z0C8tUE7dx4)GNU$)iHCIp4Yo4RREGeZNDUQq{nf&sp^>Z-df+aMo7|2ZBDhxtoN4M
zX{t%)Hb0Haji+$Wx3W1EBd6D6+uN7iyWeIltzaJ5yt7`ck#T563ujOxxnajwSM3<1
z9WI!tfHr?DmM&E(FNMQ_u<}ra^-diQ^gd!8E>?|=0+Hd&r~9M+OGBJ5uonQz{UKIp
zI8KG=KwS-~rHt#QNdukwQa(&82G?Lctx`-Oe+-#os+Pi|QZyrmn7)*6h=SARV7d(V
z2T|?m)xk~K_%zMGVRTbAz#hJFbjwf|MmO$8x2=DaLQW{+T!U*gn!;Dsdkc*!EEddU
zZu!&h$8)(QSDx#nM{im0tzP7q^bbVyWE{MK6H^S#<?eirmm*w_dfxHGt%`y<;?Arm
z#kU-Fe5+}kK(rigwdwdVPPk3{z;*yFE+Jgy!GlQA6oQ^Nv?LP5Xisr#cBOoHf3K8!
zr)PhIRUH$StJuVZu8Ip|LRZS-n9v;*4u^Ws^dXFb6GpYW#?%#_<{*~pKjyLj7|xuJ
z_f)@i)4YRIK1{2tM0R*S-*XfrB%Q4F8%C>GR`=9<N#?3|8qbUeDqWtc`+)OtD!!go
zPSHazR@vlq**;>Jt&3e->5aF+|JE#8>4bkAi#cG36U2))Be+T^qx({((=yT+BG)N~
z{p3TYm|C_kprE6+IPp+dMC&)uj*}?LmccVak*Aqam1WtePEjN-iihrda>M1)-08nh
z_mCD&=(TdSaLOv#rNZC3I|B)?UaN*3LsMvvd%At9U)Ol9?wzSUkblFl{{b>ySxT1@
zkS!7dGdPzKULO+zF*cXcxHBi0i_IA?f6z)xkr4P11U`V4-Oo;TuZQ#7;kOs1J1ZQ4
z#qRA#2mC_vUy$5rsIvgVDCpkayRT1DqT%bzmzGi~zAQgGzSHrO0>?ez#1|YtSq-_r
zLcV~XFX8{bUmQ*8EuNwT<)5--$yexUL8>5=`vH%J^YPo;mxIje1K(kEe||gse|wlu
zfUFnfvHSjh2rsk(p&asSDKx6?*WcZb!|z2zFIoy0p7`(Uf=pwLaA<?fuis?%7r*?%
zot1apf3fR(`Ht-V>Hg#otO<5PM!|~F_O}Gf4uy9<iycIe@<M8VzlciB3<~3JO9d%B
z&BaC=bHFwxjUk%Y7~2+-V%a1-e+$nx7FJ)g70~(k$;e<u#1^NXdHuOn%g~D4DxdqO
z@}A=RXd|Q9OzfiI)H<ORoCt(P*a;{iA*C-s@#Z4ld^$|~Ghg93wAlh2Z%@KXK$<mQ
zK}1Vo4Q`DH36hqDYayuzBupN3Zx!&lJO%%O`O=M6m0F!GJ;S(>)H;xxAHi4w8NF~)
z?F6z?LVB2>OxGx1;|UWy!OT~o3gydamsQUhk$;S0&uiy{D5felJ_Q!V3cT>$2SxfM
zCel<+CA2Lqm(&u+s3|8)Jpw1w=dqNDe3yG%Qd^M4AvT$%H*tyh_B#7-IrWho21(6R
zALO^5`pjo0wPv5Ok&y0)<w#dpdsJ49l$ZIQi79VBDbJ}rPS$2{c+&C`CwTP*FdPww
z^?$gHlE&P}<&X4f1V7Ma{_!I`U44HUimEJA0o#r-S6Qav8CctvW$Li!h_=r%bx4Vg
zeZ<Y&EuvC0<AjkrrJ347H2mHv6ymC`_II$Fp1=+SHx@ra!K0Ffw$Ek-C9=Y3znR>k
z@5XYAzS;=mcUn|-+TOYk7S(1KV6$7OjDPRV3a{XxNLaObZNSHjQU~2(s0~oI<Gx1l
zGOi2gZ1E8kNN+Tg$!XSHnbWpUYXvUU!dl!3E({+jXxV<(j{+AA#xum3Qao5TudAdu
z7C-w!=_M7nLWz?}O&z$+Agq83N;ptEfeTiEgh@fKw?&gM)2GCMvOW5YeZaU-^M5f@
z$}NkYdc=^{&}^Rd3YrolW7!8y$s|IT!hp<0^DDBYGnt`EEShz`jfL_l%x)c>ww7D;
zbv#CyeZrG+S|IA{@PuZZJ)~9CBY!gNjoZAplr7I`Cw%PlrAftsLUvc|qaBwOyiJPO
zyX+NZl|WNCUhupTmB}7i^Wn<C)PISiPp4e*Y|7-5EDBCMwd>x=V|MZo9DXIt&2X&1
zOo<Qd1G7P)hVY3QygGQb12c`0kf4h&<^#h3t#YHd84nRFA#I-e3Zy!6hJ6qv&g8T$
zhKozbI@yr@#69@I^~W{fH*BYFO6A<B$mJ_hY@O{2h$?Y|&BMdSa*Mv0qJQr`;Q_tS
zY9V<!W&a*b;Q3?LK4PtvsZINI)?iWsNIg<hxLZVB8ATzFxy^GntX-n>!VUSTx-Ry8
zNZm@gabZWpq#&8fT1QMZtkQZK`-emh#zo2UpG)=BP#KTkDdox7N=s?aSjSTM>;r3_
zBlR@Dr!%I#bjkFH7kZ(d7Jn(pqoqqUmRZ73S-b{3DXn=}`Sa^%H*5Lhb<kU(05o&S
z0J9&g^t}yF2j;yahdXakjs)h)TH85%nuI~$R*gi(sgrZtS0hnL@lY-DP63r#g}AJD
z`W*+dr(;rCtR!Txawtb@H`+`v=J3i-=|CSN=s8z>KHRve59Tc!Tz>^#hA3iEdQhE5
zx1+OyI^V6^>9-;XC}d$+lXK!qO4P32?P-JOJ%&5|dIKd0HoS?k-Fe>GuHV|cwQx#5
zvia{8QRnS~SfgkQ(r_#GEqs`y*emP-8RjN-d}4Py$H<3jSK)E#BrhGPZE~0fvYcU0
z&MlR&Y{yN}3wSU|+&Vsbg%f85kD2hjA}TeLUMQk{A{ekQHACjol3%v#w>EDkSH?WF
zBKL}@G{R8GHFguHGQ0l|d|oa-myzHZ6qoVNCli-G)EUiS96W{3_2-xGZ)6r<3U9*K
z@EHGk|8n#1n{hG5C?T!8{N>Y4ctHw;a(DqFyfT;H{&x9z^TRl6t11VqDrt<?ovJ{x
z{)lJyLWNgJRoY8Qycl}s@2n!?k7U)6NehyBmu=M<aewR%5X_JH<FgAa0(OS3OaE*c
z57rJ&IHE7I%4?@M>4!Y6A^kayI68bZlXe#`2vM3XkmK+a9HiAu5yHar`i37epi&58
zwixj%VkFXRMQzh}3P?o()3i=u9!8b#aiqyqJcEX4^CR5V+6k~!L<%@dkA-)^&>%s8
zmiWZX^M9$pP8emwsE72+LKjyN=pa9J9$~ZmwsLTd^kMWv_E>(QT(F2X&hXQCiI?Ak
zK{wxg_^z55iJckDxWb^}$i;5U6bwS|;=0-#P-nI5kL+Z_tN>pwuOs0BSMnB}KV;1m
z522zzn=KUW=Bh?bri*L5V#tyqC8QC7X0R1ye}9Qcv+@_^c->yD*WGDr&)Z09VJ-C3
zpdK-;hh^H*HGvft-Hxtfu+2Q1q+lP|^=vxj*=bbk+3A;d>_g+XIT1%IJ(bZ^Yyp{5
zq0|fa8V`v=xAf?*M%_2Q%JV$<><8U?l)9qHwvE=NTol+wZwH8gD3o-&dUJkPZwzDi
zjeqXaNR#NR4~IE?z`Etv@=Kbn{*|>Bb?LL`CQ2%^(-B^#nTr2c%)BDr@8V{WTrHH5
zhc4dAf6F@Rp}~{K!V%vk{lp4WT|w*QxS0r_uPJP2qNnLUOY~~2MO{e|8IY5U*7xtT
zYgU+vj<RwL2;ik^-6^vab}?meU_vSqK!5owjx8?v*o2F#<FjI;!|JRZXcGcY!4k6(
z+H?^pe0K5Y%I%?mVJE{w%dduX49WUycEBu?$#PMzYw}sXr!Z3GaBU$#>++VG3cUlD
z)|pHvRpU&HR0Z6#WRHs-wW{MV%2viu{Z?l6E1j5P!%d5mcd-;I7SHo57rdE#V`MQq
zB${lW;PEB`lCoDY(V;_pyKY$(SxgqT=C|i;bcu}t1N)RaqPlIN>~dffF*>wb!m7IA
zbLCc$S#B87j-jPHC76j4mJqFUnYVsyWC_oAUDdO@mbj&B&`ySZs?1Iloa(o`*cs9X
z1ZH{`c9)Uh85EZhDh3k+Gc_`o@mB>Vm*L<UG6bRR*_oFY;Ta`=_|n;nS4udoE}#B;
z`RB8g<Zyk&Wu(*^m&fOT|J&=6RsnaY7hLN4<c-SrYr>`Q`IY}(?@QEn`ITPM6Lfse
zPnNj&POqfqXW}VJ^L72#r@y|FdHL_(l)Xd|9XV63og4`j!^IgV@Y}j}gLNBlGk;D`
zdH?40#fj*3uq}0e2nN(rDAI}g;q!;z9<_Y<@S`MpWa{ni;IROIaW)DWmYaWl%+C*p
zs|#8Xgzl^pGA@#R_{U>+Ng#>dU7R)|u%4LYKbgdbV-hEPv^B~1YGp%*D~nMO9#hKg
z;d6IoTC%VRBvvThuFM~C(J~05bj{_v?!sU(Li7Gh5~#v|{+7v(Z5XB`$f%1WDI#vk
zc3@#m7B*`v+X62B%-?yoP0x40sx4i+vI<ccvemN(vZAPkgh4q4Qd5h{*)xT}5o;#A
z?@V=U{#r&zNw?ePZCbdY{&?-Ghe&8^Tb@>`$0s7%G<Cw;XrY5|d9#yyIJPx?-ZluQ
z_N==?((qS*6Lt%hvTsrzwxMewJ^bFcbq^a?;aGnsOknDIx3nBWK;;hk1ePKAkQLJS
zh3eX86cMXbvzxE|-9#$_+t$HuPPD_;NM@%V#n<DMvq&pqTnK*~SU#4FuS$w!_)p~h
z!!Ltq0fS&Os%fcjMam;8)l2-P8(r26X}uRx#?91!6~_)-$Z7J8BZ|opCFP&M-{>{Q
za0Gf?qL2Au{e{1=uDXx5zwDJ%c4iYZCwpC@mdtVO4VTG06F)j!L5g{_1cek6*A<z(
z29>{a{YV%MY|mG67)C-byfPrM!m6oIsHH|e3pJN<C{#x1#HiBIbSU)60UMr6C{*K6
z2>bAVac3Wsq0niBe%kXM#6p(#ZpFh`XoRQQl83P{a-=?1;4zaxBGPVY!&nF*LhC^y
zrJ1=<Xat<N=F<;_(1t*&J>Jc+dJN3bDYsKk84HiY2B41!$j*&faJWcLm)x{;{>EuP
zX6J{KTfN=Kur_w6LJkohnd;{Dt0N?e*X;U#Z`TswZeh+L8(F=%kSm|k4-kN$62h<q
znT0H?zx4xzK?nzWwZuq1YvkWw|JT(C0X-;QY02mNewuLr96{A2KlQnI+=^9KB&A~>
z%K+N8+SgQHKRiAu=iq8j+C8K{{KK~ac!pqhywxCdqBJp7Q>||2%^Bb}wvo%<y9I}T
z23i3=2MO#&=1M^KH>0_DXhiv^zIPiXRA_Yk`ydVnRA#m4?~B7)-KlNuJXYE0>1$sW
z9<qRBE0Xtvu+5%%EQ%8*At#ml><8%0o1Mjz>8>sXRU2!nZ(T+P*h{y<{<+UZNvMa8
zO{kCCOk`}Z*#l+7V|)&<VfoA$4u!&h^HML`*DI&~=Ov6Ll;xIdMNH;}?@$GDqaFTQ
zzTu$kri1dj5?YhR_MF5$RNt(;%u>~6emA_hev_Y^L;k6ukrNVmv6R@TR)A|$OY;4K
z-qpav85^pa@<ZQ1EwY4;H3?9UTEgr4$uFNqjw=z78s_OSzjm#PozX<qTGh>edJgnI
zg}=L+F&b$F3#r!I$KHk`s0Rh&4NXCB1C5Y1t#}8}+gRw)xAK@o-f%o&rEXaQdG|55
z0bJyUNyBwA)>|Z3eZaIU9V?w6xs%QSa$BIlXd7OA$Ze66gAON}JLbah=|FRR%*`2G
zNoM44b$%aoTMba#imng3t&ReJoQ}71q935bV0PP?0Nq>6M-hUIzFFf|X@}J|O|GH&
zmsmu|EhZoK2+#V^I#pxI<y^Sa8VMy9RkUMc<VLadwxt>vBSo1?l8S0n!5d+S0}$#!
zKc&4?^t1goJDTe&!e2b{`yUgfCLmzd5hk6(Q&&@c>6gVi2+PYn63a<{@-jy3mo}={
zElR3D_h?Kb9^=kC6sDaNG5Q0A^;;0&qhm|8p~W-!i`vaLhCAugd_qNMg>+Af3W1K>
zv*i;83RVboij5e!Q|T!&Dn3U7e6G0k%Ti?(%&4Xq+Jv%Ez|d=|eo9f!EGX@0Hz!;i
z7qY5d+aRStb;1N@WCfXjsg$BP{K;lJ{gi@|OGHKuN833uq7Vd8YeRf~X#-<4?Qbm?
zj{m8n8%$9RXs=cz?ekSs0-#@W#Y^tvZufJZVeFqU{@V@f+v>?{7TS8nCZAlbS*hQh
z)fN(`#`?YuB1MFt+=-1J7yphBRcuY727p-#X?O0MwwlVY{<cDYP9}G33qS)Ry=z<1
z3KeS#>UZ}5A+1-tptK%2y4>B{%%UjMx<DV$$Hlw$%dlPRyY>saLKHsM;z?~_gV1sp
zsz$Ye<Bq~k_mkP&u$#Lb&p|df@bu7z;eN1ZoN3L8Bb-$OQkZ6EGu41Zt>kK|&x2|J
z7}%S85)o__HnbCeTi;DB0q8YLn|ISSNAuC)hC2FYtN8RTMvwpz_&8&>3Rwl9ye8=%
z_bF6H7L`%8u%X>@_LsgQnm1czkZPiCSq>1r;@*byE!6}Th|1Mep9fe#r^An9xgYPy
zIaGaWx1#}WL*Vz)wo+WqN`2AY%2nA*&U1eKras=duW}1g&y(k-uWN7S+MVUWUZSOx
z%zVgA?@$j2t+zNO0?ev_KD%2GOatWu|L5ad0475;-KS`j?b6n*(DM>AAbO_}3Z-Yy
zG6MHnsiyie7+?Z6#aEW;8R&msT33_Mocha!joVn(0NccxYOal?rNi1>eH>Xfn-k=D
zy*{O;fZKj?D?BQIi&|jQ0;x_QQ#rqpM$oo<V}7D74mN3BO{k_#PM-ai?^#=Mt~kuR
z@hL~yUY5UU1#|9*x6j9}D`nzQwnAhjnU`DB#QLpSQtZKoqV7Iv;d_6eV!&@g9&u6M
zJ4)_hZ)7e6vry~l^yd%X9-oYQIGZTfQl-?utd((7Bjucea4SPVW&PAhdE~(=a6wfP
zb(r!Z$-v#$B!iK1?%S(a%WstD6i=?R&|;e6pXpTd%re}q;_}_Fxlf^4*u+S=RSC$F
zw!((+2S>`!W_zT7^2&clPaIMJNEl;ksxQM_pp0MUvf|t!1uAUEHd(K21ai_yP)#(@
zJ<<W`aI&;hMOs^HWQ1`z9RyvIsNvjEZ1*yUaV*CGd%UF+M+1%sNv8$LAoX$|PU)TT
zjLvFMW3{rQ;jE}`1u?)lflqiF35>A#>I?+;G}EsFtj<tVjm&>kQA+g{)hI?;#Hb4a
z@Tf^7zP!m%@fcVsA@yimlZ=pf5EeCa!(|GI^B{>o+|h}V8Yy&i6~zFD15z>4X>~fV
zQ9cW6B{#6qRP;4s7~$|JbhH043|~wXXxvW$SKLSvb_OdY>E=>yK=4pC$q0m-)Ltu=
zHG<$qICsxJ5Tt+a+XX_jLJ?k(4B)p04<PnpQDg9JVtUoBpiwgl1Tqt9eMN4NI(gW~
zYmoSq4`;B6;`gZ<F8$+QhBF~?DM-d)O!JI9)l`Eieh%VL$l-R*xP&w-QgQj(pW%n!
z14msHw^xE|4u5r*gx_M6b3Hl42{P0QZ3nIwl_QH0g_(ahk=O-!8mc0heiezwQdz(a
z@xb^wjN4dK&Ap7I3?H!jgJYT}3aY5qY_5iuZ`UUIRu#&F0oU@d`Q$(#^2xEOR<X^1
zpEQ%X5vJ>fZ`IC8n+_gDuBPaRI3gvC)iueW7f2pCq-FAizX$a^!E`B7tBHwY{8bn2
z(b`$F5>S8H=zC<%@QBFVBWvy;IurkZ9N37Z-WlZlJ7-xOvOmM4J;H}>#PDT^N1bK^
z(!$%hca}pCo*Hu)1aa1n_BF{ch#SeJN#hRvAkHVV!5{DDY-rvJ6?RJ-%<vfEvN0WF
z21hd^JcisZD2-mlO4VFE-p$F#g$`+EyS22}(u#k!R@Yme;m@wN{T?%g&Gr|zm!Z~s
z=cl!IL0Rh*64p=<ikbbkK~S{kw)4m}WW@Zw!0WCpxV7Hw`u3?mI?IIhy)M&DhIBFA
zw65sZFwFeALDh04sH)xF`c$U_u_>39BtJ@R2d9)a4L(;?gCHEF2tb^T<mb<r*)nd;
z%te0-%W1LMDE%t$+<HvmJS^{595zZPQozdnmbUlp+J4+j`dwsW!v#PhdLDQ_fC>zg
z!{#p(Z>>_>PO+J(wys&)x@AkZ)6QM)ZeM#eoAk|Tn|h&ii@&IKWn0_&Is=q|cy&}&
zw|#TPgD(5_Qg_w$e$9#x<+nE{t`XG9J^M!P`D=l{iq3ZRTF`-OFK*4FiTdSgm$8U#
zM=m=$<;#^EwafodOKpJL?h9u)KQ+j|G^&W|c{9lMpKV$H|HG$u{|C1Vr<YObDH8%S
zIG2&zG!z0eFqa_<0VaRN8`+K=x$S+vq8@DoBa6BV7!1;L8Q3>({I*#jffGBx+CkPy
zu)iNtrIU4Z^)Z_PdS+ThQM^e}Dmi=^<naFH;p5G_7i|VFf(5UJkDmv?2NHjS#6f2V
z4<MX|;p4aAr#mT?!_NbLjFejA$K(G2|JwUo4FMtGfFHQO^#*?uyjJ)D{Qr@@XSg`q
zPEXY#V$kDV!YuJakRGHaoQMT*5X}A0AAh)!iF_$WNB`0v{_XkW&966{8D~I%gW=P+
z8+u?g2p!a*1M+W%?|&aY-+bU$%UTkr2ur{4zwciLHu7T<=O|H}(m888ae|C%I5!GS
zk0x<i6rDxp42ploqKM3)WANQefJJQDA>qKWsiz5J6rY|j?I8Oo%{B;)!bqz`T5$!7
z%UD#N#V^+HquWR{Cd)CB85ol!;1gDPJ^GCoM*C~h-(%D+8M;d}FBuRTBsWJ~T>yA8
z$6p5od_P2JM4-oDhu?-cEI?T^B07h#j?ilSAyM4C``3S$&wu>%R}eVu-vo*_EtG{Z
z4Bw`&<0y>xP7PnD?;t@0Wsx!vBNA+IL<_G*^ol&Sj(i$!jPe4s4ejt(8b-vyDAmr7
z?&KqAUZsu|N(gBg&1-&b2Qz7Pk;qx0guW4$G?%{O{5*@kMpm2e85C#zW4QS|+<Zu~
zwN7~g337izJUaYYk}U`WUPCM2JQTD54@zgbj2>-jWwhnD3@YJ*Ga5Mz!efxrLs_j#
zKaDhe=OM(EG325AmyfT=f$@qX#Z;}ZBMc*&qzh!Ikfb|Fxc>j?TblO;XT5L!PIzun
z62e$F+&c71gVv3>V!X6rG5sX)44orfj*}z#ZuEb)ywZ}NF(kO;hoK>9;_bFF_hG+4
z#E52jTppRKxg0S32&u;+hm|y(z|jOmfKVb1MoIIKVp&Y7ma>UY8O7JgMi3NR%}asG
zIVa?~?yec}EygHkd#@;!cTrg(LFAUvceK@0WS))wSUFc(35(2ePGpMxi6bfpVr2GJ
z+wp%{Nc@b6pGa>Pwew0EqmPp#Ux|tmj3s166!9N3=O$s#m`Oh7Wm;9fWGI$fvwXrm
zc#_Qn&nH|jQCkzeDZ9*cZ4_1~-Y-cg7D<n3_Ohj5J|da(*`!ySGM=#$h1C`!zpncR
zMJ{HNL$)Ah6_L~Afy`FD!}O*L1(o|-<?VmPIwwuy-=~R>2_wxe5YLlN;`c+Ifx4Hm
z>^$6Cf0wB-ch$b54Y+LQNn#;aB?UUsZl1fUEz`mLGL6b2ue4IeM)u>1{oKcgS>(Aa
z^cl^xTtWJnkVYXDn8d(j)MYdoeEMe6bHB+#SXQnxm{*H=WmAY$za)wdJzhPj<E4KY
zMam!<Cp;|`O&hg*CIo9>+!&NvUSjx>_}F26<X*RyS56p%o@SI#qrOI6X`N8Yaf}zn
z##<+$@o^{BJesTsXg!xUX=^c;=Dawi(tpO^Te_`^mBTI8KVUuIp6yX-B_!3>w@q;!
zb-5z>v_~*w5K8ackr=X){3{wyH`0IPyu43i*fAO{<*mHCV9+@;#}DNsq)R@WPK$H6
z_f<1mY}&FL+Ga1)6-(KX&y;9^@VsF~$65=<X(~^P%rP>D6djGuHA%E@Q@jbE_$HvG
zZ;j{0FcbRdZCRj8vrJwm2C*FJT?4PCfIGoYRaKdNXtG20zwE1_U&G%uzWINAuWiSr
z@y$%rS7KCu80YU=Yb??^<UrkLmgZ@t!_{OFCFDD<LJiAxH0NxF!KR7_S5!G`zBC+E
zxpnCy>)e9ywVgzIELG>B+WAz41V%|pBx8az4fPHG6O<jR-IN;GHum9iMQGZ3I^=ba
z*+haW^>;|xPXA?SUR(G=pKpI`29nlYr8md<`I<4^jxC;^@(H>dX)1o>JK9qu5(uJc
z|En!|B*vjpswti<K560#3Ki>4u2?(Upk%krRl-ymgu5nTx>>@^i&$v{s+6yi1L>)K
zK&>qnF|$4(;3^Jlo8O}mNW1@F#;%T9H@AJwsZoHLW8Mr@_)$kf&_jQk8#Q!aDheKN
z-N_kAVJRs=bApv*P2LkQdfa1T{lPHC+(zwVy=Yu2Sb-lti(>CoGiCOdBhT6VYs~j{
zye9veOhB&izr0trUQp|fO?WP>yXwc$M*Zn*j<_8+?uu+utq*plu&82KWz~QZBW{ej
znjhId2SWtbD+X{IU4Vb|!}+yehlp`qeb*3B4{Lc53fvplTrnFj>^9zYIbg^fo49^B
zUl2W}Ba&dt9udWv%;^!g7F3+3RUXe~j$ztnjxU>b`V^^4LO!Lir%t!$Sl2cgX&b<x
zNtovtxevJEvomZ@D;o(qHZe2)59<mFE~bdf-H>6TI3b@eib8*2HJ#JUFy|=ra6V>_
zBj7mZHK}GD^w9s8=s_*uTnbs4o5IaVn-`=KTIv~>rBzY;R4p~Jc6`h=%CbbV^>`~}
zJs!QpdU{MN)-$t=e%5L(-7mAE>mGBJ*}@uGcFjg1!RkhuP@C3C`l;KPU}v+}Euy8R
zYs?<e7L|j|q40kN+UhYK3C2BeWYSf(roh;_gldXucal?`Tp=3_lt>$+kzIo*(6~7w
z+vtI*O;fz{g2VKzCA4j*N8h&Hz$XjXyI-Nw(`t^93H$MC*8sFp0{y2by0-Q=?vxY}
z_s5!xyGk`;SsDW0RVpn!K;BgH<4UA0E*H99XxXSD?KFQGQU3I^9R7w6e;DHRfZ?|Y
z`mHQOXW)OLUF#9V0VRZpaRICJ7GgPIy^Ibrp{*#a9tC>;=_vM5rm#4$&TOMav5c1E
z-E8xcYiLA1Qa|!Kd*f`Q6aW@6#Z#QphCk9!N-;B^oy2&XN1?`dS^bSdRX*$Ka&Q%I
zXQx?l=SzS1nC7&w-d<u;C!^TsMub*-ca9rzf8B9qPEkP)W~r~8dUl7V&*!_YquD$m
z2CL%X3tPeFL%%Tv;*@!8?MtlEJYo@M6~}nI``LEHn~rJf5W9Tp$-Bd~iyNB4;GvAV
z31cbtT;D0JrAp!ocQPY}YK7+9aHS0h(<|c5f8Kwl8FqOF3J2JfKTBoPELfgJ_t@x~
z4U=AyTq<GSy?DPk+XMQI=4)CBZ*P@SZ~m~oHfoO2JhwK~fG262&9!NNtzn$6{dP;P
z2E!<fjN<>k-CPJ9R#|Sg5YXrC=7k2j`($Rq<q@(m#Y1O~-O)NL+nPh6dO4;$=78>)
z;H-Zxb1c^9w@v-j!u~y)%*xPVt$KwfGsC<cfl`eu+p7uF^QO@AUB@-7wgi;IRN?D3
z_v!3y0>d<G1vrmWfiBNw6?7v%&bAU9p+0sDwRzmiupJknem;GJp)gRo*;=!E)~8^!
z!Yebw_;dkAx+fWC813>VNE@N_%uMLayqbSkcT*)-YpPz5vTSc~3bkXHDw!k3bwdAK
zV!3AhjQ{O!HwJg1)y&9Cjd+4_XUBse0BmdBZL~a+s^?uRgDV2Zx+74yq#v1}pBk=C
z(31|r1kcxpm&*>IrY$%}YBp7s&%wATE>IO_Ah7P*s_L=s2}q=>^Mg{q9xS(UN_c<h
zp;JCNz2l;UH#1(>IxeX{!zELV_36B@oGoG>6)p(;aA*^+3SHbjQ30(?&m_{`9yn4n
z(C*8RhGUrBL^~rSGWoztD8y<A=3Wu@7KhW@9W4yme%Dw#-A4g;;V}9u`zXWuIz<B=
zq&mLTntOs53WFMhKG!{yv-7phW-Wifg;LARGu?BF@|=j3g2Y;MGsB^;<(<j*IrbC0
zGx;~eaDjJbf9LX9_Rd@|aeug`cV6B+2D&4nU$J=%S_@^a+B~LXDDPjPc?_1Wg<aG<
zJ{iSM$`HL%uhcvSLl;P3XV-0ab6D+a4#RqL*c@vP?`;g9FTG=0aq(jFCH8-EDvF&h
zf@qDtUSoK<ZB|y3N|?!h-uxwZ&B8#JA#`9Sr?3F@I0L1fwghifnzHi79B%mH>zHMY
z=6Eh%EopnPx%>Ro+zK_}uDzaFkJ(GSbs4I|+gt-E?=shp(y?xNpi<Y&UTjOw5MRKG
z?}g9$e%kWuuhYfqhg#-XUEqK6rCDwEzmP?Rjz^jHBwo;MS9H{gJDVQiw)4+qP&r+=
z)Sb9pI$)SM&&~Dp)=(a07mg{AJSNDe#mn7!=B&7!X2q-3@f9canBf;+g<EMfdrDod
z71R-rG2V+&`5`D_VeB1drN4|R9ZLtU-S%Glme=gN@9xo$Z7myIe};c3U)Ha2pRgHN
zjXNO-jT|;ay!pO)3<9I&WKX9cx61oBzib{3C`^;uX+BqNE~vl>4Lym!Y$}?%16VT-
zUdBVr9g0?pYGloH(JFfGFa5Rt?hbpo<X^7Tx8<kruHmPnFdWTq0a&}F;O1Oxm)d95
zbo0j6{{U+<q$3JtZe*7QZw?OvIWU)TZwnIvHkVO9IVXR$T1#)_HWI$~uh1Vbhs8Gm
z!`N<*9U!+!d<$|2COfl-@nVxb?BCxaRYg(Mv(;?_vbx!&BI~Osi;~^<PIf<?E^nt#
zuiA881PfkuZ{Iq=A0++<iG$8|9zZw^-P?Qj+oP1q;dQ~wNU1em9zO@X+ouN&0V&{&
z7d$<914MtX6<&a!pVN1y%TL?>S9Q)DbbKVv60bz*LF&X4xd0A=efs_F=aZ!6OA%fA
zm~QR2m$%bjr&bsfgK*JyU*AvZ2ctpgpgIV6SKZ(L?7p48v|P(v5+4yZ{liaxfA5-+
zUkJ`95l-oxwIet|BF3h3E()D~>ftm|N{jGWgvWoNkQU)I2$NN5Y$WWb*8sDeO%!0J
zoNdoJaWk)YCC;k<oVc4a%|zo!-jAp6p{J9mXpNVAeR^;*snb65H}!~G!qv;?yvni~
zg-D|%IU9sV0BJQzzKV<PB@-rL$Nw^<qY;F3BITH`k|T@?we`YTJz0e#nLqb37Wo!r
zFkXKc?aR{>gszvuYFtlfIWI}!Kt$<VYQ3N-l0pjSabnON6hj@2)azkMwG+mNvL2km
zqHYbkqP1j0rnjkVOJQNXPJo+mOKY38F180PQ;>MM8cijG@Y1#OfoX!o6&yh&uWuQC
zo)uSt7Yb}kZ4o@g>Lzt4D=fr&sbkD-QyqUfB;2Bo@qBksN5gPK9b@VoS!;D{hw;T2
z&guaptQ&>!6hQ6HaBW10Esr}ajzdesc#=o<C}45%Y*M31ZqP7z90AudoYydTgRTP&
zF^SjWE49ERRfxjzSUu*{#OkX!ay%b%1i9|MV`|P>`lZ*dJdlMVmULkv*%Io}ifDgy
zOKs$V0D)02PUM&8;{k$0PQ!5{LX4<$_9@Zfin&!3(TBZr>mW>+$e!vc)+8a?7_yB@
zIE23{VRGb-eVoqO=9P*_w?De`lvO#GdE7Bf+K|l56ocV1*Ei}`mibzmX*9AHj(8@F
zB3_l!S2~s`?Ikh_o~6ef1WRiPoil$G%CF-Ln~OxVT9O~h?bb%&ox4+RZ@8^!|ERMz
zGvr+ur27q7y?4Ww^Xp<E%&yivQETvqYRj4+g7mdUjAMP1$9bB~a7s;Mt4XTiy6u<A
zCu@n|rEIfY6>g9Fr7})LET@{4&6EgB{-zTCHntqJqg7X^B^k9MiAdh`S=E0ufj2Id
z{<fpRM1LCZCd_z0;~WRAp+9jPRODr!tJSlw$BE}1vMl992;(~bx>?&gso}W8s^lPh
z73m#i3`sovu@q>r)Rtzls+V~$c1B>_XsES}*U#&fuxf(uQtHJ#BsH|uv(e)YSr4#W
z((NQUfRGF9?dwnvqhYwzgt&iU<<y!&O=6UV&3rxA9Rs{Wb)WCPSr7n|?wiuW1910^
zISliKJYB%gGc0}TPKLztx8G&=C;s>|5~sTTf0^{XOe4D=-LI*=x4}3dLde)*y=K?=
z$__WDXih!XfrHgiV92?e!_IY~RhS6tY*C@nwtRDQa>3L9Fv@CNGS+`lKHjf|$WbW8
zm`ojp$X}`*$5b25P>x;otQQ`|Jt+S!2bVLKVMWEq2Y5$TTJ&}p;VsSKt`#f)TTJR&
z7CW5?*=nheI8p1ZI}Q|#LSu7f2c*wWHazx;S%ZIEFcdDt10Nr-dWVs0c1&le+e?N3
zeauCLx69F(l}wCD_`iQTl;fu^3v<fyTLd5tL&DCvobR;(h$+IJKK<wWw|{*71G7u_
z{lDo6jA`gcfVjQWS1lrFG^QW)-6henph=AYoVmk73Y<j(ey|LD?M?<$3TPXu(I06v
z;!e_F`Wqh3+VP2yC#Z&a4T<A0O7s1w2g|C;6nW8Q8R=W%LeGDuulPJamA<B{{?4TM
z)W5pZx9;?%;2Aak<su3<@H21%7Cc@<4Ji*{BFC*A7=|NyG?InUTF`+MH>PP*ZB*(X
zfJr4B%~pivkxF15LSD>)hsx5L@dW@HS$jw*?Te&Rjhla&vj-5YJ+#B;ukZBSIP)@}
z2f(~)tm__hjOc&fCh^MdX~V305;;>3?y2qLus_XC$+h&E8W!UiP&-+}6qoR&tCrZy
z`Y;IlWj8>WvaH<$P+JaGz<9J6I{>o~w^#<E=3?HCVwPiwo)kO|+YU=<kf<&k(FQ6x
z{5mv}Iii)8#1ps@8JRw%M0(PcQjMddF2}5n4q?l|YuSIManqwbW#X{K?N`l_)~d<V
zz#Eo+eRbr9afLf)5Cimc1~DkNH;8}d|7*v+LspaNaVM<tlVe^iaGeb6=e)yOC_y|_
z8Dh1D>Z@o>4<b&FJ=MWuPpt$fH|Bav3)*17)ppDeEcJEt!eMIP2p>Ijr0q=Y@A2Gb
zd0s|Tql$k_*_R9l(izj|<Cu>5E*-)$Vq@y{B8BSYm4YPix|>7ss^`<wUdR>W&~3VA
zmrVQZbvcjhyBm(Z4rlv<!yY_}sVY<B2VSmk{IFq2EQ^(%Hh%P(8$D_v-Vx+Lznn*n
z5#Hj&noC`865n<kRBc&|n{6|5H^5czB9zSy8$o|gv#M?Zr%lbOi>R3P80LnS3f8nj
zJKb=)Db!Fyq~W;3Sc+5uYcXl>`Tm7=m<gceyy*J*1g0Ur&x<urFcZo~??lYJTf?|u
z@&q$}t0hzuDzYQqKFjf2FUh`kG!BQl;p)KCAKN#}Y>I;RG@zl@dxThk$5{a(EW+a|
z-35Q{<*z`pFbi&)l^SyL=iVyvn`;O1JUh#n>jGw+!5Zo<#pmO`bt#wLeY^5KJB4?c
zDZm`=`rLZ`wWGvRfJJ19`{{V(e0ezlxfTvW!)-AZFysyOjcTX2+8VA)YNtZ1q2}X5
zJ_a@i-gTfutDWocK}1Z>4fT>~-S8?=yPkigbNxQvOw3~Z;?Lo}HP%sDnTat2qJY+T
zsBaAccc}6HZEL_N8gFYdUd>|X%3S{%$%izq!9`J-iJ?mlMrcCS8SYPSInl-ONt~rq
zOZ$TE9jm-fl|0}W;61%;!3Me2Q$8;I2B??-VlejxHJ!4ff!iRISJRXKYRBjaX;*)Z
zxoRw3&vd1+zV5->KG+V6w*ea#c2e07g3WlLK~G>aWICpSMOW~#c?G&(7|tHHxD&R+
z;%&gDbrSo7U`wU>&|WLBIfCQNJqr4KW{V=43&F=w#w*P1uv{B>DXbZ!zc*g#%-th%
zi=6E65@CZG<8+}*)7@6MoeA6L3cG)ur)0-v+rSJXI+*Z5n1K;SyG6{NtGlUes>Y0?
z9AlR7b2~9REZYWV=~b-|KL|64Rh9R%nCZ%X4B?@xyP@A1FUBwMZ#!{2EZ;hAAr}oF
z63wC#7Q$>W^HuOV(^zbP<Xqt@M?LQY>!?ucQ2A5}z{glvNX`B@YhWC_OyxU-whHo`
z9DH?g-rO1d4@<TVnCpvx4>2w;#q@nlTlt)=N{CqbpQ(HOb?I?cd2v{@$rh=({~vK|
zP;Zw3Zw?azIWU(|lNS@0NgFpPm!$?8p??tG-+ul0>4g*f0^ddP{{Hdu-<MVZqa3p!
zr+<Eb2|p+h44j<sO*)wV`1kbN%U@>-nKEZCe8RuqDw2-zK++g|w}Ip@zw+r{_~TEw
z7`&VQ7ngny%Q^GfPyd<zhV*%<a3@gV+!(DZg$vKDkx1M#2tOr-N3l_P0+lQ*eSg1^
zO2Ye<Ug38|*ML4>$s8)Ftq`hGl?Kna@=wUMqc5!)e*Tm6r5#c-oxXtNhe9W|(2=M2
z8|f%tbJ5L$?+X2(+D?nNr#8ui+$JlTcbV!MXU8b)@kJ4(1@oI!bCB_dZX0F~;ExfG
zIi+U&K{IKK$lwDmC^V*q^iqz#BY&o7lUHaoxw_IXw!7-)E2f|lUU97ozaTy-KksPL
z6^$;YeOjyao3P=x3w;-9-9la6DxL(<u-jb5yEyTsXm4Bc7gq&&Z!@=@WHZOaCB94f
zN>uK7PueX?Hgi=;zCzg+cdMksC_jGZ_qQ)f$WmIp%W;Ud&i>pmhYm{i&wn<GI#x-7
zKmqB6yMMeRB6#MFG^+9@3<zf{Vn6(yOZ}I0)4&ClW(K{uVF8od=2pbGb{O%N_WB2H
zWZh)&D!2y9BtvkTwDnB6<m*2Ox<$KH-VptytTuT~z8~$1o~TRvYO?Rw!h2N6!cl*c
zeD_9iK;EECURBYh&vB)x{C{RsnOQAHGL4CsIEKrcj(h~Ka-#H?OSlp})i)l@8|iFy
zf9a>`{yiDvs)!>P8zOtYrX7N;e5Rgh#9i`Lr8ZqcdSF=RRI=~5g!GmPT&j&=+_~B%
z)-?=Vv0lMhq{S^_Zd>(?H6%#7u{F{K3xpca2>qIBz#@W2T2Ndwg?~xufEE<@%pB2y
z5;j&ayGl5sfqZB5Oq1>|DMn;>yv!nV9$LJG5UPs@&6$xda~fCUu8OyqOIIX$s$KF!
z!ohh3;eGUa*UPie3D`c|r;=~c*i0?p?`4y#4V}_mn-NDIW@ArqC%<!)S>Gvo14lQ?
zHWK=5GWJFEHf<~I?|=CsHbSAw)5<de@R&DZ>!D<8<02zcGzE>wqE1Nqsq%lJmtdtC
zDT{i5?ot&ERr*EB&g*^%j+wxI<%dbSSb(pjAjgy!?VZ32n_p9ZmwiLtYOZ7K`K7v<
ztv1LZUST=g@fy2LUm}qvuZ@0TWybB`-Aa}<QN{r~|MFn2QGcq<HR=9U1en+FWg?={
zqE%6`tz_NI%TE5HRZadH$&tPh$1ro2QA1T(Z{Q!<v!r1btDC|<V$Xj|jT+3+khBd{
zU5DLeQEl(}hQo&5xOTU)CZ)m+D<%`W8+`lLd|RC$NC~(fv=apB@h0x`2%#Z6cTybh
zVLJT5Pk;PAHGfxrrY^|y^(GgaFs&wyxn1xjIz9QgD7u2W4$BM-+vfmds;8{U-)}iG
z@Fhonoqp@QI^V<JW@ogEJMjPt*A$$F#}&8y4V?@)3dfsdDbW&WZ-s@FP*4JCK?TUX
zDyTO9i5jo2{QT{uRYF5o@@C9LtF^@f3py@LU0l!d4S!^C*ps`tp^X4jY3R5V8e;d|
zP##Hv)s*M9T@8HQQX~Z?FpC=tn8fVI)s>Q;EUCqil=CD@0=3LV(G`Jz2C4zSh6&SA
z3%uzyp0hsy52MHT!`hS_lQuW)etCvA$Mpe<*xvb`;MY}gC04Lf=yr9CJ^q|I3`c~t
zpJk4k+JCM&A}Ty%j(WdOFh@gh#~k(Q2)v~^_ECNa@!33xMZjGJ@u?7b49*pEz9aGp
z>}7CiDR-IX5gQEI&t6sU@`ROz#Vf*e6xYoxEcVV;w5v(NR>@d?f?9?y?mS}Glp4f7
zL@}d!O;PaeAg)*E%mvh5@e{eo3uas=*+}hPGk=env{|kA37A0a^PT+savml!@H7;6
zaH3irbI^<t2lkfOvs(A)-ndG+PWDh$WlNGqTSrFd0bkZJslca!1r$RB^Ekxk9gCVS
z()diQ6o#qgP;wp!+s4|MKZ&lUovJHafRuL%L#w6L<R(iLjwz!SY?dCuDod+RjhbrS
zkAEk4!1h=R_@o5iP}{Kn*;s9MSWfPqNbM;KTgvxgHJl@@xudk;6D1coU|OrvBGmyB
zwQg}YwG7s@-Aqyq)%~~(F<D9^-sfJ+6li-sE~PTeV@=g?wo)XF;!SGLtoAL1Wms_B
z%dBKtb~X@V-4Co`*4l3WhldknhvPjcfPY~CuCwa<=_r7J?y8feDW;ayJ`Mod{cnZm
z@8f9a`9b?$ui$3(P+ts^-JC)4;3^+fmBS6+Vb-abG~t03Q=#OsUE30~q4+>N@<sL{
zN9LmiPBiSf+CgiyiR!p|U-x^X7$R=78n{0nZ_U1gn#G}gFoQR%S8@0GdI7=!-hbM5
zmx#SIhJJKxMl+z6;%JKV1n)HZ13+XH{?Dv+@=lmN9T)4hPSn%27!2NuVc;!$_Re53
zdwS5j%-_mkJY~PUPAmu;LM49B_0$|q^Zj&GQetvGYzVP6>Or#f;gnx3gV2sE8fq3i
z<%wU^fuyKC9yxPfuQ?E3JD$<!e1CQ!v~I$?^2(M!A-)Ui+g-rdn0H|2kUON7@Auly
zr)HWG6q6uZ+D~-IQb}f2uVt0w5B??mI#%V{q`vbEHX9cYDYm8RYLa?&n<sKjug&q=
z11W3W?J%g3AViAjY?8eGls+KfGxXb|O2kktb9vLK#V+UGX9^dB0qGR;Qh)Uz2$x)_
z$o7=oS0zkwmR77?ITIxK+c^_lX(`7~;LJxM3n;bcfJCnY<eulqFRyqVKJhh!F`Yis
zdc72Txr~m%CN`m1T43JjCyJ5qcoZ0DC_Ob`qf(-w7#}BHZ(KX63?NRMDvSpwYM*f^
zL~VSt>*!F3Qc$z{b4~-(1Aj+yP!*pz4a^GFk|C|Nqu#jN)<Y?jwIg<|hvz2KODS*Y
z8TaGhn-fwa#;8H17Q0yuj|e$XpIRQPT2r4X>|PW>W0jTqO6n_BHedo`vzEQ{Yx1|)
z_MbS;%`sXL9f1vmXQNHCc+Y9#T=RM)dgsUO=A^X#oWQ|{5(H*H>3>!GP=W_$)Xp;r
zn-A3|vk2$ozeg-%?9282gguZO;Tmc~3?hTHtDp1<I^u7rj+}xHW}4M`jGD*c%AGFq
zALPS2tU`3V`k3b>e$Sk_(t$$D>!6h<9P53iJR<Ep3?+o6&dL>wDxVO$(}?6s5?Dy1
zo%kn8P+eI;0%0uH0e{OFO4CR$KS`(O=UU4Flpi*rRm0acu=@=t?ZWk77?awAmh1cK
zgCs&5w3~UdVWrrnf#3bSrd4>lcX=K_XYE9v&3xH)JrDFT$4o<epbu;rRgur&Ie8Gf
zwDi3^$`}7@QVfuCMIQC2x1+kDeP53dA(^&ax4D`<ns>|?{{;t?-$Fu{0dEcy0XdhE
zojMhlUlAJ40qnOf5*oe>1S34)R+j;94if=6m$B^~6qgtt8qESEf4Aoz8kh|PdHl?I
zmjQ1M6PHkO3Kf^DEE>jt!c7X#KfQf-Cky*h1RKA`kN(%^x4Zw`^}-l$1zS7(?dx59
zpqT`FH3Sw`hvBy`!_Rkr?y1gI<q4{i)*92OiX`->2Wx}i_$aH&=*+Cp7&iX-?U$ju
z^ZmCIp9Bh)drC@nVqZW}L2_8Oqe!l!s0^kWng1k;&7C<#h7O^Bh>oI!xvW(9rr{DA
zriN{$hHDs@-q><a#ouvcbdLB-hjE_I<IH@%F28*pl}2!WqHTIImt<iR;|0Tz)h2ws
zO<GFtVUw03z20t<mf~cSWCLGxYKpGSjvZc$meaTud%9wQ<G4b|Pj#_ivb2fz+P9B(
zLbDs~1KL{6Z2x+HxxK0`TFSTBOKagnPmt{O<}wU3`JUp;Tt2IRGC?&}G+t6|F@$#G
zoH4YL5syWI)!@ls`=)^6wDXfGt`C)#(tCtTOOYTMnSE(U3o}LnhwFuD2z9b7_-qEJ
z$kEkM`BTQo!aaE$qc3K=Ln7M?VS++t;z}>M!ePR^D-_v(QCyhD227PQJ8z=c+yTW8
z#~@d;V2QhK1)ib!h$Jo2ZRS-=B1qR$G@py~H}Gn$AiLB?$bpmau<%S_ku5j*4Mqe|
z>!<<{lhxI+NKXYRb=HW|C0t+1eN#Q>Vk;{XgfpS%fT**wlogm{q1fD+Q?y#3^jc`@
zuo?!+dV#BdA1$XW(RzOT1&vU(=vv|R;b{Eeg$WM9xKf}RqlEW*<SasiEO5xnY9eMV
z*1py$eyUBP{tM`xDvAwjq5TAKhYY<c=)}RZiqesMekAv_#tI%hP3+`p2<klgWa=Gf
ztO}47tU|1%%}Xww^7urhI^bdU`%^P21!_?PVkE$S9Pom<8dR`CN!_U7{`140=2&V$
zta{DOSdGHGE*WVmz51loE7Q1Xrg<tgh$=OSp42#Ev0bQTflzr9qsczyO-@?jkm{K^
z7jgWjw(?1%Lc>ox&ra%3hK`8S;eYirrG()L`DT)Gc9!Z&6;Eh<QgoCzTkT9=E5cNx
zJ&%-s&YM{ZlB5}Ys<HSmBL$md+l(leBO-AbG4)e%FJ)KV3+QZ>8?+}OC+WGyCKR(a
z%fZ8dee+rOzdr2xu?vE9JLAj4?joLra~q4GKCvP=_NJZUg|nDpuvRV&?$yKG+I55j
z%DMRfkaX+uheQ)v1>Jvq*wZOXD-3XaOQ*DdDlU@|PMUs}4Kb@#YxvcfDvBFfgo>rg
zEdCW9ziSXy$X^|o_g_Yn@S1*^J(NMMg5Mr?nbQVoYRMe!YMqg2?7Gf3v}aim*?sM;
z>%k%Jk@lL!_5rp+hBhQz7y*_S;TzMFs_Dr;oye3W;?GY_83EeVR50qL!6%+@VMNk@
zW0^H#kv`Npsbv(KPcF8OM$W<(iWSV=vyala31`wM+NBy*medtFI*n2$a6j9q=RI##
zuq-QhD|vV;b>6yxQmOiB7P|j^*BKP5eTRJrdlXZ0${)61@*-=ER&jF){JPwz@eJga
zECRVzKr2Hyw%dGhU%N3RR-A>3-VJnr6^AmyNh+Z$Q2u*2&|^;U!3@}J10#y+{<6P{
zGs0p|&dw#IPgDM}kMJ;%j@q72`AGXl4X})k77AmA2Lvgz&m0Ob2hkZB1AE$*xeX|h
z(3;cjLjT=IRMcwGYc5&^=2RCgLX6D|L`lt8HxGpQ5JWhrVzcGWVgq}2>$&BB<3##U
z8Ud+VVSha-tb@Zvd=KQ$TBS;B<0@)P+B}4kH=V;s6?e|a?<d4H2ub*zMGz_7t+Jji
z8QW=;>`lOP%X=Bp1iafAG7EUKTWm2yyx<xXi!tI(q-iWf+|D{tb{LcnLAO)Bd3X#)
zPsQE(Vn7QA7;k2Ne0WsO-R}c`00$C^!(vUc_Wn=t!{v|h*Ejj&Umc9*B*fam)_w^i
z8>KSjb+nkZuULfw%X-(&k`L$PNM;h|;vC701ux5y^M&Sm*pkRW54N4vSnuK!Ceku>
zKDrYoa#A3$Z$4s*XpZW2ELC(qKg;FY6|Hh52$t)pZuK~$Xl(k?T$(L^wd9LMTx@rx
zV$G&=yo>pQHG^8L)U~n3(b$s_(r(Y!T3I`GJ8}WL(RNix9}y&HWmB5r%Hc%qw;bC8
zktlU*k#PlR&OuDT5{)gl=Fa!a0DVA$zYcaPE7Wv=!Rrl;sc_z~THNC<sIBrkHRdSa
z<<FPf4!CB4^WA!+uRjAQZPlLQf2Z{pB)out1l{PgZIN>sYoph^%r-P?^@Qc(Hf(8w
z_7=SU97H)3er_QZ@wyJuwviS#KW?8dy_@U|l!%bOFZ(n)@4$O&Esg<iizMrWw*I{$
zsik%zlKgd1<m1WUVy&B^sKjMQaXpI0m}^JJrrsYIMfvEVx85yLq>$&;e?kWvQB-5D
zv-b^tXiyHxF!#f;RZY{&4`b8&Ft$}zZDgFV*g|(yPxzyqUIa)-k<=VOUZhQhjfxHZ
z`gGikcEuVsuj*%{GmE<Z>7%L5L%phHZN&6m%B(|l))`(oC+&#519#<kv?n|rv|r%R
zo7`83Y8JT#HOiX`9l=lge-7CUF0Vl?6bxT5VSKO5>>Lwc#BX7trM^|smSf;<biwQH
z^$(>O<*&Nsje=^OmXh4kW>I}R`j0q8j{Oa3`;|A!Z$&C@W4gihC9s5JFBNYtXN6H0
zV9(jJ!q?{jjUDHbUm%238v>R?^=?K+&dVjLIjC1IF11n38G&nUfBkDvJFsx!Tqjh)
zrxx7p&d(B?ddZ7*U)Jlwh~>1<()0q%RaF&7>{+m|fo6=&vGhwtjYV{FZ@CH;2np+W
zo$EXiWG>$oYfz=0`W6_MJ#^2DyrLG(`|FO<hULaOvmI?_viC}>89|b8q!*lCP-hiN
z6eJ?nyjPu7^xjcDe^F<>Pd=*^Q!f$SRtFah?5J+l!G_}NRK56P9HbYy%gLOYsX4m4
ze9dq8ykmI>R|J5$Cb(uBjOk6blWudW*+$sLdv3Rlr{DS5%jF%#eHY@{>pHWodtbo&
zM#BpkObm&s)IQM^OS`-6Hr9fAbR;z=zOS6Js}J>nxg1COf3-!QX};N7>z5KfY4*9=
zkVU$aNVi=-7MSj6Y{ymtlk@nrn#cE*7B8_d+C-8+{v?OL<KcH09PRtRH1zAZj2!+s
z{1E%SCkP|SCc>a%+#X>$px8o&$z&%IhC|BJ7n3-Tvk(%AU8-0)n4auzOPkUOFrhH8
zQ?r$PDTZ|UI+u_5uOFsdwp^k7K9t|SjJMOfnlpJCzo?qJTBH-2vcTGKiwQ5KaGuk~
zH}>cnmtDk0_xXSO=aa#g0dEcym%;2o6#+Jvp=tsrRkd47ZzQ)7zUx=?QF{>0Di&XW
zEI>0_8SyQ#4gnmW659xp1!BWU{(h?1WH;IGX{`>ep6MZrRbRc?l<t1*bobre<?Zg9
zS3{jw!Gg!`?UxSt2hEqgHyRs%&w=mu>A^tY5%A0xJUw`VJl-H*z@M-3y^Kq<ZT~Ai
zCkiG$X3Cl`8R<bAp~+Oh2<@kz-oC%nnSHIIlaF$ne*64(_uJi680U=AK6HQoawk8K
z0c8Sq;8ZlO`~Bzcm%Betsn$>hV5qcn);3UqX8ee$M`h%v9x5BPCKiT&UimXaM4@M1
z5l<Wb{Qhe<dGeR<M!pRyKsTk#@YXPlqaBL_)6tsMWf^`es9rM<OCkc_p}H?ZXfY@g
zX9z-n&kz<0qoP7uF7W3CtuWs<kkI_`C*A#nfBc>Yn;5$PrJ-Mjb#(Xd?g!D)4&lIB
z62gY5xGloELlr~JWJeNz)<mT#-cI5mXNZc7-zLju744MnrX7MpYnZrI9!G0o{j>Vn
z{(4)KyUpK|UHGR5r^Q)K%%hw<<%?a@@=*Jk-7;oN#@?#sci?BhcOI_cC0o`gc4I#S
zpr<9@L6n83-nG9bqE=b$nK&u$+<>SRALQ%W%b)HcKc{@T{LJKk=RSXy%N@k(v`jl#
zX{{{855h_t75HATqKwsZzOEUAYlYW!w54I3sh)*{c%7DO8!sE8vhHKxvJq9F&^Rk(
zSWFiGk|8N(<txaZ^WF60#C^rtAs&P5v}iks`Jgzad=O*-RN#PHAoG=l6|13qclr6H
zvZms#akR*rVLL5<-wtlhD@XW2xOtvvW<iWb>q*aIzOMP1&tuAGn?GNU2eR{G?Z9Q7
za>0BMTsErcVHU`0Y>GG5rdEC{LGD^TKWbj5<=VlE`pV#A@FHd&aE6zy@F@wgwz*<c
zp^A3Jg`W@Nby}_+yiA|>KL#(ui`cn8+6p6^uaYcX)eBL7wkw~X*@IY}mT3<wP~Q9Z
z6x%nGIM!bXS-V|Et@RhvMkcA%hAt=@y)#P<GexVAhJHCGmQ+fczHFvzF2GZB$XzZ;
zKTJNC<gOn<&r+Tzo$mGM<Q>-#<C@oMro)cIm_tXc4SIGMXlCAj{<OImH+~2-T@3Wo
zL+a-@B}Bh}Dj^R`BNC10)F$WgiNduXGf<TKOEXHSUdj{qn46l^x;#PpuH_e-d8D{L
zGjUgD*(b!ufl{WBS|c2rd-+9Lrz8($UR}eD9vdcUrji7gB6Y0!puQQ)yj**v7RI#*
zQBSWRWJ8g-NK~*M8ouR?iJ4V7s>`g5-JVrNjLpn{U8}PEFiaQAI~+5Xh~sG-I$Pr^
z11X^E8C2;B`ASKp`}Kp$^Qg2};<F)M4^Yh=apK^4H1DLwevYd;?rZMEYn8dQEAwDL
z*9N+_UNJuabHU1`W(W-3{%c;xWX>>4=8G!rk}63%g@GExcT`E-BnLOuQ7_{lk?9=^
z>VQ!?b{R_GD0r6Op~GX@kJ#opkb>Y_p8xe}%VLS!A8>J5%#~H53|(1;vy9`S(S#=B
zTa8M4Ud5uqzB88wJQ^<_cO*+Sp&i(PfQ59;jBV6`i%a8{*H{J}EGa{>hT-0qR6H7D
z82sucO3qJ8hD^*2RI=*xm>X3;QWgdF-<Qrj8Y%>?q3;cs{5%>}59&Ncsy1w6DmQLR
ztDwA;mtZ{_8Am>tS%^9uE=rtwN^x4;N=6p<9!6JhPpU||5McUvKdpRHAQ|4q3|Zk?
z=7e0|qL){t!y0jHZT#bGrW@YFeDey6O2;T~6qmp~8Y>H}(|}h~?=1!wBO90YJsLv+
z#+N!i8Y_QaED2Fyo1zIHi{>t-5y{SA_1oODeUrm>)*pXrnaQpI5~*>nuVr>5l<7E|
z<u^kp_xfzDww<N(6m(mT%1uSkIJWRR{>1$R_u__ozX<k7vNkBGj+%vP7}f;hVjf!d
z+MICn$qwME!O{F=hZqx?cOUIQFK?~Ka2fh@D}8?oIen~SLKXG$u?}`WXY6#e$-9ar
z?MO0_?TeH&Tsl8<Ghd!v5N9i`9XAO^0e7rZzLrw!uxr;R(R>VLhO(3Q-B>4yp@POL
zvA)*nx0kxrNOjP~c@~px4V2E(H&1Hm3l~=2B9FVNX1RhM6ZneOBKMFpYKy$Sp4U`M
z9n62^eAEpTgJ&(y_=VS1yJQIzL_eKTjvATIH%o<p;+azx=(JGC@fKkY)623zT&fbM
z#;HJ3{ps6<F?~gT1E9olcP@Ww)v#^!-eca}EZOmN*n7>prM6<;`FzoP9TVlXi>o;w
z_Z`Z<YnFoQ=6qQKE)5*lP7|v7(uO9<y&Qj%#FCux%L!m`;n^(Cq|OhAJ4`&0D*7{h
z6H5AJMJvRW(R47rRyzq{R5g|kIB!LX%6sr`H}(<>+Lc!_2WlxaW<u3lQK@;-gF0^F
z1?9bcnDKHxHlpDwBFv;?p*5~ZXr=3&PUu79?E*RJy3HJ1zHoVY@6wCZh4lyjyJ3HQ
zIU_^|O%P*9?{eR7`cfSlJri3$pVp&ve*tE03CrPF%58&l*Snmm`J`3TftHlGf~0gG
zfo4ifl}|phbV8ZNIR!9}5im<jH&Nw!Bm}eLbHiw<9L-F!l^_byaxXc&8cgTL^Tr5P
z;XmP#QsbObP9PBE=)<tw`Ah^}&NzR<SI$Hesyn&Vm2KXY6W+H3%)aiKtiM&M<l1I3
zqlqskP<3(?5}V2dm-dJ<qH(=Cu_hhwm{8nC)YSD=dWvn$GUL}mGWzNT`Bvk@MdVhi
zGNaCIUmQ<y?<0|`#M#UJrB$Ec@oW4uK#-zn+H<39{lK|@1(3r;th}%PuK-QJB}a7j
zJ*SgyCUK0j$jM#1kNHo4$n8(8y2D&F`06yCt4`dN&!IX6dYK%aJ$MxRv{XA_L3;o2
z4lUkF;h5b<DJ;Iq3A5?_|8}q0mm%6c69O|imr(~76_)|50w<P28Z-&4u|yv7eCJJ4
zmpwun7k@7}{tAwV)(lPvW|bH|z70P;b1p5eBQDLkQn(y`d)&>-7bQGehyyP1^2KRE
z?UlqO@cT9XU)rVKCcl-3L_vjTqRer{mQHXLX@ZKtLveih`QwKNPSfWs*!VNv`v2ZP
z9{%&tGh?k{-1*^`Zx8W>R)Q%n2Vq%IcKH1t!+)<2f9|Q)P<a8UxV6SKPzetFG^hur
z;;Rgm37l&%qfht^h_ug>XByqcKfnJrbWi^B{WNbSnUJ=p40vl8rq&khh`u$4xD4Po
z5!F|k!xD=EcTwFJAvDr6y#S%ecfzocm?Y)XcpkqO#Da-is2^zl^fMp+j+Z~6u?l|p
zUw>-)Z5)RW{~CUbLfRngA*M)JG41slVLnI}d?2zD38MnzI$TfU!kJ0Qj5O<Hp;y85
zbXOHoFliKs8|H-?)ddOw`9lG-GRC{1|C3_4moJt_YgK_-+4vL}Y}25rd52vFQe~Od
zI=&ZpCU8e{F5XDBCbk;|8UfwS`39n3-hWv8*@y<_tORkLOuQg!q7mYHUAT^f%YlBo
z>Rq6FeC961Yd2pTcyZ1wcXz^zE9Qfl;T19Q^nb)glV8__jiok+3NthL{Y|WP^R$kY
z@qt<GKN~B<3?^%WOyQ?s#R&&Nm?*-*linQYZt2n$j-GJ!xgA1|C?pS;LAamQ4Sxuo
zXBJNKJ`j4rf^!Xonjmxp$$?bKRge{x3l+qwdhS{X!-Wv;W_1HX>zGsbF<9dS1OGFF
z)s>+ZXBOyvF;d+rXff_fHEu6k8)zBJoD+9~%W{SUZw8m8-^IrjMe-GNHAKaJFN_8I
zdD_5A=Y-)tSn0sV`XQN@AOwn1y?<5_GgL^03c+Kk0OCEH<b^?DKXV(Pshkts2Q=lG
z&=YLMoB0a5-ixWqgB>cRJ6;z=_w%)dmq^+E4axY`BrdBzB3ZkhVx6l$`a&m(*7(6Q
z6P#5`6}1kADL2M@iR?QU&YTi%m6L1zN@;qkAyEwL*h)J}+-+{54>5^Pl7HPQ<+Ved
zJG%H-x$9#Ja9@z`Kj*>H1Z=Jew2fwF4m6OPDhTsB_9;_|Z>Byr*AS)rP4$8ymfoiN
zA^k5C14J)9v4u`U<1~(vnT3@E6tL3jnpu=7%!W`i`}p>>29M&Bp}=d=5uBv5CJ~zC
zqD0Uvwg|Q`HL{ohJyP4)=YQBh-H0RitVS*X`mxVUjN^RD-Mk3f)6-Mwr%fO}bxv?*
zdSeaaa$47@MpVW<Pou=&xJy}ZGh+)&PeQ8<WgZEKkzT<#m7qm!j_YEF-)v|4_`WK2
zDIL=?^vuPTE2Cv?AEc2TBOSX%0g|dnV#?J$b3HxMIiobAfDW;xKYs%^jeX4HW%w&4
zsp|Trv9}cK+_+JANlj=$D~~n4%?dQ{y{J+qF2s!*1rS9IVsSIB1w@8M1VX5zte~v)
zNQUrSEqI)Wjw1T9xU|^P(;Ar$lh#syo!{qipn-Cv>UPYgEWLzjX8bj)oK6Cjl_Hd-
zmzMv~2`0~ennQ8kt$&5n_+>q^S==G3Lh-a<RoZAKqBEVu^(ekRtr?=W=z>>umv5_<
zsGUixR;^Bu&s-hKYvxR&OdCe8QEFS?4;39Xg1~d-jz+~OQ0n(a6pK2o1w%o)=q8hR
zW5^K2QVv~Iq}GO))}uX%nKh1qNypxA?V4SpMk<y#%dHP}-+vNUV^E^^YKr~U7`+di
z+LFoUS)_b?T3Q^3(WRT~wKQZmmg3T}W_cECS5k&^(`lDEX<_o~39?y=98SlFg9?fn
zWOzmYBHNML8-!-<^JLZP7@s72J>~5S8;YL#r4f`=Oh}jnl4eUa_T%5U=EacIfYie&
z-&6uhStbm0mVaAjcup4||NXQ&5Rz2RVXhqNt;r((x-p7?4dL#QSxf5jPwL|Qau>~v
zYkPbrR-J;<vD9fG1_08mPt71NL$Rx1wA}l4YG=kleJs$l8R)<3qu99c;jS75E)F1E
zHww}d8F_KYPDz*_PqiC~O$2(Y5Zy@VCbBGVcup5wtAAL3DqZa2M7)9rM7E2f%+P{1
zM8WyG$UR0EqU|Nt^D9WtDX?Sneo^s?CA`l(*XRUtuCkuvi==@S=|-Iv?HpbL6?K^P
za!+d!P(^4#u<2F6l6Gx&0D#LW#~E2#R1hQ+3}HHJu&(vvtQNw^jWf+SH45kWk?<<I
z?e*Q$7k}v-3H0hsRVgg)VZI}1n6d8WT~#IFq1+f&V}7N^II(+2wWBKe@{X#8^p)vb
zD=L)hk~n6N%F>3ny?$QKvmj?9o!jY!6$=+;^ID6aR3y+2@C@a67q&Ljv4)gYnR2Sj
z3(A{aJMy^44B^A%)DVMG=DvASb5FB?1>io~jDJ+i72_1)&qxiPRL`g_&h=EbKFbcd
zdcR(FFnEj5(Tyi`cUa*XiiQQPW)$BR`IXVLAq;e_Q)$R+9?SM^P6i26%9zq~Kc>Ie
zVj$rkam^&icVi%3T<e~3CDbwo$VRvlGnpseWm%|FUBUc0yeoJK6VwLpQ)_4^mosSz
zvVXL)_Hn$;lDubrTF$!Fl3HpTmUN>Ja1OfPOnkG_S)M$Kytsth5?r6Vs?ncx=SGff
zTv-$~OG_1dl8c_o%o5#ZY$1h2x!h6Rkcpj^Y!=T0U167qIU3=<8(*BxYP>|jAeHQU
zq&3pzXe5yZaXHpva>vOj5tu@%)l)hX=YJ(q8RV~BYZZ3n7x@fUM~T0ku`dm~noaUl
ztu^?`Xk}RmC-<FPS5lhmg;+vHJMx)yLvqcunK7kZ<wy=ixppWN@>HbO{an=k(4mm5
zdp8$8t}`5(cyg-mkga4#EYv6|)f`-0a(Va?>*nL{u~1ZNb0a#Vf-?BDwRXGe$A5hk
z6eY8951kx$I8bd+AE?iEpe@yn!0=8d-rjfE8d*;}UY8zDtvz$x&RW0M!DzO9a;Wyl
zWVOU>Ghgw9Zd_{C<o+Vg$XVi_B7^1$X8-jR_)=L`l&hUywOU&kI#<_?YSL0jM}Ff`
z;Oq5WiAZS<*|Mpt7#ZYD9P%_R%zw&i2TfLr7N4mN_~8Ikz||S?a{&$^QE)vi4y+e$
zDUUy2YR#y}1xIaa>RT(tS}q+}UK%yAcl$YFVd3@FY}EiWcS7ttMl1p}!`|T-vG9&b
zchfQAt4UlqGXxH(|9y@T3m2H>YIbzTR!4Choinj?jK@~Tcgl%7TOE3Oy?^Ws%6`7K
z@Ip->?M;%rmc)h1>*6MX`;0RvYb$sQ10(%%`GET!zKMwWFt>O3B%&p}@kK=2_;Wvq
zh?@a&2A`Hp>YblM2o4Q%?hbw|X8l_SQ#Z8#S`wGd3`vDl_wiyuGE*IF{X=~Vv6t<2
zFr7!J5%)Q1?FBQ{xmQ&kvO+f1vBP6^*xFPlIZbzb@%8dk)w{Xc_zc1!(5U-><xtuz
zelk*y>-V2Nl;6iHn!XaO_QFpf_VP3HQQiIj%J@=lmjQ1M69PFjm!X|96ahDvAqxQ}
zm()xepnqP{arejbPd6foPq<TYe|*0AcGD_ggc_Z<!>6w|;e%ot*~!7`k;n~y{cHGq
z^UK^r?pcg3JmL34)ukIn=+YS6W0$}EA%=hA<4<VJ-VOhaO}~cbV)WV%{~7*){zVD6
zBLFUp(YgYh^r(eJ=YHhyX#$*dfHSC$3(%0qbrMK0hi|rccSz4RwLPW_{}l#_mjg~3
zB7b}=pxLK;hRvjRhxDFH=&r!+#rkdFrqoF2|7~znc2sOv&CV6raD!*Y&?M8_r}p+j
zyk4x=242h~d3qzf*o?}$MLXLkEKGrgFQ%qTnCbLv^8G@*UaZ#!UQ&*d<s0E8aVLL<
z*S^$0;o_2wsnkvvbJi8#FU9M{dTro^(0?c`zY$)fMtyQ)oZvEzn=*yx=faig!p!#R
z-SI-WUaZyzT)0c8`u)n$yWH)9XxDbJC0R}pAfeN}ZgbionvUDVA&#tegCCWY!YpM}
z8UbFZQ-I%RPrJJhQRZPve-uVeQ@V%$m;mYtDUk}vQwT}wDP5rjNVJ*W>uFkSC4Y7r
z4&M-GF+w7z4ym^}Nz-_%UAiMt(|T$#OG7ra+D~szbZVbup>)@&MGB|N&Kz8lJnOqp
zR*Dbo!YF9t4DQAYIs6p{aq-KqALe0|ibhAK)lg=v?6OQ@5Y+pqSDj?0UM*}8hSl_n
zRQJ2(b!zk^&*aUR?-Mdl)BlLQoqw^<<K|{WO`?lB?qo>$l*uguO*2+1*-LatXZFGl
zdV4<CyFzQt+LX|vwNz6RA7rv!w}`bOJ6h>hus*Y7=||(JZ^_^^;dMy*i)q#*DKat{
zJ2B6tOz%%<PA%pUNb@ez+Qc`spt&SEq$zIGUCm<^aPaG6sdVZqw0q2NWq+0ueQ7`G
zW}<Gnr%+TypI{3Q67KH{n?I8?ayx#vJuA3VjR4Y&i_E$lOZ$z|Q0Woa@2si6&_m%)
zbe_J`fJRqyC<jp-QUK&r4lj<2hNh89S-AN|)MRMUlS3D84KKP-AxE-vkd}U@#KXx^
zMia}9`W*C2oV0R~Z%#5pnSUHR@EML`4mZ}QZDV5-J0lUAypcv#><m1RawWUluXkMm
zt%*5kE#O92y42`>#xR+6sJ-5I6_Cir$v_;1R8LLgc>9<%uy&CIRfq63#qZ#C*X09;
z<ngMa+W&BOPsZGSjCXtcsn6HS(b;pxbayYUz5Oxs*NennjBs!@@qg#O$X<`g1#6Q3
zbwy;ssBIb%-*vgLK$IZ0%*B|&jNY=VGF?W7Yef*u%nKg3yFM35C{nqBe?u9VhC_6%
zX#NJUrM#7WZkk4m9K53Z*w;xLjE}rZCnLedUNt@zPB(amTBR-*2NKzE`P8*Fb_l}4
zw1z5GUp9{vT{duwE`K8hbCaSO)h}IjLWm*PAn?Q8MJ|;XwZ!bSYxff!5+ZvHIBOaB
z6n0!{wdn5VzPQdBV(@xO4l!jAQ#Le3x;ihbJiP1iaybjn`R$?1;G6t#p}NqsJTvE<
zIl3yk<la;nVdl_mPFd=L+do5Et;e%&AX&`>*@so#QwDv^Qh!En$7#eSiK=XmN*Pjx
zlb(w>8uRh_A7`9S`J{49b&)eJcim{5EUk;~sbad=MpEYA4KeT((tEfvMpa@j3NcQj
zEUzj09!z?q4T@@#cN?M4j}pmNMe^sL8&OHY_cv^QMwDi^<zl56Rg1!gi)E0QhS|5i
zi)8?3&UK2Jz<;)@)yQC+s;Q~ZQ2VgtRFiz!h)>|6?DA0cT`fyB@+mu<3$tpfFF4?(
z;zbpu=1LN4ZCp7kxni)pf`$<1xD~XhJi2C&+7|6<hTpmjB}wE3w8(IFxPA9%Plej5
zQ&HUl(L9{o0u?SuMb}k?qj|ApHJt5h2S|^#TQ(RGV}IcvY}{<iAvcYuJ}eZcE<sMq
zPJrxa5WJfnvy0@zuIqI$Lj&q1b(!<8@MJYgEnA*pIVqIc#`k$D7s`w^mMfNp%+q56
zRa8IsJt}6lRr-t(#nPfX2@DcsF%>3c)jUcjRj)cN^vp{fWio((*j0q|=#i(DeM2hF
zjtzD7LVp%5;d~*h3~6rC<lp_+AkC=c85^nlkfAd0s#g4VtNz2lAiJ$dLfPlo%2~)|
zb|_px>ucH{CtpwfY*&TlWOL3HMd!Pz=6qNEOND4I;f?FgB{+D})6Pt)h|ZdzoB_sD
z6V*oub4+*)9x5TsoW^wB;&3yp(yC~}PE~@e4u3d`ca5T{;_^HE$i}o%aSy7Y+OkIN
zAfArYmVIN7%ZyOdh1HRY(Z0{cc+_-!6&DA=&XA7D%|_aRg3GYG5<N!YWCUG_9$dxf
zQAA&%SE%yC22{o9W1madT~vOV%S!fyN#$D;61DjwO2eN<Ya{0`b>0%SDLz$H7mbTV
zHGk_(%Wl)CK_t1Q+B9klj9p}Dyt>S&#i+fm8GYO^YUJORmbt7rWfLm<bIhr`?h-ts
zh?v8b2-U2+=(X8ngUPI*y;is&(s)->(e`ZfeC4pl*{02v!;188L~(r0u!h;jveD?%
z=4><UuNhWmhw8y!Ih7a;Yztj8mDXw_J%6QoSD8^Lk7BB~WJYWGosY{&l4a>e?v--#
z0tJN(yZjaLW%?npqLA*Dl%;-p4NnzpzNXqTMJ*7`Tq|WSq%+)$5AdVFJrC(_xz0qY
z2xY5uge!kMH!0H*-PhQ{4qTc>Cyk})qwBy($eK!YvuO4VgYy+h+KY-O9nG|%n}7D9
zRx>>rHa_cKE=?;-lIzmSlJbhQvS0j%>e!x+Ih?Of#zOXxSFKVh%Jb%4E;4+v&X!Le
z=+nM!TG8t+40wsz>?(;$SGuA{SjhI&>T%)LAXzsrOg~HqU87zq6;;QiQ9cWy$~DEf
z()XC2_g9_jpqNAJS2@+;nq>mBy?<qwS0D4ah4VGaTF2eI2Yk|V-s}@0844TDxP55;
zEgtp7hzV=jlfJNWN1RXkHl5DrgM<h>+u;XKahtn(^J_ZV<Hho!ErB6X1EG?U^YZiB
z(U!z>^#mBoQg+mICf`pT#-q`k@Yxo6Q({!k($01|?+0+PjcqRzTcKKK-A48}pSq#E
zmoW7zRkD^e?r}*=&!<sdMK{DqCi$AW)jcv7)^6uH`YN&uHL~W{l)dgT1|$=Y*)Pns
zJT00#X5Ie*jJn?~myzHZ6ah1r(Y7=rmkl5%440{o9Vma*TH9_Mxe<NWSMYK3U^a_)
z28#uJi8c!)h+#QE5<xIVqqcTpX<F!MC31r7x9=&=ty`_h^t3En`yi=?Q%x4@RGlia
zWui_(2@?&8#D8@ZgdYq9q4S8<!X(clMp*4wu)-z7f)l}I<s@R#tgJ-R?nuWVk`FvU
zf|O#(cu0RX@fakEJQF^WPQ;QGWK^C47Fa9+$w3M{=J2hHS;DtI#v@3WIVfJ;X+gXY
zB+Q%wo92wrO0n5q3ynuuvO?=@uO~=(Q{D@0mwFM>ax4k$bix|mK?*!Ge(8|+f$4Qf
zSyDo?4<JD=EywUn#Vlb9q}hWJ6KInqOm7UY5nz9YaUN>~xWsARZ-n7KVsZ9J&^$@R
zc<}7B>@Q+dqFU|`QWREYB!HBqm?Z*A#7n>m;Ki;=PFR~KNJ&_i9S*px&+}md8?sT@
zflY=TlrVD&c?@7H!dOxyoTkK*5zeSADM(fyL2|;`koAIa&SxnK=h?YQVHR(Bj37l)
zEG2)05#?`L0>&6q!jmSJNO;XVN~DFiPGJowd&gEK3c`EEu5u7ue!xItXCyoiAxlOC
zrL$xqdDa6Yfm9yzA^-!Hq6pRkjf8XvmLWkHBe=6E$c+&4vzRo*ycz;D#K?L8Hn0jL
zFCuY#u@9rp+vUvAF@NhsL>OQV#ClwA1Mz>4xZ?yv;vmLY4}lzWP$Xm?T%Gki(h5FA
zkU_{9X^~-?!y{pxWOP<DkbF4${`=zn=bN&4JGm^2*VVi!=gopsP$Adf74OPLbz4u%
z#WLS{=+EWFZ1SqQJ7o<ZfGz1WJYVBUL?A*5T0i{2Gru`COVW5({O#}mKu&wfSrmWZ
zs`>5J)mhdOSoW%_Pi`jDQW#bin5}8bdM@<x;~(ytA5WS|QwmxR9K%q|e+<`>!p?Yw
zV`MKeWcdSu5E8}JH~c}<@&nS2IH2+exSO;!OOuPYbu~RHo73X$o8#hrdDkp0+~J|v
z^AG?S?4jc2d|u^=tX_HSG!rLpFcW|0N^mTkZOq7ziQ-jNUzBxTL7f$UEPg6p;{$ec
z28*W6DVRzkT6D6KsZ&QK6{oeb(nfQ3RGi$NH$2~;X7f+Q%gyZ5W>(FMlj5)Me&V0+
zer}qZ#qW#aa@PEOdoHKdb@8z-FY4K+@>9)E11A6@|9SO0?4aeYp>{?iIaz<#hJu=8
z&?5|OZkwu}O|BNjw3^SR#iTx;HT9(a{Nbi9r?UmLcNp1O;bDtu`5{|u97zS7$XaQT
zfuL0xxWYK>B)GHUrNHzj#gEnds(2&5BLn_36p|zFqXK({2s5@BV!|dvP&&HFabJN{
zkX&jkd5M_@Xc`mckQph&48(sEBxWngies+Iv?7IL)FN^!j!8codg*7~sPs$Q^t1MX
zWIHfEG~$6kv{pyy(MK3-Bme^ph@5_Ofao<M1N@B82;@<nI?-YM*W%AdaG`TOFq|0x
z25MUu3>UXxI9gGl(~n1mf|VY1-%*grNc3-72J~1?!>76ri=u#bz<GZZN*5IhuL>Jj
z9$^HtFxP8PMs5xXL7_8+Sl=_k28(VFeQn~-60xLwCp4VhO&m{?yqyw@4o<5Tlk6<9
zhenb&y-4yjJd(D~@%lHAH@;6|-i_L=-%~@p?=ikML2|G9TVed5d(H}KjI#oLgV4Cw
zcn->>gFWQ?9<rl6<i39{vmCn7*bjxM#?+hWxAi8(RlN!6F)g=qPC7Ia>z3PjC(&Fz
zzm|J#=(^=T*wFhe_f>T{TQsxjhq}D1%SEr{=E#csa<8woWGk<fL`yY5(|RQxO8kCJ
zA}7&deO(6ZW`2hg+BMYYTdB1g!by=^eUsV3KR_MbLoW_hnnr)|NKF0o3S)RA#yw@J
zrLYAhNxt}c0#LgCX*+ngPlv*__4YluTHqgwL+bN+vSVm2waz$6uV^lrFL2l~B>8^I
zMK6gvTcxGv`@Q7QUiUiMMVUE7Y8;UsXs7pcm$gaq-3b3BzGYNSDx5y7!(j_9IEZAy
zu^%V$d^VZi$a#O+6z3qY0>GJKetTWkv+3mOMRn86u4n%#>&1(YRsCX3aWA$MS9A$3
zxgiAC(OSp7mBzcLi2GL4HJF(?(cMX$P7<_o<-q7ps#nC58q?45ZS&K2D}#J^unT<O
z1$LARhD2RgYZ-MeX~{R#-|po2g^mY_dUp_f&%fW7>$rbjljY{%dUTp!H)OKpw+$<S
z)`EUamuNs(3H-jfnDrpAXLv8WwsUhYd!U{7d&KhY2F!X5&wscm7qiRxXbVY4=@h>?
zBebpXBxz~+U6NKf)`r7fc~Q+|RbLjQxU_`g^{c<65S1f}%34cb;O=-`E-3`*_)XcW
zTv7~JGYx;Mmz>FD|9@lER@ijV#!C~1Xj+)1ZTOab@$R;EciT3sIeP@f=zhi0YB<Gg
zeXfU`g1TDUQ&4dJxggU4qYG0T8GhsFSGV)`U}_`Ny+hpg&<Dd5`Y^dsFzvZjYb$ID
zs`t`%PM=SMP^TZ+4OhqYT?e;;?0~Wx$QIs*f?|LAxnoA<jv16Yrk^|T7S1H;8|Kb#
z|94OO!@y`kdoURN3^AiJ#O%uu;~&I<Gv}Gyx(tf;`*7`~txmo@qSV-aH)}_Av#r;g
z{GTJs-qo&~So^3Z)>$i4Y+ZFae#sV|;Z^6#(9Konp@!aXW3OlP59QU>>}E0EIE#F6
z2t{3|gG86f{~aX*{|^=VG;j6cUi^1?E*H&Y{t*ky^5VAm?=Sy5uRdQC(|_HaSJkKP
zO+~VH2$9-;d+1`4{6BC5G|9o=et;%Y5SxFVH?i$xA9IDcp9X#fT`GQtmjQ1M6PLSi
z8V#5H<qZgzYrHlnm;NLO443X-8km2D_Z2+<>+SDPlEnY%)rM0LHhKqZvO8}CN;t`m
zvKpjfhtGl1J?)*f!kL@)@!4zqG^mM1*{vI?$=9TnPk%|YgMtfcWmWo~^&PlghC3r?
zr%zEtr=2M{dnJN)1?NvcJ~xF8Af$_2*g;4X!J;*7NCHek;&0}at}Bgj<9&ZSIY?fz
zz)9I>t-jtB9=$C>SeM=V)hE0b5KBXU`uX!0?foa$)LI9;8G&mpuXA@K%?6iS`jn)P
z&w7yszKyOLbiCQg4EAVg49*K=UWorck{c(b6VXGk#S<Qck_I1%Lfdh2etP@dWa6U$
zX4Z3?NpNT}jgJ_i3^Dn$%_M(Hq1Y^%R9SZ(#2_U8sd(gX=sc}I=_WIyaLQ_Cw|_TP
zGRg}b{hHmDLy)=&k7#W)2<u^s)&`U?fh*l)V(<yWnEaL>vFnc@u4!F9@=pZEwrFS@
z$!CgeE+gX*Bh>;t)VC37kw*gC^zohj_VV`h-zN^3xMWb(aQf}XQ~G~HvK0h<(h`J5
z!T$Qs>G!99aI*7g6ckD+9g0&xoI+r+xM(2F8<(Zb_&Z>60yQ_noRdTBslwdzeAXz9
zqN7oq*2wr4Eeu(};IQe(Uw@p~onOA+D1jC#VRuPMeY@F{RFwi(niMk8&~~I+-1#L%
zw<uR_fhKlfIhPbOb98^3VC}-<WyQMA;nH7Nz`l?LcF5|{MrvUZ9Y?QNeUn+?J>?&g
z7oQR;t@Dzqlk0xP3`WT?g+<%%*OJnE`L*N%kpgy}>v{;soBUc*JkqZv<)i#sQmy$l
zjW%)etLyh`N$I`(T5@^NR>Z=}k=ekxxHH#>EMr4NJKhq(NC|(Sy3L{i9fllr(@bKN
zbt93Kmo5y|(g<S}=X9`7w9+Z|FeW=76oVTk1)ftpV5w=N1@5I5rr){LFz-(J1*3|-
z8?t97bstn}q=#|(oxPXxPFzXcef>xYLmFt`!F`a{BpmELuwFQI$9dcF5OBZjba-d)
z0d7|n_=8Pi(u#jp3)$s0d(fj`3QlT5(c;b=MUBCY<;nVhDXX5u%R5Oo+bi24-JFYe
z3y?uXv;(6uAc`Y&2_wa%w4{>CB<_Y4r*U)P7V^rNu2tDIm|-N>@by+R$g(9lHOXks
zbl>LvbT>MbjFnm!GlKBZR85$!>$s(#Z|gH~Clpwi^?rXWtw-^3R^WbL<9;y%gq<<Y
zGvdLJniiW^3yNfYd!C8^nzuB`_VXovjA}V;uugz7lc96Uup=R%ZJRY#Kw{dYd~fq-
z5v0wnoOh>+wK*9y2pnd?Ma%W|oe<ap9}O?7sj5YAGCn+-Ho?nw>2rcsavF-JXlaG>
zMg=O)M&^Id4P2rX(ZGU>?53Dgeohfc?o{}_klbz^CbSh^n+1jKHk@$rIPG#)xLLbR
zTpR+V3r#LLn4a5ETT(DZLeViRA+21ctCCsLJ$>3$$p8?OV4)jIwON(~;HK`2v{J5Q
zrR%yZxaq|dsbZzg^5itAA<reh`dF36$RM0stVVyTfUn|^RKlwG5ja$GnsCU|_W_Ug
z;}GL=01h#VcIJ(9zS=m%DD1$Y<Fv~-v{}0@9AaD|aXHVf(8i&X!f`lsm_<R^l&Dna
zxuxsUMz!+kJ*ak^W+@MnOIu}J0?H(86_=)M<f<39<D?gW8@if2j%2x3XVtVdWY(_Z
zq=|o+&bkkP$&%4Dd(NbvS<Ijwlw?e@;V#6$R~?zRmW)6LqUk6iMySn_Y+_=g@HVz=
z8gkCHu5PdbFGgTL_(&~4F5IwVg-R!^^)+jH`tIwA3}_*PVFbF+2MI)|nLKYeOf`9j
zv<_5UH~IGRRS+Rm=ewov^$>1S?OAeZS~!2<qriFb5kOILnv}YdLVl4!cOeDivOk}y
z$$g$V#5VIOPN6@aI#4^xr?zX?lTUFj^X299)y$_Dg~Rfxn*Kog$7)Q@84#neBLk|*
z5=ahP2O-+mty`&gUssdIQVfdKnV<J{+%^py!kOX>d;|`aoF=Uz&HC#x84Kfb01kgu
z<j!e_t*@bsLyW=>96C<Bj6<8X>%yUi?{gK)v~j4Ua2yUDX3@N_tLf@#^+J93lXIey
zLKm`)JFx?`vmn4OwKFr^HaT&tU?t^PtDfAfVN^(^Ztc9cyHr<aYWY)31Z-hts@w<j
z5m;GrDpqC2%Kf!?&gB5CY{>1Kk#v8;N=9J^RvxEa#>&mwbzvprVrN*HE4z2FvZQbv
zRvu>22`eiWov+^5#>$ey9;~d$67HZHWq_+T@6G@#vUpni_bdC{E-R=;??#Prxbo~=
z@@dtr_1fM$_+Zq03k>IH8@}e#3`1jmoV|&kH8SLTtLGl*&SS?9hS%wC){=j4UZ?2k
zEQxs?+xLgAAD=1B4*{gt<>uvCIcn|C4Gq1$&Kjc75_qq(vJoeQZudGr@)mvQhpxP`
zC%32KXf*c7PCt4L&L|Hye7#xM=(ZnI=;Y7L+8ElM**#T1Chtnm5{GE^L^Bb%uBH;;
zorvL9CvLWCbfghVMClh?Dh+=L)RD_ZA8Bby4;Y79_YU=OR&AsYmJR_hM)#<6R;4aA
zZY=v=!ck0rP&jID<kGqINr>5|)VTu%WKSEkL*P=3Qgxz@jZkdxbihj8a8+Kmi;`z?
zNu8bh<1JKdbVtZeN9dGaXtgNbZD@T$bTJrXxt$UwUU#SHT;rD=6oY^8a>e5F*L}@o
zpFjDM5ip<58R9y0AiG7wnmf~STXLSxIzf-FK?w=VnU<UH5KCwlSdyW*v(lpv)((d@
zGbJ~r_KqT^95Wn>CacYv6I9h(G-<@x4?S-jv~)t%#-f>Z<Oer-remZ9g(1FQ(pnq4
zFwqj6wk8?;OmZ`u5w3p<fRl|7YSgV6`!c2-_S(_gH)2*nZ^uMZsg4puNicwEwhA)t
zrn|0C5`xz@HWq`sAbIR$eoWVlql|0cv8z-a3V{<l??=y)U?ok=z_?JQ=Tr;5Nq@+7
z`E7YQct}n<+QaGiNJ&h|W?ED-c72R9*p`e{-Qbw+c+b5vZ1jKJ4Lamy1~IO?3c{sY
zFAGqCV=)C_IBRj8r~9|9YIp78F`1}@+mPwneeh%HtI9;lO*y;KC0BqSLKjFS&;2->
zU-Lh-7<2PbV=;Od50VT&2F}HuIp+iedI3OZa@@Hc4FR#$-sEMw>m4ocjB+h1=b%@P
zUhMMCz1{yUkavF-GAAr$Qk!;`T8{46Bjl1a=UYKDDmSyyw=(V#mdKS1CQ+rKf{y$h
zV2PX!KV}JPt6H(dxL43Qzcu5+8aLkCp@h!xF+Phl@9{lc?vvs(Ud`%Ym!d+s-s7Z(
za^ceMdrs<&ui@_Ly1mWvF`rJby|l~_sN=LNIO=c`XpMilo8EBns^_2*C`^C=b{<qp
z$J{V?SVze2Q|gRrwLCMPHfN)ch!#|jWRz>sf;KBnTa1yStQgGx9nWT5O8@^w*#|9^
zBLf@Vk;LSLom(bSO<}!yl{0hBzXOlQw0$Fc%Fj9bgO+Vx3J6p%>+Fo%{(%K8eop0H
zWp;<$d6F<Qj@QlSIv<vKT8s;=U4mr)(XU{R^}lt(K-deY#tP$O6Hr&*va!#Xx2OLD
z5$)ZVu^0*y12Z`_mr+dy6qmi}IVYELYZ{q<D30UbPrvVUQNKRFKJ9w7kx^OA!#~fv
zmz{DD();m?OxMjXI}x5H>r-BS`Ry5bc=&Jk=eTbAR?phk!}hb&5~S=0uQ)hs)*wY>
zH32FRumDH`4n4|nVm3v9SfC|=n*2B;urS6oaW<|`aJ=S1)K)o7^~C6kD1}cV7ww^c
zoa^I>+wzQ47O0n}Pp|9>$~n{T;^<!FczT1taUW@uP)KhZ6q2?}6lCHtzD*cKJ8YK5
zVkwPQE+vsIGgQ}RiJ(P`$i;L_I5Hj5B)T{k$_6efQ#N4)(H1y7tifRnDnh?-icsFs
zCXj|m5kbEbb=hTc2r5{#)a+uTxIONF(;uJn4(A7}f-|0*9Tt?~Fd%4UFz>+6o0?g#
zj9=I^!s@mW@topv1rdrg-A0A9;Aw9Jg(RtQSfH{8<3GJ@no39t{DRFf;ZcfFb2@T-
z3d=U-5GrHf##o|Afdx{Dd5<H_`G|NZp<zp?hr&(19t?4YYbZ{$h3)Z-@+K;O*ryl>
zq?~A0gFDB~me&EC3K;sG#%w2(m^vBzU2kL>P>j8WvfBg$3d)jMJ5o#pSHZ0CPBRtW
zRPMOAA^}<4A*@N|Ob8lFOMHCmeo1M2S)gs5LUg%^>4A*Is~59?6<as_`giydt5qv?
zUhxw;+FIt;BU-GdLF`vxrbNAeUUak~**1Rs`Z;tne@%rFj1>)`mt2ChFRLIV_n}{G
z*0TONwh5)lu(zw_jRLngsd}YX%R5D80Z(jS*c6XnGF<8TP&pwJm!g~W6k4&)A4#`n
zX0BUxh9qNfJmv?fnpR`VGnrlaS0cYJ!^K&sA&ujjS{cV(MB}@neX4+e=a$jaipt+=
z*BExK)KHQR&-F;raLIMQtcy}=_0<{0Dm~WoaxTuJside=mDOM{i@AJm>MXEW4uv~Y
zU(mFY_xOfo4cMh`DnHY>6n@ybX3>JA8N{5QI<L_uC2hNsaA}gomvI$0h;P~%8C}OZ
zv#UMQsUf?mL#th3F2u=yPi`s{i7%yKRKCQBywacL2#`ilOZeoEgmN=8Ls=v!2(H&k
zbJ40|ibb1x60h2b#w+qmX}2dcB<YzXr*@L!^(6O>GQzqJmXsDFdmS!qNv+;MuTXlF
zE8eI*?ii1d-(B^<l~be)8x!WqeinOhN2-_Xh_;t|HPvgmTynpEdTtY)rpVUkt=P&>
zeW@0}VRjREnD`m<Q<~#amEBt<-;9Ql=G17@+jk`njkI+vyy&r`yLf6KNc#?pvuXWo
zlxu=pcw)3!`ic~wpmaI6nOVz@Iu`xW2=m&_=?X!Dl765QwvlDB-wnjS9KPPp-T=zV
zMW#Kj|JUIrfviG*Y#v*I59|4%X1r!`YAE#WX$mlQew`O!5sEj-l<wPM?+q~P)}UOx
zT#w~D2qjPw;s+6`74|Bj%KiLKQe_UbN4|qpl_#CRYowAZBSW`1^&NNAKFJOU{DGz*
z$n8D6(Ud~Jr75I)U!`k-dl%ol02lwj`w+gh!rsQW{O`vlU7V{FX^(vu=Q52tf!8?a
zmD9FcJUZ&KcI)l_(w+Ua`#(^tl9z!B4ip19H87U}q#ibx-v=8rw<mELnunLdp$ZZL
zGBuY$oF^xj&oDv-mjU@2nPI#S`t-~Dk54>`&zU#$l|K4kZ|_fkKJ}B)S~BJQ@UQPr
z^gu~q(uo0tVOp5suiu8>pZ?yH&CEEy!!-TDZ-0Fox;tMZo}MF~+!(DJJb@#~YB0t#
zNsmT6mFHX|0=k`-`}rDVe=ie(QiMqQ6M$;&AW)f$0p_`#aN!Lod9=m%!Sth-CwH?v
zs9FsCL>_=yc?4HXX;{Wc+*vwhnD=I{bQ+k@6R`B7g!7VFul5S(9YbrLNN47iP+6v=
zO{3(??TRK1`Yz}xEh(B(4E<iu_ARnSRHNDJ$hNujbxH`iE9<iVe=%gL*@&`Fi#8l>
z7hcmU(Nh|w{HnA<GY!6WauhVII&PIMMcA;)X0!wfV{fBcDSo>Q9s{X!lUcnkSYou>
zc66}B%jp>$Hx(p`&{hY`PcRJAJ7G!YcUM$|6&QNOUQ`?~WZ91nF5r^XPrsJ15!BF(
zKv7sJjw1Z(B1BX>e^5-DUh57*48|}pq9eMfc#0Vd?ck5=T@1ACv|iT(Kr=Yvd-K|Y
zpcJZOe4b}4fGINt|8}OR8dj)vz;7eynMYUG5md(sN{30P-KP>cLf%s)gmDb7oho4s
zv$hvkndpc_NYFoEA(CVAt?fQFNhZoALo-f;w=Euu#Xl<nf9hi*=ggoLV`&LfX3&mq
zLfOp$SbM<?TJ;A4u%1t3Qi`G-#VeH;JgJ$nC=<v5a?#M-`8q=h<Ea%n#TRl^pHYT!
zwMe>@=8r`L-CK2%9o4YjJf;@5B!XrA#kJ55Pu0kiVrobi&+)!avK%lh#W*6-$=0Qe
zh?q~c77iLQf2{AU6kf@6I}VaCyhH?2*_1H7luBl!C`-0DgROiy6a%D?Oj|uc<T^=J
zZ=y=cubiL&a#mxcLL>5%RVleZXhl$C!;TG$q?xJM<d^#t&9McbwP)6tzT{PQt+|m`
zZoOTj5pOGtlR2NX3p}XpVx!{@nh)_3!O~(=CRolTf3zb+ZX-ndBpcC}M5i7RiY)w{
z3b0&=L!uqtAQ~@87MIEZ($LqA6G`B6(e~vEdcYuT(7YO3uogX(>F1m{$D-=X6@%B@
z+n%y~*S!7dpJ+XwS9yBHgcm(e@-5WJGJch(<c=KW5=+bz=U&1z3TM>5_0#87K7bnL
z^n?-6e-+=H*CUBOTT9u`k8CBbVz7_Ag6DcYi{~!-k^Hfa>svH&st<Kkh+Zr4sr9k6
z%*JOtQomMhqD!F!F|WC`S*Wly@lN#3Ld@RCh&OtL+fg#~3Kx1~$x(1%U`-u3gITD`
zv8!l8*V5yd<V>Dn*3eVi4w%8<MGTD9r~t4Sf8Azwi#MbNadx<lOn|#rTA<MxXxUTU
zo@L0Rcgz&t4%gS6OkM5Jr~3Wqx17NG;wDQoWDHL^Jv$zrQdZmXzoDm_(ua;h=;l+G
z|3LbHj!cRvg7c|sN;Z1Us#aH2E8B<qkVL8IOp=Ycs+72VOy8}?<jyzNye|MwbzL&)
ze_c;;hpwk$;HBxA-eTwTNy&WV%AQn7CSv0C>5tFP;6MHQ+iDoXvJ#7K*}O@Tiql%8
z1VOctYx({9+ke(sp+m<&cOxgXY_!*w(@*Ps5*R`wzhWFI2Rl!rEG-R|*xB?~GAY?b
z+;g>U!jH#5=`^MTwvT~S*m=Dk19!wXf5p*1+ET`gIFV!lJbp&&;A%z3yH;N0AU!Y#
zxlAsw#vP_ERZy{ckanMP@vO)4s^xyFH`-PW1n?88&snU&%8<_Xo0HBlUcgQyN~+-k
zKZoCE)=#@o`XJ4pf8oQg`0z&@OwQxKVd(eFhYx=Y|De^}AdJ!k!Wvz*UL(v0f6AYH
zBwI*m=|ex{W)eFkQ;I2N*2$uJc-_-o)mZ9%OC)YFEml>_ER~S8X8F{PN`4obabZt9
zZ7r6mG)j>{Iy{>Vl%xzz$a~Sv^4^(lmRHa|fqT%ZqDyvQ7FN}kvO<r$%llUDu*4fk
zpwQa*dm}+D2`p{^Bd9JLgt`>;f2qWxf<D#<DSxU;{n%VB3ZIhQO}GwEYXdHhnHf9?
zEiOa8k<hYL8=6<m#z|3{Gx~Y`O3)QO6xC^)%b3Se+frP2f_oH%b!>(6SaUrHT;b98
zOJ5feSruasOU3R()%8w006cfPt6VAGnRV4wj<f;h@mX)6X&p0OJP1t(f2?RKsplLF
zlu>cL6Nba$ZoptD#LBxzrhC)P@@m>2(~pTvm@42yF^x>3YU8G+t+Mqwsm8}F=Ddks
z$ZS`=J&$Mba-zGXABS<-z*~*M`Y~ogy)rKAJ-XCsOSOQyZh3EMpmpEEkB20~(MA&w
zLh7+GLHARUJvK?kno!!Ee=r}0V*_S6W&)3a8EbB8Nv2kaElrngcok0f{|E137&hP)
zW9sl2c(FbQm-AV<-DRpfjH>twdV;9BHlw3{0_`SKZ(YXk<Q(t?h(|N|TYM&W!h9Hx
z4VYm}A|3-X7=}=9Xz?!a#RB+!BDjdXAHN^M{xC)x*z=Hb+&zxpf0t~MmX7Cf0gnme
z;1!eH&lH&Fa#BOREf-c*VX?FPJU*WdoZOg^Jj9~tSh;KuH8dG&HmbTq=$ahHRC^@l
zpk!r1I$BTiSoH4-m+?Ej7%WWguUn7fv4OKioUHsvgY9M#`+ACC4Rcl<=_&Tj%b*h4
z4b-HxM$Wa%G38`fe{rpMB(hm-HQzx1#u-_T)1IUbrDNPmwxx5|H{zoXrkehoTkq3m
z{G>l!kF7S(=lb2MI=gPufs2`~_MHFMTjZiXvLV_|<eap2u|9~~4kB@|WhbzvG}27H
zgLfF+7w;Y0HX7TuZKH8x+Y_g8W81cE+ewo&Nn<CC_2&1ycinaWfU{=Knb~{xY<#{*
z&)R9t<LVNETis3J1)_01Iteih7@=Y!Y5tgUgiSMXoZem5xWGsIjIs>W_9H^vWeIoS
zaH!{;dE}$Krqt`f1$3xYEoMv|MbkfI$QxTjKJ8LMb%9f<>#@bAL-X;yz_*n30ZK<+
z;x(kp$4TsmD1wTg(=&%-URN+Ca+csj?$`0fQJ)A4s2w%g`4=VL;Tn-rTY3XgmF$dG
zd~E`@z4hfQ08GL7J2IS&lllK?Oyx?wabv*%WlKdBf}{ep6&&_CQMzBXj*&ysNy80=
zjr!cr^L#c}T8DEsE)DQ7*5y9^cchc()tqg4I$<nj|6V&ivHKE)Wn3Y0FjZbS-FrGN
zh$r$(Z4&fyL?aC_q2mQRd~1Y_&D^)SJBS#AN@9pa;R#2($x{9r8r_G0-_E;!A%(;C
zu~5O8<?{osE;i<S&YCl^`VUHj<f*N3)3Z##LM6`bSNBUXF&}X`;!fY4?prFbQL=DS
zMemNzWch|z4D86ZAu=<G0j?d*T<8#@bom9Ux1K4H5u=0{xaR{p4j}&yJ^H)iG*EqV
z#*E1E$qpC=+4PMo!=DSg`y;2}**H%?K0xe#=LE>Tc!pwfAZ<jwiId083<q0CQ*cNn
zNj(896Bvdw!n*Ep5GK+Q7&viw*a2Cj?-!GuQsHrD;lg<OjM$ZV)zmN)$ZEY(I0yDk
zEfKEo(VUWqE_+9eNs^U<@VSArKG^=DO0$%@AL}9B`!ry`@eBA-Dkj}J=q436vdb*T
zQ-B5=my@Vrd$xwfhUFK@uM^+F+zydPc)-4)G#Q&+IFwWmMO6@R+g#Y$w3Fgw2(>}f
zR^)2%c%-eZJ1U5n_mehE5&cbF9&oR68Na04>A<}N77@=+9XowysWMrkhO-={>7_0-
zCkZXkw5O($R=``qUY+c!Ie4r943!^_0h-m}Dasf!@oM#JnEIF1yhX+{G?9E(nH9!X
zTT5)s8)@4&VUh+{6F!{}{pjB_4fPPWu@t;(y*V^52z_B)=n;7Rsk`lmf^V^;th2=M
z$;%8XTGs~&aVYDnkso1hL}pZx9^vk>1VAcMN2oYy+6Co*<6d<9+-ovgyT04^0o0n`
zKAdkibp|<?#>P@-?lQJ{jd=9+UMG+Tqzde!n4_EiK-b6FCsykdc%=~^u(jcZ%w0@+
zFT#ol(b^>^9uR~Wam!oqOu3*8hm5%h%5aAg`Z`^-_voF+g>p;6-6P01erU&zG}Kmi
zn^(Ucxah^1s#3a4zWv71WLaK$0iHDqI*Kg2V1))`n@03}=;(eb|9ujQN=}N&P1rjs
zotYV;cPWBl$k~YJi2ZG5e>qu&j06jzUPQz(r^p+>N=$o~+idhPDptZn<Giw~pB0k%
zS8!9{?kFLT=P$b8YNe>9?{A^;Fe^050|@+A_erg`W8jbCWo7Li)~usw2ms@hLn62n
z2UG6mgx9=}z`7Ne^DTk2XYa}=+sP``ir+r<Dm=P#JvrM`?(LuFT1wcvy5PO``qp%;
z@PJs_x$eeJ+}M~g55rT&7dXa>s}M4q^2YyTY*=t$*mX_lu?UMohg+(_l?yuh3}xX7
z?pk!L9xiaqm6Sq-@#IJD6VP0`7<xXq+wjdvn3|^LNibPoF~;LUp+Gc8*~JHk4T5HW
zM4PnbGpDXl{WS?;*12SpUE!_5>!eZTT8u^kDYgA?OKm*6(xsq@uMZQTmktBDw7r`@
zUD%oLP?=XbYG_z^chjn6Ry+F{TVt3@mH2S~$?%3X@IMs;7gy&Wsjv=6@L)WgDMdtF
ztzUm^nblt%^TtSQ7^!j1_)}nJv#sA-@Fn3O$nNRk{_|g7ATr}JQS{WAKKv1|excTs
zetaQlh<k?1!~f&nc>Z&5sbJ&yoM5=lsl4O(WWYyQB$`ktnxXq>=~ihY=YWjTA7%b`
zls+yDAZSlPZU9ueQiNY1${{!Ti_kVUNnOmrG`b^TX!V)b-v~kA+5gdxOMQ#y1q$Tu
zUD2lIl=xyJ!-|B@Lh9T^CLkrnrbUBwAsc^QEgJG^Ox3P_U1Y(QJV`VD>7&bMT?U#I
z*m5xR>=b-!FuiWlsr9{ET(xlB`}cWyMaczJb`VuGge&U7(Rl4rKw8{uN~6lmfC|t^
z^t4s2|3D8m1-o)K`UyS+$P8q8IidU%voCQbRu$fs{YFcd&8lG`FX{cFoETr{H`Ehd
znoS-=k;B_{qi%Fk=NqX?(^i%p8Ln0j3~_W8^v+$8U&YODbTmd9rKqh-?KfDUa-|fD
zpp!YSsSP}0Gf3GKwQm?UX>Kfwjn})ltfydzv$?sGv}+<ZPzNETFdXOexaeDY_kHaD
zJEE{*cLp!iNoSLKC=k6UM&_n{^Vj*3LK9n4wHy%iQlg*{=O8+3QjJiK#t)?jklTM_
zZsPi|81&HlvtV~st@U|F5CkpV%i<ua`d|cTqUqk83EhyjJpSBJD}r`c*O6mi;-DH#
zMEw(^F4)vzpkctWacz3Bzp7Nf_H1E}HJ8E=Lcv-Q_>s}Xl2uq-w`~dH7Tu+*y>8%;
zvWoUwR&9D>f&Qv!zVceUM+wsqu#!(cD_?<dkRRqG=L}N+kZwWb>$c;-jkQ~<R!cXn
z8d$n%Z@aBkf9|2j_Q=N|HBP)~hdC;Zwt9@bIs7%*!gQ=tkmBH0dH?(8Mq5uS8OLCf
zPQ#c)lBozylgj*Z`^j)m`UC`zdKD?u_sJjK5#jZoNX4!tx9`lc+i^uBz_(Q7Di^g$
zw(9bv@C$}0qQ73)>s-_2DT$&bHB%^Trt?chmOq~)giJ+No4>Cu6N5zr-0@ekCDn4z
zo;H;&!QZRr;yc@@mQUa+x^U<F7EvSfeNVy_5p|TS#K5eymY*yyuqsd>j`!Mm;@G&?
z4;z{2OX?7g9Bmgrv39FE14?U6s8Tu2oV9-p4#6$e{vdnL>F^Kny;?@-Du4_g5*AqX
z=!PaX^1g)45rO}$m7X)&Tat|FUNk3r7b96cc>t!0=@XV+Z2-yC+%iq{J*@7;dOfL!
zx7BoVW1GZ_mtme!-S5KVL0rgyc)HJm@GAACkG|~;PduUj7<1366ELwsj$Me*TC#io
zlNjxOY4D6QzdaM)!$Mrx<MY=v@JUio@`PPNu+J>u#@18F0)+q_zk3x|a10;6C5y#c
zI)$;6-$QL2k`D2SPvbYt0}@zUOb`E!k6v5i1=$*(Z7;*+jPxmkLv5t8LF8}WjH=R<
z+wGp)fUG8dbe}N~c8yn4;Jj$@KBj=b_jz6A!nm`{Md?1_dab;s^pSgg33`TzIy;c1
zs1C1IO9bua#(fIy*lvefdNh<RB%cuLg>jg1*e#&4yt1I0Ycb|C=pD+|`91nSe*@0L
z&HjJWP3*0-Gx(&Psd?WJF;ag8U{j}tM-ZTcnU1GsMG)WtX0m^sLHXg&1v>-W9%~gn
z052Y(`^}Z&GV2zKz}>Ldg9N$h4U&PY`h*~$0b*-4?lP6DdVb=Fo_HlgT2Qgar#)kl
zE)W~mGZi_@fKV)jDhYY)QSR;g--XQ*Ao5BEJhZh+XOc>mcMiiX66;Ud>%X(rStec&
z!%Cz5RZlAb@bm9}J1QoOc9N`j-*ON#*aN3j{;NXN*s$RNyfwbQAJkTFY^1q4BbR{}
zyFE%_r2T%@8(9qgQ(nzs9z<mcn$BkvT<}D_Pb2mNHyafr+Q@9rqkjp8m7eTSuIGx9
ztX4NhZAH0KBRA1CSroS?Z56Lp8Y;QCC}N3G&W9f0{9F0Ejj?xJk){2X(r+Z^ULaCz
zbz(>eGG&}F@SxOYE-I}~Sh{3+JFdN+UCQl1$@+&gRkJg=ml&({P7BzZBRty@ehl#>
zXG30)e1{}&0TWbD1h>xcjMk8yNO#YTavTq}k$v;_sgsB;?exoyJG`oz!i%5K!wPL3
zoA19=wkU!sq&tO*8*2_TUewIMSx@NHizoteqT4617K8Cp6&S}s2Qwx%yDhUq<2Z8j
zti{K8k^Syerf33Y?3n}m49%L@%8N$WGI}*PGE`=~dzB-bRQG5CVxYaLnktclP<zHy
zxQ4ldky6%7X1hSA;P3D%v>IA*fr(?BVDr(ztLW{pL7JWV^JHzuYdHU_j#;7GXC^tY
zWd1)A&=m`+-aK8ljCahR$VZGb#h;1DIv05Z%F^{`#+dj%l<cFVTUvCDWNDgLTjH$F
z*k}g}Fn!d#M>2~0Kt%F;8OsmjK5@q4BMK8FvGl13xP{C4syy=DRUJWjhl{IzZyO*q
z)W<HzOrx%0;5ba<yE`ajXJ~cG=GHzYIXm_2gC%)T8lpsdJkH=gOscuFdxVvHCu=30
z)lG6<7b}<&rEk7LrQ-$BieDS|SmG;@qmVv<_U)oBXn`mQFh%X&&+Hc_twzE=iC$&I
zB`;2(@xka<6vM&SWiRF-1V?w=kq7odl~1=WG!SyWJZ{Ox<*!6vB0u)4J`U%M<@1kA
z{9p=XNWVFSjlP7-W0}?*h9X`Lheqf?L3wB$w@|9l&m`H*80PbfOE|caxO%DZlEV9y
zkA^<`Yo2@=;JA~#$(SgExi;XhVS~eNJQc3yu-}Z{KMHf7)vdEGV&#a-jelz~lW^Nh
zQfEi>=qLQoveyN(3T66X`T?~z+nMz8TU=rOgOqlSYh5&^1+SM^E>2fzTYTbLy{Ly<
zj#Yip0+^miQ^%P5zp6S&*$l!{1=c=^yE=`v1F^kNf$Ogrfu<#SZo!){PnakYwL#=>
zMBZs>5~F4egeVGVUe|M)kcuC@MbRsM`mH~;CdrBb;E6O;Z(rf{1Ihu6$N^+0p)ay?
z(lZJR=G_AOvh$`rjxRLFnt-q`Zbk3)jA8}qGYdp9C;7&mwAFW(lzS#_c9UnBx_HfU
z^?hFl=pVn=|4VZgK|m8_-^IXwP~xT#6BX<)YIt7e@5j>}31o~!G#N$i?{)o}c`rz(
z&x_$X^B2|(Xi<z0yuw%|(yXt}!dV{hYi>P_u-jsthTZ3@m3L&(oQ~_4p3-Fe%_O<>
zqu#wF45T<Wc<>NCJ&7%6VLZ(^f4Gs=o<@!ySby{3QDB>)zsuJ7;Z1K-_n}+iYVnxs
zFLff9>`&a<pZvUfbQto)3CF_CAj){ZvNeTZw%dicF2a=VVLg)Jy%iBrtg=~wiC)pI
zVD>|<#&-U8jWVm^4n6VhWPkRtY7AzA&$61^O!#p#Q3JF=nF{g4y;p>CxKT%q4gxs`
z2ozPC2+U1>sD?AUA`aYe?4q$J3}+^AaE>ExLSd3`fYDR%_6nvZPPHn+=APOnV~aV&
z-*o;_SP@V3z}VYK7CazN*s2-ziiPuVF$x2y{Mmg%#NEXU0`;RZ=dUs`oi(F4*+(DG
zF(B2G5c>wZ7bz+ouDc;1!!_}jOy}twxEo>Darz|_vQca@dg%S<1GR=ks_b8*vp9KT
zimym4D_4Pvs#eVr(fUtXBY<0-tJbk6!`af+ho}Kaqq~-w3fRBv^+f59aK_r<fC#hm
z95UPJX874-p^0}(hhK^Q8r}v*=t*Jlp||~41rwi;x~+NRRTIjaDRLG+wiIt2xK0#p
z!!c~%9pc*DBJG-QLL1OKAzUIQYQLScE;w`UoRX#Ah9T}JxJ6x-ykj^dNyne;uqsl+
z@`7;zvBTi=A$J2wNk>-Vo<z0VWA?SE7y93xr7atMkX%3_H_`p42!xSGz1C9E2J70h
zaV65Q?qpGqq$TTp{?K`TqwNK3;BCdkA)@h7jnPo;kX+1lZDo3A3n%LFQ4x2<6?#z{
zwoLW<<qxBT{C${ADYeoPU}k=0v(cLm^}>m9BZuxl9Px;K@5$v7ZSk?B`!&8?O9-ax
zL>*VLb}S}Q-Cg^T|FS&M>ogvOgTon?IM{kp{_C$*&9m*Uo4yEcuAH=%>Tfx_3eoEG
zr$gR_ANspGbE3gaQQHpXSQ>IQAN~qe4;|}L(`fl8wK+?j+^Z91wXXNe<><v{UKNe8
zP2c`XO0nCx#Yo~D_}v`}K}~3^On-tz4KrA$8Yd87rmDI!qo?-c^HQhQClI875g@0^
zBoep)IcUiu*87+d6>eNRO!pb8RQC!PmF6~)j-b$@ke08@kAG%$qqqJg{k!8dtjGTY
zD$=sa-KZvCQ=NcrJVQ0)VPxQDpw?|hqfvWr?LGb$!(E4(yWfbA%Wz2aFG~wlit)%U
z-B*8DkJYKkXex!i;aAJFw?jNMD#qrVOPi4d@KXfx=yiEM@%X@f`!A*s*Y%;@cC_r?
zfe~3jiy$#k)yUo|J?T>4k@Ah?cpFx-@C4+?H+AX4ZwlZJjpKUk>LYCo>e@Rq`0@x_
z;0|S`dSN&bP=Wa8ytgbja@Y@*xW=EJNZxc-C^{1CY53tLFE-MQ#YC*+`Mp}Q0k2)Y
z0G9K_`CmCGPZy(J>R!_yPQswamrEjOA#a{0*S<m$N^q90KWM|yb3g1)O{U?ZpM|%*
zI3_<xi%#~ZFZ(v=Y=oSLZRpEXAk;xeK=9BfPyBsNP)0%h-fv+eC3TRwvMe@fj5oqi
z-__uy#TKxcEM#b_$;V|ly*%Z+9zKS8ff8R*Uy*25mDPLe#qoHPQzuhF3SyiI%-u)T
zMYPRV^gG=OC5&hWFOM=?A5=6t@^h`yhQTD9QaREXirpFMEH*G(=lY?e&3j}6nW9hS
z&ojT;;)%;FV?jB=nFDYkJF3o%#1(bjmd(?sQ?-}so=q?dilL`9yY8!`;9uLV0C73&
zWaf?sr!{5nOHyiFZ?}xO=Lc7d*^jM9k-y;+=f?dE_tqmy2I+suN9>xK>)l&_B6>51
zUESuFY{Ozv5(mQHWV08aYejV6MCB?JUQmyy{eCOd#4jS0$a+onwj4lL(8h-T^FVwP
z(^*h-Lmd81M0T(c!hm2wm`S7{4j6v6Q0py%jwA{IM>Uky>iy~QE&bot9yV(i2FrR!
z_<FlDPP&XS6#vgGdnD!^jv-4T2OKnlVfZ?tx=9c8`=5IR%OmThDfGNPaL_$;!!6BJ
z7^Q(Cd7hZxbbNAa70^0Mzlo;VF@#~G+V#TFwPaZZ*XoL3uzjywGsmvUQUk);re<b(
zvv3(Yv_HzL8(gH)z8VD6XIQgbx9ppLz`zk!iJpO-RylNBdd>V^=#QV;C{kJOvcw|Q
zl{J``Nii%|dWvu`f?M>zVERbtRi8SSW!o{gEJ<hPbjh`KO>Bi&p}&%ax1>75KGc{$
zg5;pa+39N?pPcNbgCFnmfd}e8frVxJMTxBw5(z2PL2{!vf8)G5JCIxQx=FcmSMr|i
zPlo&wf*D4FEsp#e<>NnNS>EWAjWQqmzVmgzrfM@UG8lTlS$m~e10dm8U!pVLMob9Z
z2MwbHmIP=7OUn|1Q@R<AWJ=Az9|o2(PF6ZKurXfe46B1Q0?$ipHSxMtW-@9==>j2N
zRo{_LE+;lDRKVY%DN^bh{&O3k|8EsqK8?Wc_kUuIsWEfV)T#em35x+2&ozI)kFMR5
zhrn&)e;-d>&pvDy0g{xE?!A6HPqQjuiJGo>7>h_9z{hcqo&UPdF@uZwJL>zp>r~PF
z6x_a@|MNuW&dKDgYlCacD_7Qwp9kUg`|`Ui_84wUH1uFX)<p`!og%ZvcUHbM<pq}5
zOIP$zj}pj5_jOvw!zEzJHXsc$2xnEdT;Qh)0tT0v9*fin`q0zpMa<!VSV%xI40q%}
z?OVM(Q&9A`V`?zsg(FK}cx0|m#09P=7Qezp9SDJgJqzZwBUpcAq6XLV**?q7*6Y#!
z4K>@7*}KQ!la5ot0jIz10x*{*s%KsD@)*oyc`C|DCRv?=)Te-z4|veN-M_IJv2ApZ
zMd}bXqM*{rH?^(!=-^*$g%V`Yq^va$Z>H@zd9r08RcQL&k*=^O%sf)I;9aEenNvU0
zmCHsgC&0OiASBg<b3wap3~<t{q0Pa_2paV{;i8kG`M87hZ`ECg#b!a$SU32nm?MA2
zs?$hDFim7>Y?TAijeV?9Yhya1%)j<vqNUh<X<nAcPtvHlRO+=Lju$FRjXIO)Y0BbA
ztu$pqPI1gk&M-qYVoc-~5To@?^${;I3pHsBMav|4mT24J-;wq*;TgN{`m2P9?CH`F
zvN9zf=a>j77`CZg>b|9Gj(X#YO;{bKEFa_I>}A3BMU4RgFR^9>W(7$)DMNh(Yvo(c
zQZ#32%EjKoNnbscf{8@eUmChoNOcIM6?k?iY(YrRi!VA|Tl3NqSc`~W5Es6TkZJ?a
zE3CV&S9f&n>qF4QIluCzvsfk+K~1R(N4P#0XRp-I`M1T@65p||w&N@Yo_|-&1|Ixg
z5P{^{B`O9G6Bi8h$21;k_GvNQtndwKri+{=c`ug{U;V)BeVn{L>T<d788fnWXEf3B
zh`?|yjn!}L@<Jnb>v2W(Sd_PlqbtQR*zJY`#KZ6oeOqD~f7t^G%cV~AC_|*@(Rd;l
zpv+p7e9pTicTHZXB;?QfQ;{Gz->J8h9YKi{6}o{@?2hh@FM8Odj+$chZ)Iw#5eCI%
z^n0DAtZe>@V@`3X5S*k*`4w$l^+$(whvNcv+kGx1c^!T8#ti3qxLXr9+S?=XOMWo}
zf&@i&o=LtR$K`R(Py#FP7m}C|aG;B#-(@gQ>ARY$M8|uMc(Zx?f6B`jK3P(?Ch{R&
zgvA5E-<s`^0df6z)>MWN0}gT4IX0*51t{+kmg;+nNUeLqac-D>>Tb)_qF)>(2DJq(
zQ**B`!<D#Gkll}PAy)I2%w+N^xaM;PI&RkeK9wwUY5rkua+$KJg~RvP&s%kUf=4ps
z{io)HI(|Fz@r|uHrZXmXf|8=?oI5r*mwv#!1QZpff$7p8@n-;vh9NOK8m@cVk_5By
z_V#SxK6`>2zM#%pcU(9SL*mBSZMe`EB*B})0yf_phdW9qE~<-rsam!^7(qZ%BML)L
zQj6>nOLe?$xLjWqd4KrdFUH7TPW=H$jv+a0iL=+Fo8*W=1vT3ESOo6nHtG$FUteXp
zjia8EWzDH814jXb^?Bi_ob1zJ-g4P{9S#Zx=8m`I`lIFY7g0N|2XkDV%VzF)`dR6v
zA7FDL@SMA{L0@^{F0=LK4j<fFV+<)fD>WV{qo~T-R4h|82Ji!ee4I?hK0C0reHz=I
z%9dq?xVZXf`=kssZyJi2Z>rKV`&dA!s;VCvJymLSMgt<1%pk-p{T!AYj|$mAMQ5s0
z=Ff?yn9UG#6Mv?@{EWpDSd#86=Up8VLd;EerrQPER=&T*bfFSK-{PWAVa$JlssODb
zP~Gr!QRo#no`ze?(ePeL_ICPX`-#@Y<M(9MZ4c>e2!8E1U1>T&<NJSAVM>70Jf`3X
zZ$eimbY@tp^rcffA){UYzBY1>nj0UN2Iwc**UwY2oWO(#mk>_nh9J__)Ec+tFKeNb
zSg5^7Qou3#d1&{#GqL%3Xn(w4=hB*6G`-X_){ju_m#u^l?a*Uu(7L)e<n`Sxhu(sC
z1XmotL{v(}ok_>4;>vKuffYDC<MdBVp}A}6t@Q<VK=&~U3V9=OU4L%H>9Y%mGs}bP
zKSK3f2Xs%qhlpW#b&&K8S8{b~WW0FqG_9qMT#8C-gbRYx{6zoB`Kw@8hR!QK#_OI?
z#~r1O`gVg><`O7@!nd2^)2H52NO@GB(}6FhP)as1g4Wv7rQtS{tN^rnHFYK9SinN9
zUC%REIXYJN#5uerTNo_rW8m6hrvsCxaQtovE&lf=%0XtYCEfJFnD^_ihOy1{RW&%+
zWN9^y@We`!zDDb1O$aXc8O~6BfnX4ysXMA%QW-b@J1JZWQ+&hu>CUMMxy|M&TiElA
z71qJ&l$EYC@KgXa<mW;gB1y=PnE^S?gwxm58+>uhl@SMFY}CE#B-fFFAMEX8<e@MX
zv93@e{1Y3TFnb?zZ-7ST3u^(+;eKcPv;UvhXRiyFtowSsk&vj0ZdmdxbLQvbY0cB(
z4*bb&7L`bj?63R~$bH-k%n=P_bh>V=2-Hef&L~TiOg>3jNArlrj1Qdi-_Hiz&9?ps
z^p?{Qz9-V_@mFBjy7}>cpK%N@malR|7#~LBi56w5w;y34C{ZhsKjAnugiqccL~39h
z;TKA>k`X->J&bT10GSNp{``1*4jVaS+N@|ISu0r5T;&Y3!+g?@Siz}F<Qt@WLf|oQ
zyGSap>MT^#sWCww*#MiAv8`uj%S5SB<k(Uxq8r@by`1O@f$@7j81KXSkY_EJu8?B~
zbH5TEaly}|7vtJ$cRW(ckEqGA$+Bs2xc%;X{E+9u_%ojlAn^;&UJf^sD1%9Bywo)s
zQCwfM(@rwk#q%3T$jUXT^qY(g4-kNe>N)ZaSse;w9~Xt9p^(QLgp!3I3X>D`6${rf
ze808A_CJS7Iih$=pQ`h?=vAme9X&#6Z#--1JXgTXyE~endN8%9SBbPK#>t0~!ZW#A
zW#qi45hU&fL~iZyJ23^gQ#zD(<X@oLDbyaEnX{N%Y^NqBH7pYzFH6Vw2{pLm?xb#A
z6UA^*wuRLM!%SwfJRDFbKn%d&)c<iSL!g+AU`ApFqB+q-bF!BCd{{4qkmz5dmTflv
zThRxx_Q^0Br~$+c1Qpc()Rq`=)M4uO_~P|y`#jhI2r6s_@fImvYmei1zt>-(!FDg=
z#9AbvBy&e}#}Dm3LfqnREjk)#(7N8sf6#y9UoUnF`Gy-=nk~4)FkevGg{UkSk-xnw
zUHePn?@_J!v&AAlH`(vB`B$UBMum6Rbw;mkfAh$udlZ=^S}BUl{l&e_{kv&kB_==1
zi2UsgIE(3I<ThGd1BukG33C7tfbBOID|S^(3fX;*Kj0{`;99daz7ev+E-)Ih6S(0Z
z3+))e14@x$iK@FAg-zvD<W#>iiY2^T55)T8FhiyF{UI*jmIDg&2LFwE*t;K0{?nI4
zoL6Vah5t53a;;;-)G$RpZySp0w{|WhhTF#t=%q;ZcC|@?X+#GsyAW^sd;>+lgZtJ4
zJFopUkuT+-%Q<+EI?CWtD_?-N95`{Tof9felRa{saJR&LSiYGg>7uG2<HF?j)Z|md
zgBPp?CsBsQFt3N2j@C?{T8M9#Kk(p9cy;lxZt=7=)+L0a-={Pt*s6ng$Dp_St%!62
zAlFX_9M6scGWzcj%lgs4C(_?8Jb2&m9sEZIVjVutqWlSo#lm~T4GFIy2chyk!Z1~)
z2yjcl$yU}Z-C0T7DbmUqxm4C!X(Lo0*63d%ljS<YXDA|y@M7?*($w%0oHDvDC_303
zV4A)wq6;=8Y-y>L%857<-MeS=POIbsURT^X(|yFqWp!W&4!JiNn&>vuBc75|X7_Dj
zR`Qy1Qdp=i4<{TgM^3sv^HXHxab_#oO4GNL@2)piCVn2*`ocX0@Epi){lSiVr9YUk
zSv?asS$6XT!6k=9E~k7Y!;14+5>u|NE~P<9Acq_c@=QpurWP(Cg{|aY&BGFbd}~x_
z`}%$#eLMCGPm8}?Ddx76nbkwhp<4$@-z|`C@)BVz9DdvzEq#*x<<?4^im4>;`=h|%
zW;RJ!woFm4E7$lB8_{ItDfjb+R1<YmBWXWICni^LN&}#2+7SE0-9)|(Ytjn-C@msf
zk9Zkx<?NZu@77{QN%c%y2o=!*6vl>U@M(a+U+DeH?nBFBp5THx7ASqk2s$caFwA)%
ztK5c4<U3zi0AueI?r+vszVy?Fe@-FgsQzAAOJI4a21<uVu3*R$0F@T+lhz%}{>NQ#
zT-J$T+abDEoG}%w9VwvO;|D%+k8e)O0NruABw>JJU;4%}=q~|k&M~n%puXn}Md<k9
z3o^1o7l{xiD-plirmoT?HE-<deCFng{cV2*t>zy~{QWZ4^cG#7c@XD0cqWUu?TlTE
z#t6sqa3y*?9eUaw<f(}QPyb)WGj(32J8JAMDd_yIQJ*GC(VlpYGVFG_Sq|zP@0tee
zy7Z{)F}4JAuix~#1e~bRfR#`%UHY$Y3z{3pCbR;z>tb*@r2|#PaYr-~A)%F3-<v&(
zKv0lrUgj82NebOMwGYsQc+ToPoA+G|&bz@ibdwR>)oHMUq^Eb*PM1KvYe$56-lIiV
zGw1Fu+;$Hxnrh9FWo9D4>Z8nD2Eh|L?U+<7nptPR=PRbqTEG3v2Tn<GWs{M|ZS<)>
z4g2IqeW67P9%W#nQGv4w_{fWFUZ^M?rs@txGsjjdp3OGrd$(Z)$-JcF0nP0a@#}B)
zTPa?o>)o{DilVMjl}Iuzv9+w8bmio=PNd0AJ~vK|X?Y(V%&=@JmUcDW<Qv<iV4Ac=
znKkkQO{7q^M-=lIfbiug2F*XW1U=HO#J!H(pu%7rKc^z4(;}J$j{^)f`$vo}mRnz!
zh91?;1IrbsgQ}5Y{dJ<UQ3#S?jB~LRRaCk4M}2p@D??B&q9ql>J&a@dAg$V71HG7b
zfti(`9%UmbE{^B|=Qn7A#Mr+#%*XV61rN(9idPd<Ilm<r0ZFTENl6>Qf4JMUQIh?n
zBVdE9KDo;(h6!v?UA`W8pJXuS1vtK=M<ZBE+0jVu7D0H{Cu}8Lh8-^U4<}B0<mc_E
zP>qCwbKGe%&vRU(V6dxw+$7Yz4&J;M%D0CaEE0B+F1J7wdK8or-nNh2?e>zfbX7x|
zTrIXJu66}RAbDnq{}|F@K?LW~iBb#}=halq=pd}`T9elBz~j)<`C4lOWKHldq?W32
z?A&N_5=FG3Pkm*#9Aufevj6uZTM|U$jQ6$A$BEa^2fTm3-nAe(FZVYys)Oc2z=7Oe
zyfJ)3uKLX7Hb;}kAZeahVIoRYzp+TVf#Ro}N`;GefbP#o(9mTtexjw!$J&8b@1%nS
zZ}R0le>CJ{0JGomJ`u^!Kc{wA`*)vYV<{i0YAD_nCIOW@yU|zH8>??A@ct}*S{>?l
zbJ+TkVKU;xsRdau)wK;mwcTW+AXZd&d9^7`)cSp{8`-V?I=qUh7Ock-Oma5<y56DR
z_3f_YI1y23E;@nlx}!11AhgStVe;8H({%b}5}V)`%>Ek6qu-wt)Cy{PR;${@DuM2X
zr#IrT)nf1eKZ;RxhC>HqVNJc!f}~8Hk0UGr_CL<<W<8wG1?Q;Da&vR8K6*BZ10?b>
zk$U}lqgGXd=L`_BqU%7zfDfFWt#{T|=4SfPIE#OecLg#`X0{7BB1iSpOoj*7e16Yt
zm@ECrpCtu5C(0yXPrJ3dTGoI)7xXqbC=)|-3rLQ)c@cgUqUdNXj^~~0CstRJE3FW)
z-P^1xfdK0zHikDaI^-fi6jDywg`fj{!rP%uwgG<ebfG@!BtyP*<J6_N`o~n%`!Y#$
zpp4sDX|PGHG$9*9538&MY5%<>Wz=6h(*!5(hOs^i&a`*e|1t|dN<RB-X!ts!b~v^K
zX@Mh8sp4Dt+HYI_Bl}{?qZ}Q3!I(S1N;EVRV2Okk;p60Uqj6;VY92j*^0debn1Oz3
zV+)utQo5&FBW{-===EN^pP4WV^iYBcM(yX>f7}`ZWu(8|S8^Dk^=ykQrk%tS>KX^9
z^>~$F#wCP6yu_F_jH4C@UHY)@B#~e?uHDLz7Nt_|1dmTrOdhfXW79EfIvfO|<{~JF
zj^?OMm-$P;G<D%gPEUiATfnk~!Z33suat)DSA*$PeCf{&Bm;PZTAzoUI?--#tyssO
zpGuQ`RV_x98GOVW58#OMlazJcZeb*HQC146Zj2y~oH8<Q$H^5WtjT}R^P_&8N?Q|s
zS^NU$#ED`XD4gjkI8Elb&lK=SZ4)~JEoZD{YC!^Pmr^UDXdy5I@w+35Obm#-Gf6ol
zuH$VUe0fidAo`>@AhTmCc`}A>Za(I@KPb!4IX#AP;%Ix#PS?fGFRv0zRxa;RdmmxG
zS^XMvQ?=Ro6ap)&y#%j$Qf#;ZtqL;MKycRK&6<-^O@ut>aOtcPA`Y~wHQ#_r`7uVM
z6lNU$4k}?h_hepr$uB10pvq#EN_Bp<jmFlQs1{aHKzzPEz1}#L_ux5(U8ZaO4dO=E
zDiISy*NR?XFOY%Ag!3|&4LhLY6L9$gH}fEZ*(bWP;MXV0)aAM=hvOEHlxT0H$rI(8
z_GD~-OCh1&A=AhMZx0kKF676hT2k@9$We)9gsTg~X(Qh+qfNkFiIGw{S5#@c7`iV{
zN=GfaQ2busM}HIH5Y>k6=BU*#`Nd<(N1GqZTj{>B6Vv3pGD~Zl{cm&7RN$Pv@=Oqc
zBEBxbhB9kov^5r=cZKcy2^=^aYz6O~>xqT*aWl!*8(c0ap&!85bdZnIXnNB=r=p>{
zmrAhB8>F+eHpWcl`OmvZ8>{g<4pcT8YuY_bd5Tzt2vRe0EeL*t`iUGf7KEi9=*}}M
zV<Uz$Uqq9<3OyFgyeMG<iv?+bwnGX9VtQ0?mUR)xk$top6N6w>90@IjZmc;)=pjdo
zs-@3VjZ2_nM?HWXth$5wWDRvex0)fP3xc98eNUlw5XzdLAzw@LyASsLb0s%({iU%{
zibCdM!lAhAL0UVGd3>S~SxSbEUDVC&9@BT4?7NgQbV|c_l%@Wn>4Z#_t|6>^KZCSY
z_0~3tGJin1lLcOJPp9tD2`vsR*&epBry|d936h=ArW+8n<Yk!4NcpwdkuGiL-{I2j
zDGTbqVP4?Z)1KQ>>aZohWd9jth`X}n_coCbmiR!K5cEA4UD3ldj^{c{S)uVK9_*M!
zw%^~)vzr(4nKcE^l=VHDX)Fd$xZ8~UxKbPxs4dJy(~Cj6z7lN|_wTb-Ndulh{a@_}
zrAK`aSq-3=q{T;qi((2Z$yd+&OgcfnlB5bw%V%tDhM&)qHZ|;7`3x&O3J*UuoC;#a
zyAP~Z=;f;rY`hg?R<|LgE3DkWeHEPN`j^Miro0k==iY!k)Qd%L7Cu>V<s9Kmu|!nA
z(RE_4O!%MKlAW2;|5nU3@_wx-<sm&Tp`R%S_Q+i>{g78o5oTys!?)P28QW7kVnDcM
zXVc!S9s`3McvkNGd?)j-@f?_*7X?qjGR9cT&+tTb7`i%#$>&IQs-p{bJG8oTk4Ou%
z1-ifAz|Y|c<d~|Lj$(6gp5CB;CF-Bu5~lppAj*YCa*yva-HK_*lQsX5n<4Rd%ViV+
z*9RjgrgS=$n16z{FyQY6!xq(9La+u8bP-3bwhuBr&BF#ljPVVVRUE=fLHO!iFTgt!
zmY&U*HZ7t$e)2qQR>*)-3RcHNniIf3#M4K=u1TpL31rO*XO>l8HD<R&zs2WX^u4(R
zHnLhZ+O7IQfx&Mx3WE2;E1!zzo^*f#_QU|F)~Iu7WewC7gti}&FH?G#P)fTY#0>1&
zfmi`i7N+c8<&JX4P58p7?oQEtg%%{wm|JDgm*rOmp^MrIxH8Z^3MN^`C|k}*Mffcu
zU1(!4j}$$;G!%;DcL@;l8IWRwmW675U><&JsillMPcT*)XY`7xK3_P`KZ^xwwIv@(
zv2XG02A_B4T}3tXonn?^)j$CwnGI;b`DZY`YwcT6-SaXPj|6=x$25Js01ZJdg1G}h
z#I(Y!3N2gm4<+js&;CU6s4Kx7@h=WU@s9PJV2N85e^gG*q8|)c@adasFxj0Q?W5Bp
z8Mn(T6Bl()azmZqb6WNG(kVw&dKb3j8NLsb+Ce1cnNRlLF*;jyf(wxbJfBz%N9+t|
zMV#<e-$s$UV}mYIpQ}3|ch4B6|FRaz#?)V9yv$sw@jTd60D(-x-<F&0+A?H>=%Lu3
z5s_szml-fy^s@^2iMG!sgjG?!SMn$z^}mLjH}}D9Nc+XVJwlTAzTFKl?HCyJ_P!mE
zE)TPo_D4=W?l1*lbNhphCQ2I8gRr47?tHer-6p#jUwaOVIWbZYVF7$^E2~Us1hE15
zM2_z!m_s+Jz#q`3SH;i8Hpf<?nn=9>lJ`}Z^nDRB21glYR5;6*o{yLJQf#7m=(n)F
z*Z0@SUc(JYoQw1u$LEVU?2bHg=93`_N6OHSx97OYD<)tX1<h=R;C*1$0aJREJCGf!
zmm%=&K+FG`&)^LfTJDPR6ISdqZ#T>9g6Xs44RQAlNce;F>(AGO9$iqdd4?XuWqOeu
zc>n$W3nT&8z0!yKyGa`f>o+TQR^drc5+05OSu#_M9IX^dG`ge+t$5501TxDa(V$8n
z{<tHfkZCfX%OTab3$iI!kE|%|5?}sr<akuRLx>|j1NR}rd@=1%e~AXIL@7xETfzMI
zzY|EA19ks8#m?-~80j;GrD$k8wPnIma!6~S*V@R55}eL*1~I2PA>84bV@Oh%qdJTr
zTvmj@Gyes11?VB_81qJ7PdTMiJWJYsyKf!3)-%gMoQl+)A518tg((bdmih?xjPL4Q
zRwWRplYKy6=Ncpw(IZk(=9>KAAZ3mGWLu9b0@5iP_xh?(7_8<V7QPd*&BgL%LXB@~
zB%)IKQ(N`M+ARK>vCNr4pc|*7a=l`w6>dW&i+I?ri1O$#Q(2JCK1Wu%>kvmMn6^>f
z<L~W>G{XBWycO7}CEW&@MMSnvJ81$V%bd}#_`_Bqi>5|yi&x*8z|%}2p+t&EufKX#
z6R4t=8DbD?f`m%vB>&#8k%16@gegTU-=N_$%Wvj@y7DM(&b-TiZR6Pa647Is*sK=r
zLq8qOtHQVSGw>I}9IJH{Xpl7M0Nv_hVX+a`K`W7hqG9>iKD`KSO<`U(2({HwK_K=<
ze7Q*^ea_PDE!BF@<@xN+;YCzq|7MG4HSo`uJ<N<vKMpJK8p&o(#2^9YjQ2~_)Owd-
zuu#I1WW;I`YnwG-SYuK&;j+6qsbucaC!b=zlnQEF%7s3Xggl^*W4=LEP?D2LSspi>
zv4d=_!t$v6s-Kj4w<VO`tf-H?H~dFFk^CAG|4mslw-w}Y*@&57UU<SWfpGi-BG9&%
zJs+2BabwZ-mHAdkfw1`dSUMvXV6u|we)y*lz0*KQ`6{G=yY@`eUw6oBn$+O{Pg5po
zh<TiM{hLPvk}Mis4VFxu9xU644sINsF;)ARlsu~3mUgBE|9m|439FQ8(1KLXM_^Xs
zi~A3D=w2r4U~vcQDl0AU>)&^<1i;0V?yOdf!PRrfGxuz1_r*2u^qrl%kEDnC3rK6}
zxE6iYGrD^0M<q)g(qFx^t2qYauAPKai12V+$U}h8Ydd_7xkTmIn|s-FI}z-()v6Q<
z4;QRhO|47IaxH%jzGj+@L7ngOsGXC6xsY?~1wdI1!xN4-jQM?1GtEph419z2w_P2f
zFrPX!)z9=?`dvK%8&}VdhE~rlMk{ul<cvLTZcqqIP70w}W~bWLiX*cvG`>R{MWmjV
zG6esj$rBOyu_50`)V;AI`)>S!HOLQ1Xi&P6Ub9|e#}2RGn9^5@56KE8$%Go<4I0cv
zFUwc|HD9qeK9Nki-h8pz0;nd6$%t>lev&p<LQ=$J)}QTex7JhBn1a^)1P?mt3=2Sh
zDpJx&WAhYx@zsFPtzYBBa11)QIfAX7mlx)HZ+{Q$ZlEIwVec-UUe70s4g7&7OsX~B
zvXk+xI2R!%btQe8WCS~lvHD8}=9swj&&uAt5K|imjGuhJoz-%AFn}4gHxNto;RVG)
zJ^Ml~>OFehV!wQr@ROXVS%yDKQ`R(^m$XAj93=Z*))))wA$SNf$p{X-$UH^tR#jpk
zxXYHQ1x{h4W!R?VsHp6SToOuj{t`AuIL-3sm`c+K3uJ8=mU>Tu?H7|0my*LG4B!b<
zq#3&S?W<H^ZL?ba0DLy4%?L!c+F>`+=ue0%ZIl<J<MFCmiLc*9+PavoRs^s^`BVqU
zT-|i446vF-Z;49hFWech!8MC2%?T6UWJSzv^;JE09ecHdZFQ(`>9P_!OQ&jCL|(Sw
z9V(`y4~~WW#qRCl_@v|-1<~keKH(gNiTP%?aG3^mp3zNThQ+D-y{NfNRwt6Oqa1uC
z(r_d$FJL)6^*x7B6^Ij#Ih7Yd4C(9J#G<PiZDTW{j+Box&xkb(buCK~Xu236OM<QU
z0gu45Hok+t;J97n#u?^jY#25U!`Eetc+1|EyajPPd<X%5Y*TGQ3S8>*)VFdgNg54-
zr44BE_B(TyjNup4u+C@DYIn6XvtOk|p87>va|Et&B$r8I4ICz)lPEgajos?S%^<FV
zzx_&}VYav_3~LD@KE!zdD^!hkRUGIEYXS4oGh$-jt9ZUfHEr3O!g#=0puW2ErNm~B
z!c+WBdlD7H;`=)dBdTze6XfsVp?;#Ft=(-LUVU&W^_gz=@B-YLF5Y=7^0KQ%S~%XH
zAh$9osRuiE8GwYkg4d1$jJc=}!fs0z!aC#zoi5@VHe%?g6*sZEElS0$ymSnIZX|f)
zSq-K3{NNYf;Dkg?8(lS7^7tW>OQhJO)s$(lNzrl=!8{gmNu!g9Q%(hA+#PaGVua7E
z5fsi@H>KcD^W@ed=Z5K4sM}OD?3w?WV9WGSMaw!51JF3=u8*^!OfieKjdv>bkHbUi
z2GL%j#%;{jDi3uU!XlkP6@^1$y~VgQoi)J>R2=%1bKLZFl+>M&Fqi8m*S>f@=#?X{
zWkyk;a<8RRZi_-vGw4$+W3G87yEko3>}E;x!Sl^LOae`Zs(4C7YO8CBXWE|Ko<1}r
zYZ8LX8&JF|tUX|w#4jjoT7t(<D_2><pk2_tFje6c+_ghw7i57}ToTBx<&2iXL$@*S
zQ=m^XKp6qG5YMcYkC@aEJeCV}e<5paZ*P!%Kx$#<l;$<oNBhrr&ME#XZIy+oqnWG4
zNpISrY_A!`LK96bIwiv=NFdaUttyKP(y5eH1&|&N!Jtb|YY5nvpQ`EAd3pW7(OOhX
zB}5mw3dfNPrCT?JTolckCT2>%w`KrDnbvaqh{qul^>fP1c#3o+6l6+tT`TITlD@Gq
zYU#?#D8Qa{Sg8>Q-eC|IS@!VWk)svE6{&TfM#l|S!}7Q^EeW?1k0MvMH+83Wd%XYS
z0ETjhb35o3a*kmsE<|tYR7J>_Gu18JMSVveQ}O41Vs%!{POMp%P2%ds{(O)VVH8?&
zQIhVrRmJF5M>Mj{cW4a&)1lz~Hp~7u<a$bPb-LGzZB_cb@FeKh6zuEC$$*dVn<qkh
z6d@<6h#&q+$?eGd*j~FuiHCy)9)56O3E-$00TDWd3$zR@D`Rh#ZBG84Mv$r)lu^OY
zI)d9mTGU*OtvMifXieYpK+DXzYiJVah`5mz!{)|-J(MmCq{aUUo%JJ9@f4RcO(ued
z)FYdmxGs^W&(ezL0!EFcG#z;4Y*N*{9hp`+h)Y*Nni`b3=)t^-)mB+oa!~hV0N_bq
zvLuN-4eT8a!|$ViB<?1YAP!BRZRV^YJ(Q$BhS+O&CbYY+p|G(RCcf{)6Sv+Gjj1K~
z&3z$Crr1qQa(>SUb$^GXmlX2<UEExUNTYs{<DhY3&Ya_<W2Xe?Aqdw?l7YL=v$jnF
zLi>m}Kf|k47e!H$&CO`Pk-ov(3Y-pW{1c`A)v=jLHApsrZ>WzcA{)aJwR!iEdsNF`
zpj+G^we8u2IUw5w5i&}s<gbR^96pBz?;S3RN`t@x^Lt-zZt5B@_c|8|`?I&kM_}3i
zH?}1hV^UTuvkl$P<h=-^;$OM{ylG3*mG>EUZes+VDcm-<`P92NEBxoEx@rANef1Ia
zVdsZNCq`QTUB6_@(a=VuZH>rN@Q>YXBr`2=$*uyRAJkY>{{{A`pt0N-pe+BX*QNFs
z5wgOcOYi-BJItIP%_mR&C?doHETKz#`>YBzvq6n?g&+S_zxs~ySikVkNjWn@_#wQv
z{Hu0v1*0bfef$@=RUQ8;7F(J)Z^!2o0Q4=Z_YAs_u|hiKQh5B{;-T8EImhAV>W$nD
zbNqXD%U`p2E`XQX1AsW!{-a$7Nz$gjt$yoj$Nmk8CIB{y(DE?kvgZ?6_v?QCVdVyw
z3>g$@Q@w;h`zIuq{45r>mrk_U5+jnA!D=lstQLk=CODa0xP0-`Kex~!^P#*x=HKhp
zzF@J(S9CYFXi&p^@i0wfH7K~}+;k+o!~855i9o{Ie~*SMF{d3Pz}_7(56YP}yKqHj
z2|uh4@_L88{VwwAUt$VC+P!cQJlk<2UVU8p-_2Q^>w@iUN|+PW;2sh#lz;U?Whf7@
zFmAb{hYE-pY5sI;9KtG1ao0L2PTLEMri0=Z|MpqexlG_k?!~$;&W*ahD^uhhh65BG
z#mQnM5-;SCv4_kvD$Tb)q7u&T>p_4CjW_^A#PtisO5ZwLLl}X3cr1$n;X@R3;(g=o
zEZO>9WbmWlY?o>DdZ{Ige}r#3h`uBFf=w~Df0chBP`AxES_WXEr@-$068g)3RmHtL
zT|Sp<ld0w*&^eJMyW`7#1`xZ#A5t!^iwP!$7Zty0k0e?HHFx4Va(1qixgah^sqS3Y
zq3~h)1iMk$%zuD|uA4R_f&8ZOUtwpz@{|0O7)Q&u=GsZMvL?%ooR4+lA{E0KJuy_X
zVzAeD@ocdL6m69{9l?Gu8&jq_rnl^TO+A5UT@h2LB=DFfedpZ>$m)OHSGrO~_v(r$
zp}fNq>dK5Nl9$iw-?~p@4r{+Z1N<)4v2Zq2Yy>XtLEC{YK8AeCzmHPq(9TDo6-3fd
zB*nO`b}mr$Nw7<7S@f$>c@mg|;Bt#!q|)J8kQWjzQg)<}0e=K={$fWZC{N}S(=4H!
zuAds=Zt~n>W~*Ut$G_+0%Pkf2rr{@(P-QGBp;NjevcI1wFL<pFaf>uEvncA>Dnu{e
zv=cqBo<0GbUMjkj<%{87ERH*3y;p^nc=IBZMLNZ@A4-jZ4Foy@WeD0h97>zyGzf{O
z(y_jZ5uq6}(6!1Y42iu8RWEH64x<KGbPWkzgbGA5TgBrRd+`H*+3e~$o|n)8fjko0
zVb1kY0G=OB#tu^VK8;oThMINLXqZAxGHZY~f*r_~E%O#=6J^nmB5jEdSd63_xsV+*
z)T64x3i9QWJrlXl(X)U+q#HY-jQ*v_jOZ-jy<sqNHk#T$Esnba#T6w_*Gwc++uqTl
zqe!e@(r87Vp+8L?ics<38YaREx5<gX+SjzGRa3>WJEfhuVOa4EpXv`v>RFHsLFu~~
ztP$XYOFvs|oomrn6twLiMlqd^dPo~ue(GfQ2UTU5BilvhB#os-!uXV0F1&5cAfTfM
zYXYuOeTawCDQSQe{<3<{H9m3?W4b|Rr)qdaW(POEm1rL42+xj{=;gq_&3nAuUm3T~
zGbzz-F}L?9WPh@PM@F$JWzSBhcDi93V;dOIXZ<>?)TKpF4*rlR-F6v9R7F;cgJyW?
zd(#P{;=(bj4B3$6Xmy2>UrT1qwYf4##mjeEuMH+==gP#XZ;YKwRMFRp-slF{s0M26
zlL-f3w_)ZbnO8#vf;nuIX;vN&$~CLQR7GqE#f5sHvN=>95ZE;AHfS@+LdpYv25hH*
z7*3)6St1LSvDz@PQLR7@y8km!M!QK);flaksXMHTWYeU7f6Q7Yl2l@+3_dw$1Zumg
zjeVZj=Lml(XMjy_h*9-u6o$(Wcg;EVCk&Pvxol+h^)$am>?fmjD<w+7qCB+@f&OP{
z@qnQ72;I)QT&SNpuKj)0ciJisZy-#b^K_5XtRa$KV1L@A_KXLqi`Yz`d2*C>!0}?f
zUgy|v!Cb?&!rdoKkx~(TV{TlUGPK9NztUeGx5_oC)2?nINXB#ntK-_=>?|R90O5v3
zUhid4!>*oAaP9eN;1Y^@S^XYq7DXn^N*tevCc`k;Z>B@${+hK6Rw%~O6|muvHIi*_
z-K?4*HsO~>>-<Q6IuSotMebkm5gMe8R{O>I{{e|WcE79=oauHYxUeiB=k@&X1oUI}
zX=k|n<EQmVW;_IOhSSDF;Fy+BUeb~2@@J-NK|>am0j~bPZ57fyz|}xx0OnEkw*i<=
zqAfO^pPms1Y7{{sLzQy98nSg)fQyQMxrr;l8D_AmRPci4eEt}J$rBl;`?DTi!{WTY
zaVD(yVul?y&Jo(M!)j)<DrQI1I(Pj@GYIM^^O`Qbfl-Vw+J@wXCU_a`H?uV^LWLOC
zi$l+v@Eu$711*`Aum5r3Eg5LV`liHFyoxskxoj10)?(u--Wl<<9;eC+=QeqNP6=kM
zt;ut3r_Hrv&LAnGz-@!2MCQ=dWV)?74>ya7=Uc&x&y;L>Ay6(C>;|2?MTb=TKEo-D
zn!XLE1f>hJz69}tX9o4y<S<uEJ9*A~;WnEndlQ~FE7#4QdKKRCm3sxRY;?%vSl;w^
zzsQHMjB+bM+)duy7NXT}9rR~^dY|BMc{Y8UAblRja+aUMCu_sjehYqxmI7bIU_tG$
z7gIimL`LqyyCB0m4eg|76PCu_sc{sMW2mf0i%Ac<P@?<p;Xl~=>!&`N4&V|*8$hvy
z>j=rud**E@Bldkyo;>A7iJWmF*t}n5iGAkG{otY#6`6kT;G#YUKMY$-f{wcw<UK<z
z9j;G5XYky&_!hYdX%I@-#aISo=3=0sGhwc@4%&3xYTGL7@G{22feYuDB2l~;c%kqJ
zH3yi##x;G@#o|BfR%0xeVUipa0XLU%m@X8Th$=fMm!GX2qyf8^M6Mm7f56z;w~VQ+
z6EyL?d}AAE8rh)rKANl_WLp;fFk2n+y@-y0Ln@=9iwbHcS@FteUr^H{)!t|@`p_cF
zLK}xU!Lm_8vR4W=iuPGAd<k)6gIs)=Z$SrK+P3eBZ*Pnw=m*u0OLk2@wo|CIkkC0C
zb%A$vK&wU+nKao38$@Ane`4m?-@vq)tP*er{ka5-{4@}Tf0!GzcLMaP8)j`XF#SeD
zVC==Ty=@PUNcXc4W)lzC__a~j_iBe|RZc%oVy8x#vidp0VssauGrwfO1a{1K)ie_G
zlb787nw+TYh$k>HXlnbw=nBk93LCRLjyP*zZH)jYg-6%uQzuKPe<q-jv8(wr>BULs
z65}Q4OL>~A(^sOn;=mztRF!F_<BhHukln14W>Ke=K)`C>!4$6|Ro|XAE1ExP7gLe9
zG_Iu6@|13k@Y1%uuO$Ku3299G7L9a`77X@JQ~FB-(=@r1J9$h|V<?hTwmfsxP)aYe
zpBf(qb`OHQo{TJ~e`FCUYLM8dTNC6qC=&`R{iL^ovNTFYu6my9zqWXMx`e>(-4vA%
zn^IH655nsHDYc5+=>-FtGcjn#3ZJtzy$J%s>1*@r%lp%RCV@;|3C}B!zkhoQ4^V^9
zY<I;^J$1kSv-{=gZ>`kfx+Xe=ql=TnVaTyV95D;?894L~e*=>nzV_|W_8EGZ&5VSR
zo3jZ}cKx3SrqP`r`ZWzBGK{HFGtac`0!V=s3aD*;8&&(&LFTvZ`VerGOw`$wQtVgF
zYhgWZRgIG?Cc=qsk4{2a#&tYDW#EzL1(fU;eJZndS?bAr{+y-Y0?@8=k+l+~HN1=t
z4=W=L=iENCe;sP&60}opJmAVfjvO_7>v^@862}>yez}7YGzqOt%Lu^G)aGtRAOvA#
zzZ94ZC$T|ml5h`bH)<Jsj6c=(s^olygIBl4!ciR$0SNdEgY&FQ&^ckqPe__W5}>%G
zNGOuo-myE!*E3$0UI5Erkw5(`yMOS*@3}Frin{;Cf2QARH?sS8_haz+ieL`;D43YT
za7nQ2n01}cVtWy!ePoT@BDQIUo`lr2ES6#Fsg=8k>p;Sqg*%QKU3pQb903?Fz~ln2
zp{|zCCSQl6tbBUKM+hh;Ij;MfO?W+u<{oeD7O@SW$hoaUdjz0uBy1?aBJr}(x*GB?
z^9$t6f6BVabA2WPHOTUF5ixOdd}@cR-ZS+vmJVZ=geS2yYAdupMACRK*=ffsWoI8r
zb|_JJcs$pk1YwoBD|+^d*ru8GQdpzU;X3V+u{WW*Yu=U!^k(z#Yo@vI`*j=D9anwD
zB+d$J@0yr)i`WLwAeJnB5Y=ho{7%-KTxY6Fe`IBcmAi3*x+4wkNud&8ybJmjT2k%T
z)Jh$rLQYnNO{vTYRH(IPu=g{A?w2A(@wj}&T+MAalNOe?Y~Wyn^YTW+Osa6oq&(Mg
zdV7dUdHBL;_joQ<-U_69qI0i=je0pFaC!1L!lcA5uq@m&p;U0m&!MJK_DpY;PfM{k
ze^oyy0}Ma+`>-ttp_kikZ<xLqT?@*T_J>2^8U3I@%#u{ufrF-}en*kMq-K#CR9~Zx
zsW6|7+%^v!H~fl2E!$qJlN@TDaNzH8=+-@<1*iHk6iQ$FoO&tf+U8U~%U*1_o~Q+5
zk-2w#ncs%5^|Ljc9QNl5PDl#_Ym+^2e`4ucxjXP;0<FRfUfciykH*cYVy2ZK6^0x3
zR&MWu9>&zADrX1Fq`NQ`MNW2T%f@|!BW?PkbDe=*qw%$FmPh4B8@#zTStf8S+hv(C
zUEX{`hNSS;hFoFnXv-)Xn)lCt`7XbL^RI-@UB++x>APXUYCR^7V|O_wP>F!#e{;AK
z*%;WbI^QmzKuKKws-u9P(oS<NdafDh@IY1!q<Hb7?a8E}c{%HFWGE+16SbjezU@Pi
zeoL#yN=U~`l&6CGBvZKdad2}B<Yt6Nks-*_HjY~Q@?v*fk9O<otGnYmDmP1&ir2r2
zeq&TQw;}e(e4y1PDZ8LhPV`ope{o>Vjm$WFV=prv576bnmnoA!hZ#Y~FBqoGSij1-
zmM|K9{6YPylFB!lZU;*q9@Mh2z@@C`T$Yqdz+eloXUYF#mIWRJp!*$WSuOp$4684D
znEO&{KHHMAEzF%n&XD&;`Wbp;8~u#l*hxQkAb&ZcY)TeSp&t$#0nX1vf6}Qr+VL^1
z*vh0l^Bh|lFO<Gzs+^_OujQ6wD}KChQYO|0!0nVPh3!Jbe{`a-SPBYiY^?|#ciajf
zDCN`V0D_!QRxJn5xzqUZ@vYbmQyzAX-OvgRE0OejuCMLRrzf-YDvczOi%hZ+L5DrP
zf*_p|+j9`4jnIT2Tw6N@e=~h`>iVdfY+8Xs`uDK%H9}eU*BZe7LfAT%4*PrsODg62
z=U_<@OQH{{O&uS|ih*QG<4#?50iZx_&Ztez;<QG_{u<QBXG5#0S+3*Et{kNnXzd?G
zr-u}shSuD;>KneXchz@1@)jf7#kIyMgy$J`;6A+KbaX_^b}}iIDLj?<90(%++p0EM
zDa;=mdgpBDRoiZ(kG}a%?l_Fy$#*@?`F~xR_!XC7k{lG563;Fa0XCP>xHBi0JBA$!
zm*BP?nt!k?vyZ>L{cuO4`3!$q_51nl?!R}-4sfjq6Vt!H-nru-7!ig~Km{>IPJjM7
z{dV`S+(K%Z665yp@0VGVhEQx$DTTk<wfXZebovjT`~eFKYo`DCqF-HcO02ZgAJb3R
zKCK;Y7&@FPCFRWFOcF^oHg1V<r-H-f>G7T!s(;chr+~YF&XBkPx)d-3fDQ!X_8;#R
zWvK+7Wv{_yRflG7lEM-S_KH63_zHk9Y=5&|&WL4Z+3o?KNrNq_HJ^oD4=NsE=)lAR
zJS85JxG&wlj(&yUTYp@KvO)zPyZ^b4^>jS09=+3DwNt6;!opWVB|qai8_K(;TwQ-D
zJb%XUm!486Ys0-EB3@j@E9v-aL(ADvT3ywF!zRmuv8k$=`@uJ%E2E)w#&^zm;~c7T
z90VtlyI^YqLZ>*K86FLU<Y^wR;!@rlXDip4;2vp;!K&(O3fhhrvC5VwyA9g1CK$V6
zkfPpupH6`JN37P;EK#i)%)I&CKfjMv;D2%zg;nM?{_%aymHPzHpQzgeAOvBS4I)dS
z2xzlSmOc+$7+xF>_`^K4UA%pubcij)>}e*>?*8rg4F_JX#bDy(_{;1zanjf1H(@HO
zbo(N1F>ex`RKrj!!M<K56b3}955^~qmS)|4Fv~ZYEmILlP&m^A?DG*i4^}oYdVedO
z=V5Be=&hVo0@6rHRfVIB)Sm&{nMJV??8_KNLSc!?vYk=QN{;s^x$f?5yG@!POozBl
zT22&*2De2qr{df~Z6}<cWvC$7mRYf{%zU;S6lX&uMU$S`MMi%FcaWij{2qf0gA-5#
z%?UMjwD0fgjtf;#Zev`iDbZXuCx2Hp;Ex#SQ`_&FIc4D68_k=U#W#7GTpGdD9rPVA
z$)JkIfJs`S)up-iJL$xs@X3sSofs@;@u#ffe6zGHBIL10g=*gVD9yUy)E)L6I6;^s
zjDZsvVyJ%a<*xod?f8m2`Nzi2+N99LfEl~o#h(gl&(JADp4@g>Ye}?lD}Vl9m2q)(
z=+ewnaz%`>e=lX-WGSomX4|}RR^#Br@roqePHq+^DPxEbf35hID_Q;B+i@EW3Vw{+
zC?JN|hgiW{Qzkbhwm@dLP$p5#%5d+(_06pl!E@pH{mV2Ws~Z8^$U@KgiOmpXfdxX@
zw_M5T@7|8vRM1>QMG6=}sDFg^6`gDKN``a6vW)kZeI2h~0@P!Y`y`YfSH}cn+oi!J
z9Rr#ZYUQeq={EK%9n&ZqYUA5v!>q<Z*>FW#Y^R$Ha(xU-DMpx3A7pu~zpFbg6kZ&S
zB`iq^)4FGs?7eS`0lJ$r^TCq3;l9|MjOR?;E<SvdmHybJk;+|-G=EapI7%bkSlvzm
z9)#`~7U70)X`8V>R`zuFbjO1<D3Y-(!ZZ=b{CK0slU5c0e7V=<9j}n+x6wPp?|5a=
z)}nW)A_CYdHL~hL(ID>vIDd}HdHlo4yjIGq+|WpQb&aEx*PmIN87pj1b3-^!Sm_@H
z?cltzBC5aFJKi)wDu0ggrU3%Hi}{Y;ZsdnoWxm~w`XE&ftjg>YRG^E@QH_}Dj{Od%
zbP&yBFr@_nB|r4q-c3<-v$Z`T)APx*``-E9wW_R6cNukP!uP*kAnCXh*^i1Hw8<dL
z$DmE2f}rXfuJ`qKZ^v!IY0{yxS4c)A=yzM~^KXvCcT0Gcdw=z@F?T744I4doDFsj>
zjrm}ADH}CvZG4+Xoz*y~QM;31JGFRF&0`peTcWuAAZvU5UEOgZ3p#v^3z;EG|NZ1!
zmxx$Z358d$6$;(4-@z0Ovym~Fq8iDl{g(4z9!a}TfgGRIgeF%PUK-KU9p!EGaO>tB
zgC1Dm7Kbv)Z-0*-?>(5mx91dSdT4b6HPwo|4|Ajc`~9@ze-o7W82=ko9~U3A`HMKc
z+c`T<*Oo}OclVQ6lbXF?+R-NVEK6^2TOzinMd`I}x@dU$UU$O_?;I52kn=O&m3r|Z
zx);(sRE$OQ`>^*yxikDVK*xt$<nnPjIe{k}1&AT%D1U0vcmmD`LW=26=Z&+YKYmH*
zKC>QyMKsR1m)0+z`B{6~L2XUJ%Dc^no)NutsoDGcYoT!KKKeWqVL;0+m5)3<S?z|k
zkfV?O+X5~uUgDnWwGUTQr{F!OBc^%ab*1f6RP!pB$+G;Scs}CBO?$IoXQ&Dw+^oMI
zDa%XheSa%JS?68Sb~{1jI7J0<W@M|zNjG1(a36q>PbCZeyyY3sFrLns00%$gG3g8>
zZ6oWD`;u6KZ9qt`LMUW!sJ(<iyeeeNDj38g0S49iok8YSfN_tq8KlcSz;rDO#`hbV
zK#`HlIT7WnB1b^M*nA}<wTN2KjR>i0suH>?!hgblU!>0da>T*N2QeI>WvUD&kI(UJ
z81EBw@=@m)h@g2X${;C$)CUi0DKm6)eU-SgK9a+rq7^3R=CVH!p%&^EINziAugksE
zeR04``Xvo&pcc@VTYjnfQ#^Y!;DSXG;b+mk32+aimRY2<@{(#3Wt6r3d*MrEAyGyu
zK7YqkEN7T_h8X#mIfhnpfN4PeRle2p60d8;Lp9X1;){S!R|2Z|d3(}rr{@r12$WTi
zoW50$pVV6A4^=NmRXHeZO@!gPe&j5zHJ~xtdsG3`u?Rj%vB+0XZF&nU9WVVXH`hzO
z(z$L#mCsR|2O8zSUPR-0QBVpm7+oIr-GA4Tl%%!TG^@RfgWUOwaKeDQi%~uww3MSL
zrxMqsNF(d9dbm=lR36wsJd|$_<Zli5|6|?P$Vjk6NvzA_P)DOz(=Z&)reV?l2dfe_
zmx#EIwa>0L`#J&?9C&sNb-Iz0_)b4mZ-YXNdH(88zD}xVYHN#6ox<|W@f^`{I)7(_
z7_l@<%uE(*vP(-SmysbUQGMK%7)#evnX0#u%W8JWM*_ym7#hyZGKlHwq3n#!gtHVu
zGvmz~!B2@_O@)qyO>@?bOW@1rq>`5-zkWJNGvNTlxw|cRVLafBFrIxHzI=uoxD7J{
zed>}rbpR?M<C6i+fz+9~NKa=Vg?}MFY?L49s*~}Y-h%PlgPFC2SI>oa_$fy&E#sBw
zu8cLch>lk{M^n%actRLA%a;lTD<aoY9`wV-Ge3UGc;~=~VQSGqewu$GWoeQN9K8B&
z%|9_08`YLjV;7A^oLptC0b&_nTB~bl&lg`8DKk$vRfg|OB5Ecp=7EBUm1WEFOBn36
z6=%JkxoQn)Kqb+d%__%1?{p@;awVpT=H{}Wmk>6L2*qy#;XHiD0nOHgv*PPbxsdDC
zQ~zr~jWam5=&{^lmeQ{!v#=|fWf?P6-$WZ&7{}&?)9>fEyZ-~-aheL30dEcy135J|
zmqD)qB$p6Z8VUh4mjMVcDSzK@8@UmF&tJjE>kF40k~72Na6m4puK`*#h*RKN!zgS=
z&H3V3hAac=f8TGoT3hm3jxEc|<stFxE_XQm=GRBAQZ3|yOSNQI+u!0MB@?2HTue}x
zirwMha&AI&S))A?SK44sPJ#`_8&b;)2<(*9?vPV4%>soX+aO6U#(&hL07*=Ph>M9T
zB*oL*p0Yw}Pr1UF+_YpDrB=)!u(WEsR~<#0J|qaqX2pGq!X`yW6zv`^iP&CHNQK%{
zb;PkrP{Kd~Qh`(%Q&haGw*Fe7x#lKiNNx3oS7^(xP!KDbgdwC&!dep9Y!#Lzz>q_L
z`Vyl_7{^gwK!PexVt@DWnqZ}n%%ITT5J&(6(gLL%OhL?SQgWG6li(WyHjr?c?J{7|
zxrBf@A{3N_!2t$kZgar!EMl4?UK2GfgefbvcQM2~k9+J&sHPe2EK!?O;P+^@WJCa>
z19Oz%N-j-vb%jl7NVq52#--?rs%cK-I7C|;48pKBE;}YM3xBXw_}R2zS#U=iCyrSP
z=74KigISfT<C+agM1_R5a$E|9W)(R}*=9q$iiHrQ?1Y2Offy9&pH?T@KV`&ijF<xv
ztY#3T>NK`-077F!4GYv1Y+eMC=F}u{TAF`3L#l=;41!{N3YV!i{~~Qvt6_z~aBsVp
zT#XGiq(Y@?wtpZ|YNn<I<4XUO9g@K2fVoH@vkLKtNVXN>wOYlV0^*2JkrzMy*uDSX
z#iV<C^l{R?nx4-m=krTT*@Sd|*S(uuPOoMslgmES?bhEXA5M>co_;w9xMX-pDjnke
z8Q#PwL=#5+^pk0ReUSRP@vi&zU;lPuz(;p}b#``m=YN10#4CVh6<<`AMmpGbHl6KX
z9Gy&DGtFVj`Ft`vchq10<;(n+{rS;+;$nYe539i(dvH<=h)0`c4=#!kk=HJOZ?rgW
z7gkwn9G)PLLs&dvt02^`eh(=fDfj5?0oqu?V#VFt+4N+8GC%0vzTWHJPrl3-tGpZF
zItExa0)JSS@MCYDz6)g?p-f3VBeb6z<rTzMAxBlUwC&;h^&iK-PfkqPo6pobN>~Rh
zMgY!N_u<QVgfG)3_;UJ6051&~pp7MYMHPAmXqQ6n7PQgAH0oW^ybSu1Zyp=7LL*Ei
zY=Wu8uf*)u;JPPWSK`e&Da#Y$O>~S6KLi`?L4SRDem-rAG%c_?Xw1fz)`(3nNFIb@
z!r`?&?S0q%Je_@*%-U4pu={KGrhA18fN%)YPUZ)cgU5`>dtVSEq`?Oe87|dRJbcmZ
zUmed)-QP~nKXos!m7mN{r{~>%_s@54?B5T+&F2@Ff9|@Er}N*gj{V8>bNAob<iqUr
z(|_dC%-)8l;2Qt?*`KhICD<NnrY2uOQ3gA6R4?>+JLgyP>Fo6A?6NzVo}Zp{le4qa
zi_6J>jxH`{(=VHw7n0A&dzj`GoFGy0;6~ss;&YP6;RqNf7gxuZ{_^zbv!BjBb|0pv
z-SKq#sf)pb0ZSJ>;wxxXiK<lZGJ0jkyMI>@`-?>!s5f>A+5n)z86$ZH*DDObLq6<Y
zI;h|8ewn_XcCXzJmy>3?7jLsKWwLywIxY>kY47;~qY5lsB?0S30uyhPZMThVnG~!i
z5=$-a7OOxelQHi8iS0RJyrYnFQ^1Q0yptEL{3+x^WmK%RJiJ)lctiOi<YQ&jZhtA>
z`gXa_XOG3Qj+<f)n_yYvS5m3QZ^q$Pz*>Uwb{n;AXs=+yHd@D#SvQKz`k=_H<H$@o
z`;_V~$;f)J9~X?+{g+$twF7}ivygW1uwWtW=)u<1P9Dab<FSp0*xPPCl9l&})`^xG
z>{LfRrBUA@&XEdpDvDMGzr=Xme}6W&0qg=4-OicN#?>XMp4D8CJ&BBrV2%jPLPCPN
zB`QUinqWJxxi52923$8Sh7A(GS(x3<TstiZn{T+Xoz>h9y|_UgZ0A<rVsnsjk!jIr
zEkHBLCMg(Yj6>N72VNgle{2XAv}gr_x=9tHbW??5etPlAitR%Y#}a)i8GnBZm?N$=
z`jF!&@naN8fvaj|q<^Ip%1Y)hTH^N_t2B;P+9+1(yN%VXBM35Y605^$mBngbV*8Wr
zg)sEcM1zn9ArC?s1Rew#1RaDya~?G3L3187=RtEGH0Sd4+=<G_8SEHYeT6*)2G4MG
z5l~Pn<J=MKT*bawB=ma$K!37@&RxMCPzeSj(7ipn`C22ynh=vRh(ogfYT1GjnLR29
z_Rj3>xF#{dDC}6-?vL36!ZZ~Ic?i|MSmjWkZbOoQ%B9yfluTZ*i&<mCv*2J6!61i)
z3MgG{5utV}lCqb&yMQRQUfYmM^~nNcjSZYAafPKd!7GaN-v*y7Xn!YWL<Fa^uYYiY
zA`1(V*Hm^D5e54ezBZy{?65;wYs52!gzZO*efRZ=$R#lBp!Sh4_*%wBpr@+98<Zl3
zszl)}AORY$p!L^AATGtO*BBufOaLLHeHU1dqOt`f7<&o(y2%!h$&u|b>`N&sRV})k
zm>}>SwXeDhBNAh?Wq(<+#)ur@36UU-Jww$JBX6fnSmVgKdTY`{<FHi4OuKCygAu8U
zPxdK!jS<hX@60IH#lEF4j3`3BeL`Pj#IwW%sw5`Iq{+^zv;D%sx1R~1UPQWs0}Hlx
zS+E1HfFl&}<!zMA1b^5ML_+m3*{5qo^@{DFUx^=8t&D(X`F{g^QF;{dWF8UE#W%})
zi6=(sD)HocLp;mC6U)F8IXs>ErC^~U4%I~LM<&s}dSq8EifKmy(|ff#K`qx-O^p=W
z_m*AJ9V;Ok1`DtO@b<%tH5NprqN)KqM-<Y4)KyvuST5h|j!DL6jFL79T#=i=<zW}Y
z!!Cx0U5pGn9e)|d$TrR)#dZ$49R=jjp8<k0*$;<2BKL6)-8`yMgcV-0@0!LSQ#SJE
z82uW%w~-MQT&l7lr_kA%GFa~x=n`6?QRl&i%)uAswMN7O9;1~g5_TRGlw}*)1%DJ$
zhra72H+M*dP0`xIj=Vu2h^1oXgjhg*M#uHGtwsegvVSnP0z@AyfhgIYJz1!30SRk~
zq<U=ygU%KZ`gK;^7Zg@$B!AauRCRoqsCAQaQ<n}Cb*L#i)D#`+jt+H4htm@s&Sm=V
zm`^-LLD#efhD0y+lTnnPT0p{*r;zSAfJCyvwOMU;t|?0&=l=)3f$q5qWo~41baG{3
zZ3<;>WS6ka6a<%{O+ys|GclLJTLdVVSj!!te?SM?=igs`ypvh{QC4G>V2O&XK{my*
zDtCTBqxp$u!kVSkI)7%~+H@oMiSqSZp>Pf3+NY6<%@lMRp;&VD97dTk3d<Z7D1|vP
z&w<1_j?!sv<W?mKeB;Qcj&G$HX9sDY`Q4Y9rmRw*5z2aW8ddqnm8i$OE=cn`TKT!E
zf4o)Z7Uxr;cZSH1B?Wm#`<Q2|Ovxpl%2s|^3xQhFX?~9ysVmDgI=!}QFwqKTyKhiP
zhh_I81xVZpY`dE-S&=~Pc)S#fUyd}L{$jhDKmFw+dx?WZFHurxnDNF-mQ*K=#WmV#
zH9@Xu|G}GV@{(c|MY?R`YyOE~1Smgpe-56fF!&6atn1rXTX~Qd05n_OiY-JPU#hkt
zdx^}u7xhuRKU%NLwn~#9X}jp*q=2QHZLu#CD^ppYMDq_$k$$NTXtP0Gvyqypw7~T3
zvvwyZw6}(qLZKloBvK78Wl<`d`Zb$uHhEGmJ7eS*w&O$o{L~rpUI;e4)duHbe=P^4
ztk9jHRP=Lxfb1OV!<nLMyFw?9a8pRklygX&IZ^&yHNhilGzP6ch-6BgE>iGF`Zz-s
zgty~;wpee18HBfw!?zDFcYn?TfHo#5@$|=+yYPb|5X8npj-RU2U;jRRy8C&gn(a&q
z$y0<5fA~}MjPA)VIazwiJ0*?Lf4bW#kdcS>Y#~aQovON=G|{_u3@USKWLR(*sp0!X
zF(yaKD%ux&S7yga)0yg1*U7P0oN`re4u<R_BSm~}G~lMtjyX*fZwNsmrNVEwHcJ8~
z&-oT>>nX+nZX#G&Fxg6TJFaL_!l_CV{aMM!Mnak6%X0%0rlff!f@qE)e@agvPUTrE
zNiB?)cATZ`HZKGL{L?TCWRAUA_tD8=O#M@wQW5agn3|JijcziGP>&OMnt&4%f6fvb
zJbB&&1|40>4C#y|YbAX18w1e$t#zSNf+UXXl0jB1&?w1SWJ~F;tR-b0pbIuWVB4|!
zTED6KX*0?01Fpn-0wNize=UNMs&zT(v*eGE(DScP&D_ge;ugv(Q#7Ea46<c&nmDuL
z%u+#Z+MMP{Y4vPxaz3AVfmA~IDK4pOMWz8NQ`URx(8DEb##cbK>OGdTko6-sf(eOi
z`;p^(g|qp$`cYoRh9Bj8zTijU+2l)l$(KkvVU_BneDmpk%l!yQe^~6dp>;nFjz)M(
zPWQ7|9oFzH;-H_M_CNQVvLOoa)HC|L%>}Ih8<yi3@n$YR-M3aW4FTaL_JUZqOqHdJ
zf;3a!!DhIjP*E_sVf$r;a3zIC7rV?sS~cDb5&LmHXzbvtY9QRTj>yTo1&etEC|vh7
z<yc#e7E5s*2|s0_e@oG<+hyc$-7LFf&w0+wo-0S&6rHAtaI(QC^ZZ)qZWh%=K6Op_
zI+oblwnclA!8C<^+k8mBs*W2r-d$w!?f5u?rc$%U?o<Jp*gD}nVYUi$UFxQa%}P9M
zYN#|yUNXwzpUE-IA#E7vF)|9)9TD2%FHK3wN*30bnSF#}e;8ACBOfsCVi~B3gmHeH
zEw8)rs@kR10yBQwiB|v^VOclj!)6!(<bs-mlwy=<iH-y-KoS~U&*k$D7ssK2wO~0k
zDq0Fw(-36Xs*VIJ2<1`~q?xi4tU!gpMxF3ouqw7R1S{KQRNIE8A%yics4o7JadDfA
zR}de|haPX)e?nAF7-g2Xa0MZ14%OO^T9~3&iX&_FbM0A}RvU})rfY<)XYg%3_nJ-B
z*sNP}EiyMrV@uPDEL)A~HNBhMl|o^iy+zqd839s_%3e!D8U!be3i%du=R)vlb8XW(
zt)R>978~GEAgE4)z3Ab+;IPR{4DW|5PjZNIvt>45f1ywJs($*)c5b5HRu^`ox|n)%
z^Xj5V>t-!mH~P@Bl`gDq8aa07&WMjo+d1=2RXYQ#w(Go5jPqHQF@QHwn5Vc!C*ibZ
z!)`gDF>mnSw;jY8p|y^Vg3cX@<3TQBoZ3yPaeUcW13(4DTZ}azWyQFB$ylvjWCALT
z?0CG(e=}GcOEXXlY0YTfYP;#78FechC#?ktPL$r_gj3lE{9}S_6i=|$AR9&a8;vf0
znn#gr+L1~Px^Zm$rW)f*Pi^6gt-D+XnEJj@qA{JMc{CaNEOA^XL?5|bTi#u&nB<(+
z`ZNWBR5;IOPf~0Zo@vBV+Dd#(g|Nvf8@E+uf7VcgJSeZ1dgGN~jzx<$09Httx`eum
zG1D1C?V!xMaa~kQ7<wMTZtIuZ2H$JDU;<sGlB<p>*6thu<`(<gXIKQKXtdyTOrhBc
z6=sKJ0JF1dGcckBp`1FA;n-yd(6t;I7^8{Ma3$;6pEC|fA*<{3qvwi(M`>Xgb^h_?
zf9}`c<>_Di(~qI#@b2`#IPeSekUmWM@97_RKc~+93PMa2Sybd$-pBCePJ!g1tP_6B
z(aanP`@~sVb0I+vCx!fThYNA=_AD5Ka(yW72f;wPeoBq|%ljOfE1K0dY%~{b(#Z4H
zju9HpGQOM1cu&ji77&F4M;d4CwSY(qf5niG*su%?%2;FMghl8XZ6`EhU@5I$=CYy$
z&V9t?#i)esqAB=?l|IS{rly{B5Sg619!Rcum{$y!nZIV^1%1Q@(Q`KBSVxL+dchvs
zHW;QjY2#Lh<A4J;7z7K{yD2-^gE$HUm`)vgTg6Bs_!Ajnj6JkdT$l$O)iK!ye;>s4
zu(U94Qw1L@@g$%^sbjN~xAIWM=|dBn?E!!z%Auw~h$x}*@3{OX(QYe&M^j|f%#u5;
zR7gRko#M9lllR_?vhC~+h2>yB<=y=QTKi_lc(__;FY6}$aQ^_ty!l&rv+1pV`ysvk
zbpN1<pB*{%XACWjpHAmm$f6FNf4=b)K@yfRrl!ZHtx-@C)nBkS<M@d>klG{WG#z?N
z^UBeaYmOk6k1#N_*?7sL;83IrtyGZb?WPFFCWiY1jhl`z^xjcmUfIeKMu3Hya9QPD
zL%Uabi`WjRyxYX<SW}k`p>9K{ziF{~G6sxx0GBqy$gnt+<rC;*_Nar?f0%`fQk&;m
zT#NyO!g02}T@H3yYy3zlC%v1meISDhuB^vvAHboO%=Wd92n<&a{PPt83`dM+YE(no
z(6rF3GUgxdB$Pj~5#Bv%8sNaO4mjfsbg^%E-$XHAO&BQVO_<%4R7x<g90uoR)=g|R
zMaaF)c3JPh@6fz2?kfa@e?7m>gCs0K+1Em0;^!he$DbEWMFxzgwkn6hrqm95gk6MH
zsS#(k2+I=+&zetU|7^}QxL8sCb`r-V14eNk?KWAqb+1Rdo6Z3d$Ai~g4AmP-@{+FU
zCI^AX{Oy`;YZ9~D@1lU$OS+>0PT-#W*X+&BB#xP5AZxeDu9@R#fA<G&a)Y)pB9=(@
zbWVqkZhkb;HUFQcOg$bn9>5a8i^CDo$bIMHlBXCYhgp?3BqWhq*xOhJlgkg3b}nIr
zoZ=*ov%dEwjMv_TG)fSvF>*R&RQFx(LB<=5O~cmb1w+%o>u<2_-@o&MQ`bm&*&z5c
zuG|p2q;WYWP1uI@f5sLNFb?_Y{^sQs={?=u94MjOO2#AGaZ!*2U%q#pI>g8-hVZg&
z_^r!r5KBHTUr|7)tQ&v3zFq-6r5yzfY%i}yBLgM*ewnQa=hpUH@HHlO2S&b60}A?z
zyR2nvBnf40XC!UOsE_*+2Q?G2cTe%v_jHeI3jssC85jc$f6C%0mGy%<5$v99C=0<x
zKr$|E_x~{tDGoIwQmIDj!OLFZDQKMF*GwC@XtHP-X7i};ziFX6Bc0mWwP`f_Kh0KE
z4t<lZx)}5=E_!>cXRwbATse5RI}@?tC?Jub*AztH1=sm4nbokDeYzb=AHF)BeT^oW
zg8*+kk(#rwLPPG6ov$PD4J702Xi*JhXHI)$t5zi^7h4D|i+uI?H5ECAYb^?Ta?6<r
zl@XppVrL0q_vOYv6pZhT0sjvT0>p`zVUipZm!V)k6qn1_9nJysx0=`;5ds6UoIJO#
z+8wD01C#O}mm#zW69P0bmyw-16PJSI9V>rHbL_SezUx=;7?(?U=nacYHC2+H##O#0
z(arHGiGA@q*cbo(G(d0xNpaIm4kONxU@spV-5(m1ZGLNP^W($I`@@$japO%8@YuZn
z+JIiL`WLL8bkcYL;|QAfZ_Uq7)}qtTi+<QzBK>&%9`q=eM}nXu;8{O#dGrirJfR_e
zK)=7H-`%);x3pLBIZL4UR3uydq>&yhC7CJ(bS%7HmrCUw8-Kd>%hloHEr2GluW$LG
z-131I5aWRBTUd}|1o058_Xw_tSO!^LHyCTJ>MrA@C=QJ?(NmE6E;fjycj==EMKN^l
z$&#&TgK-gSTmPjfuX?EF8)i6yBO(|^Jc1)48;JEC05LE)kC7n)dsu~$)VtvEAdIAs
z+{aMEh^wX&cz=T~)V6-VjtHTijNu3(L~7YIeo=8!-Qd7Gia@J~u=>W%9YjPCH<Y7f
zgb1&i%4l_};UPqr$VfaCBTQ<UOoGI+*PIX-14Kk0gjX>UbjIdsa|9DfkB?)G39T9{
z#D-xSCHa<FhH_|Zz(W~oe9O)ZHNFK<KuFWAPF1}ltbbw#Q4E2rBy3{_F&a#HL!$9i
z^vXtpQbk-V+LNGI6&yiPeKuOsj-aR$&#;dZ6kUuSB}5O-1<SmVpuEwEdk{v_(rY#3
zDj3}=j444unp%foB%K-J4@Dyp7?DHKNIESDC(&qi=IRL!nkKza;El8vjkW$Dj#`gQ
zB&u-?8-Hmn=BZoR`u%s0;CKTVW-@H?2r$gfz(f~$4(y_1?}Hiw8>x*n3kw~@CFLof
zE{$>V>-nI|TdXx(vtm({*qtbuk>`%xi84A4>6d>h*Ob(WvPD)FsA;=Beodp4*L`hy
zonn)yT{`?Y`Stex@W(uTCMFbP^Y3pD=>ls|l7G67dNoS(=YN}DAO1d-I;Oq>)Jf}{
z)aeTnni{CeWfY^%$t@?W5?A}uo=;=hx99n)CtBh<)1b#R@j4Bz+^Kg+7->qHAwADB
zCKN46scIGdBCk{Jm&Cs(1}tL%dW6w&5o;v|iZxHMh7gP$!0tqJqSeo=FX;EVO!3O6
zl7GaXf3eMf^wLi{l$0#}l?Q(7wb<tO=AY?VZirP|vREy=uqxO#niE1SVh0gII`K8I
z4~y7onImI`*Trg3M48H6jaM+JvtOK&B<`1L_Uq-*SxnP&99OXo3Xya*g$PvR@`BBk
zU>RJgLZS!hv^N_l3Fer(mE7Lm8_VtOWq&Y0?oLltQ!OiPU4E**WB^~X!rmu$+Tsl)
zv=D(-xf>&)^+~I(=(>SjpDk1KAV+e^h&j4_7u8&1KBEHd6;{4j1w8OhxK6vZ0hfA)
zx>C3gTuOVyx*jgyP5e>mf+2i{LeoJg%K_G5Hv}S%dO@#k_a=35@VXP-(<p49tA9t7
zcqepqdJ6Ze$mDY$jIPCu3kNCz84LCaW7}bu@qOvW;gkf#;hFUuX$#HMUazCc(HpuG
zmR4y{2z`o^WfT&jva%0_v(|245ZVfR>%w$vxxKw!GJU#BVPYSQ^YgldCH!-8DT#$g
zX=IoZ>wLG9ms#y<uD7W@%p;P!6@MJ3VcCG&wWazllZ2hmor^BvP(kH&SO@TRD|M5D
zI(K7i`}eu)?n;kdQtocgbC=t(<aS7TekaVQ;n;v#S{`s8n8g_{{Sq1rc>Cbx0Pg<>
z?`aq|;N_M#+y`D3<73Zeli!@8xU8Ysjj}+jaz*SOLU|y1MNthkpL<DrPk;K+!JuM*
z%6Be(23yYU<bCI%*npT?@^BxBiA@;V79T|LF5vM4;HM1mL)_&g?x*qDz#Utnar63p
zW4XP(HX4WVPW2nmL|7=`^ju1CdU|gkFQ`WAiJi;N^Zu;kqyiEZxzBX6pf~p9blywp
zr_{JF7%KMShDoLz<q?9d%YQ_kGf?)<rR;Z8d%l`KJ4?f9JhosK;{tr9uXb3(zD8ma
z%i&ZjaoVxm@lFIeG(1^ZzjXcX-2Kkv?1DJXv|Ns_ZSdq!NxAuT+$HDO$w;fmHs6j`
zmED@X@3Z%_QZAW#bXu#NejRJcGTix|JEh%QnD>F5iR3=rb0>`xAAe!ob6+o0r7J!w
zQcJz~wj+L%Za4-pA<}H;fig!w)eI-=41>FUJDe;T#PF$UdsM_e%P{RJxI5K1)SQsE
zC`%~i{0W7tEbznpJCt+JB?ousrOV6v^6z=SHt-73eC|OWm9Q^#lT^E&)o+-7QI}{s
zt^4%*+@;^#HE6dlbAM>eC6M}N!%RZoYrwuS2Xa~62vrHfJe6dxeKPkWt28OMn{jm#
z?vpt8-`)zj!gdQabwRo@L%X(3=tCxCH_-c1fHpcSWmH>mNS@A@14M5C=uA1VT11=S
z&!4|FlQaL=RWeWu62;jx57n!Jdvx`J)Bdr?iD3J7USN8ANq?RAz*6~~wc?FS>8`f+
zYbCHB?3@8Nb#1(-oT?^XM^qyWd(2p3uWf9%l{qgGx@FydtC4~dHa~>~7Oh5IhI@u3
zt7BW+tNQP<7i9H2y&bnqE4zNCmhrM>n)qfR$o9BWCG0sp*Glc8(nJ(eVXFq8b(cWW
z)S7CRWzD)KB7cOLzll-Zf>nW!h9lmRRi#&K)t%e3Du~Vm^q(olhehnPEOAp8?!&4;
zBVLQ&E><OlNriGfzrkI8`w`1}-p8uW`?ZA^YQB%3dXh&)><iu`cHQ9{(kkF+oE=_w
zqJg+bx5TQ}pB2sFc0;C|eIBod^Umj}bHu(UAoG>;(tk@<2dGV7LuTG7P|gqk#}F?b
zpXy!kT-pk?S|W1$ZERcmmwZ3Pv^#vO)i+ozh2c(%Z?GE8!`$8=kw+^{iZ9@q7F;fJ
z9j%#+rnkk8^kG{987($b=^yY92;MNwEVnO3uT2Uh{jo)Fb!MzLrnkt$R1$|}y}E*p
zJj0-`vqal^zg(kTdgAF{NLXNIN9?;vH67ifX*ss?KZLlr;qj+@NpI-;Ii{~el=lBG
z^ns0qUrr-4jJVw50%mg58T&u^M!%QAX(1Gs!D>Gh0yHv{;ix8;@R9@yf4OOqEX7Ch
z6|4u@8SM)3A&3&)93KKDatv5O>>Tp@sp6|iHm9e1c4PsqnPHQ~s$#ua%cmbEKE1y?
ze!P48qF}PjX<_B`@$)3`1;_sd$3defD+Du2Odr2Y-#v0J4L(PFfODnrvH0ELr#?I=
z;c$j{!bcn)EQmB;Nqhu;e?QUhJTI))@vVFc5R`vRkU2gy(+aK#rb&-FeE;#=JD$)N
z_$jK-=Z|+k-}M9-AsHAqefn}oFF*;VoSckh8tnATKc~-k9}0vFnKMV<@c;KCkk&9K
zErI;*dp`XGUw(^|m3Pzs!qhKi8$SIs{S`O31lTbi04rp+Ujxi170JBw3G56)E6+Ii
z{UA1CW+aPDJP(%&?;Rw6OZ?R7x5gLR9v%!Q;Zi=$GBk=0iju~u`Z<aaKpMv}K;hHJ
z_>P1!%*8ocRlAU2BTPXLuES~J5$~edRUB$Wt?>R@CSBk?aAu9XaHN8H;d;1R4nZ@`
z?S#XObDs}EJ(ISPCgXF8><FJOX*eAcB&|H9=PA8c>7n<>d{}~i(>M88!By#4c$rhZ
z36O1iboz7&s;}ncouAS>8h^@r@#n>^x`UC|n~I_g3(|WP_m0@&_vf#JapDsRTR9&k
znU*93z_>0tax-Fvgyciagb;*d8qWmj1TBx-G)~|BmP5zSNb1`cKpkS<TJ3t#v;u0<
z-~8@At$^_ggDaYU1;4qkJ;r4fzFybXbyzs&5b3Y0!%8vk#ZZU09_v#jOuCxy4&gj~
zInOQX%&$-oS4po<zx4*W+NudE&4IN4aW-xP^}am?L*<r|;rD!SuP|C?sSC$9A5Q~(
z&A!{to@XOs(niPIs(QtUjk;d1*sJW<E9C`c*BDTnSA}tZdqHfBToJ5U2YW7Z?L*`q
zw3cXmAOyufPdMxS6sFz<=o-O?VLU7RH<6JAJUx7=jWg3PIPvX&e)#-{Pk$E-m;INf
zVz7?!p+MZe#Gl462|`%JEdF*(JFgK3<}qKuJccvyN|>^+41AjIKw8{gyK3|&8ubhu
zNLeio9I!Zlv20KcWgG}%8O{6sR1apX<}CtE1r7Q|P|{lZiR<$!`bkgyJ5R+`|C;VT
zPj?>@SG6(5(y2SCG=3UOay{UxU?41}a}{UHki#$1s>~H{F>pRNTG3kuDfl&@kV3~S
zS_(B4(1qqaQqi4<5LSkehw^7XeufVntvFJI6dfdgI!0}Stw0eKiMx@*^Y>3*(z-UY
z%y1qAN;#kcJ!prrY(t(EJTn@nQ5K!1=h@ja+hiG4atEzyp|Ls#0Jr-vRp>;z9%jNm
zY-R>#B774wdX2A`KmkX^gfpdtOoKMe7amAT%v?;;LIQS5$=d1gAT>NzsTrkl^w#7z
z0Q60NKUVMcL0J0zM5$bXpbwIK6D4~w$)h(adqR<+Xfw~zKEEYsWyhQ{iue{xSt}!^
z+E(6<KHtp=X!tGjjsY{CcC>S%OIT3l1VYV2rZ<5LDfI>0<5n=3A=8{=!-bwYd=t1r
zNW9=IV>3r=O>e-y@CbARljn3hav0w<flWSta=@%om9zz%P)A76Pi;!Cynf7SeEBuN
zTzS(`8RtSHPXz+4_KQ_Xn$jHANs6uu4>SuNXTIxHY=FF%EzrwedWd1%r;?H+`E=$+
z(-}~Otb&kFp4Rxn?;z)SeUJpx@S&kCpJDwjJuX5g2>_9~9jF>{ITk=9+Hi7X<c^$w
z(u!QnVDvBVmy(c7#hUhyHKwijC8+4E1v=CYs!mYrVo;E?EECa2%)Css$bmHHxmgQs
zf^uqfwZ?5^wP*V{vf8r+j#m5MQ#&n1Ru5HZm3<8cCf*t~s@Q@@@aQZvrxS;>%!UG)
z=s2Bq3tYpW#2HuF+2z?>svv_!=Y-gQ*S;_7MdbECwpaGds~~D>7|r-gS!rs}i<bl|
zQ*RQ^A(LL<EEQl7xWsv#QVa;01=FvR$IS6cXHPRdMK`1685AtR6s#+%qbDfTLV=KH
z+cF}fr4KeGu79cc<BIRhZ?k8B^S(LFto&iNhyiP;ETye{nj#CMo(|e&{gS+Y^D~hi
zZDtEuB7kTYBNtDtr^!;LJ(F|oqp`km-6bm@^<5__qg^QWpp8bd|4M#1%7&2>M#4s7
ztX|6o_ZYaf_`P{fnJp?XhD27mrko=&v(?h$63eVNamk<VS0zy55)|p43s05A9oO+h
zPi@scJT;Y@#%NkQ+{B<=6sp~S<eU3dysTorTk*%dLJwCsOu%?Y$=^_#)`{_gVIVXC
z=6qJ`mKtNmOB_>Dw@a#xtQ)OT9GKR!T_y5HZb*>_W)$IOSTOpu<j)t#K~4kiqIsw=
zg(W|(A|^eiwnnr|z7eaZqtJy32YaLj^^9rdTCw?YRl3ns;NU{XgI$q-s`e>4_E~DL
znSDyMOGV?89+yiV#EF6${CMB}8s0DB-ponp?-wn)1k6;^{bC1oOG%eVh5*j_5<oB0
zhJ-Xw#+LsJL0L0W2V}}1f})2k`8d!5z*+rdHOOB0k?w5Mjj!2{mQx|Y?RFAp;y-5@
zs0Dk}S}vAiAKv+o%xIK<F|SNvw5j4gIeVqzF_M2Hsz<rpxm`FFOXdi{?dOy5Ord7=
zkj@;RL?K0G>Sn9wv~MUH2`a4cv~M%gh>gUE+bs4~Df)sg+AiMpZ2S&l0qra|U&@v5
z^KOnwPUWb@UIGqnv3LhWAxV*YfbWK+O0;)U=zcbl$M<Wm4AnM&cS~Y0<dsTprA9dh
z&W@-_C4NqVgB~Q5kka1G2=4v%#q*KXlbSjL7Qj!7Jn)>^MEAtFSZc?y%6*1Wc<M(n
zF*B8QwZ}!d<fo$5uOU$Z#)Evp?kWe`LZ`w=04>3iQ?JL*Vyq$EhBMeAWM}b5`>oKE
z-%gAX8Z)p}DHY}i$P4bJEN{=3oc0|nfANeLN+Cx-dzg#P6^mHQO*vV}E%)O|QsG>`
zv*;z+_^cvmexe_&M^^MjkR*Xx(00+{Y_n&{ICs#=9{md3qa?z5xX8WB)}~IG=z6o~
z*6Dh^=a~6wyTEZQiArc<az8s~2Xs<_x307IJL06QYYr*ZzK)Z&)l9}3gvpZ88mfnS
zIbK@m_$!=qilaIl_IPJ07@%VHyt%EHVD}vwf2jjidd`$IeTJy1WaF;uKvjU6Db4LW
zP!;fyoxh@WVK<14m<e=sU#$r;cA_dwa-1=@?L^&gsAK!g7PZjNp|3|iRd<B0tsjSN
ze0*eTa?LkU<7U`HDVlWsuvHqin)1cQ?m&}|9keOanZpOiB-(5`FK+9q!r7MHt0)YV
zf9&b_y`IAQNkQlNW+ya*X#Q&QFE%^LaG}&J+1c%My#lT4C7L8-#I0MyV@D&^7d<px
z$-yo^$4)dTt%hY(=h3$Nx4gIY`?ldvo`o~_wynWg%rViul*IVb0bR|>^%J7PuI2LU
zoRGvw%C~Lf{M;c`eT^`)_H|6J5AQsze{F(LcYU^pIA|w_?FW1P0bDv*i~=wQeTB2$
z4Psk-LSVIkTl5)j_!}j(?hKS0-Z00TfPCNgu65JR-FKMT+(*Xy0hn+6LuX2yryKag
zJ4>OCGVPaK%e=O_V@$Y}inCt!Af=g^E=YI05!z|-YrflX)m`psuGF?w>>55{e{=0k
zh{O}su!wVo35E65n~*IT)`r^>sSp#Xn&R~3qaP<+DoASwaYKPR2QkMyDpRW!R2S;M
zRs!!d!>6a>%%zHSo-LUOIh>8R9f~jV%O+Fak-1Y<NG(__<T!Juizu-D?28%NxvS_R
zv|M0j9c{-OzIxe7L0CYcUh@0Ze+zH&k_PQyvqVX&zz8Wk=6@V@WBqEMb0&SeYw%--
z)yrHCOd^{Ex@W~*d&_+0a-M35UbeXYcAX_5qnC0IOQZxQe8+c9e=OjIu@<b+R_XQo
zV4<SA?_vpEJmZ^&BXSj0B_)&KcX321z^(lrjwlpqY9DHaM1b*%8`ORje;_T^If7?^
z=iqm81SD!8TGECku-aS!D&BTCxYj)ncD%|lJk>qQ27nPncg7d^2!Q1D!q_X@DjPL2
z1ii*iHfp+)LosRCs8&(c)Mti?Rx4TeqsAtIR@rj3$3Qhpwi&2q3HbIO+;8dg6V&oB
zX!Fa#l*<iCb8}@AY?biqe>i(;cwP^FW_a##_Lyx*(}jY0Pu`D~Mq}5v99|z135<cf
z>?0z&Jn{O~y&;0G67}30a-sPlmpXT}D+ttE;hUz2H7?~wiCjuAEsZ)h@VC2{GE;@X
zE2^wUGrRmVe;^AkuPd?&^r23Q+ZR~{X}x+?-nJXWHq1PQ-TLeNP?<lbglK+BB>gVO
z#%j^zA6rsWF5b**nu{)E-B;7H&LWG3Qka{8&Dh=;Z240}YBIE6=RmDQ%;C&|+Nzd2
z_g!E9lgw-JU8r{V`G1eC-<MJKE)xMXm!JqY6qjrJ9V>s$Zrd;n$M1Oxy+Acmq`rV6
zK$12EdK;EovD3EhvL3d}UVoG&E4Gu=jvEiCDo0ZOk$$2`TA+R^P*1DvVfATGth^2u
zy-|l(h0+1?A3++ORUT10LUnjkFZTe(N!dza03s<~o`ckOeMcB%1a73@`pz?!@noc+
zJnz!yc3gk+ZJT?|rct1HUz9;g8tD-zX}VH){q^v?f--%S8awslVfA}8T!1w}V_@~|
zZIwEhP*X4pE$Sfa`tw7*u0B@)tw2)0iB=kPNKf+rsizq<h^BK+YM(S;zJmHLozF6u
zVo-mZp>Jt8=xC$*rM`$hbcqKP;^3UMJ>nR(cSgjcrqq>xi6=%ZgsrUKO3izz{rb*9
z;>)Sk<Q|K;hYfOq_K*gd`y;l_ntG@#MGYK3u&fbW)@pfP2$v0_<SH&-a0xc~1KwpY
z%n45UoPc4AumS6jBN~*hva1vuGj2_OmyI+TtJzPaJDb7?U1p7<zYJaWn#otX&@o1r
zneisIRfCLwg?(ZSExU&AwjF0$>ZX`;JnotBo<?)_sz+!NX^glKO>&xjXlW(`J;#B4
z>YK=6_IyV;7^4lDo9Wj1-dMNJSK^fMbnAT+QA;Au+Ey0>)g_p61R4iZoC|kc$?;h9
zwmF=0MM)2~`4C+8_s~Q^sdp32V>Xr%+#n*z3mp@GAT2<u)1g9TS6q5>I{PvAO%D71
z+r*rzGzo?)>4;7KJGXP|-%auFn0)<2%(HR4kJ*Nt3oL@!Mxz~5P3;Q2X?QuH`)_#X
zV;I5fbMCMV-XNK?B%ZAmr_-OT_@Ov?*YN_e=PP2n31vg1N2rE=PPa{IwiFqRB7;<(
zDa%WLS8kh_`^`r&g4pE*Vj099v^RFdf+vAD7r+m*z>D1NOzyM!jBvL()wp&1zO`<h
zFI#OfE)%~k2OYql25{Px5S(`J@$ySm=bge$+}GLb8R0}tDzeORB6(%znx4{0YFdw!
z??mI!%U;~^kV%emhXMLwqB%2|&r5So+A=(UKT|`)TpnYXQCmNfrWzeXD|>EcS4xb{
ztlVd$3yk+=+dB>z2%lbZh{I36B-QZK=?`=8nzE<rU8Fc2$+h<=1Ia12CkwT#I7-fk
zHCm24?VM4Zy6j?yP&9HD-kj=vCv84A%SqQsEW7K`4JjCs)~j!milhU1C5m%)^nD{x
zW-Guge#D5cmY6s*b{e8t#@jPXwphJb<4mvw?rzXd&SqEIO5%j&?W(!M+?BGK9?a)2
zYYdmX<O6+ZBd?X1XuB-m0ltYh8r1&)um6Yemtm3|69O?ammx$Z6qiSHGbp#P{~eMD
z4f<yXpx5`O{{ctf9uJpck{lBQF*7-r0gyry0Wz0C)EOt2p9mhGfBwAu^7T*eWSQu<
zUo1Wa5iDBW&k~gq9>S9~R-jR7mf!s4eH8uni}O!b1gE^=A|Wlv8ZDd&Sp=(yn~tKj
z06xag(#RrnF*EIGY{R8~T@PcFur{0qg7+WMxo^HN!gG6}^+FmhO!_2sOHKMV@Uoh%
zvW}F(NB(<VpPi{ue>o6A<CC>gkaQM#S!%wC&^q0q_={;dPa3g&UjMzP2Mt2W(DXn)
zWynu8DMbH3Kk@hY%tgO-u?2ag!L^nk2xT<u`<vJGn$Zf6v;y)v59ZdQ6$lKLk5U&+
zJ8~*Q6O)FE{7jo~J&>k3A#ka`tp_4QC>Y0)>#d+HS<k75f7Is4MYQ2+gRQS148~b+
zI2Ay+LD4_02MJmj^rHe4EXSc3D2yGZ_RD%86E`G@kpC!aZaIN9!s<bUoWQJg;H&`4
zg*N8&AV32b`EI!>YcjDCg@I#JQc}=wc{3XGf-wPTzubbLjF28!PrU&y#-{*|;K%V(
zU`)wJmjIAwe-odW_`NGZgAAy@)TbuoC&uI+(zo&VFHBaKpc8*E2b<bDjTHNJt<Tpo
zn>2^^^gN@|n}O1^EA!Ab%$H|2Ct+)%6q+O8rSW}karE0pY8o-mr@wU7RMZ7LA0ujC
zHPUpc4vVJMzaI09DM?Zf8g%W$ok!H!7xFk&m!EoDf0bZ8`WTrfZMHt}W42DRoKKpW
zN>b1P-BKq(8l7j6*OTn44@;(%zxL^)*lOv&mY@V68*7tkfijv#MIx2}i@%d;4-qM9
zZeAxHaE9jN`MlPd7CZ{??WnfqqO!OF9!`?ov#BhaB~;wVoIO9mS-vjH4}Hsma5_2T
zi#r}qf9iI9?aWs<c5XE~ZQImwHEq^>tFDx)B|%KZ9czP5HIVmeM<N8fwdbar+d+e)
zpoXIz%%`$mUsh4W!Vk>(Y0bh)vZ5lqs@$Wvk;V?vGwB_0M5I_8o{UBUYH$lURuKyR
z+UZ+p06k$%%_p`&1B#%UK`!!PgP=$RhEdiAe^v$sbweUn=;{+|`Vp~!42F-gB_N7`
zd2kOYfN+|%`ZUfuw<3ilLqx}rkwQ?vGvF@n)%D#Lae;yoDhXU<kAkA@JzvBQ2GzMq
zA{;tAG`C))*<Q~lW`uLYD1WV^sHun1#v+zQA%TEhS;~KI!87d@&98BNHG%>bCGj#!
zf2N51r9|D>;&YXE6-MwjuJv70tL#!PkL1~Dy7psomo1*hnWa2%^#988ONA~KiVY*U
z3IR{!rReAf*wo;Fe?FpagIUM;H**I%p$)gmE_Wy;kb<rzfbQw@PVUfBph9kWP;Vt)
zR)~24m^3kt(2&*wWiFA2R<z}O-&xy6e;4jbA=G#rhHqYKef6^Z+XOPF=~XOZ8T!!u
z)AW}rzUU9wpU24<CPUPOKI{p{<R?Aycq<x}-FQ&^&-QyjJJv_VN*7~1LJVk!tUZs?
z00J|}&haSXL@y$6tF^i1ctq@7fZt-njQny6d8>BNp*S22I1!m`&in7=(}vv9e;@Rr
z(bA$<HKsCsFO@+ow)nvM-}<>BKUd?Fa)4_bvpNsh6NsL2*xYW-Vb_JtNJEe>WXr&j
z){<fqUrUP>(B^CU*{4KMLuoZ>O$?tphUj{E;)&~O8D(xlp2G@uRV+=Jn9FQxSkyFX
zK~uP@nR~6I2-yJq$p>2wGF{HCe{t;KK8>Vj6ZZhDqu<&cyx`zt=7=%N*g!knA>Spt
z;6NOd5G7N?8t-c2qEgz?J{!ENou>^3^9^<}kk<*ijsvZ&M4?6mJh}3zgp+l;jvvzq
z6p0ar!TQVCLQvEZ6YnK=u?3YNdO3_0OGZRmd+g`=8d99a0|8dVg-(|Ee;u<FTG^qk
z{8!6%g}I{%#b3+!whw4s4r~RpD#K8p?_+(-#mfK|v_Qi6Q*F&5H3sU(y9Aergl3U{
zt@ko<{}k`JhO@4Kh3dS2%i6fky6{+ZdP2M}yIL)=V7Al^KlM&v4DbC?abHG{fT15{
z&rbtow_qPJgO>4RWNwA4f6BWR5YAKE?C?mPq#kY8r`uw!i&$K&r<<3D4nJ0W*OiP*
z`yo{qiw9o)D!XqP4{9V)Qb>1B(R1wrC}E<rpcl2wxb|_%RpeckjZ>otsu)Fl$LSb9
zV&2a=Qm7gz+VBmvPFWm>5GyQf7;f_&7o9L3VzOQ7@Fk^GGJK2We~pnUdT1LO`_$1|
z^I~HtjZi-?vBO)QQ{sbW90kZ1l|J7NO58$W4{ZBV>M}p-Sgu9x6r0qoy|6J-b{?^T
z^#t=QH61*n!pl~fb(6ArUthqUwJeuq$Oj$pRjs3lL1HQL5kEJWk5@-O|6v1^y$BSh
zVt531t2l`9PJ(t6e}sdu{#J320*pYge3bp|1gnG$z@;{~6D%p(o6|tBEO{YeIU<y!
zR4d0_OwXHiQ{WNR#Uy0(L#hi_AgTvd7YpVx__^vXDsqyjLqI;ab>dPW)7Cp{`y(PA
zW$DX0qoYVZ2=54vqJ425^2i$Va(+z4lp|w0CmZl`{#Aiee<ot2eB8R4tK*7%MN~4Z
zXh**yRk2*@uqm_$LJN=XB^CFQyYPp(eOxTf`$z5LWh|9E;&<MrO9YSj9rbRMz24=Q
zSWXYo<PWz>&WZ@7MOH&hr@XK(`y5m&X50)7+!-2tF0!$YGxX5YM8t~S{Ry|xJv{4Y
z&|1G8;5;9<f3<H8+SH&&X}2!MDU3&6!1T1CcJclYoK<@$iI<J85mlMq7O~w?5(hoa
zxSm~Cn6crs^Q9V#%#9`bt|F_$^YL_g4sH328i>l0pIAQ?QyTf)N%`&~!*~{*<V-hy
zatJ7`s%Ci0-VyWDzG5(5JIESbUmT)9pghytj`CTge{>OYg-hNl(Zu_L@kEx5vX7Jw
zD0M0>_3QR}k|X9l4zwwxc|{OUjm~o_(db%&D5|ec{5o@{>>h*T82SHIQ+4p^(0N!N
zu3@&3N_c6<^`$LtPFwtNXO)P}bekzk-{ro+^8-wT^O#SJTm^oe;<!SgO?N=XWAl_S
zx>>kbe}iubQbZsf!5RTcwhJ!kFIhqdlL|Lr%IVpU>uWX<K>7$IoiHxkffy+By#zL~
z&COn*EFRHgC-T4z4WfOZx%BRahPFaSu+?i~b3>zuOf_*8sA75C^NH;)C{X<HmWzDY
zhnyJOdXeq^sw$5>Bx3#F^-hC9nu%t8@9?4}e~eB1lZo88*-1-B;8JpXpG8_ntc?)5
z(YNMibi)5kWOJv10CP-HOk{JXAs!+`2`z|+@;eRJj6o8)NC!`FvitR7;)Sh2$ZYUG
zH8#M6{`|fVvPju7c$&WQgv0UXWFmEr>3EddQRhf%4VT(cXYxqG{HUA&@9$D)Bkef1
ze;d?U4QkjiXPYW6Y-)Sv9Nc+oBXhoh+vGKP65erOMGTM@fD~(8u(vp{qNZ_6E1Q7S
zorT3C^RpN2bgtt|C3vCqi(Nn*(2e74FZoB}hz}Vzf=ILv8P`!ZLq9R@azO+Keaue1
zl;4o;K(0D$z8!UT!j5|TKEJ!OPK6p&e+q<5^1JsR!CIox^x^VH`|G>>@nihBNwg8P
z@XD_5cJ((!QgVCktKDvUv6}%wJLi+L`&NAgcvg$&H+M9@#nmO{^$p%jjtOKrUEVUO
z75z{y_n=GY`L&~+BL$2HI9^ULYC7!mJ;BcH9z4XucrZgw-%oP`W)W#O8j47ge=vZ$
zgJm(k+oWj>ohq+4ZY!Vyq18!y^|nG;>eUIEZzp8__UW&O2fSmVYJXz>>!KDCxf;~I
ziD)6-2}KeyB2+A}K3mWdiOQ)_i`jMD?3~!c*`d0cOnEZY7FOCQr{>sOX0YO84%_w>
zAx~}ZK#+L$Xeq1$Me1O0ojS+Ee@CF+`$4wZ5D*h4?uFJS1wsT444>IL5dh_O3#XZF
zUQvqo!2_FFJxWb(z%FD!;ri+QlqqLcVGOOR9$2R!={T29Wj+<|zI2gWo(^j3H>~=g
z{OlH*3j57jdpp+=*4tL4Q-%A^r$V6kYUa*&wQJ+0RXtd%v&rkIj}*+@fAOU0pszV8
zUKs7|m@a@F`shxjH~KKNwIVg2>Er5v6yn_=APD3T?<5b(P6=c(kq^K~kf8(<Zvdv(
zhNQZaVOkt<D)HaRJ&b}%#+s==G6&QsPyBH;Mton}B0Gq{?oO%eDzBT_BDs@oPJyo5
z9#;3}On*Ie$l*-yTv>iRf8?<G9RHW_g5WngVDsKFMj{oGfGR<8qGxdL9ePKIWEf?C
zPEAxK9v?HQ-COj;Ch#=Q=9t>*^wQuUwd2~v(7iuPZHh}2W^6e52|FUnmc4mIBqLS9
zxsERC<6y{crzym2V=VrL>wn6f-k{BHc*-}&8=WZ+v{u`klUvSmf6mdz<_3%>&ya5J
z=odH62*^=7r|e7&&BL8os0lU(@Ci7`a4XJHY$NmoqdCSmZPMaDb?07o&SrpkD8Lad
zNkv``%FF%4Z=in@r{{b5n^PQYz11Y`VKlvmyGA|Ds~Pp^Bh1H;-L3z2QkmQBS>Q)f
z&~yHuNRBleX?k`_e^?wn*zTUIVpzPJW$fwmi!#q_wz{A9>ONrUX$Oz;tiwZh0wMaE
zF}LmrL%iwFMJ~>Dv>e|)(9%6X)?(|u;nG{<=+C5RokH@feQe?D0}!;3?tuVBR&=*s
zQ9VRrpip-XKvGpWa*>%uE^?cT$1Ggp!=@ZeQ7H2IUSwYmfA(n<Ex_JO?nGV{3=_Li
z4kn%X@Q4CpJ=l|a$ALD-!32SWQ71<(RjTZKad#FsoW#lUjpy%!#W-1{CvmdKl-*kX
z8n(#2IC-8|jgws;;d+cAPR_DMWKHwa-`iz07yp}=aO<XqvA^}EMv^@09esZ+KjaZ+
zdF^~ae)KumQMU9l=K|fR3C*)L9d|_vnwR}OnI`+nYdfWndW@d`swn=@TbShJD-FCJ
zTF|BgUb+2sIbddB0?7*J|Fecar_16c^}q0{pP#?J`yVp#$cC3;k{lBQF*G%o2^b#Q
z0X?_m7#^1ef3VM)l!UDbw`JU3y#i;78Qr9{$?BxiF^?kLm->omHKsGyky~>=+WV!f
zEA_mPb<qf{G;%B^`fP()qFHRX=Z8<VSOZWOI;+m9T_;*kW92K-v-OwF{xxqWX<w;<
zs)1ttSU$S2Th@JPZnGRWzO-5Xd+596y!Nh1Y&X=Pe>I}fPBhzo4fUxS79*-_5&>80
zaM7F3_1lcE8sB(}>k^1Uy{(=Com`_d<GjVIj3d?06kl?z)n?tp#OHnWI|a$2hs2Q7
z?;!1TpwpxEJLN`;;l+K5f$j$H8>*Lq`=+CsOrHRItmI<-7Vjj;9jHV;aW^v&SExM$
z?)bjre@W(bZJi0<KQ}e!QsaonG=Vud&pdT4-Kjh)-+IQI`v)pV6isdfR7<*e3MSN0
zcV<qfuq#_G@?C~K7+BF$Ki?5lCKwjr^g=0qo|PBmVInx%ds<MdxXLxW$aq?tNGHW1
z)I@5@98jYY&f^+VTNc2R>53MIv;j{VJnq2Rf705PuZVI4KIit*Kse`9%d58@$laXz
z;2~i&ye9k|@B{<XczG*N=&7xGg5q|+7ZA?HV#t$R!#*qM2#kA2GzYc<J9wwLvisyH
zrR^kpnn!{YZcHey#}U=scuB4D*i{fI1=C>Cl#|E({#U%3P+W*G3X6Y^_TEvVubxci
ze|?^xw|HmQ8?1A&RIQX@!j7j?HTp|AcPy^o`2tEx;hovcL<$c<ZIYQ_E#B&qnNXMp
z_>R+g_&k&BaKqrdj3j^kjSv5TAAcDlCOZ5TyS~?H5Cr@?rami#Aw)&O0*)x|5@9|_
z79b}{WJeNsfeP}=>q%_mOa}PH`O9Qsf4PI9bT6iN0(Jrg#C2;Qqo?*cKtkJLx6Hl^
z_L)BkS3cFAX13=TLq}mdt0`*SFYk&T9k{p%AEZaGm~7z5ChzZ~Y}IQ*9=2&@H|HxF
zf$cfiTPp-Zlmnh6lj8-6sTqM1yc5-o)J9e8WqCyt2Hv^&AXQkf5g#V1Fb7q!f1BqO
zU1*R+%Ma3p0ZPrrm4$A_=kuk#RdwuUdqo{^WaVCckUH@0?FetodP%oglPkyDG_zZR
z70pP<Eu{V+%}7YGguLi9$2GHCf;G)ZtkxY#Caxy2Eq}tyIVig)f1-)n2J{ZkKqibA
z978JFjdg}V>j*MFNg(>?DH4_*e_FK9r(8DGE^PB0L`$Kqv+xeZk7!`2xY)Te7N5gs
z$5!1xrI0E$?Ht>Y!aUB7wS$6GbkAdqvSk!F;=_Ix`OVHXty#foz{9zPtV=$(ELOU-
zg7ibsQ6sl?b=ulXd4i?jtzx#KGfFjwr0EJWPAm&jBCE&na9JlU2HZTEe-WQunZ<dV
zDxGcc{{IWS*J`Fx%xVu+f@@~Ezl`^02*~gR2E~(U*YY*H^*lIK4wV8j59wrAAX#>%
zEr3x0W(ehOSxX62&Z&7`#qntSG(8;WD@p=V1|`%j@$C$e5Xws!x5Yatj`sc_pu50J
zWVQvcTXG+-Mo`{92=Fc|e}eS^u&{$LU%qHasBHke&$|Hcs;hp!1Aw;_5<Qm^TtK6Q
zV^SE}d-|-!KD;|4RCH+$;@u`N2U!)BZ~^azn8(p<$bv4oHDq-v^!FW_P;}MZ+gOAm
z1Seu&Vdo`25Q?hbXvzjtO51uKey0TwLUrBV4_vU827!;slnWMGe^$Hiz2oVhQduCB
z1l|~USw_~qffM+I+Z|8t$e@@DdNTj~yoyVvxSA~Y{^vjx-q)34quv#I#ANzrN8A;o
z1?(fr-lCw8uWu8IhYJcjmNf(Vl#>Gmg%UOi#m^UkebPHV_`49~xMJGX$Mnp~zRdLo
zp#OSX2$C7$yD+Bve_L+(PR6D5rzg^0>**eqp1VXQ$psKMUb9QmKGf`cPT(t+(*cum
zA~0VKIsJh6cM!_oPB<q+PJbsMbF`kHFDB=CNjx0Vnp-}G9>=!l<oUsmo*;i}-TnF<
zH9ud`6J5%W^VJL@(rU2Dh7rblZYE;^y$d{u_Kysl`%G}hf7ak5a<fS^#cjL=MMRsh
zeEkNILVizoA;(urGVb-<3ZJV&w%TPSiYbn<JyAspnpa5#VFJ&++w|)2z8DorXX|eO
zg{wFKJ5)hzR4Uf}wVhFFo8j?u6NU~reLj(c)Z^pj^(HY<`79zp-k;x2{{ucR5djKi
zZe(+Ga%Ev{mur$72>~&efa*OImvOZ~D3>%P9+?6Dx6UOV>ji%yJPMl>VUyPnjpvZC
z3Ed~eb0|T8zbl?QOk&^66xn~JJ}I74giU5Vt4P=cJw9=L*utsv^W0cJp^U=2qNBqk
z_9<f|ONQbnA(bMe(pEL41lI{ZZ5$t>r@e=q*47cCCJs2@y=Gz(dZQK(2^m^0#X|!s
zDpMz!#+DVPx$1w2Uge`ptkCTFadD6(!=su~F7R|$cY}~3a~!<hmxDf>yacTUsWux#
z=EAqFJPjxr%L>$tkJ6`ARO5G~PO;(vXPlMgPV_})yi60p<+N%%If7<I56Gw%fhe7-
zBkBXkv2I}Va53j)IhEJ@=ZBW`4jRhOGa24(uVCwNl#PD^`n2Bx29sUn6kk>k23hL|
zW(OKc0Y)~|HA0KE3bD#Z=VMYiW)aJI{=zYnl1h**86?(hr<^%$_r>9ChVvp2*Vu?l
zhp_ojleONig31+Ub2U&C&RmAFk!9w@1Pti$0OOXK@p|F}I@vl#aPa`T*GfKov!)X=
z(84<4lox-_uZqN2YjzOdFwK*pAAR`o)aNwQrUbW`#)iME1aVw!FZbm|+d!_|Fld7>
z58p-80XH_$v~dcq)o3?Q<PWC=v!8Y@AxWN}w4C6){Z&BzGVrzEu`44`E9BD4^>oeX
zS_!9tkA7agOD(~a;70ij8+|Z`^IB=VP~*vk9z1`T(_3*$aBC}0Tc@q`c{^iJrRk_3
z*L8`pXgB_yDnz1@gi(!PK+VsQym;~sc<X~aY?MMffq<V>T6TngL$CcSs*7iI?T|mN
zY9aZuM%O+pE0<dKPo=OF&vOcQY`g4nGK%mvzwfLSV4D>@E_c}bOsdHpax0+>?SJ!G
zVPk(m!09Lvt+5Wrmku6N)#;#lvuE>lDm=egssGNDRx7=iL&nH5?V8m*K%cWPK$8zJ
zR#KV~v(f@y&llj74I@!#=Qav(wt~_?mT|1y*LFA<KM7S+dQEQnKzU_GGe!`r{0@x8
z-NjUn%D!4PHzm3lyA>@%d8S;q_gmQfVG)16i(fMBaF3&73BN4o;F6=0A(D+gb#xAm
z?t?IVHS+bm%bIKM*lOl^i4?w~#K-Y6{`XMK2(0q*$Sy_*0u@sayn{ceP#9-8!_UVE
z82=^6HxhhVu%|Np-xkw$eLCdHIlX;@M*xN}x2tc*2PT~J9vew7>sl-q(OfmA!E0EB
zEC(!LHqqpc8WScKumEaGL>{iP<|cMJfXunFWC`cdyTNIW8xJWo@6RUL<o2OkDNJUa
zZ}m(onH;G6&HseKS$5bCY^9tX>BXtqT7G~1c=|tPX0`yAVUipZ0Wp{1rvVd}4ud8t
zm)$QOn1At)zrOu^;E7OypQ8DEemnee=t<SaA*pKk`hK7vKnbRt9E6CUfBrsvJA5f9
zGHT8oeZ#-6GdPVk%t=c)fBBUU|G*!AMquUL@ZSLXUe@8mkKymQz!GA|ctEU>Sbq&M
zA0+e6C$trXR-SS2+fl5f%t&V3b*zFjuY0t?ntz-?YvRUeJ+da$fz%3dYvvJowkEU$
zjg|mtK^iRsAidK#T8l<I5|XQ9G@>wc%HWR=hD-WIr(bAP;IB4J0{ltQkc!XpB>aUR
zqp1c$R5ZaxSR=D5YS=5U1vO~FtP~%t2hdD9hsZVK+-E%qC?Z2y;48T>K5G1s2>Ek*
zM}Makj%%87O5nbEP93-k;O5w1H|e{L-$&qjrR;YLX+ikFnKklGP5{8T&azj}#U`JU
zE&6Di;xqk{qbS()WT;b~7d+-n0?Z>|l4aXGB8uf}in`BH^uY5dFI0&01X9mJDsIv@
zXGNJkXA7I2m5Sf3pasQC&m32G@;ap$_kSgql1^H$vq^MMNfdKlr>oaxH_{2*Ty)Zj
z*FB|aQzu;~{M|xY5WaBC0o=*!q+~+Lzu4{CXzEtGplxS}y{>3hYIC>L7qm$wl+$-|
zCDf|0s<@K5oWRb@u5?b|=SA?5D?xH2h#t7hm39m1y709(%wMltZ*V2fnH2Dsx_{EV
ziLaFMerYU7*~N19E+(#>M`@QjLD*`5c13-rs(G&m57^Zz@O6UUEuaO#+gQKe#k{q3
zOnOtyyV;=NziPhiW~G>Oh}Nex8qE5#k%tY4YOwH#6<!J^6QSTGyTfR6v`z=RWwoRO
z$+)~_1G^Q&x-Gl50=Mj9&z8OK4S%fYj}R|pG<iqsA8m+VjggS@f&y`jzfhmVW+QRA
zYeMlZ7?j8(LG3!J95Ti-yD_E;f615<#2RBlyC>1&$7}aC+j9Z$nNiU9TqOSLb#A$I
zSl4(WDp$jxg<;NS+0JrEpPy53LT>pvjt4%I;}XC8DeyO0fMa|Tr6Q*?3xDKQSw<-V
zqA7p={62Iee|xPfX%$SbTS@Y+zm=ekV4m9vUA|fpX%4l%nGUD0VDdd?A?E3DmVuhc
zCrI~Pq&ctGRlPb;oEe<f*$t*X++6Hy)qzD0)g4-OU_A41Q>$(>ij|aU$NZhLAQ;J{
zb#=2EOY7`VAtR=D3Uskhqko*GQYjotIGdb_0@*W#$7IB^=FhNs#ZSXS!MG}??Ph#I
zo`9d~qkC>bGmSnC8jVlXo?9uiR<gXCwgsLQ(ro3Ocm~8|YMXH*%ezTiVi_2B>L!QY
ziehC9Qwq0(@Da7kwNyS`SXs}&%RA+9h5IinuP5?cD5Yotar3u1&3~l?+O2pg-CCS4
z%!IzSP|DFvrg4|bGL@oQD%0ASnA2n|B_2Z|e}UZ#b&0}#ldb0@R5{+MF6mZUs7rdf
z?x#qJzG}JYSG-WKNFm86Ct|y`Oi`#0t>Gif6s3_=uRp#_Q9xZ;drMT?jAA8aiaLyU
zI#N_XeO>+kS*Pshe1Acr6qudAPtsP3daGt{p|YR01)gy;=+&KAmekR7yII~#+7ibK
zcMakukKT%6We`KO7UYNZShO1))7~02*JF_uccWajcTn2C3Z$C@p@;JP_@E$kJF?TK
zPD$(t1br^`bfiY%aXV@No||2_)Q%WqK=M%-gF+^ojJDZU0)Gl;g8AzG2OqmT67+R$
zCb`-RrjfC3Ep5i(uDxRs*T@+>+?JRPoe6`b87AGLfO3PPX_K0Q(>ANhieIO4s<cT*
z)DZ5_p)%w4YQhQSTfP*M>JYkjgd!{_5e3dQ7IMxjz~b$b*snu-Yx;FbANo;>gqgr*
zOGZ1c?Xq<N27e-54kAaLJgbgEopT~enkx8nCop#oEga{H(n#5~Fpc<gvdi<~)abC6
zL|KJo9%t81Wj)OSJ2&TgzcFe{hg)Cr++FVRm6p&kb=jlaz8-Qaf~8PbG<03*HW_K2
zE(<9&ONSp@X(z*EADQP26D}ap5#ur5dJ8w&=+@gg(0@0pWu+OfhX{13NV^&B<eqT2
z<W%KKBQnj4+~6n0HePlU4W__V)`QDXLGuv%qgV<jjKvd0wW_T1>}%Q<99Ls)l21I$
zI3+uNYUCJkv_0^k<ub=*45GEmNM|`QEWc+9J&kxtmet^CDS62f>5pL4^nk_JTVe_S
z(+IDw_kYC#EhRFQeaXZMq+$ZZB$H&kLQZ=BL1H?0_GM%DHRcpJhtVQy<0;x_Tj)lp
z!gX0g9w%=rGEt>wCKF{my2kr}BS-N%|J1&Xc`tOO^W$jKWo8p|^Nqf_L%78Csh0Gi
zDu_6GPp@m55lm|nBix-G&1#)4C%EGKBTtoCGk-Q&bd6|bltz3xwdqcJHspsee-abp
zc@tzh+)RE<qbu@bvU{wGdm#u7%!sgc{$s!*<#h;^${|#V8q35mj9WMSZ=lOuU<H+s
z+MwjHpgMf|_xEo<ef>=^oc13A#r-!HSpsX!_w=f;hLhT2g&E%sL$W;5z&uw525?NE
zNq^chu?&124uJa<+&H>s^aqW41`ecbCWlJa5@24S8Ok&e#yXnT`_T>-s>%|9=t2g)
z5tcNUUU7b&MX&U<?->+l{c|{c8xCKRRJAe2QXT3r58<b=bi-#xs)B*AO0<&AL((uV
z0X3MU;w@Uih0&JZ3d&HGppZjq<p}&%R(~rz=aGi*JcPJ1hCEbA>&7q0q2m=tYWE9w
z8)8s^yO^+V6G^&J!u9*t_cU+ZEXsHu1O|A}2ERCuwi{X}c}Fw1!${y6y3Gqu<;rot
zMQXjPLx9VD85(jTUk{bBFRO)tK`XVo7KG7nI2s`s!$iF9aqZ|l^N^tH(Yeycl7H%Q
zFq}<(wmo9^1)|KDyGUMxYdvINE^qRYz3UQCEKjZ+$E#5=-n((6#sr66RBFCzF)H1E
zE26REY&jYeZ6;b#=gsXk0jpeUpm+3jUD`SOt(G!pztt9_{r+nhgQXnv&?K1^cQu5f
z|CyuMTS@x1EnIwWZ`Utv%$3r*Y=5&aCGk`gK`y8ErS|n(zouik@q&i|rsSjnZZKh6
zdGooD5Wt|G4~{o1=?GzRnPbOMhh+xcs$1>iQ3qM3?4i}%rIPbxgSPjWqg(CzF-K3@
z!o8TVi_S+Op)3%TI(Z^(G}4MKD~Y4a+S;>H^*)MDJJ?{F#o*?}3LbYt6Mt+Z!gXqo
zZrEOj-Y)DUzA*^`Mg(s|Xqgmxg7Ndb_9WIT#`n<Bs!gVoL%y9kP3IvKZHbAYZ$gUh
z&&f7A2Gu-+#xZ;*I?gVOymK-#N_oll+7Vp$(~IYG0Yh4Y0mn|btNn_Wo=agzV$n1;
z=o?(aOQ!3o!Hu3VMz+f_=zo&*PN(Qt5f|g5l$u%DlS|?820=Q?_kJYEoWnlPw5|PR
zVWwm4F~WUOMm3dG94#FSS;NK_QG>*?T=c}9_Pb~F`3~QO=%YQe2sH-Q=j~i1FYi`;
zo?kUN?g&_gGt~K;4;GE35T~jm<pYRKs=Rjo-+yBD^FRL_J;VxSZkP3kMGpcrF_!_3
zI24x#J089P=eNZ>9vlb*0lAo$u^0*y0yHs~fuSiBm&TbGDVH`x9w&cn$q(5s+j12!
zi)vTdG(}Q;9=mn(N28nXFW0Y^pFRyWPI(50&DZY@$OSEbLCZ<YjRR2DK=bvZ`Q=S(
zwDP&ihqN|EK2E-SIr8PzfENmos{of*NAT9s;ir7wg$5&?e7~jpgi=^&W98mpL!<|r
zKG(~yUq8R-Bq+!cM#q1Tub2P5l%iTg%6QxS_Twd9AOrFf8)9WKG=Ki5`TgY|J%`L$
zD<8kf-|tH-omC!PWckaly7{+U{8@-i@XcQ>>5qOH-Tb%tmn^Ww*el&KHe%LrkFjo0
z1s@XIfd!kOG=<|Vc1mVZS%04^Xcf3*+d7j6aVBjo^Te6NiI9Jd@MaZ+d2%KYX8>^k
z5(lEs?#Ssk%LR6qSF592n2@M!JS!h@B^FZt5H||l-SHezcOB#$$Wh2OeCu3kr1ivo
zdVAI?zF|)W<q#fiP*$v+FlVK8=nZOO?8W0wm8)gj(3!My7N4Xuk<#3&Ob1o27Hvai
z<CJseR6rkPu`_>%30rnllf&l8GT3${Ih4QV3$sVXzD&m`7X#;XTyWgu1)?WY!yzaS
zHoHNa5KI^RbW3T3Qd&vLqD?C<xd{oo{?DI(G{u>Jejn7apn~{au_EfCLM3V^_X|}i
zpG^rWCaRpwFrAOzImyq7j;AS6OReN<x(6>YC$6chZ>@h|GlO|AN7oB<%goj|9XAx4
z)~kw@)(n-+TFCXhjYJ7mr){~SG%HK;Az*aMnc32^OK!WXOJOh)?eNGhg{2CR$H#Xm
z%&0KXnWT1@#ZJjc^fCOLHif0aSTz5iZA!i78x}Rf#M$T6Y&O!XTB_|+>Sf!|8G<Bh
z_avPWCEb5;wa(SDZRu<kp+6PVM_KGlW5O0OJfcx)w?}#puQ)hsmIhmLK{5BOa~w)z
zFvmcT^`{MV>Oz^k45$=`+qNRP6#6^_a(h0EDf<KwuU=EwcQtEk<R+<%VEgnF$7XKV
z6W$o!hPv3xIZH@|WRZi6<T9|Q3QHmL%JQ|658Quy9WSpju`TM^-M_47=5<Z?+PpeZ
ze`nk7ToB%tTOLV>uSE7Ooe6{0r3hzT#$Cc7(|h{W(LhSdek@s+o_m^Ed$w+$dCRu0
zgZ_*w;2_haGt2u;K}m1Yw-`F@x9v)cVfQwyLV0E1Y2L2ZQ+0b6f3@BkrqZ#FS8URu
zEmD7DAGV#Q5xkKK_|}0mt?sw9gTorOU&CLZmG7GTG5eq$drPu}b=j97OYJ98x7Tf^
z>92FLfBS2)Bi$XTX>Rw_jdlQra!enS)nJ|5#25TR6m|ENJ8#pPw)=Eej5}}MFP)S=
zTYseCqCd#M7Tb{I594T`>}K}U&i>;p0senXmVj*YHqJWkezc>}s_&U@?3bX|C|zmk
z?%v&zlz9*j*`|xAtj~mJF=(azShTfNS%ZBM2p5OhbeuylId7-L`&fZ4ghqjx&AGZQ
zV6Hgj-5Sjf+d=P>`}dh8r`XN}eq`D-I`!S<`T)N;tr)UWLQOfXSgo8Cud_qE33h+n
zBDLPy){oq5AD=;u6#=2-yf%YMqc2TLL2PDXbn|z_up^coac}%ME}D0W%uMN(@e=|l
z>-|i&rSUi(zYXP4=NZ*H5dusDN+;PSXU4MS4e_iT4H$<>K}4xk+>%H2Ubte`pv1H#
zw-&=#vh8P{pJEX)6_(Amnyv>lk?wyOM^=H%jN`Xus6dNRgDa_NyhoVu8+{zMz@UPg
zX@@X#>qA^grW1mhhGtnqNdrNmyWG(npf<<?I<A78JZ*eN*8S--=%rME9H*tZ=iU0*
zdMwXcLP6`5d}i{Uhy+K<dH1MDm~z{U1fy+z6cm!fYWhP>iWtf}V*Wd0IctBEM?bM#
zu3X=y_{eCSShQXO6bp{qdBo_U0qM?SZo6>E3x{G56p2fhlDyZ)Uyka6<4T4xUXkoj
zLCx3Wn&e0>*8NysH%I4$+O&YxGR$lkrE%!T>WpdAZD=}I_wO{eipD5}*&_!ZWb>#F
zTQw1e3G{eu)qr#=+MdB-4zqvQDVb(v$UO&JHL{2`lU^$iWaA#7d$1?fGR3a=XZi*%
zf9olEk#fkP8%SHdzW42jA-<0go?}N0PBL*{>C9-9nr7MFt{vd9%OdrXZ&<_|X@Q>u
zx1yIu8HNWnGdB_=cn7^3?Sx^gMc%N4Lq+aMZv~)}E{Ofi6nxSY)iQr?*usKhc+%Sf
zXl2A3>Q>Tuti4Vz5t#hhX^m>BH>|;0F+VFvq@(f#cY|b7tJ^&C-h;vWEK{w)hGm35
zZ#_z^wv5a}-s9bvId7S24Yn*}mDPNvmpIH~U;7l<()||Y&@-L;Pfk$Fz%V>G=>f-K
z5*R_FFv8t3Qs&0EUHgBKWB@}afcqj8`UrsYHUVL&{}vLi2gUS+W8D@I%B4pIFag~=
zNpavky8GJu_W<%%B8ec4v?VHga|g;9IE^HN*HZm?^199DYc&;bxJc6_nXOoD^$i!R
zz+y@LBzQ5u-BPS}|LvAiwCAFWV1x7}k3ts<ZtYaL_id^P;z)n#@lF5g(X9%FL_-P<
zmU5jyA@kH<a$La(5~5d7FQbs>_L0Ta!lSUUf?MT2R_<8}gQ*B-whWv?38<3e%9bTq
zsi2;1*{u#Z<~S<OcIzZ=Ed8{OzELQ+f*U7Xw%p?rzDH6{%miKRV|P^1^Mh?##*t?U
zeaf}Y2jl-bRO5e(9U555aUTsVsF!ouaar=76|$@#gigk8H<P8e1s`8O81WU_X$zxK
z89m#<3cXGAFd66yPVGBepQ}aQ#P^wJ%hSn1DP6O7;eil(!LiQ4H(>ROt<EK@knRMm
zo^oEna3zj7R?1LP$8@`bZUeFkxCw6Wwd`rY+jb9dik?$i4k$iZg?9&3O#FDb+XrRX
zotaRNa@Z%A^TOgu7_E;F<m@`sEfJHU?Ar$DZD*7`>IpVrhiV<S{)bHwJ;!oR#u9YT
z0k><#p2bmd{2K#Af-9F{k{lBOF_#fu9}@#JF)^1BU_vK<8@Y~SxiRmr&<_mF;!R-C
zK&y`dHuJc%#YStd6Hnkx{QEAFMLu5dk~ABrB|VYFx=%L0{QJT$-=40YPk;N6>SCGG
zV8!M0j|<=r9RC8xhels405cLUpFb|Ye&t*kd|mOSIF}M%7C$?D*Uw)hI9vhvE5P#?
ztKe%|CqD3h^@3|eScjiqX&<gc5L_DE>x~cV!O-jV`M1wso;V>2_zu<k`{&btpL$e{
zFwB9v{Ql!9{GcQ->BL1#CN6*e>+;9b-^>0CmopbW;ool!K*x9hXbisF0Pxq}_~oDY
z<1e^adUyF>T>2v{=gez=`OoDa2%p!Cb%2cJ#%SGtFcvh^+#+z#WcVpDRwFYMGDsou
z8s9Z?Mh8Z{BQubRia&obJhCW$5|LT?)8=lScZZeyz3dsFix#>bR_Y25XRAGNbre{|
z&_+mFdd8K%gQq*P(wgDtpCl{o5R%EV3iSR}@>+K9$?K4m%t0yL%*`DsDa8Qc{~ReP
z%apW##iC}?a25(b<UwBCjPnjJq-5aXC-FiWrmP!yq3`6y%{cGzf-}LzHeA8H%V1}y
zodmIm83h#t%eTSiK*Kx0eXuzYZM8Sdi=x6IN~+*0P<%Q>Ny+i4^5iu1GngRU;Epf`
zD~z)Wm0erf5!RrHrSM;C7{MuNL+kOWdP%2$#C;z@S;|Qf638q14&~yBBJ7TzA?*mU
zUYYPzzC3Fie=TQT2_sKW<Jl1tNX{>w=jf#PUd~o!ed8nx1;Ks#yjo{Yl~eS!{N@P@
zP4otbpoxB$c)>rS0gzb@@=?Mu$9*rKxf>@)b4e?08Cd8c;rWQI{7RTY=0v1~a3ua^
zY~iUnFp%N_cxoxwNN^He+&HMNQ_S0vC^iCts1`_KRvr{wd=?dmSBjDePrn)xn9Xop
zE%@|pJSP(g2|`#AMXwgsi_c6{-YG2xU90rP3<U`RGa_MRHbJ+jac^>IRXjzMZ<kR}
z9!`IMQI&2CJ2#EnqtSrtrGAUi2w+;-0aE9HFlU=;9VT7e$zp4Rn+kNfhFfmnk6YzW
zWjkBk0Eh9^_Q;?(wg?BWna3#Eli-v32s%2c>kMfUdZj=y$<C&wNMRR3HjN^ypY8l9
z?ajWlaulkltZP)()eeQzxsI@M>mms7BI__!a(uMTElw?-`Z9r#5;37&NGX^KnTgc$
z@@L3VhNs`Zt*f;YOp8G#Q3+jzp;^9{PEj5>EU#z6(Hlu0Yt0~TeEZgYn|3}FX8nen
zzVi`_0j#yD`O$$b=`Y7USTtPXr<bx(9#a7mmj+TELIM7lNm3p<f8TTTNc29J&^IIb
zxY&hFVWKCQhu4%m;WWcnzazGHb4&O&wRThcmvGwmGpkM}i^Uogo$W(eVU9lTT(Ar?
z7(Y9R{MssIvN>fmv*4S!7LI;9Z(~_AT8Jx*28;W_te;=yF<F0Vk2s`(Gh0a>FzQYz
zgM0e5Vs=sE^Ew7vf5whU3uuk%%%k?yF{eF^g~Qaq_T;R&TNj#%s$2X(cPlXU5ukf)
z=H6CdVWG?@339B9<c?TbFI~)-9M6Tj<nIjzH;K7{xGU#&^)317DYlD{|8$40C_#}?
zyZVt@fN86r;IpCz8n6TyIJOmYCq|u}sT9m9QHXlI-%V34f2dGfT~!6TQh~LaHA_bi
zP5LC(*B&EKm+uORZW7MU`Us@MykmRfvDW9^Zf#49mh!I7_E`uA@xpXZnum`cf`Uo}
zj^S1L$K>oeYYb+Wx&TY9ReHK6ui9(P(Xi}=p2li!GV<h0Y7yIHWa=^hdObEQxt+s%
zl0!YT_o>ube-H<a2*HAtf6=f+O5}%Mh9$Ycf5Fl_4kW0Jv)MKX4ABED{<70-KsCqm
zq1&}|N&^P`Yuh`G*^8wSM#7K!c<EbFN4GQxa+^t-(n6y?bhQVlStgeePzsuPmC#G)
zJvX~bAz}(F+qK1VyGv9}tFBCm?UmI3)wwZNu<eIee}(2!$hl!{!C9Fe-fA0eHu74y
zvz)x${hoDXr7#SZ5P7sSZh%LwFZ_%+mN5=u<SPBdZ+t%Hc|;pfh!B)}Qhq`gP&iBy
zj2uqNB60k9OI~u%1XlyB^L;_7c<fK~kdDm+!@Z<lJsn?Zv6E2uyPhHQvgle{oO0&9
zIynrKe~dgSO76<kw8M~$|CL5->DMaq24g54Niw8GrOOVnI{K5K7}iVk(H}olu{JAr
zdd3)JuxuYrJ(f#saX3_2fQR;wyIg@Vt&QK1`z^HLD()8-zeSI50)xqT>h{D%I4m5b
z>vzS$T*Rx~9q1fYb$cMrWnxz#_PQIF1bO~nf60B-#*AXz4!G<DQE3bla+ISj9PUPi
z^fU*}^sT!4P&7(b_ZWarHL~sWlLe(;bmddlCM)Qj{D3)7!z*|wP}_q-OyvY0EN78J
zI}9%tE9UmcQG;abaosS3$iXyCL#kk-&9=R1$E<+il&blLg&%sa{j>@Brd<!H3_;A2
zfBz&`RFK_w3Ifwg?qv@<=0$Tvb5Ze@Q~raZ6J9W_%y80;qnnV_dp(YB9E6&0b8@p;
zhowqfSkDYB4I{ZkyqvH^?T6C(<P7uo*c*?3V!F{h85>C5A)K!GPIK8RM+3pbkUrcF
zPHB7bRaqdq(q*y*+!W^g(h)K8$R_+He<oxvgh=R2>uG~UhPE&a6g{prB8+23yJk$5
zobjQqHMMSfDDU^<Ck^Ty43J-kj{1@A*4{|3JB2Lno|a<|ab;*lUH@)%o>a)|GPwvE
z7T$wLC>PcpU0_$$ltZM-Q<`a$O@tG@f8aQ4I6)a^mvQ7^O_VW@W^sN#m?nPhe}QYw
zU231Viyt?+Xy>*ooRJ#S?6FOH!)TNoT?=rH^sJKI@edea(1f;wyf(qe@@iK{KCaJS
zg+NlpB+B2+v=-nV)4JmyG3;;Frm?ABzcsW>sFSj+a|$Fz@a0~@5Rx8mdF_dNrq`~H
z75<XmoElPajQrmA5GmWc#cd1Be|ju81H`w@F-*_#+YMSs6_dCA{P+3!3hQ+R!}+({
zhR{0J@*TvrF_iYqYCoLc&Y=hW#01szB{G!k=VnW80q$9HyE<Cp&ohytQ0=L}w-u^m
z(c$2E*G-EaN3n9t0xKE`{pyoE?3DsMFt7AxK_E7I(RJ_lO75{Z%~Skae_ZWhit;Ye
zsPkj;^UNQIMV2#l{!5kT{&lr<o2!R#huKlK+ChYmm2tXZU<L}}4ICZv)0C@*ax3Px
zl+OlEuixBRa$otHQZA9|sPzAzv&$mk(h9Hs_29uJCxyc9ln}k;k?U1->6SEaIs~kn
zMf7BrwNq=AAk}$_o7l=qf4Lc0rtCn?1N51MfsuooQhr2}(p8J4zw2@0@<=uq<ru)u
zsDH1X1!);*s8bbvSJK%5>_XM48-WzgdWiDq&EC0lr;MtDx?1{WMU&jzaK$8LFFggY
ze!-8jlWxa7`3@G``0639<vSE8`}1$VcPB#UaUf_d-SaLE$sb3)e+=Cp5If|bF+T}9
zm{SMwp37SjJd<KjlRJ`p8LV3l0OvWr3=JshrWSGjYPAfFCO%>KYDy>#q<(BC^rjBH
z>rOF*xt$k{2gxDHm)`!u=o3LP`>6H=POYlwN#+4?2z9;dZLx`*)-^Wy$wny3cNcr?
zj)e#zG_yhtyRS8_A(dQ?b1h%J8@P7bDEFGa;r}p%yzM?*|Nj2@^ndZdsDPJYk{lBP
zF*ujuG6oZqT_-4)95V(9m+)C0kbm;^w~wD+cp?<&4$b}j<K=%ZBdJPTB2`ZR`u^fy
zpadqZm;mDa&;Lxnzx<=1$f!B9{ullIwt!O`&8#pU=dZu<>EHD7X9AYaPJac^?`0i6
z{df8&EwF{yG9D01BG#=T=96H~x`ei(P|`7u?tT<IDbs>!yN(q&b818zw13G0X%p94
zsZN_n20}@~t(YU^S(```6iFLNgA{27VtV%Dq%Df{2n3g>s6=7tErY*)(Omd9D!uWe
zKyOW$g!D@KAsMgjN%V%FqN)Z$k_tA=>eQ}?;V8d0#Grr~0S}e~DkiKY<ce|bvK#;l
z%ZL~B=uOxgVSjlR^7r&iMt=*ID?jBefxG5<8^C1%*QW-%PQOk3-GQr}yg$sO4dx?f
zMvEsI0V3ln^IpCeoxCJnbWu0qtACTdD4KL-sJA>XSj^coFpr>4nr-q36wO!O>%K?U
zBhRC}Pyy!|q*{cOT&G{wK$*Ry3wK>R1%FsT8-f>(SuUSsby70!N`EZ%GHI>OB-ULL
zE0(NIRjbQpq!PHkX{0l&yQFDzBi+aR!%W&>K3HZEpJa7HFp%O0n_cTo?QRn^?F=|-
zih8Fs4|9D(nm&cH>Pe<VQUxQ6De2n*?7VGC=LCM<1n*1<lNn*Oz(b~Vm`Ur*S57mx
zp0|F&lsIPs@Q0exvVV&2<nnQDY;f7eboMDKuB;<&mplPXRY0>NKa*9x_k#!Qau@hM
z#vf+T2IEc4U!S7h%2+0xF6!MBDD<zYZ(FPsbq?P8EsX}XzOCe;0FezA9<k!JVA2r^
zuh|^B;%J`?4)bbD27+<%nF97+4EsuUWq?X{I8w5Yy@4J55r4lIv_)Bjdo{s-HA>=z
z7X(OS{ObE8CM$`}T@&x`f<i$a31U}C<d`v**^M$)_*=@9AXb$rc<ZQ8r2C$|bo(|u
zE%sf|Ip)l)?;_}}Q@QOjVBKSlBw&RTF`fxtS_X^T^!KmHL(!J`IgW?TChPT^=9jR+
z*$P}Ufu$JsxqlXlRfa}3A)zUM{`_MaMt=ELS5h+Q->{P4Lz62(YhaF>8C|ZblNTR}
z{%$&~q^!w}m;sjQu!f<W@h5QiT)4R`*j){~peZow4TV1{FpjnB&`4$!Jic8=YQ&V5
zpYe8gqu5EAa!m6l6-}rG8L_ApYs{~+L4~lGp1lW&*?*c3Tr%auA&)c3n7lW8_dYTy
zv8X{bRIqf{yzxUDf$jLtH#mxxX?%3y0*WazYE)$0vKB5S+1k11!?bPeSsJD(j6G-1
zND~d#ZsVPMK1|!zo{?#1K4sJQqS)!i2rWAMh!kcmn$H)OmR7Owq);v$2@>>bD9`y)
z@)sCuz<;{~T?(PYinjtTd1fT2k-j!x%Kl8GahtL;nfzMH)>^L^0%d`H4f*^U__gi!
ztY;uMa~Wwh)5?@BT%MN83@dHqG9z8j6FudM%y2!fcq7LVAiNsS*!J1fMp6JZU>}*=
zNQDfPdwgyqm3PnZ89#V8ik*~E;NbA2&oB~6k$?X`Gn?c4-r%Po{7OD2%1bGjL^zt+
z9H(t#&p>yFPujDH26uGNhiTi|vtq#2r!4wj6g!<5AjJdygPJqK3Z{%`nln=6=%Zf0
z2;=*m!Q3&7gw_f+Ck!G)jM}dPPRo(o2YDa8g~g#YBTT@ksv|)Q@+_i5iEjf&SDt8{
z(0@R((IJIKjqbaQiK^K;DmRh%INGNxw{}uEg7r{D!U2hErmPOh4cd8AKLC3`gul%R
z&aphDy#)7|;bm@&wQ~mEnKUnVK6If{l|Y@$bfyPNC~vznJ;0b{0H)a+T14rghLRlA
zMx32B;>_^rqBi#ex4rB!HTy1oa7ox~Bd@M?*tLK8Qu8jKg$cOcA{}8|FH_gj5>_!?
zv(c5O71^de;tZc0sci67D%PYWeIJoL-N0{kUX>D(lz<S9vs||7#2*~Az3^zWvAWxS
zN|rg(ASIY>n#7paN1k~Zekt9g79!hV-ACYWoU`arQhOOy15p~w*^k(=T5dpxgL=`z
z3;us=LqX#Y)67mk>OK-5<CBfv)YmfE&VwAXFZ1)HJ5_$7?fHwgawJ;G@&U1gC2HK7
zej#NTD9n;Hc6b9s&4jVz?4mu9Gp+VSKIm$%RXUaTd~E&~`KoN$*TCgPz4DEfdN^~Z
zoq>={pqb5{b<EejBRVEMKYA=p>{z8iQ^9|QdPc*p(t4FH0h)?=Iit&!xDR7Inrl_|
z59YKSfs^OR{zcXAedQRVYW<=38u5hVuCntimqEMz`GfblH)^qAimcUJ=h(dDzgsU{
zXGQ-o_ks3QtMRg*OYp)Oc7c=0@BMUDcJf9hD8*Zfr2a55ujz6=cj3=g5SDblxch%b
z#j{!1HZ_CQGoyQgI#|byf$k&kIbrF-X9>1q2qyiPJnTZs8lAmR=E@Z-2YMl`&y1CC
zeNt$1#JAA+SbWnqKoFP(J?>Ckn~%~9G<iYb*V+;q8zqM70!sD=thuf78G0?11XS#f
z^qP9$H*lt@t0Vtb!gWrPawySVg4Tb_3fGeoqKZ2?-*?BY+sTOvv%1+aTt|1d$ZG_{
zjIcL&J=0|W6&+`a9HNWJ)k{9VAZzy5K;&!<yK@TJ6w2tTDPZ~R-Omx_ooOnj;$;V~
zX*#;Py~s9`p1B9H-#ujLoL|57JIbpjzxD;4?s7KK<vJqeIy(1lQ~gzP9-DvTzBx6S
zPu$-^gSqEk6U{$kVntOnCnHwvf95DMUCi@U)8q<9^j^$ty^>t)1v@fy9ErH4)hK8?
z_z?fxq`vV`&28_IgIg)QM01>{3;sdCSlCiLOHo37{6vMOLP+D-JSM&wBi6sd5Nmx)
zT=_pdb;`SuUsB`5@wG2CAP9fh)Gr}7qEuS+W|vIr@HJ_`HY7ulL)&C;(##!$p-ktE
z%Tx7p!WP|>I%=s#jG`?-Fm`*3i{Zt5AGIH6L%aWR+vyRCWKI6RyM+IV^*G9{Vd8;w
zx9Fs7?9x1x-^QPEfd|svBb2)OjDReawjw`gO}aF@S~T(DP9EfMJyQZZR+nLt91{XD
zIhT=SHWUFim*F-9Czr!x9x@1NQNZrE@5NG={bL>&e}$yKAnBns2L~W5!r|@j!!J)#
zDofXiE-fV@U5@_ubT`kB2%c5|eGBmX=rlYH>(qs=Bdt+!9sPgK`)DPRkZ5UdupzDo
zq7U>P=;vSGeteJ_=yw<64_=tyEY6{{P79)0Q4$wToHbF?uXGW2L(dE$Fof0sJ#&e}
zd-7Xje}vUl{)k%>ZtaV_kw>EGhD)C_{bO9=Xoa!)TU8L1y#^z$XB8z+-$vY*NS!{*
zvOiI@m7gBgRax9^NYaYSu%CDil#fM{gmaAKGF$^+SDuD3@<a(U+G_v8b&$BOkJw~J
ze;G+>I*RAvY0{D(c9AbdJ5B6@URb&pyBwcQf8tKkD4+gwnFeHCx$<pz%AZc%lfu*0
zCt2eB2{Rp=*Ph4r>Zn8wyktq-c>h$s2_8L0pmqIzVw)<b&AXuPGGj!|E9w(wp|;DM
zb4*xu2qIWz^*AQbJK=)vbpPqE&v}cP-4aj)zD4CN&dqJn5(QoT7WjM(n3i6kXI#Gj
zfBXp6e)><mx#3nneLviO_<St|Xr$0*RG)_Ara?aqiLa)a)3TwjDph+cz|ZS+0apOw
zz2dm4``GFl&3%p;lWVRfFIREZKvPz@A^&~)f3X5tM+Ar7bUQ$ktw|JRgaUhuC;((Q
z=$JeuWF__KnyhR3tN)b%m*aKs%jJy#f9X59{(D`nAcYFJN-lDH3rwu!I0|R0$kagf
zn?CPO>(9!Z&Z^4Qc&FMQg6&lG^HsS~DdD`I3O<{{We|?&loVbHn(0K3$?T;?Q!Q*^
zET%p`KPE9-x$4W09F@cK=<`cwgsE8|W0EDUPVhC;TrZ!HW$?{&n^(`E^{pI@e+qv!
z;k5`@(6<R>9MhVlQ<W#-u6b&~rOBk)J5EI&-`PbuN>6zAc47|gN*NqSJ1>x<u<>IJ
zjif2Qam+TEv_nnn35hRFK;Ozt5OnQWqTySE7u(Df$k|%(GBmOjU9p#qw=e9|>)XTs
zroo{FDt-9v?}zwYBT#`;2SfQ%f0Bm(`Tg+6!%v**xU2U-=8#k#Sc;ZpBYiLqDD-SI
zM;5>g!VH1-{AguVCC(SLW@182n23mQu9^6l1=qB!{`u$61AFrIOQVEJkehKz@Y2+n
zMU=t@+AbpvWd^F<ldtj25ag^5G_6I$xu}psRyFsd%ulBgb`n|8Y|(-af5KBeFGbmH
z*FsO}qqNX-QMET@oDxYR2jYoS_hgK}_!zY(-h`@&{@nGhJ_{9~imcUbo~ruPG3%w$
zh~Tl(oA1h3+svRqYJC{>y=`Wx9!tl$VCc)^kcMv4GIjKtGB?-C>s-`JY2~a=bufZX
z`lGYhR-b2-KFH@8mm4F0e@hH#uL*{ECk8fAwxls#vzHE|$;P-%QmKU9<O`GuKu;)_
zS&C_yf0bn|9dD?;WK^NYR98FNPc1fEop_(JG*r%IDZ^?t9-<&1xCx$pJx$Sp)X9k)
zY4W9RvW5ajhYiCi#G{dwYFlMdikf`qZ;-w2pd22XJ}yU>X9?ALe?Bi5prfrH=i1U9
zmH9)8I~ZHE9^j}tBZ&a{Pm-*AGWIg(MIe(fj<GXlH6d4u6fZ-b%=#QRqFKXN=4pA)
z%w#WrgKgs2PxS_UOEar)L@N;eM2-dhQa&S_eoOuK97&2tP*M*`KZ~v;3!_9dxuY4g
zDNk+*BgWR<7(yyGf5v?CX=Chw&=mfR+Cm;v$_PglhONe9wdES^E7U@vn$iB%m}RRM
ze28*_a^KR0I)*VI%Iu@eJ%$KfoCm+WFL}>mLGUzsQr}YYo?{9gFKroL4F4u&sNqpb
z^ySI2;M%4^4d<{RIUO<4;J1{#=Qv6mk;j_rJq0jECDz{Wf3FsM#;i09l$nF>`mGgx
zw${Q2PNZmeDC))O_W<p*uns1m&$l;&`Cz~)cs3LB*>_5T_1hC{PI8RnQi9E)&Q>oa
z4IT;E9Nv(SW<wy9F5=SOgr(7QGw9adjA3GDkcqPT)$+V4OiT1|YNp7!6m=#DXC`_a
z^{6xYF}Pkge*x2bC$5_X*9cxvd`_%;++-Jw=Pj6M8vK0!1}4VRPIkeVdvLW`kP5#E
z_~B;3m}Qi|(JV-{z)z(ay|G!vJAklyqG+)5yy?InW{t9Afgo*YXZJkn7s?ief!3py
z_Yh_BY$jQK%Rm8*62|xs7*!zB!C8O6q=FGXHoc^Ne~BMbPm3K9LJ-!&*eBT-an;D9
z!0V~9b5R=|I$^10VXN_V1K**l#Y2;-O&}a&wu)g%7cyh6pW|@qu_Hfir#e7Z!3oza
zxie%=-OY^l)<gDIcxgTgGNW$qqnaT*BfPdB1({Kw{)lS}WJ@>iJ_<6U9<?0qhHQ||
zJnglOf6s*BzCbCE8Ff3Ud0QasI4B9Tkay5BqrQ)pQ3(khV!a>Yp7=hVtUyqG>kwl`
zZ0<+QsM|@0x&<=TZyEYjZ}x*cL;ZL!WC&Dx5&OKkTkvt}yA%@eUVv+<K5F>9A7n;7
zLO$OO*}3{dr~<iYS>F$tQQyZgh`4IMKS{|Vf9&SVocb;)4vZ0WmhB+B4B6ignNfH5
zDI!Cr8)E1n@0XT2^<8Qo#k0VshEqm+hxPJHdPdmo;>Ah{tzd%I+>e$~-lc#5#CIkw
z#fSZr%P8-Y)?_@!VppQ?O=%e}qdW~Zej@cj7~PJxa<?&s^TK$$VwSLY2+!x%o}ijJ
ze{|!*uSteMalbyU-8w;~KGSJnfSC38$g^J@0gtA;J7UWdOoq)|9bnF2yMJ6&=9!H6
zy4M+bFv)1_04~4&=7eu{N{piyIgu_J@iclHFynZeW{`2Jc$2&7_^T6w+Aoe1$n0`!
zd24V2CgGflEsnY|oJSL1o@hNmbAOV_7!yBBsiD1qT^uFO{dkq%U*8`71q%@VdzWF7
z921ujl{yxe#B3g00T7oUZ5|y>Tp;-`NWN)J=KzFN&^>?d-aSf*mOm%{XepKAkK^ye
zS97>mK>P%7DZt_0X?Tp!;S+z3{0xTY`1?84<0sjKRF>;{>%;TF_JPa59G8i09wk5*
zdeeRRqx*34)70pg2>~;a)*4eW0m+~=8z;dIXC|8c)!a|hU!&hbbCvSv%g1hV=jS(r
z%PP2Pr&du`1j^Hw?rk1De+%F8!^%_lPyguW(vIx@>V63W*Dy>xbiptPQfkSt>`-{}
z$?RalC@-Y;FE_E$v!ElVm(B7s-c0Q-iW3AH#m2P&i$zAYIPa|t(@cB5!!j`@+`-3G
zehMPWcQj8$lO)xb^0YM&q*m(KREqIw{koK>U^|uYRX-(pvI^l<f0E&Vs1jR&znRi9
zedE3GV7@hdFC0_1QVEnQ(^tn#oTqbQ!!uZdN*@T<bv(BF-QYpQb?iQBG#&+Noq3pJ
z(owIr%qFNM+zkrpbwp8HYn8IS2MsdL3C~8pEg95;!yG%!gvgWy{Cv0vk~Txacq>gs
zSMIC>vf1$l5X&7>e}12-!<rBZOGE?tlFv#moBeNaF;>XEbYi?v))g$gh673+zOqhO
zGI<dxx5Laitpu{w@dgiVIHiG|fN0_jiV6_Z1QPN|w-U%^&l^NkjLQ2EQBLTxl7L;M
zkhM`~v*QgQ7z5=#K+p>9%cg|~e~m=+N+O#*ZxDePD)%7*e?}NpvO=I$8u2TMZ1%iC
zM8*)g4-u(_aV1mZzLu_2>OS-fX+^}YB(mA_Wg?un_c3x8BHAm#yN5#LI8S7oMz(w2
zAmU<p+=mFE=!+7OQ%yvF)jG1>^9B(c<KsR=90)u7>;UJQNZ!(zS|hZQ$Y#$QL`)wa
zcgb3LzapF(f2C(d4=VTG%tJY@gtFE71`{1a<UUX|UmaCYbPYvauCFgh)3$rxprT@!
z+=q%HLEtMYW}p&ea-OHMa@)7v`vw(^p>iK8;YV2u6%|`k$y(9dCeUW@8&n{M%YCST
z6<%UR#SK)#zkYsMSsRZxd*7fUW60cxiZsG&RZ*due~L>~sGdP)C6&$IFH=!IhRt25
zD6fSlT~YBh6+NH3t;}Y(d*7f!F?Q}lg*anDQ&BmV-`L_QRSn9mUsJYw-XLOQ;M|9Z
zQ=zaTa=Ij9>Oj3RciZlHgNTVia~~oW1;{#i8-9l^E9^ugE2Hst&l^N^44C^6F(5#X
zeq2k7e?}rgRU}p~TwS+&-Xfv}IsBTFz89NVceokM+GVrX4mYQ#sQus>>PQ&xteQKk
zSAoZnG=#JDa`bl^4&chYb2@I(;ehUNuQXLhgYoHb1lS+njcWDf&<}d+{Q)E|tSpO;
zatE+J9C9ExNOgk;aF16SrTp~xoKkt&IHy$Pf3KWVrjq`g@__v*)Wun4@i2H{bDeOC
znrh3`+UcNrPbePMfE>Jt-j5FGO|kPd@|996=Bj7MI3)d3Y+v$#>EXDD2JcdTGZi^5
zE%bV>7k6oBb0FUPSogl?Zt_XpU?=z)^Un(zILC=q$@0tCrZLyK4^aW*em}*Ju}N;c
ze@J=ZEm~d4jdz^VQq|7wpLel}bF)>96-G{~{BU>A|KGyhVq*?@p4qzQ!`(etd;5L3
zxOgf5J;ck~yQLY!>!|=#Q&ds71q}`0!&^4Ixp|Cl)06>(D316xP7w!`p-iHzAiFf>
zY&8rn){m`Inrda}>_`H(v==3ipLe4}f8BF>epo+p9iPRJ)+9Zdh@+$6>Dw^K%s5&g
z55F=wke_7z_R!UFRLD6CR<&>c%|ov`-hw$rb)4iqyyhGV`%o5lBDbHi&{*iR+9bh=
zXw}NEoSHV&NgR-ET}v*@(kkJ!tE)ULYL25fZ>Vs>Z)x#bbinf@G&t|TJR6i&f8f+-
zTI?nQ2uG7vJ}tsTa2&cPRaQ-PR7z;)r(J4J^&ju<9fg0iHjQ-%oe>($7Uw*V9^i|e
zi|f!ZJFaLW&n@&qE46iQjcVWi*O*&j;bSczkCvi(4Zk+EIKRS4D;U{*Xv{CxDokp&
z=a=`ouJW+lO87+^p{@eUUE*Duf0nAQ^05(*h*QWbj$LVmj4-Yq+{U~E&szhSw9*K%
zp0iamsj^PmpyN_fcWH&RLG!KI6hyNBSyl}t0*lx5>zPe2FO{BNnJm120xlXiGRmFM
zQOcdtH#5rpQs^`1Q0kijG258ta>;VGI!W;JQXakW>B$hPleJ^E?W;7Sf0SS3WQ(Cd
zNfhX@hQ5qYaVYSUae9oo43f>MZQr%!rjdlTu5Nlcsw<_1vh{d>YIv)aFl=o);5KD9
zG|IFF0i06lv0#-$zeWO>eIc&Y!D`@gPW7rEt|c<sdVvaRGM^T^sU7c&d4D|V=DhX>
z4N<i(D~j{Aj*Oh$D*$g9Oo7U;yf0K9APU2a;n+%R@^ZjI4m^{F0p{Iv;nl{nq0NI`
zUVz`cD~vm#YrDc|dvRA-Y@Dl<e0FzN?Yj}vy-Av#udknP{s#(dzLyb@EfWJYGBB61
z@i7#a6?Gn2f34|eJy`Ez_3+`#z=!YeUOwHuJE0s5vq~5kK7AepJ>c|LaJs40V1!^A
z#qjCt@beSrpy~5MAIUi)eGL6?>8cKoC@ifI^i+t$qmkk%u7eYOo@ovEb@cx+!O==W
z!BG>e)6TCKgkKncJ^b?N!yQkIzB>suIOa5H*>Gtke*<m?Efq7yWG*FPmPAdy1}yJR
z8VAjsL91$Lf@A{ipp|8S0r>vILxQk#^ne4CMiGQ!Afr_7|N5{_1OjUsCF1}Q<X;KI
z*m)35R63CFL?#2XoC(Vm=tdIv|6At_kVzp3goJ3Z<!s(66mYTz*~^>1WWCdeG|rDf
z?xU70f76PfxTAe#Af+RKGQnWhWvH1TacdCFWljCKPg5ipM;!0c6fzRypqycb>@9%&
zwy7zy-xg%ArbcPp|FNMcMe=h5Vd$Z58CxI%HY}}){CAxQ8JpHUk!%in*8PAiM=ETn
z{GEJ$oi!!trfY!y^6)6oZP}|K<0FWM!*vEFf9a>CWLJ5Uf~*Er15td{%tJV4O$+he
z!y_1TzitLZh9P-Z4fvkKp!NNq{Kdpe`Sw0uKJ4?)#?u=2RK-2F!lJVNA)FHzl||;P
zJSQ$HO;;_?iHi-$PH$62G9}zDPXLOCKoCgLSM75IAE~))sgNhAb>7H*6tqCUuCpc`
ze=}TnyluDunS$=9Y<U5Bs533dZ|iPAky*G-%sc%+GwnLj?Qn!2b~!?gwf6*Fd&0be
zIA@l^x{{W0Wp^15*{I@<K<Junbjx~#X5g?TQ(4cv$N@C7fl>v<R|!$|#kjLA$hs{c
zRXQ3w-qvjawP3af*<}l84C?`BOrcmre=>z1A08FvFZ;sv`Ml+Ly~W&R0lUW9+o{=a
zr-)4CGGwQnBGvbBo|)EEAd6eh+bGqE99odSZP+NrR8oU%*{GxF>spX48>Iri2ia$%
zmd6lWL+&urA08f!J<~|fX090N26Go0X@#{nQoG+sDI*Ibc6BJ6f{!CDk>%-qe-$54
z1*dgK1)}UeZ^TnWF4#fx^#qO(%l{hW{W=v&cam;3AY!A+7EH>H$``LVYSMQWv<>V|
z>#IS|v%Xdwx3azpYj1sS(sLFl-_*f;Df~R`RMV-4G*L?l%AZz*GX;BiM2Wt6^`{HU
zpD_5l^wF`Al}Ea|dS$rdPdKD}LyBt@6f8VF7wMY6zc#cQ>H&*yn{8UJA_qz}yt(pC
z9YEx}r}&BdG`OYzzubLp=-b$=dn)`ycNKq15T0){&u6~lmn(T5I5bngk5hU29Q7$p
zhRvsBMMvRV;6g@D#x+4(GVmR+$>)ahM6dZtqnSvpmrlG}@Y!@66Q7rtc^*oC%hS5F
z<x0RwJR%-1a*BlhoYG&u@4mf#y8F*$t3Zvlp%D6yuXp}|q!J7*41#JvDRKDypTp<7
zpL(?EB$Ntxf+3?E!hgSi8M-?k)3L4<6ihj#!X~3q+@lC9sb3?*mbx749-HG<&ZNhd
z{{$k<9U?MfbPOqd_ZWGDv8<+lBx@<YcdLW8l9PKg&l>UJTAE6eGD-Xl{-^`gM)VYu
z6;AjwmN{yyyVW679LI2xLr9-KvUCM9WlPe5*Yml~QtdGD%k`8c-(vK~wf?AFVE%Lz
zNEpeuRHD$_sVOMKw3_hD$)hy)o{u^B^XS)_izdvCAug_8NOQA#%aZ4RRQ^Hfqh@^g
zY)w$XhF0{Lm+V_gW?jD+Cu~I1h7@%CbR<ke^!e;L>ffyGDK43}Q_N9_>G^|b;w4x%
zbx2grf~ljYuU&npA0SLLtML8UI=Fx4cdzE8k^&LAENxZbAWD>L?tGl8J~U?IpbAqV
zmM!a;Qj*HdEa=ZM(qV^xY?0ejHe#Rg@a<ylwd91n+{=@KJA|(t!tESVcWuI%Ub@Fw
z=>hli0MSeDe5dnW9HqURi3NmVK)vUxSjI+>`Ykf)Zp+>m`1w0Q$*4{2k%_BfIZ&<S
zrt-o0?I__x_JU4mpGo<&Ci+=@=cAv!IzCrQfO5>JiI|Sr$-4o6?tDE02AG0gJl8}4
z?YX*H7-+HWhy%y#;y_F97J;C{xM1?x%RA`>o3Hl=YV26A(UP)M>_ela?T&g@QwkY6
z+hg3;6jr@=ybDiNhshI;qu=LsdzB^}hgW-D?}BF?s4ucQtULDm7CP*>i|zQo==C|o
zxAY^|XItxHSjh2zx}|bW2FO9Z!+1qz^k>^J)$c-T1jn4Q6S+0sHL<>zthdXus3n_3
zv9afIwPN&{<?whuZD$#^*c0i9#nKtQzX+Jnzpnj5-}X%0WuZJqf1Yb6DqZUSvT80U
z3(#<eM(*Xy_po5+L~~7GqRz$|dF#N`V&yMw+mO0YB<FE|<0xSvbj?Eg;j}8<WzfOv
z)nSd=V5U*^*l!%xO0sIHT$ARr)H{sxE-tiSj+=gUmF8z9&XKB3Teo<1kh=DKCiB!%
zS!!UlW<pPSy1j>%M7N7C;5JMJd%m78Xnj)ZYccVqUL_@=j&7Sc@<~hG!dy`T<ifwD
zS)Ii0tNe9;c>kxza)S~jC$b*%ZA+A#;5z%b+?SC9BWx3lmzOB$hx}?H<AoBXB;d~S
z&QJd6PcLEWJE!I3pEGZ#vn!LoZ{s-Z^kM(z0jh})<M}=u-hk12eYn&Doev*#cAJmR
zDT{+cD)PoRoCr$1&CMl>Kd!&<4L+loargODzIb<kstWS`tR0)jJTc;cgKES??wE#u
zacFc{8~6$*Ne|F-!k;&w$kv7Eq5QZjQdfE$mgNyGSUofUN!(VGC9`9)G=a+4tFdDo
z4~`S0j6(*h@~Nt_^f*uqaxPgd<IN}HMXF^?TX;F<{p)%>`F23{RAUR@WRa5_4}JaM
zFHt*xjg_H%E^F%p<3^r-D7qx~#Tg*@*UF)3YSA2ojAE?kkj*SN6^{Y=ab$6v&`xKz
zW!B8=Cq*?pK8*s5Zwr(#c;>LrAH;xh{jd*h@ZQx~-;Oz6c!%HM;_-W<pY)@MQ+LQ;
zAJr#&ipL3uGGCsS<(Kr{9karw_b?XDEz`-D*M1&4e|6UAG(}LV3i|=Ge{tTaY;+mJ
zSXtOW&eH!gcC0CvqVPC>{)G>}(!&q5*tau(gr#2#9X|Yf__Md_4a8JH1;p{Bas@FT
zfH~_D+7X397izGxQS9W*|5Cxqb+lw1PW5<KJ=Y({Q&;YPbY!5a4Z1059WQ-0i8Z_^
z>lTNHf8&-K+bF`icvl<Jk}|xLek;6%-y7pC{33eod?W4psj-p^+O4YEn6-Cm?qZHN
zBmqEWvbnV+TzJVvP=|`udp)Xp0V~&xrgW(J{HctJ504UrYz#lQSexRycv~CdA{`59
zzmd3vWD<4r8dNUE>bC`F3ef5dd1_!;cB3JhfAmXC9)H2<^HunzX1*7GrMXkm7fWPA
z(ta#u-8Ygn$M9>S>K%<@n5m>5Uq=zSv{(5RmNGG_>dR_|Gzs9<>tb8dym;5^(zMS1
z%JrMb()r?R?9{xEzl|9xi18(z$jua73o%h-`>zUEm?>Vl+9%gb_rESx<I}jfdDgoz
zINHpuHVg%Ro1Eu=o;gi;pzAk9$M<2!gM%h=EmqVSU#Oc`lL*(%uaGzG;$oRh3h3_o
ze=IfFv6o?z925dMHJ3rJ0VJ0YR~ieKJCH3Wf7MxSZyUK0{_bDFkJo<Sk~4fg90KH}
zz8=uSxpA&&YZ!(iZynXP<Vq_v{r7!_E8E(v70D~*E(wrixFR{jd3k1rbS7D4lrxEx
zWfp=Jg&>4eoRbh<`QU{JDl&n{q_XvjO;p*HDOu%=ZLBfMTFytHg0kewvIr_Vp^~x=
ze~YjKD9>dD7>9N(D;p^Z<v_*q%`2O1;h8~sC*QInlO1?itH>t#hABig%C8BO&+^SF
z^0L<2fHRpw$?2TzY_UtiA)F<A%VgzS1SKc57_#vJJcvN0vR_hcS>_SW&V?^U#V!^K
z>T0=R0}fQEptwB|{^96FKFUxl$qG5Xf5{#^7(<F+i32&t=~)D0SN0I9@DkRJf>78p
zcv#EIQKBmpevL__W3Wk7zB%O$iKe_!zA&`d)R)QD1?4G-OfD)Iqx_mc6~5Tiw|v3K
zdy-$hc!Dhyb_rs73sw6VEqDaXOHuao*g54wMFqu+EWxTEEI|>dC>r%iK~W+LfBY+O
z(GLU)&K1gmN@AgqRR|`52b2^-E_^}7VvT@ZViX<`yJRCf$E2bu){O*;3&kqRWKa=~
zY-BA=7nx-Rjv8Gc@PNNYU*;tPs+<FgjZuQYB2Z#3@*x~jaRs<ilC$6wgUV$)eB;74
zTa_FYia-Sx7D**<$~UJ{$Rd2ke|9kyDyo#Ms0u%%RCdOu*+_sA<j&%=pd<{TS5U07
z`~%8?a{lOtAF6l%yQ{00)9I|KPM7Z+8U1B4y;jw$+5EDepBV(cIj{a){Zzd^v+dw*
zJ#NkryBb0%Vygp;BL7F&g`Cyo(s}h#Vf)kS$Jx7C^+tV1L^*~)a9t9&e>RT(@B)AK
z62fKH8E3s0!Y8kzg#HF0;8qA9NAO-BB#M1HfDkTehbV^9p#gZhGhhx}u~u^!D8tii
z0i4RN*b>;l&vMWKK1a!D0bwx<<$QkBIYjmHVe@e_nN1<iuWx^nzu$dmn!CmKRdqFK
zJ}lqs@$9zx*Sx-*Pp<3hf4Qt~>iMXt^}lZ*<<S5!6HEj@UC9Zz){zFt*({scd@{OO
zRO8unGOqr!98H_$?ZxF}F`iFuC)1HE7%rw`osbM7#4FegTOyuQ(oF98+N`Jgzmx09
zU41zj>Dm0Ml2P^7mCTC<8vR7_N3q|W0r5r9x7FYO@lU{!ih#fAfAZ$${K*N{L_{J~
z%DHIzai*$X&u(V()4S2QRxArtbuwyT(W#<#_@|G}kEbGzvh9cC*|Y(VW274^>0P)S
zLlekd+zyZztfZKB;A9Zn0YJz~?!w^3i|S-P8=uzAS#|Q}xO!KAY|hurax%KAA>Fj8
zr_Dk#a$yf!EN093f4E+>$y`4DEINEO`&eujtRO3$WBnW}F$)+hvx=F^MxkYYbXJNA
zIa8@AS|S(Xd_Uxao;~t>@L4<LJ(LvMqKHAx(Z;!6!29UDiwy5iKxZU@$oqp1ane$X
z+zY`#10eq<8O9$*cQCuQ!?4DitR-nDr@dr)b~;GX%?>9&e^_{&{PMS-J7FM|;-i;K
zz6npnLU&dh3#bXSM^@iH(_<#*u)|CwF5vo?kFT{Ts*VxosLNU(KSoi6`oOFPUIunG
zfE${X%*Uu}kmpx18OmT4t*{|Vjn&o&VGi)yGO)6nO+!N0nomR5t_0pGbk}sAKM#8A
z6CTW$Nqdw~e_sXYeQK_Cqjc@ivCz6zH(O}U-p{s>?^($55DQtiB_0o^{Ei;F8tUk7
zJqS7mTeHM-)kS@WwrNN%B7Vq4g4O$H9wIFb8LHhJr6o>UdPpfajk7(_U(^MEb=3X=
z^@|m^#hHB~&X575rT<Fetc!#l@as5E2P1>D|8^Tjf9~Trq31Z{hQt~9-zv`5l>gk*
zGWSl)c1W1;W636WAf-}US-QOe=?Biv?U9S5(ZAofM=p`($X{Q3w373_J$kmBkL}Uu
z<_c8v;dW8Iucu(XxVWq#_uXtU*{?;4=xgQf2@VNSk(qRq+ok|a3WurzUp*yqw<u)V
z0$*zre;(>V`1|%*-lCpGk+FRy1A1dP6dZP-Z8J80rWV`?ZJXgxqQY3;gV1NgK<>jp
z90Gcrb}9BdvsR6-m&P}K33%LGd2|)KuK%61yjQdA`t_XKrJl1ccahhdN91+1e}O{G
zs7O-5h;sSQv9!tp65hm370MqqJMNSy&TWZtf9IZvaWG49hnb$NmRsE~;4qy~K2$;K
zUC-X0cDMJVE<FDZb;Vvs5B*x)ry;H$)@pGdR<(scNKww3b5`U7QnsLi(E-6k!Rs8-
z=X~Doe-FgAT=<|DVD!PcL>`JKWg~0Lo9}^QC*1tDj==Xj0>U}7r{C>5<k_CoFV2G-
ze-cE&e``S$`#}`aprdc|lhF4Qd83gFp>M@cV&6}q8-h47Jt9t|72-tRA8}l=+N0O>
zo;0z4cyRK8fCau6^wFSKLxo_Cb@BvfY&gILG=N8*+8OXi8C6bfZ(Nzh-0oRIi08y_
zE~-+$3QL2ku=bgvb(CaQj*{#>LHqQ-f7o2hOa8<AU=BJ)|MfM%_kun=5I$GXCp+Nr
zvS}vMtBXZ58(&|HXE)2+>BUt&6`A+xh3{X#Q9)?uMOq|##wkKX2*yKo!>>Ey-_2+5
zZ|d7_H=MRr%Bf%f5*uyvH-pHzpT5$d^p)~19cIcqv=(2cVc7h@(``4``^ZDVM_KN}
zS>_>H&f7dP_dSx)k@?@cnB4sE(tWuZ{P04i@40!9=jNR}cWss@bRp0~2uANXlN>q+
zx5G@IvUX6>YUcj|<o7|BmmzQ<AeT!@7z&rc!U!jSTT5@`xDmelSM<?yFo#tvzOr5e
z-96()_LLw75@fT7Ae#*)$N-aElAoXAQ>0p=)NT79%aT|{@~a1{ifnoK$3Y%`xI4Yv
zeSB7Qu)=AuczF5s0Q^DnUy%IL=z|3iM#16b>*1$IDbetC;)|qIiZ9FmJHG3u2L+B>
z0DlF4czUn|kEtC#^L6Al7~1jw=S;^fnT1q_X}$5GJurW;;d=V{<;Odj2MT<L=KlHR
z?%#Lwp=vD!NId-Q>s|Oj3WRca0QACxI{fGFhhOi0Us)8^Qn>IR|NWuk>4apS#u&ca
z!1Je{<>43p_#-z~-W~oMn|=+=rSRGx{&o0&1M`<HwhkCuX^hq#w%~;|$jrS^;Zwpk
z;K_54#Q5+{5@SctPY*^0WHckAZF(E-+J$$FoB<5LeF_c3f9R7)qKHg_8+8@~zbDM1
z!z>C2;z)2eSac9;<%J~w3OF`k)|y4<-U(*yn3Jh0oWa`%FF7zByb&(UFq#Y_L#teW
z*6MTCbl3^$%~Riilt=*ce;-K63R1ReZongG?kBziv<1wd$v(WRF#7_076N}^MRgs-
z(E%{bK>1SB71+zq`C0FT_+}_JAXZ9%hj)Tl89~-v4{>OvF>5fRNOlQgogqGFpZi*f
zZ-!z6Vnl%o-w9&WLW8dm>kRGl<OUXh;63KAik$-jEyEfX!DFMnF0tlM>6v!Idh_%*
zVC7XEt$H)Qb-ZuPw~kjb6QH+N10Jka%v=c;Ur_@&x%y+aq@Wm|s{)wbzM`7A)QVS(
zeR92dK4YA;OA*Dpn2Yu(45?}^8Heb~dna$3AjTCm<teSQ#e2!~(*xT@zq$s0Lx~0y
zjKt?>pO%EqFR6V%-wwzI&^E;s?*p{snA=ecbb(W=?4N?dovKQndo}F57iKjz72^qQ
zD$#bnjyqAkdG;HqniN*N52{uQi#6KDf|H)2?u6f~QI-20{OvQ|KvJjZ;eC+Q;a&sk
z5D`a5V7qnl!c9IsXae6uwbzUlZ=TK;OiGa3hzwD9SJ*BM?MfC~BbVun9x{J(YC-4x
z#nd-c$^_)g+WHg|tmxd9g0E<37kyGvsQmU~B`STomUD2Ur?&sd#q>s=<`-Mg6hCK9
z7rndUCQRE#V;z|*Yu?1Q{@0G$<#9s&TCj(a1FF56GQZqf=6EwDp&y^MJ2;`eHEK!z
z<gHNT+ak~6$mpt(QHPnEVkuvJ&Z8P@B^x{&^b~J2`i53N#+XdSyNf*eEfsk{#^jvG
zdrA^RYkJ))oK|L>zwVrIZk_cZJr#Y(Pi!Ob*ko{+xJS-`?o&IzxxQuoS;LF=#yRQ|
z7K~@qjx6DH(U+I`jUFXo!wEz8^~q0bb(}7>Q|=~J#ALcPDU9zzy^Imw*91FKxI7*^
zsSzEkF{@_8)n#P&l@%@<+(loZG>Mag^+|Dy5@zS<o>D(W<g<8ha=j4)Uyc9RIhg0n
z7gN%b+vIp9ewTla9u*_Az5i!c4r`Z5oFFqJ`The-)84~St<WiEcLD9v`qcT=ZUvXO
zjvgTg^%|_8P*|Rq-;N#_e_jKYNC;yj+L3%W?uAG=sg`?LmN#hRsHvaIE^%{uBBW9?
z2xC$SRiw4#b?(9g?5ssiu|9Q6vot9qlIO7ygH3yXs}LLNy+V{6{7RkSRU1+aZCy{Z
zB}K<0)3a_iq{u=5=qIyNj2vl7Jp-{3qg@oEyFrX#Hbto%Co;;de~5A}y1sSXU&&6f
zQM84qc;qr6_&?Ac@2VCnB4!9dhQT{tic^)~{>z4pt><ildYTOxJ0rA$eq>q3^246^
zO&KkWH7#Bm<|vf~aG$TWQ+zd)D}gYeZZy%BdZ&ph;zcM|lJ8OOGSQA^pNV!P`%JW(
zOfgZvVWS+VIWt2sf7@u-GI&FGt5Z2|%TD=D8u!|1*ZPg^wBfziP7yfnb=TQxOS98X
z+mT&mr|7*<ee(CpIJlbXa~6J0)W#rtDKua^pe_6JMxZoI5tb9ZC(dEn3t%WalCe}!
zqR^z4ds&HVC3{WR&17Fx%lVbHY9^DW6fumyi8C!u&i3Pne;hnp9!WPd9OI`+*g51I
z_-QMo#jEf$BorIZ^#D3#LR<oHy*>qdvH_d-R;wqYrTWfg5>*SKGSi=@O2SlIs;MmG
zih%8^Rb7oKes3!Y)EVMZtqg{W!yvLudv*Qe5N3@|XbCfzSF+zA6Ak)eKLeU20oCJ0
zSdb@Jz7`?Qf8VD6K%KB%j!GLUUdc5gC_trN%Yv2Kmi^rVd3rl9{Uyp{+e=&Vj^eEC
zrrJVnOO;2SQIlgAs`IqL7)y$C%H}LZR9D-LWX*>s+6^Bb)948g*UjW&j<wH4rEDa%
zs2!Zy=^7=L9l{7z+#IapiBHt7=IOi2LozFto9IT6e^fq8kl1Ni8#`Z>Y-pO!{kO4?
z(;&#%j$-EH*6J+mWL<|Gd*BB!=KeR~X49=ceV1;(8#_*3N<F9My*Q#C$XaDOH;yU6
z+8mk6@?9Hs2wr_y1)5R+b-2_hj+Bv84>UgR*5nB4+i)IAbEL-cP(P0<?kxK1z!z14
z>IO(ve-1gV!m)mq-n!o7M1QfMv<pG-u*e$LNmLzDU&ih@9AV%<d1gz;Weccp&UIzD
zFiJ+z$m+7kwlNOs6o<3`9mo5z2t!>7!~FX}gj1yyJ&Ec{d+nyptQKnO^0%<<WY@Nx
zT$d>HDXOfMY9))OvuV$__1E$Jah~87PhFRPf9iOvS;h6(Y&FPf3#}_PethPIbuj?t
zEHlKcz=mb&S%QdK;9aO!f;?W{U5E?ObJ~;6`Bnd6++v5wabb`%a>tm1h(C=(NhWb&
zs9GpB^O$aJSe}Q|QhdMRE7|LWJ?j>Sjl>N@g3eZipat1I3d_<GCcxE$9W&<BbRsNg
zf4Q|EkQ@$k0>>k?wp{=rYf;_48%qhiVm}GjNGYURNO*;mfSoYVRbuK$cFL(mV`F&;
za+0iRF==+Wn=jwS-RP7hcin8LgATJWeK)liy|zGygN0o$f03xiW3!wlDcw{3G9DX8
z)`&9%hC@@%GKJ|oal>&bk1QJKNxlq2e;egl17@A<3T+68<>PwF9kI=CciHF6)H?e-
zSNmL*4=TZ~`v9B)m&4Cny2F-WNylvuPy2L783Dm~+FlynS&1dt77V+@j!JH50#65L
z9c^fW<E$?jX-o2LLlcZpG$Yxm3c%jY_mr;=b2$%Xpn9%zRiLgYxv35u;E-2Wf2o7A
z!k2xMo$Ao`QYti-*2+LSn0QO_=Y6`+lI+xlmSm?cv?SN*LQDB}y3lZ4(gnI$x=WbM
zlHtr`I;rWVVp2WqVw*gnDb%E$_#C^MVf8~(lYiCZH@;2PjQSL61G=7=aP_!N`TksE
z&Y0;_iI%k*RY`PvA|sEc9jocaf4Sq7M4B_!EuYdy-AK^X7QGVR)PLe*n{9o4Q7g$Y
zGc`Zy=?Vh6J(!dGa<lo*kNH%(rU0?15mLvhe|xHA`54YHX(APmEw|h@j0-dfP&1NU
z#>Hln){d6))y5S{^EyOYJs>M3!mux?X!8J?gKG|;IcT*gJ<-LtwG+Pie@Od87p30*
zB;Y*RLCnt_E>#NM#m~!hQ!Pke=e3pGXGbk(#jpCMmdhNpb~??lt%;teAHL=Gd+pVm
z`JqS$OUG<nIsIEp+Kn&D=Z}_Dz#4(JGn`t6!Dgc2s2BOiM*PR}W_ac;(UjA8sXgU_
zkX`=X*7Eb4B7|~wF1`)1RYLrgW%wDOs%_08uX2_zy**Ccn*z2Si~w_?z+V1PwC#6&
z7rp7ZRbZT|Lm8biTHlw*7LnX`pGzVSn>;wxCfvfhH8?6;ok=}Wo$sGt?*0$cdu*3M
z!ZQ;BG%}a5?Hv;WGBcM!HySCI5tbgC0eQErmL5_Ae?~Lb@1~~-J@?c_0C&Z8V#a^r
z$!TP}hhSr91t%8FlZTcGLP4Q%O&G?~7`flwN_)(7pMX)}mtXN?T!U{i(T#Y62@H3x
z&N)nQMVQb96Y?+<-H11sVC*Com?X@Yf(3P$g)X!k3}9@+{Cg4$ATTM}f(3n;g>J+f
zOkj$pe}<pL1m%R6lfoP(q8so!5yDsk>uJBA2w@o!00k5FC=>mNH<-{qVV}bUZk~kX
z1rwLUOmrjOU_$x)b`BF7w<cB;Ok58$(T#Y63F-6OIZWWDUUFG5aXZXJH{uN@gwJv3
zFd+<X;hP1OTQ3JT_{4@Bk9Kyl(T#e84eoQ@e>rS$O%&*c4RM%_ZqyrWurc4ALk3es
zNz;%a50lXgdV>q_DeoLIKoVtbLxx%;L+>P`8}$Yo)TO<X7My|#qD4c7UL?crB%>Sj
zIvKdx9y8x5WN^!gHqfwPj<V5@dV>w^v)?&v7)A_Z4I6fmjoVH(x>0Yip?n5BhYbzH
ze<<Ftaam+T>|b#7Bi~>|`Yd=3BMK{0*)VclWkl{|q#OALBf_V`a~KhpSkW%QZmW!#
zos4uN-(ZCMe0UBc+z@M;-?mm3rQ(4&(v5tJ5k@o|`K|p?3;RwNoK?BLz*st6sE;vr
z2+eSaNR|sPuX0r5m|U&nl{iUVV}k-Gf9QC;ekj~~%hgG(T53r<q?6WZ(k&J0AR9D3
z_w4lj5`MIQ>l4x_x55|A!4Q)Kq*@o!P@*~BgYYt*d=+<o-zBzT+93C7jXEB1j8}E|
zwcS{K1k04`<0Q9O6m|JC#z<)z>{y0n613}hEEg4HGMuNwo4=SY?@wnRYZ{_hfBT$+
z8h=QWoe9xfKeKVngqOs`gHw^T6*d#7shYz?#*8e^*PlNo_md#BklK>7XI`m|AchNN
zvq>22Y>xKSVm94^2yn^jh?h-jH;={JPi2;S-%5(QMhCa;)`-^c6P{Zf4VE3?i76D7
zc;Yvn^)HN15Ef=7=AAro$=9cXe|ug@47wO_DVfR5m~#B+o&k3cb79nq|L>A}1bJX>
z`DxK>oI^1v^r(P>q^Sbvqu7EPX_PKi6?%f@<I*6ih?M#PupH!AvZ`^i+=e?@9%jKB
z=YoVu<5EbAy%?9&0d!t~$S5PYK1iOo9Vs7|pW+={GFL5CBE%z7!YD_re_P0ei?J_W
zitvz-v$8tsm*^~)U;$@y`Z>~Pr4T13N&Lh71xZ+Cz2Wb1S!iW(RG_#e3V&YYjLaB}
z-o_S@(`WE~QBn}BTp6Jutxln2l$bq1sTFjvHR#*^upacum4$JyaEm(b^HqIPVs{%9
zORB!0BApqklzwyw0&gDwe_C=+<M2RO3|h?wZ@d;1dy3$0yz!1F_(5<xWeMZGuB(7p
z3?7vc_5>&OQMEGM`skh!SoUMD$}bx^^2dQVF-n8;$=QV`WlY^gRmR>(ur!2e_Uzaj
zA(YnX`=W@RQ9*=+y2ep4HUlCs9adJCrZVp6HE#54PM$BuX3G5>f5!LXQdmkgY|6|0
zhzd<O@EpC2O5iK?a^UOEweY>%Db{}R)NtYAk=)kzKJed|m_%_{xs_*F26S4ko~U21
zEbd-b9^a$^<wR)xOqPOW1Gu}eC+z-8!{aoKKSAhB62JJbw0yI7AjJ5fQG!|9=y8Ez
znN=0~eif5yaf5<+HBQY_)BR_>+~meT@$#a)8knxVkl(4~W*35G%)NP0A*aQn8*+G`
z`yn{Gc=i7N@$!G7%UYMwi8m7iG%_-m;cF)o0x>g}5kCbdVpz>?<2Vey`ziDSQALT=
z4^RYX+<LHkn?Y`aJq>2&R4h90e^|<vY)i76r0E_UJ2FL5A3u?_l=v-__|l!u-NRr|
zIOz$T7Uy3A_=VzMP&{Z!H~?u4i1VxX{-~6;e4Y50-=7{8f8!*l2fm(IMn^e*?$bP$
zWEIL-o*Qh4^1%9m-$6M3bN;(i$x$VPjgRqG|9d%iZ(Z#+67rgY`0?7s7id5luZ7km
zfvouRQ~c__)@s+NcwkhNwL}Ff$ZCVIHb@iSOrQ`i1bLCj3*+@EvXk=?dDi^%=eMXv
zzI|Ezc02;Mf7Yy30mCBgWzbYhtJCdvy^-LW6@l5d>zzdAOKo>J9<0)j2?eGD0pB^~
z#G%Af^o*Z5)`-*CI3DF_k|hqi>TmX5TQB)){CTO!bAv`_rE{*q12^%|9;D*UQQPlF
zOyZ21IOHW$eI`!-dBiC-Um{K^eGhR;O_zvM*<Fa^e-?5e-cAJ2aSzwdz;@KYNevAM
z=IPzYVO)W0wmkccG?5vH4@3{8bx>=sqNk)YJQdQ<sIeH?ic6_6(#yQF0TpdSr9q-?
zwQWOWtVHJ<a(@y;rS{K&sMI(@6yeax?~$DFURK)y&3K94HTo;iG*>FFxt2>3L0ED}
zLP$j@e<=Q_gitZU>v<T0+ALC8GoYlI-WaB(QR(!hN>}KJX&n2!ZzD7@segvcImo5U
zF)hM%?&btjDo8;iKjqiV_&L+~n@UrSjWUw6*KBM^+_5VIeF7ammfZ8IN=tjmJ&+7&
z8oz7;z^!~rOh$c;-(!J*CQI+|phAg)mX^KBe_7Tn7DtahfQq#etgbchWcyqoD{0I$
z7<n<5z{Lj0pqwDC@Yds;(<<-lJ)LK<O{ZzS^)P3{VQU+`*Ed}ipMJY`Wq&A0XP34k
zWn&5}s@r8dTx$_twymL;Ub!u#Bj4yLp^^D`d-I_cN+hT)H8tVp2BK03Ceo_;8d<6%
ze{q<xp0yVQ!)Y1)I$hjb8dG|xG&eBZk(X0SseKR0_aPJKswABouS(k<UfN3ULy^W-
z=_=Q;kKk2mzNeMQ_>>yQViUh)`-ww&BSqZe*ly#1QXAc9+w8Uo*tGc{+6Lmy6&%zx
z6y43Nsx@OksX#Yr$fW{B*Oz;8(voR3e-`Qaendb?PbiQZMn!l5kzHT27Une=yH`ra
z)Oe{6IbEv3SAAe_9eU*Z%egsA0o43;YvWnUTPGd&s`35&Fdnf?lixZUv5f~V>CP`7
z{(99xJpUWdTaw(@hWdwW?69&n9>kCrUQ%$@Y#mPl#B-jRtR~e>PDlrXeqDX+e|j`R
zMY-3fRsEe(!KFj7z>WR#xo><@KQ~tKEX+MFWqdR|o{r5j{Zdw^OqwNfA!h#Ok?3TY
z|7}M3IN8gQ9%c`o3hTL`!+L~gYe>9aI^(Qik{{W=@@E9lEhiV_{dsL`l$w0Yw(3w!
zaw@IR<vPeY^0~}43pf7*QoSs@mx2lo5(G3dGBvjWq#go0mr)}Q6ah4sL1`Z-mq6bh
zoPWU2;Qac_x1U~QW>gAq!oT4%e*60E<v%YY+uB>9P@n$s{Utmg1wuJIITXg9{`lAF
z*O$MHM0-peFecI%ttXg(WT5C|j29|AN|>-nQnPHtd=q0qx5T?3ey#rb<G0gr=dV9D
zwv}Q2V5E$4f<hflB%B`ys~H=7<C0*eYkyjT2)TKjJ<m$u+&CjI^CEj*3r}jsAM>V}
zNo^B90k(MklK9o(#x+u$0j`~34XJ>e{6?UJ`TI3V9PdMV7Nplps9ZAw#<HlOP9S7P
zU~;ENpm0J~kg|1wf;^t7QsO^y!T*yk+^j617rclgwN-v^uOE#J=Apy-eLcQ@(|`GZ
z1|xffVqW8c;+I8vSsnr2j-&haqH|u)OU6NERBgVWH6*UB32Vsr<ikjyHNx5%CkQiR
zdx)y@Vy+yZmcj5!SU+F<;x*Vo$)J5wu>VR*+27%5Ki#}s@=r>lP;#~?!QUp6nDDda
z<rL^}bH?8rjE21%SvN5n%tn%ppMOl}82q~7XvL1ejTE;HM-^uS<t}DNqFs77(zqnW
zdW)>6g$Cct8fC+#x>2Qdb`;BN)8w_nXf;yZmbcuEPL{HX+2B0FoLRc4Qd%bEiqcK)
z+|e{dFFfNv@Rg)_*o@XR5eWr{5#uaxa9c=5c-}}xieCnb+sIUk8iG}h%YS}dHhs;b
z<V;COPLy)P^Q9#`<o&8?59)M245&gSw4KQ>m-hjsP)IkEUtb=Vgm^Xl%rF-Gczq+d
z4n31S8cIf+=y5)*6w1ydT8WH!q%b%$o4DfF<_P18zip)nP<U`V`693)#QzJ}tVLu%
zLbrfyBeaVeem@wSO(2zEdw=LdzWC-Xd+ah(_NyffEZ)5Oy<j-karW2fBFAWODH{O3
zLU&zEHWYy%R_h_zK)$ky=4_)-XjbOpbdLXD$X{5m5N#m&^DpxBPk#6l4-R|Zf8)^a
zbsBm4_vvpTo!KDl*z}pOVMnX{J;L(Dy2JTJBHKu4<%J~wIEkH{X@A4c;XYMX4zEYD
zcQdx&ofXsORK(cMJ!NOBdcE?+*_K@t>bTV;s!23ajf&M#p*FB0Edb%&=vH>$nQrA*
zkpP2FT3p3}s$N;!&b>$9uy@@KGrR!=@0Z9S`a1*Rg=J<na#ZYVttQl3%2uMYv*S3k
zM2ICStqJq%M?#>9+JBc&YZI=+Vr{_1+ZYl(2`-L?w5nT->SL`gU8vDP-EL|&zN}yc
zyQYZP70a$E8A*k)o#-A$VFO*s1^}QyU%%w^lhBnMAz0N{r>ZiVocx?)qE{5sXY2#U
zvPQ$XCXmMA7z+5dW+jN<MDwuJ_t9kjg+lWjG|6)~1XP_|Fzv5xTZ(^3UUJm95M-^m
zLRc%yQ4WptK1h#?z5%IC+xB~JbMH;J@@pK&;ZKTT$bvAf2-@~JG}9bEDny^wxQWAB
zjiL*^7%Si1J7lQRQx`gK>BnJQHt^PQ=YNjH87XWle+B<tD!k1^wsrz#V7e8cPv<4^
z{KTZ~0CheiaVd|9TLpjStpdWfu@mK^Xl#H?DXMr5%EZ^MG?=m5hb}oV(~TAqauG$7
zQ4H?Sb(?Q^20IZwEcyndDn%a8fmAu+S`x3eE3E})-#mX@1b)JL>ZTE4IGkgcK_?dg
zs(oUr|22Q#yLB9fVgq7K;m30z#*n4M;}Dac2Hj-;42Tayu>pTEq?qJ65Q7oE)qhGZ
z6OW8&72b1xs+MvYm0?x$&jePr6V}7xZ@?;11W?{R=y_+lg<s`UT=JBlM|l$7(L!>6
z&XyD-;q|1p%jtHe{>SCnz{#b6;wk2$90?92s=3T^r0V4RFyBo=KGVDx2X-I9$?5JI
z{ZTO>7kV`E)un$I!_iuNwliKi4#)=3CdCxb0otl~gwg?B;EZMba-@B!DtVo%;k=x%
zTr}A0hK@_Sfuc@f#B)$IC_uIoYTZ@qbvq1cmDs~5%Kb&G;}UNmNhw-*4w4!KPz|6%
z1l*GUd$m4I;EyTXVR5!#BH=8)<z(Vv68jEg0$FpP?Am`}%y{avqvKBT)~CIllqGYS
zCulf&vm=v7Z$C%-!E)54DM$6!bRWz8N;AftG3FfKFEhp2OzQ8Wn0vj2yY`OldND$X
z3^Hs#YX^%tH_SwIcWz20H_XXPBHUFFGkf!fy+Y0ACvL)8HmYDYiNd6%!}hHl%d^^2
z-z*{Qt%rXqj7fWQy&HJ&(}U5h9_YO=1pV2u#;&e<ophh&rp)D@f>)^0nR+%no6mu;
zvYoJB@_gq&yJ1#mI=6X2U9jaj={Zpy37L;`eKx9%Oo-<31yM>!<=JlyzqkrES6-m3
zLgOtJ$?CUm+wK*;)Rd{tv32;27?JyQ)9&Pg?HYfzWZ0-lprpQ)mI|6(7h~#%<GE>u
z60M_D<J&Fwf}C(W>3S{}F(+FRD%WwLm(N2QXETXPC5-ColYKz0`ZONo*S9;zR#U>u
zh~H?3S*GhWH;A2wx|KfYTCB03ac2`(Sv9PUP@itROMBrB8tT_0lxaNr$$EVF-w*ee
zka&ORRT{@7Y_;FY?5!;VeO*$kYD5blxnk6<5Vz<UP%(@3fM-V=>>3n6D0nBCQ2g+B
zj%6|$Y)(datK2jKVl*rmehV3u<e)zMeiy*dKB&=+tVgy@XXmDgP_?9agL4;-UaB`)
zY`7ae+m05%HIPEvsHXv>`{H6~g13GYcg=s%OU(*d{epj0x@fYl-qz(kqxj(=?Fn6M
z&GS~3f(4Z-UcK#fIn<qtA&uamXkhSOhhwaonc6tDice_H;M8d#e9imem!-a-?Vlci
z)&c;=ZlF+=)%51y9)MFq$@bd=*OaPniC=PZdDDNmdEA)T3Ng&Q*8FezeYVPBbnAbv
zUF`Pag%OIL*-r*0EJKN^RVelaoRX#-_`L?_uV@CU>%Z5K!YDnCo*G1wXkDuhH~kN@
z_7bnO>5qKM_h^y}S8{&dB*#u1jy1^va^?v5{!Ma#oJS(PB^7;`#7@qDUAcKuu@yoa
zXtz(O?i|AHD-ImzZ%b}|V(pt&<WYYTJCzW^j{@wUR2;Cj?}}A!o;DEDu>BJ1&)vZk
zb_s3vt7OW1<Gv9qa%!j~*ySW3&zv^&bv}kLAdkBnJ*L2dG;|4Ore<1NmnoGy*BK8~
zy4-Y5MH<H`9VyT1@!Yk%(PI-)>=O`bX?#Djep?9*GatSNSxZrw(VXp@^-h2B`++qp
zUPwkeB3lu%vK(c)aoThD1G^BqY1s$0Cg&0w`XoDj>jK6W^xhA2+2m<pdNUb7Ai-(G
z$?!$vuHYD@K;AqCN$*s=G?rhqrf!&d8D)Mlna5spXsjnC5)Fd}r#rm{C!}YkkBoGE
zqtoF0#-PFZjU5`C*Y;XCl4B?^pD1O!WJ|B%eyFuqE=&58-Zx!WQl$~&ovHR$mz%G=
z$NV47!#8A?A#flR0y8+51?nC?e<Qc;K3~zt*~8ET3BKa-R7u*E%f5NDRY}G1DT%jS
zl|954SN=ZWv-#+jx+QH6JuS6J0K|&`h#o$Dp7`|r)A7^OH)n~HVOD_=)2Cl2pdUE>
z1x^pOnhXHb5~fdIrk`Iq7n-gkUC6nVbQ$_=>8{>hBv={&^ee#Ii$Qpef8)fNt^<t`
zejI&1rFk@xKyayPuG7ws2Z9gALwWn<(~nO)5!zFw4Kjj}sY3e(hVF_WB4H#XToGtI
z``_rx`RzqG?MD)9Wy1`5BK#cxMEGyg|BmH3#JTE_pP-I&!wkCcltz^i0Vh@?{AiVE
zqmlgNNBVWDL!2KdJZJ1mf2aAbq0mFofgj(TwU``pT4=1LC4n<Tk7~X=|D8*CmuAUH
zGiRh#?JUg@=_ezYqI;P2Tbi=07E%bt74`z4MT$VG$V-Zg@rr<;UZcSASXd(9HGW1H
zXJrVjxB{DIMjR4*z_WL=$P!I~jb1UJ7CmN_G36;wPmW2%Jl)?Xf9O#ZgGy@Rj1`3=
zWxd`oOfOtg1jrO00?afjjL_*4aRot9a8@Za+1RogMQr71gWJ|>k5S#yIG5n683(G_
zY&;G8+~7J^YFnV!@(gs#M5i7NCGle9j!Al81}d|um|>lsKW=^1nXXzZ;e9eBP)mj+
zNSdJZq{%eE*y3Qke^VnG48C15C?Uy^fMyF@oaXZn*#_+*LSdLxhNdW{F{;n=UoV=i
zAn4q=NpoZ<I<qC37YXCcfo-RJSG>cbPZ1~?a@(-tx=#`4$P8#bfJ|F#iCaOW1R=r}
z(SI*vQHBX?m$58+E(OG3>3YfXW8SjpBE16@+;nZ>0whyXe^1H+BXJcA*KwBDJMwc|
z^g=mF<nB=mGEt`8F11kFF*IV2S~OEOszuF>=RdAUi_`uxDdU)ta+^v>lAvjkTvG|l
z3>nZemStnt$gF4uOByclGHpgN=@Hel{<2B;It?0hPZ!-=Mef~%BgQDwVj(s$hJ;Lr
zX$skju$veoe>KzeQ-g&Vqp~)cWz8Vp84y*o_(eoSD*2-hk-f-%NfzOBJL9*DbV)~a
zbu+$u_acmWzO3p9f@#$;``dRfp!M^g{LRE$`SxwR{jg0LDyU1!V74m*0=YD|OBtkF
zaL<>Op~fOI$8lK`G^4$;o$|MBnov*KrU~_oZJJQef7qu9^~}d<LIrh66YQ`hm6<{_
zA=T<kH_Xi8Dmq6#xoDry$$Tg!unDD3F*Hq2WfxM_p(0OVyUt;E4C3YyK4|Tr+ZJ*P
z^K>5`@52wTRZt&jcR8$v0x8dazYwUphB7y=q~whe;-O1qu*vyol<8B993(6&CsLS!
zlt9#ye<^CYtUw5sAsKon%vlhs&bsF1ADV^asNGcgPFc(#w1$y^H6~qilpkp(H*=w|
zfYy*F(j}V8u5P6qK?=_);Ok~(AunYWw$A@7$Rk$Sl+9UWo<HUjKCevAtGu#<+vSyA
z#&WLn7&Mi#^@^kWT&T-rHYjYuq5>~7BtB=?e-Yn+0~^X$IT$r$jZ-y@x)>Aqt1>Um
zT22PbE{dX3JIh)O)q|2VpPCFh%f>VuOl6+*t(Kqb$923PoKv06ZgEU0=M1~m=PBI2
z@4kKb^z@&B0m9G@P7(b-zC8IKPy&<WASlLSQcS=9Yx?!+rygo5m~qBw3RC_Mefs_L
zf7Ctsn5ut{leux)-&JdFoNN#$q>xUlvNMHZO){NE1x4|0nSf;eBV@`C0SCyVK<E;e
zm>+jLG?lZ$#Afts%zwhtn<pRVdZLe@j2Nha_p{Kt3Z#%oJFe(wh=z{ff+H>0#V{u@
zl(-PY7P90Y8aj5Zj`AB0a>MF;l9@@Pe;_PCZJvDmepOe5uO5nlBht05D!HL8IwBlp
zvgigKNu0QYR>=jEeBg+1N%O=kMISqX*+j3DzVtU}k<aE*^tDG1fKq=;yB|{AZRYsw
zW-S1MBZ=@99$&7nBW1*{9a=nDu2m)v$&dS&W+(+UEh)W%&tolfB+Z7FKA6p{fAVIG
z<&YxpdS%Bw?{*BJB$Lkekk{CzfjMsa(Q=?8IxUexYkw#b4ng#O$Hgr(T=y6OmYo)*
zFK!qtvwt-!g>JRZa6o3NVz*8+O!C^gE)&n8mS4PDVoJ6WZD^(Bv7o3@^JK|NxMWsf
z5s<uA)?N`H<>GpPwn{Grb()9mf8L*EeYjnGEk3~8eSH#f2e5bZJGLV*On@2~d|K8p
z0hEcjel+qMwq(kUj2-a8V12uwq6n(r^G7bP27TZ)aI<XccWjnQ5wmgRYB|-~v_Wdi
z=o;v|-HfCYA>^#gs6_rS=Z7(tpTn?@-rmo5P&b&9p7W_kgm+2m9UW_4e=<qE8y!xw
zPIEAm^JV!VJ-<Q6tzvX-rOW#4ejw|xA8m@Oda*1=x{_<Alo{}z7M6b9t*l05UIAr~
zq&gMe6oC1#1qV{`bU%4M4p4f8JS03@1hGoL-+o{a%pKc-G@MB@D9s)Ffi&^>p!XEZ
z`+=nB&5m*{9%{!*tf5G3f3)HAYBWavf`J;W>rP*Sv>(FgDOYS3{eCD0jg*J`ytMUk
zrf!0r>y@Ep4^8U>-9x<Yp=H$_5ub{|Z|_fiv!8CLS<4KHpKfWt#~-Iex#z}82_PlB
z-RK89M&Tqg4tp$nJ)<w=z(DIGJR>+J1&oj%>lssO-C?=MGx~H`YuW+g9i9>Cu^Y5!
z%`@h`s?)X8YO>{hxx=~d;7Z!Y@oJ(qU{Dlu+tL#eBD-YG!MCJjFr;=O>R|Y+#*VL&
z{)JAbw^HObHzWcvrRa!Tar!xJ=$CTUJtmTebDo}w^HR>c$~vnh*35g~m!9$-G=JX#
z+BLR@d@<>ZO3<4pABW-`IWNZNikQwjvrtu?E_GJM-oUa;r&F8sZLwK3p5(+4`CKK9
zFK0>dbR<3P&XH0cSDm6HSlv0&5=H)$C_S%E+|rXqTbnhUON(qBjfi9c_HKTYp#s3j
z@Lbg{RW3ERr^ZBa_m;BiED|ZccYjUIWj8gK-PA0-O}-;%=q*I4E4@&2VNc}md5Nde
z2YbtLONVtW@69f~TlET!!CfaUDTh=yI4;_kxrD7W=1jgFU1H!nDn%*lZvpQt6y@`d
zp1j$B#{P!g30e-(HZiP(Up>R3Qt?w03O!_l<%)rm^Sc7AYTOX;Wyr!|-G6<v!5p`(
zXT@ZdKV6aoR;U`Mk3$>AL%dHq4Hi~1Vf0l_E*(e%-J!ZKA9zd7q}Ba_x1daj9B?}D
z&JbpNHlVZgc>$y%LU=es|M?d_{gZzDkp}y-oj=0RF9i*s{yqKKpQv8}Oxv#qjD~y}
zTmsA|L9wVyU}q4%`Oeqjc7KD|NSQw|)8aB%5~))?+|UuQqJ7|FCDJ|$I=Q<2*gkKb
zp}Ko$BbIqai3lz?9OR9-dkZ4Y1*{{Nay)gMv!dc;W^OiWS>$vxL$3*n6x-lkDzq2G
zMynyP>N40htLe>i!#iez6DT$GBBq9+L(`1&e{9@oC>UO{>6)H$Fn<yh6y&jcb*F6u
zTX_{{?3M`lZ*3nUZZnH^XdlwdNqxC}=%OQtn*kWdU_N_#Oao6-X}slFmD5&ldfGxO
zEDWc8`;J%RX=o2wtVFIc&h17&pz4;>>K^jGs$fcdzXEwF0UjVfv?!Phu!ERy7q1}?
zZ$`zr<M=$K5go*}<A3v%M)Z8EJBV(Ms+>Q<8r*==Oj&zdPz6ZriXa~wQ~~;Z=bjC5
zFNBSBQI4YlzD`I5Qc-e0SBsSQRIyr=>y5dSbZ(|8UUFB)9{u3joaL51-*{Iip1w6M
zR0YEuDkYO_2gF}BzY?OgB^y&0nBp4)DGaRJwN=%^t*hPfN`DY{y9lqSh5$;-9v*GG
z5OI<(U=HgOt{Bxq?5%kL!7<jv{YO<f!K>`Jnd~d<0AJ#eUx^(c$#kLR2VIB44LNfI
z_i*E8sxNcH`F9cVD=~vh#^?9P0yC29SYd|V#*CYZzQPOIGf>~-+xJ4)wwV!Bbs6dd
zQyq;=RzdUPKTiqpuPMN5RoL#5-u|VB?q!wX1&hmh49qigqJ3I_)njai%D?z_XTbaq
z1I$-|?9F`scLb%<Uf_XV#@D<ev<GM2h`Y!&-N|<T2ku4x0+%6hAQS>LH<tm19Xgll
z_8ypjya$)lZ!e$kXdGxiDy0QM8F^GFQCO5hDgKh!moWJkN|e~Mz)?us+`)dtxw(QT
zi6azFo(g9Pcz$SGoR}M{mnwr~fi{rz^nqEMMsP_zq-TZ5-|)IL?uvUl%eFy|>)Ac&
z<V%_FvPQ?p*k?5srz!Q)c%z-uVP0pifr)y58sKLqNze2;wiv{x8w)wTyaqat64ZSd
zN3f8$j>=<57F6}s+?$xuz<J2XGZZ7#u;dv+729~NKzprDADq>^X8wxD#`|xB?sGW2
zRb#uTv(D{RkQaZ|7V=kw4Osi9BC=k47PV@6^;U@9;jCt&jnOJj+mFu%rlOT;N`&Zt
zmG=*_YEZ?UgRtq)NB<#`_rl;YL?p4w1*JQE>V7@F-2FWUgmWSk8-D+B=WYlNgrihY
z#Bw$K{B!tr_e(FerYEm3RZ7Z+o`B-OQDbh2<|b=A3U)l$QeY#v=Ma#|Y2DcYt|&Bt
zRt4EPnu6kBh;$$L?dSKQyYi(F7h`~bR<Qke*)+{DCW&Sv?dZiFQ!vEmKR!z{7$U7!
ze2UC*?1U>*(`E_L$^w^)DuuaEL>256XsISRPkl=smHL+kh}GGpc=FGO90rNHL$IkT
z_(L$IaHf&C5{^0C{9O}?Gdz<tQ%yS>JtgYzX}FA`ChwWRUYslnHJC7hG;y$hnTaE6
zd}*|8b>D(kV<jI2ChA(`Jk*3gA*&@O6}3xK2@U7HbijcThOyky0h0zNbhgX~TLR8=
zhNVlKA*CgX70!4l>=+PfV)B|O2d+>X81<kdf0zmBOeR?}N^h=w`83ag3{8sBB!wXg
zxh1iLa@^(oj8CLKk<}`14sE@E=9l6X-XmnXD`kq3c;`InGm|M>8YS06=R8s-QcePB
zJV&~vA`tU9Oe+dpH{ECuOc5<%!c;LN(*%$LF__C`2hk8!z7q}C<#MszbS}5svC{Yf
zCkC4cl+&k9qHvF<4==H~GUcj`M9Szih(Zu!`^KH>0hB9-pl~*zx|a%nGk&BL0o9wJ
z%-)Hca)Nf7qfk_s^V?!-X@+Q9Ow0&Nu_P?Vv#9gD;FKlX8K>YSLa)I{8)DH|U5|5a
zVl)w$0`%j1y{zo&yh}48S#PmEa7t^W2d(?@-{42WFk(T@_)#JTo{45@j&D3i#F97q
zQC;lGhSy`ERp`qVD|2Ljw&pokP}l4%oGha;`lZTT67;EinVt6pC5p>l!jVy6fhc48
z9<h^n>?U?b4bfCh>9SC{m7bJFk-!JYJvCRpj3^xAj96P%PIx8Udgo|6&^(CqAHIX;
zN<$TBHc)*Jj+J?KJGj#A%Bn`?=;<!4#y%M+GpWvMG2<p7H}cnibNXm}i499V=MxTD
z@-6XnbAlkSVx*-5lUFM?1y&Nm3;|YhH6F`|SO(<WY!hm5RcPKuwxE5AFd|#GYFQgP
z^tma!n=YjmCugb%qkPqR=W0T`B8+1i)b--x*iWsZno>obR7_8JnC1w~hdXJrnRiCi
z+k>ToJ@%P$XjZL%c<HOjmKjU1O6wJ*yHj0ev9=Vpmk4~)B%CtBu~l!9%T2Dytt3o{
zi9Eo>aL`k(IgwMQO!2WTFzQu+&UH<oc7IfvRw4(tEE;=b(>cYfh6gu-FeB^SFfDp|
zYFO1TI@(CPQ%a!53U2Xe7mxgDbAma?^eKhx)Vz|3V+{0vLTE)9+x<UzbSDkdC_l~3
z#R1k?IIjMwTVg<S!kK!77)aTX8$F#`qiwQZnD^)%F+!`UD<ux+SI>^QLd;Z7qC#pm
zT0)r5kbSz*3868OtXN#NY%S}e`35Coy4)*q8Ii-R|HpZ$li-2A1ZQ#D_#%pzSuc?Z
zOTSYIb;&z_C|6A!v3Tr8F|$CCBYzBC&PsK(^J8SB?IjD;v3Eu6OHCAvHAu&F=wl|&
zvHb1*=hE)DMJJPheVSQ5hmvHc*vV^+iQ{$8>4D1^19Exkhd;&k^Z<^})Yz(iCC76Y
z>U<=G{ucq7SVp*QLwjSnz?mh&w&g;#H{(6wIVSgiY3J;UvNEm`cD*>nnH?pYyGl&R
zGX&RF-Q6zeE*3kf-%*z8weoQ(YdzDIMJKa3Tf#{1t<J@X)N=Inxmsh4D{X03+*H!l
zS@3oYQ6d339*vx&AK}QK+EMFQ@cXy}#KEBT*^1v!Z$bf{p*m0Qa-lF{$Gco$uoAH{
z?Q$i55v4=p`!o1`EEAJ3Jo}CGufNgZ54`yq8*^)he|*y)xgR?GHT>$f(N+ZGXuM#C
z<)&B?Ob13Rl6??cL`b*XfCO^2h>auDKs3l@vEWSI%U%5I0AQ-hQ8=<VEju4?kxXKH
zkS<-1?q3YnxL$^0x6M|tP#E7*9fXAvIC%blZm^IWvDj_16)c1$+UkR_ketv`@V$8~
z)J80J+iV33ZZM5A2Vo%?fl@NRc`WosEOy&$1q+{%+Jmsb$rVa9^yaZJ8?o4JvlT4d
zKm6E1P%uhZshQq96m}yNdrh{21eKV=!a-0_OIRr(-#io#8==^3vt=y6hNSZlEWjFn
z!rl)Q=WS5zH`xjlCghyQKw&iD)oT-P9gE9GEOy&$1q<zRPH_+xS`n__N1><6e#|Ns
zyKT0Dg$g<6F<2-`1n_r*#m1t0zs**#kRj(h1`FvnS;V`+Vq?+0-)1XVh>&z1gN5Kk
zSov<S*jRM$x7i97EM%R>V8I!YT)!KCEH)P1`)#&@1^BG<7$_JJsm;5AVq?v{+hi+9
z(2#T<0|h>)AXgXsDgU{y`TMAUGbkE|@BX*(|M$1gjR5Yp^csSKaKUcbin&_Ew%sQu
z7M3&rx!tFJn`_fs=R~QTCGv-p^lz?}Hv=b#Aimx9Tmss^N}S&cEHC>uFD>UjclqXD
zS`OamNJIGd_t?38`8xi&a!x5G(N3(>i9W16!-zoQYaa><=?!mLhgM<VsJyI~kd9|`
z?SBDPsM?o7!ZQ;CG%_`p0sS@;0yHw0@mB>Vm-;3m41cgzlt?{*g#mkZ9W=MJdkb=E
zn_RjmHkbbWQm>Il9*@0t_Anle9whaNq&~@V_%X=g)8_E9c{pkutZ)QY4KLpZ;0DRR
zAi2`eU;%{DFuc4DUmvAZhMxmJD5ci?SS~w$({8K5aSz~DfZf(2Jf?o?$j_epsL+qg
zV;;vnS$~AohR1s2Lw{iT!16%vzP)_j$ZV(--h^U!^<U2~o3~AGwlP{r>xMsHH=zMF
z2<_AWMtE(8pMMYEH(z?GEhr8c6lo063>1)z6tZ+LbZAPT5YGa6qGZ#9JOSc~67a(l
z($8r!0$I%T^Yh2h)%^ac&I7#wJkbO!6>tX4G=GAuFr@ntD6{URsRF;Dc`rgq4uz4_
zLrF%Ea|OyGp)Qz<>KvCBR!2J_lgO{%<nR|ae&){FyWzjs^>tDshku4Y!dhJr%mSle
z6b|f~U^yt^ozG&c2-03i^zRpOrDaO9>$r8X>|UOFxvTyNLWMO~mc7Nyyo7y{-7oqg
z9)Gh>=I`j2cDFS$gq${w2UGIc&`XtX%57t6;~+RYM*FPM<Y-X2{W*nw(#$gaE=D^S
z<uMqd9{Ucd)yQHdaR$LL2iT#E0aHrUsf=dW(psmTTgO}%CiIj;q^{pA5GK#Kc?mDL
z1}NdY|JAPHg*A$W8v)K1*U-EWFcwTXIe%<HNetkpB0bgwa}kf5?Y@AA1mW?6;E*Jg
zax*wEZ95qZUBSQ@OILz%tI_LNkmoTxr4IpvJPKv}2^jmcU|i`Lw_AMy4QGY-`kwUu
zZV}h*XhE!97kh3;>$lD;NEz|wiw!9|?!cBLXX6r`m7t{H4BktS`wOF#;4QHmM1O&|
zCB=ul6~y#km>}X*I_A904?frSv<;=Nl-Jww%d{#@HBqagKYj)I`yE9;eOe+tmsukz
zCs_9fu|{HML4M8}3ov1N>N1m*0U%rZgm>CN;R+rq#}zJE-V96+ivoVf-1fE+7JO`;
z3K6U{_HDyjVkPD`H$MFKs^#$XZ-3+RUryNTw&>bcWf4-|Zylp%ja~`_rU}B>CJ<Ub
z%Y1<+k9+nMocSD{_*|+nd32EEa*r`fh4mmQHd!T<r8|PNlW<*dc7vMW$lR<6%)=Be
zCDdewCV0v?)dnE^G|Htuf!hTFmTP;!WCW^6*v|qQiY3X5Zw=E(d)>*2gny4r4)on&
z%Iul5oUcG$^oUyycAXX_R}Xc4?=A%>-_Z0z^3A(#;~5F?w$}!*Xd6=kfdseM7SdQg
z6jv|iwPnW!-luWmXQAWytoDBJ4U=KhD>;X~#h6>>or5(^#--{U(Lrtiomn)y=v5{u
z1p=9{btd^7nAt-M!gQ#cihmrN^dEAwpztHJakCmW?&0X=Qj920GNgI{(U)V_%<6N2
zrkocc&K6g?!FFc;-KOi{V7j+_tbTOCYK^SMp9m=D%Jz6oO$0P+ktcDIZ<utDf_h<d
zz9(;u#O~X`9?!AId0<2lhw7ZPzadcgG$U=VF<dApSdW(Zh<j{cN`H;|WY%Z05qJ>B
zxpVsO3Q#K)P%pW|;aw6H39C+Lrq2Yu7v7%Ctlpcf?@$+L-z7HVwBYS_IwG2eRGGB5
zqx8*Ak05)Ke7%fGt>IvhOUFi@EB8s0OdY=z6fUx#ZzTZ&)GY)l1Scypf2m*OTKUV|
zC2^;0Uw5#{wau1RZhyL3z*90_#A0E2(w!PFrVlo0KEfon)moXj*`(i8Tqe})Y)nPC
zxm1gN_tQaYmSU#segR}=8Kx-6OOI486en7@&XCF(k!IAgSysw96m7@Y6MYS>dgYX;
z>ZuT0dXQm76)rJ!88;Sc+*%RGVBS5)R6pGsPDDEXH(}!~$PdCSYKE!A_Vj4=Kd5LQ
zHJ2f9AQS>MFqZ+O9%{FZ1|LB~0X3IV2?7(BZrlVYmkLTBn*qL;rb-{Of1G`4+h@=U
zt?fxV<AY?>2LESmy)SMa1BEXg#aaoko#3FTFVlhWnuK=hWZ_r*tME2o2f`~W7;k;q
z90;%YQ=P5(75`52`dD}%A!$Kw4}@oaXwTBT^g?pV9r;u5Iz@Q5Ej-67NL^oc`@$2a
zHkRRJ;RpVdT?buE@7MsKe@y`7K>*f71O5Pjv(eh80vP#Ib{~kzEdnUM?{}?0)(h~9
z7A^-MLSd2SWDo;?N*TIXfpQ!`O^Pc3#tNkuEnE)(aAhzACj%JxQ&e?l1uSl~Zv`U$
zT(mGBfFN#j{Ml9*_)~P}3W6FCW(~r!2Dpk7YN6Gl39NZ*8;JITe_OAUfebQG^oXt=
zK+W<i0GexjpxXdAsKmsn+q$zYF~~sC8#;HIOIqEx1aOnW`LYFaP?d?(xYK7_Vvr$L
zv%BM>0O8&qO>dEd3Qe&Eq?3URGMHTC8X%Vp#O?w?Pl5?`Kn|)ka%%S{my1yZlgmZU
zv?iB{jc%i0a9R-cf5tc{+Q^yRo0E+(%3%6((R!8L2Elzi`Q8u*g&U3I_UvQ~g9s*9
zjEuoH81(UM$@&dn!^il)4ow-6xqB7twgFCffFaj#Hx5R4ZKgl}J$+{c1J!l_cfh!$
zF^skyz(F!tWOA&t+Q|~bqA+Xn+$?#3K$B)qoWszSa0z4zfAya~|Cok5|NLozfGC_F
zc<}Vz>SsC2Jq3;e$!liHqk<GNuuznkc6XTRmp3mrk&=$K?+GsxW+%<Ax{P5Xs8PDN
z*JmSDQCyzFqXKPGTDV8>u;P?-!gAMkm*y0xVD`b1G_D0$+&NrXm@R(+{)Fap(;vAk
z3qm#eBhOhDf1vndp|E(pi|ibK-*9m$V~`z0Y++O8^IL>nB&faN+HzwCVePS#j_Z!l
z=}7dClekkdN5X@<k)+r@G}1i`YT#hdOxz2_?$agQW$V=l)>*hh{;+JfFteVN|CBWz
zC2^kv2C`Ov#Q#P|zhmdAInx<hG7;toUG4dn1y@WZe?I5paCT0~N2imQ@T6==W6s93
zCro$BnlQZ0L=&~VT?@}tSs~h#yi(S$)C2!($Dw&YN_0v21~ZcQJkXk3nNljusnRxY
zw9=UR24)S|Do2B&OSk@5+OSLQ7NwM>%ykaFZe_&Vby*pX663}^$bEv<@aej&!-`4!
zwlHI<e=BO+1lf5WF4AP#9reg@uS^?moVXz!PKRcJuC=_Z&~UdR>wp(MK{uXVeDZ4o
zAu$t_i(eMc-(w(2fP$`%Ij<a--elmS<hF{6LWk3~iDyiU>=FY}1FkObA8Qq)0B%eN
zZhO5#SaTD6M7CfN{jJwNJAvaQr#!~)sC9<Tf4^XSO#8X_8o2Cf-=Z)U#?wKb)ULs1
zcl5a4Q`dE$P)Y%hqkbinUy^1_yl#mk%t6vMZL!@YnHt|HUU%o93szPMQY$5*&p|<V
zwZY1T-fz#|jea`-pB6qO8>d&Y&x<?VkXs}3@oNe$XHnOLAT^=jZe}*cv#nYd(jjy0
zf1P9tsHH&<nona@A0RdPWf!H(2LdML=z(J>{R)nQrhnrzGB_(xuTbw*6%WEVx5xie
zVQPf}8YvHYc!;7RVbvO$-idrKyj>x?cXGTju-$4km2T^<F1^v*c3Lq2l99JMEzGHD
zUV6n`sq*QxlgcZduHPKz&kuK6{0Ywee?=P*1w1!1mz@>J*{wv0&7SLSx0OrZAmiq6
z|JlJFh?BdqXAbt@IJNup=X$Vgd#OG%Nj^;CPRYVlfcmeQ>%s6`ulr?i^S8G71DCyT
zCGc@rx75Wd<{<oxAs;1ir|U9VOY}GG1c1wgv2~o%B`4cF$y(pewy`}4cU^n8e`NJ^
zR_1eay;EtHJ8=?TQ?Ga8Q2EwO&9PPOIOH+3OmQi3_Vk;Lge!w=rE%%_VOx7DzE1A4
z8;CQZZ6vs~IAevXcW_MMvTP#iiM;uU&EJ~hN;7v9g_VW55wXj!b>iFp^%>*u*jf(x
zPSNi5Rm)s+gOua`W*GkYZFzI&e-Y)&xn}+Arr?RRmUlIaXDU0tks`y<gCiArU3OCm
zXWbcH${@1rkMNqjDWCk-_9`Qd+8VMsv<rbs#+mqM0#5R=ByT=xO0y=;aSleAgYN23
z0NI<b9q+uS%9hg$io8kND($FH5Ulhp=luIduX7%VHuKE{U45mosmV<Me@7=;wdqmo
zD0qE_mwkVg+~%6vJ60cqjWKEQ{AZMRTU8d%58KSRFPN8HQCpiiodkVVMbjyA`Or(l
zL1h~x{znm#PLypb=A1OQQBtM5(VQnaDh}1+U$jv(`L(g)rDW`uU%#2nnlQ^=W<$cG
zT^in%nY<Pbbh{`yT$jS;37<E)YxqB)+~Jo|sxcD;G%_|fm+@dd69F)n;Wh*&f2~_d
zZzH)CzWZ0`=tj^SlJ9Xt908KrYG7{DF>;yADU%7hkv?QD`TKd?A6X^Z&gf8<RK?|c
z`@*t3{OceOpC8UI4?kQ14^DWaoIbq#c2M*{(qAR%rZoqrl(3)<FW(P8KS`-AU1z#b
zN&vbXefRWfjvqmJ8lmW^Qpb-Df7MeSr>}IKXpD~I==-J2qmi_P1WR*+4RO4p^%bqB
z%<-3(A0K1^dh7G~<>5aM6|xQ>(D;LOT1Yg9e|&%V`4@Tkl^*_hkOG1~{5K8#9)~Lt
zj5+-K@Yjc*R(#MQ5dDT0HJlHIfFsdnT)vPn8*@}HJ|pTi7<@{%bR^3Vf9ZJ5m$8Yo
z5xP?X6woe8c(g0%N}@pT=zGS#L%$)&LEVRbkRZGZ2WvrS=bIi}BAuCb73gwAv$PN&
zg+iz9EGk28(@l^u==adQl9|^+drPZI5j0lJ3U>fho}m)}?oK?TfyU~c{Xc>Wp{jW@
z6uKb6+JFZy$SB%mvj7i9e}LC_2EW<Afp_IvfY(kCy=}5xfY<b?@6Gwm{sZS>4PF@`
zO{d%~z|(%{?!|c-gd~?c?{7xc!8LfYS0$lM_P4<sa&3g+&fus0&8Rw;C3q8*P}rP|
z3JdVwB}v~E{Jg&@s)Jrt&c8x{3qplX{b@l!3@FmxnSg13lRfi0e+aNnXy_a`F9@KB
zV2!#n0n`2_sqjF6T@zpoeRd9976cedZNj}JFzs)W8V>}V_6cxAIYn)QAV7BxU>mtl
z1O!2;)}4u%FqotX&4|sK0HBy2*JT7OBCjD=?fku^FkvuB6W(G6l?X)%3He4&#3BM4
za@ij4Erkg~id=89e}zf}<?v`nNh~6<NvWQ>GZ7O8TY?!Q;=CjRgAugIK*S;<Bf0Q!
z=ddtCuwz)TCBI$J;6WHO3StqMk-T`YcNWBq!EQr?UlT#Ox(TBo76BTqrxfhYG)xfe
z7#-XuJGxaj{%|FPKCWe9uj)lgeqB$&K$Qyb9Lt%sVJzw_f2;kV6?T1(8<jOByo%OF
zP3*8WQk>;j@wU;Rg`)+=(W#a;x#lD^ww9RW=kQZTgcV(^)M82I>qrb`u9`;aT)P6s
zDmLA*ZS`2^>iUef$<I~!tiM0;ebTz;`A*u!<<bzZZADcu$wI}Jw$R4z7S*}yU}TgG
zoXXBv<z)wof80CJrJ2_}P8Fnhs`Ix9)2~z6n`wyF5nYU*vG0-gISqE!BrUz5t4bOR
zs>G3qUJd9gg{k+Fq^UC(TZf`>E795JCRwIA10UL*4IlFQRoi7*TjhI1+pg{wO$YCV
zm->yGK8_|?9x=C4iUb|gy%Qk=JyQW#B$BMs?dT}(e`tN0mFX=K*3+(0ZnBN6+5}K{
zQ$-gq4MUSbky1mJbS2N+Adgb4vaR}!zH1)(tn1_rOGhrtqOD)4>MK^ch2k5ol|K~)
zVjEknZ|rQ<wX0@dIR2zf4SUr<C3p23oE09M6HmJ8yDve@<l{H`-nG<r4`BR`Jee*J
zbE;N9e;>g-6@HwVH)%BI@kO^&#GrgkYpU6HA_h4XDyDngRKgG`2z|(?$L^?-`6rda
zQ5b_lC-(%FDS$7~VT9Hlc5-@9p4324XUamry3q@o>QUE2US)Vy^UhAsTtqRC?k3#e
z{+i=BF%|ZV2Zh<#T$m}eGL9r?A?N|sg%kvhf2C?cQsqWI;&*Z(`%k`f$D&aQuk1{r
zzB1F^!4&0}g6VbFU1>_vq!06|WSD;#uSXG--r%h~KQUKRdaRj$YeOmNLJkI&8`mT&
zEc)WrZZ77D?f_f52X5WcHk5{(fEs5jIO`T<_jpZJkZ@Fw?(;gC=1Q~aGl09P+4v^S
zf6k-0X6HM6?1||ubKd4sM>qFaTIX%1VNR^tL!M#yeC=4^y;T1ZpN8r)YVfFjLiWQ6
z)SPK+cc6n5D%jiZK)uNwXdZnEJWh1aYSPG_8I?QTf9t0n$CD$xvRuShmU3jp4F0$|
zTA`_M+HiyzYTx3|6QFnb^E?BN%9eCje?Q&y>l<64^f3fA!_A25+WJ!#TsnS(#_X-{
ztxtHc4Narcd5_}704s-@zjuRxhLoTxoAZLabi;Zr9Q~AaA0^DTW10P6-#^{6aR%-r
z$z>f^Qlw&~H8Dgkp=(L@+wO@keoX0jno@SOnTyG49rwGvM6sz@jp}1lI-uiZf88>#
zwv?<k)${iP1^C?p{L&#x0*|_7YkkEk-nP<~(fDpg^Bj=P_m$;2<0Tib_C;@P>e~)#
znrqZV8+MiK%0Gm*o9u2Oan8+EbLG`h_Iyz>o>nupEQ`PNFj>``&&_pH8&f=7s<`!T
zEmrsQ-B8dqFTLRwdSMWUQ`jE*f0=xf^4i-;K}FO;R7osG)Ay`~NgR1TwuGUjbWyE8
z6)#KvCc0Gm_&og@dPJ2R<sIb`7@I`#0miParwTYwz6OWi|9SZB;iq}k+Q|`8M&}*>
z(6{8thdW=&>kUCtn3L8TGlV%MDfZz(Yf`^{;&UteY%QIX6Kz)mOTg8<f7;|_GmBn_
z{VHY^j~K#WG9@V1AYA<HtGH37mm8)P<I5T`JB}5Fe!d@Wi2~rBu!jY~j7^DLhTaXm
z3LwC#JHnq<1YoA<dOLAf{a<TfDuc9f(?pOgTAimRj+^6O%QnLiCChPZsouiISaQE5
ztyB;*_}1Lf*j;RE|B;i@e*t#lce#7wW9Al@4*H2@;-NeEIzE2Gy=K|9H$K?<Cl>1=
zwfr)jL`>|`Ym4KnC-F5IHPHF^`9q%K-2RjxocH6DzZ0x7%j(@(q&${nKg{~uhmXcQ
z{`td4W!>ZRa#zah$YtSXSM^~fMhXeS*jb80DDARX!3U%omyk{Hf0uh3BS7N3d96q$
zmFY@UBY9V#(tj(z0!zZVDR{oFJhuqW)Q9&YYb=zPK{a!Ruwn(0B(<4X@w{Y3JgJNC
zLjC`&*h?;AyVmvd#<8W{5|k4w+R~HEnv!z#mOXFMWAVs|J+)CEvWH1cxXN0<){?6!
z$GJ7^fu*3gjt17ke_R@r_X5Mng~zYSmh9uV4=bZ5trW!JajIULFDRNaP8f{$>v}_o
zZ!V0}?b_UTX8dK15Vde}BI?^K7D~aC!qluto0FDNnT|`~Bo?xw5yq?D5z;BJT?=!6
zLyYekny}eWEhU2S#k1hwqXMXDO7v%Z%G&M`v`_|S29HN|f5>C1?v85HZ|z4_7k-@O
z>Ty7S?pc59eA|Bz4_rU$=Pr{SA6>>ad_9Crh3jQ$p~}isD)N0Ef2}-*gQbW9eoVEj
z5$JLfr%O8Y_;qc~%26_GMsk`c9qvx?NWmD#DPQUozl)4C)JHX=)Eiup{-2i8SAjf~
zgPG4aO2t44e_+qEW<NWYh41L`=k@|Cq5V)0@8ohf`_eE)iLtk}dstapB&__edAMZ9
zR;GJ?hIaqRwCJO}uJaY^g%6H#^%cznM<c104O$p&C&jzv>VPC?`5v#hXc`&o<+p>1
zZ+u!efci#Z=e2kfdx{#7Z{*j$O_76gflqi_C`BDle^vOrSDYx2P55D>bry+eg;6+~
z#|;y5RL!&Vm)b!!YSw3?-_;N{^1Th=yaAN|FSywdF6wRqH_kM7^Cm#T(427Y?c@{O
z8Xoe9IXuv6`M9RFb_dE-4IV~6%{{rsx}7y|yi-!k(Y)?-4nEDF%{%WESa0;w`<8_(
zeo7fHf4h2M$;|p|{F@xp7qOh%d0M;E^Vl@9H6Vw!q}_PYv90KK8<ng9+&s29_C}<A
z^Xnd^gV9^N?Mcn8)#{|{9UiuwH`ZtRJoVtOOlcI$Th#E1OU>%XX7H?5`R+=sSQMKx
zoPLtK{g<Qv0bior&kAL3WOH<KWnpa!Wo~3|moRW31pzmgf$BXImkVkxD3?uMAD(~t
z`1j?vx1Szl5p<NwIutZVl}fu8WdnUNKvVfB2JxXMDUlc<O^DaXM8;p@o7ecZy&)-#
zc>(oYXiOZR>Eb++Xi?hu>zUIbDPmNfv7?XAk*6)AeUuHc(V3YRG1AiKQpcGvzm22f
zcP<f06HW8o)xs1+uwaodoCYt9)}4QePJcfZ5_&I!)82F>8U=tO5)cM;CVBs;G-j0z
z8Ua@IY~YdXq)&eoO6OK7trS5Hl=5*?Cu@ZFgHE5exi~5Gz(j3p<Vd3cJ;|S0Bfl-4
z55hz5m8bVlk59_l)3zDOt%P=VG~*UGRE~xY+<qw}>%oj*gtTMSWVd!aB%*)YxzW8-
z2wjRhV;LPo!#39@=Z(&fp}p~iE(grkN*U$p<zRxZ6nL9fFaW~iw#jes(}qm07yQ(w
z30vl|Af$(C!(^kBzd}hQEudYruB<5`v?rE+<~MOnOc|6$7*%{KGpq8(<?L!*9HY}J
zu$4M5xs(wzx1ID7M<>sKb!mTNuEmrKO_?av4BHEm1=q!C`1Hw4zL?1t#Z+UUUsQ5R
z%BZlKk?Y*eAscNP=bO=FcFTSrsLt7D!Mucv<%#aFSbiggf)+v;%Iw$tNt&DoJ0xvt
zh}8We$#AZAXrs6deF-Y$BNltrnXQCO&XL{N_)Rwk6wc|kfmR!TE|P!QBITv7FY(8?
zuR?giWLsaP<RkMs9txwB_t*x`+{_|h^$m&88Ww{2DF;(~lFXhL_WYU8U>4k)uc>0;
zG@^~K0eypd5zuLzd}YYM-;j4H36>bKWj;CkZ=s356y55t8Ik1oF_?{nbxAF=F4T12
z&0HdG);x{5Cd9H9AtryIK$sIGn>q1EhMVjrl2NFI39g&*^#1q+&Yhm)AKx<l4QI0`
zg14hclGv-yo$%nSu!@8?K}gp|&hAqM^=pDcFv3PYHbO6(1^xFX<ccJQ{NJDxv}IM@
z3W;4O@q%D-PXgP(PHN%ensAZ|71njpDhMszJ-TQr5j+lDJ~n@K(Tda`Np^Grf{hAC
zU7oBueR_ZX*am1wU_pg$c~8G@vqLXv@rlHxg&HvNIvYbU3slc$2YM42cFFPXU}K$d
zdSLUfL+Hg25R68wX_ujjl;Tqbw?Kns1FKJ-<DUFfHu$`mw|R7#-Oeaw-pDyOM#J#R
zAWeI7Y}54zH<5p$UD({xxYnv@R5x+VX?@Hn9)zYvLhl4E*qs?gOC?akgux15s_(rh
zBoYWXA(b&`)sc*%B?_gWC)vv=L7-PX$zDeB#6{hu3zay=4ub7u6d?8OeYyZ8eSJff
z^~j^rAy9He+E6H@j^z<0WQZ#Xw8Ox$0H@reWa!!K<Pm?Pg_nAj%@1!6|9e=d5NlTB
zrvH3<`1LnA{Z2prG*QVM{Pb5E`fVbJ7-%Xv{dfBE@N;e!-6*z^>T}9C6c4HQaG;9S
zv5um|$K#_Vv2E;%286@FTT%^S*ctnDNW(F-v7I7OqabT8k?D-PS=zwWx1q;><*h9#
ziy^v-@pFGFbqrgUgmYZCE=BPKI7`8kC%wu1WI?1?R@`4^uBDz>ScxK31=H3byf~=b
z%`9ugo<)0xe8<G^R?|GMbXw&&AmBoK00IC1o5}raQEi(?w|QO}$9dzp6(|s6Doj-B
z6CP1VashZNz~iJHK{z>LiuDKvO3sq<cgN=Y?FfIS63#?6$0gGW1i|)RvX93n<K~@e
zy*CHsZfJX6J54!Fnb51A<XBTq6T}Q7=}tN${HDV|Lg>_(gUC!9s!{BW%vKAtE}#PI
z5oO1!XvF4XwFF0C&6bSw^843qw<By2`&?YEwn*!$0I6aKTn?%*ksIq-&?M5e^uJps
zh!KA{=Z3zSxE*1RcY*R4J9K_Hcyo*VgI%ohm{**B35(ql%)-!!GdCCIGM^Td{a~{N
z>}hdazc$;YFb0B0RL%`+Gw&A59GtY{-6fseYO`Xm&P?@|k$Hzc{kZM_R0-M;I+ACe
z6#cEFNJvF1LU4eG3T~o6oeOU;cmz^X4i$e&_Y|<hgwQc5(~)dLN(dA*qzPG0nQ$nH
zQc(|%!c3CotT~k4wbDY4Y`)K}6xhy%+EOLrddi*5OOQ7p^dvj^JtAIIJ;{E?HKwPy
zC*eYD|4UflyvvM@`h?rg{&HsMBfS;U9KDp7Z#b*Z6%B1b!8-OvNem@7*oNg2{VsoQ
zAYY^{lcJs~gF~Jb-Inq<!~+slQ~vw5W5sO(b<d?ZmfYRaH5Eo3<^Gl)U{Gntv2+k>
zB!S)P%V^V++~rw{a;<~4-ZsVTqCS|r%Vf-5oXJlS8bxi!NW)IPySk*GSr-4it{?1S
zIb_O5K!AMaB*kY2J+ek=4^#RTXFq>U{_V=nE_vU)Jau-#yN;I91os>*6^bKAOR}}y
z>oL7$uw!08b_hdsYekX@fdXmn{cwe?F7wPlsJ>l5k^J_|DH|B6Af{2K#F3l(de$__
zALc_1Hg8=nGag*bXMx(`O}VOt@;Ji)7Y8|cof?gs=~}_5Mmxake5{q+^vZuERP*U&
zTHicxDFj=o&CBaNf<w?X38HsdcxIlnu+kC~HNdv;fYjY+w`WQKY(gk1tOJCOat}g*
z5sqdXnBAz{m6HObzS#oSt1V!?sGhZ3tI-4N9@LQ0cURBa@f}t#cGx`Qxhx5?<>05Q
zcWF~Ag<6>#ztRNGqt3HCCYOKC(ilDyPY|;#U4LihxDk9Q3wvB#anc=JzjTK7L$!#y
zh*u8m_Q1O`UI6;+d0%$GJ8hBRI}(<vAYUW7vP%Z#CRX_c>$lDhHGR1Ft70}QV{g_@
z%x21ry*ri7K^SK!o->cJc!t%RVGPwhLB6fEF8MaV2tsODH}nk{LqmVj>z?E*O|sKt
z^&(I?(ot>S6w=NK+VRH~oh`?srdaMk2d%949XX&-9Efq~Z}nlb|H84+3J5Lp<&4tx
z5LY9S_Zrgdmn~#&4-~IPrK@uaWGbzWRIfzh2*!rMZaZxkuT<#A6IM>KQ@omx&IUQf
zm(7!%SVGyjSKXnAr9FR2LC0jL3U^A#fnrAyXNC4^P{hV#;zxD**e%CHaFp33)v)?#
zJ#IsEvsiHXW49-%9!IL-J2qb_@N(s}b36oAI4b5l;+_1Vi65$-<b#+25L+p?Ydk85
zqRt)h&Rt4NF0KX<cZxe3wbFNKMGGCF%wWa6tmugm^hEcRcaeXoYmKV?lA=FXL0+_6
z+4+Pq9OUOaRvcMj$xjy!>g%m64r|-7yW()4UG}@Ojw^PVO?7W8QpvKcTqrEO??HCR
zZrp7yZBW7}&5gYm6#)f75l-;S&edBGTG~|>H=3x?1JvLEUG2glAM(mE!xEzkl*<?U
z*s)@*(HJmkGp#?l$0`tQ#!dCr`X*>mXuxGjI$pcD@3hM;?}pVq^2{+9@0X$PahtK-
zZFrUb|3E2X>i+|%e}8y;_y=+%2?CcPa3B<yk+CrompH>TDVJwyA25H1YIUi|w|sA<
zM9S0eCwcnvVf^;t(`RE(K}4s5K7IS`r054pf0d-0-kpL{!W(t^_WktBi<H{aXQYpn
z(ir*#`ajauy?!(*(g;PrD)ss?SoJcG)1T=x&={S^(f{W%k46#+X*|tMKIQR>=qsY9
z!t1Z!e*Pefpr1awc#40*rQpq$ls;HNG@F#fffz|g<SlWxasEDC6@7@iwdI<}>U@9V
zcnuHloba~ck@c`+q3m9D*1^V%MR+MZR(CU3H)O8u&QQi?S$suux*O{^>&qC`?!=;9
z(RR8j(5%gMG}q<Pu5j1$cju7bJimU^Z}XVRLix|<HqI3alkR^_2fAZZC;K?loFpH+
zU(QAHj%F`!o;`6HI-NTi8<$DrHYOEmKF^;<CHvM}=U2_T+6d*+$?G7Dt$s90w>{cu
z5y(CI(X(u!`U3N?wAjm?JG&G~rxDeH!{R}0tSXd;c1M=kq0(-@giTy3^}=<szFGcX
z{HF6lgFXy$Rv&-zs8uL@s!T1~&dQ9s$Qh3L<>Yf0twk|uwK;07ZgkNaBXyEFa4Sr5
zg3gmy0a8b`X;!e^w@U2|{i0b&6{=yeP?c`F=cimd=Pk;R7xI>iN;sukhQthL4`n{*
z33cVynJ;H=`?C1BbBo_kv0wdZ8nw!a%U!=<ShMwn?o59@l#Dr#ulXXm7w}#+TX4co
zf?v*as%;U`bFo)s7M*9*Y%_CNX`CzHWO2V_;uejzo9h6NKybfHHL?pP>izYjH5w?_
zc*?ofDzoD0T2&~#O8#?AVmYq$Y{(S8vLRA(cDDY?+D-}OTb-*K^|^dTtH=>C{=vg1
zJm%=HIc$j;77dqwGWYD~QY4=PJYT-BZ|~ne{CA}Yg*-Jd>GaR<AMy{@C}E;LDdmLo
z;q=G9PrrTmJ11LGh!m3M2$%oRe}DXbVt2ll9x6$S9O<b!=psi+3MYGVnpzjzB4^Sv
zLM~o+AxqyA)h1<|gtBYO!;*DZRVd@GuX~5Bv14D=OCKnIZk!z~56{l)rMeV+x)g`l
zj6NICX7K8t3DfS**Cllum4wlXtEpuN(oGbpWJkjZD!W-Z3{?&pS142bbP<zDQp%*f
zDO_ybz?22&DsrpZWLXgtC&;vJgPjp3MLybWTIF_8LKI0mPF*Qx0iSAU(BO$;cjxPu
z{UC`{R>DeuTO*;EcxWbyYn!mNlFgOpCj3YMY3oIzSZ7=(?>_9$`>r6p@H#eK#@lf>
zp3;=!lpgT>>Qp^JMT#nuKlc;%N+cOLWx3O72bxb^1RKpkuZEZ-Xs(DQ-KbsIZKAaq
z`CWzmd^hZj(&Ml*f=`5<QN98@8~q&EqZMK46vF*~sCR_+@?8P68*AQ67O+eEDND67
z<NW*7Fx`;>39)1iW2m-;)6xp-xF$`4w=6-!@fInxb|PrYu~v7xkJd?~5V>0Jeu5?n
zOPRDMDBrhnD(ROE&N!VPvX%~7IIRy_i`GQatgLl@*jh$vVWfGUwG3rT=fY)cEnO3}
zor+_B(A2oz((aY&+Kw@`Qd>+Psk|l!yWx&!pCDaM3^uF?ubS#3!0$@$g76#x);`{3
zSjytlDy1j5oy3jzv;yb3Z_3P5p3X~2`}R1HF;GXTek_m_SwcDYJS1&Sy)qp`(#l9>
z&}<;ta?btuwYnIL2u^jiwmfENkA)5{_p;Z2Im0RDy0TR0oVUOa!sB|B<*W5tgk<nh
zz}zn!lf6>0bPaC7`u0kk(@!>9PwA27tXit4NX-)O%vRdu@(B;aM#O%xgu1gKDm_lm
z<jMk&jehR4faw`iTkW%F0j;!bJws7+RdC&WNTZz-l)Cvd8iGN)RG-mQ3X7%V#4gf*
z|7(&aQOUF_)BNREdHNUq_%jX8Nd0db`n~ayr~jP(ksktGAsorOGhxD;g7E@jdD0@0
z0cWxc31^a^dU-pEW0L7yOsv05R*}+<)(wm>z7gCDWJT66R32yv5p~zJz2|16yuHg0
zhf-PIc=&|!W-pbGD`P_WexCd*e_bMfsZFdzV_7>5WwI^5=j$1+35W-b#)+aW6dh1X
zFmnX46Z}5R5&$(*C=F${DObm+)ZZdd99FW0`uen7Q;`*^8JK-86wNZOyfRrbTogK-
zOrScqy!)64RuBwRyxOIXiRFHTx3lM201&;3qL?uf17QRz!(w1Y>0BtC0lENxE-;qQ
znE~Zj269X;^ghdA^&tBIzpv#R!6otI;eV8G09}Sn19__K-WCG|P3#mcCSO2r1xQA)
zQ2?7bW}|UsCkxU?u($;r&VYdf)P1?Prf8Yv$3m&<p1?y!u*9&0GedX)ggtN>8HLIb
zwB88;r$v-Yjpt5b0cyu$J`@Ijm@m2--{4W8yxUM3DTh8susjgKND0hqAPXjtDS{KH
zRR|0aP#)y6@GL5U$uUpGmAMDd)B$c5WQdTN6NIU`n5XPq!O%znn4N*?2>x>hFe0cN
z$`XEQlHb9PG5cW|NWg)+#<-aiM#x#HbYx`-qiKNOd-ErzwPEoW00jqs@lmA+yYL!n
z9pEW`fY2Slb{J5;SGw~SxAKetAVg5M8qxt3-}_iqn*rgVRtJ`-(G@VV#>KOlkp`fK
zjY(h{53H{kR#6o}wg98BOeJujv?R|fMxtd-f+zG2V71^g2H+dP=Pjh3w5Y~_7(w(A
zjB$o2Vl}!^9;V`CSe;6LP^T<FjD19eL^VRXStnG>A}RueBTJ$IV0DX5JMdDooZv~k
z2S%P1l+arry|gmgX<VScYocNt7#3}7-;YoO4D(hj*jQl<7;Q~$WU`H{MMKYZtf}cE
zX#i7WLpQRTGy<l+Z;`f*qZ6%URG0oORIbSCO#~MTkP|R0);L&y=K2W&W?(fjZsdow
zYxQv_vkidy0RFXKqUyTBw3QHc7y$y)uh#&poY-3E!XwA+7W$EveNJ~Tsjv-!wBJC%
z?lXYeH5P5>G=^1O9La+Y`^b9#eY4?u6pl6AQ(`N?2j((o)?=LE9wv+$z~d&=!3&xC
z1(}0NoFGSV2970vPF$G1gN3UC@9n?m8uW2s*0QesKEd5HTdqcyr3~X_C+)Z3_f9Bd
z8~hLZX4h;#Y{(5@9)e-*=iRHVPoJ$iIgvtekw5XVE_F{2{`$9{$`rdjMbi}DPx;GF
zhq%BJ_$U{ga*zuw1CMgSDF?a0I_4-BoN|y0wDU)~;FN=ZT%aF4$_1w!<N|HkQ7$;;
z7#HX-4l=<R2YEmhJ<0^99OME8<R}-Ma*zw;$)j9w%FJc!>2>!SpgtFW<N@~OC*v-P
z=)xn<>~AJ<>@`Iu>j}v_UQ_V*AqP(L<ZUEwd3LYUt8us&UP`{_c&Z!yS&%kP(&F!h
zpFt=mMW!`>VLQP>RLTxuSKl^ODs8v@8Fr1^x}Ummc&l`_<=Z?7A)9QjZ03l|e%pLP
z$xeHYyAwQWPx>+w&OvmiY<A?bj-ASJWj}g(F}TQ7-@?H@;Hp7d<~D2`oNMxpUfamk
zux+OrHurdIWrWwuj#ZRj%M0i&CtC7Ib$w5Bq1Q=&IN!PLq(j*+uPcr-4Zg!1&u?he
zYc{8_=?&K^N(@DZ32R)&i{~TqE@7|YYL;AlWO-@YA>l7e6SPn^aJt5qb7B(CwBcz6
ze^FYitnf*3s*~5xSPkd_ww}6}{Xx}}Uv0GNHh9-@ZI<f2uu_e$^O>;`&gh*X5_bGm
zD&a$a%U;KQXXjkhCG_d5ux7>6!Dd2u#XoHotGzmjl;72tZw@y9!S;-0d)uC+v<t#f
z3(X|B_Z&o`;_oAW#h%CuZ2D79L8w55)P)~#3W8HE=Xu&G!%%iI1UQGIx2nc+e&*1|
zSLLa@ND(8RRiT?Vew4i&jPl;jkFo`rjW#=fdgU{?`7CZe&8gpS-&$GrwlZ$)D8Mp0
ziXi<;HeF5>26k@sZP}&Lwv+fkLMU@m&F#878yk7@Mk!^MecR?*{oKLT3*}eTrtL0r
zyWc6kGR68G$QG=6eh+Mh)6|ToHNox8a9RbSmE)sJGj#h-<#}WA6O}PmXrqs6Q}!W$
zZ)np$XVuW0H9hY?$C6i9A*}cwEZ0|e%_^5~u3YkgP_g9Xv5S=J-QK061wFy*!D3{G
ze=V1A0CY~qw|k(=8<m(m`LWl|*$C||E%6PTFl}(VW21g`gPyy--}|Ez^x*{E&W`pZ
za;fc2@qm*UVy+c>$+50$$5W{H42n~KJ>V+uPuU?(JcXrrX8fPLv$>_Kp36fCu%t9b
zamSsl7QwDiUMD~vYOJ@t+4J+0GMf%cS&v9?TikizqNYdYuksm(MY%%m;$czl+z$kw
zCP(u6*}>OS{quvb=l`REujlfN;A0f81YbXTBluPeNwzG!E%b~C$*w%+LH92*hmN@B
z=HjU%?qBDpou^RBOey1e=P5YFy#?IkXDC!och+7xL$R~|W>;F?{~uAM(G-^<a3B<y
zVW>hA0x~w2(YP}wm;H7hI)6_zS2n*lvUz)Wet-D!6{)epX|Sq!|E&SOA^BgBe9`E}
z0tkc9ynk<gev(od{yg)Cq(tP8<-Z+&>&qj8;}O7j0WOc0;He#_Uiow7F)EDXzi;U|
z9?2vm8h+LrAI1aI2Nyn<U*3Oukcm;?Zy0=keSi4ZL(QtuMxdnT*MIL1;RYcHbgEG(
zcoEGXe`|hw_;W`gQ<lPozxdy`DU#L*r)(hk^DnabE8qNy2cvh*zvIyF-88cKXY&`H
zV2Q93G7?4>tN)0wY?ScMC$c>Wjb2Fd+exhC%qU^p$5cTIuWPaqOb!@K(ip7^FfoXf
zMy9QWXW|J=#9+|8*?+`<AO;CBD2PF#`CBs(G^XDS60OqxF#h%OXk>tkWC+o<y{aL7
z>qODyQRLfs;t12y4<iCi4PlhxQTm#$hP%)CDo}S{Ci?My&yl>fP#s9MfM>T~dj-3&
zU)l@SokK|4N@At*Fj?|ykMh)7y24NyWdwoA*g|^&<HNS8%YUQN!7>F%Mfgwv$wftI
zn0B+<5fLaV-sLOtY2QO7LX;8!0v+2uXEK-)aBYGig{Lo%VWxO50b4SFm6b^aB26=-
z+49|jGyO`SBbTqVqX+Oeqp^|KZv1f7uHZZ>%?0dJJ3>A63KePby7jGl;`T9L+Vs-W
zZmewWsGztre1G0*84VS#mD9pX^A#V*2IU-q1xS$TyjmPdJ<sW>A771XJX<VfsDK=!
z)-q~`RKEP@hx7=WIpHn3(gLCtWDP&UdSqIZe);=9(lgdhPD55GQ_k|IA6O{%3#12U
zRDvyy)wW)xytPtA(AnI%@Qaevvw7|IJzVpwMo;bjBY$AE<B5_P?2qpCqt%2wDkANc
zH12gccelR@oso`9$bPOz_46*6^;SaBA&n%2R;BGIagelu(ke11VO0UK$<)uuoj?nz
zYN8<+qNr4O%}W?~E1$hqw=5b8cAA}T+cMNm_q0x?!F56yfnO@2WoId;w;u3W^+u+J
zjqSKwHh+4V)@Za&fEpT_<m)pvvONX5=9yaw41e`-sEl_zgY0=uGE|*W7;qKk2i;g*
z+^59Nz~r}gdT!h#H{+NkBB6~Gyoqj;zuu}}lYuRA6pDHr*fP$*+Z@?eWIGov`5E5t
zCjRxu?@e{(Tk@`6PDgI+-Fa$ly<qpBA+xM`d4K-dkdV9pstGSC;Dpm_B}!p6`JcGR
z;>x$TPN+(<c50|5hL`XuIF}p~Wnn4Z5-qNLdz%%Tie|o=1Cy#}i6yYEk(NRy$k|$2
zU(@OjFiPwysFp^^&Z5|{v(>A(G$bV89k?`(J)x@ztKP`0;?9APSTB_=+hOjU7W_HL
zoqu;R7B7i}I4FE%m+=IW$qjX?q}WWbC(<k>k7}q(^KLn1Rv+nGKWFv~z7Nm3OsTbs
zo3)TQND3NK@~mW)_NUQWv%R8vGz~({7N)&&$>;4nIWP2R5>5b1tFVEfM`8V#;>+|R
zGXkn@gf4PB3RDd7i5UqUO-^iAAr%AMuzw~Jt@FA2fyM{`f8vI)T@z^6ZU~PQM!DGy
z$$9^>gPrxRS#z}wDML3yjuUT_=6B7CXWg#Gn1EL^N}7}a@F=6P9^Kghm*N~u*5j5E
zyG463wqZoU>}gJUc!N7Z*j^Bkh?>eHYV%0Bq?E;|2S#v5z>s}!$y;Trjyd6E41cQ`
zHhK{4IqMtA5Wlhn<Sui50N80e(Fhst+#lDh*`fMV%{YhapxGAc(TliV1lJhX285Rk
z9>u2Do5FJ?KdbdH_u33k@LG7ndWJbqnxo%xbeDaFdORIpZOBZ}7SvrIk#sX`R_}Sa
zonwzP{}KgQS7gY3U0RcoRZ(~yWPknFe&w5_aNBOWa@<sMqhNnXlSBNHBFyO-drr1}
z<dsn1KSbLapYho^AWj)nue5ms*V*HaNC-M448w=BJycTaYNi<z&}|C)3nW$eiI?M|
z^B|0KHT6MpW|c3KPFQJbDl)0b00=LqY<TwZHP^S*^$exM5!Z;oT)pA|tCuT%A5nie
z6H$7PmrZ#|P2qp!WwY@pxq;W|2rLErTX<hx$+^JOOfhFVP@`~YYRow$qaA`Hp*XRw
zxR?uFQ>n7%dDX_ArVJ1wDto3fq#6LR6W8SKKp8y>J*da(zK@|Zy0A&_YKmvT=D^mZ
zGYWJ{7+e(K)G5iT{vuQyl%P7ZzX*THbJU>E?=M2OGSb*1T_KxEtmF)};J@$FV1#Un
z(VOMtr1cF%>Ocb#oZ3LNUvf)3q6DZzP}xXfrE)OxKdpTz#Ilb1+S_Z}50mQR`Ow+r
zqxN_>Z2J;-*BdnGuo)egKlew$@no$#%6G@Q8G;}64(c^@Jw;V1z6%l6RDyr9s?!01
z)Bo{=RXa`++I3Lzd+nB6oU`|0>qS$xTJ>#6!>y=K<#TgVqI#|Q8TJX}Q;I0yYfEkK
z3BBSNS)#6816!lum_FlxKBR?8-Q$G5qI?8ZddHz0SfN1IT%3Jgue7jgwANUkqCsoR
ztzW~sdO$6_6VAd@O|PP?e2;(T%rfvbSbI3$o?%y1`?F_No;{y&>ZJ_NX*YzWYs6NE
zs^ug*8yj-Ttm63iW`YZ4R9_L^c<_@?07@cfG&V>X8;lNdms(*;8Ua_K;CR1+UQ)ke
z)~!Xz)OAu3xt&(Rs|im4e*cMBs`8a{oY2fr5&XE9p;8zqclvmR4ZN3le;+G<I!YL6
zbytzxAgN&4YV))I@G6v8NY(@aL)<s<%296WwE{-o>WuPgj@&c}1h`+_BA9UkJhN`0
z6f=T%crPl1dzRdhWOOr$m7Ga7+WtOe3&jsgYmbH4XO}eK6W)p1E!3sGLgD10ccM0v
zSgANnmPPkTMUBuEqx%DFRKrex^`iJNTbi(oSSd;(gBPf(zJEFB0~z)AyM^BGn47=s
z%gx!o3u8yQc}POM{x~=PP<31D9bVG$t6w15kpSm3yqxId{5NMy<s=8v-<;S~;yEK@
zvVyP{$I<LJ6&XsRG5Otb9H_kT*3}%&9BQSr!;LUn?I79%!BH8eXa~%Hr^hep1hgNq
zIbI99j`eD0nuCyc5<OId+s@iM@kc@0V_ll&k4P~p{Um?oVQMx79M^mODYl@T>0ZX5
zoJHlgsOopU((fb1vaElV1iz-2W@r$>BrY=aF~9HT_bY6AF{EDajK;Qk_14Xve#3N6
zXG^{LFq1d7>0LFtQ=VZ{1N-jGg4JE>Q~$#S;P^<}J0unyYVcEs{=<dMB-U}pcp|({
zUf~t*g#Eud4K+)t!=SBrF@4x8x`hof>?95)4_hg$(+#Go=#<o<{{AmmtQxbIA#flR
zm9Zrgm+)H(DVI=Z1q7GKfghTG5FUSj`Q=V#Mn9a4KgXqYD2Qg|l4dY80}>@eGa}|^
zZV%i}Ga_?KU#4FCE%9TBM|5%SOPaC92#d33cn)AKxQ9HGOEa-tn_R@+<|%)BELx0w
zg!053;wg`m`5gv|{QOkMlO7F^GN}jCGg%wPR=UUbspuf_6~7OelT&(sMJiA2e9_jd
zt%Y{96OI*^wR31j`yzkhN1h(<m6^3hPpBL2(vvZd3W_c;OQ>c&u%wYs&m-WW$X%(G
zF<CRo@|Aj#9SaAlZS6Q2_=3!bThQ{Aj?8^*C!Ys>V9sr#8TA1jrP(hW(}9m<8E?ur
zhj~n|(^=-qLFw~UX>WRe$jsjkuZtE~_W)&thq@-Qu~*W=yebN1mVER~mu-hGzkFPw
zjnA@;w>-FLzK%R4f7|TO9O_2CBKOBtxeK?&bL^5Y_b|&F`<PD7Fne~SDeIUeZ_S#S
zBSrLkpVng4BkWtF^d7KUNDpg319U!EG^ZBa!Na^BW1MI_v7vH*ERm$PHfx%-Ue|WD
zbEX%il@Gb)m}ZOXCTjFCd5J5%q95B#BV<3XvsTUI=vKx@Td&$!J~S)MHfsgieWr;|
zuI0DH-9Ac3`SqZK(9G<MKglia+CoPjI%$fW{t+yXEjFc&AXh!X1S3MI(K_1Vlx9gb
zudDEs_A{JYOFVUdiOX}Vt7@Q)!KROV&Hh#7^`lL?S*px&hSRXmn0B4CcEiH&v96Ej
zm!)FBieT*Y<;UG06Ty*BQ^X{9Qt05#^yd_t1t^1IlEMNwOYcF`-}I9@e)#XVuYddU
z4-ho$f8140D3pQVryuFjT0#321<LRF*$d-?As$2^Y(yA;L<$TVY#mq|zD#$hoB(Zn
z)B2rTgFuf;HG?$>?HzTj6EsbQ2c}qC%Y5J4#=_M+MU<(s!B0#jkL5=iUq<mGzuM>0
zl}7!0y8Alaea>1nRypqCWg;9b{jF&g1cD=w+Gw^m1u_OYU$hb&`NURhBX4a~!g`CE
zB=o{bpih;5bd|ZZG;)_FqE#rArt&_vZ|Gr=m0GHk8esxVhS^Z;qL8(lG(3O)@*|J?
zMx#D8?}^SmC@qk@{N6B*CV_}n%ndeBckW(#x2U_u)f{KiOa;fcD{LS_Q+}QrmXd7u
zvv8kxD}+)OV{nH;3R^3RAZybKDSFt*1d0eo2~;(I6VMeAW(*i`fiQCfld?ud?q}+r
z(8tR7d5Fm>gcdP}f(lEFi+$@iDR!5*T3b!w&}1yqqU*R{x2U?x(NL(!Y@7*Cm8%Q6
zNa`gI4Sul=2Edkj$%_&%_;KFr8!fc>yK+#nJ7bzXyC0`!oGgzmwd&1--I0PTMWnK8
zw)>rbHXG=Zb7ORFM^2<sZefHLd$gz`&N^yEYrD=}r<I(i4w}gnZNh)i7f*$TZE<><
zBUxw0uRbFvJ^R6S&XebAaZH-SEVDm+UX#48x<j-3>m5G0YTXn9O8SBYHgrjQ2WcbB
zcz)Da7>%%Y9niZ~_h{iyO*?!Cie$UQs+fU)T_wYqbHTQTwY69~$ITRL{}cc7zRMK(
z?bGn|M<pVsdnJ*8P~h?0mLP+gASj_+n123i`g-?yB)5pb8F<|{$Sa0m4-|r>wX`L-
zAs7fA5YHG;Al9Xb7Yji2QmYn-8WhO;MEUpg+cdQNI&%!pi=fQNfs&mqYH#lZ>LtH_
zCc9_xCztulAT<VmwUN#q(ldt)D<LQQ6u>&CE_(9h2)?F#XVYB{gV+S-i~4+ot&Dqv
z8K;CHhiT*Iajwa+z?&2MTBp0tSk%H`xSx5;SjDshpDVdoF;>g}WyL+#kfyFvoTZi8
zS62Gye6?P*rL_tJ;Vrcw3Ll2P=K}114LnFcUoDeD3Lq3@dOqRfs!W8u1zWn8>5vvc
zwgLf1p2b86nJ4dBe&s@SkZ0qRiAiFe;b|@~wx-M6?Gbk~G8U8eGXh>c!)W(`;A!1S
zjOG-1>|M94T$w7HVsw#(A2+JM!tS7WZ`sGDQj~q!g-~NjlQKO|*dKKe)^ufmk`}6)
zQm*3od&rDCO&mi07a8wbTK!tiirqL1uLl*z;zfz|YtYO})j;L+w66HU7b0g-rKaVz
zw4Cv8WrA=fEO>XrG0wWLDoZ&aDPtUcc#Xp}PP1S4nk~F!z^por){0VAT=YJ!I}Kqj
z8+6SM?dzj+Pw7H<9;SVs<M51s6gXA5MRY40Wj|*F<@b02a1k<BS{f?v#<fyLIE4$Y
zk9GZD^jqyCni>tWvDek~^LiIh&I;!iGYyEp_Vys8UN}sNS7mdJ`*lvWH`mi?JJQ_5
z3fiHt%CA`g!!9dm!qhszlET$m*6e``9k1^kNZAg`MQ`>^%nMUkmjVrcRIPL^?Gi7+
z0vyFCte&KIADpcNSkkS7k!s$X#(*K&#mw7NsOU5+6R>tY*by-@f&B<^oG?M?S0cn%
z<n<$CTM$vBP+Bd-$RJTauZ5}V3U3HgG5c7|du4D{J_y2}1Ce)&tr~*R3KmT7kF8~s
z9Vlq~Dp07|1<@@`ecIfAMzwK-2p0mpKYr!K%7x~URrzY*j<nHPzmR1}1?Jv-IARyB
za#}?<+LL1ElxYJ4X{RXy_}iBq((*r_J--Rr6ag$H6}pxH;$eF}YD80yMt7kc_A3I@
z1I53f_@T(L6qs>l`MkLXYJePg{mTm=H!&fl_`$<1O*l}zK^?|_Yg+qZq`Zan1JJtm
zWAEFPVv0SIiQZU>30k;-BcodzZSaD&w}n`@uf}L39HwdlhjVtl%MA(@+Ur1K8{DEp
z<zQA2_Cmd(IhONVD<-M`8thfnvtDNJPk^6ani75~y(?=ds~OAqicAPls`%r<E@>jp
z=DS0IO3_=gh-Q+1zx^(!f6|w~Oz{fE^k?k)qfW!O1^&**1Q!UCi$ubT_9%0Xu$(A%
z5&BGIXA*{zclNTYN$ljz0Xc^~PnH}gn33+;(+;juY9ej~f$F-B_`<Bo@8rM0SH5(<
zOnc0q`Q@6I$^GJcc9~hA&M~U0UM<rFU0F|V;@)J%Yf0>XbX+29t$&T<E?3B&W3)Q4
zLJ{O*5%QTBL9??*X**8K6p;guZ^j<44l&@INRD)?5Cgnv_;&~~5IA{#Q;4~m#7@o#
zEU-7cONilvN^-&)Vss8IZ5TP`7v^?;4>{}{a(Fp5VTi~0eWB*{vR%>{jY0n=YhFuY
zrvnpNhc^j-H9LOR0q7{cVE5yN=DKygae&Raw2ikL+8YYkTs3yxrDQu_!bsz~E1OS?
zYt6?7wqkSbyubgj@47a2Hz|gz?ekrX=Gu63@g~<D)5Y22;_e1^msDYK$$R5M{p!yx
z#*00+zt~gI@tJ*NHMUw5THzR5Y3t@YhG%z<J&-zo`qp_=4a$()nyygQUY=h5bl@V=
z#t`w&K)GCrtJi+idcva*ZG}tMdY8&s#YWKr00nBhKHK%FU6G-ux|Q1H6m#gU2Z>*C
zl2FcS2Fe>J2_+#ttBa)N8>Z?YwLs@ydybUK%>$8lJp4*U_L1rRrOt9GXrNGk!EhQ9
zNk3nI5UQV_<{-k7n$G3>>ttPz>r?rZoDaNi)DpE+LQ~>5+TyrebjfHz2?PF??ryCB
zs|Lbb-kz@q^_>^ao~x5w@z;mB@*j&6bC@aLd`#cW->Q5Z%YkS)Ry{P|G-Khp`t7Ot
z#m`W(ctuIP7gmjB<~N)H@D&g<Y7;|$8+o`k^0IjxkdFy_VmFLAo-aR!3#Fhv6ODw)
zr{=h6=AWC^>!)y4(Xk`9wk56Z?87?&hyMdWEEAU@v<MReG%`7t;lL#m0XUZdXE-Mn
zy;{wV+c*rq=PB|6qlyysV_*;<lb7g?A1hqT`|iWp;y$S8AejC8mp#w}O`o7?(7SyA
zP~O7+{oDTUXRVE=>xC}V+FH5-{f_kQ4o?;$%>a4|aCiy`&v~9X({-ddhUd}mldPke
zL_%9nYX?8@d?5Nj^x$}xZ;l@pe-RXT*nj@E;|FL#*=Y7Q(oS*z<KO+4-N&A4<)vwf
za{NKRetg|`BcDWZgC@mk@11KD2RiZaeo&SlWyK+F$^m_cW@8`D5iU}}pwNaw#gV>{
z)agi-j#P;9o$i?fsaLaF>C)%Jlh>?Ndn$K^X$-Uug-&G2vP24bf8`@>f4do>Z5Qc?
zn}Q+q(vjJxdmCq?LI?e9mJ?6150RbaO7|y`Gs+%y-XqFfmfgPHE_d)Mgt^?W%W_9h
z+Ro&5EH}xcS*s_JVbPJ~n;f)vPivy`F0gxk->CpYdR~7dug#0U?>svn`>=e|!w-Lc
zBe`DxV;`1zA7|n}6=qKpe{RWfUAz?WRvQ(tZkMAsndOzP6GbX=l%p4?BbRnwoKOsy
zjK|}l%x3c?6SMhViYb%ouHlb~NMSVM4<4S-*yqtDoI4mu0me^~gBRIs=9c8~qhqn`
z)zM;-gjV9Sj9R&bZju#c)l21}W%TV}T?ubrWs8|(mdO?>L*>)ye>Bh8rqlc^Bdq4&
z4hQf<Vo?dXhV`PN>~p)#qzFGMi|-jT5EJd=bRC8%w_?=1vTU)MdZNaJHpfPdWo;B^
zNNBk348bVM6g3jP&~$SYWYj1~ktEZj5*w5eg{S%c0M8>hrpryN|CT{Zf_}|&jWf!l
z?}g1qnZWlkE7%UEe<=86;-91<8&af+TJv2M_bkzvN1bdO>PVqi(oP%!)wyO?<hvk_
zkxT^>UyP&Qjce0vtW*m9V<YoqRWf>B?b&63jZxqo#o<n(qjPnKnfW~dIyTlfV;<`*
zTC&m2r9+^~!S$5`P%^OQa^;Y~^>RKpmYA5aDrkUUE!W+gf9Jpj0({$ZGkP)4f%mjy
zGm}l6DjDW1@|Y2V61=YD=9JsRv<6{|)NFaZz`9ml8RN-VV{i?ltD^6HVOzz8g94mU
zek{gheDT*YCjBlf!h{@FHCDk5nT;UGfK!7Gjo_1&IVaukN-gzkuGG%Hwqq<=Z|X*7
zX%9((JbYD2f7SE0#n@xKvzUS^-T{ZY{ZmT*+q_UyyXATck1}SU5%4{AE@#&a8m)pb
z^ea-^X1JBRU50^N18q-nwRgy!MmIaPp{2Ft%gH^Di$hru8Kp%{V_sTM%-=Sq_A|MO
zk<Jvxu9mx2>2818H4s%CX!lkPqgmZp7Z*_J;7Th(6=dPygl9AI>4Gzbs0NzcGS$9x
zg0`0vk{>jG!C5K-V^4V%J|V$oH~Uf!syKKJUTfE%A@t71-#M}h_5!X2uNzcwLqGfR
z(0#{IfaP?rXC_}c3b1TO=VyjlItqAzGm{$t*bC`n%TBMQn2w9aWtSA(>L$P39tl!k
zb(8>a-dJCn^c2q*OOsx?=LuRI$Bwi(X&e6&Gv5$@G{&i5aHE{YqB3?S=cDRpF{h)n
z>hO2xp_c2lM3mqv<nx)5i0ZG6=TBqG;?oyNPhrv<%TdWBwfPlp3yI*73CvP>4DM$W
zn2F>(etKM-b~sJhXK|(y!oiY|ARkHgX5`aCdTO~aB%7f<FUU57W!)7!n5k8k1F_R{
zk}QRPn48L4MlR^N=AusaZTdAHl_q8Mt6%N`kVznK*S%#6&joPH>B>@?ia49S+&BUR
zWkPM%>>YiD^~y%*DXmpGf<YbIxAB$0qyl(-4D1yIYI)*Nk485|nY1n)m0u=xO&p!n
zr_F~N#zuuP%nY;g+9m7auGX@=TIo2gRfh6^YNeZ9@T$~zE{?}JtMPPRN1BIN!3I>`
z#3d_iZ0c<c)C9ns!tGTnd*DpR%Y_8;Yd1Zc&Ay3x83;6mF~p|*nH-?qRP5ytZwYpI
zKxJsiI0w}z&B#(8)lszT=ad^ughlr(f!A`Al-X^POk!uph1{jo?C@(k7^9;K+PC~}
z*Is3U;iKwSRB2uqc<1mXF7i|G<$z1XhUtxnYdj?|A6pu~4sVC3NWxCy?PzQHa2iaz
zo{yaFXd0}s%7spQa9&NSmQ%3pH1!*rn~X@Ul=|+Lu3%P64_ELlD(`LQ8|Ou2i%b@-
z)Iq^2MDv%hlpi7|kpDRV-DK0rYc{^-UOpzur&sk&$SnVxu2sJD$BfAzH2w#iw{|p_
zA#flR0y#970i+&jw|JEwAyk(E{WcSqfQ>0Af3;W5ZW}oazRy$W1+0h?^#=q7Vkcf~
zZ{78+$Z5Cjr9qKg`ua;rltv?u9XpF2#*!rt`H><&ag_Kjl=x%X9jCj4LE&Yv;I%k@
z6F?h^enHWqv%&*NXFwcZ#Fu-ew4-N74=QC0J)Y))M!Ve@2qXct3$WdIg!>|=5A-~c
zT#Qa~G~e@kBuOfiarAC<G06kzhmf>vpO2p=#XyyeE=|)g{(Cx3uhR(IdMi<R@z2YY
zHqd}HLDM!$r-FF<D!xshN43qEqIaZCKQw#$E{2uAIC4=@<dk#PE|;+CA1r?tn(qz)
zR<Tws=p`E8u(}y5(y1XbZqI(u38G~`e34b<#7?4MoJ1cQz5{D}q-S5WlOko471oqw
z)L~tlw$ehZ+H&^jlu^Z7I&v}p<oGRC`uL%uytbo8Sf0o%jWWbhSP#ym;z*9914tok
z;ldKKyzYx!UQ6#Ewkhwiilu+8rxZ<k_UGdOD6Q*{D;-5X63-Yo#q#lSP%{QKW2wiQ
z|7Y9<i)I`R;tIY2SIr}3sJ?UsruM5+Hfrh6H2BH1>7Z1IkDOUp;&EN>WEajE4n$@F
zd&Ai=HiWJN5qVAXoP&|$9M%A}*sbPs&CoT%0UDtj*BOnD-Ul}_nl^uuD7q3_)i5(k
z$mRmgS5zW#hl)w%muDsZrOi(yY+?|<vgk`Qqr?yKcVf(U6bIs%q}WQ(a7nQeT1G0V
zqIOncO_U1b?JBNBW>A9Dm(^0XL_6xujDL<rL27GAIk9LX{CtgB9;w=Zu0Sw+#qD{X
z9^XBd<?_fWGyAhhj23^X54-eUs6c`@|LYKMRdHqB8uG^DNAh+okSjVMj5Nk}gQd;|
zcT5tlbUk2py575W)!#FjB-v2ABg?#+Ic{EEW57%}O=3ao+pHZP*Er?<-Uz8NQR;+A
zz{zOb<<{t85fqL^>b!KA$=xt->bad5@2JG3W|uFOjZ=w6&AxyB)Awuqd-HWq)+9tf
zqC0clYibkG!YF<HB}5)P-Hy>xccD%qSdV$cXruI0(qp`L<-W|#!=+27j+7!}f`D}d
zB$w^$MAu8q`)!i*ro8+*xzwPhEaH(i@BVx-T0H*j%I@<C{7p5s>8xE*ji8-$&`8#}
zuA4S<cbe<0wmE+foI0FTr8Y5g6577T#>(adMz+B9iM8RX$uR3U?MLd)18dG!>T3Ef
zN^NBNmeNlZUpUJhS@)sqyyw=fe{VlB%<PkAEXDWQ>z`P~6P>#gee7dxahI)AACdMg
zv74jwQT8G>_S*a=@}lp<R!ECDa#y@;_-_8@uqIzpofRUj*1tC#$Vi{BJ)Zvb`{s1O
zc^yfWX7z9-M|52AAL<Q3x0hj(91{aGFff+^q#kLvc$FU^RF{&XLlT#&LLk9^fS%aT
zzy0{-i5Ecyd<ydU^5f~xr%_aEEK^)>|Ni+DA7~|*@^X`k!6vr9{=5DD^p8pbDRbuI
zH+cWPBIzvi(nXTL{>Hcez{6i)vI>6tJ5Bvr+3@Xu+rNOpEyAAhL|B1Z!!^QulPvg9
z$Tkv21;%xFoWz}yS;?%wPE~M!7R*RCIg=OQOx#*yI%fhWl12e-!vYX5&LmWH6?mCw
zaPq0hcW@LD%IBqw>oRsJb1vCg$i6xUsTzCh)^2Fb1P0#wE?YiI;!gJkvO37OxbGMz
zcccm{Ay{0sZVYoHVp?$U%kcA;7~lqyfPo-4u0Dw)U|kLAKtt%8no2Z(%+os0Pfu)p
z5=X&vw8N1^6DVm&HYOmYJkDS1y63|0Qf82uZB~<}u6r#!zrkzo=EYrdIyXs+e0*73
zcdJcDH?$olYTf>;92zxwAl*(^m-#4L=w#z%mTNL<mlAf2ps%mSZ{RXga=Wsy2@d>8
z*Z6%d@e8|y4}OYg!dY5>G*A&GDd06TPM>?(FfUB6yOnW<S*wt#BHNcHIgrYVPUcq4
z#qtbaTug^LA1UnFrOF&vsJKGG<QaJfxOTBd)@PSq<#~?Q^_wA7eJ7C|oHn#s=5nwL
zb+?w0e+kI^OMWBnxucv8ZD+exuH%YQy25$S!FZ$><dn{Ol>-fbb<@JQ*(R?{&;2>4
zi08ujm2ynpGkJ@J<DNV87EB96h!#zj1T+Oy=OH3nwk02wu>@Zq(7Gg-WSO%RCw%HU
zzKPC-B)yg>FU?DG5SPT6<fyZ*TAxajMlZcj*YM|EQbjw4A(ifwyPqsdlM5yk^unlr
znp`}C91??h5$DQ(G8tN&^X8wJ%wm$MvYWz17mid)JslEx#F4mgBpOYN{Y0~^ApTSH
zOwaAE946p1xH*TdyyU=lT=5Orj?^w1sCVupQtj;Z{=xh~{bTfjmOsEg-v-o<@XxXU
zih<LPi58m)#)<U`C7GL7-1Z67+R`ts%H|?2l2tZyTeg0Gg?z-fvYo4Ml?XrfuhE~h
z1|-Yn1htf;Q0vW_Dqu(%jOu|>&Bf6>l{&SvAf3A}#p}|O$(yr`s1;G??cFb3DcfGM
zp6}`uD;!kgn>@dQNR&>FIim4<K9rlD6>YvICrlnipv7C}1>h)jBofPGZGC4r^;mjZ
zI)&oQ>KZJ6w^LU!a|1=A0d-gzFeS#$fQC~8rd2oHCe2!!FHyU0K9bsWJxc4rRwc)s
zA~wd9NXJB}MT*opbQY$=8MdWPDpRznNr_Pn@UqBa363w)k%Qe9O?i~8^6T^fE26Ok
zUps*Csfr{HuvA&Y2DVpnjSXS1<dM#d4UIgPbEm|AD{i=<k$PU)0bTjAOiOZjoY`S3
zyFQJQ<J1x_@h|8~@E*15fMz{+r)V{UpY!u!Zl;B5G2c+i<*wEms{&W%y`jCeS~$uv
z^8qFfW5xowbiyzsE^UYunmp@hsUe%Sk%k`XC{`Z%*CUljUa@02-0gxSl{||2n<cGv
zSNp|(NwsJC-I}T!jVKYr&y(8lKnG}d48$T^w+FRC>WxD4v6bDG_bkXz<J8l*0H5-D
zVTn9ilS=-TtSZ#A7_lVkAxD)?wRv~eRUS(m>X#??Ay0W4l4+j#O4=hku@L&AY;|K<
zF-eH_%XtF0mS#~~Q>(E&f6}Evx~e5pi9St#vQbRnHa(Fynnaq55ef9N=JwXo@-CuH
z-DPNO!|Wvga87NV_>Flj<(nO!PRFHbF`45qobEXYOBUw3*3Oj!@&7V{nR7>pYvqf6
zP&L))Of|hHQN#y7YE<drp0#6W>#$NEEG-t6uIFH;UZvwyLoI`svyh%wsLxu#>tl<5
zTf?}hv2JK3jxZ~qHZ5o1DY&;LA{n(!8^*6AidNiLs#2u_6Cr#+#eGi=tR$Rvob>TP
zv!S3GrS+*#&PpE@9}g6KNOwYqX2PBY_i4n$lgn#M<f$F(L@9CK7hUYU2YM76x^9Vn
z^O~i^UCK%FbP$D)Ee}JJuxNUPG>Qm+7EPD#Z6?!c6Fd5o>7oU$IY!*EjIAMY)iDhB
zKwXX0lm7F5Wv9nxjb)%kYP#M&jFyt#o(K$XBTc8*G$Bl25*`g)q6L*NSDbsk?$q&`
zQ{?d&Qt3Bdcba@%nyy7%y867PA*lrbL_cz^DH5A=6vB%+jmmTgnm}M1HnvTFgYsD)
z6)?4Q)Igdy^e)5`(EL1rwnS^$B8s)INTqu?D{AE;<=C9hQn^+I8b9uEodi>xxdOGe
zL*fG~Kav4_zO^Cy=CCW8z`E^T3zRbHh6PBYPqt<n3pQQx!N>RE*YX-%8pouO(=~YP
zKpMOTAE@4Qg$qGyG?39u`_^QCm{V95TEnvX_SYZVF!Im7pC+UoDa*e5#u<-GO?=~p
zVmCr5YR;@O6GdMFGpzf1Mx~q4VU>C*JAzEbnJ90Ge>Xz;g9WTnBc<pS?@t(!&QgA<
zeBw4*h(9hr+S)2+TAdQ*kJs=6|FtXKLIX1TBUl}^yq|?aSZ=x!yubZ_q#!PRf2ngT
zY#$pADGK-Ta%Y7d2prRgt@+T}?)5^di(<blW#1mr0_sKK&1Xe4H%v?M84=x6mIwWG
zjMjd(Is_Mw%i~}~O!6G{_aT0)6XWc9)EodF4&!`(T()8b5g=-9yMJ&W5M#o4=1f@L
zoHyXANzaY~rUpbyjVCLA2u(M%84%Nc%7;L9sWrW;*7oSGI`05sLjGksN$HL8*j3UG
zfIns8jnAPzSKbi+79rI&kGTr`Vk=*%i6iZx(x&jJGlTWy2umE!tsT8wM(C^iH(W9=
zCTWH)*CEW0H*Day!|Lc0$w#o;sotI-t!7G2fBf^3+3@c4D3oP?%IHr>j|1fi@K~2m
zW>Eo8H?jacBSkg+N9j>W_4PIIrFN`k0@Br}9dz?c_a}{(unu%x0F#ggAXsgPC;h%P
z<>w?mP?UVsallPWQ*=y6u4%w4CR~^}>^0Qx+IcW6=xOKbr2@A}M*<ECWxG0e&!7>*
zso<iAbruA(P;Cx>90r|LxXHm4gBxdm;IS=bV8KoD2=QSylu9$H<g^Erom80he|6=z
zT<H3St)&kvCX}8T!3~qCxnkP6&p<@pfed=&Al{LJ(0O`_cbtJAk&ddX^xTIN>ZXBo
zIjIIJi0a-UpD{2^&y=SWd5;wj<+q-}YqF|-m_w!LBvgQZa-z8YpO17FaFY8(zg65a
zW84kDt+V41E+*+8Lh|o8BFhmJK!u|9?k*%x*9+A9+>=A%FcZO@Nd9{MV(Z<^m1AL)
z(F;&(_>3H0WbPRp*h_Lih};g7d-1b*j<@$Aro%uQ1j&`peaCvvhERC6lUk8-DY}@>
zds!RsBW+M0c0+|ZvOtOVPVOt$Kk+k%3(@-uXpf%&<NpuVu%0fLVUipZ0W+7;TQ(E{
zIF|uuI46JI8p&=WyYcR?=pQhL#hrj5#M0OU<TinCK@PznGdauzc`t|jeyX^4v9;R)
z5(BcN5?QQ`wYFS-EOPnsuz!2_^s3F`MX=!2^7efJ{6XS>kO*{k@c_bUSl)gue}0lu
zIehK$Wu(*^UmiaPe7Czt4FOjG{tB>r^ah^tI`w~wuV-ANXdQk&lzq4oQAq7@Z*-B?
z1LB8>FTMNA+aDifAykU!=p()HZ!d2TzddlW9YmpwUcUW$pdX9|p@UkS6dw8f{cp?n
zhp!xKJ4@mgVd)=!`u)ekAb%=?ixLGVowIf&IB$gWus9cmrk}Fl3<}8tzWWtppI!r;
z#A1JgK-p9*7R6&;Gq!}0<hRclg(3RtzeM&5f@D|hs0abkbL?P9b(8ShE>I*Xwn-k!
z@#m#>RAkl*XZ7NY7e@OtBD>_IhS^B8UNi^ogz@1l=JO&vsxUM-?T`m$g@rhawAl7@
z`AoOuc~i_y8f#2R^`v%ux2n=l@ZF=6ba#JlFa(HJL^D(E2FmHP-8KjX!nnwHn-`);
z(I->P^)=H>v=huC-R98wKpw6(VejxN`-$?;S8KFk3|8Rls0R1dJ8jBBgA@T2-%a<E
zI92HP<m=U3YORC}W}?XsR%LJISeqz7PBhD~%^}$(Ov<k2D6=;rW|PV%HCHojP?>)X
zx;c<7XM8R^*|74m6jsc>;HR*nJkb~n4R+@=iBEQuOfOwmZu!@YC#>-h#l;oOw8bc*
zG@LdtW!E-G3OQ2quI+2z=q-{h^o4TeXwSZJe<(5!iSkEy+P;Jow2jq+6m7`oqv1Yc
zurIhwTw`vh@5||pIumx_h|2v=!_$AjN9Ys{ycABY$t$Xpu5_-P$$?YV=}36?g|ATd
zbeam4HYWP&9$)ak?6hqGBI#?gIFCW9X{Xv7VlpY=vDL-N6%-2B1hrik<yV!ZI$${B
z=<|w%zkTa8B%CTsntqja?F6lR`Y%;?6?5Ag%P+&dP8^h5l2W}$R-+Etwz7XiA2eL>
zlD>?GL0_HzdJ`#hbeSa1leQ}q@6bw0#hRI<`bjHWcrGW?s^>|1=|*g8+*x6a-O{->
zcDi9lLtI&%=wP}kc<{dXOYTd?<GLTMc3$l`@fi9YsD2G435Fswsm)44qn~%yH!{rA
z5T&6V6$j?B3&<DyQ1@8wcr$-Pwytp%@4mBTUe*dJXLa$Oa<WDRjMj!Lr@1c6jXFn6
zd*t`T6`9Vy1*Bt++JeKFtS7T5mw1zl$)v0nCPZ#h^L3=E%Fkw6pcgyXvYLg5RIgoP
zyfv%3whF`KPy{(uVIG9XlJaB~mMW}csWzrs7FDnD-8WxsJ`esr#fyLI(_u`vKsk<C
z!FHlviVIH1q509P&##<eF%h9>4*up855QuI7blEzo1;?1u+=7Xe_yk<9GMG;SOb(A
zOxaA&3Uada(^yOfexi<pG5b~4b-r>ABI2P8qkXxb)}RIWNtWu+^OV%OxU9m*Tu&Ga
z9GjCjRP#Z2=_Vb2zTbap5OcJ>KVa|1XrdP}hIh1V6HvH1vMl>9jGi2PuZ2%_q6!qf
zANPG4?a;?azVEjM7k*jNYD9qcZkK;A1jbFc*g22y+Tisg3LifG^T+o;eETa1-0&~j
zfvyv66Hsoy@>hpvD1!#|Gk^CIEmOq>u<+0h!xol}o(!(;Yk+@m%Y#u~fVN?P{v>F`
zkYtn^HYb1WX<@<NA&mk;*r4^epChmm)lWzqH7e*EVJUm*EAFqm=qtU(?{z8e`up<m
zeR=rWq!~5-(WRyvu^~(l1}OLp0&O6?KAO5jMH<md62dlgV6*|1aKRa!aubi1sF#Mi
zF<T)6!xe<MDu#aws_z_rL;#G}ERy19g^lQJkTCQRXp>OHnnl|`fBRMT{lHychWCu;
z5fo-F=n@`{_QHZ8Tq#!DuvrV2<#l!S@6V;qGCAo;DPT_pbIO!Y-nwE8DU$GY8A!~$
zK6YlmUMwI0VXe79Ks<DI!C>sd0$w-49BE)w9sqTTA-8`CW+>)TVQ)&wkVD``)~s4l
zQ8T%H+jp|eD<yQ^U$ec?Akf<&xqP;*SgZJJftYgN!)l!!a?Q}zvqG&=yegC`U9H8i
zl0&UBdXuBk^*qvcuEZTU@IwY%+F14o$LS0^X3gvTzG_t}p?eHzIK1KoWQMS&L4n48
zL0SW(mEM1}N<R%1!Ns`sMiVa8gTc3S<~Y7eMbcMXQPRrSeqw3~HhQ6zzbooj7BpB~
zIj_;eZ7O3pYf}wVtwuApFqF{U?Hq0D70uS$UGJ+MnuR3^Y*keGuEy^7%gNdjg26Bs
z`6;P2r&4RKHits3-&mz71u~{7ifW57Bc-=e<-~ty^OrG^L>?8!G0r1zFdT9V_0$bx
zQk^CT!3sB@v}%2E9fBXLY|eXusJRHNb*}GPuu~zk-LNh@d)v33D;uP@aB_`Z{fJ+;
z&2s-?C5w%KHiV9?fYyUk4ru>a*2_xf`8cGWr$Td0RIL&rG+(8v4BTMNhjM$SMpW1s
z0knTr{AQhQ<!g-vQB=K?vjZ2OisQ`6554c#m3lF;SwZKCtxD!0^@xF$aFzE=6kgcQ
zQqhTMJQu<h$~YxrZ832)z_BOiY?@=r@}cjgLwRf4M@zij5_sDOkVWn8Zu@Aw2EqFe
zX!|hCtnG8xguAqgIPHhQ5KDe%FMU7&xqyF}=IadRwxOHy7BlPSwN;gT8=jD7P){4J
z>8m0~xmYHQS%cAAxv9}(meszs#?@<uZpMd1YabId404F$hX&Q1?iSU{wQnOV*Ceo_
zsqyGbOpjsez^kbPuUqsQv97!Ax=Qgd4TS2icEyDoqMqGwfybn^)y>BGMR?n6aCU#3
z{&GhMM&a!Qjx$+}t|Md(ry~mP4RWGu<@yKR4#6Su*~eHzT+d^))NLor+{z$w7-(bJ
zmMt{NuY=fSc_-cPC^hHh@_U}=9;4-D3$E6?%X6}m0yWa5Q3qGbjqw4r6AB{%j}MgR
zIDdD5PzD5=(FX(whWP^m1gHJ}0KtDS(nEWxh48$-%2jss+?=u>A-gr^Gcy&odS$Yo
z$0Ff+ddib^oSq{Sk<#;LtC<vUdE|;=gl&1$C-p9tM;&1z#+OYEBQ~MsN2E^-^BE1Z
z)l-YBj3aNEq7@BudQNMYuf8^sx?*0F8peU2c+3JYXIENHdOPaSOAGCWr$T?d5>`zo
zu^A120FSSJsX;nvu#WrV@bt*@O<L7hBU^9JXjKYN-G3<m62=u|<Ka4nksL0i8z4{L
zv+a`;<#s%|LQZ&pD%eBN6D5~<dQBFrCrywact^i!z|#!qLaR0OXsDWKsugEV2HUfL
zmut;ng6LwLwYE)%!J^a7l0ARiJe!B$1lA<)IKku?PtWGr>i@!Qo}Ks(vw4Q`9cS|l
z^PRJKhV72oJlFC4vw6TGT!;9Mvw3jHwLUucz5yXU<>cW#uL@p%3oV(2aNBI&MNEm|
zV@C5viwHVG9nGTyQjSu$HU<v@9QAYS#s*l;<{NZ1o9~<Z%)K5kfGU3t|3AZd<%Gmi
z#;}R|CiH)FIFE@QbuHdwI1dsnz_<&C^L6#R4(A6Fwq_Zf*(-<h_O9W4ppfD&oXwZT
zRV+|q-Ow&7ts=EK_?jDEbta(sbJ;Cs507dlNGuqKhGT-utzE3~c5>c8@hBR`;`y}q
z^q2v?iqRZtYf|QBnAs(q#7I7LvvOU3)u;2#QyWa3V{jnR*0qyNk_jiaZQB#uwl%@T
znj{_Dwr$(CF|lnsdFS4$`_=dNoUZPw{&8yUz1LZfMmAQFvnoMmp(ekyRct}W*Y|ly
z_xS$`XD}A#WYBW@{}>E<wf`XiB1wM0Br|IgMuHwE_h=Fx0t=&bTn=|DER!Ucr?2EH
zNECapyY(9PB63>-XZG|e6YEtawlQE3)HtDG2sOm9tB;sz6<)-#ZDA4v*hE{9<k&hy
zxAB-L0u25NlVLN5HcTa9sXH<wWp#Cwz84~wVPJ=UX2Y!oMcdG0eF%Qyc~1r<L*Ur_
zet(X9e|`6O0j_9o{8&!TPqu)r0SqnZqMuNt9q@<u%Ya;W_q)H)MW+HcW%E+XP7cT-
zj#nMXa0s>XD+vK5G9OX&rk|yT+SmKTaA?H)DQ{@pPi|=<>DqHw*6$1!MJui-@0{M7
zOqE6P;b`w*ZkAUk=dzFlJ%x6H8fFp1k2%|Q6YHJN0N4n)A2PM5vl=g%Q{C{5wu5(a
zk?jgnUUYXNmgqs8Ob^BH-!tKe5>;JC77ClO>A0=Ku=uk*3%bG&P%D@Ti+Pw?ZKta;
zcu_fq<M|<0gZYEB9}@Xu-O8+NAs?AHGP%bf_C1Uj8WC2M6q5NI0g?PgDd%Dzr|DJ|
z36Gk$K;)1|%=y3Xb41qse%@f!%Uh6->3PS}GgemJ&({dV=JZXKowYauW1lQXn&5%y
zGJa3OL0w`_5m$%Z97(}fjgbOZfQ?ufL{Yonvk2+lZw<Bd3k#yWRQj<}pBJX~z@RDE
zfO2#>*@e~LsV-c*Ic*7-R-e|v!D$&Jj3A8#_>(xS3{=<3ti|BM=X$7OHIA~_^-WmR
zkJxXi{DEPnxulYqOj!95S9VOD?b5*@f)Nv7d30Q6lH~LoE`4lZniT#?5#*oGV(pbu
zkDZ%JQvG7ahUI=gj;fq;pg%UU?QMUK3%yTd4YLV5os`Lh#jouP=J-J?5WB2`pJR*y
zpgQsj_?Prhf4;BZxxV+N4a{*Pun`>Xvx;&$nVYS`M|$=(v<Z}pM^avq4YlWTweIs~
zYA*GIoy=jKuhsb`Au_6L=|u(!eHEgT`L2@X&3(kGkOh}?GcJuHo`yXHv~AbMn;vK`
z&bQ;Qop#$wRHPj}U^Xk4>5>VeQ*M<5t(kEFqQ_6OHDX`^L8t337DZku!Ljn_oy{su
znsWY;GgQ`>mG%_*WGQ@UR+Q$cK5hy7*u+ve0r%77INU$M-cVYHpi&C3Gj3Yu1_W?H
ze~@$=G2%TFAEsm>PCQN8h0m!=?DE1ofW2?NWJ0MZL5!CM@n`*#uuO?h_`XX5@CWFy
zqr!mtK_4^L3DTFc#&0)42~??yUJ^wvpua<!Qk$vM#W_s~4Ha~PK+)0GX68R?PCvhW
z<H|GYgMh&+AmS3tRdW#S_sg;?bKgNR(vUmjM2RGqB%P7OabkQLBQWa)-$gjwP-P#&
zQ@^!^7W@VtJs;Y9d-+?z5hZ1YkIui0=)sGX&P?tPc}JIi>6RW{^bmjPqY{4>bN`Wi
zXd_(M*4>iKSdU*iVf@&MVW_V)Q`!j9vD7zVu(Re>JwgizaQX5Wn*WvQxlkPcZiR9s
z-&}%IBv%cQe^1UdAxr|UZ{&fG=cDg|5nXT3j4bFC`pQ4Xi~kkJ3?R?mr85Y>8o6E$
zwxsd&8UwpUE16oLDAmq{Va5qSp2r2cJwNA*x9(x)KUIG&?yBuM%Z+n*HsoOJ?cJhf
z_rnmIY#UDYG-~T;H9CO&L~-AqzulGrKjjadtu|$?rspyqFewMr(~Uq$!Hh!OC^6qC
z=iaElr_zusrkb&jDfjw0j+0CjYaY~k?z8yzRdh2TWG4qBl&BI2g@DAx)#7&v%S>HZ
zs#eR1GZN%#Vlx{;+|5Ka#O2?98mNq|394&InG5zch!yT&C>y6*ij2q#`-C&jAl9=-
z5*!=Gp^5#K_&5a&u`0Dmjv-`J@?Fw{-mc`g>^3n{9`rmpc%xVUF22<820>|6m6XlD
zHk>s;MfM!m4=&?AXEZMshYV%?j=b~z8!UP~7vOQO<kg~yf+*V2Es8R*7zL>lKv3c9
zUn7dlY7nONGV1aCxUJhIR;H1K8l8gtM<<U%RU5mDMfe(k`;IgjFDC1+Y(6yFf?W{)
zUNqrOPEDqi*3782Jq4fcct124v=#`dcv2)*zPoZglXn2q!)~A8FFG-X+RW>AXSG}p
zkt}JO&q<7oEqE2H8XX~unmEXcIKfuhM)?;;Lg~yB>T&?=d9}<rLu2NZ5hcRP%b;@O
zHm&i~80(*6$>a_56v9se5r%VNpo2^#oW^viU2q~L1?QlW*4(ciKbR$)dPaiPZ!C8U
zVohLo_;okB4*sajhEqOvm0VWR|JNtu4g3aWpL9xIq%sX@xA39T%F7nAHL()ABECpx
zzSZ<{rWaRp6pO1v$Ke@@Y&iBTO@%63vV4+hqx1wQ@2q4Gj?y@JtGXLgwxGAFYZp|D
zziptFFr~rkYg)NyJ%cm^lj^ZclpfM^S{p9EM{4DKGEg}A=ed<2{tCKguNJwaQh-}m
zZ^0Nv(!^Y09$jLU_b6$4b|@hG{d}@OA{LIyQW-=rKW75LFZ^Sdbm+A@7vb8XQ?ni2
z{nx+$56)DP&MtnY_7B|3)<Tn|{a|yEP9%$1bYCFcD~M}MTJU01MFg#va#%WFu4YX^
z$8rASgYtw=pzg<*p1B!33_XK;?t^XP&{Vy263x0gToF3Tz}@E<=K^DO&`K@S6jjyu
zNiY^$U+f33e!|)y)QOET7y3)NlhRSe%<@lQ`Y(My`e9yG2ZxoMdp=F)?^(HO#b_>L
zN@fKYY**s$tn^{F_7yiL8ZM!y_gAJCe6b5LF7LhLWpbs(SzW*qbyOdFo$hmY0XG{i
zdVrom{f8s1Z35+3Lu<csBegK@55+1$3dJeGC`RUHR=B<S3~9T-vL-8;(K-V-<z7l)
zy!Y4DTN3D>G^jtkVa`TL2%Qr5){So(ISZ0Ol{4F?oG2!>Yb`yBO{p`biOvq2aaV<q
z<rnoQ<w2Yj|DvZ*qGd#Dk0=!g-%$mlYjmdkG2g7{_puo0;t0idP;5*ZToXFxrTmK^
zZI(0??J_ajuf+rG_m;rO_(FWL=u8@b=ScQPwao*#t<;MnqmOTknw2FP>k^#n1Dq@`
z#f|png_ZDYe<Ny~;j<u;<}I|B?9UC}8<Zp_VYK1BhG$J;c+up=;dpK;?KR6)a1e-7
zbE;oP;gOBrIp+p`eD%#CMpvBhbgP%Qd3{!C3Wr%OB~|+Vs=RAJ5k|1qmaD@8!h+>Q
zz>yqXu<hlJ7F(WKaP0A$%&3L(2E{Vu%O86+Y}FtzAIMPVMes4P?RmLbcVA6x3z<Nf
zM0nq~*mh}Cq$g^g2->XP^1t#UIWd-E?hq^oTDO<;lGeEG!9O-CIR$O2+h-x#>oJb>
z7sDOVW#m$E((R|PG&fUN(8XJTM9TE#+*ZU}6=a{)&3krHc$1`UPIwVA7=EbCB}I9L
z2nd#4I366uKA{tRmJ3=bj7?O3+Mig2HwFF5GhM2FBM=nIX?}0z9v&;cUn!bjI)0Sz
z3ah;;dhr{ipbtb(IIafy<gKZ4ToAXl+qFij?=956>IF{*2W>x~qt2m$%IJNhK@Ku5
zJ^4xK&O$Ctt!XOxq({fm*=wEkCCB3C?J*hqCLTNv7gy<I<(75B^m+Z?j$zhbRGvgF
zlP-a2#un&fe!qWxi=tD=O(xhPZI~6*IX{Z)@yF4Ild|?;?M91C%oTA_X+bv8T^5;)
zq7UCI3LajHJe$m~TbWS<M92b)9__Ge2yBz=SxB_{ZL6{Rn3)#8eQ>mBeqH9^1sye0
z8@(-t%GS=m4m~DW*{mrdn5{0+ynEBU3_LEG4GLQ}*->xM9i|VXQMYo?+p9$4oFoNt
zbe*+d;x!GyI@thx^4j_A#ukapjkFve#9?#f7Gdyl(wlb2tXf(C?dt4rCWd_lKhYNt
z0Y1U|XRmzzOhUl`E$Re}N-ehVs2Bl1z0c>h+n1gG2=soW1@J<~>qc$`x)(T-?p<BK
ziXL-SD8$f$>0eN~n~OsQk>~Z^e=LZN37cM;{#t`fJw({lT^8k^S?eFPE?j#|ut`}f
z@^Ls<-NL^^QqEBT5eMiZy;`;s3phxI4c@!L6cXPa{T?_%b&(=JRYQZ36!B{C#V5=J
zhOLg&{=UofxKOy290})M8ZlT=>dnd6OC{j7C5*cCZw>m)C&_o7)8M`jnNqmUM!8Ij
zwS2hqrqRt^1*#M%xmZZAz$JU&urWqRz!OrC2QRL9lUp$X9h~;~XF(380O%b|IHxhs
zK8Bx)0=ECVRg1hr?f|2mJYQ+EaZCkGd^Fj0Qv<45#4=e;+4ofTUt+E7yu$WKP92s0
zeZ}CenjE9dN@*F@3HPY=5#nCsvC^Ql36Me=R2npL+V3{LG(~LRYBg!DM##IehfUP+
z?!;g{k)3b=;P?#%-PxNBsOc)7!of>hpYAL1*J7yL7^*IfjGtUh9_vY}tXXXyHCP5&
zY?A?DphIV3uz7QDrfZtvN8%j&X3|2^(CO*pg27{gN)kwQ-92?q*svBUMR!Br$3-x-
zTfd_C(0D-KZd7dk>s09ZrpU$F+t0QZGL_QW`OhC9MSKqnY16LAL*B-n%{yYDIR3I6
zRz#2A(B0h`$x_8KB#7?N+cdn<&I>7eFH}$WuZ3WHJC##}cchtdNW;T3eFWy68UBv(
z(>K}H7|1_>vfL!XIAd3w-6=#S%dT6Xhqg)@-*{gR?ZCOe$Wj>_dDYxUqL5N+dEr)&
z-;@=={Aljh8jWaqF0)hOR3u~=fFB<gW6N<8f_#H^DHOzOKgZ*{{G>sL?YmXz4VrXC
zxI0S9u(a4_%bL`V;rC&&|9yt0z8iC>?~`mW5?r0eY&~&*Q|>lpOsw8>*Z=ufs({JH
zNFGs$AKA^4Uo^|!1}HEU!VTdb;Sl0p?QRcL#+Q6-pX>*F7L>;ycQo?jJ$dg7GW09z
z`_LckWNzj<CS11M6Tkj6>mZuA0^=Lpz(>~gfKHcYxV>~>qh_EF;MMOgUH`3eg-<M>
zY+AIoxLX&oFQb#R(yHpUBQ(({-CmezYxK^byZYdWakE>vLBn1~SVgvBgUYIgAUOue
z7H1Tp*N?u)5!nr`Z>pbeC&BfNw<mIqL*6eS7HNw^9RzZbg;rwMD&57=+*}MsE>LrE
zoespuqjKd!Qmq_UII)u1-`)J}5!jn<*+#=5$adDNVJvmsuj`-~8V-@e)g|MaKG-04
zW3NHGvY)wWe0#ALE#BMwL}Ymjod>`{FvSj%Z$t(DWBF%HX8Hp`mi$u#9y)oifRPGV
z)LOG$6GQZxR==+G8gaNvCT5H&aK_0I_2-0`!wwRZj;Pcih6t02_x75uC`e(Y3(q6@
z?q0(uJvDWi#l@1b(U`PO$~zmU{hrrNf0Oq4Dk8y)0_ykRMl6!&M<htLDlSMNHlR+<
zlvDwgu0fyu(=K$;<oi`6#hM?5TwEQ1D#^H@TP9+MM+QGi6X!-85HbDF=rR%j7pjm%
z8rV$D`FD6BK+QKxaw0n35Q2tXWR$dz3R@Lgf}Y>MGsiOSn7ox3T9!6rXpy)V?q58X
zPv9)3mf$Tbq#E%k>Z_In{iBbGTf;6<AZ2e|WcWZ`f%E-DPge9Xw#E&Slp79U(t`=P
z7nGPKpAbS*2cjjW6A@@q@_j~CYfgKQP*Z&R+J&u37y*Y)JCzdSroS!$VIS5nO54Hq
zK0nICWDFu@SSz8RaF$jDyg#oy%3wySGNjFd8=vgnHI01xdfZ(r$lsX&)K9kG7=fWD
z>TkcaM6B*L6Q*!gfIOT#(zG>Tlu(?ylRP`CnUl=qw`t@x#2w89R?*Vp2SSrIZ*d)U
zlXe)fn7(bn%Gh(;(uIgj!#RoC(FnFV8#X>bO<cK)je4MnETVP;j{FSEau26g@qKAK
zH!_5}Vvt)%&v$(3&f-T9Uq*;IHR#K(b}|T69~#l#Py0ByT7n8CjYMT2t_B?$`6Vz-
z4#hEXfJNAV(%_gD6daOtYH-y5Y)bij1}1|dJLa3$KI7y9iBvZ4pQKOEupW4*bB>et
z?k~`}49$Sx+Aw=U1=`~I`uzMi)<$#Og!9}|R1^(81lgTb5Y^wT@%N@&bmj5x>|-dR
zcyag6YRm4bqz;Sk2g)D7nPldoJFyx<)Og+y<}mdk9G*DW&@(Ye{9h2vI1*-cc==vi
zb$ARM+YiW4l95PktQez8u~MM`=WmMRftqYRmV)j?>_24<-Ei-ZDZO+P^W!`CW%a*2
zGrA@P7OfaJy|a!q7EGfW(Vg`cV5NfWolnv|73s?ryEG2OJ?_N;{wx!QG5O+!29a38
z>6vaJ`8k;Koh3$PrOh5VlT6g+_ZgL+l#J1V#MG$>pZJrJ!!dUn+g_}qjOU-Tk&H8N
z33;ivove{&*qhuS`dZ%LlQ)ylt~GhG0qE{&xK{Pmk}d3&wkZt^`X<Q?4v6|D@juX=
zYFuN*TqON<Mj`}&5SFIBuf{&vS9(tuL{D4s7B*D&5$KcD!`*FG+wJ*2RD;g9yIo=a
zyw7U0+K?D+O(=Mur(1s3#M7M80YuisbK#++B^adywF#_;Ny5&Ln-)pPH)dt$)KW+$
z<qIP-VZ(R>MJOgBQv_L*^Gf-rf6Y_OV3I!;*lVs%k?wgw6OT`O^UF<F!N22z*LEjI
zH<)1+ZYR9S%c#1r@e>hh(9dnZ`t}<l%Z9Znq$v%Ft4K;KiK6|a7AoLWkya{j=?)M4
zJmtuREjmLmd+necvlt0uYbmx*hJ867&rzh3)K6ICtj^=}PWnvhlG4xfHO)4BSM(}V
zz=yO(t$hIucxSTa7)gj}DMdF{;_pq|72|uiC<dKh-@{)N`KZG0Nk7geX<lD;%BqgS
zv*%bD{&vTojb^Czjh!+Vs=eloh0QLyt3^y7Et({Jg-CLEUb+U%_6o?n4f$V~AzB5s
z%&_TLs7r?y|F9B7a41{&n=Zkf)bkksLk=@Iy!Hk#T`K?3_24S4PQtKOnVNf2<*y=N
zesLD<^c4UxCn(~$T(?Tgip$xUP&PJ~HV6MRU9Y!f$~IaQ&YG?#R90;k-k>)zUAzKh
zPm7z@mXuqA4f7ck1t|tdmTmRQMNXI;gVS=7M|^21-ZqW#AI_f?wD9Vi%zHA8!vsjX
zGfe{K7e2Kv;^>VV@=@$@vnS)WvVnbws<H^Of@ysOo&xoqS?hXS)(lcwY@+|>uN=^v
z8<JphKVB%g7mFh-EGx&MOtCV*Uo9=5zhs;ZSAI6}6nDZ~*gT$CYlIMeUU!@!Yz-5X
zC$e6{^7NJG)_7TU%g%!j9-qI=2+oJatAPWStoI&Q-W6<YkPtps-VmKwU@gsC1t2XH
zqO7B@KHNkh^g@PELcUiyCc-;GaY2DOU79YNI2H7neT3YBdTVud!^!drNxUO!#(!9C
z1x&+S1qBhdeo=0&G|%!)e#Xp2POE+5;pVT^%-A3tA@~kD8#4JDob)#|EggdNC>;QQ
zh3AiZj$Le<i<rUKAD$dad<@Rq1V=vgp6f#M+WRKSr7ER@KbmEfm{(J1Jclw1c+B8w
zeW|ED$tFJ0K3i}NyK6Iy^yAQyYlo)8@AD-T7Z~a-xwYlaM(1D4#_E}1J{wfG=lPC1
zPr6F&l$*}9McnWA8<XD(kK2HeW<@~qGZT-X#Q9=Tr)@zvrJH+6pJl6B4*r<=&x~RW
zxJwZIO=|I%X->FVTXZq32vN5;cm&1YWO@1|XMS&PYV$kOjVS-l!>F{?*JlC@nVgix
zCFWNQ<b64czE8)mnvBU6&aad%hH}$Cca>_({H_CsUw1~h0soIUlKd+U7v`VCMDaTY
z>OWp?Yz`eovVa>Q<F^JYsbmK?LKdK#<P-HsX?J>e>qqwMi=<yawvRq&R(6HobZHjn
z-W9}IhyubF|AC7Jilm)oO(h5a!uJ9Me%|b_m(Qnf(-Gd>R<5Da#%RK61BC;dsE>KI
z7|tT6)OJC-j*%T@j-P&PKFV~SMi6LSB-7NuvW8%%h+Y+uX@i-V;K7b&021Kr=@aV7
zD-g(#>^@qF@zNV+o>w%{;Zm?G1dk(yHYsZz>3k2Hh$w60K$B>Nm4~>tGeMQDXjc>9
zcO3^!C5B3TNPI^y{0B)II!ls7oH}%z89(_tyaiKesY#=qH@z&cTvV)cW;8#K$!!0u
zBvIZOQI|xZ_$Bm?>=q#qqYEJAGTj7MgpY3Cxp?@vJ~Bi_H-z7Net$ee<fnv|;bxLF
zPDNt;c(^)w+c}eHT3qFXT_;q?qy-aKSTrdxDJh5OW>8%PL#IRvwF;aZ1Avc|pt!9~
z>OCdKpJCrjK{K-)hJ=;zKKD|M)Ni$COs6s2)p6+}@&7=1em}u&s0Flw&`2W;4F)2%
zB19T^{uKa*!RQDrZ&2qRV+6h%J#SQ2Ii!+59K3P7(ZxX>v!X+P)f~-wr;EZh-of#D
zf&_g#0PRPL&UI;UAhxQtviYXj`}X1x=lj6Y^~~5jH17nHN$EnCe7Lj+bH4spSi>tH
zf>#vQc{~H=GUb2;?t#h*|6z|82C-Kxl5l3%A$<foHx-R#XXohR9D2GVQ;^b=?YKcj
z<Y-SUdCY@Q5%9)<RnSM_p-DkEI*%mL;B||w9ycmxjrX1TG+I{fB6sn+z$;EwtH=$0
z;nlW=aqz1fb0L%V<Bd?Ly25H$vH3S!!y~WaE^J;|uzrE0E{3~;d@L{?Tj4r5HhLn|
z$sTaVffali;!<^WRe=75yCFj`5)FhH=|PI&|4Slp1_bkys}BkajaqIx>(o6)a89VI
zQ}u3XUsoN5#Ovs3y0&yRVrTC8`syYTr5MB8dt$pgo~So*rl%R4kA~$t(MSgZP0EI#
zcf0)Fe=0s<%~G?g@kHn`<kro>{;I08t;BF(8}mT)ScnuqUGne@+DG8SqQ4b^U>rF4
z3?<CH05o;eq4F_d?liQ8%+iWM5ucSvc^Ylk&m0D?FdRxz9ml5J)WbWZ12?fSjBVsn
zHfnKa?i~!UJ)3c#c`I2nsY#F_EJa^f?|zBds4nbQ41)~Wqdajk$)rcQ$EHrc=F8-3
zdpm>th1t;KLF}07-gjovH*LjQj4(dKQqxkU0ANNtSbej33pBFOG-HVgv~9sCtUj=Z
zBg*h^(dzwwe02s11r{q4?7(q_2gyb9RUC8BfeT-F6U;=|+Go#596(kRS3Z%Dun%3?
zhHH~%%ToELkJe!ihvrMn)uvXIBl+rS;4?OiZeczP-dbhV{jBZZvIbXgWUV$_L7fdb
z1j<Jpq0zGI;Wznzi{_DC?#ROPqHnaX(cyFakjhy+dW1Q!@>Lv!aGU~q*e!BAO>tNk
zK0|s`@M}dj9&i2BJqTt|vk5|~T9S{gvFwCRk$FX<{LlPFw24hMq27KJUZEXEZJ@Ny
z^Y!BGnwn&Nmm`2ap(<iMpC%fF`P0Xi2e|4N&|V4sG5Q1kc<XJn?O?3KfU})tp#cTg
zz4fWxu<NpK*Y8qalCu|L8Oq%^zK0rUIu#bX+lTCov1ol_<_XDQLO^1oD3<daXwXbD
zqr4958$11h*rSWP{42OdLNKSe1p|yE`;90EcdrXlS8fgInCddHXXMBmN+Nu%6Tp_T
z|NoYt{L7tg;C5Q@e6x1-pdvgkeSFM%W(9hSL7J1}#VPMqo`eiTmp@EEKZtU}_2E5u
zVE6_*6)<1c;eBI5lt19Q-QV##GsOh1TekjY8q^iR`>yc6{(IFP!HbTd0oB{lgIAIa
zIDV%_$fyV)r*UHggf+(c+{?*-r3*2!_l=PUeXsp}?Yr1tEa2_=wbl%P_CtQb(><et
zqjaQu7phm*reZj}rf$u0L6&a>e9|V1$p-CxVmx^i8d5#8<0U|;>9-f%n^Qg<H_;HK
z4KaNi8{!RK7gQGE1rIAsczs(zM^<kP=Pc?IbAwfJSi25`x+qn3gYSOKOvVKWB|!O-
zO#wn;Q1j*xfbbV6;1Rwxgpx?!&wq#9eOPhNk{IVz>LfYLVl$7An&d%zP;$gQ=+Usn
zpctvKKbzFwy42NXGR+)4zOraV9Z`{I&YG@ne|JuXhfq8Pp{XS;_X72#F;yHFrjN>s
zHx(2}fHjtTkjk0&Wn+a~rsGm`AobHXo*qU-`&sFq%Sx>UxcntLG4AkiD+2h<D=BeB
z7U}8QVN~1i9~|wfMmfFs(sSyxP1%nR1_utr9}-L8O?twZ>nAM~aV>k?zXsSdl>#k%
zzUk26yR{Fu;2>4Xw<J?g-^bLapoUkQ(?`VXA%gIm3*-(X{OsW{(14<c_5#I2(mH#!
zQ7<>mQOyqCy#Q_k9>|>v-%9bT%6*dItiOGYiF|TrorXd^_|2kqAP?Z`D3HziPWEZg
zBvnONw3<FQp^%4PIW)gn{x@W@m#08Y#t{>sO!o34Oae<LNJjG~tOSHO>f~~4UMWc!
zwA}U?j!K%<7Go4;mKV#kZk7Z)U=X4;j(&AIpno^_H!533%Bz6A?yS(nAFU8;9~rSw
zEl50=0AXRObBl{prMn1(Hn(JOyx%=oFdIgVOZ~a~{{B}219^KcDQLoLuz_E$`+-Wn
z0PJxf!N6RZIHB4{y$8TAqU^3DeBG>@O(!yOQqXJf;;`kUoE#XjRw5n`9p{Uh);ffI
zAn9`n!NEyvwhv@*$2M1)z5Q4f+KU^WrMfkgLxQo9J$ub8#49qleAI4;Rm6vR@f)b2
zAe<8Bk54e83MU1bE?20=2a88`H9sd_zkA+wB5vUG{=pklXAAglK&ff^i&75LJa;l1
z@c1|17;8FER+@2@qZvhp@#tzjXOKw!pj=TcN>~$yy^px$6n7*Wuwfmux24|O%&98k
z8F{>+)mIE`xKa=`P3N#)`F%OwDHc!M-sBel1KCdK<@{?V?of<cCdxYq)&$3hbu7JU
zb`j$98Ir<ppa`fqSce~P-cJ-&vq$yz?>MF1ic#0S8Z5$oz2N5i@~IGl;IFc?xzZG6
ze&__U&si{0A-1C3ET@r)7T@5>{#v$8lYZ6wOcjw_q|8R7jFyHon%%~&S{Kya<~3ll
zF{;s0a{CAnt)4Tq4UA=2*J0nqTL@gNTMW1Z$&?1cDF%c>)Q7w$sLyWqAEr(N37I<9
z$fD{0s53(l87FdzLJ!1kak9us8EiVpYw1RX*r$G76qH^`1(4Cv(yca68z(yIn6C^J
zeREiUo^*U3%v30lF<(3xMu`FywPm}L9}qT0xR!**NC-HZi8K9q-zWn1XqiM6H>VJ&
zO&`TO$_${$tWj(B2u#;;NDZqILQU77`#8HVqNirb*I8@cytxqb@J=&}{fI~kh@P&u
z1`GCpjhRQG)rR0cgg3#WaO-_kdpI9o2_mAz;qzaoq|AF7i_^?7|LAC7=^`u?GLlOz
zmSozesN$EDq+ZhJ<lzd-pbzb$9I9Y1GD*16r2^cDX*6<Nx_BgVSw_~D#|uZsW}CH=
zpq6zG29$`%R$xyHdag_v#-=Dgn2!l>p~pX&u9&{5Z}y%GeH)YwK1`vlG&3K*r_J0e
zx=Gh6<CKQDtogYR+7vOe5zG~hB4+o}!A*Q_&a*G2pgz5zkA+rgt`BiCIe3uo($gfo
zX928#+P*UGFlTkCxgdKcHld_ijTbFn=EaA~De?&B+n$-o@4?EtRaJ!DLme#WYhNN@
zdy|~wSdAGHtA2<5eT0sQfhcDlkPPx7@57n5p$<xlh>X;MN>9yK+bgj^O7J9vz(QhY
zK}x|<j#0=lvT+dSk>rUJ`&cQK|5qK8vlPg6?lm<moJ`>0$esw~kVgMjtKT4^OcOHM
z!E3jQzPD5cQeK!HcvNifp@zm(Vb?~Lal6GvWr@deqiMqlRBtD0DVlwyndbKw{Z*$$
z7zmzV%swS2&=2e-a<Lyu{iU`XhlIn2oHfI#^J+ffa;U|j3Es)ebGU`mSR-;B0u8|G
z<*x-i8tG*m?S-A2Rj9T{58T`j8LA!fsc%hkti?YYVJ)&S=#O2OC#IJCvW6lbh3^8p
z6{MPXT70d{VxT_~Mo&h^cwvNwBH8^eSab4tCEFKdwmPX0lbEKTmHRs|o!YRe-r>{V
z*VHIDFC)1sQE!!B`)in+&1Gy;(}k_8Ej?M)qFujyV<~Bd8mBMM+#PS0s$5uLOZDUV
z@crtRUTj(l`E~Xcz&v=n@Ol!cO<*Sb-<i5o42qI$$%#dqED%b_448Ei=&MR+b!G0K
zjQi7GEjI@f=tT-p(M32<Wo>q4q3Xuot4mUJuu;$mT1X~)^L*@SO4TUsChAcZ&fYuR
zcNe=leiqEm<DI&EzFbrX)}U=8ZA8>C$>sEmwICN@bj_WKF=%79zOR6EqIO!FBY}gB
zlpbRee^=$A%*#1$0Q&AIJmT*YmjaJXO9RQ5$m<$p#78x(c#$yjab+}lNUY~&+l?AT
zVl+xdh~wD0I>d=&X0h`jc6BdXLxSFM)eKe9(lkT)^Q~|_)d+7QUG?Lxm1ZC=<drOp
zI}2TLNHadA1(~{U=aYlW2y>A2Rs#JLaF_B8QK{1K^y9Q70Uu-_vIJJ>wT7^4fgpJ)
z9||UEuI=){Bx(7b3u}|DU%mLnOip|1K;j=<QfOhB1y<ZJ6zyMoD38w)?>=+MzjO%<
zz7Q$OD4js<6^6|HePOC3!*2V?MQ)`5Z7!j@_hm6<$aT=ED}31LGid9NAyT+#?s1-W
zCk|aXL{ug#kojjkz_aiA?$yPm>`=UY_m>4Y`ScOBbbf$noml!WiNjAVzivykdE(oj
z^Z#g|n^c4nik0W$Du~tx;isD{)p-*hw3;j*PPbYprrS_-J?Vv0T&5<oPl?Lb3)BSK
z<$v<IY5MsN4ozFe6ljQ|v8Trrs=X46XixO=_nmXpfzC|+Ug;&uSB<lJ<|2xT%yEza
zwxZT^2X@UK37O@ZZqwz0)mWW?ai6NgolMo#R!|OMDh)$tjWM#xL#jBTo=oOZIUR`+
zmvip{Y|BOMw85WX@N3q1FwmzZHF_|jWB`@T^z7t(00Qf&Kd&SM5*)pJDt2%w{3e&0
z%b!AAAdb&rvMFKJV`r8iG432|d+Oz^Hk$UQ)tS{~)k%C6VRblL{e<IJkvcT(rUX*Y
zT5*}m<Xs+2BC3KzmI<}XOV5KPk~x=@^^5f~%_d>BuB<g)LkMSnU$EU$(L)7Ki(QUT
z{B`xk)HnzmwZq++U4h!x5`nL~U4>hk`_5e}Aeb0Lnt#avVFQ7qDD3$UA9Vk6u_N#_
z1WJ_=IM2|{Hvn6bID>CtP3R^>d7)oijR0sXRT_uHF#EMU2Y*V3*ghW?Gv|Rz64Bd)
z8LMqkqVkdvKZxzT5!OO_z|Q*(U6xu(`<0V5ihdg*U&Z*NZiWJt^62MGQ(F2bC84ec
zpy2?8D4&cY=2Y`$f~1pXTEYGZKJm{A%lnY<5(`sMf~9`}K78w}z?yo!3eHJHD>ePv
zq8oWQ_0sr-9%J*G_=0|=>MWWk?ORq{_ko)~z#aQhXbV2w;m?c^zQ6g3?OFLVzIIxJ
zuJdcR(ZEprD-dM_-OIO@x)dl4FBM+^t=*!PtbCWo<dx6-%sDSOS+AZNERmsh+AEwC
z9$PLNO<x^{dsU_GV;<4?UliI@ZPbDm#nw~<*7+#*yjQY=$XzmIHtCA5^JLSe6Soj=
zchD8rP1aK$h8BymZ(K8^RyxBX5S2d76Bw>ma7{qj9opa}QGlX6tAJTl{7Ese&XMVT
z+R6=MlEIjhBA*h2vCf$t#79>=EiIO95N_V7Z8W5<$y{N5NSx&hLX}^q-wXeabCm>o
zqLq<Q+Nl=xCpPoKjOR0y<xl1_Y%_nloH0mb4Rr*{H~l?QYbV2SeEqKWwT40zSDJ!$
zT|sg8=7VOLrKxddgm%N;2A3vaqrdnl%ie~KRi<qv+sokuY35XDjr{t=J;Ok)mW7P6
z@n%zww0_$;cHVaV^|5o>C;mI`d6rcro(nWit+KUA>-1Sr$j%k)8a}5P2HW$}rJ;RL
z1>i>JB!)RDq|+H0UoDhl&&U4p13Et{)`-vM<K79x*M12B%;v|@aJK-Ei1KP8YPXtx
z6M!zUD9^=Ul3IWdREnR!hoB+a@q(r6G^q5=74D4apCYA{<`q)lEf|Ch>o>JhN*D*h
zQ0n-2?ysRB47kAVz@5s%Iv3;rBZ<Po3VupPt)mktXp{~sZ#4}D6SC%V9zYY8DzY9G
z$*{u6i$IbCt4Xj-7YIO93YxzHIcYt$s`cYr(_~4phfh9xJ>mPDS+{@}*wgb-@ly`x
zKNWNuS9aFuJgX!<t&L2{Y!NmuXEDqvHE)1?H%v*?iPhOeOxH?Curs%5Q>sQvinBwx
z?V>_;7C;;Wi2N)bkL3Yv$AbPdZVD=9V3iFXpaHAh6!MAqQx-5VDMIV5kGKFIo6yyv
z{vEki<D+hMw&(O)-ypI5Jx=`~_Jjt~1U9hhA}L-$n)+vcviHV4veGt^P%(;Hi9*MH
zH}Ja+8oFWR4~GpeLWQuwY4!>9@HXLEExlIIA=uiF!%LpC3riDlN7k%fsKN(34~Byh
zxwuQJXpmut24r9+p@7@7Iv}Vf*FHah=e+`nT<sQGn3<FJs^T9YEykf9PRX0W|3~lK
z!!Yy3X0qfp+}SeIFeITXC*m~B!|NUf#LudQ{#_$}6lRWlb4`?D+!fqz8G;t-nbKHD
zk_=zs)y9g;E88WkQaFz0;&4wT4)}31l;7ekJJRB^8L_|xyolWVB*f`~liy6JbGeVx
zMQJ=b<hoEeuIC+e(azVI>n|v^LjAZ7ec_rU7)m+T2mxgKmb)2A=%a_Wpod|^19zJ2
zcIkb1s=Awd$JMT&BiKgGi%;f)%5N_sb#oIjH9W`P9z`1Z6lJ$UQxfGQg4uHOV;m3g
zK=3Njg|h&}Iq-FnGzq9-DxPIa`lRqZC{xe}+`Fn)P(CFTPHM~zx{KI*QVJNfVgHM5
z-C62^^{A#OfflU9v@!DpqN|9Yw;RiK%4!uqrFUHTX!?%&XLycc{q&1u0~ny#UF$%;
zBZnu&xz%%v_G2#bgML`alD~S;kVm1?;1}Q6g&-j1S0e~d;o>hrgS`-fFVMhJjRfNl
z@_ofZ$`PiODm_t>8Rk&FHb1~`z3Vcf9CgV9)%p+m0{Aj%V>y#kQa$1MwBM?KjGXME
zwjo|K1MJWhc<PN{6Su+W^BNsX<K9@s&mZi9(*b)Vnf~o6v-2vx0x{Ii{Q|p9aX09J
zkQs~`(O)39Q0yDJMQW(0{2+n!XDMn)W35Fah$>&J^5;hLWYoXjbvKWi|HZkT2A*C2
zPey@8aiS*YDgCorE8y~jC-1}&{-ah36zl7Id+f#@-%lmLjDl?BL>hrJD0%Gstr)8>
zGzikLF-P&D2)!4}`!=0HvkU!AKB$YjAed8edJiX5sqmv>pz({Lq7do6J5rxOU3bs*
z6_Nl<^(w0o57$x4LRDH<q90dMTiaknH|KJ43kZ?k&q-yOGpZNSEJ`2>@&WA8lHYmg
zW;v1x8FzwD$&9dIe<^)iq<_)YBDTHFgRqRTCH*)&V@9`0fs~d87berzi)alUoVlb~
zGkO_kHcUzaX^{TZ7}_O8DO#-M0CWe4TTO-x+xMGAAIX|0+|?vtFk8RsBEV7Yw}@Vt
z=v!_6BI7ErKSClVe8!>ay8(M-4|*b1tR#Eu1G&q*Rkrn<!VIh1A>1DRiYF#Hn+;;8
z^@3?5u+=h4`-&KoY>wFQT4;A<6o)6LczNViv1jLq8_<cxJyDSQlvsEn3Wg0s`NYn%
zbS3rKiJ1rmUme<L<k2_wyf+8Bu}1*zWJB;GcPnlBFiLlm1aXuO4v=vQ?#=cyPsdZ@
z>&4!kKUX$E?AJ3(WZs=)ZI*PL?{(|H?E3HeuecBpq1%LIvqxE0GY{xu*GG_={#vJX
z(4h1?T>4-gWU9(q%u4m%;l`w?`lcy0D1#~Cl!_HcVUF5H5afd)wWLbCibAEBMp0w@
z9(g)`325C+5GW)bQGiZS$*<LX9_DZ6m|DkFZ6qv$Q;OOgJA0nU#MGw$JW3_sNgB7G
zQaK@<Jz%NCQ<zj{CWbIiU*Lki-4e$w#LF;B`V?0SepQhK5n^D;Qb^OK7{=9=6`noN
zrE9rxDt`|g6+AJ^!sn8cqin6g!k^_DYAoHyI4|Ed`nF7$IS3%f!>YPtEciV}**XUh
z)Z@w3_7@v|K9s&)?Tya^6xNc`hLfQ?*;`rfK5;z-Xo6M&4gFVN*j>CI`QOv$goiBQ
zHFpucKR%t0yF8=ZtM>|0b*!w3Q9&r2Rli}&x*J8YSKVfM80k3Yy<DbN?4TPQP^vun
zn37y+EsgUNO9H9-TAhfbKbhGUMrcjiVHBzZRir?IZYe@aUT#~*lF^C>5dMT*=bqc5
zceqF{fr*z)CQoM{7G?#^8zIp{aw~Eop!sT2I0(Naq|9u-Es(OPG~DIIrVhKWFw~>(
zT8&l=IwVYJl{hjh$7xbnu|Rty8C5UaDyoX-SFh2^3j_K;pZ-QAneTQC-sbQ^9NY)V
zTV%&ms1`>Fnav4e*c(NJmhH0$c6}dN!Vp$jaY~6!s{9uA)HHlsVhzDqNYU_ACySqH
z#^dk4#5Q*7-I8)ZeSxi@2v?XOm0D3c=9ef4Hr!U?|M!o(Xp^M1BQfFylAK<=Mau~h
zjCXrr+yr3%<I4$F%rCLfda#v**P67&-z|7s^SvHCUK-?Kb(3yYEZjN>O-ML3c13iB
zHd&>|C-Can30z%<=sxAh4RS}hBX~69p$&!&MF{r<D{op6CUu64Kk4>QB8P-RfQgdn
zbR+%SW>xU)lhfZrYbh&8t^4|e;WgwUx&$?Xj0eD0xz&P`jx09`{7X-z(?wAwU-sCz
z<aD-N!mmR<Y;FQi7?${?#v=e-dPlMksUb#sh>Gfw9y-n{h;v2WJG0qAPMdJz%l}7H
zLIC~{OfCi5FZC-WG4a%x%$Zt2&^Q=Axp1|FlM!e-*Xv;7Rb1aKqvK7wQKLX~7bhL_
zdIjLb?eF0J_)Fyv$)VdswWOyDJ!nGZrr(JQ>$81@a*|*QO71u{ny+5ph#~xHUv9^%
zJ)a|oyz^S;R(7Hp@3e>dg1_Jd-sC3mT9S;?gUeH5!+l<UekLi!Az6d^&IKA5PZFvr
zi4CSQIbhM+<n-g)9+OUr{>~|rK(-Um;HCh&{70ox1PGM=F8K3xF0Q-c7t(x3k+uHf
z%u0le9*`cQQ0^q686oH_#y)GM7mU`pl^6}M6d8zxPc?=AEpO{mnHx~Qjpg05pfa@_
zlYw&lzOQGZ7%-Go7k04}$iporS_Z+ZHQYhK4xUlRq{sYMTKR`qsuFp!3;Ul&tvFSH
zFUdq(Z9M}1;WTk^HX*4!#ub%R`PJt8N1GCk%hf8LO^$)BVs{)nRKc!7)e>YU<%fVc
zjh;Np0=&oaul`Hj_I&v5>6S_1SbLg~#8wdB#nnqm$kzhnK~oaLQbgB6f}Goz&sJ^*
zhT><}`FpQnc_HF-6P3B3cbl7Y%UO0{_+8#Y(lmpb!iCM)iYA&%nlq3G^82IZNGv83
z(d+hMO-vZH)G0S@G<&eI;jG|fQI{ahCwpAV+%SgK9{$xqyVFv^?-Q!}IJM9?$y@V{
zbFa!q+O^Ozn%<V#N!fg<gKrIuNgW3>5==HZ!~DxrVZ(&~o;hsQ?T0E?M9~7&8wA0*
z&Fq9+mu@#q$)5GNhDVWj$%aARPNjPBR$}?OYEV@Il?~hpJ!}JOqUh};)_cwp61P1~
zTo7K9Eq50w#-`8a^FQ7<<MRm#5!imP`)d@>;nQ-yWX_Qg8aE%PD9BtoB&JJv{~5*^
zvzIGGr+h3FU5Z&_?i8QlT`vYc@m(E$Vh3&O?I_88*QA&F&DxX-Ta>Q*<<>WYWakuw
zrag?8B+fz$q+<NZQEglzzgU~EsvUkC%M1P3G8WxSue-M=PT#x4^_hW_R=@G#`Hna6
z&6zjT&K2_#*WR`M_q@^*9Km9<TTyM?n1<f{+YDL6_bjrv%P4xju1s^`(JB27y424!
z#@d73wPQJLQb#J+g?5))ujO{(j~K2+5*RPS_LKcei0htSZE@%-UPsu=CK4m6*{dzo
z#jo$%;G07KlNY|ZRH(@d>4ZaIdiu>78H85Q;L^QN{QunsW8p~lJ>^3IWlt7NWTZ?c
zz$1nRt_d*d1(7{foLkE-va-t7*?F2(<W^+qd4ea&00BY>B~OValk%sWCha_s>wbSi
zsPTD%e-PwHhnLUWXapNMIXCcPxh+5dlrQEUgLI?8%>ktE7pK>k{p*3Ml<7&^%)gj~
zG<xwZKCj)Ea7cNQZ7Gb5Zq9Gji=1r>-*wf2BEzYvuG=>!d73sKdTA#=C1#~Od&J|m
z4B+E)KHMbNmp}LQ{q^GjVCek2<hIkm{qeX*#?Ocy@OUahFm^Vd{l)e1c{?riAB@-V
z@B1fh;W=0=NiFGY`ndn{LOJ-}Lm|($<$=Onfn7T)u%LNB{)_mvE@<UjFKwkLU~NF|
zhW3jXKOicaJmn!hA;>_`Z=mH4gN7tU!Y3yyD+Qqo$^b<?9(kKDN(G4<FA-7T4zEp6
zg@Ul$l6^b(hfvz}fXo$P2<7n`1yM*A<%%C67J8gwu5koEbJ)-=3J+hVVtmx3y6bYl
zKKnU-)*jz^0*G^Z1#+)>nRMI*kZ&hX@4^a>WUDmUMZwe7`9y}7FW~OmtJ#SfK}&ej
zHbAUTh4QRi_h*39$cUFYN|%Z!+Au%oYmzLw@7%7cD;<tv)4iC&n?2QuU<K3*D-(}B
zs&AAmo;T~DTqbtwSO&?4x$X}WmVaSz(RR!;vaFTf9h^p$({Mnj^bf31z!=Y9b4&t&
z#IV$v((0<)cT)>9o#d2ja2~QbYzSGdGvOq3QM>l|-Z)S%TB#7t0w)^RoZ814u6cK&
zJ3)dL39d}a>^M_-Y?SpR0xisPW;LtlB|i0N1}RD5wUgIn2lr3=L(*J}JtNkk!5PAH
z8Ys#{TuKvcwSA$h%N2G4AjE6c8LLp{j0~B0E6CKOog|E(xxax=`_s$!30wSq2g>Bp
zs#$8SqHr&Yw}-?+ljtnF3P=A%(LQicvz!d4Uo2zs-jS<De7seS0~TGk%$@WCr>gDd
z#MKoUZ*G--g%bLx1TkPWS^h@QyitPv7lgSeNosRZdGwHD3;VnPMk8nmG*VV|D9Si6
z=b6qU0MI;^jBF?1AAX-9^LKB+7-|QF=lY(iNfIq#9GQb_P|msp+lF2kdaRR`NN%y?
zw?Zv@O$B-^Cf+Rdx5id^Tg{lT)pWF-Y<UW*9MWX}EpLC-3e;of^jnDK6(yB#WXP<9
zVI<x?h6XSqDHabvR^+()6d%=CP}bIw_0xGDHMzSMf$dWbEfX;^u#UEG*jfwz)8RSA
z75l_Zs6m!1*aLQVWvO8Mg+DoDe9)=)(w(aZter`U+7k%Qvv4m0d&;uTR`>JVc`(J=
zsjmv;(aIE39r4-);CxVEYL1%FAyJ`>`$_6VZF<y`qTd1ENp#JwL%B#i2F0LT)eI}B
zxg^;WOa%u8IOjtZhbGYp7K~^x<GzljrmJPX(}wsHO4qhzzn`YR$w`q152xssn;QPa
z*={2q%J7|$s|vBVdBEv=$7lm&znCHJgW0ey(vx}x<4>pP!!Z!2W?wSlxww;AWNK5Y
zTN99u9M%B*iXtboDdOK!oP$uCu;)!EN0tZ?e8VIc0Z9zP>3v!KjxKR~;^GfbJc6Q*
z{a@{w+axYVKRIn&*P!`rDd>i$YvQB&1zqNp`1<4;P^Vgr&rN94E*rt2`uw+yq5s~l
z+aPJ`(;bn}5$8aOA>d{;Kn3=d3SN@X3VDp(_BjAmHG1!#q~SBYp)&<z9Aj7CSB3kZ
zP%3f@fe1QQR)iYx3%b!VbXr4W%kCvt+Ka9EMiT3eDpyjv<iWJ*Pd0;9BB%8qy5B73
z&-;3IR@i<HYUztNC_8qm&Z7sakkxY9F;vVJ7{|7WaM~iNq3r)_4oEd7g4Z`?dV2p@
z0p|uTmvfbx2W!-hQiLVP3p31LHZsDfB&Y~md~e@SP$L<k%IngGsT@xoaw<9^FR4}^
zGinsP39qpn3@kCgvX3B2HtVLP^_^oFBext02Y=H-&bTY;?!?oC{_VD8FTR}LxXqpp
zrj>bEofwT=IQ-_j2qw|1P6Jv*nkm}Ud@%@oeGg-ESKhSOry@loI5XjU>@jlf9nH@v
zPySW8z3(5A+lH5cAaN<Ilbkg#R8PgL&NciPnhM1cg~Adpr@I+uMGQwR3%mPp(-8~?
z=hhtuddFqU@2lfHnr^<wO5*-yl3{7J6#OfEgHqb<cxvk9k`RSHsG6H)3!?^|xhoy8
zl`r*IvSF_tJ2|CS&yxC!*V`D!h=HWOG962g^@KN=sPvfA;y6T-hu?W#+p~oG%-WuM
zp@JS2af`EfBZ<4Q{H&=?4W4^gQhem%$J%;qnEW^>NjCN;7Zjfd2XkJ7{4`^xBn_mA
z4|TCqJ31Aocrz2zN-!BE%hm~11=s+l)1F-Te%CwnA&S^}Z_`DCfyEjXt%QL??l9s8
zf)kuPb|g=eCMbg?YKaT+h5lQx6&8{vYgqWO@Z^!e#_Lr3jTVG1eKHxx=vAUg-GXcM
z&$=kcK9;WRueT|C;ETCER1@c)(b&N3IdsX%9X9*CFIk&+ZJuG1$2g@WRS&eQOj>C)
zCnQ4A1<sRlUT3prA5QfdS`BMDP{+X9N+N4hAk^3e=sY-AaZ<Ua+^_Sep_;uYcqG>!
z8qD1KCQfYm$6!O=2Rpt}&u&E7O$hPQ66yr|@_%QFLh(1nb)K-B`N3L^rX{^$l#bm>
z6m!6{)tpa7hg*5agVWmD4-ZiGCS6NRpI+eWio${Ie2Bzf+9I!!`WuK6CVwVOowcZ|
zI_5rN$!F5=*9xI&f4wm4R4lCLdR1OR;NJ~R%ko17q(634X203*Us{U9b8NFT)uj)x
zKZTfNHi|bmZT;fcHL#{MoS$jujsa`v+;H=lmnvR_o0&U3$^M0-WdwLtoAceOS6NH$
z_M)i$X;EqIa+&<AEbXy@nQ=*JTP_!@-WE|#^;*=KVJT!Q`k9jct0l^-dXX2TklK-k
zQbkf8Rr*Xzfquu*=Fq`Kz|33Pm2WGTde{B3NRH&J%BiGn&ajk5x4ezUE3TjlUq`#j
z^399xYEj-Uj)Wq6egaTC%yZ*+HVDtYl7<+gUC!Gr!fKYFM-_Ho-!QV+No?O!H&dW+
zj$rt^>^|K!=YJ{mek1Nu`o0DxCS3HAz};6!Z8V^-QL@8<(uJyiVoAC2`~X1Hs84oK
zza{BImmD^u)ln0+9QfV5o9+5*CXHU!dABsY4-zif%r1-VE&{Grt%vUJ-V~NbSqI{(
z=CF?Smby%Pv+7G${$`TsUSf373@$N4%a2*qewmXP1-YynKLkeYE#b@7IO`2%L1sAI
zbRBzme>p+0XP;QQm}Bm4uzrlHRGYq2is~wsKhYp%SkV|N9g0wVDVXf0!{%J6ip83v
z{(}DfWk57d`^DU2v8%pt1S3%8a*%yS>YqBC_=V1tjo3P2lR@@5onNNAIIxFnu>So>
zek|$}5`?rq<9||9@tqGXdHs=sB6*^QFdGP!+#Rpp0=(x|jw|opo{GN&A`f{KAWI$b
z6PgzOiMEquedi~F&zmDTnuy%D^53r{Y#h|c&R=JrANSSd<g~N_JwlfM2cke-zm5s9
zoi-ipt7RW;c=-ZG6aQq{J$oLS5Drbq=jIHfWw+tPF4~tbh5+~6Lj5gp;VpE^ZA@xE
zY8!8VIcM8HvLC5=*t|7qO&SQYrs6sbTKnti?EBEwp~0wPWDt!C*0myo&T|kge;z`!
z5?+eRMRQPiT~KJCa}E{wFj77az9pXuvOO&s3h+puG{I=&Crzw{&!kuvKa+ywK54QU
z3P#aJ*b|BtjHFd`$Wr+Xw!p{~U+OPmv|b^9`lEWOQ{FAqV;YwD2q`3py+!#l&VEM5
znsh9~kNAnZN{SS_xcJF|p=Tk;+I~b#=!dy3?nBRGvxLe5I}Ch|qw&<h4oP8vW3dzJ
zi%;?X*qwx)3$O#P*uK$t*T4=?SvZ(?LVfY+^cd`R^3IUbGzH!_uwz`TKbUtyeevmk
z_5Rqs?&qCGWi*b)hi&XAM#T^ui=9wkd^$e{yM4T)Xq2(zl*<FnHg>@)a>w#cs<)F*
z_r~tLn|I`+GI0w0sexTeT(vtEJE6W{XCH&zPTp}^12_f#wSgUHg^tXz*a`IoJGvuw
z{AEo~edj86L<+(b_;Uk0;vzC0%sZifzF_Abhh6v_J4=k6@5rJ;r~{7;{5VhNdiq%W
zR!lIZWVt_n_A>4j{Lm;&Zq1Bu)EC4#nGXl^Z^eXr*B1i69sDy~@o@q&H@;D05a(q9
z59Z&B3HPosX86%Q{%NnQtApRDGl+Au;OSWWR!q2eogwhMoPRcwVh1YY8?}ai<@_uj
zoARw$aPLxM2A|tSKO0aH)v<4s8<tbF=#Hh|nhEzVHw1n==;w{1SO>pRa#$C*q(2ru
zVFAtD0x}!;6zFVDY}e3@!CeQT-$%i;@@h>5VWaY}<hZgwn1aFtnktV+Vkl?g28Pxv
z$7#tN!$t+dWsW+Sg~9-u>W@c%BIqt>A(HY~v#?QzSmd*{$6_c<ps5adB!>QS76K~o
zYZ5eS5er(G5Rb)h#e`goJQBlrIS<+405u63^@!Dh>;8jzxMD)CM;?jc>0%yYj4HUA
z1dW;mqf_9I<>8tMxh8oehUbfUh{3B^bD&X{V6Y0|SQ@TakgJkMAsAMFR%aRrdg}e1
zJJ6_1(1SvEFbh{q$fe06F{H~`=xk+tqclOM6)QRxKcPOCC6C0;UCuumHmuAYXp|;s
zgR&Hk#ZQ>v9$%Y;W_h=@BS`LLLc;|E;(RYtzfnccw$0EBB=gfA9yC60bLltHgtEiQ
ziu)+!R%MUOsk-+(w4B|4-?y7f>ovQ1n|}939KS7XNE>b_+Y#k2Wv9+yMQYdY<@p#w
z)88}MsIxX-S`D@sGhoZCOmL>tSMjZ8WCq!db4t2y(<*6z4lJq=sQdH3-S137A+^L!
zX}(8M`oo|8{G$_BzU3`zQ8RtDb7_#A(bs6^wCx~zMRt6OY-0|8k;$PeL4iIDa{~SJ
z5zMM&8~eIJNTy7L9<n%XnCj1kBxWK@zsbtwHj6!5FXnl<ZL4sqZ&LPVxj}_~tw#{x
z0C0zf;MDXCy=>&_y-#PFwgvTeB@($DYMXE_e|q?<E8k{a%BOML31#Lj9d(~nJUiwv
z0u4f|E8o6u2h<0D6}d^R1mg_h(Tj`X2-@;!X@o%iv5uqI<yY;3+rEqW?>sK{29^-m
z9D+>1Eh)qW(X@$VWryMJfE5bYl8=rUdF2SD7AT`t2^mdHPTy|{squ>Ejs;@x8KLWa
z0WivPW9dtKf8D3{l1X}nQ1*L4g)*91P=3k=^?98zJkI)mI94vbaS7zNyoD3WG_{{N
zj<W~tr3TOkcJn-z2n{y^t+g_7VK=_BG`)@?FyiQ|4u2zd9OaaX3CTxdSE3^{+=iWr
zM!B%CyM>+R%6Zx|YfnSsb~Ao?IH{!{il0-QOD{CsiXSKQc42n|KNESXGu5n}(&U5i
z6Vm?y{G^6|WAGDFXZZDY<;0LO3e}Z43I*e|G%*eY+HuvPA-{vr=UaU`madx5)5d&n
zN%4AT&*Rn8lw8^OA{C0%VRkEReQ7^x-jT9RA@uAw7t&_ToU1EqaCk;uPj7bMRELn8
zG*cZiIq(Q6Lv~W?ps8}vub28YPH}sP-cOR&EPT>`D`ejxmC`d}E2SD*N2K7&U4lFA
zIE^gotE;CGAz65~h)mrYSL(bJ70Ap_@06s^xW5<x9hBEjYBXlmJ}{8Fp)?P2F+DGt
z)A?;8%SP^FI7~X!DR(K!^Lr@c>n)jn>9^$kTJgK3B|UwEF%LttKaA+O7B*7SRv5We
zh_dm2q5bQMF$0l@dp935`oecs)s&Tqd7ZLgXr44`X`2E0*f%ofiQ+t)EUxJDwQXBm
zrpU>8l&wed?k5SPH8}4V`ad>_0wF3|7@VLQVAyQ}YOaoHVZsdqDW-k>EJ-?LZ;%)j
zf>}s>TZIS#6?u}a(+h`JPy|yy#;_wqn2K+I4Qny_hv{t7iieID(m6esbPiYPoa}nN
zZSVk4Fm1WIb?~UAH3pCU)7elt3Jr3x`I-q*NNjZKy!asQpwoVDAgahU$PV2lnG(92
zOIg<sbi%HUfi^I!aA}iVl)HhQ#weKH0qNN}UquO`42atd&FdLLqFk=c^E;D`NAT@`
zY;n0TG~CFe{`4pV^@PKEa~~A?>XqfbxY7Gkhg&AryiivgvK8)3e_?fVwAg(6c$bx#
zXSHJe)|z?iw$sErgX#0<8e0W6re#0_Qc2vkNQ#N$C}o^)H!jbFQt;xz$|WJ#G(JzB
zf;czi4yl|XCk1Y#4rvKR@*yP-S5&Dx$O8j<Nf`a!%Kx3`u>{t~yQ|l7YoQx`uuQbV
zf;;p<Amzad#an$4tpbD<oW_}d^-<&UpGW)+I2n1Dq0AH$0yHz1ar*}p0yi>~;ixB<
zXSyIT0_y3PnYti41b<ul#Fy8)AS{0mJpKChrw5*h=FDmTr~eqgeg69JpNEkJ={QqH
zZ~yrA;D1mc7+AT1bqv96fBtj({o(H;*j7STkc7C_h@FHafhVD9?HKr<qJ;P<Qp>a<
zU;;M5IN$#Ckh?cuetg(LLd$^AXIYcX6v+TdHb}BT|2N4Ck=~PB)hWF<oqU&qyC6(|
zS;h{^>Fx9x$xzBYWH0T(5|sy+{|Q*SH(xH$Z;{b6ViD>vh6<7FS?vJ6PK`r4V8-oM
zdlxJQd^>93njxB<sePW}27z84<7#|9j#mOhvlAj&5j}I<pc#PO;nvaByfb*0^6nMf
z%qv|Ur^|2#X0Hn|r^!32+|iVs)4O4Rphv`a{Ou-Q0**3Ffw|KKg;LDA9S|=@8KZ_!
z@$VvD6=RLEZ)d?#Ub1$Feeb1959Ep|6eENstq0<H5)9dzGPVupA?SL$dQE{hh4etS
ztH{PsZas-=gvejQ>at1;<~VtwM25l@hUUS3x>G2IH5l-kV^$kDFwNy%x<3$q*|=qf
z>k%7Ygf?NA6u6`9xp8k>vuMbuGJf~8!sp}k<y_Qb9AbH0Z5cf~`uK6YIa3XI=V3+~
z&5YUUESOyr?t-2$>niCpS0TDq8%M%n4!Sq#54`s%Gt=K4x&Z{L&h(AC0W?gDcTf#z
z%>?=2$nrcRW_Lm-nO5*UOg6cH)PPao!lk4yi%0VAu-N3ABie<<O2L$N<Ld8VvC>jZ
zr><wfyoxxn@|tMx9?dw*C==hP8Q_uuhj&m7m43_$xnuc?#af?+RjuJ_%23C}E{tVP
z;s|wN9>M|>#_eeM4(6dGqyBcpyfP0IY|oSz)tFUd-9YZmj2s)vaS=LyW^^rJJu}c0
z6FGOMLW3+~-TBQCD9bFbWf7#N(5Ld324Gv!$V7;ZgfuPX)w(C`cC}FOPy*>ovX`U-
zmAA@XCH?s#$tzNxFQS&3S?jK$?h2$s=<kgxV41RJq;FNCWc%({d*Uy!PHmlon>fe+
z2j?Amt3;O>?-mbyzenePh17gFQ^sl~%bVLeuLMOoUwF!{V|2%(uA){3aD=w#poMab
zG`?GHYHi{_rEBUHo;cLUX-jS1fNhl#9I1_R*B!KZ3lpDbcC9GQa8y_O!y{>q-UvNH
zUI{!U#jA9z@`<80fjjK1<dkwBG?Gc4>OE6F*Vu4KTS@*s#sy=4xd9*PTgS#3b5w{3
z(1)4B`<d%8USn>cZyo+-%ps==#~fkqjTKlSyT`m@GWG?-;zeGa@du9D0o$5%;Omr3
z`kc)Bxv3D|iN7B675?^QzT>bk&%*~StDYzplOsuQlu8;T>`0(zA(=G3p3!le>ZK(<
zYN3={TGK?dw-@SvC4KbRmOe{%?97g~%O!5dTu#i?=fdtp=<Q{L;+lafWkayKL<yz=
z!Hv^T;U^skBgC$Rcmut@nWS6=6f^x<+nr@jhGKvzja`^9)$i`H3qO&vVuZXbd4kD~
zWVEeTsvyS7I5VKhzxZ&Jlww_uak}YF@2umJN(XE!ru>9|O3QIICWi~vchi}bt}P!%
z%GzX>v2T~3o_sMsOHr5DA4l5B?t!8nDsyx&w(w4%f*0)~c-9H_H4H20(lIUFZP}tg
zIi%=}Nmzddk)t)Fh;8paLd>1O0QmI{0Hf#33BCWUvzYq35t`;%13;Z9g^MTb{ie|@
zatz>x86j4GQ=PO*I?W8qu8YR@s6P!M+N%KBbf1g+aAqJLO@)5~=iQqx`#3OQ=2#~!
z!DJzsvJK%o_v%9oBn{9nGVGtHSWryD9@{zZyAW0Hsb~f~%W5W$dq?T-5>4Z-w@Wxr
zR(BFEJnMaY^?@jau*KXr9czp*+bPBDK8v7F(R|JqFU%+zjB&UyBOw@Yx-XL<RBy8Z
zou$8*+`u4Tf9ZKi-+kPp9m1vWDt4{m(?2%-{+#yM&`~${hvw#-46{nDGryvbdj)q*
zADWS6zsXXLg1FH+2rR!0cI6zS7v6BAFE+KcJakWhkLBXWM^G+@KHl_V=km*L6#w)w
zYu$%&K$Ggeu;R6jfr|}>WaU@#w&_n6qfP=|Lg$w;e_p-=m$0Ygh-FvdCt@+F{<P)I
z^{W^eze_xbVO5|h!;6J)$CYeGpM(F^l`0=sWyrSGL%sNO0CHDYWUX^bakcyiK6*}&
zo-0<$o2|4|<S@O(Cn-|2@mE5;^umBXk81|~N+SlzM>x3>E|eFz1=HlX$KNSc+}|*j
zF9Y4K9M<UG0x6qk$!QM}rc}hzDq=3SaL!k*3iY}@N0(8;AQXQT-N{PssDy>L^)z@*
z>^rHc4N#I}9Q!s@=X<-wOKt1*^!?;ln6;%k-b|~G-pv(0_v&4#AvANK@4uRxs%c-_
zs%E%xr0<@g{B%aUS%%JNn1xm$sS4zEoRAf4DWvfn-dYFnZr%#NPCLslGSHO2^>e=T
z_O^C7@O3RYh;V<Opfs=08l=F-YjXg<N?)B+XINKVP^bh8WeuAVL&0rYXodTFlQEcX
zgobT0;mskrX~L&yVHUU3O{&pt_|G)q<8VvUJ1XeLsj&?T;g{k}&4ugAC60mEw0Lq-
zPV&Cp0!EEHDm0o?CEn(1GO!riFqx#3w;V&H$}_E28B&*H!yq1i-j+k^p1xCW8X@z5
zZrveA&Pq8<%vn`~PjW2##9+b+Qa|m^Z+-Ysi||n?5Ute9rQy?5>BiqQuIKuVQevTj
zWhC#vWm{F)*eZEe^mM;M-Rvt-oCs!CdSZ+)5i-8i6WS87Z7MRgO5#lM`qNp<%~7<L
zluWwI{J=Q3EyEv0e)}Qrpmub&!(Y>pvfJ^Ov_sqJ4kwwRr52e@NBSTBmUqBLsmy@w
z@)L0Ch6SksS;|fy{*^cA`3B8|kKA>Gyo`YwMVCdyARrZnysu9e>7&icYEPd{3A2b`
zo}v6DmwLn?I31~|=Jw2~sc}kPEydm_npogn$ReE(0*s-2oSog5&%_`i7JF6Z(>(5E
zaN|M5aG0AvsU-d%k)AP^!D%5B0W+5oDh3sos*fEfmsrIhp8<TA_{AV$e-AFcQ>5X@
zBh){lcRoVTaf}M%M{VOH=V<#&^dk&+n~w&qK?CWiDY>o(t^M_IwteXG(0~`-Gl)t!
zG_}Zpu^dF?&qHWd!b@HluNsBdH~~s`#(>b)jg<F;Z^5U8tm12mq0koS!VvK2FAPD{
zm!zQam!trh3qv+T!6>5Fe@8-5gOQ~023RVeH6|G2EMA|bXa#_6HA{`KW}zO!vc!iA
zApz(uikD$_8yIVnJ}p1REuJbNQtaa5meGnH3qgqeiWty$b6q@#9>-w;l?8U-`0PjH
zse&Dl0H@nxSEx@u`P*anT<F<>owkDQ8;v&=>@*0}jd@q7Pd+{0e+RpRyi+73O@X%+
z>=+lcjd@q7Pd=S)kKO4w?<5E%VKm-tV@KW#hM+BWh5F>v`99bk;~jY=6ppW4UTC(l
zb5@XP%ezuNPCnflyYpe*k@Z6PDe%t~>_XxyO<U{=^$9z?4|WH6$7zj*De$KXcAOO&
zQf;v-)F<rdhS>3!e>FY%ldISf3GJr9pDWlA7m>a(?+Wz^J99tm+-2;L7(3e$f&iTd
z9xC{8p3e1jTl`i`P^o0OJ$|^4dkH^K0)ksJ<16(AaZYAkWB#p}aO?V_!0!P66jywl
zfXt1r)ELBhnS;jsTQT9*HO34-I>tX~1)4ngl{$ksH*=QSf8w`d!maC!0>AzILr;Po
z$c(Sl8p!#XZ=3S1S#ax8V+Nl&L_c&Oi1OH1$_?Zc&6~FLTQlL-<wk+u0s2`b$mhYY
zlpJUrm-KDnD=eTpR2(z->@oU<T~|~Ye3hyL$#7NQn0|!`G*um_No@d>u)|mwtikJ@
zxgNvvEnLA6f2}Z_lFTuzR3BX8h{g;o44|p@xF>>UKMR3`g`9<z8U%sQh;1<}OrWU-
zxhIBpKMOSoYjYA*>JZdYDCgT^xMD)AL+**8-_JvK*y)@Em0AR)<C?!Q4_8cxwa7g&
zeBRAN@Lo8Rlb}+MfHy+hwme)jA=V@J#PGbEhv2O6e>n##B?&m>^1m$&S1gDX$-NMC
zD?2k41TFRc&K;;!C0zXrFpXKbVnQrS?ujAoXQ9E&_)1v<MhaHc7QaG$EJ^N(o!QSn
z8k$w+4phn#Kqrt~Tl@+W%*?MlVz^z}ZFK{ceN3RZV1RMESoi%bp=Zg)R+hd|#-%PM
z^s$5De`V^X$ogQO$C2(y1<&Xd`z>VbsCgMWXynVN<vD2x-};uBp!Wmcabvqol)AKN
z*={phU<XY>b8kWH(}{N%S9-rF_AQ0y!e<}*eWvG5$?MVL^AO&_%E1Q(Cl#MG(MB`s
zDf))u)e9X*N)2)484*GmPM?Zz`D};*Hskv&e_Y3~JL`0(5rlKp{rtQ8#uK2BI&Xt>
zil>D%CM>?QX`v*~73{RPf^ht0dL8oEF12U}i+xeyzK_l!iy!~|{N5E;exDK8I^i{y
z7HFB#e;m!PV1$Le+RnPdaA*<=VNZnZeKCh^)7lGfU8yaFu&|A@Ue0{=HMYs{lk&%W
zf6>$&cV5G^GlDJz<~ZoY#W6u4yue%ve)&rw!{Z#FUSM{G1-^gra+O!c2wula1@+#o
zq<E98p5G%_S`8mrnCus#Rs8a1#seLTS2H=+*b>#m{zAjsVAaYBenpj1FX74ylCf|Y
zSGpe8T;s36wb0NA*Ieo)TnU7o`m3(!e|<fSZ}L9Z+!Eh3aSIJ^#8;EBdSUk^d{GNK
zRUlV12w#nFuK8EsTWDy6Z!Yx`zMPp@KkhRe=%PNhCWJYqN^MQo1Q5#lg@(7T2|DCv
zh0rZXQ*x&H4C9X}@*Lsny&yUG(>s@4sW}%VE6+=-nz9QYb9FW4yFKxBUrp?Ke@CP5
zx1kHk)ueGH(F_UE_)M}0W8!pvQZdaB#fOo4LqQ9mBCEq3o{^fORW{;9`|*jHS1u4X
zU5$R8R{ss-cs{EH@&w40%Ma`6Y)pO&*Q=Huw}{`3T<R7|wi0aP1RurVO$zngQ0;97
zFap5svl)<i=1T|jH%i+nZTY60fAU?IzqiJf@v^3cWFjG_j<mFeK(vqeX(km?4po|O
zM=7rAr4{GJ1m44lONC0IW!&1K!<xn^Af2$%EF^!pUgGdZ+A<Zd7aY@vhNh+!sgO9e
z_N{IiB;tjinDvzS#eJQAD(dEo#7MMU>(;i0SHd{Au<=*FHIMG|#C2)`e=V@HCVNK=
zn~TR`V*s2l1#Y3niGSD)-c&!^_tUdoJUuwtpJhz_%?#N^xz!_=Dy{%>st!gH(Uc7i
zYvLec_6{mm{9f4(DKiO&2)i9Ct%aT%!-A7(d$#v3<e{fx`+kUtQbJGvL~#8U;WE#S
z3MA*`ZcG_VD~uAJ?RH*FTHD;ecoN68F;lsfo5W&28(T?D)?BkrfK@GKQLK(~fg6^m
z?Uj6R;3j;$y4Ec;V|CRlG|NqZ3=3DfAwapn3jpPV-w05lnW_L4nkRtP_xo4=bNqi_
zTC9MVVUipZmw~ht6qnD<AS{29+cpfo`ziDS%Mb+kH?y-dYj-y@xwYL}Ca1K?r8zX0
zzJ5W8vP?*nESsAH{UcET_~9c6kkZ}vPIn&<$G5|WSESCXV8NU2?OO+YL-QwSzUXY{
z0hB}N-hOnSAGJ1)+mRb-jmXXO@4(;o=>Z{d4>)myrw32)c+roWdF48G$};H5yT|-I
zUqizUxH*;+jnPRG_k(9{_UX&pr-RN$YZYDknZD$|m$$>e2RQ^AoI>ZiuRjjy2A8zX
zAR}P627oLZZFw;CIvA~pyPrRPe(%JUKR!<OdQ#4Vl*nLS4u9$$R5W&`p8?Z@?wO@F
zDCFTaw6v7CWvS55mJ++(vfGJiKqlLS!*zNcT5UAs6`^lAno{D{XoP-0=uh95H_sp`
zQGMt<+~Nuk_rq({i$n&^)1NvAr0m%{6h_Y!C{$-d7Gk5}S%qgQK1<RAD$>^poikb?
zBbM&RJglWjS^F+0zthU=*jdZ*Ync3>zv%8SzWJB2&mblx+oub1Q>IB8{BanxNpX()
zM{v(0c9)iN%AK_>&AQ$z&y!ac9+&*jAT<Fsmom^GLx1$0kjKR3^b;PZ%ls5op0JVn
z4CUgIGmzW|cHSEU^zvPnkD?9T%4H#g2ke?N2Xi8{j2&2x^VAr-Qhnl35JU6P^<JId
zqF_QDKDYtCoDPQTpwo^M*C~r~E7)mNyyP^-PUw#d1IyqoQL;67UI4ca9z{-x!sPfm
zb`(MBh<|Od6Y8Jqyz{?+cOl=tj(4nHc{BMvtYGIk;9_WuozQ=(^Dh1Z-o;(<ju$xP
zrwlx<VCTGo&}_|*%D9<&%Jxd`ai(6O<!<<>fUjJCE+M==?h#j60ce^$D;TCU)`Z3w
z3jI{!M+WiKfP>e7_`EOv5hwZ=E8lSiL_%eJYkwjN{gmOTF!8vLiKwYEYsU#)Bu5lS
z9Xp~R5?ARH3r;Hlru`Ub%|fA{7A=23HsW%oOD1x)(-g$hmC=wZae|JFQISd(;0lny
zWsozCfn2a5E->*rAQzVFoj?M@QfR&cOwcOEk`XJl2x+glH3oCRhU3kwjCZs00!+J$
znSU6g(pa)$r8vQ871_qjT(aSGGc%uCGSls1CI+u`z%tmcfEiFxY|Kn)gB6JF1M}39
znPHWg)lI>@rv#1^8j^<usDbZ>+fNDdz9@34WziHnhFhC;IBntNEwp{R+I$6^)NJ(X
zuD0lnZra-GxBA?0lr!0tuI;wXn6&E`rGGZD<E|AGQ*(vX!Ff}=xKc>8T=mRwHdyW_
zi%}a4ih`F<>Eq(oJEr;S8|-aYH}?@yxhZ#UfON{oIL=7PwX+S2eH;<Jt*|5Gv(JwH
zoC7(Va2H2gzb%_F8eR6T0ls{sg6ZILI&H$2+p#Mp)r=moD|=tR`F|eBu9>6*FMkT|
zN*fKB6uE9Sk%3QER`AxmJWc0e=Q9MG+5dVQfAiU#kY0;oyflKJ*>d(hPu@Oy^dG(6
z;>xAokWTkVN=n>2drKZmUoZXawTg5*lRk@c0`00>Oo+-)CyRFUwI2i1%?pc=EL^e+
z?7a90i3yJMen8K`H1F*+_vw}M27leZVX(p88jvaEbAc#zE`#XPo<=OV9j5cK%X88j
zxGkj5fZQQ5^Iu4;MXbzmIXZi>;*7_8dVs$_A$%Tyt)FIdLdi)J6mX44hhaKOq^y8$
zSWZUMOHKDc$0bT6loE3yK4TA0Kj0i(^sc~Ror?OLUD)GNbiAECrSyBI;eW`UQeq#K
zYD;fra)DQz)XyX>!$Sk|BRr?J@RZV9z*9;*!&8kxW2C|tZu|f~=YU-Sj~Fywou!`T
zl+mP(u0U&Vj^q-=7*cvV7|3sq%z9~~xFlMAN$1YOBqvHjQ*BAlultDJ?^T#+Qe4DJ
zZTA!=d8wD9-6%|Qfh3-)6n`cVffLp6j)e&q;R-^0hr&ci-Y85&?{<YrzYl^})+J@<
zePs#Qpo}*S$`T=Iqbw1<_rEQ4TX~Wfg6nJ@Suc9;e?4g}AqMjO%R=u{rwHjAb&BYH
zN1Y-hZ7)jq)i1p07&J>-al=`)MbA4<!~M)VbvL3@Xz3m58X>tsU4J7a_Atr0Re>e1
zM^eHvBwQO}wC_}<2<Z)~6d`e=N|B?xyGj9IY|;IKDn)2&P^Aco8&!(vU4cWRqUq&_
z##M@t-XIMZ66;mU-t;@~2}}8lphcrBgm)@kO5qJ4D+Tt5g$u&lVb<8wjfe^H%b8>V
zWo^8oT;XW!W^cJ)^9nln%iH09Kk_*?mtm3|69F@qAbKwqm(jR0Czr+5Aen!sct8I1
z`t3oo(ozH)KgQem--p-3=fh}%H7Ee5U)~OJT`LgE<An4unCZ`7r;ms4#!XxQcp(2s
zYmMpr1CpGkX>%_WF|mK5oto@4WRD@+4C^j5vDH6+ewv1r@4sHuR#9C$?o2*yy|7SQ
zG6zO#Gw2~HW}5DLjymykou7Yc`d`QAxWe~xe6lh|(wvwZc&9b$nXO%OX1P<0#t11b
z1xg=7^!Xqhg!H_ANRi9TpAV#LTGV`}hhKd2rquNQj~GX$`|iR~Pes^MBkhg@K&TYf
z>l$Fk&Pt9vKB4u`jXMpybk1%yK07EsE7Z|_kut4DxANx{h!2&@G6jFsl}P8S%;@wG
zSJK`H%#+q-p>)h==DG_9F~O7hB2UuFv)mHsQ@cl9Etbx$MY`oJcVt<%)4Ez@iZ^`%
z&b3%*v_cXXhFC~*gDa%eHE<Q3Al?zw8Mb$!?165tXyZii_CmB}llkODclWN*p}O?M
zp@kCSh7UOaw&sNqwitiBa2T$^Hs|UHlM?zAzL)qUg_}Z~fMM$j!!g16h3Q@0sc0Ja
zud6K_6C`KWUoN&a)VgGs3w~H2RdlV8uX-dcUz)s@smfk+1|I8MfyX=h`UIAIR-Rt-
zv8NQLu-3+KRHgtd%C+%79oRc&keVoB4eTt@O`hfhnD;3MN2Pz`*c=?*r9IXKFC{-G
zkIa?QFoomv<4ZKEMqRG2d>e>d4tu56BY&^6du}jMkR#bwvhpb2_jKVkWLn2mp0Sjr
z56Cf1%Iy-MZTF+ow;E^bMR2af`QDQ$(`i!0p9x_&#yhk)#nm`2)HQ!p%)>SO6a?5x
zm9#5Ts9#e$cNKs998Gh+n-Zo!Bulokoq`~whS5(qa->2O)~Jz7osQW0A=`-;`2nBD
zC>(PJI5F0<ujNv%sI;c@S+O5a_WR?LPTi)~b=OjOZ5K7W<m^-`=Netk+w56)1RIi=
zV@@dNWxE@WG+<KM*yl(?jwA&}uGDJ#a;*cK%9O;Kz*K)UaqeN3j$|zY_@2u`IehH&
z(iKGT$XY~Vcaog*l7Xf%K`s=U(mBN|KacJN!)?Fh5x}+d>1m|b{JB(b<0mgna8|8#
zaTt36SlaZlT<IO<YAY619*s?nW4p|UYYL=QytiK3=&0D!$9*D~oXZ}oH9gU{4P~Y+
z4JC6Tw~BwR&yzLVWS-EsjX75*?V~PkTMum7QPyf_JE|;fw4*FLG^c)MYfR}|s$@Qt
zhP_@Swf%#3s|N={5J>mZ*PWJF=`-ywb(@cESv7Z6k7EGN%sN5-qIb*tT=PJZebWPZ
zj}{SG=b}l6fkIz8hEzv7IcW;?O~(}TDtD5c9kYMLg7++Kt`qXrD&dWwyfJ$GR&DWA
zk<>+;V$!%(%Yqe#`u%<Kv_#?x7r1E8eYW^^^jlBa>gST0VVT#vQ&@7z`{ukZ5#^bk
zZNEUXcS6Z9YHj?|$6&y5B=x2pNxgl(;|+tadhm|5(>^yn%c&V6eMem=bw&~EFkk4&
zn9F}!wNYraUK?gJ{f)EJO5Mhilfp<G{d)_$P744va>b%|3%eeKm1eIzqSjPBub^P?
z^&R+SER+`{iEo*zs+a!*qS5z=UPrStLYX>q*BV-znAMpA{MJmdOpwd$Y)dEE&Xl>s
z_8hM(b0bjOWys^TJK&BnF((t(OL2U9F-U*q71<oN^U}Gs<JmI6N3J@W=Ac2@rF&yF
z4tCzO?XEYgy5yFm)vRv=@<FAH*$zg&9Xs<YhSeSN!Omfv*L5&^$3Jg7tL*BIwYj-s
zRe6$|iuJH-L&sB9U^d-xeTjOi<ue~yi&GH6qSft@MXQ99mUh;{`+Q^{K0PvvoGgFM
z0i<@%;)MrNLQ-XHl;HQix(Za+xKoaxI;SzW(GR~Bfo4@luhR>tPRAd~w8u@$sBk;z
znf`v<R4q`05YwepLWNv<AO3o`t(=lpc&A3?P;V+n=|fOwamDSc+QlIkH1yit&<#C}
z4rZ@z$_uAq4<wXUR1MY;5x>K6dJ}&Uky!_|2NBe{Q*)$tB0^a!Q0qOAP!2@U{wgHa
z<#;LVAm-iUJ+q!81zqT7f%Ew8X7f0M(xThuhJBE}z)%x43eMk!DbQ9#BN`F-Naa*L
zd}Kc7R{5+X5vpU9n{EVagtNV<ciS?kOM-wM2W(qI6&_qK<I}cQ6kIU9kH3H14sM(~
zG!DVVQiszkBQp+W{dX`oYEOEnfla%*o$;wDq-=UWNYDo16|RD0JrGE44Yrr@RYr;2
z^c4mW2EyR0Y#)p-+u^vKV#yT5_0V3)u+;QMFx+jYieLT0elugrQ+BX2jp-`A(kWfi
zU~^r;l`jA?V$K(1+WpATlstc=^gEEDRjjeF!(*B{7q9)A%V5)Gno_<B{{KLR_EfSZ
z>_CPlSO#LCa}e@4@bLEA?~iS37D<eu{HBEUcNH;;k&t~}xLuqn^uq2@-5|{{7I$QL
zqk!sh*r!+hg@hIg^*xFKAT1Jy6G5?<x1ay%oEm8?xQSx#_?IJ9`70F37DM_?Mx)2&
zFCLoT5Bv+Ffgh@uVUipZm%Gg%4VMd~9$&X(-XL^~me~gsm+NUE$^q=RwrU|$3j{`K
zUt^cC*CZ5|k)lHrmo#=E$N`<V{dOU@0)KrPHFCG|@$-Av-1+tOJaBDzgwP@@*`$c6
zHHT5`XWPbhKq8-&9}t^gI&=NdoE4}?lg_GXhvqr2c5chqB)=~3T${XNeX_S>m&Z=M
zn0$5;6HfA~nUUqSN4Mw==~5WaU1jL6s7I{sFye<7FDDJg7y)K9{xJLCRHDO6tbYLC
z16xU+HABxSCXU7(@(TtFE_0GYX@n`O1FlSef=!212XMRLRD$<pH}g;nxQ$$;*hi>#
z)C=4qkHk3GOfGFEw~>4RyRC$M=xtD!_;tv=RQ3MPCXTlmt#!gPF^$AMdCi(@B;3d=
z`vL`{aWC;|HD|kcHLtXt*IMEcyni+lcH*^0UE-BPad$3stl_l@ls5Z=TJjP8ppmeX
zKd4cc_$A;x*^~fmp|r6lzqRBe_-!QY#BYtd#4mE7>ulk7R1QhOZdDA~;1f_73$3(W
zeS#KX{Yb*CK0$?)p#hzvH02DaJUhS=5X+}TSDXdOem-$>)tyU|%gCX~bAOIKd``^P
z#gbX$HOORuon9n!&c6Gq!R<52YN`hj-`2(!maS_kYOaNp3LRZ7sF(u-T4`F~Pul^L
zeN7J(x8r7Cu-ia`O<UIj=3g`HbcUELPJ31om0qP3@X(l9E)!tN7-M;0ga7VLtdTmG
zV5cm$9plLK2TA|L>k=~B`G0ww@~<1`AA+uX?}`^0QR|G5URBGZiZ!*XY3nNu4d5JW
zq+{WBRy?2mU`4LVhRup{xo8+GgO6*gQ^VgtsrmgV3SEIRK)A|K_wn>uua*R8J#>({
z*0kj2d95;<T7+9HBP+W`oGS~ZQp=S5L1~C?D^WG*f&3X$<wpX%ZGU^j?HYQpeM9zJ
zJyf%%r?<i&`;-#1mf~xQ>SHwUSa&oaykxOVo&qypKpck=xhE=)c}mdH;-@T2YY@u2
z|H!xIWuTEdV+u;6P+GOEXMO0)^bD00TKhc=Is5t=akjqMM-i?3uvh5$dgdewj?720
zamHx-K%;3r6?A5nq<<IM?MYGvxWrK{{YHvv$#j{as__Ma*Z>78*pZ<Iae<<yby$gR
z;HZ^-*^FA*7g!4Z)u=sLsu7o1s$1X6QX~7aDYddMuw)G{=6X+-YQ!a$8u<_NBQ+!l
zanF&~h)W}F<WIz8Qzf0b*$@hHwC8MU#HHCZ@>dm0W*2ycT7MXIUf`d`E&J$$*HB^1
zt_6h@W-0}2`sek<)m5{V6$J%SG{F%yzp=mc-%}97%{7k^(nlP6b!Vul%vL2N1P^*L
zq?|}|BXl$Z8VbvIF7T9t$42<d;!*ZOb?3*M3^!;Djc-8b0<csDXcRQVTd}M4%2DJQ
zxE;-)FqQXGU4K_)${R1`Zi>{L6_MB56sf~&h;~Ae8?*RpD8^t#(vR$aQsJRIW6Wrp
z`&A&pv%=5&4`&?4^mdDN+ul)VHQLS*o_bH)w!MS#%qi@@?Hx?NEZnfy?rMmYBjbh2
zyMuO60y&}V`G?1OH=MUa&*$2?8diA>waH^Z?=jk)t$(y`KjwDE7Bms|L38MGj3&?u
z4f>sE!mQN9X2upY5w=bnbC4#&t}eoNq6xcF6Pp=Z&_vjRE$u;?;Fw2ydMBE2D>bp1
zu?0;qX@Cd)KMH^&7j52&Cj3fGY-Vgh6Cu-Z{vb_AEzsI`qKTFNKH1LLWlcCQ)ItB=
z@k$W5cRHem$8~DhO4ns2*brhK^!Gpqg2=xUEj+E$!e+jf{@5Gi%R|&)rTD+E4S3~$
zy0)@4%4eJNzjX1Tm+^;169P0emk}@}6qgNuA;o_SU-9j?kDp$2R@(60jP9R5UjBNK
zYCCW$7~A~*^(DQqXcPu$2o*Gc|F8M-@^jz5;A$1pU;O)H1sIf0fX?yVC%|8S)6F0J
z@+TI?7@B`t(XUBdt4Oi=fAe3)*JF}{VUo3TL>0+QE)X(!RG40MBqI-;m>6rjJ2AnO
z$oPNZ?bYc-YfBf7=+hn5@SY@qC2Zt2EJf5mNr(waXJnI~L5FT#_uRfGxzZEp(%#r)
z*mUpAF`}z70bx|!{R50`Sb|9L-%qjx1x9k;i6v;KCGKWy!xGji!}!m!g!Rgz|4uC7
zc3R?Y#x^X0s7y3ZvIIJ1efUl+;dff%ZpMGMETI)>SjPnwLxG)B?I6UKW_d|uSiKHg
zmPsvur%NS=$kY?x<iDmJQy_j7<XUW;is;H}EFkxICl8ctLlG2MTf8%QW9uuB{U;|6
zv+(849wL>t!6{B6lG1;@a4JFV`~Hu1EH%h5aInf*Xo6N0x1z3L52H+XY5!W3cpZOf
zv6ic0P%bbPo+&YXt~NiV-d+tsJL6oQgF1`c@_g30naT5<W#H1T`>843)0Bp$K8q&^
z9HK~Tv+JZcIMU=;FLcQ=H7$LUTMjnwIq+P%u`M)By|xs<5FOj5FrurCoUK3BNh`ZD
zIt~Di{Tyh`V~BkVahRba5ClcqH$;C{1#b?x-I`}vt=OT6vm#6VS`4upTN+bg*2FgU
za`GPBhKSR_Z-@~8ZgynjhKRL5TOuxQp&+>pk!6FjE!?^-q(Rq=0US8_6<NpW92RTj
ze7shmID_T8@n<;$DsnLCKdt?kue1Uq+mGr+nI8qfwmck3IYKN_62Lkuq$z*46{Vr6
zv;#vz+FBb5b}TJ;rq*JKxCy<Oa<o3($w~G+y}YC*M$V;I^q>4JJVGzc0GvfHW&4Ny
z^a60=HS(SndJV5@`fTqN{l`bW*FVvF-FSBFqnASu>pFPP3cZ?>3o<9uEBe1Y>b>EK
z-ka#X9`arrgMu)4PgthcMh<`e(Va`L?Em(t_g<dpz5a0QP5Cau;5~7L-T=zm)4bPk
zg)!~F%suIVCUgr|9@I9?*C%>$7{}mkiajSBY4l}+u29VNgb(gyibel+e!K@L&diLn
zhziKMLo|0rsk<#y$HjhTsdk{O>C<L&U7?yY?0{!`v*_OnN7+ZU-Q|DHJRQSQGKkt%
zbl3g#lvi}e($4|UCarkLLu_23n^V}}&-QH5zn#_m2;J+RZF#NTOZiqHiDfa_;oXK~
zB*w;KXt}In<qGAT)<!s)@fj83Lk_-=@wt@z7~=-bxSzP6b#RKPcCwWM&~mmL9i2@3
zj1Jdt;OOyTM<3Bn$DDuNSVgh+w3WipGNt<LXg;ID?Hf4z<xFRH$DCb*a@N!f*fQ%F
zqtbM?vkM();e*B5wBOEj_OR;g$3BPud~h2;H%S$xLzFuJ2-xPAt6KouAH~=A2QW#y
zzG6D^>AL+4{9eB4=6wXq@cqx<e>URDFN4O$2VU_=Y5-asXUKojA?<fg^quUsxo&T5
z&?`MVtQ$a7(FfU^F||V|=eeoDhUY>?te%V!d3^_j$!M|OhY_O&PQ)VHTeQ+wk>RoF
zd3vHJ<@L&554U`uZ|N*oi^B<v#~Cy>Dr2og77X^%gl!xW(j@fA<36T`RXnwjy17);
z_+eg+qI502wE2H3$;@3@PyvRW`Oo)rS0gPcUAU{IkMCdg13ENQj8`07cj^jUi&z&Q
z)|Fo*br%f4@iVyoD@DIUMR|wn{XU2G-DZVAUkq2MjGySzJA=nAOte33ztxjpW&&T0
zl(^~5Mj|ZjM+Y4^WVl6inxGG+<X{66y;jTuymaZ<&uxD{*g2nCIBpVtVw^-4COJfQ
zS(9qSmX4i1x=2@EQ3UclRhosVa^Ns*vF2^nyB)X!pOAF`u=scdxD=&|8`ndl(IVqV
z1XJE&zH}SXv>0t6vc;mQxGG!EVc3ecxdVI03X(M!Jv~BPBo!@6#BxeXdX8o<<-$ft
zFVGQ!(%663o^+jye|Ni%K!0SSW#18q?L=ERmyr)oeeMoS;U&kKh3SSt+2AMm^Q3uB
zko}N1-OeSgMn8kJ8gYMgDkPSKrp*rXIgs|A5KK8UMxVP<NWYIyk-25kk!wNJIMHuD
z=LF}JF?}e|x2a^VQ6L@ABsWOTOCgT8)^?q7d9;6m>DC|7^HNHX*A^lDv<jxnv_x-@
zR4{qLm!sXNV9G?STP;*D0HDCfJ1Q73Tr|TI3YZ#rqkgG{&nsWj8jw!R>?>b{^s(~A
zgvdI;3FV7G+9+Ry^!|fz_mwR<@7<>(nDm+WZWZZ16<tt#?N!l(B7l2ztdHxhrSK_)
zS&M&9s;mV1Mr9?WAE>MZ(#e(85;{GGp38|DbQI=X%8+<bB|RDait!!O73c}|n?OFH
zeiMj&TDdqAj;5EkJ6$3BnJ11MPfhova!#P1P|pd(je<^&?%|5gSawY3zd=POdOD$^
z6NnoXoseFj1KG|xJgK4+=qD5>0&%&bTb^QE`IDkVnx_X)IUAo;y9x9&=&2E(5Sh9i
z1KH1f)@(SjadAfpX0+j<nT-<6M->d7nS>)K!@;EnUP{8zDNMUvmGCqPzjv4G&3x+P
z*7(0=9V#+=?$-ay&1kRjU(;A3;Fq!2BoqNLm*K!A6PJ$6KPZ>=har@IyF9l4p8i^C
zHd?N4r&VvPO<tk4!}s^q$4UiA<$1lKX?gi$Mc`!;`AQF8zP=L~um7+9uIr{d*=sGE
zt()BnkuDBVxfHxzBGUS(VoE|}cGi*8j;=_=;DPUnOeC9JCKNNFWyVjd%T<qWps%X}
zW8{^VF~ayJ?hApJLT#P52#Fzv0|exsHtLtL*CZ4IF))__q#kLv5se|zkC(FT9TEaE
zG?y{hE+>Dz8_ACBHtybE;kiwKJUvmOHlWc!Rrl)vxlPiyAg64Xd56hmem|5<d7&*=
z*{<Z^u4-E*Dek0Ly8ENk-FHvdkEhS?7Q3LLgJ8Ok-#efOn*M^Oo8EN+KzR$@$Jg$s
zXRVE=>q-}DZ7p4azDN3WFP|(#S^=+g!ON$B@H~H6N0)i0>y=^POYqP9KHam1E}+X3
zoMep7kZ2uT=(@iA{PF#hW}>x9K7Y+0^1m-1Pyc(8Td-i1P1ybN^(j9<3(7{*Ss#=)
z(Ea&a_xsa-B-jC&rgxOff9Ttve{|x`56s1+iHq8M=Vn|4O{~SvyGSJX%0*0CJK~z7
z9V373(h(PsxZq5_H{0#!_Xw+yF)6?WWaoP*NW6B01qln99+TtG{H$ODeoiuSu3uOF
z_X8zagypO5U&)WkvPy7XY8gb}=cFFG7ui6pAxV&ac?=3|7}#*>5mrJ95e%tdbs0m?
z4uzTF^-qz*><-Ccz!vBZc!>?(@+DNsrz(F?<9evGJ3oBa7Hx<s(Gr2hrU?^=ecUBd
zqfwy)2~oeV^U`{SAr6ASOoSTRZyBc?q^Y_TnN=xNRbXX=FY8bMm7D2YQI7W!kPG|8
z<}7J)6u=)XH4iH3Adncxf-06Hklo%v(?+9J5Rycu1%lfirn#F$&N)h!W$%&&S$uy$
z5$=8JM{bfF$sCqcqw3GJRbHE6)$6L7xMY+^JLrjK4hJ@o_rp3OAPbgCVT;vvK}U92
zeH_ZptZ0Hgq{Bu>6||Qa0aoc_j!W_e^lXRhd)?dUC7I!;bf*d_W*x#od9op=J3|oJ
zoplnzhC3MFL={s<J5+(}b^2oDjhlZKpk7vPO@<F-ic2c^(W0t*4AKz=0tvQ*0n_Zx
zmT#z-Fsi5rXKyC@w;mE3n(h_&1+*ro-d{d>Jp^~l{H_5!4R*gv_bn=87c>B_>%NT&
zV^~mJC^OGY&2{8LF$4B$B?YNjiXSy*J+B(G7-+jHOP@GIQw#fD6UAdovKD`kTCRyL
zV7_>&={Y&7JjimqFIk#c!EN1KFu`~gW18t=nkNNC0<KcJt&4MLw{`K}9ojEUi{7uJ
zj?y2ElD3wA&R)PJxq5KO`<285uR;h5{O?UgF)=B(AULzCXTxq1toF*e1qq9+WCuiW
z*}p9|g(n~;WvMii5Z_i=WQ~7SE;zY5my-un#AZ9TafgU(F2%H2lhNO-&X_9(o$gd;
zq~yeVyQ<EauIoAHL0XXPqkP8<*oN!apP<nuQV<Bh9#+*M<C_EAFlYk`sa%=A;4b^O
zVv<T!0N#%ojS2`7^JWQn+{4YOYB*TR9LY`>ybz`A38f?hNdaVbRkVNbZC<DsjH#`e
z+*A+{Wuv1H>zGh!Fzv!;Sy=IlR55CY(@LdsqVYm7N~c*0T@oO@FKIXvQXwb<PEyj6
zIBUP8d(%~{C4BqoWL*=O_1jk1m1MTE<?A#gXEr{h8nIc%7$g-X%e}v?WT^_yZBPm2
z0rZhg*nmsOi!SDwE{cE6T@<_E!kP3U24&Mi6<X;<ZnB`4t1V1qygy`1HT<$wCLoEB
z>Wg&;xxZ53wupe*u-F=sB@O?8#TuuAL5aC#v0Q$x@T$=ZC2|#nAevZ5!VKC(J}^$h
z%-BR9OVsLptL$88gc>A_Afet-%&&O=g=ar*?gqU|$xFI+x+8z%b8NH=DCgagA(L{(
zEFnMC9hvA9T@uR)cT{7I0N7CTWK~Fu)d#&%Ee55kt$U+d=!w-u?bsCz{kh9AdbpI5
z2Bq7@SB-8<njV<d+C&NJLPD)Fi+qJ!P*hEJD7WU>?ZTy)sEkW8^Jq1NOUm*0_eGJC
zy)jRBW53qbF5Q2fr<X1$@9Z32BbO!>wFqNXefZEoK)P{gAXdBSp-LRfggtcdagc!A
zin}qN0-{`9{knc_l>}K=^ZEAi%Ca&{K)S&&iFd)Usb^>sA1OUD%~G%)3zQ_h9t|8g
zM&ffbfvr_HbI(B1_Ta*q^A~}oRY!6Ae3M@X+M|8LQ__FX;EN|0<riE$DEMHhFdQqb
zTA+rluK#ht&#~PSyrAu72Z0r}#fX`YE~gW{n{H;R)>CGq{UTF6;n4XY*E*owBiuOX
zYP#_X#U$lTS{*`(>*FnZgvO&T0!zysRh~i2dHciXV*1l0I64=z)|2T~VC^htOzl@3
z;XTcr$$5X;jZu$5sn*7zfIeaJxjJWM98!E+j9KMkrVd%v_s29j7G#mWuK=V}+dGpX
z{`9l%ex--s)8ZVO{$G}U9kA%`zuk{{it7g8NJ^Xm8!~9z0<1fu68WVb><Ge9R!1ez
zeh_z3=5?{hZLkz@DX|oAr!##ma|OyNEfq>gHH3eirxaDw$gEgna_I>43<A&L0ng;c
z+%e}#zgNCjXZ0?V^`v1i(?Ll5^vci@uZ(>EmG329bDcnRJ(kj2o7hE#J<?t9M#gXv
zkgoh3C$wKDd$Ik!ZOdCpK}PP)9XN%^KvK0TS{!S|31fk`h9kp<=T{~u@jW|Xe3}t?
zU$}o<h<IRwWoqJtb&j7^Dv;Ub#<lD;KELN<1(iZ@Q7Rl#J~7?@zrX#4U%~ZP$gZ59
zzw)nd=C$=`K{H1**yeE~+BQzAS{#vk#%=5cai^ywu$CBoz*9<>e8)_%F{oUebwZ1h
zUFr<*Fvk#BJsR+AW(G)j`f}qfN>X)_Z@GUwISgSkM?|?*rRORs9H&c$JM+}JzWG`q
z$LHJOy<i>R#o4=9>ANxgTj?qDENyz_{7dZ@?VXPHsH|BGM*~uQfp&X1dNb7YnW51}
z!8qQtn#P<=>A6@Hm`=<y)Phz(EItrU9TwNlu7vz{0%AXfEFm86M%FIH18L+=#)5yw
z8qwDCc-Ej?vH`|T4cauz!}H?;^FpJ}&yGtBi3=1n?x9ba?wlX*M-c?nasr?qRxwi8
zS1_4!+1LHA)tQ9i&(p{AcM|gJ?w?c$2zUCd#SVQ^^JdX>x_nppTN)&Dc&*}>onZ_4
zBYqSz7QrWf6vF#*IYl;mr>54sNPvHo2f;W>EktMTshF=l#qCyA(KLrge?!4Qj3Y_8
zNB@R`1B`(F$e$c=c#!n-@#eHOe$!p`-8BKJ(E<ZN3L^f-FAhLkh2xQKIWdXM-RtC(
zWPcEWXW2|SXwG$POOV>^^Y&d~xv|SbyMW0kH#w;~)&!)eIfy6!+^<Hmq8cFwe#K=e
zTQGhGwslm(u7A4J2@3ReqrD|RDN+6lY4CgUm$BC*6ah1r@#PH@mkhHRD3|&sBA0*a
zoG70DeE<C*NoXmeOaG?V{Pp^N_<opGa54xTVEOCYAx(f$!UT<)36F~X_<Q+s_%u`H
z1_}xdB%QN%Fpx&-Xfx-cFlnk7$fCf23RzSJP&q)w02Q(Lw&{&y@|G1^QL;f8WPgv6
zr?Er{O$tybCJundqF%QtgBM_Yql}jjC?c|dl2C0Ecj_DQY(_I3t-zx}F#CQG9+{@;
z_D{yRviJMJC95RFqc=`I|Mtz;<>jA*c(m@*$_<_>ucyu|BcTmT6oBaLwo-*gFEz3$
zSP}s~m>Tev5Iv!Ma#CBJR#vkk>O;yNc~%qnp%YnKW|TSK>P)uVM|n#FpbIxxPI)GO
z0b~i4nyZo=HqblKwq-?=N%;!I!1qCwPE{#c;3e1hjYA-Zsz83GQ^yGt@Ys!Y!QtCQ
zsf#Ob&afZ=kw9+0(V#?7Zq_hm*UZIl%ao1+UV2#0q3@%RA<k4WlmE250Yp@KC9m%%
zoh?J>n{Dr#?d!RPX3?4mig)M)Z{X^Rv}Zaie;|L!A-wVz6HxDH)_M=t+VeJ$7$07Z
zyOp_Ug2S+BYfxTz9d>NJ@?`CV1^xec^01Z{90Z2>RIbgI)%J=w)t$C;AN@biwEg;p
z_)GH9CWnqEV~=BsZ}FdHm#OMBbSgNTASJ=}dPl<6eQRPd=w#fpCK~$zYijL}jJrLl
zf4obUT{RIli85VVJyR9?6dKi=vz;Sf-#l6Iue`3?rktu`{5sB%F8R#(YK$3u?D%DU
z)A|lNle8P?oMzDD+f1dF&kKrWQt_rF&Ui`!xgCf(33AdTS~iHSAyfpBNeC!LYyww<
z*XmbxSIFa}2)X2zJE-WeR)bans_x)$e`DqhWa8H5RC!gC_oR+o<Q4&ORIJQwVH@E_
z^&rn<6k?+nIu~$L96BEHft~UY+ks_>YoFhV807<^RqFsvHBZA_Q;{Xy&&(-rVX-vP
zlON=B#7K<Nc~^;*%AH4|zY~eYrh1A*(qJPJ+3avu=3!Sco365U?ZJxttF|OJe=f-o
z&#LOgvf^Zyh5>G%qT3fqm9+$>3qup42<^W!an#&S5EvRW3@InD?O#|nde__D=IuNA
zzgqBRjDgH84l8+4ANWj>i096cM~;>o<2A(87EanB<|{>h>REzwY09>A^SoLN+p%Bf
z$f{QYxKMDTotGc>&%SqNm7(mEe?Gzh2qGA>!T{u>ZH40QIcc<kIw4KV9j@;4Oh2MB
zXJhLh$;Nrr#vWbiR+`2BJznc~&VhMfs!$TI+$@{B$YhifT7}s*AK)2=i7?*JyNcxQ
z$Xe(t!xdkus9CqHclE+r3-g$`)qVEidZnGk2dGsfw#V*2HckkNoYj%le<RlM8dB1Q
zMX8gEL^+9Xd!TWcJ;H$hLu~V|0erVZlcIKEx?-khbOxVln%12|XQ77r_O3RK3q!${
zb=xdB;-gtd*?3yVi<|+$+y;V`ZcW<|bh=)_Cm2iDORLe(Z~J|b1EH`ln(VGGy>@8W
z;k)gU{Q17ul@3okuGe&_f3nk4ffh!uIPc;99%88=BDAbB^cSgWjf$OT9&I+PD@s12
zEPG*?+0U8iG!%=v#+5E-+C1?#)#hd8nOX2sNB4ZnlTXpA!N=GBO&sp}ciELo>*!g2
zDVmv|^Q_(@BToAHKn##u18XNKiIV897bFduy6RPB$3jYu4Qxx$f5jwQtvF5zWArTD
zm~!wa^Di4a$RwSJt)U!~Q!Vbo3Or_W|8yWXu{;K@^^;io1D`UjleD$Iu)G5g42g1P
zFg+JrzjLFM;@8afaT{gZ2+3o~vn~K7^O~?~czTj!>uH1%<eVmW*;{XbcVH#m7TK_O
zdjJw1s8{-`Jn~wQf2pEA$XaL9y3O2C`swu*PUz88Q(_^LNtlha4n4IH>s8Y*)2io1
zU#eTto|!zae1+A_-h-*&(hG7Lb*!+i&U)?b@y7np^yF7mtUb0nK38-mpW8Zd?}d_j
zcHf68@Ms>Xz>BjzxqSbv?_?wsw+>Qo9Ax3#ZV2Ph@x{kGf9CvK>>Pwyibt+L;;uLG
zwlW&#QS^1kqloI%a5FW}gVB1#m&{R%K|dL3>*Spk3ffPIt*>`1)KZ-gfY;=~9@j$m
z(IMO8&gteXuiJBB-kIky@*K#Ww}vXq`0E5I%(_@w^#CBy%Gy~KEvEIQ(r5KZ>ZO8V
zcfYic`A`|DLcLb&#*nkps~)Uc*f7~%ONE`k&n#X3_g^W%2LT~YhmP!f*?8`yuaP67
z^{jFCM;41Et+#q&BFDw-x4(Y@yN1!1u}uvVm*5;C4wnn09$>e5FCrQ?0Wp`9Wg=~V
z-}t)G0L2e~9QfhO%lX^O&+kef9P>svdHDAGLBJOr{uLY^T61thFssDj+xNq-N6w{%
z*BM@#bEV+r;IoH!^ZH2%4<p0{UgGtW)8g2TgO_}V*9B?lS7(nnAD$@*F99zLbb^#T
zQi5>;cnS6T+qYj{I0{WAv`#aiQzX=Xz|jDIr6PX=bPe#<K0XCn9NsNHkI&4jjN`P8
zW6CTY;x|Z(<Lnd8iqGFFeHWj`zhvTFDVRn(&We<@tFsDi5J#VS&XHD@pgvK^_k?no
z%uT3s{icrP$j#-k`n^gPiuA9-m62%#N)l&WaEypHs+#CLs(hmxw-e5ZbI*l;Q?e>j
zOVlsj-+rLnWGZt98{^!}Ge*GVc`F(BvUq-yCuv%2k)G-z7h2h)SF%k+8##G9#kO{x
zwRDQ}DA&*L#veR0!CAGmhOE-e`B3Ck2Ki|8!3FsgX|zYbCwVQ}LYG)cHq3mt*RB<o
zsh}#~WEbRTi%v<~qIJ>uQv6<jZBy3CBzO1R*3pHNuU<f3r-*p(BKp!rm6-K@74}GV
z2fKgHh`@r4pb+1_kiXu(z5I6y7^4&u!5{wl{Uv^(m0$|u#|h>D;{N>i;rExnQ>utM
zoN<`KO#BBv{rSg%Jo$=n8aTLwTWieV62ZYw`e1@)($-Kgs1OPC!u02VcP|#HDR3-&
zMX1TonIc#a02fWPFZQm?Ua^(|lBm4HrHFLyI<Yl}Tw(~pm<}ooAuwuF0qkj7ILVZG
zsv09jNx(raB@z2PeXQsg_cQbvDaN#1si{Wo8L}JBG+@^gWgjWZ_ij!37F!aNSpk!r
zrc>g6nr<pE?aVWE8!0M(+dwi=vso8e{w^_h>Edk(()Q98&|zAHAsz%t=p{*{l+_aY
zUV(8{`qCcKyKL#J#$mH+;jhhFsT98%OQmjXU!WP3)uZe-j^w0F$#b=DV`G|1sk5X_
zJ4xlnf-jYyyrkg{|FdQKS-1j6x|l#Olh#<mw)rImqtsWVV4Pcj5<F%nS{y?9{bU0#
zCD*xlxN^D_`w>Z&ia4mLAoP9Bh`ecIQfX$fJa12P8LvI~Rg7LpoI_uD3>yHm7>mdw
z*(Owut3Jlk(YBQkeSx!1ZL2{!)|Fu^$5SiFGOsE%<_p~w)2&PzEFzj!YWBp@>6slY
z(X<^(wDhDz*kHeZTEg>AUpw1ur+{AcyzxwkmcfcfJz9!p&&(NcNG?+YwCqvx^lB0&
z`E9I`oX+H@of(n|08TP1uPU&FQ+-|M>yzHz=}FosbJVSF+UoeN%}Ar@yOolj7Mwut
zS(B7sCQYQZOzJRcGEm5JwvFYcZ4|dGgvl3mlbj{=xCK3b<H~%m_$uYG*g*C+D;t)q
z_YQK$=7aH}-|$qc#`Rg(q{3}trdlF-#arwC0G843-jy|UZrF;MZi-b2X}4w8%xT<A
zWt#HX;cK!0@1$!ME0Xdfji-JdKX)7M^d!|kzWF-7^N40PpeJezz@oMCH-S<bEyl}c
z`O+|Y*eVKtjM+pjc@)ud%w~hu)kAy-X{S2F%0qzKM_S!StCosct_N@Y^yP0`0S+Z5
zQ(967Cd00p9qPEwox)_Jb`CC`AXjH3H<U4IrHBxGIk;fgZ)`1vOGxTc&U4~|v>5q0
zOSPy)8X+9Yle>JO|7*}7e}C>YSQ%4!@ZsAnmr3V;nT2Y#*YT&%n=X?+FzYon%c{$=
zvc7XtesbDT_>vVC%PU?5&w{jTtr%$nLB_IcEC?CbmCPA!oTJO9q4^iroop>791+@D
zp;af2z%_<{jg=d0gFIIrwh_u$Z{Zz^K$&+*i{E^0&01-}aav7w;JFp-&?c!>99zMs
z+E!kFYW+g-!33YE(Ts@px;`T+qnHei(w#Hnb{2#n4YSW-i9KpXm3Bm7l23QJpkMBx
z`Homn$K9&-lZ=<%YCU3CY^@RHOa8c&c#)V*QoHh8pr=K5@k;30$}wAOu5m>ca1<)T
zW(LCL73HDUaIK%yXCyIH53N<%f74e^Le6o2Ql4sZNN}dR3^~-n^4NDw=A9~zSZeBv
zsj1(_?(9^RSZv5=4Y0oIccx0H4LOrfQZ&si<D2a_%&dnA7Ar!56<^*>oIBd0!>G{q
zV`e29oM8|xH5xq5D*P5A7?J!^c100`<Xe|!a|2rGW0BQO$e(b_ylQ=H{iG_r7#i7s
zHPJ3a`Dr_M7{>rpY9D5*7OQNvnp>9EUv<ba?MGV1n>f}(6|j@3wcWtoPC2tXP|~OF
zC~i2CKn>ermF=_f0_A0E^apLFf?dhrxAtj}4(T+m`A`MsRO_XMY_En`^^nAOM>4a$
zdqfT`4rEtXYuoK0+8*B5zT;}8#6>!PRUoqN?1v>+%)n?etLr<6z6U<|_&hujs5S4_
zQ9V{}2<unN)Nwtpb6MNR-FvJA=*vKTdaMLRN>FzDSScMDt+ao!p$t}JR<y>3Kp}PP
zNtb_8`7k``$_60<Q&2S%N|Fzy!A%q;Xby2eC}LqWozyGI1QZ!`^xI~0db>G)ETI%K
z?GS0$>T|mXblgp+x%H{-?hx<-BpqGb!nE#O5YP)$)D)9v%!ho1a^+gVMt7|w?zWZ%
z9~yAnH6YB%bT|gUq%KlhWKkYF7yN5wvX-%NWT_9#%E3fU-`l6w+6g8?pj0a;3ex5+
zFq|aMH#S<4WcZCb)wps{hCBOzz+ua3<XTwd-cb&<!Cn(m^J9)u+KL?#n<?$*rL%KK
z&dys(FsC88_c5QJoM@Trm(TStoz$0hk47AK$`D$djRn96_RFb4b~k!Hk0nu8uF*AK
z?UZ<2#@6z^2|#Gbf!a(7Kp1AM+q+mGO*vsS2)o0RuUBOWsXf!C-i6_RU`xwociW_t
z{+5fY+h=&x!~1PSzh%ka#FAE*HigNb<3<uWb-vHz(`r`zu?(eTJf$Q{K_@#jaP5GT
zW}IRb<(Tenz(kHfLH630BQbM(Ue$pKJ#is<wT^}|-WoN@bkA|_^)gECnd6pn#WI!3
zD20sCNe1r&dv=|AyInkg16895lX!YrJgZ@sH<^&X-_F-Tl96)ayWDS*&NAoRGx&PL
z;5G4G4&&InoTE6llX+dqcNje$(-`Asqp#kzRDfgvJx?%zk^q&?+`?PMTYmZ(yfpx7
zP`B`oE1b(64~V%*RqzKYCDX2@tl>|_yZiBH<g$l9M5YhIpONc-4*raTKLUS<csuwr
zj(!Ax2;Uw2d7N*-AHsJVf5ySj!yl@AJb2l_AlSu9@dP8NxO-Kz4X`7(1_cNnkFzhJ
zvY`HHoxc@%gR<$221VD@l@naWg0;H|v0Owh+_m{@6_s2w+j`AP|Cpk3<gy3qM5Ye{
z^^xlisE>m`0@R6rcsrmzj(!Bx3Ev%1f1Gatb;5TW)W^Zk19c+A{Y^#{*PEEl>9<SN
zhs{|_)aPk}3FkURO1=BIA|+)#PZ6z|*X~+PN}}5gwe{**1O?#c_JH=m>X~ra0a_~4
z#{ezix(&2s@CN`b6>l48$>;}wmh#;O+9&Wuprw2_ftCz^ej3nDLTs0>SUo>(1O&4i
z*&`G^fjxrCjr_5C-gd@g)wA%D+29wdXSG!2iuEgSU3EOH)c4#ZZ*{F1n!808c_iOV
z6??Ds0$*|yyUKB{Bgij3_+_(V7V#H$a~EX&{@-0VJ}}vX@lm1=Lh*s=4u}t9KLW!?
z;rup)4}%|nLGY35X&b;F=34N3<a*hJ?_uoo(3_Y$Ua;M)Yadp=l(mF&*loN%&bfd;
zarSQExM};1y3umxyqN07C*#e?We;$OOdo_eBi9|s83%s^<`D6AKxZ8N2<i~NJFxRO
z--0`Y?>6v^gP(^yM2L9$ao&cWhYe7%hwwJ5n}rI0K9BP*=tFqh-BX`!4{YEM%=c3E
zYyC&y&&XvDe~3&^uEBy~K~G%x8|gIc|Gc^(m}c4tN-<wkzPo=Rg=12fNzTv93sD;8
z<m}ID^2yRw4*~C@dtbau0$`_rC%H;uVC{NeZjuxN%>+Ji^YD4$b_yu%GRa?m<A;C2
zm%qe+9~r<_<$r1D_sWJJ{&V<8ysEN8*aI#^!kQUnwg~eBq!u5LY$FkW(+<0c+eut0
zS-g}A(7j0(R%<Yn?&eh+0ee6QasX0qyYjL89(lZ%-~Y%8A>*c`AJg^v92LlKlgn=G
zE&q=i<Nocn{_yUML%N$!sRV>~JK)>$h`POhBtJFCPet9F8X{J?5s6#Q{wKYr|G$B{
zJ6O%X!#b&vbGM7PLnpu{UifE64%9pG7j1x&yxrt|%Z0xNs|v|g2J!e&*DXQB(2}D^
z^xLz~-Fj@B#|ua&;m4TAYe*t?Fwbe#JR$dcl6g!v(Jkh=TaRt?IH(wYjB%U-6!e2I
zaW1RIIpz3C#=*weEylT1i*3_b$U;Iqk}Phea9w%?sz%6rvg4G7$@~8SlCk~)m$BC*
z6qR8p6aqChmqBwPDSt}DKoErQ^A!C8neD&%34{T6*A=}j<R+d3Jq6kKzgSt55JWsw
zx~Zw@o_tK7#Pk%7V>n#Qsl=M3D2*Q^IADB*Vdg|dBxVyDCwd*2HA6?pna!YrzM`hq
znN%d?1WD_P%dpEqdVx;M3~wb5%YAsJkU$x8dR3k$At%@+-hbZ4m%z)T*ZMey-;b5T
z(b(vH3bz2Kn%xt-7DwxVTkt-|dbho=vCnJb@H|<j^H;cT;p?`nTeo$)b^nmmQe##n
zNgAzeAp2Gn)$#NGz?=XvrKWeU&M?acb4CE2Tj%P-52*QGk~LcwHvyxU*+rlK#J0ot
z4Gs5agbHPDWS6zqBoG2KHj^=uD~G0ZB8R4QBe$k>B+5RQQ8h6nm;V<m3jr{fp>I1U
zf7M#sZX7ofeb-m;F?k4^MHX+1K!9UgUIQ^ANM6Ldf)I?Pjy?8hT9|2VoWI{woYtk?
zy3DjBd;K7&*j+4Eu}&Rcdf3^djACcAVvGOIsKoN06xkpzSW!swBq(R(6_a9;J`OEQ
zA|&O#6Co>~X$v|=1y`PU6=La?R4To^e^;5rGCP%hd4(Z}WDNEgHHk?MikRfBH;Oz*
zk>Hb02IjqmQ&#-NuoObE5*0-whJ99)N;Cm)GmB*pd%48nqq5{gFOag%SS7()aap{z
zE+&i+tg>v2zX?((*6=o(BEd$?MPg->^&(}=4@u0Clp`Y&26QCz<)Ty07C)nxe{mRN
zM9ip55?PGOu`LqZVVHshd%=ihObD!t6k#uxc}=jFyhM>;FLRM#FGuN{;F__Lh=h=>
zlfJp6x!8-VslhYoagx3nuK8FbxEqXE$<T{VqReO}%(t`RPzquyLpd9}#cW7<varVp
z+FL4;fh1lrQsQNDAOR-qiv*mof0$?319addlU2SzMbZveiUi9sM~nf)_-v#HfZ}uM
zOIZaYu`vN<CdxD5F|5XgLqR~xlE$#|0Ko(=NMpez6y>c*f+_~yLlU$|Dxx9jf%1q1
z;6chN+A;=eBuapgSVdnFikG1T4lKps#VG(PrZNOz9kY`m91Le)x}qLTf0n_3)MR6k
zoJx*lFqA|RY~xXel#9KnlILO%&w^AR!&C}UvWS>dB;NY@=jz2lNwiUW)w_THN0Es-
z0lUGH&H&^5d^S7W-o8uIK+~_IiTr5`6y2|hlG6z@e5I+X=gq8H?4M1JYvrVas&*%B
zTQBCyu9`1Dx3BiwNn0yge?8b~<}Jq9LGWZiR5HpADwGK6DUBq=#ctKe!W~8!Zf|?L
z+Fdlq`*nLz?Y`Kl-qxSn&f~X#pVb)oeGP}^Z9Q+7LR_(!*sGV#`Qo@<uDDDdy{S*8
zlV6(82L{am*n$ZT(SL!SD7Mb4XY+Ye3^A>S-#RFa3Hk7H(4tiBf1e+<()x8e|D>v4
zn#D=ID8|iU_0Q^c_56TVjlKG~J)jgd0dputq!T=pbOsw(^4dE(tez=Y->+UZZ=33c
z`T-ORSb)o8WCArB%Rg>!U!@8aa+NBiC<MTKkoulh2a<&Gvs*)xlGEn<$vo#l<~gpB
zd60TP^WI-&e&yF{f3fQXe<6e$Kyfn;5R{@02OutOL2@=bon)o-PGbSWBIGp6@^Gt%
ztLoXM`Qvul%z==9?Y)-2A3n70+43J%^?urZI6u<I&1v=V{PgUnlls5atMf&3G}Rwx
zz|7XQIa}Aq=rm!)yztQ*qm2!Z>*Mp?Uthkv#l=l)t5iT%e>P+dd}GUyGzv!+d(GKz
zR9vH-0MBTwr<YDRIFbGBPS_j?3v;xYW5N;~Z0=P!r^|}O&g6#TzNT<dve*S^lh@LW
zgcPr(X@kb%_Y^#ETj}>0VpARnX21hJjp2ddu6V$QuIhapRrQvQmfal+%V4Kq6ACKH
zp3LgN{GfvRe{)fvET*68PYdadBw*G@SH1=3mu-IMD7I}n+7XVn=WVl?PG-w$a&|VG
z9#7=`vN~)2suzDO|6aEB>9XggpwH)Z7PX$%#iGV*XnA6dbHW;$AY{$)u`u6T?$#^-
z5*DMbf%Awp-3q+}SXyID*9gj5qOO7CRkjyz#DMMIe}D^Nt@8i{0^0GTYAXzCEAaZY
zbv5U~-%f6iMlWNgtv6G>j9kmrrqdROloZmaX}hmLo}8LD4xI;~M>{6;=<Zn#uPIy|
zosv##=;0I$c7FPK>=D5OCN&N56i4AHel<KjRw<<+p3*2hrTg%dZorAd0^<3bZH4aQ
zs&&b3e?0l8;Q=efQB$A3Hx|oo5PGIDp=bLQ*1boIIt`-^qoR)M?CW}3tAaxmnu7nj
z#N6LTUDeB%Wp&F23$y(0x1*2PO=Ha0r{&r=-&%0cz^p&NBMofqmucY6hr83jrk@76
z@O^slJrdEgiNmP<sHoSuDeG%>$A!$=xbJ0Cf50^0C?*;+Vf2+bYU4nM$e>ZkptX=e
zLu6pfp_$W@Uya^TOFfDy-(1WCD8>#c#*U&G`z<M^+F>;0d(f)E(uiXq=7zQ8HmIe`
zJ0x~jcgbwKo8!*G)%U7SNEbrx1ZN&{&t#c&Vt;whWWC&H<o~~WrVB?e?wK}r^p<<3
zf99;6o=#?etnQosoHc(v&P}rm*pOfoR89h1zKLSK9e{=tBoFR*ePqwkh2T>*E|HJ!
z;TO$V{k6p@mMCZ^A6O}4okH7solP7#v8>Ca>8ObwK03*_0Zf2lUB1LeuU&Bl3H5d`
z`0<Cy8L*-$-7xezUm~Jr?7oB5@&VYpf2bpfH(E_*1e|A+`Ru33d^yz}u%}J6D-e0p
zoYXV9kgS&N<Y-n`y^Bft0KLXa(bMWSuqgo7`CDF|_f`t|AZ_^pE5+)wXrCGL7-r;a
z%;+w7(4}##7orX<KcfiR;uzt`Yl9rgXYQ2J<~TW>w1qizvv^;PV>nbhs?uKiYlO`h
z7=p+9bd8GNJbSzQ+wU*;F8!?9MO~w{1dOVaNjvGom(AUG9oH2u@6I@B`F+IwnqEpj
ze7oIH@5Hw0O3#BcE^`>O^Wk;QxYxCub=u2sqXL-JKEWCH%i5RU)T}Ou?ymj+0T|!Z
zm%(Wv6#+GqG1(KBt-B<}e*vup>ipZ+AD(zZy2nr($p}uy&_k3oOWN~CVI)0~Z)a5g
zoA6ZE^3=p9l2(y~fy&>6iGn;zIeJPoD5B5P<45@4Qqkv@VCn)hjuc^JSzpaCn#Lq(
zxAF!P2CoUWDzH6n#3A{4!aG1NI0sBTr#{`ZI63CD(6W{(3eC7Re+|c;|MnqG!INgr
zpjFKjL}Id^VwhGKr+i6M#+rjtFe5}Sm%tm3zggr$O_6K(^*y;Fm19P=_&)!*$`z@J
z8^RvpiX`YLT+fw&qwA?^HAcN;Es39+6U0rw>dYpzAW4g|imLekn&4`e)|amQYS#5O
zC^N?)<Cq|sgolcjf8#A)a+54RVwz!uaRqXsGpy&WNZDk2*^(kAe0rlyfmCUXN|Fv3
zU9;r<eEZU_7j!}Gp{TL{h)1#iHQVPenIm}~VCIR(23wdlmx~l;9E#XpYRe&}1g^S{
z!pOg|CMeE8O!@|sUC@6!<V*fIc;h@X`%O<dMrmXmlg*~5e=`0Bc=2#9XXGzMXEV%!
zA4G7Sw}|lec#8ye;4KnP8FxPCYBvHZsZ4gZ=tF3iG%d3?9IFJ!$a9eQU`9qmeg3@e
zL4;$%%p_WIY86ANnd<am%gZUrq=Z8_YRSOrnbwi)L~XY?@Ycx-1`l`5vgDWjnvRwQ
zbSl17dohApf6_%TdAuo643_8DXuI)wZM*SFh#zFZERRcmb(&TC;NUTNnsZ5OVA9L1
zeFc}s7UGjVRvTBISuHIQBeWq81!t&+p_GP1Y*?T;ThVMH`BxV3%(;dQD37d7SzgN1
z9mFMUu6fH4dCRCs%`RFNJfoPVUuC9@n@zyQ%(0fbf2`=zwgjV4W6Y26-ZYqw!au#_
zx+<>P1PQ0V;f+REs1&suU%1s9p1ekLkb0~N+%j}U7nG^D-C`V_^80myGYxO&#u7NA
z7Q!{xnAHZSE5j1y7q7JJTK)dqKu*Cvd`=D%)W~59>6oR&M`_0ZY;h}|9#2J{ZOm}C
z)=Z*ne`Y`P4{IJ;7+~aOn;C!ou)2ux-qyBL06u(FN<IHd<gc&ikSW!nxRDKTVCbnb
z$V5sCOWbShXo_&Ow9>SSrchZW99PLg(mXC(yoefDT?c6gb|72S#QHj5nQ|Tb{_^3Y
z5aL;_vd(8ks+ZLU>m;#`vZ4d5BTh4i_7Ll6f8Q0Tha^GKhw<J`mPMJq>=oYcz%i$%
zkMz68hT_|h5-BB^)~3N=Lmw?M)tPqYh5=a3kRCNtxK*^AG20e^<m$-7HM2LQ2#Og`
zoMy)E6CzOFCrKh+GNjde^il}&9nj*nZ2*+qBUdeM8w;Ue&ijvntYDpUk~oLrh^)M7
ze|EwZ>uV^*w1q?B(a5BRL-O&NvN4C`V+2E^hHRsKWuK8%b9*CG`;81W6L6Q2)iZY*
zSv>`W*llG%TISWcp1&bS;*tTHA}JijeP-s!y?VMRS5ZlfxM607Nf0ew+h&Hqh$9VZ
zTUJK4VtQHavaSOhBV#e1nEf16z93}}e{qbmJ1xvm?GV*I#job}im!{}&t?hYy20Tr
zGn<t|SV<9%Q&a8wc8)=72%XPs*aEFf(0d5{bz_#G478sV3MD04?qmNLS2sy0G?6My
z8AK^S-SP<WIY!A8)^o%<4sJ7{^5(UIb}3JDR>#*<*pjO1f}yFXkR_((q&vP^fA7D(
zbidxdKK*+pDfA%@MQ?xq_T*pC?#IxI6VP_6$V>kD&(rTuKlWhjLkx2EhH)Ljq)H&b
zNlNl0DYRBoHFVCNh_s|x<)1(QICW2cs?Huri_t;{Sn%;dg_O*?ZY&P6Rj=0KT4jM)
z6!j2t4rt@F-GV9OpCRWIa}s;(f6D5kOe$u`z0qY}Yfdt&v_BhM%$xEtODUj2d>0Bm
zBc5Cn{iFt*0iJ;IM1S+7raxms8Jy<DX$agZ=PpQ(!eKdswump;n&#fZ{gIcaSk(0l
zk=CNA_?B?shb1gfQkXA#oZz;ogF@04KxWg;sBw(=BwV;(RQZ20x0@$~f6J$v<0G3B
z6dfrs!!1=OYTWivykyzJD%ytCj05uU3KK-f=1P+tTF*{$@y5+Y`xDGN#ED!}CT<Rh
zGln~`CeB*rLUF56xyK9(Dg2bOF|6&v&d|6NGNcr|gUBzfYoSQZ(pD?!TU)j4$AkF|
zX96VzLf6hWZ>~!-I0aO7e}I0+9zoFnt=U0}Cqprpen4Fd=Q=#c`U)DD<E9_on*VVM
zjb%=#8HI9;1En5dTtE5_djU(4JB469B)FfOdCTl5QTMwOPA24L9OrKjK(x&8xjxzZ
zA%dY;T<z2$%J|5J<U8w-@ZACucIr^s#vaxoK?#%M@;&84*%4Xyf1DC8qi~`$q{xtG
zGIKq-9;Ya8iK`U#ehwW4r`*1SpX(qMKZmJ{$3EceVw**e>yoe24D^h$M}1xLUOkly
zy-CSKb))a`b;&dJkRI`MG2&mR<d|oYOj5$z%jW@G@-?HmO37RyyC1M{%o)2=l>~()
z(wcWxrR;k4sgeloe@X5gRLNHj>GPgqO3tgN#3i#|aCtG(?$Nlz<2ip&<TePJc|5XY
zOS9g=<F$~Q$WEot170u1Y(2gUUeC~cnBnxW*Gqn^qf{@{5|XcKN8jc3l8@>j!Nc)b
zPl<Xs9_!^Z*1jBTy{~+7Az+_;iaUE)J~<9sr~NLc*i(`|f8+eIxU>$B>pZVZxvM|C
zvCTku8`Uh+LR;<8zPJxK)tGG6&gtv8%A_{Okf6HK!7{ZaDP2`yL-Uhsdu0*~SwmE7
z^|?VaW159iD^^L^Z&OQL=-kjc*NJUeTleeN4VY#9x|*GEE`QwyEqR4ErB@|4IZZT9
zxYxZT4OdG?e@dCP*<lh#x~Z);6i6D#YJ8-Clmt=QTUR%8ai!shnL2bDSuXU&4EYn^
zjQCUP^VL~F*Gsj#QEcr%Lpa|s-V?oX3gUNVfo{&KLjJJp(=2svMq&pY%<&Ds)ZNjg
ze^18!nBST%I^7(tn5K;=t{3ru2H(<p)gH`j9E~)se^JTR+<83WQ#%EsKOXS`jZ+Vw
zjv)6b_>@em3h~rfo$J=EaMJU+hTa`!w3sZNr{v(E9X+?_q#V2PuB-Rkek4Iz?Nh7I
z3C?wYaXi{+HlSWCf}$y7C;P$a*diaA1k#|2VGYL0ZTKg!ym|7|>`<y(#G{HkJZR$4
zz=LWof3i`|^eDd_dAkU4D3fcR*A?RvaY3I{O>Ta&a!P@o$z)MI{pS7jzr}J@C`%TK
zWEbdQE~W9IfDgZ=o{he)#5X|SP<@Zx55<jWVCH@>G-)LFlOarW+pwIazXu?d4Ze#&
z{_-0?{excqK!bfn`#)jmx3UaB{p<9z_ir13e<>FXfK`afR{-;qZ+mhH>;yug3+d!x
z5IZp=i!;)$gC(_es)xJk!+{A>bs=eys&o7Q*@p99GJP)T<5TC?Db%l+*=S`E)7|{s
zk`si2%Oe_e6vR%WA+So{V>7*R?zs5r)7vRgni{RmOWW=}HLsL7sJTg@UWd0Opl%_l
ze_FVq{#8PJQ#GKgS=?3N&0A;=G@N%=8%9kq%PPBgqr%rjxOR2hfH37w+7t)p3l%7j
zq!szv0p_GF-^r72>B2$0`Kqe6e_7wT%X4KqZ(;Nbt=B50t@dZZj*GZy4c{~wJlUGS
zp5k4XD<YQbE&szUPe@wJ%+|~oFd|%;fB%YZEb48^ByO1x>Sb4h$YR$X<voU8)wn$W
zcm28)k3G9R`dyb+LJcPctvkZXRoN)WTuHEi7dc&Tu6u1N#mF~kuv8&tS1QYutf6`v
z-z+ZME{}yyH$BRQtC~n<z$uwqV>cLo9BCU23cL02N(WfUz_l}e<wC7CJ%W=Ve;R0J
ztvC6Cb4**)Gsk<2g-%6Z>lJQlB+FNMY`zLp-*^mnK%K07tFW%~t!d_K>2}OxWPVrd
zGKaWtWsI8g-2|sAYo(_P-Z??pC=sPmt}RI^Uhu0gpT)72-xR&yw6(V{O=qgI=-y{q
zH`udAYulwOTk*~4%9foMUy|3Uf4$%G{cYcLhSe?c?lyGVZ+YH<Cn;AIi>ZWU&k)S}
za-Q~J+uwg2T-LgyPP25w_TI_|B|wJ4+b;9)SF?OBwO07p)7Z45I08a5$93K7R3ZiF
zT&AqE^M!t=c#_>g-EJn#5kV-^UfOemdrr>MF=$U6tTYhDK8<e1B3?^3f32^JjibKV
z*Gb#_pAlACP(u^jdei7^1ujF(T^G7${_Fb}yV5CHhs_e-)45eQT8)yd%l@)eoA#Zv
znJ+Um-ZV`aUJl(}o7=62+aA3!zy==<0rZ_8euHlzE*z|*p;*?;v|>Sbod4f;eBWAJ
zv&6es;szV_eVP1ssl&Cj9-H?y&iAOqrP^92qlG#gIkIl#S`DuMrj77_%4Ox3!Tkjj
z0yH$2!J!Hi0X3HqM>Z#y$H^p}f4GFt)6d_2eB)VYfuExL{POMXKW|iQqYYCW4*&S=
zExZ6Fm~!$U6+?v%zyIs-%iCWng+iG#7yiS)UuPt(VNTjW@~5Bq;h*^ON35*8JN!3R
z{jIX$hkqabh74{Ic8o{D3Qg=+g!w@-?|dfPNND952mg5zcS>d?Gj2^)e{kkCB^!a|
z1Okg2qxAqR1eDYYX=~<@cm`Gg0BC%Ezy|>Z=zNXf00f8jq3rZ-cz{&=hOh#JlXm5e
z#5eqsX}Hujwh#a7QX{rOPWs%x2~<P7_$HJh5CPlr0OcP_TY)viIm0qMSWCF@BzSg~
z93WYJ^r$g!_(*$Zym$&df1f3BXNN(u;N>G7hW5-m0+o@5fiN{tU5KZB@i_YRBLpjZ
z%m}$83`K0|lZ-fVrA9znC>5L1fDkAMEuS)@C*re&M6I@{3vYzF)HkJW#C@isa~}vY
zM~c^69?DFmUUUeO+FHspmQ7!pdPmjXlvaUuwtzvYWbT7DhVp>!e~q;fec)8>z3Qa&
zaiV1^rLtft(OreX0DQA`e8a;VICge*rEV-4IY{EE_y&qJqWQn+TqswnB9%ai9Hw%}
zRvzSYnf`^=YW#cuRiE<A+hiYX<Op%9c)Q7Aj*Sl3s-f#kq86b<)%OJ#xqU4!I`V<)
z%-h-<re%%Q)JkEvf7crFa9;H{;@B`;ZQkDa(z}wb5#c)PUizb{wF5ujf71nHBOlj#
zr@o#<&{^<3Iry;~OWJ553Kj;Wgzx9t%2WwbU#gL#^wA<2kLgmX6;dC5EpMW%ALDpe
ztzo(-k~9?E$aXGx$FeV#9*tG0v_ur1B2l()k~*YU{3Pvke_Y+ig9D~*?fs!;^v_4L
zaq4#{1IU}cj(!e&{_tVs!DCt=4Gb~fI;KDje82nq_o<0>2gAHoZcH<<s?ZXfD8_Aq
z?sww*B^T$Ik%Y^izJC|SeD}xjVTTWs-q|XMQ8PiLz`kRNAb-wH=0lp4C_8pHV&>4z
zz{jlFpR7IZe;GY@n`rm~qc{ZF3fTq`7?K~GgIQ_|)qT9tk1)J8YG8Xu58)}bRo(Dp
z22hkPoxb&<1JowWoK_YCBbdim#A5t>qZKgj)dtSZdiQqFq`@r2BswvC%)64stWdPj
zT78o_&xR^d2NoH=8EU;$(ot$jfbQ+S*D}<;jv+N5e<0#dsywSpl9*#3+$x7@!n04i
zyshEK^ty#$Wpc2pZE}jPO>YmPhv-l}*w1XOK&%B<;!lx1?h3!`)6)K&O9nc%uBVfW
zs@nn?hBIR-Dr0Z7{MaR(Jywc`?cIY@tQiL<+YL;%3>gE5OfU8O(zYTn@r1th{(w_k
zL#7EWf2pl;<(iXp7m^D2()se7t!~TE9jzSs4hTr^ubmVnRt?_l`BX&MYE{VfE`!cD
z)480y$U#<;(#537$uVh~;s6}8+)g7z@snykj#~7w$+=!Pf1h1KJ6~#=+HU5FR1b!F
zB?04|PbanyHZaOZ8(<&zMbLB0bdq7(ybbsmf3&(m#q)c>7sXwHF9!G0eVcOqpqXe?
z-nm2jp;M%mU7poDsRrea$K3hKFpFtS2?95?lU^Ukxh|bw<Qf?cGNltT&u#c}xRhKm
zsU(%@#vm)X#wov`t%BxBmO9Lx(`pL{b!P|*?Z<g!*6zk!E%SvYLF%xrd)7DCBB?nx
ze-Bogs%7DOD*)0%hl)(63sk8jg>=-V&QeEhOQ-Cl5|x@?CxXr6tqlooQ{-V1YJCkm
z!z@$SVE}cT;nz(99{-?g6IJ4ChvsY7qw1Q4zApIOA8Zm%wL3Vby$xgQb{z@4Wzx5m
zjn%9R+_4Luw)Wy%YgK^r8P{wpKN7|hf1fo4$r)-wDctCTLP{nqZIt45T~shFoJCW1
zQ*q_#Vp%F7s+W4_p;b0u3zN3bh;)%58dsC$v@+%1<9{pIw{A<bQEX3^phk)}m*4W}
z6{pzh4$GL0)^nX`kQt`QeA-pP;Jg#WW{?ouTs8&-+vmUOTMUMFSbEAC<*fD)f50rw
zpKIu|n#>3@hwC__te(yb1fjEMs;VnpNWYb53_g_eDj$rq91(_cbv|i#EbVpYj~Zr6
zAl{Rr3(dpULx=5Lli;PJNS4Jq%HvY$6d4+rcXd%*kC|rPqM3vX9O%K))=_g!EOx0I
zD4z&})-sEDrP7(n)=!<Y4OuK*f4pC5Bb8#_Y`E2xk>dAjTSkU;5SA7gK9nMYD(-9A
z91-HaXV}*~_j{xy`Xx<xzkDAflWe^nlcKc2<W1ad!H=)x+jI|9Vdka5(mhb2ECUB+
z6O@p>X9JI;Nl<$w8BUef=)gx#vB4Q2x%%h#Uk{|@uV0%&VRZ**K!l^@e`H^{po_<i
zAX<Bn0l)iBUOp^CenvJaWUw4zmU9g%niQB@2>pm?VWO_SjGbJfpAnfL1uER~@yq88
zjPeL+C?x@N17jwy5F3>5$LIM1U_!ECYv*jiGi>k*w%RhI)rPOio4RU^tLOP+9_FRq
z=bty0!-UAsfy9A<euWDPe||q@3Mn<Sa_fp~6=8!>3>u)-D=<1z+5=0$DB=4!MhRQD
z#_02G3q}dw9wQJivV+x(u;=A)I9e!J$`Ou9=pCGfwJY-iIcn(Kkpps{xy<EIVY}LH
z<k$zS^5O6Tya2F-(oT;qfTesN1}tT}2J8#iX24RuJy61=>AD2$e+%TufZda0ztfbU
z`n(J|sP2hiZeSJSG8+|RfmP?2R^}C09Vu;2-Vv-4zK>&-u=Q)KKF_vbmGHeU0eAub
zg7!WB1t2#Qfahf?SpAeKN&-OHFEEO8K5gwQFgjAY6;BD@$1zISp4S+Co^8P>;Y$+$
z*pI1BF=72HWbuL-f9G~(f&9PP4{CY%eqNRey&YL}NlNB&fMMQmK(qp`G5rSpGr%1w
z?O3V;m+*ZYxP<Ku;6Bf`0GIIXfva+Gb8Fx}FG~ULLYDJ>_L@WE(_+XlPcnX2jL?7V
z0<$tMmr*$unAI9x;9r5+k<yN#DrO1a$1zLTUe=g>o^8P_f8l#)LhwBQk`V0i593@8
zZf>0rJS|H>{ijS(qcUN?z$kk9CFwv<=T=M|cs`EKf#r3L&1YE_WDY#(q)M+RRnN*$
zrc_fIAY&7jD*(^RP!ou$45svv<?MfyVF690V!(X?I7do17ON=V#{o*%-hk-yYzs&U
z-yv7sfhaIze_x_{Ng5V1TyBTYr{yT&b4LtVuj`h%KM`Ys)w~O%Bqj&09T4LA+R~SK
zZ6TPG>i@haBD9LPd<eIlO_2(ov9{LTp_6_i1lmQfJNKJJ@?^fpm3L7-<x?H*tZnr9
zVeUg50yW>-+9%m0%B;d}pm_{lPBVP&mBo3sb$dLnfAG90N_eKExKk7*70$Ref=3y2
zl1J&s<Op$$%X{Sr;M8U*$K^3O!tolX?v*3ND`i)5g#RO~W2c^N0+)MuJD<b0aM%YA
zn~u_WoD<!+?T12nt@EwC#qB$(eB5J}Rx#Obb8f+YdKX&CDfmwBQp8i8BNnjO#KUu=
zYp@t3e@rbEEi~#Z48zG+)EV6JDxwr&Bx+X?Cx*j7rKEe2;Q(i*U{EsjrKqdCvO8lG
zv#M}vhR4Gvvsc1#SInKymz20yt0c4EsG`<Y_6>_R5o?=J!>)`H&Z%Kj365z)wxQ=>
z#)QXPc$H5ups6kQci>8<^zB_aBgg5`yR_-LfBJm;cKdo5nXAf}0~)T$M0>6?ej?@9
z>P*<%rgS)&CZ(BgK4NF&a=st9c{Og)Z44k0CYtO#%=MRoar0uIkKFxD_0v|?#<|5R
zq}Z%g$CS}{O=bym-FX=6+-r3w>`hbNe+uxIPXV5DOmWHf<Wy(jF~(qm_)?um?w!Bm
ze<|`9Rd<%!--R`Lv-+J*kmnU}4B1W{yW;J_G&M)Kd}y5}yBc$QU<w$TywV<&{pF@X
z+I%W#3-eXgX$-LYMu<PL;mK0<+z27zfY*Y8ve(TBm%BScTc*3SBM5_Y_701eh0+G+
zA@0SV->|PrU-v`bj%(98(B{>+JhL%?f5sWJ@MN`OZh$EBi@i?R+tC0TV7rtt^?d0K
z6T56@{PdF^-V{<ZkaC!^m2={lET<Eq<$_N<H_FSI*mO;yl$6up<yv<<<0|s$e{TC9
z0x-FDx#6lCKXOu?+vSDyeA$Eha^utGbM5(zez@N^obTtGv+#WP2O;?8XAqj^e;ci%
zreblfnD;jXHb1bzhl?(IQf_{%q1=@f6VG)Qo=m;&vGKl842-4cwue*p?Z-$~9!^(e
z&f^d1Y=8WuyJ|Bz;hj%k-+f26|B03RX_gB~^E<bedVPAyDE!nzT?vAA!Wzh_`}PAU
zx<fv=c2j=m`VjS$*G#J#J^L#IXqiQ3oUQ;2cb0~W&Te*9rrvn!>U1~5z_;6CVscyT
zbI0UyCM7vl2=sVLKF+~;_yvM^)_r+OKHRDoju4E%?5`Nr>)|$EN8a44Ocv{}&TODl
ze&M5CeC^#`f5~YUfN`Ap9~U!_H<z*3Bo&u}p(zuW+1Dh)0hqU)*dz@I1o9cRewVS=
zBo&tuULO;eT)71(m;T@+p8@cfp5Y|HfA&^6u|}(e^eWGe7!d;G2d{$#MvQOe>1D7B
z#xXr*TTpxll|A9;eYVE_2__qkV=vWee19*Ld_vt<nuN5TEwpaEAIz-<uLBy9l1+Ye
zuXK1D)`t(8W$irA_q<lm3CKh{Yuvm6&t!SZxC_nJA#+)H9yF72f<?@?<co-?e}rkl
zFOCiA&3*7_ye!*;kpfh)rflCVm&@M59v1IPbZ|MEW<JdDq>e)I$r!^jZkf-tTFr|s
zym0iUe{S>AXeH3-;#pH(LS+m0BA3b9DdQfEkG(8w@sS)ph+vh)+MAUIfiULyFTebC
zEdY?_MUbHtG8<V1+BLgP5hp?kfBLEhQy!EUS+{)>BThMw@y}1q-N*Nr|4b2=T7&@I
zKYqOM3nZ~Lg%_|QXxshyukQDkzqM5B>>eYfwAPqWb_YrFV`u4CSwBR|#3a(N^5@S#
zy5`BZDhopp+Uk~)lJhxRkyM2GWZhAO-nBlW@S!CrKjxP=TIk@Hme1)Ne?I1Pm1upv
z8{c_hf+MMC3MS)#9NrWl-h0!hTt`I^Sp@njNn%cS_tn%?Xd5M*3We_+95^64HKdzW
zTG|WB=PfOHS)YYc7=U&bCwY~UeBz*_{mRKX>^i8ETH&_1vl~&V1F{<xaCPC}QxLrt
z`Z^<0LR*|qo~c``mP*3NfAy_NiZ#0VF8g^-H-+;`!kJ>I=QOTjhvb=8E(K6dS0g%=
zy(}YDC3v*spqWds5_0FmydiBnhegXw<u;&{O9z1Y*-DGjFzCyh@vqxfspNRxmOUyZ
zfE;l&r%5`uRBmQHCR@!o=F)#ih=`-Kn+qf}*iw3%+k!s2@35mLe}&^HvCrr@tzdEy
z5R3ZRLck^KkL^g(RW7W@X}2jbRGQb?;B8t!rWuuon^X|T(Ipnl6_(4^B8|W*Ufz43
zj<emC%?QacX-we}%W>_bD#6>T5;3)cq8e`HE>~Q)0kBdL^RxGH#hMjRn^@mPEgJxV
z_I|8vKx+k2Ys&^sf4{+;E*sEmLDf<LZ$6vtO5@rRgV0i5DkJ)&-wf6>`0;K4?=O_Z
zpo>}UVlRNO-nT0zlb)~_j0mVF{FU{SWys1>lKZuFcLv-kp{lOZW(TPXms{1>CAVYg
z=d4QJRtuc<A_O&C22Oi`x*&d}5TSA#f^O+;79u*9FU_nbfArgFA(9J>wv2B(t-9k&
zZ)S>$<jt1>@8tkmcVs!LlxSfpt=Z%LuBo#1)VnnPPPM4hu<q|tSVzx5cR(Sd_k+=s
zZ;!<UAfFuSf4wE|f@23k&fO4`*RfNiCIY!NQY%r29^7YV>Y%t|EGBGwH&Ucd{ShmN
zUI~&Tf#E+Hf0)seZ|p2V&Z3e!jb^2^?NE#UNElkgo}xV1i9CS2S11%6JpntR_&%>L
zhh!=+!IRxVp2yHTffNO{F<iojbV?z{wG`BBFTs68zfoN_vXWbK7?rnWYx0UD(UT3-
zOBEDURmJ&IgY8a3&XdKPcEzPEneiKsjuIwoFTpEqe-tr~y$4hpe4IZ;y(N0ruDH;R
zE3_g|C~En`=kXW3pvry1-+SXP+crDehBp(5`}965wFbk6I^ki#wkU#=EtVMPQ##Vm
zv#s%pTedYGy<}UlNBAV$8n3uzTMd8sJpQt6HT->oZH*~DVp|Ol_t~P7(w>lh=$qDx
zLZDx$e|~3+5G6gO13THPq>f8tsX{V|0TLhA^Ys5rVZ)A~Vqx>^Z?gL*z5JO5b8h`_
z9QvcIBfEchf9Eo1gD~kS5~kdRutr#Rls2f(=*bQwj1Il3_mX&!Gtks*=ykHB!(dvv
zt7RGpWi$~d$0)}U@vv5|9SXgDv69PZ9hw~_f0f$fkH(vPz1!3u0P~>$Un=F;7{T(~
zm0YZ74um&e9$YkIVn$_VRw`+o&+KR%(T`WDtOw{vMqHG782t^cO5Qoe&(^A;$l+qB
z;H{9dEn98CrY^WGa)IB;==Mp)6iO4mr8`Y$Mvu*}1JfqThtXI^Sy7|zta%Q~iu!so
zf7V?no0BM;GbkTNV*_PRZ=`(^${vLbxC>=_5@mY|<-=%fplpNi=$?eK)70RbT`0Sg
zD7#ZAA4X#XW#fhS{z)iX-fXtJQ1&NL_NP!jjK&7a=%~90PeR#HZ3Au>%Hbr+;S|b;
z(bzy)$KvieAfu!L((eNKauVdrDUgqXf3bnF>PxNXfUJU0A?yPAdJ^R8DUc7Nv4Jx5
zrPgy$22Ty?E|700LB5>=`7jt;Agg$Wa-s#^P2#bxUs1Q}*U2t*{q6R`1K&7<3&IVx
zQ1!_|J}qiK^=J<M7d}ZD<}REoEq<X_AKAw<yy<Dx>?awcx2`?+dh`s6x4Ms>e?tMM
zVuwdQSNo)MC@@YK1D|vd)lgeJP`%gtB>Lm<0duq7nUkD!2;EYgKgr1_CC%thI?3TE
zLAq{;?>Wd(CT<tB$~}INlhifywZ~r7Y5oFI!Gt$?SaosFbE-M)Kh4b_Ngg50-%a2|
z%grA8E{c54X(p-#y};{^-EThIe}ph=Gz+`md>W}HK6HLx+af7=64U4@P{KX5s|{q&
zDwxmT)s9~`Np~it?<VmeXOz+b{d0D;0Z^c}|KHYLtRauXzoAx69R5j-JxDFktvB}J
z;CA-xJhihjKQ8?`s-D_F{Txh#5(fN3keY*#nlq3d6@490=cz1xQq%5%Ez(VR2vK_w
zQF{i_!?LeKYMs#VBqMcH`acAyI|!&d1L$GNCq+W@`#&5$04|rY*CZ4HIG1siGZX<b
zmw^#JCzs9YB&2_sJ^uRs(-RF$GyG)P=j;2^f1YZ#6_Yq8b@<2kC-*=ABVhPJ3Bsg3
z{PEl2_ou%l3Xw7;#(l%T-wKk3P)AZJ{M3%*FTc{mKk?yD*chxi{C8^l-8HAg;sF0X
z{EYNz4zpnhGgV5;0%qh>QjNqd0r$wmEG4A^VHF6UK-hmM!T=VpSA@ge$6NGr{GzA>
znZCMj9KQ*K61)0-r(6Bz)*jq!>II+@ETO{gnEeW>QWE_BAXKG6N~#N0yAf5p3)RiD
zUqMx%q9gsspei&Gpu12#ZAA673)RiDUqKasFv}i<Dkvhf=|c6q5!Lf9R5#Cl1y#-o
z2mT;bIevdK+lA_7BdV8OsBWJ93aZQ!F5os@POOP!8?23zSmP|ZW=CG!A{*T`tq9i+
zyCZ&)#^SO{j@1cJVw@qgi&wN$35+}>c(s_+!?n;si2V4)DZ<oaQkmFGsL&IiNacIP
z-cwH&?lfI<a3@i-jcwbuxv_0~V{2pl;*D+FwzIKq+qRRJ?|bk4GgDLd*1a`7x6gES
z^*LRJg$a%8Stv<Ot-8ejoyD(ECXu_V%Gu-!D4wIU$Y#NbMl5;Yu|Kttb#6A2>U>SB
zP04LN;mkt1-Pd`nz;=`xgr0AYZcQb@P>pP*DVH)}s?81R^+%RQ1hhg7X7$umx~Q^K
z$AXFk1)-G-*QA!;uqY;_Qx&Q!xzKL={|XbDo|$d2V9%68N&zBaQJ6)-w@BA?grs}|
zJWkFnRdS(oF;omc>tf=mkl?kwnBx?+vPR7fMYGCrB|K8e!o|c=If)0=!-g};C33Hs
zPh!?P>-dircZ1_W4q;~QY#B52x7vm^EUlYUWGQIf(5xVYCtI`)NZXZ!Cf;Z({?TDv
zOZMv}hk)(0^N?ot>Qp)ff+T6`Bn|lonEm>PPYeb(0EB_`n`tHhk!gudm^0-x$<GMU
zTfCsa<T+@1?@Pawcki!Czg+W$VlQzy94UOWstt1TqiZ1f3>dYuf-HDT!hA=Rr4@Xu
zj@F7_As+qT)g+K!KbXURm<ZVD$uY-*0SUu1vCb<f9h(Rm<P1D8P719&viS+DfbeyK
zwt6v0dAK2d^v!Hr7!jk~$Gs3g6e$&5gs7?BL9~+KZDNzKUh&AS-wyp<r^z+^(@UrY
z!fcOA&w;M)`wPssQ6Elho>+WP=>gWcS})#DQhb=&<FN-aLUwRXwua)bUUtMOvV<ik
z?=ahOCI>3n=3g}pj*-BhIFz}4fa@J#%J>xwQK*t&LL5-fC*3PtT~0fnhA@$&YwEt6
zn(j$7xBIpp87gMJIMoN58&zcDg`b4znn$9^3IQ0HUBksI{4a^e&XByjNco=rBibi~
zxV>dg%mL&Ic7|@NQ7BAS)Ar|n0|1<!Xb0>tL;1yB3So!%sNWSn`SqRzz)~ySjKlD6
zU7O`V(BNd6k0QUKrc}`BIAl21n_~H(VYcZkrWY1L&npM>n85o}14Xf*P{?)dt`qC|
z#JP=xw^%$@wexK*ULjLJF6~)-;RdnPYZ+M^_AR=Um2|6z;+BfVhIv_cXUQ6|)Z_mR
z>~Roq{gR9iM;&YjfR(vHDFbMP>>#|wV*CE?#vm6CWN};#heW$`O3r!HNZE=1?6D)#
zH7u#Dm`@jpPir4r_^BGd5&05M)`nO@NGfF)2Epu_*d-KSqCh@68CoM5|LW6cOg0KP
zs~wG>V;_6)sv_mYx<?yd=$mtcCDpD=s$wkhgiGe}su~aYStp$gyO;UtK<}^ALs2@u
z)=Mc3@C;m)MEFl%qQrUZHeUMUo6Mqri8U}l48QgKpi|`C9ydaIR}c55ZqsT6B<V$(
zUB|Gq=XULG**@bX_z$r~ikFV>qB8}iJ&bau$331NL75{mTWeXnEYWb@M^q#yg6WE>
zxN=H6(LVsek;Xd|%~4XiICEaUe+=v$<&8JX%|>R%%#mnlic8oikcvuvE&ifoIn+0+
z=AslKH`1ceDB#$6B54q$?#W~+HKJgvn1y;ouf9{G(O4ZbSKZSzV491<Qs`0ra2c&t
zyW^e^LBG_@PKvoq4V^CWC|eL1ExIyrpg+=*dpifbM_a5dh>WD|!vT9OjKbtpcqT)n
z)Em{o%5PGoux(2Z|JLUdAcZdATDXW*(K+h#F?+uNh693u+sNuecrjos;(wu>=TU6%
zo$OLMobo;%D_3?@>pd$6ahzme<8tL6^Y{g%q7#y@p%5oW{LZEx&qw#bpsw@EEmREL
zK3N5b9h6;}NZ^%{v3^ihb6-7I2O_I$Yy?RmsxpD?rz**x8y{sMK+zi`O%&$sLRU)b
zl<jz$m|x<yBAH{7$-uB)&_7a3YmsaET?Xk%O*M^toKP-GU(IiD$SAA|7FL~&%s^)O
zw3y^r5q3A&`O4f}@m6~$ayI4%{X4SLSx*Zf=F(Mt;mO3SPl|5UNJCIg65cLeVbxU^
z6daiLM8;nJN5cau<%-CMnb<;s3e9Pe#G&L0dBSB<s}vgicUyZ)dZOy`;=!A*Wk-7A
zugWs1RnMAq)zbQoQYrR4boMgQ*b1rfi%MG#E;|n0DE$o^ljvCM_sn3{67E`8h0Zj<
zlfq4OUQ+{d6bmSN*%Yul7BQl^>a231eG<qpNWjUQmBy||IIspDC+x(X;wOsEjI;F=
zzr{3CoEBDIpIho?xemz2$uHHizU%dqAtvf$S8Ni<g>CItQ(`&VT!+Q%e@BgufiXMM
z)(vsCR`q=@Yk#im%bQ0))vllJ|BY`0+NMT3EoDrvsXq)^e`9NuaR15Th%sGZi5acW
zeNMCuHgF4q`3G%!r$8s<rXHc=Yfh-BbV()T_Pej-gU!yZj2Ne2%SzU@$ek-OMx$!}
zg{ML<W4jaUh#G4zopK2dSDA4#?7+J>1D{YJQId10L}P4?9?2FH@3?&@|9+GcV9%Ie
z{g^)q+p1<XfC70v_Jo;EBjguhnpKwFc~s1!B2Ul2;`{w`Cui2a7<EHPIN65_1pi&-
zYm-IYPzss|Xy#Eddw8H&`#zWDTuS3$mDs8h@Ivu`KSn3E5NmC-ud_$)d^_!W4R!zh
zOmQyLEl~NywmdvlK9etFaA^rZaKEPVNa<ah$A>s+Hrh{SYEl-U^uvmr<-p<qSd1i!
zQnFp2SV$^m_4UNn-m=Hl(RPk@v^VM3*nHl%<}r4e8prIkH<IlLv#3Ic5eMQ_W4W@n
z=?xsoN#3$>6vUwRYPPX-uyWd3IFdtKd2Wf|s>DMy#w=VR552?SrmO)rGac4=7L83m
zGyd(DAU#psR5Y5yy(FT_z4V0}#EY^jRaU6Q<kiZo8c)&(3$ehT<B?s@a!ZDo9TR!P
zC@r^aif#V!S@PP441VanZzHSvAszBAr@^&GsT=urO=Kj#0v)R9IP15thlR~rR;@)0
zeyf+sJ~_R~S^uMoyYB%^;&|ciEtM8rYg7hy?S0mSdj$nD4?xm4UNp`up_IMG`X@hg
z%%o}mNpI?u$hn+2TICeqGMaI+YjJ2XXn$<}$kTx^4oz>k9gGI9rxO1gcy+-y+{mye
zP@1sIA~K}za)rg8nW2_p#On%|(PmezcgeE*`@yAHZQl5u1I|r7;onM!>c!LsTbx=r
z6>SIJVr=-sB_==Ks!P(!X_RUSXS_w_F(7~WFlr&IOR2f^4Nq(<ZyM8dZIM%SE6+X@
zZls&|4x!-M-2PD}Vn62W9sqSn5B~q*ZeA7(dWzl?**HLyZno#^<LMJNhlhTz#BUnW
zo)OT!@n?NFX6EyHe*Dy{3QR@cH5rzo@$&1f>HYHXuQhC$>D0$*$;-7{1O?xIos=>D
z{8-l7$Eo?&Cko)zr&9}j`_rfUasJemAoVuyz2Kc0L>E`w#|!y!vWnd-&)z6?x4MCY
z$>nWPY#Sh)GbC_IlF7#Fqv3E`YS3@6Jvcjz5f!#QfIVOGbJ<nur)jR_AL!V(8K~io
zLHStE@^WeCY@(aOs+X!!$w=u|;5*TIN(OM%lvjKunTV%P;SiQAG|zvNdT!&n*pW>>
zbnDZTzcF69sDHoJk5hc+tc3=)MxOjBvsATA1bhx4@3Z}8-E@9?AeuK<FUvZp-86D}
zRM7!$OuTs;f41DXUnzCIe7MBRmZ+oz<9r<Z=jQp=6G(7&`n{slQetVE=IUEFw^+!|
zw_9FT1<ZCX1Z&hMtNkm=%bTxaryOZD{?A97W0%FRf{ruCjorpGZmr7Xi^>rTNKwg#
zIY1<PXt>CpAowFwX(PJQtjL<Ho1~P&WwGs1an7*RmO@X9AM+II-3!LLo)T$NYLM9w
zgL|2$p_{taBiRCtjWc=F^W`+LO;#3MwPsMKlLjlrv;*cM*cQ3-R(hhk*SWhgix5kN
znz9>fNu;DSJv}i^by$<mb%Pl}0wqeqEr410{%y$EYDk{83cpW&GI@c3w~51P727eB
zyDf4z9qZq{E44WVPEid`&SUc8bD09OS;{BQmHM^TiQX$W0(TR!Ic(V!&-YOQMwU@8
z_x`eDBdRCzu+QtGMndRR#0ph?;>#W_3KSTvC{iL(fM1S0m=H7xRBP6MA143|Xvm;o
zp-4MeNX&`>5CKQA5cRP{0YFu#LW%)UAhKd$LSe9y3_i`0W7z0e%Xu_`&1vcL3{2*i
zuKiogxR5svW>cxKBRnAPG}k8U__VbR(MaHpuAU9e>RvzG(v>v|1V>9U#lp^Mz6hg4
zV|_1WjDAcmnyVfRCN%25I7|S^@lM25w2HLqgZyG8>R-{onC#%o2r8Quhex{vz;-oh
zYsyTp&;pY?Wba&Yw#Ay03#@fGr>!`7fi4A(lNXje>Cq1G88yCNK-l36l{%(kbS5+(
zooo$`f~?Aj8IJ$7Y!{7N8zjq1e&-~#M(ameOi3B#BxKvUDREoYbvOZ}mxWjv4&ayB
zbi`>x>^dR|3{O5kn-Vg)(i0jrB5{4VsM%Sq^Z2=UX%kSh(=%I@3p*|??_ku~nsD>|
z17+Z56$7v!1A4^&Yt<M0|FtSbN+>XhC=?T!C|qK$Kat~{AO?yk(hdj-lr&n%H@ecB
z*+UJUc^|%BDxe_C<ktqq5tSqb@6Hj>2A*Hbr1yrJVk#<3Wh6P;(XS2bc)7~6OTG?M
zaMZE?^Qxww0XCigcdDcM@ukz6<S*$tb=I69Jm%y|O$AAxrU(0m24rrvaMN3P)a@3w
zOY}8@XMTJ4vLf@j85lR=_r9!>)sRU*XoM$!b+GXn1(4i+tzurjp)BxRh%36V!5PnV
zg-t7swU)YtAUenbArWurYM%+F_`#dRiZ$zYFy9@j)fhL|i5g8ruUI3lhSPXrcd<sb
zDSJQ7b{yIz(YBSw-|f$#JU3RH>*oZ$pRdp_u6AELE=rT?+~j3Vab_7-ZN3iBQWajC
zbfK<EOKwq}w?Q|xqFJxZU*h*s7;(lQ6&s^2uHw#&NMEh*<{w6<an{Xq1V$LL9@qep
zMnc6xN<{kaDedM!1|tt8+Tlm}&!dw|a46FKrznsENGKAO5U9|*IH;B+l91m-7!j2a
z7;rX8K(46KR~3aqbDAF@fq#xcp!Gw16;`qus3#a->SAd^J+|w)dly>S8bE5T7iB2k
zk#V<VWl%WK{$v}|l(R6u{L4;VJ56r;N%GyYS$~u!|JB)RT?E3+bi@o#Ik?Slp*ld?
z9x)BOW2`z=Mi-66F>dsilQA(G?k+PftlsQ+p`1tOCMqf-aCkAmb3_xHs=E&Xa>l=e
z>!VNNyxx!u)-lJAr?T<iub@IwI#o`_#ybgY8BI$nFOutB#|@+6GwM}y%B%cwlJtgo
z!D*3SyS@c?i*`K1At%f@X+8|4R!Vvgzpqre*mN``JV5)qIdAM0lOLEN{PzBla#y6m
zZ)|xM{t`1tD>;b(SUFC`=$fjR_om>-T073%lMh)_3BkeimXQYJf+2%tqWzp%Bq6g3
z*#RJ660j945(w!T!B{CvBG?}kxdHw$Jd86iI`O+1jUd-(09QMP1kYsW9PAaGv!9Zp
z1}hpCLW7Ei^CfVYYeg(8hnv4H%h9BMd|Zw8^cqh^sv98W;lo|&+JfPaj7qi_O8b~D
zW>I*^m`LiyLF#z&C*w<)$ol>6(yf-OXQN^73vPAAlAfgvZzCU#Z+28#7W$wTR33K$
zdS)GBMGN!}s12y<uDkoQbiA|MR>+|cG{}~8u7*8@q65-)KPX*w$m|oT_GrO<JRdNY
zVKHbSEInYN+uw-_o_fro<iRrN!2fSAuOf{Zu!AYS%%vA&Tj|v&0i_sZw3nkZh$1_z
zg!9}Bw{6u%W#~>?Y<*$^$F|Ob4Y{e6M!ANJQaONYC(DpTT(YnH*tF2h)jBXvQY$Zy
z<)?YG`SGY0h52}0Bw_@u4pTa@_!*UuETo@?JQ(0VVEmH&zt~(M`X3Y-xxhhU7W9Jx
z7LN(_M`1%caKIT6{=wqLA1wZve8K%66gg_Jx|7d<rngUBI9i#*Fk&RKX1(+1jCB6B
zQ^iuXWu3x~A<lYg`(#*rmgQCi$!h%0Irj10Q^m$hs?AJpY06;y-ZJ`Yk^?}eB`3E<
z=KQA0;a!aUs_9uO>rPRtMwgTYPo4VZ7I0_Q$v#5>ZId*?#Q-)bAG#xpJDb@8!Lup=
zK+`V7BC!wt_a}B0h7XbbC)ZEr-6Zd)ajrWr@SLODCoo~C5u>uEW@+Lli)@y$L5l84
z$~>#W_r!J+)tL7@L6!75j{{KSDI+S)l$V5#sADl=uI9Wk!<Na>*R17{Y;b>&W-t~)
zOu<Ue17rRIEs25(_(4$z!O;J>aoQ!#e}7zt)S*~EP7b_OV3Z3458}!N4+0J&@-<9)
zf9YqyV(E$k9Up9sLp)#Chw=30@*1IDr`Ek`^+v1Vy?8a5(*~vO^95>q-T(dH$SkGT
zl@L9JmRf)^>5@DH(76Tpd^;6mnWGO`!WMF%327t|r`kE0n#mIK6ux1e`3|c1plc<m
zva7ze^vS<-zo(C3PnOsm`(T-y!JmT)GE>!sJ}cGfrPxesj~;D6^<_h4{x-->^^G=7
zwTu$h=wns2c&R37k@J7i8n(>#8MU`06YcJAnNUHtF|qytyw<h7zj(eH_t`WGl_Ir&
zdjrB4BO&s-&H^x&g07~&U3$7cDuQ{?I1tHf=ClXmYW>Ne*|NmQ6Q50K3udr{_4Tl{
zMEcKj1OV6RlmduBBFNojv+$zi$6``CB%o`vmMSzdQV!Khg0S*`0LGn_q+o(>L}dHy
zo-6|uidLZjM;g`62a4H@LHwqf-WDWN^>k=~MzpT#87mL1VHsbF%hW6=A*@nYznjU>
z-aEL{V4+2#hZzbha=}#$loJc16P8;X=0oC>U`P1u?W0v2A+Xc6*{;yJZ^9UrlTIiY
zO0;wjyiSZAVcxZo;pInpES3*?vX$vcLEk=wR_Im$E{UU>V5)<zFP`5?aDs}*-4D40
z6~>_ejykxH+PNN%x&VxOJ&L6;5}kMj)kKmQCh=(Ev+iJY)Iy#r%<G3rl1*YB?5x7w
z#wQ;q!SdZItYm%66w3ha6Od(1q{dsqjc^BPlZA@$2)0bFV-CIoZ7N}luTzsOty}A|
zD$EWjcwXUYmZdKwZAK7d@(hk2rb{3W3mpBzn;}<Z{;*PFqf>@dR1*R_<_wRYE0o^K
zqLUtcut6mc7UF`V42`eXpK&?jsJpeIhOT^qrI?2FN4<O2?A!rAM5YCkqswjqVb3<Y
z9f}o?!@qO+cjh>;wAW$yh&TBcTK;bp!FnAaM42~6q)u&rwxg0@y%YikFM3#8`o`hc
z($#e5Lz3qC#;4L6RMFzlXW7xk%~vAV1^;D7<Zl^hgZoosc1w7I7EAr%GFl$(;niKq
zm(&W18)<8FipVfpRrj~M=(4N0TfNLG6J8a4b+gz>q3x+NYsxk=khk*I@_tJ|=Z!jm
zmd3VfFluXvSM8idw?hg^SjW6iH-=gvcOlwcZK`^l=Tp*jYhJr9`lsWt+F*vNV1!t6
zH&vmRiM8!+!noj_AhaY+9(yBs?g=gk0Ru2`taXD=urM=m5t8y5f-V{HqKs+Us`QYp
zzVGOnVvvCb)>WMSx%+B!A@f9YkpC!Pkhrs{qDN!)1&`~1TOa7NZ^tcpnTBePefbXO
zSKDRwD!8aUySMiA{I^Q#P@KdaMc>Dv#_DH^{+DQ=G4(+W+k=;$gzNY97rvV_IFnHV
zPy#hM&NWanzE}}I(@&Q2EOsyUuaM2agcXXG8zEho#6OHHy)f`RsuGq&dM$Q9B!hmH
zs(qU(*V0$zpt^ANzq?k=?`oGdvU55PJ}J`O(up;>MVOXYv`mtZh8=Zpg%F)v@<2PH
zUzToBd02z=7uDw#Pow-RQM@g<`{?o(FfbRZk+~P6Gyzq?tUk8c*878O6vS>~t><~*
zgs}sh;{Mp9K4pqkPS%WSv=iZgZ1KK?R<YiY`zD+kJ*97d*udL31#Ye}yaA6}n#%L2
zVe_ZYSH&pA)x8ECS3M=!`S%yPgMOZ-#eWw^Gcd=em{Fc6$5M`61X(I*%uRT8dQN0R
z`8e)6B=4q1=Qk#?ITPCVLRd{h6Pu#j=EcPe_}3lxpc93e(;c2?sHSd!sS0!<5&mZB
zl;iQo9+Z|_D*-v$&bw8$oXgdzpI0qX9oK%xn=am8MLjDsSE-~N6->H6@uxm>IrhEK
z-6?{4t?r2BSoXI1*GWOx@mk+!pON*c@<&QX0{gqvF0R$-+&SFc4zT@U4IA~z$gydH
zE(?z_;+tDqY)0s?PZ0r#dIAXWs;@+HC%PkK#Oa@<Z_#pEG-{gG+!pcf<rrSuaa7hW
zAFS+iU$Co7YU~?z<r#!%5xj*<uOWGg=sxbp&TvTRV}6-Zk_fzP4heb*(5h=zs>C|B
z6LMqP462(le&KQEM*_jILeB-`3!%lt98MS_VcTymZa!4rNI?LQccdK4An4ynztj4N
zDn@H(4kZ{Co@m6Q@6bCN<@nVQo~7;KXlLjZBCB|$JetLgI#wNXCU0k`Aqb|88dQbT
zx($#a7kQqV_jzQ0KM6vWP_KILCAH!<J30vrT4#zP(-D-l^1=PqiDM3MdG2LE$1Yi%
zI0Si|=o5!6dq@D3&!gca%-;8({kioO`Bt|c8n3(ElCOgZED90lcd@x0th<%tUUiW<
zC!SZely$K?K(lWH%83pM&red_o1@ihdRfQLi$dONMHl}C#>@edZ%fSZ2X)|OPMFty
zw^Ar=gF)v-Ao4Zh{{4Vq%qjfpj~%`o0A;U$QY0<3@`oWuLM<hLmi)*+9MlV(c=%ay
zEV(}S@1m;w74KS^g#Hp5A&g$ZH+uN5&{SDrG#$nDg7UREEgf|uL?j|*LybhDXY4<_
zZBe=Ma!^?}NBy_ofqrRwJ}-sP!WT#?s2x(&F#oy(DZusp`tK3b(XgB(uVUA|&<4aO
z3W=}xb5Jb-w0q|_xgUh{j!jy_M(LCAtlU(q8$-30oSe(9&`@8*w;o7+;=y5y>dxX5
zjl2B>wL+S_{CTZQjUTouKn<-B`Ai1|8SG0qZ29I%%~|My_CpqL%&<MohtM0a2jeGX
zk1cm#tz{xz?4GL9xhdTyS~$$^i;6588?ng7xuuYRkLF~>^Mv^KF<Dq^&1|nCC7#-5
z_3n<_IB;X$*<<3Ap#}I?*Bp@dWJEzM*lnG_iC~@*QTLQ$>=5(DJjFQ9H9g4UX4$mc
zi>^&7h7iGG7afF=?Y><H!awrS5KRL#Y804|6@`m|#ls?AUr0q-@P@lyIU8eLpe<C`
z8|90D=05@X?viUpYmvNVqe7OjiUeQsCI0Rw<Ai2LptoNs0+SY9ABU)l4nkHeA7)PF
zPfCZcXoe<pPL#$RW*r^;I&en~19L80k=bTpF>ai9DGcxj;BgG}hm$90kqROyU6Xsb
zj=N^J*t$bGUwouZ!9<M#Nt<3!zY|Boxadp)&DzkXl#%Jy;Pe3u?2{*6!Wev*gah|J
zDCIhUyLi5_ecW;QQFx4qzU+*sjM)+>-iT|gyLdmxen(awkITEd`c2{kW{H(>;n`4X
z81s{ik`>K<b={YY3T)kcF7jP3d2A^dI8QNmf<P>=k<lTB<T$6<vLY|7RpAd45}R27
zJA0^ZM1xdFg}7aZi*4jUczG0LPN}8wKiY5(0^JPcRu^7^<Wx}a-KSSuEN*heNh1>q
z-sX3eY`N3{6%_hMJo{)|AXlDsB%>TwF_Y5I={g`?=?>bt7@){W<;2(Kpy~6Tu(}kR
zmh&ldOcS+U+w$k8o<z=1_(jEBd>GS|hGjAoVD2Bq2DIe4ANJ8SlQ1}-1-W~==45}u
z$bvt-rjkT(Zi>zj3<9MF6-+<pcTJrmhLE@9moGnm#sy^2t-RBTWtkYp`CxfXIG^{M
z$KHHL$NhUpO*Xn*GB<r?<`y6m%y;L7mWu#~Pm4HJ>9^;%$J^JxHn!HO)==jn2Q6>c
z*7N-W6oel=;S?@@ZIr(NGdzaZ$n$v1nb~xa>=<+DJ(O0s6L)Hl+#tt2JSqZ!`}ZUM
zwCwd}&erzld3F8uuX%mhA?YD01;D%a^CzPB8?t~epJQ=n#uw^eUxa?}du@g9)b6hL
zeMQPGW?E3C%~Mh5>O4^Rt-9(6t3HVh_JO_{T@mkqmxJiOkmF-OZhQfY=>+vh2rtC=
z9kw)@_E;`CppgR+-h{1De1`5YFIT_E+9Hm@Qg$cYM)yfwxz{jmpv)l>*0^XgM`0GL
zt3J;pKNqtXFD<SfJw=u#Y+7m^7h^sfd&6M#ce<7_r7>E<l3Ijbe5{cBuA`QF7DWQS
zw_y|O^RNVIC!z?zM?FuL#-<G0e8GLB8zT?s?v7j=&+;FSc%MIwxxVtOUoJdEIQR%l
z(Z#u@pwnRO_tLmP{1J`sK#}moh#QrD>#Ns;DMHj+tIpYsO#WVE%cU{;p|(^F-u=+>
z;iI5uQJT}8OOXNncWEZG7BUIHRUv}oHQP*3<khkz2pT?sGyg=TC9#>EguZ`eEFRlJ
zR2b@Y?5kG$3t9J6B>83GB@o-952|AUj3Y;dgQm1dcwHMQ-H{J>2%!-OdbGRVpfY*5
z+Tm{I8tQ8Sz&NLKVrDII@{r0YuxfkS8@yA_aQ#%YOyG7q#{aXXOc7#--~iZ$6rlX8
zYXL~qH~?U!kK~&RY@WtZWBvVjXHf1x^lt0xz17syoDD62yTcaCFGyggukp5cXF$`X
zC39`nXdq=z7k^UuaWClk53*qJT)Id53%!UjF6L*WbSX~SZ<bj70<fDJoA<>##&~_|
zv%>9Y&+e=DuJqjs^Ggay>MA9j#TfB36?PfwECBP3v$*py_M+I5Nk~#&XdY5XX|APC
zjtLHO_IkkEc+F28G_w7@baCDlYHZHwz@x4faW!sGm-?Kg{KPz+S}tI!F^m32)AViE
zZk%zAEh*5xRiUSyEx?I6F`EKkbijEeK2+b;(c+;b2ALa{|DI{OOqEJ!SJiA%Od*y$
z7x1hYlrK$tr;1&wX?s*l^hbqCBdi=E{GQ!9h2&iMG}CERERFQcw5XC0;ccKybJk@;
zPkXA|h5CKa*}9~)y^@EyZHBjS5qA2(`O)bVROksJ5oe+!T<ABsXSPqHFM_)I*$h9t
zg2gGiT^r$8DUH^ftJOm_ZWX_I6vY_mCjjz29$f~}7g(a~tQzn2Fq#I1k*jBdw_1}u
z>Qi83@$Fs3NBe_lb@<uWAp()B_XY2`JUC*bFy)<MX^cEM$xe*G{8%z|Z?W@wJavvG
zucXr;@H+c^rJfX}OPnzslKO=^g^Q-7wYG%8#TC<e-%4T9mrco8ny`NA4Cg*`1&}%?
zkv}05wVJ;gMWI6!t?l>+2Cpcqx^eA*=51Q6w4sz&He94Z8NUtRMQ8n{QY6b+yg;Vh
z!^qTSAG^incpR?_hHx=30&g}jaTLB&=ECr>B<oKdJXY!GvGK$7%ErD2LecfC{{04x
z*E`#ob$!-Ri7j5Fwc5ABHV!=cHh}YQ=X;RetJ_1R+$Y#Z|D#!U<+<{r`@Cm?O;|jn
z*k$W0ONwkD=3KO#Zua@h5TTtgo|H)OXe)1erlVw1;pj#KyjNE;7y7>XJ}G*ND)syt
z`1DCUpIO^{0!@*wr_Y|wG`5OHwzN$aZ*i5Vj40U_#mg{|f&WY2<>sCaBBw{hsmORg
z_yBHp0!ke@8QlGtY|j^)57>wvlC92}JP^Rq7${WdgPcnc+rQOZJ!4N}J(S5sFU0|w
zoC)AFVI72@hy!I!v5}PR3&K85(V!0HabO_7Cq9M8OkJZU%@m_Qhu<D$s7#C0BFjn7
zs>g%-O0!r!Ud_b1hTU1d!+Gy12Wj}M{rnQ{qA$HC{x_0p_!Hx=uICp#q<Q!+R+Hlk
z|CcM)lh`%;e{_~>4}^Uhk{xqn6xhpm($fHc%+uB)acW%(D|*f8eNoWqY+NoK*pxA;
zjC?0zZ9`R^A@)x5*p{7<q;2w|Tt;IB&Jpz)AufW4q{x~k!!hJv$MeTk7aGa5T*3rA
zj+WJ<rO3JJ5?y=ivvN_G673{3*B*4`>J<}>!8D2W6YgmrASX3HG7;y$3srs;Z<+yP
zRSNd84&vhL$G&>em@Q|&F}ZsSWy7(x_$hhKbOkPy3@J((sBh+zwBn&VG8g>v<pdC?
zozfnf1DR}Bt$7|h4JFue58}IfSL%VDG3FNbuFH1s7W^j&TrqNM%o)~Dw+Ejj!&#l~
z{J`fFE4WONP`qrnzYmEgYh9f~$cq7&zsIf#Ln62)ThAw*zmNY)P?Otv6sG}I(}gs%
zhQU%G)uB6tjz6A$LIy?rjpIaG7Noa36ln@M^~8f8o3e0JKrn(QWtGBAjc=inad&S(
z`%J8>H)Ji2dvB-aq?Q56wmN1~VU;o~qPrNcdPKW`R$Qxk2IhYyii9Q{NYw)vr;0{3
zmZ24JyuwL9|6`sb5jui+`7BWo5tH^ya`<Sh8nNvKx{cL%qXrPd?(fR!(zytrh&zHC
z9SElq8;1u*KdeV;wRB0lQ}#U-C+*8ohV}-0(%2IabQRM+DM=eEoCZtwL>7Qx7fW^z
z%qDSY4)sW`inAEE{MjVdHN^v*RqV`V3#B4&fp#|uKVB<ch>Zz9xVnmz!ckbfewtRI
z!IAGLtK+sOU{1{awr;(pxp6+lSb76HR+OHEv#S>09q_D675!8H^tC@DB0Spxm%TqD
z94g!RK+S}`H+S7PS*-K<M6ub84Qq(xa}jVhjo6ZzAsH1ewa8l0ecS}d(HP<ugFIJ~
zc$dz^Tw}Gab!1y10T?R>Wixrw4-{D|O=^-Z#`>W;=wAE?diQR+Gn5#Jgf+d6=WJ~@
zl}*1tdcp_5{MAJs&cyk*mREa)G{}auKJjY_rpg3U5L}xf!^^&Z=Nj$adl-e+JZ;63
zIsQ4~((iW}+=cqN(CQQ5WQmAz0kdiPV$D~3USuMdkr=mSu`TD_-FCX_8f)&mFn$KN
zgt|$j4KbgU=$h9byZdi9-)6{SVX~|YBxmD9G0Df*wy|cwWntqW91%`Nd-pQ{O9bwM
z`pUay1QM-{nn1r7C-ed4YGDe$Bq8<#T(`s60gk=US=LP^(or1XV1Ye{=+H7=a?d)o
z;;*F{Fs}{S4+i!N)#3JXNl7^P$%OLVZJ*umt4Vbe2m4#M*E|n)&})1naW7FF2=i0a
z5garagPPvsqaLmV$ZMjHq=lTz2>7qfB~vqfG+y!9c-`yFmMwvLA?btB8fbh$?!MU|
zEdkNqJIn4Bu>Kq1vhDe-BHA8>aWKEyynFmN7v}Pn7#&1Z!Hliehh3(SFn#W1_+WDi
z6X8Ea>%UK2X8T+hiDobcTSX%(XJt)YyBj6~8hgYCs=GymE-HT>2P!2>RWp@*_iv^D
zz!#y1x<%;5&{28C8og<0oMkyI1$vekU(PpS5y#PO$fS`1Ms-)NTF&K=b;B#N+i3j}
zEksv?VYW6ffVe6fZy82RPk*~)q%<5IBy;ZneQk8+SEwxUD|Wm=Ab$k5+{9~;^UAQf
zXHdYYc#=oRaEn9H_z1zzRYF*UUJM#zep;u+CF<N1AZAgNs@x_@clP?8CIom|-2RsA
zV-N%k5Zb*PvGpXl?>s=1RT~{iU(y4UC)!i$Qk$xALBTK%_xn%!9!AB}??B2dPucgK
zZ#y(r-)tc>%-gs}U6$o}k1_kt)ur7pcLz78^oq=jGA^MQFB1JFU9V<uxc!#;26vdd
zOi6JJBu7hjAwW;5n-ir>8Nehj1jyxlf4v=Zve5EbS>VG)s<4>^BC`FK#ElFWnBN}4
zfQLUiD0yMycT|E%zkU2*u(F{V7jZT|!@aa{46*i2xOZRv;LV;LF1*%YRL=^4QrEJh
zugFBU$!%E**i)E!-(|FK2<6v8&U`9Hy_JrALUI&tvn)g0?)HI1{q=F20}!BGebqyO
z=+~wXPM9C!^B(ILaTOz_*Rl}n5+~g1?pH+=FdDUgP6o|@ktoOX-O`IzSC_OV2H75~
zczv)0U{nRg_tOD3bApR<kvt2!u5bgY49(xQ41sA-e_C=D)DQAfJH6Ag2yvl$mJx2#
zJm0t^uOrGv$Tg%}8}gfS0yxdlH?gL=IC-fr(huws|Bhd=K)W_xc*XUbqBP{TI#w;r
zUopfQAtzpn@tYuPFHe#bYetFJ0TX&`WldP-B@$@3>^vM~#`fB$PFJ^)dSDC*^}(8s
zu_<3G><1D#r(A+$>a7hrSuB?%1|f$W{xVYpwE^+<T#<F4i%j`P2B2=N+uFljQZLyC
zo;&aH77%Fx0eY!**mZEVBx!YcQ)f-5tc*H)$w%RvqS9sJ-D|qz89U=fo8z*Jm*cup
zzlTOr<Z59Ojz_3Mo*?I^S6G`sGM8%`$e}jlyjNwD^B_HjNw}9t%w$3J{i)|6HtXW@
zWVvsOZaOBJ5I-S~20YKE0BdlFA_FNTzhM5$Cj9C;C$rJw>iUDid97F<m~nsE=Qx&t
zZv9W)7KfI}9Hh}D?@Ig}w%X*_q`^9^?7-tF{nj?2j&ZU2P$S*B<#nasV&ixMXbh)e
z>PY*r;jhV;ig}h<r9JzW1p_1;G1O_94sWvsX8<1%bSzUUEdZOFVJ%9OBAp0Zg@XLq
zxdu79O~ZU!gN-y9E#?f#^aP3n!r<f0?B)FRaT1Us53~%*pZRsWp78=2$Z1jI_eNsj
z_4au@&uHI`D`88`U~@{c&%{TP5OkzCCxsMLZc^C5(tduAmaLAUO;w}iGX(Vb)AMXI
z0pj%c8|Yk=3Sh)V+`4hbZ+F+$mbKpDsWTF{{KNLn9#wa{B47AdL!RSOy+&ZPwD_Lk
zaLZ&!^la>#2MdL-6&y|u#d2o$Ru#Ji15HTqQfGA&CrHlirBp_!WZe_iS9sF!eR|JP
zT?|zEtmI4g5_x-WP!WgqORxTn&`+)HoZ<J7EC-4LJ79{85mNb|W9_HVSWsk`R8OJ8
zsPOKuJR9*RFbehPq(n8)WO=nkO{Gc6ETjcjj}6m-tUsEO*-WX<6vt*EzHI9|ZMOAr
zEv%O!1TYNytqsT{uXdM6|15u(wvZ6Ip5l`xk38|--he9CDZi|dCRA!th4WskXtB+W
zM!^wC0RWmWuxE>IwWG!fYR7tYF6NsT$n*2OHhQiPbFV8!M%;|&s|IDw?}}aL%B%o-
zw>z$1b=4mggwEd(oFidOmI<Ai4+&Oj;=`EYb{>{Ynp32ctnKmPkKt_Q6fdgGZbsll
zjT+q~qlY#h#_K&;bXpG+bP+kzYY*3$9WO2j4S+$*O=I)T8Aa%n<MznS<m`Z#OTKke
zQDPzqmk=%(i8+v`3$dquC0hDOr>+sX-BK4IT_Q))+_cB2U91jxj=LCNFG{4!QS)iC
zY~lRFI@fcX`q@%HIl&#?u$)4tzv{P=-dPKi>d+TN&=p9Vn)g;r#jUWL-q-e4VN9Vk
zv4A?9dR1i~okI8PeAw2r6g~Uv&UHeeDMfiV?Z%_c8GJbGY4J_kvFa1;J@mI?ng(@4
zGjz8%8WU#ad9L)Mc=#hdnGbcjKiTU~_G(WzX{l>Raa3*oL9+6iYWM#7l@vy?qsGD2
z!(P_K>DDdS{n}ef>N+of2~DkmmicQ95CMi@$?!Wk`e0z1x0w7Vi_(4&L4p7^jQl`|
zDpb*COzu!!Q!vLuA*#pMrIDym(NJAmrEZ>bNuhsP*l8a;BS8aF5<00qzvdBwsnw6L
z1W_@NTNV$%&FkA!;yHmM&FInIpGt{m<y)ZU1{1X3o>|B*qpty$f1^PBiQbg<H38BX
z^+rzpHtAk`hn&NdjjLQLOQN(f&VwSky7EyAv#*yFinAruc&^kv6}xLzzP1)4Yy`_5
zmRH^jPx)F`lCUQDzl_Y{S7vSTqW4_=t<+YJs7KW@=W*6C|7hsd?5Smtw_Z!5i$8Ua
z0WLFH&oY)NhO367e-y?})T4Y6tpJ>o=vwaiS7LtTR-BT?e^WEHax4nKi`iq97;Tno
zZ*lyb%`V-qrL#Ak=0?NwJOvY|&BS{QjB!x3*L}ls5aiP{nF>PF5zc&9A^9XuWs4Gi
zjo>1r2_blW5;)jh_AlR#%Nab@JkKw;bjV1u*AR$*y7bt9+mbVF8W|(bR{+rPM5OYA
z8o!1f%0ZN2P-0<5yn}rR4mG%bl)y%md`<IA{X#L*-O~&>qveufdj+MT<INmWqY`am
zHC%WEU5l7rjQ|)@H_Y<z9BOm>Kk9&ae#BSiueXNunbDhXTKty@VEFf01{WvDQhrfB
zs#6+g?zmD=T@$^`t-u-g3xJ3X^=2wi&@!<_rz)yBmRH8TLSC-rC4Sv@s4FV)5DM^i
zW^e`zH<5Y!@vAaq?S1dh6U;5p`E=^(i6Wt2DiQP=?HslXQK>N><euKFeYeQ9AmfkB
zAv$F~B{9c5af#I7_$#-2{UJ_U|4tu;+dgT3nprY}u}Uv0YH^yv(E&srz>6bOgk*Dh
z<*@zfg<U9J{FKK^WbnJFpG9y^`@yyvO@6)US^v&ufLx&hzoCY#JOkqOn>J(W%Xy-=
zCldapovn5SJA|vO@Sf~;%eK5>{pud$-=~)hIpQVfrnMC~QP;D4C%L0s<}z+i2T7eO
z@!WCA1MiXY#+1Rg1Ow1=<^Ofq&3KXSo1ph^Q`&>&S=9bqp6;!@`5frI4z#mRVN-X>
zc9Qd<bFJE~3~P-+QU6kWI0Psmn#OYe?U!4w9_w<w`C)su{PwD=^KsUv;cNSn=NeO#
z^7iIcx#Rq-yjNv<2e{tn19k~Sh2AxN#I|w<RkjWGwaSo9r7nQqOg4qS|Hv`;wBkL}
z;F>?N<#vuJ#{3Rg4C`ehC3_T=mubH`ly=#j0(1+2QbUhc=y*_7@N&B0>#oJPkSgDr
zWc#!*p*J$E+LSqh*g%*$64PGZ`Dizk^YH0d{c#^05`0@w2G_Fr7ss@`?C*MqtXcG_
z)0O>FrCQ}KTQWfIHOD0CO3FOSJ!U>iTWWsipUlz)OHPAh_2c=3*{0_Z6e*_lS0*m>
zZ9I%Bp*YDD{w0=`>%ME>pA@47BV#u-EF&gB<K}6T_GZ@1259X%Z(m(_;b3S^9cjp(
z-ndb&4#&U{Mx8-vZ5^?paOEYTt%zrpI_H50?=4~QI1}IvGe|~Yj&GOcMA6P%ixV5%
z-W0!`(pH{|m#WE7V>D;ZnOn~-a($vLjuph>>Ngv=u8pc>&Q9*>#<V?UtJQ)M)7CZT
z=3#AX%2_<4HahKyg%j#Gwfaf6Hz6Yvcw2hlJ$s{Xelf2uhVQtmJ%wjP`oHK3>}vbX
zq;34{SbPBFM{BoQ^e!6P8jGI9yVHAKqM<78->3C@Ip|$qpe7gLdp727X+evjME5Y;
z{wTw093K2-ddksqZmGvdgyGG`aSvpmzRN#4nY8OwK5FNyUFUo^AV$Tl-VL^HK(S8&
zUFXFK)40^5#m1xZtY6=j7@dO67wqp`qC<6nDO*B(Sb;N49_GGBuN#xO@2vUBuX<->
zrw{R_8v&$iJg@(T%qemCBB&{t)mYRieWc{MAQamzCS>H4z@XjH$SpDC<W?|Xu{j@M
z|DF5OC=a-T0X5}YlZXm%sXHED*^0dRtg(Qo8%2=jDf;J+s8Y+j_{7xHrrgBX)5MV5
z?&`~P7KjuS8B%G)U{F_SNWLh^U8Z-aZ2Gou&+U3k%I^EP0wIiDh!&doR`<6n>H}y=
zPr$rT6elUzy!1x*ckAcfSlhw__6cYvtr(%mdvf#3lNhuN=|CWWfA2B^_%=PNAm$@?
znsmgEv%sML^^dI78`PUc{C?qqemuh5&E@j7K+g@6rI1dX?RHPk$Ga(??Mytd-21hF
z{s&^?cw!PKHJk`+i@=x1Pk}9?lqlL{+N0OQ^SW@cX{L!)m8Es(gn{p23t%QjVBYZX
z;!<FLhR~PG9AOi1Y0%OW=be4N3Ny#LjN_PR9{6V_&S5z5+j5T8Y0Cf4-~M&X$u0_i
zHvd9;B3sh8g1PgQVLqV(aLY8n0BLCnYwVv08vH?`Rj{;rl5tPCu1=}@G;NGE@k1=g
zXXr^bw&)Ry?@=7gXl#0^R&psGEANGh<%p~Y5JIB}+AI&irgv0nWIH-+)qSqU4|_-$
z_~FkN)%=jN#!47;j}m{*Q!q7IOy)`gog4Z6Gu10m{U#J!EVyC7fq}7%;mFIRx$S9g
zNc(kR9&u5s!=9znymt;A=cI--MUr=O6%2(^oP!CKZv7&<FPf?Mi4Z$B*x1R(R<|EI
z|CC?Z2N^ivCFjbYfCm3AKs@95&PQ}k7=pN|f2-KUmwWKKfNrAm#(hpn-sQaBKTe}W
z3Dbzo&&a8y0M={lSnikF#&^(2AlrGdNT69}5L7Fbjq~7$3}trQqH_D5t!^vQx$etG
z37!-3^N7u;y4pfzcK?2=6+)7t<*-Vp!Og8Zo^KAosc?`fm0W}axdY~uOQz$mQGa6(
z7Q}rmT@a?T%6+Xnt!KKq<8t!;IpynI)H%PwjC|WQxwjBy8bs0y2&;ip6d#CgKfeLV
zpsZxP^%w^$MJuS@P)cU7nHZR!G;5NPA|XUYgxQ6cNq;SmP3AgMNUx3Tsu<#JG2u3Z
z{X`<5*PtI)2G&5EMto~$BEup!#;`wz!=A_)kH&l0c1zUeqS1i)6`Vu!mKELd874^c
z4ZCAHBeT*jgEK}}syu;n^M%8`pyaWay^}c;I}PKm_wj6l$%tdBykyEfFfGDqx1&58
z<eawDc1sY$J;ZkFV9WY1Cy()N!Px=E?35ZHpCvKVHD^bTvm1UZ*UXT?6VLyo#rexd
zdaHDmbig|gW0#mz=!O;766pY@r9q}#3_XiGj@|qME@u?Vx<w?7o>6<;(n-O~kG0up
zwjlF#sJRzARlR!I0fMgK7vBk%qvl|#q3xod%C4*={gBGldZs^|Y11TjhwKPu{{|c&
zXMAJUp9umTLU<GZ^5vf{j;m8K?mC3|LTh|?2+J4~EQiuiQes{2Sj*flx~YxpSb-dN
zW<!KFQ{5pA1$;+JVwqMaJ#`@Zm`x#D3{3;f*e@RHd05*(<~hdjL9xWRbgc}6?kFZz
zs{MeF9W2wnHFr{A?E}In%{xtLJvdyzSRo`Wg#9`3aFa)SmWlv9i;MC4PGl;^(6WO#
zk@J1N*Gh3DU-HDju9*PvR8A7ydg3hhNQ~{+!4y=*ZN*3~0>k$8dgAE4$-GpWods^<
z?tJ<S&X|MYRH1Z(`-SsR=002_>dqQVNj;9Issr|zId_v0wz=+6)Q=@bHLL)wlh|De
zLimTWoXzi<z?@Zo8Oyl+K!<`q0JFQ(Dt}InMFP#uv;_?MSu3!HAD9%2*7nP{avq`l
zr1(IdD}QrWcChYoV}*yGYqN!v7uDPp$kfN2w(C&B>z!L}dPG%6nQTmEF%5oFb`2q5
zbkDxuXsX+MaLwH?Pr7Uy$v6W>DlLAQ#ZIhJHFGp2kZ7|;;H!*0eEsHdRk}^usgD{v
zy_Ig%0?;y6R+^eAyKI$%M_(xq%2y?gklN(ql1G4_){r=HM45V{ds+x9;=h~!wx7)B
zwYoaM#-<A7Q^<leW(|Xr6K27x8-YW!;5eNo?AcbA_<Ex+6OktC(Ub*f-l^T|V9A|J
zZWGsbcl{<9cqPV)?OCkZb#-x#m@ONrC0q*|5~t$b+M^19WHwvEdd7IlZnp{$9Qn|z
zZ`lu5j7CA8ffvh+ljIx#n>8k{JE4s%@Sk!!pZ#EtW_TfIb9UzDafuEJWukc{eF%jY
z!W5rH6MJEu@^&Q4XuAP8Gl8*|94_b%JQ$Wz6psGLia{WrI;uTCOb+rLVr&M`zTa$S
z&DWqugzA<e-!i*S(JSFydk?@l>kW_=M&$70)o8Wmy9CDP&qR47Cn+r9_kCn1G1{A9
zuF$)In@k^g!)BhtsU-soGQSm!DT;B|XTYZU54-&JENxNuH{1Z6<+rFalMbvdtr00F
zOx{=TG!WXO$Cnqr$aeD4@9j)@C**gR8psnXF%~HTv^9`XxifX_l!YN1K<&w$Kg;ce
zB;)cI1r~{xwU8k%!v>^5rzMXNI!SXwUHXdBL1w`&&qWEyC(bvDB1yW*ks6@${n5tj
zI3gFF2|4)ZJaz%VlWben%aE>@^Yf(vD|x}r?Gy~v$VZJ7rrL*_`yQ4-Qpe#;84{^l
z@i42})Aq(>8>2}TTN>J<XpEgT3L=SLf7<D&u_Y?Zs7gjs2uSc1+JB$b$G65!!xWuz
zh&uY<2XF;#aQtK3e_l!bK9Ob*evASc)Js?{pPHpCUaAI2><{4$7rsKDgTzkwadAmN
z4}gOBuzP08oUSWO2PYp<9nQ94O-3>|wYlr-HMrkYs=&`>p`3)3!OU<jA;3^sy9_>l
zR{8Q@-*NPnaI^;>$~ZyIq`6CpQao>8pU8BQNe>;G3-4<|LmnR{XwX+nb&v(?fjbu^
z8oJrhn92g^1dgYRWzEm=1{?nVbsXiEtyz9`{=nGoR}YD;*sBmWs?5lKZvW6LxZNzY
z`NRP%-m{LLp-$Gl0WG#nGXW@Q_UonTC<O)&=vX-B+I^{T(Ucms_Z6lm;)p5DOVOm#
zJK|F4!OJXA4OVdo?U!^qWVv-W8dTKL3uk(CGJgXwOG-8D4kSKpbhvS!87=C=bv$~F
zsoV-?1+};+vJJivsK!qp=W!<3vo>1thV;}=U?hVd;K;=KmP$6<glOI<t)77*D)9Q7
zbiUzrEork23J_Z33m{1=TF>sSE^15hDf1eU&UdR>sC}l)cFC$Y-?oR<xcU$7&qshY
z<UIfxiaJl)eudM!Fmf66)l3H*<9dT;xY>Bo?hrl*>%MR}E83F>^V^k(D=_C+Q%&m>
z%};F2^ckCZB9ms<$Dm&D>j|#wF+!j^fwFwx_+Nfic`i%3wETGqG|kod9xXK<ZUekF
z#j5>nJ6EnMe~p>k`e*&L6l8L&4VOYJi&_AN3_vG-9$e;qv}Oys4>KkZc)iai?He&7
zpayu$^vnMu(k+b<b(pXE*O75?)xD9nD&$%)1AaIrUusj}GaxLa8w~+9+=SkfZ9C`f
z0oM?y(Y<R`-w@SgggYuRJMs}+PcyR-N$)Z}l<Fm5%aI_GZ`;JN1cSpg^5sIN-=u(n
zNz*Y!QtgGBVgMG(JKyiF6S=5$;&Yk6$?(LquV>}d{ueS^rDZ2AJ=%rHp}+gCdh^oL
z(?p3L0(Wof=~Y)X+?JDxRIpzUq=))B^xRrvryknM<7#QL?H6TK3Y@O(bF6Rj9qlXX
zd63kbf{V<o2J1-I;~bHSkoQFiuTkt!VP*UG4^=?1Tg!KEw0KJFGX8aY%j6Eq+tNGB
z?bS_&(IA6PwJ2p-MN<`z$rkSh@sza<{~JcL{I?azIu9sSN)InN0>GeLw>;j&tNbYW
zphxa7(c~ze+l0a6(~nFrz{G96K1oDJof8*Kkd9DLI@?!ZbbeZ4nNUHX3)c&Vga`fM
zbl9E}9Fgop4aLsW4MrN;e&KYOu*ATiaDkwjR##oxcn9VN5TxMd%R$oz<q{~9l@qAO
zXoIKwhF^_4S9`o;0J!mKHD}X>VL<#7zr><)Cpr0(H{Va843m|K^Zw!lOl)uY{`x$&
z#OT!_Pt+&@eB34yY8C{Ay_t|m+A09P&L7t24a~}Qg_QgQ3ZJLoVdez!{~Y<CPzaR6
zfN&Rq=Us|PLxN}_{qv7SNHp1<AEJq@6)gUIy?>tnb6g|f2H*zSQ1Z=|8W*DX`GY3a
z;nC>qa5^TjOS>@%*-5Vo!)mH|LC@dV*vG>jldQ3z@IL@uK%&3q$KTI`R>J!9Sst_#
zLE9}JTtYtHac%Bj7wwc-6}`ubbHd3@Rt&HrZ}n1d=Vvdo=M-bPRYPh<bicyDsYF!z
z1LL`mB2d!H!Ya2Zx6wv_xUguc$opR2SRBelVZaNCQ;P<Bsf%S{v2>`P_Hw^7c@M#0
zon_CNtnfk2qaWZ*Dmjs?nJMo&6T03wx)`!ULk@WjD9C(CH`-XCjs7f85I}9*<jF2;
zc0VuP3aLJ=A0Gu=oBDZxqbDOmvY+K8MImk4;)Q$q|Hs>15MF(MVBB{R%<iHVVfLII
zGL;F7qPCnphb73q-UeY!=p#|&tz^9SIVAmRH(&|U37uR{`oI2Np8g3Be*uCf#nXSw
z&_D8UB@)D>f1mya2;cw=>0pG)ODX{~0CXU|It&2-T2xH&bLszCIc+`{*ha~$2RKtD
zlc1Ob$&@p&4iWx;ucEE$aY;k?QIAqjj!q7Udf{|z>bX?`v-}DyUxo9t1_k9u2^~^n
zu;4?Y*T+M4Rg<L(R^OwVBe9$o(Slgf3P=Ra_U#4(q}C{fi}6uv%?UAX%>QfwDCd7o
zjzWl%!Lk5vmD`d%GK1eWnSvG?{Cf1~ZJ8jpdSf0e(=z{mBGa{+14M^N#X!D&$>MIy
z<Rv75t|xyWlk<?*!-Hj7=HHabcbOdC&SrhxmdQdSm0eHaKqhMe?CJ;0w9LOL(`0)V
zAGvLrj0Y&#^%M_eGT_ku(d)F#zblg}LA}UCYh_ymSme?cIP1Df2Xg6(${#D&GXJJr
zsoUQeJ~m~4Qd(GmsYLT~AX83`?$K+t%)cuWTPwrXDs9UJfZE2Es?hs3<+6~~l6>@9
zE%Sf0R`x@gfSz!s>*{tO6Qq=6{bOZX=fB6VLbJg=oHzG%tXBf)H@K0`#=T%+R@zY^
z1M7qs%~c3tuN7gr*08b&Q_9HtZh6zm4m`@3{7eyl-Ttl6i1<z}I1B?WH}-zPdhLCW
zBQ5gg8Cbj#-Ey}UEqF)2vi?bvAr}`#lLjrsVS%AW)J>YRfM;b}DMNa>RV-_yFGcQI
zm(nC6KDAZ8O`ViCEIlnWij+;`bcae+D$58U5@@a?J%6{%*C<{FYM!X@=XDj~$y?)$
z_~kNxd7R@NHC|_B1(?0^c?wn8*SD{EdlxOJP#k2Mb>eiR%q<7eKeiNRJ#?uzEY<N<
z8dhhmy~qlTf~uy)Acy+kI->ey@SN@goU?l89-C2Q4^7_cOBdAKZSXL(o3@gZjJ3q)
zqw_kI3wZ*a8-<!KZ*u=`5_~NhG>@dFP?KhV)KNUN5<Y_BLTL&hb*d7bZ=U_ryNf3X
zyD50H)xL)SEWB-4$2eS@L{X5fvuBoS<GeQY&QUZiBN|1|wBZKojqihP2NR)u@3u*m
zwtcH}mbiFDN;?}E7D_R-Py%0o3Nx>nxtx`<wU6_xOOn<0TtTd&f%%)ha3h)ozYQaQ
zH{UNkfxPGO*3GM2-93?|Jfd=wuzv28N|;ad%!ISD6s0vX*8C%K!i5EMR&JFwur3zV
zt<uB{0-)gY@aZ$?Db1nHC%j2y0i{Q~EZY0IrAi9a*+r`aWFXX+n;klvA&$(WvgA=t
z4%MF*M-QW8m$~XY21i}Gx#oA3xwa>N!&tW3gX78Fu3t0UV1!PF_dxRgSiQ)>+3K+@
zNGkH%%9wFSsOxI%%PJf`_0>QrvXcn;yKJ$t;D#$=h|s!q^fX}2eMNL3r+e334B`h2
z@}b8&Cm5I;Va}nWYbPAQ&yhfuz?MFnaTOP4)3;JR*7S2bNhvjRA3p~Lt<;}?fP!@t
zrNjCyFKuoXf&mbfPp95$PE{4;TS$ZX7S5w_@3xQmntzxCU|r9vG`j~*`W5%+@-QtK
zRuVAx=sG-bk1fb#061I8_PnYikm*mZW4j|1SCP}b+pg=_`&<V+z=qzIeq);R1qN$h
z0nniznJ83BGqsXLLe{NGXQd;5PjBURNlIR#LO)^|zD^S~Ov128Lv)q>Beu1cWs7d2
z`T*b7+6x2NX;!aib%|l7<46x>3E_@<&4(ErLWmX!97|7CZ(9j0f4*+JZ>MXHQeUOL
zlOeQAQ<J@<KzgU4d3~Ro+W-(iE=Hs=cvboof9E=oBhbnpL!g`Ps0?6#ywK~|2^95W
z3W<8hBB}QNZspTkbD5}ppt!HtE}o+Dy@Pp*{gsYpg-TKHjTN;^lq0wEWMtRc5(789
zyayITBBeXRpCyj_8iPvixZ&9x+4SH?KHEZg4m-Z0vw;`S@Zha*UT4BwSLw6jYjfG9
z63**+pUDQ=j-x`Bl}}@TY1j(Ru(t-0m!yUlf7k8=T-K78R$xUD{qn7l^ic-T5z0KB
zxFeY603Pd6J9JEZ+Y0~bJ;_btPC@)jGx3o^?H-r3srx|)j0(G{CMKc0pDE|KRxw*K
z0A6y*jh@4|<GWEp$F!w9pJwHix=wqv*sefsLxjLQYp125J%VU|3$5xPc;kp2{DO0~
z;9a?Lt+`mv8Z_iqw$(dD<v$4Rtk{^FVYLOhQ`ce9u_@JNvT%@KgIUdByJ-o*8<x;_
z*wA6Uab?Fi(6<8nU;F%phq%$qK_Fwi!o=R+*KwdGSu5j)`Xb$@mmXM$Tt=bdm8Q{%
zmyqZx!s{?8+B$WA9?9$woPMihHx9s_S(Ga4(aeCem!q56kS^_*XX&981mYTqrc(>l
z+r43ZCE0V{X{*wqp|qrPkn-vR8N*5&#SU0dXT8w<05fQ)*n(Bd@iIChbYlfrb_A1n
zOel;|*P=X*W4@vn_B^*f-K{*@2zR!Y#UN5|LpB~g(kwZDt-P;e&xd<>g3M^H)berJ
zuqtZqWSI2jPDWw6buzxCFtm@NtdejOmQ$9MWA!QSi?lNAwrw9Eeu%@dt#6cOS{rz1
z?(i?l)7)q5(0|3<jSfJL>my~$Zui6qc<}e@BX|ZS{a0jM!mx`ik!&BT3&K0UNb$z^
z)5_thm!X?~A6a{aT!BHtGMe^(RxrrQhhoD$1s=DgEUbPqX69CY@RkBtYZsf*vn62i
zKbPveC8Nlm@O{v_qwqOfvB3+O)V+L1T^a||TPx5mQPjT78ybyqa?YuZkc0(O^IPtv
z5*T;r$N@X%`HW4bG?g_4jPZO?cPuW{e;D>XX!po}A2&^ik)UDi!tRc0+Y6#rciW@9
z%>c?>*e+_{5sST?9J1w5s8IRCr2Mv*Lp+Si&l2i)1`RiO%FAG1_x(8}Apv_&R3Fi`
z$wwQMUib|?w-n8=qlMi<H+JiWeJlp><FJF-*UF;3lwXZ3_F|y&k$}4vxd(0S#}y|+
zP~+Qw%3Sq$%zEz0s2~Ew>1|yvmh7zX%00r8+g_B!0)?TvR-iMR>4ir_y_aD|J)etc
ze3<R>ei-X`IYwtG^4q0B(kS#=wl~w6<cTEh>)(nihZ)Y7SVyzqPoVVvc?@N|@Nwyu
z89Q%y8H@R?7ID;tET<CAl+5ZtIRr9#wH=>-u&s`5`v&bT+H9rNGVEFqr1A^3VMi~G
z^RzLnFVj_tMpMO=E!k?S!pm<PB-=f0$E5l8GvB>ywe5wOnOu0kvX0xeaZ9FNIc7UE
zM1@Rm=<B_*VBGb1<C3$+#bN?FpWbZF*_*OVLg`+nV+QS@h~wRbIxg#6+Tl0>0-lY3
zBL$^Hr>0Ve2=pQ!7phJ-oJ(<Ao_6a!-J*Co<R}{Hm-h#;`uU#HNxxPWJGvw;8_p!}
zt2Fs(7E3<?VU3-tqUF?&!}w6Bl^-WIemsk0Y_}bMvO%M6BX(Shu%4Y4Q&1YBe2WI_
zDH}qr+ijuMKyZ1r<?*~{IIf}4GF{DHhl(Q%|H6MxgFM9L^o&0^`*OJH-$o5>vJEOd
zL>PHbh^R8n8<<0g*v{&sCG5s=K%>zlb|5jdom<z_o*bcG7mMu{^z^kRshwj}c2miz
zcfLF;`_`Y({0|N4FSwU+gC!IJGnet@4HE)2GnWC<HYk_oNF|v8J-2{KC5{1qF6wcK
z*$@PTQqEe$*!oJ&A5l68AkGug*H3CJiSfcYI*Q3}ho{PCsWok+J3VdCdmH^F-PBYs
zmx*N6E62j5Q_3KucTJCTYZs(@(i}1|l62~Q>~YfljU=Dao9b9(5UDC^FSHOfwd)%+
zdeIJt(NjjE->81{gk4*UJ^h}4@Ql~G&IkhWrtB%|jQ0({&Rzh`R~5_8qbi+!tx%SP
zQJ{b4sCStr<CzoUwi;iP!hzzZ4F@EB9SiDFvV}y$P;}TwPAJWcF?~Ni5jiOolWJ(O
zgwfQu><_j^FWIS)s^@})sZH%iY+aV-D$5nG6%TA!62=cJ+!x}ZxnYWb+W`+vd6FUh
zSAPuB20KJ4+84`@zIK7}z%u~Xf63qWzY>@vSu?a(@d71EZ!Obc)}2!X+vJ3m6-OSh
z^?N-N(Kn5R#xv3#zsgOFoMnzF&lF7+;#r$IyQY)-5aED2+X=adE55AymZJ%4mE4%F
z>TFS;%H3Sx@RdG80v<tsOA;=G0eDBM@sVmyB*U()*94u(IqJ7dE><zEP2Z^r_Z2}K
zoWV8g0WE})G#sdwa%m&8L8PD!hO*S~Rzyfuv>WBg-lz(~5wmkO2JwB}j!#Vqo@gUy
z!08Q6Pmc%5v+0Ks;O&US`)^fJlfIU<apOqESpYt_V{4e@rFM;fK3SojiVZvUp~m}Q
z=B;DQ$%T;3fTXLAJL~MAlnSTP!W+F|Q-;uMGu6bb!%0!%u96}Yl+F!P7m@=$c3)Cc
zD6JST3GaS#b!Cc<G_MB17sx<~xxw?}%zqE?T3nc8Fk9B2h(mLys!N4*i76Z1;60L2
zI|pTa##87dw7Ii?Q}qaC;w(YiNE^_vLQ)tM0J0Obp9v5o4`}ZlLo?{LFX_cqn#*8Z
zAUYu-z2v9ItIAMuBV8*c8LTsx5nSJm;$_pgRB3jta5bOGQ3~tEL8W7k8hHLrPH@^X
z>XLvbfUHCvfqtUu&7In+xkJ)5gQ}WbN#yj5S`#VRD=9C3xh2=rt`#fnq(5OG2{b8K
z48J+E>)C#wo@?y8G=eFCvCy}2hdYMpI^*1+?<f1Cz)VD_LI2wvvJhV@C3@uXW`hF*
z%wq+iLT(PSHaKn8iMBF=W-zXsNSkB(erOGEbJ5$y4Ne0h#M4O^LJW-g(~11Ytwy=!
zBO1am_QKA8_ah6zE>wf8UTc;dDT4((1KGB<Cm}oOLe4kMut4(7@R(d(%~#<Iv60+w
zjFJq2Lk^0a+e|i*lJuSlub;_w6t*N56;hfyG0F=U@%qVOOj~BO>|5OmV`P5)+dJ_}
ze!tzwy=GpGIz1^~C131e=GkDf0n$9xL);vy9Xr5(B`Z(umNqu|*-K`<8X$REkd@Xl
zJU%(dcC9=RWP5$TCdl^s-xp*%DGvr&X$(`~z9`6ctvx3cOuBG?knM%tA7rK0Oqnl=
zv29~d!gc7tZSfNd9hPVc^j?XtVOphn<|vBmY6W>C+J@~6()CKtqb;5^htG(%QZlW?
zI1O=s9UEmSLNM*t$!AUz6?x9=eDh2MXvw5^y=1}n<8u499P~)BdUNZ=EoPf)%C1F#
zxoeU=cIt_B;x@1O#`Ak+Jg3$kGT>oXl2GfkbhTO~s&&?LyV@BOl^H16po~qv1jV45
zJM{o%P0)YQEXn{$nwwH<Ac{Pu%k=kr01X3wGf3Oi_nh5RAMYQ!PIK5aFM)N7!dTgT
z@x+ubb9dMM=VW{`EDdSfOIQ3Uxq$@Fq<{g=GI@F3j2IT)+^GS9x%ms{;}BMhynBum
zySi`sLkkv-L{0L++nFp;cqg#4v_}RYQ)S=69m%Cd<-rV{|K#M7E2cCI`rnvbf@K1K
zbT8%J<dO!>SZ3X4=a%3k13R?#U=Eb4$XOo*&9^F=Z`HfB*Go9x4W~BD3E!wv@2{wq
z+X}pST}o_0AJ;k_NkD~X-nw473Hs{-hCma8{)<apX_nc<jwKz}C7gk*9)e$K<hEx1
zJFQ(-tV!EMclCP!>y!s|itIs6g=hDF#a(I-0pO5l)GAnqI3N}CAco8vDUI@MO?__e
zOcEs_B=dTm{46zv%r<dO%dS=3L92|p03@|@y-_W}QY{i-33p_<Q&v^#>7B44s+#Os
zYnkJ^ia%60zS=}gTZW-$r^oA&XdzzhELvS!km@DWYfNI^kGXlw^8<1vj_Hbj)U)ZL
zdnWv-eCeb{2R&}mbDmV{zEb@56eS0sR?=3nMN?2c1`}W-sK=)ktf6&o&4q&XnwAF1
z4#x#<nF72tsk|xiX*-9(cm~zZ;Vm>~8zcbFP@U&HFhDb5hCSVqS%AF%A`4iXbFnpt
z*As~nDIJSo6#2`qeEBy%{5foYmB$SB&-m(((hOhzyZke3@VFq@AwWSeR@Ywd3FZrO
z1G;Y#yNb}#$3^{{MO>+w2>(N5_Qm3edfm&78&o!o3_{6>MkX*qty2N_Z6wK2Cpc``
z;RIF184HOL74~<*<kGy`IF!DHHKNp8Y>zAR7P43&ZA`Na*4?OH7%d1zJOA86c66(V
zD_sZ0s(rCruG9PI7619>G?TXam!f|AnO(QF=53y)O2y8@R<YB%g}T~WwtoG#+C#tH
zh|XvSe*xm}IF})`2onM{HJ6bO0Th>g1{x@r6jLRh0baMWQzh;R0}MFnmw`1e69Y6g
zG?!767ZaCB8#gMK_gy8QfB5kG>n{h%OiK|ReWl0v>+|d3&%?L`XOvJ*O<&&*^ulN*
zbbv_*T(z10`seiR@MWZ0=aM)?So(wC{`x)*Ex)o5qeLM}=dA672*?4-)ww7%y~;u~
zC@_nQHGWvw`N<ZPZ4eq+X3Ewwbrh=&`L|WwET3M2TIJPx;jCUze{@WdD8=$UQ{2Jt
zreu^%(s#xSqy1UQtZk0Mr)r+7n{T5QR>yJkK?~IEviXI=JYPuD1^m9EPs-qefJpxK
zyPW=kFMq+#I)>@Lsq0&vMovGbzmayGAsmn(5ym&(oFXhI5HUn1JCndA3TfiyB=&J;
zV-!)@(`0e#Xh*uqe>yfa4qXe7#z`$e>V?J?9Y)t-U{e$CoFqph(J7h7MuWJgK?h<S
z^#anb$TX>k%X3f<3&~EHmVM+_dizwyDpS58*~+1L%A*Z_mq)gXS(KelmQA`|n&)=O
zi3t7{{kfRLK9eD`{xsPwlNoJuL0SSDcoOEH6ZZOfN?i|$f4W8T)U1dz=bWcWjZ%@k
z)#1~>zkmDb>)(~Y0so;9=pRrc6>{~Ke>$84465yq{M$=`aWYOyL=06JI6+8HYF)<F
zhOg7XppRDChNkrgwMJor0cZxNrF-hg^yik!0THaNWx5}2V_|BZA+4x#LBEI!kENeD
zzKo)u^l5*Wf37&{kLmDjI((_pj2i#p(|PIEgsFs49&tk>(AtFBuZ<5y8qrG<!g_17
z(Ao&X1!puaEeMZxsi#W1F1EBZ3ocDat57IS<vF%*2w+%i770gDVI!IoGL{Wou~@Tc
z=kH(Ncw8UD3Y|NJ5z1IM-C0Lol1L?2YQrj8<KEnve?6&we9)G>o+(Ix{Te+Vldl@B
z3;opwO=RH9)bN)VdN;51%WmnQ-?!H6kj`lHjZ`QGhgik30b<J%v==eP3SyrKcs?l4
z095M-WeJTc63jDxULNki$NTKolY1w3k6{F{?tK<u3K_n(EP_*gV;?{jXk&8t$<#Q+
zxiVJ?e?objF$>yy-pcg!-8>~2;e5dJQ7>8n;c@hOyj<$V!LxFf0weUJB}pN;ju-P%
z_{`xJA9>9jR0kC+qn=rcwsOlMmgpsE=YDpcfx@%-Jg%-;ujKU|E5u!FJ2(t#rUHu-
z5~8){{Dq|4_NY)pHE@Afmq#QZ0S&0gvQ6sMe^F9dV|0EK2CZ?GUYz$P(Uo-!Qztp4
z^)pCW=j*)Yv1Wk2&j#VL8rJC^i&D^5<ju|s*f2LE_A;t$)iF=eh_+QYGoyTgL^K5?
zZPEQwP>B+9KC$|y$%Bly0uYN2TfNJ|GrNV&D%-HGOYCrX0`<tb=~I`e)KP{aWEe-W
ze}xrsJT`Xh64bm^9LjrP5#<c?0$XUcxP`tXLCFG_hb<m`8}}(bk+s&0ia@!4vL?;5
zmT_hT-W{E6&-@FXne#H<tjZe!i;(ExhjkZAf`ujZbRX^%uK?WahjUMbt5%wxxmWt$
zr_^gfbB<4IQFD%fIcokZ{XJFLm@&1Sf6xPp5Mj(|6+(CIP8T$0P8dGlHTd~tu9ow4
z<0*BCyQ%|*VtM(K)VFK-wKo@sJPg!9ZWvb-%53wpD{sshWA+o(M_kG`00Ko9M@|Oy
zn0s4Hf);!yg)eh^!qkfYq`AA6c5aQ}RKrM(;sS~@tL<ZUcv)qer=fo9JO+eee*}C7
zsxOzzCPcAR81=+#r02RVONYkc7c7+I`sz}Up%?-1o~0mTkXK{hzZ7JQMkjw$Dt|GF
zot$Y%G(MyhWVFQ$Y))WYiv=6=c3F{A_1JNR`GlEkL}MjHp-lk32A^LgJBdlXxcZZj
zNn0CYVtcDQ*9+*3;3=(_@g|kHf0V>dD?qZK-lMovQrv)H(mTC&AfY3WUnv^W<OSLk
zG+xm{i@MX>6bz1xfHKlMcLJilN#Ul<SaQD^XBQq!hwmSmtBh*O``ERbN@GiVUpllH
z+MFXJt_6jP!77@XdD@YSa?e=t*0{>!B3IcgsG?~-uX*kBXEw?PG?2xae+^Kn7!mM?
z9@;R}(H21(gbqt?myUNGu`qBMDEB+Zpv7+QQYv-(21lE#fy=;&dY*Tz3KP4r{J>_a
zUCW6>9!M3tFwO$L$us;e$pOaMh?_j}`?)&pMt&*hfH}h{b6gHy&WooGh#`5{VT3M*
zWfzsVOW4tL+NT$Cl*3N#e+Q3RJE8SrXQX8<1FplA&}TQYE+?p#K2kXj8FW6X)Clse
zi_4?RE7Apn`OipjD_gKSg`%<dZOC7FY*B8cx`F%^h7jk~rU+Jz{NjwDhy|_ADb?<2
z!bgzRkruqMpLJ0Kg#!N3+S!R(7Ejl*!ioyQS*V0uF;_hV6FygSfAe`-P0oFH>eiP_
z%9{oR<QAQ-B+!+Edh*wlC%EOb%V^I6E8Pqmaxkd&96G9VA+>M^wItZCb0ydh+H=j`
zWB7DYw|k!i$~8h-W^!V{pB;%JIy|*V-12b41s;-wf;HZg<=B?@(sC+n*eJYRF2?V3
z^A|{OGu#w!K8kZAe_=(yFZH7LOYa8r4gsBn>KRRbzQW8iiWn%-UPDufw!W`hQ}LF}
z&K`NwmVW)B2&sTj*T4kME#Z1CE)|lMrI`zgiAgt~=TK(V*}WTPqyovt{sm;w*9?{F
zh{>y_`%9OH5l6?Zex$k;MkwAo7<(_4y1<7Nla~wPxC|Jae}Zz0I?D~0dkSTz5lQ>D
z&}HIIq?`+a1BF{K1H*uU;+CZOGSoP~n<ZMyjP7z?0$qACBX!cc1*W^eA|;wr^otdi
z`n4V?UmVTB0zSgtXkL%S7LmLr!+Lf_`SuOWYO;v!W&68`-nCRe!1r^K*_Y_+=hG;T
z?5Zg!bthdNe?qYQfpSdO(x|t9+)DJm!GKuc2EBuh7hLY!ldX-QbeQFe>`X-^TnHnj
zYand#hZ=MGWm>W@z1u=qF_><$TSFMf#l->=WHj>`oDK7iZWZHgx_L3sdbwVzAX5Rn
z<;u<1*KR}vRSWuAR^&NvfDy*3al{Q?9+VfL>_~C*e?TTzLB{#FRP6@I%ZO&v<8nJo
zo-o;aNWPzCkL3Fi>*b1!C~nymuHFo@ZS4wcecyQ@uHzO$QwZLbG#`LxC~hnB5!bkt
zM`Rb?^BVU+@m8G5v0|XQS?&lT;uempYVI$U6gZvEe5}^(<C(Q?=)9_R$9KE^uACJB
zA9(wnf6mo3yzA|Eg*1ZwfZOj1f^hDp+wT{X*vT126)!*N_B$r4sHXb5!+`(C+wa#)
z^V}{O)M|H={#;68r^z5$tNgnike)uT3<zq>{2#Vn@-4W=iEiJPMZbk^p(h|8&kO>A
zOY6rp=ZLJ%^T}mWukN%oY^H>K(m$L?&9|=p1PA_=^_M|=HxmOiH8q#P5&{#K0@pbz
ze@%1THVnP*uh1VbO8^8vW;{FN-PoDQt*vjFoZ6<B=FnXF_X|p-q@{LO-%fm>C6m%4
z_y_`zOX=$GN>?8br`N+TFJvmGJcGmP^=k#vpye-UskB@<0A&fP*KgJ5N3GFHIY}{E
z6G?G$@8!zJ2Ldloz*!0$ADn^5-g8p$e?`hUVVI+{kLi7>Ba{MCtl$Jh9g%E8dpdr3
z{dCaDXsv>cxA9W`_xyVJ_fQVOT2k8i>W^=S*kA}0d91*yK(_k+SM~MqYl+$?$b%51
zt!2I-2(++f1U)E<O$|XTMjIi7WiUntXJlZ^sZPjXpYG*3mUew=2B&P0&b44Gf8^?W
zwB(}Zw~t~;&{r+7R)NLR5{$hIeWvc|g2J0)@}$6p5~;P(kr)H%Pi^Bxjo8>BIb~?T
zj*d)-8*pUhDKWtfONLW@k3G%B)SF00{!?3UQPZ;kjm)o1X32_xA}mMcpd1ev0=htK
zpSDIx+kl00EQ!-=ZU=IO-u0?mf3uU7iBP(=b&8G4Zk-XjvHH3@FZWkg3b6|HLizkf
zSAR;|CwZ2n>c9HQZ*4ES`ltFmE}IctFHRP5QMrKkfOUn^bx5!s1XEDjg!vE?EkjVA
z=sv8h^uQ(V-OvECWM$fs6(C$+kW!vwyi51owAGF;3q!vgAFR$1G}f}Te_9=D@i8r)
zbMvk_<yvt0E{C8#b`Cr6A-P^h%!PxslrQFHg!33Vo!OdmS(q^hf~kXHFW}@*YfR!J
z@gTwF!6BM1DkSoBibUo)y>#v((I{z)2X~(|4nYOrok-K9P}2ozr-mTVNJEBzVv_IT
z-WV&(tsh~<dwIw1!wLsfe`w4FE7V)6y77*f&+06H$IaxmaMjH)H2SI5NjB*tPFl$~
zz_otDq?4f0OjF%)IYA@DJ15!I??k+p_~|c1LrrWF3pMdMR%Ws>{eN>Jj_}ybMMM5@
zDhPX$Hd3vdX^B%YneN53jM2@a1jCvHIvZk|WMF5MzYlv$GYqxKe{ErZrDGEH>3Anq
z&m@fuKY^+*ELoxI3#1jQzC?O^RbR5QjjAtD!K5t5*VVm4#W<LuZPdMyWh)!%E!90s
zmbcDsFYjfFmGOR6bean>%`)SIOi|cJ-W$;e9^8AC_sh8%;oNz}ZXab}L>YMYKTO>(
z9@_|IvaBt@PMkUEe>Q{};-2S;`{mY;u)<bZrkz-kkRRT+wqMTA2=9hPrR=mnH%2xj
z9^Rq0Urx&y-Kd=AHQ94E#5BPWVOiTPwEcvW2BU+B%Ql<Fj27adg3I~P=OYHwIz>Hm
zH-QbwWKaRDpGIC1*$r5{BuYD5a<LO=EiwtTcK#u#7uFmUf1uMP^IeOPP{7MkzTBWL
z_vLy*Vdi-X$!M7NI!_^QMOB%fVCPj>vSQ~|Ag$PWl}J5Z$)T6C&ONp$g-cFXy>17_
zEJ0yl$=V&5^-4;KI<^B7IVfp!z{(~&Fp9A}gwoAp77*3EXa`12%9!mpV8qd<SDmlS
zhL~uXz+>3ye|&{ve)4a)|Kct8U(2Z(p&Y#;+DX0<HHVH9H(-h5fgUqCR-IEC^j%b+
zs>ha(bA&%|l23O!X%Qi+ti3#GN#m~hvz^?%$6qUxZOV|XZ?_&gU0kGI?&XQmxme8E
z*jyM1hjvw9n+q{9l*e;M@7_oo$3!PlZwGN)AWu}sf5lrH$>RcbGkt8XaoB|R9JaFk
z&7$hro$l&j@6*EL&YnhR`(1Rlq+^8;cGB6%UKx4m$Ls9Ek`+3;Kw6=*OQg3y;V4<T
zrmv@uE1z&&8CR#V2X>b_yL@{Soedt9^;^npOyAvXuCozA)*h`%N3$U&S{9S*_TSVr
zIvXX|A7S@R&3hBt%efihTuYeS>ysSRq^S5L=LZSx<>S21muzk&7=QF!*KS(hzojWA
z&dYNC#^%M!)vbwgvmvGhcJM4$|3;!4xcTHXhrMug?F2*bx3-sbhIs2ISDRc8TB&Fo
z>B&hTwvzoRKUJe-_i8__k00KXJJzl7B!lPzF(09XeBV5|RDJ@=4q+$p2_yo=wMXYQ
zMR}#Aq3F}CVQnEgoPSm52BVV^JJ}&D5*@)RuW@_L5lC4uYZl{NglTp~)yBM!;E-h=
zy0Z*=ZV&eI(hH5Rh-;@fOV<}bj<n5wII9bj&bXiQId^f5?H{C^QeAz`Jq>)d`{`v;
z)LzWn{poN)S-a|olDP=emW7}iyUEWa<@JB#!;$9-Wo~41beDI7B@dTkm?jbeGccDS
zEj}lg&2J^3e=kozfBD;k?nHn0>eD4DpJQ-4ibDj2`x3=vS3n9qQtcz|4uwh|%CA1+
z%I|u{Tha*Ls{m6P`N{dR8&dD91Y-H6#}iv*0EtE*^by@o1(XO}+jD?P5k!(Eu5Ek1
z^cj3*s&vz-<Fr0X`{Dk+$l!U*jD-q1p=i^^c1qEPe{*AA=x5-eKGXN7Z%rT5uLr(N
zmu#oO4?Zd-FO{FFONCQhx@hHmNCE$ctnKAD)bM`f#~&U@7cN<4lR4<($cOO-Vx}<v
zeIaO&8ymUMNi9OH%$ZIo7hPpb0FVfQkX1&UEQFXg`v4smRn&RF5=jmDku<^O-SASi
z|I!JVf4U=*<{ZL~8BeZ^m9pP+)8RzeK`U26QhoZ|pVZN@D~xZ1F5jeG_9tvKD(&}8
zun5$Sr^M_6;zVh~P+IW|)dm_;tWU`z4gf#NdDl>Kc*+qHyaHoakz2C&(5>#7q*f*C
zri6xPKn@Fp8m<R?GPn5qB(meZ`<w%;a#UG|f0g<UtLpu-0(6mK>!EZ~7M&Qax;6ta
zXo>hwcjW2Z^;~Ew&xbz*qP}>_$UG$(43H-Z(?}o;zv{ZbunX?=-9FZ*pg_B6C|iui
zZn}|l(J29L=oQJp*A2~>>D%_6*E6Ja|M^Efmj$De2`h*8mD$^dCg#9)c(E}no$|VC
zf5*!oqBnucXh$o(B=hGknb)pwAo)e<@kyc-B@(mtAmXFc6?D9yv)|ywCFDE1Bi}+=
zXJ`ZMRg6ZiHtxkzY~9<YL7TMcMAio+hom4LfIJ1_NErRQr64ebSjfq5xa&7fVFU%a
zW~2b*lt$uWnAvXwhYU<|$L8*Onjl=Be^9d##@3-xp_Ii*AWnF0r<mfuPW_g7m*1MI
z=$PF3t}_jLv?3Qj#8;)M{FY&zTM|W}iqzL>jWX9uEUHxre100@AAYdMAeBRbdinRm
z^6T@LhyNTHhY*TQUH<XwLm2>P6y{}EEvSga!f*e&{POVkb=Pi7T4`OJPx*u2f2t=E
zPk!u$N!Dn%I$(Mw!i2$q@8RMthDmEl^k?pR6b;ph9Cvd82{-F*O9InG^fuXV(rg0C
zVz`Y;t7f+EYs!2V-GJ`e`E^H2Su3~R0yRG&II~)&{tw_y2b}3Yk~g`d?S^Had$Rv=
zv_tj^P2>Y<Uv_kk8EX7cxdR4if1Tn-(k^#&K~_ZtTmMgfqEiM#><98AcU<}5_Cm$@
z-YwqQfB@%|3$yI-+y&3jGEyNkik-JoV=$PIMVW}DY31blZ3;`+=ue)hz-*M1;>%{c
zl$C96FK0#r!zSOOu}w9I_~~=~b)4X4*aRli)$dSo)K~Q^r~N9epc=N;e^gldEt>70
z03KOkmnQA@afbfVlERvBt-L$4w3upXC&YFTBy`tW`y7$idqNn3t=eP&zT5I_K(8uh
zmu+Cisk9kD7E1HL5;>p>w<m3Flv1?*i;LNe9bk-a_uy(DkRuobu{O>8M1ZsH6OI*s
zHS;1p@2Fy?lU`(Y)_Q2;e>B%AO@fA{i!tZ=gn7OZt3ifLq{4uG?5NSUHZohQ4;4Cd
zaYFGTI-q9Rvml099<|SG2;}2pnOmEQ(1CSjKI63#wF8CA23_CGDg1;0wHjxmFg;Fy
z55~z-MhEPWc!Fr<-Lg-Se6p5No{D45%{Rcbty~O69g*nIS|%}de-<^G^ZLw{z~Hv^
zS?!dpz;fCwz^qi%r%mo-j+5}-$-2nF<~$m9W38aW$cA#`*l=n0CYlDW{n?`G=MA2C
z15uleBk26_Cf9~8P3{Jo(e_omveRa5PqtOmlu4{)JK=n^xihOF2i3JQo|JHhYTC2r
z^=q5^TIX+J5R5X>fA8F?tF8zrqJ8c@ppMu_R?=CT%PA;3Row);a@Xc2R%=r^7su#|
zzD5h#Vw-c=6B5%vn?sU1Tfwz>aJ}=af$+KNx`a(?fAVHpm6O0lJL4#2@-$h<=y@sI
zlh`HB43W_WBC9&+8?s`Bqh8XIB8F%mN9&({vAA82^kZc0e;xF+Oe8@4gaSxkQz<)2
zGTH~OT~TpO$@=&XhS97rY;#8l>JVHqnKRDBomDg(&{sjnpo&0D-I(>p2qu4l!)ggn
zzJoj-;BqDn)=lh&vsJc>%7$5%t{@Zbs@B1jBWHJcTindBCVn-oZTmCl`wR0wk6|V&
z=Vs&LMw;UEe+bI8|IgDDH*N%EnleU)yUmZR7#jg1Z?ex3*2#|ATD_A8-2|CKh{`we
z+P)*T9w%*NSMAAF^qpzX&n%gEo}p&u^LyU!El>LFyUG*NjpPKLMbfm@KK^$x9-3f@
z-_G(LX)B)#wY|$IZ-%!we_o`LqG?RL1HX;rok=+Ne}g-IY|08?++NwlO6QWp=Oip+
zBv4^C&wD$9=JFb;&ebI?@MD~6qAamV9+r};)hnItE-N2?%B=YY)jgEAIhC%-k>H#X
znYgz2_@x2Ri5#<SZ%V-W*h$*aM}>6LsdR@dUB6)j`aUT2Gjz*@mtCqK5D=q+G@2(Q
zIZ-QCf6Dxczp^<B{=){V{pl^Nc2A&3!bwSbTU$?lVMW@@I9XtVSN@}tuJerV@^WL|
zTF2&aK*Bg>)5__I@$vkFp6hE7MPI4WFjv>7*Y&AHemGEn60Z;kffZKdold$N7^Rh;
zH`n7XFldTefyxJ;c%j%SjfdC~2JK<KmomH<f7^HDQ0!)g;*w9|exKI8tI<BrphyzV
zixwI^_~OpZ7}qR*URlzv_op579gXB)v-o#qMO!;<>}7=Z<z6cf{Ra1-Y44PenP=d2
z?aPLf(Q;_lpu=*~@HX2MTzMsDw%?zhd>Uo2JCvnv`-TP;)rV_W?RHteXXUCM=TI)#
ze|`bqb6jSoTbG#{-gB9`C+&EdiKdx*Rm?;n>Tl@R;Vm_%gvvec=p<(`UDIIsG<Aa^
z5XZY-z+PVJ@NWVHk&109<Sh`aQz083PHKV`{5IV8E_|?;ta7jc<b8ru8l2ek)9>Hh
zY}`Xq(Z!XFS_RhDpFrjrQ8=>2VF>qRf8BN0U1P||-AQdBFhn0Kx<Mxgi=brjXSYP^
zBB;E`=hHvCNQ1kCa@~=<eC<OkV~ZB7GP#-fr$61)0`8zNr8rf>u!0lDMgqfLb8Jq|
z-?D}%tYOS$>ESKWcto%PiQrHFy(t=RtkO2WM>NQe3L86Fg85m<Hs!uQh>&gge@j3k
zKm}HQ%`UVNh4LoNT)M|zyqvY`oSAfg>1?u7iM?pXO>w-|y1ZJMmrkr+5(L>i*uM_)
zR^ZsZ3U%_rN9FGOjw9oDII`BLoYKYkpgcI<2q}qPq?fGrkSe&aC@QZNc;8*#Uk41~
zg7%w?6PU6Nl!<`O4~sdUB@Dx!e@OI`nE4ALKYc~v_Ek8vTjg;T$VK*>TSjFAvSW>0
z^84*WqCtO?<&r<s64)+DM(Mnq^7*Fe+7!`}94;yE4J0YE6T%R_UY_xjzsWr&^C+){
zmD?`G>7a~*JzaZUI)myr{++sX7L;BQpl`Edv%;Vxm;B}SDbA>5gIr?1f1=&pKB>G(
zknA2dCsKg&hQO&1PR>C<dvD*RAVf5{_#WW6kd@W2(&o+2w`~VPy=XL;TSv(E5NFYC
zWF!g^gkHqo<q14GjUsA(%2#dS4ZFsNImuOTC<x6b1i2#J)=thCR=r$un;X!m(Bg0S
z5_?lS<rUNHR{l*}bkGHcfA7>^k($<WB3O)e#>ypk2o{vHNv?WB=>=ZppwgK(<}Rg2
z^nMWEOIt$szY5`Y@>qN(6hPij8{vc>awkssk~?w2m)wUFDKca4vtf-Haj^H;iS>Y_
zfn0*R@wbChDDFruy6yXH)R??lxukjX@#bj%|M|<q{{W<9C(;UKmve(96PF_FDifFE
zAq6OZty)WO+c*%u`&al6gyjt1Kwuzt;$rvKBDck!7TcbJMf?8`J0yn^ACeu*J}9yz
za`<uHGmPB-+{^vf-SK_*<*fAH32(sR{{6=u=!2xcAnBzwdj}w_g8lo){;y{#(b9FK
zOG~K~U5<Wxx|_q30#760L>C;MoQCIW99?*Srt8Es^ugI@o=@)-(gk!`q7x8hq-2@e
z>G1dapF7E*znqQVdtrjJY6-D*S`giWSP>OXD2$|wP-sG-J*}dB!go-H?k2>ybT@gl
zekr5eGtXCi(Fj9m4RJxAGTrpu@pJqJ3Gp@FT4RLOGjD0O4u>a=2~nhX%qTQQUg@%b
zg3VtO);K33@&f{%6`wNIiH5IGLwr}()ThH!l$x>A8(ZjA^nuqyzS9_arWr;xxO)_7
zteM^<ULFt%MpI-H(M{_g8Rv{h*K{);P)Ej?o?Xqg39aNcS`^suJTQHe`H;A)J8T`2
z!av(_3uMxPLFQTs>ur&FBTHTJ*X3_H(UOdG$x;qrLxLRDG64v@oH;-o64}OIbo|vo
zV)dUCWl&iLaL$)G6<Q5EJI|h(Qr5aG(yS!PV3*o`B{zS8aI-_b&8V|NXjJSvb#X$r
zxZm!{wo*-;b>*vBbB%efzTg(;5vrH_L8bd;_ed0?#>!)M`4F-%enQVdL*_7&9KQ%Q
z`u=$P+C9C!?|uy|Hrfkf;QpVFUHqVl<H{q2vc&k`|L%Y6zV%e=i0Fa5wzTA)8?OzL
zT)Ey`83cd8WWnedq^4ohpWi?C-J9=UXPgj(qTtn2l1-eOMkDQt0?4rLZHhxmaWhV@
z1h~+zT*T7j^iF6s@JH3yipPA(7?Z;3{Jr0gJl>2K4G1&?Tv`BVF;EeDyf4B>hr-~%
zX7X7#!|*@}2n;yPr*J5VLP_-i94aa-;35un9fyBHA^jjud1}2H8-)p8fSHpn*lZLX
zrOqgBqR58RQB3-vR`)&ah}0Q^pZ)aq9krCMn0Ytal=yAfEOhpoQ&fT+L*lz=LNUf#
z3C}s5j$Sl*(zj))IV(^FUJK86?6INr5UI5kL2su>nha<Mim%IZP?n+v^rKm$2M$^z
za1wu(Qrj8W5W1!#&ES}dB=5>WeuL)(O-8Q!?cBq2<AgQ+wg8v6Wvu|)Z!78Vpv`NV
zY_w1ab<tO6xzUOgXE@24+_r#dlu%@w-D+zAvGAZBJ?CtU134g?IU9@C9Icvm%lp@+
z&XaGCNlKG8V~^zyJ4KCxYW#s0W<A2@@Z5h*$RdtKLCFS6m8Q3Z-l#$5>NcICsla~{
z#pX>-F(~1}pkO&?z8Jt)_nA`T`}eA}6eg$#?o~xEAaQB0xY##X=RwRIl^ZCXmJ<4c
zl<$<%_X#~JBMj*pCG@{`!a+#irynWL&y#=c$g0vy&3F3v<&O`-^7>!Q`gz>90rG!S
zS?sA=EtLb8+f68a-EP9yoZ6DJiQ$0TOi5dCZoja&uQiCkhq^(es+Ley?e*2GEX|j-
zN`I>VN;RCj2UUxPzdNpakt?+5gH7E6Xv)FHq1n23GXv05X*x6u8W&hVWtUf;2If{J
z?;Ox?Ef<{<()qqx1gV9U{_b)SEPa2(x#6>{3Ag1lIYa3vujLF{3m^KyYJkBhBnKya
z$-mWxAQ?bd@B0<kf@ClvXf@zN)(}^<5#s%IZb+<@B48{@!_^72F7+_-^QkfyDD~!&
za4MfgB$LCW;*L4cy!k$s5TFhD(>bYAb%$}OySO62oSSvjR`E_LlR0jeU59@h?n@QO
zl(d1K>ROsW4%JCJ?Dq~iQ1U}ue3TqoY8T|NwcpaM(D7-LJUfc#Rd+GTs-0ot8O<AI
z;iW{3(N&oEr%0iB^F5x^1i9oG5kY7>?2<iwd(yM)T1zt9QlRoRR>!yJjM|DoUi}6h
znIN1Qci;+-a+Q9M#v*A#qp*L?LCor*!NoeHC3@X|JWAe0qnhP!5jQx2U|Iq!?Z)4o
zYW{EN9#YyTd){AC1M3A{-GrFk-y?ZI(rsu&?=#hM`l$IZq(rjn#Bo*hE!9uNj@Jju
zpZ$4@;T8ob9;2a9YryU0uov399uq8k{!;38qwm!>tA&I?HgHFr<rX(~Pq8$5x3_bU
z3XxGc<#GoNLp**7i%NIg+{M1krg?No1?NRkmA>e9Bb6X9mTXl*QO-L#LDiSPf+Z?{
zL?avyB9U*)>z2%+Ti4G2Ban_H)qR#-?6AYMVlI4K)MI*Gz_8s5m#$Q|O7$-UpYg<#
zc{NlX5hYY?ZIAFiA0$dUtfi2GL}l&0B?l=K8G7OQm8x?(d|9s(`vn{^ECOv$b5b9A
z;gGfzs6`p4cALpQtX=$&8pC~^da1VvR`uBa2iW7Fml3Ws69hChHa3?5q#kLv@Pj3m
zBmy=tmqD)qB$uqHB@34kM>Z#a)mm$F+r}0Bu3xb;o#~_>y4d%llceLsv18Y%RY@JE
z)o3^n35&2Of&&7P{qKA3@*%;LV-g~r+GEpq7yGz-_uRYZE?|t+OfX}dWKuW?7J~F7
zSf-UB!7*)9jb}PW5(0iC2`B|cf?_r`cA9x*k04kUq@^0i0w#f=Scs{ACStBMqSu(r
zrH7uTr4c+7(1xD604b1m-ZE)olF-fufeAdGI!T(Kn{<wzqOlBuRD@PCBxOBji<*=S
zhA9z9@8Fn{K1rG>wdfI;(uVZIM9OGdMDUOt%@NVgM(Pn!=OXEcqVgg=wM>OT>xdXR
zs?<(FqP1BenU>CwWMET&PnyO6DbhNiB$g}asb<<4MUsGoMMHw|qeyjute+%25Sn^O
z{75pcm3RQ9=_wYYJhsv*6w&1fQdoAglA4hARze}~QXPJxPNO)6F-cQEz^a()0<*M<
z7zK4sryh}E6R4dM48u8^Loz2#s<X`L#Zw0fE7H)+>5!@v!y0LSIP8hDjuxanbB@f6
zDIo=tus#<|YG;@i!ea`|?S-L*YY8c&z7c>U7+4_EYmxc}NU1OO(-Yo+X2HBoCWSuU
zCG`y2;SWiGZ8QOM0C_>G)GiW&)5wMeC1_p?NoyJzUJ(}Sfc>E;jRy<{8)!U2zf0zW
zKLrnUArbxsw%h`L``A<Jf#IV}n+iijowfs*iJI^MRia7D#pp3PnuW*3<Y^Z6ir&(|
z08b1iNfMHeN8f&%oqd=U*>QeVWItBZrkFN!a#I6mKgmvtdDYg#V!rTPs`{lEmH7|V
z`*VS!1DbqP7wBK3Ct8U`$0yTil?Fn%K>7RckIqxXfPArko-{RL*$-7cD(W<zxX6Ca
zUSvO_0O!8I{KMv4naCp`B}3qVDcnJG1yITuxX>3z*=c*((0H%P>2>yGJ^HXIt7&$c
z{rltv{rhIzG_(2NvaFc$+w!`c(L#7tUu9Iv{?+EAI&aG1;B7liUH|mF&S&F6^>)yV
zi$Oarf6MECGH;4?gTG)|_=^dj(&}_t`LF-^FN1XfH8yQ0lZyu%z#BO54&N|bU_-ZP
z={$FGt1+$QaL+9oY6m;|do39tFsbU(Sw1XS>cFtGO;glUrWVDY-Z#%r8(4wKMa46K
z26H^afKr6T?mxql09LUqAR-E}EFgr)2ro<piqSxS^|FAlD&XL$fZ!`BN~HqAtfZ*6
z7>x!{OH^gYbu~OKn)B@V>9g#tc;76S{2-{^CaB#GpuRJvyE7)vLLLzG9@y_rE00!|
zVkSp0IE2YN7VFr*W7y7SAgD;GVjo+2v1!Y&jV-<E1_-b30)+Fu0)&y86OF)m!c)u*
zlGDk5jRS!K$G#%Oi|Mc&71LocxXGKkeBX6{7eDR(huGwuH_UBpm^a&R*xf>LG;RU@
zPOByO5$uSod{OY<gLo&9>}tLsH_aFOm@kMgi<6JSCr*D!Y-CX|RPep*VK?j%bT50D
z&%o(0M%kMk;-(#ZFFS-k5Er|yI&Pk%QT2&`{?Xq>|JaWnB75TzELef}5L;JXagFU&
zPH_?U75uUIckLCizoyQo^I0`7y7515#R=jAoB%W|07eJc?oROg9fEM5hCy$oq}(7|
zA5n*Uq+&<#SfM@!k6n(s8+h!RmYt_Gne33V<X%YzxI+M}odA1AKBYmDDgAQb7E-8x
zeZlBiU0oK_vYijg>9nW^v$`6!!{!c3J0U*`j=e(u2SN`{t(@Z#d7rkZm-|xrbU1G7
z!LYiSO^O?AL<jbCWpVhh_33n?3*Gx@?Iao<P<I_6;n+REl>3KAEBq$HZI45NE_MEh
z!|@SImvVFIeRvYxSba6;A>FflylT~d=RBnQw^c}^8@LqG=mKsj*p=W`f?o;C_-EC+
zCa!A4O6YgM;xOeN^h5`h!i+_@QJ_pZs$OAcRFx^ZAQzTW#O&<&&B-ZO0+CiJZ>;w{
z!=Y6e&?5XPj;vA#$GxPRQt$z#_Nb?%YZKwU>Zd0iH98$JiU+!h^_mN8yw-|;8*Lt`
zsP>uU(Ff_Ij2=n3=EJeHJx!W|PGl+wa4fnX#}tn4t8qHU9(#X#d&`kMkLSD?@qEbJ
z>#Ul=i;DW{L-wi~=9Aa6CX1Ktd|EWORejCJ&CR5zoi%Vum+MRA;Tpij!Ikf)tk#Ch
z^u_>SaDMG%#AclhdqgkB6$sORE<6$vbj^rpuXW^faig5zzRz{+p%hgza;+sD{D>>C
zWDKw*$OB*zq}^RnZ2hddE$XE0s2a|*U!I&Dzkc)6$zN~sX`4^@Z1lFLuGnpQ5#^A|
z5Rn671u|^uBS>fuq+XTHxV_{cZQ0wp7}e!<aa~hy9FaH^`MbU@g=i9garmqhH8OCo
zdMMhoO;wlqWS-@-*`yri^n9M(7MBB5bo3)Kpteb_x~&$;m60B9rMGCQi0Hd`aF(&B
z3n{Dh2IbIl3(8@aWOq$D^j!*W)rvD{(JvRr+2G)JPNcW0oHog+NL%nI5Wj3EW&WnP
z%zi#QJMN8MEfp6a2rCVLDQ`lQ*q??kUS}^)U;lFhzFPhMtD*Z1*4F}{EyjdH;IVTd
z_$hGiibo?h@ffzp<2_r%uPx^>8&~V}BDjYuT5;!!MlD^@teE{m)7{^Wc*7Nph@i)f
zq_-pn=ODfQ1Je+0uYQ@8^Lbm$Gavi13vgmX-xf&vPZGD##M~r*7mqv#_phpVoZh1^
zb)_`J#z*K%?D9j^Htl7R6u!-y;rQ>reb>Gk|9aI7uKw|I_~+)k2pG(}(Z$%$L>II9
zF2=yK#9=NS4-#GM#_I-ab)V>B&DR!P+~$pnjeVkvvv<*D@hQgQTa%@*x;L<)-z`A^
z&Boe6v2n;MWVOhDx*UC_g9N5eUhs|_7KD&5ow%&lt~+n?%SrKIrq!9{VXft&u+Z8W
z1V#Gz$a3K$Ge{EXD=KH`UFAjggkkj4?0I!oWlz~R^CD@>4W~Pb-aOxeMtg78Jl{ry
zX5s(;c>YS!70=&U(NFMvr1k@B;#Gm)YB<j4O*!0PES+O;CSBXLW83D$wr$(CH8HPb
zVmnuC+s4G6*tV^Szj>eg`_(^och#=0-c@I>TI*N}ihc@|wOe-eQFe@8io}`@bE@8o
zd7A#cB`k3Zgb&#MA`73#7F{zo1Sl5vUc0Q_?*)kdThtD9+Rb;Z3%om22wn9GqANB}
zyHMZA08jR||0rvyhQt`4%)q&JHYz|**7(l@P88vrFSxJYB*I*Ha;cLVNh7(Hn@NAu
zCTPYFlX(6~IRSY1u&VGRF{j`Xk;PfR&2I9K9#?*pF)59{lo4+$D`|48VwJwsCGVpr
z38fwehf!9gFqfCUI4#u8sQ%i!8p}#UOXQl|A-p*|LZ=<oDo1xZKT2LY@8|@?ad{nF
zrC6L|+vPs&Jg`7)V>pE!d4b^mxZ?>m@qNAe9Hq*DuJNj?{JeNy5K`JOv1=9$$_V&=
z#!$3Y2P2fv7vVlB1AJ%q3VyOjQx}#v@8JUiI!G3@$q80T9#(b^AEEO;5}71#O279@
zc8=f2&Th`Idqw2x(OKanEiW~kVAIfHX26w9vHz(|(pNXwb3Clz1MM5CsTjGfrbU)K
z!<iJZq`Ed(b{Hs)_Gr1GR?|s&MuNDoB2?N!5l<pby$L4*7uYJ?`0TRNE06+<B6&z>
zz_lg{Qh<PdO=vOQOxl(1MRVD$y>H&eZ6H3$Z+NP^ITAmDY7vb2nJ;vjizXS@cwWb7
z$BXLM#3)%ScIqBg-x59-M5U@AIi9`dZCLC;Wf(R_y{SB|j46RYnsamjv~eX#@g$t)
zDnk^E1_(T;bg40<Q;u6W4f|=B@qsl}r+v=EK0sbebHW;p%^zmeGW0RS-<Tf0YwSh4
zHdyp?%0g<}pFP~lKa0|v{NpO&laRdx!#=wy8NT$hGo(cNK7AEpK2d4;<o8(=ow^PW
zEFC8Bb*~$e$6tE1Y+LwTbv3{(-M4wuyE`o;B_NMkY8m$SnE#ImXW41>z1rf^prO3I
z1%UTDNe}-1kOg<~X_Zs~2C-Yn{d|-e2<3!M{>_LbH3B!pV}bC+C#sVnO^24Ra-j?&
zxWwF`$<iTNpkD{^heBN#39~@4Aty!3hMh!hg7&#UUNZm5bsV_%G`FOf5nsd?QxK$8
zMsH5sof%@tnr-|h5-xNyW=xB{s}(oEge5{Eh%7j$yv57oR`7Nl%cR*)PVn#rjBTOd
z0u=-fw@|nt9yfxjaFS*95ibzmR{&*ODQE!wR}ehBdr(rcYlDI$<8W0ca8Ph~y0jVY
zUA{O!JVXg{9fAuLVFv^;T2SSJ&}hR-I(}9GWp!UUDAsnA2=i2;Q!t8eeKi5sSaWT=
zm_z$l7#HimRy-_f%t}-q41*0?0$h_>YUt;eYPH)31W!&b?{RhIZC-u<FitX6PfY`+
zq{Jkl9_oIEuV5FSuwv;=&8s6S{$U1_-u7Sn=xi5^Ty_c6+IW(&2$51tqvBHay$mL%
zID^f>`ZHsA%&#z2aHHK4&A3%*SApq$3F6hd7&So4(K;VxGA@@jTGw3Y@oUUKzGjP|
z(AC@;@-=5YWi0G+(_WXww?BKjyz&QtdM)<Pj?ycEW{KFUe%>fSo?3J&R!zc^SA?hR
zsrtsOr}BC52ZbyJ%%eL%=j~641{W2MTgO7S|LMBiy5IJ4Z|h&B=b8js%}^<6mS>hP
zRQmdNXTDb{<C}|5eZw3j))op_;8I;ZTH4PBe$r{OlX$>)CZ|-uSL4(;P60Q-Zfz6u
z@VF(-uf<}i+18h%cN?%n0%}bsQ*+}B6b~15*>6deK0xb18|kPKe|;sx4?q(2_I>{F
z+1L9{2g%vB{cA5@2OJ&hD7S*VBwRSTrRT9i?4+!^8)rs=BW8_vat|4O8O#C#oAJ7%
z6&=`qwR_;}K~}jKH6f~FAUOhxclDRz;*rJbW^>O>bSH1{@iW`6o6<m>68Bz?nIz#M
z{Euz8L*X!Ql)u0-Ll*x2{h;XXv(%k#+bFnkM9m>Tcug<h)RX!GBhc>25;K@mVs>gg
z=LatK(Y~s(L<DU8E8k}KQGi#UlQpTRm7~P)5$(BQjzPChfTX!Rz>ftCl}-vGH7Cdw
zu~>SS-J@A;Uu=ReRM-%#5x8Q>yk-tuu8!M!DI#jl!`p2sR%1va>u9I0sL26t6XH86
zrON5T!n;lOR{rmvHXRxye8txsZ5b3>8=OBH!v(rAv)*i&O&tScE6V=8@hgA3@w)Y$
zuWgA9Ns(3MjGUK=rvTmhXDn7>?vPsCYZfMDSxdLsc-8b&67e@*7MgKQ!Y4D@Yg3iA
zh86oqv~i<;{*UfC)nO-##+ci|%kAiN2avnCv_>8D)#|4S8ZcGL%6K5*0WX4rB~Y}o
z-oFLQ$6)L0O)Y3!3MT|GiCXNqQlr~G9ggmJIJD0&)?Q3XY5>BaS~@wEma2Q|T<K9~
z9Z?2WYBqj`h=%Ef?}FW;pkhA5lbYT7C2Bs&ra7Ev5ZnF#dm9P=R;vTG(Q+DiHLBxM
z1>VlLptRPkWO|VP+g^#PIpEP3F+PniUaR!9bEYYky4he}6hJY}qd|&Khd-O@AL5#G
z!x<wmbH+)B9elx9%nze_T_h5#c*<FNo7B8T<KiKU@*024&R?`F1U6FM2~SNI`f4pd
z_`qY==J5Oto(VtvADI-4n+52SMS~8?0~E0#rUHI-QZND{e!vjuXTL*Og!e)7`K@2&
zc@oDD1#^k~KDhnx2k|#~!mjq<#pMrlff#&<-<!;7Ic>gbX=o8&@7{*)nfQeMl~2B#
zk9aNQ71lM*?clZOL5<^gvt^J`=O3~zoTZi>VU3>D3!Ve*_x<$x@CU0vq>i~mP~hfk
z>&cW4Wh)tg-}|{*f-P@My}P>_IJ9?SH2AXRgjhhPhd!V46_6<cWrkkt>hlD9KdHs{
zQNbj1VH{=hmkIeZF-#BLJqQpcgt4P&gOty%cOq`dUX^Xy^GizXY)P6;+(EUqu34D+
zC464UUy<F#rlH%F+#(rq+8Qlt<mX4IK+c}*(CH@t--SXXakI<S{VDg4^RR77)HkPo
z1DE{QR!66gl2tnyE6QGMJDJsLxaZ0F`M-zY=1BJNT|#|A|6c1G^pvsbi?EM9vNC4k
z&|XGa14Xlwt0@~HW|;by4=wq21$%nPHsnyy4O~I_w}V&cs-lLhFdfM9IV#_X`USHy
zLZJYDLNcnT3D{UIxXX<q%<#&x2!h2_Et#XH9GM7c!9K5x0yS9BA%Dc9{9OX^=xWMz
z3Ek=y^mqsGsD*?U0(E9?tUjbL8Tcc}r(agBlh!Ru{@7vhsZ?aCc}OVN1k<s~?aB2d
zrAj;>t;G_^dgqPE;Jn}a5}ar=Je%irC_VuQ_-$jcq@s7>g2-d|D~5-X2hyW6_fH>r
zH`fn7p{>wWE1%JHZM7lfi>~&3;AlI{ITG5yMC=p{(9v<<$zB`MHP`FaBe}8QcAMMK
zF)_(st}m|f?n9ROwJKLqzDDma*KOI+CKs70>0w&8Bf&=3bfSiw?3IuuB6}xREEfP{
z>?glJE~aeRRfz@7{-v_;EF=bM6n;9&Ejc1w>|asc*e<iJVOHw@i%tqQS9OWpLW#W1
zK>+Kj#upBL{nOEU8D0K(f1Z9?H!$#HF-={r!VZQP7BU;+`vM;%Wb*10Ae=QERrN?i
zD2!HWat81j(hSRDU}6#F23?0gk;wrLv9ye}UVN1pDoUC6vX$CWJW?0xEl<8}`@1oA
zLA`RRKsW?h-24k+E`1R>t^?`Leswd;$oKI)8%m8hV{AFBX2ItMy=r*Nvi`H~1=qoM
z#)0DU2W2uuEsGf$#<P?j9PIU%__Uh|!G&#A1i8x-f{0-{yU6?IxN~iCkXnGmkR8r`
zc^SjCN?QumHg^Hgyq%XeJ_G*RP$>cGE_%>v_DA#TmDiMZd_Kz|T2EG@!9{y5G3mh)
zDIM2nBW}gh;1RFtkb1M{SWX<{#5j`tVv%<EUsb_pTFmz07(^r-h5D;xb)<i-Nd?+F
z1fM`JXQ)ll?IG|QJcBB{+-1P>6zj}#wLs_bs9Cd}g@7(6{p_SL&G_GF&sKLY;oojJ
zXILg(70%+!m;1jm5koI{sGaGuj-2&EXXcy=Zh1^Kxw9KRQ(x70Jtf!_!7}J(TGP%p
zvf*%2hWAIGFlx&7Q52u<CCS?6#H(%cCNf^hlGJKShy<k8r$99J9_9f4U%8Us1g~4}
zYZa0?ueN%yn@ui6Z*v+fI;}db9VAQDO?4_dJ(d0wof5;UwiQU;cB6*~6@#8dC5z7t
zRR6T3H=i6)5m$n0?BFaSCBveaW}U~B7$<#;BIkm~aFHP{=@`O$M6`ckd3)+ljZ`!p
zfsZz`fKI1K5zYmuZ4LnB<_`yDW&*4v$FkTMnrIF4R)D!x;c;ZZx9iaK54MrEfcr0T
znRm&3C374OeLH(JNh+&5O97@tl;767hB(6Sc=x;wMa%4o*guJ**Becgkko)U7(vC5
z*sI~wKSZP@yQ`&}&ehNIC^EIO(Cds|J6#<Rt~GVJ*C-<JYLe)4rcf`!KZYCZ`3K#L
zxEd)-0P*~J<C)?qeWTrOUk0DQHVpD|J-@*d%=tt9gG6xu=l!4b<E91@^iyPm<;Md%
z`zfpeoTy<nAq}PlYgI|irdv4q%WBa!?#uhAV)tXA25-uTQ3x72hfNS=ale|UeOanC
z@IAYIWQ_(GdEJ+TEja3g-sE-qrhP0v&mUF@biB%L0~mWp4OAQzJtJA=a^;%)bz@{L
z{bA9GJ)mvT*Yx4W^FAI~`oOM!iAeAl;qAi$_`tykw2b>DmNC`h^~Zgt25E{we%j(r
zRLc933>7KSAiHRdQz*DjuzoOtYl74m@*R_0NAcY<n<jTVS&UA37^>b=tf!b!lHKF5
zS9TS7gyzkk#7HpdS1}y(p5<5;z4%AI%Xsc7&D%)+Ap8Cqdd?CCe+kdYTcrD1*)(kf
zSo$FV$|jWOz`ki+Z@Iy)Gl-?;i&G;Rw}aWIk1uyOxyY2^Z+2O`e-<@7FxyqSKfm(&
zMq>mosIE-c8?%WJKs8Pq+99`^h<o?4H6~;^Mz1>$Yi4^eoOvq1%O^HF3p1+`IKK@+
zkqj<ZVd-V+vaMy{C@&SrR>zq#mRUy&=&byQ;+*?#DMf=r>fvn;scKyH0an;o0vE9%
zI#-9(#qilwI}%97&*y7dB=9D&)yF|ndRmvte-`m+ZAfg;k>yc<-_N!t!Qr19UsRmY
z!geK1yi*}IAjKCC0#gYlwH*ofnDmk+5qS`>j^dXsGn~~EUZNi*FG5;Yb7y@3XLxcn
zyliT-mAEDBWK|9fJ8plOxxMxf@y?Nmb4TMt{s6P61Ut5B33Zf{gq0Uy<^$`o!3#Ml
zedm53j^&)+iE0|D`FA9ihIN;75V#hGFfzDfvK@6Ba~eiw+;bA$P^J$DYya806tQOn
zvq!N@DBc`Q`)gV3TuEd;cqVWLXg(@mVe2I<H^vV0b2yq-PCW70j4`tg2pkglS>01g
zrgfnIk_uPa`f$>yB)%NPo0{t^1O)d?3{JJ3?g8DdX@q&*JxL)%Nk{%eiM6$J&9M3L
zvrB3yYbRhLWZyRub(^+~4>AlkR3ylPw-O#tm(OmQ%n07!>G{rMNv-M?&}r_I7=NKj
z0EdiDm&Z+S=hQo$ly61RLHbIRx(Pb?xj65Fh(;0nk3lO}1R@8CMxk|}Ce5!ydDo)a
z6h{j=p!o1IWUdD(nTR%4syw<WK^lSPNB7l@-$uH_(_xMZ!4>1I=Kt{{T<GsXxCq=o
z6R*?e8^3ig1$0ASc;4$g0RN^5-UKTLWZ(j~ceXnPThWyG_+7)4c6!CGFgM+4*@Q)w
z8y;O#(50o$>d$>D&s4z$t5t_{A>biH@BFO*uwhpz?9EU*>msilvPthF_|akVYg?<z
zLz*?KI7@Rw?&}mJ&H(3z-1X(tCmT=c^?E;rcoe4;+_Qtp6Iyfwuzh!q+P^sWljnS9
zTG{2G;V1udN+~h^g8=|y-M*I^<`YC5A=p>p0TL7igSucuv1!RK7PUN|t82~=@kNaB
zMbW(IRkzoCU}yq9iwSNzpgD=@ruXz!-5Liy*S_3cE!J0pYsZhqWnOsTVeBei%CGn^
zJT&8Z{YAm7fUN%(K;7h@y|EM@#gytu+p^S_iuF4b3@;?nvBQOKc><AW3~t!ugDJM_
z)Lg%$^AHcZf!PM*1$7podcyy3#a+1NWTeK|Izap8NfKUacBNl(-Tqg7)4jzO92;I;
zv+7v?+47;0IR%nl>o02ay_+mV*3h$ZUpghUr%tCKdEs<f69-f-Rk$)$Ju|r`40-;K
z_C&S)ZOhvJpdVaFJNPbD0f}y4&ih)dKNAOTtw<{BWJ%+>K-y_TYf>pg<xbkO2Q+QT
z0lA0;=Mt^L^VvUFqPzDzDxG?J4gS`peuKQD>x%xTQb{*bAOPWuiA>u5j#3zM0TazR
zF@U?EbX35^Z;DL7Bf;w6OB)k1D=dRV{YR?MIiROo;NwA=so$CcffMlIxBHGKlCOoB
zkG_6!ykayRAHHL|=)ov612YIX-Kcvh?R?;%VUboXY+g<mjU<t2>TKYIIm}~}T@;ZO
zlVxi*R+wOJR#uQ@j=I5&{(L;&qisY2dj!eQhz_aMxsn9P%9IYAz!%QayMCbqRW}Pt
znokwsFY+bBt$JzARukL{ryyI}AD$lCJ^10A3vwwYZuq9fYtnUeRp#NT9Ad2vlc1I-
zD}@`Z7SAnJjF`w8Cv@~-E;}BEC`OB!D^zT@RC|etYSgQd^z8*LemPTauK7*)h^*uu
z!&B?ezAyufZhs9pTD=#L{pz<n4yM0n_ir}NbPUHsq)>;-x`8Thxn18{L!2ckxFN(F
z+YAI>JL2j3McxmBZ~dgZdFG0XANezj!;mD@ISn~k|M_RbB;Mf9+GQ?n6p>CJ%HK~B
zao2%_B)ysu4?PIk*g*)oWR7>Nw<%H;VK`&bylGVcm2o?WkOD^cAny7GOC5bViF*Er
zf|7F(+hSV^1FTXQ_$Olz-<*YamXa4W#gQ6x<dCw8mIsE5aSVROyR(b?VP|V7Bl36#
z$}T4!Jm{bLngbdn*bL`huW=sUxf5rd*{Hjfemsy``z}8HhKLjkC$88FW^h`PI50H>
zZ1Bv0B-w94;PK&(*WGcZDyK8LwF#UpYej_FT@lKoK^FZ~#Ah{#UT~}|K0Lzis*qE%
zL1DhE1=F5fv&>mh&YeZiMe2jbM<WbNhZW>s{{2ZsYUU^F>%eCHUK~vE{Dg7^&<$#C
zt+J)4+qbA@Oq2F6#5jvzXqsk;^$w0ns{0;*@o_geADkbzoh22@WrPSuIKCb{T)IKD
zD4DjEwqMbU(GKjoB1Kqm(48Cm%cm6z9^gZ$eTsr=vAutvnC;O8s&BoN41<HcYrXuo
z^iq_JFY@3gY%3sS39Jv_y2tcDnCoG%E_Se}is3KZMSjzwUc0VEOE|$F{K%e|hWl3{
z7&9jRx6JUQQ?jk*I}Po(2UZm-mv{)MW2CKcODE;bl}}`Q3FS<Ux5HRszPl#($@yHO
z@#pYkgWIqIX|!>Ao#jLbT&N2;gHcKBQ*k!mp>0~b5Eozrtin0&JqB~)xi|<T;q(#3
zd!;zh_v%v=qJnCO8KDzZ4;rroI=5d2P^6WGWF7Bzpx!3KJ_jwHmQ%U8qF0$3WwzG3
z<Vmlns$_DHG&n_Q_|K2SEu3Bayx3EW$A8`bw>a4J6Wa1xKqG;(EB1L{vKIeOPJLHP
z^iJmxAE|qu*2{xC31tCV)A}sch%<<{kPx}s!z*`A*^O2lOhh&U54uReJ7bm*;B1tM
zJw59La4?&9#oqYZ^_qi>Kp7}t6A@IQoxj=%N{??#;P)@@7t)XNLXRyr^^c_CQ*P4^
zD@Tb~IuZ*_M6L**jX!Qfp|O#(kT#0eVQcJp_?fsY=2vIY?EUS#U;iVPb<~aPy5+D-
z+~n1B&F)%>5VAR7r=s$<V;QCj(Bi%U$9wy=jP;@&#mq#4gE^4tRSxbnf?L!YMtNa-
za4;;1-52pJ^TjDHteg3WDax25d9tLhEGFReNbKt?c-)F|P6exs_Ib4J<z^Ty`qtmA
z&qCgQIV_l0+`3gOU%~M1#7E}qdZT#%pdBmb3+xft^nx>MC^EGw1yP~}IINx&4Te(H
zW_qeHrDzaz=%UVU3%c~kFajZpDvH}?HN;xiT~x<OCXxTl2RL%{uLjtx8`MY{@%(a_
z>DLFQxHD?Yw<?Vq_xyp=BBVUpN>^=FU<fThpm_>$Oyi)@O-QsIqD>pgOg*ni8`aXn
zJngWD_0`1Gg@OZb0caEeN@B~t_2x%ZBmTj{mMnLt{VU-_QP<t8FBTFD6^LAAcV4$y
zywRG{$3qf&%{m9VH~Rt<)H+pDpf0iK7{k&}+BXwJcb}!d+kTFwM$+`5k#Ezu<j=7X
zx#pPgNjg?d7>44aM;rK`Yff|itW4eLwsp%G)v`>y%BH62qHd!CP!)nT-hQP>Lt^Yz
zDfBmEn3kFkHH+-$P^13MCh#Z{7fMFcs<V;Rq6SM>(A0-xS#YA@>v_$OlfFB~a-(|o
z>0}Ec=^i*AA6Jx4bh6Q#!PbH9>J=7nXS!#mYu#E4k^cijVGv71=pfm4SmATu8{o6z
zc$UuP!*^`e7;Co(cufuD=XJ_EN*?K+UUf{$yIE4ALCnawPQb#XF!=4<lsCZesm{?a
zE~F9vCG@0LV(<H?*5=++npcU^Wqo0=VpODx;UWfDHIx+IU6;Jkk)$S>VRiiG6}@}E
z_E?D9G>ynek~!{wPeF}JwmijIFMattGw%7PLjU&cAze1=A9R6O9xvJGo`0H-T$Aan
zwUmT=55vA;&osi@ak5^rBHpPwT|c%@4Ga|Bl=;~g*g(*c-gCgRghU{YNk@7!E|PE0
zU$mVQ99S&ccQ$LyG9*7P3|n*UW({FHhVxt`WBr*)=|&MeyYa^ZOKi$b|Ch#Fur98Z
zP3<)zL)=X*K=C>?<9O37_TZ{_fI5id(Y#9RhedhLKhpju_?sIR&&ehYN*0Q1nlU;e
zB9)GWqPk3e^ut{PZfd9fusYlM2Aityi7mRDL$dFA3vYdX;(09fM%A;$)T0OF*dTex
zCZqvR{7(rZO;7zbTFI7ySMb|!eD*R<u0u(=_idIs1i~+3;G+z|EivM*9nb`W8Uu`j
z=YM}}V61FFmpV{Npcf2fGQeK7@B8h;MoPj)b*^M|ap^lZ*R!t62mTMPgdjVh?{Yp6
zN0ah2EGKG&L+IO~_eyZ3+zcEtENG0!jta{3*Am33N(TEPo^(AA{+!vta^Z>vdcjr*
z8vP9*Ai$}T3lKo?h^<1p492cdJ*U(vnSo>3R1imVj#Dmnn-gkK18{tQG*qm~TyCZ<
zW0A3oyCUGdx*?!&qGY;PTZ5s9b$BaiOdQXelWhLISbw3;C8q!S<g;LHM=H&@YKdNm
z4k@)1vF6U-Uf0<9tmu7nq2Wf)sk@I+bfVHXrPC@~UB^=uI}miqy+Mlbj;N#m`vI)D
zf%aJK_K1$-+xQaw7(n^wV}i}dhi36(<s<da5aec{TB38yy-FT<sic@2eLZ*xhf0v`
z25If0d=uPAcK9$$T<eH{@3oE%e4%xf_DfJADuj+?NWaLwRI_%jn73fc?_;Tl1wNo|
z!E}8EERT<CkGB!_<**23Vitb~=DzGT*m9&ly!VCJPl-QGpn$pIVZqAm=SwwfWEvt%
z>>4ubAwhy!xF8~?>iH=Dp0QH(bjY&b*}vTIo4drb244!wk5qT!i@gh`oR5Uu*{GVl
zT1%A<_vPT6DJ)i>`&Lok8U{gqh3FxBCeL)VQ!EAKW;n^jFR~t*-1G6@{#2^zIsO@s
zP_H(K?MzDvRR-{{`Lw!--A|{vJJ=FFP6Vu1>oMo6RF|+YU0`*iu{9}KF0Niw;CRt*
z1Jk73#v;J|_oSKEbtN-zl6RL`i!iX*)J-6xpWZOBC`WpRCDAa?EAiZ0%I_ygm-FUG
zudFX*aHHx4gzPH^TOGB+%P;NQ90YY4txd;Vx^+cH4FMlhLQtg06v|dvYHKkL2eI~A
zopcs-hdM4Nh-p7x$0L%t%*3HBlj=~}GOij=1e#5`^cqaTp{lqar-Ztb3E|#LPr26m
zX*}5clL@ED&fW&TZDcH$y~&z<!82aUq~%<6lJgsnbfLk06L-3sT2A9bl2|1u#I2Y1
z@HkRQUIELcIxgI_lt0S1_eNjb|6+U!5C}?4A)TOQThH=Pd@t`YApR3>8}hZ=DiliY
z<-IKClgcQ3N+PT17VR~iWuvck9*}gJ%pc*r8SL=F&qFTOUPl0U7NNon8)<AMUR7?)
zlN{r;XAoNIMd7sjHb-`1y&7(D{0{<!+@9`XHh{MJBp`)^jTJy_vo2xfr)%Wd(j}v6
z&t8JMw6WKlxXL9ALsU<re@Baw#0P^ahhiv|XUpQ3*pP$j26+dqIuks{g;jSAvn}T=
zZ?weVjpIPpTt!2CiO2mGq7`4f>~-yrty`A}e`S^ohp5%96{}Qty>P9jm3^XXKC&;>
zO_9N2RZ(u=Xq`?HcL7Jkr=I9Yk>4L8r|xC^qx{=KhZoxa+A&Y&%X-{+=mel{zP(r)
zUT@9ZmcIDWH=e%1yT}88{hplv=Tqci{Xd_gZQB4MrO`jj;{R2TLGck$1NFrzv%n_u
z+Fr#ek--70=3eMPW`#b+Tj~4O#z#Y94^3{Z7u(UlSF!n3qEoQUt-jdZ`8M(k7t2<Z
zDCEKSgBv}WvI{?it2Q!YQ9GGrX<DH0ppho%WShI@V0}xVQNoW(U8|v47s$Osd2#h=
z`YRJ9Mq#hI-;Tva@%PxZF8?g@s4Wt>(zr-65YPfFV+fH7TYj(Uu8PP=N4|Yiky^a#
zB_)6rq=s}8C5JG7R=vwetAI;xIN|V5v|?L(5a3MA-}>m2Ra?)#bmr6KSf9D;W)T8m
z2Us<5QBe4${C<DmPwaW?B_*QgHEszEx)jnwS+7%0DC(tcGlRuHj$r&<cI29wi<-~U
zXI%iSyTGx_p2||t$Lc_*Q-&a@unoOmYDloF?L52Bu{3798ahdFy5eU&3m>BJeD*GQ
zl!7A;a6jbBgXl^d@fasNw8f*Ncn_%Wap=VZ%&S4DAh{VYP!DjpnYzPedq+oEBBDjd
zd{^Vmno=4~O=X*7a?><!=x3jY08TP3Ed>B)r*I^PM9M+AZMlrAJDn{^(IqT+i2FPe
z|DfBY8S!l5FkK9S8eJSuF@2qjEM(#qSEF%ct_`*{`cRAMIwvkkano4Fs@?D`KA`#{
zm~+pV-8RtnlC#A={2`}CF#e&bhhFHCTJeb`Mk%2bK43K``^x(4RZ(YWU;}nv@c=*)
z?VI}y$8}>6glc`!LzfvI+|%R!^t%;<f`k28UXi%5aNaU|=3$LbvF5ba^2JhXY%-**
z-Q0jQ&Q12H);z=p=GNae_lv7pY#e$Og%t2c-(_@m;Hl1iEb{#*(HAp4c0F@FT3*HX
zo4w*DC|0?+3;%axk@Ev-zak2hXdi&F9&evzoW#-6)*nA@3Ljy#=M0&}3<S(`XwF0y
z_mp=Ov#1+W8v8Y((Z+qKl&%jTb7P`OiAdd~h`F7()Y4=0@+e_I{<Uf{zJwCb;}nuA
z?t`BVlo|&wa}GQ%>yMF>P;Q{?jI0^)zM;qgyImUpHPuU@)%m!k5u=xFpbT)nmLnWa
z(uEAFwUlwNk>62D7XE?2mP!975MiVC3Z2A8u6174N9fd?za0&&T2Clkx`~AHbl#13
z+D+R=mK9qva9Yi8d7zeOni+C00=`L97AwM^cF^H8a^X-klbm51&L6=>!^Xp=-&iV>
ze!M02fT2d=U;&~;oaaZZQZfLSaVLshkmyo)d+>GN0<$zzRa~eh=?@Z32I2{SIkY$l
zf%_d!R5AO~W)kR=ZjwUsY#Ym#`iva?08NPKfrEY>UrOD18mFVukvE=N2I~@*Ms1J_
z%ci=lVftJUSdh>9z7o|<SgwO-INDS0zqnh!qVCndkAovjrq-c06A7SMznm8LLxm_#
z`3!nyibKzeW+vWC>Za5|YijiUzSwT?j#-&9X_!z<de2zf=L+C6w69>65k@*fbg@?J
z)_}9YT}ZuQreZj|dwKwNshAt(N@BMsGAcfSVH!3TYcy6*psKoHzxcIsot@UGV|V_1
zEkyU3z;}?obvu}lIR+S)UZIRnysp&Eg_3$4rHU$>IzUWwf-?<f4#*xOMDbXoM(~ze
zrKu7p!DJf}ynItT84Q3)AD*3%A3?>6F{4)AEx#nWLICE=6^Z<u(?g?=j#c;OP2;mX
zN$?|R%W4$gb`9bFcOg(xrnlGOxs+?t)klu>W!K&-MQ?^jz+9Sfuq-pv3l4b5=IUBt
ze=yn!<o*dXNg_Z8=Vayt(kW0Xn&W$_wY150v<PU}Dyg)5f6s$jGAm|Z5k)wp$B$-l
zp%i^l7YQnrO+LDL7)s=z6lwr>;H>*+nQ$@@%E<zekQR?Bdx>Kkf<_|4xJG#J61UL>
zTm-qL%|YQ2(~;)}UFwTWW`J7?lsJHdlRRKuvdQedUx$2~u>sOLep!G)*@}$Rn0rs(
zY%_PCB1EC0Pr&^C0<enfd|~HIuT5{~$BIZ2VVHfYv^=fN6id4RnSnI#5xscYv=13l
zq<l=XjnUYka<A8L>do{rG1eq6VtT4rQ6fqU0aY;mjpJ2y9EP!-LP>f*FcUyBuPrN3
zBwTpzRY_#&quNp*_|Fysc^j$q>ok90bPWqr9bnILz8v%SZBw)1Viaf%-=2&z#B$EZ
zti#~+S~<t<V|qy!mc;eKv>y2~-J5@5-RW<@;C-NHeaV<<*^yRHayMi;92cW*PwyMP
zPS+)ol*CJ`Du2UpHqz-<@&bryWj1&|fh2Bn1RcREWj&~z+0+)iYT$e;xDxEODZcMz
zCt^_V{Q>s}C7ppeW(4W7q48OHiQ9fh#}R=@f5M_~My&LGGtQHw1^$=sR_NA|=>oW?
z`=xhUX{(e<>uOgHCX;6sS=&1DX3E>n5@8ZjYa45+hS9H$I{yQl3r)a#5~;m!Ei3j=
zrlQ?@si6bqkI(e?Jj1r-r?hecy=(ezjUchT>PB*V+~wzv_sd+wB7gw)`U(RvH*J=~
zkHN{yTWamyX9zpC8=ih$q%)c2r6}X*dr<gx+fe7b>c?FlJja%ax$MV-^({d}pi<Sw
z0*`j8I}+#LyMa3%+h`#ZwLWAagx%5O7Jx!Q0|xE3X3rzjKlRJ#*!2v1FDF2CcCN0I
zn8#L^IMpw0?4``PuWvJFAp9Hu8-xyjbNxRZk{^che>$XbTfH)+Ej6&|TNn*U?@h@M
zpb*B0<iR%$jeM`uP9Wp$=ATUL4cUc-00#LGZ!j~KfYD*aJ6j{~b{Ftlz_}sU8=Ke6
z8ai*FSNz#uvlK5eH{Gs~a6D|b>}|6PCoIVv#-c2qFy1de6iVlv$au?fcvZge9F;%>
zv)fAms662Qgv?2h&`KarrX}Va>p*mZ@Mz_T{@>XG<7EAxJlclgLpcru*F~u3^$n`f
zou~f4RZvc#3IPiS5ILCgH(+Q%>>R$VtPk*UA%YgHDh})|MgDrczq@@8YPzUK%TH6<
zL8{oYM&T`&*qgKl3Vk*B3Ht4dZ>Xb%r+ioIfBkK&PcxD+QYq{$c3oC{1B;Yp9}`Bw
zgH7(8@}*&zuvfNp9S~ZK`z15tSqI8VgBO|EmR&fE$t*orxQy`6PFyOaFvA_3`uLjr
zRuX1rqvRjUbQMYd`u4bMKJTl$z%DBQylEHAkTOYneCn%CsenvXebvAU(QSZls_$u%
z$f;c{sY4BuH+RLHqi&bcb_$&v{}EtRzr!SV9dqE0;PHBslm%)15x&^XMnnVSVBjpX
zJJ<yPynX<c`htF&0KR1m=4zFdBTQa3iE2>s5cL?0&emX0ay;N8A~M<G;WA=`TY_r#
z7Jf2Y=j1!!==PN^Q_C~UA_g~cye~E=GP7$Cp$#wVp1kvNYSzAC=#-8~`pBwV+xraN
zxvTrg0}foTqKdvc2n3nJk&87hvL99{AVSk(o-c!0^mHLj0Kdtn5Hf2N)?v|$o;2v@
zif}KR<L+A}DV%BMz+&JA$OX?zrkBlY8G<)kU~M$uS}FqWp9uw+bs9?6sC`6+Tr{a8
z#>Io0_1HK20B@29y}RW`VKbs9pea}*ok#sNi;d&Wth#IUk=Kho&$ZWDTZ&fLRZaFh
z*yMo={@PF;fGB-rgL-5wI`AM2mt8muXZbVPdQgMyiqeoA6sawTN1IRGhzE#D$PN#w
z3u=rJEKXg)JYtJkO@v#OWi#$(M#vZ-JDba`z8=FsBTw3`%G;Zi==dU@tkA5~QojIb
zkNc1U5(}LphEHN#rfZQ_L19=VPegXAIwTW@E>uwtfFe+QkKRgGcaK@{w#6m%m&=dW
z$Z`QAbiPcCI3L_Mex^uJ1`shwBQ}|J(#RAa9~fT2WAm}%q0(Ls2lsGHIu?1-zdt?{
zxbd2oD28fv!<p@7Tg~+CMvyYSz30=8nK~WQe*2fcf_NjyvBuy(R={y!&2b5vIQKCq
zpCTCw1NMw2H)Hpo#gI2K$C=S2V|?|zD>+2QX3=4fnb(Uq*`xQ)7fuBq1*75KyrsD^
z(PSvvr>qFO?}PV7IgB+Ls->4r6jMKU6^4oIz<N^4ONeN6iil$G<Q4kKVG}DjN#%B8
zD6R-$y&wdAFdCm)1(v2@Fl=We4U(e>*W@2Y0UUKTM{q({W;LB_Y91^UP^jHja!sHf
zt{%PM|LW4<Gg0a64z04uZRCe*R}7I7YOg5@?%<kOjpertS@cUlt5}YPqO5XbW=)Qe
zv?WbV)wlMv!3m~u!XJq=%72mnYoN{ZvxZYpT60&gz-Gcli!DjW)iZE#Fj;QH(yDBc
z2WV({)(A{Aa%R?wnIclR{$gfuDAU(n=MPp>o>X+#WN(<uo0M0Sa#||yNaGJl46mKJ
zU44Y~=7fOc2*kAc$k04&`2oTHr?#EfL%YJx-ZMtj^vCa{hEk|n*A{rK3^A`h{Hfm~
zmr#l}g}x~sU2U2JMKUuzC1h3P5w$@~_5d+MS4Xz~eGZGui|19^vK?*M{m2@K*e3L`
z>$)N7qjs>zmTvQv=OJTnkj`$!0OI?)ht#d<U#+#8@MnAL?m0e2`YX7(H&El+*rtG)
zj#tI>YLsMw+i@@T{nF!WOWX6V4BcQW30#%UeG}`oa+6bqYCRXvm@X`}_0(-920$hI
z3Z(8Wn5&7RKh->5Xlinn+}i=~23tdoyJ5=|^7)u_S87IBbq6y}EyZ7WcMEyM^O`Sv
zCm!=K?se04UOI#wayp3A=V47BUp?J&_bI!f@+JAs(Du@CT8%XK`BjLmB};!L(*zsG
zSBbh7%9g=aSM8c3*DHwvDD7t=N<bg$!m65)4OnR^rs?!$e~dGiFJlST90I1CcjkQK
zyr_-N9BRK?DkD3si?mMRk&RkGa}eC4r7(G+)$zHkgn$W#TsIu%Kou)szv14%(|o=m
z0@Rlima)P?@!J#!eUEIZB|EMW3y$G&I{PAX73<As<i7SyylGm7dy9u!2JpMGR!1ic
zV$Wf771?LD5p&NZeSvK<WiLvWQ?Kfg>&JPhV}LK&6r9-+gZESt#~*WC#TrPT5_`+e
zN&)xpKKwY`YR8t7w@<~$FwjHB`|(DXs$yybu7s~4NuZ(*@huP9QD&Mssit0#!uU!^
zim5!f+v5=)=XpCS`Fs=!AD}xn=qVLp|8KEgmQw`^xW}=6)@wD_0M6h^w%4}&HFZPP
zEyH-Z(I`W%E={h&;z$0T;oa1RyqIkQ54nQymbg%#Z&?mnsnhn7^;sm!jt*-T8jf#J
zwP9Lsa6mwX4jd--PpV*gJwV+-fb$MwQ(0B(doznDFWdEqrFgxiIG}m82ypxxy|q2$
zcT`-|gG)!PSy}dE&sqivZ;&{59`5i}f*$`2+nt&orgQU&I$kSgTwBvrZYp%`6}#$#
z{>XX{{r0_=g07`X{p@B`e=@^bS`Y}Zm?x#C>vb6+Lz)0?gymd+-_J9-B50WM&{zll
z=LEG}&SBd{-c=0U1mJT*Bp8)<#E$EXJtoTy?~<3TSa72yJxh2<z{5ciOFTm&@k#bm
z!$(2Nv4p&<sz|2un3#gfOi<MAm;uF;NWnUWe;(>s88D|amjzDgQ-bV=`$RDA0p1DD
z4mK>cKs5P9O&7D`w<yS7H2a#AKyQ_*wfoR<Rx?5?*M*<q2zX}HK&oCUQv4(sM5>sy
zWrJf;u9KWpQvf`cKf02tz3|r1PH~6Drw}3Vk=5{?lV@J7vUJy>WyQh&TtDx0zHSc2
zqRnryS(rvxi2cibjiBhrA9i04s%M2gWn*Hxt7fQcc>)PS>?XA!FU&&FaC13^8++cF
z=x$LMfn&wx1!!ZtzRBy$5qMEr3p6~ltlVE*yUaJlPd>}?x$IDA$%L1y494uQepKZO
zbn1C7t44>T?|Xq(6b}g}cDYe1LEs!9y!m8_@~{#+55xD_%?26rlGM3cA^Zb=Q`4qc
z8TZm5QVvoTIA*okC0bs6Yg_VaSBo+r$FYkq|D|cu44C)cc<yfP$~kh{`mmK_d~I`J
zDhXk&c5vFjWLhJ%L_b?5FG~a8n9QI90tc#PftN8koWoZen*`nHhEUhXyyn;+q>hZS
z5vp4Y(^gS0ZVwsv6kK_}Hyy9$NI(rAnLLz~i2hBLyqNu&0E5g1-@Q)ui_ZsRLKEnN
zn_O)m;37G-cRR{m1b#MGjnB8`1MN@Hops>e+x=QyylP%}((WU~>j`>K<tpDm?*03g
z>&#PZagQ3Vhc#2=bfz9&0F`sK!rPx0_1DYH_3s-p@jl6)5ApKz2EKzu3DAkxf!`6-
z5b1BH);E!vY{Pt#q>`_q7^Xchh044T(w;#(fL~qt^O^OL&(7F>wRGHiLWtMD6zqIN
z#sP4iG|@+tFOr)aP?Lf)6<a(N5oflR?frc7$ZBbi`1y});F)f=uTPE8)5)}3ToD_(
zGkWYBjws^l*{er>Lqy@b77I7&=K!6w3X4Y-Rc#QHaPq-}j-A~l&F4trSD|5L)AD~L
z46`Hx)PDo1ixL$&Fs~AV1{e`XnF?^thClyX7vCo&BdHu^eBkhrk+&_B`}N6QWEX?B
z9zGb~Cpe6)8>CEx9H@!Y9KI94`|<G@k88grKcD$AIyUIkM^KtboNgw9Hd@>=f<OQ+
zW4{P(7p#1zahx8IJdd)JS-uQ!Fhm>@DSKfJtqdLETM`?Ny8d@#ZSc4@^B6$Ob`v$^
zrqIYb<O%LJ?7QE@f*TwmgXalOiCO6>T7}7w^g{^*O<iG@eSu;IH)=Lf?MSK64Cl4r
zDiB6=gZT!;5}o*8u)6k^rR@DhLpOh5saG6wLWb5;cu5mNwG_8<yL2S*ck0+7tKzjA
z6WuWTyn~U(j8_3nY$3!r^DAKB1Kw%WiY`dO+el!)s~yZp`?5fA*we*Oog>e=?k|ZB
z`73%=5njhHB{v9EAN>j?+_H9FL9;e9^{NmVV|#lda@KK|p%b$o|2j6=q?7$sb<yxZ
zU2wzfgU&Kw6>i|~Zd??hye(;Q<zUtX=4fFXgJNIa7k?!9LP$8JZ*KtnXrO??T9<IP
zskf+IWD5guNm7Yu%mt2mq7!geejL=3)vA6bVOkNhwbvL@KtpV_fBRpq4e_w151f1Q
z=d<I+LCt0ha4Rjo!h%A4docP&1J0xM^guw8vNFD^>HPEu#=*B;&ohm8sgJsZLmIhV
z3>S*PFCh5fBGjMO*8xN$<pnX5FPnR!u8B+#!{P=v6)<VE*}I{rqfSgTi3}bMOaC^=
zsNvCAYNpocN}bw_J+nFwG!x<zt%a<-Bf5%80WbI441JL`Pg6}~C}6bR{Z}7yNc&iS
zJVRx`!@iGmly&#l9PV#y9BvrmGo2Hf>3y2siT93tleIwO+5@CsYDyQGUO(l>y(Y|i
zYSD13^SQstkPlje4VU<kk|gLdC2x3sb8kQP4Aqle(GUKJFC>JWp=<S>t8E2;LYeiA
z)rV4OM}gp-0us$%`Hgcn+psl*T6i<K3bI_s-PN=B7r;MguwIxK&`K@XQjpH{v!7^B
zCa_*H6k(R7Yyzyrfo<9&ALXS4C!{FIUx5K;z4;zuHj3n#xrc5hs;PX_byj<^$6WNd
z(>J7+R==EE!<e6PDU=x*>M;FMZAG`>(X%!5BxvvA>utP`B^HXvG~P}1sWdgR5b68Y
zkO~I%jLYS)4FkbbiSLi6zAhkJ6A#lsEH~NMX_HF&&H?iKF}L$2sRZ&rDRs9?2Vw_}
zi3<3JL=%ru+79o~ge%Aj)~-v`j5l~n6wjIr9@|j}>Y2t1`sJDyB1Htedm{_M7f`h=
zS7g^cZtYF9Rx#8lZS~YKqk}{o`ikK?y-h7itR(0z<*^fa2)QOtZ?qrRp-uEWr2Ym^
zYShq@{i~^qw&$UqFpY_&EB8?;EfUsY^YuKAZ+s40EzZvS+0f1ClKns*rtW37)_u=!
za8tW%h$(w)QF}D!72hyWI!Rab^QZ^Uq1MVmR~GUf)8n{K*PW4arPHB2q!6US$~ud&
z_)cKr{ImQ~F1q96x#rlVI{8R}1s^zd+&JylL<%^0ni~G&wy@bPcRpc(kGtY{QT(sB
z_dKo@%sA0vp6b>huDJ2bgGbiaNcI4tA!R-mo7DZAaW_R&nn6_sm}8qu@A^{zSej&M
zq3<_Yrv~bj)yw&z+iS$)q0kxW<h^~%e*Tpn{_tT+w=1-ln*Iy5Zl7j4TOcM?uZZky
z<pgk)l+E4U;IO?Gs{sSn^PRh;U6T#+6Z^%@Y!a!_!z?_p_ikis#T!^+QDWfF?~?^@
zgKlAlcZq<igMG)bLpO=GW;M?k{u@U9_~m%Ec@mtPT~ys5$M8`nHS^TrJe+=7x4!tX
z4iUYldD+0md!t}f6F$$Uy?~e!1v7=@y&E7#;u{W*F78T-vORYjhJ(TIC!q|>2R#F8
zxu0Lpfsi8mukN+em70hcqxt#|pYce8hl&9x-C*e(9)nv^-?o9hfiJw!1-sTupQEY5
zXu&-XztkT}_Kw3^5|EG|7`<{-sz2?nlm?t|*0@*m%SlsU{h<=3k=M9?x){m3Q7r)8
zF`$sduLw!s<1%IF8E)zfesDdiKX<W<^_2~8*6rHMY<H%n74VU0FF%Q2*I~qHQ8nbP
z8bLQgbteI^ujQRsv1EGhve!-=4R{BV_(#I&t^49A)~ZF8@|;&p-xjUh+pOPa#as=x
zxK)qz_-IwIN`mAIY!@kP4{r1XEWQD(EvF$3Z$4GUsw%eHJZiIk=8k8l)6~3S$|>;?
z^P`!pPWFj>mRkk7TKN>LMjN<RvO7u~s}}`Kc8<3;iMcJvy=5g5$3fIrntCuj1-vV*
z$%pqE61?PqIF+lYzwMJ8=qm9Tj_2|PODV1FWn3M)Q%Ck%##jAIxdWqcwd{ZuE!-cB
zYu*|x)jqb<jP0CqAMWMbs*;5_!xYLmx~`?>A!Ebt-3!C8^jo;}w@uZqdVAguM)xvg
ze|;Nbo6PXdb=_s?xKE8ju!JGHu|s!NU0rocbfwNs1c+y8#8$~Y2IGoOa*FCk<Y<@I
zGaP5%dN`9bQ^3{c*~!N!e3k*}hh?3ditxnlP;D5iZ;mk1{I2|p>$XiKI(RmRw+zB7
zk=MF-T35Br()ufcwgLy6<rRB{gB*krevNY=6UwC~TCq?v{J<~de@VTt$MMC3%GQ4|
z_^MXed&MQr>FwHe$yiqNXY)E3btIk#FS99BU$rM9P5c66vI+O%JyigW6h$I@)f!i-
zq(q$Jk<!<EUdTH0JkX(-l(PCSox%%?7!Nzcl*kuIKEYJ4=E-LU@;1yb5Y-J>@RU4T
zEQ3Y{Lb|t8wo!v~EA<nWn9Eg-$KNQL@_Lmpno!c)!+|+Mv+v@|h@>lPJ9Pn&6I2Q?
z-)5NhKSkcUFzby|>xlqglkltQq<ll)$zWLQ5s$dPSKwCVKuyc6y9`-^2x8?uLu;MF
zeKD&l#uQw!1%Ktl)e)sp2g4tBT*cVc>}|FjdTJP5xj?nZhL=4!9`s`nj~wOtP7{s}
zHu?K?X7CbhBx(tubveVdnk*rlE!y%C?J1qyvltNeq_+SVs#?S)xqEm8qKV}oAHolt
za|i0QkmYTp0+FYzrGE3lUU9$9(+rWTN`Fy>yNI<1;IWu^nz<5gx4GHB#Wa2-wSGeD
zZ+c1oe<hFur~(Uy38Wu{qy(Z=P(uK=I#IXtj0%K|!`S@5M?|iYxrIGpyXiXHVemW?
zF?!mfzdn)1NhK2Jy>$?4L0_dYcs{t<vau=3+GG^-kZ0<`k(p$b6urQa%2saBXW?bO
zMO2G8lEG0l`ybz2-!_U;tj~N(|D%mv3w&lC1cNCF1a5=$fN~3RBQA;pkX1>a4toEJ
z&OHCk7tte>YUe)x3u5zlcT)!tkcOg`sK4X<dQ{3YMD8Q@Y1fnqLMw9;LcR;tBuN>S
zf!-5Q0DPFe1^k^?t8taVOZvv;`Ml$FXE4ez5kVXLH|x>kKqa-8Q~ht&vrPY@E-8?(
z(`^=<leiSoroSJI4hLvN>7rdID9t6nu&b~SamHqtb}Iq*bmVh6o(ha)$ciX!OaYG3
z5r=Y;WZ8urVBd0bM$NPH@U%lDza9iCr3F=)n(&rLBATF%K?|0Y8oqx)r*?Avstxk#
z*D+7LwnR-$tIOj$3KoIImVAX6c~s^0O|{2JU*+EAZ{eP)oCin<%qv<SJueo;?Jwks
z#ix}&FBirAb5KVHPV|Mi&du*u-z#fc5BTP6S7)WmjijBzKz2?QC`u{+)y6?%sE$_C
z!@|PI+5ltGe#6ahu*KBGu;xDYER(&0YrxmDWL4^r@_CqgfhPz{Rn5OM_b!=r^VV49
zZw{!5;;V<OK?iUm*vo7x$LytGvVI1bH&a$k1iCBVo0-py`~vdX88}Y(7T9Zp<Ivbb
zVL)Qh3`$u}$0{o5kr-BNSAgo8%zmNOS?Xg*tj^)NV>F#N;)Rn*9|PmGC;OU1B;>mE
z2wMWLDvI=5(003FxJ_R%;nHGh?uiD<{OT^J<up~q0|B38AeumZ0<9H2gN8%CnC_@d
zJ7}oBg|UAXaIqxpohqbIq5ewF1*uplPy(jOnI04FE}@*lcU8nVl7)dvUBg%-ID7n;
z9a(%)FC$nNnjl0s@GR1_<@L~QdJ(8YW0RiZIFfdIsQ>bs25ulK5EU<rK@`JX4=LaA
zNc8<z)c|n9JAp<_CYT|T65l4w2~`xR>~{;gAw|^gEwwD(C;(%0BSc#hr~A-N8#Q}f
z!fEXvo$Wg4bstb)-ZYASmX~e;(PquS;kaAN!17S{S$9o4+}qMx_ibMRCiVb0jI#V@
zJMV&0X`^XDOU8ydi3;3L>DuWMW_AmcZVSbuT)<EEE8;BUr4=k=ht%xWLx%ZBuCnR{
z1zR_o)FsvE4#r=UW!W$+IW>M}{`9nHWrYf_fwJzx-uioW&`;GosDdo?#^tS#iUMMO
zvLfw&%+S~@qe_OCR3PL3a3?#=xf)l~`4=Xsomp+)9!GTKa0Ja!41zE;iq}2YveiWu
z!~lYaqnuHuVk#$Pj`?lrrk5pE{;)^HIHcLcFobB=2+#o(XjqhGyMLs4IfF`QYY<?0
zum6o=h_osN@P@0(($&qYV=AXJ18;LrW=n9!@NIEZd3izqJT0!Qs9{uSI<|V4*Y;}%
zRQ(mz{C@zRKw-ZWx9(L4i3$UfWR$Wk9yk)Mo~?nXe`+lTj5gt{SV0f;?$j*s<b=pA
zcGQ;DuvF19Empl|6@A|Fyc*wZ)?zE6vz5Emw7eF<gn=hdvl*);hEqnbHk{t2;uNJ5
z;MdqSTcj?_D9?GfCT8ASvqDY!uZNNq7-6jYB_%6mkV5D^3+Nri9fOgM@`Y%m?|qZ*
zM7zm2f5}WUQ_*Cci<Wcg2^a4>B&u_kMbUE5c@?kGYC*xx!0R0_Qw&Gfz`eq)Sb;m0
zK;k;Ae5BM-jHSJJMNhJG;YKOSNbr_2Eb+(XGzMxCx@h@&h2Ae`MaAK*Zgfqv-F+F-
zioV#|Q9E;_^k5}$fc1_fC?>x!E7J4sIS*3ge}V>ou~52MQ8j^)V!^CNqU*x(@tkZ}
z%Vpv8bIRL|AZjZDd4O*2;ZuMcC>nAza##&uyzZyU2BfEI@2BXja1<o=T>p6XOtb6#
z&h3HT3NmDi+!Fb+^A_BH*N~xCZqJHQM{!78xez0tSG4vW{7c!>Rn9nmG?Q}PkUxZJ
ze-XrFW^|vaPBnW_&Uscn7W<sKG0#Q?XRS8_;RJmh*cz(&uxkZazIHexM_6dTS_|Hd
zNm=%V#DoF{4Lv>6*2p}w0XI^f4WvN|6|GN48HcQ#k=EAv_<S>cP!s}c-_yP`r9zjq
zx+9iYX=gTaZe{FaQx*wLMrE6fDlZ6|e{Hv_J1xCZfXOX}^R-e>O*&aA*W51m^-Zq6
zoE?}Ur#e&q`k`m}k`~Nv&X_`Bp<HYEGD<Sc!Cg4YZokziiI$(0t)k5p0RECAZK;-y
ztv$x9oQqJ<eu7ZyC9;mYCfnVhm{6RWP{38H!?szk71WvoC5=5BL3^Qn`pn_?e~IkD
z=P4NlvJ)zh+xzbnn<3`d)~0AlX@C%|^x?z5-@gCl+iw&*O>h5+BPiu0m4@J_pZusO
z?Y1&ZKls_kJ*s286{NCyqRCNu>FdDS@NGKiSj|zC=FhD^VrwA48nGFi5?vQNmI<1s
zf+13zTY0`8ZDU1Mr%0eHjQA9(fAHA&NaM?>_!zJDxpbvb|C|orr^A=Z&!Ffpg`m=|
zu@j<$Q_!-sG@*_Dm~NOI!I8pjw6+VKy9(=V<Pv(}DEf#3;o0h9OG~rh(nPchh0;`^
zb?qAr=+>H9lJ$~;$Pnu&!c%fbS_#b7%-Z?>+fN?%Mx#DAznRVhAcY9Vf6|RtB7et|
z>KDeZx)1Knor|A)I;v`p4R77vEaiENCvCUbk|qK8GBx<}LLcUpez}@E0Kyo31$Snz
ze|s^t;h-gw%2gI_c;K>J?p=5YOXpDe=W9)#jn-P(q~rDwol8t-ZQY#JnbuGD44*(I
zr_-JpaJ1Ap0ye)jmvhTdf6k1i*p-g}S4x;$afN~yJL?|(opFWIAqI6FvF)n1+x8RZ
zlzG;LHEo_XGi&p|re!zwIuDK8DZBE_Q5<<3%5wbbQJkwsaZ1fn;)2rl@!NVC@lqTd
z^7%q{z{WAg`2)7IVKJpMvHrFszX-~n&eR1?=i^PnZlvpp^EqXXCxVu-N5?|`Vs$tP
zk*rn4I95)SBeQEGoMp^3m%oegnziwN!e;7lmtm3|69G1tfuSiBm*Al$!U6WTqoO8a
z2m~(1@spQfk{lBPHaM42lNS>LG&Yx!b0;U4&f7Ewf3C0S9}vwVi!Z<sB8@!;a`R$t
zK@NctI|n<+ZVvhVRFNz;o81~oc6+lipplx*N4;Mx>*>p+rw@nA%i+zl#VM%hAeiap
z^91xl(_hdu=-m_ml(#Uwe4GCAR%_$wbD<AvZ7qEQ{f_kQPDcxoW&phbdINAe1_T<R
z2W<__f2(uNGd=s$5dl5&iomN*N1hA*^w*a^A9Mxq{mWql$Q!57+v(G{!><!f{4nL&
z2d!+1Zu-xZhZR`*H)$0-2HHYd=cAu~qls_+_2u)QKK%_8E&Df5)e|b?AjRohd35v~
zUBu~YefC5GgfxLl1UXJFDrW;tkcDmF({w-+f9P}<yV0*aniNJfrdw=5*%)}D9MBC3
z47_6-mGi##g96ojM8>Is!B4zPUCWQOURLoVfBI*cN~``c9X?NokCm#<dmm^@jK(_p
z_Jjv+Q56(IAhGEvD?`>RbR-!<l@zpdS4Mk$+n`azO0<MwRDmSNPVKaYvy60j%Mk4)
ze>=#Ke2?Q-=rG_)BTY!1a*5VC1tGAALZxn@@cjGJx4P~Nt@_Y?mwk?4EYZVHM<0mF
z$XjJ)b#YT!>+CwY>MH9dK56Z2)+sq-OShKm>9CJe2TMWfIAitkZb|?lYt0>$5U4gN
zU`gSVYb3a0uLK1Q^xa56rpJ-#ab&h!e;|`%*vRB38bvZhF3~p`CFgjYXBeJwE~Mib
z_$;G{bxmI1Mla<}*=e3z3>irfDGOoBniZ|HtSuX(zs#r`G%AJ>m2IF;-m=xy^}Q|Q
z=ox&l7oP7M+NR8KA{YD&3wtXdkZD(-S0DtzJ8x$*0Q1(V(kuua;B5iH!5UmVe;1l(
znm)~QuyF~!fLKrN&MG`voLu%7{)Nhy#*aKN)}8G98=xx3AVyW_I=9zg@nv;pL}0fU
z7AG=3#hE!LZ>_7Y4~rXQxim4Kd`+N@CVQvN2#G$wz6y-eEAUyc3m}*(BF$eHc2xv$
z7JEh$;Hozs=yb(o%<5lE#y_X!f3)h;kDXH&Ge^G67(!(uCUMEBL`J@1H%A=wNkn?0
z64TrZVuUKjqLcMkQvqFMq`rUdMxXM^*J$0q38=t@W^-a89>#|)z?<g?+s4fi9mj*5
z?>J{hq{`SDr8?J(rC4zIx>%7tfH*>+pLavikjMb6P>i$A{TQmJVp8tHf7X)fd1p4T
z6)nPQu)=QmTMFkAtj_$2JS7~Ii(%_THuBZ1=?+^gW;-hkY^o7#tI^V$sN@+m{`xRH
zk%jqn74f4cf5@hRhfCOND83*ALFVzdzfYeJAID8A>STR#kimbPE=*aB{8DqEq=`b>
zdx|c5wIivIj>3Hh%ApYof8o2?0c2~*MjZy4XD+XIL<+DdavpYcHB1p(jG|Y_aLS(y
z{$ivH8o+x6-i$jNz3Jb9c<w1L8fM!S95iq#86RV6&gHlD55s2+1=T){C&G68VaVVG
zv!c8mBN&mecXJS@=^5Tv_-?cRChsZj@CNtf{c^{Bta#y*)29|8fAuVUT%0KHA@eoD
z3aRt7cAIlPD*B8o_0w(~!Q@rIxFbh~9e<1iD@}$TeaAHk`IFIKjPwdE=tfe7nC@V~
zF2wbgau3*(xXT^)^XS>pW#VqqeP5YF^3KGc9mz86xE*?aj|UibzTp9e+55<T3ATO$
z8w}?YZcjE0JMPGaf91|b%V$exT_wa!6&dMwj#fcP<IY)|OqYR5HhmsW`VQMefv^sK
z5gNKm^Em9j6@T5zoNO>T=oK0<4=I5|?fA5(E^Ft^Ui*!Zvy5)=P2N1!x=!8`tM!rE
zLNgche3bcquHsxxS`by#afL@srJbbKzSwJP=P*SIF!*|oe`z(U0BJlSuXIHNL<RXY
zo$De&?5yu;w5W*`?9yCP$|=8@D_>QDgl@Is1#i3hPT`W&gvyS6ousd<E%_0K)vwVQ
zlOd`1D;lR6W}$WVjjz+wnW&IbyJ5GkEV5h9E;8;(DgGx(I}qg(Y4HF7y!*a1jD_7M
zTqR;8p`LW$fA>hltc}V(l!!fKU#m|TPpO>9iHB+Oe3zFABrD*@9oP0@HxFsSE8I0*
z8FuWQzGynu1V0A>m&ZM3h$@-YN$_r#)iHR9L7`n?dzG#a<UN}2l>7>oD`_eimh)lq
zwwdn&p$EW0uhRO2RS@w=V&_PdR;1$^D#OgR)8P*Gf1%w2Wl|xGyX$Q>8cJ>G^Ap%;
zwO7uCJK1PidB_=zqMo`dpIBhYBJR4tq>6tL{o;VVhMlnBI`^G8a5WFyik4RQ7g-Vx
zZL2U>PJSR!wVG@A?zW*23Ng(}J+Z$cs_^DnMqG0rwk>b3>Z}&F<RC{*;I4xDUUzQy
zvewBRe^Y2fVY<VxqX`~%{Cz4$m4vtK{f^2ypm{@ChuP-RXA2D5<8WA`bX+kI@X8gb
zSLD~n*+q$V&3cUo<xSdCpq4xSu?k+sabHQJzVB*Y*XmwdA?HTx5C<AD-fkT-4ih)G
z0OuXFZEhvHuXP9w2?Y^ET8EIUEHV1INrz-Te;S$~3){e=b*POlT8HMvb4QRI&<zO;
zf3Vge6LQ|9%f`vmzXw}~j91zq;Ls1WIr@LvI%KUf#_ZBM#L+2ecW3XA%<gJ2+196)
zq2JL!)E&9i4lpD4KnD?!bHE?!AR_xg=|Z@(gGgpe1SO8I-EoFRH%@K!7}|*HJ#o8B
ze~pVmT^p&bBssNQrkCG)!+NCPud#2j%VjP`bvt<}dChOfd+s*36ym^_)|R=grnR|@
zZBuofTiqF7hSs5Tq0a4ola7^}{U%LVUhg;QJi#6PCV+g`oHqMS&U3%X`BaGDI**xq
zOTx1EWbH9qS4tHoQ%1m^-&bPByE;*Te~g$5b4*kth1`H^S1gB~Z7Zl@8&(nTxM&q<
z-8E&rCQdDwm%gi~Hp7%kFF(~bwdXkj?W%FhCWO!{SF3Oi_3k~IF^G<dYg<ctuIV%f
z;29%RU~eiA?`t<n<Us9s*iYy&;c|X<XM>oyr43-Y-VlkqdP<znxBSBiW0ReTe=GOd
z%<Y9Hovd5il%a)l7Gkl=Z3e6bs?C2@q<Wjzm&vc~R)kslE!*hI`h}JCTXfrXg?qV#
z=%DTGqrL`RX>H1HE?p7^oobkcZ5??)vKRWQ`W6fNBQnuyHB@xAa~PNACQ1)ybC_}u
zQ-PLj?YwPEu3MyPxoUC|wT-R0E}tTHZkQ}d#@3rf8O%~hIACAwWckkt4E2Kb+XXCC
z3H-3lP_*p4*?$3}<Lt|qaf2lk0XLUXH8CWYVCyprmn3yAD1X&jTW?&q5q{UN;G@$A
z9?lHO83us>$F5TY4Qf~}kW>*0McF!P9IYX*5%=Hs8}1%i>!X$B)q33mz6_VQ;c&jW
z@#LZ$Qplxh$c+EB1SWDqgQQO6Ad*3tk5eZ&l<b7IP^|bbIcPE`YS3l}AT*fxbdz8z
zPP9NIPBjX|+<$*FP}PUE97xQ!X*E#RPD!AWuu)AZP_5z=2ht)3pwO$W&&A3ny|)Bv
ztvIEDlXc4cMCMcmRdI?+DX>s=iUa4=DFKyn*g_zz`W7v4VV_E%syM}%pcsbVu$)2`
zpdhr?8O55Ca^C{-BR@uEf&kGgrMM&$hz3F-rV73!vwx2l-?=y=8RKj9bs-~B?G$<q
z=LCf_CF#FmFjYSy`{gZr!D8;B;O4hlu!s2az1I(5zCiV>X*eh-TxL?|9h?<TzOvL%
zq@P<9RCTTdcb!xJO+Xc2G+QX8OR8X%+I%iV1D*1fb8+qH6xM6~GBB-D7d$#3K?nlZ
z3i0+~8h>zPXTDg#hCTtxfl1_BU|bPDM#-U8?c++Q+WK-eV1Mt$1#1F70+rU(Aa13g
z)I|pM)#ats7F5U&0r@KAh#jg^UCZ!~mcDL<X6tg@64YE*Yjq{>++k%S^a@~O)bFfX
zFiyS&4s6n80yVenumIb^O21Kv+-h^VD|)S85`XYytup|uwbXl!LgUT=RIN$h1DI;L
zGXR)1=TR`VY2t%#z8M}L<Khjdem?y0<39o=jRxkKDLOtc-oAQua(IZ5VR$ruHNQN1
zeR?(v;uFH~>~yi1U0wuUHvjx?@!iqlbTJFG{Nc&`Vu3xLKwT*TufEF@To<_*dcz#u
zS$~!dh#wH{%p4vLKm6_If6vZ*{?p&Z4G1K;?0I%MKRcQ&j)!NDpA0W%?-m$w`f~PY
z$>{LnKd)zC?d1#{Uo2)9i#Kk4`sCsH?9Ke`<=O1b(pLFT-_L$IKmB(8?l=Y4H(YDu
z6U@KFOpKD%uA{JUd84pgRiM0|K}NTTWq)$`+fHpXJ$2jW)RwNPNQH{j|M{br$<a-=
zI(puG;U5kd;UmT|!{B{x_yFM4IISA+Nhz%wuo$qDt6}7xj%+o|8%(QVa7&5GssZPh
z0JY^X^x$l(M(?RNA=4eX+C7lj()6NitBY#;($Q;bxu&q6s-tbHqwP*Q+I0QsEq{Mm
zqqL?G9!oUdUL$InTCLW$5KB!Tvs%+pOZ>uBMvqFrl>7St+1J4Qi7Hmp#P4}zyNxW%
z?Pl=$fG=nUKTt-Gt`SY%Lm7#t0^CX&=}yW>Q)SFeVk&j3k@u1JnuxCn|5H_vCkDQf
z-Ccsi!;6dga-o$`IXLe5lr!7YsectwN1{(Vx#FkKACBJsyztgv&o6!p!?*LxUuKtm
z-*htkWq3L~I;LggIdtckq(!$w2%%#(PX##|wi4x-w44kN1C~D;zMH?850ArxH<)xj
zzlaj^SBHn!gt5)$tTlR8fh;ZdWG-=JBt4Q^<f4<!4GC6iY#GUi5*u^5KYtUHB{HOp
zL?t%z1Q|>r7bQpl+f10C1f*vvNR4w=E654!4n$!!CM4S<7s!nx^0<vmaA`zMEHilJ
zgqB`0(LvKbV#TQ4;0*E|M&Z=at@r1}-p=CVDwEj5bM-Ayndv<$SymY25uQB4lSg>6
zj)aXwTM?2*V(bzc30@I7-G3k5P}JCP4#Z7SA>G0`_Ci6*F*9vMHzoYJzfwzs%HSRg
z!wq;LjRmUHym%68ll|a3E5!z%Ju*?GMCimO9zGAe9mmU(UW#7iV=$q;z?Nc?eQ}XB
zW4Cg}MInntX)_nMS()v%f+&iYwxA8k%cu&=hge!TR^5|QNQP#5Nq?~!%q@5UPcXqD
zCxgAEeccC+OkV$7aZxhv#jB_rxLCpSPY|E&Kqq@}))2gE^zxtUXv@w`jauvH-o2HW
zy#AXhW|$@hJlVtx%`tJ=yn9YNuuK_FU*Tz77m;*~f)|Er)>{NmEV~oDhdt(z2e;_O
zYY%&Js(0Jny|d)IM1R^cE!#kkqdUfrj}(MxfZslg3u{5ar+WL`Cf+{lHr~Emi^22g
z<n!p{bG^TF7O@D61N0??3Q{v{vLUtcn0QZCkR<v`13VM#5O@*zXUC$Z*lu8gEHP(=
zKOp#nQXsUp3*TrFISnEB*Z#@`-Sj6bWFlqC12=!QNXiX%IDaSPEo}`G**+5Z=IJ9q
z@pBiw!$<+l;1u@@2ly$<%dirS><zs)P|6f2Wh<c6YhiI2!{Ra`ri_RwJ3tIqkJ$|2
ztVV1M&$Tq96}N+2a$ouiPvtL^)&LTJK{9wmI4)T8azoO3_Uq|uXalC~B1;tP$;<2<
zS&U@FOwI}*bAJ<FkwrEkZG~zzgl|Q6;klV1eJCp~!Jm;6GF2!fRT#2No6o<F(_(k=
zU1bWvoH07(JB;+&212a=8{ehV`cd7gex%oYR~_@oI%2bq*sLQq>n^ZqDVhf{@Mj2=
zdzpa!a|vLTZ3+ve#sV;FgI8-Msia+~W?tf=ScD_(wSOEi1m_BjW}7fkpel_EdHFcQ
zIFRkYtFn7=<aY2Av8jU<(LI_K(H`gh@GuGjt}<eA#P$;5Io)+2-o2@)g^`+46{TvF
znx&n@_=x49!Xp=)8=~&InG63SL^&}PSgR~1Ev*@i`mw53oz|<U2QyzTg8lGSa_uE^
z?TK>x$A1r-K7Lr<#&Vclvm9gnO0S5qTBTzl%f>>MjrA)V?alHL`kqDwu^?DN01oIN
zTW5{_aTCIen2B#IU-9lRR1`^}fAoZ0k&vzxcw4$5NvuspY9g)bwj8nPE<8SEZ2F!F
z!Pz{?*~kQ9;>zZqEaA3`mYw7Y0ti8iub4onlYek8o0y<f5SSY&j63wZAg!hy)+=D#
zVW9vP1VhEYXac_XbRB}&uek+-ZukdKFi`MzK)&P%Y3#?#d3ne`NgJ8SWf=}{VgeQh
zA+Sh8f}bkVExxIWck{lmaDhf$EHL9L!Tk0)K+B*N5*pH2%_j$Irx+eyE#5H+cK-hS
zsYCyL@atmn`pwtFFuRC<od0(I+AUO^U%ni?HGKc<=$}75_}BdM_tV8sUxCkq_txFY
zU1CO?zRy;wZZO7?{{>u6gO`zR2onM{H<yv3Llc*}z9zhX1ElZ!KVE;`>B68^!N#xg
zqyP2ty8C<Avuh0~urU1jeHULC0!1DN@*<%be*86j-Tl^69rx0*B>!PO{P;F>BVP(&
zK?`8oTIL2Y&}k2AgQEB<17@DJW)a$m(B>$Qz=O~T`RTKVRf<Ln^xRWqfs+MJBe0o7
zIa5<6@>&0XB=(yTVSl7-t)PTGWz-E{<7{V?b<oeI1s^gXrlCbjcqu%chV&$!#wkPY
zo_Mwt`AHtm)VRum`FoNtKB4UPIlqr`@|yfr7mh?gD1VaN&B#m_h(QJCd#)n+wH*YO
zJTCCFjbB!EhR!PI+=A+{@V9mVPPqart^h@hs{ew2kvU}Ucys}guc)KMBV`#P!Xvsa
z&UBAJ2dtRZwGrt&e-E};d(?5#hRy4Wt^^5V^ZH~Fg*rU1TQQOX4g-5kcxNdN9e%}j
zmKYEUp4hEU+DJS^`w5BCCW>qnINMX0$y;QbFfF<g;mC`!h^@EsX@kaWQ>$<0=Un#k
z+;16weR$duicycSB4;n0)LBJ4%#;wtd>LnlJw^Xr)#V#?wSIg{8J3imLw=<)n$=ku
z(HQ7$MVO$B>GytTML4fS=6=r<3F(@{a2~pbqO-XY-1*|k?G$I2uzoH_J=xORm^&GS
zp6I(mj@l{{jlbEFX6__zl$;pl-_f4=MthcjPt<%Z<jv}gIgA1Rq&>H*XW-s!y;k@$
z(C?6PK%MhIJi((U6(rbn6!g=HRg$Ji4DVJ=VSpg*oKDMooF6Uwdbn1b&FZX8F242Y
zyin0n_>fT1Zb@9y`=@tvQbtTen7R_=c6A2X=3F2iYhz9T?N!jGr;@mbl&8Gy)Wkf0
zB`@w)7YVjk1?4=gah@`m;+#MyLw}1BYE)Wsyvz1SEuqLSWqwy*Oja0Ka7s28g%ScS
zC5ICCwzpVWpR*cfDFkrb(aGwNeEj|a(XOWd@l6FF&r1xLaoiT;@+Ct$7KW?3HcZw#
zF?iFnl=j7#w&n+qF)F>SLSQ9hJtZlBJ}QU49D}MlWG_{)l!5Yi;!=_<M!#$yI3$*E
z+HZ+mv|vl_4Xc2*CwfCDaX^lt-yyaUC{VGM7WR~LXInkal~$}ajkLBjQ|z%)8x_nA
z9hxL#JF7UJ>Wt!B?g6d=tH|KOT5IFvn32V<3dJz1t(rP%YyKV{*Or{6pz%n5#e_;v
z1$n6zlFjL)u$2<?bmxYg_z&MVZmdpj{v#SUQs*lZ@Q#g!j^Z5~Hyz~@jhk-v2Q_Xe
zc*V}#5zn4t@|>4r$P$Et@2Pbd#voSZMa~qPk+%@AAZdkL>~Ji{nF|u3Jbud3PazRl
z1w%dM_t2gsH+GgD+(COS#kID7Ta=^5!x+u1hR2OI*h_qiO=7V0Xct%onq1>_TyC5!
zWQaD<O`iPq<=b1Mw3MkGMS%6~tyl6am_0i_T+AOqj7;`S{s@91gf;#mKC!BZSfv8e
z<qm5tBx966(WPWRNX3g%;ID-zBBKIM9lG&4LY~9^A?DSkut~OLN;}ej)EqB=Tk+L(
zPAGszV09h^dU_d?j5m}2$|R7|VFIR{t_;H1MqRP{Sqs$GLS;;mxk;F9@+w&Sd1KdO
z)_|?j<XNLcXB2vFvo$;Bo>RhfuQVy5hKN0S+~{&@*iTwoYvP}k^N<JSgX?!(6o?!A
zh9J>;sqM9-xwXf)Y{(dYP)>)Q>a}^7w-QlIpLZb`1tR!IBwpecq_luLhMJDz9TTsP
z@`=Q&=i@3vjbZWfTP5<cv+Fsy^;XHVl4n7+`fHAi)IYg1Fdv!YS#|G%Vw@6LQ^R`i
z5_2j*8=4#q@(K<lma=Ud#K-A!gzN9;nF~FRP4Akb(4@uWhKuHZqC3s<#B^s&bq<SL
zs;f9>r|i@gr7*jsDd(o)#im^g$3VC5ZzQ@KJHHQ1`#n!J`iNG@fh6ypY<_mdw8Rt_
zReI$eOXImF&#^nnhpK_wWg=GAyX{hGD=BTcBcC%H!Zy>1B>>vnxkRHJHWzwH+vU1$
z3cc+G$HJb=t!<HisA^)CC`?ktRy&p3@FW#Y5+w=a=kxf_AU8`W!Ocm%HR}{|Nk}*>
zIETWuUgJ8p@QOv9g#|v&(i+zxTke^=y!Rdj7X*m0{*F<wqgd08^t^QXF1Pbd0e0Vl
z5^PX`=$`Vm6j$RKUmoe5jDYQ_Z$^4A)y9&SF4C_Tw)7!?g>Csykn`lJ5x$#!OJy6p
z5?OkRTa}|mMVc1ezGG!;BuUqJE`OYFFjn}xsItA>Bc{8mSu^S5GbJ#hN#OrsrUWEz
zgx$ZEDFIq3)%;JDDFLw59N|BnDS?(c_GeC%LTrh^|0YjbZ6kD_FR#0Q08}IXO$udh
zWOH<KWnpcXEZRK?0yH<5G5ZY@m)tA_DVMIsCJg~7m%PR%8GjE-DpgfKcFI2N06pDe
zNfb%(t&-FCNlq_^=P!qM7u1sz-hflnmv0l$14;iu6408-0SJpQeR-Y!_9UgU^m(R_
zmJ*RZj(&UkHm4(krx`#`fSv%Hj!pwXXdqG1yfRl^XlzeM4ch9s!o^jmW1I{A^!G1+
zJ;*}f$1jJGAb%@SAlT{a>*2SFCcaE*_CX2^-c0|S5?FvTI!sbn0B7ktApK3h)5LfG
z`Tp%MU;hDumi;$Qm6I0AK=9LR9y%*DsxbX1Lob{sd=rR3gzL!@T2I({8IQkC2d$g{
zG`<0UL|_owYo*#s1cBbsw2Hg~mF*+C3go?h^gt%4ZGVussWik<#8TGAA+6_CaTq`S
zn5WXJe@%yP)8SK*r?JY#DPCs8!P4KFJV9u1q%fUK8D!~<te2K$5}Y(<1KRUjfr_MO
zH4=v;HPRWJ>TZ>1hID*0h;T6m2J!a@-;l$=D?zGy8esyl=txo~Dkl`_W(l|7Uti0*
z@3iWr`+v@Oj-XIzW2d8aByA57SF%&%SM9V+=h?+h5u`L^bB1h7XPe@k)@*A)LR)^C
zI+8MP$Hv^Jy9ENoGa24(fiPJsf&jAkXkM>Z9TxMTgHWKW!!l6-^dZNk$u$}JZEO<C
z5q)?9d?qpJOM}p_P=serU};P!@>&<y##zzP>3=v<yv9_46C}td-X&t8?{mhG_MzYA
zL%&}-q~`)jYa_J@8<eNItb>L#Q<Bz44g!zVhs<SiyIALh!h^|<wo+u6`yOMO8TWD(
z(o<xMEX<s;EFaqTLe5>fRn8JtU1U1nGhr#JI-Dswvd6$@I>YLGIiVy4S5Es<pS&-;
zUVmoL{<IP#$)VvhCzYfWaTN^Ab(a50DSqy9u(L7-Gpw`?Sej1Nl~GO@;>euK(?kKw
zn@?edT3u$aj@cr~%3UvGUDq`&xS|1dhpGV?a<>M|c1ydYQ+;<K*K*Brl@Tbb<u1>Z
zr2CO%9rZg(SQe}s{=hCUrc*Le-s(K(BY)x7d9S3sfKM}Dk&+V9LGNRXO@R<i$|qK!
zpI<h4RYZ}-u;BH^xoDNgQBSlQPk5FxZOMan6JXt9=d6=^@rUI0RlsqVdv#|AJ~ics
z+AXz-y>-yn>?b`<BkbbFxkQOJ;~qQ8Gn;%D`d;GNgEETG05;R)0r^HKZ(%uy@_&C%
z^Jyj7K6cqQCn1z-Dn;qqO0i%Z3Q9w(z132HK4*$Mbc)eMnr6^Rh|eS}?Vko!D6Y4P
zfJ0{#!TGxBGdV^H+PvmO`<pW*?!t#K?qn7m7S(=?zgw!VWDr)_#qLGbDN7zg8N1Y^
zzMf~)FmsNO@Yp#pjt7E6*&{UC(tk=>Del61fvkjr<fe^db7fO~o^fTbaa2My&*%bP
z!C<<is~CGZnT4ilvL`A~>iKsv@1=1rl5z!el^9uoGAs#*BYT&dX|7<8;3U1!HY~aM
zHQF4gtR>%cO_r_%WlGFz8dqgUX$JjrtXsaE+(O<}x~ylqq*e*q60Zv3)PFFKE+v81
zvvhD+ozu-O{mMJOZ&HbxTA&~1>0$~~8Cy=+A@TIS@WhF8v#3>#q!M}I5inuuB?_`7
zK8a3Pr0sCsi}dRr3Sl3L-Ub^LBYwRMBMNK3Ef_-zaNeef6d&Yfh?*cMp<I}L|L64W
z@M%O_RF4!Z)tvF4k=a0IkblpmCl(}WsI)e!Vdw2BYI>t5BPogVRYP?txLD^Sw+aTG
zPZqbtNV;y(j2KtSnJg<E>EwTj{2AnPE1xO-rF2h?n3bH-RcAXbtb=hjuK8Wk3rSqU
z$3oL>k=~N0_9zUcruWN}q~-`c#D<M5yUrk=Z_9kxRpBjvyCFF*f`3|Ii&EN(T2>}2
ze@>$;LmOCPger5>L$1g$78Fh`WLOIvcb6V2B+u{Nf{Wt01;vf_FVQ!H^m@l<>Xfk3
zSB-ui&Tj{;F;PmeqZJMqll<7@ss~#ji}ODv_gPgSR-t$!yw^lI^omp4>F0-?0!k~4
zaU-L!Fs?tWju;OI)PJ~x+D^ajIUuA1GE`L#Sl^<a7=**FQrTZUf#Ox`L++uTbBr~?
z$-({c01^BCZp6A7AZR_+fg=rXP7Qlg$tKc!%YEYvhFGoH9gk$d_~-Q)1EW8|V+=d&
z<S_=UZ}1rCwFp>4BjD~H11(Y+ANEwwImWri;8bou^U@$d?tdAO2aJDW^O`gIGt6tg
z!(Qe!$NCoYLQ38>ZU*GFaxS$aJrUw8uJ!*^kW?GiV1;(5z-_5=xySRK`;bYgu<IMh
z{@Sk8`?+J;u61~AUpr^WT9D+rInTh-T~b4TN8Z)vRE|Qquer9;r=!YkZZ;6vRqYU;
zn(>lL7Za*8^nbwJ%9@^^PKIUe13koUg{?c4o<$zbIKkouvKjM?(i_V78RxvBYn_|1
zS+8odK3=3Y3iu?iaeu-Ja}^WLyPNY<%~w$xBF^|WP2!9#FtjUPnfM4#-Py>(gJ13X
zD4YBGT^}Jym1vaxb)~%C5min-b*BWKj-;p9Q50wOG=C>)rWw)Y&NL`NqvN%u{^hHk
zblGxgm!ir;dDke(%_h&n>sc={72VZnfo${(Q(p@zZ=<#>dGc$0TBeV$a4Y9f={px4
z${2ec%8;v?F{4j)`nkbv?O%p%ozPxLSG~g{cDd5~yQ`aBBiMqL<mFp`>Cww_T`#}G
zsdXy-u76jUG!#fB&Oqz0>Vg{+Ux!-k-oGHASywgt4yG$b6t(C>?|8lI1Uk2%&Nh2X
z*S5Fg=UrFU9)wY9vD?^1&m8MsGE2`?pN?IBjVt!nVe*9+WSv~B>-<Ph0eR}^x<Rg+
z0(j9S`!_sJ>B|o{UD-bQrYnz<*SutUx2#acQ-5#0!f-<K)a5a`m+y;(UGdXp=H|;3
z)q~`?ns)i4eSZK-l^u+Ie@cDg`+E1Z{Y|N7T5-dW``Z5CoUrY7yS^RLQlGnO8?;0?
zjM(OHZT00v3xo5+WU8zc*5jU62Tb7{^Hyg_YpSrv$y+xE12Yei<xzlJ>@?I$ZMG8{
z7k}cxo1UqwVs76tEDpouZu}eX6m%eex4`zx?i3&x5#r0R12+%eDaiPeu3d0(URPsM
zBWKNC%}rDSlpV2ds&H;Mj5Pf0=H|K}5Jh|&x_>|Gio!~LMKLyZP)d017HqmMxRmj%
z%TnBX+O6*#kunum%AKPbWX*)<6iCYr7#_t%gz18<y{i>)y+rmiFTl~BV-cfnuT%XW
zOI_x(mtoK*6_?UW1rz}`li{c)m++DV1(&waCX{~&hrfUN+nr<rr3g0uj9>H9^QXJt
z?q+gLa6&7yeExnHU!Vq|y;{5yHY~sYbNO=jVMbb)k~Bk@_=kS|{&krkADLf462GLi
z#ti%dNle7W+8}g%mHB08zlL~*C;A5ZejN7q?#06<Hf@P+n6as+j4^RE{&_jvTL~-|
zH1vP51*!^^xNUS=bw#W3LK71Ttu)n;oRH*!KF-Xv3XB{me-dK`2r*_w_lQ}^Vb^su
zqVQoJtXE_Z;Mi~RcCx${;8rV<if)qSoj|=}PgG$;#2Lrpl>Gga=`;~=V%n@DB0w)t
zb!mX~v=w1sT-8;U#gzhK#*8_x6n`fUyu^PMcUT$(v3154`DeEHS{29~(-Tu4it>{w
zW$Js#$M^eY5R3?@SG!^LH|{F+SJDe7?P?dNZ8avPL+1yRgS=0pUZKKDfU+~zT@@zY
zXtSchG#)1j5_eHGH(-hk9(gSC*i(LIPQA<yQ9y&(b{L_IFc2ycmKD?G#mzMed{KW<
zCZBQ(YZa*Ib0;!)tWbdy3d}tE1}j{#2T+*i?Jgjac!nE03rH=<@vfA>Wu-*ho|7_I
zrov__(rU-fiIAyr_%Y1c0kY-bMhc`Ykw8E4T9KybxaPSyTg}t>B?wfCY;hwuFcyS&
zx(c#J{_>k*IuQ|-rs#NN!Sg}c2bO<UvC~Wrh4gt13`%Gn<N0q35%zvb)Dgitc(eR(
zNnkQqgV94-0B7SiH2sS#yL<PauV4Q9`5)wUm#_bgLn$dBHbU^r_xxz+*&C0`H-5Hc
z(MTExB2b=!3xxrr9StZGE8z2Thsu#xH@<^@i%?>lC&qP?Q!3)(z_LIGDl~sYj1^ke
z`_=;sRc(p1r%FbAidcAVe5CnhR(y<~{<#dLS^rw@zASein5)Jr7l(M65eG~Ewzvuc
z!BLXfakhZ86Tt<vg)7044(>qv{Z=4lZr)NvK^0>HV2q%=THSMjbo(wK!c{O7Q2CzW
zJA9b-N|36f<e37^Gh}-%swjUjcQc3E@6X?P-dCFSzWdI2?m=mR<eKhDRmidvu9Rz%
z{cYvuJiLsaf-Hs__{vJfqI(0IkkFDJmX4&nxBIra4_7mUjts__D;P5S{EiXi$h<d|
z5j%&JvIkn2jHiT%=wFZQ&B+5Q^j$PW+xh(H5uuF00<}DfbN3oYE3SX!%3PPALGRt`
zsK5~9pq|3hLBYd}FU6Nop2jTy0LJ7fNv~ae9S+)*0A?x*V5E-~;ycn!{vE@(vN0f-
zvVC92OD6SHA&+f)CT%M`@K)5o)@Np%uFxArkoSiv!O*7z<zNq$ydlySfwOwB#B-{b
zWEn^As*Cm_y5w=Pd^~?Yi84G*p>>)S`Jq8|oG-%HpgS(*rxbF5iF~ZkV-Hbl(HC5@
zQ!~t2V^JU;p_^(YWvZtHPGD?V0ZUSH><*=EFGX&?Vzy1StB%>7UD{E>i{fW&>Y>|3
zTalBXlv0L%<$Yev=~%=uWszHsuDO&gN7tNIc69$<#$lX~+_!&(<cjJEI;2FUd1$0<
zcwBF;<a^%7aGqkWHoD||6ZB&}IBjaUq>OA@SMQwcYo52W3yZ=jH)AVCop*wDPlM4z
zghn@2k3gs5irUOkY1_eVjT^@VGC4q8jW2$*(UaVqZpY;};IXDeB&KQ`JWf@vQ_g_5
z${31|c*$Y<t7LzmgBB?_8ds~lMs`Jc$F`W{I?`ybX9M@<(m4p%v65Fh>u^S&U2MTB
zfmqKbqbqO^O}@qz7gJ(wdm0^-e+~WMH{DL!Pk?qHUD1zpd~Ai#>brSz=+2)p$_CG-
zk^+0sbIc9ZV{!zn!ZYo-n4)eTi51J*n9G=|Ud2E-vTc7WGM(fpe}-veyX7~wZ1tFb
zYs-3m_8gr`7u}hBSIx=QfbtV9jgh_Fd9xQhf(okhRm{2)f_t`}V-YPPkj^}7+278J
zq+sWGU=*G4qVwM20kUFn5pAWPT2TbUW?#jwxQKM0MR8E-q1cbBD#&jod&{Cowh!XH
ztyY_V8dHC|J=*~kR#n@l6xOMZnj{lRxP#iMa_g}2UJ25?Zh_2tfpetDdQo-;!@P>O
zVyL-)m%h46&$_z5ru5C?jy6IDHS==Z@37aa;)?^JY*=9pT|C}N*KIlusFy?5qo$dA
zYawM~GDgmEtqTb_DkD@Z0F1LzObEI^R#fRFr`&%>$Dbt=^iEi5-dHAv_~iUb(lxv3
zJ=e<8q!O`8v)M6vyF60`0ankey*AG@B;kNFwsIDSC#L(xUZ9FKGa2SyAebO%rK`Gu
zvM;_4kn6gE+S82Y%aNr~{VVB+g6leh+-R6EzD3W@T9T!jG4C}zC=ay`=F!(@4Yzj<
zC@g;>d8lc{y|w{)3#V3`zpiaSSf81`%i_N%d*dm+!nc(DgwfnCYi<utnD%6C$h(mv
zSgpI~-7a=ant2i0_9xLiKsFkr8l3=Vw$585?392h+fmjRI3BOgQ_b^;I(9A+>T-;I
z%6WW**S@T4HJP?d?%M6>T3qJ#H1bdnbOe8k<vi}#Hn&Wkk9@1vRw%INSfz{5PUG3d
z*w&3yq3cY(B!z>AQT8*tD*CN5%`>BX;dfZs1R-AoiF?YQ$uN+jq6p3zKbjhlwZO37
zKmX;IoK29LZ%$F<7u-1IPrux%oV%u4lr<(TW-OUNX~vb4USA*M=4zMr#xk}(<yL=T
zWN*E0bIN1V_fx%o^TXyv$p)P>b|@aiS*gV=@I<L%s<2V1?2l6=wa~h?v>#iK9FO(x
z!p4a>(~Yx3@U<cOW;SuHK^$tW<f2W?2Jr)I3_??$ree|3_sRYK9tu};8z~p;&hSjW
zW>Y_`7s^vopx}aDTt1I)?CkX({}z9DpS0on!ok~Ufs54)b)IRlYtKf1{`yfdu^qM;
z`{VRdrf#IZmfpQ$TB5sMK>7pdMlTHF)qLKmr8GD6N6;MGz}}vxxpEO^1KM}NvFE|R
z)@Ol>v#*tTay6yd1-UxFBf2_shIE`at%$qPMB2L*bH3-J<lvgL&y=@nbqs$*sokkG
zhx{6BV6JnQzDSaE!}2&(z;&sTPc<s8pPw%R)7e64JyZRYN}QS!Ho$qYRN%{H#Z`I#
zV&f+lpH5@Tpt44GSFer(bZC)`#n5XDq5g6qG+p5!Zx9^W*oLxz(;|TDuUEhB0CSP`
zbwy)@=>&aVJpe2bkMpdba`=BbOTHpMrIeZpDS}7kSn#O)PS-KNnLB4Zo@3i8Zgl;q
zyhg90JU49$^s#xx)_JQpLn@7?-O4P@qhk9^;bw1YlZIgrrrcT4{Nek4-7ZgEMYpF?
zHfmwJdk3h6X|=g>e2Jb~n=t$E$PG1_P2Yn{Gh(fvPckz!s)OST$!ULXZ;GCsg$;uw
z@}(7~;|v#g1-`vg@B5U>eV%_d>tQ#ON2(SH<2}cuKc@r4%Z<dv``RcegJOXxk?-W(
z0Bxl22%bmGZQl&-MTzY3PE)hTtlFoA-MRGxotT<m+*Ab(!nV}=d?&bf(9k-)eIK@W
z6eYLJZ_EEON!^|7+HHS-bSco4Cgri$XwnpRSd&_38CV>rz8^hwhIF|M<kvlh-CmAC
z5pZT@^~F2ElhR@9^fZY}K)Wkh6xQN22`_|c2f}+dPr{*#vAd!Vg1q!R*`8l9SfQPt
zbxU@)=S&xh>m{UYVp)Y#Z)DQ1dZ`||D@mr$oCe+PuH(nGQ$>GxyQ2x9Nf>^(d)d;F
z<JP@}lx3<_)@DkCX`PO_4og8e5Wl>kL;QH%dgp(zXeHcgg2Fz{8OtKY8hqb9^IIv>
z#tn>(?0vM>me<xUEA$=hB-fYTj@GwBqAwXLC?&@DRU&vH!nN2{ZF|W2+ZD3^15afl
zEDB|AWOH<KWnq_U&?XC&VJH-r(cC7%0eZKS-6q-!1pkq)7MF48CKZ<gjyMySndm0L
ze;&ubzy5W{Gk`rb#wvz;7HDvm<N}`*!liK!&GQwWq4c^DEZ#Lju2dd@V5vjw!%<=H
zFh}_^r>>2|HC}4M3t?3cG%NOEt2XqO_HFxhm_hs6(rOwArZlRq634=92gobA1dH}*
zlsYUZ0C{zrp$?;_MBTIs;Tz^cUSB<re^Mt(1^V9~9ISDL;(~#J#W~1R`&SJ3V?802
zF$z@R2#tww1zs=_C%|^W=}kY<6MrskQ-pWHCq^*t1i?372g3GK#&d)FrDWdQew`K%
zEds~ee8T#HTHOA8T7*8t`xwOq80NL>A-OHZHz=E*1k9+XPEt=AV8Se^r?PC=f1?;n
z>zJ1Mdt!Vszu-npD!<@6Sm#cq7~tqUO1gwi)ESX~iq1GLjd&DH0rcxdYS?^<E&Y}B
zO`=A{RTaHbU^<4kkb+qu77!F|5~xbj5ug~8eAt=n6uH5E0W^=axF^ckIb}%J*09UG
zpdQMoL*~DUbZD{&@RaNg^Cp;_e{g&0-kx7~KQE3@A&WaHhfiO3p+iYv63-n?8AI+L
ze);F{dH11THXAk0IL=A`G#`F>8@iT{*)n^MH0Q=>Jv(*aco1qZ0#Hs=q4^VH<*G#q
z*b10g)67T96_JljkhuAZO9{%fQUOl-7r~#kd~EEtF=#JDzcTXMs+C8pe_|~3qve_;
zKPEpUU@9KcGx`*l!Y6PiWqiLQz$$z~Xyo<<<%4e%Y<c9(DbX$7pQ^ovXOguGr1Mg2
zq>xNoeS+GR&*D>MKWh2-vdMc=2K~0=y>+c|itnVBOls4wmUS%xB!~Jh!Z~Yc`HW-&
z?|}l1f0%7pmX7F)N;(U<f5G8}QkO%Zy=MkYfxzdLRi@08PCVXQ-)}0GNoEf(^G5Pf
z16LJ0^-$*pT}V3Y8mOIP2E7d8=zN}qFI5MNiMx`&Ca>S$T4uc8miR)Qt|vhePgo^O
z8X}S)kztM47NU-zCCv3uqBgUt4TU0MOg2fy=$xYx3PRb9N#Vf)f2qQ<8L=rTjfkME
z$VFts{i%dwS0k%Y*|`$<$?SSX5%&u$paxbQ1{+HH0$X^>5NdSRosl;*g#C)TIqBz&
zJ0vFDbT!|v#Iz91^O^aJ;q#Oj9N-$>8{=^W{qnqaIBEHFEL!!Y*8r46qN!z<tOK*9
z5RKJ{S4U?3JZ>((e?swykk8MeN)-5%VhFL8;c-dBuNMyh=9QjDhDAd!gt~az-%+4B
z`k~HjKs+O8f%e<$RCFB}7bTN0w{2TmKiLDyddvjDQO~#I$}N^0OA_gFu6#ckJq8?d
zP}ZY5^J@;-5{cr9`Vy8MBig%U?Rv&AjscwDi?htEn0tate}l9t11*=W?+&u%!eSgL
zb_I&C#XR>xS%os2yuBDDn?>}o9z>qX+lGj=(NE$?4nROFNkJzF05W^7^tW+~OPJIi
z!Y#FI$MH0(uwNdTa<`>NW#cK(F%==s!OKZvoki=%m+kzS#fU{hPmh+-qe0!IPwuDu
z!Bfg1)bUa3f7>Z%_<nO=Wc?t~Dlj2U54d(UkjN>HTaijB$$-0l$MXW#^KkNb7)q_}
z&?E+Lfs5vjp-x;-Pj4{gb$DsxnU|v9Z6p1sP-pwK?BA`B2cemgzT1cb%UL-=RJ&04
z^)yXV;Bppe1Uj=0y^anRO8Z&9C-ietFSR^Wzc!h6f9g4w%*-QHtAMc8_*QPR3d;M;
z;b_!WV;fUyk~dLB<JRTX+q7g_YsG4Yk{GVBkM5jO`<fb<L)4^xFJXr&Wx!y;XTmk^
zz?9j6?AFwR5YY=2JigApC|ZcXoUKTUj}oZPMygLWtd**AtGPdIneeb!-R7gSbR11a
z;!v`?e~lxP^TfQgR7QwBbRguPQzoXdy!Qv~2FJ9wAqHI9RutLbw6p>YjT1=;YvmTP
z<0#C&jUSH}p=diJ7&Pg;iHd`6Uym?F-qcDuo63Z7x0d_U+(ssVH!gsw#*arLa|#&B
z+*<Z@6#u)rM}*4MBvMFk<&h+D>FuF8*5|ghe|5L&%pzkI2B^Eo<4<Fc+2AEZ(ySE+
zFz-a;TMa=v^APa^C6da&L%9oVVMj>1A!2Ly4IkuUPs#t71&&iuunTU`oPqV*NqRc7
z`3c|7xM&$ubv+!bQc$hPdj8}*6M}lo*~O0UE!}{zF}d>{OE*$jrp+x&H~orNWUQ`;
zf7v7$X1M+xGFR==!52U7qi;FrBYpm`mf(6{f@~~tk50jTuJ2dh47m^aTXM~eGRH{u
zzHG$^`*RZ(=f?GM$$7$J5#rw;6#Bu;TN}y<Ez1LPEzI!!^Iv{Qb`L{=)hQSyXgi<s
zrynk5&P|cY%qqNM#@Rnm22VK2tnmM3e;K?GVcY*F@UBeYiSR)|&3F9|B=AuE;$7S-
zgLfN^e~G$J)rSj8BDY`ziN21P$BEXY4v!XUi;SK4#NZlug=4<?WaP|5>G2P^eh;|8
zqT$hJ)0w>3r_T4E>IX{S9%gi4dHu9t9p<%|=u~?951kypzJ3f&Gw$6Dz430vfA_^k
zTb_DUOIMVa!&g?IRz8tgJ$tzD?8Wykj9|>yM|{qqX;mTM`+2}|FGC!R^QCus_-zQy
z5JF#Lj2*(4oG5aEn7jL*-ah~M=^yCA4sXAP85l2Ov<!H*U(>7BIL!+=e97;Qp>9Ek
z7MMp8s=;8;DgX!2#M1C-*eQ$#e?VgNU0c6|R?n1E;k(K*IYy}#2bK+*CW0ZXSXybn
zFKr`1Rkw&Z6)?hESS8O5uQ;D(h1YPk?=%!={cG5L9(ErJJ-x&~yc^5S8aaU}u%R|n
zNlHyJsArGPe<p{w$PF8<Euy7Spl@#!5)KWN<$zCRw}LFSGze1@u0<<Qe-qvN_6-7b
zxF(Y13<}dB(`lGBfhi$VHBq)cfBKrIz2T&n=AM?hL%v{28+?$sqKyn5lMl!ot5tzC
zoJLdR`H;WWrBod(ic7Tfnuv9aA_yOb2E>HwUBUIk+2p~vQEPPueX2URJXPM%kcae-
zD>U4ytYmaCM3v%<4}{K2e{ccHw05vq_4Z{kR2--XT))$|Ox`KU44138HMx02;*7Ml
z(0<?5j9uf1L0hvO?-Z9CrIx%-Nhp`(ZthXJ+MrD5rZ&l{SUE52gd}f%YYccBo|bF^
zRmx3Q?xDRYZQg!4YM1F?;>66ibdvNNrtWL`ieZ-+R@FtVY2y~oe>|Rk*L29q{PH`O
z8l^6Oy^w19dF(`x#W9q8uV40)D?#NxaEspu%oFHM^1OWGPwql2XVBNGzDegpTQuoD
zW;B6c!p!pPEGlVCARi+?lq!;MPTKSZg;jTDO>0)Av!*qx3QTJA-|>IZclrJw8TL4f
z3T19&b98cLVQmU!mu=7{2$$ikFB6wx^Cl~QOK%*#?e1SOy|q9+(fa`yi`o-!fZh`1
z_MwL&XwpMc^!s|~-!CPR5~Z2hhrLeG2H31eBT*#9x42^ZHi_xO-P7m0cSmg|KL;zl
znm&J>B)kytzZ5X&?Bu1KotD$*AJZ=nLMR8HC-@j4w1$s|zXN>R!@ZUPW=MD=;f<7k
zhkI`%jKG7?3g+Qlb%bYkxHr-`lUDe&>TsXt%H{Bv&wsuXh~T@U4O5(rQo?Ov=e(H#
z2g0uHjejSou=_*V+$=&^3*Uf*0Ya<?q)kE^289I_O_~)l3g;N(teP`cWor#Et;0QF
zPxE-VS6aavV-Q}DoA7j)wUi^g#AsuG3ZCYP!!69TI^mDw0>9rE+N?cEfs%IiFf{B4
zt5n7pDY1!5$Fhrcm70u4deRRC!~Q9Akno~!g}A5p#|p{en?%V%F!CwH`s9)NK&#40
z6=5l-)xo6G?M@*>a!83PluD%3c)qsak)G81!htNjx>DFt&#*gn!X~KMsujO~laTgu
zSNLrdo;pAM3uV%XSy@z8sh<>s>WfMml`z--AYFVYtklxeQAR{(DU2x^UCCUEAmFGf
z^HJcWdmvPsAc$jy?s@^syA@@t`xE4m^|;@PRrSqU^|B2CR)nA?F&p><ZT0jUSe5nY
zrw_YF2PO_w#rKS50^S*uasY-vdB1cUe|Yv5ZTkpy@Us<a+)2JiTB1t;($|OK>G|{B
zuXm#)SnFmB-(P;*r58raSp(Hx&K`{UZ-1M<-hCXQB8FlXu%!HFMd4XD@(CSB6hMe@
z&e~2EC9u_+$w?26qe{DZ-z-WNFAxAF2iD|6SVIJsIBiFWdyJk<!jt~X=yxN4e}0AR
z+$^0MAuG}0RcC{<abB({yGj^f;HkLKv_?Iazrh_|i#rGj2+D|zl$)q4YvJTphjQML
z6m-RsK8PY{O#3gR*o|=Zk|+oVjH_N+fu=jLWn`V8c;SQwy#xar!R5(NF^J3xq2<V~
zV`&1Koy9^pWiDC7%i^rWAwjg*e@Eu@o|@*q2J2vEZ(}{Jh#1?ZLLDwH?}`{p{rdgA
zo-5BbPK9_gyeT*+47${kqC<@QTy{OL#>=TWm}(<61`;_~X&#BN!@wiUc8}eT5?0K%
ztRq<4(bgL8Awo^2_SXsKeVFoOSg&KS)4!)QER(Zw&UyIOM#`8dy?gi1e{Wy^^yRN|
zh6VphE5K4G>mS4PqdbD%MrTEwzT<NW7PZlnoFgEdbZb@+K*zeU4t$yJjPkS8;P2Y#
z_cRI-bucL+sf)9J@G!A%z=jBi1hJ0Ne&6{ZVb!+CG*vXxQ-XqP(<7{>Rp~K(_%ls~
zRezuEzD{=^D?g)#fAH-*e*{$<?3A;S$=flO(h$Vl!h>)jH=+lD*&gI%)<HG9;1Z8i
zm_1|}={Rt&N-B*oa%l))5d<2d?{R#G{DIQKkRp^|=p(Fipefn?OJr(cZ2SAm4_x;R
zR()XK1<t)xdN$Tg_f`QCKEujxA2PD75EQ4=Y|_oggSHRNZe?xkf8#XTc}v3vpn$^1
ziJ@3%y>G;RyqG~Ltrg({292(?F~)>AgITML*@?}G1_2Tq^l77o0k$_h>^7LGdW+FU
z4%+imVq*@(>wC)FvcuAp?{pA|{jSf1$UtV8zGZ8O4L9-&<`#|x@-@#}V`{F&^i(zl
zMs*HCj{JN<NWOPse<lS>YIR~Md$Q3yl;Q`Y6B8i+(D;d2Cei-LLW|1LlB=Q$^s+*d
zMUgH_0wqux`eWf~OR+4^A#)H@vSiAGaTc9Z;OSjS1G{9ZzCTmHfK)g~#RJZ7SijJq
zyGa6u9cc&3^m!sWgvA>shAH!wiD~Adn3#V|{WJ^(?prKye?e`{*yJT^T&!8K!wS#J
z8JTgRC|y#Z@|H>^`1E#Ju#9F75V8#TzCjk8^AzP*J7MJkDIE;sFdDd7OK)BEKjUaI
zN|oE8R59!mDupdo350M&C`)gw%i5GBnb;J64UuBdB={7ui{t6S5u|u-<i#n)vaeGt
z3vry1s~wT1f0IQ_l4#>0W>*zg;@SJcH)UD&;tS3RuU8rb86m3$#D?7=W2M>>>5)fc
z97wOun6?OT!wC0EccPq~LDdEzrE*GTHb2xl(H85$qeRBK(5*esz$}O_EhiaKI6Rb-
zAlMDj{*-c(fe<jtTb7eXh?eQqGqbzCX!m9*=jAmKe<SrAZP=)Zth3UcJiBDz>)Hd2
zN=<(m9&BQ$S6pSC!NhS~O?8hPa=4W<-9GknzE6Xr^igyAjgU2LYynQK&tSE<%p%{{
z))Y^~3j27$WAgRSaKS_C(61oO!LJOXgI_6d2MFj-5c-z2V;2)RYvorCQDV61yzb(H
zfx_l*e`;Mlg@dLU42fsH)ZJWrQ?Eow@|PC;GnyWopD8}4!ZKSNsr42*li2jdE`7L5
zvXCsKZsG23VlbLLy0RJWDVhP@sSHtI+vAU>^z80*k&;t^Sn}H%@JsUh{^foj(90HT
z;LF-FluM*!3(xo<ZNnT*7$K6bmFi*T08iBlf2Wo;4Sl*;Id0G{W8h7)MbcpE{n6!;
z_;R-;T?*UH=(c23v-0t$_$_Vptdzc?-x?s^@ET-*eC8Ttr0l<X4Kg5e<{D&}ecd$(
zKz~Lw&r8Tz@lYU=!roFmdWa>LXGh^!ErL~h&W1Rzr&Yz6%EVXDnrgcMUZ#4ez+JnM
ze^2N6G3=@|x`N${lpUc51FBZgcDdFrN0L}SUoX~hipIrpKWPU3Tm#Y0Uad}Hbxm~*
z#8eLtD~NVm_q3T6<UgYnh@0z_N;Mc4f3}?UyCyli%*AcXTL5T!Mavt?8=hsq-ym<R
zvm7`ruHRROTxu90$(Ku|QL{I4MTvbmf5hTq#-n)NaWOI9*$%T`dMqx?@Hs<Fo$X7b
z(G8BpyWID3!($1cNcyJ765_npc6nR{F4q9nLV@?x7cyK6H(R@6ljxfmC9`0~x(p*5
z<pBjDS?dBS9hUe}<-6VVt1EIk%;+(D#wU;00(#>@0s?7BC34_o74HS8QT<aNe?Td=
zBLz<;vf4^0Sxd@ZARC}WvZld_#Is36Y2mPe-DK4}s6)-fpRv>9h!i6T8B5P8tP)Z*
z9;LsxDLYUh;8NlgYss*VrSOJO%OFo<N9S8Zl~7NtWoVnHWz1|_ZRL`D+O>+s!sUYU
zX&l+Pw3K6r3ij86Kk=!+#T_&Vf4-BQH2waMe<+EoG22VjERu%tCBLI3r*~moxO5l3
zr0i0S$Q8=G8!C`;D+S?24bhWOF>PcQba*NO?<Dk)Sp{CNyY#d=tykULooizZEGk|d
zl-<543QE>AQ5$cmdesW#Q|m!(G*E*v<Ah}u$%+Rx=8pSyb0VWlu%OVae_NGNqF0&)
zvgKvpKi5^~SS{AIik#c5z41~$<LE{sIq{<YMTRiqC$zkq>M?j~H_})}_oceTm%N=W
zNr(b`3|n`ouGLo;@{HRau*#ib$#I_)p0R*wyvaK`v|59Gtmm;=so&7u^ZkaM*Y3PK
zm&m1)gdH!yFke=;0vGWGe{1GjI+If8pj!+{x2qoU$n`z=sPdt7jp~h3GA)T^PU#yx
z)E}2<UI;ByY8SxJBq`NKr+}%w8P(s8_5i+9z@RUw4ht?PC3YA@<x2JAEWA+tx+T5(
z-9+si9B$~EnS6hL)<vPN_FlfI!_q_QALH9}SSCW00V=V>684f_f5qB_2sxGTj&%ao
zfv&-_j&==}ZE<apB=~KlV(|adUuhz&HhyTE+!_7f?69<en7mDgr3A9IG#7SQ((Idd
zSTYp0+MgP+S9Mt0i&{|kJ1m2mr8O5eSV~C7g){9&ndeE}4eGrX=7O8%GK(UGwr209
zSqr^!(~Rrrw^L%Xf3>q8ZL8=mg~!GVyKUUSu-?FtyIJ~J5@$94RGI|CMjL0tTz0q3
z<(5$FG=GZvQ^0uW8WS1DT{9`0xopU8f6CF!QpB+{GXFl%_>G>KF?cGo=s1))aZT(J
zA=R0)<cs&e=Bwa*@vVoS$raeOM^@I6BkxVDIAA)4bB3PMf7<oQLQUm(r)h)I_;KnA
zyuIDJ4Wgz2oW;9@eNC&%g|8fbZE5_&ielbztXelx_k%!GB0R;&g6KeXl}&*_8^I!x
z(%p4+_aiUp+vrEYXYL3!Gc~nE@5oNOU8bKr+wH=x0bG<EP~!cDy6R;A1J~M-u%X3n
zSBo!cm~FOs5T{?(@|OJ4FFPiH{{c)Lke6{IB@_ZTIG54P1{Ad|CrbwlJF1R|LEpb*
zPrm+_VbCTN0Xdg(m@X8TtOqB-e~)tf{q>&*o|qju(*_yQ&zwnf1nrkA;aM^MigS)c
z2w#p;(LLJo50Sv9JmT3#nvBGl0{S)0^<$pl(UhQ%Itu8}=qjQTq9UQ-DXvGIa0ri|
z3hl>^u1=^ntuy7Hr^vl>?R>vlsfrtRVk58cDGH1^?PcbFJ`Z;<uZO?pf06{jgm%-n
zp9lX3N?_882_3T_r@#N4z8}7fs7WcDaat0;6rcWnn}(6E$;uojE8|)#)ma&EGCiEM
z0)|3lUE-%kAkR?@0y=B>aI&z7h@9+Ijfh!!n8byrBr5cL=C8nhH}bWyJFOXLF(M1T
zY^7wK+Q2wl&iJ4KQZ%Psf5J=R>3xyDD^KBQChjTE9#K7AFF&#;z!`6Y{D_*NrDS*D
zHyjYVkrq)Irj?u_%5-S39+lt*RFZ&Q)|gYw;S9}dM=fcM4z%Wwjy-=)@*sBVSKz)I
zX}Qn30?QPtQL;rIOULd#udS@0BV4?;-?`VnGIIyG=+L}#(Z%IJe|D<*F37;P&MLT~
zGABCEa`#wWRHT>maSAryAoMP|xNlu~tj--x(ur+(jc&2bam8h4%rHAdCokoh+%z&o
zTTbg^BPA%Px|Lb7vsuUpGqKP;YmYfRgqhDdAhaPOtgRAF(o?$Aj85Ul*3g44g{J4C
z`ST~E35Pi><&YTUe|BSt5k`=a%Q3_V<Cs%-2{8(RSt)mPi38$}A!a~+C&Uc1?-XJb
zkm^8PA<s)(`@1slsov`9jb%>E&@OZJD3W7Q)e&xR|B4eYN>}&4L}8HQcI~01Kc%~~
z_6DjOwKt$$p}kXbZjU7ZUVBmnYUapIjXQ#cgysrV<(N@!e?$ejC<j5oivF9);dWa@
zqZzj|Q(O_(p5~}9IKDfZV^4LfIreCmn`4Nvvc~Q$=GgN9r+v(EK*jUsILKjxIih5O
z%h4X!<h7?g3Sh$CJGc#0x7uTmcE%pBwl)wxMH>i!y4}zS3ALLco>p@oA(06w#ude7
zrBu9BM^qK%e+_w=u%to>7w~r#s*}vPnX!Rh-m$ZcvVxgfHykhel|tVlyB&|3w3g&z
z?|jq-svD2mfOcIT8RY#`9;k11{o{`KF0Ow-1#{N*k8-%(Hl;BP)lyQtB(6QpQ7K|+
z&r0Ng7$TSIkJ+lhY!DzKQ>H#bx~5i<YXjNPPfa=be|eDT4V0y18cO|4>LfuGrJ1V}
zzDko0L#!R$PFFGDWU4~2pn`X8CefwiL04jK);Cf2eG;cH$ekl~YoJh<+o^NTOznkS
z=*s7usLh-ROOC_X3v#why=nosQzwUP!QCi_%89<92K(|lt8!Ui<O+S1zb2aZc?y$h
zy>t|@fB!f6VOm3TbqH7zftaP89X|Z$?fXBz{XyYtdi$?m!6!;^6gWHmjJLivWI+i#
z{m283@@z?&HZVuW3@IGZdTnS*URy`LO$QW)LZq_o@DD%i7+T+>Xx4}Nobgi&m3Az%
z;!(?RCi<ft%)C`9VQ+;Mf9FxsT7OIH^D2KEfA-qk=c%;nU(@0Hbof%_>d5!VkPUFX
zBv#5Y!T3C`iZ%}chET4uM4TZjSs86NS_aAaSdH}fBP=sy%RU!L)T-RmNVDNQL|7T2
zu#|qs@f86KdyOJVeG;ZzY?lzL-aH}6HOjVLfBTshzR|MJ&2I#H03j(X>*+yIa1-S0
ze+etpUX@*JN2KX_b}6f;er(FNLV>v!&tqXeQeHZ211PcZWolTAn|)|@`{iO0kun8b
zAfPb;s)BKXp($cn>C)sK0HqD}`F89!pu`Ki4Jc*mHee9!do})u9AeUQ?z|Lw6-BpH
z#?=dvEp+*TXMS?$C85nDK$^eYZScu5e`=A<1Y+b)oHxCqe#e0;+wC&s0P8E~w95`t
z7c8xWG|!Zk*!PgKR=J`FwHg5-Ve_W@6<!WFK1drSd3=D6jV`(#S)S!{#IB!Z(<=?2
zbELzRMar(P<atav>dX@T8A78*o>L@VvtL18Q3#yvRFqoGDm{9v5XQ{u`Ho!4e}q=)
z8ZTituJ(<MU@aw^EPccvqo7sH8YrX5F4A4M3NPuC#dW;)6hq(97RT1udc<idhhE+S
zW%aS@uJrCqIwep2Sx9{5Y(81e*x5{02~INZs;`j=+cvL}%=(uLJ=sx~aa6SZ(S=o6
zhRMUN?x^%J>yEn4u5LS^O+PgHf0Ia^e5I(?=)6t%caboR3U@4CA$Enw7kE}qQM0$E
z&%NfG?X?}?v!@0%Fhl2U&|!L0VC}RZWof8jTy=E^HyMz#p);Nvxb(mwIQWcLyYv0q
z5C~Tukp;0xkNIG{8>%%RyuNcPkvWAU(uN8MWpE6Aa@`}&9#XD&T_x|9f1IF#&0R2!
z^vE|uyQKDXeT|8pZ-j9(&Np_u1Z7kxM%1<Tf}p&Y3~StPE|CZoYdp2|(~WesEo7X6
z(1@hd<KN@yDL)8GFU6Ono9FGCr@&cTy;~LEf$dn{r)LZ1#na&rA^IkLQ<5DX347o(
zMSgmeQVNCz-Ro92XuTKPe;t%A4WpB{m)3i<UwhmK&f$!v(oybL_<e0r)v(t)=!$+d
zioJBzqrI=I8v<X^RZr{esY<%)S8P32!(OwlxN}8U%@b6z&7HMJ4n|wKm-c$JUwhI9
z&UXr|J?boVxdt(L;sH_IjwpRTI0ID6CX2u%;7lbO%ymGzgKOa!Ns{=TjI&4k1<}7J
zu*G(=jSq-odm=F)Zg~MAC6@!@xfc+k{q_sCI?+L^+nwm*3$^Pt)l$7&s;HLYXsoCH
zJA99z{|95me$kg<&?XcCIhRpQ1``4~G?yU@0VbFB5GRuYpwyR)5hp%>x-8i(-8mxV
zY_GedEBlftlyxaXA(&1p%Pk;b_mfvT@BNn69$=A^vC2g!IT^gQ2BFyt>u^z7=OuL}
zQm7>j)5(~uztf`$$;$c7d(u(swNLd|(j3-iss((pLY-SFCa+0)&hq<_-JT~SO<hg3
zO73f;o8?_ELA7w1Z(fUkJS+<PaX;p!-HTsaNHp;IMB%B|<|QriU=?8I4f^@UEK<><
zoK@B<wUFh`K0VR5skREOH?5^#TNvvjk3UbNg<kr3FFpO<Ht?3^pZUB+<$~*dMOf4N
zlqdXC<+3drdY;V78Y_8nzVZI)s`pnJ?0q?694n2tF`!7pY`yM(k}vi=WApx&F_VpI
zPt{dF)>k=3vbES^KU~zmA>~<3O&8}ApsDUplZ9vXKojC(=nZqSI#0!TRDYjq5-q(g
ziDrs2#Oh-239;Ae5}etR5Afrw`1SVj@b7Hxq@#ie-CsW*(hG~AkevrKCrKT;pMQ7X
zAAXl)t6jCyJg0nrkS@MA>G@W-%cwbAwd1&)q)K4-Bz8rO3NZBHT6K_DS66F5B;TCc
z=M=5xjh7_F<}A%MI2jaO*(7KC$l_0W#$n@}0%s&;qZb_=cw2QI<Y=79!5tjV`rN06
zdsKQ$pA<kDvqGJ$G|$ulJ4k#o(@D>_+bW+jUAdVKdJ8Il8c8gvh?Z2xIo;_Le8eO0
zZM9KTKC&(fc3_2a))3@GVx$FHXl^1Hc}EM9AT6g?w9257+FS4J#u~*i^gz8>ry{40
z<McvIi0)JiqXE*hu+S;)s%kj*4<DKFq-UH5T8w$#X5EOj)-h%gs!O6C_JbAy$xZ1k
zZMN6*c>?@@1!`!`seP`fHS&@I%`Q$+q3*Z15-{b^q2J5X>eC|_jZ3d7zuql`4GOKD
z2|?F|pesTgkS&d4K;#bGeFVjg$q6iP%yuT9h{Ap}p~4vCXuV~p0~NKO@m&tCwl7L8
z>g|!m&XNvLmppaJ3zopit%IiZlZY@4!WR5iu-Qw0kJ8)#DhqZC%ZQK~`pdWt>U&+|
z*_Taggbks=`!>5^&Ow%|+#1xy`kYW3AaocV9dfT5V^k^S9yE!ZmwlL{Nwz7mO+N!o
zT5^22@CU}Tpm-pf6qIkENiq79(4=Kl;r%D0$tlyFXmU#I4m6pLb#|z=q#2224;R)0
zpQ#9c@P5(OhWBYtQA}t_&~A{xJ0|3%M)h{+6-aTDOWFYrLsT<A+k*`wDj#PAFO(!=
zP>i#rm+RGfM(q@lzu(dby#E-GoYz|z4Jz`|AQm<&5Dh1Dlq*@@{}GMSK`MI(u(l>u
z{$Ma0H$6mY-k|ib-I4DBcbry{`fBO_1Gp`J7+#NAf-wgmX{$knh)YQN0P3tR6>*N!
z-7f*)=%QlU2X5Gc;!XfAs9gfW6@bgtYzM+foSMkrW#cUXn3K=ghj0dQ{T&dufmk}r
z66RpPF$B%UPn>|{R@1g6nrrWkyN@azbA?7Dj7qHHMq?|S1y1T-f>}PDDubU>8Z|M0
zD&vnVjS7kff>}ZN2ACD2KMBkl?-jaF2D4M9JHhOf*uACEZP2k6ABFropd$ul`2aH|
zC=%+HLZu-1y1P&*2wuRBr3fi3k~MF#cdggB3yRx;8n5VItY;5cXRpU2HC~Ss$N5A#
zP8=s-Jx<)`67@K-<P#aBwp8Y^<dFP-k@-)0#%MgHE-;d0yOnNbrR(r~CeLaqIF}KF
zg}D?hZ{<W9E=!Vj36PpeMZGPDEl!3ZNNEi0iWQ8YoSnr!E-xpNnwfd~smHK(7}L*A
zJq^c<PdsjlZOVT~nu8fHMuzggA*wZu)Mr|4V|*dK@|2WIvF-#<^(zd`>3;8jd9F+|
z#Ow3oq|#cdtmVzL%DV3umd>z@d(-8QwZpySO@%huOTD>Cmj7DyIt%+v5?RZeAa0O|
zCKA~y(Sed7Dl&G1WNs{3LL2k+G#;^>Hg+CeYakh;h>?3+!>GnyWVL>DM3~q(9aIKu
zm)pGYt)6*0Ds;wr+g)y`>rQ`vIn+aK_^o)wh}WyMfV7X*{VMmD5ZFIBnTBgued#%t
z;o8-oTO7VuNMNYoLR_kItP#@vA|Z=q!Fh&ou*R|%V(U2T0=bMyZdg~!afgK$rtB8h
z5-SpKS@|Ae-AZBiA6UVlibw|*00hOI#W6wcMsZAx-f=rE`P!?u)0g;6-j9D<*+Npf
z4~9cRixy|r++G|6Fnl(5EA#9AqM@Gz{@pdh>lf8wRO1^ita1A1%3L!!k1ohZRlpIT
zJXii<8A8_Q+sDIyZdsT&m$8v26qiwSFcp`v(nKehFghp%m)#gAkbk`sdi%%u*9}j?
za^|#u^tb-6<9YM<rWYLzGT>(T^V`P10412Rl3oMT+z#LW8oq3P>&X^ooW6-a$cOJ=
zhi>F^vYz8)J+3w4$a;d4Nnz01G38%H>%j--NZJBv4P*uIRuCgCKb@?Yr3jqX0egyM
zCXSLQk<B1bj?_r;p@04a@kb*hexa;26I%9^B_GLEW*f4e5&dj6k%I=vPdI5Qj!CB5
zCcSA-FbtG=z@B6^L*#`Tgl81Ejz&J;Ve>IpNy?&&+z=~gz$A&iU^&a2oKW*t!a3`m
zwr`sI`0*O5qv!MR+Dhq|Gj^f@?`-q{MP(US(^C}iJVlU<5`V>n;>QAtnmG#Ro+8e!
zi$uX@mL-^91}A&m?AkKAZkWByHkuhTG3hhQZONt8s;0Ea=$cT}^O{)4W}=k)Jdl(@
zHPoNLdNk7Jr5y80Gaw(%Mocm9Gmfu7G%ZvmA3sRXXi_vD;@>`=e3YV&bCsW~a`bLP
zfay5R7bV=MPk(j#8&i<`XFq=Q1?_6^hnj~el3g3gzN%Hn^0m`z5~MjLNsqBUW=({P
z^{{?5_m*U(E6>1PQo&ndUwGT%(6&-2D%fP9-c=r1FRBPy4O${Y;YwvVMm*41!JCVu
z6_P`Y3jc^rri=5-%*7H3V5qMMgm(*s>j=aZK@fRGB7c3iM0yp8&`G92jU>VcZG@}Z
z2q~9o3s)r4cS|&9qdC-f3k)NSTs!0dpJI#$zb1#C%}^M0o)bre+RGT~XmVmIb8V9Z
zco(F)%sNzRIlx<vIg^dQ!W7&kCGKV@c<Uq}G|^u=!yEyoPX7e#E?G@+B$TE5ue>C9
z7$<9=b$@f}c@*YT=4IAE5A!WKw0jB2M=6=(up-FsD6R<dJIXgeemDC?AfMD_&gvI~
z{Eq8Fkl#_e1LV8KAivjOImquQt^o3TipxQMPw{XK<oA=01v52^!xIGLrzZe-0Kh$)
z3uxIXX53^pf7}BUmrQXr^?*fQpalAGnP{VCwSQ5*Zp+CEF9iRDpBb0trvU%(Ecm|#
z?`8C{*T8!)j=W+F?`M%1G+AvWlnS)37GlKO-DK_7VU30f9P>Wh<bAjsc%mH~11Y(l
z1D-&kNH1$5&OO%9)p(;DZ~bmp-MAvn^v+{du<y63TfmRk&|MZyQ5Hj24YSLLG9p!o
zxPSPNV$|jlkBun;rO4P9vaEiFzDvFsPXKa~84BN)!v(8&oSZnCMs{8#%@LUpG!!XK
zhr!dx6s_eqGdR4blHTITPBLaIoir*@p0Ziaeu3kNwr=q@kM_f~2MG=OU>&*3yN1Fw
zR{F_W`#qT|x&5|vtE|l)Nu+h<qn*ptHh<T4B*Ix{C>(o<W-wL=MDh+eCaG#rUh-{8
zTM7fNr~d2?MG%bE(}M^J1V@c}$sR=^Wo;=|+@o|9SKOm?lyB@&y4jcaC{JS$N|$7Y
zl6)nUR$P~JVx3nPEb&zFNfIb3KD=~iq-%0!&Pq8;o+>)PL)l%Zl<Q=dLoc9;g?~sP
zNiAtTq3#~<$mhg0qb0SRKNU@3uN+Yo>vV6T6fz%K+KASaj5NZSNynIaxs)83l)61&
z4VKZVB9wd>B?v2-^E^-YkJ?R28tJCDGSNzjNDE}rN>z=5hW`Fh4tAGn==V8N7Ywg-
ztwHD}l8>rGFEz^IuXou(zE>NRy?=pjYOM)%aIE3-4u&+8;laJQgVAwa*um(i-O<5V
zmSxHDnW{Jv_v^2U?`jyeVh!vlUaWynSVJ0y0{+9ABZX$z{5!z^qvlACv>Va))|{r^
zz+&FQ5hAI_o<C^45mr%Vu|ILWAu!KOa{&3H*PACa)PkAn<edtri*k211b>p?>2N72
z+-z*8J<^>1tGWJ#ge{a~7S>GI9mV<W;?o<Kj(V5QYJsKig`UW=uf3v2cvCnF<9ceb
z^?UhWPB<Y8XQnbdU9g0-%+}-D+<Xk{Wmh<Qb>o=EEx-ev2T*P%)<fqf{EK4h2d8dE
znV;MvQqFTrry_NDG52_Y=YN{k_P^q2+y8<|#>7mYXVb|tZ&#~V>5qkZAq=2)hG17S
z(OCZ0y*r)L>2)JJ<nHmjJfVZs_3hwTIEIdPk-hDWMOyPXrC)5zDZSML)waQ}JKxex
zN|x>6I5drPFMe!3<pm3hPrjXRlBfIey!i)3o~=*{Wo~41baG{3ZI`=|CkmG#b|Dm(
z{wgOne~;n|5QMcIJ3wv|_!i`_7-VJ-GmD+<VgG)LWEII~Q*O)BP7JivO_B9t@vBE$
zE?*Y8{Cs$LIed6Fuy_%y@_KpsyeRxZ;-`{0=<MQ^63(dQ<=gVxQA+LbdcX^$GzKq^
z_W{4{>CUKtGZfIJPIn#@j+l6I=16NSem|Cbf1U21X}vq$fz&D{3P@Ag^nLpM<(Gpb
zVWo&J-KMYc>FMS0{V<AdWfacB^6A?l{QyP@6ZC@Qg>h>6^N;28;p0fQGLyKbd|0{s
z`DGbKekJQgiPn?OS-Y~Hl4w#`oQuMwpNjS1Cn+sz3#c`q6(Ctb4O%jXeAgKB2r{l*
ze?a2i*bSFhxLAevxL?*6ZdRwU6}J!1K~3fNK}5ZwY>^rSqNoc+utl#lzTy_km_%>E
zDP^vwdCx@(0ZA$T$hxjZI$gDm!rM4Xq<%`*N~gNQ1%cPN!fa&0fk%?Zd1S22qN?pz
zBfagdN9Qt&vq2bi9X=oCEmoP!E0Iz3e<Q`Y{czy<bmycZjWTVOM=!UmMdyN%ud;(*
zUO>@-kZNNad1jxy7FO$V_DpB<bh&mh2s{xR?dGx=sfCedD$67JJnjd`i{&sHL1}A}
zr*|9K7$=Mg8^vI*7(#`OTmJ0$`=E0c>i8gl7hrrfnQc&!!N_jF=Uq^TT7wT2De4&J
z#!!I*wU11BHS+5#(8tiN@mQ$%e(2-lnGP`Ace0W-asY5?mfn|HEhip-Mj+D*Lfcu3
z^WLs0-XLfw=Y(>4q^#xij3^HX_iCIy4Hyr}rfB)6{`_2JPa81N3UE=Xuph4WY(8+!
z;VTwy)klFm{rPO@;&fr4wja7iV7;|XQ_|qv4SjT=c)NsX#NzApHW+Lm+*0o~MQfUW
zB*aQO?&`IV$s0Yi6CuQZ{v;NT>_3>Kbja5{iAOQMHjKP9LTsc_R$vHaVH_ECv)$FX
zQ|QyL)iBlgHY`zq*3!*tFz^`kLTDkLD=fS)fTXmK%B#E=n!X>~e&oo4Qw^jN?e!C$
zXC|TdN<*-Sm$pWH<4?+^`VDoyGi6ikNh=D|K#2gj5*#tjI<`Q6)kdn5KAj94n;YdZ
z3Tef~Dwp&W(^H-g#KX}V-$HNZ204^4H%n68m%}XcFOKL?YQdCk+vWUgU;tohnhsiW
zRAYGE$_gC785lS<a#frC&0%@?@U1khnnC4%27*aXH%JH_WF`IY`@KpD2>JBJ9Y71;
zh3~iMkvj%sQe%{VA3dYYW4p`MsG5xy8`mc&pg~XB=BL61HD@2<F|G|R?KfzrM8q8b
zZ|paKzBGy01O9>H4U@8g@=j7V%>F@1nTeMs5VuIm2Ci2oWdpT0BxMg*1#J_82(qOL
z4>vD840=$@3Vzh4te9TACo5tXZHem4Rdg9fW?N!Y=F*xCxS=#+^QFr-f*$eGz?Xq9
zClr66Xvv*|{G39}BTK@esD^FFA)$n2qF>Ui?kUe%@tRpzRUK19y3=A!VYQ8y*giFU
zFmW;fBKlS*SGm-2oj<o;MoXBA2WSq{dKU0=;9}WB`;~+DpjB8|cPC$*^UfBxEuA~l
zt8hfKOT4llH=ROBVpCG2y5{9I);6W}$ku;0?0>3+7~XNSpP;;cDiam+M2R?_4U_bc
z{h7__IbA6%fvJ0O-~70E|9qETIa7M-v0?wz=AHL6BizQrwGq;`6oGxpuRP6^tEQNC
zg$k22_N;EQp+WXSozL>sT1bWApaoq<r#?wnLwX#WS0qpG@VYAvNK4cDdlm7{6a#<z
z@gR8|_e(mzS|MTTC}!%GIjm4(SB|SKsv`zi-#Wru1Patn?|E;LYkC55zm#_0%GbSR
zj{N!6N%hsK&%w%S^T<NB%c$|(%iPX@KCb!Kj#21a1tj;?R`*fpu5W#&Bc+~JG04BJ
z59e5eA90VLqek30x$JMT$G+C{9%6r!U!9%l>!5ZL685Hf*B}99OXbC3tu&t2P!94<
z9i7)(C)!xyT0RC`^wvm=9<4g>Eyb%pPD*t*&d4XDl(T)ot2*jo$U5v$?w7)P@b}%~
zc~g4(W6SEmPkFq#Nx*$SF6issb889Nw*MQ0tAJVj3f&WJ4csazccr~S@}z%_mo-{t
ztM}I9{7hRvY5fpAGpvn6=9Xp9Z_LJs=B~{O!eQM<|FAJ_?}Wy=Qg*Gmm8nsjUh~Zs
z4=oF48{S6#O3#)`fYxZj|HrN(Xx=p4ap{B$Y8HLyB-OZ5&RYA0^j5h&c2DxlbsO=$
zO(F$OST**DAc}xl>yKy=sepgb<8Ij^8YtedMKn;}X%P*xe^84^IU_vW@*UGaksBzN
z*bn_l!A3SMFYbzrZ}~@x?`(raD<W*D{U6>SDXWDKH3v;^hq!9HNg!7`nai)@%HBJe
zf<Cd~D`#Qz=;x2!{}1>YIVW%67?HcC_n8*B9eNF1Bh(C`MvO75D0zP&&AQ+4B-HMs
zwD0*PX@TR%I=0u!FXOIlF9P`W&bikDJMoY*0z~6+^Bf4H^Zy5aUvS5NUh=l1Czu%Q
z^1mew;~Jdn^0_deFtJ9zJAC-hm(TzB^iMn`%a{MA6$s@*TNT6djUJ7N%0#t%rDumZ
zrPk;%B1(fO1py(>;sk$QSO-2W2f$=r85_FMuW2+22%vShI0q1)CNhI=h-F9=+bHq-
z+7DK#DvK<16(c<*Dzr8|;(A_{9@Eu7^Hf~*ujTN0Iee`4j0XQQ*hn`srxKvNQN2ld
zNIQXtOaYrCdffLjqrIYSDkcREp$Bvj9&sQ+9dWD5@<^vJ4<Uaorocm0v|;>=0R~-*
zNeE*_+EN1|NSnM;ky(pr+x@3+v<|D8jH>)y=-er71X#D+MGONvCQ>;uuwe>iG@B-K
z^t?YAdu%7Gp58v=Dts9$`xCaNi2{6FI{AvCcTLleZ?=w7!j}K(-LcLXbDCAW#X>n$
ztDFINnoX1IJMLpKi-bq2)H?j~XYeb$1HUR2u0$UD7oXf?ARKMzT`-jDGp-7(X_2X5
zw(q(|=kS%Oa^9cgS&JlBYqzcL=C6LM47Gc3ilxuGisKV6Y#ToF!sb!M3;&%zYg$0x
ze*t%gcd(aXk{lBPHaVA3H8CWYuof~4mlKZ#D3>oJD+qtCzk(l~0_AczoFQ2R2AsG`
z12nB+xj<V-2xRGOtC4hqw2IyTzHhj<`r*5?q|;id1H_u$<&tm7nKy69<H9buq$Rl&
zUCKcn<pP6(i*{3Ma;(`5agx{!b(yLe*_AA26h|R7t2&A-Tj(h`N}=^hj)cXm=t$MN
z4|pS^F_(YrsBmj0uyV4oi;DpOQjX*zk=j#+m0P>yA_p6Sa4&H~i+j?ZlDKGJ0ZZ)*
zS}*vPik`WW<3MPp_*nGO2FtLjy})=H8x(+L=7J%jS;eqNbu46qMZBS6?P779i@h81
zDK*+)j429u%9yVd+N+Fl2yekc3sTHoPPk+y3?YBGm$*Ge??PmjO>eu`J|Nq@gnw$V
z{s3M=`);g<P_`y1IpH?lsjZWNO4W(vu!YY>I}Pm(g=Ir#MY{n!I*?NBUBSI-OPWQU
zQZS^FoyO)IQXD|p8&VxQv8Px~)wbXooMtto_(w~%$<^qv==L3OA*@Inx0?H~S;^&S
zlaqhZDZaU3!Bc79p$v<yzYI$?VGJxHT3CN@(p=#l3@6G}3<_|=rR*Ubw@|xRsw+|L
zgGzwaKoxQ&MH{R{S7d0Xt^o5^f}fQfZTw>1-4K`JN-<2qjMVx-v(hFHT~@O7%D`V^
zY!<_^pvj1gYJDbqzhcVp5%9t)Lp$74%n^T{23HL%EIcg@06wJ_%n|M-#25ZEQ_ZaA
zM+hc3bFrIjW^T_d6glYDx8HWpUn8&u@%^&<=fD1q``H)F1su>@$emmq9qm4Sik7bX
zettAR+d19apE)Chm#%xU*Y~rt6BmceKY#3h*y;EBnWN#x_WY#B7~5>vFyS`JHavfX
z2%pw2Fwcbe94>&E1mJM-^lA6vY`(uU>tA;-o^N-rW*_?jNB8Q#r!$QFZU#Un{p_Sa
zw}8*Bi<h(W`Ni4(?0nz~yY<uT;BfD|`N!7*mx>9&jPBz7GrVc71|ZK)PUfv63d8VQ
zuN$q|XEa({7&;nxG|Fhy!7x%=7&LzxZGLty!MDL&_uc&LV0PAUh28Ei-H+Y(xWI($
z;yd>H*Hp+C@Q>JEgm)~_SFAtc(ATuP)$Lro>8-!N9-jQ#J-h6EzdxLxbUWQYUjAtR
zKY8Exr{{m_x`X-txj&wt&wlm$^W*La3w7Ri`+KK{-Ts@me|$SXJKpR4`~H9U=o@_h
zwaLCK-kd!}<Bs4HNHH>bMl^z6e2_9J;G@%r<BdUlviVHG11&s0z)?kjw$lC4{^#QO
z)E~@#>wenn-~auz-){K24d+tHH<peOXprj?<`zT;{`N7{`=gtk_z#D_9-dk>`1#qp
z&aS$@9sTn2lVKu%{$@j@qnCe%=qX#a6>mvc2uM@BKhh5$KKOSRXY)6QZPnof7srUz
zz1F;X>CR?vkM~aS=ff`#_W>609T_S5mb3$Twe(zrXS%1)0{khQbnf@gFZ}G_qI=t)
zcIOADdn3;NW4{XKjWJeZrr~0NIl*s%QKE-F057m_T0!9}s=M7YhwguOx*z7R=G}Am
z1SyaU8dl{L6w)nbpm-pNMB*7jVFY186fNzo0YOMp0huOHg10F21wR}FQ}of~SMxOp
zEvzCVS~vzzOAr_-xM<WEVgwM0d<6k%&=ldA2En$bE3dNOrwztD$%1JR`Vx8aS~mP7
zPhQKQ_sx@3m+~a#P4a(aq{}>Mx_go6775c-_rkcmBcq`!lHq+LAR+=FHx=xdph6*P
z;dIcO96g7!yz)_O)dg)S)l^VcOyPXVqhz=Qk%FbWvK|ED5H$r<cvR@Lf~j~o2snkW
z!NRY4f*OwSOH?_?V#sEcmcvjKH4T^!l%SPRH${Co3L7;U=QDpshZKuYD8-ZjvR07#
zNGK?iakh&ww}pwqUV$nC6_)*xP%x=ygzJcc^Dq}{QG}Gh&hSi5s2N1~8VX|_D%V&?
z!8*E%n{KokxJfD;6n}4T$$@<ZiGM9>pW_mXaoB{#Sgu$sjtmubH1cSa(Wnn6FM=b4
zY=s{r9!g`lowt8t1+?VBAxjQYlAxH2;0tqA6ai5t@KSiElyV7zD?$D8S`b8%6KeAs
zppI0KN2g#fhLUOfDn!1b&}C?FaI;I1fR0zh3fQwMs$NZ@GAiU0Cy+8lz1l{_0)=37
z@WCUaz{-NbEscbx_zCP~8wVHyEOsxiuqbD`Ma*1la7lj{C@>?I<Mx(RK=`V^o9yYf
z6P~0YQgM;&B5y(^sVq@R(oLvDt~L&AlWyP(9ybNjxG9jvO@S0gW7JA%)JjRFU>TH3
zZQBaPZK_B|?1kG#k*B&W49;h*qWJ@&KTHr(8HtLJ1i@`cYAvJ~Qynvk#JD1)q96fR
zwmmngD9nE=O;V#EfU|8BtB2ZmpfLD11c9aKEGq(A3j#WZK!_|<EHa5iV^C~XhgtUm
zsx@(B+nO;z#Ksj&3)dnl$#x<kT3LA#RT6wf?2I6=D3}6T3ql1tqf$k^R<c%9Y!gfb
zG%6ec;XTx^l*JSg5?})BKT_T^mq4WezA<h`xr%>qu&q%ls3gk@$E>*^?H3@>s+g2A
zg<~QpEyIi`lqyC~o1rj@PWdhx@kXl!ktg<%zMlG3b&0MuZ9>=i8ePk%{A5&qGAchA
znOsID*YS>yl&&MCYn-68%_hX41|teMJE(wbw1DM;LN7m-gC_8_9fFN0s@1l-7Dden
z(u#jt3af9n#tC+J#Hvb*(FF+GN3QWRQ^AUL7i78=*!H@hs;8;2lbvaj)o(~15h(2V
zGpdL5TS5U0wPr!4gR5k$jg$%2qk^W)3l<<QciYy0SY&gL))Qp&fl{iSSwbFwG(^0e
zOC?GW%=t@?E><70<CHx!wFR&vWfq%@<x+noS{g4JI21yGA;{uWcmafr9jjI78Jhu=
zPgn;s;x6)7B?9yd*LaBnoq`+ES`?p~3Cjm6x6}x_d@#rm5Fqt0kT2G44tycupO;Sz
z^F}9zxn5-w9kqPEG}6O!0)qhUL|PeyXpvBgPqjKtw%ND5?uC#iYpHe6PKhUpfXaWq
zL_2jrwT!w3M2iqWm8=Rd^hh93)o6gPnv#Ts1`TSA4vLyc*76YK{>?45tap)1d3n3v
zegpG+ZIVsr`it}a>#kpT6W1@-T)&KRsEl%`Ody9?LOa_6h511Cvzg?ryk;aqxc>?v
z5|aIVs4~<5m{%sC-AK{fPcv*sw+MgRuZWO{YTMCGW%;NJ&r(e}a>%G*Kn0U*v7=a0
zOh9IQcXdVy8j>rJ?<=;bJ~IvG*$cm0PyE*!taklRTM0>H1!Du@e3zC<J?hIlj2vwX
zBTV&MlpC>jBMKqk$$p>JE(l4!F1E_I+rdg$JONomE1B9w%fdsofG5UCvUq=|?K|+|
z36c$ZTMTMtSB>=fD(Sv4U6*6J-Xo^37NDkExbgqHK&;CLoOP1}&NBRHXW+IvQd*sW
z(kLQ?Eqe=l`=KFpy^pL;H6pa-ap(l8!za{IY7#QnCaMDxtAcMemG!BVJl*7yT5jQz
zT2^q0oj&BR<-DncrQj-Ili(vNA>BcE+-x<>ShC-#B>h^#u!OoiV|NL|LW|!O{114Y
zHa3^BktY?G;lL#mmzFdq!2;k<w^=nO)dmEc!IW~Bv5_Yfmy!Q26PH;$CoO*?IS#)2
zSJWSfS`_sJ0%JPvB*5Ofk=tSqgT{8xp?k42d)U9fq$H|Ts&f2@CzuW3*k(x-9}@YI
zvYx(AdV1bHyzD+ap`C(?4uY9pzD+<6H2nunMDL~mpgh9#@^kw8z1GIl>w#WYYeael
z`W)%q?e7pGjR0F;u)hlyh=PA<@A1CQarAyDpACP0s`LH+&T0dRIS@~vXOQ;)c=_8-
zS7ArVe6Whv7s(K`c7|Rw+eVQle<OLV<<CqFOod9(JeJ<QjL%doFH5qyL(j>XHZGR2
z)-(SF^{Mg{-XJSyE+nB%E9siN%r9#J^(n7$s1h?bX8Y9UH=Ubi3n_oE;eSNZRBhwR
z_&m>b+7e&*RQ#oqXP5EY^YHEQW%v8ak<LXGwVl5H-0=g8pfDPGY?aq>`s1I|x83KF
zZQT@FX_{01IX(UHeHtiV>Ml>3cDeT6xy3FAO<Uhi-We5~k-cO;kw)Uns;z;=Zv5~R
zVUwfPbY9%Zaolm5yyAc8g^$%{mgSXyS&aqdOSk_6QNWKYfZaYVbV_`Hm4+mOKGd}x
zPV>LK_JY!{E%vgYLPFujYO!XHZDp3b1s_X1$Sb;qjORG-q~{>17~CyAx8%q>JMZvN
zb-SS_V^9`XIvn$4qDUO$N@JZPLRmYo%b@<tYcD9>+J;V-bufQ&TY2(f<+aAaB$dKS
zp0=yr?pv3FN-o~ew*yD_aP<jr(W{hF^BLzW{dxxQx@5yT%1g8^=fpkMXRkzA&YpQL
zjic!;;<od2i5^aYPLm5>QS23iEL{I~=H4kCVUzn9B3Ly&thg2NDTRy)1ViH8P5I;Z
ziN-xoc`(h#6y1OH-zk%61pV=e@Rs5sJ#;>L)?@eK-`~Ie^7U6xH0?h;l#-A#4pN+c
zmapKFBB_0xepG^|0O*s(2^9VJmMBPp*3b}{J5#<+J8J?-;9^hy!DM0_tug)loB%@}
zDywv~CJJ+HWLfk_PZZhaRkChP6#m9s>RkRx^JNx)&ANYf-(@Jx`fb{Mn|7b8b)EM<
z(2y7noijRZAzHg-U7+L@2vBww77|caamg4bmPvv3$1!rRaU&WEk=~NMOjLm)Df_3c
zH#?wGI?aVdyo5s&s=kl(BbFGnB}&_^j<n!6+Ij%ITcH}b7<vBu^=F;<ou)l^?}gDF
z7^K+kr#pYgezlob200YjwxE0+E~5o7hYI~LpE_A(>zoY?Xt~c*Cr(l9uG8%E`Q{+s
zlHX2SV@yA@6)0+wZ(Os&k%C5)(4IZl3YxS>cQ0X>;BK%r>1T?r;5j&nUkr=G9O!vu
z2IMgfV8MkBT7PUjdPrU<k_g=9_yyc(Q~~Gt+p2#UDPvwuCT*^-qhKK+kA2*uy_d21
zlkKIFGJ@0ILjKgP_ta)!m+HhaX0h%<rwu!WuD~n2%f4Ou*Csldlrbdd$XeBY7P5cc
z-axgSvlUhx3f+iB1CPe8s48>WhB7|eGBmE8%$2gSV{&1f`C{0EV=-K*ylPF?mNkG6
z1>b*ET(jELq^Lp)jDym+y`>7L?c;$I9uuaO`8F!$&J2g9DAsYl9)fYzZTYQ66<2GH
zh<j6;Ei^6|%V_?T9hqda2;Wz&;3z7E;I=jqZGeIm4doC{q?`~V3KUvVzdRq+x^2~H
z6wU$J;K@U~#JWcUonLXjxH1<34JGz~tulYtq^BmY5#Z<g;?FG1ji*#Yjv%x{py(DQ
zfTOrioCc2T6M97%2f4H!I2(PWj3Xm@7)Rqqq~Ow0la}!K_L55MiL_(%B_$O@xl%m>
zFO*kZP`;dD^W~}+ZuHr5R?z_+x#a?U%X&`|4ssFuBq8Aq;pNhufjSWqUrB8ARvUlB
zRaee_70N4qMIg?}uKc<k_D-D<L@K2G%3s$sie$ghd$g@gW}D6pp);zzQ)o`GUzhl@
z|Kr8Mx^ES@6h_WQaD?ALQYCR8!*R!2&f85DPnGj(b@pdNsD+<yl!|2JNn3lK1HR5E
zZSWp_WjxYLvVr?MfW3+P8-y>g|6hO8a#*d>@8&XHij~Ng_PBNLfX*nG`zzA<=nb5|
zr>eS*0#I!{2mF@t6~ojy>%)HPb_9{|i^f^Xy|wozBrQ&|=XKUw-#*RYv=k;AoYtvK
zp2SU5b_*<HJQ7mE(GsWb3pfxI>mLuaCqe*XRLR6~<!f&tOmfPQ+QTh`;rM^*-sKVk
z!BBp?WxI3W_%^4EnRXBNi)4Fo2ih7HEQ~^4(t-9-al>rnaQ(doYK~>;taj){=}I;k
zNsR3hhA%?P6<gT6tXtLoB@`U`3(MsdV3bO`uWx<dhHkDpdZ2+&mi2?EuD@M<IJ>Ge
zj=gS1W*@ui`Kbjmx6r-UAk2Rya?7Q$1iGt_I~|cgDQLWO^i&1=cRB9H(nGxE&DFs1
zuX)@DA$f%?l&y(tOOPYSKXSxgsojqMRR<JMg|3lu*wglvoI7XlzG>EO%d^<+Hv@Sb
zZEldKCb!h&u*ckZ?T<861N1Gly75~p?aaIQ_sIK+v7sFe_8xiPC#8R599QPHnl&ih
zFVjKob<7gG%USDNoy%a{f^Ap?uHx4@cx{B?k8{J*K3*TFfNDr*R=9-B@2ReOk3a>9
z(<|ICPz@Y!(Sz|+H+eI1KGuV&KZ+Pe|Jgklv1i~46W^-`BQ{^IdTkFTFz8-!JKclf
zs^1<i_Ny^Y1+%i*s)K)W0Y|jT5EdCY4nMlMjAK+bh84$8*K@47a^!d<4_f{yf?w?`
z-P>Pstf0Z0bhnH`z5zesoAWUO@(F>JZ(jR((GhHb)~aY%K6yiF!1wMR?f?A`KEOtb
z9qz2(WNdZmu2>O+3gEB)703-DE??`+&7@mouqDU8Ce$8;yyAZtnc!6jVa4e!5Ia)N
zodGr}>+{XI+20v}3d(xGY`}DYxqRcMR~=o!Ua2JuglqKSp1I7Bd*;gbj32gg6g34P
zhF;Ph^mS7DtNPDJfp)Zt4$;DD?}4)O^^{$%%NQGDtgVZzzL+`AC0sEV4n@5EI5*H`
z2Dh`TDO>ZS^fpWfb9TtRKAD>(uih{7WgDe|rS&#|N)A4s3R2H999ke+{UV+sfOcKB
zlY6E9&Ly61AzREpz+AG+);TPHpUoYxG4RjL`u_po_L^Rov5_YfmqGR{6ah1rfgDCB
zf7M$_k7FkgzW1+qf1nu(C>F`GWV_oV<(8;#QBF}NryMf>|B!+QJeIe*JtO7drJtcF
zYW=F9`SAC^hp#W!kC!iZsRqL=f)T^Vw*lw}PVd0!p~k@gFfC#D_-FY2#JSM)y3$K=
zE+xGTeYf<9=T`}qMu3VhIKLVN^hCO|f0m~>2kG-Rei#1xJ<iwXSH%T5Pl0Fx{RWrw
zA0NNG@W?D@PW!L^qy6js<K^F%mUTqQly$?O|GfAglmsTN7(i%-LJ!~n8os^!+LFzS
zahfx}As@c~Jv2{#i_&wPq{p>JtfU8=M5+d@9h3g2QF`i)bEMrW+O;AnC@%%5e|Guh
zZef<9;-m*`DUz5t3Zg_ZQ-ShGjVaG(dJF7VPl)|YS!)Jbw3G#}_{yV=V%k7ET6TQU
z1F{oNyM*`1)3!)A<*5u)(mYe1HAS4+hkSorrZshVTaYQu2i@^P@>Bfy+Ey|!uBF!K
zlXh=tD#!u60s_VANlDRB46wGNe@IG4H%c9G(tyIK5j(qEAsln2qvow;PWt?Qe$_n0
zk$1nwVR;CnnK5Ql9tS8|Wc2E^t3E-|vEIBJb2&uNNoHJ65$9FnEaR#(8Pa7T{-9Q`
zY8{nBib9Cb_@wRwTuO@267{BPj2sO!yakHYlaiu_5c^x>!f0H#$ai73e|%mh(jn_4
z!jt)-Upl=7j;bf6^Q6!vHB)5vSrhc)lbQQZjDLHnqRD3O;wbH(jt=7a#`pN!qWg{M
zHq51QG`+W|Bl|!GoFpQUA-F1F&PdfetA_!Ke||+!jErQ@f;h9_=d%tv!L*W3%sLIl
zCug080Bk^$zxqbjX?ZEL&T%tG>wj~F`rn&5vA~SLpJ?WYjv4*9ozKD8FL$(qV@e3F
z=g>WZCD)pQQpWD|vgcQ?Ij@?ek|_QV^)CeYl?VKF4jA);&l>SpJ#HiXiqXuI<LG@f
z!Y_V=$xlj7Ov#m>8w9ge*eFM^KxT3KqJ)fbbtXtgbwe|eUp6Kod6{izp?|M;v_3_9
z7g66a4j|-TUGjwVBy7`*=#rn<DJch!$+Sm5%CzhmO!RP+)A*o@=#MJU>3D<4Nm}XF
z@=5%O01}d)t8w+xT$IL+ClO<*uOFn$)HeLx^Wf`CtGzMMJf%+43?EbCTHe@PDGhZk
z_vfXMI1gjXHF%;znDZ4*t$$o2C9nI=w5(#r=9(Fk2Pb*o<7%pi_`$SIrOoUd<1SXM
z%tu`z+m$1<S)N5WR`&dEWBMhE0yJ}QkPRWoudtgnnobrAFO2G1B$C3=g7K<Q1Q((U
zN=~N8<L6srqS44h%t1UQI)(XPGAEZfCvcKiDvT?}s6ZR!H92hIB7dc;Nu^Unv(Y^+
z(I{+^!2Ajg`cC|%aVkf&&OnV|R^(u$lyBacO7)}3&!|6+3=+_(=~KCXl26PUY474x
zm^x%rhsg;rH<7A{54Lx?2(WrvLhH-eW`hd}Y}F`zMJ8-ZsT29gToHk%?Uk`Rj*~SH
z*rYCid37Nn1ej9QwSP6JbuSlGcv#a?*~(jj1gW~UN0pmwG&#lFDTC!We9I$Z9PU(^
z)V<49V%%RqvlK)C-JpbF22v2YE);hl1v9kadUThKTBn$DL7$N(xm21-Pj%xYYI2Uo
zA#`V>oWi5E!sC5<S(y-J_w=`~%Z{NXRUWe2@v$DxlC?-qm4AmZEp>o(U2#jFX0NWQ
z(a+J9PN61SB36Fgz|tZdiMeo<l0^c%;qM8rtjR)J@;5Gb%Ay|4<t(2jNGW2M4M{CY
zDeNvK6IaQm_u?tujVti=P;IGL>9(?Mkb7E%c+I_<MU(Xf9)X``eGe;@2*(xfELEjy
zvu+Z5K<iuMPJdLM%>`eDgoNZEsi3#1I=kHUE?)#xOuKq+IH%c>KEVJj+)aKUeRAJa
zFb$!F?3uHJnYF8p_>`=#1|%tCQ~-kNj<wxO!I|zv7kY=+O%yAoqKC~~fMfbbVEL@=
zsJ_d&T&O7mwB4$ma$WUZs%pD6@wx%?1=R-4CJVJ4NPm+x!YxC0(G$phfCXh!$`aBZ
z+k*tjl!kUvENPbvxKCAk;>A(hk@P9ugT_-$E~CY4*jtYwPBTQ&(uB8d3!w~;yqj5@
z$YZ6(cyKZ0>-_EgEHNR?TUFe^a;BR(bxpst#o-bKSCp7`Q>Ka|<DHr|#HqMr7EGsz
z4Il!Faevqg%IeYKSbrS|wPER)PyBdL;vC~tay9!RNfWXV<JnaD_8FwIflVBNb#7ZO
zwy=rXE&Y8k`7mA)PB7urV=<_nd<3yrt-INeK^Du~P4t*Wm=d#y&g{f4%pxkxGSVzD
zi`b1>C=+SZa?NoEV4<o}h<0W$GW13!FGZ^jGJj8VJq$aVcW9qUUP>hIaAv<{l`p5h
zu_`+$Vy5M`8YvCoFQ@0^;ZIV=I|T7)dtE;0yDzen50+ACJ23M+AwfeF&{9(CK_IyW
zl}U0j8TRq6JQcBL&@p<&@t=a)hcSuJiW&W20aH)r@^+@AlVV=oSP_#cR_*-z{7!2?
z6Mv(h?PX+F(cED`z1S?Y9YC4U>&l7OQLQM)z&)y!YI5j26wr6JzI#Ti_($WnO;{v^
zVani6SX7VRoy$nJ)<8RZ=f#^v#W_WFS5LK)+!)sqP#Sx(%U!biL@NTTWb|eMkgdZm
zLlIJSayy>IDbX<Ljl8DRTjm~DKNt=>n}4AU;>6xwbUk`4-ZjeL1@f~~SrJN$+kvua
zp~F?U-Q#7E%dv75%t!)rq_jp9^Veg>l2W}-Caies|4k@;vCvw-9RQSd@zX~x4V9fo
zF3qgwXkltq=j=Y|E5j(e+$+gY^Ey=qVUB4n%7+Mbnx$6Xw<v8ht`5afNDAjERDYzZ
z;^>8L0tAwRM0=l`t57WcjQiYxmI<%5r`-f=C_eE%cSCvOK6f+vC*9`;shLFeOnKH6
zgL8;A`1J7o>Sm=U+^=pae(rvCLv8!e<VRfa<-QkaHf43-f-j&~ez&{2N#9-O6~311
zCOv+e7X&itu<67-^U^7BNdV@V|6c0QP+SZ($E87aErKo29%75EY4XLJo(l^<ba!%O
zzyg_Bu91Z+P7mU7dI|tZegRhA@_AuwYRp{%i#VccL45Lx>BzJpwd8oag*4w1tKEMY
z%%;2MlhWybo?`M_m$8v26ah4s;lL#mmzvTrD3`5ICz%2Ew?|MX0s((HJ7K1oPP#TT
zTiyUtJcXOjWg%Ud#|(KY9=<qaOqmc1N`aK~({W4}=5cZuMfXIYO?4WSb#B~YF1dz!
za!hxpa6o0rbaYB1G1(den`gGTc?xIgSELYoX#}1c_jg5ygL2mPL}w|nO>BL54`Q<J
zcs*+5mDjcK#I56wn)`pFhgNytGvOPFylP&OiJuBG6crsmlKzXCIWpENu-HnMoSo}d
zS37wg>`Is7wV3Tmm*?U!J2jhhPk&fHL{MZyyCdpN{Sbo+&Y#oo9(f3?g1KWFQXr)P
zI|foCaUDokgF)UX&9x{YRn}Crj$Fc++Dy*ud;_x}hf0Ua@UVZdiBx86L97J9h+tv@
z$I3R$)V$KlQwd*PTh{aRTx1PRs$8YD1qG%S#(6^>eLUA?YR5pg@6B3)bN_KpOeQD<
z|3Bu$gkB+sUxpK3VIeqGE2U{HU|Vgw)I>5F9Q96~ulBl8>pHX1CMUGs*SNaZbQ3uJ
z_U%{G8;PCCK_h?t0y$`P`6xN~cw`w*itwIY*jSX6J;`t*vA4&T(l2?e(dGVhuGtaw
zcx}@<;}v;#PR~2gI>VrtYFA8^k&DlL|L79hYO&{!g+}5HrvnDOvhJ5V9oRZWd_1uq
zf>a8OF|<BhMdua!YU^}#Dp%M)$e;`NsF^Wi4qBL!A^(50%;A(P>lHKtHO?&Z|CkMo
zthC&HlMz;_BdjwwaOPf@&Xtj^a<h|B8e_dU>o}e#C-!@o*>#whgaMaqt7|W5omM3Y
zlo>#osS?%}RoQ-G^ov|a_R@Q#?#SiMnD49EuQi8On4o6`WLr!oJxkR&72Td^6%jXe
zUS{sLCyamWCS0xERiC=qW#m_eXm`o3q_@de2fN}qe_-6OtH#SxUI$BwNq#NB%F^S^
zdt(7PY!;BS)&%)|iN2LQ_ScWT69GnLG_>zTf>D;jrB@=C`%Zc#@^<S%!>ViNX#`%K
zG|hQnNF^q<fKr!kI0i%{Wu~`iRO(bx&fcZNqeg#WT0M`NTZva^@e=6#+<rU?86Uj?
zYzRX6rbD4SQz@`P^l%-4ZY921MAlKex_<+bEd0dzcOkjWf8HhmAZZb%w-=ab$UWM6
zimW1GJer6H4%JgM5{h%QCby3y>pWGN9R_3@c%JBfu5F@Tq3(Fhob&EfBR2i`^1DI4
zGqrypEfgL?yGuuEA%#eGSlQ8=N?2A_Z`pB2D&b>ptoB@FHf~1Ndn^)QQW0;YUrQ&f
zv^8A)hXn=YWsiS@O@KCvT-WN`NW7Lml-!*kUvS8LqPXpb!Yrhar^D&=&oc2hqv_t=
zoPg31+_|`G3ByN_-Hg0ul)(N6#HDi!3T2l>I2jNEG&q;Q5&{$gHZ_+KU_vN=Nv|6>
z48HqUOm8jFr--Chz%cOa)CGEL@NPv<MUztw&Hq1?8Z9%@%-9}BFNfK*km4fp6Ky$s
z8RYQ$;rw>^a3wWZ;WStr-aZe&9g^=LxzXss0tiDeynP-1I!cL#uQOjtN+P~2zdL^F
z(+h#)5unlqrx&Y$Tgaab9n&0t&Cg4G$MAkl^Udi+NdzA#Fiqe-xSale`{N+fVx{mV
zyoN{n_1oLwzeBrpLsFo`;h(RE(4hzfIUJN14ruuH@9_EXXG^v4(lkuBZ(oL{<x^6g
z16H0iM(av>AX%eoFvbfCT~T>pf?D{ZMli|K;!VrQtZHFqAFf{B<fN>BP-HVF^-)UA
zYE`UW#p+f;-B5AeT)UENaJ``DBPGqoSIAEvsRj-sC%)u)>aWI!em1I127N%e3;D<h
zWnIyv-O%KkG4HuJTSXA4Oq<jsk!Cm4mm~@EQ?J?0HLpO8vV|6VT+>JyJL)5UDOKKd
z><gnu>UKdrtfjOD9gehr!gy20LZV{xN|-a4W72(CxtaIs*W?qyH?D<>J|P@4?a6y)
zR7V|5Qsgh2XX(V1DPfYD-+g$8=zFc^wL{*LQ*g1#SFVYot+bf*8*Gw&I%qJC-qvZ>
z)hK(CZf4Kuk*FPq1}PL@j<Q!ST0fP2v+yQFEl8PGCbcS4+9TJ0wAR<j+s5{#U(`Ey
z-Jgl~isG(j1@hWWMglDq?vinof#TIcMz*=ouqI>nL{-!4Ahkji)-HIKm6u5Om;{O=
zne;6+_?XtaGs!8=&zw!$BnR3v7rd9{?kAox%iYdBnt54cI_931VwXLX^V4Yza~o4j
ziu(SPtcfk8Tu-%sG!%1!Fwd%DK5taZsw(zpYmG)Y=PT@xGou-!b#MHmr4rskPh~e0
zyTrJK53Wt^=!43UkXm(AOG~lw!TqY3&x<}tU3^ek6w>P-=z}by0NWFTHx!o{X9wo)
zARju;o4%)f4aLhWn61h3e&-7XR?x$~^2{lNbG-m6IgLzzjw1z|M!cNSZrVs=+)=%;
zuzP1Ltchi9Mb>5(I^O~Qq7sr1tPR(m@E1K}Eq_n0&XAY97oPNrT<;5c!7q63g}Qub
zCu>?N&q7_2jxnukp{|jP&VLNjCGCX9o+zWC`1}@m4ASMjpn}->Hfp14nFUDX<BDfp
zHVcgvzVNJn#GL{F19%SX&8&{eh}gG`q@JdzDL%ieA44|ZVTyX8W2y83Q`ED1ngWzC
zPWLoLLvat@*{@d)biylYxt1E^hF&=uS!vo?MX%3b1hmz5L4UG%oXI$=LLxw_B_iMh
zh#-whwZB+j%&nnbXczUuRel?YqNbiJi>Wo$GINE0F~+k=f>=k)D~WYPyIn_oAs5gK
z*%KH)b{A3`<->XzQ_CG;Xi{c|W)r2z6^}q?Y>g9hGZM`~R)mn7`PT7{bEb8Gh2!5e
za&i8JTNxv?DK5HJx`*FRFGfP*8Rg~oxN`0Za-+;|qXzV>@PA3tt-Ci$920l5j#L!h
zjRTy25~%#5;$@{E13v@@pd#UW2E3YJNDpJ3zc}>%n8vCGgAHM};94jH%OT7{Gnf{u
zh)yUkqr3FV9>`Er67EnYtIRTaK43s#G;*z4K0ToDnk&-X2)eE~Cp<2wjazk1d)K_l
zGJxKb{SkyRez9^`m~|Ib;?2a=8me14j~Ib}JG+qg`Exgh^0l2n*SULD2Amh5dse4g
ziYw^k85F|tOMz*rgTA?Co6)lV$sA_diaVugS+Pj)Y4D}y6o1@L7se+a;F2FxPMr<f
zn57BKU$etw?mw@OWSR3Eopz5&dvi*sdBi2wP|{Z^N0xNvZsVH7{|<VZR<i^Q5tT%L
zZsG3{;|Q2)D>(w_Nh&iX;OvL93QVyGa=znC3dZstoyz_kt1x+Sh|%G17ywU!g&d7`
z0)h~6)<8g5ThZuivc@INI#)e2l$f||fVZTvhpe~-=o(*|%j4!!uHo#-KNEOdwU9Of
zt#1L=mgb?2Hwp$mf>^6v9l?0F8mR1l|MRcEM5LwE`V<b#+SXHk%2RH0naS`dm-&>)
zrEBEo+!PR9$H?7K>>Rmk>h~gdGx`<)-;EX0*|g!YzzNwHvf!wy<%W$A*uh|-(K;qP
zWSfjY8cH4O6CfEmxt>3QwG?;!H|jgElIIRIi?Yune}QVBe6IQ*s4Ed83T1AWw2>zf
z0XLW7YbO(zf~7zxmk_Ba1ea1=CzF2xK0<ANefVj`ld#tht5!H86?5Kqk1wk)9Twhp
z{puBG$~)ct+x1{36#hEStRV{$Oljjx_cs=P{OkGY$H(6U1KK}fDes6$ExhYq;-l8E
z4d=G|oSq%SuLK#Ght$*=(B~2hGU5Pu>{cW#xNBEHKL@C1<VeZ_SqY|`#lnAqK>^A#
z1d0PR^ZQ%^vs5FCUZ+8f@D!*d-0%qej1?ZkRX*pX!1|+GJ$0+QWL0g9u~_1qR2p|<
z9nq9k1tVb*LWN~BNI3?lgHy7Kw_NKPXsPc2lnkyU1q~guh&$B{+y<R9q|iHqKr2(o
zpnPxPGj?cn1!M?P&9n!eVMu?25TYVkH(Gf7{`ivco`H2++#{W}K$Md<-P%}KnF6Uk
zakZOF+_X9^PI(*hl#q8i=2Ye}m0LAFnB8@SEK#ZTsMX!kmOv%ZnvSqUmTJNRO5=+f
zOfXZsvI=AwaW~iibw;DEkc)K%4n^#OaR_v$Ko-F}0lmKkCnqL35fXpT%a#-QWbJKL
z!Di(}&WDU|sRRll%B+L`AU~mQl0~f)1}n4-ym}sU5=t&mB5g}sqIQ#s6iOaiCmxwM
zU>x-x;K(h@fx69aHfzH}g@>N_CIej_uKXkd+AhgAt-F%M99OM;{qE84Cw=5@d#@#}
zgLQkk^(I<@N15=975aa&k#;B4R;CqZu$+&2_WJIU&N}K9IoZny={dP*jX*D0DrgRj
zK<6q+w<~db4x`*L<C@q^k(_;fljO0ZuXVqt_Ht0t#YP?CPD_-(^|A$wtKqks(T6#1
z^sq-C=lOwp_*Irq*&hj_dfHOth#c$~Jpo_N?c<FxW{S;NIAVXX`LnB+rEXwdv;(<7
zVi%N>o2kx5dtzRW5jXTLxL`-zf`g?l44DE(@1Q!lgJIM=Ybp2#QLJJdNt%;$j;A=1
z=MW&#F(DPByX-;dL-8V@$Y!jAL0}Shn>02O#`i{sJ*X&07JIw3mBt_byGST{62%#&
zm8((Y#adeX;VFM%g5m@w;<EQ}g~>0L<WpQTgj|L3!!i?rhO!k;ENY|>*h=^s<@sW{
zOoB@rCn>S^qAB!?#*{64PS2qf*eW81LPE2VXOAeFBaI?jRl+eTMl$uQ<!y^#H&1)I
z4Dv1Jl{5zJsA#dH9A!}joh+wJg)BEpa<l^7RHaT9nk;{=PWczPU8QX#Z&qn@6`xWw
z;!UWJ;da|Rp<sU5b253&B>HdEy6{7s03^W72{balcTcOkmTD@%=(PHFD|~1ko1O}C
zpC&`kVO(wu`s}i3A>bX#u~vlQI@tD8)Gu#5L3b&d_NVvtK0!jKi4meExZiOkLs5`2
zfAqpvko$kJZxjdvfjqAXqdh658WvlMF;gv8Q+>}ibTx3@b&}|?IS<-WNUVH~F~hvo
zh0!)N(1~jWxj(|-HP_G+9!T+N!#PyhJ!;2vjrmD7n_lV}VZAW6=Z{{;3R3zyTN3@Q
zrlGHDI`HX)IJ7;JvCQl09Z1~#;$}+ML&ER+etdtgtR7i7ch%;-+yuDEa(arHMv`JO
z47|10Q99h^tZc{3I?`!o?Ve5FTn42_8T4%##O&|9K1&%SjKZ4h6V}0&;z7fwWZS+g
zaR@!QreUwlU16__J4F&9F})sMNXVij#J;gg1salQPvg{~?vCqc_lB9Tm^dmls;bpA
z508Iihy4B|@<>tbTTX9HA{E|Z!~yFY^%Ba3#qNYj)KGk#6DAVs{d>I94&&@8Rjdh~
zr%Wia<#_=4-1OG(*4|Yrlpw~HKlwOp%BFvKuaYDmhua?8RqskB$ER{UZib#B%7zgk
z7MuM6dT9&_!=<&Rwa7y0m>nP~oa4|QrR#r<(n+%rA>n;iez(?Z-@X7?l-60MwQPAz
zY$bD25;u&DXeqq4ut}{LeapHgJ$hoVPMGu!#S>0R4dnx4QnUI@bCL{)5>}siPHMQm
zb53fgy>?DoY8zhRb*Tm07p8Ig=o>y5uYxSNOltI#e9vVIedU}Rq(!Jnxo5byZ(@Jw
zotPEoq(h>VNFzCReX@GtUcX2CdOfJn#3ZC>r8!~miJp^SnTEq#OSHhWY53lIEGt+z
zbUjOr=AojUviEE#p0Y;Qln>SDdi9xVv~ZGHbLJ9VGks@;u8AF4px<7N1cKdhF<XsJ
z;F8etxKp+v4aGN)RnwJr`QT1)@ScA;xZ92gcf|>I@3O1nRP~gHcW8Qjadx-F_7ww*
zEza-6cXWQ|kdeOB`QYiD$4nq944DtqXcEIC3r&sj^YY;s*Vsd3<B6HmF&q95Ouy!E
zteO}L15dm7+y`aap+S(XTH|}K#d9(nj=+6`{5>C&<)S$LFTwu+I62}9Wp0<WktY!W
zIhR479ut@SV<$EPAur#THe@F|e+x_9!&4{l1IMS}5E|WCA()Y(dwA_WzvY|^t{Ys6
zb189Ie0R9(%~}eF7DA;<Y}Qr@NTlBw`8M^@xZn2QDSW?6{mo{rI0+v#K&LQ4%jV0&
zZ!4ah<;<J#HQd^#4-czvtJZZRIU`Qp<LfH?prl~ZQKvj}LU!MO@19oof34PHi=#{b
zp?UZHxoae!5_p~iJZ_BEBk%+VqN+2-GZ}vB!Bg)!*9cZ2Xax{d06~m`d~@f-q(;Rt
z47$|_1i2<?0%R&`?zJ%${LCNG{gDLj7tO{nVW`zieB@VYZ51<Cw5{cq9}r-ga6~CS
zB%k&}x))DnnUeMhJb{{pe|n6!2!^87k>t|>n2)Uzg4MparuAT1>Ewyqft+KWwzLJ5
z^b~q`o3-H~ee_Ly>M=AvBL-D{G__;~Ln6&AZDAOaQ43u5pm|8rG3D%56t~GgExX|(
z`Zto)^Dd96G)!4nX7tTm(iiXMNejR+%WX|s>Ly_uzJJJM701qpe<vHBRF>}<->SDZ
zqo;^uvh=RM>q3oTu4*&1RPPdMq#cL}nTgBEjNF!Vn5$cnlQK+WPB)SF9^y{3{*6a|
z8e8&`S*KbH;(dzYNOIPOnDQ+*6MppiBHXO4M?Ngc&8^0Ruulwdb5;&%<Crz3<dE$(
z$9UWV`J|;}-VbX-f5lTV3OmY*kT;^_)a_;+S!`!!#8#$fz}U(R5yZS380r#vq?q-h
zjW}9NzLkj2&nd|bgT>aG5U1sovV{9j-0x^XjKkJ~IW^XJ{5VYH*ZKPp<<Z<P$sMi6
z%W;PSYYADnL#353#L}a@0DmZCl{9*L&#`l*oLztOs@pS1f6p%s>r@eFD$JxuJY}Lq
zioIRZkUg?RQ{d6t;#W2AV%vd%atoXH8WFc$JR?O|O?`{Z9zNO0Bgk=Sw{XTjN9q)v
z4w~O3merdcff8kmhvq%@nqd#n+L%~tc6q6U!g)-z8cy9oNGBP|pU8HK2qu?gyGG-J
zr@Kb;p3_~^muqJy7JrHM=l?MyPg(=_EmWJ49|njQ1V@s&6Q#y}SNqn&%<gmNsnQx|
zt=&TJ^i@qdj}XkFr%I`@H<Ux1RIgaWI$?k84pqPl!a2V6X4#!7g`FwbwXt$4SCbXY
zxUEtgq*jWHkhPSA$+{$2HyRft>qhe)vTj;mh^(cEQ}x1R-G5j;ldKzU2a@&aq(mN@
z6T6#~&L0)BmU-<KoaK$ivkqQU=?%JQ$wGL=;A$X?Zyg$b>|$#uklyL;-!6(VZSV(m
z)+l5Fe$)nsU%<9{^Vjp!ua93bsk-OCLsKvgI97Pqz4nh-8y!5f-Aj6Q48JP)z&w{G
zRJoWy;@Q%%2!9^Cl_Cr4;Os&A5~Q9fhpki9Ir5xE!(yNkWgmjZBF*+*YGL+jFcD)a
zZiJ_xCG`!D=%0Iq$8gopxhZ=6(XF1k)qO%$8xxZf3lc^@WT?YXOrR>55`nxcWV1**
zhG_;%2^DWKA9hIVc#Dv*SQv$~203QAP=O$lwGwldRDW@25p-n@S(NXsd<PCqtdI=L
zFVi0VuwlVyVv34{ZjbQr{qZ&Rdq%I1)xFoV79^R{rdty-&0=0dS6s)CY*WjAYMxtX
zw$F!98FMUKnaZw40nFWZl{<Z0>mjiFgTWAjX{`?cqb1@B1RhlR0>X}+SdKwhSX3hr
zhBNmfyML^Jl5nl1xX5HeRXx4!qQO$XOynE)#kYs;z7VICTh>EuCRf0esVWeZY9BII
zAnKY%@pB`yEY?n>tEwc}szBo39u-IKaZu%UWc4<C>&NwyEZMOHOQo9VqFy)DGa@c^
zMJ7dAzRly;Cr~rf(!8wBHql#^>(RM-J=CA3D}PM`)^&x+4YilLwk`VTD`|~VF9|-&
zVX-aMt9m~4*6+r(t*XkF#!Za31XH9jq~{cA3<S`V#-H7|Sz6fF)ykf!4htGlIH+x(
zx3E?)r^?25_{RcV;jIFuMM(a<BjJ7Ug70}1r-qjlX&BU~v{&n?CwEoui-Sb@m{&CH
zFMnFjmB&ihN-{2}klQ(a9Z^gzs35^&DZ-6k&Qz~)cBJ{ZaH{uB;IflUsNmp(_1_S7
z2%|!&!`^p{%GH0@sf(WMLIy%J?n2jVY-O*WQcr6qdnysOk>j3x-WV24u*<m1GmH^k
zeV#XFLXpjnd*&`Xz%ZJL*1wl)<ki(L_<uC^>YsU&IQXcx|Cj68(y|k=5YWs@(a>d{
zyQE>j*Dv0uOY9S`+<)*qpDE2um+lvNo#?Q^#EGxLrOoLB<2>k<q9&ZzsI^$1+fFUw
zfQ}`ppIC4g9IKJF3!ZKpjSJRr8qIrZI8Eyd)o@%`w*7@mIAfz{R&d734k_T&;8UGe
z^(LGhjc>ubEnCR_|3U>$kklwRjS@o4$G;iyIEkw^g0i)mMm*NaMC&y_K0&1Lpb^np
zH}lpdG49$K8`9y7a-7A4Jn98Odk0JY5B`?4`<G#o91{XLFqhFGEEJc%rZ*>-9Bd~I
ze_|QVaQI3W2KMCuy>+p-Ag7|qr8%@ff&TmsUy?{kwzO7!ur1T%4Eb>$-;kEw_fB@7
zZyw)nJ{}cyRyYk7ySHy0@DGyTL2{$fodpm^LHG8f`*JTO8h$<Ui=<SFUzVR8f9k`X
z0>?do@(T`kmVjH7KNxkN$7ue1O3xUce~)>*Ioy#%@IeC01l9xk@b&Gt8<{8D&dgX6
ztiQ}m@d#o`EL0L}fc&ZXLEH2jJ>)+9Y_;bW3N5L3c&6}t%sox%nR*>CtcYA_wc&}i
zYRp<p(iCOUpgn4~Wkl?Oz5Xz>$09Mdly%s+`j1D0ofqDqq4EAGsRZ%XjZwJ$fA2#c
zqGyNjRytn;cZ(nH(AwK?d8D<oBhd(SrXDFzD^noQ*m}r!R)%={C4Cf~3k%dJ(ED1z
zUmxzk-~P()#`2ewb2eI_b@ljPl4dxq*-62@>XA{Wx&4$>f2zLa`TTnl5Ry(vLoFa_
z$h;2{D8UxiS~`1H7&Y++Z3IX?fA-YDUbFq$%64gDb36J@-^*qB#Ty>Ro{()n`nf>u
z9ClI=zns#l`-eS$+WPvL5pyty7#c57POxZW!-=(xBhJ#kC{)6>ItIUG{GRiQya}Sh
z$LHqT%iGQ0lP%Pqgi@ya<Ht?-f!K!>ubsSbz-av0eY^R+m73=&h2$aWe@)ATT-EZL
z!@ZZhBx#J+Rk#NRSF?304lB-=MbUHGz__J(ixC=-;#N#0lH|mpTZv%{LcpW~{V_vD
zVV>nJPQPk-9VGz6z_aT%l2Oj53w9o3q3x|E^WX+HMj6ciCC}+Rq-XKun5%>Jtj3d3
zvn-zX8-k*!wQ6}?fqAVJf7MxMokz|Ij}vJDB|V4U@o;Bk2siq1jGxYs4fvIo6Zj#5
zAmb<aMaNxbf2c^H!dAliP!LwS**4K})^QWe5+UN3nnfBt=LVXrR&MtVC5_P9PV}3T
z`W1Mg5GMMc!Yj{#1Mlz_=TNnb{j=D<WNT!rEuR0_*!f%-`*_rjf3qhcN;5}CwRXZe
zs&FmXQe59%NM3oP<ECb#*MWi2IBx}`jn0^O-l<I7t5<Akx?hs&Gg!?HQL7oom|7$M
zdq^A*wZ2c^V&A94H&N4wC+2TzkQMo+&*p1ujlE1YxOz18xZYmc1e-cPk9y&NQ{OA{
z-J&UkgC0#Yo^#;Ff903Lrvnd8zlH!DDgmX^s=~JEnLW#M!NfeWWa`@=dYr+O^fI$L
zHsgYkpGeE>D9-n>NAOH#P;z`5G?^QK_MK+BXLU(8k8+`wQ79~%THSj+XIkPxV43Q(
zC$(~?_@1<SZ-sA{BI)IzbcQXO>_ZP|DwO(CCr9gPsk0m1f8%JNnNv42+?ekcX&#qY
zMoy&n?qtR;dD0ljanw?@(irOAbH=b2e!vu#xO{3gj;Ji*Lb$-WK)MRo5=<FDp=laR
zg9pl0QA*B6^Zu6*yL|bUVuoB%x8(jgdZkDgTXZFzM%klC>LZ-|xf`drG6dToxbrrJ
z0ve2Kl!Rj)e~K(G1n&Xf$f0m~>&|Z%<wTNF-~f1r7{;=Tb@mMDa!AQpfy}{vDyF^Z
z69+Cg)Q%6?evT<i=YTn?-Xj}uLy*lRc+s&#$*5qW2Xd$`zE)f2;;bEVAuy4a&mlHf
zxky~)YG|um@F>vwtCYEfT6nu-m0JtkvB<3jpQ&-{f6iAbaY+($;%k?<jn)fG+(z1(
z68ECgNO>isUoMR<-xb9S5Uk$O#2bn0BV)#MjMxtA;|Pb^UfJpztruEdBW;D%J(X{`
z%;q;8S7;E^i8N7Wn?N7Aaw8FF{@lZ`+cEqt5EnR>ass{Ujr3`M#{$FY3ifoYTH<EM
zy6luje@PvVTh5$R?SULR#XV_9<Qj?ZuUke)Zco+atg@NS!u2I`j<z^KW5>{l2AHOq
zJ#nrjUfDvK_6vhUEYS-VI)9#cnusG1@7>IS-f4P`#A>J=z4ds7gRkV-Hc~d6uZ#}9
z!nw6B=i08EX^*d|OEH^Ua)Jp5g{6Oz?Cysnf4-$l>b`KuB{`Y)y1W$M<??*nrD1ig
zZl~AjVbx(3@4aZV%#*p~Xe$sjLZi7#1p*#~Be$cIHWGKNK^n<tijb!Fm8uXhUJQi#
zN_4PRdLbLEg{>fiyhimB)Vw2Ou`27@`CmZ$n0Aq~jlhQWZ4yW*EI}*q`2vAPcy{DQ
zC*VSkPPDkODY-ZKc3w;@cfq;J4Pi5-r*~WsqEJ$`g5cf2mK(=am&0-=C4b`TN0oC)
z?#IKYX1Qe@T7cwzCEj1?S$Og7z{W{lF1^0C?4|1&={tntpjGX8^w3D$dMLXk<vAPw
zR@Z0*m=E2Y>Wp)3?-;wEFSBb&0<78BuF+>~aQHhI`1FU%o$S~vf#A*-liLM5<cQqg
z!NtFbuB`#uPiBPw1K47#_?Kal91{aMF*lbXrXUoTo6$cgm$`E%nSVO|_4UsONz6XI
z9%kkoXkm<7K7T*_zF^0fCHFo^p<}Sif0opYgRrx69`D)!#N}Vu<=sEOe)-eqzk$F3
z|4t+1VuZ2~!}7hnIwv&pxO}76Ae@y*6+}dKQE5_GYiu3H@1K{0QC@(yp*4O>jZv6j
zlxib22pv3jtHiq|ntwi#tBpL@x4w}XY7=B>Di7%`QK7Nv702_a^qQ{zox9?w-<HFd
z<?yj;GHU$8yYtd*DH4PMkN(mrvo-~MP+qe#qL<e0wRVEmMkQQuMx$s!cyuzoRMJ&g
zZs`Q(CZxp_xQXv``;G!8S&K@FM_OScEEd^_Cix?fT8nDi^MB{>H0~>odg<<k&K)Rp
zY3r6dRIqTFR9>YvY_3x3O`U0x>ieU%csIDnEG{w&`20xs`@E=l-mS$o(nJP6E**bG
z(7Pt-$BVTCr~q`H^V(sqHnfr+e5`8O30^FM@wi4})e8m{tMHPjhk)V9rEp`Z)nF*H
zG5tbX!;s{?On-XxsRzF$EoLfyppZssx)f3Hbxi8vJBzfjv9#vd1I_MA_yq4$mlQ8e
ztjw5!=!$VBT$pBpo6I*Vd(1cSX6DP-9p>fp@h$|#E(qD>-aQ96X)JAp(anh^jZQOT
z#?gxS^F<?B65q^am={;t0Dsqk`nh9DMB8GP?nzyWrhm(qEo;VD8pl&TF>8h7Teyo&
zr+KPEGR>G2n!2SF%RgJ$l&x1Lk<v+0^Y)s+%pijHvzp0iz`xITm5f$&gC_J@#Cu{d
z@$^j5V=^7BN~USc`qG6esVGqgrhZ>k1hAXpC(=7%sD+dghW@ei-6VFv5CGs$Lj;M*
zqog&DA%7#KIc+f^TnH3EI~?bwW~cx<;+$5ynpDByc!;uPmew-hh1NVEhHiXI`*oJS
zjam6Wf(?jv2!3Hq&^CjIV9b^10U#jw2zJ5*JI#Mk{D43RfuO$$fUr@E7~F*jf_roH
z^%QjUrEsB}ZgZhF$*uqlAiRGYU@$%ikm^FffPb+{c=VL2g(H*{h}mCDPKG&V`V`+}
z7vFNkD}B0TFh=e=rcd3W>#1_Oq+Tg$3{g|^3<(4fCEfaDzG6p-Qh(!ogbQ^ZL;XOG
zw1H^e4}4E~r$d(9YXcU@EPKtuS&Jq1<5I!6fYwYwBxtKsVAyE=^s);f(h|qr=WIUa
zA%9=0@zoOUm~HL2;oQxn=;=zDc{b~NKQZ^SpP*j1a;LQzjAG554CIVsiX|6QE=Bs;
z(ve%5jJyFK{IRs0mz#PPTb`gK$%hy<J+o!U{sR?8P$}PkAe%>*l$>ra(;J4C-`7gQ
z4S&oaAhjd=XaHDrK;vK@`TO$a@Np*Emw!02*GT7*vHC^tgu!h%rceQ)A|m@hJem|@
z_+R$Pq2*HxH=vD($jL-W$-$p$Z-YWPHTS;WTT%B9J2rWoe2pYkRw&gx_JL+P9gVKm
zhR5}R$#9x&Pa6S)yj3gSRx#viK9tk~G5t(=!kn$Q7Yc#q)H`eVVPB(wsUEQnbAMg4
zL#Np}UaFf&hiDLIhM9vM>2R+uEUfsg(%~9*QeEs5RUg}n?p&JXq|Bbdn()(p1iCHY
zV;37!-$#x-{^#VBq%m(+y#Ba_FU~It^NJ4h#OJq49n{%GU=>VlZI$l0rg~>m<x#~R
zH{u3CDe1QA!$y1DJv1(B^_QP9VSk~el9jvoR5h3~iwTT6%=v6}a*tb?LYbYdMQi;N
zyhEXP;Od)ZGTSm<Vs)?<cbzMWj4D*n`}FZwLpm%^XLQK@<Fx0Ob)H6PBfF&9lmYc~
zG^<mp4|=89N1AC`TUk1Pk%rN|Fdz6c8g>oC<macVEh@wHy4IKL+IF!^F@It%*o!n>
z`IzXhEYSK89ba1>Dw-d%6|W1-h%VMmUSA9(G|60sf-?(>^Sho`+Bq~`jOvITg)Z+w
zS6G)z6AIB(3MprB9d7>e4lw<*lq!doPlWxV+{At$Gj_!Jgw-`4nhzAMpPMZeOy0GF
zCXEzM+nM5q^{&qJ7?1(+;(tVW4^Md`oQJu$-a?5P*P#8Lh^O-^J*!WLT7Y<kK2tSl
zNd>sHucujumQNSxe4JJ3i?YaCb3sJ4qPB~%GiHMK13pAOr7|M_%hC)jy<J&nCJN`~
z?Z*{<e%2_;Bb3p%bn=Pf$>97eR0vM^ZVTtzwsSJ}E%d1n8ir=E8-M6?rnnPquhioW
zB=t%}6K*N#MDc&^Ik<&FxX}Vu58K^__guBz;XJ1z0*@A+wyfJ1(dgtXJ;4fKPyuKH
z8KviZ=2s{UX}+}ZxmZ^?H22LK-A0Ug=u1I`KUa%W_3)?`AzgRK24uNGLfLQc<Un_;
z=()qg9yoR2xQ6{uC4Xs>#>#nyzO_6VJ<HYiHd_{gY9TciO1EH_0$dPY&!vq$z({2?
z)tC7;N^x$|UTKBR*6|2$D%Xt%bA4g474c3S6Xv$73wxxRm{NRLA;7d}MG){S5FkV=
zBA8!{Sf*Luh*&0SR|QSzlvbpfS~h|vgjDaW>8gAHMwcpgbbkn%OOYc{eMym2ANo4?
z@N2AZlQVOLil0ijs&5>&ZCkiGAU-lDraRxShw;4EYdp|Cti9zv!(6+8C;!SHY-~~(
z?0Ajt5po60FR8j3qaPD!m2m#_06HFhU+1gg8z_`Kww@{M6Yrt@%E51vDh%3%j_PC6
z6Q4wF&pp63w|_njBcF9d6S|7qP`l(0D^e;;clrEwZ96TbZg54j|3TS-vBmZIdj08D
znX4&JR5Zi2p_3f4U$eQSMpxRk8C`90-zU0ZtqfK=v8L3f$A<H(b^~_2S?5+^x{O;@
ziq6yMJD!@N?GP1KUVTC~*`h)ALba}<|A-VC6M}>N8h`g?^=<O~R%c@=br0iCLAXou
zLGn2Ju=1-R5*j{PwmicSCC>8kc8x+5s;k7WBT%68i+=8P<l0;T1$H*bdCFV1BY*zX
zx%%p+&zV7M+nHs8YK`Z%2Gog4F!@X4`^{RHA)n^4);A+Tt|9d~$e)&nQP$)~<UciQ
z#4{_`Hh)0hVjo%9t#Xqe*3O0<bNr>1kYaD@_m)UNXeoN3T5Tdv(uHy(Z?MC3MrhU0
zq@nMB<FNt#(0@l@+;rYNejR+16B6r%HZyB)IUosEc-$+D1lt+lDAUd4P|^t({50(k
zguLD3KI?=*H<>2juSwRM>uGhI+8{CpGq#=P<{H)L+SeK5#+WzDjHlXKYMVTE)piv5
zUw(ramoa=s69Y6kGnXNP2osmFJ_sq7EPf}If6B)1{hwd|xRZrhDT0kZ<JWxk{JQ&X
zH?wZF6d+;w_H!3+pjJYASTG2$^z!>(m+!kTGuBE=(w6+s$mRF9WrBPy$_tW|C#^MR
zP@a;cQCzGILdRQKdH5uyAz_h(Mp6PsC8$BjPcL3=Vnj*DU}lUFn^E*3Rbs^akTDxl
zf1hy~7Vk+qviN0yU<DqzkT03Kz$h30OX3a?5|=pv<pih|C+caCqpk)e(1sP?#E*O|
zw9l*hte(*c)b0vBlQtcpCp%D)P@+0OzFu$^BUK=GFd<V*=TRu5DV+FC4JqeEfEBhA
zCw6D0nssRO(*7lB21s{)RcG1?biSR)e<^1xP;#@`$qabmypB%x+d>OJFKIO)(7~JK
zUrPcD1qMUZu~N=Pw3_}Rqq|T4c>Df`Z+{`BFK_>hOUY9O7!~~TGhdQ|KznPKANkso
zNZ!#g1qvIucn}!qkm|My__pj29383g9rQ<p27w*{y3Hv%xVSK*Z-7ca;;T@mfA*sX
zGEa>_lBY5euOgPRH(t_y-W4z7r@!W<wCiun?)$R)QZ#BTxVXg2j5t{O+tMgGpmG$!
zJFPOv5+x1bStr4f<sO0d@>U=S>n&=MkgR~hesrEft;)=hPHqMfF2=wh{+{7S^f1Xv
zkZPbtm_Q;1vb_+^Qx)xI4Y%Lle}0yIZ?x-k_nq<FDbNCqUG7Y9BB+#4QtUT=GjW#X
zyqcrKV~*5hZRFF**O=0lcTGr$<d>ymDa&@(g!^)_LMWdbI=S--X`uk<7^5X8bJHAP
zR?x4TN~2(@wXaG`A@#V@oTZelYas6eUrJ?e6-JNNF9yj*!xqLAQX{rpe{<ZI6bcbP
zLsAeg6w@79V6FsqTY~#b{umC#m+R`7E`SOYLjii?{IXPkk!~Q6|59+Cb;Fo#C%;V^
zXFWaiS6P?wN^k4XO&qcdJe2fY?%{!PFb7HN4~3f~cYjieH`$+Iq8}+7_>_ioG<%2%
zvsli`<}Ae)tcma)Cx1t&fB(LxbM$lo_`c&^QRa*z?uAmiP%T0AHA9()NSmo7-bi|Y
zz+5ZFtBz=@+?h00_!u?CGRcNJo!LrqbtB0uSB-YHzl5Qzn|9gGVg?)AK5GL4Li4H6
zBr?9~Nn_$R;kub$E6S^eqV>!J6uW<F-j^hS=0Vtd5Oim2TI4zBfAgq>UlET@8U|PM
zQxshnV@SlA=tMkLp&I?M5Zw-;;$x+rsgOJRY|dq0HII{gR;o~`a&SsJAFksg9Zp#D
zux(r=)$zgF-T`BK>{f2cO>=gbdWA*Cd{x{oIn*OVQ+UJyQo?bEQ59$xGE&gyAYZvk
zPF0*Y^WkP36|-SGe<*HQ=-IJXQDeWgRok@6V5_!ibxN*n{=SUMO8;@!H6UD-QSA*N
z?bbL0DfMX?s{PktMfoOqqIu`3g~9uCM&dfrSygG(V`p?GDK)C&+_VLq>zlGU8|aYC
zP)}JP6z*IX!GdB3$a^FScHiaqL+4~1ttGF{MQS<T46##`f5#11Z!4i?i6N$SAtZAN
zy_1FWj>-FUi_g5TIOo8nXH6s%e$2WXv5eHwb+R?>S$3r6JjPWQsPy8x7#dMYcF{vu
zH5@9whjkEP3$SGWP}_z<n$U7#2`1Gs9b2lM9rCnQR!3d&fsXu;+0b2b*4Ca<w>8Jq
zo%U@MWJeRGe-A=hJ+T6u_e>3dr#uT3b&?@WNXkK-q7Fcr(yXYnkEvFBB#sWvkFMp=
zdx2VSb9kqnAM@cUo;3weyF#UHIS6d0egnR>w}RS~HQTYMFgCKSANI!I_l7~gvgIBM
zPcOY?RwMDP-c4SUO^!WvFhK0Ol(KukWmcxk(Jj(ee{$RUbyOtQp5UeU(!<!r6yMnx
zi)l|qA)2#eu`~vg!dY~LLTJ*Py>KjsK<Sg%;;|T_u+G!=I)tM0@jwfBV+9O@u?igq
zV{LPrcrk$v#60|ujmB)6<x*{~O&aWa#4AT*fWk<iN@1LLZ9@M~M`KzG3UW6YjiJ&a
z>in|Ne;BXslS@h#myX3ca$BWi6YPqy7|E;k6=Sir5_FKMcw;3TDfP;7Ud>8)+$+u1
zkZFavo!fz-m&Vc}<(MjK)mFRsRvDxcpUZZ4U6n946@A0ewL|Zj6H+TGq|>u9;PQ%k
zJ`U<AgrqKe!pJ#qs}eoms&ps2*{W3AIo+yse@TJ8#1jTbHO@vaps7_4sJ82#eV&j!
zUffw|O`!|x1GqjjgHQIv9YJ6ycRwlnO|^e&BOiHh)oy9d!RvXVld+xZB7pbyJwe9A
z9?*CqEZOE;t+LvQTl2Kc*#$jDS^3ZnKn@#{oh-+qbt^aPKGr_^xip+lvFRjC^5iKZ
ze;J~cxWZ(M4Rn4Iv8S<)g1&Hn?4|0XN`aW+ocb!Qt9a|M$8A<kno8!b6daqAV@L0Y
z@RcS>xm9Jui2Zn-@5rhAo(5uZZ#P3(8A(6gsO4l@ziPE}h&yX2McI&yE$p?q?^@?9
zd8bUC{M6Fdw<qV<zTN;7=>plr^m>C+e<BtIzvA_V^c=z6^7V!Z<N5*k)vqfkYlX5f
zV+MKTVSKD{)x277J%~?_-|6_CZ}u7kILck-SHH$EQT^a+3=_7?USrVD*T05RUM!jD
zDV+3k4MrmztWIHivE@39(Te;EquG8{hGIKtSZVMbujo))=BRSGp>C)h3=~3cf2bRy
z#M)ersm4&HCD#rjKlBC+Sn`VC8*M2=$TQ$<n!0=R$F7i4;&U7ipUw>0w4q%b<*Zb7
z4ZlVx>N&wsDAQr9jF2z+8{?Q4^7GL7cB?Wz#01i}U-4j0^I%T!u$7?BRkPHBQ)Okx
z+W0@8?E9{%b!D!rpT>tTIpXb=f6XsgY1KUg?Pj+8peu7>yNGZ5^?p6$4dbkxXH=;y
zd1qOn6bel#a4IW&9Hah{szP~7u|nOjs+cgoGvg=>Lg9+PDXO-yf{vrUd9{2`1rvu}
zZbf%HRC2vDYf-*COe$xnYlWj>XUnHQUJFJ28Ngdw!HDsrRxn~ab{cF2f8*+pSiwa1
z>9iG0TKs?&P)%X1>64|B)1&{sIde6MhH>(eo1_0)5>*t>th8qm*oUbA*&Mo~RBn|A
zJqq;oZ4Mpw&YgQ?-JbVDd#NmqbB?69{w<ICJWmaS$8v{9IIHpCR$A=RTc5629c>u*
zZ!=lNYYJs<WOH<KWnpa!mu!+82m(1Xml1po6PHYgC&B^lw-1RYo(Kb>psAN}V;&Tj
z=_)4;m)y2A7?+w9C=8dtTLdV7)mqDLTgegK`zv~xWYO$;b{8-ZFf)!O0TN(fdyz!&
z!toG^5koR$8F2nSr!ILV(Il;huN~N4s4v-%y46*u&grHOYNZt=FCs-p>!ln1;+cuE
z5k;tHL(pStXwlxXF^V@la4W8yA6W6BPjv7ogR_fp33fCXjh+jz6rz@YORyGV$u-yt
ziEFmJkVbwht&sE7BD8Gex4lqr4GRcH49SUQ6pCqe*iT`(2F&C(@GX{L;+$Br<vBp{
zX3}ziMqzC^v1Ds9KwDu8!xBn`JLk8Uorg~&q;SMEi^46Euc)I4=UC20q=(etD#@Yv
zT(sxJ-iz4C0R~sQIF>MfeAYekGmMSND8plud~ct&Wts&}l9prh4OFt9<3%^y(6OXb
z&XUvj$!BS8cflHy(<hI@3~QYBVTX-!($S+B2fAQRliVpd)x@V>Xl>jx(95RR4s`32
zokP&lzzPHNjxl|P{cn_O!TC-)jqSyG`Zfq*lEI@HFE$N81Zm)Zqd-Ksrz-+0dY@tt
zzH|f*mI!<nmTrtFy<txDYo#+n5G^WQjIxF3h5NCDB8+M3maTMZAf$VjTnx_Yxzn01
z?<24m(WH!)8yX0Egui5q5yt7eK=6;;V6?Q1?R*Gw4~)Qjj=uTQ(8^5-jxwUbWq}o8
zA>JZ5TQ+@b1;f*SVz@jcof6oYz>ek!Dq-mh_$WNR9hpb2Q@4TENShGlF*qkd_y){=
z^1~0Sr)S9FfsB5(`rF_CQ97lQ%K3-4Z(p9A9Oyu;*Wo#x)l8n7o@Mp(#oLR^=kH#<
zUKhy$W%cW;tE=_pdEw3DA3t9G^8D)6)w<B;gVT%iE39#Uig%`leaKZ#;Uco_lBalS
zB&2K}5cNdx6dq1aR=-|eynepEI$Qnv^mO%N{qbsR@r(byTVv&atzppl)%yJEeL_v2
zrg*k~fAQh+_4@q=ee$Efu77*;>Zgm3XNhZ$E$qQBG5!)Gv)5$m>c{i*i_Jva<)55Q
z5SOnpK^z)?FRxdhdas^;_~%tV{pHR1f6D5oi_72Em(wck<?2tXKd*j1qs_y!_3NuM
z?ycGcnym&Bii2u~zi6wy_m`_53#Na*`sL!q#p-GK9W#*57-<qcA_KsGe{yom1&=X8
zT2vZ#ZG`BknER*^goJM*gl!qdVnCvPZAFF-lISLXbi5`Y9SE?UUfTd`8DQgU1lUeY
z?L^xb1lVTOPJnTLOn?dB1Q_!+z<7={*12JG!@oTCh`DOwb*pHie0trKt*D~5x9cJS
z6z(XCfJFmH@&chp5*5{;$QZSm6=ubA*129?_e1YJ2EDmB{{t)@U%om!*DS_U;9Dj@
z)CAyv41AEBeHXAfH#RlYtw?-{1&j!617eT~MkG%FB(#yPfDzxwgdw$|!1t&fp(>8N
zM^%ezt`D*x)0*!pEWBS>#I2$xe>a6id@t`sAC>pxChtWSs3(YMPXf+{AVOZj!IbPy
z6kGeSq|Ut(yAP7@^xEvG$*9R%zn2}=eX5{;kD>~Ciz-NL;ac0=pt&)Fyv`u6OVEj-
z6%2<>YSLWPWuR(dm)b4%Ha}CB1R|&*0Azra+X7?);L%bvZEb*Pm|*Rg5gOp9pz@#t
z3IT_|3jn)falYjUu-^*EI%)&~@jGZ$N<$O`bQ!dOE6l1-0+s=<;N{?1PAOvvF7bVT
zE08>bNK_W^D(IfJ#D*yGX4TmW)Om;KIBEqq@1S?skaUDY`7BUkpf)S?&$ShX)!~oh
zRy20#9?FD*yn(D@-9aYxJIaji(@Ty~!L+^?cImc<NiL7UB=0wvWWL24>vLnwjTx?d
zhAXE9FbR|n1;Ris3?&Y-p#Y{pS-&d8t0m1Ge|8`Vghn^-)(``NC+-WRA}LnAHbNUg
zc74tWsB}WlKt2u-Q6$j<Ilz>9>c$G~$N@iU1$G1}GO(T00<dJH3vfN-3b+H!92FQ-
zB$-0)3m7=<vpw?Bzz+!jyNkeNUBC!o(0yd$Y=ky&-+Vq|1QJM7B&l;ZRe<#}yFn2*
ze>O<3(vR8@P|11D;?&ink1n=@N3M4;rEh>WU>BqtX^02ShyEDg>0=Lgx<3YZM!NwX
zpBYb|8Bd>OZMOx~EHs-yzPehafx=>50da0feBuXf=i_rrp(pcL7IKD~<bf~}o(htt
zf@$p&unwIo@NL%)80Ibu*dPXQ0r!i5e+`bBr<E;_Rt9W10Cjn2?A?)B5DEE23nUjM
z0fW$E#$Ut^1Y_0U0-gz7vS04a0kzPS3RH%4Fljr<)%zXA{#J*%*n>M%{JFa$z@?gf
zjZD2&`#(urBm)Jv?StjRCUs6>AO*#N>_)n9+nirMG@+yICG;4NN$9wBPBrE{e;#ul
zkNJ3N%p_y<1qcK(+6ZM5>@7J;Ms&|xrBU4vaC!BPNa`q9jRX8!b&0f6D2K3#w-Ur+
z3Tse%6zCdufWCJ221x}yj&LV<#@cdOB}O$}%V|X3y*X8n16d^-HN(w-aM;-x%N*Ht
zMk*t}dIC%``b=@~JhS~1D*&-ze|_#G!vkE#mpC}gBP)0!<wTysrbqH71ah{bu<ui;
z24pi$Rl^ac_Q{F$i2Iy#?DX1#2NZXP`~T~l<DO9RK)}dLsZRhS@5mkw7}0JWaK)@R
zVpbfnRUCOww{0w;I2?=)+R~9O2nJAGl;z^k{>=}hs`Weoqd{WOSNNy2e+13$)qE5~
zH04$a3Si!7Lse7|>2GwpZdRxJRjV;FLm_|1bc;7+KT(029SDOW+&~mOYt0hEi#G?5
z@CO7vpbGxTi*Sd3bf0Ehz)PH$Ab?B;u(>UB_`r5!M4~q9<BR|f<URCf>B#L4*9W2y
zebT<l5%)dG#Md~Zi2W<pe|XdtYuvhG%?tM}8f+;>5Y3JGE4uz_fpb&M%s@>u1jOdN
z5K7%ssrSR@h1O9~1A98%fG7&<0;-rIPbZ-k4ABa**9n5OPw9)Ibl}jkqqd`}ig)TU
zz~YxM0*t-qRV)&B!2bjR1*a=8)a}3-MNCD#3Fr=aBz!D!A?*OHe|~L)D0Ea_O&&4A
z0#1lXdP<T91xp;1q<|Df6#=(OVVznoe<y-X;qh(!0Nw_?nk_&~)ck$R5erbj2(`lz
zDlDDg2-`)id)>;%p<DU9M)*4CrpFc1Er5C`Cdr=K3BHfoiOmh%4JK{w^fs6@-@(Ux
z2OsQnqc3(DZ(VBAe^^7{14k`c?tzQwuM=%LlEBY4KsnU>Ey6JyJ{xb3_k*|TY}7QW
zs+BxaNvUHN35xqiD1!1lPz?azjZ^UOGFS2q$SZj;n?Db5I3%VYB^+Q2umAn(9U{Nd
zZn|;YK@xahO&CF2?PS`qp2?&1ySZKs**j4T@%yLbjVq*o2oA*l2b}wt1(#ux91{UK
zmvOuo6PMW6Bq^6LIw%B}Y@8>RfBD0|zy9$ev#?Tl6K=!b{Pg+t^1qi^bkb571<QYa
ze+e%TlmMp}MP4wg<&Xa^zrOr)CR>>iEbZry-<GN6m#jUfSbNeKqCtB~vPy9=#tR6q
zlJ*qOSV<|7XNar?3imvF;zlyLO{ZsM4dLnVYGlx%wo^}+bVwh0l*!MOf37f^Li=aX
ze}1OaH&PDwr$i^%B>W@ylcyl%A+U_Ms$E7FVdy<G4MKf{R{fS+RGwTAwZ16=hyLMS
z{g@29uj{H<&9-GvHRJ@a#fn?*DzwNRTvMKkWYHi$g1g->5n6#@Z;*%vi7-p_0Lh9I
zf)Qw4l?;oWiUiSOnC&^}e+RMGl=sm&atp{ds<>8g$7@}Vm2#IyC=yEg#b^?we3@mC
z1U`#)lU|WMSZN7cuT6Y?w-5>inUxIYybR`y43Zh7ACgIip@n7m>Ez|P7(`wRgHVJ*
zb}Dk&6xlRd4P3vCYRCZ<XhrdqUHUz{dN8g|29NEoI$Y$!(RXP-e{oJjORrZK7S(DF
zYLvJ3_Pkd)H}{qwN!Rk2`Cq?;$OX#y*T!+}o36{kQ$FpNDQ?d}JI3)8xAkV+R?m*x
zOiEZyf1|ifoDNuH{)Taz&<Q8)GvoF-7kf~gCSbjz8<tXVR;Y0?=6|W_Zf4pD=lqJ9
zTQhU#lFb?z$t_Nhf42s<l4ay@WmOFmz$znXJK<#u{s8hVFZ$#|dnr1GBvVz5LR{T}
z6Q|@luH(`YXmWCA-;x)Bj$UM1uqvKZV?qq)J)I~QY*LT@Qd)K1W`ef?)V`|o*g(nW
zJRtbovGy`~(|~;Gte+T%N-yj+6gRAuhCm~mU#~QDlU<^Of9LE0o`hv*o45ragkT2C
z6rOc1*eo-IeN(NJdZc?d*sivAx^=yUW85!%)yi(k^H*L)R)A@ne7{Q^6X!IK!>b0&
z(TH?;ir%D?N96auSn{BPHb{mykfT)|)z8H=e~g^pyTm1FBaiR*si18J2EhaUA<@)*
zU!HB!a`9N1e|;fmrXhwhFr8OToFx@IKsV+%>tWV*p8dZ5ibOtHIJaWuT=r*ZI}>*4
z1e$(aT|c^)BDQooj=YXtnr60r0f`eccZn~gB0xyFVk>4XojRY4J^J0}oP{Apt070p
z{`1kbzlZcLuQd%@@h5TJi!8}ECd;R>QKI&gy}8V;e<Z+to^)-&0!RsE=X!<XWY+Ih
zEB0YRtX7;y3amMyABYnMdU2uY=CYg<UyTNg5z6S1u9n+az@ji#jeTJyLvq#VI-gmd
z`w<sPCA7N+ak0KElUJ~)z$kJZT7J1e+ap&Ed3k>e!@tp+6JD>VMGUp2y^3rKn#JSQ
zcGV2sf305WP0v{P8j4<c>t>1yZ>I<-H0Tw@A9ql6f)mCSMcTBBMlCF=m0eoO|KobS
z6R}6{>Am%4phSM<-8b0Vpy2HOk-9Wa@O7y!trnzLwhRdBjv4`YDLCP*S{yc@1Tyq)
zs+~6wH7BVN$v?L68%d6}-)(3c($=zRmW@1uf5l^)NvlJYQCn{Vgxo7fFmNs&?HYwT
z1$TcbLiw23f(`$*6cH<NIB21j>o6^ga7%sV4u`_t-4d}EqcmqEqKhS>$4JB=0MQN-
z(UlU>8Hu(*_IbF==tXo`?)^Y6V<K(LTgy9R)@$rr%R*!Jr~7`jNM}1&TovycKF##L
ze??t|Htp2EW7Z#Ol?*)Q`B~aRWy)bAzfg80(@;R1?NPx;*qBY_Z*mmq0!=VT%#Wq*
zvFLG-W^-KCSdV==1k%QLvnbk?Bi1f4tKyQ{yI`hx|G9}nfJ3crA$C=H$d!$4oYZpn
z85=c3QtKwEMKz2#qL+OHgk80+8`CgGe@1KSZPL$-;M1IZ^SGqd6;^Rq$VR!^k|piY
zQS2$x$XMbw4d;ydk{@SlFKCqTq-UGuH2b?nj-28f>*rAuwwsWB*3k~KzZI2fZG=ZN
z!!oOc1@#nMrk&$_$vzO7O%xx9%_hn_=xpl!Bz(rvOi1sajL#;ncjB{&+Eeh^e>{Wx
z@!3T20r+gDct1XyDcZC6Z0`M8v)#@*GH<-{`gNuq`x~tJe1mixF~V7U%MCu@yr2yi
zxX%qdFMsr_$-w4hKxzRJR?GK3*cL#r`u{*}cq?k#!)56p@kzK$BXWMPX3Vih891Sg
z2jCqIz-fmbc-vjx_F`m1?i{jFf1ZTF#vQuvDSTI~mR)*Tj*-yKM|t_sx~9W49lh0l
z7acMxYlX&!O{YT+RQ1ro^kw<_rb|<E%WkrnljBOg6yVEKy35e#)Y4|k2o=F^90cUi
z$FYB(b~heTN4_0{HaQX~3@x;ia(A+4IFMUh+yA5VZc7T<)ZOLKwT@_qe{UWC*e61-
zO}M2D?;b7Ac8CtCoAicLb-8*_hhJl?=`>T>tuxM)&v)+Lp_b`HC}sWw2K@Fbj^bE7
z9^1fc6&H2qQK|3yN@_WN>O}3FOAqtkI(a^}ll9=XlcvTqsIIoru-pdrbG$5yagSrj
z&L2;$jEz=se6JfJV=Q_af5tKxOQ(Mw#8RB%f~N;Utcl_SA=X5B2VzaVp9Ha#1cCOE
zkaeO6>U#a5=S*)U+a39<B3?6`NlX;uw_tGh_g`KM>OLdcB6`1}ENl0HR_e)tx!FWL
zneY5N>we()S<6zlEUcc?1V!7b;=#%cU2i~f1;t;&(ScCb-_0?9BpZ7PM@~aMj+&Ln
z9%+Pt0ct=S>!)MOink;8^v@f^pT+Jfa+hI}91{XLHJ2fR2osk%LNY3sXr(8Zf1js+
ze*g7N60uT5m%gSC{_Fbv_MbOSIs`Agal^m<cuNn!C}DygRMbNIIQ;qV;kUQHaj*@U
zHN*Jl&))}j=kLV5ppbjgIcq1}QxaJUgL6@s^r&!8;f|G*5^%?GE1#BrTYO;s>D|e~
zGg{@+UA#12xYS+?&ux0cbWpB^f3F1!Vu^*1@>U_<f2!|1Mb!CHUrY|qSu^s%3MJ#<
zY!C(+Oni{%q9Y@9<1bx*UG`%IFTi+?RT++XyrcU&oC1m9%O5KX1(Q2S!H-*%5A{J?
zVO6XY!B>%FmLf_3l2s^%)p-Q3%%BpYjCFD+W4)t=fNPEAShXo~P8f^!e{IItFgPca
z(~B5l*-z5ix;L;@V<#9ZG%Z1ZRUBH9TcR`B0Qk{*Mj2X?thpv!1-5`Bmii~}VsfY3
zg|>)K&ZMIy<I5I^QFt3UL<O%ybRbYGXAr-ZIq8DHi?24S=Nf!%=Txu)uOm4XqL>@T
z9!5nYe1x@(ia|(-PZ)LGf6OR@E(q=MKt>sbp+u_#Ic2mF*4vewG8&_*Sz?sBRz@XD
zO}0Mz(&XW~U)RyAHn%C8mzmso;7v_1=#2c#o8}lcA^x6cG|`H=&@tGhz|$z)5^+#C
z4!NU1*cRqe5K)T4zM-V%OaT^}^KxCN*Ch-mH)maCTMk_5-3!60e=zOH*!B&`#oowa
zI&#C+;S+|06(wDRpd}27uzcq9i(sZ$nBGkZTNfrTbg3?6hGsK$H<A*S3(+;BrNWzr
zq@4)>g<EL^0&mtw;WiO!NH)6TdZqxXB2i*xiC3y9NZv~hYo}g#DWA2Z{pp^j$d{vE
z`hKC@?QvAhsh=f%e~6Pb#r3@3o&xF}DPYsoPugJO)4A!vsD)vK&^8#FuDveJVq6vz
zcL%1K>-AX@d3skt`a~FRL1A`LJaGNmXAR}m=_knjBnLb_1>hvF%|O@DUEW>(KHJ0o
z9M-1X#<^dd!!oQtA%|VU@Olo*psmbdZIS8$k;p<2sQj}~e`M%<9q7|kN53HyMS8!6
zA%9pHsv+o;F4!1?LJdKBa|lvDCj?o8X%mDoU84#B!&G_E^QQ2yo!>oMd05opAt&=a
zr*k)AEnUwdL5rJ})b^wjoFl`|Xpr(ykLIeK2w}{upE7hgN+ffemLv3?azuY;UP{nU
zN(no|2`Lkve=1@NfLYp^-7V?zopw*Q05FE@H8m*xtOm6~qYkPe;R#(t-zt&4zg$;O
z+9`rU1;lg?*N>=JUte0FXsH<atVRnd){NCPlZ`S*s9cj(=B&p=Hp$nhloe%pnz=75
zb==}p>aT1{<jb@1=hV2ZwWFqFtq9fRC(!&d><o+Ef67l{7e6|!BuYpVt43|D3Tp#N
zw5%ws@!+PSe^wNJo;~HF@Z$suOu&bJLjv_IIbsm!iN@yoMNNtrWLCdut36<PrTA*I
zl<=I3;;U<42H2I5f<js$9?8!s#FB7t=b<&i#K57=YCta5{c208YJcYu2b>ZSlRS=r
z47xb(e|0<eK}0>@d6m)*8dI^xM-Dj^j^?M>RkUgugqQ9vts82EN+%0vfRtO#?3EW<
z%~&LFyIG{Q65wVm8V}^_*n*Bt8_l7l=$jI`tr5Ez>I>%&GD?Z@@{UIS`Zm`1{vI|Q
zc!*9VOzriiL-g~4DfeR>ORP!UI-g*@a4204e`S9Gw?>y4<XHiCjWs{>X$;Gbih1>A
z9O81tT!qwroXOg#{rszv@md%KugaIAHiz=HpT5c`2Kiw5rXoJfxUi>pBxTM5ImEAr
zr5?8W2>DjYV0`b!2e!&Lrs89mLAmaxukndNK32XIof(on;-v`AA>I)$MP;`72=P|I
ze^}X1MP$bxSEfLudQ%9G`fuvE{+k9TVv@|9{+0c=+8<{SOv$`>qGwF(=QyU_{%J>>
zcN^>Z`i0&-!#YpfS67&8H=aX$t-@@i6V`eT?RkY6YE`gSdAmAeUqj#0LZ(jaGhTU9
zIFtWmh?6@##AF*VlX)U=l=~9&Or7naf7d$?G#OCCoHH|QeoS*V?BSla1Nxm*y5U%E
z$L%BeD(x5d@parjbQICJpScf{$}o%DxE{9+(0DYPVaW{_yJ&LgY6Uu@8C{p-LV*qf
zd}|ZbvKRQ*&j2-tK;`R)Bo2UFv(My++>wCk*rGA9LzlG<^21{oKXz$vklp6Me^L>H
zI4KovKQ(Koh1c-^er#@Spa1Q^fJU=rJ=hgfr|%kM%t0(G!@`K+PZ}0LynV~|g>ls?
z&PKW3J1+1_J3=SF5LI8s!nC$F9n3ga)17COqq;TJrdDeo_<6$s+PA(@?OI!+V4@j5
zf)MELxicLyt)`->!EtGCxM`cZe=4bMEwR2O(KlTBerLO*D;EOqlwLa89bZu~XXAp>
z?BfvgzCzdUOprhCQCHeQd9S{c#r!=-4diYzD(g~lkUwrV3;_l0QFRxz0MGw;d}<L*
z#}mkpv55$rR`8rw;1XV*&q^y}Z46zih7paAH8s`GWmlQ`tk|g$rrQ{Se?4jSF-oDe
z%qsQeDNNpZMs;4<q?mtAC)C=SC(hT>Y?#sJK8(+90OHo@q`fpvkhLY)-plQ_eqV`0
zd2hOF^0l>1dYP8WCm_x?Jg&#Ffm>4<bG*Qz7H={+R5sZ|6iSVU%1M*EN*jYgv}|WO
zx&<zmyK5ws{xJr!zpIT}f7{Ej9Mz~@N>G{X=&oUU8Aoo7R43V$^>_8S?UJ}vcL<#?
zZ3883Z5Qz8Jac3X1Znl?%3W;TsKbHK7xlvUT$`>b;)8t}-=a~FAACM^Gc^j5$*3H-
z6JUL2{q{Jz>#QnMK`iy&ee9LD2GLAxGZ@naWkoCUd(E86uf=Jqf7AQOjJPyG)##kL
zr2>^qt_8nLTc$%#u+uz1=0%-s8rM_efQLk5C&K9rcTw2d;=#^yVNd5IdZz~$+A)bX
z8*_Gb#$ytbS)lu@r47M_X%^92AB)qM&Ku0KoEqcXjFr8gg;bsG2k}};K8f{xc2vtj
z{Q=|EVG9Ss;BkfxVOyZ*4e8MS3uY~pG<dE)a@K-DJaE>6L4Go8!B+pGS&N{AwMULx
zOabZ_rYuq%2Ts|Jf3&IPjVZuAT|=7EU;w{%04xf_0NUO2g&t)xnN<B+0J~_d>=T{A
ze*uUiB9@n7k{lBOIhTOyJrn^sm%&>ECzs$zGz6ESuP2s&^>X;@mp?zqLaY?Q#*guB
zzIy%g@cY9|ItC|<#^vkJhxi2;C5+dL*TRP7kH0V99zM@ND=JAR@*kATAK#a0<+mcc
zAW3%8TEsziN|HcfvFL>%z(scOQA#9Ufp`Xz0Yn+7iAU=QgEEVXxDw)Fg%E#u^J)_-
z&|ZTxR!v%e!jQ2cHq3y95<gH-p*b&tUO|(jdpaml(@7UkXDIUR@MvY!Hqs9Bnl()S
zr95W%0?pF?B+iDF-#XkZKX3w!W8BHAWWjsPbVU3}i)hQ5&ew^<N|ScePM_(Ew5C(G
zTw{?@{2bR`qxGJXUAE{Xg%G6>sXRoK>7!~dE31Hio(s}Q)Jt5|Ph~|lp<?}p36A^Y
zXZ3M1nJWaV$ur2$t+4`}uO@O)piFyIK~}Y9g^*DzyqzLewMKy|REmJ5!I+a=ph#_?
z0-(}^#Xx*h&%l%8ok^*2mo0MDMfq1MoQ~h64~0U>uvi5GtB<h<spLWT1D8e+*DL6_
z0evff7+%T{Xa_Gfwln5rx`#*YlHn(V=appIfjzZE#`;Z+Gte19gOL2vWED_sVMs9^
zg~UyXk1Kk$)<XL*)2pGJ+U8R0AQbrdRE`>zu3}*DS`1BRH;;~!>C2{D_nT&7*vsp=
z@-w+)G<WS&EGt=rKu`u|h(MA^kk>SgvA%<UkGh>Mdy88j4s^U!`p?N10EI<;rF;Q3
zIlewG-{|5^&xulBT%DtJ36jlLm?z{c^;~(2^sKBHQGjn=f}Q=9ld+;PAyz1+T5^J?
z6TT7>aws`y{nTj=!<Oa%o=OPnVL+ofNg(fn^R{iAP{GVx+?Myw>dR0xF>ZyaahdRc
zu^lO<pY13~hqdt%zeU!St5>$$_f~Y8p$y@gWj5o;lH+fXt^K)JeBOJ$B5J;rrGI2z
zkgai`r$s7;dQdhct!-NaBT!dcU)l8j`sSQXC`THir3u@W`c{GBpL8=6rM5YXX*@v9
z1<wccMMs#EwPk6=6~pDtOO;#IY2MF&b`T>H%A|`a#&TLtQhVmJW*xiAD%)CjW$Ou3
z;Mykf-YOd0Xi1I!Et{NZ=xj27Ge?86|3dN6Xw6Y?xKhJavB6HytOMjLGnbcc%tbQ~
z=k1McKvBJ6YAIK}Sv6m&I9fbDX@8U-g0w5BuLpx1ZI0~SnQLFlMJ~*)?Hd7q6;;V|
z+Xz8;>cW0S-w1(vJq^FOZ=A5+**8wmcJz(Msc&or{yqzm0-}`pPm;l~($&V=S>c>p
z@jTbVHn8%Dr7YEcB}sK#msG3?Y19hCcJytv_O_0U)PgeaN~*W_)+9JWgEgM-YMs>o
zU+V-m=BL&>9pANg0`ye<yA||*_v)QMD-pb@)-mM-V`4?svK_Bnsu{~Js1P=FR=?G`
z-l6`ftel`@M{Y78``X)f^*7{7j&b(*5~b<n{)+XW+?d@Im|U1Vg7;-qSxZj6ZnPM6
zPMqb63GW(HZmk;eI|353IW(-BB@>WC%B~nJ)1id<jfbkWd@hYeHHRdBgvvuseZEir
z-F`d--AcC)Z|E0K)ta+r%QPwHQNdSD)A+7bz^+iGr1<>Wcu!ISze`HVWX>jkrp1C#
zT?u<+ewq73iR<QS-Q=oCFdF(O&sdHXsbNX7MfPz=;_*1hc8}&8zHU?zWha|HnLgh`
zp7eOq$(i#8g)zIQariKQQk`vT3<Z5tksLVF$L^V~R^cG;xfl+rtx=h^QYpQIva!ng
zeW7^yX-Pqo;;0E8m;Wqjnd%q%M^V*Ta0EgawBE+p{qW&m-@pCo>)$9JF5mwhw^FG_
zIavk2{4AG-EJ~wWew1q~h^;_0P(?{5MLCdyJkSO{SQox751^@kr8elh)gN&+P}>Vy
zcZXAf=;B63&@C|y5n@{{_<r<@m8yy&5naiMR}o4%8!zcRpNg09)nD^gI`xm`;oI`?
zxkweQP1T|InhL*IDtdq|sVdTui&adMEH7yrD-i`OmEgz{POSF)wylCx2~^W)FB~Q8
zSg96lm1kb*_~s>lg2fPcsggFWpOC}EE3LGqgW@7pT_7CEg-lcx>1GMH&tHF*bKmLI
zr|vz&c~sg6K)X;#)FNAZs{bkUZiTzDyKLsj%U)_q{>V7E#OQr%JWt3zFC8tJsmI3E
z=iQP}N+9l$#M};a5>V6_xGI4Y0iQX@n|leov}V1;{chxc$(f+Wrqal_A>b<V!>~GK
zw^by}k=3=aRu^kSJC`y?sW;-Ff(qP6(6Alxr-4U-$i-fxt;+#**uXKjFR)yN@e6aE
z+st(p#sG23BMPyVBj?PyTF#a^H`C1K{Lf|FR`ShbmvDC6_SQZI;Z~T3JZ`9kbGA2M
z?1Jq{78$&Mwb5^r(4>Pf=Hxog>MPVSd%NNMr50fF2(FH{@_xM)ajZ5o-J>jA*rI1y
zJWqC`qiT_LciF|1^)@?GIe6SkR_ohD<DJlJ2Rx*65%fIZp=>U!^uJ`nqpc-M{1p?P
zOVgVP;(FF{^^8OjL}m31d76>5xerOtsd=~An)$qc#K{Gz@3qi$S)>=`2_OQ#r#J_I
zJ38O1rR$AQE=-8taR8-)qB6+O2!=4QOC(yx`pnKEVddR1B2zj#*9h0t7*q0;ZpM3w
zP~$0Py9uCtyPr_1ryQtRl(&uYH1)DlnUy%NebVG=i!&++qh~IrXE*9*%Dw1jKZreS
ze94!8ag>2hSR|H=5h18Hqv!ej_1?kJWQ0hzaLeSo`CPom3@fO&O}e#4;q(gH<w>`B
zy?$QJMl<44Fv;Q?h+fcZ1@YtU5M3M?=@rCn1LFz&EWPXPT6+}qdQaI<@rUr9q6tR0
z0DDCfydWoExs!aCU}j><R4X<0arZ=Bx9diK1;lZ@-cTz)$2C8&(bu4|IolP|rISJ-
zu7ky-FK2RCNPiN<VP$~mg|#bpleh&jQG`82UOA%$h`CR#+)wUH$)KEwjmt*pX+gKU
zJ#+=<F^2+N;JiKZ6Q8ABS!%`06`ZW0d007Dy>LpQJS0th%%pq3W>+=GX=H*jC#E}p
z=;SJE!jSvkedQHUm%F0zmL$Q2+6ZZiTXXd_Zq{&H4CY+84|SRWu7sxBUA0v?PQ*(0
z9``-9R6?l}^>C@fHcQYUgUfYW8;1b5+ylMzU@#>Kz7Ou(Fs4HH_dgUI>p}{4D#S2q
zo9@!Hdj)r?ylMqgp5}k$M}q4eSKdE=e~USRQO|dF$Ld=g&);=VxA5sGMYZ8#;uFvC
ziD&u5i+ti)K5-|Xx7=M;u=4FTUL2laoJW4L)v(>L({{s~IyNdpo+FPO-BUlP_2XE!
zA7#Qg%)t}m%XWKj|Ba{Zn|{N%?%3U(9!73%_pQtRhF+JLwy<ps2X7r-IVp^PT>YrU
zQc5Qy*fYC2o9A}Su9TMXzDn&}{p!~GuGDNS5w2!v(};M;1&m(?N|^6D>0(+&$fqV<
z)DrtJZ$3Tf;);ELn_3v9HUL7n_j#;Xki!gY@6ULy7iFo3Ms#=dVk)2rU}k`K^kT7F
zpzuEdJkg4k;%wZ^!TB1jx!Rh4RH`<*l<SMGi4$|S=9=t)ty3e%0Up5)D;w779rehq
z<hM1Wy-hbGu>K0`Sbx&=+@6_kKg!`~%|Gj@asTFdPa)M=f!#y#_c_uC?0@b^BRXa8
zcBBd5{f;yNbY~rD+<dzuO)7mKN18Y}??{s!9JjJL(=xqNUjHuwsP-h6VUipZ0Xdgp
zm?jmMxdT8af9+aJZ{)ZUzW1-tA25eiEWQB4Si8qPKyI^`TaZIA$aoLqAiFu_@283n
ziK3)_xZ7R?hyhtGiYyk3#j1}VrH7Y64_|gq-*%r4h6b;K1&_nG9|Q0U&Hq7jp|gVr
zP|m>c?REI;z1HaXdg6<;Hij?HzXRXx!<~V^HGs+me-C$_fJ;n#aOS?Wv3%d>-zdKi
zrM-K&BaIM60?P#M2lfwt`}UWeF3R?rIZulFubPP_#F`OHx2|e4J$Zhzz!&o6>w9ZT
ziOZMNC)SHtFA8I=$08~9F_zXI$9mGzh9c#VYjBsjE!0~0IoIN@*@k0VS2TelnQA-f
zf$z5Mf7kA_?&*!o)+VI;JUdCwV|g2Ed+zQqT4G;ng|>g`bJEim*^M3S1ImnwJbEl+
zJ(PB~%C6%}Z^RO`t@&)8pv;3WJ%|Owf}gV=sOZEM<o2a|dj7Wiec>P!L3xDX`|D0#
zkO5@^4(wjaI2iu;$M9qKwHI5QO)Jf9@;|GGe?MM^uH@IUdZOmlqn)#Mwt7JGlA^&`
zP=r{0lopm@Yr)LRKOF+BN+iwe(Do8#Qe_^g+U!tlMWSc@m(-t?{MxK1=M*^XC8J(n
zeZ1<NhrYI1=)nc-iJC_V&qb%xCOv6Sj9im$Y0tSt`*gYeC@P?ZZd>-VUO8(P{jS*0
ze>`8VOE3z2?4{Qg)aw$wB120XaR2(^zFB(((VwsbTitV0FayZMCyBF?nZ%&Hb_<E7
zL^aA$t<HQCx7S9C<W!6cY2u}8)|Xi6bjn|s&jSXl_*%>Zb+))fB1@tZ6MK=sXq1mj
ziFVz1F(^$7$6~8DhDs@~O+7LpSkfYve}p3G%wKF!4%j*ApKz`VBuX0a%ki7l@f!(y
z7#7Zx#*QWRCx>*!g-R+<WqNYT3p6CO%5n5MHD+(}@=5=a#92uz(GV}u#UN~6^Ll;i
zN!HCWV(Eg;l$Cz!ztsJ#q>WBBibG&uwv)>hlvE>ms3xmuXk3W}%TPfH$*Dy9f9LXU
zGa_W<nc8waZoXVPmyE+I&}>Hr8aDVgZkp$eeo0>~-A0#Ou9t3Cq+g#|Ejze}8z$CH
zAQh}_rg@cUn<B_l@xO{)&d(4vgUvg8Yl7hQ=CK#{KKa4%-i60A@9LWL;DX|-7kEW_
z@Te@~yH4OncJNMl@B2Cbu;jole`jNF&1-UlKsvn`{6=c9Mj7jRw|UphpovT3sR3V*
z7ywm--hEPH*l#C>lApRp`<tVFYmuSKEq76!#9N{7vXknhZ=r7nAOvoqON+iQ&#jgi
zC>b;M?f*w}D<=vSJiS+Lg~<MH&&sWDkk0#-$PuAR-``Ysik|l|VIh4@fA=+N<I};0
zA<E{Rw^l2RA6dt#Bu+`G@_FU%9`3}W#ARILQ2V-kz&t*e2v;dHA&@8$Kqnl@x-@`O
zq@>_jGNTF5Pp;e0QH83iv{a^{68R-AP9=z})=eX#vJi@{b|U^HRK#sM&!0EamK4L5
zLTm`PhMpX)xv!*NJDzDCe|UQCr@45-mAT0FNYTCRsb;CMH);1+W+VmF#f%U-MZxql
zlBi*!iv7{~Br*zY>|19&rg@u@l31)#Ezx9CI-IoIyte4J@GZnNqE;9CkTofk=RAPE
zouD)(aOC$rLFs|>^vCd%CPvmI+%U6t5^u;)c9Ku<lU?l(;wO#gf7om<Fy}d&os%Y?
zb^6B54|vX5*b6(o`N^sC@(mDzB2F^%?ylJYmCaix@QMu(GDPs97r3zj;sAyQd+S~9
z>_`j-32tbY!g}7$E*8|nPqxfrZK902Wue>c(UdB#;*^Y&aYD-}#h4T;j*5r*d0Jf`
zNM&7=eQ8RIWy>a*e@ApnXI8oi{0%Hfc=<w%+;9i0K;vzw`$ou{_p(YGTtnXaFrwHy
zKyB7u-v&Y(vPyfisf1lu!X8m#9Sa`^nnsYkj>pfK>*;Wsm+^#EfEL_0r<(Z;7jX_W
ztHk63N1+zT3BXdPa$>bPlIWBVp>Nfli*NdP0h7hgr}N_sf65YD)1QV@d%jN8+GPoz
zD>!RTg6sMDV*|3{Bq=4-`PuP0gooxl9i}tuy4offoleiih78hr2atHoD@!2S?*qHd
zmm&EMyOW8*4*wZau>!OdC3ORMCr)PgPdt71>0d8De);})&gH|)zojbUJdU5h8@Si}
zXi2eB4ETPQe`m+Ck^2UoFynYj9~4IU2BS6{3%(CKLeE&ihNkqVl=22I$ocHlIH5~%
zg|UOCsOlgR$I{y0&$h5?HFc53G^!y_BBivIM{ZwR$)jBDbFIp)ej9c_hTYd<RqN18
zm7ufW==g8Vs*EMTHwJp0iVSBSAIVj#M9;pmQQD4oe=J1hf+ODHpd3a52`($WjbpAQ
z&A4k3%c>G;QN8!&8+Pc%$|ciTva+1=I8=g9y9g-O%@&@1e}66QUbxkl=APwn7&VHw
z{lgsw9ujYUSXSCS*svN?O7&X1M#u5%nh|Bzo@%^pS6C;6yZkydD&?`=O@sS-wm|@t
zWk_(^e;_?xX$)ZvyWePsjfh-_;J9CK=SF09t+3KcPhZ?f{6sJf9Z;UA{yIisAYgl{
zR%2beX?2_1;jOW9fK!2|snZ$@KB<Vw?vIWggM1@^i?NrSkTu88C37~UN`5on*Oc6}
zAXd_j?ha|nFQfh3l)2h(!H(Oy(o3V|LK*jne*~%>GELCfqUwz~m-dnqwlve2dg-9&
z6h-y7?3}YyzNrF7tot#;9b1Q^$FD8JX-VZ}_|IWpt)$DlCSTS7a)fGR>oER<O{&q)
zsK(<c{Kga`M6LX}1S5hQ=QW&Mo$d(1h%=0Yv+wDqwQ%M*OZ!SN;*4O#6Uc9}Yi=i0
ze?a}dF|V2ai^{2Fv{l8xl~io?J=7KA&mcM{yRH$=OOi9~C5!AzbmBEc=ja!nVZik5
z!Wn*f)Y~jF5OTfL4<kV52>M+}&d3`%8~;R(6BzG;xgo_l7kER4b1wJ<!8xz|LHs5N
z&jt9w1ZUUll?-Pm?cFKPuD=^_oV~>BWJ%6m;&GO<x8U1p&KugM@SHcaO(r_eG)-YT
z`vs!g&zfRgY%De}gtp|!@~dC>bkfgEKBwZE(T>5LUrg)U$ymXZj3@t3#$L^X0wZLM
z^yiYXpgCz9_ugb|e`h%HU+UYBB9~#391{UKmjM|K6ah7tfu;>7mk-4!mw$Qs_4(%m
zPr|-AD?2%c09>z0stq!dCreu8NbmM^q>ua|3XQZ0w1L5!v*W9TwPV)lkucD<m7~=}
zRM8f2j-SzgK6|P(t`p^TkpUcYT#S^7K%}H2Wd(`KWff@!EHFs&d3l^yhw(ILPlsMM
zD=6$7(;mub=B<QZ6M<)-tbc`xGo+W5N>2ZfDwd-ky>hHrmC;ycxLnA16~jUG>j(fy
zW^h5>;}@u%<4e(Ze$G3Ew;m0$0`5(?;N&GzC-4Rv@N!8iQ#6`qR_htCi+am6cEeDj
zW^g}(#R{*XD4k+X+mS=rr{`eu?lZPZ6nmN@0;>qxf?E2;%N67_jeopzl6j6?$^MLY
z7~*ZhkPe{-jo#i5-`+hR{yvNcf@ssR!s*kugZ}}gV3N=#pTeXT(?7pYpAR2KvI!(w
zNo(RC#iO#ik&nr3lFu-VYpqmg?}GHHaME<bja0=Vlg;H_`kjLqv=g+yq==jxuo@`_
z#G%?Jo)cE+<;;JA_<!99iC_JjW<tYA8F(jGJ;}(l5#wyRkb?%uYB=puyh}1&hx96*
zUbVv9fhVD6h`f-K@Qec2Zsg+)nD<;?aOMg{E158Er09`SYY3iO<WvHyB9})lAP2%Q
z5Q{woSv{A>blz^XLS>jHH>I2=eb{1!Phs_9mQ@rvU71CE<bONQ?|vJlGmAox6bBo<
zk7Bgrfug`yP_)c}T2OrZa*D}XMv7{~whTjZcf)Mt_oMA20WS0hV$&-C_wjw9A8%Pc
z;%Y9kmFpBy(%GSia<6ot1Y5$qC72>NrQBt~!ePKDTsRaE6QD50YZ`KgkWwgrdafF*
zk<k@}KY?O5(tlE<BuAOWzP*6bJ0Mu293)dM<veo1Wk{)+ly<?%)T4Wc{9CR>A~K6+
zxt~=%N_0$cSj?W+9S<FDhsE!_ZIunG_js@HtvNm@Et#XtA{&=Y=3{R3sgg-&M=Csf
zNhPQHC&<^0wC6TONktSXN_ElYsl`BEiC!{C2NykdG=H#IK>6?fo&G0^@X*XqX@-h3
z>MW5pF`CfpvjRn#o#1<BSp&tG(`LJg9&9sazd~<7hX!kI?`obLT+_UnC-;cHfHSA#
zWlMG-bwW@@G=mg6=Ko2Prq(tu;;RT>H=(l(pyUzjc9wWrBeJzEqVx+mQ?vx_7Mwf#
zUE!==&wtsFdll6=WzyIfSP74#YS(I&YmEUS#!wf>GPFH*Der;&t;Uhlj>V}@*NlXb
z!&q`dEV);T&0w%LjK$wXTg)StB0kTPr3I0Hfv%ylEdQ_XwBga0)M4~7%_~wD>4JL^
z@+s#fHjatt)TsC!RW2z^lIi-=_wR|oeLn?+Nq^ZCY1H)J6o#1qh0Yg5LKyAMO42{C
z{==KUzI^`a({EHEO<(@?Q>lI?lMocAzNN3gxb>m<Yx<6qeH!VjDn|~4jDgG?IZen5
z%fP4UfM6&;AQP*jU;U_KXc3@U9IAHXCuRoCP?o_{tfPtF*LJW-RanGLWHJ0VPbIDO
zUw>(RUgf{~SNlCrrB#1ShtJdDLqb()FyYkjNZ4umw}z^qn#PbTXpqfApecf*z>JDB
zWEUHw?SAW2rmdzN9>Hhk0=-@mx3cFv((F4A5v@!i4;9gd@eLe?z0yccc$jiN%_P?k
z-b6)0HzHg=fBKfzeW6wFoA*fPC@3XHrGKX*3d<}dJVYxrWvDV;^-^(qo}K-nNR28}
zd}8L5y;6I^1|r1r!_<fr_jatC`*1TLLNG;|yAH^h99JNuBi+)648(>rP1E!2?5^LP
z&}DZyM~4)EA(t6Ce6{<B)sgxRLU5S8X1>9R)=nu;Gj!|jVh5sp(}|a;^<fEyHh=Rj
zZgOs%=-hT8gWJZz^^4YB*D}w)i#{)0QFH_8nxd=HqN4wt`f14_9~+OHP4vPghE$sJ
zF}Hb%;p!!ZJ1Km)A@ZOZVNRvEQX6n%j(P!JENW6sT58fdC<WEEY(z?AvbK((H0X=D
zbSf&NThVv797dCK<+zY<rz;Aw6Ms%M^+0{&%7AvMQ``^x*6T;C9*lsAZgU?jGPjsz
zU1$A!)Wt)i%bIgErB0JkWxdD-k>%RO#Y-zyaw#`uO)bUo>R_<Fj2V{Vd)-zR_6Wd`
zmuOHfB}{W79+y^VAG(XO#;L_?wpNOAj#PJ9(rWRsIfs2|x8@)>PHuF&#eb<ScFK`_
zRa0J`r0OIyG3$&weX9MhnhTef^!6zVX~BZm4VA!!CU`D7pK`5~T9LUZlWi*k+9G|v
z=#zlTPer;Pq~+)C6qLL@SU#rGc&<AtepYn(d^)`^y}lC8G4AwuedQ$6l(hed+bgg>
zzuvR2GEm&{_G+Mf<@Rcr{eOdQuRv4DrGND8)xh=6fHvN44AgEpw6E%k0XhbJylL9A
zdk9nyf!ou_JAx%-n%H3q0TJ4AKaDR!N7uQ%`<4XA5o<RH<2_|aBIWvR31ufBVOUhO
zw*%H(F{<JB1iPoWKzk+7X4DJ1^O}RcvAO1-4pN}nr*6tq5RmV<ntxQmBnz0^XHoiQ
zlDU{j^a;-QP9{38%f?ijEZI>?N|b5>X2(?Yp61!&nI+v<kvaTcO&9yd*+Q2}kaw-A
z|I=_%U1K|>J2bYW-Q}Y2h^U&;oM5+Wz4U2TTcx&%MQUY6dO9^IE9Ge6`M1DJ)-FB9
zC%-~E?Cq>gV-t-A|9>>n;a`h_RtIrrE=o%2lIrgyODY$V)T-OVW0N-7Zq;K`op-dV
zl==oMNqpvvv3^KRQ&qYTt9qA?nc5J#!FrAu^dlcn%5*0}<>#X2?PhO(0<uQ$ZzFG<
z$$LtC(UthsH(V^Ww$Bg69Bb+=Qr+9dk^vNz#^d7VigwclOMlIL%d%g%(I48cO&}>W
zijnJHZ31qYlnd&85uO_cnCwPmA8_f}#5Es*x$RPB%zX#W7hK6I7KQswcGeZpcV3e0
zW|iDOUHj^)cN5wSP3d+(V}#x2FFBUgyC~OQy)HT@T`%G_w@;=Gq~{7~qqbsz_WN|-
zkONZ=24zXz07w8eFP6!tm#1=7d?dZhx$U%}kqn(1n7*`L*T^I+o;IaS+B)?^%602o
z(R&s$J;5D|^`7vRTD_lqQ>nfR^6*YN{2$0G-Sd}Wk{lD4!M!UKm&4vHE0-_NCz^kF
z`uFoc50ZtIBG~vfezbRw&xh}aR&=X^0Mx_hZ-@8*wG!IHV5M-R>yO`uFNaSp)yhlK
zlKjue;m6C+jQq;R3zCc{tu<z5JSEAZI9MBmj*qhO@E0i!DT|~uk`XW(K{ZNRM;w$M
zROFQy&qj#xhbONVxgyQgrsWE>O(uV2ZpMt9(Xb?bCPAg<ya>=yleGD?P#U0>F0Rg0
z<kRVBW!!E<^z~z=DKOfr{7c%*Mt&Xn0VshswEQ|J*h0|=2i-2HUv8inL~y>NXcsLA
z8U<+Ryg;&TiZ98!)3cGZ)64z^2!hU5N0}^d=Ki$iD_{&U>TIJe@2Jf?Y*c^VWew~*
zyyNjZOy~IQ8S)%2*3al+{Fn5bjZExX&x`Rw(?P6urAv-JmN@f$<Z4{;IUOc)JC=*4
znn!-N>ian)rL{DVmiPX#MGqsCBp*g6P`jc>J`p)AmfgRFy0!wz3tQ@DAL?{!j0zof
zyFh&@bY4m*#?PZuIB9cNb}WA;5qCHDzqFr*_mKBDMw^^&;vGJuC{{*zPMXjDoA>{e
zX}iKD%{-4>(yWI<C!Ho+@37n@8i6C=Q@)a{_nBhxpRMT%y%x@>HeCfREQWVUSH@Zj
zEAO1G=1gx)R&!!k<f@08a@CyO4ymdkczdR52)>@En*EnI#d1R##(97Bw~yrpQFwqF
z4<B}o<wk2d{JNvQPb@b8B1r70FU9i95#(a{kQA1@eed`&=NbdZEPL2_2EiKdmjj89
znY+{Fnn-dxsw-!<vEv3*$GEVe@E4;xI*dX>MMKd8=clNT$mIl-C(wCs^5Fc$1C}{j
za(rM^kTlf!9;TU|wMBo9C&(5ek*~DYDG9q`uqa|N*iB#w%`7}Xf;0p|tYt`&z(-Z+
zq~dszq!_`M*{fhDKTUx)NzBgG%M^kz23naSjcj~&)Up+9Mm}h3nvzOXq=%Mj1z%I?
z=q|xFpS8)lxj6`YO#=vYJSZHt8eAu-mO(h%t;WX;U=xKi!@z$GW;1YfRf`xr0v{tV
zp1>WYZMwWJN2wbf17I(Yu@onJAK8YWR<#YDLUFe*v8zMYzRR7TWb?XHuHQEfnK=me
z`^P{!1=ITUzU*c3^;!^^698Mf`Zs$t-Y2cP^Mi^KU@etONl=r+c_O2q0{qc}r=z-C
z1{=44eN7c2*sgy!<w4l@!TK1RV?l37)|V1-L$Obnn_;#hJgO|TltHf407xFI+Sz(3
zZF5pnkJF@m6Zp}=3GK{|_G~CtfU&`T7=*%3HjtY^Sfyy3uoRYBif^pif)N2bYL~&*
zTS_*raT;E?)NI-d^sb|LL(#_Yt);k9wV@S(LPP6$b2xuB&uyEeaa0XDcfot=Y;Ocz
z?d-o8dTMV4s=t3jkt{0ER0xf*YZ0m~3JkuZzE2TKFovFlj{3C?MarLT_5@EPm5JH&
z`#=*v4rw+KAlPDt{|#waC@_R9DVizeY$U7cFA9D5@aM~ye|-Lp!rSoj-#C?0Fo7>c
zM)=0Jq>g{J0f(=APnEPX;D-n(bcBH@yz(@mEUW{chXaC(*vwBxzsAu(5d#4xi&Nrv
zabjUG87eYFisLBr`#K%0QdJgN=pshEi&S`RyruPJRlJSY^uA1`RsR_dUxvdcwyLq<
z5<ZogaIo~ZWK~5+&uKNaY#9Qiv<l=LY?a{XG_8M(PW2t93hOPU4g^n-USUjC%&qEN
zMkacfA)-|%l%XnGGrnSnMpqgcbEGhVg18bcI;djXX8VrcKY!y@FSO?4<TFz_D$sP~
z?LdH(B5QW`gzJo7tj>6PnO)*<LA$PI;lgR@d^JRf<)>jHQr7HPRQq(bF(@2Mt*<ag
zn^u2DGZ3T`27NWdO5x1Y%j{x?^GRtauNnOGHAAfls-<en-!++{)YHl>zIeE@?KFF#
zz!$JA=V|3$URC%~6`c#$kw$&~I{MlmVZ#@=3LDo`6{eP|Fs7^KD0jiTCOf(0U8ApL
z|N3K?PvabLoFsvAE<90iu~_XpDL~aS$yR?+jL!|Fq=C#$qO#AFua9W)j$21uzgFv?
z+1k;n=%*M<vhCCEDSiLQ7CldV2QTkdszYDvL{yzLq4Dvom+laNS?LbBwmUS@C9f)-
zmhrCmIFGD!kpOJds@%}wn7#!u(5j_Z%=-Ynd0EKo%-hBqt$^6OmVX4(fO;AbEYp9&
z>2VkhU*(u|Ba1BASY~VovFU&d##pXuy+M<1<I+Z$nLRGJ@wl~Xc2;t+y%IOfR&?F@
zdX6A|!g%b|?9l;B^(}>pxftnZN%3b&t}Q99YJ4dV0-z=O);^Dl4Jd+3O`pe61|@9q
z9*rK5T97~N*qCT2?%3#QC|_vwG_!x-snMgnqQb(x^INWl>y59;8)`rOJ$b`#hZp57
z#oOPMw-n8*ugY5sZuj}p?Ag&a+z{H)Hb#lwUDiy)KPhVu$PPbuvq;!Yca%5080B78
zLt#Bt?9zQwj?TFCqurzoA=!_Q6+AV#n~yUvcEfufHzQ6juPE!g3prLNH57l??VYMg
zcXbiAysh`OQ~yhuAYx-+;YS$qsrJFBSd@C{@ji_#g+88!H5tsZ4CWOXmR||KUdSlv
z1qRqy$Xs-oolBGW%76SJC-9HwF!>j9oig5SsZ(fZ0(UcwX`O<#@4Zx~FwYz76f{xu
zMoE2zjaj%i)+gWuq1`#xcx+fPS0ERvlWe$HonZbw#Y%O;GIDj2@7n6*YCR?!G;UX^
zPMWP+x7$+fK~RRkbv=zgvC1IKZd}9fnh3lWN7Gt8{o%<~g$Y%5LQxi+3(^MHvF-AI
ziJkzLu}uvV0yH_5u_Y9jJJ=`3e>na9^{)q>7t5J9;cNJycbC`0uLrtx!yS`OcVE6A
z!Vg*rrkunbF^e1e?VsJ(!)GeBpNr#=^aptN+gnFke$L=|j^J@)00!U*j)>~c7|&Gr
zDS@Yd;v8_Z8aJ&G1RX#SgH1lXI58<vb36v15)nZT5+xyAI!aE`m=5bKe}Cfg2Q9e#
zUa~Pv7)d2dK3G@Z+iGU4puN>-9oT?s!f}`4lGn5h=^~!mGOg?eJP|dE;)Uu&m?+LV
zX!&^s=EGSfZWer+4WQ1fA+N;EAm^BuBWVF8m_qO6bTmAe4`2^<*?g{23Z{*x^9fDx
z*&-2`m7`M-KiwhGGjH8Ue<DmGD3xYX$q|CvbzW-8&^PUgjW*1q3)vAH5Nxb2bI<^w
z35^Yex26Z#+N$Z5mMf+wjb<E;g-S-Hhjp$?TEnchRD!}t0_Uv!JSjfd#-vPgE78oF
z_Gk`atO~!Kf`x>xVR-R4{G7K9s;htEwg)Y(7c3q|8KzMqd$>9Ge?G#Vtx>*|&1YN9
zULXXtX#G#Q9rBW6C|tK)xP)MN&9kHc8s+=eN6Z1WRU$4TUL(ETf+R4((eI$Q0O9*c
z6}?!Z2K*&`qIw>M<Z9|?sM>qaB7|mOX4e4IYk(_jm|_bLFW@@Iz+0oHI22f55<_eW
zATG@n4~1t<&WC}(e;>8cVKqUiFDGdpv$CLK37*=oD-tp>Z6B9{RFD++w!{Dd41V??
z$`YhZmK0|=$!Lf!u)r`kT4v^hFsYiQUkrP`^p!cF2X$7em$1B#(lg8<%U$f_`ri_s
z3U0wOt<_~I#r&VLn~~s<SE2*Q%c-K9qt;3_Mp!FwOoOFZf2+9Wy59SIX7rWGZ5E$N
z9Qd>%Xk^TuJG!{yBev3VZ50L{@n59nfJ6eU+>w?Oi96DAB6)$9linXh%RxwHHl;ax
z&~h^CowS@tdv{t+=G%dmQ;GM}aw_pUEvE?HPRn=1Ce!jAu}QRiB{Z3qQv=zLmQ%^w
zX*rRwPslm-e->hB$k~OaWBg#on|YdkjSxqw?$cGcSXg7|w-fnA(HPBY)`nb0J7M0B
z=xXcnG;7|%7uhbZX_?t#0H@(wc{C|4L-_u+!;YW2*c$~X@1Xmmi_J_(&^~x|A*=~v
z75)o*IXwLG_Vvdv|3Ek2z5O1#A_qW|65e&+(=DKGe>;Q2zvX+($&_{>G<eh~JbuTN
za@e5^>>Ivx2Q4jz5hP!=^;>B5Ogk-QGdMC28#<N=nx?XbAhB;P^L=X@%c|-WaZmj+
z!d;M($A()RUq*%7aJBcPD~|eCclg>JK4(<HNE`ViPlCO{0DVA$zkgGxVvrGmPBqAu
zCgm80W1K0Y;(sk_{!DAT-@Xz31tNoBNyjXzS2&kETfgVh((Jo5p{ydIG?h=<K7#|<
zE4K7XivHH)ILx;&^+XjF8Qp|%{r=^99`}Z$J~f|74jPF<LT<XF#+W2bWUG-`Vx~8h
zE9LH`cX6g7F$l0d|0pv}Sr`$5Q-1E6C8fO`>*hYM27g2d26%GIfY1om0KqXu@wNwI
zCTT2tkQg{&nDY7@J;doKO-a}EbB4&}+R1ZHN~aU0Ky;l|R6aakPh}k@7hKa(F({a_
znvk+eV^cIQZJ~;DE_byZX2MILhtj&7*v2IBrIpaPx_*W>ojx}SrW_!3o!AyxWItJY
znZdTWV}F`_4QGu3)+5QLI8P@wvzcS40a-z%VCr#{PB$7*_fSg-73DtRu8DU(sY1*9
z-Y)fsJ2ctgh-PLz9y6v{ol}YlK?g3W9A+sY;?e1HwL}i9B{5UN(ELaXi+%0E)^cLp
zph^5CstKGDwP=*Gs!*wqjH@jz`1yB3*6~zTEPwY{sKtmO`JD1@)sIYhx9X{qcl&oY
z?3P)`v7sTwFJGy^DdSNQ&9(}he9j658QXKkHl0JLw7FbViD;W)ujKOjUZzphb1@>J
zo-U1g+rfx^DtHQW7UBNa3wZT7PHE=3BT2t59f`PhB+qi<NYaS+D*__tLEfY3aV=_L
zet!~g($R(|^i7yqRp9fAVP7)wt@S+_%7kEPea|7e^B+>)Q^-r5vpbgeh{PSsdqna=
zd5`q|pz@yL=xUAs;PM`s_0IAhk+xjkllPVP$b37L_o&4C%X?Jf_3|D?@b>cFj@ab#
z-j3L$^4>~la(Ry$$bRKLDtUW(kDAe4(SM}4mss8_<-41zR>`wSByn8q|FC_{E@LI|
zsiq92e<^CtR|!*b^U{DrC}ij=!$y31k*#@7zwudowzA5hF~nl*mJS#$x{U*~#m-Gz
z>>m+E<UVAJGp`&xyo3h<=%*|)`$15`0Aj=df<OP|heU(5o*p8xs~ryc(+^vfcYiiu
z<r1vKAkBSEm7(fv3CwcxC5UKBp)9NU+Hp|FCDi#Pe3QjTl*IXI&G<U%?lQDAm?~q|
zy9I;MUIf2=FQV=xBD%B)7ztVPy2tf&OQMyUnry0)W}n+kr>)M^1YOuZ&uHo(Y2|BE
z@Mey4$%sh0zgNa06~rnL-EqkNWPf_mcda{Cv-#Soz@R3Rwd}&qG+`UZE;W2SRc3CA
z9IKKg?U(^W_k3%%6cPccXtJa|!z^?mTiQ2kk{obqwPV&qC6;4?{L+n0wpi>kUNXN4
zCu^CjEq#?^2Ad+4xk}~Jff!$~Kw_6tk+x+83nLp*z7|kY!6KULDpuw^8-HVc%UgOg
z%s_a}G-eAUOWdBJ=2IjRNrA8S*35a@`98xhR(~4QgYPo@zDg8DK2>;|=g~{dqut|5
z<}nyTn8z**B$=>q3X{+JDF3pRg$&+r*%Z51kA%6A+v9zDoE_(MQcs(MmFNM@6j2VL
zFqlSUyEX-@^5~!^%6|CfLVpcrr?S!8)tc&bujaR_HPPur=Og6jP2{?TZLNy8Lnyhv
z<)zv1)J1alZYBGDO6x+P-xKoxthPR`flqVH+1Fi>Luu(o=q;b#wKGvZxWAFqC?PwP
zTUi<s0VPNOlT5kjc0Zz~?a{hwX?_QCHWavHMSCcCp`bnN{6Y0>i&+SKa51}`^3F<j
zE9%|fQf(*Op@uyWcYgtUAntngdVt#PrR%-$$aU+z@JL1LHE`sL^}q!9PRWvEcnwv3
zokL&pS)qmv$T0?oc%Jg)lvLN(5dH(paX&AYab74B0x&R_QIi)F0W_DPsstyOnc^pz
ze}2sAUmyQ?l9^d4ya`{!$NaCikEj1W&8!<G1xh^p`<JKi11S*7;eotxpbo$O=kU|h
zKWD0CuezsJhxX^!pAXZ>4;e%U45Bnf>j6ZNj4U0D@j`{45{SSD_da}{o{bC;1sI57
z(ZHV!2QfTaCzvSw_L9eS8m9<wx{5F!f5!N~M8l@fNQ39-H=dW*7y6M#m}BxDK0V<o
zG}F{7%sB<t;&Ps*F>MikiY(=+7!Kmojmh_NAm!U33NwNEvNFg)JP^-g;$daD>k1t5
z6|z7Y?S$f+$-pU@soF+4vb+0;>1Zk=_a#Rx4}2TM`ulf+gBIRe?MB%F7AP1XfB5U)
zz6(ov39T<d!PtyT{_@?ZM7IU{vNUS}(jN?2LCTg((YncOIJ<mJsHwMP2;_VE3}5B>
z^bE*fx31G|T|!;Hg>QJwm!Ab9=*R8#ZPjTnCJPLd5oFyi3>Ier`T4UzJ4;0L7880w
z<ce*rqIumnurvz=RJAi0p4ZHle<}W;{mUM+N(H5H`4`{#CC5iOW}{`xim2;Yso8u+
zlh<?#7&ZT;%+!S&S4XZY?7%N1bOdV09zzP^$EhM5i_xYu3#`ytVF_ofwixb9)=rsh
zq5}UWS~6yTv?=Qs$L>#{qnVjvk@h-REwy0RHal3}ga@ztS0Q;t`lDq+e{~ose$~rs
zepOq`TlsPh`j!Rjy#a|ZA#=gDk>oH(&lICKz)tMmGg1`zEPo>V!$@y2j1gdPri`Nv
zs<SN#V_}}H7JMZ2mV8Qh%RJqN^d_EK3+rG9o{XAh@*Gb=*eLEgjQp?x^Kn;HH&t~V
zIVU_Ws6`{VhF<mK$wcnHe^*2h#Bx6E2Q=ZchaywL#Vt(n!aj<gqXf60s8<n0MFJHr
z5Zo{GNXSMH9*GSZ;n{>d5E~6*DK3|w;aOTP6-4%`r?a*-)hi=cR1Xaa$;vWQ?!3?q
z(}mUuYwbdf9^zq=cQJ`nu{ew1VNd*&Ee(-yh<|28dov=f6#HKFe@D)UdI&aVM8lHD
zXG9uNcu9Zgj7anCc+$W3jELpS3GHnVre#J%76rJP8mpI-gJ@<@QOrSt0-2m|(Cw@w
z=T31Z6@`REqv||OrqWj0arUsjL!_9+LKE?nbm@B5M_LK%o1gI~rORGyz6q!coD-j6
zR1c@mX2vA*LCTlQf3t-G*&MWG3^E1CbQ{gy0@^~m{2CGUb#6JWt9*9%k>_-IW2UhK
z_e>+>-mTXsgySXbD>o+d2zdr2d5yKyIHTs+ZldYy4It*RgsfXXWtPm!i_4Oj=nkHz
z+&ckhyR{a#&TGxuJETKi4eB)`7^O4i<CBc!{JTfncOLBAf4+zP0M@xg#im&GQBb7z
zGM_QwldBTvRCx%Jvd;g2;iGB;;2fvpGs{#&L$^{#1Q;*pqc`$?B+X%{nRjXctQ@zf
z)wy1b9?B%0!1XaAT`1fPhVMdwW|x6WdlUoL*56RHyza81tqPS4vBjd!6&RQP*fc1D
z@Y>0%DZZ!7e@vIl+5k%(j;jhpvO%7KC;_j_2T40`@g5-ef}3b#ztbjg6W7#fdc$W#
z%yhs*#)VRdZH{GapoX~!5h7ishV^UKNBX|zIPOl_&a5H_a`oz!5BXUQYgX~~d%Z+j
zQBm_-y#zCJ?4>F_+GM|dp#n*sgBUxXB;tvw8y6C%e@M~h2lt#O%um_FZDG6?U?|5S
zHs^*DqtT5e_xREZ%(ZCiXQ!P<t_YYd0v6$12v}RcCSdilc3eRfCdcWB*?I$a!pZrW
zu&Aa@2j#4Gx`q^RU6jxXjwN@VV~ZrQUS1UqP=Yx27&pFPt(76c>2lw)9=6XFz+egx
zaFU)he^RoKA@7xA0J5f_nI5K40s)F7VcyesZ4O;$)M7qAZ`lA+sd3a<osZ63Qg1Tq
zyy9uQmPk8i@W_p?<WTxkC%$pmukebirx||j-C5t}>B7<*y227Jl)?EH|NqRDW^+a4
zXGWo@5NC>B5Mq$m6;q%{7_C}qw7fNuKl#%5e`|2uY?II&J65TaMcXG+o6<K0C>+-G
zO?6eCXg!_0c+1({xp>3Hq585eOJTApubu9cKq#ko-3~5SoK%S=%M2JJ%sf)`<T?sD
zuJva27%@1j)F<6ZT&+n_ts2^Le%T7Ng|puOq>*K$UQRxTBMv%`(rDIrJ0r_MBQ+!k
ze@pZp<MdTqC^-!=EosY$lO0pP<_)ffoGET!lMQpx8Y7!*uUg}C_O`9*dc)qfmhv_^
z=ugyTF|^uBK6f^A$SF29jdx`(?w-tL%&MB)Jt`?B6>=S#?@isrd{ZXMV$8AKf_g0q
zwZwjRHyW%JO3lq^uY{qm=|*#)A^2Jzf7p$lC_b<oJyE{Vjh<$IQ8yZuWA4AY8$IoM
zXE%DHwyhieXdJ-^a_Xyb9Cy7J$C26w991gw)A<vf{nRy|Par?vYYtN2>`d{let7Dz
z&lJDi-C>_8UTTQP?)fZQ59o($FRb<p%6t0ZC^X_sL2U>cWDXVkCif?r+bA`2e?Wb6
z8)OsGWpFbzP`i?E4T|6PH&+*;u)n1To`g27rKei^jqYz4f@9a*lKDzquTcNY+ldff
zQzabA0Ba$ZtAQeQ{Uyi=u<VlHPX?fC3HfUHcCqYYK{&ZMQ45-`CVuFG@xggDy`{cg
zq;O8beonV)c>LKH1H-5GFvnuye-#F=za$m|4DXki2gYJE#YtFV!ER%WyvHg_5~N|?
zec_94`|C=P^M6KsCipq`oZY}gY=d_mMRyM(N}^Cw-9V%j4p|-m=h@qw!G7Mf5=t+?
zt%V)6&SQwl!7os}&j|s9S6*5+g4w*<2bovSwrp!9s^6eHidygn<33v~TW?=Gk3|Z%
z`2}h4Y%3K`n|3wxhrJfgPSDP0>yqGY=H#xkD(N^Chnb05Ux==&)NqSaI}h~>J#=>$
zb*0;{;&&`}bW<CCtk!I>tsYwWb^af@*iHQpb6OJ7mvLSw69F)nVdy*)mrcJqDVIy^
zCzXF+$G_fwf8uGeoH^~^`rG*F_3i23r*Y{@A(Pfke}8}SUq}=TEldJ5Q=q3G|4d(>
z{usp;Wt^UgFN#k;zD>i(mxP|<gdW#gsSZ8C2~s&}?HKy6B6{+Lb49zAv};KyNRNW(
zcKPXI#iB&X$qs6is4z-VD&0BVNrN)!?#F+D?k;}sHHQ`0sFg{-pZpTG*^SUP^9q1V
zCY2ahAo}&!*_Iiu$JzW8*UZmJzrv}$1pVDeqrYIPEJHF@)=xVWY1aW)+n^vk632t)
zk%wuPj`%2u_^4O1f=k#y!qM~3MIO?0E?x#MZdtctg1bpe!z7xr?%*SI;p37~GiiTp
zhscuT`;@-RXCax=d(ycRsC2_#c!b@aX-AP91Yft49VG*VVSPd^a~&-O8Z)9pYv7md
znw4NM0weUINy{@@GWu?KsQB39c2^FVmbjzg2Z_HRjtwig6UT<xJH#<%>14I?e;3C{
zBr~9XLL5`dF{AI1e)+)ubWwJ)OgVp}aVd`>Kp2;0)sUDQ5I!fKCA`9CNT}<e`(Bo$
zIN4IVEFJME<@0?;lk@`ul$l5H_9af|(X_(g9gCbuOfdcl%yM-3WGYjFEZy!d&&oa(
zR+_w7szzg#rz$@)Lh^Hx8QvF}ftT$kd1|~Bl%-6JJd-YqnIE&!FC~u5kVk)G7<kT%
z$lpVKh*HaTl>iLvOnQ4h`TSO*&sTjuE8W6g^N3FVcu82tb+q5wXyb5YZ#FIRKLNqa
zT}qLeb|nt^poS24<-KnwM1>K-#|LdKqI6guK<dnU;xvG-D1t_>`J6{aUdPX6e&SyG
zRJ^M3BblBJ&zR?WrpQatan64cpotD5KNoG2G(fa})t>DS;8vY&<=LB_SXQQ5EP8#m
z#DIo(x=L#!Mw+eLXxHyG@!Z>Y#v1ZJgT*t74+t~*K&T3@`hdvx4BzHx6Q2=jWu}v|
z=qqIgu8F#lYf1OHdp<?8%!M>ZCXSgKJ6vJqDy{BP$`*y2(ORtGNWXuNxXw5#@Z=?(
zj>w;zb0h30a0(K-B3VRVw?3WJ7NEq8mV*wH&jxuqd2$$S&$BkpRlfBzkB6MXmrPSH
zozq2Y_C(jk%yaTq*Pe+PU4MR63G$qD>X{VTt`AZM9HJ$*bK7{6R~n^3vNkxmHY!vs
z^QoKn-C~mc*<l+UoGgF(*5m?7*KKw!azXlgscS2{O~-B}(=L)Pm3^CR+2RNGDEuj)
zLXI7)xHgFEaq-!mO`eehj2PWDQ#>Rsr{xfk>5>=IEdAUUAZ==%Hks3&(X}s4l4MJ|
z*KF974aeQWjos!_WottWxnq*b#9!DDlb{&XR>cm5<RCEzo<e{3?G?Dtnh6b~WZ;Vh
zqA-$aBgWY{TSqJ!xlgNJcPq>#M}}J(G6b3->BIuz^#rBe$d?t6cS-?Qg^f9GmN^)O
zg}QnPx=XHp)&3wNRpN3}cViC}2dxjnI7nN8BewXd1X;x|kKDiyfOKj%!!HE7RZg7S
zY=Z<`k;U=T_RfEg(aab#YP;6eqgfoq4B7Mk{G)72!JfYo3=+ekvK=p>$TgiEr44~_
zjW5*DDwyMDCJ~ce%K>=p@VnWg8>DXVfz9pNQPs=x!hAzZxtTdm4>b-^TBd57EVLnC
zs)2WG_akKnTJCvT4-!MZl+#(86ZqO*eH1C7n6|Y!G^Br^&`~avf}ffU?_3jF%|lM+
za_N#s;B(CcORGASDz`P>Z`oF=-qdoZfrq(m=aw|a1bgTc`zr(gP~R1dq6ex8KIcj?
zrRTCso8zZko!LAR5^1@z%3jm21!b5r_^y7fFiZW~n7|KJZz>9G?Q;ba*xXw%xl(Tt
zS?-s#9oB!oJ0k9t7fUg)DR9oM-lksDL-Aw#{#X7bVQUehB}0(28iWp6$^$;41o1hC
ztG~Of%J%N(NMoQIw5X=zknFfBY^G9Vkk)q}x-fBWx+@r1xL1|}92gmK$HQQdxMS%y
zNWN3L4YRM6ZdbA0kUgY~dC++6w?`@|4a|_oqJMw*4eO=Jj`z(_g?5yu?`Zm7;OuvQ
z-E?7_-@2}5bo>3%El@L5@`G!^LF<FH;2>>nP5tgPOAt@dHR<i{iKyVnq%j|Dm0s(k
zR_UdEXoWuFx4l9iB>dzGeb9F0KL)i@r}An>l31v}o74`N617m-|G!j*NcO>mZdF1$
z$pTP&U;2NPlt%XzMq6aUxK7-CIwP;?PaK8(-08h#c)J3@Z`R=L)b5`<8<c9x=;|#u
zT4@W}iDCf#Ca(0Nl3#R)pO+HwypG{csrY|(5Z5%9ab74B0Wg<Qatafd%kw9-f0oA|
zmoEoUuwaKML+Q3}&zHmR!xlOY*~Wp)&$ok2U~o2MZ(QcyoA+Pl<?wauwa21h=EwW1
z*;u|QdD$^}5F^owJO|7univNgWU7)!OkBe87{N=L&_~op#FWA(St%AF51yBEyn9Uv
zaKQQ?s4!O+hWT(F;I9}lSq!ORe?*DS1wk$ha-mQP<yKKY%o>5ELR*jID|SlNBUiol
z90{|$aGR=K)?%u5K>(*3d5=6DT8ew7xpGKDbs`Rn<E&Xti!APo>IyXk%jIuT>cQd`
zcGm&%v4|OWBxvK2Vr8Z1Ks6y=iL{5sd{(>vh}qG#<8oGVQF3jMLcz+ie<0U~Y%!-T
zK$mZl+jo(({wk{-6-t<GhCNT>#^Zl-63LMr!k%fa9R8z|NPu=o?v7Y)e@^awWHGMl
z6tbQcsSU@#C;qsmCir`2hN7&NF9t?k<=TFfeYw9WQxa?T&~_KgbmbwIHNK3q&Zb*f
z5j$$9zUMry6xJ~hv;U%3e~=ky_eWgho;1U$BYsjGsacDcth}YgK8cw+(&Y?xB&6#a
zi>)f1Kd;i2TPTF??Nn+Ys<^5%MHWH7&5KG&rJvgml%OClZ%NmPq6=FmMP1h|s!;O#
zs2b|0?{k_9%sEp3eJgQ>{<c&`P;Hf&8lkTX6_0No+Fc)SD7oP2H|m=6zNLv8t5rgI
zm^1I$b)D+|RlJUO=FYn?xogJrZDdiAo%MF9j17Xpt*@%q^AG7Cb!SW6mvLSw6PHjN
zF&CHJ_9rNRIdkN?5zh517-|FI0qAZVwyY{iD{YlCua(&^@N;=x4%St6v_Ibs;vlZ!
za7OY5Gb57d1KnRAXi6^sT;%fQMyB7Vx0_#H$t?;58Eeh*_H9x8MDn+iRyae8ky?P{
zN8gvf+)JsAl=#FSl#+lymj8DAOAmK|j#nstqUhm&*S9|==0<3TXlKlctU!x3bJF}#
z{4tR^Q51&H>3+Cp5_`af8IiJj#vQ#E)~Le$GgGpvthEGSb)C9Td`-BL;VMZB(!;jU
z=5<l8a%Gbl+^UHbnw-;%6v!V31j~Qf#N*wsUv7$)9)&i&l0msufg}L*W@lZQAljUN
zeKOB~s&L+hd`jEYkxD*i#WjY_hr3bngoLaVND$HwS}Bz|Yf6SONl%xJRT~QZV5sM~
zPQLOJZ%?{o<CopinLE49CrPjIv%2STg9$zs+61Ik+wdg+O%E#Q5nU`Rp7{yM{A2~4
zUM)6&%<=CQA55!P{ao@JU+a9G_qF<vc}FRKuu-i$qO_~9QbsEEIRReL;)OTbU;}us
zgtC4?bOLxu(q*++;FzfmEJk<>jv8v`1zE?`O1MaT`{PrhFjESK?uwF%LMIAKa|$mo
zSgWG=nq}aE!aD^`5NnBoq+$vX4k42KJCo>{88mMMp*6$px2%_&>7{um>;@(=!s&>A
z_gk4s>{N<(!aCDR`766>S2y4^YK&^Be(VYY$x8_bLF&_9@*_pUz<`C25n^Tz(H#Gl
z&n==o2qe9rw!u&Z(I!%a7~njx>i{nW`IbX*z#M$q69a?Q)rKp9izNR#=73ttaR<~?
z2OZD}$__BoU<Acz_MLkA)19{V_RqV2w&1+X_;qtz%y)%cx}CJGuzmA!Mi8<m`9suC
z?F$1dR0!gl@^_!^lrizg=Bw|=Ald?LfGCX+O-UCxod9i>-Bm^KDKsK$3O?9Pa7o(R
zwX&QdIJwp`&w!e9F_(VJBw_dVh*y-;oP@EJau(FeF;dXARQsS-UJHeK5VQ|}>X5$`
zgSw$&P=oSkK&{!?%zbW)IX<-FxHGo*^4dC}t$Q}qwV!1wrn>9juxU#o4HlI6)O4uC
z$tM1}-%IOr^HAt8KPzQDCdfPV6fo%@fbL5}k*Qp2EpiK;bj}68+hc`X%osnlqtuY)
zhNx<Uc4M0JuW_AY|F&EGS!dyYL!ID^>P_r?ZXb4WN*uR*z?kE7jyU3T15poCjr($R
zI8Su3`rCdo0_3D?da$v3r-_HVJ@BR5D}9<tby8NnYJ!a7E>gJ-2?p-t!j&+!(AjgD
z0XLg>8sV}mcE`t!`_u-zOi#H{e-3x>zEjW^InTqD$rJ(W6}K8`*zJjb3(`}L6_@Tk
zXG0t@*gN<ARF0#m4G;WucZ5j0+K>EwoGCsrLj#9%QKLhL8K9{Ir-?&R1AH#${-crd
z<I|C&FIaA?a6@xGwY_de#mAgP&33}6uF839y>rBIiMFJ}hutoY+Yrx9K&@$0@Munj
z!YS;jSfBlq0djTKtyyk=e;<35-}akLQbOxa&0qJ+7s1*9c0LHDfn)$GJU*;|jn<|6
z?i78(1C%B&xuB@~G1tW6Kxe_712b_+*OlkysHo1i>)fHteP~#V$45Cm#k`D9v$5e5
zi`&lLxge%<eUQZINcAaUIDrQ}o>L*tbBB_$BS&|@rW;-3kI1QiY7*j!YQ7`na;$Zk
z1~t=@;s;1}d~70=TMLs~i_%im);0wj8AXpL{QQ`F^<1HB!=cY<zuLVUa!GH`_@uNf
z_BxC_*O{i9ASt3es)~<?4vZ4mOA35#gCO@<arI>|PQ4pJ<?MG%2T3XsDo78ALHj}T
zDXh4X7-Q=p<LU~3yOo8mU+f2i3XPDVgKIgR?oz8^0XfdB$O(TsU<QdoV>t!3Lkji{
z>zd}4tGYJNnPUU)4m;`4$XN3+3{^3=#)mDJY;(dO+3Dkwt10FhWv(0`;!Jj!gsu(C
z;G|(jI?g(;A4~u8>eRHFIqr|89xMYF**(MQl>Gs?`}Wv>OCh-(?6(=|!<tb;pc@5s
zmUP8-CS@R1&3>Qeei$d%+2SRCrKxe8N9(<p0kwpZ0k_F`Of{SVXF;khdxW;oI*P4W
zUrHt}Rqb(`y01f#xGwL#)TxN=He8(`Z@IsnMF@gm_iquh6PNP@^?!(3?g~x&j;iTZ
zu&QA(o=<6irrf1<DOz<Fk)^lz=!oQ03|0-h{@{q@Q~t!Z^N1{OD7dFLfpPfx(p|m0
z-TY?&y+HvQm%o3%3HK2~)ze&U9JkJFn}00dZvNDB&2KRohfDe*8SU8I`I_EvddVA<
z#!x$?Q@lYwU1_NWYg585CF2Z{uvx7-ldX;!IieJQ2B_yaN#xN~81&3vtNP8I`dy(h
zLh;e2<E-VpR$F#i07$!fR`tWoRq`&?OVa7QNzd6+IVqI$7ub`nW|_T$MFl~t)aK6D
z^EU6KhG6I$)fKko<=P3$IpOudEm(<W*5i_Fq)JUUyj<*8t=0PWqm!qwv(F_QNmxC2
zA&2LGRUG+PY~8@oOf9H^yE)JU3gpM*jGWL`|7@JKQm;(9+x*@Ngkdk|@fKvRukhA8
zrR7Plj92=n^7iB+ZyVOcxLzamI%J!RxH`5ksDUfaG2}?_XRu1%4VTC5Yn9vFsq<#W
z3R(dQ>eecpEu!T(HnI)ZD2|pznQbj!hUb)jpX>Q^Q$2UKtQ_L{1EPTA9X$Q)C}^dA
zR1~ya4@LnsQE+7*=9Vgq>^coj;WNwx2P2@oTr7EE$qyTUxjxWx_;wqi>p5B@EFb)P
z9Y61fZoB$gw|EfN55PE8db|^Sw)z0SfDG@RhCOTbt|SPG&$wEh@l9E+L699R`JN<y
zsN$QhItkL{NpPb%k?vHr@}tw8*MgM$$=R+I`a#LAWqL5znZ1gS*Hn8Hk}c6Gb0W5F
zc!{FB#Tf*v{bk+VQS$BP=zM3#(K+NhM{oa`cjO=j0WGT6wAkl*?Q#hBjLQqdOY*`=
z?LiCadK0|kOaNd6fa@7ve<T15|L4_N!{$U<21Mb|_87YytUw@uOdpQ&eMvc-td&2w
z6)Oc;Jhhd}Q!=;3M83($=0h^_i3WA!KF!_1klV`Eq|*t^xfzl&3~yM#3#GRSzfB+C
zh<JIs`5)zmc9&7<DHH-QG?!767Za6rD8qmCu>bAl=Z!3bmLk}A8-MuI^ULPjhO1zW
z7D{)Y-!`!wK?(4>)ATZ!?)x9zm(3^6w26|`CI2(B`~KRomS2hugCrYDYmFJ%P)RZ@
zcGd;~@m01VK1pfFW_YBFbjN4|#dNfeBq-=ql$8XJT1fDRqgS(1k!ZnishVaB$@+hW
zX2OuQO}r=T8k0v5N0QpzXe5oRcLHIei=?$c3=&r)Zdmk^$Ne_XVw4$Yky;X8fzfvL
zU(#&U@=GsMgBR!>m#G}H)fF7kia=A$Ins!=yq&8F!kIAPYHsIfy`XDi?Wftf(as9z
zCU!D3H}HJD)x8mPozbSJ4q|4B%j<u|1}`Lrm8S;c7NpJ3x%ydUK$*CoCT4#u+8F5i
ze@X7BWgxe0ed7htf~$5(n;~uu9Y^uh?o7SQla^g5Sut-Sud%*Q?_Awrsju(tJ?|n*
z>xZN9ofjrJ3v;gCDo>8X*Re*IoScQpTRb_5;<C(BevS?#JtwD<#&9m6kOhB)M%~;}
zafalbD4^`OIwUJJ*?VPoC*=F+#A=VrT-&He)ax-m!LD~=CiL|9P~SG$$3`C=UTEWu
zb~wZ_wfDstMh#r$KR;SG9_3Ve8WR?MO2CMtPa|qps8WqSJ*wJ+MxI9%!>cS*Sme6a
z-xhgp#A8G^wT>QH2gc)wD};Zq2;<4w`OzNXGg)O)a3YxDn0@WQzWv*1hYU%Eb|@{?
zk`MXKDLWKyBRgbhgBo`kvZVe#Q^j3bBs>h|B{{yxDNg)a7-J*~%TbI|!Vx;V2Z}M)
z5C&0KpqO(Ag>jtXh^vE$vQ{XoIcF9?8nwej#i(Hp&m3YHpvkN8^t^vb&)O4#Py^f<
z6*8$26~6CYPFT*1peNkUqe1fn?5LISVDD|mvDB5&HRF02y1ofnqtKwayi*vj;BZE5
z4%f>q_?zqbya0O}S|=0BDO`ZPQAk<|oa4eXLc!>qt2=?JX&sVZU%}!TpRZ`}g3~Wk
z1f&)qC&pZWs7?GOLpFc3FoUM$7*Ury?_{H}E8u0M;Hy-_%Q2OaHfh8Sn4&(`JA=h3
z6`JZ6Dt?Yyls=a`k#k5JjTTP#NIUhem<HQu3*Rx4A0Ve5M<FUWRCLs1n^5qNDx)89
z;<$+8{mMBzwxn^7+ic+s<*KI<SFxVI68RTA_b|};(4<{5M?HU@oM8Kk3nSvyVGp=#
zXP9bpU+62!lv%He%5O!6w802K1ILC6D5!Y63phle=qcTb;E-`#5gankXTTw&z7se^
z>4oy)&Y*@#T?h^tx68m`%x~c1^<hpM@^urtEeY5<h~PZuxCB@f(8W1EE#R1e8RvKc
z#v#Q6RFkzO7sY>L2#&PsMVEuLAEYPBb3MT4$%as<s>`m<{kgn*C*c96rZJ?9(Pd)7
zYt%5by+a4y3hnebcbcmiljPbB+?^4iV8Y$Za9{2`z<_{o?|K81y3lzrZWlNYcZ1VV
z0&SR#TrG1PWt%gu3f?_iaeWtTNZRH&I5W8Q*Ow;B9{7LWYkiHSu5{vNTrXH>@om&n
zXUNzm;%}|DloKo)2@5p5*2L%FhJs5u=W<SJan*`zViup8`7Ax3I5p?W6{Eg&UIGXg
z@QgP6SiF2*>XCW9c>!sKF>b|qsf9Am@puwWL#($?l)Axb185;Vu4x{y0>EE5(LnCV
zgo!3g9JYT9G(BN-(rFH+x}83|F@`2mx>g&cZ82L#9MY(k1BuSEwGVS6a6xW_&Y}Di
zsx`$uiePjK{nioT$1YJ50|>M!|LsyUd3^dKcFrQ5uIQz)-bO=hKK%Xq<)_cTQ#|Ni
z|B0k@loi^j;Jdf{NS@M5N6__ac_tuhtd21xNU(p{S!y+b2x`aH@VVO{EkOmw4_d#*
z)(}t0S`Rv>1GFQ;CScH1#}GNT)++Dou(6J+7m;*o!-!9jOX(XQsekPiALBKAuB6oK
z&u;Uj+k7fkHP)th>b<5=8H1}woU<w+9u+G_1IyMXaCG1aXscC%qhNZXb*OJ^)PklG
z$yk5d3rFiS&P$QjbgnG}y=xQksvK%l{pIZ^?7(!TmbOSK#s=zVg^k_bHmX=RTe!df
z{8swCQm>DL-^^#LGzc`d+iFK!BBZ2C%8wX7n~N*tmAYmTYK{y_MKnv9+g*$SiTl(I
zhLiQ$_KJPF*ceKY+rbNr!Cm0M2td1mqF#SB!ij`r)dA@xA)SigOn(AI?2D)|wTQW>
zTku^<4ThR@%c(rfDYe>H)s|LHf8K-H4Y{?%UKkdXOZ&q}ZYf_6I37lvo(G>OJhb!7
zHdSQYgqs@_mR%b;Hg1=p*xK)LsQ>w-Wk#r&zQjvORI}G<oaOvn5}C|KEiIfsL>PaS
z?dxooxvydCqV%b|I$@g<uD&nA4@M-(3y{N$0X@1tr~!A<+Mg~aYf}S|i}zfc8&4Of
zc0fd?YHG8b7I!;ltwTR6>;ndDpV_RTJ3#WvbGpZmfAn!x8xRT*F#<4YY7E@wTwm{G
z=3HOz#<J`Cs~eLQc(NVflj^W3ENg!)?OgFh1r%QQhq3xmYpyKC6j#b&?G5hc^H{^r
zlyXVv%C&z<iHDj+=u>&{xK~iWzcGHe;i{87Rc{&Bgl?;?Oe;q7)t;H{T;GHXjhRPx
z9Pscp)x>qI-`=F}y~^18FL+|5#@ww^V;w<5!wWTZ(6qF^?@_0rMJ&MKigkY)#&N|u
z4dZ;KPQ$40RHsn{b)%2EZz_yQU6^WR-0o7R;TkNj(=d)J)M+@!<#ig)@%=gtUt7!S
zG)g&v!Ts+GH7X{v!2h5^O(0O=!ky|gj{Lq|^MlGULygzFUJODZAK59-Yn-rT#xd~?
znO0?Ej>9oV3wE%g!!n&`I72S;ZUAA#H>lz1C_#&dF|xO-5tOimm>I(p1r1ud0r_mS
z{PQfPV7~-e{sR?S)t9k@LK6cuFg2GEd<_$q{s}0?0i3tR3MiQg1o13$%$H%391{aM
zG%=TPvm_(|G?(#cG%<hGT2XHtxeb2Luh3)rq7_Mzlz@YQyGwm7&;mh{hii>Iq`5^K
zH1Gz_21x&XKWcUoZ^k~a?X?TP*qj-TL{a3&kD9H$mQ+&hVyQCzS4re81Wjy1uq1^-
z@FaK)fs-^1p(o{dq?C*(aiN~9_Gft1lJ)UOQezK6Q>P(V>KcD@DyQCKQY(pwV>07J
zOpu^Nj6XG=agS%(6G<%?9ClH4JnT-1w8a?eiA;tbp1Q{bhQ;&}CR>GZ60;lDMblGD
zEaNFIxrPTqts3md46J47J0vU=T0J}x>uyl<u{xK;ZOrPp7v?}xIBZx969rR*1mB5{
zcL`o>hE>39QLcY+Pas)5h3mz}yuBpx;qH!6+OVL7G<K&!rBJN9;gV{+8?Iy@@8UI~
z@Knd}m?Z8&Mf#WxL&a*f6x>r-3JLeT#M2%U_7nGDc*A-j#U%|rPD&ULl3?#!eFn*s
zRvP0$r7A{oXux%Gt0q;6SF!i%7*h-5)mGvu;c4R`VV-~3VI3h<dQ8CuQN@ZWIN}RC
zfv1*?OX%S>lN#nBatx~sNg-iRv`9^+!-9%;y^QfNvPaAo?1?pxHw5>vX3Qa!oOry;
z$ysAIc*!L;8awF~d!-RXUM+^hs$9cotw5+_u8>-+RwTJetQIDEYl0aN^d8>FJx;BR
zRbYo3)tG-3>4iiYqqNi{TvK3dlgI)@T$8a=6;iD+Ff433UO<jCYy9koA6C!Lkjsvg
ze!2Sf@Bc_tJhzA}uUVMW`R5NG-kh93Gp&B!eAryP`uO(UI!U~cRxjRex9f}Z#M9-U
zzHEPawSBu?Cz@`YZqB!u;}p)N0+V8vQv@n0c7A_cAi4>uG+jg@B6s8B<Ye{YV)O3R
zdV98d@%(i4di`ZP*|B>4-;Zm|{I4|(I^V9(x1R!8romsXKW#o=yjy>oP$zEvZT;K(
zw?A#ZoCR(nJm6A(gZD4+X5g4i)6Krv*%H1$aykO(n+v{In4N_v%M9$eS(EW!yFY|m
z_dI{O=i__0Rm;+??M`m(mwaLHg<a{}zVH}z$pHmV#$(Gs)H?;D-V}(s3@_=Hi$9&L
zHCj5W-^E#04hgS4tr_PmTjv8bSvUtLc{0B33*VX_?%UOm=jWR-kkiiBXM;4wG6rXw
z1d~~)v!JtJA-wr&;TUoC>hnLharM{t=l_36tDiO(zpXEZCFRZP&#S+zem<k=;^q3?
z_6)=<I|u_vVpeb!X3ktkt<FquRzD_;{%ZBh=JjUvJpDnySin2Mp}bY-3)>%0PA+-r
zE;fKPWJcYdZNMBk2;Vg}5RtfuxPert;Vts}8v&+~gU41uI7*Oi$p}d>GlwzuSXO_S
z=HPE4lpQn*%qlCkLn>u4ew_UvLkxhP62bpqK^z$`(FnBM+9L%8*mMrQNcdl)-2q_=
z21`_O?@Ldh-2pLgUK&CP1RkBhS0e`Ev>V|5lEMs01^~HP3n7pd<o5x9>k>R&k7Bn4
zuD0YH-tGcila&*2`3zG&!<5f3<u-o{et*0|&z1>2LRC;g_)2=uhN;k&Vl(`4Vw1V%
zXcS>U9z7rluF-kc8UVI1;uBtazd6rDu>A=6SsTU)tOj)hQZ*nInzSgO#>>nA3qXGa
zJVWzn=oQiO9Ki)xYD2Yd10B&-H?%XBx@WXOzw3Z{1P4fgpDS9stZ~Z0o(X@=TSI7+
zE7o7EqA@0ou1|%SXb&0uAnNQ(>dXdkwgVIiMK#*qr$TM|k~X$}7li2c4I$bL18wFw
ztsMgoA^<6=R^WlhNk{GlEu%?p+?Hh<2+-qLjI{NF(ru+2p;(dSN-YkOvg$aS%W*Iu
zicGD`kVNombukMgq#-nq01JNujR1{-PUFN0bjr{h#eG|P!cfW!V~fPxK+#d%_tK)5
z8wzG$a(HcE2y}?10JMnh08M~bcJ=>T@~lpgx!6S8UDUYn+ozS5V&A3JIE}k?T8+cO
zuW8kXJe>~i1eyh&1-XBsal<ymnHnHIcu{O1!)JzVV!whPiy^uUaRYxEFMw!qJp%Hh
z-T=!}WkR9RI2_Kg&WeiA%d+q^ir8))T*aNe0xph_?U88FE47j(&Kn(dfItUY&#mjy
z6{QuEqA0?(w$iLhJ4V9-jEPgtsSs1Q9D&km^mA9>9yoEKiaJRZV2PS|1l+@da-34{
zY!H-BMf6^01J?%NMPh%0sZ~?urk5aR9Dm<fiNDj@oid>G15OXn1OdnZ$Ok{}x(+jr
zbr3d=|G!5QvYXLF3-dmj@O;*DHqQSi$RIsJZAN3n<C{p&k)vGlblTrCy0ynjlMzU0
z7=Rol%84UsDL$+is^b;7RV<G4(KQfgeMe<4dofD>yJc}t;cI_c6!ZtZ9-tWzqS)D|
z!4wfYz6e6r!-MdfFLtSNed<5>sq}4^F1lFyu1?=F|0;dUKc1_(%+*}J-kHm~@DYyv
zI$+Qu3(z`YWR$Tyed9xwXSue4<DTjMISFLW(t<AYV5%rb<Iqu*v$ndJ{oTLce#Ay+
z>c_)_IAMz~hBSYA_OPX(d+rZ9H-j^w^2WF7?D4f(JWKbH_2P0RhpJ9!u3H6NN7?Lt
zjMiN-8?Xi(YS0+Kvy5-=gm$*w&WgxgsW(94C53B`*7mfNG+UhVpo?SehiFD9?S00&
zZf0=N+`s>rc%>$*jc)EZAhg#~>>&~bd)t-sQ&`43E{}f;i(Q(!@zdC{syh}jei)(i
zJB$H%_l~#emle~6$I!b;$Jail<HBHFHZ`^01qP>2p%V<&xpS$<=v;&|86m!R=Va()
z;wKY~kZD)(kO#>+5gP}WvlVf0sqwp&Ypj3{dWqr^ABl+_$h?<y^&qZXmkij#9nBBW
ztl;rxhiD=)vM_?_sn8Z(a%a60ceYDGt8>s!AC5zq*|bh<`hQHuVV6NT850CGFg7=r
zkt`S!0Wg>0HUuV@p&clge;hw==YN0x^GPz%QUn{n#*g{G_s^&QJ<Tjwg97mMpWmP2
zxK<#P#}m@aU`~Jh_w?)2zh<Bnl~Jem=a1h`)5y;(Lk}cFX{|AX48aJH;bd(PD!$4x
zG$cRuo+YK_`CH~zzQ=DE`8MTSooUKU8ubq9Og~9xpWFp)#tTJEe=}`h7r3{mbcd-7
zm-dfGfBG=~<b?^&sx>{dC<TBkbgFPRnm#q@o+h0#q$h@S%tRVUpUiiDM^q~9B9%nE
z);zP!dW~<R)+QKnJ?Wvb-7L*BfyZp(WzT9zSi)hmngk(QNT%fcEy^<COSsCLNJXO2
zX6oao{3w=X-gu2Ke<O81@Ph?fn(qswIzk-i*I@)2-w-8aBHw2$^d<h?NLhlyGSM%U
zE6WtJq|)ry%2_D;Q)z!;9)UlLv>U%GJEl9yL=U4eY5|6`GS5ARgwvnFvVJBw1kb35
zA;W<0Sud~kC&_~ZKRXmKL1>A49!1ic{Kv>_qQgk>C_%1!e-uedr%$xC0(4O?{XK7X
zV(cw5i!d17ri;>&+`E|oCP^Xhd;-jp(viyYClNo4^!~+K0Tw5~C<m(QWT%C7Filnq
zDrkWAQWA^s&J^v7b61w+0TJ;EVUoxkfd_Shh$vwlMt(jZ@DVE#vnH^fnua;rW7gLV
z!FdsI0WC_&f2>LGqxf-^eD>A5iU>m6FyjL~$>$n~hCJnB7m97(1Tl!<`~sp`l^`lD
zP~k#?*K0g#6cq40Qa0L(K<?v#vN0f*>aryoW5|S-1|p*gvUsiuIN02@6By6<f?LWE
zO3(vr_RE8sp7(g{Uow#xR<g3wym)plM0I_DWRw+3f7aklbzzbGXyFl{RBNwuOmsI9
zWF<=51rg>l5#|CB4{)4P361`%9p{KOI?iLk+qwhwPH45@<%k2l!FjS{nmB;Pxp<S?
z8KHt&x9Q?iCVs~h3p!oM`!tQL*tHIrj)uY&TVr}Kf08w_^ibzOl@o|A2+e^Rr;*>w
z46Ux&f9Zq!$FN!cC0S89z=#+de&suyK8KO%;grL;mTqMDTj>V#!wCWNAJ!9@7?78s
zAX1;KYs6B*VVA>U=7<Ay-6Y&nhq%vm?_v^68I1NI4D``oLZb_j@LUM10K43RN4ngC
znKUe%gLwqQ*~lX@6=_?bY+C$(6ln+9H>_Ccf5sw8*Rqa;Uw8C`rZAC`rZg#gD7i|>
zDpbk}mx3;ImE)>ON-`eHG8|-(xYWHg@&>POoFix5cCO-j;T@7K$1+uEs)K$XcI#_R
z?4z2N%2lDLv}i|1P@?^8kX0s}<u#3X9NoB2_v$`S?q@=a=3hJQlHHEGb~HwM_IW?2
zf1*qAM{`aW-98;D&iL_sCU0$_x9Y$C{cpK!q4oL-6hz#x;xA>zWgMLX9!1hGIf$hb
z?@=uMl8oh=(`~z?$ETh46php<R@vsPl8bwv6wr<(hkMRh<tS!qN%DJkh~-(tdF#ct
z%jGLg(Sh@3EMKj=)@U~9Lw7)+k1=Q|e<JB}TGIdc+=@57TS9;7#LV!jlZ@*$pM|<P
zvuavCyJM!JbI0h5^DT#0#;u;a6ZVB^cq>rm!;ay0hh3gi+=yPdT|UQOn$vG^ZYZ~9
z<xkqV*gx&G(BS9&13oM{%G}z0?MbKSwe0XVF$N_RMSW`+VHcwzeT%MOiL>Zbe|Ghk
zC@_e^I9Q_qKK3b~?9Lh7;Ci+T6+rlE^kY49thTyQNph-y#(RtbJUw2O+;1`38@-|2
z$odJfYC%CMbsn$|^Hxb+53$zEBT4|tXC#0RqJ?-!{S4HkLRJ2IcR*VUP-B&OG!gp3
zjs%daxJjC?=T`GBPNkSeckaiaf1NP1P;@A829wPry^FFfc};Ess!3jePz;{t)E!S%
z^w)OA>jur1uza9GcJWDZUYis%gqQqg8y{}s({DWvtuLE59dlmeytDW*s$7XXOLSdo
z`y~U2?}592N%3UoBwdPeT7myigGE`gdW%EKvew+m`aZS=F0%}VJVAgNf5WXT&!J$p
z#>8@q1TV6T)(8T1jOEfVa;yEKy5{ZA=@H%Io^J2gAZYSoZqcuyKuUZ4CH)$u39g5$
zN@WFutfjJHq(3FNG&W(WdpiaXyDcH^+)U7}H-LwAXcPo``N($H1o46Wt_ku^!)u!T
zMIA2;PC%<Be^KFf!g^=Ne`|vF!#iFRz6W%?W{CH9yk>}(J6^K{Z|`_LP&QniexPh}
z$LpYIxZ^b&i0s5ey1>RpV6!3gBNxmMtrzsn(AR^aMMSjc+LW}EOS#3i-rlK0PcF^7
z3!S=M%CGO*bq4#uKH!6Tf#{6D);0NadVy%kep{nu1Irip0;4NXe+It#NExv{s24av
z{h__UnFF~QGE`?ft++KTyMor>1zFOJ8Ey|)l_W?~EqHlb<-B^fX<zQ}Z0QCU_BGTk
zTjr~%tqGu$Y3WVNYmwSLHjcCjU3q!K#+3!!w(%9(`wN;_n@s9^m!tBfL~cLaUrPo$
zMT~9%vFi#hs;k|`e{@rtIRTY+Jv=XOEEhtstFV@oGtXFswpuGl*ts)%ZNYA<nn3Id
zfG|s8De>uy-zWtHC7fBvG^PA<NZz@~!xCl5B!5t4SJQS^B~#i@ADoU4670)m-EMV-
zfK<3kSasJA*xHAKo_crJEZ5OX2VLu-tg%+8jOmy0^#~nHe;&hS)^b&MJua))cDoC`
z`xl679cnCqGUi^&jyVC|Dc$uH(&bk*TrOT*WbSAvG<DaB7loc5cKNqzTH)XKSAzX;
zEqL`jTjlPely5H=1rR|ksznZjQ}89lqF8~H@~iJvO;{fkOHELJXe>41nz~+FH=$;}
ze|$VuqMrr5e|N-FIYOFw*@&m2o&hfQ=IxHKg*BHVv=}Mgbhi@^+IFAYrCe}bpVqEv
z02#9keYLIL`+3dR)Jj(gduLW|{Di5Q4~6L5!Pud@fGl*@Ty~X=bLoa1r6e^4ZC6+C
z6*uG_)zm4((48-KrJeSk=cIT58<i0953{Vy*Hboxe;t3Z(|$Sj5NW6WK^9HdLQv(-
z+%CQO>AcN`(3Bl7ctrv>>Ii#xS2~?Lcy*;;aqNCfFY2R6`X%H2m?M_I_mekzL<(Ln
zS4#Lc=<)2T-puN5W!;@^QGScCn`|v;RZCjU4z2nD36<4Cp<XuG`}SG+xl@ZPOO_Pl
z|Ktb8DOfZdx@!U62y5*`!sA+_)#)8=jRq^cpXu5C-PlwJ<WC9){}0Yl;yjmeUMLd*
zFqaWt9~1#Im*M9ZCzl{7D4c&ln~BqJZ$IDBBs3*X`;Yz&e|vqq`|A!*K`IIJbHndn
z@B9x^Fd{4;I9G%zJ$(CP`19@;OtvtiFebhzI(+*wpq9@`i;jX8sn$w0T4WSh%Ry^L
zg#RgOQNB>BU}o|R56oVAEig38%prUl8Rr&YT)0RIYost2&q3+PL4kjM<zvI))7i2~
znYAWNb4-n&D@r!XtW;dVFptO7u422Ck=p_VUe%s7L<+M)d$V)PnrbP6xKx3KN|WYr
zPO?icWPS;<H!bxVWUaN7Ic?-By^SQ=Fx*>oazF!k8--c2*W>gFmamvwWsTYZc_s!{
ztg&hNyg=tIR}Pa^ZnS@+%-P#dki^bt&JxEvYF<lfN_*$ieT*godWhH6>&iJJQaj9#
zNu-}O6csVNTLt#AgQ6qO*p8x_)F2m<aKSqbuD5wEIe`#=q;8}p4g&Cjx>1bu<}#-m
z1-)qRAhdS9iy?W=^|+P<HnNU#2&juf7dM(1V>&l_ipK6;$UA>PplC*S^5Hsc)PCbO
zaWnQ#lEZimFDLII@YZTsiD7&R(`{Pn=$9i4lLWTZqGd{*3))UjxI|j@3jNY%lBFyG
zLZ=PK#mA=}%DcWuG+(P3(%2q@N$khHG%ltVRcN>vhJ3P;0t)m4;`K<HMw)SSkmud!
zaLO4{^4qc!9Fl(?NaZvYL{naNmX%Y?5AwWEXK;4~!Nw@t?kuniXx4E3;m41DC=8L)
zLpTAwwL?C9Tx)@|yAFfwj=DiPPP7zdr&nY4l6Tj*GsMi5p5wd!2>x)-IDA|>PL1QD
zdC*t?iDySo#^N)j>*L2!JG?UKz(Njh<tiO?b76G*T-bj&m!Q#WLa2_&L~6V%1yafp
zqxFqbAjH+DK(dO*K3Ol3M2_jUGSjJ$^~qy5mnEsthBPRq7<dMBHpR`!0Uoinj94M7
zvjhSe=FGMiSb}mQsOaYMJUMegHBk`R@Vcbl8@5#?EwOYg6JFN`GVH{Vq}@9$#CMKu
zV_mZ+z6pQW^Q8SZ$G5yH)%t|IGk&vN=63N?Wl%iEjFiJ29wR}$PTV>~HcXF-kPXrI
z4Ui2hQ03(q8Ot3s>GuzjC7>Z`cSOh(^V*rr1W0dSjN@h5Uw%OC28+Pv>yL~1`XW1P
zQ)NF+Oq<eFQZAjZ0(lMhWNGA61W7{`U?M1*IhKFwAtmwNBYa!1t{`=&2CfO7NoGOw
z#HYoGBv(k{E6ngcn`v#fq1HK-IGHTlmoh4sNO~*<%Q9Rog<=_$D!NpgO={i>s5a!G
zn)koSsHQ5=dlWm!yP_1$)m9aOW|SG)pVD^`sco#)YGYhjt2IMD<RO>2REl59r5<v8
zMJRs_uyJZ@MVNq$CXoP<KyJVCdp`T#Rbu*9NhWq@R#G$Ay!AgLvznOSzHF;m#0y7E
z#i77n9g&<<J#|zjIVGIh5eCf%Ven`W6<9x@ZN{`wx+He~6ioWlw6L+b{7o+0qF{V<
zw999E0tzKz%Ib;~3^VdhxU!b}+yNddvdwg@qtEq!oVbKO8gts1*3HhxI8Z#pT#BW6
z>gvRUhjP$mSckyVrm!!PEUM@adhv&LY{|t;bEcN)<@tnYo#OoDo>pj=wT)K3cx~g^
zs$~BBR6L$kunc=~#f5{A$~o2$eY`RXqJZu68fVza?Xvxr`Bl3g$d~)83Wc;;1Ev3S
z>lx~Qsf;Vo@wggLJHoH}p<g3u+B6_ryw%{pB|%i%Eq=6cLAW%S^0H*opP3uH_FYVX
z(D6{>!eki%=`V+E7|%}6Sz+Ppvfby?`&<`|`Ye5OI#kK$Eg{(B%e1z&B~ZFHI+^mN
zkw?gKPT_ObR<H=2dla8xcCF8(>}c)n0p`(vB(Cx80j6Llp!_Dc2LzOfM%x3=zY)a)
z&%Y7nE9c*+_l?fKh2NFbje(7p>CV&dme_SbL+v~KZrL4h_KgVcKl(-l^~IBKwEpyX
z1EZWcack%daXkn-LsYNZn&;{BZuz|4-tq#Wnb7U_AuN&BV2Zo0D8zFOOz~y+r3XxZ
zamI+_g$8V51wG)!-IKA=^~+!tfu|GbJP%~1BNr^y^JZicQ(~Ldo1sZO8+8$z#Ad%A
zoaBx{sFLII`pS58ppaY9@ydYD4y)@4RUTI=mD_-p+(H0i>VMF34QT07a?D35N3bv6
zwwckkft_Z&LUV|=V?M!hiKVMjUoMq@`8xoVwc#y<4x1s%0Loh7OS#lTjt3|?lor&D
z*Io4pki`>07WW$z?AsGVGmmYphXJ{?kcik$B6UR~H6xMPFOe{W8P`dquSle4Bs%`|
zl!K&PvjdAZOi^#j>6vtd$xgnR?T{%!gtHxQy7?{xrxG@J{%*E3joe-aLYcpR>Ry$h
zdW1Sg=pjeD2%p#lWkvA9K2)*vIk)g`%?r$860jj^3aXCErnFuIDx-AI2_~vH1KICR
zTVW6ls|JM|LP4A$8q&U7fFMj7BA7VPpAf|ZK@ddw3J8LF-v|U@P>Kt8YZL@=-3fyr
zYVVGNAbtk`A(-NRBm`5u4258SD>xntVN?)l%&qOL<$4enf~ejE3&AtlkA)zLXy@O8
zg<y*Nu@Fr0A{K&|rQjgN)s2Ye6JCM<O-n66`lAM>sj5R}f9rnc4<rdW=_Oh%2W+jz
z$tgK<a13r_XJru#k{aGEj(2fM`=GcPEfUHQ-P^!CSOmo<5vGp}8^9uezk_nyQQ83(
z!Omi>xZ)$Fw-vB#oktx27CBB7v-L%`fEAhQ4uu^<S)Jg;g1p>QF{r1F*oI&8Fe?2S
zMJaURwh3E`7{fYv2`YA_-2dp(P0%Iqy#F4wlrQ&$cQ!t{l#}YQ`m5Pfa~r8y{8Dr_
z1>kDnsxs+?h*CXb8EjP(R&pVSu|Ec^CTlkI4kh8=lfL7Zai}E|12r%>mjSCT6PIQ&
zC@X)D<TecM_g6Sw0_1dhka_@%1@>ll1Efo^U63lsU6msF|3#>$^{{8`8Q&Di6`mQ(
z5=DNZNa|ZJpT10d`sv}*`@_dK1(RVGg^|<y=Sk25PM?BPqsGYy!L$<7`(M*<kDN<Q
z*C)DwbEW7q^xM)mK7CNaQV&5-1kxp*J{W%xX^fI|3A%I~L;aO|jB?~DJ@f6F<XFx9
zI6?QqdWlcJzyJKelhT|y?SJ~O{_get;qQmu1k}I)?DWT95B>ot!IYJg5XexQ>7PHR
z&kw)!V#{QxrupNaFH_g@YeLZqLXm5Y*q|s$1p-!U$CQ7RP$bKMtR|Q^kcCLrqfUQq
z(%(OQNEi@4{LSCV$As$(UAjZ$vXX;FXqh5Y_iKc;M(UM^L5Oi2n0P<;@OR*^H~MK~
zuh7O@8?LcMaB^IjG#CcT6k8zb4<_sEwF2LozGv+KYHXeJ-^DgQj|%pAEXeoqHTrbK
zCE`^2K<X!Ao65MjIL^!T<AO#uqC$V#cqE1uQksFW*{WI__!M^(s(3G}8N=#foH1yl
zV;g~$w0`oCTv9xX%!+2im(_?OcWC0{K)8p9K@f>??LvPSL}rcrFzKQLh5MxW?4D^^
zTyiuyYbjg|C7K73YiLV~2ErYRwztbhXmq?(5(ltV=Xgbj^^(?r3+-95ib{Xv(c-1q
zY9HS~Ovs!uXq#X~p`L5eINtX^{@{nw5Lg}}87bTj`7RGRKT8Gp)>-Nyah5${V>L)q
zA`sk7S^y=4>;WERAKhZSYa<-Rt{e?ta<uRpB=?Wcsnr@@;tNG1Avr9F?!zO@gZ7BO
zIwfTLxnd-206^~qK(A<c8UcUMR{*dsz6F4jb@2mv1fXdN4&nH+L&Us-mTDgh1(27d
z;adzmeyd5*ZLJj-15Q;G4iW<nQ33<6SpEq5HZApnfnx#&{U~$Qx|H6iDAU`%cZ!3Y
z8SfDyeS~?NmU><-nB!)UH01wfJ*;BFxB-Jy2V9Z&DN-^c@hk4wGj)GwS@HArGb*U9
zRVhD&wCihKVIJpz?yIa=MVGXz=6cN6Gw~^=C%H09t-b}HWJsL8hpQ5#ZRV?!`)STv
zK1JnZvA4=rn%vB)p~^-_-YYLv*_StKsIo8YSIQHj@`nTVYbz0l>TUnt)tU({d&!b7
z?-@hhV?^J3v5F*b=45}R$noU%_(w$zlicjUlTb4;YzRM5n6+v7^)lw&tWqUUhPet4
z#ZSr^RADV6XPJ|OG>;N4*Wf8W#Hg@)U8NKZwCm>^&aQhTB1N}h6}n<>pF~H+l^rDN
zMFb^9E~(_;1^jqF)0Ro=?(JvotlrOH)2VJHyZJNdV9RBH%UOR5PWyc>uTW<RzurTX
zWnfJ&@fI%z$qbzYdWko9F=*x}ee@Dnych+O!(vwiz{`HvH84(&b{_V*H}%qXaPljb
z+en5|9tow<%owv3N-EsA8kQ3H7(Ua+<-2WEbQYl<J81N_im&nqc(oKm`00itQHo4x
zm{MhhQ(*Gx*OPyrGo?}y0?@J+K753-Dv=(cpbM$25CehSfCya}s?$Br(Fadkc&C}!
zxs7@P+G5JcZ`9LWs3%a!K&by5^)!#{IAb%PU}6QAw5g2CTFj%Co<<!@=y^`75-csJ
zI9Ecwoh6h?vOD!`q{!+GvXQ)MhHRtAYRA|*OFiUBip+m8I^azpfdS5^+1L6n6#4HJ
z{w$q{9vOdPeL9DDI?p{<Pvx+3Du)}Qda^<&1`f|yft#(sE3LrWt>9D+D1F8Xz1a$V
zr4{;SE0hxqjcKgFooj&`*8;!VS}03r3SrM!p*LHhue9QTwa|`fbd42yrxkjm6?&r;
z*Klr1sd|4k0BZis^Iuq4vBI}?44hD(WXg>t)t0`t%3Hk4E4uVLHQZdzzGX%nn83p9
z>8kp8UXTzIL5S1?28vVepc7n&K{~CkA*Z-yRM~ffLg{d7Yg@m7*HVPSArlS;ico4g
z*XX-YgvzH@bO%y|ox}qv!cOu9im>Z_CyJ1cTugs}yfbyv&3Yq6*h%~96k#{t0Tf{`
z@pg)^mv}l$5%ykiJVhwAVN$}q{j8n!Ad0Y)dKZeYx8!z;u#<QIMc7Ncog(Zdp3YH(
zedpIwvs8=26Gy}MhjqV6Mo(Vj+MHs>jlQzq0~8$-6j!4M+~ftBr|Ydz%N(6AFy*kq
z$1;D1{?FVvYIKG<^m$geL95dt)S*Ppq{japcPIhL-}E=`@Gjh;kM9c1f0;XkYq`Uj
z9h8ggU_8-&1$MCB0vma|X2>>nuwMGDv(!V5X9uP9uYlN|KNuN<{vEJGn1cO$ha^})
zwtR=XsIR-AuUpaA-L5a4?V+X0P8ZcZT~L4bwDEPiWg>?B%n^ptyY5KD_?LgItWG-l
zUjx$XkhDP`t7P*$=ddP7*~HyIW@@&)d5!uk<55&phDycq{|)Lnvn+|dLD%y|p(SX6
zPclc25_H=hoAiwFO=&lj31$B3t*$kc0*8bxQ{-#cqR**)s~;L|8-6Obj6_C$vc!K;
zbyCt(9MGWB#XPpyp|0O%o8L;CW!svJ;_+aQiJ$ZTXM*5m%ID0em06)pnt9EoW_)JO
zFQZ$n^UNWNwCcQJ^+s&Yv30^(W>5?oEtFefvyykfYtm{ilh=G)BBz6u>!n{y-)Tx7
za$xU(wzz+B^bX{m;ELj(Igq^TBpzN!-gS~Mkau10b@J}&cV(SqI5WGl{M$<(a8kLp
z0>lmHAuAYGWTva)W{ya?>Ux_iH9VE$p-OR1*2G?na7z{50)en!X+v&O7Ue9%SPI3Q
zY1M}O*?-fv`xhuME0?iN4HE-3FgTatG6oZu2s$V>0urg0KsqQfe*tiD!W-rE^7ds>
z^gz<5k~C<|;*=5=)bjRi`E4hqwsbwwg;E00<><SoyV-vN<!OeZC(6*J_MaRoeg`yN
ziY}Yqq4^MYnT~$uXK~yT9dBiPPtkAXBenhSZ$CdsW?G71<JWkbf4#mv{Pi%4fS^S{
zwfynzAwED*0=!-{e>x!@EZ_fJzC8RgldX#xSlY+;ugf&@D{Ij!(xSB1m_dt5l4^0W
zHVBB1vKC2k(vU`wBt()B5BpD9W-~oeFY$?HO@U`E=}aSyxvW8LCoP?4e8r{l0!aMM
z)<`R0ypzq5RxxBJe9RmeV!=2bmwbQ8d>d0GvOEj<K1h}9e;(!4EEP0~l7#X8Qhn@Z
z3CZRNuG+z}whwd<E6x>jEw{Kfy8JTl(nXwEbWrqBHS4p*DP~3qC9$7BjLZ@aBP8MH
zLy#vuPa9!WC=w&_WPFx5@))_$^QzVnc?FA#3#C<uk+kX&e@B`NW?oeT>o>}8+RA8{
z6F+cJsW~r#e_m0OPwO~|RAh@HnjxPv&1<|wp)fTJ%+uYCVuBVH;RhRKS~VI4IRK)o
zmq1Mh!WQnvZ5zkOP+H0I4&%h9&pG7d`t&3l3ZnwcN0M+D=?taOia?2KT6-!Kb+%DL
zuf#8Ay<}Mvgfn5aEQzP(WUdRSIxURWE2_9HjI+Wy6Sq?P_1<ngRd|n8FN1lC(OT|P
zmw!AcB7bA8-pqBZX#hp5aAf5hDx3jodeB!6xT>YhYguC@1G5HG&_CDkgq#8rUo}5R
zEmgZew)bN7BHy@~Yi))1HN(TnT2Z`FZTZ_3Ybzr{%f*n}Dh&CGjYljeMJ@5Uk}W%?
z?~qMdTb!rTAnlVzxu40&wDbhLMCk&eRP<aIZGYQ*0h3Fl<NL3PWHRf4jl`>}kXvTU
z91Th;4anvB{wkG96v3%~dAo}Ng;e@~-0q?WAr1a#+TAN8q~a+<+gMF9m|DP+m-}qZ
z#I2ozG4WJ6PqwcTmmu_YWTV4r`r7DlYGZo$@<@46DCh229w&-hvcl`d&6BcLD641c
zQ-ASjtv|i5>?SXrW>H?wN1$|1jnhO~%W`eET}nq_;3PX&cMS+ZXjN-=Ny^EL{uK66
z4Ql)n1Zpkq7KQ_XTJ=_48$uKDmHnq!!Z_om{9o$BbF;nSSOWC#t|IA;$ox5o37iqu
zRPn*>l?}U`a=_UXynCI(k`qVLjLiYbGk>}e1?P5M0KD$O?3(MQu_qQl=b&dA@mHwt
zmxNWirjjgFm(UJ%cCpuER!Izb2x`|VcV4H=#UcR~<@UVK-C`b%T2Q;wX)GDZnX^VB
ztM3Lwu@0*F2C17)DlbR2cr3G>xW%qqq>`E(W90}Igp)AD!6W;Da@WX|-WvPWf`3yE
z9+O9GUp(rl+8WKeufH&t3;vahpmLq(Ca!s`q`XUUOhuauvxYGx2o3%mjCsD$BpCB_
ztpQX*dVj}SGf_Oc6}!C7Oq7r0b>&c4>QtxMu;tzfR3h`0+IM^H1{J8B{9qVpT=Yh0
zZY1{M$Q(#dl~R2mG^f24I{av8Zhu<!PSD&$^}Kbih2sDz1?hRcV{jaJ>PFl=`x-os
zVZeFa8}T?s`DUs|@HiVck+9PB^**kt9RjFZnvB8ee-?w&2B;nX(V1#m^<GgiQvLX-
z7}q-&731t%qGDS2PEj#YJrfm^u5(nV&i}G+Oe0%Sfm&eOBxHbuFP-lV@KMq@Y~Kr#
z`O^As<A$f9{+hbI0-FuMfb9W}1g4nQ=i-2-o`yhrfT06eTn9kelaND|F#<=O6eESf
zLYL@XQ{zc{HchQ2hIWSd+yg5(Bb%z1?LR0X0gIOpKqwX_O<5UxJUS<NmAlwt@Uwl}
zxzv`By;p_<UKl%*ed0?10bjB6MHrV$Kqw?MLB8sS){JQ3deeLgJuJzf9JNeX=rvNr
z0%vV&B1>}W;Q=*m#zXUp=;&Re-hmP<CGe;^m!m)^IDZPNuWtGr4qDqTERWrB;GyTa
z#em+8>nFKguZfQDo3(C@a9A7Q?M#sN;e82mr{yUv>*PnK9B#6-;Y{Qfe+%QuL9(!2
zl_odD&L)oDA_2>}-3^(5n<ZsXIN$!}^g8Bpl<xwSPouU<d11BgkGdz2PBI`TTUiCQ
z0y6nvdVideVk1LlJ*cKS>VD%uwTrw7<Qw6-;{e?X-m6fs<7>6X1!#Cry}^Z8za1JZ
zZJH;nG%sm;)ZqNo>=JEMlBwly89HTNUiqXNoA!GxmPQR>>(2^I52km+)WUT<oNvo`
z=DT&re<;l-702LZKXUL<{M6YdxB9Ff-JOM&mMgylrK892%rRHa_Vr#kz`hyTz@M?X
za5{7F-JvHdZT?>d-(TMz{sy3o)0go@KNA5pml4k{6_+hSD8qjcHrT@-Z@=Ej!n72@
z#-H(v|2w|j{e8!mV5Jwzo8iwdckzbNAhcJsGT~@pU;i3D-TlT<M_Q7m<PVg?*UtkR
z`B;P)BneSki`WPeBuUl`)&`;Dtt>?IA|=wo%qv~AcoT&H%_2X%dRRm>lC}dmBFS4K
ziP(_HOhV3R!bE>Q>qlb086oyZWNQUj#gUb4_!^&fMpy^@Y2EQb1EemJRtd+#(`Auf
zl&5jRXtzP0BsE9ypa+PA64hqp;|iI#R23~2L#Bo}Q8rjO@h2qbMNl1DloBpU@AYt>
zj{|-KIVuzgVNbneh<3XknjpY*yfHmov!5d(I$=?Fh&q4rmnDv5elY%aWiL&ZKpJ<8
z`>3Q527#lV*~hJtP}&RQ9j5@e%#Trm>@9y9pSIIQ5USfscwN|p4WQNt3Ed`uSze4K
zlWlz`+~=|w+Rf=B(QQWNn_pEOj6ge+A)P9f<&%GTC+iqvYS}eha-4LKgm#VNL_DZ?
zl9NuDtZjc}CCG8QIBKd3ZM4$0+8E~|);USb!Z^>AId$&yq>J%fqg|wMQtz7?81cq`
zZdy&k(fJy=d>lE9{6E!o#<}Uf$(&82dFCjK!0_`qTwL{xLfo}tp}KuBS_|(q-xqx>
zvzB+$&>6IR4!jgK+!%<MctpMDfe0uoP}z5hL=b;zJ@mbw#b<~+Mj{OPLL|bTey2zT
zwWTQ1zH=nPsNNWfFtl}%h(2!jD++|Mfz#`(?G{87PIbFb&*xQ$K?LVJM7!*!v=K_H
zP7CDzdPeTcgWvPqGQ{g6HHNf9q{b2ViPRYCWj)u&h<$DRbV)2*0E2KCIN~h_r$l2M
z@rHkcQzABwc;etVMJ`WuN+_%N)o5mlg3Ha@pe4CgbR9>pCtZmKN@i83D3N}RJV(iF
zk3dmU0D@EGOG*rKQo9t`mH#!|DQXzXQ&sI%R((8{uyREpr?Fs!JWroaQAWNMMJ*H!
znV>|NyVWsteady66#Thd#~DLbU-&3)De`|F$4D53=L0-kX1hXGHls>Z_Ioxm<OWV+
zH7fl#We<u2ja8}id`OV<*jt4XOT)Aw>3-dWV}2&WB>@*MA>@LVl?n2ctUZWeU{`79
z03eCR=1bRz*{tO`Uo*>iLU3ukBJ0dVoz0BPhgVuBlCyQ_TaQOx!NN@Ke*F2T9HM_m
zU$Znp31W9X;fFfmJcXIC*V33z$yBzOgp)oQP=E{rsksQ05sbNeHh{6DYGjf9xjc1?
zQ7RNIg&|6J6!8Q_yqzLN4bqy9qCG*;uFAz}yMi!ENxPJb8%eo=n)0`?L>+bfa39U^
z!&+NPX{YJa*pw=tOC&2R+Eb|Lry_q;NqbdLC4o<-t{RobxWuRQL-hf5&OE^`_wAJW
zr?g$AG>kuQ@@n#&Ed_yz*QuZ}qR&elk7cf%Td9Lm^?stB4AbLmy-(TDsNgsAl~z*M
z7Up~zbG|r0i3H^lY;*LS;8fW-TQ5KqJ_L@$YaVB};(0FST(_AEOxOBI3Il&yH&T_1
za$>KiGp5@$jB3?XYN|IdZmg_4&gOM<{7e<&KhRHV%r|RQXW}zgZe|WGt_on8f64>Q
zl=wdu-HFQw#aLBPU0eH1b1w1---J9`ib)lD{I${=6qLG6t(r(m1UwmZicRVAx*ExD
zbaD=zN+}1zlPa}B;pQi1Gw^?sFIBR`7(T=%-4t{!EvE@_c&y0c-cxF?gQXP9GahH#
zw3jdpIbt6S7sj~z^sR0Y&J-uD2&8!sSGCumFdnC*DqqXiWVMLNVF`*nI`HdCkw6<3
zjP6R2n#AhOyObie(!yD{V@ZP{?pTUs$QMeH?CDjd$n`}MGB|NRI7@%m)c{Maf`NUP
zW~7#6=kT2?sV&hP3r*l-OP8<YvM}c)=jD%UD^D%`of>xx^nBlrN#Odn9i^pY?_sJ^
zPQh?_N#tcyC+~JOjY|wp?`XCd;*#@fG~@<pqNlwa-DV`wAmjeh(pJb|_mozPxW@If
z^OdFNbhZ@fBdx6&nV)|ENh85=%8^7m*Uy>qoXR36C;Il3F`(=uHSY2bdjLj&+p~AT
z5vNmmwPs2!lxjTpDzRQ^2Cix7gAGECr!<w~!kE<N?56tNrmx?3z(+4o{dL(5bo2gv
zoRfUUb1mqNSy$Wt?)vF&IU(8BFQ?3$ys2W#HpLT;DQUkW=hA;;9UEoPfgTSUOHm*d
zE*__>vC8Y(K$pCiR^7XQK?`fbyM(!5oY1hRzhH<vhPe#+LYT{*{x)GQXX&Oemw|nk
zFc%aG9rk<|lp}I$yg?62|CJ%$($TG&#Sm}o=oZ~*LBC-`H@oOdV%gr&Wr#a;bUEVf
z9bJxyXFIxF8RdViDGixw7dqLW>*AKaFX!ZW&dEwYwyEE2A;$?9v}AqYsa<^StXuoJ
z!1F|qY2qsoveUO6y4paYpqafWga4AQHc-H6ocq78{Vsj&K*nKuq2j;X)}HI#%BFKO
zWnF7~ChwI*<}>zhY<nP+94@7IU9Wp<TDQF$O%~R!Vhay9u!SqKP5w>E%RfPS#08ge
zUMLg+FqZ+VE)$oSN+`>Ject`^>F)<wgzhDD-UuAa5)vhg+$@eZfu%}sD7P{6omxG$
zLTQoI7e5{MPpvvX<`F-SEhVdJzmWSHNAwr1kA)`F2IicopoAt4(p7nhUYZ4zUx~-D
zNKDj->|-5Y=UivDkn$4x>yj?Uxx&XT$EP;@4L;;`6VcJq(Z|Ps?(N~}@XN5ER$Jly
z^yT}3+KmRGgPK4&VU?SH{dxL&_|#J^wjqVYG5MdB)347{*Ydd-eUxbQ(m87vMh_Bq
zf|;Bpqql5Q<^xWMm-Lt?t6*W2?;ZoJQZzDA>L~VxC-0Z4)el8=hUDgdX*!FR&&_u5
zK}0oh0<*q(f)?O^<1kYcMIH-ptEc&2;=O2Td9&VVw%#!{njX#2lFn2=!|W;&N4t>m
zoJBmjf)Iqa;hHPBq(}yyHj2-4`z=;9S`m%zId7~e@(~dIFmst4Mds~k>7e{eycaDE
zZv~2WR!59xN0aIntPWDRan+#wYEZ9VRvVoN2ClfQJ;l0zWSvRw-wZZb5zuk*vQbQw
zFUN))_jD}1*uSdmW9>tZrRVrxZ3~T<$7xwSxt}E{W>Mu<=zb9dSd6Dp=^9TZN6j2V
zD3Yhh=yT+*5u`*V;#eahMQ>oK@8I^d;7pPQwfFM`MlmwU9g-0KNbjbsg2{aehot^7
zk!(YuirL72(tD9S{m#qqkozQ<pk^r{W(ebCBDtk)r*Wg1t7<!0S5}S7k8a`5Su*E`
zAVx9D4aKGQ`x!8i5ImXzK;4SC&(Nj$<Qv+}4Y)sHVR(Q2+YdAphnV#tN}#~YA%Ec^
zFLP-WelC}GNanJ~F>JaZumWV9tkviP=NRlr=jt|p>92gV_vKsgvi@pb;i3v>gA+70
zUcVTt$=D=Y!wJdyc)kYJ5c>7NPb1XBErG!AmkIlQoKj~TuVb*&|EAQ8feQa9bUEXZ
zg3#6lNAo?r`_J>&AHV#B>pVUGm%1W)6v{#j)A#%e0l}mWK1|=ZK?@f=Aifk4joDy}
z6*fwL1fV>2Zuv4D4DMUdHnh#(s5uG~j8g6Rgj{{<%3Rtu7KNzRW@h@WZDiSIE@^I6
z6nY~nWh}kocpgQsX|ByXcg0cvnhsy5!>6KM>zwo01tS+mM}l(!H^+*0K^X9u7&aQ_
zCcyoIc7x-KW}-)5vDMtpxHc-(mqYAh6haAqG<7{ye)9s9mS*PMgtVAK6{>vp?Hiit
zW{WM=T4{X=Z!4ixAZvx9;H>2N`<L%!;4=>U*nDO}H=wjI)=f87qJuGsR?1G$4Xvf*
zr91cLkiqdtmjIKwtnIUo0J8fuHR9xn-83`%biO)3SX{T$+UP;!tOaefP~KK8Y!oJc
zN_Px22r^?WOvF2FZy796y6QOoDF#|KIbH!n)H5g4zC-sl-c$U4-zL~|O3h;`su}w)
z@`Ku1lWVtHOz>x?+->h1<rk7t--oQB%suflt#r(TK<Z*1aR%3B){t(1=>idrbKB6D
zZwfV;_7WhKb#_4nSpibzRrF}U+@=+OS5P!)O-Txlxm~elJ@wX2aU_p5cJ7~$jjvQ2
znM=!x<)lJaqO1KK*aW+B1*OzxDq98mKIat$5)8(zFdNaB#CMa{>!y_{t#1>JZZhgP
zsBfj>)u`cEC#p-o0w4#j>+&1)EMef;r^>6dSxciCZm0?^D&Bo$x-V+M0+o_~?3xjw
z0pzSyS<2Ed`KG9+89EKEGmzVMNtSZS*(K?-H}4WAZ#X`0k&EfM?5?sEVYSx+BXTXX
zNu0C9=^+HesQ8;!*_0E;;G$@Zz1mzOnMai<<$8+aZrxvlL6qjL>M5f^u2$6|DiL#x
zuJ>y4eNTlET)v0b7e`p1n699IW8a}qf_J&}1H>qCU9@~YL+4AbWb`)9`C*M5mdhR+
znlp$EhdGdz?0V^<xCVQDJ?fY;jyKk0$Mr(zF;@dIr&Ok=dRm_?&)izFyuCX|xm;N5
zB!I_Z_w1tTqi}MV7i<(HG4Bcprh6^S^SV4V?bQdyaAEWy%b*K3R9P*5Y0>h6{b~$y
z(fGlB*F)whpGs5|#6fo#oDT55!nHHH2Cj=M2d6Rv&LbM|l<Yk04)|@c*hb$|R*Xg6
z;N$r4Y)4Lzz}R|O-3gEb_f*=g*-o3+%xJh~1<ZRMlk=mt+}PAKVd>via>yoGTg@SB
zI+#<)e^1M9rQ&llmwY&X@eQPD#3d_d(OMdz<(&B1CQ0K^6H54@shhw_jtP4bsocBZ
z;rzC;`A?mkt6)V0HA{PE%<Tz@^J1BOr>(VjoD-g$^o|r)AQAr(R+0WS@UcQr2@ZOV
zoO_m>%8_<1O|CiO6Rts1kVVYgNpR^2wK7P~oBoKJO8U{h$qBW8qFS1aS3C&oDPCo9
z7;)soi1qWQUB{9Z(DHg&PaR7-BP7!4M_qo-t*k=#2CuHz?}~10d*G4`9D|PARhF<?
zcM2R2-1_yAz}20$rsCRz%DQx$8y*nN8`sNf+p9VEXO}XRB{NV1wf9ZUrq-oW44!A-
z&cV-xnHqgVHpxqW9RY+Oz`NnDiW5=4%~cgjoQzdhbOb%c-?Jm=WKE8MW!t)#vEd}5
z7CsL8tWM&RtqPXzLDXbcAMenNIXH*)iY;HcN@CZGXQ?0BIBPFy@S*PdkgEUf-sThZ
zwtidef3D;1yU_IpKtW5~KMH!qy(Lfxn5zr|P>I(l&6QYx)j;;qm9b)Z2QJF)i`#hB
z3Hw!dw&9xJ&xLpMaJlQkps3XVVS*AazK!G3PH1bc>A02@uTlMTn0YzXQ{Jlnc_W;M
zzV}(xzti=j?q*gm(q&~gcB4yTZ(DU^dqB)!w7aG~bQHU*U3=`ndg0J@brgr%&H1ZX
zEvuvW%Bo*~v5h6s)5}R0Jj`)_iO*uTFU?7{?4cObp>9WU0M2{LkQw~<6hFLzVi5QY
z6gO~%$5cTL3vJ|U%d3o6Iw>AE-IJCMlNLWna%?4G^u&PwDuqwTXy}BMoDcrAmqPnt
z?k`<8-avguA_Y0v8>7Bg-2Mcqh4KCmcl$G<xkC6k^YZ7LT>cDHC5?aH{e;%+&b*Y_
zjyUh;XQS;id9)+0fA15$*m+kk&5o;==aiWGPs|Vg4@NxMEthd#C>58ZU_TQ9G?y{=
zDJPfOR45J@yp%U-6lsz-LE;9Q+b2#4IB|ggf1t`+mQ^Soe}a=`P75nf@4ue}KH&H%
zI1U<pvO+MU#OeLd)Bj#Lmj<s3Uck9hcv-x6_^r>+N;sS$@QKiPiSx4skzka>OW<W9
z49-{nCE}6S^o+(0;&Fx(oWS40(RKdi{l_Psh~~^2x}~rF*W3HkUr+rO3<xHSIQ{zP
z6FmSWm~!%@e;n=_PJjP)`u*vr9&1@ktJC!H_aCQj<mbdhCy<HU7_A#73e5!fCu2NQ
z^e8eBxx+PX8jvY~4AJ<li2-ksb->Thi3OkOP7=%|-8swin9v<#I4&DMG>^nKBS`G0
zo9FYC^1uxPmVvUFQuvH3qX^)J7g0>&63wcMdk2n+f1?MC)qXYIqmZ?Zm}!MDZWC|x
z#3vf3ImA0aiui@@&%`7`2R059q5KvJWR`>^(nnxNGlJ~M#7Zrg9B8G*mSn|S-WNnp
zGrR^&*F@5sl5Lrw5t142GcIM%!uYW7^D|ie5_cU4BO>2u)VxH&X_bEgl)JQzAk^FW
znRpm@f0x7&#1OVwFwtz^!*7X@o}nzBwFpo)Qyegggu^z}8lohzN@|Bz${FD0RJOih
z^yi|20|ON@&JFx~j0Sc#FfT#nE%VCt6p(9lo}8t;wsZtBKxEd$I<3#L2<bIZSwZ0Q
zqzGZ;l14d=+M2a#HFzydFZ08GmKa$yk)sI#e>8rqGrr)%K9d?IDz3FZlf`#wpLTjn
z9wt*T&q-W`32O%mOkAPJjO23(esjj&wnGI7&$u;xsK7|X6h@GZSh4MXL#$Z$z}h-y
z?IS&KhjElz;uR2``6H0uj8yJY7=sa2_Lxz+b3Rd6V8)7mwwlo%^6DISDc%xI#ztQ)
ze~62Lre>se;b>zhkDeqosai1d#6mLcz*0Bw#$!C_kj0oY8*q|->z@9PM`W!He)~u2
zLyK|VitpEn7JIhDK_?Jim|=I7Zw7vx29L?Ab3N8hFAr%I`5kWujB0NjnO83`%==;Y
z+xkV`6|RT*BAsWWDh+HofY2*u{eV%xe+Mu+hB?ND`W(TixMYf}0pon)rnq5R4}wNn
zCibT!{3z>{#K<xdaa;733awQ80fIDY(gR3(789oB{NxCsUNgU`BKuQn;D9R(81Klb
z0V_3tLD_OneV|kud@InNi(=r#C##D&RQis#oUPEe33H&I_VpHv-96c*(g}x;e{hN~
z7BYouQYB1_v;yWq)NwS(J>_y#YHJmL%Q`o5iISXoCEY=0B&uUIwLy4BnF*zkmp3qy
z5m6qXgc=0lcghWO;HE}FnC}D1lFA~O^eM#^b2Us7i=V2bOr`L#$6#U^xx{U$n8zIO
zk=%}nhsX(R>WPFz9QM*2uPQVwe=shLXUr-Rk)b>9`ss#Rg%)A0c5NzxVG+~cTJgU9
z{=1m3lVQ$*NXE&X)*#<xko8q6hVNaa28pX2aSL1R8!Z&WkiV;a0|(kS@Bs&wD8m@X
z4SWTSB~&{m;E`obhw;H_Y=?a#jqP_glg54^Oo8!CU|RdN1uD>vN$LL!e^mcUpaQT=
z+W+E|X!LGp1ZhC_#i@$ldl#u79v`JZ8K#8zKa5f!HB(xC@hCOlTt``W22ye}*V0rD
zu<jF61)EQ=@OnKzlb=mWTl`GbMOENl#<Pjs_)brAkCLZ;vaG_M*@q*W8`v^ZQ(|kX
z2zIrXTDEQ^vqdVwDcE&Df6RlV>EQV528_lC={g{!@L3vHhe|$$a5{KMPkjzrc_jN%
z{aQxcR0oIPR&eWd<RHUw33)P`dUV;rl$K!Sm2n<I*glN=u8-c(*>vb{mFKN=rKiaq
zJW8Qc1K>oLQfGjd$#qh|j}+&FnrW@n-A_CS5ddkGd9JQgh(gu1f9K+a#l2*G6Y3($
z`9s-##l<E_Wu4QUSyN6zAte`KkFSY;E^;;=C)ca1D8IJP(a(d(k2p$)Ug>UMw=U04
z97wpg&tC#Z!*g5cbY1?TC#ouI?svI%4x@}<7)LrLE;y$(t5ja1OY}Xak+rrnu!-c_
zG}9(jOM~v*k`VJMf1%K;$mtD(ORrSx{zV!F&h}WgypnXxz_4}{xos3D2--~!su<n1
z%t%Hzr?C6YB^miboq+|e&^~l2p)?~d3k@@&(AVOI+u6G_{+bJo9P+5?OaiLIOFL*y
ztC5d5*QL`LW!$j8`<n8LHo_IIV*6EuSe1};nB!5Zu}ebjf9!iCWU@Oc+NK+~*rgSf
z%aL;~sjPmj!m%p>Xm#g?4!iemyq+B1{u!y9LU4RX;Q&{1An>yx!p-=wsIz?4{Swj8
zZS;WF?K$89Q=d1KvNd_u)bd{T>X>iKp1C1!kcvq4X9bPu?W&3#%1-(c4b$#9#|xTS
zr@{%CDJ8+{f6eSNNuhxm-vLiG>BdV8R^ZYj^zrB^xy0yUdrS+7Nr?`2UI#ytnB>UO
zBob&Zt4RdLKCMhoJ&_%>Pp|73paF|v??N0CKfPE{j&hEQzX9eFGIQn7R4i=Seg$CB
z=&^tq7lri_OG?SSPy^Jp_uV_<=zL<^m4Bt=TaY=%e=w=s!Cfe<_rQoADwi-z8lA;J
zEhsEi4my*gnj{RhMDEn22Do-4^R>pRd+~IDEPI^cU8oIZ;b@S2fNYiJz8WN+5zkaZ
zZ)c@*fVr00*X7E~&A&O6pJa*x%nF93x4!0i3S2Wjh-rhJz&&&1p*lgVXD`A7>XX}<
zzO=JLf6sS4dusEUy}6A0g1L;-;S{CY2{oi*?lHZp_uvfYSWW{1a~VI-xs}wwL5b|m
zFb^j>!YVZGJ0o2Sd^FMNq3=(0I!s?#qSH|y^<X{lDN?!~@L*&Tbfh5cosw^Wxw@Xw
z^<*t~v);r`K_a%8=S9(mwfy#Xd9u;VxI4wlf3xO$NlloVQJs}$_fe@}{^}*%DU}8|
zl&w`FG;0-)=p=sWD83yWr3me^ox!o0z1wR6hHWc)=$&u6Eea09h8H8&hI280$-y(W
z?3NoWVV*DYO4(Ubdb<#$S3|M4Oruuye7mEqiylvrlDC)^+PYh9%li`cL3ijiOyl?J
zWKJ`C&F^(W;Z@9c)#>uBHjO=hq?m7tROWYolI)AmTn$@TZ8@uZR{V>9wg6HeFM0|V
z`uA2he-ON{ecvM4*}ZRpy0-Py5OwGHFPh6F(s2HFo3{r0=q7&tetUoV2c~1A^_Ov8
zC>59Cz$FuxnO-Qte<9xf_4NDBR-vtlE}f>s`0w#)_wR000XP^7)8F5AX`TrT1)PM?
zSS!=_f2J?HKSrvZm(^#=kJ#z^*J+sfTtSRhLA1_^EI<US$nxY|G?XqC5cxqFt75#@
zp~`at0TxPr+y_{t$X3lEBSppAibBi^8E49cn&Bj8`^n-je`ZwtPT4sF4oAvp7jpHx
zJsam?yqk{X&<S-Ht6IWimFYaBXYpijnEVBJDr#QI15J<^Wv+{v&o97y$rV*uiG`cU
z%$cQ15>6l&OvEE<2_;!WZ-4utEC<OrRt$xv`6p|$m^rj-!;G9^AG*if4ft$e6<{N7
zODA6v)X>3LfALj<0u`xt{^@d`4`jf{fzjadFL9ktFlw|6VI)XJ`efTAsibpXy@qEE
z#IJe`QZO9+K=c`hxfBNUXasR!0J1KPcgFisn&wWIa1CN~j9m9C2DpG|)L74>MS)jU
zA&Z7Ajlkv=hd0&JxY#ShTkV(U?7YG+kdK5kUb=H0f9I+ASN_Qa7c++h9pof8=pe=8
z?79spMQ<G4^fkf7l2Od&rP?5bhkhpQOns%C1#kOR-aH&hNNbTdj7I~-lS%h-{4^RE
zk8{%Hh6`4yQ(`4*sen#!?&w`1fkflo?2)ljZtj0S2i}HY4q^8_Jy8j0f`=(Eq15gE
zX?~9Hf3k19T52urg6&C^o=wNi8)41guHa2tUUE;UOr@ITSVpRfQ_60*r0ZbkV4u&a
z*fPspQAw4u&~1j)m*TkOh6<-t7F;u8QcJI6gWHusm0TvBo`1|Fvt9SqN|h_bQfs>^
z(E_Cw3NKGzS7D0Igb+`clik~fD|4L5ok^(Oe*_j9d9H$q?)6fFm`H~$QMbn3VLj4}
zS*>?pvqm~TG;4N+DyZER7E0yz2ulXaO3|!PRxlkOdMoy71udKCd(%hN_}Ovdk>rbv
zI7%(pm;Z~f$j(+VXn1sFLoqnOt2N}v+7tGKMS~9Fx^`0ew&3N4wU}k<l4sj@joTUw
ze|=`9V4LxJgPNSRw6k%`&U2(#mJ=LER*K|dX3Yj#x&X~7C;M(4SdMcjN=v#|{Gr6a
z!*Gw)9xVll93soL0%z)2L%q*84u_Qc8ZOq|+$t0eLfFzO1I3s2N>{8L3<f2>t5T1x
zGsLhZo=b{jj7!bed5m9GQd5+?PxX97f4>EK<jJtxGDZv(*P<RhUcEy#P#?<3d7nj}
z2X~}hwc6RZZZB_W5tLILIiA=-LyVeZ!zuaRVC9n2Hs88;wQ@;4M#7enaG<zhlSXPW
zhw%qCX#^o{J^frg8d+~Vzz^xsFgoJ}|L@T@?9tM`ouAvEciN+&XA_8j;U*0of0^Lr
z7jDu}&_wX>t4Vul0de@^X7jX(WKE25!?Zb2+_+>LsGlj>M(r*w*{<+3RQUN_RYnkl
z0mKbWJW$-&!~^v+CLXnW`>K54h_;Pq-?^QT^9`zV2wrjQX|*NqZ1b?acR4Z0TlRb-
z#q0NlK-ggT`Syj@J|OwHmoP=ce;Q!i+UGR@+LK9`UsSe_XLM+sNd@T5L7`Y2Ascs%
zs8$48@(eUz{`NVi)WTq0Y9ux;!pK4FTmlrFE0yOzEsE)#*WAq{w~Og3BA-a!SAj$O
zuBaEvvE=G9+gN7>+GY!C|A#3M=m^vn;L+^-$;YvXJ{WIl6!E(E_@{6{ULvFMWnjXs
zQr*{D&XroC;u6~H=#IeBH@46oH53WEtF`ZbbPfSxZO`0@z|x&=vDBp<0pI?Go4H0U
z9o#KKt6{sB<Y(%Y)-AI?=sS<QWsk=RtQ{YpcK-oMYTw?Mab74Cmy!Q26PMLwC@g<(
z9Jvv`&#&k|Aeu!MUw|NFG_nlj=7nxS4uKIn2Ya!*IppuB_$ZRi=Da*(3203>o2<vj
zVm;WD%eO@?pLWl0yAQ7#7bm;{r<S*`3(yNm|3OltHH!lfR>Si4WBJFUl*-ceOczS2
zHC>MGJ^h-e2MwNj0KEZ27d$;Ugff3dE4qL#o5xUp9Ue0sbtvy*yCpi_$~X?_8HhIQ
zpZ@vw_nj<4OA&0mjX&el$G6>|yHN!rgRlmdFF$ti1!@r5tA)r5ZDINA-^<tC=aFn>
zCTUFmXXNtNw`FMgt)Lhrp(w32W`!b1f)*ERgV6C+MiF16G^826(nZ>1L;-(RT7G!-
zFr|o+=EIB>32<o$V<gNlQO?u^MmhOk5`Wb~;#bPn3b1OVtmI0r+S{S94#wVkBL_8*
zy-3m$J{FnILwXiZbON=z0#8EC5qZ!HL`I40s^zx}Fkf?3q^t><n=y?unj(omf}9sY
zZKy?*%nZHPr-!^A%rC%E8DW1Qyf=<(t`SGHS#v{vz{Go3FvToMYL9Jpj%gq-<^4$c
z1g0H=rhwKE<dzc%D&pkS7C}?E>{;#sz4PnwD0;?`8`Fyg6vi?`FmpgGh#=k2**stG
z7062>*pch&#xYz))N27Qj6`v+3hOPkuu@f|?Rstegl1(OuIVbcO*MbE#5Gm8g`yDw
zH~J4~-D1nUE=^Fv5+_F3=XHnfvdCn#(7}&X<e|(AO=0_&<Xg4$8W-<me8dv_)lBA2
zGp_QC=R5Btx5eC}*#z3!{xTT_^{p%;4>Lq>@~_4O;jF!5+#e_!-C_^>$fI8GhM7Pw
zyljy#6snlLb<%|+V9|dqdFv6;pWv)1i9=uW3tJF08O<a`iWk)rM84_<zD9ryi&0j1
z-~9Sjwsbpp`VG4C@h6S`RC-*^>U5fqZe><6702h6!U$zNrOf1-=<=G?*tE9rwdJIc
zM8c8N4Z;7YP96n#J93IkB2kPS%#%vSqGH)%A6rXudM+CowSj+oWVvF`_=L!%PVJ{_
z!pCpo(hEGoG&IZmD>UwSB^sEGy|nI&#8vt#Wm@>4p9`0?P+Tg@B8DGcjbDO5rKFwO
zgQ3|C(1HyYb#xt`9-^7Y$twUPvA65~tkEeiuvtQJj<JN$n)GlP?JRdfOLwTa4YIDz
zl`3xzjB<UO=jDG|o32S8Xmb2&5WJtv<N>RAcpM+@5qe{cMjGGPQc2~>&MN8-0M(gD
zp8iaU&HG5@p(&FB1B3Vg>^fC=(#V1s7NiOeozpiJQN30mCcbRRDF~G`O$Xq$<9tzk
z8bV(C*<1tI&tjWX#A0l7zhe}6fq<j2wJh(fm4c;7H?x1)KE8BGI2Bh&YO(xOjT1UJ
z_uV`nN=51Gc&s^CZK-$XAk4i+2(u&%A=5ksp(Z3Cs1b>-2snnJhCmh<ei??k+}pOc
z8iv`lQ72DzC=NN(=-o!<af3v8j_Olfk9g8AZY)ev8FVE$WmRQgA3gFT$A#woOJu5P
z{3@QX0knTO^33il-q%RQkaqDA?Zm$pNqQdGC8m`QWRov(=;PZ44Ham?Y9F`j{#?k1
zeOi(uBZp1Zg<1Z$#Ab-z+1j-8pb@A5C+EHU@W;2WfBW(;5H#SwafDou96C@{@*_V|
zbrvY4#N~T=Mxn`w{SwP%P3cLnLXmR14y+AdmK}d8S_ql~L~H#XTPcuw3Z>oPw8!XT
zM`qA9u?&%7ZDoGn`$m?ku}C^qG2&CCQpUzd8qcHRW4!uj?n<Nnv+TYuyU!KRpy@Ad
z=TJ^CffEFR)6hf8O%ADI(ZMhr!I8lnY3&uQjZ`xFSTsg2?I_)dn@NRRRhC;ig}I4n
zF$I5a;@`3T2mphw8In@h3PYKYS5$|XXq8Z4&Cs^{FF(q-Z#3#t_nYZFfYJhuT^=;i
zGG0eoY44a)v1!7byK`?&&G%2*Jhoj&&N}kKD*h#N=J?+)vVkax@OkM(%(^}_Wj|jG
z5P)z_UjWdUmUO@n4FgbtITDyz2P%cNR!4vG$CIHQ6<L8BDaieqQJA*EhEd7i1vSgL
z4--F`de-`&jyswr<{pO?pWdfVygc8hcrf9&3V-eWrlcMclH4<=mpR@5KbI^LZC^WL
zFG1EIM=kQ#b1689*`L7bT+%Z|v{a583qmI-iL8;g&@?3iqix$XoNC8JOyw+e-_3vU
zGq=X{tlV9gQYd#WVyjiwtDGOu>nL0hsa4AA8CA8Zqpp?S4{53z>aZ1F&#l6)_vW}1
zo9d#@bS{;OtcT+>tW7L($2|a3a-H-Xe_w_qe3jMIdMH21ZUu7Yb&}4y>YjKOg;s#u
zNLyDDuYO$RSm?PNz&8J&_b**h*n58h-0VDrzQ>?Bw(_(qO*4&S6=8bITawdiP_BYi
zC;27mGS6xJwQber&nyopUGc`Pn9jD8-&D<I+aAs1>Uv_o_FSG>^Qg(kybkJ2&SQi!
zC;)Jjv6de$c{Ik^b4&G?>}#E_)v~IHH?O;{!_@_Kc&Zk>8B5B<XDg4{w|;-7^;_+>
zsD@fz7gN-#24`bq{Vcbc?~)bp8t$cqqh6x>5S!vCxS3vUA5YaMU!&Z;_CDjsLhoUt
ziET4q_qd{jYQ3>P+0r>`e?D#?tC3ST=l9Zysv3%&8<kAkYBZijFVnBOMVR;OFHuQb
zNcrWc(qSE(cu91doTu{NmT7;^^52%r=KS~fW!<fmst;YQdcj#5<w1Bk+hl)G3MYFt
zWH*oPj6pJG&;89c2}`CkCM=mJR%(}q{M@^^-ArW_x-&PqnrUtwl9N6_6-um2(RP!*
zWT|c2cPa9i5t-D7dDxb!c8K_H@2G^NjXw8UFjT(_B0O5bFz1)72qS;v*&Dy16pBVL
zOZ@7p*jS|v#G)GI{k<QM_mg_yQzM=&j$xj$pZ(V3sIYVc?dH!p?sKl0?myQ3DUQnE
zr(bC%hpKpFmA<xMzhk!AaLu&Nm^FV}>XtiyOvRzXnZ+#La4dNrMLbAn@6rya(q(hs
zY?9eNc-JINw<H^0iWz@gdFIX<VX0IekK6^SyIpvfBX^2wJEiY=<ZhsN$0K(G<uga_
zhTg}H+^JGIsC~ZI_G<Z*0?J~;$Z-g-<B_OQzV5@-rsgX4dgo(pQ4+H6d(QouB8L)F
z(we#x@w%?4&K||MgRYV?ns+|Hx8ix@F~}9IOAkT*^g)*Bfe?RTJP>!sBR?@B8KvKO
zMh32zpQ@P8jgLoDH@cjkot|fIRX=w)PZ^*bV-(=VV~5I81{OgbsLszFD(eMZ8xU?V
zJ`;qoxc$(_@J-Ja9Y5%E2V>PvShVkLtZS}HU$g-|H`IKPbf9?NP<6KvC9e<MN_&Cc
zjeKu7<*85!M~+izq&Pp`8#fFCWlclxTb-k_!g|-V8IGRa-TJ7GarcbU1HrQ;#xFT5
z8MN@-Hbq!791Qs$FgmINgT5n-Yl7xX{wDDP7^jK;@Jofn{{dn3mExCiUMLioal983
z0x~d{5e-2nmymBLmjUUQC~zo=f1e9xWocPIGO3#pCUv6AcxA}fUD$-x3g$3K?KUHC
zauKbv5Rc^I0CC`2;+S!TV-`NgDKN@5aV1#ANcf?{8D%3XXXpV&F*`hiXDfk(C}Z7>
z{C1m}16CwaHCuL&)y)+pNxQIdQj!KSkOCh^h!++3{q?!H3`SYZ;FL#;f3QqqT>Z$D
zZALm_Aq8cEAMt|S-6a}S?BT|@GwyF!6dTy#2GikY+AzaCUQgHsdUvE-8X#uFp-$#n
zzC|Cjxx+0TKqqk|FyTl#2Z-O#aaeRJgm{<d*o$Dd0=$V@d9)Qi`B!i;i;g3)4mGx9
zcD+6)BeRrd@;gByP7uGJf1)?&H1k&0WhiB&n&!E#&W<6}`78{9x56<|R*B;DBC~db
z?JR0`YVbP&<TWeQL0omMJr`#}kn<K%ljKy08ZOG15vqDef$r=6Mbu~mXBq5b_z6_S
zJ&mr8brKOpQoMsvM496f?=vMR9H_=-E4`D0y`&7ru-OJ0F|E9)fAVZ_PmLeKi%h%l
zmB(@}m^hzhS>$1w-o(m#(Wo_FnAFn;<u?gNOLaVhLo0&LBhW^wYL$Oy*^2K}mFBHy
zC10T$*?j)jFmIvi3knfLWkbcBR++Y!^v~@=y(407sA}taTAb_(s|Yq2yqM)mC2U~=
zojS8=Tyvcd7vasg0Chl$zkjniNdzuZl2zE_$v%HG(s5!}4IFzxFdGwaodGj-=yl~^
z?Ehst@_bdbo-+2#@^0m)tn4HsE2~*m*XD~iZu$sCdt|-q-NL*(m~iO9Wm&f-)S1ps
zivf;vGQ`;n;Vy@dbutuzv#SwNZ@j_;yJ1-uV+wnr7wmh9D80BFoPYO~X(p^Oqd>Tq
z)#A@5g|vn#Y+-qdK;VnVOKUH&9>^#{Kf(guH)>gVA$S{Ne+}bGNqH~gGs&0IxC-Gm
zKCU)#-f4*szH;7P$L@L0#!cAuZS27+7pIj?g5rdWd_@kfb1)?qiQ?7(P@R+PJm!uA
z?8};*ir9h>R!0#=7k^$x_*db{Y0X_S=IW?Ib-YQ_`;c)<evzwfW5)g5nS5U2ZYpDt
zCaAaKy4d5j{UXP1(l|oiNntPUEk<Rm-`gA3fb-TYyV#tcbIHQ-=6o;6?P{^I6Ut<j
z>%zSJPQ5P!^iq0d&2@$@Ik3fgFj60SF73>9$pE%cv#YCURDbj^dDof(6#2caS$(k(
zKPDp^o4J=GvTEW;ohg=iU1uz^q-msME$pxBi}MP!8CSPfz~{pydQ>|3qAiB+5iV1o
z68wbeX1>k(_oLAJV5gMW(E`HQDMg-*_DdS2Mx!EoWJh9v_?Vy$t>jbwLveM3k5%zp
z2L@%MceGc<lYdQZ`ZcId?O!%Q6FMg-|Kc`m)}T~FRe7t0shT_c&Wx{j7l=sV+8f(+
z7YyaLmZ|lh6lUi%LxVpmBil&V-n&XfAzs#-8WV&OuInar1)w3blhaig_u*~t?=UZU
ze&g^(3N2eR&<0tGz>L|f@nPJ-eK~A{+Lvi94y}<i?|;tqd$)brmuBDk9ZF%f7ljBx
zVF-O2kd^Kz9}K|Rs52uL3Sz=tmWMmsCtG=NmmdZnl?OLsJ20j|cDoY>d>?X$W9UKS
zl=X3ko@ZKXD|=kTbjX7fand$RN%ErvupdRB%LMe*IJs>#FNuoNFGC06IB_Olj1<QG
z4bJPX6n}t|4fgnZ=Co3=N!uI@>2aST_-*R`Og7$hBGRC%JuIgmawH`cwl<&5Y&$%M
z7*2IJIw~SuQ>03xaNMo`4R#b4r+b$ZaLH(J#FVJF9mP|LU~y=D%ZS_d4_kf{IX=GB
zH#a?^vy?Jr{m^r)6t;pVzg0V!^6y+^C=f2Bk$=8x+YUPjo83=2RGOui-E_CEs|DD)
zJE*HNtesw`ygRwQ8D<TQGJZWc?*%R7@|xCObL&yIs}1$ks^|RL0ulMuVhZZvN4P?r
z%<fPpGkq)J+}US3&Odj3zy}lkND9n+BO;<gjtpOs=%Y8vJ3KJaA0W<0iVq(LC!+_4
zBY$M+R!xW2{Fd`DzD;{ao2SwlNR;AR`AzQe<7y5I#>EjM!2WSHTIF%#Jl!~=&7F#*
zYCj!z9GhzVV9b%S-$HC12yytW8UNlN$$ZJtGm_r)>b~YG#(*_n#klPdXN?XFWrBA5
z8b-aU@GZK$#ntt~SN=S5p~G-0h7(k?Z+{_j{bQu}+e?0g$lh~J+>XD(0Wm0TzgWGt
zNDsK^vjuqPJs&AbKXK1zixxhiXL1#20C?%%kW-Ok>h|SNUK@hI@`|wSv46*_L*Te(
z^>m;QA0YmpTpda$ZD{S0bPf=APH!IXWC$-#ukcdB^=DLxI1P+nbyqQclRxrO=NbU<
zv1+g~@>3j4r&jpQH@3!a&fxzA>v210mvLSw6ah7t5qu34m&YGKCzlL(D4Ksv8&3QG
zKL2|ov(WFJ$W30LQqrtRY%B_<RU~fm<a@o}@~z}oEWacc44DzK5a{{&!~$TL2x7f6
zsUPt;3iqKu=4V@a#a0%Vm>J0^@-=V`DG_IpL=6DoG08#!2&s38W1x$L^QlM^dSH;x
z?ZJkgJeGJ;_>|;}J)x%+sK<Y?hcY%$BvpAW5ZQy-WmHm+BBQG`uc1@<By_T@vW`E^
z=7N{#25e_Dvq{kqe2-t^qhL3a6W3bfj9tR4qFXR@Rr*67KVO-O9cKDL=CXoO3O}PV
zU*Di13ynZ26jgJ>LtYO|62XUJW>KO*IieXfz?{{9(yQ=qDD*TU)_Q-`v4lNiob;JZ
z&k6F@=v93NB`j9c)2kGVy+mVt&ueFJ*CCfR2h04LZ5?MVlu<t9IR#CS1C3~#zSz(y
z>$SVdHp!#VZ6&lTmYAFouaa|Wk^=d1n)LK1yQgLA=ism$ta4H)Nc)jfvG+LVvIeU7
zm8J_3(f4^3OJB-pn8AN#uFq9P=dKqLWW82IXMBRqJ|Xw$RwuXIxH^i=a&<BddA#OZ
zxX-E5Sn)4e78PaAPr<*W5fwwo7$*PFjY?0EI%Y3{Ny_4fie{s!^zZRn(Tn^H%_=(V
zl+eEJFF#R}z#2ga3y#G@rZp)jmR<5xy;*PHaF>0j-h&2pRA+xWQ9YRyEM$O{|2_^+
zAD%aVEF3eksqj)wUw&-D3)CPquLlrEB2rC%{yKf#d>YAS(4>$&CjF31%Q<WLoI^1$
z8Cz)#(F|MQ6(^jCd0AJthWBmbic=4f6by?M)niGJHyNX8AUL!0NH`;fLCW$Ig`c&s
z@RhJJ0*o36D>;9Ys;LYLV`1p57g8ioAWw1kOA=j&^D34)EUUf)Ooq%ddC(ICMv?2R
z<?{`I&$%j{BLwX9J=i4G_+uS7C%jrv3n<YVdQbap3aIHP;0Po<(p%#==NfRl&F`+`
z>8wD#n9FrMd!F)sYf%`YB^dbvtxBoT2+wY1=uOA_`4WHe>MrIBPYYAMi>h&guLV_H
zpN&<5lv-e(k9r&zP3<T0owf8nr)Z)?;a%N%+S-)W3bCe<Dj}Tj7`=4otgLTwz~J>(
zf==_xN4oBT;$vdQ5_9cg_YlP%gamU@v4`GO6&?V;0Ri+#^ydVds=ZWs+2afF@BrmC
zk5^vBhHZcGk?_qazD`*?O?|oZK(Io$6$O*?MS4JUExO}yaSRr<)&flo7XO&I<HsrX
zW&+wfn*N?bGlW3in7H?84ljKWb@T4`Z(slQ<rm%wr*D6S5gcoVjDUC3kMu}<R&!pP
z>3e=Q!fI=I^Yl_1c7#qC1b$z})rK$A29@QFmXCiDt@V3o<tWx+@W*v<3~+7enCW*d
zMS?)Ewib5Z`^F+v;}B<5r3g=fN*)^?d3+fa9>di?m##eOf79mcwE0xg^o+ZX;e<46
zV1nRv*3bi5nzRw<J=ubWx2$SsTKkFCMkS0hsH1LM-a|q#leyK&mX>b9rHN@33Z<$1
zkL`bF02t<)AW2%u+dcahB%#<YStlXGnxJj>Uw-6q-+0u=?myAlf?~BMGi^0TpcZ|g
z70)f?*7F;wd+A*w{KH<;u0_4^Bv;Pihny8loYzcjAj%?qnmQ5FT(@nqpKb<-eSu{C
zxeieC2e%*9S3pF~9vCW!=aNKuz8tiL|9^ibNpx9&m=Se2p5s%?t>^gEdpyUxv$<oZ
zyWGwicbaW2Ed#I)O`&*C{>G@7zcDqFdT;)wK#ax<>2btZ-y%IO@~f+7oT5fq>B?sA
zTc%4_9=WAu%2STnb}E<Ah*|TNAHHO-P21P(wb@>>*Wadjx4f}#yPdUoo(uVCE%|?R
zVerO$H1+vtYFWAX9`xZuIzd-`NqOjt=l5!4*Rsuu&!E(TUv<hkOM8z_20P3nhFZr!
zM|M?$Q_j6(v$L?$;gqYK8gzuCwxQk>$My7FKbHE6%;`Sq;=O5jr(QLin<*8UA_gwo
z2vvWv7=`CJ<t%W+a)yEH>pSaI>Ck`GerEluuUoJAC`-aX%SB<exNDWQMV8ZK3l5Uc
zYd&diWlYo>+uMAqZq@hH=TTGUAM`OJ(CTHq3@OiB>ZR3Bb^V);&8k=h9s7uEu)Qf8
zY|1W@a4uE7h7Bz7>vPThN}#P89qBeXI?EaJ&kG$OD?0P{DMYHSS@~u8u(W@gvYf3%
zF_!S6I_RLtDxcm$npNx2DtBG@W%9zh&fS~RYX(K+U7g*j+mvipq&g-l2KQ$LQt8y1
z-(7c|-OMAEBrVo<-xQ5XF*DJ29Wp7f98r8+xdp+C7^%+|F|2-Mf4nKoLOyKJ3JG#;
zZN_ztMI>@bW2`DGd0d<K619JzS+F~sc3!9`$JAEQxd0G>cAW;amZ$Q`L37IaO=q~N
zCC*jn&1Bw}G8MNNlXas#6&*#W#l^fyrB+(ioczT~d|N&wC|VTkhkKtbQ!fyJHphB*
z#hG{V4^9dJJEz*wrrC^&w#zk#+ardl^tQUj99~)5cB1SC?4VG|yJ3Izh-L3Q)RxOt
zjpAvCI$m)QBe(gE<ra0tx<A4A?MJ^?+B3_Z<JT>!0JB>Aeq?Qzz}bF!B_iUu@<mlg
zAC9qpQFK)7DwKJXqGLGy0X;H#I#7II*>RwJrS3TNext%8I4@NDI_-@LW&_t(79Iy`
z|488xOvu>!x5jbK^|gOE&Z*s3aIAZX8b3XJRVEzB2)$B8yTtI}B?r(b5Y<~R?wsp`
zs>%b^H>)ZSe6LiMhu*KKDvx}h&N|{!F&<a@4xPJ0J`EI>ez6`OUR?3kN(-_Jy*X3y
z){3`uy<q<eOAVPdr(akz<l}R>yAklWD3pT?xn}dgXk(=4+Zli9><5Nx9`yx=C_^R4
zj1t7pUYE7SI!b{+<Hu1|Pu4=c>z|pkXt4Oo-1YgI8@=4wG6ipCOnP87RL0NB>2x^s
z4r6P$ZTm!iI}o_;gTm81w4Vq5-GSlK6cqF4wn2=L74D2Cd9z|*ba7hBJ7MI4;IMJE
zl==SwO^o}B3T2nX!Z8p6H8Gcgp(zxX2$%{em)!_A3zzJHD3N~zy7b}ammhbMnSOlP
z^-Nl&g|#qz`nLOXppMT&>b;Xf2XBUdhS*GlsI#?>?ivBY@E3J?_s5sdKYaQH1P%B*
zj*tTiWgz(BTYfZFNbj5*zVfpds4dX!A`rhgSfZx`btnUC!>3_~$_dcMH?3b|YY^yB
zsb;VSp}nJyb%KAUsqhdf*48rL*S4{6HBOOas%*rkNQKA7M;c#7#m9KH&!sDk`fb>K
z9(Erpo<Y+e-K~>$2~H3Qj;yDFv@~fiAUNipcnwY(v(nlMS{s$H-l8Vayl`Z8dZ@Ei
zVM|LRaA_i1g+gg6_r84vfKJv7Nw!NPOrUX=SPjuCfnk5m(2l=9edBRoXw-9a&vf=c
zhK<G!djeRHI*3&AS>qRdmV0w&c3FLY(1z|vu4c$F4Y?W~`F@{VtIWIPVpE#Lz=xsX
zFAKUKCH-(QIuux((-)vqxLQC(8*-HqmX%OeL_kMA5^5+52dC~Bi-&`bAqOdxp`#F#
zf|Cg&YIuJ<?153l`23W6n;5WAR(uaC)2TEFWkbQwM5-wqj(zC2W=fH&oy~vG+)*IJ
zhcZ6)OB5qz%WsjNUCU7#|G+ar<+F^JSt1ff8b8`a;scrPXeWiHD>Q419%YJjC+Ns_
zoKYiI6%FPlj!1a>L@lHRfOy6b`)k)o(%=V@?YDo^G}{GhnrLK+X{6$j^69Wwp*+=@
zX-4I3BWD^HwUV$*v%P!ruxMw&2t_fvpM6PRitnC%QZnUb3s#82g}GeEtnJ7#=m(YU
zSXMhOQYN(6exn^`3ER;PXGhlFQ|*WK7jv8}<e0~#yJFfSdK1r?drgBrTQ0f@C+Z~9
z7(;&ut@qa<z0c`|d>?;B9?Zi@^wDb(;sU;oK1)}yD_-MkR4DemmdJF5yAo!{keS2m
z@58hkr$hTDA1cd*CkoE$r4-KuE0nIwg<_X3?ypMlNZB313IrTlETKcOVI0}fj-~H3
zRQTeECDhp299ap=%qpWOYwLKG_h-f+ElYo@8<gXtKO0FWVqE5@6;Vg`(Nc5^gw!oG
z$7CIly{o=mQ-Q?B?nX)}uBLYh+_a8lOQXca`<Ma7OwMbgW0JPe`tf<y*A&T=K%B~y
zJlgT43Lp9RX*F}7iv0oZ7U6GQWesqxC`TjJcFd)P1HvNoXgOG&Qhe@NmfvP^=v9A!
zJrrdD&+4ek(Y-(9k*EEkP)U%Evz-i$^ehY|<Gs2jGo3VIqb=pVGb8So+G6U~2&y|K
zK@n>2)uj?*qw7Z-PJvkat8PY7G8P>D+zalU79bq?ifz$4yi1koG2@KZb^1Z;o5Ue6
zh)n%{nil1;GzJol2eu^HBzL^vnYn+pOw;f+w4#+$`0vC8a}p$~&V`mU1r?KA64_l~
zlH99^X-`n{l#{@iF+|78uoi>nk!?QA3rcdi%fKOHk7e2`(J_F+gSr@<MbkHw7c}ci
z$pAR(PNpUstT@Zj*-$ulDnWx5)f}-Y&5I$LZqMBl<qLn#n9yDcr%hby<BNa151^>V
zqUv9&;jdqZ&$|yjS&phyW2HhX{xdQ!4U?9SyhRiwVJNMos(a$Zw0am?lW9(D-0(eT
z%ui$Pr=SbxNtbhqNJS<I(@~tH8HqAug^|zlFR?#q`8d{$V66bFddf;p_^Q1f3hSWn
zt!I2t18G;1l<>$rorm-+o-%(>cKZrA<Nga#hqz-QdD2PC#|t2zOse#`9DW5iPae?|
zix8P5=$=X{Zng&zy%QGIjZy4ojWAxeDlNlF)M7QzQ9X}mJEwfTvnZsPg3!;)A+_gH
z#mygzD_w6cs$Mm(p*ly-<b|i1qrQr&^@6S$Ra}<FDPfeFVSZ$3Ox%C&EdSEHla@B*
z6iakgSg+?<O3@b7LX;35OVa$yJ6YdihrtG3LKTvuRb;!+-BaAGOi0NF+|Bp9qC33R
z;ERb&nO`&bVj|Pt)kHQ>i&ny;+fpqZMLQ|FuQF>&O1<|Rn6>;armaK#PyicZ*io!I
zyC#R@Wa?%2=l!?`hunWB<%D*@Uat==+6N(F3~7=-zrpOI*>VDSZXZKahO5TajH@M1
zVB2aI%63XF3fk6Mz2-hm9acA0LIJO(bV((b;H;yTIXYSAk388*Wvv|XZTg;K7jGw*
z&2)a_n-2w+H}3NrGj1p6V&M9;Us3%MWSFJ>%2H^uH%a?rf$4u7Zs}lCiW{c=9py7=
zf7km)X}^}9?zhhSyIJ3u_;=L)lgwXv%13p$F^=7=8)g0-)thDh-Mq8K<TW`x2sMM}
z%_%+7x~FGUSMoK^?z-OB6eyI+$5R-+S{|>L#2Z1^OcK1FV$i|`+!%2!X<;$$Ld11+
zMaoNhoUYGwlrw)KT%BwxEN%L6Op~(LC_b}<a`O~4et(iyZMZ`A6#rM%RmVKolhsw%
zd&BBVQ7r_xW#ZCPe0z1JsJa4gHZ*uQF?g8|>CKFOxk@m82m+OocBx7L7laK>>?#KU
z`8V+r=j$`6+4tgnD8!#VKG1)@O&=QMcq{?^WNkh7`5S+X6{^Yh>{!W^qdCT$`eeLZ
znOyXt7_@6YKWKLVn@;|ycs@cdeq!ejO~=}!DSP;=M+GL9A_NXDfy1X4@`&+-OliUT
zYxzV2za0xX=1Z10vt(t${q!T+c}do!DCbB@nb~u1kiDBjWl>f=>tat!BHNQ5d8Xz>
z7mu_YYBhgfWmmv8CvlGBNgO^YQ#f7vwgk$w-<B>lyL7Hb>+f?C0IU-#%y=6=+VUdw
zihYq$zu5-UbvzW+FL?90tqanaP!f73<2PUJ{mjQ;Vk2ckBI*iMZ;`15i>`~6G-t}H
zMZ#saTlDpvPHvz!LR0MP_4Q``w6f%+;d1>L9f*G^v!I`=_Ccr+&<gu)5Db)l>YMC>
zfYburmRq$Q#SM2sI?88uLAu^I+6D1Ib;hlW_vxf>+yj{syKD<YKfXSzjz^zt=%#y1
z;Yc3oRIh=uL3tIyxL@vjf-6P3MnP$GuYcJY*rF$QcM25$0=zmts|saqWOH<KWnpa!
zWp0<hO$`MEH83_YmqD)qE&?+*m*F-9CV$;p-;W%(4SvsGp~tolT8X4Y5(o?=ah(=u
z0mJq~+QU7#&Z01C+y>`0;QsghsNLDS_1rr5WA4zzu)UlajYLr&Ka$e4(kPYGO5;?S
zLy$zhhtLwch2W|2XoK55+TgiXf}kZ?jW)P#JOoMJ`5Oq9YVqi(d#aTsqy|ajo`1F&
zE?!Wy6N#rvkou6IOmhuMQ&o~AhGyEvErH~YA*`h8z2crxqS9&-55YW|d7{eWA!bA#
zdRtG##%KmXYAAv$k`{`1Nz^=s$A6@y#0UzBqC1RF)??1-iF`~9gK-^}DG+*ef}xoz
zZq>NQm=YuyQgjRfsf}UrK6fz;fq!YT>KN8w@S^bsB`Gp`fwEx^JoO~Cwzvn0G}>Ve
zB<6#AJ&BYdX_Cf0p28koV;ZoZ7K?`lsYS=oLOj;6)|g%_A{Ge}4+;8I9qlX@Kjs)h
zAY>@P8k!L9;rg344Bcz!L?MO7mC(DD(1dX59D`vgd%y=$OTA)Jkb3GI9e;!^yY$g6
z^{^@gg(QR_gj4Tf#wxg{7AgW&-D34BqdWI_l?OOu2-c*udumB}_tcX%3guy6?c=G`
zr0QdgmLQC$=*4$WDOnl10De}7X`!Q$M1KICQ2?&OkxXdZgEO=iy}+5QTfBiwvtcT@
zlX(ChOJHvCLNCb+#SlQVmw#aaIGfipnt5^>XaG!{Whl@fjWvKVeJoDx3fF_FzWsLf
z=D)wJSKnP+Y__Y{pZ>LttG~U!_)l7Wzq$N*eR)>k^W}W?m(^caKb+Ow#jEwZ?HPcS
z4Fd{mW)8=p42%O@E#u$$>br#QU$1`LyxFXtr$1o4&RB2wi(`eF6o3EYGyHjw62v73
za1Kgn%Pgmr(10mXRWxP+WQ-oJ1RN{-l_6S_jLuK$K<cK1zrhG%3YvSk4)D_4;5<T{
z4Jd_6W&jdOaTcHHASw;CCIMjuFr-}R@caz?t?2vJ>ZgDF6CMKBsOjR<hY#nYL0bK=
z`LMZs{ma{T>%`G6t$$v=-EP;H7m0S4fBwAv@%8rYcAe_(#*59x7G1x9S!%FsbpFDi
zCJFB@z^OIBIWA%XfXn_u;$kTD?Ahw&<>uY%_4aJ_^7)I^oAu}I`8CVEeEaV;w7b}@
zFSZ{eitMKGYW;EZ>GIwB<1Vnot-r5-e*gCS&F5j!4ibSS&VTX#CEi3U8WY*A#K&Fq
z8pD5cHUe18X#_9`=MP&q@r<yQ*gPX3vE(X2?v{r_Uo6{4yz+i?kt^Y)1-z$YaDb*7
zXZR$7m1KAbLM3NRZ#H~$FU3GhfNp^kfq;)t)B6E6J9g+_Kxf`c(W(WhXE64ogA6Q<
zNa&A<IPEjUsejxHadw3`r9I-57eHJv1Y#R|%PNsHVGSa80CTd<)*Ml+h*l;u>1GPb
zUjTWC_?dt-HjEm`D*!y!fPp-uDPIM?9AQq|X{P$J!2~ztKbeqWJp$a0Ky`$@QwTdk
z<z3-sbq+VbJ>0IWWaT!ld(>)Y%WAMCx_@lfJ#D@tlYc+NmThKB<9pdseg%@(8XO>5
z6%@=*8)hG1ET03<{60MAa>a9|d!94f3%&s1A>qu1j*xQZ7KK<4AM{66m8HysERiu%
zbo1&c9=&y6AU{Jv1|{sEL<<T6j87?nn!^f89juQbuS2obmZ05z&dx$en^8o!iqTPE
zHbSbZcYo6&knFdj4WHA7kRiqR9Z+yBb5HQwctTWc*F7PM|AQxZn;f7|!sHunOXL7p
zZ&XbHBeGc(AX#Edq$4w^^{G+-_S|7Sh9`7X(?n3FgonU0SWl$@e?ksm3#7gvg^C0o
zzMdnI2e9Ff(8A#vI!{2@3C&3b_(6$pu*bud;D1sw6Jls<*<<Ra!yQqxXRUT8CE#bV
zk@VHTbG^ajh>XBvO@9ZDSj(4=Soee9(Ge>h95HswhTFwv*@NAVBM7wE*K+*UQRaDU
zOjWo-uTKLyf`7$sBNupkLs%@587#n+5d-9I-akO)w}=79t@behBRA{+A6p9#gWhbP
zA%C_$?*l!*f?nP4GsG6_!x<k{o&v*WIYJ2v_}mdmTyr;nBE~1{d2DVD6<AuPep&@@
zDAA-j9K~Ev#_<vAYX23ktMhFJOenk6r{%f_z-qr!aUE8Bu-*|?T(4k7bq_0|Ntl2}
zG6`)G`Z601-s3nybL@#%qZ~Wk0yrqyoqtyT{@AND@=c+Zk=`TTTaMEK;g*F?@3a9i
zOK>i78$*4`#X$5tvW{MYb1lma*#=e~wI`RTJ)~6pP+ub5s9=6MXaOhByOXJ#wa`+L
z`wJt*744l$U7`?*gbTrdG&VltfGhSMVG%>_IX6T%)Fp=65>|i$(w8s<fSys-B!B7(
zT~G#f6eDqd*CRa7PcAyEBQ@2o)nTyaM@&R^oA`IsY9If)=Qxl2t<aIkjE;ovLr0=l
zbi@<q@WeShrCh9&V3Xi`LCPf51=2F){H3G*LHcbPhoU8`dXf>E^r^fvLNewaRhR=q
z?8wtvoX0Y>=})8uY0a<%>ca>NOn*pJE2tZ~IX@LoA)P{Ipo5p<*fq8ok&Hka%XGu3
z2(l(D7gz6~DmR1=UVyW!Nv?cQf<(a)?X(gCI`taozapTC!d;d)q1jao2Q2_qDt=On
zCStW!x6UjGp{N`dKxS0ru{pn2f&(M+mGyWW(w_FN@GIcCD3Y56N;<TVJAXy5AmlwE
zo=zaIH8>Q@7_*xFW*EEKjImq44`cUhC2^Y)$TlUAZAu{9MCPq6lQ%Yr??Eb3H$>}N
zavxuXAdPABb9LYy6S1yDNd9ZJX3L?jAzf)QzAVZmKIbsc?0&=MF;G{!RR<jza=uE-
zoR-OMmo|3DKmgDsHn8LN(|_0jWndnk9qn~!v0E3!bh8eJ%{oxw5}W6i75M=*7(X^D
zS+YkC*`b8Y3COuP5;l;hBF9q0yGO2n?dyiGWNb&R_UF{_)?2iqV5EMFmWWxoJ<h4m
z(NXuQ?U1|jiO#p!`n|T;^y<{7_9^k}Q{vYb*lQy!tjA|h3ap^o<XZ5d_<#*zU470T
zjX>;FG;)2$OV-)~j@ttXo`TBaAeumBNLF?m5p+Xl*^udRVM96!?qKb$Eb)C(?C{Y+
zi#U7Y_{`&!7K}WUBbgI0nBsT;vC01c7Nc+_mvLSw6ahGwQIi)Gmnj@RCznqmI0k<V
zzR$1l9~eaxsTVL9G&|V@=q<tCiXMtZvppn5(;oWwOTGM{Jv%RFfEHcAJGLy5(kGIl
zK3R7!o$fwt9-cOD-XnEh1q<GEPoF#B7n=Wr=0ayX51<@E_w=><X{WVud_C|*S|jr1
z`9AQk-ER>B*8qM4mM_?EJwa`QhA)5M%at}<ALFjbF^Bptl{=OboKEe4pCKlR`=6hF
z-00eAt)fe}=|?_2K5f2j<PfYQMHIT{ubcFO5EKT}0a#_7@BaL?`@H!m#rDcHw<-Uu
z?*4q~M9HU$V$_VHcFx)fil7-<>YR%T)2o6az0=w<YV@8j9z7)#K#Y>RrWJo+2T4~}
zJSkztZ{7!3)Jj|zt<-9GzKHR+JQtSnw&|WRwzl*j^(2@3Nl$vI!7IYqSTtJ$3&>K5
zrBNN1e!t(Ac^vEH(kxq1!ms_yqnng`8uS`6iVaEX1-sZwLrz7WLQ&gtFY)nm-Ii@c
z&&Dh)a;P1Zw{bSx`f|N4D87GYSU)c7^-g*3*BW!Mim!9=4pA)__!UErL1~(S2W(14
z3)eVbDkcX6bLIphhXnSf8SIBT8jIcVFYBF@jP&-7Z-YWM$A(Mq3e7p-UAi*`;Cfdk
zu7pxXvX=7X%lCHlKKQ=oEMT6FVJ8C^Iz>KQ5Wr3br>Vpgj8f=G0f~QFfHEaGrSOsi
zRMlqR)i6`*HdJ$2OxHfgokMw*e?yCHC1Qz~R43zBkC_hXgvSU(^;N+UwNa>XrW%*+
zTRUZ~m$iHQUcpX)cu`%`Or`<{|GC;DH7a(K6+H6$dYlcMcm-M!N+Iz!50mbdJ?{H?
zSX~dr5RQ46H_93_gHV6p6v}yE%9~*c!Ib09Tx~gx<h)d7rNrr!w>c~Q)^WuTU9Qan
zFmf#UyGzB~h%39C_#pwr)YZ+KKVCln@cav>f!)hrsRbvs%2<e@`&u3?af%z!bzkZ;
zCkV(H0;tF-b}}yI2>5+#*B3l@8#11=OdCd}Us7pQ6ez>m8s~pZ+NX+D{-~t&APMYC
zo4POK!q%xlLmpAzi}aMF)VAr7+qYKfF<s;5R+U@*)@?p_n~y!35&19Y(%LP735vk;
zerE(|D@s8#Mh>49jp*5Tj+BlE+7}wOMn}j)3(9lMNSS=q)(^Ioj0W6_SXPtJirTL%
zKLUUlYk{O#CFg&+k=vz(ESV%!R131h{qxt_?1dYB82uJ8TQI1I<Mq}eDveoI-XF20
zFYPjG^;Wz37QWr1-Ib7(!xId<6MCEdt!ZBJFd>w?@Ua_pu?%(FkM-l(@NgvHo$ML-
z4AH>wq$Yz-h8eKMnpx?mB@8KYh=@IeIl3Q*Q1j-4?A?F!DX|1Cvk8Vc4fgn2CoS*%
zjdi=)<%hhe)RLh?fDg3Ti`LfU?jDsiz_5q>d=!e*!tT!9F&DmrH0}z=k!6Is)?97_
zJfUoyUgCwL-uq*{r#kM#G*uF7qy@t(3rG!Vx~qobWm6-o7TKmx_mxdHkdR2A)Uiec
z<c?IJQYU{CfH3XQg5wT$zvQ=r6Bft)y_;66>+5#3z_th!Sb}%B#B=d9D;idJjC)JF
zI*)DE%TTvS{sOyTyH3MB#j_C^rA-eldbZD)4Z2}EAz2Q^D|7jwuQ4+BLdKnToZ@(;
zc5|NEmU_8y5eS!z9rWI~R;QX`&Qq%sReLxStdxH!R-K`Rhl`!)SfoUb_`p^mpk6Oh
zmM9sI?>wt5oQjsj`77JM=x6hrOX*L8K|!=L`ZF43Sj#==PlBidthtvF66XZu#aj$;
zPJuI0vf!Tj+MYOt1X&xGh|q_$?cn))Ty2+)93Vq<ccznt)<h>aug|_*HIuD92}h6%
zDw=<pw9_=QeYk1+hQV?2?+0>p^&LApt2D%W&&pKmN+iFKdNIjgd68?j`jx;JJvg@2
zx1h`ngf_VsG5v+Rmji#intE=Wg&c@`m9v<Q<hnDH%1BDx{xwE&L3PjLH#g9|olZ(#
z@s7q&O7A$#QqyCW-dK4dA;sEstY{9Gv!;K%-43%Z!E&aI$jfk;!PDpBQ1_+DCEkks
zyCzP_3H9YRiR_FR;Y?THn#$1(rnL0j5;iR{_~)>tnI8>f#&n1SY<Q#cfk8FYx7H)*
z9?HyGEYD_?A!ayjO|x1?8wwS7`IMA1lm^e5YXKWgB6!xzc!b0|o|iPJjx)OBfPa6T
zrMDn49ptfNx_XQuYF~4F(n5>S>lV3P89s#igP@pKup<67#;$2aLgMY`DJDi<=FX20
z?*9l;aPu78Y0VQ=A@RQ%A;ch0L<mv)h6o|`+H)d=RDT%5at#PXDv+OdX1Nm-6=<^t
z2-YhHGqvG$47^t8X$gdt!({qOXVrhTRtQ`nNb<yyhqHsEJrX3tzbZ&BF-n<-jf#v(
zS>>DuH5wKB+~D6MU!#Id3NC4<JKBf#aV{~R)U%fqb{Fl2Pm#0`KvC!o1J~y$!I1Jn
zDx&3~p2H;qn=(czu&}uMtqJb9`1Y3iyYrB2wc~IykXz~W#`Y5ZZkbca2PuDm-v?g=
zr+5{S_#UbvGbOTf_&D!`3oQ$dcminG2>N>e*uOc{bF14Rw*!o=Rr1_?wieu^KYpTT
zfk5s1QswyRiBNM17nL<b8uL(FxSq&w(Zu%$3$2%-M8I0QXoRKwTI$=8LPL`Ij>!Au
z`w-tiy!OGGOOueeBJnLGpCEs}MeREg-(XTSyZ3k0LhFshw~%%&@x3mNyvmaI_r`Im
zbtS4=NWI(l=tA#_@6ko=s~GC?ExOcu+XvE(rIu3ZRhC*}em%b!Ss;W<zs9`F7nr<%
zS@TQQiMUr2j*E@>u(UijB*ZT-l>Y#JcyC4uWo~41baG{3Z3<;>WN(*%UMK_tGBB5M
zZwnIwH#C>Qi32B>3Z5t~1CXU+mph&)8-Ly4^cOfSYFrEe(-M}KU&}uqITxBfd-^EO
zrKFFc-<H1d=|O^}9YA*g=>ty>MnQT;3i<$jbbN;POZOPzh(o&P*EPYhn)~N~uEBX|
zPyc%P$BrkWIdj_o^k3uEr<dJdyO9OWEfZEP-+t}<4JCm|D~K~@h_S!_vwYuu86cq+
zTqu|N=kFiOu<|tt(E<sPYmL|m5jaU!Em}J!{jHbvo+u_M`jDuV7ddoTH|bZLl#dfc
z<LOQiZRE#)b21EKi=*iG@Z-}%RH@SOE|)}~C`SRSmyDk%ihrv~2Jya5H1cj0R#%gb
zgPx9MHNotp{+uxeD3JFiSw>ot{E{rg8hlL7cx6^0EX;B<W)%Pm9NL^qE9YZgNGqqq
z9x7x8)D>Zmf}0@>AQ$ks0^)Av>-o?)U={LW`)DaCP_0ib6E9Z=)ML`2^76}n0(ad?
z6HtW`jLMS{?|)Eh)ZLmnQsJ;0_t)b^vJXD-3&fk?gRZWCs;R`$c7m$bxiNyF5R=Sf
zT7t$Dqm2wJD`^R(Vx|^2u_8<#5x>dLlIBG6PtCJ0`e^K}H)vVMSkl~e<!Xo#akW}P
zSvkaPlCE-w#2Oz%-Z>OoN)a=}6_w(9io>eX5^`006@T`0xVGYZyz497hMOKjr>Sdn
zASfVFDn}uiak?*A3WvoDn=F=TG}*7jO03T%PR_20c3bb}95ZlbJLy6L^-$%6w$s!`
z9Km8231iVVw0Rw>265l<WW5$*#Q;sbxJB9^m~Bd88KOb9a9%VnYrz=oFU93x@yAQb
zW)DSvN`HOT$Ji;+&n6Ojj`D|l__FBYxveK!pXs%<?OTKY%};H5FU-~5%`)de-?l{G
zrb=uqO~y~Dgs_VWlU1MKjn`;^sa^!QnF~erwbQku&e6$KJr!MKdw10;yroy&7eB1g
zP`1_bsY>~hEbA?<Z(Tb3>LOyOmJf-8&JHf4=YMv|QkzWzZQJZ264@9TI^1&vThB7C
zS^@IYR<oV~JO#4Y?5gdlF88fxn4u!BG*@2`n@xJrNTvH`siS3N6GiH)5rJgzsmn#?
zG(WkP#l9Ly_EWNlJLADP_HlC$*H!?X53T4i#Nojsd@T0+Z;1@4ic*ToZd>ZHfUSw`
zihtjsyH}LA1lSsVo+JkAeyW{S?xSNjrJLi9W9Kii_7!NU@1cs6K33*~jA^xY7Bg*V
zm{b#O8y4Ij<7{wisj}sEgcy3}6|U~aiQt-PYeu}Uh<{t03Dit^TNyB<NGwaATjJf|
zE2p~hlcyta-Ra2dJ8OyLD$<RewQ|f-n}7NqjkR>t5j%Ou^TYw-9TN1_jVBl~n2D0C
zq_IU5z$l$gl6af*Tj|?ZgYSHhF<`yZLB;^}-3~Gayk`zFhTS(DWQ=(4bC5AYJ$;Zd
z%KY``>(Zwp(6hbQf;blmP@Ca>`ejWi3<rwW!xe?UG586jJrIK5Ukh#gyoB8ZKz}-m
z%m0?J%s`-W-sS;}_FyIb^DOQ@{Q2YiZ{Pk+PO$v=uir`$lyVq2yZnlmhJ+=p*2~X$
zZAloEvw%4+wOXuT6xr!^eyl8fTXu?yK9H0V>*`Ox>X@>MG*KT8QQP>95qw=J_&vO`
znz8+C7c)&2LExtl;jcWD^sK+6=YR84{?dQi>%5ho`hD4bUv^)TL{SS9Qaq1<ou+?l
z5~aWhesI`GH7{wxB!#e9CeBdw*jR1(EsKJx2TirK5|-*0P(GL?Y=!5%((s*^2v&xW
zm-6?xz9EN!S6Zn>8be1cI<mcBo{*#)C0u`h`<0%fk~jEj{*G`SKuD@!^ndao1#MT>
z!wNYW?P@8Pw&&eBTIVt}j7HKBBVt<ft^o+4{IWD8Mcy7Nb6+kN2mnJ=7YL+=g3Ig7
zS!hC&hZ$Lfwoy^1?ZR0mn>o%66*Wq~xCXJE-w_i-H!9HO{>tV$wVtiy4t(q~@NmDH
zTp{TkN6PP6lRHKIi<@MFwSNOH{HLxn$>}*?xgyJawIV~ab#jr)r97nGmnX35IpWGy
zdW)r=bz~j7l{1M(kEPzvd}Qu`_tV5|g;rBMSADN?kqtZuCmt`F`j!sFpVtcqiTemc
z`hby&N~-d>lm&8k#@9>}bu7_@)6EHXge|w^=GM+~;%b%~WVv<NgMY5=TXNs_dkJc>
z#@OvAI<dp78~ZU+??36;C`P%g=(&A6!8S8~RVNbnJ42u>>%-GP!p32g64jmN!9qQ*
z&td3X08sv5dcy9H{Zur*LmEk|GLSNjLl4-)F@-W)0lOXRM^UQs>(W8Jg2AjKW!S!e
z0PVhx-VN-{O9fZc(|-&bR!u9Vy_KWgbf9s?O9lBrHC;`w3xpLN9ie#*L3JzTz{L*Z
zwKj}|kZzk7M&jX99%(&iI?M>8K*mEeC@U}7Ww~bi7!Ds-DxaoqSgU+$6%#9;KbC%5
z>P{Y-_9WMARGACG_1a;hKSj{4`a^j&F#RUpy}3We_X5l84u7*z*zG9d**$mG3C_`X
zDH$p1&u@XaeikGko3wx&0eiZ77f8D%8uJ~#@88f0`5naCxoVEyAX<aVZ?5olTKJ*R
zT;CDO7*vbv!efmsnNoejB-zm*Q~rGVDfCfUjZcMa8K@4^x-=K{Q<V~Kj#U>;?u9(r
zsM)g9l&CrwgMVtoxiP!GR~Ay#+nN=2o9Hbzj7qGU(M6?sHg{{tH|tql72MsDtUzi)
z<%ntIuUF8U@GQVFZS+V<ii4A645z}p4pvkXN^-%Y+*l?lcS!(<yVe2#^3g4o%WfHe
z-R_n`ODe5l<nWAJ3Qb|38kOOx;3AQq&Mg~PQ*Lxg<$u;(GVp%dB@3>%b4P!A?ayQ<
zKJRLaN>Te~CuzQc|0ybR0qlh7?fng<W(LKGdCT9B8e!wwvzq3GQ0VHLwZsp0!>3a9
zliRCW!(rdtHZPhcyW`W#qfM2_oA|YFDz{Rb)gg6CQODI$n<;f#;xu{ThII<T!2515
zEo@&A*ndxcv7Ks>2{S><mEqO@X@YE8O#f%X3aaPcWdsLga1zxWM{ongc_?%B2n4id
zpv4Gz7=M)A)EsKBpnkVp8xHN`n>O(2ZX-4TYE7kn=Mme0_59m#8<Mlki3!^I(OP&5
z@b$wW;M6-AGY-k__+NFbZ@b((E2T$zr?F=b^g8OPXPu;7KZgX$p(9gp;}lXcP7>{b
zhL%;H97GOMc)jUSn!RWU*^C~cjvgmEH$wIBO^M+Da=%BZmvLSw69F=p5ilhb0Wp^W
ztO6&0&05Kh8#xla>nr#JqPYbZpdfUWTxDQxZq#ishlVk(KG=<J&tZOlBFG?<$xNxV
z<QQ!rOA?ce<uO*$O1udrKJA`fcOPD?6G2AM!HCzdLgNdCf3?Cv4<cwSz18CNyZGmW
zQpV%+gpX6oT6_Y2kN63Pz10zCXndmqAAQ(=2d8<BHTY<Jd|89@ZF(ql%#q*GxJNqC
zD_W=Vw@&!@!@pktu~RIxlF8Fo`WSybzwZ9pjVgdq(rYWeeBaRvXSKA^2n|U(WyFtv
zi?6%SBiY7G;hORXl=$%`hLK+x#iS5L<vqX*MXeB8Cp<)Hkuam^J}L!>ntQ<qwRePn
zf}Uyl;U(%-iVl^rFj54>6+)o|z~2#bNM(#TWsdkL+&@S2={@4Ej)n6Ds=lkI_0om7
zrJi$&q$lan^t(sz7uMU6g7r_<dp3f4SE-CiGNCksRN3Y&**tyD^lK}fmkIZ4Sh*$k
z+DX{d#AXCHu`sY=`jfSsjdVyit(2929=-H^q_@BjB1XIDHUx2`7?>g>cApI&^tL7h
z7!rfu(uM|#qx;?mM^S$RuLdp6LRJmyjtqckF0k5i7cAy2w`C9=DgGag@Olpi<7JF#
zg8hVduchbrsu!1(47jyS+onSZ4@C*b3TGwY;2rT+2Se}Ily-qCoV>cTwh9}6L6%n>
zd7Q?4t%;Dn(aaYd(>#`OlgArv*Vd#}IrA+2k<q@g5;B0nRJ*YfG6tl^NdGe`A?Ksi
zdPCcu6U1%ENRf$W^sbKv+Iy+J83-G-8Vz)C(g!_EwhJkW(<oF?KW`Q&9M%*%nJd6V
z$O4kbVptQS%yl;M>ji+%xf(lv`AhZ!=Inqai6?-IGMNQ6LJ8K;dpYb&R{%J}KHpHe
zY3Y0#`Dr)ca}7nn*s|D#>YF!FEY;;m(XS^W>!h(}VFl{0XzVN?RXg#fs5tFIj$r(|
z#JRdJb~Rq%^UUu~+6Q%mycXx-@+F4X=z+0J;XWHgC$ze&SWT%Pow}cYy0+44SWt&;
zs4F9FS9w?Jeuet|QN%{HLmPi`6dAZu6uB&~u1Av?klI3*_oK;BIpV0vi1#rX7|;pb
zI&ubGt|>=wvl<uWH{+eeB50&wedlH@IH_jY-Kv`O-Sl(fl!ERE57sAaUt6)(Re=mB
zx-NrACOdb#ImErotiJJo<qeuVzhU$$eZ@hAG=5?PZ24`vGOVKEYEj1sYJCpPT-9Cz
zD2Z7C-O?*vz~<X*SSXDY|JI$T#pLSUmW}v8@#l1+H8<W+uW*|=h@H<r&YUt~8H&%i
z>Xx23co{-i@Q3sKW0K3=yM=v<5<Pao-k)V}$$GR|SdZG)HkHMH!f(q{l9A%ysy{c$
z$bjj6(~8g}7THWeJfHQ4%_=|>wTvph<2g4`4v36<|HR@3%8+Dt6ymQq3c-edZ1**m
z5Ll#QxBr*Gd7ng{jO%TRApR@zFcxk252*$3u#r_%c}7p-?mqna_Vt%9zhOxy-u^>V
zmB6AJbc*78`H8K6vWXT8uW$T2(q$0_AT1LnS0@5#3OJ!IYy)4!&Y6JRA$FtRXcS9~
z=#1$W=YxuX6B~nWsLDVT+i2zYtsksXO)RobO$_=)RCq1@#PxL*{iLV=U8mxz-^K2$
z*nMVH!TVei$7s;OsXmSvkXBH|wp0hJdt}>Cln&6&7?c!$(0xt}cG|W<BXJ|0Mg5`-
zX#Z54GvTIj9qF{LLrAMqs6+L096y1=pec?t3&rG?v;Y(%WCPWVZNYc?{^dKbd&5<q
zy3ayquZ=}Gyx3EJk71u^<x?n#tLK6|z0R&hb>QN%I9lbh=2`;*WcOKg;uO91lVYDQ
z1_Q&bvi1UhFvhGmfk3BGI@$)piNjHzt|`YRJArV}p6+D99QMYgS|lC0hq7IYOE{I@
zA7?YUoPsd*Ywi(efL^^nYHwCj7@eIOP!FYS=d(|q-_qjCNa57^LZMdImLk8=zv48P
zX>c3D>s^!LM5k%;*>Xjwo`<2!U~HhmhoXe78DcSiI`z_vjU+b?I5ceJJ$Izwhr(su
zga8W`rxqqecmA&Aes0Qc&~{;_(POZh8em(_!0TRvs#+k@0meNJeo;`>QFyGFFhx}=
zn>D=P)EOKWn%E8xG6M7KtDa_;jeo=5v(J^$N7gHKw#)OIhN0WxY^R5Pw#=95u40(!
z+0uc3XC9|!Ie+ISo{=%x$19B|e*f6shnd93q0z2QdF<&meW0zo-1ACQ9_)gH-TY{U
z?@?f~zj~jtycwu(SR4M;yr+(2Ib$ay$*lVMV_fTQa2EcRy`;CLmF&{R@wMzSWr~tr
z{t)wNT>kB+`md%$EPISWzg@X1)0S>B{H$Dmu7xdqN43Z~9_FVYXU^&N*2Qs`!+ch?
z7~bLcZu7J@Sa-LqZE7}pK3I6TGu8mI^&^LNUwHa%V{g)A%f+`Y&~y<;l$Hj&?SVgK
z51Zqu;U<#~y)XhI*KA>Sx?5~McI>^r$>-V{HFag#Lu#}<@@}d+bC!8i#sQ|g%dG!@
znCZ*9$3d#dk~V2td`a=z44h^+u(d96Q<s^&?Re1nzn<lUluH#02j?E9YX%NQeErHQ
z_*oKMn4eNzb2dbc7H?_<Y0={h8XT(_jV{_;b-T<tjBl75)&9=8=A)YF^x{stMbFE9
zmf<zf<@=s%CkyfCj8pBBal1kH>!hTA+37S&zS|Mool(>*5!zYAsVc0RM~pJrDLB=1
zP!r<ue2Bvy6_g%`kaT=_c9OfpUNgFpd*}{i*s8NcPm`lxDIul}8N6MWu>J3aR(YM1
z*;p&J9!b$ze0An(U>nG7m6OV>h+T}=66ePQg!zBNi0mj&UE|}%BE!Y*M&W54-K5V1
zhMCLF3zpD@-_q*T608rni&CO1DgFnyNS8q=EEEGZGBTHOl`|BV$gwEMf1Lk&{M*Z4
z?|5SR>E&)@QiEYA?DYBj-LDgM{4j;ycbtK<YWm+4nwbFlH*sb(8Y00!X{)FIP{()w
z{`%!FpZ_ix4g0U}$|qz}31_G8=~gmYuaeWZbdO9MOTz_o#5GQ+nX-~P<bkE(^K^&O
z(Co@Kt>1jBV`x#zW-tf_f7bYpd4i@X^WZ7g)->O@wy|(krihEkZ1_7*C5`pBG(L~=
zxBhDHb5|Pm&*|>VboVi#s<hUIx>zeg(O=>kaS2sHdt^wI8f0^mM5egpE~DZMS8AoT
zU2pBmXe$wDg_aqTL72*HWzD&zS$A$CTA4y_Dy$9dD>w{mrIsS3e~Kw5CIBXB8%RV&
zLN_8@e}Dd-#(kwxADX|B&b^SpQ0eJj8$ua9si45hE-G;9ojND|^*V;6Zwi7=<vEsg
zs~=C3eViIz;-c>Bsy=Q;LeOGi3lf>D1q6^3UKIpZOjuq)?3@ue9R*1Gp~xgD>&=#-
zJ0!z|WQ>%^hD@oWfAsy(#wTcUU$Sm-H1!anN1@{TsZ5(Llc{<7p;h?S)bkYY{S3;F
z(xvyQ<HYxW2=fF(ND@fOMo4+hw4lIi+d}K73Y~PoxPbu+xs9pT`euRvmT~C@f=Va6
zbRu}F6VQRMX2$!CS`oD>Hc{06HTB)nS#aMR23b8UQ7{^of6g~eWVLk#5lpLA9*rY#
zzo=6|;nHcToMZ%bku{mIEL-_LiP@<L6ZJt?Ha-Yiy7ErEuTRUb_f`Cgu)zw=c9+w}
z#P!uVLYr=tA&l^X4yk9}(&x;55mdb(sQI}|+&rg|GGS2GMuPE}#M)^u8~)FS;f{pm
z*V(C$N--_Ke{ZyZ5Dz~|(nN=;q;vfr)0exCBiST<Bul~6`2Q4-`qQ<1O5Ty<wBKB7
zI%C@XCaJLsCuQhZFgEWuKE|C0t35^BWH0vbo~>A<h$M$}ehw6~sR{BBC?~AY^O^q(
z>~}4V>I<!z&~l_K`J!G6<NXdX_J$cB)IdAUX_ewBfAMq~(u;VaBY$ECoZj7-yl3an
zk2Go5@@WI)-K3J1s$9}iRVVglJuqTdaI?(Gk(h@_EkXBuypK+<`T-!4$KySgH%76a
z)x)^W@GQe=$%w;jEtm0Z=Y%Ie&u)d0<c&MB7UPQefY}a;WT_ONtw!Jt6dg0l%_u%z
zPf;V|f3)8t#pUB)YRl-F$r|tW;tK37sn2}dAD5YM<b&W01og0cTHthMC^K99^QxT$
z;TT9U17QD>fhgD+?)>TModnPnC)x|rR(g~EDmS11G|aDj<edKt^LH)nfs^f!(BxWS
zl_a`&)!E=+ru$?GlH7pny~yZm)AcmCtY@4Ve^7~ri{oI~YQ!KYyP>oZ<ku;#n31YI
z`odw5oSWd(LA;mEjM-sER~;tbMsOsqkHO$yU_6t|jQRs2FJR6_B)NkVpMc4+PS}NX
z%z4K*^mj*b=%WDXl7A_d9s95?uKIVG%b2_2T=PJ31#NW5jMlG;c08s<3P@DSIR(!Y
zf4A}+uWHQJfD7%JGDIfw=)oMndqy%C%o#&5Nq2D5A*%qOKwrN_CwPDS^PghCfw(+G
zGVTplJLC^}$k|z1z&Fm)4T-btF$}A&1BG^iy9q5Y%T2MN!x8Q?ipcXm5T^7?MmryW
zS;KRiP&b&Z#Ep^6qFv=iEptONpW{p4&X*R<)_+KK#+P2<OP}S-416sGOP8WvpN7-6
zW)+gd=~Wp*OsWrUvLxBQpuoo%H;NoZs-Y*ktm-uxlkHIWT}Oq3qRiGsqHRk@y*%9)
zBkL|jqYou@6qPAR`wT6zWXw1-L#xX;gxLCArUATjDsZHtTKi@)CghFWM=MzPY?D(W
zCx5ki{LCVLau!AuWG_ccYnad`P>eBKP2G*NSa`EU%#Wje<pCue8moWmFu*8zVVynd
zGszZ)%xkA}HA!d^o=vtRA4Ekuo0OPodsXahPMc+1m;2Wo%c)IW=A2UA(?G>^*``z8
z%{ECH<hf*VmiN{+Sm_=s1BdJLHozr4HGi|b4M(QAeUrQmxTr<1m$w-ycGh`+&PI_l
zPRGDNb;X4j+81ifjE5AtT;aNdC2v5fA95n!p5Oo<s={k#-A0P9w0k|(84`PPcxi_R
zijOx?j9I~vqF%{v(RiU}IPsf<L{Ppk;VMSIlOPdLG9iD)h;I@j0v!>g0zYF!(SOR_
z#^9|Tu`NPu;odq)JLL5vMDIIGE`M_|8F0sBll}Yc#AMQ#)>6wX(^(%+O&L4KQ5N<T
zs!Q+U5nDV|*;hKn6R}_Rljn0Zc22dW<Lm=Rsw(-bal9F-pOEybt(aOl#xY@Rc#ip0
z$@G0;A=fIED)ZPBB|OwIVV-m{C4Y0Z8r`6!vrUY9u{FBUl1UP~VKhJWn09>E*AW(0
zkJ;8?LK@q+dNKZyS}|^`d^d4w*!ia;s5Yk8*GC1y0uyc)6&S(%TQ_eJ6iCES(|A3%
za-`UA{53A+Kyg-+x?IN_9l(*|ZFM^-d~7!oy<E1twp5qOV3%OoQwEELt$%^yEu}gm
znUXW0Zz<Io&5SWKifVaFENy*w$F3K~TT695j0Zjw$(33icqJ1v6t1b&rOMm5`ZKjU
zAScDn75s`dr()T-<{OH2=*W|h7v!z<rf3P&T)%EF)=41B{P!1!Nl8-pvsU*uahRk4
z0o>0xi~^lT?ly)<?TBr0Sbq!m)=AnSuOEjI`V2uf*84|P;DmJy4L0Z6o+$2`Xvw=*
zQoFH&6?Ol3Pj1GcaRiIkyZd$AwMapLqsBQO``$%cU8N<-b8e*W^+g)KP_c{a&DDeW
z?q)%uMjNpz!2J`utp?^_gyO=^*14|+Car&G*Njy1Td#@F9TmN1y?+n+j4O0boeVVb
zZfY?#{z;X}@{)M9%pFFBT7I`S#?-l=J2ZAXMX7WsEmZlMz7OqaY|o`rS1nzN@&qad
zP2B*8Sd!~Q#nQKgYz9wqh(nEwn?sLqc=EoWLp^L;O%lyhovU}Et=`#rO*qH9P3*cV
zbB5v{)U#DC)HK9nT7MHiUxOu|35**lj{1_2&;$4Rz_g{uhT1X3;j|6ojC*Ux)-s23
zq{0KQF30z_1b3@>4emW~bqP`2jwt%KX4OVSOCHO#(ju`)maeCJko;^lyFICLyQ~x5
zA6B*U+kOMEV#9?dD?1xFfFcK#34Wfe4jd#GI)uz3z8ic2Jb%d{<Vc2@i#f{%fDTV-
z^^#$?CK;nPsfMN|<br9j8((xFUyU|#hQwZB6%b001C_Ah!l%EyhZ)R$;C^<5+5W4Z
zR&2`Ct)|=RCF5(y7bOzab>o|R3<fsQZ=_(wxGoB+6ckcoC#Y+NSF9V}{Bpo75gK0u
zu<5X0PlV=|3xC$WTpGW#(%ok<@#sEp#x2ukzTWGN4v@m4lOsj1adkcOmznI6VX7b6
zteN`s>cx38?=+)XMvkOu0`DXZL7U2}ExWY~H8R8Ps`Ru;`1DDV$Xb8?gDQt%o33Vk
z_cvEYGg%AB&`Ml*O(m@5)E6{-AOQvsgRIAvEyanp(0>p|;j0A$Qd-8j85VYf5L}86
zq%({JVVJOdL}p;`3vv;WQ+uaeQ$9MCR^7O{r#4tGyO&=}i~Vo^5)vuYQnbopv8zG?
zuOTE$xjY3(v9ul-xnLp6XNyQ4*pD<2g4j|hhKcHjh34Xmf5n~e!t?s|=l=nnO#QVA
zWo~41beDHtC=CKKGnawtJre;rmjR;~Czmh*IRux;y(po7__*UgUjKgJnQ6|v317oU
z|8{;o{Cwy|(A+blVfga>5Pkq9m~wJJ@=O^q{PNTA_3*hTTPEW;CjF=R@XOoKwfvA#
z^c+#-#%Mi4QE-G724g%^;irTmeB@jsYVd*&N)G`AF_QAbixaaHf#Z7Uo+1LyHDU~i
z8AQs78iT-pC;b=MAGIL+g|abB7}-;ne8gAjZGjmp`rc~72R5L-aFkM<Gf(S~uHp%n
zfwDXBMAR&j7ith;qB!fQ<%bQJk7t!AtoCU(t?SI7Gbe5VImf)5PzxyOIrLtR4{1N>
zUjav{nHK)uIF4rxID$!%YqSG8e76}>NSwrWJCCP-3ph?>S;w=TQ=P*+yZ4L)!#o;{
zULxLMwu7QZN_--XfGa3^X0@MC{P=!~1{lZn?kVco^^wLgd?xD?*Fi?D+mE}`44+%#
z?iI5hR%F^o(J>CU#9eX86rbX5W^q^CFs<!mkMj}s`*mJPbgL64_{VuBlwnFu7GFQ_
zpB8|BIhcqeUKU-H(w3orpV+wlQwPr|CM(I$?_bH_J*8XO<+ZnCZU*vGH@{Xx{V&cx
zYH58v*?$2{o2u;i@N6zm$$SR0Nk%!gWrG<Fb1|jTc0FAR?SjjSqfCC+)vW8#eoo=x
zy6F!gV>aUc7`@?nA=*fhPmwfeD~7g)D~2k6-^MACi7@gx#iv@OciBuDg$fgAKMz?r
z1TZt6m|5YSND`v~7ZPB~Mk2X`lMq$McgVgwLmeQG{S#!wQCuCLTz!1coVV*6SH>rD
z5#r2c49M}*@JMn2ZS5s9pl*tGil`zqCOs;Xi=;R4oSqs2ECCnWiLy#eq>*Wt#7|Lw
zA4G$Kw5N<?<gJ?7e*Nhe+AUFY%)~u>cmXkJ=B?Fk9Kd0)6^vV>)_?x%U&2rv0`o&8
z<G~-aL;gfVF3!>}=f+t^L((jJ4Ao)?NebrhSn8lS{#A=b>l2Qr)%{7bMeSyib(aJU
z3*$pD8Q1*^lS`cAvC~TZ4$gJUd9>AkA1BWL8fTzGplFIWkU%<H0%qH6;}MPkEnnWp
zz$v%1o1czJWdHWt{@sRO13FJ`|1kXbG5iWY!Ts(Ge|UD9=2r2x8qIddM$_B?zIB#%
z$PG=i&`fYYXlXE7t?_nu%1#D2844(F6R{<SNUqUYUW|%4KA=jwOXLiZlk4Pv>=l7?
z77{<bT9i>=ma!VgC257^aEb7Ov@3+|r9^5`Yt$OYw6{jhC4|8m=5_JBVZ+3YLec%E
z=>MqrV?F^S|FxEe28}G~YFEerxZ?vJG08Nv76lRp>DTBt0*<=0EO?%n{%Xx)Fq2PZ
zR6d|FshyMISci{Oq&1V)x?LuJl1bbq*rONhU2q1`JKHo5CsJnInts-3GyNJgy>HW|
zj5xJ#F(;Z0I`Y*lPziKSt<^0(VzAY|<t5$=IZ2D;8BCSuS5}p|6TDWyKuvJZ>a?UH
zMft&C#IA0nBa2%NL+X&H>t-vuA?)L7zWaf^w8&ZGy|o%8d`yh=wHh>kx7ygd)M`Ki
zZp>TOYC4KHtkrasS86p~?>p6Mfa6I}+`CrO&3a|6rla<o)M}JN$^5-*HQlT?s?~H<
z?^>(rCAz*=(^0%Zt){1VeXXXac)3>7ug@)NH8;dP*J^Hvd#=^o4tJ{6Y{Q*uHGS{%
zRdpK0w4LA{3P4K*lU%HSnXxq2SB;EF4!+}Np==cHX4uIJzo%@JYbQOj7s^InFau_a
z!tS{dG$!;^=J(f)@C?k1Q7f8XdgFoLi|KB@zHX!p+Nqj-&8{c}6yZ`(&5@0R>^2QW
zfh%RPYCu~uJ+~_&)l}VexkQEGyLIA^T;g4Fi3;#j%HLrTX+EWYABznx_kGENoX5fu
z7flIc=GVmI<&soQ6_v6>lc0}n`^!=72HBX691Yo)^K2Dk>n!b%Tjo4HQ^6ji=X9o?
z(9dn5iB>u!>~daDrmy;T`O)`V^elly^tc{LphptGW(g=s3Gd2&>J3TC^XnvYAT<Vs
z%7)Cr&jWV+G{oM2NI-dvKL0m_W)ZSP#>7LAf^eq;!8v^R&)e6(efcLU*YNh=FoF-j
zq!!){-}9?NM~Z>L@GZX^43i}u3<ZLw3?6bFp2szIpn;{~%WwcRXae=TY5f*jJp%`*
zlyz`Sci7O8CTN<7hd{Bm=J~$0jfJZ+MP$0L5#9ooJT|<4;&>VrUc=MAQ&$}Iui@}@
zID9U6dWHYu$&#BnI0XY?m1rTQCTSSD7Xbu^w`hPCT3bPDS7uzOGJ`}Tf(jSTwRyH8
zEVVQOQxno63e;3s>)ICp=wwZhB)!D5p%1o-IowvLl_JBMpsnA(e9z<FaMY*fGtt3a
zrWj~5JV@t%m<AbW#WWULy16u7?oPdFLq8o=+^iHpiAmCuW}e$*9Z_WA^U$!EWPPZ@
ze%=fa+H{>RXRiR%N4E_y22Yq_V|>4Oqn%{Py(xG-jq;lmy*O$q9^B`0L+Zzyh~ZTF
zjv};U>iJxL-EgfcX3<mcru<GV<y+02r}w9pxA+TxCpk(!hBzPRbLVo&IaUZ*1pA+R
z>Z24nT8NR3eScm)tho{WN+^|8v<cvHCO-aUL5q0`>Wai&^dBRbHWY-J*mVBZq?Q7W
z@q{-q&aBOCjjDumTcfIZ+8X~J#@(_PduZGkZSo7nxKKm_Q|bxQAAvs-Y8j=rhy`|B
zS;3ZnzNk8XSPMP<W7Ufk&u%K+Nj$T1+ZZC6JzZ`=LrolX8rGNKa!(kIS>YNTL!<NQ
zZF_wy=|5;aIph)#6kEp~`BG+}>n@~&oQ^7GT&{Ihjshl;D@mu{wlA$$mM0?v8Ot({
zp*h@UeC%{p+gZr^*%tM*3&)4HyxPKtyRvqFRBN+kALsQ&ZQ`QW#B>(fu{Pl^j6lS{
zA|mXbvSdvKGFq?{r_#ZnBybgJ#Z}7>)iV=M%^8l)am7~W>rdIjCMuKK9QD(k2WWUZ
zwG<#Dme#b47FkN$&Ku9EPfa$&6h5-OgfrRIdE8Zl4QWta4bsb~Z{;XGzxEDdRjNyW
z3l*BtHl=z+suYemOTV6qX5FI5jz<xU6k)>IzG~2-h4YGO;d<(ERl)Sp1c>qOp%&lm
z4%6Z>w$*%QuO=$RtAp5V&*M1P`26A}#I*jtgmB0$hFHDBU;_u2d<WoZG*q8@s^=y`
zI3>MrBZRPTqE+xRdBfPdRPz&lOV$5>gj^eS&EYW7c`KRZ<<jKDhnaefGsh}qg`V4Z
zu|!!T3q<i(vBr5f!{a#KYvV%lA=odEEY?Z5+uW;=I;2AEMD~@t0jjJ~<j*X|V2b-#
zCG6;DtFsmp2>?8ky7n(tH{WK7(NLc+%dMC0ucRodjs6>Jx$%mV2p59tfo)rV0{w`G
zA5)HX)xTmiTCGCWtQ>RYUtP_!2MtV=C^E@(m*TQA$DZ3lc?MZeQzGq|B1P#_uEn>b
zf`!XH7qQM3fSV0*fwbM3@1jjoA$KPFb>dGeqa-uFA4K8eNVJ7z>JOlGVcMJPjVhdI
zv>|1FZa{jTH)G)sfn43n(_?Udr(n}VA%#YBp?hcCsxi2<j33{3g~$L?a`M;S%EiLf
zJuVP=8KZ*R-5(kgyy5ksG2xZlL*veOx;$hNfji%LX{WsM+E7bs+nphG-haPB3I=Fm
z+R3iD3W1@Qm1E?uxqlLKwQfvq;qr+BCY2noIqy_zcHefNXcqnte!_5<v8*f;12r-<
zmr;`!6PJL}Hz=3TRy_olr^zTK0!7)E&dDfpe|)%J{`dKxC!RaanK$7!{G-1<KA(Pn
zq9$l>nG|~Z|8Gy>g_6Le6O$9n`00<Ir(d4_ONCZd9KT3^G@t(Xbs|rG%ouu(7;<B@
zZZHInz^ch;$0Q2Q7^-*9HDac&c%khoU;xBMV^IUh34*SuxOsw#zh51!YE{@)YN}P>
ze`sl+v}rwSRyy~@H9e@(tYcnu(m2im{E%xLS7pQ3NeaKdd`qLJ!>E2FnM?(|%AaUW
z^W?`_pAf(ljtkKTis18zM9r+|MgiVWNc7BG*Gbf?2nwl~kfNIf2o^_-pkQ==s@<Ju
z(lCYn`r>)kY7O&DnKUyrdbJMsHqSYge|i$1hi7Je&UuVE&QZxX=G+27ok5!sEKhDA
zSmy{qP-iW#rQc0|H<8hn87(J+QZ@J@O@0I^4|@a{!@vkCP4L6>7XvF(V5r0sJ_c6d
zyn#;Q`w59T?huLVJ_b%Qu%5O~(MUvJcpP`HnOAZk?l^B5csmex!4*?_B<_Mce`dMa
ziMs$1R?KbKk1PP$701C|!u2@VOGqCFyL3!ByBBsP80J>AbzD+r>>m#`!YIa}oAUg9
zy%$z6EjkqRTMl*ap+}GR!AoX+H~w(#=0N#Ea>KOlu*-9=3(BCFHlt#4BWy%SZUl~q
z$gRKu0l5`;CmuJA$Asfikk@EDe+u#%j7LFUWAPryODNt#{6r)khIk3Y!w@fVco^a(
z4DUj`MB%3KxFFn)<eeDYG)~;>d<bq)*b#wSBtA~(<&|S_$q>*g*(DC4?9O~#BJ*-o
zO2fX^=>NUih&GlfY3o`NILTXP3Z~RU=3n+XzmE+31<Z4_Ais%xNR+Sef4zWgX=F{c
zUHlrFC)Z`WrS4XyQuXed`I)UP1E=!!ec^ARW^j3lT7akDQGe!iL2Ib<C?R7lr%KB#
zZMLFX+1b@qAxg{Ftn0FlB*Dw2UZwAewZ8Vo$ZSbPAX`~t#-p_*g$td<m``d;7MPXr
zCB-GJ6mvp6MEdVLBy~Wle`SXB5kgF*97()@d2XqsGnrRT<%q;nsaMM7sKocv$sCoa
zZ<g#x>yz16D#;AQJ?C@da4!knINVD{w-5J{u5IIfyc{|X_mb5O#672V198uJ-EG|W
z6fe67BqhdeIj~q+N4=H&R~9RC-bpkbEmoG;B9K3DbtX1v(nln1e`KyTqO`5gRu2*j
zB0_JZuGJlg`J{MIKmuHTlpX<6E9o9cZR@l7eFmu&x(6g;{(uC(_$9_}@66EzzpS{e
z6n+^-KrrE9T-ZDP+oBQLO`a(sFifn_2&Ehoeo>b!8>smnQ`(TS=H1c|uHuz5-k9~~
zFr{8G=T%Jxe?O>|e=mG$5x?T~sVaAUFpH+&<48&vT)^s%ic~p0by?eG$yFjda!rPR
z$*jXQyryL<KIi|C?Y7y*Y=6x6FY~=F=-H0%>_KbhBG7Y<6_hALna3(kPN5w#2Asfn
zIf=dyCusL{-ihn5mfB##XDQZ&0#vpp3UcXwuhsE8o0#Bpe={7dZ>~=SM1n~Sc%*Yj
z372pdUEET_1t?t$^CgT+aa?z`e1LJ?+d?H0H(c!qt09s%;(J-%NNc&4zq}3Kww4Z8
z6{){?+nQ)Sc|3wh+i^VN?eOM9S&gX$=L@q$;(*F5m3Y21OC{c}&C<DbAX(DD`T-WO
zoh;@0EmeJ4e<c6N=(3BT=rnzoPi`gu)-Y>37(|WNy8@4Jp7owL+OG{Z2m952-D=&>
z@HRHGLW#GtWv_o9@a^o0tv=$3otF`$56tr7FT~D|b-mdq+Qvss)}C4IJCpTo?>cXC
z-a4UrujfLk(+sb@1DO4gpg!0SF>>;mUku1ssKlX(73NDhJvb+tHyo}kpP04Ge<DoH
zlVz99&nP8-3n}IK!lXG)&^N)>aCIx?BL->)932WU2XwHES&=~cbcJKP!DZtnk8b+J
za~_l_nJimDcDc~ZBu?eLsnF8ET*bAesC9EJZ(jvegr7HJlD;_h@yxGHFqH8<0|Q5o
zT$E$q@(^w&df{|(9%4Tnme>5wEo9~PGepS+Hc=XX4w3M<ju`bffrzw=^~~J9v-QZf
zEmG@c^b-C>>gT$j^-U^M3m@^W)Kh(l2H}lemJN8urcirDix5RtEnKAUZx$}*o-8(e
z<9AH~L#>rc!iY#M+l|p-<%UoyRg&Tu$E-w6ak#F&HG0w`!?)0BE!)5qtG+<CM5&gT
z(s(R?`>e}6tL@v-dd!|~)+vm}YA!IVwt_fdTmX3?NV%pddhPWu85i&lql%$g=k0lW
zYg|?iCd`?{dbw{N#oSh}9-8oVG=*4JY!^K^T7v>i(zWY0{edfHxcibCln@e=*O56L
zk$Bvj*OIlkpv?qHJfB%$!lEJ`FPoG@O2klq=qTu6ZZ5;%fS?+-7qndiByoVWmsIMB
z%s1@+XuG+qdY*A$dcuqr17|0d*v!Vmr$~s@&FW~{U7~$B99RFg9wH4xYDWzn=^=S8
zQ{Yc}NZexRrroK;I8S8j(Kb1|$_SDuc21P#_(l@BeVkxU0dR%F^%;y~R;pzo|8!a$
z9(z4AlN%i`B1=aj%lz2tMiTizPwx4JxYS-QmkH7+B!6pTF`%bFGZlXL*FV#qE#_&f
zp9qEqwV(W>d~#dM!mhnqmY*aokGSgQijg@jU_R==&~qhWM_utU6V$%IbA(fa)EGis
zr(Yk^e}w1yCU!~m3MmTD4GhS`^E~?^-}G{^u4$7q`1G^{twpGlUQkvQ+tz?r4)<Q`
zn=QWO#ZxlRZ#C}Ol=ot}DPMa*`n6H9R{;6^lrF3~wZAOU5=DYr@RQV2mySr6YFf>6
z%67lah+b+>vwznk$foJ%>&$SLY-fQ~VpnTC)1g;M*Ck4>*_Zy(C?X_=QtH-jWt(>G
zlNBedJ-E(8xHyUaK&nRzx%od5%Rd1XVkPjGab74B0x~t15qu33muu4~$bWj;|MT_l
zJIRZdBG`Bvf5xZh*WGWskp&AuD2>zS@4NUyBM9_rvQjwW`S)L^FT0N;)jF4?A^DGy
z)9+uWq2&{U7$kuxtu<x_B1l43Pu2#3@l^&<zn9VwX!?aNVqQlOV8-OT7Y`>XYDwZS
zBSiz5BH=DIL9Gd9qqsfZ6MvM(<eB1(somvw@{q^H{+imf-$(T0;^2AzV5NHO2>LcS
zPGmwRnus$i+AwRL7sC^#{x%c$jl}x3W`u-}*q1*^B(oM0NkI!e3P3larR1PcMVjfZ
z3Awa<`fygl7-)NFyO0O-kfiKaNmffQti_R%JTlh?r}ihUan{mtU4OOCc!8wOObPrc
zt}m+gn|h5_Xy>=o>p8{0^!l=1&(Zo3dOb*(VTmwAp4RF%C=tMq)b7!3WrTrHbz5I@
z5p{K!U7ZOi5bDm0?noaJkL=nK$C|Y<wYtYS9!8bEsUPO6?gyWwU3vu855p)Z<H;?N
ztm*PtkXW>SsyHt=QGevDrL!X^WrgXwK{MaLPk9i*<2Wn4XTGj<*V<Sz1$uPXpsX;;
z{uF<0oDw>O4ei=U@n(svwQI#TSyoP8I$vwwGI3S*`_Hj&Y<&jQHneZFMge9ams|Wc
zI1x9*K<N(KW-JK$scxGt$x>nI7=F$WN0~<<Vc=dx7>b#B&VMoZ8eySf8bmAVTG_-E
z@@v>J0%VwI8ex3RAPlK5KIc3^%d&P^HMhPu<vq1}SMHXk{)QD{5LT&+EAkGuoKeN#
zPA_IOXJ$0_&q!+Nq`8cGFZm?(C)TSC=9gADZ7*`Xe*XJIRqeI#YRgb{r1+OfUt`h-
z)P8_TALVjm$bYIpXyry~cbIkUkm8#zWc@$4>)OVk{)d_MTOxRkvNw9Mu%ttY)6KD@
zdNrO01|{vJ+zNP5?F|_si;pa*rzNy)q<b8;PZGplMHJGueyQt7J?a{sILkQl&OX%q
zVEtw+wA?HMxnM8kOC96uwy60`eRsm56xHM&Dj|&Kj(>O`O%RTf!v9@)PRk(tmeKy4
zU=kkhZh1(A>Hy<}rR{~mi_og9+m$_PRU&c|rJnbt3gs#8y;zj3hV6C(9P>blFfudd
zxo=@uAn{HFSA(?3R$Db}w`!nyG?|rQKK7JrR;;4q32s4w^`{6dIb6jK=?bByaaoYb
zgkdQpoPUZNSkU*wf@lSzKLrbI_%*Pg4&*1)NI`|`V~3~`u+Rv-2p{SICk48NO!18f
zsX58{Cgoj?j<^KeAQV7rh3rziEqos$g%_=O<>%4P2E`>%S_J8biLTOO#dyE(1DbU;
zCgRd+944<0F9*madvLWyv#aF*dj9~A^Q9J-rhkhI%)yG{z@ewI6^<y8Hj5irQLKv#
z4EcFoptlN3+6GZC2PjX=rT|iEp?cOj3rKUUQ)`^_aTVHi`ciur5%KS>F@~bRkk}hz
zjDVz6;itqH2qbmX8wMF8#R}}z_O-g)ImI)^707sjRp;lw)evNVrt?WkPYR!}W=UC{
zY=8Bz$idRK-d@nPU`gI*<)sqRHIr=F%#=a_d!n@|Tuk?x2iuqW$$2L>3zlTJkq-Gh
zJ{?cEoF?^Em^684<|nohPm6x7T9b3i6phT%z#qSYcIsyG0_3sDY{T-A^0Bz!Hr&1=
z#e23d3W~W$iU6m#FK_m7{?HCVwY^#Am4Bt6udAoes{)%#MLVHdX?Y8&3=}tz%ADd`
zQknN|1-jcM5(D{G?A~@8w)FC8+=i<{TH~i6v{KTZ3TY{+B2P1~n3=qUbWaglEX>@^
z6-68R%8E;5?yZO%oJlEEai^v^yR_D>@}hLCGAY*I$PHp?%{gR!;`J2Gh}Eb3rhn9x
z^3%O?j)t3}+M#a7Yp^Y}<`&Jdn(JHA9LIjt+IkjqR0=12j$R9&uIF7dDs85FwSi|Z
z@+oz+r+HRbq--(8l3hyDz8Y`~%gQ~D4U1G;>{65X-0k~PHMA@V=n|Jh{>)A*ofGR+
zhvR5`QO%pzBZdlI_l3h-h<9HvTz}xI`_7@HSZ?+mh3`sMtv1?a5m~3MPP_W!07G_f
zkPnWDpKB3S!pqI%rrH}L>nd@m?V7%Xf6~6|K(V0m<7(I&P|mXWXi<~ZP9H)1HW9TC
zQzA)7kWVqwpHpfUpbX_N<hlTyjpmh-x`fT{-G9D*`P=7TK+v#1Vpqy~2!AgikMuo1
z`sium*COBeIUb4zS^;JGX!R6;Jc^?CWn69eJngh1(~JavYyB2m$(nnz=iT6xYB}my
z=yxq;1(9HFEz^DL8_S~F42h$<iue?%@YwiB<IAY{7_a`hbfr=Mns#5N-ABgMSmlyJ
z$xJv{`gaOb5E>jpk85diAb%7XVwOP(juhrhYtL_ORKogL@6gH%N8S#nI$Pzrv~+xz
zCZbg+l&11Kwx7Xa;FVfhs3IZ{&J%(3Q7tgK8R2&S`8$vMN~1n>znRVhC=}XQy3tBF
zbdgrdq$w&mS%1piOYgFwfw!rv)vhvA*@XS}8i>%8AE$0fnYV|=+<(W_fRN}EtiWS5
z4GjlJvdBe>OCM$z4iwRD!4W>~lebQO$JzPGPL5X2l=Y0E$ki?A$?P}P@p7!=lrVZR
zI!7pFCAAFa`tRlHiIS~vLg&3wOXp25@ereC7t|>p?uY{}@r{b2YQK;|UyL{1!7e~e
zgIzoCn<ErY2>YbIdw+YJp_EoUKc`GyE5FT=&?U4q<r)|kGCL9Pl+&VlTzUx?Bek$Z
zbD#}Sc){ZyuoigyJfteBEh&iv-GiC{Ill49<pP-5QK*|AjMiY}*nI(P#=}?s8lw+o
z79VhyuRao$r@X4G3zf_|CFM%B|EfSSjFPeBMxK4$r;is8dq6JLDM_I<86R(F-zgqh
zr)RO?%5ywRtk{9zy`edOH^g$_<IB0JIs?ynU(n3%DOXU<diDhWPJc(L{{`8wO3#;Z
zUMLd-GB!7t@nAg@0W_D9J~$_T9LsK8yKO&Tak^`PS`qaYU>LP0c7W~@cvo~$1Wmd~
zir(8rf4`KDs3Y+_JdfkKn;T#}GddDUQGAM=$l?1Shfjy6*TaVwZ3Zub1+RwJZv)^1
ziGM-jptFMq5KhDJ`g8c(y_Cw~>j_^*O0Dtb@pHg;`*5cr;0(YgVDSZi4|m=`S)&!c
z0AFrggY$K~AK|EDd6vc<!U;~Mb%5U?;_K<*@2`J7$U-!y(0L<pZKW_iXp8R_G1^R(
z$8RP)<adj-8^nnON{jTHm-Gz3ywFUGSj|&jQFdhF2rbWjDUcuWGtzC5Zu3Yp)7tdJ
zs=C&d`8dzIGQ*BE(E2=o=IRQQ=~i%0G>r7)XnGRLx>#To`Uwtv9C;<8OYBN-pV+tO
z*TXL}CR!VXbYb}V^N=1G4MGQnkBx9L48Q(!_;&cr!IlG&LgJe7kCnr(-v>7GrAS(o
zC~49;Ydc8;iQ;DlXSI;v=p4<*vfGz&$7#0;I!XBO5?~Rcktnm5vfd~a0hgD^-Y6X!
zw^{loKk*D@3}^5Yx#kf(=m8R<jI|s2asrqC-Y6n}>~mELgmIDM1H<@p4n!-wnw<ja
z=S2{sh(639x<w8uZG_TlCIxbTIg?R9fLqI1k2T5JX|{G*-hvlK&-St*Zx>5gj{)rU
z^1x{RxFQchnBe*BOZCM#5zz_o-I!(E7KEA|nUUwCMK9&;oQW?p6MPm_uY^_V)XjJ&
zyq_@yQXb20m!#h)B!B68Wwy9GEzOOUnPJ^YnHkh~P-cc<2W93oZ!0sKeWfz<-E1l|
z7Zln06I-Ni5<wc-5J6II4mlt8NpT-3GJ}xft(1SL<N2%!R^V&K8dzpch+?ioib8ZZ
zzBDBmW|U(2bT(O9VbreY2bL(qmtLe!qq5mJTa}`LYmAq(LVp{(gYjBQE#AsbtyaQ?
z8r|sQ187Cb-;_K#;&z*4<y6j5vX)c7a!O6=SI)>DYl`Q}Y5F=R+Y-=E#ad;%GCj5J
z>LiL8W|4s;mMINQ-&K<iW(;;rJu|HZ>!s<PaUd+77EAA(L|3M6lHJ*KkKo*`9ZQdO
zFUQhDJtZr-z<<RQhtT5!x$O}8*%vtjCA@(d-W6v+u7l0K0I#(uOf%jxmpbzm*W+~<
ze7QNFSF$SdFie_v@O}YhTF>}?j`xdN_&DR>72Z!vhgM@IRV&_4CryVLZ_>?P)i*Ma
zkBu3m+pJ=IFL8(IRtDmUXNdk+Xg}BDEB&BETPi0{^nbkK0h0z-4@fqm^Ou%ptA4g!
zob6U-x8p$>8o$~B!Ko+u0Qc>ojwBI-cg$Ff4=DCG_A5Qoo!m+f^bD_Zh1DruhB5IB
zqf^{P58#@JyZC_~;99_4+yFa{FzQE^>n13xvVEKaK;`;5bPs!lB`_`p=df-G=J%DP
z_3Ik`et+708n;GmUS-$`p*+T=JSx@R3YUY@Uk04_X~>gty^ff3{cp&_0+cl|=_r7A
z3980_=_n2#{`39YU%vhWm3H|4-!v5?Tg*)%hT-S<6&-^zxNi7Sz6W7Lv`B)8QX_8k
z1tw~6bzvR&Ivfn9S)grbMt`Kyh_RNzC}3Kg<9}(76Dxvdh-gR<>nP#-(GC(-6N<d2
zN=EvYpp>=gE3T(i>1%qm-)SnY`qyyyHXJ?|u3D#jniAv)IJ)Eu7q|+-fJbp|I9rF%
zcxco;;!5-=Wox6&e(PXL-kj06L)2f?j%0uf=(=YbY4%M+2#X-l5WVy96+W=N;z*TI
zG=D=8!?8q$N+Jq{yCa9E&tHF*b>DH-r{;Zxa|cQbW8H9PFjr047Q)K;k`0TQZJAE9
z$)-`QG1nrqx)G)&1!F)E<>#RxY24f0w7JhGGX!cP`36I{t)3WxvMwN+%7~Q|pZ0(l
zLDma~R6x%DNfy)eQcI?oV>>Fs)|@_>#ecD2OCzd;2-!*LI&UG1AB(cAD4bYdyy1*y
zWsw2ObDEp`>9yS|Owgl?d?npK7{86zN1;^Us};PBH{yyJ!8w3w{fEiXMqzONR#}6-
zp}A(MAyAyWNIoX|?AoffrKs`PGI~5bM|9{C`{aXvxXX?xmTEqEX@!;|8Ov<ZG=F*O
z+2|D6%@|?OY&jg_;#MP>;w?8X^H|8V%yKJ(2<YFok9v<ikqYS;1*~uBu>=R92_{Ow
zaKZK<IuQXF($o7ON-u_Tyca{6`?fQpn1E(|*N-vIc9ZPX%%oD0X4xs+r8R`}*v`pP
zXiA<_siC+$<8e)e+tMsc{UeS1FMm^FdKn<RYns^8YHD&ZJNsg6=^SWw?7B<n_`~dy
z)Hm6O=N&QDww*7mYvwF#5Xz-3YBpalr#xt)Z7DNa=*J;v;Y(I%mbgDlFBUv$<Fl4L
z>EsnhPx|knpH>3iyD8$O5C8)`rKDTB+}6gSdg<<Z-<B)>^clCA$_L~W>3>q%g!_gU
zZQqb-_~eV#HDO)Kwj4(#Qdx9)u-1L008Th91?nt`8Eu5pnn~UWXLC1ej@t=~4wUak
zpKiHM%GXQYRva^BddD{Jo6d5Ujjo-A0V5!IPCZ$$i7Pc2vZY&0_NAz#;p2%l8otFh
zgMo)c>3tu*b@q(p4RpGNn14#1^SiGk*W)L$>=$!3^&zDInqf`7jYcKus}3^szNz7#
zQullI7*a8-&isu#2Qh0CTIyNnphe3jZ&O=RQ3*NKAb02-WDs}k9AuE+=p1CTZ`3)c
ze7be<H@;J4Sg-6HWYFHdbCBWNp>vQ!yuNdgL$sTngIo&s?;KPn2!928s%;0*o`3rc
zm<592l&vYL^}=C%)=$1bh9D*X(lbDyORN{RO+>fof@UTNT7Zvh*dJFZKvlw+n3J94
zWKS7UPM{UMwJqwg?o{AmP~Ww0fbRfS#h6MsYz2!-uc+vh3Oj?pR?;aIb~gJKRpha$
zij%o)B(+Az9Og%QU4Q9FVLq9P8|M^}tCcvbnA~aiZ&@nNg{A)FO09eUQZYnK=E2_B
zmN2Y4*%Aiz1Y2^v*p@I{3+>k2mar8`9Cx)PoT6=8!e-yjmT)q+Z3#cp>ud=Gm2a+o
zw<YgmOI%|7t$$s`u<m6`dZ_OjmoQv!^fR*A)rMWs@lrY#`9Arb(ehH7P8fu1tQy%V
z#v9oD0Sd=|1FJ@ciqW}d)*J>ivlGA8tYL;`n@*$O&DxbM^Z=+3YBA7G*TeQ>m-gl;
zB7b974B{m_STAZ^e$nZS&}!B~j{%R;{wCL)(TXT<{X5GZ=}r!(2l}r5hS3k#{Zg=m
zw>{7|dgIw<>!x-HgBcy$QE)%`c0n*BaytfQ%sS4D95k9U$}}5U@EO5~XjI!Cs%?3x
z@f~dfor;=TPDrZzl5s6A>B(Z{OKB=rlYdg`I@3;tr(D!&O6?=o1WU)hS)p3L=kjdz
zB7EEfm7UUIjktle6ijrt&ME&F8@{8r*wNxB8#`T|n}tGE7?l_5o;0({1sav|-9pO(
z$_qLvuSE>?5@hTOX#D8JG^$+c=E$H!_SjVCpslyk55%T<#M%+hwAK3QG_yyvk$+Ww
z?x5-ID)4j!9p!coWBItI18<Ki3z3~Vt(qq@!RV>zkItCI<8oR^Wwsi9+NhwBkLK5Q
zWz$VrbkdwmzG~W0=+}I%mC3b6nMuEzQZX!-i~9pf0U%>VyQeTw1Uzv_zn?<y<8A2u
zIGQR4t$hFS&wm=r&&DjTK&xndw;SRQv|<xSb0{a`XjhClw&P0vjxp2!0g&f|fR}Mz
zC=&uQH<!T@0uuo?mx0nZCx6vhOOM<(48G5=@G&{8B2uCr=wgA)W;Z}@E%w&-R5Sqs
z<j^1m`t$qAmLHNm_Plo59=tOnOQiUbBB^h6^SRN@$KC#A_wL!)#;YLYxq1225MF5f
zC5=HRjVDshQ1kM&`Q=_~c6j!9tk%Zh@%SF_DGzss0%j22kl>+*JAZE}uQ7~=@VIUb
z=9_rm$>GQRE{!|L2~MYVgx?hLJU;yP^5ahDO=}fh`kFrKUr#T)KX<hWPFN8)zkl7O
zacfAKzzySNl;+zX&8OW@wa~1Jl*W?b&$rJ_HS%-DFc2`*&WRKlk_NE0aV{#8UUe8I
z3<S1W*lzLNK6SrECVykVls*)8AG4X6W(mr`q+at=^2_*Gy@qTa-HH&$tlX#q8<*CA
zJ|HaIE7Xm)W`}I1A5A>fr4kG*w&(a%Pe79v#%iSZY1FGe)mpKMz1j8yj7x?+cLpZ|
znJRRhhXyj8-&)1R$j=`SN0P-{BehmGU9<?75D$^mL+{H->VKsMudJEbqH(?<pp|NS
zhrurW{BYNuOxb6**@RjDr?sdUnM5RWRMCf8R~w0S6rLc}vO<!XNww!o6jD3~8r$Rx
zw&|?WaC5y)v*;FfLE)LXrG=l<8Pi=t<;T<A5Tcq#FVXTwqESB1W~W>|L>%S#c;7`_
zd%#oF#cn)Bq<=m=T8xc~G|%=ai~BTUkYC*lu*dXd-C~;YPpi8aS;(CNs5nehr|feI
zwQcImhk<nOmoz?lX_zhjY@W>?#sk}_cLyl<ou|cgzj-*&8Qvs2R%L_EaZw8{-H(WE
z)C37lRF+x0rJ5iXpJqIceM^wP?;q~)Rp(mRZf=sr+JC~3^G|EwU@S_CZA;2@c32uc
zoC(vr*21m2CM+t(F$LtDg6t4bY7nt_g6X4aHaceWxdsyDjj1&p!otjO=z6LN0#lD2
zv<vY}W@gt>sHLuq{QVe|V|kYxe(joE`K;th7$#^hW5GdG;8^b}SFA&}q|Ag}&up<h
zYO6YNI)9X*xz7EnP5^+^$@85i{8sSEUX`)kdH0O)LOFD(3xx)46p?EzFtOUV8s2>Q
z@;0yG<uyS$p>oaNZ&%fjRmREzHK|6UQkx*<ny5m+5~Ct|>2?XPYe6$MGv-$uqjhtP
zN29~#s_#F%J?6k#amxCGv9PR=;7!6=oGgCxeSg(3LbGXOC|1BO>GAW6i_oYb_Z5ru
zx>Xi#K+-L(O1CP@=02AKKQ!5{w|E$lq6yCp_t4Mz>6`9+x4&ZU`<}a}%?$6mC_cAY
zH~3`qR8_N1aJ3EgCfz!^T;z)rn)R(lyc9U^mQU_@$T>v_(Nb{La}JZe#npkT$!4i%
z-+y;~5IyA`ZNs^VCE+@4_{KM%N^83LtXG}`zy%d~ChdqDkH4sww3-xoQ&qQ9f_%#>
z#-m#mxv^I&trvQwl6K5JG_}}g+(nVWU99z8a%)5t@jUZF#o(NCU#=xC-+&m2fDzAi
z6zGeh)})K*$R4#VydM3~b~X8TW*M?yOn;novVQF3<te{g=@*UUwV0P~CNm6Au>-YT
z!_VRs8jG>dE3R=DOGlvw?lbAKApHvGoGoIPZhg*6tykimDyiG&-DIW|)ogOf?ol3K
z)PtSH$O^id^{K&a%{Zzr=oy|$!ueOav2<T@fXTpmE;tN7PoVu#ILuzWB2#fPLVqnw
zF$Ia-l~@b*+}<h=(H4&*Z6SHLeu6;XR@>>d?>|UvuxRw-^oqnrrQtAIr}+P&|6kIH
zyazM4*N<vb|1$3ZA-GA$dRH5~hhd|fb@eL!ysnD_vbdk=3hTOzQDz=4DMfUc9Lu(d
zD2Lvt*1o~EI99>V_k2CJg`+3|lz$o-LE@2Q3kqhGa?J+WqMh;~9#C_LY+(gYH=NS9
zFW8pZ6y31|#}G0UXZB9`m}TC=7?e(5t>O&|N03E*xn!MEF=#V_i}l_#tt+yExG5L?
zbre%X0c}m>GpVBK>cmam;qE>tKEN4S!WoF=(rwZx{o~q}+udS~(CQNV+JD0TBW;9^
z3f<I<HPTi&9d5!Jk>LL0`HHkrCGm>9Q6+hUII3pfh&dvIn-V^6dv~d{UdT~Y(q5f8
zN=e5=c}3=kv{BakTYII{dZ|}RX-9Y^T;!F=N+?u~)HR*rfzpjVGF5q@JEo#uoyD6f
zy8?w*h`XG>E5xmncZJWV%njXQiK{+QLjM8phzc&3ab74B0Wz2I&L<R?4`4zlf8|@v
zZfq$I-sdUm1ul&-=0_?;x~ZGAG`kty-Dp;$>0b3>`u!Jez;J*={iV{Gx$4D9r8r<?
zW1H`TL?3=V=)=!XFW;YjdP95gDiC;c`2NcQ_=V=5pt;e}!2>9VaQOb);h#sXjpOHq
zAFDMYKc0UN{7KJW5CZoAegni0e>{Kj7Gw+>et;iW#&Ca($0Em^<XviaEGIaf#sPna
zul(TizrO$NNz0_QiZ1<{KIU()-=F^bG@IabP$t;Jzkhp5FD!z>VAwL{*|6XLbNJ=y
z@3YjVi}0}h`2E+z)bbB;VgPocc8+LpB53xkJvbK?rdP#@**#io%Z{^ef8~?pJMjzh
z&tHlQFiJ0%TKQo&5_hFnd1uRadn&(M+9V>$SnkKeHGG@gc-^$k4pN}w!BJ4i4iY<v
z0|8&ll3V?_bnBHxzmrZ2>6(X@AEb-)p@pUY^cMBSH8G?JV%289vBtU{;!OjSR_9qf
z-~p1NtaWI~TF*p+<-SO>e^!ICwnkRIt~1Gco?1loGig#o^4QK6{Zis7Nwt29GyT@$
z>b2;B*VJN;#rE#o%jWauRNbr|S}yq7IIB2)cfQV64HljAoAFM0?`H`qwzJQ?<UqC3
zmlD%*M*N!FC3mVvh@NY8YTZI{o<5~V%)cj)CA7dQYFSc!y?H<RH@bW?yDVD1q4I`T
zq`V2n);_2^54?z#SiYMUJ=Nk;4RU0OmjLedX7<|Jp=)oZI$piGuHh{RmjUu98Gn{p
zCoj~QxBAGwMVq#%)#QuKt%q2i^U>AToKvy&@z0QVJvvo|^%dA`@wNSxh)aiSMXxLw
z%kVaN+1B4K8dQ3i64L%Sgt$pT(hT!B7*@wNqlzRpifA?}4}~_NTw}^GaK|pjp!RK(
zfG)b)M5<947`U<@no~a^qGdlT>3^9F58+rNwvCj!S1Rcs52_{hczN%Y9!_Mp<*4~#
zU!ReVy<fJ4YSu!riN0-1k#pbVu-JtGMg?Hn-dD5qO?IW6sq2vjl+nHiXsg|X8j7eu
zZR!0IJ~!=1k`*QVL+M%H+OO@6)Ft*+DW`@4gD<TtKLKk=-L-Ce=`M8)@qezTVYYmX
zl4*BYC!z5{1-<hjb(vpL)4iY*BW8FxfzE_=h|~QF>bwqbGU~jHaBK{WIl>(X=NfvM
zB<?YW_{yev)*pHLLrcG9I;X&yS+db<yyGn+E|`1c3hT%=&X}e{F%Ka|o3M!aWWYEo
zO?Z(m;{G#!3o4pf*mam#04&Jei)uKXtc#PnDJSm}NGxbxfgPc_M_^RZhgqOode5L$
z2F<+w`XP*H9h;ZK^C&idnzNH1?;w-bC~skwu!EP6GeDNY)`WoqE@>VtXx@SVx`_Z*
zt3Q$iPF_g$#6*FhCPQ2hJO{Cmx>Q1RXI3U+*qheq8TMx7VkfPA8NEUdFB2(rsG@CW
zY^I`TaY+vN6qaAp_sShMZqix=r)O7L8FN}tsgRL!?6^u{Begt#+YJ%hT8<^vbHWLS
zvDyjZ%uarj{aqF+BkUr#XCtMBw(Y?Nz?Z>}JhR3*QhE-2sY#(FZJdyNwzZMO5Jy~J
zC0*HkHd|86EIb#=Eb`ftm4brX_~I?vbBbLgFr|)hrc#%tWy#G;kK>XzN+5f28Ic`!
zV%Nh+^2VNY95b$e-?j6M+ARFBzHJ(43&A{NIY_b*I_fJUC+aI!b=0@?J2w|PHLYE+
z$r!2<R<ol0te~XHW>XCok#Ms=#<{p{F3Y9oT~Fh-qu+8Sg8JB1L$?^!r$X*s*9Z}M
zx-7P2+L31KP)<`^Jk|kO89N$|+sS0*<~f(zbaIX&bY{4JvpFi~A=u6$%D5#|?4{;>
zF5kXN-&3uT9mj8Gj`ONPArf$^I8mf^r-}V6w;+-6eMQ;kf32cm3%Qnj>(DSkX&>(j
zjq>)^(11RD=hKh2i^!W^S}i_(vB}=-POnEzZdkDrP5r8!IV(bjC(NPS9vRCyR*Q>b
z*9~=8de*~#%K4;cw(-h2(tFQXjMobNe8gh-Wu#9zVKEGcou*HQ^pn;Pg7lNrPlfce
zAMc0slY|wdf6qaN3yPmvdc&`FeO?fFC2=<=egEm02~q9CtU+yBsz+@UYWx88saymw
zJN*^Zua8U6pd!Bu(*pD#VpZuf&{_9aVpT2D23BQ%w@a^%RkdaLKvu;xN>O)2GSqlK
z?9s9dt@r726}CRl2M1g%;!fAg<qI^<O^;?CW$UgLhgsqX&g&BOcT|GFF3r?n$Q*Mw
z)7`Pu*&7E+jW|n?<UFg_85J(*b=UQBP4f5a4UVhpdA;_kUeu&qO5IU{;FD_?z6L;n
ziDxf=eC_|$g)iN7;R{^T&31|1E_^9e>`vjDK@@Ji@TD+c_|jk$g|ipFnSBA(PPruQ
zt_;i5E@Ub87qUgwd?8B}EG7V(p>i?(?8&#R2JMbPDD)(R!cEt&GE*035>lj{ddKwe
zy`vZ6n~8Kt>)x_)PD@}8-Ye|Rxfioe{@b8`_b{_|ahrV|I$y)!p(QA91W_VC?}2d`
zu^>Q&iax{f9uCx6uJYNDVazOfU}sbv(DJ9~Sn}EScFm0Q@x>YL1jL1^yP<sGp8bcq
zYmdr<o2?&gqTBA;L#pL4OZterc7Vvl2>*vf5YX{VKKqXeC#@eu1ShGVN(5&=-cJM&
zCkbPqK1>8B>DxudMItzRmiQ<a8KOQ;1i_Gs@Wb*zdOr`uB@e_U5B$j6HIA1s_$V$y
zOu2*07$RjIOaiV_H%=yw;xk%rU~WDt+-p7c$I}hNLb&dSM9B=nxfxyOuxPAeNN|@M
zkPAGVS9;wI%(1xW@4T0w_$U=mFrEf<W`w@fd*1o>Blj`_5K`hI4w4pVcM5!%akRgV
zqxZML7gvj{LNt>TbGiA}u{61Qs_x84;iVOvb1P}JW8t&x=`WGwli}VUgVLjy(XucT
z1T`@?IG1s=BqW!d6etP;G%=TPr#L8oO>Y~?5xx6Y%rVKK+4bEGEClSXqfLMWSlB*n
zBKY8ViNuH@8Zr$y|30spq$Dm$OP0poYyd;9nd#}SuC99Zs)j~EQ=N>0Cy~jxPGNZ|
z6H!3$Hj#A*%_i>osP#!J<roOs@(w0wc?W-@`V$B|H5qTo)MGpfOjtD~%t*?ASW6+a
zaSIZ&=khd>cs|5fQqSo-?wREzohK-ySfE3{HWcg_*EQFIc3iO_B%hcmmF+RH=or@~
z)}dwZ6FX_HArnV#f$=c%A=f%_V|ijsoEyf(Ig?bkd>WAYc(qN!oUhgn%UR8FI$<p=
zCNQSu`b-4LaEoDC@t!8eMkRWG&n0lLNj@dq>z)c=Q(DJ3YLhbOV$74u(31tyP!tbT
zd5p_*80z5!)j8*b@iip3flG}%q10A8rEM9zs$*Trv-g}ua?+k>Aec%yF9nx6Opt@E
zj(bqUbi9G7tQpo&yb)<-P9>Y=Q^nxc#xe8cbyx$h!SJ~c40g_^Si)t0^H3oTQ{j)}
zm>_wb+&PypOyWaP%<hq%$c8uiI0LhEyj%?0M!J|OJbB|&=8)2p*db{;Av=Z{rByIz
zm{cs`I)(dq+NKywK&NP91=u%Q<O`uwtGRc^Q8!9^FkHNm4GEslJq;B;HRYV7L2#S}
z_Ry@R<_NE5!(cee7HM&RMi8{7<4wHKhBF|<S|9$Ffke3-?52;1gl#&HBWWqAJo)CE
z#nTHMXGcUoTm1C5zrz&RI!%`!-oAZ#c6MX{)hG>n^V$oxe2<ZckW%k8C+-!5(yfux
z;``Ox)z$NNuU;=F$rq-@kFVD2<<;fHo6A3bT>tQV{c63OXmjI#e090T8s~yIudq?B
za&Czsyjk)bCZu$&anUQl0>O!kv$MsISF6|0m+Ongk5A7RFP0zITa91*=iL$jl|X90
zEB|u|g)Z02%k}$2As<sbTfSd?xO%;OzX7Ja_1ERkZ(e=3`gl=s=?J0d{1We9;Z2OH
zn^E6hUamG1_0956E(S>Ee>Mh4ZUx#2yc34*FRv9F6D^*9_{Tc0{_^JX-_zo|)z#0-
ztD!)Bx%kuK&x`LbXmjyw`Fec;{45Ihs9suvo<@;$Nb(ZeUoO6#F#Gex533ie#nb6`
zWDzb^T6IK=Nx0SDpPk*Z&O?+?vof${ql6KamP(ya;yvDZv%364e^Kg>UtClRA!<~5
zNh1UQO5jXYd#NG^t6OhK1yL0!Bb9`kJzfdoTokSvtpLs7i%<%etrcx?po8}29eGip
z0>Z__bOu9^<%X|yz@#_^IjMwoBq5+o>r60W6S`3%LmQd@CbV&~iK2%9us3Q00C@r(
zbmH!j$ZG|*_!3ZYe+v+P0Ts7E<TIgyEB_KIk~n-mRPg3{0~LJ-41EU-U3P-*1ltL|
z6JjT{ozQ216Z0aPOs^M9KW+$ND2qrfpd(+$k-1v&8y%4EOSjWHAn4)p>Tpu<6@o9@
zED=3`6;!Y|axWQ|HVC1?K=PFg0cfi@jhzzQr3lGMCBBd)f4ml>q|8*3?}OMOc#&%w
z6*)xrCW743Nx+jUSR4MEA;Ux<l_cd4u#n`W#Szb(xD<yj5#W-PX8<39OM@=W;t{}f
z5MlsXsRO*A1b82!v~)VeSJFHTsEA}E(Yi#PBi*%LG_~*ms%M%5go3IPEF{v7pgh&&
z=O_a#nLYUof2L5l1Ont*2@r{r1=T4f2m=uz0#h0MgVRB8=bGfRpX9E$2V5r)r4s%{
zlE*1(`KaVUa*#ZT9+EuBu9HVrrUU)Cww*9T8UQy!Spd`n@Qf<2YSFBk0X(%IadnA6
zT?Ibj0qP5O89EyTgC`fmu29gA|B)UGog$Sb3eOYje+|VJ0KKHS0c3?oZv>@HjgyOm
z_|m>(6W@bG?T^}4QI@9ezZq)8bm-mv9^TDt``*od<K46^p0ZD}6?Xnk?eT8Tol8YE
z@0zR!c~!g^w(npZN{4CX^}k-dgSpEr*+kmL#juv^aB=~;BiHn_gQZ2fb`6$_oPr`w
z*k|Aue~fGj3<6*GEzQWH6siDQ4--4m4!5xUCra*pj?xikFjT}_6-;;(1SeFusT`~A
zg&Eg7^2e<K*S!w}c{a@M9*NK?*?9ZX9gWk0skqQhWZ)uOq#O^4QqhAb6@EyR%1y6>
zc^8ws%R#*Deq^6<E6@z#%7%hKrffHKPJcyVe>-}gxPKU}rl6j#2%8$cBd(*(;jf@y
zS)6QjNQVuSX(x2J<*+lvt0Em)G9TySqwUEh+m9SbPbQyQJ!U{G?I`P$|Dq0%4ezyV
za&%-px32_QC{QaxoC28{Kp`+m_Bf(LhxRE8v^~HOxldl+8?|k2beutzdyHhQK-aj(
ze@L8Rc4)qCu0I7Sk75CO&`Zt!{sZtu+YP?xwn36#6t=<)om0sO1}rHf_3)&qZY6NO
zf_mymNr`H1RQ8U@H?LNa3TLS1m<nQK;_RRrO$eFMI%Urq^DqIzV;YXIIVydG0?pLw
z*c7%fYspVRqsSI0VIbOiq?$A9k)J^-e@e^I?}+{2QoAD?>yf6Rw-P8Re=WW{UyikQ
z5Ci59i2<)SF<?8_H`xig6Ksam3K)fj4B)1twZjGA>a`Z3&vbxC_0W5MI29_j@@Vel
z_csBK!rc$G@E(pjpmoariW<$Jgj62+Wv4{8TL{KKsYJHT6A(f?zhzU*(#r20e{3SF
zfm#A5t|)mc+#|D|sGh7pp)~D9qQ!mE+}@~d@*<Q%#J`#H)8ioM+#VEkw}`FpV$F4i
zo{F~O;S-n$5={0~0u^7jwHu(`k3c;Kg-W$(pb_FI3uo)8pwxtejsA{Uw%WbW5$_Hp
z8tx=WIBv*%0O>vVa=!!UZw=Uaf0j6kZR-CAKp%GCb0H7$xriH|iyhesp5b#2X^H$b
zNA|Pw_cSev=QXO9euQ2)QSaWOHm=#P?X$StRteA+96XP{d2rH;DAbNHtuP?FwFFZz
zvg{}Np4Wxm2h{dPZCua5F_P%qJu*8(`5pb&>MbwMvn=S_rrNn@sBRk*f9HXMFMq|e
zxrnU49t4@~@MC7H4{_h7x7>Hv0*D#f&{^zh9X~!RC?HYi0y1m5b30n?0V!3ZftP?5
zSak0;N5z@MbjmNIt!Ay=Q7+&N6UO?HUswRE135cTA~XYS@V&3<-0u3)(GI_KSWjZE
z_ZZ3C5$OEC_D$r!Xw(P~mvLSw69O_hm+|Ed6PMae4JemM11XyU__yT)DGUMxDcr7^
zxAX-miUWVcg_|u9W75(?{B#4v*aYYy>g9NwQk`CcS0rnxm2gMdq@?POveCdsb6FEj
z>SpvC<hE2qNL0A@6*uSJ$4eK<G#5i*OibjNf=D(c7^O^usJ76n8E`app_nTTVA<SZ
z;m3SM2eT8w<<jGDPv>?TKr-iGAVQxeLZ2YwcE5iK9y3%UAM7_ftamEmI;fZU&#Rl0
zJqK`Gagn@E2Q{7Rua#*%zm>1mZ0jzU-N2EaReRa(svT7mfE#udShRPgZ`xJOMLQ=-
zMRi_^7t~Cce-0Pd6q6V$G5P#`!g-&kFd5fN@AUN76ozp(;Xj0Cjl!e~A8O+aiDdWT
zxA%W<|N8nL)WP)rCr!l+7UKbUH~olTEoRAv<FfDZyAvcTj{A;zWJyhqde#Cb<b`G6
z>$FqCB6iIA>*#kH#mdiN?bj?0d9$U789^N?G!Voxn(=*a2QyU_Mc}EBLEi{UT1#JX
zeO^Uh>D7MEQ*qV*rroz`_a#YH8xwpL7Bqk8hRAU_rIw@`dXW-Z$>t%d>7@`ZOT}As
zR4bz`zhzJ`RHdRL1BcZNDDs;LTj4p6G<@eFgq0!Wq5Rp8ugIa}6-SDVVzS`Tb)#f-
z=qy5#Zj^BS`|FRi?j2WsZayO%+;1{m9eDyw&AinFE0jf;f-KeLX?mVr>hk3LWx9W=
z6tOIAt_}dA`!Y4;M82LXV_z;726Z*Gx<DAcpRPi%nxUl^8LadsG7E^%s_I_bgMob0
z%F(qu-8Bw!^PtV8C@3hc^31Y6-3g8`5tcd8@(2lmX_^w`1V7R$o7x|7{aX7Y&Px|i
zw>BDTyPo>mc9717mPn%VO!k>NNWFhD>B5ndgLX;_WhNIhKPI?Wa|Ou^zsl%n*jRS1
z*`M-;X{;-H4tc#h-W?0gt`8>b_2~Cqy~3pD7uAcw+1W{oiv6@WT2Glev09VO&DsrT
zacOqWXb&$^J)gfNYtffdyK~0Q<%#Nr>1S@<!t8e9?DG7EWNuZn%3|ab9W#G)ch?6(
z#J-ik&PaILHnY)iQ`Og>7SQ;LDPUUeWOF>ZtX3(8Lp8BQy=lGAc7M0sMl~aj%(=L-
zR9QWg#WYDc+lwln$%<mR`4qX1l9S6n>Dbpxxxw_P!z+1zGrx#6@3$nnCMX0P$xx$k
zP4nZE1ff@C{uK2DR2|Fg*~L}f;sI*E+%><5P)+}O(OCmuw6qjoTT7dJ^yUD&hy&c#
zpIS7H`_6{bK|muIH#^*)wf+~bfiL(8n3vrPxlYY0mQtOXw!~ED*YIAdmGt?4pRh<k
zm!TIa6_+7`2oslc4Jp5W1!%9oy#M$lGtnc3#%jTP3xxrSW(u`?eOAWs9nJ6w!X3Uc
zKH{t9tG?!Eyn`+g$Cs;vMQIok+N&9^^Xs#bD)7PZl?=ZE<yfGt$kvnadIn8XrZLd-
zmi~^vA#g%3(-TsKH!?j>M6W7PkhqU~WPTF%QsP*f4B2dQjetUbJ6zy;i#tAQp}d`O
zr}HV;ZJcNYkeT7!mMCfz3n(v<#Ucn-2*}J+7GDsIN*^Y!D2*<A<1drfK(mLk8Wr^_
z*SAdu85G|%>#78@{I$chc1Do(Bc#Mn;W{Z*RTs>Bm`U~H)yGu<+swR-64I#(y{L}N
zsuZOOcvg)RvV_@x1w_LY-UQnXmmh{--`=18nxP~zLR)wG_0Om9KqClb3Oov<{ps)D
zPQO3>G(yeJK?=!R(ih3o-+!EjJKwX{^O8j>jUnnR6T;!>lR+=EnH1@3wy^9m!m%sp
zyiz{yIYcelf6@psAFUKR&{e)7x!s-bjla`gfS%FDUT4vNJCYTKaEABW?RX6cFYO2k
zbcgpQh?ehwnL%9f^Q_*+!ff>&oH!H^3}zGWO%OdtN9T5A)BF9|ndp3if{$zZig>#_
zE#9K1%CS<ri4!Tkj~taqwBY@Tqa%GwaW^`&5NCagw`|=Kz7F!Zg)R~xS;Aliv8_(>
zcm^f)lCbZ804~|@`R4R0>`^86eZih1AQzNR-QWn=+%1C#JV8Jdv37U9?>mbDtHNUu
zi0rc3IV&{Ic#-Gi#Z%nN=?3#9uT^=@H=CGJo&&L)aR951N@*oHPaJpPrBD(hA*R$(
z*_2KeNd%e$A2O`I+-*i=QCQHElMKA4Ye;A1HC=UoP4K1|-Kh3#3zaQ>SdZE9w-1{E
zabUJVe@X(8vC|I0G$)vL2-eMgr}Xht8JJv(*k7*XPVB|JHMmrOkW#|)j>2ZZbIF=3
zn}nH~B!0W3A<3H7!Ba`|Qx;S*L(ZAJE-ALlxl_rXsL7pD&X-psvUp#LT*}>J*4trq
z6XsfftXw0Hyn<4S5B4RybIkS|VOJO>6HGLeOfc<!Ycs!GWTbVv?w<vZ#2cMk46)eK
zE#8`UM@-b5SfHUuyA~od?)0#0Jm<1A_y=ZuhmN>zDYNB*juXmiK-!nHWvD6?r?Ch-
zH<U8ptn)2CSMH3b5ehN&Qp&6*<5o%52}3%6`})gN3Ls|dBMfY8Y|uVbq4*}nnK%U7
zomNEozBN0eyeGOF?JKfPrD0hyl$BXFtpw7Ye3s?3LIXfkEb<GNNp>o0H!aDoK!_TB
z&Aud3s}`|_z2>moYsov)1;>%M<`hvyeduV|MG+-4I8Z@@<}92m{T_VqjVVY}_e|A)
zTdsO9%T96tbg-rld8ajQi3;kd(7RM9WrR_7iH<7xBDYCtCiR`1RPTPW@o#aVIM?=m
zz`P!FXHjpve*hdy?S0}!@<J+FqijDOZx;E|3F}7<TJ%gwSx0M4ccZwIZ%7fYIhs-e
zz_Mk;G^1&+vAsoeSH&o!Q>UgK@KzjuBhFl<E}<uoiDk`J=W7WH9uBDm=Tf2@ueQRQ
z4!hud>w`*$G^2zTQ-}(8a8RuUPzhpaR^b#^jBP-^v{Xl)Dt(EM3Hxr(b9UgwNwJ(t
zp5ukbs{dh#<Y$m?u-(#Sm8Hw7HZ4ruVqwZyq4?22T}hs_G<|ZAyqrlIke<?ic2WvD
zlEH^!T(CxTrH{7lR}8ytFe72=+PAcm#iv$!Jl<kJG#wPcQs(>$FJn@*?Tf5U&VN%<
zV_Mp#dZjs-Q#~z9SG(HeN~(rVINh~4s|_Y>CeaX`+GvMq#gHPVX+;mot@O+3T%MNH
zHm^O9q?T^VtxZq_5XezwXo7}+>W_^5lrE?S!<KPd^P^ypwfWKA>H3+e*)d|dObOsr
zY8~sf=Dy+p@pgmBC-(975qd3fYAigUk32xcuXp#62S_`8<PmRek=%|ypV9zvt!J9M
zzr*TW(%sYjJ%~3q&BhZPqR;IOGO2|oI7o#G*`G8I;9dboIhg6sO3z(?DG=KDIrp)1
z^+5^bhTlO6XcOB%50C({=al-W1URX;<R5?j)(C6uQ7Rk=iI32J<-DN7%zNabLoXfw
z2Zj~M28dI`^0r|G`pI$Et_DqsJmwN2Bx(;DVxUz5o6h4(M;MgE^=$UTL>RWjc1*mS
zGmk{j1`!VkYskt3{*$MFA_LY3O+^N%`%{s-(}_ErL{A&cWa`VgaiW-H+TPhX$?8~o
z47i)n9!gdaJm;>kBwPI2Olpoxgezo$=w>=FpDW!2HKamZ$c%EvMU6oU{>#j``m8Ax
zhGDNcGNjscNe>x5aZM%6eks4{mj2bD()I`_sDJBKA>*-KB{-{p80})k80$?nb*_k7
z2yMVh&HBb!>UQ;JD?Y7QZth(&RG4}crMahV*TrszJySiKI^BIWoO?`LSa(~BV{M*)
zP0f$Crk*PDx28`p)!k*d;F{iX){J{AYz`p8sM%{qq1C|F_8A4XBF3dY`|W}O>x1Ix
z0QEl=M+e&eg*ZBYV0dsGU63uFKNLqt9e7L}9p$_oM@OzcD~>81>>vLdfmHE<c&Y!l
zfmBK^<~zmQj-_sQUW#5VfSo3l>e!?*%u5x)0IHq+RKtWW{&ch7g=l|t?$q{+H|LsM
znWG&=xlvB84ly|=v%$7|;aw{)^<;M5z;;49y+%(zY}urL#~7it9RP(<L_IokCoi;8
z(kv+#SP;rg4TJih1^LQS4!~X-{T|*N=#RMXAmkC)GirC+8#+hcF$zuA5EYAmWvkRs
zM|f$ZP~X1@Nm_cUKfs3u{g`o`q7aJoj0c~idh#79il6PJ*d%TLTY&OTfbt{Mk>2vo
zPI((DOD!9JW#F-S*W4a{>mIEZ0CvQ_Lo=s?Lb5v=Y0mGnWZD`Rk<Xyfmpa7t_05~j
z{|?X~T*&2mjKT7vE_Cl}P|l8a>`+_JdEW>!fZ-CTdBLr`NRXHqtDhOOOLICvP6n+X
zU#Y6Ecc~`7FE1{oYT{g0<{4g|jt_0Ticg^cVr7zloOrn6A<mMP(yF<UqT;kr4ZD{W
ze3&`slx98X@J*OO&Q;cC<a5N~O3!x2L$5ROYh!GMqUl~#hd}}>bY}T+=uzfQ>9u>;
zWCwT1u7i7n<obP`-1;-jMt@XBhQl-7E2M7oij?4bhm?B2H;Ad<nDl5tMsr;%SFWZ1
z0sim;V&<2j7bz8&!T1^!mk!4bCztUTDV~4y*z+%+KflN#G?&mi6f|~_@Lft1C6ZCX
z^Q%%wt;Xh7qsv3`7?RSs_%8O+u@{X&lEz%lDoTem0$qKJwv3aR2HtXyG%9{WriYkm
zCrXk2wMsEifFhAAJ+btLR-z<#$Zrsn4t~0Mn6nnCUPz_5F7A~$$|9pxL;g1ONaBAg
zZ|B!+i`-IVnD-b26k2I6L)P@F^r%n(bR(zI6iF^F{ZdJ$b$P57)`w0fNza3RqrEOs
z8`;N%GTy6`6iAQPMxXv7uj^ZXd|#f8cS2#uSLu80Kk@dh^=YjVl4PsemW7&4T35+-
z>YP_5%^H&@>v7g))wbT3Ji`><i?n}kwO5omy>+YhR8k2Q?KzcAe-sZi6VLLw1?8$W
z6)Itr<usGrF=l`(zJ_L8g>_bS9hbR=RyR>+!}v}wRR?uFC#t5dq=BL-l**bySEcfm
z2du9E(wIiqS#!g~$p@rEJ_<6g&Qmb1&v9{`wHcRZQ2<RTE8<e7%>swKwW@z2$EpfQ
zZUg$Fs=|lK_bSy>^^fXIrdV|>3_`^aGnq=<ZAeokDu5h*Vfl4`%uvCL5Ey;)HpX(i
zY{p?g<OHg4UR9COVgTJN1h4&Oh}G5z1k1>eMoE50DuwqF&4P_#Bz}Eoe|>zu{MosP
zA^p<Z|NZe2+fjqi<W^n@8%}?J{d)TC<v*NeiBeKX8j(MY%rSG)@}q!_AZcFGT4N@`
z3@EG`Qk#OP{<^lyPM(~jtUU!RIwm{x1W^)dF^pjHW>vj41-7gvj8fK*B>bf1M*}*n
z6<`%3tmHgGp&h_FV4Za*Mc(G(h6^7>qHT3<>QYj7r*<flROX001#W+cj1t#L%a1!0
zKIIxWy^Rn$uQ*zUs-u>`z?}9Pn6lfVij0Hg-&}I&%({!JMxkhlGOD0w=c}|A1RJ{f
zUdwC<aEu>`@1&*m&tiS!LxNP(jbzVq-SRJEod5~~U+Z6t>p@U8P$j5_x|_?2pB?os
z#sSQ{A1EW7>f|}%)y04RINt|6ngAH(a~(yTpNRni&8+ieAAZcejiu?i7Fm)?$31=q
z3-D|P<;<pB&~pVXF>6h5#pNuOBV-IC+Kz1iIosWnb4B!0H8GfFZjar{1c)i^)8jT%
zf!B3{MGN>+^Wdw(xeEUVWKH>JrBE?$q>E+I`!+*JbZ@1nD?NWR2CnpRJBKh75-B>c
zS&g<;qiuE8JopW42f~<+;sRPWgvdkVVY4uH$dq)OfwXU`H{Fp1g9qUZFwSw!tj!8M
zZIzS*FnN<lV%51TIZBhUc<#*^kdK*?YjWi>;}Q~ckZmcUYpNZ!vNHIbtlk7|7S<;Q
zs@p>98*NIJi8Oz!Zn+xJWM0>nrnwdi)ZFiFJ@X=vYi*}Led87;Pq{Ujr5U!j#q2DC
z7;QP_1%E?t1h{cyo7xV`igH<BDTbMnszDR8t_5!ScH~g=YLl)SwJu%mixcY>izYpc
zc{UU~kJfDbXNU^dG1iO{v5L!LCM9Vrq~Jf*%hqbpK)QdX-b|axW3MF!n;M}8iF{Mp
zurvO-DyIt+O}LypDsSsPMB6xll#Y66;;@~IvV-mIr0P*3bMhsXGut@xqxmAD5f*s9
zsOV)dUy?AQRl=kBVi3x>J_&>}s@uuVzUuCN2<c+QO>w!FpsTaOImde+fv)81^mKe5
zfUZFc7dn3pFG1Jf1+A9O9;_Dr5Pn61$KKoj<JUHrfE60Yno7kU`Rt`uSY6q(hVzsh
z<!baU)i7l^!pPNrIo%QO#0Uzgk>UfFaC$zmiam7@%61e6Y@Sb_*csIj#VRJtbFNXH
zpI@YwQi3%NO3V7gj2LYE{Z{xei)PIYJfP(|<)eRFUAl@SCi<-7wu?WD$c*rlk)G>p
zt0qp-BgLVoX?hl|NfeqQN|kAXW!-R6l95`wH^fN3fB?0(w7ZO_CE@H`#n*O{MOe(S
zLAh`3DNW7ACJ6q4Gs}V-cAQx@+1-E#I}B^vgM+nm*_v4#MJEso*IsjY0znDozNVn9
zE%|?m{OTes<N9C`mQj0VaV(;B%Nabf(NmyS(22|-%f9+uuE=qUx5c7=t>v;2G5ydV
z`&EQigc%*-@fqdH94Gkx06{|w0ymUC_70TSg>GfY{i?B^_Edj|Y4)K{T4fsUolU{2
ze2pAOhT3S_w{&~9336OzeeAB7P=~p_f}Vem@@{eU=lV(-OZ_5*>bzmkJSbC@WM5Pi
zZE@=U8ONf0oI1sFRHmbDW~e`Y)S+ddp6-d~l8mC>@BM%woisB_ODB=W2}`Qvd_N8x
zfPx4_fl+!OAc2$VP@7XUt7odb_ko#sEr4(Ur(}=MBme;p-%(u1`-2eyJDUuVaDRX7
zh?%QF-^Pv*YGw&6KOE5BCC7+1KZKBP-#sZ7DDQ;vq4AYhP!i{M<^#lCgw2ih2HNiF
zfj}7)Zah_eLLgui4IT~zjM7FRV7(s*1dI|afXQ-ig7rQC8fyUHe8Vb;<(7M3k)b94
zgU{Wd|H;Ltm2kl<*yj*M1;T5)p!t7itTPBf1n^&+HZiVGD(f<;+hyGYU_dG2;e5yI
zeKkX2z0LJ-cEb67EI|L9C~&@W*YqAMuckv$6Wa1B1E59MBxx<MnwED1CQED6(^qS$
zvf;3CxG>J$%^|cLrgUfDnEYm#*S}p7c6XPAjXwoK@=Dsdk}x3cR(b?9!@qwPl<!zz
zBUW*-*7~p-#T(AQac3QC^}M}`ri}0n?#A<Q1GA~V4JX*E^6jXzP0!Syc4u7yVNlAY
z*{st4P^j!E$r0Mq-BQofbu0VmXKCeKciAa(PT3seKHfHOPyC0|+P2ob>txvkAqhY*
zZty;UrbRg&Z6&H3xpC*CN;iMKMLg6>H*O~o*g)HP+eFXn^>7NhV0xKrCEzXVECCPM
z+yH>qkE_1XbyvInc&=dN@mg)HpD5&|T7tIm+~q4OwIH=X_skrhQGDX%V@7%7_A&4M
zHU{m%W3M^>%w0`3H8rrIDwURM23T#&3alI836rGv;w{|{fMZH2?mI&Nz!hpAFN}pR
zMxj%t57zFd#O}F2eXL4dZn;j`JyN7DWp_sKfg*KrK}K+)s#`A4GW$#BV6)MPeN(Xe
z8_6q_-j|^lDHH-RF_&?<EfbgGohT@m5F#m;fARbD_xtZpvIs3ju<>X7;{V>>pZ<H|
zDxeI)A`Jihe2Oo?C}F%Fh`caX4Zr>!zCZoJ$u?$^*5nVA!>=C$8~L-K7$l)6twrom
zRFa^D!P+29e3emzS1FNDg9|;RJwy}~)6zQPpv<5muf(_;A;w=XUM+G3nro1A)nqRq
zfBcbbfrK~4--Iz@mLN)!#(TUQRqq6tnJ$vNfe0k3K-8e<B`>Eh*^Wpv%v+`@AmUa(
zl2+ZwpCdo5g9y%ZehL<)(*}e#dWN*ToiAuXj>{?GB|gi%5y=&1$lR;91tH1GTyI?F
zZM~r92)&&Z&bb+1@>MOFZ&vS@%{RM|f8?9X5utq$^MI&sn1aH0%+{RTNGlL$;>wqz
zIGSPnNFsJ46UAGQq8EmuzcXJ>Xar|<=IjRaPfzDyofAiO`?aChHJN2Mtr#nSLgbUv
zO3(GEw9-@k__WgV+e<49=?BsZ(=)_pJFWC99!)Du@13;rxZX9bFuos?R+zY<f4JZ!
zyXCd1C5D%p8U2yeqREshY^Rpu3Rk9<9o-H|FVYBNc0+pEO&vYOd4S)IFa|9wn3j~F
zAy%ZOEJYP-#oDZJZs^@1fu~B@Lz5sB%24bR5ySFb5KL+Er6n{|HVa*5UNQw7x0B~_
z9h_qjJuWp>!A`gB(-f~l9Btgme_|8YA~8!Ick#y{oNb#f5C;0W@KWWw%3^b$P&b-}
zqM9I$xgUwqgXWZAjgY?Wklj(+U||t5$w;nPDp`xB5tZ~r4U`-uQ)MiATw+md7dxV4
zahFKywo6W=t08Xl7-5!`gA#}({A1&kL}_AWDJPO8sit<TS>&pXJHC6|f70NEm#%r*
zJOXPRus2z5UAjf%OQ)zajWp&mfcgL_4K<D6r~Pg-!Fl3u&r=JQ@oM$gYQm;bLAKey
z(<=48QD0rOGNVRG!}$H$`s{}_)K+-k_9A0PPr)#gTotM1lGNmAH8Wj>Me?|ULB*pX
z>$Am`dR#+=BDAf*Fcrv^f2(pes%u)cEdG*6P4Hy;X=R_qZX*lT>IfH}w+=Dd_PKo+
z1FlP!JO#BRNK$HlV|dew>Z(z%w5o3QuF}Jre%naOlD_?eqQ4u_D|G$14dwu9G)E*O
zLZyV5VPGV6XVj{23}mY#*U+Pw5|UoAmDOdNwX0+5?5<gEsD4lqe?ilu>C6j^qSf<c
ztG~2SJn2N$=NWas$zIp6n)uc&c5E_3OW12@A{C@mrEyG>wCA>Z=6{yHRXYq+V_|}A
z-K<bmX45)U-wJAr80pyNfol0eWmhu6?3ly^xY6wk&05pGolK&z?Tp$Pp{3_i!&r67
zu2#qScrujTaa?M`e<(pPE}awAQs5<z>u(hYT`J04)=XBYse@JIrdZ@4XvnxRIkq%f
z%cSSA5UDO)EKqEeXqCb%bKiDL%G{PZFYXQJ1#A}`lv-wwqdQ~j+LmpyjlBXR0M(4*
z=#5)?C`(l}^?S3`%YKmb!!<cu=eZ`U)>_0N#x0(}gEp<Wf2=3;U35$$qehCn?D#$k
zg;(j;`+jNpul4<SXm4g*UW%t?O?U)Mdg9U7<7_wj_2wo+U8D;$u8aOx_2_(T)bR=*
z0|cSL&g@pYx?Ar#;)y6IJG<wgiWaCZ{fv7KFq9YJzypJx;Ocq7^~X`lTA{3FgxN_{
zo9sYXr`Tk(e{?K)j$S~vs|RzBpT2-1`)TvZhr5hxJlw_2=6D4-FM^&?`@ln8CBO#G
zj~(jXMiGV6Gb_waTy8&j<`i-B;F)`YJlxNW4<sob7_^Myq&U1eTG6FlfKfc~;JD@k
zj>6vQ>V8+cic-a$Vg$fm&G7(BqZOz>`?`p6J?OfKe^K4fENRA_*`>rH<GFfeM0WLV
zhW0&IMpgvcra6Lgb;Hu}G?i*Iw&Lz9qkIX(^<K$XPSY$s*L!kvaszLVf}UG+4_;6d
z&2(?s=b;-M11Qjj&fgYqar)}Fzs2b)#Yuylk9L(=ZJrO!0@_ecRiB*&dag%hfu8Dz
zUm7!hLpzto9O?TnjX8Q-86aO8vlS0s8gsq3FO3h-yIdM`zOU<g=L=x2_Cdj~w6H;M
zU*B#8{vFr1i(MM~rZ(_@e(3?Pm!TIa6qhmk4HUNrDJgCS3H5yd2me3L^Q)IJd`1%j
zH8Yoy|1J}kUhO(6m(VXMn175<eEQe>FB{1WOA#IYNk8M;>-*;S&B#7@5LWu>>yHh6
zuto_JG=8?i$uRx#@9Ep-%ZPR+CUH*w59IX6$25%m$`XtcB`BQ(=p?8liq%feMPcYu
zmY{u+5>Nv64If;+MR{p8tkQX;pv<I*RwCSuAmLAML9J4?2y0-Zs((@T5HmG^O4$N)
z5;_iMY(9vn7uY0f6DLYQ%~?;gGZy*w^x!1v97x}<L?`1#B~10dw4!e0*Gg0CD54LG
z705wGXJ9lz*x*Nu%3hAqYJtYR!1!_nMkgYg--ywz>P3SROSParr9{Q5R~Gtl`#oyB
z)(dO=y^Xi0O2$hjMt?>Hh7FkBU4DoYGDT49MjQy+VlBswc%*r$L9?=kMO|Uf5QT?0
zV!XzlofQ%lG-ABMo?Q_5EHLh|XRifly)ZXf)S(|Q6MZzo$EE6DR=F@z3nLd&NM;tY
zu_U7@vF(L8Y+Vi5yQ-^|kpr$~(C8s9sM^Rz4IQuhT*rVHPJb=^^?LT46G|<NE%}@}
zjWygvLRc>Fe07>&FpT&cnknWR!bEkVnIfcc)-B9wcl)@jK>$kq^tgL8rVbdp*v5Q=
znd*Ou?{1`#+svo30$ddvKC*^V%U+Xy@dNl0T1{jq@-1I19iP{nTaxwDqlXZT;7fI%
z!!`&d(U_&d{C`i+9?l+3xhr}!^C(|S<e9%i`L5T$yBZO$;0~fjGc&VPVLu-|pDU0z
z;wYfu2iv&EDRkD7YRu9XIXadY(KyP{f1eQW^OT?wfr$a8zos;dj*tHkok#CO6qxw9
z;E4Urr~iF?`^VScFuRyO{v=RL^l`^kF-$-B6$86LBY&H|^E;;hKHzkujCF^dF#Cd_
z5ui+*2fj`lt36VO(2Rbk(TG{4#a*jTPUVFsz(SxIDm)OzI?D9Ew}UfRGl{&2!Unw&
z7M@G5IKRxIS9;p_0*bT#Gi|<2n=dR?a4uzMA!rN|l&S@5*QBb1RUQk$MzUo{V+_E2
zCKjngkAJ?pXS7{!1x4c`gSR>ek6uWbB2$~!Tt=F8mm$PeFqEO0wPAdZ9EP>x$ZVt-
zHF1A1Mrc$6ny6ywW(n87zy9EPA2{oC^PBNJC~btrQ278BeF_5MO4TF&xmJHOc!6fm
zqk21Sy(alup|a*(0|F%XWop>T>w2gQ`*N}{Fn{)AF@J(E#{IJqf=26faTda!CPY~S
zLl+_Jo|H+^hjxuXod@a}&|Z@2TUB}yxUp%?n_a4QeSs~BKSAVFf5yU9amJnZtc)$~
zRl4kC47tniX@{d(Zie1<oDlrB*4b!==y8;eknfoihZmzmE-Sqmg`rR5YBim~x>)Ks
zVt-5}RkJtO@+x(8T1;#9RyDh3Z`F3Pz5RFU!FW%2Xf}o7B>^#bXV$t@7$i|%^L|bp
zB`|gU$tvb719f6Uv7XpKy>T&Uxgac3d`@dV)+w`?C|fh<LVe~^tZc1lWvkIJ+OT%T
z9^ZJLo=1UgHov)^Mp<wA+ZQN!W~yw9oqvUA2+8ysd<lyCc{a&Sm$}cpqDF>Ptr?x9
zs~F0+oX&n=Te`EaWSC;nvdrwf+f_L>)C%5Rk)~UhW+%541=!N<wopnG-i9>-$J;^Y
zU~z+<9cNtJ9G0r0*$|-;hThal_~-%|vaH-=F{mf?c+95eI2W2PGXZ~G*Evd^*?(<C
z`3XvT#%?P)KV}*C=(bv;vEny3ESv_6%a+mEWvrr-+UXJV&=IOVSjr{U2yNC{SKFHx
zqLzOBy}Q=~)hoN#1Gdw;*K&7pQEVb6w#wD&a*)sUS~1>>7{i^lGEluzD+9I@w6eQS
zD-M@tYL>IRy;dmI*6N0_XT%t)HGiEGZ@^F~Jz~3}1Faktmwv>kU|r)J?*`~gR>HAx
zz$9PyH2#fHJ}&5<-*{Bi_zMdOsw;Z<5?wDYC1|hc)vy&Bo20$cJzD8p?dIOm+}B$0
zu(n+v%gA*E;rnyrV@c5519T_l+ZyACF?ElynAvzU=7R8)lQ2JjeddU@Yk%QWX6^04
zT$&(so0?^KtIXDxdaK~1gv~$p6@WgJt>sxId(Cf9xU|a(k|W1LItFtsr_EA2`+BMe
zWo>{O-V0gVlePI2Nsd%IK!~M(u`QCak0`rY>zp*_wcD(_l(RqA^Ov@h%-Wb`Hb|*p
zTH)2&8OvH$uN`3C+GX{jntu}Tp;T=bB-Q(sQom_K%&EE@O7ulyUY73nF1@rF8t9>k
zad=ziCOr$rhHSu=O|lbl=HE0t!*wNsGC}wN|IZRZTO(Z5H{7NNj5jP12FyoFgaP_a
zC4$yUcyI5Vj1E+<OhyN6KfOfI^sC?}mIx{Y5#qgXZVXhf)XISEr+=3Sib|3u^54Eh
zP~e2TX^AjmysktTu<eUCoc@o{2M%uK1w3I28R_h%r*ea&d&Edr7OA;y)-R_hP5aO=
zsoqPuK-}ey+d|wcPI>;Vv~hWGczJ+Zue1l&cfq|IzRz>x^#u_4oU6VV1e)YxeOH7D
ztl8upDl7~o2P!Njoquq((OC{H=+QL=xBAq(bCeRERCr-f*2Z=xlAP{hxNmPRmD4Am
zyX)Ghq5jJ7rT*gCnA)n9v%6A*F%;Ke8RuAoG0d4Y7|-H*f-ytK>9U)C?rGlMx=lM_
z&`NvZ-cYO6tk$(_l}dM-dIqR+GtsscysQO5S10>*eL`m+^M5JBfzvsj*PYHiIa5Rb
z;*Po#+Si*+`;YM+48+H#nraTh2r^Z5F0LIXrFn@_(;w;0l*$oOhV19XdUVxPljSSa
z<JJUKj4X86e8&a5nB8S(vA&~s?uiq`w$5b*mv&6a-BK$eLu%&wWw^w}$AplM3C-n4
z-us-y1}`)^Nl35o3$F(s3v<p2qitK$Esx6<$84b<w%=gA3b$q(^uWp;g8~(lA%>BG
zVzJxCj(qMaRCnih$Ce$MD+z9Kk)XLy@Hf`hC(oCm7bz41G%=TPxh)fyM>#3Ie~#Dg
z_s3s1vM4M?u<>iWwNH1Cn;)B2eGXpOV7jN*O?<(cg!ZZ<kGtxA{?R>e9$K!wEQu-k
z!^rOEOV?1o76%3i2TE&=nK+Onyvm)mLFo7@JCHw1Y2X09#|7gV?ou={^3%PiMUE^H
z3e$3gN2P%a87||*Idc;n^{gLBfBgvs>i3+j6=YS*S;>iB{b<L+I%-GjnI0e@>=IVe
zUD4^VOAq>!onY<W;ZJzYfj#MtqM#&oLiu`x&!<uqj5Q%kGiH#?rfA~#c+QKUX56Be
z%#!!+cUvoC37O=_;zOgm;Uifyp^W2NeWtBo=lhJgY5cQ7REUMKmk!EAf9R<(nJ_yd
zB)6x9L~3EB>O*q7dPod<ob|k7<7qkiA(h_KY>XB<_?FwgW@AYN>1Ly2U_-u?&jn(1
z4!rz;PQi%4x|Kq*iOTUDkvD?EOz*ThVyz>+Xssd@7old-32A1W6i%7Jps2$7sm5M-
z>DElOh9j3Uj2B!BvvO&wf0~9v_|vn^SCk&dnP@oYvPK{1Tk89lN~OJE?^*@Pyd76V
zLd~qo%|xBv-;D;^VLvD{@U7pF%JthB;!h|ay=M5Q4F7-8pXhg}s7vrCb4Ia0Z6Ps|
zMR4?1?B(4c>xg_<0>c2;GiXgguyPSv|L8FAOP5B20lK^C{_P?ef0HoSrj=bYQHVxu
zKK=dj{L9lftn=N=KXC?-Ib1|o6<+hBu|j(1T=%^^!|&P>q+>PEHm;vS>wp1uVo!ML
zHc$&V!uS#Od!#~P&q|FZr~QEQeh@GsRUIP7o?7L79}}xqgNS5PuSR@|T*}<|i1~F^
ze2mxlxempwKfBFyf46z){TVd=5?qpY$(<ya9JXvDmsS$e7mHHS9=ZmHqS-?o8`={U
z)+LLDl9d+@sT09kk=7eqDI<d`3A`$YN~-skzQ+NLt=T07R|pd@&qCD@vyzHivulU%
zPp@U(3ub*8-7}vpDJ>XKaLd>hF?<YOX`^TS;!e2?ucPaBf9QIr&D8)$V-rdZDc<w0
zwn|<qvW6*Cc<4qdW?Q#|u@6T(2pf9m^brnf``E}aNPN}1WzPXZwScvt(Js;Km<xE!
zeSNnj=~GXR?LD{XNJ{<6mXWr+?mK$vmkJ3Ni!Bz7*R^NPs1sloKy$5L>UI<&H`+av
zMa+Djgs&=?f1|sjcxI87AI=Tk*yW`l0N7z8)(DR%HkoB6mM4;&I3zkE$*^^)8)l?c
zaTsZqCz_q2>i<v|>OKZ_!<0lv{%+uZUBi?7-;@1lu^n>zJ)tRRV1EjwbXye+G-zZw
zg_ZQeFi+*v6v-^rY^{!^iXEyiSL|QibXZwjx1-gyfA$_I6P(pcr*s;56_g%|tk_<5
zuxVR7N?y(Jq%NhjFWpJ?R5zl^K9dif6Kbc0Q9ZQPH7opz*<gg!t)Q3&rY5W|DKRlD
zD{^GW$pac7hK-b3Te=kbP(u?J`+h|4#2wkI-jU^Ps=LA?*LeSib8Y3LRA$mQT*B7#
zjtJIXe~lA;r_|NHWbZY)9`&}HK|%@cvdgt|Ky9w|M!~uJe5gGOGn>Svp|Q_KaZxk3
zldM>+t@obtD#a11%}#l-Edm<5$T2TUP6VPA_dpHD6_0rv&IgWpo6#3K=G6e0A@GGy
z6dS209w|25&VI~`nns5-XFTRr0U1P>K74Paf5xaBx6&_1$H&9>nJtH-Q)P~iZQg%^
z5FtAnP%HiJ<0p*<&_d4anQiOe!O^0fB0eq0<vyXDLZ4Ud=o52Uc>{#(M@!Bd#~Noi
z(o$9nwdUb@!|-U>Qje_~kFH=L$KOEJnkF_J$G;DJAOkd@nALmpd1)~}gp(;iQ)$M(
ze`fcXL@r~JBk?cD0$a=T<KITm0IP-{M3M<JQ;x&GE})pXmGTDub!7`!H^DT)z&91S
z`lBG=fJ=vVo@Uq~DJ!<-e)JsMW-Q&me|rh9GH<8Y`)wrd3leunBk_;JTfnxMv<<oq
zT38Mj0k_1+im+xkY&fn6ZX3=Az-=@7KtkY_l!Nq3|6JZkJrUeC+|CYeiL7vAc(MAD
zA6U2$E*zAp&=X@aWp?)eVencEKT^;BgJH2BHFvbT{{U#9yO&`UC=&rSl~FAgms|}%
zE0>KxDU*MDIsN(m>p>TRwF)-g#!LJ3`hNK5(2CFADr0T;*N;Q|gA6F+u?t373f<4Y
zyKjfzTB?<omRIr*>+a|GuIc%`C@^SIpthEIqCn81RqCt_%EZ620_m&PEDE4EX)<{d
zbs;p9v`#pH=^*k-j3+(9_{*DzMXn@c4YyoTR8N1xmBkD!_sr$l<`yMkF#}H}sZ_mF
zWM*^Gk($nLlFJn1y!6}Y$?9nOo~QIAx{9Li`kyqjNzd;CPc&YUcP&rgMG28miMO?*
z`Q?7T3<_ro)US<+?jabI<1m{DEl<Rf(qibA|G7+rO*i}`zLTC2-|B$ul@ZS#6ez~Q
ztek&xC$lSejZ|x5;AQ|AtH}8~0^p`|HxzAg!ALcfcFz6j8l6z`3%l&HBd08Gnw~aj
z7+EodLBlcoNi+NFQcL2}xTh`~iraO0-lR(oD5=?0U78EJT*}yA-q?3R5u&zBYnfy{
z6@Z2a>{qHB6>`ziZKx=fa#l;ZlJb<ZDrA3|B7a<)QR8zG{?<~<7?k#P>ew6g^<+44
zi+oMIBQP(NY3JVZKe*#qq7&lrnpD|rvLqkN?0SRBn@~B@vCMzco8$<ET7IXD<>xZ~
zoMv96OHLIkJv)_yi=LC-?V#&4!{<SYN^O;m=x(jlx6b1yEMX?Er%eoet#arhgB^eL
zlPFDfi<x)U?I3CD&+FOD7r-gNn&Q&PFrx5QG+O3A=&JBdTd_4-&N7T7PTHd0yIcum
zWG~w$S=yv|(+Bgcci5TCyj->*67sH{S)D;_ILJh}KE5<HDQ`K>46u?^-AVvZqEnKG
zR~(!*OVLR_$lEnK&BXG()mOziIoW?<psSQ><n5+wK$6TzEvE-}ffwiFB-hHurs2}Y
zMA~X#^(@<aSM^dZIVC-DktZ!cuJ*c~M)m8OjYg60doHW;L#B0i)~qv$RzFudeFWMP
zoxY9CI+xY5v8(ww%c!$F#bfUVLz1<X(=#vx5*I80<1^3@kyjS>+`u#x=XihQZUg|<
zDzMm6M!k%U90@vucC=Z-MQNKU_A6jwhYca2RW{HkqsxYC%qwYiqT<bkGnJL9&ulZz
zc@=O*?KZ01xLqglcV}#i5qI3Cn_!d&|H)QqxNfvcLv4jswr%_+6UnuLC_e9R6)QQ7
zV8`6kQuGg19Dt1gbl*~I^38u54;eBTsG!}9;<hRfYM+*mhU%Jly}t@WR)sL*MX;d^
zlj=GR!KR$_NL{B@$A(Ig^-9{FGCHfAn>m91x@=(2N?JSm5%Wp%aDi6xH5!%CX2$n?
zH(wJ&T^8Z%HW@|-Wf1n@i$<~I9o^NS?AgqYu3(H+#`v|WAyS^)4z7P-Vs>M@mak;4
zv{l`15~r1MGZzH?xCUiq^b0mwF60_-BsI@NZ^H^nK?OhTbS3XPS}5U;3cNAoGD^vr
zco-E|>!r7sio*GQ%-XwDD85*byMn1>xFXh_pSfNqDOI=iwUAY49D{h=Pz|^#11`&O
zdg^OG3v!gVCfFB4U;Tf0D=RS8-X0O-A|g&>Dk9ErTtmdT8g*iii0z$q#Q3p~n5Wf@
zUebQndeS@Aaj5#OIJZ`y>|_?twHBF2!jhg-VY;_~tua5dybH>^Y3X}B$6A4K<BH&b
z%wNWJi&;(fVs!zoRUh-{p8L2G+Sj^<iUSpzX0(U(+bRUWvJ!tY{FfC1WGDMHx8p9o
zp}1p(prL%DLePx<qzVBVR_3z{1XHFPk?54zhnELrK@z8j4=fL$?03nU&$h{w>sF&o
zseO2P0A!hFO&?eu02>s0yJO;MDQ+tdG}JC7n*#lt(U0&gNv@s6<Cq3yOHN>0iW@*p
z?~M(`4H&#X>au^MjDZCJn3;?hFt{c~3_I8l0pkL_k}ZOSbC|KcA9o=NN^$;%N!n9B
zJMb+hfcA6Oz6Q0QD<E@9Vk>o1jsorA3Jw(`uI1}xNind8O}tQ;$I{~v9D`Q^7iJV&
zF#5py!d}_YN6h~}xCZcY4dauW>kk90g3dxoksCgVuYiA;l%VDP5VO8Hg3zU>fzVVx
z&PEtU^`rALJkAi0fX6_41Ss?SJpDv}>9e~*6usP13r;HVR~2onk)<QF6xW~YI0ZPC
zqqq<nbRMTBn00jigWXuE5o4$O=b{=tKK@llo+sw1rg3vjyoABMIPmC0*_|Kb<L;7B
zPY<j-k=B0=L@zyK@ZBF$I4=|tLO4##49hAghXKZ;hXYcf7CCw@qeC{74+=eBIC2%j
zg8U1n8s<4UB+&m;O_QJKs_%*#m23hUgUn7o#besN0*W}~MwOg$>cLDVe$~fW%MM`X
z2qSaO)Uo78&MLM;X>?pb!4at;ntdtQjkGo@^w=0Hg`RpsGCTLu$w+869p(D^e)t#O
z>yGo6p%*C>m%*V569O?gmjNFbD3@UKG6|Q2NGXng!WCV+?XM3%Z$wsDaBsqG_!^&X
zA2#1MqxxtKx7M=9mreKq6>#MwlR|N+9Q*z|d)nNOVyDRjjY%I`u<y@oX!(*6^nwr+
z#%SFk2!dcmW{l@5{FD&H8zD3yfOmA!>XGo0(5>>*or6V*D99Filt_>YO&B6!hEZ~o
z#xdxBSw5orT??r{Nj8Rqk)vcOI=!ae7P+x7_Es}JsDbP*Xq9lAbvg{`K|IlNRQ3cs
z2{lXNK`{su#aX+SFGpbB&njuL+NasHt}~;~nz%j4Iqv05T0luJp?9~vFe2<B9dXPU
zYQ^eDTus-~+O=S&3`grGkGUT3S-~oRT$}e`HcB_?xi;y99buB)c_yI{WQ3edvRmCG
z+EM6sxMJWLB?fSPzGT8EII0eY(5DBXY=~i&eL={DEi_xDY;=;HKEb9Z>m(&`G=7`~
z#9*ycBx-KuOz)UBDG7pFTrDLlw{A))eFX&Pn(-@_p%*C>0XLUH@GKJnHkV<JCMSQr
zTFq|bI1Ik$Df|LeC6Uw*5Cmw`c7eU^Ah*Sy7Bh32!_NC3mXameW}F|_ZVparSs|s5
z<R_A@?tXT<`(yWd+C6^&c3uTC-gc)So#6+XKaJ)=N1ZoDIWXPnSNHc`YwP%W<qNe2
z;LG#hfxqeS1Saqd!%qy)GUo8)(d>WA^Vc$;%Dw9^d*pA+mr3_5BY5kR6qalFHwF&>
zoc`SDvS6*EOMj*>dG~(W{kN0LC$x&7`}fx_jUyNZ!FC3`!hqfH@7<5xUs7t*rTI7g
z2kGwj&rXbdD*#5#0BYxmCIA}EsAA__RFEF^0OFh0!~oz2U%Yu_xU89M^7(%wm|2Oa
z*%TxtGNPIoLS)D=OYWs9OzW&4dHItOUVfA8oHEWz$yQI+)n<EC&Ko%!&ep*LY(mYu
zH1B0iw;{cWC$jCpFTj&g^DN##XOg0<bu#kp4$P-oEpL`$uZ?KXj8s+<E+7|Fv}e*1
zN;HSw$KlE86hf47EEuvs4<CPdH<OXET`Okb6#CFS=2hUcfmIADa*sC3L}{`~LM~xN
zQ0$WKU4Ny8B!tP#+a@HN$er9aO;y>C*b%%70?;x|F<D9&C1gtkIu)7DrUDIpULX;>
z!Wlg8vshY(im6*n{tPaj>D$`tS{Dfn>|7v$Hn>7yRA>Hg=84nBpfrER^{{}p3N&oz
zE})$v^4Hy^O-?ay&Wy5xo-`;9ou(FsNYdJ4y<@Wv5q36qX2bQQM(iw((}eAgZYk0C
zG?i&7(O3F@E1XitnYuLgOw_pMxY=<^6qR>w%RLH-VdUb=eS<Klp#3OVZd-mXn|uDC
z&;>=&JDAIJNyi+Uqdb2%bAB)>b=6o7kh~ktRVi7{WxcmDXpD>tZ?2|E<w&Y2g0d7i
zCFO|ZD?r8NY;-se&ox<MQj?0XZGo>?uH)QRgn2J)$1~Aclv*+nHYNWlZ6`X|b}0JZ
zm3XXwX4<TDkY;zTMFWoE#@RL2nK+#2`4?_*RH_Gbn@um}+|Pd+tA51x97^7H%<HV9
zNp5U+SM$8=26b)6y2~)tf+eVM2>G_&Y?1bxP_Sw_%)#3Ir937VS-lf2EQ>4`LXM3o
z>LSZ|srK-Yx(FQ~hKus}Tf`$IZdexy$v5gEG5bMv5pu?7?1MA#snV5I(Nx%%7exrf
z#x-A96rllS5gvbSld0CVMwv>x!z#yhR^imv4jB?1H@6BOzzW=>YgM6jqheJ^T~w_u
zuEiiI3|{J7Q&X0zSMK6%zIrMpt(g6ps*()nSNEzdg|Ae5T&}iM#ZvEzYPOWe(j;z$
zM5ny>_hM73=`)k8q5yD_SLH_zHYx}8TgJY)^QZ`Lvi*OuC1s?3U5FdzDpKOzxw<F>
zr-$Ltf?7wy0FIxS!2&7MoN2~yfUM!1$-tJ#3W-}IyVAE)tAlW8b$l7JDK<T9S=dX7
z>ybTfh^+NGGuPMqtW7Nm+L8^V#BF?b1(x7d98Qx9NM3BSFnwFA-CQ%oM9%U3&;YSO
zPQU8Ik$h%w{?hw@{%v3_aVxC(?@;qtE|@Xw@6Dm^?$x9c4zLm5J?r38{S$?IY8qGv
zqozSG#};QwMI%6)gX7jp9i22=W^|6-q_#zhJn9e8vUc3869qad2yQu!{afJ?diS)X
zfYYC-qtuo|hx)+G8eES-dLxaR_WhgZ{A76#uIfe;nJ(aea~=c5W84NFQ}g_6<<iQk
ztqxf=R4;cnTdC4bELx~F=ro7ozKxc3wf5PW#(s9%FS*u~x=97V)-z_73VfpiTLC<d
z`eir*_iY67OQVJ2RDGxI0J@wu_tJpjytchAl`#oEqfmC8>n#|y1NDV2@rB;4kBV^?
zD&<Q6UH;{NqMim;&|cyCOZ(yiUZIytGDs%#!nGAz$JLTCa)owq?3Jqn%ZCV8SVBQ<
zEN8THL{y%G>Y1f6?C09Fc}8u|<{8<~i)AV&$OtPlkFD~)gbE_5V54-EOXOBP_N9yu
zzIhE97NaamCHcr7$87J9BR(nbAEjHzsEnnF`GxaLrVL`}v*&!n%}`Q}wF}*RvIbsL
z80$h7HD+H%ev=WI@3=QqE;*#m|2;9?j=iqvssn0H9VI+Dj!Toi`BZJ&ivCdfen0L0
z2f!wVmqB|s69P3em+>es6_=`3DJz$LNGS$?%pa&$q)0u0Ai&xi7wD~v+=`xx?w<D0
z{r`tj&zaFYoY+p<9L8hMNPLsx$3zd`2R;0HI6fafoed36c?O5W^S1%yg_cjy^3d|&
z0F*T_JpUN}c+eWHTt~S`Yh&bc^4-gwAAd36Wd+C^kW~PWzc_-2^8Q%XbNOxag(tax
zqg)Vw3k~nkMJhrUj*;5qpU=M?bP-sqVB^<#>wi5xAO1P?;#&v`$qaw}IK&rZKpBsN
z)yj#!KmQ)S9e(erHeOn`<PX-v&+kL`<ZCft&|*MsE%VHPpv9^*SR0gyud)H@i`Fa#
zptD?}Jc+pwW=eiKdswANT3o?BMHY&GBotXp%VJ7h+(RK>(cCQ0*yJ_(l({K3!}yWZ
zn>`Wr7P_!kMvPQAgXx@e*<cr~Wvn9S?`f?$#b_ahY_Yer6tZFn%~HEz+%(xwMfNkl
z$23gZd@lYEV;tzSq>{m?uOt#3v-LP8n9<R$HrtY+K4E^uxIR&^qt)5T3Jec_g+g0T
z)OOa@T)0JF`L>J8+s|!h^kb;+`Bb#<y=;ljxp)BCOWJBha(Pp|uz32l#lymiMth~b
z?X8$NxpHCFAj%WJQx^}-Hnv!!#FC~iBG>&WnWH2SI{^|AZsw@&@-_6lHeS>-Y(;np
z>7cQM;M$u>C=#Obj(UEgOL9nm;zZd&*=0EylA@<kk{C@?*PpU`QXnOH>p6g@nF*>4
zVgjWDuIdaeL~<a;{fm%yPMlj^<wr@IR8T@Rojpl531q0NsmL9_r&X?oWR`a;Sin1l
z*3bF(g;uPPX{@^&!N}e@h3xA+&WJ_@^eg+I2H-wd60aLO-w$6(G`;edSXn6-e?At|
zP?W2FBynd?rZt{|0$Rw1dNl=row8DzdcKQbdck>gezmNKpvZ=vJMgNkD5sXgg=ROL
zgG@$w96JWl0`z_=TOHHal+!Gww5zEj*VDd@q8V$rw@`eI+^j&0y^N%hb(@{e6%@r2
zLHjrCohSqq!W~nkBY2mSGDo>&e_e9&(-i?NGRY}Am#yos+G0Zr4fcxY>r#oV5U~r+
z5$bG9z(F|<-KHB7isU5KBOcpef0uBwP63#E#?zc)E4^1(fvp0j4_8)*<URpVTi&~Z
zatd{Kk@7Xb!vtxa<Ezp^Lb2PVBTNl{?!g)?sr^<4DD{2xODY>_H<%-~fBtN@FVb=(
zS=-AJePo`pzNGg%Bk8QH^ury61eM0V&`<lxx#Sg;q{4>No-=jqRC=_}FVe>VDksOm
zmOgw@A0zFUg@r{57+LtxYXzoNAk)D@YQ9T)n0(dzsr@rSg>*HC(&|#|3uf8&AWI3P
zUXm*tp_0!q!t_nQOU1Nge|o-im^taIxBMlU2N$J(;YeZAr;Y{GF1*$H|B`9hTe*H!
zucVyf39c(}z4(mUx+ZbPNtuG*#aG$lMD#F;^A-;yX5GYb#(D*wrncY0X$f2Ae5?bC
z@t3KVtg=s9j#G#C&I#q!8dMvRFH<yFvs&nZeJFHj+*hj8d9(}7e``a-$&Js5-|M=X
zC3;p%c5DqBHsm<)z^x44s!?tyug85$ZG5eW6I-_yu<B&TQy|yA#&MZlQ;d$IwqvaJ
zbbK8%E%!}EUGo;R>4Ry=w42eAV@RR5YiZ0kGt>+Cl}FvK7a)5R1Wbn|KicI-TFwb{
zi(OYDdXOeogIhrQe}D>Ico#@-Pyn&9ckC-WiZ?*|j#3Bd<BYAUi<~LWp6qqiWem`9
zSno$)roA@`#Am(sFX&_w1gmFzt(W0CRKzT<Us!Hyz);{&xB3P=WWY*S=RO<~FwzaV
z@ec&ru6g8UE*N6h{3<5514neIJmDP^yrXzSf_Ic6L|P>Ie=DfkIMZ<ps<%vwiVcOb
z_vXWeHOes+knT7)Llfi};y0X|AqapXch7k3DE_Y|XPs~tMrXY$x6RHLyk9gt>vr96
zTA;K++WU8yo`IL}b#TX^>L}hYJp(7c1^jTu1+0-S%e=w#3|RW#KyNZV1L;u4p{H2_
z#3>N2AV<$#e?W^rVkW>wGl=W@vD7-mUc&1s40bCpxWTCjcuf+T!u2%1DL^_oKUJ!!
z5Nm~Pg7;4g^gp7;P=u&>vi##iq;@S9vaF=Qbx6qglp^l|gn~&ErTbNgIl&vcq&cBC
zKtXc`cK|`1>i-P}nhS9O0?ilSaHe@w%>3bsxdX+)54{SStma~OmrVZ`&S^Sqm!TIa
z6qm8~0}}x;mk{^_D1X&h$&MR05WVXw`Ui$)ZCrqXfRRQrkXr)Xf}DcbCm-VfKe)KH
zME11C(kKCZV6VkmAB(TsPG2WGectR&n@>mL$(tbHaXNjOKwnrrg4KskCJ$g7!F2jI
z{rPMyIvqP5Y%P%vuh&7(vV9^5DgpWiR0?4G<Qbmp`(2fndVlZAz`dSP2c&z|5j^^&
zMfHNt0ond_`fX#Y!`4KXF4JkeJDfKEZbtWoz=%fn_qR=Y!Fo#pC*WY1Y5M-p^kwt=
zsCG6cG@TFMzfQxGKWYX8Xa=owB3%YS4A?k17Y(IXC4*L@7R`WY#H>-WUUM9xw5p;U
z#1_Nd6E*zl7=P{CR;?D3C&%6}`DT@0i&Z}A(44bY3E?fPQVpH0wkK+Jw!&6XAqy)}
zt!Oo+ui9#c#2cr19JTDr?4R&wH4DxeaJbZ*^8Rza`CuY0BMc>XtX1S)(|E3==rg;k
z=Mu*vT5^VcXd06w9coo<jhs`6?T{nyo?KR<`Ppkl+kZtlONSH<FQmEDE)A?v2&Fov
zXm0ii5e&OHS^-L3A)Vj!4T%jBj<f6o(9qC|Vg%x4ElcfgUVV$zX302<H>|c*+%;bZ
zja)I7Mp|LMth98^L&fsFD<k6*OPmU?ip_0m%XGM-u+c;<&EOpH^@0wes2BElyKQzR
zYJXnXu799f@j;^T631CHsNV1JW2q&gak_O}xBrM1t~E9u<$<6YG|KM&ETp_TH)la%
zml4+dOAxf@B_@9R<ZMP=tWxd_;sjbbTyH}vZsB#v@hyIr%~*8f57xh1y*o6MG7i3I
zcGrl@T(9foORSQwz+Ia@mb~T6$`%q?%1w;x^MAD9<GB*D&#O(zS(dD=fkUHfxjMB$
zqzb00Whp=COg3KTSy(Dlyci_2wplXQxZP#BN)j@Ledig~2Dh}Ea#_*6$QGr}z&vKF
zZLF!rbz@!2H@RR)Yd2_U_H$mSBD-z=;f%*yNht{$REI}Wst97W{thX9zp-RedcC)>
zkAM0yKCrh88s&D{02~eQ!nZXlQ|Z)b_w|g*j6_#9y{i=GcsIovtPh_nXKB&CyNK)q
zyAUg40e-pm%UU!LSZ$JbXw=0I5mz;Sd>dv|8sqtqOQqLX8rA*FrPsDXgh0qp42x*)
z_~LM2qHdvsCVN;c)!V*f*f<Z%XdbXZ*MB;KW&q@WAZ`@3p-2Bf#0(lgK-?&3<|+Jg
zor7G|ufq5r5I2%w1mU5?jSvmVOpmFNEZ<QtU;iO+$#Wq>7ynQiX{xj%(LI;$Uz(OI
zA1Vo)bG}7<UjGUdH&ZTu@ejkzUcFS5CacHU2kAU_m0V(#51qfh;yo-$U}ra6(l)|k
zuTn#Il~Q8<V|F>WeHrh^IZbz$^NR7DExaJG3C;V?ij>vc;lBLa>@ZT#O7?yL-ZDsS
zm!TIa6ahJx!4d)!mr*S_D1Xga&5q+X48G4(_ytBwQlumUg8=iF0eb6TZbeT;x2HXH
z-~Uj`mj7r^>?Gby4~`vK`bd7HNJfF`p9-qCJ1F<3kGpS2Nfml0A;wUBe6LX7fPNxG
zmx5FQ$a&CLKdL_-04M<PlU@uUUazRnN$+yF_ndSF>KlxQKR<pe;D5nUVvbATBGGZ=
z4&0N1iC!-)D2F28S?_fb6biKGN<jo8Eg^GVyF|U(J{Gl~8-#m~^w1iR4dKU(okzWn
za%WN~8c4gY(JW2YLs{R<IuPbD8qa8G!i?FYs~wWc=-57L$42M{v0>s^u!?RiUNbA^
zwY|J!LBleSTh|oX6Mt=*)!b^+SP_e@JZQzCTPv%+xxm6QlC>6{0ms9=V8db>-gLew
zo}YO-={;iS=?1g&c4{q>$IkPuo*r|c_B{2rT8=f4QFWzk71amN9exgWj1R@0krhka
zx6-!oSM*xtuVt!e5DH#u*a1h95>?<>ujIv7KXtRuN8jIiCVxGHtA#9+ezR@5javKS
zC|bMOkF8!M<#aL`qXUXort;k<j@mI1EG8wE<hb_zswU;+)%p2wuQ$kQ6>21Qxrw=d
z`ph8S=&Vgi$Ojh!^@JuHf<vI5?YwWtgp0D(nT%*<lizGr=c-SwtHEJRR>E>J?87up
zzKuPDy=U;GXMc9yMkUl{KEtGUUhrA#@lRMMao}f*p)=)lxG=g1(XO6SHm0?wC3XhK
z355b93ZF;cU#8i;rD8j^O4>^a)H>uS$+tK2?e*jC-x;gki({bbuOD~$1$%T}LprMA
z9;=^!SKse`x2RpS;y{<Q9|G-$n2fxaMy3WO7eWx3kbjH1XM`(X2W>U_l#z{;QL4(c
z5SsXJd_70h9TvP;M42VQLKSB{yN^cw$;c=XV+am)M;T;-#T1H>UC`8NW-ldU&>RD5
zCB7D&)+t@(Q>9COwV6+4Eh;?v3bUX_>SW}71LsqzO;)SB95G04!uUALXajS})iguP
zEXfjkT7PT<H2Lfo@F=8~#ar_@m3l#S=n<3GYEWCmbuRrVNx}Jhg}k-^DO$NGi&_$s
zi;Q-*Je@CB(9w)Qcr`9M2e5WgW&~;<vuH?Q2kK}m|G5~V&B5@|$eoOglT8<ZT780+
z^MOEh5Vho#C$+9T8LgHx_F7w{KPNRz(`6;rE`KgvNAcuZi8ZUQEV0~Xg#Sm0B_byr
zzOclSrh<TR#}aE0cPz06d8Ndf**7Y&6m)0S^^Hq|k$Po`HE3T_V##ZUO?Yb_r&4z)
zu?BHti7iR$iV|z&u9R3a`|=WN&Aq<FT0>5kSS$bX5=*2%?1*+Pv8NA6Fh$2Giw7jK
z>3@J^%gNxbhb0)qN#ZX#EHScw!C{G!u-9S9CNJIQ-~>79b_thtb!IVRpJqNK)(2N7
zyU;Xr-5(WG`v&$-*R^g|nDRVn5#!|zo2D60{Du8m=HAz`W4~q)ckI^;@=CvEX5Xk^
zL-3B{t$)fJsaN)E2JK7wwS-RO)@|=p=zk8~+7$5Y^N^uKIJ<o+jhLYkFK?e%Q>`YT
zDWGWs&+Esu7=m^5H4R>A*-mOWOioTOGx-vtV{lRT%6?`fae7%ZK50`^?Pi<hiY7Jc
zJ{ZUVh`i97Uf;Q{HB26ZZc<-q^Nlq8!<Xx<DurPpLhI^(9Lk<*3T19&b98cLVV8Pd
zC<+2HIhSy=Bqf#@ED8ZJmjP!uCzqgSDFv4x20jIU;A8qwUe0jH83us?C$7=}4PeMF
z&{WWaEH7fEveuARh5PSwhIe<R&9$PHv=*uWj(B;8<lNyo^FeOm#vz5mtp-XJ4bwm(
z-mt)`-e^IR_cRD*?`aZhZJ|^Nn$S@#s5E~;K^yI%(L!Uq$%!y8^%kR5Fok*p%R<Tm
zQ5JuHSpu>8{0V(oU6Te)GW0@U4ut=E3w^cby#fhROpY!=OEOxN0+}{%VTtA%=F|cw
zER8;V4{<?@IWT7*OM+JQiKzt6t<O>e7p{AdOMz>#P|O>yi9uew!c1-*R}LS?hRM-~
z6<YDP5Xq^}%NecK-=dG%SA`Ne7xoou$YOneL<+f*YX`-06(5lqt#0&;S+r{mqh?jt
zQ!t_>|3)d;pmD6XioeY@2&rKW?3lP|(6S)ndz*!W6!kS=IH}~{ffv%hHfL<OCLd8K
zh-H6UFiR7+BUF=Cdkc%FhRw*nB>R2?G-Whsw?j94DdX&D`ApWQgNU#%Rx#9pgccru
zYsY|Vp~>Ht=0*pivQOHokF6ByXwq6j<&F!ys9bxWhB|&ObgVG5V>Sm^)ZYTowY0uk
z1BTg6g(a&N#Tts4ZA5tivb$NZ$C~_0fmlrdIJ5{#xsOB_sehiKmB}~AFj6|;fkfca
zTfmQ&+OPulT1T)NsG`<KC`jW3%V3&+wyqtFWT_Jfk4A2;3_cil6GBB}^L3$*Ne2Pc
zGnjPmyYGg#|NAr>ULXHD8-AFdFJ|Y9OJ`FC&ku%!+2#D|;$(I?vemnOp1nUk{(k=X
zFyXO+I@<UM<1aAMx3zSsm*?m6aiZ$aFJ9~&c4vY1=q#3vwrn(??civwPai*jyt#U}
z@TY&Ap8qEd-_I}J&o279>1g=-@YC?aA&n0Qvy;Uka2h+)Ok;wvv3*hC44^}-RgQ+2
z0n@)3ew@FZ53j;Et&&tGFw_iWVr}-97x?dSN>GihyRjAt4nq=@z{U|-RI!N?ml$+9
zKZk6j{P^WXTCDIV>*bq7(%YAR2YVr+poJPZuJAFBR|4)bDjb-V$jN=6Muo$HD%#3K
z1p-+G6agxvnyJ;MYaLzzHa+GX4FCG&-yU?LfxU`tKEFCUJGw<mAqBQ)i$dE-!i~lR
zs9Z}1u;FZe@#fR<$t-j-1b}?8m|dI)9v}bqdGX_$#qnYmXzbVrf3U`Xez!lk>#OVo
z*&w8i4-N#Dkz4!4;~BS+AH(a5`N^Bv;&6EVYJYe;`@A^1u?v5oXHIEKPurfJH*C??
zwg5`LWY#t@7QA6Bh$44OQcKaD{8upRh9x&_{1~&=KbqC^h5LKGJnAxfy|(mvo|in@
z$BEU*h@Vvnz6~)-Qe;Gb%W43y(EZ5uaDu~|^u!EACj?I5F*al;SSK1ywGfrULIR@|
zMMfs6-ekDvb9Xa*Q)&sYiocDX4Zw(|0Bc(SEK}a#t*0!v^7gB{u6uuVn|w9t_FsKX
zE9Nz=JQ*KpMP=2fJKz_PY}i|2ieL=#Ucs}c#1YBalwPP-P*@0mxFuW_8Dc5B$RQ9Z
z*^_ewbI=UwdbH#zdP6S36D>+<NRh2(58tSvN|GHYQLs4DUZmU+u3aeO(d!C$bh@7*
zI$&)Zif>3QgE-K5x0F=mc);1+lDWZm)v_bL-0%fk`j7jPho|J>2azYsHF-EMIm0U!
z@rp&f0~R60LI&=CwLjQ|0xWyRz+UoOY<U8nlMlzAVC_h|;0~AQ1txM&FN=g2#FfxW
zM2Y`2%vr@~P^Ee)xD3&j@rRzsg%JD}^Z+cPpeDt_0KHeDk?nqFa2m1wd6sxnV9iMI
z%*)ohB@qIl)fR#TIFYtzlcmcv+UNEV>Poq$Ze9^J*HvSGm0)Ju0d<oh@C5;A3tR!4
z*N=uc2-B0^UcTnI1LhEetspJjgdfE)fKgyXK^=!Q^J&<NosX%Xpd|!fY`$7dei~;9
zjG9)^WW?62fNdr%TS2b7J1wZf1c3ZX3&7v&`KcDQx)ZUMvewkHwze66C2bRbH>8$4
zP6aN6w~$1Ctq&J2&xW3bC-f}oA@nSIO;1@KZTWHM<H^V)0SXdA_(MNis_aMM07rW&
zIw1Y4Z~zr)lb^t=##a2u7o4<q|3t9R9kJww!1}}iam(G@7Lo}Sspc&tvma7zCFxbI
zkUbk*SDE0t@<VW4uTL@jkbT5;SsvN?5&N=1k4HLxAwyJyw?PC**aJU8bRXq~7c$7F
z!TK`xk;guG{#EQFkA1a0P3%K3iGF{!ihYFG$0Sc1`w;s4{%93+5c$fNGS@$ge9ons
zw~09$Qp=cAAT6?mBq2J%9X=aw>Xe)MAa3f7m{V7j)D<Ol7nEFr0Z9)~kk@FCrYN=L
zeksU*!|S=^WrQa{6qQg);;$hG2KXZK;xELXqS;KS7HJ4x%DbrO{JhgIlb1@YO1j&W
zSbK&RHw!7vcLYD3LcG8+c;vlXl93;jR<@9&f?|}P&wYnYt|Ps)d;o3iMkS>y+So2G
zA7NR)Sw-#dNXg54x6*3NU#Oe5bKhTH76uM~v_htzQ~K5Xa@|7khwD}MHa=o_xt_ek
z6YBi);3xn0&4<O})8)6r@ayT~!_~WZGXFUIcJ=YoU*FIEJN$gLIDU6FdwGF(A7_ix
zlgs$q8I;|-Mws1oL-&7oBt70hZvv`5<RE^$)Bef_@w+-pY-RG*%ZvVd(x}~tK5Z3Q
zbjBYDj6W`TJfCtyoq+_<FOmHWfW|1~7uuHj_A|t`gER<h3n2~39swC{-eQHf5Xq>&
zrstQn7+kOV&ts6bNyeDnDM6K^|4eA51S$LN>()yA4*eT$ESHhVDHWG8`6?2Z5O68P
ze~0wB|MT;=oz4xGt#jTg*6Lelj(af3Ux-aoNKNwm%d$bs%_8U<nSR+bm$+1SsCUTq
zxIs1yb-C&G*ws&;1nF+BIlMOcu9Qo_rQbTK56YD9axHq#&xgwvwKiMDY|ZAU&bcPH
z6i%HpRCLK7>Eny|_3^p;r|=S4uaIy5fAwRRZde1#1Z)6Ek@L;ZzngEn-=%E!xK^5<
z>Hd&zetvI6$=94lQF9t;=ZGe01e$ZpHjYrC#y(K?l;Sh`G`%gtB-F3x0JDl#b3&3-
zZ2iqqw>b{>oPA|eJNrLv^Cu-=FOteR1rDV$>iHDp)N>wWZBw*^3plSdza@O+e>UBB
z>0N&stuW97{$#ItmIu>7GSpg~lzd&_^Ql!d_wx3N=Yon@s3kAaocGTA7pK!;L9HLN
z56Lv+pZ0AgC6{sSeFjk(bVJ4*gnyoo3PsO36*i8Fu=-?FgxZ3r9G{8`1*fcyLsX9S
zqv9;vhOMJAwc17na}-Y5k1o#}f5&0YnT<>^3Vf7`_al?WqBaW-eJNh#{R1N-TBjmU
z3$g{L+4dswyrK)g_mO?XE-PsExwQK+Z`Y$D#KP{1_YE9nR#<2^L`5^N`&_pmqsWib
zTMFT+bz%-KH)afI{&rC&yV2a-{(7d+L}d}Rn`1u$Ob_HHj9|wXlMR39f0SgdM;k^N
z+rd@;Eum72VsHbWx|++n#ywiK5ftDsn3xPsBh0m$hq0?Eq`)Y|3BXq{eRPV_vBQ_P
zxx-rkX?sP7zxsBOM;;%f1<KvKRi!<)gYf{fEozRxE~5lPlwfg`U}2Pc_1F!vt<Ghw
zY-cUX^30?VjYL@w%5&G-e}ZJKiwx7YBFPHHiX<zP?~tshy%EV;UYr>3HlCe@)hC%r
zs4XGc$LAyj#SxOW;9SJ@Ays1)8Z1=p{7h$pvcVVcTR9_;l%=gHT3D^XYeI1?ueJLW
z>6!Pek^A5iaACnJ8;6Nx-a-X{Fy5#T{47EEksxdiaOgT;Vw_9`f8%VuoD+Hqxj6$Z
zq6x!Gd!9LGXrdZBaEfRn4c&i9zHrei-a`vy#B*Sa7EIwG!^N<JmG`;YdvFOHL$Eps
zaH1~jV6Ou`jQF}>0qej5jZ^3gumH_-Bd><HAPbZBLl7#Gg;1<W7DD+BS%}&jk%eJa
zG~BzkokxV#C&@yne=Xf3tLjG#j{CyadyT?sh51-0E}W0^Vn{#US9)3!c7f*E%A)cX
zWbHMiNWha&@wy-DA(5i;R5<V&B<4*UX@5Pgpl6Umf8oFzJ0GG|RXRZ0i_x*-z$_H&
zO!$3D@8NAGtDkT`XC~1~@2|l|)S#jXv)C90Y)nUn{m9x9e{8>9;U~cO>kx9-?kYFG
zVzSeZSjT~OmOv*29wxQ__0TB{6|P*kH^qYuI4IO60BP@qYB-d=I1sIbgcJ3--LEoC
z{T<ic?X9<cJ!X_WUhI>-4;Wkcus;MI+KH3>v?YhWD}k0rVT1O+<XND=NL};HS14BG
zS)qJ~XGQIee|Q#niDJlZ%wfXnlRPWbmhr6lSiM^&>=o0~BZ{Ov7EyrH6N*tq9}2}Q
ze3*CYWESZ9d6ssE%+@=y!fFNH7mAzlzOZ|T_eJfs`Q9<VG;MDl(8@sT6=u7YfmDL^
zog%LuZaeWI#e6I*)8X1-Zg=!sDJ<R5@WQY}qkOCpe-@*$-8H!V$`2`~u>lGvf*G)U
zwKArI?jm)s{BMClrW;c!IwlRJyv|tGVGR#d6ipePrxsVBjk11rZJXYyTm2*(-p-x;
zU)Gsfjg8@vPtcfH6<5mKd5CH*l!ah?9Pn&ZU}ej$@f~^PI8M{b^;YNI%3N1noqKlF
z{~~*43|A83lUFML0<(tjv6oSUDHNBHojMZ&F*KJE_yi}HrFJQt0XVlpcPU;0f4^Cx
zdafxnQCVX+bxf353aEU95x&p(kxL~zpDL*xs3%^~bmEF>Q`Joax3B3*e<#sGT_B|X
zU5d?A#Zga}5useS1DaC?<UkG4jA8WUR1X3dTAq1s0SnL7-H$#FYeUSrMqy`uSlrCj
zK@Ib9WpKP!EgEUmC&NkM`1Fa6f2Iwoke{@(aqON!-|KQlT4mI_dfAt8%=3ktzW?r$
zxr6DsYn4|w1$SyCYl-udms$z9JUShA5le(op|u+(F@eaWcI^7h?0L0V639X<$y6tL
zW93JeOgAnH9NwIB{2hEUYo6h`2tzW|iM-1@9}74iTQxWokCe5F<u=04f7Q|)m<uYJ
z3au?UQ|$Grp0?-FMo#}a?#!dGYH(3zm@c2Fz(<KV&t;0!Qe5@@YsDa^)8M7uYw|CP
zP|=%SSIACC8xlF(_w~vz>sUDudsD|s`|I_G4OXQf>G|trdAZ`j<tyRTCh_kKN2J*>
zFAK!=Wqm4&k8OEAKgKwRe~|NdnXuc#3g{d`mk%vqxQ4f`r2~&vnU(>RMf8@Fpzj<!
zbv{#u5OO82K&~rgUU37VX*F*TElmo|mqBP=(s}CC^B^|cF~A){bBUb95EQ^rAekc<
zpGUf@aw_YnQ|Zw=24P`LrA#-dv$DPERl(yTHf{1b6Nu5RNafe~e={sc0;@J*fnG9F
zD?_h~(iCuQ;JqxTt*0zzK?|Z+4(hc+)Og@z2ftH@q9bMFmb+<zxS<dg$TLDz^xjE`
zqD4gkb{0^=>Vra5p#2shiiWF_!@YSNTirm23dG%nsIWUDL`CmzK36LWmI}&iJ1OF}
z7Ho#Afe&uyI1yGie~1%-xG+v^o~!k7BJ6@*zjMg3BoVJgo>P$vn&|yt<QdO}b=X3u
zC3GYDre)=A^=9liLzHu3Uk)qF@mNv7LfWUkry|GNp8vynqK^v5NniUdh0j}zg<FwY
z{RmaROunT+XTOD0Yo!xthGRygbTUqnal1$-15q(a8%igEf4HG^638>sN%Y=HIzeoA
z5c_8GvC#*G(-`cx=qB)_44A!H99!L>9ty<Wbd#_<qnkwU2dfu2%XuDm(M`Z3K%@<m
zmp~l$+e@p>ivKfkkVY>p*Z=02jc@s%2Z@MzEU)>zi}GietLk(q9L`kJ>1f6IjhMch
zi4avisgH=cN|=78tr2N8B5Fp`Sds8HNm<%L1YTG&s2*HkNhNr<P#$iu$aGYe;+BhS
zjJTboOMO~dSI)E+RIj^N_{HQ+>N4@p!tXzu+5KjhQG+QIm*N~E69F-o(a<y~m$ZB-
z5>v?7Gd7T$2fhV41V-!}>>x8a<oENc7cNaIlr4@X0w{^a>e{<&VJDtNC+>FF54#Vy
z+6XU$C9lN8lOR0k@J$_n&I(T?ohI?{Dt@`@I^}R(<ChtIDII^5ahZnphdn=2H~1e5
z4F?+0TYXh1fDgsPuMa=(x+3t0TN^}_Mybv%N$b3kpk1XMWMOLH*mWHXgdYo`3?3N>
zU?70K0dvC&7%Z|>?-Cw_HxS-HNCV+*S!W}|`gN{g2hxx{-6!gS>G7(h%9_EQRWfOp
zRnWZJB|gazQrUkHl<1_v6|~jj8%#5Yeqmz;TI8fGjMOs9Op0>q6vZ>%pmsK-i3fIU
z(tOzhgEN;cp^TNjGcy4_GvYAwd^hvqf!tN`W5VeDNGJO-uXCAC(t>?ymNZELo;?yB
zE5tu{&E4mR-Pc{~U`twh<;3Ibjvp9JQp0zooU{Oc`(1xL?e1H$#bllAuqJ=mF1vmL
z`BFA@>|oi>Icq1LCq##ZWmS|m_M4h*V%a9QMN4R`sPR>br8z~@;drnu#ff(k<zg$t
za@MaT{saQ?d&<s9a;l}Qx`|vpHa5QV)MD$I902g<gqHNV$h2<KReQp<&g(bW6ISye
zPg)Q&N?d;@kS{0Le9Ben63HC*cHo?%{&qM)W|j*wsu?x5l3lXi?P2da-Wzs%JKBJq
z&yOx@$A)XN<N9f_Xv?4_`Pfoiw<X7EqD=|<*1e*1kS6$+FWE&Yl~EFzxy8=@EWD9H
zhOfkX0vYeq>O1T}AN`0obW@3L<FO6m1eqbg&f9;TUk%0Wk|E>dgt(AGPTXYg2#AOr
z6K94Zw;T{LNQl~E_rb7%!!#$dEfY>d@ni?Z%$Yqf!BJ)NC66SB&75+^-cW2UHFCqT
zy|I+^QAM;hgfT%XvRp%a6+pZb2^$1-3|9Oj5?G?rgIi%RBJY^2#)n1TefaD7>8Hov
za72HK=f4>hg$RmBib1@VTP0E8TP0q~Jtx=@9FEn9Qr!rHdL<|z)P+6ZQS6NJID%1>
z4$v2dqIM6c-bag5`QsU}G8loX3`}u=7Je^dU@ukAA}dkF;5(*L*77aZ*HwJW*Z97o
zV$~mF_at`r#j4gRpX%%`6OIm_C99G&<k5dSjVxP1DE81pk*QXR9y$96bgb`yDqV0!
zL&Jb#nGC19!mZc2LPmO55NK5j71WE?z>nCW(G`$BNNp;?w3bQ{hbR^6W(yDhA79J5
zXRNv#{bo9QQd%18#2#h1G&+^csk%iIxY*ZZ^om{cdC9A3>#MZiDyO9#)(`>9doh0!
zDeJZ$l)FFKAn2qm4uf@rwCD545H?EhVsD5etz^}qVAo=p53}o%f$Tx$2B&X(*i5c(
zuc$U#sd@TX%F3G_;wR4+nlV3V#%#~Ku9>vUi!Iiqb&LHv)5A^4A|`WAMXwpobmPX-
z7Onf{k>?Xe@vK4j$$o(cm1iH{)<}PbtD&H?ZjhkLO4Dq&6o7-PmjZAIrW}BOiYZ!2
z6#G%8s9Wbh<|l@5nwZ7$$<2x5lZmI_I6~>mE}9HI)<z~eI-J|_`;SZ-M(?c>kEppi
z7a=b->h@NOH_k=oTk-Ueo=g7zIf#NyE$8-T4uZHhsFf};2hr$RR5Tlwr44_@4d);Y
z<&`-|gMFbnh(-gUy}NM#-f%s04$@HjA#)HEy~qLK(srD3-Cz#VP`ub2q~W_V2Wha+
zoP%h>`4!_u<{&D1Zlg9lzZ#0~&uB~D&U29ST4@!WbV`p8w)75(vLIZ`rFTg*{HIU(
z;`dNij})lE_Nms5{hNzA4$yz|OkE-s=Wp>V>FxdM44OK-rC+BMx2s&(&v{(3y0Ozc
z!|UUI+V~Jn+dKFzjX0(FKinLsp@IsN&9P$i@4Y$92?MUR`ktbd&JCrH$%dMBbu{h&
zdLJ<EO^(Ll6`v_o`jTveF2Qs66le0jJ&_)50aTEHwH)%RPL8C$$k%@b>S|XXmb>s!
zY<FEJ03*4W;>Vui>3#3YjP<<@gO8Frn)M4ED&8(!q+cMf)1Z92Bc-9ZVZYE&Ug;Ma
z><je^WC;Z-UAWF|xSrWBG}L}bzd$NVMVEg6+;H8XUuY;^tY2vOuJj8H_L=<xQ9zP0
z7wH$GlUiT8J)bh&AW9~t1WybS)4xlt=H2v-;nM5Ev%(@+sr@N$(-+fS^NV!)FE~I1
zWS3EcDHWIT<qZ=7G?$P$7bur-#5n|)rh_S$e**G>r|rMq|FMxI@TZ*(Qxrxi>DHum
z-Uy^!rR}oNca2=7lu0NQ!XQ(Nf@qTjH>Q5Gu{<*_J;w_bu&6+KH9tdAW{2Mc&1*E$
zOoKDc>zsmlsZ_H%$Bgc*5?MnsdZG|OKPNdLQz2t#nZYN{lH}QSjv!Ipc$!#5FsrJE
zf2Wc@UyIspJd4uaN^jV$f|1vb3@odNY1F=q?_`RuT$yT8?rBB2)dd-4HgR@$P}jl_
zUW;dm=#q1#kI(Gx?S1p-hOdo@N;m`4*B_hoz-SOUs0mCE#^~v<|4iRDUpQHEUnwNM
zQ~p^w{q-@ikzXk$qeS<R&RIKi50L0He`a!8VhGsk9!b->^hwKPRRt_=0<2Or5~JI4
zirJZx9X;7_N47Fa&iXHlKN~^u8)fGNIK?R|Ig_i+#<+DJ_-wt90|ziG$mHB_B-3$|
z9<`?oIMbhDPqdmx@}MV3j562R$gd~Ze9l$ro?KG0shoQ(?Xg@CQ7xz?E7>*cf9<vp
zPUbkp8T#$)byW(4agp-_!|ZboMJv2otU`^?izr4BeOOSms}@w+2&L7+3*^oH%mycH
zh<Elg#x?udY1`Do*Mb({Bd2)W*R<^Uy^N~I%e>Dhfkg$E;2Q6<Ak@N0@@?H!Y!u!u
z3~*^ExghX5qD-6Zm9R=JoJeMqf1b@dSb^7qx4qnuQ41dzPA*?pwb4=uE%l8osZ)S~
zI-*UVZBw+YS+WE@@wa;F|1$sC$hg_n0QVL{uCp&CV~vMW4bdl*(v#w2Pr+&5N~LXL
zxl)y!OjPPEDRKHTPo5*T6(G@_UD<tSi%=?NBdd%U5)`+KXuE);MQ6bUf9nfHv>R`m
zdC&RWAOmBy=Ho3ZCr&Y*Un!X=z$|3jS1DO992%9)KBek4s(6tl|C`OBqlJ9l!sDKF
zenng;{M7F67Z@^%;Qbu~!$d6SAe<Q(YiA8Syey6`%I7l*9$o&NaN_4F&qe||;$HB-
zDGy_9fqzKOJA+99A6ge&e-eN5>5q?ZzkK~SRvy#Gf74Vv|KRQkF-$+`R^x&z>!<H@
z@9-ciBg!Bmp1IA0$0R^c_3gzr@O9cG0YTd^jDAm}c;XLu<{w_pr!Jo+RtCdRl_62=
zqr~s~IM}3WEV9r|jC7Z%(7Wjt->+|_+jNchbt=B~+qC&MZN89If30)5atlFW+DK(w
zz@T5TD%RKFHH?<6L&^zsdoa~1(WAHT86EZQQw1(1dVo<uc=Xp)MH08Ba~&D<u0u$x
zQm8{sG&a7+4op`Z>4{BPTt+O$eQ{AB>lO?5?_Yn=dms4L=i#%^c>tw_v2J=WP6<HI
zMOyh#ZNn<&LeuN)e|q$+iBacRt6bKiixEI}U#3BvqSr%L?90i<01y__=W%23U1BhT
z#)J}mGr}I=sp>$pM<YD<^8l_Xx^LRSk|XPy)E1*-V$+yH@w~_>?}MvOMA06UPPx~-
z(N<<C_d7Z?r_Vh#JPNW<H1928zQ?o!_%3miD==4<<)!bEf7aERSsrM2VGE9Oczz-Y
z{@Rf)Dh@DK(JaGaiVPIz@EGZCnq1dne<bEh-G0kxX7WU;*?mSxGqSlzF__m+Yk|^T
zaxGB0)0aT`eVR|B&Fo>=%}&6v4VEIZ!`LqbUj{o=JK5$5%MEF@jOzep1r5{T)kuBF
z=EZ@@ddD!=f5QG;%r%R*aF9Tyk!v%a$US$(iD{J9&V<43?Y780Hwja3blDXd!m^s2
zo6NX|+!j<9B~g#vxK{T!jc15(k;)g_68DN^41Qa*Z&n3L-g95TYDc8qa0q_Na)dW)
zy*0Xo7_6C0HAp*lHJZ{SI)zP4Z>DC5QF`b*T#-_he*{z`a5eO?B`ZeXyJW>>%d+i3
zAhf{L!}(?#G(g@DiDY<E@}Jz-MDUNirH~dRk36;K&GOk*P4B<hS!1E(Uh2w}xlcn@
ze;Y@$6nXjE=us<%6CEG+IA6fpPc@K01HDB_S4BIzq$wFm#caRIA+!;awq`DR+w}zA
zNArFWe{0zG;2QR!^7NDqW8l*@t~^u7>)d5q>S3#YxY@jgW-a>yU23#MqZvoZbAw6s
z6(uPL@bQ%NO0)(q>bB0K&nMAYU3-VMhYRN%+R2kJdzj=cPg~pCbj_Q6m3OR7!Wad9
zT#FY<lO<YqOdEfw(brpezOGS?Q?+R0l}=}de@_<MvIZqIXl{`X6d&x9YLzMo{_;HV
z{bOyGl9P{0u|6ZY*vol6#5Q|y-XXP^4rb8kJb@i@r*L_H5)WM&PU8JZizi&E?$V#s
zcqkA3D0KsrjZwT|f09u?(w}6r@6?~v%HTSMJNJzk*DL#zjM`7{PcnWt=udKr*Y_tm
zf5pT7N$v%=?@wx}v9N=C``Mgp@-xHW+;&26i!FB9qr_98Ug&ayzYRG*0lb$q)i66m
zOsI_aRozKE6w9SHd!P+x@QkbZ>`Nnw8jgpx>9d{i-Y<OUx`vvHO30`#2H>xc3FXn5
zy}5HtFs?U>2}bqIn6OvH1mk^8qnToYe{1u4FecdSD`SFdb2uir4_z4(Sjj1{2U<kt
zr}R#h_Cm#b|4NE+y}_?n7{$wfy>gp9gMI|Uqyl;HM<8rAkJD>COgW)-_}jH>F)D=i
z@s90UC5<rsXAH82BOD9IwbR!z2pces$hh}M!DEAf!9|4Mp9t?zBcsD0`OZ@#f5ShH
zYsU~=$6>GzK2;pL!g`%9>Y(Seb3U;HdQHGYI-w5`5;VC6uy7w#)&Zpv2Ad=3mEPJp
zX9@AmEE?P0fF=)co2fC=Mi2gGy5*9E6WEaYw9UsCAm#}2R*w4QwpwXDH98~kEr+I8
z7*Uob;M$OWsL*8Z5j8&zGl+JEe~n}*yWB9tO$IQxf{vZW2sT#f%5#d8=PkA5W;oHx
zkNRsL>s<T=>kT}qarh-Dfrrw2fG?mid0TTs@P$#lA^5^59|2$3>^p%k0GQ&nyK~vY
zxLygqFls*?d|~`<0KRaF*Ml#d;$iTGd%^9&7d+`&xHrZ(XL_S!=A7tVE|Hfx>l*+s
zOqavB%Y5>cN6dq&@DnC;p6km0kbZ!1c@@37x3N#$IsXqfTUf)FQG+QJm*HzC69O<c
zmx1LoD3_{@DGGm8MTwLwpa{^k$$;Hf<TmUyY&-3+z5ij<uPxba;-m|<gB+WZ`SXb%
z%}9wKLWvKX-Enh&FescH2%Hwj?*e#2@f8#cEeQu8tpRcTDLy|ar7e#g4^+w+9**w^
z{-y1mfq_>5&j3pSws#JJB}Vfa;_;l;1|7z2@4}pZ;(JM-by(Kn9hrx$VHd{s%kkqz
zWrvmYHr$4v{%L>Q{Mz*HBfHPeimyL6VFC?EGiaghAiW3i`<wW_`P8ee#uUGkE~Lco
zAEJ}Imwk>YFMs|pz@k;;wT61Dn)eFXDlw+S_eA;GrV%_L-Y053sagjo(ad5|L7Pf%
zqDvY>T<WmBvnuRwoS_soP@IYSB?b)3B}UjtUPjv*E$@}>Z3A_ft<Xrs+c%Hqc`<fv
zQ7T5hTOE)z`}NT{ElKNHG`MWrS?Qdc<q$iop0aH+Vt>3u+l^#myJjQ}(lDYUlDu2G
z#*{dWs>Am*8BQ|6Scz`9H4|E^@5O{x>q;irGNZ4^q>{z0C$D7^dFjZ{_CT$-%e9lo
z41xXI8QiYEH0FUMk6FyLY<uu>A?>atk7n4xw}Dj3<jrO<dP>pXp_DR`4c(ejt<^xO
zX?n-x4}WxWnNB{)UZInb;C!#~G@aOF(hXB81*99IVMDa}$7pHBb~&OcZO6DvjbxG|
zMswF;X2IC%bO>cP)|@avodsTTG*_(p#vxa)KDE{-HhC>An#w7gd@Il=EBP0Ke>E!K
zdL650@Nu8|MD%IsuIh#&2+vpeO%jwBG;oJH;D5P<?CRr17M+q#)5`DX1$IKsw)P(>
zHDoxsaoyMZG<8mun$BwCQlV_jq{5R@5Hg1%hhHlrPt#HoBvD&fRnZLbq#TddGIJ9X
z<M-U=Hx3Dh>WA)&Qk=nz)P5=@U7ze^-9CG|8!Z&?Bd=`JS5(htJ<FQXRq@tV4uhZL
zT`lp;JyLqEUX;kkO7)mAgdWE(>VjcY=Md`}$gw@1Q3H!Nj}uI&sh`b0Hj<~8)sQJ&
ze~<58_>g-PMR?;LtF!tJJyxgjHa*s*os_bY$jyANoCvK{pn0%&ehUSEw}nFfbFciQ
zG5AkfD73(-X>Zs<G4Py|?&e1jsQJ$;`CD_jwHl8g&Zy6vA8#E&bXwOQL0oVUr=>&a
z?E>4->gvO|UTgh&ymBa)4k3CBF8^PT8sMb4u+>h|!Kv}v_tvp>k#@iP<K_=RzPjL-
zQG+QIm0>6rm(^83Cbt`sDKiNSbcWINDU<v^?-Vlwmr;W$6ah7t;lL#mmxG%rIe)+~
z(C)Sa<d&dsK~BNUoaT`E|3j#)s-!Bf?e0Sk<+8Pq6v@X$*29lM55H~p@0;hNp}{H7
z;Ba{VK7dSU`2;P4mInu*tbyVE=kV87Yqaw0<sq$&k;ln*FK^yG8St_KWCmmeVE5z*
z${vIAKwQ6bcANL(YaYw{<*`B|AQ@w%B6LCK!R-Ei|9zv2z*+?xzs6Vp>+rspRGcY8
zJe$@mC#Iu3azculfaa9XM-QtMNsBGGr^sRtbI24FMcL{{61f=>ktb?F0WIW%w?>w6
zH5+Au?^o;N?U&P>DKLM#C#$0$n^0ESI4Y0$G|N&xWE@1RvYAmiWFOB|giFYNEjlX{
z*Jg!ydOADYc2ZWL!(hoNGH$fhKBIHupMJf5oR6`JoNxH>MLw6obrD%Wg5pO4t0DMN
zTs6di&%>%&{1H^OT;I-vs#V|zFr|&M8b`MFe~T#&s4STqef@t9K3?*!r!#!`F?^g>
z4*3ipyTWHJT|OV#4_<L_)~r<$Vn*38dK(^#wXs;Z*o|HiuS<0a11cUvn~APrql=Wy
z3Fphr$V!`TZ*o7%E=c>CAg;LF{=7h<K*3#0GrMvuiQG0j?{w5!D5YP;{dp{(;?5+6
zuOp{Hl+Rq9`qqEaNOGVATem2UGzpabmy|~2g9<m5MiYnhZDL$Unn0YRde0@)FmqjQ
zV5h8Lr>Al#mMGA;LavkAaH4Ss3l7N-uanWoM8DMMMbFNFPw#w=#i^ZFVJem)P-A$e
zS&&WM4HG4Em}@7dl_`LHL`&70)(eIk^yV|&912frvwDB0=s<U6&11`_!iOXq<0od6
z!p2C%w{Ye-{1`q+3N?OW<|=%^DC{yZUCdl}@X@I7Y$kqGhBQ`k2u>ODrMS0LONJK7
zA&smgtGd?K-SPWXIn@v;F!$xuc5jS2ye|S=U@qPtV^>R{O`tJc6PNST7#dHf%tk9a
zu-l@U{4IYU>It?FNTQN0(S<&VqRa;Vk{oIf6azi53+gF0wp?V4dmv@x&D<27$c0mJ
zTY!9Nl+5T$Gk`Pry*5SvExT>m{RqfnET6-ejiMZ>eZk_Wu4LaLTXNOg3SDHBRu5VZ
zJB01kTxK9@>rGd!J3=X5%gJ1ZN!97=NTKi@!%BaqltBmy>1V0Bmg-%Ai-n1kp0xv<
zm#!>)3EmKca{iIG4jsj0!Iz>XQ}K4kg+T7{JL>SZ(gus_iW;O7OjMhZD{9w{%d&O9
zP9uR;kd)q08zas2r$>p)#0$Aluil5CG;iW7J==W|eg>;4_|3ylY$+1p_{7TlLnVS&
z9^HSsb5#CG(wA8=?58rFY+Qx4@u@53VFEK)^QuJswPw{i3N+5^DFDQ*n}U%0l>0r9
z4mIfTK&W*TJ8j+!vC<@UQ~^$0y5$?XZuJAwDR_Ym&o8t?y!f)?C7UisbhGYlfA1w5
ze*Ps}j4C0(#m?&q1m}yI98WB;D+@j>sU?4@e(_>cbw*OYBq=|rR)N@<Ihtbh4$zlm
z9join{MU60pjr8;J?`3tm^i#7X2Fa+-c*GyMyE#nRBO>Oihn0E&h8&*R_*vH2*(<*
z)?yLl<61tElh%)m;FKZgrRZ+><}!tu9B!paIoD00V@1v_&P;_~TRokU*=SHUdS`!2
zWqxZjlbved@nS?Vi(xBCFxCLl#b*?(&CJkXP<lAVg-eKsSzh(9Y@Qg=>9`s#jn(NE
zp+7=Zv@Riwraaae_LPcATWN3nLfA9^M8Gi@PVwLB#IgRj;noAF)o0DaVjgW5F4YIl
z@|H6AKd?Vy2MT3wWOH<KWnpa!Wp0;0d`1NVH8hu@k{}e9zF{sXm+;Ut1b^c;5WV+T
z=m(m^VzIaY4Fhet9U!+1`WECAWac!7%>N%!TqRMmWn1dOB*;NtD5_WsS@j+%*?sL~
z_si4q{pr&gb!UYmu&R6i(gEL){0Wj94Rsbk7=-TqTlf2ml*;gR<V#D5$d~2cj=$;f
zjNrHj@Ezb5fWx!Zke@-t7k`BQ2Wwu^^Qup6>3O~k%kYX+kiv4I57G{Qy#M+nv%*q%
z6MlxT_TSt4)89|6`s9p2W4b@TJ%t+@L7-EeM`5(@zW>#IdHSuDTE~+7NctezegE1u
zEuXUky<`VUV~8dW1j%02oiScuxRo5JPf`*atj~P0fjT$>CPscbJAYV|s3nJjT8U(n
z6ir#GOVWf6dX|r<{-lM~uab=sU{otv$%$UIx7ET}nDyo`6Ni~#q*I^%YqpqJKQedg
zroajXGphj>byEt2Hh$(K%DTNS`AbU4Q=0gKx5KlMVbbbjx)XcaV!_?oqf`PV-so8j
za|UK{z_<Iy1qu^r=6|(Qvz*FYj=zQ}URov+S(2JZy<fIOp&&bpnmpfc2`V9-Ua@58
z&FdZAOv}?_V{0!!s$0h<<X$tMlh3p0oL7FhSRI)?CkVUYBkD708L!ilQ^*PAWL5Ox
z5qOa$$$3Uo{-0K3nbqiXFO$|mToYk!yrone^ej(?Ss&7S(0`e5B>63^7dWL9&VJ3(
zW6P<usNAw-zwF1oE{&r4A)7=S?Rd1yclVXau9{r4HFQt>i02b$CvAmBTUW&~uA`!Z
ztlraOi`1$9y3k|p>G=~ea+9axzkZmHTkPW1r>$GHi$2IM`Y5{;2?D=Eyl97b(SCM;
z7G!n^7wr(P9e>S|4DUE?Zx5H*AzWrRyT~hst++$H%ntFoWY-dwUSk<VhM{e>OwmxL
zVo3_g+a%2r%ZR|zhJW@Y-7CpHUZt$r2s!gJ5%?WP0GS_Qs31epc80%==Sh<yv2<q)
zN_mE9t6RO9-GHH&Ilc@E0wnaW&&$Xa8Rsf&Me0#SDt{<imw=gEP0!U4eKvk@RB<6x
zH(+itW9JMGEv_O}@9C^VUt4^3I1E@qm$9L0^O_yrXFZN%iLoVH5Q8=eme!N7P5I4O
zuc-x)Ww~6#bFIFYHZr(k*iJ2N1nq?81^6R)BcW#8>+j%=_n`U0CEHQ@g#V_C=r(q0
z89jq(C4Z_My%I&`^Bf)_&A2)W$60sYs^>reyg)<039zsu<8CD0X4CyF+#!$R+tDsC
zU2p+I>PK{oY<GzTKh8)<0<&~AW+Xn~XH(15s$QGRBPS)HXh_DD^Jg7A`C8aQCNqG!
zKJw2Ch45jWn6-j3owi;mud_}Q4O*sgcm^&jmv8B1R%3tFd4uOt^t53<iBz?;^ULFM
z`}%c&m(?y08=x5Jf$KC`fp#FodK_}(C6pU;pu^AX(Mx=2yGok55LAu_w4@SUREa)9
zB_yG#(zCiaFL{Zsd5QK>NihSd3fUDO(KR3Kq!6GWJtSYZ;v;j-M`j<Dq+&oi%B?G2
zGS|FhZbpCXJW;&RN=dVX*wOF<q=yw(cKvghc$kv)%Pv&LdSUCIGAayh@jsPna6lPx
zY*cZLHXwS=vIA|%hXq^(9Q%vwD_;(apJ%O~YvrGhz%|7=Hyh!mJ?`X&b=wc{(wbjQ
zUt>-Rjj%9NFbjT0v>E&p9aP-VcB29uxA#h~3e$gG2JhIgr6P@sW{?z(jZzK?<{zu1
z0_W{S3q97N_8seG26sm3uOrBo{S!%!4s$p_mIUytu8_kr*rWnUS+=DTm<VvQ4Kh!-
z<>^3Di_p|~JI?x3^xC!QO2Fln*Gl^|`&v89Wqrg<XVY9+0mikIa25ktjKKvO7=d4<
zQq_N7U>6Brl-6h9;RaB8FR@aLi^yY3tej$s{X=LKawN@Yy=*53jYN*b-M~K_Gw7La
zmyX{U>ln^nZSM~{ev?vwx9w=qQcq*~QNRu=^_;X!Bx*(>Fw-A%9=K}5^41AMEzgss
zkr=8UzQw$Cx7CJ8EZ|B+?viME;wv@tquzfYkt=|mN9OdnM8~nhS!CdX*BuGIW?rvY
zL!TZ#F1>9fl;hH-_lD!r>%MG{LR&5MrUT9!$Ib_w?bwtK730BTJ<gIDYK&tt-L-vy
zEUNJGgc*<@<-<oiY#&WvPd}{bM6M@W0(XJKT!Y>oJ>El;CCH2-RC?M(uNADEQEY#{
zj4V$_oYWUIQ9U}T#;FqLef3Et-ga=v2*5F@P+Tajq{(K_RiD?R+38sa4O_rEqjpq7
zp2d7LT=_%4U2_Y<YV&~R6aEtNc|2y;?J;v&0h_l%xQq(x+;~l==++lQnvJZ{+R|e#
z8n?fw>Me(-;#}_2JpY)vywfqYVqAaX)a2N6m`C-%KkZFiVZk{#I~;&1bUc|R`p_-<
z&`tVWq`{6Mt2Ma9K*9U*zGnzZg_35J9A0nYv&^8mpiLLaO$joSfN_#E70j<e|5oQj
z4j+Q|44Y8t*|ABeRnq+sd=KP=vT8?s-$<n6LkOc_{6^Y^Di7gRBJbMX51@YwFd)3N
zw~m`%1Z;YTNmg>}134A66$n$-{9i!~`X@&W5b)x#Pxo&?3}x0R;x$0qEAjfPacb?=
zxSzE`12C6Hb5n)hJXU<?;kN3#a4fy)NSBEzk>4??PX%tw=~H11rEgREwDX3H-i&%Z
zp*IeEJeOPc-~Wkxb0S|vK+Op^BdmV_VYT_Mmr;W$6ahDvLGUaSm*J@?zXD)Sx1y>k
zBm{qBp08KA^|WgoBAkDfD?ex;i(K!JX}jgd_M>BQRR^2{=PxzK29KHAA~Gg=+7QU@
z5$I8UHw5mxO_=*juCCqG@#IRj>A0<*lI9#1<*mlUcGdk{c_~MDTld{<UP9b1Ypjan
ztTJ>#v6~0GY4Tm#fsecJ$yid<&ka^6wg!Kz`#)?y=G2R9VQe)<E{W^&hbA-}B7tkT
zO&fk_H)E}$3nI4;6+nZ6+y0Y#NXwF{A4dDDCfnOBL0?;STvjZ;S*b~t)KPmtl>`_S
zMw(YiQ%ZmvQ^NXxboe>{(E$WyjNLUDZV-4eUq2uvB&Rpt?V27p0@+;xEvJO7KlMkI
zpGq|28G_=mLBO>nv4Dhr1qt01NZ8h`g1&V)+>KaTbS^d$D*KWOa#M|nofVQ}16-Bc
z3>swvb%eTZr`{O|{|B$6y=j+GgDDgNIG0hA7ZaBoJT54gTd*mYf8hII4*&W3=bOw7
zOW{rU8NSAUpI>kPdmGs|S_&oA^xwbU!VT0Qv{Ms$;XqG+|33Zk_OFp@T}$#K>4W6-
z_n*@+@>^D*m#jc(3{j^*kgQdljPXK;TS<X<mXcUtyzs>WVo(HhLjJfoSfz-PO+h0?
zI;N)u2ENKiWVah(e|Bd|D4ZV&LC?lK*$aMPA&5=PQ~b^>ke=p`$U-;LvQX4RPH5J9
z;eq|Y`|y2uH!^q+J6T<aC@GED1L-q)Ur;9LtWcN6gJ%(>XTT?Kl9xa+_?BjWq%b2;
z<_}WoX&#OjoClwKL@E*>S+i&bYFng|$FwgJ4?!w8r#&D~I)`^q5*L$HnM85Y%M58F
z(7KKxx`!dUi=nms^W#Epa)P`ydM&b$H^RB<PaGLQg#Z$a9n4Y<nnJ+0DzYh21_X2;
z2&k7)gDDgNIhVl_0uz^~xGBMZ@%!=5w_gvs2&`4G@n`&M|DE0r{~TKJEd+&zy1%|1
z;tfNf$YW=<a<bp=e|KLFzqM3*URsvq57yoHuU#|pdr@G}qCjme^F)E5MXRQ>HYkd>
zvI5PE)+`$~uX4!-O;iL-Hu?G0!&!<(%TaJkF{+nEy)3F^mnlZd@xKv&PAsZrn>i>G
zWv?G;OOp}VQqRqk0vB3t0L_6|Crg1kh<ZsBCnFPzP8phM1jUTZ#XU^8R*Hy01?O9e
zS`3s<k;3L(qp>Q88MHF$X4e}iGK;ayb`9nRia|M^pUlgCiq<GCC)QG2(7N?Xu9?<f
zA3s@*$;P+q-LsPGg5LXo*@2V`bKlx=-oYzv?x{=Cs7PO{OVU;`yV*|kSfA0Jqc`Lc
zC{aYmk-B51Zum&@O-9CESmP3VMPlhy(<Ls6Gp4`}aBXDCb1BZHJV(4W3K?(3X))Gw
z*+Lvk|2bc2dK^FHToOe{WL^u4Q_d3%nPf~0A*Y8y8~8lPSnt|@i3Wkk|JE6u<AN((
ztgdqKQ`uzH)v@Pod%y4lg|zwSGG@<M+?Ep++p%JCCdOA+Pc%lrn2ZNK#eYVm1=J=x
zSNzGv!sc_Ub=@KpS!E<bFU!OmWa9hCgc6Y<Ol0B>GVw*3&cTUGK#Tn(0#LG{Xdj7o
z!36WwXoQWKtv-!^Jk=FTvm<Le^^s)c<8>z2?>erF4Ci-%h?424l_B!s=V|odV}yzB
zp$3Br8rcuwB1%!cVJqBY%}PRP3-=r-Cm8l4;2vA6oVg>~n-UZSZU%BDC@`&L3z-$z
zDJyQ%Onx7TDHIh1*D$*QVT!*h!_Z291HdeBa7yqNd|8lx*6LwwBIr;tVynX&(4`YX
zZgwa)fJ-mOMD3w~J+JVB0~>|k0xc^byUPA4x-qosq|<h;z|p7I)i_L+$cZz%&B<>w
zXOeL2&6S)bKNI<uGiL~z(**8=)?^{a<SA*`kF1?m<jlIM50j0u*lz3ta5YQ4=REuf
zxSDDPmqEjSS>{OKzS#N_r;M+08DF~(!q;q!@)BwnaU+@yxbdYp`0aR>iNw)pMK0bb
z*L`rdq$L%!naE`~=*zyBz5rA(ydoEGl<QtKfN|5<aqPhfrs(7aFWbJ-O#hPIAB)3>
z@hA@!M4dOT2>c`1hl0rMIIJY^v8z=FO4Y%VuNrxOB}r5b_CA~1!7pR2l{8+f>V9TA
zhnGI>)%7p^d92kSgs~&k3KYqdt%o6~#4Cm0r&@6IvtsCZUfnu5DYC(}oZz%Bo=3sp
zd{#b359Rn^EX#0wDN0ufP5<kP!fl|R_Iex@g2$eYFNR}^W_*#x)!><SHZxT#K?nQx
zQ8F5TTrX9@E0j9q+)6Hd{i+s$g0jZXC}5A@7kRVP3(mGH&3NDTGJhwl0a_|0wJh>>
zLC$=pv5~XD;4JHfwkBn8#vaNd_MKN8TucRLVoOAN*qdI=xP4V>N3r2)T?LgII2Dku
zpjImUq2713H&9SQy0935yS)cMY?YL;Z4W?ys$egL*GKdKD6kqHncblWFeS*(yL$jr
zLKk`f4cFtYItT=RgT;o}LNy|wP=c#%Chw?$!1-jJ*KsXRx*FvSNoP+PK;9_gpQ#Km
zd&_{Un&|vNAYam@qA&ybb2w<;8$L^VR)(mBsVkD)0!-JF2MLt7Zg?zL5fqja-j+On
zNE?N2Xu$g>`!2P*S!rq)n*vudsA(~7-!NTH=@Q&sOSGVw+Nz%Hwr^Qu1>DIa>*yGC
z=fvkn9-tfhwfoZ?N?v2e|M1O**x0jnORsz2p!;6#WMq=Ea_(!ngbCogm_bV}YpI;d
zLDW=wS*}HG#72{Y8HXHIuadF67tvCGxM<ng?Rn@|&2}x;?ey?OhK7*YC!N<)Dilho
zuia~>bj<FcrFxq2-34REdn{$o)^NF~&1#*-3+#5|p8Y$7>cWe}C(%!My<aef_|AeG
z;}ou0*%b4zCH31;O?Z8LjunfYc7tmR;z}+R=FeiJt2JzjjZ;)}3DT5?MGtF#dM<r?
zq(s%sEyrURmmT{#yD8^g5!*}t!r8M)FPSxbNW6|wL5|~&bxA`p8NyRh5|x;8++8v>
z)KeF!1$d&mq6R_x`Ptxj<chf6aG3+w>og*@Lon1*(>tCJmS>m^rKLBa0Hdw6Y>!3w
zk$2|9D33RW2Ogo=Ixo!J{~!c^j2z>6)2Q5&oKK@P;76E#cx@(>BZ1QOnsm7?JRC6n
zeLX~om`+n9b7w;GWna7KXlN98yr-j0DK^^N>|g~RG*p`h?}Iyn3X*S{*$wpAWg$Qc
zk^ToYaxQ^0$)*b4^cvcaQ6qy!h1P7Skt_VU_p4k%a;WAFCGuPldnGCfYriGM{|gwz
zg_ki-8506EHJ35@DioKucn&C+KxjS$m$<+wlz*2%>GaR9-`->v_?J&|gA?Sf(M!@A
zi-LI+X_*_SAQUn?WES|z7qcT1$v#rv5D9}-c^&^5C?uhTBInc~1@h%I>EUk{bH4gV
z5Zu8cCxrsM>p2CJ5u1$IJY=){m-Ic#Pe~Gh{5jnFqq(>0+L>=zUj@wF+Fpu4jnwH~
zL4R-~wleuh`#r2j{<ho|B{)~k_w-g41jozF!hwfQIusp)z9C%boNZw8cS={h_p<6w
zjaoQ2<5ny>>ZxL^Wbcw=kFyUo8E5$2c&U<y7k^<dSANC{Y3AB-I2u|E$K*=9hP8>y
zDoO8!ce9YJPLYk_Y*MWYf_tt&B!M&EYk!|aiwtaL5-fgWQJE-Q)8>r~Wxm&@gg`C}
zlhreF)jNbzNYK3maLs=3CfFjp{n~xIeZBqnt>3g{l~B$P|N84K{D7JxjrSu5oaH?B
z_rHff-+u3@W@ED){Fn4XGWmM4@<+~@UJk*-Eb(MB18;c3!DubCsV=%Ei$rT&l7Gb%
zbo(;8W=@bTV>ERH?`?8{LCW$Hg`cc60Sd+lFsdW0<YMRnCskvizj;%NdRj6CR&YA%
zm;o$GrW6P?zPHb6LgZLIr(^=R)4P!|wViVSV+!IV|I@N-sf3n#n^koP7U*4|Z_jh(
z<js-OTkJaFI)-rkG)8t=nGp0stACllFUx8uEy(u%lAq73p_Pyh*VPzrHbQd>?nJ!4
z#u>c;sWx&Z9nyXuay|#oXWr}KVqxT^)Dk_6FR6yKEA7@SNkl1@bIgVT5S2Vz%0~X4
zwqOb7wamRWV~2agHVIiiz%;gMrXE+WWkuzcCH>{4z3Uzq)zo0{Nzaj6#D4>qEN|Y{
z;osjymzO8kY!^M_N8+gB+{6GHQPrJsj$0|q!MUMWky@?PwSJPVodfO}@0weE{;%jo
z>irezT*mm^qo3sc<n%%M(LwsrVfvBtf)qZ~&wQi*%s2XvEca6kn~u?s4$|+Q|CUhO
zqb`mZ{e!L=2HGV*##}gdb$_{qjWEWBZ}uboRx(VyN<Os_;N<^web1imK}pSk9DBL<
zbPq~1s#3Zq*)~80vte;sZoPtIvLP;6`O#3tgiy^;dJxkZXEdm5+RFSJW*IJZ#b<-V
zSj4dKhAPdEZ178#<J?k=D_Mi+w28C0#cSO*#nP=@Qw<=~a=EO$*ngm)L~i9=-=Kgp
zoM_BX=}_2|@qK%X@(h41ZA;JPGWMzCCfe2rSuQeQdWy<69dD4?HzNVr?2dw}9~<wH
z<<L9f4{h!`iraL0zT<Pe^R4f>-D#}AS%Df~ku7#tOI_FNP#9SnZ*2>OeSmbPmZ^tw
z?JkpO*@fV0GOj&lmVd#Muf;87G6Oo-NmPK##qedBxK-k8F(7yGOgGCkk)Q{^r+478
zau@l=pK%#ZTlKladf9`-R;+T``Q`QU`s3>aADeG|Sq*M-4LctZ1vC>XY@M%V9Do8<
zV=AoPUxVTM#o}HRTx~Ie1gm1vBeCchvABa(BXX|Wbt0CzYkw{CUDo1;OpSmS>;tjr
zkyx}Dr>;M|&=wdMjK%GCia;kO{7W8Z<+6zZOrdRnNgZ3Ao3$UCo^phd201qldIAJ1
zFsr02{G0%z^GW%ST8x@|iB2j`(Om^<lbJ($W`Oo-U^z$Wx09wt`((RS1V6P=`gLlj
z*TI}3FO)Ssn16GMVf{~mayoF7M*@<LVz89H1!~WO4zj}VF6=0F6VX=&*BwQmJCo}K
z8@s#5{fpNZfK-A;2kPWWc$``3&H$1lT!6PT{$ZcWmm!vBV9g1oodS$?0oD@men_nt
zWTHMWfTsixh~Fu}1HyMo@Q$ZT^iFR+BzTWgD97$5g?~!ucHC)K<f@>Oh4vXh2p`6;
z@qj`Yj6Cx+efk9Y1Ru{lgw8|u)sgh6JSBf_D&1fXe0BbvrrL#D9H3mv<Z7JL*6?!8
z({ak1u{6|s=Hgyx20&ai1%VnLe5H)Tkx-G>`0-9Y+KE$(B!&c$o=9R=B{7>Mxr+h4
zU}zjt_kSFVF<$1ZK0^6=m?k%I%qGa=PZ-`S95fTqm!V-ZlH?WFAK_Ts2;oj&W^|Vv
zR&EoFR8qAYQR@@p$7C9~b_xNM6;_*`$`*<HQMI<~@_G=)&Ey@Yl6DF=SC6p(XEL%g
z-Pa{hidce~s@P)biSuBi^E{-H4mT`1g^9)*CVv$Tw+I-Uqa!(cyo!VkM=p1J|Hr8&
z?-cHX{eeiLqxel{9?ER4;|Bc%)>!^4=mF{kcCxZi=X?3}rZ8GQYT<>quBRwp2UiM>
z@EiucjIV<cMV!dxOl!a$6uq!C_ssQmio8w2VW*zr9jhBB_?lUL4+d0@1Ybwm-D_d}
zp?`K~gedVyyO*;$+w9Jb6+;3$%P12d_Cm~Su5xZBguJaDa#=`gjSw5R^D{z*0*Q||
zrtu|1E6VApU0Lb;Ff;-*55eVAosW_=hErf_=BynL9Kd*?#%_l4^M;BwLUU1`Hxy_b
z%IO;LKHw+`5N88Bne%2;3T8iXv_gY!r+>ADxIMh(2Sz~J(~*^WO7XiZ^_1HYmAaGv
zfJ!~3ctoY{WIUi!cXm9WQg;*&sMH<B2bH?B`XQD2IJ=iheVpA(r9RB=wNh`gd#%)6
zMmefdckX;pspo0!=Tzzk0dhd4?l{=-;LXM(90*o~KDf8&jw%hvsnaaWv}F``Z&VI<
zuY^Ckqn>+Po~+rQG+Bdqer!6+$H~MhO{?lzyjX1OIltZfy%IE@lvrSoE9_}Edk35l
z{K}JzHSO(g_Leu-3UhqI^oygC?)~=l_CIQqHmaAQ7bz41IXIU=uK^<hHa3^ZCL%F^
z)mh(fBgYMX_g^uO>C5bp9L|p*FyQ3sG(`&>&R3wV;Rl@~ByL^lV9UVezwgIcNtSNq
zYAwq_4V-VOUG9)G!|x+Gt5TB-E`{Kv;1XR<xKu|}oJti(DL9Qm977XAb}_I$sxAuI
z9XwCb1ahh76A0>xU{DfQlHi(M#l6IT<Z8^O$*z_Pfg%jCnv@-JKOT&8GfKgcR2!Pe
zk+O|V2_9+1U<#pT2++^cMuWGx*cc3{l9}S<IN&FqG9+cY7aUVDtMCX$i%IG@DxD<3
zOk;^INp`H<^zmjtMS+xU!cm}Kf~^Sbb77M(XRghHL?<D`%ScYBIU**hVhR|4Sh$*-
z@LHrfEXj$jhDI!qsMs5dH&Qg$BX%cMo1%?ZvN;eL8xh)5ahSR}5TnC{D{Mq{(KZAU
z_DSJSvRj%N0`4TTF*HDG4#xzkEww_j-P4Gzi?xlbCTWwP5{g++tQu>31CDVqk8Rg5
zHIrakve{4E?+o;G^<RRrwiPgc728GJ!%(@cRDwz=3>yh67zmLA2U1WQBwMi7Hb}Av
zvllX+a&AwdE2&9%DlK9&VkC;~aYls4VpFrN!QR4zEl9t&2#D10ZH0|~Z;MH>J#E2l
z)(X2!(PmY|Wo)xa+2z=LDFsqC{}A9=DW4?8L$&3yu*#`70V%h&g^#&^4<9T;F1DDl
zVvySFVUXI40gAa|099;yA;d7G;tF#*`R=>!-K82-g8R_@{U85y6p}Akp^ONhJ>1@2
zou2O0%Ft4q;|MLRq^DR$Xfm!jY>w%=ALh67yNl10>!}la!FBH^i^cSA=D7d)$1jT?
zFBX%<)KP!qd_G&i#yOUM4kS^Vm2*Tgh0>cm$EFZR>3?9?$W{bDPEWh{ck}Cu>Eg0`
z|L(l|IQ_C%di?RX&r?|bWeSI8i|K4}Z<N!I_hEWJf4IAz-uDz`xBfc)`DXI{{L5u9
zmIQ_5%2#;*4sV)PYFcNr*}NYp_vTM78$sGM8bL0FHnFR<8|}4!?&9Io!hZebX7+E_
zeLuhZd3x8(g{$sQ-A~;Qm(>6GFuh(}f)ITKBoJucqJwZv2xr8M50$REGl$_Xx*z8s
z=iNK^P13@YG8jNOR8W-gA5Kp<*5V*1IQjshlvq7u$(hI#*h{qfKxN<rFasQ?nBYsz
z1=!N#IY9_Z13AHenFO|>f-oH}1i?B0oLLFpq*Q~1wG)VAOL40>1}hDLv#GEObg-ST
z*p2;~NTt32$on;!$rqsgIN;gB%LYWg0?!7<o(mpK|7-A&90DGC1dkN~J!WVH9y$hi
zmLY@WgAhTe)a0xEPQ_Q=5i^XSvP6!mLK!I-Y2f973p{#%vI?Q7x0Jw1y@2wist;Lz
zCkW@_QA@EDkGY|OiQx8XrH@BxHSi&qhTBXiG=T70zXA`9Nj=+I%hVI1$10D|#KgU%
z@@8<s!=%CoB^6GOxFBA_fmgWTBjN(3p%+Y<BQ^y|F&R$Mkk2^)6O|Pe2r4*~(v**w
zR4T5hADI+?j7g|+!JK&rPR@$Rtw!fHp$b9)b_`gca#8CZ1kM1PoRQJ8^$<I8@xQQX
zam@RY3d@LUY?l_A2kcFJ4LEEST*s=hL<VpWIs^`)k8t2snvige(h6LNHUiC9BN`UO
zLP0we)mN>9y}!pxH}jcSFy|4nLI@V-Eeo`+HIvVOV&ShAflpeuWU|f0!eh92OKVr)
zy-qkiMYYckr!owuG(Ma*)*(NQ&{t0|&z*G~&N`+;K&12tqPR*WkygSg%k*CMCksnK
zcK~g7bYsqrR?&_21YC_aCF5GrIUwY$Cr0QI+Ln_ov~*|d;aiX+?Em)RC;R*6*Tv%V
z{kL6z_sh-V*N0F3dj6m8&kwgZle4?K$#3`F*_WI9UiAOE1u9Nnipm1sM<*`;Qbk@6
z<O61u^MIZ1$OnS#RbyrpjU_5td`>T?20=Qm8$MtyYoRB7uo4YvS-2GRJOvb1kC2|Z
zvbPeQk)G{T$Ql}JJOu@u@~}oA6a`5Qs1iAU*Oa6jAw4%fLz${D4svYi*(zwkuJFKz
z0(5ghoq&H5$(p_+ECfQ=PJLEBzyyjVebNb|T22vRVMUWN3h6Dnj(H4SZ)rVQC+yJT
zX!%4(XcB6iTHem7jphLvsvbgyNgl~iSB$6ecy%?QghgkyE@`nazLa68NJjRT0{BLM
zL^+VFy9$f_Y%7dC*R-~an%?KYYr8LI+jo=2r29DebUWSVS&J-9drIC7h4ncK7*5fS
z?$?x@tVtfBX)Ue9F=G-6y~1zj#3c=@fOJq5kT<G;w94@5zzh%euoHGrz7NL9qCTSs
zR?s3c|Leuv;pn93OJygh_w7L5JPyf!PBjr?MW2;TEV9xlB2Po&l|o)bM`$ttoD;p7
zNX^4Dkj#fv>3Lrtop<!n6cBklLKDI76O|E~U}$f>$9_#houuVej;o&9ofl=(k_KC~
zwD(L>so{}S%Br;}tJb2dT8kpb$gcw6h|`R`s<q`;LVi^{S*@>_UlX?z+m&5^>)Dfi
zCS2P69k^~hAxAiy^pTLG)vNLUlaR4}V9i>QLF_C#fR9_?`}FLxy66}--qL!c4f}w<
zUsJ4k)e(vm1qDWYGqhcYpLFV&PdXaHHf?Ww8{SRZyFN3=>PsG>XnC7Cj?lF4lZonE
zS$`dVkJeG&qnGQy=HhiKFfOhRD+;bl3fATK=>Gzfm<g9bdp8pSHJ5Q@I~A8M(kaP*
z@bvdDf8EK#Fz3!X6hyOfrw>m`$Ki&iGd{B9=o=CPND~@WnT+4?jaqE%Tl*PZ`o&h8
zCjRi|)v9`25ZWlt01(^g<4l=K{2WC!q%u6@w)EZPSDNimW?EkeKre_cpD9UbHhBc&
zg)vJd*-lb(H$F84ffo2|WCnSZ@e_4_0FZXg%q*fXR<}{jh~oRx1HH+|Y<WrWJuMlg
zg^UV6D7}9viZ|T?xYo;-ajCT|{~>?RqiWCit2q<}`D6x_23G1lRjeG@I`5{Enz#VM
zX=r9cGh!gR#uOeR-87T2J^!+BcF5-2cyGO$nWvW6mwpFT4JK5orb^^a&(WxVxuP|>
zXnp3?R%EF9PA>A49qzfwT=nOQ&hr?Y`JefoTME~%%1!V4==g>6vSRX$Mp}IAdbwj<
z+02!0^j*`INiNVhYwcWbSh;GtXtE{SSlEJ+!nwh*lP%=Xb?B_tXZfm~Cp~37@-@h<
zy>3kZOX;^6JR}H^o^z*gJ6K77D}r9gL@SV@xS){aVUhyr<+ah%9~9S@t)D+0=Zp5E
z0KjK}lOCS+5d#Y7t%&ecm%c>Ty7e55drAxI7Yvzw=R$97NjlR_rk>|Xds5z=n%NI~
zRqT_;w^*!D=kWS~5I^Z^1Bc!+#&3gHD&NNpTU+a>*#=WQgvKnh%!NOHH`^QhdA7UR
zpVz#>1TQ=^ypbYS*^m-fri)6dMQk`OVTxAOCtEBWI6AS_haem}&RN=`UQR}QAoB<i
z7NcK^M+6(~E53bX-(J7$eqX8wtdIexuRnM32h>XFXnfK3VEW@9)3@CxPPG`Y6q1JI
ze@14{8(KaWKo}$mBCREV=+LWy?AsfhjI$zb;#m-U=>qN{D-;y-oTS(#MU(_B4X1dX
zD;M=)<h}k&A`LB{JILs@Ak@ZZQzNg$k)xqXPb%sqQ5;$ZicX+jXk<IQtj3mREQFzg
z^PHk2gVwK+!o^Yp63{@k;88NEFQG^wo(zlY5wE9c6m0<uf-g6J5Hw`Zi-l2kUFnAW
z+bj$tq6RA=nfT>ucOVSyQg>84QHtBk9o6p1ptZWE-n|j@T2S=sdiO!B#1mOnpEilh
zEndesr537}ud7UGMR^>kwWucR+yqGGe`)TarH=?TMp{w?&Y+2+K?pgM$#Q8~L0H+-
z4>|WTuQ?g0aYKcFoC*_x<-%OTmk~;mi5sK^@{KRr%oRw?zR^Cq`J<mdhTCUHEwt{u
z@#Ku{<No%^H3;qI{r8M1+s9`%``Ye$D&xa27+6W)dx)nNn4jl1cEB|`PuMO}Sb-t%
zxxw_9FJod2^X8;8qS{0?peY)0!q<Wt9cYu4z(GwQV2$B_PXRUPv@k*45!5h>EC;*@
zxELoa88xHG_LA1yTA?f~dJk)UHF#oF8F=Eri1l9DScEp0@2v`=4RBVDZ0^blP>bcb
zA-1ChT41=$nKFip`#_!f0{cB_ykv4t;XRFar1*;wIh*Ych@A6(J0fQk?|{fTkL&JS
z(KhSdr`H{STDl&nZgX=DQ01$7J5c2mZwIQJ;wDh#ZgwxA%1vt<sIEBsEx`<8?qmM%
zfGR34TyXCMR59Wb?cWKhY64ZB!s2@&Rc*&mP-=nTfz{V?%c-*o@S2@X6iP?VCdk`Y
zD%zRK%Irs~kY$b;b@lyi`~8ZO1d0arG+>~dJOKuOZq-aVpCuKJaQ3Bi2zh8U;JB`n
z7DLM%vZQho|D&Opfyh_%%*rJ?LmL(F(esMHiBqFaigs$$&H@%zhdxzd>IT<jA2K7n
z+c|>LhWt<Qc`rIgIC|Kd&k(lHQBS!en3!8F6|;J-fQA?Kn-F#m-ljM6dbpKG@RZXR
z(cSfbSsl<G0|cE|R;R!c;;?^8TBm6)h4)0pjN%>AI!5sZ<W*BYM)3yZMLW7cWfv4*
zZlH*S-qnJly<yudYaO>wZ+MVfP`*9if}jOhcSoyY6n}BP#b&!hzQy^!J>Oy!?~reC
z9=G!?*8A@H);2fS<Xe1IZ_l?l#oO~OPH{7T-{Nj|uY8M})^@(N;_Nr%TLg)OmR$j!
zUb5a|nlVyC6o=V~@|CqJ|8*D5j^}OiEpwCJ)%>!u&RBs*MrQ47Y{1_cdzb#h4S?IS
zLLfvz=m;D^2n@qzyG^ZCUgBmjpXVLj8m{1ou(9r{tmkuNdX6qCBKnmg;uV&NfkO9x
zT*mBfMimcpbLNkO{^H%IqRF9N0Pj~X0PO`j{Z91)AZsuR->F^z%BE|czp7q9;GRQ!
zCA|pwz~=XfI)0o|?_G2T3iJP+QnOHC$Xf|MDCKM<tLZNawfpez@8ACN^*7pQrtklW
zT`51H^i2gn{mfsT7b?W#pC9?VC-D(~eZ(L^YKs#^Qu<CE>cH0Ub=o1V2xWM_YyA;h
z$>u%M_B#(wS&54s>jZOC;~`RPt#!U1ePiWno+8QA*ofaEl`=MdrSWxC{2E{V_u7?4
z{d3xVn|7ayR*eOhGF+Jn2TOlzT2+MIU96kVnypP%2_K}YN(oNJd%|mty|k@=5rp*?
zNmW3(DaF&cDrK%Fac$|Ou1!R%Qm9S!owpy+0+W<l=CmrE^VlGychNvq(QVf6{Q2w8
zGVVK#`q;f^I^-f+pt1D9IT3t_v{K4x{AxX+bg#X;>RF1bp`EGLnb!P^5uqtRO&v>F
zwuf1`PiHGcDGCyJhC(<H>J$NgGJU9uI8sKeG8jt$s%<cOKNj>5(&gFOqPr{CYOYRn
zz5H0N1Bj0on6h=HGMekIW7n&)&yj|$@WnTZuU)}zUt3wxajUeZ6;jAITuN3CIC^JP
z10n2%j<#~96PFaV^)bwHiVW%;<%i;-3&2OmEgua$E>>-$=vTDOu8K2%{rc^)CBZyv
z!a}$FYCdGQilAu7&$gdaX!eOHbO-F~McG_v#qEm~7{f4f8oF9<fyvs8<>bU61434w
zyio0MF;l$uYm+Z%x9%TZsTkm~m0lloh~;Zo)GVUpI2P(R`3j0MI>q^t`xB!`_d4)V
z>0){=nIC&#VJ!dSFs8qM@|$EZ3%$2r7FXiORX^$X2((EO4&&>weJwujoF@7g(^gH^
zW?3reFUJCrYo(uIPO>t$*vmawZo8=M+KS@x(Szy(6o*jF9##)E)Joa@Q0>R?2PwRK
z?B$S;b2=EUF!l9wj843|&|lroW43)R)^T*uoabNg#1$pz!nnnMOv&eD7<{hr&acs@
zR<mF8%_jYRpwJGlN`$MPX}h$$T16W__=-CnSq&iv!v(`b%R}Xn9QmkW2g*SxL~}#h
z*WeSk+9z%1r)gHrumFq7kLk|Mit&ukCtQ5X|IDDK2zE)w`H|+tT{wbP%JWgx7~7H@
z8Lf?gi>E}&JKSP_)S3HhJXwk-N^JD0M|-R$ObUy|8Xh^?e2A<)uF=O8R()Or!!zt$
z1H&`0&0zRn)7Y(~tPfqzdd6d~A?8&xif_*y7H1A~AwC5~ql+IU7mrWQpdFr^i7jpg
zc$R*AW5b1&<Kmf{Lw~YdDPaw%x&QLKxc2M4;RvK8=wX9X<EpqYmTD<t(mxWJb+)3R
z{wYQ-MGm1g7NA+z(pgV?x&@?ct1(tOHv_770x&S~&S9XGapme^xT7%L-N3j^<w}!B
zP#QlH;YGID-~I>g3xF<{G3F^1mmz`(69F)n5e-2nmj&J_odM~$quwd`0Rw&BgSVdG
zDUSsMTwXPoQS~kp0yQ<4!B{yH0yi?3(QP&-mx1Of6n~mzz-}9I8+IDno_bi{|1ipu
z9b0y~rpbYpOi7=}kCZ4huLjMxZhP$R59~~UC={$Yz8KO0)Gwhby_i4*p2-~F%&!MT
z>s7WY4v|?UXg_Kzn>(hcGe}2BHAI`c;Ha))t3rAHCisW4-aZvwS+C+1BU!N-MR7^T
zX}kG-{D0nIH5eiJ+|5n<w?B4&yViV30oaQ9^Va17#{?X$i57e^@6YC?`_Y1qTcNHf
zA3^i}Y8uVY%3ng|54{(e@kgknI^%r;&O?Piw?h>5*Bw;UKbLu+nUD8}NUMkrH4?Hw
zRCYz#6=l+ef+|B>^|&qnc-Y)|q-qB~m3_fwkAH?w@90dUjt(|hj1bVbHdvf_FwUj`
zmX;Q0%c|AA#bILM7i<#Ia4eX{lx5v6A3^1uT~6a8skQ@qQ@(`F>C@Xb(@bm|vp^!S
zrq`!qHpHpl`|)H;-%8CLpJ(OM;!??~euh7XC6AMxw!>O-Lr-A~em{PS&*1k32jjn|
zf`2`@5N~-Z=7@H}aPN<+zE3z5e{rr+WT24R-G~~7e9)!DhB7HV;!HC`4B&Xdh~;^f
ztoyK@-8Ml|qy{l(67Ms@<tNN}NSeK^jo5hog)`gm=&nFa5M&vd-PaMl1JsDMi1P8B
z(y)FQ&*<)Q-lKrZEqPB2A>EYs6hZHw8-Mc7Xn}FbdsyY2(St}>%-)Z@Ur`^m#aRq5
zKRXHlB#Rj0e$^YN_0@Y`yC1KqK2}c~r7?f%ZL3#I|Ar&_?0s40a-Pd&j^06iX{lv1
zKkH<BDy99u)RL9^S-&q&zhNs%<+*nvpJYmUeq#2;jcb93pTSrgRfAN<a<^U%WkU>h
zJRyuFH&z<j^u8o0`sm0ZCI>RU2=dLGU(Rz9oEl3OT{-l*D|y6Uzd16fmmlmN&h8#R
zdXW#>rq#FkKeGXTlb12(DHNALpB@wfH#3*<R|O}Rqv<Ive?hTL)3x!{#)=+(4tn_d
zq|@v1{pria_yI^6>zp0le+(esX!#4;DR0eSH7e5b)vw{N7p>81%}04yYeVuldGF=b
z9G;22>;U-&!r|}tKN4}P&@;O=;>I||jx%u*;s%A_MP7X(ewx+{$YT>(;z0-)aJWX5
zLr^wAq3V*zmrv>`5)Mz9ijONHuIOY2n52#Gmv8DRD1WzyXLwn&Z_mA<PvbEoWZRgD
z*C><N0)zqsMUX_AuGq=X)?V1ZvggPt%<_*qj5bLUij8@;k4`(2jYX%3iw+0gDMtmm
z<5>xR0+&BQ{510R2e6THhPIyc&HBN4<e*is(Rbp@*XGmP`_sRj2xvhW&%-~zp5hlw
zpvdEZz<&yY8-9NuemwozlFeSKwESlJWAyO*=g_QtOCB7wc!Rdqn93VKi;M8U)+oT@
z4SY^$y(Bz}wg{7UHDi-_&5ZMkFfLqJD@&&1l2Fd&{3;G#E*@58mUv(d4Q#|d7BjVt
z^VW)^K>5!oXSK4F(~|-hTFR&^xytSU%6i}KrhiTjSrEiviZ1v%y(fFOd`MmiUR#a;
ztMVbEatxh(xCqLp6rtklIR|N3IMu$AEzYD+Q?YpV*GLJ|=hMR9Vfrp=Z`Q1GN*1tA
zR0SHKAgNkeo=}#AP$8F`fRv&cFJdS!qtC_(8Wp&swkfYfMXSfic6|Ko8p)x<*IiT%
zD1S69QJ{4|7$e@vow_@Y0foIcpLGm1wHuB>=+j$#?mv#~95v4z!aG;8DoEUIc1PXx
z{b2@8P|k!~d0SR;q8;yO)9AYTe*Rw0lX~4R`=A&|xxgJ&Ue_vkCCR)y^Bq`}R7Pkk
zEan2_yi&%6O+3o4#>CrD{1}gvi@oE6xPM6gS@VTG;UOWdp%|-gvw=3R<Y|BLNWgPe
zIp;o88*fIev=q&D#LJj$v^hN#GidB`I;3s*t@J}mT9<bB7!UO>q?5ERB%2mUAJgqI
zm*2}4peo+i(awquv{Y}1zIw7`O#LnYN6??}(#Vmz#7^oI!<yZX;M4~t6>zKm+kcKg
zIEA)Wk>dtTCQ!ACJP6TP&jWV6Hy{come=j>&w!|xYDfDDqITJ%rw-C2Zl_!=UWaH+
zbq_=zfF~-0*lD)2eoG;(p*jX=x)RnB!Ww(*I4|eK!k>Ly7157RE3yRMl{#yd@VZe4
zPg$|v{7C3X$+A&&N9f2guvX(rx_{32sa1I_c;sx^a6eziV6}M)9gl~t8ixCPi;$FY
zX4RORu0>=OQFo;4{V-Xpc)mk4A9U7ZP`%y@cN`+FX};s2xv{#ryZ-o|X7hF^-P;he
zShDo}xwmK6oQ()P{~0+|E8ni%>7;uoB2pH3d6zO3)Hz0gF8&}Z_>#ZFseiQ0@1_IJ
zJ5rmt1m_(xcQPWXQT)Yyv|np9N*0>#N+0KI6ln1@0A+(+6h@r|B=lqp95=L|WbzKm
z&h=dLkc84WIhyN;F`k!`I~R_xS0%4cx04=w>CLA|9}7X}{30Lp+&+Pe^AhSp9jd$2
znUT!<Iof#53>`;vBgeIC=YMf%6BHxr)qxPAm2v~Cpc~>%<%Edn#kY-e(!9tdCmREu
zo^O~;4svL`$Cv`-@sv%|t<E86b1)RE7t*l2<h&+hzWBT*T`}6mNJ^4;2p#8flv4%}
zu3VVsal~?D5=S^`NRb94<E(93$<w$LnE0(8mn3S7QA=9c)++Bv!+)<e3T`tYrqR+l
zQEbLb>6afOW4`vjcidu)5>Pb577)N1>C*p?VGH;;IIwpNTQ<{XO3)3(eVA=aiUYi?
zR2BE<O>mMOaPF{6KA@GEq%nQo*qWg1$mPHijdj?N$;}X*oNWic*kMcmMbZCD>9D-1
zNd}(FhSGtRQOe<&5r4SOrMzDzfxd4h;b{(3Dp*)|)>PlD+bT;q+f;r<G;xU2-&oyY
z1p8&b4wCBvRWrEi3XIc~(HG-}q+!&g<i2GHODn6`?7E_dWR`Fr=BhdxJBH?r@=@O7
zB*EHDaUG#~zIVulXNkL<9N6r38c#|%^*)1k{$B@9ta%~fY+U{80jku5$L2D#a*=UI
zzA-rMJ`IOThZ3A<*=Xtc-!`rTo-EXeM?nnw3FYy;BI?mZ#82y-Jpz+QK=h6q>aUMV
z3IZ$HBHk)0zydLez#2~RsaQys{149`H=CC+<|!1HQ3n?jmy7Tz!hduf|9bxYV2iTW
zM3;W1ulBE}=fgjTR&^vXE<pG9*F(C&1cm~3%xp-F{`RT+eE6e%?I2938$Z5%>6(#`
z1wDPgH`Y0k8F~;6IqIC#?{ur6r&fzv$BGeOjt|ZzdqwdPpGT$45XFN7_yos?G!fI|
zF@xQM?=pZW3D`W&Sbr7;3%Chb-+*}Au7wd23@JLri1c<=sr*5&B(<uq5+$aw!LV3O
zEoSQ-)jiCNQXMTw@|Uveq4<!#5?mHxpUOSiZw!0G<nK2o%@QoDqq?%+GKMC`G}p`+
z+HVXw!#)hgNHO9vI@5|FkL50@iy`pb#m>dLxL-LGBM@)+{(p<`#|>vHp#OMZvPziT
zIZH+bR}1Z+2R#;FNtQhgz`MBz^>&6lOZWR;yd|P>OoKBNX+2)^ZCUZPQxxqfGf`ay
zdp<rWer^tMEZ{yC5cM^1%73Q^2mjyCIubPS1fH5xl@=;Kp1?Z|@Sf+8&UOG6$lZ$5
z>?ai!V?2&DUw@@b(d1mU+O6w;&Jc;lyOGVIxQU-}#KKjN55f?ayNw%#tV+utVzWZC
z5mo5jCEE3vnRk_thHoB7Eir>sQafj;5N1+0qcB^|U8H@ZOm{|D0-wTYKPMD-vy<gK
zmz`oOY{wq2q98oYVqaa|8an#zl~p&K5UC$MP_vQA%73AE#(UqIfp!ggDQJ(-*O4pG
zV-Ta)0w`aM4x`>TASa84EVXY^GRaooPqKo}gb-Iqc0q8{9X1JRr@XxQn17SoP5ffn
zXs)`U(=wIYyJtDFv4CxyuTI({;IHl7MVw#k+y9_cAY)}D=xH>wWk<mpwcW~lThb$(
z)EX(RwSPGU)_P*M4IYh+196j`0P9M=@i-&KSS(DI*daGbR-9(p%BpkrR_)f=fdLu_
z-1a%f=;pn?C6<$&l6%%>-9<CXni4J&gpzXE{pk`@Gzek2Yp_6_)9PZzM8!$!a+SJg
zsm$gc>1Vi{7*<}|vm6)<%D_*_QESW}@w?^n<bMEvhEk8g8~U}YRkqF$!`rA<DN5h7
zcoWqs1vZTIb5^V5$*_|*S*;ce9cnJ<?jWk0M7t-Np(cnPI}$L$Lq8^$a}7>hFbV}5
zDMe>&Fyj5%XvH95On?(DJr!*1GPN8u`;MEZYKGA`j0?@=|I@7c6$In_3xdWEaVP7I
zhkt?nxsUT#;CY;EIm!KuSA!>KJ*l?xk$#dU5qdU(_^mXF(2*{?{1Qzf6f_b1FVQ4I
z1`{QGuO`vjiOai-%8QyHV2V-8gD4eba3sL-n5kc!ha{(Ls=oSB$0Sba&R9^?k!iJ7
zMmAarle@mys$PZIn{(^r4178^88Hk^1AhoL&|qrz=oP50%wqK8&}>;sLXGIEL?cOF
zT!~M5UDulyTehN-9B4_mqJdyK5lkh)n1fEKY$CJdr}_L)rYi`G^@6MG1#9aiCFUjh
zP&$YAo+Sc>K22%P5>wQPa_cDZc2BfFO1$0{1<^PX=oQ<d%^P3GZEAb;IONNd-hac6
zHXyJDF63z?wB8d+<dd&EI#Td{yD@@pjr`DVw$j$w?FLy`O`_bamv(w(u3*)OTS0cL
zzX9%nY_r!1?-!80h)Gz~DPyA_*U%DjyDbySKqRgNp&CAip3gRQ^bNeaZ!W`V*+to>
zho`YGEgM>C{bNHSPv0~RmAx!K*nc83Vv{4*^MpZ$95ao=g2q}eis=~2PybgV_bJEo
zij-_$7uDZ`fx8!V?9|x7J=|o*xaM2Fh{~R5xl#8<n|IT392<=7%k1U41}+*5t_AE;
zc7q$qq#rw%Zq@LQEc#u$TmGtcL}J*v`?O0{D2s2kD?PI7eYNY_1g(xVheo<jn}|BQ
zRmSIxrozPdDhjgZ^A7z|!Ma<<bm5xHtTk6yEbq**N&~Xjsg9k82IQ80LS!K2sk0Zu
zJnq(SO(gyWvV(&Gmoerk6ah7t0jn+(mt-RjD1X&kTW=h@4Sx5pm_J}eH|hcyM(xe9
zfxackThWJuph+K+qUZFXf4|g)lt!c39oy@kHo$s!xe`TE6#0>w$mRPYm*<<uFE<~a
zj9t74puAqbd|MR$An~t~NOZ7xrGztT`SNr5%bk?k;p-7!R!U>=<?(aCceuYbD&Po(
zzkjHRFLi(GtwM^yIY#5Sr+XXI_fY4Z3f-tQ2It4<(|CoyE0Z|g|MknCZ)6qrmoGP?
za84NkGAv(z-u$-Uz~?28zLCPj0L#CZOcqLmjf)gcDev$-8vM8XjsrjZ^ZU0yef?Dl
zoc6CYR4!I%P%$h&%U9=wwm5J3QNITPA%6nOAR<bQwv)m}qa06U!q?@-YA=+5&{2OR
zY7{nDt-Hxa2@^aH<OMns$&ffQmDc;w6G^JJMAnJLNZ%5dIyZgA`7|qiO|Skt4aHf1
zUv9oFH=n9i!D*j{1X&nIhyP|)Rl+KdF4S3;Nf9^0`ao8RUIOe<JAIj`g$vFa)PFDt
zj~-(dX3aVirPG^8$cs1-iQY&0jvWSFk-|<20-{<ViJ_B=s%qV0;r99K&pPir&U)_N
z3!hu1jj-UBTZis~yMnCp!G$nAxOH?LOy?HQ8sIszcrrofb4BaS&exg;F8FEb<SB~X
zHqAaAwgyHM0DHh1B-O+U8iaJ3&3~{Hzeoie&{Z_s!D5`rQ3>E)r+16d1~3{y$g~{8
zK&4SKeRAbjYr9b?R#}~X)ybmSnjfBmn#vIY4=;>z<hW<LI4Da@#TZ$*&9%~a(9z3B
zZXk;uSxQ9`W-XKKkLg#YxGMJ(zQ?LR(67Zms-DSTJn1SQ8~;Mh+)T4%o_|+4T3lD2
z$cmP&ZXfE(NdhPw)Avj}F;`tVC+SaBzdVvH-)j-Nk-A8jvX71SB|Bcu#Na)~qrE-b
zdB+>HhJ)S!LAa<#zWS+VNJm~!X{eFaeFnw%STe5m=RHL`Ji42eW}Z@|?>q%3pYfD3
ze&#7<UfS(ChIZO*ce<`|W`CzAQfzjbR!ZIp4^@LxUGb55RIX*+ELoMNaA(PZ^oPs7
zI`x+to5<R`+Tn-f%gWbPi)MtO^4RvxHB0C9`Io$dx*t0~kF*lmQl9b&*+W-@GCSeF
zh4K>q|FQJLWJ*^;MmP`7R00r)8$7<_DOK?xECax4Q5%9~c&@ba(0>4sORx_$;fYLG
z!7`?<U|E~Yz={{>NF>Am5G+#+Z!vRw`BFTF!6N!--5O-gssE>V%sSznx(XhHXoT`d
z0U3=Z1n_WgC=vnG7)mJZ;V44uM3|k?rtAwL8oUXj0Yby*!a;}z4!sADQ4i6iIReq7
zJ_peX4-D=F*j#{VR(~MRjV!*DtCW;eGR|-fH%U_>wk1I)x`vv(!};Wlq0yKyu9iSk
zPBv=|`XHYvDKSB4uEWXP*W?~}E+uu_&tHMULNS%o2)3l9bkyXbgbyPCXwWaiFO`dZ
zO3H!uY%YQIO(**_O|IlsWG#}aB+;(GVSwW*7=W5ML-B={dMj6QbP04~iI$ihkG*J#
zeja)t&GVg2*vDCua9?D5r1h#_H(f3By`b$8otOOkDH?yn0W*vZ+?X<<TA-XfiP<59
zlxA~_!a(l#OsJoq_hUlBKW1Er58Oiu2Rq@XH`%T#-U=W6W(8luR^99e3V^)ncDi*9
zp}TaTaOcYJrn*RW4`m)R9$z6s-?>N0KD1KjikZ(-b)M5EUlG50+;c5AI6G5js5rfU
zY*y?Mey4w{?`E4W+<dH?9$hkddV3zeq1*np=^xsF<ydyOL@#hnF9=iu;@02)wtTz!
zG%~GQifJ(C!t@7Z@vRl*bM3B1i90Wy!voOoyh8Uw@C-6xNXH@VV~6rj#Tr@-zLcGH
z%h4j>Gca;I=B>4UlGU3id_CnFzE~C#qNG1r-4%c3bAth-8<yXQSGn>`?Cme5dz^J6
zPjRLDF|KTOBsF>e@42}#mQ}Da*L?RfR6``LH6r9mY_X6~6=#~((#97aopiMBO8;gW
zR_&3nT4O{OC19ZR=!i2>5L$a+i13y*SEJ?;>r&9g3QZXB?;muBh2|OT?EyHXHd@%M
zwhDj1dnnWs3avSrn@HC%gEm2=>S;8%tc{>k@kRrL7F*y*GVUE7?fQdu$VSG1P$>5x
ziLF>CbCzrc-{i?{i@awQfx|RmY}3OYp6BE-LsHkpIxkHm#v{hTSn|hZ5`(Vnt-fL=
z892WD2);{JG2ROwT7h=92&>&w(+*9(Ys`O$p}f`yC}))yo}pVO(=091Cw;fMw%p4=
z=L4;=N;nnz>w=+bxHaDP%|Y2roug9z^?U_g&y_dahTvM{T&=8?L${6_8C8*U)3DxL
z_g<)t19A@|_gu3TjoT%}St+B&M6!|cV)cHi&T}eKQwu(~=0kCqqD^*|%;wZO_vL>+
z{H%f&Jy|P13Z{6|Odr&cQZRe&5xH6il+B&p`jlVBvVTntOqtgYtKyp+5^;!gJFN;%
zeY<XLQVpm^F7n~REF-*(0D3@$zvHZ|k<Ly_I#QQgTdI$=J%Wd{?bE$B7t?)pL_IiS
z$B8Y>5>WL;Nyf7L(;d{)9h~rAH*QLjU8TH#GUAX8xUn%<&^q!8nn@L8SF1H?C^?t8
z>y}Qx($J_;a_(~qW4#Rh?He~ZPA#|PUE+GP+*2~VSY@pWL>LnI7d3IXmO8C&P)4bB
zM8xia!y*O>Lzl&nE|-kF>kO!vGG0j;b=R`aH6gawm|Wc`D7%}hNj-Kl1@1~7!Hime
zMG>&{8&*h1N0$<0c2$?HaOwDQp|TV8J!TKP(T!RhMUV6O2WNuY$W9%+D&<ZGukK_$
zOl1f2b*NECt}@rW>(Q@VR{o;;%gVyrcDqCegf8OQ?gw+HMGpxWa7P=&wU=J7q>Vus
zwb_adl&S7;YSHqwFkD0Gr^6oz!!^QxQ+3UL&cN`{8GRSvt<zIzQk^+b2YN1N8Xf6@
z1#3-@c(aFp>=G!Q7fSafmh8u=ODJ2*AkaR%X8#C8<ZAPR6zJbmDONlpeaq3H%$R_$
z{18Q)UF5f}eo16H`Y|eGt{fFs@>3nd$(Pq+*|U3IEyoz^gu)?%g)iH4&#w=E<sGZi
zNg5X|mr;JY!M3me*dVD-$=7ZVWL_G-dFq8t-L|1aYBP+Wq7yP6cuoYJ+^B<&XbQ}j
zIHK945VJ}lh5^bsXQ`50-&u4p0u%SlDo31UvL}ZcTVRC>o4v^1;%qGyGnzG<C+Unn
z_at3Q_MzQXfNm%{Bki&S6~E?xs=~m0o!uQ}yuBmgxl2{@#DG>X?zqkSr;^y4_&KKa
zEzenRQsJGvDJ2rD8V1eQwy0s|eZvvvFp(b>UrX{@bc4_q%Ja58xQRwM8~-cdMmenF
zwYnm@892UvKW31JyLoRwJqq%J?!0x#3`ym`Cxm(<lnPj1>5rBEaEU%tAg0Uvb8&NY
zKu3UUZqAzH>%;M}%|YCrb!{eCQM%31Rts;{D4=xIf~7k~{ZKv|M{Qc}!PKWv%XrGU
ze8(MHZ-ySM0p@-rMwZoPXrZo1PtO8Tm)rp=A_0b%{{bo&1o)HJ6_+vQDHNB1p(zxX
zLINtm0p6Dc11dd#rAt(ulAE(uIhy6hxG9Q^{B1eyLGV?nE$E5mq$bm}meSc3LT%?E
zKH)4r^C~Zp_Ga%;^gw03&9vn2SXcV7@e%R~%tz9cLK^~3S`k0NSTlWN-L0X&h&BUk
z$q%DfjzNR`)|wb>V_O>rjSAYVur+@DZNclxQ#k4*z0d1^;cekDIyqE6-Vbl@o_1e%
zD?<RQe1z%w%TB%^1In;8kiU&~`u4~4ara@wTDa0m^OpR}>MY9z<d@>YQ8V+}F_#Nd
z1QxfY$r+>AaVzumwutaCGbijL4B|}6Sz{#4!N86fx7<Xt$y$E&zKD(0KL1LrEFjFv
zif%+Ep*eJa3M>*D*aSfo!qyHC1$JM)KLy_}_}muzDc82as>nuFUE9tslUz+C;fpy@
zF(-dD2LX!8k2Ji{J-}N&4_DZ53_{6GTy`2LVIYMPDh>j5b>{b_&SEtx1V1{92jw{j
z{U6R^olzLV6`f_kct}->gwFaak-vcS4&s2-<wwkau%v5V?G!j1S0@MEzsf<aMP)E9
z98?$Muh}RRV{v`nnX{DBVovCXl_6Ic5<t=}SKe*Ob(Cqh?-(A7EQ{-<SRJ#=>&Hx0
z`h@!2$&pUW;V25ZpMyL-RFaxiSJEEZde^1K&eoe9ULZ)x0_>8O*v?y^!;|u>L}uP9
zVdkBG5R;`aLlZ<~re98G)wf|wgt?}nDxxMdKkHygh14sB9$T!WdQ})W^^Knrf>Pk|
zg?D>e>%9C`O5ubJI!+d%B8v;_7W*-;$E<^*Xij3@4s%yXAtl<9U-<7!lL!|BT)B_f
z8Yb7*TH<U~kMypuvz8SppAGIdM}S=6-XY$9_Vw8^p1daMkU>R0{J;v0B&P#iG3g8#
z4_0V82U5Zs<>rRs6^VgNP+kb@4BV6lF09xZ{0fyDad}}E7Fy$1U*gObm7<PamN{eC
zztq}{L*?lQ8k_M^MQ^WYYy-wa#wOBD8SCYSWn5Gm+`_PIXISS%P}a(*6KOgjq_XpW
z>%<G+PiZwHFfrKa-zk9=pry#@9DsL%*6@!IzI*fcr;k59{|<_`{Ub|RvmBBkhUrUr
z<>2ViPoK*xDUYCOl1k1vb{VG}l5gAU`}4FTHV!Zx-y86=069AaLY#LdQ}VK`BJV)u
zctJM<%UPd$U?Zp<kjbw!$eUm(d*zjX_t#zWDo_7jm-4PZr`^YC_o2wsI`k=D>5Mo!
z{@IczaM>XZKqu1*LML{~e-xQS&mMaK+VeXvRW3L(%pu1twkR{T(;Cke((zqEjH_a(
zphnsNKOl#JSB5n6v>dHq;{1xvh;c%ZZkF);_w$#s?-TEO-+g90H-Lsq88_X37^&c_
zph|(lhRtYMR+q)y1S>_$nDV*!Sv_aVvz9pczz<VLPiE|9=Iq1S(g0A_(izei`=Cw=
zTz`3jCWZq^qq0y|cN5SW-D=ECa=E`=i_d)(htuKU7x(x<$<ZUWR&M#lIPts`cxqJW
z>x{r{k39XwCnd8hEkbPw)`QG{M8_zn+K_~r>O)Qbz8`=d8OW{%r>kar%k*a}+p_)H
zj<fCmH7%Ew)O6G3rs}gCm=U$xs&LSGuKSivORO`qyBle|N{3lRbM8#Mn|vRi!y~63
z42y2|(z6z;x}MUWKYvHGTv0W1H`kw3b2Xr}84scbyScJH!p3Rujmjc_ji)RJ*vGxF
zA9Gu{-k_fIr3-N-Xk9LQ_Sx!;?V7eyW(tjFQzkjQCd2R-4EZhBJ7J}XI#5Uhh9(6+
z)05&zxjG@c2<6A)!djPI6PvGjws+I$3>U?kOIr;HTn7EfRs%IBcz?~ChY{o1-q!WI
zI9q}2c||({k;`hBLtV^&m!au$)MX!#+W~ycaG@?;nil^^7oR?l=;T@YN*YsM=q`1h
z+J>7zS7@8TItBA2Ovs;2yIc`?>QlO&MV$gr`u5blVsayIYN%?fOT0f2hQP5o#}$o`
z=}aPT_s?5*&quQfIxe21+FxQ%H=vp0d4_P5N;VX2R#a9aIA}P3`(ix%OtmSansf^v
zvcY6-lCdN-`|>r_%lV@W5?iudxD13rQ<~{x7(>83;C#=MDyXd7I?|<Hnbb!?TC5-N
z9=rn2jmBV*Tw*}<iztAd?;KYu7%9DA#G>y{>@wUR#&K8T2u*)U9}8DVwX#M!r{qEE
zGkd~$s=iK!h%^p=vv4KhMnM~>W53=U(jabXa09>D%5E3%ov(S;Qfrs;w(iUp*lt&S
zPeIGS`Q$_Y8Zn<(8rsS?=xuv5sF0d>)Zufj8Cn}C{O@a4Pa`kt>yC}gUppFG<Pny)
zPSkZ;^R<`Wx0+x@DUdR{jW`Vl27|TTV=<w_+fH~O5VJ9Vt<TW!@c$j)4LhG137!AL
zr3|Bo#C~SEEZIP_a}L<WO)v+U`*PO4Na3OJdOABDjqq}nPGLN`Hnm;DevtOanGA6y
znev}Qh00~#tr@cFwK!4LY1?(e1qa;4%?p-=tt7{>^~|)3y8n%;4iT~jXStu5C*=;w
z>6H{IE-mytgd`~jC3Cnz6|1tPyZbFs=B1UkCvN%U53IB!C>O7JfixgI>7L)Ez7AS>
zV~_aKiM!I_lU4U0z)C!Rmoerk6qoVNClr^+&@(BQKMN|K0r!{R3o1B&%-Edcm$Ozl
zX#7Ih5Q#VbzU;Oqc+9N<pICOPyi9W;?ad=l?`HbcMfBxav{u(A`=E*MQ0Ci=Lw<*{
zk;jUE$U7kZNiss)5cDDi;1!HD%P%Oa_2`czy`Z-Erb9%L&TN~A8`WT20H{SGMwlTr
zQvH2_=gL)x>6m-S^VR!*LSA&VDgAwzo}Qj}KkrtuS58DE=F_(wy#S+xK}KkN8}0JT
z-<Qw3k2BOll@t<}<R2?D8;6k}i?v3H#7l=%4%&yr2aCiT5DJ;VY<yqXc+A8R`C3iv
zOmbO61c`QUR}crmpl<9lGs)`Zp|A}l-hL7(!w8bH!dn}mjpJQ^k`F=6Q;+4Igktg9
z7w>S<_(112+s|3EMeXdA$w221a9?}qEvi|z-{4eJvt*yKJ{4a>@K{D*i_EH@ppj(5
zA=_kv-XOAF9?ILt+u5hr!o(FP^s(Uc*tnrLa5Of*M|v+ZVq-*rFz+UhH>1axfOFzX
zvV^nA*A3}GSqdG0^#(niL&~skMj>b|tW_(C#iAek>#4A4HVS5luXU4vfm(x~-K3Hr
zoD9E;pR_LA*iX(7H~i%FHj~%c{R4$(d3ckHwbl<V))W_OCO3JWzh{|itU$kXOXf}x
zH?ji(gdqPUyoZsVBghGdOq!so5vMxaDG{ZgXOm!F>ol%^z*TD{z{Lx3)!GWLVGCD#
zc)JE^5C+qmK1j<<VR9_8a6xs!NEa@YB-Ke(QuASE$OQ$trnJhHXIsJ|Wc=+@abv}h
z(zq|>h9|dL5|Lxk?Q|vMagsLI_r>}S9I+Xt8aWxYeVy;R%-A)rR}JP0Mp7n-uGK0d
zps!P2W*_5!m`P*2%oE{~nb1bl2(e7Q9!nOt)B?Kr(IOS|%r_B?SWuNx@UcZn*a3uz
zQeWvQ)emKpIcYBna7O4j{Y^4aRwx-4h-hdSg;=>|=OV)^GEu_rK;v(~EDD3AE;2<}
zZO%IGt#FC)caFXW{h14fqVLZOPW-T>*$BYIV3+@YEHtbr&v4Fp{5OD@5URU(|N8Rz
z&!7IL1TOeDt-#=osfmhV`IaA%iP0Jl%h&RZ9z$vj>f}R>gC#nlQGBK@tOK8x9iV+E
zgQ-p%{Ys+|C4*T_w>YH?o+k1J-4NqI5bG%K_q88ntlAcdriunV5tOo)9&tUdqDQ*=
zXP%0GtNyX<J}<kEm7fulTavQ#(rvL5^-p=DdY1AKbv7Jj;e{R1qiLKN?Fp@e8b$5^
z8K;BrsIX?KyH#R&q$8Mz5Eet=A^xAoPcUGjHM68dW32zx5;P|Mmb6k8wr1A0`%mA>
zx-YouL-(KI+$pRf!Mf!R0MK9p!Ahy14Vxu@UYX9bxe$1KF!sKcWXa`?WfkwdWn>dj
zMB(Gop_pmCYs7xMm_bSj@681anv;UY7~}H|Gv-9?%QRr>p?&M8l{60Bk2L1xUKnW9
z0zX5Lk7rf^>d<Q6ROLp^@g^v<*PU4A{#Ke<E9|<i&&7FNs^0UA!?4es3z!zKz>l$i
zYiqxu7tr3@oC^0bF6|VDnYp`mr|!q}h`Ol`+<oqx*1n)0DabJx{q;5$C98nP7kI@{
z7`pj=H)4~cr8m0LAc+x7*O<yMYmd^kg1{EOQg(sD-^kPQC}kGq&b9Q?a?VhaMV6qg
z&`!ycq+a$b8l?T)&*>l>#%}2#$GYBsbzCj5*U&D;K8#%ghN#N=kzf0ktJZsa8jDoC
z6nd(T)-0{!dN?Uz6FufA%~oZ>b>HQ*^#8t-^3aX0oVWE!b{2Rjmhd8+S`M9Z^%|T|
zeuZtdB({F=y)S^;Z-*v-Wgg$RVsG#Sd44a~*w2{1Gx%a{mJ3E%VH`USur1nu`fb9r
zHOZ2>K#~*h9(&AB_`gk=28({US?wWX4AK{G%-WLkcz#nUFL*GeAgevGU)z^wrkp5w
zD%F*GF-B+$LMfPWtY2At@@UNGz$4XPvaj`qWt-n0J6)T(o?*_GGI2h2^n#siHSe>=
zC)p}=x|8Si9^322wQLugT&om+oQ;jOi@K6`$qLwpdu-K7FVS_Wa~Iri(Y5up={tNw
z(pCg~HB;i6<iU5nH{Gl##h(tFGo4T!s9MzXPT!%6i&c#<enk>@lGvnOdj{PZVG1di
zwcN2Dr2KLw=`aQ-HVN8k>OO7czg?!emH)O}Hs`<pOusKRl3N#9;&KRom?5fF1KD`0
z;@;jsZupj(`K~LPFX%2CkbSH51`W^&;&qyRv#)dp>q$&VzUx=Ix3`~-q$qz&<_`d`
zGePd)zlrZUIssB`_NH#%zSkb>uwZV;zzO1OoY*z({YbVN=3P&<_oEdN%sjib%wI1x
zXN1-8|91IX+vVD7VbKeJ|K1zs8c^}1`=4)^=eD#qXx!jeZRy{tS)SYY8nJUBUgRpb
z4}@R|83kxff?OX6Q6AwNJs5Rl%CUgJ{qDODep5{}cafmhjbvslKzPC`vwkbS=+qf~
z(hMy^nDsfcxodF_tD4XESnW;4iq9{^zJA|4(bm_a(-}6d6YI`@$2O;{qVn9;*_$oc
zXHM9KQ$2`U{kclJ>#N)j@|qrwj86{+?)x)9YiyWbX-$6~phA7a@iPzx<IR*62Q6Iu
zRSs1s<Akx-9I9wLq`S5;kMtH^l8Jd2Hnqyc*Y8ly1t=^h>9a~#39SywoJe@QOfBpQ
z0@qJTZTl9@Mk@<{y3~>1bIF{N68TW;aNoj2%~P&8S`d_w^d)CR76Ie6eY_wV;Uml|
z(ivXUo#~Z1`01(NfEkwaq%N3gb@_BxuhdT2v2-l<g(^=TUzMLk)iBaiCFu*7n|C3;
zn=_uEOKn_DXSky3lMX)9V^44?2vnFA1}vfT3VG7dSV1L!R@dzd3F`o-Ue_pWz>*n5
z%LMUu3j3fo-TdnmmKi%D`+~yKnc$%Gykbz)kV&A!S#^y5uP;0diAT+wNjk+9x6wm)
zK_9ATdG5RKFZ$GJban@)Svyxc#r`5@)9q}~cQU_mlUG}xD-_$8EU_lhTu+9&zH}Z7
zZ0N^Z2~pQ<AE&Pten`<R<GoJW2PUrYve!^Pc+67CqNBF$+*NI4mrq3rZ5Th^oN`SD
zgvRW31X%Rt-4WD+#jy970uMnAunhD4qCfuWhq3~;S6EzOh82JK;o7i#&I_XPxNR4B
zL7qSVV?y;`E6qb5moerk6qgYy1{4A^GM6F$IVYFI6)Kf~5Nt4~|9<@Yi6>+^bK3v)
zxAE!y<LQs55p{)}X%Nf*et+^4QUDVc7f_B#j?16FEWbYeX9U}F#sYDMYpqnr86%lC
zuxRa=@Kcd9lCV}PT^R5D94TN%&<2ox{`9+bBj0{1$kR#&&WzBxiae8NztibibI&Qi
zh2Kk>xWBxA`Wf^yuk=KcP`Zpj!poEqomOAz?h=GDD|MyAUt!jdnY7~KW$Y2?q?G9p
z$`JEcWLy$MIK9P`|Dk_@A>9Zu<j1SyQ}%nnN-SDgCP-#2`ymj}qO&Q)3`H6yg~?Eq
zNW@+e!!oh(DPe*m<FR(!g;@3s$mL7ozu)j};P54X%cRKKMTrJ<Y@YFwLZ1I^$O5fm
zbgj^++OmrUT{y$Rvt_D>5Tt#q#D!+k*dlR)F5P|7Hk?V&BZN-cypk43%`A}d?#cCI
zb8;Ixr6(x=1UkEs*3`)n8RbT-2(TCFRFDRlLR_tI(Fq5qtfON^<sEJu_w)p3jmhbH
zha2sGWaZRQF$L}~vIpdr8Lr1&T6>s^q}V{~v23HG7*T?^qS5EOtwVtF8nZ&GTE~{w
zpO05@kWB6Qw0kyM0p@}cHD9rv0Yb8g?`|`|P>smkG8ruyWfQ9v$C-?B$g~uHvB?a(
zG$zwa;&D^)&K~3g=E9i;Tp?7<Wz-BW{kX?};pU>OkFWAbdy$kg3d$Wx>kX=1$jVDB
z{{+b0$hRAI<@u=OqI?w4#pcCLxn=2w5#5IGpmt{EI3LBfVn=zKCJsx>EU)>|q3MdZ
z2v8~6w=taz<%3D%=QRGhH~w`x1=S34Co?!E3<ZocwGUI8#xGx|7&Wuj&HedtUL;_D
z4G&N@9luowmiVRYLX{mNdCN&5D=7&ET?*M+rusWMQ<Cya^!F=M^7o|hqEu0_7B1oM
zTI(kv?M6LFChG~=tGT5XOcc+GrdklyuP-UKqJI)ea`FP=40+QNV(YVX`mgi)7@Eq{
zIU)F(A<hcJgnnL=N!PU`ovfW`%!oyQdL!ZuIu)udTYaJFql<86EO}!V6IWd-L>VXv
zeo_{^CZ%%1Lq;g(rNd>UGz+=z5u+F?YtiQ^bXk$DZ#7yY!Ak6u$p=g9#jbWVk;@@{
zo9mP}b0ky-;ID~$*AgXGN=TWOFeSEnt%9soS);tNqsn~CdVqS1-(jYed(h&4_<n^D
zO=Qk1<chM75jkR0$<(|-3v9YmmpC8pZ%uZW-!t(!bhc!uTSF3<5<xUKO;!Mfb|+Tc
z8Zu#NV?#^^*ML!PGv(Pj?3x$JP9xuPR-JW}X9a23H5`?-TnW~DDKk4cMlUDnvI9#t
z!W+68%jO$;B*-b(vMQ{En=+<<yq1A<<H`eE^DRjt6$qR{$B3_OC{La>apPIU(cYF)
z8#H7|Ha}&FW-E&RHpNhw^9_MXTk^uJGcvSPTb+?Id$DSx`Mtm&Q+1$|o2!+W*Nq5g
ztu?)0P5bX?P{QdU4JtUxl51z2CO+T1Y~kj8{$yRQIY~m&;D&uvUS?^3CP<BcOCy6n
zRGE_|#yEoULNU1n{x}nN7i$}zY5;eXyOoS(n+rt~TQ3~-PV_n#wDMa0$A<0+2TVCv
zi{+D0E4`;(;;!c;v*eF5{1CI_o7v;*s*Qoirr?Qzp0Eq$lrgeC*zWfQl~y>m`%zow
z9NcdAr-Cq7g|$S?Bj3`0k-$bhnrb=W_p&9*`YI{wMUWWLjBsDqCDow=#7R_rw1Wa#
zQyz>X<c;=8ZgLE7i<xDe<I#|(6kp?ip&#$?c)zFJxFf*4y@QJC^cCm#VnsNd@yWt_
z+l#m<{$1J2FnOfxdx$mbs@=8{cWlk6cDro{&`Ml~S}or}D^X;BVs8sd52hRWb_YND
zT~X97L$1|>S1CSjL!(Jka0P9XYqb_We_odd%wc734?z`@p}-L;h`EIbOL>G@(R;A=
zQ}%?1I=&9zelhf2^d2q66u<+?93Zy2TA>l<`SyOt{ZzvKj2(}(j1u{VGhfR#PIgsg
z$yhqUTfD^G1fBAKrvfv2P`89+i)_10vySiAWG%^VggR2zJ;aqiw@JrEldq?oP33ur
zJSfLc*H8PT0h@t?V|*RnKGLR8GX?fYKKBqqe%c*ge>!IxAn)W%BOW)85x>ySO?(8!
zl*&1~(<0l3c&#xq=ufGba{o41UiWbc-u$>kp2v?%Wd1;Z#O70qdEJ)o<wKKPOBS4T
zhrtS`pcOLvy|(w<)@*gP?}nsz#C=gjii{UO*NW#eFxH%c{xh7e@<J+N3~3lA-6^)P
z;wsEKlHF%J>QcQ*i19@lxr0#O%N>M-YtEzU$OGMjMBPsJVDjP7?K7^fd$8R(XefY=
z8xsv&<CMF923vZ+?rkM$ffG#2TA@!m3K{xLI|qAH+uejjU$UREqcB2jxj5&*4J;<q
z!{~pBWW}5@%Jv?eGI<S+4*&44f5w)Zl=B712!(dRKk|Z`JG6YibB}IG+~vqqT(HQr
z#o~SIxnJ=Gm?B$fRu?a$eFrdi1Cy4^_c+$Ch^_5^aPoS+Z$9XVN@sJp@4I5gaf0dL
zOj18t50tv?-kKx3;lqI@C5WZpFfs{dV1Uw&+MEbqn5GNky$M}kZVX3fKLp`V;jK9{
zT6Tx|rs3LqGHEq%^6l$8OUEoP37U9MN1N%A{O|-hI6-R;Oxy;DfA3-lKr@H#W-)Zv
zk<)l9-W1d!d<Xqd{j^s+MYA>=Ut);6PS|fBTiZ_eWlX<1u<8E(@$`Sp4PIxLG3F^0
z0Xdi9z$Fuxn;t5`f4c8}yZd<~va*5)OOJHx-`?MCK5cr{wb5u-VE0cO`k)nX<t3BS
za4?pA{)0ViZhNtHGJ#{#j}h$iBkL%?WbXxm-V@duGxHt@oLVz$16TAZc~9R6VbD5#
z#0#z0=qm{`E8iVGtV+}Z<6wG;M&N}$)aXwgB`0ZIlb+=#f2u#Dp!!v^wH&PMB}*~W
zt8zQdt%IIhE%ZRbKIIBoDzDQs#`#kHqy(;Q=w}32)sLd<<ihimX9krtF6l#li}#zP
zuOR}y!~3k?`S72Oy>?tHH!}8fiN<E#$tIcK4|xWnL;6N>Fl+HSjI$NJZXjG>D_Z1G
zJSYY<QJi%~fBEGF{+Q1yDZU}iM)52&Yg2r3BBMPIav|-c*N(fb6;!9p{`I-sX%36^
zc9B%2xm2>33Tl0YD9^Qb3wsA=zfjx?IWqg_63v+vE%bAlKh##^agX$Rpsk%mv(GC$
zXZIIW`A+hgD&Ol-=g$`U?Bjh?O^jmGPt<8fX>}_8e{%wB_Egzv?}tVH4vCcZ+>4Pl
zxszzzw4c<Bgj|zKrpipqH9OMtchZJvDRXyl^-S4*g`*XojI?hz#d?T4q)nU-b`ocS
zn{=rc8?-BDy)}F2+&CFhDkx?jKFN4$93N{iJgAYOo(Wv+zCPC@H93svwELLy9g+v?
znmN9De+QW>)48ud49%Gb?)y4uE?{Jp?<DqKj+KYJd3}+W>yF*m(Xrlel8Z3%wBJd5
z+to!|z|)(3wz|;$Cbf*Zw0<&I8JXM5T!G?+e9}=RippMAkj_=3C~_)Da$b<sbIoTB
z-}BA0mYQQvTQ^%ug0S04Pi?R%isI}t2-4+!e?GLw1TWNRz`ht!sfQ-vO_iyp<8N_>
zwoMx9%$;PEqI}Ft!Y{ryMJ>*j<r-VL^mG)6Uk^=4u2*OBi=Cs`U&e(p+YVfAa#!}7
zY;#=5A#=LMisX^@_;&!cm&qe*=@OU5k+d=E@hRh~#wDRjYr&=QUQ*G<%d4_-7SyS^
ze>t5yps*0u)ufBACa+c7Y&|DHfOA=DvGQOe(9IO%D+$gakLn!LnprkQKY=|pH<SIR
zY?4y)urP@VLd1m*>Ad&)f)Q8pyk=y5)-6}qFu~IBu3C^QvGAn8UWh$`!yQ+c?}<IB
zKxYRSxEpm62Vzg1<TG7KZ`nmlV;nAJe*hPDugob}%3!lC7D0ItEVWM!>)@tUS%LvZ
zo{u|;ZwQ`J+HzyfH-@LWS-%jT>ZE;Dc&gWY3_R6IJp)hm_72W|Jek?46CRQMbBSLD
zPmQxvNbZI%@|nQ5-H^xt+&TDvz#Y;CypIfZC-FN29-TI45RZ;~;EIyk@t{UFfBBun
zfnaYZ`OLAZ*W>Gey;521iS0K4dnFi-R*a0)T;dM}PC-x{^vHIgllZpx0`QXK^&{*B
zYds-0nSF~HEE;C1!~DdM?*og>I9>&B*uRX(K!Ag_@#LyBKp<yr-u?af@YDV8c#dU{
z|Ii3*P_dPP;Mr4p#csvhAQk(Vf8RZq*wT7D`vx?W{=COQ0y30=m2l5CS~@%@8{ber
z5;fq#PD|Mgj%~3+#yUYmDmswFN-gt!Y>7pyX^Lo5RfFD0N*+tEIKGUcSGwBw!iuB*
zWSa-Jxvl;T3jf7+K-e{R;&ug2LF>{&QdnSJ(6+cEI83e+)Yi~Sl-zo&f6;N$bBCEj
z%}AcD8e1rh!G(mfiiAQc_nv;j0iCVMB`Gbj!3{Xhau<^lvR9L5r|0{pJnRt%eQ54U
z%og*4Ys~8{TE{I!SrP5P<X8i-GJD}J>-&1A%vA%1!VH$A1y^QXTVkD1RN<C2Dkf96
z)z;e=y9a<9tVZYk({Iv^e_sSPNKoB!0^2JVNL=_$CudR((hyoa)9hr#!hObG6PSZs
zKEo@qel@6W8eh@BL#lM~zkEs8Al;d-!>Tj&g;;gwK7mzdONx#@p_6$`!P4DK$||N^
zInLnOE848!*;%U#@$BsHjAz?9o}E&*i*8DVH9YPN)J?~Z4DcS3TKO5!Evg+V<pgXK
zO`o`$6sntFF&AcYK@M~45F9P9TJ)}rc;3-7C_2gOfa#)F`IAc4Ll>?5Rdw)8lPoRB
zC8W1jC+d=1D-+Y2Pj%w2_*?h;eE$!Y6jy$iab74B0XCN*b|Dm(!DKurm+%k-1eg9L
zDwBU1pgjKb?e9CDiT!laZt+YB!Oe_PV--WVh0;a`K2MJi_`&yyaAop?h_p$MLSi(|
zQ8-nqz=*<`+U7^6845qOOJiwuB9RI0oLD$h_&BNM@)r`l9{uuqR^Ubp7CB68pd2bC
z>W~DcWqKEx=1NVV(x>@}zfOP^lXc8^F=BsG5emqI$dX9>JqQt`p$i=W-;s`LQP$#f
zD!*JPy#gn5msV96rr@k2I&xK3sRXa`e#}(Zq<+$?tbT#U$bq|!zH^jB{=;tQM-fyw
zQeiTrQ<0e6lf4=BFVY?NYo!q?Xi_!7NaZn@TZH$N;aI8%#TbJ&M2>*abIkI9ggJjm
zN{J|`vJ0-zRN)*OjL45`dk)WYEfbP~_KQIW;&n001J~W`>nz8?)N-U?7KeIFNwOl-
z>$8LfkpW*>NOLtCHkx_TUZATHAHY+ODQT)pJXNo7?D>lt5q^`m@n!`B?mmLI4pZ~O
zJbAdtW6;Dy=Ic0kIyKYh+jvG#2u^=~1kW9&<V95ChkCI1{r}7}QY7&619<K*C7Nl>
z5ZO10=Bu>5D)eiPn4F%5{zDD@vbbCz%0>++zenfVQUJ(KJGybQ($rZ#6R*pNIECye
zvT!ckBSGl2szQjT-k@W^F=|i3kI@Z7{w*d1IrK!)YGOXThY-We6s9JrMRk9WDhKt1
zhq}}fC{bDkFQL{{nZz&^s1th<@?=xs8a#9Qm2rig5QR3SD&%4Ybsvz-K%tF1Jms*Q
zR?)DXpwg6K6%gyXLmp|=@VqWSJT$60*hTWsLjq7F<ZHlLX@POf`|1jd^kO<z5{4tu
zUfe;&?Y`-q;%-kkQbe+<?Sg-*>xj;#{oxg2aYIj84k4N1Bt3%y#}TKUjJZO;i2ksi
zAJVVERF!|UK}W)O+|lb3>$w|}8t14{#Kj9m&&!*MM`d_!d8Ep7qA1MZmFA|cz%njI
zg#1XQQ#2<@A!!wn?-P}#ej>i=KpbV;1>dxIl^P_gnrfX!LRIxW$%}uiZ9k=e)x%{?
z%J$IBQ6&u__Nj_KodyqEq2j)mx0--fh=a8&Q|C^7QS&O8)=qhro<Yr}WzmAHzEX)v
zHIhf`+%i&iCI~>ujhr)Tn5k<>H4+8xHnF<`b<5=qL8{T!1thU=6~<+uPo52Z6n{^z
z*px!0uDHYZE7t-mJKlff+8T^Wp2#}~lY;L?W>S<vSj>eqt+ivL;+UL}Fny&|Yem&e
zE7swY{gH`YprvMO7;L!?Aq-li8VSp~{PMO^XKAtKF=qMtbNA<hp6PXo?g<^I^q3I+
zcZtIcB>thVv_@G3!K60M1XsHI^vCyafBX6yFkJSZFqJPVRGWY3xPInWiF69AtA6Bn
z!%!w3X<#1NL}DGL@#stH!aDGE*+otyy6ZOjBaC7;=P;imizP7WY?xROkfB0DfY?Se
zz8`e3QdLnzqAMBUEkMa@!z->YtHNt|=zE!ptNy+0zAd}YS*qF?V?zXTLTdap*5Tew
zqzVjUf&?nrG8BI{TqsaQD&C@sZH!WW+f;^IiB4Pz$1JA4a;exV&t-)0U4{@=fl!7j
zX~X!290p!-q$;Uq8n;m@W>GGvqL8JVB;3A#{h8N&$5pT7Gr@TXHH*?_!Ek37DlmeI
ztxK#5HY%JjJ5MggUEH0_aUtO-!JOJwzaEi&UI;CTsfT~Y)#r;P0l+kF*?CEf&6Fkq
z=_J#liD0|P3)9Q&A_M)X)V{^&7&21+*PQsZc~`}0kP#zRIIjskU{|3h2aB=t0=vo{
zLXD8Wok;HEY7Sg7hp}3EBE>bIHnSuc=C~Vq>jG`gqf^xaUCmQDBcX4dfzQf&%4vFY
z*M@U&7leQF9AVzE2BLT~b7-oa%;EQ?pO)j*0|{6coIXe$dit{2@FN(%Pa;;%m;Prh
zm~qtetP1JdysK2_xh(vW?F~6a$a1t4q_E8ENWUV^wGHz`(9d)&&6fy`7j9=>krjUv
z;ya~$!?~S4Q^--)8+eKmr!2S>&3!qaW`;tqX3&2jJ?GQ(9Me0-t?SPuNS7NXy)c)G
ztEA|qG18m5m^MaQ?;2GyXny#39p2Cs{#=m>5<ipo4?m>rLoDD06V@+({cHKQ`#gfJ
ztu-EfefXz&@~hp*mwYhiIhqDHCgg#{Gyq+%R_Ky2zcH<o_)N6%n4}|M=;;*<@moV&
zMyP)?SD4OVQZr)FWIy5_4;69Dx7Rddp=mH$NbV4sss-oz^eF9y3IvsZk<M;}(urOP
zP~rn*F?ZwO0dpER&Xy}WlD>w6&cJ18%7yRJTuq;l4zo}@VjSFUosHUW6|Cr+Su^QV
z1ZzMG=<F#=TzJd0(p`_eSi2x)^@-L_*;;>RQaNS`#Jm?mZdf#_8=*T?C~<~qS5F&-
z%)U*LmXo3uy0L?rd9{MJNiX_4!y$Wk`Ev9A4fsF}mURWhyvHNN=Q|)8=F!EB5cOQg
zG+-JxPDaUZs*u&p2)Cs|o!h5J7rJlhtD;u7ZUmMeQjFAWF<G{)pB=fp-pLe^4XA(l
z?K`G*wC@q3%x4yJ-516&e8yE<+h!Cq#&24_%RWVZ%(nH^oZGg3_MO`{pgt#L%s62s
zzIz;>mZS^~eotN0SDKT@6=vfw06p*iVvk*w5pjn@Eb{3H82sT%W4p<UUys#dDQ@k^
z_RAcr(+r<s@~!-O1me}m?}q+*$1r~<zMo&o(L3vrUpTEYo2!6gE+++CwsjKAK5a)K
z-eU$K?&)7{2R+ma4MG1(L+B|uX9@l6drV=#%ip(!e)he_FyKC84R`Zg&0)a%i|pZ|
z9jpd1XxtqZfmNgRqRWUkkC40L<Fz#)B5y~C=dGdME<F57i|DC1XA%AE_Y!|m5t@$k
z<H}3xmh%<xX5F1DV!=Igw$^5~D`5vc)Zx6P{EL{nkuUF5%ele_v>y+8brbZ^*EA`x
zD}FWp6+UN~*M>H3V@F}7j}?FF4&9DdGvIy!8|oKJFF&BpWE@x*by``u&7+tDjK6F0
zyR&uXs%E=^QezmG-8BvLeHVXH3j8A}#R`L5ty>|4fDjsKn|bj~$dzIs;0*T0?iS`*
z_v~E9IDAqirX6Dm`b>vbf>~WdFjqI+O?Nb5=Mc|a*-EsQu5(&5>pHhz%FW&Q{bkOJ
z?QSjfg?ocWYr=YPt;s!v)e(i+lb$S)ey!={k?#$FT^*j73%l_vu4Y?ZWRGUbVZ2RM
zuMj}Hqm<5Lw(lv?)28mC#m0S=OK(q+a~yJeDxcL{aC_>)+cTFZ*qgVfEQUK_CnC*W
zYRe#{hnS}h3soo4jpKuOBK17I*81|c`yZNlG<TOV<|z~bFqbiV1tgbX>oW@iG&Yx^
zvm_{gQEwc_4Sx5pn8);GZieKL!yqu=#8sN21q|B-+6sD*l|$lI)fv(maQ}Tj&S~Yl
z-l>wMwVfK4G`lyu!zDj{{5X3}yrz<xIHk&*S}Qe?_ZYM!I?iZ{vltjAa#F=XZ*hW>
z5>B?t$;RBOsjEN1prqbC&eT*3QH)xuwF$0&k3k>TVO63r&9T15O)<BM#d_5+>TxOI
zCR!!Vb&MKe1(H<F#ub`s-(1TW4eN1JDin8NN>qoH_(hfDQ^rG*!Ft6shOKIhfs89V
zMqaQ2YofS0Mwq4?mcU1Z^<yN7tj9y+mgF#24GVZHajIeDF%5@YDQ?_aVyuCYRE#`-
z+$5A(hTq{>ZW?BaB(@TF5nR#7vZ9H7Jfc{VG9D3D2#inKD+U@~Y@CE^JT?3*+&GqS
zj65#I6p2S~%_L$w7VAl@#1e?AbdIN0+}JG)h?v$ACM$l?@LgEJHm=YC6RZ*8+T1Tb
zP+Y445UVi_u5H%FO>n(=Tm_Sxk0--_JT03W!-K_R9-^u=d}f-IHNp+ws*G!`BW{~d
z9X{7_7r?TQhwK2D^Ekv{;G>shrG|CAC#x2oLTro%h`VCsiX|NL1R%ln7-%>ED77}+
zgQZQw3(Y;54fn)6i$`e)h<S}T!U*#g{wO{55-?Xpq*sq%;PE;|lobK)j2mEogii;|
z6bp5Yb>WfJE%2l8e)r9%EqQ4kBTZgC)&O*!1A1;4wDRP;?-sBB{cgE<c5$&<FJ8TW
zvyP)by}kHnT718{{CRnKT#)ePWbuc^Pm3Rp>*nC)@@#!vi8H|(ZJbMuv}w5%4{v9c
zcCvVuaQjz_A6Ktei|6SZHsFeXJ4_TLeYf7^cTe$e*NDUGmV}hPUYAvKiEO-Hm)$eh
zezWeAo#$9QySe?@`t9lh2L0vbPx153uj}=@t8W*JU*4{Neg7t(t^T$6<NNcsr_WE<
zr;FF8Z_bxj`5))7@yVwf2{y?B=rt)hD3Q2jqrgMS#qIMM`TQ75FwA;?x%lhf{thzL
z43Ge`HC?<vKR>x;mgS&C?Un<?chBVBa<JRoGf5;jYFjhY;)m7w>hjgQ)3aq77o^3D
z)Af3Jd68&y`1=p*A78Cc*UMBlCyrJZYpihuLaYTz8><{OuzLk{j{_uDCE2WhgxstR
z8HIzVPZuvPS7)!5>*K|Li|0p+*UJy<ldXWhIQ?Y_gD%#~i}h8MjLpqoF0WSaFVB`&
zo4OOH{=EG8?dkWc4<qV|lSsJ}Tz`oxBd}%68UVT46t}VbljBj};xR^j!{FpmiAS~M
z$Z=3cM(vA&kt?`Lt7YT9mv|MWQKMvZ24A@<_XCOdsoBp6E26=Fq8Us^O|pS<xCPGZ
zo()}41iH2O&r(rI%Z&{Ioca+ss166vS`VOAIiqNO4rJ|sX$NMHKo*sKWc7z2i)ag3
z_IN-s(mu)uBUOi30H9F);oFS5xED(xiAY6Uin&8OS>KJN-$6gKg=s4oJ2Cs@$h5u4
z5Z^yCA0M6Mb{PME=AN4O@kkHCBX8l6X29bHH6h7sh6II3mmFe_YRz2Q0fHmcSLjX0
z2)Y-_z;+NxexHK4Gixe{a*tUA4h+3MPpo(7=JN#J_5#oCL4o&=foHxhaV#?+fNG9L
zsTyPk9E?h*(aS+~u<pfQ32{RIg8mM}+Sz1=0tb2PXh9->$k9B7j;VNHG^JS&RJe-H
z1@JRMYg^Dl6?A1gAY{Fx#BEKuK^B#)U!cV8AZrVNpNgzj_mCz25M;IX5m{2NktH+@
zCaj4}gEkFx8q7gNgn}6THX-KtTMaD_dMBhy7NLC%NQfNtljK~u5UZ3qkhrlzkOd75
zRz<lMw53viJF?*b60t5vBVN$8cjoFkLp~7OP@9t#%%KkGRx2VKA^H#yLFXE@;4`?w
zzzdQ<@=(Aw(xap0WsthrzEUxCn(Vz;2{*vnHiIKrTCNS%sNk<}!cgKk`ro$i2!cKb
zfDDxjJciER+6>X?l<||Q1Cg->1l?Vc(p1nCH{LgYqKc?$0Tk2$<!(HK5l9P(O}SoS
zOiv8fc0i<h1yuHNX%LK^6zGBJ&Csy#mg!dRQ>J%jZD>-Em$5N&kC}p=q0l{M5_^t4
zx&N2G)%Mk#K1j`(Y*BB&W>cLwSSJqFiGy|GV4XNvb%rJvDXy`zH|_EeA#i-_?ot)B
zDrQE1=&4AfzvMZnS325SNev3ryrBR{=GcgI%3kbCS#hn0k-F&1V(Xirhe5^Wu!g^s
z6|Lsjm<_V6k@aSoK|xSG*}hyvu8ScQe}>G06<}YqT?C(WKVSs}A(UhTkyuwl8zHkW
zpXJ#1ovdJ3MdMY#VJkjUF(fEc1(%>rLW1Id3DLg8C7@xh`>;@Db1mJ}s)K`@@361>
z9r{%Lz8i)IW}-eoKI?sEHvHY0`P(MmPJ6`L%R`8_Z)sk&Yc92kRBaNrcJt9h&gL`l
zE#ebzD1vVd`NhS-G3|D~jh4%L>0Z$F6eK(840#28$6z>MYQy?08S8*=Q4Pu=rekP-
zn%YY#A^>&aIixN_>UJPwvO>y6-`r{iG#ykwbT-%G$bU$_$6F!M1EUYRu>$-XZ*guh
z0=-ZNMLQHPi!?`nG6&DW|G*`B*&J`EsE5!3I(mWfJFwME9#R`}7(5GLsLz24hX2vi
z-&lc|gXrj1E38z=eL*nLumHfN>JHO?^?Dzjxijm3(%pfXHC{?J`D!G#I6~K4h4lJB
z;=mq6Z@y1&b{~54x=U}qwM*D%P>2Hk(ZN^(G@vDiK7=0x5FtO*6T==fOC2tOB?U*j
zy+`kNRKKkWQ;-`vz}Drf?as^>WZ&Nr^!A`2Z|w*=&$ksUwdAVf+Z-bU1)YC?+24Ta
z75NSZ_6~HA#h&0!lwH5}tBJ>Kuc6facbK$1Bo}q_8uB_!*ZrH8G7ai9plQI<;AV20
z20CAqRXm5jsUx*J#E3`K_riR&9Iwfd>ABAFDie)Oe02npr;h!#3Q-R)_UwnY*Q`;z
zVsHl882aGa+k8fVZ*v8RAZDO{ze4#q`W-#gU6Bh5fq)>uSPILGR-n_Q*wX45ty6pr
zCtAM!uXMnKBHE;wP)Ap>m2Qu>L87~DAB_kw98#jlId91H*l#sKN$@%20gOp}D{TV`
zgdL6EAy8<z0Q>{oLZgtg&9T+!L;i}Q<!+mu#u|OU=5m1jU>LNQ_Ue!>q8+-e&L9D6
zh%NABh^-DDiFTi3|2F{I>DJtAnRecPXt|@pZcjMy{_Q`!;~JMSd`1%kH8wJnG1(IW
zF*uju@)svpTuX0jw-Mg^SE#!N=w&#36EF-&id~?)7Ty(I6-`!IH2;6mAvq+6Jd}JS
z%jwNRmc++!p5F|~Bg?}t2YLAG)9L-`$2V;bUIYtX9o~OF0Cq_H3lbZb3_B_ye{cj~
z4@7*y%QtTTTQttm3dg;@n3%o?ns+Yk8qye?AEQs>0s8@I!|~-`?|*xe#H<w2rH|=t
z{`UI*^xvnMb?v->5Dx$T^(l224MGQX(4a(6{_xv>4nIHreJ0x$lQ^gRvGVZSFNdk+
zAEdn~QG3!kYX|LtM3tI@b5ZEje^s<+ev;Cnbmk3TD7`^lDHx>u@h!lTqLF9^J5#i%
zKa0AuzEHH1-f<-p(ieQ4N_!e>d}&V{$a7hFho~RQX}mDnKT^(-qK$=0qL?WM;@u=B
z>z~Z~(1N@-$__-7ex?i~UwyO_;M6=i>+-~OXZ4}~$?Fd-eEq5`?S%26f31(E0okpR
z-YtE%kSzQ1&B+``kw&><?`(S<YF|;ik#wVqo>G5~LVU;J5`LOu=@!1x<D5P?Wrc;<
z_)c*SDN;W{*$R;H;J^s$8evJ{zasv5SG7?IXY>PIb)uL|l8vXCiTv&E>ahjfOkK&f
zyZRua+VGNH0{JkB567I!e-TVn*oghRJ=6F>ju&{YqFh51G}-IJPz9F~UOf^$CyFVO
zlOHE>YxKys8gE1IDXn?C5u+6m%`vTe!H7y4VJvL8Tsjou!E51BzuwH3GD=`_o>v01
z=F2z{F+I-q$m!*79JDqH?;J<g9QqAcxGhqxkQ~><j=dEW9tjQ_e@h1w9_eu-#cV@M
z!3Gq7Z{@Q*kuH5w8ql9C$Ivo>gb~ieyh;krRAFVW4b7jEAKM}o&m#>0@<{beJ&Hy}
zdYwyqe&6J#JUZ6-#VaX#<V>|`)VChbI3jP)S<DSwCBL6+Uo?|F+;UyjmMb<#`U336
zLelTDuSv0rxoe(%f4YDw=bLS7bP&dDe6k=w(@Bksi<*PSYYr50gxdwHoCqP_s5#I|
zfO4DJrxoJfrMX(^O)Df+k@iCRq+~nsRM!6SEkUJ--rF#u2B3H_%=+_RexNC4Jw*u&
zn{mpYez>)Qx~SWVV^~?mRt5ZiP@r}|0Xz;0=-Ze!I3ITge+tdvNoU$7JR;ecXoLjy
zz+^K~%;99OmN)hx<;}G22j$Jo<=2%r(@J2kmN&CKK3CqDpfS?TzVc>h{krnTATWY<
zkMy{a;un=S(<<Lhd1L<o<;|?S+sd1H_UQtu++W_z9`yZ`H%L~1_ji^z^X$_KaX(ew
zOe^dtZ)TqMf0Z|U*3R-~nDY4Y#^@*%gzu`pv46b!hG(6xsBgfFN1n5d^xUPoruE!e
z(}Z;3J8^&M)UDP(eC)M)Ch<m<@ZEKVbV3{Wd|mMX3+sScLOs&P6U8s8C#F@_^@O(N
zsw-Vhts=kJ@;lA`8aII0xRG>U4%6w#Uv2etTEyM$fB8$BC_t)>ts9=fE<F2+nhDca
z#JqB!))ES&Thwb>yFcq2+|#%iXX;n7KJK-Il3Qu+zq-<13*+qNy!(HNnVnhkX0O_l
zC0BRHA6=<y#5xGD(xSGuW)MT19(hmY)z+%Tb)>abXTH%nvRj^0r_uW9Qney%zF@WW
z8nTT~e-vDsa|5^4_LpYX3*Cw(ncq<YO}UhEsHM-wGMUdb8+NbjnqPp+61=Ev_duGh
znH$IxtJfS81+%ZdOn$Plyb40G3zSxSf<Bp5%~>9!KpvjUV-lGwPqJvkr8A{Q)mqOo
z=W%(hP-OPA%rMO5IagLIwcg4zUBquJP`E<sfAGvMvklAPuF+aOyfCgdcyOh$k^MYS
z-S}g^l>Dai!zgJy_ca@<hi#~aO7T{I0lBt5IHCG863nTvrX$d^p{t;@exz_3MbCX?
zfltMpMpkEU4#GmTQ|F-AdlF30rz*ayblooJZ6nD%wBogp4x~Vo%Y{JC4)y0+UcLcc
ze`+__dMg*!spK^(Oy;8%?Ir2=!YR?2;FR=h*2b1bU=cQldcAzllG`aT(n`P~*A#c-
z62V1@N)JI_p|hQN4RVRXjgY%@-A_QS0rSj~RP%6pZif4Im!ewIuekUKaD9eIbsi)f
z!DeN16_sEoS*yHQu$LoDerS%Pv%7@Oe|hc<5@r5}s9Z*Ugvy1xi>O?}jQFs1rAB=t
z)m3~iBQE8dTclb*U)Smu`dSsYF`u_WLErPAi%TmryEL`vuF+Btyfk#m3a$fdG}E_=
zE5~l$8gCqTlB$)B!_wS@iJqw#Hndt&N-e9K+H^1NXw(%05x15X$b%d#=BVYwe<X6l
zI$VOWXOXP?8Y|OMPb-E7PQ$>R8!P}F&o+-ft;;p#v{1G`4Z<ak=DBO7**n$(BJ-wN
z3p6_o#aBh(Z0nd^Hufz`nTx!YIn|LRx8yL_v^j6%dQG#Ls$S%Dns*T&OXjUmRA}@}
zLA^V&1<jO{$!FV^UT)jMSG6B|e+smryXhry*MAGuXg+|nT2+)s-x?1kpDn!tx7Awq
z>&pQc;!ZFxBQL=p^)~n;4mS`;+k>+8md$fI$SYpplCQXWT%5tp%IZ1!(-zcbM9ZF5
zb>^DPx1gFcJdriMmAEOKp#Rls;LPm1;jwPM3dWAQ8KuoyFVeai%dMrxf0kxm-Fn9@
zR;8=QXkB`WjzXJW@$ZRbQ)_Rf!X4F^QgMSzWG6M~IDgw6S7mhLYRC#_PJh@^0bjLk
zysDR+1}=P~?fZpHjsT!bs`a6Yjv%Cuv##awgIS2|I?S^=7m{q0IW>LbN%BllQXik4
zD_7NaBrPA5`B(Eno43r3fAo~?g%U=2p`+?YONc;uL0noZy$GIeEUKK;c4tj>pK_ZM
zG!}&H6m6mr$Q1e1bYlvol#s~s#Cml<0>Ep=R7MMH;)cTOi?&tDb4j#y3sM}dayq9<
zyLbe@%Y?ct=cia21eFe!@fCDEL*2CY`^>EyYALX7!}(V{wLKkSf3H<(R1`jJ7|L13
z=ynLzGv>nF^&m>w)e%h@yDHgLtN08z-rCQ>4!;iJTMy;0)RlS@ku?4|?ib5AUX9T>
znB2p%>*BVzCjFf2EWQz`yYH4IkJABTS@}Gk6*O5~$AN4WtJo2;4J=%>Y-rl#W|b9j
zW;^P)z4u$?z3Iw!f7*R6*E&4&@b}lytMkV8{ydKlXB6kP{iPNr^csu+mOyF0%T>ad
zTdsurY4o`!C4}Ae%doL+l=t}rU(JTCb%%p@xWNYsG~zZT3Zs;Ct1E|e@o7B_zuk#+
zt&wUu<eYxoR|*Nx$G0v-b^&)4gWEzjJ$cvi&ynv6Q#}Q9#hm?j`+r-K)>CVnHnwR+
z_X7yl<K}TDP6_HfR<)&?v#j1Oc5EE*LkLz&vz3N=Qn8(I0rE?-uo}0?FzN<4b-s7D
z{Az8Qd7Bo_EdJ{b$ns0Dx)yQ^qkj`(t>QCX=ljr4Khu?y3&S}m($k}hWn>7j$|bJF
zgX7~mT}yke%yo^AHh;5S2)WHcm5aMiXDNe>G)4|rMNvf#=U4gA#rv!TMp0t6KBZLU
zd){vVHs4(}W#rQ{7dOfXR~&v}o-<bo3xlOgsDSklFYodKj`9-(c7{V3?zS1z^us;F
z4*}C*UH(0>t1YrJ$ZG6nm%8`d;6YoPm$_;xzg*wXj)$|e7k}D$w9MD<Ia^++;L}5Q
zE^H4WgJbmTK9xUSRJEgm=g}MH8-&t-&B$~D((&d|<t}c11Yn^)$!Hkvc2p;c)Mx_}
z*)Fx>hS^oyrr&*g>*m$lj<{)RxM|=gXTdLZd6ECs{ygy85}I#n`!#gT)n2{{(P}zl
z4Q_@1h1>Z&hkw-$7j7{kZUIynt|=*~rb8LW5}oIF1J=}|9dgg};*Q`^P6fD?2OKi8
zole8@{~S1%Do1_bL)witE8f4$uT)46?f6-5l@NC*%blCw1~CZaT6K9N#LB^(#GoGb
zvb&?D(=@Fg`++#kiTyC9-CC~!DJWt4!zvArO5R$K$$tpq2gWpE(I(-P7LMJ0OpZy{
z&YfI4Hwkn`Qd(A(W9bRg?e_6;IEK}9^m<awDDEEU+6P<d-vy}I4qeOk$o4jl&ee(r
zc)i)`oK~s2&D~OO-8^>Ldg|Ro`F3REJh$oEf(1a@9n|PUjq0WX3KvgUy%df;Pg9-1
zvSY?!dk($*M+o%(`u_Ak=TP`JmvLSw69P6fm+|Ed6PHRyDliStQ}_i|5k>t1!2mnE
zmxo6xDSt1wkISb!>c$Cgz^Ufrdjs@B(qEAD(3-{p2#e5s{A~XCETyt^-RRO%BGTpP
ze@}OFeL?Wl0_Y6{y5RcaG|&^IZ%|RY`&Ebd+vo4z+UE$d4SgSii|v8l1IdQl_0Nyr
zE;1`CMX>STc;mmmd|dvyaMe|iLOa|1_45*6XnzEOUNz{0Fc_L&e>dMRzjLueWIPZu
z(pqCGWIz&Jy0JD0jIUZ`^t+UXG}m{!$TF>4?S-b#x<9{uH0;T@uf5`^1SrcTD_}Ut
z_^FA1WR_SxA&H;vK~CzfC5@)xJw&~YZS^)Z8v3>BaTW<{1z0uGHzPqqNIsGzjj+>B
zXn%Ciz^<134D6Wh5i_#XkhzAu+0YN;*5T*K3K;ABnn%R6619oFOyj7GFc6-BV=bBC
zIKrfEo$!d7Q<UqZ&vreEZ!taOv>r|i>wve$kskEm_7LGMIx6bTsnICFG*(Il5u6|C
zsIu?4-j~w-z12DDC}&0=v@o`VMqgj7RDa_Dq^B?m#<Wn0g2Z%ELvF466w$)fyIEA-
zyp?{EF+*VQ*$`RuGq3%3`nDZ*P%&9EMd5wgGqPDGC}A~rsE2G)F-0?F?RJmVL4Y3h
z#Pr%E(KulZjwIq^OFhllUeCusH&E4^?e-cJ7(DN9ZX7I#ARTX=ctqv{i3`a5EPwL@
zWhb4KJclCB3|UDU!Faz7=dm^M)!|`tE1mD#F;P+rZGh`HgGqK|ZcG=s8J6gtD#L>V
z>Lm01y`x~0P?;c{?XpB9V`<RI(M=dZ0NH%G(<QRQ)_u;pM4RGS>kZSNL1KE^KVkY~
zk{6~w);i@$pPfhz@@7dc(7uaPk$;(*qGqfGBLY6d*h=t-GFeAo0yfZ5rs~PJ2|Q|n
zm|bl)-_FeGUIfL3&0^N4wH-5x0-<df0Wsnmn}nhk3KbJNle`srp}phA&Nu>P8CoiJ
z?sl7Or9BF1^b`DrNj|}0xI}DodE(R1XXfNRy3g;7XLnM4<kKSrmR%?I8GqNW?Gv@r
znY0xlOrUhkn_`kQflCE5?M_w;a<vfx)BdEL5;F8F*@VJ&?@+|Yz;kJH$-nqg-FgS!
zoex!#yo3F`D>FH6iHY2Pk$+KI6X2e)wq=sfh&v|nq`0f4u8cc#R@~LS!~LANV^a0w
zxMK*+#2p5qz*li+_Qzc<ZGS25>eff%u9D=PJRWz<61^_&z=>zDqe|`q5v<ypg_VcY
zn<*d$lWLgfTgfCZ*va)MIiu9U3i<Ud0UqWQ9R~X5{8A%vO(|go7!HZGJ);22BtM~u
z$aRnUhIK{5yzJ%@Uh6eSS;ThfJJ)T7L$jQi(>gSk?spW&j3GcqqklA49Ls5zBfsD{
z$d>G-kR2_PZ=L6C(HqeU#1Xzq&K4grrY))X7YDFI3uy;q9d;Pk(J+Z!*<OiNJasQ6
zm9~Zuy(B(QZG3L$>PXA$d9;eg(pKL3me;jvadKCCD}1cm8TiZMjVoyf8Tf5w9$J}=
z3oHhN(cLUx3ZFeoYJVlJf~z>m38{td#`mlu7aa7;J&BTBRS;H%9cAFjR1H3I(P0hB
zX90OKY=@v*!@D#Q)S?VU84a^^!T}s!tT^GF;?ibG6XgWLi?!ZGnzZoTh(l|YG955O
zmewjQm?~IGyXr6lZNX&H7Qke!_k&prIz&J-nB0(Q0y@2OynnQzt|(1e*JtT==6_Kb
zx5i_lN(st$PawyKPa#~M%CgybRVl}&_Wm<lzPY|dTk|cgQj0<LH%oQ7eBSlmg?5h4
z+)fR3UbkROTPnOCJ(QY$VS9C{rqivJ^gz>23&0z)v>xp?`txAd$@e<uQvblV{{WdP
z(-luKabKW|O@H0x@fQRaJJUIhmSXu^IfF4!w~Z-96M~BFPM1fTx>6X7o__1*g|Yiu
zLvCxCQa#6MBSfvVc$(K;OD$lY+OauoI@CxjVNgv0xWI@@%G!F=BprZ2e$piF?-P|f
ze9|QD96zU$%Oo$X`nVod2xM-f@_?>jTf<#Ct@^=VAAd2MIHjYU?%c@a8B*zuplj56
zrG98~4vYNyOI+Vh(jrB)Xi#80Svt@x_MAc_ll+ABG1onG^WC~Na!b&elyvkXRO?%N
z4)xB3Pp)-4XRXngY!eXNNWz8?d*m_cJC{N%Q4B1n92VC+hhQng5QAw@=iE4$`D_oD
z(wTl-&utLTB=}JeTZXUQ_vaFH9>V*!;hG<pG=zPby!U2xqBj@7>8?!JQ==;Jz$EOf
zQ4L``r#9t5NE}%>SDVUn0c(AX>r4@yYq#`0UdGbVBWZ=lm7XtS#oBcwQtXQZFza~Y
z4ed&p$zdo3jud{EkWMNU0qd8hPAbO+arqZHshXETATkpJH8wPtQIi)Fmo8%%D1W^h
z$&TE*@jhRnA84Azoxq@x?w)P~x!IUoki&x!I|n<+OAh({6qh0?YUxem0W@k+WNoaK
zr1<b-;KS#a<NM2}i&TSUPJ<Q0`}YCx1;_uuvC-(k0x%<Cc>gv0?Ui$3@H*n9IF}MH
zi@zP-^?oP8VGqDJ;PHa}&MLqbiGO3Hz<!s#^5NO#ai`QSA@srV-rLY0@I4@GIPL%b
z{?`|tiT!lZZtzSA!Oa|{#wvzziPB0({w?=A{K*3Wc+-ePUGR>CC?rJ7^sF2bq5>lt
z$<`wGa!pTc(-7WA@~#QcXu%@Ii7c6+YEHO(kIRQQc-Ij}+ruR~<0WzrTYq}!y$Iq{
znbz1_@8hHddL!-OekdGXm4m-Y4r#J=#3WC9km5DJF2YpUNvSY^z{ZwiA~*g%mp8{e
z`cO#VG#and?SvJGsufS6No8THG%m6oe(Q+{f;dHaQdw*?Nl8y96R$WEx>!!6(2S-b
zYB%<AP7ItWyqwf>_z&v)w14$$umA_MswRPHJ3**3iKd%nkp!M5D2vK=rnaEd&N1Ol
zj|pdgRscdU=T&hXey`PQ#TjXsnKA)!nmp-vXc4nwAM14PWl60{%MJR=FW6hkS-c1d
zY=CpjD#KeViDS2#NQBC;doQ_Y?>bYRO(9|znLC>)Bda$z&89ShaeoCpwk`QFhDl?4
z3@X#P5g!U1Vlf}c2gV#`yqL_3uRJ8pyp3LkQW5qe>D8PW?6NBN+h{eOX(MK6C40$k
zS=sHV_Z;P;CeprQ$3=@=wowWi)kR<RZ{)dvGcSFS8|mqVk{r@h3|dlCP>Eim1UaA(
z6Qr())}Ch7Laoq}cYo7|;)~LIJHmpk<~900c&%)!u}fbPvcQQILC9lP=Yz&-q7$4J
zcX9&hkR9=fVtparOvI!~gszR0j#zeW42F`N$tHUUSbTQWC&DyOgX@!}F|RasjPFWv
zl|$h9eG&zcN@hR~-+sOPIiLo99-<nerAY7e@V_B6Lof?t(0^58vjXtJYF<A5`^Wdc
zeES0!4)|{v!3P!Mu6M()^k^J<aLN7fGd<e?6_gx+c}z8NzA*Xp8avRy((r9~QNl8i
zA>fqOpP?0#Y=<td8H@>rbT)J(1huKq5FnOT!uPXnB&sTk$WtLBJOwCuY<R@+G%7rX
zt9_=fIO?Cn%YXOb<x7#Lm-vTwV=?=l5hpMO7Au=Zq|}7&6EqIa((o4NdcLt`vovC`
za3;t|IA(d8j*P3YN-Yh$)P%4I0yWXSZ(m@)<1`5*X{Bb`BZDr8C+H+(p(erB-`{@a
zaX)a>=jNW^>_AASv>A5VVT4d7z>4K07HM;yHFu}pRDTT}vd1Go4U+_!Fq7|YO{@co
zBzzef5)-ex%Iue$1%y?vwYLa}$eIz15~%i7Fbg4vc)(OCq}G%yyQUCCKZ$ksYw>cM
z)eRa=33}8qMp2H<oB2>HG54{|O-PTvW5RsIVAjKzvY7Mfeq>xukj?S}PfHczvKqor
z+BgXlXn!%wiGtT%)h-O=5*Md~p4!SIjampAmMZcb;Kn%C%GunLGMf|OrU05Lp@63I
zXxTt(>W5IZD<14ml#)AB+pIz4ymX?lT7^*6&RLX&oXA4L;tB`7EmfB**(AMDDJbWs
zTyw=%LvPA*1B@^(az0DbjKt&omjwV?tkAHknST<UzVfn{&q89gGdxftc=#pavekOB
zWi3{S6X1tt1>@3+E}}V*x@09v%&%rNBk{v_BJGz7NSU&)Br`4<;C8-FVJ0AYM3K}m
zGjBDOEF@<wjXv(NN>6}dT*wL5l3JEr-qe&g{T@{s$5DE$y6QRC6)Ej?jbB=X&`v$C
z3x8mfRl?+{Zd?Cr<uMndYc!mT(KX(t82zvKACou}K!)4r?&;40gbzCv9Jljfd}vD<
zr@%T`qd!M&`N#16<x3AP6N|w-M)dGc^Q2j$malDe&sqJhN$X>D_n3S$ZVX0|R_m}P
za#~^Th&hO^0q(6jfoQiOM2$sQ(Z^qvDt~HesEWERb&q9f#jz-Dw3I2$C`$;YGKMBH
z%^{>pneFnr<gQ6ZIegia+8-mC9?PjN<24KEH%SL8AO)s}IJc!BZ*VSxjQCI~QKx+y
zLnEPvB1Q5R6fNRpgsl~|-Gms;K)^>&NQY&E2^m8ax$GB*^jn(kB%~x0T;3s}MSt2!
z8e@o25+bh}xw=-zR3MW6McPL#t(>zFcaSm9Xz8jlR9GfM-&;=Tz=mVKPIuEJKm^a*
zAqja_C$2I;Xl8|)BWtAecaR`4IdgVG!vcPcSyoNOGM5E~`&*MVA6HcHc6uc@OzT_e
zl|ey4JVfs)U2xvcOax#|)?1p%7=L0(m}par3Q$|OOJWJ+YEjFNbLN#&{p4Vu(Js=L
zF`bLrWtb24Fz!7c5ys&l8Q?}#OU%Jg<Tp|+!80i)JPO;cvS!L{Et*5^QK&v`reo6k
zxWTlYj&%@w9ix1T3emLEzM@wf@`|OO(s#w>xSkAQxsc-WsO9So;us+nvVWFB#lprr
z=JVkNGfB|{yKahM#}yth!yGm`+F=s@9*G=dpdeQRN5?Sff)2~l^e<u^wX|WJ<{rbj
z^0nS?1an6fb5+BXV%o~ztzu)W&+6D1+KQ4rn7(ufIt{o_Pk|}pClD`-d76-l2`Ty!
zQ62OH;VK27>-g`2sy(xM%72xa=g26AGMr4?6Q&|17UpVBHd~rXI2Py4?>3b&Rv~X$
z!J0Q0CmC4ZL%c0Lq6LN28tc&W!^u)d!IGYku8w;@8X-tWJMPa5I_el+v$*wRuEgn>
zJoUUUSbf~y+K6?@1(R4CtfRg!e@s<J4^c1Vk15++5T@eK89(RDHh*V8b*e8(jn(vw
z1gW(jn@?GUIOmMqrz`Hu9qW;-aCQvdI@#Srw`Rb7oMLNTV+7f-@+@a6Q9;{w{XAZb
znhtzhi;pov-I&m;xSw~aLZgXu@ta+0hc&p=aySy|=4Q-9nq~5CajLpiO!(x4mo2l9
zoHTS8mfU7eg}dI5x_`}_1NXT3s>`0ktsi#TJ?<xy6)Tt)Vgl9Ab7T!#>4g6${FP9e
zc_r8EUW}|TOD|ccv(-BN_Vy^k=GB(#HyuQMWnC%;gO9JP1x7^RTdpB=5T84^*vnu~
zJ9lvh|6FS2<%4?<8N`@{XH3kYD-K7O#t^&mwa<*Y-cOhqcz<A^N@(S8C1+hYk&px5
z+jNg5<hjkh>&gP1;9GAA#o2_{c1tLI+j7&&TU}DQVJ$ciKV7>H=J-UuktZXZI+Y7^
zv3`c=K6GIb50e0mnY%5HTCSeKnE57Rrv#{xGQb8zg`-PQSLbg&ZdK-RURXq773%t7
zsHb?4hDoj39e<g6bRnC_Asso$!u6TMsFfv}HAfS2%fg3HPvzK|X3T}8ud8u-&IL@A
zq5If__7MPxVli`csp!faS91rAj`=f`V5|ChE8~(WYo}%=OnYEI`;A7ED*x*g6TC?G
zqEW>jCVhR!PfyRE$H)Z5kIUI_Vy>BpG%&Uik!!l>++|I&6A^siCABNEA(3q;e5+`m
z<^^M-*1p|5X}-(2!h>@0SWaW-kOPZozqOqE&hlxxEhaeaw&InQC*ov!dTNIPcVjNj
z`KwRJ_wI@r;kiEZ%YFSP(QEcBP<KDSzx*HKkm9J9ab74B0yZ_5Q6ml%mq2MBE0^h7
zDi42EPC3Bb64Y%lhe2VIIV6qlp2Ph8A+;~>c`3USz-2!@i6TEzBt<EF`M&Vw^X~a|
z_u-^qvCL^<<?{M%5%_}RU%|1_=*0@bj1tT1&*dK<IhO`M&-ek(mBNq3-wwa^VXuV4
z9s=J8j~{W^TM*cyaEy}J?{ol9&n}HS#&&-ts1J_!-co<z5*2m$=j-2hJTc3eH}sRf
z#;2Fp-5<M=bwJjU(dF&uj$VKgOgXto<Cyni`Sah)x80YKY?(}}rT+Qz`!ckAP3Sp+
z=y7AT?$8sSDWeu+@S9#m^hjFJ`1yn%loAjbjYt3@p*^)7_J$J%)JtP;jlDGD1NMKg
zN6Z7jpQw{0axVlDL7l#b4^6>e>N7v#H-3rZ^<#h!O1O;61cG)<A-Wj^orFhj05$=5
zN^jL+4^luUI9X0W#ox~ySO&@#`?e8tni@`-8GV)Bh_mnnQMf>-ys6yZrh+>p<hZP1
zF;X)ve3p<`^La`(Mv}c0;7xCNK~8@cduek&-U67ewvZ+ZaCQ1G@bVT!{!C0%CGg{6
zkM6?s_(H-KI}02yp3-;1HsDX#K0q3-7*5)XlbO~knK)thT+luCh;|D=CdAE(&A{Ef
zYA=}qRV+Qvt`jH6%+=kzs{46$7<C*JTtzYy+qsCzQV)9#!q1Cs1drov@sEGzk)f^y
z8TxSYeBH6fMWwZ8))YG?$Dp%MJ4UFaSesR+iR!GmrG)`F23<YP;`4&-C~pQKZJ9xQ
zM=HlS`U-wFQ80!HBWKE_h(T%~a`{h~rEB?mhJTBJT23fT+PpkNq39G7)=v~abrfrF
z3JsI4b&6l)hqBJYvXJY}Gx&cj-WqTm!=oN1ZGmHsmw-fWBrnh6j!*G7e@>dTw30;n
z6A<WHS_C{TEH}<V40gWQjZ1yNJkSpa>vqQ6a>jO~0w$H5IIz_+`&rJ)wPiaJB*W};
zmUC6p5)7t<X6n76A%*!RxW8IM{Vd3Nj&w1aggl6sz_=lp^vv6t*y(>vz7iR}eo;wU
z9@>v{2(iwQ0frG`;^;Z1v-&y4Z(eh+6ve4gycE7y7vaskD|EElPuxhp%DV#31V;Oz
zHz}SUx6MTw=A8el&844(>w;9tT{EKBTg^c8G}GM5hod{3t!s3LhY-k<0!2weAEjQ3
zQ<(Y=f^4bR0yA<Fdr*JxU2!wVb)lIrnHbbP-mQ7l_Wm=aDI8FfVl~~KU3>!no?KI6
zJ}le<o>qCui);Q{ZExJXR_kh8@7jK9T&;cq4EkJ`T=Lv{_ys&gy?~XBdwru`^!<7<
z0{w@1r(Vos^`g<2S$nl!%qmYjSDPNE8|9f4?nd31`*kB7!&rZMQ8(sS=tfwkuGS2s
z{q?$bj%PSqE4{tEPL(_bDwOQre(wHQu;b??^hN^8JH7mG3C(EP_>W?x0rYwJQX6MT
z;N6GczkmDN+rI?EVgIGBe8DVK3-6Yn=~1EbrMAnD^lTW4jdKd-(FV{g93lhQArCAK
zZ_5s(#o4uMT7Q2~t7qUq%4TpVLR#vWCuo{74@9xHrulxfjfJZ+MPRzHK~F>_jipB%
zpGVOnecI>T6-WJN*?n7fUlOX?7-O-Ea}q-r1~BUY)=(7;%|?k9+1!K?l}iwiQSlZ-
z{6=dlZ|#an4iztGkyx}iwUpV4%(<nJJ2xS%Od&TF)`oxf4IBnpv8CpDm_kx=h#L+?
zCMpuT5#jp#?PnVI9Y=j`z9XGIIRemT*(==mS``Y{p~kG;YK1;^&%Lw1%Kk~EY&6cY
zwz>uaINg_}VJFJ9uY!HK8Vn(r)^G(FV-D7Uppg{l0>Z{IMV2A;Zb5i%wlfk9gw_bz
zx>MN=POX1K4MwBojg<MY>nHcdX>WFAoj9J7z)Q;cAngcD%9Ks0#jbBsaqXdV7#PyF
z#-&otJ=O}g61MC*4l-A>12@N7kf1tV>d8xyNyg+nzrkO^q#aq{d#rL+8Zy-8ldKwQ
z_8+Q^$GZE0r*2*sdF~3pWrosoGa{#vBK8?}plg4YUu|s7@~ag`%l~cZyQS!~Z^Bbv
zujrMf<Yw(eO;H70H07)&gMpBJ?<sDk!?tL}*)Rs+z@@4nq$3NR3eb>^;8e@F`j|%)
z^~9&M2~H{cUpgq1R9<$?3PfY^AEQ<<ET?*HFAlxYS=Y}=$D!pWp%k=2#qm(u+x<3g
z)<u7u`GW8~^bB&ieDrX$OG&%52|DQ=S^|vFt#h0Tj9255cHla$+e|h!X**Pq*L4fR
zFoPS_fx!q$u$`x*xMNmtr&0(?jta>RA}9qQ+0;5PfDsAidFLwXS<BR1GJ*T~P41(S
z1nz8=p>rqUFoE}Voow(fRAn-6qB%|>qYQsX*;#!7do`?HD?e6c7W8aemg}6kZX99k
zWz+JCl%UMIW%6;<g$+&5L3@xH-%tUkXdv_1M8u%wfj==W^2qV28ZsT;tnio3-CoCi
zs??CIF1mqweBPo|xog6gZu-}SzGb7u3=C#Zwpdp_B#V^Kdrr)5r5y=e1#gUl5#E2Z
zOj;9;P=-0`{{B3JUf{OHOdK9dY@NWsJw5J5R?@J&if4Adxzk`)j4ze{a_-UOH~J8>
ziFE3yLWm6NjFhJ(oU#GAvN+=`YtD;L%hr_ES(<VWVZ$q*y56P@s!6hBbhR41hC!xJ
zZL+Nxn#QpPJj#;ea-HLhx2Z0=G5CMnOE=g*B_3R-p+3Q$>_9o(FWU1FjS>*P<!a)a
zlBn=en$U7JqOo`D3}Q^5D2Df0ldTnI7Zt~)P`Wb5F`;ycW1Vv8R->i~(xTB-4j6H6
zQ*}dktk5<nRfIC<dEQLRN#}q~<7z8jrkCuZt*=@GZuGs9Z9zTm4(y0g+#7#gveDQp
zOB0`-h7|+t-Gyl`NLq>>W#S+oJCv|}s`n!TK4sh;0R&kJ_i{#bt`8te8Qf;LkwV`0
z+<ITm7^amPugeL*TnzXB7p}`)zETmkM(1x;NDd5Px#3pD<^f1LX&ESPWRN%AAxL*0
z2Fe>}N`lUmtQdQ%wfg|i@oIl1Nh+V>Z)I*jj+KIe@#BnJ$FWi~D<|aXqGB-q_7)a>
zf2yV)(twxtOzf^3HNr`zynmqX2a2uZ3|k}9Y@4pXukjFB4QkDan{E$<8myo?Wj=N6
zTE1R^M?b4n7xx+K79DTp4AeGJYYy_?ZObF)Vp?xEQB5~c2CBocXYGHD{tr``QQqu1
zkiDdP1SRUFZ8>)nCfzz2rK^Ga8f(wjFGl_zygCsDZPc%}*nZY~Ew-n2g~c8pk0Vu7
zYVr10;n>egI9{1LpSu?~P`!*N%~>z^$angLylRHoH;U#|;pG#r`gr@i>EhBrNu2KU
zMYut?)=pIO6=t%&b*q1tUNP%uEj-p<Jz#q9$b_sXxUyM!Z$8!&o>cOFH=R?sVS=vQ
za2*8cs#Z@ZESq;M15DUJPh9a8|8WCR+>%H!DybUd+1p}+Sappb-@V2kD_jbcgK+Kt
zpYiJgrZqYU_}GPJCA%phytK|bdXGI(D0lYy6NRfH>a^aQym}0EV>x`|#`XW!SpO53
zab74B0yZ|65pOaSmzQNKDSyju+&J>yU!fl`!(tU*fYDgnNe9?l0^bI67z7jVAzAEX
z5BvLD#g`~b-R-s;PA~@Ak|?rR?-!+~r%6vAclVFGcL$`&t6;&K>GA6X{6h18&^+ku
z<N=gJm>$1RzujtW9RKe5M_MED&+~oYtKHuq1fBu>1|t7pf8z;w1b=xA8lHF9Q<UdW
z);;8L1ewF@WAri~LV-g2-yi?E(*;?rqLW*>j87jPcYp0h)NS+%jh()H-^mL?P#DY<
zjB;`M@#pk)_h|&$wz4?Q|9(78!^r2ty8z5P?VPm>-hpO@(&SuJD6b0dm}X*{iTPpq
z-hRk@v`oL{aedg|IDai1Fw8$X@F;VKD9=tj<*U;&%kV6Fm*L4<y04>P>m7vxTlTVL
zZyDqL;r+nd^90Y-F;xyHXM;j!TcOT23=&RdU)hiD%D#-5rKBWGmp0H)Mp2gyfxiGL
z@`}(`q=eNBweIu2v92MjKeVxjau<AKaZ*FU>_ODBiDw&o+<%}zGkW=-)tQ3D2vhLx
z5cNqVGFDlLjiUFAX~^o41WE~)e%GUFNt#XBju0)*>cN);F+msQi$#Q#)h6v&Rp~IR
zpQx|QXAwiyJk0tlpk<1_&)}0zSIs!|T3=W6`U6co)eh2(Lnc^}@Ui|&>|Km>3W8{q
z4l8yVJqYXqh=0sZ8*zlVPzSq!-d0+#9E=O~Lm}!CtuHzPq^x^mKSRg-jO=j@|DsXx
z2ABDllLmS+o;2e*9S^NrI2egx6eafg+sYxqrbffHPS-bS#I>hmhIYu~BC|3xNwPhK
zVpUiTvAr3zAnj)|772!3%wAx_oHR|cK06#S?L;FKOn?0><%=V;@yZ52MrIxj(Vh#<
z=Ur+F&8NMpB~rndD|*!d;?MW0Su&jJ6y9(ql(@ARIT3(N0L8i4NZ8W5Jc~sp2E{~=
zkP`0qmwR#%A($0jdL^=ppc+)xuf*6A0NnQE=oDgDQFiE(ii_#9$XTPiHhs=1_N*)V
z@HBlssDBV}Mc>1#!j(m2H-OiUS6QPfUa#`<#HE9aEi|LwgOk$5n&#?DBHg+Ah4c4E
zeL$P<YBE!xEu3^87{nc?1WDfL3Y};ZQXB(DURAxy2DZYkV!A7aFa^r96Brn>5!N!4
zZMLFu%vLi($zdpm6$#?D1hYP&cxi|sw2G6nA%DV9WY>Xj5@E<&PB`(35oUmx<cgi%
zR2jc78bBw)Lp9#l*C)j83Y@zpolA&6Ig;iI1g=nlXroZ$NZxBIkP-<8i06EIiFCuN
zSNL-xqm1^yvOm|3GZ52vv6>nkW;aH6g$Kq4_7==<{spXjSdD)Ec+_!%WCkXTJHLrC
z4}aFS%^G7lA@D2w7PVdh8gk$;0G9*~P;ry`OTva%W`uD*XjDvVO=G|;0(J?iAqV5~
z_O4Qug73AofW+45A}z4BUZbw|a?cVe1*hCv+Y28~NsA(Jtqm(XiMhL*y1gz}yT(c<
zH;)UxsS9jMHy2W_bV{bxB9V90!7qZDJ%5QKfb<)2Tq%1t91Ct(EaY<hlSQ2&!2oA5
zZiPtBxO^+oC5*IetLAlcws<$sTV`cB$84p(o7FboZ|9rO%P3P;N=5l;bSryE{d&my
z&uP_4XofQo&fhqz`fFl!{5U0@WK+Qy?DW4$hC%GYIp_I`BsB+{&F<ZQp1%I$%LN}?
zQch2oylN^b0d|+>YAQYi?Y>UCmnds09s!P*M{6n~0Y{f=YbqRn#H&7b*9_+d*h3Rn
z+c)S%A|S9*c5lO`6Drf|?CON>_GoXr#u}`Win?=k;=OqdKpBNkQ-@-v^=2ma)3X@_
zpe$!-=NU9+i5+7&lW`UsV~+UsN&^LsBu6S|Y+eoXFL|nm>m^-m*HTmrNL(-JO5q`a
zdSTM#*=58Pmzvug$Jq6^PxplD?5&zbW~XpOGzOn}O${g#d`REUmmzE_E`M&$2<zKE
z<6rj4yS5G#D*d8^e~SLR&dhsBcT7&d$<TK1sIi4g3R^cl(NF5*duhifl*o-}<l~6&
z97WpDalPnIrB&f#`)p|?t2yCHgr?E~fcH2Qi*B=MrNG5Am%0NaSaue+eCq7IwtVXB
zAG1&W_q3c=5}%tc^LfTc6Mx9puCzwnCqta`xzmX*2Ujm#TS7_l_)w&UrC6vvAGe4%
z?HH`Jq3h9H=nA)E*{$Wm6;``1&18W~(>FI8LM>yO;dO_0T@G8zuA%{Qyto0&_b}Sp
zILF&&P2G{%My5=Q132oeC~P|b->m3#`9xd^(28Z3RlDT4tne$bzkjp=D)l|u)fpS7
zf6}sfiw>v>#Kr|z><kVN|2}7I17WW?V;dm7@{Dc7>pEv_15~h^=kwzk+eq0%|8>YY
z=QudXOZh>XcqFd*vGuu+Ev7QtO&vL*2?U1g&lQy$j{BS@r237+$BJJs?(Z7Ua*O6~
z_I-(QT!kY3V&9h-<9`&nxqf=s1De}8fj7Y|l583A8_9bf?N2H5%08<sD@!vpnZ}Wi
zMN*@5o?ZN<E2V@8+{9HIwTrFWdvo?z-E1jDW?Ig+crzbri#OC0(dW?AxHIGh1xwx4
zs15dWvW*#PCjmcmBw=9No9)x#MF~`uLcZ?;YN{GLd585!om!Ws3Sf=$WPjpeDMU~q
zhAXBD1H@nCu#_u76~i@Cq5<MtrbO}LFlua6-o_DWSf>?p!^9|S%!<qn3&=Q?k5lH4
zN>2&&@=yp&fi~tR9(9gaw&?!@g?gQ;mvLSw69P6jm*Fx76t~T8Ds>77TL<<30=KJ)
zmoZKm69F}saQg=omvVV3HGhVe?*||Vk;amN>}+@!WDyv#v$zg&Zx;Fe6v-;G$nKtL
z$z#I?#-m}A#bP}Vi)^`kUF7oq?&;IrZ!g9!K}4s5UOs(U6g-gdzmm}C-4c`%-l*l%
zx8-m5Qfd#MC-_(?je$>qzaxCRhm%nedMJ3J9^s=NPQfZ@F)+qx=vw#kpk&GosSUC+
z50UWu48P$IO#}^o^EQ<y%o7D-#b&bjaryUy^7QOEPyKMxQF)zbiO8IP`1_~7-bs|f
zykH-!fLUn421YrcW`REk^V>(DXfoH<))sfBmw<XI9)Ay~EI^mnF0)LiFNt5bxn&-i
zkcT!e&?@q9PdL9sUS2AP@&$TYkE^i(KO1Xpe|elkRk>uY^Qy{t9LHAph^=L9RqjH&
z81*8Bg-<ljEPsL~)~z4kgK<W+Wt=8?VS}G_rOjdfxCQLFfOWuCoF^FtTm-7S*g&1d
zINf;?V-QhW7o$tvAj-JRo+HH+_EwiIdnzIykK_W_zUKCb3bb}Y%P`L>&$QOEPM?42
zdrm#iuFMmcZhI;jf6Uh5+Pr9VW%a2$KR%U6>D*e-)uB)QT_o1lRAz15R$A4nc>l(!
z#wY5&>%#F-g_c}XCj0cws%x;oS&EE&bC;Dpf_o?(xFs|hkVTxkGZtP;h9D0zZB31H
zU&j_2Eznm!X>|4>q=U9+A8Y6<s~bi~9#$=HC4VDH7MXc^f2!YgO5~l8&o-rCv~apv
zrf$538h%ft(JV>uMp`w7{)xvpw{Ypyr(xgTyuoDJG#l`l!EfHC1{r(g@&!MRas6Cp
zWEvck9CFxb*{#+h;m6GPib{)<TnF9L8U^l0wGU|=BMm$Wc~<pqJH=}wluL`3PB@Dy
zdy;PGe3MWKTSgI}X_)A$j4h=t-L+KOFpX6UM(^?m<7X09PAeW9d>t&R614X{VB9K&
z=%9p+W@4v68(n=zsS4gM$;nJIGJ|{`2r*@`Y0tYJ5?R3(D;}H2<Zs!;!~N{sm*9LV
zAAidjpU?zdGQvfBEK2u?i@?*5F&O((%%v<>BPzzBvVH{29nrM-x#yD(P?nbdCphIe
z(=o41y-$rd$GlI1x!X9;c*OmB%pCIqw~?@8BRF>3Mp*8=YzJ|brC8>@h+cPdrL#e3
zZF#ON{*d$(@3DoD8=^OC4K;`0uXP5%Cx65CEa|=NSsIW~nvjs3{Fz}HojeOPHuibj
zeAlw{f$Xf4x@x2IDq6mXeOs4z<F;;%<A_Ht^2(ngr{%*?(-7xK0-M@5faQJKj6`pg
z&Jpa=td}t0DyALNI!A5{jXp8--f>RHILtn9#CgD>1GZiJyKz-xrO**-uGDecKYzJx
z7rHGrYqlfI)tllm+X$Sr>-&hN?el}OrINDa-YN+1SMF`vhPcDINZVA)l^NIK+udI8
zhsYZE>XMuu15-}jmp2vH(g6d(>lMkaa2ANyCeUYlK^eT`u=ZFDqv_%K(@GWACh#@8
zeExR##{zlk{Zf(@<WZBNTmHA?W`C8FruRO;x3Nm4oKx?9`_I=efBF0mC1AjR^9Z>>
z_Hru4<y-yf91Kg^EWhLLD6|SLE=nW-V-}zjYz!f>w}GwU^Kxf(5XwN_?ybM)RsgEe
zYRv{4B}@#^u}#35h=)wEwbFdQ^G4#TPf^y?*vP*!6&#y?!gv~$f94PWO@Ccs)W4Ry
zFU#GBCeLKxAAI`&`F=y300>ktii?z*Ob{s~>q#6bfYNtb`9fQx7Cw3lg40n1so<|R
zTU#u(umw{S&>{-dMBj7!jsT{$R*_)sNVo*!ya+xUr9!Dz$@afLf5Ty4VbFW_UCEp>
zTC~nDCkrSYf?}psir8QrTYr!byHju4a6CR3ch4%aYDi->w$+&l;&C6Fhyn{A7RF*-
z>Zz^u!(s6#B^>DSx_o93&Ex{^6Ty>7E<442T0+Q>xXT3sylF@&wmOmLl_PuKgHy@f
zXR4O<S|Iha6V|}0+O>Q=&fy!TJlS$wu-YGvs@p9ZOV}^iG%uQQU4KXKYzvhKH(L!3
zz1(Uk##dob>kGo}zNihbQwk%Fm2Kl(Dy%JBsLrthRVz4}139dUQt@(Cr97VjtYPnv
z_sNRT=4eB8E$?5*hd;ltbt#{B!|#O|hE%w@vvq`@2yR!~>a94{L&8>^>OrRls(&wI
zH$)UC1}P}wLO#`HUtYUDHiTU>1V=|U(GLa7`P(VW;JzN)4e-al2L50T?Ai$?AReJo
zho$qJGQ4h18ZlQn<p}n0UlGy$>(fWJ`^;`9SE#?zPOSdiM{|Rd1O9*O8hD*Y_T39Y
z@Y?64m)3wPC4UO;v*%;LYPF6oP0ea12-B6^+{T#bg*H8$y=>P2au9W??9pbe22a7D
z{AK8P+fd8yc~S7weynFtIw8ru+(4h|BxMh}roVR<ljJBH;``!}G0i@mF|O9!_>C;F
zd8Q%XcpYcQ8qSWKSgm26@#?lOG{_g&=uHdA^*UBF9e?W0`9gtZc3FQ7)!gW4C61jT
zq863p=Zg4y<1Dw+9nXNM`XpH`q0Tr%4&;+Gl{j}JzjlgTrSh}4F3)Nys>2?}uWqwU
zIw2i)Nt##2t%-X}>$mrGy2K)9e?s-oX9WInV||;J;I%M1@mle<>Z|tYKz&sQHC3Rl
z)Xq<~L?=6)>I&K|gkiR_&Pdd$c*3((?Y~v|3KuG0I>`CV6!Q>1nQ|(&9+x13Dj$Dr
zKKa-;OBQx!%BhuZUBA6hYMFs9?l|2`I=C{&*9mdm5`3n7k562EQtNFkW^A2<I0O4d
zr)REqN^Qnk5m+U?0nhI#C><+-t-*0?aYGqy1&`gvRjRGJC(qdw%UT3^oX0n~%}=Oo
z&0lm5ERxUS&R_4RUjUf@SRd+J&_REJy1BiUlJuf|6bT^npZ{FG+<lm-wv#GZw(9vm
zC#yvcEg$jPU6KG=rT2inIcQY?)}38o^C!w9G|r6+30h4@WHdwht4~ZV%mo0B3vb+N
zmUB<{y~IBqb(PKan(rWmReq+J3(pyFm%D1usEq3Zx+)#m%7=a>>wt82`6ho!j1-5u
zb8LsJu*S?J{a1J&TDXiogc+}A%35w8g97C6LCw9*ijR8JZ&R+vdDAKgtyZhvIil5C
z3G35}H;tAO*dgP&h8{YOlle&hc6!$#4wZ%@UQx>1D%UQ_D~6Vj*KK2(HQzEkaD8vB
zVTc}jMJ;R7ZD|>To|h%;J-vTkzo@ZkF7>jqzgJN;0E>-YQO$dMvu!z6i(u7i&E!qM
zBI+LmQ+2~zE?9(iO00#c_bzyDD_BFE2i8~Yoe5*)N<RHpu?278&Q1pEa{_6%rAc+K
z$PHIrz&eqv`PnWo&Z-UDW4+@y1&6)HrW-^fR7f+$>uz9V&+D~S<1T;od-Me+o!CzA
z2k1+g{O|`S@_ClT_L6>u=Z__H3|AxNqN6k3={<0BW}lOKPfOv=b&tbI{k8)+RBt~k
z34k!bX{8o`*M{o^JD22MleJ?I*4UZ)Ue-?4h^OASx#%}sWV<S)o-Ve$=*&~_Ma$Z?
zXB4|iN+=trV*hI(;zTFmMv8+K3zfo5UCUkNX{9rCp9b{Hh;PWcu58LCio2*PpWDV2
z!<(1jgen$)5A^O$eykJTB|%c9>$#FCuh^6++eJ^Z35t3{%f~frRS~2gdj|Dnjn=&N
zhTW@d{;5JuYbY<!djXgw*x?O(zH^_k_9z(%t64~8s<L2fU^~^gQS*xkgVpLr<Ey3Y
z69BNakn2qwo)&po7hB@xw}5al)9Vd5b}A}h>%pRb0vf_2JL%$L3Nxn;U(aY)`_8Pb
z8<t(AvNwq|-z^Dx!*F*E1WUFTOLlE7JC`ugY!8&_XT5OgQ~3p3+Uad?X@40e!_dp}
zMH{BA`HAT!ZkX(Mqz{s}4Z<kXuflHe^maHzNy>O(#>?A9#=c*UJDRi*4@?wqm9%gl
zK2g7al(Z;oj3d4$YvCSkqPUl}tQRR~&57fiWUZR9BFkE5dg&8yK|Zz@N8g;OAW56#
ze$`C%3*@C6U_@oXY5cbP-EMD68kI)aFunweLAjr2t!e@tH|EQ7ex`CO2RBjJ7*;%9
zaPV5#m<D{0e`AqjqPXYZ9`d1&D~5jzDmC<Kg7<SDo=?+_db58gEu4zKj(by8)Z!%p
z20<Nn*Mis>KQk8S$U)z8#B>#_+}^!ozA=>MH>2#qg90Q1o55aGc$#JZhJAZ2D{->B
zn|399j6>JkT1MTu?-<9{i4-Yk`hNa&_kZ%7IwF^GUMLd-HaRnw^oA-ue~u%$?cQHe
zKVYri51`SA+tXekw^__B$YC+aOb$Cic5}$@hmuH%QkC1)ZBI6v0o-=6BuadUq(o}u
z@^z8Rj}OnE9)36*y95!P3VQkUWl``#!e1o;=-m>O65gog)3@a>k5Xz6pJ(`3DUE?o
zfZq|m-P6IS2rU%6QAhZwf2TvR3Lpmh7!B>tPfDf?NUf8Vxr>D76FkEYO$ZHsbC}8l
z=79oD#b&hlarya4d3yJfhkiQfsJzajM4ViB`s=4ZKS<<(6v^kG`D=c9`SkGX!^{yE
zy)fP{pT9li7uG0YqF#~_-lpZZzb#)LKF(O%SQ5Gjm;b=8-@Yyrf8-~WNs=HF>AiD<
zOk5D&nZ^4gOny~mVu5uw4MhELj%pJ_kisfIV{jl*3lef57z-@2=Za0(JTLlFraM4j
zx(#!3BH7j4w3?W`AETC5Sf^%;%Do?>(per8LizH3gjxyf(`t&Dg|X_j8U@n@VTBJs
zoQuN1oJf?Qw=N4pf4xHhuux;1F={zc=L2+AW>HN82go1*0H?6R2rJXVT5}Oq57wRy
zRe}2a;gf#@+&NUL6T-tLM|9A_X+4i7pR;<tnHxGf0iPM0I&W~~4OmkYu+&yq-6lmw
zYGI^VQFJP;SDIjhuqIF|U~)B@r?+cD%$V$qQK@ZBB)u71e;4URl(a<Y1(b)ZQZ7=D
zRRiSt>ENZhZRXlrAwd)O&0PB^;4`<*z98$H`v9L)^##{;vQn1yH&a}6IXq=@3aU`Q
z<%tu@m0c3C*A4tt>u$WTAvTi`>j-vsTh~Wwm@O=F1B*dB;Z(Yc879&H6wCwUr&s)o
zk`PLO>M1K+f7AG)FQ;xHs*Ax7Ng~5vuuZDES0;xNE0U^~5!qklb!22jlxrTzHz8fc
zm16^U>>6&L394*+FA;xv-?g8_J_?28(1k*f*Tj#kb4f!WVVaOW;5s}bWp?FMFA4+2
zK4HvjpwXhcg12>X>h|5?3je;V-f97n;ylkwtKNE%f54hft>3%q^`MV+#ZM}ki62+~
zeHEHnkFB5)g3wh6Dx*j_0R44=W<M^aEp#^tT;9Kz0v1X;a8;lJN(G;h8u*tb@$kbx
zzkd1C=bypfFJJ$an?e*TG+2hXe5;SnX73s8@(VtD&?F0{Lx}`IHE3xBOAw#|b!>po
z%Y)Scf8q%j8T6M7O~OVCF1~jzv2(}`X#fVIIb@6tO5^>)fi$Z=M4_ooBR^#<*f&2y
zf9jPV^M^lEQ|R^g<>AZn@DaJ{ybifZlofIE@NbK&5E-ff8z;tD0+}FENY)cqQUKMw
z0Oj;HAZUKfx^Hz90nAj6-Eg!<GeH>L1Oi-ye}O>sJ;N{XVWJfv-HQpApq~>egsh?p
zxm!8he}Ddl{k}r4AK7=sb5Pnq`0tm414WP(SJ@l6xVc}&=G2<Z0xtEk0yPF+b)>EJ
zH6ei^KQ3%am9|4C?&EHTz&0AN?qJAl?-(OASk8*$wT(N`U^K|Eg_G#GY=jMQFptI9
ze;rzUQ;cSNvoDB;*7ot5{i9aJfm#~Hl9i1YvBVOW&|5o<nAm(atSRnbY2*xbtuHhE
zWef@+T8r|QxjE+>Urse-z;**ii6WX8S^C^UNd^n}>~Lm124i?(Hd<kXRyLAq+%k}e
zQ6W|`<{0R>q|DyaQ>Bv5Y%&*0kYPs(e;St}wQN+QL?&o!cTFp>Rhv_MMF;ThFk+!G
z^`I~770$)<Fov`)YXj1ax}EX20OwR!0_o2bcx>5+z7n_WLtpFFKKx_h&E#^>-U-<9
zz3RYKb~4~w^5z^oBe=$Wt{c2Q_`u+s@7Ca3qj9ObNevD-z>c(g?+zK<cLxvtf1CAh
zF1U}vVKkNhAs4P%D3nh=<Xe5WhTWM<A%@8#Z-f7!-dlr*Gr`}*ecK4pUOg_|;l8Og
znY81RagW_F6;nwr{kYeivngyXS4QoAe-6Ub>Lz!toe@f@o$lO%$uO!NF&gr9XS97W
zAe)Nc3LVmZokw4f(4D>>Ti26je|&N$=EGcrHv5ZKmD}KW*#`G5Ikrk2+UCxDGPdHL
z=M5Cq{De1obDxt2rDwoTS1`cow?|oVGd{lQ_;V#+&<5lAkv_1rx-4+znp~c*3@)yb
zhd!xJ&IXT;EQ`s8fZ>uQ=kELdzMb9Txkn}g`?FxJ;@|;EI~hDz)u??ae>lSSjA4kA
z3ms>xDrRcI#|TZWTDZc+VneeEZOr!Gx1dk?iGP)6n{$dA+j3A@r#uTHainwSs12Ov
z98yg@RE3Y5BDY|trCXo<zpl7Fh^ry=q_8bmYU&*Etm@oCB#$&}cJej$5B+Ufk{80f
zt4o=MWK{}|j!a0Cd=I#Ge`n&5f!ZY@V``Z=DTS6b$z{BHJr!3>?uBzR?S-}AycPK8
zG#p|M$6s$hn0FALK-@PXp7If`v4`$)YHPO6;Bq?lSWm!8*7W7=CvegTFoV-zmk!`$
z5E@!dGt(&Nz(Vg+HINjfw_~pTIjAO!eDcN8(55FV+nJ@s6;W2_e+8ae2wYd#uAgqc
zCj&)n8ppad0|h$|Y4r+g&TE@(aY2;noLbRL>gyAoPiu#=;E-2NthLF&m1wrpe3urR
zwn;>7R?~aGs=&rX#2iF1Rne##A1=DiZFPwvfqj`U?mG-Gs`*S%jal}4<cWyR=tNRI
zQSf@AV^_iem&c6Re=&#h?i25Nw9dWjD&ICI+mKbDg~%AnsSJ=I(FKmd>%{~sQa>1x
zo<yJn8~<2i0n8Jj_eN2Xr#n>-CGEC?(0Ja6l-xbC_|@u92yaN@HyxfZey1cpgfvFk
zd9?Q?@$ndT!gi|(^{|?)T48k;$`XL1&}yb=3h77@Ac~hge~iB+iSNA7+Rqp-CGk;h
z#>T%-4&UgU9N9T>^dVE@`=s!nXkGvJQut(=t>#IHd!+CQvr$5^6#gBJ{)G96SoAH1
zzldP9t=en&Zx95nM+ldce@~$tEx!@L4hHlZI$V5|KIh*59m{E$$Sq}cWTFRz7px!I
z6`9J;>JI)be*|IE3gayfE_#536~^!6;BqaO(er5Ub#UZdMr^k_xM)OlD;{5Ra52KV
zH!F%J2Uis#7r<$&@8REq0ZeAbc*(z^+Dxr~9|wm!mVGm)7-njHzaq7A#@o#7clU5;
zSapQmZ8YD(!|C9K(e?`FYiBn(!)Cm@f6Y^2b>Gw7f1*dYF`uS-F;J?0U}}AzvM2In
zPoy#1Wl#5s3Eo)#^h#cbWl*;AIxR(WAqyLl0ET{-B&-i$A}pvT=3oQ3uAqDQRLJq6
zX9>av81k<TI;3BF=axMs{2kI}@c)x~1-qL}u8bX1tlQ8(X5uv`*42ITu3Ce^aC`Hv
z)cUS@e^)kzt@YW7ePbDwyD9Igf)hTuoq1Q4lJGrAqg!HXOq=L5r@aueN_$}<rW#b6
z+r<m*N=;J5!mFp!kRA0D^$@0tvt?rSRI@eC(+<!hB?%iYPcE^w(aJn4da{nsX&RZQ
z^H?s|QB76a%A+E5Z#2|7nOLP)oqNR8A<-!@e|36pQrJq$<;xS@y|xS+dN~&<mdN+Y
z^epmd9$Bu{v8rtnXX&#dI=-pwF6voWrT@Tij9W;RN_kmq=bT##TydC<T<@foIdKlI
z5RdwEo*DZNXQqu-V!!E`YoaiVPbWE)5;Xc`=XRxfrjg3(sNkW}$K<goUR?Yn-#Lx1
zf8JO^aHUX;*_w-mV(wCia-eeWLq4NXmYZ_$)LMfF-MF!-j1|6@(DeJ5<?7=hvn^Zr
z$fDw0ykW-$Uu2l6yLLfTDV+<yn&+7C>SR`UN%E1GXO?!&oUz4P&B7HdOmCaspD1q6
zt#mU`hOeHEJFVt281n*I(PmlRpY=Lne_`n}D_K#O@z%_3d(VqM&pnaUj>uSkz0%d@
zH8>=*!Iz#Ntvt@kFxQmW6PS2i@6aIMUV2$<;oLd*9M>0VgM@&2xVuzDaKr$YKxn_B
zfl{`ysv!-l8k^egm6DO#qM2pICfG6Rp)J*#)_0NB3iqTjn@g#V?LAd5q^UT4t3+mM
zCx2Hoc4{`xBeE%V#pCG@qq#%;<UU?79P8Mo3rnm$H)oAY`wC0YqgIoAWYxr~jbQ1Q
zp-bClQxRf>-IpB>YBs_f6Wy%wwiizKQZL&|+1w+$?RDgk&fM1HL<MClkrc?S1=}6(
zT(F(lXl2WI_O?fM8hLK>IF40Zm*zX$pMQ&VdA7iat^sdR?)o&<=FFC}V`P2-`HPsg
z#2-BD;)mLqB3P!!+NaqV7DyXvtNe{ZpxSrSP(GOkYc0#E*{)jy<uKv^QT~qNCY<kN
z9ad#_XSyRG`5l5a_4PeoE^MKaZ*8%t59fx)oDTikEVv7(^RskoL0-H6P{+Aa6n|I3
zKbf7bPF0%CZPxy>LFfj;mV$UEAT()?BCzv0+t&!Y9AKw4*!~}gjmrlCy_lLWLrm*o
zsG=*{eC%tO)vwL#PQ|S?(XT0};o`z7#*FW(_M?5rc&%fDdOJij!=9ioqDlE<KVcO>
zl*z|D#`)+=J&@FR4S4GfJJPg43nF7z!gameu$CC|-6?v!tuxPj%`rGVp_`&^7JdH*
z4sFdwmvLSw6PMvdK@<ZsFff;qITt6F&f7E!mlKsLvH>`^#+53d0e{bCbW-3ICQB!y
z0!~p40I6&dhdiSg%Er~(tA(l>40bsI<<)A{)xyLD`LlXy|1y6agd067s*P^-pkm}s
z&NEm2B2%#7FnqLsuqh0)Dhw1^_0jBeLP0JeZkt2UtMqZCc-tJ3OQZl_pz0B?Ua((I
zuR*)F+%F^7D8m~E^nY=dh$)vG_C@8SZC_w7Zfl^N_bBnL#)NT977=HYlVkRAvTS`G
zVsQv%a`=tgTN$%ptgxF9g4SqT<UW_t<e`5ia+7NIl(!;xgc2i!5qpnN8jq+9VM_`{
zNF)dr8n(rv1ZRcs_&uV)w1iQ+3Lv+=%9UUlHC>jPmw!P~`G2NYg>OvpJqBFUih#kx
zx42-zObnkXdzPOFr7o=S;glv}v|<NXv+m@TG{2TFm{b>Du9YCAOI!OX;tKdYG0auz
z<tO&Er0N63jpVOtK#y&)Y|MK{iJ#x=zkiZ*W!xI$Ud}P;7zNkW!KKzN_3s`J#23QV
zvxc7AE_*XjpnnTDF{~og)WML`^n=;m4sx3=9Yc}{ewE6+zDqA1fg+_W(-T`qU|F-P
zvb;$v2l-P|G;=1XH1)oqenaUckP_1Fyka;v&aO(hSFAt>imi8o(ndY3PVmwkr}Zs%
zyrJ=_)f+kTxn+E4ssr|O)~9ahoH&mb__`X)ta#Sg`G4^Fs!yaLHb|*bpXj#*5I-&%
z8wlta?DF3wfng9JyBrg>kuoMRKfL?*x37Qs@+(A(<=cPK2(dtMWn~P@k8<f?4rM@B
zzL#t8+S02CK{X2P%mOuw@JW|wTm^ht9*hFh);2Kc_XLfr2?i>iHaPdTd_qL}4N^Hs
z1S^!L`+v@X#8jIhuc&g7t`Zd-n=WBIjY^m4!><VyM*VYn__{oNs?v-a{&JcDiZu}?
z)V|U~5M~0cpa`Rl4<wD~MH0ds$`)FI3UWPXG%PKw9$ZgKa*%XgEJ4_U2?Vr=0)gmz
zgl`C7SZfgpW(uo~5U7DMX_HP$6l)P}`~Bqyj(__Gqdv0lLgyfrUX69j!6*kB;1jLf
zvb15^Ye#g#CR0BhwS8=Dl5BLXbz5?rGV{#D2BN^ir-iXt)_Q1*{j^&^Qmz&}_-z4=
zKCKBx`?!KhSHWx?LeLt7Uaj|YA9y@K3jE0X@aZl<RU{7=DE%eeocJmZ$zJYE-FS+)
z6@Q=g))B%w^9afQk>)j*()X_&^}wRItymH7nAX*2T`)Em_3>-8y7Z}iI&RT1sY^O3
zv8ELSJUylyB2#!OzCT3+WhN&aTXJpk!jp@({w%pnw?gevNy@@7va+VexYIh-f=YGe
zH;Y5#%CSB@XcO*`la&TrTxq{>Ea9co6MqxY>~ap^3CEcFi=X+J?)1fU(9P8sS?VnJ
zdID5vU$(WTaQ6H<iB8TN1DmhjsnQG2bDwFtio<M%(dqivc-F;>v8VYyqy1B^)}3AX
zag9Y^iaptiL2ax{J;mbE_9z>BEZDN1x%KC57naQ^h7YvVBX7t}Ih$zNkk3tAGJkV^
zGnd@lYjMf{S~!|m4v-NV_k7L-sfe;B^?;p}-qo1DFLSbR9c(a^Iavn@wvjD$f~aA&
z2Lu{e0bS};p<U|K2G{JTPQXCK!=EU3g4)I@pHPA*griFvd0z^pJ)}YrcCgEf=FtDs
z6v|jgh5f6fP~ajVe~CL&D8e3&F@MTQ_}-L>O*ymAwzxY}Ci}9KNy2UjzcXdhsvMqD
z%ETP3DS>&%;|!(?_4C3oVadTA7YIGdut6o1(2BT*W|Vh?ebXg#RIsOzg>){()gy<U
z42VOThk`)Dkb=QHURz2KYCDPM_kHL|pZxSoe(a=4o?}(1T!`Pcl6k<y-G8UV=!6NL
z=~qybh89T2(7yGs$c<fQ;nHo6!9B&h-Hcr-VtLr!o#YkHb!*>{&at~M56;cZHiNU>
z3eE>=Y#g)zUuqS1TM(R7?&MrQWu?4cl{DRf@TndMACmq(mItr0_*e(fGJzSlKNHSk
zxb^T?Y5?ojNmb8erpo)eIe$gT^ZX~Ko`N;H@#!Pw-89!j6>&tjs0hpMOgr!C>lvcS
z^SXLE<guzMG;d0ib1<3_LaBc!E7kr!uNw<w;;irv)sGt^U+SzH-HEg6Tn_ENP8u7o
zLkJpgtCBGH$LhpqCzh|IKC8w7j_9-qD~3C39I8t)CaZPPS(Ua)Pk)c^pR7=gj?1)Q
zfn2^&+DQM@?v#VMnUrXCOcv*6WsF<yv>TV*%%z^!)<5>HdGhSvbk8AyG`>A{L3L^k
zC7I{yw(Hpro)u&*kYPj5@y@ONVpSRsC;jZzN@vpA3UL$vRRhmcI~if2>4co<Z7UHm
zM+UMIsv`x3d3}ekF@M^yW9lZEO6M65jAM7$O~K?6M(UwFJ(~_z$uh7e##%UGs!q17
z(`BEHDOx9a`B7ft#wY`&h!5i^;;bbgv<5mM<GN9;#Fq6{ZB_MDaBF@jgH{Xna39x3
z8Nj|8bZ(F2mJJ()Uc;nr`BXAz<79HB8n;p-1`DLPrItZW&406IYAd6MlDjovo_!pz
zs8!~~XmaXjxaxAQs`4`KqDbRAMZC`9BBz>Ka12LW=bu@!8CT<D<@A`RxI8nX>Xn+(
z57S%;8l)jZkp7-ZMeYr1TO9J^%i2!VST%76Io<7Gp|Y=<$Wyc~X!R9CAcyW*EyZ$s
zx#?J+*g@`zr+;bHIiG5^D2;mD^sdtR)yMY4ET9($M6KHFAiGwe_MAS=svQp@`KoL1
zYEKPb?RGirIuoXCIKzcW)pDGi3dl<TG9OhrA#Z@%k9dQnA$5oT09%#S+Mf7q)Bx{;
zW}T1iV3SZ6LJMR}@VP~cj!)mxW3mi8)WEoVeQG^b1AlARV9c~#Y~yAL0e2m*vddN1
z$hMTkR*m8O720_4of9iB9(ym26n1!V?c%uCZahjsYQu9|o%T&i9$eQXZ9sZ^E9^m*
z7}&Sc?<PC)A5Gf3ya8K5p><+qovUT)l{Of?2KMs8a|b3e>@6nbSZKC{_S!3MJ1pf^
zAttX6^=cjkQ~jA&xpBxq$BOfpjW+u>e}Wf{{2RPz<jks-8Q(K9aSc-cA+H%pp&>zi
z>(`9NSzi8{(LnKyUNZ`*TlSi78TABT@g<|4@QrtjdgM2L#i*a`#v4WhXZOBfG&Pjr
z8|?6ZBDc0_mvLSw6PLjx2o#s+peikY+{*DjU%?-kc5l@SAPgdnH3o9Cp<9qcAjHnW
z3i4hK`TbONSFb~ADA|6<AqFT8i`}(vU9?=jEpqvEcz8N|xEQ;55kPspJbhghzL5A=
zNd!7ryi&p$wLHBnf4P%VJG>t7vQip@m&e}$?{GRA6>x>ZH|icQbvk;h5MpqDkI}g9
za#Au9Al6M*(iVx&XMDyVx*#-s?_r7v5+Pq$v4oZ{py!iv<=rEZJ{@&XP7{^D%*E4R
zpZ<K1tOF^cqn~t}pB|qMza3_c*cgP1uspvU=!G>(n4lLG@UEBN|Gs=Ze4epZS`xS9
zKOmRizbzBw3)>`0w25>MV6cgQl4vNqI2VPXSJ@`^qm+QcSyX~WHCj@NnuK&7DJZij
zqLl~_5G4HJ64WYHi)#(cR5cn1Vx|VrC|h7o!oWer=7Wg(3Y$cq!i^HpPpqfi8H>D}
zj!u#n!9&I;Ea|HJNs<hZFO8PgN>~?XT8f-SPWpKRR%3-mOV3!zwG8=xJ-OPd>>89<
z5&gn~A*bvKxG0;bha?lNJBEfRJX{G%F$KsD_<@}*Lu;=E=qthGnfU{=v_FZz0n(|c
zIw}NVFaj7;G2t7c;btfpC#(;DbSR8eQz&S5{g3Rtj`W_nNMW)O4#5cJ<BaikI3UR!
zdJhGv@A&Bl?4?0~ioeo-UeX9;^-cCF=w-Fn0O{o@_L-=}Ood`$u!|<xuGpw)Ys8#v
zM11RvIVHdgU%<}<f8;froSagQQ`z)s3-A`thUPn)w=xD^IJ722Eu1yYba{8l(Dw4?
z>{(oM!aQZpx;kQ?5W>8Ak5U?s><nQ`3t34;5nXB6R!bnZY*mYYYy0V*Y~kLa8ZRvN
zO%wJ_)qsfhiWiFD)E1rJUf5(1Ub-2hQtM?$X=+2(R1va<$&Hg6deZ!;NN+Jcss&#O
zuVHZfSCi9wMN?jA1v9-Z(YV4-tEodJu9A7VH-RqP$g`*`O%kxzjY*;Rz~pu7p*RTr
z$YCe@j7(69`b0*5^95DSKT`D#g`9O)!oRKRTWQe*S1KH9${eO7xs~e?N4<!1Q}=MS
z$_5dnukK_zLn+qqr5J)W<xQ^A=}*$s0Qs_#j7BQtwdj&Xu8jbI;?llY784Keu{RxS
z!2X(hT^e?-2fpt0h0=iSI&YP?_xuJ|sgDIJVKro|jv^X=ccq(_O;>4keZ}p1yx9eC
zeFlR^DcicXP&F`C{@WeQX1(Tq1GEz%#GhZUQ|G`MOR3_%lcV#F>=Ej253Xv<bMb)!
zRn-HNr-tSuB?%Yn{KQ@jFf&B2-q5Sv>YFEw<W*G`-Fckv^aox1guR{ddy9|N!do@-
zxmA3u6Ojsk>$rMP@zFPY4|85}tn-kFe?hZZbhI|DxHTHGZpa{}%UYOZ`aDKNv5eBx
zhO(U1q2xWt^BRht#Hwsj1a)Qd2c0-&54++C#l?(p#nXC@XH4tT;n7}bJP=NpvK5PD
zR1sLlIG*=%eXnnm8OoF92udDm1`L}+6wbZq1Qz^%R4pbgSG_00>*j$amlWRj>iuWP
zyw==v$9?5bfVmoUe{^8X$#d|o^<C2mm{v`#Z@q)PFB3xc8y-w1j!TF<6zDDObL?;V
zUFF^2`LLawTMo(tp?CNpbxNd-3G<+ZQk{#2RJNC~P(6j7TvubWwac37J2u;$*K%GZ
z4Y%5VTyVR)R<ro&zUz#9_ZzP0gu18ik7B>x-F0?yjg!7FV;rwzUFo`LC=Jq2*^2iT
ze{x)z&ciUXtU~>(75yh)QtN(=JxFwR_Uo!;SNxSDv*O2IZuLB=@n8-S`~N_btBm$x
zYWFrB7So$Y(6IT+U(<-6GF6(=^K{<vxc|(5^-K4oeO{a1r}9Ez8rajaG^XdWSBCrL
z%YmMcDI4KXg!Xv!weVi4huyzg?1UoZyr1sw{bB=FBdy*m_OSbxi(Nk3Jn|<Sg5A^a
z#Ps9SRfK3C1>h0x`Q`B2f>M22l4#hMpqs+-pM}7<2^YIGt2I_(O2l67@Zn$IzW(Wd
z`EN?#hJVuzxmY|ms~DD-{0M;CqBhHSehxw_4`ERvqDVB86iTC_>&6OrUJh1!JX}KP
zpx+4^g$<U5<#luV4&>?7SqOBX!UJKf&_eHf4=h|Y5=o}Y20alL-b;_TzwDw%y87p`
z6nFh&Iec9XpQ}Ej!GCE0F5Q|sm9WZx8`Z0{fba+xeYCEqBYMnT8_-_R3e<Q+r%569
z8p7iVhsN(rtqNNpoxlZzxC({>D)$+_p@2!&tP-xh0wV5n!aH(G#ahjpZNHyic-uGJ
z^r^dNG)IbF>}r;y!4jZ-a`5zh51nHp*lBrLUBdPLWNJYNF2}9e$t5RKE(xK3kFg0U
zvhaE7Sj?gxtE``Qt4ApT06X+E7qF%m9dWq}RWI8~Fo__Ikh<xmG>9@6Xl>$VW0A`&
zuPC2R&IPjGZq)Ur)4lK7#LFtpm9`@dW~g0?fpg{9CJo&64Tj~m@!U6lv@&o*&iu#L
z-mbMSys96rg>Ga4`{ck?>2rO5Jzms(2}`f*s@a|G^=#?l7u`scjuhM`ow@nUZM|=Y
z9m^J?;}W;`rfC$)5c6D`uhw`~Y@JUfyI;DtHdL&)+swW^uS8?#_-oPFLMw;kKbL-)
zT#k-D)T!4CCRHY|G)#6%VQ`mpzq4%ObAVuL*>u8T*^RZdV3`B#r)=GSSOG)Xtk9us
zw#}_Z76Kip@bEuWHl+ZkJuL|`BaYaV{J(l$S5FYi`=3%i>Hjy-r^tQfv#g%Jhvc-i
zw_J84SIcD;?vJgP^no7y&U&fQpFrug0iDS;X2<6YqxFu;kXOAvm8`hB`pbMk;c692
zIrQhZWmJ)$B6DunhU%?<PG>r)`prl0jOX0TF$@;+duVjZJe5``*^lJ-8M^VCD!i9Z
z*L{8G3V5`szcb_LgjcQCNA`nhIBD7#sExgN#8<1Zow|Tl)#hQ)2n~IWeAUV-3_OL`
zY^uWDOmVH7;5xn6+FkAE74quskJs=PjQeYN{XoO_{{qkoReTDUb0Z}Y0yQ_65nwA5
zmnf<#DgjxSQ>rQ~e>!-~@o`SBoc@~R^!6n4_u=E|+ey0#l+eanGkyG+fM+Csg0#XJ
zJsF8YNuK&S{kD@5jg)lYhosbsAIrZTf9w6TBF8I$XW+2^{o|KR+z50;c8$2!R*<Eh
zxJG_tB2K4F{7{H%;K!vEq(|1;<XM%UKuZT9vp9+-W~{;Df2uT63P4w+Q8uf343wFN
z)w?Vgt4J?_dl)sl_zP^9J#>;&-V0?v?J{78+1^#H8$FPFjK6)Rc}iLa-N{zD@@&vI
zaE&W<hoPzbD9P0JRtX?uXfvH&13N>(c4lA;W&EBbV<n_gb?w<xNpk@EXYGTA==Un_
zz<jC^F``0Wf4{Wd^qSTt6aq?f)S_f_M1u{CXHL!uu&SJSl+!e=>4kYGn)TfBig11q
ztd4w^g0fD@CJ}aRwLDci-Au&SE;&R=r9kZ*PErs*PH8p$!*07AeSfQ3Av)FyN39fP
zi=pl;qZd`)`P7RJop!c(K(>=?U(o{#N@drv*iq6Te}|bO9XqMe>|Z11MH<_k?y=cA
zKno^b5TaZSRHVI-fQ4!bLt%ILy?+k&u^D2L37vJa6kskp*=GU`DJ#IiEbI+t(GI~D
z7f8o(Z?FJ+QIEoF#Tl+O%x@=Fa|H=BMvykENSrwx)rHk4jhh+SIm{I*=71w7_?~s~
zCJ+c;e{aLT*N>;a7x97vkpP<h`1usZNr6y-xEBr>hCintPrr^tOVSmRx1<j(6J)cI
z_neTuWId!YTF-0;7*$OZffnd%%A_1}BQp6U5e8)987wK16h`ThV#FS^<44LFDGG9y
zk0}0Zqzgm9rwbtr%_*Z?4GUmN7z^WSwUC1cf7lKrYYDGerpu6C!c%!Alr>l2$*5T-
z&*u$+QRF%sdB2SLz!lX+pIk@I36BeE0VTbL9w#9qAx4TGKb;|K_?3|j{18Eq@eBNp
zi|~IA3sfXf;R1`RVnDMAdCWR-<Y^yKr`bqvr^N;w3Iba_KzHdmTb+(v3V!l6{K8$A
ze+qL+IA7aV3>LH&ATDJMMK(%X`~xyhIa46oWnq@G(RvI9(!z5Z9_Z@Bb`H~wm2=~+
zQY2?7s%&Azr406{+`&;?Sii*MLT*;&RQ;#&1}eo|Xyn3?=NYx#^`(95f>J@3JPpZ|
zRSkuTlB<Py!Hfvhq2;!+W@s!zoq%0oe_8W0_OwEwoHz$-*t7!Wv;Y{@+;T$w8Y=Z%
zT3{<zfX9HpjodUKu3&_vC9qt$n2o|EgOueWm%yT|ev}Qmp5oT@E}gIn#{7;v%NF+)
z`CTD<rLhmp?*qjM(J>#Z%W(Qx<hnVIat4>OHqiLe4kk6~SqDBKq}(|1gRZyle{5h2
z1Kihs>X7WO*jr0$pAP}mO79EFEV&@kRpCPa&=t+LmeYObU%Reo!97T|TXbt3b!q}i
zfH$+YV6}x9JIvHWSDD6c^z@h(V$@>@NM#gSpFV4!)<YFB>S$?1J#0Upvcvp#JmQ3=
zU_rIJCE1k8c%gY}MW*wm!9|%gf1m(d*w?L54GdS<i=YOQN7-AyBC2tI7g|3us*M!$
z!N&(bX`bim<l7@4a#GX2MnKTzVyfdT`*bx6T`oIvcM%d|;dNoHnMpmuYxLl1Kzgoz
zj4<?xg7|JmoSaN+%QwpDIj8#@Ka^bH8TCk$oKu`fx9gi|ZG~ab2CA@rf42q(gqL<)
zEf0@$jhKS(?6tt7aFQPEA^oLwy{&qfH-wJEU7c|5;%FHyXRb^q_Z2h1KCiG{8ZR!o
zhYRZSeI=VRw-f`tIbz#ACL6PZg+!lB$zZ*ZPAzuN@Z03imc>1;T~P4(P86CIw%)6h
zV|F*Y1V4KNszDIa0;TQle>-G7Lkr20KdilC=w3*qIauB)-(CccJ!>vTuk2Y@q+Z$8
zw|L9fGgWpoD=;sFx?2k>uBQ|(Zh=E%%g6N<^%MXJQs#llYoNH<C2qyOJ2C^OF0!?v
zJKrJJTev>C9a~G3@7;WRk(fsA>PI&&JX!;~#o?MN?mWksg3~QaJbM0v%80X_*8YLE
zH7B@bTANAkcGKc)YW}yY`|N5pnq7MN28OLdP0DjS+e%oMU(deGSa78M7s$W<;g@k<
zC=&uVF_+OHEEAW4r9dc`|F9~df6MBJzd!zXlFVe!k##6&Ug3x^fDQCuR74dKAuVXk
zLB$1-kRl;9E<RU?I0Egzh=G6K=_C0+%{R8NVgf_cYkbeo7@yC1C3&upX<k_~jYJBL
zX(Y1ua`v#aampD%671Q`xFQywy}a!z(004f&&Jb18<s}m?v1%KheR8?e?}HGYUA@8
z^C5XAWG>)S-XIZd+@tvOx%+qec=~U?bD)H>Vfg3!Q+$CMg!XE{AUx>d$G^kZr$2kF
ztOSx|GVqs?!;f!6H}Zw8AxIL7wAPr$8bFe)aIgjhDpkeuDKnV1kGE4nqkX(hbBrj-
zu1t@ys0}(+y)%mnBc0_df05see3|IEwF0c_F)O)j6|~XTK@Y8GbkG1PpCm5f#58S-
zbW@%x5G3v?&laOO-Y!4VC%_qPo&0DuM^j1fATpdVHX~Dv#tCcm45Qw12+KPbj0iZx
z7I()i?HaihD`D;eZjp&XD*_qRWtmiHV-i#kZP{r=D)s2}%rX1*fApVSX>x|FZ}Oe#
zB@vEaiQZ;pD)FLU;Dtd`^(#h>Q;rfSi(KR>KB?(@Q{Jc1<j%X@Uve@pXXA$u%}Cm%
zSZbUlPgR}j7%Ah5hiCec8)AweMP=f6G8WJN7q=r!fr=wFIzd(#ic*&%LuJCC2;SCB
zHB5o!btiF4VU4DVe;;D;b;{ErMU2eXLAlp;-ZG}G=o#az%KWh^C8|z!6e}jjG`X@~
zfhkcf#V<CavR;zZTQ(`$r8Xv)HPRf=N6rP(JjcH(XAR>@7+Tt{*+G`-YzHf*P`Q}l
z#zoz_dLN>G%=7^@WePPx+@hXLteBLo|B!QpGN<GXYSgEce@^OXj_N(jEzdPYt`&FU
z3-he$8u2OVUzgcM%82i0gfQ+u!#*glDSsx<TF!|o?Jnn`Jnur*<zB1R7RoKJ9d-WL
ztZgr5jObA~a+cjrCSFutwOWB%)Vxc6!)Qy`RgYbsFVa-Ci*h{assm4JYTk?cX#ztj
zqs1P3v};T{e^$egyG*6m!f8LF(vF@bQkb6Al9pV~#kC>qZk<ZS=~6DNXjCRWcw;L%
zXTO$56D@kwV)I_%Cq4{8E?P*0^{reAUaoZ#+}p%Y7T(`0R^3V@X>O$;Y5fX4qcyYF
zw4-k^ZeneD{<3}F?AL(OKy{s4t1jxUvWB*ua)!RDf6<z(l2VsnQPyCv1Z;Lp!#j*S
zmNgybjk2bL{zYXC((1;$8{~OSIk748cGY{!3+F{pGi>tMS_zjlXtl7)ukUm9^s|{p
zMGpHqtw>o`F*GBWgkKb-L7e^3b-G(PZxGyFwXCp>AROEqWN)h3q!*3~k#5CV9eV>t
za6*CUe=(*yCe|=|xf-M@FgV}OT9c2KDP6cP;NK;mcEvs9(_)O(<br%|!c4(HY1Lzv
zr~5IYqVNJUTI~w0+jiMk{~hLqD!R_FC}_nQcPkn~f!PJUo^0J7S}2VJ7kbPOXadAz
zRPTF?;O>@cIrZr<UXrRrfd={+y;Mu^AZze?e>_dP|I9@PN^2&`uxV4VbWW9g<ihz&
zmrS;>t7kEf2W;&~y2kC80CgCrftbs&+mas;LAoC6YVt$#!OpnCQWUi1$fcZi^;Gcz
z{UB=(!8hJfj+k+wIL>;a!+6wG4{)M9PyFrzFS=+U!9})83!>`KEqoXOG7Q?p7D>25
zf6pIYDVEN@HTgv&+1)G@P)eouG>Z=7&r5&6JArz}1Hj6mdUn_ybdqgjx{s+o{i^Xm
zd>_?UVU%O3h#|yavzw}`5QG|!(N_|_yD)n%)OeZ<`cnGwkS;4K5480=>T-wiC#I@J
zm%9}=bh(4xS(js)hcf(rx{NU`+RV-Ge?Y9YzyQ}}(vPV!fG}>)u4#wy#skc|RJmPo
zOO;#by;WI~6u|s;sti6*MYQca@4)Q5uwm>}x)sd#bU{H|n&20^pmwJ&XoBQ!WKDay
z6BkdUb=wVtQRS1@#_=56T;CZ@!W-|*^r!P0RDU`h+I~*r)lZI;+}PLDaH$_<f3NIE
zF_taqfWA#S%v&?63ZC@ut&FG(C-iok@~de|239(Lr5EcfO%n=I%%=b9bu4?Px0S}u
zZ>7Jzi8a#$sRj7)2y3C6LK*!+M<m_;GF|_PU`@{h;v+od*^@3yUg4Tf4%5=n9ecAz
z;%mMR7_pYFZRThf%}rEq+qJn>e{(ZlJGs-BahR!WYUk<96t4G;C}Jtq)ZF<>V7VYS
zqT=fip`?oyqHGQ3Si{%<Y@1@%SmT-thSP1$Be%E1rq>`h#W-C#HK{rM<%mAJPpX9q
z<CUP0JKUw~&I?aE?B#C2GzRpclYdViPyYk&7oY|TWo~41baG{3Z3<;>mmgOb1pzge
zv6&bYm(O!ODVM%9CkU6`xGI`|9E<eFhreI{*x52kYm(2G`Ah$Le%<}M(^F=fjAu9e
z^K+LUI08cv2ME@XhvC<^;rs4S4SNt~<(=t+?eOczpzeGvT2EHAZoOx&v<_BG;Rf%M
zq5Nplx_h#gWp++9=N_fWED@?%etwCti0G`?k2Rtw*|O--iT&J&JfRtXm&ZAOjKi<)
z$nX!y-W%|!k<nJ;I`j_C_yD>$HRF&Q`^m@hO!3QfP5$QbA%bx}X^8-f<3nT<u+e<>
zHI_kgn8uU!Hhw=%<K=uqe#3Ja-*xj}wY>|*QRr#=8j*b&d!7@P-;YH?z7Ny0YzJPl
z&tt#2iq;v@4Re(l1%U^D8XyZArB-*ozCj<gRLu5GleBElS+e6<Cucl5nuHD7S*^W1
zeDXHOEYAM?yxyN{dEk_OZb2+QuMnR$ASRPSY!G>Q&5<)m*m!~Mzosl1Z}HxKR!hx(
z=KSOt;~MkRIvah?mAEply=8xkc@;Nvx$bXvzrxpwXVkK`_}XlLYc=}k^@t9QBpA&Y
zSs~7DVh?t#?o^20B-zfjw~TRr*@Ia;*S0o!yj|no87sb}k)D?21~iO~HrGa$C>~94
z#oEW-Y@{_>vbW1kWI}QseS$0+vW*tOuSd)g1n`Y^A&~3=p_5$*B_X9N4u0rNXl!&H
z%t8{rI5l$7Hp$(8O{h$gST&0a!g@pxW4_+eeR7H)+OQW_hZM?bFUA`}3PqAVCY%i^
zOvCYGlwI99P0-8c#|e#hbh=^4(w5BuCvBB0;A4s_(yC--Dy+np+PD0Cb?C{bJ@NR?
zezbOr5G%P0n%=ba%ML4BS<g~RA@Y?HVa#XVKb2+|O|(6K%Lauwr^p#`M2;&;MleYb
z74n&qY%VZW?@^cgZ7YH#u)27RsM74wDq|63Pm7^>HBc(O=F|<R^o_;uF%{ilDJA4N
zr0nz7O7}LczGd8Oz@~xGn;26yZdiy<E0E<o9YO@q8Q>+#;~InHLNdYoD)jl8+Zk2t
zibFltPIjt)euB*KOe!yCzme6+2^I&^tPV<;MT&P}bsPr+3>}xKLhQ)uDC8@wj_Q2}
zR!2#^!s_Vp-JjKQBv~_2PutgsABfd)AgLbkovUpnbt9{z(B7TZQSv&lIvR01tD_Mw
zvpU)f`m;KeB!;JOXFpR?JFz+n^@6{x^)Mx2lZQ=z{-DQvhO3}xTnxJR+VxU6n#;2}
z@-twtG1(fbf+dOJMu}VKD<6%tW+LmU3Jl(a5Z-E*LgREYOQCY~Y|ey;6peMMgI&ma
z!`#eyD_2oy96g)mB}_<?SvHzHY^dUZg&UXyU3KSFE_p-LF05mX%_S6!De}Da9Kw$+
zX-x@#MJEaL`ud=R+9*1rSbNDSD*!85frkI-uq|*jjBxL5p_bZfTc}arV_T>tH1zq3
zZ9$Fs%C=DV-nuQQ*<9Whl%s6e7L@U~g`b(-*OQ57{tGB!U#b1}@Pjd#cKsZq&!JSx
z{`=*g(Z>bD{*(dN(WQz*C($MLk1mp4d*?cT7IF&lgYb!DWk*NfbDlL~M?O&@U*Qu~
z?+$z-CN#l^{~e#0|M-hGz5}0#K{VsiyAX=G!1VEWL?`?Iz9AN8&l;oCOdyVWq^TV|
zIacrGJ8AyKGj=@OxkxGK5imdC7$f6h1S(n<Dq^BH^FJi^0sZ3Ym{$61^;_xBAVqn9
z^v#(zC!4UNyRe@KXuAGCQ3H15K5O7B)MwrK4$NmQ@Cxy%Mt6VSQzDiD#*SuJh#!de
zgj~{-yK~{8q;BLr71~>P&r|(f{8Rj#{$|hC`Ws86H~GfQIci-j2HTQ0&cEF;UA6nk
z|9h_iDNYFdz5Sw=+NmH_sPB6EOv&nhaQaM-e*5V&J;v>)&$PE(KYgZWwf^*(M!J0Z
zO!emd>XT<3?K0J{+AJGb)xuT2O*Jzgxm`6RnY66!W*Ki}Hm4(HT0I*MVHL`<p9RS|
zDIFRWZaIYIk`yI)!^$EK#>3f#b$f*f#701?Gg#rxXRs>Px2jkhgb~@EawI27=s}~m
ztj-ep(itwe=NT>lGF>T`8t{WKE(yk{+HaOM;8I1|f%}N^z#mTzxl-u+1t*zFfK;y6
zG`F~J&lIleTMFa<01<`)Lzi)0C=&uVGM9lxFBF&GzbYwzOK;;g5WeSEsJAZAFq|0<
zA3$IrcH#wkYmi&fQ_<bi9=iYk(4k1`LCLXXsXoY(C31%Ik^JT%>E?T*o39VYkB6r-
znZ_y4;IR4l)_^?F@(WsST5cSGvINb?kLJ&3t<lPHl*4FEB!`o0FK0eH5_o9=@&u#>
zz~Rvu$bATZ<<O}Qo^zjL=|dogPc4vZiDdIr$_KVjU6f5Ow!+||W@!V`8|d)Y$L|k1
z8(CXqol(-SSi~<{8yT-T$zkJllo;j3J8a7+j7Vp(jeUL+<ShTqIX@E{*p=2SxG<)1
zs9@tz;^}Mk@9pE^-^l<Cd*zL3{{Hb09~c5f9vjquq>y~`^PlG1!ymQSY^+vV`lKJ$
z&Cl;mb>}_LdeAcK+FIt(tb>-NU>eJy3}V5^yk@)487HwDVTAJ2*~6?vqh<YYEm5XP
z@DWKap=Zt+D-*|Ae%j%W?z|7?(^>@<YsskRXJ2OBI;dOIWE^tinAUi-6kpOc+12Ai
z6iqgN)EWV1$A`$u;RN}()|t|JYFC}k=hwXI;tlx^Z+U#zr~m5QW000+&$%B<WShqx
zZNmKjRPUrT4?H(Nr~Pb|Eyv!iuvIK_A|9xLsHoLCy7RsOkFi#yo=xI>7R`nWCKpz;
zR{<x|E~0imJX#$&laCp9&QDFjb=P&oo|Tn<^?zS%E3F%`t)wl-w%e;w*USFcT2hYT
z-q?<{Mr@<UjA*ob)@HxQ+3r*46=yN_P8pima_jnC&(FfNI57%z1=|{dK_rEVM7~|3
z^-7KjM6dHM#98H>t6R@6$<=U?wu*ToUBBII9RbI+Ho?o@E)hww<Arz;q0-)5!fWGy
z@*)duG*@pLJqb}(NsVC)PF4iWZ!r=Y^PVPn6bH)G^gpfM(VbziRfvaCxqt(`vMKR0
zYI)1buFCCXH&i{Ub(x=H-@{dC?LeO-XJ_tBpJT0i(dSs|tuE?j`mEHyZKuz%#69S<
z8sioE9Jk(!4hntlhF3?Q6<#~g=OVm+I{KW)Ypc5{^jU41o#?Z|%zsXw6&@St6M_oB
zgSEYgK4q2kWVn_-{qFQ>f>R;*9W%sAB45dRwYf`nXoL~;u4aw|4l78lV4tnJn4hgD
zX)PWqxJq$FI(=iWoVF`RIn8v|k23RgHOkkyh0A#k`w1;~-ltSvO<+6HO6dB3J!~v1
zz&`Yf^-gGcce-0CKMVQ#mcv{a`CR$$>CC<fv4AA=I8)6v!tgW*-@00DViq~+=Un{C
z#iefRxqj6%93IKErCCblS#;$N4kDUXMsi2j%FITCV%tZ_d9S6CaaUBe>26C6l=X0q
znrv6Ljji~A@Z4{ojz}PpWNKM|mTebxl@hQY7Jj<|u^-Cqw3)fPJ%R_Y&zDf_i-BuP
z#28owi6*KNaUf+)xJ!jN`o>`Fj^tQL+_6GjNxo7cu3BHJ5bs>kA(n^e@0^lUS~ph9
zDrwhKk+!n_c5`LD(g)`DOd(g_Yl*KLBub>`{6wOicfDj35>61jp{iwnBtbW@CpN}P
zY-abwCZ&!Gp<!-c9@HB|54YZ@QNg9Hht?=@u@lLypIAg)CC;d~(N+IB^qf<lnds#J
zgIr@nyh1X6y02k_C<|d2ui$y$9DTq0vAo@U$@jT0PQ-{|iJAPC-l3KEI;DMawXKej
zj`-F$MwP8==hpnTdqGEkr$^muqWGFF<lxE;X@x$eu=w&g8zk*Sc-dwha41ez!W1G2
z^kC=*>R_QA?P?2?Wt4%ti-qN=Hto|{m_i~6=Ok<#IVEu*7KRA?`rDTj#~Cx<1C>BC
z>ha5$wM+M+y8@A_h49fR7*uNO1nF2@N(ycDIfu`>-AIpsa`~)(9iG$mQW3TnQAhZb
z?B{BK+#?goZ~#emN2}oA+LpdFyU3PF|7^=62`5||1aAcNV*kAW)hKS*tOC?-1k`Q;
z)J}vf$q_$K#n6uY1s46sftMJ4=aQ16E~&DEB&tgS6M3aR^T}N+#f-kVxs=HHGJ;kg
z482!3uBu-|P9_F_-Gz!qFRzPKZTSKypD`!CN&XcAj=c)mu1)Tu|6w2M<POj(Mthg+
z4h<>8dwv5_ODwU)IoG-t;CZ#;w>~zKm|L09*JEQ&r+dk*W<CDCYv6$MDoNjZ_3nJF
zbDl)f>xnC7L`?QZ$GBO$pvlhN=!dpyhY*wnzeqdpuiblpW$ld~W3zJSl+&=v^c}T}
zqi_ZnX}9}pm%Pw<8W(pL2QQno8>2j$RoZv7ov8H`OSJR<5F&)pvMagXz=xV#^lg2@
zR7_-DQ%TvCQQK@Z<ynyaw!wyo#MAQRl}e4dgaGc)1!#G+WCvd2H1JDd5S5(zxTQ|@
zDTy2!mTJ@+qk`L|`7mrHG=O?O7JXDO^*i}2G0A_f;X#a-ab74B12!=?mvOTsBmpv)
zzmGjLf8TEuITC*7ukhpEL)<P`*=3(rkyuzbbet3cCvg|$AqgYMUYLnyCVTAt_4k#B
zPG;x{B$EzWkeIl;?RL3-eD#&x8ps8ga!@A`e<M56VvX!#5j3bcj`oPEla%cd29#<I
za@ron<Vq5!0e@?Jgd5RSw$U6Ng=(WXIf|k6f39c|+dXl_(w-7pYA+S^ORb+2M>&`<
z9MCFukJVA3j42pTYckonx)_?Q5?qXABZOS!)ZP+ZlwxxTyJ)LGOI_3i=>vvpTZ;(R
z8x<{#iZwH)NXc=??P*1;^}~!DDOpP#Gh|dOg}IR3BFEA+lF>@ZB+2MzKX{rQYg<`T
zf5(NBO#$uJHbsJZsXfgMTQpH}Kr7p1Do!NW912UN(4HnUi0LtBCmL;y8Ba@T^WrBY
zGlsoBl08+Zi`B*|E)(9TgsTKQD2$nsMl@_bGJ%y>Qm0(d(%f43DQaRW2Gm$E0WNdM
zO}OHq;U;i(Fun;=g3FR^1ayRD+S8I8f9~64=vKt62=nIBY+P|aG%Lb!l%imY8cfeN
zv_e*fd6gOe)#R+WlD5DKb0jmn1`IIUqYE0vW~gX2e>Ed$G3#i++p1s@YRaf)MRSTm
zrWefZN{qI4^X#~Kadw5Zja7I(*$DiUHknuhlPSEoD79V%tsF1|yrvY31yg#df6%1D
zuWEoxYGO2CZafoLn;$6d*U0t`*rqmR+AAxYArxXk@M^8(YSQ}X=byW$hsh!)xL4iV
zfBhSwr@ol~MDh$nPA|{TkM{S`({)c~=d=0ii{sOYV;kVQgX86LGM~D*y8QLi^7-rK
z@p9s5b>rD=y2KpMENU&@Y?f!xe<BgA*e;+xn2T2zFc|{53A4Z79n5E^uP4jH?%?UO
z?#<-WawSLi=8uaBW_~|`K-1-9x?C7StX}?VvY1`YPbckGldAj2<lWivFSAdF0hbx}
z(Hf61evXmw-`Yg&9#5yU)r<0K`bUQi$ZQ!6$X1OQ`bHaebR|z4-@U#(e_7hq%d_c!
zT=&at{%$gF^M<4DZ{3UT$sw&SUQJGyha9Vio&iAO6VOCBctzYq7LO1=>K;41{&n|!
z_GZ>Sbw80P!>AmD;1sa{GW0L|`)gX+MT!`UCpZ%f<bg1bDfk4JB`Q7_9Y|3lz$4<H
z2wC{n!uJ<~YJ?E#s``M4e~ena`7wp;mE%wo0ni_laMn~}DGV$dK@-Zp0_HF|CIv#P
zH)t-fQjsk(6wIDOBA^z~5g{&DGFVhaUW=-XsLBm8uOn*MYZZH@3}EL&#NJwjeGz-W
zA(EBz<3j902l0B>sAL{QCDSfZ3BSCG%9SYh3q8sL=e9Hte?tV;f9eBF9+P^FTWk$x
zfaHS=8IYJ#Kt>E3w&r+cMWrvyJeu*w!b-^ltBCAk6)D$N;T{=T`$mtR+@q)TeT4c(
z>>E6SQGuTc={;NE1|S+DH-Z{~y!n{Y(5kRv0F^-+1)rfzT*kmt_)@Z2WAFj0=FIRd
z#dY-n%Wb{|_og7|e+-JpRI}V)3eL$hT(OrzsCv%vD^g@6fnvl#folRGMu2qzi^!mF
zS5m+dj8uK26p~9-P_9J9(o78;Mvy|PHv)Yng~CsOYBx#&ekoeO-UFo=AXLmFNC9u;
zcr683wy)_cx=7LL5iS+23Q`RoK0=~R!PiFB*$d(~B37yqe{AHwn1)hd=}-nIRij3<
zk>u_Xs4xiJ0HNpMbS@GSo{?r4it9jz$OIw{2~nJcQmAU$0AESBNwgciR*4o%`(X%f
z8m7{;Qz}hssZ@GoPGJoAg*8!=!fD}R8Lm|urvvPOP!5UY0)x(GQ=dG^7_h4zsDY9k
z;RLu9Qe^nre^NXC4rW(N=Yk>}l^H5b#QUaGrLSc4FFW-9gyP;;`JSo&UhFK#flKAE
zi%Vs>cBzzmm+E;y>UlsKgSavB8rT-{yC>)xdEMLIj}RNG-pS1ONNe!ubS;FG1Q695
zg?N}0kJ$32Q<mg4gTyMlFk1X251V}_RUo5F+t%_Sf0`!PGhCB}Aw#+`<_WoPBiu55
z1H=E^p`>XLO3J&0lInFR<@7mlavxK<kEz_pRPGU6w;}j;`(e$F))*QX4yyia`M)u=
z3c}7Y`6Bh`(S|>CkFQ?e><G*L^VJLc|I>%%@?!DVu6uvB{BU{VPiG&y-!4C1?7f@(
z-aR>ZfBEZM|Jyk%{^(18yM>5}d@>^4iwLR$7BH@e&tD$=eRl~Fc~Z)iged@&V6<^1
zyqwQY&UThD1*4zoT1F5*l`^i3Z!hMP#bRd}5tkn5<UsU*WXX}CNV7aFUgcggE>6!b
zW*30BCoc|e5b&-gRwzCe+rIX;#SBz!BvuJ`fAiPa&L;Aye1mALb=Xc^Z|cPvbx!45
z^t8P`J98NaZa4Vv8us25RLgKrx9qg1Th{h;wcmQse(OPdEY%)MOTSN1#@OYp$%~zB
zz%e8k!d6_{VFozEke3pyCQ1y?LNQ!WjMq`hKFiBuhbcb7BryU70z2dnAGUZaN?=HV
zf4YpKW+Zn}8Kv$cubN#B(c77gZ9OxQSr7=%L&gHu1oY1lWCbo-Xq;W%0I=38A<RB(
zr%N{oK>~_b;8gf2H9%WsL{%iUa~3<8tSB7-%LVX3Tehl#)xd4LDkq7Fwr-5vB2|Gb
zlAS}s5pu29X&~EP+1(L#Hb0vlPiH6pf1S<G-}#gC+55%CY}pVuLgX%#?s74i@10&Q
z7kj@?=8Mb4-udzW{@6R6O&81g@fkwhe^`E817yRfa;v@bGTb{C37)<KaooVTdzAHc
zSk~9ul=Zf&wt;bV4dd#LT+n<`fK;&sgfRwfr`2{U@~3f)`mMse4hwTaNo0rdf8X^)
z<sRiktwXQS+k3^DF3N4<s1DD6V`ZO!l0S&XGz{WM1MQ$OJLrG-lbi=p0fnIo$_`Ls
z-!Sx0VqY;dDr6rqd~WFdIZ=$C(YhqdxsFkFh6rwiTudxCl4tcqBQSsrMwXqh_+;PT
zSk<9M%<#8tIVMVYA|+~s+E_}^e^5st0N{UGSGzYcN;~vd+q@B9w(P@K4MHgUD;q(q
zoZ@;BVIQD;JI*(I-Bd|z>tzKlc*g_R2cvD4D-BLK?IW2T(16m{lg)aEt50XD`Y5;P
zXd8HpZ1sa|a%mg_43(X<TEEw3-%UV!=cNxyQN7jYMT4cUeeSX8Q2z&U1NDvqmvLSw
z69P9gm+|Ed69F`r0SGWDmpaiZr+=Qo<KJHXvXfazDT3v%{L#NXz3hJ5^&(q?0&seL
z-*LNEAe6@m>BnHE-~Tav+kNSYR#uXRq<=<EzrRgg%h#;)AW7%aT4NSE2T6v|leIx8
z{we8P|0<;+*)-`*ALt<sBP?Y6@a$ogqLySm(^E8%D3Y?ICT(icu)*BU-+#%f#w1S6
zF}Az(o7|^y++S0h_K$1`o}7G6lM*%*nb9OIELy&PKB@v#UTE+739G=2@<d&pthq=9
zXtj`_1{FMIsyd6B6{}5FbacVYmHM&dB;SLT(d}e%Lp)@OHH%jMMTr+JBypLKB9oXf
z%m<bRoqKEH^`H+9L7o}UJV_eykn~cdV1sd}L_OK%?<VV#!z$8{$$*Dq&m?fMdmZXi
zBo{x4B-~PYvHcS(Fp;D#vS`StY5o^Av}kEGRKvh{p(wtJhE9gJDAOwc0e=MMw6`vM
z-{P$u-!tCYY4}EOMMLXt_5bOv>njuEryw+<v}-F9?P=X$=PFaTQFvp6AQnyTqJ<w7
zAnkiJqqm`%(}<x=*iu<k>IW~S0;#0qbm`J5N#En=i|{c@C1e;lFOh9QMiBs1OuT}3
z{k#-vaPFbSRObHL!$9#UVSmt`y3`XcTlyJVNsyy=T%uhv>|rE`8N*%+cThV4df1aA
zGokW71@34?#Tfh@3rI(Cxv3V@)aJxZfE6tElu@oHZdx!{2Yqif%p(;s$t&Da#BMLh
z8pyI6v`IR1L>$xv%qHzBwk=;T(0I#}1Hqik@<+6mit>WixG7q9M1ONVG1ox=L=f7B
zCAVQGYzsX@kh0T$3CgSJ1reO@=&b`3R61@pEG>T<Owf{&oQ9b*ug^{|sGIse>?mKi
zFFfdPiyTIP43i=d#i_mIg6J&?1!#FYb;1?_^+4UMP`^F0m8Ie-?WWi+<YkTRMN2DF
zj^&=qVFq~f_ReMb*ng2%(^9&lh|kRys-HaqbIGqej%6oO-z_&PO&QJKv(|JJ^QO?I
z+M)wh;9+jZYKw8gk~8+buhr>>j0DK_zhewTF}ePTiqaRtib~HS^NWtFj3Fl()GTwz
zURy>S3WGRkWqn15&I;$;;D&NLUvC6GXZUUjfAm4DL`aPOb$?-k7nG@oDQs7wqFQ6j
zpcar<S5L+94tPiPD9ZTcQy1uG(>(d`Tb&VnJ`6(YHJIp1&n?HLJ#TaJxODK`CD6F4
zl;+f|2yr$$coga(#Q9`ku|36IejpfkGp^bRNe4?$r?<=Ru@f>LBC+q#no$AO57#a-
z%5HoiTzBj@Nq?9mJGtF+ZwbrO;l2;g1VjzNVQ2iZs*^KXXf^Q1KmC%nSg4?`j-U|Y
zOw)|`!!KK1_b4VPs8D(;WN-v?J^-z*kEM?75EKJd&1NrAR>}|HrOi!07@49oLMii;
zSDhW#D>q=~ZC)p|8e}=Y0<D<ud^I@=ee!Cr*W=-Gcz>f&I1PhNZrakD_a|;xN%B1*
zH6L=%VCzUk%n;(1s4$9_9&S@qq`5?!der<4@20^E_OYK{hRr#!L9m@)-*>-F)baBa
zdlLZ_2>ktbip^9ujb@B(0Z7k8R`G@*>1p@j-*4ak{Q7qasOjxL+*MARvIPV`y{A`0
z#v!?I`hT9^EiW&~fj|TqPrZG_W@T7@eVuli>JKR6Yuk5j3qpHMfVd2c8;UNT&}G6+
zQ{m0jYHQ*2y=^SBnHfY`>J<2mspPTzO5@8Ye&wfqFI{QWKd0TdY4;^-(^zzI8(1d7
z!P4KFHi11%L95Zyq^JOBNlV|Xli<jIj<mM&)_+D6)?2MeLN6S}2`9zOt;$?l8o5go
z(JB;5Q~B)MkLaP3m0IQ%DB(sLXDC}bc1_6IO&XrRzrN>jZ#3$2^O@*S)KiQaJ3Sy{
z>ku!7#{FWd*{g>Zse9>N@(JX4b9HC2a9WG3jtH^*GBs05vOUbgeYsd61V)@w7bv7J
zJb$wy_|6Dfyei_zIZ^HoK~Z$9ofAD!h%{#fODuTit_zSqn4b2zyDlyW=dmBvl&_=w
zf~2yw!Y*gua+q`XTzvG{g-<^0pn!+~%7q(O6kLv@@|!*>s?s4YoqAdCwya5jecTy2
zCMsXNvw&-&u6RxHsqQqMZ=;8J3WcY|uYX>feZ2UH$x!i-#dTY5PM?$yCEvgrIi`DU
z>6TP;f|l;OwRflDe4dY-J?3aDx3MCHPm2@GL%dlwo<kNcgP!w!zkN#~-4#o>!|Jza
zV^I6Lj%#j&L3?BHm@Vfyb+&9_%1YX764YASoKs**oBxXcE9}Xi{{!nu-k1tyZkN@v
zFb@JXH<xiBK@^v6_%A58>e?!E2n+csoT*a(A7ef%sh4qHC=&uVHJ9Ns1{0T%<0{bs
zv6o`xDvy76lb!TXy%R<^v`7*c0+65@f@TU`k`C6&c(h}j`Z?ms@P&yytRIQO$;hVx
zw=z})Iv7roWIS5|5s{9sp$5zjh(S0E4aBGI5UmlC&YMHT+4E7Bq}=*O)?_-1)f@nW
zKzzUVBxR%>mvOIyh@^#$R2rD4#s|?V4n9;7iiE8uN=J2n1l5QUG@{h6P+FjCH%<8E
zBVz^XZum(1n2e0}(VGSILX%kc%%w*TDipu5{4a`GR7!_3tE#ZdYK+g>hF@3lXp;+!
zkk{#XC5|-q`&ML7$&pwSO@XWXjw>Vd2-Q4f`;I#QS0jvx4$K(SZfqU{sJlK#;;wDH
zf~*^jLQx4yXt$9;uGopT<-aMXVDjAPoibk&$ZEUOPW&)Ug>2i?c%dTAD@U@#Qf-BC
z{!o^FjPLSh{RmuHq6_FW4n=lCCM?N8nGZAu^i8)p<|-}$f0~lUchTLJtQL`b!I)?E
znwWk~HE}@;^O_oV=mB!lMmu))=j=!>8x0e*@N#hcRa^aGS_<S#q~?f}UPZ=L8%P?9
z%5W?>IGq?TJgYV&nKy<@hA<z&8SIDg{?AH%SfruCkNSXF6DI+*-EL+GISDm3W`;5-
zt?puGj05tve|VsoVGvtWvf5yPwF0c-kQeeUa3*;cHhVTld5A`Yj;|WA>ny#>PZNX%
zcjs~q!x|M7!w^+h%t?G-8kah7&WoTLXxAM=4x!dNf@0bkzCDE$vR6(ur?97xq5?(o
zzJa(<NYO?p(p)YD@_0F;<nPGfU<z+y+&Abeh#=h*e`Zr~yDs&EToMMb4Bj>;#DkCt
z%o{?V5ymuoqvSg)oMRkX3V=#EZF)lE@Mi^Ha<MKj>5<R#^((1_j?RBPH*vREmFWC-
z@IY%yhdtZNf?jD#nYH$FJQ(p|eHD$QN;m^^az9YUb6&_itLft+M<yxdRQC7lzYa!t
zr4O_Ke+**Mg9u5D>O3{PQu}R0KoN}QSa(<nC5<o$4aA*x0)ZlC4)nUA*m_$LypSeb
zWy=aRlU5))4snN*t)*}d@o5J{jB?pPTrdU@C^~Kq+Dr<J<FdtAC3E?Ix8cy9ausWi
zEx+rl8BD(T9E<WyVRTDrO*=&{8^>mTj(a0Rf4CcC3=)jJit!CiC>cKD=bb}!hIOZT
zsyC|)>J1_L*t|3RTWwdlbL156xM?M+M5Q>yJqbbyyg0-i2|@|aIK+hnq1<G0+3RhW
zx`XYalt$X{<Jd0n0s!u5yQC9<YP0RyazmhE&&d#~Dsx<skEn85a}4WG^yX@x^}w;!
zf6Kt3-j<3L{iR2L6%nz!zLNJiSjHnRB{c0#HJnK0l!fYF+CjMemcm=%Lvxyam&E}z
z1Q6caIT|)}@GQ}x4aGlxj*fbIrrxv<RoBX?ddjtM-upYZaK3e}8g8}oRrFz8qoWnd
z`B|<VSLsGdsmTLzEUAJF+bYP!KH-#Ye=5ir$WPs;gscqVviFXo7}gCsI>b;~kX7Ok
zZ^-fF$nN*vduL)?yURFNVRv^KgSe+3tMrR;jy>;=%lIBIm&7>!?J<$kULcH*UOv2K
zJ!sr!Ftj2JZ3gqONBtE?CqoLXz@YA}#8a#XwV5)gZ&$1-)qYOJj)cdk<QN#le?19L
zX@BJqcO*QyBE|tO<U7@xG4Yz_J6ChjjV+tGs>FBta-n6DoT_d%drZfslCj}He|H_5
z5aNBR`M+cHxQ@*jaX7$3S(>kPYw0sS-$j?^Sh{ZPpPVeYtxI#v@LRErmmJ@vNtpo!
zWp+i;qBb^O>b8^zc3Z|<To>Gme<DY_LtL@L{V@zs`tDu0a}2|<9#qg}P;V&c&L+~^
zW1EsRaO&jlq=7-)Rm{!hVGi+WcN)SW;>9$C%iZORsGEYF-9=tdv6ttEsqq3h5$tls
z^{V+ir0yGwE(Cb>F7U!Gw$y@SXWMYNZ`b!Z#Zx=VQvBd=E_njw3*<{UFR)2>XG)m;
z5aHlsU-Rqhef>zgjZuBjglU_+U)>q%Zte_iCA)k-zn=aBe=+1$mvLSw69G4u0ggBm
zmo)4u&wsi3r@!8QKj^wGt)feR(?#CBz8(HO$W7Yll|eiF{qv9p$bd2dhk(k(;nzRI
z_ro8OXlJE)MftIM`1NBDJ>RO9qh>8@=d7J+88oY&2IrzOX{czKzG`iGZ^YUn-Z!y^
z5N3P)atd&kB5Agll@u-8jhLcktrC{|{GTaXJAdhqr7v;L5^&ROZ_CZR76}xky|tgb
z<yjAJxk-cI6~aOqpxIvfYy?HDaj6X!L7%qsHl)ztl6Ue=ZsxoU<Wyu6i&gqq`LCOM
z)z;({8Na8k=M*p6aEs0P)H$8borUT-ZFQRAZ}SEV3K!yDat3Qzp-Xk?i7~7h#-Gf3
z)_*f$-IZ^gA|E<$Q=Q>C1(6z1VhbwOS=u$$7A7sNJBr}5gNpzu>-O}J8NnP3j>WXi
zF~<ANW(7~mc@Rui&KZPcR=HJuywAGYD5vp%weGnr@a1H~5EUV6cP@!qXA}X>RCBDd
zc}{P<>6#}k))RLcO_EPsKC>zFr5u~jV}CN7DJQv~&t}L<#k#&C)9QDAk`1VORf(``
zc)$`U8#*4iN@gfBKjr*B%Ke1+xFUOA(Phtcu%3pD6HKRGcp(%{b>-M|o@c(oG_G}?
zDLx)Y>eVID2F3+^0JsIy+;Y+Tt)Iw+9u1)K;RZbxG&IrP{QRG{n(<C~V<fl4DSz!S
zMV8E=#&&?}1MJR0F)DKAtPi9!OH)WV8m$n@>19GVkK1h2Yn+UtQ$X~gPUb4D?r2#-
zq2IZ=JLS0xx3OU!P3gI<&vRcg#l2h!8MRtR+qHF@D>-%{Fp?W>vOc@m*Si0W5S_qq
z&?gDeJ+ObnhhTI{X-@Bn(FH};P=7g@&8@a$UT|z2hmtbt^EtRCZZX>|<R~y9C1-fu
z4F{9u@+xe>!K^jU<k<-!G0I$LJ>Ra-`J5}75&z~)S~=&#e;SYLLUTbyTu{qk>YVmQ
zH2pv5(P-tlYTD5+1;sPLVv|pTirP!67k!e~gOjtj2eq4^EMioV%LBo^1%EABqSOjg
z%$~vNc7}PlT9w_Vc<!|49Xq!|Na~Mj&^=ydTXaEjTj=$*=<Lq2iXf^9ypZ+9%Se0$
zN578?OmK=({S?NGxj!G{?C?DsV{wM;z;?{pVqGhN+pJUuV|nP5LxG>>^V>6+;o^k2
zIQhH`W*l+xYj%tjg5r)DOn*>b$zWpk2W2orsO+Yvc6($n!S&7zCa8T%1~Zs;7WVW@
zL2;i9CaA7tFfseRpm;;XH9E5A#2rHJ`?VhW^h~xNELUqBQyWemCALb4#)7;=%)CO+
z%k3*GL9~Xg{MnV@w8oh{GMUVvo=tQedzt1JiajU6L?3p{I!T@m7Jv45%n&&sF7%f`
z1I5yAsR0@8+ND+Yl-qJmYu&=`R6|0|$G6BHtumY<ma;dyS3p&&`Hl>F-98B&_wSR4
z!#%m}{(aNltwu#LN^Z9z89=8T1c_t-7^reK`v@q5%+n(6m@)*#9ifb%yaHv!><@x6
zn9_-M9}Hy#*E^w%pnvu$P$s?BaAD74TvF^2e7iq#Y;_!#1jS4KSiM)wwzt0@5|s3-
z@TUzFgHgcXM^L=*Cb7M0lCyWTy0glA*wgBBiguR&ZU(S|>I#6BGTdQ#p&z*2^7X6>
zieT^XrRp7Xio!elY(^20%u+Y|`V$$x?pdfyiu)8qg6f(-34a+Hs&;1zq79GiG^d=n
zkVRj3WT%7jdLhfAd=hx`7P-pD4pOYw3m-c~<!xN3NCx}ubs`v4A<{G{UQm2ZimW*P
zhZkx}Pi>TU(&oMN7x#;?vToU*UV-mYpT-nn?DqQf{bz$$*Qeeo9IJ3MzT|r#Fpm=u
zpI?Qw-c}OUyMHT$ZtKcp$rZudY+Xl>y=MZh#@=3rCMwUBXDH?EL_*9UxbTnj!Zm;E
zr_&dC_+0lJr!?qsEU8Bg<h{0stF3JQI#NzRjf|hBzaV9|7c^ewyie=N0i?klDKn`O
z*1f3JlFK*4d)1QQA?L)oX#G3N`k@KVJh6<W*s^|P@*UTj@VH)?n(NNgYNbw21&m@p
z;vG!8a)J6E;~}1x0hK5e0yj68QB4LDm%>IXD3@jPDyM(pzW@9DWhe8pr0^8JhFkmV
z{k;2c*DeyB7a(c)=f^G#Xas>y4a)ON(hooX9lq`UY{ibsB)^kBj2wP`ADSm$GtOQz
z&XUNO3TKcEz8;9Yz%Z0>)^Ac8UY%yVG_PDUMhco$77g*+q_{j`#iJH2*^6eOg(Wk5
z(lF?TF*kqw!`Mp8r;~$=c3UYlP%ANbX_SVOv<E=Ci;%WLqy1=v8Z8?jR7(u;9c(om
z-};bZJt39RgFNPvHto}VP2t)8g=E+bQ38cuYQATV+o7$DFyL39eHDj-Y%OtNvku0U
zk7#rCWbP}20=;V`LT%49+X^zcSkY_7n6Z|J8<Bsz!Y*%w*Z81atVe*i3tVRLP@*s1
z-=>%d_DX&xPbD5Y&{X6EUyoXCl`u-RvsI1@L&@@BONd;g6gP22)gDT#(KcoE_#8@n
z?ZcAdPnSd8Ou&pABX6QnB+Y;+FtFdJj|fBcWbV8Y_@ITM5rsoOWn?zylnuzg({3HT
z;PZbhYD|02NECmXG!~=<tKqH5TWicVe02GNBnnOjqum@W^1WNC@k$U!^h_$qeyNT*
zh0}AXF~dbC_DkoyYn7Qxk5Q>FX769pV7#zImHm|M-$;XY!by7>p}=|iH6T+T?J5oB
z^zr@D6$$5jCH<6R{<U8^-#uOyo(U-SE0KR%cnepxtam|=HQJ%z5Uzr{91PICR?SIW
zdb5xow#=blI)MPJwu|)P>X&Tz2({HVF~LWte#z*Cf@*s}C|&xKpH_U_nrdrA`XSR*
zPOcseU`4e%6l`O;B!L=gOuD5alcrE?RrcO0=SI4DyKqHQc`#9HE*(`UGUI`u$}xXF
zKFtSu+*N*J9)<8G(h_EeLN}&-on%ESK#7o{3Z-eG_cQ4?`O7Hvn23FPM@Nl-2Q}kE
zb{w7Ue!<40W0260BfgVoHjq4UY2@oj*cS*QFqN<^fRVr8#BT~HAt`d9Jb_0u?wZYx
zXBMO?LZ6zd;j$Yw^BL_HZVX)oZDN1TWA_l707eBosQ8a6%ttTGM<UJ-m^|}l8JQ%@
zNnbwj7ZJ#XpQ3RmvhLX6Foz6eDWx))(Sne@j-?fHqxih$JFqC6#a8#$&Rg4Hah&tm
z&uxE&jg)NcGx{s2G&5QE!QF#aOMmyEk@$iA6_iR?+_2!U1tyDsxz?|RHK%{sIGLG-
zXWCp(?&&mGMxRMwE$~{S!8+kpyY3I@Gbldb6`04YQVN77-(vFgE`#%85oFg)_9(1x
zC;xUW2Dblwv)!j0-$AX)PxsNkw?!MRO&?;Ktpyizt%>snB~6vW18MUr7&+;+nwC}0
z)rE1L1a@I-m3-FsgcNzcPpN;*=5TQFRKIX&^o0T+iyqChL!s7sClneJZQd)dGinVR
zghR@{t+qAc5PI(mNPEKpNq9@o3I{Hfr12ljTv{#t%%zdoM?p8sv%qHyOhxPlG^ddm
zQB$RKn^EQ!tb<4ZsaEne(vx>dGaDCnOlQ}I=N4EHO5v*L<l<5azyW{8=&YVRkgOW5
z3NIJzp<kLh_BRPzBQ2n0E}=K_Aam&YivdV-aW*#d7A>|7yn;@kre}Ml)~X7V$^~!q
zOevXZzQZqjC+F@)>V5BQl4({m+s4d0)V6idf=lIv_Wz#+%5t%7AJepbAW9>&eRdpH
zO1BEbO6YGJg_V`}g0LBO&P5WpSsZ#T48Y^c9%uo#_xJPef83;Em!T{I69Y9kFqd&}
z3lo=a+yp3>ZTTv#e<!^B_WqwI$<zAjZ2Z9s6P(qT=42fTlDy7|WJ01&gz}p>Bl0``
zCTR?DHa2nD(wH?ySgab;mMKK)KFSWyOpobe;UB+y<{kBc`Z<kBdNHJRqtibC&$*|u
z`I)3-9%z}C>iD@z*YxtlzLBze?qqQM-BqbW=7m_OB$}F~e=%)bYBzNp@YvGpoZDh=
zblvP<v)^Edo&@1V<8x0-u|6#8BXN@~!8y~H=h!oj&vH~5WXGqs_V?^xoo%8-$I_B?
z#TLT17j24$xJ}9A6^j?37S`M11(9Bw2e3Y~P2;anR!+2t@mUMWin%q(Tz(n0de-EN
zjnXN}_5;>tf6+qb@yavF)QZ+Z-#U0lZrQR}@rz8g^YLU0<WVQ~(cU++RF>x&=W?uc
zuv=wbU0P*2SV1vq)~`C>T$?Dh2=v!%6<!@QlUqCylmH3yXo%5B-sY>dP-YpK*B*Q5
zD{IpiJ4gM%_pF~34RvpoH1<1>u}p*B>U)lXBG?!Pf8*=t>DSx))BhGnH5i0~!|#7T
z#dg$6Xb%Su!iK{?e;odN`tMAWqpTE?M&u785C8mim|A|}1QR5wMp}xCL%39u^hG8i
z<tkG(!}G2y4x^r@=$@>!5=2Rv%1i{4OLhwb&#WelQkIV-{Ln(eSHji`Wnm)BmM>D(
z&XilMe^XOuy+}a~v=n6wAa5+uwmP?TX+jX7;S<Uvl{q4>^g%^NiR;kv%eKNNuAn(|
zbU`464O4>j6~%cGV4)VZWG-os0-TkK6T#kkbWUi6J38e=(HuYTg>jwYA{z(3X3tMh
zG)`Ef7m7IAr_)~ckg27oSoB9LMw=EW6F}q#fAX8c726N4z!WPh%wcazNoW8kwU>^X
zi0|5RBgAID5ya%a8{S$dwq#LTDh0)<S$3RF0y7WmlHjBc6|CyFNVPpEC;MLt)mxC&
zs6f8a>~$*Xx&win_(xz-Azw>TJ$TAVy#s92FD=;)X|5QyFlflfjfC^k0plrAkyF#-
ze>M4Wa`vUQ1+ZoM5SP?C`vPYx;Jzy87y&NpSQGTHxNu3Hf(>{Vn0#*<M{R@)Vb0@D
z2}irPuSn!b$jrMPW9CS3$3#97-bmzQ=Pyd+NX&!z>O?+qeK3(v)P6`JN26({zh@$!
zDGq_+4v$nKNH<dj`DxoW?Gq1W7mD`<f3+GZl@p+r*?+{QpVkSOAe;#^uLpn*STE>V
zDB|V0IuHh6nd^yXLOG><B)&t-g%pkmLOR=lKZXtJT)*KCX`IrFkiP9uLatWS`bo>9
zb?4pcA*EV`=bFmrdeE?wAD{sA`MGfOoAsf-&QZr>sYAHuS_^|(p~?Intyh#fe~|4!
zrK=I`YVE;W;n2-`Hj%>Z`a`T*fK>BHdA?nDkhF&kdKnesbGz<9o@ZA{o5em_`OuK#
z(?}=zOK9soSEO!BM3D|`m?;&2{8_l80R%e#1L2PLS{MvFrk07~j(}&Pyn%S8-d_ZH
zv>gr_z6kP6Tpxry6SW@#d9<eVe@PnH(Jv>8`#_$F>JE@+rnnE}nW%2t^m1_Ik$1o$
zAH-_bN6=94f&^SB%DX9Q34*q{3q`!2VsvH&GsP~`&gRlULCY=Xx<lcba-Q^y#=C0&
zT{A)y9^HnHTJIo~T;~4}u8zSerOs(NtNZ5^(~DX?$y}ZjEmQ6A!kj)me-jF~-mm_q
zxh$qH^A8l9aS0PK{62u~HF(;s;B92Y3G-O6zZ$W>%%|ogMJXp{Kj&X<95g?y`v#st
z?H$9N!@i~IrOouAeFg*&&bGZ}KDGM>K~`iLf2rrLZ0qXIpd$F;j6Jk-O9*B-=`6IS
z3gFeM9$7}y{&ALaI8n<Qf2q3@xQSiCkOvw!Q;c>_QmS?mH=e`~^Lb_IPwmu`b<~x&
z(_>m`5Z{Vk%W3wai}L>jTapyMSVn#Ll(x5ADR;AF%bx0W-7&IMo4dXH3!KUtp+U`v
zUnwIDA%m}g{6G;5!0!mHCW<>k{)zGi<ez$f5#$H-;>!{L$n-(Le?JoYA$T9a3LRii
zr<^G6gZC$@R~<5A`Kfmc)VEUXcoBzBHwnWF)cO*o5|-p~YA@F+Obc`=;6`pImg4-f
zoM6FJ#Zo1A--U7*_)qdb0Mj98;lfJsN*R>52~+O|Hq>mF*S5#%uaBj+3##cYkjBTp
z*iLwp2SZCB52!L0e_(A0ADlU&6=C(93=}K>#m^)PbPt{Imz6S7dH|r;WH3^^0?_N)
z8++fO@1he$9%tGum<@nl_)nu|eRGsj9)vM&hw{8~N|@mPRS;fjBdl!4n(L`OcK0q$
zyc(x_5nM{%guw&N?offD;MR3rB|Rnhc3xX9p_VM?K72Koe<G@DH*>*vIQzKUtpWu>
zAh?N{-n{Ue1)4MMmOW{$XlEy8HO$)n)F)lofNTt<k1(sSR^!(@;KbNo@ReKJFa=01
ziPPY8W#!mkDY<PpL$Hn<yC*V>K3o^+eC@V$-je@^d1^K%_;QWMK*iu!_lC*|Bn&zd
zx_p>5Tzk22E0r6073C|P7cRbh=uoY{PMow<FIV^1b%zZnAAgmtS&1zCrtR4O0Y&Y%
zml5MG69F}sF>^B(m*4>_)qi>a``a%&S;XS|>)Y<zjs;Rm31!`2y%tu&@bT;Jx8LRP
z2fh4dkirEw{5MVgIw+}xae<z{4FB4FxTGefyijNc>y(hFtD0V<G`<L<Su_lJ)Q}iR
ziZmo5l2}H?F%nY-56x;k<@boJ+!YE@3CluTFKFQwp`}p9dquO5E`P6$9=?-W=cB(C
zKk{%<LB<Nh>PhU9Zcr5gNh~Y^8ylY|_GDVx(gs>hX<b^sL4+B<h{E79$5;&DE^|4h
zAVa+F=S!2B!V+8Q*<5C6x;Sd<bU#N*5<)C%>N&DhY=gyJeHl~b_o2~f6(c_HV=ScU
zSmFUoOq8WYU6yvrn171kA@3JgdSU&dw5P(ZEnc`PVJlvicB6z7Uf}c)0Zp*NDO{BB
zc4Kr@Y0*P?^(dTO4u=AdNz=y$E~D2U*ZebEV`jk1B2Jl6aKCtzRXpI$;+=KK>TGKw
zO&NgG`4nSl3BabjTJyCCHa^B%|7+ZcD-r`Jf^>1C_yRQuO@EqH^fs8``=7(7l)WY-
z>qe56<UdAcJ4{A6mjy|=Oj>Kqq$UC~mBGOfjKY}gah9$}<Cl#~cAQhh3{gGBXtYB%
zHbc6=$RqzrtWQQ7_W|RCC6DT;!m_QwLegi?!(ogI&UAu^j(-vllaZDO)|CPQe)OQ1
z;)=(a?=72bt$#j6Z-j&zbeh98R{-a#5>Vtr7W>rzB`Qq823x3{;lY5A#O;I%=z|)o
z5z-tNf5$|R3pu>ZpSM(`hM9C+=Wy_sOk!Sh6&*8741Oyw@~x0;B6UwaQB{SCE0#Z&
zyv6rtmdA{mt}<ohwC{@cq8H2J&4p&x0I#Q<ah6&IJAbYhu$4LQ0x!ic&*R4gtE$sz
z6bgC-(LQ<Xk3bN#ieeTYh#)=1#~?^Ybp=6s9_EpB3xXg;8IE;yI$z2r!e(B!f1>l<
zrf3;GshW8UpN67|VwEg`-qvJ&xC-6qP!<F#CG8Tr>EMMkej@uZ;}rOKV3aZn>iO7{
z1x)HmrGE_Q=hqUrzJL*~2iH3AF?-I6pj@qZGN?akqqvrJ<vPw!yHxYEFi_UiaXW%g
zK5UHuli<|VH75qwZ2UYLcfnAp`3A;9<C@2|@M}3Q2RvRc#{jKzT%@C~(=u~w_KHtC
z48tjGnCF+qf|<v-{#e&vm-f*3l;S(P)s{<ft$)~N;Y3ii3r(49E_#QdicnK>?RH7_
z@tJeT32Np~meR7T&8DPvm2GdDaF-`e63cxS-YJ2fZuITi6l^9@v$xS=ZR8wO7deUc
z`QFUBwJU&9QtCxlbUspuF(-W+2p`@L2_y(N2m=h2&~X^zS;1PWwrLoFQa2&f?x%k~
ze}DS<<G-mE89x6fuAr1g#|sFaYVP#Y7+MyTrn)=-rsD-fT#2|HO=YD31vG&d)`5@1
z4ym*TO~;jS^h+Ei-13B6&Ek|u9ZjqoG(*foq*zCJzc1||b2V;}WU6e$Uy({#8-LPz
zUKM}FtNqPWY1LoD?$faQQ1ux!{iC~e1b@K`>ZIfYPD86w9@37yHK1j61V@HGGujGT
z2Nj*(EyW|!dC$ohnRT}+ERQq-^AOQu3OvO3e*BCAI$5(yN(CTHpmlU=x6vt~Xf<ng
zeE#^gEc;B0-Z%G5=8615ps@ob9!*73a`03jk#j5-=6O2LW`ug)YtxhzT}hg;B!BPR
zG=I+#B^Evm4U1XSrz-1*tJMP_jKM4P)0a$*UI+kO@KrBoGA@e%feM0pNYfiJ3<0mq
zOPC65&U>6&5lok@Ou=T(%Q3U(T_Q3TDmG`e+A_*Hgx{({!4<SF&*=j~nxVp8SH8^x
z{pY5s_F6Q6BcFALQb)x6fjYb!b$@q9!x-t<?2P+imYL*D_KM;|oHw7FitA9b;$k^G
z$zBUq;KIEx`2%5C2>s680bwZyM|WR}!7-q65dJ+(r<H8^)TB#B6HgSJ)k}4%vV^Z=
z@39u9Tn?8`*j2^xRfVxKF+q=?gOH2XNk;(hD@9iL>}g**!Ebf`vtN5mNPk`4UVeR*
zR@ZtyTE+(JjJhe|o*cUg`CGdK@mrai^%ud>rWSrkdq7e4cgk;Q4~#pcr*UH*X%DQW
zWzg-i;3!u$AgmQ&RZp#}%(EfZm9wT92uOGJz{W{OF=e{u69<1DkcZq*k;m=L6t4xi
z(C@^bKW4}ZRJ-{ZUx)4c27e<$tN}Zn^`3J*MG7hb!Sxiu+|fKKb|oG8pf6j28#~b!
z@pQ~Bp9(4IJ4l?RuK3!XXZ5)<?*Lfe%dyUhV9DoE_Ja+9zv|nujpocFeY>L=8b3a<
ztg!`li&s0OGZXZCobzgqf$8kb&F!8&r3Ug2&qgXwX3_vIN~2^vuYcK@G~2aE=jNC}
zdPCh2qTuN0Jv03*L^n`WCgt>A*w<y%#MZi2i-x{IJ#;1ubN$?7XhW5z#1;Ka>KPLr
zVQ7m&f_|i>sZOD5)zUYBar~|Yc7zQE!*$?=JAhaH-+)&IA%lM;@J=bVjm-SrVmsuO
zGOMF}4sIEor?d5pkAG}a85_T^4=d8**-XKv>NT}AV|di+bu#kny2_`UZai^}dh;Tv
zo>&5}CDn~NwEv$OOBLjHqci+W**TY#3c_@}pK3sz<Lh|yr#Wv3fyS1{>Su%enc-4{
zBdIzun*NK*f)2Std%05U$16U=E&RvP{{gslrWXojZe(+Ga+hhh3JU`@I5U?af(R3r
zHFhB>m+uNI5r1mQc7Waz_*V2#7>#>~izYqv?{|i;CAqs^OI8ddz*-WQ91g!3&NG5N
zF~L6W4iCEz_ewLzy#XiL!xsklLf~HzIA{%X0Ng5I56|qkYayh?=YWqELMeP4e)srk
z_E!o#&H#J^M|@y^<uu@k0(@lpl=t`?>HGeQ-qG4fD}SzcG+U(EGE<iKdFC-wxWSoP
z*g$*XdocJM_P;;;x)b?Wf(J`q>0|tL_ptl68xLu`<;q(2_`IVRS^-yHvY@!tfqnmz
zec62)sn)#&E=hlkVBeqEF!DLW90bBFtTm>C83balnYDo{dX-?-w?Y^^oJP1b9$X_t
z5{5%shkrYOVnBPvjk^)t_`|)2#a%VdRc72(BK8KMH~6V>cVp7d#^rI-aZsr_&x4#%
z6R0LgAg<IdQKa(-6sQ#>(z+MT2=Ps%&oyCJ{=}1YBcG?erSe>RH|`C~oJuV@>g0^m
z$dSCxTpTkOMzQHn<f0pCxTsPzo-0(3OqbL(o_`T0NyhI!<7$$suj8Fi%-Ceq;C2DI
z<DeVcO1Y_#%5v@eduwFBYt+avp;dO)CUfFn*2tdPD>M>ksJqj(Py*LBjQfEe@HvH|
z;ZDw^0Jm?T7<fQRBgK`QqBNIE&a^;~uHmYJ(FOEOpp_N!z4ff$wOGEWSGQR+Y`Tf@
z)_;a|UGs{A3nRsCT_Z<VEW%`gW;gqAzE)b#@fkTAw6e~2k{c;!@(^p>8Nx?rX_fow
z-^8&(ZBuH_@v@F$6u?bTiyR-zbG9uvL8A9ph3dLoQphvm1SjK^f|!Z#mx7r*x)rt=
zW4VQj3599;5Op|*X^E4447v18;*Rt+`hOwFKFF4!EP14ir8FLcBa|IJUKKoNqth*%
zbLj<Fa>nI}@y8klWINRygM4C1@*(PjEuG|V(N0QcBt{liO}8e7w2LbgMiuiFJ?0Tb
zkwV$tCJY#7`Iwu$`j{+{KGZieS#DrfBz`OsE%J{%6h7Sg`3fS!b35J1Z9LlfQh!s}
zPC>hzS$KS7#C5(Y$E=4Y<8)u&F52h%b_uVzA5ksHBs-pQi0dL&^;d^&8i71V<k3V^
zCQ0g0N9HSiyZ7EILQNFJK^<fh4Z4Z)hAb1;(Qk~1#?(pGIerf|#x(n=bgEbX2)50n
zGAs0(@-`Dyq$+Sy91`^?Vjb{`>3<qX&%nxRKY<YimXx*x$8{tM*T10VcBJlG0rsnO
zx0lSKsuyiAh9PqZgWJ!zY+MZOEO)-<$Y3|G#;{_+R6t{rO;w`Vc`7a$n|P`8?Ak+?
zzwhuq*##^hdm@(II^y?kuS}znTS%ix58x`&ZX4--MKnY4Jd0*1@m4U%XMaoPdfPC^
z)5@&x74630oP}6D;oZZmZlOHJ^fFN0;u}K<KC*}d`e+rrVgE5226D*O#t2zyfI$9w
z_u-$XFTXtgf$=hX`j=J+h7rGk;MsHjsSr37fPKxsEk={*TQSBCLRrnU=U@#^C=2Vr
zBim`|FdoO4U>f~OqXE-(jelY0v^W-998D}6G(&|4qBxD_{l2z?m8-f%WV*6Jzlcg+
zOFwaaSw%nTX@8fgxau#q`@(jgDxN{%znFFiy96hugWwdjkd`57Icg#30!MIo30Fqj
zgHBVq^%R(D3^yFcC6x5!-Rfb>NORyagtQ8UGE`X`##aCs&YB=edVgs+i4=23FAQlV
zWLOil)Az^cyzU8CeQZ7xohy)v<IQDPT59ev%tKnSie~&`JejAL*~P5hWaE;f?s&mU
z4dy00w6t?EAMGU@h$0K0Si@pE>(zAFPp<|DfE)7~fJUp<03(3ZyaKb54wN(G*#($W
zLzCnIHcC}Za9KPxjDLRK&HQ|u9P7qnE(D}Y!kJ)L63zs463*YOpO)?DRbxnBQ4?2~
zDr2Bk(LC<a+Z2GR3!gyf+_q_)Uy0(<{K)||VVVwAOj2G?&N))|E4jm&8A6R%r#mvI
zNCH{Q>lC_aiH2hUXsHeovtEki?=n~9PDQ4^3_nIm+{qdIHGgY>30cakOMZ81GXNRI
z`#x(1fI+_^<hyhOFs$d=g_CMxWd~7fV%<nv;m9?xVV4Ez_u2MJMa^3D<Z`4o<Rq7>
z7LtaYRxRAB1P&Cpsx`Hu^!ln+sSpl3NwFNb20%CLXM7z_-!~W@(4CK%!TB|0yc!VP
zNO7pool|X7!hdwtwnZtF>RRlH+%&e(H{Ai-_HtfIL~=#*CXc!l%I6-7uJ0F<fgjRw
zk*FtR-!7-y1I4)=K3%8sZ67GF>9W%pBEvXaE!D*j7w;@cgP%IQb7x`L^^HB%f!b+3
z)m2+|MxPbEpEF_PXRViDH5Z84$o<quSmW&cdc9PE`+soh)@x!k70~7=u5@I#PIpP0
z0WKCsE@(4)imy2ry(>Q8B>$8=-LT_|Eh)Zd<Zm&yJfnk~`mF=?l^#LqrJjS8IA6mt
zK&4ytQ>plJ^9s{=j~SYyZpl2q0_&(!HP!4&tAYyNwQ>pN*Jg5~{mjvqWG&~Vy4L-4
z4i<9zA%EAsw=zP|Xs)L^xa!lTi?qy8L1Sg7-|0dvGjM?1$dP)FNJ>g$ni*`{Pzr7(
z4Qs<L9cPw$b<~@irdAzk#E`Aw`U1E~j{ytSwEB~E?JwC>DPFitao!@<`|5dL2e%4n
z%k;<-bzZEMjsBFBjP9YH_b!6F?m4F#&brLv?0?WS!|YI*53cPk1UFCoXL_$J4^sv?
zw3xJ=N;!Q^!R7vx%m3%5v~?o1_ibe(DcSkiJVqi)>M@zOA~px!BbNuzTr2y7bNOMH
zH|FvKwI7trdn(R_3x+N|!5^5(_Zw{_@&i}r=kWtxZ%^ZUf)~i*otPT_KVb^LXa$B(
zGzT*O0|fXIYnP!+D-@Se2Nx5Un-eQ5e@l<tHVnScukas;B8qwf!2-Lp*#f;a*jv$4
z(d3jv^ZyT}Ubalx<FVIWpv}SK8Cj<I68VW7IeZ)B@XO`t_44rr_27gz;MDN?bpU!G
z=@TRkS~EBRVG)Mc@5Aq(q*Ru!C%Uwhh;%vn?&)qGJ|K9S0rUi}biu<1r-4Qgf9O*A
zu3qT6#_taw;<NZ}qLtovakk8}RiUh2%FJt_@JKVYv>~oV&%wm?9{za!^&*Q{QUn{n
z#vA|j`StSmg$t=kNMd97^ZO+}&<Fy(8kF@S1Q>q&HGIAN#>tM0NqQ%L7&-j-Hn5Q|
zMbAN!o~5<M%=8SBl&=SCyg(vcfAp-Mr8Fd+CVgp=T$7Fz%%uGI;$e}ZmSi#0Q4FM5
zvYeqyl1{CRS~Wyl7VQZgIj$c`kl6?cGOppYunzbdO3pYOXL}Uj0-v3^*}rx$B535)
zKr$b=pQ4glXwyjpdQloBy>QZY((Kow0u_CU8w!pK0!}FCOb_NbMP-Cjf1P{E&&#So
zEflInLmyC&wzU21nT9rocBJ0W&ony-XTlkVwpvKQGc8S_iLOmc8yu#OL~%CKZmSxI
zv=bWL6gFb0P{oS+kuC~L2Fl%(@MLR1iDAan8I&xH%QK9MW0jV9ETi%1QyrxZSD(v5
zKRwTHYd7hBN9^4UyyW2qf5Mr8cTO?NQu)`3^8S!#tpKYy<+X5IgY$egc6?-;%9yOn
zem+5;CMaPkJfEjeMlpF338UjZdYCRN=;DulzkW-CVk@obHxMN9zLQiJ-MYdXRIi_;
zulr`?RW<U|{S=?M2dJW&5<r*Cx8Iczf<OhF`ITIU3Waf<2{L|2FB^5y0Lh^$!hrHb
z9x@GjM*$8tOtXt(#1EbP&x`UC-vZ157i`kR{t#?@bTB<`mv9y<7APaJiTT!^34s~K
zr7eV?4@YB0eZ_^q(feRKUhdGbmIokZMu2R!l}MMa7Aq@%5}*oRkcv3-D`x<dN}xnO
zdrE*2@8)EuPDplI3$M4#u#2KWfu=LnN~}WTgoWF_^|s6glzZ{)-Jh6R-R3RJCvIsS
z;+amJxTP7zld~10zLBjsFHD~YGE*(vFv@G0sgz`H@^RP=HQ{{<{r?X?q|gnx2Q`tm
zIU)}@KwKz)tvVa|a)+<ZS5+jV2~AipD{E6Y@hjfuMNl2J8_~WVDxMDQ1*Je}8~A$c
z@N=X-#wk9(fnpHB`HrHwS#jfDAWuKnYvqM9;l2HAzH0I_Rs8C?ct%@g3hTaB&iIVt
zwy(vyNsS22!I$ssr{i8(YI%&awequ5!YI|*L#}auw+eHQbBt&OV$bmkh9X-U%KA$j
zU+Z-|meYv`SE0%%V{S}MT7unL%B`k}>su|SG2<IQcoN=OTxzbNi=8(!=An%oV{B}F
z`Gxvv87<XP+5vUM&hTg*1wI-Ie4kcGzhM03vlh~t?S6}t9>-a?Yrj!nTkS@qlc;X?
zyRL12!9*|8k+~gM_iW-Z&q!K(0Q)K3)wmDLYz8#6%5USww-uCm+Kq2ri(a79F^p2H
z5(!@l<;PY(Q$E<+n2ZSvg5!-zAb?IF{aqT9i18uriu-j1ImZ3jNV|(moM)>f%h;KZ
z-G!tJ)@9UG;W@RI@wAMHbeI8cPCNlHiYIoL)fg*ze^bJ)@ZV2>sX5M8S4&DB`5G&r
z?ehv3rV`=pVnYW{myKH2#9V4fWf{D^MHaYW0CBO5SuLLd;>=)2w2xkBoV87-t-S>_
z?+Fv>zh`geQaPdBR3eJ-AEryoyWszAup1l6E{y-J!)l!w;H<FLt-{*0S8dURZ;lH2
zeKQ@Cf3<F)l`nui#hQ*s1wit5JZ=Y@9Wi7K+$(!~O@D}r;7Q=TO?t~<zZ?|mq3$}B
zsPZ1J;D6L(k>My<2a5v0cVOL$fE!pn;qaB9UmOx_3|co*&3;@F)kvoZMwulno_shT
z2bxV3enEE<t>LKHlOV9c*FCM-lp9U%-)S~^1|#U=%&(=FQW`5Ge+sII!Zm#vQ$n7*
z8SX%#(A~S?cFh85M=0#xDgkCzO(@=Ku1$E#L5C%~CV=T#7}xci>UVxyiP%nyoyKBQ
z4yF))&A1%_%O8Yut)z~mozn!;v0~<GRYZ4n61)T3m-S}ArK!`nkxskw+Zt3_H^1A?
z%{s)krj<F|N@Fi|3nr@{FivybXB>4om;V|oFae5}FdHi$1XxuOF_)oCD-;1Vm*Fx7
z6PICeIVhK-8!M}S5J#Xd|NQvdljLJb5p4V#KgM5gA5Z^z8V_lW5}>Ak{qYpXQG?K4
zO^RLyGyVDR>9?o9k3{QMl9uE@MoxeJJ`E#3Sjs_?l%=)CbW#RMqQ}YFAas0{rHtRD
zG;}s3X(62(NkR&`Gyd}KVUr?C(wA9L3?x?4o1sU}nUXMndRqP@*1Hj6eZ{BoLX!ZP
zYNY1ymh3CHw^3%55=yH=SGx-&1siqwrU3W=hG1dLMWx@QcJ$?~Vu2;_=rU?tp`gmI
zvf5{%8kd-7VXoQpK{yje_Av0OD~AeJ!0JqCJ6o7RtW5eio>?OHtd6**W1x#;pkL#m
zfmwESRUFlSWmP#$6;a)V*SabM;5n`rOP;$3HLafWTS*qEv}SsVuWmKIwyTbDR!==M
zhcAptWjvXl6iAQP1(&PQAHQBz8X6$8@LTfFywjr^Nx))fRnEw~t83@ER3(_fEO&rT
zhNPVBK*0*i?r*L3;#ebO^0&gKd8s3kM_3(^vrIF8y?tRN&8*_rZXNSr_~1zx^q9Hs
zY_}|jceIsre3%L~X!0Lwl$JiAvKTU=TZ$?>s2lB!K)`V~_|E4!8BB~sVcja41T8Fv
zFNr3`J7Fz6Fq(`M6^jNVU*%6?w;O3A2nGST;!Vkpk2>4XEW*gR@55~Pvc!ixl!Buy
zmct}}-jnk?dsO9(pl4(raPz2&V$51+(dSyJIPk$+`~}7Iu^9Ym*0fea`aLb2qDTla
zVWp{JJi}o%qVVWedJT7FCB|9dq&nj!^vx)h;gGm`FOrAvJ)b8`cGq4F3g;i0{f!iB
z_IKWHji!vxj})UcuILynh#=j#B4+t$a?~GxbB_4CPvPjX9IQ%6V$D^YoG?hVB9NmU
zIokF;${5nyN-O7sPn7w23k4I^x|a90JppC;3O4m!9G49RjmqS`yOA6Ctlna+$>e$G
zLzt{KKgjbf%ZiBIW&5awSGGyb*>5HbeqjrT#?n&~RIECc6=en^F=;kwy3k4q)Lhbk
zt1=3vzTZs0GS+{FzfjS_{0V<SI(I}K^aOFEjDmLL#{r+x@2ZOTZ19ep)z{2EOL5an
zPy|4|I?IhPTr~F1Z(kZOT`Sx2d>&8q87NOHvfbV7Q1xs4<S9%#tG9{*bRZxs(fFnD
zj*Q2G;^aAF%8h<3AJT|=U2CBx`1W>x*;=?UO1Q9CyWvtmo99L+f$6e-N}Z^2s~pd3
zixuSpBaYgx{w*VgZjHyXDLL4Bd1e}+VyW8qQlu1VRS681%{+07fa%#~$Asx()4rWw
z>Rn!QQ(N7cil)-Ybxqa1RQ6}1oR0Mg78*;QQ&uG$tn`ktFbiF2+GIp=@7U9SqdmAA
zmw3Zi`k5lvybF)+`Bsg#bQx@=)buBq3O4?gv)`CUBoz6&uILdDBM9AH2x7b&y)*(=
z(x|)GMz?k3D$=k+43={ID*sJnGhyI0EGXlQV5N8Eg{LTAdU`GQsjxRPcVH9UxsK}6
z(<)5>KUQe-T4wn)?QD`;ZXdjVuoUEmw}VE<k7_a~%2OSc{locVT-)hk9cn5_s!7a>
zdc`U1q98YDT7N?$Z0AiLP1f8>6S~F^`-{2{WX&vo;}~)PIC4-NyA?U462aB8>SD1i
zbu-wLud(c=w>TWVv`1mwlBI;p;s;~*(c*Kn&kkqGnnRp%kJQ=H+0E;Jb|S+O>Nnfd
z%2V!!IGow1%<OiHS_MTb!YHli{nYc4%d$U*>X~hk;Bv~+I!tzJsuwM93x2UZbLVCC
zu&-U)KWnY+H4~j}TY#{r8+sR6nLh~q817J}OR4z+x^}nyrXST#39EFKWr;a6M~ckM
z%u(Qwx0KaWUh}pqKQXs|2d}&CMdH0SajUNe7rwzoZ2~t<>MdMsP~1U4b4Sobg4QeE
z@^pj<&o@c%yp0MCXF};``@%rpG1gJzEu-;5YV{?JVQu1ybA@BsLDs^t8|hlSxlkN-
zxpvjF*UGY1fK?;4p|r@FxWA+(2v9Bj$d>s?aq9y1L>rEqY6I1OYwdGK>%oq*Khyqm
z@iGHer-4s5jKO&k)Ji^J28!oP_n2~}IQKtj1>s&R=qI#+6a{Z(@PU}^DRLrlCvLM9
z54_%S;ZP7cP#o|xqHUO6E#W-Ml+!}h_DD&#{^Ce86sw$jB+@#HNs12dM;rUSf#18(
z#`x(eKEPWkjJ*1PwFC-E_OYb?FR=ukW<D~~4iwkz<Ij7op7Pab%O0tpt+(3P+%|@{
zhh;9{4V4}E#g_rYt|yx&qje3`_AdeIUTV1PI&mqey+H5Ay|3NUqY@5b#c6j<Uw<u;
zvvAydxVbj5N+w;j-;K1_f|T1Bv*K1(KcM*vCP^l+?uX)kY2>8x&MjE>1uWYut-E{u
zUa}Tze*R_s3L}sAWe~4Q(0#S-^}8n9+h}4nP4b(z*Hb%Sd-wedDIf^v{g{b8QY#RI
zcktC(>31z2F=^p@)12tK30JO0HPta;q`0q{7;%dIES_m)2OHOypV6Zv{6MX?uJ3Z-
zf^__R`*`{VFZ^a>lb3N`C=-{-zbXxvd$S}OmKZDw0X3K5p(ZDP-CA33T(=Q^*RSAX
zTA(}}ayXm`0s&53rwMAnkX#_Gpa(@>#7Y&dAgvnr-}f6Hos~8#S+|wi^uaREIlO$F
z%QrLRw%S4pwdEW*<NuOFPV9{u3iU576jpC&VC`=#2o>*>LSyj;2gRo*6pVBWFVGP0
zi%E&H&&)NT;S&{qL9117wLr!C3>v78?<`PD-j7ot#UolA$XdYy74VX5H8ID@q2X;#
z>QgX<OnrgMA!qSjV0XE+4n#uc;tMd|Rp}!%<d*tZOsB>7R8z<%f<06UC7CZ@b0~TI
zDxnmBXw9XP`m2H#`^0!(BsH`e2DC1ss%S}{#6q#;l7L`;FtlO`$cw3e70ADIC{hA*
z>HS*ZipQqtR=b!4I*{T^U=&O9-C}#(av#cQi4SeAD|1a?Y)xIG4HG1CpsqzSAAxPu
zt`tq^SFmNY(2uLHp_q!jOLH*@^{-{rDq4L~Y(o<!RcihYlPY$KN3_iPhYEA?>6=lI
z+(lqHNa@;tuo8s32#kZs_!T4EO3;!Z+7k9*EkISixM_fF#aF|-EPXXdsbB)U!`Di_
zBg$?s_@mI5_;!`L!F2HEI`jdsW&pzyu78NZC(acu*QsU+*qJW@KhWYj#owUJzFkbE
z-6s?>)arJwK&Db10Bl;_#;O~!HWgn3Giw((j91NnjZ|TLmfT1c7H{3;6vk}b<P@?t
zH#xK<7>{3Du-f;5=?y;Ptpb;$y7kpp!|OdbX@Otw3_tw+AAwSffaGQs75h86I6mIr
z-bT+bd^10upY5I=9L_=?5QcXLi^c5hB;@h&&z~0G?k)}%vq0mEH}jJP)_4P-r=;9h
zc?0QxQibUI15A$KYvThX0Rph`0lOlGNXG~E2dAH%?d{>++5B*Kw%8lqy?!&ipM6@4
z2#5FoKAi#6PcuM0S<Frr=gtx1<U6zT`Ni4ctiS5;hPSg1M+aZeKkX$vikpDteT+ZD
zNSH7z%X_^#Ihl_W)p7Z)y-sw#!A^9S4ZCcA6rXKpe~G7$A9gQ(T=>&pk52v>hOg&m
zA7*EL-LyaaW%zFRW{<{)o!Q}Hk8+7_wn7#w)K&_#0z=OhbCLbwRlw}K!?*MI^Wk;)
zLKH5P&7?>ZF=w_vY;Rwa$zw3Eq)2pyK}wGKGax|0lq9PZ19bec`g9O32<t@VU_jM>
ziDHFfzydK_T?65vgFy#yDoX`)(#q38V4>)|HGwE>(b?uU5ZHcV&`wskp}I3$TcU7(
zEBTtg3H(jKO?1uI)_aX!Srf;q`hVb+yzZ6!D6iyeURh-F!O1<#<ep`6Pu_%kYK>S6
z0XHIyrxe*4QzRyb`WtIF{CIE*b;O*1Zm`6g*McenEj=J(dL&#a3K)o*MswRMDI#y#
z-&%n_yjsqFxHy`h0LI^TzVm-yd|WI}&%YdopN<wEFMf=N^IwLaFMc`w(}&r=hVL)l
zzKcH}4@c+c7qj!>WcKTD_;Efzp8@gKlRcPY0tXj@p_^}@z-f>XWr?gea2eu%BMCAk
z<Tc5#&5}3Z!4=XYPXi?)PeaybQ5D%B)o3pRVUh#Y+5@XAGG#cBFcjJzP9l6o4~=f(
z4_@aFfw$k?t5h&PnEvjbWY~mlDoa=)niX<d;iGFz^-$KSb&Yjuc?|2+cEvh8u?|nH
z!xQW9#5y%i4VoJH={%IVMd^iq5QqrHn46TRcM_@cnnkg}hKLdj3(km~u=3{2nvEh%
zdTG|EBO(&%b+blxkCKgH(+DFdKuX<2nm`<#{gL1!LQ_XUI!Xi);sz8R8jv_k1#nb-
zc>|)zMWY}ZP9U<vNl?bz2I6^`&Jq?~VnDSCqPrc#XIbX)sUs4k?5kdXqNoX+-#({~
zXEBOP(-cHZ6uJaZB3|MlgIte0gvAYIUPeKf1zB_|X@zeq-zRBX>2;a36)-*s<3*FU
z+w!y(mVGFFimcJ6@IC0WHCm<5mapiuPGngpvXqG|Wg<(-Q=>ecH49N}NTMtJ)%>`o
zRTMO!ELIPDFGLmw%J#y4#!9Tz4})Uz69c#kvAzt5LK+#Q?dcIi0;K|sxZeVd+yo*M
z>MB1v(1sLV8rcR=cp5ENARw92ZpDDKjijWUP}mp$e-lxR8H?2=1Xlth<liobXHW(R
zFj_w@jbdJ8f<1CYZWqDMYY?)4C`~yrLcIv;|5dh<dPjD6t;3Rks1&wm_vx5D?Yl?E
zj6x~ht0y=O^$h>lD%*!rVY4+VZ0#{rsB}ezO>SVfyGz7E8EGI&EO#Uj8;k;_6@ou1
z(wy^?YQ*m^j*kxB9xOf%yQj0mql4o&7bkt#ISR(B^3b)~^o{`e>>7Qb5;f)DD$x(6
zAFf0{245;sr&wiwwuua1J(mv^)pNf`NEzY5JrE>$BQrfuCWScRC)sVxu9532Y!yWP
z(Bv9+;<pE%rKam0t_s%%({A_Zi4YY<pXK0sr9RzINvcHaHfWErLArJjW>ey1OJgik
zgQrHF8dLphQ~hd}DVblIj14N&8{L3W=N7Y{C|8O`2=&c>6EY}Skbb1~!qwE*)yycD
zI2TkBn>f^+KoslcL9VE!6L(NG-Ebb$8Y%*jq=l=c>fW_M6dqBNUp8jARs*UQ5#85o
z_?$>kpTO^mC4#Bc4T1p4Bu|JDN3=l{cn|U1;NZqw#n1b#Rz~ewHz^84ujB`WjHu>!
ze^lYnWgSI-Lh0;Rw?$j6P>m9A4n;L5S})=W#R0N1q=fB;JdELoC>vQ(!WRPVW!oSK
zO_9K5lPD^Px3zUql$In_xQR7sNQiUWq$E&3XOKU#P*f`YYBnmZOQKRiy|3K7I<N*e
z*}!oIqE1AlDTK)5r%EX45mox(aijInb3W_2$sc2X7TrYH>qqY;)kys}1W3@$f9k-%
z^vrALs%~$=kNBG3Ir$NaXVjo609yA3mpNy@-(|RhpFse6K!v~RO4OxvkK>EgUY9q)
zWH7?}^vvovg!UYv_qJ&JlpJ2io7$DoJ6`8LH1w9T482~mjgd0T)My*Hae~to>=HzW
zxPWMtOIKm_e~{VuVdn6p6EzTHAeK(i<mY=p<#!s!_ChGUs3dMn6vdC8{dYpQp#U#~
z%oyQ95W9gvWyLivQ@Yy|(rX=-X{M%rxcfV)S9QD*U9t7?_YC*5^($G7cdU2u$V#uv
zTP3IomFe$BV!U=&q+A}u4s2KKz?Vmx$}}}La1X>aTgB9J|EWfSP%09*By{UfH9qSR
zPdH7?$fS|RL*WAdEKMRKVGsoG{6dFleqz^11B&06YwfW+BVk}agJ*eo*krzg8+iS|
zpA`uL-=`xo2|x1tA9IQS0IH#Oml2RH6PLTBD-M@Uf*=Q%fZ97Lf889(j_kPY-d|Ba
z&>HOv&?um*x_f}!^3b;+ryw(@Ib{C-5K1OxiC(MhH$ZZ5Ew)68J4H*9m)|e)^6Sg(
z`{kD}V=qBOr-HtG|8`OEK*FbzaMQa>P)c~CF5iD#{_!TI_Hf<cvQioYSAgFUzTNfJ
zs0b|-JW+SJ)b%x3e+74pf=dtI`U|dm{(XJT&vM@atMm`KwH#V&gsi`?<sBisLrW`t
z%DCXU@)_s#pWlCbkx14rUoNiE<bxG3a-u0?bO6c1*c@QBfXOBJhH*JJ{4c-rvzW1G
zI&B(yIT(}mO;0`CbH0pei@)AE;qA!ZbT#e`p=6%e%oF$Ue+bjA5dv7?u5tbipWZqO
z-w`@Mw%FGqJQwz1K^{ak#{*!zNRIO$YLlU^a?dCul(n3D#3$!zgpnxAG<*)cZ78(c
zWAs5JJrPp|=Lk{5Cj4*Gz$*HW;#?!i8ZV-mC`1{e??`Ecq|8kZ*hi8R^>Wu@P(_S!
zU`{W-&_XQge^W*{QBiAzkG_uRjwAAfKdWWAGR0wdg6#m8%0e;N0(q2MW(VOy7Kmx7
zDu%6ZI95MgRl!w-zM<Tp*lS|d-U~p(Xe+80?56sGUlJ->&Y5CNQO+r1p=wqUeb}jp
z7K}wrq_L@qa&!&kD%@>d6J-(!FmSosQSM9={j4VXf7ocEU)02oZwxBBybSpr`?#w&
z+4akLT{TLCF<iA&4?G@6LHg)w=*oCN@)(E;#vAs{I(VaY)dJL_1*lyX@G*ofk$jF^
z`RiBq@ALcRKNBKBONEB`^{+o(@&gMro2V})CgDuF{Q2+8Z!dr6U@<mIQ2PshIC=T=
z_Y1r8e*@EkBtbW%2Y1wkEhRx=K=VOIlp1y8qv0v?TB|N7j~L3Y5i#8X#1#wZy3f9_
z$3hlGe6{Z^>=(<Kju~uSdCj3|9CJxH3`$~~Ia{1@!Gmzp1DshLH*$eP1kWdL)rzIC
zd2sYm#@yX$`Ra@?DRPKP(bbKM^N;nC_s8)=f8ynX;*hkQ#~UHYhcv<a1l=(RYbLT`
zY2!l>lM~6B3B-9yw6wyy3Ed8)Xg(jNMk}3ddYzcDaS&Qfl}J7%K&_n!daC5=X+~-d
z!6r;q(mfB+L}6o?sLki|5Um$!j^l8dd_51@!%Rd?)JT5J%Q9%;^wbb<&#O{&0<I12
ze~+ypz)XcNA4i*_06(_VdW`hKxp0#880E67-Dy5XC1JEWfo7wPfOPjn(=i&-0kczf
z#%Lj<J;8|^n5Y;N&m;r_ol9oZ2RL&XKMKU|PV1g282hZ>W8}QPda2Q5?E9F74w)t8
zdbyR_p4uX;k|*9Nccr_aPxPF!UCeDuf0h|W)<2M#ZRQ@3u^WigKQ}AmShfoQo0(3C
zFv;eSu@FGgdOYe=jPA#)KE}+ua=|Ae9EMS#c6U~IUt@cZ<v@jCi(55l^cdaeFq6(z
zojju&bohxZk=*<y+a~V3?_t8@zC~(b<Wywq32JreeJ?Pu)<kcOna9eW%Cvquf9Zno
z;8GlTO)G6x_4lVjPEDYDXbT}n1?Ugc6pmXyk#3_|^^jP){Zt~&C<(_iAu^*L@=_**
zDRAHT*(69SVZq$@iKnK(hs9G<Uk2$wGHomb(5a&N1F5eT!A``Yo>w95l$h$|85-oH
zfNQS?L9P)(I7@{nkKU(3=wOd)e-Js|T_o2$`V^9@(`6*P^*s78&6M>rlAoX5)qHl9
zkjL^_E20g1#gdCM?ks0!yec)qeVS-;WK!u2vE<T@jwd<_lgmFQ>{PHCv9mL`?yQL2
zLtDa45u4#Y5V3nmXGCmH$$1gGhrAK7S>pqx3yQ<LL{ww#VG-5x62rStfBB=*QQ!Jm
znw^7**_7smFyoR4E}cO$%{Uk&bv#(_lO7emnD*S;s{QI5c5&uRZUbh}c6`@0OcHQr
zhLZ+&eCBKu1b$r?9@OeXc(~B9IFrUofQ?Z&AK5GQ|L^LD1*h%ZPcD8iB<IyWI>60~
z=s{fiiD|r)LZ!e)e0udGe~KVILvrAy^mg%Ma2!3t^J*W-h>+|y(|E<A@9<RVW5~`K
z=B>)dkXHAdPAGiXyb2|a>s@Y#(kKRPqwv96<4TTa(x{zy@jWAJ{Y?6nl<+Z5xOZT-
zOh(KpIVXj~$_MB8$7&A6T0fi1U>>zkMpTD5E+fjwI3#tnqQMC{fA9tq6E_Y~9!|G7
z#1pD6SozqgLgZdgQK9O4S2;g%u!mKMN*|o&12qVIIk=@mpk|nzP?FT$`7xajTWduy
zV@=#sXc%1j;7Z1jnVkx7gxb_!c?naaEtE)p)Iiff%l#G14s*9J>{<ckCA=j&3biX{
zpu^g!@mtj#QzuN}f66MW8o`>iVne&>S}|OEwR061jt5r|7HIe7RWZb111cTNS&_1b
zwo!0kt&c=XhIb=U_N|XZO2)ktA|<0l9xiZjh~d1r%jCBacl*{y<L)kr*SI??@rf;H
zEuyk#*+LI(%NF|9M{HpqUb2P0^$}ZO+&jS*I3=XHIH@oXe@}<E>L^wzn2Z1PIttJi
z(_c|X$!ltpi8JfNF$!(hQuO~~EyX~>DD{bpSbKnFeY|es{nyi8q6fURloS`9_KnSC
z6oWivj{2Xctq3;831!Vgg()8ZyfPP6ZKAvLV+X@~tXXr5wwXpp?c&-K?h_fbQn#Pu
z-k6;^gzFpIe{kddvIhS1DL+lsXf~#$+A&UjI0av=ZPsy*YmVr7Uy7rO^ZrDf7T$6E
z)i>mAbd_6_>4t8Vdo#At`{-sTLX)|g>0X#pdiy%On9gWhcD<1Gs6JP&<Bj9SG_$I4
z=si<60r_}9cYcE3gDIu<;jgyU9&54HneBa<2g5Fte|1~!q3y8MW9w!Zz5}c3>75>j
z_;Gl7IfuC5D{%b`*MfGj{Nr|`A-^*+c3GGl<zQ#_<_&V0Un7XM+g1I@8)~s{8mi(G
z9zH?`crKaBt~X=Z`s+I7KMdU`VWz;{4DjP-u5A!D=)c&^8CD9(MQk6vsOt{jxH|b@
z5VuXef4q6Q*DaD1IQBZ+1F!3`^2bKJH0{KNO<Yd$1h+-+O!!C^s)ksZn;Xgfm|HMj
zIAz+Y(ql3p)0^^y>}DoJ&g}`@+M}MQla1k<`&F5mZ^Y}XRdsffJ<>hjF?Gi%Ue(Ld
z)3$y$+*#Z8%W%|+-f%cT-O;+R+ZSf10d6rufAso3SnXY@eXw4Xa}6V^Sr5(Ia707S
zh4(+kouNnvo9Q7(KI$G<t=>j;$l|rezctm%P<iY|$5Ecr9?7`M5v#yYnFkqSApuM&
zE;FPM;fJzm@k7mob(|Ep1i5>wMrKkDu6Mhl?CjQG6(a-?+im^T*H@EFw@G)nLxQ7!
zf8EC8MB^Z=rHDKcPHw4Jw<tfb0l7eOp+Zpg+y{@P@UpSrOvgF@F1?|v@Yr;HOmCcZ
z+wu%&aNGB~GRd+CfJtCWn^Xk<7kMv=TXou?hI?hhbsBYHS#s43C}nCmL}eLlNq5rK
z+2u{;DtKW{Y!`jBn$T`#?@r=q3Oq)Kf7Uj9BWv{EFb$*@S+jI(WZNAWFgzibTE|dv
zwEWPs8}oJCdqhNq*o=w6-#Dy6#~tM_Mw7vd#!b@KJJ0N}UPE@o#r6|UN`O(885hGO
zZ{*$Q)yZ37qiJhl6gQ0@eR}?9t4d6(e{FqW=r_DqYgc?!80{$^>&(kDCX*B&f8{)P
zhPg<2KZj0`+Nd<VDZ%w6iwC(tD1l^;W0QD*kn^gk({%<%)e@{M6cY-Tg%spV$6>pY
znIO0NU_(%*C#5VaAk2*8MABOh5oSOV<`i5%*3DvV98P+nT%3d0<|V@dvzjb!rGv`l
z)tIq67i6ryFO5ZMFQnp~Yqr}Kf4s;u<{{bjA$f=)kV`YD<?JaJhx3#|m!{eAu#T)K
zB@=PnZ(5yJBC9Q)x@J8()q_Sx^&C%Q!YoFo<_H!17aDJid>F<Pn@BLiucI3xMO#)&
z)j(?AP{raqn%?_Y={OR{As1+P1{L;1UyamnpMJ$KVyZjq5cB@u=H((Cf4YK;*GjR3
z;7i7{X_=E9zEY?`4O8U$z$QR(Qt)P0!Ic3+tlDH^qo=MJPgLgpv&mS;6Q?%Hp3Q+y
zY$AYn=QoGasZ}$S?L2KP<<}s3p0F`#CE{+nCr(?xlQ?acPSDYdM{*(brJ9$gzGOrL
zjwv(hlkVi9t9qM1d}stkf5m9xHiC%*2uH5JJ<KO=9yDd86JBe^`SA=FT3otZ?c4Cm
zIoUBbFa&uCXvc<{B_f-fWjH93l`zvR$(gg?s6|j?_4+=VXYGtf$dxEPxTtZSiUzAE
zMYwg94pu~I8FKl)(aDPhM&EZD&T^^=dUH}!kpp`dgJ(riDz8RuMRxUssG2eluE;M1
z2f_=qQroRmS<SkQI(n)DOi#Dka}pO#Hmx>uhY>0=h_G%D)N)Z1*oj*O3$IY)_viP^
ze*q*1r6ZT2Oe++Zvb8f40x~z3AuT>9m%uwKt$zj1Na)=^U;n=0d0Cu!6aEYz<G17M
z=C{qbNMjV^)=l4jZo&&tf+;5_VVGBD`u*4G`{v6?wH2WgC_-+G)}06iN4Wx0_e_OX
zNrd2Wg(yP(oF>fxyO&?TzfD8S*UxpnQZXakNKw3mXApt$IGhI>QYWfi3sU`b@_cdL
zN`EpT&B#UJya5FcC~%FNqj8gTFZfkSXv>UNBSn6=T>{21VdO|j@@^ibw*_Xb7<=om
z0$hpav=g|xBd!cp4EW7+lH4)NXYvBEI43D&(4et1i6ZWg=4q!Q+pguzwn-@%;2tli
z+YMVKm1E%SN6Zh<oR(IOa}$H|UKal9{C~A)R?nm$;ARemq5;N*nj78n_8Q0VnXU4a
zj7OsfgMBg}c&_Tl4~;o2`I$?8+9*JNrtk~&olws*^K^B_H+x5}apT~lyd&C!w2yR-
z9YJK1pJflB3@%0tgWltq0GRUwoi(?l6;pnk>4UwJaF&6Z$#+Ez9;3V)XY>f8=YO0D
zZx?p?7M~<6lR|&M?4)K^&Sdmpk3=4|*ULK`5=e%b)ktu~ACcGXm-x}{NL$=8Thm5J
zyyB*3UXg!E8zJ;i`O!Nmx2bn%4;cqMQ-T<`$I!Txi4-a`x7YSdTuX~d+@dTXQ?fb7
zv4x_~Rw-)c6K4=Pbh9^QhA?|aYkv=)hh*dNyKC2Sp4a;wI8tX8_i1q85>Q*MI`6@=
zF_A@8^P-lo*~_owlG)2^t(kB&9KNKLM%f>djO1u?m+iVbU(9Zm9=Rmxl0gSezb;$N
zIp#o3<~c`zqFUc`RyY?WBPCPZ-yoUQd6N`+ill2mx_x?g);&<ZkaZ8e)qiXr$=?KD
zBaRP~CpeFBPu`?y#v*x4@kG(IuI1}xnGalL@M1pOzOI}>lkZ)ToMT?js70^5Bt3-2
zh6iw*a)sNGONI;szk6a+j;JHWmpdqW<`J8Z6qh9^F%6d3Yt#bX#4`k<^;hKGwX}Lq
zI)(cJ=qU?te0bdl$XOdjRDW%0UzRa5KF9oSRX;*WB#*^@ID9U{Th@V|3@s6iWm-V-
z3^~ILB^%)PQ`d&+24xd<{TS+?nf7Kz-8_am>Wxu$M%_M!dVnU@&#1cv>JRXMTrg|I
z10%{v@eYudq0d0^(G@HsnCEuna!&;-qp5-7dw`-xG!1H`v=U1t2Y&`yjJ+RS!O9S2
z+~<`FR>mJ=?+4&9EyB(&sM{@QF;`$dV0zYI9z7S+gYhDUuDKQ86O+AS*3T)f07mZ^
zihQ=nTM*;z5f)98IbP$n4-r)6+$ZoWyj>G94aM0@C@s;_``l_1o=mMMS$a<~Jg<%N
zEn3$DeY8R)bq@aGEPr+28naZHQ839BR(<1`r4H1t$Wn=y%rWHjZcjDvUdslvry6H^
zYPIVfQH+Yr<1EhC8l{yCG0|)`x5p6aaL3hbT_GoxE4;H081*hk@@O+M>))x5lD&oB
zXXV;P&7pYQR7FVh$;K%&R%i7yBgk+B=y#&q+K^Jr6z&5IS$`SqZvSdS?zxg7NBHp4
zZ(kIuUx%DdF@dS+K=HCM2RKg~R;50Z_&W_OpScXPx@cf2K96g;Q)v;>;`&SKwYb6-
zNX>ycfuK%O9e`PUV^`m}9?~wKdzU;(yU8JTvPLbHYNa#n`eLcpHAcp((NWM<P>QVw
zq4qD0c@_v<^?!c|#$m>n%E!`#17o>TdV5l_JPdh;Kt<G4G%jW4&?4Gonaaxq!fm0X
zYZ*8en~xdb>sVY?Jj%Xs$Qn@AMV9cgwBEFTd8En3>ckbu;g<JQuuP?Kqq0+9I?aQ+
zzRpFa+Y+OjO!5&_4$OS21gPv@==r>VHkZuf8}bB{fPa+|SLbKi-N?Cdw|+{rq9)9C
zUTcJTOD9FKv>rx=n<a>-gI@TZ(%W`Nn=}xba{eim-Q!L~Ix~%`M*@qQ(v|72u@KQ(
z$a6_tlZ3CZl046--@eT6R{bb6&V|rGF?-bG>uf`W!;Ha(vxDCz^eLaG=vdI#AdJ`3
zpHpmRLVs!q;M@pdP4GYpe}aGAeEQGZ_rHDn1tZ(^_DAT-CqR%VyqkWeuNuEQjY-Lm
z{M}%T1L3Ct;+26(qxVu4JJ7(=@NL?Fv=}+HYg&JVR*!iiV765U$72s0Iue7XiRKU}
zmR92Tqiv*FRTgngWg6jIppwUiuQ;AYg|FdhzkgF#9QChh^L^TU$x_uu+R(-EMA#er
zS(2(?AS@<?jby1wVakP>pCPGui<`dE+OD@WN@ko11q&Ijxd;eZO6FG9OfAj2sR?Nj
z1!|(tv3*4j!&<SWR+BKS;W&*cOjJ>krJE#N|Nizfk9)&WpPSD_=PBGB0BxqHP(Mb~
z34gTWsTPLLr7)GdQ*R1Wl)px&mEBw7M8UW<uz?7&{4zCDN^5(ni~Dl5K!k{e*vkSL
zGpI%o9CI?Miy&6wHLXENw2qrn2nqK($QT03X>g?12Y=W$E|GA*yOuX}Cx<!gfLuNa
zToaH&1PYkl_X6Vk3xj|eIZCyz@^S!blYfg4`pU>A0LK95hBNOfnXAKmnNg7+?M<^A
zU8pCe$tpI~H0L0u1hW3Df4DJ_QkR{gw7aBctHCwQH1?+CqPcf~fBOg{d#JHwztvKf
z?6=xtvfqDCeYYHDpK6@F;#RJ(p-`$<^J7qBwqOYVIIcn{Q5t=3EU9yuanRD6u77lo
zjswJu_n3q2i#n7XNH(Sx@LSy8(`vyR!`;ZGY<*f;o6XEJ#L!D#o2Ny3cex>uZJsul
z{55$R$ZPYoBlgy2Tj@u0OI+J4E}$N;H}_}@X{z>2wYL(Zq?+T4PR7e<aaOZ1o;m3T
z+EowAHJtuhP?ni#9GhCZ(9`GLRDTTC`{ip#L9T_SZ7&%$bGUes$hJ=KP$kk_8fa<B
z+4ZRp#w5iVuBJt_=PKRAZ2o7U1RdU5j@vfq*)%IY?xfi^K?W?9@9E-L9Y4yxUiQHr
zp7sFd_^G$a{c*HOx4o)7zJQqwFpP`mF_Q>+J*M-a@`qpq(a4o(Ni#1*1%FEpUwf+y
zpfCu=jH+0PzIRZG8at#^Mpjfr#G9tf{(lH3iGnvnA(h_@kR0Pu7#IN3V(b4aKr-=g
z?GXS;U_KN6IpjKkM7^otzfbXqiu)xoub1<#08Gsk76DUP?2Uj4?gC6!WBTT<#7hp}
zsS>&*oo@!W<79h-G<|b@j(;k)jr(_E+Z;C>vw&T8)bSW$J`FFJ6>^wHI*uuo>&3&+
z?x>U^gvoxn@w};*^cjH2+FSTk*$#E{bnPSco98HPev)EdctVaMbDb}NoOJMWI18zO
zdvtE2+{0o;2sBFX0Jiny3Cd2Cq>6B+lWp>4R?5A;z4@D{sXl32LP11Xhnk$XURjPA
z&Aw&_?Ws&XILm;hzlv<{&*o5J!asZT6MS^)kFYl!+w@E>1F&j&tN05>{8f%j^5=g6
zAY5^mK{y!`0yQ|7A?zv?0yH+4VR;xQm+M9=q<?LEc7K2Vv4OlTz6a~HAWh32qRyK3
zp#APmx`<;)rXiVzu#BTo4>YQ69))HG)~_VnnBAl^y`}t`XPAPjj}fuVH$|$>GLqzD
zUgh)3G}k9xvJ9)@UP%KTF+8fnq)Ga&XRJ7?%Bmu%(JJX(v`T9*f1h`~&j{Ec9iug>
zc7OYfGVQ1$kqxle#m<^&+wFJKXRwK{Bbav|cWbHz+SSJy8%}YZ8?76oVTb}N1$bGa
zAlV^dq{!2*`Z%t?m>Eh59~8%f{lY3M2%WErmdVm}wwrA%;7eBV=wYLE-n1)B7Mf7H
zxzX|$SBAU-i_bDzj{2-+ik`GnkLPwuk`M*@!9J4k^(o<%$V3`Lch$F^NGtyX1WZr7
zWS5~#D;1Y=`v()3>~}3GxAIIY<_HNlm-7DwC;X<fm!V866ahJx5qu33mkCxYU4L0>
zimEEso2-iLzIC$uczAq0e0b5ObHXDy)xCc0fL=)Y3z7ycb`C&T4c+UH?jQG3DofWR
zT}Dc+>2mb!>5j)c4W4EI(nW6oj(5%gjcA}t^>_6`*NLBxcl?gOjj-bPj%Uk$wn~uI
zOPYB~5FYYO`B=Q=nGX!>7~d&-jDOT<`bwUTcPdB|m^zvVmhZ=ZzW#cUi4h-O(04%?
zrKFwXf^|mFS~C~MHS#l2KuGTUg&4;dk!TpbG5HB;4yyichzaJDo;}Z^XB@t<<4ojd
z$8Y;t`73mh{7b^Ks6<>*Pqy|wM3WXT6nd{ZDGXg+qwfA7cC@3PnL!?AF@Gr}#?*+E
z9;JQ7az<$^EX#WloUgG(h$?G^DTM%o5MZ?sV7(BcmiMd-mWk)v$LiPf>*4o9Eg6Hf
z6f&zXKMwrDXb_s14xa9+`{Uo;*Tbh8YT}L*l0NDGpzK998Tp(HJ4j-pv=(u~L?Ef6
z>8uI@!(iCvi5BEXw(vp~RDWWLI?zHil3WDV5Q$rng;5w0a$=QXI`f}M{$zy6A0S&R
zz^WRuk`uay*<_^FK|NbHbo8!6(kkIO(R7)m7x_t&wN78bPvV**@Sr<jL=o#`<ns!i
zPqDJ$VaUonVm6p4@e7<IZBz|3<`T|vkDRWR>~@i6=;y2DsuT!gLw}7AOvRsTAfj-p
zkpetzfEapbQ$fU84k~SgBFn9{K<=(*=xL?=>Rt7$!n)DH56}aOU~~=rvfl9s>mKSf
z`$j!PB(66+t@Lo59cWf)BoJkMJ#ME2D%IeAY5dz&Pzf4cYggpSQ#Q*V7DJ(b{1eem
zMoLGWm6Kes3qzhXFMlQ`XJA_rC`f)%_x-z=SLrn3-S|#djM>4QD3|aqytg?`vF>e7
zQ|On?iDhDLP+_A4rTEQiHmENGrJpsxEpZ`Rcm;7oT*zLag18VDENvX15lK}3m#o&=
zETY26>%|Z>8C|Q&Fi%nNg7!kq?CZs^r4@*-5kD+uXc_{gw|~&$-sW@+Q1&v}o5@kj
zqYp~IlkH@!4{27mGSyoeKKIsmey7myh1cv6G^X!Dg3Vhn+Mnjl%F75czIsf*b^jR+
z#jDZTYdD1hB*|R!USiD9C+>KNt+42d&%0?@bkjFbs@RLGhEiOh6wgqKS5Ue=pcu9<
zZ{9heR9JToD1R0775Q0#^VZ;!Ja@ga4Z$V(;tJxH=wiK~tAV&7x}>bIhPV)2oE6Tw
zM(5;YkEWAbOVTP6G1kr@;$$V4gBDIk&gj0@D&nu{2-&HwI!DkZknOA1g07EDNV_oD
zQFSyQL_pY4b+nU0J99yG#@R+#)hOB|a9)AiJE6l2e}5JOH))gr#IyLX58Q_60EPdL
z+?V5_aeNYHOAjoc`m=?bH&!g{{UWxpwDXkid~PA1>YR6s=og~)Icn#QNuOVvLs#uV
z7^V5p_4ke@e(d6G0-yu!ng8nIFsmfx#}roqoMo_@{xK5{AO8F8>n~saMM1gy_9su3
zouT~-1T^3M=zpT21lN-j`<{MV3guQ?3gm%wg!kHl(*ZSnI1GI04n~ndX!OPCcODJG
zcoWYG^2e7DTPqfSaZLjm{EMNaZ}}&EpTFXte3ifRRQl?-?(nrcd`etJt6Z!o$R2R8
z^ly%<BqMMdO3vmXP2NWVVM<&Hj%;sXv|R5nsD$;F6B^}(Bi-Z#D`6YfoJWdv=OKcX
zA>^Taug4eoP_30l1|`LKXl=-fDA16KLgH@E;qm#)kM!MtH~Q*haqr>WfzraD?e2^s
z-7tz_#d-kx`3g=yJ<ra;AiK>N!=t@rOmh}i0U?y1x`L#BZFl41KCNa50BthO3Wn4%
zYh;8DisbJGMl4j9@*1QelOGtNPe;(vV{yRn$2%i!%5bBv2Oe?v)1ZSV9vuovcvSh+
zGJX?t<_u|nO;M_VwIXbx9cX$;Zy4iasu|8#@-!$x*=N1{;yiMRakQfsPBn~^FX!=*
z!C`2=Dozg;@o7pPaRwb}e7MhIi50-J1XyxLp}{XN<LA`Pn1)lls|>2BIE`{2wB}(<
zXW~6e?>w)J)}+7<t(Q}sv+LlT0x2BU>yRTPkmu=t@_{(v8oF#+-T5jtO4_}s=Ic_+
z6aLynvq9%v<AN=Wh2c^n{petQL%BFY^J(9ti{vuzbc?d@vL`Hj&YUM4y;|=H|I<yU
zrHtjSNLf}iYiLWz$AdUm<<c%Me5{JQ#sT&H6!VpFchL#;U*~K+!Ov|68Ro-o?FY*E
zSn{2JF5Jv7d`*jia^zd+)f8Y(;IFOnYi&r}zNzwC(s%nVReo*g?r_H{e+_Y#*VmsB
zU4K3Vj*^Lno~24xsyNrwm$kb&F<fd4QXvR1cD6Cdj*US!*+`WDY-kcHDr*^Xw8m^_
z5z4A2Ay#<}_{L2_)<TVGa!9#;Jr}!#0#RpwYGgx`P`H$p>EPCiqKD;=VDC%vk4oJy
zE+NBLsVn2DI?AvqldyQ-x7Wo}1)s4+h(&yhv^?_LklnT&C_9a|H_Shcxhp&S1l0{$
z{vHB1h?JrEIf3seXq21>+8Rqt1R+Z%^7`>Ogf5dKyPIYf*%&MeJ?ShP4t2(ORtg1w
zZc>4f9Fs?y1mBg(%x$qcJ7XH@>63$k-4xFpAM@f*Y+CCp&U=Ql3gV7UYZc@RO>5Qc
zcWPRruqM26?_8{^ux@NxtDwDk(^`dZho-d};`XMs8shn;wOR}IZ(4&uvXWs(yR9Kk
z!paRc%V(IbhT2q&F6;iNorD8y194k_&Qohn1^xXR=Te(lh5vk;SuL{L`sTD$LQCB&
z{du9|gcJ}|Jv*}5jQjcxtl%95hK8O6lheisZ`?dDHWkyrI}vcQK%ac*Dm|MPkl^XO
z#>|zPNCdgKv&i0D@EMo7W9tP+URfL@?cimm5Tk7|BsHEg4%wL6?uAmjMTK>LgBMcL
z2!qf-+{p{2R=5iKoAn#!hO<ljxdI{}3%`cArLLb^+-iuo$8@c<K-cKvl7mr{sy8ZB
zb%Yd$ag%P4(>|VvcYn_OgtGohN4HZ=9>|viY<Ia>yWA!D6fSo&?OF>!yP0;on)Wwz
z`v8>2`gh*MR9JU%sY3{uLSM07sVPHKh5yzLywr<UuXM}*P9W}b(R>PV%N8mpq7}pq
zTd0(*Rsa`vP^tC0R>5g|_Zz!gDTOK<(f>!?E!P^=W{2f|g0ppA=%VZTR?_`uUpsp;
zSKXgq4}Sx%D0G)$w+a&jH90Vs5qu33mnC2;D}Qd}y6rt*;U9>OD3Q7VLC{P#3DCD)
z<gMt#!J@l;*e!ZaANu>Hu9jtc#vae60UVEIiQ-KpMapvcKFHzi?(}~5+pEUG3P)ho
z@cwN8ej)ikNNzMVSO8%(4DUaOzdcB)3|}X{P)e=&vi#feogVKsIPL+&7ry~`ytfG4
zqJM!emG0`5ujlal@jko@V<W6M-iF>X^;Q|Odd)qbGlZwoQ=Xc)q32V8^%&nN^BA$w
zJW4)4-YYNB2kf{XFb&`z|Nj2hJDC~LQ*A5?rmavN<|W37_>N2gKIh*I4H<i4>=meo
z(30jS#*K7*qf$HnHtC9^BeXCoV2>ulfl>;GBCPyGd32o5);<x7M2Ar1oEoG+zMLjK
z{K=HfTfe-q#2qY2T%i+E>7I6_N67`OBW9m6E`cJOG4T=OmaA*Jr6mfBzMo0H-i#J4
zy^xm9m!V-R7ANkN)(=ItMRrj1A#p6~@2Cq!uL@3!GB0|ea-UNkLpS3fS|y}E&zIg|
zD;Iygbx$wvcYn;5Flj5abi>!5ci{zU5Spn6Wmr~b`12pbx4X|h*o>wWlE<WfBop%{
zEnl+j^pdHO#t=<xCoq2r2SZwb)rFd{YQjn`79?nen({{qV|fP91BAKa#FLg$V)I#%
z`)cZEJ+OW&WY0dpxgK(`zYHuvEt>k8^xl6+$hN$!rPNw&;b<R|bTM$Y{EKu?TE2|f
zD8o9!{OGycb8-(vC-~|i0$mT$8ey%SaqQ)Kh>iqbGaPYIV#;|jS3q`ziP@L6`=V;|
za#G>d$Ex;p)`RAJRBsX<bq{@mdcOqN`xO`!yFZvwBd=Eu3Xpn$oo?&{NUR7mF~fhS
z0}#Ei<Yy3RS%1qK@6=4Tb|eZXD4Ux5q9mFyVz+p-VqawE+<cvhsE*J^Xzga$J~vB_
zz3uof&Od2s=C4akvBze&R5`_$9Mv#VoS(CD7q#?L>mHW{ccqstk!*_bFE6DKv*&R>
z!w!C*<L(^rMDQ78e09ptIt}Bdc+G#5Y7eVOfMliOa4!g;2^*p{qOCznQW&sNVKN&*
z7DZ$8!(m9+*5npv$9MWL_T-(=)*r^+tzb_(us`k@h+Q8V_+Zb_&3Z6Q?L9*W{jz6R
z4&tx(7&$=csg#EUln&w%m*H4e8fFkb9S(VVh;%OG>D5n8#A2Fw^t9E31S@}j$dZ!@
z-c<b`Ecr(BaU(NKge(Gm*pT_eb;f8-Zifu~D9a<XB1Z`h57YYe%V>2gv@UIUFJ%D<
zII^)-gLTbE-7Mb#ir9Sg6U9wwjwK=!wk4Q=<6&SU&_jz$sH6`G3(UNzo=XaIOsUkC
zIUeOrFde=;A=dT{DZ@NwSIU2IT&Q~!A_yVOm?cD<62^Z>I)vjyw(xF8WN^vrcMwa8
zG|Epk-Zz`W+CGzJUTUYXboP5^utLEMo3*XE0(MY-#tqpdKzGcoQ7EnEbJv(}5eGEF
zvxvJ<H_Q(pM=a=XDe4KpRSE8q_=ae0zdX`Te2VRwCge?a$18a#)<8G1{ALW6bXJ6?
z<o#m(!{->pW<@D?pD)*3Vchf0Hwmao7|^%Wk150@FFHQ&7AsifEeZ6qTv(STWh+G$
z3URd2q9$9Ef7Db8EMX>(TXFLAu$O>kD;X}JQj-cs6<Xw3f@>1IlAwt=oeGqgrco`H
zA@VRy<WZMjSuFM}=QLUY9;cj>=a<K2D;|HIa&TA}bs7bcg11!Pt88D$rj>eNf?mhx
z+Flms7=#8wX6nk0#$e4g&J}JJ%wpPErr@`&IOi~w+OQTKs~nE1Fz!u@+E(~33(&R;
zI@K&&$eG%w{MNN|Q|3{G>_=yuYkxD^Iq0!Q@2Spf<2NZ<mWP!~@{BMlhfSN}EWLj+
z-KNkT`4pkm8h*j<slJ&?`Zo1%#qL=5oSM`Ew=KGhwQrNohjcSVXjy;Ch9+*k5)Yet
zc;-RwANiH~6n#vxC|tAchUrnu3tX=hQZZ=_#Z&n-n+Mi`MZHssjn^2m7b;3%88rAQ
z>td1iYrnA^H=r|PKiO&$jj>R~zGQ!j%(&TXxOml8^4uDeB)qD&QO~x-RU*%04LNb*
z{HEE?Hq8s$?aR7ZddiX4WS4DO;RF=d>oc|!OJyI%Q#q@66`5oo5PHN+-l>;{ZDLTg
z>j{Hy?XRB3P+q`XDh#S5f!Ee2A_#!l$co!jPAXh^%0dN0V|AZY@a$IcW3qn=AxkWc
zy5Ux$gSg;YcW)nzVezRR@=7_eY&N>yda2@YO4|u|x!KNLaYzJ+A6&obW(|(&ydhlS
z97lC(25m#V<`?jh7oHjy+(+?oI_m+4-yi<zAaCHQK72x=r{?epi4NcmPDu0+kDrj}
zA)Y@W(W}}1^A5@?K}yY;y=s4;N@1>_`&eFsbyX;1KfX$)H|z>-R3TH~<DtaO8{b^Z
z!M3!YB{M&qa2Px3gl>|oTd|`v*wG4h*Hodvn#oyI)9t+4bW+;o;xuU!lOwW?(6>jN
zrMF$%%k7Ehl!cfE=_pz$cdhY7c`Lj&JrsON$*YL>bJ}y~#7~Nj?e%{}B+y~XSrJL`
zTTdSn?*SX$eTQl<`OF?-r!=vCu+ew!F}o5467;jreB9-jV5~!_!0{g7@erW+nhx@Y
z5MZTzd&OzIOKjQ!QS+9?%@k>+NK?-C^|W8_=QuDG3UIfTHCOn6rFF>bdiJ>s3av0K
z>OThV_cp}NJYIWt^Z$PxVzU56>>ODCfDKN7s`)R_eD~YGzkmD7*WXzohwuLhUD+~l
zxCPz~Khq;&h!~jbKk~ETrH9~IiM`YY2Xf3V%N@$V+VFL_L$)uV$u+G%LaP_(kVBVo
zaI{A@bSx7zO$CPlv9^}^ezc7Rt8t1*Q$-^@1t@uJc;xYARCs?3SNmML@~D3fci)D)
z&zY-al#MApnE-pke@k2ifvkYga<()Dqij8z!WC~>ZB|-adTXQN5Z|C?3Z1a5PQke3
z*(%MYrJ=htF{}ciG?jbbzQTu&R&E*3atQJ~E-XJAR1`9I6Nks&Uw`It-+9zqb5C&Y
zfh`G=;UChoyMccIEBYI9%Lh_Y_tLu<p5pRjH6^Fc(o9PWtOLSRejXZ>5^eV*ai2Fc
z1V9ks28Q$&ZDfQc;nIPSa~}fc-j3ht(f(riwuv_8>Qc`bEtxv5xL8}?moMP7DY-b^
ze4kRjCHd4udhM5V3F?LHDa87UDT1tX_j$yGL;VVm`+$Ewsd>&>X>9Vkg^o(YLIyO?
z_ta!gPv>)gU<7cW&<S8e>Q9j{9Ar)hPB{=Vkr|`H%PEdH?zU#sZHp#)Mao*&oXJ`!
zPd(Qp{OW-*Vowx4P@RBlPU){Iv|*bB{lFX`pPkW$tdi~qD!4qO;~loQ-9Y);7UM2$
zWVV$D`s6wM<U|X$rGkYm%*~2V;gezP*|aaCw_ZO+0j>uQ(|~`$0FS0_$^MSEG}+&O
z#s4*o%J=^PJ%&jEm*G__6qnIkHWZgkh$ktRWNRzIf4kGKPd^`I5yZ%~bts58=33J%
zL$lE5Swb1_)g%Sd&ugQnKMBtg`sEYJ*TbrOm7x{Jg-oDDNE}FFGJamLkh!TW0|G3J
z*diuFB189fWLqy#g$3P7ICLeYp}Lt51zMn8VkfLdvDytXPP&q@nR_-p6IZT=7{?eL
zxi6A%e~Au$8=uY{WJz;_7Xv>bR=6&hO}WZO3W>T?h{j2-kmVn26hORu8lFBr9sXEY
zp;0JI@YC1lLwtc6geFaab^<+4fBrswJA59|77<G!X-@uUWL9(s`BL0JNYaF~)|gJu
zK(=6TvSeAPRJEvwA{&ws-e;~A47yF0v&4v!e*%daF>YCkz(^<mOXPQuFOAb$5J9>T
zGr+owTT8y8#?djDM14}Q$m^3I%>UB*9i&~KWknpK)ph+C2#ZA|dHKddAX>@kXk`kJ
zp^UO@>&3D^l5|6njhd|(2nKYe;SwG?F0Z582$1A$Dv&{7r5mH{Vi12lX7IwG=_Im-
zfBCYhUb0gYtgy<zp`8vGFGsl9hV;klZ8%Vx^?qCx%(@K+BLeCL*F=JRyn1aVy>QYl
z*0!tKs<RZNSUt>oh5s0Dge8nGum2|2VzqDv-qKonjBgm#2Fw?t+A#VJqM9*ExS;-W
zQOzJwHKX4ls*ym2cX-iG*V6K(avK8*e^=EP>)j+an2j^tdoDcomm)EQ)zPw)M>Z=g
z7glJxqQ}oc4_$@c6)L!VN)*+%`{|X%L2QC5;PTDZGOxiyB_5!ft^dQL@e`$%;9|O6
zk)m}%$(ncc9oKq=SoCdGi9NdvNp&9Xtn)He7j%nWcp2VXxn*t=X;!Bj)fqV6f7Q36
zgz~z!c<5t^XiVYVZ3<7g+#EX^nmaRUDs0ZWZOsqaY~nq~iEArs<*AJUSX9=eA*8&z
zr?NIB=gbL+A^R$8R1~p8jUrV)vy6A;A8m<p7_q66(NZ(5Q&mu+a%`icJteK_<o<D;
z1$7kbl&&7M)ZX>^j4ACgE4iM}e-MQa<6ymFL}?dbovB+2uwlimbXy9rk-oZqQKT%_
zhuKj}DqaX&qEOCmmYc;iYRZNDE+zOisc^!#j#vdfLABKmvXbz1(1N_LDoD@eObS4y
zKuX!-anah(3JijjIAYtRw;E`a)oZBtN@x>W>UrO3XBc0WR=E%xFz?{xe|$qJjsmHC
z8@Z4RN4W|sP&x!NV%;c-<^n&6;QWH`SLnu3%dn~$RqW}ZFBy~Xgl!vR&9rzg>qW~s
z%x0+Kc@NiXjHT4l0P8_bX-Y4Cpx2xmnO?CjS`jEOb+{02w-EzEnSqUUhfv0({eoWX
z>tRh>S_tR9=t6)@E$lmle>4h4=v5ceY_50`MH`#JC1VY<H#`3}aFHlWLBKs~2We?^
zNc1mOk+x>+w*JAna>`~qJMkh}T}1^;N;zRH_Ef00TD|K++{PYjp40BJ7RbsTYk5=l
zSj*G7#SyFY@-$YYb$+ZF%U!XpSmk1v6YsJ^O_pey@YD3OTZy0Le*{+8xx?7KJ9NHj
zhlacRVbye)u^qs{c@d&d_p`P<RT^Q@{+~TnQU7bbyU|mJ+3)13!z#Ny)nBns0Ca`s
zy57Z5?(urj#pp;T_b`lLUe2f0Jl(Hgq>QRbJ%sU>U*+@<digUA*1=EzO+%mSDsuWU
z{XI5G7Z7`DS_rZ6e*#FbFCdna5`k*|jCMw0bP(jo*Q3};89MSU`!ZUx&R|Bosma@D
z%OGY`O*E4$8@<yY*^J~exEEuc8R-n7is)uuF35=dOHrZqJ~ATWv1x>_m65+vMmO_v
zNk*EAZ2u|?zZS(#rzW%#!YxicMvoCSAXVTw0{t=3s83TGe|kvR4WUoq=(fZjMSlQl
zaY?qC{$j8_eE9eGZ-4vx8>NHk`+wq43S(pyK(syQM;kXp11O>_KSxt?l&E6uuc=hA
zl;s@I0G?O@U#A07Jp<bK2KpmHW33fW|K`a_0xk~Z1sV|35HVILulJ(`GF9^uiKj|N
ze2Q4g-1tcIe|c7XjKB7ohtjNnPKR&P;d7CyvC5@0dzldjOMh!pr7|5HX{nKH1;u?Z
z<upqrII4)Ypbftjs3;ZMxD%q{ArfFr?FDt^nIR3|3?f{NfkAv7;Tv)ocqPdERfI5s
zyo>6ZWT8-mn*}`Ie|;{~`jpFPj-WIJaZ5LWiepT;e^OjCe*MdiJe)^!DZ?9kAr-W_
z1na$v;?))sTJrPMkd#?FR^C4E76*kksr3$dj6UDUfqKP2*2!|%*!E@?oN13s^m3!~
zs5S20OwJ?sZiuJRX|ov126Kk+Nk@L--7VYg1@T(k#Pk?!yUE@(KOI+^H?u2$3HQwL
za$B{bf2lhck&G|N&NjTe!r{4M?apW_bhTA0XFGY#_6_z;0_)Oh5uok`p_%t<5Qjq)
z?H3!-3{BM8`qb|x7HB=5sbfqi(z<7=)3`<piN8&4xT=kMbis$0rqz1)$&aoFA#Pry
zp`8ydw66{2N!5my+d{i$^NY`rwBNJ|-%2~we@~s}L!Qjtx&7#CZM|$4>$)voT^m5q
zXQ&56^%Ja(kEa%m`SSCp7oyx8P}ZW{9I|ql``6SDE5*pMsYW<Hhxm<;v}?z4l=W$G
z*Bxx@QcDtFadR++tH{{>yy2e2d#)#g9J1Ir?5>l*K&6nXt+sBZnZ8mqn~R2@=5y0(
ze_eX;W+Pk6%oz8?);v<?FDl#9S=FA-W*Y9>C$`znBJ-QpJNZKZK1Hwt-RRfiZQ{mU
zE2GeS4w&WfBG=8Ql8nD?<lPW&6ZMt0&9m%k`-`t?ecDN==hG+S`_tOau0|)z>8?9F
zmCXT|X#@g%z+c0Yn#GWmVf*9%9B?WN3wNt7+_>`p0E9T;m(i<169F}sP;v?tm*Hm!
zD1W_MOOHIM4ZqK)=-jeW{w!dg(x|2Bp6QixOEkAArzpFpJ!HTCAqETurmElDhnfck
zeqdwx8JFDswv)SG-VWbyKcB$tya-l#z5D)kr|^ZupGsn*vpcVpaG-YIzwiF_E~R$(
z9Plwx0^sBE?|{GU{s*XlJ(R@<-zc^J;eU<77Et);@>`$qIi|n&KhnE2Hq*-Eed;Ys
zZ%vf-iFzK1!bj~X58d0;^CQvP=XcuevowIC<Z=H)M`>b`4*Mz8K^^x0{{H)$q>1?X
zWWz298@+?IS#aJ6)V0n6AO({il-!)kZ<{nQiKj*BEvjKcYNfZRjXBag)WW)<w|^>h
z>@iDq$tKQP;ml%_#OF}-&QgWK-+JUBA+<*0lVp_ko#Z*BzNy8e7DrN_7KxkGNxqLZ
zHua>|+{0x4M90)IG)Fej2$IbB*3(kUiKIjO$y0ip=B3T;sr3&<M|&Xs>|X14(zkNp
zb6j+jog9pRuT>r`yLlL9XVv0m?0;1jFKfc6nEi6EK-Zy-O=Xb>D0=I&Q%p`^K)B$l
z6IAt3<l;g>t-6plLfgnkC!=+|HXRUxFoqA#HY=8qo$sW1IF{9!ET-gMyYzP?t2^pf
z{jyLXyDFNRELh*$RetNVh~Glbk-d&}ZBP%_%bo4!OAoYBjtnBY6dcm8Uw_ztU*B*4
zo9tmC&aCmf|NQ=zUKmgUg2pNH!UVJX<G;IK-~PeLQuvWV;+XQo%H1Eo?O4k{DON{`
zW|0m9z+e_7(N<^|t%UR~E|^c6yQ)80Mk^NgvYX5ljl`&LImNs-$SP&!(p;G&XZ?}I
zA6iiSM%g)`oaU6Z9OP=fF@Nql?8SP+BuCi|Bx<R?NT$nEdTBmwj6zwsVLoxzJd#(i
zQ;AXLI<)-r@|-iS8pzL0T~ic=Y>QFrGZ#eE6SXu;c5U`>EjU@%Dtz5+SB(}jE^>a#
zu=%-#q7`0GR#2Z0P>dq_Fj2ItGpGR<H0a3-<n#T^d*oza?Pnv`M}Pfnq;`d$9iKms
zcrL<2=o8}~r|52vcLu%3OtNjoJL`o*Ut_)RSdYpn5vAvp*m~qqne~=;#CM9=oUc9a
zJ#3!kbNvE9m{+f^o@ZKH75(HDVT&C7*(IM9<Muj>h{Hk{>0VxCX{W3w_|mHH^J8t5
z&`d4)aepYbQFuIuj(_w%*8{N&0-v$3J>0H>cfxzmcz)flL$Cs$k)CqvI)o^=Czc2}
zcJF+<cphu%S6cOMt?^Jkph2+o=<*x)%<2z;&|Fw>)Zm4L(8y0{FKC$NSl_@`jwSS8
z!u&?IQQ6<>e+|Z{WzAIyDvBzy^)gwzjq6mS4Ns??FeAw=xqr@ij`WCo7hTmlmS2Du
z!URN9R^d4baY6Ng4P1Gv)C+EDI~%%$#p8~*VT_6BQBDnV{Pk}?71WpZskI?mpu*#j
zzx;Ho-)=_C>c<k}xrSxWOFAu&k!+!^0jW!w6XT2k3_F#F-z$@Wr0sCG8<QY82~fb~
zCUm_i?NcCUg;~aGfvy#d?hCj*bcu)xUymI|Uq|@bp$Vhrs@j^0krgVPMfy5wq44m{
zsa%`r_`DNDgGCfEgJ>rrxR+pPjHJSwS7ScL^@-;NM)}?o19zV!MOvYGPUP*Ezj-Se
zU~^goWmDti0ZZ=m8sq?g!7|K#bi8?v0DZ$on0!OtePDrvMWp5wAHe8f0mH*IXAfX>
zNC-$x1lQpMAjrvKgSpWilkKGdKbh@;(;*<V31Kqd#`zEeV7%~?i+4xF`<FO+D=dF|
zv|&qbW3R%4gE5n?_wGSpKm?`u9)xQrU_^h#9)#?W?yC_i<9er&8r9&1fk~Ik`w!>&
z4AjbncpSHKisz9l>unF+s>T(x{K&kAmKDGU9=8~2`8KFb1!1a0t_IH9CSH5mmE-i|
zX%<A9xb^v0yY0yJiAN_!`R-)P;Lv}!yKwa(Cd~tR3(A8qikq1Wv;-*L`-+|E{FOQy
zNq~<1mHLJ2=WDLUHQFx10wR}dZdz?YLO@aR&=OqowutAUvg+2}^s7%>0a0-dS6!QX
zeFi&iA1-GtV-(;bmtld)=P{iMLLrW4qswJ|lael9+j>df{z1B=o}b+{zE^);x~Ggc
zrcsKEt|^H!)3>g_DxQ;2G{P+xwCgiFlsA+XRw_`jy_v?VI=&{x97nW^W|>3A=6Oa4
z4w=xQc)H}b!Oh+``KmA1QPmDNl{fSk+^haq=P;l2mb!x0Qi~tdN)bvs4Yv2S<U^K`
z*(**|(4b9sCA-+SQK@OOv9N!FavXiu<L2Uy<;*&9?W!lL4p)2j{Y3!EYGDu$@%#ls
zON$%ZD-Mqu0Tej>#7i(n@rgwMjPivd0M`3OMF2Xbq|B=?XBpQAFJ~FGKfMTm@q0oM
z0H^qP5df#?&K3c1FL-_tfL0)^e)Z{@aqUmf!*xf$Q=4TErCjuUg0_Dm8sOB=_}%TE
zMpk4TKF?cmMV>qE5sen!C{B%?!X_Ck?g4VL^BuV=?TnCWnq75zC^=YB0jFw6D>!$r
zxuw+Y;8shGTV=QaZa+65Ew!$^cl~>{{I#f8T7qBR!P6nK3b^cEy_bwkpS;9J;(x*l
zXAI6S@L8IB*LX=Bjr4z%emHN$2yRrK(E?}M9PiE@Fs>H@6#@RoxtG|$m_HV!0K)@F
zJ~icggUujSZ+6x-dZ6?|xhPz`vGk%{r2##fF!EG>(dXx>C9SL2{_BePyKA}&*@&)G
zVKW=i7g=bEQu)ISm@WiYv?xUs>S}rP3Igr8tl-XyvkMnJUeSN?k_%;n@US2v%Ex7S
z2247np@GWQ`w|2gN?hDvm9HV^XK8kpQW|KnsBThU{U(%8lLId<P1?q?L~WcsQmNd!
zj4kKU%a9OnzQBzn&vE9R{x5V?R|M2~S;N*hm!0&sxU%eI7&4ZfoXI>!VYHTREk~#n
zZkBSX*s@&!!L@&tCZH}?w=9s<T(&d?A75Rak*{C!=(x|{7;UGLveR6X#>CB4`Q^%&
ztw0@ttT5DH#O`yN{K=EKVx3a2E|M(c40$l|qu4|5Lzg*OL8i5Gru6Q8k*LY#>@#R!
z#!p8^T5vI2H)wT3E|X~6I(IP6%DIXggAVE>H(L9SjTV2`6}4*TJR?>4fisLrwXVqy
zx|(GSAhhyp$(wfc43xCY-(BgN&dr5U`6TGO#3@gWmh!1uz|{WB(^eD9h0>`jdaB=H
zmHy;d)bFBs=>UvCbHB?S3-7HUs*W8vHh1sRhh4&CUH0B(K9(=p>gHU&^0V*k+qSxK
z%0_319(xDxV+1*WXdXe_m9>)B&k}{^`HN0G6E0b_TLmpS*DKcGlKnlKc@FH;Koz=p
zk@x<Ox(Zjb%g3iyEa-^m^bpu2J#eYGwVuH`VeyPVGHN*KaZOTfjFt9`CU>B}Ep2eG
z714%So1v~GTliV}8Dn!`JEPz8N=&SD33iXX*wLr{{`!6e`#+kIS(ovLMH2!wIhRof
z7ZaC<e=F{P6wE;#c7Oi-$0v!3m}l*S6)?9p>km@8h+{LmXDf4`4*Iho-@fHfF4a(K
zVKy$*kA*g%4(UvOrW)l}J=Bj>TI-XHBHh>L+6$B)TK{gR^LOsW;#iz#Jv*vt6ZWC1
zJd$1dQ@K?e&4G{9k|g}5wzYnde5m?&s*Q@|wh@$nIji1VnaKhFjyT&^)5dB+_kwzZ
zS!$eRnVp}<4hqZKFSUwIUYO=y+6bt9T<J4=!pZ}@7BXt`fMOLID{4%8{hrxNZOVpP
zm)VNxWKXTErCv^3u+=&YBO{#$UQ$L7yKVEgD%Nv(y@Z|e4i%Z56bAxE@x>^0Sryer
za_qH#Gtuz*#9L`LO=?4G?$S#3Ord?~fscnAqx_v)CzKn8o;)*qTV73Bv+7grSIKRM
za8fAULPDE$>aLmHum-`>QwwP{WBe_ki$*FrXI!j>#ziU2xcIy6b7)~H|LySaI7m|4
z<0(te<2ZEHbZ?nUPR-<!WsYOGa+!mRW~;n^eqI^IMJ1wC%a<Qdzbz2aKQ1M#KS^Oy
zbjv@NTnvq%viAYLja4cwzeAOGfB*LN4_|&&0vh}yw~&h!+Nl(mANA2WXqL2DzT<Ng
zS_KyuC6YAWE<SsOf(q2I6?|EqtPVmM7g_1|T$+T95U*HcqlAe8Dz*+-5se{BY^BtH
z-*;X}qv~ChIkjQrrz{2A=0|8xt@303@Mo$Dt^T?^eO;bDHGL)n|KQsP>9^FWgjE5c
z15+s#nIKX~){{C?00%fy$|u?iweZnfP@Ik;NCkhj-r8iTgiV-=Ko&`$BKn@ok0@Z8
zYgI}0OD9}{c3uRZol>Dzt7iM(Uw&YJvv1JoBm1smp207Kb$)sFFh}Q$gBQRC&anm9
zusYSoIA!)b<611n)z}kFjdZnV0z-U_O+sM^pB6U6I@RYk*H0I#2VkWGJD%6i4AHt?
zTr$Eq-}G_>$kG&~6GnBtV4Ojq<V^;LlCpv_FPcTJoP}(>5J>3)Gz;u&xd%#rot$7m
z$w;+#jGe6)+v@uX0AdWqJ~V@lrvI%P7}Kd*pYl;{SL9a-d-@GZUr_^g%B?itP%|9n
zV^2_`0%=$!G3IcS!$vgD<cP3Q`b)Q@1y`{4L6+&c3TI<k&Ub@l2!569HRgKs74j^~
z^=X2?Lw3)oM7`g(1oXN|Ga_hz!@9X2%yB2yA#?eyb=VhGU5I~M#%h=`pIOSJhyv?7
z1#h+}pmSC@$fQJu&Jq^TO!ebZqRjpSvr8mohqdhr?_?dEqovMxp<+5l(2NP#e?cq}
zky9xhkdq<2;|R?f<}<BE!%!C2Bdrxz>~Cbn#!%ZcUu^S|`2AsPzK9!tEv>5TPuQ#N
z<Pt}E<=Q|w7ep<TKSeDpJ%hIEtdHU8HBxTlA{^SSK>idrF}Tx^jlnU{>SbPaEh-v)
zj(uK+RqSK&R<UtKB5G)bWVgI2q?2CK*7>pAtgRL?tjB-BN`fGH`%1#rgxYQ8aDxwx
z{<D7_{RewL??0#3EC>gGjHu_?cYWu4RCR0<;|e`LHU^27KrV<+eY0WHXOtE?v`ArJ
zSp?V<GeQmQ2Nv`&hSk{Yp_MRSFImg>thJU_YH3E<-^R@XtmTnS`)TvZpJGqeLhdl(
zG<(W4#o3gxzEXhn2>V~iq2M{XXd8#}!UG3Ojb{MofgwNzYS;>Y1`gFq2M*O6cfv#+
zup%16e-MX~0I78-$VgVf2^i}CsnnWEIiViJr69nFpv;9_ifR+epfy+W&ZXRmb-Br%
z+T^YzPxdbIWL*$4sSC-I^+`a`x(Whn%M#a%YM`YPpfkzKXv4R?io78xLkWXC1nJ|Y
zeOu4NDAJ|^!zKWKa4NY0ZWGprSwnJvrj-?5uk_mIdCRE)wwzLn<NNnv(9NY_t&fPO
z;$j_Uk1ZlaUe@M^d*r#-51drBq)Vy2Cr&d1(4A4ZaoE81miN@H*NEhdAR=xvB5X|h
zBHG0k%Td~8&(kiJYcDu?%kfxBf0W`xIfFSQn?TlpMCXlvP6zQfC+#G4rQ@5wFiNf*
zlo`M3dxp~c<`0U9o@T2}DYWm~CjHew_&wx-sh4K^Ix|k@8PG2fR#L4?Ahe0<u1Ey)
zdCgE*joTh5ci04v>WsF+1=_m9eh|LE70s;KMudsN$Z6i&Bz(4;1G~X!SJD&P5<#mB
zT8~E$Re+Iy2<VN+hcZTrO6J&MNI!@u8hs%^&yOa094XseZpsmgw%V>aglQ4%&^5OK
z|I_LS=%k<47*%|=dC1Mas!mc-(!2pw0vM2=>$j{S*{m%Qkm1BK&-!(z{d<-rY~bgw
zacDp5pOcriC5d&fpr_`7P1$3_$7q$o?N2*4N=jLOern`WaKot)P?X`&C_k^xjB?@X
z$f&d&PK=81;l!wP|37qK1QACoZzq<sS^D35U=$&mKZ4~L5A=@#jngcLYCrVEhz;S$
zi4jir-^zN7$<|62vYvv{ZQHQYdpfwu*{$8(QSB*?I8*V905u!p#tp>Is^W;TI;zca
zBn8ZWQP1Nrs^&Id@7qXrN9Tn$it?|+?UX?GVaDiNYw3h6LigGV%2cSgtkj44(P<0S
za72S!c&VZq9r&G?T^_DHoZc&x3F6yI^Le%f`5Cvf_I=+Zw9OU`hY??l!!eaVvhAMu
z23?hA*2h~b&p<!omqCUr6q!#FPT1VGBVOWv2ZRvrNATb?$*e~BXao=TkuuD^oM%#O
z0p7!;Sn|GrNf~K`mwo*(Yw4ug1}j+v4{Vc~*)+pNI`I|n*UXaI(~Yb&a0N?!=#3k5
zSbFU-Xi}Hl6H9tMDt=F_{=R3sBy5+!`Zl`p5S2<fPW7F`*+Hl6Dp%6e+F|1%mg4$<
zUB+JD9lGptB%Q)qyIXt%I)Jzm8*@4Ei(CB2@Z2$^F(M{2%R95{qfBNydT!NsKg!5G
zJ8F`{F~H<KN3WA%jOn{SM_ADf7tuoZo>Q3%dKN0K^m9DVoZc-1ZR@a-M$twjp94<*
z`Zzs71p0051Qw!<EHNnmP{FXyU@}5~?D_pK%h#t*vs8Xzx=dMm52%tGSIN>|`vd^f
zaB`eesC&V|+6-0bzY@}i5+2GSW_lH8i8UDyfi|as41i?0aZ#lXB_Q?PTaugOhKQ`R
zpL;oDD(d;DHhks;VU~*7|0*p6C{gS2$w>p(NQ@$ZQf9rn+a<;ztXbJ!e%QW${gHs4
z0b!VGHxGfKL7tcOKfFc5N`B!RX_?HdbR#_hJ2QbIzj<-p%lW!6mdv+<wzq|v=K~WB
z);`knM~Oen^T%F{WAO6rGH<tlTJ1^$_jvm#=@H&O>GA>IKF#)ww~w{&@%E|TGv3~x
z2-*~%*6r&1NuRbsgnsv_`}BQ()*Yoj!n&ixM_6~1c+R?G?R%^{^?RLl@kosBQC(#7
z?Non`RA^;{Om;XDZ4s1hg0>s){Dyl2dIPZH??UvOAEj=;Il|!?yPe5j`kCO}@ZP}5
zO$rY0w%=V?jj(RlvzJ>e-(nB%$D@ySGjr;LULxxE{_P-Rrbt(9*>Ajm62y)d`;q1;
zKx<Rpuv%OFHbCoi0{XC$b}OLuRv2w(JgZE|Te_zKEhfsc9wn~Oq2w0?y+=|ghRM9^
zpct<@zaxKXN}VOsC8;b?-v=@<bWS}eMV})94>(BPD;`_P=`Au1mI0gRi@xlds=Fa&
z0B1B&&61{+(Y0p5u~yxGvRi}21imVb?!M6l2#Odz!?7%PMI_fsSic&@*+}lkm--4>
zb-0bjUOeoP-l^fq@%Yj(EFKhMJ#%`qGyg8_lpW5mkfY0k+yL;oHXG-qm#u_q(^19^
zp5#nE+L{F2gFTSrjRKXuL!cMR&+1Y~kOyL^NiSzzdR(Wd+kA$97W&Jc`Bmdr%oy@|
zMbk7!smGRVu)+tm8pF$dW1uiKO10`FikoYblRaPs0ID<nCS5a8xO%OvB<BF>I<Ii^
z+clS`92bOlW+jnr#uYc>QSlDVQ0mkUmZf{7W=OI$!iMlOlJX>Z!q@jE=4gZpY3^l%
z{ja=zY+=F;>jqbU!qoR^P(a*eY9$YRC#5rjifnN&=Jlp=YNvAkorn4M&<zspk_c{P
zh#NLYbnqg8gh`T@o2V&i;Z%G8Xc`5b#!OSEM<A2Im*5bx{3HNn@{1C)9y)%~XOU7H
z;7urLJYwP4sUSDgN)F=~`l0MkK~XUu4yOB1)Fk6}6g3JrdH{+V1wH^pO#)A&s7Wt3
zp{V?t&x0Wp7kLIh4ZS}GJaLg{kke50utnp^lG8U3`hNlDCSE$1;Z-XXm0>6o0y8m}
z(T_AIm%uwK1ea%xE4zPz@f@E2{PxEaPm18oY5&##8GpULoqjuwYS9P`G)&)qo%{oo
zz@!zEw4_O#e*f$A{q$v|TIS+3gem_Yef#~#G_-t8hH;z>!?jkbGmIb($W~fACjFym
z7$m_IsRzv<*#2ZQ%E~fHEGNoJCF-O$W#exJ-7KH378ZF`hG~B#XB2g!#G?^aG*X~=
zfZtWk01w*NXc$VftC=O^e7Y(-S*D!Pa%nkeZW%z4L?+!JxgK7UWG2n&T^hX;B9?D1
zN$&Z{ef>C5bh@`7m37QNNeq}XXz$oc_&G4x>nz}Lk1tjD(vOKJU7l;{mkRef^KBrX
zMdH~8J?8P-(xiW0D)Zo<KIa*+aFd=o&rd_crRTbg(1|kfrDU4f%y{zNc`0MC_RC}2
zSx|TUTu4$VS}0od@|6)d%1I<C2uTH<9*bs!)S)H%w@{V>1!qVfD`{~AAD^{;Evstg
z35o;xHw>Pjwq3Dbt~)Mf3^N)FWnKe-7sZPm6dodNP>6rd(Lc-#`Bx)+e(uuvYy1Qi
zW{$n;pGr>~Be{A`b0o8ekv@vQB!%7>xyeD8LOVuo0L7Fx+o3yz<&{vEf{#*)p~5cI
zB_|A0BrAom_uPa%^zkkK7hIxiX~VRV6in?nli;0=OK%O#aWmuV`Sy9Wrj#c}N(e((
zy@8^k77Bl+14U5Fl4K1dGOWqOdsNXTEkw=3LcH9kTcT~<IK`T3vmyH4k(v9|+fxJn
z4)r3_24)m-SsG)R6KW>S9ZIy4fthLXJ&t01MxU8}aFm-UI>j8ynY6Eqx)GduDOj}h
z?fMa5m;m*GmKH)$CfL(Y?d_N|_5&;}Br+kz2U>qx5Q0W)4oXr&f)GrRL`}kqNh+np
z>+L=)HKhd@DX&F>6gK+Ka>hn<nWX5M5S%lukEQeEdrK4CA!}aw9RoSPj-uwp-E+H^
z=C1dzj-DG_ab<!+dgLai0kOdsyyiqPB1FtM&*M*bGxw4UCb=9K68z>(+1e&YdT#+S
z>fe8)eVMIDb7yZR{cdk$bj3B%BTWwfMc*7a*m2I0UmgJgiUb$#@JibE8Lnr1tVCE4
z9yku2t6O>|q6><TcR^t&3JEdJ{((LyEHF+RkWtOIJE3rvk&H7<<y1s~oHLLU6dt>l
zuk%!Zn)WW_*I8=WY%NvY!uf4t+)1nibjN=vsE^b(96RGq>LFimcXJ0!TRYA-M7?_e
z&Peg){?ts70-TvqRGW}#=#Bi2gb)|gU|3ji>Dv+0C`>3^`ZD3NPzjxkZt)IxRhn6W
zt-IzjX>c;{rZ36S^jIC)-0%&Q_g&Y+mAnmaGUymPpYL{ZftopQ$)2;d)UJT}z~6s&
zMt&r{zk|t~z(*#Ry<Bq5lroE4Uh-$DPI0gjgM@Ent-i~qo`RUGFQMcmic`=N+my?q
zu}gnF6bP|N-MUiE-^5F-uX_!L%Dl`^7C?5Ic`W#){vGe3-@b!qevXt)my2bR0ED)q
zBCCUV{|jL$tpecLxqQ1fAVM&IYUO_=ZFe+OG<x5+p`|)nTXkLq6qzV&3>^F86vBPI
z6LeTDy9{?snz|PY%OaN2JYLIn_qMJX%XisGN_<$Q60~GO=#i%nC{+qBnUZcr+Tz}8
zI1wlk8Gl>@1g2GuL_yDsBcxzvK0<Z{&bTbnf*F=ZDwd}5rz*!~VX-{tYIJ|9kZ%=F
zSC>IohVArFc-uc25F$S}>$=F=(}Dm&vq#_8UOz%$=1jb#Ab!mdt!P2V{1bdu<>GO9
zx)?$Z=vLR=N_WywwWrtZ5cH3E9d*38b|np?ZarfGbL5O=q3oWunFIE`pdQXo8fynF
z>(QI`Jpg^dqk4ee_ke!H!uo%KCPt~|#A?uSmAX9*8vOr;sE7HO!Q8w4x^X_Iyq=|M
z^t-1^K`>X9Q%f-Wy6i{Hb=fBGer1`>W1a1_vvm%)wQ3F|@m1YWw3o6J*1{qFj&Ax9
zInvN7lgT<yEvMo1wD&1JY-qd4vb@3uN~?NWJL0Ny6u(vJNiGS@{MUaTI^|4M@}b-1
z3@9Fp5?|kDC_)GffcD$3({B@L`STQ1O{hRRtEPWVp_z6z^pAjm(P)SSlS*6dal+}-
zKYo1w!?(ZCzGwRJPv4bKNZ0|Koqoli1ev+ev_<`yep|}qXw2mIFH+;A32Yea_poVT
zY4|psP#Bt>ut;tF>05stZO~ANW^g#*Gt{w6&@^QpJjK$Q=lj_<X0GZKL8iin|K+Kq
zvHmBG&!hZL|7m~ct~Baz)9L$k`jWJ&v<~}50$NgqO{7I!(kd_n;|noWv$=_Os<c^8
zLWF`pInzpOD{pBO1e?8PQ25Fhpx?u$*^11$rI9-~5v@!iH|2l7WBZC823e`49=tK-
zd})C>2nV^4v>P>CzkmCc#{HmCpPRpt&I1U^XzMmT2m{QZ@U+4iud<79Hg(Uv+raC?
zhD?Aw&XPk@ySfGfG~Jh}VJFJ<PzC#PSQ$VRjXa=?F=94KklJ5-DN0z`nr0c$($-fl
zW*@HBk=xd04M=}40>nrgUdB4a_DvaQm_Rd9!CYR=cxU^*jS#d~mor}Y)6OW#xPx(2
zi0VNr1!uUdE8S*}sJS+N=iJNnm{nun5sSut87IX)-6;Le4YfauovCHVB+?P^pPR&Q
zY7zYPCGp(c1+fh%SLu2UFUi?#&V=XzfW6Udh)Bks@fUwmY%xjNxYbQnd@@v1(99~#
z)Ds_bO~06+$}(G8X(6sSZqS;Lxn!51CiYvse8q0Q3Cw1$E%nmUmf^i?JIeM0cQ-6d
zfhvHqQ9dje^x92;Xl%X0kukoUHj>+6!YT%KCPv*st?b@-gVB;w1B-KHzZCFmXk7~U
zHF(E>|M!2X?Z&4EV+ob%Z8YZbmrck)B7i%H!?f{8k1~G3VTNFx5w1WLBN#WjfeEFd
zgJGq$gJE@W0aBTMZ7Cf551}xxO_$KE_9uV0!7n5jCDDvPYq|CR6Mhk#HlRKResO<O
zNgjn?sdq*mPidcx`EGJ|o$x3mt1K+7&?@XrkW7Cm^)5)}FHkFx2O$|)vZ<(wxwYVA
z#L*(7k&ks-8Cq+^&U91?pP_G8l;yXUA`8tnQ$)IHY~0$G){YyQbDXT%xopHE-AgQ#
zgzOwzx-1zf@bPs13;}9g%9u83)LCJ;?0il6vc2&ddM^|~m(T%d3VqtP_aC?H<<Wvx
znBIS&2ogFqO6bZ(?QBm|S~D~azakuMEb)}&b>%{N9CfSjIXqHebLz1v3z~`-V_~Gs
zuInx1THcLF5@B{pK9_j19~FK&i#oM(>Xn}4vi=}{qnkx*rWM{`z3pr09+2n-;17=-
zGVI<97-3$i8<lT|rZwxeZ`q@wwssX%wp)LZN-)2GSG0HCnXkuwv=1|*c4ZdJv+|TR
z>C15XaACZZ=jO%drGg|92Owsg>_f&T^shk1lw6h|<3q0FJpf^Dr)-;AH(qZ6podod
zhhA?f{hKHrTQ*CbhhB>Te>=sxQt)*V6vS28#5Q#<4i9`>&*ulB@!)GCLaTSZHqw91
zrOU6&_@+ZD;SYxwIc}y=>7|#*)0XTe@hlp?{e_osfOhp1WLu=Da$@<mR@Pp4F(xkY
z*~nDpj=YbDFR^b}b8V1_hYL#w-qtc9>(z^ElUmzrA=jiLxAabun1%b)Z{;*fl#uTe
zYW%fy8kL@yYS?1L<<mAwPx>*w5*dG#U(=bb$=63p;an{uv>fP;chNpykdF^0iFAB|
zPJ7*9DYp8<tWj)Wf^cf*>?8F!aJtcHdCBxfGhxb_zZG*N6&}I6oOksbJ}av6t)oj0
zT}h6w9bLvJM~^Nmo%-uKACERWmPr^eMB-kCLIqk}ugfKFdLm~UZfDNd9<oVOq}CBp
zmRV!5lP^GrfY&ZfL=IM}+D_c6MTgMHWW9!!3yMu$r=7?)#)kI%a`~o4DNCl1r0fv%
z`q$TX{d)!i8Zhv0hGu56m*G__6qjMIClr?|DL*KeC7LV0e*iIqdieX>Ur#&>;?t{k
zm!x2<V#sTef<!DDX;Fw>d;CQXv`+8wZyZ56q*DbN6hi3I!{o1_r$}wuL|WD~VYFsO
zjhgTeQr~ARReE|WWw4Dhl-_HBj$hAR|Hq6SF{|8HWqjHenl6u$d3P|!yYRrYFobs}
zWWah+YLZK)e=O}8Bqf5qL@I5jOD9{&x+qku$~R;<6o%5dubxOn6`oxlN0f0S8P~K@
zYRBv)Z0|vyMUt@2as!U`?nkvJQf*6A-$G2mHt#M?9e7Stfq4^VL-_ej-d^5Lzn`c*
zY75M3J$(Cl3LlgNCY?Zs!7TcVKmIX%KYgKCTfY-Ef6fgmsIzw9s7y5&DH+b<qPJi}
zVZQ5goX?41g<a}@f4mQ*<?H8>W+@m1JL<6t-g$uvtvLETO`-Mq`f-(F&QF+uZrJM(
zdWJ|TM$q$&6k!U1lD7TyIVezr+6ID|3UrUCdW1`i6*2VCSwj={aSa3RDilxCr%V#{
zx}W_XfAgX}CDwXiIUyryBx_^B1n0=IA01il^_XXLajbdPN8Qeiu!C9am=~H-Q2YjE
z)=37|Pv{M_gylhGT2*qYb%AobXcNLSZ{378HA|b2iU}zwbXejH7sbt#w>nSYCA38M
zovDs)TQrNFuM5&O)XSh-n9wtsMLcxX;Xj@+e|9ZjLjdFXoI91^nG`V5hRw^0i|ari
z3e7ua8BAVIBwc+6&O-bk2dFf5EnlZzNjsdxehg^je8>`)+}8=6)D;kH7wkXla*sR5
zaKtz_p&Sl*KEV}J8Yotf^ES*Kv)oMhJ)j4i-(glXdMKfKlwOekbC#f)#1OMOF&G+f
ze@ij1>m>NZh#T5jxyg6bwXBiFT<ozHW^2{2XirV3IUl3~Xr?g4Bt2JrMdldorf@l(
zlZ&)EoQulQ5mTrh)1T*rCni&}2Bmqeh_h)7>uX3_!+Jt_41AmDQJD%ETQE$88i9^s
zE;K#VB<ycOUM(s<GmaR;W=B(&nIz;qe}>hOk1`&_Bs9tKa(bu1FwnMJnX_$n*8XRt
zdCFU6&2pNjU6~QrP^kSi>7Maa(y9Miy0>AJi!(`bz@$wP8Ei`TKr*W=m0pokijE3}
zk)puEOOkEk(3Pa`3SN^B9Qhoz<jXuV-KdQ`X>!-~Dw*A5lX`Ze&!(QK|5LTKe~E25
z%^c5JLmEH%g3LutUyBchJP#v;N_+;$y*Z;)7Q>NA&WNd5O!JQ>jmpS&P3&@<aovB~
zm+_<^o(!=R@oUMXcN97ibN$c;-IEa%8yrZstirGWCK%_#=8^zVZgvPUCkIqf2NGgu
znUWLLlf3b|AWt)|<b>Oog>Wnde{XY|zFs~8EEAw_sA(ZIvtms?G@UxG>5^*QPSZj<
zrkq{X^!{Y2x)Ckfuyw<Nww64VB&EmPZN#X@{eGdR7oI1{IVJ(V2m&W^N(<VSOHwJm
zq8~-%)U0eg)WtHt%lV~oi?pik^dst*RBs4T%Aqt)XwR(n8{}4d7))f0f1^DDHSNN3
zJtepx<8uuq9s?b(W+{29sl;Ow7V8vpWE!gs`v5`U`iv13rZG%dSuG#&$d}!_V+7mM
z>XLU`>DYY75}jvRkLk#!q_)gxIpO5M>Qd%xkC88Irj@xzpm)X*WvwLK@Z)RcCQXa5
zb>!`Yo2PkOl$=RWx3a|^f7{}|a<)Y>_5oY$G2OxzFLJWW7JH0qwn(7&MLN|M2{$*e
z#Z8*7wn(@soE(TEiv5PI*YK&|upg_Aljvc~sFPTQJIpl3(o%_2?6$ZhW0B5a;{6d7
z5`y`1g@gcKSs{^DGY!A8LP985sgO{5Kej?bY10=gBm{b`LULgDe_0yEw7F3cphvc)
z2C#2Y$#sB!II`;P@m&Pr^MRT|a&H8ozL>-Iz&jQ_0V3?#5S1RRj^qTEa3U|}ufz@x
zSf`c@P5lFEBLwrGQ5)G|=_T<Y)Gh^hRxm9Dg;u<ycqb(}@u8F~n6!5cJ<}{DYZaoG
zx5>VxFqkoV7xpcGe;*L=^AMpCf%KTZ{dWk>*a?A6G%2Y7y(cns`t<Mj?|=FB8yeW~
z{-4l;59ngix4PkHdesWSEx6%Fem9t#YaNWzbF}0^JEpxyfYPxxd>c**!wrzyHLX8F
zt7pn7)Wz62ZnoM0SO#dC3Jn2cZ7t*dXd4Sv;}DUjN=A4Ke^_$g@QVFQukae4_Pu~&
zuYV4w@5AX!m8X~ZFV0hL=EMn1K{#&Ih_p0e0HU;2P{iRaPKbrpmeJZM(8rq)P>N7-
z1r-#MN2|=1mWJWdgt!WZ(o}wF`+@)nuSq017eJ+kEg4*_qg6r{YZC4F_qU(9-#hmD
z-25gycOWFnf5QxS5>cW;0<O5xzECjB;03w_YLB^)qDovlx~jadRYsnhWCSUa@MUO7
zOuXKW%zoJ{AoSuKqnUL9QU5UtMp)0Bta4c>-eu@;UeP+;2C%fyq>Q@HuDUCj)-O$&
zD6~w=M9S!<Ok^&UiAIcCTs~)sxurq_?CIZ9coP;Ue<ugZdF4;$r|^)^J0{9fnf7u?
zP(Tc=3H}4SkBcQl=NRzGNMYg9<Kdr_TNNGT;&pn9QWve{QvtA%1}}Qm6HV);>^7FZ
zyD#SQF@$1+$zW^!I9G|qSP_PBrpaEeM&lZx6-5SpzB)69$JsJBg~#cbrtrUp9xSi6
zcg=EJf1FWR#jKHY6?35k!|0$+x{G;Vt=6lpb^BNmSL?@}JIKaVa?9c=g-xdd1j^U>
z@r1(D14vPh1;ir=J|CaRNbjUBHXGG%@-Yj|$tmbXF-6E!5a;Vvkb=2ji?&^;=u6-H
zemnzR4}3rp>H7KRM9jwX%^A8ZMZU*N&$V<Kf7h<vNPo)0J&!ok>FHG8??n}vGo$dc
zt1c?>?KuIy{m}kgMPC5AN42q3)gI`#)Jnnihu%%fD`I}Zo2@HN&j7Vjtok^2R}4gV
zVE)>(E1Rz>Y&`7~PEYZsbn+`MbAYkTGy#-+Q2m0@Uqi54CLk<YY`u~&clE(8Tz#<X
ze^(z`jFMb>Fw;|DIUJn5?BEVLg)1&S>_5DM$t@;Q^z4cgp^|#^<O*PP%LUx{x&y(w
za@~PI@4xOqnK^#lfnZ#@?f_Duq{RnZcjz(S;JO1rxN_ZrKrdW(2(cJ&*|KldaS;?N
z>98LYLFTH*xTahXvDfWO6$JC;QUw9NS+Z0C92hwL3riJ*f|XJQrT1e?6_mEimpKUZ
zYN?`ZsP+?M{}!?5Oo&?*FnVMkdWnJ1eVaQBJ<^pc420SJD!`;uG)MMtw@Ke4{=c}t
zKz}h*`5y)lY4w+(Oe+)uH8z()uK_LrGndhY1t*u!kqHHt%z`Wmf3&yrvTsEYz=^9g
zK?@j?3nVr4fstujDNVX06+8dF&+PFeF%&IIrsVpA<>B4j?#%4WGcR|sL~_B!62;Z_
zzdGV#M0AwfmEf2dL?S0@w^DKnKHSPqix^Q{ijAla4;+9;9EF}6D~3WwVM$_RO92#q
zf+~(A*j;Q!QX5}(f23SNVR4GlHZQoAqOF$+g0(#|J1W`M!ygVI8Ksa`?AaV0b8agq
z#~N(6oE=ALe8q7}jjuY+)mACcv~o5oItj5J<WfOYcE%tO)b25;Y%<mCL;_^0A;2_b
zCMjm5CNaz@Fo;pOmdSR-N>bEztys7gImlQ<1CuK-0)%Y$e_&t2l2{iv+2pMhoQk$}
z1ynUlRK+<+yB3`$vuCwca&0oP(<*!b3d)6Ks{|}h*s+*3SlSl2&ZbT!uxV3=V1sDt
z)PO%x8iglPYIjYi(4H`zG#nIK83vc^kDYPl<Wke83c1uyyHa(DdzRpms4WcLaj?e6
z>{8~?FObYLe?h^sYB4R~5INZ16>{eQ-zyL?huR*EE|b~=uvQk+xWZalL%Rz;jb;FN
zMvkTN;cgD?nz#&a!4^=ogk}pVuJl&X;EI@j70VZi?QR5BO?ZVAvDpzf;Pkeh3aF+o
z^eg>gD4u`@*aN2zTiRU&Rt?4OLetuut7-z(_61Arf1L}O)^^|({*Tpd57-rZ+cks|
z_O@pcpEV{v`1<Sa`S0(>-Lui_arfP9x*SiJ3yV!gq#t+3<HhV^el}k8v2B<BIevLQ
z`gZo=h5bxq?#`!pevT(y%-JB~_?vHR<)<%V4;jbZPe1<~k)6EZLB=UgFD8@IEnL}K
z^y5r@e}(_v&PkknqH>MXL6ymDe)4W~Hg;kUI7ql$j^|Uy{m;LBSbl%994*I=`U^);
z9(x=~gCK3Y9Knb{q1yj|D-f&nKM=7LVe+PI_iR2pI~gxubkCk1b<f8imMe*Gv~Ak7
zZMuVPyC>7>tY0XtZ2h9atEti8bwxC>PcOx2e~-ED+u8i(c;0phr`_MXAG+`G1FD@u
zqO;`-(&_`8ky7yqamhhE3<4Z2o++OmbSD?TEN#JmoKOGLJ-J-|Y<WJLb|>Axj(@O!
zU%go_-!1;yb+6BtZ!UiEXS286j~A2k(bMx+uVC%@`_cTR(f-%TpP|DwgY7D5^QVe)
zf57=0(nDgwVNpt+6W?9Z_wV2PUrof9Cf02J`VV4uzXh1Wb9b>A&%4R_+jBS&mRk5X
z%eVJcq6QCi$l@8%&Ki9Hju!SsA3%ZmcrqR>#)qcIZCY5grK@Ozfo}-hv>A?EtB=SV
zw{sG-jaeQdaJR+kT1ekHaO-B^=KFwRe=4_t;)C1auBxxryRLfS9aJwRtz%ZnSD4kC
z#kJyWMaV$tK*WJa12G_79*{2A6;Z+skhc~@HwPs3Kl*@V_Q<TN;wvgL4&)<|6;JTh
zblN>}SpB5?e)fFUJ#}B5y&1j3v3RP8oIi&DUANX^SRjI}QCc#>#L8HKD3Qv>f7Wmv
zMXkg!g@CFqp!N!%E(Bl*XnIYG3ZZ1X6p)Z9Wo#}SP;ms|{aJ$4D@sk2fnKc6X2?)=
zLk6a52FRlLON@#??ZFTco1;V?gg}Ai#ftJRLXf1OnS>E634RW^$a}K{SM|aAkiinj
z$|-Nkz=FcBV(Xe?Opt;0XNW<De@jDbk%0oC8&nuUIpxF>;~wnLhb43IIAH7Qyuk)b
z_HolG7vq*9JqN#+y`eZNfAl3NEU-D!bB@+psxR@;0y74bxz-c`70mk8i)17YR1R`C
z87v1Cxh1t<W4Nlxbq6j+jYM%@>1?e8e~{c(%SLXi=^nYwI{VAqR)*YGf5SjrUg3RY
zsoo_M(7M{Zqhya>9^G;EsDmE~Ar%P<2^zqMgLjM$0nN1J;PnQ4RD&ge1Gt4!jS_V~
zp_vl!5uunM0~O!{`gkttIX@CWDgZK|z8!>sANqh0-73V#5YmFG=3ScRHLk4zdIOgb
zD)juN^vgAD#HS6S?h&7Ke=R=iP-)ho(yXgW)5c;ef(JrAJz_0b0x}mmRd16s3DIY@
zBSQmN>mwx*%iY0nVvqv=N7k)UJP2-}O7b>}8-##rKI(QM9wcZ71hrsI@KT$5q{I$2
zE`oNm&?iJ|%{em|IHd%q*vomq6v1Z0><OVHz;A3}6Jq;+QwrD$e@!Q8jwmk`<(Tfl
z7J`hO^l}OTIP!$(wJAj54H-l#@IABG?r3g>({=Q+LG(u}i|Y&nBxyh?74%CP1ytJT
zhi#gDA56OHhi!nq!kf6q3IRbzrC$1XD)2a^NAhoUh8~UKr4&r$YpUC&$PB!X0SJ$J
z7Y!9xT5~Rv9a7^uf7i;x04_j}9XQd#!SBBqSLD3miX89bib7YeNUM>Gy?a>^LssM=
zEAk$)B7n@Rg+Ke+4_j2pDZ1QvlXvXaTJA7)MR3?l87^p|klm?54AnO5Kt*u($?MD6
z*`go4PDaz$ZS*>xcE3^gb~L>hO}ba3|8{@!-XAXBP0p8#f5X{yd^noFzIcnHI6R*&
zXNO??{m|yNV>_P9n+x1Me04Ex8y}vHCX>Y>-85>xYdHc<;SSEG{6Gnb=%WGF$bGDj
zRg0i)K%@^MC<}Z*x5@A@m58E3x0i~&2+<~PhA3Oi@R2&iu!PesiSHW2RmE-tjJyM<
z&7ExKc{?Wwf4rGLUxD8pA%;m;k{#L0B;|Y`Y?DoFlY3#C@)fr6kY;57IT;{LWFYj@
zu}y7oh!Y03C3}r*tt$K*!lJ;GZ8NLCcur++^D@*Ds3WZ}CSTe+?`ke`*Ufi(ZHPWx
zIeN+cTSEY~eTRU2U;dBY8X~{=Xdk1jB#n?XP`baQf1mNKp-Au{UA{F$VGa+Y&;Q=g
zKCIZGN|R0Kia=tLxC5t^3{v!qnPTYX6hrSd#faCSOb5Ku0n>C3twRg}MZ0~Wl+i*Y
z`_cz}p?$)V49PZAbIO5Z6s~|&bPp8v(m4PoQ5C>_QngRE*c+Zu@*?GNM39U<Gdv@r
zIW3SSTH4D8dk+9$&-$^q4t1?C(Azp<y<wEc<n7C#+c{}9J9vOdSWD__k?;u(OH3P)
z5Ow!RxUxlzyP(V(*T=psx`{K|`z8o-J14vk8Fm*=Yc}dt$@f1n;f}|b;Z-XXm(mA9
z6PKW_EA4+mTIJK<zy9?>A|ZZxbNw0>e6Rv=B`TyfI)G#W8wb!V%;F+^118Ru-}&$K
ztcUg5v=$r7C<j<rUyanI&H22hEsl8SgtsF{;*)Wo5K5w=Nprb(uTbt<A%J~Lz3nG_
zdTYk#U!`CA46~@TuDCe^;@_(|{B;b%M9uK;H86i8*aUf%dPZ5HfrE^CM5nysK&#BF
zN-$pwZ8L_Mu87UrI*so|i2a0537(^wkPCLCLXs)s2yw{WR&lZ>zzm!s1&VKQAud6S
zGB;gjXmKu*@uYnE`x&Vsk)Nu-DZDo=)U?8hj9Mdn?}rt=A%~w4o_q2Mijpn9g}USX
zX&is93-^420-Q!&li}TPY?8Md^~==rrC-4~!+I<8+dvR<KZPqX<BS<v(iJa7Pl)J~
zsz_g-*{|2H4}VMuV09EW`0m@!hxEW2B}}h7Ee%Lv-~IWI?)$?R4z||HS+=AH);2mo
zNe~0ud98(nZ=!E3+7RVM3Jcwb|Ni{Zv6g>dKaUI6z$w*ph)U6?(UZd{;{hMVM}ek&
z_x#q1K5{@>&QI$MsDR00<pH!EY#nDNXxnDqcR}=CD6c0F7cc>X;=_c14wz_O-;Ygi
zgbHy3o3VEg)=Y&+9%Ep1BH9ThtvXCe*sMauFu}&1fCwsC`ib1U9*3xc(8^6Ij?RCN
zIS$z*Lp6~j`7+B(ZM6U|Yq}#2%aR6T;-uZ6{@6Q;;Nn!4-ajvxP5A8GKRS33Lbz!E
z=-vtV?A$-Pk77m#A3!-8@D%KoTPOActiev^8L5Sl<^nBlq!-SG3o$Xt(kZ>E1Y{cO
z(ti>mHZ8v<-#*G!TM7}5jIOq&8y<h^7&R<dX&VLDH3m6*s${8?hm*W1FV&}Gc02SU
zYR6Z<Ww#?lbA8W8A;(9b0Gt?bHGNXVX7u8oj~OTO6<TOZ#>q;EH{E1zdBC$kF0_!C
zYjMbFV3>r{N`G(%TF47Ze=zrL@1xEG$;3YXkhwq2pTW#CNM{@eE3BJJnm>PG=5_MT
zIl$v0q_q;(#}5!82rGo`!%h*>){8jB?5xkQON0iW;k*cu&#(_?@)@qdnS2I@vs?8U
z&ZA5go}ql)l^tyNDkAT*-7Xo$ooq1>fx6%>Vv8~0FZ6pw(Ines`V07CQXFf;g{s1&
zVAuJb0!B95fuN}nF0$DS+R}gAXtVj;$89!)bb-yDU~-wwW{@{*Ht&5qVYiDV-p3=d
zw&!_7!8D(DAyC6Wne{%ZbhQ&fPvm3C(C8>)=&!&TgK`c@oi^7yxrbpsmmZ4LPDDYZ
zHz1zSsvr2{@T4|iG~_1H&z@O%tmQmN1DJlIAH`YWJY)wNOR5iu;#z;m)qMlrK;6zp
za-<nr3^pS<-2LsSg-Fg>CydpDKE0i@3N5saNRIUeS{=3u`{u<%V%GbJHS$VC879;_
zVT}T`C_lyI8jFOH9H;opx<zPvAJxoc6?X9m<Ery48q5oWq)ta@9FY^J&t!%eM0wtL
zFo+jKkr>JC6d`W~%yxg32n8*i)*F=P(cA?xgpnM_`NZu29_(((kalF$971ebew}ge
z!4WaN8P@2N;n9OiFRElTM@gvwMX1m4SGw-3qy<}n%lMb8pr!Fj%XH(f-f-vuC}m`1
zn;M3p3h!9+J?r?L@<DJJ5pQhsjqG4Kt9$CwZ+dI|z8hm)59WU^WN7zgdrAa`Xw+(Z
zm8EQ<ZN*P*?=zN?&AVYK+umm^C8OR2mXcv2&&O*FVz|!kGX8DY-M05xySt0yWOrv7
zKG_P;*?M=?R%oGZZH2b?nXRyIUfK$6?=xG0QSXASz%e1g#X!MacQr;5AQg1?9~=}y
zBqf{;8$sb#N{oN?z3@6ViCwxfL2rj8mqA|2O(}^ith1RcqeP87P9r4JX;t9#PR6Yy
zG*QA{d6o~4<|cSMe4W@XK&_u;O;?+Zt*?0K#`c_2^?GP^bo|0rx8N9}bKG_=K9Qzm
zDRMi;OIdo}2KaQq0)WyfY^MgPZd&G~+Jb+o#`HfkpMZavC{)tY7Nn|ZF$Qgg^TR>J
zsmnk%-<iB9JgDFhUry%-afnL?pXE}J^=`7A4sh2v5uB;HOX<y-CyA>s$QRF2a2Gg@
zc7s9Q=5$UR#WcoI%6IR^+v#|Xt$4}xz<7=I&U`p4wR9=QSF)b=GPja@B^7hInXg|b
zsh$P|EckyurF#C+LC4Qs?hOFIp>W-QyVMMWptH|nWvo&$A@uOezkhuH%eUV^$hsf@
zNh4$jiq0u;M1Gc^N*duKaIE{C8oZFE?|qlV#dw?75v_(UG<IzH);(CT<;u9eZvH(r
zN6?ZM*jJ6uS*js-rCI7WT1tXCG}ENN*NwEcah89K+pv`MD?!1r=_ib*QR!#Kwfaq6
zVbnjnhwt6Pmw~R)z`wj_!Bv?ds{-#S7*&&&njrFry%8KyBnH?Z3#~PxL!(Y<&PluJ
z<c2GoGzQK#(j8kWgsBN&5d><YZ{EH@0ETNJ3A(hxMb$+s95^u`oeIfXNIU-h?I(`=
z0i%CDSKkHBV@eUW&UcR{MiAbZVC5X;g_;TMPQA$!dd6IOOOE5Hw(K){LuKZ<&BlPj
z625fR5(}=6BeGw16G(xN3!L005Z5CWVH9|Z+6;tQDBvI*KoiP166XG-%y3eXvh;OC
z2R75{o-*{+kk^l&Av-@km6JSwg#p`gHHLqz=a{QGSg6AADX~3pB=mZvLRS9Jn~fT>
z^z&*4X}FIDGrLG{*!EIo?`I6%bJhDcVD&Oo8EExsG=sT%!I)3gsZki_cB8S*PkS<*
ztJHuX8VrwmCpEI3^k$tJnQM16{;~?Z+*t2WFv_(!_ey@s_~smr1^6h4-ioBM$bo-v
zPRgDhbs)KN8h-iMR8r9M!nL|LNg=kV5pq7gn{cPV9&0<6AYD;Kf-_;&Q=Z?Iim^;*
zU(PA5%#~<UT;}&nO{qK9k&I(ZyK_l`=ctF3JF61sywG~WyDf?HMo4I4qsXBwE2ceB
zq;TZPEO%$+$!VebzNteZsnsp@iUfbBx;#tkxPx@5RyOL|qo!i*Uac%c7<Nk?BFn<z
zv`ci1)GNtbbQL%2Dz6G!vyS0OEfWA`T1{18Q#Dmstfs0VR%>GL*R?p#wSL!4kO?P)
zA6hr@dtCQ=ecQloD<LR1A6Va(g3$nGmCYn?r1g`OI%X@i_k%vCUxn|j9;JV}G4wyE
zr>=QGI>XnT*--(gqUgcJXX~D)0DNBYPK1I-?4x2io%V9QvN1}a(u5!KPP)va6g9dt
zR>w$LmwRSK5bG&r7rc;)exnVU)e_(;>o8`T)N~E&l4rHhb^>8T_oXA0UWLiL>-%AK
z9laMlxZ*PqYd>Yk4!N)UMp}Q1-e?Z3@=;d(G2prdhj?x+eWeRjrR~$2?E0E|En+kr
zYLX}CLa90qy7t-ZSYer@D0?QkN8^`fd%a{ujn}mo<Y{zm4aS;f^xn!CTaOv=R3jbA
zdQ4>la4pU1^ry-eI@GMX^$`ZX_+>kY;Q;aCxxMei3RTolSjkEfNL7E_r4=<e)~asT
zkJojvwO?GuQ>83c%_hIe5)8>`R>3X2<rTAe#VBFyPIWGB8zZE-xNVHVO5FCZuI+|L
zEsxb%3q_zkKnUPW&s56m5b|Lr>EFKE7p}s;SLXS(@|eRrD)zDGgyN+&wnu;ickL1-
zcfJE4rO@lB=P?_-V#0s!rd#ZE?y(0u?d?LHH9ncCl*cEtxYtl=*W#}UHJW~_zElk-
zdk5pbB^W*KV6rrIy>~xcN-TQa(t+1Kq5D{Ooes3i$1LEzVFheD{+&#aUtLiXVRT1}
z`l`||knAgFtxCn1E85pK2vZeG-A8pWjJ5@W9Mo79D5HYVFI9iom7Udfkq%b$((>8m
z^TsGIBJ6rLs;Iy~bf^mYGb=*t@}aV%&_Rvj>;|Pt?l~$yzhsUEwQl#R>t$6<tEZ!g
z_9Y1Y&Q~=Ip{zD@F9|7Y-QE=S>l9{Xp*B_c3XXTbEJ1_6Y<16HmH;Lr0&CB{D<S8M
z%PZfNz#x{L&#r&7e59LiP#fh+scu=jd^Zq9gFnw^nb#}(uE?<lf<Zh|?2j)wV373s
zOAc7?I&v7VpHNvGuSei~R%;nByz}`)ddeG}5w~}+iY~7|V0~EeH-S8p!?8{L#r0!t
zAKxO!?~f_Rbh7CdGEz%LAjk5pI^cR9-F_0k&3nHshb(^|RY=_)thJ8U(O1bRUy?~b
zMmwhDA{j1JjxE5Q?~!Q-9e<5X3uogkGHvIJwGzE#h0%AtET470VAEr&#FWeL(MgzN
zfBovPnu`OzRwJQCWpa5!Az>{R1Grvi%~1Ssbp-oL`Z4mY&pZ822X*xX%)^3$=ksj5
zgH3^QRxy80M)~Sqa_(%TZ*NZObBNI;=HW+vkDwV~Ow$2tXS$_H^|mlNGtKVwK?M#5
z%;<^!1_}ls&uRUwEa(~k0cnp6@_O1M>%5WncqHR0rJvZUB=}rD9fS7a`E+c)3-akW
z#Pj)d9AdwkPse&6s@=-2IA^Rnl`K8j6o!mhtq_0a;xM0`8?|t%ztC4|0q*pb+Cj&C
zr54VHuhe$_AYX~mcimTFov-tird&SYD_y@jlxq(7LwqGBm*;&Y25{Y1VkqwSmH2+`
z)2%}MKW5Xn3)sk}Z#!d7`q7s{u$lYSN~E@Pn<=kehhHq6gRYKJui@nHJ(Z^Aq@bOA
zqWL{f>=-->U{ZjWj`@VT-?yi+K9rH^xGMv(fd7EZ@fz&WwDA*0jcjSXj17}U>Y-Wr
zR_w#S!RP)Tz_g`Em*G__6aqImmr(O26qne&D_8>6%$G90D>Z+=3;r1?wZ=b>-vfTy
z)1!ufGr-~>J^?sAdILD30smC_R3Gr~n!cYN)3dZT)5`08nk~z0MU>SCnt3G(=Q>kf
z*wZxgoM@f$Gv!X18;z^v_4KHsG%<09^8hJ>JN^ChUmqk1;^p(hd~m0&2+?g{zCHZ5
z;lz(yp8X(&j=_I!|J(Ag0FD1`Qg~-Pk`UUu;I`j!;!pqi<?CO*`~w6o`)`^mHzSmV
z7`AWpQ={CK_1mxX+X*~)L>@$x+L_G+0n*@vy08s=*&dAYxVsH(^w%^Rg$YI}wpfGE
z!KaCJ12$B7NEG`h?e{AmY+SW1GSiKX^ea)Jwdp6Wud9F3&-BNC*QvPbU)#gi?cozi
z)jH=qP6<J2i=WO1V^&fHVZft#GRf8<<pi1om@1X%rL_m6{JedtKvf!zCk(<%z;A80
zcIG<5&RvI)R;5sf8f(+|fgGl@;z%=6D=ex}8{ttdXre&UEfVhEzkH*0zu>Bm?6c5$
z1f_+sZhM#GzboAZe;BFoLzl(CE69I+K6jVrmO)evN%Y)3A<rWENy>$OO7q5D-O5@*
zp^=ikvaV>RNPES*Ti7R^=vWqkb~D}ohBJ~C5uN>hH`p!h6v?mP7`AukT<Y0L%x9Wc
z8&~B`zKm9`c(~Av!x?{|S#@9<i<1murv_QyhSK*&TNjy@fYnijg$fE*%_4u2tgizH
zSEi>LT&^{og+cmdW(lBS_Al6>WjisaR=t2YxR$b`_#zLW6$|qj>3i~2q7s|*7pHHS
zq&wQcGX-RfNGSs2$H(axM#A6rYF8kF%;7FQC|SJPFmVmhZomI?`}*){rpl|r02nHy
zbC?J7PyrbCF^F0nh18LUihO@8UtQz$=p<0RnCrxlM$}VP1sS5NqgA}+SFZImRgJmI
ztWPo+pkhE3_0j5Aou%^frt>z}DlmezV)8_l@ZLe%2hJ4PQw3>Kv>#orN(T!D8Q_nS
zPOYQ0zWw~_uJhvxmr;SA-O|}pc5~?!^wNahN;OQxj13wnVS1GF=IXZ$!z;!Fe=%K>
zyeb6~ET-ZxQ^^-RW98HgsMK(l-^K;EcdjG8o5`vulxCQ-K(sgBn9xCZWkyPvnJA_?
zUnt9U$3g=L5BexCrc&b%x*K#xAky!y2DK9aZ`=){<>uF^(mz%f&>BX5ru)MvU%Nk4
zAqww%yY89Y){uiKm~iOeM$%xhf2r+Eu^D<w<<$!I$<*rmPA2hR|H;Q1M)>Ul1{Tbj
z=N;j>D`a;!U_~@~UeRIZZXT3OuOzx153t-{fl6AOx4#?Jnl<0Ws&}}npLbVs&1yo0
zuvh2(YDyILmUbkesD+PazwHth(9qK>zxAKe0O2FtTB^l+jMflux6gV<f1Iqe-J-!*
z9vXMw4fxy^E^EQG+vkG@Z&mvpAn+yvSSKQ@1Kt4vMv4awi@QACAtGWrJW)O{UBL3o
zd6;MGRT7_F2_^qRJiUWFwGpwuymx~<(_7VBdVxK?CQG{xBR}6D^Kn<D-^$1iBo{<f
zJ8Bu94@w}iXAVi<;geYIfADtq<^8Q(Dy(+>pRJW~*B{l&NbLr#Tz^O_5HTS%9fRK=
zd%&f%?N3(9xM|W!Evwh<gW;NJomGHv>3=~O2*0f!&E9b<qwU3`I(qrxXg@La@@lXj
zC+|497O-sGyn_K4FEr?P3_cUZed5qAuI~y!9!J2<)DHttaoCxce`nTvR@4u_d`HS@
zgt0JF2fJ1+C8*QvAAtodPqa0|rHN$q)~?&y>^+Wd)!w^*Np%;ykv6<`l`@HGO6c$7
zJ2e2eKuEtj;|9tK9sB<vP%nD)k5dNJh<0yM{&9baN?3T$YHcK#>sxi)o4d<nN)k41
zaO5yS3-EE~=El--aVvk9KQ0+1yqP&Ys!t5&lvu(|ia082FB4+k?`^3)%idx$lSwUG
z1Jmcr=ZF6T9m>cRm*G__6ahJx0sS@;0y#97kvSJ9e_Bg#<Tebx=U4a-L=`Dg4<HyI
zGf4{c)<tebPepf6d+7fELn&FZE!obzltp{+JY<Qa$WMGo(!;kw55GPf-yeQ?Gc-8m
z85|DpUk8vET0TKbqvgQ?C~IJN|33Wjq%~T3j`EP!#>nI3x0i1|JR0!Q16Uq;18{hB
z1kz$ae;ypa@lBpn{C#+g@8Z}*D~)%tw;p?Ari^d7=P6ToE<N>e@iz8+j<gQx9oj>Z
z#>gmrIy_>~6ryy}4_tf>f4=|rptB;h3O4?XU+urw_lJKTTD2fM#UTuTeSe5A$bd2)
z2SQW?cKGr4@b%&Mmh8k#%b4+l_3-1{(6oHXf0hYaETgSut}Fvu%teE>L7Dh!v<$sy
z%`!U@4U^28M5_qZEPr|Put|}$*qB?29I4<)jYab$dL!Y$$C0X{C~x^l)2Lcx8WRUc
z$~tH{pswVbdV8b5g_S%!&WfM+<jR=Hnn*}8SIdfoN`TvP2iTMq$*LgcZLP5*8z(#_
ze@u>Yu^Hx%Bv;k4)M2qZdu4=Uk&7E7p9j}jWErc-`JTwCDRLO8M2d-W)R1@<^Ezs3
z&$^_qx2E>WY1>NBAnz)Q;1t4YPRZ{<^nDHG8!U{h7{Xb~9W1SlvKo6@x~BM{rB9z@
z=~{vwmaZv+?Ph5Xs4SVYl<&7RJFOh{e~cC_MR$EvprC+Ovh76$3At9(*z_)0Ma)KG
z7O+BGj2MC}#xa5<FRaJ^1#z+O9UhrEC}-_bqQnhm{ifeVn!KLROM7^<I%YR1;+@7f
zbd0U|l>3(Pr_3w<=KR~2@e~%^-N-3L)4Uqt<qjlE+_Pd8^2%#hbf8~~l<+cUe`Y0m
zle?dnktB@#4zHKY$KKBet@1vc<0BR?`t86ISOv4g@;IA}JFwEEf`dzVb)6ywQW|iF
zZw>%#M#wCo6Wi$6-8GU!r9)5G*if8Gn$z_zl8q#LzNNU6j20K!&Zsz`;zCZd(@AYF
zqiv&Wj4n~X6~|#-uMfirifmX(fBdqYK1-_OaAkwqVZ?M`;h;NMoV64fuBXLoiajh|
zQ#@z!W=2hDiD^~}cs$N_T*)&UIP4){FM+!wT#^ZrR?Uesm82(^=O&lvvHdy!w)y**
zAHOK&Q(J$pAjzNUiI63|sKK%-^$q28vHFtcPMI*5;5uzhRAclF_;g+ce{ANXUv~<5
zr^ry(xuZlaVrb@WKb(6><hSR31A?_xhTuPrVB<wo*b~DwMUOEahnqcIBi~-yZ$Sk~
zESugXX2mFxl4X0_O3Dars5hG`Q||!E`jzB&Ajvo*(UPk*tmmR6%`pFu>xJDEnM3Wx
zOL@amSYyck8t0TB!icq~f2O<AeO92uU@Qs>=S?Wnp(M_f85EN{OS!}Pz@RLKp6jEg
z*dYsOD4xjz+8N!L1*A=xx@3c#M$SOoNEWg~C@pwN`q06Q+9?On(~K3xX;$ZVC%1HQ
z@lMzE&I5y0j^{$^I69AM$&?Tj@^^6q#~hBrA_P4z`fxfQi6);7e_0VeWg@4=!+G#2
z6Arn5iG^-vR*c_qEu$ml#gF#`K}*rz7iaPcZKz(Dvo)jA0m5XNJ0+8m3W8fnb}pmB
zc&N1!>~<_$qrBt)KbE&n{LUS+BF3%jxq4h|9&$Xqgyc%kZY!3a6Cxn*-?)V@Kg|xD
zZ5fd^MbqZujHS4Je~182iz>Wqm%2%&dSh}lKC<^1aVC#91puqoFVh-a=>@%H7wB8P
zM3-8&doP<~v*Nd{V92Q8q2*Rq85&bITcOMyHtjJUMXzm=TC2D__S}lSqfHWCkGa>T
z_TlriPk|-NbsuT}d=_z$=9%MhxmjyIjb5!1sKTA_NJsrCe~6w63d-ex5d-B?Lcw;_
zd~UG|44LW;rm7~>S&Zz4Zp})^SWwUBx`T|1^ovqEITu<^Ig4!d3n3^SrbNp0pE)3j
zW26<Q{xb)u(7@Z46J;u8rDYqJmFx-z=mRNcq)cfXFOW~$Yr4f&8o%Sk=e~=7G)qAX
zUpMXNPCTyhfBu`CE~pUFq0tVCJgE6Z^OHaX#x#ufE~%Wu2?Zg74Y8@@UWCNi?Utj6
z<TgrzSzP+;oD<ZhkW8Ny#wWUbK*-Q7=Y6papXREKs<6Io=q-Blnc-nxFjJr^A!tq~
zYGV{PUFpv*G}A3`am*p9h2T_$j~U;^$FnQ-r!;bFe<|T@;FHxP?hqPz_tupaq|OO4
zhJ)d1?YcH5Qa<Q5^IlTf`MLJGn<ZtC+S_iyC`;j|Ty_I;%0SP(%!cCnO2G9sp0x@r
zwv_kwHS?9YrnlL;ns6MmQF!gPX#&W~Ty^FtT602PB!$Q*ajjauT%hxsYs@I8n8}NC
z$@v1qf41hv6NrY|?!}JzEb9~WsEw>j_p|j_bG@@Jns>KZ1@S`-_49gddB);<u?1M2
z_Y^B!zfa+VS}Pph?z-V+G%BN4QwG{pkkM8#yOo0aLP~-GbNI=Ks+o2MR89N_JaL7m
z8}_)YSMs!iyxx8@&`Cih2&f5yzn!3GB~NPyAbsEc#o(3Xf~}OayQ^0K-sowG@68zB
zd`k-PFOGw^wU^Q7EE54Rml0qq6PFOOGAWm@(ksD#p1zNNd;ZHo7C}f69sQ)Q@#*pT
z@atg|i}h9*>6g#n5A?!l5IQJ)ZG@9y`RyOem&3;qZCy+vO#aWx<+rcPF!HHrCQ8(d
zbk5pNGayl2W^pbGO|P<M%zG&<a%WJ&2DvsURRtZF@16ooF&c?Fwj)N1a<YiYqDl;6
zwx|_<6SLfn%g+`yV(~YkAUXFkwR&0qOKf%{$mWJ88-zw-6$(h-ybAM`euvDTwb>$D
zN)pvJzvuZhh89MKl0cCO`-np>Gm@r0ttiv@sXQ0vojl$-3B~LxL=m)g^j6&$+(eG$
zk^Kbr)R8$Su@{DIVc?vJrhSq$;tSYQ0gL5-w<@HOwa0Ar2qxuI(o&VYLW9Pw?~E5l
z`zm?4LX#+HN0KP$Sv5pw!ZOi`A(zJhSbK@+mlQ}-_sS?Qo#nu*pAt_+IoGLlW3%s`
z3@lbe?``P!9N@NrPD+3L(=P;yAZ8At1k&7t{Na~=hrI3xUiAXX3JdX;USPn;Dr@C`
z{fY;8%gzTQth>GWt@d&=^CMYQU&7*;I=~-SuZ>DLqi?9*5#wI<+8<J{4VXp*^M9;g
zJJIhfy1$lCS!2Z{!X;)gPf_xI6-yQokG7$u%u_4BATw{tEI1}qSo>-urJ0H8owA~L
z<ad>(yh;_rOxc^RtGZ&YcEH1UUFFSx6wK!e%PKD!UcwVvn4<hqW~&r!k;X%QlOmyP
zBnzE{I@#6H2gcg2Rb?cL*@m({OZB0Tv8;M+aPO}fTgbU8nOMSNk4mxGxs{@B8zKz`
zI3G?F1#{on(5WU1_(m5Fiq)jmOsl&P_gYcLwa1DQJ^QVQaToY(gCzf;?$<bfp7dZ!
zK3ZW~th`dQOR{w3h6|ta5?+_BA@aJJnmT*073JBz6-})`QDRV&?37e<Nn_tv^(8@2
z`b6UZb5vMvs3vJ+11Lh$t`z{AM@|~)ZD`lePo~*uHjBG4>N3l;p833P)w%`Zx>Xy9
zb8q781d@8sOl?wM<({aoZwzC98=Xy|eUlJZ$a8TaRVJNz#wO1`ufJ`TQR{fi$Sg)R
zj%_QmWWZ<ELe)*<v@UWn)4ahbjyt4};_MnoQU=t0zx8Y*a~s;tmhHhnb$Df?k&7_e
z4iYG9p!0IYd8PTq6(+Y37}m^TIg-1Hvnx`|^j~9X88_Q=!@pkb&747hpJ>K9Kdz|)
zXH*ofm=AWkUXmPCJSKT;W#RY9qSF&=r{kzvP`25%n3>x_kV32K0W9ZL>;+`eHSQwi
zWs@eE$780b%_fU?U=L9sLQ7S%hHt|Y%}zWufh^%^WhvXx(<d+t&%P_KU9_1k%{O7#
z#py0wExJ-1I$q8lt!>JGkVe#LYmwQ>SD1$TYL%yopA)tV_-&Ktwz4E&Rb~{=l)PdX
zrMxZQ*$!L~+E;&2)0WaKNFE1#V^73UZrCLa1%4VM0DHs~G4hPC-L2)#%5Sa{lnv;k
zq%iminrns4=%-q>Y_evhweUub{BbLhT9DNGTa(2?J!>fM)4o1`n&;;#B1JBYE0xIm
z?g^{=C=BL*ZkE*>9YhSZs#=?%_@SMQin}i@9wLoIW<pb9<ZW{)_p7tn;A_sAh<kkA
z*n@UX$f&44`Rjtbe^`<LP$xPDyZm>dVIT!K=RAIEbV-r)b9ndfuV4QB`R^bQ;6DT*
z7tAs&#IStNkJgHRAbp6-xAGhWSf}&?B1-MBn8#5YaRLL@fzQjqC@(;x4XmTzXcVOz
zP`(XtYN~piSQlu9m<GaFM|r((?I2UNE|GYuWY7~~DX{d2@C-$d^l6`YDnk8pIeb|T
zAFDj0#(((jJQ~VGoFEK%4XsFdNP7_>fMRh(j5+qgXv=7S9mK*uP-$;e5MF{Ig_PAQ
zvpmu;%tMHaG4K$7kK-2vFz}j1lI_w88xhV47b96E6tQN}uD?HjFK}NG>O=FL@!WyZ
z!dSOZfEON7BCgad$Nf30y32H)&Hm#d$8nDQocndfTV9#{M!U%dq)5WYr6Do%dbcwB
zakqc~B^>pC4goP)Tfu1Wg~dJ2g1OL+DARR_!Wy_(LV)3_CJgB<d!{_24ZR#f&SOnW
zv$@twb!cpADAxv>q@2g5;(S=sh?EpQaL_Vc5oMb4U~Q5W7e1GShAe?W67u0?C<RHo
z0~GR@q2}RC<IYspk5H<M0xz#?a3_<4?=hslK$5h7J|97I_GLq-LeB44IDV$StQ=O=
zCZ;2>CMT&ax&(4`QL54doV{s!fkrFpe7($jpTKQ8Agu*)<(*kDlQaP2aMn?5zU&Y#
zp0>$sn=q^mR0=tIgnoH%gl&o6mV6NA_R6VQNnwoRG`=lQ=V>@W<{~FKwn7#3JWZTg
z<2Gi0&TMr|*_r)o>8F*v<E}|OcAPF_T7zY5H3iPjEi_J|_728+!$RB4vG8*;Q1GaB
z@FS^uO+t6(aJ!OFjP`0=tEBulNv92nfN}gXGqq-WKQZgLO7;~-5My?#-9pW-y%k5e
zsc-@Nl6rQvBKL8H)wYL`t6IdlzVxl7uB)Ygzpuv4Wv=bIoUSuwz2d3GXf3BNYyds9
z&*IS4PO~mX$O~J5cdK?nEBl711sM0RL>e(%GPWccfT3{+7aE5uw~8!3!MsSl*&ZT?
z8O_K`=dRSAUU{B3_Z~;Mf0F|tqN3TT>V_@^3q^EMuu-LJy3ALwf7#z0YkO{=%P|&z
zGcp@zdbzRBp%#IR_GT6^bx_9|1F!Q@bkb^#8;QGcf`HkAu%Ird|8m}T;5W@-OX|^4
zw_e7!Pez^nyp?O8Nwyo5Z@h5F9_l5RSm$;rlN`Fw`;N|-uJM?u;M?4_X_Bj2l*oyL
zCCEBkZo8x1Wt6A2udZviI-M%axIVXkIsA<4u}8O-QI*@_pps^eNikc3bi&%&Nv@Y?
zrYUXhRJ;CQwxqb@S;yEN<dGq8mJ=`TR8Flflp9#pcCOF#sv~zw1f@pK)lhb3uKoQW
z^Nr<?rEJH?R#vkTLwgVqHWYQ|PxVl<j5L<qO6?+jV?X)a_c+lVMDO%0{3{=S1?YM)
zNL6=N+mnWW?KvpEssZ|9a>4XD%+A)EExLM`+tmWf3akK!g>D!5Yt(;B>0l&XLK}(W
z^srb*t23xNgWWpIFjrk%Xbay=$z^XaUTEa9_1f#XpfN{Nu(_T~)vWWs%k^BP1IFKQ
z!<$<J#tSFKXvst;|Ch+`M%vqdSy-%tQzK?2=M4~Vg!3@Yo?S#G51l;AW1;ECZ^9{X
zV(l935GVEE5j#-VT9!q-kxx5B-YrU1<;&;A7vvU1R3ow!#Ac$acUjZbv$tQlQ_>2|
ziL1GsZhkotg2uIPc{y>!xUDd+GjBSy@gvs3ysyIOaRl6m5#GPJVTRd%@^n2X#e_yo
ze!ecw0w1^Q4mP!aeY8;laU)6jI9*>!I)B<osjbJ~AOnTL*DJGA1}dG3g<HzNC=9F{
zZ5nv9H7SF|dfZx*opXfs?ZgDmKjb7Zi4fMC)sR_4^{*Ws`R^Ed_3<2)MNod_Te3gp
zCN!eP^4{Ol*$f!RbWE*(87Wp}Z{D)~DwkqwpuZB>a{04U%RO;uMD`=*rlW8%v>t))
z2k$EcFx_~}m+k4kJ^I}mE7=6EH`sJZL*M9nH%N_s$6Rixg!}f%&(2JnQ!Gl4`44Dv
z0o<h0Z%U1w!Q{=1nKAtpP_(~GVod!hd&@sq=@H&AE$$J&k`?z5lXoS>Qy~xEsKfsO
zY9XD$m(k}e69F)nvF#lbmyq5o=zmq*{_^zWMrJ`s5nZ}XKm6(8Y4c^n#p1LSB&mCT
z+oTspE1?5*CW@eR_w~2#W%G$+t+XVj4F9a`zP@&B<fp8dD4`hXoV7E>C<$$u&bcUb
zdNnA<e2~(DIRgP3Fl``IQjAIGK>_PiiB?Z|HUh%$9)nt>Y5=P(r%KS4jei3*Ye&k)
z=9Prm3Mw`qM5G2=QYa=$3-ueHW)D~-l<Z_uv%O2|g!&!^yhn)gpd5`CM*9=wcu+<R
zV;I(Lb(Rt3-$W&wp)hH|<zLUoY~-f}+0ChtJXzNjV@Q}9NhIBtE!*&rp51UIDJ6|r
zwWw<{B;W@)_G5H?vTP_=fq$!^^{~9MsD+OW-<2noS9$t6w6e)O8cX2&@f3{MDbz?&
z*+tGYX<>aUhrt;R+*sX>|9aZ9k*R%@ouZvEK1}7*?-YBieZZa`u;%0+un_fOqG{aG
zps9SfpoRPfQp|UKs8K_>DSRzE^gIez&K<dRMnv5^@-H0#KXyGd5P#?hOZvO(hlL_*
zV%kGWd6&>?{3Wq%-u?Od^2775*i_x?UkMb^K#)~2bZ_~QgbX3Z?sI-NNhvyQ#3;2h
zopC}N34k)O4m@`oLmr#fhRNvXG#Z5ohG;Umk3l{G76Ox@vWA3l9L@ATPY3I&Dv5!K
zvW)bUu;jVv5%bHe^naMH>2m?atlzuMOSk!yrD`4d1QO(cI6C|-NmU7>JhG69WXlja
zfvB%ckxKNiz#XGgen(K@f-?{n1>s?xNiUgN<++SZ_%1`hRWOvHx->SvLk^5rj8r4F
z!bWUoEr@y(Rav@&gva~mw><9^vp!CK2Rt_l4#8O0-Pjm}!hc={u0F@GVKMp3-~~GG
z_!JHLo$PC5FO-&c7()V*pSsDC2HtK)<~|)Q5ags5nma5I?ieS6@Nj0qS3&H^f(s8B
zx(MQat90+Bk~Bg|Nc(tJJfANKaq(g{N4ex5dp}7z%P>`<1y6dqD<cWV{v0b&tkis_
zXt$(t_F-Jc34dsNU#LviWvNnXXya~JRrT*%qT9VWK+cdFaiUs)XKKVbj~*GrvO;t~
z95j|=Q1i&TWQC)?m#lEKsKE;V=;mNKTHH*+McF?A7`)R<#go;Dzf>cpcA3{?8>W%=
zFhW_gADHg$vlUcG6SK7`1~t{g{^AbaNTArN>&qG~S${j+RS3!b_dZ&eJfRsK>xP95
zgiJ_tLbeJbr^9MY5lZ6S--n6W(~@wu>b8;3JXc?q03nwlM0CEex)*$1=1^b``?=pY
zrtX!4%dU_IwL04Hsimk<x;HXbiXypnVDK8HsEIB`xD`t%hOsRaPA{9B6Uq_C%*v**
z$TBFG6n}@-O9jjNC8+^*gIw*pr9Gp1VQJ5>eV5YSSQ0j<HSLmNYy;P6Hkq<y&#^Y8
z{3*==j3~!3qSV@J<#T9xcXm^PWAe*%gBHrimTs|v(+htIQ;&@#S>b;g##9ZsBRL{P
z$+ND=5aOn%)`ZXRAW&cMgvK<MI|!svtROCOj(^VY9vGv9BN|O|TGy+2?Z2beZ)*G%
zTm<63mu5{DF~_)Gs?+Bn>UN$q9QLazW|iJ~LY$J&zGJV(`vIC3nw3@IOX<U*^ipG=
zPpZd~>)y}mmHOlhd6RW;KL(&2aE44wO|!9_v7MDU>0|_mkSud@Qg|?Qjm!xO=>6t;
z=6}a9u9!J7%m*?j27RT>No%C@(qB1qVpK28oEWz6k~v{-2^p@PfikMo1BfHa;zgN8
z1f38#=Ht9(Z9hLw0@KNX+*-)Ik?5`?31@_a#2EBBbX^d*8ckl1KzO7A2;dCkt^~q6
zcv0W*y)T|DwCBlKp~-&Z_PmxJNz2%thktzrKW)!j8!b)S^TPmd&kv)#J)b8}+w%i~
zS$lq1!~d~8Plhm!x91l||GVvZV^-<T6GDJ@_Q>vhfxf6aKOwoJJ8#U%&3W(jMa_A#
za2hcj*_<!X=QQW-yydyGjR`3!7aLLsbL+irSONni8?61h?cVhe^y0N&Tu=I%=6~+I
z^(0&H4c2;4;KoCYKA#zraH^j&>}>s$6ciREIN;yk_SX*Hr(K!J#gpGp+oodHEYdbL
zy6&WnNSb;oZ1&+Jhlv&MIC9AQWdD)FBm^FD<UokGp;p;sa)MJ&9Qy9{aOcKKw5YaE
zSIzWn=%{l96%mWbeEV|*M#xn5PJb(|Th@njgk|mzo+TiCjo3D9ar3hTGu4C&XPzS*
zr?d0V5(43;`KnDxWGpgVr6NZDhI}KgSP?UfD^|n|^MQ((L0_pNR`7ixsw-D=jOv9I
zF~jy<Dq^G**yL+A3OUAg&Jq~b70wbE#sg;w4Elnz1deI{Swchi;<E&f@qhBOgyT5M
zX9+7Zl+O~l!Ciiqz)fTSSpw(i{Idi_BtY$7c9wwHi+pd*V{ndfwRER<G3vHW9QOFu
zlg%!7fEW`y{09ZZmQ<Y2q2>7>0dq_{c3X&|)t$WCa};v=SG!yUj~6*+t<%Sx&6<^2
z{uUUelh&c(@Um}#AvLAL+#xwFe8hYBqusn=m8|bgd7npuAK&RCKJw+iNOb=J@cT2D
z0ofN50yZ$0k^e3eml&urDVJO4E5`wPw<G8)+z13PT=IjL(dR4^mrxur6_XWaD3??5
zE4%^Mmk{zRXn$L`@oi!T%-S)DTO!O7v1v!xK8XrHX3y`2d*KAU;t9N(0iTfPmMLiy
zUN3&-ca6#dD*o2jDDH@tfBA3XqnfAdd1<bn0X%<GUU|I|(}=U>nOsvF8vW(2-N`XM
zStI96%HVyHnSGjAb|E2|$S^;q(=%kHRE{>{nXMr)nSWM^IHpaSsV_z7>YCT9{Cpy1
zNlKzfgkD)(o&|)D)6+nef9kEtV^vduP0%jY8lp9*fC<do@;KT}r~oSS$B{xO%rf8#
z90GZzeP=Go-CPPY5u)&_5}+$m<XapMisLyUZQ>lRA)ME<<R3lWMQ+8XiCdAz8_k!r
zrcfXF7=QiPGB2ZtRZe|Gj~w=Pp;=SWAAa}pksqX4qJovUl71JiBwo2Za;aR&y%|-B
zsyrIaJ^jk^3(_B?Z7tJwO41Ho4m+<UGjAK!bDLJJOFqvchp4{JQV|UJjHZ1w5@>}!
z--tdEfI%BKTakd{lK7!{bb`%DiUbp1iv(Jue}9yPg`cFx_hQ%r8o#v|TNKA&uLp=z
z@|wgV<>^Zwf$V0a_T5@ytZ+TV;9tYl(3MvigV9y=tN86Weyf<V&_fb@9<P%OP#L?~
z+7u%UQOCsy;!QV26Zu8l=7(t%QZXTi-74r}%@K%>DQ5?+>)BlkB^YLEi`-7y7`_!H
zO@AXleqGd1p@E56bqe;VcPXVXj^Q;xc>#IWFl+4yIrxgB%wMG6$FspoK|L<rNisW)
z5I^UoJ2{3dA2xX}0VM>PkS6rm5kmc%bBkGI8=)L9N6m@ikdZ`R96U5eRT%wxjw>-D
z%tlv=ThK&UI2cD~BNGSALrR`T4DuO)aDR0?^{muxj5WfEsHdSq@l?i<6dwu7cATr$
zjTD4k@_KY52F4sL=;|OiM1-mKuy^F?TIS|89G&5~D%*MD1r;csDz5>wi$v?K%E8JW
z7ji0kN)o4?&*W6}WT!6pYzepXK~FucUj?qZG_r5BbwRr8*^lzR!Md63XMrcA=YP=3
zo%FrqF3s(-ypT1DkTp(csXvO0$R6Dp%tv$r35Vxrau|>1NrGdZ#BHA^dG9_=@KX{c
zWh}Zt_rgS}gDBMT1Vv2@)Ii%(f}$-mXldPSpM1!#<G3l!>Y#O@siac49?ZZ+9<3%8
z$d6Frd|9vdQZWHD&h@jAg2QPS=YI?)JK$YNlDJ|@!^?Ls1TbZ$y;KOztWdAmJtdtz
zUNeuE63Q3gRah8JgBF3E#~TyKmouGG%6%#)tYBJ*S8Uz^;}Z5V-D!r`EAUdrxRdFW
zF+Pu1a>KM9P0Y>WmEKK*n3J}@awfVR3pN9RLe7d;#(@T42$>=sQiTJPhkspj19&Ma
zP?6hCmID<ryKl1k7w}SGh0l*L6Io^vJjfH|lC2r}v4i0)Rv|YJH0K$I$%h!Ag&Js%
zd-r^zeSOZ^xA>l^1yexiAbvgpabgpuq&-Bp5Ev+obXve)uV>sRjQW*E+d=HhCiA$}
zwvaosMeI}BJ1Sq)|9WJN8-K~vTMRNCklTG6NGVqfMJH)JL0?MGM*8%v6mhX^-go>1
zmI)Z7W%&QZn@KyNh1x^jZ2JJJ*SsFfu>-MG1oTkU(+6Tne|yN|f!HE#<WV~aR2U6A
zz&uRz66wE4iyuGd?I6C=^5|^1!|;q|V?}Ja=^^rT%5G(hiQyyA+JB7HJ{yEDn0r7v
zpWGA%1Q3CNx(1p?)=o<)c=cHJxttWJhRv=YMS6py7qy3(LTUTJm4zffo(1=_pFd^J
z%8qxfnp5UMI**?+r|P`J!qsZV*w4Ps7~vB;pER4)?9O+(lWZa#gdtRCNew&1dQ+D-
zIkGyEyHa6sT>zM~y??@zWi(RPZ8Cb}rT%t?+2G8z?k-a)Fs2Y2(^^%IsWaXi^+rs`
zYdZtsT(8^t;Bg;UABie0tr~c1>Db>5dHs?aNz8ZTG)U={#x1-al>9fz$R3O6fOR^Q
z19U}Fvax$598{-<hkJD_?W)R0kOt>ghmaLQQc6{wlG{~9e1G(iq1(j9YpHHFzdmrt
zAh7fpn8)W2H`0+r&+6LT9-slsA8qws8iZwTB)EQ)tR8xkEd60JR$;ve*59^gOYZ2o
zJ@m$1dXFgFK*N2;HynVPEVzR@AKs*|LX=tE(GxbGvUpTHtJQ|h@2N`qDnvNEOZ&oF
zw}Tkspx)^U5`V{zcM1wKF;`x1{lp$cp|K8u>8JN73T0VojSK-;wWbX!9gif`54IjY
zK%7Jk;m}a07pUuwy5h`2xA>`a3xlyjG<2JeZoNpi>0cP+C-(ohBfN9hyeS8}TN2wl
z8lSkmdxZxZUa{U60#o&wHW09q8E(qEt<)h-xDhXY^?&V@;;YJXF4VFY8&PG|5}VIU
z%4_<RxtG?}oOdVNce`6qK?4?pYOeKzUaIr*sy(#;_?Q0S4LYPw1ywp%V~65m-zPME
zli`;jryu&^<$C{xL?^FYPd2}7tP<gce77og)8u}e)pT>%lWm2RyAeu{Lh4w7VWf0Q
z<o=z^nSWW~$8L^U1&Eytg96m_mOi1<?Yfw)f^X+U_5tR3Q}LkDa^_bcckQ?$D7CKo
zpbE8WOSwoYLl;%ct01Ya+ofPx>`)GCLm7FkN@3gg+2&e1!D~9Ms-C4R8*SD($`u1G
z8sD(N(RM}BpWK-oF(;E%jfJgpHfUBJ^|R-Bo_|j;M&56BA2-%g9SX!!Z|q~+plftY
zRNQ-u34Z9_c7#U`6)N(t3-z&y;=9V+V3(#qraq#%jwWlEDDB*h=Tss!&+A;5;wUVc
zsuI*57>G&@>R~jyCR@)7b7Yf=r};?$s&AyL;LZNc0h+%#i2g4Frex+>*wiRSYb%p5
z#eXE~eM2+1r7JnN{G=PZ)drAxQMGWzOh5gZp>VnWJ4F18Ku|SZ1tnD<8q~DK@mcX3
z#iUU6QBCexG<vO@&zQQKFaDbMCIl(5uA19m5>3$*iKbU1i`|btB0|3j;Tle^hEy#P
zK+l9)UuL9Hw;<MX&26Mc&9kO*Z*kPi-7_ALduRpZW~LR8SX{xIYXuC=l6Ic`$8>O2
zEWWK&$$_n1Q~q2Ne#%<u{`LL#HyHwb5SP*CEE57SG?&3o0uuo+mqDB-Czt5^D-C}k
z^`LCv!oZ9Z2S~Y?Zn14rrOE#np&q2@$0SI$(Abtpiab0bH86Y|VEFv@c)oqud>GO=
zMNDpZejSh(fPWEqlH!mMNAfs4zYqVu132RG;1K}4=aKk*;%8iL9w(N-$b(l9ms<+R
zGah-k`g9wQHNP*nyq0x~t)ic@G*o|S*DSlONv_SpT_oXQR<q<i+gi$+qosJ`Sqs+X
z=4J?U_KxL{E#tcU_x$n(4Tulhgug<8QW!Z_K_Sclk$sHpHG#`BYk%W;%wO3r<k;|I
zeFoQ%Y2)|})u&e(aWIe0&nI7~`NG&ljw}T6fd_HqlyMM=FzF0bxv;gdiqwBjA~dUy
ztb3SS>+WU8O|?>!j5QNkBeE1^iIisoPsJ>eb~<H;hQM_;0{AgbM*wo_BP&N=HvJWZ
z!G$r-{%C=<TuI%eRvgMVwWonsg70u8NxT9cKDHpq1{#8fOg&BE-cnG*Kar2z2$NH9
zsab|{$SyTP7%|ME;XAxHH)(%Tobj&ADwV)Upj|D|8&pbCXD<Y^M2WhW%!Zw{jKowh
z*+g44Ts~queYO}Z>Qlp7CE{9hXEE*GX>KOTl(DZib>^P^Z(7OX*1Z;<IPq6>a+Gix
zd&gN&Odap*L)Fk(Uwc;Ky5xX&BhEp~MDZN*d1=fnh1Sb#_8BD>L41Fc)#4rXul$CZ
zn|9D25+l5$e>mZE%R@!&-tdqpAG-f83)py}s3@b#{<`Dhp`Ft}9n<`s+fTr}>TT((
zmW{_XUyVDPNI~aJX%k#_z)u5=KGH<|q#4}sgPX_o%0HhqA3o7YNXL9+^tGjJwV6s8
z&IJ;&))3BtZB$yxa2|hQK3RQ@@85Awa;(h8Rx`Y59_4IabI?I)S>hcW`Bhm}FAmZ;
zYjZa(p{>{U-gv88r4m`+LuE>{*&DSsBRPw8A8nNd7v1Y{Mw&I|4V`$iC|O%!d{vae
zjUGtbmK-S}Z4AZ5)*Y3Mv@3~Z!I3tQ31j-}x|6jBi?sY`ixYnjPNhW)sai6v@m%t3
zGy<mC?@y^$HSfLo))HV#f=bx}<PYtfOsjSFRz!z;+#cC({ZuZ=5~-9OgGs;X&>s6M
zM#MRn4YZBDWYrl4q}_q2nJXkzY|_P%sYT1hiLVVX1|E}-z80+0tSMHQrV-uAGcSnd
z{q_EBkHhbAKOKKKueZRf0_P}<l+w|YezP?fgHTIKDfuq9dp9pWy!YIC++BJM_xC=%
zE*E<S<qS8Q=Bilz_}>j9Fgqek9CL8EjhD9O>$fvSOwY;O8QLQ)lGHDer9D&opR$(!
zMdccR_sgtA7suJ3_|Yn!NXzrn^X(_bP3@8jWo~41baI!OW<d)BHZU}o;lL#nl?*Jw
z3GE=4Hhlmg^*Wa!gFh1kHZV1p0sS@;wf`*51__JCk6ix(5WKdR5my%z0XCPx7Yr1a
z{Rb?;e}|5LzyEQN*;pyO2|vSE{qODl@ZX^~U1>qacjT?min@Qk9YO^H31fZdl;Fzm
z|8`#vf7U{KRiwM&<NMdHD*2qTbHLb1V~8f$DI<tnXN(s*R3+?)$BsN^;xS9ckob`=
z#|I+=N+828^;FQ~N9Ph}hdNzhB3WhnAejuEf0VGHk7qA0T8UPmw#7<(Wy=_SJSS_B
ztx-^EHI~JL?jR_NS|=q@>jE}pgx0>+s&w{+pp|hVm2e7}5&H{KvZUZgc4ASHu7bCs
zsijA>Ws-_44fFWqxuCD#g6t$yU;wXt#L-PkK9BT~6NuD|Mpk&OkZ6qX>|vEe!0i$(
ze+VU;d4|^=5=o#@jYKm)C)NuKw(&DL#=l-08|H$;5Bpv__ph^7>EQj)%v@fRt95HH
zc$^I)=Uu*_NePMSj=sR*m)ATr64g9L9KR$rqYCrXV{|=GHt|Lqfz}OP6Jp=6Cc`)7
z|9PdCHR^M#5))L43-p#^M~N9Fa>d^+f5nZ^PTdq&Qf`X<e3osFL48mgo1O=eA)VoK
z{;yLvE{adHz*az|rEtI#uZW%0Pr_KWP^QTOlEdd@F(epuIl%-6XzCXF>J`ap!-xSn
zcV8SZ-Fd+Da|bN7U||<WZ0<Z_bH@=c&?hIg<6=$5nZOP_4>92}S2OU1mWLKHe`ls~
zx~yfF*KokhO>E1&XOY<wf*`enfrPrKanbgoiN^pV4xJfRfL~{<cWIwz@GL=qwUzeM
zIWTRoPf3528FbK5#XPKzu?{lhTg$o8k#H7j&aIU2fLBO&{e1(=yn%$HR|4=J(_E)V
za7OS7t`uD7nk9MQX%M}JeQC?{f2wP}amB(gb2SS?FT6I5wJ&q?+^?X8y<wSbBH^9x
zp~d7oanuR6>}e5?j6(o4f~!luFFQxNn!`)W6%$uvN4_wzXEHy%WK4`B_bS$y)J_iV
zqL#7rNCSqkWU`Q$OyA4r(n1?vijIkXGb1<667M&h3Le3a(<R3=r5EBUf1(ziH^ueb
z?c0AXc?yLHXtCBtRpe1^<Y|0OGrRA%(s~Ry;nI{2HLeV$lp>YhR8Jt+gMKdLa@s4o
zVRG|RmOCM<*-4Bpt%Lj6+LpzIkhH9E@4iRURz@_8uh?CdWkhGNsJP!6`{Lp}KETtk
zKnGjsG*}z@7G8{YequRJe@jo}%|%)c5o5fZU@_9FT14URr$JmkV;N#(<}+pDvT@6~
z6yu&{PSZI%UYLTnW&f2<fKjs!+{dJuY+Pz{Nwn7%-jf_LxycB!am;gbk#^!>cIsID
z_o>(2M7M46x8EoY@%5BP)Q_$oMRe4>3`-8MJ-Ux)&g>U56K@rTe^UL%DV$K!^AM3o
zh-_jxU`g=auVFdN<rTBrD`$(>kL)6?xqND;nGf@3X~n7CwG+8Y;!HqY%T&NY<iC~V
z>O`b(&2y$XS5&R_9G8l_9`=T}Y}5WzP>no?-hA*WXr(n|Ve%t}m^GI;Q|)jD<WLyZ
zjNv*Hr5}JJFA+POf9`w~j#I6BahyuK0mqm7<0!-0fse*<s&y}pQ)xHgXzz|=IDPi0
zKN!bK>rNaiX*b~b8g4W;?<^~Qd+r_(1S;D|+;R61Vw{6UB3<0|)?Og*?`OpO6GeB-
z$F)RUpMMPs$9w%s(o+7lI-JVYtvC1)*pR`YmIQX;7$vcUf2y_Iz&A33sP$f1T@}vq
zzJ8&N0Qu3ktCiL}9(pO}x{dq`5ov@YW#|SROUVTs`xRW_*zWOeT3f}(6$!vT!<W;H
z4cxYpq~HR-&Hh&D4s($6Ld<{h&4Bf>Td{*sV-LdiNUM0yD}61oc`IX0ehJ6G4Fd9Q
zA$I57QQ@`Wf0n-FjQe3zyz(M8!&?F<paC);y*921+}Qac{BF7o&u7z>upM{Nz4uNf
z?%)RCop4S!5_fO|8E!gkBXMC?q!rpX7*+2jyp|~lH5)uH`*3B^c`-^9U#W~LhfJsZ
zdIXNm8Xe<r543&ih=&-x;_3O=)%a>WM`=|FDb{Vde<sao6uc73POeE~ptSAlQwssL
z9QZ+d@moqb5EU6Om)a-tc;nS6jO&tPg?~Tpg@yRCLVPh{#S06;nIcfx-^2Ye$`;((
z5y3W~OoAF}L6%~+)G)C18}FLOD{*2S73<Vx>)fU(o79b$SMU7@9n&Cd*)uE7XAMq9
z`|EF?Ptt)U;(Qm&U0~Yfmrp<EjvcUNCj0K#0en0Wo#UbTOg3H-P`N}Hz5SqiR(I`w
zoh9-<N8NnCzDD!H4X7u9wc?dm>(zDl+xy{v=T^d<m(k}e69O<dm+|Ed6PKtDEW82c
zw`CA4T?Yi_E@?iO;Z-XX0XCP>wlpJ`KZG_6m+7!LD1X&j-)|c^4u0=nF~Gj;9Z-!(
zQKEDx0(7_Ay&l@zqU{2A7v#Z7oYcL<4*a9&{rC4#_Bio6ZXCxcx({_ZGnB}osE;2t
zi6u(on^;n8C?pfKVXi`MI2IF16Fiw9O^^UVwGp*P8AEflN1;j4=273MrY8`n$wloB
zH{_GesDGg$D%m7~wNUk8R}!048j3}lOu3;HxXFx>Nvpxskdi<uxR;}cfr5J_+Pv_9
zJ=q*NHB4d-WqXaek9ZFu^FJ57QGJ2p8%}CtVUV*oy+V+Lo8)oH2*{J<VKR8Nv2Z((
znvH^R1TWBo$q>mT@0*|;Orj>_7)`=cGY%;<LVxT{l1AWyNwJZT6*hcoB+j2AtVz^n
zCImPmf>;U3RO}We$%#xYH&VPAV|eD18M75bgiH_drBWrlRp0;<@GJ^K1qm^fZ1XXq
z6ocu(c8qE-&KRrST(-5*QYF~Pp^n9BX07+?+uM&lATO{H;5XqXV`w5Z69GOln+d^B
zh=2BGf>C1G;<3U>ikS_VGu5X7!!+AdEK{<~gM^GiXkbE8qeOL<+$j4h6-ZFY@JTt_
zbXb24Y~Dp~RI0O>U87i#k);}=IfZr5l#y_-ugd(yRv*b*wJ=BaHXFjG8D2+^L7KBg
z!lc3$31n{e$t+;IWYRj;q++gP`(&~A^naK%$C^@5&RU;h;c}ioO^~$Cf&do6VrDE-
zF0tC6rdWn0k43^m8-X2IIMt^GQcW`$FA=Zt>{-+PJh~W--(U4k`c1(8e%G{bd(&xu
zJZyM=`OBB-ukWY5X}_WQjhCb0^x3oaCBj8M8z{VMU&1~Sm-z+qkyyg~0y~HzQhx;(
z2M6uj@#y4ze|p%yeetsW(El<$IyitI?T7ze^<nwE4~K@+{%|@mtVnf=cm2s|Ha_W3
z=E!2V-t<ogy`M&34t--u#1`cvj2~m9g(S_**~~8vD`wfKfBpME$P27KHN)A(#nG(_
z<ke-{Dml7R!WNob$T1T7TJ!r3tAFOTuxi>qtnKsRa8xJCei7!witF|o71tL6bpm(7
zQb#(WM)PRtYMr`$KRce<)$76VzfJqoXnfiqSC{-z`}g)$`|}~qFW&V}riVaa#}zHa
zWh`w7!3hdVisFE#NA2?l)4y+j9eo(JFPa}G7&I6S|LP>izZ@Lg2-GgtiGPDLr~wK)
z;1LyZ1jNqS3DcVGlaIYCWPuB}7{1*WmbszsO(nQYwK6!oFp7&)z?y{UfWAjhR$OnN
zFK4gs3om|u_sahL_;EVDn*6P8FE3`tgP}VgOh3+!-C)#Sou0LyW|vn_CxdbSv%@9)
zeDc5u1%=lrJ0Y<Xwc=D98-G<$cBuDjG14Hag+v5^87Rg*tP-jlwkmvx-bEC?kl96S
zvXDHMYki%#U<(qet1k?FVcmldIk^yd=>w>SGqSM3g`|>_q6!6&0cDgOez*@30a<0~
z0t#K-I-7hD)M!r$i~tpdg9o`;`e0y);&Kg!sYdt~a9z<_fGa1bVSfunA$9!cfz}e%
zptaBz&{|4dh1Qa;p*6<^TF&uz&RG{i?u4=sir)jGW^y2eM(>?RwFp2~TO0<^q*T^-
z3%GiT9vc_G#0sjphg1cUCZpO5C%{3S7~$UF1QJjZI8QX*3qN8I|8OtZi!)A$X2AjK
z(G0LtR4wcVr!jpK;D4|ZV@(#J&{+pW0Iy^X47Qvv@tgTzZ68^gL+e74oq~*!6}lkb
z3cc7_U(q=cLA(kUXQLBY(BTMr#Eh+uj^c(BhBt*b)55*s%ndYI!I}F(3%V%(2WYAN
z-Myiuwqr|ZDII>4IeyeubuM_{3DgP8N1+p>6JjT%PS8EzP=CN?LoX*O;(xA08+|Xh
zR&+skO#HiPmIS1g9buqg7dsDNAQ5`Nfc+Oe-=Allj!q`-)9hl<yXXzi-DrH?_J{3f
zYA<`kS?{8~>`!}F<I(xJcX>G&p0_{z^5tsWpG+{@<Y{ks`n3PKcQNbP?6>i;yWfZy
z)C=qfar5A+;D1)m4kkht7e&9j7rHMt>4NwRDg&T7WbjMldJ*Jiozwuz&@Q8x7PKER
zX->3<Isqn?MVCI5;?Tu!_Tf>ARUmXiM(^fM1T7$0veAi0DcfwRX0+D2Ftn8+s4LjJ
z0liD$ydH_jnNeYhNdvq9`QL*V!H6x&bmxVD@EY82@_&M(k*e&0Fr*R1KJB4qViMHn
zvUDO^)A5y11XhXYdK3vw*q=V`BQX=ryDtwC0VW+%0eOHE6wm^cP2I6$bOe4MyaD_v
zfH&2~!!~C@EYuEF0qJ_bba-)Ck6iC;IsTxv=p^7lieU>)F2=9&w<H7YOEFHxX>lQT
zp=*uGeSbLv^BQMhzYAyJWjj*OjTcm*JiNqr`4ItKsMyF40k1~M1+@+7+w9ZM2jkZf
z`*fM3$S=9)KDCr$`s!)pib9uj(Ev(y3rz+)d7Q(>&B^(Y6fEm0sN1LDjZoxmHvWp%
zBD9K%6!}jRu@KiH7Q!wO3vx#u-cYzcm;uGsvVVmp6)Pz67-8KM$p;+GBroe&C_7=H
z$Tb#*t|bZ`D2CfW@ot80)ZD(;@SHc=v6p>(0}@e8_b2F!Bi)nHWebLW0nR-cv~OPh
z^Yz=;gJbvUVlS*)9oculj7p9|!znn@`pwpXTzrjKl-=4g?0Tb5ok6P+#XdSN3_c`C
z^?%U&wEww%)0=+$?P~f8X)``U9wS!mU^1EYCoQ30xQ?6c3ZXk7=`nW1z4Knz^FHnZ
zCz9@flN$<4(gl#P52L$iJ|uW6pm8IT_YGbQYr%`O3ur{W293CfBU<s`sFZ3e52`9L
zx#ZE=+cz)0rQZye2!yU93#cu6-l8j<C4cfZ)pbRy`%EfZXhl0d^<@i9&?*37+(Og7
zhC$v&6R1k*v2$x*UlEc|>sgm}%DR+qWL;OGVt3S+{7$9H4wIt&u}w<OSaW1o>feK~
z%e#DqybXr`F?;jYp!fJ<bljeuAze?-kh{l3XB18)oJr_wUyo1qRQSxFNjWRjPJhNH
z?Pc#|Jepwa53hqS{PcAo3VSGbBKDdEI}Ob{U?F0c1eHb-?u+brxUzk;ZKGSk>B3F+
z2~YSt3O6MSHz~_%l}vm`-AK3Kg?;SFyp1L(QkHGzBs5M`!Zw;<UnKPC)OMwPSb>#I
zrvdjx5?x0U-3dwbM*T`1fat>jgfamjXrz}Z%|MV6<M)Ihlp+v>z26IhM5n?2cF~Rl
z0hqqjBM$;`_P2e|+;C3)fA4&rmyur@69G1tQ86eKm*XKU!GFu)j(@-Zagaq=DT0k(
z<B9)zeLwtnXre1Ej13(IFKjT~Ki>}V27(dRg$_Ijt-A03c3%#Ea;lA&q(}0Hk=^&N
z9lP?mfEOfzC#^MR1fH_O7~feNgpRi|czBl5kYe~o2PuyPlY)_wpWZyoQZ!a*9XLfo
zQ5u342~tdyGk-OKQO^32#2;ND@f&4=BO01fR&pd)TOFj(4*6<5k%KOfn@G|UUW-iE
zA-#$xI)U14z*CaRi9F~IBBR80bmj93%*R|6DQiOJW=zi+O_9VeAcw-Li4GAZGl$;W
z@uQV-`<NeqqcXxkcr=b<u6uDDQ(J-K%l&bbT3D$bjep~q>s}nk)K=g~TI$!$aYkt;
zz-i78km0j|B7#8WCKS&*C{l_b@Hs^@i=e_Fg0>SckazbpLowQVykj2W6mfgruqZq_
zPH8D`I8FJ&^3^x^5jZC-0H?%?1)km5YI#GRESZn<y}Y|P&hox~VS;ehJ~;j`t}*_k
zGVU!$i+`A}j4rV#{<isu62+)uApV};QB<N$=wFCP0~}3(uE()nxqxHc!5qgTIX)Mu
zSMIIq<LHH#Zd9X1Yp5wcgQyx&DiPecqV?(KD9dC(*d1jVMHUmcQ7ovBDazM*pD<n^
znhB~e+p)2V`hjpyY^)W*=quQ?)eeM`exi5QWPjpCWJ!;HjTgmhvE|aQ(O<XYqL8ix
z&XGm(ZC<EE0M9Az$@SH*8O8H{zh)Gda(%80*y@z)(;ik&IVz<n^_=%FHk;I{V6$z=
z!Xo&IZD2mjY8eD|y%XO#59Qb$HCrB_EaZLPMfe=s(XES?)IkDOO^iy#s9BV@SO1aZ
z9DiLIX6Tv~V;0e)^kGY4H0x7T1T1c<l(NrP!gRSyp`^G%nt~7&(De$1xAKibS9O?~
z^|^TLohqi*DpO7D>AUYsezifT^;jH}!b3O=hl6`OxS<+NIXRASsD*J9(ItxDfO}~L
zo#n<`%x8!4&h5}NW*W&0S>9@@^H}rYE`M(RjT9D05l61aYBXnYq~))_eTa|JBRKzv
zqeo!;$S)r*48i(c;-&U<nKFmBv8D^OweoJ6yVF;B)4^Yzp3(lau8$$MVCxI+cG6jv
zc(FK2M6!F(ey(mzHZi$^O^$8z_DU8Fjnkscl|!5fVBZ;+dN>+44lARjZ{oTzB7dch
zz$wkRKEK6LCb%*cOPzS6YmO@Kr(?PYrRvx^)YRTnYMl%<RZO^eWvI!|D|&%6<Xi~g
zCEV&@1O(V&)@ietTRcX2#c0CzX{df5r012PqowIN5Eq@H+J{;RA9^RpZ0;nltByhK
zcA1u&tI@r!ypCmE>5`jW+AvDT;D4mKv<u_pC7okI?Sq+6scloFa;iEfE%iWVS<dI8
zsT*3M=`ykCMZGjN3l1g%EfJYlOM4|WT27_x*hSW~F=6Te8+jw3&=O`1Q{A<9BXboH
zq+TOT>D9<oI?P3BEJZdq)k!7CyvWyCs_yJ8-@c<><V%rNdB11@iaM9JAb%?Av-&Jh
zwtdo9W+xoZMYl%aIHZuyPF^d~*--d=#Cs(T)IocF$9W*5c*pljjPjNDN^JFm-Yco#
z>85$`9F1|^Imc(z*1lIN7;f*fly_-|SC2U{eC}Wsiay}|qpdRLdWUjGQFW?=;*_sj
zsE^V&O|0v@i}78WaFkgVTYr7a4AL2acB=2|?Xw!MQ$xfx*)dNFN}#_YA|JwVXl!4v
z<vBp<-Cyn(n5bn^eT#gDK+h<)TDa)ZW*T#jxg%I;DycxtlpnR_p3X~J)1Ft2XDIm}
zCJXbGS-b2WD=7ukME9v_IFuL#gce}?I{fkNJ9sN7DBZDr2TM|nUw_}Xz^DgltCU%(
z=q=uJrC4kn5JXUVb^E@S>*e-+%=HJi?;wp(!Tv<;I{;9PuXmVrseNaZS5zbuemB4W
z;#$eY%Co+WVo%gcEKLl3871r)3{XR)Tp0e3uKHB#^<D1bmB}-LHcqE2;VG|ga|~Da
za`wmAG1uqX`W3SUWq<tB?)@9nvqxYmE?9H1>z09%qGd!XE=iKI7x7eda83RkX$QaZ
znuf&ETw|e?4~`;(<N9nHjv(1RR(*^Jn~h|Nn`K`5f-*Z_(`n+4O}9E=;AmXZ(0~>I
z9PAk;mozkSh-A2iB{FjNw<xsg|M#(e>~+;O)4vJo>RMR5uq+u#z3+R#@;J^vjU~xX
z_D0!0s3UH9OSVztVB60A7otxmEW}d^`5y(<&hnSh=PVNfFgce2{WcQ<I53y-R|O`Q
zFp~%bmzOCllYiwp|M~XYNoHX`y^@=iRB1uRcgCV%>|RnvAeIza7M@HJON+03WmMr!
z01mgW)vuSg(?6#go}3iGkN^643J(xS80$Ny1V8-sclZ7Ddkp}TKx@C61(!ndi1b0S
z`}L!%TE69#@RIeD#t=>QGlE!kLrN%RihlTob#hs!YtmCeWifvTvxsDb*1krJR^oL+
z8>|$Nv!yV|Sw5oplNJ`gM|Kdsrb1S7B3IwrOQ9{+y>%l8H}FbI))HQ_Ot(#X)1I1V
zmu`nWC9gQcgYF<q6la~ZeA{62>8wiU7<kPd8oZSTuMPTBwgb_u6%|LS_Nqn>7Na>!
zzMh|q3~@=9-Rpm=&<n3kqlIkNco(V(LDbkL?jm|#XA45fW)0;vYec#mKH}Vyma)ec
zZvZbe`aTve!Afe-E!!-~6n3{pwj+F!ycIa?wg<#`g-AZ31jLB!vZq|`vK~1jFl@`?
zm0PgV{1&-)H%4H!T)i8EiSD_&9mlP9!DvMtw88?uFGznkg<i7^<w<h2;CQn$sf8uH
z&K+vKi!2=R#g?Fj20K3Dd8<)DnkYFI?aoh%f@HA;f^MNjKTrFIzLd7Bm=F+5$CWV!
zC0HrmuY=X%c+;bgdX-R!--~qA_uRIH=JjtZqO}pq>DnSKYDSs2t2Q9Kq#C8zG0GCw
zkL`??ti^v|$<RZMf2BTs%lsj|q!ES<n@pq863V<rXi1QD4a(=d7E#K6UCZi}C)eVy
zy$-GShuIsylpwU}Hx)8vEkmkl2V1S_59kD<Zm>4T)~a221fxTZ9#UUaK~vp$?)eFx
zTkp;K-1goy+Vmr2?s+ZgH&XJJ^ee^D;A6RUB+-9p<TXWS?z180@;G-+vRsk|VJ}c+
zFEDxj>nP4mv{16^dP<K@3lHj1DLoSoA?<&c(Ie+QN1pW)yaTiO3GR(te$w}YGI=(Y
z0H=<$<P`B8nKBp@4sDIpWXZP~MUQBC^X09)Mm&H)C}imyt>2Z#>}C`p#4X&k(+)(h
z1U7$8u|g);Xu`Wz{r9KCs5}!|y?WYIIir3_GNK$Z!ErYDVB&#+L~_n@U<geSA4EK~
z=h@JF9P!Y^M?D`&k0BnKtRUwfNIVu2pOB2kQLp7l&24sict<yd;aYOnx#f=YxN)Lz
zCvnky2d}me7wu$4TvE#taj}`WlyTu(?6-f(PNjG3>HA+>QYr@<Hw;b!SE&r5@Cs@h
z-M9YhBq6K{hhjD3E!vaUA=93$|A_YFmVLB`LRfN?*4po*J-wd+2&GqO58R|ZYJv3U
zCnIW;TW(Vu#bius-AK17^44h$=WvQAt&q#^QV2)ZyZEHe_r5Too`-nI1%P5BJEngP
z1!=HC`W~~yK&9f2qDNFFl?GHpkb7J9F*C-dJbUrPG;@^8XGv*{QZ&L$U^tSK#MhA~
zpPTj1x|tzwto=!OAMM)^+cC9^5}Ld_Dsr5daHhkT)8%k%T8H{`>&wJ*bC+uFaKaRi
ziF-ovTAEU;TRSoz!_vDNeIw0NfoFe=-Z33a^MFXe%#qD^;$-_Hm>wTZLV`aqy{Zro
zWP26z4Zc_P{vgJuv=boVgWug$vmW^Frb6D3$6Zq{hS&X!i$Y0^8g1@*f5X{t*juKE
zfbU1l?{{j%FLxl4K+{+n%)<L6FD%%`&*VG!t&8tA8*ATX@j_W^ZRid}FVKHC7MEY<
zDKj{ZnMSJ+dv}mIrK|uPD0?Ur%E9|%9LXWe2-#AK2JNRAYdXvMdYLQ1)>(J@C&R6L
zJ0v8H|LxFG^BK*DkYc6??%2`*EfC&!NGFNC6?aQVE7IA+qyvwF_nO_(nHA~$!%8Zg
z><EqSz%#!>Tr8=?vtj@FpHF}EY8Ie{^B;v#t(Nkx9RLgnJ=8xIedmx;4sa-H!UvUd
zK&i<erw%XW{I5ze;gr+3r4ob6Rh5|3a;p+E9v6Q3jjxdNxW2xTUZJhkZn8Wo_w;?s
zhWbsCYh(Rpti?)&BxB|Z$?&W2-rTbD#jt1g<7~BXI}0SlK2G)GSLJ_^IRb2r;0$Y=
zH7r$5v|o5(%uld>DeU#em%`o&ho%zah{6dn{gt)?uInaA&ZW9Z)}quAqn1l`liYWq
zZemEz)J?M8#uc^!WmGpA>^oOC$vTYcCIiZ}Zqi55Z0&@}&%WA;L7_(p%14$?rc)nW
zIhle(MPBorkFkFNYdHbTm(htg69hIfFfo^LbO;lZ5g8|!@GvY60hE{jFf0~-rDRIB
zWjl_OY!6OiS*FO3|BUW_bh`U;c>6efx)|Ge6$m_bAKyFR2bw=Y^P;2911M*p`}o=Y
z^{O>GzTWt<S{uWc=ih<9>G)zG@D3op_$k2g#al@4@E(KxyT<kJ7~X|v{s#U&rFDKk
zYNr#0AXhy8{qe^^3$vdt6c#yuMWe{Y&UvetcIA{Q5OZpo7d}iAb89brg`TR4E)h;&
zUz)r3kHf!*mONRf2weBi&qI1(ft2&H1O6uGyI<eB?}tBIvckGnn&0U^NO!+}bj`|F
z(M8lOp>~dFDj^@u!aD+G5sQTOGfNdjs&g+XLW5a=MT+E<F|nnXabit>S)Xu9lm#oB
z_$(ip{mBZmUnvL3cGFNsJ>jeFj#?RS+udfy2QRQLHA@Na!qYaSn|Na6o%We{PASrP
zKYq@tV0}xe!B4a8@vC4uNCIbyla(<=Z<HZ7qiDqYF5(->PiXL$ggXiTM2p%(w9NfX
z>ZB=J(%|Lz;&h4#<~<>Qyc-+zpiqZRZ1zx$%EJt8u=i68Mnw-Zik}~#7!=}6f8dsN
z5>eT?Y3to!vq`8J;vDRU05@QiC!7gG_ftev=w^WH&x`PI&QgA+@8s>e*dv*4_{c|3
zR>q!H;s^vqmIlr6r3bsu9#lLEl2=tfh_cJ|zLq>D`5?XbzM1cTVTeficNl!tZ2GO{
zl)7MjJ7i?Q0<glk9-vUOE8TE%^72#(u<?n!Vmy3@rBYytz&y+yMW-Bw-Q3X??)>w(
z3tTI339*+ucZIvS!rkLpkz8_7cw$y;C{paHB#N1Gkerd~!T0?cQ)-Cj6rh!g!)8lq
zrPL2a03uXA*7{|C)pnMPLk7VFUgj(vT*~B8qnx|+<ZDHIwHD?=b>a_BVu@H+iZ`VW
zt~$(CvDyb%Mmc55yQ+u8K|XDj?{bPqWpL;=#P(7>m||VR^zM^a`kYA0wg;assT~0?
zrD!%*m&+hbu1I*y&j9XgC9ENK7HC@uPGk}6MkmpJ-%tpD=FVcpLnT643+$=kLG8|b
zEjTqNzFi*H<4yzVJq45Tii_U%-0YArXss=rY*f*>8Mo&p9apDWF{+|yO_`HX^IUAG
zrtyrx@dcquo|3#TE93pSMCi-$Y<~oGmmY$2rg)w-yu1rvrTDm7Qq-NdDP49KIr3BS
z75gT1eWjUy3!Hkl+-j}`2*W89IrcS(u2Nani`78|)@?&2SsohmS`Zm?+{;mMc4IuM
z{F2b9$bBw2PY^1Wj1&SUtK}=0l4547SHe6WuQ+PV?CbuPZp)HQGzuY%abJpxVTg(|
z3Z69>YLRS4*m8}qtjqC}AZg5(RZH@=)s3#wb1-idrI?<+)!R9^^!s$ojL+)x-Xx5*
zmmxJQ6Mt+!0%za`$8T-&*3)jBKl9OZeHw9W<lX+fMWu2TWz4vZN~~Dj19v^;bDRwv
zO&by<=bZ*}eNq^OZ>fQczJ#|m6GqK78v3P!E@E*B{C6Q8tc)!4@ooeoltwK{NIO(4
zyGx`M^~tJ4#s01+eMJ8<imB5#af~+gCEW`w@qgtDRlA~mIo(d~sy3<dFL@pXy~}_k
zokp}z%O*pdUww@OOs~&ni=*20Xkv`A78j#QPj5Lwm8Vy;syIZxV=I=ycHK92r~Jga
zA+OdPSWp<yC}&T24SPaG<--=Vd|f@&xGs`a2{ILlhs6ecMRsefZqk5T$Ny`r4l$@`
z=YL~$&25?vzh$hRMrI*Ca}3;2eBxNWp}a9xZ+8Etu{sA&6}{Vm=Pg~#Cg#&qpVY!&
z`umH`v!Q;ahZ&(r0^ERNS-FBDTK0edGZg2E@g|Fu;Xk(?N{{NdxBD${O6kFQa2dxm
zRP!H*X4$rjMBJ}wiUU6POEjhQEQzL6vwuYMKQ$k*lxq1Vs9}&bEs8en!$k$$S50U<
zJ05^j5T5zBWkYdwp!oLCP!GAgLNrsG+%?$Sn+qu^q0NLYk0mw|;j;#d!>nkvAOH;9
z3gQ!Ii)?_(Xc&j?6z;Mr&tfPk$YN7!m48iWT5^%UY1WwjsMEAHYP_G3zPH@yO%lMe
z3c{3^ozur=BFL8YAGXcSmyvD=69P6dmr#=z6qjx{EIk7#SKF7aH!LcDuOCle-<7^t
z;Rq~VK7PFbzmWVFBsUtmSO8%ZTt0rg{PS5#H2l2sqoq`eAIskzf6~hr1&(_F@xyNc
zUcOijVGQ?Cn8qrT#whNEH~s|vyoPoj@1>Dp2A7}f%fCMU{v`8arSK+v4IkrwZy!(p
zd#X!ECCK=TAuB+_<=?+YJ%tw<NEquch{A)u{QjTIuTTFNXDzhGPaesAKwUscdHFpI
z=vscs!gRpGl*SNs!bBs8+{G9rl(kiuEI`eIAZAvxF!Vc@o;WNb4tjRhyaHYo@fcF6
zkGv(9&NwVBNj-fA8hBo^TU;R-VVKfDG0sZ6YenN&XXdr~B=5>Qe%e}<W=dvWt4lMM
zS@g9?JMMqb6&G6QIrj_FlU=LK>oB>0naSn7yp@+kIV?7Rn(1bl#<iu<>@IDVO|*I2
z??rN^?OvpolN>hV3F}y0+hv+1Qgz8b*kP}6MBWO8bY!`s2MN}TfE${I07aJcf{<;e
z%A)9-NxY=qk*ZqsEB0t?WuX|YJ^0hN7Zv@7+qX%_Hc2>gO0O?pj12PuuNrtxl%XZ8
z1@ykng%~3^%{XK7Kre`CO=Y#&KSge_a%-~WRDKGVWyfqgKFYPW6s?PLc(47;BJDVX
zF0<ABNb3Xh+(`IVmoYjlEPuYU+t_2zaFkm{VUEtlFGtIGueyFWSke}S^*Dkhhk{@3
zW2Jz9@*m3BsB0<M$jd!!L|O}nSZ(E7Mgzr`nNNhlb8nz6j?PN?amLx_Wl8C<&hjTN
zziauWK~@eTC>~~Jzp#$OByY1Gd#e@S8?c#VF5xY&=~$$P@{B%6-+v*`Ga&a+<~`LA
zlgTgUUML68-)ie5IV}uMJOnANMK1YX{4)IjIZD=dXZ~n8_EdjRrhw>#F?ysXU+=aI
zR3WRHqy@B2S8--$)!3UJpsU$Wu_#A*!DN+p%}P!%be*`->)e>&B&R(q`BJ6NsCM}i
z>31!ypUJwX*@JNU{C_z?l%KbhxN`)L;odJ{q$YXQl$2~B!fVRlV=tRB_nImeIOdyl
zz$MTWGW+<hgBz($6=<8`OI)1KPzaa)kQNL4m1mAm@B^$_t<Asp*i~~-v&2Eovx)iE
z^L{Q%acPfnwXhV~(HUl{IA!o*gZ_Md*R1BkTN~1g-cMP!;eVFbzy9r~1Szy$A7Xh3
z^h5sg(@wEv(P1Z*<7P#MHJ$yMmzZ^V?afXWSAmD_d_0*ZOQItCK@!npZ<eey7~#Es
zFfZI(Z=PLwC&<?f4|6;Sx)lmyQAh5jxB+Y#CVD?oyuO=~JE|rFU-GDz_%5T$yK$g0
zzyZkpiHc^5fq$<7{hew4QY=zIMaNkaw0Ia04zt0x@a%%Zk*jO=Ax!}#270pr7^|Pw
zs~$Wid~Z1w#0G#}n^MeqK9wYaT1|XHc~*H<;CP#7S$|I<Q7I65sYd9e?WuyY)Kng;
zmJ|V#DK?ciftH**Q^@tL%_@yc>smHu$-Cn^N<Ad5ZGYGAD5->FZ`L#3@A_de!m%&z
zCwSODNQ_oF_=n(fmD%?XXKY<M;nV}Rz9a|9_C||Jf%>#j)d<gFTD_%FC|iZ^4DH+2
zTX^nUwd}?{V7dd9?=dChl;7CXrgZI#b*VTN-kPN_beGws!qPpa=N>dYe9-h5Mk?(F
zLEU3|Mt=)SI7|>rHM%LW$vmlD_@-|geD9XtI0*v|uETn7yKB_S2&acJ8BcE#(j4~G
z$d~@MPe=zF_cNueYC|xYw3S$vO_{FSc$+uge9p!P<V9dcns@9xN_L0{(+s7u4a=(g
z6M_`{+fCo9kth<5fdu<Vp|RvRuW~)(2F{bi8Gl*dyPB;g7*Bqvq=>nanv6>B%gt+E
z6OL5kEX5U@3S4eU4$0$@DMoN`^2Rg6OyPvX=}x-#_^m6=Jx)3c-f3p#fcIT*VL39=
zG($PV$t$EKUKNgej+VYNf0L9q?Ax#-u`sRV+#ZFrG9Z?5^mREWOlO9br+mhc|68|+
zK7YT@+ROk`+oi^-KkbYWNC}B%q!6<_Ox1#z>E&AHq;vUy)l2LsG9$wvGZD=O_(f^^
z+-$mQTVm|hQ$&pK`xLeIGtH4?Ian%(PWDp|Y!!zKGHMd0X---SSNX0;6l7CZX-iyF
zIbcH#0h!uJRK9=SIW*W5-3PWPFV*%%FMn<CVL1ZU2e>BGycjc^Mm;3_G!rRC$f1s4
zDYB`=X1Mk9!1}>DL3Wxo!UScWmqJ{?qp<Iik#iOBFwXTzR<WHbg!CX(?-uYLAVR?A
zL+dTK1}btr2tIBR%Mp@1fO|%A`ZdY-lsp1jS2+3@rJnQq87Pdp5&%N*1U>xNg?|*p
zP>f(48KgeT8D=@}gESJ1v&OkI=qdp1y@ild-ci9^y!)n?qo)nH7WCWH!@wiB*7siE
z8u&51$CehZnR9Rr(2Z0=&ewplTh9HU>g)mZRUo{kan`Uy`GV243+%Zta&T>mcS_?V
zkA8i-t8!M5))TJ<dL)T9v6WNrbAKADFzl=uXAT7c+}A(FI-%5xp!SneV5RlLC}-i|
z7MKxi<oD7d6dcbc*5H^m1FXRQUV7n<gLM;c2ed33i}$k{cOh?+rVsh$q!{_W&#o(F
z7ph=V%GGn)Huu;t-u{kMJs;1Cc(#(69RZJn?3Q7}6`O!HZI`EDlQj=Fwtv)gH!!u-
zb;+Tyacj<Uf`d$V2Frc4G&aM}B&FKwTgXf6R6n=KV;Llj@;JERMIey(d>VC5=L~C%
zchl&DYLZo)PtiCkLL-v*K_lhEZ6ZC+^$0ZTZ(nFAXR~2H<Xa1W`ghqumOFO(5V};&
z4K)?OJE6;dy@x@knr6=T5q~z<<4FX+KP=~*3Lu08(@iAzMSvJEI!vSLHBZ}Kf2=jp
zYJ?j#(^{aob;lvah)Rq$3mQ5?Im!uZ;VjDOdN28`oQz+ZM~YG$inJf8RFo6KODVh4
zN}D-AI~)YOUgP2DDmaH^hh}G8u(B>VyDsKB%sR{-jrM;0rI0}f4S$mz#$|LVnvV>g
z^O50Z&q(r2SXRVHPtq#q<1w5%c9c#D=6+)R0B(7?u5+hY*U~benD*`j><T3bX(yeM
zw_;6Q0XkU|wYF6+YnQW!)MvUJ(|CmDwhIG=_I7c(<wr|eI+;GQ5rBELj+-3bibGS)
znD)-I-5&Fam;$}$WPbxE?hfL+3^#db9JS`$6S`}XRBDcp^1)|1{5-%RB+$XM&I9aY
zkqu1Gqt4m2^Q@Y)Yz}JUoT^5Q-;X*9Lm>H+&W53`_Y#Jz9Q7r(%*pmK&hX*vTaA>y
z4|Of+I>1`Wj#)nr&{eps0kkHBgE*$p3mE|PAJ5c&pW>m8Dt}_m#_M7u=PSn*dE+;J
zwz83Ec_(~dXlsAlvgwWPyScLU+m2fc{_OmVAlLqet7|P~FS%T!Xnt4J9rv&J-;Ezp
zr27l2=et%51F(I(+p(90llFS;#5_91`tzUsGnVUOu<6c0`y;G^CgYqF<E8dzmzwQv
z0I-HrH9atOa3>*a+cHr|80XR*cw5yY+jyz;Tc9j}YhP&?uBw&Kw~wd)0}6#Y<(JXt
zEEAWY2sad$8bd6@0q(cnLo9U*1QZw@9+%PQEE57TGnYZ|EE54am!WSvCzr?yISrSz
zPb`anm#@RqAMd|D$UIpoya|ut*8h8ZfB5^MPF;f_<0m9!x%~9!_lNKT6$xYgq%{de
zX8Q5h^!4Glo@`|%c})5tIsN!HbuC{qdR{Vm(iox!J#7S$n~d=S!mETHzDh|f86Np!
z>5*~Ls*}=K=Ad9wfmdYQv@qjON2gA5B_jZTe9u+$I+0J@h-S`6c_yxzL8azV7(K`!
zdG&dqBwkUpVRWJ*k57+AhV>@yA(@AIOLQ-Q@<SZa3Iuy^99yn?acrqwfaBrzII`}O
z^7qEE<+>Ngmf8h4vXoFT!~j<LiPAJJUuJ$9tWXwuehNE$uAqoYptb|W*Bunur!n|{
zJw<a8L5*H`FoPDz+v}MGVWhsdp0!+qo}o#aLA=jUc(z&RFwG>khqe{Ocozh(d&*~Y
zZh;_W&}IZn%NYc#j&ul8=JH(T-OP6t84K*Y%|u=cgHWi^%hlGPoQm;4M@fhMTW;3h
zV$&Vf&c+65crM0amFBQrnwFB1E1pMx3a~iveZHQrcfvUs_#Stms6&i~11^frqiChD
zy6)@II$Y9g)Y&R@e$EyYG!Wi`243$?HFAsZx$co#J7Em=6!-XN(84P>P~734^B&Su
zT=UP#_PgHEKX<!X_M`Ipb+9DR=$?msJ3S)@!do@y%7*o43r9CD|1K2(NkSuke5eB3
z-ErAkh`g}I)FpA*Anavsm_umA&b*^?<4u=X*&Jhh`KJ5PhT2gim;<)5ajtHKMMVgv
z`6tTNv@D%tk%RBf&EcZYR53ZJC`$<&vux2>_37dU&YU9px(h+}h>_WAo<xixsx3wz
zxt%RY;S{Td#Asz3JaNdh*$d=<`LL(rC!H)bKe1((w%2})P98Wv4Q6w_B*^=@RC<#n
zv_TU3d6KZ_lZ2ZjF&iW?KTi^?h4i$D{Ss|w)h)Rg#i6>HBdHi4jq)BU_gO|dSLd1M
z<Rmz5=GeuVkR5pTsA+De-pu*JSlcQEB90)yw%V%mI#aUsE*H%%c(q!8h8(pk?V1e4
zrHV*(o-T9Bt}JO!<gz;7OdT8wyQsxw2hW6KEl#No99OD;jvb-zIX)jll}_2uiD%5A
z0$qC2(HrWg_)g_b=GuB4g!B|~tk-c+h98vc`uhdhWGvsU?0<PbV1*J|tBuRK7I=45
z!wy<YA<uIvx*?6~C<ldqNY~lqwx7@8)bdpWSW%d=hI~hwOJ+{NOIDM(U%NSS2zz&M
z2Ov#OaN}lMcJ-8$^IHmUvGzO8@kfmcsJxBn&;Kyf#7r+I{*im&V=7gbRCiw4q|z&^
zTV|E#90G#eQnm$GW*_IF@<?0x<r!gZ$)QB~H$+r*<RQP6`Hh`_qdQu}UyQ6moviT@
zmH(XT?aMj{E})T-LFm<pZi_d3inaYTI;zL8@SKCvT(5}qUm~#CV+4pz2eCc}aaxe)
zOy`i5=%t)9#D7@&x`yYHLaLn9Q40((`}vB6ZF^O^x}#0YIb2p<sz&(f$bRLF0ckd6
z^kz1ogjkeRCghfXHPUN+EN3Ax1L=HhsE2^Q%aMp?FBp>C+hr*o#S^x(wLH{O*5}JZ
zU2iy*72W=5L{l)ldzbV(t~*Qm9kt7jn0A!(JAOAP>Gu@(m-Ksz?rcfF*Mi%Z^tJYa
zyt=oZwOse=SxfC-If4sOm$7hf99yn?acrqwb_92097BqKV()&m-f`WT5$KREg6*kZ
zfTzjf2+ll97?%_6MZqFw7+XrVhLb-WX7_qhmQsuDdM8iam8-YcV<D`*qG9!Yjr}i)
z8~`~Qf?=)|qaE`jEhOrZ@}L8kti<a44GJd*6n5-mV7y?DFi_mSk3q<G)u25q`xscu
z@#$dI^02FalB_HWI|tK67u%xgT%1x=t>@)bfpO8(3_Z<CPdoIqD?Q!N)6F(>MG*d>
z=VTpq_TCX=$My2V`SYH$rMlH~Uhf-?5uTN<=f84{mJi*!-dm2*@_}5B=DuV2>gijm
zYqlymGjzj>?y*yjqP_l)o9Ve)w_1$8xjF05{odby+H_oRWzH?t`<rve{{wSwd(-Q7
zb8ac^{T`&4b4N9Nl?&&Cw@&^YCM`LjSblNzKe6f86X#bWuG@;|)(AcOQ<zJrXAg(K
z3y0m<Y6|}RF%OX#_?BZGEG1H?pU6FGr58oiWRj(FUf$Ny>WMMoA9Us6#V9Ow?b`1-
zw;n2P9jqnjHAH}&Nl@litP4}#HK+hNFm<*g8XS=ZDDz5RgcOOgjL{{UrBn@mhXv9z
zT8rX#9zzXUZpJxrga%_hU<_x<)l@%h7&6vgMLF>RJ)er}sA8p7-_-ow)iwWIC+|{@
ze1Cg?_y=z$%l4Pi=PVPK!A?dKmw=@}1eY?HJ13WQS1g-<`LYjxz5agE#aOMPOFz?1
z{&#sj{d*crH&&60or9<h+V1afr}O}Ul=IxImFKa3|EK$M`a?<`bZLI2|3kX_{<RY=
zpNsRNX6I?=h#Kb^r-(x5TvRqa>YazrS`!ZrU-;s|BfH6f7^L&80c;0JS61A#u;Py|
z0T#8AQ`Sa*snzf_A!{X`Hsaqz<$u>-$r9q}q1KWr)o9Dgy_BkxG~K_6zL%KBrC%OC
zIGv_l>a>@ntJk40`LX`Vqik9}PxL#dBH7ufwD8*oiJlZXBP9YJkQkMRnUCOkzr<iw
z^e~g?mggu0WsIBQO*iwg-S<{Ug>i;s{k&)euYX2==*)_Yv(Td@qvok#;W=4YY^v)@
z52$ZiGU}Jtnq}0RI3If~sHVo&wY6itRTgl*VvpMwKm{|yZ(en<86T_mM^qHt9WiZV
zPyz1<Y2-knBi#sT<osiC7R>ZHU1d@@DvW=!?xtm;dkDxBlx46Cy7|&$@WKMqeCG$`
z>*C>m+|>6<VD=SSd|monnBbN!BZ3}xpZjcs-Bzmr9IDVO<RDy^R9QkP&+-x~HtI|I
zWv0)mJ~zwH-M9pgSZAF=9|o_|OwWp)3q4mXUg&uz+YKp#I6>pBw!52ccQ@I-#6wou
zS)n)-+N6jDx2D|BW0KAxII%47H7K7e%6lq*LQj>$<>2Io|FB;j`5jYRhp=wTRaZnM
z1AavQR^~?Flr!NcGB;w@;c)|TdyA9_flZj$oR`6S<p6}x86|p~N)7^pQ@j|2yo1Ei
z8Wlmh@@?MO@WlOjw{kM!`1__^6OBSS#3Mv9vLIgaLg4<S%hRS}knmf{xmHsyn1suJ
z1{P^{vdFZifFuxBNTemjaZGgn(Hyd{KQZu)6f$aF>1Gx>j$79=u^$R#Ol7{t`yn~S
z>@W+YN9d#B$Tf>mYwXdWvQ94kgXEEjln?OBkw-%03dH<R$s<{-g0sIQc_hOrPUHWS
zJhIh6+rZ6SBXe!Y=|5^yMz-{1j1Q}SgcHLoELUzxJ-{)DXqkiB>i#0X!r+YCIY(xw
zNz<yyRReO4wgbRS;c_fI?_dJ#)VjjVISY6$_gk%xF9Y*VdNZSTIkT7q3})5&8Nfci
zjGVZ^&vrT!IytLd=gabb18dCud4=wlSefz((t%l(5xr90R`3nGkXY0`j-t1J5Hb^#
zb12j%%=FhX^F{7#ytMVaG`Tkxde$>F3zI^x2TUHbZwfm-mMXavJNGW{Gt^Mhh5r3X
z4PwlvtCLdY`=$t-`F>h7rVLp66*+x?=;vorYQoSj$$6@l)<QMQ(vAJGw7SxNEJddJ
z-GbbH4h)50(roh;Y>oG`)+-TzF>O#84ux`N_7zQQ^N$XccR^~Ql^_zvATmP`d5-o2
z3$Rm9TeCKo<C8iDUt5uyw;Or(6WFK9a@x1@dMxOD99jLynhH8B)!qp=UfQyX?_LYz
z+ik_S&GDk*4VK55z3wMo^8I)x_}gOVOI-ogw&DH9f*l;USc<;I*+o)+HhNJVlr@BS
zJyBxjH&R3aZknVjt`Dh`=vRxvFig|6+-NlBeZnP8F9m$b{g-tR8?VsOB8Z3Zx3Dnk
z)6$ad^_91jP}qbycG)C!)O{`yMBM~HWe;tb%zA&E2F}seuotAQtTxC#mUo9y4x`Cn
zElUz6`*Tp{i!=7M<f9{h17rvtwM?mDdE%K}>;-sI4Ul9n;F+E}c$e)y!?ZSnzvF(N
zkhl<BH*fl-gCxejkSuonraaVgdJ?_u5?OlcOJlf9w<@l>^i6*{R58YPKKK?|Q!1u7
zp@2nmjW?oF89j%2^I1rHn`+Vz@FQBK$1R8V@?)#@s2^KtH~8^?e1AV0P%v&Xyq6za
ztw;UXO1r_2&uPP8d&y&cE+&P(BfHwA9+DCdd>+9J!UdQ~q?LLss%YH|1^VtdgP1-!
zqC4ghDG|5l4US_0y_B?;H_}!(IoW#$j%vj&B!OSnTe6$zt#{<flKG_G>#IxpS>G2I
z><eqcJI5cP^$xp#FeZJK5sj35B^Dh%#w)QydT?Hy+_b7Hx*dDS#fF`wc`$PiP4oNq
zSjt>BN^J=H<6b%h;UNi$BJS<k;1Uw`W#vq_&kfRlJ3i~H^Ljtul$>zgH!~y<DTG^q
zklNCI{@(8x4Hv&U3WY~LV=U+Et-vE`x=wB@@cD4w7XsIRl60BArn5MF9^5}U_TB!O
zrm#m}{iad?<axn@{n2fR)X3l<Xg&`a_fy`CYA!0!MoKww)CtxqjHZe2e0f|rE<C69
z=`3A4(#!`~di3LXKqij8{8VD|2CW>ah05rD($QFK64VVxWAc7^J^cqjjFE#1Wo~41
zbaG{3Z3<<V-RCR^0x>n0@#PH@0x&g~!LAA?mq=eMEq|KbweO@n{B@9ruMem9hc9PE
z2P>Qgi--5`2jB(CpCI|B(FY44jDo}akHc?IQljDO#1~1a6knEqJO0+kM+J_b0L>S#
z0vsPLL3oFsQJCIUCcUHhDSY8?;P2;fpWpY=$N+)ife%oQzrX+TATzOF&R%X&YAgwc
zJyFWkX@AB_jEP?PVyMK}HDf2ntbMo_F?=~Y*o3ULFf_}6q3OW0Ofmqt65EQz9SAh8
z!}fd*k^%da@DXLW`iMcc8jo>}r5$A@yb0n8Utha_Z|@I(KTOC_32pqrfD?+j|KrC)
zSRi2B)*lds2Ri)x^YH!Q*B)xt$MDT>@-v_gAb+Gh{0uKNcfMtVi^7(M`E1xnBeZ7r
zk%Zyz>(y8YnIIZ3Kw=M3viUNw1l4AU=PW+rSErB2LUZR^1>G}L9O#ADa4ph<Qn4)T
zz~OVEK|}u5iL=0^8tCL3=i{T1QTb+mv;0o0$gAX&nMSz0DqogNWOA+*qs(@qofG6$
zp?_Hw%$AMQ1OtrvI7MCX9rtvR*2<FuAJdABCcfpQY#>6B=_y=*x#fJQjc~RH61**n
zZ)US$>i8Jcu*{fUP{H&RhI>~qskCPg8tH4`Z!)N?WtSer6r!}Y>M2F#`ztWKlcYG^
z(apNc@Fwgrwr$Bmb8ylY29uN-5^TyrD}N)b#{O-2-p&NY4J-#K<r<uUyVG`LSC(=1
zFYlow6qaxaWlW|oDe;8=C6N*)&x{Ef%I|Hvmn6VnOB+g#tj=6STPPGxx6x)pyc}qY
z&a-VI=#(%<?a_xm(GzJj=|1Md1kYKnp;&q8=|95LMJy=CiBt_rgtwB@i?_r$l7G{f
zKsHVc35PGIv`UF+xclwvssM=9st|S+wCmwjsBZ)5Ug99<1q}q`GG%Zp?Szqf%7`yH
zU}!z-JBq91L9M6N6g<nnIxlI{=cV?UI#{J=Rb5!|%5%=flq-hf?sm@5u^>CAOTVQ2
zs|-D*Cm`qa@SRp71<zHmgYCditbcgay@amJZps+0IIxEj)ohAfh{`W~O$MoR#SwX(
zR-<81Ri{ukqH`n5%wyD^Y2@!`sYMLaJ1sq}=FED%vuIHiIBr4)WKb~WabJ3t232#X
zm;<@}Zh<XcIoF}Sqsy(wNh7>fJ<o%fvvB9QQ`A$`Yzy~Oj!v?0v$%r;LVqz?u3@)w
zB|2eY?#BX@Q{_x*GMUZXS)hRw*3DwIM=tN5vhOFBrUD>;m;bLzimgmx$sWO7VkFz1
zu$iGOa9lJ!TRVkB&@_g;*1i=Z%M>4zL|iFf&GgS_j!Lw`<fTEirDM6aeIzh~`^2t$
z<fO-I+s$KW2=j-|omO+FG=Gb+=U;6WiK{ragDIfB_Cbs74!V4`(tdISpp@?+-lI|E
zzCKJ5I>;!On^w>5jHUkR#YXTHnz~p`B%1a`x+_mk6v&C<a(Q0T_3z5O#Tr}web*W#
zmm2a8+H$+!w9%)kIp(@*{)pT(cWS*BnQ1MsYCKP!eD0tXrPwC)XMby@#rly!jCaC0
z)kECZjH*KVJlOcCzPNwnxV5UUTgD^GH8!`FohYZkPsvV1piWd5KB?hOm2{w^)Vugp
z%7dKf=MlwbZOUCB!oYkO0JBiDvOrG1XuC$`CRuZX(?R?1a+~6EisU#NW!kx6{*3IV
z47ah$e6rtnSnuHX9e>0N{XRR3yguWi4nvI|E$HQ`EcZl2x_`Qag=^4q;_-T57i6cR
zD<E1e$lw5xmN~IL7TmqKZx5A9b0ue8CBZ2Kuln}~11T4j8q`)Gty}yQ7-$NDls$xU
zxX^vW-UQFR+2$N*pze-Z3RYT!i{UShBht#f#t~ikw4V|_k$)8TD5ejehC3Hl{bz(v
zF3LDmyE`@q9mJI>%q`792YIJC=$^iC27F5dz&OD4mt!hBU}@43P1r6}JQOFW?S9DO
z;3H=QoWqQ|dwq?Db0fkh=h+VH9rA1k@q*50a|7`MDX^C)d>{paKso5BOXk->Olh)*
z_<=Ody*i@Vp?`6D+H?OgFm!{k*<gfsgCWVcn;8ieJ^4}C%k_o=6}QrQ{I}N|sM&7~
z&N7kGQZd%5(pacNnColm5T?LM<@mPJhO-pAua}sqTw2-KY?(414B4TKS7X{<+C14l
z(gM!1hKI8dadhh1^5%kFu?ed2LVLXAILj`km1-{R`hO5~nEvIXETE9%>i_8|3#hQ&
za;^HtV=N>b*&f;j(R~wFb5x{C(d_{e^;X?+frI6{meLBzbbaWs2}q8RG&KQ8)#e}9
z+~`zSbEAtJMo{haoU6?MS7wRRleaq&DiDfeeMjR5oF!y9Wo-|!Im@uN&CSL1D6f^7
z_3Ge78h>sVM<~zr$&Bqm5NZzSFmNzg)i!<t;c1X))>zW~a}J}js#k0ZvB3_}%dS&P
zdalFEWCga^jLN;^Y&{Y7^Zj<=b*Re*&8l5Hor~uS;>4aeanC@mIXy9!m{VKWw2<Xk
zLKNv{X5n}r^7e@8R-XV9C&ev5D(3>gE>Ap7$z--ii2KF$Y!hleW3g*m5?9V9Sman<
z?K}1Z+MP3A+hWrKxzuP;kl#5b%En}%f5|P3iYIGtc77LU6&re5pM%JGZ@8L;RQ?vB
zWQD{vSeV(E&g_!kd#Fl*E<JPl{|BwN4*r+X=PVNeF_)nVKNJEuH<!Ub0x6fCXDpb1
zL`wTVUw_+4X0~z$S{NG!>rfD_!daw>#2Nbi0|cNuhR!s^qlr(B{F}~0;to$`71Bw@
z^Ap>zP8@tVd9{QQNG}5VjuQ_$Qste?=mJP#gY9L&$VVs~X1?OziKqC?R@pjfl|Dp%
zO<WV7R_V%4!%~F_#qyh;%Y=t}<fjpT3wQ3Uh|T^XYW9bsi#iC2OW-=Z3NRWYj56B+
zoPmvNk{D1yid@W0(}EyY>!~*?L0tx<Qb!aD!<~qbPYOWW?Uuw+=H(b2I2nM+QfGHg
zDmHsi#i2sxxyTgFB=4)3oC>wLKdlXy-AEf@om!IXnj4`L7W$nTQ(NUJy0+PWaBr1s
zq7UOQ>I19*H3p($?o&;(=0{s2KG3sOnPc9Op2QW0%bG0bQuqOnH8fvGT4_Z%nSLi0
zJCyqeD^tjp@ru8q<tx0*kEJkZp{E){3R_0?rR4NQ0}~J9JCfJ%lc<+$Uj?az;nNss
z;>r|WSd+zvM~fsxiY2nINB_ltL%n-BKov@oS8;IVy@vhvc{x#$U3-Uv@md6D8JB}>
z)7n{Gp%Lg^w>$AmH0aEH#k41n)89ph=4paD`*hZ)dCB|KiT9WZI&{y<j>9v0FY=N8
zrN*`LHM`mC|9dhTis(VFm9g19iP-NZ=d~DSzzOF~jbUu>DQ>!Vh-F5926_l9Vp+Fy
zA(p9|;}{CcMLos;;~4t@da9{#t)I8MtL>Z4{*@Z9f7$2J24Uq|pKo_JaV=`T$Q2tY
ztYte(grg31m{grERWJzzGm1-3l94hzw*sozIEwkQ<Kk0=a}j9u1AsPKP|3U&=w|mU
zxWeHMu;s)ZO#(W>a`*{<z`hn?5rWW0*9hO7w};fM#R#TNAJfvE6s-ze99z(rSYP+e
ztTj=;T>VRd3VQwsnTwUbrhGM*YTCSJj-beBCy^I9CT5-+r5BA~<XI~#!klNgoa;1D
zesPN|cM8bed{l0_{8XS~I?=*n6O&eybda0K2XsmFu+IW!CD49<TvY9Oh&cISoXVJz
zO#BqMCYf}cw;V5+)>7+?$_X8OZBNzMSSE>|N0U{LrJ+l0-icQ})fHAh@-EkvEs71=
zkN72oG9^EjFm+c=$bR1HDooWy1Rmm%+K&0Z=3(}B&bZiA5J0f8fYrB<S?tCtoX0(P
zuzCw&vyt&gM|OUHUF(K%N>Hkh7$WWCXufP{fzi07?$jX&w6*Sxqr@0f(^*pk&&DFD
z&x~0ietqnIy}a)Jo>`MaVZ9#y`nHSrQN=+69jZWh^2G0-hcCO|d#apvq>!{He;7G@
z|2lMcKJolCND@$5YfLMk6^3Nj&Wf?F#^;G1SU;ea(%qbYps_;h&=E`^!DMW4VxTZ`
zSw0f;&7Dt^^3JO?>1_#vT&@L!6q-_YyI9Y-$hkMAz!&D|I-FOrG$?|GJHV8rZz8Ys
zpyEO4)aK5o8vt)ljrGM+E0l`piK)(jBU6N#45<WOLa!+?<dHF?v@w1pX_`AztrpBM
zjt^?t&57-Q9wWc16`<zU*ey-GqgY#o`)!+wa3!Iqn9H)>QbLg%3|q?d&Lt$A6IQAI
z6bNHhDg{P*ui<V&17ysEGv2E$+>QcJ+?%0##=(J#6qK{oE3gOfRF~sSo4tqLK~V#h
zT{Da8-V?3z5=GG<jIRkoDa%W{R#M`0jE}bX=(Xm5ds5;bO9eC;N|fh|V}nuS8~QAG
z@(UL>^x2kT)9$YETc@c;wmqL)`z>d_?v%IsZG2;jGv3$uEl{E|bN=Ld4`QF}aW!x7
zNk^3QOxop95>U@>3AS!^wUGNg7n~OW-8&<8%j?byKc(|Vmy@ui^EMP`Ba~GRoKs@Q
z>*fxBYz4L9-63!G-<T~snyb7wXVRWK`R4rBgKBu_q~DweqkmGxxbSNE7k4S@7ZL0`
zcNX2Lsk4A-Q$HC@jf1pvokd6OhR&j#1?=kg_l`i_jG76;?^$NS-*|y&W{jSuWmNtJ
z(-NY&Q{&S5mI$c*M?1c+SM5NjPHX(w<mrolYGJnsX<%I0O~G!4@%FYGTXKWpoo%-}
zwb*Xzf}h(V@c1t-XYAJt?{!b{y8WhU{c)UAa@6P%Ka?K&O-4i2iZElra`uqz)oUn|
zx-GMPM{zF8>TKUpK9iXGSyH#_ElNTwoT{UB<<2C=4YWeZ-GwdZdL6|j)7LMnhkG4C
z^>c4}b3N6w(isTK^$Jp2c>T+S-k_)++p>#p32tlyE%D1H&@nQzfO*JUVGTZw-0uyk
zi^KZv`{i}_4`m4y{+H3`EEAXUIvy04b{a4$mppGQmI3Uy({C(o27h{uKM;z&nf(%w
z49bqh(xwEO)g_j+7Fc+<YHvT8tWfe5EqQ29*+XP?gPgXaH(*d`CmrG$Pa&VXkv)u%
zjMPs!-DsbB*^S+`g*;K8uSCOz92vIdWamEHJ<on_6&mP^6gL>z(g1Fw2Hg~sgCyl<
zP5oCxXEBFn^TX`BmM^9TQgg00op85pdr;u*owPNZA5I&bKS-O6Wi#JNTeJD`a?=~M
z)%V-`(?5O(mzNPfK@$TuF*lb{2?7%VIF|u?JtvnwcPuOdF~^r|cPu!6xzi&#N2JHl
z-<Cer?FEIU9zxMWZ-uzM7%4mr^+AZVs3Q$hFL=<WpwCy|PV|mz?nj{UgrmpY{`L9y
zUpz6JKMRMbm^JdEjbt=t;_%K@L-ogp6|3?|#iTVIH3yPO(cF?GP{gBr#J{Eg`-2|h
zjppF=l`E1-n%L&w@sHboi{@Tpf&MS>e^EQ|{7v20+r7T{Bk;Ys9)L_~chQP7tz_mm
z@F_eWUjODAIrW?=o8Qo<&Z{GJrYS8+0)^?bjOO2lkxF0X``&%OLj71~3A)m06kU%X
zUFL3O1f6F|Z)G*AAE_XVXoL=0jNH;2fNP#FK7lfiM&HF-gN6ZrJ75@`hE7fVUC7~{
zatl4QPu}CmG(g;CZ}fP%c^RRh@qG)jl$|1rBeS;1gLx$*evV*MS<S!}TY-<8K*%7n
z1)w7YX<<#Zut8lJOpgQ$YEzE_BfS7N_aczRcaLqbsUlPjX$9rmYof@r-?yPZU+f~A
zT9!EjQ@$Por_#WGtUHU&min9ExfrmxMe{crJev52H~jIlY-k&2%(I6P@7_qpzW%e?
z0J3U*95#yz9aEy1MQ}G%8F{MUftMB#*St!mHe;^zC=@|srZz2QymjOFXw08S>Z3Jp
z9_L=-)T3~d^plP~2FhBva3<-osPOV%TKTc{9}9B8IwqaBpm{7#0e^a+@$_s)&D`i2
z!3?UNqwn#bp3n>uADj<-M68=up3z60qJ%<wLrvYtvf?dEspdRNw2TJm2YQdt4%_n)
zEp~RQV#L|IUD*aME>Jo7H|@Ier>;J0*Bk1}T|H*kTWSwUI_;mZ>(Cl2>s#P}T~ED|
zU0=R+qf|joYDqD8{C_4p=iHy7#Xv_h`2vV5=Cq=3`UQtQJyk1@*`TgPHZ7RH$?Nb=
zt<OQ)wsysYChopyLakx~<$#bMFriYAu@)O9v|6L1CbS*zcQBzXN8e;Zk87rw(3X$y
zG@<Rt+dFG+MkO*-!wUX2b`&&&5^p_iM;q#pGYXvYD)MM-;C~#+)-pnCKWg-weN<Ci
z$D9n}GDj0`2~Y!;dXG5tEa@Ga0#=TD0c(gdN>5ov@-#9cXozM)-#T@W<@f26M4z*w
zv2WQGw-<5U3eXdBxlTPq3dFP3jc+;`=kB5+6c73iluM3vid<_L_Z~nuLQa81>58nf
zk7lA)tE5YifPYW9F4z;&M=QmCR+go>FbRV`YnrpA!S(uRzstEZ58&PEaYm^0vtbEQ
zG9l%&y!IL1%(HwO+jn?JAQx@a)?x!=UST(R)24{zklxzf@ww;Le&}uq_EW1Vc{;ai
z=q$5ZoV05#b-d~FH($s6!;ae9F9?<#r2k7F^b~`JTz~TH(n(My%1xEo(QSG%_D+7{
z6rUQHhC$q~nQ^4Uk({@F)V17t66-+5vAE7beR5IciXqOq-K2&M;3fL(8ppmn4tQ|e
z^%Dw~PBX*hH*2`S)N1Cqo?HeV(X3v-3yC)u<=s{)(yiXF@FYh7slYo_Y0m719!u?_
zrcYSn@PGHX1xBYR$t}(IrF4<>?1WyuRj~EV!pv1O#Z}YKCu!x{ZN=iYro+=VY2ZhF
zN0C#+VT6z24o+7Fb)HB?0HzSn8q|h5Vyl{8LVpk*aumB!y`fbK$ID9!82eBeZE#K6
zIdd!4T2JWpN0k-5_D!M5dne3P9Dh;S4OJhfg@32LRPG9&<p6xqVZ|XI;c;%=xl{c$
zIaU`4gMJnnR<t$ebuGxH*pq_O#qsI&9SS}w2Qak+{p#huFgwTT?OqV+ac;!53VZKW
zisN-jVRXG~iS?mGd+ar=j(;Q8_OEjNEwttbPZA;GK5R9*zmt;tTcv1rTt;bup>e}Z
z!+$E3Ff-qjYu=ZH6o=)IOAmBPYN)cnCFL`<2IRyzr+ez%!jo_ldqZCK#@tonj&Y;<
zS@L~uGs0FFVVJWxN;$Q=V37fwO?0-@OM)N=A10NM_`t#~mu%TiRb^P4QuzWgJ<Jr7
zOgO34(34sXbFJ$IlKr&JTU>D8SmWB0nSb8)&}vvN)-;q)4hicT3++`kL*i_HA0!y?
zel^=YRs5P}PaUZ>nc_ziJK)aZr=dPltaaqv?neC@R*n~uW!m7OVevv!GBo=PCoV_}
zri5dL;;Bx)E>R8nlx`ACUoMYpCEKX6Ab*zl>Q~i$tx`sBR?6&KHw8>0oJD6ty??oP
zm1;`{*3QUc1;Z`%EjthLBbPW;b$DsW1KS=9HA>wx8UXLa%JOD~=ebewKBw2XGrmXT
zx+&2+7z%#o_8%P&)ghg1!}b4;vxcMFzgH6A?;EYQ{mdoEuLJ3{PF2wPqT+VPn=IMm
zx_UHj$V2Xm^cpZ<hhMqdSn4+G<9|i(<4pN9!@yBnYFV}~Z$lgfbr@d`*W@m5ZgjC!
z`=(9ahtD7k<E}V52rK9<;j=t;(;=&%`fT`|3)pOR$Fxxpn(LaCuGTAK>$OTdnIR4G
zs2m5=S0*<Fr@807aZN?l|52Ibwz~RArgR?I3x~{{O_VmoZH-pO53JXu&wt&{V1V~L
zPXarUcRVc4L|0^-cT*`zO?VdF8o9nEZO@Su$TWpXEgP{?Hf(CqY;3<Ub1w-cZD!r;
zwz{?KlN&n)%9V?{o^3x{+B~dwojM%Wd+rGMTWr$gSr|6dabE1~*w5pzW#az_#7-Ia
z^`8a=NE}3BJL2yL#c{#E0)IuHPY`D#?v{F^s;c)li^Dg?83zUe`Wb!%XI0k!goA|K
zk{hCWP7jvP`mC;2x~JYn-ONe;nxShd`Zwz2>RR|^#;WvY4|UQoCC<j_E%mg->lE@d
zbLje6en;ey4Wxb2b9%7A&*y~D_G|I`AaA?8=*#FsU;@s@;VpHerGJdJ;TJ+l-r-g<
zeUI6g7>6G`ep@Df6rA=sP@;zw3~Ji=F;x}4ABl#`5(+x3L7iQ~hI%)ykg%F?+BG2U
zELGyquBCBhV_M-Y_6tzL)gKRT)<i5<W~nCf8<Ks9jNsbK1a{;_nmIl+jU)!`?AJ>k
z%a+h#Oyl&U{{>0Ey??8Hif>SgUU6g5?xDMSqe9K;=C7~X-x8%Oh2i%^^UOh5&s=Qs
zXx-W+1JN7{GVKw)?(iO}J?v|)_A#-EG;-+C+D=*Jc5dTTR2a%qJEHb_ZSC@Yg9l5_
zAwJ84Q%viDhar1no{US&|5|`Uo+N+1fByO(3zbO`3T19&bC=fVEDixNm9Zrgm;Q?^
z#ee?%$J?(@IuBN>VB^<#tABldd;0rn7&@S08#;?xfrjqS?@#do8CKSXj)L+AyB~jb
zU!Q)fQ3qLCUg<wrcR#*$Rm+#Gyr4yS+FItB^1v!)-&q?J#z(#K=v8Z$$<akFnLLS_
zz|1E3@#4)@M79bs)QF-*&7xWol~O|P(SH=hah8uX{8@_(e?ayo8M;CSJsa21J7|TD
z>fYFlLmI?Qw9L|c&SSc*(wq7urwH8&eQGhez>B?26qKaSTE5(&^SM-z$#UqWF^`go
z^GN)H=15s+<`A{S*R*$e{$O=nK73kjQ$VMT!)D~BvFj>SSvdi1jcV}fe1S=sMt^zw
zoF~KBjUP$utYzw;#rS9yIX}4m={<JX${8#+k97C%aRvug&eMUMQ6V1485Qyk&Zv4n
zh%?xsjMYAvGb*V&Iio_mhcn7A_|!j(efg#Dn??Y=^wb{POvGBxGyxPf1KdF}q9_%7
zqre4{LCp#X8wG7=8MdH;c0ZS8$bX{}fA}|920($xI*?`Nh)1%_9QhW@%zHnIWyEbE
z9GU4=h#Rs1V@Y{(HBz%UZ?oNiD9a4lHi+BU4Pt3Q4ccvXOSn-0vp`J0md#Hg^B47R
z7I|xU%zFi3Z|cfnubK3b5N9njX0-B*n%Qw?XI>z*peQu{G@1L)<8qXoKC;p+9jt#q
zbWZrLnavWJ$Mp`_%zZnVlb0$5?J6=~EQkwtv#f!EU82ULkDiJ(Vj1i{B_A|4N8;j_
zjAcFazH46d%vOdEh*8fs;R7T=HijP(KS*k8M@MF<72<*Mqe8wBKdRmjiXWuC`2M4_
z)Jp2kEVV+rFH6PGQO<h4CZh;kGlPFU)vc@|QCEd%Xm`C_Bff4&>}%>8&@L3<UJ-HQ
zT6(>*tyMu7ww8NYTuk7Ca;7m)bSPaGvJhEbBEZ>wr!DU+np||$wm<igkKJ9oQ%Z;0
zYlo^SGIZkhyc&2zM8%-Mg&STv^d=|uaig>}j$sV_x=M;XKPWGtxSNCrE;WCOORc3J
zb6qVI%c6i+eQA4YfFbU0kDFLA9P5#!k_z+h>{MCN7;>-d9!a@T$)EGHMe^)-z0<qW
zPhM}SV8uKnOnC#6G|Af3lU=c%QsWLh>HX-ARiMzI^D*dBpoA9OUIE`9VUeWfcKF~3
zTS?s+VJoydBJ6p0gso?=Kf-@jhzCU28gX}otpVSUu(fOH88O?;5oQc0ltWc68CHiC
zgUT`7FwM|AZiv`M?cgTtmJ8{4s>eCJ#5O2lrxe_?kW=2;5Ew~2g%s|bK`d~5yM$wE
zp3W85J2!!s1=YHuxh@|;dDplZzK)?bS{PxpYAF%+L@1GJ42{bqM<{=n?V~IEoku@Q
zJ~r5CtEJaQH1Q0R2dXbond4>k;508;XsVir9qO_m%1}?!EUb63dCj~f?1=f;$#Mmw
z%;co=rb5fu`<a!ILA47N=Q(u&%*jD?g&5O{D{L8gaS0>ACyXf=4LMyQo06~Q`{WFO
z%D4Mc9r*91IuInA9*=(#OKdwR)k!;mljFtnp{WiyAv_Zr^q-UK0J9LDd2p^%N!^+2
zRA_hPI<LEPoq7iQbDawDfLy0W+@0&xfbZuzwQK3?8A0BZFT}wUihns|!@iW(w@g-6
zjUL>3*7D9`D&=@R<51#jneTnu4S93DcI)l=0T78r@PA*y0-}GS$sZmdQ{&4q#4Fh^
z12^cWW6+IJJo$4$K|fzXzrGtUyo_F&k*&9B4@^UreeKCe!;1$LMrwA(+3r!<|KZ-~
zjmxG5u=p21Bg|uI#c5xZnj|a^wIHKWFm5kEj#DYI$3CN4C2is4#)}l|D`_tmTh3~t
zyo6~}IR=T!;5mOiD<#3&Ga89v+GBG>11d;L|NaS0eZM0#rxTln=$z?7bSbVBHdwKV
z&_ZQKe6d2j<J`8$dKt0T#4FG#Z?EKzb;k9;g;D3t{0HLQat3HgA&dt~)VoTWbShJd
zV<mNDHE=02hC{&ekJ&rSP<ke|9)7h)WaaRYdcM#7O0-eU(OF;0lRhR-(ffnUp~cQ(
zba^XG3ZK)wP|7{19(Bg;RW7-NhiG28JQl8PTHoP14C2(m)`nP@aV}YzDE{TS2ImEm
z@a?`qxeGD;1FBdUm$5r76az3fIF}K+F)o+0A`K{)pl2xsf5#1d*RL3$FKL0yo#Bwf
zVG$T`;yO)`0EXoPZ4IHYq*dZn)(X<9aQ}S|XT8#Hc4h6CYx%)4@4dV;91hPJl6PxT
z4k^^65=!&`YiRvChfD;aQ6TXrt%XwA9~o3#aK)@HaA=ZUkkA?*NQ1Ib^*c3KZ9yz!
zz?6OH=92i>e@2)j`%oq!7wS(961!w6WY*#mhg?(F!@HH~pyq_9sxT}k4Y_rLtcGm4
zzo7vGy8+e$QMWYtN_@7Q3M5vS*jHyRNuXM?OBIrq1H5a2w7-$@w26<BanGc>9vqS~
zxde~&*)kSbN}mnuEz+Oj4;&{hl~7cDHn^?8C0Iwse<;RSv@zZ=a|<vfr-F54lEB&g
zU9!NK&ZP{g_#D_EX5W)cC2%$O25I2d*Fg<Z#cpIcmMyt`gprM155tRZ93%~r&Bw!Q
zLVb=RLAdz5r9#3k5W6I~-y=xk{$ohh;4AD@5H(*L;z%0xIdGY(`@lCen_Iw`lKWG4
z30I%Ze_bcZ*9#w1>Avy+sm-7IMB3Enz-uk{Q82P~k7H9bcb^IeExErTkfuJx)6#v4
zEmQVIH~2tV`y4r_cm!t{sa1R(aHm4i@8PMc&rvm~^}PfvTFP#HO^`~L5cK{4l7+_I
z&x)svrDiLBuk}#?Eun@v5Vo!MeFM6hYR3_zf7We*vzv7bfK5x?0+?&lox!D*zB33j
zWAT-%LNd1x+h|zV!8hLw&t3zn1aQ9`{{D}D0&+EC(x3>Ri<|TFqbE;RYALi3>{24O
zZXL~H5y3F08HVr1=i{rx%Qq*ZQ2d23ym&L2jIJ&MO)vlQe)9d{<jrIh^7O{@@x=sl
ze>^X6nMbA1^4y66VK=>SHbB5mFF;7y5rK;*Plgv)<CDYD<n{34+4JGm=>6n_#jpN*
zIl|1pj9}2kWOOmPc7B<T_i}VSzPUOXT~7(gZ~Z(vef#Fy@%z^)bZ}wM=?L#%;Z4`l
zuJ!cdVmuwFO{YJ2-6_$R(J2wa(e3PAe>)7{j;~HfSAF1gH2ih=VfYRgSl$r~Ihni$
zlB1|&xZ!^+l%hF&RU(-9=wLX!IiC1<KfS&9_we*~^pnZk@x^dB{O#or{_m?_CzH$T
zKM%t%ZzsRr9LJOKyWzK+cb8wEj{Y<Jc=Pj%_}lq#d_F!N&dyr3lQU}VIOnrme@?la
zaT)1Ik5BcK=`5YGomn2PuTF+{Z%(ep*Le32KOs!(*Z3>-k68Jfn>So&ZX68|pRE+M
z9a+jFT4_n5U*oUp-AWw-?G6IK{0cO+WsI;$Y*yCM<QzTp9@fN8L;^vqp(z@zeG-H&
zfN}w@AaWcI^}_}Gh3<@9TgI+!e}!F~sJGZ{Gwj+b>@LLP>iEvni)nl#p8aGb)95E(
z_WzN-6tFK6h@;_Y!0?B|_v2UN;aT{q_}R}&h_l9`TF8O*dt7a4M5X9y&CCYPu~6H{
z1dk|t=nUCgG|ffE{h_sC%pm7njv3n`W<1TjjTx)6n9-Tgb|QY{RdJ3Ee+9P_JtpEW
z#0_*<NLLa6h5^cEk&#7l!s<Z|A}xBN$r%x=x)h>W3Q{O!;y8Cq*#kANfXN2O4{7w=
z|MvJPp6<wFI%sPZ9JJNa!tvu}&qt4+d4<^lAaHY4T0@gti*>CrQKLix$2_cwy|z}r
z7{18$_k7{y*cN<|?Upa{f7GsNr8pIGnhAL(&`c=1pF=1Hqk8>L90e(_P>Jf5aly9b
z6$mS+uc3fAMGo2&L906<%o7>4xm1>0Bx{ceVax>x<4SvAG+(B76+q)&XcOFSh)g1h
zD5IC$EeX?Rv%;uLaqwDcyC3_7Q<SQ=MsO~XYrWS94^gX4kuxT?e^yM(?nZE~QBe-g
zjPR38QPAIKMDCQ1qTK6VFUG+Yd&q#8BkF%jB{pv;sr8^FcmxJ58DZXn04#9IH19mX
zi1`%AyFU+?(TiFi_>@}o=FS?L@K*s%eCjg#e$KovW8aiOX_-K2D*~lC0;L&C(u^f(
z4=h=l#nO5w2-`%cf46H#OhvG&@rpbI5ruN9q*TRHX3*+REQx`75jSH=FB4SjV`9mo
z*;%p^)UGUvv7%>P$THwMphG~K@QO9uE}I=njw6x>k|A0K8a3@X=Ng)dGqc!dgXE^b
zE4&1}LfZhZ+*LZ>65gqyg+!a`U4SzZZLN<g(Z&JniFTsJe?+^KbvS=P>yF92%eaOn
z(C<~q|Fto*efuTzQv1dG^;-lO&j;8nH7>K%n0M@<Hi@;|)zWMzD$8SPEoS6g2sz99
z;8ga`l4>Z5B2SR@RU9c5M@gedydu-jDQ8U#M#X8bm5RttXP9CBxkfH(O?_0Ii(E?w
zcrA+n=LZ_OfAb2rh=r4Np_O{CrZgNL<%_|#EMr@?!nVv|n`hYO8Mb8)Jyu3mUvOR{
z_+mBxvKjrO8_L|cTwjpXiXx{OcwR&j@{FCLoiZZ8Zo?f#9LidHROFI8+yU-=kc%Li
z&09Dva}X4m<ds3P$g8Nuztk6Lm69e@qhJ?V$k$|@f6tMNT2nMtaJnshKv}7Y6mvzz
zwL&v{V~W=qGdKTylfM`&*JUi%4X}LYi~13YeB8NkZObmG+t8WjJ36bgo?Vqa)Sd-I
z;=dZ;9Q}74sIa1c>X#fr#w%(eB=h9ITPU@#aBD03BpZ~TR-$3T+cXKuD64*MCc8Ml
z{MYFHf8|x@GBfXiV^0=O?yBq1$b!}s$fd!#)Yj3g8oB8w?^@sY6NgRXefk50g?Lw8
zBi{3e-#*;N`}F(qG~RXAYtb?jcqYh9sJlY}kr5?PKxjw*As18bp($Jul646S*P|Yb
zfAXHq?!7{QDkj0K+9LUn+{n8T{w+iSjOAt_f6@=AWfH&;Oi-%rp>0~Squ+mcCJ_e{
zT)>gNfeho+?)LzveKd}nIrIauMREVv>DMt5+%I(n7YT{LRL1$42oBJ+2P;5`{<S)n
z*kFv;H7IAYf4HEJc%W5U$T75i#xrYE*xJ&^>D+7^*cwl#{~sfMzt0L~Ze(+Ga%Ev{
zm)qwo2?8=OmjSCT6qhmgDJPd%lPsNoU-$of{`)}^v4t{ig!5_g!3u!Ilp$0Ll-UT7
zOoA0aAsyhg_yl0u^Wq7tIH1M_Tt&*QTb%CTDf0lZoNG_vK5KFHUU~fcc#jNc&~irV
z)C@`@a%=g3Lb7^t?3vJlV4Ru1Hv}a{Ht)c{Gs$Pd<O@ljv6Pj<S|kdxP*^a3l@i^a
z+DBLZb|Rh@bhsl?Yv*(M>;m6>E?@KNdc77hYQA1;^R7?lsD|olq=><<PC3GOEV%Bx
z#5+=oU$~+_*V_W>W=tD~k*tW?4KbZb;ZXih8TODfYlX6DsJJSS-18mztXewjlrB$c
zCDR*7+du*HGez>x-#}($lx1Ikdwcmw#-Xd8X1six{CO<67qrH`T3KkwXDyk}yn3X4
zg{wk+)=n}z#hO#3`swWA@P1gz+6%3Udq5}U7`t3ZB0Cw8Jo*{@_|koQdOrNVx<rtf
z@XAczUk~`eDkpqMQ!>KC)&B9X>Br$~50xC56cUciZ%$5syiDE7H}cGXNdkk?d*_x6
zdf|X~F9DaV3iw!!l`lf8g$E_$SwnPQfIapQ2mW&wl|?%FEt9`k`PTUAsB#qPAZt1H
z>`4j}Y`@#A=zs;dN(o%5C!*=HNH59L0j)mTE940p5CA-AfkH$X>tf~GWtn$awb^A-
zW9dZ^qv;`{q})h?=T>rmj)YYu$1870P8%(fT1#&2ba!cCFv2)Lfu9fo94u?$!_tZJ
zE|)~`zU;zW;tBCxucM@O9mGyYuY*W8v4p7XyCG_@2nD%L5ApE>5Q8vDuOPaiZfVQ{
z?MmuaUM+16epA84_tI7$l6cX?jvF)P#smR+)bGX=ZB7od&G;IB5OdZSuIDv4VZFYf
zS!H)LA_cXQ@d-WM*aj@<nJgjBJu(lN&S+Ny`0Y7k2`(x4$PHQupx_ufh+8*klSpPI
zfy%w*w9}I>;v~XyYJN4(;F(2==_1b{mK*`{TZXt;SwKv}rnuq)2sC(?V_P&@3+-2u
zlBW;*^AeNbcB74dMFW?K@kKxGlTZ-7J1B)dhl1d@Oy6Rq^`jh_;WNgxh-eS3R+|eE
z({MyebClp{fpseUZ}g@-ar2Y<G~;=!e3bhhrFf7s_6ehPP8gdL#@{E{_+`qw0f0%-
zP5+(pvQXMV{)oAyf^r7H8u*7|I(+)=<;UN?{{o4{^zt8n+zMHz&<=vl^h#f`B8e$Y
zKk2&_)&zseNP-I6P8ROhO4z^$%fk2RV8O>*P_Aa%Ph2%3M974i!<g$vt4&4FEHMok
zVp%18KikDZRY8%7E@a?0ghFTWE1c(3_!VF6cisx8{yiOjOoy)|Rp)iUEr=!N%YuI+
zsRC~BYpRuhEH8l)N^O;$nSc!*Zgsf0;FdQ6)83+_AX23Or9wef80M9RX<h<Q41t#l
zUETVI5IRO-rJB8PX);D=tx-J{Np_ZQ{rmeX9eaU8Uz*Pu%$+h|_yFn7#t0I32rJhy
z8-*He=iQv`D=J7s@zRWGWL+H)K>2lQNXo3;Ro=dT?iB}IigRv<IC}Toh{1!~!i#u)
zv(d&J&~6SZPxm&l^F2{7MbDLfuG>Sc@@xH%Dfg)RRC@xe)|XnezApC!xPMct_fu2b
zQ;&t7nnGReSg5v(Py=+OFuV8W1UhZ%5HwwgxgQdTJqh@0-Z!rI8@VChnNz}%OUh-r
zGE%mGda(m#tC}+_+pqZ@d{C4IeGg=A5k0$45G3dLH+u4|RVj)gcTEt<Dz(uJ<B$Up
z7`#i#5BXXnMM_`upc2z_X0X`}MqlM>ORi}nY7!<#;6U|Xe7kK&a7}{xM%F5rYHzIk
zvCFK~mDn~+4eC_wV~6c{E~=-~y))dvRb&W%-1sDeVD-4RVPwGZOBwry5vWpd7y3or
zXTAB&UWV}+GKuRh3N3G0W>}ych%-(J<!{;+@)lCV&_h;_tCF@adAbw&*rcbi;mVKO
zM(%0uw|ZF`AS-UI6i{v#448~xUmz)md64a)4LC$|=Nw%-Y9h5OZzb#=oJTS}7Ua->
zaj!n^Y`z3cIeSR%>D?{Ap7Zne-BJUl=R(rh==zopw@Z2&)VP?WR><JyM^?h^w|A7w
zJL$2I%IOvLn&TBYC7!IXliH#eb)DLzD_z$~Xj>b16d+2}P1Tz6XB)pMMeFtK8mUlI
zsiWe?Ho^_34$C`iq}<{XQuZkx<$S$=1e_Hr^#Hf}L*UbWN+U*kSH@95MWLf>DsZlH
zk4*wbsoz}TvV)Z8j;T>#+KjUl@^oyYkhb@At`pu)3%~;4a1X%IwG}7+cC`19{pg;G
znmewByjr%?OwNz<YPO10_H26w-rvu<=5y^_wm_=5U6T*n(85|$jG3@%Cdv7KOse(4
zoadJlvF_p-V1Ds9JM<VA`G4<iPF+<opJ-PLtsSq3+tN_wRL2};tf5LPa-~QCY`B()
zYhIgCjqOvmQZ=x27gNpnx$>GbgwwDJx+H}e&K?QRHsD%JgQ=<}2+WD3&&N^=CG~l8
z!mJ_t(54R1Sf`e?&$*Ta>~Fz;mqxHV@Ke|<G2EbAt=7TT!0a@{7S+?>s0GK_BUfA(
zQu#}+Mr$jiHa)hTGh%J5Fn`vBSg9ZaBq3eG4~2*DPzU@KBjOxkYXG>xh`57z!x3=@
z`ND{}+x-`fh*iYVOt_&udx*O$wauX}x2ivoJ<}+rqir)=okua<?ptPm&jBE-u$LYh
zqxJOc^5)kMg-|n;+ew=-QiK&F`8RR&`?xuPHeFEFk<QI9PyoZhl$1uh`PXM7eL3CE
z20V^+(k*SEgZR#F1$77zYMm9m@ymzRK~Vctf9d!E15(jr#@PYf&<XZv-R@i34^=?>
z=)s6;Wm?hm=h%<-UL;6=D0b2ope}&e*<Y1vy56pqHABO3O%})oR%D$Gs1<i#IJ2%m
z;*f&=e-+N`pqEsvCea@#n&-l_N5Jx2WJl#EIv49__RLGDrN{IEaw8PFdUY*2HNVNE
zx*-3sc;6r!%V(%Ed#F@)uDlp#qROns-Q)D``eDGOW`A5b(H*@D{{9n^*#80LcSCEJ
z(dR4^12Qo&mvOl*69G1t;Wh*&mq48?1b^E&5WV+T=m)~E$l?YV2HH;Cz}%+Mx51nS
zNlrOT{{JwK;$n%CY{{+kX%4d7G;3j1@zv6@`_akn<>~bH^!cK>bHW>Ps(bt15q^;P
zB#A|9I!7d|rta-$_t!y6W$~Qwu#{Tkaro}>HpfqzJa&-5gI|e`pBz)9!5*y=t$((O
zMq`&c<4t%!$92#LX=UU9eiZOLAOC*)<4I;^KVO2prKxo+z`JP5(1BfkAe|xE;B6q$
z#P3))Cht4vb=OP$eDQQsk}<;a)h1s+8EqhwO^DtpecmIW7UWzb{rM6tM=t6Sjz%An
zE|DK+IX(X@8(!&cDuRt>jjxyHuYa#^PyasEBP%V84V@(~G_3N^&!_kU6LfH)Q%VHl
z?$@{O`_rE-*X+0U-3MK7U}<rdk%5&J#`w-E78dV!xf0$$pkle-+td1Ej7Pp?8!)8S
zk6%B!X65yznnA@|z_wM4zQkt~rIkX#TR2@q5o4W;e|%Q^oQ|YNszyDraDO>|va;8@
z9(ud>K?(L7l!qoF_%KmE+Skn3JpZlM`lTMq<f{_B1n+D;7)vc=FdI(B(UFo{3O&>|
zYs$~t_0dj{o|sLJRFT4IP*VQr8ID$l44H84CI?JWJ%2zB!iR|t@;rC%E3z)+fPDFl
z_kzK*hQYDNn{*`$`Y@Lauzv^A1}Q92E4jG?%Sh>j@)Ny#J=9IPikz_4wUpDFSp#_b
z%T3Lk5@e_6;v90-n$d1OK17{uwd}cF8DF3Qz9R`}4@YZby~W!GsZy-X{_^^=?28fw
z4z#u!v$&#wn7!WMl?mH$4ACW!@1rEYBNEnY>GO%ry=0KUep_lc$bUe}@RroRP0?e4
zy+^A6c7g%sV?{rx3Sfk72>d}+00$$K+dn>%3s%(&j};4s1%*caTcv_kxP&diooXO@
zH>#`SQ8ufnWdQ#8`#MH=0JV%Plu1lSE3fsQPy7gk!ZzCoQ8SuXz?csRt&Kp|X&AjL
zrLhTuq+OfR?g!h(<bPKrWasv9Zxf$_TCvjV`CoU(+Us0(inOO5b@NI=tD@l`CHr)~
zf#H=pTxIw2&rNcjM@o%o1=H4oDIjk`mqvr8?1XYfTGpQX^1G$nmV8-ylm)8glEH8i
zAt=4bT}yys?Q=@MCSAjD3uxmmbr{m9fz}y}%BsI6{lYbt8Gl($^k%4g2cFLocJwc4
zScls4ukkN%n~@z_%CW`f8X(SR%F&)rna|OqwMvwAcdmgeWp1JRTXsp0IVZq0?3xlQ
z^Dp-BRX2OxS8x1`hcfOcRB<XoD;}>+F6AeyD@rR?K`6R53_1@+E}M$!yV&O=jH*@|
zm3TnPy`>{MCx7K@xu&aqN|deB0Z{H8<3?K%T*30wLNh~b(73v0x76n&X?3`|FW1cS
z>ZAe0o)e^&BwvEA4=<I&r3PT-H33%>26e?bJHTXmKdRCT2;|`H!@7XOr9ikljEfII
z;3BZ^dA;K-z)nm$Q%@|yC;~w{({flDMlpR@>)UFeUVmy`Ec#f5svIg*+`pzmb%Yk1
z?o_C$rs+RSuAfFLueoZSuXak+su>jKihh%vj6>6^vWJWnT8Eb74izeS>bsqWb2F$I
zHsd<p+41h#0`(N@G6cidtkZpy-nA#HNYZwLJyD^$G)bW-DAlaPmDd$EA4`phd>a53
zEQh*9O@GcX-)D^GV*A@s1v^Cj9pbX(xJ!eUnlrbhLF*0o;7qhR!%~NaT@Gm2?3Sko
z8m}~2Z8^DxIV>6;)>FA9Lbl6CudXajIDTzz&<2iL3S@zm2g;q?jt62VoS(>Ndi+2}
z2?XXOG|b~1EiS2<7@aONv*8U~okk-zu_f`m<$s++f}8|=1et1e69>@cBh3njE5ig*
zKm{eGGGXk;0%8a%eCzA9Q~3tJ<Zu1|3F9nE_ZpN{S-)}lJ+F9o{Y^Jko6qO_&p;Ii
zAp`F?1XbWnp5G0kqxx-F8D6KL_lv`-lw@2g|68V@sgI<?oUSv9-kySv75$(oC|O)3
zWPkOKnS!E|7GC~_07*WW*nRIYXrup%IcP)iA%oESTJA9kZKS+o6xwil*DRE@6PAC&
zD0IKBRcR%>*Nv(hn${SM3P3Stu;xF47A3s3E#bz0`*|-kGR&CC!r+l82q)dr;E|IJ
z?+hMU>BEpl&kP>(>cZeL%gNJ|IDAYi!*A(Cma^k?LZVS(YHhH=(XkxWQ!6&sTQP*p
z*4b$YiDcCbAv6EM5HeeHVF=lJYxf~!O#aTO*&*a&busbm+96~kZRPaH2jMB~G=sE8
z7^0OU$kFEiCy+Z#AjcMcU;sI0KuyP&->BgJ0}qKtI+xMsEE57UGM53XE)$ncq%6b%
z!?zNpESv`ecpd+jq0AH$12!@<mmz`(69F)jaq%ZQ-C9eJ<2Vk!*RSv&D77ewdH_Ly
zs#MnibK61P7JC|O&uI?Z|NpSmTc#hWB<|jsJvfeKnWD&#Pg|Eftt=UTFGw1+I5_}e
zHB29WPX9bfsVp6@bQmeMro+*HPiH(oYw$D!kPf;FaDH|M;xjZytNg6C`58^K)P>GK
z=a)E6&j)E`q(Jl_&~Z8c>*LoaS(ud~*!VY|_|?~sr~f=P)|D1$!(_A)!NBzUpHK0I
z0TkASNf{@k)6@5VPv4$@{=vz%#UwqGe^5@}|CrdwZ$)}RlJumth%@OaD-iu;Z4f%%
z%F;7$QX=inywgFuHzX$ovt539_pnM4t<XAfiX=vfBwIr=HHmVjCQRhB{v`HiBgB5A
z>>+8IQC4!sSDPKA(9ZDLdcg+`5KbiR622Cmu8Z`lJV|#(yBp+xDM{x<9`popp|t93
z<hM`Ae7>rp&0@%_A@-FG7Eb(z<czRtv4<!nT$A4W`Ps^7KbYiLWQcm|KWR6!kx{Qz
zX4(qld_&ADlg|NK0Ye6NB*^CkLy^+!?bs%if>ewkyv2jVB-BkSEHD{I_Qc?vumEfk
z6|xADUL#Um@oMgWQ=(Ahe|6xTHyC_g1d7}qm*|FVD`n_mKKTx@E9Hf?hU<}A<_!d+
zD+i7=Kt}Km<B(cd0!V{eG&)tx(!6AIL@4KK?6#U_geKM*HY+*Kh)PGOwD6^j#=Ea|
z6imA>3;lGzy8?MR+*FaAz4himTb)z%(K$Y1uufB0v7FC;qb;_CJhsZCCL2>Fv*7(6
zn^hE_aLJ#%-++oma=GFICg}ueYq8t6b=P7j`698g*$pALTP`U_IE8cERlB@yz3Oji
zy~YbK-I3NiqnM=V%!3x3qaow<a0|8q6kp_F1CF7R(Y(uia~$Q|$%o?~7iz4v2)v*l
z4unYN?2TWS)2}Qj0dtr9uPiHnyXpx^t*7Qr?*6(8r4hlk&WQemVD(I>>k=iDz-&9w
zbB?SY2!7+WkZ68#In7p0<+Lj0^IH+v=sLnqNlot1YfkZcNgdl2RY6_Sb7QR$FgGzC
z0E%%+$bido_Q&WHNg3|Q9Dz|>O6hL5#wc%uGv0b)Pd%a;9fX3`8UeR|5l+`Tgd|q&
zX_)F2Z&U6<g=AiL?M}ChWwW(*EXk!f+a+y_w5z+Y3ri?y*lgcw7X%y-4fJqr#sHlo
zLk<#=vCS=3hAgG+q31*ep3xv<KRJs3F(wm3OFkKs>rHQE6x9^Kn0$GU&1rBc!91fr
z?WH)W@tyH_W4l^Y*67H8neldY%>2xiZZ$in&h&;+7wBtE?%E!Yi;f4^<J`JKzCEut
zfdal|ysuOi?E&<{7-SFV{cdaN?4~-xcZ0hnn~_17;IH%fNCi~0AhtK`N~#jR;^wtR
zA(df-npa)J+vZE70%aAfuv7q5BPXMb){_xdZ7Jae>T9EX)lr9kmkya^ZOsPPebM<j
zDX-gkot@IFx=Jl5;`}nQ0)}S6$9s4c{jo6zdU%DI@|Go+^fE<jpR<lAK`cN?9N4)B
zljc*dToG_$2j~`8eiJv@m5&T<xNRW60FE*AliW50(kO<X<hGSWVYEF~+bM@@e)}_z
zbZ<F$`D9A=cCStwCgGy?=37gN*q|5g+Jh2iA&-Z=7Nvh8`2I?lKC&!2Deq9^TdAL3
z*T6EWxoK}}|JrFmTjAWr4-JeAr~F|-^6r-$FO+kq@v4`hvMeA01DC?GEH!_5rCnop
zL#%JDqb;e#Hj!Tl<$vu1BLhSC39rh|c$bQ@`9AW38C5wW$?>m3uqHu)B{FfP1#Hqc
zE;yzcW2&TdIE@f4J1pr>0<3OH*%f!^ft-@&D?WnMB+u9gx>BrJMRC&bm7^vU;*e4r
zx9{{kxfNK@`D?)H3Xlw??B0Jb)LoFIPCd8fFe(rj^5HVVp<?7dR94{^6EQVP@*0Y?
zp;+zsgAak|N-Ra;onZ@>Tw}2vaKaAXIt_(!3Ks=!?O!OYzQhr`^FAXfqGNRoDS6bD
zu%d?Y+ZBvk*TPXLeri@b=PPt4Tt+T6F!lDnEtz(y8)>Gg`Lb=Unbv>wbGs@Od#j*u
zva|XOa!EQh1mg(LJolx}Eh$c#AEOs!<5<v~t!;aeSysZCbT7*Q1ach#1Q#2^_|;tx
z?^Q2E7pnf?w3Q4D`T@nI3H41Ij8SC6Zz+OqjKFi=jz~0}p}1rhlxrfDcUh&Y7hk`X
zone0GRj=&)T}Nvvy|iQ*BQo;HrrXJ>hW36X)o5DC4S@Sn4MQ#!&@E(eSX4;OASity
zRbEHMLoqUiQ4YzVg){5}nfxv7`WssETf5&Z*51(>_$#yniTBrAS5kfDg$aC<9f0aI
zbc?#-t-8x*XxQJE*R(8C3Hj^C(|-Zmk%X7gXfzW7HZqry4*?UGgI+Nyf7M#aj^j2E
zz56Tl1I=c!iW|@{(CuyykXr_Q3vvoFbDBfu{|_l{T1Z*4WyOOWY}pi9ReZHmlHK=C
zc3%!}pNHo&QfGzJU{&||tpmOx`5Pn;8r@j{VGz2{AKf1>QYyorH~x^6i2Skq+woN&
zpAa0+0L>r13vhh01Yr%&e?gVjqDgCzXQ>lkfv>M&oY#A4WT3$GfX4uje}4XUkXhL0
zvzJSf8cTv{7fFd_Ae%qTUGr6|^qV~YW-gk!lS%jJl<vndmB;gc9p@Y&lZ`_*3io_=
zuqc$)!q5QUdt%JUMfd>j6*LLe$T}nFyq*<VLF6hyU(Z1wV1vm1f2i?gSq#>7)Rgch
zC@WmPRR2Cc5C0r`bQB70d}qK3WB~vAaR@gESO)7mrGy9Se*WEkJN#Z#&DKy~qbWyS
zNW*)X=t3ExHE&UoFnrzJK%x#>+*!wO^SaSGvp-FABn7d={nyX$T{ZIcOAn#4T98ro
zrr<q%CVtbea45Wme|m+wO1|W=<PtRUe0(x8?%L;>=LA@oAy0GG(0lQPc#};+J159v
zFC87D6(&V3yBvW~#?%yA2^akVj1}l@LnRtV@rkJCw@52Ty=8;xXD$0_o)t#}EuHqx
zz!wi|(6W-RMaU3LwJHRTPlReLEF~s{Py|+yNttJofCa2Lf4{#=SDHz<7tDhUH1V$R
z)~{(z<Z|vg1WK>z!+ceg(&aH4@2bS%YW)hhBQz>-wdo~Oqn)S*!2wg9ckr6-gc-bS
zz!{>sj*&fOBOYkPI5WdzR?*rCBXu9Ut_vzp%KH+Ug7M0bUi<rIKqJM%GNVYA|H(X(
zO`7k33FD`Me-^#=yvDAP_^rg!U@a(dc5Ot?W#>pEmF;96da`PSM%ig^(PhVWyjGtU
zLkqDMq|&t(g77zn7t+8P&oz`dL{Le$NH2qLCRCYfvesf>fivb^xiN;6@riP|9DLF~
zY2LEPa5}iruomO6;Vdp~ML3j&`Yoi5VC}$iD_N2Vf0*Tv>=`Rx6XY!~NueYP<C^U~
zF6R$95^SFnFKrRI3iIAo(l)l9(!-2DM%y@*^c1C@q!af3Mf(QM(tPWZ%3z{0l}#UL
z+0@Ai+w3VPvgb2bAWLO{;>FMx^P$C<xBX?yDb#p>*7Y(3AJZcr%Sru0{=1yIm-62z
zmG^~0f6s;gMf0%}jhTtYu8Chsvi10d(g=mB#$iMJLL{NRZYaHPYXHzHfs*P5AXe<k
ztaJcEhO#(*>0B*OM!x2*AhwMcy8_}u4_K5EA#%kM@jfMN6jY7qg+4<Yf#_?B6Mv1A
zZ!Y%1LF^_g%IuDFg+6#G&{91c8#z+2lDtd!f5<YeL%NEmVO!Jc20SI}oXG=r5Ew<S
zlaa4iU_RxlaIoyYR-;@e3~{4Hj$0$u>>;3}=g=G4p#K0zBHkH28pkQufa5I<Y~6Bl
z5(cz2)pb0_Jmu~FqPV&u(ly`eK<OrmY!O~#L-E{r;cZR4k6Es!suvshYBs0noAL2V
ze^6s$uze$JCR{)?=94RgZ4^EC-)?8XQ7NGeTnRV^SA@ol`tuF8hv6&oos5hFcyVe(
z4$U*wc3)+^_Z5~dzxuM-DJo<5kxp5Y%f$JiNT`w7Aq)Ws(ss}ESW(=iBv0iyTz3q&
zC88AHn+gvX;vQ#S#q~=yyEmNs)q>8me~Il`wbT>`)7)z8Xq^}xx87&#GFkY@-evND
z#^$%2IwEmE1hdj0<d(Wv;&6Ab2I0Ch!nR?U_|T*|7|t!-hI7m9g*`YW*gKlSo<d^A
zhZOb*UN{MRy2zB`j)i?id8M$gW<RK~SDqcn;e!j(itC+)eMRk;6!ziCjUOFJe?17t
zDc5^(oKm|2$J72eY7)%t(Kt@I?$jVvRBvvOMwxO;J77U5*&r%E=4H-JOEo)cc<ML3
zT=7wX9BW=@We=J!BMX}6l*KblqxsWNrmRGAL@8_|I+{^pFQsLEX(1}Ie`Yj&TpCZ_
zbAY&`DNuw$x~mGMtl<2Q6*K~Pe+2g&_-$DwM<<GJdAFGUt5LyHc24D_FryXTUAPaq
zBjmoNfVX4cq7(?QduIQd;xFl2vPIy>h7Kn6B-J*m7QH$ls3vb{oXW{S&GtgaEwx~G
z*n=;iuv%Tqw?vjo$8y)E4ZIgh<Bm32Q@p=vgJ33}?h8ik*z~MoZdG{Ne>4tS^E3W$
zfo#6NVXdR%VbsQ%P=t7?vU=w33}wm11{SmmRZPjnMZ&UU(^Pb8Z_@>PP~a~{yY<>}
z?bWz=Rd~~@ao2FY^bnadKXaRMeb9Su@hu&E&}(ki2!|ULS4z(fjjR!%D&7k(qm@Qt
zx*)s>N^bD1UPyDH%N@OkVz&7q!PPRCX8@JV8NjCU&;M}^@L>-yJm4<gdP~MLY-Jj*
z-wzru7`SQyj`uOV-gC1wtYFO{tOBb(th1QfulN-SPDUmKop4mqOI_0J?uZ>7sove<
zyNUIcefeJ~wHjlW(dR4^0x~t10jn+(m#n}oD3>d}ESmzLT(`)*Eb#+>0D74o0zo*V
z=RL+*%&5bA`<QVW<R);Yg_R>hG)8zm<Na(ql#V$)Gx|>fo^p(ljxZJ@kJ^lK(;jkk
z?09k&PIKRJ1b|#e8|e51Sk$2kQs&}g`fxZ9(YmW?^T9~1!gKvqyYvs7Cs@nHMz3o0
zxzxpx2+o|`8tgk$kX*TcP_OQwwQ{Ud9fWOo{7C1h?Mj>Hvr#Ipy1a$aLHLrR_Q{Qv
zehyEqHOvl4f)w9(>Jav+^-xynCNAf=TT_lD1=v;`RlnC{aiL-i-?6RV%b{4dWiq9U
z>8Vcj{R-g)WNopoBLnMXB#=8Ox#wvk`g-CRC|DDfHG-))Ra6syx2mFA_pB`^NVXj9
zd5kqCBP>BJOCzoq$hr9{ySsz2#~Z<**|;s6c^8hYp@{|k<6r+wEeg@{7Ry^;y5)zm
zWgAN){JmJ(EeXpJ4`0n1A-OT}%5W3vsY%%>QacMJW_LB*8OcwvF@iNGX1^ZdUpD=9
zjNZwA-P7vp=$1i$916#`181ZU#WXfYT;EOlK;TsF|4tv%^ih;^f;hGeFUR^t#efuQ
za%%X=S)eh5z45u-vNH=*uJ+;>H=LgV(0eW@VU_`sk-RkT>(d$W*|`9@N+vwB6Y3#;
zb}k6xEYaxwcu*M~M4uJ=r4~r4XKXF@OE1XJ1dAmCn{<7DM#DpFug=aef2`vvZBz&K
zq?RXi`()_5%UP&3JDC29=B$-a-+Eep1zU0!-g#TOr|Vb~HIdL{P8IdU86y+lMVr)G
z-Ol7S*|IM|wgTUarQPxgNm8*XINO|q=BVB@OtLmi)J3J`w@BxYyMgCOX%9@h9e!^H
zN(;xaxO;DZMxLdlwqKj=n8bqcMik+k#J%XsV2IJ<nXkwgq9fZw-Wb*btAb|x9Y?3N
z@>%qZL(LtZ9E+cqF(k%F;u<J;?`2h%`%5$MVa6`OzVE&9k%N1Wf?Utj*<)XlE!FuR
z3+7fO%Kf&lQUS*ZO%JJ0E5+Ig+u_uAOvRPx)ImCbSw5h+K|tuSd|)oF`pgvKdn>08
zji`Y%f|JPZ5w&hGd^Vz%o-8A3bvUPvsJ&yr$BqMMn}gq4#>4xLqm>x9K9xekNZUhM
z8SHURpsK}lsD8+A&sVX@Doz#Cjps$F6rFHBv2{e`xvi7&rDI9-`FZW`PJF3QEZAN+
zp-$s}GGa2|c2}oLbI+ro1GWF~ag$C0h-bf@E)EC^q0Ia090}+eR`IEh;B5{bI>V9o
zlep8$u{-oy0J&(#)(Et|H%E|nY&mT1_VyhBJ=eLr&Hwi>+_W3p6GH7UD#rrt?s<z%
zqlKDm>?tk`I0mG$xt;4<3~;&xu!y2ey=vcoPp78mOy?P<g<3*i*?%bP#U1iG_Dn&N
z;T{}GnH736cM;2Tf#lkW-33ne(^mI4Jz%j@eO)vy&ULC8x~rLCC67S!db<8wR60!>
zS)=EXI4#p`LKI6H%2dVs?s{ODgEOut@|<v7yZy7maXpyy(}B@D;W*53crHMuJa&f)
zXXVaa&;J3{t`nE>heZ<tHZqrCm?jgK;voeomx{(Lnt$*VDINcO`|TiE*iWy<uS%-4
zFg8rqp&;5-Qler=QY2Z3nKUFVzS8A0RS|3?9KXJFe|>&C{B!8ZBNA_KroVn1;td2N
ztP7I@6cF6>^Y7`~;rE^_i!OzvXYz-U)6efyxAK*J5hQ6Stu<z;p%sRdw*~|%RW!sG
z(#o?|r+=)eg3e|hRw)`Qv<^MRWGC_zx@f0JnQetp&hnAOU#yV$m9mF~n~t)Q3%Tm<
zAcc0=@79AHv_QU;q$PZ2nf6V(YflwuCthJsNj@j?peKlo64%Ab*9&aE<f=$n6Bwsv
z=8R^E*THfqtQvHPRx)eWdp$l_8KcoWU+q(+oqqtQ2gzaZ>s3@Sh@fpx74)jU%EUMu
z)GYv|0A<%J!)|F;&VWH#MSDlLkQ)psXxv2srJNCH)ga%U0ZU2Qc=yqOU2=_pMVld(
zEEIn`1}|nwRxDR7B{ea3J=HV7eK%Z}9J+Ec!!V|vL7w-7Nda`n_w}|gDfsND;<`50
zihp1R-S3P;<Ei*jfPE|uJFXFjseO384Ue(GLEBURHHnlF(%TQ-kB*!cJAxh5-vhUv
z;*B8Y^w?9~0I>_gfg4<MXAoO0oPj%bQ$0mwYVlH+f;8Wo;dhK!YY|8$U<8(jBH?%^
ziC=cAy|#jdDl}AP9M_`|*9m-eBp~G{jemp528b1Z%9=G6D`(ss8+1Rqx#*B=<u7tC
zY9+k0JtcW!tts8kA4$f=O4}S4=ouwksO>J0Ok~RH;p7AUn===CTnFA&yrEc|&-r+D
ziMQg>E72fY5oz$!+58H9v=kmFtf_{2t(){i@~WI8^DpmYS?hxSUHsy=!=_djE`LzD
z=-asfWos}4mxVQxC7;+==60UaQ~kCpJZH_~0UMHrj0$LM+urN!uGeDgrWA^vZt8Hy
z6xvh#KZWk5(9UGPAcgi$3A1Cez89uZ7rVWV*bP|}ofXp0N#{SFM9~@n3O+1}Vr(6)
z`jAw!HWb^N=OH)|LOsI7X=6^Z*?;QQYmy_hM6o|UXpED{KK4@lx`qmJoSxc!N%GC9
zXpv1)?ATfw;xQyqaLivDZZ)t>AYELw?dmCU<JF{*pS4)ZRd+3(N!UE`WYcH#7a9e-
z-6W~dF>?eY>3of3mK{&WhjhqsRU30uDPwYWGr8b_d#si#&*DO6mSBjcB7b4bBB$17
ztxh=GZ8d6-A)%7Ep>sU8!BC%B#SF{5h8xy>;h7w1Jp#1%0_Lt#X9J~hRm&0)8O2kh
z2HXW@*;O4<O3M>xNoBlOu@undbqV!hbv{iG5LSv3g6rfj6}!Slu^!>NGKRU<_FgR2
ztw#QObdkXYdhBH@PKr<foqt_YthpC9c~=Sc;`&m|v(1Al5K8{jT*4C|vNuY|`m4QA
zi=PPswdqvT`j`>7f>FPP(PV<EsAjCr=1{c}9WYA;6U;pC_H2&lxs`2U+2zDx0&P91
zq<gBR16p^7D*+P9`=Z9Uyn-CviNa9D*Yb9Ka@N04mvx1+SCOIyoqwBQTsj(n(v*0q
zJ-JgG>unk;OU2c7?arC7t3EJZsOF7J39N#EwSFpfXPa;6G#yVhY)%~v%$}@a4(y}S
z*`5vYBRsA)tDCA8!9cg?C&j0O_;M!}^VmGxCTz|%mvz=#IY-e^99%E#)2l5wQQ>Ix
z<GD1`;N{A9ii$!fc7MtJJ4HpJ;La<>?mN9c+&EitzpOmb)$0C0$$r>9hTZgDk&>%G
zG`$k=GW6d4D7<?@J<Zvq=G{|PYxTY3Y0bAgm8?1#t#EekVa*Y7rH4wHDi$Eg9S8mL
z+dU86ZS!c}JG-;q-*sr&dExlHNY!+L7-|_!#S?v7t#WN*t$$8M$3E+`v?E4}bbC*7
z#k7yhRx0zI_vb#7pP5HCwJtC>WGZg%K>3;Ef%dPaDHuGjF`cqD->m=ADx<&(@4_9c
zOng%x?B-x*Z%%Sy$)IIwPW4qK<5JEWks6G^p795zf6#_1xBj5?4+w;(8$5WwTplM^
zLG==y%jIO6(0}j0Hj~HsPW<1as7TG~S4i!tLvV0`)HnS*LQr%D>Y1#MtS%b6Q6Efv
zZBH=`yUrPK2M+B3EE_g;0nxU$&Vf>veQt>R`tr2;;)d!9l44hx%c^TRC%Jln^uxio
zo*EV>l(w?+{r#L%IN4ba`#9Fg#RLS}YPWSd+NdOSctDN1>&1RI$eO1*XXM4Tv(iR|
zt(xScx#OdD)}>D1+HL_i&xB=_&w89v_hy9N37p@f?fuJNmoq6R{|l#1Lxz{p=PVNe
zGMACDF%*~m%`CwIvbVp^ESLuaxAG*Handjo12!@_m+|Ed6PE}&C@7b!)+{f7Q({%|
z)m{o8ehhqgJzd^U-)<5I!>kephWGD-pc|b23Qmh!4Mqs2r5N6S4u3sz4w^m}`XJ|0
z(#OzuOILM%kiyavLeYor3UPigDEuCJMnc>xb=)KADY(*A(Dli$(|*S__Y{bppfzF6
zf4~26;z`(VH^<i`)dm^S&XOX3aX^wosrW(ShBkeN_*<E34byMGPWKf_rgV_t+s%qq
z&Pp+jBSz1((__LTMnZg6nzM(o5olg%ARObi7wGBc!4<}f=FUn!M)5p7)YC(g;sWf<
zX^*_Wymr4{-cSFX3JM^Z(r(bgGD$%F^Yi3ypdcBH8vvLSIQ;r|_<s6-v!|L=+h2oS
z2F$spvox54W=h#X12Rq5L$yr`yF{#cT*kBoZwV75DaxN;KZb7Q?X|!HV42p~3o4%d
zr!nlZLmDQG?00zgz^pG@9ZyFSYvrVc>zw|I6D_!VQlhe!p_3EdCfHuGq5#~*-^orN
z?8qkEk{k`hu6{azot_SViVLSIcl~)>p<EPuj=US?W{Fz$o*%pp7aTh8FiD0)FQEHW
zWSE>VTZ9T8Q&!a(0xuge$a;!Tm~<K?D@k~|KF>>-yl<ix!|9sU;-hGQ>6W2QmOrY9
zM-ajhbuCa9!V4ADloc_Gz83QU1an6B0`U~$uQA-giaWwxSBx%y7`(BV^OE68DVcD+
zJi%U}fRR9`w3+pL&7cKaa~mTgJtQr;mQ>nmx@yNEMuXWqZ?DTKB<oiJ(@=a~B?T7?
zjneN)hl&EF5luQQn(*gJo*jatZi8SzFKH(GMv4Gi(xKv%2;goy{J1pHKFsEH*b4MT
z=@5;e?DbXY5RGAfQrcs+{X5cOC+LQB*zNezbciT|3^~j_<+KpYu#Tvr1d(h%U23RO
zLMJ7^&7~7n5>!fjVm&TPO>wyw*hjRW*IS8zs}7?nVsYJ5S;$F9SHLA~K-Jj9Um?%n
zSO7%P{N0@ReezB_F`eI}JG@O5sB5Y{p44-!;2DdMk<%)FvSd^Ybaz^x;DS_8ykRjz
zQN|RuhpW^aW;EX>6Q+739kB-2RNo6+b;VM_xm0;Hdvg{{lv1XQLAFsxkheT%s~BPv
zFPYUi=6E#o)<caYP;kz8Y|lXt(iDk`f}7P<vl%@ic*WK4v}-ZbWZI>wh|ha2iCd|0
zu_~%^9cVIts&`Pzbax}6`mhw~;N(~4olq+ll24s+3TW!bmUosXR7ufQwNIXc&+1n?
zM`61ZG<>$ut0U-GhG7Nfb!;X=E0Dnbjing%uuM90U@0c49Y@=iVzxS)*^`(zm0*%`
zCwKQtu+K`dkVd!*Gox0DFRHtMz;;gcb(f}USRGz}c1f-oVSbEeL%G$s64H@KYH*4P
z9E&@sIzCmLac@N?v?JM}r#@4WdEcS!+|vLRsWvNb#RsNHQT3*U=()gZtJQheybi!m
zCQ<j4`E4FUiy;f=Y6bV>g=FfoC$v()3Z*Fmx_$@Tstj+(J{fZE$ayE{qy3^5XgXQN
zVQK|`a;wS7!dN1Y7kE{6l@fiMVii%H5LYk6z{ToLbCHG)zDMD=59l&kg`t{koNliw
z6R4NOen*&r>p0_v%F9z7+9t*rGe83aA25@>Uj2EXPva`4;ds(`_K$mb%Te5xPno^7
z-2o~4PGVR3-RBpNNK<Gf6DLHX;`UoK38bBW+i;XmSvB?S!0q1S%c{63^Tf+0TPL)b
zF+Ia<>pEEYC;C@_CO=7aMqKHKQ0e>qkkAxdgq+ZJb+0=XxNtx{ft(_*?m6Z`F^+A^
zWh``ajYcZB7Q8_Xo{EyuBqpmXl_3~M+L`dbAMTb+BY8-X_MCc{)?yagNpevnAv$4y
zwv-rkUx|~h5!)7hYRgOTpi#a-?$N<jouW+<xYKEBv7#`j8qwr9f-eLA@#Vr?69R&A
zh=RJ7FZ1r)=2;;xPyu9`4x^_ioMs@#ls6e9o4bSL(&sc?&p4z`&0u-`5n9y#wZvhz
zEs=^8*`EvKu1?^d(So(ig6X+rV+)*r>ZUZHqH{0U4(Oq3-XUg9FI`Rpd;10QZLM_Q
zE(nfHQzlq%7-CKde^g3f&B|M9A{I*HR1gR%DKsA^L(kPawA7VDU%6!t;oHbKue?pG
z(4I^68fctFY|2@}lFLF*@WW;8eGKmcs3;>@y#v6-qh+vKiU-WUg%zPQ?<fO*pRdCl
zXVPH*^z3{)ydxv~e>o0MaLj18%OJHO@>CUwVvx<s+Xl>Au08_gDNsG28mB-D$YFSq
z)=~2)sX6q<Zo>aZ%gg@4G`*$h_zCRr*+CHnLmw+UiqD^*NI}ht2}L#k-US_VRBiPL
ze&C&tvJ6ojiA`z)4#Y-l=9JujicM+|dJo~p`ydZ#^s9Za<@&)YV4MR=Vs3h>_kHmz
zMkEdxPGCNALPDjEp=zk(aL|lIO2r6{IcR+#D=$6o*9lAVOF!^Hm>A-9^MFojWR|N5
z0&63al+23!<eB52IM%OM>I9n{hhVp^RQ&QKNA~z%iToC?=>I>4?DKIo=C+Csot?SZ
z!VX!etG84lSw)d_e0x00Xex(l<?xS$0g(mFqoVs5cCv4ocT;)%KkL$c>6g*xEE5AU
zIW?DYf-ElqF_)1!7bkyO?{AyA4gQ|LLco2w4TxS*q$p`P6zE;o+YLD&XgXk95ghWv
zzRNAO1IIaZ`|I~m&KujcgF12AxIpXBulb4;MLr}YDOGX7DV6ME`&(R00-{F8g^ge*
zk(|^VohTVkbt&Md81P0^BgAEn#;h(Y?SaUy)?`nLBZ|eCq9cD!jh#VZ6v>$!3EQBx
zg32~31xFgHQQ1+ZWN#!;smW9TjzDq02i~v;3Mx~K_6$mMB6}mcC`F87Or-rv0!rJn
z<f38#4H!A9m{z2qYOrVcwap?zc+6QK12&13Y?@IV2QuxcfQs3k!A7yoC^9IsOJVRa
zH(e^IHlqkCwHa}8tZhbr(I{y$6qe3yyzGQbzlZ^Z+4O@#!nr-eLI|~I0xCgC!6=e~
zNrQ<ciAiG^R@L5+WG0S~!3xTjGYVnxY^m`wDA+W?vY>3WqS#k9T}14MrgL#Hq{&FZ
zB~JD$TqKD$Y1t)c02!h#<)*XXS;=NeGjRwua*DG-LD_2s!lNyJ8Ap325X4Pqg=(q!
zi$aicdjkV#F&Sp;((Gr#t2C{$%R!q~1#Z~vr_sS(8XsIUX&Wz@Z#I`TFNH!<VTg=j
z?A_{eW`zt)I?uE(c%$a_ZgnM?^T!-qiPfILcT0+9Autzh*1&R?+$@>JmD;=n)~&&&
z6|syJfn_of9&jChlR>Vg-Xt|ulYxb*%I%qe%C?NKbIsJoE1;6eMU2*B(%_knLTHYF
zNCwGu?_G=P;NHx_8pQG5ci(ky|9ds)UXDiNsq4NUPtFIEcL6_!<L+<WPu&mrg0YUf
zH-oe3JA{|Vnj-eZ3w)eX^c9OkCE%|d?{!Bvr&Am7;9~TDpYG+|@MqJD@u)lM{_*A~
z`~U54)9KaqU%Kwsi|KDSr~YjGzWe>={nh^Y;E(R_H$NZx-!Hq<%kgP992Tu-!&u5G
z(UACDcqlx`gM50P&m|7SQ0lN!cRe}l-k+RJ#@BfFn**$i?jOKj^rsgu_V90Mrb}wB
zpEk|Dv}tpHO&(rE-dsZ@cngV49wC0*y>uA*sQYpJcHF&k-!fOJk|T#)B*BHELW7wC
zudoEwq`c|=`LBOtS=1NvJp|L}=JN9R9<B^7;GN0egio@q#R8}6ei&blCr4K&X9LI0
z1G>YL>2xp|xj6g!*ALSlkESQnfuq@j*W=L?GF~Hp-Y67lQeHzErBG&H;HYHyfG;mz
zbcd7i+0kJ7t~-47x_dkLFg;$d`Qgd01L!cC4o1^!!_y`Nw!I$TOwI<^Gq|@$KM&3?
zPQD+1XltQB?_A?C-k;!2bKBXAZ4Iwyd8Uas-$tsL`F8D{yl&^@!s<E9om?dP5_ZYS
zkrritAD{dbPHz5{fKtF-&scF}B##sgAoM5-R{Wd*&tw^uB`{z)<dE(h{W0b)>ND`W
zw{~R@?|N=y-@f_HffDT>DCfDM5q?(i^x+5eUyb8MhCV)ln_iH?HsA~}eMz4qk{5!v
zd!1TIR##&Dp<zucVQrPuBf^^UU07SC*P_{f3cXt*_r!MbowQ;h=JB0!@-;nse4mX+
z*VD<#1#CL=ylS3T(^G_ak8DwWwE`P47f?YO($Q)vL&b_}B4ieP66%(qTSa)Z%w*-r
zt-(<>dMV|(4L-WOYK9i!{R7dbG&F_ettIIVO(Z_k8ZbqXo=Sz%sN|C7qO@2vSPS%j
z>7rqoYX%`&;M3^XHsC}(lcDOTIiaA}kT&5&#$?`v6O9mxs=5hh4ES5WWO=iMkPijL
zxC)B&C{T>^TE|+>?U2uE*EB~yg>9qmN+5Em`v<&O)R-^CPvzRU&H)RG3f_d1#Z%zV
z;a2yh*R4uu4l8aI%VXS%)7-6MpKD`(pKD{@u7nQkfmL@EG@*4;+5tcNW$@u@(wwbi
zpnm1L4)q_9+H&zIgXgKOh)<k%C87YNXog9)o1~&7r}-RFK4p6LDV_|6^Tiaa(j}&k
zkfUyrJ(sv<pum+pN!f%mtCvlNjGBB}>dx=wOu=VVOFP`h%PI)*qd-U&K!|mJyRE52
zXj|iaYbs?JoOi9NuMwVF^&?N^`iUIJb<+}W9q0NIZw)7L@swo~PIR+1$X8R2rPXpw
z9#xK|g>sDhj#T=*DR)n{tJE|l<du2Lq_#=J&)zcqaWc6$IlUa*ZbRjm_U*GiMeCH1
zNo#hAqPGq5nAs;gq<}?e&3-F?p{QWes*=!{gcM@QYr512Y7v3_3?JyVU=?rs8~17v
za)EyqEpSYf*HC7wR+%dDiOx|h2K?2XLow@HvGP&tGLEH<`qi|i3q{9h-OUD^F#~wY
zXWtxKLZPKn-Zm>`Rw?6)*q8hmNYYhEN{@mhUD!5DeJP)MNK)>f^w^Yt)?w2_1BGha
zHD+WXJ7q(=@9i|zj!cO73x%ryGz6tbuSJsJ?QrovEv^SGUjFP$j@wJ`gO)Sq4&*0Q
zUzzQ^9Mbin(l<v5Mr}hRcPPQA(3R58p#<jwl#obUHd&N>&I_Oz4V-ybnk?jseSCa$
zM4+GAw$N2<V?nl2-GnoLRuq4maJCGx+IXztq*A;l`D(U8%4!#qAJv7Fg)T(vElB3j
z6Vwy2CwfH9(i1)E<{owPb{i;(dHIZj>n@y8EZ2XLtO`JMZn=pfL45dVyC~_i?=W0V
zfMl}>my8HcPjT9cR>5bsdqvd~Vg$+CD=G@8gdQ!e=1mPi`Vbj^*uWY6g|cp+xwzN{
zeLINJknK7V5?HaTQ1H6CqC}yrT1&0Pt`TKH{&|8@JKC9u-Cgo_2PYzFN;@2aQv~@k
zI&ffb7nz8IN0*FDjy5aW4YbfU4^~gcf|jCzLm{k4Do>$=2#OVncS<BERrFpRK8qpx
zO8^vX^T{p&RG+PXq_I$}k2S;tE&+(QD^GWq0N{^+>|MD8fbw$xK*C4OPTpvHF|Y2O
zq!`v6q^;w`xN8p5^7<{yH4W4{&Wto)%2%^<Q+XxLSK1)W^X<)wKF#y&_R%cO_ZQ;g
z>_VIoed3>8t$KXsP6i?K#BJLe@!5|)##hse_ZR;gOnyNWP8&>ySn1g%gAn%daa+6o
zAA}5km+?hE69YCgF_&>~3lo?5=Pbg1&-uRl{q2`0&BBh7fmP_!;Db{{t4WrrI+D!A
z-3N${MqJVwi7(>L1p1A7d_wX$<fmz|hGasYe_>t`^RId@YYz3oeVgBAxI80_>z<n|
z^xceedkM?)`UEj^-+eG`+>h%Z@1^MC;~|<gYguqA#7-rS7p<GvTBCxKi}F5yNn=pu
z$THY6YV3=v($-P@G-|iW>1LX0J94~l(yQ~+zIVpw6{HQcl^=}zW{vKdzNx?UfmWw{
zK8oe?%&@T9sD0;|uUXjpz_ekUheeXCqV8;YP&z`>iL|n<=Sf((p8vU(>qu|q$-+EC
z=$z@N_Jr5)sV%2)vA$4-k=iqVKM%)cnz&UzDf8!6Fs#Q(ntBRXDVFD~9};jn$9Yzy
zlf!nbcABqperaW#otrY7^_HzaY15dx%Y3INB}%B|bD+q#FXGqh+tXj2iP1W`4}bi4
z%Hz(U@*xd|?&!nt^UvYi)31^$M<T5>t*IZVho9dEG4ho|YtrO*+Iz%*MQ8=D5XoC@
zQqE6~_nF=2Fqs?7A_TJg_Y~19Ym=myO;OAi*w9T?F3Z_Iiuj9>ud`qqO-TcSGU)Zt
zoV2nzq%3FKP7WHN!=q`J=9OhSuhO&nG(n+D8}zBk=0x7u!DL2->tf{V1v+1H1$)kM
zk7&**58WQJmT*mb<PKhcn-b&cdfjYSutAw%B|qbYel}2aMmY!_#m_fTB%dVZONzLv
zL2yYW>pKl@Kl6Z5IfwVu4W$#G(mi$Ky-H{=)lD!8bnhX0rXLE4>}`PT3G<N?{znyp
zN};n;F7<HF=U2CsH1i67QEamS3VcUh?^$kA!t_z(UX09RS&I{Yn(i}|X`ND09lI19
zrrhJ=<jT-0K+dnF4&hvtGtv1C)Av`JcP6h}obvL_(e+eKwUwoQ+DI{%Qu*o*-v&im
zt)hVe6V7{s8j5Q5uMhR^sOC1H?b>53k8c(tQsfM9R0v!b0#68Wdn!V5_>dpHYbp|4
zb1J&5zFH3&+7k1BAgNxGlIAHLS#^^h+sLVq!l7U4u}FMT*l0mqP$)gCuRSTubdz_{
zN|GBw3nH~tR4X=auUC4-R2SXi8IJDrZtx1vs8hVAz953*Zx?7$2)|xMlS|l0!xnw?
zCE;k`t*t$oJfKG_T2ZeT!2p|Vkx#wv3b4(6gQKoF(ZLdbjf2-J5aW}5L+J^QxG*0e
z7pd{YsVCr8w*-v?I~;6in_iaft)o2z930gHI>nI-v<8r`+SYIG6?{^L>?Ek0&nwC1
z`f9_3pu91X3nVr(H?9qIs9L%#WMrtzc}bZoq)vxE$L1ukmQDI{bds$kbZ>QWZV+6c
zAz<?G@bC41YU4sRqU!kJ6)g%71?&`U4xN6c6;XNbw(b=;JLf&O$ks5V;&fbDdA&2&
zbrj>7p&o@g>Jz4(%$5NtCs&*dyyBd;h+?AxfUa&{3s6VF;1>JzB05>zxe>+;EzIsJ
zK$HP7oeH>h61p^)6r|A6W>Bgbsm7$`uFn)S$99u{dtd9V<TwIcMd|`)y|D4^ls3rS
zVuWnrA!pZ?i@VVi$qIs1j?$Xkv0p-{$VL0H!rvZ|@Qf`fr78$fa=#>_jQ8hJOA8oK
zGv;(8a2(SI`KI#NpMcZowL=;-ofbzZw4Oa&$acOX<(aNQ-C-3Tro&k>UE^t4cYX$O
z?x$CO3v+<90;3fqJq3`K7bvT|11QS0-D{711B63+^>I8KHYjz!0FgV8JGKgBzXJQ-
z%^R$v1ZCyEu54b8`h1KkLXceAX)T->60bnIZ>L}a>G^fi_!0=}D}XFni{e|0NuXi|
z!O3DWX1M5GFlbBp1<~CRgBBF;h(Qa=XE11gG5eh`XbV)GTK~?Czu<Z!1}&&P9D^48
z?tno{inn9XlHzF$T59lm4Em0`VGR0?x?v3ZLftS1Ej>iv-v>3SVBY&Zkl=bJ@J&#C
zSMW{haXa`XC_ET^1BzeaUxK~mcOp^u&^VqJ<iCu)fm=BCBd|9UoJu-92zvuSMUmlu
z!PpxZM<T$3us1>dLD-w%a!2e<P(FgamFTGL?bw?jbP|xMVP34*{;qI)DbQ3SWHom)
zYenqj&R%8*Q!g7^{&&@7slX}I5pORobFj1gLh=S&%_QpToZEn_i52HJdlmMUI~NoE
z4141szWD1HV-z)iSIu1Q|AONQu(bJqvRH91u#_}X-&$`!?_8*!-^xvAp3BjV+rI%9
zCv8!hk_?+C)YtMiV6(<8ve#K`HoyhgEM-<8DJ1PjVzZHYeaUcyW`lh+G)s|?W()oo
zg@>~+ZE=V#NNvW6(Me%8Behj&bELM8T!GY5l(E)y=iv-e8>ml=u`_BzYKt^~BT`#_
zO$u*+-9DOtIzO;B``$l-&E|iP694}lHY@h^AZ%7}S;uCZR^A7n^|b#Up3O0e{PI2U
zSt?V|#7$7f`Nm(?w1P={Lt>}M-dXZUCl$S-=66Z&>Js^OBq&fla7So+PVSDl_MGe)
zP<uZ0B1C)jXo@?X{J#Nax_dX5(dR4_m$B^~69X|cHkXk(7bll}@GLI^EiIR>@GKyI
z-v~PBr16NxF*fhtnm?Zb$m!>x9}B>&AFua8ud;t)4EhF&e)K8g{>fWRbMzgiZZ5lS
z4(nTV)GO-sC5`L+2o912m4`ZpasSu*?>orGf{8BOrc3?T+xza{U2i(E5f>W=!BDWx
z-`{rWfkiRSNA<{fb?Ey)&6nLDwbWjJ7wDAkOQ8AwwW&rvXYe8@c;KAK2t0B|Bs5Nf
zVHKRgv#$W6&e@ZG)Oo8g5mrt*uNsgWOuAadqY+j7=@f8Qt2o0PYptZqB(c*bGHAAv
zEeRtFdTG%bM-v}F7hflW=z?0GW_MEHw10AtmYY*(ymc~-O?Z{B)|=7D=RqHTu{Y?s
z)`zyjXMseE#*&{%e0@NohLD0^OO#mz$wm{|O%OaS$3`VsD{Ero$ha8S#Kt)jg>S`1
zvyfn#Aqo#@d}4{8>uGp32iY0x!;j9vvDTD>)^%AUTJj{_`<#!qv!osO^JoF9a$ifm
zo{MxEWVcUq4M^p?Mc^fmOR;5td2B{9b^<d_AdO|&NHV17BOfdzMW+SZP9QyWrYEx#
z2(6obSX3EXaB0Dq#V8ze(56BL$MZmx9D73`mXQzaXCVWYCV_ao-9~3k09lSe-lJW&
z2h&*qrg_Q7lx)3^n!UqU>*HwTY|;DVA~AxoSj2Co{C7P7oVu%p{S#_`5I@g6mM-f{
z#`giUz!(M-Pyt4(`+jfBtfHwMGPWbV;((qZkj-~10sT=kGk6_<gATK};Mybs%HGdC
z87k4Vfy-Tx_-KIpC6Z!W_9s=PssmWuXxXUSqHa%jG+J#Fb)l5z7!xh29)~SLasfZ)
z5|$Q%ZY{5AF<NJPEmnPh@Y0BG?OuGZh0c~-vSe9Y9MMICx=xjJe4W{9@2`;Py>W=O
z7K(Y9F{6`1wnOQr%0sF9x2~3)TbK!kY>kgYg^SDavKq-;3tAhlyhm4BDEGvZ;K_e;
z%zI^deijl}VH7jbCwKJ7q)muSfX34VpJ7HY&e^5DQtpzpsP&b9t*sQLZ^(UwgFW8Y
z()KZ<H6#q>X=bU1Wz&6B>0C7<F~~meOPlkOO)L?gmoB+`DRz-e7(zF_d^pGEj?~vp
z+xklI>Y2}JL#UK+N&02UUKrn4n7rjMNDThCJY-7+8N-0{(c0z7=+*nuGKXx)HUD%j
zrw!SAMWqO;L#S+jRDdQ@e^k=9Rdf5^#bcMUr7D}U0|e=C#izJD3|+o(*`mtPOWT`d
zJJXvTFX*4<(JnFiQP~sbh{wbwUWQWw@g;5MJh`DSP)Z-_t^i$2imx2MTxVc6bPm>9
z1KzJh!b+}HQg1jAIoy#oDG_S>(^(mOSmvu4d_1>fTGSqY4N?e-vy2%{NuB#bhP7Lb
z3i8hj8V=47)5<|pN!>7LDouDPV64ubyv>i(U6p1cPH<d6cgrbTEn#W+Zo#HlK*<Ph
z$)?D8qt$*RHpN;g_^i>{hE1vc#!}pfO;yr2VpElt71>lJ`9ep^kWPCZCsSiN8*!<z
zlFg~qSlt7EOsWFsI+0rD=#S%36w^r@F5pqsj1iCGeDB7GT{rM33Rde3Z9t?d$r~}L
zO3Q;(s*<)6m#UHYh)h-Q{yJ=mbU`HYDr|~4Xj@FbI-eqlT6^dxHpeNZyp1b2$CcC@
zn&ac0&2gnE)%z=qs+z%vO&JvityARJWK<~FmZG0&7c?l|M6fFdO(k_hMpbFL!l<gZ
zS7cO`<U1HuE#d#bsE|kr2`^w&8K;XsS&ABV@hZc);U_T{O)tNV3Wi|VRjOVO@j>PD
zr9x*Lr6^<NFDF;FG>Usk|KPmLRi^))&sVJ<{{e8_r<Vc41rq}{Gc%Wg>OB*eRrV~!
ze?5ig_|Lb$A7o}$3U9)n;b(k%c{}`ZsH}q)Wc*|#2rbR@_2(gcKt;k>KUpiR^wXdJ
zp1vJEk5tQC!*2x2DF#X!Lo@>lj39E8F<NMA3sA@c5(|PXB);;)LL-Afo_`r0qWle0
z-`I!y%jYO`rH(9%rO6NoRIXBo)H_g#e@~!H2`+&=wEp+!_i1SP@~KfbUMLA8b@eek
z1NaQ66f)2NN>pbpO!dR7m#YXd;zp@)7VrrnViiDE3$h83D2tlSzJ)ym<wP@5mapp=
zq7vM>KzPbi2R7ga=%gU82cj#;EwG*q&==GJW?XKIrkNXAt+Y~PD<X5dY;&qgf0bul
zK+6U|W-;nzonz$%AykfEnA4{_aGpV)Q@_NyQ_<HAqeiTYp-U&o>Dud*xm@!jo8C3M
zD6%ufWiN{Om`&Se^Ppq&t85bNVQpO9;%lR3gg)6*vLQML0Tq6nCY{h)w0@~3wdBk?
zp-7FKLDoykh8ouho8g|<<H4BMe=z|v0^DNep!Jc`$0Tc+F|XwUQ9^;P*;jBX&)M2J
z#(>s=rfe98uW=4rznaz>UHI$lD`eccl}@EKa_H*YDa}E;S#q6I<#}CCHDcSS<?k_A
zWDerIx^w$gnurJ;L1oQ}>tH(%_Y`_USjCS?A|}JpyN-89WrAeSjkCz5f9V}^dJj=u
z`=F!IGwsDp&E;Wk*Ksphb6y*6Ol5VQ5d`U3v^^CAiDqNwP-TM&3+>C>0EZ%ata|m~
zl}-Je<pQ#goY_b9TPv$weAy+xph-B36Ax!w@R|Cvq_uFk9F4@fZ->$6l7&q^*q1x>
zBP9-P4+bIoAcW|R&WG{be=A}Db<Av?YwM}$S+x*MdW6s|U*dcn1Mo&5E+Dv0e9l&d
zv}Pf`Iy<!nC(!CUz|UE-P;LQ#G;Eb0$RJ^7w5EqUx;<CH&7mZIWoH48m6I~2sI1A>
zgGrqmjToR$g?goo*<so4Yv(r7LJvtmQdD$KXog@lCF}T=gHwZ9e>iNti<NiM=a!jA
zIC}NCoeX3XL=HpR>l;mQK6)+L+%!FdJd#EV(e$F@O<KU{6RsHzHdBl%M|y5J_ec`h
zTMO%9<ng*>c3)E9Ihj-oJVO|ATCN26ByViNSqm1#7CW$#Y@G(Q;+xXX(1I}9LK7c{
zp!>QWc#)uYe~Er?f1T?*&C{nc&g9@~vi)VrYIwP*mF!ngyDfWDs!KJVn%Badf|cMN
z$Jn!43>+qnTF!{W=nefs)<K86_@%BrBS@$7ym?^idZd2IA0Q`|-rnjDW~0}(IT!uG
zoI>yi2sI(qn|?QYz@1*Y_)DF?m_5CdG|5+=Nq{!}^;QCke}gUM?5<hQz5E$U5Ma!R
zG3bkyxyOeg+E|AgsF$eLr|;||QW5QfE^?+pWNy{P+30gy?@Q_7K<R3od^zg6-905P
zO5Bg%ZQJtxRT^e~jfQDMoYAQfv3qHlVs2nTeTRmP+ZvQ80IQDboHM0Ns%CqyITk^y
zcHN`&V0cSsf7Qstl^DIeErini>n(*s@E;R*{4~Yh%(V87rvFW$Sp^P_F_wQdD&>O!
zhY!Dh|Ms`9|5Abn{5On{6EjDOcYZ(9t6@n1iKZXv-E!(~!B0wfhL4vE!d8O6mvOb>
z>vZ6GI0ON7-F}2tPOqG#u5T<}J2tLLzqM2jBEiyHf2RA<Hx{Pa3{j%WMR*HT^4Rdo
z<IAY<8lL{WbmdY1nGWBk!{>^o*ZeQki7h+76)<^Hb|k6?v^IGjjW%<S!8!Yl#$L?a
z1}v>XhCqSv-rBGjnQoP|v~<#zCMH!Vl&11Kwr@y32%7LnDoKKPU+08XQ6E)?G{M^b
z{`xZyfBVjZK6Sr|%!ATiD5mtF9fU$WlZs~)8%2xUJ9jR|{dCmSB_q)ZHfNaJLz;KZ
zxaKsOfzMONUzGEq3Ho_2HoQVaxC5IpIChwDm?UVRY}Hz_0OUi(pm-a%rx1300E_3!
zUh!)x<E0%$DuIPBWjB&3Y?UI@*I~z<TEk&le>^yIwcpNsE%TM5so9LCh0n5Y>Et<O
z-><v<jw#%YyPG_2DEDNZ=d^J?3!|p$e!cFpGIz7cRhE&nEM#xq$G!!wY3GQN@=@l+
zvZ+zsjvEZ*T*6)~mzJ>=4a~Sk8jGFdsa0_DIOyk$#pF^5E<EbCKg_-MYC)3Yq89Z0
zf7Eg;wq0<Z)pj+Fjr^KJ6d5<Jk=|G%xUO18;n;@b_NW17T~D%e==PMUB5e0mj=KF(
z+IM$f&366i%KY_I>RTJttfxMoTkMaz%2_=~Sg(1lx%1{bSNwT%&dHzuHqCqGgzKTp
zxXy*8lAlQ<WIJ~&E;xcVJ4IPLX}f$6e*mnFM_R)@0A)CxFc<Cth?ozm5b<R`dz`g=
z*>yJnoE8YTX0&}Jz;RwYP?OIYCTQy&s7SVSBUScP=?%)cJLKc!_Z(hwj?Vd3t;v;G
zRlI!GWz+ml@*_7cqH;ppn-+1;0s#txceuPUmiv$D@#fnZz%u%gwL9O<2-Hz7e`dMI
z-3-qC5NV{`wDO%7FoG3$t9Ibi*Zpv@);He3xQ10t5dGV)UCMB1M6>`z+gx1HE9Hds
zrPDTr>wyxz%$*18#rqtHa;zflCWY4d_Cp_7`%=2BMR48j0@O9*^-{7q94ZGOG1=Rj
z8ll81!gcDsD9bk*=Hh`Nm=Ol3e--WXXN1VwjA|U6)pK)7<S892<~+@E>+N$%o|8BJ
z%Q+tq%>Memc7~N1mI{z6PH>(J)&pUr8%Ng@ptL9u>x->=cq)8SU&Vr6y=d)>GcD$h
zN@Y^dwX9mn)modf9DKknb}UuFy6W1M(NY2N#@mh2QaeTy|B{qP5xZxlfA5s?4Agd~
zJPEvagvWfraFk|lE*C>#4zi;Z>nT~_{4|qACZ$^z>aP!)-U!Q9HOeE0C={U~nLvsZ
zXD#QVkR$=2LN=eF+!-UcKMFxOlu--RPPO1%I3ih!&0R}>E3oJ!yG&2D)FVy7zfaJv
zFoOkst%Ub<FP)!Ldsmv@HFWCT77?zm!h*YNErL)ezYwu9|0`5}VE7lP{CtM@sQiG-
zdy*_q!riz>8!o|fr1?LQf4-Li0SprZHZwJsL7yHI0y#97!NLe8m5wc)0mrv20WAXo
ze}1ago(hb!>7vCPk%HK#i!3)G;j_%T@bl3|t3ZA>THpvU+{^A2ip3>%s)p0{@o1$}
z)J`?zxwnWtrWEFDESu00TGC`&bcdYSlBU_rIrP|pdmn*{yKl05m1Q0)E`F>b-yL+8
z@CWy)W|A_)CMD&IIFmVyrYgZzye6SofBp5?-Y#ojbv@J|)>UUk+PWSaU-gIaUE9yL
zw?VgA4r|3)V)DL<Ap2|36xhYcy5MpO(pci+<dc^#=RK&df{k7g_THZm$adB2ni3P^
zbk(-P7AZq?qiw}{qda`XU1e(7jW(63k{$Pyso-HtSwb>-(9h&=AmuGCgVql~e^lr|
z3`pft%F{Vtx0v)JYg0Kjj<kB;KGv~RXXNtwXV<tUy;};(nqecf2V*{s%ql@%5&+kK
z2wAJSocyg<-qB3<iiJTpBsJ?s5SY>C5%Ttv459jJ9%kz!bJDHO0YMrR8j!}ej9ez1
z^KQ%XG-RB#KLkv^(xv^*Qt_eLf87gtj>z&mGpw$3lfl^PCL**>V_}GE6i#`<m>j6~
zfX(412&V;k=y6(H;`Ao=mB2+evVaPLpFyK*^YK*b2O6z=(!)9~t+%xtjx<~Enh|E<
zk&<{EmoEU5qTy_Ov3NOLU#@7tpyD}8ow0zA4g7Ff2*+NGv{AD&3dE2#e@j^lgO<-n
zaXcy{`eG(|v?x)kOZBD5`sg3Y;V>OEP0OlLA=P&qcB1W^TURU2l!jIZ<wm7p)AKYb
ztS;|$Z3a&>EZuCDc8uL<R(mBOB%_EcATckbB|lh|R?u?;<7d)aLD|*;bK33EGh<v@
z!%)ilis3`Ss_nm9%Ra*ue@bc|McL;e=ZA}BU#)=hA6pBD$(@&48lJ9=T4`v4%RONW
z_0P37JPuIfx}pfI;{}XTV%|P*=s3*wx(5Pe+DJ%7eX{6*$Q8af@@<;gQ7T&flzZ4>
z(=w!sc&0oy^a?BR<a*Y#c%TlFqEg|G@4T$Qe5!S0&<w~K<#0BKe}s}fg<fCpQX=l2
zx3>V~6>v}reP9GW8zdS~2K`Lp=PM*~NHX|RqCHLRn3x+jiQsmhZ!N}-*f`|QDej4l
z<M!0jT5LG0p`J&G9^rFEuUBa}A$_@sU6tEfBPG|hwLJ{N@TYJM<2&PA*7AfU1<nna
zX&qz6uErD|MZr;He;I3y**L1aG1mem*P<)RcKf{C3iUN@oJ9JWf`6;T)BvYz=6xHG
zBrYnrRyAC+qW1lY=Uh~NF3`vFX*3sKsOi*Wn;&0o1f>=(oHfFpHo{&o;`FA!Kf6WI
zD$VWR4cRT?L;(?R#BPyCWnB1(^p-H~M*No0vLnGIB)=KMf3=L8%*<j$CLo@xEf3d_
z8HMDT&|W)mUPN5DX|tB|BIj3dUXC3G{7m)-<-EA_L6pY!ze2chW6p~!#mGH{1Zy|H
z$v;2mzW&4nmPjxTKhm?6Ttk8tPg%vBxUh}{>wT8f4VkzT2?l9T;1)+n)x!^F8<;L`
z^$hPMQq}_Hf5eR<OP8ZIvR$aHvKTk&#(V-|X7R7Zc%|`$_1cB;BF(D|?w4b{#JqN9
zyoB`kX1tL1YftY;c%?=$XxNeP5)yYLyoBTh!b`mUMuZm^Un&gk!y6G^LhF@;mymXI
z!b|Aef$)+NuP3~u#LI-2MDXVcuN|?egx8MPRKjZ|e>Royk`bcEv1e<z4YAkmTx$F{
zw|kha`lyy$Z1Bn%G$G(Ex~7_euAt7)^3H(fJau-7q{{SJELS>Xn(tG2Zzn<uDdyIq
z*+++a6Y@!@8<9_lksW5>67?bh#WuYU*#zu&Vf5qB4ag>N#6fx^t_eKXc<4VO*Cb54
z5!WQNf9%LL3CVB9H7(;Nxh9Fh<7grE^mJr9tNNDbset<m?eHe&sif==eV$4f`_|{F
zg!*gU<ht`z!kBkHPbJ-GpYv2g>Y4LYz)=jt??syrREhB`PgF^*KX;^RqwajBN{spc
zP}K#B2{a?-%`ZnWiFxfzF$w8=Q%tRTsh?c*2K@)83Hvvf(dR4_m%*V56PI8>DJhqD
z2rZa@F?c!t{rbm)%*;yRO?V8q{;$*P!+#Hzb)^LvKajUVD;oa!@ep1hkTBK{P6>Yb
z^WWjy!=F9Tx|QS=>4)U-^ZU?^e9hQ-$=FF_h$h%6BZ%B!j2Ak*O4#8SDT!snGhZw_
zGD-?2GJZZgSfofsXzhE745B1P78y~DloK_7fq~ER6WO1PF#DCVgLuvzWhEzk)!AMO
zZLy!NXMFGgYl&nj;gos04(U}q(F)Xd2cD8ooXLY8AZ!$Oos4|lfcbP+#ln(LyAidW
zkuoRl0&*yfn#mAQq9yds$44WB_2^(}r`_kOv=v}=zaL<S&kl+R0+pRn{Bi?Dwnzql
zzo$rx2rBf#YcoS|zs@Ui0+H^CjR{W0-xC{SgeScf8w*0oxrNBF?NK(jW_x91#rBl-
z!ouv-gVxhF-=(|~IAbjYk)Z6sgn|J?g0`6xL2@hc*)~(7=M+RpXR|ZMU?qW6@i?-N
z8w_LniPBF-#@?`a8f^qx*HD+z`-pyjwn}v_7-XBx-XO%q=mnngIAzvjMBFAeol;DU
z^PtU-b4=dDrz^zX014>`LnM1jp*Fab=vs7fOlhhLf`rRX+{~&=O949y#PgVVF}n&J
za?G`CQi=qD9|wk|6MUP9VNMeBLqsfBXT_h-2m=XkEh|bFrL4gT1i$|FDJ<oGCAhrA
z@)DRX`Q_7I?Y$Vmv=k21U;sHGs!0Pil`Vlf1(}q%$#c4gM>eHLQ2e-c9kq1{ba@I-
zs<@1-55YXHuWzf~ZqNk+sRa|<g~3b~kZ+d-nk*5~YfR9L$Vx8UfN@eg9@b=>3Av7E
znG--uZ<_AR^Z$}Wn~B?|Vs*uTV9hM!*`KFtCkSVGI+z&lR}=M^gkMnp?8*sII8{$R
zI%Jm}z&R&f>RL*d4kyg3&O3V|6(4tJAS|QrS%WVN83W^!gEdz&DSsW0tVM2xyz6Y2
z;2Wv9UDKjZlw@N>&-qhFxp~@aU<NIEL*AxA3rlJPZgquHLa~GGR>aDG)uhU_#*rL`
zilr^;+r?_RsIqjzP|X{-Dn`uNNeqE5DNd-98EdYXZFrmV&Wn~vp7ZV*=?xH%MI$qc
z!Z=P|QvQ@{^e_~cjeOIBhFroyvzxWb^vuTU#us@5EBKsg>*haLpPf_T)cJC{xc1rT
zir3<LjiW5gzzPy&mmS@IwwtCT!4B4v%_VNHXaxIAHqsdqxaCO7{iz~mGY_tgOpc#J
z%3(<brYTz1QS_Xrj6z}HKvi9&vP;xaow+sBOf~M_W|_QS&Sz5{3;J1&<jJ)yz%ozP
z30s4VF5<pKAqs2jjw>_M)<o7^V-Ibe6a4o}ozuX*ml`UKLTlZBiM6wHa7nX5MW$^}
zyk;ZOc#ldQCdZ0nR3efpZ+gxMJ|(1zBi$RswI?B+?t?hjrB=g-blZ@O0Kz+V+d7JO
z?6!53uXNkG*&ozxLsnw-;oY`w*BiTS9kqAwwsrg6q1)C|yuI7jQ(W)1^$5Pb+jd86
za<}b{*yL{8Mr?9_x2?AjInkt1j#+cc{hcUe|Ek`j6SXBL3VoJ6U6YaYD~+GG^kHUa
zyT1j1poAW^rWRWsu{hU<OV#}COh^#gT;ze4d7u>@KB^I*4GHjOM<XByn2VnlL$a<M
zjZ@>&3W*oBj0R7+N{7yQtCmJrb!(~bMVSf*iciH=l~6f<#0BLI<_mBuSI3f<&ltOq
z+o{kB>G*m&*Bn^+Bk99h9$ZuG-5H#6nq#K#GdQlGJ>ZoLuDfq&%P7b3MZUw~r?!mA
z9UGHLm1WqbkJ=TtCw9t`K%;uGR$3DQ?og_BE7rQ=i;3SY?FvO)t|M%1SD^Amo#Ae^
z5^(4RGh;=6=KmRZ$=^+`=!fofLr?S~=|tBRNmI9#T;UTI_B@tjqFnn1g+j82Q9QQo
z@Fty<T-~FVd3NZIq`$_seg@pMqcDe7-qrnWnVCM3bnWSLeQQ!l7=6+?)9i}q(GmOB
zm57K>v+lyXPyEzgsJx$I6eZ<M*gCVzQ}{L~)-2n9jv4rvIZK}HIBi3`6x6d37K5d6
zd0jKH+&LNNH`PJkKS(RSJBIasK^nXj9`zm2zZQ(adX9be=3snj_Y0;ixjePD+x>$=
zE+8wMvmX?40XexaHy-6uT%BY&U1)uNLoFS|pRv0ZN^`B;tym4Ed#ajSLM();!~1<-
z1g8CeUbo`eTL-o@^p?DOo>jZ>E`4=a>?O}V{iv}DBadb$Z?QgIQ7eFTZ;)~G$obUN
z=`ZGStHvEZod6p{M{MHH=U68!QmhT$y~Ou%ylrb+lXZd8Di)J=?231h=Lu&pZe=fM
zip_Pso}FH60O#11=8STcxr9><n(sE&JYORhp^u---!INt_Go%fnCtKCj{gT`_NnHV
z(dR4_0XUaY@i!C!Fqe@z7bliCHwu3rSVg2nJ%GgkyR(@9y*1cd(Nodnltc6X52c>A
zWm{uUcDCrj<MBfjpC3uda`-;T;mgCz+ry`e(t{O_z~b=sZ2*2C`4=SLG&EQMVH6B+
zKZZY^q(sBV3m;larTDP?-|?AFj|v>W0mKJC1voug4PgwwqcDwCCXG@27M_3k419hL
z_j$aRMg|H@4|onxr$66*dytt~DZB}P!>PUc{PytALuDP6AmayPl#td9fBkp}4>XW4
z)(>EWS7!M6_weoE_m=8ZF%DQU(iozdV$cX8$JD)0;ZahIX2CRztuOhYd>!t8=7SYL
zetd^mNX>_Q$?s5lN83}-S$G<aY&y%!>Cwn|mj@6JAl5ul(vHx~SN-|<eV2_9EgOHk
z!&E<Ayj)Wdk{00GNmO8+jjh?}6m$Wz3AeP3mV`FgQlrnwZFHd;&XlMJZ3g$$J{>qy
z+EkrA3QA`i1xj{Fg<p~KoK3LiLYskAyi<{otf!(MI%!=@Y}4Fq;uun_t4C$&DZ51Q
z7#HrCr;(qiS6TckJhkDeN!l(R7sr2(7AQ2r8@^mc*Msp2)Qc`UnvNi<voyLQnmBOO
zXQk#&5J!woj98>L>6r5uq>guBd>ZCRyx7QLrQVIf`WD}%bG>?==-iHT>E|lzQs;{}
zuDtlFo8TD>oAX&3T`|K=_<IQeYA4jIH|i|OCC?OnBEF&0QRq}ZDEy~4&sl$$$(N?m
zr00ih5`jI)mLqoCza@i>n!#g*xGBD6VjICiUI2%ELo9G(s=k&I=@dJ~zR~mL80%!G
zV+lKyu6a~?%jY>DJ()*YLYxU_5K(^99+9VWvFyx>re4`BjhY~sU6l~~)}JASFF1Je
z0UxuarC{NdV}Mg+oQbJGVoHAst53dux#}I{9484=@jP=u1%*xIah@S!y4fsL$e=J<
z&m8V0$HpWFM+>?vFP2!zUXH~h@_(8CC%lOn_|kX&!i?4ewX0RrYZ8sE6{{wIUu7^c
zv*2;U;Ceuzvch}aF!+3_(S-g)>Z$L~)h`YfWF(znuWJi3MBzdA3$lOqyrV!P(1QFO
z93tu7j>BRNH^YD%N;41|ZOimZX*Q1Vv5sv<V9pkl?M4&C3fHYGhJt*Ca=NXOHazTz
zbc_xHy*VJAGFmuEN2OB>UVpJlV<KGcj#XO5&yXXbpzCv{j|^%UWaq;vQb{L^yFGLy
zDS!i{owZvLq$Y~8tEzv<(OA=RrkSJY1}>nBc+kuoXY(uh@x3HQ0`AeEotGLZh`MNZ
zO*+Np12HVG%rcP<Csl6BjrI9P?!01`F<)P!)UP1ADMxc`X<AxtiAP)7(KaM4T(kY8
z%Bnur<%T2(@0OZAmZb<eGb|XoHSK6Zt~Ktp4#2t>+R{g*88d$lUw~j~lWc+37BC~p
zMMl_3vX>FICj1D5MlC>nBtmn++fMp10B!kJGjj!Kf_nip2M;XN{O$m4gySGunPno~
z6rk&!Rq4x}JpfI(gVMgZ>jSm1LaBuTj}YEKOJKs}*IXkB2>bE7MisExoxTyjL0f@b
zy){(s*iS02#Rz}G@6b*v<73ij%oK&s3yWo`FrM8mAE?wBMjT6ar9V70<Ay05al_Zn
z_(2Vy;_BUTb@wq1pF$9zZ!lR(tpO}RAMS0y83EW%-f3-sUDe=%-R@%awfsV(m|<x=
zBtau%woh3bGZ@10O#^mEuIT&$0hRPZ`#TAkN|-8REFOPKS@JxVUXhTiqK`Vcz0g25
zv;*9^VBbZ7kW;ZGIHtGckAdM_8EpqQN0N0drCbzrS@3M<B?~D8%Jzs1$BO%On@D9_
zs%iPt&X+%X%pqfyX4|yfy^gjG%G#-<Kt;NBb=$eeL=QUWqTE&7TX;g)jIypRIZMJt
zKV7gz@AH2qMgHgeOUmP0Ws@b*Tk`vUH6AjmI@QW`Fauq!U|m?^a_ua2VZIdZxy%?f
zSr+M&X6kl?%*)@G>hP&iJy#2u+461CY~BvZ)SgG%m2H^(WKsiWdM<4#?A^S|+~-ZX
z312khM=vDHtzq8qS@l4&tV=bG+a7%0vG79WV|stsUwIv_A`n1$^mppt5nQZQZ>$>h
zEfoXABCwgcTQaa=tZql#x&t0}&qMCKq|CcgDjZ+Hc8XIG91sdU3cb*O;Ad}2Z)CVV
zwF5Kn%>KRj&Te7yowR4Z0+iEzKr`8WGh^VAtqTKXy|APuTlR^suReAYtddSxty{Ut
zcN2f9;=uF9wxh=V<*nV=<_iK?f@xluN=r5yc-rl)ff0=3yJfAvyBv@NIZHO9?ueFP
zjL?eiiI$)oV+Gv_EkSehLhv7;Wi#^*v}{J)5iM_O9MLiGp*bKJdLyh_2o@ppm)kss
zVV3OxH!8RU&?-0Q%FQZ(!@C1G$G-yeKLda8+5D9wH-5ZAvJc~27)frl8|mj=oP0gd
z!6|Ko;ZC5Tcn#ureE<eG)X-Nqz8FheUaw%j0RaZvQ5&5emnfl)DsDAd#&3-FQUvaI
z-QkN=j}6)5AUfNO@#%Zp76GpJ#oULHqn4Q~-&RrnRv@))^Kj@pD|!YP|NgSm5{n(L
z%C41)Xg6o(j)^~f#yT#A?aBMinRETWb-F}Rm(k}e6ahJx;lL#m0XLW7lqx5e5UD5x
zmyH-Ll>y3^BN;7Wf0-hjGO}#@^W)ppjePk$%jF@Nq}v%j#%rajMiI0kXL^dL@XAlR
z^UN!W={U+DcT<eZ!Y@;v$Dgtx^+o1}Q?ze<(wEbt)hJIKZWau^SBUlo!7y0mXb?;g
zr1iOR62*L8i0I>+$njc;C_&+Q3kz30q{56yL1-Q#o?y&Vf7#V#Wip?VZ!knJaZmY8
z=B@h83?QIqvjv<sGn&Yj3l}p@SvQ)m%}!c4*B#qvDD!!yHSzZ)@f5XD9AJ=}v{%lU
zUhZP|YuY05zF=KXl{nkeipko@Q0;Xk3};~C0?aF2CH;h5c?_bm-VU6ZU3m;TCA{@i
z67@2Y#O&k+e@eu0ZLMt3n7v^0&6m$t`D-{64Wa~|s_&|Uif-F1HzPObz1YrPyvMjc
zgf*@D1)gy69!hw-%wZ#iHM6VdwyP5d<2j>z%sa`3DWd?qm6H^d^uGB}lnJ5Z<-^5R
zBw{!*J(ZndmNDVA?A&a9V_sr`j1|aqc}ZXx!0hLxfBt?TPca2q!{5)yIe)QEYEJ%D
z$uGA@el#)?dM5WGKSebSV(q3h-Se5U#Lq{3)}+LVmo?G0N(D`*b1V^Dg{AsI;u@_`
zTj!or#41m#h?qwPX_l01S5MW5e$E<c{TAtSon|Of;A-bIAKI>1WUD@)%l%{3ccXi*
zdo@}pfAq7ATzK-UJ#5XIZRFgKB<R^nU2G^((PPAxT%MRR=7!Oo7;CIiLHpjaGe_Au
z4rpN=f46bO;+kZk-%X!$9n~;%FX<|7O^^c8G@qej7rKvB;Jid3>USwC<owPj#EK@a
zv9{qhXY7LbIJZWaSi>wL?jV;Vi3xKRxb&G^e+ky;Z5B0|ZEvyQK0giAyFS}yG=Iq>
z+^<mAg!=sHMEta<)^cUqS8F{_yOkM&5)k+5(B=4b&dC*DiSV?otleTrN<udM#+#8U
z>!U917Ll6Q$tcC8goxRmnd#&s+HTs4uQNZO#zGAH6{cZ;WjzJiW#PQck9D&of|*0^
zf1*@h0wa&`vXJF^+FGWpBjx)sK*VE`9IiT1_W*HSq#iga{oeyS1PToG<~W9sUsY=x
zW0XdB$68yUtN_MZ8#>Hp_#CODDFo&4SRK8$%e81OQC!N=^)ByJA(KTVczCA@nXDXW
z>7Z$9_ojAvZtvnarz}9f>s}YfTLst?f9zGfXyzu=^#)GuhU5;&q9(qfes?T}!W)D^
zFFC>u;-h^-Jx+2aU`O}rz+CZ`bw^<g#5HYfcP#t~QpReQ$dypy(ax3m;5qk?Z3ISH
zb3<sm*?rNoawhzu#w!Mt*v#+juHCM$yK6^qpPuVoTvh_QB%<R2)%I*J(OX;=f0-qT
z!T-Z$xzm3werh^1_J8QcNKOU3XFCu<><fO+v9`*>)HULASg+-wDaS~EsFZ;~$|Bxc
z${gn=_o|i*U26*lIYIRdy;Ic#Mv|!*zYCj0%1d&)chv(@V_CYhwPUz#x_)zS&?_s6
zs$+DwH|P|IsyYfR-!=xZ>Y-(;e?NFxof%PpN&zQeNH(cXH$weqx}V*u2fCjf1=;mU
zDe?8;=h8#eT{yy13{^s|b?QdtRaYZlTF4ndh;q6>J<3tNs(Divz_uYB6O$Pht6B31
zs)^?>w~H*>8kZ$wEL8iZPlh0_&0r2^KOKf<M|hfBWbK=??=lj!JWwfGf2(U)5%ewH
zjZ8oECiLhOW$x<-`u*0H)C)0BcADGFHCH}BI;-RwR>0Qayt=m(bbj3XIyzqYaEOxY
zht*o!KWjFv-E_BWmE2NXTV~aSU)#DhTlKd72CQnibGWfF^JtJ3t-GO_Yv<$AsykMY
zZ+Dx;xt4Rw27Q>Dt4%YPG)8k+e=E+a4#Vr{BqO%uURSrEQ}7*NTibT5e~TU4Kiv#I
zH~*u+!6sT)KuCe|AH%Utzn9VHEEEASm(jK~CJQz%Ol59obZ8(lHkYBQ1SfyZT3v5j
z#}R$^uUMcj>4Wy}?9S}$7J&gLsnQe*U|25D)(`?pq7o&ts8Uqm_Sg5!QcFsuTvH-#
zF>vhRy?gg-X3jaY_vToN3oge}U5ph*4KC$q20N}|hB%4ENhTO-jcaz9S{rp)*o@+`
zU5&LmP2~j)a$3+h5xE+Y8SH;31dNe_Sy{{yN3q)7kR5T%W~n2Q)~-0x(ymoUmFyZN
z7lWi1umUU$lT##29^J#rE~eOqRao4fC@zYsSy*HVHf~HVg=FJKhLvrMC=PuVv#<xm
z6=R@Dii)MClwC^N?y5^1TQ9{Wk#?<O*V<eXo-NUOPUKP*YnQ}vDAs>2MaR@uiK$_^
z5n~9hrQGh~8jJP9!ErA3ER*9xc9)|$q}@$UNNsJ*PLjDIF`UC}d%zjKYIA0FQbM*d
z$VpA@33#%$yBU)V=HWJY4AygVrrI@|xCCn_m@b|(3!A8IuWZq^wrg-WMR+{r;526I
z1%D~GyKu5<+bcY#iXnd`7q!S}cg?biXM~@YVm?CbX|+g9h@KMKjw1xH1#DwfSEAYs
z5Qs&Zx0GDTY~x~CMceE(xl-qM8R5sa4&11*#Q{2~Oco{hSu?jMimOG<=OrRKiLAqt
z%xt?_EM}<%Bm6=kA`HZK7-qKV!XFe;dkmRCfo;n;SP#NBR<VBuVp)=qWQepBFdBk}
z0_I>>F|gTY;xS6?G3+CeXlsCcpwgCwJc9{WV;;@4@Zx}esUienl9}dch8GF$<riOn
zJ-q$*)p&Syc{!Q6;hV|ybUb|*007}=_{;Fa@GXAexufCD_+<8uV)TNYgeN>aLK1zo
zkT2P5rK1<a;mv>X%=$Yxzx-`@wdnn1c0Rcr4u`+J`N97G>X+H<>iSQ^@bmfXmz!gM
zGI>Azdh`Bj?{xh8@YkCkU;AG#hU1IL@o;ulay>Z%CLI%<Q99*x#%Zi$JwDY_j%VSF
z>sh7YdV12I{KJ9dx_2O2{`m6c3;ema$a_<Kc*QjT(N=#PwRqdl7?Wp21fs_mz+Ozr
zhg^?_R}S4D4&P7SPKNK?R{%+*FOndHpcMcq0khb25*W<j29ES%h^5{zCx+yy(8?SL
ziN{I+BZLET2wSL&NqqMXNF34mo8j;O_$M+IDN640=HlY$!3Gr67enSEz9I<MXsP5i
z#x+_>Dn5VfI(<rHZ&CZ{mVgk?yjDxW_UYZ0ZuoX`F_|7-jZVfc*#mBPJ(|tN(@PiU
zKmYt;_Wj{(G#fjbU)Y~q&M?M)GBf}r*eLs!%?3s7M?wBXZr#0nIlP`uP7cSjcf;%N
z_J_CQ53{2?!SQ<Z^B8luoQ*GM*G5nCS-%-yPi}vvC*yW0v`asZPtQl+Og^-rG3o_G
z9^v^Zo-}rxw`#$5J!jfB{EK&uYi;I@YHy9)4edthMm6K;sX!qJQNdn!q!0>7oBZ5J
zaedlw0;#%pz?!y#6VZdffx;eij@SZBfF(FoA7~XOfWF`>AyXh3Yyqy=B2Iu5J$Sq0
z1Ui5Blpcu_Pw{r-?BxzGL}XY<yW;GUKnh_y&SqbY^ErCsXb29h8_OnKPOJM2xmH6Y
z=wZ_b0%0m!Bn86eho!g884z)gW(5-}ONh*X$oxOlur+wRgNXOA^0Dw3SK%=}2_9o!
z!XtGIq>h2KAqJ}Y3hBSVzydByyOI`}krsbB%R_1L2tmlzXH&La5lZ%uw$HxC@$7n@
z{ES70&DKbfyj4PdjTF_RfZ3TfYS=p{dAUZ*=}7-y>D_J6brJH^3+Ut1Eh~E~k9Sx=
zC0;>@6rX|+=?+4~ju5dUM69<)3f-UwFyoO3PXG=S5GaPLS38b@s8H0R0zT&$R}g<*
zQ$T^JXkJjfW(}m<pqYr;P8{1LP?W|BuA}4_lSoxpCy}!W$`Ap03~eQvH=hJ*SEQa}
zCKO+6M<TvJt1<{z!m6-ID3L2`gq|%7_yxXiK8KVM?I3-Q7#wiGNE0h_AR-}gzswvc
zW5H^&5*%*=3p{P-5=_`1dE7==PN#qL+>}4;iydLi6X>zoEny|QxjZGiB`&i&^<3tj
z%S_#f-ALW&wG{VSihC`^x|wRI*+)o_&Nj#>uc#!RnQh*WPNtJN-%wTEJ3t|`4M`28
z_9fZbrT9u2Sw-H?_5`Ue{Dse11i8@C3jl1XE+jtZr{c^bR8&pAYF_x9L-2pBx+6+4
zg}8SBRlE&vbMO`31pok}KwQ81Y<LE_$Ye4S4hUssqjNPjp$fof{HMCj0t&3@?VUpR
zcA^A&{p`*r5UCnQb~#y#k7-wqMv#wiwuCLFHX)@^_bgHy4sB;aU$|U`lmMtKybb#F
zq~+1n$@SyPdglt|yefn9lVor%OQMv2&VACkPdfKWSIE*8vUFC>omDe!#*aC}pUa|J
zBZ7=}C3|v4_LR7l>?z;`S#VO5oAgmi`mDUj3c1OP)Zrp1fcf*&B9yniEL|KfXz)_%
zuAD0d)ZI+71z%GZPgY~0u7M^-t!1a>_%{a+%6kAQL5W4g{wg(uk_&I4$mpkkxfXfJ
zh^dY4BBS7W;3yk)W|SC$NXmlhh*H?>6dC_2FI7;C)kl7I6?8JXxHul2{B|pTY43pY
zW(P_t-o9_>>nwzSc5Immv7Pk(nNt2#(YV^Gzi>m9%Ih>t8JM?5!x*YBwXD&SD)2IX
z(X*;yABAK4f@S`L?T0>vz6-5?(6`p7(093Wij_MfYiDHbb*}b0S9_hSo$Ix8y>_nG
z&h=V0BlMy``Q^(aX)n8XD;-4k4)Qk2UE=D=4v#>kO!l!-2If^zs`{8$iw`MTqlOxa
z0Bbow?ree91@r=k_DQ6`CdfZJ36h!_rCZ>PN(}T~(-saGZ7b|s8n)wq^=eyzIIuG<
zlnKP?OKk<s{c$^q6u`ukK01jtx8iZ7f980v^tXYZIsAVgP0vTi7vozth!u2@>NZ}X
zBH%+&NSk@!$6~Q)M=p@{tCEtSbu}<V4uHotyH?puWI)`MvfS^;f+*az{m#Yccf)sc
zrYh}K#p0kr$(Q)-$T~WIJNs$;;cD7e3VpYCfF#^dP_e&rpn`}h{OlFszivjS%XJdN
zqa=J^QI+tCw{&+JEWU`sVx`t~S_(u`TB9L!h9_C8WiEhA-sK~Ysw<kQ$`dq`eXqID
zOr>k4+BH+{nyGfpRJ&%XypbNt#PG)Q{67YtW!J)4s+B?8P(+-6ldo`wfcWeiX@Lx|
zb0qs>O!EV^Z(uH*BvWEm6v5kh!FId(U>Y9@Via^F*}i)RT|98w0uciJOqi&^InW}T
zsQ_kk_EH3}z36>*pHHr4=kL$|GoJo*j`W(IA;sq>9||zzQa1c_grM{9oa7}0dVmDj
zC3;F3<$l`qBbkUkvi6?jNZ{qTTNxcKqvPF-HVUYDos$1vtJ`mZ1Njuv8ZBG-m*N^N
zIhj;kt0k!yUT1Xsyt^*HME)<9QeT&GBPA09HZ(Dp0jn+(m)O}lD1Xgb$&MR05WUY=
z^beR}v5FfY2$+#)0^}BBZb42#?2`}i{~uCZn`AFiTe9JU)Do9kUTs`jPG2TD{r>dw
z{`A`$(qx6xVAb^gc>?Z`{0)*Djh-xkFbLE8*Xgg5l*;hug+C-EB7ZFZcKp=m4+xG&
zfaVYP0-Qfsf-r~2pnpnp(WE)Zqtu0;z|ZH<&hx!AGEiW8@DeDU|9=1DNoHZC@Fx5W
zkMZB@`_q3<Rdj_y8$TK8gmN_f^Ytlo5NKhnXN`nqg}(heeSZ3Lq}l@GfPs<5Xx#y$
zjL_Om#(06DD*;2Sm}Vt4E2P73R#LN)#8A*Bw{siu7~*z)Nq@gXdl<t_l2~#1at@e<
zzkwXt)J$)T$H(*u*hdm0oBSUab0bOl7U3*<C^j&9H5u*rX^I}BJcTK?%(6c8yiJv?
zAhgwOd_&15?KLJ2`jy+WdgwCcOM~WXQ8^*Ap|#@a+9mCyXPVR($Gl~w^ip7}1sN6?
zjUela!0d|&yMMw>rP-~p*X%G$6b0WCikmhCQ(9>7RWO-@=@LS0DMTlLPXa_lOMio`
zNrJ60FJ=VdP^!GmjaJ;1F<Ei+S(C}wEYGE3wdE8w<klUdiMo((qB(t3^@EYoB9I9L
zTwZ#=W^7{eSxpa{Ni%0223=@UEpCyvOwZZ!=MRj_seiaw_y?y(9#{M23{U7u;a$d4
z!8hoV9XDu6j!uyZ*g%6bi_4iR(!ywsMuNzbN^7KiSsTN0t0-sS_Z*NfBjRkR|6Cjk
z=9K2BGN<Yt(eMiyT;+dLR&A*zSs-V!WZgl!VAq`87ZxOIh3Brpg39Mv>vkungq3cb
z1#*@W<bOV2&ryzQ={dW6wM1drGly9nYU_x#!Ct-AIHBEK37(5JX6vzK_O#&IA4}{&
z=?Z$l6^O<;#g^zGds6bnPA2YEgWsg_eK9Dl#rM*zIZM_4E~i!MD*5O<rsqXhvTTS%
z^sf3}jNCmG(Vzh5Mntao0w-kQVIRh&@I44>6@LWiKv)M{H~BlP{J3zOQjw^M)zsM=
z$t=T}#;!W3Htkxpq@h};RbDq0`p~R_n7*>UNjbb~J!MV9JXc!iwMiY%bzGzI+|uHt
zI-Is<?K?{yH9Jb7RPpUvE;FTu2ivl`v#H}7v|>+A&aN!_I+T{$$<w@aNs>lty62o}
z`F}b2+f$o9Iv&rncf^udQH&_1U|sbCB2F`NqH4V{@!Rq+aI59A^vEm?WITc=lcFN)
zrKhM}O{H6*tjGyVZj|^+LR4A{Z(*EuWBO1Qg~A2TE9pap!e`7lE3Cs}0!z4dNV~ka
zh~@pIUC-6=yFS-a?J`d4FBvEOTXMc}X@8J-RL``n%(d6d=(fv-%t^PtwdUhy;p5SF
zRu+)gcdbYu2<*6P1+N5$&HlDk3dwWQmzLAFFVoQSDQ|7P+}qk}7KQz#`du!jF-9*5
zDFxl$d~>tR85_}$6qmc(j1>hr%U2Y?Yhm##We4$mVxX*KCs!Tqr9eyLXk6K4hkqOp
zcOS1=rt6Si#Z!U+a0l=V%P1xfI6+_(xppm|Zos_fsxW6enm6$!6Xxu@jvVI$wa_7;
zw01@@Z?1I2F=ME4n!lnpT}x}%j2VZQfZ1VQ26*l_3h_coHEZUdW0ZkwFv^XIY|S`|
z*Z+Waq<Y;rEzi`d8k7%&p)yJszkkq+M&yoAG{Un_3=|iKfd?q&lavvAxkwj>s5+*m
z%tB6Dw-l#uS~%T3@u96y4wqZLVq*9>n^j6n<sYk@e+{BOB&bjeH2Q&+W2E?yW7c#Y
zx*a9SC>B#}$FB%S*Rtlby|96N+)k88)9b!bJZYGHc~}7=KtW_5X_kTFuYYZpUf5_B
zbGuoF*>}XO3`xNKVlj*1@aU1gF%bNvU89%vpl1w|Yjut1-9CN7<ybIgJZ;r&wQY`G
zj#gbHE3C7F0067fqE><t9R^3aDBC*UMqs(<C`(`7j$~I-I<I%ymL2CpXN0t)gyBLW
zFp37Gd|DK)hU`ra7WP7J;eWB|y5M2C615XX<0Io?p!jPCL@(^oh!|$iS1qlp*&h`a
z#nXE!Hsitw<rgyl=)ed&QC!G)Z)7O8CoHITp>e!kg*0P4J+xk}A6qck=(w3};!H$S
zsx0GL&%Dvvyn;in+Q!z!ls3(oovtUQJ+{$en{Pq{9|!kcD^Z<8Ie#RR`Br)VDy48W
zo>P6j{vB_*0q&I~-JRO~!$BZ6fE$E)elz7zO!>3M)BZik_a2BTBq5pUa3H3X@IY72
z!-nS#uXS+k=0&)9oR#yba>lvj@%%bLx|2S>>&Bq`LfCX+zMi1NsiLyu40WS=-TnFH
zj+=2MxhBqwY1C(;n|}_7`@>&7T<_?7h!tyPy~RI%R2)a!IRz8dV_)CgAKrZ=?Z)dX
zKi;#p7dz$h!k)^CVE-;zg76+jtOtMFyvqh7Lent2M(^WOHE?3}-ajx^gJH_je^(GU
zo=$3c@e@~Nmvec+pgQ2oP1zhsbDhSAW%Grw=lQLTR^AO^w|}mm6nnrZ>`%(}AiW@<
zpOoz(kpBUGlFYCQ{no1eco7J%O^_aR*sB3~iQi8+{x)HEki70%-?M7JSr0UvYcvcK
z?%6k36v*B+NU*VOu$CQlbe&rZsao@gdkU%pjt@9iPl>L)#pB~zp(lNOsj#19SbpRr
z(=*AJzPs1kh(v;LRD2_zwo{xJnrHV!YW%Ofwh&duW(8-*m4=<)<J*0^M>(OD8d)Dp
z(IIAl^H)#g2i9oA8vsQIO7!^TWdA?Z2x6B3!vzxpHZ+$}ataiefHEz{0nWEaGc9Kb
z1WLZlzn9VHEE54Vmyr(v6qnjNEx~_4dw@Iq^XuPtGBYbhu>2=K#<$<T?*80W)|D2<
zhKZOEl0^9SYsX(87-3zQlnWx5>92oJ-*=x!s&imG5HQkOV>)1z6~_3<+8{K4m4LyI
zQW|0$-{?bvAtvSp(z5#NuOHJe^5s+G8wqkf5?0WmP$wHJv<~BJf>j#A8VP@6jO^Ub
zhKH#6kq^1fN03509B1nVA2e`Gx@MS+$=@4>>t-Gmf<WcQU11$7g`|1X=Ex)MZ-`Oj
zlFAsn`23i7W*#NI!*7x5!T?BJsNFjlKu`FR0razt&CIh<D!-pHTRaZwJ648SQ_^W&
zSY?EP@D9TEga=_Ic`U62J%N9anP4~aWd)FdB5OD>i%Fx2-E4UglnuK^VbwzLP1xQJ
zFIIB2GsoMlSvvu41&f8(-R2_{=d>NEf?oF#g8~-dZCL{#Mc8U`$JI$DC}B}Q-(&{D
zPcWI0BAE=@q~z!?Hu+36AIw1_`q4U+QqyK`q<YzOcu!89X++&=OqPFk4NI9GG$v;R
zS~XGyvyK3zSO-f4)~(4<!m%E!p+ppJHx5m~UBO5VKCZiWHl=>@9dK6@>L_^!cS8*p
zg-5p#WF6`Xq9MqweFtyr2|vTz28vJeww~&h-e#WVZT;@>_8e~;sN#8V8xYv=HuS<v
z*Lj<{-`jd>7rd>X{D^<Ib(Fl5k9*r72%h&gdV2p3<<LO*LhU!qHWvz7i8t4MMxqD~
z-a*iw@FhV9HcGgVS>RIe`c(AJ!wH5NyE(KZAP%s<2@CXENQ2MEVj#Q-v65FB{#8P3
zH$ntIyal=1tY-X1)|uVV?XubKQqKfZEhxP9ZAPrF-7IjQQiXp#*UY-<MJ==#%#C9_
zEV!d7bRd*m=xM7d%rSNZ|AOi0{Hy}|l(ASiZWiV3d~2fwfw<sAZZ_STgO)VoiU+XM
z&QV2wEq0oZ2$WvKj`xp;I)t5iONTrvANsJN@OYf#^E}u4n$y&x+X)?E5=>{Xr`*k|
zi!GSRqoFnOrT2eD+rmVQkE}<1hpnZYJMg)oFXg>ZYAKbS#q86Sp=;qp(Sg>&R)Nw2
z6-cdH)6GNYhXGeXTny_85y?=E=vo<4)Z?y;N+l#y8KS^f__M;JQf-Su^#J6QopvL@
zD^ioG=`^~!&lGYBO=_ry#<WyVOH1K_oEe<ZkG3?Zg>iqTr?pjLaS)+yp__Y8g=I;N
zL3eZ>>t=HP=w#UFWeRqZYy{!a61AY$Q@ey;Kl>2`J4%n$s2RgS5<GW4UPsA<wWnc<
z1>SOd7Kl{bZ}pom)$5jb0{LK_V)@*6O$k{g)Y`?VT8;*rw+tFls9E>A78sfOu~ieL
zRg6Ae0%m`SS6GsC<YH5vvb!?ExAWFq*TrnAIR`$<P>RqaPwV$CiZEs$=y-a4<LeP{
zF~l8=-e<6(<%RjKC2QY$VZv$3!s5O}i2!*1zGQq}aetx$)iB<@gUTVd|NU@RAV4Pg
zx|QqhC<szHIfw-xHzSUs;)#2jVTH<P(~Dz}Qh<M+FMS@DL{qeplm)Msq!f?OkmR&x
zx;2Vy+*KC)HC{NSYW_9{#C_{iQhYkZTS^0m7t}KPCI6)sCdi|Zn`-Kn${X`~jPWgx
zZ~<obmkhP;c71eGT)#;>O0`O1BbEg6U<o6IOA>9^&=#MXDmiPHyKGTe=ap^9QZGDG
zNfm$AS@hji=))4bD>SUSXmKY()jrV*XItL9E(&d0y|n^Bk5qH;U+hh_t=Fwp!;(;9
z6XcuRiGy;RA8|ak{2r_5`faR?eSO2sd*Q#$3z>qM&y9LIm61N}7VfU~t}IK5G^+1$
z@U`wcSU}!|xPW~}sae5U+Ydn)7{P6M>#TnQ5lDd)F1yvBkE(@3W;ZL8GWMubKrg6n
z<%Ip`L{ak76hkFNPaV9O{x|V3EutdF#zT239wBh_-+lPw$M?T|`xgjW@Lyg*`JQ?o
z2!8sNUM+E~kaG9W^d4(IMVpF1oe>E{^|IsKUe?utZ_|$U5#&5-Hti>mYT-O-dA5I;
ziwuuSSNd(Fau5lY(X!po`CwtH?GPoZT<{xH$!qzQ)|XZM%AfgrnM$kvGwr@lyU!KP
zpy?;xb5L|%0h2bR-W%qC)*-%<MVGmcHE-A$oQs*q6iLh1X$Vx31?N~9nQoP|jLf7h
zLqw`jC`08rj&Ddm2%7N7?GyDp6UTpUi2rZ{8PWu6`}^CkyzB=p`ZRkcGB13u07U6U
zQLp34Xo*%VCfR6KVql(LW|z$WdeCOy?)+m6Xt5y0$Fy^)E#~eu5G58qPcs&ys4q>{
z&)333N-_qofM+bQ8Z5wfJY;qSmW|YqrvsIk7TQf)Nv3FW`iX(km63V~^O}Ey`D;#t
zX!ixcy4s~Wor|S3$K=DMIB#C+ohk)W-psz;P{cLmzf)xYUn=te-uA^~IjSm@w~|=i
zX_kK0xbG;bdNjTj^idn3sZ<%zy`eBe<0wq|kwTRX=UOo(%~g~!Bhw_D>9RkJ%68x&
zmS}TMeabYY2$5V4$*_`$sk48%y3k5UDu-uEheWSW)d!V}xw3XeKY87N(J<<hg=L->
za@KE>EuT&8q#I4ydBAS%SQOy;oKJ0i;K18c<pz%W!F}_bP9n1^J=D>PU-H?CHkP$G
z)9Y|wwXl^OVde#Rr-{X_z)I+xg{hU$+2UXdoxjK5_ncHU%DE(ZnbkQ&?GR$KBIgrF
z)&AY-2`5<~k-D~U?pWj2?jck5WI3R3J*B#}sqjFxhVM7K^Z02v6CyMDKZbAeml0PN
z69P6gmtn6b6qoHpEx~^u{`>vkpZ|E{iCNB^_Lu&v|MmUz?SF5Db%9K2chSl);BfiR
zA8-DK1jV#*7a^DfdHM6dm*3w0*;CDPb>Bfz7D!QCYo!`R3C)zU7p;+@F<DWP1XU!o
zA|)vjRQhWk$0R90l92L210S>TC^(ZeMfvmR@0V`m=Z{5ZCz*d10~heke;PIBHv~rp
zX!<w2F+2|n(l;?D?@=C{@DJbLzi95wEM0n%MwO}F7L_)Ls?za(xF_Q~Vcd`D(~=`P
zCdjP&l5<8*!5Q~7go;#~5u*C_;rPtTC=D}q&=zJ*-LrhnjB7rF+A*#Mp4Iml5P_1{
z;K!mJNp|7cFP?vwL;7&$a<DqtPGwn=eNM_GC50CbGHHI=qEb3$9S%J5&YD47#iQVX
zVlz@xRFZ)lqC>6fdKxj(B6DJpkl&MpvqZ9*UUQBTh6%vpPwHLDYl<C{Y@*&nVkJ9#
z5gPqn>96FFW%6i<QRjcDYbBj6_viU0S$5!6ASsnfK*xXUDwVm!LTM)CK*eW(pmS4X
zl1ZOJSu4DE(8r>}Xyak@=Z|&Hjbo06x;Zy<oJ!KT>RuHY{@Ulkz&Cc0F_#<-P$?bt
zm+j&yncAXJCuKx6wzDyV1|G{>!B<3Mf0bq^8Mc15^9Psa%zBV%)<^wRWhpQ9;K@{8
z$u>j%+suFXL>qkVARiZS?KQuQE!r<{WFL>eYMDS>E8E`6$mp->PT`dzO($+do>`~p
z5?T5tPgL^<{YQGTT;0;sSti(wG}M+EjlC{UMjAT7(Ak~}zwNvw3CEaWZoOwCaUCPX
zgFm!BYl+uIGr1%~J7`BFuTo|15q)`-6%j#%t|5Oqdan$>-@S-4pp(=5(Zo*h3VD+(
zb5i$?8mW8CY_3%r&>2Msh8~GYQATreimsGL@5x0lhio+;EMUw_LlB~V%QU~r>_L@@
zYKC+%L6j$=!*4n>0b`JZCXNB)ZC5plAiNnUtCSuWy_!VVHqw={Uq<RwGby`~TAgIZ
z_qKmSqR=R%BLvATkt(YQUKH=X$2^CwtL0m&2ai^z=}5g0psyBrB+F2d?<S6}6HRhU
z>TMh}%U9em{u;7pbYNP^U8{`zh}Z|6?u`^iPKcK0b)_C;s<#Cp27`VK7Ac(TOko{Q
z=XY1!&SWRmn^q=bj8;FLva@)~R&KJLtCW91MmjANZdKGjN-U|`^Xn)M&!+t16TcK5
zE?}>i_}-pkCfR6GL`#+&_}PIQDGE=(jU;`^&Zcr`z3R{<y1zcCQf=v8DJH1|<^@2#
zBspk3G)z&R%@n3E$ITWCf>*kZ&(ygw-VfTe>8%WeR9#gH{rfJ!5mJz}rsvsrh(doi
zNX2ea=oFdlOk)HizwIgYmBgnDftfpi3Z6Kq13URul0+8iXWE?TZO#8e6zWb}k+QAo
zYyp#QO!6;*PPAmT+V<KiU^e55%;|bf-ormi5cq^~rb<7+=XFmgrkq@W=Uj_9F$-HV
zMWyq=sZ4qfPn3m%{fl%<NS~gOJ^z3Boc;vT5@4iu#J`eef?tO`3|R5e<Tp`eQ<b8`
zVWrQ^b;$3w8_Sq772c%Tl);nM>$o!2waU+zMQ*WD$S$Tup6XluUZI?_ahe$y>563c
zO7rsts{#xSs~V*E3c=M%+#}aVBmKTS6A(AIJv0fl>P2cAna-<bRXK8#R}O!!ryEuc
z@z<_yYcAV7+1yDMJ4*E_&P~mE!5op!zMb=v%j#sXItQk$9GWl-La%KSQU1&(3r>N0
zP4w#A!X48}4qWf(p37sohOmbcx~HC7P|;q8k&JJT8nx^cc^zgOH>=Y_s77kVEZBM(
z6f`YdZ?!OzA$#UIXz12f*Ytm+O#pk!Tk>~=j4qKs(J<AVfsnlnXB(zgGn50Bf{n#l
z+>vybwPMmTMr6d3?r%2(x8#<b79tu4xj#1LlvF>pQG=XSW4Pz>All#L)i|cqz-ypf
zTEN?UJIkaNJ%vztmB~K4aNpRisfqJGfK(o{P5UOximR+h>V>Qr*GGS6J(weNUOO!h
zOKyCsZzH5-&W%)ik6i1vjG6gFOR`(N=>>g<6&%zWJ=)=Nk^yDsUY?JnorT{`KcZzk
z#rk0Hw%xw!#~x>omTLX@96u(s6FP#a2%O%87Y1k@VERpXFq~2a?CiqF4ZXapw`}p{
zk)LU2e!kb4cjhGmBiDZ`s`-}E<6}v1;@BQ1C>=U39mQAd-&+XBEjr5Gp69aLPuerG
ze;QZL7w~)7r;j}Ct8`bN1WFK3sFRb)i8|l~%2sVN^7DYon@I&8DpaH1;wg!i*!&1a
zL;~?>gZikpc`4YIKS7peq&8|&7*eG`16@5l<K=rML8BQ3?B9QElWtaTloOSC?LbOL
zb>BLDu8Vg>rRfua=<}7}gGOoc@c_aJs{ZJJfu1U;xhAY-f^d1SiC^$Q!YH!qCM+l?
z=@GtFdP`R%bGqtyeTmmfsu9(Zsb5db9mRdT(nR@zt^Z7=-u^%12b!?18~9vF^TiiC
zJ^j+rvX_?w6tsVORPL2lY$>kbsb|V-M|r2b_I#Z74=Tyu6aW1A*agnaUsyYQwlT{N
z#0%`#Q{2a*&m48^W?74&R8_jCpS9&6%ky>>hQPPLm*(U-iY9ZSrD0T>)*1e=AQy?p
z*`1AkR6+H!jf|X;vL^Ka!IBrQo--D-FfbKJukuytA@6@|dM8TY49T<jMe=74juBID
zbBp>)Uy3p}>W=vBQ5KcLy(_$KZEillh5}6i?Zn@7kQp>Zv3A>*c-_HeiUJML&$@AN
z8LVa?`WMcgTkhavPDx^D|C|GlDzt=-3zLCiFU`R-3-xKmUpj{h$O&r5dP8;+AY-FD
zw^R;qzuSMV0i)94lU!<=h9%gsJ+#@jG87UhdTEZH0GU{S^$3zrfZPl-xo(Ss&)Zq%
zd`nJp$qa{n#*-~LDQGE*1K%r0kpKh}X6<mo7nE*1$GVxKHPLn0(|sl#JFg8fqtqPu
zlKr6uikK=(Hx0z>XSzNfHAra-4Oli1meHvPNIQS0u#%Th*=u$5FVdCix$gXFzFW&$
znxygG(b$YUrS0<ctL8b<*NxOuui*qJKsSr9T2ImYheJlSjv5W#n#S#N_{WYXcJmgv
z$c=b``M)&uYd7YkU|{7SYw@d1pE|TT=YTl{!wjXtZ_fJhR$INXPdR|<l6D6w(}*3B
z&fXQ4p(RYU>^tu481_(xpwj;Vb_E$Gm(k}e6O$pa6qj;mEGd_OPc0Jz0q@3_mQO7x
z0eY9ePc0+?;+N!4EhK-P`3!u14&(g1mqrE(Ob>Vs@K1le{r(^`vr>2yF2ia3_wx4e
z??Yu>X+g$MXh=A4)89WH!VLrx#`;MC3h;LN`Ooz2;g6AQT}<*b>4)U>^ZPW6e9hYP
zlC>v|A)0AV8A0SGW4zGeR?;4Rk&;+CeC30sN7j>qnUtSj9jt#+BqOx;BSluDB-Smm
zPBBtW)C2}T%THu~Hp1*z$_}EX2g*v$_^PwL6x!lATQB(F0k)xJDd8pabX}!a^@&!X
zwmbAG+2%|h^aNp{xaw@=>kT@euc}yB@@X}qwlY%Y#BFE}g;5I`f|h7ad#_KQj12KZ
zzd(<Q1cCo(J<fku4SGbABqz=T6s~R7DJD+hxLwt!lA0H{Bh~Br3~S2$aiY)*ug$m~
z&^mV%vdlP{PWzWNN*KSO1h0z_-9^=sK%*8^aa|beg@vW6&siGMQ~QbQ&PMuVQyh?!
zFq!IqCfG&=2ENoQDp#zubmgq9cX1H?{V_qF6^;79Ts?nM?7T6YDgVFGW6y_0KQAn-
zI<hJa6a%#wxs;`O&e3d|)ZyL*$#;i|R>H$VHuk4ght^hr)eB7=tY>t>NqeN8=M+h|
z3Y0SC1)bg>ti;-K=xW~fnJOlWMp;T7%(BHiejZ}<j+OcBTunFuJw^>u(q`l$`c$oz
zVO@26ZjgT{yoDnzIj6X0G5vBkmK&&VL;*M>KrMs|CQX$EIz^?TKa8xH3>oG~rdpiE
zC9oe4mCh;2$JXE_CxdgE8JsyDDxF(7z&f*)ioS3}_gBRg;uAiP<Ya^cjJQWOr-dY(
zxueZs#UvhQ0LCPqNA=&cvQ<L*2De{AgVhd{ms)?N7JyD_$HSV8GXcPPsd6B$-D!|+
z*&dc2OB;c<rUo*d&Z%hNoo!s`L>YGtZ5)`0#x8bou|wO4!=l*G4K{b(SEKjGr?tWW
zjH?D$z8s^&1EFQc;@7TgL(E1iHiHrcaHHx<bchzSzwBmYcwM02Q^m)*;-&_oAGwf*
zm2-bWDKoC3eP04f*0?DRYtX{dLa($x<!W#&!h*Me&o@k}gX;k;Y<g2Glrm_sSimc_
zu&+hH=JrkwE^VDaUxBz<b^oVN3QR7pc<M*+MxNQT_z8M!prK?6G_8b|VB<_(R(x)7
z0#om_m64R?+zbneZt(Xl9cfSuOVom*k6wSKV$}&0Wsz;EpN*8UC4*)el+xphn^ZRV
z_6(Y6p^2~|r=2f#OC>9lZy6+6sqh^wa;}y>XFhc29kOL1YBn=@mYZ9*Lc=fB$`NCX
z9%(o_uot~L;;E@2u?l^er!b<na(H$|d@i%1FnLy;QH86tb~$))3H0^f%q3J-PgH-c
z+*&7Ivx7zHW<gjH?zyQrq;{osHWB245YFvR45L#q)<qXY<%E$qn4vUpOE%|aa>3lG
znWEi$Dw4(97XviyeB~6k<?zHs$nnseXMlD8B4o7Fkd*CIlh8`u$nIl6gMG|x$LxM~
zylkpE&w)N+=lxxb<Y&vO7!6WJd!v8ruF0x-F>A7MWV+aToxAz4Qsdk7^P0n_yEksg
zmDwBhqSN3?@NQX#v80o#N7s>d(`%Nr)Q(bL89g|sLS^Q|>C#*-Mla!6!?~V&pn-?m
zHZsOGcQ?vjXaan!Wxo(rtGJK<t5CJ#5@bk!FsdF_y^gAf2kuAJmtO~riEV#B%4R+-
z=f2wo)RKfky-UI?GW59fU6@lD?sxT7%(<4o>{VUY%yc8p%&Y9cnTFW!X_Vp)57@+S
z3WehQFs!hKJF`ZXIesVZbXIU;8b#U%I}!M4SSV&A*|T$Ug07CK>-J3&10{}S(q0q+
z3<~6AG|=5BP?zKYPs00EN>P6sE}^~5VYlssl4d2N74wa<F-qxrsy#+bl#2sE=KyOI
z=yCgf0;&qpRi%r0-hh)6ZNy5mB#q}fawQZqcNr1qXO_Iq5%H?GXb{S`ZKgp#ISih=
zJ?Pglfy^}b+(U{6z_MLZ8`FmxJ$aQ|C{zJa>joE(f*=Z&0EY`%_d|b>EzedSgeO?6
zkoL@r7I*mXX8Kv!r=(=|gSM40qq5nxWiSe*Z((?Gski{20sHM|IyG*(4sWG^U58Da
zWjH-AP?NR-*>oTANeQqKp}mYb;8nyx@xV_?2Fh1HDH&#e&?hBeyiodszl4}C`pyqZ
z=EQFOp5*Sgfj~|utHxAM8V-Eh2Hb?{Zu8*)7au6z{U&~<y8|QzMlBu0J3!Jo#sBe5
z{J_bTH}NCi$LiU0cMnJA%sIgidJoT8mB{}Oui-6cxOcvTU#*D4nJeyp(0-SfF+3*|
z1U57{F_&?42osn4UO6Y1Ay+Mze@5mougAaMe?RD~tX6^JpYhf{y}cj)Jyg{JmF3W}
z2?_|izkeL!3o=%*lV+`)*!uII?%UyyR%)F~%aHVAb@%go*R*`i&I?+cr<tvtI1gA^
z>pQkyp$N{-qc2)pF^(>BiSZ<E0y8o4`Qpu@M9T^>v=YUO+KOvQTuNCof4(J&ilVON
zr#E5JA|_Pn-t@t4q?<1j-G<(*49rxKh>`9Gn31n^KXubGp1Qb-oFa5J&wR>}{(^KD
zb)H!F8TFRTwdd5++I&H}r=Ad9UVD}n6N7LnPuAp7u8s(~#izIArcQXoHe^N47u#YD
zSeL?2oOF!OQS;j7=k9-Ge`c`Cnowtci*-3%IbJIjR$Ijt=%pDgdzL!~4Jn}(-+ZlU
zJV%vWn$`yyV{7j4$mh0r;A7P<G8OTx`VFgqe7*Yn%3i6%>{!dJ9M!@em0zE2`B|Tq
zN_2a9I;Yr}X406**DG||Kf)}Pq%Mv=CuidVj_6A%Hycu7))6sDf9+r{<+AJlj^)Ff
zO%w%G@Zq+JF~>^IgH1vwEq(Sa!!qiiD73-DkTuU`nLy+&R7SK)mcmA$-44J4?F!t?
z{L5;DQQ*>fw<W|kf{3F4N_VZ#*@~qKS!E5S_)xIH)%u3T`ZzuT_vsa%PLq}!x=MoO
zo_#MK5Tr+rJfc*ge;{~~`<mXS8GHNIkP5`H0bFazgQc+MFQHJf>a~;f1MLy|1u6>E
zS6sCWFKcFD0A>8#tTidiYDX~jFw;9yKGIZ{I7n`>VMxXlILI^hjy%^;p0ge}3bY;)
z+PbmLHH{)s%(SCTLgN)fH8GpltKcRpy{#>iJUu>Dm!ES@f4!ETy`s5i?SaEcaCmPX
zCwV2Xlnw32r?Dv!l2jN4E50$4H{=bTXjBDb*UjfKRnICZTjARZ%9h4pxxOc7Yc2zw
z9Xy0>s;0Sl+;5w&mLpV=WHJ1jl>qiQ#m7lg^1+zJ(d4~y2xHFH&+Fqp$Gr^J7vw2G
zEo+eX5D{K3e~%Uu^3PhYu_Oj5E7R&w7-_|qf@VBKs#&-m;chceIN30T@R?kV$f%(G
zCvi0*FPQhM$yzh(ZMN17xK7DvJH-Of%7};VR}4S^lxe0}*)y4@_We-7z(b#spLKt8
zuBa4E8555Wkq$B-Z0r{2NT(EIbV~JXSFK9xK*r)yfBpB_ca>p?R~=EKHeDwSV68aC
zwS(q_;p)szNF&-t$VLtj89PRhnh42g`Wbsh45YMo8*nJ^16I$rkR`93y9EqXhzw)~
zS!RT^LM@yYtlCm$rXcivI91CBL$T)Lsan4`|ByC)Wla=X$tM%UsjWh<7cPko;b?KR
z_>^5zf1l*J0BX=T!ahpsSTHT@1E^r=#sOV(zhfDY!k2Yy=V-@eBl$XP2jqCPO57@T
z6_68zHzW4i_I(I_!GmJNN8xww^3r#q#LAodK)B2q0~qR9s_i#Cv_cSpNx6IJ;eOQ6
zO1%c0&JGLd3Ja+^#JYjW6>N}fvyz{ax%6c>e_J3PPRdd+kzY3_{1EWov?;J6vY%!X
zY9kx^#%DX*AS%clpJ;>iilDK<@~DpYj|)Lih)rJS%6q;DdS62x5<0_^PY1SSQG1@i
zkF8Wjdj-N)+6_uvT7ZAaw<G`(?1vq{B{dTF*!$$~NR4FB%ln=X86>s67dJkqxH?Hj
zf0U!u+b+H7PYnu?HuzJhsuS@B(}$OI^gM3)dWX-avjW~c_*nKFDTbL1%RAV~N&aa&
zOF<_j8E2ZCSzI?s6}}nw_Kno^;aDzQ=07YPBh0+Gb7mMXd1?0iYTHVDc#X?3K8?7V
zVEOQqHdy19g^z-U<HxrF*)wahe&HZzKL>Bd54*IdGp7Rg*|eFL1=sFie55$ACup}4
z|L0UPB<mxN&mqRd_xicuCw^ac@m~`*9a%PaVFCXE!u1v4m(k}e69P0jmr)4<69F-o
zfif*8m%CjpEPo)&rpRLLuUIsneoTCNK3!f<_g9INVO9wP)9d$1&<9RG1*bu+CL;vX
zQcSNur$5e|gQn*~4{|OgJ%)a_^i~gdQdpWH6g~7+h=)6a!mpt@65?8^;~Gh`V5YaA
z_oE-D^^R-qDG)tz@z2AbufLsmV)p*(_$^MgK}NJQr+-MhAW_0=e5?F7aiaX+GCtv(
zisMLcmA?y5BXDqkwc_T^N->R9N8cg^A|7-k#AmsA@M`QBS~nU9$G9CCdc1n&3gcOG
z4@#aJ0d-<!PFsm!xS)3CwAZ|UJr939y`KI#6)k{dO1nu5%Ool4ub(IX0R{13+yua!
z!0FfD(|`BV?<3V@+VLIyGU%LZlBLr*Xr`2%G$7OTKGfP|uuIgnOFBT>wfy+?V;V+Y
zo=;Wx6ey@i#>6-ObdFqB0B3@v8&>#mz|2x+caFBA5aEc;7)CL&*gmX+*ffvNYj!Nb
zv=R2nn95p)P8Ob4)tf^;{B_AA=DeozJa;9;@_!dyeNZL8402-fPd{E}`<dv<S}fx`
z8cN>1kk@c}7jYxYQ55w6kT&#0RSBj*-n9%xlNF{+V$<eTHidDRTLrgir%Mtv3BSdv
z0;Zw3i_fURG)4JA<ZGe>$q9fg3f^hyu~NzC%!|>N=k08#D7lbJjQg)d7pz29b<H8G
zZ-1mR3Jyi#H0;K&hsgt=n38tfAviKd?R@|%cI9%KA*wHwOLD<bE;M(r9K;0e@}+Vp
zN<pxcLpLetc>5qYCgsA*Q!L(JDNo-@o03wwQ(Yd({=^fH*OC!F1hbMH2nbLIm<_l{
z+PaOijC<e;KUBiAZsa9lZ>Y33u~!M8GJm54Oe%MRU@7B*7?iyM=1P;Y-9RvC3R+Bm
zHy8^mLS@zy2EK~J!f_@I4wL1A3tt_0SS4?GN*wL0b5ECq)J$l2hcY?Yo5%|}3BOU;
zwHtZaSNO0i$UWfPP#lIB8K`+JsWt6gAMP{{<<;f=<&`u{7&-1Gt_!oiiz+(iD1TRu
zR54cVLarBXnsS3%Xz^&z(3Sz>fnhdM{4mlb8(*3tlX7|e5#;Sg+SttUDZ{{)T$#%~
zWf%T-h&H8SQkx^SIZ!<2yeZBr$~{1QL2#{(l)fkUKf4kKI<1rgIJ&+FUbxdbHWE`*
z4iXk9ZM9FTjziji?aQrFIb*dZWq*NgOP3<OiYK0dx4kBZL`q>ocTP7aYPLDgdtKi+
zM+O_M!|YzOL(+ks!)~wR!%W5V_ZnYY2L($9g`TbdsN(^OsthZ~QcNjk$Qh6Bcz`2x
zmtSD|pz&DV^aoBPF^!t2ie=oCM5o&5vQ3in34G2m|ACTCGXdjfEA0~>et$XnflBpH
zJ8~->rix|QMXs>ElLvFVXH?Epq(+L)XQ72UwPQlm@;SFSyltKuyP5Z^&`6)6#vN|7
zpmLo&tO(KUJGS%tdB_?bcn}*Eqzw_Yvj12;LdL43qf}`vqJ&P3H9r!jP{|qEH;=hT
zl|!Z!qrqhF7u2&%8K>XS*MFl&`l~Ta3o&d`S}9l$M|-fH&C(Il{1Yug0Kr^6>q_;2
zc&@dNN<FmL^Qyh6whtEOKIV}gOUk?@Oyw^&!Fy^B;^T3>Aw}kKW|VO+R`+^8mH@(g
zdrviJ3rg!Ww>}~E9>uw9tah!|+xiIwS=VCC1xQHUuIp9KH+zo2k$-%uqCTdMCYH`<
zD)z$fxg%-Hj4UC_UWUVZlaxn1HV%BsjZVUD?-ALO*j9g9HQ<|@#Qvx&v9^1&=)rNa
z17$N7C5C45sG|IR&CHSJ+!<rleL7?03e?o_L9@ckQN}?;n!nWN=Yk16@Jh9y+B$ud
zrdR;-{ex<xe(c(DqkrQI2bASaM?GB#G{J+7MpJne#!ORorHUzp-rTn8%Q-0nob5u^
z9NNSOsR>Z*Lat9HHo`(f)CQ_=Y?=y#jGhv-<$|QM%pM&1gDD`iC0&Ui>&~^IRT{{m
zoWT6ma~p$cejZ<q+PI>NuyR$4y1DfU#JYfvOL>~Zo4cjCY=7y}XfCW}b=z_g7THwQ
z`Fz?zUh{dGlT$>bJ=bn@+1b?4sq0GKj_aDTBV6cJYND**4IY~!>(UsmB2fqlNF`d2
z<qISGRAQ^!&O=|l3AXW&*WAFBt3#91dTYjZZR8|l?@CqLP+e9d<Mrs=&adaUW^bME
z1=sf5s^O(#mVd%W2*;esHS~q6_1UNxk_P+PT(;%0I1ZN$C20!EVYknw8?KnZ1<~QU
z6_gsTTM2|RLfWBZQSFQv2xUGDeXJGd@_`51Hq>LT=qOg6cnmAixB)q2H2U1TP+al0
zyYX(1i2u$0VJ9<E*KfEp_d5#cxe$Cf+$Oi;REUA>I)7d_c$FYCcUgYZ{Kr7}x+q!;
z|EbOb)!0lzIfSfU;Fmq;mV+FkIT&*~O5LG5q8@um5<5z8D=kD$w$e*Rd5~Ml1HPzO
zMo?5S;b^N57R#sv;jS2|y}MctRve?JdO$v)_^%tj=QN3ByHafx19SuF@notCx)oF~
z?GZa?Wq*BcP|740<NS=?LNAK&RV?Ywae8#4S;!0QVVP;sH_Cu5sx3+up%<q@eXjUh
zlEe*eLx(o6Yl9hXNS4osrXqHDe=C8yW=>00p{SW;2~eA$<2y6fjs$kN;JEXzbv|eY
znibJIonLF}M6=!dO1P~p3wyl@w!6uxRPHL7HzxlB)#^cLQM>dt%Qke5$Cerowv8rN
z4@)ts7TU@0r`OZJ1^J$Vm*K806qj;jI}-yjI5(5wIwqG-W-Xe3m<OB=f4=?pq>Hfi
z+9Wijg14RXRx!;Y;Zf7%sW#h^vcZelSr*H@Y)JBZ9wC;LSTjm@SV9(*uZQQP34Ka4
z`AThe`seyl>#GfH8etm{lux2L+@;E|%{>nD?Te&F6B+Y}?rB<I=AIk1B(WRQ&b*<Q
zGEpMxrMBgN&nvHg*P`93l_@A4-Bv44W-nQJT?3cxi^<;kjvdQcpWkcYx~j=OQe(yD
zFrm}RkPD8lE^6$)vu4Y8zI^6g6JT-Ipi$&v=ZsN4gsMMP2|{oGEPa@Iw_<Z8mVA-_
zUf-VneVVb*Rt3KO*SDv1!vZPiV+T$}<GSyEcVC}=msH1p+deCxog<nm;1p5l9B+H@
zrYd0TW}sre>O43&Y<AwN;HfBya;Xn_UC%srEY`QoxqW4;Pe#5B-i54Vl%m0>^bxjX
z`)H*T1*D$lWF)U()Jq0Hw9IwJC3!HNiVsa!Zq1ZvL0R;Y6211+*g9IX*E^jp5_^BH
zhRxoRcIs(=zm+2L9^7E2`T(n>VwkI}OW58of&q4a!kG-qA%=G;uZ)}VUkt?KAk1@K
z%2)|7Sb{|jblDF}z~k^xV@leOx~jFbLsn6-Y(DbDB8M?5F2}6HGeRD<g&R@`m<-0B
zV94}_oQ+<JG31qxLCWx$E!s@P0eR}=%LYMWIMr5v3e#ZmR6#9MTT=<WW03JPl`e0J
zHJOZ#yqYs=X|tg*!(yyS!;YuCn;Cf656P47_0MeRxr<i+yr>{g=Ws(aqhmq7Ph$^g
z9(V;;%dBzCx!+fbx3-T<f?l&+c1vNs?8>tperR&fEqKj+ong3~FLeo2AI^+~m!bgg
zMD1IDmXq>es+pWFSaBZkVlTp*K*~dyAv{G6w|E)i3)bZ?D$#l}@};B!%P3gW05Xh*
zO~K??Idwc2w`D~U1oT!0`Paa=j*L(O$SGr@WO5p;*EYQ0gt!o#0u=O8wUTJ2DkJ?i
z#WOY-=&E%96<M>xJgbsdQ%osd-zdAG8Dda>AsERC^&&JQSAiztez~@6;42}prHD;*
zZ&D)Okp@o8?bQX|V-%|#21$9XB&sHy&fbCtx}d7!G)F5aUs<-v$XK?bHG?6`NDa-X
zaHGLq835S%DE4EbNZXY}G6n+ssNl*B<~x47E@)6ij%Z7|V8Q!1lY5+NK%6~%UUU+F
z>!WhcJs_DaQAH43^$ji-xm=d)OD@#IK`B{%^6*h2x@UC)Q=f5JT%GIjc(oKN1&rV_
zX(DVag(xU%Nm7zaSAw8*oHXdxqR}bedVN^Xw$^cJ9v>LA1;vHd#(tNB+<X(1SAul&
zIicA**6erG*X#lw{7ki2Le~f(ST7BK+mJt{{5H-Occ`zoAXeLmlp<IC>ia<Bk?}%R
zr%GM20ZeUTi|JSKt{i?$6~?T1NZYeF2!NI=D43YDCi!ka*cO8dD08Y`4OFnC6%jR3
z>8gta^I6m72C2a?9$<@-rQU1aXoV~;b1f2sxu3sgJR&;=3E7k^))9AlDX=tuw{s(#
z`T`&~x&7UenFuOI8smpV_tb<g(!o%42<cXGqvfXK#^#5iv9k{yXJ5N3wuIDE5!SXk
z=(&nZt&>{|`|3`6t0dGMg(J+WB;%Fk<m~^bl8iA5f!?`Hl3ebtlLWzcER+m@!ss7}
zxq{*YF;`H&g1J(&pBi(eYLCHxTtRif+y@k9ywr-Nmps-Q$(BQcF>>~%!tB^5ns?*?
zsf}(-_<>0PhR9ACf3YM0Jr@TEKP(AA$4KGuCrkp6gB@w}?n%J?0UnK26zG8gUQn#>
z*jq5Y_@bw@KaT4`b$=fQEhiQMq+mDnVH(8{6r!6^yg|0eWe+E9OR`;m@2k}JZ}REW
z!|)#gx$b`?{tJqacaO4n?&urUriT!91-sDktT<Eba^q;^EVs9Lj3$NF%uKq!&0`Q1
zI*^nuzG;1r(-+_;adX4zi(;GAQ&2!GO~fr}-byOEDs$yvP{D9wW|}kfFC1^;x>Cw<
z`s&dLt}o|S$6-8>sV{PWVZ`e+oG5DNXLeR>(dHv6qF}?+8opMqiR*@gETB0T1asTU
z_4_PmALTW)eXaYj4;KE0LoA@2byBbk=QC<AUOaJ6XexbpDZ@%=;!--D%UIwq&4o3h
z7c<>;$ik6gB8?p#to=BI`F=Ma`txwm`VY39k*j6aOC6DU__eb^bSsA#9`Rlr(;NKg
zg^J6LHh9^!vlt;|C`J(N`zx*6GC4|7PXmm5F`G{s+9q6`J4<~grxNF94C|Gn#DD0H
zlb12`3KIl2G%z)nAwM}J0x&a|arG`Ue_d}}Hx7NzuMp6e?t^tjN|eY&5g=_g-7Z`p
zNV35027a)eaqF(@F`P+}{`(&4`g)v<jf_1L$7zu<=Z-!|i99@%ZfTrD3N@RA!l4#Q
zfvUJr12=+@YRD<M(1PUVf`-<Lx8Xo3nM)ZG1`w_UN%JE-D}jjnSOTf`t`>-Ue-EyK
za`q`R+Lf%{t_2e6YG|NVvP<Y;eKt;kDfuW&fjKMYU=Ey{YX>`6lK15jxaR7T!iMT2
zvjx`V&oV~Q;?FWhQS-h~gKg9Wqxfu6(Kja&pF<!??`t8GdAou|Ve6wn%G9M6vh<M)
zh1{B3jwQ*O`m+p4e6~^`VH9S3e`L!AbHMy;&F2to1obr#!<O`sOM|4o8W_I#GC{(x
z(#OL`soHx`C`vxBa41%Moh1jM=4&Osf8<*#3dsjBtfXY?&tSBhUrVh)q<UX`NbPoD
zSEW(^6`O2*-iohDeP=2@G85+Li!Hr}f2e;|G_LIJym8Ck*SAdjc<`~7fBLF54vo4U
zEr&K*iLGcD&?SM>HNzZOre=MGS__)SUo|K-?|}i-y%<u4<O`<IN=x7AW<jOzG&WDG
zt1GDl9sOA)s5akot-(^?H*LYFq|q4T=6%(ImF{a8)J9*!B3kR7h+Q>HZku^jr157s
zd7|-mL_^96AtHz?gpD_lf8=rFD~`G0@=1KB+@T?24<Y(sSb84;agwUXqV&LP%^P_1
zByGNNhTuxZeO!+{5_$C9cf-@y2=t10e=+>$zyI@&QL#@5$k;Ia>ARDYy~mHyG7LY=
zPv++@&-M>z!2#k$7@qGh7PIryz~k3{ykGqIa<RXd1sWebo1ZQ)f5$TqXLk#q<(a}N
zE_wWd#UjC?;>+X5!}Ig`;mg_L_3-@Zv*Fe3{bHPYc=hMm43qsbgAS*Q+3Dh~qd=cz
zcrklBe|LU3dpkmjKl*w0=6L`6`TN(&aSU$JjQ7z096jNcDfL!QPEY6KKy{q{(d&+E
zu6akc3!y8wcd172e-?)C=jU%`=Y58>H~e$>Y4`zOVA>uuI$XRa<|x>MO!y0*BH?Hb
zAY?IB+j}&;e0Q+$0e6p2|35sr9R6@|JU<;?4*!1flmGkn*Tv%O?caytm*d5+?+)VO
z{P*FvcfX(QyqWzm{OjG%&*N_=!@<e?U^qI0mmD6k>H*Oaf8{qNAC)|7vv%;Ny(xT@
zj*1-_4R6m6hr{{l+r|0*F_sboe6!0{cXo5KzhN1!tobdCC589E*C%7iN(7ocG-IWu
z9ek9{jIB8`TxYViB4DUJG*ffQ*w@mQrXt`S3#Wj0tOepP<NV_W6CxmNBGUJ#8?s8o
zzcpG?ljuode~p$k#fs>Dz;MKE1&+WH?f^%*Ea8YI6e67jyAU`{0!;!>LY{=O5q5#v
zMFu6BTpVN|w*1Az`Rp9--Mt$Nxr2`lQ40_ch%T3Af|~-O<>8nK+Ps4gz0S-JbiCeX
z8ozX#L5!o5F}4cy0*9;|!Hk{FMwPwcNx<kYhacy!f9Au}@GVGqvjk!(AikzzL>XFa
zs)W`k2oA1LB9tt#fj)tj$0WR>1GrSqq982blLV?(%jPP$-P*iTNYM$CFh!7SE-B`^
zx(*<cwVIkK2qvQnk+!)GuqG;4J_vnmV6doKqBCt~jfGZq0ILUq8#N$SX4!%c@ck-d
zrdFR@e{v*q{#W29ZPg*zeZaD!SQBHAPujvtkYd@rB7`d>_Y02I8E+}k(yB**#iDw^
zf2ey<jHv0-DLsMkRX$t0UDxC|US+)%smm;{)3Q;7TdvVk7%PdAV~v)SV$yY5qBKXy
ztkX~_qTc)rY4b{^{3wUIHxb)ia%f2_IW+4Ue>t=i`7no;E0jJ@ls>o3n0)QbeaYm(
zUQhskdB%74=Sbm8a|$mTQh2uLXAPGW9wAvNKNu_P&hFJj=zgE=6KwWMgUi4Wq#ovV
zS{8~X>x@Z+F>6_;Wx|+N?vu`L6qiz0;!?_8;!<oGmu2F%W#YDF;<hqzTba16Ox#xT
ze>FjT+oabtnxA0-nne-WAVtm3(sFT9bWbb@<HfAaS1tO{A_~%8HzBjZjaIW>21h`;
zgO?RB{6gSG+v7n3(_5m62MIE-NWNblBrx2(fV|*A;Cs+T@0SP_1Ioj>Cn6;NyvSXN
z(E7mZ)Kg?xYyr+Y9eyalYZ=I!+ZJjNe+HtGpPT@<KuEu~PNCSmB;gK@h|JZM2K8@R
zDySSh1EGid<pa7WvJTz*pw-1m1vfZ-S*ImRLTX<RVAs{dh$?ph?0N;T>jcJi0^>Rz
zG}O%w8oUI=)-Ee>cnDDG!5yYmXIR%CH{7=3slZt{R0i$>M*J45V_^0pmRfEX1a~`^
zZhsK~wq&zjrqF|1@<Oz;TbQ;8J6>K_Dwnv!StG#w_PA5)&Ajtr*IBxHb<*rE@Ss=D
zD%(Wb+JqBr!ihHFL@SdZlQ30hZK8v1vkMndY-ef9LrVj*DPR1;<&XXI<Nbq^*#(lA
zxm#}&Ns88($`VPyR%5k$KoaB#Oth>Rv45ptPWfRiI4Cm~T+(L6aW<HE<t`6g5wY~8
zEj>Ls8*Nmir()Z{N)nC9T3|SrI!1a~P}pDpaK?09g;bM~ri3zy%rCgNa7fEupA}ka
z0mkSrQrx5ksuQCOCZ8JTtchd`g@NCDm2#;=7Uao%vknBxh@zmOL_>2`O<PeyV1IeO
zRHEb*ojKp6L)8TECH&vOuu*B3vQdaPFmqdS1qDYjepmtC-1(K#het!2)+r72n%hoI
z5w5gO%PRQ$^_;Pqt>k`dcgX$zj?885*D0Orl+JZZ=Q^G4=ybYc6H%~<DA+_4Y$6Ia
z5e1XY&wC~`lVW~sK7pBtMbT@BAAebEmdR?SWNN~hwg+aD5*Ju@lM>h`<<>;WOE<+|
z-5Qo%s4zj<jqAXvwge!h3Pr4{RofbrUFdLuWH)aGTPcd-q;iu|O4{s@#e<dU>KQ%|
z^noFniSF!<S6Z8wJQgV-S&_x;w|k0e6X@b4kX?7Xmyju}Hf#PaRfgvm-haaF9$Y<T
zFw;b_zfH-0z32Cy6jQDDbG}zlSaZG90aZrKPH1k>;q$Bn`(+C94=ISFgMc4A14qfm
zm%{F8Kl~!ARAi7A{gqF@>m|SH;To?i7(R;<1#p=By|^U)idgW|@apd<`pXyGZsio7
znN2@&33R2wC$4+1YuRyv*K25*fh>NMwMNUVJvhG-;viH$CJz2y_8o}B_)^nI9PENG
z_&cc=!sU}T37Z`V)C{;;7Hu^H6ZMN@{zcDkc1Z#EOnEC+8z`<{;xvvlzRQtw4Wb+q
zUGrj%GrycsDj~F!ily;Z>o2c6ZI`iwLK6ZuHJ4Fx3KW-v<Q*uN9jE~amoNx01b;@6
zC~5-+4Rlve1;{Ofz6Ci2nVja3{Qn`;zF021md<jh&6X(ckEDw)e=L0Y@#*#b>4&q#
z#W1Ucf#v<%BIpLEcfskRR*MmWX(^WX@5?WrIR{OzS9&4mQqs%NcT1n@`I8iuRtQBe
zx+}!<CxgQ8p*0fXUa8|ANvq&QpMQcrU;K62@3`ik0?`w6PsrzA-+y}IiP^##NT#$~
zv_VF+3TJet@>CG$m!k8OOfUbP&PAe&rmQgjj{haoFMRcD;+YJTwXkr8^s-XP<uBr5
zef8^)q(LjD8g!QU(<7h6P~Zqn;^^E-xn&OkWNTumlXxPE7=$9#Dx2<^bbp_&@v|4!
zn<%U<qWg7-=`*UJQ6jpDit_!nD4FJ|Ge*bBGkB*Mi~*77uu8}Mk?O$9ql|-yw0^Bm
z)#&0&<kD-4I6Yz{T29o|iYgW=S(;)2NT+%W6{%m+$vT|6a(oSplcX(m(rTb&a~OuJ
zw$VFJ*GxJmTm2kg>g0Etyd^V*PPlt&tQBad(0P<CKx(FiUCDKyEkHYFxEZNZTJpPp
zI4e7s!Feq#JD)~v+|fmt_kw4UH`o-Gfmj#NaVt-2cugC@G2)RfT}_GzPLkX<IJ7Bj
zs_J|`?I5=~DbzJNMMT=CmneEI9DlBjk>o2O;rV!J_P)%+Q7A@ORou3bq=b*|gRiFS
z5>naBF$DEs#&&PyB&}Sh;Et-T>9p>%t@NjMJ?1movvtz`^gQu4$7rdTHMl4zm?;79
z%GG=|Du&h!tGRs177EsU7%m%1(g@>e_of@Jn7|d$lznsHfl?*=N3{|NWq*XUL&>7r
z6)_OX`c&Qox@zuRKJY->hI-5u9mUEMk6|SmHz0?MMnCl~6nAXZ9zlsEC3I&*{BO<=
zCz+9YkkI9zPQ0W{8A2+#wSb-r!TSKsZe<*^H9+@xl^`>BS^k{$RlciqO%;LD!hfo>
zKs7d#fS(J6NHO7c)<`<@)PHi2Lo^3tPDepfqDs_#+3TJqc9h^!T8Nx%rI(EI@%&lx
zfDbB`5h#gj9Bp-Dv5ZPEKrvFgx>^oa9HXatKm@dScgy#*XU}%E+7twIgXzI%;uV!^
zI`g^FLaN(hS`N|jQn?8{#`PJ#g<e(RQ!r^EG45&ge-iQ)_Q1@v@P8X+z!uh4HH*+=
zg&eOW04`bL0=J&evagM1xFA_>5KV>b@cC9Eb!tzgnBz5<tRZR>cBmc|)~+b*@WJt7
zr3t?H&?C``>YdupHGO)Orc!tlyOA@B;G@{xtxg4VSI_K1MX5q&?a-=%=_{9Q03Dl`
z8Wgq-DOX#j7+U-Kp?{PGMjq8qx4LfyqKi%)Ap&yalzFusW*Si%-Bip?kx?aY8pIrg
zSfOj(=2af;R>MEB#*svO)C;*|P965bZuLyd+Y;=7Vi!wCHCzg{(^_JR7cMP<qg2{z
z`qYj?gwE-mOEqokMbK(yhEgYwR#a&%TMcq*Q2kf`G5-4U{(to6s@*|l8KPZ&`~Ku_
zASEDe9Hlqrq+b5|efjqE^GLNFHwFQ?)=KpPXr{<;sF4}KwtkKcN!yIjC(LUc{#rx1
zNA%DGQ`|?*BVX!U(0F&{YxOMf^H0<xW8!VI_W|HckaWWiU&@3j=nXtbU{0)*vk^HU
z4HJ}U#`AXpFMoAhryBW?VxZ*fQye8U?R~Drc@}GC_`Nv^0)=fe%H0e_>JEg6HO0f)
zL1mb6cI4qY&pXy4Qj7>nQo~7;I{Q)`0n&NIAK9Hceew8BmJE=uj>JOeC}q|2CNGv8
zQ;BhZb#?Eg_qtw;+*}W=T<@ZMD1aJ)waOvo>bHKz0e=DxAVG3L)#Q^0G(eVO#z*a%
z(>>K8j154%hDW3KkoPJ__3PT!2j{8Y5VGL1e`2I^?a|K4>@11wV-odxl%&^~q#R{Z
zxC~01ai4L+n@WkeyYe*wm*ba!OFBYpSHkhFmXdjT0b&5&d_eeC&&^jF01!$uNz{i!
zrZdyQn3p|%Ei!-SsLKth60Ya#Nc`@kKwtOt-KvqkbX+81n-aKd;A@nEOoJUMZx!)b
zh5hOyc)$~;GSVg0Q|T@6)?I0N%Tl9vy7KJClP`}d>|}Q5yKFVY<`u<RD))v{>40^_
zAD1PUd^1@f!M$7YU3M#E!mT!+=T1VMmZFq+f1R4MN>6`q(>EyhP=hxfpP&Ao>j*wP
zMtW^(%xkP;nbxaQrq<cd$6IO6;}3!vPwCoTdnmm@=&saAX*69<f<V#O)NQ`r`trD>
z&4HRp_sFE(Q{1YfuW~sN{-!ce-bm8L=LX4c=rT9PHT|@vaarYKJio;rzrE6ml++L5
zQBwVT%kzIH&3$<a$qpe<rN1inC*?OWKqaZ=^{+p!DTj5%q7ton8h*+=<*ZxE?Q@;h
zMp;JK8(Qj{xYmwYtb?{A&yn6T$HQNaZaQCi%)<*$FtT>mk5aRRP5N_`iluV+<CM9d
zGc?RK|7n=updT<M?nEum2kthii1|S2-~<4KBm{pn8v|_5Ssu=w>a)SrGJMMXHjk-O
z3!MGw+J-_SqujdcC!}FWru`@S2|>{>(0*`3F`jaFM==omz?Q;4Pk~2v6g|NQVpC6e
z2bu=4Za}67RQaGabCg7SX13;yMlj%`K@QdUud!yi10)4yTQKl1qNJd#D(lGoe}a-i
zlGA@H{MS%YfTDI;{RdG}2xKVrhoj{E8PEYW)%JlIuqRmllW4ZaiR)QJsEj*{#gXIr
zriLDnaF<x_`MRN>uk&ik7NAFhH*n@O-NPAY``x(|mdw23;`A_RG@R)Nx<yOw@lKKA
z<(|i5sb_Ba+`=jiOuJsB%M*vjJ#L>5NUj?gP*2in^7`R`1Yn>=N9N*lxABu|-M;~G
zsTP;P{RI;OHZ_-#4*?UGU>GecmsWu-lmV5O>wzs*f1P#Rl($kRtr%_d##5G62q$j<
z`~=4h|9dNXxQ^q)Ir<4)<Q<NN4#ifGZDoWvp9&91F_3Jhf{o^jub1w(=hwS`?|QeG
zV3cDX{`!6wKQI_&T^JBm03Lq)J$$?Sy{B5Eq<g0jPe#?&8Z!?9tBmo3)u{AOcubzh
zvi_L&e;QNtW#Q2NNfA0*k;rU91XdP%A>b)Kv%sqdP?G`)+4+H(LtH&kqxY1(5_6S9
zesCtOXmDEdP>c1A*r;x@L9}VLL23Z=q43wmQXyqfiJ251NxEbHJ??4|l_Cdf9Z7p1
z>*=_{4D`rV$F}MVt<eXFRSKJ*F(#WN4`nrRf7#7F3|0kgdToJr^M^Dm5WjY(xKtSS
zSr;zSRo_K<qQot8KECD|gNn&qQS+|PUo;9N*}D7%F(ldj{JkJw4S^^Czc^t#DK5y^
z?!;co5@&}nXZ=1jZ8=Db0{pDBHJvU=Pac~E+fPen0;!qS_ZE07l_4k$(4YDrq_VJ9
zf1d3J!WTAH;NC2?s_V`dArvOBQ0({?Lg}e)gsjvmL4_UFPeLIqc`L>8lTZljAb=~Z
zo4!;oJ1Ui`oXOF~;rc4(oU#Bt*PSXB0igtccK|0MnPU5z;JGm5HEN;?gXiPD)p0VY
z3Z#=hD{9=;%AhlfEnc8ZOf5OCBMmf}f9h=IYfF)ZqtKHIZecQPT6Ub?mVYZGBu`ZE
zVZljX&0(Y{C?qU`teddLf=9}kg+1TZbVvdY*P)`T73-M$g;aKdtX(h3D_!Z3)f7fy
z%#-xB?N9YN)O~zWu<nCF6QoL0r_LLRxn&y^C_#ihF&?{JlE3y99y>9+rs2H%e^rY}
z=Y+#*a@`|^!>BL2+;?s2ySuuxKw9KZ+MN{MQk#h-7@;TBT(xvP9464DI1cn&&D`vo
z6N{S7h!QMN`^2n2BvBX9YK>Hb5hG8{xd_+MB?S;g(0SScL`|j&v=X%tIlzWIEd9^7
zOr&D)(%d(DGHd0C`qN385@a!2f3NAs0yxCdJ;h`bZ2H@(arybV0lZW8ah>OyMRvlq
z(40WS<hz=_c53z7VUlMX&9<ZJ{5%Q&hbGN<u1d8iD>Fd1@R7iK!iJrw#N(WQqE-A~
zdy_<(_ZVxoJU&*JRLd8udw|fQGSZtm8P=1tczp^Mm;GzAW+gCedrtcNe_^^HN}2I~
zp^F_OH$^O#;)ml{RL7?U=-tpIMl(>)>4t!vKq|g*x17Dv3!-rDMzc2tO8svarCyIl
z1?FqdJ@q{_YQ+hG`V;Xwxi#YRB+ZMA(a4uW)?@1};hpG5cuFPaOX1IC?4I9%b?~Ru
zDV$c-`6*XA)Vfh}Pt1mye^&4Jf+TT@52n|;1jaTR^2#Fir*4>y$bxgN|G@DX!KaaN
zBRx{B!qTgr#tzvKNkIdQtDyFh`CCpZF!u8K`&y}vaM%!R!>uSL;ZCkkn&3fT9J|l@
zI|4Y6u?arj`K=Dc@>+2)*HPen{k%+0fummFy(>%>%X#Qe>|S9mf8@Nu<S!wuq}RQz
z$a@BVB7r=yg?7r{VDL9KfvcX=I|hHt%4@T*(8l0zgOb?mcfYIkfmja+FP{9dZ+&HD
zZN|rG`&&ENzVGJTrIZ%84|yMr!$#wEuJ`G6*TqwPmZYERGpD_&9^J}R?=G0?r5G4=
zy`O$&uD4Fg>ypx2fBv~{uwNU!0AoR0+=JIx=?4A^i#CRO_gO=|^kRiD-iZ~O(^!ge
zVC;V5H%4lPTTs*GKV6)1(<eWHbiPC_)sc4^gZU4}$s!f{opJ{<6LZG*^O6mAO^-H{
z@7QkcB0O0cwXNR&tBg0wdFu79R}5@y=O;F9+fQYG9V~O?L$@;(K)XhF(~1kurzxeH
z?Vwj9LJ;uu*2zseYG?b6kdlFRys!NwHxM%kZg~GN<8#9qLHZ{lUo)_GZvm$N0g!x>
zm%+j@69hIjH8Ga~Z!8m+t$Q#hmq~~%n}2W#l1_hr{_(<-v9DLp=PWf=F_fJxRj5Uy
zB!#x9D}K-djoQ_@yQa~4<)VhytAkm|S~H{i7`%6qQ9z6Vz)Ecs5qH3ZaYWnOHADt%
z(CbfH9y>ic=tTJ%^(u^&7Tq<BdW;$KCWI|~eUM*YKVSZPDN2Q8+W3P3hZc#$KYu@8
z!VLvzgY^d?m<M(E_3z>P%b!$gniu*VK@|ha4f;~KC=AmYT@;#uaO<WDqkWzXDrX8+
zlz)ExIFKvfK6;<MVn#Gpz<c;qkjo<kmkbbEWJE{;r8}F#L};CSNWbNn50};U3L}{D
zevQ2rp~T&Gd&j_3(s2UD2~pa4W`Ce04HgKcj$I6Nns2R7v1A0`>HzZ&-Zoj`!j&*)
zWX|)mh!o|^bIr&aSwneIPV#-AsmPnQ<kraYD@6Cb^EgL%;AQ>%y26A{F(C|%@yddu
zaHgCQR7xn^TE@a1uyPs3GWz0^2G?b4@Qa7o8WOWv4?Hk$8xQs<qZxnJGJm5@iA|1o
zETu~2<!TgL@lGtOrut>m^T~6RXGMw_2J@IQ$PJ}b80sXsQ?x;6tvqSIrtByMzvyZT
zeUYVmYvP@^a>u7DWnqrGV3`#qPK^}(mi)P-3^+HQws9-JVx|Xqw)oh_$0UoA_i0Ic
zz2d?#H{aSbUn-rO^lTA!lz(8L<gn?<3v9IN(q}`6j8f4PnNSwy;xh1+rxB?Jbx$iI
zv#Q6KbC9Ezszw?$n2)4&WImLl$V;P1HB478SwkGf2Ufdm!!oN!Emo;nv5(L|kRIl?
znM=tTxUvS0<97myboCfR^uk&1vAxW@TvgO@ypI@USe2tS5)+Ol?SC3po04~=x(uVC
z9LRIBKh_;vRhP!5m~*8zCBNS9Yf&*4%vncUR4=hx_VX?3f|gXrUf&Lv(mXl3l;G1B
z!$UyjvPrMtsL%_map5z}^S<>>fG7*AA3*amOpqAfds|(ib)<FV*ka;v9%KmJNv*DY
zD-5iOVQ!C!7U@az(|;#v-$ahjiV%(&qo_pvHAP^9tuj!WCL+T3K#fpFdjYS3U1o1;
z(TP;U$1&hE^QtUXr65kH=<0b#E*Y3B(X{itV}O{nifY+1@0irSiO<)LYtO|fMLL<S
zdFMd%rn)lcO{xUSD&~vwoPcT>5Q-2JHY3EI<z$ARt>BVaM1K|c)IPXQb4h6#SV>3R
z$t9aq(2{X0pOFf+rbnehtz~;EkoY_~6+#~jYHKb?U=VJNZJ({5&y1W}YTo!F=mmw#
zx}*<lotvn(v5J&)blTDe)Omf<byG<U-gynqiCt^i0zid?N*XFLP<_M2ahI()VLh{8
z7sRHFa4v3rP=7z~E}?gA$)Om8HZ;)Yqc%(98Di>Qn>*Cy$iqI7S2(-Q%Y|vAv_eAM
zr1I@@J&5i-n~2h5lP~*jc|aytY~-|?7^4)lA~1K8A^Q|EB*J<l9z}+vl?**u*u6{#
zb8N+z*eXl?Yyw2IK8OGjEjth(B5_9oL?rJ-fQZxy0)HemFzKH`fQY6Cb$vw3=B|&%
z=eb><R2a>zO`jB=VS(IQLiQm&;T^Uz)TTMole#G~Vkgpr%L3CvJOlo<rbofQ*0LS`
zBtB1uKexEV%8mFJ-Y~<}&ZLJ(ohLoy;om}fsKJjRJw(e6q=!g+2<f>U8%lbppR3B|
zeMASLWPe>_>lD$^hT)TI?0gKlL0m{<7p98}y_SF`Umi_vgaO7~>9S>c{}gG*?7<`z
zjtD6#VUM@xNVpV0BUz2nm+^Ppc8nffrru!p@8!qemQT~VTO-@IceWk1?(JPgKj+?M
z-to5YULq)fX$j52Eu34gHxof29ka&#=7U<qihl(MwaADk?+yuzNn`}IVxnjYCz-d<
z*u1v5?-TR9vqfVL6FXaqIj-r`D@tdBxPV-4e>dW4E0!Rx9`S^wIa6+;c?xCeNFcRy
zOJ-CtKf!T3=t$UafUbUe^LmXoI4&Jt#y-ZG*oSh_Sw4E_+sE_@oys$KfgMrBf<-?3
zSbvTXt~uol2J?j9tjKXWREclfL4%)z;TE5M2IUysb@+Le5##-c;65HduZ(!onP>oD
zQoxhWL<1o0^md<#?UJ+WKTi25nG!t-#1?4%tK)7<k{CKy8`G;}GD0kDi=^(ig*P?Y
zPO!nLl-YyGYY&69Q&R4<Ku@yyxjU8>Q-3J@UBsn&Z(s)zxupH3Wb2KL70xpxu(iKS
zqb&&FWb4^nqDmyS9MMtf=9h`Kd-TxfZ7b)G6l=k-X4Mu9<&NNg@@TZv3=m_LtW-yu
zKb=6XDSED3PPnGitMB=2aQJhVK?ve=`+_S<IybMMvdgaGU8YX~e!T%J8(+fJjypcE
z=u4xqo1_Zk$<0gVp~Vk<5@IBs?eQjyR=B`fw1B7?PT;vmpL&}|jN;4*NraIOCV%5{
z^FNe{%$GqRG7|zeHJ1Sy4HN<~FqdKNIx3fDkS&)1&X@j>Etr2P`Nk%{g*tn$^{#sF
zm$aK5H+rv9KXBqkyf_lxk0(CAnBtA#&!kV8mP)#cXI!%mRcASGg8IVk+wi}yZ%_X&
zLV?%;hz<J3KTqKSDiX%}2|##-)Av88zn*>_sb>3(_m~o*Ql&9Ob0;u@$W1JPaEk6X
z8Ec(W2m47BIP8Cm0sE4LMme=V(M@M7uU!i8PB^DWPSlrhttJg%htM9z9U=8#F-URZ
zYch^r@X6G_o<A8Gl3IK-OGWu9*}EkH4+7l*K|+l3ZYg87N<?QXuQk>dhGR*DcYY!K
zH3?qWt|#4Qf6GO<1TaHpagx0Igi0|u9a9=LgsW!epca4DjxWpWT3Fpkt~It0g{h>t
z&LN&!rcasg94jBGR(T#;ZIC$}*pRuTUQ-g&tKOK3;#wSn>3#cPu`vRz85k+S5v!l+
z0w~{OU_BRX5LvmD8tqP-nU<gWQ)bcc?9L&YsMK9_29<4urUiI<!!A<Rj2CECc%GdP
z^HiyvKgEAU;Y(^MZ-gX>*>K@WxXY9T7tW`FqrBDIcG|=IOFO{_3<r(=1w)S4+XeEb
zA8{r&V8#<GB|CSbY6utOxPx|4(M8$8%-5vwLMfWRnw*}qF{281%-1>3klV}I#jR89
zB<h!4IHD~HwNUa-3E5H}O7;r{Bc+TR7T%*!SOb4Y5pt_JkMJu4f~*>Opp8U@*@zWQ
zF6+{!Ln$wSQ~Gey2clr~MmX>fM8WJ!4Am;d0!88lnJVO<@U6dyu1GD@)qX-Q()%r>
zjKuZ=y^ECJ#noAzHW=G5D>y+o@b#1MbO6v`^iRUm2#V4B2N?fc64`@bH)t!A)g!}u
zup56_b1eO^<6hX!)iJ~FpnzuayCu+ULNMxp1n<Jz355=a#F}|chCr$yVV)*7U^Qc}
zkm3f4@4~g38&<AM)g&3n+FS$S)Vz;OU;7yR>?RcE)oL@+-*CU}b{aS>*^x^0c{8uX
zrE<5Jn8UwTxrgl2?W4{K=a7}5eEFB@dfa~s8lywz6QjOlJ#B$4=FYnYFylUzF^o)m
zqbzn;=O{Tz@}p5(%V60A7VKH*7GALdEVQu#u>lL$#$UL=Nd=1q?=qcyNsl2W)-l>>
z-?1dg2Hc^C*D<y#&H1kSj?27UI{s9aqAMWH*CKD1I7sZ92&g=$yvx>==WItcTb+OJ
z$}os@zDmT|vBl}rVdBc|MgZ(YK=G;4etZ~E(&z=rM*<}sD0|;~c$D~Z2;^wVbzSa*
zoRQ*2enQ8k&<zxS5<CG8KaW2Po`6LGYDMK4SNQwKcgPymBjFtw5o(I&x;O0Dg%Q4u
z+2b1Kp~Opiw@gg9nuTCU2m?;rv-p2|aFPk#nLi-ldr$^g#jdxoPxzt|oi4+SoO+MI
z42Bq7x^cnx*P@$PP+%FnKQPbxSlwt%Lc&54D3E#yz1}tT%<ybHZJc^W#z`g-;t8{r
zJ$VcYCuubnyB>a#OgH#-Mo0#z1GRU-NzhJ6YH-q@Eu3m9ZFA95)^@|Qhfsg5ST1+B
z7){?B5wO4!oeNwY25pUel)ga9W-H*&^iZH1PGJvk`eK|WgZ9++b?~nP9SO*0il%^>
zy07w>;lOj-(a|#l*qzDR^j$$^p!PfiO=<=K7B&Sb6hQyIAjk(e>JiW%uxWa$+HK%G
z!l`Sa9w}pjI{Ca~)M-9Lid271e|B2kAJ;Kl0FR^m%j_3od@~BSB1Th~SHVa<cmNlU
zMWZqE%U!NLLwIhyR=7R@Xpm(lJ8mwHS9+eg>1Tb<b2S|%Oqx}`*;k%DoO0Mo+1zWl
z&vBl+6m`@vC0x5Qz9l|sE`79Sl-0nHjQ)&m6Pq#xo(J|kq1Z!d&+UIW@w`hK`cN8s
z+R`nc>*wKp-_(V}YI3v>Opb@R&S?NUZc|#<1=zrt&hIHHJA9OM<Ey8PHdhReK()wF
zvtDr1VpT4GZc~B!ImjNEFytcmsKGpNr|<kyp2GP&UECbLBMH(pQrjpf+%XBT{(t+)
z&Tu|-txVSs9Fbc&E0BN8NKw8#j_2^p0CeLC6$Ze(70a`iLpIaQ33tOE$2&0m2CgO}
z?>)co_}P=?MX5fLe+R;Hi5^b)!5Gof2!w?tbBug`;U0SjSSuriMl+k!UfP5(0^O43
z&5z9!3<=Bn9L5PQG_g0P38{f8T}*3f%F1h=FsyeqWx}Ae<-UKj{>Bi$v~PnYk%Kza
z*_1)SK=BbjIdUqFkOs;_kJo~ijb?Ftd-dB!n4?@{mz{XJt<szNWD7c_@6acA)|Sch
z9#wnO*~;qyozG7NeX}I5YYv5hg$(Oou=+1U^msjb&6UzT&@BgoyQw_OeChgfPFn3x
z8qD@9y|-owIEE5^`(fyfalx^PA!m^P09fPpj+f!CEfklLu`v^ur2j4{mw-<#1b-wa
z5Wf3Y^dDH30*Z$;8p$(jN6IZL-=dtN?4I_J{r`s)P|!4V&&%!2q1`qO_5A7qSWe$3
zIsNu@dVBi(s?FqtM{sI-`!)f6kn|HI4O*NWfUp{-w;$7=2Pu`M=R}W@Qfqn~{qE_F
z$4?qO%>bl_z5*OSIRkMG&Cx2a)qggx(KJh)=?(OLj^ngGNGl@+q6hR1-0`os-=8E4
zD@CyJ*ZAuHdU<>L_o<4mv_KmstEEs<O@IG*iXRN1ur5p{2oHMt`Ooz2>5raj<t1rJ
z{(*A(`F-j}zOwOxB;!eI5f{c&Rv`Mx+8}iNl#OS;NQtDJd8LOGG-M_P3x6d)zj|1u
zh*oGF28tn5LUM^b@%uGX6eh}Aek2tZBcwv5?yUgT6VG0xM=wOIi-H&4>wc9yGyCfn
zylOt<gwbwK^K}%X=?IZVElp`67uT4y=!ANje<Yn2Bdtk`4%!OieBp>h6_NGum+<{4
z>WvG+S=*yN$)oHqfjtInbAQGTe(NZb0VS&JqvWecAP*_Ob(EFdz`7LLnSQn&<j9zj
zF?+r276<6GaA5ARrzD>fc}jB-86~cZk*_z{e92XjvKTVAh~UE6AWPymEN6sOgAOUO
z)~t8GCzbNTTGMmrvesP`wG-g<K=E}GABu`f(6(QJr~9=*FO0z(cz@qT)d7{l!3$v4
zwQ(qnRfErVMm2(Plzw*15*@{l&+cfX@J&78_=q|r&V7d8!D`?|fUu|4I*Plt6dmy!
zTZ(?w%kh(yiiL|QBPF25z;^|=BHDu?hu|tcku`xXMlQHY84yOfp2LOGzvZHu43?C3
zY_UwBk8;ma-BbMdAb&IjHKPBH7K^58Ps19mx3N?P+BRg~YhUK#Ic}Ru7Z!IJP1|P~
zw|>NKwQKU!=0?mA<zW3<n3_^Vi>)nl@!1;5*6M2ch9qbbYE!);j6{L1np`n#Vm}mj
zg)zk6Ctwn0l(7+TecKvKcL=5v@cA`vFTq7o!x8m?Jpn9A5ot~3F<`{s+j?v3g$C$b
zZxs`cxzc%e@6=^)1RBGh`cZ=C2LK}5zS=v-sDA0DH_KDG^N(d?Yek^2?v#Nq@#$OI
zP;#3>ZfV<*OBtCPnvRPKmvC@*v$f?g4b(>A$bG$))C3o0p`SgMPn<0ne}it%j2_^3
z0hry`G@aKXu4cxKBdQX1nyRGItDCYj=#%O)$3|+yAdotij@pJ*H)SGw7(Wn+B}pAW
z8D$O(HBL3jt!yB4bp|_{n9&FxI0c^Bpd5mDZs(^juFmJVHQI|qu&lXU43fE+)&=QZ
z!^t-#&OX59nQ29ur@THpe~)%iI-U#WmZ8dl4L`^ZYWSvh<FdmGd?=q9I|SNxk{g<q
zZ7@(`3fObt^78qWlK3ULq^E|b?q+n8ze^i|aG5_-qC6+8oC5@p#vFQT$8>s|JQ<6n
z@d;P)w@@1I?NrILvhUb_uCBEOp>-tb(tpp0pZ&}<7r1DAKGSFPf5&)eMo1%j$jDFg
z<(!hxwOST@Ge0~kILyT_f6e1*3*y^#N!zMP-I|;^;+lw!vCOkp5@0+n$JC(cmMY3>
zt;VvUu{|aP3?6r0STkcfM+3J&RqiueA$k6)hx4q(nl|=~=V*n-gLO^j?n`cSO@Gog
zU)5H5wr%u=CXdyIe~x|4KC3#)hLf-I7Ei<AKbDPjDDvATdsba;$!CqFV4GLLV?}SV
z1CQUBC6txIK|Ots_bJJeZc+5EmQszhT`|-?Q>Kz7=bR+2Z_^F6Q{Hv6)16S%m{Q)M
zR-5b%jvYp9nZa06!8R5uzD#qU3q{l5i0518A@diIx7Vi3e}9%{lir0Y$~v?hOSYh|
zP>_CJW*lDCVV335WA>RdR<S~=J(hLR{HVEuU~L%xF()!lEWB(Vg&E<CJW@QYj4%l8
z9-~;T%OzFbTk&7>&0K-1=INvn*0EQR8mlAR3e<LT!fENQv~h<fVotMvxdN{Gcip#{
z){u|#C|3l8f4Q2#p{;e6eOoJF$n^{B*J{yReKgNC<v5|*jVg(1E6I6L)6a8`uRV=x
zvR9UO*gZD5bPLVyw@^7;dO2!StrRs(x5J8WgN^Sx+u4$}=A<}XMJ|N6OP()EOAq3D
z1m7<I%{iHdg);S2INpzAsbRA5oPX??fws)n=>C);e`p#FzvO-oOb}M99S;pViaXx#
zb(F8%?{%|((fuAM_dne4bzI+hzt>T_aKBfMD^BIFn5$H?znw=*Jk}kryZ?a6d!dva
zw0*~28f6woZB$2b<1US?DuN%Zy*Gj+-g%d{3xbwh-;1CnwGX%<?l^QFd*p_=r+EE_
zxLbkee}`{~uk(K7hIp{i=4G9V|7Zek@LTcEJC18h%y#&y)%d2`QX_AWZ@=uKsT-w!
za**_KKWIZmUXRFHZ*Zw463&l4*f+a)-QH#WqU5q<s(6=^a><u&hkh>Ss24AlmjtgD
zC@oh@(gXj}VHH8Y*GRr>L{*N%<zzf}PQB$035lA={{Y+PbdQ(et}PS+IhO(bHWQbS
zpe?(9@Dj)UKOcYYbP-moVB>8Z`lq*#-G93#IvT|`4CI}1#t#4f+{F(BtgH(I1XL*F
zuW!Tm-CsS~#!Sna{KI<q^<(JXycg>QE!NZ4GB2!$Rx$g*+MrDQl&wdvTC-?J2f0K;
z5<4L*w0t~xI7*SNGA8sCBQ+KqlGv2R#v~Vii*2cL5e__N4vM1e^(U#ccq1xRZm5j{
z<9cqW7qYh5(MQOxv<+EH9<y9g#`G*cKz5RoU6D7+9*WF7S%~8OVIN&7O_!y5%wt+w
zrl<QEiKu=uwp{wKf3P}Ikcrt)ZZeXN&h}P}bbjB@ZaO(dV!Rt}`jR4>$v|=Qqvpwf
zB1d&Ada)U%pQP8~&200V9>ywizH$|&g=4ej81GQ-^jp2-mo8{O!53LEgvQqv`*F5q
z(k$(LlOs9*=lL3~Jm6q~ByTh$H|50Z%N@|ynMh;_vBu@)2p8_SeBCA4ON4s5p-XlY
zFUv#OO`A*JKrar;S$j*pmK5(4YD`{#dF}2VYHXy+lfQeYvGDgktfIg>xM8fZRQzC$
zFqs9kT;}VHG^yZV9n^ETNydCGHH^Q^lL9FXxK39zbJtcIUrx1BlN4SXFRu+-Z)LPW
zGF6u5*q8SfPV4ru_F%kluvD|Lqzlf@MZOEr6!M!5DH)JW^;=EGy80*Tn{rlvsvz=G
zvnG~A5(6=|xd2yi+Gwbqgd9P~>G=>FUvQCU*t40Z(%Rp=6q`Y6s&k6DvK2PM9K`}%
zD~zZD(3wuS8&lW2kwZ&XF&@)+@fMYw8m{hB!I|iiJx6$r*U?@Tq;ACx?^a+9OsD}`
zDoDgDn>}fsYigYfvl?*Bl$)-9QFn=?28C8r1yd0{O+9iBntei9u`*=0NiH%b_bUkO
zh-20-IeXWq=d!ZsNn`y&)Tp1-z0FTueSQKnVoFCw7ujQtERMU<7`-Apq06FS;>>$Y
zRW5ZBRh}e6Xc}PMu_j6FKl~(7Nx_)wSefSwmvO=*d_qo;6o`vDNzJl<Nv>*dJJm;p
z>^(-GmCKu&*jH~)rBUfdw3M)pKh7d*^1d#OZupvjVu}&7q$mLCW-<%DCQ11nhlc#H
z%LvIsE|MLk^kIzM-zGa6micZadz1GvIVlKS^S@J0CgIRCXqDqiah;c3zA|eGg~x+r
z$rDKnWxF}>B8111G<0r%plhO(fc#YTLhV)1p|4&JlrwOba~!lm1%E@Bu_U-g6W3S5
zQrDJ*Hx6tdNGIp0n{19@n(^kxBS_KR@Csu}yXk~Hm;oZA#C7rJ{Q`q8xuQ8++m0KK
zl*N?}n~3l*!*5rg!3j2#{$?e)UTy1DR3$eBV|%I{+M4GXSQ+ns)(#;f&@B&1dy1Df
zpyQW7JW5^yZzq#x+_1uz_rhM_q<KKNA@%|`3IG;%t%o~qLb<U9X}Sn^fg$C9&*H9g
zT(&S5ri1Avmpfsu5F0JN6XXJIQZc+}XR^0bsnq$1>(d3N<9rHwm8#RNI0?OA*5n^%
z=Nf)W-PrU<e-XrgS6@P5*2&-&t@cq%PObk6W^773hadK7VgCen>i-+5QEoms2S3Rp
z=Y18mNUpT=c~4y<B?u<qG(pBH+jE0@FyrqQjrmp-8zSXBcP7(sCxg;3D$u@Ll<WiN
zImW-3ob5Tjp5_>jUP8aKb+>B4X4~6&372R4Z0Gjt!EoIdq1oBn9Il#~m(zB(=~94X
znEwH_lK$hD(a$Xu0Wg<QlNS`1ymAF6m#C&KoPTio`|Xb>ot6D~VZW*=8pSpY){zp~
zMN?5uvy93z8i_Ox@=LSpcn{JDn-~<yY!zspBtv;xBwez!_}}Ld1PgM}67nZ-^2>r4
zn{)^9dF;t{{8U}ywLED<>|w{=@H~_2@_$!)v7gC^8oZQ|lSGq8;`)4Q=S8nD>vb91
zVt;RPYjUGrNft~koa`)DMMs~<j}p(*crdelZq4{C<sMT@vJdhYl+iA|Y<Yg<n)mSu
z3@1a+lbvyPyV)dCK%va$e2$WAT{VyL^(Aw}!zxECD&x78I?i@RQ#90(0k5YIR;L*a
zbtbG%j5ybSBf8^NT1lVe6J?xt*@_$#k$>y#+hLOMZO=9gS_Qe)NYtN-Jqev`Q{Rf4
zt?BAqKD$;KZ<GPsS#>TeX2v8Mxg)+NGk`+8irS?<uOm})i=O5nV>3Nz8pOhK9T69H
z?4rk7yoR`}o6yTwt9&)6Akf6D{<(bwft*6|UX2_mzVC`x1yX2n^8bEorHsB>1%Dge
zb^Q6#Tz!6f`mZyN*vK|xhJSuM#Tx{ytP7&E7cQIO=fA_Zr$1Y<c|*`j%QNE_>*443
zp=tS=cetR%LTxSc(n70PjOQRb9NCFkpGLn^JbSbhq1lPjjQ0{{pD-r05)0EN-+0Mb
zQPi`1CG{6AqW&b=L)wiS$*324O@F;*-5N)2Z`0{9GBK_~_?&gR4(U}qZBP!w9e8S)
za}f__fH+VZb<y(m2F#bEqM7aZTt{x!ZjT%(i=7Q=y{@6Rwhjd+Gt{qS&7x&KuLU#q
z%DKA48+Z<w6k}zL<B=p^O62qXq_(8IY*>w?EmAw#2-tSxu5Yw6GINHi;(u0Jc5p*&
zgngZEvyJ-}tPq96($q>hKdv8oWi9oRr=Yg7=lYCn>QA5uI~0KK9-&GU2QP!Q5<fj4
zF{ofVUVvL=5wdK?9eE#FxyZ@~Gxro_cXyN|Pw~+*`l%68Z9jh53yp*0BbR+6G0dWJ
zH;XJKHq(@`I6s6i1*0_Bg?}&JMfbNLOHOvC;15K$m3RxXv)oU=t~tXKJRjo^vnN<l
z@*aQ4F|6ZG$^rh+S`~ypo#bQvp_8xxIusu9hdwSR=p;Vi54~52wv)Kw4+ARd*xcl1
z?^?AshHo_=ZkYPMRa`ZifLTurWGNai%@Qzp8>7jfG3&i{r%w519)Co*@Ovrm#9X1^
zQjKU9L89trWz7UH3|n~8sor7RI+gs#n(dZ|HrHl$>l0-(p{}(xKF<Cs<^-{h2XKG}
z8+8dyDrjmnc)7%iU#h%dII(%&A3JBOcjuQz2O%i7`6aV*0j{K%&2LI2H(Yooy^*xl
z`U9pkx$Z?VfW#i5pMNn<5|fS8@*NSlEC>l+8mXz^)OsKRGrMEtoTANJ_O?t(ra)F%
zGfJdhDgnn|f7@2+t%6QaTQ@0C$)q^z)+girP%WkYDyWvDn4<TxZhaN5TnA{R@#l8^
z-;D;sSXpS%fb~sAR$XZOV$Q}F3xMiEX+2atIVjcVDK6q^_J1Ei9fN3QI-E?d`xTAu
z2y_ghLJK|ecOxAoFm!B&OUJwE3ByXz?eKnV#towqhjrtO%kHWK)n7H|(j=c0+jd3M
z?1_bPUi&V=TXXc+j(=j*XK5;&8a@$~LmXE^fKxjBiiP`eQzP2nmSeGA;^|iQ{o89?
z<S3y`BWZEe)PEGT+16Bon`53)`$f5Yv}boOyK;<x%IeP66^Fce2mmB0Aapja9Dil~
z%@RX)<#=*s$K2989z96d*NwU+VO?+1D5)K`5~g70_%KzRj$|lsV3>*~uB1s~j7<vj
z376^VRA}Wq;&qItQNW4A;ob1`y(4d}8&=c~{F|au;D4B-xi%^#QeN0{bW|FnCc+11
zaH{}*f7l?iZ34C$tpria(4py%FV%ld2Xp6fip*}AerfCd`j6T2?dpp*D)`QTMC<rQ
z#QjBc14cRSA>7!9L2^uOI{>fj!)qYoE=^8lZ^=7t$3Z8ntsW!3n}e_S2ktotZZz1L
zG?2vHdw<hDvCF#Mqo(n7mvWSR7#k;Y-E3LRhADKj@dzDrpxf%OLB|}-w#Z(i;}vw5
z)Eud}MxYI9T2iddqtwjrS=>+0@x-ywV$UV!qp9WzwDb3f8X(U#*!v|^EM?3)tjMl(
z`M@Jg)U1$9rvmJ+ga*WG3Cc|%CkEmzym=%o0e|%A+@N(&Zas)k!EdO!R64Dwcgou`
z=}MB9(w+@YPfYK@oNSDYVktM>H#JXAR)!SNPl?b&0V4XKxHEQzuA8}o_Tx-1C1hT7
zphDzTtx**S&*RH^$o{#ziiS`I^t=SQW?7~8SV)+johP@KF6WDd3-2*<T|VBF;(OP+
zlYbeY0VM=)4VxBP8(cTZ!A4EjS_R}<sr-^nNU+UKE<DU;Q+@w2vCCG+K7TK8>^*Y|
zMPeU{&Lu-W{fe#Qn}xIY?XrX`Tgq&DXKy9eNxQ|<#(PzV2AehVN!&qPbGJSRzYVsd
zm;*8YCEKXgGUw{%e6BN+kbS+GVMch{U@S5#!5y(SoRvOs#ll;6)^}QJ#8TVAeB<kR
zUQ0mbZ+$zj$sHv({U331P6?Oct}PS-F*uhIyfq}38?P-30X3JQf-Ek7?q6|0zoZ|$
z+<DpA-6AmH#C4iN0R-Cx+8RbcNmMLX7FCKeoc{Vgv*eMYDcY1oTZ#eO!@Ieg*?E6<
z&r-B5c9vr9NW_WKi8(>3)2dEbmxMd2nYxtQ748gIEwObuu@ki`srciytHO`K6FZ8Q
zoSQnL>Nlynkj9U@+U_!c?xuE^A)|1k1k#M0Djmn#_)^Ez?n>>LO(8>ZO|q-c+OZ~A
zTVlsnT!v7`wW$e_Ldp5CMB(CV5bZ=MJC(5F<TAuMDYLt|(@5qrs8d$wQwCKMjG@-4
zaJwd;taOPWLzP^HDkxYbC5xb%4W)GIf*=RiteGm|EqS&Mr*gZ0i^pPYlz@`Pr=W_z
z8A`(TU}1O-e(^VJo&B`rYKZ#E@hWKH&q8rLRa^${TJ^gL*J{(=)ri*KT;b8`_Vb^!
zwKWH1f5PxjX}0pnsLhfPo?~fGq%JdM$S{P1w7YOh)^-;`lbe$(w9oAY{?cr1m03VF
z>msnSmD*|XyIEX+xMfheJyE+{V}%U3r$lPcW>D-aFVq!wB@-N981W(ue-x4S<^l>Y
zFsvF;+MY$g6=RnHhZf7uSF9_Q+#e-(HLCNK(p8$>^}vo!)!0>So`GwtHNQb5(gVRz
zl2a|3s)4HQ3B0dJ(||w~dSt$QiqVZ~NGFQKzRp^m^pLiH=n14zog~jC4<C!po2@#D
zEQvKXUSuyt&tF6Sf`lhP;$@)oqwl{Ty!!ooG<bS;Hl7Y%ULH*S*WSt5f4adB<H^xz
z@;agl#5aS#4SpW{_?l)vUW^W>uM_2vu~0ZIcrr;26&{DB19W~fc-rCZF9$!3UyTRP
zy6*@sLApqPQF5ywuyXzL6Z~0K;(EDQaYu8xi1Xm#K{*A1c`0$I$bTvnO0_qSR@O0i
zx_JHJ^kjSnjsEfCXaD=%+v)WD;xB{2Zzt2Ymj~f+{C@E6^8NYaqtS=K-!FgJ5ARL~
z2dCqM!SQjidU(vG91tCo9H|_ujPfWS9OWbB<9Mup^|;dDVsh9&`KLV=dA!G&vB*cC
zuJVSaW~~9<?~H?@RH9A=LbzwQ^tz#AMr|<!m}#uA5GyYX{p>Z0t>D!!2LJrmzdd+D
z@#I9rpIx4wzS+Q)!39f0zL$$FIJE>ub}Tu&!H?t9@#N+C@Nm?%2fD%ja5^1L&N`m`
z{OiYm=}#}G!|AA_*@frhvngaeM@6QH^}3YjNNHl7iF}TXBX3OYNAdtizrl|uPX_yw
z@!`wS^z~r>+4I4x(Z}hVD^A!S{x*U_XVcNy^uoJDZm)PTx)@(h4o4TWw&It586BMr
ze;9vk&hdoI6|ucPoZ!jMP|h1ybLz#c-8J!lk6yQq=gVjvZ%$B8^tZUb#r-Yjo@i@)
zbEm*TYYx%tOQxW(6C-yF?EpnPL|u=aqG&QmCNv$+XsXQVnjF2M5~>os8qammTte$c
z8+7<={N6r=B!yXBufgX5&N1XxLo2k<Z=?k^X=u>}3GL%bL?)wlq3wh$-$)5eEg|}U
zOptRcBM=d+x>6_xly4an+ZS%M!hq>wQMk3DfyDe^3bzmnw@BJS;id$WMjsTBhZq!?
zQt;%*Xc+2s`9~Q6S!$gbTP5Wk_&e87QH*;fFhB$p=#5Hzwo`k-XIm6}AL3Se$JeRz
zS4#Uz(Z6Y(*SJ>arQD{@&wkJ9JnOuFVoP7lr7z~v7ju!GSTLuau%5`*6#VVe$}kRK
zQ17z}RU@d-!^tWs$f}&L%c|`cnGyx@VQ;=lV3vTn{=Qc^JU;$)^znQ$8~jqj;Fs#%
zR;mRJ3N2z$B})#h_r1{nUJj3zD<!kXdjKW7sFX&*p&pzY6v&A=1_mrc*udd`>rDyR
z3vd{aKYC8#X(>A>mCQ1e+H29k%w{FH0_jQwenx{PhtYPY)`-><cnN8QYRKu{bNFz4
zb}^j{Pf$Z=ZpYXLzsD5s7UGy@Fpbh;xeewR-}iFQPS4vilksExV>^s)jJ}G|w-4?D
zlvMS$v3-NqfT*J7z=a_+)4k7sTjPUw(DCHMXlCsEc+ajh2cI*fZUa<TxMuLv47GCw
zes1Jc3}m_mqgn~R5wrzoN^Q*gGABz6zQ4K!C+0@}rf~YElU1U%$x7>OCaYx2$x7``
z6A@Jcz~SW%5zw%ed+(taZ-?ijc_po=Fl9CBQ`|gVf6A~3V?e5k;cp9nPM~nN>z6sn
zi<R*<oCZKi>1%}N$_CFSNO03wwYnComTwcQ`3efqJRjq=%!$-rNsBA}wvY8l!Fr@%
zJyNh9DOisbtVarFJJ=CQB}CtQ&Yl}Uq)CIxXLe&0aiAQn<|w&T$?<O6wlEMvH7}bH
zTBFwT&}N*B38&Cbwg8ZSMgW-!!%jppyTR)m$-owC22VLB75p_V0`x6fcd+l|oAZ8L
zgpOkfEci1J`|+J`{nTr=0IH^1h^nFlFh&>M#pc&+2uehuk}js8#1yjG*B^tlrjU8r
z4^rfSiT~aQq4Vj<`;-5TCch&3Mw4UI%=!E&=nk3gHe`z;{^D|f(c<wJAMBzzMF-Di
z7pEqOuj*%Ea62Fi;Dv>D_8TQoUe7!&NuwXa5yi?Cz-t#lKd-lbTBQ(9nECw-n9#Ag
ztZ(`H@hl~(#YE6=kbk?_cE%vogwB;<p}-EjHYh=&F()ptoWKy-OTjL-;MLCo7D`~g
z@Kex@N<?-%O<J^nR*MQOz5{<?QINXSLNLQ9>qad;7Zg4YS=k1otZ?<ng9B;ZhLiDs
z-iA{PgHyNR1e}3Qx8M{3LP}e4GV)674|C{O);ELmHk-kz$m$q4({eTj=DTg;BKv_{
zmz#dcao@e~h<q4MPKF1kJ!pwZeEX-`Dsaj53&15`1C|kgU6o}UPWE|8{xBe;%35?)
zx=nP|vW%|MLxIvC8k8PnlpbW19%R&bw@^$LvMzV$)OV5efyE+o_OC32cQvYeulAl`
zWznjI&?4%$?bLG>`bND$g|FJDdlgbdqts0*REgf5p=4HI#FAwPUkq>($drW&xrVse
z3fs+XH~FxCC^vkv{IE!aQhH!Tm>i)3pvBc9TzQ*$WQA*9UsXa`-$J^MlK~oe8&0wm
z{EM?KIAQDbb!LaJQQa-tr`6r3shd}KXINDCI<MURh2NZ@p5UI4yA^f(Uu=1P>b?}K
z58%jUd8(+SkM|IccRXCO*e2Y?;gV&bz<bZ)^9k^O?$zU^x(lQ$F~k=1Ib`;=1M!l7
z`kCR~vfeigR*8&px2P2Vz$5QslL_cU{Ug^P>N;pB?MqYut(2~*+7crfri_@Gy^E9`
zG!CG;0z=Go5cnK8YoiWGeu}Yh^-Zo=6Z(0+ithYd+|Q1*R~Z)Nk>Cv8f>C@sYO!nn
ze-}{!QI}EmE)xPaHkYyO9TNdDmoWD!D3`0bEt~=Pmq@xT_kW&*emYpiAo<%z8)da9
z+XQ{lqN9-F_&!woU|}SWhfUjnwzPH-MBg_JFR72a9AHw(da;~wLP-Y*x~LGH<o{(`
zSnpLLtgI)GxHz(Ea0)$0G0si!I7Bo?1<>dQw?#>yPy+BCUm+ZL4HJ|-8ya~1m+8)p
zccig$7lWRjVt;q<aulC^#Ev;cQhY4XW`9AbaV;$T=efi-H_nM$uJMrlVe$^BNg?O`
z=i|_-fEKJ8vx8`$MBfO;wM7<2&rEYdx&t62zX~v=y`qEuCda6v4n0jSDBf!Fr)BiJ
z$G&J3LXdpz*%zsDBLT&1doc<B8v9$SiZL$jpLOyJFn>VRYO*~cuqJpT0&2$Vh=3Ja
z?*U*n8LmqBzXk^yRSIT_gRB{MsbiT&NvSr{uysCqu6+{aoU-oLlerpdyR@;@@hi`8
zqh?~&h;Pc$<J}ZBgyeFvo1#)CQrUK~o64Qi4mzts5xp`A?t^C*aH5RFcW?t*r@-2t
z9g{~20)IhLHZQqO5QPAnKxDse8r`0<Mk#ccBbNd}unYsaij4)>8Q>ZG*F}NaeMoJ*
zmi8M*s&S!Tzth*9?{oZGeoF1_cz!Snbi+5V?QFIEb9HULzRWV%$2!Jy>Xt=|p<K5t
zVZ^?}u3YQ-22Owq!dyJ9h~5At@_CAAf@*IlQL}C+kcfX(ZiZ6X0ua>P*yi8C=Lpg@
zWvF?*iShNf^=qSMnI5l_1sqb?kqh37@Da1ZsCx)6UDQ#buke5w@df2OfpZJ(V^}s6
zgp6xitF0)%yOUTGhz&s*H}51)A)a>*!<X4nDO1(G#BrO2!%i-L<|iDK()bH;8n2lz
zzmY~&Ja~W02GI4nt-;J!eWcRr6)xUSHCYFH85v?CkCwu2go!<MSxv1^ijqP_Poq?q
z6qu<XwWA6d3A0GkudFN<1gk=y*GeAqKru?JMsYB`>lbs+t-Fzf#|Z2Z^oe#Xr<&0V
zB@IP+3NNvdZ1P9e#j-*Wj=rr`paLm*lvLd+N9=zCG272C;N|B!d~_q6bhd~cqZT*v
z=$=KXFornq!)Ab9gq*QP8ETfrn+Gsw94Lp)+5$y5rHEZl0`H%gMuyu1_^v%|QVf#y
zLfn}wAU_#IIKZ-RNOQPs4Ozk83)+^VY4l!lEun3zdKjGD!u(ZUf)*4lEg3YU)hQ!f
z3h;lx73);P^RW;&$7Ac2!|q~#{AdGW3{^4Yy7D+D0p-o~A?G<HfeHuDAI25*wAH;D
zSr&t}Q<tke&-%--bsJG{r&Np7Re>Z3H9>F&=6mbA9ne6aK9mZoe%ls;qMQ8Tc&%L<
zDnW7YpsOB>?XJJ?kKC|2(R^$-v1QZLwGDp=Z&B=;>3Wel-~{qWA;eB4N->XY-6-gq
z5HuZv1Y3@A@>k|<HL~Q5yJ+1gIB%)JtdYF(awi~FIAN&)^zEl6Tya|;T0q?7#e1%S
zHn8LpM->CLs=dY|kb4zue5~^kE0Eu%#Susc4_;I3cw{XozjI_QX73$YFFm}sZY_VR
z)b%JwQtW<|bJVNHom#=zonAfnMwHiMBlaB|xj->M&j&ge+_DP*G$_;cU`JB?AD`t3
zR^K_x6SHrxiMIqV0?vEJhw>I_9;mn!40b{|^53mAHq`Gst}w~u)bTMW=j}%5d2}TR
z-iq;MV80=<aUQ3KYi2r`z5bCw(7sT^;N)`<;G+y~$xg0kUXv92n7qfSW%=ih;BPOM
zmEC1A`F|A3FsKlNWUyB#(^l!^8tQk3GH}YK>k{@yoxaAU*Isk>Dz2|@r~d$^1;r(o
z(a$Xuma8rlmp<}5D1W_L$!;4t8oui(>IKYjUjRdBONk83ErD-?IShhH`j9lbdk*vZ
z^Zne5RV9~VC4l8}QRKhxG@rgse0n~dUJmarN>7e?Bb=OGzDxpNaQqe=i`Gm|2xgU-
zUcOKNeB@kOd`|di&XvN);ct(pIX)=iu|pVq@U0NX2d4#{!G9hl(^+NH8HHW)jHkf!
ziPmvGaLWk-;0b&uPRD<}{NungvF|Q{U!&AI%>cVlssU94k^wdi?1nxnzXLi5NYD3q
zpGo_)PJ80Vq&*r><B!BlJnkc5?=D_kCCC_Nbraot0s>Zf5}golC9)DUr*j~3UB6OG
z=A;;ie7cYxgnuJ)!_lb6439{)9a%E1LB~-;ELc(!J)VcR4=;zm4plNzF=NAIykUmx
z>GSskeW3+ta9B3X1u^~opXtlt;|Mid!T5|O9QD90c$w)zT4s!&tklfnd3VEz@}J8b
zXOq=dA~aZbFb1KR4WST0NC>gcr}QODQsY?t_4n6lSbzES+#p-cymljG@t!_2&*AtU
z$2q}}CPAI8e5&5H6rxj189QQ8#VV>b48FRdk^ogmzmUd<6a{pF>N}gZ$z|r!!onrH
z5=EBt&C-+BvS8FgdUHPNZYP-J3%b<~(@{@(jrQcoxr{H#l{+&sU>iSRbKRUc+*enV
zp27S9OMl|b1T_G?w;|cU1eXJnAGRaT6rjw3nGc!uH&9s&bOxAo7X93Idebc?Dm&L_
z<mhP^C(>vp!hr9swq2}suFoygHcVT2E5|!5zgc>+?$so<Zsj?0n;@9;HG}f=TDqwM
z-Y8^B&Y?snY&AS{)cUbyf=ChEkV+UBVNQa@X@8MSb9rMz7Q`i;Y2QRAwD}{Gs3O0*
z@v2>gGYboKb%rleP@r8h?nI}h)GT=ZUS;SJqfn=ay+ByWTMx^`b+w{V7B&*(D8rSo
zv!w;)HH=h@5VB1!bfp}#z8ND)F<h^?6as=!<H;K+MBU1~Hwy@)Bgj$;QQ(Q|lsLo~
zw|}7I4IyN+gj1Gjv%myjdXP%>)->lK>%UteE6~}V&vP3mON_fkbmV*@QYx?8jBre^
zXyj*><y3vG)mBL;mebA_S5pbDvLeVKZ;NW%AuUoNGF+b5&NYB-Db%74mU1Do>?Ek9
zM?cZlGbcRN(SJ>dgwIn9hlr3Wz?A<t#eZcc1m#N391orz-b+mx<nZp#uU~%q{BOZ<
z>c6xTwKx)_z_#z{74GW2M}P1wzk3F=b9NFez-B1}Gjw~po-6O4r-Mf8sFjp$(>GdF
zAb^{;Ggkjz6W;}Rw^Z;)cxAO9`_?ZOiW-K<PbGrh2ueOnuQ*>$(JOuW_p%kIet(}1
zU#7#yEKp;tci7^Mrn<rhE##U&fg1I*=!IIAlt<ckg$Hb~(C6ynjM=;<!L!v6gpe!*
zhu*28s{EFfj_0xjPz6F+D$j9!Lk0t*SZQ$-wInQ)g=Cl_3pNS2{r&lSKK2!dp1WrP
z^B@%JML>Emq+w+VR&4g5*wqA(wttu1-Mk`(7@pH)YMfKjNf=83(fii`45&X&9chup
zhsNW_y<&lK9CE|!Vi~KnCJw|B<YH09v60R4c4Td)bsdVRF(VlJ*g{mv7b$qsx2~&H
zVm7$cRdbXzckrAq&Z%}vP=`0YfDp={2y~s+;HLc3Dy!5NRjOsI3i8`HU4P*fPYcd<
z<y_t>_k`ixpP;<A&S^BK(CWE$ZY!+NQbVpWGqj37<e?c2R8(DkMGj--RP>##qv-qT
zX#C@>p6ucF&cVC(+tuB<6CIlx1O+4!YCI_XDHI{GAz7?`dfu!Lq(nD_BsEYx)(864
zbUU<J8?_QT3E0H~Tjr}&Wq(D=R5}d((n@&fTyZ5lG+$5Q;g4y)E9dZsE{R`r=#ugc
zCGgh9WOHZ=uFXkSZfW7~pfX`4Gfs>&RM-3&1GRnHrG__{kuZL_pS3#6Bq0AZ);ESg
zyVzGgFsIx|oG6XTM#d}_r6j&oHDeOqQ>$HL*1R-T+&b%38qQ+fIDhMf#;ClwKI=^r
zoA}G*na6D9Q=b2B%yRM+K?~?wwfS$(6p^XipGz^!xeA8(S3_J@`r(<eOsTNIa&F>_
zR$*%4@#=8PTZVda07(5x1*RIX9Wr`NK+1cj<PGxKq&&-F4)7kcZZG1M${euds}rp<
z9Gq`nogtsD$n}PK8-MyW6w3p`t$CY9X)WDI-HyOP9i;ehKdoB}KJKda0peX+{ge~7
zn(Upe{>>`IQ5O7yr45}bj>a8L)396Fu5$KehLR}n#sN^6n%b{O*KDPCgrYn&##+~)
zt;0O=4|#Jdyh5CcI$?!=*2XD0ml^-sH`hvo675A`(uF=hMt`*z_1cx3e*Ick2E~-E
z5cR36gR|z1@44S~H!GcH!KtfvvwJaZ6l|y%j1~i?Uv$Ar5y>Hk3`ti?bgcUB7pyS-
z=q$TcwB@|F=Y<^V2MSiMuG$IDyjSCiTW=o`hjWmLJ9C><;^&X#5D$8s*ulwLuD2wO
z0~QLC=Q`h=BY%3G{4LW^JfJmNV&XSrX!B$b4(&ofdOae;)TLdB^G!EMl46I{J8zJ#
zMj=CEHa9{n-9Wxqx*lEVe;O7v>2y;<|4K97CZ(12+ob$$bKJV3>t8b`p2N%cNtZ4^
zx9i+;{@lOKW-4>Xc|bSQb>_+Gj%gZkdgYc%!{*dNs(;%Ky4g?#gFv~IuI}zeN?y<R
zI_0jUNuK1A-^$%e4lL%JsuazAo?|sl<sIdA(9WRtQtK?v?Xb9P!vdW7E}O2`4VzY{
z9h2OQP&RJygmWyY-J^>8jVVlE9Ck3^A1ruJV=WxT963U|<6J>_8sO^@L)68o&7z`>
zJmoT4&41I=-WiDsC<(#}Fa14nn*QBwYGMVFnqP3Y>XErv=qpCC(lc*(EV6^+3pv*M
z>Hd{2{nHG{y<?>-l#X`0RYqMpJLuabGnbflyDDHoHDPc!Rlv7WuwjWD6*1{*M!_$t
zjN#E*xnEQmdkvS<Z&zrpm%|U#$En(#6LO_IrXh@{f*cN9t&VdFscW-@nD5tpEK9^|
z71;QM2LBKL(E@#!(a$XumtmMD6PJ&mEh&H1SW9o*HW0r1SLhFvA%`=30LB7|lMT>Y
z4d05Mil(O=n*V?3M=iN4jii<3)jI7#E1Qz$@y&Z|_jhNzuMeMI9zGmBchMwqk-C>}
zox=l*UuSU`q>Ij(;9d9fz5D&qS_-&6;bLpO#})B+!d(u(c$e^o6I^)e+~JqVu6%!k
zcRZcG_2Kl5$6NG-yTkosImh?e23shAo_i|S;g6SJA8Zx&EE#ztA$K9N0j(w(Bw)s!
z@wI@1l)D^ynZ)&!J{*&q)m6wi(gb@Vaw5d35?BB$LL_2MfuRKF>#UPC88B<!`8vOp
z46X@37={786fj*ZC)~`!N1Ve5^09wHbT$}EeNs+g4vr_|F%>vzT?c~pS>y)%C=;A*
zf}cG$JqoQ}yYv=)&QC(c#|_`T&=hXco+R(>mxdI2Zcj~~lY1JHl&G_YaV0Ol`vHfY
zo<4tFM39+H&JzTuI^s^V8V&<x31!X!T)gJuPQft=FA0WnI)_)NSfq<@RSADXk!Ck6
zoMdB*lw<xhg9Xiau9qrjeibG})(8j94W74`K15=~C(U7J+ft0+a#uYPA;(ShkZ?)4
zz5M&?1)&UkxuafQl3|)4SA41WqvJrobQc|&5piy!qiaP=(rlcxIQ7YhT;ingZgl`a
zi67_?Drz|tk;?6>`tQrj!#{tMR3IY8Ate9&{!kvU6BA-aYhWTegCBo&-yVL`RIAR`
zSbTE&5qq-E=;XO}A=yGn8v^RcNXgJE7aX8LfoR70Tod)NCTS>3PJ!K<drA=?d!IGM
z9x57|y+XxB&i2!bKRS6H^d-sl9h0I=cD{?r*7(T!Z9kC%2Qa5C{m_4xD$_coYj{d#
z<2i1?)57E+?|kP9qn_*N<ar%)#g%+%bgHY$g>zq&kLRXgiX59j6CDalGKU_Th+s<}
zDxWsnlzcR3?k1AEiCs5Q<!myNwW50JnsNDAy(y3hd4l5Jo)HP@Fz*;?n&P=h-I`3R
zjr(Y*`XaSoetLaJC&Pc}Z2ARf_*##8G3b0g!>g#~2wn*(lpx5KD(kr?*E*e+ULhKc
zKCGx`#ie$wNkQHladBwFSB99DAn!TEOW?KI-GZ?<$*^l*kiT-{lK9X=wcw~OQ&L$O
z5sORTEWtw8W6?KTRuX(Xp64IR2l0;sCu)43D#q?yXE#?xW>kMgT@#5+cDzy()xEEO
z3o%r^mc^l3#M=#8W9Xc95AJa}@oq_JluSSwQ-ZqB<(f}tHMOcQsLIh)>LY+Jmh)E&
zM(h7iYUM$um@~xUc-<@vqh{J~!7!%xhW)~rKwU2^j>s#uwT-}|r9m1k4W@pjX%7N*
z@XZOL^9c>5tCWA*ARR{E%7NV5AodBZe$W_m9?z*OI5YAIqXQokSs0LgE=!o7HXo1K
z>oII&f?XXiy;Tg)S#$q-SWL;hV>$GKjxj8DAU$PYN2T>~Zmk@8kB&b2P33T_1lP;q
zSb}#dhm<pNRD7p`=&hJ!`F4}Gg1CL$LpY@5wItUn9vpwfBdw?}s5QP$Z63pGt6g6P
z+~bYaj-nmT6}Oi=Meq&vPEpvfPuIL(w@=snuI$sN9=78?T~LMm4fqj8M7B;-xVBH<
zqR;<epPrC9CN8}(&8`JbnX+t{c+En)KnRLp<4@}sH2)bacHE~cZg<$HD-Ju`w(ZmJ
zs7KwWYdwE-Be$<!ZJ&NWz1luqak#cm-&x1DPgfGWa-U9~(Z=0%pYAa7iv5}UbaHUM
zxaU4yQQUE#t|+hU)79JWv`;VGgM@e9rz@^E?$Z^uyYAB!FL&H!pRTFCZlA98-q|xk
zLR6<6BTZAhb@aJypY9}@nEq4ybcdPK)1LeETy!gSi-^x`S!-!n+Z9C29mUEURt49X
zEloGh5GvY;*&DY4YEnme`xk*Cz%7^2&n*-IH<#gSCli<c0XHa@A<!+Gf3MR&Z+{$g
zR#vNE<HvZbzkPZ;{C5~s2UKiBXW1#EP51BjL;QgZE9*iBpn{?9$CvKw;m=xXS*yAy
zpu7>Fw6)ALpuj3--&u<aohv|*fJ6cVi@_wubLhX*Sw0J3<on;Wj1w$0%NTO~b2|5i
zE0QIKKR>>8)sw$Ik2VKVe-;`o%*S}mU@S90qcZ5~8POR@H<Rz89EqbSI<C=Eq<=a+
zSsmS+MrNxx|H^K+CE&dRup2DVN}Q~fjxvCIsk0{{^}{*nC9grS%1Pwc*1z_a7rlp-
zuIs-RA81fHhGr$bdxb%wL@-#>AViCP#u%T5>(Tl(duA#8S4p8{e-=0e28%37exhUc
zzU#fud`fh{CasgvX(h7QVj77_3YSdD5>|9_f-gWVA;oMEQbC>NrHu6PNs3mRWBHtO
zv)4I~AG5cNCmLkom=!T|`>bA+oAF2^Az@MEzm{)(=A6JB7qjOHAq?Y7=ADd<vtBMD
zi`J^Z*dV)}rE3Ude{jN>4Zt)Zwbx!y`po&WpMuXY7)uiZj4qbBI>Cp>4AbDOJmNV6
zU}p}Tg0?g0N*GH;0HN+;(UhX?^}r*+Lvt;~^lR5^v4DueNhM`$rqF1QamV?}f>xiL
zBG*GC0J%JhHxn^5Pe=M$NU)%Ez<T|*9L>mM)n+fYij+!Pe~N_^o_Et5G$;zt=)D-e
zeTk)<dh8cTe9a~hguwWErR+zoL>zk27}<K8N{oiyp2c4-`J%*v3kt=yV$)P_?(Jy&
zk;>R=m&S~WJWR<i`2$KMk0_<p+=Rm7xaZ6HQdU7$S}ZLk>|?KUYzty9Xv_^Ouf$tm
zMfJkZXIqj6f5M3qZIq71){HxG<j(eyBp$ppB0r~nWKGhn-r&KdR&syym7Bs!n|n83
zaZf?EUM0p++7rZk9skSd0eEFC)skX8u^NtZ>~3m#$Vw4Jzp>>RIQ>fV6BL1KWWOg*
z1anD{N>Z$mP_)Fh(V&4TXDemxXH8T8MCUm=*ZJHuf0f23jgE3~DkIhY9+{a$Kl9J!
zW`?yg(u36Y)w5R`7-@xk>q9Omw3g!>{nUu{`nLJCrZb@xpEQdu30PK}n3eI3+7B#D
z;#0|AzGGn`AZb6ip|!k&ywbxCQnYYy2%8E|EqNtj*eQ!xN&ZQd1+7;k4*EHj1sy48
zF1Vnte;mm`-hc35CETRHtX8>j@L+66TwTooI|4haa5T-s{^Tfi&SApnJ1ShR7~Ay)
zbV@dv^^6pJ@l@*_I#7^=wv7&Ed3~1J7TyaNEa%1hiXftwkTUy<U`zV-Zr1g?*)ksT
z&V8k1&?d{9`80hx$l$#Cv{2ey0ASRTZB@`Ne-<l22{qPeO%iOiP;qRX+L^DH@tlNW
zv9I6}8T-=-`8YY%a}TW{qKLA_@2DZW`4fUsuI<c?O<8r3KR56t1TRGpukj^^{andb
zA%=Q|i(E;{ldET3JfzJ#`{CK_Oy+j}Bmvu-9X4eCswUh*pZR2~R$+xcol&40w_bzk
zf3Ca<)+oPs^Kft8<ck*nQoa-qOX5~2<pf*hN-4{h8*G0ot^`6AcJ<r)xl+hcOIP_I
zt^`T0aFr_Y{Zu(yU<*^uCdAi^LtLIG8u~F8;tpcS3z}?l-AJyt5am`23Qx`rxC2ej
z6}ZBZJK@`{73!cU_-$I1cW0GW8yt35fBKH6-V*V@n7*(3qKv$X&!AC$K|Q>*vSVo|
zqiJlV%X^rl+m)A3LTnyt<T4SgpVN5dbc6Q;3D{WK+XZ|i0h_cNFb^VNL!{XGtDF02
z3hdw9SF3E;+*dQU@GxY3hr>5N2Iu4rsyV)oj(c|9ZV6aOCquO*HxRO8MYh3aRh5y0
z?DyS%B==LYiJh_Cs8{*#gKG3P@7<f2wO1s~&2K`?qwMiGe%3|X>VE}>7sC%CX=#}}
z)+lZ`J&k8k(tp=8A4t?eg4UZoJYC&|oBs#GH{h2sd`1%iHka{+DioKl)-AvRDz|Uf
zErSFEoi|m>m(k}e69G4ufr2j-lc-=Qf5lnNj^j2AzUL|Y0;7r)sUKhwU}knRKyMw)
zt>~%f?r9I*_dk@9<v%)hoQ#tm96O3ciXSPGpJSUJjcwizFCT~BPL7&jB16FD<9h@0
zgOyLPGU&Mp0LD9LK7KZTez6w4TrYBwwa&>E<T=Wnk53Mw%m9{4eg!x_1%kXrf95!B
z*E-*>aWV^E<qmRx&f~H^S#Pt1B$vyQj(>gpez0}3)+C=F^G$#I_Hp?4&~F``VV@e$
z!GMM4@1KYK0|6T!QiEs`Q1k1b=KJ9fEw$67?ggj_0+jWhc>)xA!yFs$opB*_K#>^7
zVk}{O7DHJKCjpC|+dB&gR>0)ce;R5?MD`!Qel%+2{cW%@1mgj;R`{B)32gm}=nbG=
zfZbML*izmuWj~1@h(=@bO!@iL&dok7=A>hGvwf{roC9<*%sO9TFbJ;OmV^`0m%1fE
zjmy@KCBPVpp6ZXeM8;z0RA3Rptssh3z`~Y-YzawAC<vr=)iGPmGu>%je_xD7u5<c0
zp9%>1n>=NZ8EYu{!KUJu*TIJrx>aQcb|<s=F81FRgYuZNCTL-vIm#x3vG<d&i(|i0
zY6GNf5Nkvr$A~$G<%*778{4zd_qP?VCS#LR>T%^@jiPgW9G1L*)*F|7H&I|uyIFve
zNF95nfc+5U>#1U!75Q}Ye?6`6(Kwn}5r?SU(He)zgrs8*y7MndcO|2661-_hbXo07
zO=toMyCIRuF*}pFE*BV*R1<yB1*TL4^cO0Ou{o=k53b36GV)%lPO@uNhs7VkVV^X<
z))aRdQZ|ADd!df88~~TJ#DZ0JW#_`v^|dNOX(g0owCC$ovQnT(f6sQaeYCaHik0xK
z?&&h57x5%-qI3t_@wDP|5if}wvZA7OGV;Cw^Ql&JxoE5v-v#8zc%0c#I*fDZosLi5
zmZEY89Fa^+Vb8c!60hm6UPj%C<hS9Y*ftp%$8Bc|IFnqTe<)3>obtvV=utvr_dL{7
zf!ovc8T@vzF3927f7d)E@o}+1(w6|OwV4a_r5}~45?eO+2GQr-k;ITz_qkuzdZex_
zWriblXyonEHg@g(JZ7uM7D=E>8T@z--%>|uXhbuteQXV-esRy8h7cud4z;1iM9%t|
zW&+%(b)n^!8WYedt<R^v-NtleH)?FCw%cV7wFR$z$kknge<Ls?4COJjR9Yb8W4g_C
z-4VH=<QKB3l&`VVTD#PWedy`vgl{(i0~_*NY{)0G%YewW0~n92<V1{8TLR4N<{D=>
zbX==><PyJj!ag#o-Hl%>fj97LCE*Hwt#!X_;@4;N-IZTc_p~8dtuby)a!>fbQEsh&
zL)N&KdrJz8e=xIjPuhwy2S97kjnI~iSmupez<8#dF7R29#301*es;QCOd!teyt7>~
z6JFL@Hxl)nXg#NU5_J$*1b5BTDH@XA?GZd3*%>iu$DUV7+>xg%$rpIKn*AW2jt-;>
z`3LiKrS+3MT}j)()7=<9u8<A*bt7y2h6_dt@C?7#f4=Ttia(?zY}jiFZc{UPeHSLg
zm1a<>e};z1-ULZe8rauQQM_C`z=t81?oRh1MI*58m>jgkTe^PtDbyeagJ}~l-T^?*
zl8f-c*+6OiWHwOJw!XEy-9b!3Lv<6qt5=+y-Q@QSYbEg}GW$8yP)LLeo8JXK>YY!V
z1kZ+73O@QSebay9Zh2al(a$Xu0x>n0AwM}J0x&a|?g$t$f7M!TZydJ~{_bBvpkLAt
z;&M1Md?^A0PF$xc6u_`upsgVkl6*>($~xhE3Y`A>KEoBIvvp^mbt_nbVVk!*OAhDt
znITi6;DSqu$VtTiL0l31AIZgNFQPb($taTJWUnZ@6tWG9%hEosu4Fc1c0`mOVH7Y@
zBL@Y=YGjffe{r?BFd<29)8m;ficxWtx!9;WD(x)?7cq^Ej32SL99`r>ScBO`CF2^b
z$E;){Jj-ndA}*H3xhO`dI3@d{QsX6Iq|r%HtWm3d!PRkK`yvI$lq1$i8QafoT?j&g
zy=8T*ZR2b*X**}omGNzHTrpKjm{~$LuE3gbduvO|fB2T2q-?T+n4~nWRGm~YUNkuk
zZC%YyAp{#?3*6qO=v1@KlGtg}2)2ZgwrA=z2eQ2pM#8oa%$EyZsLfKMeF0)MxnPr~
z(x!!i%~0dTUQm-{!lX=%4lhJa;%!^SR?GNVW55~^L>_gX#X*(rt-74XZ>xi1OqV5K
z)b>`$f92ZDtVE1#TNNCX5KN{yVbu0p)fGzlQL=4;d3y`i3a2(7-b!ur;R`bwj518J
znQJMohH%AhCAKeMjx{EeK_you*cUars%$#1?5YVCqwH!O-(r>u)k*=+wU+h_b{ax#
zMigym>t&ErAShc3Cc;qKI$}d`1<4sLM`$HOe{mE9_KwetR&C!o7_C_phKgZMjQ~Cq
zYKS(>=4RZU!fPnz_7v<;v(QYV_XQhbGRj#Wef#b3^ffG7;qNbofBx&=HnT5~V3h1L
z4DS5$?Cjw2<4;jCdnv1wZuowFHebBFI6R&?wh!F!{BXIPEzVsWfByCT@`snp!{yA;
zfB4|p{CtTup0Vu{YP40JY1>F0e?Z0*g~uOI7Af|LACDgo&lmIKm$T*T;rY{N!>if*
z<rT-Te!rMu<)3F@=zKXlU%s<Q(pVT?%-+o}7ss>qD6~gE&EA|Iem8&rx`jV7>x&2Y
zet|C?JZ)V2$@%$woJhvyAH8mY(6-S6f8o`j^g-kQ;F_QIy&JxpFW$@+ZK-fD{B8Jg
z_#Qv7#{;l-ynIbDdd)2!c&JvRuZV_{#m7JgkA|0*M@yS-|MdL7!;|aDkC&(O^Wo+2
zj~74M-*0|dE-&8wWf*=wUH)=;<d5fXhhHz>UhKV@{Wkpl@~7wi*R$d1Y<@JHf1H$5
zk56JLM?@#YZ-h^TXL**7-sCqDPr`}RlS;$8#qlt|Sf0K;{oicypHs+naRNnSqCf5D
z6!!LWr9VR|D=KJP{($54`={|o+kfQf84-Xrd4+?*HNC)@I3-WzV0hv%`^({n`K$Tx
z)O|w<A_@`ltOys?nsWsgJ2Jr(e_`WD0bG27n+a?^Sj;aT4zh*`DHgRuP(^~MF;ZN^
z#P-TB#gpaFEA3bDu)2q;{1WosevuXBTf~U85-}2_nOw%Rdx#b*G2x?Vu@XnVG+GGV
ziWXe&5-lX(M2py?MeNZc_GmG3Il`Q)0rf%b16~tRzSO;9@&U<Da`p`%f0YV1GtdDy
zQ@(B5z!3@@gWMYdlo+NScK0uME$ketBHLIdMbh#BV@?!7Z=Hhy7gD5BVn9Zw8sK8P
za{-u$Rj--g8hkcUY+wSaW#B~S%mJ|&P)2uha=<80z)s!EIl|2|KU6n|TkY+WP7&it
zix0q41O<d*SigASC?I|De^u6m*8&|Cm~F6NPXb4pzy=Fp(HkbBK>esMHZg&^1qKi_
zRk#4$qE5j?Ou<)VyUqk`+p_)!7IL->13N{*An+kBc_#}rJropS-2y48=bFE^1s>1O
z-z^u1Fu!-B<%L*bdML*}OfeF!Qw-eb!|d`gK}W495TJG5hBA}Se*lf2r-V^>IPVse
zh;csXHk2@OAdb2XC4)~?P=A!nnr>%Sy;EjQE18w0=Sb~?o@=@1TJE`4dy3VbVzsAO
z<v!>WYM)U1gxV+6vU@%j;THPL+Q{;9MKZE#8PO>St;vJ|)ilb12($}#DhLP@#KNn9
zjQxA-kVK^C)&gRWe_VH)3E>nfp)?09KsJVF$^@|x`A>nnIul3=sKGZgaW9gHkb7Ho
zWP(+^Fl}Zc0U06k;7<lVLz<EaB*8!+@U$x)L?9RX1P_P+1vPutfJn*s6?aPu3Psd#
zlNJ;aM#5K53%?yMP7jaHW@A!F2{|E^G9et{ia6d{ilHF`e?eSFD!wJ~jB=Aun~J~*
zGy#$92n2J*x;Nz>?AaHVhb3=d1KzWL>jV-v$PMoui~Cv~EZ1AdETSsmP}hL3U|lh`
z#$VQ&09xvFQ7;NW7jjS!dr_-Tv@Z5qcL5L$Wl%tnRyf;qVeguYtGKt3i+hnpk9(bq
z9{4uPB36+Cf7pX!7M_6kFj?H|JoLc#34Nd*K|zM<O$sV)*|F_{Fz%z3038oURD$%t
z0$-81w_nmn?Uoci({8y<HNr=w=wa7eP#QF``uOvd)~ic#D@sTjh0GR|K&4(o+JdqQ
zkQ3hr9^M3jcj`;$atjaVbQe54{vPpg>R31RzEbX>e>Zo~n>*;u9rWf7db941Y<qN7
z=tHs<wTyZSbyW~lOKW1hAe7|{5-QOc4?IDq3CS55CS7x31$QLb%*E6-wj!LzauZV&
zwS6w>fhG4$mfUj;>yrBy99oDq6w1x9g<3<Qe(lX(EKbjt<3UA4@XQS#KiZ{pSCN=j
z8h5DGe~BJi+k~;z)t1R!b{Zx6yrNJ`&GG9i^h4{FlK1vy{pr&eB>!Zc@s(y+#`E&k
zlmGeIMcX96=^p;k-8NZJY2=lIB@`N29?BL`+`?kRM_s<0Khxz~A-D!LNdYADHk3*L
zAlpz1!#;Ey!pL6Zz2lD^AV*ENAV-xuKn}Zse{l^tN(T?6cg#xfn3di!D;<`U4ohl>
zCAGtn+F?oM2|<XRreixXL~qSPMWLzjb}vO^i7`|`hOUrw33(C>qTUYgu9-k>Rg9@k
zOe9n^AQBWm%0(x^K`5=P)!%tw0z{|Hf*|35I@69S1vm)5Xg3@Lcl1?m!mAZ`A`h&#
ze<DtXgP`{*njEa-%>@+rb|Fnl@OJr#S$qm>j#@*sLnV?TwE~~Q6lJ<52U}H0cE(n`
zUE68~u;!2x5DrCkEv=gg)GQMmw4zQQ*Sz8(LNc|TcCF>%egqM@V9^^vL|{`ZJvT5>
zIoRT0BTK5eWS*kC3}x8e$<Bh6L<of&e^`*B5+Elc3N=>PknGsrh(2mrX>Szkvpg^t
zM4++5J_#SffU4UxII3=Y;<t`+{fXZiN{I@mD%(&}@Fi|T2_1XcW<vmNB!86SJ`|Q8
zV~tF77py@y$9;A|vS9?f9(gs;K1h9#`=Io}73j$Ap7TV=BEcBN8mG_>Bb+4oe=@43
zb!}U?aUiy9yU6g8(-Z*(=4H2(I}=Gf>t-g@Zm{I6KqQE0kasOp++o1Z^_mCffRa%+
z3nE8kzk<jbQEnj}sfOK<MI=-)H|03;g%vpuEUd+t_UsPLlm}MPZY<f|cCvh87ZHNj
zs%ubfLiR%eLqQAT462qXOs`^5QMqP<iM(C1-Nc0L9Y6q4tR$pV1=Q9lZfe@yNV?&o
z7@nogT;y7SWf7Jui6XzQzt8ps6}9-7ZrXdsMtW%7J!Q44Tj|4;YGuIs%KHD53-_0y
z%oGy>HaC}%|1J~(G?xLR7$=v&>@AuCU=NpE?JXdG<dWp4lZRD_%*xW>`>RNbzUtn0
zfGf2utgr*+ff_B(r??W}qP}oov+s3s)Fpb@Wu4fM$ttHcQo+V$insUX*W1VLzg<;|
zMzIZ@1+R>F|Id$Id_W*JxX>Z0ASV6%xBIsHvz40T(B4xLrG?a1e3`h2Rx$g|A}K5P
zt3o4xt1Ag6`H`HAzsbkv*-;YF_0P}mT{H6aeel&=1=uQw*LY1bt4;{=CeF-_`AH=P
z@y^*)+vmd6Oc^*GzE~YIKSeA@Dmlyft+3e{^hN~-8@$1QissIYOyZnS$t=oofqu+X
zQ|0&DW^)ar5|?M@o)*m&WRP5$IT(S4_>8T8`S-a1$Rn+f=L(}j$wEoy>R|5`1ygJg
za3k91;JaLqd`NJ|unr}%4w8e7z~V?0ME(Lwh)GDTgp_|_UA2#1^K9mo32-J~muy4|
z3Bg(~`GBw=(x}u9SA<KtvHT(`J>(3M<|o%#U*A4n43M#cd@x(31^8K70po*0o~=NC
zMP~`L@sfpxt+#TwL1SM)XP<n1Uyf=?Q73b4HMcBFfhSuNtHSlx7QrN)U+ztd0IPzr
z>n-AnmcZmxApSiqp;3NCOK4`?-V%y)bw<Xw<gEX6n=qOclG)&)0izt_CJ!(8Fzy|n
z!(kzx&chvXcq;G~9G*(pg2RpO)kR!?$6Ex2&d(%10B<i?xdCS}2;rQs!bY5BqclP*
z<$*XWXs$f@M*=NbQbDl%F+fYQ%|K&*570KsuK?O+#wC8eokydWu0|UyL*tZG|I9KN
zar@K2K9ZHWVlI>I1MKH~`%{=nh9w!o{V=l;cmrlO5;kCFJJYV~6ZR$7%;HCX;Nr#T
zKaR1IkxH4Z>Q*GRPN8juJQ7Gj%E)YCtd|aaOpt=aIfaZ8DJN(N9%c>Hgzh+u*9BWY
z6fzE#$8go5&`p*kNVpI^1d;&w#+YApc4Rb4)UoYX%*qA0``M9^ZHOm(zbMBZmmIH}
zq@dVIWQ|td4+n^p;bJ=r2e!(8n9xdGYk$H4Jh^tp8Vkp2Uw;J4v2&XYZEt<^(WjoR
zrxTIWd9NX;fCIt7oG33v<96?hFH^~D-b<qpJZWd#&-<z8C9|}b?5ve6T~}xY#x-~z
z``;7Yyn8z@`L<4tv3ND#sZ)DDleno>$3B3qgsWP$k<Eyi&74-xB9Li+tsI1NdSiPQ
zY}ucy_4vdNjeE^LB&&#Lo-O40qMa#27rZ=F5|JepyWK!Ol0MU4l`K4$H~F>B9o|m<
z(|*9SNI!bN$qh(2p*PQ!e)Lqvoz+(Tt|Qpbsj_=p0`BvNoFpywxG`;sSjxv53qK%p
zIyLqsQfVzh)iSNcTnkozaE~S=ll(xK1|J^EQIO*I4|hk7g2H+PCThN9wvfo0$sH@K
zwdt!k2?W{HtbeDvRq+B`1eT(j>itFR11v~ff6<345SDDw-5$QYp?>+~rVb1!5_ldr
z#$(=}B;CW_rTp@vf-MDa3JAIW>uWvhF9{xok1NCJPTEy)?|mwN?&Re<=yFgfWU#TD
zU)DiZh=T^-&Wii}yiXfJ-MB@%{df(=@YZ`Ut%$nU8n=~Wv)l8->f}tF#RTH+cGzea
z==0-PB#6u~Qpk@&_8uc`lFP@gl0a)Vx&~Mlg}gy)L)`-&udeV*8Hf-gX<JPX1Yv;|
z4qGs|y$fvo4}k*#s+Yn21rq}{H#C>Q5&{#KkNiI=m&EcdnE`Q^Uh^$7f5#^ez8t;T
z9hq5KnlLIOi-ClZF}Hd%Mi>_kZ7&*_ptSEDdOAjMjf<{@k@2uUtSnrDT?HGFkGHq(
z*X#SkKMxHBK*cr;)_8>i)L%az;tMk3g9`(I3WkPXe-Gaue($Z8rS0!2E)(Y3%3Kzh
zgH_Caut3Vn{Zm(4IzN}mf0GQ?{`~qebWgs$wfTcY0S4$jGGF7hh+wS*CarX|0i3Na
zo_uRjzM3}(oS`2B*ReWf%Qk+1aXnV}@oTmwFlqfVK3bhZPBN38GcwF+40!{K_FKWy
zd^o`^C1TinMFCshJeBncWkgfMmZ<)U3={JzJ#pE)$So0p_&ah+e_Y%2u6iAE`xpap
z2H8rC0^%Gwv1N{s_V#iX02dU3>t};*KUKZ36;t0!AM2c%M~G{XX5mUC^O)Z?WGV*J
zyq4@!pC&7m-!zOPCdZ0(wpEgUwNg&Yy-m#)$ozbPqA9i+*c$i})ZRl(8M!<%=Ad6a
z{q}a6WGyImrgI_Xe>vp2+`#PTA{9uo1)&?fF{F%$Ugm&oy-mriA<uC0&E704&<x7L
zl$HfguAyu<T;DH?EQIOIR2Gl3<cU_FX8=MTC4uX^whhdSN+N412C!ceZ}KFUM02Jb
zk}QDloVimuBxi(u?^O=TNfAQ-d8-^tVQiCej$Ml>I$mEPf2G!t4!x5b6|%2VU5}_b
zBWH72r<4*jtsCmz*Hq0uUZIGLGH&AZr;+$Djmx;73?!-NR#sSVB6`vnhsyYsm9$i`
z4x|VsivQ8Na+UX)NZPMjHK^3Q)H7Lh&7z&v-Ta;x7IC9S6}1|3%jnvs`Z|RprBR?(
zcDsurk{5wQe=8Jcp`9uhY=!+UvKYiNUjl5JuqnCWG7P3blAguEUt6DJDNL^;@?j2I
zlqZ!7TPrGM-@>70=VD2kFL5K4MzYz0(VNAZASf_?9&YmzpJmY)%;99YC)xCY>?0;A
zV{AC&2m)hzbz-Jij;@y$Gi3q>Hsd}|91s_lTKkm+f3j0gI<|IIH>URcTFrXpMXPrc
z{5-m>GaM^8GYUExh8`3W^-fLD)R>roKn;Yw=|+nTW;NaFx-B$Or}UDiaXB1$ninB5
zhvO_HH4rg~wq9pulfH@4vcPS=mB6%feY%#(r!iP;5W;*zu(Ys2<8xn%AIT@(baEv^
zTXNqme^vCt!X={Q^A)IU;blAPK2d=pvm(D#paqA78x_m2+_oir?L}%6QKV5guQsn8
z^+a$6tsG!C>YZdINJe<Pm#jo>m4N)Qr7L7YWd|Ov){Q1m+6;AG?CC}=S_xg>Bb_Yx
z$?vL3;=(}lFP~&4JyCQ^JmrX|mXV6{i3o;ve@60gRS7nzK$EJ(vF-Sq1PF72NFn)%
zg!8kI{OhGV>!3=EMMe<KE&BSjD!fEs-r#dRA9?A{1M~<<@rK!I7txaR;IR-G)ry^h
z9%m;DXVe`?MJDbsX^~U*_<7Cy)o6uj)*`OZnd(A98^*bQ%vN3qd%{mBZbyG>j}t;k
ze?>RsQxQ33lQQ<%W?LC4thUVWgvbMSD)>p{;S@g^^{L3*p=d}FiIV0kipKTCCmP0B
z?1|4<70~Vz-vvYm#wn3tH}G6LU8AJ`uID<*1CN~}yHuBtotGW<m<|^n-i_^y2nuj~
zHQO=I*^b>MY{zzs?bv;cn}CfXu>A}Ye>XMmoT^8~dBvgI+_lT7SMeDJk!;5FeomW5
z+jCEPa+1$7UOk^RY!jbt*i5~;7@MzB#;1+$OB|f$IN5Y6g#QV}83_jTpQ(!th%R<(
zaaGk~Jy!^&$k}P0s{^q40n65B!;7VKyla1%l>U&T<>0pWb;Hq9NAi*HeCO!Bf4*^w
z_=bClZ)atD@gIc8XbxMbpW3irf85KYOz7O-t7H}=3;Pbcm2B^v_i{&NudSrOo%i&X
z54RXs({MKkhWS^=R>3a$f{P4FR<Se4uVwl-c*<5m=;6Rs-Mq|8=OknNf8eDQIWN2K
z4KJ_z5*|-~J#}{U@n#1}_MkfO7wry8Rx4xrg^qc;`_;X_3z7W`=iNrJm(k}e69G4u
z0ggBnm%{liz<*-e;0}L1{eB}evz0T2LL0vrOM+<?&X@xtz7un!c?BKr`Ve{2%nMo-
z-ctC6;oBo$%z-vxe0i#wPx9ecpjGK$9v{A&^;pbf={}69EBG`faACs;8Rhx(=wJ(R
zrM+;LX1QcQn<>f1!$%_no&iI7%(S+(r#uKSW~Qa@3V)B0D-{uOXSrgf3I#JE<<BmU
zO0)6Uc~*kE!c$eD9umxqG8@$ZUtp9?giHqbm?4E8c74n$YS2@jXU_^pXoZCp<ZO)R
zq$AP-FACQ)83hYF;8@u<#Zu1;60OrE0O&rlwb}<p3F%bTaj`>+)X#|z#4}+g5GF=r
z50}c=tAB*2&U(-~k2~h)HoXhR%CFMaum{NdF_3EZ+pJkV)0(Y`zG#+|Ef`Mj_^ge%
z^P;J>=UJFIPbS0@`eoGU8=@~u{Ass|W6=bW0P=N{1o}l2viBm%uL*C0YYVq8)347@
zH~+2_t)&)*?|=We2`>ovr}g~W2(8TW^PlD0&3_*=+3Y-}kUS@SXu16SzD!TP=4jw0
zODK)edX$jYVBw7dp;<}ARPx3xcm+un3|>3i-kPE|0@2SDoADIO$yia~vwTGMhbLdV
z_~0N}%Sc(t?P%g9Z)TcD<BE@D+>)h)&zYyokY2)5doK)_9e7ICIg<xmAS@JD9iDu>
zjDPvGstWe+vz_X;GU?G?SIChtY9+(y0=D+QzTd4=#f};*t~iI4UUyL?FFcx=D&VHP
zO7resf!7?F(#;=U+f1w##{2g+lX2BxCeMon+we)`#qCV>jJdqnNX8XFZ}USSNUivJ
ziE$P)n$~@MopExo3)E(+v@MM>!t0gn*MGJ{t%Y@PMfX2boC0siKT?{#Cl^V1hnjP$
ztrVp*xMPNiQEVR(rr}8+X$mYH_noSF3Ggbx2NcRB&mdPBESbE|qlzdhiK9QKDrr@F
zl;@pHU0pD}3Bra89wsB)At6fPKrI@PV=4|c31zkTmg&SyN)J5Al8L6QG-Sag-+xkF
z@|Y@;Op$WF>q6$m;~ty#bgmz3h6@RmkW&VfqE_>I;G-dcO)wM%zXk@SgqPIAfM)||
zDs-cY{ncT)%kNxAD~}%`iZGl;>ADaB2v7R|0TDj5_d#atGXIsVTWRy9jGye_z7R)#
z>VB48NosO^N3qhF99qPumx|`XH-FYO=c=X3o936~YWO}*c=eP6Uw4O(1d#h0+Se{Y
zdD`_i2-HRgrOggqk@V|Q5uYiwa=lRL>f^Tl=*#OG$MUOmc(!K*pImsU0cOl|dj#;j
z7Gi%(>f94Mt;0&wwYMYQiDPvct*z=mG$Wj^o|y`_rRP-kT;j!tRRZUvFn`zHap$>O
zA2%(hXsEQ~ikp)Yz8n@WuzEVwUsh<gyNNTz1M1tY{gEs7tv!-tC@5g5>D|~}D^FaO
z`M$ICPAAj;nNuhP;=OEUzKbPpwxH5_r4KkGSoq43opLW{%mucr>$Hp|b7{Sg+T*aT
z6H?Nl=<7}GU$D=15*_OdwSQqrxg_vXWxu}lrWFaNy^ONKWp-di_KK3ts(j2j=E;UL
zNhPH26T(RFuCpAK?7S{?xzSgMqeEeI>j?~v#}qSmuXw@>J1xAQSG;?yJt^f68gI|-
z_fBs+J4HyGU3;24MIn0uy3i<^?y)I!B8SiEFKQKP%iS8wxwwCEhks{W7#>!s?%d=V
zC#bu&4=bm~q={WQ%W+Ff)!ffVX#@sQnrEM4G><s?m=)^v&Dw5@FR6ZO>XBZMJ;e_y
z9Zu%X+_v+C=C5py&7E(o`B(3?A|I~z?X^p5d*5!dZ{O(5&ZsKygtjgA3~|nOP448d
zcE`GtenxvCId)9ztAC9>p*HqIv#QBup6H7!_S<CU)oSbH_3|jCInQ_V3bnK_d2jFY
znHS~qMsO5<L%C3bAY38Wvd%Y=3#}4J_?B{6D-c)6^?i;ipL-J-$tc0e<;=k|Pf*yQ
z2t$_-P+AhVH#C9|jj~l}v>+_PyPTo0J_YaoD>l(4idQ^CnSUr>I7696ztb6tLL)f8
zd*|l(wCXF*P$p_`dWJG{ew{OviE8=>$mz3`+}xh!zOrR-MF-AbbEi!d_vX6yuj(_0
zBNT-fAU@^0u3!3riAY_s_5%XJP!I58ZsnVeg1(eXzWojjA5yI|8k+{C4fnhv5x7pW
z=3<TbV#ey<B7Zh{JCPD0tDb$-z}q`jla@HkggIh{69M)nyNbyjMWyF}Y~R2VQZHPm
zA=`fT1qXoXC~@TZOg&nh<~ZSOJ-QJM%4twM<M~`7L+K{P(feg#FDDeM_t~-_Sz+BQ
zJlg#$vybJ>e+K15%0X<KRxd7`I<#1(ySl{f&w30yLw~N`Mxp@Q92gRg>@>Jomd=@V
z)mO<4ite)Y1^tDnSEIrXt5Z%GPL?Nb&sbef0LBTpV>{udQw^4QC<l;6mj6SHe>P0T
zcTQfa;bx!5-bD*<E${ou71&Y0Uw`|MXVCTxwnYQOj9)%HBRWoZc7y)`wpg>H3T19&
zb98cLVV7I=E(rrRH#e7o>OB*a@Ixq<Isz`8e~j19zkK}sBw5%h8K5vWT&&ZAXjRFG
ziYAJNR%?>T5Pyx4zt-^=#9ud_qw*Y0avD?a8chEnt}!I9eu@8imQ1qzbn|M>S^!d@
z*P~3dI8$=W@5WTtZ4TNAttKKxu}I{)q7<3b_$0b=&sl{u?z5uCl{Hb6q@-htGAl=7
ze-&CqUCT4*D&yVcm|~nn*$rHixn-6txlMra9xhU7`gk;W`I8(ukAD5KX;v*kpqgAA
zgQ#j}8|xJJ^MCW211-%(*54$Hw1Ktd4(+XwZgK~(ncmo^HZyC?6diWWV#K@Tw8(CX
zUkgH#DVkVGrpK(5dRvcvG{QTIG<3qze>JbZO0zy&8=<`$W~Yp;c@hosO4%n<4up~w
zEWc)LW9Iles|f^M>GWZjG+$OkcR=|UgBK+}H(`X8$8Mf^6v3ELA9>{*gG?p*DEh<7
zBxSk7K9R4{y7_|>fBOBN$K@?wt=!F843xFfPBPu9ww_bu+VIg05|!&qU2)}Lf2|`4
zNqmi`#;QpxU+uXOoeMJdu~^~*D5@yYSoX9kQt}F~nz_FAkOH>y3-w<kPaY|hB23s3
zoNSYpiv85s0_-CZx$sG|R2<P|<XT2Enm6-JhKJ}SaB;q|y@dyjbkSO`>RF&AN%+#}
z!SK3p?cueqZ=qH8@zKoM1n{)qf9~Ri40QF6;zc)m9WT6W{mW3C^4SdXyE6L_Zn)6m
zhL;65GNpsK5u#qBTK_0)Xz4|u!fKT1R+T*h8_8H5Y-BTyuu+XyVS^phgzh!ZP;wP2
z48dK1x@~nS79E{sE`Z^fODYWuSn$7AiZs%Kg*ylqneW5lSosyuXhv2*e}nHfpusu}
zpfPBs3pcl}?+P3+g*$M-6z{{qSacYN!m6bZFieGOOVnmHxtOcA&$Vb~`RbZBwQoiD
zz9dg}Gw|G$)zv1s5YHSs`>B?1@<7>~ZY(7i*RkO+(sO;7KqE<}R~mytv2wh>ZI(3;
zY>f@gunS9CTLIFKUdioKe{6#G5yYZU(u}LSY|IwWtz{3uCXo@|PaGm{5`)~NMLDx?
zSL4l6`)F!m(XDvh)lO*;+PRTJ&b{?~ddQ0%((HP110!neuHr-Ojb2!W9C3SAMgS3l
zpKP^60q;nJ$n*<3JMNdJ)>G>;QD%L1W%AU}63X9|NxmPHNt3qTf7*%izC1OaGztf{
z)5H}#xD!uW7^Nl(r3^D+EbQ_KF1urguiVxf-xPUy#SSVjooUOlpjj!eYINBJH^HeP
zw|Tv7_Wgpr%U#*vY$<HFG`%ZHt6?how<xs;y_9+7F7M_kJ6zAb{Hy!8WLs98Uq=~K
z)OyuJsva5K#<|AUe-5coc&znUJfNyOq)%=6Ck?5XCm*I$6}8+=#c^07;z(7}?*G#s
zo9IMY;z1g^vHnOl>P|R{eVfcOelN2$mB@j1leXk9d$62#3@L(*y+gcw8UDS0JpD15
zRVhn=5-$Jw{VCqi$_VQMA(Rsq)aB3rUVeT0*GQJT6DcIEf5~4)=3}I8<SQR#21y!9
zYmMnOw8D_`nszZ*K=7?e815^f@1Yh3>DR)yjC^f6Cqw#S7@52_F}!v9ln`&0+FPMu
zpa<NG#KX?`8!da|zJcyq$0&gg4wwLrKyklkQ}SYpp5n$~s1!O1S-z6&-AHR>uhOCP
zKpEtEGaG0Rqt<trjenCIxuV&$?c?$RI-wkn95-5X&`M}<lYDFYN4&l*7?XO@_w#%Y
zL1<&n^nFjU;Jx9Vy7q^+w&u!yFl+vKJx^EZgjtSrAA$kvh`hp!sxZ)ve0{(kJy*cH
z2V3hAtqCnB9X73YJi`FIuvU*8lpHK_Bqyl4Rg=t0e>=sJH-7`g+a?l#(Lzm5FxTVe
zu^?;+53oK&)fu5JOjOs;=)eWx)TDk*5rdcIEAj0{S~%sDz(CjK2&&lS!0vl5=?I1?
ztp4Vl>zM*Rd^$d-ou@V!2ZB3U$Gs8CrM}Cfz~_12<1Fg>om4;4S(IOn)ID!E7$}-M
zJDk*i%T9nQq<>EIL_K?oX6M=Cp{gfUf&=l*q9<<=qjEPc;i1`#A9#GKVN(sYlLP9#
zo}$kk`Vrm9ssdcNqx>O?+EFaJkzzdANc~+r*LqnsEm|!+bx6a?lWQ@E=j+;^7?mSM
z$DQ5gQ;&K#Rm8_P$@t2Mwj<xIy%`A9N$_Fzde9N3zJIfP@|zJ9D4&84(`jZs+iL9r
z<rza*u;)aq1#5&{PA<;yc2BW8zbYpcH{O#<0pu|6tm!N1(T%h=q;?dI%@viZJxc@u
zY))OW?H)Og&3oc<9?K5~L-YE?a4yTcD4#)b)FMX(wiL^`EswZH3e&s*WIIXn<IUdB
z$r>rrO@C{jGP2`mOz)7m$zE<P57)TvdSxpX;^~Ty(baS1XKSqK@Cf)whU?hp;-u8u
zVFjkJ(nw%{ZCKTZe4B1l=ew=6L8Clw8z4_`8@DL~(}+o1kLyd$Q;$U({|T)CfD(YU
zZ~wo_>GGUHj?Xb*%8f%o$7|Z1LaEbt3KfDtb$@1EFj6e9WnTFWyUqGwn)Qu!0w>uZ
z$5x%dXKTaky*dGI<$Z|d<g7mN_RD(&Xg;@ZrKF<vnhU${C^MAzglSWm(G?WctjadU
z$N6Q3vQ}t)W|=Wi{6S>~6^26lJ7tER+}tI8=j<73;54ZD0s9crR$ve(JGi}cJ#9b2
zcz@gu6ZZuw)k9GEy~T!B0==1R@cq?BdL#^ewQ;Wh$|H!8le-V3XVep*NyspAf1sdH
z7@yjmQBWuVVdYLi@jsZoO`))$^|uxZD+Ne{^;4m+{``|nfS!gaYQ2p|p+N%Mf9=lB
zEA{1htncC5Hl{Z9d4DI`aDD%H`U{BGCIOC@(dR4^0yj99!4d)!mu?F#Kt;QSyTet7
zOR5x?_b-mpNFEPo{5Mv-DN@MVXe5gI_=%(#-F@qH_rBWfSI;{`omatv$8P_%1OA}-
z7c>t#+j#)x40QXK^9wFKf8llg_x_KSF3LXdqCTgo^Q2gI)s#2T#ClkkM7%rk&NlU|
zEydEd{dn$q+_E;7wX-ho{JaZrc1LTKqXC;QyfR^-du#w2X)$jh)G=FTti|gtAqhU|
z6W7+f^g1P(d2iqElXOU9Y^98`iY~#Nx_9xfxBcqB)#2!9ly$Lle-M;mbpH9iN*@SV
z2Om2`6@j`R|8`$je@dyvj<U}r%s^|$XBh#FPFWi|hol_$O-##+H)&2eIPq48pC8{k
zG4k=g8UQy6m|#L1_>$U)s#-BbTInPMZ7+2+@^PT^<jj~Z967lQ#&d=g7iAtzBsLO%
z5|Z_9{p2*hBusC|f5ZkwWUs*Fu);YNt(lzfb+A+J?O?;?Y^q(8v$31r_GymZ&Jp;4
z#0lqCSZo7pAI(9B!l}%w9rxSwZh(S9gHlRi*2%s3x)K{ktjgTPdUdmP6H7yNs{_Zu
zZ0MxJsI#dc>E-%7xAKST+Bz$bX7ty+9{QjW=97)&7vz}(e*qISxCO(e&5W6he!QPV
z((pY8m|(Ju6SJ%{k(7@xu@#-Q6=_r)il(cM(aq@CHKtIPwe;ntRt%xe;@v9F^+c}+
zL_?W9oCC2+bbXs~+!IE7&3#ogMceEY!=OT;cA-PPqz#M+4wdyFRJhky=QZ1mVgk5c
zO{!NQB>BSqf3a6Ncj+xEhlArO<jG3vqaDJs$Ve$lgdig$2M$HyQfTZ0ME<eG^NvP7
zO1PX49W`9?5i)HnT}xh)Pqpn&ir0k<uud5hrNoR=GE8w7w7WTQK!BV>3CXBuuP>*B
z%6O8q%|wqrxNANoycL};LwXTU9h8pt20S_3dlt`{e?r=*PXLZbKCZxgyelTB5DQQ3
zurt{)Unj^>X1kD<P_lFA<zxVgBWjEz1$Oh^H{_+1czXf^j#MBoC7t4-H_EVcle3p5
zFPqyFF7_j=h18&baiw?De9}cxMKjTRNd4>rO-dp+L2l+Y+#uCQ<-y;2zKHByC&#O~
zA!|AOf8^~)Bgb^8RP|ON3@?sN1xmlwzktLc8(a{}^lEuMj^#CZHM!-4J=O*8BuS<7
zNTyvbue@SL^7Q%l%h*<K{EoVhdxXjiWd%Oha7m{uwC!1lj1M;Dp^!++vC&bHvp3z#
z`AmzCWC9PfI}a|u&oveu=DwUud`sqM{PAt8fA6gt4%{R}-<(<+A|~!MwRHJ1oXsq+
zD6$yKM-BEDDY6J{eo~P|?9#BvA|y8{vRuYQIw^U+GkJ=ka?PpF2%8pK43M%2cNSWN
zlq(A@V%6_L3rsLs&WjVpmZ&TzPtsnOiY-og4ttYTZHq0w6kFW=#TJb0IYao=YKzd}
zf7dIwpbskh$e&Yg0gRle+s`bwAZIz68m3EQfeni-W5E|{E#t`#Dz%^?1&9wWwTNBs
zEVT$}pHgZ;K!t5dDI_j0wFpUqhS%6rmSdzuQykk-`ZDj@=aVOXt#I-nrCwvl#i^Y{
z_l~zQ2L=}>_PXbF?13dsIHlj_(JKWUe=&OrHi6SVbQku$Ng@YJ&)J8>hUU{4&G8~k
zki6q%kD8IBqt~m}o+^8W7wm(oO~95i!+(S{@DW1dCs&+=HjQddQtJIxCn51d-AP)l
zW#tK6P==GLnaa~VQy*TN0&@JQgO;2X5|`Je!0>}JVw2nbsG3zAU)U>(R-)&m2K;|D
z)wL6s(dR4^0yr?2L7yHImj#M3D3^N?E}a3}w;2*H_5=j<I?b_{(dR4^0XUao=sXhx
zF*q}q;Wh*(mt+?%mw!b5uxxWV|NZ{Qldi@-U!&f#)Ol1aySJ1TXkx9%+Aw5E&(~l!
zB@iix!(KnLgtRwZ_G-+E)D;gBDBh8f`f^P>H9pCYXR@cFgIG4QZz##V6dh@dy_kKV
zqDyW_k2mqxm-nato|;myid^iR2`Xau&yS}xj)qO}u>-cqV1M`X-|pMfpHgUbvwY?l
z$$rz0w=&rwI7JjX2c*EbWzX{;uk$gy)_oz1YXAKF-ieW~Z^PDHR7Qi;W<ICSBvRT6
z@0Bw~u1N8Qz0cE=iqDh(?aWRdxGbGPi^`SLuLnUPssO!eP7}maRqN}~(>W&&nLaRb
z5WKbk(+ViC41abAe}W}j95oyL`nDe8&MVSJnzOK5JCE>Ndj$iEPzu+thglCPel-dO
z30E?PP{E*#FMglJX!#huwD@|Q#mM<b^7u1qhKhhte4WSYPQdN6J{mquJGZ`%AIMR4
z)LtyZqXt-roou@)drIHoy32lRN!dWs{>#{YjGK^c$$u!VgAlQTGOKnlV>pyy)gYIX
z(+2jBB^Mk&U1aph-dp@-EBksd@St=YD9Y8d>OsgYq}(>?5jthpNUtO`LZy~UDHI3T
zBfpCn2|TmO9mIHt<`wviq);mUAnrt+o<Wi)?~)Im4rfn{49)B8dE1c<YMW$6H}ceB
zNE-Z&?0>kAgHZ+aQt~DT+mt49MM4XtngYItKMv=($ivs=U3yX!L)#@4@+5cv?kFWw
zloA(qmr|aWYn`FANS%y)ZK*py_uhw7Ew+#37Qq4tR-%agYGCGsFH=`J4&7)wWYvy0
zxbqNT9CC8brH=a%Z~lr?s)b~g3k_)!x#Ser#edQg^mjRCDp1ByuNFahfZdJ}0GMty
z8N9LPE0#5D+N@FD*)fym`_bk8i0!j$D$$tDdRWZ9nTu{HSl4n8xQyTetxDttvza~G
zY35o+_JQ;tudNe*BgXOFbcJd@afv!Dta&8eO$(}?KjrVi*w?!6d>;ryh%=h>E3oEi
zlYfc}Y0}Rc*1}NO^y}g_e+}#AI%#w!qj{HdjhO9rvy}3IsTQR-1&*%PP!W}#t@XvL
z#bZ;ik3F{T@$nug?aJQV2QOA8%%UPkLe0zM?aU@1CD&W37%S%`Z?I=`8ea}u5U;t2
zaamceCxl@&NuRHr6<a~hq~xKM2MKQadVevQnS;gMZbtOdY-BiwNy^X}+KzA(hAH4)
zxZnywgrE(vj%0eC%`d;UZ@XBe&Aya?mQwPnmzu0;!01u7C5PBH)ef5?=*#Ll+m+6g
zv+h*xU43MxwH%F5UF*h1E$f*e{<UC8t2%!D+J#*EastUC%Sat+zFe#OVYi$2+<(tV
zL&j*KIbASOyTxD-SThSd-V4tjuw@9jY->%X?D6pxP&P!cpA*KRQraiB{_X~J#5T7g
z(qY$oX0y~$68oB@#Ia`r5tNNHjTJ;@{$`f7(D*oYY_lDG=WE1{xorUcG!$5UTvM1p
z%OPzjhC;^`kg)FqVj8?tA<Pu}o_~mHC;^?SSg4^^Mk=M_m$Ri)1&raB9VClEk#$o4
z0}Tt$g|yrSj;$}rIr)puIgRjTR85CT{m!$Upz1Pmo2uu!YvPrW(GeQG7_7h-VC&+f
zE3!~RNni-l-U14nMO9l@I%NG{t(a;fy6*C&7C{5(P_NCOdhTP-v{v#6*ni9f(8dsE
zwqpetc7vPw<<LT=6fyp;LI#l;$)PlC1%tpx<!~rm+EIy;BevrJsAHLF(24`b%==s1
zf$eZ>`|ud7=Nqh*+xZB{im8V@*Ia3@xMQm=)Kef<x~wINX(Kq6GA?DRs|V6tja3eX
z7-xrla5oSgU(eAjoxySlwSQ1}AbhHS7em^4HA=|Uio+jYaip#hSP~}u&Vs|TVFrc5
z-`E@njQO3+am{f6_L;uf9G^I9xXA2jjDsc>%tMXwO|5H9an1B^h1L)1is#p_=7ZiS
z8=u|~xyMd!>5cRKoizsE5UjN1;9f9CCs*|#wd8kem--C5+MOxZQGZ7%DKV+ja7_0f
z2$hLDT+$XWZo4>J@lf;sR~m(eFS0S-_Y$7TaT5V3D!!ph(T^oHuy3Y<#?{ccpMhai
zu9{JyJS%f|zDNbfr);J6`>rLB-t9Oegyh=!GUEBQ|KG4)2Rp@cH*MR#E+KN9NZ0;G
z3Ns9ca43aE_Ur<KT_7smZ(nxyFqjXF>iY=n9;Mp^Hc|Q@e+g1y&iWrodEu8a^9mCK
zHa9Yt5WF=d3pp=LWo~D5Xdp5)m(g=QD1Tb}ZyU!E{@s7Y0r@2@pu3&N?#>i}fjsIo
zfdUw@3$!(iKo2TWs)(vY1<rrpZ<ah!G9{Z9=}3tMlf%1x%=??2yHY84sVkLArySqZ
znc|zRV~!hAI^hyms1u7Jj1Tf4q_f-xYe&f>sN9iIc>=0*q`9q-Ix20ZP)8N3s()h*
zs-;~6Whqt>0j05(U_G-M6xMTYn^jPiK&f_|T(Ji>2a|1&U{~7=?^rCZY3jI8aHZ6-
zX`9QPq?&_*T#{mCOcQCstAT3bt9Q6mo+wCA#c&P6NTyghLlrs4417p-?J~!VBEh{1
zs2Vd$hRRaxTQS#D@RzZ#*EY9ynt$9v4(n8;-StkjNm-z#()3cO)7n;s!qiIeTxtha
zLN>6L39_kW0oAO4A(>$cSR$xWu}1-ub9pPq&G)FK%eltB1=q3%g&R2x1uA#ho8-Z6
zYI1~fsBvmoiZynN1=t;XgwPNquHkMj!E>0xnvDz#c)JT{sHub%taOz<_J4r6;Q?)i
zfol4LFtAw6b?<7<@GFDRwKUIhtWqh)Ij&i=2TW1jVr9>radT5Ioyq)eg>#37!YCF#
z4~ju)8<#*iK@Kov7MvPkhKI%t{_M?j!dacHQMq#o4lLoW!Yx46&U2{Al679^cfE5D
z?=-j%rDWK|5NVW{+AfHJB!BY(2!f<dX%taa<^ysBSj2`wB%)M<gk}i;uopv*pmMuP
zL55B(qKsJ6lg~aIzIuh=X2kmU!@vIhA1qgUg9+e)eM0=6UY?x1e){wsBgM6^@Ui8{
zZuoL~GChBBws$b;Bqnsj^S#+@a(>$J{O8}_%)Wjx+nY@~nqPP}J%62Hk7v?ShPGXv
z!80g)sr^7OfN7uqfK4bvk@)fS>G1q~dhlX0do?`&>e=ve@@BT+_~mbB6YTus1Pq<d
zCa1HDNEq`azMou7FV7Dq7xNqvm%f`E9`AiIee)_s>Vv3orPr8$j+t;<n=|tH>FKmB
z1gC7<KY7)XV2II@;D3UUk<gLIBhe<lzT&X0*bQGy&krZ(ZS(Yc_{Z?g@Fjjg1+T&E
z!R!?=_c?;Jc&|`6W$qpMIM>{J*4Iyl7nl39SaA3F^uNRBSIZyFj;E)?i{YQ&e-r;d
z{b@EkyZGBM{CGV3>2kk6nEo>SeEG}S&f(<O;oHmap7%eW41fD4)BWM-2wCOeh|Tti
zjz|twj#MUfQuhz*q4H5WQa^GUF3t~zgX!tT?0oMSTE+sO?5gA)eE46Wp=;*&Kw$w1
zedOs;Sat7N`JsjVb?^LmZ~tVn5VaJDTGNK2QbEYsQq;K4{?MYHo}A5vtP(=j&4gs7
zKJwluB&+p84}UG=-qFzylQ(DQP0*}q2OoBupf}jOU22;Dm`hy<Qfp-99w|DilAeK2
zgl~1|=JO8AzZkxrzMKwUb)OPSk7Vous7M41(IjBw{^~|<R$FQt7}L8yDs2OecJOgc
z+krD|*Kno)XQCdUx6$avs5u%HLo#$(XJn;5()IgC!hZ-o**d$6S?}u%WX8a<)=1L6
z+*osK@OlkIZ(!|*BC6&UL|uUPyAd_2h8u{Qb&05A3$zqp5Sm6}1Y6}>fb^arnoI4Y
zynzlDs9KSIc_l;<w(JQrYT~S@0ogW+(44V9FTp$;K;UsDM7SS9)HlrhIPoC&T?r~y
zdQ{a9Er00QISQ60CDol>E*qkoAgocZFVQXcs*fw-;s~{~D<+#sip0@ht5tBs=y8yH
zXhHvXxp%mfl%p);N;jcJ^(x2WOnVEeTpREnRJm>B?Wl6g$P8aWc3Wp81I5!iYYLK>
z>RKZ+Fs-gL636;pWv!8nH48sxs48Uzs*2tRs()%)LRBscJj!(>M)1lbc$E>n$_QR%
z8{mZ`8Vx4}XmLS~C4A$|uLw)ac?G;KCM^#C+FRv~P)iv_*|jsMiuCgM&Y&a-&_%+&
zjbc8a*=xt1`#8)q1G!t{FheJGwU}2a$;nbms^lq2(OZ=Mfm|?>L5{}d$c0D9)16!x
z0e@wMlcmZMUB{d^X9A~`KH9Ad9;Cwhdbc8?H-<!uzgk38kvz7*LoK~TS9igL7U|z*
z4H7bhYR!ofBT&H1Mj=2Y+Ai15R+W1#kE=0Z1OznEn+RC)I)|$pqZS7toJCT~Ym5~4
zKAQCJA4vg3w9f7Xz>IGFLp2(QXVHh!d4Kl~KQw;S`=C(O6%?v`FBH1gf-GahvC`(i
zqKSH8Y2}PUd^JI_(i+iy(pwrAi0OpLDvYuwbJ8rdQFNi;5?SJlQ*V+VI=D{Gez=&N
zUL4Pke?6Z4#&d}Ao!$CY_oB+(CFbL9j@#-{-tKjjH(z>Y9m}FRmPK_ei)<{5Y=10^
zYzt+Pa7Iq?F2joq^_@7%BBw3`D)g4Jh!WK}&KFn7mf{IY-5j7+Itz%M=#{k2LeW*E
zE;Y&p%W!z0wYzQ>LIOY{U&as)emx|7T!+z=aTwi7c{Und2)P2r7Du^85F2Hh+G-;Q
zTeE3VQ;kGgVD;_UoZ)ftwJ^CP>3_5HM>jb{X1><#Ra8cg?NCmr$XW8ujA^<~Nn&fC
z%x#^M`A0vU-pI=zn2O`s_}i)2?H;K(o{V3mVjmw*`dBUa_;}LC$CExDXTAPuQ?Yo@
ziSBIjddh`yddt}zbq66m{OMCM*9c6{f7VpIFnLSG`J-F>50HrwXfSwstA9P}{+{;e
zwE^!*!nck5gC$|Ix05iA)b~lkygaGJi$X2I#(P2wqKrfx2^)z;sT{8eEeIM39tjzV
zapm^*-o{VHN}s`EHY82d^FZ!6`^C#&z42=OiiDP!k-E7k(tHxxLZ_)c{)95Fm{9FK
z-b%ZJ3Acbr5!|a%GD}KZV1Gg@HsvJ~&b`9B?_wg?;t4sQYdp+%sg<_TbBotQ$`unL
zr5C%~9E4Ps6Jr>e5WuwRW+^S#czNL3Ye%)c!|=d4Z87tcT|mgrZWh@Benoei5kjG2
z$|xIv8PRzXg$YtwRW(A9+}s$&z99l&k8N~%b6|+DqIU3xV}klkB!8N4zUuj<G*@%g
zTwr>$vIP+JibAD!;e1+4$2GUUkPw`Y?f23Id7qH%xyCbNgDOgo69OB))PE-?sGxF<
z_lb+b8KQ<nF@s!;O?o7nd)nhe-kh>NUgsPiEoWGwJb3v1vUhMkT|8_dE<5<lyy2r|
zREe|{lo&2p9?^>ek}acWk^M&96mJB-{{(w;)VqqOTSr!SU|wfryhRf4nXNHWk=p2L
z&Hn?og@mY=(dR4^0XUbD4*?UGtSBzPf8e%%z5TwiRak4Hq`&FbKYf1N{JR-NM`u{#
zAnXlT82<jTNdp9I#K!^AM4;j4x8eKdkABy7R(H=p1yG=@2=fdm6vG?_5oetD4Jg&m
zvMNogB&EA5#R?Rw!lWRuFJoy=dS?|>*30gT(nrNQX{@F(dJ_ppW9{l0dN$>=f0L2C
zju16-;Y*QAZK(!rZ?oj80=<-#MdB=3#bF@>S~52DBJguvbX;(>!3693Rra+B%Dlzo
zLs@)ZA>?ft8`a5RKBrktfus)6G?M&}U3pG^mmtq$dB0+wG(Ra?cF8d5zjQiNIe7##
z&3tX3mYvk!lco!kfHi1i)67W)e{?#d?!{7IhnMOA$RVJ50O=JRyWwZ@hsE2kUrw3?
z1Py!WWuR*E?4s@x_ngJyNn4^i0J{FPii<V6y$dR}V+o>ni581aM@QE5s@QTwL!L)p
zs|a$lX|d&8_Gs7Gjzx^O?+)TLCwu9>A{$miW;~UGaWHZ0EF=?xw@R`te<8FUia=_B
z&<V49mFB1zx9pzQj~$JDB~`O#8CJi33{|V@s(<)6D<}G{ijy6tpX}YUH|o_B_c;U3
z_tL=2!~tXGQ;4jwHphS2b@t}#=-CUMG39TP{uH<L(kVucqlXxb2rbb2Ty|5SQu6vF
zHZ$g~F>;N?kVbGe|As_Ie?cQ(x#nJpu8m$E%ZdE}xR=cN^`KKs1G#?Huap1Qa$n7j
z>-mp_jD-I;|8+84pZ`uW``MgG-Uu&HxMNNv5)=6&b0Q%a2q)Q+rFrK?xsEIvAMu)8
zw>aShH;9r}@ZAXyN?=Y42P=<CV9uj)E}l0@=_^Pk1-UAnoq?+Ie=vgko&?rOb0mTF
z$*5bAn#n}M3)$*3UL^mGy_0hqIF?4D;w{k;;tiE}Rh5o-5%6i|tnE#vm+@(_;tK1W
zOg==0G(<ZcuG;wofkmf@i55HW*HJh}r>Ws>5~Tl2)bb8kFqiD(?X;ZMWSmE<$PlQf
z9kK3B>9NpM@!DKkf1h$(cNufBn$P3HeDCHWZK@qn&qbh{J6Y$VOSnh-1i?$GCp{NJ
zI2Yy_NEZ*wu;u}WK$haD6q(e#${rl@QG9?tYTDz=_@H>?S}~SCfbGY^^1$B@3eh+g
zMF*PA3N2Z*<tI6TQ#zDhk7?)ixcAO@E%y7>GDT%8PqCpQe>H|$$>gUPr`;hsjE4pG
z9IF#$`ej9iQ??2V)2V6JRCQ7N8u0z9_eLMG7His*-6#5x6GhxD{G?J`tMTbb_}vs#
zT1%4nM+Mcd?yibT^b1tZ`K$Jf=^ItI>W)1{D)hW_AJ|iX_v$!PghKT-d+|3j;Fx7w
z>m!csdcr?he+zXh-L4Pn9YDK;Ko4pi0F~uHjw8Qp7YN(;o1C7mZ4r{41)x85rA2@s
z2C!dnX$Rm&QEwzFO(YUv_I+t&mic$34=A#e&e4YMP`f*`C?+d+iw+ob?)^>LTJ(-m
z?rTyAQ^9*jCh=?{`74N|ziR;f2Z}n+ObTUgWOH<Km)Pel3<5Ycm*K!A6PE-oF28?*
zKL7pp$4O>orSK+v4Y&T+*SFKZr%`oOf{Y)GQ9@cb{PXh^9%vw8tRKJ#ugvi4+wlGL
zXD_t{#sLE(jUk!<gGLZJmhOcLj}kD%hG<rr?UI@Skqi>~WPs2-b^Q4CW9UX+U#jke
zLOR_`hv#rj;238ZDafmSMgTSF5%qrxcUDX@;5s<KXI;w}f7jvrExj9@6YiSdlh@rg
zpa!AQ?yv#1k3Q*UgkT|C^^CS$vOY_0TMgzC5lRX0FvG}(;I)qdi4w-NiBZ&Hbv5gS
z)X=XcFWa*m@Y^#T$rRnD5mxk!BX){Y(34ym3-)m;Cd_zHIR;IUafa{4OH6;hTzvBS
zQnFwjQBEr4Y#6oW3-jufqAjEK;WQ>^gfixn(wIC7qt#{_%Vd<!aO&IswC2DDMzmXP
zE3K^+2B4pTISi>y4V{1vztv3FCWSweGDJpbJ35$oUhBt_HpS8Blrk`eYFPcLx;})6
z!+;83PLWDz%|^EZ&R9VbX8M1~TwAUo%u_vSlB7j;%~A}N4s<X1{lprr4S4&U+{@Y{
zX<=}t{Yqr5YHA&0W_u~`5DS-Q!AUb)+Z^T_bev!zx;q4$*wkN}a%AuF3%gZFKa(nN
zE$&2>2*+=>mapi=oCk^K;~UrLvGU7hPpIte>lrjYekyO0a)LDN;LU$8H!rsuH3B9I
zHR(qtp@o?w$|?{6*$UbsS$i!|`H_O2lZ2RhDy7}7)z<4D8rQ9MxsCM?{h8Ebl~*V)
zb_STnMvv{RF`8q{b5ZGP45IMN44XBkzu!@FG&iP5J6&~LE%JejW?F2uXOdeIW&0VM
zE#}d6h#N{~gN4>DwMc(xXj*r8#N?5Pqs?4&8=pW%S7xS4m8?8XoSx-I-p_1SLNBZX
zvKh1aH%&dqY}c;XV3PDUk%7*M(Vh|uMYc-phL{;D$)U_|<YsMgV{wN;_C@L$#?e-`
zsh*^9oj6A|Nhb6CxmL%Te4{Pb$uC{LsWSWNtYoz*M8wS3oPK}6y<PG3wdO#ab*bJ;
z)S%QETAL!um&Rvuq=tSi<*nfOo@_ZVd%)MYZrX5pet<$lX0my>);dzQQl*c)gKiOl
zImpa3TIg`vD8dz{$l12q6?hWL&g@V5GTNGLtlNz-x~>J2SV8@rw0wLaWg7(VHIJ*K
z-0O%MdnIS1Ww(DO-(0gjZ21}%dh<4Z@eGgME7P6rd(~Je&CIFqsm32uIx0BV`yHiM
zqRfQTqfY1FRMU|i>CEi-#~5Mh@RF{$M0$68)TC(HbU;Y=DYa<^fNeytghY5}v$1a*
ztA*3Hm%O=+RUxJO8M|AD+p>iY*(JDa;RmcsCuyD31?ztTJ@IrY<3=e-os7JWJTs4&
zd=${F60D8~qT${%)=KO-BVZ<I4Jr$~?8x)l?1x8lf~#K8mhZW#R@}+=OJgl!TM2@j
z$&nJH)Vt2+oRlm<K;2~hul^IUP4N4UL7LOVW$G+Xxov?WO-)-fnzUxijo0KDqA%30
zJwL3NrzU@9j{B(lHu?`0{>`eT2!wXFgD*$l>dv2>u+Yz;N(ZAEmmW@AdzdY?G;nlt
zpD|rx=W&eYss@Oika|8bYFGM5Sa8m!mevUan@Vmg$h{qb>jZpaGltR#R_wrT44jNx
z&%K2;=VaRQO{-^luWiSCeKk&VceP@`-Hp|7NGpG~sBo?R6P-5{UTMR^@dmqas_<{>
z!vHns8vLfS6;P7<yH5z1X(u+r|FGj3G|2hpdqcL4XSlwioSA<35scTdJ^JWB0cEdf
z>btf(hbZR{dG0*O3x95uIyc!f-~Tt+VW;6rqCl9zwXluHTOtaiTGq;8Id%b={tF%V
zW{DjP?m9=;GL&G~>GO$Iz#3u9o__?pJG01ti@=yim(k}e69G7v;jAwdmq2z8D3?HJ
zJ_MJkG%l9`j<-QIF0KXxi_318!F4(l0ya36u_Y9jv3&|CmqR=*m;#`Wm)$%rEq__$
z`qo=m<)#$Va<t2r$c;=A>;kZZ1oJB+xMYsYN`h1GEeOWT;*JUri8my^Ch<u8X?22a
znbY39{^x7;-!C68f4>ZB3uH>WPJ?BZ<off+i+>@3WH7E1f;o`g&%e5FFTd4P)B4rF
z0er#8T$3(^k%eYT*-i_^0O0;5rGE+MIM0ul=IC)LiInV<9VBrczJ7l1sweMX2U7^r
z+0~-K7yl{nI*b>RC^R!X>SbObn*_f^rEm~5OGyv<<E!RDZnwU+-ZVjN{|qCAake)`
z;Aa*_UN&0<t!CVq8*IWEVTd#CX$Tdm7Ej)Xecpfqon*@D#*UMP4xYnQ9DnDqxajo~
z%$0TFH}N3vCgowYTd=-527jl>&!NzVWYM!9<)4@uuqv;!e;a1{6?S>-2JbMPkt7M+
z)KQ4Th^in@?^k!2amn(Ud3hV$4R20)Y~&O7&oZTTj1HsDQNK~A5{;{#n8-4zWW83~
zcgL!?Retvx-EfhKT0~oeAAe(*CWn|E*VI07t%hBg<7Q!th^*Uf0VJJOy5YOJ*Vm(r
zAW8;niYab1`YO+V*pry}j`{`^_X*h|c^ILr72ccB$D+dM51OaJCoJeqe7VD#zvmRz
zTu*ct5gS_q$7r(<FtUqObz-t$LZcYP=~}iDIeS<^IKhzHIZnY6{eP$Ok8A3oMn<u)
z=9&=l)>Hgm^RQAdZtdX3y;<`P;K|hefYQx*3UbASZqpPB9kfl>^OP8j!k0>suXysF
za+moqm2(#bjv3;W7b7Kz8Jb>E0hOT_vLU4zqN^#+IP0x}8!UumwxTTfY$8L^m7$UK
z=-FiD>=OwB!!Jpu>whM_W>4t|O=vgRlTr&q<N-VIjAF7eee!<Y<`q{#$|{#&sz%Ns
zxKH5PSWYr68Xd}<bS~9X=k4h!8V2OJYssNm*G*KVV-9sq6>vV~Rgyttyr`DXP=p#i
z&fYK~r|ZPCQBp9y7bFF-8!&R&--xt+w6ULU#geNx<82{p9)E)ziMgq%-eS+G9aS>j
zXvh3y#g0N!ydh<4JI3_;7N1YypGm1|3ZcB<ODWqAX~F;#yjfd-dFv4fr54SaoRa-*
zg0~qDuV^Motl!w%eHVKCL8EmnCoOs#!VVLr!4H4HzCfR}sAunslIvdkqNKWEUu=mQ
zX&;olpuaB`<$t|VQC&yLfd5MGoe=`GwAwryAtl$n5K>ZILCB|bl#&Ced3hu-DH?e`
zlN;qMrs8`i9<^ZFHLCz}UK_2MQ;qIB;}ET7r0tHUkBZ_xwb)pPopm?3nnA@5LYU>I
zndKb;5`~~6Bh-p61jrn!s?l2lq{_>EM@`fiI_@^hEq`39wun)RncC>ml1pwF%4W&^
z$_>?FLh$$%WG<de51AQ~B&%=>dwmzoD@Es1+!OO9#s4wpOGU3?z8sBTMvQxNpcuOv
z_uxfJCMT6gxj$e?d+GKp(Nq*44jA8t4Mj~ime73Z%oE)QAij<!u~uveAl%WY#;~H#
zb<_BHgnuKHh8b%b&i6UO!M&DL?{$O#^u{cD!ug^k;Rr1~MiNK|%W#qS&AnMGnx`76
zKjS&4rp<x?&fJXDO{rpY8CKA;AT3K(hXozdu+=i`4eeo}kZFzo#rAL|+70dDYR1;~
zFy)T9Jv?2<X(QuGa0w>6Oqyg>^Jw0j(Nqf^gntcYlVk>@bojdA?r{_T@Mn!~jGZD^
zFlv8WjzmWZ>#(LYxP{MjodsA1LfqVShNkzL-Ys;Uq3AltuH%-jbL`|@j<0YS3R#Oi
zDLUpz+WD1E3r&l~7Mi0&lE9kELKl+gk`r{(n*pThYTp!rA!Rhx%>A~|;7QhZ(Y(P^
zjDP3hbedv$9<S`x7B2d37oD3c<$PX%0azn+t}Z8Kp7gLW@$Cugs-%*#4CR-3VHs`#
zC%PQNY>HiHuO~kYZS~d51lA{;`a$|Bq|MO64EQXJfsS30%d%=bj9!MapIuL$-i`@^
zo(6xgCNOBrG^$1vS>FB-N=yU{M(@)CWPg4p-UO5}5v9ng!#as!a<kR^8$3)dY@{nJ
zas1TXrEv@HZlvPfUHVT~J4jDmYi^n6(Od8rXYjO}d>zu8+W-3fM&58<_GHY!<&-Ij
z$yJ>ttvA&78hjZ-<D`<q$J{vMT%CvTQicmT{MAnu-VV^nK?Y0R<Y|;n@l>1d2Y()z
z_vqDqD(L8GOryiTIZwRoafids&RV+i=F1g_?Q;s{pjPh28x^2v&szPc8x@-JGtpf9
zw7&HDl=~DwJ`|PQ@vvS|-0?m|Mfu8oifZ(e?o$9z#6$h$PDRD_&Q^Iv?V<N6YRUJx
zPf<}_zfV!C+uvO_$*qf8x%mdsVrsKGHH&Nc_64jHOu@77|CCI3YOIw+AJA7T8C+?r
zm4ol-s#RR~X{uF3*L!M}VfV$KA@uAReWv6dR44woYC^jdgwR=LWnrcUPM^P_lUrQH
zDzr!It5u*>wVIMyEdkZPlxqD03pL6Jm(k}e69YLgF_&?hDHNBHL@qObktk9NU>In-
zy9b!t1brLKX^`ZU!{q-DLrGLED%av}$A@y+l*FA+l&rfSo$mfPoZk;$-Yj-r1u`DG
z_wSwI0nMLAGtp7!jZw~;?)_)?_fczb{G9ngt*zz9^Y_4SIz3qvc!nW<c+{BFlSh-*
z@Ei+et#xIM<ymmyx8e7HbE5P5sGUw6qHyr1f8PH*Xi-+H=+f8pYJa`HAO1V^s)JSJ
zV&|+@prQNs=OGQCA?1ARFsfi|_v>5t{qR@IwU<S<8$N#h=o-pbab7U&JnbCO#ChNp
zQRtkD%BG>>JmVUJG*2%dq>GRVg@t^1i+WW8dWF_k3HYKlaX9dQjURR|vKzrnq%P`A
zPFWjUj${f)j+aIbB60u;gr2^0Fo}w!Br(%{8Pn&oj^LE?0GW)UoOwMxIi0DMVHg%C
zK@<zq%v?-C64QbxBLee$SLk*9enpFW6eh3Nj~t8%g@aLDWn)kQTTg+$7G08n4$-ZY
zmFOZDpT$`$P+XFKTu{v26<LI!9GI4yzKG%5sXog(>lFIX=QXYs>QGWa+rVT?E;xR>
z$i-RYT^sq@TQmo!TH%`480>R(oJv$uVCJLj@kO{Jj~r~fl>Pa)ylai+?_rdztI1a?
zVV-}BrdGwcJhdK@O$wbMIODJ}pH((1wzF84HVq(9#<$9UKEEY!H$JHyPh^inv7^Y!
zGDKA)cy%k>lL^YBZ{Z$5MNW}*xNqJ&pq+pe?}GPFloE$cL><HF%>V`z6e6`on3wcS
z>ZM=doQezv&5E>jV;D^9!P=G7oa#BkoNG~7Y(*r%7#MabR@vNgARY(MFI9iaO-ONP
z&AOan=Zpn^(xgnnMOF8HG-(|tb5}!IXWp>oh-P`c8;~m_*5&kM>R9!i?`?S&UfHHz
z$HBriFtI)=G`{uZLPjmcgYi-E{FX~!xA(_+R!l|kcf@(le#QORV)<?~A3H%C87v&!
zgxPxqFO*6og^k<?!LJS6>$(S_uVp*A3Hf;V?gyoRsa1>6YP1HW?3&ol|1gl=ps6=}
z911%$O@&#zRiMIkLRo2Dfg#2-#I*K};4h!={En7!YeTTaaOU9y8^SH{zg>|P{<Dp%
zzyREWTbOBnBLG8!Lxz`z%Wni=i-wN^@s0p&6F|Vg5yY!6;sG~0$@!rcTC}Q>q025n
zhUgW4Si6eIpVFjsZ<qD(8RP(jMwth}G;q#DZy$uysci(jW0Wgy7e9Xzx*?tQVyomD
zblc>nnOP6*g26Hl2igaErzZqzI^aUtSyVjRjo0U%wCAdYVhapjljSO0Tn4z!{&?%>
zSBZMEMnG7!#h$(a$2##PxS<;#R^-`KC9y7lWwSVog>(Wq4h=-Uf$6<Aww@2ui@xZ!
zBH=#bYRV1LEaH5yHFko%b`&f>I+8s1JKRF032e+nXst}_8*IC<It3Z>9j%}pnBpj*
z4wR&e=;elE%-Gm;q+u``xeT0<PzyJ)v}%ACW6$}JROc=wmT)Tjapg7&g{+qJ&DM#3
z<|H?X>Gqn>IR%U@j*_G3oRY&ya@6F_x(;c$wjZGeWt?gK4t7U^U+r5<8kZZ@q?3J>
zJM@=wBVR`0IHmH}3PxM7WB9NKICLzVkzPK4`yqdd52Yv*+>EmKmJV#I=K^pu3(gx?
zS-d&m+-}L$#fydO8M%5WH-?YXGf63bMjJYEnJVI4yIpy`#tnCkRu9)FRrkvzQUbze
zWL?R$euf*U{a+2f?}2ITjgp_GYWpjbVQw6%N+>9>e3|$&h%deP){Zkta%I)SkgCnK
z9stRs+2$a>?QLDT4OTfb^Yd@)$_;_C-rcb)PkkFYx7n{<q$=zGFf{!S2mP)=Zrtdg
zHm<ZktH45WYgAF=RDlOI;G7-omB+N-p2xx+qS@{`rylx0&~@kQkEv#BEw?o3(*<|!
zxZ5{}?SFQENS6Tt3=;x2IhT>0Iun<oN-n$saJOGeF6;q+!oJ}6a`tkoG-B!nfxBL`
zJa9MUZ-$$KS)-6xm5R>^B1<_IWdI+g<^YcaLF*9-xfYOaB<Sf}ZH6;|j&^^M;YR7+
z1j&cT*WuU8+rxhg(<TSPG4j8^KZF+qS{TcU*1~$2e*80id-!97npHbKV<8-6E{)N;
zw;v<4c9TYb;PS)YSKFZ|>fM%@H_HMA$%3uO7FAQ081b(95#6Jkd2Om@L&A|A0K;<_
z>t(i8!N-OTp;oV+TwLdrSW|1Y-FcNg@?HSeu0q+{`5`t(H3C}UCqxL|EdvPD31~O-
zT0MQKdoFAg>wH0O2w5ep4cPWsxM*qC3Ii}=L%FVhAIC)(j*IUUgknO)oKR150KZ(q
zFIAoWhMW^HC!k#Hue;^N!l!t=2I0NCt>ueJ47OXW?yc{dp<d6IA~SMy;PPxi;dTLy
z=s1JeaaUK;sPwy}`&6Da>9eHTr}Y-sZW=`@vF60T-;!J`H%W1OX6BZnEyFVXr47h<
zPNS@Um+kU(rAjFX)m*@1(W=d*EjP7^S7=%UJwuF`&Yjd!D}1{xD9+N{b&L|(u>}Q~
zEU4cFB>}T5ZFvT8G*H+{zZ9r4Q+=MJ)vc)(5gE4~Q!Q`?W~AGxHU--Fp>ss+W?pmp
zy%2(P7*U=nnY$UvB~N7Plo&-A%C`$)HV5^80pdM!Y$!-rbpzx}0o-#DYiV4)nnjN(
z#KKFaX<wwf^0Xkd_czFs)23zcpe6{2QX*|<UaydOj}_J2CDuxUI%iy$90`L99xf&)
z?#GOb&VOtYkXF<Dky?v(rqyeanPHBp8q9GJ&mEMaoUn6U|5-}uv7)*v_O2_&aYe3w
zMu_{0QF6{?@^;5VO&mK%w4hgsM9oQ;^tigH^A@z>{7Wy{eYv?LtPri{)PW05Xq^-5
zh2_XG;{Sk!qZQVuMuMy$Ow!5Un7jOKpz1CTAT@U>=AKc$hj_>Oc;Bj#5}+_jMo#tI
zhm?dI{+zSFOi5^wQ@p>Uk_L#Msw72!nS4qs?x`q)Lc*`p6Y+A7_taAlaqJFt)=3ol
z)jCXPO`As~AJ^f>^oZPJs~Q!zu@#!+vAJJ5&1b1e-O+ThtjE2Rl^y=}?iHGlw1s7}
z5=(csQTBesn9d6pml<gk!B@+EPv;N5ixBGRpkwD3x;0y`^vXr3wdDZ&R65IlFuUk8
z<GOkQSWW_L?f#x}2WaOmGMl0-=L$(7vHP@k<!T^djiN4HCsc@~1GaXRva34a1b^om
zamF`ON(Aaivnd24Q63+gV}ILTosBSDE0b%wYTI?ynTFM~gwRNFM67XUF#l+v)L=2!
zo?XaucV=M6%kci1zd4Jb`3kFlx<`NILG2ktx@Nam(|(l>XFSkiUO72w&yUO;^_vrs
z{={UL#*3#uV3d$<KC3*=f0^iPUKiQhrR+COZ*I1+_D>9*F4{j$TncdSy@&un!6nRi
zI>J@C)AT8)BfyE7%suPo0pjPDWH%(DoL!)iTIz#`nU%B-oYu9p?U`ABfE)*=Cp_20
z*<8n#leNMi7K_yLRe#!vJN6r7p=9H2$Ar6Fvto1oTw_m^PD;6^81BB%aJ-W}wjxje
z*y5p9N>e?)ix-WUV7?poF=3VIvT>*q-lb1)LeqTq{*uYggX9%;$j`WWb|{pi8!n#5
zCkLJX0fV{(_zGoiWOH<v*XJw^0XdhU2sad$V^S`?e~9hhUw`a$G4^pE^_r#5i-PRl
zQVWd&E%Js7778?5XclO-u{=v-@ZI9OHPsj@&(b|VZ_{F|Ew9t7=WO8PKA2U*tW{19
zE_~}0!h6|;#(dY<&ajES3&zgJn=C2gYh$1HDK5q*vtw)dY28<dd5)NhE(I;!KB`~u
zU%UTye}jjJm36UkCMeOq`RCUzy$}N%d~AqRG}8S3xB0RAvzA)2qQ2*p&hgL=YZ=9Z
zoU%4Fjzqb@*H4r#4zs>TtWh<1)T#5GSiNr_oX)eVH?l9D6xcp|{QlWgPd+~eXHsAl
zy{VUH-qJOyq+v!RB@=(7(oX&uEkq7$=836zfAG55Tt>pdXB*4~b|8L}1@dyT&!@~H
z<dtl+AwZ7|@@*^f)U7q^Nq5_3u(PHw50P~&uu%#5iR|r#3UpKyTrH`OMv7XFRFNoS
z$;H~7yDw=*58J1!pdYzj2P1iL_%+oxlSKo3Ks*P8jjh8<wQ^BYmrb*8K0lWI-pADI
ze`6ZUxlMg6yLG^}Ep%f=l&9O!P0EOZ>O2^Bie@a}jrp!LlkKdGT%Z>rc*|5q-icl$
zM5W`s=tZJc>~BUdBcIRH$|d9>QAKvO@oVH^!P#Q#8L63*ei`8ihurm;!?5y}HrUDt
zH90e}^#s$^CBy|=^O${i2NPV*Gy+y2f7!@1GgvyV!!r=ruO?>ukTWeFk3N;2H^goc
zqnNIV-3bB2Y{u@)t;Gk{oej(!ffjTXU|zIf0d=*gsHL*tPN>UBZXfQ2y6hZY^7lhp
zLuy9dOHmbTj<soPn2OF;^kl{zFqJ%lg^Ji<EBa3&SI@{<oi_Uy;JQW{gsH#Ge>|pZ
z3+&{He(G>=2(~Www}RJIcJq8iU54jOiABGU=V$lgbLo3SBxZv>o4F<uV<)5VeRJ8y
z0}ChZXCg5<k1W3q5-(Y>j>KaN?u5j|C^hfygv7)S?HPYRq}NB{%Cs|(xSFvQiOu_P
zz+(^lb6eh8e*s2DB{7__l2p~0f3A}ZvYph?lh2_#C!qwmV3?b%zMLcUb`P3P+rN}F
z%od%AOf>1u8U~4iq-UOZNak5Yd~Q;z(F22C!UD(v<>+h{pg<K1P~bCJpv;|OQIU}u
z&mE9DBlPPL6NX$^y~gJvJ=-!!w=MORS7mP*F`O@DiLG&I*2^j)GlMD3f5cZSRkGxD
z_TV$nS1qaReipr3s<q$y89iR995d^%(!r0#)AiX`kqtRC6Y<RU=Gg0bl+C9-Q%lpT
z>)bNvx*YBm17~|;k4bGbhrVed0sivXxA&D6TekhVj%^==-P7R|vD)dnv^t|(@=x=C
zomN#OoB{huOF9A<=Z;sNe@3_@$DeW9*o_+H=%mJadG7Q%Ei<cNX1r9^Nr&3We^Z^v
zP4uQwPo(9)#)4k<ejNL(yuF$CeP6@=>i!(E&IaX&wB^|a;(A|mH5f^sI2xDS=Zuvt
z)q!aAg#P#>Uc%_aqxWv9v7TSSogq--9bADy8kvtKPuKcsN6LCrQQI(&e&u-ZxYwR?
z@M7O-q9U+YrRj!y1e4mz!<p6<_6X_lVub7YoQn#R#Y4BX;Y`K@Mh`|VZ3NCzw2+G)
z-LROkq6_=|ss~4p|3A_^YiO6z=PVNgI5RPq5xg}d0Wg;g84WRi)mq<g8%GX)_g^u<
zz4RVDJLHhVnL}Y9Z7#hYTp&nNpm&B*_=ns&=U9d!2kHIw_i^pDq)3ew%dK?}mZ=q&
zLvn`XNA8x&Iiyf3OAz+|8mfe1*#dFF1uB7xfygc7ltD-dlA2%PAVr<XL7OvY2{or3
z5E@!>=8W5^_4h-6qJRMc2qH^$iUV09tfvY}dhr=lb+ICWD*L^HUu$VWDyKk=$d{;~
z)bG#|IGL|sBrtCAJ1j7(Uy~YGn%gKXaADU#Q(#lqKq+vmZjExt-0KISS&Co{m5{a9
z9?ZGk+zUU{!a)mMlUwhSl0b?7S_o9_b%A#h`x<bBBy$aaV5Jbbeo$99E7pLia0@w~
z1rc(~;T$N?OM%f={ay%=M!uR+P|B{eK@GNE9=1UHT`_@b(%(^oR`WHiVS|$E0#Vx3
zf31+xny>2)OW7q?d?ZtdHyBOimiuua7_9{F7WUnOP*DADxnpxFSfk-r>2-1Qwdbo4
zP_kQ8vA7j~*ROG~)a%kRC|Icsg)G%xzZQ%#gtmlhTp+3usBG+OXfPqy9>>j!^ztlN
zb#-g2p_2QcCB`|#0V%#cIC<n0&ZfBkTJkfFtpHU(`Tn+K0sG!|q(NQmc7F%t6?VtM
z{95{>-auLMI|%Y>uPK4A<z7<)r9H&~*IgH!k|xZ5P8q0@c2po4o6uj&p#0=C_K@+N
zb`6@<UP=qiqYJ;{+^1A}Mubn|=CUD{l4chKJt>vmUkDjW&~)ePue;|52w_8rzv_Pa
z*S~SyVuc}`OC+4j%j=7a!@a%FaWX{&+Zvo$GYU@q{hTewwjeTe-;OWFlh;>AC!^pA
z=~d`|ULH-Sqse99o3DTWIQ{PR^k_N?baUaw_;QLRULa(#vh<6*z;Psncl$5cMDkPG
ze?b5v@|u6y+v{FV#wV{w(}V8i^B3Kl(Z}hn!f$@N8e!#MMo{Q-I=Y;`_q>sM5wAw?
z$JdjSkzdL;OZ+f8JwN(p{PDm!3ijL@AL9Oh3GVcycC+@g%ggbNpmMYP&Ox6%-HbkY
z-V!npIuK<bdLJI%Y1m8bUSA(i{oDTe<$pr=&3JM;n)I5a!|os5_uaP#bo1ra=wy08
zg`y%kAi!frnnHSvegw%P*<trAK=|wKyYZWG_dI+hl2kn47$-ZkB&mPh+naMzufpno
zRj~(whE<uds@Dg?syLX+u^}rWtKuls!<_{yx@ULdPp0SNODOc?tMC2qt6!(ntM`BF
zx?j$xzg{25lkvOm?e)8>-P6&B?w{8`yo_%zy5o!Sad&oxq<M12wH_0lQ9hM?CV6Bd
zJ3h5j;j?rm?W|Gvesa<sot^zW`gk>e!QFq^FOqlh(O~G#r<Oj@*@FLG?^tRqp8XBc
zq_pJjW^jxkB_J{7(s1NtBX^5v^vXB5Tkt<i+b(yTyW3W}dwokeaV*H4gQc8WB(|rv
z^kjVbemXfiM+WJQRh8X+Ew-7lCfL<GQ!7GBXxl6WhT6qPomuKtAh)|JfNE@is9~R!
zi8+R$OsFAfmW$LX&LWZ8I-HVo1Pon=ldw8M_<l~*`PkO72(Suz@*h$87K+_m8T`^>
zr?yn=R3B38ip+|g*li(`IuPxa&@>P<5JP3up)#thD1YlmO(C3)S%v_H?Q`%G$IXYM
z$@$Uo#ppIaa6^8``B7?5i2#W}cE1XYf2*2f-}9srGkI5#BO0is+JjbtiE)s)riMCg
z)>ajtFVuo-i2$s3T4dlzhxos%MKY0wasXJLS_beC-VAYMi;`z0Pz9(nKY0nfnjlm5
zTaGSwcb{tBtR!S4Nk{U7nYjv*raZNo<E!cUyYv5!CO^a6(c}#2?WX7xqv+Gde??zn
zn-xMPit_Z8aJIF%+U|aC<q@JeCyI<duX)(0j}-d<gv-Ft%VET5#1&{=fY~$J(wb?b
zFfxq@;Th)Oh{oWlJ$-p`b(1wKA!}A$#2Q-W3U5K-Evo>>0@p2!V1ZNDVWbo@ufdp$
zBKPV#oQnJ?>u>@<#MIVV-Qh;%f31cawhTAyL2#q$3~rPGZj=FTlmTwo05@!a8#cfV
z8{md*1~-&zMD)!HPe!uhvOU>JGNJ|z{v%{3ZVk42)zQdgA*0Rb>~m4roMbsYA#Vw?
zl4$c;t|jzEkAO9REi4J5%_6olQ;v!QlX3jBK3It_sKL3#equh<U@7oIf8CZWSpYNa
zF*aK&5EP@_S_Hg<e*x`Kj2TT0tX*=mnw5+qlnNm{vlUGnxtUS(#B8qBx7e(1TLWAv
z@8aW8#z-i)a4%9{Ok4t&7PC*KYphmR%<L3-64?*&K!vF4W-~h=b@s<UL*Il$PFpuv
zsXj%5Hx?m)m3zhwwp=Yye?nth9wMybG@rbz!KkFs``I-(aki<oH8`_}TUlrIOx)+e
z`#BMxp03y7G>NSAXU^`w6jC%=f)uqq1X8rQ2}T`8@pZr!wLZd3UMrxSp-DD0%gAuZ
zs49de8L(A#Ii(7vOr;f3)slQ(Y>!YreKrqo3|i!ZJ*LlSq3Ar6f6WplXaR+6vqZx_
z*Q~Qvoqe)d+qR_|0`f*FWtPQc)6zQq#sb%8GtMebWE@rge|coq(sB&8bz-okwh)8<
z?DsAP+wfeh4Z+xkU~F|W$PO6GX|B{I1)$uzT1NnyIDm-H&BiB!QVxJJdpcwV?d-O|
zRcSs7O`FceBt-#Of1StVE$8i;`Z_XCo7qW<j;qY}5~w@clb_t%I|0x9nQGO=jOp)o
zd={R~Gu$DhibJ|Iomt6g+>e^?=LBFz2eHoTA{_lqrTaM@pJ|nKIMJv2VE%qiv{z`e
z*I6CeJJKJ6&mL*y(?~z94gMC)Luy04s}26<O<xK8#hcA4e}SUoqPa>y5&hJYkJFFe
zLd)|HquWQUyZd>=8UTpJHmi8mvxaS#nbgRiB-3K<AXx)#EpiPBgkxTMpu3W@)Ci#c
zX9m$mXp75IM6o<1qO#6K)a}UBUp>1eh8SZWh`d>_0f#8h9xf<wNMhNd=Qzcm;}~rS
zw*W<$W@ZYYUx<-xS+H4qjHi*5qQ7%_hq}n3q%AYmba&q#^DQqg&k6CujE!!<oPs=J
z%WruRTIQshTPm6bcjuMu`xk&err5t?J%B&1F+cs_dC}e8v&_F#=Si3KcL3o301Ovb
zmjMMh69YCmG?!8FHx!pK*e)lR=3y?Ef7|KLkKYbF3Co$&{?=dmr}vM;zlS0^AXC~+
zS{Wv|nEv{4@GqoLOdB@=Fel{n^Y7{V;rE_u=G9#TP*xC7Tx+FfK!IjT*-2|;XiNr5
zk-~}ql5~}%s3fH%0VS{0uH!u?ASEfvuct4Xd#KboKY@Ruar_cNSQPW-<YffFe=)SS
zOrG?Uq>oV=MG`22Uy?M+L@9&R!uK-A)S6^eCbtGnWYi>TQ{Jfb#_RXqQc)1JNEi<C
z0!k*;AnyxsyM+*P#Zc>>qo+}I2?&{tnR-iT*cPLpYlB9RwO|#-Tr!enjLdiNHF6<C
zq2*LYC4`kq7|4WeqF)gY^Q8)1e|w6p=V*KM2H=>-B>ZNxtP{!$b8Z8EgGC9$Bx1x%
z;2ENnfqJ!&Ap0p&e@i&>j9lbVk@ixGETu`kX|#7Hp|@^ui)#UtwLptO($Av8>8D>B
zL+iI+Z(AIB0qx9OpXiFRl;N6ZKu1><kxMgisr~Y7c~=<GKZ&SX>%ndse<KM2=z3wu
zgGLi3@3OYYIcMBe6_N-+B%YkmwM^*UUQ{aD03`HYR3OP}x7;Y>G$#Hj9%pIR@-4Ym
z$Jbn|GL}hQT&qlrCQpLb6hw#g4QH)1LqfWv7;y5A+GJ;wu##TR{1?QZwUGFQvK3*!
z>?i}D$yItgPMd}7d*dKSe{^=7v=r}2rfo<!@gx+Kba&v%Y0infz)5&Uf$Oa0+YZd<
zTmiQn-3{a<(=h1ZQBq6j&9{gb<4D_96C1t4F6R{Y1re=2DydNQ$t`R8ILoZXfxAi0
z-43&HOAjFLFlfQ=UrhiM5b%Lgj4SDfi?#velZ3%n@R#J}NX4GNe_H3SXf;bU@XsL}
zZAtYn@SnA`-XvQXIbrSH{;K{jv1KlpQSz0xJg2xfAWos&-81I%6sdczTNab;Y(DA{
z{^-++0mm(i$#&N9f81gSR%U1u-Vs9Apjph2YD4KNf0gQmRwp%i9%El<bsfb|_MkoY
zM?Gjqr%7!dTph_X6?Swtt<y1wdXVM`>!g(2iLz^}^IEryWiC_%1lEJ)P?yo?EE57b
zHJ9Ns1{0U{^E)Y*G-@uJ0q3{SYA(+Le-^@*>jzawSLz?7h@b;FsP&Ifuh2Q6MLT~-
zH7<bdTnL*++q_d6iHEr5I}mCfu$@`v=i`V;zl1s$lxc=t`SD$pwA^P!8femTgF>CH
zxlkI|%)8Txyjr9>dfrP@*r2QmU|6}Qf7owa;k0(n(YT7iMSjKeABFiyD$`Ete?EFM
z>`x3Z?A)TpBzG_jB~exPlvj)=2(>O7M!%P`$5AGpmpj|wpw!9+j^ap`sc3#Ys{EzQ
zANZG6xNo1r&$J3((2jJT!mE|tI-Zr0kuQOh+3QahrMzI)NXpp};w9^S4szw)m+cvx
zVA@)`dx`W6jZZmr4hR>Wi760zf57DAmhSxgv3ky`LDDfT2N)S?FJ&^bkQ~JbyjqLZ
zJcJJWP~q*A$91hxt<s%$0S8F8SS~SU`8>Pxim&yW0p0j?Ges?>KWg={<sdsjK3F8i
z<OqYNr$u6=5Ra*yA~E@l1>qIMlLI0ks5hXWm*5&ucpMr8EbAquXe+-!f7W<9KlA}W
zFMy!@TP;DF3-P#_H-NCing1u=j1=wjW}F}T8c<Z$b+OfK^5r?4f;x<TP4^|3;z5Jw
zhX}UzPzFSx373o@OdT<`+8ggA?2;l!T6tWdj~w|vPmv`)-6T`)o_e&9KYfj@WtL4d
z!#epqfDyCAP*<sS855jye>D!HbDUjZrWsvFcTQ-ryju@4&2~C<X=|JsSgq&K!|5*V
zZjAzM9oJD6&T`uJ)LG`rqpNYP{9?N^_Jy8a(k150@mw5sJT&I>T+tmP-b;9gAejye
zr(Z>5pu?*PLC7Y*HiG%zItuCZAn*Q!#PgE{*Uspu(}RYN8V;`ge@+*x>-`*;?FX@`
zEi0NDq9!R~lBR7-Fa1&Jk4gjGdq94Ln-lLgwFkqjl0f%W%YJiBBZ7}a2Spa?6ymZd
z!~uxzCOr$_wYj3v4LvD5om0hIZk4Da%u4m8B^OO?dT~Tii)IJ|G=?rk5rcu_EJK`a
zUSTP0KvAvXw6gIIf7~wxv&u-uyT;wu!K_k-L022d;eO!xrKMQaL2ZNASjz`{4|v_(
zX~h*<1P2HW$DRHrv?_&R0t7{?&M{-`%h76R?+7dI7g23h)s+wtX{!$snI=Vb6EDDM
z!T@u~ezNigMk~z#^2DPkO<b2xXbh9Os&lJpbci*_-90|Wf7o1bLiql86RZ>&2~{ig
zcL{=lv~>2X&@`^<aaFSVN{wAwfk<Pb{BPhH`Uht1%(8Ml+?%8A4SJ-$lz5)N-_#_@
z9X%?fOMQ7{Qc10u^iWl@h%pycz;_DoTCuDY^Qq5@&v)q^vRHg{wqu`um4^_zm1m`<
z7<r(Tke*jge?r2Ja+3W;$E8~*NmmiTvt8`_=u$#4t0kSimz;<D+KqMhUh-j{tQwP~
z{gBc;-Ukcvh3VKUeoGj%H;=*tM6^o3Kpkp~7+2hB7}Nb<uMVlca357tawagKQ>sj^
zr$Fl(fJ<*;j`?<Xk9hwO5%?u4!kW)f$Us|5AaEUve>5>9nkRz2X99sLSv@v-#JYH#
zh?tN(MJX>}M@j~*xca4HjBx1j+m328c&(>+OZ~u=BvRP$u1;&6*X*k$=R?-85lVG^
zm3JXIt}(05DZs`Jd(4f33mxfO_qAm%wR1AaA1|j$pK%)9$XKKcLB&#z^lzQs1Hy`t
z?npeue|7<MvmBILO8Bb7g)~f~A&~uw$5F326kwL&QQ^rhd4;5`xxzA5DM+8G)VpeT
zEZ1$sx*BWAEC`~;n~5KyHo#+$Ctr-ea|=EL!Q8rDc$sRoT7q)3exfqy2!bGD{8HLo
zGYzwfiS~4+XNw&2oQW5uIFgD9QW{dh%M=VWe=R$t<sD0TAa*Z(zf*NbrhhWHs{G`{
zhvgI2pkkKw2oTG6y@bwf<>xzh5dKp^`OvkN*CY6%>6cPvOj;i~Ii4ZgLgcOM%0#8>
zY&JDlD$Nw47eebs$XwSh%`}K<B~r~!a;_M_bnbVMd8n&S(%9_mVu1DY{!iv?tFG1u
ze|`boB}q?ZxEn%H*7^-^MRnG0daG(7xm35~kpNMl_4*fcqiVWz7&QjJx1)DLNyLSg
zKDm>QuF_#F^HNQ84E=z}?dws&j%{^tW#6{O*|glNh$Q{^rV7{^gTCDm{cUkr2&P2K
zMO^DM^z|Z2Vd<ACa9!S1yO@0rW$u!+f5#lpo2kL)XbWI@tG*#pnN;$<%TQhKU-dcB
z=0b-K^>D8XD(f$pB&DLSzjXWiE+ch|f~9H7)Gui5BRl_xl+f_8iaxA~FRd01t7CkX
zsX5DSv%(BWV7fIf^17nSu{L6_Q_*ELFldK)h$S>tA8$Q@>Um`J`Y!?H-Z3F4Y$5%t
zKOv9XPn^@dv96K+W#UVvjU~CX|HRlRZR`MP{LQSMbYdu?h4<?gwb7(DT@StohgzG)
zV^3&!g{!_W#j{Iv-K&e)?GiWU$!UG>H5S^l`^AnjlepyQ+DmV1Y?0Dl#r5;;f62Yu
zwU^Q7EE57bH<tmcE)$m|buKD@OONBY5x)1Y&>t8Mi^bvtXc%a_#{=wb2Ym~23bHxv
zA^ZOiDLy4q4@+)!l7lT<B8$a(epQql{x!(q=galu<@+7=V1*;FYWVnL0Dd6(8zkQ}
zG*|#(5QdLGhrhi_sSJOv{LxY(^2hRT$5*<1LvTC-h(G)k;PTCC2y=LU45~C2O`3x|
zO5OMhe0>Y|dA^rM1`0k?xLyAK@!N~c%vR163Nn5$Rtu(8Ib&+tU%n{_YZCt>UXxt7
zqrx4!sPIHHx1_@p{(VdLRlHy4hI!1Fz6Uyyhwpa>i&FAhDAGMI@SdbXEGfdUM%rb#
zYMt)sYo@#`f2Bc?qMovUyfc@GubFpSb{GGdx&wLnR(I0uFZ}|%G8fvUc~K%TrA9W)
zt<*4cZQ>eR@<2_3neCX?%!amkF`qv!i=!@{pX>2=RYaH{MR$oB41E6>r)uWg$Y^OY
z{0ci@mXpohf+Z$9o9Z=^heRVTNom|;ttGmneegnjPu@1QX+R2p>1H;|J`yclurZ<=
zq(Cyx@YQ&!hQHYEm!JIlxo(lgJIDy<u&fkyr;^pj{g`MGQ~&yY_i`%|a1;=Ps!~*0
zN<~>RU*9LnRTD|ueE+)bQDrqi0cfgszbD?KUNPzF(;TdLp47ri-Er`C4-pD%;Kg4k
zcpw$~Gjo%zn)N1sh&Nn*cK?2Sy!?9>)W{g&Jr4i)^CdjcK*Cr*APQ%3`0Jm;A1}Z6
zRI~p2YYgsD$kG_1)<}#XasxRIU&?=8rO64hV5QK!O#O;x+}!y!3Fh8KGwnDD0TkXB
z!@pt!L9hB5!BJ~KD7@a1a?p=XUSdp)C-*cKa~$II5u!+c_T?knP|h(k=Eyd@5ta?G
zn=!_%!#tq$Li?FgizRmJBU`tnnHV35N-g`+%&;s9W&a=Ss$pc0Sv3=s8C{yPo0oJ@
zxFtKaApZa3xcLOHrx~~Rb#5&QZD!nV`G8D2DY6X;<=f8~J0b}Nfcs^Onu8nm+xv{E
zb8=$`>N!|{v2?>(l;BCtt{D7s2wnx#CAy;z6N&)nqIE%}w&dnZM9~B#+eD|iCj6EV
zcIzgWH0nAVuM?unb4*2pX;ejpdHXoQM>b)}ksf2gt(?sOUpVNu_bAO;C)n$|NojBE
zdw(2C>c|Cy&lWOMi&DbLQ7?JbQZd9e-zxT%jQE{@Q^6?t<B<I}8td%t7D^)wo9-9R
zpavgrvITByK!%f&q<$4$3MFS=vz@E@Z-;o($Xalud1f@@Sfcn?qUaHz=W*+z<q=fp
zJSZ}}ux7R|l_OsKs0`n87SV-2?4jDGjcRZLJUo!##7NJQVtr0IEZF%BTeTOTFB&$N
z9lWW3%6G1<B44Edyk_^XUs{f6k380%SdM7p1Zr~>V+3nUrE}i9YUNUMCPger*{D;T
z+*P?~HcwAy6fVoWCcb9uuJF1N8b}GS9k2hT5=Vn(_ceWq!!G7QpI+i<1u@N>ZYHS}
zu_*eCTKA&fibRrTCtxJ@sLKBw(n5Gm-HzUWY-@fx4x~CMEnvCe>O5V_nCTML=FYF`
zvNK(v(HuUhUj=ajx2Rj~@gCMRj}C-WINKAH&oWB%nuFU+5)~dl1NYEbuJLdR&zr;-
z6H-=sX}0D3R~m79GD&Lct>D?Y302ySiBDU9HJc^MrCny47NIKbR`t3TK@Ww`&^lCq
z96mdttr+>5=#jI+Oz@!<jm}LFR^n;4XFY4FE;LxxQWH6s&819UY^QGRURy7m)jj94
zHs+3)f?eoRX;(FCJeM&HXE$O9G#Bj>@L&in$c|lww-IP0<1T&W#vgp<tRQw<N7jCH
zXQG=`3BzA0=e*47g6&L+C~!{Y33sc1|DZLAGwKQDy#T<T%PB06S?!#{B39I#!MoK~
z&n8S;Z`KySYg|e<wz0`+KHaF&jJAB6u2-FL+UtMYc0ms;^O>fPr|~U|NNz~G4h@Kv
z>K!hKdC}#_38UJ$nQO1Um@Xx@>r5mS{cgGGurI=w`56awqCdsdiK{;e{eF9YQ-93w
zK2v8n0HpiX*210!q0Ja-GEYg0_H{JNQvVc?bw*3UD7F&ldH5o^D5ZsDd338~+dM_j
z$_ok9ub|()W<Q>}GOk|@N`HVWi7lYV(m<fTk26{SAynk_3I_TXRKzHry0T#6RplvU
zNytf9^=69G(zqGjy?#9zT@TWK8{RP6dh~k0YYd6gbz=gLJe4Gn`VS2imHnfh5MMlG
z1jmKq>*Vm&)g9(8MuF5Z<lD+|AC~h7vdc0t>zr(MiW~~A8ag$~w+E9b2KH{&MBEug
z=*)OkFN{qbkc**+2?#Z|th%OG+uMynJE?3;8K<OyY<z9sx{jN^k&(@Rd#A(2<}s<J
z^>-azAR6O+bpE3)3UHcpjP6?$PKA<pyF-!c;rK<laBc4V%3X`)O1X3`6s6O_67=3D
zrkzja6m%7HSt0%!7cO_Ehht0~(a%=7jnXUKhTK>j!{<u3#TfyrCm3E73PeNU8-8PH
zP&}5=Ljj={4$J`ot+HN!gwaym3zeTZlk8Sfa`Qm)FkgEbSQUn@H?Z2MVllLQIVmGM
zrG2R2BSO&(tXxldrAt#q*pA*)7FTFuk(q$}6?tp!Opj$X`@MsHwLreSA5>#kmoDV-
ze%1&@l+ZJQF#Wm9@81eaP6rwnx1z?kvO2a$6wF@7_@<wIpTBN@b-V?st+3ASGyo(O
z+jG4ElvmmAZGxOsS%-LHG>E8n>ELnLU5jN}qL6+j)uwQCyJ8qSH<$lvK|hTXmq-Un
zI6q?c&|CU(T4V;6op*3zWbP=^as%g3&<QPKLp|k{NE}9?)&1zDH!Ak_{TrD-(^>|y
z3`w9#-SzZlyE~bGn<r3S4;X>qq#iv8m<-oH>VJrr0NGUNtMO95%BS&CNAL^r5`Y#O
z&%{d|#WTIKrToY%+tF)Yc>+{X$YmNQ*lxv@Och?>*2>QKI~pB6uY1N51nxxDw*-%O
zm~xzKG?k81@1MeHEniQ-wElI@j^GoV<+nT+INO1`ARJ&q%(%FnW*AnAGb6&3-1$cF
zd8Dbx39YH&bIo~8NL}rgwlU_!Jf25{1*JEVJPrp@DWMEBWHt}7`zz=8{{UF6ijJ4j
z=PVNeIhS$!2NRd2elEU$*$TCPzW;X8*;%cO*W??m$XIXXZiB|=r#xML^7bmT^RPHO
zBxP)9tYanql-U{a&WNXCP8;!n<$v}QFP^SJZ{cOxE5N~Gv?lX25G`e^u}yc+*mdoE
zO{1*(TdZSHZWN8Lov#vz6b*n6jJOhiZ&V2Nh%eU|Cvb7>&>DJw&R9W*m5H`5*#}4*
zNZ6@hBYg4pTK)I!{q*mt4vPt>*g^Bxk5hcWV3c*CK~w>_`T2MA{q%cnwXmeRCyFOf
zwYA1fNnn*RUPklE@drXC4d+OjJV_R_{ZFFhq($gBfD0*{)yVKQUK6_YJ%W(iaor=P
zA)8EGmwn|S7&SqEV)`XgbLW4&$3FV96sVr^D-<cwvM32i#V3&h=I4x^E6LO*aWGKm
zlc;h|>6L|JfuQBdI3kyPf9apjEiKM+Gr`bWWkoZpx#ZZcA%G4P;S7PE_QwmBVOBz%
z<MhfvBvfyZ22P~A9^yF3g)pES2hW4<$(Qt2#JnkAk9x3w5-H6pXs$f?BspVCN_+k!
zIb$0X*~cTA)$M%0x`{@4$%d&T8{&pS4+>ayh-Cr$N{269WGIYgb;+KaobA3VFOr6Y
zHdY`H$r}?X2=-xln-g{<Z{p~poVmpiwqM}64&#ozY2^jlYA!jrt4U*xAV77moCY(T
zgw|+#5yXstOPYac5sQ$g<*Ljr1utxihp|$D*xLS%?;trBVm3^BrF6DE)}>x@y+7PP
z+fAG{z&aWfgww;CXfq%EFi@sKlJHDVG~Q#(rzVEWP?u-3)~$8I1f>Jj!`~nRqm-Gh
zA1mAtTs_qD)Yrpif++6*R0DMJ%aXMpp_^vnoHD|HuU|J4FVzA1Y5-$L-r~ax-`*_A
zp)7ngF*13du97Jm5=c&jIk-t`p9eRS>sfjGYWU7I5@nwe^R-$dft+H$w@4}-SE2M{
z^5QgUdCLoDDuq@TPKrV!Q7u%<)k&IJUSkj{uveC1u|yPViz{2OkL(;O%cwVZ4b|JU
zvXjw&GE4ZDecFd~7f)jGmG8h)>R>14ff|SdrBNp>Z#Q5*9TjOvndNF>1v#m+aW;ok
z5iOx7dljo8EgNI}OonX0ue5C7hr)SqG{Z0Ld~V_&vLsWKw5f$#mH~Z#y@zqPK7$I(
za;+EQYR)rYFQY0=<aGGywoBudaXJ*LiHZb&J?x9>lcNL3tU(3uj}(Wg#fJ-ZdI&)b
zl3=ox?W5Oi$KZG*QL20ZI1tIo;xQ|0TT1L6-22rU4NKu%``2T;EIsikB1n*`UVUs2
zhyaVg1UfPTQ;YvMm&qz5-obmVmN@><MrNw?k%!FSgjXgk!ClB7;bl9VUCd4h^EsM-
zY<+GyxdCI88*T1(PHsNz#oFmjNFO1lUVPH&4S<)tqE9-#0c0uFj)`oSOzHhj&XO5Z
z_Tl%-nfS&!R6Z^o?8MY4l7uGLapRWF>L&tQ1A$>HxVIHWSy!iqyUGpAaY`1SkATxv
zLBA9m@k~mFIsmxf_zo5%oYvabSUhrHOy<3^&{tqIJH?8B!Sg`P#KTK>J_LCanab)x
z5=zg|Gd7>i^auO>CxTlOB8_BPEQ@>pd7u>YZt2u5rC$Hx;pT0#lgJKV1+XlLFPX}E
zINwa}+9^-+T{P)G)V2MWv4cVr12#E1m!KjI6_?1F7$%pBgf5nUz*wIC_4?aE7Gn8c
zDO5&yrx)wAAe@E5lQaWVovn?~-c7R;Et>ezK0&>E@vtcki^6z2Lu-tXQW{e(jQ4_R
zN3KB$t@R2M@oXX{l7}W4nMl{UV1y`J_gF?OOWKh9wT>hCij-)_Zxdaiqc6f9ncGAs
zLYl|e<C+G0>^vubQ4#X`+VwEn6l|{0oH6$_n5T@LaN%Q>5w9uBEarju(MKNJ)+FoQ
zc;Uh9=zUj{YzG5$Uhh=7zNkTKg%7a8U<+@Rqbh|kaW8v`UtxWyKp?JYM%AqfMTRbG
zJIV1}#@$7J&>P`!#ZTfxmldiVJY^pzi`1wF|JkyrWS!f8c6MCZ8g^E2E|wMVFNKrh
z5o|5+Jn7r`O>G$?B6Au=Ft#vve6iOVi`4?aP0%+8o@98+qKtEOagfN8e0h4X5>}>1
zZ_YO{hCwLn?qzI%eq+W`EYr;+)O)sw?Bhw(@(NMxdLsKWC}E_pT%3oP$#FSZYV}Fm
zVkcV^(Q(m#3ndDt&Ai$Tkpq#WzD7`k`V<n_Ib<*Vo&gRw)qS9llbxR7qk=+xTH(J|
z;pFt0xgC2%;>w1F@`TV>3x=5RQ({vcou(y=l(2N7%$6io)=TO+T{Zw-(~*Bds0Q$y
z*QDjBNU4Uvlu$ylT|#OS%LZ)DRTmd%@6zyEqM=BC{iR^Ul6nhw`gqyUh#QSKb9IZM
zQIX-%&KMF+QM1oQ_N?u7IGvNN(Ls2n?o75Lbe790%P>1W&<M)-l#-;W@5fbktbl1$
zT3EQ3#2$EqLK58@PzV%my}y^l0eXu@cncaO1u1LqrE!2J8kw4=Vd#Q#OdQeQ<~fOe
ze-qt*+Itc7y%st`$F&O;O{V*Fk8;;9x6$oXD!=Zedw^cs#L1n1(jGLGg*x18vm^9|
z&2m=zIU2PR=<c=I5qeFdNBy((p%>EVd+8gXrFpJ)7h@9Ca_jx|x!lUz=m(_ueYcGc
z(5dFjJix#M-p9j_Y6Im%D%)7QdZ@OJt#FNhv8v>6J#63f>b)7a#%YEdi^`$C$!mqK
z{5_Z7&}HE3>MKP#9b2SAt_oaN$=JSmP_6Zck}{ekg&VDr2LycV>X*<o(rd`{HHGgh
ztcszu3ld(&<}O!5MBSRW+WRnF?*rdtKTMcKpfo$J$YnZ0+0Eu6auU5Rg==E;C6C8{
z-Lmqo6vK(qlG+3s?{T4mnJ%x4TK*uHT91C(cQWOO%3$ACK4orJKClnCt73+Bv7N9|
z7+X7Ic>~sUK&my&FJYZqQ7@VOD^of<lBS)aw?l7l=z8&fnYTII=jg(K6y!n5lNgk-
zgiDeWNX$aod~-oil`b>_ee3YBKuU*yFflzBT|r*z4r2(ydi`reU**2b7gOPZ`10G^
zzbMh4ju-HFfw-MLn4+q}x{0{hpeaUx;mN4SOKsoI7|6q;=IQofdV2qQ`28?z8fX))
zNH3qiAL0uQMpzdX=Y*4C`Qty!m&3;y>(pkK{TV@$zqHnv(XOm8l&#Q!(8j!fy%iMa
zD~+Yx3ux5^vFb?qnxaLpSG445o?P0$<VwTHr*7YCg8=fX2|GNmcFNKH2wch(y)xEe
z<WqH{wE#&W7GPclFCPYM)Ltvh7>{kVoiUdd*Gh&Mgy{|=&P8$9ju2{I$5z3Z=UU_3
zEZ^XT_iA4Gjpd6JAXLD46=b`Az>UzpBTv-CL6|WWdEUf1%N#u2{jahtgF;dOm{w7t
z=Cy0Mj%{ME-Hatx=}CSnR^j_rC8KqRp#rF_-#a{#5?@d=Rsg=X`&YC-rZhQ2%@mhF
z2EmHfyx>OsW|;g@4G>QZrE^11y0mJ!F03a-W4qRM<*{9zla;r*3DLtEw9}L!miFf|
z-0N1n-cG}Lj^C|qx!F?I`eL`RiY~zd1B<3#mywAV69YFeF_!_WE)$m%j|C`~#f>hU
ze?0&F{jU$QFw4)%p*F%hT&zPuGz*8Xavb&H&BLaAXassUjh0A!BFSS+`Kb6z6n8b{
zwD1^aO7QXd5hX^N`246`h~G#u6RDHbNSZLD2|SfqfP1CZ*ElYIkFEYGzr``>6KUw&
z)7VNK3!mE(6B%LG66?Z8ls54ad3!mOe^|>PlwL`k_>H40&nyXPP(>spIbWFPz6bg$
z{Wd>U#!+I{ThdUn+cVgwJfq?niIRMyUf!}E-&p5NJMr>TW=uQ{{KUSp=Mr6+uUMpV
zOI87yw6{W<RYH2j+N!n>nXU%+)JYsF_Od*Z@h~DLkBV;?cm?)4EiJO)r)j3ne;FnV
z$x8R7P&yUdy)wnDr%L~&S!U#6WqC-Rp<kh^^^8^k!maCh7@hgFd0l5sx0T1(NPRw*
z-`pYcm9!#6<d3tYwXSqo>qSHCJygb9qK<4e)={>C?=9~-eY3@V--|4Lw9d6O!ueT`
znHPP$uE7bX*CTDQrOMtbN#iPdf6Tz28Eu|W=Q?RD<Hpt1S@i*>#U0~AOQCK>M>qBo
zjetQBNE5@jX6<>UfqpHRYI3q?Vhm(IQlxDwXm14{dUWCrjheWFqUmUt>}wAZ?WHh&
z&}x<fSC+nsN3BTncG7&z7|d&lVaL%`6BfOjEYMa??=y|7F(Bb@jFiWie*zTBgj-QH
zjGnZgEdtH&?dK6*7`LLfiMAt+LkQ%fU`6LjTqrA~)gzU=B7@RGT3cr3^9pPzAE*)6
zl9o}%Z|X;BVQg3ojSi-IIc*ABT*=W)LDjLIDm6#Z9ekzcTI9@8(Or=LwP<bp>_aN5
z2OgiZ{vqMy2;m0Fx64}*e?E>*wqY!pvdU8CTyiDOj^JbdTw_xO0u;<M##LaY>T*7m
zNu$1KpiysHvW2@&!zk@j<Fk4V?BZ{ibCx+uUTaR=!ZKGo==w{EXed`Ud+D+bH_y{^
zg(cz_ic>XETA^s&=l&66t$*~5nFEf_MH1Wc-<p%S&1}^WZXB&GfA(o*b#!@Y$|BGC
zBC3rPt+5SiDU&#{CmHdR*Iic!VDIH8<nF9X#yyaa0DUX&htX&Gz6b8d86)nsgTB^*
zqW4@EbhuXsxX+_jm1U+ip`v=corq2xQZ0Y2ZjNnAF7NeFKPW*^zj@Z7KHugSK@#@-
zz4?vt^c>H@2?J|Qf4P>Ia3+XuRei6p7<XihA0&t4)n~|uQVgF)-n%9u=@giGmJ)^M
zIEkK~rr&(XZmfn#F3*b@IINKy@l-rZf%Ng(=;aSu!t2)G4i+^P;D%Dkk%e=y>>LhK
z-<<96GR4bWl(HZw^B&oeZ+DYdSxTL?nkKlT?~Rp%)Sg(;f5v)CPaCAljBft&+vhbN
ztPvJmUu$xmnP#^&#<W?3jbGzqe*5(P@cU{C9TQDb&DWm~@dpH&(S^k;Vru#0pUbz0
zFEi18e>5pqS{qf^=Y5ux>K8;U$kE>wP01f;G!>Uz98GfxB$+b{YDzj$U$A*eh>=VC
zFPUg)=}lvUe*h9`qMn1=WD3%tB>59EGu5Fbp)<%$1v*EDhj|utoFZ=o?-gYgx2)05
zl*`Im>5u}i^ffiP!XaLzQ*SA}&y(F-r2&Oiex5}QJ5krEaHLe7(-nfVnkfo{GQUc4
zqY|IfEIM#NxkzfTNDkS}RLm#dE0U5>u?HkgbvIaBe;4Ch*)_BKBfJ+%Hs5FVM-*?P
zY0)UGF;<?qDyloR>^FV<1y<Y_`)xV9a}}<}5mn(iz|&MunC_h76NPl2_R3%Tvw?0W
z%+wyC+Y1bm>kZC_*q@&GSC*$ri{oO=yDp>wlzlDE`}LmZO?z>tydRP@?}qt(eM_8_
z-M!b#f4QjR-^x6np0zn&HKkSORtneGB-3RZp({#K9C>3{G<a4Bwe8TA=@mg+g-xKI
zj<-ip33P6z5cxRV0-laNaO{6h>3P%jI;AT_p($CPIK8jdnfustauD2O(Cu0M>R{6L
z?Zov?@~z3OL#QrFo%%zhdSKy(gPv~Vo)=KIe_lQK@ta#z^LEFO;h!lnm9|0>pA)|$
zX5&mCD|*e0XDrh7=-$T-s`j3~cUwHkTAR7mrfrWchRO)=Fw3@MF%`5>R&UfxcdLq!
z%43)(x?NqDl|F!i{OLUF5C<|?Tr&eNH!ctP9Px%KWLFuic4vgvR}>{BG1<LFp)d$k
zfA_CBhk|ogDrQ1cKv6BAe9R(6%*MMMaYy6Nah7YAJ?3_xpd$p_#N5kU-882}!=&Sn
z-KwC2!j+{V5dNISxinNt1i<6kaDHA;D-yF*ko?k(ha+cob?&Z^)Zt;``_8F*Jj8@B
z&2f(^FecZ^PK0l$U{tW_RQ#7}As*T*f2vi-Nv^D4C<@bDv&gXk7}sMhV>lOF9}8Zi
zAZ>%7x_MU6c<MQyOKA_nt6AU@J8vscz-r;$nBB;-&W;b?HxKJ-U)QcbpgX?2apnk`
zDX%Z+%)X#I9NM}dFkU3tIs`j94CpTRoL2eX<zXe3DZ%?$?^pPK>LNVZc%o1se?uO7
zHnG>_D|B?%zb}M~&r1Ra;Y9~;mj5lW8Di37Erk?nMj+op=zaL`pYPxP^7S7SW0vp#
zjU&k0$D^s>m!J975`*ON%<?0@J1K}=f|UrAcM<=Jin6%kcwBAxx;)T40<}_Iw;!>U
z!i$%t>zhK1M?yZ+Z!MLBM37r~f4U!iBQw=zND@^p;#;Ir#>Q6~&!ggN{PgeKl}7z*
zdHA+Ge5u+Dn*K#+=E%8hD3c}yr=eG9ZHjjh&=u~fgN=K>n~lAgrA?8vmb@YY1?1pd
zG)AFYCFPb*T5cjzOo5yDIk)dfKM9)kQFfe!iFXJnceW9aP*j@L+P;7Ne_4ior$L{)
zXD0IqnhH0f^avQO5hAUWO6*)-DBZa?2a1<c<6^tGusTl4GqX4CSJy<8SopGZEGAJO
zo2*|hRSy}-L^z*S&m0|^S^^4(zG`Lfu(t?60noN1sCxYBb`RRh;p*U#oqD=B;Vr%A
z3zonaEZfnot8=aLzDk98f0F|z9F>M+;PP9E=O%PpY_{%FGu@qHZiopT^?q#Kyt4zB
zRaY!g-8t}GBs_JyH+3xOWvF!7c9PO8sfR<#A)@B;aEPvv11Yy+nN(9;<U`N->V-}j
z&{zs_YwCX*uZcujacdDFZXjwsUG8Li=Df;vhpo1|)MJ^$ZN8@qe^h_wm`yt^g29z`
zS*xM#QjI97GT8pgz;vNEUl!g<?<k51Bh6~w{c9wwb6t)nP20ia3#fYs<OM;+RZO?h
z)2=bNjr%oI@`uBX&9lkl)@|ouhq>LnUL}#^{>8pSZq9GRA2;Wk{qet-u~$i%9$U_I
z6jw@bOJ#^{f9=s14F=m9*!l|%lH~t<t}&O9i5C+CH!w1nAw(t=mq@EUCzq_4E|~(N
zPq#;zE(Hh#ZXiMTm(k}e69YLjHkXmE3N8UNm%)|+CzqgSDFv4x20jUY!Mx;f$Qc%a
z04J`~6bfosE|63Z3Vo~+rLtC7tpc~dzGt|5cO|W4UG2)YY{0&=yhCz^9G*GD>rzQw
zaHUdpOn9c?l<gT^q+$ZOSkz^X>L{9ON{%vGYc7szQ)ZV^aam(=NpuUtvEx|riUYdZ
zz?@xSHDz(7G(nw)Xs<ATSiq=|5+rGD>>~b{T7{JDZAKeu#qcG#m})jXagk>eC=3-d
zhtZ=3n>`0dp}ox!*0;B2bqY-qNM=sW@GVK}Q5`YGEqF(eN}Jal4mO)$yuxOcU=-oj
z7kV0PjzTVlY;EY4V1h|l8cw#5CA*ZRwb?2KGC#y6O=gRrZ*9weBFS+G)&u5pXlo`J
z62^;F3l=uD*|1|Xj}$#p+Zw1a$2OZn5}QNBu|)Gj(AM^$!qptaMpo<(HS0CHoZ4H>
zE@y3R*whlymJ}gSnr|wQsttf{SvX=2dn;;lC|)SFQSera^{6l?gybaHcE`3e)ugt5
z6%&y*>oEW~OD(E@O0%WHNDYEHFlAA+VH7B<nwKpE5Wy1Qz62~bB>WS_e%h3TAclmt
zgsp{y0k}3>@It7KCr&i$79<PFYJK6GT9iXdkhJNcEs>a1Ar%{s&}OkA0VhTaA{IR|
zq+)v53f0iY6Q?*im;>jihS}TdaPHbD!6C}bDjXcdmc504Y1kbb87@@i)+Ug&^}r~0
zNFiYq3mRB}0KpRHt#7~WUhQFFGLH7!?&n|rWv9w3BCu%ow$A5QXJ`A{+aJ@E=q+p>
z*37y1g54gV$>cc<8IwtTk|ARvV0;+l5moh}3>lL&c?;)<HDge3@%^x7Wi01)Mb~{l
zJ{wPVFAk1>MlPAY>)sqpr=!WaquI~DewhBSJ3W|=T%0w$9-mJ!#cMzYX681@Yr};c
zboN6Hx48|sz1_W;jE{Cl)4lG^tJmGT(TC~W#&`d{7-6zsN3h|1Iy#?T8f>;ny0@dt
z@zvyLbUDLbYx-$)e0uQR_`_bXT!Et@mG;qpf}YTS3Nx*j=jY?uKzTO()?Pz<v%H}_
zg#E?rO}pE@I-J_CozwIGxbD00<ajh`vxNQbZ{3gG_j_^n<L&5Zx<`bVY#72`gJG<C
zGH@mKoMgXy=`j3m_rv(zxO?TkDL8<@M9K=A&>83W&)eJAmY^AN_q<qC7w7XvBA*SH
z@l>3D_?XEX>@ZcIDD7|E@S}UV82)H_IzER%|9tzS{r%>*>Gb0AFJ1TR>GZd&Lw_`W
z-~E2|{^G^)=#TF2S3kY+zn^u7XXC@}<U~t7Iw94=7*Aq4=5)eoBqKRImSd)qaKiPZ
z#_n=*)V)7Anv5^e_fI>l{9*^t4mY=!;mF~CS~wPjBcPAs?V!D()gP|ld4L{rvCr7*
zPY~VL0=j#RJRKv|mAHRwsOP*A>N%|w>Ln~go%>K{?S(oQa_EJ4$0-;dfElm|cCrXS
z<P1FN83kZ0>BSB>*D?Ut@iUd+8?~GJnF{=7{nCDpFGiDt>3A}8m71%P?sE09z^3ef
z3n)=EKfxgqtB(M-DB`ibw{a~7ECwe~vjmc-IVUQ20NW#tNG19VR@588F?+6ZpAi%k
zM;`Pj3z%U9@s-VOpeo)DT>_qikH+HRB)FDGN?x%bry!}|Pzrb}I8daxu)u&S^8%<)
zQ`~`g*Ax~M#Zt>E@wiJf1Ru2w(6rruOdzrYH0?a6_&Hg^4N!e7j>;-!C~Hv$lv4{F
z%`^wgvJg@)a4+OuD7_#XA#2Im3#{l7rxfxe^5qd{diwtKf1}C2PqFBe6GTD79u%uD
zb~xx0U{CY7SZ+Z)qgW&`1dGiKawA38Jok?vRTWRvGz(HyD+v8_U3l;(puSyyQalCh
zqaul@3@yD0j(tO5yF~!vqKf`ZxDR>^R&BQvb_DRm#-M~v1)NUv>i*-@CkjGH(28oV
zP~F^yXr7B*BdIR{YNG5>ZX0qU5qusSh)^G*CSTZ4qvv*ibGH!@;h#%VYzh@FDHaYl
z@{bIn-Lguj3nTxj4ACUBtq#zCWDP!rA+s|!-mX4AHJQJWkv2#|^TDqtA*FSaP`yUx
zOV8n@XZ5mZ=~=z@tX?-`^+kYuX_<PYws?D{2w0GK%4YOFBfc{)J<d6JoT4x7@w}&j
zJd=U^-F`kgzXDc3qf=!ZYH|cptI;lW*{8&-td<HTpnO4U)fhl*o`P3@$BIdLX@n>W
zcDmPy2!)tq2AxYlBcUAh39QiAX6>y=;Q5I^Vgzff5xAs9fJ&J(Z$vdB@~yGTKJF-F
zL-;QpcYhpAP7e;xMtubxB7(2btxFtD4^%VP6c*_Lb#z$0n!QVtv$wAp4{M_8LamyI
zXd=>?hG?RW0udUdDbXu`^H-Ccs;nfZnjfE>+NYC6a;kBjcxqoE)xJWiy!kb&9ZRVo
zLK$m~0a<dpI5&EVb9fh6;6zi&6}1DK7&gZW2K$O-X#@sucLMhrp+p=(TM!&l>@$+m
zMp%)EQ4Z<y3d+J>>NC1(OKPb&>Uz028R@XV4%FB`XZV|v1-b`+y<W>toH>8`9^Rdu
z9{l5Cy72cVkkvAT7T?gC-;TuS3lC7FiV`;IS8^SVTKe`iO8dgN21rEiL85*AG&kDE
z&$*C$q4a|E!V`R*>1#NvGz7&Pc#PL1DCmRFnLiI~U^q-#7Dgm1rGQ=c8iAw-T7Zf{
z*gjkrCEn~hqp;6^;o5Zu5*9Z28R3RVda<LSJ_WZ;)eC}-u)2)=O#qJ)0iGdvl<=z`
z<f469uMCcpvFHZY8!a|A4f8Au39?-$enyL}R!L~y7^K+;3$6u7CtphbWFX!CKgGw`
z<_(1f>;eFi<RO|Y9_7tf1JIOK0Gi@D0G;9zpkoimv4`V-*u!zuUZ`_{!V`QxNBJWJ
zEM!!qz{pgqx7!oW*cH_^Db>1JB4Bck%PfHkRciV?Ml7<#+=xY%xW|Ym&k|Mv6D>b*
zaxmc1=kz}bDA`31ZFbW`*>1L%pH&Fik&-+kKd}p-(S*#gnWvGE@_1P<L@KCLo>RR*
z95OuR7;Zu_J<-E_{Fc_7k1I+sN<%aY2)Xb8P2!9Ip&^<m$XJKWZcpJMv!lkW%d3t5
z1G<GK-Ivi!FB1YVGnWy34HTCZt1iF+?zh&fF8l$1Opb+{f>yq)7QyLe%O6=vB9x+a
z^a{^0n;Bd1yTR-_G|(VHgEM=&+PXI}d>Hk<7d|?)983zdi{U@;q`*1hD2BZ{DKJWj
z5MGayY=%|U#}p!|rtzXl@~CdofSfbA$Ut&v#a-&-Q5DF5UOP^CC#ph>SjO@hF$1jb
zz9v3@NpPF7JMu~Yf95QbUe|13AuS+9IUH1vy{mm(+L7`-(vCB*c+*7~DF36IrBCla
z_cWBCa4RKI1u7}<4$7;)k0&3>2ohx>oLo;jqFI#UqZ9J4GwRbW$NimB-FE_lubyGf
zk|8J2W(7C$vW*uB0_H%K7;<%?2e3lHDkx5Wbq_0+f-yxiHbXstQ}SgNfK%Eq3{W-*
zi)GD^t^&cV(1?>A1roFb7~Kb1{aGxiisfYa-!3w&!$JlvEQNHo>Z?l|sVGBod=)fZ
zQhQ~q-Iigq+$n_=Urn+V65~c@5z-37lRbG<41No(=`imwvz82Pm1mUVE!gGoD3MTq
zJHj`GQ(kQxFu9_XmY~Wgz1ZZSf=~#2mC+~Pr5-R=C>1y*^I0T+g2&=rIFGu-w#$9|
zmFUDR!$ZO&^`O?v_<kvMg)e1|Nkv<%6iq63Sk&v2y;hc8_OsyNg`)z2t-cZaH`7x@
z%I!6$603G)Q|WjrrC8xspQ{WtRHYSvZF_nnICmp!L2LYGy2z>4FqkP-KGTR=0GA+{
z>z!O|Jz;^m{;8|(Z7HNPozD+t>bjRIpKJY&%2j>JO4|jN6i~|0@Oh+=Mr$jlm%2`U
zns|ZsU)Qs_FAZ-qPot)5=1^J?G|Kcc4y*P}i1Nf1JE;;j%bJO@>*v)JsAYqHdM=iV
zCb#y9@{V>sE;PH$c{ZziX$4x~6k^YpUcF}z!X#VZwQ10{Zcy$4kv8#tMs%xhV8&=|
z%*@)2`=-$Ib3U!l>zD%adD!;$h}eirsF+CDMcy#VPx}2;_)Kf}`zO>H@EFQKV^OTZ
zC>%Zj)XeTEPELzKopCNIBEZUjTfHxOGeZ@11YaFxI_FagnA!brD0S26;RcAxo7$V>
zbXUzUodA2#r4YP)bXTdks2%<NjRu}FI&JxSsU3sama!#+ce*sSuAALp=DY}gA+fwv
zX`1Dde4aa{`~*6zanLuttc2MzUzfm<$05;$(%ngTsV>fSt3c!L2GYuZ3YR)(U4q`l
zxMMD64L>RvF?){+hPZ5LOMSmm%Da;vDXP)PGz}0?mUk^9xTnVoRDyfTR0#8wp4B$f
zS(^1qPaC1VW1|xYY2)m|*ia^9k<UC>;*~QGDre|d-J~g}=r_^^j5U&^o^z0Lo!tCC
z-d^u#xN+o4>;Ik{Zfw+l>w4Va*D{|B!)ABZ^xw0FR#f_Wv#`doESJQmYHERonkyx_
zX6>5(pd`INb=>#akJ~(VYG_qdjI?h2O|%Z9521CO`7ByDisbDkw@io6jEmmg7TxOA
z11Q(w?Ani54UYNkYGbUOxb_IMBgJ|UQ*hopV>j$FJy7yXhdQ5s>bX@k`30Klei^tp
zjZOY7Bq3#=-rb9&VRS{(uoDNZuu%5A_STOapSEXUJjnI+lKqBO>tKP(fnPartWTCF
z2KCLUOBt8Ehc3Nb)1vPZs{3Oo1}mJK<5)RmPkhL(dh(AQ6rn@`w71zok;mAk**_=B
zJ)GK<n>%#%$<K0scC^vmDIGJ6C#ZIA(v`EkmM)Q;2!}dRjN&U!fIOsgM@D(8*BoT(
ztgz8*Cp<C9j?HGM-Ll&%y{S)%bg1nCeM+)8kq0$FYqbtfe%_(;an*yT>%85~sBNBe
zJa9t>y|82xwgQw>WAKOD+;i?mJIBL<@^up@_1+JgVa^YK-}||}_93b+2uHaSr>fVb
z@k#`@kpI=_R<yUeJmUr4Vx;(*!|Qyr2lo!IZ$SEq)0`ME$fGmC9%$_=P{JAHloId9
zL8_t&!7TWaD-P+%`D~yPb797|q1Th8SPqbV7Lr}MBOBUJ^PhY#r+?CihYC&zLUkqQ
z+5vYvyy!1~$i+*9DTTnfJ(;eD<^zgaNU{cReoEW1rtZEg08L(Fab$Ers(A8EaR4Q`
zDd@aYX4*SwEX>}sJSj%y@9o}A*O|qaUyP#1gNMC+GwN_2<vo2ZuG;|ZMA%QG?zmn9
zI`o&5|G)QZKv4w!B{xnO<yW70mUBS1$}8^gN+(o*Ku&q%0s)|-sH%R!{Uk>DjvG5{
z#xwVmN{nAf3lxmHBly?cH(^xoxNpK{JaONIQQE(6!UjKZ-vp!)E|~xEeG|6oFS>8S
zR{6#EP55fBzi+};c<#OlBerqhgzV+jaNg173g>Pc`<iRan|9~o1OKLw_g@g_U8f3V
zZe(+p$NdEj0yi+1L4qI?ms_$WDVGnmE|&p#m#4KZDSu1_>X%D2i+p>he3ZbQ&EV+p
zIfJ7oF^CuXL5~h&zvg_u2YGMVlg|*#?%eOsWF7h=1uXQSU0}MF^#L&`B$6dzBs)AY
zG<O6jkOxW3Ws>)hyh#>?ph8&2ZiZdw<7<98Usp1Pm9fHHW;BfOXM01XC9lB2*i+@L
zV4R&|-GAU4)1{dQGuwIQl}a&-d8B|fI`W%k*jS1Or6W|CA@knKq0*L>mF+o9c<ilv
z$|J8-srU($P~TqXwRlRec^4NgvT}i_JjTJ^GlUak(5$ITp{#g~N<KSIe*7Z-_x|zn
zZzCiJQ6&?)KmU2jFVK)5ghX+Gu<F?T_^bQ=@_(BI%dV-FrZfE?r@J42cVgum$C{*x
zL+!nDa}K?7#JmS6F{kS3YcH?rM43}CvzWkM+(L9-S(_xpA}3%AD3DoJnCYy43i-2@
zZ<FW~O@SyNgI<nRNh=#r?zRmbv_QJl#HD#>ny!oVsysF63BwL~YSKA@H?}hwQNcP}
z`G0nU%;#7k$G9|lR>fRG<iuVZ$&uYb!$X#mTaw;>O#&Ib_b4$Kn?170KD!`e&LO6O
zd|9;<a8V&TdBUCcbLG8C&fb7~7u1kcpx_{(;xaeUsuUXjNk^|<!(kH7ahCh`P(<OC
z?9n&Yc0Z3cdyR1|k8BTqkyo;KMbc?_eSfScJCeVK^uWj{AU>wR=?fc9`fY);;lAgo
zs2odY6jfc>5EUe^0@Y4ZRGBciNi<d`g{cv?OP{9A1gos`o%3Y0qjy&!1+EKRsl{#~
z6||+093;TTtymjrRA@1q$eKtGk^Ep%E+}J$R-u+S7=%kwbQ=E@OtY1(6jR*+y?-*a
zwK}~h=-hh%l0a?0-=LHN>s$j*X+;uC0a!|b+YNepJ%p#X(WvoMS1H)>zox}oUo6tw
zue-S%xU!B|C1ui+(%vNa_f*HATN;Rg?DBQpE(WQ!YK@G*Mp`DDj6SFMq=9K%dNXQd
zwKydy(91NlL>SE)eUs?fnpPW63{F{nQwM)Gv-|2ye!I=9SjcwMXtTh9H5X`>WqLQ}
zmBu<bg%a`G#n*lV($U$&R<|~E47+Q5B3UI8h^X<YI86qxl=$LV%K2?CMF@%toBK7m
zmM``fMWk(;KvHDBHhw#tAkiebKMyK~s1_&XtY0qasQ}=(GlY6SChmKRo?K+abOV1M
z*>h_M2gil{bY&x+*E}e>KM(Es-V$py<8sL*F@2(l%dr%&kq~YcZ({e56PFxe)vwvd
z$j(%PAvalfL_Ekuf9yvG6u$w}IJn4~i4iha^ozD;j%IZ$B2z12U1;dgJcK*<QOzz|
zWIv!=@<i~}zIl2jg89nS${~eWNlSnAky1?hi66Of<b!WjHq(&{IDJ@+y7euOgjP?p
zr`Dkqueu}Du^jsOFp429#;+MoFxFUdz{W5tZT*>L&MK5NjP>|!j)utk%Wv|N8{ftX
zV>a6{_KuhyW@k21qbU(JX0{Ua?Rw(8YEy<HNtB2OKw)lAgDzt6>*-j|Pvn2?$3|)x
zG=`dzWj(k$3Y^%Gi1c-sSgpI@;PLcnvr@4!58$xT=z2z36C4`yLaH+bBrYZz3-a`|
zLp5sam@`fXqXjF=ux6k~*p>(7f{`@1m6(Mv8NbHGEVXV-lmRHd3rs_S_4rEJ=HBIj
zbuxLIa=2V&nTdQ*l#;cn2ljuH>NBAqRO9V6wADb}7~EV#@20BHI7@`BC>0`!M9}q<
zO6)v;14?y2r^JQ;RH~)NEB|N?V+C&Ct8DHbvI;^UT(N=;$^<L1lnX|9ShY$iTqWxp
zEZZYpsiLZeF!B668OxoJ9{PpP_dIn#dR_9e8khX-?0P;`Jp)x0#*cql$8wIG8TFWf
zhJ0SfkiNc;lWJ=zUl*vG*Ao5Qh~ZBz_o@cVDV#go<OIsAK*!FOhquFp;mje2v+MC#
z@`6)-`LdB~kU>g*>^j0>dQChD5|`<fXRZ0iDku)M%SQFH)B!&)j{@brVE1Iw-Lc$I
z+v3|BQD36KW+=ZQben$^XIDM{+t4iteNyNaSe_KRC6?{bEmichL-%#b%g`<P+Z(#?
zpf!4>LwK^4*n|`)8i`o`tfqYX5O2CU!TddwZmhao8Kis((6@=K%h}g8eX!#zf?@wB
zk{u;F<qbR|*=?tgAe&X>hUz>|4`T~F`KD@?;O<CnSH47AOWl7$&#UcnFx1dlDO*UJ
z9fJ*I#UKwoF?$)zVJ|*2dl4@dP-ZU@PtKa3&R*b@P421LYd)4h;wn&tGpLXvdG(q5
z^IN^P6c8UAxw1(`iZcSsgY%mSfehC)SXRDDqL@uqE;e6DZ;Q&%R$&N_{Jg4)YYurM
z_0_v|apKO@Cpmwl3ecG-qR1|W$kC$}a2{=AY{_#ZcO=Iu%Q>JG(L_U$AdU0Edqp#c
z^CZy-V(%Bt9FTk8&o#TMVEeLe;JvpB4u(1VegdfSiLKntaw*Ion#?x?H_A;b?8i%_
zdf`<nDtSrKCy_vUyy?zOJ}KY*2i4nd;|gVNWOH<KWnq`vOfL!nFqZ+2I1~XjmvO*1
zCzmjj2n3gny)K)7c>34t4+lxYR>`yx;Ns-G708PvGv=f`J^?8$at7oKK1r2Yt!4S_
zmA*Uvo>~Frv$6JpnU6(jElOxZqG5gjC!OY&TXD7!h%2^hYPa*~v5Wv!ON~5^zA2W8
z#<t@;mU8C4sgFq`W^KK@2DN5SV?bB~qt<jJs&y|gRtn{R^hkvwRoRr>9~ZQU!7Z%i
z_QwS;yfTY%Rq+Uc>Na7L!hl2GQn)Df^b7jQ`j;O+Y|2-?37k+8cqCw}1hR&!8fJf#
zBD(ZvdW>%`uZN!(4hz8u<NfsM+aZ0h3WW1GvJ@U=`sLr#=flU5Yr9bM1f+8S{REr<
z2$M5L*bwV~1kAV;@NZ3SMQ)bAMRHcX<@P5<`jD4A|2nY6X?gT0tp&^x!n_b;O}4Vw
zz5bVN)V2KBtZcM_F>&Omp3`s8J?c{mnRsX@T)UQR@hDdn=Bx<TkE7&hq=!*%%f4wA
z?}bxpF^XJ|X#u@YkDWOjBwJKx#J^<<VRGFV7_K3Ic{p*96f!(n`TDp~SY18m6{E~k
zH);8e6q<vmq9=P=N(l=&(=69#DACjy8u~!;(2Oj`0W&gsYcFG89>iTnY#wdflSrLC
zCwO^$;_Q>d%&*&~B)=RcAt|9hJuX5FBCIH1tq!lH5emcq&^v*T9IM83N$<vyhvDBT
zZjGLQL=h806b>wNl{}lZ0K5s4IJQF<egDp2Bx_91K!mvZB{a#o!8W6l-m}!<5dbqw
z5Ec^T2zO~YLbcy6v_SsYxYYeVw6GEoB>0i)PlOhL&<nsn04*q}Y=Ux-=(*P2%=H_d
zlfRv3E9GEx+0`|VY@<qnj~C)|nMMu>CMofMj|x-3Mk~*{6}5^&hC!b5wJr(7wP4<p
zlGOlt$XTn&AH;DMJtxKy#$oD?%I?Fs@wo@<3_3T4UKCtt=ik=U3Cks<mbLwzU`=a_
zEpZg^jnk+$Nv&ZVWd*ElDN<V{to9=Z!ca#i)!X1C?Xkde`VM_m3akgBSBtqlij)3-
zNZzt&MJ4sEyk%7!B6Vp=Z!?_98OIqe5SI~#soIR;hfKlGbM0*@<6N;t%bR&A=li3z
zaCdRtQCJ$^*U@Ns*=N%EqIzoe(`mz19;`Xr>eDuvAk8kX#`!A{?@o$^zQRZYPXpQ^
zODio9`Ld-~14g6!87mCA`G)Rup?H0NY;$8l)!3jxM7uU%JQP;6-Egf)sm{<0_{q(y
zuR&~C)*Y>^tV>ESm_kNV7%XM9R4Ie}JmFIPFy-}%OV~urX}?XW8Hq6CI*be0Y?X>B
z{2bo>@5|?ZeELZVB=|K^U;@Grim-b6R$j60kimODeWmvxtVyU}iHO+9ru_zgEKINi
z6V`@L(*Xy<?ikzW)~~762p=RgeRJ9iR-1W%rio?99J!Up``R|LRPzvdPF0NbmbegY
zdPRDsO0VhHzH?Wk`p<OuJRLq(eMW=-QuguEt*H|S#cE5nD&-~<go>V}#||zTa;>qO
zCbemtt+OUC!iegNQyr?#lv|pA*>V%IVh-HI&#`@D_QOPrKB!O?kg_<YqKs)2skEqd
z{`=E6BKv|wADU-jb5te@g`CnhuE30qtTMboi>~D#>P?+F{hUvx24Pqvgc8=uxSZhH
z<<;{<0X|NRe0iS7d7>Zh)eTz6dvk|w#suH!1S}3~t5P<SGEcyvkUF-1dj>h&Je44)
zL)}x}M7OC|c;C|0H%_`t>*pjwy_(%>8jFg=8vGTPEiE0i;gibqbP9m0fFN5cY70kT
zNkgSR&%9ygNu?rDl~U=*A)Hz%!K9#z$*-a0AWieyDWr+hkT{5LNF1aS$0gjU7U5iM
zR~=uM>6mRy9>KW2(Vg{woD6;Ug<zhB5Oji{DM1)81(q&a_X<-V5fZzF3~#n&oORlR
z^NljjI|R)zN>U|ov#+ub4qHWQ>wjKsD*<3$%_RW*cj|jljy^U?dM}P~O4opqz6P>%
zBnB5kg4brz*yTbsw%hZmZ@k-^&C>3M>?3-EmAWy|O2P(A2%MaMO0{^UBBY9P*7s!|
zMf-%ETn#77h2m&_dJaa-zSgi)buR{e9DmLf{W0A=*njOPZgWRd84v+|Gu3`lGaA20
zSnqAIrVM`QEloUm|NhqA+|tf1MqWq5HQD;P=Bz&3?cp_*##Wfo9xDz73+Z1{)tx;+
z-W4RO&K#Ea!QHEWv#xhOe7^6P_5ZK8fY`$5aR&0jdgF@!UcG_B$`9_<Th}}9^WR<p
ze*<6h0suzo;mSIL>s`@PYlrF6FQ8XH%TOx5QlWRp`c2}hDRGXUi&z|Zx6{@g`~5Qj
zrhXxy%sIk!TJ1HjM(IeZGM_moEUp#|HSS-KGiuyi*hL0^Eb%HpFeD4GEJm%>{e3HI
z#^~?KWk$Bz?Ky|aHS%sw+0iNe7N<Ogn)RJ2YZi2k_q6f6j#wpQKZ`?tm1{Vof#c5R
zQj29L`>o7K?SSUII*a7Xy;G*`HSUHJscx;WoYED`tBA#ro$ocAyVdCC{gzr7xGs{f
zpDtZkXH7AG)SbcY`3J99=AAav^A6Ticj1R3ULHp-<muPx+7|t_mAE-Cla2kKZb#qR
z|EY~8jXnZj<XM&>8>=xD5K_735>bvu6+1=Q_(d1_Lu}(v<GF6fbMx-3Y)=L=ub3~F
zU5nS}*7Hn1(`YCMQ>7)+g|y83+?>XBJ&Zotd=;&K8(p7=T;^u%)%1EE$Lsbpvu={q
zj&yIEl56IYcKXw5E&Y>+<tuS!e#)n6pUtyQtb`mk4fY}guuT<&4tAt6viJIg406=+
zrMT}l7NA(S=BtoTL5}D=1<9{^Rl+;2-hQ#Z8PB#Wo}(Z%c!Wyv5o0Va-oY>n8<u{@
zNW}JkRe@<dlP|zMZ`5=(v$s*8+b5SbBJW?o)A>|s(9I1CVi768k3=TWz02ml$9Eie
z*FjuFHwg=~UC1C@x;i=p)8)qtD7!xgLM&(kEjJDb92S{|FMMP^ycBpjDEd2n_w+jt
zhmdc7u&-<sN{2T**iW4GydPRGbMPe%-7m3Z;zV~H+$(8C(0cde{*lo(U^JF4X&a;;
z?Jqi)M?VCo{_Jykx;*3!)(H<9ijYFr_d9M-p`&6|*9)+B9gnVD^#r@8%)@Nj-K;#D
zmr_s5E1#cZmMS-`_RgZmCmQj0PS9k>m(ffw69O<dmr+dy69PCim%)hxD1Xat<u(l6
z=PUdJqY)|U1q2tk@#GHBT?g3}T@{_KX3_coMJb7rZTWE@&1B&?u_ck>;~a{P9{w5h
z@b&5C?dkKy(BM_D;Bk2SJ^;Va{1Y@kbawCn${85mehhy+YmJVt7rsbqWBBs?d*Hi0
zeKHVu1X#ZKEx_rMCrER6jDNv0*SIpr@F*Pl4tzhS=R7}Zr_%~iRym*keEaQ5i?U5K
zG|IX-I8Tb#s+t*d0?&7APl)Nx5-qPpEQMGiv6hylkg@!hSVIfxFZW!=*gD3_TQ6(a
zI;yQ0>eDPQjZ0718gyxLlIEW;0k&+5mXx)=-+^Ykk}Go3Diq}!%YXBHPpd<b3SY{V
z?5Z!PPfjOmXAMdAf#M4p$re5uX}gi5>|#4Y2!n$4F0w6~WwTY&zCwFK#a>Br^2^-b
zWv6YVne?-M?gfe)>nB}TG2&3^fK|6!WSbmCydLEBUy2?zs)~2Zm3NB89~|vDRuwOA
z!l0G0E;zossB!qo5r4e>=G)iZC~12jQiej?gG7j}vFh6HYo>g@M17Rg4gzb_a=KaJ
zT;d>0@Utwq9?@-OtrY~@(Bey(%?8Y90dS|Z9c&7q(1lLg`O>_H@np@jtOb%)e#c|c
zC9|d5*Y2;^x2JzsA!VIWWNG;8$5VPiV3qT6KvWTE`1$wn{eS8AepT^Hf6wrn4XT~9
zcIkY_5D{2fRNma_ssbfEv?PlF8(n0>#3I^{9GZ(K-)2x7NjY3;Q9xo7oKYJE3<lkd
z1o0KR8psN)nU$UmdbZ8D8y3|=h7`sYgiK9DHFc|v>?xB~0V-Mn6$z1m%**zzi8swT
zmpOK<w6d{E1%Gsg-3GdiK-!2(Hb!7C$@Gn^#<^4)WhGw;jVd{v>o_ag34s+zTa3?S
zGnSN0($YOq1G#N$Bri3nd?|#Uwe^L_D=9Va3b_b8Rh`x`IVWT<SB(xU-trCNu7RcH
zJmlzTuAkDy?4(M4UWMimJs;x_rv-V}3De^*r2(rs7k{;TM;ag>R4`$IEi-Z?EMCm?
zi^X^KS@zE5x>u?s$H+o>l@=bAzU?MV^x%Cb{ex)2_^1NMNJkUSDR1AMB?zO6;eUEE
zVZySABHjvXOAr;?sug(MtZX@wRm1%p<GpeKy_5rvf#4KiKVdtMd6C!cP3fStpYN$)
zrR*bv#DD#JVb%jq=jGx@TdA90OvrhW4ou?%jLpd6sn-Cur3<|;2V>`7+_X(A((cg?
z2^68fO*<q`nE(%IcfV#46|F3xHx*;|aJI$wbD9CesDQVbg-_W^GY@>2(~Kh(w7pR~
z&H%1)HndB&Nqn5_l)DE4O!@a9T>)IaPlbh?ZhyCO*hH@YT9HYwQj*m(?fIeo)+Fv$
z2dLK$C_$#WZdR|k0JcdGW^pPV{ztjXMpYjCt>vzqZ>jccmSkL?ZWU_srYnn6nJR0>
z>20Bz<1PpV#y;xxzkyzM2|olC8D9Do$NA3ZrES^K1pad7U&6?q{)SLFyQ%*UAnoaG
zJbx$+E4?MsR{l18)C{KO)+qfDz3R8{6EH8hYfB#CpJ#9C0YSUTYKI;K&$vasMj*^<
zzZ`qZfQrWTv-2SYL+i@5ot<l3l;cl6IuuRqwBN3nlVYcKaG>wt)MM5((=AQ`@&@y6
z<c%r4O|ILoSl?f@zB{nK9P*Amz>^tn-G8PBY|$C)aThOg&N#Zo3x45+4G(NaD{^dO
zi|Hfp;Es1BZ#TRn2$oo7?j5_IJK89WZI}5F+kBR_I6TfUVi~(y<$!l~v71i^m@>Xw
ziRN+Ic5Y@3h_zjE>!6|YvLCwNfo9fAJk;4^yJfVS!JgSHtnJ~Z<CC3L<+=0fmNcDp
z(OP=Cy9n*1(Snl&3ZnP;?Vp$R&M!QWW3bHS@f{E3XruBGSI0O%*{=Sbk8f_W$Q74&
z%`PZ^`*MG&T28{9@T2plocOvKhwa(<SE5mBK0C3RvD4Qo`&#ODr%y01YwtLRaS4a$
z?${DWr##1BH^cb!nG0uK+P1OKK7dW?K3;x<XBe3Ja<`S=YlvHv)vKs!CH6A;Dge&N
z!Q$*=Ax%A5YNIQK7af{Lhi*PDgynNmX9<aa=iQ>0?Pql8XYCL-SS~gF6tH3TF>jo-
zpO;#8k0R?li3dAw6%`vE5)x`34|B)8y*10UnaykDw!Lock~iRs3-WJ%k+OH~n;IK-
zMnEw4c<uGiw7{C|O^YDpDPtRPPoG4oU$=}q-p8UZzVa$@`8su=O^x?3mzj21BfDZ@
zfI=GrCiLJd7;6futezgxxI<+<^g3*NS-!TdeY~XioiYiy*kerTB>lqD1k#5IP48Hm
zFvwctR`hmW2Fooqbp5}Qpnz_wLylzZRo9>buRfRYVAau80$geQ2k)vQdzaBnFB1VU
zmk}xk6qhE?F2w=Um+a3jV1KAwrD9QL4}$|OtO}J1E(A24Ej#VoXcA#@Tm9$f_kmq`
z|7D`=tO#fYMp=DLpU%K_heu7o@f~I+3to6z*1{@bT~z`%fw^69WG{nd#1`b10uNgX
znMi0DiFQ|z=$E%B7f;Vg0nsof$xGopn%_cE`>kWT>JGsPr5U|7Gk+OmE^BOM!xGoz
zCTet>d@b2f-EPC6auhR8dmAxIbbZUYr1;J{_$Cyx`(I^At));flYPyWDhjZ=+A`j5
zD~lP=a6)^V3Y;(|axNNE9+V@ulquM;Z#f<v0+auB*<3ORKyArfw($2&jGvPk?=z+M
z!YReM>}f}G?4E-z!GF3Cu-C}H>Wb0`W&DhrWE`^KAV(E4hwjP+eRP6ZOc?xJ;&?VG
z*XC3Zt_vNbMa0a_sPWJ#$OREkOWV5YT+Vw57+1>$!d%zoQ8-xpatR1i)!<*sl~Y!b
zu7s-cv&w*aaEX?Z1VPH2bP1V6-_uUJ9Zbb=Gc`2C43@nbC4U0ebk`1#fbXdarO5J~
zn&~`W4~7f_?xILWMub7&B3*|SYa$!4h*C46Hk)}g;;|U9W#F^4WajgxIW791cZ!>4
zxu{u_)fy~d5TX}kb<6IP%1~H`@KtG}7N9w!rg~YE5Yb6Q&pkjX7HuU$QU=zwPBe|=
zT8GEnH&t#J?thHMhWo03zyW+yKBPe5O0wl&>8dFoJyx^R3d&Y!Yqlq!f-k3Hs_)Y(
z-3Yo8oLJ90L^y6is<iE<D;cE`BWJpjFHIV3N5R<Om}p9a1G;LWie0izVx`|$PqydH
zGX=Xk`303zron30@qmL?FBK)$><EZv&S`FiRD)!k(SIaMX~q0y=QmDxm{C*YsR%Ha
z@InLD-xYgIwOyARH_^v{2+1=r5OtskJV{vYZ!8EKJX?+W9}G~yXJCr$Ih;$r(ekD7
zhjN$F86qsLpok6ao;IS!t9Vvp20N9gcq^3`C39a+<y<9l#CrobjPMb-mK&Acvy*D9
zZ168#^ncG09vZ(|i7s!2mzr~&Y+@-=T_Z<<_+le9MpF3|8f_8&O@aR$0qE<RPAe82
zd|L`cieD(Xfi$ggTvxG-t0b$LM}M&)V(s1m{FQ0h86go^Is!qP?bM7Gjh=0sT(PF8
zd?ik*DYfkQB}{B;@I$;ZO#0QiH>N`bs=s_&mwy$LlT5a;Rcs-e5?VIDo|0}3E<3Yu
zTXwh#^CUqHkxZFnlYP3JoP+ppK#s9C0|v9f71dV_q&Aq8y<F)i>=j_fZR<3_3v0v5
zX|DPtoo=`LnrcU-9cH%EQbGw`Vb*|Qu{uFr=L=9Ew5$PsZ(iJG8uE&gk7(NMQX5<6
z2!D2_+2tTL+6spR*{+=SMvn#Zz|QmgU)7`QlteYLREw(C%c+iTV4tjBuQaL87Q4Q+
zD@zPqkn3tFw+5T!isWf145#UsDz=5C?s9vWt0P(&A|K}F9#*@JY3|sDQ!hR>=5*z~
zOn6sj4W&=c;fYNXxy9FliiFsxslIYnQ-7vFjN%^C-ZB?ql-Z<$&y3jW#YyieO$O>b
z=^B%6+0&#+y!1ENQ{tH;d1VGwCoJ8S_idXqt_g)JOu@vt36*8bp*!lC4#`TeJSQ)J
z0H<cc0%n;0$=Y;R+H)nFrO8^8u1`x7sK-tPKzOsGlQV*`Sm|`flZ+@7)|WlOi+>L?
z#xZ2SA$8n%0(2fn0vCE_Q!C%swDx8xHNt_FTy6EX>-dufd`z=>C%ic+v<P{|F+xt>
zt~LZjKVXDb@{V>6f!46acesT1*lNg`)2f>%zY{?$i`>k|fDybSaIuo~UNtpEakCT*
zi<qCAKyuTKp%A(3v1uAlP1Q;F*MHj9^Zsc>Yb+J)o_VCFxcz{2H5K0-(!B&LSx4*i
zfZ5J05T{2I_sBqj%4D$1GNk$LOyCiPbP)!ToPO>5^v5WZCqj(-j!#PxW$o0=CFZk=
zE$2eV657u#b>6Sakuo3LCs#Q=*Hq=?S1oA12csmlk?$MyMVFFZn5a4B?SG0VRg=`}
z_a0co;Y4-O#?z;py9R~FYNVTW4K{(5`IN3fJKWKX4jxa0jN*=61EajrHL%s6)HP@m
zgld5FbK;3{eXwg_)UN9q%<Ju-VLb9NoF4;?+h9lutNo1P*9Ry<9-nZE%R@<mLGA_e
zp{`16FSMH8nSe6RB)^(H0e`BwE%rIIJ`Z~>_g4L*sn%mExEE}AgC0hTfbwhQuyJS#
z>f|-!>v&c^PF;e~)3cgN9TuoR9_K1_@6+v&H#x=Q!<{IjcttD9b{OA9`T2H9N=CU}
z5_*?qNs@EdNX)uo^*~o#;+D&~b6Z@RP_Ek@eR1gn8Krv~W26xM?0@9fI^(E?ogL>L
zE}F}<Fmrmj>=fxG6C3Az?@y`9cQ23H<Nv(h#VjcH91VB${)1k(<G<1A_PQVM>~F`y
zqUXC-1dJs^x3+05j{jzJn?Mj-hmQB2t!-(Z`}4o2v6aFzbMDyHDxKae)Sq01Gp;*!
zwT$xJUF~t|X78!jeM=Nxc`nV`kcW_korP>Ax>J(s=1dZ}3?B=UZ1?o&I3^6~B{UaY
zMtdx+Mvz&$^M2u>c4_-#|DP!6Z2&@=`HQst`}O1L9}=)smzU8@FB1YVIhTRzJrkGV
zohT@mz&k4he<V2$z56Tj2THAl+5p3VGs|wEZ(XR{U`~VXKJ{Vx{|`ecsEvvxw<=%`
zX*(k%ilWG;M@p8bzfSV>>&yB3%P*e>PF@77ygq&Z^Q7>F#GgvyLuXH3DdCJdegFIP
zuU9Fx!_OH%Af++*@%VSZ-}dc`Q2|>hiywTe)Y}&iH7fPNHbyUfja&K{Y^5*wTjB5b
z^c?#~>12`+Ny_Ez-`{_Gk)*5?(WO7rFaGW0`^$e{s_K^!*)AY|m)aX9C^U?8&f38+
z+6imJ$+;*@dbJqFqQZcN0Gtfq#4JwK`sG=AC(n5@RSKYv-kzKd!r;(F3O2N8DxjY5
zxtKrzcB?=B`Rl}<eEYRYb|3<HE?Ip|pTVHDhc`GA$9tGb7ku!!J`3;yjIXnBw`+|O
zg5iQzl$HWlRtlMaL}<GR9iE`jUp}K;T{?>nVIm(T`J}xp#s^E2{v1Z#bcUdElrV`~
zW-?BI52aN$Y#BGXiCWwyKhx>^?Pe_cj3YhUj!_EZfsc~nInTj&p^)AGDqCtTg|d_G
zTej3zSQV=+<J~Sd3i?qlH;$qa;w&vW6r^Cwe&l*(5k|>>Tjo-Lm`II+HuqV~Y9RxR
z!k%t1CxwlM4e}I&fp9u-?LXG-C2^OKy1Ez8@V>*uOEThphV))IrT8e%v|<5*aRg29
zwgPe2=)W1dIEq5}<D*G7LRQ?$WeRzO;mOW?bfO{Prge$S+N8ZHLua@SYK#_91ul)p
z4(;?dY8=~t)K)}GR*AzVohqCJ@xaL%tw@^yUiX;8+ZPS-m7bjQ48LfbXjS${SKB$0
zJV=)PTA8z^a7nU#kpstpL{^*Tpsg1ksUVvJ`Ty3;HS!TaXJx(Wd{Sg3?W`<M(|Ufq
z8B2B{HnF9Zkye<XC#|mAT(`6~5Jf-~;!+{oTKU(1WxcW``w@KBmu8d4+P5j72ZEop
z-{65;+JcTk;$E<(ts87RdeAoU+0-{!p)hi|zHVQ?fYyY`2U3avQdR0u?FKaSvWA0?
zuEZ(D79hY|s5+UvR$9MO&q$c<7aN#m6`>?9c|C16`lVmj+UkM$NvIP8)3e`<H)9C2
z0BiVv43%291LBsqjo$X_9@OC+4kJ$aAiU$l985t4boAh-UO;O2ebY(Yf_UevET=T?
zOp~%XzCyv;Vwe0akmne(hb=AiYujruO6W=HljFxzJei7IMY8TR-N3QL<LqQ0Q}RkU
zL;(XBU4GrfRg;2XZWqBQHkG9G-Re4DMRpl~Y?7_udTDj$9OoA)gHP&;m1OtOA!)C*
z&VPq}?<`S(G%s{qBoa%|MHebG_<grch-D{fszf@{gB=|pIuT{95twSbVP`w!4#tXr
z=#KjY^1qT8ouysR=E3Gb8(U|DMGqv@>+SzIuHhag<uu}9WFm(}I?j7tmUSmRi$)iJ
z!fzmqzO+?rbCBodjUv{g+Fr4#JMv`h+@!@Zd-&p(ol9n(i{FwLD>#TyF94>|<mFIL
zV{uhuWIeAzK;zPw)9P>pBzxBwqX-O*@1!}K1vl!Rne7r9gD9Mpq`i}uT^65$bM4oU
z)9usHbJ~tiO<p_ZY~=BnIr_rWiT~Pv#-|mo(lWU2NSx<N_fOO-b3NaEuM~p7y-e+?
zp@qaLJsPHV99q#Jd5yf6QtN<G;04J=aOzrk8Cq9(0R_wk%ae!T#R}Xb_=tz_Vlc4h
zUIZ_sCj~*?7)T>@6P=uV8*Yw4#YZ;oRa7M^;pJpm9;jr`vZcns%1~!UV3oIjc-6G0
zfG!&;wl*|Iou6w#n|r?RCj>}aeDv${B}jfRMWW^00J@;8gqyybNOY&X*Kdld>-J;e
zviN4Ma*<OehArKgt&tI2QzA)8dqeqPMo(y75i696enIv!y80v;120Qj?Jur3m^VeP
zq5P~oX&AYnA-?sB^Chn*bX%T(UXz~|i6sz9zd8~*RBq#1#fyr#n{J>Yt^iCc@P3fD
zI$<y`WW6TNxT5ipAzv1I)Jli%8D(1F%*a2Rs}(zv-Aj&nUiT_3MYL+tuH}%snd33=
zBEU%(uLa(rbphU8Mr^e(mXC8()2#faMgWCt2F$SU?Fb4eV|BzUl08s=!pSi8{z05f
zdf~$~T8>PxL!Awu=&)kB0X{R=sV$3lsutOm_o%K1Oj{yKaIX~vi#*px>@%gN+kh0>
zzcxNq`>I8+%!lGOA6g3q6BCTVuq2`Fk^4LeE(@$=1>!g^H(^mUr=i+h=ja*};8dLA
zTHfgB8%5<!2hzB2Tg^v*O%BxfVV-l6cUf5B20e{8(9<K6hYkaKbGuD*54`=-DjhVk
z?IrisXE??<<Gs+vY+&3bauv55dLLCA6W)2nJ3(3fFr*Y+IfG2bm{nu-DGaHN<?lwu
zpA7@86w;oqlv|5TI4*T7WWw4_GdoBCachp2FG1&@U`u=qZZm#=bA6e#z*lv|S{7>_
z-nQtTZ_a6E-86|AW;q>Z^tsiVURM`mj5R{)cA}VcKeu(N6d|k~?;PJYNhfYz2Y%Z`
zStLeM?_W2ebFgw4)=tLClW%k-W#h1q4yugFRnp=XUt4lh_85KTQ7N5Tj{=`@IfafT
zN5Slzke_gmjaRRKrJR={3LWNNW6ncc%7DN14aStX-Y8y~ld8v?!;^1U7<?QxWl?K+
zXcs5zhFea$nY3gjyMC#sq(A5Gc+;9w`;lieJn2`JY^I|ZHnszp^V;$6L#v{FiYoU+
z7v~ZwXDzzlbnz+*g^|uA?q0lFt^u$fDb2h-CB0cstpT5ZOfz2++oxSXC`askZu4=*
z;bA@C{$(7OL<#pzGq*tirkhOC2K;E}aNjh%^&Osp`K!=!_Fi@K1&k&UiIJOq%`%C1
z#IOS-PyMMLd0VOEOnW~iJd3v;)cK0%j!Ab*)XIC=O3=yo3u4L3^AQD*EXMduL>VLw
zm9TqIg)V1*GClWDh4xt1qoIo4L>2W2ss!O2OnN>WRlGopOmn*fRV)a_7waxiap_!T
zk~H0Yx&tnlDeVIiV-9L2QLa0X4k4!*taWwXJ3z&tOY#d3<Zs8gqW>>(MdGfd^)qp0
zlz4pMe3R|n%@{@vHt?q+0kc4CR}b&ep|Aj^1FY+RK+!!mUWjgSr~0izG^P(_8+-13
z&y>M<&((S2^+@y&kxreskIW_8BX7gKM(8-%xn>2AM}|UpL3Ms|X2nKLOP>Gtq+AKH
zizfZqQ&&z~^)cy%iJD73&azSlTMBcdOV;=}mt5@Ny6MeT3S@Ui!#)2J^Q0O@x^-`q
zc_J=<GKo*DQZUH}WeV2%C)FvGw&;cSXGhaG>VtI(ChfX9h55LB4<Yu*#$$WPZM%{Z
zp#3EA>jM%oB!ikHt{)1PU6Ff%e5j726z=z?%PQ$&dGV{2LuEp?>GZxky5IGWj>E_0
zh<~!-4fDY5HYh~qe7U;SFSFMy|D6c(Vd+tS3#-B;+^z_JJj^}D=o2w>5cmFY$*UK4
zMZK$Up3~A6*q7UNAP~7yhw`r~lMX*HHKrgPG<<SiOqH!+R(B-DR95Q6+>;UqBcdPI
zd8^2VQ3T`S-QkD0{fHPwx86u@4f0xuY{XrP#eO5L`XA`AXqO6QZe(+Ga%Ev{3T1AW
z-b^nB0x&t3Q3(PR0Wp_xr#L5<W8*HFe?{{^zyJL8&lg#Ql_J>qF~0F{pI<Nkd1<06
zEsPD5K_!fye);1ijw2XhU6?>SfgYzHzfQls{EJg<ylCP4=g0388~Le#=YhbJ)*3T_
zr>ro>Pu2#Z<4+kp{CI!0Qqd=Qir?r(&>1N21PKx(XydF191;+w&LBXH%QK9Me`A%F
z`6#3D*Haw@Y`@DwKmDFpBYl=<2CpRH^B*JgmWL7A^0!NnyZc4LR=c*}?^%All%0`T
zC|Kia3Zt289U-obi?KrMu;8+fQqOHuUjc<xTz;=~)>771rX(;5R{tb<hmlTR6<r4{
zmlwXr7ej%}wQO9OOm!I9!ComTe}IM<m*}8Ksy79ZWL-ETM9!A1qJB?knVZm3E(iyX
zb2OjXwTYmVmukc2bT@8V8lh0%fLmJuS~#QliYQH%B>BPleo3?x&fO!34uarW73AC?
z`N)pgoZdGP6NLAR>8+3}a$XKKG)+!o{KA~IDTWk{>vEF;VRQ_MoRXDkAa!#}rzv<6
zKWzm%KL=6b`FS@+;34HRMvcPcnjV1=1-F>xF17|911tYLbH5*XI+>fbo8dpRE3F>c
z*gG8w{|lqDw}Y3_OfM7xFqcu27ZaDJ^FJt;yX-ES0dlui?Jms&f0wNA#cVVJy&Gmr
zwV_w#^n*H44o@{}=h~G=)v0&YQXi*WosP)L>%oeN{=QBtP1EGgrYg%aEI87qrs8-@
zbxEy-V6Gza`fIAI@;Yd2=2O)bHDlNGm)*AT=^IV@*2y}qF0E__8t5v@VqwY@R@RK1
z`exOS?yzWIMHLygf9g0~sV3(tIy=8}Q=p)%EOyRsK`00v$3Oihzoa!*t*`Mk==(K4
zf7vDDw9u3hx+U=Ym)*GJ=@^bUDidJe6@gMu2=}^TH+xZ*?}w~oV`gx6Vau&JL>}FA
z(Q5j)nTO~y#pUK1UFK|#mDdJbZ}re};M^HrtImCB2}RjZe}{X6Vm~`}|2eSJ@Tnd(
zz%_ZX94qpS>|!d|#zl&b<J`BhBeo$W&9}{Q@5%?o!LqJ#GT&&!7Wy+coBOR`s~n-S
z469hm$I7*EpH_{*HSweS21exw5?YcLV%ce%V{<t=n=;Fu=5|I5s)?p4v~1zcHTZR_
zL6I_y!1<Wne;*C|#!TV24{$pHFcd$S!+MdNJG@6o<u0SXH)iy+BT{RjgtZ;i%EA)j
z!vzkmMv63=(`KBnhE^5F0<_t>C6;g;NS;xiR-K}6n;uOAB5Ur|0Q;LR5|w0zZR&|@
zO2fw0YQ6C5=9foN2Q4yPC)wi~DsMHupEXu3*<9dpf73<J+_pt()ntij$y3*{YX~X6
z_SZ{^m=be~;rscos2tBid>(6#^ZQNB&KBG##L)V~>N&g2su;zBi_pqe=M``7pO^A@
z##}hcP-%U(LCm4^h)MX2s9tHj%~IB4DA5~YTw{@1*v;!|)JSVpllk8IB0PC>IITB7
zlUKObe=A~>QdVs>nhlNa&AP>v6m)iM+p=>T(TslAN4zZy&vDasExLkr-?Y8`cx+45
z{TSP?%Z^ReRXxB}xuMB%2bE84n7gh$c~tfy$pFI}#lBuTJP_WSVG15AdfWD;ufF{C
zX8c~CLrEpj&Q(s+1dlk=-Lc{W-y_iZAICOWf9zW+`zhZefF~$7U4$S19%0z=3Gcdq
z4i?-u9d=f7J?#f6G!+TM?9~=nexVHCpY|%BTUv^-^r^4%2dexk|Bh4eQgbM*TF7wx
z`aZsxudnUf{>Zbnt4UsipRL`V>W^x--|ea2<D@Ta1`3OhrXJW)I{{8FcD!wRU-=$q
ze^_Czk4r-+L^OBMerQ(*LP@{W5nXy+A{T@Mx3F8W4~6lJVP!JD+L0YpD^Gr=7P^|7
z8h6EALZb+NfzoGt9&ppyezxcJRF(elr%Bwj=L5m8<3~SD%8mVpkY^PkV<D6LzJa=o
z(|&gBZ9mttK?yT|+w_D2px>}-LaFqoIBod`?!(`|^u)G;QvUR1;OuGd0b|(<r}6#T
zQ_bYad!A^dcDvMj>IFaejO4y|vsCl`A3wVIqL<N3FBF$C`wbHUGBuYmm^&zcTFH(Z
zITF3=EAj`5<`Ue1LSa;L$-vy)sM}x;4WsQj*o|He{reL^GDs#fQ%bqy2D$*1B#LA#
zjO9fz<@9Be(}%<J>*3wmz~n`+%IoR%^Q7<xiT{<vL1!nglyF8(uivKsJW8n@zMk;~
zQW}FVkDmj++ouPk0?tqtU-+wkQcn*a6s^HIM(4H0<uwLp=@Y&yd_U4Su8-15+9Bf0
z(N0hQef{@A7G|Z0jy}@c`0eHO@cW^%u8puRP7s4|(N3Sf9q0#8RyZFg6_hY~`s0`B
z^Wo#TYo%q))c*bPWg13)D#{BAl_#CEcBVY-gtcLEE($|GW#ys1B%siL7FFXUsm4Cj
zCyk%v3xBewLG0_SEa=)G3<|__QJEGG2~g?q9$)mC#(d=ikHqTlr)#7LN>2{F0OOf%
zl`eS2(kb(HD#i)$p;9@mP=U8o(N<U$S7ed*q@n<Abw!kp$coY=?vnpmnF-BCkkGqx
zl$T7?MlBo|rc7p$Oyrz@O%#x-Rx_%b#%`T3CXPGOZWg^8l4;>PzN{_Acg_?nMiNP{
zV)FNp%S++4y{BCB9jz&HoU(l6IwS)QWjTG5vC4J33tJn7ckXWw9TtKJ(*NzD!y3#t
z@S~mDqLd-ToBi!OC}FmVW<h9XHc6Ss{BG^rNlFmWt3|}hhgax-CbzRIrm~R+gERq$
zjdj7{yNgm6Zq|PJsk+k=P;gduGXt|OyG;rgHX9d&!9kPDZ$cCREkbNNt^}DBe2zs6
zq<J1QMf*iFv?dw!iTo{bu_+Hz*BHTN14t#{ki!x{QtGW7JykF3nOa9}8kunb15?K{
zp)2Kyz4&BJADvx)A%7~zpb~P+QR_y$WxxoK3%uo=@b(Ja8s2w6D-)5Z`Vogt>Xy$*
z1(hgDW>E+ebRy0~7a_kW4k%yozvL~mk=}P5VA%8H&gwnAW?oQNctbunuF%I$sL)ar
z9k5wRP4*=Nar{}z^SZQN2^(v$G%wq#OO$Bb-L8S)1O~5vf!K-yD8w)pz;v88osC>V
z2NC^3hmnf*O-Y#1_Miey^$@smCsL<&&4+C9t8@%75Br8~HInMvG2k?*Ue=Gop{&rK
zL_Nz`44%)`-+lt#3#ZgT<lZMxj(IK!n^KKF`_2GEuZiPk$)Z)+h~1sD-mzNwTZGVn
zAuWuWs^<xREd*SXz#&#f>zQxl<jrTN(G0Z=!$w-J)eE*j>4vL$WVRvVPK>LAZa4M1
z8l#Mi0;}&~6S8%XbNl9K4lyZ~XI8kB;`ehG(n7_ZHGr*UtE&Hv?gpnUHOnH}a~Nyh
zYr~KwV;4kE-1V5H0*WZ{DdpQd!#{Sa3#)5#`Y6ADuzHm&A-80;0*-inM&4!r#G`%E
zq#L%z3^MBss4G{(c%sP&-`AP_j(u*05jRvhb&|g0aV|CNjFR+QK_ZAY#)H#<00*%y
zVvXM?Jnj!uI&(Aw6N8=pnrIjoAw+Y|6HEn##XOq+;oWavKL6{}&r0BeKWPO*74&<B
z_WCV<f5pNND+2GQujP9XU?@T=5wRkI3C$WSL!7`1>%gb!fP`>5BC$65l}3&5LE3I^
z0!RV0X#~0<<{|CKqfGB>Kge7)5=o}Y27RMl%3Au0>v<J@rC0x*r{b#rO^46Z;bYZj
zH26m$(@VFeP9%y5MfED>Arpj(p1H>fjy!OGZEz-L8N+}XsJ#G6i-_urQzca;<&jQW
z9>T472Oi@0aePDiLC~y^(kVbFfQ1cKUx6%A6_sYSw$Gowm1SRW(TDCmZ}Xr`6bg5G
z&=J7}MA|DsVLZ~cAXcXHY)(hVCu1MGN>44~HB;s=**n;<DYEc!>R3#oJ~UZB?o|(e
z9+LNFhkC|j(y4_yGFaWPS~g-kOMr!@?%->my09L?s`0V^W=e!CpLS;g-iae+EoObE
z_BM;TfZRhVqcTTw#gtFVJnIUn$F+gET+1MxTo*D3QA!>9NHJOGjKy>~riWe|9X{LA
zJQ$Bb*^HW($1AH4agZC%S9RJs7JVsy9yML|>oqvfzpmt805#X%41ut<6x(9zx+|fM
z^Y%&0)2ltdbIpXsez={1VI>4j5RUSa;kM3V^g<IwD~EIZ$ZgH3CZAraayN?8E650A
zt#A?+`uJ|FG3Tw0b~RAeHSgBbsMzGZgq(NKrU?ND7oy&`e{@aZ;DbPnGIF7R=B3bD
zgl^GrB|^7wn<MnE(|lIJQ4bxEx|C`db|GGOCM+tS*gPiLU9UdBsPC?!ax7?sm$vUg
zCo1>zBLxzxoCP6SE-FvCZ)dtNUQKY;?I`oyV6G|irbd#vU?9fs!qt0>2%So(g*=^e
zu~Ttc^9^t(mRt>KSK#~1?c4r;EgWtheYNTFxK)(NR=Eg^8NV>+TJ>Qj>G_#P_3mQg
zavwKFVWc0qNOVq@o2{g6r^c0(0JJfK_A<ZQ@q*<{GoVQ0S~*tMcb6^qNGA~o)k1=Q
zt{;F0Vf@0R3d*}=1`xC`$}UzOo5Q4AzN-z;WjmG490Pr5y@qUH*<5meM`OFCS8ozG
zBWJ*ak9MR%2oJRG5S~$H;nry+cJc)N1@`GqLw-hNpnj!kwCPrOhF&jr%eV@9?BJT7
zI-Iqc?Xi-K0oKpqB1eX94de0z`DQzC05vnIc?9{&ShQ9j{A)w%mpPAs8GmuQJDc&B
za5v$IQ7Yhj`^)}9=p(9sHJxm<rHhs8dei!y_F)ZODCrgZSoJ!@v%%>e$JTzL+V0Km
zOe}rBuBA}2sFyNo+BuPK6rc2EoTVU$sN;yU5*LBXdZY(o-FNq(EV?$1?b>v#kF^1Y
z5Sw>}HPLmTpz&Bbpo;e^752qQg0MA$9eJDI)4b)jB_$(zcr6)!Xv`0<+2{6HM`<T-
zh61225sVCuWqnbLEG0dZ{hJy_G?SIR)9A|VQ+vRsg_s_DP{Eex&XCBQ79SHHp?>U|
z9;We<3|FqZmx=-_p;@BsU3dd?QKMj7+n$l=9%y-g*_53|H*pItpy*or8qn@nIs_|J
zTyVc_Q~Szp1n@e4EOw;?ZlWm(JUN^WO{Vg#me=f@vNbBojB?r>B1UAMFe@Yr;-DSs
zb0WM)Z$o+`@N^sz+!$BZM9p%<%a0=DQ|vE0$~v~*(%z~vg)7{injSVF=EB0I7c#iM
zF_eLnU-9GHt~hokgFtzDdD~dgGUtKRbAfu?<*6+$B2u<&f?0_Nx829=$|`gpsZ~oV
z7?<K>-#c2rQJVLS{3RJy%G6u#l;nAK3sy_5y4RHqMeQ1I2RYve#SALXt8r5pvR>~P
z-D6x;a}~GsI3!#xfDIObDws9mKNiS=$1DIa5FQ#|T?6d@G|O;(m(ffw6qiwQ3KN%~
z9y}?RFp~%bm%v3Y1b^?ZP;V{JFr48GA3$JWy=w>Pt-;=ko{A=?9Gd@s=#V01k&<l7
zQraB+qQ!a7@Eel4`_k#|<J0l&>HXPI=agq~*u8!3K>pD34q65+cMd>V1Kr!#?vD>z
zqm|cDUZk}#@^bRs%O@Y+8SwH1u)O47fWtdS5a-A<2K%|j_J4DXJcTdv3G(?I$7O!d
zR!0g!UN*iCf4=?pq_eQ^&+KO<MWfh;&N@;;dyrJ5Vv&?Y7Gb7Yq@}aGd{0#c8wtmc
zkIk>s+tWW!EqMr5fp^_sU!US11gxwJ9a*c8vHSLS_xb7fmTVSXD=l;SH`d*^FI_Y8
zDf=R5(NJ5<Tz_k56^rtgK{?@>HKbS3%4e-!dQA~3n_W1hZx_a8L>LcT*eYYfgiGkT
zuy~5ZE9`o1jB%{-kcV0M11XCW`4h!RS^Q6!y(FLH`thij77yx|#ATr7j52IHOImpx
zh)pJEH&fKKyd?uvBVz*uue>)6Vbqn5<zbZkFaeu~&3|<jw1kYX1=GRcj%s8qO;Hnr
zB;q9jg$0Kk)_ie1aUI^RxG{}sLM_TRYxeWuoz*dUd872;RnS;Tk6GH+%x@f2;fe;V
zLQC(q{Lv%HR>w5DHYU)4LhW0=!1FRXR%Ee#nvd)GdZU7aiN2>My;o2SsA#GP%_}0s
z&dI{ivVY$Y2$C;>%s3E|<FT&;oP!{RH05B3<8EF)6lPLn8mhO$A@Yo?)0R-EdEN?z
z`spp9(D2opL!lu^4-*t;s^G(f;(1a&S5XC{9OI6P3qq0bw0FADAS$#I!*56emb40<
zc20tZ>YYizo06d6*+8{637V%jBtdII*c4%<%zw4onE80fQQq3S#KUc<wFvM^e0Zxg
z@?a;<0Lf(t#&<(5)J6i)O(-s*qz$rX*vR5*`E!7c1v0wEJ8ZhkTjErWe7XW7GYn=+
z3M2;ahLI)iV`k9vNH#yhP&Ph=w*ha#MbCLO{(CHQrH1BjQHtU_I$k6y(Xyg5xYZUg
zp?`ukUBRM(iz_RFbG0PexvUqf5VlsvQNv&XVHY^wNOO17AQGO&(Y?*aQftxSo<^=I
zruMW%9rct`-oIS*OgPr8ubsJ7X*e!5HVJ39((%am8BHgtsw~K3S2WR7)~7>CqS>%Y
zjw?%5`M0z@z7Sk(1`~@c&<!)JqUb?-KY!CwOgkX3yo^XUEM>W!wR6CdWRGFmF<={t
z_Xn&rVW*`slENUm+?>|OHGv{@tzhsvAJ@dniiWLWazj8VII|45*QklPeax|8os9;F
zO2=(1(z{oDvVyCe+tax<#UIGIwP3fL+ZZMT-L1}L$=J~fTb#?^$GHZT6u&*4TYpoG
zPAjmdZ!eachEkCSko$o3Fm63nTV(wbQfj=kUj;N>fRrA-4gWTFzvoFieo~)CFl{X;
zooEHp?)jl9jQiTjOOuVQ0Wik&!N>U;tvsNt9zi**%h#Z{WJmdCr#8LlKeePt=p78R
z&2~AxSI%->W@qHQ>iel5lbj>pZ-4V;k<~fnwOz>X?_wQml_$5bT-@fky~-FO6>!h4
zp`o}~P59xqq2X|&Z)oNDe;NmaMrqiyacC&+1I^P?eL?3-t{d$c*-#9DoOi#*BUem=
zPD&WKRuOci6QAsOE$(vU{hNl;Egc?!BgU}OIN`JaN5oA6?I*6fz>)R*#eeyt?Zjq!
z=E!daRU4G%iJa;1^-2Si+n(0yFEhIv>lao7C^DJ(sF2sGKS;j4XWAb!m*$aA2Ut3w
zJJ;unWE4B@j0U9xO>j%jCzJCj{>Bm0VN)r0tWoed77bbpei&fEjO>ME5j3WC)cv{`
zJ;^&HiFhUbrcveK{#-vjAAgECiDaCHSNFxTXF@J2c#$$4=gX%|Su(XuQ($;kQ$LSN
zBMqi5Q%tFfbieRffpW08V8JwgytGG=V?A?h<8n)(U~)TPn_C-<rH+>Ximh2wj^3XM
zI#$6s$*X2<a+azH|M}Nn;#8SJ^HZEmfqKeMWy)19W9Qz<rJB-n*?(mNITms2(gFm|
zPBAoW!qC`aw)`y@>G^qk<2qy~%dg)NZ!{EFcaOZOdqj#0UwF8CwAZ`GU3)^5_%igL
zw}gnKjK?3oB}6ca&^*u*?%(pEMn#q$Zu#bKd5OD?mtjF`-X=t>G0QXe&!to?UTOm?
znK?Q+EdZjde0FkQCVv-*#S48vpbWan8Dmc0{svBOkr=td`IswbXNSImFUVcEAgVCB
zHkiv$Oy78IFs{%Kb6&Wx7+6uu`1u^~6@rT*gaexo*C7nRV|N3sB%m}0?bSe1-rTww
z*nFm2fY*;d11bjpxc_3go<(5->zTuz7wW_=ahIbe46loQQ#v^ZuP5N*k?lO5PcYeD
z1)NSYa3G)H;Ckg4Q)Z4y&8W7eQI<YTaWEANlXu~JXnEg4?wqm!6&YUNmIp5sIWG?!
zV)hprD(-OgYG&)?xl~}gUJ~O3J({~D$%hz~B>A%Dq;bu^mn4nRJnT!|Q%a;#{R_y^
zH&~a^OfM9dQ3n?kmvsj(!~y=dItVX62LxuP2gjGuOfM4xFf*6YwlpJ?O#Ta(NS*;H
zf7M#wZyPxde(zr~z`g7}L?cQRDJ_Zs-R*U64_TmSd%*1lLE$))*4@<3IZn~Nzy3bT
zzw7kk#OVN&Ky1Ip`w)jS!;wge{78<ZK2YNu^-&v=0D?mkB-@?XgrY`p!?E(DMud$h
zP4QyPzA3l}L32|w*&}JlhxkN&qJ}6H+)KC@B!3xyY$(>Tm>WtOjY<uH$0+dSXmc~9
zDO$zJ7@bTE#yreMB}gW!HU~FB2?LZ$rTR31Dy3oX@kVBoDtS=ZW`h{flSzf)BL$l`
zWl(iCh#T1(CBBIg33CWFQKi0%H%h2awP8v&GEBvk%_xE?LolnT2UTsT1k;h}D`v!G
z?|;Hn%-Npi)Nm?nbHs+FN(z&3);bC|i9)@XniNwVPnyKV?q%O34RsU(C3uiv=ahZD
z7eGO2Rx+vt%mMXNv9%;Epz0hDTs&o5!lq(MV-9VEe4QgV5`yW5DP_8DSwK}ui*F=}
zSv=zn@g8#&jH1y-F7TCHtx*D~>Ndqh^M4qn4o}sC3Vz7em+V30dJmToLgF5lCt-cT
zf~OL&9%Rc?ZABi7lBM!#lj{S5`kCOCNhpHK^<HVR>A{L02xU`xP|<8fP(QPpR9M;!
zQ(_LHCQGz;VZWSBOTu_%Dm`&i;KoL!rc}=+5{m>IFCZpTeHV*Y5}t;D)hgn3s()$`
zE|>$xDVptF7`9|vRf2q_R6Ve?uu@es6eHC<T8#{$-c3fDUs8=ZHdiJ@FN_wA0Y4z6
zJbC)GX@4ADjK;4od;PA7i1$}b`?5EgbmL(Yrk8*FJo)MMq&Mjrn%;OZ8cv=*ZC}7+
z$ovF1+ZXUP$QY*=@JKS@>IJ-k5`XMjFCaD~k=X^?h1$j8Vf%7C>c8$L$L-7KFWNWV
z=gG<8AtY?y{C?R%rgt4g9ZtI8<jTNgI^L`9YIHsBcUMz<v0J}%XM^4kqtC}4mx4vo
z5>D{`7;hR@R3+MH!{Mk7#9=!9ljDkOreMXj8Ig=QS%^_zYhPcVPVDMvFn|25X@3}v
z&$@A)&Y!e@Z+~uoJf`WztFAve=8zp>L<#H?LJhv=L@jLC0l!Y#XAOpb-TpLsGiskV
z-^EzCfGGkGIH|@Ue?2^0u|*Z)=6PZ2hGz4^X`SJyc+8Lgl!7a89wEm3eDY*XkM`MO
z`2J)t8bY9dzWUkzzI#8JTz_7D-?r}tllRxBu0Q(Nez^X4d2rT!YX5Ql%S-p+qCLGB
zown!ah`Ihb@br}EoZ=bBbB>*MdU~d39M1hYm-9mH)wth&?DfZ^E4=%cBVc1VuwR9u
zPwuZNDZ$ctp(&|4d2~h3yWZ8{bZ{}4{QkE8zU%+ijjPrq2S<sP8h<}OuDRB{^IyOI
z8)4cg!g+XoadC1Z1E9@iU<eS}MTGbkoSB>@!6oOn3IDZ-U%`UA26+k_L7x15f;`DG
z$ax;*JP&eCb3*2X&WSuH$_+rUy=QVMPOa>jd1mtBLJIrpn#u=9CEpZ`cf>vvat^p0
zi+2GPKuFQ$V%G&`*nfmRpk3mKKj9opN|DPA(05lEqB<)Aqh?niOrQiDM(LJuU&DqF
zq8V8l<t(F!jge6DJb(~Fa0Cb;-l66};x~(+YE<u1%o{=gx;#P^h7(u;Z$sLhvOr{Z
zr7W{ROyrPe9@_&)my^NA!GF8)+X2FUd=4C{9+;8`I;Ol^LVpNUoP?#)NHGpD_1H?!
z-b9?mcz~ZBwImj2{lp^eCA%N?M@vr7vWD_Xs!Ozzp9J|yZpwFST+=LuCdi>+*oG68
z7Q^Rn=Zq01^ILI}Iy58SJcoX*guZJI71_w4!uQFcx>6{od0I?mMu?viG$+DN8B>zs
zl`gARF}%wA<A2l%<(o>{Ep&nHj!|1I>H_x2Rr{l)DwNFWAu0i6U-))}qSoIA#qwG>
zVOf!sH<PdsRzC#|Dp@saO$B#1bd+54EC~#azSjz|+CgJ}w99w5bZ`85@X3wF@7mt!
zRU5ny4(%fE+PQFsI4cN_Rx#`kf=e-H54?`xl5S~MR(}{4El40Iek)G=NA<M@C&g47
z+$}g!b(J-k-|q%U%LaI>-UqyMT7tLC8z5Z}^G-<Tosi0I;LXtmURET+Xn*+8{rNxF
zy|cI7=gVo66EeD-kjst;?xW?UmCg>WO(}a@R|g<^iebmPIs;A960btBv=tw{ogG|s
zZ_h{LkAJ<%^uto2gCp|%5b0~yzgaU$WzoJ_>20gM(Z0bVIY#@)D(ngWcKnD(z)eEu
zyvPtiofLg+i+}2k2ffpaZuT7yIE#=XlJ!KPd7Dyyjcbb660JLyZ8!-%kmwei3}4mF
zrr;({6&C|<!5Nf*Z`;fs5}Fs>znv4Q#o30=?SGtw0k-9jOlo({V|g==<$L9^Ld!gs
zcgs<QTwG?T0#gz_b~p_$E~oWYLTZAB>tU4h*XynQu44*stbYmHum92nCEkK0dv1r#
zI4Jjt1KTBC#6j6D4iX`=#k4G72(1V@vONg}bR7tV@<r=W{;Z*3Tiai_fNeP6!v)AM
zTz|m!w`<*Q``$a%?SwjRrEDj4yg#UzgFiURyKWLPa&_>$P?A-h`bYofGCKVLDF>gX
zWjX~EuIun`CmwFEqSm?Q8-y7e=WRH1bQ;qZoSd>tK5WCu_N66%ivV4V$+c)zJ0(Q3
zucv}Ky#&8==+Xvy+nJ4fgl<@VO11MO(|^#Vc??sW6Feu<oY)QOS>Wm`j*q(0)1I{u
zfl56J@DRI?w$o>)ylmT%ktTyAXD%8fy2zWGZW1NNggOunGEqk7!MpGL**0UgGiWm*
zSc?-0Uljs82y5n2qImWx#{3X_pv5WtDR|o|k|v^E5~^FFNT!=dP$=8ZCjz`^$$#Q0
zJb)oW0chEYyo>?^9RVxA2JTO&y;bFp5rQ(Kk1oIfL23(ihcHy=fMI;xY~;~+=YkrG
zQdZ?CO3N*2Lp;j=|0C!s!xB6hE)lljWHi7z{AsG2>0z!lhz#jI5E)mho4`Av9wF-O
zOylxYP;jh|zLV4+4X-BS-e5TUR&GUbE+qX#x}NrDhvjZb`)gdY4+sF9ki#~dF*}6V
zxA3yBp9Ea<zXv)NNO9Bu;x_(AU;5w5u|PX6IZeNDo}%Xm1u63G=L6@T7E><MEQKQ6
z&N`=3Tts2|B_tJ94J9c`IyPEDtcdhKB4zyem(ffw6ah7tfkiJAmslAuR)3qsmPCp?
zA`j`x?q?^v_ub3q?)j{-v%(Qr)qVcxfFDTy3zBad>MVdT8oJM4-CuhtmEr4!FO*Vi
zzAS%t{7Hu=4UR_u@x@O84o?;#%;7Ow^>el9=V%_KPW%b{c?|b?zL!RZ6+B-&4&34I
z&p&oDDf@i(a!pfXQSjPXQ-5L!$eIzagsd^LPNe&L$l8!k-}yPQMilN*l=yshuqug2
z7#wW!$}@uvWN#CIJEhNighvRlwvm232g`vCdiVqBeXk3_ur{93`VhSmZ7RG8W(^<j
z&9Ar5-G95XWu*lf-x+X1vsM22wF?grSO@DnrGy9Ue*fG3*!|hAnt%M(en!wm153ku
z85vj^LF784P#FHa%a!bnm-g8q;4wdrc6m*YZNC)y;W^1<D`f4;pWi>b=FZ1^F_$6~
z;9LFSHC!VwOTk8!LWU(2^^1<~d=$V>tfY+Mu@fnJ8B)U$nzy@hPAFyC73KNxWMnki
z#PV00hNA;-&TOzvnGq4t0MyP1p$DngDejYcz4E@7I~y+=e;qeE#NbDe4`Y-LIVc8S
z1fB79gvRtV0ue^P!n&rOodJu-JU;FEoFy*!XepR+R2D|Lc6W=bECeNQgr$j%ik1nJ
zm0hO0TR6Sr<5{dz4(I{Qz8un}U24y8=V00Ha+A!(kg#M8G*)<=*fa&*=QD$N4QK2T
zG6m+rY}pYqe>vlP(II6qQqjG6wHb$yqzKe>G!$#C!m;JcY0~|1p4|O-U(S!@IRj%W
zI~+(lrCrGC;ABXfK6l2QR`<j1$A>wFZ)|D70y4=8LftA8PomG8--^9tstS8{1NO8L
zI$*Cc$6HU3hUgk7Q~yitd+LMhIMdb%=crvdtiiYTe<|(1L|ZkwM+lwMKI&L<EF(=C
zY3upCX9Y7+FWLe6^Gp(FWz(#4*Qr*9F>;i$`Wh&OtiK4SX#3jk@K)?jV-J+9v5YVf
zN<yn+&SuwAEM%v8i7mxKFC6F#6sxBV@eiduJi7CdXxBWU$h3<>il)M#G4a|6nNx32
z1)B&3e-J}o`*y{9h@5`Ao{ceaKsnI@aF%W;<%BWVj^;ykwOnnPsq2YcnwN_pP}xVw
zWvqB@CUVWZ+@7npPM38odU)s#zPz3~GMboZY_uoKWi-RYo~MciavmH0MA@S|mF#jh
zxb~BNpsl=7?{m<IRqnY8rY>!p5?wtfO5_rCe@&D%REk?nOmBU)90RRQ6|${U$dVJ5
zW`;N!u1Q&Pg=I7|0#F>FjNlq@I9yMlVj^h#bJN<61URTXUKTga;`4El<Ck`Pn<3;(
ztg)m@qUWBhk((T*`H->nnlHyN<&xpe3`1FOTp+Iu(#!?P9DXTt(OOz`>86GSN)&c;
zf4^e0&;<tSrLanmg?>|;LNftJ`lc?0wwgD3qu)|@Le^X>R&!SeeaYz2w$)~(u2!L?
zho+sOj=3~QpG}M(>v?BQtvN-KQP<VfHM2~$Bc!o1)z|&Jy6X1jSIuA>ACVc=?LO7k
zrxdzV$03zpdA&$bif))zOqQclsaP>;f6-T~w$`ziSH78C(dVkZlJck$EgTT5u(f~0
zrmoJR9M?b+H`QDStc`K)>g6;)J{oKATmkt9Mq}mqRDhqHYG3)*RFu^Wr-sbMu1k?(
zNob`vh2(1`FNI|ap>Ztuo5HfvOvM-v-xHS=Vk|yx_TWt<6&%wb)e3)b>lZx6e}cb}
zqJ3!wZff^32(_Q+8n~iP+>nENZv7r}(9NdI#lbNCjzplO<5~&V`c!ib)sQ?h+YZb=
zj?Ez$hnh8p&k~BM;@Xjit75aoHmUYPyNRB9odX5S0@!wz9GbPO?UZqH$cZkFJ+8N$
zWJv9LqU{(Ti*2Rwl1D*1d%1aSBKK7jDJJJSZ8fPnrx=~Sn`)l>3cr47;t4EQp?YBA
z0TWJ_a33#18aw{#ilDS3#(TY9f3F0(E7WIggm;q-L{x{D%pWgBe;KQo^h4V%H>`)c
zt^cQ!XlDOFtQP2P-kFwOpr3f-9<VZ#Rt6?I!0LPT*0PttdcnG5;PybB1JLkK1>1qk
zHCcGEcqeY-PUr^g_N05P#9A;-Et&BIWrln<v2oR+{SuTa$G*03#UA<)JRW}!L?tm_
z(yaopdO{ilUI6Ge69Y$Jsx64w0@Q8ffw{^56O8smm(ffw6qf-R4HTD9>_8}&^B^yq
zf6t$fzrFwbpbN7zXBL$6Y48CR(V948%1E5Jz?_7i_#2Hp^XSBzBg#m;;hDb2>Iw0O
zY5IjYkNfiMTvqcA?$HDDsmpU1$B{WWF&+Lsal!jBC;8>d3^(qdIgDA#&*x~i>|kS5
z7#nG8s)fR~B|#?@6G{qsrM<W}8BJLlK9Vb2EneIoK_wj}6~YIvYOLzKGTO~nooskw
zT|a)ZF59R6#hS#z%c<rsKQkjL#@R*gX{}$)J(u+~AD>3IOCm3h1_MCxte5eIKobNv
zF*q@oak(uMmj)>>!GApO|9bs>XN$0pr)1Y8^?{7gu9C7)Ad!rP8wq!!?tA+PcM|R-
zZ?1Fl+qWDaPZ3r*F&j^VPQG+hL>;Y80N%)b-XcQ9KnRNa^C@c%LY5=#8J`MGk%4@n
zExg!uV`7z7rzZKV*8KXa{(5=c{k!Y8?9i}J4Wl&yn!kVU@_#sih!9dEJ2f1+`Snlp
zefNi^TI{CZt<$v%wqEu!Qn2%eIX2#r@$!C?Da9IxcC$TolDz~mAE5CmRqAxMqE{CT
zTn6<O)sc|vKYsma)X2B51IJ_y2GShiOMXVqZRK}aZZi$X9o5muw>|)-T1(Nyh?)yl
zH&F8T;WTdp=YPb_c%mhK8SB!E<&*HV&d3I9Bs>>YLU~u?_%1rK7**YLl7Z7KQBGh&
z?BlDAtDHZMw;Ld!LH3#=j8!qr4#opa_-EzenKRh({xa+mpUYUDQ}W}~MfdPD<E0Bq
zh}J~su2O9(-de|~i-)XYie0j8Ri2i)`6Q_g&(nNO8Gkh{TIw{!2ZwdaCE6sOO|rd<
z<P9&)q+k+p?{Rp>kNfm^lleWFNI_HHI{vDYjk(}Re#Pzb9u{+C$tikIP@h@Xb`}lb
zL=)N)CI2#dlcMiu9p^x!&m#*3L`4oe_2aCORQ-AXKsLh{QT~=oh}Lh9)${6Tn>^Aw
zKu0tdG=Iy>aIt^zwyjiTuMYBG?x~1Koj7@}%v;6h00uShsS~$Ke;~}2P7dD6q2;n(
ze+bNrEBXnKJK)hCH1M^#X&L)Eu{$+g#o)B8#@?frRLL$HVmC2lD$=m>+RoD`3J&_d
z0BrLT6(t{Pku)~QKG^{dCiH^1I`)|hi<+~$bbqw^U6pKEbBY)#MwcU5o`{THgil6?
zD4%+=cajr9%W%7?73MFvuGCxBq2^Yy!xsB-E=gpv=0MNyvmeQk<fG<R?EH9=6t|MG
zh$V~Ip}fz$4@EXxJm@sYa%(9qb(wLJz*Q|RKpaG#^-{WnqxORfZj460mGXK4s7iX5
z<bPvuNbgkomzs9-Mja@?svO#H1LHb!4!Z0OVw!L{4(jWIxXn7(l%x>70jCM(_OU01
z1F<V~bWuSzw*GUyf)+)u${17RNilqTZ4;qwMipge=S$mCHK^IMMfqVxx7W+(Fi5+K
zn&VP^LSV>+iF{X--d6sGm(6T19%h&~G=Ke=;QhVqxz}`HJ?^3;f1vCV%168T(a5=R
zC`n@UN;b<y4p#`@UbPOz5Q`n1Vo>U<yNL?GJEx`2hfP@(xkQ05>s@HsI6}SM+x3Rh
z6bE*5db7orhho>`nf>@py2RY0F>I?#%uNf>j4mM`Vd1<UWwH%HZ~rarfpm})lz-??
zv|8m+iNNEyl+0yV+fGB*B$V<ls8Z}kaCXcl(Bz;iy4($Z1U91!QU}00W3%S+W^7gj
z-w~U0EJwyI6ZDwkhS;nqU%+NH`vb69alC-dTE36RW+(l9!Y#pBQM|X69{Ws(&i5j-
zWSj_EunW-KulC*~F_B(<y9-wUSAUeQ09%)wp^2?}kl4Trq&F7kTYABm;{V7C#)92Z
zXz1EWQ@q17k{7v2Q@T_-*kmY8jJH}l<jW<CQHGJd)Cl=V*1Ar**<aQUR4Tx`ak8%m
z-StS_&m1zW-#bE*TAFatc~~8*Wg8x^x@XKV!W3@0XI$g9P>Bf_CoR{Q+kXe$D*};m
zG3a{BIg_L~-TGb;q}XJCqp3=4r#mj6y6Ysxb>{X#6;o=#HdjAwI(k#74nPKn8$n0=
zu1F7UN7|o>)_(+`tL~8M>~ceSR>tRRwC=1mo}D7kZ*(=B919?k7N67lX+9_8#Rhs+
z6^zcijxGoHqS_KGel!}2(0?y`<frpv<hcX#t_O<!k^G3tSTVg5m;FfMkc8x&r_tR(
zVgo6R_Velc=WdC#bTdn+{fC1%USi7<DGg?q2p2zI@!JI^NpiW83qG%Hg=DJs$4@^#
zuUV5I<%U%mbK%@y@Qu8`_Sk>BBY*KYn^3WL3pwd5u&+81xB1_AqCg_P<pIC-#yhRN
zdW&}fnl4IDsqI(ZPkJWtTUHDz;N1YSmv(|6!X!|PuDqBC(J<oWdx_z$TtTEj@PAkp
ztIU_tOfM9dk=ryAmm)APK190`1t_z?%y48#iWFG?AGo@w=PJ%Ynj^}=aCZBs`*>5f
z`QF&()8XxWcs`M9f{6?PoAb8@WW&lYSXuPk1OVd+n)8>WFfS~B-{lSRe$DH0KUr@x
zg(Q!hgSX>f=g$XQg|#O6{AYgY|301%{~UVJkqrCPFq3gHHGls&<P8UGd`PnG4Wn;<
zer>)T{?KH5W>)UBKWv+y-y5~^rCKjpv7Yswd1AfHgJa`;GL*Mk>$wlsvRKQV<Pq(i
z*a=~x<?|`REJgNzhA3%@Vn)kiTPHSkg>p;HxGZP)Q^cRFi1<C_2*uNiGTMn;{pn;4
z1*cEbg&eXVabiVF_*iAyHtD84T`-Qq4feF+b0H7ZKxS09PFB8LVe=_h6lFQJ+?WT>
zj;kbb!E(-cywD+AiRY|$IzD-u^M`+d9gz(~+A}T{MWMoflEV(GUGo1h*o@z+=0&%v
z7gIC*6ul-Z<CyLA04J&5Gp$a^Y82*Cg~I)Y<AlaL+R<=|Vk@`2!;Qw<LAPK{z&(qV
zrudkkU`{6ZJ}R;$3-ve?5%aZQj#89mVjq1j7|CMgJ-0KIkjpmO^&&}+fNt#NobXwu
zQbI1HwR&rRu`$angP0cMCRB}UKIFB+HH_?C&+<5L1}iP<R3&h^aZn$*SmPzPf(#7+
z?v?78tKY0(^OyxO8g@#q*HeZT5VC~n$n;XJ4z1M-xA*D|D6plCyBooVbKU@W4K@%>
z0=I!R)Vtt{g~X~M?idXv+TZ!D$zz*`&JT7eN4d6tr*t#OKqvMPos{74AwRS264@E)
zC+Y%@MCvpuV!5H5aK)FcsFO~55$K~rGv}H`bliCzqLhLj1odt>K1qG6wa2}cz^yA&
z*YxKuoYU`I=4Rc=9xAz(q4j%D-&Lc85jd&UuW(q;ZH!c4-Bht_t&S@0=u)k<{G~t!
zg5qs|d({d;N-L(lO3GFBg^(eyMcBxLC3Rf%CM8hBid48RMT(L>T8hOgJ<pw`^jN_a
zQzH4REMm46g(LX%(hGAD+E#ja8o2WjyV}sf)gn>ss;hJTUn<7i55u__uRAj_KHTXt
zs$D6L9BytK69o+<DIi)Os5*3E)M1g=^%j$V$k>RQTJ0Lzh7^<d0CasA+iXCqRS)Uv
z9EF2aP<bMCsOMsJ4RwglR&-YUcfbT&g%QaK&Iy`x-&OZVRun)wOV|-vQEuf47n|#-
z;1zGa)_uMTH4z3AlD`QxZE;*^QjsO{HOB|xA(YP=8y^IRFy$KOZ}(XJ-dOyY@8QsY
zz3@knsrOC`x(0tJNG~LP13S@SP@#yN=W44ni7_TBYTX5V&^trWf6t&h_Lbj-adl&u
zRBq*kzCEA1%AUUh`y1Ca^vbP0IqnP{JeYQwm66R-?n+)tJq4{bZy5XrtROBnQ9d6O
zhlRd^QlW+t^6<7YWsfV_MJi>sx<~GRs!}7T4D4w)4G^GY)0DwvF0o!~H#Oi`hJ3$H
z15d_-4uP(M-`F{}v>{|;WogQ+2IccfD(_l}z+}7~!X&nXosuvaK+=#X?EhPn$vx(`
z&i7H8dfIiquPE*~-&d43=KJdOgXa4npG@}Q^L@qj&iTHg_FLxrAZ-M2$FNiv1g|QX
zy9{%I$$*Ckj`)GV;I1QnMe$~rUNtWm4AY=rzopat3wozyK$p=>FB1VWmw|#W6qoN;
z1t^zrA3X$@)-^Aae-EBdf4~26&{^2WOVD$YTE|Lgcaj>Rz#^FuZY<pSB7e)>ob$TL
z>n5+qHIas0;`(^;Fw4ytW%+8CZ!HBeNwN#T4iZdz1~sdoxsu>~i53JGWeF!k$AX^C
zGCXHKki3&@tAdT@jknk8UoY>6e-FLdNXppISn!I8nty&Bf8qlMk->!qQ3c@U_xI+<
z;m?|C_FsKZ!IzAzt>`i^GFoMfZ!EHMI^6G4@rnek30Y|tHvP}<pG~#$_IlMjc;!9S
zlEAlk4Z`#49;YZ^y#qTLp^){tb&dqSW-iva)gkJ~zD<^R=WvSoq%WrjtMNExfXxaL
zka7lBD5#bAe_tydX`pGT4p!d!93457BtXzY%_U}ka+1yp7drG=KOG{(PC4&tiu!pf
zu6tTHur3F;@m^W5y>xxJb(xg{?<uZ!XgA@3?#Vc-^OLwLuI%39etjVfBBC<#u|>|7
zf;(Pwx{%6cE|ItG=|g_Emqd276rM6VWTc1@*3wSce_>7t#~gy3t>F7NxTK#rp6{_v
z;+Qq+<VhCcEF-jQ8No(AbXwv8Yaf>KX1Nld94#oW$Cw*(c{0>S_h|L)^>$q_LFqs>
zf1H3QoPehy#fQ;<qXwW>hHcGF{I-`qh(;OmHHQ@aoS}GoO<U9}SjB8UQ7_UU1#D!w
zo%IkQe}v1Y>S3f1<5&-u<LFY9O8+UoVwQ}9k&ZbpA1_kB=djkG;P<sW`iw%{Rs%zY
z4)$Zc3Ii(<y=qlbZy|=8?^W8OmVQjGpo5jS+`mkhyRLr`Efng$pm1&ylVTa*5muEu
zFjf(Tnqo2)mTuqP;dn?5ZABUNWOpOIxgy8)f7>*YqwTn~XbCS_re%{Z+EaUl-fys{
zv?xx<12qsCwM|R8^0va}L#{|mrTT{DSXmtDkowXo>!l_SQgPnwQzWOr(MWO_`E?Uj
z4k|?G98zr$dveiqKWtIj#E}(ujtutVE4O2!RRp)5G@go9RDk^Yle8PGl)2bx<BVc}
ze+3KgVz*eRq;1*Ja6^iBpV^*@vbjA2DrwSiVkTB|gjc(IPwrrE0%hEu6``j1#gQtF
zs0FuR-&;jigpn3gDvGh*rsJga5iQ@TCv)w0uKDCys{0`NdaX@+xTx@XX{;0*EsHnD
zO39YFX~~qAMn}X_J89GCXp+l~!>p5fe-2!6`J(X<X;K>Ti^fBwgTfHjxuy;G@^qQr
zsm@)SMni2uucVkA4UzO6uHJzwPrTPN;-EdS+rd#n1oaxp@0t0a3(CP*+uSORG|m5}
zay@I$=XEn75>77E9PgS5p(C-CxjhrQ&Og>oF6NWh8|B3_6<2*;*yThiTkkS`f67<z
z(NCVTjD@w6XNq%U4+Q%|H5wyc!VOTp8IDkZf@uUtU{jT!g`*YlgM@R_=_&a6$OPI3
z1!B>P+YHQrb5fI`-m%XF`d;X<VtZ*NmbJMLdZd50Bb$9H^cX38Tywk&dLUTw<0|M`
zaUQ+hF9>)r)Z7*!M;sAGGXBpYe+Qf|M^!s?hhR_u+st7=-)>xrsPcus^2y)$>u@^l
zP0^UX_;j!KWJ58>_NyCDAeM9tTN}^y3iQGbb=aW5jj^!gvlm1gD)3dGJXsNr#)wk9
zT!7tMT96izo!dX2PH^1fj##dS-&f+a*O?}uGT`@|KZ<=lyNd8^d83&4f7J$>o}o>t
z;N`$<24#C_lMkyfB)oG?>(m}Hp)TP2KwmdJ$Y+<z6fgOcEEb<S`#M<6VeY8R6pw;6
z<J5lmYuY9O)_n)Y_#K2tPJOvw(mb~u62@>Hv2?{O{eKm=PFe|CuC*7X(^ZFSh6E=v
zuj(i2yP(;B%K<%M3T19&m-df54g)tbGM9nsJre>mH<#h^7bkyPJ8vU763+cA^aqB^
zVijM2A;8vH18lYfXNQXpw-;@2`~QP0KAVqjNiFvT7qlg@$zt)VC&{ur{pTc4KVIHH
zUw*qHovd&gtU7)Ec>;bR`7cOrH2P!#gh4oc{&o7ttCY&{^Ue=RiO7%T?~Xt9`5S`c
z9-#T*w*cpFmLPwO;XbI+STt!2axZn^PvFnD(9Yw%G%`@|{BR$@^FKd-f03El%9%o;
zjXxPnf@u}b0%c-Ov<;lt*YJ&7-a~8QO*3U;-ssBTUH69hBVT__BYCN54VTgq=!Aap
z;nfhYhIFDE$GgP6P}(_ll7jegsPXhSuV-%k`GaNQU`>A(+*c@G>A2#`Q$Raoc};Rr
zI<GRUD<Jcc<PcY#5eSGyPU#%P6c)kYwUBQq2Wl&{G$TVS@-TW5lr{pSu57p9Iu&NF
zj8rbl8=r(qxkp%z*5z$%O)W{_`mAER%QP{+XJ9CdHkEFq)&A-^3NS{P-m-8hQ+dmR
zj=w7(1F(P1JVmWG<_R<_hu1_qes=|pX|1r}RX8i7VjI7c*I-OhNiT&@T%W%|k8DOR
z&KWeKthPv@S#glk<uC=WnnEOie9@GGXMEHnXr+ELR#m}gxE@k?MIbEMk-_NWJhe(q
zNqvb_fj<j`Yy|w0S#zOJ<Vt3rVcL;Fbt5g4)G2@UWlJsZO|$&B497xq#h~)I4TH+`
zcN)~fL7h1pR^hh@u+E<iV(Fa~a@BF565y$z{gAP!eZ~?-U~4F8Je6--IOkD5%MKPq
zsYn=FZdz4_kMKqh+K?%l#o5&L+KwC7wX`OkD0I%K9XYw!*nc%9s?K?(e!l|!RjDg>
z9$0^Etj@{v@JDeyQ9nVRzt!+vJ2CJCdWba2M{rzL>|&}io7%PXD&8$FD+f^hR^JpK
zDYuuXOCM8beIMx*GfD4@`5%}v@N40n8Nocxqb4Wu`m{$SP94aoN!k^ds!8mF;I(1N
zcA}yRsm$*&>cy&NNzS9(;!m*mU@2=X!jFHPyybaI@Y~1^AF_><H9lk;m2SIsU&Hg_
zle6C`>Xuhfk`;S{<28S<<j8Y`@UWfW@eqN!?hz2rM6X@^3ZpiIKcOHRi7d`9v%Ux_
z5!=|dSTA^^F;Oe|Ub23-2w>SVxZ?N|4A{dTVX5xAg?omw0`=QfyIc8=wn`AVy19RH
zo1Q881#f~?g^wS@uaD1{|4wp2MFKfn`PW}B;RQQIVXQwX<a`|T>F<A^{(SjkB%AG4
z3dv*AA1#vu>RNv0?W32xLTQZF{R)lHym(DsIEmH&e9uAVy5!Lc28R*VWTwcw2xLbw
zP%NIfcxDcH2`dVGmOqjGuH|R5p>%%`x4HvmCD*r~L9;B4y>a4$8(5a|5(9qX={}@)
z@l>o%PVc}|vdo!0;0eM+an`Qo=L4AcvnpKDhPw5o>^d8Losc78)U<|xl3qiPlNTc)
zM{E4pEmz($3$QrO2Q=WbgCcJONZE<v_b*Unn`H1uih316#X9HFOj;n1*E4^q*&nzg
z?Y5^_LS>Q09(%ROzTJ!z2YY=)?<Ws1P{id><-x$cDz&1AJ@sFw_Z$N)*m1r=?`v<z
znWBYB;&T7Io}<~MI&j@|rN})MO~6EWn_wD&dJYOi&FKJ5Jgu7-tEph{W+;+*9c*Ic
zNJxF4h4lohf$66WtS5MYeGPvEk2SBJ_%FAve!4rxHOTsD+Zq6byQVcvoD<BDSboW>
zq_7hLTQ-#}a75tOn3Pehe(Fe@>M5>lo1V=!2g<W``LcGUvvsvCx6sHL>NrC_oJt$J
zhfq+`3iN8E6jkT>_^p%^#?1W5mE3iIZZ=Q_`Q_Yfq)6$yW#f7(T-JXOP4aF0&hw7B
zx|3s(^F~;to$EyD1^LP5j<#rGZ3jBiqI-&KyZ3l9wtft<TVmn*l*~BmhSye#!zZ&n
z^u~{MK0|vOIBQ%f0LvPgyOtZcD+@x&8F$|!aL;czCzP62z3}>FK5Kj%X1FZW@9>D=
zc#fIJLr+NS<?3`rx0QdMTpBTIJS~4BJY7p0REoJM&R}&Dlk=H*F`0aHUCs`}!B=uh
zxT@~_gnMJ(+$HA|*m6f7mNIRBcks=bgaZyREDUx-i__k8Kl`Y2?!h@Jg)oNhj$=E*
zU7^z)gbwspJ;CL`a;I|kcS2Di)g9m0T?svuPw(=iLDdh75@>&rdmf5=Pw_dRN0gf~
zizk4-Jna=wE1?Ph@4NzXN~f3mnuEPU*u@y7?|Oyx({UCTdz5g=V4R7`IcrOrckRss
z#ShAd8!_k-lXBUh%RSGa#L+A^YqykfVhIM%I*%w|ovf-=2D&7D_LZ;A6&8MUNN<)T
zD4S`lW5R^~*O-4yo^9fDU~;R(LkX*3VDjx+?|GCzYka+R;a~54cHh*1@w_{pZfXoC
z=J)PthR=8>v*FvaJ5b;SSU6rU&b6kaDH@XR{hcB4ig{5g+{J=A@*2mGaAZfy@O|<{
z<2lzb0@uUb+A6?X&bM%a5{3)?{i_%#Kd_2nL{qW2|A>FW2!>;10dC%hjnk`qhg#G@
z=FjiVHGX3`I6=1!I<u#ivY_@*>xms27<SnnN2)7;f9Xty4IDTQ46~>e-tT27SvWnR
z5vpwJU^S;R%c1CH%{|?z)TllRJGb#rz3szw5!*qgI`6(x_g1_Ku3Uj0Io|Lwn*o8N
zC-T;Pq|$$v5=?6jB}V486Kh{XXSy1(TkGIGi|`zZ1`bvf{#K!`GdNO+1)_=nQsNoc
zaid;jmpQjpCM*7MYrUZ+i2F{6Z&BH(EMiAwoE@ob)UJR6p_aWNw&f<F?lQ^lPMseu
zHH4=8swHBju2c1;UJKP&9hv62oWI2s#+Q~>5_x}?riAUQ`o(b%2u2L{=ojbEIHNeA
z_tsP^xw%k|7Id1Tn`o(R`L%3nyqcrz_P{38Q^s0bazh<drr%cI-mJGArFD3zGe|F#
z!?nAx2Ylfp6?Wj5Gq1;)9xP4rAeOj+Z84U6xW+H+v>c}fhQnAr#CVtM7ezo?kh6Vf
ziM&+Fz*?#9s@ochrgPz2HlK3PR+O!jEz`9xA-wl*BI~s+rS{9cHbu~Fp<WSJRLF(5
z<^WJ_JK7`7JKPzX5z!siO~6XP30F^G$OFqK?`!@CmFSI^;oK+_0yi?3(Mtsr0x>w3
zfif*8moSqE1ed@?FP4Ah{O8+mCz*uJ_adQ2;hpKMGlFO)39k|-PnF#Uh4HT1E%nP8
zw4x740-|Rn9RDZzKe|X<bY(TI7#~@Dx_Ov09Z?dhVI&8Ijp_{%4F~yiqB(yYio{p7
zuD#H>E)1e@*3F8;XLx-`9C}=M*-$m4AVM}98;UOCVg?fuFB5-OCDuj>J!`+7{JLQ1
zMF_26rWcF^S!M04P~`M;yqUsvop+GJ7+XILSz`3Ffu^)T8#-%(a4M8QV_&-_zU+-8
zgt-;plOkC5aK3%1{`>rP`sY-WN9P1l?f&|4;x7!Ku#UJ&;YgmJe|O(bzt>cA;L(F!
z`h&9j`Ms-Fz9xSp36d0))*?<7v;s+YtDFEZ<w#!pfQm}J^olB&#BMsKmLgiAb*L#u
zGqHbV&$}g48J08uiQ-RIzK&4BLqxrbvXb+dLI<uHfu_54BL^+eQA*MhJ|~&BO}c4M
zi2~@L_S;j^krR2)9WbNFb+Ynxhs~#46)B4$xhW2u4NiaEwXvKLRy8`Xm2l2_xAT*g
zoJ;U{JEGS@Ie(yERuub`kq}WOgiQC$C4TlJJE~*_>?I;Qvic?i>q|1b-($4x+!J@-
z8!Va<FeQ;%U3i^6axx6<YK$YV_wzLbX<9dw+&o|7grmgY@XgM2KYwc|`C&uxa}{iN
zP;^>oFnfPBA}B}ekJN~PjB;)r3h=l_6yWG>;~VqEnGRm~R{fWeGV_Uv0#$oj)?^RM
zXcj2NMZspj8Srf}4WpUn6q`%K6qmNaw1h8<6*MXkO6pBCz&?y4dS+C3P2rsOnj}qJ
z3+Z|($PnSmCz9)xP7zM3Sjc5KPvFOh!R0-`dj)@v!nHE52)s}L6SndZP6fW)Y_2R-
zIHS%*I#f`Z#7?Ey-Nlo!D9ADD;z=nH$W_*hr|KRHrG@m75|VLLgeY^o!}N}_X#XBU
z@*|j%Mf7@Ks`h(8KsomOMj?1zOcInJgacPUp+++0-Xh9nMm<OBDzl%BYK9&4$z5RG
z!|#87t+cm7x>l>pV1469@^O#6D)W}6yerl)egi=lyZ`L?l;SOxrbHK4Ed3li+BMoS
z_5Xv*fxepAI8Qrw^n!|EN?|R>juTyOq&=0pzNVO3%T2A;J#f9bB-Bs<4d;6(FEpN1
zkObG9R-~jh8eXIf>6Yv}_|Zn=%T2ecsU&~2ZF~x-AX7tgo#D*&_Eyf0Pr&pZlUDU`
z=4tjZ+E2RHPxC&@tL}L^m0(I7H_DqP?G-AQ<XyMeOQzLmmrH*Ue-$y+r!~{)A%||F
zWTT;#AIlpN;5>|<qTE<K<opyPJR#lKJIeji=s$3<!zf|2x`8r0!!scJ?5gp(wmN@z
zCGL4?P*IwS#sjG9bFR!3&1Krq#5fEg+M4)Fc})BPdnskyiaXCM@yZCQ`XWAk0-M*_
z?r@7!fczb9)G8DQBt*q2Sz{6FEC}No`F>&guiEZgQQwucvSb27-5zoxuav1mDjl<}
zLTMC6{f@CiALkIs9d*#cLm5J_b`O6ZJP!aE0Bed5mYvENLDfcGc7k@oc=I7cI~_|n
zIB+;pQ9N*HS5e*=+Eu$hXlSQR5X$Qh9@<r0-#N6asO=rvB^#pdhCh9)K3*B0Yqpv8
zLXY~PYh|AUP6aIal>6vYnR4Y*xdYXj)=6q&?rN&rr&A&R-Ax{PVGP!M7bbr_$|_+)
zqdQ(7ZZ2zmAD_bI=Cjts&THZ|72R-zCP5LNH*9YWgx%GpO8&u6?>rVQgD{~r<~xFx
z_#5NfE~g~c*OF9y{#5EP@9LC$WVrLup*}FDO$i=Tu?^MQdn&fOha1UlrW%1;^_uDC
z;xq>w6qUpl)R8AuQ5@XzQAdAQNXS3(rZq=*cRBh;jl2~Y!k=*9$Q!&DxZ?+~qetGc
zYu+>VrW#*>SO1H#_eki4c9MNoS2BOl<l6@!$(ic62CSe)_2{`bP|pY<{DQ`pS*0H$
zhxdiZ0Hh=+FNWIlz~MOXZ+>Vi1}jtK+~!OlcyKg6t139E>tWxJb(w#PB&1a(w90V%
zvKG&)adB5I#jmRwTT3xxR>sGijTEK|e!Q}hpOm6d#70-023R=&N3UmX<tKi>i$_Ma
z^>XL=4q2&N3%aogzl4XreXnRs&Qx{(2XZ<p7_F))tC0`)F9z<DGY~`(oL1E2t=F96
zov&s<*&|Uvp>12x?7;~+LeYN#XSiDXm(ffw6_-&pF(sE9uPqCg5l1#Bf7M!1Zyd)B
ze%G%UpfBlzc844ehb#gEPEw~S6u_`kpsgVkl6*>($~xh64&46we%#YachSi?>&}&8
z*k|r8cZS2^eEc}GjS8on8kLf`41%PhB?L>=2f>rFLXezPE$&qOgmNv(BpxXzuN*p(
z+<gy1O%#no(nLimq?SlLe_j?#)MRJ^dfp@?O`Iv-s0~sb?inP<19wgwHsPK_vWz}7
zC&r)XpCG9us?=g2VH51(BJl={X0d3mL9crBD`-V)*iz9-v!P`?b&jX-K!F9(2mFEF
z<6ccgh7qZzQZ?RfITg!89}SYkC=K_#wU9JPw(wyqNy3qkBnh|Be`_^~<cdBT^t2B5
zT9USS!;(0KWSnXqyo?7m597d3rD0O6Lc#z`IHK|}EzKbf(*b8!bhp<y=TvJM+G~(J
zY{sCg4U=hj!!_)Ka}>@GiISGVX82yE4eb?#z~EhPNf+ZqQ)*52YP_aS!&AhJWAeme
zx^oVz9EN8>D?Cgxe+CE%&2x+hGoM<KxXgGe;RDY2(P)n|RGN5<1MQhbD=@<d(+#O5
z3?o`0Nmeadp_)t??ljDisSfw>MkB&KCP+2<Km*poNW@e!D~dj#Vm8>vC3&W38BIAw
zY%r$VW4a-E@*y_D?JnV6hKg4Wo6(0aTIPyH!J&sKb64Kee+*j)ei@qP5Mze95JVb=
zS3-epumY<kmpq6Lv6ko^-BE6_bP(nc4Q{OwOArTVU}hn@?tS@X|HFSSC;jt_i|IV|
zUrlFkC$l#h3qK$Bf9b#Lzs3c6KI~sjPUdfjxvPj+Onyg(`;|L~4Ml3FRu1?2*YA(#
z(cr<^#lQRKe>csa%+ID5{p<d3ufB_aU;H$mUtayG?|(d-|MdR2JDI-ge}4b&a{uk*
zm;SHszkk{NeBK|QPmlZ4Q&T@V#b}R-PN}?=aw=ugChhobdn<gJPsLB2`m5Q=_~suD
zRBZo1jsN)U*&hC`jB>@q^+Tq`#bU^{bP~?@+%c9de~WjHSwtXr-pXPBJfZp5{kPK}
zru{eR3&w0$)>1Kr%@rt?Va2bF*xc1bxr<dix|&Ek#O!0c`s$~n%gNQk*|dMq?AF!a
zyg?pNjO$hZ_ka8|1)?F~FW#S@AFfqEsCE_*%Gwpg>js*{UDjpI)PFrapUz%i9-T~5
zi5F7;fAVNPpUf^2FE0Q5VgBvw`O$omXmR7k^kR-aUM!Y~486RFpk^tH3m_hHfGsX+
z_#TT47tfycFK5$}*OU31{^d6>`X44A=7&pD{PO6>2@JZJPcG(Hkv$fTznWZ4-_K4a
zSBn%Bx4xgeJv;hp`eBFz4fJz9#QQV68L4P6e{~TuSBqpibiemz$fq%kA)j6gItpzR
z2HOv}TmaT(z)Fq42cmP8Auy3CIsoctUjQPlbKvQ?00jLDJ)!tF#J}4Ppr)vH4cP>X
zpmWd(A?kv<6oJS2F4PbR3nB5EfUkg^;Hquq>OzK&_!ic=c9|a?)~A>Avv+6zoy`7q
ze}*MBJH;9w9A8QH@#E`U{|NwV$Yp^6BfvUxS*edb=yyjav*{wh7&&Jjzg1R2ww5Ji
zR3M|SL)sOs1%9ekEakF^W+Q|oZ=i`q4@};qcj4SM-$WBcyVE8cXtGl$c@s_0Ams73
ziKZgUfQ)^HEVGhWR#MIuSw{U%meKZ*e=O5zDa-hEDyexC=vvS+3S%bHF%zk7pLbBJ
zA@4}iAYiD7fN@=rdewqJ(>rJg){bRdQF#<FKU6+c)_~!sCO_j2`BepdqCf!}cZS*n
zC0U1%xnl#0lZ=p&syQlz>_cAI(gtp|vv9<<5k*_4`2Rj55I!gx40%{lbD-8jf7QHQ
z21Y(I)H^pBSg0#3Pc<!=A}yHjg8XyguAt@$!Wl&&!}T)M{kSjk|B1oq9Bu)|N!PHl
zP)#BZ0FJ?f=<f8R_z{Z=vQ}>J1B&Xcrr)firP+rS`gA8BGze_0(rTtuyT2UyJf*`D
zj<3M|r-HiC9Z+xn5KuS01@$%}e|H;^yN$@*#u}oHHALG*4WWckk|c^+B8upNSSD%6
zw61GP2$X#c2z6;DxhVS%V3g4}`K=8>8L*b`GlHtts>QN^UW*p2tL+LI(73L}mgn_C
z;sSav?a8hr!%Ct|btjuB@VU3-rBjNHTL#6BwS&~ocE|OVEixA-UmYvTe{#z^nExP?
zrQFz1a|bqYp9$8Aw_*+SvspE6I{<P9lL?k~Z9zeKXLi2{Bp?$C#OH<#99K)_U7(uf
zP5A$c)<^mOI?eU`e~l)`Xxui?BnN+Cdo*6%zZi_*1wg(4(2>*lgyTtm2vWP=L28!=
zAa(4=-ypS(C~P(gBO&*Ze~|k~$el-_tby#!fvN(iVsm{2stTY2^L}ig`sHYLc65Ax
zz3W273-C!ucZg5Lv<vSsirL2CSgxR2_3qL{WhHI}GhD3{#`4>aW0jV>db`P6kdv}K
z`X<-O^x|qhJ3?N#_IU=gCZZjBoJ9~@sH^}EaTFLBiKV%wsL2FQe@mr;cOxxJ)4Bsh
zri@;1ji?^n_I@Ko8&_$>Ewe*YWEW{ckqtqxM=N3<JjcQ0eO5HqjHHP7z+-?6+wQQU
zpys9JuHLmLx7s;Q8HlZ{5__+|an7)$Fy^0%dMb`de_K<KsY<CEXcmNX?6z;9i7Gx$
z27f2JN=)U~X)O*xe;<(6$ZyhG9O(|3Do%8-g)#~<3Uw586xxnXGk`l4m3*9qqB`!X
zaH>GumUSHo)9j!z8Fc`1E{YM^MJ*!em2Ry-ph;ZcV}+sUi;4(Lm<yIZ)g6Ez*d;o>
zv7yAdO1a+#z%K}~G~r?(=w^8*MpTTv+!!Ho)IlFJqJR@Yf4gSND4+_Uc)P@FAm2;i
zSIj$qYUtN;-9Ko&<v~Xr+y(C4)SsF=R02<pWN_8}1K76ZsSX0D`qHV63D$jmTJs9v
z*~gEr<h1w(=C{kbRjrSeb!#=Ftjn8dE{G{_qFEJ*Z4*t>*vQ;O5j8s4*6$`)@||=3
z_=4vFxsuybf3CDK{asgn@k!A_;2VK<aNYtW#sI)@JH${Bni?k^MyP<+@o_`#lf{Gu
zWYmG9w1eZ8k&(s;14sf3cTMg;Z=vi$O+bh%Ie2ij-Q*;=z%fl=e<-e{#kLJ4<#5o0
zjrath_R^uwxdm$XNRE$<T^Hw<3)J$wf1su8q6!?*e;PQaF=9j;!{zaj6R$a6fu1NS
zQbf+?HKQ2J%UfdcRo+MIfn|l8oht*xW!*$mVb)4L@T3n9xbPi3D)}LJRBjIs;*-N`
z(DChdnFM8SUfN~C|24O#66I7tLrBY(oQtrOmp+t471WvWQ{!B~E)Y!dRU%So6C$#r
zbzNZ>7a(T!O*F~i58FVKQLWV5uLJ)F*;yF=m(ffw69O_cm%&d06_*060w$NOS1*}=
zj}4Tb|9t=LNoHY3$xsP0elS)Grd1>hlr>92mMLj_2)}sb6h_O4<~f961k_2VXMN7=
zAgv%Xt`9SPl;P{e!4gUcD4cSwEW|s|OrI#z54^(fOOz&2T*e(q3B}ABih&Bj?VZJ4
zn7k9poX=%<SNEb21Ze6wSClnos;)wRW_cfqjPM3KWP)6HT@GUf*=`|ZH64=}lPZHK
z`Ey=X=`~TcRsyANIm!!zOZt#`o-$2(#$eP5dUMDU6k|GtQCCJUd5x^)8C8~4dJT%v
z)C~&h@Ds2=$?onuj#1X&7!W1b#n~gpKwXW-UEUPT&J{B0jm*xW3QgCVbQ;NjNJry4
z7T(!_@|+}zonbP*+u-bw@bB=}nq9$<p#`!Ow&-2e66>PXII1SMk87x&A0?l!F36@)
zQ@A8b(FU~-yD3Q1<k0$9)@^kEdiH3?We1haP;<tY#5stg)F|mOqZzb*87T*T$Ml(T
zI96ODf|xCVvY(9nYfGAtN}iX0G%X8a*zVlUciiJtS=plA6$<cr^fQyCmc=e@7u=t}
zz@ZKBqg_c3f#LKDU*DRi*Y~G?IsvppVTp#nem;d48b}!H2T)dMr-xsE58t1DZ^`EK
zQ3}ax(jSt;uOCA*@|F*LFWK{@G0bNcYm6Y`qfB}ZzoYHzmq|zEC_|Wk9n3j;^%VIC
zP|CLyqy8{ibc&K0D;?x4f1>!ak+&)Za1fJ|hBC_KguzRptZrwkP7WSm+mO6Vc+E21
zhV&Mm<g73T?!Z&B&Y3)@0Rp4QbvE*L8}o)Ms!d5rkDL|;yFCPyXbC-z14hP}9dXPU
zk{YHzaW}J(TCW8&YlM}5(|LDl@Yz8tq=j|#*;<)%4O+P)aci}MQZaMeQoXI6nr9jf
zkyO3j?T5%&LdX^x304w!G@KH$!kgJeyZ7+5Bs3?N?ap5tjom@fDd8=2J2A^UaZVtb
zekT#ndw0WDAiV2QT-LghBv7j*-dfgCi?T~h{!vD*_y?y>JfquxS(U5aiu!yj6t*+U
zm^_3#;Q4-dI6WuTZ5Ut~9&LG6&mSr`gN3BzoCBmwZYb|JO3m&lJ#BOb&G&20uR0Sx
ze=Kk^R*ns`ySo+FhnLCS`N@^1Y`Xc0O|Kefvr116^|)GA*qxGYW9n<uy?#u`O3Lc1
zOHxdb9??mvE);x!_B;;w?(3!7G3jY2(pa6yx9GBWnl40z8@*&KY<s<Xj?w$(4KG2}
zM|tQJ9gN@sQy-(me3PJ;T@q0sXAjNnM<Qj8Ee%^r_5^ArBINkdus_$ZJOYAY>mul-
zS(d_R;Uw(=f>lC+?)bXvC^#czk6l24O>c@hANM8gz3_W~-X9u*Q%rj_N^0kYoNAgk
zqa>|FN%y~slFN5S){SyJtjIVMQd-Y>cS3RL&6MP>7OONNU(ReD`J((~kEXkBWQ$s@
zL2C?&quC>tMhQ+;8Ej=Qg|l>GVmz;eB=vkum=w@Yx65+C7V8>$j+e0_hyIyjeIuU(
z+VmBjM{5#)%7RdBMS?@Rhg$`S<~tT2R$M)Mrm~^-YxmW3jP7u|lzrEIL0JJ-wF1LS
z3?7Sj2F(DHKyAOsIcQhlru$!Nn3Vc%b(;GLgcxOA70Q=%vfAz`%GMw>;9^q+>`I+B
zNA_ItjdfL7)kXu_cAQ-@7CWQS)%Zq!8=5ej@&|irWWTE=f3o6s-U^M_Qr^<U2mwSw
zx3bgM^(Knq)ylL~R;t#DzqMVhTBDK+RqH&uP_;${Rn@v#ovYS)bfIdEu?y1Hjx)`s
zcAThetsNu!L*@FHLUl>s#r~?Du2hrr+p3bhuxmH&V57}X_|!H(@nXxaaGrR`&Gx%)
z4FH00tYz~Ff1&75Y1dNfVkHz#Q>A;Vz-LXm&;!%QCe?sI!{i=J{`K)DBx*kBI8AP(
z4+#yO#$^3<Ou@!f3ofz#Tb}5;U&&ChQ!SUAjIaAN39GPOwcjCC5R8Y_yf_#7;W@Cc
zwlVo~<a+2jXPZ>D)B<wFG-d|5mwE}kuwJjIcgc*Af0O_4U1pAIhN(ohdZ=mSfM{D9
zv9&ce4O57`o(&q#Xj$4;mV1rz#$f||L*&Dz<?>+K28M0ixvrHIGA%Tp(n?~p;k;e%
zcol3Y?$}CdDBox$HM2jdm4rlmZ#AFXv}m|~u$9zM+uBOn7e}Q8fctD5=UjJc{xnqY
zZvNbTfAJwFw5FEtYRe}zh??0CwR}LJ>~!~$mXGDL2s#9tBIbQfq|_E_DPFpG@<@AV
z8{(xs)Z+AbPXYO^59GI0%j!y5sfJluor`<d?O6`qE$(^3%qi|vYl6dKP5xo$n+v(p
zDIA?|K6}SfFIXoO?s>qs6hB_<1!ILWw1fHDJO8Ml3Mzy?*s-7rAxQXD8^?T_7kf5s
zrFm2B)EVKX|E1bgJKf57CEfQVbi?<-8fjCx<OKU)YEmIam(ffw69O_dm*HzC6qk)T
zJtvpMUoV$`MRSBc|NZjEi6>?`bK3v(U+urAm(zcz!n#xpwHq`VCb$^>`FZj;Kp@k`
z4N@{EVEFa#@cr~>yKAP^d<R8YL5kv9BUXx%njzXjYsHi?Sy4!W0x1Sb5Fp`^E+C=d
zYa9oi?vWDE#W7Bbr73Z)N%O8z2@Hr_`S|r?XhxoYzZLl<nNB&`awNX`t5RuML8@`>
zd~agUF(yvYa}Oius^gRfuID?=#Z-QgfA5xq2ry^e4mlKe3|wam&uIYQ<9;3pEmDh)
z2Qg3IV?kg?-`ihJ?9J2n*Qn7lE~CCwr@N_UfzX2KuD1oxQq2K*@p$JmA`O#56;3@T
zmtZY_I?iT>TFYxd#5EuOz%|4b1=$d>Q6w^Xc0i9S|3=zT9*NdUZxOnzLKQ95jXr<N
zXS0)1A~~;KN~yM-$I8M>dQ@C6P`afMES%2@o@F=@@*Evx#7<`k!mhN%5EJ)-<Qt_B
zwTfs&-ft2k&+HFmx)UJiO=<S;dK(@U?F4*(Os*o#%#>|fVbffH|1pAIDmBYv*iV;#
z8-bVjF~J;pt)9%iruClI!(l)K0~AGdW8~Mwh%j%#w{AT1=qNr&%v?-Vdz82o|7nhC
zDZN(S9%N^6U*T+v_!2xbZjmXKRDlzxoU!_N79hv7`{haj|CWZ03lTW`8^8S}ZOapX
zsq4hk?M!UIAv0XJiH#A=P^NY}v0>Z?x$uE0_EjU#DcN!SR<eVrnHd9|H();3-YQ;D
z*;-{GeWIO|a_x=>q(Q=pkij-POFkRADe0t;4ejjNB<JFTE`i}G@wBeeRedTTG`Su6
z<YaQ%o<ffBgaXxS<avY6YpNkEy2&YjNR-17y>@P9)ExOX*~u`U%vLGlWs<y*Lm5AU
zG}Xw|u32Yv8VWVMG}<}9CKTG8RYzjH-tm|cA-ck%SeZbZiz~mePbNZ<CdS_jW|;RG
zB)76OtWaC+f#$eOMw+{GD>7skrjc1uMZ%H*U<%1?R14->EL1@`=J=aqD<n>T!he@8
zZ*E;;&Lp3R!C*nu%$O8Z6pX<j;V2X(%7>J<G8-A|_+4OjAvmIB!nDjjTy?Gkq&uTe
z#BS9BxhmQl&y)p9FXRUm>WVx)%69<D2#`+q$JO>|8349I+^X%hW{x~)v%)uBGJjwv
z4qvZU_8UUPo(fwNUnq7+q_ksy#KkU&Zt@)4QQ44_N_){4%$}`1Qlsx}fZiar_o+E>
zQPvs*{dD9;C_~>5!454&S{O0762g!U510-`Z|XfL1g13fC2#gQSal=peN&#<Y}2Zr
zuu`_BeTrp;y1=YO^Grz5DX-8Yf=Y=nb~GLmwZQq!@Sr3x(7MyY`srGK>_upx)V0%V
zDt2%3Vvq?m?aH0_U*Zd-q1`)47y2Md5z@wnv(H&Vd`Y^d({Tq<P0^7W0m)t~4T;)L
zvE*+%P-Dl&&n%rVnC=ig%sJwe=I6FaM5D=(z>PR6RK%ovb`+i~9ZW-7iOI+3a6~M0
zTaU}(q?5#rJ-%O%F%Uw3G>T1mn>4_Ld!M{5FqoK`#p%slO&R7KGz@-Ft|mE9@WnP)
zLuAH@!*ewWib-zYy@7OGeg>ouD_X{H<!9)v*e-Z|dREK2KOIw?g2)d3sjy|zbOOBP
znG4k-WVVing=a2QTuZ`PtH70~S90aKM#^kDKq|>3xZe<vHRBt909h~ibU<$AzW}lk
zU=0wBe-G*C^vOUB&EkF;8&o^X-w_%W-y4uo&Hr>@ta1OA0;*uCBJzs*sd%Ysqa`2-
zO2)<%_%}jQ&G-f=)eAlyORM}J0;ZaZa<v&#6|o6bRPkAk9bleb(S?aUl~x1SWjAXK
zC4?PVk8!h{`xhI3-qc+XK0DTC>47?!_JKN>t&hIr%(Y50QAqDNq}?leV*WwbWUhZm
zy|fY@>LVX%$_?*q_*b*6#fer9eCGl<k`u#L{~!FU8yw-;K*vSVY4whhnbGoGZ$fBF
zDaKa*Cafr&efc82AZKj#*3I~8MqHPY4jDWAW%QbR6SNC|pS<MYCKytB0}?zn@kFy7
zv=|HPN7`dIC?J+k9QC9NmZW`1oIjS4toOL<RPS-5qKutx&v&ICJFDN^|46&lUAUgc
zVedI)T%ccP-r@ZsVH9~+-BQ@_e$jh$d<3P`O2UC1AHg*P0q^7`)>=H)GCn?pJhH=+
zn8+QTil>Dat`6zgl5eSadSz3mX|z1Ooc;%JBHbpJ(M&HB0x~z30jn+(0Wp*DLnoJ@
z88`%&IA<@D0~mjtx6fxUrw0U>rwOl@(M&HBmyz!%6#_Ogm%*GjCzs1_FP(qEIS?NI
z{`%L0EW$oqqTG_yc_WZEOKOpmLCGvqGRWQ1N0(>hi~{8EPi3A>+#W6gHn~_UoEhzb
zt1am-vpoRb!FgpTq7|lR{B$9MfhQ5-qt&M@hDG+qHLO$~!_7#h6w#42>FcF?`uuwM
z{ZM%-Ev$=!MVne={_WR+elUN4GWa++C%hg0_{Z@5@TDhOcB}tRP)nwj4po^bsGP7i
z3{Dy0@cYdhhy%|_&q~>#8Yrm)EnJM1vdq+s&nwNPSjXV9`p+LfhHm8R%ShQ75sdFC
z!zcPoU|UysEwpw03If03ohw?9R<w(HCMOPDmNk@0cvBVJVJb4I!IOVx%m|sZ4u7*a
z7d|gCvJr$275{0qFCU_48x?4b)pNI-I7*>Wr;g)-i6t5omrJ6dY^FLH`C7fl9DFf|
z5OmK)*<uYf!lQD#uEQGCm68JN%!*2O9FpKxW)%{L$;gI2N(JHE%xY<OZFUl-_p2+i
z8a(XyQ%9)>U^%=<SOk~laWBY!1SV2F6-v%bRHvMfk<5t}+29QNS<R7<_TsWBv6fCm
za@d{3??%Bk`Y}vMSTs{$@Yn6Myj`#ac2j0t+FACAVL;nmw|d-Cl{d8VFljWUW)%Hw
z@kAWt3&0j3iP6CuzX0Hp>a|`l(Fr3j@_l(0d%~QL-{=Vy=DAviUY7rV6Hn*@UIBX9
zOY%FjHl|DgYR*JjTse}UF$Cj^Zsm<R&}#Gx2J@XTqkb+?PWxFLPe<~otSM;bbbn^v
z^8|Z^R*NsC1G^<!^J#Plbt_^Mar10;fF_@@hHAxWl^cq5bzYJ&l<n5dac9R?rK%xn
z%5|i@!|fJIi?XHB&^Voc7VGCFSGb5LggkC)P3+fDkWRC9yqAiflB7kFYH#LReFx3n
zYrR4@@~_2D6wk;WZ{ezVG(OUq*N>)VR<4A|eF*^ynSxhull>><=p^LK82?hI#-m&~
zSZ|&tqoN3UzALWrkJ&!w#y?hX&UmL;lqPYAWZ%XXb*Ba|O>&igSDDeL%+TV4h|;wA
zc}*su?yxHQ7Fpm_k_g~4HQB*9UK8oOWdFZ~l$6W>pNx#}x-O?E{4|k|<4VQmJihU)
zgqLlsC=k}b%p$iXIvQ^);C)YJ#rNYjVX9K^G`d77z8T@2F(7*JHiw^}t<HC_+30xF
zn|pJEF$j7_eNBUZx34<ub2$_|r@yMLWO<S@-<etsP#aLy<iSsla|@RV?9zHi6(*47
z_r-j%^3d#P%eQQFE=Ont$TgVlaPAStjefY`1z_W9X^h4UVXVFoh38SuB>SZS!j1JZ
zZGzCz^!2i!s2;IhFUvQ4zUGg2ejfQ+YG^BEjB*V}Pzj@dbVb0Y#@U~<o(gU*HdHRw
z2~AG5<I=j&>Rxsg86Y?TswYSlW+F<(G5T9`_2{~dt{F(jfHU0<23i%p7^9<jkK$s0
zRBP@icU7XkPS(v{EFPxIbgk{(#hxC&L%63srxf2-2JZJE-8xGBv?LTwW#N|1!5|=v
zvAvk&8#_gR>k1D-$(gBEL4JcQQ3*363@c(pCp^ZFZnoZOPr^+#*Bx+UF=Y9B!e&D7
z!%%aNU2{%E%pRr^!xivMChyraJme*b{=g!=c2QuS(1~})dbai+BkK$=%a?&`SRoe~
zOh;?8RE;tQsRIscl)|OPf>c?B#b35kLQWE38Zh{O;?CELC4IJApX3K@wC@4`b~gH0
z3!Y(DmsEKpplNd5rAyihSq0k2npd-Bql8z`3Hc{=R3T~sQvW#}RfXBE0(XC0WXCu$
z`D*i3X@xf@{{hX{j`F=bgQa2HanTHl3Wj}QDx%L(VP=ibZ)pLSlSJ=SoZlRNdh2b+
zpWX_8JD{T{mGfNxox7KOlH2<ss*V4xb-I;8f#S5wQ;EI-O6bR<?y24zN*E;pgG0yV
z%}~k-e@D%7QvaT2cme5@vMP;{)vCRlZfa8?yxL^8<b!*L??pIcw%gP-gj->px>@NR
zCtyT;M)_i_+D1Fbl-=?kw|URl0$6IuV`LM5I^3ln4{j1hbPXf5<TEeZyt|CFvoyBa
z$1<os`m0X%1$VKUCU_}4;-aT;y`rfCA`)!B^45x`#EmpF4NBKwv!V*<;nke~3_MpV
zs^}UW^i)<Vs$i^8LEqR-)1dp6H|!8tDMGL_3fwUR7`vgHlHg7&Ka=DgZf_M++6iNS
zXPi<OPwVz1O?5bz+W$DR*)1xwOUEd$*o7O9xSdJ&q1TC}Yj=C+hEp2~^3_3eW%=n=
zH@xnHlN2zmSj+Q*cP((&m>jCG>6gz}zzZU3)l(|R3-1m+OGgckNEe6S2Xx&pL-OC~
z((zVnhyM;~SO8FJfSyZ6x|{kCl%Z&Ud3gBukMDo^_8SOX^&gsvmk&$_0H^)RkJjR8
z3I>Ls<v9q%t9JtmieNVb=1O3ssq6Up+i)-d!f2(&Nk3^&qX5#5Gt+e^P^GsLZ-$tE
zqQ|4m>}Na35{(>^K2->MA}VDqJ>q&^MUV7ppLr^-`sZ-?J{-OjgIc`ZdusT9SVLJ<
zhz8V>L6J4y7@##O59xq8j%OWkf}<+H(%?kdGNw>xXV4I!M2sFGP+3)dd8E;qhmaIg
z;359*$5(97DT*T_hDGW+i3py63B|D4xZ~g7ewAfEaM4Tioyi<QV}6X3j(|E!%uSG1
zy56odbINp{-M%NNnnsSPZ3yOn$Qe(2ygFjY`peLWmQ*~BI)2$}7Tg@~^$xT23F&AK
zPd7DHt+C=r%IrXm&+&rR2NGLb)nN=X+2_))zIoM&8^L9?ed$QHFO)FV?#aw3ME+j_
zP_E}Cy4*<SR}JcmCRyszBg{{IF-cZFqff6RXMUPA>d)<p^diWjrsHdF0BoftZ7Z$V
zyhakb7fmw%SpN42svqe(dc_WX!d14rV1u{&>CuMM<o$t`Y<n`AUVZasSy+6(;~t%G
zdXZ6LgqjNcQJwIJ2TbdZ+3Z!=Fc!tTTj7Fd7*FAXV;E0)b`1YY|Ms=I{{ixzWE_{#
zOfM4wGMACtG!vKGOg|}?TzW5<e;nNpzdroD<5}1$nN-Z!&{?M$Xcftbvi5Y5Bn~4Q
zQs^V+2EQ2a7q^KXlg4}5ai(#3WmNY!ILt@C+`&3tELz7(W}IkcFj)oz>9tH;jmpo^
z7D}i4ZJuY+Jd%{=#EZuI0~t=^t10B@ZYKs=bDeaSaVcBj_;EA!r5B6~f7No(*ITJy
zbdP38>Yt^$@_{)B)Ch)Hq_;mEjbHR;i6~>4;HJhpiFlL6<4F~Yry>grmc*gkUGwz$
zVfTF=u-q^iO!xS_qaU;|%(?)XA(8I>{=NIUyKkvxeL3TJC4Cs495xyGk{5H}P>@?|
z%v3?k40N|jG3`V3*KLZgf2dTS6qTYebc}#GMVLyd&{B+MB9G>X)g|UM$yq+4_>++@
zqveg*qQuismV7xG1kRMx?QAuZ0|Rg+hnC`Vmg%%fPudg7(NHXHXU{3cZ4ydKCjC62
zkV(vvsg8%e<sut0=I;JtN8(i${7fYw79FW*acI@fj#<4g>kg8Ne<XczW=(3vYw_FQ
z%rf$VGnIDCc{ekCPDaKrWS?axDyx}C{%9!iekotnl??{UV^|sME$%jOnKW4Z<t0?g
ziN-hsUrbOb22KeOyP#3D&1w~3bYMEX1I{fZ`igU_Y8IiCK)N1Elf3$3$SMl7w+dQ?
zn##H$=n4=cvb6S%f1uY#TPhS84n|#Nnj(3oiYC)oSudBO4QSz4-J*>#%mr}*ZJK*T
zmX*AfGa(Gag!s5Zh%!L5OBF;zZD%@w022^sLqMWctzAo5+zu6GUCXe0PCU{is*{nI
zAjK(K=PegiBBMscd2kXmv(nG3@i|r7C6iXu_bK)kVe|MVe-x0`vVdx`wcaucNGF)&
zGr83n;iIne7|qE$lzFa6OS%aqilz)Cvv6{ri7?bRDsL<L3=a17;WVZEUg0$5q9*b;
z+)nGT9&V`$YjqOkT6Q*9OsJV9T2JH*qpd@Ma2BSn?rBSEL_X%uR^TX+6Uxd!ewH!s
zX!<6vhLSE~f2k|K$F&k(1}hKKAon*s*O9R9cX9;=7F0yyiR6efmVNC|nXDoZj247$
zxzEEm)qGi3JMc*R=NC0pLNcZ8jK_~IT~C|1%w9HZ4YHHrB#^9RtwVIk%9`4U@R>H5
z<fb%ArcO8nF{DKRddozaw8Q}C-#bM5T^GY0D?kNrf4YCWI1CJ?Nt`(zsB(1J(`#DG
zyH9^Tef{b2H@K*K`iEwsBtQ-knD(3=Q5ksewdlU(XV0XvaHwDbbvX7|Ag{Tu@0I7r
zZl_W4YDCX4=^G6y2*AxSGZrCUOEnN54Hf>8UKuUSzKsWqL<5JYPZffmh)RB!9`Sy8
ziyrA3e?OP0c<YaD_qE&IXM-Bl-i}%#D1qq~w2(^%1#9G7F*a%$Qk33Y(E$^z4a<WQ
zW%HPX&K4UZ@MR%5lyH?*)whg{bS^_6RVb99a&O01Y|to*kzv!y3>BJ4F^&RLWWy%o
z4&NW2^J`Cd>27pSWcET4+J|nh0&+a6Q_zYvf9FcOCQmQ3N3F@Y?)b2_kg5}wB#s3j
z#_Ovg2G;l8NVKTpe$es$T(iJA&Z#rZ(!%1%99Cl8xN44-G*mQj8oP{P=bY=Hgj$Ea
zG&a}6H7nng^hv)`qB7T`-S)ALnnOZEpWiP@OBAZjlb%rsr6&xzZkFhl{e5(rt-Grt
ze|5sy{M2src;z#s_d3#-r>Z@1=*Mpm)-G!r!zot<+O26@X@vn?(q6kztNcUmk5f^k
z>gWkOWKfv-J4Z+H_w~W}t8xE(LU+74_-F^<>g-$$C+(=2yIs3jqGkk0H4?WK!a0C_
zZVNnMn`=eOvFK*(Kru$En-!@`DPmZTe+wyMSnnxC{Mk)sWgOp+fc&LJU0`2HrnUH?
zha8Oz@4B}?^MFIdMtURImIoZpGU;v0VcP?aW-of?0f%9H=8`kdHxz0L!^3->Z%B<L
z;6>YULh<Jy&qthZ0Kie9sSEn?=QNJn9(~KDAIz8P{#tjOY}0s)okL5I&P6LRe>D)>
z;J`xKz`|%3`t}2hhT?_?Kn>*+2SCm2wF974+g{MoQjVcsZaJ<sIj|_SbTj$R$Q&CV
zR!Qv{A9`wA_<l<9op=1H>ASk?PiBH&u-D8V(M*$(u^R(Y=*%kIWI1Fee0h^>=!<|d
zw=IX<&v`XfMd`F9O2u_76*>KNe`5K7Egy5W3SVwH1^3Hf7fn|R?3?V`e&j#%CUa`8
z7zgNYYfb_wKVKYiv!Y^=zGO=+J+o5GEbx}5q!e>hvEvSs(Rso4({2^jg0Yeixiy)}
zdDql~7lOGE)Nb`Hn0rYr=#Rj1;RBP-zcZG%T;7c34Z(NBa$y})TSEDif8d5lJ|%nt
z$fuJ(0LP~cPr!I1-N&Q2&<2&i+S2V)ia)Mze=&l4&!n`i0w*AR_`?bPrD0}JetDP|
zpl>LhgUJfJIqKgwGW6Sp3#D^uwhWgk#s6WrOci)V+0xWvEx}hrkRs5S+4e*gqKsq^
zcgFvt|8C?buyOj&M(rK-2FHIn74&<T(M&HB12Zu(mvM9m6PIC*CMSQzSxax+I1Ike
zukas;B9fvWKrlcu%>?MJL2ipZE!v)X*#7^+QV+|vJobzeH+%3nktLG)_!OnPpPlYr
z4o`1~`!ku&DbL`rd;8IWe9-a>S_Un54nSFg?(J9i*Mrt*<$03FXiX%KlfS*Z`FKa*
zWd^W3@)h8C=M2O(GKVEFuO*w;kj%o9yg}ZNaa`61ZFQs&<dHe>c>MeA>p>S`@6SQ6
zmtli1A%C3ZZy&!UDeli6R=G1PYX+RY<kiA7;RNteiH;G<nNXLNhe*l;HbOm~Bgo*Q
zc;U$OkZG}qW^7?ADq+lHqQC^d3N~sWzFwODp5G4t9x4*Tlx^rNctrx%Kfey~g8`Aj
z1(8VwFx~I(-H*e!mTFOP`%d<maMxD$GQl0KVt@9XMWd{|_t$#FTZ^glQ^COq=lb*e
zXV;9pzEmhs_8eN#;7fcaCOzoO2!+{-=oqHgP%rE$Ym35YW2YiK=M5O29W1_$%+u%N
zoz-za^KHu4(<MDur&|aQMcM4VGJ$%D$rs37kuJg9TV49^sUL~K;<|eJnaM(nWpc6B
zet%W6+Bohi-88fQH|M*G-m1gPR+JTtnw=LbS;C7%vjJv!rCs1UKVpT!%~mdjsRjP*
zRUm4q6d#{y)oFQGGt7#}D&!+#4=_hn^vX>zhV@v<Rync_`76kQB^AUcri+rwZB-`}
ziCmS&Y%s9^D^M^hi`6Bm#FxRSs}gKgrGHi^IAaV8@_1s(;bJ!W`m%ymP+An#^1mUT
z4X)qh&xK?m!OFvIrxG;w+6$u00cbg79o`zJZ`+G=ilc@7xxIK)ER-7#AFM8}?MsAG
zhPdTgZ1(;v{;?LMNzzK-i9bILawgeJ+7O?5SJEdhQA#p&lk|a{GX7fDS30K9^nb%8
zvra}{OC^x%zr{*`0hJz%s}x{qV<o~PG+-~3780h2LPIbig(bBm!kfzi@Q`Z7DPhzX
zYXYf)6*<#PrkNDENYS!IcrFrMhw~biQZ*>BxB*Np`WEs)9YjWn>ty8hI^Ydgq?;1;
z95|DFq@k8dxl8B`#SxA1;{w@$Uw=Qcf**>#gQFRKsX)2O_|ODN)V8H?OWfD*C{zQA
zf40nkFITA{rf)L?_B5&?C3p@$iS=Y;><qoBxu8%FbqzIUwSzIUQ|Nie$k0%HfABA6
zNgqSdh*W3??d9Vp`~f7(Ib}yArxf=&zU8Cx7~Ji#8rdkA$yc$x9D2}Nsee_eT~&3v
zg9fxXY=AvAFs1lE+cE`^FL(v-r8+_cAuq$k*4vnXg2ux4(wDXJyRU1Q71>EsDIGL=
zB<zI}ZoN%K7?i}1MT@sZljq^=FUOi&T2KU%XzjKZ)Ec*i^OzJdXnn)&lX^?&NO}EP
zy(KA>m3A9dwNz<ruNF%k(|<^@)<)7K<(6w#thSLZRYV_Y2%(YVIt&dVcLdA*G5PsO
z_Msj$?hJE@IDK0oiO?Uh-AL^0XZ?yk{*(?;9!=3Is^B!4S9?FRxo*@&I_0vG{m`>2
zGsk@=mnDTlJ1PD=^q<UQ<s3#(bD>boB~dxGQX5y-77l5>Dbe*b#eXQHXPVwtS#JmF
zvK+@dco^@#be$Uf>#khkSur=Y#d5Jlf&{5U)4K@{I40WV?}Y{nf>9dz)&|R9{ChN6
ztQ_VYv@-OOYeTfW-ei#~R~)I~RO8_OnG2*^*8z9TT;Qbc_qY}g*$vf09e>}23NR?=
z?bd9zNrDAkX`_NRjei86l;J=Ogr%7OoD2uCGpBb@Y0i2>u*@ZRX%df{ky}deQbjjH
zt|dT<FwM4wH{Bh4+MS7)yiT)@a_U{v?7D(n4bf5963TTiQe89CRQ)nH{_>^v3wG#=
zlFT!{-I^{Jzo+c4H|v3htUTH(6ttRs-qb%eUTP`V$R}Ft0e{8}tXOG3GEu^Y8>qF|
zX2gxD5n@dCg2OEB$go$!x*>MMeUh<Wa!R{yTZA+L6#_Rz;*G!Yrmy-!!L{Hyu4?qS
zhPzl5I8@fxDl*^ZOg*9T0$B6{SoDIrrutvLYCKI=UWXKkm9x|cJzPo^OB$-g)$Aks
z@!S%FrrTQHk_JNl1N8{hmqBYW69YFhGncVUKoyt4!U!gR?OVx?+%^!s&sXRN%#g($
z7&PE+PaDWB#@vFOf;cB1;{QLSxJhDB)Y9GK1<65HkCtNX^{OcI;qQSDAD_;zPrqEH
z8Z2{KSQuVE4+3v+{8w<iX!Kx(U`C4J^~>=4Gv{FNIOCx>ml6+)za5_S>75h~pAZ@k
zyeq`%omC=#yo1k3NbgFM-jVneT<|RL{61XA_dPc}Q1E!*Gva*u<Mr1ko|zpv1Ie`U
zgRzPst-@KLtXwz(Cl;?L<U(CJ*LX)mqQyWv2Lopc4=1%8z9RMZ>ZgyW87CGsqdntZ
z()HsiHm$04{*6*=KjS3vH<zlXe?Yx+4|EmOYLWwgNlAy?N^P9Mv+*2~EQ@#6<SF{p
zh6%1Kc_8r|<WZ*KoilI3&v2rDy}UmC)0zw{10jb$zdVH-N@!-RAB16En&InT!{?{p
zDAl>EMgh1nTGs*?rnMW4luS847a%yqXo%eg;+&tS>sp_J7pqHtOs7YSE|}uO^iQ-v
zb>-84<hcNj8KWqp;zKwVeI2A_!b<u^h?ppv$Y-7T3(@5#5nG;9<i$AgQ7nEvy)zu1
z`;K}*K&jBqQtl8ixFx>Mc?`ZpwYm}=*7GHMOAQPyM^}+WA(!m9f&|*K_ed+56pc~z
zOP2l)KCMNsZ~SsJK-Bd<9=*b_Hk1;-rUV*)r48PE&e8`hpo^e4aDK^hECp3J`R91m
z-ZBZArdVxDiXB&VCH=QiHr{FdE8hvxhP5OWiA%W}DcF1zXVy6(+0j3HX$>%)8BBy!
zhOO|B(>*+m9LrX&wED6Jr7}!m9U-gpQMzC=48c~9Ycd6oQo4~zugbK_rxKP~D;kP_
zb}X4aUKA!Xp?r@Ql0~}fYr(8jlu9mhjXzq?Bo@VDOL<bNcuKI3S>;$@c2bHG(cJ)A
zWn3T?L^KVxq_|?%oRV}EIPUJi=_|+xrp&0HxoB?8!F~OpMWG?`Z2~4#Iv=c4Ew0TR
z%wZ*lSc*r8H-Z#CtM!Eud`i?b%*0-QVP9>;u@nr@FKf2A-grqNinLUbJVRcYq!4@=
zlZ;;6A&&LRa&eF&^iszi)h<MxuD<9INFp{nwwjk{ODey9KjEopyW8Yf(lPM%7)MG8
z<~eokUJwNo1K+Q)AnU(TV?kDV!^T2`X4<O*jYURO-KJD|&x})#K>AxXqpawEFb_C4
z0`L!sLPRYxXzxK>bZo|bY{F|U-e1Se2L7bwtl%g!%Y~4xjTS!Gv`LErXBBPQT<B%&
zdecVFJDNBwEllzUnKnSnI>dw1TQsM{s;wr_!ZRStg|fi%1FerRnDa5`5y~sQod}JG
zHuR0v-mc*jslO~si1180vkxMFW*v4a`Lf=RcfS&t7mjlF;N}o8(<@ACWR)A6Lok3V
z`$n5D>x~z`-rqcOj3w8s41O)h_cfDJBXU)gvf($k`;<q%=<lU%?|5c4ez`9DG&U=0
zDp|@*$*uTj^j|euW6Te`d^=7=ZgubCcr}v`cW)nOWn`6G-P>Y4L0CwCsC^Un_DuO^
z<wv-8amt$VE$&?$mhjkdZ-WR)i$<3R+?yK&r|uT_4h2EGw@cYIXK$6V>s9Ve*)_Fq
znX>DZADOai%J-yf!sABDCZ0*Wt>e$qBkt$E8MKJ(qn8QfjL_r_eK`-`30IHy<vg$_
ztK90#+A*bCO}@P^gE!27%0^oc@nz1erG#&BWyH;ZjR<A2<I2i1>+IV)Ww@R&#Ez_=
z8y@LYtQ*sW`1X0dR5j?`Gp}C+M%4^n#3@<1QYoT8ysQSPD}&55opaCS`SqE;^Gx_g
z&QAN*9Zz8giz$#2$sJ?O_OjBHUvIT1p^|#jz)zPx^)h4CcE?bEAyC2=Lh<X%e8Zp^
zMo-%%FDI#=vr;c<+?EG?=?zOpsp<FoED4{w&dFV(OzQ&@kTOmO2)8#*)xj7hO+$@e
zS8QP>In8s@Hat%gfM|gd;dar!2e+(dnwh!3ddXL1I<~AUfN2oR)?EnG(!<OL+~NG8
zObMz01sYtFJ*tp@brZiyhxCaud+7x9j$MpJjW4IiKGp8lVH_%7%1D3Ifu=ldSf@L1
z%+`Xq0qY(=O`#p<R)z;_SS+YH!_UJDjST`e(^Gg#(Eh^KX(~;(%ae11q!oHqBj4R{
zg4t6r8MWndlqH3Cd9USW!yI9y@HvunK8IRx@~nl+v`dtKuY@;LcC<_FeB+*ynA0em
zewkys%T%YnN9+`+un){N^@iJrrMq?xMjteaO+%QOeR66n=pb-&rXk$3(JcI7p3V*>
zp-1T#blB}swWQM*3TfVd-^OVAriipN8wQX@Fvkg7+n-5vYY->|`Np<2$d?KiY60+h
zy-R<o@z(}_5{hP>D2((XWGfZgj%%K?JU-{|^z6JIOsm3tD{%uhaweRn@;-JZekz$D
zemiz9o#l}avPSgZ|60N_(}-bbiW^l~n2iEVhNH2W6ihf2eA!*Xwzhz|<27HI^u8<r
z=I|1spwH_y52KKK-Mm$QVsQfm+3tG|d~yN8plrT>smK#lw3B^RsVmnx(3-*QhwLL;
zA3ICJ*{%D<Tw>3qD7BryNqEQnwB!D`k^LBMJHIa(8FGSgq5cO@b8hyUWgSRW8e9cY
za{vY31wc8CT`N#;_(lLq7|)Q#Q3$oyN<SEsa@ffWDl-LBuv4~WP!*Fw=Yz9NP=(o2
zfNqR`ayzUNS|CM3VHG$%gF)cTJZREiYW%eWR)L3x9)(qTO#-X(v-31Ct$F}f#qXPe
z%KN4{KEW%M=X3>P?e?BD7ZA_;(3W-a!we!H#al_o6mMmX)`7g}EfW3DieuVaM_WcS
z!|f7fb=ml+6)mcpH-9k1B|tEc?gvC%gtI$W5f`Dh7jfa+a{6yPkK08)I;z4FD=mC)
zphfp9eine`Du4w)9l*j&BB&2gc(s31?QY0rB~!xwrw~^y@~waih>h-<UU_CL<?8^>
zqF<A~ygvO4Yhmn{AxA0`1UE7>H<uB-H6#l&FHB`_XLM*FGBuZxco!#s)mq<g8_5lR
z_g^tUU#<_C9S&zWWN`?PyENA;d_WMVK-<75{6oHTnzI614$}Vh{kYOfRwUQ5r6kTn
z9PWxc!{LzpKF)3>OCg1pIfat(f1yB4U1%X|@n1A>>IDn6xNhOlT63WUr4%X&2d(|%
zHK>(62oj9N@1$V;fVBmGqC}9OO*!=gC?xfJ#R3)bX=(~oi~m|-i%g%*fsEKC4w;H#
zhFU_NK3i+Zvu7=2DgJ~cNGh;6V<9c~_Xtvhy@jBs<OWNIgrT6Pkjzc460)(+0cB3j
zM{W|B)K@PhaL#U@)W9tLJ^WhC?UNQ5e_ev#wEwl`P?EX@jYFY->^5W20=HofF?c>z
z+*AC@#eaozDZLNiX9zwX98n~hOL)R<hF%TDw0{;$O&N|gUtp5{Uc=U;wkPhAO1)n+
zLsB2bYN(uD&wOLKpVyt!`v8uul3fqVwTZ6{p(3dd1*0IuKAz<uSRmGc<xA<;VJ%=v
zx84*I2F0ZXY1uu0ZUqvY*chQBmXb@DG9~<Bzf6r>a)T1Pgva3MJxI7La9HajviWT6
zf0X-s5;SQ)at@lQmk^mAJ~j1{_4g3A7!U4j2v_z9V}gWj*vsy5E#gzc*;;)fh9guw
z_L(sXOz>Z!r?r<9LGM0gYQeJaK4Jy&U;BF&th!G*dn#mq!#WTGru`gniJAF2eDe}{
zpmHt{JzSe_5GD4|VvI|v`i?O+i?GMd#q1)shAT|Kc5^?hy4cFz_uqHV-ynAhS^lc~
z`#=8a_oE^{NF{nIUR=L>clh)vG(-2}_}%#O;Qi6bD6l^ex|c_j$>{PT<mu&~KTdu+
zm>f+;fu=WqUW_j$nBxVz$?(0;@&ds_UM2dA3R4IU>@Uod8~(z@)2H3b%kjy<X!53e
z`TRxqdh~HJwWE9epZ6oo{Obq?T}(z7lPj+fy$Ri`(bf3+@?>;1tu21*m(l6@(GTN~
zZ~Rg`tt}to`Aa+*8p=wq_3YweJRPV_r{8<i7dKyjMqk_@9L{I&wL|yA`0{jg*$<o!
zyT5clcR%6+%R7W2CzCgnbF7}#m14toQ_a!f{8AOpJ>2UKu8$`^-v0T;zq@Di(N8Al
z<BRT~``fFZ{oi-LO(yTJ{?v89o=<+eK8`2j-@CWhzrTNSI{MK4_4=2W@$I|r_}%!p
zJ3CW<v6C}Yh-0EN;?u%sg-31Fj!)ZZ$!F=T*qKpxb$QaAj4!Sxmq+JtAO`rueiM7L
z->m%+i@)LNd#;r@#e|3))=Dg~T6t`XRnAZLwbsRAJrT+x5xmME5f}0Qslt#cg*6(&
zY6Hz&qLfX>Bx=#oas$l-H@EuP^K^^Uth$nankbA9NKN5IYL+22i_L_Z27v}44+0ND
z8HBnfp_d|beZ<rBqK|mY7zGh8*wzYk22w_2pc|J+tsZvI0){{6ej2|XchAFjQh_=E
z6)vdtRsdHS(f4Vl-P4HDI10*>B#9Uouw0{a10mG!l2?q#ttwRm%4ZXRQ7gQ?5yTdM
z5j9BjSOAcS+z5Y2DiJJW#e}Ayd<qyxX-Hrj#%YZS_o%K?A_iMj*G4f@{@N91eEb$x
zo_`qi^#-K%$v!pSuDF6iM7G6JK?|bVBU>s`^2vU_Sy~?+-3X!Tz=<t&ss#2K86=vN
zB3Q#0itYi~N{e8#y)od0*aQ@5(g)^$`F#-8<5_MXYD2v1<Jxo+gm;o<IIZmlUJ39@
z)?w<B)(j~@m~@GK-l_>_RNF+8J(kKQnrLOBH)D5e5-H-*HqpfF$?Tiq(~VX@TWkZ+
zMg}f}c9R8Y%K*Pp2SEm*4T5fu%-jSSVm&gm2Z`p`;D_6gc{4<MVA()Z+|z`Ap=<~t
zx}<#evQZWAi!O2~8x>KF&iZkc4FxJ1?Eop}WS{^kO#Nsm5?b^El7{Yk#oLst5gDVc
zmuflTs%_<NBk(Vsd)pegpnYv;&z?kit^`M`F4yOl?r3*zC=7+HB8&^wO*DazRK93y
zvQ+=>Tk>Xu-z<5WA5!vE7bUNM4Rx&z4NJ4_3!D>f^hf<c0%f!wi6eNApIIU7c$F7f
zBUBYXUTazsOSY|oeab`i6CV22!+rdIa{l}Ie@B=9LZ3gnJVTiE(a#y|glpSDodKzg
z$xriuB&qnBa74up%F(FY%kQ2w%9{|RC9P?0iPmVVv}c_r^3{)4@7C;p2~yoavFAtn
zIr4fbm2cxU_swgQmAtm{ki0fq<h2f`4LY1Q=x`8W!%sH+WWzy(@eXod2sv2#jrFs7
zl$Z$8f{*~x*(%aiWzJTa{ir-q6RO*j8>}Cbwz(1P<N{lHn-Qd?2xd+-wW3r|EVgU<
z;eog~WUQE06Ft`Y<KWwWWQWy<t3?k-1akB@0q8-T?X0@J7OSPTgeqhSTwJB?fZ5I>
z5M%beS8h3aurp8zMQ+s!0;@)+tT~CHjZc<$aPo!p#H@28SmBz~ZZ~4r5$G>S>LLUU
zEk-SmH3FNVJttnKe^&3He>W87qz-#uYTiV%fc|OzRytm$m5$ecU-8`^Cj52f02BVg
zvakI9zH%nyLEu3sgHQ)Sb^sHghRK@0v5nxKXf30XlX8o~tLMTMts<HsMt52jr9|2e
z3PmQvq-AbHQ;?;6s|^GNs^GDx5>nWY^mP|dMblZQMsO|BUuWNHL^H*175D(nOD##t
zc6Guh%!|`I7H`9UzdBTXIJ!JPI(|30nw?HdMS-ZhX~k#okO|ztUk7B+MNpueDW)}t
zw#1xT%PAvdQ$-WaTUKTQD*AU1gUi8sVo`alwR!q#KxilY!20TpkVl)zh<IeRA5F-b
zS3v?8wp<#jC;k@!JGj}L_y-twHz&(l|F1b&tGTv0S*OW=#g82}(#(GBu!$znDH69$
zG{ygK-%87agj&I}tPjC5<{Mbn26&nsW%vgzGhu#b)&%+2EHhyMzB~Y59)K?oz?TQ$
z%R2<Vd4EJX`uA0Wq)3HUk<a|QnW8ystr-9)b*t(lsP|@VUVX$Flj&po00M6Es<Vr(
zWI;9jq<i6i@mdp*_w#!kW1+q;JYLz)eD#q*tn?5oJLv;U6Er!_s2F|kl;n9@X5}b#
z6=MCB6jd;h0#;K6fVhh@08Co3&0vnv8!r$Wr?A;;dRjdTBalmwY@w1+;u@E_lU~$0
zrGKVLHG)I{>r3?SH~m0T*K~nJ)LQe8L==!GmUi)fs;Bs?7L-mm2Kzic`Sxp7KbrYM
z+kr_zSb<fD4R4H_zpe)HM64vaAN+5<9R$utJnRN#UHdW0XW9)c7!)u7-BB4rZRpJ5
z4>ai9WQ`55x%kOMOWf4BCEML*G?0CD(vQGEbZWxZtQJJjj8vHo8Iif7PfuKzsMxH%
zy|q&tzo{_qzd^tK%b>e8i?vVQVt=h~$o>aN#{ABg(M&HB0y8z2A%X}Lm#&;I!hdtb
z+x6G?U!Hhkwr~cLDeW%WAR}6ZGopr$s2D?cNO#KkJDcxxxx+Gxc2dN>A}&z-6=^QS
zF@E!0$%!5@AJ^$_msd}lMu;lfyhot<nn!Hj)$gFLBFx}H&aXEsmNWqtX>b%bu$$Oa
ziFKtS1A$qQwx&5~BD>+)Eso_ibblCLM}Dcl<*$*nCn0`{KRCFR*D2a4AAd_SEn62F
z%^Vk_Q01&JpQ}ly9Y$oQ#VnBL_BHG_S%?P4E!=lxo4OEMG0n{qW!#on7{ipBRt6rx
z`_5!5!JDd@s*W<R_>yP!3NCA1Re5-7gI6w|6nYygT*;0PlXl3u4ZPSy*MAfqqK^fS
zWuPp{$QjblN+mCUlMm0M-@mP&v~hG^cI0f3w2cp|C+{pnZAMDzB);ApUtE<>1`N|s
z^(Cqe1y{<tuvJI<E*E|OxiPmD)-w}=GwH;no)p$l+1+5~XPMCnf&*M3Z~{r8I?rs4
zG3_vzSsdLn(PkD@W1P7gLVsgkH*vAAf<o<J;HqZ`Da@z!{rP&Xv+2L}?uE95kA`pW
z*p-tq_tu@yX%3TYMR{|Vx)`iZ+@)ObxB*9lT~anI#hOFD2^3GIl_S$DQA%0Cfa@{i
zBgdeu>zCwG<A^RJ=>Jm(Ge1I$B;vX;^CxOU=1+55)2+MSllcv!(|;Z1K0OA-v>!x+
zCe+^=2=G#EJX}@W)*qeh)&-J&=^g^GKJ()w+kgO~-g1!2h8#4I&rl2+UbGzT$eBt2
zje4q@&mkw}Ivrg%;-a(*aR8YEov=vO!IYCtm)Y%NW}kDmS@f?$0TRAlqjxyit^1cE
zZ1VpI0(!>Iv;Q2wqJJMe7Q!S4f)_aOm`<28q9iSeDwVdHuG-P0a1e`6js$p#%(ynb
zhr`HTsTVk3fjFteQ`}Jy#?_3`8dEXznoYt{^6DC4JU{bl(51pC8tRk(CgM&^lFcz^
zZHfIqMUj2*@pY&WA7wlSu8H>pjIX6OO{z5m<1aUzy$Re^`G3-KD-9Jha@s4?U%m}@
zKfgcyHF+{BU#E!6|NeaP4^Yr8*$e=40xy66cKQA3KO@y-E#o!x2!d|7)=Kpjse^l>
z7R(xZz5x}TZ0lhyf2FfSeyM+g>vbdFYij|FPezkKF$*p>BY+~55oTO*JYS|{NeO`T
zwGoBxkV0w-E`OBl_NDMahn905;&P<gjeIZBV^Y4PluF!AlB=W@SSZkz%}83!xS7r{
z0f=QQ-K&@oECXAjq%0;$-WMT7o82Nf+sbZn23A1e4LEo%@6msvo|E+m1UKQH;!IE+
z9(aG;oi^i>q?5vVH~jf3RR%hq`b&kZ5w>I*<K8PWo_~@3sW4#1+i}k!$m%7<ue`^o
zT#Fsb+n%bXxID^_8PuI72cn`uxg^IOEvY?uXcYrgllAc0CTnZ`&`9I50@ckVoV`j=
zPLtTd=s8xc`X(BRu4<&>8^stn8T%Tb%yWCEpKI|IJk{hm8EmH&rtz4zRNyUGEEHd|
z<+fHF3x9$|>X>wu+N2>8Y=ao@Yt~MZ%SBJ#_F9Duw45|C!<j7I#Cz;P6xlXCsnbLC
zj70mM{z{|E?XyxsG9mON;c|9g^#-XacEEzIEfqe}IVQO5a@d=1{p-!V5c2k>-x!MM
zWbp(2d;P{~8pM%GQ){he6s{S_NxrP71A7A~CV$Y56ui%B<(}iqpkM<T@wke3LK8*k
zp|2;MNiBjgpJuO%=f2dB2vN)F{f^qICmPBrZj$HRnkT~6hahV)smVmG35v+dZ%rsM
zzteFMiYPfVQPSD9$w>ud!{VvO(WZ>l-2xFtWCLxdzE%@1bY2C|wXYNJS`e0==a8bi
zd4CB`rMspZRBC+P_h1r_9ep|Ual-kTCgi;}retCWSlnOshA}q08%cAxjO*GKxNDM3
zoPSGC3xx#YzLdUeHu;xoc%Rs3@BrTldt#$OA~Gdl)#*&Ny`Udx^ueoMvV+DB-l6V+
zq)OE%m3z7ecX4wq2C28>si3OKjtu>vu73bH&J4I8)D-|n4=MSbs%;-PV>{`DYTs7?
z8b?OAs(neJ)5vkf<JJBw@cq@ku*8c#AjOBOeXmrZCdpkBxKfgfpKlzi{1xF7H}U?V
z%Ae3NLtz#x|FxIwohg=H_N<%qgugb`Z+T5$UMMuL1^hUFJKDluNf1t~BUeC*BY#%6
zpXcOco_hAG+rd_}UQflgb=CQACV>mWTobP|FOLR>gtv~padeg#HWM@58=wGY^ER(L
zrJW_mn1ow#?5(nn($vV*qV4X~>e9b60i0;%mnLnEy<cyl0H(!6ZOt0#Fr3n(?=^3b
z8p!Io$r0*|cSRbQGGgT6U$QbqiGM=Kkr{qF&#6dUUs{ghK8~SwUv!1W(zN3zB2WVH
zgje*52$W2i$#no~6S!|n8;CKqY+QSTd!SQj!yMpUpz~f1R?Lqha~uZ_eh|tOk{ppD
z?2kd2LQ*^z^5ZH^6}h_`fcfvstd%1tTQHIgL^>E(Y{JN$HeLPpGK8s!Fn{5B=>Q%;
zm@@Rz_67VH!UP6djJx;`gv)yt)yuGlF3~V!%}DV>=#oM%15kIN%UJ^%Va{H?h!n#I
z%iSlPYi1C7>45F$(2*;sUGL`e6R17jM8BYfGS8-#PU*{+&S(3%Oxu#l&8{x%H(2?B
z^EldRUcn`n!l1*Wu;e^QqJOB^cXHf*Atb><IBHz_5aa+#w5W&8&O{ubD6ZRKgYvPK
zekg<O!*$IN)#phisW|ap9`+lYECumEn|!7e$})seiniy_z7P8ZJt4}_ALtn<-&j+k
z-UzCOGovd`vMJhNZ9_M2qq=;Pu7pGW0cxU7q6%ehWOH<KWnpa!WtZJdF9!iLm!S$j
z6qoy@FTQ_EXab)8di(v!7GbSPUVrLW``72Ur+=OX(H$E0sbfY%F?E0cc+wjLY<x&^
z+Z%Si`}wu|_Vh<fc3@^@PyfNT`}w_VR=yPLB`emm-ZRgvr#v`z-Y0{4tG6C~vX(_N
zy2vHkli10@Ov{g#2&)v?8^qL76f;^D+mhInGUb1snjw|5eq`}yD<b|tIYRbyL)qDx
zT*K4J8VuAvjSD$sLE^-UmhidAbls#^?McB9h8yf@#pgmEumfe3xz1L;9I*MEs}p58
z_1u_8&csEMxM4Xm-Ys-cEAg83E~gLPYX0!+%|3M)3<PX>!DZpso2XJU3H^d<&h|yV
z4!gIvr!Nu(1bWq09hcEeFB1YYH<!T@0uz_@tS`ZTa{BxAkAqCg=C)|4NjQsxu_#z(
z((odwnN>z333Rp_9d$vr2OY03k$aF=CtjUcGh}Tkj9^Vkg}yL~codn`EBRqR;EmgB
zmJ#L<e?lDP#L>6(9^?i2rk921JUV}gy3k1lFUIv{wC(fF!BWiFnL?`GM6{zpVR5`A
z0n0&u*BXSL3rJR=Ek3fNKAjzqv@GMu=Vg<Hw)R1p*Tf*&ZL+`9oNFJ4ye_77gC_B@
zxviy>Dw5BWGuX2x%!kpdvrI8cNT+%)<oHExweR&Gdra~TRGE_X_tSgKm^dT$RMJEC
zguKepkA1cG>WtJb<p+9xL%FWgCB-AfkIv<Pt4<kv9lamc3p*H{>vzehuc;z?%Q`pK
zm?FEj!KAfR>m8&?j=Tjc@R+q2qZZogi%&AB(dd$56aM}ZB%eM(&XDtRlt~IuCD-XZ
zNtt|@)&bjlHX<ISY>gP;>+!D3<?1V#l#L}@i}aYS@omhxvW`l=vlI5DO`S(nQJlVi
zh76y<Ks6)wTx*?+XGaULm0#ge&bd+d^5&|H#$<D}rH=M>!iBUK0M#+3-6DWzxaStp
zY=!$~bO-lMcoLA)y^SVao!sLaRi+i-DNZc2T9C^LODP#_+9v8)0Y{Wn92?eonbNy~
zlOl5lJg(Lz24_K=*dx=J3~#pt05RBqfvOi~x1P9UPt?9Gz0jq*bTy4ahMi)`q~YSW
zRa&4g$C5&>Ci!jj=)%U7Q|iw*at~C=4>3JWk$SF@w@8ceakJz%Wi!oUoTK<wzDp+@
zz}9?kMu%fr0n*~5%NmVfohRC4PhbrkJ$acP;!Jz|sflf*gh0dcG6HjRw1{7S3gFfm
z5m#v`&D^ClTA?&K>0Btv9COx#zNqRrG3m6l$43o{?@Ub|dBsc7Wocg)3GDBs94Bx1
zh*q$x%~!i6I&%VAjZTxfW`h1Dw3%{N@Y321QfM;HaBDmtK!0&=E`Ry<b@lKHyraqZ
zf%x#rO*T3!pUy-nCF(8k`R3(+y$XiV^PyJP5(_BlB5hTzEJaJ<r*ndO8FgX-u?_;b
zBYCC5BFmMP?j)aY1!$B%kv`t5Myfz~6WkG+uie|1*Ta8@-plAo7>C0@KM&yp0>4^6
zC?!1L@Bo)UXus?G@Z<1jFSV<~WL-o?=SgFT+L#(a<OY%uWUNLY079@Y*@8T~fBX}Z
zsp<Jv!5$gG$G>g?cnQxkP8t*rZ8srQvtg6Zwnutb8v%~t<xvY1hCsk}3lyOM|DXYS
zIW?C9yqd5%cZ`ee0>_Y@&NPux;r_9?)2TXsP6x>a+}nm|_qMe%Ai(P0mXv~P9#DqF
z%>EC=LL{!oz1|TFk#{7gbH5cwf9B-B)x>tCh;mgSuBMYwJTb~#N=L@o0;DJd>}|o(
z-&5)Y#}&O+VyZZr*sAp;5Y|=+%?a@w5Z2ZSWAy%^A3c6eARhhlI1DP}9^WgufP-Q8
z@N^kuC>o_UJ>SYMn7Kmf;M$oh;zgwD_R+F46@ow}>rO(bDvVT%b=2d&f3P5wbQ^pz
z&G5p)Oe=M+Lg?DluwJP53Vr<!D`+M%Wy^J$*16*8Pnov8Qdmcmjcl-ZEfU(7aTFZu
z>zi4MUz_WpRlabZt)t$`m1wWtuR_Q_=%pn5c#91*BbD^Dne!jh4ID8usD72I?yUhM
z*OP7CP+f*&ifMD30gJZ>e~|AV>Pxq*`3JIVxGwjwXQ`psheYZw#BQ#Ymmv34istQv
ztBe&em64kncY9GEmD+MLSYcqa0?~DfK8b^s=5sWtCvmXSPJr|KCrCN<w0?gdCnv7!
zl-`?@xjgsU^s79=$>cbHsFyHuGVH|3&IzT=249pJ-U!PH-`n*te`>#zmz6|ey(_$Y
zJ|We-Y@5o<mP}cN6|%8%>f}~AwyKfwg0-5>#OA*Cx6cAD{{)$thx(slx*sSW*k~Wf
zLg2XYP_QElfl<8Y@E*v#{fj~Zqs1S?Pv!R?*rg-{)cnys%23;l7!|(59`9i<$Fb+g
zw?;gPy+HDTALvQ!Wd(r3dDuU(#`ruGllL)Os<1kP_GY$Jm36B;!fc6kItjZn8-pPm
zV>bAr%n0egN!Q!;Y-V#HXm?txg+9P(&9prUZOc!;Bb~kV|CT0CT&H{Y4<t5&LMwcL
z*ZP(b?|=Gb(dU=ZOfM4wGnWx61{0Syv@g7Wm-Q>{{{H;qi6^ixC&#ytYJ-f{&PWw~
z3F!;;A+881p6H@g6`@Cwo|!a8lMvHyKTgjT_%#B8my?CfFDu0~RvEp~n_eb7$_U_I
zG&?WG4$!jF1USa+2xx!u!Uf}@xt)@)5uRTU^?cK$xFB}sw3ocUy$`>>eLnsBRHPt(
zBvaZ=I(%GD|NMIL4-g20agz|t0ZzZaPd}di99K=M9q+*|gUq?6y>v1cnki)`EfkYl
zK33cG@*&Ea$6-vb;3>%j2}Swy`{y)_e7qO22w|DlI0A(?e|3gk-jIfYk>eZQJuvGl
z!#_sd$r(->I4<e0IMIr`CnYM!8E89y`?d+Tm#ioV&I{INrw?{yliU)HPQ<RA9fX}`
zqsqG~PyK#e`L(FnUBq$7h%R*IjMk*@CrJj$?#mZvyUW&~lD7a4c!jr=Aaz|eq)d^L
zh=>agp6+kU_>$kvG-3op=dwn801Xm4WvRw}G}__2XUIOTTSSWv!aM9+;Z6vDV9w|n
zBA&wgF@`$1;fms{o}&8&Pb{XqX0%dD25tr@7^^7aNMdNznH7BOjKW<Q7HN{S<X2K@
zigui7$3+YUi%&k@*YihKu|j5Rj#Wv;#Y7|eYjU9yjseV83`S$;d&6wp*@byhjW?|0
zMLCd`VFuK>9C$a1mBhzgU@1L+iZ4ohNW=raD({i!FqHZ+nvGr-!&`8?yYd}n4#GSD
z+{$+;Lk@3ToA2P;?pgB?HgYb}H4CMfcFj&CNM@9{U?<+i>2>bc5Gt<PjeHbWohek3
zfd~z=-5<MHT)%W4|C;GY>6jpk9KXb`{z6$14KvovB!r84mP#-(#Bnx%KY6{0*FKMV
zK9F;54tm!hBmN@g|E@b{nVlPMhAyfDJwScv9Sg)+^#$CD#~p+(bR!$vLGOrSW{N%|
z`dAU%XekG+;6=x*lQVVSLrW1A6ScykY7_Cbp{9{QCURNm4z?sR%u^{a6k62r0c=M@
z*2Qi;)v<`eD||RAqq3)emCbrRn&lj~Oo8WwbTjIAbPviq4PjjfkZbuJ0hy+ZI%r@c
z)UN3gkusvO<4P;E7lmQU$TlpJL^hd-q}=atd~-dxcpn<<!fx&WEUgSPY(9_)8v@zB
z(Y}cq6Gpx^e9n}SMyqM0wH=;)&p*WrbRO3TELOyFDvj8oP&k%<@40GwA;Ds@@l!?>
zd<;`6-awuUs)|_Nh!{5!qwcHqth?~t;&DMEi#Fw{8I<C-DI=@LC#-13(y%Rjn*^1;
zYu{-7P<-CjGALj@wk2?oEU$i)JmKX;G;bpoN@tkEQtgP`<-l2yylKmMHHy;0*2ucg
zy6qx~u9+a0bJ#?Gw#8^#WD|OQHr4^+6#Zp0vTer3p=!fwTR|(lFDggmjRt0sI`6Ae
zg=;GTA0QCTSPFmzMx1236o}b6r|g=)^T<ak50+DSE)S%#j7qX%;%`kqCb_tPi_$nK
zO-#y-D?Xqz=EdPl#$n}w(3IJp(HR_PFLu5{%OF$)5z~2pk)D;O)C{b9q&)i-A5*}^
zuE|;>P|2zOiJgjeq*YDcwxAqc6Wm*7>dW1yTz^5G2eAc{`2AKVtgIkjtLuiDnB4c!
zI3^6F_lB);$t2O~<-M5(gDw5`o4zL%8G)IE6HF>usp}$fN)p7j^^@o$3cyadAnym|
zBNB0?4612=$ra_mVtX!^+RMVq2&+&<b53-OZX51zF(-KQ_Qn$CnfN=LF=5OV;qpw)
z+`nhz>k-ukkfMm3CH65FC1toI5fPiJ$(Wm*l}1)tva^Eo2F%JaG_{6O=bWkN2hvPa
zTx&6h`!(-&DSvRGl$*Ph&@F#VA+xcX2JacJ7}pDb!EV!mxQ?ffIriNnd)i%*yj<bt
z+uk5Gh*@0prXI{*r@lFDK-$Jdf=k>FANoVdfue5%M+^GPfhTJMCz)AHKKooL5RPx_
zA+tLa>CxQMdjKLHN|dTW+O8Y2T8$!Aqf(A<FZ2gCS-RGNDvhgjCmc-$C9*^!qh3C&
zTV%d}^JB0}dJx-8-BnCEhPIuFdbo$CJ50oGXImZn8+p1$;^R2dbSr!V@GDq45C@bB
zAIK_)6(7jbhn3H;^kMcFv2+0J8kgtgc}cr;Cgb5No5p#W_8c-T?BdF)kz5aG7Rh~C
z$w+_fI03O+G-f^Aq!n+fI(GSxr9U|ufi;y-$L&4oydh%Gs<v!O@^4P?)%}_!Q*Zc7
zIefaFbDlfBBMQ2p<w9N#nSw5zeib6OZFiZJ4h^Y9%&|P@Rg$@-73kpcf|NPlDaif@
zJx{odmoX186ag@oL7yHIm-N6TD3=AiFP8!Sx1GE%<^g}U3neBIBdI~d#`);!?i1C<
zBI??gFJrr2!Y9EXHq27OwK6(M#@GRX3@J7NnRSLfE-1D|$6{9&3WrTuD`kooI}%30
zDqJ5q<^k?xwt*|DlA(`tW~U3M2n1%Ej?O$yUbMb10#oz97Z92h6{OSyLPV9oUcrOi
zJ!8KOq$YpSYqeE2-07$-6ekK>MxXU32j*;~&7owK$AuK|vgk*31rinj!g}2bJV_eC
zluo&R`w|<QXmibsn^93c<O$uvwA)}liHeX36%!^YTxaQC{OqPqrm3J=&-)O<l9<av
zYzmq?<~>R-55CfEJGqSbf?D(X(O<LqQ?%3Qf}ekx3REpaHjH}k=y{!c+{6*bl#w;>
zs$wpJ?r80beqqvYFAq5vJS;3OC}3-q)u_-c++-)#44twldI}pSo5a!lWU6sHAnS62
z@%XeJR9H7muSZjqW@CAwQ9>k&G2OX1;9AwIA;%l52A6MvIB9!Vwl90eLoyx_>N&EW
znPY#23eI=HO-2OhWC>;g#xrVy<c{JyNLyk2X{6NzT^VUNTw{v6286}R#pnjOJ<qLz
z=VkA;sgW#-C0DF$!qM>dzb%GF*6>;`j8a@S!;^zLhTg(gg4$`qNoldlO7Xd7G0>U%
zBCafzQ9b6!u3by!oW=iBiK_n3Bh@ZBwxxgE^;y<$kN*QiW2n7Rg?&dvL;B=~xzi{!
zWo$O1x1$=2sKDdBux6Lz7PKL_ntF~Ogl&+*jXn4=*ajUa2*2R6cf>ceD~qO=WEbBA
zWfJx)J`~>|xrO$nJ-(3wa3j7ku!nD&{x5W!(zTZ`2!VWyy^LE8(cGRG=*jZ1&^~`R
zjTQ=+X2TW=MiLFM+9~#>xQxW+Y@zz_QlP-B`IXt%+jzxVYlR&WB}RnlAa5Oq#U6)k
zlqqUD2t85l+M`o=*Q3)b-H@GSO)pTf;(q)DUSa!{`JVzg8_4GZTE+w};1!%o6%24e
zRiNXZJjS5%6wkfCDt{<)l@rE&89aY^e&M6Oh&2g!nk>$TR&7VWBM95RKjMNL55^Zk
z8<K+Q4CvvYrA~*Tr<$YorgAZ&xq5bU1NxLh*5!fb&A4dfCL6HLgWRjWDY)U<W{CpO
zP`T@@RB<In=(uCV8*L#oB{IreYnkSIXPFJO3Y7IQowfZBxDsqq_?`!4OxR?&8NqtK
zS-0ibG+a4wz7bV|Cb;G7RaAKo$9+^e=lDUWlA^*V`$3>G7?r~P{wlR(JpBckCw~u6
zrrirID^Q%tvj;5sn0h`N8)>R0St2_D5*Ma_0eO$S3oQFPdFx+8#cAr7F%K{lmm-7~
z6qj7a87r5x!Y`A5p$YEr=j(4zx-dI)hNO%QlXa|wR^?1MV3v_(r;!s(9L*u|WFT_q
zzT`Lj5?57z4aIdbBizWKS$OBu{MME|oL3n)zX^41K5L<<UozhcjrTZd#qxMEH2oAw
z#@TPbiWNMZvjQ?Ik1)upi$;%8&snnuqnr!JNhU6GL7t<3LDoYwP!xeiL0jc-xruC3
zAq>XQ$64XR7Kd4BgZlX@))OyVN@9|;c~DYInXypyiJMIAyhLoB0vf1^JI7sv6B?@w
zE{1_}rpsr+*TW}9`jSo6=pn6pKNn0oj_d6_Q>FLyI#k;A!syekmN6J5Go8RAb4A%B
zYkNwUaG(%>c%c5Al@h5m3NZ)7GdJDko4zLm1ytaNV_K`dEPlt&`<{YE36YN2mZMNm
z647QyvDF;)qRiS#Zu_tZ0!Gda)SurPKH2LwJ6;M4NV-3<=tV0bLtZ#1W4)Ep28~&+
zvv2lftGY!{M*EddV)ct02s*wcXxaEN4*h?>ygvPZJDAS|QY3-+uOCnG1%px6g$b+*
zWT&5hPv4(@@2R#Rn*`9-8nYHaT*r8!9+U}pT&f5*alWXQ^|bs+DqYQdYe3XPg7@xZ
z_?lwYS=bF?7{~pN2;MoSl(EFIsB{X83~7!_96IH-G)Lrw7+HpMD%?y`s&y=@;OWh&
zz+GT}-@@@6ZMmH68tF~p#5Lh?Bi7oW{2<jOZJtGu9@!J%&w#oWjf1jadIqisb+gOr
zKA0mbH(1nMz-$zsgh68)q1dof>}UtG_ZS5B3UH+I5bInt4U$uVxj#MQe1(w+kSVX5
z!DSAKx(WN06Ilgo8g}8Y^3NAshCm^jo-b{GF6?3fFgNH@m*Rl39H7s~MB#Xsx+CQ@
z+{%-ayC-1MYdB?)`#WQV6g0T;JVBPggAP*a!%bpypAMM3!r*S?i5XPTgO6|BC%JvW
z(*Ks)bmxCSYSW$gj;W0SRBY&m)aLUk=)HoTy;Yu<T*EE6ouWa_eYuGeZ57pOvtp}%
z-H({{Wa~wirIqkC?@s7RxsZJuHyPXf#Y=%8j^lfVo=<ZC#BeZFX+bF^yW?l0-<U}<
zp%RqrGf6}hqyXB@B>VA-%rf486Zo;#PAv|oK`G_}g)E<LPDByCvX*;BcM{PEfgnMt
z+p(L74t%~b;bbjA+h3b+A|y2C2Sz@Br<}x}j*YG>L^l|l21929+oie0?)n(%oU+<<
z{8jCz0BzTaaZ_>^btM3d#mzC!V3%i>nabL!g!8QU_N^;2QhrJ*Q&#9w>_hRHIO>zU
zC0d{ND8n%ufh_z8!(C;l>d&z+q9uK+mbzHJ#QXEEz$c}qr1a*2z(<m}ho(z^uu6e*
zxJ$|Iu~B-HLZmg<0iB?k@QxRO?%vZkE}hmA7hQMy2ca4UR`gMzCOTBs*tc1-v`$vY
zP`cr_gIN5Mgl*5*9*ChBOQh>|d?<*awMvxo=B`OsOO_7EV1Tw`k|SE*g<f^+UCA`&
zp32g6aAR=Z?4)=~*;Xg(M&h!6S)*yIf<*R?%0@ke5@-xvsn^Tdv#|=HCwgbg8UiRM
zPSR21<4U!D`8uXDy(OxY{TlJw(r>vAk4S<nD&wH%VJQFZkF*!l94U{j8ML5JkNf`5
zpqwzuc*a8a$(keHb>fnU3sP@)%)Da~fIuyfJ|GDI?<5IWK}XxM%U7s>>3ZRkSY8SK
zU}G=x687JG=+=GK$|0=}{UFE(lF%yo#>ZejP>`EI_Ze=%`Z=?$HOI^CU@wN2e7Wc7
z4A@6Um+e}oTc4?}X1>i1JIIZq&d{GhI}*$n;JhA@K{2|X;+ql(&{EA%xH^F#?h^<y
z)j6B`ELL#k&90T>D%&Q125FC!*>(4MZPj&cvSFLz%6G+9sTj+y=GPSI3;h1f#^>hR
z(=5A7s~uViJ8j_7^qnhaGiz^d?aOB_jmzWuJMo5Flk4aF-#Qj8yw`-Q&a_?@c>{eG
z`vWQM=k^QLBHQIU)^y!GuH9C<N)JNUS>K4Q-YaLYKXTtYc55Xq401(GN3ey{_!;Vt
z2_*jPTmxq$4_W$9kwE?Vg)Uxw8oz-qqlDJ~0O7@+oR={VFcbkZm*HzC6_=37FD8G*
zSxax+HW0r1SLhFz;c#a70E`8^j<-N>4SXwlDw>>fX#W49!>1%l+SSI2d(g_J$l>st
z_fQYt20i@taDIFEbU_-N@(j)lZ(j$H4_baf%b?}K0Vs<wynP@3eA3!jdCu~X)`;>r
z`P<8zPmc&*W&q10Uja^!jv%g)IcR_KTC{l$$}ICDZ;<zM9GCS$TOBC`d3=0Mf4%+w
zpbN8F1sng2ul~Q+w}*cp8tVoX+c2<WMcNI2e}9M{2v}Jc<Yc6*_QQ{VhOZBQ^kmy+
zTGr$r*29l)LpSoJpck~Dr>$k4p=YdO_Jg%SVf>WQqt9BioQy8=$mvONGBAHT<)@2>
zRf=pCV(2LfjG6^o5=<#k&eY`K!}ycfpN$Cnjk1R%>5j6|GrrpFpcOjmXXAno84x|u
za!PnDJl%%$CZ6OJp}PZ5EixDKfCI#h(yp_SFFP=w?`q_*9P)0=<Ico|6Tg9+=$%=d
zA)>@<=v_{aR!93mnSrl&%hi9NQ{b@Q52(ZE4vGYdj9XCr{0@p@lLCHE(XJwBD5!vT
zf#Cf*Z`dm$J`x*ilnvoXY^+s*aW6Iwl+jBI(Gz^cCl+fnTr6Qaa>f3PM-|*cPSKn5
zOVMAQb(h`5D(o)bhQ=sk7u1jT9flP{_-NlT-}Sw|V@~b1?}*aSF9m-?QKs=Hon|&N
zjR_SqAPJ={*tia!_X~xCm<{0v3#HpN3WaQjktHEvJ5giGoTPELZh9JJbJtVd;n-jE
zKBD9@VxBtVp}Z?SSJjZm6!r0$<G2TwFiU!BDqI9HN~MlmkqSns057DH&cDx;F(~5}
z2*>mwrGPJYyJ}>k<lKKfMSxXVph(zACfD!JMY<+qY~1Lq%tZ2W3~$IlpYJ{cN#I1k
zz;(lwQw)krEsZjMhbgn9dDc@yr08CsX5{2uU-G+zkD{H68N4}zZ_cPR>~sdK(Xg8|
z?0kk;K%U_ZwYj=TF-rWLwUDt$zS(urhRAJk&e6K!7(lN*r&@p0{Ja7c)?cpGH*Hk>
zA$l1W)%G^UY;B6j*S?JIZ&}mT!ib`gG*#fre}%v<;!trIn^mGL7quI!5?@Xoiu1@O
zv&mPbqK!DxF;<pfkeml*QEXD)sAAcPDPc$X`4X#5aB+`tVw`eZTz{fnWLjO#z4a$q
zsXxVYb@ittI{|;_M)}#jQ>nQT(~5nQB!CY5wg^f@3ilU?C2Dp%z~@DJ+g4aoo?O2o
z7W(8zLmAUV5%ATdDBEC)dST)P*Pz=}g(?-4)0mN`No1}DEy`b-8P#j#5%&amHW|yB
z6VP*=yEB8Rlw(#9+FEMkH*hG*xHuJfW%XF1CUj3p#pQo(OxcmSDt@9JrrczyG>%;1
zF;cQ^9yBjl=lQZr#iX2ki%+eU_dG2i68gUJMft4RDHS<Ox#msOOL;ujh_@KM=K44&
z9kAcuOsr$nu+H^u=#!UXMRMbgVk@=|;9257k?(h?D3h66cjDdc;$0G6DY*C-@@|l2
zzFY7<<=20V+MXotSHjK)|4s%q{7U}4Q7)Km{`c`2q=a|WQz^Fb=^}3Xil;&;y20Yi
zrq9JY=$yqguVsHOM8*D#DE^{F8e@vFAJQTf7ofs{4bmCC1KXoBx;L7mv$-GC8b$Of
zh<<-;CEe}%&h|z}ZAW|K{L^jGo^!pB>V1vO+?s#vsCEs-vHeI27T_JN5D-5edH2v$
zyc&p#_|EStT2WgCdeVT(YPAw7jzG+K5sUW+BGysdOv<My9dHh7lyOHMV@~kb%iS<2
z1ek^0Nn%L;f2y2~1ju0Ez{)v>46vrx5U$DS^E|E2Wz()|8e;Kf?Vf6S*KS*BFi0xt
zyWM|2sKN%xGPS%vrox8U#Up#4sJ5cm{dq`<ksTLTs_@#E?p9o_!mmbKdMzvN@8X(x
z7cXjloBJcKsk&XJ*Hlz`c~t_p`q?@t3JcWV)qwylPGr1Q!KOzilTh9^#7Atm1g$HP
zM_{ZAiUZoLY$T+5UU1RLcC(_9GZ6U2>M(!upNv{x=Yc5MZAWFpO17&&Q3{U>D!Fse
zV@0KFp|SQ`(qGLAp4-Tc4otKM$xC*3e5NZ}4<>Ul>7AsWbm`ROimrsZ<|gZjQjnoK
zlIU>N1*}|6b%{nDBG+)K8kX=`?bV3G5$l}hUei`sVQ*Br+7Nm@Aw4x^0>Y=wZjgVo
z>9VLA0*n_liJs*(0a`m9eJSm!JX#KwM0IJS)23CTNwcQk>9O9FuI%a?6;puzJYK(*
z3&HIs0xIFVJLXzN6K^v{3v73JTbeKNThb%$2*+zg^r2>G?m8+pEzFAB=1vr+GMX<O
z*OB3F4Ta{cMpcHE8CCr0CUzuea~x5Q$BvDtHTKYnEwPPCjqU_CJhkRyud<dDns#+U
zcXf}ql&E-!yvEN{+#`Bf5K}YU$NH|e78yxJpeIPRQ|zkJ`BCX$d1qVSlG^<TEA}$a
zm(ffw69Y3fHJ2ejIV6{V;Vlc7RrV|>m!M}U1%EdVefO`pKwr8K(G@9@k`_gPwApmK
zaDgDn0=pX+g+IovyN<^gkBfAF{T^!O&e$EhzRtL@TOc)OhF2mf>hMsyIZ5qemy=pY
z(I9Z{SQ3OJUCP-7?KD;wtTQ1Oa#u=mq2MWQU)7FC6_Uh`Qjr}<)RB@8D%_EY4~C&B
z7k{5oq~pl#3GJA){c0U^8(G{8@>!KSR&6#iB>W2d@T;YWIW$4a+=-RiaVBv+jGW<+
zl#FL{ZR6pwB0h>?l$zZt%Uz0v{Rv3KZI05xVYf<ahfnT@MMQ#p6hlmkxK*Z*@T^w+
zkShDUW{~g$RjZxE=pGqCYENY8go=-h|9=TrkD8TE5JaB?PiPBwjxfB%8`f)VHp4-M
zb1qqjV6^Wtq~uoR0;$C`*G{di2R^EquVcX|IV$FGb8}mTk|4D>7wNR5_AI_D*`t9U
zv{tvYrp_p~@v!zp{%eh$rM6md!Th5d!fuinb5w=|SDCSRwpjg=A*s(+@tsw~N`H`R
zR+=C9XocNhpE6q0^T#N;9wu+kO2p(^Z3nYKvedqVin-^AahEf*OBf|%p`;RiC@rZy
z1&a6+zKqo)DM-B%Tf{5U8A<8SLP|tCkV;oXnx(ZXhQQ+4h_v^~g{Z4-RRtfB0d6VG
zt;zirM5Dq=m$a)3fAK>t?7t$-2!FZZlVGI~U6@W4c??mS0=tMs5UZ}hzC6ONz*rw7
zjf;ueu-QfR$C8Uu-8|oJQQc`_{}zd`^VL`VyI;>I{p<Z7C;d0mv-#v~e&J~zvH#!n
zZzdPh%h}=NVzF2JtM4aAC;MMdKfZU37WB${cz%W_yNn&yv|gN@O&0^1e}8`deCK@w
z5y%S=SqeG`HV7-7G6;1kNE`$j1RjJm2*b!Sj4Z>*(nj7}x!g?b-(DWf{nPHr*?+tK
z>*?%hGHb5Iz5eh0cl|f-Y4PFB<Z%8T`4F&`Jgox)iWRCsu%D39p{VWkFFFkWw*Pke
zZrZ=>z9d7sAlpifr0grn_J5b>_-{Kan1lMptQs2fnFZhE7Bz@$7=hhwCR#I>T(F&h
zz7Jr-G7>5^kmp}ahy)O)hRB;X6o5KuF)`Qy@Bv2MV8hMRVrO*pTh~(xE#s9#WgwHp
zKt=X;uDQ{_SPg$TKbfAvpntyk&i{S+(|mq@@z=ip@nrtf<v}={et+nHzWi|h>}c{!
z|BuV>Ux%Mh`-9WzL4SN)tR5bt{u~e;Q#z7#ENPM_`QRuY2_MH}smGQ2i`ij6J)fU^
zIQidX_QMIb((HKuaDsup*d-K{XS+la;q}gzp$B3IpdYG*FE=U)=ir4ylHzaLh&^_Y
zRLKAgpdG47dWakZ3V$Nw^V*17f@A4zMiA<yWJ-u5iZT+>YThaX0C=I$Dgjc~fMWgF
z39x+fi(Mq-vt4aFZQJ$c7(+pw1HlOni{L52agsF!svD68<V52FxM9GvW@UYV?QVpq
zdjYqAWOl|CQ0^G3UG4LjfT1vTU73)$1}?YSfKnHcBb0NkRDTgGCfX|Dm|`#xg(2gZ
zGD1+FEac<@#3}g!5)i;0qSGM_)VMfu0j0M}sI_bco+@IVKz9?Ct5oIm0QH7hFe$BU
zz#eeexy1&-mz=W}@`_XNw`^}h!ahXJUs`~EB1pQ`0`v`*BepA9c7j^K_#VInh1+Q`
z!7*5gerF+2>VHuwEBJhI{!P3x;U)N3Yx3Uo|N8fToDGJ8P9=kUo?V`v?%kr5p@qf|
z8Q{EZgeJ!Ij_PhrFhvlEQJO+Q&=uURNhnIrNJnTEU@I_ot=WMNpH63Q&#|Gp1-%5z
zonwogbqm_r|Lx=a+qd)m`J`JA(*6|)H0F2(ehY+ZRDZO8<wc4Y1hs#KzKAp~nDi?&
zFle6Yf*|*=XVb&ClllAp^~+cNyUEA--j!teEkC!k!Ox98g`Znq^K%`zw+`G}2kve0
z2J(Clx`y4GbzQABQDA-aYW>Up>}3DobaK%Uvt)a=TjPxs?H*-9P@d|VnZnJ0X790G
zJv%*bPJgn3mvF<Pt$7JJctO7ms0)7F!$8E0W)>+OO6Jei3@wWj`VmweRAm4e8thVn
z#Z4NTBPI06$-C-o1}p3uBpR=v!H)wr(eR4G3gd`__nUWXidQWgq3EH_C6Cdh5)jWn
z0=TYW)io5`0=Sm=a}~I9`V`<Qx&|&A@X7|f%74RIK_1QuGH-`qIic6S(r^|e{o^;B
zhtso*`D`E5?_$|#S_SQkmA6AN(Fk&_D|ZFe89m!$yL<7|{`q9^l-yC>&d(HqRdl6|
z1bjS!Hu8Li6<y%9vEy~w&P^%?rYI{nQJ?}q_zAfw3OA*0=cc6q{9k!#HDt<9$VtZD
z1b?)zopeyoGKPg;(^}RYG{l_C7)@{z$s-gc<xtC*$ps;=DkC(#Rns_Tb}k%bj6akm
zaI9lAwFW<k|BSr;z7?_5jfz<Er&Pq6))leXcGXV7KX{E7YA6S+PxY)rOL?}dRj+TX
zQtse#R}_YPF|^jTrZJisU8t53nu>lkntzSaWWaR#(~vodo3}*uQ*Md;-fxw<b-2Q*
z!xc{5E}7AD1k?4wo%r#|<5QXG>7`Hf!^P!?5BsxUTOudqag{A84QuejfEA@duU(9v
z?B4aR8_ar3f+Ho6uIV|U7d3sd_d%ST)d^n*tUBsEu5O-QiD_f8BQ%xTG3wo#V1L1&
zjA?`>`zf1@(WHdbj5<b>ftg7^ClcNl{Ai<Y`@Q?!;3q8C!H=(WvUPa`=kFXWg)#`s
z-qv3@SPC=<JP2tJWDs-^Z2RkPMkOmS-)kR%w*Y2>zY<~vt0lB~4^%;USu*Mi{+eYg
z4|m{W$pGWE;}l^8K7-^&-ZUb!W`96~3L;4AM?SjMJrG6+Yi499*)`HrI|pDA01V3m
z-u@{2eV4A{t%!fM{D;R|iGshyvErfxh>-lW&W_2FpSzfp2Z^|aY`K#N&;3QWj7TRC
z7eeIahJ#gv3DE(qvJ@rNm~woW8~zCDHCUkn@YdU{FvFUw-~1LgM<m&H7k~UY1xLU2
z4S&T>d$RYfQZsl=<cC=Q)nz+KjsF!Vu)g!K6ZwYg_UbQCMMBPq<tQ;d0L(I&g6Z0b
z<P59cZo{Z6|NGpwbMTk;*0=3`&TUVY&_%31)AbJ2Un~6G4)m(R`i`Kqw>?H^7HKa;
z+^vb^z+V}o>9<#TjGg}KTTZc2nyud|f0nydS-g38jo2+-PrNU&TfCmwh@A}YJ_-+f
zjt4%M20oVtK9>eQm(q5}IDl*_S4yYv>GJ5M^Wtdvh6*?e)n(`Re?#*mmjT%q69P9h
zmvK)g6PI)cFDaL0+b@&@kB`r{{o5}^2Lw&^>EV|#4=@w~HJ4Ed0u+~T<1fB{Z?C_7
z{OgG)WD8~}n9^?1MlynxF;fIp5@ynRNXD-a97&icx=Wtpcj>=h{4QG8WheSw{+Fjf
z>a+rWPlOy`$OtE_oA!KME$3x{>rab#n};gTAWrDN4>yDM=S;#&y*hk}9Mn2+aDboo
znsjr1CY<@@^;v@X1W)n}l^^DRW_}dx(n_S}HR5sdZ#|6!6+K0PP7f#28+SqCl*zk2
zX6hWZfT6M$7tSPotWaY4i}Ye%{qdc2+QO<%XU);bIGswG(_<X4ZS`)`LKNL@)dWpu
zZr~&Vzk=_PYElcbAipYJ8l7~B2bThYCz$85SGDk>7LZ;+(VoiTA{j`3xG2m}X3R#(
z=^{BuIw9RUU82>JEKR}#GaktfvKAH1_21tL%w65YPB(E=H(#IOWlcY^v(<qPFY{f^
zA(;bKpJVII;kGP2Oqk1L_ur~KNW0@r8(L)@Gjh7g`vQrc&!S-5%G!J1;)!(j#Q`@g
zzY?!JKb0N*M%iV$Gx@fEb37mMkRFdD*Lb%bbpnc~x}-OwrOO5b*U}L1^jv&PA6sQx
z{6mDVvb7{C-pZSEuM=Tc@|#41m%t=jflLXPIWBbd+pXSu?<$nKLwAf|6s8974^SPH
zNDc|7n_34VZR<edFB|s+1NFi`cSL4$i+e9EW_dL$R=BCa4KT5PaZs4f$vI>Nl`W@A
zfUB^wQp!ou!ph`SXAF{>cm-A<8$(h_y@=*jwLs34mBqbhm2-tR$<4h!qsi$*R{Zd(
zj_$geSY{=R_dP^U3`4G`o3cC#CE?CfUQN`P_%_PZyXt6S>(n0>x&<nRIHDZHjWIt0
zNIog4R$+bqZL<e|t11}YWd@#JgDb^EuV}QCr{{ZdW49y-Qagq)X;9YX#IHuaYH;aj
zx+G3!KD*iZOj^q=)95;_t#&KaOQ$XC!9ypVfXtwZU2%vFS1KI!*0BCmR;TbVSuI?#
z+NeoRd#3xx_u<#KkEj1lY><NFBK7k7pHKdU1jV#*l)5>8COIyD{b%{(>E9#RBv#Hi
zor!-Gk1E@({740q<0O5qwNib`pqWy3QI?+87D>D}wv>__gifF84dGk8IIsL@iXo+$
zAYVV;a&CNjwo@}`ZLqW%Z-N9k0>qR9_*%_>r+O_4WS>X74APQ>0(UTI6ZOoD9^$M}
z33us{%zpxZ?%hhuW(%Pj9v~xMuTUMQh%d+8Vnwm}=?ZUe#}7%*DbE;h1X#rnB``_3
z<%r*|+oOeCjgh;qe*$jZO3S+}XC;_{QYRr*$(pZKM4H1Yw+18m1O}8K8G4m5a5XOs
zq3)a<<q$o5;uLWT+))Xywv)g!Exqd_3A}$DYXQuExEk|zyBguvgfWP(-;W_YZ)<Vk
z&hRE;5hN^QY!;~#<Y9z06Js~(uVcv_vqralsv6nbB<vPFFBIvTBdD3GqO*yzb4(us
zg`!dysP)hH(kuHawbEJUz_hhWsqEZDMb3EmM8SsVxg!GZd3R|vbIu*;-3N#@2fW{_
zQ-TJ6lm-kCH@s%3<CLE6-_1Gi_vXImD)o(cAtf~vSJpG!ThrSZmwJ1S+3}?b0S73D
zQeErp0Oc4U<~9KInfoU~{{VR>^bfl~SH;~XhEI48DJjJhE0K8i2OGXUdVO3;_Twsu
z{7sHsCtgebv_sZHWIUatt-mGppL!3qmbq_#_SSM$K(PWV{W+p_ol|T*X>5q3g6$mp
zps3wgG{=>$?Gg6I^5<N3G=)%H88e_2NI74;tQ3HGe}liQQJ{8Vb0HT_*s0|AvzJn6
zx(;$KSvT<CD=*_Np0o>+F#?pP%U0R%%}5F{N^HZdHw$qxVIX5%g>r?eGQFtrGfZ=T
zE}LbyaE1|UTQ4kCuTA`13gWgg2VG+h$`mF=7tbm=WFn;-S$eItoFAoIOZv5jPJ-~D
z)(!Wyi5_`#T<<2j#M8088$~6ORjUGh>qiS4?POsbYpCt|=*00b*9%R=#xkdleB4#e
zVB1sIsU;frtOuA50%QnXBPO11JzwL0^-6y~%w@~eru8#=Zsz*@lAevPO!v0>sA0!f
zrX4e`*D1Z~iITmT>=7YV1j!TSH>$;8nzrVdReHNFd~KM)+VabO%AezR%~ZLm{`YG8
z*hwa}9B;nUQzBT#-GqPO4IjOv;Kgg48vpawOxOA3=OXp0g-~kmOFOkpT3*k8m^y_j
zV}xOH9()<stqj9ie2){suGz>-t1?g1pyhW@<q)B-#w^1bh)EI~U8qe5<TS;sb$NE2
zoklFS3;-srq-bh%9Sr5sPMx}8%<SN$nMBu#*)JDL9<St8LRiANcE_ga-{$1+>O?d{
z_igmr2f@t=U!f>Ytgl*)Rt!OZkFXyxbP>RWn{JBnyLg1wvzAC-cF64(d3j^2JvBq8
zV@5DQ+9O5^YE&2^0&aI+cY-ga0?hf~JKQ9vfyHAo773D($&J@v7yNEUAsw?g6LtTf
zT5v%{c5QXJ80LC&zMi~xig(qJY&E2COKxc1JBa})LJmGoVx(nKik~uni9xScT0M9Y
zGv4y$Nz4H7fs+_ij*-|8Y{v``51hmdkayZK!|tCji5YI(nZ%6v-86|IsNas0y~kS|
zDf$ke()EA-LS)#;`R0b#;Y~;1FGaw++<-QVPwEyZZlGKFaRWR0TT$%(zuUS#OQvz6
zVUufhHgo#CCJ&gyWNmkUL8UP0gdU*|bE>jiNd^=1zR>zaPNBVy0Ed6|JVL!F$dSO+
zftEBtJYe84K-?L4jQ2lo;BgyXao{oX`5!Ux08OOBkr8%;m?D@=b?qmFkt6H?c_+dS
zyZ>GT4<M%_m3`>IL;j5i9zX?+s`S@?-Tiz%uO$;yBMs@iu!Mw7{3+SIL#E)-|5i0z
z^5tB<hxfo#-cW{epqFXE9hk(oV0ZjWN6=j@hvQOvt(hYS)1tB!q6{$Py%@xeQOGCu
zU2?x_M8lJ8^#9hF;g<mc3=;!4G&h$4{WcSq674U(e`pVXef+-DSy-)tjkobqKfQhI
z{@L}SBPq6_VQ&>#+x-1=7hf1)WnE|jDQiOW>s#}E_eV{&=cQ#!`eEJt`q5M~@7Z`k
zi}AFz%p>EWRm{Gz>=g;&Y&`R-HH%{ABtw)pVkU%<lFug(vlLl$p-@v~F{crVEG9LW
zpO2YCf1HQ(PK(RL;+kcRap_%LqmPOE73!|)N>&-?Z_rhXJA^z?1Cde3bu{ySg+0bx
zQ9_bK;>J91He75nlAxy+jIub<VbN-*!;{r9v^j^UGSN`wr_*FK^FHuGata(xB##F{
zSN6GqstGEPt*HXd>Wgd?OpMnD15-2nbZkd6e`joa=OZf?iX|-Tle<B9X*sqUw>(M}
z#WquVN%w4Tb}e}5a%M%ja_bOuT9GpoIXH)GJj!pwF*es6W8`Ge!Oo+Z^MH{=V$4KV
zZA)m!%*+^{aGxdcdANSoov-O6HrXeqewOlabU95G-P)Y8;-LKQnDpzeuUU6aG@wo%
zf44^|2b9!Ly<bd`5wq`OC5rddTVZPLUiX(QXMEF^Sup3;=ysk`q;`~`i>tE26wmg%
zsi)4_GA}8qPqd$}TEd9DOcPsgV>%7u+GLx#H$#G**8fsO<+QLnuQ#1o#W8l0Q}aPy
z1Qg<H$?^lmQl+X&W^(rK43x7ur!Zd>f4Oh~ROkoDl2e7mPzPB6?&P`vl;jvYq>Agy
zDFPaBk$*TaooU(yMhSLIbH%;4LLB2*Y42)?gQx<S6^K(^J2<nDG4uH`lHbUh`&^f^
z*KBb<#G_U)OP3Tl--kQuYgB$BEWE}sUkwc?gRb5YlR=++{h7!S!d}D}3En38f6tQh
zoEH|OsyNuN!qCUEz-(joS)QDlI+OYmZ7Cg<8NXkQ>Ab$Asp7~kO_vgMEC{mDN-O)P
z39l=#*V5@FZ4}q~@`{~jVr9vqWI-WLIOLNzu{%0NzAqJIOZn^VmJ40Vh<*9>9%Dqx
z@KnmuWcE_p<)Cz+nm;ZaDXPxNe@axLj?J8}sSBRnN`BN<ko3fUo$Pis$F9mRC^ur~
zFO)wbW2qq(ha2j|nhqpBFoM)84ml2=Pm*UZ<pUiAvU497QKmM_pc9bldp~<$O+hEk
z7fssct_YxgNZ(A-xra@`enoKx*4+#CE6OWizgkUmV7`qtLn7jBv)){)f0_#3Jh(Tn
zxPG!Xuc%$qo0li7GU(RUH56ZN8Lbt?)$Ez=(SE9lCqYhvq5CtEkpTONZ#J<SNHF=Z
zrPan1?}Df8rr*TSdnM^~Lmg|1kBJO)QmbI7=3$I~CeiVz<}@OAG#fUjD8-J<dUJXl
zKCO0%r?d)uRf7saiU8l&f1s{$UTRWVnVRu~dejJM9Y3l^1#1PpnEO4r)TgE%l{f5D
z(|6!X;ER2#SWdLBcKk2;RKOq`KG&ztxTNnyPs7KyYkSwj6YS6XSNFyJD+HOW4Y6ed
z>*AL<TXR@wU;$7`J1}Y#_>hdi2gG6wPG2OmL64mjqh^2PB9{S|VlTUVF)^NhCXgM8
zQ#+KXXzT@91>u3=|E$*@v&RdW;pTQb{sGAqgDtyNfLJpt1|v{#xn8Vlz`g_DU#Ed3
zWuXp%8>`-G%N<J&)LLBGam%TbRDDasT@@`UivI#IU08RQF%K{l0XdgJpB@tdGnYZ(
z1t)*SS<7zYHWc0aEA#`y@V@u}0t1ONF3?>E*%e(ComtJI^Z$!pQWPaomTV_Vx{zg4
z(tVu!I9Igoess3`?eP41csLVx-UI=U-Rt)b^n%q-u)64^^8m&Xbgw_VKOe0{r^mA%
zY%P%<UcU!D%kc|A&^ti%(5nE)FP<SiqxXLh_Ro^*pCP>qPkIJDKc#hjK3Zp!gs6wg
zgwye_*WVAe7;8;*>1#T*zrMa6{ykLFk&L+5iICwCyT5-P(gg-F&PToMjG*g&{nLFv
z{LzZ7%B()qe~9gV{pgyNZw0+*6+P>m$PB%tLqg|VG?Xs&=<$oSqMq?t5A~iECxm~Q
zmk;Lv7bS`_M6s2qFj`S;Suy1-xtC@***Sb<^Jgn+ew7>`N4k-Wc4k+-J6c1Y+q-FD
zhc0NISoIRV7Mrd^dKFLhhRNT7r`4FNc%TlFqO5hc^6du9=UP#nmDp<&88x#MOTr3r
zY#c6pNGQoA^v=gG&Zhd2=lrY3Q1gGfAcj)BnXYVLhzy3&FT7FfMXON!O?imI1{RR)
z9cC0Gs%6KrX(T46wG}az?-_ss{!@QWq0Pa`X<U30>c)@k+iYc8R)a6V+Kp<x`d)B$
zCp4M*e&8gcaZEb`*GNp#GFYyF<lg||4t{!TJnq?7TZx~SAR^ItS2Ki5XGwp)^eAt!
zU|m)BDd;kZ3odsBi$(vXD`VGNGXFrI$R_$wBk%eiCB<`I+E${rI!ka=&KR4<S}x_^
zr=ge5y3?x7gNC5M@HnlMyh!YnrWJrUYF(L;V#*n6v_75FECW92k4K@2rwOyH?^Lja
zJSX*;bLVu!taK8voo*0K1ipVV+%)Gr$gGq{NmAuD6D7PYbZaDc_o#xFl(&6r4ZNh|
zBI#lpU)i-!;L~W|P|DeEw=&9{E(NKk{YrWakpgF+W<3om!^j3hz304m$#qqx=?v`i
zu#$2reS=4!=OU-iso`oiJYINut9p<?J@SpAI}`}&XzBIRTyW}e>lJ@d{p50oAH1oG
z{bh8*q)X-Q8HeCj30yGulaR|8i+s1Mce}Al`zMYF1IFiy_|k?rNh&Ea3V2!0=?ILE
zwe;J3p%{#VO1V_xFiC+4JtW~n?&|;;v?FNM=n-fjpvCG|^LJ>(Lu<nwxp1v2&_)v&
zTMe}h?oVW-S!gG(Oz(duGJWwe+HkLS!y&P4i6BnO+>QxR5zXa%6VWM+8avLjcM;u9
zRvlBOUo>nbtQd@NW7j89p*n25#809^(hz7-?jQUfa<?B_YAnE0Yw#9ZvLdJ6H&1*j
zTcSsUT$!2{@NP8T8RRqCnCJFz{~B3tlSzd2-kn~LUHo5L_uPMk9edK%&d%OAX)&lf
z7w`8FMc+kb;o--5q-mf=wF*Cws&dH^?MPK=0Op9ZZbki3OBp9~N~lxUEi?p?jE(Lp
z4c()0r9A{OF}4~%iD1$uLD2V(&mkCa`j(?|?+#gEj`-#+;70QXQw-a{%@dnUvAJd1
z|0R%XGrp5SKu3QO8n<#tj?9l^7wn7a*mro?V3P&zl~I<dyI_>Q+0R}HHhf=U6RjWP
z-T2m8EYJ4p%Lo>Yy=A^^TTK`Zhe%Y*eX()&qLcX*r_7j!5nr~Rhpn)L(2AW}liFu)
zWxf3vPonk$YSmw;7ram_W4_(SZm4aYU88JIc1v5-${Jg91vw{q+>E@lonG}SI>Rg<
z=M`DyA8KJ+OeOnzMLI(g(+7Ixy&-C0o6f*I*;4ob4_Svj#@549o)2ebk&OoFmA-^#
zU1i8JTP!@+9x>Z|E@vgl{{Wiu9U_;JDlilQFqaX04HE(}H<w|XK`4K@jwCk>&F3rX
z2UaUeqz<6bz|8E70cT!JW;imu8*SkI|G?B)C8>_?nQl9A(3h%;;>|~@<@9}$(=QKC
zZx5eeHBMGI0;{ICZxiqZ$?qU}(a>Z8gwZg){g{3~NvRCKp7@1QYRxap-yMI_`I836
zcL4E=uL7JuS%mNm-=lw3dRCkCjOM%4Gk*eqK8AIE-b*6`1)pE~{Kwm`4>B_=g*V}A
z_~?JVygmH=P+3=6knxj|PAG@dpFbYL1q2etdX`977U<_+)3=A;da89V$&aKrlGD%c
zQ@8RpE6+<-o-~H2QJyk_$W6w0p~I!5Jp3Xhv9j=$Uo1SbniPLDLVkXAuu74P(AxJD
zS&Ncbwa6;PNI6jx82Bu2k^N?c*{_rxL_>F!m2CK`cY7(c#r|$R@WBi0CX%Ium(0_#
zN)PpkR-m?5=u@)EnLOwT!bWk|X65SzI&XJXEG+r78_~QoQs%^MXby!@gA74Sw5Gk+
z^Cu%i_|VhUHdRB~3b4B03tk7iUPTqX@Y)PiW3Uh5TAwmX7(Zz7)}EnA5cpe0T1zod
zBb{EsM$z&rSv}Y;UWa7EXw&HZSw7O!{1%s4`Y#><E0=isFN1&U{@s<P+Zn!u*KW1x
zo=^*(C>tR1h1n30)^z1H=9-dE^J;uD=LoXb=UD571$%)sA<RlPQ<Q7aa%CsKDaRN_
znLz_p%B!>amS1Q|LZKdZdVjsppoJwhXb4wr&FGO8GDuo;=$hMaMsbNKV}wUlQ3{<A
zPucpmq_1;?t;BzDm2W@gaDIxJf>%P&$Go2_eW=wbPU_V$!|;_zQ^h#R_u)leyPNhD
zr{SQRPOgI(-O`fg8sm%}sA1+WQ;o?z*H9`5#a8JbGxt$zb(PKvPOEX2wOk}lnCBdm
zYeqU-(xc}5;i%_#C!V(?N258X4ti|uwc-aErxojgY$bn9Ww5d2;E{yFyP=R4-)T!p
zDLRcd7k(SB!EBLqcSI!5r5#&Jkyn;``hsT8KJ1W7*T6O7Nl94asL;e@QEhdxs7c6G
zJC|3g4{j84wuIwE)a~$2j|T6Bvje}0v%+Vi5sRIEan^`Zr`etv;uIR$Q?le(DBU2&
zHP>gpZ;*e%FG0tCTqA}w*%1iNT37(N&jN!W%Z+nKfr@+Y2bbms+nd3p$t`R#hh0<t
zn3N9;w~~4r1MQ|06B?3_C>os?ircK7f?MINOP<AXwi5-nAJV(3`zDOH%1#dO;+$Iv
zV7GbA8jp8I09X=y2@EJ#bD~M*S&?%pKmdL@UD|(~VRfh1P=<t;I7q$$OcT&b)1w-s
zZXs2eG!2rT;nWJd#E1oJ9H0yB*+ksQc8tc3b|Zgsy8@<WeeFsx-AsAhAqvZVOSe<a
z(kAE86=jmSNa}#asiz+UR@KU5imT=CFV$6^mo6FeWpc$GoRDc&Y%o`txT{&#yJEqZ
zdn<p<dPmW~%K~>iR&M6!x&Udlv}s)((g5qeSU0E1cK=8#SBKIc+`{fg39``hu#SoH
z#KN=_i>`;*vcCe}fy+v_Q`X7nS8geHka#ExvxXpZV2}Ip@;qsw%{+tiXHBPt-J->#
zjm@m@6#C(+y0|tkFpL_CPFz`|6L*5_w`qU3f;r-^iX8o-1f(k^g6gll+df9{0;pHY
zrS%z!VgK60Gat1?ugOxa*Kmz7-lJ7&nV6r3Ueb8KS@p)|xfRWnm5r_}UtOR341R?m
z5QYwb5$*su;0hlKhacqhnx{s}_1?~ZN(Yg^5)C}a%ES2+oIZv#r4z=uUISseeW`zx
z$Fu8j7{py2{|YvdFXbDU)T0dRSiUjIFncKLu2$!5Fb2Ys?d6%Vf3awhq)4~O|6p>g
z<``Y|7>+$8+t8oAd$-qN&OlO6Vb&3>>1a1QS`^?%ikCxua132xkSb14Rhlz}DQY<4
zBb=dZu(GyEZxM{yL;ENqVMbnS-T{BR3szTP?mN5UW}oLJ+-;%%S~u=a{WBxSyYPRY
zc1fe+kTHSY_8WDr-|%mtaimxV*S9}F&;WuIUU7hM_4r&l6-W2o_#8|;wAvk?Uob8Y
zgIGTwHEeYyO>d<(%g|M0^E?FaIJohk8N3tD>3$cr9}<@ws3{!|Fa`~#BB6gyH$7rd
z3^SE<7wD=V<M(pE<A~uyqiRJ;Se%dFcV`G8dBBi-c0p0Pe>It^W@fNX_ZGuzkDPP{
zN#8rX=G5gJ_SC4f)^nS;Kj{js{J_(36wlFq1WSM^!yq)48nR)=RYSqht(FO8Mbl5c
zsdxUNu%#5@fR&sr2BqxR6m5UvX;1gm9czlxDBJ?O-=n5T5?<Y_q_As%x5p*r`jV0V
z!_e)pchYKDmuUMZt{bjxPuA{s`r7mRfq%E;lh$5m>!+2<hrciMH-D%CfHi_a_r-^m
zI2=;p7GTa-%JCMuk0l5dhK-ziJB|+$MELs<*R}qsi;o*|^|%p_#)yCEx&FA(93MAo
zLxRZ<QeJ=Ds6E5&QS#Xy8<xE8xUtZ=ur11Z$!$^Iu^m4R9?D5VTDy99dD%fCpb>!h
zFAT!Foo+q|?*N}bV87`9KL`)jRh&ZSg8ZKh!aE-BI0)}{x+Bm5U|m>$PoU#KXlL&P
zbet8&;D6XUw+!5=b8a`eVdLB+5}y0!cK}0$Y*^myP$BcwE=(AE*=~dici~e1$!_}(
zQ4!z)mys$k6aq3amvM9m6PG;!Fu;HC{O8ASCz+M)G}9oA4UKguNLHsAX<=v;(%yKZ
z?>>Duv>W=UoqNrHUCNcHZ?&fXg>=Ty{WZJNkS%!4QcRLWcSYSl-MpIet~Nn{)Z7LW
z@~Y{WXlsShhOcUMu*>^7CxV80zkvmM$`?|~OS&$vi@E`zuvkTfvY?HjAaQ@-8!5n8
zZ|Q6Ung5d_JN?W1*I8S-_!}!Un$8xRA{O;-?Xg`HN(M<0Z2THe{I4$`r+?}c(~i`0
z&0jxG@ds2!SQi>)MF89U{MLLw{m!KhORF_NT5HVM01Bh=jRl|*10IYi=Pw5-yBJA)
z_tCm`bme_eLV?`vkZYhm$Af>ljNBlFj4~+mTB)NeZJ>gjv@>|&J@8$0$U#ymNNr(V
zbMVw?_c;%0&?2bHgYtHMu~NS#-<iwZe$g1($gcaPFJv!}l91c@mUU41Tp+?G=){nY
z7PMlFA>EUUY&xZ8p*m-)A%AP7Hv-}kTqsNVVKtXH@an8CP76@g$~J%AHpe;o<ZXHX
zV1-n!lCmFMX=8+j%4Zo$A7~aanr#MmjS#cHTOhjZ&@`xG>7uE<vJA{F5QbF|IS`m>
zoj`pgE*qnSqx?Jum+27E@>N{EA9tx`LCm`G4h#6AEX5ohU-2lus4o#A<51SF$W_@|
z6|HmkBWA|Gb&H<qByoT62?i=JWHc#ycRQ`?KigS6h~QMUle~L}PGZmktE8HlZ8ZU(
zqdI1HKi3vCL9*{AGMdy83z)l7m*)gdG_z|T81zxYo`Jr6Z=RF$ef`SCiIi&AdBe4P
zEU`OHR_5oRXkyfS&jk8CQlagkIW<c<T(`$tZeuRNdbZk}wR(S@EUCP%YkjHxrUr9a
z^btx;B~tllp>*dgM;cE6ppsM~Hy3l6u{)1xN{%f}Sl$(3kyHGbGi74Ht8!FP*+Abj
zhTg1lw2w$?)3JS;>jF<3>Jv&{orCJzAlC<}Xe|MCshMnN&m+4^$g&IA5s-Omwqy+*
z9tg8qaHy^>wNZbog&_!HI!s4B4MI5Vu{9vh)`JRy5Qpz{;Ans@wX*|f14Jv74+qT#
zXb8=%`GuT~X>br{W4r9g*_h<rINQ2k_{_p>TxdOjJ%2*CGjBu6O|NiA-o_-}!Q0r%
z2YDNpvOjO*7XBZ2TjiBayp63o5S$?dA@!le4O~o6JP3cB!C7JPK;AaqU;}L%FL(xF
z_GfN34woBjdWyI?nz$v8%pttZ3#~2R=o#9Um>6@<G;O08iQu^}pK1z}53R_ptPQIP
zK3$n&ZAK1fe{&aWLyBjm-Q$TJ_4Heq8wh8C@4K0~;Usf2tIQ1_W^N#Yv0UGWm>YTA
z`ASlWULAjT^AK~x2bmjaY8rt58RiBmar~*ef!wZfV82}%s7RUuE*0Mf7EX`$vzaWF
zFcGMj#L*byx(J?MRPjUSwQeZuhZx(}LdmZg|HMi{{I$%M>2M&kWjh@>JZVRx3|0Jr
z!;>av8-hP*c+v(4Xx5o6JUl(I9X5<lY(;!oAVGgoF68R~RPt;dUrOlabWCMbO0~46
zry8?XPFRgw8}l3^LEGQ!GH`Y*SdGek{sn~$?io0GaPI;3TC|t}*j0~qcdq(%nFGZK
z+A*SI@8xWo<v;`&&LLlw%lOlc(sm(IiujBlyt<MHkD`6o4*XO$te2ybCjOg?Rilqn
z^>TkP=E=GJFu8U{SKf2M&OD;Y1*@DB{p%0-xd+cO0_kl}W=egI1(pc{%N%vPzwo_k
z-ys?cW<xoDzBRrXDlHNW$!klAl1A(9T=k_f!t~FdH#n0zgCp@Y8>qNZVnxF%?;9LG
z-c>_>AL_Qb+^R=g(g_HfOKyRC%9FQ_RKjM%58k*H@;Tm>X(^N1zEK9#HfN5>Z7v{B
zGfDrEdi?OpFkeeHAkv%y1CYv(D6?(ZXGy0BN0n~7<fJA~m|Y8eEOEZYKj*#p#$d4H
z$jx|Z8Xotb&WM-G`(71eXX*g|0v1JA0hf^~Fcg=7MK2YX0SGX_e>(s1@ynCU%1Yr)
z_&5C2XYU_Re?RrAD->jWXABByUH9krr*J_731fW+MtC&ckH5OFPrueu$G|vXV5Bib
z6JV4PM2@9<f#FgDhFD?EX6p<8&wyw?<ZJq+*JOy$ETaGO<6Bpa{PtzA!AL<K>s8=$
zI3}<RGrRz)>KOrGe`4?nl5YAf|L>B%M9G1~x)P@IsIy5&yg$xA2VcVb`PoQys%~`I
zaC^0(%AxQUb{HxJ`&0@UWGG5fCnGHN)5Xg<rzx+5ZhWRP7DerTp>6e-q64KD+P5po
zt5@hHf@2F!1CXAL0#r&U)4(a;vSa3geK7@;eT2v|RUc7wf7bexoo|!#iOzBJbCf(I
zhO6O<j`1Hx%<K*cSLl@DOZ)z8m>G1=NH^Lku!1Z$+8q+_G1ikYbee49C9neUoHJ{6
z;*wc(G&3&|$8A>2_NKpN1tJ9fD4i9YW_MfBuNWC6p240*`p(0rQ48Prc4bp98LkZp
z@2HWq<kK=Df3u%Vx(ae?Xavel_Gc593IsCbWyJz<j<2}pkiED(*&57I^u}v2a!4G}
ztR}sc(l)Uojf~gySt^}<OqDRNwM2oO+2?i>=9twZ`^~Bw!GKVJwN(8wvasEEYyMOc
z1S!)<^chL&k&GTZKSRNxM~!j>4LC(>bQw**3;2+ue~_vRJb6MAWltkhGW4ZNa0PHq
zh4jgXL&B6+o6#Y8&w!~or9h>^{GQ<7yqzN|ZnlyvVhmm=&cV}#G9X;|okYj_A{l35
zmE*b|t(m>DW>EpvaI!SbR!PBMg6+O;i_i%xAJuJ<=5*!sR<~8gqfu%x_OI&1Ji~2u
zt--+!e+qafoWpt*<%ihPrpam3Ser*sp%NY%1m&ePayNFUO8e;9ZCv^p<b>8#&p5PV
z-`VPrWH<PM?YNl`>MQp~V&<^na#%^+(Mj;BKxk(6NwvW9vcA4ILV|6iY;KTYBV{BP
zlNgau*kG2@d|wLMk~!A~rXn#ZG43>kK+I3Cf3<{YL++(Qp^m|`?iJq1JV-4n>%E=z
zoOv6MJTPA~6wHd;BlET9Vh7+Kl&`gM+&yfwV{?5zCL<7iy}~LpW~Xag&)AiB%y3WX
zRHFm*a6je88KLpcQG7g0PW*M-=6h-o=T^?k4YH45$(hsEE8W48L)-y12raN=l)^Rv
ze{omk&|GFw?k34f*MvfmT*!`GX-9-ApY0{dT5u(3v^zBP&pl`B)zfB!ESu1b*J|x0
z%Nc@-Ea%Y`vYbG#$+Fo)mN|%_f$=7`Y&m<il9rrXM($_JT<rzw{~KG@k`o{5pGpTS
z9UjOBtCb$uV4%~?^5KyU26BIB3_Pg8e?S8)$NfY?SZVMeLRhVGAR(+Ij|ic;fef;9
zdAV&&2KEptQ$68fh3U@fj*7U2qyu?jC2^x*LU?}TsGfB-Cj?G2jgTWLA!j6j#<>4B
zC6oe`ZBf~vgo#L2vmTkTrRANXuE^LqwJg5cRZa=sC2J`(#zWX&Z~5u;MsOF&e+GqN
zj#jqI=OmE@#FxeZSFX0ODIABhan))smp;;zKNFtT66VNbS5`O0bx31V-Bie}dcC^U
zzdNd3$k}%Es$&;;VcT9`-7JRCtXDj`y0PnGpiEc6*7;l2tklUGq4(=+m_z|+8_O@p
zzKIUsQ|Ja?EFpSZEpMz(UueXJf4nY{`5ZH)(1loCLJu%EFO|{p)YWWTo`$6C%~RrW
zSNRPOGY#n7<l5J7p!3KAn`!5bg0PGxhISOUP;41+*jAP;wWnMBL3<8esHKgoXSQO)
zbD>I4?B%n``&*rvbh>9Igu-V(I&$jx+{hLQ{XIp45px=?YU8u~RW1(Uf6EwE+Zr0<
zC{GQoSp0}+gO79L>b-7|dn?xK^2ESS&P<O?=~$=N<>+m$OXmYWFTDAo{Q}tVJ`N7-
z7l4#-a=veziKWPgDeWa)dU{>3*wse>YEII-`iQ%<Sb7ug<EL*^E&iqc!f1g?*X;#U
zYU|<f_5!%{X={)1Y~ObDf2(LV$x`PPlY@b=H_79MkL1#W{i)!!Hvzf1nNGN^QqG#l
zZ4FT$uQRcB%Qt{ip*ubSoKC>8Sv^r~O~|ppbjUPM#IH~qx%h#ah&j4>M9+HzCQIDd
zKNX0_qINY;yE~xRJcYihf3#6b>BtpEt~5vWFQ<!t07pjJ*$QQDmoyJB5|>dmF(VE!
zFfcPOOl59obZ8(lH<vM(J1BozZEqaM4gRiQF+jegA3QfC=Y=c+15RA0DICCvQ=qLO
z1bR6oN@bmTIt6ZjeIM>=?oN<opM7gvHeelhcXo!uA^AL<*_K&D3N5n*5`iEoaBVJ7
z$jKiGhl2kQ@VuG}7HoKwQ!vpW%oMEoqfCJk(_S_fNOkD42BK8_sRn;yG-<go#>>Mv
z8It-_DS-<4y^MRc3?ty4s!O=!MA;>Rl-pizjUnkUZVOB@^sIrGGpJ}M%{9n1aOR;G
z30#WbD;X2j>QV`;<h>NMv*L5DDM-?~hkGpul*T=bCS1LjhI>-|T})8i43Kh=+6Jiv
z(d1*GiJ17iP$;M68ybIL<l(7>4C7!7yoRyWd(Ht@!x#k;OrBL4QgaKnic451tD&71
z*STSxEcwh~h?2x-ZY31TK1S0}Sc;E<r`{OtQYdv;52a9aurtOerv3*0R2o)6bEs0?
z8d#p{okFUiRvw<VP+J*Bqfk{1V;F~Oc#7$woej=n=rj?OtcHJq(qLU^-_mjlkVxDv
zis=*7Z6%>o@s%#*>JIB>)gOqDfu~K}R)#;?hW;#QO8y2+rzAr!6ja=5C2`O?^arzR
zbFV9CtcvMFf-@8O8w78L1MZPwT1KvTNj1Qey%%<0cl*Jf!Q2J#v;-@~?-3&abm+zX
zJ~6upH%?Ssb%K8%CvjITR6Q^_?=^RJO05j9AvBVizlL{-n1;6IqFcGZ6^LX|O8l-*
z8f`HI6?gK&7=oG`udqj)xmgQV1AW!9_sutxpZ|L?ojf@|pDiXYuMQV}_5Aew-(m9Y
zY<@hQze-32X@Bz9$q$q7UeWU6r|Hq+6%~yMh8M#xJPdy!89jt^Y7r~HKY0?+{maSs
zv!7>^r{U{@NQWg&s~|dx<;;J2jDNc+VKfZdm0+RBmF`sn`31K1$VU1y1du81t_4fR
zsB0yf#%Aq)B@!bd;C+hqstM_q^9V{PVbl#Rti`t|VUm1ssNqQmF(c&;9qt}B1!VNT
z0?g5UjVXUeb>}V|jXmE|6e|0BH>NRpvg-b5aXLGPLVy412mk-|Z;Qpn<)0^$Ur!gm
zT^+`w+3U&gSFbM~9Z&z5{O#(;7xDMA$>G`TaB^~@)s9YB^^oX<<T%SomT8-|!{c_G
z`6QiWJ29GE&W|R49L!G-4$r2Sc>SyAh>-LMf2)7bb#GULQ;H4gyHc#kTfneL9^#lc
z>n(dDGgai2BsB?DxL;YX`O{Ybx`wWQn*8IR|ME-_)o}p`dVY0wwtr76EG%FJjvTXX
z(~vB-ty*&8=;`m>ElK^CH(G|tceAtE{N=^L(KLv^5GF4U7K`cpJn-`JFK-s#zg!$F
zrh$K!H=fPT7Z~H2!+rw7_$beS3{;XXFOZdix3auIZbSk!zj*w3@?t(adO2OZn!I@W
zZ1VH;&0>FJU0)phI)y^#i|P5|(m{WC3+s0|yP6+OFPALAZ~ZtuK0Ww$_GVb`iqzeV
z_woK5Z#v*CwHo~7a>-SO;rCt*Y{jQBkd=R>ko%oH+}YnP$HBo0I6#GG5zxYO9#Vcq
zyr)d><!C$P*^H>Rl@8omWW8Sr1r<U&&_{+d75~{OyGTlis<z55;vs@4`^?$p`pvJN
zYiW<3ga3|@R{C<o!v}mNtFZ#@Nm`^@??4lG>k7cBIhlZ}ZUC!?ZPr!TI#s!yR5pJa
ztTF)N+S%b2Et^D6_AxTReL;UCH@t5KD7P~}`JfC?X`KO3ODyBPNEdnrC_N{XIzlSA
z4u@JS`dFY4Fg~bug|Fm%%Qvc*8{pRo@q<#OF^|Bxgu!-T4KRn;z$a0?fN%s<0I{gv
z;h#ndNGl*+0`E&T*AbLJ0w6f3ATfVskj`uoc@Jy{x;6S#O0Y8kuLWTV_<%Gl3+f_q
zQ&_v56$yO7)J(lhb-ZZ1E0GBSh$!hwAV`ra>%B_6GhY;a2NU@?0e%DTcYydNw*tJ;
zhXi<Q>j1Aku4|9$+T*(R*srY%+6B87TDl|lw@1HHO3Ywc9sLR*!m2zN^vi$5kDepR
zjDddFRV7v<g1QV;w8CjXvZCa2yQ)O-1o+7fdc&18wKPI0l?n!xwGt&o-(k5&2h>j1
zM3D)gMo}Q*2r3i-+S^J6fgOCa?Or=nF=5o$1*N}ek+fSimoZ~mdmQKL7I|RR94Zo2
zbAa>_9S`Ihl{5UVMNnAZtBHTen76YiV3$%NDZZHj4{f!Ku<9r&fv8;vpnT<gzYa(_
z?E81z5*u3PJcj30M4WGDEtNnG5i8xVgmZh}NI}drZOEvmH_)W<JJ{z&gH@%PJ?MFx
zmJF~%UHa~p3^+s(ZPOBfno|2>YFbOT08jNp0ME)A@LI>P)-g;|7pQ*=+=Y&g+R;%v
zI%-EpZM#7rOVoA!hx`@=Jnv!UBS{(gNLWWK13}QhG2JUuR8cToU=!U%>o9OjslqI%
z%9Cl*-AZndS@Xl0UUF-VAda8g<g<&#>Fd+~P3OOyA_C?o2+5`KYlZEbjA(i;6cxn<
zgHcyM5L)i^-LA5L{!)JvIRQYmm6PZg1OS7TwGu)WK`ri8g1pMZ!XpJ68R!EO?M4Bl
z#^i?vD++-7+#c-u8`d9eaZD=UU@XH<HW0~Aa3a%f>O}<nGptV)4?7f8GRB#L6TeI=
zC9K8eLF(N~WE7PQ7*;}MTfjm|$0)KB2JI?HlQ^K+eP&=o?J0ktirhd%a~h$zl~#f8
z)>S|q7T>-SR6u`0g%!w?8w!&fWJGP7P4*kD`rf&-A>N{8_U*ddwrD98jAW~Zh`<K0
zes@a-O8O?<7A=!TO<&B$eeZ9B<oGe{`;#0$K3<U=>*U0G>296&*fs6(!{cj#y1-qK
zF67;~&9IqnYjS_15}Q3NZiBoAcui@HQj2Sjh6T--zweRw9$p!0TSke&chxzH6pLDi
zlxwY)WP~E0#A1^+Y#EB5w5x&kQ^=I#9sJ6vteEdFsGufN9_3i2lRR)M|K5c;4QWP@
zL|E=1c?}G&6tPrMcj9f)wj0R<tIV`<u7YaM=(C@z9L;~uFBkIzApEuWHGTBl>`v!<
zA4Y??N@au5z}l%kw8<jMd|M;YC|u^+%P9*ytFFA9e19qHr*(7U0vOLLHzz5;@%r)I
z9J~xMz243uLJ`#-+IAKa+av$G;pRYZ|McmtJ1CnCR(u61*6HFFEm2<>(HBDmm~IgP
zvJW8wEGvH^;NNU5k(@fVQ^$77U0AiLU1(j<o$w#EWh7ZA;p!!x=mYc4Hf(ORq`M{I
z(hP9Bah9PaDt-1)JUBV|W%}k~KFn1zYDp<$p5}vtgVWI}^?_ibOpuSC`ZB7vCTnt;
z6$<1BW$v25Ei2XC983Irw~Y_boY&>g{{h;6t`2{W*Hdi-Xh#YkbC(LWaZPK_8jAMv
z2|SCN?@+8g%RfL&56#{J;@EQA&9yL5|D?T76!XmT+3U0Ovx{Mh%$OqQe3LYA`jQXv
z%kq=D8z?XQAt*1lMtSZ~&K=6RLpgUSXYE4Y0Fb@`AT-{$d_|WhTs>%up%m8q@q@P1
z8vQjJs1^J+{8oA>&MY3UvTM9L*lMFS6rh}z#t&waUd@W~sHa30i`F&0K*5tmxA&+2
z1HANzmtnUG69P9imvL7w6_*rmEGU<t8Zepx*|$3zFf#*x%-Rh%ZIz2{pmwj+?Db)4
zR5W{(xm6NcBQD9glz<X*M4!LoRKCQv3Ez1;RqP+R+vM-jh+X0mxEJbRhGF&aJp9It
zaw&i!^1d^SAmNG&;335km^Klt`G}C`s=2AWH5W#f2Ca<%Kh}g%{rjx;9M}N6K=@F>
zhm&da;W*NN+1+TVJ`LTBA0})}erz&ZH^rPH#rWH$`GCuZK1T_-@9dk()txDtk4-*2
zZOEeAQ_33M!1>FsA5-<K^)iWN5@<<2e7x49j-GbK!xRZrs_`(h>|sZu!{(-OtzGQK
z4$RCnuC<HAvNSFa<;ck~(3ECRwd8ZYs67g%IJp~teKeSn3c-8X15pc{3W1=Q(Tks}
z?=C|oFEhxB`?&^Llf0rqf)Wa-G016uGn6C>MN|5B;yAc2xQuZs#^-a422z85)c(E=
zzbYG*s&XNj?~j#nJ#Az9kX|w9Y_EMvLbhc)iIZ4X63ORVgLE?e3vg7co)=l-it`mL
z&r1k@CYC31f^y{)zBfzdK`1|_$J^%($t*8Q9k$6E;B?-=&<d^G|0i$Ii>|&aCnw>$
ztIu-i07Phts>?!p6$;M{={3oRLUMAI+Op4Fr6D<UmCB6HU8N>@z2iw4FSzt;jwe^=
zOI!2{iYwwn_1e@`Jk61I`nE-YA}0vp4V@W(^VArv)pF*x((EhFJo75b5qy7<SJfDt
zYl$^UYl)j!ptgbxHWug`3$#<*0a-Or{P+T>v_eeI0<fG(%Q-%k7tBza7_;cKMK4bw
z38S}sOVDwgOy}}fU{=LY*>RlpXxIDO{Y$l;<8}RPL|%ku-2R(>wnw4eDpVJAJv}>r
zqtdg@<uH1r>!~D%YCn_XhIY0BEsPs$6FO|}0s%T}?1nDTaeSZc>`urP#g=w<Pj$Uw
zG5Df%uc=t_!uc)r+AtjdjABoyty29px&l7)x@DuH9*-*uoiKcec0`BGUiPE1d=9HE
zVQo3SQ+H{FJl9=1stwfuMo6M9)zET(e5Yz?g<NxnqK1y@hNvY;NI$))eB3;0fl~sk
zd*e6!-Ksaf;qSPe@n@P)fIMBE^{{(!me{YC+q)A!7jV=dG};-*#~V0uD&%0s@rcE0
zFT64{E08x?Em#x?ep)+MDXywi<W%>jDkQ!}F7=D4YMjz4T?r83g>?B-a%g*N%`m)R
zAGr<#!~Q%upKD&l-prh&P!eX`X@UC&ZqDbla$6Q&9mn?>zdIo%^+2!wQ>N!|gI7TK
zTREMj@a2D_GKq3L%vGv0A?kXGdF`bqQyf0}2VB<}c<B?ksU6cr-v0jda`!JO65?)`
zkt#40mw@U$6_@R14kwqRA~2Hz<4?CZBQUlBf6TdNruU$gjYCd(DEuY*YH%G>n{%MZ
zFk!0DFKxzT`^djUj=sv#rWpfOLavwR7~jq;#VJ>D0O{D^Y_OS494QFC-_`O1kK4c=
zWXT1W!h<$BP%O^8zN{zFDrRSu_SnQ+O|d$&TL>K6#Q<5aI5JGqL>Qcq%wS6Nv^&`o
ze?(<8y_@X!U|I<{p3@v>@F}J*%H$aM?Y}6IV~8rSZSK~PB<d2LisQEV>k~LcRL;z(
zxOD<AWkQF=l;A~f#0L(%+YKIfCe99HOYT>cI3Cyn_ef~i5tWA-w^H(_@Q9<5>->pi
zj9leLaN{FMOE_^T0m!$YI|S%WFZV`af3!XP4U1cipsaUG&_hZ)q1t!#Xv9Y6q7SrW
zQ7i;qn@LF+{QwysCP~6<@@utIks0Wyoa<|oR}Op<G&&%^F<GDuidapzOBP(7C@|wB
z-}LN9$B3d4`JRpv=T?5~$sLv>jyJfOJ3fke#a|*R>=A!aDzO@siP+~7@0Hi_f9AQw
za>c^W$2HgvfekWBDrDgia$KCmg08T;I`cT>=;(XM8*vr9L=)HVkhV!JyxXzooYiP&
zbhaEY<B;Eot;*}=QjZ$tSa3DBtZ@g>W7*<1IB~{qgLPd1PQi|WJTEiFXI%_o%v3EX
zhH(A?)kOsyq&n!Esw?bIcO<8me~#Gs1BDAmOfHps8ekcJDRa8jrc<(6lutU5U0pUH
zC^UX2eYIooA^NVTEee(yGQCI|>qx~AZ>T!UX~vV;DH=}tsjc>Ps<RTAH0`C2oq;Qs
zCHh1@(%=<YrAD#OF*kh=r2%ZF^U@5KN^UJm8Bo-6K0cSTc<l&0e2C%Zf6o)(0QFgk
zPmuHZWK&OeO-vxgQ9Fiys_Zp<G#quB?ZXM_*eAM$lDB62u9LIcso3l%Fh9s8en-nH
zl$G+>${}~g(T<YUAeXFWhJQ!WTBRj;#l=w%@Z2|CrVXAUp_E40G$^fB+$cI<B!4Bm
ztybM^y4Lo9ipkC_hj=jzf9)TUR6O0IGaYm2>=EpvS(Rj3F`CJNrjDLlFM8d6A`8Iv
z6hX=6^z@G1paQN5y4~!1_ZVZPz&FkwMpkiYRMLc#QP*>|16TU81!@@e<;syO&zOhO
zs;He?o4cQZ!bIwJCUvta^$e-5`h?`s2d+fIqU0&l4YVnJj9!s(e>ai-qLrfOBEB8N
zuF*_!(*@0x*eA*8sm$1NN4erFe4N!bwkT`Hx=K`R9`-PBsXl|Uk+0Spfl(QWv>hZI
zOQLaZr(Nz&F0I{gMKgp&#)ro+lgE{jUfP*qE)O@oBdC=<tu{Xr&bSefOVc9la@9|k
z!nN;OI=`_~VZbO4e{{o61*B|c_cjvEik#D$){NTKz5o}}v2~+;@i}Tv!)4bycimif
zhMu5NTf_PjklZtk?|47j#{g^D4dE6&@zOCSoqNa0vL++1rPs0CORgm)*f=@bZVJ8Y
zWWP^9-=*lV(e+ok<via3x}oKMteOT*EKWbBC&de*_k~(lf6)+iQzRm{a0gk|#!*B|
z>xTQ=wq_sCi+I{y>vqFesxPy|`^}=rO1T<|4~sowDPS_|G!6Pgddsupa6gE1<6?bK
zW4+`qLaV7^P-5LVDlZzuOBV6$aF<OeTIKZ6Hm>AcO5O8(vUD6Z%?yQL<P6GBuzYk(
z4mca#$}3!5e?<1KDdK(V$tUvsH<-?E*W#=kmqX8`+|d(^m68uDWwftPxQwXzHgrO+
z|69n=#xrqPJ*j!C@?3*kDw~3Pry%lvvubwbUl9pIJqT9r8odiQ{BV5N1KG_uLiVQf
z1rA7-6tL&bw7F#1LH>3$bc83jRSB1w+^&`W^H6k#Em_f74_(7eR+{&tTdl`;AYh|F
z=W#er``f><kmF=%%{6~d=-*sh{0~F3W)zo^Dliq70sS@;0X3I_rVS^TUW676m+mSs
zhJVP9-~avo$4+KppU**VNot)IM7u~zGHBvrh@&BnhJNcfmcQ|p-cfRyHL>GdT5}%X
z#v}=z&mJ~88zZcqY}8N&xK)~cG~-O5zS9tbkiHe@aE`_W_YzIN2czkSA=?|`qkZyJ
zva<*_nmWGTny+u~yZ?4YZB)Y8Fj(+Hk$+|W`L&A^8i=V21EL6^hu{AWKX!k%RI@MJ
zcM8O0WNAtGnUT>7WBg!I3rp|&yH^TLtjeF?KZj=I`&-c+$>l0wD-94uc%6l7*@033
zsnqO<az*)~tUBwG<np6#NuN`mrvpcyX)b-AvYz&O`GXaXf~S=qy0VjyBKUBv?SI#G
z>jhdl-YY<qqs@3PbWoSc!11p0Xk10yGeo8J<pA)gCde_EqJl>_R?cN-KZ?3^i+>iH
zI7#mk2-(3&5NR@#XuhOgO)|a-OZ1_v#3=SQMg<TaUBmQ753b^mH3C$_%N%mVvL%z~
zlWnuri&8WDo|6^l>egmLJ{g-vP=CgX5KL=?d2nngrkocTTGyqB@!2VMI-5x<V97N{
z2F5STQ8~HW>@necbU9O?G2}6(oaUDlM{5>pHBKC}&0JZRLZ}~=#tSP=jZa8aF{qb`
zS~8c5qfv%30$K8NiTk{wJ!yGVO<J=D(|cyW!dsSeneD_g94RvkW}lSkIe+nLdE|5#
zkI-14Xt1?*pOi3LmG|1LZ$@#(o;)kriWSBm*Z4-OnkoiH%g%YmFV*GAnPhL{8jyXz
zYfIQ(4lAFDQx4w*qbbN4mGs=W87f^C*z>*wh`?qmBL{XSt^KwRz4P?F4zM(Rs^NMJ
z#RC6+KRWoB_hhi)#9C=sqkr(os|SG$W4yocz5t0eRno99ctYYZKCzHyQ9r2kG!bz|
zH^eFnY)WFOQbuexD_uF*T_uf?lw?$^TB$3Es`8ZY35LC;w+#}<GNn87{cSy)anPa{
z4oa<GQi0ErH6i+?-Gh-RymVJ3$b3K(?3^#^6tvfZ;^3A}VU_br`F|9Q&@JOfbp%>F
zs&&7rBhW?(jkmh&bJeH@C(zES(TK(wc0R04*Q?Mr(^hroaSHMTLS?qp6J>3;eDpJD
zTtF^^y!M<!3Vqw;SY2g8E=UvAK5R!Ab_~PW`eS|6)=Lis4Hbe=>W-M9TnxaA<FDI0
z<nfefE`kt6y*=@|34cO1B}tbJMVPgV22JMojWCab6(T8|)n5cFike-i@9fpjp$vMO
zboTxz)9H27xde60XK{YExtL2D9WLyK?bL}(g%;PX-7;?s8R$1CXoCoPg4|QC(NxuV
z$O>FOI8N4+w-;2o&`*^f&XtG_1iX&VWnoe7?fQ|Cir3AyQ-2b`fLRF1z29h2TOnwu
z@Y^kN%11XKCpI0^pUP_fN}h_Bxu>3=*IZ{%$WryN0Y=8IG7)1kU!@5&-ebJPqQ@(v
z#ssh!-2#7$!Z>I|Sa)-i0M?5tYg?iZOkJo5liWS)U6h`>5qA~Laap21yl`PI;Dc}u
zZcbJ`&YSh_oPY775)+XB4SZF-3x=8=f2+$rmx+LMvUvIJtBSW*j7H`50;@ZZ*Tu~u
z?^?*6?`mwy82L7_ARX|&bO)x?3*|6X)z2G}jO%CFaI8~hoEh(6_D>1J!^EZLnJ^@3
z@LJPny5+udMX&89oh%L%MUCanb;m(#Bb;9NPsV##)mz?uO**lc^r5N*biTixbn9x4
z_03#AAJDlc{4D=ZnEDr2Q7#uRr>XXs_dMowzck?ub*|V-mh9mBF`4p4KdvsxA!vZQ
zkp;&GCXs-`DRV^*Y2So_{{wI|9hX5MG7|$gH942T5&{#KrhYCcmlZKEl>r{NtT8as
z1OsgI{g?4YKNABtHZhkGZ!#2@tOqA4m((^emVfR1`^PUYG7&3<H{oaas_))EUjBJ0
zq@xmK{9womkTCr5_e=Og0|{gOfG9lZ;pd;jkC$KTRr6lecNCNZ7L+uGXdx&xg2)ZV
zD50z^f?@_WGf2#!W^oYT({*HAv%oYnC|Qu?M<#{l<;us;zlLh$`?rZd%R7;&cZ9Fu
zT7QVF%y1BuS=0D7(h_NOBaMM(9hBrJy5kuC*5UVUe0IUdHSw?}%C3QBEuj@wn@kfM
zlGdk04;u<ay3WzG2!fH>@UPD17HB0Lc_5SKP@*@<pKN}bS3)>Vl#g7W5l=HonD9IA
z6xYKtpYQY~|9@<Md-v({E1VJy*>Xk~NPo81$dJ8>Edgp7CY$O>UKuOM>Sk8qG*U_n
zE?2Gj4AH-Xr)yNFF%kr2+0!J}$ZmMtl%<0?`65!x23R(;<7iZfDbGQAwkVTlj5}Qt
zaY>*|Phh_>H0gyWUkuG8GjNkpjpiNK!DP}%&S7vtzMLX8x@I%_{aaT_Kwe0%>VN$+
z#TvWfox5^RnJ7bicJ=D=_B8rY8ey=JADPKXALW6R@EoZBLmtQp=dt0zO|JMPUQ|FC
zX!Mb6^2?2pVj~!mM*((1TbHam(7*yAH3eYyjFq5!9~DVDVbRs7XhB$p`E67bGN4iX
zpafEjk?#p9z1%`drb1XhIl|Xb5r2`fgfs}1ushgGMkwX$6{Dzd3?$LTZ^tn<A!fBQ
z%5{A89EUlgR!^&jjuG=Ea|!Q>ru!njD^JZ9<=}ukC99m52Q@(0DDJu#`MyKu%Uw|g
z+}X+*I+D{ITpAu?X0#@~+xgW<NI`$R!z$Z)wseu$3)C~Mb-Cj1Oz|?yw0}6<O&^i3
z#mL-}Gau~eyjby(OP1Y^VAl%=EC`04kb@X!f(?0ziJG;4wm>pPCv81@lA41|1pS<y
zn#Pin{<;N>rm*rNFE92PEGy-}b$*Nc^17|;tWcn65!90fh3F7UD1b+)I>V_$SO;&>
z3DOl^Z?}@5DSHMANq8{Bh<`b*AlPBNQKW@tYp?Le7-8w@MVOYFQ(JnzvZIK@u)Hjj
zOReID$}5Kbl1uCT)^iM%i~y`R+baB0c&%mxep7bUFyc^f;anK$Ne~lL#X^oVlR&fD
zag!U);+(2y1TCk4-*Cfejq$ElODX^SRPe^{oDbTkaEPII!Ua}E#eWKiAwXtWLP~%<
zAK3sZM8<7ucvUO3-vKC8mIH#SSEBcEH6xEpfSn(Fy<n9^J;c0WQ}FE$K9zEqFwLfx
zlHDf=PQ$gU7p*-e#hg%tTQyZJ*9@%mmYW61hg?Zs-h&Sw_ZUtufpN6j5pI2tv6lTF
z>KT~gUlW?r{XX!E5`T#eK00m~S#Nh)dB{3gfhZdlea1DM6Nsi>iQbQMo@{Hk4w*>d
zT_dWPK7_97^)Lwv6oYSTN3gy>IF@z8dJkLaxLqqw{;jin=^+#xoe$e|hu#2%@>WH-
z&onX>r}ZG+oF0Pbb&)t(3&Ez{<Z(yd*Q7z@>P6KQEV`#^*?(azZIukEdB?G{`O5e>
z@2Tp$88vz%G?|(PeI!@!v~4sA<)PtjNn4^grVFEbc?4(O(&=tBX|jtauK8}>Ro+^m
zJda}ob(&p8v!47&Dvn}}4)-T{BXn)2QD->a%cP2@_0(ZB=Z1|oI*nb$b(X@ZzS3h`
zIC5eaz4YnyV1Ki23xNfwS@*xF9D5B>4v1W#RTNYm`f>)c((lSG^6R{X9+jd~|Kd9J
zYf4LTc~+9z@EMkl7Jg`$39eS^<2UBl->Lj|RC?&N@Om;~JySULP5ORg>4-CM)sbiY
zzm<hjY@)UnLZ>^alouEzUy1A?l1neDQH)hFUFj)iZ+~jpC~Ax=w|U!d_hQ=zl-oUd
zR8A;z_j~f2E?rjRYK@jBu-Ow`I^onNd$*?3>~!`m{}*x9%zQU??H2#K<jLd6euO|}
zgps-?)1(%hnZiQgT}@kUhO^X6UoJ|S_lSO(+mUb6$H2|=&_KQ2OFeFF&`A?G0Rb>I
zCsme^IDeUC37_UjlNTiG4RQN`R~s#mQ?=!!&>GK90djj+E*qJX=UM@U1QJ}e=-h=9
zRgKF?$3-kHdcqw?r?Ih>&+g|1IP}l(X`ZXn`Z4t9aS*t=ksJ^!9MoPoP!Ho^XbB=$
zLr*;V3iXxK^~$}YFHV-otNQ8w(O079{sC0OsedJbPCa_y#J=IQtykN5;Pe3Xystt%
zr(Pc~&gxfys#W)J@rf=b3E%AWUNvbBtcRLMniWA;E9NsgS!BEz*=qZ!@QPO>^VG@T
z)$ZYP&^Tzcb~ge}2fUs12#7Ujrk8FomCd<O$wB&SO0x8ZRv&ro3V^e7rF&qk2~QJz
zX@62XXbqt=Jy`+bI9k8glb)Tu*ONxFlvTR&8$@@S1I`P>#cjpe9jrvp8m4;E0jvx)
z8gR~0Z~GWnDmD|e4PAQ^-i(9VnF{feDt5lLmK2hVBZk#mUvaq$GTn8$AZsWOlDMR4
zUd@?acbUIJHgPoeQ37Ivx|CPV7O#oTGk@jVwh1u}k`vX6mgg0h$5PSa;(aRt&ely$
zZ>LY<tnhxq^J_|qj45yF%^D{;6*6YtcrYeM1-E1%AF`AJvIsmWsLUJ%IbS)M=B8rE
z367vO3r|e!gP)<RHdx^s=KQQSURESl645N3Q0^EzMcB<`obzUmZ8h(AN&sY3gGyRp
zM@j(2`KjJ1@SQ<TJI=Xkno5hRvXC4sQbKRt(tg$q+!`TOTa`^Oc>1KS+jKfufUk52
zb4RVb5O|rCT%S#)Z_fMw1*PtkmqEfa69YFkGMDk?4HE${mofJ#Cztg<FqZ-ImyJL$
zw0~r_cD)42h>ln?UTCrn-x4YMOd)+DW%Ibe#JiF(Q&(^CGkG0(QZYW5Om$fK_Vd0Q
zI47)9^DXdoH-H$GXlKgux~)TW0=3(A!%ygr!XW6G>W1!C+8$5D9Fn|d_KNcN=+rrP
z#Aj*GHGju5ds1sF(0RTs-jY1WJub1KM1P{gOz}|S5~kycm;saKS>kAMStRo%4kl?h
zm4FdOg=*=`w#*7C^;JXP6?%`WBS@hgPM3k))<p2adp&L5GbP-eEZB`-V};>JBHcY(
zsRnCTUEPz@cO#kgE2(&pC$mqPc8lSxbY~)WnzE;zDMHC>DT~IuaLPAeiIt=*e}Au{
zl5&!+?$v8m;GV1QG5B97xwyIsK*>#4FXmL`j^IF5Hxbx^GmHt$5RegmB*~&AKGEL=
zL`*?@VQ{bVKBHEN=U3&TgPfUm(|t?z@ozqno}9}Q&m}ZIg#&s;C&kHDJcCh`H1sjg
zvCe;`eCG&shsx6ig>q_UgbBwB&3~mTFR~C=^d#=Nh}`y+#F1Ens-~nGrO!;)-&7GA
zLWO#5)|;97P+p|0r##M7V|bwnT!S-qndGDuP{r+wORiin21;m2s5^kHl4h~J5(Pn6
zZx<s$fUgOtK=;iPXassUpNKG?)Ispy72`#(&dq?)jlalKpTeR#6Tb!ipnvUYL<cA|
zWQ&Ljl{C}gU5`Xuz9p@MOha)pzZO0*M_<`lU9j?0YFCn>uFkoj@y|tqBRo3lo|cPU
zFm^9k8?V$>9ouC(0Z1Hp!yH_SXV{kU@igpM@mQapy1a0$t2KAV75mcBZO)zbC*uyB
zHJQ$EkX2A`D}COJe23Xc2Y-2WzA;v%#v7lDiK6VE?L5)0`^c;8O|Pob$!6V}!yBzR
zIW;ScP_9_hp&%92C~v$SrLzdmz4i0yuQg|fsSYT`g|NEuvtCaP4Yq`->}{_fPr$yG
zpV49}9sJv#*O#ZK&eLj+jZgJUS6(iYPnvqZ^QLn1apTA5R96}b$A4w@Qujb9k$X4c
z6brtSHrup2YdE9&f=GHAliWKcmd{)UNk*StSq=${`BuktYya@Q@Q#cyx9iK$LtKY~
zPmt3~?Fk?f%qGu@Z9C!Ygh)4^bR7;TBi8&4nyR%y7oPg&^|I-)U_szam2HbV6%<{|
z-8EW$ci*YmpqwZ+D1X&wRlB%KC4F6TJhZCDQcl(N)a8$o2RSZ3q37M~RR9hG5PVX7
z>psYBt4}rDpL&YjF{TbyfE&h`4Cxw{$a!Att|wKuGcMoY=-VA)>`&dg(~VFM?m$ns
zo~{+$XlnjU3c@b9tH~amq-rGFDV!&%WTil>$CK3QJ55b@T7Tlw|7nU4AHHI%z25M+
zv!iB^bn1@pWt2I@|HuJH6XV^t5}}rHvfn~CLq5y#>q_PBXm)4`2U-c+R2r+EzzsPR
zhrPFD3~fS_Ge(c(Z|0qB5ip*521yx^5Q;WBXNcA6L3SX(LVbr_*-78zaLg1tEt@q$
z!-_h0DXi7hs((S&Pzk47GY1GDT)@h2@m=m8I=$aBe*nT+-masgL#3RdjoGdOu0zaR
zITMZ3)Em^!iZSq-=hk07np6}4Hz43RB^4!6Sn2P2Kf{xyHfG|L``*uhp-vVXN?uh)
z((j$)yfPx5DE1veipA@v-_1;vcgL-HHaqRsD?ajEP=At?>b;-b9jf%EJ~iDA{sDbb
zU2#NS=|$Cf;jr@U4xNv;K9~=h=E(10B}3Gb+0tGcHJ`sc`2_8Ra(bn|%l$8eFo>Y-
zN*52+&gK(!noAkE{m7WEU+MjneLw}FZJ0;ojM9@n(=Dt&`5u1*mRB}hJML{eCJ*|$
z->TpQ8h`a4=7E`F-=n}S|IC?u?Rja^nw0o?^Te087+@GH$>!F$(I{p!>CnR%Zc*v;
zUY;F@quB{)UzBY1j`aaK$icmWML9!H@D4Zo@=X@gyPrNNx1IE!*AvIZU4a*Ef~QZf
zrc~f(4gG7ayaGMy0+{Q3!72qcx-yOR1#i};X@AnXahjd2&Xnt^0X530haD;5aGT2*
zBVpCNGU3!TUCff+jF3v>w4Tuv?st)JokcyPHZS>KQbxUPcFFm^+qGp6!J3FIotw;D
z3do*s80Cf3YC43b!)h++W9wCcQWWm(pEO#h;9mMpBY7DreO`4j!vs|4cRig-hmkT2
zjYi9*|I0K;n#sv|6yFW}`*E5b+MnKm?rKeP(5wdeL(|2BCmyNvV?m~C-ex|=p$dMo
zS^QteD8UN@@W2#h{wCc1FF3KdmmzB!69P9jmr(~769O?amyw_?D3?u0Fq41fIqv`d
z{9~sJ*ymH!Ye=0Jg}H}FEmp-s+5)MZ@Bpa=daR@c7%g7PvHdN!`zEjYd<wL>nzhQw
znU^=rmUz%DFGP1hFmD;wqN0`x!Rsk8Nd6`9yC*C7VUK0ukj;_FRM90qQ}bT^_w94{
z->%;$8)aQ=90g@S#XmoH>4kp~7WJ`VR*|In^>6ci_h&t8F<9MZZ_BJ|2k;|S*(qy7
z<5-kK`(+!V|6A*A`qVP=QhtWgVz|?V69J0;&#xa%)${S*)50cgd;*j(BfX8>wI3m1
z47;wkR^ZySMPIg&`pW+s9AuvL#egmbwIL{p=8U6rn<0|q$}N&EtW$qa`XW&)=U#-a
z917vo^=7QTJjn+U4D#B)oi*yd#c=4+Tr#igeR{G(uhDiJG#%*BO`WGYl@0URaEieV
zCzIm|#+ud@CcPi;%j)1*Sx*&&RX6Vbm%>x1L~lU^C@O~C#|!mA4r`eJyvDlV(7LF(
z-OcQqwH<Y&T2_vG_w9eU%Wh#?qfVf&>$OG9dz$Js%>+epq8u1hzAITJxnN6NcFZl>
zZ8bEFw`uoHXtLUj04-ZlhZ@kMq+BE_5>rdL2;7?QF6H9AGG^>BGwR|r-Oc!|=(D?C
z)r&?~nAcR@LPf>aBR22D7R<M0A%pbGZVTQN;25rT7P5BMQO|!zAE*;c8e<jm!(@*|
zxughda!cDOecpQ|0}Fm4@JbH`ol^#GSZ0@v3<{V2%;NPH*G3VF#cFkWTef&AgvHe(
zdT+~wTHR8c*B<q(@Z3iX3OeDQh)`VHR|dT$=}Sr%Y1NVweV4ZToH$GS{)t{{>em%E
zu=3i~>wUqfi_(8-33VygE9HVDrP5wP=L?s1)pNnp7c{y>noE~7;?#eWU7XCJ$PL(G
z8rZ2VmGX*dcf86Is!;G`C+I6l%UtjrDs&t!@tU*HJhd2pT%|W>q!b@xfDbG%M8lwL
z?pR<Lk9Z=!U7I%*6UHK>$=#)x*yj2av9EgJM->eQ4&Z-LVg8=;pFs=SzG$-BG?GJv
zLC<OKP-5;>N(*;BKlExb(0~+Foi6ZDzui!0i#BxQM>>2#f27X`i=%rzu;n;4#NtY-
z=5C()aF%^Xm%Ezdb*{_D!L0pZ?xBLgo>VmUj)K7g@8f<(j%AOBt3`p?<Fpi4MPY6B
z1y9B4(zt(W@<HHp+*{M7JUZVcoGY{2b$}u)g!cF707Zvl>w+cP9B8jE{>Z`WqN!I|
zDgMae=mfW{alH=WJPa-?+~xZzGfl>~=~mi}g#I*Z>|UgcdEdpuzEupGYS^~g>~V^>
zeSQCe>Wj<aV{)xDqor|h*Aqr>_+liZwnQZRnq7ap9-$_9c$U<o7A~qtSXmh#tSaTU
zrED8ypJT>8*h#rI8q}$If1w){x-yW5mK>^dX0vd=gf(4XlHGc%cE1}n8<kG6I$pm+
zEL{AssKemW<zoCGmLt^@>(kw!Gb-AY%WCLDxIF8on{EI_l;Ax2K>QMwAT=-I--go}
zjE*sn#OVTNo>;Qaf>+{=tXO<KTM3Kj6SC8q7;7YmU?z7ZGIRUu#s~Y-BK!Uiwsn>M
zmys$k6ah4s5WF=d3p6iGWo~D5Xdp5)m(hg<Cx6XaZExGi5&rI9K_FkQAH3Xo*|*{l
zAa|*+SNMP+PJy<8agc2)(N&QQSq_qa-)AV%k|Om&TTZMX21ev?cV>3>nP-+vfpg`n
z6u4B3zZAq*8nOvgg~BF;DzoWGta1<&Qk9FEkgI~LnwYDSl3n47JmsSlR4O7RT+>()
ztAE|aPvUlT4Hao=@8XHjPK~iCw$~!0RN5y}MOkew%@zA(dJ0>`2h0(C#i5-eP{lEs
zaUoP(yv-3Jq{ZC?DYtiXC7y#x1yZy*qOXK}sniA#3ateMsmM{_$d$0PPngttnINgR
zm|#!|KH3ujQnY}@SV?L_hvC(?v5UD<YJY<lF_C1un=I6%@FOW=sO_}?DcchXan0Ob
z!zejg3=&2u?Q95y7Bw%StfjQJV8y)H6Bs`D)INc0L=kgf!2%277d<3V^C&}7Zbu1K
zFsC8K3O2TPlVw1AH&@ZOzU5LyZdV3VMirYEOBQ3WHHXu2;dz0Pp3PEI_6T+qq<>99
za@aKtN@~$4Fo-ZYtin*t$t0+7Y%Rr5LU1A3(n^pcTr&yw#bkF?T3c`pVZ#$pUJMWk
zlPBMfhvAVmCgECM%!U;9rV^1V8L4Pesj_6mDv}`6sU1%t1=AyJgVyj!zQS_A+=%7L
zDcN1PnzOBHB-<Re<`DxF;`Z5M`+r1C&sRkYayPPK%ZOQubSw$$i^Y<vWZ#ZQkZgS;
z_E&)j$@Yx*rJXq0HnW_FSRnG;8irj&9!qOad59^NqaS~)-+aDq>KBvAbY9gzO=nll
z?5)S9_viKB>tE}i@dMlPynfwW&fgNV3$_iiIE6Xbn$B`Sa&d+9{HQ*=y?>b7fTyF$
zf9n@_!(Yxv(@A|+|MT^)_V>g4`TY9kuXX)yG=G14;V!2i>OXEjTpwRGAM1bI{`SiK
zF|IGh(~EjI%&}Y!Ip+(aA<31>P^C#ty0}VL%0oZYGAvZz%r5K6_!<vBI|X1Ip8}Bn
z^8EP`{y)@7A6>>e+&M|&gnz@qIH>|BHSKb8!@}Jj)8a?#F~eiReg>quf(-yTb{QGX
zUL5lDynaz(__O+#>6>Z&vigBEW&nC(>;jH47B3!VJUrBtk~0YELJd%NzC3xTmrp)B
zP3ibF^D5FW-=dZYyn0>#>)-zYMgfwjn%s`Z=Q|a!*<6lRdJy2!M}ILDmwcZgvCSPY
zriWti4)t=SSk*sI$J5!__29CpTK%eiHJHzv*`(sd&%b?||8h1T%$tf97fz;=Ii@%P
zijW8F*d!;}>b5RsJOKb<CZi3npFgi(&8C-U&HQcs>g7rOruj5KUrUl#gLe&Vn9Q5W
z{KhKxV!YSQ&GdG5*?-(D%D-Lut+^Tvewu!2Gbs{)wD38epW#VhJ1-t=lkR3gDcb3e
z-nOIy^7%tEGka%+xl7C(T?|i-a~}t@(cog-+$?|~@bM{1)@FL@dIkW<Ksz>b^xG^y
z7!(mI&6l}QpBTGSj>{ivKhi_8By49%=#wSJzBaNX%37ARrhhDW<WeY0A^1+%FO$HZ
zT)`@IsRYvJz$p(amzYpFnRUO)3EJf}hdWnQt**qUcJ=0}^^xfKl)`=<GLbWG*m=m5
z9h!uv_HZ~FH@^?3vk!y$Vm$@EYIp>SK%fK#JiT~Vmj0_Z6!0!Y-C_tV5p4?^$KV;P
z9vxTQiy%aCOn<Ar6Hy)M$+ynsbaFGF4Mvk?uvN1~J8S}R_!>_L+qF980G=RyG!r?Z
zKAJ!@XT3@fO^IOlsfVU=a2`!*4*<D9M!!u$AuWAor&6Gq`)HyXTkrS{YT8ESS}Sn(
zEmldkidCYA6sx>etl;{hBE{GVUHOV#`HEfnN`Akp#R)CIs%U?SBHid<h<Pan=HB}~
zTfLGU^NQvH+FW8ioc_ZfpC)|_7zYWl^CSis5ONG#5`;EY95Qg7w{ScZU!nlQHK#1a
z&LH-}aTWwPt;(hp^dwK7PB+aG!4VeFpYRwxwY9JWxNX6T5ny&=oh@NMaoKLE(MNMu
ztVIz@5y`h-pbLNa39kZ;{01w3@;JXATrY`cOTyx2rA+*lmStP(Ps`;l%_Jb*eKZ4E
zU(J0qJv){@V+!n)?Q*$GGqKCsN0S+amHXJqh^bF>$mOzxt#Vo90p-#TvG2;|!pOxb
z_EPBTX71`{?&@al>SpQMzCxYAJ1XgYPue^P6C-Dk3CVxQijj|0>?A#aoNb@90eWM}
z!TXqf<d{@>2Q1vz2uy1IEpWmmbhQFdW-tek3^X;ejZK+3r6?ppqn1S3m%YF#gyf8>
ztXg4pK9<93g(;vE&|{pg0w1zZ=z`U$+M6iUV2+grbzHtfxRs(4zs7FZ4AMh8dGHqR
z_?RRoCe(l2vLZPtIg;}2NlrjdrmJohgu+G52g7o#_OgG3Kuc0rKyFAtpJj2*L&1*r
zP;o%(wldpON`SvuR004kb9-vJ$9n@Nmt_@c;_O78Jko>V@b~7^^=xrGn*k(9u}!X1
z^ed+VYYJbXIlDBC=7c_)L9NqCeKf^h66GG6>WzP7aSu(g<0S55C(;WT7W-&2U{$q;
zCin~D*=NjxJXbzo=46cM3+7}mq94GVX!Y*IUX8cZ$KH%Dh13bT6UzRqh|qm{upkU?
zrwf%;%!n|-7+1EN(ww9u$>@?a<&Z&V9)S~qm_?BII}=cxvzfHpgb<|&bTkkJ(E6S4
zKkt8Q0?I^2I(8-qrnWB({*MWy0$P~RHYOFdU$h^XBAB3oYp3?vI-q!9tPCwOI!Dd&
z7(_wTS$iOC1Q^JHrrA&lNaxcIUM8d=V0i@F2rMoIy$d*i!@e&0djGnqwFJxv_P^D`
z?bHOdkd0pR`@Y7zFD&ua`KNum@&H)kHCTV54uq%!AzF5A_5sa9PCMC}ee{e*V9*?L
zL|H<tt%yG0aO!G$iDbPUj|Stx<ei(&-qp>d{z&zQ!Q^%@u5ad_$IbD(+tF21k0w{m
zC-;8-Vf^gH$LZ+m*d{u@9?b@qpO0zJ*Ko*Z`_5~{`_Xo#aB#0pLIy!1euNFJR2+Y2
z`*d$nP&8sakvc-jrR`2jq-fE^;!{@pij4PrOI4CnS^+452NgOr07dBpfR>Us2&sLn
z_NRn&U13G^?ASc^&_tjeCH<ZZY6HaXi(8Vma7)5Na7)7qx46SC?r@7c+!Egj)Ct@P
z(h2$qOl#R12eifobJ^)37q(9*cfL-?MK_0kpjZqyMYB(PAN|<j$Li$Am><)D9YuKG
zcE~>MiN7AfUBd$fDM(L`2Vdf^l;cy~^IJ(y9zeVHJw$4cUJlLG^-n+k2cWuLX_wJV
zFB1bcH8+=mJpmJ!tbYV4e>smVw-L_wS9B=BSUy%2PvBvI-Ps)jj%;Bw!+{}Kf&&Fs
z<e%>%yT~Tl#~izI(9<*QCaYM-R|iK=zfOAk^V93c)3*zm$*W+&<Mi>%1pGkrCukmY
zcJctq5lkPyO@IHcH9Edt`7&A)@#Xn@;JZCN69j$(EMNQ-;PmVbf22A54&gkP+&qW)
zEj;rb`2Lc{d4AMRrxhY!JO*_7$H!luv@okxbm?oljlbSMp8oSxSw~XV#mVWQFqrA*
z-=5L~16Da7c}L3ghJOFo^vl!V##JjV%~R$NtEb<8oraO0M0ioN@U(N*cEUratPPWM
zQBis{3(vf1ZCP06f5I1U-msVux=nt&1lU~BIz<%66<LbfvS<y9)Z~@t)s#tmmXFMS
zH^S`Kl>=l&4=bbY_^NM5t;n14ZCdfc18gRmcM0#p({+(vm8bE_kiSEonpMur15J<!
zWvboCPdCWCr;5B;8)s_PE|iTGPQs4ljB>czLz0r+lHTR?f9!N}ANzW@PLaF<Pb<##
zO0RdVYNCqd##O;=?ia05uoB)^8w6TqnaUSBSid61c_sTgV^&sA6t!gom%Uh;mXExb
zZlra%!aEp-sp`qEN#8hoo_rf5y9ZUgV+uS>)&-aBJ8Bdbww^2&zXp9lPV0j*%wplq
z&Gwl@n=Rh0f2>>-wHrpP#b&b!@slJi$i29cn<Gn-aJo`tNm91S<!mOn%X%S1Y<A96
zN@5WSpi;a7_$2l1XvxB1hw}sVmrIIj;1fXl6IhdEd7c+cgh5R6-oR#@QJNVT0-LqU
z$7_LYxaT0Y;}K$~R+B>PLr#*6U%_4hEH^(*a?ReDe}7l=en|R!z0Vm$21@b5lu5J#
z)yxMq(43?`$lU3>(Uk2*CEfON0Hy_Ro6tvw^+YtzK#^7}pDaQ7J=$u@m9ncDk+7+D
zSap;rc{`bi3SIRtQydl+YI!zfP28O#c&{A5NCn<@B>=%Gz;VK{`nEU5*$L?yKCEAi
zCL5x=e}6OVJi+*J<)h7bN(eg6NGWYrVZ&42nV(bMV$@mH@1tRkhMNubntfN}Eg@3Z
zq^29WA#yU8icE$0>XH72kh5De6ntV7MB`e!w@{Ff6%FothpdQ&12<ne;$l0?_NF4W
zMw$y{?#5bWS60-Kclc5{seFa}3+UApT>2!pe@Tb2a+#>VwtOeu<{9=hzptzDeY?0Y
zah~Ncg^@+J6f9#2%hquD0<+$J`0I}{1vjTePQA44)bYbV%M{KIRrI`UJH=?^P{NK&
zsJz8_%D)e(Fa;Gu=&Ahom`bWC!wwZj+AS*J6>`qrQTfv`W0;a)*H8KR$D=TA2$Tta
ze>7l;)88)aH%Dd<DTN#}$CA(|SSWCH&5y{cz(p1@d{(@9)pje9T~9VP@1dKV^PU@`
z!Y)T4D6ZAVMc+=(25H;GZy~n51D3qv+;<>m+;IkGIo8s+?b+#D5E;0<%e|M3ZHy2_
zVV=XDeOP1N?5wh#c(<!2j#R<K)k;ZEf1nr0J4~b_156C#8n;KgV>NMH_i?q?RPAES
zns^b3XI-VytdI8-*`k+omIFV51!@oNiE+@{70vy4e7FN*(2BHLt=KjP(85_<Qn#E2
zhr&1B#y15{yBIkKN;hOtyZFYC=p(Io_f5!jO-gS2uqpd$yh!%qgt|y`4!!khe-vIz
zE`-;Mx#AI=tD(k4#>dugHOB}u&Z<?`*VoBJ9Cr+SVVz9CrJbQ$3T4{v{aq`NF><wN
z{G$c(F4Y~KM6R8X2C9ElVT`GH!};$2r!eMxLwOw@)0NZ&J|*?3owQl~q@d<aZAQ_%
z*cUY^t(N;+LlvcrHlx?gVg_Evf9P$BhEuZGf?8(T%c8c~RvLS$uy%JDCXM8t;6@de
zys^|bTuaX#+;-%s*oi@fKm!r9A`V;{=6AG1_sjxOGpLRtT=iQYFWnG>a&8n^*J%ki
z^*WK^)dd(9(wE*Mij{Lt90->NZfWG?Nd4Mo1GNE^E&GAWJSt$|9Vl)ze_B3e6Vl8W
zsMO}JSsBGH4oF9TSg982kCe{9IO#EVvA!vCbA<I>fof`YTX5&4X1C|<{9EEuDNnm%
zbzU#5B13(91IYxgS9@Ve2Uud+`D(@k#GsH)X8M_*$Gqd56p-t-Fj6%o&xCB#v}hu)
z8ygsuJUC~uG}c2hU!R_9f4;iJPI+q1^3*LOqO0?=Z?Vni`uwfXu2jbC8F0l7-Xzib
zX4_)%CAT?s7Prtht+cmwY#iryU`J_ZE5q4J0QPOSl|`wJ@?lK3qq*(-Kxov*rs5bB
zVQngV4q&4S(ck3&pg8#*Xa3<1z)2?q??AOUzydFMJzt^VMwgKJf3+tk@Q5}9OhJIJ
zK;_hIHF}VSV<Z**O3t<OCt1r4#(qmzhOAeS;eMo+<y?1JY43OcZgD;V<L1Os$|iBf
z3jgi-z@;nKzDml5rJVi&a;HsdbCXmpu}Arp#qW8P0Tlz@=}~B5oOx$vyWbb1lx_@*
zXQL;x6+;OJ;@QCnf0pX@OB{nzrszh#MvTSXP0RCg==3&&<MMvS*LDhhk&gu))GIw#
z9}__>Cy?0+brTNGYoB2B6Tx?>RMUeG&ADbhX)9?KjZF_38PG3T5a?~I<+O%VxoUeC
zk%?X<Y8{p=saVUa`(9=iys_o<gBry5OK6tWtVrT<RLbq2e;b?Oz)2MVEX}r4uJZea
z7}GWzr#tidJB<=g`A}nA|4RQ_xQgjNDa*f>=wu30UAMO!6UC~A7`G0g=22|LDezt2
z&!gA7u80hO76%6N8Nx;EOKdX#JMK!8)j6d-4U5gYOuVfOVCFV%sAp^6hdoP*T-J~K
zxS}04f_%0$fAD*5`K^a_nWKWB^?z8ksu7c4G($~_)PO|O(SK&B{66<$9+XHK)k~Tq
zeelr$11dD_Unwe2ZqC?;lM2(O660w_+g06jcE5W%V@lRs4@w_^0y+P|@wBJ(xIUEz
zmGm=6u*q&v08bCcyDQypfM>IoY5&4<7H-{c;^B;<Jw<VnckE`ySDEdy<1j$bGGiq?
z^HrJUQQTOBUQ{Yk^w9AgG3NMDP>%r^MNSIcxY%&`R_^Z~PyYk1!5Npq!Z8yAH#Rqy
z;W7pjmr`Lc!vV^-7Gf~g2n5#X_G*{WOfM4xHaV9ef(R6s;gl*UmqBYVoqzKbW;*`x
z{?|L1g)P@AA(C)bPsVD&b7l#jWk0Gutx0HSho@PO^t4dqoSLN2{BcM){l?>$cfWpN
z#2l<5MktW<Vg=SjGeQLU%9?A|Tn8i=C*pgWuR0@1j4ok%j@}Yl^2Aqv3}70$3d*J#
z3rOUI^yX5Kx~KmyAPp<*fq%sJNaf#7>3(@G02iK}fr}tj?EC_-O`sGCGJZ0M!a7$3
zqafIG5$dSaA^aa_+*VNvaN|V5jKI>vYa)M<UZ3)08)7*Umbc+BThu#XV0T4P5}As)
zl;!wnWKe}+<xlxWzQ~km89o7dmYrS_EyF$LPlQ&!?8>g36QE#Roqrue8bxl`5WvhO
zsxoJGkV&$|ErOXDlw^$ZvTmq@BZ|LZ!zZ0JDI?52@>ve@R(Pc<Qv)k~aCY{PSB>;2
z?ysyGe+ih$e0{AX%1`PnOT269_`-Hu?`&$kZ+jx4_@_bsLgwUh_U*@x4BP?%c`Di`
z7T(M^C=Wt2*N633<$n}DUIKTnVM6OWqF>l@AGk}Ei~NUX*QwM3tZec`mh3<Btc5~P
z+0~t*GbrSsj#QauVFoBOJS9Eyc3RNZdf|QReOfmLhdjNfLxv;<U{e`bzUikqL{(^@
zLD|(ObrL%Y=Q4xSjI9%edbh|_SGVT?i_$ItOXH}OWf!ow<A0+@UG$!%lv%r%s?mjW
zUa04cl+Or3ig49esCHay6~O1}wd&4anA~toWp*|(BJ2U(ddBLtqJ=whHFT5NGnJ92
zvg=((U)N$-M^+3kCUT0|(CvsB##^};l%8SIBTVNUw$vBxe`I&7(JYKmjF=pznFDN~
zE*xf8bzxYfkbhmsIUME0CxR}s7dtgM%9-GXN`PtH#F4)p3OmXkEMlg#v?M`^izRC#
zyB-tmdA=t>jhfhF8O0WAV}jCxpp*0d+d~>*?dk|*I+%XMKNy>b-DOY<MCwQVzQ2x^
zg5h5*1P_nO+WWukuSb?c>mcQSfBMh<iWY=2?poC>34hRdm1-{RCcC0q9>v+M@D?5Q
zqqN{^kw<aL>e3FlR}cFcq7SL^z``cM$T1jRvlrlsA#8iTd>OvIz2E)1@Q7S(3I}TX
z@t3>sgXWadSkD={7n=9bZ~vOU-F-h!TPiCdc})64GP#@X&evQ&d&#OvV~F}4W(4t8
zrr>pG_J0SjIh^K9B<x!SgNKR!YdIqs=BOWMTsrHFl@4;2KT-Vd&ex{wa}aC&VP+-Q
zi$pI4a!ENnjcY;UITQ&o<R+QULwXKRE>DF)a|ND~ZO+pJP7oG~t9Ez3p2vJxRpDGL
z)lxO5KaKZYBBzB>3mpPVTOrrT7#p8!vBfyyG=Kkz+H`kXyB5r>VU24?f}RI_ZeSJS
zy-)szEFWj|*R$Ga;j|rR+G4d0jR`Y89Yb8M7^0n8OI#s~cfvVbXf-nR@)<*E54g~7
z?3sOpEgFS8qkrZl%b!~ZvFjI-YrH+T6h!Q9$gN?<PjIWB=^VF)r*GEuN+r`_ZhCDy
zrhlgU6q{qgtk<hw6ao&Ks)m5j+Zrf6hL~Z5JoS}Ko)>+{O6ZRd6jwb=^@TzupBX?x
zr8vQZ@#$_44l$uZ_!09Vr(Z$3>N{5V@N7aOIF}j*jTX#X3@t0JYsmRT@Iu2o+1<_|
zt>=2|QPA;ihK18|P7r*^52Ji*zwC9mOMmW(9n)joUOFR`OF=idt>_3hk77`gV0}J+
zu(030&Z{Xlw}JT>BfH($jl+cPrzO~^3ZxcZ>Oq31Q!S%InM$F7(>A36K2{?rFhaYr
zMBr-PdN@-d4A)o<Y($<+=FpTi*yrvp7sP%3Sax|Xmt*V4mGZf^?>~9Aq|Drc4u4(_
z$439D^^(zXj*pb%m1~?L7l@K~#sEe-9a(pb+Hy&0#}TS@92=0Ly7~La*>L#SwP1AR
z;(P+etgy^}&8fPju9xONmMUbAILsJg<%YH;xfn+58CzBPt}%3qgsZmRH1q<faJU0C
z9F(?jTv*mamXG9C!_0et<#;xXz<=B_)s0N0+otRcUZl=jbFi>FE*&7)#H7fFm;Rt}
z8@M)XpWAYEKa~tw)}pn$b3x*)IGwO$yQ7Ht%y2&<o@*S2WPrSCcjDZ@4Pco==kL3o
zVU6LcYx{Oc)Q<7R{l_Nvdm2ut%3Om>Y>#>p89u3jDrbXkyAPVW=TJ@@GJmDNI~lju
zieu>nGR~oOAWd$n{cX45r$MoaTxX*DU&|0&Yu#yJ8u7*nr_Iu*T<iX)KmWYH4$}<%
zu+H!M>!9NK)v(U~DjY1d^jEE-L4d@5m7jMzo|*#;0rji=>0#G0WQFG0uCa{q!jSG)
z`RBeamUE%&0OQPCp}h()hkwE&tgo^G7*_$>s13l@1>>qlt3mr8cl*O6*`cEuz0HM3
zC19K|bI;D_O<sBjc8*RLzC;5Ykda{HC7R8II=@uYkB$lSkXCH2&k+$lEoOK<K5B)x
zKZpmXn5Sx7?_&wbMF=Ybfy#|5c5WAkU~ty?uQ*^0_*`C{nM2I@@_#_**xeb&?sB-W
z7l(_vR@r!?4U`l5!!&CLN0x2mWSeZ`Y|3drPV^d1q6e~V%Hr9a;|!OlWBb)1C!XlW
z$wPvr($VEpIg}1o;=2?r|8_(w7RhI_ip*@=-B)S65fGw7ug`9}cx|qDEWFMM;O2WK
zQV9sYx|5}OA1o`p_HHg$QrcdowC1`+a=hpl+D~8iCHKrLw`sQuUoO(BccE@?_cchS
zd??}PSH+h5KJG!BVNbrJ?%vWiG~KHV%RS$|n$qs04c4F&R+)u#`YgRQyXRq4?QYz@
zQRSIXbEoctKGk@Ad%yc1qag%Dm(ffw6PLjx2o#rObTGsLx3?*EFjocxr0{l^A=*6?
z0yj68k)1jdmmr8ODVNlHFq{D#mvMYBEPr4F=a-8WS2-dw4HcU=8tt**k+KjwNin}d
zlDn`nV<}D-kEAdMkMmI=&;xp=Dhfi&5<q26dtCkPbNAo-$J0MgWv)^&)NatiGDUFw
z`R&QS06|QR8>D1T!0`RA;p@|HJ=L^u{XJ-1K$dF~euS*l4ABl+0@HNgZ!PCVTYuF4
zG;6e$fm2gCjsf9$SqOiv4iWMog&y}p7o?(Ict~$;r3Jp173!Sj&m+SKznn0hU263$
zKt$jz(hI#zdFv^hlZ2+FRAu(k1PYa=t@6Q$7Rr${c$D7KLqC|v3;Age{Ep9_Q^gRr
zVd!;XvdV_EP8DtQ)DYtAgG4ktd4EX!s@!RwUMGuG=iDQ36xk;h^s_EXQY=Zz2y|J6
zYWVI|bJF_d^P&%u9GO!S*`-R@K$a{%hjqPF32x}WL^2dJp`L9G%zpFGmPs&k^Jw)v
ztVu>2GQL5V*UU$epQ2ziS?RQntqy06OOEp8LT>1qhd|3T$0E1a!fa<exqscuO<!K9
z0VlVr+GyJ%P<1=VJJqhNeJ=TfOS}S&Q$-C<c`S_)`miK3q40KJxR4x3J!Ky*6umZ9
z--s8C)0p^!JgU^F<x7eoAs<+bAxII6uqnEmqH~y8vI;3W?OBLQsG0J%Hl`S`l66Jk
zlCT0VXZ{J|k6K9lLfMLtNq=;dB_GLEdOOaPfxfqz$UzOXo}6YW-jhtXA-#nsIL#>3
zuZPpCGEo<55S~w9I%@fH8}c#Jjoz)|M%jeyXW${iIp{9OXU#<lSL%K9qDp0$AR`{L
z6VQRPfglJ5()0x19w12ONWgal^(=Z49GSB2(c7-<6c!i&?ud()On*DKBQ9DqhiWA*
zMli`IcEMYN!(?OC;Bw0qgOkQFBl_h|8O0iZC7oo@3-(wJT|Eka?cgQ6>Q9*HsHOFG
z*^^jg<oPPToI4kA&W3ksVwJ7-?yJsV=cm+;H6vJCZQ*;YI72^C`-eU{@o@3V(Hzw=
zqz0IVXK1=$>AOmo3V&GBdJ=WjX~l9lOo;js@oZfMIH0I%$a1K}l~T5jf1HLUaqpU9
zOv?}d)fDs9nd{5Nr)iXY8ejg6qma=sg5c+0KE<R8V19_9A!0n_r%zk8cQHgM%5H@m
zKr1G7$=MLA&gMuQ14-N$;g@*#kKES{lq&H9H}Vpn*OJhcM1NcW{_dY-b`?3bJ*^pE
zN8NAS84i?8ao7Y0_Hz+@(_ALw(_EevxrJLo;DdnTj|P;YzXIF@sD2Po{Rp6XLSziV
z4ymYcAtBl?weKlHEuKK$N;+e4Hq9r4Hx2M42=0lMoRTJ8DpuihdR~*&vg=_>J>2@^
zvJaHJc?w`#eSh^NrP{8=${AW*cc+(pn7j38-szJbW^Rz7ldHJn{@gVcKT6%{J|5<d
zifTUCH1{33gQQG)-MZ`jxhug8>QMr}K9W7mU2<le*~ERvxpT*CvvIgBfR4ED7&oLf
ztQz7DVWT71MU8bEn0@q!QDeLz5u=B8dAbtQc1qiFt$(CMV4@dyK0>UkFjI*-CU{S1
zS*UcRp-Nnfw$cfo5MQ@%vE`*2Qxsx;t4NbBIo0=GTRbm+?umtU4A;k;QI%D8O}$&b
zRZMGeWyZk9z1*<#m_NTH>$z1%wewq9MvaDaWz-(!RJKvANuyQW9@-<}kt;2+)y|;J
z^ZSFd8h?w(h~8Xn?l<SX_J$hU#1G$@W2wzu{S*<u)Amxw+I&IqOtO3(IM)d*ixEk#
zkYov4PtNzRR$p~{E=68|4uElsyFxcI*do;ZHrpF)C3&~a#`kqEB#_*=9nxyuT03rs
z{Pby+w`dk<Ethu45AYT#&yY$nlCr4I`_|H;-G9Q`Ru`-uiAy8p=6rlJPraF^o*g-B
zwo)>o&3*Z8JCkj>>FSX)7(m_zu@dIzB;uz2|I!tC^uqeTZu3En9<22a?U0o|s9Bk{
zT-qW1#_v7HfrJ^y_|p;7NAzcj?Lm~-g3Un?iMsn#AK!=du_>ym{M^2!1<wCXMu(My
zoPX0~F3@h?^Y++A$Rm1c&~g;7<$ZUkhmy=Z%bUAlWxl6!SRv8MG?tGIMLGTPK?vST
zO);DO?s4_V9vHzaDOk6H@_hAJ1{h0TVKQn|*Ucql1!DlA&M-QNHqj}qC{Wf)*`}jW
z^_xsOe3Vu>yN8+=E+0ASj#|~eS+Dv7B!5|h%VEyf(r3k*chvGxv)HPd@zCV(?Hazd
zyu-1v@U;5a*uvqgh-NwYNG7H%nZ818bts~_OxefWn=6Nx${GSPQFnS!H+GFV8f-eu
zDuiRQZu<AJ22M4m)S6}0p;e+nmf)@)THV5J!DCu-z1nu%n!~Bm9Z}^YhN4y$tqyK?
zHG00ke?0vQ4TuJPm(ffw69P9gmjV4Y69P3hmyvTPD1S?D+&B!r&#&+wh$>Q~9>8FL
zOxh{1w+(Vz>}j#>sfX?VKP)B7vMi0~kvJ)O@Ys=Miv0LytB3D{9=`65FS}1CLxWSE
z!Qt@oZ2)<p<rlOpS{@vLvId5iAH!dJt<lPLl#8@BMlL7MUhaH&FyLheuw3#M;PBuG
z;y1F#;D7k7vEw&JcHxuULGF)nUA_<6>PR8TC2Qahf4}^()0tVVf{l;y*8cV7W%u8%
zu#QHt4TEK;K*R9Qk6nC0z{<KXAgTZ~{QP(Lw)?ZCI-S+r6Hp!mC~YnC0w}bK*$>tl
zWt=NOiEu2!u?R~75Iv4}=27^xCW2}Klfx*eB7Y%Setv!*nw96Tvy9Fu3(%elpX0TF
zt$g9F0_@nF*n*io64oi|N!TN**5sMu^J&~Wi*bl*)=ycAbAYaz*`zBB%2}G&5_2Nj
zXO;v(POKf0fiXl=bMY-rMi<#-%!u3k4QG~%CZ^p&31L6$U=;j-xUayl@p(Ea(Zlk6
z)PL!5dVk;f?&|Kt1MD*o`Ad3C$Yu-dFchCshM~nUTh$u!#yZ80y040)iBn!kxoXpg
z5P24tk0*Az1yqc)$z@%-{fv4-gCiQ{<fmj<6XH0TXyL_;tLc1O_CelTbS+<uCml2Y
z8SfQdIk3V|%+O{*{cLih{U<V*4IbHhjDIdFmp5!i*X>rHzpknfI4B)VbNs|F;YI9g
zZ(q|kYO9Qd^W_2hw3!4bdBU)54q!X`@D@_U;KlDLh4so%CxxGVbwg=P^t{m8TsF*4
zC&?)n0wz1Jv<Oo>)iK9wfrQN*r^P+a1m^4clY}ou+vZOP%1J?RIe#i0b5{D{oPWv|
zE6+Kn27M`~GRr-b$U&b+T_nMBLPBV;Udhrz$(YbmOi<w%O_Byr_h(`csZeStqduR7
zgT!iyy3KCW@iA5gTBL+8nWyWJUd7W;QsM1*YS}O3!3+=&N~bPXo^QZ>c`BNin=X|W
zE652I>|{vw(>e4`hX<=6NB$Bb#DA<aV#ryVdgTI!NT5ivSS!t(-EXl&WKz&UvbCBe
z;W`Ujy3LZ}UB&IODh(xI%|A)D#Y*i~lPp@?OJi}^V^k~dZw|g|@kszNeu#yW^){J-
z95Qlr?#-p#J%vlj9pD$KFO?!wRc$-d&U#$!y@&*L=L_nmswDKY0~t4?AAemTX{TCC
z``lDyT`VQb$w<ohf;wgkieV~teE@IeCUp!8hE_5@NMYn#h8|C)PBcP_W2?YVIqUh7
zL+{Rb#(Nc{Ir^B6(Zsx8rSMJEE7z9<ZtU!0myZ7Kl!hqrU3hL7gwtwsiRhJ;CZZMT
zLU$Q?OU+l9tCy(1xX0=@w11Muf|~molpYbu?U$Wa3Ws*=6Y|@sr=xl`>7y;D^ej}O
zXXEasFTCVrXsiSk-!MumYAwW893+-w*h)PDS@l*^x0HjD)H$_;P*4~>66}7?rt-JE
zwN-CeeR-bY2@lMfzECN>wC_sPiUSH4@gpNa8mSRE=9(g&pKWWhw|{i6)K(P;SLPf!
zdtJ3V3d%~Ts`Ux=0O1DH01`ja)Y5vLA6*Hoq@3?gDv@*h{Adg+BEMmNG;){lSO?HV
zv8qOpt5aY!>S}IvOD;r73YOqnE_~I=D;6bDl-h%<oQL7Srs_*6N48{gc~bJ)6e!VK
zNr3vG6o?#@)RwQ;;D74}sJRhqIJh)G4PJR~I-WO~puTZ@ipa``@Y{}0JH3}To%U}V
zqINuAPb*SqDGJ<Td2Nj9q~bx-ezh`QL>=oS{4GYI5Msm2zibxToL`EwX&`zgR;2ab
zVs08O8r`0_<y=wv{SdH|X=53@l~dkR451u+Cq_3H-Bv7PQGWvOd&(96EyR21)Kkrq
zihTZEWF(dZ+dJKj!(ecv!29<c1`A)-!Y`Z!gCOJ#ZaCCxC~i0lZYW=w1vk4tXci38
zMR}kPo&`6jzHt`ZP`hOoeE%@SI0ephoI|JAyQoUr6{O5=sE+?}LN0TD%p#G{h7QH8
zEklYFFD|^Hrzb5bzPikR+bB6Wf!uYIWba^_DFgkY*|8;MknU)W+q=Z<{{bY+K9>>W
zE))YcG&YwJyfq{XFfUAHZfA68ATl?XVRQs1f8APLZ(O$#eb=ub(3kW<yyVP~Gb{oF
zPEw~S9Kf($psgVk`dB4S6|E|*0_WfN40o?qidOPUyRnQIus!!KKZe7ZGiP|Im2*fz
zD^tiUE^we~E=Z7+U9eE*KebSlAV>-|`y0ujmg)k9+Ps%=P)R#lDt@=#Sb~v+l1mC!
ze~XV(LQ4t(U32mtwdRnsG;gkuS|8Iw7WQB9L$0;=NFiIwJCM+$u@6@)P^#Xyv_Lsg
zVC#~&iZwy9ig%cZh+N51AgM1~C6E+<rx>LxT?@eiDn18#*j!VzYvR_Z7(QtqnFDj`
zk^~m^c32;mK61sBn)=!lle4b~cI7tzf3*e4i7^Krh5D~rgK#&!Y7j{kCKbFB>F*d)
z^S*|@mAps8J7qVrW=PF7zy?X<-i|{_)oo>N{^pum!CrP7z$B#<x0-1vJ`~2wp)~3e
zLBjS_?eF^5nuVIM^{Oy&CAWsA7T_p~Ia&$ui{2OA+^ryKsPiqy0Che}!<6LOe`~@}
z>>8*Wu|Qo&-k9K3(zmwucfE1h!LIt`(jT!@b%i+N(YR}c;SI|#DdVx_-WO^#Y_}@>
z)pGW31V1ESrDm6~-H>uHc*qVU3RbfDyBcgh8@icl-!e!>zH&4%gMPJ3E%iAVlK33h
zl$MjP4bjsm`>*JRje)JO{?Me`e-CQ3qOJ=<q|KK~2#+THoySMk{t{u4xQ~eKL)6S0
zATUzVJ|iL{`8ps9L`<;r<(I?TU(Y7P^V8GmY<P2dF!QIE?@#|5hOegQN0ak+3DKYS
zhQAHp4PU>@^M}`y!`ZuBEGBGVrX1l_2sVu#wV89Qx$O<l1BQPyd^>$Re;vLFUjRLl
z7P8#nU`@3&%YS}`KO1R5DPh)nr3HdD`qJ;zA`yHS;AGMoyp;ewThKxbv(i$FDww_A
zsRg##XNMl}p9ZJovIQloVzRPSqH&~qOd(bzHG|W`8L%TIFD>!SJD_mI;ID^&{_EfH
zmKqI!<t}=9d2+IMM=OmUf6V}q*nk&nwA36OH9y|6utkTsHCj@QNDFJUWYfqn)EX^S
zBEo--mQ)Z0Yc+I7E!USGhOeh5)AKiH`-hXj-XRRH_Gh!n`Dvi}<6l3_zI`*>pG`uZ
zzj!e{onej_J)(i)KFbRv0pwDfKN!-WdqICdBFh<=fQM(#hF9m)f5SJE*}LJ@H!p^_
zlMl1KYaaM&|Az?_I-N~UXBVE}v=8xmaxuL;Kb&05v!uWJesc7F|EuYT?wJ+vsEzi}
z{~SF5+qLUH_vnjxChyblyz9B#m(f%CLTC%2q*3VY_il#9>S(~aLmuIYC^~Qg{31IS
zC=?rz5osh^p}n0Of6+BOzZ(8<_I`SbF!{&p@BHr<Kh9=n7k?RsAKuS?ygZ19)1QZ*
zE`L7TJ(~P7{QdI#SMjHl;oxL?FdQFSZHLEXc98RN<|E-_;fYT4;7E^zj?=NU<Cceu
z^TXkn{qy(x2PcyYbpO*!Cp5b+5x5l*xbs^aSbwpMC^7eif5l=MPR3lGda?V*$N!mp
zI6I%OHI-do@T!$Q6!99C30ja5bP1lrrAG1WR{>XL75IZHfjCs;B2g<yM0IoA6=>m<
zb=0D;#tgo7rxsKi)(<v8Af%M=q+39L%<8N$bBj*Y*=i)vEo3@CEU(bQ^A0rwp#dWU
zx|`mB|FXm^e=KGN@|P9Zi@XOOzfQ}NkTL5TEtx!<*g7pU15)YJLTdHmVu9~3fOQ!=
zx_CfwkkaBoogn45F$`<D1;fgZfnl{J3>O-)mPbJrLLGmq<4^VKr!oq46h=6&Bb?U}
z&NZz-@dvs%XLthY0cQ<@%|I~(R|nky1#BzgOh{4ze;sg=!v?~a(iWl}7}I2_1Pb2B
z?H(-{#43RSv{ryu__u6<_M#$;@J-+cJRU_}wa9QW*c%WekA<8EWL2PyIr$&6s#I9P
zo;uMidLqfN=Tns}B$*7tz(O5R-PK7;1jKOzzOLGDs(@li8C7&G&>bb_=KGyRBAkve
zsb%STe?5^OG<2PAD}g8lzB7tiAUta2v<=rEl|V-EI)e)y;dF#TP^#Wm0=ySFu_6Gg
zp|Kz-ZXwTsimM;g_<vwLxkd6~hb1zt@caTdg%RMi0yT<?*%p*Ai&&5!D@9Z6M?oE9
zexy`mNh?eNJ4gl)0SFioXmuN=5ahdt;(yelf3y_+eB&-H;F(pMNs^3st(Ldq6b(pV
zUUjyJ5P^%L+HBl;NTN>_9)KZ6g;ITPKe3aTy<E)^v#g+j*^j6SnsPCC1*MzRM`fML
z_eQI+_5}4yv`))ReeZq;u5wq!cZ0<|ly|gri+7kkhIcex@(#@-I~ZApjV!}PmSH2y
zf3T5dm~Kubt{i5O#U0b^o-EF(kO|6Ea8`B3W3)xDEyV#eQ03?+$6!`5-%_Eb3=RRe
zwFZ()*=QR;2FL(;@UvWGGe1G*=!b?i3G({w(6O3OPr4pTD4+;1vbhqRfTpel)(yr6
zG-X5)Vgyj%TK)am3<JB&kL$*=8N`yae?B#vIXyXBmd)TZ+O|SxfRX-PMRN3WW)iI3
zYi6V{6i^QX%f}cS`5|$;vUO8r(x<lc|1S5BmZoOjeJQrIwL`eIk5t3g8Z7HklnT|>
zX_;V*te-X;+$dIW)Ti2OT3u24-mG?)1?~gWiGTO{O*-)}F%M5C{(b2xowPAkf3z`G
zw6P#<V?o*$1*xXA*|B1A?nTQ2r~*ippW2ZRr)R&;eK_yFWZmpEQ5Z$$wNMpFsDwVR
zlT(tii+|dd3O(^`;9*-DDm;rBWP=GsmfTK5AVA=HNKjn;po8+tg#3Uklz_-(=~A8+
zsXVz$T^w~omm<3_;pN*fbVcpvf3XAXN4WqHLFs@%_By8I+cg|>lO{@>)TH>g)4GwN
zL9^&uwy!Pf(<l4U%%11zM(**i=^rmgZnnDbL)^-inxcQn|9Hy+-q6ZA4GHe6>kKKB
zLBDjJmVVxm>l!V|TGVw`?&k?m^~t`&4KD=0fXOZ%#2@rAIW^xp8ut4Je;?1Ox=8z1
zIhE2?PW2lQ^L&}}DA4BR7kPOh{&5aj)FU4xROa92iqIng;1Qq8spKmi7+&aiR_;<F
zGm0Yha|;xLX6k1b8^MqtrzO9`w7?<w3MOu!3lL?n;Q3kgH7Lq*KMlU#;3FusswK8U
z>N+j`P6MTNX3tq;F8T+Ee{a6cnK?fuzIplS$9HfZ)<mPgqadSDMxl;^j)INC$g4SR
z<8)?zwFt$1zgB~iGjh$LqD(@_uK3bZp(KKnfbgx7i0~EwB7^SMr4nE*ev#lVB?`~4
zHBl<+m%*snc0h{Dl@83Frpw)iz!oL8M3s`&Y*tFz2%})wm~44_e**kC7W>?8PcVa<
zW-h$x?TH+(U!Gtfh~&@e@&s4&x{kNO!pHl2{w>2l<FFrn`Z0h-q1#8GU}z25FIfUa
zz{hgg#+@~kn5dwZN|;}tx?711SZ5OzVgqY4OIy*RVd=Kif>9y#ZyI-LQ4Lf81s?#X
zvfm|<EnFgGzb}H^D>7;!e#B7k(E>@(v{d!`ti*UC{o~a|H$z=ZLw$7pAk@=7-&-L4
zAGg);mr<%Q69P9kmvK)g6qi7iFu;F4{`vmfK_+2eu3p}f)L12$c9xWdlbX3I=B8Le
zb&b?st`2Te(OMW)_xr|F+(<3#7vKiv6-{c(8l;Z$=^AznHb`oZpcVhntPo8jQC45v
zyznOMKfJzm|N8oV_~%f#APH^!V899DE&lcE5I!hirq&OL!h;(A{d@Rv_`QFpniSl>
zqjyAYNyEZV_Jl@g?FNHN7=EwnoCnxj=<7HhHmAe4uu+-bZ{u9e11hX#>nn8(j-$*t
zJucOwet=0SdHiBzoS_mkn-=Rt=#s`xiMUDHR4jk_^Y7208~OgNC={I&h^F5nyoG1t
zm{PFbrI3LJ>XGVX<a=@6nIeDD8)4aRIxeWwuCcgR<#vA$Yi8pJWuaz!LfLVMf@?W$
zA##|E+}IBV#{k`Ph*s2NC@I&4^@^;D9Xf=~l#;wG#YLgD5`rl>G}}~pYQaqFBSueh
zqc=tObU8HH5z{XX$&*eQ=%VvN@U65;rCBCZ2(zLEeOaUOuPLyU*$aOGl#pPy7t01Q
zWh<A+<|e|C6a>^Q+!vNhF?kdG5Ks`gj%j4T_=gZwAK51}M6$jKU^l+ym1h$rS5oIu
z#c4gmL0;Qna{8(+>8q(D$xxTN?YOn;IJ4cpvW3g3S#b`d*h*6*!c9soGaL7dY8}5$
zT&~KXT7@rLnH3<&W2b-c=w&#%3BbDVw(8JHl>xx(hK@x;yB(zj@0zo4o*{rwsj2L1
z{;yFObh`q8$t&Hj{PYOmv_xXhHHSb&mq^bRB-j`fX}ue5nX#2m9p^Bd2mCnFG@|%y
zw4N1nxjhMQXL=sFByHg)SD&K&vX+RGB+p!WZc)!=mxXuHQUiYm6>sl_UczFXH0iS~
zb0`+{HyIexlY!&(G%fRK6w~XSJ}K2Et!gp??9AiWB8vE=^?GMSp@x3t&a7<p)OHlI
zr;d1ZdQwYAC}GDu1{jTr;`0K7oG`}pa`h04tD8>WuL)+~Mr^F}M~H2f(34`!@t~a9
zxx^Udj9R+};bebv8eHZv!RVJ1KgpzOOx3(5<uj8kEApH~F)gJ8!Gl09do7nXRu$K~
zCY8upnl^!wnuxWcn%BHC!%b8Lr6&Wvt><#0rjsHb2PF?RFkh~m@7_vdrD8*tN;lNs
z`kWLiQwe!(4uhQ~Bypii#`Yy7kQK_^t0BiqA(soacF2Dt-&0+=Tt4KwQaKdBSxmO-
zTjM?{Jwe!AwA6~+a8l}LTrO3A)z+Je6>to+*b~G{LyRi14LZ83Aj6>5xcz+vt%{zN
zF{P-Nn;d&8XmAi)Djnsn+AOm@C$2oJo4wXjSk>*03ZESsEcF}#7Ue?)LZO|>w;_9!
zv2XD_+w*?`L`^aE1lvz3M&QGQ?|zyE$NVP&O{or+Y?SMCK1}PMxZlZ0>o0kKXt{(b
z@Fw+=wZr%l0DMfdQZvMTHtI~7d>nKp^g?qs&<su0@SDuj{El2WA3htDdoCZ=Pe+iN
zR3ry2r4&Qus^2sLodN)J(sPP=M9fTd+Y(4}nzny8IsOaV^iVy|9tMO_&pZP<icdTP
zI?8v>fNu8M84#b#e%!~9z|Xiau=V=tv426XiA{INpcQ8dGZ9wUYbTVNDRI-$bG`uy
z+r=eLveCdsd2<u^5qZa>^7GsKDmVec9mo7gD_n*2f@V?bYyH%AZ0<y;3N9eAr#8tf
zmh*pch6Hjv%;&o^@n(X_jOXyumm7s|sSl7394-jDY(eKNVQE*lukO(MNcl#*PscW0
z<soBF$nSPt<*jg9=}=g9bOjx!&$`+}MZ&X3{XbWGsHFh6*CvnM%fucj*3MJxMwjB+
zr2>9Z*4J|0y;iAWM41L8^rQ@*7v^pM+pd4#+ss#@soh|8ZhM29`ZWu8NVU5=Z&K{O
zE&7bO2XBO|)$J4F9&GTjnwq7yScuNp)M&A?xt~kznk^Ur8iaM`CI&phb;X}Kf3Zj%
z*9H&JA2sEEJ#>j?s;(Y|2ODplJ8&>YSp9Tt0F=VX2kjP;l4&O|-1M@PAiZ#ko|%7r
z8pRZCPkG@Ezal!tO1N9Gn1Nmwcj29It{(S!(Ip-IZ8_~$F5Mbu$XQQ;ny*J_Fi}0w
z^-$~?Y~HDyH&l{R)#?1&?cOf{`=sZ}zY}03^|c^Sx}G~U8fq_Qva{*^87>=KdACg}
z0^+d3ZgQ*;bv;zHo?aN@hSoY3Um0QTyvtqoC&9tAs^9clac21F@HDt^Q>UNc2szfK
z1XolJ)(QjAbAP@gAS0*WCKECY0wZK6lS!$bO5KWdYf0ILjZMq94qsAb3$JsSNf6AL
zr#E)f|BC6~e0ux`PBs!$m(ffw69P9jm*K!A6PKsBDJXxvT1ju?HWa@5SLhFzL-IXt
zKwva>7NEBbaw~c$f@XS{6rJgzf4`4AMM{p7*a@I0qR6)|AF1%^+r+2)&Hi!o<qmK%
z%qn5z^!Pjpdco<p;54Y!WQ1TE#Ps+w{dUE<)O79XLe2r`GW55lyV`9*SehXeUG!Fn
z-PWM+YiNHC$ha1CTmv*q9_TLUe(lF;z2lmD3XZO8f9-yM{B^^VuthQnOldc1gN$fp
z$w&f3x+6)Wh=cNXwF@#S;iWVFErzE+mq}%<oH#?eEGU?M5~p$W>pfY?ibX4FWXNHV
zg(r;<-VV`PNE$1hWK&TROwYjzchREWn#!g3cV2&woydawr82118fMLfM2UgwWeYq#
z5iys?+bhfCZHT7W%mtRxixF1xz<$jp<XR@(z?WEr)tKqu1uD0Xqhp0*nq2VoviUKQ
zmiJRol^h1ns_DPU4-4-dt+k;$pb*YGZ}a7!Z_mFx{UI1F_}8!C6Ou!Pv(rm@1m;L>
zH+_Fk&zb=l7bd|RMIlaTndFw+vak$1O&emw5TtCTefOilj3o;;i-jT`P(o$U3}qPt
z#WI@seQyVgRE0$lx`^SQJe9Q8KhpZV%0K$6ea=&9)nC)*dD=V_dpe*Wx@*Jrk~xW~
zFjNZ~ljb3*8P~o}1nFslCPyv~&eY5kNn3v%OeC3e#<;1lDw^hzM%z3@q%ws(ROIT$
zSG4aGjrNE^shDyTAhT$EM2ck6Xx94o(@R?RjTYTE_ef?dz%fCTwpt-m(s^2;{E|zT
z3QL-vXXnCly#tl2fr?3thjfqSGIP$=p-r%kD2ebeH6jL8w^i4NQ_Vw8QVLHnPhWq_
z8?%655_B=kN>|Dnq}jPEsZ>*3DUhLBq0KccbOYTQi^|bmd3#9fZ0=&ATG?D7^JrOp
zNOkg1>vR~q$vguMKV#pH;85F#!k*U-@}8|{0aYX?WN*X1XPv0)DUHY%FYX48Ld;e#
z1!@h7g+Q%Q7X$V0X+A5plWo&aj%9yb?mj%!j1pSj5#&S#$$cuR&(pI==E2g^R)aTj
zP?oBDP3aOP$>TLzT#9O(AsQ)jB<T{rJk2_kwK`G78t+YH&SDMK>|M#7+Ow4#Fu@_C
zDtlGAH2htXw~@QcNLLM7B3-sr16pu-zPyQ)LY%@*D6AngSXf)KFcuR-f`)&X5?UVx
zFoOCsVowA!b~z-^H1f{^;ZQwRbxP|4?u&TNt9)CRl4*~{j#Q3;HN8-=a1@j38mFkU
zs5_yrvTMw8Sqh>G*o&-E69PxoD&f!oTD(m?s%q93;Sx%A(4oLn9(Bn<)JfZBqU$7c
zHJNgWVMGQzJObr&n;?$kGM0bN#W%+we~vLl8zVvz*%MjrESX*?kB%SPYAM1?sLNK%
zZwfp`UzCF-OjCJE0WQ0NRe3XYZV8R4QrBkb6ifb_Ij+<>s&CHq9Ix+6xdM~YH}u^U
zu*FqQdwcoYefLBi;>VEdXn+7V`0DOo1n5ivQWBCBIQ{%{dfq(rRFi*0`g`aO!HNW1
zm7059Q?nslZyZI;%<04aX}8r}yxoD+y**aSzd~@%M!r_uB+02j9d#~_pw=vIW-;gE
z*#j^hKtd!s8~Iw|a?n{yPF+}aOmaiSl*lv9WZhjC0Rkt*BhMhfY^2pnQ|ooEW7t*5
zN_feouk~Ie@Ix=>Ip%-kQs=gK+)-&}7}agv&BqS@*n>81v&McJ{oPhv&3Nc?T#-lA
ztbsy>1EonHnhmxoe9cVDD%aT{CDlTbbFl96mj5+hfiX;&f!Zkt7Zup(tNtk*z#Q5>
zQ{e2~MSwOM#iSczu8kHC3e<_^;T5YX(99^=bGu*%MCs_II%9taM2<`g(|5q&m)1wN
z?ttY=R0Q81Dt?g>u!<Hy>`!W(U^SCcoYFYoCukIEkaa=O6>Nxx@T%<uy;Q&TxsLji
z+AS!<Ota5_ld}f{zOQ9~RMe+RP)eKu-x;R_wf8O5j)gbwlFImMyu>CDV`hNFN3|YK
z9dJ0g(2IM#;jDjr*z3i@dA4Q&R2HS}?HaT(>5?aFebV^@r;z0X$Cb?^c-SgxzF&9S
z3h~P3(K2JPr-G*0r1-BRH}P6&k*lfBfk|Tr>ey`}mu2&SPEqdmt96y<gBiBk(m9n0
z7_8Eqz}3@_qu&-J6$|;Aap5*~AAVdY4>!5$sdanmK2?9xuCpBFtsM8w5?7_dFgv_S
zU0*hayZ{EVgyE^9_CbBV<Wl4kutP{*5O+vNXj7aMcSx<6G4>L;10rS*A>>ZN8@`<b
zIr#JN29)FT<HZLI?d!1<@KQKK3d&CTNjO6a%dEQq@C>%@?7JfM1u{htA{5%3F;j&0
zp9$Y&r{;eE{ludxfs!e}4`;v<=M}8y6vLkJr6Sv_Y0PJZwh+iD@;>OY_d#bX7|0el
zXO8hCe)bOr^7ZggzWzW*^}0DBYEFfHY`;U-a0El}0bHF@zI>@f#<nTVxqby^2w;vH
z<V@@a`3%SL<1xN~eL%$L!{C~NL*gY&8BGW$pIalK_OT-8qKsb(%<KfmE`1PEmxUb~
zNkDu<E$#1A4E_gJ+l`lDw+a&iH#e8D_5%|GF*rGwfln+bmky;cm;tA^r=>7H1%HBS
z?HyfJIS}a0j=nmk=wonv#9^s~r<#(>xwprbl*hO{>aykC8>4!U&z6%}E}y*Lfr>|R
zz2XBd=>%D8vfG8XjD~724}5k*$Q_<*$=(XbxT}7-?!D@N>Al(uC*7XjJEEAh=*)v2
zoU<X5b$1W80u*QRumi_b$!Ol?zJEE7a_Qv5`Hw3V%VCawJQ1eX|9%RD-`%AoQv3Rg
z8;UX0jg2_(C&t8XXeTNwRs4K=%c#%@;M)K*ANkP>V~~q^8J1A(tzBO}j7C~Wy1xsG
z+VbwB-oo~;9b=rEXn23vf}?<dd)9HBBC`MRY_H)M#iav<Qj4+K`x)pygnv_=qU1!+
zOaGyANNtJYvzKsi4}2|%5b(Ewuc7JgSD)iBt~Kr^sK4}Y1X|7tSI%<0@D$A$#r5E;
zWpAv(e-*GdTQ0lu$=M@-Q9;1q^(VLKX?(7<r){MZYR|wlqBqji{h?_{bD@f2rRHeK
z1BMA!sQxdx;1(*l`9tab{}aS!jnK+$_!Aqye|bOr2LXULu$R$HFB6vmUo#Y!o2)Rv
zfAKTeZ~y%GZKJcXkB6v_S?W9~mR&3r$(FTR9_hf>KHsyo#P(U+@1WeLQTOky&!bHi
zJRSlZ)oiVDH2CO+SBg(`uL0mzYqsLXi9yRjLx@U;)}ZHuI2XJ)g+*E78!P%vFO(yd
z{H&rASIh9${CfS^{IjXsMx(5YokOiGf7tG?9~*f=z)F4W5LE=~e*WEk+x*^2&9Q99
zl!(d6+OhEyC!<r=2G-`3kMIfM)rweE{qysC*F1TDt9wTdxrx|X18f1jOwzTkU@rl0
zsaYWhh}a<OdV7kElz+ELBnL;{r8$SPucg4%<FazFbM$PKab!=)k+DT$L2pqxf8Oa0
z)?apuH8)NF8_RE(M=0w>I#FGFGJ<vH={QEQA}8`}r`{>TFov+P-AQ7SG_}#6r@7cz
zc3gJxoMVGF1e84DD;R6~{F*&|e_MqaM6UqMY#%1W=iw|nr%<|rAHlFSGmlk(;7yvq
z78~n=<IzO`oIqp0js1QW-yGRfjg5^8Ie%}nrj&b_y=imChl|R9HAW2r$1qWo>v<G(
z30Eu$dluR=<gw{=F4~{1b>+B5tvs4W>t{hauT{i>{2EB7z>o-ZU98yM%tQ441_&f;
z6oo;*=-Bl@fG*5pTb?HUl)<sslwf)LP{A;RUcW+iMX~-G^xO|3q9TX09qG)x_J3FB
z<tbkpJ{72`)m*`!rE2yI`sBF;j@1v(*$?}C(0B4CPCZn7UaZq2{}#B#A~hBruC6O^
z94ioldzuz1N^wfSABWPAr0Rv370CsvSfYQLm%daEOL^nfpcji15h#nX_-#}^N->HV
z4q+e%!>;R5e?0JkI+p85Ok6513V&y}yP;Sl;5dwO@<QN1gNp2o3qk;8@C0xuEj%mz
zv6+5Np1fx;Nw~ibCfab7wAGRM=oV%n8%1bRdIg@WQ--TxBQd!Nb1Eju7qdrkhm?>;
zGU`R<D9~>_wX@C4jsk_846oUy(=I*bPs6rC^cVP(tG;LP%v3-WC9RVu?|-L#-e?t$
zQ{Xl6fvj1W^Eh2+o+IUOwg>T&UGg3Sl+)?Z^9=KP`B#BqhUQJH52*3a1riw+2>ndr
z%MB9wNV4-=iFWy`g`$ea&9Xp`N-meA;ZM&3$4|z2UvkQGSu+E#lvaTr;*g+IPYY-t
z7n9+;0ALTie7)v(tNGEl8h_GSo+qqf=EDp2VYn0nc1BpMub|y$P2ct)eY<oD0-CLG
zyCD1y$DBvf_|?aJs#T78t*5Z?QM|7Sv6!p^kjA%C&pGa6xU|`IgCEP2BYv#Ms_|pb
zbj~R^%<~kHZ6#dChHpEbT(yZzwcxc)?KxDIyiB>tZ<qlEX(gn)RDY_fy}O=^CSK%w
zPTI$`Ftg0ksC@Qa!&$`-saqyy>HXc03rg1n-5ps*BN6FxR^dnm@-s;*til`lN^fUh
zQT4!k<p5eqH&i{~4?+gOR-ipfkJ{N+Y$OI1f|*Hj*B9H|CTu0H;Ft8+X%-SWRd*zG
z=vj*n_e|YJqP;$Kr+<=8rS5e0EvY+|1S@nh)AnYzzTyL*<ZPoi9;LCZpYNO(O!Zv+
zVgfK#dZ+q76?(e-UxBry+&8LLTK|@qJU4SI=eR(rLl!V|8mt}^^ftAFn$X&t*Z~uF
z<%j<S&e=h6t9~u@pOHJ?H7vHgAMQd44ED&|uA}j~kp_RZhJW~9aFoPF*67+IM_F!>
z+`sdV+A=`E9XszFLj}$*64=rVD=oY0TJPXM`OyhD)p|cy>+#n6RO@H8-sPA%Cf{Gv
zdIuwN<8`sRL+d><FopO(>by4_I@WkUFDuV&yiXPX>bsxWcMqP68GU-&-7_E$AbYCo
z-kf8`iGKlLh5;^@!Tkjj1UEP`F_&?42oskv_bDfr{$D{1e=NfxXZQdD16l97KyMB9
zR`gUPIpxs&|3imQiIil^wszZtESaKa$ZsC}IF!SWK@Ptip5G3i&Kd_R9D!BC+xG$Z
zh2&q5+-PX90K#Y(-hK{$ewI=hzMlC)DYfRy^0VVR9iKEf?g7LXzXdoxS%mNn_t7eS
zt4;bwb1!w`e>?E~OK9isy)-gN@O-fj(8s^tem}@OSSh>-kKxw;dwo0nbErdCT9EMr
zdMAvh;qRY^@B)E^v3>vtLaX7|x8eKYk6vokCHa%|Lvr}_W9VAGX7IdZ@T4(B4S323
zA~zW0g$}P0c=#eEF<5x!izkl^lY(ZFpUw^zC6W<ZfBRk{qbP}?MTQik<RndC7-#v3
z!*5!6_*F8)5e(f)R<apa?d_$|7W>|MHV$rJ6H1;Xyyh`ohx96*Xa#C}1D=vi&f-B2
z5EMnNP0QB}n73LLPnLYrMl@<h$|G?r$e}Q5Hiv)`EunWlJ{cLphq#A+z1glxTLD)0
z`T#q8e{PV7AW+$v#Fqyoa!4}xy+m3>P@xxIo0%8L;Aa^h(!=-@bvG?jzb`(ejX>*0
z-F(T>4&hGuKk*B)#V*?CUM93A3kH9n*ZgkPKT`UO-*P|xe2(D_^UzW_P=iq<2z<>9
z(Q8sFM)|mOK+dT%;j|$9@h~T^ghBnmVcuyCe_<X?tFqyXnJJs|R2^c`S*2#>wwJmk
z+^6Z1GIS%3rR4&SC9HZJ<A^scH;SA<q+UW;tt%cnKbtJs=6eo{&|a2+2f~whc`^>`
z2wr>Fo~O#<m(e%*jruH|!u%AMsu-GKyw8c5#JSkSNRmzS94Liow6{rZ0QcZ<gq&Ln
ze-wL$3wdFU*^y;*5*6OgDhvvTb|z_sRanhz!t}lGSoOeKVE}qb53G9NjNk;=3#5zm
zsPBEpM3}rfr)QGP{fjL#dhO>kS;=34_v6eYA*yyHbi^dZq1)4SJBf6^>$Z}vxo+G0
z5!Y=c!3v$sw7u&s@7MqopY8Pei!`>!e|cxW;Jupva01XM-D~}CgkE3%@5r@vxo=gy
zERTy&KtI>YIc`vDm>Kw95S&*7MOhMPRBu{(j2%c6-oiRND6q=cvi@`C&QBeSE$_!C
zR}7<4=Iy`3>30(icC!ZmUvQMjMb_xrB1c&+k#I<A$IYDnoC)#frOIYodNW4gf4taK
zDmIq0gc++FCH%@X&1`p_6Ul1%3K7|HM+cbz8D}r^{bd69X#`H2$<A-ozB-ShmEDs+
zsHr69uVB_tG>hhWXe7~ubuh~nJjGi!%`wsr=F2%$+Q0@d>=6h9&Zjh%<T^PRZ!ETH
z=LOS*uGXfM@PJopZQVWA)mAcfe-3-?DOLjZwd(TlRtuzM;)d6kan-=M$I1v-C9#N5
zDx@J=Q?a15lsM;sl$H|F=e+cdZ^F{y7rsnW;GLUOAzSn1bjhQ$)(gR?w2vXvPAp&?
zD;O_r)nl4N)^NRoq?S%dYdeL-aCOZ+r2#}NZR4z;&g>JHOoW<uvvnr^e;B>Kg*Qgk
zrEsdk6or>nnm%g!Jf_#O4V+X1vop+18g&+@h}EK_g6uZ1ynKUZ?M&9RRDz()=~+^j
zu^)E>U{-<7cA8=iD}{Sh$>#9@0Lt?wn}cJUYv`r(tP5JRkiak5`6Gt4^)y8>)#2(t
zInuOQUT_zJ@2amip`1Owf9a)WrIy};%2~k}e}DyBP051c_~v;R3wF}q!Gf5G)c?<d
z7KF9tBP`fxlFBEGMJt*Vk18LisdJ9caL9Ig$igaQI^N=S+Ampn;XJC~oGavQ)}umF
zZSD7*E#6wGF01=n_Z@hwLN{4(cA&9}iPXdd7YUD>aDSUlUCXjle~ur}ZlbY@cURio
z&0?<(B(Ye#iN!W!c`p(RCge2yAQlUjrv%WrNhUi~y}F(@?PK!WP7uSqXX6Ze{MmUG
zgO>|DigNX3+OpxWqSF%cRm}Q@Q|gPkdCPH0)yONCQl{5fm@C>1yWeXp5h97ENd>*+
z%n8Sf7S9!+lmYD~f8txSn<<%bF_Ti0nQ_88i`3*7P5)+lq9)tSUA7h;dEbUTI&{Ui
zC2reCAdf(rUi!!{>;$tE8^@<&i?BYy`>hC|Ksea5vf+(@={6)U?K%9*R(E5;!dhYJ
za=qf3f|YnM9P-Ry8U<U^cb8H*W&aCp$Y!w$Wo~41baG{3m)cA(3IRBmv4b!am!KXz
zEPo?648HqU%pZs<Qg0wI5IgY#y>*dW(NodwX%F51e<<~S%#6o&X19-nJ+Ve2De@yl
zQsL8|6Q4dEPVa}GE<#R*Sp^2C_b(IB3r^p_X;7=l05C0JdjC58`ph|M`kd$^ITwOH
zhW@tnRL3U)mS%vW54{CAJ{bvq4b2f4*MABf*9e-0XL<rXU;H?&cU*H%!O`dBsT_ZM
z|M|d^u$(#V|MZXX*W3HyuR{?X1ykBhS{Vi$rr*CF{0j++Y2zkD=0Hy0{+PZTei^A|
zUc)me$^t2hYpqnLDAY_TJ83PLFeWQ1C`ZT5`;9A7fTGPx8#lGccGVbpM`##=sDC_;
zLOmzi#6yxol}S$*cs&WLpVM116KOrXm4225amrHVm0^XU6|$E(jrtKuMC5Cv7iO9A
z)>9zEN!gD2Ibj6Ki0WwCDxd6VDeR>=K6yBKiIh$=RHPl`uk>Z1;^)c8BgmkR=xH>X
z;aT29%~Lw9NYTqJwlr@xG!HyrVSi?20xA-sl9fmMNa39%?^p_GHJ{N%>n#$zBz>$<
zV*2L6HXHr?DdT#MIc*tD9)v;BLu1(x!Thc7t^pfSGH~-PC!+*TWHJoUJAzkeh$}k8
z6|YX071F36D`myBNr4%tx>ce&N5dvXt|>=I$7tRtw|DqJfhX>^=Pq@kx__(<;Useu
zLPl+ron@0R?>E~;UXoMsY@@QqwQifNm^JfSwH1D23ow3~fouj_P63dY;Izz(=t<rL
zl0dV;O%4KaR+w=vzuJ(b8)<DCYo+(UU;IiKb=A?D>x%w3cT&2tCOKiyXQXRW7!XN{
zy1&-yGRu<!(WL~VJX_?E7=Jh$eudX9Ue{|c2d}D2nYb4%z6^7h?5Z3vA<Z0(S3LxT
zV}>w1tfq~|T;S|IIUbYiK{R^QJbVVLz;m<lbA>{G+QxKa$rQm1_PRrYU?^srC^4NV
z^NUxZ&Zltes@|9{*^CK%2qTbi-F5xjhuyYvnsK=aLI@kMFpdd_S$`*M@Dk9Sqaj7S
z6+*@`W!1>(pMcOTjo5~c1ZF834#Kz4p_439AUC;^^eGKY>T&@4#rf#sDRgWjxe$}|
zMAN48Eh1X|D79d(WKZcTUW>(F`p_HK(i#Zfu_!Ys;dLn=-cIgOBY|U<+f&lw9Bd@_
zjMJL<rFhJYx{=Q*;eU7hTEb6zrb41Rz-trTQ>Z~0vY?d`LTM(18z}}=VFTDpQkC>_
z=C2@rH$vhU$`(QuH&902$yH`MP9=;SXN#E}F(2UG0pF5L+bZ4Cr*z(S@(z7+vN@5b
zf(SgLz_lCsysh(rtIuyk;E?^H5yD$_v$^IZ(>U8h$eNb4M}LQg=8)Wi@{OoNN+|1B
z*iARm+BNIUT1K^WLCk5;&kk16ngN&{Q9e?X_oG@%X6ZN{D6OH|P+qF^I9t4V8(^cg
zdn8DnGSCEd9UcAbNBA1UG{7X4Vo)oDHNMIFhuOXc0^~ABWxlT)3(A?~o_OgAs>k8d
z5!ymaH(9(rZhxwd(5$C*Pb??&;7CtBPA?0H$k4(b6qAGlL38!68D{T|MowJxu5cVE
z>YNUSjPhZW^i&6M-g3~5Mi>Y_*0(8r8sVF!Gj(pCZx2LwG>Y=w`3_*!XOT`pL_IT&
zY9(23$kuG;11GVtj`pEyz=&}+@&=yI8uI@ql-xx}tA7VjeqAT3>IVe{cZel8vjhV{
zHI%H?!xqoxU|4wVEu4Q@lzDtxv%6;{6rrh%u{*l;3dK^cU@CfS?LYM;cie05V@G&i
zbF)Ztxu4#UDv{;m0F^2fP9-XP(^i))6|yaXAUn=V_d&hIYalP1|K!RvG{oq=@1*7G
z?jr~mlz%)yo4d?#DL>AHs!FZkW;gjTKR~8ySK#~e<G_rNSG8z%_fj;Dxa%Fgbf8$y
zo^x`iwlR0Q`NRtc)sOW?->9&vYm1gH-YNFCf*+aK=PEHH^%bS77bKyJQu<&q<lNl~
zyVeIg=3BS*R#?=LKeJ}R{*il5pKdhnAzqHk=6_o|v`rhWj+7ffbYsUVj$jJjCTZ5D
zJIY8eHTu_9mzKS5R~2~i=CTRWEoM(C--s^Iw^uv7>CSVd%X^#jt9P7LbGq#GjvjI4
z7kayq+FBpDox9XeM&uNt?!RWR%&Bql<1OJOKM45Wp(RWQBr}wr4#K83@ZB9y0}S1)
z1Ahv_Flprv>41W>RBGT|9lOkdmqMl+X+vd-`2&~o?*g^eeA5Y%dLaWfybQ2TrVNSA
zC0!KRbQ7tXdInXNIxDanY3msYOrjY%bDd9KboF*LrChh&Bky@ugVBsq{~gcjk>Wmu
z@<8?Dh4M&npF(+{dZkc~iXeYZ<{}tHv47i<D+WZZDFTX-<DC#%Gp68*p%hE|d7!ix
zU};Nwqz21nz3|WkH3L$UW!w->*7QaYWoYIwjd{dPx>3j|Z=7lV`!*UtdyD+*8V#T%
zL;JsLGywUts8#bGVR4z_|EJLaNatNGrS*TL&p;&<lj?gn8k}SD2iE=8(hG4{*$aQm
zzr#{K`Zs#o=a<3#1rq}|I5d~hAuJP@0$w*Lm$;xU41dkAiYzWbU?6s)2Ie+FZi6`u
zru)=~>Hj}WahE7BvaRGG)27HO)?@9K!_Pqu-*<<P-P5S_V1*;FIDGsVfL}=d1j&<z
z1`8mJg5l%W@Yl1HX!tqsqoq`eAIskzkF@`$!0`$oe)uiG{+rbh_V5~oX|FPAkK$E$
z<Pms$34imv-%BF{1<%hRP}=|f@yAZ)!%E>z_!@@(*W1VLzg;~#DnZ5%hEM<w|NPp8
zc?~3t^#ih8o`&E54nKB(_CzZy$y?F~$>I0Up__Tn*m=p=Nn?m+*r5?bZZO6R6<#Im
z^sAJ_Sm}`;mR&PS2s0U<Mh7P;k`YSzo+5)N34b#+9*NP_%z+qT$p~tb-V>yEVV)>)
z4twIf-G@tx+XE>|2MuJj=}1wY=L?d@@T9{FQbYTQXw7CAtpzmXgksFIAgoF8P=T91
zBy}N`eq%jo7LL*hq7YeNfTWIuQJ9FsF>eW@Ps>!nM`p}y6R6-(_qn98g(PGW1;TuR
zaew|LU#d%31rsRPoL1g?-DixiXfTi9uo_?B$A-c~I8os0!mlc_o6aPG>idjrdLQ>=
zQpczis(i2OgD&{fyCd~6%WulAK(i{G%pezpf)<Uu3h^YhazRDoB8`kPy-sxFSQL6H
z#pIk>*~97Rc>eTJR+XNlQ<dJC`>lqn%YW)fN0pgX*aeMP+%Y>?L_0O<B5g(B5S*)x
zce?+UZoAH`F!6}2Ee<;2TYl+c)8hoam(y!INwTwAXfIkVi1oaaEZOH4#+e*ewa<xh
zhTl40mM#t}ANI7VX)r=Ve^!9v*_>34%~p|dhDYNiW)FAj&HMMWsFeFA&m%j<a(~}u
zVu@5dS{D>zm^mAw$?%ANigOiSXFT21e_p)3N>zKIRXr!>;1xV|;7~m7Do$s-oaI=+
z*v~n=j8(T4-!$m}(y^Ipy_y@d0Rm*X^dcQV;Xz+X2f8ta{xg9($2S5mnC$8+Nr8h(
zSR1nryGf9}CxB?Dgh$f?NP#vQ;eVky+Hew(*`4G-kR60CtIoIqDUcioDJe8RPL&j3
zF?-24ycOQL{=8AyFr@%DDBLx4(17sD^xUS|Q`WMlk?-;nzQ(Ye%}B!TLpxj4*>Qr@
zjXKL&GH0-h6c(fwM!GM_VYSvm+6`0KIl-F27H2<Zq;xoC`G|*~%}i5D2Y=CF%?YEN
zovJAsdu*$_#cFYiWRD@qg|~d7>u_GhlAR3@-T)?t70cv74G<VbuCtl<6@bsVqH2C-
zn&`wtU|a;B1{~`gE@X&a>lAvN?3tAQ9H#hDhOFUNW-j1|$_lV*fnQ2+*C-!yh=~?I
zr?Ad1->i}7QRoF!T{+tp8-HQhPxZh)R$pn(XTY~C<It(q<I1d@7ihOEZaa!GKc{J-
zdO_XpM@m_)M4^SqQi<r6sp?`TW6NBrukY^Q%z2xl=7e(ST~^@Llu|Z+H($kp7?7h1
z<L%u<H<&)xWiF%#@!Mwgc&?UWzJA8O=q-bu3y{J#c7b?ginXW?WPd@eEUrHzc~{70
z<@AQhJ9-k@+~SZ_?ZT|e<Ix6_(~srxC%B$TJUPBvR2|nzPLF<1s#M_6D*jg4-=E^6
zW>4(ti&A{l?6@u7cXip79q!0>OlWF$%;7aW7Xx><17Gs7U}D9`bRX(sdeDjSNO^)B
zr2>=%LfUKn++a&)wSP}V2h@&-uBFMk_GT$tXpg0FpnWYEVn<3(v@vvva*=ed`*Ab>
z>z$h1*MUi|<K?1#bGTsl1wDpYrOqKAVh(Zs+wZE9+S(JUQdYd_dimv{bx-=+=6poB
z<~%Xqu6=IzP|MNQ-#VsVT0TK-_SDnmZeiqa>N?14TN&#zr+-66*vtyO#aS&Eb}$Xi
zrMX&^Fh?xYLh=@MFnJ+4ZJV<%=he9C4px=G>>C~Qa>{WtNf1M_0Ud*BiM>vmyl&AL
zStX?c4((jvf0~pkgcxwWl%FgWhX6wAn6B(t069(z?e&gDG^e}5u+c){qS5A%v#c*^
zS%BjFmh^^|No_}Q!=^<?`AXNKTm40C3ji&&GGE-b=(s-Ew&<v>Xj`bw+7>;2x3?`i
z3ZL7y=$Kt;TlBy_*0untgto0as5Wm~xRCew4SmC$;9XUKdI&Siwlx9l6KnbOM>uU+
z%SV#y%K2{O(DOfE3jhR{(M&HB0yi_4F?$6im;Zns3YS4IJt%+OT3v5kxea~yuMp6e
z?SnOvC{kpx2$0=0+Z0Y4Bq`7~2p0YrZ@hIp#(2EQ_Sg4N<2$x@Ch^2GZJa*D;oQqB
zQ6zPEsC!!@4k@%o5^C}PThJOfR|~m_U&tYgLZ})jRR~-|&HhYHp>lH}2ey6%LloM9
zKtboXYRE}l;uL=}Wp)We5{AoGFbnl3xrAJ)1kutUwLTGr%vmtD;x2Qy0(T`POd%4o
zo86!ovUck$B;zhgC`rjj;ZRC;J;t?Ky9F&2Zay9cEiL;KY@t}USrS%i+2_cCN^x7H
z1gf=56iAvsAvKVuE@`NV`dx#ByHb<8Df_hsKV|NgW=MbL^FpPT>g#9(NqrP78ZLnM
z38qyNe?mCan*0ekv2wS-@Px2?tzov-eX5v4t6Q%KMr!^n2IZ0rbHEWyh!P|WUt52I
z14}ZOEU@;w2nB2Ry?IEx*9@Z=+$0$ug6n-=6-e%J#ZSpicOc<z(f(`=Le$4&I70e0
zJR#<*wB~<7b_?7Q$vxhkm{nau$ciVPdaf1e*Pu)}`5d@G+~>t(goz+QrBWXci>b}W
zD=<<rx#X_PxS=`ET?eTI%*ZYwO?zY%fr75D1MWmJCBG&}V(w31S9lm9X;5?1w3J}J
zO&}2@kDwxW*7L;|XvHI_Sd5Vy>6vZBX@`7mM(%$Ng=tpF=Ro=xi;ssrXyP$aYzPze
zQIP1SeH~D?zPA*qY}Qxgdn+aDx3RAhHp|Yp-;SQVL3S1F@0X*$|Kp#y-lBR!ag7bb
zoL!!tzJ2uQQ<^nLhV6H2Qr2i#(HfeGW0Q76GmO5UoKB{%&i4<;f&76mda*y7ji+a!
z%zuCW@x$zgSF`=uIOO?-r<1c8=6FgSUuvJ_DN;X|O7kB`*_;6|{Ra{|`$F*J(WB9e
z>E!U$c=l%W;>pv|>+y%#mBp`rJs)G{pT;ogY&JfdT{x1=hj=-@m|RW|#}{)D@=MRh
zM<@H=O+LIy!FK_|Y2_`RpW?}R*16W>v$KCmABgLG`kgl&I(-=(I<JH>2s8*?|LwxC
zKG5jZ<-yE<?VX(cD~!IIOpnIXZa{iF`qSvy==(Q${^RBNaQ3EHiRQb%S_G&NO9a>i
zW{a%$cJw%4^jD)FCa))>C*hl%Do_Mkz~|aL`HTGi(W4vQxsMSQMhJH3l@Wlqn1O$U
zA2p%?d9o39p|F*NUu|<E3T6cYVfGvyC`l!j<ete<xUI0l0J+5!)j@m{imtL^^~A`w
zkgq1rr9=kE1KA@rFJuS_VnCFlHhQ~rYaye@i{TGvCzCT+^tYGK{NFe4X0!8)Ka55{
zoy^`{9>l}R`_a#r@6UIS#=ngIeEEO;Mf~}6bZ|O37#$z0*x@nY>>%gkOh=WDD~;Q@
z9UQfzT8`6kwPVYpi|OI;<nQ*1rrkZjEEc(Q|5awINH|`svZNfV%7a-YW0731*XSTs
zbpcXH%U1GlQS64ofMTK?9edZ%Y!s`kqe&TPi|TQ=CejkK^BS6{OEFVhM^k?+O8T$h
z@(uL;I7ojkE=ya&W#Rkaa^oAg%met!mB7OlUR-H|po3snf()=O1FXvc>oUN)YysA@
zAg2&3@X&}PEKQ?=Rg31BYF@&!M%5V|_$wF`I5_j>cuEx;09*;2ucuATH6~CvA&sRj
z!>MKv<Yb&lV2vb^yb+u-r-*;bumBq778Bz9g@NYm^t|JYYTn(mv=U7}*V9;OVkcwp
zFe(-`j)IzAnGkheay57%xFD!FnhX>#5Y;V&8y#6BOg<RlMch&3V@9mWB-W2NXiMzf
z-&MRZHu6lrbi60$vy=BH{~k|&JOMIIj{%iE{^<X9@rSn&f2y&7NTYufjoyRRRHAp6
zXtAQnD*l8LW=tjpy%PGWvK127ECNcMy;E;275~GC%n|LzvJu$gMXj(htPn<<wpsi!
zSTxj-1+g$t*<{;^|MUIXyT6{#p4|xHJQse+Ac=7v>c33RPEXFp`_qw6chSR~iFWsL
z-IO$eOrW(LmX-}XK=OY>fimRVF8<VQq_Qv=j`yz74@#t9*=WX*N`;nEND?TCp;1?>
zTnh47H_9N?Og{%2ybQY2a(e(y{&D=_d^*pjqPw0_x9O?K4X!5OYO;!>-qM;gIz}n3
zSn(Q~g?e9jw<gBx&Rjz?QG~r~XqMJioky(F6hNW0H8eArkt%<$qY2`m_Vt`cs4nqd
zmHY5sqif#V1_rAGgVlk->cC)iV6gM$&5sZ_KcVm{W;Jm0qX}<qI9B{bdyf6TqPUR(
z5U|4x)LgLi4QZt0m~V9D7&6M2@5%@4?M-><rE^p4&E_l4AWr?*1-lOY9v3h=i=~uh
z8VQQdl@%UmP!NB#oZILcPynz81{*58??<H{vthjw@VSnt5&=1Zj5?SWgx*2L0Sjzt
zIprvWdVzbgw=P^&bqjo|csVj%SV0a)K#z}F0UxyLZz#}4i89Fd|9d$^%ecFzzdfzd
zcE@SW4TU>SYgTEt6dmdA*7Va%)Q)vDiK5l7=43@vUaEgEX7{Kt{z_<3VQeTGHXNwf
zaG+wtfr{xC>PeB9b3;ABK7!fkgR!p=1f@tTnkA;H(lU0pV1FePw@uF_H2SNo>w;3j
zbstJSfn{Krwhb;{&;eEeSLGBvWeD6f^;eS#1$?=3SCm<5Ej2;LqL}2oVMCZI0!r5n
zVTK)6zqEg+-|rtzC-deIg%icQ6O?T^+geO)3txLbAj{1+YD3iwst$L<&q!%oxEdJc
zFKzGcC~j+g3P)FJf`5ywqsdnXiEC(rDyFuMoz8V~Sw|7AhreY0TA;ZsL9^z2L37as
zH2a(Jj=%n5{7RrfsDr?RAcN2bVHIkB#=)Lw^zVNQ76*F`4FKCi+Q(9{k8K-gxy4)Z
z#DP#HYTw4}OcYIr*)shtEdJ_Iu|0&DQ>wHL`def5Z(TtD{Tmv!m`d|6USe&_uw4oR
zxiK^F1O&VRy0>QoYI8yLA8aUABX;JBBAuCYs)p{&(3hZSZ_fr#KQABP4T{Fo-=#EK
z)dhc0Vt;p1t>gkJC)=!k)fS1aJI~@L=|um-15KdX^yJ`jHk@T5v>*pLZza*u)|h{b
zLK6KA8+qdUhgxX9NaVgR{XgSO70U$KxIR=ZK>B14_I-gLBn!1$y^ep!^vPc5ro#2t
zJ@7EgI-1pAAe1#U3;7pA>&#x@cA|A=N2PyUaXz?8(>Rt}$Ab^MMiVd9XrlYnXlmCr
zdj28Rrf-em&r*HnC>`>0)cy*$#1)4x*t)5<Oh|0hXl=;yc8f+*y@N(R=z~<9zv90Y
zr}-yWpBpDmH{x{lO@PwXiK-5tdg<_~mkytL>3l`^W=_CK@3gL4P5&ySe(^J$i=z&X
zXpQ|{#ucFb2eFHFXqVAUFB1YdHJ4FM1{0TsPbVms!qPC90lT+b(=ZzXe_feJ6tedy
zu_C#=5#fmnR;w3iz(|4wAD##8KozI4AS>N3z^FBmPN7_KEq!9R4i%HW_@yxUby1AZ
z=SMPLX?`L-62y``SCMAX?qUm!qv^5*)+D{9YKlS=jB-Txu;+)YI2~%#A&u?56`#ef
z5?w`-Q2!6R8jH%==Y+CQe^NYwP@*F&X{9I*74=Wb<!)s$YhXMgksa)U<9dLp8&C#%
z@vYzR2(K0r{8$$;5!b+T7umibxmtE*F5^OmafmE;D{<0tHI1ISq*S5nHdI29XXDep
zC*Mmog)*+&;73d8OlX_5GqoX;hr^ZMg-9`t#W*Jk74o5D=OZ*6e{wZ~eVva)@>}lO
z*D9skdD`NN2$tholTxux<hMlCE-j#N5kWG-#Ym&i@gURuD-AVWhuZr3$Ssa}S6Z|@
znT|$duE{~bHd4GQ<ZyX}v|fh$QD}|Ms=Z%iZQ)5PJlTtSupggIKr5#qhC$gUc2f*i
zIsMGj2Fh&c!aU(Fe+IG!R0Xb#7P0fCV4IuzCo&4pUOs(%MC+MU3gQo%LF;GrOOyr)
z^naa35J3^Ro09rIL=zFb<(~Uc3j#TwA-9QiR#C7%6cct5jzp`(^Lq*|4rN2*6YeUv
zG&xixySlH^VWK=cyqGvtDy;is!$_x?4z%1vdf?iQOx9H!FSYg6TebO_m5qnFkrqFy
z)h12tK$scjg>}f;`;yGuccwgGyd95DJ$PpB9b)#Wm!Q`$umtTBU{IIQOfM4xI5?M~
zoiY@c7<eEjm!#M*paFB2McFWHe`e#tS6SOsSYEX>ObRFp^K^t)j(Lu%QX*01S!IE)
zE59JN1kUL}QWL7U0^%(9a&Zk|T{f>Ze?^z(D6j}?ereVwm9>%<hxAxbaQM?Nb<z6E
z&sp+|6-zCdwJ@&O!&0fpH@ym%>yqyvwk2C2%m}VVN?|79b@pLv0&s*Pf4R_N;`bxg
zVjUCUR@Y+EqpH@!%%)<Qwza3SKpl5X+M<H0>;$|eT#irPHblty9zk6qZ2(VPSEefL
z{Hy_3rxsXBmhWkn!m>($R!8aSg*H>T>i7(^wf(xS+pWY#{vz!#%K1@AIA1Sh4<=hw
zIl%y{sc=Vy&#T)h&Nkohf64$O+g9O!1oeDv3!Cd(7<*}mZyme}7oeCmcu+{N;yQ8h
z!qs5A-@i2kDu&h!l!6EG-UVgyF5c7lEn<^*I>$B;yS9gLhN!5K3a$R*G1FqU%H2k@
z6;SK@o{s`$1=pqm0GQC~9t!9vap|I82_b=w0+%W(!M6v+KvGO{fB9f`DMR|;LoX=a
zQ5o2JG)xwF;uVwcY|YR$y$Q0z67Qk6j#3Q0vfh%u;2iaJn@*47`8Ks8GNbg!&>hvf
zOPQgyhD9kr3n4}AlUqgVk=-Y@An1Knrc-+czo+LGxU930M3~3-dO1RA+W|Q)$XUZZ
zAlFf<5&tKe3@Mn9e`2J3Pm_@Y?lvCzH1k2eied?Q2VM4*uIuug^m$GhUp6f|XN=iD
z2lEl8Ij7H<6U%)nr_778vXMcra*B%vm8X`8%2SUGmB+7H<tTKc^5)x2<>#dKb8+x_
z9^C}3Aq*DqBF9iTrPg7DOeSiv&KNkZD-^9{O3G1QZ^)ZAe}kwD8469omOwX2ZN{5W
zib-R~k@o}JCdho0O+IU2j9*Cj*eAFN3`1*aSrv2nIs+!oY!RYsfr&YMNq_BvR^cKQ
z1k%_p`7Eus!(3pKBrUL_1w-h}OK`r{?nk!JjwxKuwLg2_jznE3hYd6DK(v&UNLaPD
zzV{<KqWSzLe?(IT%0R=&m`C&|^yJY-Riu^?AF1EWtJ<kowVhYBB@nlD|HQe&2-yLp
z|Fyo?R^dna9va;#_EvQ7$kLE983p&q(t1ksX#XZG#%O=jz`oCwJ58B^8krB#n}r4X
z;401_MefDM1ym*5m!s{_D<TXg5jHOow(lu(^sg!ke@aY$&RId3y)B4Z?C9;a&2d~s
zI5apFd>z#k4XaX?`*1LG#Wc8kfVra-0c_LBf=wq2U9D#?IwP2G0BA~G2ZjPbof{oK
z=hp!0?98Qy@x>5^`Ae)>z6H%)JOujMeSvBmw>P%@?j5{%Zsy1pdtw|cDGijin%>%R
z8CAOaf90177@Cj?Y8s?gWVKwOs@pYic&U60U_`4KfEefQHe8%HZ%|Xy^Pq=V)qc|2
z)wpAAGl8Q#!{UPbYcUCDj`$ize_*sNC2cB_z(8SVY{C}{`@G$?eZFET6B+pkk%-i5
zk%(Mm^08$jwoYGX8!^(^ib$ksH)8ySJ=K+nf28LfGo^LjhbU?lfL1Ug>-I@>W%dfH
zsVw=!Jq0b-@r>cB!HAndfoOwp1^WRn6)>7}<n{xCaNsnvbK}fAfDS@3feQ(K9MX5e
z90f(%V!=hg`5@th^Psa1D^aIo#2x$1@U7$w-vt{DQg9uElwui!TNtcI$z~1v!md}~
zf427RT77LFf@g6H36Ua?d(=OBO4A_xHY`TV`awarp+?jrv+pdKUJJU1^Ofsm?Q&4O
zC+Oa|g|IBnUX7lFy{$4g1s;&3-93u6J*9XU>cRKD9+<+Zf+~YA9H80+Ow$~ynvALl
zywLTgfO%l(Q|s~VnT#3EV%DIi>#0n+f9|jVlR7n!dszpRO4n=fd!i=exOxEx&*|R7
z#vKLu+y|*WFV_J^`JTH+b)h2^;bhXyu;eFigVk-AqSXSW6Fn1e&%xXbQYm{%YMa3-
zWiN&16<DQSBdj*Eyz7@*)&Nshm%-TOgQ3OcIUD<1d#O+jSPcZYW3Ea+OQt!ff9!9S
zEw}l`O}N2Bgf&NN8|UxrGA16&O5U8SZX!Nivyqna%Ubv^$HCeA=tYM5_2T&KYNb>@
zl<IB_MjiD&1=Fp;f=*d4N`Xw&8)V4R=a^28r|D^ChEjBjOG~u=EiR2v7E)Cq-pk;z
zAhqMMVQ_1+8^Re544y9K;+XeIe=+_&$8FI%;#=>GN{&p(J6#!-f?()A0RTtwypZqP
zZj%awOxD8zP)E!~$R(Tp91`2gQO9gP<xG|}o_MyfC<mxLIGSN0!~ND8>hg8SG@1j|
z{%}@xbJ{-Gt8@7qg>{bk4UYv6^_bcB$C|6Pw>eUeeA2cugU*t%O%d!|e|6ojteE&D
zT3-PeX{eRFE-P9~CaW>paEJ4wTM;gnjx)W?kBXILiU2l_%C|bfHCju*F3}kmXRh;?
z4h5pLW>=26g5T;S<e6nO(Tv;@E9Bp_)YxA&AcG)#ZokOKI7G|UA%=VmhknS9|2<A^
zo<E^qQOc<)4^${k3m@dAe?cv!JFl|{%?!dgYPjyaeTA7!s9lh}b{Hv`GIEt<dj$$?
z3w6=N1g9bXacKnRSS%ZZl3l7=`^9iv2xC1OyZSri_^z2VUq{Y)q0X^h>F-7W?yVc=
zjj6~<$ZCOky%zO8ryMP>d7D#4&5-hqfw`Ucp9M&A<+U>C>E~UWf70-#;D6Oc31q0u
z_TBTWf|%+zf&8gC=RI}hg=;tZa<!2Hr?oBbPu(brXgT}xpU0_Tp^yM6%ISvugTyhn
zEkNSewjcG^<CHhBw88X9i1RsnNTXmNB<8#+U~-VbPmfPR>uVg6ja3vd-N>P#QDWdw
zk_5$^m_-H#H&kU0f0{W1R`o-?N%j0$#*J)!z_JS^sOdnXx`}m09L0oAj;oe9T{Uq3
zU!2BOPxazU;oSDy=F5-29_QyX!0Z5;HHWZ4(Vza|AmbmqumH-0|9v(}Ma`w6ZK(El
zqxgJ<1}N^g;|hoW^pM|_I{Y8Axjk(PWo~41baG{3Z3<;>moE4)2bUpsArzNNM=B_m
z+ByW6&&MtZm+j&(j(?#&lt2IV`ul@qWu*u<-p0%L?dA31--oKYHUe#!(1HN))89WI
z;sa19tP7K(D_S@G`e*w7@W&{&ZYF6>{)2M*^<x@FzA<`15_-~F#11`e1)`s<4Z_4n
z89n$aB@$Y2r4LyTgp*dCm0zx2tx7~Ij0vMef>9!23xp{o$$wdzfT(Brlhp4<Nc}3=
ztK{iHvX-4*b#{=#I2dP}g&s6OK9ppodSRXRA>G9joB-qAfTtv%lXzt&6*o$|b|c>o
zVBYVl$yf||H^g~oK-R=>AP2(gg$)rUTtn}AezY>?4<;NLLp0O;Ct1^tw0<p^87si~
z8ZplUK6mhn0)NV=&vr_;=|_Fi2|M5t_c519BdpZbCGPGnK}u@t;gfwb>>8Z{Hsgp?
z=+S26l7k>?lE&Stb2Oo(C{YHf``$4x?4dNGOA0mL-drTLQ$ouHu1XFVzC|JmuzDd8
z*0FFwIJLy$LpB91;I1YGD;%{GqeNUaiGo7}zX%$T?SE;39QSh)l#v@rU}O*wmPwGp
z?a*m!v=-W`Q9|rmQZPHNG$s68?rhS16Y#*DPYQUR8}GF}rz%pW`T4N?Kvf}agwykp
z++~4U%L2Y5mzO)(oq@?k`J2~J#$y9eA$exAP5!QHp3e{f7i(bHl)4AJ8CJ}_<_NE9
ziuQU>S${{VRvJ&k2CcWT5C*E&*$Hb;=F+(Aqf0Yt+7kwgdIb%;IHEYXHyd-5Uf9{$
zHbW7wTtYTAVgcj<p>ZzOd_+V?#?C%DWKHS@UqCIWe;=eiU1MLbT-4B=6+*N<MuSP7
z%8uuNe`9wKal-&(cTb5M6znwj!!<SX8z&p*7Jpz6+TB2VeWy*diV<{G(r(Co)Tq;e
zcYaj-G^SKZgWBU<<4kiZ$#711&-Lc4^{FIMky+9y9mlq;Pv0(Omol1)!llm9+*N&T
z64JF2V*01_Rb@13+mB$oLrMeN>R7eWt&H1^M4bs$LTGj5^>{Noz!0w){lbUqmNx;c
zAb*q=t9t#bKeMITr&v9LPmCP<Ea%6(x<i@UAMCK_NL<7iF?fZQ!S8g*kJJ^69t)-A
zO><w)28wR!NwO@@QL~TYk&(u}_BJI%^6YX1T)8>*`S8)fo#rPgx|2@zXn3BFkrcaG
z3uIqkc;8t0pJp8ysU%66{C<Lp#wov#$bT|#MuxE<6dkW4o=;`pxL=mUS#>LBOmIT0
zg|ylJE*@qtENXx<&w4&48~j=v93yOdzZJ<9U8)`S)b}c-Nhd7VqRWIJRG4Ev>Jp;N
z&{hN5la$_#zHFa*iL-6d(AKrKv4UF}BfF{-C0yG1$s4*MKd$IjDiF{0Y-O|-N`E|#
zTDOs!G)*z&`xq+nUy5EuN+Bv~wJe^^amh2dj`6MwnOLiKU14X>rKqWuV~&-%UCQl|
zZQT>oak(7}pG{bGy{8a{+j&4+S>5;En|+B@!f10y>~4BRVh?H*mi#5LIsZ!2^^T6O
zvK~dOCya2cl=4+xXJUT=<Kj@gT7U41HP)IoH!XX=<R~}W>0WbYTixblT0i#~RpKFv
z&Vj?))QDmkZ3d<_Rr4Bwu8sJid$D5|P_s*R75Y-NFsH>gs5`cKkGvb#4t~|J2xTJU
z+T$AjxqHQ9XGH9RHRw^^UC7bwH+9-=$EgJ;lI%xCuxa~h<9&a!W6;qWeSb%`V~i32
zdbLr;Lv+U|BWUNtbO#U7onGqyNO$7vB1ca69J-^uriZuSejne_fWimA$#(+s3%;9N
zC(Sae{mQ^~0<3YJw0^;Lg5}p6iR<J!TU@6~=(tXj(09(*a-F<t<T`0p<qCPl5w4SG
zLY43it^+{D!Hr(oV>?Law0|*+RlNn(n{Cg+nr<<j<gK|=$0gHQtKHfd5x_KW8y}5_
zL%n91+4_vvHf#r3KjX_e!*e>ev(9w#nk}XSxy|CZPB>hv^bNMd8e?<TGSeygaetWR
z6{cf#e5Nn0jcH%UpUc=ibQ)fA2Qgz4eVFa5K5IP5Rv*WwwbK?EP=EaIZ~I>>5qyoq
z+Sfb1l1Vi~d}D9uRuqifR$r6n?enFKJbHVaW5y~^Niy2>UXDrJ(x?SMmvvIe=9e@$
zXk?tJnkI9Etx7sb{N|KM!S3`H@$+_#_~qT#iQj7QC@JVDSp<*Qw@9JaJEYKZ573~H
zld~m-Uhk1Y-vy*!Ab;!iND93kNTEIT$KsL{ma?K+M!2!pQvWU5S9C8gA$AU;{1b%1
z8}AHza3BIMq&0rIANhk2(&014(J}T%P7s#lkE>hy;hFxeMXw58NH_l+;w{uLAI0@`
zjT@<9eu?xJYFJ;}RcaXQ7B$STxQ<Z6Jadg2_BnRxwCKOhO@GpDYFOsC^gFzP{>7Ka
z!THfI{|Eisp(iCkycJ68Po{s$dZB!{BmGlAyxN79H<X?&6;D|H*Kg;?277cd+Ty+0
zql+KbVuQZRkbZU7Yq3o)KgKO7NT@&jBx)Hnv4%c27$0mH7XOgr;|L$qn_&45l&Da|
z3T19&b98cLVVC=dMF|2oIhV2R9TNgEIG1t2Hz$9+8M}@fH?Y62*wQXclQY91i5Lf*
z&L;!uVqBM`3gQ%jix?^L`58WU$z83adlk}ZxjXOoNKb!Fdirv6c)a;|M4FuP);O3R
zzfDFSX!$l;4%tr57{zF&$M4gxpR@*+&p|$<HKKf+{PuFSw;#}WdBa%w$Wvo(KR7ZG
zM&5rx$gs#62IVcdm#dNMT|5`~pjk%@K|Y6gz5V_1=Nny^ZJYrr%VA<iN^n)qB32S8
z@`A}mx)e`@6PvE_6ym91x5ACJ?W8@k!X0sjqGN%!4IdXsli!SZtu1fgmvfgTx4=D5
zQId~GZ?<#|4ipfL5+zQ3#!Fl<Y+DNgs$hQx3<eJnX%|`CrW|hIpVwM}&zTP~x7fVt
zD&mgPe=anVMk)=_8YEhgf6Vt#@3~wn{X(1W8gyq%Sm|N`m*kjvE^oNT<z*IVKE<H4
zhiZYwsoT}QA)~Z$J@i`pWcUK@=eL#`$R|nrF^cZnJB_;%Z7Ke+oX(Kdo0HGN28DkV
zWeThp-nAn6c(geEkhoi)ew`x`wN=iIN{eBaoP&$W^k8mUnQtr1k-f4SS0*V2&q55Y
zxo0Gys>HiSeERMT6bKg^j?vV3jiaw67c900Q`mHookTNpu}ErBnw48ytqiop5-n2R
zF1*L&?LL)&rji67Vo1&sC5Re=+2wz-<H#ZUDQcHJ%awzM>^z%K#DUI+A%Y5i8HJ6w
zEzJ4OX+9o<-m79PR~8G`TCthu%lV<)<N2&nuv;G#VB@!U%BZL(P&|=oyj3BL6z`5)
z4T$d6Sc*!!uLu{G3LNznFJFechsT?LR&HahR1;y~kMB3}fw-z-7bbvIYkq(HIeokN
zWyD$xGG0@sk+i6pt?iA@%342(nH5raqK&rMIH|`fAw-63N3+LW^jZTn_c$l}g!J_C
z@O{-FlC&juddVmvPs%ztYoRdzj`F7Gw!Cv$L=m$xj-M1!mry^^MM;S;QqX$puN7T{
zv1@pv;39MiynCMM<oqGS>SKRyZuRjt77+C1NWfVab&#IsujI+y$)b+{@QRs6CXHY;
zpEsKWkdBL_pbZdE*j+_kxxc_So|Z>QOQw`&rSZ}7#)1>O?2>aHSu_bWpWfBZtn{`y
zt5BI}#*R6|oHN{<YewGDvk1nWYxu4y==#UmCRcQfC*xh(D(lS1LnnVLoRe<IVJBkS
zNW~PJ4wPOo{~`zF^h@SnBvQz*vecs9P?@z81y^iu^X%Q;xhw}GNUy=?bMSLGIF*M<
z1x*77ua$=(h@ZY$B8DIjoqHi0J}p}TqjCZFd05W*(%a-QqjA$F?w+T!y8&EIp0qrz
z<tR_{Scj1r=a?%St~h_w&N%ivuN4pThQZGnI^~u7S>B$nguQ4b<#^r69U3>(Ai`1)
z-Kcgd3dw<FkCm<LVSilpl7BU?pqiGT(`({#Q6cj7XJw=9x&hWme)rvAZd=WW7^!G9
z)sI%GDles~wR5wuohkad8;0gk>(WjazPo9&!uy)_{q>F%G$nshjZksrLV(LP-+fVG
z7OlUdj_pHLSP`!7N?H@pttp%0E?8wrfl%b^$nCs7deziT7NpFY4P9j@$$)!_2}#Q9
zyljs2s}t&y>Cd<E4V!Vw5^XMsvy&DbC!FG-rn>Rn*kZX$$GW<vmHH{u7lU-QOHTr$
ztbvZRcYnXnAX0x6HbGcXpdREmQAdH2c3=w%rC(v+L;d%DmpAkZWat?PX}!-L`Lu}W
zpv*pR5`!@4;HUQ=4D&e-*-8nu*K+6t$_yNWXwd8vlUWxXoiLT3SOtgP_8>u;#K<2j
zT^1SdV4q5mt{}iD640&&Q93lE2a*1LzZVl|MCsa$F|mJGnC}u4ccLMp8o_PJ5T(z?
zmEVHhH5QT+74|@SUoc5Hh%3R`jB;f&7)-)8^nymNmfe|VP$s~seAzPbb|+s;gP5B9
zlMSK~msl(lskGg#B@v5(8ODYaEvID>CX5*C_O~@R$Eyc+<4qb72F#$h|FDw=JBHHm
zHm>-P1Id5SI##HM!qa(`p4F$5K!z*SDHEO(>;@+j=_I1v$=3^1-Vp`7YJ{EjCe;u|
z38ISZmg=N{u}&o{DY4GAIL%*qo?V_Sjan5Zy(6+2I?ijn{H#gh_J;7SZPHVHu2Fhy
z7g)tT&nhtmdm38S7qyD^DapODReGw?DwplgxA%X78ilwUv7WaMJ`@Y36Il5DHG51M
z6o~9d-RMDUS74N8p2JE@hPI_==}N?XM(Lj=lMBij_mVPsl-|j#FxqXIyfm1=EAQ0H
z0%41*&wgJ|cKkl}UQU{d5%c}(FX>Z0j5<hfzKWcuk5+R5d`6}SPVu~U>>WFohw2Vu
z#QH;v1~*y1VUyDHosSFdd>*TK8_^F%NX+-J__M1DOkwd4vw!X^niXbvR#<b$?H277
z{;Q6~nXcIKj#ueJ{|6hnn$njc_%IZgLG~;Y0x>g}(T_AIlfQ8cm&Nihi+_*kfXD5h
zufJ_{G4_5R^pd64krLU}(x{2VGEt;SG?Dyfv4TY-yWE$)`_dm<n>60<JuGTsR+c7A
zzU0{gN5TZ)O6|vYL`ErauG9XwkH~?GD|TC8k}n*EV>UpxA%UoZjX1{UrTg{#y7_lA
zuF;@k8wLwrp}_Fh_f333Ab(0-7>rQ?X!!B>@NM&ZFSTq`-zG0hh-xdw&k!|MG5f(9
zQdZil_D%+i*U-N7Jw_SZ=ztvM?6-GTn};$OCxp`5r(-CMj&~wN3RLX9qJX12P~DJd
z677mPoJ|K^mJ>C*8Z;E@sM?d3xl2pPmKjlF#59t}BjfNR8hX(C^?&8G`v#|i_r1Nz
zZA}ixD_!>-V#c##>FO-ts?m-9dVsOh(vBuqaGsLtJHz&kaOaWTJU^L`HX^qBaV(cN
zaRjxaXVNCwb&M;P=+fsVIOFf=r^#2JMmsP$fJrV}W~SB^c$Tx1q*SHFapzFY!yzeH
zN;2qJIWoNv%3Dru*?)py6dUR$AI96F1tf_FTrVSmIRZ7sM!WS?!1nF9HW$VmV62}n
zsBT;9<zdPUNrpY{lkAn%+}fC}igwf-Fh)yR!9jG_4B(4P)X4L7UOQ)6&Dd+_2)E9k
zWKKDTv$LJDU|z<?am5MAO*WPoz28g5UdoH4rDjZdF<QBs_J1j@W8Y-Bl@*GlX23y#
z0xgw9`!f|vX_=EhSm#VS>-k!$RyuK<tXeFtDnO1=@}=gE66vx~16^V~Q3u&7#L!DT
z1iwH<(a!pj%%Am$`J-eH>CDneHhN~)R3fxONBw9#;smkQ>R2+Ki%sVtJ%=X+tK@L;
zb~vMdiFCjLqJN&Gbk_6rJmhnw221!;4x_3#Re{D?4{#H&GyMu8M91PMpG8ud-?iXz
z!D|Aq_AG#B&`G|;9=w!x-sF3Rf(mFm0hb@aR=X$yD#xZX0O%;v7(G+>*6Q_vfzj!J
zd#!w2D{3W2?@sAcb8wWS*YuGT&wAR3tlln%A&Jg0)_-@MYrjK0Syaa2JLacOVm`6V
zi`+@4zXB;X3iTADr97t?0IEP$zqMs8HCFfaJeG>x9MKCj6DTBZgCZfw8O?ewj77Qx
zbT0E3>lHuqAfX^FdfWI$zP$K&xyaWTY|oXue1uQ?;qp;_4AgRvp_0|e){9{q<msNY
zCeHbJt|ndDrHy|nAi98rFFdr@E@-Y<BQnQ}v*<uC-N;rjkdz*@DYzYbvvlPtX9bk}
zY>a6y9OGiRVzP>DKr8oSfRpL6+X3p2-^&aG#+(eiieWgX@x&!ovQP-U#*bnZ5R8hc
z<u<HBtQ7&x`aL?vB(yXFd<2bPyiv}&pN2;0>^(~(Ec<`wS?a(^CO!b^4a|YDBuk=S
z)WI95gAS_;xr1KHI`W{KeHnYu3;93N2S!dcUXtH0`rt%9L{fp=N6`n62yXgJ{=i5z
z#Oc44cQ~JS65o+O=p>)v54zDe;t!&Bx;cT+X}yv`=%n40Lg?{Z&mnXY)}6T4U*WaH
zr9apKu*X~P@S2sOgH>Ka9TdW%*XtJ<i(c<_>;br-i1`EA17Ih*-;>?!B;JTU0Fu`E
zPhbxKrKUi*IeP%{-_%bb4;JG4EMZ{VtOouQ*@E=%q+LN4^sO8?{sTMoyik`R_%Ich
z0sS@;mr3|AJ%0ntZ5p`^<}~P@Qy-@P|1i{AD%D~;ZRb!fTP2ZCkBi8%9R3~T@V+~K
z?4Hj^gB4DLRl~=R0r-LBPmnxl^k4ymK^Q)M4u8E!sSIC7zDP<$zAS%teAoLY1jjQ#
z^TkgA_D_}|tl>GR(pofW4e~5?;ydvD5XO1EmqrE&o`0`nxb}a4{IQc6?DN^nEu_Yh
zVA%y}&_s+(vj!c$vo>w>-H*TZDWQpTc}JR&Y8{?M?ayZin_9INhDLr~d1c^3CVl{Z
z7s)<QSivkCO92Vgt^;#82U5WXGVC5Sy(Aq$cF@5Viu?+10>$C>-u-&}*!{PwTUIEv
z@q+;;kbhbE=jSdwAYi4|4@wCSH2nHL{Mh~3OU*Xy@6iDx!O}4J3Bk$;9rmP17{1@v
ztjq9LgCXx+J159vkA+;8sd$27p~RO%jl-{?VmA8ieci1{3Jbj=I%lh?yy+_mpuFjI
zkGiyTg<g2q<Efs*u8xkkh1xj7cjKi>n_hgfwSVzKvE{X<C<+^lpk3aG8QH!vEoch<
zGcnm)X5^55$C!}misLpbJ0*o7mWih0`j}-MBb)ItR&1A&hGySZnl2`dPXkRRo?%@8
zx?*;+g2;71AC*vuy~}(0qJ%<T3uW)`5Rvm0o<EmvNP7X&OrTC>KOu!PH-rcUAg?(g
znSba6{L}~M^fmI!2$gR!gS_ogzZ43p%!@>{GBhO@j@a~G#h@7REfZGO2lRe3a1c>g
z&O;p|CI6Al`&Rk;slnXl$-n|^dOk_T3E0jw@t#!18r(#ivbVS|2zo2KIcdeL?J|QI
zMW~8&j-YH!S9$QQ^~NJBA}^LrX?x}=9DlE%YZcO1JnGVN|3s$67V+4GKFWu*r!-49
z*qG0khRN0gTwW7rWGk*p(B%xu$%3&N3}=aKBp%YL;JakpOQ)|eeaOCcuKreZU>V6W
zu0yuso$0AOCX3f|_ShUrc7riHqK=3&>+v@um#1ozLvwn&>Y+0mqZZ(+M=pCA%YTFz
z%YlJY{k1t)9>@GZV2ZU%L>FWkck2mJ9zU07UWqdvMp<i~6ZKd7q+fFh=P~Yj&nbv_
zJtTi=<y_@CX?E{9iMP3B%KV(;60T*|r*neEb5|6fHu1k`ZN95oA#6KSIg}$*QyQ_7
zGDa9F89cCfBdG;ScgN}ttrzI+qkqabjkSDhz+Ece(12hbzqouWE;<3aUA|3$KB{~}
zzH_JlMdceaEF`!G)^AgyQ-0)Z{rNghdE|#<+vrLcHwNzhDViM0h@NJLyR?=@3u|1D
zM&X#>+k6@8$@4<MuDya_1ac<5oUQQ$8WT~o%kkHwHZhHPYK>A-TdZGUvwyM#*PJZ*
z*XAxKsk7p8IjbjIxy6b(*DzN&9+cXX{dmh?tQKBX1G$s~B6OME^SsttTn<<8N+_Il
zTNZ>xxDc+KO@{oTWpcsE$hREAF{!VHu+qrKMD5AxwK<iuT@-X)w~`dJ5s1E*7->>X
zUvSP;QNUR~BK*k+!>^JZ#D84hNmg=#tIqa(SX%05;|xd1ChQAu8PjD-FY<{X(AgXK
z<Rjg(c)%eX5J~G~<l7aTPqp5dM@HsI7&Y5NtVx&H<6_6in5n}X=(n5gs*pP+i@iRe
zj-MMO5(rXuCh_$*ByvbPm`T)&3@Y-%qnR;yyU+7!!exWLCpSiM@qg&=$&E3>YrK{l
z&T2}|BSf~br*pWCu~$YmVoz~l8r~E1l25gNyOg^W=V6*b`?7kbscD$b@gp?=?MRyy
zOU6-b7v)FxrSv{pp+x6O(XHRoe@EdBMYzi07T*l;Jcc;;jvMfCap*t|N_s(~b0D{5
zF7}csYovdu<T)WH`G4E`z&oL#b_AgwSkd$ZN@Bt^Rl0oM-oHl!4oReYzF`M6g3!!7
zIJXu{hoEXpz0zFk<Suc&X<j%`ZHMZgKVjNotPZ>PO4syo08Ys3-oSrqALHBL$yPqG
zkCl4cX`%XkJgUQUQdsvz{hi}y`T@?A8vUD%kV}y|_x3|IA3GG&eZlK2&yALKLMhV|
zA@iT;r<)P*QsJHN5x}B$u3m*A>?W&R$j#a^vlg>+jyGHA{?2arKNc*NqL<N3FB1be
zH8GbFyfq{XH7`tMZfA68ATl?Xpn@zhe{n#+q#wMu^RjP6V8Ds%)P;-~mJ75Mgn*JL
zTd7Q{<VmpqeV$qJB+{f}N@OF|K%%+h-0tk`?9B7bUCL4#-=wU*ArTX(kyr(8v?LRx
zi9#meDh3mBlS*!aCu;0T+$l}Y+{8qsDXe?AqfLnsKZV>7d4?1rvCnpoAq5*Me@H`G
zZ1kWFMOKqSLpc<aV#B`FXMMxMHdQ1@A@7Bhv|;u29z#+a1*WrRvp?e3z%~bL<`8Wu
zFo>hKk#Wf})Ok%R_C&@Lnat)~8ZOLOMl6l^U~?qj2-T4jTqI;8CvHT|8#ze>qpF@Z
zlBnv1MzY#?DK=74)k}>M*}S+^e^$X9sWhtI?qL88bsI7@8ms@{R!zlh&e2?6?<Gj^
zxkR>YsQ!a#p>AP;i-R!k6&M+m(;n`5NH7f&H6i&baTCmNNZ4ZXRnjIhmW5}16N9f3
zLBci%&H+-c_psTvTeue?#fUku=$LGO3t^OGUMoI9vK=fiD4Ic#uoz$MfAhHXa0w(V
zi>n2;K((MMWbu~jI?Q!G*c=EnUw>t3GDn*YW@It{6>N7FwNW@i%5}Dc7-gFSzQT5x
z*DziVRG;vWf;}O4HYs5a44-ST7cr@hf&&206_eN^yh_p($@STUV6WpzhJ;@=mk6oU
zdpHvb%38ch$<`q_9@!RHe{dE`sTLr{$$Ja@f;EsQb4tNM!ttp~RvZkk7DokPMSix8
zu#|k3I+og5i{@@SMC41@`~Lg({g=yO`(iX2ckRK|ac5Wi=cE5L?GNM0>2PxBkplct
z`?vOW`{N<aF5V4Kx<e|_6|6xsyNuQ2@HoXxSc!vUj@lOuhCgV3e;U6Zw=bLTFgFgq
zFF4UcQk0_q^K<-LHe&s}SbRhMyv#1xkz(&GJ}9uW?Gui;pO5yI*3rJW8UCa@ACF+r
zKi<8z|GxX!b(hn>wC#uU?&H<5I~jj!f4%y2`RsIf-TwXR&0F{DMSFZPK5oy>aMVxE
zIOk)cGm=x4GnHW)f2QNpbgF#j&$OHsYNwNvb~@~)c=AvC67|`>WX2-*zFy^hT^+LX
z$hVZliRCkjaJgkdaZI}EYEJl>0%8nSNaEo08eiYj3Z{Zt^~MSza*_FWtWd-S0?jh2
z#$YL8dn<$#v(l{<Ik{lhcUggIgan|7chZ1V74<gmULK+te+7*GuKnk~{@qwU!BC^C
zi;JVR3RsJ?k`YpI_+b;xXs0A>pegvPw23Acb3t0rY}y~k7vssn<=|x4h&|A>ZwFmB
zoQxWtUH<iR_tQZ)=!Ol=ZoC?gI?VA(EQeq%Hp{DmlZIfKU7*fj4`p^yOEnuHc=!By
z`*t!uIT&__f9>0suiE#+&)w0&w!R&F7{Z`YHym|StD^NOY~OTzH8~khXNA>ny&0aK
z4}KVbuG<}{mVzTZKf#ktGQeP3HB6>6kWi=JJFGClmQhi{T+m(!y%0?}dT55Ry+dm5
z6HG`*bV84vV6F#~^TF}OaF%5NkmT7uOx#7DkwOSoe>R-q1b)@1F57j42Xb0DTR=aI
zs;S89F(nlGn(ZE*(p7L+fGvBEtBO<A$M*PtuLh^T3_o8^s>cDEp5cFKoxe9<Dwp8m
zZSC^xP}85*LOG#;aVpy_l$0Hoce7N|?10i9d#NYm%P+GsO{mffn`t>W<b|AzQfpXy
zwJt1af6cQ!4i50pCYn-!mNrlX4?yh?n`n}E0c7CYnheO6!X{<{M7!iSSvqQiLw(&u
z6X?az_#arl?pv{PSgBZ9HmF#@y~`CVm)nXJ%e*Wl-wV_WeKAXYF-v_hOMNj*u@_P=
z<X-4U&i%-_A35`y5_n(W7s(l|1gCQ$&yaTne?5z$fLyC(A*Kw{5+l&Qdaz;8#~t84
z$xh&d8zX@GK;!SSfif5oSQLRH0i*`d#_eq&RR{hSG&(jsR5DIGD07Go<Ypb?Nwy~?
zf_Ns6E0*0EA>Ii{A``n90anyI;QrOkKtWAV3=9wAA0@v#8-T@aPgn!;32eaLP>!pP
ze*kq=O^AbNO$zZE^UjP=+XWDGM!_Mn=0|ogXbl}<Fm<p1M2-i#Vb8nsgppt^ihK!Z
z9zlu=s-CWT0u_-%NnmCXj(}5mhmJr*L>yBt3A;hoH^S|8M*}8A&+tM9c0*_s-klLZ
zGO5r_NT`J;kj>XdB(ipz_u`Rk3kB#se+W9-BCQwF&kHQ-$o|gfy*JDlOg^7qyYb{h
zJ2;-Uq=+U&YuE*gpLs}v2be~@CIBInCy5@GuM$_(0_<i))~Md`QFfO+M1n)`-RzR!
zHi*=k43;=p#3G#+1|KUsOrt`@S@V~H`ngyGC%ZzthQV8?1}i5xIyp+nv=z+&e`mqp
zasx>jw8!Tiq3<_??&Hsw-D_Zp$r<p>4LrRjhq^gKpTIVg^W&>-PU6^y45;n1U&f(x
zSaWx4VnN{su_bayGMi|I0*m;6fy3_V9DIe)tb14HB>z<eN3fRb>>}6&&DF3)UeW`A
zoHbRy0&>=MzCVz&=5!C_tY2Qif0eb%=K}YF^nz~JdJqtau+Vz|T_{1HOmmY$dbUrC
z@p_+CG6irz!*Do=_KpSVNCB0ls72)cR*6MkTu#Rq*F!hk<TE7ip&JdmHn`9p{Cwa(
zcAvH*vB*A#DQ@|CM72SAe=^&VLDi{*hpo~Ol)WinBnGLlE<044+Ga+^e}8)0K`|j=
zFZ2|W7$<l#+kyB-I_KrCLh+k)6_&N;y@EO4+CUR%#RHIkTT@h=l(dN^V5N0lzpV+J
z;#hPOO+}INPq_t9m{%&tyg1#na&TFw9HB1{b24vlH0MI<h1?5ed+H9V<q}Rz45ii+
zuySk>u46F(xPrB&!+KP}e=MDDg|Q+=u(n2jL}R7zPoV8Lns$@H`DlJpLqZ)3S+~na
zcI+(F`eKdpCiLjt-08=`W&f7I178yD<XvPWtXXUA_P|n%<iv$hy9E@YtYe#;62d`(
zM$l1!hO*sz-<XZr-PYn1j7B`M*D|FTpby){RkT-{w{Hn>W)r2Se*-$KbxuK+tK1y4
z-^p1P75Z`z!30Atn`l}-Tj^Y{)6^1O;IxTmeRI6Ym<9u_=eB_+%B%kaxYbg%TOwM^
z4C+pzwFS45xKbfSA5tOJ-&RQJxs~+XN_uW3vVA@5KUxVHP&dcg3h*dXp#*z-v8HKg
zpRL(<VNpe+6r{23f3g?SB-uw4efBEFMG8*`x)y%6FQmJu6mws5_RgZ0k-u`iMjqoe
zmBLf9SIH^Fpm2G@LEp;Pb&vCntF1ug@^rorV$@^_%Pk-K(rg2gAgdpR#2nyl*+dg;
z38B1!reNoA05;JyT*>nP@<~dWeKfL^y8=BVcg2O=rG4{9f3+8SystgX*R+EqCI$o)
zNi4XCDOkNLV0|@-t!wyeLfT<)0m>yZVv5pM2P?#~+?@ghdjj35i_2f8!_oA-JHI~f
zzHmKd5fH_1hr$$?W8s?=^7cmj(bLH3ymL&3Gs{kZy-`K3-Ed~LkE@+{`@|XiKdkxq
zM~Whx5mxq*D;h{9$E||Ohy}}SSKgXDk53rqeioyNUm@Mu7ihnObQcuv$kM+7^;%rA
zmm&Bt6qlirAQP7>)IlkK)mck#<Tebx=U4a-L={Eq1q1^m-pvBNHOQ^#so3pl58eNN
zDD|`@%brX!6QBo=$Fd}nAD^P-^v@)xZ+B0xcfUO&O;$J!R!y%zCg2y6e?jt~(US!b
z24Q;rIsN${r80ay@kLT1@@4tI<GViHBRHM`nlF9}aJshyVGYlJL6z2`No$a2shRJ<
z_s1~K>%BBG?BMxg8Ze*!dj0)QW@amA3WYX)GL{7IRXO`2r4F3Pz}e|uL7*K*G}EMb
zPx}d_ol}z(#Fs;j(=Vp59ew%6=r|anV^A2UdmdOQsaaC8i6TAo6`oBR*DS5}nH!$0
zZ6o2HE|fCDNwuMWX2Y&UQlc!a&9r153%tyy`#8s>vB(!+dagKO;OL~{f3-*wSG%j>
zhAaumjqak(RI&(wWL>0_=At}0^PQiA6*aO@niI#}DGKp6Xh~Utgd1p!$=GN%1*8hN
z%1`tZ36_z}QiT3`ppjQXq1*827XX6dqrqjB={c#iaB&8Ih3Zm93CVnA3Q?Bem|#13
zldF%+o!vfA-0VywfP8h<z#j()bZ%vqD@V<57MhEBM*6PdXkM>rpQ1~^!egW{uS`m3
z6XJLB`E*axgjJ8r42@fE>PWSdg?e%|op9K91TC0$k4rS*ubD$kj?p8mc7w^<F`^S}
zeIrk6#4DzM5+P1Z?>LL~0uA&`7q`fe;G)ctA_ve74F;6}PnAYiBgib^Q>GGB`<TJ8
zYg3IJfg2lchA_HLh4AD9W>6HT-Jz=h31zCQgdC`!7Wa=RJa>7=o8X7x_N{yR`g-^8
z1_JU{81JXQf8K=`1X>vDIWv<m(oMg<Pe1Pd=(o*(o-T#tHR(so6my!9mz@2*WGSUF
zTDMXfp&0<}Is22T0U_tuDA}B(RnUc)Xz<$|wGoKE-*Fp4GFBA$EI*O`X5?j64>*XG
zO1HC;<LbdnftLE&BS(|0PZ=`O*UZy-NYCPFys*k&fTv`g^Y(xf1VWK&GxG8V%v-7o
zm*f$DqrlvU!X4z;?$kzxfYK-tgkWl9423tqk=Qi>J{!lDYQV9n?D}eD&f|87r;^CT
zZNKYV@Lc1lGjt1%W#j-yMI1|5*dE8ZCa!b-%ms`fPGH>*OC@))A1ez&$&GFS#^_ls
zh~saslMrunC0vRl%oVYGx|k=FUXb7NM4fGaU9clK){WWhsKWLBT~P|VF(uGL=n%Q~
zP_!|;Ib0bcu1J@kNV*wWhu`9o$cGeN6CxWao2yEJIR67o<vqAg&?^K4PRAbUvjaj)
z^;!X;-Su+=LWko!0z$jPbt>9ZIpJPj(Ix;t5SJ))KqU2o?Zl-}rMay^UO6ridEq>N
zZg_fsT%w>psWv-)EG`kpbTAvE*a=J&V)}-K6qi00o7l6w)rYZ(im)5@_9|@FC7;T<
zoYcmd5IsEOm-rdIIcg`Bir+L5xg1k;He@n`6MgjO-S!<3h~{*;Ho~}PZkd%<Q7H&R
z(T3}T*3wPbm7>rL(nuON=28au7$>KH+y*#0t131kWYxFKJU-WU2HF#7dRX^L7N@@S
zsD{o6MbsbFloB53%t@`g$5Yu(+R5gOUJ0$>%4!|6cdliGAr)%Vg!^{ydyEbigT79w
zIR<`{+a~cX`E2UCj!^iA#SI$yPD(vAdk-aTDEmXxRfYsP)k{;FMKxKK#5I?HCf}0i
z<Ct1Ce#W>VqL8RJl-9l1$J{xQwXX}gyOEGtcQ0Hn4l);^>rUVwQg*3bkA#C?BVX{&
zoPSyMP<o9(Y>COr(K=n*uCy_g+CqQS-$7!;<#O_H%YOAtT3Ug$(qeg9OSd<&XBF{n
z>>AChf+Mse33u_W>vZay=cmJexv_LBR-NXr>wvj!;;u1B-T74HA#J)?iO&yX7O7;5
z%RMj?c8*=P%n`58J+-i^w!v3@gVKYm1vk}B;zsMJxJX>A9+!xvYxi!bnbfyyyrj-A
z8X3&Vore<Eub;A!^Ni*r!}U_Oa*pdwY@&OPH#O$0HDIouz?GGJ61Qi66ZpHiN(p?#
z*xm;)Cp2|@-71$Wwoy;^lXAI2FO<50w%0|{`%-&2kFpaUrgrU*QhR*cr1msIb>&%c
zd1m}5d{_jez4pTW!sBrffS?3O=;gje5#V4*X;cJg*B1c^!!t#Ij!B~_0ul~yEdqMo
zl_Fpi*1-pPsu7<p5Zoqz9l5g!EzYP3Ej=|msPdVp{h(D7THWz6txDhQ)Y}tUq1q>{
zj<d0ByVjW`AEi8ZSd6%)7&$W+`(k9ldr*wPj+U(#K(1=}sASpsVtlCKl)ej&JD=_@
zE?JV@+9WqTyg|!p?WQzw|F_Zv3<w40ilqsl7AW<#l2yTY#mTua(geT=Fa3>6lObb=
zrAf=ztx6MiKtukMDif!LS2xggE=<bz)q?`1zcVKP0Yt;lm(i<169YFnIhTRzJrkE(
zyBR2d-CIkK<hTvK_phiw&|1A8z@UIGPuBvuWze@Er(idyJ!Jp?A(TYQlx&ynOr;Ry
zP*Yv9L{bzVA1RN#{PQ9&zrNkR-hR0odkG>s74+rn_ltrD68=>Z2EDrkrGz)?^7Y5%
zZ{MWU9zHktSSgKxPk_H8e7ozrQ4wY+2OoHUs?_y8Se4hn9HZ-6<Lep&v-AYt3cf$`
zIIK_7%S<7`2c9W&{rlHn-Xsd!B-2JXpDsRF0klXm15!AkZex}tz=)l$@0q_-G~$4>
zy_GqIvd}I@eUZY#CmLrie*&-N=#O7PPf@Mv$vY8z9Hr5q8wdNapb6))EcI`j|Gv+E
z65!spPC+Y>KrW~QkxI>H>Q`NMsrfGFUuBNYTB!OdPaUqGNIUuDtfjw*ew&-;-P%*2
zW>fq1aF>~9o7v<l+#H=%d6p%cD(AvI_T2L>z@oCwe!082dJ<$8&I_%4JBdu`hB9XA
zrau9Zd_IKy_3QBK^Xu(j3$t@UivZ+*|MACLeqfaoJ|yrH5x`CU{O$7l+h0ejc*;^p
zSW^FSvf5fV@&jEmN!XzD-no8*UO3phH%5fmjKO|~O-0-k9)MPZc@^VbQUp~QlSYa~
zP5F>RJqUY8RoN^j{Zqv6Mt(3SjjDu?fwGqCE}f(>!H%=dLJk;!qm(%~e4<Q$`zqbl
zr^>RX6Z({J<Un4Ti^_}&*KXv;0iE|;PX>%xb5?l0*hAKmThreC`tD^87bXYr&(r6s
zO%N)Wk)N`Ieoj!dN?08h6u<ujMQ}+_{z%cSYEauGlJN@-p0{~O8^T)mOx<`bd`i#M
zjrSrsb5u7F)wEoCh~&QK(A{Q#-=mQSzNbTiD7rXxOh`v!mrh#Pg=pqL9x<!g4c8jV
ztk>)%c{xQKIpMBBee*wM*WCz@f)rsSWCArSTgsG*($N|9owkmyia{Wwtm6D1IPT`$
zt<<sIx=)z5lcSA8`l$;M8+tlR5rin8h1s4^qyZeH^u<MRR;Sx(mmpSu@us}jZ!pex
zrKMSW>}K2j86iRxDVa?n+`16%fe`Cx0kV*ZFl?NUKBt%@HHx_Q<Pl0G@M4Y73`Y_+
zjChiM1(_BH-mDK`974>TvN73YYQ-YSs?m<hZ$zU8(P*<VMDW`4#D$n|V_ztRO-X=9
zv}B4YJX>(CQi`wqlw>7;8f*%;0p0=4Xn+O1a36P1(A<Qpa`qegu&MGP$~}!sE#wMt
zw-J|W4Uxz0;L_o{V9*Wpf)UO7D2#T`KlU1S!xlc@sh65Ej23Q2ofgZSps65G$XDi3
z!;6|SCm%+U*JDvRbx4fXU=Yls1u;^&Q=aC~S7^9KwJ$%nj9No~Nf{x&^qcs36+bjq
z129O%foEIKbBaTHq1o#@)XOB{4JGlk>3teaVG~y$<+HLrA6MXNEOlgE1-@CT-W#j!
zcx}6;`h~q{h@q;PGu70Mf2*Nejr~#SxoM@h!0O~;NOU@MYTxN3yXrTv6&MR_Xw`Dt
z<<&@%%rh;LHGIr}-H(mV2dnX$i8nso&OORW+Hd^?GI}Ts``Z3lT@@<<+)1@KmmQ5t
zd#I~otrZD?`au0)e<8~jl5KvOt|f-A1cFEYM$}9MS;7n^%+<JWJHe-(=vMH-UQU0O
z1NNFB4TKsfydhwS6e&gBq#N+b6Z&LOlvyb4drqY<*!qZnQ_a}r`+E)4R@WY3nHTEs
z(u1@WHJ4A!J;r1GG}x~Js5D3c+L=>w=^4(x64=>h99}~@+8ugTNh4M7)M=W#UJBQQ
z(c1|>QzXq<PpQY}1X|SEzv7%aPV47X&ft*V6+mVe7F4<3$M|$Y=ZT?M)kV$L(JU3z
z9n-wy5sYGg_-4emU7REMcOPOeX5PK?(@EIy5<lxW&As-7{$eXi$m80PA00bOD7*uN
z$brL?&XRSjOQxad7RHR&GMeptOLaZw)aUSJB;7yLWd<gsf?3b+pSm)oW&DCu?>N<)
zmpDhu5<Q=130iq!(zw5B3FYRevxG8o$P%nRY6*pZ=E4%{C|QD~io!+a&6S5+i5zwQ
z=4rPX9NzbGwB43i((Y%5dl6Z&vrPv5le#a)wwx32Gp$4#*&|u90I5Xq9Kg?Eu9CMp
zVQ7T9{eY9yS@K5P8AD^~jNz<3S3-^m^sV&y*C$JoAABAGeOv`c;#ZEO?x_n2jeZ!@
z?OY6haUN8DPb0jo$$?4DFOa4gbj`H5A=dqIv%q-|&7$dLY9MkvWF3=;a)!lZz6*1n
zfkiQ_xpZkRHwM94c6C9PN`V_v?X<K~pv+}L9MUPfhTI96I$wCY7evgCQ3-x#1!9^B
zp{U2__u2wI13mLax*r+lSgGxpHMKp8I&mg{zvE4<x-9O}ddDHp_ROt_3;i`Z3O}-j
zRi`{VvK9$dp5%(!m*_}Xix~=Xd4RE)91k#7<<jK{x~>b221DI>M46jzF@Dvha{DZ`
zb7@f*+*I>=dtu586zaMPTrS<cosi`9b?2hG>keohMDQFw<{-#$7IWhHo|{(~ZRW&(
z@>vT2^W6hhj-EoqGH|d^6CR^%{?xJCjk&Xfoa3?3U;%a5SA+)6Xg4%G6IU4t)sh<o
z$f3eIJ%oxnawJsX%!8rAI>j?^<5IG7{Y5ROJ^bLzYPcXeWbEwMj<HvYM~t0G8n&~k
zvBE^hFGaW`=+lnCgU#5Q8_Lor6etvbIO1U!-FL$6A}jo_7Fk+F;ms?;0%vs4?GZg_
zycpHBEGz3Aj@K)uG8E1M3ugs@i$SgeEW#KqbT}F<>dZ%tX{6;CLbUU34&ydq8m_-M
zqMZ72PS?i&U~_tG;M2$siq=7~T7${5AG21Yg!ICPz7^e4{SgGNJ{Mj!b-qS_aLM;_
zaR|0-Sv5_u(;0wCU>CfLSkxM>rpaOY)^X>#slTM$@WuQ-SJE%=#g0)+%jMcv&o7%c
z-}}!Sn1tSYhW~H<%k?)kI}D7sYu!Cm=-)hLZpac$D+u>Pxe87w|BC9b=k(Bko{8@?
z?9;}Pd$FeEM{7!lZy?IrLw&P<wAQYu>sxY9N18*pX3wfp*z2iPIs<%ZcbN4lznk2$
z!{k=&@al7F+31-mZ-kbqyY)V4u{H?nUQv|voE}^3sqSxqA7>5OCJ<`ee#IWnjo8Ea
zqu8UYJz~g}<^nM|ZpcV&XaJ)$D-B?rztL?hSfXC|vVwn*qwSV*souhWD23OC*{_-^
z7@SRszo_uq(5#XO_9ceiae8d%r@D64&`;oNYejH=6sSb&XvSz=Mh->mXdjg0VJOE<
z?B0B-Yc26R+PTBo8=X6txY4=8b&t?Rk?2sNn$irPS4~_75#k(;S9*NB3|%j~IrybM
zp4#o>g+(7Pp6=sqH3|BE|7)|A%hZ3%ag!GEC0(<QQO!5)|A=9}AqbA|c908q?|+0-
zH;_Hk>Y9<i`4@T9lU}}3-v*aS<iv*od`x?<Z_TvQb$%6#UjK@VOK-|ke&xlk7{_bI
z%Aa1;zA<B^w;N--7AI46oPXl)KCWm9Up8L{PEiUO=45pKn#Ws4Ve~7kvSZX2f$^s-
zklu#q89b*jJn&lJmR+gP|A)Rf_cdoQj2sN_$kZU(hoG_3+Csa1R6w>AvFsPn_vhE!
zzX2`g*VC8bO)(P!Ff*4CbuJW_W*jlXe>wd1`uh_nVe_>}2%0&oCSw&tnk3<g>;5Sw
zwR3XfOyS3YhUpjHnUB7FM_o9vs0*c;a(3VpWsmD39jNgALhIsmwVgDp+V+xJ?M4pb
z>)~0spcAR`H-(zh_)(~1jS?u-s6vb2*E(kDcbJP(X_v<Xm0*)FQ=zQtNHYROf0S3-
zL{0NNQ=2pzXF>WXmuY2*;)o#$TzX2m!YxuR^W8z1K>0+}E?I_ylwTU^)<;Q;Bzc7=
zqR5FC&y>R&(@2#HspjXAhw{-LGs_yuIO2iy9_h%W$4DcfQ`o!`k+Gh&6;_%??zK*(
z8?+&^HeB3KEs|!tKv^aoJYpAgf1J{eCgg)7k8n<7VTatYD`fCJX2vn>3W%NsTKy||
zC9c!VEnFw+Tev1}glqcc7wnmMb#g{5HrCZdJ5n3=0L@;(HBnEvZr3GgS2^N+Cdov)
zAAGa&y4B#BP+r$%mUjx`4WuhOCGn1XL+9Nh@@}abYth~Bp@In@DrK)Ff9sT|jzrxq
z?>^hD{qP+1<{Mot(Cn;WNnTBIi1wLwmkT%%vJ!@hZy;_FD<-WoTI<q!h;93HI;t9T
z)tj^kD;n3uuA4qWa;Bw!gpcSrB^$eRjeY^;h4X#BflDTZ963eb5Uqi9A_jVdT;(6C
zo!0zA@%5|zVZn+uE|igMe@okr#J!w;ak4M3((3a;#7(+qY_6GoNIYm6rBD!(yRXUQ
z(ljB#-<>E)VHo&fb;;^hjS<1>Yb_q1-L!b3wPo>0qgXu6Uy2B6p2~A}{Xv?q8!u%W
zua;=!Rjclq%7bUGJI|a*(o4NVYgqFB+5@u)uI&9y>yvHHf6nXqe^DC|P<}7=fvl}#
z-s%S+x#v{!`{=FsyX5x+ht^xoGRcQ}ss&4ll9i_@$V)uzd{pnziiEE5cXrH>J6ZTB
zN90=L{)Og|961lzLyb{PSn~+fbes}mELF)aq315-SmdHdE&(}G>H#@Ynk$i$W^N&e
zXi-*FspkwT^D5uwe;y=1k&#rX<~*k6XSLR`d1cv`q+dCIZ6qNaR=G$IL9c68Q%W-L
z#i-9^3x#o_wrm|`!{CUvqSuW6&g+@7KGoEK{yJ1E&%i#S*U@u|RbH&{0=kb$U$OtA
zDdZoir{kz1G*@PEhI#NKx9AtS8T}%^I)meT=q6fQelgN0fAKud*T<aQ7alpXt0C9t
zUA4hG)Qcv{i@Lw)8R`_MZ%xcT7Jg87OT_=NuNmW-t8{tw$H9l%)%s2Y`Wsn0AsaQ;
za(%gnU@4iAkH|}VPAQtZrWT<NWbA0Fgzo4C=w3!P=$jJLf$Qgq@V8c!y+>~9c`@9b
z`SOL`Koxb<f5mWhy;<f^wKvk`yo1n9xyV%+W9VH=^$3i0oJvPK7CNd%6luw0^1fp%
zFSmKgp4;cDbr&*7-gi+O`{r&_(+?s9shUZ|KIyil71ksCA(rkr?OR$l??*~E^ro)o
z!11$sk}f8uks(G!yPg{y*cPF+&$f<(u3rK-#~g)Ke;1})k3dPssiMRVamDp-&*W{j
zz`*4bRgrs@VIN`0zY`^B`v|3fdTjEm%f6)zVD|Ees;}p?UiG~#n<$p<xx}i^`LXJA
z{@SW9&fKi}60HqO6OCMZ72d(p^lLqO;|w=egz1^RiqN^XA}lhLijYPq4*J5W)syGf
z)MiPZe~41*^ot_LQpT*bV9J&^<Xcr8LuTL*WuT2Scr~8Ogu%r-TR4=!wDFUYo(bM{
z<VrrL$%1acoab9_{1pclWXQ;NZ7-;KCz>&0lP`<wPH(Wb>hH`2oMyFcAG*alzaW61
zusay9Z7DR@!1s7DO$sG+>;j+bjwp8XCMYjlf4&dDzP&#E_cR&>u$*<;PJjP=3OAI{
z%ve8RY2byLe*H83c=}_cI-GLg?GcTrN^XqSJp>HX+D!-<CIH@UFjRVM9@AnKpX#3o
z&TizTal0M?i*9$2Kzxm^YSB@9$DD(i%qSDb^w9A1;`L@pB&yioAcvF+DfmS>&Pzt$
ze{<N`@@Ql98wJ;@1hiw*5N({U)?MqffN!Z`-iKnH<-9+rN#U8-bLo#wp*R$5kY8G>
zgs2ec@_`P0Wyy>QE*Lp-jR|E5TuFb)5vL-eYhSFb2*5CuZKN$k!TFk0pe*xVk5AUa
z8{2#hx|jKyBHwNQg!_nDVdhVMSm*$bf3m`Rh7M`b?+M*`q7Lin-C+MpjFKf+4(IHJ
z4rwspiv8H}7S5Zoqi5U%59tpY+HRm(uTgG|C79?lk9K`bERi4)zwmYordXo=n`6l!
z!4<J&cw*CQ`P>Mlv93`ytmB?I5M(o=>|#v&m0CDx;<|$-@r7;j@1{SnlMaM6e=%G1
z2Md~Hxk;eq(CBMgas4rv<=3>cC{<qYE*@IDQD(Wj-@!_TJ9^Zm_B(#mFl)w#X!p&`
z<OGJG3Z2!;Wjq7eoY82i5K;uyR`ulISo7pACvht0uaeT%WDW0_glXT%D7Vz_OOE{A
zk^`YV;~qx-<-CDOm9AcJ6=8e<e=t`i2f#;;{PKPY4=4=sa^&#dYj6g+eB?5`z}PTf
zhN+(B15ATC8~WXkTo)J|#9JfeY?A}J+DiI>H7bqqb-w9nYyNh@P`Ty(mBCEKb#CQU
zk2~Z;&*UK9lop^|Ak~-^SivlaoveWI5PjnxO+mREd7(@p=6b7{g2224e|*XH6lGzh
zX&aj2W5atT_rMwqz>gFcULV|?Q=oef8RI}%@|AodWgx?Bn8}fH93K8f;Ty@cuhLz8
za$u^y0>46^nCV+2FVHmyMv-eb@^XdFd#(~{`+Ye_TG5<hM$Xm{Gr=Y8l}cZqZm;yg
zGGSrlB`!0&-b7Vl1q^0JMpdPHLcnqt0X?DdGcCR|G7!XC`*+-EJ;68U_@^dEMt@%a
zq^h(XY4;?p$|6|HRzC$mV7BEfr*E_W{<*5cL09=dK*F$>G4l!&12`}>m*HzC6PM7B
zG%1%aCNYzL*C|li|M~gbPG)AM&`zQoq(E();k)rt4Zry9{G0FJSULx@(kT?$_`z5b
z%ukU{Gk2OfWpYw8hs1xAn{e^v?B$xi3^}27A;=<0qD<_Rey8yJIS@`Ni5o)R+~&2E
z)`#a|9c2}2*h98LT;?))o0q30f!{^R3!f+jXsUL9{YX-?Nuyl>9%Vi<dddGr=I5PK
zSrF1u$Nb?OoCj>sx83veuyG)WX%k39rSK*gG2Fg&e|`Pj{k!X(Lwgd$s`~5aF1#Sn
z!dO2jB|OmZ>+j*m?)RQ*@zw4g11%av8WY^3bw$Gt;|A7>;COky=R`x?pnNHaIjcWn
zd`+Hza9mY`XcFLi-hp|9k4f0n4~!9BWB-N9(p^ZB|6bR|oOx@wF2Gx{G4o?b06@RD
z!H=%&wHRDFJk&kaAXS7C%2W#Vp!6$t)@D5-Pz&!acBJjzks~LrPzA)PxX5{=Lx=J<
zuU*pwA_>cn<}?wcva3=dgJ|%Q?&}=L7d9k+FVN%YOZzBMz#OuTyoXj+I_5W~O117|
zMKjrs94why(p9D6@@_Uu+IqE0SJJvGepWf4yk_KhjH6q@ix7na)q@vE5=Oc!=p|$b
zbsc`li~y4--!r1cRJe#}T3KPWuRiG;?U;G!sZqOtm4<{P+f$w<ubV*(9K@8^Q3aTP
zMKEwiuwHxNsByRrVJ(kSS0=WaWfs5)SA#_scPYVmyJ$sWEnWCnglDiAYN+Y@P@j<-
zL8HoKlFFw%@`LldY*I5Xs!F#7(r=K|T44bCUDjmHhMa!zoV*)z?)IW*QD?*@h;tR%
zB5_Kdk&?sZmpS;zq428Kck(n(-*Pd3WAv;|RKu~q+aBExO0L<WTYP&^%?3@d1#9We
zNcDWbLv|lU*t5O>9w@ya?=JNIK!K1XgBkBwvbj-=;buw6TS|xUY+l(dKQWZt2=+RA
zl)7byiAScMqB`Y7_$Qk+<OGGivX;!s?klbBSC7?BFh(`Vmh`%qTZ<7jcS1CO)>V4k
zsWh)!b83eCp5iqqctC~%QN$}<O*1vc*skrrr>-YyVy9SW>#P^9yhz&4?L!4d(zTVL
zE9*u*?B61&&)4W>^&z`(q_bKfE9BaitJaRwSi`6^909m5Je$7ne%Ra|T#3=Cv#*x-
z#BID$7oykLFU6r7#ShaY<MBOzO_1W764bX(g$f^<9TAP{`DmPAuH8V+$)Mb&!S-*8
zbX&%o>~wN;?ns+fi{I)XQVB4o-!GQYDmi4@uXS2QrnEjMz+<i0_6ph%Gi9}In!V^s
zhoaro;$(U>-%h2sm-}yMY76h%114C>F<dO_|IvUMGmPIb0a(W$bZk6-i*1t4*jBnm
zTIx`zI27EcKD_~5Kndc`x#Q!SvL29Jc@8*ZpC@_SabGVN>h}ePu#tosKnW9#(;i4Q
z>HGzgWelB{@=u&BBYJ^1G%7{4{_3!%?w~l!l6hz*Jh$sYsIbFW&H~e2Up35Cj_Y)7
z|EFQD@*p(geYWE3Yszzf@|;df;yu;+rLZ>gfvG1eLg>*s6|u^-eV@jcuBCf<8>^^O
z`Kz0{lr>EOpHokoxk1z3oC_UOQ0+LToJ_||)7B(u7R1nYQ?g0rgN}z$*w6kF4Vh_F
z=?bqYa9a&TWYTrkXPQasb}hNAm1Co+Qbe6vysKKW6F!!-GhecQaIWR9VLC<#J5DB|
zZIm@rQ59?cO*7J}>scO@6Kafzv5ITc!g)l>I`~xDITac-z8q>yxP0++4U@So(-!7f
zm}$DVoHbmDv#=32Y192)m~0fV5}bKrvH^xm^nNRVg=5^fV;OlJCY{KZ<O7EBRN^Da
zXA>6-7Lz_6b5FT{cvUh28BQ!rpNxo022@W^Mu3d=(@JCp;+vVx-73zR>-K4(!jOl1
zR@crE2?QzIeel>ZV*0OyJmd(<s5$nAxe;<vW3S5*V!Mn?luk4!Pgdq97-m3RTy(O#
zluoEAfbThC$pfa~Jk9@oyl7hA?Q(ru-+h43n?^wlLiLh=X_NzD%y`H=X>l$~m^^5@
z;n>**;OMt`q?cn|fL1-hLwjIp=L(?Iu${{BXgcDo#lz)jzB%|@n^)-mn0oc-nPHUJ
zgFEIsu&Q&-xxv@*s`lXh`jm?!$x_e93#$O`IB`7j*-!-Zcg|J+0WQh1Tnc4wWOH<K
zWnpa!Wo~4bz)}VT12`}@mqG9>6PHgdF~I@Mw-GKeA_RYIBjd-`%IKP^6xZu%ZT8I3
z+2wp(u8{5b8Od#1HJ7>PoV$_6>IxDSG44;K>~|zTf=i`0D@d^gO7AQnN!T{aSC_!V
zxu$_8PP(+Jm3A`76qd&t!UnI(d}>nc-*PQ-gv~i+u$N;L1as?M<o3@xGj-bO83D#?
zUXz1P4#9sNY18UTu?C3@r)sL_7oj6>^CW)bpv7pTwBZfv1%{1!nAxb$@{UpST&$b+
z2Qs~JAddDT(&&``mZZsVZciGLBoOto{qIehgwFTxTh%}@S~%5ijMP&NyJFV<%p-(Q
zeYyi>fZKUxdpT|y?A~3wuvzN$<Fucg_^kF!b$)+=BQ?^7M!*tw|1KvWJbvo&4TfIG
zy9b`%AcF9&UDtoO#oTlXgOYXec=HtIUMUs=O8WTkf>w&pI!Xb<H9`HZvu;*Z7}-Wg
zwvevYe9U6vj)Mo_wvedy^Qvi*YP~7R6^@MKe>%pMD!)hN*L>ex-`lu6iks%%v@)$i
zB6oit5_GRoyN`PV-Y9ykp}>_E2Uv*^mn>k?!9zI_E!BVMbn2VcGTqK^u@`0#ejVGb
zj*jcQnjVHX-NzVe2_{W1-rm=)-W)+X;k0wQg^hsR20OR&00op63<J&NsRPiOK9FXN
zpWbR3nE0e=>f$NexV?>G2XaT)84v_+R^fkUeYGBX&T<2!9;tag21?L?+Bpm>N&Yyc
zRVsXwrz=H|hq*ACUeB8+rt@~V1y2N4D42$k9D$J;9g~%&b;0K&xJv))kagseo8dI`
zJkU(^XK2nX9Dw^Vt<;`#%#h;fRpFYC))<3U=IarxmUGZFL%s~lwRjMh_U3uimo<O#
zrSdD#F}7T!GPdvd#H2R%>8-Lv;X97e(uGx7mNrh(u3gD7(=Pam=6TOhXL!vNTkG$M
z=NeHIHn7aZQ;%8DX^8@$Ctff$nSJKmZz=qB_=B0Bgq1wsn56}1JCVZ7hpgnz*AA(*
zr_Rt!-AIX5{+5uI0UK+582iLN-Ej^hMtVn^-e0+!{2zDzxtAeFDiZ`ZIW{qu5xg}d
z3p6iGWo~D5Xdp5<m0=1emzT6U4u3&B9L|Mg5g16~I!)oAhMg8|1*6bql_+(z>S|@U
z|GwXF_pBuEDwd^_MtQI`=PWP7;mkMRkcX*CNTD>fAUPN-2fV>^36$}!H3(MyO%0iP
z(-yL37bxV$5UikcgHSjq>yy%?25p2lg4ekZm_k(_$8u=RyBlK)_C_XwM1Oq%Yd%}?
zp;Dld6)KTJvi^<&)zqFCC-klAeM^P}Eu>YnYwUBGK`I6bKV%}GtwH@vT>%Z5&D#}7
z()!32a>?GVl#r_$)=<!+)xNA2n6l4SQ{b%Lw^CrPK1a=gwU1mmVBjuE;8J@J4cy4B
zgvOGj_kg{qRLoIp5T@R)1%DB7`!)@trAz2x$@@Y{F;SN?Bn(g!L#mlw!YHk8KubX>
z_4QKF%<`lr@=bwBAW2YPFC-0Wxw>RQi;+tuXl>$B4VH56(SnhBkDLNJG?yrtWbXlk
znJQxrs9{#UFE-m+>c2wW#Qhx>ovn`oV@pad!A!dP6^dK`)i<y9zJH9qm3=lw-&zTC
zux}J5Qj+yH;+;26-q)2RaclSw*jE1)6EXQ5aFwd!>xKK(BK;kt*5`oJH@Fz)z}7ZS
z{hfxUeLOfOyuz(vDZm5#R~W3h|3N}u_dkO-z_Ho2f^#)@Bx7t8CAT7mF&E8Of-s>(
z-CS@U>{oM(xYuIgeSZM&ru{8^k5Z|9`U=q<4<g=@yE7UfP>B0_Jw8Z^j|<l(Du|WO
zKOdg$Bm6SL{MGRHfBe&jixoz)La_;jv&*+{-#mT#Gc8puTGG3;lpGaC{CG<QMRc^e
zL(5Vkr@QD0$A}pBE-hIiUsp2>U(Mgn7q8C`j%I<qeHdOIEPt1?#aYO!$3MPbe*JoR
zu$%>2y?8!9TVjgm7>H8G(O1KBY#M2*s|V~AW#QEW&<}go{Nd@-;pJj}^m?}3A6`Ct
zK72QOzg#Q)-G9$#nECq|6gpeZ&XyPM>#Olz%`WDbi=)}a3gi6Mx3lBZgD>ar_Y)qi
zBVB%j{tNW<z<*hF{Nn6v-UlkYn*Pat$7o+h$LN)yX)RbD1s;XIyf+&S`-sEq%fqEV
zy*NGlcNo5$FOFx6u1R_`{Au`R_-db453goN%YCvO4G`jv2h0W@Tf{~Ls}z&ao8gOq
z;a?A5&%c`w&%$T0oB$_aG)*vY2B7}_>C<awx{nfo?|&$O#8C->5{78GO$lP4Swl3M
z<x(1F%K7112)sXg!d+{jlA<QNT?@=DTBWgNf)-UO@6clR#PG;K;~r!LD7-+-V(~1h
zIW}Oi6plH`n<qCGGJLTa{%CnRKZ8bpef5q1`|OA1^8Dft!|?s-@`uaAcr^cM`0?_m
z^S$HQyMN)&m*2jOKfWCf-_8$*lap%g=p@&ANOVGcEPNt7E3<NVT#hB5q!Vc;jfRWG
z(Qq_ByI3v`PT}?#;CC-b#rE*ORjl^O{g=zcQDEP3xlAp#RDSK{UL5yTW=VVauRLsx
zWGNy{Hftnv)NYStjpS>j>uYpqMxwidW&<=kQ-9b3@EfgG04~ZAH0>@exy72~4lOw$
z2!YZcZ<(w0aQ%2otT^U7HDnLK_DiIs+p^G&Wc8^jsmWGKYWY4Xsnx5L1UszKP1y)G
z3S|_=v{lBmRmZedX%xo%Rmc3bvc-yzuZ3~sI*we&k!zbG*-0#Ah42CsV{fA~%i9p1
z8Gm_LN@kEf(Eq~O9t7~t#OMm&L4G4jsmB5ab?*hsF1Y*DJeu6P;^kEA1(s;%w2}*~
zP`2|ZOiNsn2<OTfOkF476eJQsqF5_T&R$(cRF0?xp_Jqu$-JuXaD|o?uFgfvbV@h_
zN8t#Tu*!UJ0IJ)TP(y)LsQ?}Ya0ZIirhn4`oT_3MP>#ofYJg!77<}6hC}Et>phz>s
z1WGHD>6D-ZVD&=_ln<^`0?*sw5vb7=t|JIg6vE(`sZORtM(}5p;~<UEz@@VXKkYDt
zW&<Sz$twcBE8yMJ=^&s!fac5uhoF#=IXP|X01}KET>(FE2jx1U6-XmS#mow86o2r3
zg^TX6f)!H(z-wzrUIm4{PNoFa8dJI23Tnu^bsL8fl!CIHMU&J<2#@Kuf=Z4|r58R`
zkfJMMGuyTy5SQdf5MHDqfizSf(<y;J2JV80kK$|*Z)t}PP}Z;wh?#`gK-wy4IvofG
z07zxPHSo`m8c~hdF1WVDMB3mCo_`ntuahZJ8-US9d}uYcwyi-$XjW9lY{Do(ftS;{
zq#<_|+n}UlMcvk*kO9Xt10x%=<Mq`(IG1d^;gX2%9qjp9gUvBb1{pEFOH1t3%`qP{
zEv<?916tb82tS#Y_A}KDEo~#gY^(M3Gu5>qqoAW;qfkbnjzU`tIgP?NV1Le|FpiwZ
zk@Glm9!JjO$ax$&Pr+E%Ub#M!7~RrpLS(dEv00Hbb3?vL02Hz5@WK2K>jtb1AO)=0
z(xD++<c7SJ6?koacsyZrmI%}XabFc?21+U0X0V3B-H<Mt5&4-v<>^#N896GC5M^$S
zw&R@4K=MQgg^*OuP<FbV0DmPyiYgm#&<cXhyJ%GNM4{Rgn$!krIh`F8NXdSp0gwhi
zf>Ll%t~*>Z8!(y>l?WfLfcNSAK?OBG;tUK17O;LOj-qW#39#u_h}<0joYieIoetDc
zRO;vu4&z#7yOvC#aa`apwPAWtj_J5PXi@CfWuqAgb#C1bO{Ufq`hSvv<)~kDOt%%%
zQY6=54G+?qSAW~d8%etW|G{RVHdqu-7Z_4>RcEbz5|OvF&O*+{7MW(w@Cm=Qo6a2o
zxuEn{aRJI4LEN|F!kH7m1auOZ-LPd3J39sv`rYK_?3i7fI?dU_w+G7~{&K$j1{r;E
z0*bLYXL#qOR9qb+fPZc*P7g1av-K%)E4=q2r(VK7^hg0Q8~xtHkBEV$0SOLF2+s&M
z5#AMS85=Ah^x(>7K%fU%ml0ro^?r3!jlwnm>POZ8xjZ=j*X;fIVsn7g??E1FZBmqF
z{Tl;puUG5rYy-~@@w5Y7TdOFOf$bn1NJhZ;hldl$^~#|~C4Z^_!$3U0pC;X|1fd=R
zt{}%@#{mK;WhMaF3{>+otWgUPbYy?MT?_DKk;+-%9_k+tOmK?ChRO^KeN@8F+5K4d
zb|o4q0G9imlu40mdH>md|4QMbC;T6@TA$mKMw2_W1Wl3;57~FN^jI$CE@Q&SV)|un
zh;P(AH?E8CeM4N6)~$cKN9{g!&*})?>z@Ao#Hzf|u}rbCOwo-n)+;*JE2=wck-Ouc
z72w|0!L=2@?yrAv{qA6KdT{u5op2hby%$Z>{Vwb@YxECt#48J@P?AvH_#rmORPTZG
zzP@-{@_YwWi98&fpDt!6Gwk}&Z1~;Z#^vnM120znBHM2_@W+2YNJ`Gk^KWYjz*+zH
zYm;!nqH^ZP<#C|!NV!yL3X0?4u0=uyk%JZR7iEPMBwxiYZr2mw`ziUQTuP|QvniJc
zQ5?SuEB;~A+TxiEPV~qOo(t@~`;zJtzzh*iS6t4pZYqy!p}A&ox!SH@%=gdfe(qH9
zMuW`<WnPYa_?ds$?v{f36$I<!Em;99?M^MZq9|;4Xz54){NYDxyIWR4K<NLI1$=7a
zFS3>ROTJIyU)`Cn62FZnjW(V%mhtG)#v?}?-;|Lh6naqNS44;CwJU2>*iGe;MT@9^
zs?#2p`ThG5|7di~{J9wbSbtM#GZz6*^VFYug@7VBmiB+!T7tmrjjr+#ik;l@BTGcJ
zh-dz$6J&n>8o<Bor&zUta9SksZ;vSMI9(@gOyXC<iM_^^GQ~Nfe~;CX(7(Ww08W8D
z@GC(?f7@2#V%CiwUBw5Hjtb7hS6|2ZbuhQb_PwJK?Y*#^ZYJZxKUYKl54$AcOA2Lf
zWOH<KWnq__Um6JkIF}L6E)<tr#u+M?Ye6xW0o%6&LNP1>fBvF~!4#AcPpg(uG!v;M
z%JxH%L6ji=8cs%rEy^2~%T8jh87Tj%2ta|A1o+Ph0IWd_z->b=sEG;c{S6LJ)(H>O
zcywk>)-O~PzlQ2mlt-FbQ+AnZ-s~!kC@qD$wsV~<H*z#ptw<hkB)%roUwl1b_Gp}_
z#V&TEOETS`e>AOgSfLlJ%mOR6g_`U~A9jzdZ4Fz{cD@n}7bujuF({?gy2H?xZ&;~@
z%EPvG`x4Y@XsMyG3xJe#(E_N9K+#JKphecDL^g-wOTucC<~d=zu(z0~(N_$UG-a<X
z?CxiZ#ZR9O;tr>**m#A$1*ceK)nFdcOkRK3N{#2ee;z4T892H<cu)q<xV_S{B>!QZ
zE&J%pw-O3c^xX^v1XfvY@tlfNrcLehMiB;M-Uu#im5IiDOfkWPMTLoSfr_G>^-mK2
z=!=Ly*J6-j{6HD?kz8$dkgVotoNX3z)F3Cu@K$6x59v8PO(5mO+dqz{<X$J_!AuYt
zC9X$bf4-f^eBg>^4xA|~7tX}T=DI+Rk_!#gB1*i5o`fc=q0~3#?P9&g;0n%+{D200
zE}%%D2;G9>>kSmeB?bJEBCjHdQp*Hm7gnIVWGV?~{*!C5{!4H7)utp(IL9+@dGxG@
z6fpI1ddzy+s_gl#jBVOSJ{zXBGB4Hx3k)yVe}`019&j=a<=xHz*=q7RmGte@twafS
z3+lW={SJO^qNc-V`}r}~=;!TpQj&VoGZ%gCn?<Zzt2V;ijZ{D4?|r=F0^N*b{c-}w
z=2rubxyboQzL7Fc5%EZ2yO_Pr2hpq0s*7k?$L6~Xht>3w`N@dXHXf<`O6e%MU#X2^
ze>IsZ9+#`ithRY86y_sjK4)xiHS(=ny{F0RIB4!zDhw2nE-wwN97oA5C~cGm&cp-n
zIQxp!1x^l!03)RvQWtouViX$)^2r)!oPEVcf|v8WSx~aqXH21F!g41YF(}?_(n2NC
zZSGi=NeQoj_>QJKQsk>mcc65}bcfkDf0*t-2{vlNQhN`x^%YfovDuMtI9ZF1IN#YW
ztWk*Wle751^-fuQpn5)wUlDF=)*gAiu1m(GV6oj+JY2xmqOxHLoZfU7bganYLcx4f
z5rB*e?TIR&LaX0uo|gb#ehmd_9iM;ssvOJ#cw9&Qx{ibUFo*De)FQhC$XJf7e_v_b
z|GpZ*i^}+p_tg+kPGf#9TI4y*SAw;-Mp$}oex5_v_1i+ePD(vIo4;f1#~l)2__2l<
zI8tW5j$ga>zD>X^%tFRaz>TX-IopRogMi)bIj%&!WQ4`ouDwVH<Ha~WtCa>sLL=PY
z7&7k1SYH$}QBsn~JO~J{i<A7ke?ow>G_Ek6<ELAOdJRj<OI6eB(G@SVRo7`tIThW|
zR}hny1-3^#-d~r7+kV}kiksJlTIm`r8gBRM8bF98mK|mNg;1VrG=v~t2%|vl#;DvN
zwM+5SF*O(|&G2(<esf3YL>4>9BBuRlh=+cYq0~RAU8LvvA}k?ZW~PR3f1cTereryX
zQWm<dX`Hw;+o62zme(gpu~949%lzv~;w}+ErD?~e@xChuZCl&7#*gJ_yG{AK>&Em~
zdwHi^yGR`dwZ~LwbR8Mct6dj49auY;SBC<{%g*R8V~4%oKsGUWbkMxnU)a5tBbPzC
zpcok6l5+k76-F{^3T19&m*q_{5SOB0KNFWnNHM`I@HgeQ|Ga$J$PD&z^zs~1vM6}%
zij*<H$Scbvy+}rAB>RLu?uJr7m^2TPw=_vHsRRRNsBo82=_wNdIG4d>C=>!QHJ8C;
zHz}7hPBE5$`fxw|^ZIosGuVfdmupCkCBb7?q``_yNbw{k;mD#-sN*b5dXGA_Ll><+
zoE)rL(^?oB{PHZN4t|iyF8~`LIL{195?)e+;OP`R2sU%1*_LKII{eKzi1M?I%tXHm
zZ-Q&X?WKA8^1AzfSEr0hXyZErPLRj$-=Dkifq?CQTi+oH57hnoulv6Hs~t6S*WRP>
zqJyR3c}yIPMriFigCs0l-G)vnh=;HL{rb^0Z@#?@^Yv0FODzI;4%bAILk}mMQ*DnR
zbYfhZvG6}heIg-}Ve@k)N{xxKNlbyfMCKu^BZ|YF<F2ldg|(as)XoX3hZPbvQB+gF
zwTdBs6Rqae9pI9$Podp$gbLXN28Z(Hlvc^nv$x+~me-fJ!hmV9exNHaI1jxSim^6;
zA{fYn%etr*r&TDh>;hvp+oTo<_aw-g)*7pKugDd59y77=O>h<k1W(zKIy*JD)Uvi5
zZ7)$Ar{o376p;-r{a7B6h|)iL4eMx`pk%*)zSY@A_A=Gpj2KK9<(TnhvUS-0C<@t^
z8B~k2*=iNPBNR#NJo|yoy#{(rUJoP6fK8j?DVwi;F2Cx@;Fo^dGqN_*gm%1xR$=+<
zMoN<lLl5u{dy>el<grX58<&G&*_T#2Q&lLMLM^S58|%}TLJ?n2<95-_7@kxn!k!_2
zB21SM$8MjDug|W|iH$zZEzaDu4z2d{tYgblZao`oEWhS$_2s%4LMxbgsS0x@)UpVo
zC(hDMM)(l(65qkc!$K>&abYg<>L&ZG%U(HVixV%EmcueRKv}eLK58!5V5{Z?JQ3x1
zpJv56!*}B)rd|B(m$jX|%ojaJd8+e&H6xGd8egu-^t4n0Tpe+7qwMo$+2@?IPaAA(
z<I*OZ{(e)&M~;KLbE_HOyO6}lbSe1*Vgg(PW|sUN_VBoU0!YNMnKf{q1dz1iarMS5
zUzZ4`w5~z2k_aMeMRAJU^a!V)$;vAk%=2;vQxgpH&a98X@_9jhimb1}x|QF5E~R*m
zQCihnzcIyQTtWhFzB?_Q23cvUH5xO$6iCLoPohs|LQ60if`-jsav06+OEO?<yAo9W
zcwDN&a^pbyJcO)D*R-D>g-<g}Qa&myDTe|OHVjLQK;Y%5bV~0d*D6;=a>}pln<9}?
zd?&go61~OxlG3CAYa8o#s5>x!OlDQzDeNf8>6yh1XuK-aIKygejjxs*NEn>K4Rtlj
z=iPX$Iwjw)pNTi9E$~S;bodNGW(su+k6nr{Rtrn5;TyvXa%v$*b+PRiIyld8XO>oD
zN<Kc#9a;PMX&k60^b`a7am3+s@+dZ}9fq@Yl_}}ZY&P3R4pcgBgxf`b4wPP?9}07q
zDnJXV1D4c5=fZrS$LQ11!sMaso6kI<6zb)AmHS;_Bd|H%1+WXtOLHW*E!aDqiGUZ0
z)%PljgW)s}-S$FB0$KL`HL(m;nEVXv+Odg15ZboK9}8xH#M$)ch-MAeJBG8y@{REf
zWXOE<{UyCPJr<)S+X4%J;SZs$AR@qp@RwUGgugUm34i%1%Wd=xRzhJv)IU@8wDvTk
z-%R463e*vYYGJ7+hGMsJz81UB%yW%MA|(v(fQ`S@LUsDQ5LX$|5``paEo85)VbdH~
zy|o`FvM@#x+I1r1)Lj7Y@eQzrk}>D{FytGRmLe0FiE=1ww*=IGuNzDDI?+`WI2K&{
zThPr504Q`!x^}nfnwNy&fu_#K3B;S!Ov(w#nw*UL1}zgA-fmTc=Z`B{8^+^d3Tr{f
z4h0TX(N@Syhl|-ygdck3XnzeF0=b=s^@Sx?5``h<yhXepb$(&O6khXtOnc`<_&21v
zR&)wwZe(+Ga%Ewc%Xte40yr_3L4qI@mtIscHh(ILFDbI*apIBowvlhEJ*|?Qa@hR;
z!xm8BZrbhGPLH!A4TGC*svfTj9-xPBgC1TtFUQUE!O-AUu;6hxejR{cX#NkH8=W0I
zfN}<g<M-jO4_c$+^1_9*HinDm-+^npeKHXE3$R@HEx`846Qn-;jlt5_xYEb)SGeaI
zxPRWIcJ3dw)5${Q!tY?W{rmXGMhn>IL)4d$I!}siH>A~iEC@4VSK2XoKcx2T;rS5Y
zqGfB9qZ0=owYF)@x^Mt+qqhad11NN%^R_!AX21*g7Pn7{Q_H?=dS)((<k@G|GiNHg
zL|dv}yQfdb&A*%G6pgYj4i17c%)&q4H-G5`0b}rSKvWTE`0-`<y7{xWDsk!SoK%@m
z?RYGcA)-^(@(_`f<NEecX+><R{rmB4=vqF%p2p^)0*$>l_?XHhwOSAFl>_K|q=Y5L
zWf=?qY1$XUGx1)Yv9y_SP<%<06etN^iXCNh=i<~deIY|6ST+Pj(KPyU{A-p}aDRPv
zkjq|WmL|9q+f(}PWrPaZ1tvnd1Y=Dhde-{+b$RZ@M5XJoz9E%AArE8B8QuZPU_gxP
zHh~t0RYTx4*LSTo3$bN`fjYaKrV>@-%e<E;)mXD1V%xM6)~qln@wB=|nB0lioC$|A
zWRiH0pXEYky|qGFd&F9lUQ{BIe1DP^ET<e3pVikizRgzH=CRlpc`H#*bz9B2y9u0i
z#=79Rc9G|G&dIZHF6Ctnphe}dP4SA=N^Z(*xkZ00FK!-g)F1YOMIDnO=HEg3S!u7&
zK4*M3<*$-0t9ebw6=Z7(WC1k8*@0Y`j)clHxGBh)+O4m2<bj3hQA$UwD1XL;^^+oV
ze}?Z{1QJ>1NXH)J1%X79XqpXcy;VK0MDXDk1(FC9`1c`@9oFA3kR6~q1(LW%dR*Qw
zj*2rDS?;HJvam*1`L3cGuNKw(Ws|D(l_{l(X|ZLc#YvPTLL+Z`k=gR(N8j%|dCque
z@?4P2Mv?K@Eke!ZCeL*Qu77%W6D)9rycZ9dn@s*J_<U%@V4Ju~3MCs&WMMoyj>xAv
zKigO;w$prcBPQK+F0`nn&BnfME-F)_=#;tNzVIW_ujAj<w@E8=?wW>U#4CkvlOw#n
z;kLy@qte2%-nPX!uZZxjnQG)<FMZ5)`w~M@G1ArcB}*t476cvhg@48r#RE1Th9`yR
zU_WN#!P&|$*9_}TNAE~^cio}5w6oAgf$@D0V@K00$TaiU>>_?~3xNSth~ZsqAv&yU
zaQ3skg15@k%=4Xl1$2gkV^>{*7NX009L}`f<LR6|5z-R3ez9+iwm$JB+HE_~?n9?C
zMl?D5UMsjK>A_L<`F}T+8YHKf-xZ|>Q7gdNnsmBDmJ+6S%U2$EQS<tWSINrN9+z%v
zt%|YtIFS&!rF{b#0mp|n_b3@j*HWJD4cSeS;nU8>^&T(X)~3aW1K;)dza2)wkh1=L
zC`pI)_a8<(Kz9lZfK!g_-3!msm91s#zPG;KrR~M@Wf71|=O@cEsdHX_Yzwno#agjs
zCNfurX*WG_y`cofpgi27Kz+?C=zkV9)t4b_8WRIJF*lb1{WcQ<GBKC2=07NZO^@3)
z5WV|X=npKHGaP;ag#p{C7U->kZ$(c<lT!}O|37p{jwp#<Y3=pLW`S^NQRK|y%p)JE
z=<a)`yO+)WxOwgkbzTJv9=qeW4)}%Uzo5C%+0FwfXP`U&=>GamYjj-pTu5tUxOn~@
zxVGCT1A%*h<-%_PwojfQjp06j2FqCE${54F@W3^2y-V#pK5D0vg~)~9!G8Pq@sEub
zu+M$eYe=0Z#kMO_vCdXATI@?(LPLy(!536A1MIW6=RUxy7i*QHF&<yEwn40!H0vuo
zFJvNGMXlpqp8^J6QexShWtTSnPHsq^eNIFUqlzv?GF30l)9=U4f1B!m4~?=eb`F9v
z9E^W{Y|;w?_U~hds3K7J^WW~<=Firuq^hkm&vIO~<9W<xhfY}=I)|hj*VkoFD`K<t
z&(H5&)AIFYoM(WN(Ya^%lFE#EV<xmRp4t(K&%_ukbDIBYI+Wn&BzSq21es}qeBn+?
zT1nzIEf>n>Udvfs2n$4ik;K^$6h%`*nYl2qWdPSQg<SS3i#WlhSf5hA{R9=V3rqp!
z5{xw^&?l{5UzV3s^vZ#2vEDJ1Zv+ox%xrG}WiTMdbzMx0(;8CXHP=^fMGJWjR%yh>
z33FCbY{qm+0#k^HozhOiXC;)<)9RXP32N?TN=Y#wUu{`waT0oe3HyXRC;n69GNmNF
z)K_Z>g_}L4XvYcWLm5}MF?MJIQ_NTw9M>*tEChV@%^C@ZY*|ze>w31shAKpqrE!TC
zS>Dw=oUcVVQPYDw!&!P{HSf}RogS@uUsx?#EAYPY{s2)T53C`N5+(9p853@h#><4s
zDZtkGsxDFHZ!_Y5cAM!9M2QHPF`q+}8mu2LN)4cQiV|^+bdx=4NtmMY)H?cggsE7r
z7I)IG%1^dyw+KSEEJKHhZ23yq*XcWxt?g6E_Ry;$Zzyv|VIyso(YF*f<C&3hE^PG*
zDA&RnVG5ZBawS<RW6lyc<9N5%Z;RVC2$lqnRMgZUc!0ov8Ls)-{2+njC@Kx%M(Mmr
z<m4{ZAXrtoph7V1h+pmYh0*ZK&mnXT){hsu2GBc&4!u<oZ?;D*eU>3b-Zjofu^SUV
zO3xVHSxDpEB#|<aiu_dXm5h@uj3kBRCb&}jZv9r5=dqXmddO7mG0EK^&(^+9^VJAV
zHrhMNC0eh4Vz{MTqCutoD3|^CCA6NoQrkf&ID8R}@*#lALd)keEB(V+(urV{3+vK(
z6`H)nXMh%#HbkXE3uvaK<)m~EH3jb!;iJ?P0y5Xa9kP9ivZ6K$jBnAbC@ZFK{N&mZ
zx4l0WNw`K?fjG?Ob0}+r_2ZSb0rXB~1q{mFWj9+xI)<}9>*s#6#(LdeD09nM>gl0T
zzoNidn+lnB6D7@di_T8HWa^ctA(>*OTq@=2Xa!Fg{AQ)b31LScDC#&q)%gjRzg;mN
zWTyPz+e!Yx@lIi@LhhRA4r(cvUNHN~`Qr3UK~#Ue6}$1ogQeAm)1{;P+2wu_S&$st
zI`g^QWF%qsX<*0%jkA;fc*{9dTR~23`!VAsb9X%f@fFL1x!?fQ)~v$+1JYn2T9@HX
zF%tqXIG0g1F(nKzFHB`_XLM*FF*rDv@mB>Vf9+aPPh7VWe&<(Mx-Y#C@v&#dW4o#%
zrOl<cSBVsbT&XSM0hXmnFWE)QQl<a?e&gN4LO2O0p)>^X0L-5KJRW}@k3aiRLRJ@?
z<ceS9O32w2vA@MdQWzz;sKt!vC<gnA93{h|4d-M=b%L*f5tpcJ!ALHvfeU<81DEV#
ze<X8V%ylx~_##Sa7Q2{YHA`TrIjadh2X-ms2urMORf|T^LxsOgEwBnJqq7)b)c}8y
zf};@0o|$ER#DD=%EM{Q@BGoL$SS<7y#8HWCR&rEDwqU8_m|NTISc3HtqhlrW4EL<s
zo-v3B?U}%`NMlT}TE7uqibWbDyp$Fxf5PAzX-f-LSgExz3j<UGOW+o%S;bngO;E7-
zBs$?>(UV!lo*^Jf=n@uVHYJ-Sn3WPR4oWV?#s;P|QHzhVOOjiQ;*vJjm|QB{EYM!m
z`e00Hjm8keOHI}rAgNYs%h_RyGWtNc8ggq<omyy!G{DjpAD&ULep7VHWbr9Ee-&;c
zf`wT&i=8T)7X+y*t^~Oh0-+C4mx;{_2ADy<S;gg4jVnxO&SY&7WUg&yKy0q<+hmC<
z?O97}V<W9;f@KgzUd03lNErJFFePhY=}5J;z(;NoU=<W;4r~Ha8>RhHZes%^H~uua
z8d~Nc+zN`(hh|riME0z>D%Jb}e_S&erx<iC&2z?D2)15eVljjjt4K(gtwXF<By(u3
zikWTVF$*wD-TLmk?&&TTEMtkk?0){`S3J+YAS5lxS1h{O+41q-<HsLEX7-|EkSVEG
z#4-XY;;rypk0fSaWlZl_+Fs&Cr0af|AI}#%C;OABYYklYVt=`uE@qBae;@z#Zu#TR
za(_8>arIz(K3k%XZ8qjn&8(Mg)HzIiS$#mUK*}_Cj~{n07W2u@bh+ETc)H!an!a1E
zgM9Vp$rPRaHbo4x<#e_@wc^z}>0VAx=Vyz_^mJ9b?9ub-;hX*M=kIodvn4}FTG~VV
z1zK9eRo_o$v-!$VU3I^;e_Pf;x>j=6hW1A8jn;neJh0}-b>GhyhtoytGwgN$?tbci
zzz2+P4>3-byQnHYfolBtYzdgiSF08!c*?rB)$N=eEY0DwH?#kBPtM&>mT%^>Zm0Y2
z%b)D$o7c<b$?3OU_uHH0>$3wtnZNCRKYM%f=y3W+_n))pFZ}Pvf8D|H{GdBJLeZHV
zQ7H#89z{MBJ`$ehX+Aj2heAi;Na|6I-RWY|oxa{bnVzEUKc1n+K6-|t|4(poVFKDX
zF9kU~=(zgO22HFUbx>5im1V+MZ84&z`Xp(ud*a}Ir~7gKYTi9{-((clg51lf3M{Cq
zg%HI1F^uOLp1jqif1Yb1_L|C<=KAgaWHDcHO_d)#<5>R!u46^TfCdN^9Am5%OBsb3
z`=pKi3fHO`ISoXPB?RFbZpO8;Cv46=)mnXy`AajsZfQA9f~3aAzhT;HUk&HA54tF0
zOL(t1UI1KE?uU>$dWO2Y9tn{K<QSk7Pdp|}C<Kp%Gy)k5e^Ms@t<}EUMv6vLGM`@h
zV+`+?{<!Yq(@MVy+rIRhv~Bg>l<nG}-r(Ml-blTny^#lYh0j~TsbtUCcls3!o#-W{
zTVKH`MUP}X-<VSIf#RKB!Z)Q_B>12kss*bH()`w`=1QJqBLcV<pGEIA*S7g*sJ5PK
zgoNsI>r~@_e+G&-R1>mX=xd>R{5Wv}^mYvD$BEMr=;b>z&XjL9K1J4uoDUBQlA$2)
z%IH)s=?c3%+3d24iO>OYaT)b7aWRO5xX45!kco@eJjQ1t@e;=P<Xk=1JO-)B+Xcz>
zNMiDKXEp*ED-IB41TqScF^xgutWrz5!)M;eSI)d^e+Bia!bjpp_)xERsMkBx>mBO#
z4)uD6dc8xrH_Bfsd{pWs;e%b^_Jt4Oh;-qYm>8?z%O4YOSNX)`DdR|<i?857u=24A
zu5cY81DRDalOnHhDzEWo(x*fpB5$M*COt*D-_l3zOpd?UwMZa0GZIea3fV<6C|0H#
zBvp;te=Y`Ypnce`1~<|NYB%Nkt>|s&71h0`8Z{HAm|JK1tGxZN*B~Fg{kRCYMxgNU
z?T3~zKCwZ2d($xj8G<iY#l!%Tp?{GRV~~ll-N*=}U5nF|`<y`}L|>J=tR)g%(Gn?q
zj+Ri`XbI^hg!B?ZdI=%DgpgiBNG~Cz^hSNSe?)n>M0vPG8F7iyD~SU6_BZTrU-vD_
zXG#fX4;nXbQUps|ZauBho4xowzsCvUW6(RVB~l!7<8cY~Q9&_;9DI8|g#5sb#}IP(
z#$y2afg6t@<nWEh0P+Jj9z)10Hy(FRQLO)u;G&|)@pBYK;f<ouUPx#!B(xV2+6xKo
zYK4UMLPG1>$l*azJSd8>R10eDhG~h|Ys39!nxLVwY?x*{U&Y&Jni95%vXK)wKXbX~
zHxrUfGHv7pj-67|gPs`H6Pd2Lo45eDMoJ7JhhLryAU_}_hLFQjVgSk2{@df+=l=n?
z7<O8h;Y~3Vmyz2v6PG7uF}#1^=sEaj_xI_K4NuCx9vxrPR2#`??V{<iO7+_M_s~Zx
zOH7(z>T_`yXoAGoqlHyTN--@fCU2Z8j{%RE0Ng44JR>xkfia!*{m~-_CUK)^lp^`c
z|0bNJOz$Qz67XbBdq{or-o1S}ZT{Vq?V?~xyFr6xf)M%V=f;1KK=OZ$8xWZTIsE!I
zeBb=pubMdP+hA}3T&`&!6L3*8rR<<lGEMFMG;$_s>GJ2-kD=@N`d;^HH3RI0!<&C5
zR#+xD8i_M?y=_ZUL6LpaKQ0du|G<Mu?};PPa%7?ndEPRQGOTruvvML9&Tb-3S<9po
z1<GIx*;=w@y`@MUq^5t<9Q1JI&5s)!t&l{Z2a_Hvl!zeDdcVFepDAaU2HorWI2?E0
z1+`FK01urD*4`-;l!3H3FNbMyTbTmGseOn~THx9`W@ImjjP^w;n}Lj!QIFe3m?R01
z`vjp%aHf%pmyxYQahZ@^G@pHZmnKUJ&|{TS`a61B>O!BV_>h0+DUposP#qUP(pM+9
zQSb0o=Y5oUnC4~)yd6>Wz_f;)>(WTc`?d)wQ=~xpl6pXlc_)#k8UM;M%~i3YEK}_e
zlT}GQ*G1`k<fO|iO%5(kyQOs??>;lc8*5GI#hG=Te6;I<PCV?>`V%Km@JLbvCK<tG
z18HQkqpE*gDKmc^wGHfzMXP;hf0+3xtL_`-RpqsQRkqqv8}fqW<PJ*sqrt+GYZ6LZ
zO|5nuqvJaJWzBhImL-if#aF1teu75&jQ3ejW)w(fG!}QrQ_AJtoz2eUs$@kv&4ku#
z$;#^L3vNdvt=cXAfn21^4=w>Mjb`L23(&ke3)vcn2mF6aJqGfE4>gwZ5iwmo9&OpU
zfOP!E>PF<m)V<4#zfd=Nj7{{~NB=%^RK9&H4z@i^klz+a%EyeiIm4$KJ9$$E@IYIa
z0vR|rv-PnC#ZY2E{(YU@gJ877tL*-gq$RVVjFl4csiY?)<RS<bPfEyzMP~6vc5ceO
zwPaSBPSAf^?rns(+FrA@gSC)0mN=>JOtpp7Ea3MWq}rXXkHYOzUTsNBijkf459ZYZ
zEi;HW`;4V%M%5pPepR1EryQlpqMNJ(XSv8>lKZ5LxVq?o=XlQXpW(f=3-4Lq&g6@3
zqLWKF!nv&S(*gp4%#3?rKtM|l!At`}cHJ27j&gs+6e_4g8sppRZg7wYm-cdSxK-+c
zc{u_IMJM$MfdbJn#AgKx6pE=kNV_R!81KtqDswGl#QUA8iVwMC*yZ}`?&})xHxJ_X
zh#{S>|NnJOr_cw(3RoaR!@E4s(mf3U?Nl$Jh8QsqF;`sF)FL|un-t2$tT0zKcm&|>
ztjK3Sv?X^fq$k$}ld`-To7pzk4m<`u*p{mukp8aY6H5}%f^v%oMj#M|T-r1OrT?wC
zqa*=vj8wfCb*_=LUXn~DJt5MNRXhHyKx3%@wKsC}k%D9*=Zbn{s@G?hBYjJp{09y4
zp5d1faWNE^Vdy*)12Hr(myvTPCzq3JF_?daWB>2#Uptw>R%EDzHoh~K1ly{}xK9lM
zh2<;i!xbOgpXUCE(lzllxq?+QD)RLzv^&J-v0s7OIqak$KAh63`^8A+cfY)_7ac5m
z(OY4_v{-{Tw8}jOq%xnMGtzP&6KWMv4<jlFvuD;~$bk`LRoi>D<WkQZsbFn&48?!n
z+6af02@JnRkT7A4On~w~TV3IoF#<rfR^hosa1@OkA9~6WV-L}>-VJ^-;}~kJS8`y&
z%+GEi?qgnXV|q^BA~TAKpJ4LYkI=M=k;^>d2Ffv8ls3-r*?5VGON&oV@u|c@d10g}
zUW+Wpj6_$KPdVb{sd;XI#3;Csf*F6Od6=q4KKCim=ts{?A_3+$;eI58%2PfQMA_36
z%OJGZ!U5GO{hWprs1_^&ycgL%<fBz^E3-%`fyOrAo{pw470obH+k6xLL*aopAsFHG
z(p>%dy8AgxaBNEud*#0$yYPTO3uAqUC_KaX^|kx9`@6N8qurikqL03p#%O;%rIZm`
zJMPU0OSnA)lFtV}^Pprm{KR=Mx$|Yn^~huPEi`zJd16T$5Q#0!-UtJTS<fUw(J@6v
zjiY7r{4qmB$&S=u$<M3Vt%NXeXPBqIwMI2qxA56HeX^RPH0$|FebRv|{`UM`9Su)R
zr&=7R1WoHb!sC*g>pOlbKrw&K=w<1n(w1_O&}R`lP6Vr;(5a*@Icc+y(4wr+h_rSp
zxuu_><v`^LN*a&5j&tORXeYEf>{Y!tPNUn)DS0QXr48kj%Pyg;yNHdY3|?bppPAX`
z(z&lnhNb4xxf1WQnLa<GQlVcYzobcchgZw**?}a`v$0$;khOwWvs8b#1DQ$n{vqJl
zTCNGWCSzRpk!wQEGO#_rQntG;EN@bFgFHQbGmcZzZh=K&Ht>&9cIA;LIDHCbZ?N7`
z*_)oPRCdMIK)c_I?CV1(4vcL<hdKa(nmT~%+cR}Qe#bN$0|wLq2W?Guh?Xoa>uIqZ
zwO(hdcBJ6LBDPrfL;Zh|->v;so!13p#y7@2j5iZMzgrgQYxPT?$B+?fmJ!0on-&C0
zTj7WG!v+O`QYbXpOE?^_UEViydEYD(GL)oc$7M>`q^<8O02B~&7OpJ-&I6Aix14Oe
z&f{1$!0)g&N9%-RF5O4$LtP`S-<~`v0p38}iT&ZA1%682xnzF>KO5;s$vbM`>5891
z-W#lUl=r6RE9D&(35E7!*}Xbc&E+P}K)!QXe@4#BQhCYZnw)29ULfbAo$Kj#0!^M5
z+2w=1n!Z<H`TkJ)bQg>NP4ybV3INzpufc<`pvR<)+d+2NBC2(yto7t2A`4+_0Q@<T
zy|}lPMCN$8WL$sK+ntueptyXpcM?Y+5;(1_F7(}76b+0Qz+`NCJE3S`Lwbf8zR|3l
ze4kqR?zf(@Q!({krDMJ^(cx1lU4!+8t~@7z-V4Q88$eeIpjU$Yb*W}0ww8TLjol<!
zKE}BG3+t=de(fP8{CdykYduP;Pz1m~mJhPS2il1M9fEEvN=rT7$5!bW#H1i9JKhn2
ztDn3d*eivTLwi>FD5-~R?O0_E@o^?qRIKPS!o%=sIEAfZ9R;$WN@D3JoUd(b4OK+k
z*BKO})rQ@L;;Bm3q>Qke#HQ}fkIRSK3b6@Q(pxst+cWFQ|D*r>U6&DYF%$tamr+dy
z6PG!0F~I?4xBGE1-2(%G_p6ujk316sI5L;vrvVg~1a>jKe}ST<-9N|A8(qLY9)g}j
zYMoJHyG0s3b_k;<ol)Z<S$#Zsn6<{Ntm)VNqBZ+$C0Yeo;9(-N)+y~><>BcNBLJ?%
zn8h}W501xT-<WdlQe;zHRIo8Lv3sq~J{>pzZ#suaifw2t1ck`Wzdtweh5_;KLW8IR
znC91i&G*fhf7)t^ecdPGqyTCw>liH$tz!0#MOLhFVLG%jGQIov>qk>v`TE+ghpi$C
zwFh{KZA6(q<VJbpPB6v5N>W+KD1X{K=OCwGyK-I<LBmA2Ooo`8Hd!XwdMhy<eJAKp
zvco~v8A3<6l<<}kEwY*p#wM@j(Mvc&cNWL(2dI#}f5#Y2X`YyJluxdHeVu+df>Rc3
zjrDqQ+{OK37|Juj!w?h+)(TJp1I4)Ro|wsHwJC6@VlbBOAQK4{dRjiQCe$p+q&_hv
zRWTUmAnmCHic8SSNY>rRP)3v$8rR$Cj-Il$L<b&NkJp}uWMLX4NVg-q5iPY3#b1@m
zZXc4Hf8>Qe#MWEsZIEXyk?YMna|CNMA$ucQ+s3X%!i8jo_LrywV-Peno7>pGWIZ#O
zwJD8r`8;V~%71A-YTF_;VNp@=?qW4rj3_-gC!i)C8VRxotU!$fX!iiy+lY`6m3D&X
zJ|bj1C}GNtvVMtZp)i30)`(_A3net+WCb5df3l3Rf{p(Q$*QovzhqT_R!WvZucX~<
zTboz|kgUPf-iH#GcaUD@++{MiFX6TwHK5FKBYA4m!sXcKT-;Ea4Cu6bUQe3HvjPmG
z9J*3w`CKZcJeA|rFqK?Oq=&3kj_9oPT;g$JTaa~xP<y<SuE<)H@ZwIoLL_}KD^qq+
zf3f2zQndCsSL}#lq{LSkuc{n**ZULkAPoW!{tEG^u)e=|RDf2B2YOP$@#>9dBEe3I
zZ0l<4Q5=v=RsSH-wy!&2J1-7-bW4h^f}wWfZ#|@{#}~SoKIqw_8BnK;FXSs{DYd&q
zMClRREsHw@12HOdS{Vr4e1z=~7ZqHwe*)_<B-%`%;!&JTEqQLZt}K-WOX3sIzOz4y
zjVU+=Xm2Q&QpL#}e?_^JokF&DdzQ6_>L<{$N)1IZ`d>5`spXWzd9@rlypvc)-`<)#
zi6sP7U{>a9jaH5wE|AI!;RjPnV5gvevHhWyvcmfQN?8F~sgyuLF~aISXI?2Qf9uyP
zqvn*d6j~+F1FEF)Z@g7J4Bityj|#Dq((IfXmUVFZKGIccxPY`%88Z}LYH=OYx3d$k
z&t}NPQpQ1T?VJt|@^n9?1q2nexuVhrt3;j7kv@IO^X1c?&tY!gEwrEe(&s{#?~`35
z#7;W2YlL_iptPZ!V48npI>NM43#%*r6n*{;VsH0(ml1I>6ql0kC=&xQHa3@$b0;U4
z{#ZE+mmqsFjRA9)u6r?Oe^)sOf>>Ss`Tlk2dOp8a|B<IodF(BMPvM$`w7$Wi@iPyC
z#cY9Qt2DR7mTH@<3@uS=bsiI@SF;xsp1ExGQgBxxSLO|^Ku{HOK@S9YS-zuFxXV@*
zEBKyF)S5vU?eO(3jC*dZLhT%exbfwbRw<sd-p{XP*5|f*D-4+4f7V;X?42l~)xy$D
zbJPl1vW)2$q^;UtVlBNucPG%Ux&RC`N1bxU=93~hov+fP4+#6E*zySF!w8?AQjT<p
zst$`%T!JXaIYubY!-%A!g{bN)^K#A+wRt>dET&P%+#Bt!x~Lsfwb*Puhf=G~qXroh
zEy**;J+{g$-uNPOf0RYzsG#XoZOz^{F-<uVCaOn;W^^mmWo9Px5wlRQ;5p4t$-9JV
zP|sC&wl6)UY-9B#XD16&j&o3H;|#aPOH3KM_+%?fi$B)K6|wkJ;+N+muN~9u^Ky$8
zwN`>0ws{{C1akslNs3cWDoDu^=`$x*d5%7}W=tY<+u2*Je^6RTjzn20KP8_SUXh#u
ze$=u<<nJ}(R<c_ufyQ>m9nYqK2is)qmhtdEFI^Y&Juzw>ng{W{mc-S?9S{&>-o*JL
z>W@U;nYKmzNG!oGIrCGfpNth=xnF|%J6j(}{hj73sh`(g!7tn&z^m?XW;!)9B-8;!
zy5pVP!%T-Oe|10}u~G-5rx8M?-eq`=&w{LQv9F1xapX0sI><C!9yvR4T<oWj>*3|_
zq@~+8GOp*i5;fweF`xBOtFfAUsK?48T6mrTuF5t}XvAJN`B@z8v~nr23x0-;pOZQh
zM@YrlY<>uy%83S@DpU0-VLU25uVvRJ?7%6(DDM>Hf4FU0=237O^p-t7Z#)RqSs~A%
ztedWQr3|&Rk~)r?MD*}fN>@79ceaY|BxAMcc1^P7JF0S<@;AOwR_v*zG|@D(pYm*v
zHoFDcp}<Ulk6tbnyTpVpeGh~uo~PNOkfJnS7!v4pT$&2#m4*vI@59R8n}QP;uL3&~
z3pQrOf5|PAP?NA;^>02N?ureOW@bo4yp~Id?<1dkW{64Swosnruc16io#KV6+%=R>
zZ7wKZLdy)Pz*QR66@PI_9E+t4R@0VGAB6I~_F>8!fzsY^AIA6*=<5zBWJo-m2uM8U
z0}7Q<i210oJ<Kp4+Gp|Xa9O>z%eq$Y3Ptg9e?U{U57<8`hL@oSZJ#(QEZ6?v#9gvC
z0taa_r<yl*N8lOaye#*&wnsT~+?ULcaNeu_53s+%rU@Od)Ea+@O;cy<9YguvUdnr+
z*w)UVEB8`pjA%}SZoC>e%>^rVO5$;;EgV{z&W?2)U&q2$?LlR#_msuvATN$m>8i(c
ze@T`e*)N8^9`zL&Npcw<4@D^xPwH{wq>lAiwIFJ*+WLdq(>Nc^JB99E8ps^0zP<)L
zE^Hh}>x%cqZhc6JW6LP{J~M}6$r<X2ZTa~4t`v*P2(NB=)QJELKWM=nL43|<{Y%B)
zIrquj+7Hoe*kGCQ*m;d;uhzb@5=be)f7(v_1NauS7K+_|4B>(SW_P^7G+su!MgaE8
zH#wJ!5S*KPGorWw;R14bHhw7mUlVR;>(3|L&Y&v^H!K;N=q8KO8tZ!AlFu|n+EuEj
z-PrRX+KnSOX*bE*qTMv2&~Cc(%|Z{YBt5fpvgOUPqqP1zqQd=jc8XPT`<rAHe;|*3
zZ4h5MAEZdCsynl6{4$+<s7FI-lfOj5ylK*|ZqwRNlIf%4n4Kl>DiqWK%w)4%tCIBA
zu^e|UWNhhaUH46x>i8tj?dKnV6FX-oqjhFMNB)qi3w-K_)N(#S+2d1JXU*+f`lJo6
z1y#Cp<J#4NYw|JM+_j#uyo_mc3qd*4w@khN0nN~Mm2m<S1UNA`G?x*)H6)i$k}nDZ
zFfo^Lr#L8o-)|f@4u1DvA)qhqLp2g9Qc_VENSmhDq6LCD2VBqagMX}3_grtVUKhFl
z{yu6}-c7Tp=bS9y?t{IY8Hyq)lKQBb5=(Hwl~|%nWWV^A@SCHHq#&v|O4St#*|CVR
z2w2+~bs}w0cDaoz#l;ZH0jTO?Owj}c7df+0kSW1`MAYnJ=4=#3<lOFI1m?AwFrr9x
zS{PIf(e7bvO>7i_Dsljd9j@Al1WKEhP&dXlLP1HJ7f;QOLJ&_J3n(fui%jQAjwCiy
zAx}qJ@wAj^b5zG6l8p_HDJ7$#W6o+6Io4qEB0J707?m8?+(y7rF|-lYC1#`GxWsBY
z6Fi=OLd6_Jpu}`0235^UsFs9HXMz;TZG^g{(e#7SNwv*}dpX!_TwE$-BVYoAG<nrY
z2<9tzQlfbr_B&xUotdzH!K5%d$=s;qq-c{TPD9g$)hUgVgVQ8tIhGc#GK&y43W1md
z%cGJ_GTfwUF+vGaw8=}5lAAYD29=tW0;)8BUGlixumB5RZQTe?%t6dv2`)!w%mFuL
zERx;B)6`ZK%voC;NJ1cvBgExe%?C|gGI;`)7Q&=>AjL+QYNMtK7HEn&O#G;+f_c<b
z@o&`B7H`zloLJgQTj<qjBQhS;HYGBs)M_K7xEk1~!mdWOxI)y`CW??6mBr-c<Z7~i
z1SBLx&7(YZwItISA%NTp%`ui{wrC)cHGxe}LMBJJ%VZ*(W5`qvzW#c6evGUlB>Ai1
zU%&j@?)w6M&e;b{c5`=qeR6pC7o1p4A28VjPO2U&vBBwvZ|B$Z#nIc-^O<8C;D(o{
z%jIlw<7mA6?%neHqvh#x=Hhtc#r$S}i8)>%6k|YHu~}YN6%k}wyD;-C+Hi4rIJ{iU
z&yQxy<KgA=7sKn>yX8vB*Z+Mx!(?w}(BWn|yII~^j%yR~YIZxnTb$2s$MRseex6-i
zoqjWacWebwk=eP@3C1rl(v?uhxu4zK%uSG$!#4fFajS@?d8>#kB6p&6qOOR48amPJ
zJ-Ju4i5`yb&X#ud)78y?-0;nOaWPvomBPvJkKu>m+v7N1yqcXak7198!x77#F*#K%
z1Qu1Q_#|>NJadqKG<-jQJs+OC&lMSHVT|A`AtbJWzBoMGa97h|`?Od97yGnC9t8u1
zq<X=^M<HKPQYEPmswW5UnK3+nyBB}Hyqe!Yp?|*m!Tvt~b-8?d`_(YKxmx~ucjnLM
zzYo9N{r>jLi`gH;k9R-6^uJvXXV>$y;qtO*J-<X%K8x`(@<rmy#IrohXBYV*(Pg+y
z^|Hp{c5&WEe)dyDn0$%<9AM_b<LyKN_Q5uum7Pi1OW-GCr_wh29v4f0Z0U}z235va
z*v`UUWBwqPqDW!Zlo`b_V&gzSe^fAWs9ww3$sE<k!he%sdYrR?WPHOStHcFr@t}8}
zxKYWou6-S~pn^YbU)!crWn<G(sH}hk@y}b+dtBqzq(ywKJcTnypNS@L3aQV4loL2(
z^0pN`%$bmV$gC-x*lY`bw#SD#39$y?GljDtJamf8m{9rS6q%958z6kTpnzieGbk`z
z{RmJ%Weo}{^?iOcv2_(IVub~)r&tlO6V!=5DyB}zC&LRC<H?$wRgKpYkR5gRHD0;+
zWIf-i?kK|DCf?^7FHGJd1^C`u7bGcoiXZs8qzn%$Qw>R~qD`oOALZ(VEd{OD8k+@F
z7A<@2YOc{+SJhq3wR+BF=e}kcE9^aV0X{_D?6t4al=}pmyRXsadf@+^`x>R$TN~Vc
zO?E$|{f3DKjqSK=1ruXLZfD*(Ohg#lPIzCftKpre4ifE@uY-^rx8TD~hI`ZpEEiR#
zaK`9kE)zHtFsGe=`c2@>!E?$JIITOAKvOtr8$#R0NdX<+l*s{M(V_7!pxg#}dk@$B
zzfg|r2T+dlW1w8B_fXD`QAY^I9j|f6Yuxb~cf7_Oud#GO|NaL(61}Y3MZiKIfMNEz
z<|e?jpnAW*rBsmjdH1*tIBnykDF&azetViKc{^qALCqn5_!4&?)MTh_-HVWcwzA)@
zjtTXG)~=4i=*M>NYU@S98XrgNMS?zgB*A6Eant);e|6lnm2>NH(>6{)?+nm4g_9E6
z`zbQ5VPcsgbNl<EZJgWR7j5Ny^6!gC!v}am%46`v__}Pw6VhRXbQmEWMo5Pd(qV*j
z7$Mq;{0Z@Y1RCTTH}FLETY9)p4=3!Dv##-k2Cus3a3c8%DB3-pP<cbxy{i*i$GlN`
z6WV7A`|WE35UKD6peSC1b`L02yyf&gqOjVIdG{Jo07L~eNcV_Blbv?%{tXrZ7~6M5
z9a2#CRLhfx6jC2JoY-J^08MP;%<Lne^uwHz(9xlPTb{s4>Z{NMPOgR#WC~}Y_8#D2
zPHc@{d5V{SFnma90w-XCwQ>_UD|@L=8K11=ZQ_$ckHIH(1E1s$LUIQoxr310K}gP>
z=x?=hKP4$0-jq-Gg-bi13+vNnW&Fn7cjJ<ar?T;dOG!SLJ%=i4Z!^{hANZ|_cIN6t
zs6@qVzI@C8h3(5m>?rotBkGPm;<)*A@$TvmKVQ5VPS0+Kn2_xWCG_Jy0pcpy*Vd3V
zh3NPE1y5nGN!H({pjXT>?H)z3_(Exoq7r!~*>4nuj(wv!h>Y!#>&{)B)k|8(G!{HO
zT|fU1%{(INm*Gt@69O|Zm%&d069F`r0cSWTmo17hngQFF$cix^e_6xx$-}B-V}#Ws
z1}{<?pQS_$0B+Q_fVc;N)_2-=r-%$V#@^ueDPn3!r-}EZS8Kj^3H5}T2sXkNhu7}z
z_`dmfQ<b6;#)iRy7m85$=hr4a&_Fi0Fd&Kmdiec)__6u3m&$(W#}rfvP-$r|lP#hZ
z#?Tf~3rpkeu98BNf2jKL`{&TjyuFTlvq6BwUK_l`YZ6*@g>%9J%-N^OE<-^}e^MVZ
z;0bug_n7BY>lBv*B|3_o7k#IbxH&)7NMDE$c9!v8=%8vmCjXZxDaQ5Dp>5L2SsKmN
zq29&!_Kj6YDlh^{msdt*pwDLCUYBzxL?~@<>nB2a0Xz&Le?@p_P%;KsxU31ZxUJ#>
zCuzK^)-3pz1bW@cr7&f$((07=?4{ysLPnDwy<^l2Bgp<VvU-@YOmC@t#jR^s-bk{-
zHpw;BOQWVLiM_zjt%`O^)=;Fz=Lb-GY_%jgmTi!1hoAa*^#3F_4;(GNsc^JZd5T13
zthY4UK)Igce{}ZATC5hQ8Wa|5+M8l4$17*hEu3Vi;5isM$ur{}^B7rC?eg~+Gl>;F
zv!u^4lrjcK=1k1%SeCttzl@~fj_dsOA<I%OE*iIxk4h2Uxz4yVx_dm;X$QniUqyAA
zC`Bp%_z1|Dk2v4Lj7WP9y=g1Fh{T6lo~?++YAy#cf1qM1ik?|7+z%i?twDhMe-L12
z>mx9`#B|OGr_VIs$#j}V)Sn-}i@Tx3^s2XGWX8sI#hGS{@NEv}x+aV~Ravo~V02q}
z*d=37jeI%bR%9K1k;ir|H@B?Yyu&WiN7jaU-Ar(|eU$3<?B@HaPh?IljZpHD%&8o;
zYoRh{f9YFd9My!r?)0WC)qA<heyh?KN|8mOY$yB+*h|q!$LcR)F9JwN@atY6AY-pd
zmppBstW4)qiaUAuK#9ko=wVahw@(Nu!b6iM_B`7Vu&oL@=Y5`yYirMZ%Gu0M3_ppH
zdTgc8Nn<U_5Q8wFdV9^jAR$BimZTll3!PB{f3)u<MNFwzVjFXPsrKzk?JIi+u3zmg
zV`a>TrSe8!i?Xz7W>g)FSvgxI$)-qdF)US-5%qkQ*pC$o+p27*)hXVOd+qC4t|Z%B
zC821AT!$Y;**Z`bk#qXk+l}N|YHmWpX5K41NQfN9_X=u}4;vJ&p>X4!pk}MLq*6Ae
zf96{5Py-OKn$PN!IehHmu(+!NfuYJw26Pn&q!yMaV7&xUu^jlraFaTjdE>H_$YHT8
z#Z*z!)XwH*B>57e2p!N1h0a*&XhSbC;Z(b>DEanZpFAY`=_I3^|K%wV={nuNt<LPo
z#a7ZT;m9^!hV&wy3WW0R20W=Xa3r1ze_!OktaUQ;b_M2Bt*9;)n+4=_I<$Av5+%){
zH&WCm;3xy7?IXi-B5;|0+Z`A-dfY*oS=)iP0<hQTxWV)@?c2QJD2c?B>jcTuhYiH|
ze+W9%bPzX$ZtV3|s@N{|{;K^772NS7_r>i@o8;_x2%fqCxYDdggc>*+o1Za1W*%#u
z#Lo}5Ug7=G)=A1ajy1>ccdI3CQ+mO|+Tcx^ynW~Dc^Z8D5e#sWx*tD*0fB2wn2m=h
zRWlH!)5NR#<s#i10ajgoD7!}C(!bTH848`k$w7_(qjtq%tvf(bd-NZPlj(Vv;Y~3U
z0W+5|`wbMA?T<0T0e-i!kTG2Xf7!l&n;}a8l?^m!mb4ZhY}+1=TD$tSsI{}-6EzLW
z8QZp8*tA@5jOX0JBm@~R!-lQ5v6=;q1;?o;7n+Fipp=RPd}s-6iA-%F>```=%sH!L
zh(J6KC*!;+R%_$oCd~?V?J$%vJUq{Nk{gthsdBLg9s!E$U=OWF(@)0sf91WPGr%L<
zS3sX>wrx~}9p=q0in!FbqHTNAgr*>g-%B%gcLQa-Y-M%}Njeim(ixgRXVznPzN{)V
z?3H8p)9Atu??f5G1In<o_DdTst_}PA`%rb*fw@s18aYRB;=;@d5m6K(@JA{{13{U?
zK{P{;)mfZTA$FFpD8z2|e?kfoCHaYXVTIVO`c8$|S^E%$h)y&M&iR*%&f+}hdp_HL
zkjNCkBz3d{eoi*Hpd7e<_9Jq|dxdCkT~`V$9mVS%Y6UH&;<${)L=u(7bN(fwv$%Mk
zJuXvg+*=2(#J!#Tp12!P839kIRQO-0RFJ9&z%!~;aH3Kt7N?oAYU^6Oo0aJ_HO8#D
zLzx1L%7bvTGQAt<4RYvmgHQ2v@*U6x<S2mbdJ&$>$kDg}wLnV0Du(%UBh$Z(0+oW|
z!2C1<w1s=npRjvCf40Vc$<M{*XP0~*sy<sRH;PX=9T0B*?_^hDq=Nki5ZKwUm*Gt@
z69F@qK%X8I0y8j|;Wh*&mqnE^G6EAxmwA;j8-F|@`4c1$8fq+nFbbN>kLHgrQljDO
z#Fv&*DZVU!cYLSAUV-BkKz#8mz+rDSglBk-!tPmRx@Q!x!ZY82@5eCC&wFWPkifrz
zWrIKbdHHQ8ld)2G6TXI9{p<a*`)Ajij!Ka64OtRe+UBnxyD*`Fgt4AI)51vC{QSH5
zwtxG*7TYV6{7m;jvibSFsYbpg^t@#Bq%lMVdT0cZYmD(ig{clb{fm^uXz7+O_Fgki
z2!)q#t%F&KWQ0<_mdG$lVr(^Is-xsi8av($AJP0`gw5|IJBT}7Nk&=N)vxwaD68w$
zYGMZuaGXf?65f+d*CD-%r?x^Xdk3D9_kWzlgK8isidu`2uQy;`YDIRIe5Z|Q(5xw0
z;uetOxWkDL0VSG4uRZLI4EaM(z!6Ch_$}kIlo-4r#z|wwgnygv3S$&JJtYhwjbXwQ
z(|Ub~?z;qUJ3Jbe%G2GUjs}JC6Jvp9Hw=vto>aYB-k)L^Dv`lk^;$11*ooeA(tpb)
z4Ifc&F)~i=yxpL=BKCV@JDel1uF96jV~ae4tks6GRY`nFY^AjO_+D2=U)k$w^#glN
zT+WTtJy@quMle{=#-g5zN@dT|DaZgR$HR(@Gof^PiOG(+nW30Y8;X1?>#|Uov^g$q
zkeLK!3b@4<g{AR<K3jUaZI99x?SINrBnX6l2QY<gIM_m?1F#}_p@kyLt&^%GAKTFC
zfKAO*lmiI|I|Oa`4k=CUZG5y*v2az552Em(uk=0DJ@z;Y*_i^44=sVX0)LTC`Aj>H
zf_%=9MNqSkO0!8rx1tYwgdkb_6im<sQv~&#fex$iOI0Im9@TzS4%)CO{eKLlnGl}u
zV=08I;l;>sxU1UUE-fPxZB9@+uZ9zn5};izDPW39B^#sXy+3qJUb^giM97q0a=Pu~
zxXml#bLWgwmR6=juVb|n8e^DJCL`(&F$qFkgm~c@?8Zmc{9Go;;A>b7d9}^u;}980
zx4QRMrzk^m4nw)-%c)#WCx4?WB}{XX_PjZD>>V@0>7J1=9lyyQ<P$Kj%cuqZG(w>)
zee<zEj(bRJYfDnWcxyQ=*K!98x4+oipY%t2WXj%AAA4{4LGR=J6M2{NYzo*7dyq=v
zhP<nie1&&at6#*slm$WFznFJbS|8+Hm9$UcT}pGS;$X|TEG5pLzJFu;DVaRjpGVg0
z&#_u*TV@iAc6%qBQ?;bN(f+uN$c>!xIKg;yrjxnZPDCe+(GxpA-k2-pghhN23sYR<
zo!s)Dr%K|pBdyZ*K%|wD<j`-O9@|ez!s#Ve-FJ6Ihj;oU(uJnX+s8>4Y9*|luRCTy
zBlkhElWe=bKXUM6nt#uIkRA0-k%&irKG67^I2<DJAxSrLxVsJQ{xq1_I%DWU4<G*f
z;~u^=^aYd<oD%@_B1#yJT;-n-99>Y%{KxYIh6&`D*V_LK!>sHb7$(>G`xs^=>3@d#
za)w#Q-sdsQ%90xyCTRBC>t``cWBE3Dv1cToNca+d31Fkj?Nqv}O5%q6vXXp-UskJM
z#4kMwYv9H7vQ+vYyDWu$3c0k3`_V1myOaXAImNpUik43C55LkYZ3VshMz37_z)PM=
z>-syMO6yv0Qs21>{tG(|%+Z(OO)(P!Gc=dMPXZGHH93>vs3(`NnlY4rP~m?1>+Sc0
z%)(0HO?V8q{_XYc@b96Dj!J0b2cx7wiNoJN4q<{o3uFC2UN}(0&wqw*hd=sND=)2v
z>F4M7p&R*{jpu-kCymj1W;`@PYd09<g$h&2c;xwF)6hjq&4wV?DJc2#^rK~BIvZi8
zUoKu=<w#bLvX$ctb8?A)BaMkW!i?*t4%!H%e7~Y*M{0HsVbVdfGqujE=l7`Jm}uyn
z=Uw6+=iMp9MI6RF^qi+T9wa|&zT_ndDk$h*exg}sBW#u;G+HglU@tW2naUHj7Zz+!
zEyv7vKL!@&S4NO^)mYD`CnNC~jh>%6SvKR(N#%H0Y2ysvjps;zPBN^`e65aL?&Pu^
zBnhJ@*-Un(%br*UWDd5?+@RTh+WWgh<C4N6(wench|+B8RirS$+g#NQl|!Bdi5Uvu
zqmsp7-4t#;eWFqVjqQXxUqYG#n-ggi3nr1u1_;lPcj>K3PoovEOnQjI1MMWe?*5Jh
zr&&$a#QVb{AlZC>nGkpr#ndr2+jLErO;{^tk`NUrK|`k}1jso$ZV6~a9*eI6RoKL{
zNU&wJ&F?A4*BPB>f|9T8DjKf_IU;W{MypQVlvskia4CsJkJlqoD*@#Cm4Zz<ShI{#
z^Qw<NXDwaPJzB`jwdADn6vu6$h~JIXm3g)K#;iQD&1#r`P_|YLjY=6jpsPh0L+mld
zMF=^b!8G24(>UK_ldgdp3bids@DvM}L|wI|9A&7T!vHvZIaL5*F?uBtljlvy&go<C
zTr8aL&4y7hX;3h{&=sqN@~g4JJb|yrf_4me*7lM>$!R=vW$%z9uAr*;nSB&T=KX(}
zu;kVR#?_&JKwKm+2<Ro<UfH%^LB*_gy@(=10*rrP5!J02?G*B#&91z?(g6n%{koM=
zu2wqWxn8i;Q>%p>DcMTa5?-@RA1_LYkT(k6U{9`UEHBTA4+5jebvE+#9X6kHMYWe~
zHY~?+hYKAdB&}I*f~ZfhBeEBHM?Www=PN!2zY}(U&kFk63wiF+yfVFjK<kCP=CEF9
zb5H+Zkn8EiN?ZMdF~W1s@A=-}Kak{x(bk5s*gtyt@y((NFrXjVEOsk?T0Gs~AZ7=5
zg6=rDqyB({mE{J}uQZ5L9rwRAh>AGX*+({r3cN5dfm5!6)7eLA&7!+=*$~4cRQ8m7
z3@_<_+OCjdXsn{x_^#ejaRG#=ZWr)ssSSopZI9jRP4gEdE~RG(>>~ZWO&@of0`ONg
zeQ3Gqn_~Ey2Om-BSj#a4J=YW)>yYgXTciKrYMH$qLAQYx2ts><s0HReB<ly-;2^P{
zwmdFed3qx9ZNuy8$VT-;ZmmJ)D(jfr7c--OYa6k=Vyx04A9rk9&9cgV%h+<Dm4g;t
z+VOEZa0xA6(yYJ?Rk>HfxTUonveia<YGE^V(_-h9OM_>tU2(F1luzdSONoVQ&CDB8
z&A;$1a`K&W+jR-kc|D;zrnCKs%IC~uOgg0V{jJtHHseq-TpKTGBzf!v=9mwKv&Tz+
zoHvJF3rzrpo2sJP6ldvqZ)M51qp8(kc64AH2^L2NRd(APUQ?5&9}mLdL-Mi@1Yxk;
zF&TFi<HqTIk;-?0_*71XJ^;j^xh}JKpO8CX%uwcR$Keyf7{GB(g<k~5fFvAEHT;ez
zA9(W6alQt~`ShXt;%-E~2kE>La@tmZ%Ac$~wroK?W0NSP?9V@K^Bn%{Oig~9=F%zb
z3REOoDrr>0R@D?~X$3yh+AVx$x*H`d<#PNxcC72pPp6eWHjc5y8hi5B8EU_TT`6%@
z%3V9I0R=~<oRUaF_jN>aIk%Qm%lW9bvp*{fZqN)xifhu_I8_mO#LV%iuGjo9nb07k
znrfP6RX-hU$8TiSX0=L42k~vC&~UKtssTO=*9-}TbOWsQch0r{0miVImvL1$69YIi
zHJ1^54HK6t!v!dpI-xNX1Cdo(mr<cHCk1YwFJC{GeW5WSe{<sxX>AODJpT{;Yu~;Y
z2s{ETfA}fD+ZRue-r+F@%e%&vcMOlhcm4(by{7y8e$-AU3VaPr8*Xp^{`|*_uFQV@
zh<cAx=SeZ`&MD78%!lO}DI|XQm;YPl#4>fuvul<FzLigsh-?{enL}IOr+RJlTM{O{
zPQH+&4<_-Rf0_LH5nz`&Yn3B7&RnIsp&W~f!4g!j9|=V8uZh`0H!)7PiECC$hEzqD
zP)$ES#MSrDm;b)>iO?wP;>Dp>9{lA$f4-y#1U#>g7i2boE`R-Y`Tgael4@0({LMC(
z!P1WTAH+hZtPMPyQ!!ElYfWp6pHM{=jF?pa`s<I2f4KAW$0VDJ3N*^uiaiXL*Wd77
zIRN>_eaqQ2n{~)G)n2t>BU?IQo1cA;=yg)KO~3Xnd!uD9G{&Z<l>WH|p6zbtF!rlz
zI28Jwb;P#}AB|z#aFW+eD(e!X^=uI+sifFKKFOD2i1l5vbmcn7uHMn*H{qV0&d#N@
zST%t*Dg;bc!k=KQDb};N`}4=9%pBwt;zDL=zl7WNd&dGebSf}*F++IEZf|dwM58fw
ze?K^k?nMx1Ial@o($$-R+s!<U0C^RC4}r@N_x1%Lt^^xIJ#*ADW^P{Pc@CvvGL!t&
zHVz9em3}s$eMJ!}Vr&J^sVvDQh%F^dE>HZvwmd59jie-Xu9NhpahG#@C_&4_Tti%&
zkrvafl2Dg^4b2tYI<_z-pM+dXJ{_$>e>u#;VsX2sgUNB86%NT6u6=?;D@*lW)8w4f
z$HSuKLWG@SG1Ckm(CY!lIHmc`rpXNyi~TWHA-qoL**V5kan2)JGR5$Vuth)~Ly~{<
z7=CDi@=Iz*?5iRZp`~RoGd_kik*<3Hu3i$&49~`OP7FcZgNoxvjUxqFtR7TQf1mr6
zpZ&LOe2J@bEtM2G77F&;{Jl<@s_&JV%_Loy?0ts3U$-~p#O=%w=TnT=cEj(Dc1Jz^
zeHx?ln?WW--I95h?%MCH&TE_8t#q2WX+=bobh(&nS9h_Ub2-Mz;YrolQK}Dd?&qAB
zBBRr}uje*$Dkc|NXKw{PMVX;UfALZ$rEFQ#2~5$akNX9&MMvdQWh1!0gT1C>v4$%i
zn~LO-dM?4J>FaPILy-#^JN4BG!=__hBgx0%s)o41i_vy!4QY&0YxzzpC$o+!&l#cX
zm|g40UZAh>R_)zDZ`1Sx4VUBA07=SlwjK>m8Cp>3jTN4)J1Tb9se+|jf2TwJd2Fp&
z^{(1d54zi^GPG5OBdV19wks#G^*L`Ba?a~+!Q+<Hnb%ek_<deIw+wc{Yo^?)p}5a?
z42m2|3OVyyKW~UJF7!6!L|Bbf<hTM@U3tlB35%8lX5q7hC)x>Ny9TU{iV2*7g())b
zR*0MSOo4G)5S8Y*)jKZ39P&fmWLxTMr8iwC+6B)rCXme)2JZWTL$eU)R+mwxF;6Ft
zDY7AmOu4ZfYB?60C@DT;=`%oHl@^pyuP=K>MvuEHF?w}b^Ol##rZG%^P2ltYYsIil
z==e7)Wt=qW<U~%Ex=HB<IPi=_@2f_Jy7B~PFrRQl^q^2nIC*aAi>HVNPNW_LQLyVU
zgpk*)1q_edp5Yn?rVT?;+f)$Y+YGe&k0q}E1glT6SL+pmd16!+6c5%Zr=m31dV!Of
zm}w(QqxQ-~Tu6Fo^X%z=QQH`+kkjiE@g6B|)I5)5K|%RW7L<IfU}?Ue4+LDPEM(l+
zhD5I4z}i(r^y7Yu#5nt3nKLvhg7{3E5ESoMq38UsWN$IcrvKG^GLC+hePB?|k)*z5
zAtr6ihr9SZZY>yr`_QKb`SqHdEyM)AZLH&#g=>bWv9h@KWM^4_tM>>tu)&T?{@8Xf
zTlaSA<Vo7=M%+DllF~ve9NY59aQfuEoROL8jQ-?0>ny{p{B-LjI0tb-A&tuCZ|Wi|
z#@W(nDA#?1t?keH!vC4^2am!aa2sLZ99nmd!+7P45tM)7aTq{uGXZ6r>Amt?Lp}RA
z3<Q^(+Eu{LM>S$Y8N8KL0oJn~G2TWFUbp6PL8(o{?LFeSkcj7*wrP8uB(;^ZIw=mc
zw8R5TpE19)m#7>^m1yKF#W#;7rh2AGP|zQ@)2K0f0R(Oz6>yi~O)(RfppqaIm#*eN
zDSzc!%}#4K5Z>o0&I`nN{BH?Gg>Zw^9arqCE-F;$0;=BIMPI+4e~*uIPDmhd7Z)iQ
z+vD-fcxJw_10S9SKHRRZ?^hQ)sRqlOCM$;f$AK`x@tZgXjUFr!Gm?h;m*KZd&V|9}
z8Xv{El=xWu?a=DYT9N|=X?!qCv{_q4VSf#1q)2O}Noypi#1<`~y$a)4@44ZD1NsEs
zwB7uE|LclpVJ~)`AClBq#lX8r3I=J{ZIA_jLrRUE75ZwI+S(6uCedj<_ElK}K3Z&o
zGFpzyK$+T`Od1AY9J)!LpjaHIaXxNml6aY=vS~U<#gon#J4Xj~)tVVat*%OC!GErk
z-bK)$e-J?1MhKtiRn~FeK1r~XO?1dU$$UENH}7?lJLWlc`d#gUcaSZbx>3`xWw0N-
zE&2-cCb&yzZoA%>`_;EqHKdSC8$TFwOv2U=FRL)22>V(;2*Et5;rpM%<La*Gn*FhF
zV=%>I;|B6ij!hV*wHu63%%GidQGe;O0t!p&ukTMoH}ds%)ImW9t9!1+6j`j18upew
z@!O7abal<k86=O1ZhW#PLRv<4Pq-9Ev)TxSgMzWU6|YC!Y`D!I@M3DrzN4LE%F8x2
z#-Q3&8!U#g;>-%sbC~+WN5@()x;4EFkuQ~SD`5tXIlU|0q>e8esV*Cf+kZf1BA3lN
z`h!1c2MKV2U+&4UcT$IwhE&YLwL3034)#vXa<u)S#SBSVeOuIz%8sf}4mgWu>+H4|
z&Xt_Ljb|WgV3?M7uVl?&+Trlwi&7hB&>GK03ZL1TueZx?R#eu488ODZ$$FLmiLP}L
z)@md)OYg?opnH&@IP>V($$x>*hO=FQIvQs+@JNaug|o}%Ib@qx09GN>s!p*F1XyJW
zc=%ZWtE^>Go*Q60u7?4(BYxW%_TXA``kTNt*m}=42f>w?61HFQso<(0LnsN}4z3-C
zIdJ`e025N;79#uI&jqWmLn=2oyS?pCMO4m^g?cUvh^o*E^;03L;(waqu5w01mDV%q
z-;byno#I^ar{HP%_A1VKgRLyzaD$ZwxWIoHvM!qGkd0~u@fygIjP=%3#}5QqiPoym
z0@;`y=-eRNaXk#O9r4?aa}I2~2bt5sHbZBx%^|S$%uCa+_*AeJUSje+VB2w+1=|me
zIiUno&h~=5A9wAMxPOb7b8zE+|5Jh2Gfh2{1>6-@GcSHB+!fj~?@q&AWGs)E6x__=
zkPr?QA5C3fufyM)HJb1wEpE~#(XVkib@DRNh#~<sj=i0Rtv310Aqh6~*2q~&P*d<x
zxASay_;H_{f$6IY$r2vh$RRsGER4e*SH*>*LzN=L<ftC$_kV&cMz+gNxGQ<A%$Bbb
z#J1AE9&Vw`CLCL*!PIDHvsbt^!$G^vvY!Ky_v}gtW90%Ke;eS6w?p(tgoN}?5C08u
zn32%R2k$4cAqu?2PY96J#XnDvzdZbb@?m)TH;@9<nb73j@RGV3eW#roo^v0fNkx7j
z=D9Su;ltIW1%E+VSO*@46(EX{)UFwQ4x<PZr{cq8SuBZ3X9Ho`pcyJM1desI?DyOb
z7OBQ9qE1ze&<kAh+R(-NvMO}Lr|lO~tomzMJr1k8Y*lT94J3{y#@^s(&Z<CaTyChf
zY#EYBGuA6ptl}-)WMQ<`w+;%1#HMf!M0&BvDDc#>TYuHLj5K<eA@C|3%1~+h@dZ0{
zx?-d{sYVXMI*qHH4K7OAy2--h-w!W&-4j;bHf`dw4oO+zmbfOPPz30}E0&MQa|<~<
zkQeF_vLzIKtiKY_rJYI+ZzHT@f-UccMx><MdQ|T2XoC=;x>ZLQq_5x_L%`n&iJ}-{
zx$25Kpnvj3Mq13iQp{!FII6bkeSM@$&|g>O5~gNnoilLhyAIH8<ul3am0!xp=*d3A
z$N3(tv_1{23Y!WA>HFifo$wo}ODz|^XO>ZHRmI_TE-O@MPFFwj@r`W0xjf<s_%2YI
z%Wl1z@EFb0mu>Obvy=2UU&=DZx~#1{b^UH#T7Q_9%K6$dokg53%zhJ<nBsFe%PCq<
zFkfZf(O@pu<8FG4wN}r^+uhjaf_4m#xu6{bFa_=3!;~z8>$*YLBmRMbI00bCbhF!x
zCB15k{N~1U*iuK-HS?E^{YpnQPaEMov!$KtIp{1soKIY~XHs`HnYuh_j?!K=?;+RY
zeP{IO$A<DBAp)6K*0+WT@u-bVyu}`VYhXMiQXR8tzDQ3-zUKFb_*Az4{y+eAIEeMN
z{x4sQmW*=M)&2_)8Pdcr{YOS!{yO;$L(5RvH4K-YFK~Y`xY#v2A77mw_ke>l`yV~t
zIhS!`9uosNGdGtZdM^~0CzU!UmoSqE1b^c;48G5=@E=%Jq)0u0!2pxY&H}x4F}I?p
zqPwR(wEzE5N|t0>vYo_99|y;YY*QrliH|SV!}mcCU!G1MPd~jI8l3VB4u_9#1IPs}
zpP*&X^56iJH86bq7=C@$8m$~BIY?_`<Z$xc%b8z38SrukupDv~;N_Deh|kD927kwA
zjUAscau=TE403*r<MMpaR!0f1S%P`_?c?Vsot14hvyPO=s+!3Y^sY6FbP|o2U%<&Y
z%M3($TxR4jUQhSRZ~Y!Mk9VdK9PcJq?1>{uM!LU_Mro%gsgCIUv|QmVe|GgX&z_%y
zx6w8gY&0*PzBIo+e?0x!8%?a(nSbH;A5ZZD8BoUK09FNKhrj+9zCHcYO3nV$O3NdP
z<A;3v>-*4*e9eFeS^%M~WnKURw4j6r%b+}k;`U@zG~HxZaWrzVa6i3!Sd~axpvA33
zd>b7o0tI5(O%!!3A4&Da$k!P>*eT1Vl`3#t1(O98wCi-R)551EqiiV@aDT~R*Osu_
zyuYoite8crUZgDB6);W!i&rX)OaeB>ZABi;De8Et0M1*0-pl=D=c7sIRX`?7@;Y@o
zBMQhETAi2c6g_;c-|A{?c}b)x1E!rZN@T3o+PN=go80eOV~ke-54J{*_Z-aVB`Orf
z#JGt2h7zFy?OSE4d(u|i1b>l)GUbD>CX~nrqgwHMjEa29YTNvq$0IrHyz8twlS2s;
z_$X)fnkJ`Rt?arpk;R#=RrdAW7Q`6$0!+_i+>cu+>QDm1)sYyDVjBiFtO9b4#K6kB
zAe%)6pbL?>Ij5+*5hKwmW<Rh11!oIzXsKnjPGf?qdxdU}uPBf(0e@SK;B883HQ|Vx
za*h#4C8>*%uU|H^4T?KMXZ1QBF69|CPH<sbY5aLCMPdP9b=tJzO>tZv!4j8C6t=hf
z%6o-qJ6}OLR&!E}e!+Lv>vPd6#qQIztzwxmTClVUt4&8TRl&KI+QPZj%xH5==3Mfm
z-?H1ZNxqBUsL$y2`hToQn^rVmZSqqYotIrnHv5t}Xa2~gT@634b<QRBiwrx3gOoAP
zV`fdiCwsW)M@vn1vyJo0dv3kBvIjR)grEVO?Ii4aZ&6RIcrM-Zx-5uOm62>4QU7A}
zb3y)C&Qq!!Vp>akg35~5;&{&!Y)COeYlqoSkOhSlSZTzC^?wTuYu%(XCJGkEI*3a#
z?wTqpVL6PRC>SEQXbTp#nQv;d<iXnfj75Jz-E`-y?poC^j#fO_ixruXh{h7+8n2{G
z5cPQ*wdkXLiXyv}mzU@+5?g<tOBL!?n;$AsK@e%5NAKbcJ+(V1J9WM@M=!gnmsn)B
z^D^Gl(nylx9)Dh(*g;&Q)(o?yTgeUDq2w>^t0XXJlRwCrt-GoZE3Mbmhn0~FeORst
zh*?RmQ&;G1)^?spOA)T?!`LcgLML$}+uqWPc}V4BBN_F2FDB)zan#*?OD}dqZZ@$}
z>-@NplS>WJ>UKvRW<KAnv(c=OIxT>jy6XmsV3cF*B!BXHWddGMYP-7+SQ1eT9WuZk
zx5UJV)u9L6x+NwkbUo(vq8>b8m+U1By@%e`wXm|`C5)`>JkjFjxt)=&nXa8%A|%^B
zj&SwD&GQGz@ZMg)(rpr@db(uLEdog!rNbrIO?Ss6(t$*mK*(m8KGM94k$E62L7Pay
z)UhY7RDWNW{6-q5K=Wwei46jtr(h~ESGGu8?opd`hUt*jp|loFK|YSwKwR`W)g~ot
zl^O)QvT8SYk0k3#V^78?G+_~ZcfD4$1X{F2pT~BU7Ncn-f&YFzQ5rW72lV8HMx5Jf
zdt9S}JAhQy%^T@l(&lw+%fwj=Jl7|Lw4sY9XMbgDLfLsl;1<eabA1mEj8+~nwM536
zS^E0~m^zm(*4!@K_#i=pW}%@9ca8ANb=;*Q&RC~NLgwR&xK`py3^<V1N+H)s-$*Us
zt%{y(RN+He1!4~Hrki>_WTOOlFs_?h9}d|uqcd*R6I&&|tfKKUEKiFTk@wP==Ucn-
z^?%DY1zTE<$inMIua|;H;XLfr&8+KoxvRl|%8PPsS98UmOIJ{s7=i+zdrRj+aV4TI
z#+s2ko+lv+^D6FM4v#`Izfu1JhldtrWkdVaH5{ICcd&J6jqH|i6I(Z9IVvAfIE?e`
z?53bp0#sodCT`^VWi3p>jJq1yg_ijn8Gi;pV^`*h)Qu%nvaHhj`|ODoT$62mmHocS
zj5ygGy@!_tQ0C)KFQ)!0HT|Q9Gp>!d|43ft<;E-PyxcW>2$J+AVE!pS#FCU3)CCR~
z^ECV`bO<6z`^OK=q^-oq(IJiG3v@_(_diRAG_zfxLt5Pq$Mmf4@jr~ppC+wFVSo8v
z$5f48g#{x;p>wU!DfR5-og?f)?%>-+*ze$1fO@EV7q}U`cJAEkw{st74`PdAb#MD|
zZYzc%--m{K=mXS}&3hlMvt=ZANgdkWt<N*Y?EV%4b$o^3kvwoKvHUxd-yum0d|GcU
z4y-%Ib&uJ2IqUce+2%`N(Wj(4U4QSOW*~q9`md>(&2Y6w(WJt6VrlUI!P0=2P2&Im
zEDhM8(A-VZP$p^kZjuJrD|G)ZNwb5u2az>EF1^du@9WB|tZ(&0lNkwNOh+cWKN3ry
zr^q~#%xNV)mdt6S?~^$zs@y^5v{UmgnbXd#4=0&M66(XH0LD^)dnfw)u{vDfB-YdP
zqQs`;`++NXU=9z&Z6olpuw4q@M{Tdj9iVL{-p6b+^Kt%Bdg#mOy5=oIY25z?y?>-{
zml1I>6ahJxQ8h6nmkuB&3ztRtDky*5T5WF}w-NrXUqPT>(hug6!<ivj1OlA6PE$C5
zVLL!tK?wA6mnfBW!s!$^{q=o@r`4T&Yj~}b>T^E$Jj<Qsym0u;45>mzTyTYox>W4{
z>`HqsE)p40T~r29gQK8EL`U3s$=PU0Av%r;M2t>i-2)Fh$zsgna>z!gt4)7G%`S#)
zGAd-G&?pGuWHW>WaVkbJD6oJM7c)0O3aZc^D9lobZ49#%RVO84tcLal=BpAgMiEpc
z8#DcreG!YJXc{C+jwq{*!4`W2g)tU|jL43(*nCkOthFy9{1n9I!y{(0FCZfkexN}G
zrEQE{GR4I{ONq06n?b>!cng0%A%P4ilr%*fGfLZAajAwjrjD07Q+7;hnu3{AHNHfS
z89$I)KxN~D6Ir5p0~)YMfegZ1H9sfDw>jA?TpbrSJCcBETH%F6GPx-`sMqEzPPo{5
zq>3etMsadd*{ozNb1iWX(vs8A{Ko>Ststw@Xp3*=Yh;swz|mMu1{Qx*sja!-ZEBV)
zIc1BRgvC>7ev#-@m0<_muDP1_0%?`(3n-a`njL~jqqfpgT^2Gv4WRI-Y?L;iY6J!C
z88d)twu!o&DMN;4mrGkgxD&#fRtZ###I_Nd*UZ4BUyLO;T`Z(BCV&MHR6-jwO3dvR
zG^HlU$e^mZJ;RPtGdzC;S7F852MLrmMKKmbQzg4f$=;TTUBo69UaF=ysOqXVH)H2k
z<*@hl*WI%N?9YU){i^%vU;oCAQ*WCjV|g#G&(9C{_XoK$xDd1|o)H5hI3@b1V>nYG
zPw>GAPR$-cHG-3fe8u)1!5NB2oUC!W?%Ub<Z2tQ4=w#yB2d;m6d9+we<`<5ZkAHu+
z`0n-MXfbhdIq-aTv4D){*pV?5Zc?5@(@4A&`5bT<1rUUX{r&Fce0K7BvN-5oK6~E1
znY>%vDEa1(%L#=2GJy^ki^;{}%J8X4?Osi;X4mtR$<-3FZ0Lu{>DkdYvv&vfs6eV*
z=@8$~@uk7alIwrz#l@^m6ys9--a*4*)4bvE4WT`ed!qD2HRAAA!!~30`uccbPcP0c
z{^PoDX7kg@ylD~+yMJ`wci$ew<-@DV$>JbZ^=gomeL_xw6km`hq#~k34!frgv%l`X
zo4uKJ&)iq4Ig|{J97_nmwLpK}-(T}m(_vUFT<u~n7Lk7-*g6)E2(l;$48R`n&j*T!
zdv~npp5D%WvN)SvK%XC9eQ*E1`gO6my!u<${c^VW_4?SK%-(jtUBA72ayt3F`{(r!
zFa2-l-SPSCxSLLk){`mn<Z+DCm`*uOIZg5;AD`w^rfHaRoz~c0%}@F-|MDUtQl8*9
z1z5TFzL|gbb2W4D0b83gdjgIy7aw!paCPAv30Q`d%~CZ2hPQGx2gtkiG{=Aq_9yi;
z*XXTyLFwZOd0I;G$fKwPf-kb^Y9t!12xTdi1(g9HX)9MpD&BHq@9dZusFpsrv)>+_
z%x724-4f-=3rW0kHh|db!XAi`2RC!iIzx}tNXUP_)-;AQN1usCa7rp(^O!k>*mD>&
zXF@Dk9WsJ5XCKO#Ib#$rs5(b*21FwBW8|~76!u<n6GUF5dY|ML>qc^m`98^w(rt3H
zn!P11E7@-d?g{A$-9D$W0j&ibDUcO)eodaRD4N8Vz>mKDS=9&BkhfTyF)p-2+g~L<
zd4Yda{vcSpbT(>RAJO&n&PK)Uvpzo5LVU31^2XN`FdoV-`I;(x&E)}oU0`!bx2{I5
zmgA>oHG`>GMzkAuxt3tYl7?V!h!a3w;+Q$nnIlDy;#9N-cUqXoX${uUbE`ffg$*dB
zavvy7yn<5dfs=aRB<+bjfS{jHkux%CTrYpggRgOyY8Uj**ifVoSSWHrdPail3q|IL
z8ofR`dJ$q-GOy4piw{|LsZdNnFSKP3=<Fz{n(E5g5t)Kg{AB#P*_$7(R@WF-l_mO3
z<uRNwduzu(&PlK*laAp;Sh46AQw8$o5eMJvh=c12x>Ao=sYk4od!qD2-3r4rAVGhv
z6<~^HH|$gaCI$VM27X8ktHs+HYrm}!4G495bPP)jXRivdgnSY5PWf6BAgs`kuRo>G
z1T>>9S{0gha<w}LC?$j^_{rN@si-K?g57SM+A7PPv5}DZfev?1EjW&;V|n2406Ry7
z$8cs;+_jG2OqO8sm^snv`m}LqKg54&wXxmRjNsJlbNRm>n{Sn1q--P@(fcGArB#BF
zaFazO_C$a1ApOCE^!2mI_U8>LR!@e!h^JWXS`Y|aEe)ZH<bfVIDY78?B1sfuk)%9;
zz@M51lzhpA`d!e6s{=Nn%0l_hz7Cx($E;dQt%dC-jg#c<b$~U#*bQV#-Y$PmDP_-g
z)yo#ZG;qW%ApmEZ3XuwtJb)6&dy2r4P>~0r1iDp#w_S82F>UA&3Ck9dK$%o31w7^L
zYI=duVO4w(Fcs<cbA#x~?DCHWq7;x0QHlNu6UFrGeDd>jHh+7xxM>w(YJTsKTk=QR
zhBbzd9C8nFX6qbf1Si=>-8_G09^y;^sGh%=wm~;hUGIhJdF{kQd+634x>dG2=-A1t
z$JyoL?CsfqC-a}r;LrIK3%uO5)-X`ra@X$U9pHydPFB&Tpxc>Df|9V@LE}_**X9Vs
zO6{h14YD+ksc>~PZfPESIXA%hfy=oe&f&|s0nQIx&JA%6U(V?ez7|?vNXBqRSd_jT
ze9fEqn(u|LW%ZXQ_V-EKplPDes;t{%08+12KleT}uN|FTOy*bJv&qHP<mtP!tEcn%
z(Xtdmu?|p;a(B1oj}-`cVEY7A->U%rAA97})|cT;F%tnZmtm+v6PFigJ}ZAquPY}E
z-k)DFf8bIK<|SQ4I>|{Yb=Rx2tGcRsyXvC-|BD(hGr+tuN!l**B8>U4&1V~s!ykhj
zetbKBzWs31I5^=AI5m8JAAtUl^bV35tr;AEuo{NXKZjqwN~tWp&h$blwWgP&@18!*
z>5B$WJ;2b5{slOFaR~7l>Z5;EepcK3jHX`dLZ3jN@3Ebp57Npg;hiPm>DSMn-Xw#S
zBG~vfe)PYtpKpJ?jnI`A#)iRoE3~!4Z-2hUKM;(tE(}2ni(&Zt_u>27&%M-fEJ=^#
zHzS9?{}{TKZ>+o^NqN#*V;bctD~$1jHC|{E%*w+LDGezLZ}g(cBTau&K{Lr8ZXOmT
z8k&m^Gl?X@nEXoLFIk!}dYIl4{HBHA&#<Bss9lkq%qU5!%UDhPPS_1K8p~}e|BfT$
z?exXUV(zA=&+cN*{FLS24Q_f}sXb)vM9Hb0WC67Seac%#l|CUkSR5{XwK5K_=$Cot
zK_ZBAfs|T+CljHa$Nql|F!qn%;iL1p1Byg)J7)%~tw86;6|<|rf-?HNycJ~n8qs7m
zOs%OUDos9*QyF0(Jb=?Qh%lo9^Ejg7f`GJeM?hGNHk#2*UeT%rafCzQLP&6mBIWc{
z^0!O&#XWzcxj*F&S_%(pu#)&js&QQ*d#^ry0VOdu`LXHQ!AyTpL~eR(FkBgInFfij
zw5Mimthe-O15G7ES^DSO$YAq;QdrqE%Okp2axc%;Nei;$#Kf|zA<EF?2}F??Lku-d
z+;}$!rBPS$m(f6p^cQ<6b9(f$7!adwm7Tb;`07YOenJ)d&e0}TI|O(+f-6#Vv`Gv2
z{T3rFxTq+10uz6IBC)T|g968+qrn!95+Pt8Y!QXmYA4uspDD1egh6Hs^dP{uPQ=SQ
z1=#ev(8l8qkx4>x@QJxf{8U#~5yoVoyro6av{dU;A*q9=)ZvEP0>zLrInPigOKn=d
z{kWQTD1(9sy&ELLj#Ef0Omp_(@5(xq3&Me$sry=#!HR!)0&=jzk^goh6z(ti%nnX1
zgTOFzi^$+s9yXR%N+|np8%t{=Xr4<mN7K4=;t&44LIS%Q3dGd35SJ`jxi@R8PZcQe
z3Z=$Wp^d|b^$V8xI_O?%N+MpTU$8*{(O)-8NpqMY?#jKXj)0k<ls;Tpb*^%>QX-^p
zQ{_-vwjh5Ob~8b!F%vJz7`i(m_FnWw`UGhwZrBVVpuC_J+CO!)E+F-cwaO`lp2~aq
z-#&5YXG*o?yOC159(zTCPuW`-CAm7HZN_vJXboC5(~8^Fik6gQ1=l~7W^gVM{-jb~
z>PJ*XC%1amF>mELSPosLE=uaB+ZpDOYCVm5;T(TA{+lIt7En--*-%!q2Ypskh7#Li
z0dIT?pv`!&I$j#9;g8~VK&)e+(Ntw?)Uq$qE+rEjWv+U*qoM(AKSq+^N~xFX(y>UC
z9Q50QE48$x4NR}LUy6P^=@+qvMZbj0z>Zfs;%GB_^2++)bS2HAuZXrWX{zzPo=ZNn
z6kC66VM1BG?3LT=Nc$iH%;(ynizV%3m$8UlLyKAojH|Pinl_d!aAuR%mCL@#*j+Uv
z7hB=$Y<?>9Lwc!c#i6UE7`5$r?A<z0q?z_)Hq*^Mz2qujIqK(1fSc|YyX}@PX5_91
z9o<_YcI2Xz{mR74U`L9N3-!J9*r0>b16zMmRLXqiXH+&*)z40JSV$ko9&DJ>-|RXk
zq7vH8%w#^QVoc@blv1H>{h>4DlwnbSc}}hh1DAgO(9YjZLA0diY`>Y4c_%1aEZ}R!
zL1OvZHBzgp;s*b!O$%y0CIc#7XxjTa09@Fnc}ST9OtMc&*r$KHWw-`m(n=KsZ_R(y
z@+3A1D3+TBdM%ZmAI)qT3XhkFi)Hib<GC_%&uOR(4(NJ#D$Rb=cd4R_`#J4{aULlq
zonJAF(@2l|6B)gcR)L}rwGcq^-a?b2q;r7+G_UBobq4oRlDMYVtiCUsMpZ)7j*a`F
z)&XVOwx3VNx)a_tRRgmplNX1=hy{O9;=?@;reGIcq>~re%&Tm*AM`Divw$<nm&RTd
zp|H+Brc^es!b}rTG)jfamWLQ8w3)WaJi8yZY*#iSj&3(mek4LL{dN;j<%ClGCx<&)
zqj1xK$qNp5dWj3A>X9<rIbi{MsY<@C#Lk6rKfv({TS=;<Vp7sdicT)i$2Wi5;*lj)
z1f8^QZ(-_9o-|WER=ZmVMy7h%TAYn+mPe-iHfGKJtS=92FM>ej=BA#x;327bJF%b4
z@&Us5khLNhJ)6{RD!z_SkGCQ3G$=81aPTUV@&;u;wp|>XSWN^UUSQI)rTJQumaWa1
zgKpZiY*oI}rm+-t{$86>BoTiSXH<jcA)9JTHIOv-+SCR?G5r-bB}R@1RWG(FZ|%-z
zvuRT#%9X*a&g%H-t)|fI=I7pHu6A8IhTwL~<*?*3SmpnxWL;gR3bQ_-t$J)8<*wfk
zP}%Zjb7!@Zeey6kZZx`-<@xE%JqEb;j33x?WCupX-ccQIKYdorSR{W4Z()YT+>OP)
z3=0`AomFn@#cW_%JJ#M$_4e{SWj%2^zA(@0C9YlKIgsmlNQcv%<Sy46dq>7_IO~1a
zvGwr?%4tqF8o)|Y<ODd~_kOB4FB_$oW<%`>s#C^OKs2ua*WN;zYVqdc)AG%m$JkVD
zuxaK=`2TrQVOl<;g0Fu|T!3=C)uQD71(yd<XsU0d+dyGgQM`_)l`;a*Hz>@VDD1z-
zij+Sn{Yp}71uiTVeSMr48{sQSCYQ4vkjGL?2wr%v`+=7;%u@b#Bg=>R2?C}0(;cN%
zH~6pad_@~q3+V-SzT$;92<}+u;<FabF<`U*U3b9%$h{=I>MnoS9N8ET7QwCTOltBX
z;*AzueeX+FS4x(5_YUvJr9)Wgx8qcd{CMp47kQPVKA;B}FHkfs-|o60=HrFP$v^J5
z`HU6#9!V=O?Z2ku%!fItLn!O>=9RQ9b4rI~I+*@=c&qF%ma@||!*V=3%=um`Xm4~3
zBIq7AHuWI#j&n6S;x*Y?m9Fk|h9h-2?~?Soe7vxax3Ba}H7h$Lyr0sIj?A)-(?D6}
zz4xp}TI~M-`Gp%>m*Gt@6PJ+NG!&Oz%rVJ-0^I-Y^)ClmgsqZkBf!PMc`J}sD;dDO
zCXN;*vPc0u-6zV_YUxhX9rZ<tK$Sk&*B<jTjRn#s5+!&jZ>ZN}9nXDzz9)KVoJ$ga
zIt8_4Qw>UJr+Uflv5Z=t2@N1(2>mE|TIc1I@KVosJ+1R{PFNimc->hUX>Jt%?(dv`
zh0n##$rjaSu~_tk_n}&bZO0ak<&ENT7JUMat6CWs1Xw%Mis+zMf=P3?rXjk50PT7T
zd4AbzswT3u9IK4Ja5F4DYfidlpQ@voW~lFHmZ^Plq>Z)TIFfXkdA`#^20i-EId4q!
zlwwFkM`1*lm+tBL_3-<gg)t(UFnoJ|JJ1WOKsX-<<*?NafBb#;e)!T;ZIeoxKIt65
zB&ld80MM1K6k0(IU2{oD;#@rDYYJ~PrFqVMa2PwKDn3dc>tyn`r?x};R(+nfBFA>z
z#HTW|=IsR;By-S`$Ec^AH)-Z+Oqz4zd0d*KV$HMpnI}r(%*1L+%CE^|W$`Y5<vB(j
zXXqp=6!|0f3mk!=L_lhp_;BuXN$$)&k3CoNNctr&eHh2s2|v&*_nbRNP;!Igs0&3@
z(~F}iKdFgLR`W~5Mznut)zRrzbBk3Nm)`fIOAPii2O(e()XHimy0Z3g<Xp3?+ZD6H
zo}t>=K0mKqOB25nTbSp{kNJ9k?kTGlmYBh#?X|x^kJYRW22skK>hrw;j^$&n=7|g3
zXPteW9n6BW3DQ_X!BY<WeZYWx8B$bYpqdz9`1g>Sh0<UnhOJY|I|^Te|As%X<EMZA
z`2NecUzNaN|Dvv#NCkp`7>2j}2v~%yMR@hIJUgVV5q2pNF_uXop(EjcFCJJMz6}S9
z*vuNOn{Gd;)d(Ljj+?=hzJt|fWY9FR42dGQGQXc~BTF^1NIF$9=!vM5vGj=Jc@#a;
zuYKmOIO@0I@O?OZDOLriJ#`^2Atgopw`5i1t$bskx3S515q-2SD}fyx&gx?0MBCD)
zQE6vQis~3WRX=7`73P+IMrm$BQcQuH_}aIhF+!&(wp0%VjG@7TP(Hb*DwfUG9q+%r
zm0>?{&`Wd8WbTxSLLsF)a8%eW(MshTXLhOPl<wS{GnXn+<r~D3IdK(hqpKr=EWZqm
zNJ+K3QMWIbngb^Vz+PaEenHWg0a1^^^ucC0Q+ing6cRn9n&E+e6JSoVILBQHbHVp;
zE!^i+|5*Doc=MC~u0GWfr7)Zt(8mjvQ%=m8;u7F4Z|u8E7K6D^M%C+@mS<0i)@oX|
z=<1ZkrlSeFl{DN+qFCxBJ*ga3Sk}OZ0=d_8u$mS%x|iY|*)b#HEcb&50eWB2&6!Ec
z^|MjSO_a`x$+4b)j;?~(XHAvrdc{#0K*NZyh$f*~qsQuL3ffEnT2;h(b`>a%P%`vf
zYy-WECS^)mM38#mG%7}35B2TR%CALz83k)h_!{FiWq&N}l=K%`zrLJb`-Uijn#+F~
z#&{2T8>5ABx~I-7t(A;7W|xJGH>R(g@%}MPdlgV}*PxPrwN%X*Mq+XCraDSl)G!>b
z86@1cOpF>YzF#FGabdBZtu7NWA*ZZN|AEYtJZbq_mXcAfEhTNx!b%wTH_FQTG_DV{
z>gM&#;*_wFi9JPq!NRm8QPi4b=l_yQla{Z|E>;X81k-ctya-XH$DxX_f-Dp#Ee*vK
zwZ_lnH&=Uqex+y$_^F;EWs$t}zFKSTupq>_jb*3x6%@S^TF(4H*$osySd8zU;?o9-
zQFt|*8ReF(9t<MXxz(E;jjQdBJ|JS))9yyb6yy8Z-O!=}_Ov?|7mN{iq1_SCr`_2;
zqsVr@=mwuR7IUSUMY?=mmI5>`$MjNOtNTP){l@=)67NY%n`JmlVw6}r!Z*wZF(ykC
zy^I`p%z1AmL9ge$PtTb{(I~t+xiB+V$YrpKE7a1VwL;y6RR{=`mNrjPz4hDq*eF1U
z#seDS^)_thV?wHn%x6Np-o}p4`-Sk`DyW=rps%QcBY9aX%cSM&1(kBjRUdDTROPG)
z*3T$^-kzUuVJ!@}p5li5gzHh-<C#&sJwK_S(l@rAqPyJ>*&oslr0JJyGs@dN&tSFk
zeoy?>QT$;1)p6SafA#Wj$6p=Aw2Iq;!JLRN(|i_d@8^deG4&LW%UTsQ)N8c`-{|Os
z$K~37IvI>0ZOw%`F&gTlVKa8Xy5i@iIs+YlaGi<9T?$Ew&Zf$!J%Zk0;sU>C&`&AC
z<IPaV0G8r7mU=plAjglRyzNrAWsYNL=1;Tr!=3MXhVuFDk+*xwGc=n$$McX$p0YIQ
z+I)6&k~#K=4cKkk{lc<Wm;z<oD=Z&-HvlpMspvE0L(x<Kqu;e!{g7(g5I$RQZjeZS
z2Qg0P6#~KD0<nY9szQ$#r8)mlCd@adPbOxe6DUQV2?E#uJp^tfMWy+01Wp-RN&Ddl
zTn7-!!k>=7wKGENJyT*w@q=+t$8853)XTpe2Xz#8*oW$6+`bRh+YvVJL-iE5??d$z
z*Y}}%tM9Q7wV&M?dG$JN--p`I?z|6wb(!6HAF9tNd+kH@^SQncwUE~~>_cfF-Tn{G
zZi7PFDuMpK8Jxjx;0%+xFt8PzUGw>pfRj9K260+NVWQdy;^2P-#9`r~11TR4u2l5G
zdi$q?D}|*H!0iaGI*K0*t~zczfU92q?cl1TxC6NAW!w&~dOL3a$k$Wc4z7BCitFI2
zxB4F7YCpR(xZ2O|46gRGJA<ps?9Sk-@BLiue#<6Z-goP1o@Tc1!PP=u+Ya-*7vN@d
zb`ItRmyRqiNU{Ou(dQh9zCDzlPMR<vyBV<M^ZgaGEyIZf7eafuWexXL5OStGyLx_q
z83AIsYyV<EO^bj1JB%$w9<Yvo;S{Aa&J{aTavpIVqdY5X9oFORY!!L~cRI@F;|g~;
z%iV3~Zj|tPg}FP++;3;DrSMKoxWk32@4y{%aFBY1`+42NX6^)FteG%(7c%c*;PN3b
z3{*qsyA9?nIz8_q>Y7Dcdb4usul_}h6-HN#3XF;hipzf-H5IM9KEECY{sSbUa_5)f
zO)(PzG?ySmCKZ?Q?gS^7YSuBG0e!a!*D=%sf4^CBDy=Cgxz1(zG1tNcpHXfzNIj6z
zbxr`dOLIO`BH^iq>Wt>VJA*Q9>nmvv9A}bHSu4x5=D<<?N>xYGI%w`Vn(Y-CnmJ>9
zsaq##qibHj{`M($QPW)vb%DCeFQ4ia29$^L6GmG-d3j!4Z&Fq<u&RunbLWB<VsiA0
ze|`x@<dA!a2=Q3b1^8aZBgTo9kaamrHNlw$aZpB5s?5L$2oQnbYA{f~H2e8zzdm6`
z)*lhimv%oN1SWZ@b1LHb(lvb6QNbD{P{J{q$FGM7YJzVJ2x?|4c%T{TN5O$c=5;jC
z$bC5&s0poLf!;CR4GL_sYC-~a<Ld!|fAeQrMZmrb08mZHo<q<v>pCZYyqq9XE>~^L
z&p`nyhZS$#^@0`ggm12=7Xu75UQsVTGQ~!<7)Rw>7fZ)cN;o6=rz8xd>+<a}d|BjW
z5g}K!5P{X`tJSdJky4W|O-mb>(jU@tQoP3)7Ah|5%uLq&G)NQ9P!7%WvU;;Fe=xBO
zVuvbyNF7P45S)4+6i5)18>URDL07ewr1aljOHvD8VjnxhUm}sDgo_|AkO;3WQS>lr
zk<P06yb_n0qs7c<HK@{y|6~#6slcE3IPgcXoXUi^2Y*saO6mH|!JkAb7|r{GKj}XZ
z=_`W2n&2CQzna+!_-lsxQQ)tUe|a7JHF94L{%S%i;IDU#cLRT$teU`I-S{T}e~t9G
z8~il{_k+KN;4=7YoTm7|zGWU|C3C`Uv$di~fG*sxw8(>gvjS{W3W8xycae154gVyN
zt@;1(??SM}X*y8*;U8WN{t0Ko1ho_X)j>9Be%kE^XqG&12TT-}YBjMJe<tEj0u%kD
z>&nN|H-m3NBcU#R9(=nZjH4Nwaci%L;~I)@4CESaD@d*#?MFemR_b*u*Ghglm}@Al
zpt%YEcsDq=&9({8HLbG@X$NQoaLJw*SkEsONaZ$I?<}f<Gl1SFb}VuU#iZ`MeGdQ>
zmTA#jWCj6BzbC2l+426_e+xkJfN8f!PWqC%KXTG4<Q4T9mmxRmb68`XSLY}UGw_?(
z^n4DsrRhPLPTeybaWi-!!<VJ2I%|r{vu4^&UXx!iJrES}5mmKMAU0KPDH*8aWvQHk
zOmWvRLBI3tG@Z$%Vc=D|O{&MUKLC#iZ~Xo{?s5Ikze13cn_QYme?8hahI@*Pu*ag*
z8Pf<@nzB7?*nqWEtK=DN22l=Gi<qriJXni9ZVQEkSM!##J-C!`qR;8TnSip><7r%r
zq{>9`3>`(e>{b-c=MK&>fpUR6&T`lLxqB?+)B<-s%iZkfZfN$GSaHXV-0gnuGzZDi
z62?=k+qQ6T7k?*rWeYM27HR1*Wm>Dl|4F!rvQ51y31D_>q_1&CzM9e;Mr5e;#6L}z
z|GoyCkPU}R{=SwL?0o#@wpeY>T>I;qrBpiFD8BQbYrR0j-mzAuxqaU<)Q^=+7yfIl
zDJ!2|Z~g-$U38amV;&O%I5d|~lNS`1$FV3Vmx<pom;tc2BH%IT3InWrLYL99FcSed
zm(WWE6az6eIhVo02qu?+Pb~zOV(T%K0sNQu>oMnl%4wP{#W`Y7;FdL?OQXf&QZf{r
z0)ywpg?SHs<oPi#L_#SpF0NAMCQ;-e+EgBo5+J=*n$@>KpH>DI`$x-6Wx$2Jg^M{`
zS#Vx!04YG!aLkeR<s?Pqk^q1}f4{9HUL?VbsiGJq3B(I5fk~&)nQLpMsHj{Bb*=1+
zn<UBQVws|`e}v>1pJnAO7tr|-Ed`XlRq|WWkd8S#PpRBQO{>IMEXiz@glj|b1{ZE9
zxfPeCLZ(p}bj|Lw0^7?*Sxz&`xuv*~_)m;eaa+0uD5!w8-o1Q<^;8BsI4KA$ryVP&
z?RkmbP!^5IV4x;8TNT&1$}8`&p-P)64(_Zd9`B%-f9~R%BG0=lwq?{B{WR|yOUlvo
z_B_Ph+gwxve@b=6M~!aN6Zo2WI}|1+fAqX3D{nJ;R~m6@N%exYfJBGToI0)NkFk67
zv;29Pe@VybsVJ~{o%a}{a(*cHAJK<Ni4QfOrQ%iWH{CtP(a<j*H$@aMDbtX@6i~S>
ztj%=Df7J@Ina&#SO)VNLpoR;d;xpeK8ZWnA>%QZ9-tks6+&ZXn(wgusO<e8B%i)U*
zN)t4vs$x@aT>|AHj1A%t<hL{mSGxp|?n5q!2X+Y+#S%ZK(oyBL8^PNUXB^F6K6+gz
zE9V8Z=q4>*jOr5qHst{MQ4-GIQ@>EsPLT6Nf6f6&z9zICEi`tBX&nw@*J0$map#GQ
zb_5ASKO#U&miwo<=vHWp(wvLE1^lQG=)9Oj)IZ8F7WL7p_wW<Ir;)Gl<HzxH5_E3T
z55?KYjr`^tZE+-eS(YopTY1}{Axks&%O&_z8h@HZf2p2aqT9{@#6Ck8Cq@xFh^p&4
zf7}GRc_EJb{m0UQ?)#etQb_m5bGhf?Xhu{Es%JqlNV}`YRy>yyA=;0&qCocv-wWEx
zf1~gVAkrD7JypU#pKO815cZ5;vj@#6&e$r9IAASGM8V8R{7Dlvv_c++K)NkBt{Jz7
zFXroq4B<+fR|$S?2$u(9)v3D%D+oamf4l30l|G+7V7MZUZm-i+4!qHYc)@+sK}>Sx
zhFOdUF*el#-Z+avK_LwEQNtJj;Vsp?N;AoPhvrjeF>yE=ba-HwRa5+!Sxm*{$}Fbd
z{jsx{irbZ0Ott%g#cwqdk$%QC#=2v6f+!7wWKT`;cn8G<Yif!MaE2t!5bNEke=&n?
za|vm?6okfYSLPCEgA&)yNV1y0G@H`-&GNqh?1D6m2u*V8hr_lWum|(Vf>KfZ|6m)4
zL>gg?I;}(KqpyG|ea5dO?aNkp2A>1ZV%gH}c6jz=KvT039*?U4Ng2aeaMglfe6wJm
z1;9*j0{S0_zhqfy+5*3*{EOfrf5&3EP&nsPyo55UvH7?g?!;?w%P=ZS^eDL13OnM$
ziM$KKF7$;A%D7tCPl8v*NNE#n_c`#&2wgjjCvNk7ZOdF+;mkd$CEwV`rJ?jxGIN}%
zXJV~<{{SbFla&iMQki8>nLxPY(wi=(g5#!y&t~HpC6xB;E`aB)qA*JTMKVqa`->!D
zqk`)$>z4oDPQMGvxwyjl+M+Y`*zXvUA&)r<h4bvS)qgTbla0BZvw1$q04X%-SoOcQ
z>7dJ7QtLkiKxdbcj3E;OI5d@EC=>xXmjR;~CztH;F`a)1$A5nO{UDjy*N;PG4p0~y
zCJRPr3DcJ!hd(Ep_&%lC2PrUkGyQLh!?cJ_Vy&YmYNf*T7ft!?k8fZ9_T^tn(1QQQ
z6>`!7j0%4Gk-r*E)1C6ucm8gL4aVT4L?C|kWV|pT(1fzE4t$vow2{yV&`kRtM^QL0
zO*2<3BdmYdfb#~;P~jm_ETd(=@9kjWYThE9sj?BjMJl{Dex>zgRs0%1?e{X3R{du>
ze4P%TDxLx9KYCgx?HZi4sdA(>4WwlVt<Z)PBXds^ED3UTusbG?0iD+B2oZ=7oC{Ok
zs-CorG^Z^?M5<6IL*+S+Z=8NOXvQNer6Gw<+6aF@A2m`jq#4%v?=L@i**9AB-aIpz
z2Zcc>qVz!E3K9j8N_I*TU9(f3&6C+o@#%<$o|??akf|9mGkxOcQ!=Zv@0xk_G>L&v
zQ^Q|6=fix^PkXUZbbN#z*c6^tFhR0-oI%-eUgZGV3G|_&LXJm2GZN9^NF6Z>^vyFi
z<eq<w;~5ag;;e*ZU1Q3c_`9i82x(0#O<>fSn)Y$LtV{dwXp)QV=u7&LQ`K)gL-JEK
zTxALJuJDp+(js!lm6ks$7k7K+b%lm)PnN$&H?388%&s|c>$BS>{r4V7EYTZyB)>$4
zzB2mxo#f!vnjADJVU}`8C_{5D-JFe)(J+6{({AR9HDgQiBp9K{ddD3X@=93e5m1kn
z{z8LT-wK_AX2W%q79+V8t_EW31~pqpHmKjHeip~PhbHKq^9v<JQXJQ|0s-t0^b;%N
zpkkzudtg>{67OPcL!m4af0kxgysj0BT!;3qUz`PA%)(URM1q@KI22xofslGRKG1*q
zWRoOqWA@m%v5!3Z%=YWL(J1TJBPL7tY`2-RW)|1a3dEvP15nA-653|rmbcDIS}VY*
z5wzlIT;Zpjq3Crw;Zx6Xd*%%JQKgg88YpX`X1viXo2&(S)Ga+y!sd<b;8f`{XXyho
ziVHv1Bk${ao|AU=ojkU|TWFY50;GQ_Xhj^-inJTi)0(}dZM<cotLv|2huMQ<-ztam
z6irI25|5rEK-#GyZ|k=(NJ-Y!Pjj5=DU@B44%f9EuddF08h6~(jmr0uryL<n@|jqp
z*IAvXO@*^KqJO1vYwH9RAqy%IlTeW=$oq<_D~O~MSM6J%k~DZ-YfPXhH=BQ0tZZBp
zGExvYDSU<|yr<e{+n8D!nqf>(>g<c-*bqP#zE-0^8E$p9-RjK4Gn>~1FKej?31*}a
z-K@^1&JN|AN+=J)C>&|Pel9?guZIN}3zsdci$Io+?WL7du7%S{BYP5)<P>F*wo+Ms
z@kzg&!vIN+fqq)xNGqQ+4)}i>_d7mVnd)X4|B5=4s$rU^Y~DYQ#@TT+IV_*`a{e{`
zq7A!c&og8@g8~}=HSZK7ld9vHm-v@_U9M7ZY2~U3p%Q3n+N1*SePV>p=PZVD7UI1|
zOEV{3#?J-K-dqJjEbS&uI_S_<p7g@vF4@pPp=ySk?I~t4wFEGg$4!6U0#PI|M&8J_
zEV>=(h4dnI4EpsLX5CMyaX>e(oB^urI;}i!bCzswiCpr5<bpM5cCqQSa<(rGW0Ayj
z8;j#P9fqZBsq-5oYpP+I0*UQG8@S#cHGqJyI9;)-xL>xFR31F_Wt8D^En2QTE$lAZ
z@p){cj-xjV$)~QFZi|1_KIkrVl4^8E1zC|zNlQ*8;*|6(aSb(68TNY|x0)0PkLt0l
zi<)3H)Q-s&=1tyMK(E|)IG-Uz#ZnZ~2Ih+^l+O$JUl``og(TSvPz$;xtARGt;?Px3
zVB1W#{@$ERUp^OF+8=T2T*bw2oo{x#IVYQ4a_2lZ@+BQRic5b?H<F6t6pf}5s<wAi
z^Y%)o*i+~X`ZV`IE;&L++`9#}lw<(5RZEu3si{TGoQcoLmM^jnH7zTnZP=fwt@2!S
z+>bqkYBr_MQe0r$zhMV-k4_3k>H09@13MCy+?&X)_G-;;P8r~o4SpNWULDTIZjDkH
zgfYWg!gF373N?QPqN|$k+@MmiR8jplzpy(6GLX}^!+v2b60%qFT&<Mg-S!`H$zccA
zy{3c3nn`=8ev4NiCfdCeHb}^Tt^RTjG!_m9E#kFG2|We3f{|UpCCGD^P0@ck<8IOJ
zeNs!Bz9P8IqZY;t!DDB$>)NiFfuCbGGqz95HnqzDMb&?i*{3&K^a{e!;N!VYSfekL
z8Aauw$NTW@_2cm8;wp};j;b0PyOQyTbT8$?<eYF+>-_ca>FeRsNR_o>e5S5ul83a`
zm|pcP<-2}DMTX?u&zWot#6wO+XNdq(8bPJ)4S7C!op_N^{+G<F8~NOXU~h$jakEtB
zy>wO+Ei!-Jj>zQPGSP12bG57`V9`dwkGrfF2y1bFK@mU-WfmZrR|epAUI8>g6h_K(
z!LpA(bYulMQn9ke<5AoYoX{rh%wVn(vBT(9#~to1W3Q5v8g^M}u3W-S1y2PZj+C}=
zBLz0230BkESmTT^bN*6V7e>!biZMTF4R_nXua<vJ(YAr3JIe+TM7-r6c0u3V$wF<P
zUDR#L{?vlydzc=1Z><;N!KTvH+_u$k`&jiTf>R^KniBs6ubzr5up{Lg)E%V}pmC(M
z<|;VufYp>P)$Eo_vh}cbDYpgaU2=0Z9Md~+Z7B#2H{1WY=aT0eZhz<bo!K<kCZ@Xn
zI#Ykw9fv}gcg}S0m3M>9PI|y#ubE#!7%9zm%8k_1Rd?##jxPf687*Ira1xskm=WTR
z3D4ajt@bqf<@^OCQ|&LP?sg6DD9NXx5$&Q0r*c-#Yf3*k@ORtvU-$J~4O&X%VMPLa
z`oEy)T;BYmJL{Bp)!gygdz60l&fKtsm2-cc*;D~s)10MP07{Jvjkn+KaqfW(*<QI%
z<#sNATWRNJDqWp=;nud#osJL5QQ3_hJ1TCU8wiRG+Mr)=mKN%5L)5G`L^J1Rv&E_$
zAq{gkv3<J+=g#7<>69uUrS&5ZKiw$><5FHSo^7>L+V>>x(<xQ7Kf0}R{nAb;QkH)U
zVTtMa>sLi75_om1+vrxG?Kl=!an<FhB{ajh%VzhqYxiH$;CgjEUbW6w-l4H7-@F%5
z{6}5iTtM8|&rN>ghCc2k>1q&&w_dLItFP(Vl*{#jLcUy=4Nef`)xM{!Dkw6Iw{^Tz
zvKQ1w+St4^QRsw`ax0NcX`vauotu9W)lAEDtNJ)q@1atn<wKijZS1=vf|Q;44H;gi
zyVIRZ*ebG7BNu$E#u1bPeZ3SPBdvDlTahllaH~fucesQmA1I78BjrmLqo{tS;^{_<
z=~myn53qSs;vFT7yC-YYj^A_38qufs2%+_0IBMN6a+=>0n^$LP@aSqPTlas-T6MR(
zBO8QY9QsbRR=moH|6pK*G}c<i=caQ?2&yXwzfyZd4&s%4G3*TJ;`QbVbLvXqa7~A`
zel1lToV`0&66opM_Zsr)iI(%dAhyjCg_U}wl>)e)UwGpb^n4IkBG?627v5q3>4Y(>
z+mCO4duFK?zSpzpeEfGB*Rp@;wOB_8ZDVrgZfv3n!aBXe<^u+N99}=%FdApc$ik*v
zeb*D&PBqE&!csgbWz}pQrnL@NUrWdu*!_cnqmA4rJ4wRLZ?zSRO|HhdYu;lEIJzyn
zYAGRJ>f&t`Q1|*|_3Qic$2QlSbH_gOn-9!&W_d%m;Y!y%-HHPlB3%gc2LJy7;F>X)
z!7~LD0ys36;imx;mor2rDSx#Z$!^>_^4?#eA83Zfoq*AJb~`pOw*<Zo<}e5*d57fT
zC3BeHpDK#QO-sAmjsdkSinX#f7A2=|lboIp4=;xgPufgg1PfkGFJC9X7ZQI#;-IsW
z2M|uf^zvi+=be<w;q`!*ky2~CJpLc>Zcj%I0cU{43*Q2qj@|&l;C~#gW?1cJ7>%>k
z8SjAidm2ahD4iq>_Zh?e>0dAZILOTG$CpE4PHACXoSe48ft$YkIQ*P&;`5YdAEeMR
z*y(>04I>ar;+)63HUM$@g;PHK{q5@?U;YgO0sc!6axy|$h++DXAB_O5jGeycXD3XM
zK29Ja@@pm&gboHLFn?ef_%a=kkT6=QX4-ce)xrm9o4E#1hG3Mg1ezh{Az>_|ncnwy
zuy9pKB&G`+^h8)PEIlGTL(wB$?K4kBs6VH}*Xi)7=rd~k!@Kj+t*H}<f=8=qRLVoj
z31yUXGxs>b;X<wsc4G2Gl-3zk0VolpPpgqh6-jxdk(P%L6@O#kp~BZNz9Ib}=&X;~
zD6Oz@LW@M_Ad7^o(pjzb|1Uo>*f#`vZtgRhBj_jqF&$C4!X;G6PjN-p{4~$ble3%Z
z`;)eKH|RzdUCpAKnRC9+ZpC!h+^ffl41Agz{*s)>D(I)z>IPTHd;JRCFi$HwF&_OU
z7Nu+iFBTxr=6~RoG(dsM<CB%ABZVi6p~+%cvYy`<mo-QCFyG}92RM7qFrkdlyayh|
z|DW*kNy{Ie0<3YUj22FXkwd0N9LR_HPFRhf-^p{vx-Pd0K{&U>YG-H@GmrM9RH9aN
zp{vA+ezlb>V;b>AVo#WMCDhIfz*i$LSt7ztScqz&7Jp`IHENW6S8AiR&~8Bq6VO)r
z7|dhcz^k>+2L9L7&&oOI*d(DnKjoAK9s`XoCP^YcWaZ1e5Kl#9jI@sUr`z-JgaZ9s
zWJDvCFi9dBPuT%uq>@@F9j0IZnZ6!AjaajS(EpMX(vOuC)rXPKxnPJA1(eQN+Y9J~
zMcHk{z<+=?$m{*s_{4b@RP`%UDWv}@VIKMX+yoLQw2mVcQeCnnm*bu4C}bO5@*E%)
zIU~>PCu;UW+xU-BO<eFbHaH1)oBmqpd9+mPCL1uX%jy$s{~0P71ks621n=pz5a%4R
zQtZfPV@P%UlQi0mG#XVl2&zK{o|Nn~SY|sXqJPzk*{|^G8+g+)qBVz++Tht`HdQ`W
zEwpR11)e!FXYF2E)rW?6TOuc^e_jX*p%;;yv(T4KeABVmXz&iUSiEfrF_-WvtEXrH
z+7AeS(-LCwit!ExVjC-{OvKW16!o^0K?~SBU&Q*~o5?yQAgUXjZxx#b46S+<Tj!YD
z7=Mfu%A41h00q-E>rkkvD0H(BkE;s-(WLTZm`%Gr%SNY`5_!yRfNKpQS!{-S4<F#L
zkGoPGeWxoZ#a2xymNqB+lPGl~pLbG1PGctGZcLYypmduSI1H!cuuMH18}7MmqVkYO
z!Iby1x-gfIw92fnEAsNCq-?wqR$Y~o(|>*$Xg|b@)s5__FfUe5bkN|c)#XUn112To
zr9Tq4Hyi%IjU2?i2I1@*Jwja8+Yts4up}H1KHiLw1N?x{uE*TuQfgFsEj1pmkNFDi
z1Z`qh<#n<_@X=cI$4u4h+33<wqS}pgRB7`wE;M-=g)O-uy}|vy@=NKjrk+#R3V(IG
zO&wE_i=*tsEnRP0gP=w9FrwbbySQ&uYGj^5Ly6s&BkCgNINFjxH9x#$&5R}bL9^w|
zSw6+nBE9m%%#btDLxytsITvavYt#70x!;(wm2Ee*q}!RJbO*jtap&ZA0Eg9rq0AqO
zzl955p+tkq^XNhrEksryx~M$nl7AYgwM<pE?`V*?TGbLt`?}jnYB|51bM!1{%nh8|
zmUD{w)rZ?R)wjr9Zs(WHRbz$=eW^2-7QF7c^yCt>a$-tKC*EUBv6lIl6{;BF!@hA+
z<@XlbK}iB&M29H6ax0PW2tm5-@N^R~8H9I!BvuGwPHG6Q5?yafV4<R^U4M;*L2y)s
zM0AIlrL{M;F4=6sXy5sib+vPsR>IZxOW8@K@@n&tL>q_H)pc8`*$c;#dL1?erWQOo
z>^b*~!fqq3Fp2q5#bORB7lCA;bdr;pc@yKO(oNj~UuArwo9VEQw0>ax{>gFIQu`BF
zbJrrQTgz)ZXDVV{tKFh?rhh~PbD?uK+*6lqe|w=KR$ZXBHde1$h!st)v$xqLXUQE1
zFlDQw^~spKwjaE2UPg(4#pZzJog~_{4P}LlFj8&$(zvO%At)-0=m1c2AX|*?L=2@3
zy3Rtj%MFPkdThwxadR!rh73oQ4o~N1@Z5W!M|itP-TOS}GCvJ$qkmzBw<ET4n!yzf
z&|AfGn@Yk|z7)?}=W?BP+VsX9+4eh~3tPrA9Aat5*swUycrV)k<5SyapvYCB;r3D9
zXItMlsorWk$_uCMNWBEV_s^P2gQ&&CJ0wjdEi!RS&eTxbWR~^q^ZA}HdW5^C^ioB&
zm+~Y8YH^$Q%9BfGPk&rZ_WW)zyRrVbF91I>`@3=@fWsoskGauW;V}lkcWwm4TYG%A
zDUy>j<)hlUD1V(PjH=D$w@;hBE{x>O3m8l<09{vi58L&9aXxw)D8b`SPj+r(4G`Lw
zkjA}K?h<w*D-EXslXPX%!iNg=G&1X(!*$;H#5@mpUHW}I>wjgGgtd(6RW|*G#;Zz~
zpu>2(mU~JoTBbnAJyq1(-0*Bi_PVURGCdWwtDW1AK-o>uIc1Z0wtC^Wl!O#jRR7(?
zzWMWa0eo<htB(A2L6g;+?*}>q>#*>A(|&U>AI6*%EWTXeQEs4fnBe*(f7L;po-4LO
zZ8d}xsQS$WLrOl?(_yPiP+c7K2N9dyR+}+4B3dSB_Op}+v~e?_zby>zXBdNq=SMky
zV?m`HeYiGCwy`>|s*xh*NaF{D!_i~&dC30(TqZq>m*Gt@69Y6iF_)SKGSLBaw`>P8
zSp@^x!V#BoV;&O%I5d~xtS=Rp@(eOFe}L4E9Xpm#C8#J8b2h{Z6aT;BmQ;e_V(EN(
zz975+;W}WS=QfN>sX&~Hnklp=P;4;X?4G<jh*2W0(B5!AA~H`|V~QY_60+&QV(Z4I
zN4s^Prd&nr4@i`%6G+y?s5>y1FHAIa37tAPhiZtz+m6&VJ$J`@VohSNV=?FMCev1C
z|LC}|fTdc<@^2@OqNoh)i8Ak%C@N3@Y5s#;+VBC?xj`bA;Y~3U0yH?6;W7pjxAF}#
zeFzAxL-fD1^y@E|;Y~3U0y8+5(Y7=rm#-Ew4wsR0Cn%So4=@jZv$L}=MPR_OV<&-v
z8kP&RHH3nqPL5oeG)O9N^6UG|o=zl9sXmf&N&~hycQ^YsJM+vld*@4t#+l41n&9!9
zqoI@`6f!;sy~BoG6@n2F74Y0?qg?cl5g%}s^-gi7c>hR(cvE6+C5D`f`V11)K}t5{
zLmkCKidAAm!I4UT$xsYc3Xl?d2@ZOhoJx3FC_aMZAo-;Cg7K6{B{AMRomMH3q~42=
z(7hM-#xslF!|FI`TM~VM7QsU$hWa!cpB(8J=p{$JhryY29f8QOuPw+M<~lCHNTOB>
zlnkd(^)jSX^@`!_RnMcB99bpwQk-B63Xp0GCWEoPO4*=)eO5ie#&~^-KY^p#0E-T>
zTEjyxsjULD5$CjL#y+|3o<PcsF*q6_uj}9hDeB7EVdvCVA>ba@m1EnmcXg%(yG=SS
zHdmtPXy{auIz_-XN+F$JU=v9S#t6YgU$iZ{iosXCVxoxlf)1jn!)W(0a@C8FlFnJ6
zm+M?$>{zIO&ILV}sMe5Rl6L`P!2c3+?Jt@XlFmw!Nr`l{NCwl>r)ZXPohgi+O0^%{
zF_l<tiU+RRPt=t~$Iz}|ag3o3=Syv?ZtHb^YA{KvryZo~2YT7{hb0u}s>KVWP+O>!
zbsY#T#DqRYU==3yE4W2r7$?Ilg{#q$n4-Iw1Lj?Sq;`^%Mv<BY(N&@bX2@hpt_y;v
zl#;82sCA?(f~g<{e)QdU&9gUIecYK>%|HJ6FWgU7eOYq00^?4uudYrXKZd4ho=&f(
zvy%^_acfu`nC9hZv1n%#<GafrJ}zFIEJlmgP<P|`bh5x0&shh=jCGXf2!Dddbr+Z@
zqMW;b3mBRRK81_NkDHgX>G-5wylGxOd)~ZmKQ6j?G_QaE&|>7bEhaQsw3EeL5u{tf
zt9CxUo{igi2P%5&m-hT}^!@bX8;8q?Xv9vRqWuglp+&B0J(*0VwWC7a@Q>b9RMTZt
zRD*DOJ9^bN&G*yUc{{5eyVK^!=BMT<F0i71Q%q#McthS>!M+0#twI2aw*}ECk!()>
z^igwieYVgJk1r?xX`bA6KVDo;C(TLo_g6pZ|8IX?EI!Qt+B9!37r$Pg+41y!^X~fn
zhokfMx8~>TUtZdGSIybg^sKqK$gzwse9mV?7sTg*F9NrzO=sunJopQD5z0lOW<DE#
zSM!Z7C+%$BJZmTO_Q}V~`IFgf^gFuy=Ge18I`&ci0$Z#Ye#@m3g0WvVVcFv7fR+ts
zo_)IRcy!EZxoC}#99QBaB0c~sh}C6mY--&^QGgXZ57G45hBTy4WJe2&4$(xPLiE+d
zaETi+T*6K<oR=|N{7noOz9WX$Jn?RSucq#zdNmx(W{<SS&c>{Xoh>6RNpRLDK5d>D
zbbr#kn7*Di&&;;~eW0s=+mwXV96^HMS-VG5Wb2V``+$nYCQ1hsPwz+L*|bIxd#7;}
z*8{1@t&_sn*JJ3xeDgkcT=Uv3eKR=z)HjtLZtt5)O?%%|a=K4i!MZGgPu}8xD~pXz
z4#*$Iz1U?GIXaGI%_Eeve;!)bwK{N%S!^jhL^H8fhWB|*qG&ys;s8y@Hu>~_gVyFb
zXf`R^?=FH~q+J}FHb7Tvjn6wa>Wy{B_PpZQxdUA-jXI$VgRU^T5qvN3PEo)PQ2=h`
z$r1_$lx#MEMJqI?81@Q+pRccfE=SKsi&67>bavIw;l2B@AP=P0`nHc$p9D<kkdS-P
zFZ7#Z0`Cj?t_5$IEh&N1AWM=&3;gsU+Nh;R1Wb9_NR|{5LJCaAp{`~)9CrAQY`JyH
zR{se7Bv=A|?w<roDiqQvNdeTQh%YJcKL=ES)}a!lfWeb=bU|ZQ*v1%tHzuJ4xX(%S
zA0t9zaeL7QBE$wl@itN5zNdrD6JX#J*3$(25908tLXXiNJ5C^H12hBbv785J5<n-%
zA)^;(nZG>0S<4o?b8mS4MjBwZOB&!M_ts;n4&u8!)g4OR2zh@tP6B+<1n;rY^-vE0
zgYhE=OG$PGnBj31xY)dZLcjtLjt`1}P1^aQJ^%M5ylk|XcHt#-be!lR@Ezrd9^NtP
zOosV0&`fKhXakkaT8NT@ss@WMQAEmqwGb81#lXv)3QEQy9FXyU!*r*!%RjIIA7<0@
z>+#}~jfm?Tp-93sgb%N2b$Ejqvqj<%P4XyShiDRZ6iIN1rk0|AUcMT*3wZ<XLfi%J
z#IL|zPj*93c0*5gLr-=C_d@6e*&hjmcUBO4ATCxgXvo<nge2$|BJ&Qa1-9(G^Cb(y
z22ta)>`!wzBrC-m7zOkKvxo^)ob@#XV3PkN!bhy9BEg{cunF7a)kWd%z$DmcXIl&h
zCN7MDmPx@m>mnV0<S7khJ>w<!0Lj2dB;l0weHyHwufggrc$8%bnjKHZwhaTIkfhkp
zK^XprNWX6mBtn!W<}4CZbRkQ6fGCK1MyRBl5Df~q$V+es)TzPAR)$cp4e2w?H*CJV
zK}5m$euyS^O~L64;P%>S#eE>YEpA7*f!mUI!R;tZ+?F1HlcmRG>4~fK#8uK>M|GUA
z`(^nmgi}IhjRy$|sD=Ov<v~D!*=D${7kv|<=N@3}4j_RTA^LMbfk4W=0+W*SpjAS)
z`ewP7hb3*I0(plvT=Xp}r&3(Vd*uV9HPoRiM3w6~$ibH1l(7B&K~jRZ31u}fh9for
zd69qtNkCM8^Bs9OW)D00kSIISBSTq3LMcP2@tW2R$ie$!NJBK0Kb1H@(^Jw?Z-}M~
zHYNFLAU4qk#Ky7<#HO%>*x2J}><<89-Vab&qjwop98lDhaNwmk3cer@5Lry{DgB-!
zYe*$<Ofc-IA3ebuatEemSA<C-3o4LEj`lrKj9P7fuDsgUcPtXWqT-v8<bqo-zLWb5
z*IH4yg$&O%tqjlvCC>Eaf#ALuNLvHJ#wRuU`O#n?Sh2;B&nMQkZcZwYG2G^v#vsjo
zY`bR86+>@p!J99TRxQ|lQnW3U=9QPmxj$UU{oz9PJ3(m(#TF{CT|$BC^(#Q@oPG_c
z{Dik=kL7pz+U#mHd26S$w>!Qt13}bJS-f}nMeao9##&Tjs%gI(`iEC9{&w_qdfqla
zUcUJ8<RQ*K0fwP|3Tshn)&ogLSW5=E^_v7|-qu-KAmtjbxe4O?QD;SA`Sv^XTK@%`
zjBb}PJSP(ZI5d~wG6ocvY9KP`e^k-n{EzoP+{vnim(R*hQADR5a206h1THW+a02g<
z3IMmSP#|AcdHL*ciiq9>+fhd&yoAZwC}haQ_g`OBiZJ9C0xqMG0{O<G^7|iOY%&Q^
zc!f+1h>#4JNPev(K8)qk!Wz?)yi&wQk0dkF3aNFciTe+mk{KC=LDzC%e}ys&$yT^1
zEu2+7m*2O!s2~j3F_*DpYmZZrLKy)jkWsBm9*f^Td+s5_YAXVn>7Zq`77;z_7RyRJ
ze*tc?EGIc?S*DKr-*$AaQGrQ=N<@HTmbC)>g3FF&wGzs?6)wZ`YP1)+x9t7T7!bBa
zGhX=2hf)2KNoG@cVU(IIe_AoQ^a1IUkDxOli4(!ZBhu@b;zaDZp!k+-?#K3v!evH`
zO75sh=bd&WVOvS4S=?h!^6R!>7+%zp<kgE5i|=<_5^zhtC;a}!37z5oyY%JDSG)SY
ze0?3;%o{=aKWH+q@L1cMOvadK(h7{t6$Hf4QT%<^S~2R5V$WDXe^~81ihWyo47i-)
zn5nE3T0=*8#8ltqRo}e7TUHxZI-$`nAHLlEXTk*icFI}$PU7-7{M(d=h0+${Bn1Ws
z@9;Ys{G0xZ6JP%O)5pJl_@@#`_P;b$PJlr{q0Yb5OO0+Nt)4#DYd~)c!A&ZKX51fU
z9~(4IAi_5AVY&l6e+kN<Z_cBi(`XbB0OROPY?Ls;r-{UX4Ur6K#d);y`^*QMRE<U1
z1c{NZ(khiUT_QbErOWi;*EAKW{x#iwobKLLty-sjni6DLI6C|rX;lfpqpi4RX-H$~
zq5<<()k^ddQ^k0c?{}Vxsmp=E9R|#@3V$_jZO=5q_Dw^$e-^DkLrt{B_=+ADd&QCF
zq*mCNP7C_HN-nCZc8iAR-ygnIx=%>;mVK9XPD*30v2Hq969u|_S}T|0HVjMgI-O=y
zDULO$#gYXge(Q6IeuWcNw)})G7D1NpCMHtZ+iBk1yTb~>3~enOPzZ@NP=xY1Pm&^5
zdJ@`0olT1He^~*wf3Qg{*gx1ny@Iu8o%#^Kb^u(t3zw?AH2ZbB*XVsI@$(%=YZN8t
zwUkS<Wa5D1*qs1WD&m~l5tIiHYc|vQr1HAw&9c;Cy^KL^u{U|yrov=pvAbAMS$DOY
zrVV&gcb=p+rSeU`lV;#e)0vs|mi1-pUPIY?t&z7Je{DAD$kAq}DUSAa>ZjF&aAFzZ
zK(`91Xt0{}kQ&0gzw~XZUc6oGFZ)HLS!o7T&1J7D^DkwEY-3;bd9~6=Q-x<*G1Drd
z%Q}B9o|q-Dl>?NS&}6;#y|$vAy*k^o?YGfpdAqAw2ePeO_0ZOJvnN~9Og%01yLL(^
z8<R{rf1ZtfXzS5Bp-TK%<&#BUq#uc0ya}P|m+6$*4d~pA>my`{gT3YVtA1;+2FB8%
zqxxmDZ?wh<pAviTVz>ofcQ!}#4oeo_FoVDq*$!DfoF!aUS9n}Mx=AH@6skGscXbR%
zhAzKGE>l*mW=$rrPu<F)I<p~Ka&=nSH`k~ke@)FYan~jHU|}d7jJVu43rb90U*z4a
zJ5zNZIwjv<e&4KdL1KxjOi%M_(@6kZ!k}YMvh5_Gf>83x{54u$If=z?z(I>vm$MBI
zJQuZdNSVerg^f<At!KR`xzbsCJo5Y9c-AE2X4dP5-Evv~&^gws%jU?~Ts2GV>xW99
ze@mMSjkBsUU$`ukuNT=al2kRu0pItgD?%{XeyGltt<xOiG!U@C)ye$nX5X@}xK#up
z<CVjx*HyG_{4)8m(ti-X1dR#Xn_EyWtb4EtvbUgIAhF+Z?iQ4*BRPn2F-;tVFCoF-
z+T8+Qawf;n%OR6Z^wM|~!02uCQjd*Pe}ECb!j8WZGGcs~Au_ocZ4)x$uv6=}{AEWO
zfH55=gZ4G#J=+eK!7rQAL7W5hk2_ovlwc;~Fi?Ac#NjY&2#p5qNcJ2KPmPzl-~SG*
ziX}-vqjy@~@i;6=lwGmq3Kvi!fE{yLL5IMLh-t9n@@fYc!quU7(-lzFSimG3e{MnO
zjKx%L%q<9A$K^1n+EE<?RXdVnplU~Q3{>q%DpDtpfT|tg9|l#8i6VrXfN*0l)P<XX
z@Q%wdShXWLhO2fY$AHz2<QTHrk=z8Wc9h4l)xAbJ1vOJq?uYV7Hy2-*MXzFOlU9o>
z4vA3v2-G7ycYq14<d^ficC#;Ye|dZO#M;`{g^)C16{tyTP_dcI=G<oQmd%dCH*L1<
zL8H8O80lcJr&jPDm>XE<gLn1ZNW0WVD>;|tn)h&FwYAXEVx`y$1Px_hkRQk&fgS+#
zij8^EU1l{9OjTfw^j@3#I;+}=WN18m-H|Sqnn!_P1K&o;PMxPH^QdRbe=O(&apYYp
zb*^Lhu1Rdkhv$<)9sG!kzDc)(o|h!8-SRAV<uSB0HXv5ER&!{v?j(mR@B9|D!LRJ(
zSl@36j(z6m@f_89)2v-;NLml118e9WyC!iDT&uE<JbcDCSZlfC<(2c}Rh+Ee!`fuY
zu@VlsQWILxlh)xafpeu?f9|xj@j|zE7!+LLvW)<XOMgY6q3zAH2zM=+tI`qFNGEQN
z_Q!M+89=z^5glaP>2TKH2#hTp{G7*%{Qf9J%V3S$yl9MHt($^ur?8l^?P*SZWopxS
zJq5GgDi~CBqaH|??X)s&u+w^t%YB0ewxJ>h)cgq$(J}lobYPOde<dQ-yYg|A*};xL
zrRM((6_rn`aZO<pD_wih&X_a61CI!}uE}vRNn;F$^UD=^wBw^K>pj)|Jh7U{TxD9-
zt?k<?vQl0uV(lBXV|h^)U(0#MxV;Ayh0ij6w^fH$C^S6AmJw)_>`{fS92qy5ePmc=
z@*1%0d>K0~tcsAce`Z>)FnZzhbQ&SDosM(4@?NO<39n=?TBfJH@@YMPw^QRsXDfRU
zRkFST!^=AKIzF>UWV`|e9;Fr2U6(QQT87<~?1Ae@<45|6>$<u|bc!KOj!vGd-MAT6
zZ`pR`WZIa!Zi>6F3uS&$)eQ62=<#cg>;k8m^lk;l@sQqJe{oUj^oqsGoxiax;1Y%z
zmeLr>bop3boh6;?T`w%KA-vA<V3d0_7y&9Sxl*lCBVOxx=6vrTt|NZ6)AEUtqb{-6
z7>g7wx97@3wRkXFQ_VYK*?gmZ--`B~H{`c6u<?)HnNcULj1tbp`5{i<w-^XKQs=90
zwVBrEBR#%Kf57FR<ta&Hvc{zLigt@hitjPcYuTh-t=oPX1r)C+pLh<I4|80gUa-B&
z(ek|Nd`}Ldc*}&I?`<LTVV%sp6h|PqC(AzKQO+sIguFdg4&zjMBBHuin#aD4oux8u
zM!Q;e++}p|3PrgCzN1!EHqS@s#DV8r7*$|3A3J+}e?8b{%RXjjeYDpJ`dwYs4$csg
zveC_e(hq&kt&!m5vz2DgZ(4A(-_#M?Nz=c|z;u_TU%r~;>B)Slq99jMH7Ad3o}PTT
zIV{r)1>DuU7YkfhdJ}v`W!+L$t)z^Uiu<;?lzVI7&~^F0evDcZ`IHlifvzxFq+Dg~
zIzHf)CxAKT`EFw)jP7c5sYD3f-K!4S4}14<=ladry2k_9>sS8=ssWMhm*Gt@69F}s
zf#v}fmu$Q?Cx69TNpEAh5x)CZ=nt69t+)V&0ZFz7$SsY&1vv$ooaT`E|A!Qp;{Nos
z*pq`=)_Y{J){j*zHb4C5zz@H@oPNFhxF~(F%z?1-@ay-3z!x0<6&xQLI9MT=QR49H
zkHbISIF|-LC;Vv6mBNq3za4(V`>PTTM+nFt@&0PHcz;V{<pn>-G*&_!uROgw2fAUt
z#V1IVD~V4s<v0^ZC=n>oD)RFFDpe+sUg38buHPdWT1+8fX)VaW+L73M|L3p2z3{9B
z`Yxgi!#rxkSj|vC5?$yhR6DcGP?Fhs2J-SMO<)&LZGb{)6b3wgBWD2bCP;wHbBzK5
z{-JGtr+@Ny`wagEOBBff@*JK7RYDktrqP<lg{R?{hRlHttLA6)+Ym*Ke+6>|KI6Kk
z^3sotv|Ju>YFA2Q%DOgxNV;P;Hp*63&r${r#Z{N{qV9B-;#{RGv&?#h_{3D%CT*Na
zzv&&nL_1IRHIi{B$-ZDJuOq*jtAUIB<hVB{+J9bG-hek6ap`&H@l{J$t^zt+h{m6n
zE5W-A7EhYXGcHvNA%u;MVd~!%kZ(YkAL1!+mY-xgMM_%Bmk}{4qb$0SJldnsE5CH>
z&xNXLXP**Gkax<)f+V%})z^{A(xSkbkWIu=R3-f;NNV+z{JOlUJVsn#5*+#mXBt0F
zfqy#uiAXU&{n=oL17@`zISKI)5hL5Bcmc^cYjK54QUjAhjMTSqjF3-pW}tw6hV!e)
z*UykTSyE|D{z*PBXD<lIlSXzPd8~0hMvBskaOhm4InLr`Rf?Y+rO|3>71?p;GVW2#
zoGZ>0T*=s*XDqy$VrRBXLZ!@cu&A$k34edV9oQ5i5@M{35t$`hL}Xg|&WK#big3($
zQ_zpJ*hj)S%Nw`po9&rsEbU2==)OA=aU)_<i&0KVR{coah4iKDKei%zi$+m4+SfnI
ze!q$!1*qLwx#YBF2Folkxl&S01D5#-uNwon>dsY}?Y5EU$Q-DdRfa#5Lscm-&VS&y
z@mwZFF5fiD)YdS;-6R)nZmKAykta!4cnf(Vxs+cj50qjav(#ag9e4xp5fptHZFb>p
zn4y{kB?f)t(v*fM!+WTBRp1A^M4KFkP)PGmd0u_eS}Q&Y0hMf?HJ6@d2xt?kh95nE
z19Yj8sxriNfD%iPh5j;?tqBAEy?=tIBKrhSiGu|^nJWoHrq}-}5R?HuCrMbNtf6BC
za^NhJMi+nmf(>3fCh*ZfIaE?5IXSgH@Gz7}iK~lgrSHR%95p{K;0~T?T;Hr}NC3;c
zQ(Za{Q;JxMy@moPxs1tKX(>bEW*#|kO}Y)`DS4*8Qe-6#uUEOC5Urpj_J3;J%z*Fd
zF^y2w&*B^3Y!ieF+SAq#BBv>3J8{_#Wj#v$iUndaUq;Aod~3H#p(!aFS%zXU^Kz=_
zTuV@zlRUIto05nkAGGwNW$G_ze8>#RlDi_qglEL<EW-wc-gmmFC;LmygfHpwl&ZKk
zsp*PENih^HTB2p*xQ{I2cz-M2)}Bmn$Af{Uysx`NLF5vlj)AvEt^8km!<;Mor|SHM
z;wuzI;-f^{)%4u8{-S9|k2cTgQt^A;76Go?Ga~}r62ICyrpxv$Ka!H-Q4+F`1<;Mb
znMXDTkwU%+b&{=`)6>3EZ=U9$OuuxlgzB$7C1fg#ZM-`VN%Ss_$$yUxI3FwbO#n>u
zpV;1hT%1^$L}5LvjTud%_6~2l9c|0rS~|wXKsz7Sd<F9wvjC-*%*)K|$6^!kw+moB
zGu+M+yh=437=4vY$TU;NFJ!u{>byy=5{SkyP&M`7&1qvBdli}{3YacATTc?kL^6S`
z%hl_to{`09%=LObyMHMOa!om}0t(tkh>;0?ygtbZ%(vV)3rx#I$>s&2XM^|CezUzd
z=+9a^q&>aKBR|W$0qnX%QeOBvw>4J}n-S}i4`N{zoDRQ!8-9KMdin2UL8)*vmFDoT
zKVHHMtpro(t&%Gyv1a`9--q8{{yvhWvv|gFPWlIUa`x`Y&wt#t@Ek3{jREK^Aut|5
zBTg7*d2vJi(HziZ<dJ>FS67IC)rILvm(gRrt6*GZ0keJR#Q2<f#*h(4j?_kL1eZ2E
z_EU0Z8e1Hv*y2cb+E?AK-neGA<^kWOw_z3E<TU2$vF4@@5!<|+DerE#_>p@tg}Oh*
zl0+LagUmr&jeinI^K*}ASoRpiX5|%g_3V{IVmfGu{)vj~p0tX~E;3M(e0^wPH8Ti9
z!|3)bGv*2zTKFkTw!yuV7}c&t^dY&uTN<Yr;_OIyOByejac&%~7Wq+-WHUH!=MO=O
zX=le3#nuVCWLArbzugjkTjN$R&;6>)q-BMe{%kr)zkkXa^C}p67DW~6AWEQiPg)BR
zCkEje9?apY0fBiv65HgqL{!+@)#u~Z>y{IGK&?9nl7V}I9~<6an8zKh9~?uD-!BSy
zX2FkU5L2w&Mbt_LQjX%NwKx{{Bs(gzB;1%)X0%7_O&--gxw7F%tP=1uinf(>u2Ell
ztOoR`pMPikU3;{icp~pFtYOj}4ipB8`(9sqoTqNDX^w&juhnM`!3P`*cLwYA#NFDb
z70058PufZGIq<Nb3~YIXl8S+aah2vk_!gH?m|}pRc-7~-brdKFE+?OOJn~fVGBFZ-
z0-2N%G2tF5ZkRD75yDSq49BnM=fXyLh8y0KrGI4&9nCa|k?IW_#t<YSL3<`h2>S*;
zHh`Z1tsni>LBg@ThKZ+pgM>nDpz=VFsBj@hPag0@kWc{k6rYI_t(nu4&bFe2!hW!$
zXd>i+2p!hGY^J*^B=6Eqwz-lnhBGm%ySDt5#1q|cKUI9(Lkm~JBHGjp2%2k8bY^{H
zZGY!1(@IYin=2aWs?+@u&K*|VW~988=nEOnDu%&Pm8Y*!T`rG9kPE0qtm}^iO9-Hr
z>p!U}-#+<yMWwmimzbJ2ZI7=hmkEwzo^ovXcTL$~O5J?D_-|_MayWLPrw9u4zP%%N
zs*X)ox#M(c?}$WN#iaFj9Fc5VA=`VT_<v=G7LV0W5}1{foC5&Y(U?qFH;(3)hwHX+
zUAS6xkGkV+*8GU{x#SwTP`O>bk?rT2>I<~sFv(Kl-t}`nzt+axn^sOTCJ&&N{U#rj
zSd4lgY}`kzDcTIy9<sYRz=w;%yuYHC%P8NMYo(v~8CnW?*O0XXE%2GZ?{E$_WPb+t
zv5kPr(S1NL-1{f<spl!{V6I*xE6S_G^}Xhe<e0-sy>$yh0^EK1G0f6$6AhC(=i;qt
z-87fdV?WH;@fvrUd(O?)LWyiy?q^)m9SI534I<MtgbFuZyBFZrHtmDjfVM@vvHt1?
zL4I@W#vO;Ynx(h8z0GGkxu9u(yMLQ+2ncJL7i#~8Kz~c%_O5?_m*;v~Ky+`q8z2zU
z@!e)ak&i8Zl`hANzRei#Yt+M-=2lH?HgcFR9$e+lJ0``%-A<=594jfS##yn`SQDXT
z!?j4-+=B~Py6io2Iq5{+_}?$o^dBG6+}r>AMVT?)v&odp``dhk<}D&M+<#ZpsrPwb
zn>gH;;nv=1JsoVW;EMH*!~Gz>xTG=px3&z7+`9adyT+zPl+ynt<fL@6-f^OI_VZoQ
zC3(8%6tSyby6$iBuCIX)5%Xwg*=S4#^w?Oof>}(N{))b`z~orfqvmhwj}EIm*;gK@
z{a^aZf-9!2{cgPtt9)Hwc~zhsQK>_jN6WR+SEh=s_LxQ2W0uzsfjkx<!hJ!DxhJ&K
zVeSV%*kHzr2f%+rf4L{T(_Zce{{`LUe!-pQ^1$8Wy=9A$^5u=?CEyO<i0A(U_m(u5
zVJ86-12{D?mvL_k6PKN%G$@y=zXuAJ%{MZ(0du#6I5M0Be-(}?d`)o-i=7{n8ibPZ
zV$mMPs*Q~58O{dd{nVkWHL}EcsbRO(AT^k#Yd6)@-lj;SbjC*9OZvN&Mkf^X4yCD?
zk?5LpHm&G@vtgvK12SKJGmeg2vTZF{kvL%-TDZh2LUrvKLQ>pvu3m8i@`-e!5KSOC
zh{M3KmT8zVe}Af1Lm?qW5?}8SE<r%_&MfNIsJaAV%&T_arjBIRg3aCD+w`C(;%IIX
z)0vY)Yg0x?>GCHdCx1?QG*y(SSx+FJC~?v*MOK?0qKuf@69ZG6UTbc?cI1{-r}Bc+
zLI*>-N*pu5w$XNJv6_d;jp;oZIXBVgQmls?<-je!f3J!SgOkZoUR3lDzO1a+)Q;LV
zm_e2`nx!o;$a))dslgmMjF_3jV!mU8W#%|HipX)@nooellM1*H(+cZ6C_5>)Pm|~r
z2a`%zH8t*!xikSqx#2-uYB-7>yeQ!;9*K5#PeLOqgIm8Ru_px0cPZARQ0;>V>X9gP
zPolyje^WjSWtUm$dc>-IbvTSGl;a$anv`+~J8b%??%Na`j>r5OoQP8o=d-lUW;sFI
zEk{l8o9__4<vT>DX~<X+s~*TFX^73To;&W-R1i_mI0N&`S&e1YFKaY?9UXKAs8fGu
zt$Z`p*h=50s)Amd(h+n^Mp|1THLw!zGgq$|f2WPBSB!c*{S*a+q~2UAaDD0VP?1%~
zXx{mGwwTX2_f|uNdPaSqvZ>RV4T^K@6{4tY#i&O|EUs@W$GlYz9JV~|wZyTo;zCkm
z&f!m3{D@91$IjwMS8A#2wqdurc}5XBA%ix0r?jP9!nNh9+IwhA!nyE#KYo!gC?U;U
ze@|45t}CWnmkH^59plvTi?lgZ)jfLtRwv`M0WFOBh(A)jnQ`p-BY^><p7p~vZkXo4
ze$#KOF2ETzdKIKlRS)mEUMlBuf_wzWGtRSfe2-nb@>6$|&2r=-Ds@mf$6mj60DVT?
z>9;Bu5qHjX?}p))cQ?f6b-WhtX)N3me>|aeNX@`qi#8h5I>Tu7b!uYOCutg+_YfZ!
zFeO5>NjGj=m}DL6*pHSvm&J)72<`QmUet2ja-){||JaXKmyOv;EO3X>ZsypGtYBYp
zzIFoLy|%jxs!am?nfb?PqvI^r3!_ejoR;I3v$WJdF2WmHP5Xo`ttPIlt$!$4f4Cq`
zfnPc5k6)KIHvRvO=CXalb!p>NE`;%!&b=><Fe{9o>3Ze9^oXiVt`kG3)YO_TUp#p|
zNNErz!imWg<G7JbG3p<lOzo4lBvV{lJ8p;?S0!4mx91KW3<GTMT8J@P$Y7VuzIj8n
zlXWSKC8sMRC)~Ja95?QlG3p}@e_W#SmSeXRLJ>Ec-%(MmbUFx%>pQX~ph&Ku*DHtR
zR^k!=p#M7Cx2)2#Ke#MY#i$^kh)K~Ebl59%!P=I3tmQQ)TT|rNFOf}Gu+&_l^)T7u
zHgjR-(!H%Djj;aA!WQGWg~cuPkH_L&(vDcnw6z0^^+wM%UBduBTqwOwe>V*BqT|9p
zLEMI$84DrQ_8tDH1*mYMty_j0rf!L!&xRJm_n;6$V!uc}Zb1G+T!}j|=&#vrMEuRn
z|8o#tZeQ(KP$`)nS<q4NPE}MD_efjf>Y6m~vg=u}ohn_D$+Cgj%8PWTppmJbeX*I&
zUf(}n{sG`OMRN*eZe(+GmpD8p3<5Ycmtp8U6aqCfmyvTPCzpz~K?{F;T&lPL4Fk5j
z9bj$=`WDP-FfXSZCjWn!B3Yy;YO`d&VCEp(Em6hVH&)BjA18VG<?Z_Q_Wf?)<b+4%
z^y%xzNznsIpGwkAizla)utuG}{yhEuE~U0~UFiZTjiJlYZ%^NN{$`Y?6$<%M=Wh<w
zdtR&WbY1dVjcL6J<=KBN!j11cx<f#v(R7C)r#ESZQ4vDfMc>cgv|#}G5q-zya86*#
zW`+Q3=TJw~E{vKx|MTk~Z<4tnMX>Q_{ObSu{CfM(Th9|jBfy8#zy5rS55Op4ygn&y
z1!{Zx>))pzZ@>0nTb6r8ESJ_Ij#%zMSf@_b24Ui(%<{^RTgHD{04=W{fBkXl?)>)4
zfUtK0NWL9J5=~2gT^&f&-uJ7c0Dw4&HeXTXKq>&lc_hxu_?x&YNl5aZAXkX!ghg|*
z=!GFRFhqHqUj-3c_X@fl-BGrOxI}m1`*=0d-bhGbq?c%mB&P3=%nu2f_IN+81()x`
zy;3NJjL2d9kc@xx=nmoh{T}43HWaQHE5P}R6|yvFH}v^$9MiR^Uqu_n5TD>Py(Ls|
zJ;Vh*YY75S^{e4hDB8cjazZP7bbqY`*wC+Dlp(noE7g-6vR=^jD27gx2hkG=1-3TK
z8W?{^ivzaI=!na+q@}<VT&q<wC3t;}Zi1wsKz!FIP=$XgNuNCe^!}C9c=$*=jqd!m
zqO(d;Y$BaiAh0@+LL{;)nwYBM-)tbj3<S&KjiUZ!WGP_0^Xs-*_E~tA{M)6W&Nl#u
zKzP5BuyIw4EpZDhD{O;`PWH*G@WOW!;^nS_`dkjyT?mUhS!7l@DrF<CL3ZWNn(UaP
z7__ips_Z^8{z<2QSjY5(eucH6gZQ@yDT^m&BoSl^`qvcncN;33UPr{Dp<3ab>sOnN
zkP*pj+{))}ERCiN#7vNch$vlV7Fk2`!7zD?WA(1>AqXX>Ov3g3t;%Urr*ZOGW;ieW
z3^xp8<L9x7Sq|72a06oWL)^JPC@q5W3*6X^19@OeX1Sz15x9&8&>JC4_)wz5w$Gk;
z6d&_P7-@wZ;%PDfIrveE_eJ)BSFztro+9QoWZpXDx05GiZ=yoX;iaVIm!w59Hh*1%
z4d->^u%gec#a79V5YJk@x6m*&{u_Cf9bo}ACn$GVn#lfrx*v=xj?vv)ip&PbSWIZ|
zV^7sR7@RYdKyt4;gi2?IE1Q@I)0p1v4q2xegUbk?k`ZtgSK=Ow*WU(c9C*Ip(VslI
zZ)eS%7C^8<hM{J%O-wZe8&a_x%YWx^26#{QLU&7{-SVgnUV1Y=7ueZe>rV55cuSYT
zIpGFfjx~LhyH+W)g7{0lD|6K)8^}`2&v8U^jw#xhv0kuG^U`=SG=tV#`nG|BS+?Z4
zC!>loh#;q!PUy`R9<wR9&!-M5HV<qMDPrE_p4nERWS*G=jh(8o?IkHxj(@fGkncPz
zYnH@}ePnLy89U&2ZdpPc+c<}uXaXgBLQ?wGbLC|@WXc~CsA;0VrlM%5fa}rf+Z;uW
zEeZ$K-M0uP6Jdu~9LKEkNLFYwIA*wI)raZ|OZP2`t^~lnO)Mq7klyu8ECWP?g!LxY
zBlSmWbgh>l07j^xxA;H_@_$It+LtBWF35qU2|EuJIo1kgbq{T^@_4>L1FZx(i=WM%
zZkP2k_90GDS|0QiGIJ_9v@6SCRN^hycVI_Nj7#}Tzn9kIY(o)<lNxJ6TJYb~NGwtx
zL(+1!k=U)S>5R(M7iU%}o5$XSAauai6^>g}2in1;yDs#Qm#E=(aDTsgNse}CPF9!g
zOA~6ILhpq!b_+7vp(&!VNgnSr6}PU=zI4q}zUc>K7J7Y|nJy;tia!!FM|awWK`_Dv
z=!uX8!P28O!|%kVFyxbwP>BmQd#<#5==Z04cP27l^?;#emK{>24O=fIc#r1m+TVsE
zFI^-5lv9w$d{MT@9)HRU@GG{|L!S7_M=W#^|9~REgbNCG@6fmGdx#&aE&FaYQ#j!=
zc}q}!(^CK&pUdZo#g^xdGud+E?&Ap*SB(dICfcK*t4DXC?jP)(@+AoUTRSYW8jZ99
z(M{M@5|S|Q5mXRHtHu3yHxVb~&u?oqrHSwgvbhb}lg(NY4}Wn@Iq$tMG!iKqtCxr%
zYOMB8;8L1@JubI-%Rm%h9|@uD$!a-uXpLm3)^bJTIj6wEf0s4_-X1yJg4+QFG6EfT
zR4IAbDHV+z!-I8&94$W7%w~OjXjcH?!S%boFwW5iI8>MmM`??3qChbu8PG=+EoM`x
zCHOIJ$dZnzrhnUKY}3WH4`x<71S7RDaze~A`c*t`2XR5xu>e!LU%iyXYAw*4JweZu
z95Lp9*LjB#$8Iz%C;x{yTcMNlA{AQX?2a;$i@grwMl7qP#rUnht-{!9AB2-bs)`Nm
zHWcItiG=Cs&YIyB=tO}7HwpoRkA2Aan3@h!D(_ZGzkekC+x*;2!Oc41{4goe#DN02
zm?-PfGaS*<Cq8$(9^i=9C>&{T2dUz?BUjTw1m}B5OF5bWp^cvGdth$Ecmdu`_U%ub
zp5QiU(iZg!ZpTw53{!3&YI&52m7enB-nSgB_Yh(AXni8XbxT1jysncd+#;Iu*BVfG
zr#8HL)PJO)HUnwuz58JfZe<p1vI8Yi0f%Xc>*3~92Wg`@)vZp2|84Lo)kb&304`EH
z$cCGGW@#FXz0LToBlR3lfpw4+u1<mV`z}Z+p4JE0jP94AMP(W<T{}Ow;oK(cNlvI1
zv+^{kwZE>Z#?zs_qsHRotAlW)#^N;U30#a7#((%%;KJ$q6Sz1jsA`<xvY<A^MN0EF
z)y6dk-HzRfUaQ=4H{^9?N6cNMvoD-3?IB)YHf^XpmE6kx4bz6o1}g6Tb4M4io*z{I
zTl0eo$m_xj#|JxrPaGE=m?bOGXpF?hU|k6%<Cl&^FwSzj-PgzJN&?yOO}5?ABb?xD
zvVX&6eW<OZc+YdfllL%c=k9zIGs}tn-G|&OSkQ#!fx6Na>v(W{o=#EtBYynf&C@BW
z{)ibrVV+JY5XQn!nWv-TU4>den9CXcdp!qRd|AuxMS0nHL8r(veknKB&fj#vcNS)9
z55+r{|3#vtkHfkYGI%5}5AW3{F#KX=;eP;>hV)9Wm;+GWV<y)q`1FH48K==0Um3~E
zBW=9*t;ulI{NUZNL!6LdOxp3Kirb!XK889=k!^<ac-_iyUbDU#&WjF~w0eX&^L=hE
zi0U*AFc$Xe%;ZKcusV0>n*;M<VCe9u(K;@3hYn{2dOA5@zCXX-{tGb%Vtoo_ZkOau
zF%SVYm%&&$6PIyJGB*R1DAAXzOfo8ehc5L@kg&V`?m!Nl{DKo>q%HyU(L;Cp+x@&a
zhf&rnD>&!9tR#DuFO6ppEFM6%U_7S)m*R`7tf}}SWq&`^vqf?BeiDm-I*(#8<a|>+
zJSakC!F4>NH-K^^Uv&>-hN)y;fH5UPl{T3)9)I0_oSbFhIpml^FM-vNHN+r)AqN(u
zL9v5c7VtBu)j6UOHPF0QTYT^)JCR#7m3{HdDn3?*qXMAAo&!Fs2@-FRl!S^z%5V0?
zxFvnzYFfljl2eN2qu;ABAD12n*W*498A%b-m%6=V(Wkxc+4N0iPyeseC+eBgcT723
zEwk1k*6Pr{y3|yXrOK#YddjVTf2h{<R#i4WX>Y>PL*R0JyZ(#VJH{;@zQ>R?O^TKK
zZEdO|%*4iMDoT{@VwtrrRdyOoSE`Xai;b|N6I(U7#G-p0>HuV8nvvSxugG@8>5vZU
zow<e$d8(|>JRxndg>CoEw3RAk{n2u}9_fg?YQxJrW7M|=oSl197flI&)`}`?7%T>x
z{LHafHlU3;TWU2o)urE#JAqsenA%P#5gS93iL8H-=h3IPa%Qz;%5SLN)9#ju^P&3P
z(Oyi1&rf*H-moEr5_LIyXD-UUb}o)O-tfNc_)Wu7df)UtO|0?N3h{;=TlHKm^CM&6
z9qz{aqL-@wWXkMb?FLFR$F=v|=LfIUp5Xq^5z<p`AT`rl{vaPJK3Uc)JGXbX7~oYM
z03QA?>-Ty){R1GNuc?>eO)(PzHJ6e9E))VcFqhHKGbop=837A_<G<L6gen1|NX*#~
zD@?rqiVLkohs#;H=sSnt1q9cD7C-f2Ka~osSUkd^KY=r$Jf#Ji7qDPtt<c|aKM<*o
z;^}5YrKDoqsl}W$A~}zbep{f9Qq5umEsRwYDvnCNLM0Md6V`riQ_hmo<-W`%b$K}_
z=IQ<|TR*d()F7}M77qFE)LevIK^wB@TFJ*$?$2!lx%A-!34=SA!NM^V0X3I#AVCzD
zAy6_Ze@l<tHVnS+ub4j&RifU&V1QXC6QIXpZ$+=|sh9r$LuHYaC|NV>AUQY_+Y(9f
z@$uo+mwzt$^8HQI@B7EwZx5d?#ws6T4403;FQ&ZF<zJ(tO2J=(24!`5_3!1+tJV;-
zcJJkKT6<frSpJ{NufKh<E|nF=mn+=<`uLrQe+LE0xws<keN-+wN!&xZ3=;N*l-tTx
zetjZl8@_k<&LS%ZpqJvi92GEq6s=rkzj3!O<wx&q--+#=&rq2I-?=klrS)F;iff9h
zw6!F%b(bQYHxZU1N3U$iEVRAfe3hb5aStJl#%KCbc$r9?t8I1_h5c#W7jb>vm-;?D
ze@$DJZb;)D?OHjxu5P4PR}B1lzrT}D47}^zzP7E~rKszSd~!;=?Tl5sW}#i<-EL*m
zJtHZfW3Wp#JN%t|&yB|1pX6ptA!=1?BJtxn!$NUXVc6MR0j?BO{sxA*e<jlgGFM%N
zD%_>mpICAapE($X>J6QzuBxVWw1T3kf0|?y@5U%&jvE;bxos>>yK5(KcjDZnxtVZf
zNm%CneAUETla2L}sy4MM>b2>jmGvQo@*A>-%YOyB)^Gm!{>B<Ff+)~t*7)QuB=5oU
zC=FY_-hE<u22|$nTvo}#%)fZ%7hi)J+R5RiRSHI3pIL@NKiV<Sc6kn-XQJ|^f8i|h
zLLcj)b%20&Mq#%_5Y`~IXuGPl2d_;_P>Lex9nn4+vo>p`^=MBJzUk8FM>BwHx(NQz
zi^&qWGSf;1AAKsaZ8Rzv+vLi(*kZ;LWOnV5z7V)8*ZCYK{n~gi(H7z3Eaxe`m=}7P
z{T`fCg~H5-tb5R!w<;uaz{6rUe`n2_Yk30#ap=yq9gIDIK8Wv!gcg_s2v4Tn4sGSW
zwfBx<J+T#uH}tf~G7z2tr>#a+lovBsV1Eem+|SZldA-*wcDoteNvW0ih}0@%dT`2q
z&d(2;hnJe4G{*V}qQD4MFH4I%qaw6ovgSJA9X;(=TP5jx6@Lf;J?0dEe*wKZo~^AW
zut(-j<Upsa-V@~0%D3raM$B5OIH3WVHIasErOLg>RV_Go%D&0ONKcwT>G`dGr}$h;
z0ZY`_mS=geFBWp!^T-6NV(xh6gnc+mM#p|4s+HYO=uL~INiX&7=W#{&nDUJ1NIeM-
zo20X^@Iyh`c*<h5t{I9^f24c17)2hO2kYKLb%8wC4k$z0;jerqvWguvM>?pd;Pwi6
zo<Jnx#eUIb71`uqabrrwl3Hds#Q*6WU79^*t^MH~9mg8Bd8(ZoOeb4VvJqqc$$SmL
zt88PFal={<vOm^RPid9fu9S8YDSdyv$rd%c!!8MDC`9YKdVk&Fe{CwtAUjc@(|&?{
zb3NTo4?BCHlTyfW@osc=4As7l4GX!K{i3vN60g!S?HRbu<&n@(kUM@+vE+)AL=Ur#
z&Nyh-;OL+NAmM58<Wf+nNHp=SdA~ML@VHT-K4>vXJ?wB>?TrP;D%7ErT{*R(NQIat
zZA&W0&YjofAvupFf2$TA6^9b{kaQ8+)^mX<ih~{rFQy953TJ-8o=;~La}SOI5npZr
zt7pLvf^soQV5y1Q3Yg0CwCD(g&*wE$BZv@x3i5PPx?6pi;~9=WN*^8$O(NOSXVP|U
zjqAF>8ht00(Y7V)UE|(xF#`nX9-b$>ZjarQHr`)LD*t(Lf8JJ&cT(TT{Y(?&;Ym`g
zS_xwjE%~)UiI;5)pSwvFx_>TYb&|#}M7rWDB~`oGH)ansdWj<@2W6n=uycMoqn`0a
zu4raz%_>O@6m&0T+uaah0%vpgor<7PJEubtJ&&r$qB>uvl0F=`Mqbyfsz-DpAIGPk
zlx>KVdX6Wge?O0FbX*MzeQ!c*Zra*P>GN8)11yQrt8C0#&sUk{#01#1(_y||dPyk<
zO>9m6YkT!}h~!$gCnf|1Hm?Q;R6<J+4z1jt^@wHiz2}E|jwoU{%E}6<>^{W9y4u{t
z-cpG@GT*)6-dfL%tQMPG>CjaE^jH+!{%+PEe3o(0f0MJh;-`~3l71|s*~Y8@^30(Q
zK5Mmdm|fufa_q$X)!28K6?p6deHcOw+7u+o9;l|OW`gSC5801(n*QiIKRtP4m-S9T
z>XWzkInAxX5xnv~&%2+6g$+dxGj}x>{0f%fP^!{Yx#|L(1U94)$Tg%E>zh5L*C#Z^
z*b~3Me|Ab_&G%^%0Ijju##vAWLkL$<DBxQ8J)=MI1FbuC*iGhxD@!;B+48h66|Egx
z7lT#CZIDy?5z>4$p@74L0)}}4bP$v8hPe*-qlGf@a%=zC!$0>=x$BA4>-N<v9o<an
z)225TFtGD3@kx>G`Vbz{i~A(X^w=zwH8)4qe^$E>KA*7-4b0C+HMppPiSvo3dCqb|
z9%3oA4z}nsyf{sadrQ;5xVZ8A%itLZ(WP8odVntNNwCMcM_eETK7ED|)NZyWysmw;
zG*{!u-RH7Oi<l*@Q51SXAY-G-MLDKN@vZW~l0-rnd?GMJo2j58iuayy-J?C_?~`q;
ze-ejL1j|zKPaxc30P(5jSXZ29I-(*P&<asPbXjDJK~KdzAD{F%m`<_siR_fusRt#8
zjp*iMADuU+v!XVC^gjFb5FW$TQn*^KGP#ohpGp^^l`B`W-eNEwp2|SnI~aiJ5b#5P
zQZNs^RuQ-OGmb;)V~PyUO`;qJ?~QF1CF!e8PkwyFKLq?ysGxKSc$-Ue)$4~l{VmLW
z*!1n=?SDbGSn`+QO)(PzHJ8Dm3KIe{IG3?BJSTtUTT71|xedP0uc$vTwR%vG0tSQa
zZciJ?&4F)04hsYJ9_%2GL;ij!iIgandZw~t8?m!M?R2pyiJ~a-BPFZl)1sEQcjxza
zzg#UWVU11)v%LSfX#Ak?|5_o@yCrD7daIZBUzWc-C}li8XM8{@Yw-#AJK|sWcyDz?
z4()&NK`yNy?*nLRV+}sK@LS|Cm&g134S(B6{agx<IbmN>nUH!rQIF@xzrO#|og%@X
z-rtRadh6D>`Q^ixyMHXm_;$(McWSj;bj!b&OkTCc{}#0d4*|Pat@F_@|3t=L{{HFX
zuOEKXD>nOAVpR*Q#_26CU+C4Mo{S62XL^74D^6=e8P-i%4-3RqD?kQrSP37NJ1_x9
zcTrG3C+fCB1Y<;Ft5+LCV&nk?iDgJVHY$zxSrV&Mb%-o<6(hZ+UZ`z)#rE7Py{1RL
zGb^_GeYyL%+`S{MI`4fzmKcq7_}7OBFw?4DK?gJ^p;;ywzoJ=bV6ALHId_99=i7f+
zS6oW4IK#LGrSVrCt<IS#V%|)`zPJY_RnZ3e4LuBVMM^bNw>n%#3u{2VpowbIZqac2
z{oxC>`-H9Diod0wdu{9r&M)^qpq6u2c@w)h-NeLA%(REoUIW}~&gMPEW6bte-yR2h
zw+LCvr0(0S-W}EidKKr6Xo97hXaIi#{-_$b;l_9dg_*q=&K|bpv67E*Nf)@!Tx1{S
z&<jDNComPDd~ugbITOz{X;Ue6tzSI#sS-z}zA4nKeO*Tj+*9#!tBdFpt8YTpZ_1?2
zbmPmPn|6$MtBPirE`%bc+l08Lh@9z8ik(G)A{Krz=7G5LbUHW8!&-DyF>Zh3u%xe|
z!K~?%)JA^06mj6%QGuvWmAYRe(pnNxvxsSOsr|W**mTS7Y~LnNIVr)H2jkYQNYmTW
zjor{JyX=O)Ej=5Bt9uczcn3%9A$U6r!(K&<W2PGLCRCTE2E{<@FUQJYigPqgOR7p7
zAy4%s2;d@mSd^5<=51i4dZm9A@e-De!pq>yZgXh1GMWuZoqO#Ty0tZL3QyW?*47@A
zdH7iiZO^9a>DAjS#MLW5YNXS4nJ*dGUgKHL=^iq$Y8}SD$sBiU63nWSsgO%ov}v9m
zt$QHtsl0PN>9k5zzg&uz@<zy1J4l{;43%GQrOaF0SaPMLZ$V9OFmHe5GPhOf>$&!R
z>MTjfYRBzgD5qvO)n3Y%8zUA!m~FD~843co<h(SekXN_sFY9OHmEGx<*W6Rjtf^B0
z#-O*fYL`cCt9xzTd-k1`LjEv!M~W_nP5yOiA|BDMip?9dan|i;Wh|@e=wnh21l=}k
z3;x>eq@!6i_?@JCGnjvE7`<(4wA<L4Hq0bLfDQ2*m~WXOIP~c%U?ijr!6Bx1-q?zw
zQC1AY$iNLNq00~|waXA%<BIX?0D{Ca{B#+D4T$tNG?)(d&#6CXhJZlIr)FOvLvR*t
z<7kFJ%zrpTXoJl#QJd9w%@CY2Ig}nt5uAo~b9$d6=so~F3z>hwt$@C0=u4emXccgO
zgs~Na?*fHNm?qFP3+jbKhS423(>o+>9)zFYWKSht=ouuqRvy<dcgcP1F8I}64Cg%N
z>*hSxArx{uoy!a?KcqVb3?=gAS!@EJ#hXbEAV9;A(2qk=*tJlf!+X}3?Q*~dd7)~4
zp8-`pUv}%)#dv?VS+ejUa+ZAbR(?w|pC|d}tsKcQgf6owYaJdc{9$~+JmGfU22Ub=
zIL&3qJ*&}*Y7TDq5|($;Cdd>b(0sCN!1>-hx#*H&-NvlHI-nn40QxUylEOi~CY2Px
z=0GY5-maSsvZ8w^83UQRK}W4xgh`5aB*g^un+}@q-5!5~k$98q0oP(t!v6O5CDC)i
zyX3R<gFw(d5<ZlO?&kh-da9j!7Lua!(GXsMqzQyZf=?E--VTCM)b(nt@eD?FfaFX!
zR}q^IKH)ys;Ao?Cc(jo_Kq}fdOj$>fG?L_h2$CkPZUpsxQ2xKfqS^s2hKZ&}m^a8j
zC>qu1DztwwUj>c={&3+a9%XI@q$jFt0J&E!bLyiYRfsTkw{5oH6{eb0%TY&RDr#W0
z#_nKh*hY7lDZuWje6Mx8e8e^bf2CD?EbjFztHk6S8Ay5Af}u<a<V-b!nw|FM25LG?
z%puh=2aKiT($gERR8c6nP++dJyj9_@@Jo+!bfA9&`x(Jpof}tiILP}tt_(IspRUtm
zlkh5_XXq-D;|kjtSL<U}y$1pQXZ0kLfOpoV;WB&EhI8_0Vt|z*fe3Yn?WosQ?-g~{
zUCD@pCc_|xW2Jc2fw+ynTm&?8qN3dZnHJikgBfwqauL>eCHv?!G&>4lC?ICoYI)kH
znEiidUUb4dH@L;7QcQJEXXQHPSJlKIExsK_xUpsWQE!L}3N!QVTZ-C^AVlqKO3S0P
zS4_vK)fo3Jr_+rfr*lNS<eioNmg03IRKVH!#Fq4M!t+$DI=`PfyFctX=lg+t7RRvv
zmjl*ezTfSHbsYC6Jz^D*=rxJ4NUsiDbuxcoR)onFbDAFKn4HT=Eg1)sh)Gn6Z`nRC
za(@<Upxy@5RH)K}7K&Wwk!O#gj33vfjaDP`6QUj4$}luei-Jr!k|K*(+qdFWnZ_q9
zN;4Qva!GG=l1nW-QB;o<ga*4l_jBeM)$)N6T4x0HhKdw7auT)=lQP@7_iTp>x%hvf
zA@_F=9)w>2n#WK!ISq$wLosFBty$$M#ndCE6{Gy>U0>Sg-5Tu-^G4WVUZ&P__GqP~
z_^msRWNxTZzRKrnSOezi!%TL?x_IN%cIHo$XK-fMr#g$0X0+2$fN_1%6GgII%4{td
zr(SA{P4>JyaW0f+%dUybjDEVbJBfeT8D#&wI$ZdoJwQ<^We&cSs%1`~ix<HTpFiuK
z9Xjg`@=*`3k(b)rBXdytL1YZ3Qm9~GqNU`k-XK;g9W0bLnQcNjJ_*e#fDcZ5v196C
z*oQ<FdZxYyPfvWJxSUr&-=1Ubseil38A!Vj--nwBW3Q%WG-Jzz!-S@WLm_|SJK*JO
zY1_|NKAJUv?U0GgBaN3nB2(h5ZeV&WLOaGtTw!F<--vl9+c#lES(-6#MIZN&V~shr
zgR(J>Q9ql+sk9p!rY5=gV9Yr#QP(fBK2Ulj``sO2aQ(%ES*f|qpQ->AyVavCLdN+{
zO@|4M%(odCH$9CzipCjst($*eFwsG+sbtH0eY}yg%fP|68%NL|(9Ex!kkyrylet5!
zVHLj9?$}9H3oj#|WUdiJ-QZQ_`wbIcAHL;<5+AUcHJWSqsB*0}B4n;`C!nmIX*J$T
z=nD9i+STt{<63{O12j^V;itP$0^7P-2t!@|2KC1*;k%ge&DW^lrwV^WOlEMXg3sK4
z_)3WwZk7YKiG9~%y;gb+LJ7|o>$P#+zn%dqs|R-0)$uW~w(|O*bW+8^!+7Sb#9t?h
z4OYT~@>WaJR*UiwyK3R=nO_EyUaqUWc!?e|g+jY4B3OoU^)hhjSGw!mZ-&$5+f_Js
zt7Vbd>dCE+TFlj=IBkEs`zBV;0EL|cD}o0(d6VUIt)``Vtt~p9R*sn!-IbHaFa@Z|
z5SA%*ZOR)E;C7-&&UZ^npxWtS&_NkT32>r!7kNQSPy@aZkhYVkzvilU<)tCy*Rk5X
zclT4&E}>E82uRuiWokFVrsw9{-{{({Eba4KaQ689f$2fSF~}?NoOf(~=rt9xhG3p^
z>HKd2(<wf@Q2%gU{bQR|FD4g6#<Duj_&;`eT(XzpO)(P!H8+=0BMuao6L=XXmlk9)
zq<`Y&^tVrczLS;lnM>zEAdSW)j<Py&<nYb<{FcWH8*siw3S(Vx_}4|LmtT?U_Lsk~
zxeQ7|x_K^mtTUtsC1TUs-{GUn1IP<p%J<wCR|{!HxKeP5Gkj$2`>uRX?}-+6k&wov
zwpjY;+6O=54xE&=50w~xfTv9{X>SBXvwvb8H_@h%o<x-xf~Uf8{{23E@eiGS2ziD*
zwK3WW1#zB9+AE`?@J_9^F`i6XT^5ne+#FH?C#-kRo!e^R{8~1)ZUt>^;z2bzUHBW5
z4U<+LQ8zJjf2qol6^2W^tnXB4sk+_v6!k+fBhcodm`I)_&F$LURlV}AJ<<G`s((G{
zWMYY0-f2%9+PwE!Dl0>BTF9WA6kK12y|mRORgGlulS0<kEIY;2w>b!VsWgXH&CEnM
zGoi1E|E!OWja8*LN~r9w4M&d50f-O`t;{%*NLn4v1KD|H<3))(XuDiYI`;)8XV_eO
zZcTl%j2s`6C3vKXjZ^J~W1Lp`XMau=?SXz*9B7skSuC{uA1pgd-FB(r3X3z0E;)+y
z=fm{&{?pwrYvqtqNM)AK-|x}~K!<2hSCkXcxaHS>EMM+E&Qy7%2@01jowIgWwvs3d
zK<ZBD*fdX`?ww2)&>diEQERDhtfkI+P14qD5mcA-WMNwA4GSx+Ko)+kl7FT>haz>Y
z)Z`h;xLIRrYt#NVE3@8geY)r&*$DZ>WR-oc8s@5~SL@b32=C`*J5{5V>Jg}`RE<o>
zQRvvgMuD+&o;5kP!gz}N$(_`NH}`N%HGlxOqAJf<Mmgd1JSfRJ?`hj5^7e&6bFO=F
z={~LPJ?@E2T46Nb?^`s;vVUm0y<ZR_dgCzjt1HAXPxT2^H$v43*T1Q!3MbP1y7CXD
zE%n#xl$S`|cXWWhb3LH8XXt`UKm^=|E~HWDbXFEBLAc~_ep%3}KP=g_(b`Q6cKOee
znsEqrb<QX321>;Q=k9*`*Viw9`uuk#aKOLQ2=rG%TNT6dy}VkB!+)YS%eVR-gjODd
zh!PRQLvlz$;Z~V8ur+*M?f{*%GB&W*Z>cp32%t3^oV}z^yGRqTCgLGcY^^lkH{M8G
z^(pe28XM^?QK@6oD~_j8=`}t4ox0+vpO?EY%iYH&&uH+!gn>N*-VJdo0m`FIGm%o0
z@gn+&dtAj4J?i96D}P^TYeYB}9H8JdIxDI0SDURZmRi_?sR?Nj1!|(dbNh|}rnMH4
z6s@$vMjYo5ttO>ZRjNg@{rBhZb=X%N^nv{?WbRQf1X#D+TZ`M9N#y{!gJ5d!+L?mY
zxq#zr%(fafIjD05cV%7*hupqKn#jP%h4EL`dEXZL@n+dzx_@YK6R!(r&QX|95KyQ>
zqr--Msi2xDwE-JgZk3%~heU@U%mop&oyd~6BqHeB4~fJer1TS!D~X7fm|vpz;)9fV
z)ZM4&tieW#^%ug0$1xw!6Ir#E(;+Y5(e}(N1z*$Z%mXkC-VnZ%b3T+?^t%nDTKc9H
zyw562Ym*g8Re!oEx*9*CsfvWRWU<S`fT&!{9qw&G^1kBf68VpWgIXt&8<}?Zd}R>o
zmDDSVHffHTR?h8IEdyk&b1Z|iVV#k$?4nxHl*xhTd1`*UOt~V>R5BI?x<}|EPtvV+
z%awG$P<+Whm$6%o_xCN}A34O@8G;ue)s>Xa4=J7NDt}Z99NImwq%v&?loTrciaaGh
zw*=4Dl%9M_rn-k=z-UZM!ZC_TaKS=MF;^+?S`5Ao%YYU}dfTuJ=DL_=1>aGFL-9we
z+rZW^V42oBV3|I+Lll~TH4zWLAuJ<>#cVot39=B5h!md-X*4FLpxp#swp)Lvkj5lj
z{w9zHoPR^33^!sL>P;rBswsNr*RFyXHszM<X={5Hit*mO3dI1LhzHf=u!9t63AHz6
zco$yl8S9JL0mbUgpL*qivdjViYuv20pRtyZM^QyB%r&D@=L<XaujcL=FSQzYh(Ur8
z3e9j{<+&F9Yt1j^vFSh{<*Dg_As;r-;X$ASgnui=28E6DM0TjbH?h8;5NU^0yR}Y9
z_4R1Uv|}e)R{cAUmMpZ~aFzo|-n+0WcZFdc;Z&s4%GCL7j;&omTpLaBu$7>Rx_W5Q
zwiomRr(a5!;bE*WKw#=^x_k@lY$G<KF)y8Wa|**;47!$8N@ha93+nQb!9zQg*;|fL
z*MB*bUVVkQWP+`?8KkQ@e^<**7cy@DMulBuMqngb6?PLAL%;gk%C3nZjBij>dY+>}
zdwZig<Z5=W(d7Kb?5yZo3d>oGLQg>maM%7&=Wi5V=)mf1Z+UX~GETNn6o>P~`n!v2
zQ^Xd5mWj8Bcn!cucTmQ)gxw8Bx;ZyuDu1W-P1|djQax=`lMWFYyxmIj5w7Ly8Y;1&
z1}A-(RBT(`MyjEW%g1-uQjJ>P^A5afN8UH48U_WW(Vi&PKo~4#!b}9}F~OC|;99Cd
z4P&Z7tz)Xu2KNktCSXm(!*49rC`ITHw6VE(8W49rJJmpYRj^7mHd=qLsYXIRZ-0|(
zXphp?x7JUoH-|sDZggv&!KT~HGup!br{&UAt8>QRm}pRMg5XaKXM_RGanNt;?_OUf
z^f4mqqa+{|MbQFh4y3<Q6-CS{{XA+c$R1NL2yRvct`%c(Gm|5or)^+sotsf&FlJA0
z0Xh##**_3ss%uqH5HZFbxgJ?j*MHa76bf0WyP;I5Rk-2yN8qJGk~EbF^UlJM+Ca~?
ztMS}&zatm=881}xI5!RHGnXTtqa1yom*FnbpnYS~=Y}*`H*6E;G^f{&=juR)=)>35
zeX(t)fOu2{8E2XN;429rjYws{X;xkh*Hbl|-M1Jj;`8RJbT1(^0BLr&GJlJmi#%-A
zB`Bkt?>(X|DH}10M4=k^eVmN1-(1wfqXfp@!}YqckhF8@Ph!MJaD!nfS5%EP<>ut7
zmEW55R%jpQxyb-6U5n_-cCOpqRciA}qee`IZhfM=;TpTke(4%Cy1Ee8^4n0%@yg<9
zFlWm3STaMa+TCA3j#RvLe}9MGh!{`(kjD>u%8tN##9zyf!UDK{FLgNf1BMHR07C^z
zYiH`Wx~}jrQy9)++AREf<7iLKA#Wr~4usM0BlDzmwL|(4+4T}-oi_Y>3HmPVz^<2+
zhWUC)X`Qc^)WQ6ENts~0UXpqEzsr;U7`IFu;;XmFlN5p~sm+af5`Xo6ah}Ac+qqRz
z7xv6N30{>aVIV?HzcEb;_|0??oTrNixvI==Dq;)ldNlevK(9jK2zuonBlN0uUu|!{
zC#gN!v((dqu3T2o0mbdqG^kyDF1%N7oFC8En89Xw+qiPk^U2gp?_CZl^5juuM}~({
zWQ!-ifyfG_pv<EhVOL*1L&*P%Wp({t13v*N6kzizM%6sZA?u9}!q2Rf?>JIFxPAWW
zf65XNFko0188h|-_JN(gT>4qpwSQoOW{Qe3(z|{N)=hFZ6NEVvoFUVH0d)PC^OxaG
zF%tnbmyxkC6qk&^DJYjrZ!$9jU}IaCf^RZ2fBp{i2T~eCe~!L;`o!JVC{Gd!>5rt8
z+HD=E(#L4}Q(4}Sgg)-J=^1@DzUVn-8NDZ6UTPP$?Ta?E`|bYcjVy-0IHA8}wRIp!
zCtGbu1JVdUBcUPfkSq<<Li13sc3ZIUvXM#hEQe$llKmK;Ah&|et&sY})=*mEko6+X
ze+BleG1-63+ZL@*606;Ar)*S<I=a^_nK-ucu#T5vX1MZ~P2_(2(vKDgUP~!Y^$8Yj
zs<!aRo?G;*EJM44w#Y=T+D%TE<vX>(k)Kx^<2c7UyJ(9k{Z)}qmiJ@ADL)0Mns}CO
zK8Idg+6rl#iD2VSiywF0uh;jR51ZM#e@PTN;PCcw6CZ$4!gxIxPfHEM$3KRL&6{3p
zdj`FtT_LSSoa_oEQ-5oW03Uc)qy;2*$wM*L8B(gse;?xrla^oZMtijpR{36|B$rrn
zfzt8ZlJ=Q9^E2Q@4EcAr)GDc(98?}yOAvr+Ht;b?&nB$3z!7vn?zUFyvYNn$fBBdV
z$$LKSp6m`FW%a-0j!8@F4s^nz>Bl4JTAD+(jvx`^`R%Ri&gQ~SDjI1I^*ab@uV|^_
z4*D>V;N6g<Nr4F-hku9IOr1#BT1TG-lnQZrn-_n-d-&<??@CaIe_{{XfI_1Ret0ZD
zjR?vFHM}pst<W@Q9FzzIjxc~1f65{WaL3y4cG!S+w2ja=t?y%Npp^uzo6c$5yC}#5
zG)-(n)Ht@5@!q$M)v7u~GP;@(e?={&Z~RI9xmWxd|JvUyO1=IZHV?z*P4#Cm^p8HR
zlXl6SN`P`kwJzl*Z3P8F1?-OC2wcvzwi6vih4mH;P0b4@6@9Ct)yZ;8e=}ijB3-P3
zo2qGD`xysxbImR(M?=IVH6Xl9xTcC*vunriZy!s)chu{y`OJE@5pn?SutjGC!A#VZ
zFz-wRFXAlCk#N5=$Z(FcFpxnckGyoqI;CX7o1rl=&viRa_RVSsMJympxbC1HO&Z5Q
z5Du|A=1hvfb12g67~DZhe_a+MT8w#iF-rIGxQYI<fc`%BY4QnI5cqXj49^j|iMOW`
zuP5=j;<apTkcSd>Vt6rff3oGf?`=sl^HZzkafZ}OOFlYxY01as%S-;((2B_@K)7`o
zm1B_27~(9rh$#l?^sU5ZCF!g>l%Bzv@;?F)1G*xZl2!!m<Or8zf6D})$L0w@H~mM9
zt)sMUY|2-J(NbWEc;0J+38_86jZp3BXanU}Q~X_S{YIfWnCPkJfoc_#61Ntri*zko
z%9%B*VR;;G>5PsOyAhW0lVBMP(wyx|ScU*pCs$aGwJd~Mz|=zQauo<jx_qxo?yvNs
zBv!e;7ZrS3znnmRe*w;EcyLV+;oLq+*!-L8fe7rm5bS2umX*+!hAC8>N{G?7l#u6x
z(^wd`W2QRgG|LD|T{ZI<A&yWnah-as_O``tU#t5|{x`u1t!CP8E+u&*SriVSmXcDY
zx<C<x8~gN9lCn|&!ZxP~_OH+QY1O5r0?T(}#27v+{&iP|f6vyt+jnP*N)u<=k9WLK
zcwIfk<;=HGcHr|+lb!ia#|e4CMQO`Ajr2=J2ETZYBKg9k#WeM5NFX}fs+`wEq+B^u
z?rOrv)qYLXOgT}0Ifdgjx|V)5AbK|a!gJ}@mW$&pXb>h>$Cm<95eLe5j14Q;KYTL#
z=L-8*D6Hxge>JAb)nmG>%yTe=#(M!~+@TX{jMrMf%Is>-;WYSN_s3+{G-vXuRG6Vd
zX|?@KRG1-j^jbY571n8dMk<WO>OQgOkxUcl>)G_q&|?NEZ?*c1(PQ0sSLiVlnI36A
zSBwF&CG>-&AY%(-ZA*ZWT&*r7GUr%I8ZNk*_Tz0_e;$=u0*(H`<^ACDzEdvGNE8J0
zD+@%HepW88G2tuX^0cR1)UPcXS(=MR9N&kfc;l9ewRySpq-5lO<@%nF-)kKHB!16Q
zvO$>f83{h5kS-m?tOD3b5Q|ou3Tjj}QKTjT?pPZpfL&WBfIW7ukXr|6n%IW#j@YAY
zGq?bDe+tISt=}kaQvkLhJPU3U!iZ$TD%@u2Clh<ka7%}CoY<GeXi`O!Sc}oL3#1G)
zVoyR}M%AMHav_IGxL)fuiVqDnhQ3!S;gcN+lV!x|SPZfyDJCEn@)2#T-TUU&#bG|1
zs%?oh7@xE^PO^rU&PZ|`n^0XQ(WDpJ-0Byje<Duf%rDiRnER*L5-r$Qn+|<TrB|qz
zGKcu=P|9WLZ<s|O!~lKkEJ8@+#@Shn_i=XlZC)lp39E87wuO8N#+IsIH=(}?u6bEu
zGTk)!o*zlA%CiI9^cu~iBIQF;jue~DNr7*#@HDK~%1c9h$BJi@%DZD>fj!ToIAn5t
zf2pQWr<pF<CrGv5G&NehJe7`@?{&9)90Gg0y`)uN!uq^M3yk%7X1(369UJKDcAfdQ
zxz@Cft7~xa_Ec)*>a-+v8P->Db;@IUyK|YV+fR79TG;Qj<<t~vYMNNK**XF9TW8j2
z4fm!fS+pe5a&OFR_ytDy+PfYZbgy&8f4d$^BKls5=NZo$`)7Kk?kw*(x!||XdTlHL
zZ@AHVZ9HuuKeK-FrmE++c_z(uKsR}xWSTcyhw})t;@aaxuWmwQyz!d$3l_Z9X5s3^
zQyEjIMex2SOsTH1;gk8&w9Qma;iRinq0M(9^RH)MJeP>dh%8E~h0&{XV~cv(e{9$)
zQvhqGQ!1gnS*5dw{Q5NMmpLQKzoLMKs;kL+z>a|@ZMK$N_4V#w^I8P?wd9^9R1*T5
zyMl*S!^H!Ni@tC{Xp3uH(VvpW1=WDK<I^4Nvg6~u>_|y~&iB33-EY3j?fE`;>wt(J
zP-B0I#3cz$E`$a#4;M(){*%tl1b_WJR+n*O9uomLmjMeZ6qmXX1u2($cruuOc>3-6
z=YwX&XDxj|MLa~y%neCPB+d>*GbCQl<X-0NLg^K*%0<#*(aJg>J$-%BX8A#~_Gf+>
zG;l_FZD$R#^dccQ@{`Lpe5Ph<%evrN=L;=qirI5yKl-d0HO;i2>wNBKUPPiXuyv`@
z7^b0VipvR{yJKcFfe*XRGYOS{({Q0WW~XHl;wns4A$+K<D0ddOy5{WW=(TdlMi&P-
zHjw*e-U!VFpDaJLQOmTpjl@kaY=Ua)iYf58;TFY`HtJ3$gR!harifjCf^zoXrw13W
zZ58jl0__?I#ahu9%l5d>KsX)I<6ZSklVrPBitlk_U+Im|$gSGRRGzheyuP8mjjBxX
zhQ<?bhttvECJt+j4v(^SEw)6Z$%Dq**@JrcihVBg8MYnu(w)ppuS5<zM!I#P#fZ(m
zreSmK9)~YP5lp5Kw8dtNW=?C<<IK_;-;5cSDTwbfv%J*iug7FqlM%_irjz()51mYV
z8=XfQzN?A3?#-SYs`pEOtTH+-&|thW(D%Ne%@%Y*x4e!>yts#)8yx>$D~F;{uoYcG
z`!MoygBN-Jh8I~Qo)H|>8Ml``D!t?a$qNs&{@uHIv|yQ2vt&C#Jf25c-Ai)N6bo&W
zJfM#0_{ftqo+TSWMjG|nWYmp!YTCb`!;N1rC2#HA#Q~r)zGa1f77geby!eDOdCTJB
zm+?sDta@=I;ay9mvv`-hvL^NO`vku38rlUh%)4MGTQZeompGqHb>uqAIHM&TiUnRL
z2qzl+GV&j!P5Hi#LL9kD<{=&>R#mewWVVu%nyxLnFyyF>o=u|X40XlweQu3n<qQ?m
z`_`n5k8scbBDY|F9hyd$mhS=~qOt*dEz+<B1J{PQi+HYM?I0HMDL<dbwOl=OzB1V&
zyPm+dP07i*iYnD^t8%wVq8A!{dx%<_1I@D&q>x&-MGbFo9FDP%l^)J}g$OrFk990P
ziY^b^bh_EnI22nSDJ&dDDko#pK{lr!OT%rkc<EOuI+$92xAFn}a7-f?9~}g<CAmiK
zg<{w_aydzzXV7dCjU*Z0i;L2C61zDymFR|XR*L5?^Ri%Oa93{kqD}ZeiI`~^FIpv^
ztNr|WoO)lL58p5L9R{aDgyrS!kSEZBve7JFt3c8I@yGIdc$&G6mq=O~l=j}a4KzR#
zUT|^7shF&PKtq#xUK<Hr>v`Y{67)-toWR+!U~XP~HaSJ4YpYUlJJN^XdktfjjFRgI
z(6n`JTaVk!?aIbz6w!P{_Ei1tgpnOi0i45U6YV|cbNIe5MEJNAYGOb$Zuxg9!wQVU
zlrOOWAzx+_RXjG|a}S^X{`&gU%kQ9Q**}?-FrbWogA|vy+I4hEM8d`Ljr(+BCquG9
zB}zEt=BP}{PX)5D4ZJJ|G(mxNaTxuUN0Y*c#te%K@<JvQ0mG2!kU6$dk@syJ6s@|5
zVpCHi_c9k=o4d4LR^@JPV_!&V)t}4ZbvZn-SDiN@lSEw^Cr^KSUIm2^1_!<MAsYyy
z3B7oKB?Us%JENn$ZO|wmJzBCbs-S^>HEGRe85!)BA>t(*WJtdA_&a`>Y^9O@M50`x
zb&l?xAwv|{yVb+}_sbiv`%0@Ghi~O`1Y;Ev%8!Hr6$rV+t7M;d4o4;zYB`wNt!Ah~
zs-(RgHZdWQPs^ZEwe8rAd-||L0BFy;4>)9hez+MNL1_j6nj_vhp~VJca*l3S|CKLb
z7uRH4E^n<|l*>sOElDqO0y}T7q8AT4b%k6DD%W*;1%mksRJEBNA$aYXWv~4pD0M(B
zKJ8Qo_>JB>rFWH8ip)wgKMw`_!3(%6MXsp{{aUqz)+nF@VTy%5-&jRuF<eBjdd7@@
zPbY3$ZB}b)*(~#|@~qYm<$Arsa?KLO2Zp0YSlJwLf-1(K=rlgJ_&7W&n0lp~pO<`U
zPYJck8;(@waPeG1Yrx)1B_>xNpY|S6{Ow1Srl}i_eyQadMa-vLcP(Dl&`AY$_RWnb
zTjME<z`SZ}NhwL7Kf(_F+gdzc0e##+!AST1{E)yrefN6d-DlkBlU%uasa$0b;&7XY
zvUc>bypD6E6ZF@CYiYPAyuCEsQ)*4af0h5jPrm;Fvxy68m*Gt@69P3kmr+dy6Sr=E
zGRg)En!^*D+y4P=_O`T_;Y~3U0W_CU5I7{4uNE>4mlKZ#D1X&j+in}l5q<YpG?14~
z0QLPEY#i8@WoxmH1ycr+2!cULV`*%X!zG8Zov+WS9u7r@mP8Gyc=y3{&8RQcbv;!*
z(M2X%w9zoD@qe;RJ5NMp#?zaam<fj7c;;*%A}|*n5r%n<H%?kSDIyX}Hax+fBvVr5
z6%|v;M4~iP+JEPz5>r-dnt*Z6MxrpzM@baMMU|KHOeKrO!HG&CZxxu9Dm?)eLHRt)
zFs<+t9cF255>a?(bf$15<7uvvOk0`XVHIa_1_v&*ceILfpy<d>C8ncx*h6DRbi|(W
zOeaa~Y0Hc>o+!@@PDGRe<t<T=n+`-7$&AVGi~<FzMSqV8HnRt@85<S0fY^;o#DK#D
zPs5zSyO3pJuy6{*l(EbzP2-G%u1F=$CT49uB>1seDb56DgUd4_gNkSmYzcu7D-fpB
z4l{xTL7RbB>_Rgk56q_hEvZ`3x==WtXjdr7xh!O8+BL$1GM#rMD9G1^W}GDJpmAWr
zOG)D#8GlL^p0mtbl_|$CmG*$W`H(qqf%zDTPslvTY*H8!v@x`YWWnS!ddC7S1<Hep
z#4~gle3k`t7<{0)5L^f`Q<xF+I4~KK&Z|JLGLrf{rfH?I2gDMMp>d=`o3#<<LI;I$
zKAQtA8nWF6{4{&OhhidEa9A|Sd`gHj$*eUno_|!Nb41OO&hOx4$z=P6%_r+<2AE8w
zc2M|hkfcfYU?Ste4Du^U?nE}DWFuAbESO{-U_in~pZxr@*gt{$TKM-n@y~z#8}9=L
z`!J8dF@%j~gTd+3rx?k^i)v6!dgK0i!3<4c;;3KO#bm^E^Z4>>{i;{@>w>A~#X&Wy
zv46(_#Gw@APP-hy#+00Ic>sq}ICArlJs9rPJiwW>Qq2QIgkkdqJk8Pbzreh3zWE&J
z1CH;S2gqA#n>l^@R2)sJ^IlP(h@<@jaa??@n^TG7Z{q^zxGZqIQC*DcDXFws;$1PV
zW|Q+`+Qcz>^`^Ke`_HSdC$ytS@PeJ3Vt@VwGcihMNr-2oQPnIIo9&;R<RA(9bg^k3
z#KiMza#2k3dh%4f7Kh>m9>Du4_CK#r5Hh%gvn%I0xu&*?lPHw;Tp4%zMD%87H7)pS
zIr=P~EtWs8%W5Qg;{Cfr`uEdSU5}@K72>k2uV!cbyc&v6v*CE}qPP}s`t{Y@aesXX
z2~0jh!Ti(U&)~Wd5t&F@HWAOs&;nF&V;<|H8beanaSa9U#*pr?on<@l6!sDB8l%V$
z;d>^wY)C66wqi(2*84z1a&pa(C_FygkSJ*`3@K&yo3uG6bW5a$a8kB}Y6<-v^INf<
z@7tnk9vWMWmQzIYHfE)2>$yg@^?wf0BBoREjA8kncvT%&VxRpqEwT}Eow$NMEeWd2
z?%D^gkyE_#TvD2JibGCbI}QeJ+x#!wO8K45@Ao{UNDjp>_CtE*38|so@WGzfpL{K^
zc{RBd{j;fnxoUW}wc*Cqx@*P;t{j0b%RTatR*4(n8<#9(<Zl;_Nojk{7=P?)kswK8
z$ZLQbTmqNb=r{uB+U5W{2o=cKGDM5?XSCm*2Wu23{4-}EO74+oZ({)&S}Px*Qln+E
z0Bd>&EF=gWpli+pAPpyQcN-7NfWx4S2Alz;2@$|<!34r7LFP7`0$;4h+vNdmyamGI
z^nQbpN^%QRxql?$BGjuNBY%dkAcLihEe;695Pt6vb(LO=x=L;yb#G)bhaY!mz12~T
z#0;{YQSLD$cAJp4%2V?!x*{>ZfyCTAdoJ?SLV_o`Sn33LLgYzOFtMD91b}WH!Jwtk
z=FM&Z&+GEl>)!vO94h(lcDSVh3pbL9dlW-|6jkeWNTSW=s=kqhzJCuAK873~eA={r
z+R*8=@gCDy8n7s$l0gyWJB@sda_gbOo15OU@T}?*6`Z_@3TBQ~seYXkwk6!0uq}*j
zFnrUom=m!jx|LJL12jtNj=*G*o1Nql3bDo`8LHg`mtSUYju2bK*`PWT!~PUX6EEKG
zzx?p_;KPfz|JZ;3;(wSAFTRKRHi?dih5(_)BT7ap2OXDy<OWgGg%0H)am|VAkJV&0
z?8cOZ{|A___CP$6LN&}fYNm<Vl(t~%{HkA1s;VPjNMf8iHBdOWPBUjEDhLQok<iQ*
ze6=bjx}zd<%tRdz*QXwmvI)H9S{1ca{Gq%V4Lv&QF&Loe#D5IXG^z-ug5i2AANgfA
zDEs@>`BdO%HY}*&u#$#5i4Z_r1y7x0PCY|KK5pG^ow=SAMeEELI6ya+sbcn;@|og+
zT3p_7Y6U8)Ub@tjSH)n=m-y~a>hgS0Oa&rzU5+mIK9+deD@S{MFnCo=_GVN3O!p?m
zxSE!AHTfoXP=9}F@-$coDg_iW45cR=%u#=z3d7sW&*;)h$z$~<h<jyJlR<s78;o+w
zas`9Ff24&CS+&WU77j}t==(?N)P%(iMj}v20F~bt9`JSz4|v@L5BP9{2W;D^axE}$
zEiiB`FmNp}a4jBiEgo<!9&jyY@Ujyy(9~yzCwhyr1Ah$X5-uj@orNLjAC|*(2psv|
zA*E9H-e_t{S>}@GsY}4PE5?=6=?tM`RDAh+T@1$q*wWw5W|to*R!r*dm?bKXVh~E>
zHg{uzxq$3o23-2|c$jM*7sX+(_x|OF<CkxaetmZQQtaSF)Ow_QMTgqdpystmn2PQ^
zH8tnWkAFd7)C0^W;s6ajWox8EPj|GCv+*6|JS#7X=}LCN&}-xs7zq;3pxXOVO$HZY
zOMXp4vH)5{@biJ1Cjlnj0gKm@Z!NQBwJlRBfeFAeFL{uLE^N7q-TZBDdOpnJJFBYC
zq8N#5y&ZFuNa*%bI|mfO3<RfrhY&?LhJPYRHh<`kE^}v^xKm<UUku7KarF68{As_q
zE(X<@umqCNXXJ$Gp6VFO<}p@#YLtdWjMdcg@$tvSSiqH#>jdzp>GK8@Q}h)9-h1U9
zBgn2cp1irIyRbB1fp{p^*iA;#Wv-3>k^8p~MH6AYyAV2c7lHtpKxV(YK@+~c?(!`z
z@GUL~-Ee<_GLrHy+>KlMoCGB;eX!tB5~Z};<h6@?pkTR<&22GV3u-O7MO8_+#WI>Z
zfOs@JxS>NK7L!}-ZWI!~-(+*!>4uJg1#E*+S!xkRauE_()I8d)G*g$ut7`BKyIhyk
zwcBv|X5|0W`n>Ia<e|<M<+`)QE`89DZjV&Izf*s&7c8+;!4myVhzX%v906qX(%%FU
zYWYg_$bp1Z7R?{cTGM*hK7|f?^K$7rDz+tKY*RyQQ$zbjg=@d4aP1crF15vIYKzg-
z@mi||mQ=_Yo2wvcfOe^n(_3PUs{h-p9A1j&<$%!I9r@$|jG}CyFCkSlNFnrvXCQ;!
ziqU_5U5yc2`($vlQGa_LEg>7ap$-7T5GH+y68u&?mZOj5sH_WcDh}uL+i|@;uc#mh
zpU_7%q<2GKY;4y}{ZIcI$Q?L+(4P*g?Kz|Z428LhiSPi@H|h$#qaMQ%^65_OQp14)
zJfF(jD4VXDk=xx1R~OY6(c9nO0eA44fKV%=4N3_hQ)=Is5VzuPNKM_*Sy>E3udYz)
zOzZL-0rhgyA5Iq>{}(R@G{Kkgr!o|mvj0F70Wp^W9~dZ?Wsx$NfAV$s=f~gPG$GvX
z9ZL(!_+=NOQwW=|JZV48>Z8G`h-A*AH6T77*LN4Z^8X=Uhi^7zOk6q?j6729fkq+L
zF4Vh^^_f4n8OVSN$I#n6%F-J6+{F$%zcGDY9(yV)w)e5l%xn47ARoSEu6F2dy^6jd
zga!8XZNd0bVH!ADf5PHI@H?etr6vDLWoCaK>I&XhEDAP*hP=K%ShH`K=Y7tc%Hs=G
z3&S#1$#BMzG*pV~Dva_K)YxdYeDyLb)3{9I3g3zhtoA-wrIT%Y@B7LV*F5ovp!*5G
zFzwZ9=N*5NAq_eR*Y^fx-v2VwwAuN+{ZZFOW}>$BRXDuMe}FsFVLN=H<3hU^+4EzE
z32cw@=+mRC<$K-H%sJ~CMS*lDknWT@n(5HsWtY3a<W5Rz$LC0#KX*(w_t|S@Y}fHZ
zakejJOKPg+BB^P=*q+xX14f#)HLw5ulFXJJwP-(`cM_k%2}`>{G#auG#nMEAx^11U
z56UeosFSUVf6NL6ZXX&Knk2ZeVW%AF`?<;!(Y)sUCU3Uw7PDDo1qqL;1;t`3>l13H
zRt~QO9(I4C4PQU~Ijp*as7wOM3t-liu=)$Qv*~9%IAz0XcZ``Ol{{>)7VoZmSh5M=
ziUuDCp5^X`O)vZ6`o5$#XV~kixgG1{@KoTMPB;5de@5{YW|)oo@uX=>{l&GVKC#2A
z))lSrar6YpE_v=-8-^{;uxAc8TZ{TjSV-etWptZ<=7VP<%!Zqxak>dK6Am($v@Nbf
zSeQwSjfKNx)=lar_A(s!FrwRHVYFIw>ITCSgft1MYr<`hBOJ*AT4m|1o^++|hPT%B
z3n&ZFf5q`cYd1k$h9G9zyl;*s^V#bl_*|bj5*Dw$5Eh!j#Dbn&4ItVGE*cfCKGppw
zYx}-7Q~R|Ony9$p_9bO)1%FotI@L@nyc_%ADV!o$4a?BiWI+#&X?T(>5S`~WZANeJ
zaWcxr#GTQWd3M!&oarsT(c})Ci!unxA%K-ye?}1W;H8gkT<0<mFc#v=Wf%@;wd5WP
zNQa5;t1(%kCroB%V_*49c#~nNb7bfxU^`BUlnNS4AyuLC{a{*%*9>bN)f^}NYKB;g
zknGG1Wa{fOozJeALA;{dVNqbaZa)~p<Rr@*IT=Vn3)_AS#qqI5GCmGcF#EU_&6s4>
ze}d7>v;xI{myt7g>!e@zJg#nA%asazXB5Tdbx!wQ6~gqZv!%r}-Ne;jU8K%rfW28W
zQ!A7PNEokx4=Lb1EZWv+X7de;XwYBam=(t}^df37s~xQ$hJ&cN>4>1K;BUq%&O~MH
zBG(%Hw#xzhei&8o`fg-TZqF~wH4v6Ue^uzIP?a#N%BCp$s2~gz)*?-y%M#U*$NfAT
zCbucnoZ{R&27{ccABM5H$KjqE9OJ4!7-;uCpYpvL39rVp`-s3>WNm(slr14|Xuka>
ze!YLZ{ddhan=q;fyMO)hmR~q93NW%ZK`jt=fBt*-``h0oR9-A97D9*M{iMJcf2$-+
zay)np#d~rsnHzMmtUCXfRVqu(7JKSj)5mSH#BVWxP`y{wa?64fV4W(A@(CEj6CNId
zcj3TMdf5~r{ZQhZ=cgXU={;NH*kIohMTRGjY%8Qd2E#Is8XHs>23@3dG_R{`U0tec
zg?-1f(HhNQR!A(38Y-xt@OcE4e=ALeOlt4q6eVsM4VRMAET*1P#9ppx^iHbT=?&c!
zny{DUlpeJh2P5PAe(`CSlwV?&xeaj)=mSzm%N7C*9N(Kjv^=k-9Z3pGebU70cIBY2
z>8!Gg9V&q26mFs$bQ&6EsXCE{hO6<XYyWXHoOUxBZ8VF>>zG+zD59+Jf5J!xi&<^%
z?oN><U<wU!MU<|^<(vg(yOA>n=SBSJM%V&jM5RUQD9wkO15_`6le-YSvdKwAi2Pe5
zK+F{#`(+ZAITCAr5FV2b7tuYyBC+5~SGRC0*_hLa+aYC>Lcl0!RN}3lO{C$q3F!5f
ztFrBFac%-C7eK22t!hI<e{8$fKJ8mAW>sM0%C|yP4rSt|bJb{<cJ#;V8YWIpev$)T
z^eqSMbW+HLPWcaJ{LhuQ)=Y8WRhRD2CV`g_ubl-{*ACqr%#!<{RBGiiwe3$~!t{j2
z$QBZEHA?)<WyD4z(zPCw-W5Ru&F4`FPGP*VLcwe=ACkgLj((9se-9OiDtOqz5|vp-
z>JW#@UMYJ%OTAgVTHI<S^JBFEIl*a?j2lp(PEUURc2%&gcFJM$DOR!P$;=2X@}xLJ
zi`o;8KM6Rbpe(piLhq(-1M*G%l*$MwA7H`@l~F*P&E@ku*komfdPROsFlb!kimJ&x
z*BYEZJ&#L5BMB>>e{D$9O8({<pEp6dy7dVgP_<g_jNY~Z=;UKu@v!sjzKq>-0Dz7g
zmh?)Qn<1_buAblP5R?g4LcYQ?6P*gt$kDKIfR<%1S18T<>-cFA`)AmROp6E|-Gb>R
zXnOMV0hUjTT3{7$KNe=W(Kg>Ol(Q08QO_;bU78Qi6&-kge_j!+Q4X#8yqs}@InMy}
zDq6V%E7$T2Kur2$k^r|`MQlF811tAmzg)oVf{EcMMf@TO6RTzs>xSHH1S8w-RvK>s
z^$}U)e7xzluLcOTtCElGO!<zXvuIeNGZMm=9Le(z5VAZNE3vpDo?v_?WX16_b7JV(
z$Ddkv=1(>Pe-?}T7<3PX;JsOcT5>DWFYH9^_sVZvE7v9KTH)?%Jgl(wSxEn93Vs$d
zLrCt?LrsXXaDg7172{^S524h0^qfWJ{|BtqnL>rPbtHJM&+VR-3Cjvo)>`MY?j?q0
zJIx4g^IV`5_-Y<}z7G3=l%b8Rhj_JscHHXt;Hxmqe^}X-sAf2KIVd~~d$ChSLg|Ih
z<GLrPNda0vXO|wz9H-J>sJNNED44Pym2uST=cDyeCR{|{g}z>M=yek+9kydUsz}u1
z$eU#^7Zqg7CFnZc=^e6YJCdBG-8RoGT1P)@!uepJw|7j;b8b5N(7oBM93L@8J3fX~
zXmnsye`d{s-4(2|mmTvN64cgYGTOV+v@Kq~->fo@K96Uck*=2YOwM0}WaEpn{(I&x
zXU%|NfmnS$c#1d6d7|0$^&E21C?6MzV-FniPTN&^`yS<9W7ZH>P(NGRGq5PPtV(-`
zdIa-JDtpKR^z^Sx>;%LoDtmuHHIIUQfzz#Oe;(EU#0+s1_cR{f<Cd)PxKh&d3kuh=
z!r?{A{e`F)ASi&elA6~Qw{m(Jym_HK3W#6a$yVGFQ12AC#OUir=u<I%D_KdLt{1sl
z`BThzP8)gDKBBtkb%Tmm_p$#{!8hd-w6!#BXP{SeI?RhOSddUYQQ!j|l=myDuN13V
ze_nQLzdn%%@mRdm@b%h*@wlV#t7{Kp)fd$s1k_(zdyss5zV;v>JgwiSFX#Top$Crt
zpSl0-?h1n%%t~Ep{Q-b$ht;^vTY#J^A6p1OkCkQbRVGs6r%%o4_4Y;AKMV@D57g1&
zhXfLwpQQ2nxj7)!R!nAD!p)U22-b^{e=DCwAHcQF&m>n3bNPe!eG@l4%R|P87@Suv
zw{dCD4G2LzvJ}4dG2h%qR?6{y^%cP?`*l&YM+!;95n-!c^aYuevC3c&Eq3C$<AbZ6
zA-Ugjk@DKJ{J_f|oW}Yq?&mSvJIlI>wn`Bgf7rP2=B?2l)LPB>cZ{{U?vc$cf8Vc_
zHgkLG!ZAt`I!^6uqj$5v<J%3adB|*ipP6R`UX;-3!Os+^;f*&XuJG(ttqS!P(=~l>
zce5{P!OR!AJ~O|hII&(3M<==qcMYGzR5}`aXEzzOqUHV`U0f1MsY&-QLeumS&gMrq
zPJwm2ZL0`UFW_Z5$0vd}nul)+CEQdxWFb%gKs1UXVDCYk+Lc7tA!}y3gMmnb+Wr0G
z?SCtjk3X03r!o|mvF#lbmmCo_DVN-xGMIlGA)WvA`TL732y^VLLqRYbcH!)RK9Qgr
z!ov^*Lp#ck^EZ^~O&l?2h<G94ftY8Yr@ei|5V8JeRGwqr+DL(yVEY2C&}tzR`tVXP
zunndyJE=Be>YnbkzN6<_u&EFt25$p=$U7Tz8>hApK{B)kKj?7-PMA3>z<3WQDUg32
zuZ=$aMZBJm{`^6@;?<I_Xs?8Ej!Q#_LYc|b#QziBADOREDGU+ELC$MhiKv9uGr1=U
zBjoQzssj%iM{yXfe1lOVGKWl&k*WNScxGLk>7WH6iB|;|9H>gLlSUJ;N!zaFoyxkI
zb-ro=x@t@2tzhmuDysDJj5gXvZgPLU&T9IZcQzf~Wv#h3h|up?=VhzO`dM%ak%y$K
zS+9bP!i*n3*thr3m;cS#NRGynNdNukOMHQvBu!k^p0of@fBon5$ICxB)S@j?NLrHr
z7@75V82MRTU67<y(pqB%SEpp!!y>JiTwPSAWKbUyBSA4WO$ITCNRVt)(-D8-nyFfu
z+$3fbjZbtVN%+^|Jp8D7#tBO{Nj^bHx1xGQ3>bM(|0LE7BR@N*N=9X!;!xH1_$<3&
zuZ41fuYkfiiVqYdcH=?nCRA`UJk%l%AW@*1Vf~2nY5J1>o7AX|Bp0}%LsjC+3#D|E
z$mlvIg<VqPtx#+tw|!_7+RlI3LER@i6tKGL0}U8G(Vg6ouO@t02vFV#i=N|%RnxCi
z2twF&Qul-cFyt>_w&O&7bDq_PF7(F8>NMky?`Lf$w;y+14RNGpU<2PR+eWe1YvFW+
z_P%A$Qd(JdT&6Du)di_1u94#g46jpCk?G1D&52fY%Vs9;${%3irtp8I4}-2H8aaHa
z#eMTP3Kgo&nfRNaFL<x2&rS&-;`<;TZc0HZCjtb%@($0rYFow#hn7Q(2^T`tqL`Q_
z>vBVo<tarTtzmr%vwh;1M=)!7;uEx$eM(yG`daM7w=mu7>owEarx-bLaGfT&8=4R?
zcr;ieu?@)Gv0$ypO00kMza&~43}hF2N3_1jzjJ@icL`lOXn8VsUvA($>F^BllQB~T
zBLZp;bv0(@2+OH^$XgS_m&NSFQ0@lMY$`&^iDJ^tN3#<Kbs;+;$%QIfVswH!jNFRj
zC8J@x!1Hnk@?D?|hfMx>k#|6|$<U6?>=@KLGCQ{7pv;craXWvrW3%^Vc8$Zk_*JE#
ztkBYam9AMUlm))am`mK}Q>wH^NcUwSMAQ4S4oV|5wjvwf%j(Mex0W^S;%&usgida~
zJj?o6{4oZkNaLIJTo*A%+5?+gi;OgeRjwbU`o0tboF+_F`I873Mg~`_ge4TlO=5V3
zF52gsjz@c&noxhqlh~Pb%a(~}rbnWV+|Z?%4U@JpoZM}xjr-mte$}f{-cJwPc<NK+
z&nQ@F76q$gI9I?izdm1)Rb6l(&`mYC;rZ?Tk=r{$5jn&y3zL_5Q|Ji#szq2OjDjzz
zI1#KcleG=a&6abP_A0<)#aSwYFRD6SpfnUXp4ds`sA_*xQ+{PU-zihRR&C~%^GdbZ
zzby^C(xSfJtfcC??P^J<B;9eGpQx5}u$05-8NWlWU@Bflu2YwtaaU)CE|VHLW2Yu#
zTs85!uUyOAmCDsmmFrRkii+|gt%v>`blSSoibs`O{b#BkU%Oy^oeI9BS|9mc>hMa(
z3q)yOJLZ3<`Ek^|b`D#W@-mOQQZmZ^Orfki%A~+);iTrQ`2v&D8iWTs+iB`?H<4zh
z`+UNqvF}MsOmj+;qL$>W>D5VKIzPE(ItWI{z)^d`a?nz?BXLITD8wS@)|}7ZN|irp
zVN;%_5M~h3vaU;@8lc&%akYO?dj>4Gzw0>Rxh{W#3W0*$tmUr)2%rq{S9B4yrzi^c
zi@OMXm*ct!4Dvx;1O|1Zi@>Sd<tkRY2igW<?R3N^ixt#krw6LK%Ciyj%_=x&d~C(j
zRdB9rIUZ|V_T-Q)pRH&7SnVZf6Cq3;<QgA!r{{J!3V&E{#+`0(w(($Gv6)M+P>4>r
z)*ydqr<n3FVoI}Q1bRoWWL%)T!@fR&;3QNRp{;O1Z4|be;*!@>vXA?zD@_Lm3FI<q
zl^1OGvu-9=&<yHY>0gzL)Bc>BsgzMb_D-+6nYqO3N~uLqcr;H&s9QW4XQv~<KBXx9
z(3`yJ-NHGgUB`F^2wqUUyOS(UISS3>v*&-+bJx||(YPss1IjOgF;nN}ur5rEFnkBM
ztBc21U${%_`Um$@U*iaap1Sh5+MWSW0__|pW)fD3K#3f9`V4l%iGfr7<sCE+J23`n
z!-?_P#~3Vwx?!-K_G6qFSjs%R-2Q$i2Jr+)g`;^gLfzuYIQ#GG#O@=jIx#l;XeWQB
zJ(+;|N+-s4*y+UB>Vuuw2p!A)zbz-ma9VL<@wiFC7qp^5SpnU-#Ph9a22qy&N9=c$
z0|;|yvl(P;_v4o|o0XG7Pp4q@-o@MP8zJCR2Y6ik%!Lptv$V1dV1JJTReymue@nF#
zmJVXQzv0?j*PYrP(4s?9R5CF}vp#>O(M_5a>yQz6qN()uTe~vfF`3i2+K(n#8O=dM
ztB-ECo{@ru=V-xZCr^{@0bd8#;V5EtPZ#|LM>zGNQFcwpi391nb!lAIR%5y#d*QR1
z8m=@T$yEhEqjw_&V!5cqj*r*qGcxvi`6jn+8TrybD4}m=!c@(poNfzTsttd4z5e3p
zrfbh;vh@Cnu^Vnox$gRUyMZ<JFYoJA&aN<bCk%O){>n3k;R3X3t?0+DH>1#iyw&L;
z-WK>3jZUSl&;}2@+{_>z(V#17J%hZF*0b4*jj>D4x5`C_dsw>3PlXhL_ce~K%L$EY
z@pl;cc?ZwOU3dC?p5ZZQDMx=>E*(lX5zkljd%EzsJBJ1%9L)H}WV;VypfKn;M1_w-
z)R>bA2J!n75Xm8WH-os6ry>Z4YNlJV$^agZ0ei64%0w01kku5{wJ>h3o7=72nxnqA
zG;bG{q{ZIrd$FXvMu$0XuU&p@^lAE&_8vxhPgT7DMNo9E|8?BX3qXHmBxON3C6DpO
zj4nxK|7*5rqp=lTcve@-<Za3~g^8w$!R#&$)8_qD2M|X2zZucyi;0fEaWhlxAAjX2
zCm*@+(~5$M4*hS^YATg4=V^WvQ+_(I)#}iNtk_GK>}b!b$klZ9@B0&sF+$03qU3Q5
zAtgveHH@0rI*hY0GRzL>3e}rsB>$vr>;Ka@-Ir1IE)xSdHZzxj>OB*e8K^SE0^kp~
z->5PM2?Wh!f-INur!o`)HIp&f6ag@oVeL96mpZaCntuzv7vcE#*S{V(3d~us@z3}{
z@19=|zaOahR&ge|oZfyO;ul&8ro5bVU|y){kAF_z4qqwN%8SDz`9H&_KfX_7<TnH`
zZ~%~7Ys>(k;DD;0tno|%VFXZr=G*`P`UNgnyaphpAd9pPIDmX2@=A<{5n%l3#fwF*
z8s;iPxqnJfl?FHsKy3hO6aNNP7?UN4lB97rNK&bK$F!QQg@YjgfkagUHI3`yFUJSV
zqw38m(_=bfR{l#X8b*F=yrc{)I6vd{yeOGeo@wvsY~TwL$Vcuizg+@jK)tio0<>V%
zaH(ZUtY>h|<US*Rqo(#>;y;XZ{Hx{Yh`KA3lz+(xn<v!UIqR_KPCp^)am{)4Nw>z9
zJ5Uo`#MbVROIfBtLj@O8=vUMu7yL`;OAcH|rSagFpb5CVDw51bo8_+tEP#Ls8>%HJ
zM8RaMPc|yIdPAF=#-q=^4C`47pAxXLKFp+vlupRGL;-hv&?adsX*k}gN0fr&3SLn6
z9DhHHbiRK+tw9dYck_%n%IK3CQHXR+;~g;)_0Att7&a%I2hZklM9U(5**}%yKq&kZ
z-&0vpdOF|}CMN@vZE;Pb8azsAK{1T!qsf({koZisG(;hpXY;H(YB9zKCLy+#2#waF
zpYl4B&*O#OyqYw%0j(=aF{VN-BPKE&m4AM$C_Ln3*Q1@d6RnHS7(;GMe$pA{T%gHM
z0iGTr)vw5Y##$D<yD9q>4!mZ+VWf{pnC%D`<2wt@k#Y3dmD~wbgq4(<M46)jDUAIa
zz`2SH*a~k}{ZYseF0-h28!{uch)BZPq34-qvr>~Rb?TxQX00TRy*)o`&<_|Zk$;9{
zHk!@p>?N(<AcK?4$QdF+EX&|5IHH?rLuURFEAhOaWl6<^5jR-YwQ7=3oY#Ce<5w+e
zq`aa-;WSjhg3eY;wdW~D=?>59wRFqbGR8nEQcN?UOABXe>?>v`pnRdkwp7eS{vGTA
z;=NSrqazLB5->Ih^m8{<44bUad4Cq<jPg4@R|La?R+Q?R=W;NW;OOjqtIA_4%aZYq
z1&VRZLe&dcGRWn9*vJJPUb&dK83+%i3Dp6i9!sfGaI_6+C>8+v$WGLcTr$ESpd|8?
zi+sO%ND0-A)RZzU@(PA7FeTbH{EUi|=-s}_k@3VipKFnQbBf5}gfQM2UVkYC(=eHE
zk;N?J!j5GHPTWZoO8k!4ULcS4yKx_7SeQWw7Zs{k-4bI674v+*JJD5NJW-*1V{SNi
zF0L_Fmwk<FCJNTNYV7c-LfxUKCr{bn*PeiyN~^X{HS14cyw}Qry^v#$8Cj{JOcH(A
zte#au;Y<FTm!l%5a(YhxC4b0=ks*|0Y#F&k6gMB1+yHq!<{0ksp-LpOsAwf($!BLW
zuHYpUGhoyOFVR+!<Qo-PR&XU5<wE0*GFvhS+ixP%YMGtt9Wwb(+94<j8{iq5D=A(n
z$+*`u?AI*F=Q3{4#fmn$WR|y*$e?AN%fZh+dg7XCC@~4Q*+Jj+l7DVB$zUcU%Q>gh
zo|{tjoWchu;j(j3F;r-MEh=L+AhVOYuPE<Y#m7%~pF>k=+=>GR=1f?WeX(H7`;0sC
zy8<mL(Bd|^6t~PkhFg&9Vt70EAuiSWGQKXlUD{YDh#4+x!N75*+zfNMkQfwN=^dKY
z(Peqqh@j@4S*sQiG=BuK^J*|sKAuRM*?Wom!$PHO!SZA~<HNmWPDLaLUB_s>QbSKR
zo;mBcdh$5?{hoY>7pKHX#TFM2t7c-~!4|^-p|lb_tY#dFt0lPh4jvc|Hp!*p5r&XU
za{Zd%QNqC{?Wt&@($m?y2M-Jfn-J77cMHFW;9~IDH#BG{#D93`LqpP=yZ^YUP*#?;
zXjY3ghxMk}e#(S9F+y-tP2F4BP1W5ykj3qnf|AMyu?^Xa$A|Q|5K=n}PW|GB1;TZw
zh6O@(!}0448}LN!j>zl*6D4M{tu#3Jz~)Lxr^N55+(v3wIvdIsmI{1+S9gO93Q~@~
zxw~Pf4g0&hAb+HXZKE+ec0oM9&fegP0Co(e)!Wa*?-NY?GNsu-Km`a^|C`b<_!<0*
z2PIZRg#=$3>up@d;nRP<fBVbZKfvOq@BfWc`J|y75y4MC^R2OrL;6?K54`s=<2eVO
z8@RH1vhjG;!-Vo+9eA4#ARKTvz8n1!M+4JdOW8e~UVlD2m{=KfLq&#2v5ppgKl;HU
zRb!FXRK<vQkqSN=Z{hj!RJ@H>e_y7;Q~#O{-=@P?MAcZEqL}v*0#poX9;)vJR0Y$*
zfk$`9mLcWi>jPXw>THTOvpm|bw+_OF*+*owg!~P56eB8ft7|SJ-MY&V&?*$lQ2C_e
zD>#s~!hc9h#mt1*rU3;;GEsr(W`x_{Z$I(5@9@-@?laSQ5K=L1?DQbDW(qI{TB*0d
z_{A;@oL**^2vVYj!u?O?vgBQa0I>Tyb>if8J+y^=-3*2h%t45}4h9Xf9SA%ySs_?y
zG$|{Pf{7M{)Kh7>$!$3*@TFR&3#qWDOf4^E?tjLdi_e+z#>V&0msTR2Q`(No;g0XC
zDwyR1X;zA%8lT1av9^jw#M`CgW%m-gW}Cd$E6p~s7Jw3aY+$<Brleu335ks8!aSoM
zx>hC2FSZZCbg)N7Ddm-2mLbG&<z_N`9+IA@DeAdaE;EZt)>o3+%b3Kju0TM&&x0JP
zK!1@?8l9rv&!tnCHf+&$ojsZY1MGe{S=IYq-fnCEvp!aBBbZ~+ycCa$=PyO07FlvM
z`u8+W!;0adD;YMlcMCV1-6(eyHaun@ltMhV_P^r8<)F6sTAoE-%HwGAK72MI%a^F;
zZ}5|pd^o#nC6q!wHoavD`XaC0Ej*#iD}RPM8k!BJNaF{3XS9|&S{iCsxvcCiq*8rj
zn%~XBwFv301yy9^@>~+)a;uE&?LBm-JPw~vh}GtV;xfv%y|#L4E|OHLYOm=QHvtmy
zJq?jvlvuZt{)_1v*p9RvyPWO6FGs3EMS*{Lzvuw-?#$Xto449HgCvSw0+&_H#eePd
zDBhjh;`J_N&m?!0OL7s7oT`&nJTxZR7Sc!9F}zn-<}I;e;J@10TxaS^tseL7hOu&x
zEcI;ldz=qn!Ch+z^llk<-@Oe*hed~&nXZ6OABOt(RNkrqZq9bPOljA-j!P}m(&(mc
z^S*FQ)RTS&s-x_cdS8aoCvA%xDSzagg*qj?ho{$@h001ReMb@mQn92M)ngC6a;CTL
zOl4Lo)NhuoJBXPa`5a?FscOT{wXvvYZdg!yC<kxRROd-eQ^$usu~u#68BfiUl15II
zN-3HJ`bFVy8z20}kux0h9PG_0UajyOyJdFPM2^O(JAYy2xnqC7IE*^*5Pv8;6?}Y%
z5Z=IEz!S!Xw<02yKFW|SRyzSdV%#jVIt$wanYek@FnbhGVw1;g^>{D$Lr6)ed>rAQ
zYAqvHj94m5_NWMrAKOr{gemj!XE`4Nqn5Bk34b1n2KF35QW$GV-2LvSc(yq&Mk!|@
zr42uHGQJgs<*i{KaHTiho_|n!JKIN~KJ&aTh2*&#cJi0@%u+cfxFmwN!gs8CZ-9K&
zV1-V<Y$Qmv?%mZwMH;ZFnWQTnGdoBMGoiAugkjP}kdM6{45k*;i0#8u-Sh3{D@HS6
z0uk#MI!pbHp=L_9p#0QFk5(r4=N!Y&-sq3%RG0AF$DM-^v4>2(_crl|fY59yPHQD%
z!i%4})0o>gQq9R=<HV_oq{gbj&0di<C-dp3XnZjJ6ob6&YzH|12Xjokmthnr69PCk
zm!Xm%6qc+4D3>77I0TnSzA}^p*!a1(;Jz~B2?WS4BY2nbr!o`)IF|vdE)xPVIhP>|
z0VbCh$ugG#Jh!jOGC2l+YID9LLEOp?VB#5eWQSWo7o&JYj4EHrFv@F@w!ETav+v9s
zj{qN_gYf-T0W>@E#yi_tuV`e4bI&*cisHQ<g8~EOCcap*`s#c4y;zeKSg-&ziLcgC
z<%)L6XJ^s*A+D7Z!S!6Y+1-^+;kE4+{rUB&NR!I0=i&)yI=a(;0jcyIoX@>U`i@PO
zF_*t>1N&(%Rj`7zKK=r)ev{OYg7SJ*SE#K<9Xak-QZ-v8s<8=sm*1sIU*kcB<dRgn
zOOzyx=hyb--=`(?M#$hNFOLh*(*do<IeRHG<vMl}#b58z1)RGMsy>=#eYUt#hezFn
zad>rL)MXS8iMsiJ6%V7lRvDHU%_W=ml@ED`0RbbJa~Dily#{cC6t3ps35?vaui716
zKN<Ie&{mVps1me#{CUOg`U7P0iO%Cau=X~9tmI_hK}8NT{?L%ahTJ5cG6b9}pmxU>
zZ0HsQo-v|v+`(+65#}WGzU>)kY=`5}IqQVen(LsU2YU2AQo`RxLlL-KccFLWxPs<#
zE;>%<YOMB!CtY!gL$!fYQ)t0^v|+ZZmsRA5sxhr+KF4EgIlq0J{s%8oV_28*r!o`)
zIg>Hj69F`rF~uM#e;dhe9J}$Zujn5T&3yp`A)2;iAU7Mn1vxwzv2(D4yq80MKSi=g
z7MGsIo+Jjwqh^!E+Bb6e_HE<ahlkVWhj(vEZ;p8*oZLQt-2}Yg@W0?NXwBw?U{;Ck
z^N;N>Pn=5&pA&pE=SsoH!QURf&GVxY9%cvw9~c$l`O#_de}wN(&%!>D`2s!>)(+pY
zJOhlNgo{vK-~;c){`}X^e}3SI@VhtTH_uFPR;{qLPBYk}Vhg)zFVxE`=LUArJAS{w
zPC8<no&X<HN922WiWu@pVVwt02JqA_&)0A8Y!hymIB1PwRyPg;4mw0`@uvvuInir?
zrv_msoZNtIf1bS$Z$ti$&k<*>BW88BEh80QcoQWOcQP__mPzBi+&I(l@ye*}58(AY
z`YCeXixs6(dImI2lq$p!a4-fjjrbVE$q%w!^HDklG97^@potatBO}0kwIn05GKxR$
zXe=_&AUP-$<E|n{Y0W@QCwlfGAdYI8_H%V4?h2}Wf6^$}y(vq?h9|Mh0?8Fo4{#O2
zwfU2vG;x$!Sv@EOp4aAi`MK?06;4z9BfO++FNrUcS!l(qw}t-MzyOvrPejRe8OeXy
z%9Tt*!g@zMeO)b+x!p}ixMbFj$V=4fNj30U&BR@$`BpO_<hWb(tR^IHx<3mgkxG+j
z*2M6Qe<*w^ZXh8jD<9E4o@A34a&4C1J-7$npDFEbX5U#S5`z!2HC)A30Lui#k5L0{
zrQUU1iddLunQQ@_Y<AThA9vtm2_|GV@eGp%D8XC{t-byht94C>F~Cj%Dvg>8Jc_&p
z#ff;lex6V<Mlh|mFFziB-++G~w!}n`xeDHFfB)X%u!sV-*19bUQHT)b|M2dg-@g9o
z%Ws0gf`7#oeAA#4BKYk`ezdTK)W&Sz^E2oG2qbKR1qk^dA<4J{Z&=sWfiK&GmW~Mp
zD%M87$I-yF*HV(jz)a;`oQUw-NW~xmtfOVS?{u&r)pke}RV?CDgo111Bdjm0;$!^K
zf9En4R{ec@__{rOtlSI={=v6(+^(1jG7(Wh2t(^ofYsXh!j0f~WWsus5LyRiRDBS+
zhS=@6fWIo}N^BV+1eYOzRUnk1@;#0pFkqlHvm`5}fdo%%1`rb_>ZCxnX4ba9zx=>;
z-(b}T@}1#43aOYjc6+n}bV4UsDdmMEe{z+u;q)>a?NYrwEA!N>Ye*!sW#^rX=b;uh
z00k62ZUn_l>tiGK<JAm;%-k4#1%nD&Gh>XEEO=cRbCAj6bmJ}aL07l!ugP1*(q+{|
zMU<xm<wO$}foVZW^-OcG6oRc!n!-yC&=JQdFfjU*G#2=hI9x-V``!}17;nK7e`HlK
z{;8x&;)#Z=?DESR{el5^uW6KTIOK3m0^Li2nK-eNfL?h+60#u~Efv7Pn-k6&(yt;*
zsFVUEyCwM0J$V?Dh@8}680izmB(pYB5<}N8NK&xV3^8-M$|hs5ce~GfsUg$`0<8ql
z1j!V++j&odS>-Z+Y6(cu$+|=3e~_VjnQmQ4neAA)6@4j*gl-AG?CQcXH5(nyAWgL6
zB28M=<`iO;UZ=!?6jE2E5_Q&;t`tkr*eB=;rcyIIx*p}3(@U75VnkIYCS?#61D83)
zT)z3#{Fw{}g0&gBP;DZRn!U64!ksE-y|Cu^;ay_c(pkr3j8wzEH66v(e*jHZui=Q-
z$Xv!Wd-9xV=1YMAqpy{BvRH<nJZDYw&IBXk_E*8E1(qC){;~DbFll^jS>qLrB%>6A
z6l-jQK>jhfpFFavO(1(2lp9wsR58}WK<WfLAXAfKzhy%}D8&Jx6T-RZ6cDQ`YT>*t
z)~LAVth=7gt5InRbkh_Te{q!^_YR#d<pVWz#L?kgqIL+;7OEHgTGcW}Le48ybwJcD
zYc&78HwEcR&p<js&DAw0T6i3BmQI!+Bzg{Z=CnvcGW^0bd?Mi_kBmtQVzrY{m1U3`
znF%(js*~&E4qKcV@=!NU?jTDC(x@dqiT1rvH&Y=mgxf}Elq8|%e}ev!nYQesZ4GnT
zEfE?jcXGP3)D8}1hpo0xWtcFwMAv=KZ|$l>8=>kTV~J7obizTcHLccZzYMvVXC%$m
zIZeZnV($pys688JsD;+xIzwTe3Fc*nuKQE(mPZz$o8U@1UI+^%x1<Mku(pIH-KnH|
zGwn?eN9+~Ov}LBAe>E)HhUdw~8l85fjcYl3&a`cvQi|8K8w5qurOfH+h;C&REhCJ)
zb7H~`jM<TbUT4;YROL?z^+pc5P*SR)joy+X&j9?yy}g4+lWMclXEKjiPo7&7yS>dw
z%}MD7dqh5)5sJL1J#iNGdk9HWDkuc%r+w$A%j|)|rN&v$e+;)PjXNu&Jj>B!vZmP=
zJ#Q%&T2g-31De>0>B;(Ub*;fI19B4+_j+RXD<-A5U{GC5v?R42TrA?PFy}WZl)%{g
ziI#Ra6Ck6NBXkTvhmE76Hd;c8+u{*6=bOrpCy?&&yis3t30I##LP5r~)2lG4)Uu(h
z{L182Seks9f8V9KZa5mVWqmrj$@AZ8{gZ337Ygkt{l&M%8K^K}T6XhKUCM=X*R)<|
z;3U!WW_kLe+}JqD2%3`fbj&?9v@_ush{2p<V7JOSz)^#sD1ng7h(OO8ni3>pI3C8k
z=Mep^;4K3!ZL`V}PtIL$H{~NQB^)R!_am<I6$So^ElP#Y2c~@JmP7!I22Ny$Q6o+6
z^Ogvy%)}|XR3URHH#1ZaDI-yKZXRVrJ~)>S+A=MFJ7rBmyLf9w3ixiX%rtWw)2iZg
zP(-VbnTBJgl_j$%CuKYIxq&kWWpKAERWg9|9t{Wb!e}M%^<j8={rvFzWWn(ogmo@<
zTjGm+WF<Edn>N;MfBbFx`tWfCi#EX_6O0C#P6c?j&Tixr-USLAsNyyrs8Ol}q!>nT
z%0W~YBwa>NJY;h{%@C7hW@yR+=3<1FvW@+h(b_T>8v`1n$tp8KJePVF9E@ew_E}Ro
zxp>3ai=5ymmjl}}B7eHyPcYgr$7^Oa-^RI3@p@=U7&2cLQ@~dFhF8=g%VkcO;`{P<
zd8bMQ)RB<#U!t!Y>FPm<aAv|luG`BJTs4^SIC&|k+fh-TX-HNF=tJ+g&W6m^=?U?3
z+NtGHxr+8Hi+3-7KZ_w*=BL>m@O^DIN&|xa&P=Kr(?dUCQhzo2el`s)TCT$jnsV#Y
zgU^S^qNHLj{JNs1c)r{KRLr+4D8KB3Z(wb48a)O_tPO?*t&Z?ji`$t;6hr+OzyFUE
z_ZVEFc#ZG%SE68t*Ci+7b;lr7R}#=lwG<ZwQlkEJJLM4R14O}L$HDF06vw=SI$e(t
zxmnY3O*I|o9e)HkX##U5OuM};U6FJF5^X(&!Bt{*={~d6f(hQPqF9n2>!DGUg6ImC
zj@Oa=2+NSJNXc~4_j}APbnAJnpvLAqWN!o1dg~xL->do)QbKIufzoJXX^zF?5p~q!
zTlYk8#Z^a8c_#H9J7(*4br&SApiOmLPmgZ$d)1#IcWxK)c%@0FE=Fq_Vvpkxs&D0Y
zo~NHktI_6>p$BGLt(FeaynZ^q&TPgz66FBxr|87vt-wq3!XI9Rlwdo~oIOF$vOHh&
zDZ^~E(;nK@ntq_C9{jej6ON^lm11$%o*Z30>QuYRkgt)|ev}kPmk`}DJ{wM?rFE*(
zdn4DK9hGZ$FG|h_!nmgRltQUG)R$@9G9VVgk!`f4h+D}3repH+mMCqiSq3Ql<(Hq`
zGJ64mmmA(PC~o(pFMw5y6hl>TxqUPi;KMFI+o!aa2`fivQ$qsci_MoCFt*5-{I{~S
z0d|&k7x@M@pn|!9Z<jig14MIh)_Y5pSNPRKK4@;m!yfCzID1*ipU&|1hK`pMOxrN-
zc+~NNSx!!zR~LL*Iy>2Gm$BY576H?j$lfw6Uk8_Ov}FMt=ty_5DnJu;0GN~ILr2bN
zg)|2AX3`3I^dr=@l%@3yJ`;ZT)m$I%5r_Y=;`W<LMJm(xi(dhRW8L+N2^R){msMBs
z7HaaS#1yP!?=s2`-w?$A16^C(m2m<S0ysC9G5IPK0Wg=q!U!jqq~9_ZNd}yGF`41W
z@H%o}|9{}(O;J)G+gig09H<XUEWRo})<H^6zfW@d;pO!4^68>+vceHqHGO=YfNx0t
z1j&ttCJP{phUw#v>F1XO;4(Htiaz}E@#BlkfTi#z{0v{?zxR)qKVL@pV7&0kO~3x}
z5^kUdp`Drtq>#=|fBiOnefen=TL+WeCw-8d{`!3yTE1kLec&>WfAYiTj4H2%^@|x*
z)M<UW%eFdj1Jo!<p0nn+=YNxRWQ8F;8FE51oeLj=js2bT&ejDpO1LspttSd&VKQh1
zYCG%c8cVhqw8fS^t)h)U>#9}q2npex*2y|55(K{LSoC;G*8}WK0zg5g*EaeCJFEz1
zeHL5X+^prx6>F4%f20;hs>)*%uCm5COhhjiJOW#pTUBO7v>MG?i;tQ;mn>RZqCJn<
zma0g|_70cs{GH6-pW8EYFI7+~Sx2-<X7tj0$xFQxN2~vi`TjA_@8w~;j&>xkgh9P-
z$Es&%xxz2<9h=Y9QoVn6Gwe3mp|ku|%M}a@504K=u3Mxff5#bz!@9Dlt*EN?oe|6h
zKZFa3H#j2kp2)Q`3m=Vm8odBsaf5Em;=SfWhxveo@n(zTzF-$zCg)&-q&@sxPJS4a
zu!?RvI%Ze<h$GEft~t7#%n^BEjd?s7Dzpj2!efWAxWFLA)P;LJeJc*(m|NpwE`0&(
zl9pXCIyEvEf6G6R=iW+kNprr9?tvtC)6!*5=Mooz_2DJfFX|X*XF0~@dPce-uu%9m
zu@(kXMmXpHC(>vaiNaghlSp%k`*{Do>K0Uji{ugGj$T|5gsfvU!sA|C87KG}CC$C~
zQ&W#l3rV=+)DIH>l~X@p+<)pvv-+HkKUViBs{|=Ee@c7810qY5z6R~E8<Z;O?6Foo
zRm@p2AV0i%spEV=?vN6AS^QOaD)}<#%MJN(POt!OK!7&=mRK(RU$q38bHnIyOID5@
zZK$Kqx#=0H6n4N21iYXRuU08)Kk;p4JM0_KlN4ZUajwqu5}}3^>J!~3lu-)}I=!A&
zMth;4e<lgK?8F@Z8&jd!rINpbz7{_PaV~<|0v-+f9c&C%)QU9h`SzgdiuIkTkJVK!
zcQ~F#BaP5l!%t8(=P1>B30ceN+B4WAHhpE^$tm?5w7^hho2T(>PjLtOKa1a?Ky@D<
zT-<#vyFhEf7kA-=HXNIzS&by?yex=ytKG>ye}TgbZ8TTRAPDUYhs`dh>|V45`&|r@
zfpGpq&MhWsK1%WLg&eJFu6~=IhoU?e|JBE-nFG590bM!>+5rMMVOqdD<{8m3gOG9Y
zuu<wpSj+%zqfjlZu$57b+A8;FE~f90L%NmH=o@fLD=ffc2@-9P4xBV^SHj?BkqJHo
zf0@0SM?EQTgr(-)sIOKdcuI+V-N&@^{mW;}Rxmgc*6X>4trhbM)3V?6&73jehkejk
z6i(v?nlAYQ_})m27S7T-66$h~?}H+fT9FbSnvWbyvFxX7Cey7fy|zLN*ZYH_XN-g(
zJh^71ar|nq8-Z^RELsZh{he4;Tl;(Ce<@R^xZ<o#)-ccQVq|m5Jezdpv7{H$wGEL=
zh+_Q#P&?9SzOH%0oWD!y-~J)ufOi6v+E`tt+j3ETHlj-OquthdIQ0Odw~>_0Cz@@;
ze0$)qMku-E`A!)}w`)mVJczNutl(ICQRCBvSF97{ZBTpp4hw`+UvFmxJ9Tl;f6jf6
zM)%h3sn(W~eb;gsyMDQ3&V6k@$<%di1TAj8YXh99fNZ4C8(<+j3FPOjuG`y)Xy}kO
zH&iCP_Qk%womQrP9Nps^I8=5}WX5-?r<`dgM{(w;!gdv3PM6tUw6?>Ng+2;+=B-}n
zqq950Zl@YI^N%fSF%rSK;V7M%e^NBf*1?YJ;q<uN&2zDioeOVvayUxAWA3lbutpOr
z{ZZ=sK~p)fI_KIi&02hy3T>}bXbmZGJau&4t~_M@;_N5?R{J#`M5ZL0R4#R=z^$b*
zx3;SI_DGR=pDh;b=E@z=^jpw%H!q@h#kE%^N3^FF+={j#S3+mYyEo_he|*_zC@d={
z;x^UF8gXr1#gXd<p&skPZvAfX-D)nk{HWnQ{$hOhK;^Yj{qp%buSj#cG7g-V@+QT}
z^V4=0xVEEUo}j40sk!tARYIo(<9JXfW`j`p+!U;l)}v@I`NRsW3!FP?46HF2oR3wj
zO?#bsQVOl?(Dz%3=c@a-e}-3C4)6x%W5(A#O?<`kd;neFC}AIhx_TZ=r7CKbBa+P#
ziR|pyKKI#PlSM}HcKM&{`~)404!~sb3!qP`8KE0x8LbP#>|OG|^D=7Aeef)g*ejYA
zDl$uyrt#EadybPS0Pc3!Hv_L!0;L;(0IV(?gUrizs^wFlt0_qeOk^DaHd2!E<f$S}
zVa(M#C19k0iS^e#$r`w|K$KeRNUetI*L)h&jzKi_Rb8;Q{dl$yrt{Hj?~1ZZg-ew=
zzRa4x-#=dd2BRi9E0^)7G86$Ymr*q_B@8t$Ol59obZ8(lH#V0ck3J}WYj0b}5&iC8
zu^-a{-MjPJM-dpvqfV1Hh!LegTSFgYnQ<7)q$*K{^Y43R$xB%wWlIz(Acib0dH3$l
z&g0CP-L2L%q)@F{pn^tDDTty`YLFtKRt`$!4=AXKcP&BDy=F=wQ}u?1!s?H5D21M&
zGd|%18W=Y%2hM>wvA0TpAdx--2h!~QNCFw&!hq;!+TU$xRiA(f()kRTF<zDAQ&hA{
z@pp6JgwAM{z@$D0<YF#;juu$^Q<lJ{z7|es6(3LVU2Z;#;EkHZpJF_$1#>XQYpuTl
zb<k3R76dQ*8&H)rUz<=6GoKgBl3aaW$xtBscv8^H{W}g)lfMgpO-fV292iQIx_r{0
zYV+}60OjiMLR`g{-hv^l_frX)tIwtyv@V|-jG{g=RA;I5ekB-_n?q|b(LP=arhVmR
z!K}G<4R81o&|<WTkDT#tP5u<dvqt?5Z*^T6!BFCoS_{yZuOq`7q}x2>sr9FsL(bXf
z#jjlYyBS7qeKx3nq@=Ew+(MzHEETRKeH8c-7yquNP?QvNKqbZ4M=pexb@@1y+-(GR
zEw$SS3oOli6jxK~KICd5?){(~>QXiQf`Pq-Uz!YaxYJnIX~|0^qE$Q~`nwovsg0_k
zrT!G=ZrnW#zT33>P%WWZUk4^{Sf;mHXtntqa2iU9T`Wj{PKmOQ?J=0R`QXqaB@wp-
zl!j(E88k#m$;WE!O;<p{e3U$n(JVdMJbaT+Q8+#M{PW?b-_9q)7qi)X5r!}47q2E4
zFA}0Y9S?sWz8${82MBRI{5UyXydW)6pnJw&xTMEZ^uUD6NBH0Ili}#<<-!O2=XCaF
z_+mZ$>0&y6pAARDzkdAI|9|$&VsU=?*I{@)UHo$OGM>)g4!>T#J^#zA$vX`3_J^O3
z{&#h9F}aMtp8e_R(<k`ZRZ=dodGscW2Crr*(d0pbzWd>PaWp@BH@WD|Zzi+J#q{)s
zq+h?N9z7}iIQ;uR{|)YsaIV?a+1c^d4loAvX~F$}=F`25ur_$-r+fOU6a3)1XBfVk
zpUp3h&QDG!LHva<JU>}1CKt27%a4D1zxd{8ak7{|o&Aev^VtG(Ji|alKscY}nJ8wd
zy8Hm|2mbZpo<1F(U(8RBCW{xt^RJ%`KTX~*jz6&9^OM&T$S_+>W{XRYygmtxb~(Sg
zIGtR7E}`AO`hN0idh+G`eK)dagVOjI?_c0eL?JI<?EY}M%rkxZCog)camjnCSv72I
ztREF`9N%=T8*Z`P%V_zbdzng($TI;z5VbjuT10{XY)2Bf;uis<9}VBkf0_?phtDo2
z-M#({+Y~s<{k>W=M+2f*&M_mXm1<;UJu5kX3YYu4_4%9EujZ!+v#Yb|$=S*5b(~+k
z9^TOaM?XJVOkYmVri<T(Z83OHRS8K%_F-zV!3NSSR+x;r#X@;QRRvI1gaLT~Is!1(
zwnbz2HCIVcAFqWlgoh`Z+k~)|8q#th1;9h%Lb3-Th$4YIRLl*pRC5`SF5o#*H$(t`
zhw#+Rl1f!8P)S~kfS`Apbsh8>j+M7S&COm*Xh5F#3GdJo#4Hlh;(kEnHr)6fym%6{
zRChrX<_$#QbO%JqtTz#bZH*{-Daw*ht{P=*tRJ<9BMKNGToiPOqGC>!7M+9SN)ZvU
z879u4QU);)QY``?c({jB)!5k{WD#(GOg4_yQKrEWnr!+ggJV@o31kd>uc-1KgeZVE
za3p1k;5TT-W<WfVXl}LN$v_pEzJRC-$^&hMYT5%}aN+2@$z!$vcST?pXhNAHnEPW%
zp(!@7Tgu=r@LJR#5$2)d6i4q38LPy(m6#kRvL6u4q8q_1yh|`ky$)s`W0=Q(80ImC
zm-w@+-&PHJcvM3b(aTJu6p>4UJGY|#N)a~N^dbRoL9XFiQOY6pfadQZpfY?Jpaq@|
z_J|yq;o3l$EO0J*EFlm;2oMAiW^W$W5|cR2fzYH)Uq+&YH!~_pPAF~BOnVdp364>%
zu7v=vMm4n|1cF)#AXtI0NYx~Nm3L!r(7d9?7R0mfE%>67`R+ghi`%VwLS)A2+u=Ls
zZzx9ARn#6B+mcT--{gp#pr%y8YB@-1F%jL%R*pL&ccCgbdaddmMHQ3z4m}ZDQOctN
z+6~@vUv#rgbhDk%&B`^p$p~ROLYUIn;IScNV=Rz#ERb{rM;*aYM{v}CJrJDHt5B|@
zWV!aL?0?Iy!FOLF8pY;wqill0Ik)KhF^-9V$qgAaqvTGFq%}6su&VngMo_z%(`%&h
z<h3IxUqxo33TWI?Jc5ElY*iGusStyZRo!1!fC)OyT4e=N-cGX)k`)AX8371aLcr3%
z=Z^^CTXn6r%0*GmS+2Z)4^6Bb6XZ&Kn;9fnTfVI>JB$nf81MrdWe`CzkUt^=NWU4}
zL4Xj9ip2IHgLB*F8-`06%+bEsBg<eCeG|4wTy9DuK?13&M6U8kGIWgShYk_Q1B`I8
zhLlB+VRNgX<=mPSa2iEWH>-6Ig_sccXk3XP?AsQ6n+PB5F5M)5{3ajYiTtp`y6Ddq
zn;s&O-pHM|Z2;Uik><LQNORsLk(P9wNNvoYHs((oQ)?a{<?&G-Q)?bmYaUZ;wuh%p
zW{btoNmKy|flG_<@{pZU(qrMY`sApGRzesvV7uNf#6d7p60m_iE;Uv&KQn_W0xXGG
z$&Z{YQ0;@g=sH_}ATCi&J(d)VLWcBO3i0h=`G^pQz)9c9E%5bG2vj(*hutp4VG2uw
zF9O|Ae)P>s_Kh>Ab_KP=Ha2w-U0_Y9^LZ_WWK?)uKO)6JoU?FD)z7oV0Hi=$zw1gr
z;f5|?k?eJ@-Ya)`ddTH2h1-o@%OoO{qR2tA$c@e7eQh??&9kV|e|E_tRj#v088bu~
zGej9PMA1FeQy}m}6kxwtwf36ED>gwXj5^_7yMqtNPM!hUwGh6)MYqau7zp6(g_9q0
zGu%Chp}s;@3pa`QYV&X>)z+F(WkFpG1WUAs42ii@uA>ZU6#bz7HW{*Hft*B!QkW`g
zP`wd(HGP7))h!1*e=M=(`iI#AX`6c>ZKpku);<WjaW|ukyBTHN%_!q;MjiQC9Xr>t
za~(U^b?1@!wUPNX+6SvcfPj62fyd-K`=IuIX$CQ!)gCE!D>9{C%OI#hS>#qB4ivnJ
zV!sEZ1dWts22D_frI_2M5cmha)hMA{Rc~9}1C_}wieFz^e+i)|yvS}Bf-_42Za^2=
zTn}3z<kK6&+yI|&HP8{v59k72fUn~-3O>yT-x$7wgbD0AsKYzpj9Z3_aNlkP-0bJt
z&v#efKaanj4KL5;FNe1$msr5Cm|QOU1K`tSc8S4mxe&BN+zI(jIg5Txkek;K4GXkr
zw5zw>;RGNhe=xB1w7&7Lyq@+UYMjn~p3bI=$sDHscGZU2cfUVXB>L4(O)b{y$3+FM
z=w~D?+QCi!kNKOEH}n12+IfJwjgcSy(uaZ#gZ%duZM*aV%BQB<&v6rod!;>W_M*h-
zM@`gxhq(oEu2M_CpjEciNNkO$NnIVM;L~2VSuuI*V26)>X+arI&)`caS~#`)3E||U
z-RVH5$(CzwB^p-$>7IUA=~oXv-IL8PE&O3>jgmHKPf5Gbo|3MQe*9i%Cnrcmdmtxd
z5xf{{Hw2Jh0|6BHRc!Xue0O&8f9Zm|-<RP{F%tqeF_&>~3lo=(&<7}&U-L4Vf1&N~
zFTZYN5?Bgv!q@Px{Pp;<`L-#=M^FMiHLo9=@PZHoI@O@D!YF9I|Ixf{o=e!SGs$h*
z4=tPT?@dwiIWg!ZGboMGdc+_|rj;6Fyuk3PWsn}E)XV_w`C;venN%>+^3&eIBqB++
zg)R{_^Qf6CtrK$my}dCqvhVqqe+p0AM<;#!)+?1uK0nQC<g6g9HR3)s+mCEvtK;-l
zTv6!3p1{=-eJcKnUVPh2)FE*hdW)<rw2XDas|q>HmKG(UQbLBrTCkW&cA+3}{)l>v
zN(Oo;1%iw(dxNp8eQhnQs<~Yto4gO~oGzbLLMv6t_Yj`KdaOn=7}FO~e@7(;Mx$~d
z*l$#vUv5nV9STcixes$n-41w4*rA`unTfSd<qHi&XLVh^*!2pu)i?+qSqFVE=&+0*
z_5<iE=Jn&y+ejJ6kh{yhj%Hu2u`y@_S=aTZ6X!PBTQ8rj*^kX2`r+(szlWKCkFBqW
zvtoDptUxEo*|Jx5ed(E=f4v%lU5iF5(5kc-R{Ii}vH|rXDwQ?D+Da)}K|ZGiKQ;1V
zGb`@|>obioJ$Y%RA)5mYyIu|c8i7)xU|+OmQZYsh-am|k$WlZ;dBB4)0*oq=m7E;t
zU`ZGYWo@kFXeXd#E#Yx!se}9-OVTt46XWQl#KTyI2W+4lAET1bf5$m5dR3U54<%=m
zdM2GDaWl-d$wz?}SkhDM?Q?3Vqf<LO{V4jYQuR|1+Z3fle7FG73(s@DL^RV*7D*^n
z)hLjopno)BI-N=9MLqatCP;Q?W$i)4?OYd!%l;#ZFe;gFD6A2iK=(xPBl|%U6_mXT
z>(>{UqGist+ThV<e*<`MFo#uRYU|r<p)&X>dT0TX*$H8dbGPiqQrw2<+%HEbC!Dku
zx?TqzdC^L)LZ=Dz6<i+o>T$&gG6p=;xd3?qR7zL)mOeE^PDEX)iadW7NMg4Ei$>4o
z79fX0>K`3+ie4`aItAK-ptHLu=oI~}5Ohk!%Y#mdcsS^ke+-_UXem;{60FG$8R5NI
zlbd!OQv&LlMGxoAvYsH<j>i}#CR8Sw3CIXupsHwJDIk+oLi$la_C7L}1FuYz*7g%O
zIIRCE@R9-cLVe)f^{!LyyAXK;O5)UOS|x(AjsmQ$gwThD(CZ3;UMO&t5Oi1wIv~Uf
zu?DPIGuV&tf6nyjIo6M`j)eMAU3t!;6olstZc(3U0XydI{Op3H5lC=kimYvUbR<aO
z56&`*UN6it3bX}T#zWe#J7b?N<pre%^5VRpK)fVAr`(}LydXZO-N_PhCU;30M_K!E
zv_ozN7P5KYwB5StLpfY2l=C>$ye2qo#lALsB3bdde^Mw%i=4amp<Mr@Yil%$eNG?Z
zA68vkBWH-3Y=wTEc!)eFjq)o-!~(IE|LmAppv@%8IVzUgoFAhQT=4Zvj4*X({`gWI
z!5apY5KiE~HC*qhiIurM3$B}gn^4ScxUPvoh9e~SAnfMT-|ug~ynf-3)4cx^TF8do
zo(5L<f1~}W6mt(@#(vEOiUm^IHX+QTAugc+%~f*SzT~ypkg^=sv>O(G4aJ__At^O%
zA1@EsST*-DEX$$<)UG&p`gK^CW$Sy1x^<$2UjZex4L`YkZWVrpXZSl;<yL<-o4023
z-0|v>|6&B^sXIX{udBf#45`gUN~#bF$z)2re`S9&S3Fd;uPcl*#N;U_ERUm*L8SgV
zb#uu;<y^#A86g*S!WHE+6cpOFq@>Bi5rT1m{f@*iKp`>PvfF=u{YVYpx!Kd;e+$H^
zJUe)8nj7UfNs{rpNxT`N-Q?YxR6W<uM;&)t)VIA?kI^qj!Miy5a?exKtpH^ao|}P)
zf331N18tudQ^dwpuz;Y_J@v$J9v5x|WL{?O9Ay=>iM}XD;a$9ia6{UqtU1O(mrGi{
zwkl>4vDC)W?)}~1=q;V=#%Q_kFkOGr8}B`YHy$T@i}mhvX^I%Odkg8!s3gW=H>ww-
zc_*P6YS1t1r?ly`xEr-ToIMSD?aun!e>+-^bmBO+&o?HL6ftk>*ckO>Qlzw_9Idh?
zvHSGZ@p5EspV02AaUSk6I2YVvaBmKC^r}inD?9da!Q+g3#F1^cCg>5fuP8U*y&Bkv
z?V|`5-j2~`e)CGXN^|NCf4L8J#Og^3%5{fUAGAJV_0-n&)=C`bH>Pjf9f`NIe<|L=
z>-kVsel)_^jpYd2P+@GL_c4oZ{aU(KkactIjlfVhwvlOEru;SMtBY^^r2)qoyzZ|1
zNggsXBvrh);ZSka>wd~`A3Ej+%B`~1!fsG0QzEW$GVefGtxDv%lX=f98YpXD`(!@z
z$u>lzf7JFx=6u}ix!V_>2P0HEe_Zcm9)n6{cNd)laLNi)AzrZ0K<pijs}Se*8N6_u
zQkR23p0^%PLkb+H-Tf)<&v(|1;ntszyJNKQy!Z@sV|j7n*FNjc=seLhUi+Xm8ytz(
zW@<&BTp;S*85_|gELO73-i|_{!eXcB_sEq+eSg_R;h1VP{`?aKc(LedEbU7n8C^oO
z<aH(r2VZN=qB%a9)y%IH`!(6rLa|%(wadi6;%ABaAE`4A%9r6yF%tqeF_+=MB@>sx
zmH{Z2=KC_40o=EM{4$#c1p1RRewX1*F%tndmy!Q26SqbKGy4h($P~id)Bi7es!Nv<
zR~HikI5(G395ED^_8BuMf5jThZX~(w-d|BaV2wz<fk7kfwtIlwCYW20!(xz`9L7O*
zbI9+9k|>p=dbHbiCkD!0rK0#IMM@>7pOc)v93H<NK3$ZaobU!5PTzh^KrbZy2T6n0
zOb$R;1=F{e>93!qL`&BrU0O<|=yG)L>DL?|6nKIFLl=zz93PwpTb`pLjLq;0>5>H=
z2_DZh0*sZP6IM%5Rs|FBp6GJ%I{xk3pAV8nD8H3&R02$xtkZ(1G2f5Ubh9e7Lg7g8
z9$KpNk>1daAy$niCX9<u#DXRk4Y6um8Pz)8mv;{{C4axrjKzDK@uz27aU^QYPsN<M
zJ7P4Cp5+6UxRH&XLjFy-FMMzK-nYCOC{HTFp7?TE6A4%0@lq&A(nR7SFb-2``y<$h
zxo|I8_DhygMjY1m7O;$kUV}^~UzpmWkRFSvBV^K8(MI7=Gne4lHpv<n)`w?KlM5$~
z;+*HBGJi#*X&#=U|IFAEj^rg3Gq#vgsE~|!;Q%1n<v9gYCkqFsRZ2MTtJ@Y^6Mb~5
z(Q4xG!qiTUm!2Y|ZoDsP&@lwEPuPmDu}QUsF8Cr;jGD+prc|*<AHe!lJ#WBb3-8!^
zt}pY};;Z%e%wxpZ>O0x%JjaT5RT_>@#;(gaK7X+VE|p`|aQJ+L5*x|CtG34o=h3Ac
zmc7Cj%|t(}@T~P&s>H#$W}{V4KoDN3;t4h41Jm_5L`Rb7JV#U1PE6SvA|z_2?N!lC
z!q8bxHr6P4$mF<0ABsLI^&)W|NqY9|@t9LRb2-pv&&uz7*8O@;s8CHVVcHzk9sGHY
z_J3rBGxHu~@eR(#jH9(<;;q=P__66_X!7eChRT;NS*Xgq<R9DF>CLJa$RNC{`CvV3
z<#Em^pDxBvUYOvlT3h~&N1<Hxt7x${_A{59@w6Xg%p+|yG0eGHS+!qVSu3uXHIxo*
zEEg$E8`mt&elcUL62ZB(d>QjEs7IVNlz;e`vwyX7&H86vXQAUYzia%o*VPfRTOrKA
zLz5NQ14~A3*O7UC(VBLx3?-^-rj$)x09|bXewB;PY8mEA4rF;woh+kWS9I0emGx%&
zoV~IB8EpmOuihDsdHp%JI$76pv4}1a*sXa>Ev;c0bUo|VHyJ0dve`R5ta&GDDt}oi
z#s+Dits<E_9cCM(F=QQ`YfjjT0VTDtXcttTa$zijjkR(7`7%5`e>?oXP`&nAptIBW
zmqUD^6$nLDm8Jv)YWm~v(~rZ~5o_CZBvnXjjp+gvByp}MLsh)SP?xhO4uz7oH8ZVg
zTK)IO&uJL>OJC_j6u-xe(03C;#($|M{*#Qc8zGFJE<vvHt|fI!W~8v;429yRCae`;
zalvo5WU>Q1l>)C1lA=oq!T{Vump$e|mkYfN(AOooi6`pdtb>sXZ1Ci~2)N+s4dEsQ
zMXnUU{}PYgNXsKzgwP6O>Y^I6GhptKYZF+OY2=h-n^eW{DM2i)8P|vS`hS`*1wC+Y
zmIQOT@=48h%)1fBW|W2t#gMP~I!=Rl@pGJ(rn@|r3U<@n2S%w9%2aa}@Q6k+m}OMF
z`o(EO1HYbGC{r!+6gWW%Hs-0{Cj$I3C1^sRf;ZEDrZh~3qQ7W5tI3JzrLo>dYaTxR
z>*tR@eg6#vP5XBQrDP}2Kv(e73qLBdDjGNa;%6&Jt$}zDfn1yH?FEovu1>52-=~8{
zN7RjPMt{Z8ASlHmHj`6|jqP3~&<qtGBF1@?>HTU4D_507V!E;spCT5X8y{(YnH3-7
z)tBrOGirbL-(Ps%Pnz|m`OSD9;;xrAc6!i;4C7tIm6lnm?$_#J1~1TUJ$z0zZeB<-
zm9^*^5D>YqQ^QV{>tPn`>#LOkl#$75*Of8m^hOEM($NktyE~jk28JG+f?D3YrueO!
zce&7$JbJEWrRj>yq);joVc2ByZAt7tua%ucyElJOOlk44sVaQe*6o>WR?eEqX7y*2
z{d4NUIITW3dG!^$$SMTUY$897>ZsjJuk_QfcRBZ6^OJ4%sa(4o#R_?2eTxo+^I@!S
zA?k^8+ItIJ<*cy5+@UBNFkdOk253_j`L-T7^S;jGhZk)qV@h7|yiv3vVAS)dx|?dY
zxSfA`yK0NktfAV@RyK!|h?_7PTGhh>qg*!@Q1n<eLTleNpR&s;Swb61`$DB|;-?$g
z>t&3J_0dRCg?^Kmw=`bW&v3;(U^LI!k;;b0+$BkA<?Rl|E;6X~X)x3+B|96+&{8(<
z9Lbhd%ex7F-mh3xt8btfOFHl!#gZM&$54NujN$D_AZ>Y^a`OBLA{Z-0XsH<6q;<2Z
zurczG){o%T9n$51`AWJRps%IN#3wcjQddQcDK+&(ACXn5KL}wV+E!Mj5)K!V-JDe~
zH+RD-+7@u6e8ti0Y!#337M>9g%ifXU(TCVqRF&y>XZR80euf{Q{|6a<p!fk9e!zct
zpA0{$_5B$hiG69!mK#>3U@5fuWcz$2FRvAd*UDw;Y}=%%*4hDOcTadKMBAdIk@0Mh
zNA><5HK|=oJ#sdm$nNevBQ@iQ%h~f*8lQ@cj@R+MQ|Y(o*$iDzV2%n}+`gG@+TD9T
z|8y=jfk&|U(5Wi2IlinaqdT*6pv!-7Tc3qwyyb`?J&gr+`p6-}om$$Z=PL)=Jem(y
z){aV|34(44{oNz@Jy!rlxr&AcjDUS0jL)wNpet@f3mdhtqb%cxHFHPZf2OfzCD{3R
z2ht}@uRJY9Nl~J^kEPbn+sg@Oav{CToZeyXHD{jSwUD#2x--ud8QGPZ^GbhL-;C#;
zIg(0`o?*&ao`T&k3?c;9pn=_>fjx8JXAV@W`0KH{O{~W##g&f?R)sh<Ym;k)h7KMo
z&TeMsdsU^4Qf~TitHiu%4;(vghbKROTE#B0Z~;fP(tIFQPhZPdWzD(uxjPkZ&*}5~
zaqWQW#r*x!Vqa$ETa$9i`__NG8Rsa<%(yk<F&01X-IxI#gm!aoS)9F;%FE?S0_r(o
zH?q=c446^z!akhrkY(RobK1DPx{=aW&B-T>12nOfqDoP~TM3`rR)p(*zxe^q&DEUJ
z0@Z>C>DOBojlyXdq4jCr6l9KV^T|a{Z}g^qn#K{%pX}VqvyxUAv($g`6CbDHdcP7Y
zeB9xkdGmD@yoo2Rh1W}-)>}SBP#_mtpF{R``s8^j(+u=CbP+ud!h(M3a!j)WUUK{6
z%{-BJ3NWwbiHcjAzsl2GwcO4Fg3$O{yJ$4R`aS<%JKqOI6$#9HbkA?-esc>nV2Tx5
zzBa#|PI;}LRcg8R6UKiWm1Q{?S6S_D=Qsdq%HPN_MjIPga4gsLiY~?Z9ae%K^F3FB
zp3=3IpojkdtOPv|*H(faI%&_<T)&Lng^hu7`F!^Gkv8rz-_yoDrE50sp;v7DzKmC6
ztf#YO$Gn>}nxPzWK9{AtbR*-ovEE3Dk`k#3ACVG!Os`Fe{j7hw2LrkBXGO0jO8nSF
zSx(q)lLO<jWf#(KB+8&E7vCX-I=pubppMR!@abUh7Car5{m|)^FF9!DjkUqqv{^Nw
zSHdCO!G=BJI~uVkbj6B2@ZHSVli9ap51WXqx{z&dqutJ$F~Z2v;G0K7K{yje-j<to
z>rS@bgO9(|;r|2|)L~wiv4cVr0XUbzUqKWDGB`Gq;ix8;9vU;6e|&;{`1{*mpCpm+
z`t~%U?35NEOy7Pz{W;;hFH@fUB!!N_PXC+IFs{M5&Uw6R0}!X*IOWrSetiGSw|{`Z
z1^-Pe<Ya`h5X1DVJX(v(qSDh(dJaN?w_ySi5nVGmDL?^E;DvSI+w^3V7octEMt`Q!
zC`>R)b&E9!9Xw8~e;af|j6;H0M|r=W{UBr2w#YP9G}2RoLTl3_uIE+hF<t#LPsLUL
znx4K-PhTrPqsBkHJ1^ZDJ3$!ms7f76c}RN^eKd~Q5xul_W3(r<4l3b-Gsrj<gqMK7
z>TZ=-9_a|?A%w*cc!=-g_yz+8S_?~xT3TTvu5-e>q?ABhe=VGCzrX#WWj}Dym+oG`
zJfmI+W8L(O{1qq-1eNU<bz{wbX*NwJ^VP?LPR3-=&@397MWZukx*xM)@xE&YHiwA_
ze4RS{3Z2iD(62Z127QdR<_6|4QY%c+Rv6v5vf<U30^`v_V(--~dL`tDw7RgwTxP66
zvSeahnu(stf2O<VRq1W=be7<xwKn2C&9wa1*bX8IAn%mn)mdkLk}N2w@9@cZf^7(r
zD-evYSSX?*&>p7Xnjx#QDtYH%RifP&icF~MGhK{9UU}-@n~__W{1$wgeWHE*NSh?g
zOx%RC<a$xaL3Si*l3nG6?I@&Ob-}4h<?swIy=2Nee|f1hJ6b6^eJ;=tJ&boeI>TJj
ze$Xhl=Q{Jun53sV&yY4;`n{O%wDLr9O}bQfww1y<k2#%S9t-?D!!d0C(rJq?>p-8<
zoYN{ry(|{GGpCHMz>?0IwH2QS0dDw@<_wAh`S&!RMp6E`h4~xO7$>cW##e(k<Ok!S
zvc4Ltf7ZP!usLdTGn+`wfLF1z$UEqwaK`0bj$~M69*gc+_d5~n*v~Vj{))}vm_|bq
zz^6-qHI8WHuLn1B1QaPCQI?*~XM&tAMbH_eneo&7qz3L#eUKwW)<t8$U`D1#L@MJm
zOBaj>nqiussQauFjC{$ye%OaSjj{V!hLnVaf5qx%PkVb?=>*v2>T~D2FJwPie7m@?
zGTW#{jH*Io#j^*M{c0e7aeNx$4Eefa=c!M<4{9Eey(bHteO2)uEc)i(^1OX%V2Upk
zrcVP6DcjF=f+}@_nW^6{SCkI|V8IpmT>OxTF1ey~`!YPCZ~AjlJy@d!1eN?sdQoy|
ze>LfVJDchEKd0|cUq`5YIhDzl&S6s8iw<O7Z#{y;^oT`A;iW30A*4-8d-&h)AJZ`M
zM@xY~D=Zq^2wBbDaGj0pNB=v{&MFdGg3(k8rG6qjRcCU#^LeK<hw%2P008=z@t%!z
zcw?QDxHljoZKuB!#iU&&MWIO0Iu{TwfA^z;DQn!orFw~NJRFv2E-Y4`({;Z%Sy8YS
zG!Uvbo7vv?L9_y7TtJlT3Ib7J$vw{IOeEX8q<i+oM4^i*PfxomTkx5UAU>ZiQLdVX
z)u}Qcchqsslg)76G&b}I5jPzm+SlS+GPFJyM^EVF-I8Jj9@zSM5Y+8zSOsOUe^|G1
zOASZ!0HC7aysNJ3E-ROh0TVBeGG7ZdGKw0ul5?IV&m36%yj0RAP?l1D<q*;_yQaNm
zUWaonS8PtAZwZDbqS?t-;rau(_)^j^Wg_Pytr`ROGN2yn%ohFjVi-6tEQ~vE%V<A*
zNhR!(j%}`#{82g6<&Vlr${$T=f8`*JQdyyFP7Yq@c_z&{&XTU({83?Dj``z&?>hoG
zP8UAL1qUec;0!dS83kKTTT&5gWTGm?D0MD5ccC*Ifa2V{uH<CKwSuK_v%sD4<XkCW
z{u#!GxhlD|4_8t_Eu2#yq%~GM;q3w%1;2%jcy#KfdX%hFq0mZ69&3c}e^a?k=aD#e
zZr+B%+F}=%dStFyVk=@>YC0=O7o%BMk6Z@VG_~%(QwF=%Ea^s0jjKuXVIwO7OaPF<
zSo&cu)lx{ay#li?zg(dfd!mMDHzd|&Bujy-uCA3t^Ql#`wK_E>)FoZ}-UXAS82yN+
zJnpYDfMtstS7i$<yLZJye>x7xIjrTkxrHA}J3qCTQpv06NPyPIQh!rW$5C7Kpg8A^
z8ndQ$xSdYjYHeL`In|J=#Mad}U5wSv3ABcV+^&1CQ3DV^)OHvlRWDk3mEs^taXk+1
z7Ow`&PZ<{Y?nH-lEMrNx3^(^&=hrtp3ZxQsq0yrkrr5F@9@#c9e^Owme5a=amySCY
z(5_js1ffi`WP9C`=PObkVJ*8FeLW{gS*%=u*;lNJ)F$yU3h-uxI`8JKcP_1-cLYkW
zZe{*%weDpJGXry8>EzFp^E)H=-AQyJge4WP)PxJ%G3`S@Ru#9?A2l4rsO6RqALUki
znmcf_<*_xAxQn;af0m_o`q)Cx(eU0T11X!<nBsd=n+!5}rIb^Nkg9}7n+)0t@6u^n
z5Kj4zY%*BfPU{i!R+9lu)}oxp**7@ZdvHSAu-=cmY??Xo26K1kLhnfPX|pYkGoL``
zpmuo1Y%Lv-%!di>OV<ELBXm-=G_Y&<DoSVAMeF`#29s9>e>v|wb{eYot39K}?n_dA
zX<x1{tD%Vs#xDFUxMVTCl<_*3*4d#QWVfT}I?s<s%6MgPZCuBNPCR?1Y<Jj@U*BN4
z$XNYyHzQX$VeoG2_Yb3v1fxQKp2~+Yj9;e+U7NNFK6cFU9eH<v0wwkJ=(<(?Z-gCo
zQ=rkrS}@zpf9B`;xL?i3x#pTX^qmUfe7%{BLUKsafajx?xC_DuJ3?#6gX1AuS7k0a
z5;}@w=YTJ><H6W+;O>af)m(S83in26uWzS(Q{gucJ~Y<(Sd|Yukb50#KNYM{KFm5r
z&;q!MX3&x{vgk&Xd407L`DCm=grHHbk+xP1N5i9+e`xD+B_Hidv&gSCi=w8U!^)ma
zd8iPF11+LFG_%jOLaZGK*q3|^0=YeM{9fr|HXRj=P$YMkVwmahyX!7vTuvZHJ<-hJ
zLkR_BE$k6g&ned*GIN+%4fa;1Lz{84Cg^%7tGiPDT#kc_J6l(Bk2${5Eer44CrRMB
z@pP=Cf5RckqbE`jqHx-cCsH0n`W|UFS)>NMm8BJ*MKv4wgA&S();t|XVBE)apdHbM
zv6j89Ew3FS4kI+KSLd={llxpKj2ouX-ZUY*c`jvdhMru@Lnm3FP)=xGDt&u4Yul&1
zDLbjIh}L%?1m%RK9;nL#V(TDEVXC6ch{xq|e}Vw-?b}ajuh+91sDYQcH8?_i2X0t0
zg)T_jdxPJ~0m(hJ2M8C!h=r1in&_@`NNI$zutwX*eV&nhAb+d|L$v`{s*f%4k}X2C
zB7$zw<dNKyy`G)fwI065p=@dB(YM@l_A!L%tAAfz_xwYK*@sQA>@@dTTKx>rF5?#+
ze~a=?H+-DYnjgz<TL~&xI&K;)=4x)_*~hq_-uGPjpvUD--GIgNLzDG})#4MM(0)JE
zYhgu#C_)Tl_)LBz+IN4@0XltNtozQ`H6KacvIQjNa<A|4RQ@gx=5V(ptog1=K7M?S
zC5iJeLi`x-b(q<X`VjB+m|u#RH??*^ELXcU?~!VfwYoUGth8i(U-I)hwP@LK9lCz{
zwe*AYihKE8Li;@P|CIzBsF&eQF%tqeHJ9-gI~11@(JCo_yRIWS4$bdZ_y<~}2dM{e
zVZhhj9$>Q_+-$MYVrHTVcK$zD>P<;iWxKl_Y*2QUN<~o=ABU9W>Cclq{rdLt`S#1r
zz{v@Z%IVYRA16gGB>gH$jTTQ%DPfH|eg1X&*SnP3()B?XNNEgRj_y7E<N1qGo_Z*x
zi&~UAe{rCH$~PwVvAMs|bjjLZQ-6J>7FDR0=S0;~FKYrux)-|WYZcD_{`|+AEK`so
z*mxWN=BIC;Z~uLpCuEHjTI$n({(6fqz$jt7K0y#(nbY6DpZ<9J=S-E0%PSI>v=(s?
zmy#qyI9VHniLbJ_;H#8K6M-9DG%0}0ya4*F{rvq~=V@yB?bpt?(L!lEQ`Tw{gN&bI
z{E>y?&_Wb{xdpk&J4jk$oJWX!qs@FHEe>3U5qBkFtx#4k{Q9B@Ix7O`ES)5oTcW(k
z5}W|zULnho@+8YeE>rLOGkJ=a`6e?y0{MuSIVUqOf0yA2^PD0h@U1$pBn{o3R*PZ%
zAWFs87I0-^p~$LeHoDe>i2%#!@gJ#8-deA(eO&f!rCubT#`kqDjGHLPV#$lBj3(Cz
zdKQ(w<)uiHfILPF1&C}$zCR~P*m572TQ#{QEs)J@_SOuv5wvb2Bg%Yerdp~{K!5TA
zgR^F9f7Q_$0T$Y-M~!0aCYG;7e0&ih2{kT&YENZqSo>lTaqo-$ZCz%_%XBo`#w#Lu
z*JgmGT7YCrdS%ZsRys^$$2<F8l%hzf&$GRih0kkiNy~WF%{l{<hac(3yEK1aDB(gs
zjvVUKo;s(t#=mWHj-$2_Jy~a@_;EJsoq2>Re+MYz-@2M56Y*i8*W_Y4oxDQlY#i5V
z?ek2z0XJDC%IQtxJrv1d=bbr@k4ZK|*sJRfr0wLBYZx9}M{`Bljecn>#kHUtxo5hp
zyQ7HM3^crFq?L4fbW_(#Dd}=CbSy^Poey3Hmu8ubj1IB&(c3|M&lZgI!c!2MWFjxe
ze_s)1C3{&UX^-8Va>8mee|6|1Ja=ag`7+~`KnnL8tK5T4wZSxI4WLzrx*iO&Uj0#}
zG4$W{NzI2Y`lwMD9k%6BY-Z1Idc*n5Pe*+A{`<a@dO?wquJmdEy-u==I_!0^>TCh+
z`zrA?=BC82Os5^^aInctt;jI(?Ym6Tf6p`2vqwtbH+-YI<<|Q=3*a7ghl^-kF+A&A
z>thDkqN;zT)ft7}h`}J_OG62iHKC_OiSR33-eAFzODo5cRfUuh#<j2;-FZ@*k1p)u
zG~1FU(KNM9L8B?&HZrzeSmoxic9a?d5=v;0rg}l=N?fd1Gl{}&dBGg7{H$BKe?^r^
z)QRNQvc74$*EZUuc~uB$tzFlq9C@CAZ`y;PWM!IaGf!I>@h!X`yHrQ+Y9=OccL1ao
zHn^FC1_X{rr2NCXu#Wh}gvo{ud9ak=P-ZakfK?L5JUB-P*vuch)zdV4*ITG2uA2O4
zIAdAIst<lLjvR+`PkGI8+sbIJf986oWcHmao|{8(R=SGPL(d3kuuYrrCgr<KQ&JA3
zx^-5cp}k^DQ0VO~fkFy-1Y~FWHIKNk(7XeQT~s8PDHN^_$M#wTSiiK)+u>d9*91dc
zROD#}PboryT}st-6RUIkb~ZQ)wk{(l94c~(z4SX<7+1g5EJo<;)-R)5e~*2}%qeQ9
zZjqTI)lBb^Nk;80z2`Uv-R0zw=J>bI=V~GB>#mHOs_c~LsT0ps>$G7X@BkX&$z{~w
z5c2XgSf9;Px8Po(?NAr$u#B1WMolDxX}5yu!tWq<C?Vmx2Nd^%2al+0HR56T(mAcm
zqNZfvd$9rmjxFf}OM8>tf3AEDD@R%OIUOqU-@*t0G)4&X7<+m^5r`)Og_E>mb}+yQ
zb<yxO%uh26K*ND&duo-MTW#G)ljsesU-JYjw{X2oEiF;zMl0LfW}M7mZ$0Heo<6GG
zubOr?J}Xqk1*&EWyY;W^CsD`L-CWM-s_vnlBV402$&;tBxleIYf2EJCj(4eUFFK$>
zDD0l3rCjcPK%_Yl-%5vy)o`cFZ5-X5_$7t(GaFXrDrw9;*Rpu2px(*$a^>}`9!PP=
zs@-F0F4>NBi~L%dZfCYt>e6xy{{sG$vjRcQBef7@_kCq8@l-3x9$>5Ldz93vuf<b7
zxqW4<70xZ&SDM}ie>~p4ivG9m2%)SVnYH}(&=yy8)Kd110UI%?zPn7ZmzpYM`MXCp
zRN54t*jTCjIwNS*rh>SgJ}MMShPihQx#_~P3mL0Ewp_gE9rC+mf$=KO@f%8Z;^gxN
zX30O0-G@TX6l8*>FO)XS+LTDum>!<qYtkMCUGpThoP_-Ne{PWuhLG_h<QMK1{gmAw
zW8(H%+tL^S&4?^_U6mm#Fm9%#@Mr2wRD!HGKUZgVy|yMJ@bjE%DAiU^4l%b34$<{p
z0QQ!_(g;2FC=E#ycd5V}`u2K&2_m?K_ha8uBIun+3A$!M=Mv=~*}jRg1ZPL&aJO8c
zVzJfe`Rku&1W#x6zL#MwGa~`hmxV1eCJl#qZy`{lj6kh;<pQslxGgg}ULU~O!&cQg
z*9*%gdlqgxo=;Zuoj6TXe9+`eUvtbzyIAB_FE|=AQGUTOqiKX6JZ3cW`}{E@jY<HT
zSIBateCL?a-1{}hjHXfU95b3_^KhhW?(BPx9%+&Y1o(fKP%bkW81i#Is}I|zwN@*4
zt3vrLj^L{lc~uHy50`;1Gb1S@m`!02(l^D^LnWX`xw-|&^Luc-;%lF*_&Qb*Yuk@%
ztji8~B9HX3m(4CSGk<@Qo*pf_F_aV!5oFz%=4ZP`xa;SAcUa&?SL(W9<~XgiU=S9N
zvjCijR9KNS^2jzXoPO^o)Y@)hQ^I~k@R&a3S3pe@{W~Xd6vWfx^GD34Z1;@Q$;#(b
zauS{}N-ZK_I=T@=_Y)r!KDe1$KN{uk<s%hE{y??Qf-o9p5K3u$3X?wyB7+vF{0nf;
zNcaU1avb0X5#+?}v9`=Ibi~jLB*@9$fsI4gW9Fa3i`1e<PIkd!hL(dYz6U987Okg$
zGL!!SrYlOqm*Gt@69G4uVVEWpm+&t$#(&Gj`~I)j-=AbrL5g7G-*_9JUS6O6dm0te
zJ_wL@^Y^!>_`+xq+N+HUG)RWc_kT8DpZ*xhR%T998<d~#u|wDLrP$O1*;HDK*x3{$
z8QN^D4MN9P*`_jDOl2D(*~r8@=KrxBRcVix$?sx2k(1^#JyU;kqU#vn@1x1%Yk%&F
zuDlAlEs}|COg^$phNWqR_#Shg`s@-(Q|2{w$8sG@cT>8CGWS^e^WBwxuDpNFQZiRp
z$myvoOOo0`G*UP1Gm{!8jCO3&1+%2l!qBk#ljg5$`O;9<N`SH>Wd+M}7)i?n$KKKY
z(O}8$u3X35E)RKWu)U4SawnNjSAWZ^q!9*TlmM3HRniM5XWE_fnIvPY!|{5%@{DOK
zjPq4aKP>BOC?c)sFrz3}JcbsZ)njkN-s01G%y)=d7Y%J7#xinOnC0nONS;rpAQzWC
zYN3N4IqFcj@_LNRy!Y}|AaPy<VmeU7mkhFr8kSj#AVJV+<@Fu~D#G7r)_>2#M4gAk
z*+eN62%*J4(}+7P^>=XqBhYpX*cLX=b3zA~XW$+Tr50AIjw53a-h~4OC9KgShv+qN
zej}RSgNvtd?@h}kg`UHN4GR_<*EV7h&e{i{@j&?u8V|kCp>dK^O}>J<Xzj8Edc@9A
z@kY=#N>GwC3c3c`#)=STMt`^(v{}x}-9t+&5Z%c6%*w5!6{TK!Hu!i`{6OP{ms6ZG
zH^+~W;@k0K==}rY$H4Ip@nfL)nD{Z8_4fFIn*7Q5&I_AF>Kr%`m7pJiO&#HL(A0JQ
zz{u3ec@CI*)BZqQ>KLDiz9TI4V{b$y%EE(#36)|6a|1FJtQn}TAb*ph90dH>`=Q9>
zgvFVBw;+?(0$dm=-W{2oP=GVL-VRMB7(rF_nkTmPUM$~%@1|6`z?&y|Y2cB^*Bbl|
zNuD7iOElLC*$Z?(X`B)=;HA<$N(u|~#$F;tazs_q?=bDxEKg|O>F|#g9_fv9_Y!`R
zGJD+}1=WETp1)Plxqlgxp!X0^dBVJ~CZOi6xZ|~JISKfPsV6AYw}|^Fm!!z74ZKIP
z2T7y@rk&olw2f)hOAaHZ_*)7qp>Td);Jmak{ON@LtzC4DuJwDxykwDqwe|b0Ha_J}
z^`FfNDz?E?*0*gi<a{fqFXwjlwWhpWReQw!cj6B&`spF_qJNBROD2mg&qX>XcZFNo
z3ntg*)+Akgf?Z?JB^^`$e%|+6(z0vies$j)Y6~)*#3|*FpbuuURzi&h$>c?nrIh>F
zQm7jslT@buE88#)zld9*zAxKMiC0O!Y8!7!9m}&VDQL31DY~ZlpJWL&rDA?^%B<7r
zicKCcn{j37K7W;av-1W%li{)=zuBu~<!GJc9mz7bJ{cYrN3dO!XJ9)z`Vfz4TUmRy
z4kuyPp6gKj6Gpl5TnDIDG~T>@u5-bejiyvl;e;NC>EwfLNM*y;O4u!**eKngUelhA
zVPrfJ%V%L)M=EO9R&S+N@_A*Zk?gz0Z!=%%>sXwM%71A0(#qu{S74^rhumXYeVqpt
zE2yA_nu;a;!sAjzfxM2hER5ca713jI-|8dBvNGL-BumtiTyj|+uY}g-7oQcjl+COB
zvYBw?->zoO>70<GL^&U<+rxI6>DaDpQxu;1GsVnx&j6l|8F3}AQGNKChOY~4sS*SG
zvxR=+D1Ts3BA=!5ltlMR$IziVd1%82!}qf8CJ2M>XY1bNRO9mykCW~f*QZH}MZGD_
zFv0&Kdh0yrmzQ_f+e0~lI(%Sz^tpN6c({JgV%-K{1%5P_ST6>qD)^b<02mIi;`J3!
zAW>KaA4>mDsnCV$J{i1TT?j^4tADb(OUIZ!+JDc?5PKZoqp;V)N%MgfZqE`1rQHPz
zzpn@egDvtUm}&5p_jPk9<^q?2h<AI3GP&_##75UzPF++A17R{uXY1PS!G>L_8FR%_
zYo$UkRSR>18iiCaBdD7ro<`-a0xTT(tFCfpvnINV9~_YR2Trxp%~Dqr@3)Zau2O#u
zrhm?@cB;a1RYo<nDwQ}~Xl||`dB1?3G8N9MQymMuv%$cU{=Za7D&$mpFlE>SC$6<K
zO+Qt_=5S7KJk}n`#onA-<54)O00)bo`Gjk2{t&$IW=8G3)vv`=4GhIErJO!)kKI`N
z0@c2#`~!|m2g+y8Ovm2$`*;pqE=&dvD}TqGY!@nq(N;WDr}E({<^`WVW{sc6k{bpp
zB>S3fOHiD?j!mHEn-kH-SKHWxEBoO_jmN=>w%%Q~;kqz<DWp-IZQ;D8y$Vx73t`u5
zvm`Yc33k&+!{4y-vGH*Vi3Xb@+bXrY)^3YRQb}zQTF`Fi5Tw_(-OdX$x7!A*M<2tt
z`<Wn>HoN!DaSi3Xdw8nRXys$&;ZC^mKdwzU=9l43F%tqeHkaYRB@+QPml5~`CzlsE
zGY)^G9zYP($!-GlmPKww4@J<VhfUGl9{T$w^%_|jJ9ZtQ0mimC6yKCc#;6x%)cxk>
zar5cUp$a+@1gjpO74QqgpI{j1Nd*Ay9jM1w^~;qp*7I}24;tefKY@Qo{-*1#gUA>_
z{4fOIdK(bx9w#`y;5*BYY511$?JYx~QmudMOw}`%cZr1Vjvw9)H`l*D{=6|&1*4Of
zpYk=YzC3QeZ)SzOGuj%fo?bWd0v%`<t-_=uIQ8Rq^}KnQsgCA~z~(aE6YZG`4BHTu
z_enc>Rm>$O%1VV|He}WDjvrQ#*c=qtNYQ#m#Z<v;2MiUJw-O0jd6#h*hQ7jV#g%`3
z;sFn0@WuhD7+=0+OGdw`WE&g7jn`XmteLjrur&)?g(s0B7RiJpxk^&V74N(;3q|eN
zmenj-5Ig95JLqGtw-K#`2JR}O_G;2)V!ZX<+IQ4HMOF+NUFfo&=>u;dt;v46P+N@P
zCp{!*#0Roowy(?Yyv7i_{8R-Q<fVUI^&S@HS%7%5{OiK3>*5TUtJAk12+y4Ht2z1J
z`Ned#1MCSIDi2W&+8IQ<k^~My^?2qT`uo1CK_m)VKIELJR_zhJ8~2jU8q~w%jjYz9
zCWy1Fw$jbE!NNxl6sjJ@F%M(q>lbfxKs4@hDlit4sDi8;%~_{Id?N^aUUPrI`$sL=
zpecs=sP#q%=6mT;OCe~VQupkam1z3BcnHmEwvhKabWQR7<HBfU3jO29i<%%`?-wR~
zs{qbSteI}G;CLt0k4$h6G7==#ncPHBTni#?OY^$Ml>idBtwGXFgKbX^l+g%+o^N!H
zC15mg^>?f+Xo-24j%Y)BdcJ?LJP`+5uPn|ywa;>c2c1lqIbPz2z&yau6hF=p`E-Y?
zGaNCP4-VcpnPlGP*GawSGejM&UGv!)m%`cp3#fg7nfoRFi%vsGGsTbfpV*)$)FuAg
zH|>5{T;NOR_>^{QzPzCFJIk+8Dy=t?T;kV6I-#2>evDsB^4K#E$fJLGx#vrFz|v><
zc>OZv%Xz+<k&na(Bv<oy#fbMRV>1DlBB_5=25Yc{iOF$a5xkIf{3kBheERF<`KPDf
zK=Zc01(ol2Z3$9TuXSlPCE(=xR@cbe6L}?vvGd3>=!lLH+QtreQXAyj0UX5v`c0ro
zBX<XO*qo~hsaiUN0n~qF5Q+=5^7}RdyHuS;QKE?<E1^>N%98iDU9yyCTsKtS^@rL#
ztIb2TDtQ|OC7MDwdH!3oD)1!}xcMF|+d$T98(Gd~l@vHu4nRkJ7gX(|=TzZVO9zfC
zr#iQ;bAt?eHxScm6dI_DHo*_rVbYZ$OO+#DQ=G#XbE6K`y2XFO%kQVxy6=T|-4EY|
z&K9_yMGlB9^0{DBu0(TXNpZDjsOSy5oky*vPrZ_FoYpdIBEl>m)F4vPZM!J<aI!%F
zxNvqS7-T;32SWs>#k0*22b!m*L&a_-Hq=HZjIB^_j<C!#Y7^y`_OiG4I7~l24_C)+
zrMom_W_eq<)h2&j%F4Tr<==0uJZXvTNs|w{><;!(7~gQLl;s)``|x)eR<tY0H5-)X
zlYFz}2@NEg$mb3m({e1y&cA4mYKw_a5C}7OfW50}Ek%}KTuYH9+)Il5srJ$8Rkj`O
zvSZ2JZa&8)f>|bXcHah68yajej`3<~uY0u8-rJ)iqN9Jd>rw%?%4TV1!!>vr!?5GG
zHV^e$?ILU~*l8+JAhOCqWNQ6FCD5R$uxr#>&(^0xEj8~t29_YKyjt}{otGc#snGcL
zw?kjp&2!tS?e?S|<}C*`J7M0k?0qL)TuhpSdp5eSJ*<ShOb#dxDTK9rduJM$=Psp<
zJ|0=~h~|H!`3xaxW8y6Mf1S@bnV?Xd=QDi$$3(~HGluVUhLI^BxhUU*d_8G~;}Z8@
zX{pnF9z>szmfpY+6=iN|uZ~TCj`l*1`@!UvF;Hy(YbXa{FZSCy)Pg=%T&hpIQn2Es
ze0W0rw+PbkT}cS3+$-z`mZ6e!3tkf=!_wKUmW?1>jc{Q<DSAPxWs}n(y)oVYGy*xP
z51PL@DgFgYG-mae;Y~3U0XLW7tS=LnVLmf0e`7ZizWZ0aKVUYO;syx9SW47DZk_Zk
z$RX&N?m4svlguH1KgFfUA|IdHjwdi=S$t$IUlnVU<@RNh+xrJu-!C5@-ra22lo8f>
z@3xPhH-%3m{#Me9;Oyq47DnQu=k4c{l-fxN7kq(~*5J$I-vQt4`O!eY9m?V>o`3oH
zmkmEN8-G<~1L;X-*%?FuTFa&I5Rtwj@<lRyDSW~C5#QupuCV7yPm=m7t+o{c!qxrh
ztfHqGDqEi9b9Gl*FFe!9u2T>;dS{k8c_U)5-^L7%Ff*_rDXY9*(zjOF5F3484SvlM
zKI-UBRm)KH$S#nc>Y{PkG2r34_F$W~xYM8N`F~51yq<N18hD1-Q`^7LxJqTK#y8Hx
zaURKTo+vdZu2!&DzTay4q`xb4aho)zo#s+@*_mhD<87Zt$<JDy7^VqJSzcL_M$x8S
zvF@-$Z7*S_D!GJJ$Kqu>tAi2Fw3|5iw`!=Di>^pweX4Ra2}gUClO$WVO0pTGCt-(R
zx__AGN6^5%E;RmCm*#De5KW))GLtN-*u9+MNL8MV`^t~%p4GiCPx$@4)m<e0cNtj<
zt_@*rgtBo141_9FfG^UnHybTH-pgTg_Xt(>4pr(M*8O%BU#pr@^nvllg6sFB>bJuy
z`Du$>JNt&&ccJ0|RhApK;pkPNS~c&jRDZ1>Kkw=6{pK#}COYAyeIe}Q?ol&et?ecZ
zjG1DXF+jS?@qT2B7+XCRBJFLsN}^!NKTW^?<uo)uFPdPimL_ac80&(=cNeAJen<V*
zPk!uL>tKb$OfqQ=6sf;IXF0NO%Nofq9gXR>uW&UORY1XtLyEN(snGA@VDBsAIe#Dy
zBqgwX>l$vVqrW)Eeb7wNv_%?<EIW_S*5>z7$uXXiCj}6$J5uCk<ms{n<U<E`R!Hw&
z>cGS95~4IEo2djz>I6GGZ(eEO5+vjjROtnngGFf2r#+4K5W4(W2IOnuiAD`WjemBK
zfd)AoUYsJ;90CKZwFxz4UAj~441ZhBz9dj)n)0&JKXLNJi2?vC@#GB?oK#iz5*oOc
z#W@EngtdEF^bXxB+ZJeAz-l44;jhevrP>xy8j>wg*@H045&}@lyzvx6*RiJ)9zgRQ
zxHA7J@?C2nEc-bsLqaOt$eY{#eVOozq>i9nm&K6L7#Lmed9KweMiK|jgMYQ;)^k0e
zC2OzJgIpABjP@*jp}#6g$`f7?u>uXAn9u4+SXPS^4d@Az@2yG2Xb=Wlx*5Y&5%_wY
zmq{i#5oDM-=3>%itbuuJ=Sj)@a%{u4fA;yZ-wx8UDAVrzl@$!WHyIUm#Y)<oRM*~!
z?YUR#xHN}~ZI5g9uGd7Z)PIsL_7d1ghFzoAZp_X{Uwy-R(OnBD-KPdDO31ENMrU7E
z9=1fh2{>#3W}jF7|LpUrWd=VZnuiv;#;>Ottu>DRAW|Jv31Bb}pfD5XYVq$s?v1Xz
zkTGtbo*#bOklgz%(?s=68Y=v;<z>t*7Gly=DW5cs7{^Qk<Kf+ZzkhuG$ERPFz+?YQ
zTjd5qTNT6hTs}G{G~l`IYyBM3Ji($2B1+?6^HM-G%C&=S;nVg2+6!fDV5?u#Y7`Ja
zYj)TuVS>kvjR9LC84|^|O8ma^MUtws$T~GK(x*hFo=qR|Je^7((~Ezmt$6C++r#JW
z;X_5$I_=YzAPeE>Qh&T!psKKH_C|4J+fqbhfDeR9^yuG5t4!avsIlgEU{J##yj1vC
zBi(eS6{a^WAuXanOY}akAK)<QiYv`Y@ft<7*(8QeE~+ZJ1>yepr{{X^3!Zw<-V2?_
zR6K)q+auyFbW-HG4Let(uezOf4|$(zkQze^C%T9HhD}6}<$s5biBxoZ?8<$3F(66_
zYvBce%;A9nf!21`G$0OUfTV-f=&Z0z4f-%?>%&5&1oA5tBa%|Wt)^C}rJrkip>cZZ
z4FYzgoR(^P%oyae40S%$x=Sj^8uyk&3yqeHy9_Mh>t)ea_RVE-;jk=RxsG;^>TZ5n
z%NUb1F;Bh!W`DZN7d1yi0YFv;4Dd3}C0fz0qGio#Y+-tVRGyYn!(Q56u2orC`AT^b
zd)m%Vj#7EqO*4~wd|o@Dw0dR1OfR)~^X#@fp(ZjT<n(DmCk|5W<lJ-PGoWTiYQ;t>
z)F77C(Ib{NuJ@iA#i7CC<y>j^c`=&WBR-SWL#x*8RewhW@HDrbUEZk~X(i#lU8Mm6
zu(@iM<Tf{@#W$uR$@1bVub8gc;9cVc(=#09JnN<J0+`&qaKFsF)7>$)%+6{db}wD@
zz`RNqJ@l7!@t<wnR?E6aR@dE2balImSpC3pZ-#5}4D!TgGxJcasQi*IlyF%@fOVK@
zCI~~{^?%liTwy`vup&R;J!#rTvdNjl=}_CKPUYLf)Nx^2SXJaYeVK~(oH&=TIS#@H
zmkQm{RB8^>#PwB^$=1}gmg9EV`YCK!bRQp?r?U;?Jq#?$FGellG&Xbl<|{0Qf5T)y
z3@z;SO%v^Rx1d%osErdo!mMvH;o23x<?!WfLw{MGbNCuP;pje?)N&ZN7iMiD<FV0U
z-8m+;JT%aY@3toIn1_FD+;yuLc8!+u3cDshv(B;H-@obs+t9c>*(0Z4deAx)v>ZJT
z0Qji*N;$#L@muu)LeKm3-5j=D$x7+%{qzZA^KUyVI1onX@SE04)2KIxLD|La_kV4l
zAAdf~MEU<#WSpgQ*miNW6t;jE)Rh7e*mtg&d7RX2T$1jG%j4p|=B>rCgv>oc{e|~v
zTFGHE{iH|vXSp)s6CEayq>9Ru4+01~KGX}LunnF)rI$d|=noKSQ}4=7+!rW0DPS=?
zW0=+&yI`9X`1u#P{SzPlu;pRM_DA0Oynk7NF0`et|0h&u#6W@07?^^=5eb=EIAi^e
z2AU58lzkJ8rpmWzjH-|wq=1AH{mPMcSo!g;t&$Eg?%Y|y5EZK_Y`4&~lEZkEcQ1^@
zn#ubXUf93eEWU(1^es#ZCa9U(Tdvq7XT<KAeS3dEL|$r4@&*x5^;ZUFSQ*M1%wl=Q
zP7D%Nnyv!1x(DAAI`mnmX9Cxy8X?u5@o!XPWFT$rMr={zP=z#W^Yh`g3LY8#jS6t`
zk~L4%RDHt=!}04XLrwp%cUW2|hkkB069x7a?FM4grPBQWIyWJI0^EEQmyu>c69hOn
zI5n36*#{GsBS<sG0p^$LNHf`gUMl?6R%<-d2;-ZE5Eo&fp+?#?z9WZ;R~%_d3Id{9
zj1d~Q0aa90=@tq1-=DtKd0%kWd-h%M+$n8@1-INek2-7<s%*XxHqEy}S75SUv@=jr
zTC`8l*{*1vEdX^uiog3Bao~a<7Di8D?5=b6;bv)|PXVwuNQ0=_kb(`k`MQZ=f5(dv
z3l*B=V3P;(_{+K^i6Rw14`i!^Ak^{n?Z{uXw+Ry8SoBeV{?1x9J7a4z5WhWF==A%E
z-ass)dHzM2v#!r=wT$sh*c(QCvd`&5BiRqbFjH2zOee>OyH39MUEvNVBZejqS}zVW
zO&xXRlezILIvae^uf*F!#!B>Le+oSD{SdRCSG)dw+PobF?8;OW@b_f|t5M>Pg^3&D
zo>R$n0%?rd<+1EV<@d4ch<zlPHi@}eONKooZSxf;;pr=e2jvpvxk@r?9$zJoUh4cl
z@3CM`7`sOJy04iBrq{RVAMTx2_X8$XIjC?&pW*yq<UQ9HPZ$E}<ER~(e@4f(JaN?4
z2L41EOwvkobbK&yf1CE{u6?_1v}yxHBfvxVAC(rzcY<~p>9ft=F4cO-vTc*0sxq;G
z`yttPCF{*)<yi{7$J&du3gu(dUe~Fx6|0R^I+saj^)Br?j<#J#vV23=m#k5~!P~wQ
zq={_?*E!uhtmBwT5@Z*7fB(x{U1c@dRMDNb{-yfk{IsPrDO0uJXDctWp;Kx@)oRZv
zKsho_54Ga$q*WsKwC65G;uh;kl2)4yTP~wTyV_G=@&ocBgvHX{6V#43tGJkkcM4{W
zs<5$plwt#Ij_y9Nj+ib}jHS(=Bv~=28tE#27~X|3h8515=gx8(e<=Hj{1TSpNqgpi
z(3?9BN@&-t{F-d4r;eogJX;`}XdIeV6rnec<paKFIMtJ`XvF-`7}q@KnMKMlW<m0q
zM~Hn(nxn!jX_lF6+S-uOZRb0_P}sAjmQh`X?2gfvIRd@|g?{yJ$Zwi_l*>sCND$Jt
z%v9M`M$~2)E&O_he=atB7!^zq3ahedrhGh}WwC(PFwe@e^YVU-QxvPZ-)K;jb$Ufv
z0Y$#@zU4gyW+U$?aAbE4^C}Z5aE}z9${q}{twC^Gud6mMU*qC=jF~OxPZYVh7b7z+
zOR#CwG<r#QMlL%)Zp!WpF4v_&yktuWhVOwNV+6%EmO~BJe_3_OVqPOdvrh8(t#(-F
zvO^8gb_q&mn&u?(2r5rW1D2EY$edhu#6{f&?tn=x1Or%er$$YbI&~6u-h6zIsvFdn
z>S}96D%MBt%#JtE)&Jy=Zr*m!E_8Il$T(xKQF*pnTvV92g5I<7>J_XvXtCWZz*Vm>
z%`iM)n0%x5f7fGMgSs($(_S+TX7dM)qEXTh?JduJDPKpM*7u|370V=FL+fSjXOVQs
zTI4`k$#~iNWP61Mo)ui#f2(j1UriFf;!Q2qlQEikR-w=^JEwZgI-<|8OtZpLy=J=a
zWRh+B-Qa7bMP91H_L^<eH#p2Lr0<kYeGbQK2<{)Ie>$1<4M=;ovi76yg<sWvJL^o=
z4Zy4GoT(pmEDkcgo$SV<tfDf!E-ZYMEM~~CG{YJvSvDU#*ZCPr@_cN_lX&;a@d}@K
z_lNzN<KD$ha;e#kh%Py%^!0vvdU`qiu@aooK0DD*-%ja8X<h3DouspI`Sb6~=hKH7
zYxR6mf5_Y;On*R@05ObwtOpxWq9ID>0EV<iNwiKhU;Ngwskhx~CX6n(uaJNBg=yp?
zbIX`KIRmRzj-4)Ft;V)B{XU4OSD3SsB)auzYVuqZNlpm>c;Ub-7@9CW4-LzuZl_Hw
z(VxueFv2-?E<3adla`b0UV+i4!w)rMR3@3>f9gQ!mR;G^1(HGtqA$cc9Jg&qC3i$Z
zlKcVcg2_#s8SzS0ZV~P-^$|q0vj~*^?%hMd359w&A>6Klw6wy)N_Nkz8S&U1*q$!a
z26!*5RzIT4B=HfGd|K)AN_^AWNTsR8hm}9B38bfJYqlp3O-BCv{S;vaz}GF10rXTE
ze<(e^(LV*01m-|9#y7Bs>QfD7_h9W+DIi<@PpKpL$dOUh7-5XT+5?V#?8BXtDFIE|
zX&-sJkPdf}uTN0=hP1mt%_VAC0j-kW@02ptGP(@50_e5oF-8@x%dIJT85$>7_}DpK
zV?`{()e|MfN(|tM%sPq7cz9}b>CNrwf4JgR>1S&`@y5hP+UPgZ?Salx4s2JbR${`V
z{VZ!Z9~TkQ35j|+&Dv`*Y4m0g7ww`@no7^J+ze|QO`}o<jI*y%s7xwzU^R=RZ&=F$
z_<@tB=9lbkG}fj)lBGs#^{}F}*&<J}DLFpoLW4r5u;OAkwki5O>mcfYy5DWtfA4XC
z$b=57<u=3Mw0ZmI-g)b7#2QZTZ2fAh2k7a{)8BXYc#YlJ-0sIVQ@L&wT3Hb=m`-$D
z+bpe<2Vc#)N#$Cd={L{Xh@;URA8(}=*3Ocjlh-|7jCL5}=E-AYyldd-lMV>pEqi<W
zv<s{p1H59}tO)|ZZERbOR`Q8{f88hs>lN04D~z=W-?##%(u8se;J<YR3@keAAlnr%
zz@75W^ebQxQlC=xD`0J4xB}Kjhbv%xa=XOU7GOg}!~fkCFjOHdx7~EvYkTy6^a_{-
zp;dg7D_|%S5xlwa4j4iI&?PXo<klse&fHJE1(pgnEyInsz*4cM*ljrKf7UriNof9c
zbGNR{JM7jaD41_elvTJhcB^|r{G6=7Z`{?vz)*@Rjq05&Je#4Ra%VMqaAaa|H?%gh
zj12J(>;Q*#*-cH-Ylyg(m!tw#MnA7Y@~XG(4&!S;O_S@rjJ=addzEh?AD@h35-rlu
z20e5N(awzN?o1f9rWeL<f3OSIV<rY>OH-2urB6PsIyV|`k$O78O?lOXHQjrGe3Gd>
za|M&lR0|`?4EMbJPJuO^q_eZ!qn*X=5-P4%OwE^6TB~>ElVp(+qFNHuD74~tH&0A=
zs^f~GTeh0zUaPC1u2;5H3$H4{{rqc8l@Ar)J5&ud)35DuLW$53f8F-XH9zLuSua`v
za=qU1;aKX~YAj`@S=TRlGEe>^mpoY_m;=I&s!`eqMtD|zZ})euU9Irit}soP5GE~h
zn^!vbE*rE;vuM)xk<N60d2n%ezRrAHl0|gbPR;z8I33SYarJcMilubN(oEqYx-hTy
zy@yOn8O&MKe8`0Pe_)94Qx2IFE~%t@#UayfOR_dUCyYnOKIZB54e;7ZpH?Zil2%7A
zbIW&{)v}XUtY6#d^ZIbOKKc#J`Q00=uTvvNt)#5n@%p(639H?fy&-!;GkT>v8snjD
z2YluKRBar8eNT%wMTO8Rg52#4Lr@=ksKfPv>z9xrwQ=;cX2`YMX5gm@tOqf7DF#nD
z&^%p=*uEf`aLDJ4|E$2gvOh}c*HpE0l81I3etmh$+T)08BC1cA#PFS&tx1*~&AiiZ
z8AT3zLh1jsnOO%8!v?Esu7n92L$a9>_tXSusm6Z*v!{EPkzX1U0ysC9@D@82mp@c9
z#DDY+@aeCwf8I%ES&Crezwyuf_VT*>Z8r;GjS`@i&);{k9W@B;)uQNQFw5_MU%u=<
z%|z=|l7{48MlQd9U8a^_S%X262Bo#e3>pMUYQ@FcAawkcHHg1RX-F<4O(4mQq)7#X
zj3163HYuW{wJ<Y9Qn)mclA^yhQ;%(u)_+OAWXb{~&*fJVWoRK$>MRb0buiB|twZbW
zMNliA<ni)(y%VTc`U0~(Zy>w${tok!)&vB%n%8d5t0W3EEA3#^_krg&dTGDX+(Sz*
zbFmTQg(l-Gb0`0yN1wC(YAxqI*eWHIR+Ur37UH58y2osV`7QdSAwNX29AlnT^nXJW
zt!f`FtM<>9z*f4R*E(`BJI_mCA<=jAi7qmF7{iZB^JePXQYi%@F8d@@H2+W-EK^_w
znk2Ag1}an;o!u`vP5Eli3a5e=4#G+}n`ezeBz4l_;r^^3nkYy@qKOjx!bWr<%ozs)
z)e7gdPxkdk;G79TSg?1*xryRwoPT4Lx#(cPzmgb3%Ybo&0z`S5A(D%HMjt^N-_K&t
zJUaV7bjp)Som0z({~PZGVL*+tDWdYD@?SEH2P@$sPb#C-V>U^n>f0ubGDG@`{ME8E
zHM3?K@+HR@_$5Co<ke2JcyvNCmvEU!)!v0JW=;!w7F}PXMdySKp*rOYqkkwrDDQx#
z27xdf>m07rPDDuQdSEfE!$a0%^#|;5?i7kzVN_VncP|trvGt1jC`DVHr;PhCMo7y5
z8*PdExy)0EEV$3>tz~$(F4%Gom1d}N6rBP~zOO9<sOSINyvc_>+b{OLV-MfQy-clW
z78;+419&qiptHhLjyY*lzJHuu1N5=?Y7IDpIXRcN0<E4$o%X%6JQxVV7`4Ge(IT?u
z-Fa&*-l2$|3;k}Pv!ZjHfU?3loGBzuz*YPzc2GmODuF+}?VC4d+#0_4mephm$<o+9
zk}V$DW?WDIe8SNBCi_3UI|&JCmjkR70{?S*m0D27S@l{x4qJRJ*nh0C%6lkW@Ej?A
zZz4_1WT!bUMw)F26i{YMN$*6xdenKeMQ<pE+f!(Zs2{tPu(RjJ9#{}6wuY+xy!QK=
z^f9f3JALLW;}<oqCujMd_O28@w6T=!QpOUb;%tBGOTXjLn6mvzI<?A}dI=Ne9*T`t
z$@V<vP;hvA*Aghd0DqWgjNXL~i=W#<wRlVRi8>b!T35YJvX^i_TXe@yT?@U*aX)jF
z{pCrU$Gd17jT6eOZOO9lAF{2h<EG}&1=??{)(WaTy7P0Efn!%el(K`GHU=wOceJ(e
z&^)9?(`KievFU+u_OYUD#W!WbW1H2u7T;>bc5y6p5j|6iIDa>IK^QEoP{uMkdky#L
z=Qwt4rt;~ZAY~EHjHrHNv>j&ZvH|87CeBL#G{crBZA+G{k+c3{uX8@392D&i`YNl=
zQWdMWrJwBL6YHs9wd~OO4Xd!U-&Ji7G(?q>YgP*zwzpmEPA#CH^{D-%o1*z+<Ee{c
zs^3_rc;i^>*nh879|yOz-{R*b+UMxF(g|wegYL785}mqeYg^o^+_aj<rjO1w5JsPq
z#Q+(1A~^GNuN>W=K-$er)UysdyI~AVp2YbJ7ZF}wSE|$A3Gk^*@Y_OQ_i;&xM*ypX
zH_LyP)GUA|4-rcNT*TC%=_kVN?!&*ne)-eqztLuF`G5NF*i|k_@Cv~%-}#l;HOjc<
z8^2pY!Up0&1VUX>mlr^0=jy=P@OjyxA_g_>hg<8n*ct@-Sl?)aQ-#CDj%9+bsqhdf
zwpO0+Ti;l@s#7GUD;x1GQsJ@jmByD*@ijjEd+AD}{;}-7EW1yvRb!QlUA)YMgQcG>
zt%5*s8h?7tmL^J*r9!}>mEg!4PPF#&wyuK0Bx;h-3nzj8tFu*^OG_trX(C#MLTRe3
zP3<T2Fv&_S)kuvnw2YJ;IU++Q3as6%;r9RM?>z1+jr!RA&2%0>X@SNr52zqjr-@do
zV4cXaxqIn7XPdrc)faen$;03DRz|oiLCHRP&wrhmBT+srov2x74^3&G-mPe$O`pMc
zD0)tcnt};6Qd(3Ld}?!-yO-VtxUq(r?fUFV2={&3?_0oo1Fki}wV%1T8(cd<o1B??
z-LfA0usW1CvNR{S$Yu8-DB)JQK8-z8O`P4-*6mPD7Q4+;<`YtG3uC=@!z#P=-?GYP
zVSm{w|GA9aI4V7KVd))blG+BLu<eFT%S^5<pKJPyWHwXz94Yyawd8Z<^ThCpLN*sB
zHH9eKmD$urT5i}*%Bm^0k}e^hQ~y|eHP4qHv;`vqVa3E%F3e6PQX5Lim7ph49a?_<
zc=A}F9F$G7s7Y>YY3qBA2IAH^o|a5h>wnSP1;>7<2r<|E$?Xlafhh0P`3mLJhDJCO
zRsy_CiobxO^@5(&a?a=G#C|%>X@@*R%SJt2Ya^^{XREf<@cYQfCk7LM^C?4x=nQ1J
zHSPzP&IR$5$A(qQU7V@PvG3p%>Krqij(fJ7bgU?FcV|ddODS{Us(Rf?+4loOK7Y%q
za%~<^@|Nq6>lu8yG}YKx1H)V0eY@HcI?W&IsOMHkN*PaRfqaTBE4_dwKXJpZ=LT)W
z-R>12QBL|MZcc#&-z1oN-x=<q$GdBeu#}@rrQ1I5c?xw~2vrpHy}}<9LgQ{YIRBG{
z(21)b6hfz2?i4~NsyBqtnZL6kbbso6LkOJ(xIKi<%XxbURfs}D__;&q_2BwH3!6&Q
zUhe-aYL*kbM774qQhDdMsX|{Wp@~;)-`<FxN+|&n?iW3;xXq^pxG-~aXM`;Ad79-`
zgj68Vu6Vx&Aye6H&EE-x)Sfn>Zti`*SeJ5#Ar-nNe*jOxPrF{(-4VfXnSZceD9LZ#
zg%bPNm*eY&l6ItV=J=@{MntS(+P=|^W7TkYKV0xAhxb?vV>mEzXAbZk)6^%fAFM`<
z2spH|F3Q5n<7Pz%wrRWf9W&0?7F=&YtScB+&JIUCN7j}a<bAn5NAyy}gM=wX3|d$W
zH>ZeTwa^NFvJ`RR>IbEW(|;^?N)acjH>8L&e`iy~sq+mf;w-@JDdN1GbBb8*%$8e=
zQ_HznR?OHap>4a`d_Dnt-+kI30N0d%*uJz?iOKCLhyYSRm974{aZ&Y}OT3g@ds<^n
zYlWu2!>i6pRdTj1HUL%$`dInUtv~j+g^wktgD%vRjIA3#74f;XXn)h6Or3qXpPO#p
znyc+tJ1@M~E3M6tbonwsJU<<;QtMk&Pc|whWryuXyWKYHwrbJryVV`;E|=R><^D;$
zAAI7rw~j`vSMnXVEnRwHlv^R@?XVD>FxC$))b3@zi2zK*TVZARNs4MCGjG(?#?F6q
zNp1d1$<2QOFO48Q3YRnU3K0Z2H#s?%0oex=mlt0%#D6ivd;II$pYJ4+{NcsLQ}p^^
z1z>CR0#pl7EYNEKw*^d2!S?|$*-V2w;Ao!tM57jHb{ameehLg<XO;Ev?DDfO)Scs^
zcV5J(E}_aQy+KtS?NmLg@KC;W8Q-UlRvw3^r;3~4SfnTKRKCD0<MJrmcbSb)Z6rAQ
zV;!YESbv+}7>=Rg%yc<D>@UARwh`EuO^w3Tck1(7$+-1G>$04beHupTvB0VPInKOQ
zg%^+aiYQ_`&o~PXPaJHt1-7HmXtb3_jyUi7dYF~vVkHOGiljZsL0NU|^cRel_DlRO
z7;T_alsNwd*GT-v-DJIsN<LJrDx<#dv(Ojtlz&5{lZsWA3U?yWfD*z+9npNivn*7V
zb25nIh|g-+&CEYbr&H#3j3jBQ&EYavVI3IFecl22e`IvfsaM`v90?8EaYF`d*Q7O!
zNh>!`*l3YL({y|2r_*p`sq|z6NvBB%)km}_XY8DEA}ULG%;p)8HB+|0y|>;@Je{na
z!hgR}^DAeXrQDEKW7EW2gh<laMmkoss=m$5BTwuc21t!_2g24Q9;J%7m!X67Ut*74
zioyef$sQjow{(nLTG3|em`+G})+C)=)sbp-o|uLnyHzSH(RLxGrd^X2IW{YTGn=d$
z<Ya$iyOqq+<~D3X-9?}?)U>Ddxifn3?0-8?=Cen>Ou9W#)ocgjY@94oXw)z%LMQ|`
z5tPLJQlwKWbfnRr7Jagtc)D-d(qI?82+}kSzS`$j8*Fv<gE~DHUy2Yw=L1TLM@cC(
zPBtBFfE3B+uqS`LPEXHocfYSlw>Al<{POwxU4CJW5+>?JX)hdjoj?A*e7XBHgMY=q
zN(u>U>VHlye|%l0k>4<0OcFFpdhgtzSxSPs*~NKRa1!@c5d!c;4G~?(Q2y$OX#@~A
zELK7^>Q-11%DtMbceR3J{0L##7X2ZDXI?=rCzI*cc)Fxrdfy`CciG~~YzLn3ByHXd
z{a5xljIcdwr^K2N^|6aZW1LUKD}Pu}fDR+Sz3v{Dc1Z-6W-KO6g5A~R90PT90zNAS
zJ|hF(L2Cr5UP+ctXiat3^OARji^w%938U2v74|LVakL9AfUhu%b5?*-w~ZaIK3V0x
zP+rfFw49Fh3W$*Hc!)M8h{|Lel)tamv|arsE?6TVj2JZUav^E-PEd%K?|+~#8lgg3
zLta^cOZdA>e{v$(6&IBK;S&Vj36;VM;;8C<Hk`9t^Y4X4PKWN3s|8w1G@Sy(s!qE1
zNHlSMge0F=QoZ8fgd3<d&*&rM`=x=TRVy$ZE#G3%<qvxxf(-u7#a3tF5woKR_d?GT
zgRl^;&Jb^)6X-2%O2X@SmVXPdt^QxZG>n|m3-tarLR<}Y{oUU4@!rcEJ7x(!l|y-{
z&-;IbHP=+?+KNk`$SY8Lt_hURMM)MsL2&AdC~24TbjBmZ_r@*e`TDp;=^Rl`gp|~C
z?TXqW?L^QkhCotvHF{%(HhP8MoyKpyNNcKp1iy1|_^X{Le;U7!0)IX$f#_k#C2F}6
zFQ6r)9uB0Pa9}GBIF~wGt%VKEa2rbNd119CIUdVwQ-Z0d1hbYwHj4{2;qokzD=E1$
zPnypflJ<xNfTAv|U1*yf)@l<wny+K80JA7(k3~9xZvd}6K`*-I7>kHqH^}%5IN+q>
zobhRRo7^u+_^ebyIe&XgQ#L|;pD9DOhO8_K|0`P^MuvbaLi+D#Dx~WL-+X}5?bx#<
zQi!Y77=SSbS7$zrZ0ItiWDp=WVXBuc_|f${m-ecQyIhcI6)(-7a)Zr@N~VpT70OWe
zJB;_ZpPKE~t0P_%qX2j1l7GaKnS3b|TEI0FF;y$KhgPWN*MAQhOU~W?Xl{e3*t2ih
zA}hurd^gxb;+nw^T}H8T7OunMCDKfxPzywR5e{d#T7tyY1zxqNW%q*}Ulm~M$f*MC
z4;nJ<ARq8yHe|^NZ__U^WX6C4cm5|=u4XK^t6WWx|96$EiIAUExtg$O%PwnePoK_K
zt-t~swPN0Ly??TnaQH^GvJg)so>(#xo82qxu4bIERzw?CoO`(`=ZrG~Y~G9|@2?4m
zYu4+cA{vNZjAn+|xwctIGujH5!VL;(6QsS81C~2xydT+Rl*Y7&rF%ChKNtn(%fxpU
z%VsFIE0(PwUv6};j59-s><XPn2BBZ`oheEgso^`_pnoVeLAszQHRJv{MJb(>2>K_~
zq#()@!K)kAq-Kb>t4V2}M3nj$)TFcnBW2D1RZYsBt4U?bZxRIO0*vb&DVlYd3?Zjw
zy%H~#gZburPQF=ZK~3Iw>E{7x4}xndGKAgJ9q{?kc@@&eF{r2RpGIRZ9q?eDPV~Uc
zSV;leEPum{-80_6+Y&m{B|5evK*T%qUD*IBXt}EWeSy<@UCNQcDVh}B@}H#)3#FaS
z?%gXDd_G$P|FRz3efZbcFMs;{Hwa&rum8?dA!vpGK&7~RuaDkAA{%vBzTtBez<1DQ
zB@$#-b^#SkAZD&DYy+Q{JF5d^#xAnaZ+SEc8-K0VY;g(IL!L+o*bvc>A+}M%_l*w{
zRUJjyQzIikWhl5dKf-!il^^qmKhsoL^^fK5%X0S#x$3+Qc}kQOaPshPi>pdl6(Brh
zoNYrur35=kTuA{$W^a`3w+(`;i5~2d)w$rH;I9U{?U_c{zG(<x5d<2d?|FQO57S;@
zq<=dU2$b^OD14<Xq6)cNIo$vL{2kYQg;lTYyTZ9w+6W7Q;eEcif+#V=Dyi5`a}!Rd
z*+Y8{#ZMSDR7_;%xQ0zY0OhBJkyQ6~-#7Q^VumPqrPT!tnFEZ@2nF6KxX>7}W1Vrj
zig1>`g;$1pQ6%=m)Q-5=i=@>rwKv#PsefBP--&)jm=*RqeR{u~A7C@92eARj@I!5W
zF)3{{zaSXaCbiZ4QW~nf9OKcwdM;SQY+z<ay~D|$D)3qmDX-j&tyAm<_Ky`KbUO#u
zjx;t$_H}Wuy_2uuhlKYPdt7Sb6d;^az1t(4Zk7YX1hMqIs<?eKIAJq5VN;vA+J7wM
zzjZU+G#AD6WZhmyp_5+iU_W==Ez%ng!Lpqpm6omO9%>HA+y8Js?I}L6R^cP65w`W=
zen=TpmuORTqws**B0JnL@CxO!N=}={8mf^7Y?Nhi<X+?9u*Q!%vDr3iHd^~mosR{L
z$Zhx6Ntlh`F14ChovMAdhWj&;j(=wMvRr>x16&ed7iMUzQhbt@;DwQCUY+yMp~j8v
z*fk*-Mo&|xK|f8leItA|rg$5kk@=u4&)B_O^^E^q#%UPA-m?()luxIC*FfCWbL%q=
z%Wh~`HvQ!6fz|kAYaynpcw*L)q(60w3OX7B*)SV7TbEBalJa#;L+$0db$?ZxvQlPf
z6r<L785jk<eI5O(n(j+6akA3$^v4q%vf}t;`klcRzY)sC{H7n2$hif73pY!-FlK0U
zv>=Cm{L$zrRQj#P=R7NJ<2`f6BZu=D_mRuanD21->4OaG+v?T1O9<V2d@=q+V7P^`
zB?zlO*NLm<<90K|)76&JAAdXF$}ZUtf_B=^l=6_g=miX$?y6i{Jhm4YEy8Mn=6v;6
z^VPQ$UD4CI&=%Kh{oKaKtyA*rCUzE!ZfmUp!x_}HS9v(I#=wckX;CF1u#SE@xS<Xc
zdnkzn<JI1@so36cUG_}|xU;wX8fy>IS%$B330hm~XFEOkC@&sTH$2^Y3NFYPu#~eG
z+mD5tvqh<~EXOPmyA9?yt~}DWj+ce%y-~l@<<QFPuwTU5w`_FqE5&18|G$*ctt^)T
zYcmy>GWjYJmnds9!vVLq>}xYF2LnV<u$SRr4-*48I53wHd<_$qN2@(3f5lqOjx47R
z-p^C?3mk$m=0`djsb{7qQg%0*-BngsNmf}T@4tuvLjhG?-BZ)aT~z;2U>p2lKD$bv
zen08ck54ZjPd~h)J$V%fyg7aRb^^Yk`4coZIy!j(<q%FEf1LjNU2Eg`dg04zjmVeh
z-+{mB`5Qvu9zc9WzTo+rf49Ib=t@6)<-YH!f9-3|+{&X(y5r?egggrP9`HTz&%b>9
z^rXe)**@p3V(Hbsq{p&-mL-y_dMw*REQ^4Dvo++Uhe>Uzm*-w|mKL6Yz4YUyU$Wze
z_W+yg6@v;GSN5=G8y2lqKCyh0?!A{bYs>w-a{TSIXxsLVaZ!2ee`am)R@O>ImzFDi
z{W$%*d_4WRUPQFYkU9PDkEe9QA}9>zWcY5R(_g=yetY`QELHrWmF6MkgY@aI-%nG^
z&jPckc_P|5qG8!U^X%-&5hxo%orrxa4ol0E9F>;GzyJhs#9E@&3|N{a@|tT(l9B_Q
zz(JS)){B$GSqUboe@}*pv}pzvUSxS~np^$IQyyA=USMj#D+4QQ*ht<WeBmazjfpDy
zFptvjpVyCpCHR$o+Skv$4Lw3VF-|G2)VqE4-Y7C=1qwDnzlsfnr{yEtFtl_wG>9=U
z$i!wt+_YgE<+q}w)0VN`War)N`nZze_5S=VMYi-^?nuAgf4E|U@*ZamsF#?L<5hy{
zONlE;%e^Jam9`vjE+yh-nEU5s68qz3$YDcxx6Fnd4LRhj^2YujLk<N<avv9QMuCqB
zIO+TIyUy^wad9#T>;&}B2&C++NuWvBaiQtZK{E@b?T{dGZ0y{k!6&p@ry@(31YNGg
zL1H}3^FHDce~E+BT>bII<L$$t>%kt8xH}1TSAw+Tz;x9kbuZ>$%NDIIy;c^WnPE=*
zF>M9CtvN)s+P`pl{^qnf<{s2Wp<b<KJiL;qog(s&bq@li;4=#in~Q`H6&m>ZhQSn)
z@xJ|n=oDi3`hLN5R>ze?&d1lzS{3Z09YUy(Rt~YNe^}cFI4FnTSN9A$rvT#pdj>gC
zaAx+*7a0|K-nklviJ$AL%70AT5(;EwgJ$Lxx_+))`fYOx#~2y$=@J7Bj4W{=-?-On
zW<NE<lf04W*+9+s&OM7O5?V$Xa^J%5`Y|-HzL$Ewx7M1Uwj8){z>y;muGw3c&Duq$
zNZ%uFf0zhcu=9%O^Nmp)-kI)Aqo}Q4<}#r$+MKx_B!V_18sKLl$me~6*uO#q!6>_w
zI3B?SM~M&3J$>4bARl?kL(4|)Ia#C6Wls9Mn|scq-aoIua<&cqbX|YRJs+UHa;^>a
z?z;NQfhJHakrFzLl{h`MY(+}VjGPUvRkTpDf3$p=^3D54A|-|{Un2oxsL``jk?OxR
zlbYA1yi)o+0C4bDuIm-7oH?d~U=C=uG`q4;GL>w8Sj^0ZxPCe{49mYL(|N7aE+Jd)
zN(k;of&|JSMzu7zS#Vc)9iec*!ECr9uq{2mNO3_f`-mgWl2tY~9Y%@_2Ku!zP-mS}
ze|^6kRqp$;K5h2Z;k;g)i+wM!$!ij-WU8fx*QQH+p8J`_qtYLw)wdRbY&Uv=>|XJM
z_X#(KdFE|awUld)LOw&dE`l|mI*s>EPs^c?+g663E8x2-^?qI1I^=wHE$K@;eAjlu
zm@EF1Mv1^HP+E?c7;o?_=$sYA{eEUZfA-SYSPK)ft%pjyjr*Ws=8}@R$+CZN#SBvz
zi))Pnds&AmQp?;27Akm{IBB-*wQ|d_u>mO8Q^iK!FMB}*oO0Bj7#JZtxpwHq7uauW
z@stg94MP`NUQ6#rmE2b4ad+*iT^NHhuAR47YnJaN@XWYqt9q~f@CEzp=oO<6f18ZH
z)?=;o&+J%t$0Q6k3C;=z2PAZizp-v7AzGKiHSY(hD+>ZvE6YI?-Qw}{61eRbcsnvr
zbOw*{_S-pj=IvKlww$?0Yzc%Ll|ukhpyS+|YEgo@rli<4g&11;vp!Wq7`>Au?J6Tl
zCvt{R4PCf^awKvfF6L0y(YgjMe;OE)om(;pP~?tm%JoN7_jtgLe5nf<M+zU!1J?0(
zyyAh+NsJ6qoF(eFZJ$<8jJ=QeR5G?%Z}aK=KzZ>UwLaO;G#w$-wfUa2c0RxE$5DRA
zsID^Mj<h75kMJ&%dZ=%EdAB?}^nK}k`w6|(w}Y(9@p;&&z|QNkUFm~Ve;SNo(*3#>
zYVTzKCPuY93pI69e<6aKRKBCtBRD1q5#7HM!HxF{mn8c}1fO~zMTH=2mi7f)q^fs-
z)oiXHLr^*E-^a#2sWVL3jq9rmecqD!Q{ivda(Q<bM>B@q=9UCBJ|lH?PZKLwVXRjE
zTNzfk@sv9<b2hUE^!s*Ee|l{t&ZG2xQVGP{SlhL(lD6CZ$Kc?pxNjWY@w)xDgpBgE
z8ol?H5PB)9yflzpe?;$jX1Hk^J)3+~gzBAqNJ&Yi<vDgKyU5!C)tBd51dyV`;$hPy
zy0I?ZnDnBv7AY1BCY*}g&Gi^;X<2kHt=Zn!rpI~Qk4Z{-r2ix&e+f9I#?>fjlSW(~
z8kp}hXV?c%-G(}-*u4sJCvc5bWN{w7SE1W_P%<Va*?p#Wd!O#M7f;;6Q95pNE2=?2
zrH$`jJ)~Oc(LG4-4q)Ky@G|2Rz%1IHZ9z%fwrggox9d4vEo)qg%)WF{<o2mw8Jk~A
zY`XKP^D?iN2}nIte|os?MIzh1{3F0ll|1Wg?}6?0sy0Z;VUbQ|Nk&U+O`A(b(MHnH
z2d4*N*f1^gH+5d;CE?D9c<I-&x6*d6Y?g9I(>8{Ht+W@&DCgY}1aYy`bau9cy$e`^
zZ(P4PN@4r;sw?R?pjAHLqtkEDOf1PCBX76!qdUkOfKdkVHCW1&Si+RDu3*}F`0lK@
zlH4|DWPmNj|0ZKt_S?EQ5~w4j-8*Ff_egscFW>CzzX6^RF%*~KO)(P#H!wJtfkg%_
zmjb;kDSy>kS#R4&5`OouXdo}M3#i%EcP|D3oWrv-NU%nn!R)MI1hT{^Ok`4^s2u-(
zziL^w6gja>*@H=dB*kWR*L^h=gCe0s49W>I#DEv%EdyF;Jj6gGjMWTSVQu7BNm!S7
z#RwM}nUwH4<9HEVJOBVhAbafrL_V3ENZJ6<QGbxi`;8Vv-to&MNE=?$3Y;V+qe+l+
zfg#`pogpJA84Cj1g1?k^IH7f5R<#$}T0YMxp}oag#FA$o6IMdz5pn{Z*~XDNs*I1*
zl4h#L3!{idFaeNb)=dPIn0I4@A;WtuARSnD7I-Gpw+c|o5R~R3Qv*9DxV(mwL&|%#
zFn`f_&~SiKU_2LvRe?Ex+AJxS)<V(RFkPpFHCbm4R&bi>ItR$*x1J5cazI_y8{P}X
ztq%z_JoZN7w_XdIvb^2^vb-0Huqo!fNjR0Y?7eX0^V=w#&NMvU>d11yV9sPb2<@y>
zpb_xaWhtQ&=N0b=BwPp?PYahk^Bj!ulz*)dESw`{2yS8{@SqzHY5=g}K%XKs>b-IY
z(SertHlOAIMb-uE>LYIm5j#?L;D{YdTJSnJ9aDfI;B~Bd2mssgDIf^jWIO{X5u^Y+
z&=J3d5kttD2O&^o2p)uWwh$YUGViQ(ZTOS~;yPQBo!2UMdxaM^We9Jb$uw}hgMS*Z
zW|pxOmoAtxgaabmFR}w<IU-PCG=>mnNk!g^piIP)N78_N-r*5oDjod%bMgAitSk=e
zy6K9O#d*h1FRJ>3D4sO!xNJ`qf?k~!|1N$lo}SY3;Z-^6PDyzgVd_Yd5Dr=yX%$D0
zq!iAILxICjieH-7P4P_pWDyx~ZhulL!KXcpXaDgSf0%`5rx4EI=vT#m|NI{;AvmDx
z#dLahBLX}gl#zy<k_w@6jVX$!&9rGxX2Ve_@&-{H54)~xYoV8q&p&s+oOHvk6trA8
zYU&O&ju5J(V0or;1OwvqxO{+jkhV+uM+iP5o%rGL<Knn&Mki%=S{y$+Dt}&=pSw$r
zvsRCXlM)QoU0HW?&Rm%VXm8#u+EF=QW<y?jS&pmWljd{QHoP9*qtCFv#ZD2uUUHSK
zG+(CIO#a|Bdj;UL{Y>eS^bky2Ej3K4os~>fJA$IOY&plCs;U18d#4Q}60O;Yy~BTH
z$-lXM!nS?f?)F*JgTKX@*MAUJ&g^554w>cwl(nXkZ)@(Z%DOynU0FKxab}~hotg5X
z%hG-c&<fBiz^nkf0^ADlH^jot?hh>px^#aOI?1^l<yxXNp`cl*`B@uD)}I}0gi~>N
zb$kv31jg^Le&xSEz3aN!{9i>esk(QIb2(}*it^&T>B`Z&q8v4I1b@rHno4(MkD`*q
zt(78)7KM~_%q!AvqP=)~zqptU#^uN2`P2WL$oJEJ>{+d3LPJ1sVC9M13vkBGmq$`D
zdvW;NVp<K4hocX}NmWmZKPQuMJDgWNIdyDbcKJ(Aoe7fo`2jgC&Zo_JaWR}DY>Vev
z)5ZCsnvUOe!}feQoqx`IE1p7>miyvhWkgq_Ig3X&1hO|u@ZRD)sp?_-<&Snao0V-(
z=QYr^MpL|E(&{L=2YDbSvx48epH6#g)lifQ$&D=H3=OZ+ayecg2tyAanSGdyn-Q$c
z%ua{(L^kcD_(;XD?W(uKQ8zDon%`+j#_?DrE#*N4?to|Qk$=4&Qn4$auIl1&C4}sn
zi#3gNk<;;(ZeSa0RSNA6SdM};$TPQN8VjE@_f%Oois?p~-J_Tqw^dAy*`b&QlQugK
zV{hp^))1}~nVd{0ueW1r0%m&f7G<NKTkBWus#R;Zv}*P4TD@*jY`AVw?5Z=d)T`Mr
z-bsro(;6i;CVyL#lH}3cBHutA4;i<QW^FTGAXs}Adh4Vzd=H|LXoOQ>7?~WCIWl~@
z&+2*HmTybg0yf*o<Ls8!?2=XN_O$5G&1w7Nc&{^C-s`B3_qw>|y>aEet4`agdzc?D
zuL3aMOO877=tTI&53Qjw-x(mm$JGGYo4Sbx)X`dCT7T7WNcPj1oUmTW=;HUNyBnzQ
z5W?^%n;@n`#YK-`3}1nuW$<1q>+Yu~e|kRJ@_nao$?j_iS8mVEE}0!LF;3*pX9rAZ
zSepmAc==ZJ>^0bX+76}Er@VT#OY+&kx@<uZ5ojNGz$CYHt{+CX+m|NZ_0sKamnMGu
z(!^T+|9`=CwTF7bwt9l=lf^^2&f;O!gM`)XNLbyDgw^dxu)W$7&2t%sNOaaR1&`jD
zWQfqT!tCsO%xh&WKUE*98B3c-AATxc3_rbT-oEMD>LaG2s{7J&SVK)WA?BMJZcTXF
zjEXa|mR`rZb9=+6R@Ru-Bzhf8JoREYE#}>qX@5DGEUIzYb96J#qUXcN1|{>S8f&an
ziN73anf9w6lct$W%O#JQ#h$|@MkY>10R@~a8h<SnDjkI+liH8NF$Tp&JzVyDM@@T?
z!?X9iu8ouY9fikXmnfUwV0a>rwp8C!G5&4w@_23V&6hdyD*?6#;NW+u7Qyc_0N;Sc
z*MF3x(hiuG@0KxR-j2y%addKj<*d9{lN`4P=V<z**SOXs$5ncbtMnRmzhr7n4j;Z?
zwtQ);Nb*PDJUZEXOi!!SbWwKin(@4N^sIba)flA)FPAGf#nxThY&7?LD^8gA9E@qa
zm6|aBA!(7{ZnINRmiX(dA>&mtPVe99+kb#+<gR&e99H(5EWr6FCY3xnE{Mhl+WcS@
zZ@Vsu+m|3-Ine%L#>liiMy6h4Wb$hfvYH-~>K7QsNQ4CjH~#g4P8b&<orP6>y3eZZ
zZ>rgk4nH>4c<`yMI*yR$W7$r>4Cb?PRK2Z6118+}h6Dql=6Zf9n2ne~`L3t0V1H}1
z=wwvAe6jrZgoSxDDeJNYzoTYc4yyTJmM=}(ay(en({et)dhwxdKGlP+8Jw3_Qf?Ni
zq<1M)>4xEwT>en}qkQ2_+aE1c=rgROcuV3)H(hJ<?Wy@uzVLD%{gtx#Hz#*jR#fJ%
z`dPp|CCzxLy$#X*4x&dgUx^;R?;$7a2v^;jM(<5}2TcBbh9<WICM7FV)PJ{Ih5rM8
z(6!T-0c$f9mpkh-442#?EC>NMmjP%BCzn8IJ_MHshBGRE=!K+;wLI2BN)CSya`@%+
z_I~<&)i^le4LCKte;a@XBz=OUPHP4SAgqSr{rm9yi<HVzd!vR@YE8}2Z%^Om{7HkS
z8DOXdYH<GK5U5AnJWr1__ZsKdx$2dA>1h>5Tp0<Fo&p*JAKUql_g_!4uq;Ke@n`(v
z|GvDR{yA}f0j$+RIzRmR{S*gKgV0_LU`5b&`0>~9?erU`+SXEr`i#sUKjMV0=W9`+
z2U4N5)|gI(AW5%C90Z}`P*$O|=TBBD`i0njnfQ;Ii#<_%B&8WeP3DHA3Io*KnmS7F
z=d1B+Mzvar0KOUZLUf}MWi%?80Hblzs#rKNk()Mu5_`tVPo8Do5?_ztWC4Xys;wN(
z+PoA=6NKDccF1etSk`ot9<cBruN6#*SAv&$EgYq%>S&ZN(Q4o`y38|Ym2BW=61|oG
zcKRkjM+vz^CODx%H;G;g?<!i8p1Ndv{VZ$^fySzhu(gi*h$j%Tz_#mo4z|V_5oCj{
zB6XR6P*s+Vl2mT8+OglKjT#04xS%HA3bc!wqLL$(XwJL=3ZSIGxY_DQ5~l0<I$JA{
zukwDuIT=++wr!rLsw4_Dz^Q`1DX)@VIB6I1j<;aYw(I()F{p$NlRV8??0^^TLvro^
zC%hEV^=>b`j3(81PHhRTJ+Sglpk8vk7gnTyvEVpW2CHp=!N{@m_q?pKxWUURXI8wd
zvZnX4IT|DkEcjI5N%jSwoTC<7a@;Exl$RPcpTNsKO=&`*^kKmPR%x*6OP8jmXb_`x
zk=LcTxF)N%Rv2<hY^4XhFJyh7n>AU{Si+v>9K-|-rG3fCS&DdLCw-0oU(On9d`zQ%
z`lZZmdPrKq2}fzJxfw5!9HUV!UKS0O$RA@d%;-f;mysL6Vv6pG8<=MTiTkSC*Mdw{
z5!z<j8Q=ZLwLJSJ@0~}8wnjup$E2GwycufOGbQDNPysllJJ|;sPuT|^oqb@8VxV|<
z&_22D)$s6y>qkvthGOKTq@Q-lKq<9<)tu4}cMOUoCrdDjz|-x=v2P)*qD%n0KzD01
zn;O-}n~DAu@Eie6bE+%()YdGDtpIj+sN+0eRx>Czdqdz>;DiFRkmR~AaH}AKTj~dI
zOQTUJtroJ^d~gGvD0K_oAC0G;>f`);!>(;OU~4_VK|gyB4th#Q;eb(nfCDCf`<}W$
zd!T?Gj;sqbd3x{c2RVLiDeH1@fa7PB4siU8;sK6d$I=(n!5fuP{g6VsZdAV3Wk(3e
zo0(DE6I9Bk<`j?F)ST-2rsjIu?%Xx)G^%Nj#!OH3V==?lT6gYj_LiVhanMsb3I~kp
z0~|2fcR6>7!s3+KmkI}!oGN*Lh}B!QV65!RXsM|%)lywe<#Gjt<s=j#l+)*gg)yI$
z)Y<9OSC?7{>&ZY}rQ))#AJK#8%E@ZCYPiK=K4WcMmUWDuq5Im_3cbum&NYFdQR&XC
zGY7eHud=3p9Y0X2y)@#=-1g**_DrJ0t)FWp*Q%3d>N!cRQwxpWGIrX3!L>$hy1rdK
ztDsVHUiZsPFRe<xe^#xirbK^qv8bdL!33T&P&{=6?f>E|Q5mGzt)4$iOg33NPh=Dy
zmRC~NYoRJe*Yh>*_aJY|P)--wiiuaBaZye;=KS8(NXcdtYKlmDN74*b2vdT|DeXM<
ztk(gI>Yc;%CXH6)SDP<?G~;$@dP+x0(^I_@C7Lt}(&=ZF=CWOyp3=^k*Q8-o4WqaN
z>vg}<x6;!AP(5vSPhwmP3!YQi6>Y&;;gmbFl#1pCg<DFggWMt*BQ*GfZtxhj552+T
zoF8+8$EZHI!Q=97PwGi~V#4WOGjznBu)?_A3O(Nk8KEP3FQKk~Ht2=jVFIm*A2`Ye
zJ+&u{aM^BzeyzQAO|=4}YSzub<lW<GLl8bphQM6bC?E1AOHS|K&eQcwNw-s)J=RjV
zJyEZ$f}E%SrBKPevn07bbt6G96l%Ja3KUK^mD*#Y+pdPnHR$V<j3@40TMwrcy<JAh
z{*{xMwx?^>V0dY7VCT{-Ur(v4q`7`7-=_6z*-`j5opxqdy{!zIUv%bK<6P_WS7+ei
zt1~xC5z2bQECcB7@8%WFyt*wnf$heye~ahV`#oI^sHdtn2|Atu_5G_gNA-)ELJ*$Y
zyGaXJvf2XJIa&FpEa6`bSnS}J0c$f9m$B^~6PLb>GsXhTdbfs+GtvhHQ`&-KmjP=t
z6qk|zE)$o7l{3VD`1t+$pO1e(NoHA!VB^pDHUIVP<LQ4-vjE0hp{>3B_phh;0yPNj
z)h%dYG2H(C{r1PxFEiD;mZT}d#DD15-+$hwC%>`^gCrG7YmFIH2$J-QH*15?@l{qK
zl28Wd+f9O<FSenKPM~&`GO67Z>ZEi-Dn(MQQSsTZ+Lw2K4@;gg2*3?xqcdfs&Bgaj
zHLif_+Nu|1+tZ=~Clr{CtUpnU$_N9YQnrP6`XqU-X-<=!(W{x7Vn;FA%ynSrVaJK@
zqf_ZP4w|#1G4XF+U0@WT{FCGwo^*04P|=PjMwFZs^$CIs1lllD1RK4aKEx_G{*E5|
z`1F>Xx)7j$F!pEd&_FD0PQ)O?==8`N@=t7xM1f{z;72xwcY2_gwgI{=aSY1%)3?P0
zMJTFIeR2h^U*3&B3OvRb;d~7|af{+!V+;`(gNI4jQ~#=KGarV>7h5s>+}0Uyj8k#O
zoqt1p8)=dDXj+*UWjYI&Mmw^xiuJnCCay80koT2;k+h&FmN$m6(>(ZXor7;Y=TYRV
zbIhv()_j#zc3iVYqSQ<)$`n|8h2DclzE!%X+HC8dXhtJhllHS_q&b;I@p+@pv}N}q
z+ENL~Ok)?U=(&79Tn;4}8+qpEm_*P*Ih<!<s7fkI_pjh0r}t)5F{Ey|OTe{V$9iG5
zD#Sv6EpE&=*2?3U<6uRQTRD%{H<C8Ps?GY3;*#`@Y@$)`$GSEI!_KHr@=zU&Vo*bz
z>eZ4}?wT)TR!-Rg!Wp2vVkAu5(B5Hjm<DRQ{J9`}@ZZH>%cIg*$M5fld@JJHXf-Vu
zvu>XNO!u%B-QUVNbcO~JQ_}W~CX~^WcNGzT63sNq^dOg?=uzB$1-6jFPwy*vZW#}q
zsu=tYt1ZFNqtY-;`P-{DFKu)|ii8Iv%tgg+EQ`DuU9;}C@6X!1m$qSv6BVEOQG%=|
z+Xb35uh&nWV|?d1rapB&-5J_Db6(PQMehSNcVU2kgKFrweP1b+;OcA8fPwpq!6Es7
zk;G}Ind((k+`%$K9!Hhjv~OpaZE1AS^Q;tEaJg&A<_Arf%y$XPeX!+JbYoi*`pkJw
zF3zw;dx|qNk!1EbZDJ*QjCllB3Sq-m-kw8N0_C8}p7Da}Q~W4RdF)0&rA{U}0&YLg
zI@y0p--V;Xf0`9>)4GTrDr6L?s3c*3QF4^n*CSjS^I)4-sz;1sT?LD!8c<+(ay{*$
zS@NYNu1q5|jI4NTx|A}H+NQTf(hm3X*b4>zZ13m;RV27)@5#$xz`TqE#gug?Lyb9g
z&Rf+6*q4&vrajC%e_U9gML}|$78aRSDu9+ANzZ0l5-$}Ah$bet#mpBKfl_vVNbus5
z9D?e*i-A1PD{x+@aH!B@l`sX!Jq%E2X>;8q#)<68VCuV#lgjq9OjtQQHaQ+YAxxS3
zD6-lqNM5C#G{zZQ)(hynYNpHFa_!jf&RKA%hWf07j>h-)YB&@U4Rg7oTRWYmG(AkZ
zrPaO;)p(tyv;Y&R^l`%YV#0TShR@AtLTE=jI*v}wk;K`fhQ8g%VYHxJb*6CUG8TlE
zn|0?xPyx()4Qgc}>1o?J*7G}COr9edG%20L4s7!pENP)+Te`x#U#Lv!19P~tD%*iv
zZ-<RnJ5WU~_qwW`-<ISFTe=s6k<xlfz?inr0Q$XUsiR_&@{(z9Y6<RtxUSiB>l&~Z
z8LI8<SOHs~4yHehxLTTF5P>SZX?6`Z?$h}Sct<5H&J?3xBZSGXbxwcXHngR*Qkt?i
z?D^T_%e*1UPfu3yB2!s)keL3WgT!ylSYmWZ$&pUA0}-frpEPjd%@k{ely<X^j8ru>
z$dW@TBCC9Q(o&cu!k0mRAFIYtDY>NH^Z=_)*sfxU*`D-eB*+ngi+y@RU^gIe^XOLk
z>g<ojf5Vv06#SX-pq~~08mxm+g2&h~8<%NKKNqzcd&%xqH!a2N%fdvtrfVmwU&%~(
z{W^EE=lMzD<1uO~$LPAI@7Ee=x4(uOr954FQ=7ew63cxjZ-|zE&Q5S&*9)Eb9e0w5
z80v|rnkuZ=;m`iiqbQdfa8fwMR=k-OxCo{SsTS4h0(TfX8^@+$^Ys*8wG6`gnM3)g
zIsRTUqYZ6{yqe66qx(wWESyk%GtWLtWQ-F!;5@6CsLGuz28iT<@8xOjq%{|%H?VN@
zl3Zc}Emdj5%N%%r+7WIwejF&5G-q6=Qp;hZ((Zaq=qleOxz97)R5jQ1rG=*4wHw!9
zI&Vzf4>AquRC3P@438;2s@g5!$Md0iDagz8mJks83Z^9&iz>X80JYvNUA9^a*H91T
zNT^R=^V72ehDW{6UD0E|wv@~vY~-+0_wl5W6SR;{&r~jdVv7bT^c5b?6Ah}*o!syN
z>u!Yx6c#JaXz07w<W1Ud6Z$iRP#$Nxr=^Y*;PpJK_V2Zy0wn6d=?=de1C@z^F<Ny^
zuk5-c9ghEWDUthrzZnvPvfe@JSa^FBPW>QBco}h^^tFpb%wzGLQcA0`MwR%jB_Moc
z($hFC#VcojILmj`0@x|PsZ6Dy<@QQc?Mrc45Vz`Hcj>xU9ryNxd+lMteSUg$77*vB
zj%U)3GX2`dZugrh*7FJda%NSP+G2-(qKx}HRJeQ$+8<6#YO(Lzn@sKG{8EV!!o1(|
z?PQjr48?7%#4c?~1;VXGfn=u=x{IMZ*ruE&Ol=o`5(-ANXfmH_fm_C+Wugpw7Z;fJ
zs_cZ%TwK@|LVf(~3W9tWhZfWRMp49AZ|U0xsq)nTo<L#0mgSr7Dou(#dZVwt+~~XV
zJ!Y<A`NbH04W(8&Y;5*P6=r@f_P`a=x2jgC${B&_j;uqMxomgfN}pRP%9Ojiqr)n%
z`GjP(QnZrXe+*rn8`QRLvQ}lMmYsPSDD-xmo)l}V_E;&5CoGl1wc90yPG7*TIJNBx
z&d8-Hkzn-cXT*fjZa8n*zsC}Up?#e+e}YXI@A#5eOtZ2KwIO0!MDAO7HN$=9hKaJB
zuM`cLLEHUPGn{qR6;};0=4Fj@-~_s%K|K2Ic0cNwe|~3^aA=dZ%W3jo^3K&-f4ZH=
z{aAA~*4fhgS*~gwQHqlbA~B;6Zl=Ih^fO63pMHH#BQqjzkEM3bNo$+pQP$kOidPT4
z_?KzE+ia~OZyvMpZ&!LU!ydWoHxM?qCJlMn6rF1v=n){tx^h5rKRF;#X`g25_LAp=
z4U}K#f6eTREpl%KD%fAU&pVE-j;a?&wa-N9%-SJP6}{E$<C>*w;{Lm3S4PJs(_-nX
zj^-$(i=lWD`LUanf%gJ5_W!A(Pk^>b!lkhJ7&q(W+z+gx1j6jJ%J#3K<aWn6JFl6N
zmt!N1tuvh0p3}J9wP}_^*k8MA8O_q{@Kbbwf0RrCe6KEs*_R!*=wh1vNnPx%ec`O_
zI7NgQHi90TBK;MKazQV(7?uIOjy}1P@<TVSLl835Ojf_Gjxs7XmwSn}>&nl09Ju8*
z>5Ar8MTbkG!&>xR=<pyMe70k)jF5VqI+V^nw_j|!0HkF66;VcS#^gXMk1rWR@RUZX
zR&l**eO)q;)9I;zI)rK+L_jl3QuzIG&RNM8g;ouoeF_2K`rDoQur9@=^~wGia;Owl
zZ!>-9seO~-)>O@p@@jlykf<o;&2G+0%J;XAr~d=Nw`0+l0c$f90yH(39-%YV0qeK*
zp)=?Ne~T5BM*@gv7{~Xg=porwYq~U7Z79*-gEEE{F4@)@=0RM+WgB7?sFPplWfaoL
zD&UYudiYwj+mUoH>8+&vG|~_)m=ds-UP1LjP_1;{-z{jf?q-yU>xyP$@(I0IHl3HP
zMSg<LxN^3%7F5UW$x=B+s=_bCI1Ks@RLWc7f2~=dusj&TtS*DBy{6u)w~Zt{fbT@;
zW~b(@-C)|bzB{c>v`p?+tNGF2&^p90#9J{8C^VxUaTcPPE<Vjf5aS-1u*hQ<u?_MZ
zcP+Df_2lGM0oc_}yQ<hx&$VIB_QyG4t=UkhZ&QeHprC{@tQJNM`0h?&a#j54jtd8F
ze-cp=O4(r)=Q=_*+_6i;!%SZThxpm#A|->(i6!bJlUu);Q~g7epF33?pThG~=Zxwf
zq~|-oBJfRo;P-gf&htBi3_{w42Skrf1Ac9V#GvH;_r4w9;)C3?+6lK{Qaf}??q%5q
zOf~!BrXHkhwM5^i0u3R@&kXEY+I`?$e?3Z~_6(4Xb1jG#6NzEynQ-%@qo_AGRa@)?
zqqRegC9Ouex%gP&O-<BMQ6udvpcz9oh_M=s+pOX{m><c?{L3h>Hba+V;sbMrxvH0j
z))^bade+<ONcV#EgN9R*0_8o2{FP?y3*t8&02SYGEPMP8!=0vYU9udH>3D6ef341L
zB~+K<NCCT*@;zNm^1@StgV8UBXj`oRpA`m(N*D+H+}!L4k{Ftu4XHQ0b+un)5VK_u
zebA#f(wf6}mgQ2$xov%;t)<nlvi3C8D$ouC3erF>&`UM!)FC=2Oi=u0Us2t{JlJZp
z(2Q4FoIR<uSeKj8*O6;vS`+)4e>iJ=57|P8Xjl&WT&M$Jh1Of7|DR%;$}nKpb7y5+
z#1TpCN!i#@H`7eS>>%=*+p7ZoLaNg?T!OICJF~0SnUI{^XphqN@si?fn_$n>1dUgi
z%bR8NsoFUcdOq1I60Ff?f}!aV%RKuCkZUU^Usa&L5j|zC&br5q&Qtefe?jlYJ^K1x
z53B5HU8%a34}k=;OGKW;6yDFJh83;+WmvppRG!K!OXE6+Z1p(@!JJ9m4j-m5iGv1n
zjS|e1&XzQoH9AShXE_g)2zqoV-(GEKHAv&w;Dj;$>FQTsO(n_rk5a$D+DTY}Ns?AN
zLSjsqZT;se%Q08f7|nFUe>FCo_f=&~uVf5-ca6h7g_i}P7r^?BG(+E}*2jWcP$Ie=
zYTpg$HIw>OSyF<qE^Y)5vTL&QyyB4%O@`L(3z(^q6M&yoir7?ey3htv9lCGD9mpia
zaiKB^O8UEcDs;zmYI4+Z`kTdg)wWhsdapjDLN#OvuyY??Z&iq%e+I;N!yBvP4H&S!
zA+2(?{V6T-lGgBA(frQLoF_9E40^#)_xc7Md7x`^Fm*zCeP%+uS~Xzdxbh;ZC5|Hj
zMiCv7qn0f8J~EGH=NHnFPTxr`X+{;k$B^!Ek16A5NZjA%HQc^jxaru@LgHMu{D`HW
z%YM^<sGUdGK4pfCe-p|m#<+*!+X#}^c1aqt@C;z}x_q(HMkpIE8Z$DC`ReMc+=Es|
z3y*=}yjp5I^>9gm3naijF!>z=ud|VS5`$^=G#b^0g~v#2_tuUM-5%>jplV+3G>xiF
z1z^q;MKXGKeq(Edw))u{PpdaKJ{YARF2nOd>X|Ob1@+TtfBnX(GW4+12gnF0p-*<}
zDjfE)9}Py1F=qv>D?YTneKS{jQeJK8>S4j?oz3l>cLs!dnss#XPF5|kWYbSIXl#9A
zQxDr9vt=IPOr@uLFz`-mBwZ`os3U1{jaG#}r<wv0!_!t~TAe$Jduia7XEd4oS$3xa
z&F*H~6;JiYf6^J3$+r+qsXRKPhl&<cZTloW=G7ReGOpKJRj;kQ&7-uX;sm5BSu0h2
z$JV2s)VpYXtoz%teyXlnQu{3Lbr>^5)EebrY{9-sTd5-X7iLNBEQl)S-DquZse|{*
z+sIf+ld*kwQ+?|;)A@WHc~rb=m41s?k1X6A%r%Y&e<R{Z7kU!(@TUQ3C7m)Y`z#N)
z88n(?mAj26{eLtl{?Q>HwdN1%nIQ;=Y2I`{ZR#xo8EQVLTFTV;cJux`^A%m?RHis+
zO7Zc{&G+ildiAFY^oL#PSxG<rmNH8(>!>m4gtrW1Y0vj4j2bzWs-zt60RhRfyyqM(
z5jV@Fe~t8(=r98;X-}af*ThK{^%L4g_&3jrZ8gIy5dqdYnbRLNNfBK#uIbNj?Awo@
z-~PKE14aa}mw*5HExkxwgd}%PMB^@h{pa%gx4(0!^jExD{H1frYST`=72cJ>Vnr$s
zNCZf~-hD7^Z7eC4r%TEb)(i!DZBy~`Hhj<)f6MU$-h;Yt^HeGg(LC$9=hgg&J|D@J
zj=O^~vx1$NNg?UXBKx`WQKQ%)w*rUOh~I2(@eRi_nE7Ybx+HN%FL9RQdBzlh+TE}Y
zPyZhMfN>x5JgoAYG->BJx21eAXcds_xSH*<Q@lj#1CjW9_S3OPKdJ2Ym-Ow+tgqS3
zVwkR7_^-4L4sR<XHw`8rf#B1oA$sBEFmGmj4n&&+2<UI1$BNSRUN@7?(gkk~vibKi
zqZU#pTwzT8CjL~2^DxbgexveU-~YOH04mC9UXT1{x!?H-^VR+$qy7iXvq>YD0c$f9
z0X3Ha{WcSqACEpLmaj7}0$CiFF|RWj2Hw#837TJ*P_Hv6VZX6%g0n}4JxBNqYxew>
zPn)>oWoLwR3V08A501M(KYn}D!t$pxg=_pekBX@kei<tnt&w@4xZ-c-5SbzYfA8b_
zj=}|v7^{HB!4WBkStAEc&FKPdjq;43g-*I6)()4~uQMKhc9L8fPm(I+GfBEk0n5`(
zt*pzGHDF13@LM&|Kqt?^b6i=TD7h2Kk$0Jqq=5t#vIGe-@wZ)yTKWiq2hnQ~>_w+Z
zpJ9?i@wN?vR+e0F{Ov|<hF>h*`p&oLGXd6o#`sa`;Cme;%bhz3e3n)uX}xl@cLGxy
z%~snq(8V@?v=>VpYweS8hb4X!-)6;PzJ^LZFZM_f$83$pHUd@j02BV%0}|w3&im$k
zj2GG&kmyLfqzEg#Tj3vOZ6_~4!F80AaMkl^B+fEnqjp<p$wh@D5Q=l;`oi@T54!SY
zthJgvw$dc|jXK(<0imE{JdQaN)p_b#wCXvo@p9}6PP}Ca)*@O@dB2zburo1#H_!tA
zG=tGr8O|HOz7Id%e(%xd09W4fJCKi*ZkjjWq+X{?qa7EPCjLQlQow-_)ySSI`k6$F
ztIZ}Gh|Rvw!bNYkMe8YEkCI_4?<A>e$q%w_DQ9BV>uLAybh6W_$}%+s%u631=Q(Qw
z4CmjH-)vn2Xo|gwpX(VYLK*RY&zz%E3`w=Ywss11@$pck9+xpV$6e}~o|TwMC~(D6
zR<dDF2_@6Br)S<!<nu*ZERwi6-=r*l5RB)kgAw?1g?~S1V9~Ixtc_xEc}{zWS7_P%
zPg-_C8FVGeV!m>jzAc%(RIkVBkpevqh)!8kDhcRJ%QRJ>wq^3<VjEh2^3>}OGlq4o
z&ecC#)m-VSgb!-$f?dqdxMR*ApSOxZg#hbTNnOb04)bZ++9(`t>F{i|ZgzW5Q|F;c
zRrxtmEY>(qG8@QNffCpHl%M$B(>w(3k3AJ1rROoaQ8{o+oj#A5f^y_<R>`$OohHzr
zX!OpmO`zOjs4$iaZ2gXZ`@zp`bKebGw}x;QT9Gc!WvfOU0VP#WPe}?R&X!#u!I6Y@
zY(eEzUF0MUCl2$us_C51w?oR*@SzP=+KCB3;H$BWLfd%8md$)_rI<%Uk!Cp*3Fh2f
z%E=oke`+mu?><4>uKAp)leD`jXneCPsUq=sY73EYbe_X|lB7s~F;+=W!ZD@BHvRNg
zH-WtjOVUT0hb!98Lw?s(ch-?&x4`Xrcm=w@jV>*8F5S5mYzOVr&XRaoJrtG_{9uFa
z+**GPqftC*-|DZmjuU+R;HQ)(nm4(>9`&{U+RO&I5NhpR(uHmvS^bu3Y_>CAyj@F<
zCU-F#*uTlUadkI;Tq^qdt<Oc??PP=wn_a1Bui#l@-OpYR^o_b772pDMPEpMSP^Y=W
zJdr#jmrPSI@=U0yk!zZ>x5<E9V|pZI6w}pZ))TC9w1Hc)dfcvRAJk~*I-gm+!msmF
z)rP`v^QJ21(SS(MV>dBy%(VD1jX?40C0#V1r$%tvU%k9#F$8MKGMwrzR*KZbFh))0
zkHm2Ere@eX<-K1pth=t`ag&)qiq}HnbqU^<p)B{amRpVv(<>rCd^ebSgi1AJ+~!r}
zxp}iahSFQ%d~u%MHP>O4r)_T=i>I{|*DqC);i<-$-<OWGGwLS28-se|RNFf%7(uxZ
zI%XD))g7&x<lhz{Y`Gumzj=`T7r9usZ<hgUGZdHMrvVcKHaR(yQ6DCk?zJ;20Zo?~
zwlidZ37a3Muanb|oy}Zc%VYd@I8Cq9m;fcVf!X|gnaTzRupxPqgJsO-?U#9)zKy8e
zmy;`&%7?tYJ)6Pus1#-=6gp<21%=>*S2WDohO#wNh>{dZ(ggA@)Q}{OLIC+<TO@mb
zOmGnyvjtosSEPd6Pz4dCP#AUXM{mwKDW=<h&t5G@$qRykp6<dT>GlY`uq%pTN;U{-
zO&(#4VkjOK=GsT~u&|tapdiR;^Z@vIL<XB(91-2&^N3L_!2dJGm~8g6;?<S4K<ziO
zG-z7!RBOMRD}mB<6tpU&dskIB=;C%8eAoj`Cm1b;*p*5eP1Go8rTstgy6UvR7Gvsv
z9JKu6zas-$;+w!yIfWqwURWNlv~(kC;mGmkj-4-zxw$r|jFp;VD!X;W4xL{{Gah7Z
zFKb4#qU>b~hUV=S%HQZ1Wp)G_Jpf#e2hpZH9<HKLrpM7%o(qXyn%}p6+lUr&@x>o)
zJZOdM=sd_TlBYIYcQfh1$A?QXwVm33NE$WyV&zuC){d>+!h0#pT?jkhESlx9?BsHP
zdp5#%Jqm5u8G)}zMbGy^aLMDDH(tcvvd22q8y2evn%{d%bH7(d&4RH7hjsL5tf)xU
z2cKtDPE{-PdvvW56)E=HCAwKBD+P$~$alpxKb4K+dmXg_w0GBf3S@J}725tTeZ1D1
zr<SLpeW1<=meL5eDn6_Tt$(Q&0&X(OjXo_NjAbACvOVPw)o&$I3uc!AYcmu9H<#fu
z1{0ScEj}ohAlWqpxBR#>ItdIvI}G~%<J<oLovxCX@sB(c1UNW2F_+OH1rwKGn?WbH
zlEO1r2MzQNhlJeTpZ)=VHY5|50c$f9m(U?B6qg>#Gra=Mzqj4WGsXl2v8x&(mjP=t
z6ahJxL7yHI0XLTs)io!V_s=t*3z6+0mpXg^AZk5Gmtp)f6ag@oac>I~x5&^lQVI#u
zw$1w=B>tvQmtp)f6ag`pac>J0mrUC;Py%Y{m!8`*d4GBT*XQ3}1PK1y!-gfSHhQOL
zQaf+fv?iC@q}8p5?<9@6rKgxIaoUD32JYUg3|80w6^t(T(mrjH-^4oJ<W)nF>r9Kz
z+SQq+MN@aP+K{xch*et`T>5rVsO7J;$N8Oq{CuI6rCq(&<H}j?vKNi=X1n-w$KVcJ
zmAvdE?SIbLK_|tEsEf2exYEtf^KN+#-8Ha+rz^3JN6p$QjmlQ7>!4&-<<U6r|6mRD
z64aG;Y86I}x+=RSDatc(t-QYn{I>9$G?C>^JO8c)jgqTSd#Oy_m%f99ak?jC<f>d4
zXUi(r&i}Muy@=$08*04he}g^24yC@U(VE}13V&)2{U_8B`G4~ukJ-3KHL1{x%helP
z&CmEqLOuAM`lMpb59ZarTF`DZmiwUrIY2h+NiP@!2Cs&RwqTElF5yW8Mey|p$fH2&
zee~_4xJV>_fY(Kr-64Pd9RB<GeEILr{8p<~3cLLK&zF3|Xu0a3mZ+2K#O1Hwmp@+q
zK7X#tQ$&yn+=NTM)`6Q83GvM05)_H3$+>iMVg%fxq6ci|qy5exf^Jn3hlcNbt+?E~
zWV*D>>&&Uw74PVxK_zu0fOf*HQd`1I&+G?+r3V#3^XIznS79fDq_1qeS0{?wnfyy3
zsN&LewYni_FaTT=gjSb=`4Pi`<_^LZ)qn77Xbn?CuS&F2tbp$y{Y-#9e3k|o6^F?>
zJMSUy`oIJ>`JVKy1HCX9+BP$&iN6SvURijqgqhiBd4`1Vt;&*v7%k8MuI?oh5)8<q
ze}KUHlsuxC-8JnsoljP~wHQEn$p15$e}LW$gP@{$E)mYlY3lG<517poIg?tjbZHjO
z?Gn_gBoQa`gM%<U+0@m@HbC!G1o~#{Q?(oSx?iR9Kk_Su`WN71VWe))_ux}%O#BeT
zdt*d|`P7Iv&sKC|Rgu!C0bj}T&}s%&lkhvQdItgvV^N#gBg!~i0D}ub(>3tkldYPU
z``$BlRhd3?WgJ;vg&h3`P-~FI&Dpk5Ylwb)LSbrNjLazM2l%FKi7%VKKs{)JtT0JM
zpl@L6JawX84`zVoD?HRnwD8P&mnBxP+$6C|F=g8()f>T4a(kD?-!p{|^P>C`JYad|
z-#AQm<(_+&THrJMf72}ko1krTAT&d9!;9w5FsAir%PJf1H(gWraqLl-35z3iiyJE&
z0x+>D3Upnqr-m!M_!(U5lpCc~7vZ`eJ-{^v7j*J|2@#ZBee_>Mgl;m@saH+ihNn%x
zHp($mVBQy!M>BK*=(=l7MH`6)JS&M*s*jygqs-!VGD{~lf5~xHlMr{5U51=lv|dir
z$5;d*i!fn=Pc!;UJ%hqfsHX<k87<y-rfAfSq3)fLX3t0P0GwECS11Ao3FIwx%vdIX
z%ZRuJqdKPR8@Z@eQA;K0;Tisl@L1GK9?w=WnBV%%4^em)4=U<4ek~gK)D{bU39@f`
z3p@~D`qty&e+%lgY*I6|l|ILl<0v@Z!M=_0QyFFl0~_;$o*r=^_B{Y&$vR`y{*Co2
zA!0fr<OWGR=bSYeu)_iO6kKz(+h8w(Yd^Xi^Y+Nqm@j9@69noQ`&7T}P!KdiJG|rx
zBWmy&?O0FTq127L6wa9X0b|(j6+=vo<vX)1*GKqLFhW!|A+(cMZPtyR+l*xm{mf>y
zuTQx~Ke1Uro7J@kep(4hd-4S?Sw~|0G<#JmiI)Q7Gl74zxRtSRq>FLejh^;Y$geR2
zXu>018pF520;U~fmR;#E#-`>z$oebVzYW^ekz<<<>}0Zgsl(B85LI}5%_VpY*^=j0
zO~?8C@V%~cZwKR)|9F>3BS)aK(*4T3{j85$4G=9yEm$3PTr<rmr}PJbnmd82bv9ie
zG!15-x|o*&<TEdSr?j6a8+HD9JH&P1v4b9TI4WlRDe;mn>E8RRmg+{IS<13=u|WgO
zg}XC`2_^Qdch|p1V|X#sS;+S<cbw_KgjLWn->D9|(Z@Qd(8t>4{X2bP+^yN2yonxO
zdw|$!x6VIF@x_A<p8ZP=vbsAYMMfO;rMhSC-c=$Ub#YlPAstJPxo*0$hbMQMkMUn+
zvz=8OFKebaD79M!Zy>ao@91(o$3p2==`VX%H~N>~<TDln6SRGo_~bKN1i*|^I+tPm
zGZdF$=sXj*e&sWs1q<FDW26oH7ivjGB9{ScGZX<amoa+<B$r_8GYprZ+d(OR+in|2
z5`FhqG~k!)1ADrw`nm{$jpOyMgEiK$6YRy%11(KxIuwT>hm!pFd#cHy$WXGVp}6pa
zt(sv?_nGRtoa&yGtXD>*WTVREZ&H?$2u|53cgQL>K?GKui{O-_EP_`qS`m!$mPJS^
zxcn?7_9}WWcY=yRo?@e{DFd;8h2)$_1Syq!tn6xH#>DHHBd}P41ZB%kesx({=bg+D
zl=W6%a1^LAJp+A;QKYQMI04{{A~NM!tBA|Ujud&8XI)Sf%VY9uEN|nYqGTC!U`k5m
zH5^0AGO{PwgJm`!727h3H;}xH?44p>-SdzN79SM5g4Dy7F321{EA}~mVh&t#NJTFx
zjwGW*j3QG)!oBR|*+9w}UnGY8Sw;z13EUJ3OW~uuC5-K&7psM+T;?EiB&-jWC$jRC
z%g9lA7i3<T#0OXAwaTYr@kq+&Bx@9d3Kr0^Aqa4Qj71E@<u`^DWrk!_;4J78)LhCc
zTio(Q5FCUc<0t3`Q^wDKkctl^uR<(1V1yt>wGqY&Da(5SLdYz3!jRFFQL<HWW$jFm
zFldI`Mxye7NHSbDP%8@{fTK(D(%Hc+iy<(23|XMyl_IQ$DP(;Ol_#KVF;52mL~!;D
zKThj$ftpPg2@^8b6cW&|m^E8mdQ&9)8bmK;tP3n4Gsva<kc|v~<;y6rtzFG3^CY>4
zV9L&yQ5+<MNJji*FA*Y!&0hRCOVozdw0JZoe)j$M)h}OWb@g^Qs;i%xX<JX*h0GZc
z=I^R^^`cqM&+El1!sXVR`eHnM(cFA6xC~hQnCT49&+(*av1tA9gN*$0Lt5pFch$fD
z^IwE+(Bf_p2AwW{CzG>#w4xUG#A<7Cm|IN+fRkqael|R>RhdR%-nOmhQ$?%GzuvTe
zd*2S*TG`c&S9m4N@k+9UNvk(_m5@-#@#+GuicFGM7szSwl(c$<>g~Kae_yvBs<$s+
zRln3X?Hci&79-lT7@0m6BfGH$;_D7$Z7Fod;^Z}B)gRV>BsN;|%Bjt|Wyu{IKiaH$
z->mU)vsOP&r%f4%{o2eQN|Awlwwbgjsp>^Dzo_SBJabn4y?R~!gbTdY8RkE4KOl-V
zBl%M_8W}}7cG_SpN=CDZXV0ql%a5%L_>b}Q_v**Z@aOHgnO5(sU*El!f8Tv-+u7pJ
zRdvyvFZAbs<zzgZ45uUA%tuu{t*&f!Ih-zslWNg^nbd>Pa(q!&&;K<axAow+<+O~p
z7+jCrPlMrnw7kR%4+`}0bTn9uFUOPNe9$(7pJn91i{YaF6W)8vEz%(|xudmm9v`BP
z91hTQF=%5B8Iye05k=r(O?Z%YD6uDS-5qbeFFibe)^(lVbKUi*k91v^Z(Wy*>#kVL
zT1cIc*Mc*hU^{{OIXsb$QranKMx|o!G-`-vuTkEg-rp~qi{+$VRL|GKz}`CUV{khk
zG~u1Gq&g{=toERe-X`r*k1)NyzSgrZ?Wbm1AYas1)vTT^<iCM2CRNYJm$PQxF8ZNm
zN?QDX4Lnc_iC}NE0rLQZmB{Fa_hNiCZuEI`S$&)|Be8tR6T|XIPn@1Lr7fI^bPh-0
zgdyumGppwdJ!+cKq%L?%((4J)8P*;S+d*`Jk+AqUQ(jBrZ9mn;d&L1(3Arai#EwNR
z0{@3Gfe@8j7&?>CogN5BC<Dun#d0OzzXZyE>g9U5>$(0s=?B!pWLbJE6o@P&+i4%9
z+Y1Py{CJ!t?28H4h{w$?%<9T$Dyg7ae1WIlRL}o%S+~Oh9QgeA?NlK3!MO)m(w4%8
zQo`0b^8uPZT7+agKvO)<#?u$29h$mTp&nb5Jnt1HZ~GJ_A8(72H@(UW#e_s>(FwAD
zHt6P<qGm7DA7zK^DST|v4nHEhh?7CDtrzX0+I1{XhmfNkt*?659h%H^Dk5{An;g~J
z1helmsZVR;^Zh}4axVASw~qVtE%u^sG2KQ=pfw{4+zGA|d?$1iFLV?ybQCXi6fbr^
z#g6O6j_bvKT+jNXgOxoyqXSZ(O<5a%kW6QgmN_3)m_9ErX9J;4KQG_Bl`x!P*(*kT
z^{h=`BB3xOryUw^`5WXf_^Gk{v{=WB#S)y<+T@Q!OTs^I!7{Snkrn@busIzG){>Sj
z&kb6%0+#Jb(4r0;iI#HEKFi*x)shGpncF^*&%PP9pZ+;(UrP!34asQ>cydR7I<V{<
z1lbn`exxiy?2$_+$89;szQ95MNV@i<86B2QD3Rz-DMX#4KSh?{zH!IdX^OX4y%X!Y
z(_rmf*-+ms!4|!CP~^~SrxoF5xH7@whQ2YdY%CgUvgiqnoEivU!!>#slhbLi5*^{P
z(UmhF9iq5w1Y?)KaEIVdQrKdD&%D4cYqnv7wR72%VBL5@&bt&ix<e>%ysN=-2WWY?
z4RL%T?zh8u@7s@lEcZ)xpZn!Lj5nDF+SDzDuUPR8EpO2*?YA3uXu8sFxNq9t`^QbG
z*Kt!i=(s8EA2&JB-4it=Pqm+%_uH^^P#c#2Xr%m*Q|5j9Wc27h^wK<kg}Xh{cyb#3
z2jj^h?7_y{baQezyy@n0*SqDecgtPxmSevzj9J@6h*?y9D>U55wxDv(>tKAJEs@Ka
zzK;%GSxrXw++<(H7diycDV8rYlG9PRYz#URpNz!zJrW=tSiaN%+VUX^Yxzh6#5>#Y
zq=mI7K$y??v;pX);gairr$mT=xT<e<Fd~ZO(}#z%=C0O#rRpxNyGzwwnhv~DK6Q9l
zlY$3}<8Lg>*!?VHdu18Bby}Uf-JiVP{SgbTh13bTU%CMmlLOs=e$)UI2ctb&r4SG;
zrbi49=Qrc4a!4}#DAdxBHtbwa>h6%8(4%bNIY*$u{t4jq!$Zk`YqUENk(p)JiM1@8
zoG-@2#~<Aw(UpV;mmA5ceeYaCA!zy1%mq`JHla(;NYO52ohW?+*^Uhy&+rBz`g5Ji
z^%EkzC)?QvKCszVxJ`6X`xFn*lw@aO_<wHhKRQdL{VWxF5J~x->??KSe;l<}Q3_>l
zWOH<KWnpa!Wo~4bZ~QX`mxB2L6ql>?Grxa(`0eTEoyd#|-Z}b8AMM@U)9%}@W!qWB
zrDD%-JAxZ6xekgcJnSs{{s(*6J+?&aRsu`XKP%YxSJps&$$IMq^d_9McBD5cpiIM@
z>$s*zNpI$h5EeQxkk>$VL$b?}j92$T&Qdf27O*Wv<<e?yh_CW5vO9u+-JVj%rI~;5
zrbMIbcTlvRJF8oY(FY_h+fW8-|1VM=L57rz9&O+n23n##XUeO48(84B^G?tCwvz#x
z4OI`YiScy7mks%{r8{`L2^G>M+gaP&(b#NP*w+@TywhK8&|C{UaahGWX+FhaZRohg
zmfJaOL($%T*kKz=jl;Ii*kB_U#k7B;j_=>!10r2ZQOf!Kz2VNwi6*OC;Zkc{2n@sN
zEaB?xgvSGAQeed;_De=3!2MzG1U;<;5Fh5I_g*4qi0Cn0@@~#YSH_HmNrKW<xi0T{
zE@o5~Fmp}3B%!i0!pCC9n83aEZOjA&wez17GfHrey*7-QF~v)K*CL~Ip{0L#%6Cib
zjInJsFS$`%Iawj#5TYj0vDcn-**Qw43p|)^!?<lIF2rrS|DxN1w=P*chTQPEN-%pD
z<MskklrtRt)Jjp53mn%(6qneGsph^<Ke6I<`8sAX&We@NP?9W(;<^$x$|g}pkY657
zMv*4FPt;C(6OS=#n6!872>pMmK+?g(?I%<z{n16aFy_G%E*Pjo+|mO_1eY9ey!$3b
zO9spCNYY>8i{5eX{W_skkX-0Xg`zV<NgGzYn`2E4giz}^)|4@AYqMHCYn`=YA9@!*
z#XHCR=-kxmS-zkeF5%@3Uo&2ui->`pI<1D7wdOC^y``dOtVKpL8y0`PhT;ZAuc5e5
z^cw7O(OX^8vUgk(U)pd5yRGA1>A>L<6YZ@iTeWMGy|jbOc)28+PTG2<AX+Off|+xJ
zIA*v_eKRo~M-8(7E{<nfSu~Ap%cyZq^&+;6-xv@dSi}Yac#UQMvIyo<<Bw4%WRR8x
zWrg5K;@#EXuP;A6|Bipu#$Nv+RP>!Fa}qxHmTn<LfcicAn(qV0Qv)7&2flHl6_k>I
zP!?9eGus&`PHGzl(60pTU}~e(U~%krJrL^#15n|CC{}3M@9Pj)xazlvGF3L{j;Q3d
zbc^+672VP`yf3I&^(WiCu-#+Us&&+tBZ=VrV5v{=0cX1ztxA8CPEVzy*K7g7=aF5o
zQ`So75q>A2LwPGuahTG;H5CpI@HWlrZdK+28OU8gpj9XoQ2A`(C-l(B3P_)haO}c5
z6j@jkE-JHjlZN~6&u@9%D^@)WJ`<h2RGJ$UvOPKwj#CoRil-SI?dC>(8NFba_$x_6
z=OUAKnTxK50JeX7WCM1RT>D<I$JNS^l3UcAWo5KUrB{MRBchzFN{F@Sq!M(exso#n
zuvDTi>|f+{B6F&#h?+h%l2r)n4$5d)jN3WTkzJfN;F}VqjFs|ba}jX4NAv}Y7)}^@
z?|JoT1W+(`qMp00RFwKuo2qM350Q^Og=keZdUB*EYAk<Dfz)StXf&~R_Ks4bKu~^3
zb4DZ1nnN1JHPw`cf;{Yv(j|aydjqBU`1npux~KDdcF43&w|k14g;j=2CEHbr3xA#Z
z1?WtbvqBwNWiwirCp-~5-em(ur!sN4H<VA-J2QVUCTsQ>(G~o>hHU726w~uyvfhrF
zcBuHoe|LY7uFRrB;l~fAQ$uMZotr_hlzCR9G3bK!a@xeqq`^VLnM;FxN=s?*FE&QY
zt;&ATtCZw@W-h1UjY{bE%(A0hRNn;>MjT3-K0qVPo%gQ70mKiYPYpK*8+QrS%Z5D3
z2~WxPci9_E?zvs>L*}khTE+Q-hsoYQ;<VHbQdWQVDtoLeR_luL$H(TGXcxHfT|3W2
z%jco-&zNUwA<;|94UcvW#SM;j4aJ3{T`N&Di(hNj6SGg%iFjbUeD<k)$CWo5PM#Wy
z3!|rY{~vbtv`(xte7aDqjiyiQcrN3o20Kn5ZZ@7jwH&P*K((Bmo}<znT1&g7JC7+Y
zx$}P*TMb^V(N#M$)BC1!XJ={(l96@hW@_{A(au*(rACQKnM0~<e%R&wn01OwLWQZR
zQ#gcE4f%{Zg=00h-o2wv^*h9G{l*0{dh47_-tIC!DJ@pO910B;o2OakI2_2H<?Ez2
zva`2*rUrZ!k4Ohbn<<a8K~30B`VE7+p}2n#)D8B>Yx%HJ#~QUf)$Pqzpz6X2sqDB`
za-|BWx3=C+OfcY;MAE9SU3H$jfuogxKn-@b4v6Y+4g3bb#gW?a(W}?fjntW68i;)U
zEBW&NgBcE8qjmm&{54MG#BBMJrJ=Y%C*Dw8=)@cBrCvEsc2>Q7i4lDEML_!roBsp{
zFvB^QA!`~F0ysIBLG~;Zm*}V{D3^dwEd+n$HVnScukas;R@7U-AZTWi3DDaXxfMMW
zL6aUfMR$AX-!Ju`UiQS^WV$(cJeDbnq{xqyWIjDle7fHq9(JEjQcadQ4OUDKPZQt^
zj=#XM(dfwnFe71lc$t3Pb1n>C2fP&LQsQOtzr(w}zLMaO0+0gm4Z!u4Rj^Mq(&c}<
z2(;Men0Sa2KAY<+#es{^`vxzM*YWzdho5&m3;OA#-Q<||*2pC(ja3Z!laygJ4V)_+
zQHB3FqDHxNcvkcdC00RVjbkd*Ngqp#O)bimme3nzS17Oc#I<{C6NQ%P3KTd-&|=NT
zMr&qNv9WAM7tZZL9z?4{Xtk}B$E<(RO`<1R=t&vkm;c|DH)spxuc2_uT4nyEjm=~5
ze80_Br1n4ug=c6nMUuw0Rkl{U^R{i?F4`8wIU$voVN{<fN<ZYE^g3i|dG32k{hPj7
z4`n+Y?6L%fMmH7>{9U4K^b*m!#n#=tkv(1XkeNp)$&54P5mDxL1}sGz>70L_T1{ls
zexM#S$-Q3DBvW);imhD}Nid6>D&Ocs!Rt;WVK@73^eRLMk&u9S6Tl$c?uVzlhuzn?
z3AtoWn(6Un7hWg{Ogb?MeCze}?T_hc_j%-+(S|dQUDA){)3@hoX!#{Wr{@^f++h56
zVGSH(T}?&+<F<sgx=jkwWCnjT2pC|_!6HY+5rp-~5$zfQRb;SaD#%AaQS7b-#V#DR
zWYW8lBlPlAYRin4GmUO?u!c}PtN=!a;ShF)!~3KcQZXULM(KrgOgTHVUYi_v1EXs0
zgtAQ@wuVU1njf!&I-Jx%JJp}4M%U6#0yzra0Yni@G`aPd3`MyTZqt9YUHB;6Ir#+z
zpG`rVW=8PvFQ#MFS#&vE*3z4{l6mDy331B6Eb`cQDPd&5qUgB0mU>T6Vkwolbr4Dk
zej5O}4FHtAYOMs($}pwm`$~{BKZ`|&n<w0+ugW$|ExN{~rm>+bNxmCS6^5MjeuPu+
znGxoPz$v;H4rwn4rvrb-3&82XaRp9CjXoMqw@FWMI$DpM&eVAck`u&Gpptg{_%6)U
z9K%f%P?-VMy34Kkn{tY2xMcbsI9|Zt297KKHfr?I{<cke^0(1?G6P))>T9X`T|TEU
zk!fA>wD4)|8`r`nuW5D4)1h5a9v3-}L(1DFhj$SZ^E3s`#qxjk1;Rjw4FGk?Fz7j6
zA`E(tYhf^G^x<K!N%|57J+;W*@gv+n%%9a|-XLin-P~j=oa$Cjba|2JmR-K}_vrp_
za(TI_M18$@!!enMn#=MkxD;-$?F@;$g1)V*=;AuQtnk;0WGtqb78Tvbg0QaUO}{E{
zTQ<l-l)kzQY5ITGm9|s@{o(@`+ERPwxxOGo4ID23Q3J;nh#EEeXo%V-JweoHy;ZM7
zngxHn0Ayi^;?lG1qs5(NxDaMY?io1uCd^o@T{bTJwx9`?Cd~R;sk95ux|r_p#m(?#
zVK;9!OC9p~^`4}4<f7|tERLG=ig7v9OQ1&yYev2F;Z=WY2%`{5<jfjx#5n;D5krqa
zXOKCL`y+r(7{|a~atS<eya3P*99MvD)aavuZkzN3bffiFoc!F9r2t*gL?7M-djhSC
z-deN>M{+!aF9lqFPD}D)LEl5Prh})uwA_h-FKUq%0Tu6|z=YcYRKG`V&4Bw!D8n$+
z>!SV)T)%&Tv(!xfAAE&kvL4-?<$N~5c&3ElW(mXrOF{#(n?;%V>+7p9qx)25k7rx`
z<n||oDaVA<s4XX()KfRxn2NH1*<9U107)liC!Bpfdz0bpSgU<mw~8qfo*I4J>-j1M
zpUS6MfvXu$!XgJbatUXrO-6jxy35<n8|*e4&t`uJD|`5d1ZM<*Vl&V7G?b((<B{EN
z7B|4F2Y9Fei1~1KzUDCd#bo$(LYKauqF<w{V-nEQzf)`uXyikh!(^$ah{3y0e?LF{
z^!Ph4r2i+-U@FC^2h1HW>Ct+Ows3a$B|SS}&Kct;z_L;rH93j-94oHUu{1nRJB2F`
zklKGWtzSYbMw-J|YdXgSZUZ3=P@9Sjfn#Z<@xHW;Bvl<E>QuxCPk~GB8y>Mg^$L&S
zYM+S|d;K}>o~GUBY)&t%3M9Bv)Pa*;vM4YGRzho4YBH#(=RgMW7AJbSvDLIRif|~c
zB<kjvMX-}o8LMbZEse(1guI9YHPL-+KVyG^K~rq01qD{>Sn7l=3J6{a*{;d9_5b5b
z?)Qwn?wfn!a|J>&r49a(o?&bRUa>SogD)XmCMW7<Tc#ZI+?1F~XZ>kJN&{_d4a{)p
z&r>5>(($S)`FXQp0GQT#gJH&6ry2x9!mDDAGk{MT;IOVP0lzm%DQg-?%}*QPPg#H4
zKx*puUP#TOH;_uwmGK%#P4CVn)kYITy%Qg^uhmsZDw$7SR^@SSYDE@KO~WTv&ij4Z
zc8+yhl~TZ-<k$P*>`{*vkF{s_Aqvf7_==)TV>yDAu}Ce0<bqM;Np2=Kt~e#M*_l?H
zGVI=ITGu*iiHjOR!WFuj+jCLuxD0;?D0TQw1_Q!#kxRCvH*Kk}6>~_@t0u?;e65pd
z^-)TqB%o;(sb-+_Y|R(f3{u(v-@~xoQLxNY1|}~7O1P-j9zgj#iF0ap0;xVKbEI_h
z^-iMCi;`MXE!db^&eewJ=1z64x8zVB;<jMpP@0M~`nap{%HZLO>x>}rV=8|mhoMtT
znUTEBnd#b188biIj~}Wf%E3K_Sx|^*p~h)L2YHMN_cc5!W!@SZMm@H&%nRZCxtdk`
z9z#oUTkSwCce61f<`%l_P9x2`!%Ja(C9EDMdP8_Ec}q$;6e|mXC)NWG4DwR8=aoGB
z%%>%0v6|<!Z`Xvo?^Y=rS2Y{6ZH0vjpyb-6{9ki<%Kv}G-?G@u_x}KqdDEA1s3j8v
zI5{<!@nr-Pms(C4DVL80G@O6j+rNMM>w_jj-`stiqVh|a-4>N|0EPG~m4MVhqQ)Ei
z;F=!ShFC|_h^bHM-HUwB+}T^E;!H-f_iBbJBPa1`*W;s+L;)=Pgs(I}qa;bmyYieQ
zmAS|q;I~gR3p~uc3Lq9pkTc_iJK^+iO-RqrlA`<uu|lX6Eeu~UrL}(`lzjxVJTWDb
zgM@p2DsY9i^wyG0y{)iFr;o&w<GBf{LUb|%#(5*Ui0_0ak0tCovdnAQ!hb)vVqWte
z2{(QlO>%`cmcI2pK__l~kz~@0l_r2wbDUWdJnbvIf9w2H;m*cP!wX3m3YCeQ@}|*}
zMttKh+k?W_E>2oGpVoiTYF@ORen%_We)G$RMq_je4RWM0$vLf=d46B`FY+yW1*CuQ
z+*u@JwN5Jw%ZRmTjg{5_WxoqgMOTYd?P5;$jph_>wx*M0v+ye0ShUi=5C_B^TM$%8
z)>4{xD?66w+=&8B-$qg|wUuaFNQ!CRoiR4MCr-<3Q>~|yhNORr^<&L1K*jS+{M^gO
z=xv>JMuq0Vi>(^>VrgHiL>y}@@Ayq<aOH_@8N9HMCQ-6QdU_uB9%v2La$i?Y_ym`y
zdog+__auC!+-)dOMbq>XX?Eg1N}dv(Wt3J7uhR=y=>-g)pZAaV+C7OKAd9b-FgI^p
ze1mKHy1C0{Q6YaEdLEbKK<8<5#NM_%e#@GjDP_6dL<}uZ#1*X}Bur~mJ;uL*q<kym
zmd0@(H#yTrMR1$Zwcuu6IX)E{#ER_88_j_^vy#{9dz;T={;qpTp25-BbU6lD!ja^=
zdCx6c;$BMJjgPz|Xwhc$8BIS)CeK7cM_U&U8?~$4L1KS;#hMpl>(0)5j95!cJ(r@&
zZOZ~V`#{b<qa6&@yxR<0#yVyEGWnFc88(|2?YE(wMdPvqG$yV@jV5)BXMs#W?4?5)
zqN>w8h<0>=<eD+Pej;21ohk6rVKEd`LuigNuO7DyoD!d9zC?GHz7^_W#fFcLqi_m2
zV3dF@)5w29D;!C+QP!z&!TZvZt1(&Q-ApOVT=gr@mvk<j+dSgEIm+Ihu+dlN(3#zx
z6798%9#e6Uds(rYrtyOz?P@!I2X|T9JLpoB^r^`s_;8}};Nv=VJ1VNIo)wrlu1>V>
z&N5n|hZtvhk>q5TEdzZx#t5Lmy4TX-v0Sz*>W_aV6gC1xk=uFL<oU%D6mLLtDHR*=
z$NK)nTM*tLtODv>N}LJi)T4BIKz2%)R+5ox5XGC;YcQ>i&3avv%BJtz3Y;JVv9!|O
z(k`Yzv)lzrOD%LM3ze-crLfYPxr`CX!LG*%ul2SjKK<eS6D*kO`Eb!c_0U`X>9Bd@
zgTjB*a>R0C-)1(-n$_9|_I-#gE=ouLS{fFwtXok&mdCExg4C-ra+IIg+8he<iFG<h
z+^o|;_TAuiwaL0YZ9;lTYguy&rXh8Q(6@lHWw|(7*4*LXmfEe3a4fiYqeoU&C_l7*
z<}rKYdnw=sK1<`q`WWq`72qd2?@H5VHTr)`YIdSq+uC>5`>?3g+hN`FgQP6c>_v7;
zW0|shqpG%7_cSf4#jjS(T42I<=^*94<Jt0VvF4d;{w{8ouGRD0nugWpU@do>m#s2t
zI1hbUSBiDMb1&w#LX^g4>F+sqSM6qfDmrPvJ)JvAkDC!LJIM|@6;Z*;0z(+Lr!Ie$
z>ykPuZjRkpP{x{Zty4FqF-3CYtnYNC)<ksY{*^RX3u>8HG_Wh>6wARPJL<Q=X?iVt
zT{jBCFZE5>=*1L_P+`tWH-eWdRyA|%io5+23du?;N|aJ?HOx0Mw1Sz>7Ul*YqzZGV
zBhRfSRyY}p;sWDEXRUSAw-P&mh^c?P?e?O&m_V|s9i4gKIh;v=Qz-6$HtdHxvEx=m
zBFOt<#$MV^r!Dut$?DHey3ms=kL3|3Jx&CGyXe^p`<5p<X9)?fqAgl<-2=-jH0g=1
zbWJLP=vfY$LQs^9QaAI~I<BP`?zrQ{yXIO4KaK76aNS`K*NSAj!Acz!9cX`OJ-(d4
z2u-L-K*idXXS3)vvhN^2Ks11y?{?3J?5dOO4!X2NyRcZMhKm}RniF<T1z@x9=Fl^n
zxy6ikLZTERrM4dEk(&WioLUCkAiCW*PVD&9HIFR(^uuSO&~;(UNiSOh`ReTPVAZ@S
zuua3RSd0U&+VR)eq^9E*-wuC0L|2KKaO)~HN`ttq&5GU63U*~Y`)z6A#`d<SHIF&h
z9{NB6_vS$f9c-8_4#K_zJ=)%*Cc8Lx*x2x$+IKCJ&>Tk&mpg50XM<X7t3!WOaYK-t
zg;~lc1d;Ww^3JW^M47QlUFsXD_@OrJO*O}oDBK1~6{;@ez%{srAgg}^sXw~!ShPM$
zoh2mL1-N%Ry^e6)l`It`7XNf)n4L}t>yWg$+s0$6Xphs?)cNsMdxT3iCp`&FV#BzE
zem-eQVCiU=6=9(eL9QoMlE}G~r>-qn-pdA3q$tSFiz)>0c2FIfdvX|4<{FW23Ys)I
ziGvg0+%7FMK#I7U(oKI9bi@6TeD<^q<yYhuU9E%UdE2eYCpSc`-{iV$4qP*bO~uln
zv<b3xog??Vp;YXSCgP<Ag)G^B*gah@^Zl5Ozs5OdT$bFx#sSOrS$2{ZOOD+2`%f!T
zTL{YOdHVeA;olRM&L5^+KR;;YmguMdPI*`vhyNz5QfCOJ-mHK59Rc6`>&tI{`TVO<
zIPE_PR8J|<F-x4j(W664wLVN=>3LR2)`B#sMPphtVU=T$%upxRfzQ(eU`R2}M=|;}
zjV=lzU|!xPJEL3-36K{MLrg=$*hXo+uX3<bRhP&_S2EI5!a{S?BhKep=`nreGlSx+
ze@qX*O%ESQs(yc-(Z`|Biq`pbPboxzEvXs>CS6<+lI0;gV~Gd`SSm|Et=Jir{5H^Z
zSqZT4w^0Rc@UOaB<(WqW-#mo47y}Pg(uVOJISjnwNDBg3{TWOsRY19*iW-t`k#PI_
z^EaCJ1!sK_-v!TOY8U|~-zn)=+9y=`!skxPPS6A_6WM=!$ANrSS%bSx(zg2fxY)-@
z&{CLsY+QZ3ED2+j_x^$;#vYiE00!1Ln<|2xNXje_G*>O0B8+!i4x=*G1AvkqaGk|+
zbWAhGfMHhGi`wwE-2t^c0=SsCZ`{|}sit>Il6cYy*^MV0nc5wq{d+FMf#8qag>*3@
zj9u+?=$n7jIjOUoWz;N>Yx|?e(plK-%EFu)>V8~z==g3~Ra4+CvuY+%%<7+04@PnC
zQ3SptSHNM?Vt80u^}dKard7`EuJzof=i&ykfY`ig22tc-<AFxbG2p1Yg>b1xw|e^)
z8Q6LgujQ|$H(<Albg*4#!O5Z#_(-csDM9NO+24P#PvVH9c53MGNc0=^t*!==ntds|
zLn+&>n`p!h-#u(A@@}jT<I0Yh!yTbxx+Typ$c!rtUgeqk<lsi1)=9o*iv4j(zjH-;
zWB&`f%4I9>HVcLl@U3nVje)uIp*L~F*N(t^@Bivw&mxCyp3}7*8#3y?6>}G}-|t^B
z0(pPqmG2??!z%zX8nmkr-`%<yySS%~DB0~^?L~0P7`tJmKYaT=pY*`E0{1pkRpeZM
zHG}kR=i1#M-Oao4Ix1^dYgX%<*B4Q?wWfCV=dMcP-#wh%B^_gI)vdnm^(-si?W*ni
z`AF97yF8~^9j9(yirGH|KC0)8nH6aBXFY%SXFd1FuIJoOsppJ0%9JYVLP5vS-(As(
z<#tOt_Z=l2{7^|}QrF%4izOXH|BRAuz;&^t8)p5al1^SWxbkC;{=7ts6f~^vLaq|b
ztHoOU$;Dcdpj}pb_<t|f(3fD;8P`28)`V*A7He=`tcj?0p;%kAf<^vJSC&!z&tF_y
z$yR0dObzn;DAVpB^|&>Br34FB=>@LbD`<+ofB4zwPCu!$eRBU`yJNC-z#HZ7yW;r!
z8j<UJ$??u*+J45|U-;hbu%G|3OWtp&+1tU1KdZe^|NjBYikLT-Vf-@`0yj060oex=
zml+Z?#R2q}-x4%af1PJ?q-Sd^(_B_P$Ce@LG)Z~OQko<kYI-g^zD-iuT6b?2t$ux4
z&w_J)uf4_AuXRKjS6Tbrc}gRh7jm2@zaF&GEV!{zo>o$fD2hRW=mJ+!jjWUGpH@@I
zW~-%&D!9_ReLqbBwM3++EY)kQQDwBbXKhP6TuhNh%hNwFe{aQ^<L0L@Z+p(1g<`z!
z$H|^pS*a_+Xsmbd?A$jIQw_JVFxLaM?ChL}_KU5H`Qg*ks12hXbJ{GQzd!wJ0jKtH
zNuKSAGZnm9{=1}NCcxao!kHDqS@;eL{w@E8itqmU^~;|=|0)<X`%kRO7tN#*!7tzQ
zBWy2cm0!M<e@DZFkU}ql1#qx>F^+KoR))>+`SbFmrDH-F-wMCQ!oalGQnr>#Fy&pW
zhMkk5GJNc=7UEdnXdw#J21v81GQ_9YOKBS)p*^;WkMTpFu`0Cs`||W<dHSX3(^%<Z
z70+|W!NR|pJ_XalL69Rl#Ucfs$3PrIodnp2W}}qOf43G&X1&!4X6RuL3;0z>t8>N@
zGH)z`zOV-@!tb$sLl47Tp`@CpVJ1L3#jHujDT;QphTHGY-%Go%(CQ=k&iy<KsX!lg
zd4}$ojy)x0vEkH8<y@D<yR=6`+Rbqv;V6g!=)UT@fZ={wh%k9p&uv=2oK}T^*}|4P
zRs{+*e^CMK7w3Fc!iJy38Q8!ajEJ6{#<Ap#)Mb2XF|U@m;=Vjaq|_>udIn3Plj0xp
zhs3!GHB<*RYCo^VorbP=U0ZCz5OwHCRp&_cAY%p``Xbb;L<il?d*}d;Kykl&CY|iH
zD2N0m>nP4mml%yB4iHDIMoi&_>c61@-J;}bwTo?()qg%CQ3)!0o1huoY#TI_&fDN0
zOJA*osAm$RPS65~s}KX~?lmPUH<Lq5NFsZ;R9#GV7m5iWWM>5zwMNd6V!3O6Vq0aq
zNT9ShM6y+;d1=J}XzmqR^bimwpir<%uJZE{1eAcAc~}hB;N}(OVFtCXiTNjxmPV{>
zByucyKYs<=w71ki(KSdX1~bnkadSx*d7DNj;OKjFT)9q~ag^5(BxsA{xYai3YILb_
zFTb&sul@OAL#{a;(ub&WOS1Z+wzeIo%4=+~e|{NcEMpnsD5S?Wlt9qcKxwMJ+G-G;
zRNMXVQmsy3sU%4sjY|Tu@xZ7d9w`G5AR@*9?0*P=&BmL*v_Pc+sR5lDS6bc5nAbv*
zaOYTDh0YDt-n?fX2-?tF3NGB!fY!9(hn{_P>z&I)iv(uxWwran5OlVevZr<ko>_&{
z2{Z27+h#9$3{UK7y~lH3TRv?gGzM{56HVMd(^2C1yjaT}b+(~_(8f&T%FaJcJsR2B
z4}aSi#2W{wdsw&Vuc9B19z{js%j5pJU_$6XVb3D7{;3H%12|ZW18$QsEj$C~xKGAJ
zqz%9pT%4*FBAldQ06K$EK_Cy)mPsq6X|%^PVH4^u4QoNi#A<0r#aiRWv2XxVgfjd@
zWK2{#3nt_ETUnU2j7trSv!_<m(myB*Q-3=0)E~mav{6hLb&`i+?Ur0prh1Cv(VAyF
z$;^ldGf>(l`vz_%m@)bxZkFr;{M9*bhP4UE#WYlMu5*{a&J6CRX-Q~D_<J0R=YdR+
zydo#?^F~L>a^&C>nTjGIrH<meaFL<OI`jExs8H52A>B&P2<Jje26hn6&|qEJsefyM
z4TGuS^N0a%G=$aq_I%lq{BV*$o#A4{z@}WRn(#gsBO&Sp7lYX|Css!XQjMAFT(W}d
zirP|NCxH$=&?8H2DYkHzrnxdI$x)P&Bk-^MDezyQx6#wN8Q&l}k>=WVPMX1;%9&QU
zq?DV^BA_ZyNJmAPfhOSU*6B|WFn_398hdLcj^LpRHttY7KX`n{8E%>4dSx-!MuTKo
zOI<vNh#GpoNdV9W6%YYuYQn!*!=v(wxDSFFdrx3e6)F`E7GIC$jI_svv8TzN2iR@T
z!ET9pGdxS`p>~7il={~^9X|VAKz@5ul0aF-09#=s_r#rRJf*-bmBUu2IDb46Lcz74
zrM&d&01OX?%7GD{4a?i9oP*MaX=zp|P#;5-**IFu!K=@wV10CDrFEH2d1^X&)Ho&A
zECAsc<Ui2y%s9b=LR?sBqL)1N=dqNyB}bg;;F`*uhZSsk>GdqD&e{qhR4~((+2QhJ
zY~<^+_0aw541E1~J-65{(SMnk<`sa7Fx~vX0ond3$oj(uYh~n$!_<?#nruHucIqwD
zqvp`e-Q#$WD`u@9<tz3O6(4`AV>y)qyyZtp#fBjOIP<ExAA)?eCy5>2%$uZ-yVDZa
zVU6wxd%zjfqwBJv0c&Rh*7}?x5ZbQdxm!LfL1!@DH`q#hO;`&_X@AxO`1?H3RmQxx
z+9&Td$iRVn`%(A5^|pdv-B#%Ch9Gyy^#&JR^?I#4_Dt}SPBKHdOUIY^kL6231!mQ1
z$WwHQ|8}}00bD5O$8me;#4&H}2-&pMd8V6Sw-=!hVWvI91GVp4?aH7nVmhCu|C?sB
z1@k_zV~MQLfs|y#_J4y{MuX>)e5In<C93rS7Fh%}=c@*G3DCx&_x+=TvTL$_J5BXK
zq0^Sf(^b)0&3q=r3}%{V?Y&3K<09LdvYTnYTCzZRNywgXjnMaEEx$j$(`?G-jq)H9
zlLfb@gUVx$1ky8O_1Ig-Rqf-^T?z($tz_oWE4?^Z09-nHl7Gcj#gx{O@BkWMSFeQc
z$W1e;MzFWiVRI#(OqRXzR@RBhw>agQ2@d@0%*z8tuu^n!YLm;2_weR)so%Zbq2HOf
zBZ6a6nsXAZ*yZXvsoOm;@3VpJY5THhU9bZmMR2vHD23vU-y`!VW9HG0THLv&SV)&H
zl2tS)TDGUhT7Pp-4yy<6#&Pw=W#8_eswNvk?AnMnV6DcDqa0kd)dmX1%Cg!$R@6a@
zsXIrkn-@YheZO!4HXwCd`MfZu!OWg#NJ+9YS80-Xxi1Pha(e_E!B8)f-gTR|%7URM
zm)a9&`6_6-iFW$#=UI+FO(&fr7mM%3d{U1blh(Hh^nVRzHKxVfO)y7JplUjgHt%bD
zod}no8h~{GZw)1uNt2e45X>`BFxaJnJ2>Tw?q9K%MDk`jSg}eb=l4^0klu_b^xQ+$
zj4{*xHZ@aQC%Z7_DAJkEM)S<J?(17UVs&HAfT$a`SC>`WZseww3T9M0J?!o#<y!l@
zWNI1Fp?~@A3dk#=7_ZbrRRPH5Ai{bjg$Hjwt2>Tn8+Q(TVBG27_5Qa8n5I>|cwj8_
zEp5AaU_6t-k^|~m&|d+^(*6#3ZQO2r;{c=xW%&QQ1@2gILEGUL__XwodJEhz;oJ|o
ze;`y6;^h4Utc{VM{z;ENd*6TzIlT*BC-=tt27l(E`vx!-;0@vAeFGt6N?T42LX>sH
z$tQ56egMk_+fi~Hx1Rl$)4(>qrWl>vF3@H;nlPKUwWNyoLfklJmu=K|lbzi6wjEf~
zSh~u=T1PJd+VLzpzaI{Pw$?>yK*9h>23fh;Z)L#M=+$aY)^4*08<^GZt>Zf2G<qd%
z4S&{lMaQdlIw)(qv&})nB^u^BLo~D_+|6rt?V%{iGtF+u<BlOYt_p+tERZd*SoRll
z(Ke6oYLLyjc~`?6-_>ZL{nTI6GOZfP>C>Sd89(9@?N1AdQuDQ>Ic?kqJ@Z4T&HJp5
zT*;iI8YI%Z_uT@Jh4>#e12ZcMWo~41beCuRGY$bcm*K!A6PG&|E-05~9yFAH7oGlm
z|MeiVp;CAgeumrf>FxdS?_p^+ve~R)`s>Fbd_WCCJ2ffeg!eH0{CoO-_-!e+jfw+~
ziZq6(iwZ~%6;8%@p~I&X6;Q!nDUmHl_Ket5<R}q)jmeWLL9u}xYC}$FR@+D*o=xG-
z@xeE~JYSdmE_^e+Bt+r-WY7wK)V2shlS~NaP%uRXA;XE<2(+$F^t1Lj39}z63uS-M
z<==0+6etn|z9<OGAB2V=Day+1bwMTEfP%VM#YnR?1moyREsRvvw>mnn(`!`q#F8|B
zmJC1>;TPZ??80EHD{q69r0UaiUZM~n-Hu5inq}2rIa$<s=pA__4C)Sl3MT6?L-)vW
zQaY@Wh3wvQNT%#T0FKY^sdIZM(Mn+z?Fn&OTp};5scRC{1=t~_C>L{R!<>k!7{D2>
z_<Dll7MKW~aMFHCU;Y+=7B3a`EkH4hj^;bMHacu~B(p$dsFtY6jz#4|Z<G<v&8e|l
z*tV%zFp%&SN+rcg?$$_u_#C}Ues8wr7LZC5-a-dbJI^LN3><nCJhs|Hi4#M}``OoR
zDAJr3lJM)60j(0!(^oD-4i_2_pl8Zj7Kd|sj)5qY@5Yiy3b3}7q+WArDM=6t=>~n*
zFlU<+mtAFB)Ltm4XpiOE<#KIPl)`hA7Npp*6y=1X3h_|rW6?K%xnY3BReL0X6&2Ef
zM=4j282?<7Y9(3wH?H#KzS#gNEr9LU%(cE??nU-}Ngb?}uspvToGRW+UD@1JtTpip
zE4k3Lr>EF7-G{&}1(pIIQBzUDZJU8(LyjdiF{_2xc<?B>sB&puNLwt`>YVe$mO0Qh
z7Uhj<TvD@3WjxM*D;h@?t4wMtiF6;^GS6ky=MJc$ref=J9&xU3a%Ionm3-&#<!h4P
z`<`{on$NAjv&=58I_mB4T|mRlq<58tx)_5!&JjqOOQRj*N+$(nge>Euk5I?CQ<Cyh
zENx;N0ao>_if4`kojgSD8H2xM3V|SJ_NGcVdUvj%=TZQF%h`i|WivS}2xVz0A;qk6
z(58Op1aYr6&_DNVciE$uCmlDpOM8adTGqd=a$f%dRxTOm5d#H1@v`M!5fZY+*Rdt%
zT>U=QG^&mrrqbNZ!HOJjz@$LFoF+Z}43xZj`}>R(Wd09Zq(x-LEnc<$x#6O={y~k~
zxy=q`hA2gUrv%6~&TeFY=DTxyGoTxNcLwycM+m&LIl6{WDOVad>q(08t$LDHtK+)3
z9;@D_B5AQVYgkZq2w2IS<FG=>Iug5>j`yRsw(mFBxf^}=I`^|T8t-kkGrV?sDdj8H
z?ymK4TXm;roNk_;djWhqRn9BC8hQa#u@^9f-9F5JHQHL<|Ke(`vgF7lF75X-`>nPw
zV;)f~Py4O-&oNIM(h<+rc+GRb92bW)O`miFyUH78)t%7o2H(}sYQ32JfQ`#0VWAa<
zKlKLetnkW^P-bNPxD(%dc9)w1Kc72%{Ge^i@v!G|xU{{Td^G@4r8P}WtRAJa!5*j;
zrripEwaFaiP&rSK!)8}G6sT*n+ssd?%hR>v1l44?>`Gz0R4ILjUz!w#a1aVh1!Ksa
zG#zKh+;Izfr5bd5{bjN!$x&Z<Hq(<@Eo$4En-8|;Z9C|WOVTZF$I_=yK-BwDxMjuG
zwHh?MAC;QQEZ;8VuKmk^cm3Qa)E&FZi5mfb=5nN6UtM|*yI#?X7L?4W^;|@sPYkhR
zUr12-n}wu~k~4P0mJzjV@o91x+?*kK#QpA-na##5+IZ<pZHt0rV`QDkC`CruZ04tC
zYdU`~%}{=`qpVu;i!FAsczcVTM>GvpyE&~IU#c&8)7<&dF-BSFKx*@UpsuaR4B>l!
zS7fv`ShOWqi-u-?vMXD69-VF0G0P=eErmA3QhVdu-2?eGfy_I=dP1=Zi)+s}YCjjx
zH`3D#xz7?f%5Y-13gS)v*w55xLO}f>qC?;T2nY52EG%3kYRvK0?E!il$+VZYgJd&{
zWSe*M)HN67ZTPn;95+BDQ~Oa9vaYCq{csYvs|m(BU$6F-;4Cx0+Gs96Rorv?&}&Gg
zwd{DuYrko$aa=#}k+yak;z?(SZ8-f5zx4-f=FVEzlfsAC+-XQB^yaFq%d7!K!jk?!
zvqsr`Zh4zEV$1r7OtPQMBniw^@W_?!G(>iZHnyjD?rwxS7Ooyi0G)z_O_lH>nNE35
zD35nrZBS_SNGj+Q)aJZ4d3skBTi)N^5B~wU_&=ACZU_?sIWU)zqC*pxxFs~i0o<2;
zCNw91%RcwG;H(u68W$2Ja7mDeYOow`8ue*6qy&&kd!>1KPu*4_7jPlshre$Xf~?<)
zAmy0}iW=4_Morp1mRaC9UgLg7R!9~9bQ)>z^sFC-nRj<>hge1c%83bNxCx6C#=79}
z+eN9D-|-x`-~967iMVaN7H-Y$`OH+J!dDD`uN8wZ@P}5^H;K2FXdU^>rHt?Z7ll_V
zJW!(3?3HYoAe4d)*(m(bHUy2-0)B$Arto4l`b(1Xpf;sYgRtJLc9P`;)5$4SckVKH
zDAx`IM3^Dwm7z#;mP=hZ9Y{Vm`K6T<(yyQ;smWX?sRAasC_ma2$!@=HRXXiU)*d*2
z7pc?EucUz1G>oq_qrIt;UtwREB$zhp*qikGm9NfkVA$<iDGCEzAW3VZclEmN=~w$z
z{v_HNkBF0%Gx)8OwIsar>RYxm3iN@f>ZF9B8T0sJ?5Q5}jj74G8L+jXJ9mZ`HG8v|
zH>}WT2MsH%E?Zc1BQ^^H8OZkDb&8sQX|UUZiwYjm6ipcP{d|R2oJre(T#5>tJlb$F
zgk@|8NpqV>4Wot`-q)%QKq}GB_x022h%KOfMP>i|{pq&_3I4ETs_}qK40ieVl81#d
z_z%y-JMgNRWIVE`cmMqQ<xih~RRWj&D@~OP;4!Kgmha`!d0b|+TE3NMkKmMlvTji#
zqU%M2qlHu!C(yz+@OgOx?Q!?S?Kb)?jYa_hv}TKq5+?XGkv8Z<4Tl7=jnaPK_#k1`
zx5zX#G}2RoQrD(OTu-agWBTxCnu@FbzC3+do<3HtTIZZkk{-0N$-(&mu*Fp+KzXA$
zXWLK|(SQ%cmFQ6}_C^`KZ4iBbXmDVV!XUg<_^W|#G}8#9n}!e;L7*Y}p2v6iFwu%5
zO-QY<Q7twIXCaBGs@yFc?!P~Ouj{_zst@eDz<E~M2ta{&wlN9}N)xOc%<hDVbvn)F
zuKJpFxme4x3t0)ST@wJ@?&HGHDRez|!ahFC42-YsL(G_s7h{4&>10fQ*zx&9gSTjk
zd#r3<0LZAN2+Et;sCn2kYTlAjV;;#ubrIa#-D(tEV4b_)EqlV9sEzx@lv&tl23Ikz
zOp&3yA>0s8-G?(eM#@}|PtwXV*A8<Fh3Z7>43C2g_UGz$8m1gdb|+<?<$GthRk%ph
zc1#nr*k%?Slw<zCxROGD!R+$sw4F)jh-XGKt=hPXS305k6l$b#c~LD=y47>_BbC||
z{9m5cX(xg!GaQTY9CNIWb|s4I(}u*AH}W#N(vEXpM={wHfuTf{n0PzNHU>S9UvVsB
zaM|Dh(&c2LSj+S)origqdD|It(DmFa<vDnAWn5NSEfDrr^JtuZbc0rdJdE|xmkQ<7
zOjqT`N?xihA?n|lp;E4!Wk-^ZbgG+XlrTs3JRL49-*7{yMK<JGr0nfs7ll)ViQIrf
z`6I>G>eQ*tau$a$82J1yd24NHl<Q$d<regylJ9BiDc4-R>h_Ui>Mmf*F?A<X9MeCR
zaas)?pIP)s?ld5O1Prydsv=eyp=@k{JDu;9ofp(I1D?mQ#Nl4$1+!F<x37CtD`g_g
zSyMXgXzOWnOPpq<69)xT4p*VLE`2T2^#pCs@XDHOx7%*}q581LT#+j4&x~W2fwDoC
zIStJAbxo_XV*ThX%PKR=T%nEdc5FAtamaov+F9SqNFaEBtYUO493FJNabP1Yi)uez
zg`h={EiE&OCx$K0zEkNytBxZRD0#lK2s2Y3=T*g^w7mFA25h_7wi#BGH)mG7@|%Wn
zZQN)LQ~@`eEWB4J1&b*_h&#oqceg0FLq~`FSS@DN!^VvC{eDQgcTvJjPeGWWTJn|e
zd41zCTVXJN<B4XoN-NDtnG1`T@^MJBzazdKv>?yT@0g`-)zrL)=24++cB1UUYiqvR
zA=9yI(a+DS8fxb--wiWYQq>(Svn;mT0qY+=)n1w3J`)2dG7#t8Q|t~Z^3!SjqP%0x
znQ_i$q`h%gjY2&|@3Lho-45m0)n2zMaSa|n_ZvTd`)X4&+=|1dJxgk2hB8xEa!;PC
z?$V@^MO}M#Qbu8vTKPy7<|=HQva)M`S7IVR$3$Z{;jU~9QP-YdEpHXecUTXlJ+%y^
zdvt3jj_<`%hdYn^<dx84QSgUZ0w~P!{fg&ihyuo9otbvXwfaYG=T1zoyoj*Eg5pFG
zT~g0~^z~tS!r1Y*l>iyzMF}3$cqyIf#Yy3D+u#4Te0lmf1Fh;Wg-k^Xn=2Kv7^z|8
zm)d2_jzv0W?O?eT+A$1FP4GGH+{ilmxleW{pU)a1MzGG<8RCz2zDlN7a)_4gwf@Vf
zl|+-X7@Cj0m1=J-^D+hUqgGcy;w$}^**%PZ@bY@V=p2KX2@i$3px$qm!&S5>Rc2l>
z(fJk}v`jTfx57c~^pSLuu4(ema#-j*NiMA<cgjXM>on>G3Xid=O=erv*p6OP)3X^D
z5)9QG6wU1gq*Vj$5$(CFI$(tD<veIX`LxrP_TGyYjda4m8{$3cvTsII)i=PZ5t2uL
zbbGg*DYRbwM$<#5^0>Ow^hj-vPz&@1l!_0wJCMIonb_JL7}#j1QFK~kAqeGCnoSGa
zz|ih!qeHu+FYa9o+JFra4*x^#j>OMc+ej`c`P=#&7z?3lE0~%Wo1_1yeU6kxO8Zm#
z96EsT%0JlWpxLCj=hEkGnW!4V+Xs7p9}I=95>qGkjlB>1L%k2}@RXc?u=hc;=M3>W
zB=^<ls}t+4uPb=NEtu_yg;hm8pP8%rYQ9mhmFPd*-rgxKw3uePJ494y_egb-F?qMT
zXzy1SnVEjz=9xAcO`2swT>4aV5rlUltfU(yRvqw4TYZODWj*n#%ugPw!aY-eHM&$c
z>nN|_0=t*edSh)S%lzsH%e5+Y)C(!^H)@*QFP7|0J3D1Vl?vV|DX{rHmWENjEmNK(
zliWa6+}_12P^j<Ma^gTIFw&=wFJ<h>0mL9Qdg)0H?v(Bjb&UJ{yvn+PHeS~++45Fi
ztu@B8t#q+!TFy6Qj?zYjPCY_@m_Ytu7N{b+=`aHwjTxM~-KQ2*D@_jt0y^N~Qc;)_
z=U!OQ?^$wPOZ^X4A+^*gzTk!3xg1kN!J%=}f=A%*x7DJ}CDz)_>r(7mXf7N<9N@2M
zJgk%IpoG$DrmvOQn;IQ*$sm^?u7E^Vad+$bO=JDBRaL>Fez>`3WBGx9Lt$ICDdwn>
zrq#-tDShWy$_&b+3xsU3alBM$fmLkkY_h{judhrUsh#%eGLD<6{z1Jn$|QZ8fOd8b
zCb$!D+(tcYP5WdDS7d13Qm%z1lC@RI&)zMg17HqeR&+6H4hv&8UC9cM;MkhFq!V=E
zL;qpiEqGxA%)9RtOfjl|7oaz;bMruH49)zh>oA)$+NVPg5HN?sJbm8#$3#(M@)P|E
z>Ue{ZwdN0%#bV+aR@~XOUON~CozkWiltc5i@}@9^%<N(ZhQ#W$QB~Nvu6kgJqC5}~
zYek4KT1n2%>{k5L!W$~-fyGhq`eNWTRQg@BU%f9;-mJxLWjtzsw_<e<qhNP@$j*y&
zIdP&8B8#M<JG_T2q88e3LhaXGJ?oj3o7bFA<L#FH{Aj-ew||-@2-CAZeQkZVU1e1u
zoQ(5IOM%LHze?4Nm|f#%Of_b?K2`*od4qJVNnVe&3)5_{YDSD=zUm+chpgJ=vMM&`
z`X>~(`MO>6R(EuNb#nW3MZT;(IG*nCcS45U)>m71Kd=}=1x=|@4>}(3WFcVCVE^#p
zM|Sv+9lCm7KE|*f!D~U^#!hn5k3`WOA!%~29hzdj5)!w{(eg;Y{(m4KW#&Br8Fy%Z
zLm@Y_M#|3Oes)G{*&Qxh^^8)F@oy^R#?6{Ny1P(S1)1`H>nuQCy)E(t=~7j4J_GYW
z2lW9n0I4?zdUeA0&ZY2p)g7(1F29lloqgA)O;yPdbpu2R(rYa>hLumQHvy^l^wu7w
z17S4GJxWy3)C4)~QRZ6HPPec&QKA-Bh5vf1ZbajOPTjcRCp79N3=aEr<J_Orrkh)E
zA=MW9<)CIdAm7g8+87=I7Rx)e%3P#AoVEq%FmlwCgD4chw27IpzSqKbsr%w*(s%vA
zmOc?J(>H3?e*mv}9*dWeel!ySFqfeSHxmOgHaV9;X&)z-7c(@K0Wr6&Gc?Tse>L;;
zmUJ94NBTKH!k0w(_v1)BW#G(&;}EcKc});L+nLadsht;CFsE}gDH;kNemOpA4v&e8
z!f0lU*&))h!Xl&3MEaiCc)lAyNy^RbY+&ngTPm`3Hw%27Pm>W#GCO4-ZNxp)TSnZ^
z&Pi&(L8U~&Hd)sLjXN95fbV0Ce_L?H6vDX6cNn)2z!Ysea_j-dP3egdlRn&Q+>**0
zYi9g@qIF9ynB;q8T2^>lfJtkAly#$FdQyfh<MtI7%mRM^o2;R-2X>k&&7`bnHr26M
zTPtVZUC>^u@Rf6Io+A>ggIP3{Q_fP*P^GWz`)1U0;Z!+-I?qr^;SzpHfB7d&1f{sv
z=n<3K3Xq^)?_em(<Akk}&b7`Ztq!I%uNGvtZd(0ToV3qgN1%aWV6tM_RFat7Qc-(E
z5@u#mkUz_7T`g`xfg@tkQ%b+*vUXFPo0bbgGC;Q|q@=>-r2NEj@ZAEf72X8sV<pOH
z8q?@+AGXCovC7a;zf<8ve=&r!C!*iIn8AifG7VCaLv=rchHj@Q37z8WM9<*&Z>Ty&
zI%E3AP|=8>oI7Ogqe0Zh?B2pGmEuOyKUK98Yhxg5SA-GXs<5vJjTl?>d^97`HX;g^
zqQ$vjBfFBRoeO2|DdSRh47(hkLX<0bY+F{ES***fTh}?yY1#%Zf6<}iSaRtvuTzs!
z3Pq|XeUtL9g@pWJ36dsdla8|8|1DvdfuujuFQXL|!1NNe)n5G$@BaMq`Ik@s21d*N
z>!<QXkxGKI%eVNb4K0%bmap-dswn_Kz4+#WBv)i8*9D{rd0`p&v>X)WlORdO>*!ZM
z>L|-lLNtp*f7SSje|dvCRB-SR%V^&3Yde^+s#^q_3L5^&LrH7>BdyP?{G<Q0&v`1X
z`h7WkUJf4<S5a$iywzGE(PwJLDz(N{O8LM@XgOPk9COn5o)TA_Ar06XZRjn76b-EJ
zW+-8q;XuC%=t6TIY3R;F1S><xL-~6g-{8YQD~%K-MJj73e?PHIQ=LL83W>Xs!}a&4
zZ)x2ZTJ@p%j&Pp5gH@yuPtuA|ArY+5KHo@7()2vL$<~9l$y7(hU9x9vt!n^4+kIRb
zbRu0(m9USynE}9%q;{JbW1wu9ASu!3wlHBkC<PjjG9swCi#tke93y#|)-9c5DiKG6
z<4rD7h2&n<f6OP($!8L!A6Ff9f_)=Us4%~-OS_l}w*oPCcvC5<Yty<WB$*0G>%H4f
zPsQywrKhd|s?r<weolL<=)@BjY~!9!*Xb;4GR}6Qe0`5|c<KuNEKx<}hi(#>V+<{)
z8rbsbP)G|!38LZ2wLwQ#>4ktRVUT;AKjPRxsdHqNe`1kw>}%&ZwuMem#;imQO8x*)
zp-Cl$a_bxC>?gH>w+up(Q}r%=U>|?Jz6!};DToZK;_f!nc4<zLFvk6cojEJ0tg@Xj
zUXn0!wh>1=A=oSg+lk~{&9UJHpM?=SEd^3FsZ|9_4yHR$xn*uiT0-Umu&+IIo6Uth
zPe-#%f4NuN2|BkqSjh`zbv8|YXIdyFfArYCbopb+z#?DEJ5o@cjHl?48EYGp_>H-@
zCYu}*$2uvF#Jvihx-a&E)Kr&GuO1;C9n2;DTBNEmVJ%YCs1qaAA4@+iWs^@$I=SOu
zUvkZaC<puUe5CrMm~&SDK>Dc~c_7`izNS-$e*j@exq7Yg7N8k7W<;=Wje52Y*gePN
zx7cWkYU~T3+T6--?r1Dsb1YoJh+|Rc$UE7aT2SN2A(ejxH9co1TSiw6wZMIDN;>G)
zh=$X-q^eLSe3EHLsOQF&WA9p$$_PhUq{i(oSm#D4S5faLqiW(^kY=YRYhA3{cRhrP
ze`Yk?^^i7n*nj9%kJQB%2AF>6O%Dopphm!NJsn8bM|z-JhXd){t)IMRaFMk8Pl3{L
z)_~2UTmQbU8mWk`VdrsKyr$3th(1L;=r~Mi=A6FGh_V1FFz``1?rN96{d8PtF9mqS
z$-ZyMQ`j0Iu6V1eZb1f-7UO#(kp2Z)f4Y%1mFsu%tdh-r5V;=A^ASmIo|;%cBt2;A
zoHQcqVN2g48nioL8}tHO{PV~aSxBIyn4-@}&%S#ElClppIITWC=6u6@Fh1$Y*7*iX
z-_}~y0W;|CetApsVbZ8s-s?E{8z#a|vmanHIZO~wjJ$o#=6g<YgwGPSz5f&^e_x(!
z^9|q9&1GCF{q~%+fRik|N*;+fSMyH2hY~j%rzLf_HRtOKksVmC78=N#0!_sdsA=g2
z{|WL0#a>Hqqi+Z2P`p~Vm!@<md{+)<%v#lry|j|GOnf%Tgmhb7p`2;;eU80+yL!+A
zK2|CG35O*8PDw_kcDQ*ULYAhKf7^!A9lFhJeO>b4SYVQj#HtRUYpNjI!l<&}*yD|o
zpg7D+<?%x$(d|@iala(c=xr#u22rIhQr-Q`GP1_$Qegy6$ETLl3F>WM5$YD>Wh`?u
z$F%E9qh>e6hAQ1OF5(Dw_&2<$cue(^lsD_A+_*BUeCRwWu3&e~$D_YQf8&oHdO6-`
zz#-Mg8fWPj-gcQrY!jyAy593;T?C}?ifz5!w89>u^~#1O%CM8%xFal~%3-=hv|$kr
znWC~~*kP*<OZHBSK>ZAJ?>36@&DN$_)W%}ByXZBNaf>71Hmug;hv+(?b<axOXctkY
zWc2`EVAD*zT-8xQg!`K~EIokIb}>zZMqTSKmX^%=iTe})HhPBb+ik{g-}l((?AC5b
zyu2R%4_Bdp9hYJJGZdGh9xf%8SPu)Apl>@Uf5lp9Z`()`{hnXZz<ilqz|B6Y`dADC
zIQHa1f*5fYm|4RZ$TC+^qL6})Z1&&ptETDLmJ-X7vH@~5sbY0?b=9j^O=eHZsO*!%
zpOl?Z$;kg!S?4gYLD@iZhm?;+Bq$1H5EUCMgQS?e46@?Hk5cwV1^kRVUWJe`@C?QH
zf14qgtvuze$|Q4nr>x2HJYnQWX7f>5U+%#yOTIjiF^V4Q@t9>>w82j1sLG?9%S<E|
z%L9Z5ViYqVVN)vVys}~*ekJ8hmW2>2=gJfml=F#Xil`h%F@>#iu`GriMnRrsuUs<X
zABL7xOlD|FDaZ&~#3|5{FpBa7RQYa^e=x(_oG}My<$ci%rJk}ZC|Kn~D)$_=q&O&e
zMaIY+*a@;h<_N6Fd(nqWVprs%qF9`a{h`DObHMr7BzXd6u#MuOXcc>sIigd{PBh1$
zI2zFhhhgF34Qw~Z;<YFnTjUTOoQeerTC$Nj60~G4tMnLUc{T?XETSVDBe7s)f3`#z
zl|4$V0=eQ1$T5kpu;>tN`8BDKQn{B^G$eCmqoONu1NG5I@qKnGvKL$6*ceMJVnbsJ
z;sZ<+Q`uiQHcHIE76==-GaCy9wnS#evS8mM{EB-XW?%%<*piI!t3(q<P!%J~2oohM
zS_~8<mU|Y<$NFV-6|4A_J2z$He^`QE=8&jIu!&s2D2kkw_$~TuQi!M^Bm5A`uh?2B
zDO&K4ge^vhXJeuWBF^AvQzWwt**^O1x9a!|xf_w%Z>vB4{1@&=Ep7?egJwCIoIig2
z9Vx_#mh8J<(r4BdcHA%Nh&5vEhNP;F+DSV*oeqYz^74SHP6mrbJ!_Qfe=dLjvUqm7
z7%XaKyBkm2W`Q}LGFEM|UYX^oWV}h#T>utrh9AoXa>xp%br(oW38C&n#9V)Ybe1?R
z7l?l-xKf=22g!FASzyN$)sG)nC$n~VS})G3ljEn=yZX!GYW?s2JFT&fQH|v_i@I6N
z1qZpPcw5ig<!o5bJ7|<!f3NGy@!(1O<xHjz2+Nq=Ii8>4NeL<6P1k$Y(Y0rh#z={~
zuD>C0z3JX*i@3gDl7rWhTlY(PW*vDGNw`cCpm;wUe`~LAY{;Fhx)C1_-6{ae-L3Mp
z`BqhjP16=d)NkEcDR#1qQtY|`mBIP78RaomJ!xl`^{mWb&a3Cue~an}7g*XkEE+D(
z9DALl*cq!4%_ty2d!(DSMknjeA62Kzi-jnBHEupuhu7+di*egjr`12-zL3AazF#b+
z^M6#;hvlcKzN|l2!}o*5tZhG4e}l<4{Ij89l$>^@yl5wp<VXY90bM$A{CQ@9#b^L(
z1fbz&uv#j63K|Hze_x{{`aR!>{-C8!uF23_Qavn3&8r!@Mv^(}yb8^0B*_w19yXBl
z@DWPt29nNrZG89<xqJ)Jzm?bz&1E(#xlC!BT;_NqmwlhY)DO5HgnkgW%5IA!iX`O7
zYJ)k!O0|<~U}&K9@QAG&{h#s2@l;l;+u5j+tLk_;{P-8te~g>ax~#f2$$F!0chMOQ
zBO^EfA;}C<9^_Vwo2jH(IIj*B)Sp(*+IMYrtbUE&`2Z3Hw1^42Ou@m5ho`nLNrR%!
zK?{ASfO&<j2iD2CpNTnM3kA0kl@H|D0Sh7+fK5i5NIMwza{2m1f0$JB#o%I6SK|hQ
z$z)I-`{i^wf1S3oMSXcNZzrGY*?c?eJp~=4Cv&X5G<z;-q1D!t2C>^eYEr(3(%cZZ
z2Qb%263jMX6G<NtM|P9hod-?gO(d;B`Jo>Vm^)x(T&1SbZ-bHP2H)X)1%$!$gPs9}
zo&kj(M?;UJp~unC<7kNeAoYV@bL=(8UUS@zQW0A%e@%2EZ3}E=jlEQqz1}8SN8`o&
z<%J$1&R;Gk<H5;b_;D}-)K-6vMwhd}d_38bVVSJuID;bSJe*I6*C5vk_?v!=rPX3q
z*VU)N9PwElA#BFY!JD^R0t;~23QoHSrU?5fYRYiD7pg=*53kt(=}bg3o3ud`GZ`(%
zP=q87e>;YN*mxi|K3mS(i!s1^%bHSJa@dHXYNh8#(4xzg{yW;d9}|yGUj6>Z))opk
zlruq3;4oPZ(im{+(J4BOQoVb4z5ry++G%w(88oBq%|s92knY~2VJe4rI>TmEYVY>s
z#*3lU=!;2v(Pz$lI2%tF^J=)0=^n)-MBAG#f2|(_`1fl3^YOFir+3XC)s6h!K6rfs
zrGT@+&=NA=Iiux>3=9x0%1_LEf=8v~_l6NAq!Vv%*f(N!OSc7VEa;=aQF4V>n{tvS
z#_kE0PV3|4>+NAlcGB=y8y!rR<J^7Ncq{2&AHI8a_+<OkaZ-EWLeeZH$8?90&XAmp
ze?8pH>BrG!JDlqeVq-~aX|6uoYB;Mw$D7f?v=q~NzHJWym<nD9q%<|ub%H_8c`_Uv
zc1W3Nv1-(!Ludxgu&#a?wzK+R2u3&?Uo02x&Bj4!ZKO?3HR76c$HeQ-D%MG&W~?4n
ztPx}zv<DZwU(!d`a>l-aBtXDyWTtajf7|$j$^1k4dyXsoJ@do)d%ijV@A!M_DS7HC
zdFm;7>M42ZDS7HCdG0BB?kRch$$mH4`A!E7LV!{BfYI3itCPpJfp2gU$XPjNnAas0
z{w}0s&H2Iu=a%L$f_Hff@-0oV*A5lTJ8*Bd-<jOotqwOVx^Q_X<UsC|3G%8^e<#GA
z%T{>cPQZHEewr+sK?kr;+u5fQv0G!S&}4^(&pGtZ#wip|P*g4iFlN7d4LAnUR`_T6
zscv>Lv_Gh_;C#$p-W+Rrft;oBqkWRU9qkL?_;Qejh6xy|ETAkB@0Z~1Xd-&I$oQXw
zHuA=d(VkG;6DK*^-9WT*lx=aLe`R@HQd+y6>_eHK91NQIxH_Cn$4$MnjdC0aT(oj-
zDKCw&*}Fwy`{m!5x>WLZQXRcHef?%<BZY@#iACu@vx=Z0DeTFQFPF`DdMBZ!^yYBS
ziQo-^>)H&<&}aOJ;=G?Xizu*nD|PuqV2es^-PcN;yGmWYUbq_gen4BDe~0-f=V4g|
zMuEjt6b+gpr$PWpGk#w$A>P&V#lhul{JC!aLe|%lb}IP3j~5as&J0ijf7X=btrZvp
z=w%>K!Wh0?tgQ`xcdB5c*U4de8$sRC3o*frC|^W@sO0ZXPrFYquRjR9OODm#(=EWV
zP6WzA^2VN5rnlZ>r>#rLe<se(MNVJ&((zgOXy#u2e@E&&{C}+^gE~Z$HjtE;JSf&1
zNP3&J0~7u~9=-4IiAXjste}J2;JX7*VRIWDx^E2x9m+>W9Ts$7<aD1HT@CWpp!+_i
z`@U#rj(+_zkPR9&sT;04^yLf?q1EWKj14mmacAJUU_tWtkb=Dxf8La@BrG6}fdoV-
zxqg+dM*}_>9mz*-E}^RkW`9N?T+zC_!kMys5986l#8poO5EbJ%rbIq5=NR|s34|!>
zx4kyP!-(Woj7Vr^JlufRBP54z+wKDpF{hPJFs(0uXft}7?h8|G!YUv-P9{Yk42l~F
zW@4B2=ZNey+piHJEidRSt)e1X!_Ir;e2h%U#E@=*qT}yUlv@H{4=?0QCvPC>q=LC?
z<oqAU-;tN`k316sIWU)@k{}e9IYKfhm*7Y=mjS!Ccu6#50e>on71XJ7<RQU~r4A8c
zX7uN0MC4!=k>bpH)ge+C)3l8j#yN%ZS9sTM98-8#7z4E*=c=diT^^@(q%&EFjS^Yn
zSdiCn+Yu4~2@J}L6IsOjLqtQrAJdyaRCtp|CZuI3?gEx(JR|}~XqhpKGo?RAmalo8
zqF<B>)cq)ufPaHT0%gv+;3%b_A(K9@7m$=MG|9{*v^ugmj*ct@qDIDeWbE-G)j2Yj
z*9AoSBH)}QKe4H3ft84Z=mP1K|AKI?1wX<&%-oD%$V*Q{5NHt5CR+iE;Zo=c6d&n@
zN}-7lNSuW?B5(!fJW5NPJ2QBSiXT^0(I-$z3N@)x$bTl9pMo^6X^uTl{2|We51EoU
z5=?A_6LlGkXLdEMR6eh~j`R367}!$P(V4(PXe?oF42}OD+G(+NxvPqo4K)~oWB^%P
zRBr`44%VLeh}ubROV(cL?wX!l@APFvjZzC<;~M#p4x>#3tg1L=rO*2kWD15_OT{<x
z2t@ly;eTVsG6DrWYVj$;<SBq8-<$Fk>_<jwb+Kw?s_38^`gluykY|QI=FrCK<~6Y`
zly#Q&M>Z1yXk%0%dQ7xNkdgfID<A%eFMk>+=~%38Y3d73m6(eMD&_uh`uLhH31^w|
z9?MRN1yiXH)TQ{@BE5%2F)0;R<f^A)A&Tw6n13-Cd~cw~Fed+FFgSxPhMmI@Bqrk}
ziq)ZytZUnPX2I{!_8#JcWz6dYYkB6Vc`+DeOz!k#9Z#zGEg@Ps-&~<csY4IhV@OFn
zg%R+E{D&}*_bYFGc|6Jlv4cf9n?Yl@CAeN?w@F=1THM8YhwblS9Zd`~4>pz5)MQ$8
zRDbCGk`(rF+5-Pu7?m6y1yw8s6TZ;$2FvXrO}Mfk=f;&4YG(W}g?foiI5H|Yld3l5
z5W}CE9J!0%7T|kf*J!4&HP=`VNmsOCON~GnX-_FX$Lda+RHyEfLx<_T2tsS-ogb7p
zOqC{A#QN<+SCdx_*HkvU<k2B~FX9jaGk+M0!zre$nhk6sW@X#)|Dl9;@oL?6sT#x0
z;_3)`H|7JOutYRWW64p)$J99lJ4z`~jJ2;!Rk5xr-)VKTDwGcEyD=h*$<YkxHB46<
zXt62MeR~xPd2t_&EN+IUIZ1foU2LoOT2D*KyAzah*zEeo+v={Wk32UNR7c4Hf`6JV
z(UlU+W4hf@qHq7uq69E$N|Wy|QlIOF({?*>sg$YPV)zi%><wY2+jIBSaF&o$L;wZE
z8NC?4Du@BJXNrsA&wmYHLc`I3%{>&S9b6r@artOZ%}8oLJV*YyF<LkL1Dc#xgY=#`
zr#5a$wdN5FbqZN|WNHE##T7!DxPL}jDWWs|7szi$s@4yi7c>CdL+qOQGTUIgBPN{P
zmraJSk;T(SLvo<qbP0}mM#F@z55*!n+5`yFVlSEvS(1MTd1yvf`#t1gr-;{qsuK<v
zm#|TF!fA%r1pjPhO?!uto^?-HbOVg44kLqf4r5K(W7fr3bZSFkvWJmyXn&sKc{TF$
zD=wljmSec@uyPEJCVY=gwNUyHU^v7bIod%T=M=4wJQw?qI6Lj{M)*tnGH;M%3Mkgn
zm&xpQDVXfXG=YrGJi@z0sA?e}5vp2<YoV&0U9@w2sOsoi4^^FLZXBw%lb1r(VXDia
zYLn`6sOs=9g{r6I#M<wsOn(Z=wACA=!<q5PlQ6CXsLjYL4lJc$+)bQ8=;%_S7~YCo
zNpdSbG3z74wbDpyE<Iyl+~)Eq5vh9!Su5*7y+EQ3+;MMRwjmtOW$U8~!77K}>hnn1
z8mm6;=q3?}=BcEkG#VSlk{W`C>hcjOdn<*?`{P1>(gxZm7W;L`4}bmIw551c(v0%P
zxP-B3M&K!Dn<TcB$7}JW?^IDGDWb-Y@9FzgQQBWEN)9Lxd1UsCkcCH4d$p^=*`K3n
zuUpEs?7yhu1O@9O1&8EgG);q$I*3d8ILWG={rWd1%F*9ZCybfDr)xRkfmO^{n8@Zv
z2^rgBg_&bJeIgtZbAJcwPGH|}6>QX<P=<LdSbF?xy~F+jp~E2fVNpOh&!*<zH3KcO
zxvhc4vubJEL7<!2?4%+DL6G;4>0m&>5s92ak(k}B?V+qW2_3V_M_Hf>(=i9eqY|M+
zvXOw-w{8R%tg5`Ghby2vJ5>$~Ri~%4Yz{ok9Fws*QGk398Gmkn{NlQFG-R3E{F%Bw
z!T`;YjMa||%`u6^cBtWOdK^Sn8;H;x0uucJpE><Tm+Tb;R3i(wwzoEWm}7+-;7t>|
zOX3aJ>CG4>m$iM^Hm<{<hJnA6%`x7y{lgM{3_u_5sQBJ}Ob^V0?YA+JQ_LWL#U>^g
ziJ8LQU}vC%e1E{sKnHPUXP}#Xpqaq=`wEv>B7sIbcAE+pY3MM0zv4!POQuf&bHd-K
zSI#`OSmbNRh?lkkCH>_@4@p11&DboCfSE`EUM($qZhY0HUi*WJ)gkwkNihpjHT)K>
zNvbJst+WQoTXfa{I!a}&2v#)KvQoJYMF1(vNzY0#lYi(5Sh)k-V4DWGV_u6}^_+1R
zy~pxexxBxl^%mQT^@e=D4qrK7te6z#D+k786OP5_s<S>iug}c^+sU!*`e1B2F;~UW
z<=QXeD96Y3eW^tC!r@@u{$<DV(KpHH55YQ4lMvgSym*EdRe*&`2ZXTq!J!=V{68SJ
zD9H+CZkI)CF%JVdF))|$WdsufGB%fCG7Tq}8&fo$f6TA><*%Rq{KPY(oXs|T3~&9n
z*H2HsKlN;z*)b)}^!fW!c%YSF%FD@1X6NaTzfWJDKJ-M(R2)Vy6aIrwe|(*~mXBF)
zGY7qKYmI62CO9ZlPu9*%g-1zmdgk1K4vc`$pdkag(4Ye&Lt7*3yNee?%6eu(FDQem
zqgLfPe^FsN@QMFL_L~-9zfiP-&D`}AMaxg^ZJDsL?`>?BOyyUQ$Rr8FGcB{kGD&Z;
z^r4T|Sqq`vy#!<7r`yGlu;yUB)_A61Em4f$!fPywvS1FP9}(0TMZm@3`5h=}pIGLX
zxAUhR)5=}pj|c0_<_5+asWvSi->(}U%GAyhe>R;tC%LrWXJ|%&R(wy%yxwPQ4qAj=
zXko?JoMzT7rJ0k_MB_PU0}7unX@z7qChhWyfvW<0yKB`zuiE+l^{TbA+_P#eCG&c_
z<i;^;)I##hV<ks-(n*d^@<!+AgPrK5^mo~;<eF)-Se+G{mE1GOZQomF+l-FSrnS|l
ze{9Dh*yepanwdE>@8!{I2FLS)$79}8KJ(?O$9pipD(ow@lU=1G<9@NM0dW|e&YkE2
zE_utywB5;;U!2R*Fe7H9#8H6lLx0tf(NPG^OgT$XoAId2n(S<O&!mez*-nzr(WnRi
zcH*cTu9)SrU>4kp?iv=$t$@N77)*f=f0$*!V82h$@%<Eg0|D?;X8O+*o0(AX56C(2
z^*{o!h4nTt{PgZ$U%&k6^WOx6VgC+Y0my+@5#Z&%=dTJFm&Q%s@^{O?1ZLm?fQ_<x
z(w+%8AzT?)8a_`?8dgpy<4Nnc&^j~iwUlIVgpV$C#0f|f@(?K2);!-g+E}<Me^W%J
z3mf5EppwUiuP`1*g|Fd7zhhSz^^fW4%k=aiYt>k59dz+tDihqY^MD-(S`~0AI{^4Z
zv)Cjp1CTB>YGrm{Z7Z!*-rAMHQfdVf2G1zqQ<<&Ej4ecNYyw(H0h<bIUHgh2I$5Em
zZYXA88I@)Z^a3U-vUZb(>*vqkfAhGnFzP*dPjsGzRB%Sv>DgGrU|j;O7#|ovjH_~Y
z>`e~*c;u<<wbCr3-PI9*={`(^og~+D73{-qWk9HB%pPU*Zki}T3+8oI!b)(7GUVPA
ze~aYK%{c^=?rMO6A*a^m)4o6hdj|wL<2QqdBYNx7nkhz}W;_)O1w0pRf9WsC33|pc
z4YgHS90rQAnd6+Obw}jLB&za9CQ+9+nZ!S*wp(`F&%|$|A$y?UtQsW<43{kUHfqSH
zEVGXMBUV3_mW(dKYDMWR#d{HDK5MHFb<N_s;)%-Z(G;~D0RYktPvd$;{R?$`&GYC!
z=mr=eqeiz<AmFLt#45+4e-fF(acMclW<{3J$Kw<GMTk<?wbMfSUiuzUkD2dNZ+k7Z
z9lNB_<%<-D4Sgm_I(?qxnm!9BjrAJTR<+=TejyUjra&@zE1$~V%BMn0DR@`pJ*?b;
z=v!y~DNkE{>6cM`Xc31jb$n80F0(~$UABfadLx!i{NUIO#0dRue?LcQVDCUVfcP*I
z_}I9{f&>3;r^4j1I-nqYagbFljj0neJB}CI0D$GXL9k%NfCmoka%;r3=d2P}dXa?y
zWO3-s;!Np;9??8hZH;C^6`i7mXKhTU?0&v<Eaf@L66b!63C5#E#JSIfiv{P``9jLt
z5kuCcRg*bfbr1C^e+C=?jsORSP%I!q1#v`-2Nta#OMty_h8p4~$LGcRZKTY&?!x!6
ztJO|30eH2ejyjGSx#XO(QNamc$vSRXpq9+&;%<JJTO7HYbmX72``|g@b6Me_%0<D#
zYF3%Tl1MYy<m-#lhZ-$Dz%t<6WwT!DwK%+#1HSI=_E`z{f5}RgGtoHpLDnA9NgFcl
zgqyo$n+1d5zMpaJ$dnp-+cD^}hq`NsU#P%j1W&cWy<m|>0T`8{%`e<}j`nRi*z#<7
z#?G_lxz2g#<e3XiOLswQJCCqrrI?j-uiXdvIOTavc#f!>$H_2r7Sa>#B*jowLaSPq
z<=thR5c4~rf7VmIeyN2^JaFQM2d5+rLBP7jQ@vT0C&TR^B(qlR<13lClo(qk@Df`5
zfpUs1(Rk-Ff*5)p(V1|i6ShYlB`Y`4WCHbS>aL6;pd!ysZvxIsZ3b|W=4hW(J_;pX
zjIDs9JkrAfrb#5?YowfeY^*flRhF4AS$8^tDLI5Pf4MKktdi2Jx-N*D-tQRH!uGMx
zIRKdU+Hx7!&tl!_Di4vaQ%%lXA`A#{m?y0!K!)){f2dpRiFTF04f(<TV&kH<fd;Yn
z&9EkMaZ?n<(awoQrPgpi5U=<{VqP^8T$Pc+6hmPO2(Eld4u@MNE|YT{J|W3%V2~L%
zTD-&@f4skjC5Op=i>esiLx4V*vc9LXKpuCn=^b;*4vowYPSW!x&(M`ESq|P;TZxw_
zrx^VWtsO-6!Z)IhY*|nrS+gG{m&4x-u@}~YQFa5KTl>3r8Ke~ufVf_I*-0OAK=P8j
zg*v{(a}I$mfn0MeX}P{kN-CI14E6mO(^6V6f9`xgLkOWlIZt@s;IY_*k<5G3WNKo5
z$<i>!VoUK-wn8@o`G*-#SMs^;YWCGN88COK&}=<~Dm_ijN$FE#NooLG@nj@8gkOr(
zh9pE1S4}psz*dqx!E{FMKLW}mI7*ptq=WmU%^J69n$k+sIV-v@keX`lOmed><%zVc
ze+T2>u20UE38#PE_5Hf*`?>G>)I;w205(9B<#yinVedEJ^^xgT@A}jazw1-5_nq`R
z@A|O!y?1>-C|&on-3|KjNxA!OkSx|ccZ0;w?<gLY&NFZy?adCvawhVABqb)EB+f25
zK6d<us3+6B(K@8re|MD8jcDew_-z<ne;YiTk!$^$WwUd4)gX<h*#JZDmuQ(&iTO=`
zVNwdB&Rx4La$**UQ~qegS`(N!reGzRxQ}?gB#~<>kf!BU;%8~$K=mbiBoPsz9n4LY
zEzB>eM${;&7{{DRJ|M`NB1c`uj8-N`msj#0%655)Vxu#`gucvznG4mYLvzcDf3K+X
z#2nMw=#6A*mIo~|(ad(&?2-Ymkyf3lv79B((N^PZQLmn)na>n^H?o|`4zy$2Fq*0Q
zGlK=5<r;O!+bmapvfIBWDlrNcbIw$>)<G<bSfk>0k;f~uSHyp<o&ap0X)StL`Eg@Z
zD885%8k#8w+p!O+Md11^=O|Y<f9g4Dt)5a>rPNk2F}oXeqX2<Ge!s1WKSk(uW6R21
z@u0a!sZVHZ3A56tv9)T+l&g7gpD$NB4_afh>i<`2L1=XV)-*lpN+8rmOW?NW`o22>
zDJAGaE~svL%)I-g>MyRA6<ke@u<3Q~aaCEvs@+l;do6Gk-oU1`zXI1^f$RT6k$<X`
zV9WE)xHWfvbJ$uZ`-v)4!zr^bVhue><U;b>=(QRz)cK@A#F)-Y^k%KMsu+n@05&=0
zru?JUh4qo~HP4&!mj1?E;-dOZj4H~SEa&4&F%;k*KR2uKsG=iqd0>Eq@OBuTt>>N)
znw&V-_8^{5W@bj&_43;5mo0lrW`B1CIeW@vg%(WcC8K>g-ssjawZ>EXhy{THZBggl
zWJ{}~efH%2;S^hx--!p@RAp&)VjdTxX4P@A^a`0bU+WX9?5X@ublQrcjNnDF0uwA|
zs0@A>fGd&$(>2hX9|uFxPT=Gf{?QeL!i1lf_Qd&R;j^238&;nbU)JUsH-93k9LqjJ
zITF~H()aw|i_yW=c%c?l=Ni{zl?G;EbbCv7-tYg>^z0SW%WHAP^qTwRhn4dINY+d1
zc+@!(O0FOp=%}w`!33NW(#RfJNM{D90m8Oxxuf`0=N~?n{8t)7Vj@CiCBK|!zcIR(
zUiVo{XE(Bvl4d2H=_{^^M1M_<kZZD5iPu4OK<!tuXKq*a%o$R2n0e52ra{pm_S>qA
zO1-+Ot^UWU**4~xNdj}0KPy>WB&Vb_*|fVZ!MhE8HT@(M;1Aug#?8hc2%Pjno^8Om
zO3(nfQuCOL$M|2zEP%2B#}BCd<S`CvX|Mg^0|1(U`6dTJP5CMA2|6V$Kf!&#ajZgJ
zkZdQ~A#wN<9r$MuYWlSNnCX&#@Scje&<g}_VbdRY*)=eaI=51H?*PyL3wO>um*HR!
z69PFgmjRAA6aq3amm!ZnD3>f?G!cKsvd4gLPUaT)5IC@N@C0#k$nU4Pv)L`FTb9R;
z0lC#BF16Pyas>Ot1bef4`M7)WppaSaG+4<VKQf>n1pN<!h(<FDz>Na-_{4s_6G9sL
zyrd5bp%i^AeRuTKhrI$vBY;K#{Q_{<TZFqbM!EcyVJyu#rZwaPy_>@x1qgo`X5W|3
z;r-*=oyd%iTp5d;)~j66YUl$8n%*?MY1&8Ao;vbn0@Wlc(vI5GbM|Yb5a)>Dm;AJu
zZpuis-7i?F4|#RWBLWXV&-*$8GsV_D!?>4W{eBFZ^h@AAzCYCQGV!O&`XWDHbsQDh
zp=hIU64#XL<eliuwvpjoM1FtlUE?9H@sLM}{ERzRi!@22Rk_s8NhUbb$0@DY-(=@G
z_2)O_pAMG%QwzgEyU`QSbb1P2A*<xQKV9;AV<Z>EOIfwkVe*^cZ_QqTM*hC&u~HGA
zo*AXNRywcvQs~L(TWAE7<d#U&oz7^4Ohq<<SLvff_TVgIuxN<RAUl7)PPAZqBSK7X
znsuDYddUVwV_>*a?m09z+39?o1$l6b<_<w!3u<`rK;C=C1t~{yi`4~HpDfUIP8kiX
zm%^`0Ad?xV-8gdi`f-+T$)7modG_IH_l1!NZ&;Lrtf0Kp>|Ykh5L9lAvGi03;DbeW
zFaG-U@uv^J0jCN7gc*N=Az2B$V^8^2aSzG^`<&kmmqtp>fO{$q$k1^xng*1KmGFV>
zP+AU(tlp@f1J!eMC}lTU0awn3fn|e^RCow0HfrAQb5E>X)h!~^m5uNgR>^b2E6p#n
z!fW{S?`0^>`h)F0vfaC?Pp{}7J&hG+Nu9tEtb$&pg(MA`3;}<OI=m%=pP=@FwxQfO
zgNj6R+zOy?b+;;Pp>zTl60KEMD5T2TKtG{?LDr;_<d>A<$XE)w6FPV$WVI&Mw(lRF
z^1M$p>rMAe>+D0aL2cMx8&CN{g|#9oIuV?Cco|)S^ZuapU8|%qEE@s{K~;7x;ok?@
zuqd(cj&&?1S@(ZU*mtWHMBq*9H3}NbR;L(>7VVm1PABasLmu4}gAoAetdJ2J3E#tm
zV*y%XUAD^ehk01(8I&a;B64bvKr!umpMTe>SOv(662@$sR?ToM0;gjEg?a6|S{!kl
zNML#?695x^S%$n$E*#x@Jr9XuCcrjwI%vZ5rD$2|rk{TzODb@OJ!G9FHy(E#fU}P<
zU<!JcoVvNsl2bPaO-}tMn-0tMXWtc|6-^7#2?BMs^8hA_BDY~zH$&-ZZc<e%z6UAq
zE)-yJD{<sGe4Y25@o8s##@&1~Tn&Z$u{WtBT_@3`lSnQAug>CIT7m~C@a@ssbf?yF
z*^F&{*U^6*^Y6N9@ERnF7Pxi(Ib@scv{AO0Y_5_|ONvJM9P&+en#ngyinev8Eq1Px
zQ7OWz&*GnxovxJedRqc=@Xz9_lbv$0oBAo1F6cU&krp|%QqLgEbf-xt7vsuaoja7(
z1l!DW$S~RInY!48UQs&FHSu${R}>tiMjtvG0Be70r?akOG{~(1=}X6Gs2vgBsM=K#
zk`|<w*1=W6q+_hqNypeG*PgF#(2)ucKTyXgxHi&8rx9skPyP7T#*sQ^47Qneh>OJk
zXWOVXH(K7LZB(eal(@2OETg+)>j=4gJ)m6CCU#pcA?haFhBlF8R`RRbMCm;@0#~+)
zik^QOYK>HjEk<gd&|=QW0llNu8F`n+E>R-##WK6h($iVF&5Bu3wbm~qic|TO@)Xw+
zMxJHHUb7mW--ApM(_`OM(Y!^h9rJ5GxKGS1D^MB9CF+rx1RTzap)3_7N>`&8_wDp>
za^a!7E*LNJg;LY%Zd{$3nR7ga6lwHqmVAHc`fIa|4@%x-?X7%RL*;)qD_m@8?5|=H
zFF96E@t_425WmF3UV|!j)@M1Yff5>IGbW7QNDXV9WeP&Q3QSEFtnFabFa=6FC&uv2
z_kjitIFyrz7+7fUmLYAq?usFbzWHcbxocPRt{0lna604SwN&wk#z~OmW>-tfkXL`s
zRJARh8{2ZGoP?#YnNrnsM&Dm`rd@YwrG?~*Qty;s%y*(mxx)?I=w>B*m$_$Us^=3I
z%-)@On0lH5{gGUx%-t&8BVRmP>PpVy(NZmH=Z}^;yt?9ONm|K;9U`)7r0?~Pb6*hk
zB&gj{SBl9_E!$xrPPKT&dDXUaAU%KU0qPpLG}KCW%+=5JcjU8EBC^_-s&Xm9`F1F%
zjyteH;lI8Y?05R{Erq4^mYRbw^odDAibyCX0O4PD@6H)UdMeMJNwgH?zE$T#Jrz66
z^X*aw0+*EP2QKiw^p96}eJv?jt~880)O-9J?BFS^2%9hLu#B2!XB5lUM#X<aPqM+<
zTey#&aRRtOXoyK5LKzm!1YI0k%2G0g$YKM$%I#x9SId?-3FMHvc0!c{bBd6ls)xyx
zHo=<^f2l-JHuu)I*IT)-49y$G5cts3R$gR2UCbGJQVB9jzf}DqT*~NH%??p6DraEm
z`zVK6Uq@Z03>84=+i7Q;Du#dEXR5FfTF&TbB^El|O^=Chp&X+)g&24WwN0w2;Q}Rz
zZBoau;SZt>^1PU&!6|Z?#vbGXDz?ixtaJlG7qc{2$HB`@S$h2+5rQ_cYb{dvgSB_(
zuQCE!>MzzR<SyCS_Mq$Qp$VxQ86aT^;m&FS!{0-yHONE>8tT#kt=E6w4VgG<4)nmp
zErE$ejwpu>Ziq|ByoAsF!UosDl4NP=xHo!>rEi5wgrkB_DHAB1#AQ@+;mU80P*Quj
z5CEGI3Z({+BctD`2y4x~BPS4=k`VNnRhL#scPCdXwKsUmMO6&De;-vSOC=Xrfb0J2
z{iig6?HIEBCBjDKO!*|)uA8MWb(TUBH%00Cs$`I>?l6Aky=*B=@U03{m$v=`Jaw0s
z;b0FF135A<mvL_k6PNmEG{6DNx4UUH^#=sWoQY4Dk$yB2128i-mmxnnB$Ene3Ia1R
zmjN_CD1Vz<j~urVh2Qln`q6nv535+b1%d$CjvOOSAZQ)rBIrSu7q%RES6Gb<=ilcm
z&T@9_>9uv~9T@U*wmIbKV%4crB)0{%DNHToGD*xdicD%sLv@O+cq2|J3XLRF&dD1(
zmDaowrxv|)vPnWVZ>343m3y=@Lu&8F;FhDe>VG7ukDA&fCHASqB(;Ax%OtHo%{r;{
zN%@-U?x{_>Lj(v-F)&ANljiLGN}$z2<veLI{~aH-7QP{Y@3!g)s!UNjs>%y$N5z8}
zo4<XvDW=`iG{u}FbJRS=()_(^!6eBM)H<1{qpxi;ZQkO4W~t9Blcnl*X*yXEN8jRP
z)qkqDY_eASH`0_uj5(M*sbxnhSi4kj>6A>wTdaV&twXD{9E)Isl-)vt4N{4Y3ge{a
zwvy1M?1MIuFlE(xj~2@pF)S^e@@^D~EUEfzfXu}$DUzmKcTe+Ff?GqROhw#QB6TX-
zM^=7k{V4;K<hF9dmfWARw59f^fGn+#jDM?Ce|=hUs;a&+(WV-4fhH_jwf+?Ut2<Oi
zzp{HuzuJ5b^QBP7*A14uAMV?<_EvS5^apN;=pPlEHe*p}@u2&k;HPc(zW8aYeO~v|
z5ESS2QCW&UvwJF5@DA>(!8$VcRb0F;fV%VVIz*bacc|W>b$jCqtV1^`u7>%$U4QU3
z?BC5+J!q`s!rQR>t^;tSbgT1fSiTz;hr{!<zZZjPH-O^cG<nC|r%=Vh{5E;BByR#;
zqm+C1?@yc0E?-?<oxOhk;(UsP|C4F+`1y8weswYF?&F_--hOemeZD=PWcQ0lmlxan
z_cxEC2YPkRc(Qqf$4d%z_rc#+Er0GlxcS@}`0(Jt=JD0#i?j3X)6L`0A8o!q|9SiD
z!2@t?zW(3qb7ubT96%S_^NZ~p58vH*PtM<5zP);J{$>}^{;R*8fBW+Jr<XrJ4SdYv
zB&5&y{uSS(R}8(^!;6c{K2VX}^!J{2GV|ByWVUN08IAEM?H}1_q|wNuQGZ6Gjz+s~
z<S-gz<UB^sW8^$W&VA%(dn0tvo3ppyZ2i+$FE9Q(Z9cuc`u6;)zifE6`OD_Z&1X+#
z_u<L;i|td1Z7vW(wmdtTY0PtE*>g&Bi}q~uaANqg%@>znUv561J`p3(Vl<C%2n=?t
z;SUcU94y5Lp~y1#%n)r>=zmU1nU(0w7C}*tQDjNDZz{D~tchSO?c-7K-m}wCDvh5|
zytO$xfwnM{NKS<!wb^sk9z{tr!{3UcmIVK$YS2K{1gN+Ih{~j%MXo`ZCMLfXgTZ2u
z1wma(GUPz6ACEx`i|-h6oUt)^mgVt`g+buuBM_P@^OFWzGx)eBU4H`vt5I-)By%S0
z_ZYyWp&en!o=HmyQLREQE)CW&M4dg8C1N#`r1l^HrD-8Xb9BXGT1fQ>8^~K$aEYKh
zztl1p(Hj(?3X+ad#HQq!N~R-gHA1@C3btrElO&R)xugSA04mxNT_I)6$YIdD;T<Wh
z;S||sGjvjja}5Wk0Do1qe2fBFW&v4JF<cEXrnri0+T2_bHkqp(*aB3s$q@?VC(^AD
zrpV)nB8_<k5Gn*IARvm`ZcO2zEH0QLyMRR?kr(mn#`kp%Q@~3kc*})+k#A6dirJ1(
zwB&+XG!rEP;lfyJm?F<4Y4gnK1;W4pCfI1fHSx>@O+Z)FA%7@x3+sU}n}Z%00#JE5
zg;i)GlgTY-WNQv-1rSx{!mNxvleuDX0!L61yK>B#?gfofuolV*EGGw9^z)~^s&q;A
zuYdo?74i(DQ9Q_W@%GiLXSeh!^rF0ra78@qxteG_kfy&9tp{@UJw+=)aU@z>{Q%MG
zh3a0k)*&S85Py<&2+2BxWE~Q<4vAWaM6E-j)*(^bkf?2p+{ValjNHb^ZH(OV$7@wn
zBlCEKL3o7%IHQz|(i!z=c_9N?qI(Hu<ikkW#urz}g9gX=tgf7qCx%aYRyNckoVK9P
z;C!@M!eFt`OI8qxZVDHLdk}<HrZ&F~1U!(8U?;>e3V)e!4c3js7qtN5(8Jy1yv)9l
zU{PGKooYR5o15IAaMLv0?RFI8Aoc(7S977XfMI!BLVJ?20$AQjE!Hsm4g->x>B+*t
zDHKlke(@Tchx_4QY+qhpfaf1izVv^ee81hke)Ffz=DU~M@85nizqtHy^TXR8U;pvj
z^Pe_fe1G-$uk#PD0C?{M5Y{{sB?Dl>KIFTV0lvMu{O08cLnPhI3Ad$knSFaADhb0s
z<4Nenbud2s@%5{}fA!C+m(ctNBaIe%eY{7C1dP7Xw<glF&(3ZE_`dZtOHaX|?xX&b
zrT0xc^@4V(oPL*(c<-=;CQ132slZE6g~c){+JCfVJrOnR<brQ?`KCrDF_S2Q?<i+N
zBA*C_TV12<QE0^`V!wf6{nemap<+pU3@*=`?<k(u5{XoqrnXC%G05DvCs`$G_fz=<
zX`z%n5y$A#NgIpf+tGCnVZjbmjD~{G8EKkp4CvHfh||{gvJ*F7(4_|*hAvIwi%{O7
zr+*Y2w%v(>y^C#Ex!z+vP_VMwt$bevo7;_oZP!}V&m(pX38NvSp`#H;!$u>GMjnkZ
zpj?YvKZUq%?0+-nies)g=89vkIOd9DE*pJp^s&*$_I(B(TR{@1;Co>@<>Il_rbL=j
zrphG!*4jz9Wgk&42Z?8Ofx)sf(VXImF@H!z4>vMj=@n?E{$-96bBk3(0s$^9%+^op
z$c43?rZm`tfZ>>1TAI~&aACL}b4NP>tr@hI0GS|!mUCVa1i6nF^AlOXAkLH5Cw7|p
zWDg>uGR?(@p@0*h!W`3r3<{~_rdk38Y{~}Rrxj5MA}mSg^hCw&^$K$67=<L?Ab%`b
zQS3PY#%w(lvBlZXWg(5(Db0c|C%zn18__|+s9EufxT0J5j*#$kglVk4t9z<~k)p#7
z`J)h>Nr_H;)Xx{B&M6-+NF;=%U;z^q$R)%z<gpVq`3DxnC$R|9C^q#g30}pX>Z~a<
zYf3jLSO>qka2txdIcsXv2L9^TAb-48KpWl);%*L`#PC2sXqi2+M>qqyiibj*3;QV{
zk-i|yR$36hFR{vAI*bivtB};)AqNITdr@83I4z+D7^3=OpiS3s3H9x*puJL4Jun0Y
zD!D8$Hx*_je}n*4WLal5U?<<F6v`C!Qel7y&J((D^oKytwfa4bL>3FShJRm%)`NrC
zr(V-N2+!*IWCvcp-3Su&IR%TGBFEX>G}QKF(^C-zgiRfEg944`H&$)|aYsCXcJP|@
z8iVh^Ic44k;%<0?%FL@gDvr6T;S_%GSiF3iO%|eqknls{i1sR4s94rJ_>&N*66*~H
zFj+cw%o1BMY*1@WQjp|fjemot=od!(K2_g+-D3dL0s^FE^5vjHL}ns-%V8BQyPGV(
zbHww@Ub_Mrl|n8rcTpg(R1*`6zpjb(`k?3iYGS{3HfmxSmx^RuDw1)jNXDfi8JCJ=
zTq=@rsYu49A{m#8WE>O8I3|+1e>5uBaZDuRm`KJkk&I&^*&P!-`+vwAFtC55n9F4*
zeEOa%-jJzaQaaf+9A=XeZ$S7?CJufZiaVh^qhfLLA0Yf#SEStv;*Ka!yDZ5`2z9t0
zofZn8JrgbN<<=~La4DI32YsZ~);r?Qu)SZ1+@k<h3T5feMb)T{IttNl)G{rvNB>!U
zzwp2oarxX$OQCd;Mt?y^DdQI&R=oH@W|{rw;T{Cq6)C!K+o452<S#+|c2rvD;>Y!+
zq>PLD7Tq2M;x*f1E$4M)GKeX_fTn4cD|?MRAQQIlSNA9up^7}EOrg1O*FYjgm-XVl
zP<%F&)QZJ5u{Re#ZmIngi#nbFSvqbZuHg%oBldroWJM5uDSyeevhB($XlYIsm$u=8
zGX&OXa3|I(w&oU?-`teFEv)$+kfnVF=+ZB&`_2n&Qmw9uB3lpB{q_?wHn*i4bK>`$
zQz`+@Q0I)e;%*1dh~k;NwDLcQC^931j<sUNdwm@$!QH50Y)uR%{eG4@iTh<SewL*R
zb4g4+p;kZh5P$V+lq;g}+hQnI*tnl-kg8D8OHmrc0%^2CnpCzB#yV$UHEJ&+lb_gE
z@|q+nTcys6C)uY~UO0LUQ;izZ|Lg>YuzpOsxJ!L20`#!|MZ#~MJsoz~pFR1+6mq}c
z)1cH6(>wIknD8Y%r3inQdshDkcJdV23T19&b98cLVV9bIGz*teatadyF*TRbk2EKj
z;eIref2VvM_rJb;-sx;;xqp1w{k<zDB9UTi2kS{$4a0BWcVB<i!|&4hdC<xQH~f=^
zejmVV1zx`XHT<&sdCPBfS_Q*{b*MB`^_%zL<ZfKP#0M51SnkG;ZR|&z6&mkREMU1j
z85(|w^~=^TpTr6eYgQ0AWnAd3;HRi1YHOo&e}%@Xfk`>Ay#{0Nr@;ns4A(TIh*7?9
zjAn|eZM3du+j=DeuE_Q^#wTQ~=LFd+AN*jARi=1DLE^<TJn)DKejejUko)z7kL=wk
z1QG<Lz3ucI_ds!hq~yqg0SC0!&JKB{{a(OC3W`~l2eUG4=CqI!2|2-KiSQVf)b4Hz
ze@CnF9sya$7fG5Kuff-xO$C`=ncys~v8mP4X>4_(2}iZ-94DI~`X?b~&zv1g1d-sl
zPh`w)f?~=`42aS1$82$^zv8?@hH#{SL+pv<J<nc}60@2?pGNvJ(i}4ee36>w>T#J@
zD~hXGrhb;uo)`0z?nCw_6g@gqaO#;fe`BT+mvts2&U+_MFR`ik%uiLuq;pKr`s!fe
zCWO!@XbJjL8J5T@D<EtIq{-Wbr+60%p(Zs8;5e~Q*EM9$kXJ;@Sk%*~=!T0eu@Lfw
zD?awYhjnYLa8L9<bI5Vb(<pzAhHS&x%$1R5v8+a%<=%zlB(<y%H}^T}Nv-$_e=BA<
z&2w{h(!A$v#Z!y7cCO?Rp{#9RV^j2%Q8=#jS*&FfV<Roa+o$eVLwh9|<ecn#Y^1%|
z!32t=Du8T<$indBkKtP?rH$CSJJ2#mnfM{!Qc-DIN?{FJK(DPerlB>Ex+7aILrz)a
zD(XYd3S_jktGlfoX+16_b5Otwf60=%v=PO;$y6NeEFa1Ire(IN0ShW<*Gq)fo@um|
zWsD2Yo~HciL&j6iVaoLL)pns~B~-Pr>%)YI&Nc*$S?oj8Q>o7P4-@_CSD0slawc40
zp738N@If<5ZCcXiWF1Ht>DT&&PcAQdTG@200|&iPa`lqT$D6=I8Wm}}e-J#RW&dXu
zP&>Ed#6l*eU<<$^|3w0nr=$fvUhhSaCqlMYKuvThFpFrCZA;onfnrpN?0Hr>Nhv6-
zCZ)xo74-O;KQk>nlhNE!Ny1pxVRYFb&sx50(2a@yQLk0o7<tS{EF~f5Vl$g%USH5+
z86JC>R(ffHsheV1=oyT1e-tW-`MM>zR#PvT11XR20OIc?-+}l$iD!tv_v~@R-`jPT
z`1{a2miW(`cMsz4eR@6dcarWz{Fi~J5PvuN(r4_47l1<NmqU)_vDey#|8GzL*rRN3
z@r?^Wr**3UT=pDQ={`S)nPV*tEyn*(1%q9rOnOAY5PA>ZKB8c>f0FM}Fj|S{3P$Hy
zQ!v`6=z?$xO5?Oey}hK=ShSKi)_28yJL5q$rFxED66qgQQ-G{I0Yk6l6*UD6DU$7A
zC+WtT5}(W&>>>&$@a0%pxA5kcniB2DkTM9F%ww6!rDheX;1uY(gS@c}2}G3-z0ggR
z<~+1BPFA7LiNZZnf1RNoq3MZ;%MofhB%Sv7HBco=cPnocjDi4fT;4jZ_bYGZsCKKg
z#YlplX2<Vjw+kP<ZQ&!&3JK2^zO}Qob7%j=3p%Q(v-88o(}#rh2)8?4zr(R*K9run
zayYdZkz^T#^m@m+vPn2-vp$+oMqxL)KQf-jq8^-9Z;$f>f9-_3dZ>HuDLjuF57tXm
zE<|>oTjJ%$mn)}Y-6M2%Y&|;)p4Rt1&)R$~LGlo01NYGzV=%K~%I767=1K-J1m!tg
zn89}R4H;yvk+CgtcUk+dbIV(=CdT}_jr_VXzl#XHfqz@;7M*NV(OjDKX7qJg=e5@I
z$|mtq*<R9hDY<X>;1%&{=HI@)yH`FV%QjJ{D9Diln{k7a2LYFxz}f#P{sn+BLK2se
zel!#TG?(Er1{9aqWGE+p-CIkK8@Um_>sRz25SPU_fq{^b)-ynEHs%)OurOlhU@vwz
zhy49i@lAI3s5R2a*)<^dh)ovj&5vY>PhTfKJ@1Y$yPuw<nk;h~te9TDOn@&q{tp}*
zjh-w3GZLnk@6$gXI2Q&#NBk(xrNocLza2jHVK2d955OLPZvYN|d#m6f_K_|<1@*--
zr$|FQ;dgV`D-K-H=tuk<iNfKZFMr?h#OQ~Y-N={{nh7DM&);{yPuTH!3cYunN$>RZ
z--()W5Oy}k;!`RBKmCDSe){d}m%n}f1sD$a4~^iHVnTy=)A#sQXvMVfPE6l&gJI4(
zFcUD394l;b3>zGOl(Dtt^R!dKGLYJ}&EKfmGvyTW-HeyOBocScENxq6g{YS1Jn6T#
zG4obu39=Pd=o?W<W9ci7&!gxoecJE2D~|g2wEHscJ{G-tiU06vEH_JP1*X7CXjPk=
zgfz^c^>tQ>w_NMB=2q0&ml<c2MBz}49PqC)UlBLAG$QAJCZv@q<fg*h(7vLALE5O4
z=%Jcvk93qq3XU8Ul4hfJ>))Thr-5H_*mLuYg!UjLQ`$^>)G=y?NQIEqZZTw2@6<U(
ztb9C3jZa02#mLiuf1lFxF+^FOcPZL>n#jP%so^h5x^G1NxLH4FNdX(QlewB}MM=%b
z@>wk_nJkNcpe*J9Tl8`|fTWb8F(qkt*c;>N8>TOH%&qu0CMad`t))~(N(*m73x7ze
z<oJ=PuZ!Qq6GG3B$k2(Lqj;paa#BkQnT6I*&nO5cg#|%`^h{Vc3PLG4@C1>2ZKI*d
z;MvgBCLuXN;I2!rYpTN`%wiT9W>h}SsBb2l6b7?@p`01XSH~q;f=s$<T4YDIN#n<D
za(T)0Osb7~q<6HqfpX*$#Den0K~9ZB-P=Kmr(;}5FYe$V;0+VFstz+Y4zhRB<sh>$
zw1VW}jlWRs{OPdw#tnL|bOQH~=L{5V4Bx~>;h>C4oCeE0C7O?2hHS>t8!htKVnth1
zh6<d2c4JEaz{@G6n9h-J?*a6b_AVYEaJ3cVjPH;l1jj{2sBxxuN)f_%Chc6V%aTNt
z-5Z*W9?!!!Txoxwt+h%4g8*!)9SJkR@hz^v(!lh`zosv{k0aG24KACscIvQz@~Gpk
z<x>Lvo}<v*7(`N?`xvGaW=5RT3FwTO(v?boLaQ8rFsL&aZ%)}LX61}B=ESfi5T~Rn
zMSSM}0{dOdrwRkWTBZ<|4HTj0r}TEftQKSMU_uv@bNGk^l5)a1Ue1?7D;hCr=W**H
z3tvCeXPGyRh}*(Ag7gx2vUL#&&!2M&4GHnh{8@)MUtPjqKjIqOC2tscSh1bKUe1qy
zz2BD9uG5*Z%wv44>kLt{&oL)-q}sKF1h*1>S<n3}v~3Q{om@!DYtfc8)>G0S-VkMy
z353|A<WK9~k|J%C;tFr^xB~_~<6*^HJZ=c>*m<k7%vTFi+g0of>zP5f-NwFf5*H&k
zQ{Tm_+?yO-lB0_YtS!lBg}<S-CFTZy<HbgGN%C|xH+NFc%+%?-G!-xg6(-e&_5|q+
zKM<m)RUm0{Tqd<P7-l9<%CoTN*WlF@kQ0qEjEjOg0{e+A?eac;TPnx_JW$M-RaeZ*
zkJtG*0Ml|d)0&@yW^`mV^nQz<&tp3nZp&IAvNCmD`$-M_bQ^PKN5Sn8H?H7+sf?xU
zuy&lEX+~Kn2Hd{`QX44W0I3ZW&mgs-w{9Zy8dL_7fxZOvbOYWSlX2WkdANkVe#9*>
z*}%gYOxE}Q0hmmAU_y!ubT*8*fzC!AUf0>s`$s|(g*y%*H}X$65geh_sJ}Zw6AVNn
zXWDxWn+Wtx<!)$iDHZbo8}%iBrDtHw%-*-dCf9_G!|jN@WcxRakS+V=v}_eCDLPN7
ziS&k8NomKVHaEmdBgOZ_O2bTNu+q@`uZfiuoyvQA7Aq|tf306&HFrfzH(2E|D&AlH
zI`zHMo%6oAbj}xX4hdK~wVB_YqZr7a?II?fHETD5vv6)qBA&aEm$;vQ)V40Ucf4QF
z0vpZ9fj$F@xY~G<j2n01X4KOex&)gK?6MAhJjsuB2kqH@YT@-vUZtzNwHzl1<$2uB
z3TBGMj$98;cLN<t!7MiQGu=#w*+<Y%Z9Pg9c`12H9=3LSfI;eKqtAoh*ZtNAx;h?+
z(j`eUDNo@L(@`?shHJBbRl{$&)JZ*6a<FD03u|)@&&1amw-M;NPWwB!O+OQEoYO6J
zGH_m$i(Iaxl`%)Yc3mE1qA&z`51G1|LRnD(75sf<GaQkbyN7JuOri8C@c+0b&l<3L
zo1BUVe18WyyP2Y#uHLVND6|dy7#62X<r2V@y^H<2nZj9o-shKpGzis(?Q*`sUl+>S
zAe4of*UQ?_r*Jby*9r!}no^9lt+-7ICY>s!ZHo4=NUm0o8Qf?QuujAB_pYifH+D5`
zWn?W=%fy?eH2wckIU530Lt7j43myr3ga0nAYBRJwc48=0TWtN4*0f37AX$Brk~S(F
z;089<wB_C{=!Lw0(r+zdYo=V<161L*7qL-wEO%E`v4x9$uQpb(<=$&mYzb4P92D~I
zsURgtK-k5inAEobY<KT^HqE?(Q6xG<h#UBAg&G8;!Of^NB}fZfhqC=T?4jr@_uPyM
zafA{o<FI<hzFK#tDUsQcSN{0=D1%OhJL6%LE*6~IEJ^i$K6X=pVPt2nSWoCT*)1eF
zoM@zS+oB;wHIojOcwMji*v7xLxLKO{-Ef4nl!a=%FJ+;cQI?J(__ViFBW~XlAQ+5G
z?{Tk|5Q1@3C+X{gY{>|K9rD1^&|!qq+JS^RxGeN#f~G0+@Rx%L3YU9Hn6d{1%dLM-
zn4oQ_3A_n^Ouzu<YBNkoz4Ph_P35WE3=`^<O95Jiy9g#|t*(Lzo=&^B*1BWRGdY~q
zcc*4)VboNi9I9IL##H7)qm&R$>9tZAAK$990IHl=d9w`V<C@qr(|9bD+XI}~r#Z9%
z-T)C_sj8=zv!#u6T&7Deo+WJO4ZugkA{Wj1{-&9KA|TKL9@aPal_yAY25|8q{YJx`
zIiE|`rAk=vt(Vi`eE3b4PQR%ls>Hi*-kJ2ftJH_usZaPFmx%8q9O6&5UPUBAAOmW~
zNG!iDXZ5SN4|jlG`f+re{w*w@<rkqX(^if%QaQO)ru;bO5KI>wAhtGGFPQ1E!O%AM
zDsH5I0v6LPh1TqGgiQfg3gMxpQyfq&A9km(joi5dgmK3P`JK|;;VIFJ_`ko9i%yjC
zNj6=3S9EHs>22e)WxEjARNWzLQC@Rtc|1t|u)N)9Vp=zj(?&==phemu8>T?{?#RV}
zJc@QE_t3)smHUB)=XzUsh%>;9?*tR;6<G&WzfHG+N?TKvj%JGkHMeuY`R06>nt!_j
z9$$5f1_dQHVa>W|tU%4QD!qQ|<p>-3b4yR^Z4{H@?EZ9ZF)2n5f|~Z~r|i|v-U`?s
zb1Tr;;eX9l=9dwWEfWMeGB+`oak(uMmkE;yCx5*eNp2%I@UB<z1LAOB06~bPu?^(r
zz_%cWz<?)*aggPZ@25ysap}d=V{2rf-D);j``&_mVS+vE&QH5f_u4SegOy&frxzyS
zg@FH3z@oFvOUa#<?CF*LdJsZ6xXy4HA+(0e!|wohd)#XoV26Yq65dF8+<PMrWseSc
zSAS_QoVgSj>K>lmac_haDxE&T1@NUg{`U0qP9ULluhc2crH#ya!vQW*cNTU$#gz>~
zF~JR7L0irKfj9HjUmidXK`zQcEAC-w#FY6kz=!z^7w~3)YYY4t;LFCcrqdmGG{B=x
zd%_}5s~R94t>IF}Tktqw0=)p;3h5`Fh<`_c&yM(iA8*=lCFCu58!%PfSX=nXda{Mr
z%$IKOGOt;W19f*#yQ#BS>$owIJ-_b$X0Y*r<=w!Dj={43SYGB*!ynM@o$&yIYwLnb
ztnEJi_2uQK=ieoVWB;bDf&tl9#=u_7BVZ6|_g8!j+(;KSlRSbl408ZZfx7Csx_|J@
zc1C$X4$^C@U(;&jCK#o<!&-73JZ!88x+Nqb0j!IJ?rXnDL^TYVqe?}3N>K1@dW7?I
zDm|vFf2OT)>JPSiVY^3^sddizWDh}U3wP&(F-tO)z>kN3*2$HYbl?z^q$8n34>qu}
z+Vfi%6$iN(kd+GDLkvr5L2Q+0T7T*IrX_$y5NL_+)AbcOOn8NrYND3g2<IG+HXEo!
z=@tpM-_Ngj?hBlH=<Ws1KDn;3j_sX*A0i}JIgi+|=ILoW?I!=DIHI#p5tb$2ng9TF
zkE|o7@U?G@J#H2Tq&DyM24PIoPZI)oE>~dJB!u;Fm$pL|;L_J5qRdhHLw`ve`J!NC
zOTx$rCEX!GdTS3Dp3;uHqc(?@QmzpwgXb1(ctYb90_B=il#Sq~GmIIVguA~QZ&0B?
zM4HL}3Q2s8H*FQSVwpF4RI2C>wCWCsFLRt%i%p!5S{#rrU-h3X<C31pZp^udlwivP
zSq)O0DEAc?CiQGzvcO2C8h=EUcqnz6b(PIkV<Ce^^Cy9dBGgpfNm-B;0`d19TnY_l
zMq+u_gmJ8#obi<%X;1P>AfRx_wwH?)^qB-w+#exINs@c#W#WSDt!PgfZXiHR*vqF?
zDs+44lgu+^Cwg0~n2O@eg*{5rk)r*!?oYlIX@9TT4&Iw_qT;%gh<|~QBx3UQy+P|X
z2<3k6)nE~8rOrh2E8)0>))J2YWaBn0nD$-8w4oWXkc(|qhpHmW>bP{qQM(blP{Vd_
zdw22mY=I|9j9L8n7Q@>T^w0$lMdynVHWga|u9i;fCTyyut2=%kgg>pBVH*@AHz=70
zM7i_p`KpF!n$M;+HGfL#Jzar`sSdqGS1@9l(EdXc1evsPibHU#6vU&pi@|~9o9Q^s
zEP2t+dkd_gL43&ijMmV%s9We}z_dvhYL+10XMa-j0LAtVcpNb0$p|Gc3itzmCtTOx
zWn(%(RCZ=LZWef*O<XWnPg25Ftw>wXRk%eE8!?r6#S;DJB7c`!a@&apYUF#DMQQE1
z22I?6-vLwh2rA!e*?|20NPeY<w0?`Z+=xkNIhKKHR_=b6=co|5bpDDJN3E}iB5f%5
z!m6Ft)_^Dw<&5mhQRIGFcYMfSL5w_va07jgn6_6PWWOXghx1F{mChk|6cdyFrU?LH
zt_?0eZUUKfhksYV;Ohr-SnjFmts0;$7pAA$YgR*%qKUAE1ixIQe#}^efuiar#xh`f
z=d5C4D0UVkwgx>JpliHE)jA-0XJP(LuJb?-+O6oFUh)eGom=b%da4g&XrZPCw+mXe
zd$aGZuv4EWTT$F#V!JjJFSAwhKMIeyx<p~~H#JIY41dn0>kl}A>3VH5_}Y{XHT3kS
z4!7)At4;cD{Bas%3<mkow*~cVU&56gG3~}KeswfU(v~U0wGt_GwiFs<t6I#UsZ`Oh
z*=T`Trl!iRp0{?KN3{=3o2(_KFFQkk&f*Pv$rOS)wEDJXVhn*XE(P^`W;Li6JQz;W
zaIV|GD1SaWV@%zv$B;}lxb1s^hRY+3y634HTY<evC3H3o(4ZA=sinZE^#)QfNei9C
zOcz@GyqYwwm;5$UNkh7fqkJD2RUQ-s&e}G$IoROq5|wr78<({i8|BhUwyFCt`_&<l
zo96izXQtj@{KT_W@B1^c<E(Il*jujl%X7a$(0_4+DlRoPOa<#wLNzmA8U~)`g2AXw
z$0q*N^@_fm{@$+!N`h>9hMFI~Ig3^bXilFr2h!hgzqeva`X&**o&uYXW8bu6ln>f+
zr?B>EV`7ApA@a38d}t?moXJL^o2F}P^`K?Tpb9I9>|Lr9)u);iwiMTZnR`g^BVCiJ
za|kzm{vTsH{v((1mNXTY5@0J512Zu)mm&W-Cx2W!k0ZGe&i7a74-AjRHvvI_rgwCJ
zGaH*3j*L4;7x?~v;NrW;W~*B}JqHI`janqDSkJGD)bh)JF8uQAxA!04etAfBvCL_(
z;_~D73*Z+V{{@Z@jlNg_W+YsG{Bil`m2+Y6^^Pyaxs>>___xD%eS4GOa0K88z;6I<
zZ+}+7HI0!jzl$&y=PYRr@xagK_NF*+5oX`<h3kX9{p-iyzVXcHFAwdWacQh#xPIjl
zDJi6=kb*)A3aM$p_d9Z_0+;%ppN5g|^(n`fOLO8kVT8<UxH`(Gm9<VIA{9N33Qx31
z>qXkQo;Gn;donj#Goz}^k%2@n@|t8AxPJ<tMc$dtJB~bM{*+1lD6X%W50ghFJ~$KF
zIdS1k;mb)aFMp!E*H8ZuG~i&yv*5sl@++R7%Bb<J3g7R8zULAWMl*+u4AjC-qpTtA
zF-Rj)R!#Rh$seVmuKAOwve7pYEH!c>(`QjW7n#L1>ioLD+gqGDv4b>nZU@OfjDOSB
zT0y?&9Z+k{{O#Cf9`)QVgM3$(rX(}YR5{LV7CqZui-e1xD^>oA5~@5CKY&Btl2%&^
zr8oyNHQT7=8ISzUJ$tx~lAP{M7zRRBi_G3rbe&c?%@05-ChSO54D1FWh>4Iz-d+?^
z30y(=O9+oqP?2YQ%u6INFmhxj5`Wd_dA<Usm08KT_>lbUmJjb&O2?EM8HvW`Z*Ri;
zYYU#i<nj~h+#!8&er;?N(@$fgv8V43g_;#r)jg%*#`LaDXfDa!@|@=(VCn#%&<y-@
z=iYQ5?F}=$h6Y)JQ=q>fS0Kp#O2Euch2N&2n1iD)PdY8c6<!J5;|vU$GJoqi&kR9T
zx*)HZU~HYU&=taL^>K?qEox?={DriTXH{D`&1ob^rnFi~_iNInL|Ejx9jO6tvITzL
z-kfC?q2fSI+<PadP(@PzfJ1vy7<QzWBbBs5)YZz?+S(`VQd!HS)_;)(HHJ3F-CBcl
zpLL2E4KxCttYd)i(2kVEgMUOb^HX|-0O_n6SLTlbKVtlH6ciAcaN7PnCN;g{;;Z1S
zmW=b2G{49{BX|z}Ip}UuLKM)Z9>?Mg$IPo%QlmbZA>`Niq(m5N?O&uc1lvpq@fRsg
zfn&(w7dV_khjV7;aH&RpUU6&f(R5pokMe7%FD2w!3)}3vveo0UI)AO{ppMMRJ=Q}!
zk6LI2cDRbxiZj!Up)CB!)?F+;paN$J!qS;_;UYsW;e^@dgflaYm*#nbv<$&)73);G
z1p%7WipcZTxn;U_=%F*A6g@<}WN(Ss+~-&ElHf6$n@Z&g97iP~T6DaSU~yVlz+mpJ
z95?N$7E4)#&sEk#DSt#XG0<$VYXRI!^`EQbLrI+r=JhB5%?TQfmO-)cR>l_&omba2
zEJe;Wa~^R&Sh=6ta=-1G0f2GDW+f>2o7AoFEZh)}iX3oS$&diFk)aJR1$Lx$x+x<g
zq1X2D`qGjnQXbYxOTH&GV^X6NC-bE|f$JG$wrGU6Mkh6A(|<c(o`zBl8Vyy9vfj|X
z@{(!U^#M#RPEP@6LNhLBbc%~UNLP1h84`dPJT#}Gj&Km_2n4`V!N5eR8#BtG^?mU@
zovprkrJ^`Cj=fhue40Hfow13XRwFfbUhQQ6@0xP=^rCx?&q$@$<G;B6y-BCyL4EVk
ztZNT|D1+~<b$=;r7lJouqX|i4$JzY8GYczQ3*)Q?aAFJ-8A>i$e@p$g>ojGkXo-x}
zAwZ*)XR5YyMD2)DHIpUrX>1(H0wx+0%2H!VhU=%Mqo=T`jvpXX=0leiJWXP<nNp)4
z&6nro=ZFBQ`Q(;U<1umUu6>Ir^9H84lq%A(+tMK|Cx1;rfm^rfLUTz8y#pq*+(<Id
zlvX|MwbA4dvCPA(v>0*ph?f!tOJyNfh4KdE>0P;%YHnM%#FFB-yt9@83J*H!__CF+
z+$5axM#Efh?UOSC-SN^%H=V8wwx^5qW==|Ev(R&oNq8;6n%GS_J=#i@;1QXM@ob{B
zDHSj6$A8*MfAkqGq;n-yp`HO&p69X;Yy&KuE4x!L(eLaT01JJ8BVgQv9^lAMl_rdf
zFm7CBYN>T+%eid>ZkSu?;xr(6yb|Ytm_m-6#1lemW65%}=DO{}DgwLbu%lUj9$j@W
z2dmE3F!;mN;-<FE9pg~X;t#<6H>rKm{ZiB6DJF;!iv@|J2cE)VMvNBUdznVP`P1|&
zlwz8zk^0%^a9T6r%}7NYm)e~)9e*^8(OwH0?bG+qqJ147qrDyvqSE)9NGFbuQu8;h
zwwv$L8^+B%9^UJ1fmV=LL?LMgaAVd4>M!J0N1xeKBp8bXR}Q=UiKo6{ycl;&0X<g-
z?{WyHoLj9_>onT<Cp&gC9D6jE59m=P2}T($-Fmd`S3Re4G{LmHc%i%DP=9x6I-Iaq
z^q`qw)^wwfN1M%PEK-hThq$-Gt=<h%QxD%OJ<vrM7i+xIp=6|+vQ*irwbgUOr&>>t
zHfrt5KgxJ@;6_s(Nsv!|ViK*$&nS~L;*d;o?u~g+<ZA?ZH`Gt9eHJ(3ek&TT@`zT{
zW^T&DUE3`?v25?T+PNjWD1Ud|tbL`Zx30%nD=R?X=f+Gk`_i&Zf7=dTJmkk@QZRoz
zD8i>=xSY{b5k<0THf^ZH@AS@DP|QK9Uc0b$&86c?=6#qJZTp*lg2tHiskD_{d9SO@
za+aCfW8y)Km0F_E?N%Px?R2%N#)V@5(B(9$l^P9;-6qdlANTF_6A_-qASao$a^$79
zY_?`;td_->g`YGue}TMzgum+v-62$WD60sk7`N?UX1Rf^w}@PKdT}6C$#4?l$3^$d
z1Vfrb$U4j-Es9b&l{UrD%dAQt?{rb|m~O>|+{4lY%sfW3BCD_A;W0JP)*!uNx^BX4
zG&t<1FCgdk`}fX;BpLCM+7-2%yxlet#$Z9^o*VMb$m_N-f2SaOdlTBeAVr}|voPLv
z6YVZuyl<pw`>Bnc(stvVc;2ULn4o=hwN9vcWCqCOlWbR)>-ARIr-0JyPqXT|v(Ec3
zHwxQ`b4v#$1+#jkMUj)!3K-<ilBUTlb+M_pTdnTg!3$OfnGz#Y`4IAy9z-!-t(a~{
zM!VZ4PFA|)fARf0xmVkXp&mu|7JTr^Fpno31Cu9;K}gN4kuQs3?-EZHLDIbbX%Qg*
z%&VCQ=>7Ne`Zm&d%*wCI<8WI?HJR8?$whe^4}S+nw?Z`fllSoHD*y$h-TSi`T{?Hm
zm{#oM^7c=qN?KvT=C6WwJBS^s-11{JJK80QA=w%df0qs|O6u@i4bF_`BH7#Y8??E0
z14c1{BL=Og?O&Wwi>MY5F4hN-)UN{HGk@NM3E}P6;nT;DZ~tAHi1!SHz5M%+Z{fL;
zsYMG&L<@8I^S8_I-~K+XN`J5x-3jZ^aP5WOOhuQ|>LNfPlkH@FdcarjTmU)7(&S#U
zNQQ8ve^Fd-X(@hb|3vk4Pl#HC0Casn@(Pb8)IkLz{JaFqnPEoAje$q8yv~CP$}}q;
zU`}5H2DfqW7pHE<1KT?M!}D&sCo#BsKC=d_<r(mmylU&G|6$1yN~4{Dmsl~h<Jec)
z=o=Oscvh>GKcVAr{1NW|4=Zly7!U%15#wmue_)Dtj8;m0-S)U*O2dk;@7wkdB1D2w
zGzTye{|WZAM@;BnVb3^*uN8l%gco{wtDb%y-;LBZ+%nwXOh_Na@X^=fqTQ2c{uXQp
zRw5g#!`gn;hde@h%~osjr9!acp_&)?MV?SU5SV!o)qoSoWxMrbYLnMco36snHFz4+
zfBc|4IWv2_5Q(;QP<Ls8fo{i*dOQ+dMWmmKe}*&g?kggzL}zke5m>v0&xottf_q`L
z-`4%>qH538eo*bj^2IUrFmolO_S~HnQHMKL#dYvqiq-~vJ=y|dWa(gBN$|lfUI<=F
zG3UnxQ|y9i>=LWk)vWFRJ$3<#j~t@!L&vU7rw?fl%hhp3_%ybxaf)kHzS@m|vqEO~
zzuiDiyq%u6LaC4E&Dzz^n)~zS*5<u8{9V%X2ng~!R&oCWE~L(Gm+_V~6ah1rf$BXI
zmz|?Dz<<t;%JE<CzZ_^}loG4lPxsY7eR@CqcIeqwk`Z9j*PjR1E;tZw*u?OqmDBJ4
zp1vJE_e5(<3?L>{DJdIFfFe+0QpytUUIk2094b*Rf$|BIQ(Ob@MBuRe=lAbZ*Yf49
zu+0@R2t8#+8x(SHAyAH?GeReEdMP72_$@B`S%1;k0;;Eu%Q?%67DVZRaktD8l|VX3
zE&(F{{V_n~c%-<@H1Ib7EDhnfDv25m6xkipFisShp5l@;PC?KPy53Snu65vgTySi-
z9$yrN=dA;dBw+Q8oHHtjkT4(zkrx{IsQI!1Ro@=hMW|}cZ0xVnT6rmV%V7<{xdu7Z
ztbZJwGQuhEd((v9oW*{)N;6p^7^7;=g(*-(DP3kx4$5=!f?w2osUuMP8t@V5nk2-P
zzf_p8(C&Y`?-RN`p<e@<30X;9tBIQHX}R&|;4(A*s3$DV<u%O9xaE+gMdRm<Z8S4f
zmd7HOQkRu?e7gQ3%!p3gbMJbp<idqWzkj4WGN;piG!L3yMn}CeX_idQMjT{ZHB>=v
zaM`$2M&Y-zlm)GseN$_XLXSYVrRGOJv*Xu=k$HT>ntNTp&TAif3!_9on+s)v@z)Ug
zykw^+x2^oyHRE?lXqW+8j}&X(XS7AqvI_NJkqOy->yf>Mw-7_5)r!yRl+uKAmVeQG
z&=f>ce>cW2#kGbiy%pQ)d6Pj1L+ld`Lh*Xhl&B(nf#1Dh;TI^^<6Kd#(Xw^##8bdD
zQNK@_6Q^Dk&tWz%!#Tf6bT73WUe-@XE?)<Eb&@%^&`Ql_)GKqH^&!s`6*7#<G}D-x
zM$w-p6-cX#nUOU){|qw*a{zv3(SMz)tnO&2BIcYP71OO+a8P2lb<wyR<E?RsH2f%&
z%-pY?DbjiLD$5b3akiI;8<?lKbJOdY9$HI3HM7o+LmM^l(gcWmv?0vHTdcCXQsEQb
z)UftoF?BFG!L>naqa`M!=WMg^3{5!7BSv161Ml6eM_0!{#-zbA0#h9P7=JE#2x3EP
z9l$^mi|J1<%Pn};p0x@Udd;)K4$nG-KY_KL5k0Rr1`%O_NFkmSL>grc;UYP?=<wW%
zW$0`!y;5&m+)R_zvc$<v!9{3Jpg{4cd9n~RbX>e-0<g*p!<<;53U$2D)5a*wVW{ax
zId>4a=#A22cu7@F?zJ=HaDR?1N?FV<(W=&zg@AD5nz9m-RuN9~^UAS?x#xncx+!CH
zn-e~yoQ~4GE@2~7%KckRqU#v+N&!}PJg-X9Ow$CA%2Xf7p#jYr&_Q)`vX5aFUtRP0
zdRQ?cwCzU*d`d9QZhodNo^fI~hq2sKmEOS}XQ%Y(eT3GP!p86Y{C_r~NN<yu3nApz
z$m!3?HxuCa57E*}XvpXxpob6t`~K}8Uw;LH!~Sqx=_Js{gEiC7_=shKWKvH*;<F+s
zwnZL@rI-Lsn34h();bT&4PU2&U>aw~m{D4PxK=D-4HmK0V83mpUB@&*)s%Q}6iaKI
z??>I3xyn;`rZXGv$$wFaW8EW;PovzU`_#{=D~|flboe$MK1Woga+NlQLBqIox~L6?
zgc($U2+$mAWK$DUgfamlp<*@a$U<u^Z|RCrObLz>8loxSuRL3pIki-BrzWJ8D5R#$
zTGzgSLnkY?WF?gtKwYFI8p)sldO(H0ib6y;AY6Wb{Tau7$5C(9cc6ds;x@K~RMU&#
zmM|tAEx&b%L3%Ep#qOzh4${%ec=S?XmgEO&>*@&Lbf2e+ogmjs5$yBLU;yR~QrrMW
z@180U0JQ^azy!iVP){<%-c>G!8E<mDjVsQHhp6(b^>B@k=q!g>Ho~>rNact^H>xWA
zO;uIpoW677iQ~A{SA~Dz9H^=+S5<ktY$}CQm8P9n>IYpa9T{seAI?R!3x0p#;;3sB
z%&=akvK>3-$@7x+xXnwr5)#);qA1^-Nfh-CCh^~??H0q<OBJ`g4qLjGm~xG&xGr^Z
z6>4R($#_$m%M}JhU5*l#E9Xm7<)-xAG`OZchtR^3b!Cm&TM&OhPU!1&l}5?@S<RUK
zN`+C>_LAL+6EhZVdY}d$=eDJ{Pp8}gjy2=)6y=1WMXa7rp+9s^QF_l9_uiY9FK?Cp
zES_O;t}$3=H)E=$D4Q{5hkfBdEFN36#OAE%2BtSTMuSo!Z&I!oc(<`Ha8<=fc~M8g
zL5Q2JI_?qaiPe8cMNod1w?Z~g*Jh;Y>5TRLcwiO}tT_4(Wq3LEx%&jNqqKek*>%p9
z8w9XYz{sG@GtVq1nu%u~r%LXU{hsoYFv_L2y6?^BPWjH8OXG6h$(0u_gNb}LEnmj-
zMrIh+A;tajNLh3tRVIuV<Up6F25Q?UmhOpviu2I@CwzYl(_v-%dEvT<IKwoN@qtsf
zi5~da&jVIS7X4(8B<-#d?4T$Cllws^@c=<g2&fn+ew>F$TB79usyooVXNfQ;m5l9K
zA|nZYh9xq!sywFj8i=8~QTol}=!HuN;mSfsaa-{@EuOye%jymjaJ8oyI5^#}E!<!>
zs0WG*+LC|NzwIs0%TCuFtd?y-@5G#oVbwAADIWe8M-emc<zB1`17olKxF!@@ZkKd6
zq0sgoWAgp>(^7WF$tc|DW;74F@BetpO-oG~GfPZvI>Pi62zLt9-;(G@t)wBlMdmIc
zz-09KB6Am=rJ9u)mSuL)4eZ}4vv7BA{9e<uWfp&+35;gvWOhz?a1?jAQ%q(J=46#S
z&3n?-9%?GS$}a5cy{?Yx4Y!{}3=r!MWw!yQ;MwxC6+GMC_XJL52z~|}7UtE8Xs!F+
z^1OfE6n;B-=QVNg3!U6@drck<=G<Mvc((F2M0e=kdYHOf_ZoJ#RIlm$h~~8*Y$#sS
z`4NA;Yh{Jtnzz)hm051D#-mz?-?nUmJxlFcsyp=V`uFHv%g>hLwY?wFybgyA)oXh{
zqI;dZuv(~O<?G~@+iOy3-=J9-uPNj$A-Qp%*^^q*lgr~<V2LZL2=bA+lGR3dy4ic<
z)|b>oA*GHYJ-29tZ62_aP%G+uy1wrdZ$5wRzi1xJt)srQNn<!BP$MtZA+oHhxw-p2
zUIk`(uNUtKul@RE>+Nwd!ks48N|LJAw0(2W=2gO_rM`vBGA~OPy9KNH3|M$&o>gRz
z%L*3mnu%_%%!f^~{qo1{xQ^nI{IT}$=O5(S&xdulp-fMC>AwzGch8IWy*-x8Rk%pI
zV8Or)*W4f9EeF4SWt;lGTt`{>e>K@@te-sr@Wuv{-JRRZ9e12VSyX*%{tu6#TEU)x
z7=}E8)p<*|OAV?s%fA42KiglI@s>0c0W_CUlNS`1Tk{Gh3(Z^2a^p4(zUL|Qmu;>z
zAb;7GllI`)mZ=Z_AP6XW_;b+1k5A|Kryt$`hoGV}!4B`=2SYz-`fD^@^lk{oC=X_M
z|7-a5i`LfD>r5}y8h~DbK1X_Yr)MycRv21g=oe#7&jHPsvIb&(w|Ony^I9<M8-4eu
zXVga9eD@2z2;YR$Z|^@n>B^{9$>)#xO@Ds-_WtzulVlsgs)T0v{jaC|1A<Wy?O-t|
z9bow9kKy~%&k}Ybrs*E#@*n#2&!2<1@=MiR(xf@<y>p%Bj3#yB;C)h%e=3@Dqya~|
zL6Y1VdVlF8{NXK{RW$I#xsho0Ye6JMlN}1u6>RVFO!`3xOa)WrJM&AtyHNSJ`G4AV
z{9{rE7s%uS1?fC%B%?bL$hkrR$rtpwP}Xm!XRot2Wv?mUvcK?JM{ALM3EkGOS&sSF
zjO0i<#JVceel7H|T585wL5_<nzH<Fmkz+yDucMwOeOPi(3N=r5#SmE^P_?Nr8;k#p
zHtepD4Hur#BU#<qWwryVnB?M7Fn>1iOEvJhQTqn-)+uLFYs^vfP|)!0c3~VSNKqnh
z3FD1QI>^<q;G;;gV;Ze|AchPKW(u6{O3No}z@Tz|ViS6A${w9RXAdbV-Vnnp<0aV0
z@tP!)<FY%#w@OOitZz%23l;@w^R}sS3sGSV;)Wc)g^Sxca7I}t6{$tqzkdafd_gZH
z!UKRw1(zh?*LC%HRJ84h#bE9;PVg#(aD#DTQcjA_DQfnT#^Voc6&wK?6lK0{tH`O=
zUAZlkbR<9?z#b<OiM-gcuCd4_`y9DFl>^>p0$nkFfw=B&CLpCKBUewm$N~Ko!?<LX
zM`+cD={n20pu7W#u(#PB$$w=!s-V53NX~dDpID>NFEp_g|23aV9$e>oa{qFXitC7z
zj(pWe&ib})A6zc7?q2H-aejPcSJRUx20K-ny!|k9x41&}Qi1r9uE1vKDIRr&y7viG
z;R;kVB-S=9DrH1_<$~%hT1AJY1>aL;e4W18t@6(0{ia(r_4a7dmw%=t`dHV(sG<(7
zYmL4w4}1HB<Zk4E-ZAbX?^k-8%4JH{s~A&P`f;p&tqN_AN@=_Nt0`X~?$>{yXCpR;
z6n{s!B$eniAUSwB9x4KHBWt2pDJ?XR9q>SH?u=6jS_7fyGiyBzgj%O#8x0qC+8%V@
zM$3cg?bZQQ7Nk6MRDZu%3YM~3=m761&r9_y<ld@&qoS8s?9O*rzPE{-K!t>o`r6rl
z{Wwd<1`30z$2HysO$c+i)$ksZ;+6wEuE55Wj}*3*hqJh=+4|ZZOfAU-qCWT9C^;G<
zDDAqMWK;w;q;JnO97Y0t>wHhEOjz;~=8ei}Q5$uwsj|_n^MBqN?=&8#JNu16yr!36
zl}@g863+Qt(A?AaV7>I{dwSXCWqtpt`k45vArZG6dDP1VT~MK6pery+f8}=hYsn%I
zx8$!orlh}+>vld;uEe86HfEiVM45)Fe6gB~r;Rj2CL-%i)z2eRMQz(6BxTKh^|RLP
zuGpIc0Mw?7L4T||#^Nu1!*-;%H&TE|XCyZ_EYBR(Cp(5?%%g<Bhj!23rpS&wY8dy{
z%+Kp|#g2y){?=^sy!DTmGFQwk`FiKr&8Ym0t8|rYOt1^dtn<p~BqckbtGuTr>m?!|
zIc^<?5J3gYaY&9Q@a9v-A>e5-+6Rq81lk9ULx8GT(|;U=A2$vW$d^53Z43h4802w7
zkq;Pk0KFUUA28|=NS8+)V)31$4lrI>9X?>xA&_@Q9b)w-jyga)W#^;y!C8ku`_y5F
zpnb)(L)>wE+#!{DeBL32IzI4_h?gfGV)en1#|<VbGY@H@OGA&_ec#fF$v(8%BM|?C
zW{>3Hcz?4;B3^Fxh}HLW399Q$y5H`Xq&l-eEYGpfE8dSB$+y5<tE1x74^UC<i9F8O
zpsfN8R|*2Tf-;K$g4-KWV3GoRtpf9=A7QD$V_ATX-SFs!Vh{SM4aJ`Hbqz(KgZmqb
zLhj2A#eVUDhT?4^bMW-kuWu;YK(X?_YbaVOC4cdu4Ml;t-cYoTjN;&f8j1par*S5E
z`7axa7Lvk*AJkA3=zDtE=H*iwiq=Nu9@<b8h&v5MN&j&TMS;B2P!y|o8j98g<vygL
zXth`V?uMd^1G$qWkhS)op<1gM_RY@=k5BzR=Q5W^-?}~+z(;pIvog6O_w8`(+u`)-
zx_=#-SG(pg8+assRX5XMo>9!_pW0U7-lZF*LEbYMKk_dlk35`zYh;RizxgsZE9U!<
z`<k_fh7+wRNyWPBbSRXf7{Rwa*Cy!^4NR$Ov);)Zy|naKG{(O!`=|U1cJ!~XP+c8Q
z#@AA=bjMECewiN{nI2CG6Zhr0n{5rwPk*eH{<5BfzF2R!?>qSb7-jUl3GMd@H@ylS
zr45jXrrDg#4o{9!2qi#)0MkPtm+4dGlGB!?&s>wE6MPCDR`sBPew?d)n5&FQt9toN
z)96EU8L$^|kxu-rvQNRXx!vqaxpIclWu{%o_W{_B6{c^Eo75|_;`j=ia$fV&41XCy
zi<QHz4l`B*ru?4-T4+;8oKO2R4_t4I?cvvML=w6&35d{xrp5`avd7j=rM)6|-NOs*
zj$Td6C3u>OsF%JQ*Q|_lUK(jz$=!-Ay4{ceGX<2zuzbjAn&I^1e02bhaAZ<EH)_76
zzU*N^i$P+`{A)!$cT{=b!Un^)Q$DF#)sY?aRqStGl%H*83YTYIy&*48Emk?hPLH0=
zJ6&<z6fXAJx6u&K+RoeDwGXF>;)vhBy+8d6H*cL*myv!nAeSpYIU1LL;VlZ6$C3mo
zf7O~>j~urVh2Qln2;{|iFq>rY8W;wAi5v$OEJzNJL|6!W8O!UecZA(p$NBg9s(aYl
z^~~tnTEGJAWlyuo)5WS&r^s$0WaCT=;*EFNG~bf(Nwkqos6`va#Nf11O`2{slPjG=
zsHXa+jb>^p+DN9h_Mn9r8Bh{G#cI5de_n4o<3p+1V!G&qwgU4-?H9!OJl*Bl>h%fX
zZeu7~VuBm*N;Etln1kWD8t(!&76l;1gzOZPR1-?nIi%^UXe&A+PTw;CNW5sjn2m(#
zZZRTm#4NO8VUEHRvGUfTBqOPgPXr@ne7D3XHS2FFnCP62f{CIRog+moSJb>If0-Da
zj*_y8X}XJDb8*a(s)=>F+e~bt?~$EJZX7v#lfpQ14kiIhTVfIqYb%<Roz9VyNoBmt
z%cPohjzYiYqP_?HTGZbdG_{H)1v6!@b{1bu4t^Xjn;gT~kKa-n5mb|NQ+zy@>v$JK
z<d$?jnlr^UeUIi%(FLUy3<gpxe>_&A`cY`nlstM1K2}N^`(;yV9s3ni-H4#6Qzd;P
z{HzAGJ!@ua^!WiTkEg{EeK2+O7r}IO^cQgv_fwA;m(fjk74CGG;YWWFy-cGG)CqZY
zZ81^K!W^D1MF$Yi|4P)yJfjp{eQ`WiMW4VYTg&61c(!vzp%N3`O;#l;f4mEI?5Z8&
zIOZ5?DdSv%4|!J<B`f1y)k#FgrkYmp1dnN*Hg(vh?ku9iHno~)d@zMRMhr1))(!yn
zUOkeg24AS@!DpWhKm70IYIuD7Vl{lVzTB)XH`f|#h4_CmJXu|@Z>~;P*HaAZPmfk-
z=f_{Je|zS*%-E^4aK!Uhe>{n8#f`nbxV&7C1Mym1fByM{XCsNwo=9ZdD7{g8qiN%4
z1~v{fJiR&D=+$@Um;W)tm+Pyu)zt{V9SwgQz8$`L=BJA%tJBRhR>dY9B<Eryae_T;
zA+B%>3soHrUl@jeI()PKVLg0pKJnRWC?)kdsW^8j?a!a{@54|;e`dw8XB4U0ir*hX
zQA_eHS5=m%P)PQ&0Eo<_kV^*<UA0+*`2dKB4??PV8le~+ELcAr1Pp2$aha31jbJAB
zEFMWz+2rExa!7(1A+FdW;S9wTw*UyKHsbeLnkBSn5MWBK?1I1}i1VU(0j~5Bv&<mA
ze%7W8`DFO#zy7T$f6^+k76{|==HlY$K(9it1fvFvm2dSlMBc^v>gmhl)0L4i5D2Kv
zW_5LG!gTr9-!|Vo-5hUL#!o*yBvzT@p=d%S=yjHdaEjL?HD07K#@loO8$+ITipAma
z)%x`5YV&M({Pn}(ht+SJqg%%PJr%^XLj~I@-jNDoovA=`fBqQnn)0`e&>PYlaodo7
zOZqLDZ>8SIy-|9j_D0(_VxKhn4<{M*r%-XA7zy*qsUx`!x3!Y|yE)YR$9G7!kVj_}
zwOO^-Arxp9=X{Qc19CE{zAX^n9*-tdWu@7y(#+Yc7$88h{XJQtgW%7YVLJ$Mg#CRj
zM{v=i?Pe6He<fSV|A$a~oaKPhDhf?XwhdP&-C7~#B1tVman}?v6%WV8h)R**fa4cH
zA>Nk4tjSrYiUl+(rX)3Zx85@fgd`HcYXM<VgXloGc#kQNiKt+&)dg#qN{kC)Am>J8
zr@AJojpBp#@Vlvp+}x-I2x+7txj1{bAc%wzs8mRJe`=<7Q4jUIhG<O%WkU;PgeC$y
z=LIp4f(xS7SI7&YtoopaJ%-TQiy}hsjF=L2Kw1z)a8|2AA9;VSR2}Rh$PgHk%w?jO
zf;38ALWxA;GHF?h89YnkEgnaN6!<dVlRS-{NGNSa0c}|=QV*f{DDpIpNi@C!L5NV!
z90c(?e|f5ME5=!F6~9l9-RjWwm>PoD`51cajs&|S&OSho6~Cj$a(f>=R?;gy7Q4hs
zU5};tmoB<e7hS1~uGB?W>Y^)k(UrRB%3XBjK636O=RR`oBj@zt!mI~)E*`pN1sZ~1
zh5tzXdz1kAw~;4>>OH$uVZH_VHYc&phne(qe<s+&F5#A(oR;WBL?pJzkPC8M7C=yX
z&ZdaohP{;+VU=2_#UXDU(x@(>M>Nmj4~d{o5z0lI7SYWpQcv~wgd&uN3YtM+YMb(F
zKzvlMV6DkIVO|hqfDL={PlMomj0LBFeqBWlEC(X;>ZDjwpzaYA6VXW26MIbIeUv6>
zf1=<h%2|;GeKYd+wkc3OwuD_%1XpoeC%P(0FjiSm9%>Fi3CxOw7qyst*A!^NTpPlP
z%AqA_vn(o=(PXgfR+C7MB*#5US@+E(>aq|YJh(`Zb_|lOvUtZpT`7B{88IPBRS6IT
z2uV3DsFZaBPZMb`83I+6YkSlgky{W)f2dPLl@G2hDU=bZ*g>gN&kVGZi)|lIAkO#~
z5tPA~Si_PaoY+8w2C*nF7R6hUu%~vCEFkeJi3HzBxhx05SFyUS-cjIxHtwkz5LAx(
zvN1XCfL}OCOYC4e+h8g3W+s0U_O_7K`6K(PNW@{mEipn9a<3=j-%*Igz0J1Wf1O^F
zO!S2W4okj8Pux}W-92vby`zT2EWg)~GKtZ5)R1~YG;7G*)nM*wFn2Xr=SG)>t;@pJ
zWnt^GuytA3x-9Gm2caJvgnn=k`oTeH+mZE9Xxs3;5qd*<Blbq>joce$fxSWJtP1EZ
z{!qF&uO62Ln`64CZ6KIS&-b9le`&iRG+Xs-kKg^?4H)=KD*=p);1Er+OYHqSf?)<B
zQmm`#FbEOYD-}#3wUUnO5(g0C1F?(DYGQdgBPCgU*^tC+zNcL|-F?&3;f`U_Ir$KV
z!iy6d5@DzWltm6=q9#;T;#&+!)FtA*+JQ6$OQO4j)nj;{vV|`aIVGn(f7;WbuqcXL
zDtZnqIg}+7a8swOGYG9noRfnnK8hizdTmvQ*n$w<k&Edd2r9&y38$Kn!9}Tep)*ST
z8eoedk`f_-`!z(Q_;Ii7z~G~U+^a_f%0eG`armi`Wzi2dlXWQ-iHtqfNNKg`*CXgt
zM<In&U0^kE%PA};5Xq_Lf02VAP`=3p`~kA`6wEFza*jeQh)@jzh7=lnLij2eVn+7W
zWtmzCx2Q2Dryq`l1uUWa^AMA|26YC08LWQb5JSd7`F@sz8pp!gOrH2l382BEm?9W<
zD*;MpbWUj)3R|#ViyWQnIo^zb8i&6J5j1!mL7I#6sAhEdGP~8yf9ggbdfH0SBQ0mW
z?iP?U+@3m}@)>`P6lSRcveqRw3v`p9q&5mEOgu&ia~NXZCLi`Vr4N)XsDC_1n$K+k
zr=04AHtyG>lPC@SbbLW+f@ImFW%SD&{Uirk(=U+-rDaYJO3pJ1-85)&7)5qLwRKeY
zMC9WaewoioVvaM2f7mLrt{nssNf%Uu{Bx<rM*iJriT(3BDk48KB$`$eHb#&<9a?xj
zzKIB$6pkL;d5__XdHB=K`T7z-|9JAP{`bjGo6XDXzYN2R^UY5;C-!vx%kcBfFE2km
zTm3qGb91#mIk!Jw3?~=sli{qLu1;2Qb?VYF$>;3H+I;!(e@EgVtn#VU<CC-FQ`g*Z
zeRVoK`u^MJzua7G&Y!POeg^URuPX-n(|7*U@5aS{FjtQHncZv8DsFd}za6OvS2Dtz
zdnU?N{mkXfJ!5Gj!N0kukfBM8Z`0H3Ie7S^oT<q4J;|MVc>3t%UF+fL6O*wX)=&PX
zk}~wtRC;MDfBQZ_d9(ZLo|Kkw<a!$@RCQyl5(XKe0M&kw`~Uss;{5pgmz&|y@#ZIt
z_u|XrYxUh%PW%3prTHqMW03w4b*ZW|qrOSzZnx^VUX5t2P(!~LNl?gIDo2E5wd6pf
zd7MmSNXTTaxC}}B{pQi*1J-#98WZS}@Q7Z<F9Ng{e+*g^hCq0~$2ux3dkfZZy*WO)
zSPh$-&HC#6_~LqavD$3TFJC<8#?@wZ_WSGS|K%N@>(j$smYP1fdc%I%-CkR&(n`O*
zP0v)FB|W}PPjrymrN2ErRZ8PuZ_(2S#*FQc@;RQwyhBynL+#g7m84EhTU%_u>pPr(
zZgqHd3|5ZlbF;Dj4}|;(-j@->G!>TujyMvRrDY;1f6W}Zjw8D<->;Y-SU!?fJQ^4b
zFr)Wmz@f33;na?t`2Pd*D3ZmaThv=E>_DqgH;=klH&?#>>rKA><=fYfZ@=9E-<<G9
zIsNwIpKmI>kl|k?!$WJ{oKnJqdi(M3x4*wjscratg%3&z5I!#a?ZdbE_zud45h{#O
z;f+!sf8QOd_dEtXed{<j%(>+?^gaBxAKy_bspIU=@CoZH|M8C>fBhy2(DGfO(;$>N
z-mF6r)+Ok^r0HhW(V&G3==Gz+{$bxE!fXP}7(W3byytIY;@cy<#UuO*@Ees!nScid
z_>GRw%BcJ+&o{R^_ls7u(>Pzpl=WnMn%^dfe{OAbq~>8=q^8VN)VFTFSMXI=?)GyT
zihz%f+=y@U-k#o?xDQP8bi};=`Uwi<+3NQo58lSnjqXQVDZzH59SrS~rC{Qt)k1n*
z_$a?J*M&AdJe2~&$Af{l{{%UhpZ@JHLF&C)N<El3X50ZPoGTc!G$-60QSqyj!_zZ%
ze>kDNpYRE*$@|B54e(C;2JUO6)33m1Y!alE+K|vZ_l0!Q&I&JaqSI1J@=1>{rq8WO
zsXm4KBcc^R-ytJ7B>?R;851c?BV+l?U<M(9?1;tKKtcOIBXR&oINT;OJj7&0@Q-rt
z2#p<r8+Qn_OSx5Qt$~ZaMjgKWuD<R@e<yvd5sKdMjSZE#h&zl*YjDk+=OrEI+v<gY
zc&{sOXzIKkBvB{~sPjo*<bBH<=&VAeW3Y*G(DwxSiM}T+PV~J-eWLGa9_jm^r*PBv
z;9W=84;}Ttl54FF;dn?W+@3g~u(KFzO`PyaBVTD14k?01GC~MT8f9)p3c+Joe|Jus
z4r&10G49^fo9V0sF0I7Yxp+0*01t6k+@;&|R$^OYA)+wSUDL3c2&^Uow@t(bg_lyV
zywK==r7zR(a4NNb%6!Ck*;el}Lq}#rV=$mz)Lv2&Kx!<CZw@Sy(#0Yg@o96OgSZz(
z$kbi|Gp~_U(I?GMtCuy-oLk!Af1$D~5T>WLuIiYPaTSXWucL`wg0DJ|HJmDHnRYle
zK-qz2FfD~mTA`0>unW%V@Y}S$A%$W6dv6G1NJC?2|7b#r@Fc8A;l#g_@^7I*IaqDJ
zDh9C4LoyTTK)N=v&g&uNF87V$;->v<Z<q`d?h1=5qhG21T#)a>j#Sr=e<9yeX*xiF
z<Kk7Lxl!06dgL=0=6zVd<=5<xo!PWe@XUb%TI8LM;TO$t>~B*K>U8{f*+`?w)}|T$
z`~Z69pYR>fJw9@cHskTH5O8%the-`~fbuNHS|PF7$l#dUtQ~?zM42>Ncod&}k^6S<
z^&zHYOLF}pSfAxdmaF(ne-Ed(IwBl3S4U!9(=aC{qVw5KDmo|(^<`$#3|tS}aeXx~
zg7#q;CZX!!EOa1^__9;#LzHFQ$N+tUb8>A}PJ}%c$HeZ4TC|)A4f4H#Bq%N$hpxBV
zi^jkP1oVi&nVcC6F`I94MocKO>&+>{%iCgz6PxpSKlMREr0A3ye|J@Z5?a#Cv)o9C
z{f=?sOPzWX#^bnyzBfL{4{$5K>E8ed=_aJ*t&(?Bb{a$I3}2BtSQ_E<jE{OqYLG{A
zihmugL203_onZSUy!~KA%%7q=SPZ-1E9jm*QQLc>{SmRf=QirQ>K$cck`&&^&hm>-
z<t~?H_V#mqnq!6pf2#HG?EO(=@2Y3wQ|?sKm}ypZCyP_fV}pBhp@UPXlEzPdd(z$T
z`C<Hfw-|YViDD&0&l~Fws1q-oRvAOuG-t;*&+}t4rdTPOTgKWUS@K!;CI{exP_`v=
zYWiV8;uK6aEGtu+ZKEA7QwDtPMj2)gC-Dg3u5!)o+6ZK`fBUtQ;A!z1u1<;arxrcu
z9PgcXSEm(nG2d)-S#H^nP0VV3-Rk6Y{#hSNoqL!$2<sp#;mMr{=*Hb!Q0lv!UA4o%
z-Hm_q!W_=Rs?KJl7S=WQ<M^n~jwzloI&OW77mj@#9|JW&Pi|NdrfN>`;LxHk)W)ZL
zZ1m?|!lhLwe{9hQRow&=x`U-wamt}4jN((>9W_;tntmqmJJf~GLu=j2;ck0$8uD|s
zD-No~xU!}s*-p}A#=Pu+*~`qZRRY71f%>m=bt#8u$lMJ>>G#jLy2=T}V?-cZA2VyA
zl}<#b1?bLXa>54^(jhaxi60o`C4-}k#doHP?xa<1e;HrRd|tCprIS#Ux7F(1`RuBo
z@)IU(wB`eQQQZ#HN{OVlGDgpMqCHe4tCiuHU_A{X@+y>v>oG2N33J@^-GHJ{*d#xx
zu<1593itdjhNEEh`;b$xa~-*ef|UZSTZ=+(j|P@ymyc<qV%*9p&!`gve&^Q072@13
zG7_Uue?Diwe0G(v_<nw1<icY0s}(!btYN0|NlY1{eu3e{PTD;0oA}c05^1SfZHjiZ
zOlTda|BwXTPRLHjSn(t$%kK=;G|VZ?Wo=SfaRjY|3WcgDE1XK7uSwzO^jW{zEA@5J
z)r;j&Z=2Aq=qxxmN7UUt;bK<^NG^e7Y0>Qzf8A-+!=zSOQ;&%!cV+wOEmNEHWDFVl
z3}_f7r+xGg>77(TF_1}loLu(Xuw#;DyVXN8X|g`Q`CL$Zbtb^aFu$_a{L(YgZ&rM%
z6m}1ViKJIvE6iGJ5nvuM4#I4tT^pca<F>0wW~sc={L!7H;2^6>b(tZ8pr&;o?m#)B
ze}eH3y?3x6juGiDm#-@DokMVNIg=!tC9cBxR1>O2>zkb}JQ~%xI0x@Os83eTVG*iW
zYnNt6s>&(U$>YFUBGXgdMPH=QbDxJPTD`7Nt!_EpWP#(#xxRy^`?Q2aeY&9^SA3w^
z<Y$?Z7%8!k)ei;oVTKrEm3kt4K|5`0e;*1ahQP)IqcX0-gyA{Ha5flyS=tKnE}%(_
zBI-3?Zwwlmho|c=RxnAk4+fB*YGSv^k9(T8k99eCfx6<Q?5|bk{cevu93ChXf9+hu
ze-tCmn@Y-*M-4@3%p4eOc|6((^ln6sN#*MrZ9sT`{5mo=Cvor!eu6ycjc^hse@TlT
zp*o4;zN3YOEjaO|KwE)bcsE&E4exz8vaqRP)TldX-YQ_!z)hEi&8D?HvNK=#$ukkU
zhE;^#Lx1b`JB!jQG(|p2O$6Z>L%drDJr9%f&a495@Tv;1$FlU5uw>bQ)w(4dLR4W?
zXSl1;UxbiGuNn{hg`s9a<+88me<Ve~jpN|T>-RA&Fc<~yJDw#9*kuA*qe!^FhHMIZ
zSxuq^?1>7EYDIVUsc&=H*r2@?TGtCCAQu@hRJ)!r>Y;B55O_$B+wKnOyM!x_ol0z-
z(@v!_s&Wls<qjE^ggH3AlhFrOz;J9<!}}O!$hsDn47~Dd0C_h&jb$mGf10Llg~vi;
zyk5Inv`GS7tjikSb+*|#Q41#*r38)9|1v`+CKw8&UimJ-v0TF!iuj59(`ozJf``_J
zILy(tvL9-+Azfngy0HRLNNLA)%@W^h2m^Pl!EvQ>s;kW?esn5bv;xR^*+lVvMqnM8
zj+ate4OmSTR`tFrjHs#Pe>2b8O@$8gu`37-dXtJagEgbEyvy+nlPRHE-lW>Jdz9e5
zmWLu|$_Ke<sX}TyO8Zg}pu#itKqgQIS(sMKcA+M)ySBRLlFvySQ(Hnx2=|S96!kS*
zIHYWEgC76nrQgmrnkmiI_mpLu`Bkl_BD`cf0i_JduD487mojzyf1ET`<%=BVQWr66
zm@nLo*`IoyFj4tV=}1{lBkjRcgh%U$)~k-_dOa|82hvKdCPORoBxZiH;mx+yYU1={
zwP*a>VGZmkgDWatC2Fys+twe7%=ENehaB-($%FD<gg~0>(v#%Ra<-E~c~}+4!MJ73
zvMUeu3AkIAQ{0t7f8%&lM&5Pw<BeZjH!2Ma^<?t~Tr*x^&1)3;vtNqCP_0Ng^Q-|R
zop3W*@I?~RKS;xw<mII9?26q$Jz!V2CQ2-@C|&)ZOslXySjJ7gmIUL2?_4|CiUk?U
zSzG0>F&e2IR_;0q!OI-N8tPy@^kJ6r+eHiw4_E!+xp5lpf4Wg<PNSWt+J>k}@q+Dx
z@@m(Zv@Xmx&!*1q@0na@<v}P3Bh~ctzGl1!i!=StxTLiJ3;nHxdMlQ-9ntOA%iipA
zjcUrVs>>W}-m7mXt~Yc^dawPSKqe7bGh*|)i^v4UA0Op^GO!PtY0V)dsu3V1+lESB
z3q)X}w{~LCe`RBG1aS*K)J7OR5p26xO7W1FM5A4%UN<*hVQYdo+5pG@4Fj#R0=%0U
zxIRTn3Y0a+xXEIr;zKvi-tULrX1DTIShN$I`!RTkq}`~+Vf9}T0KiU}F3)9`ujxsR
zA7UB%K@0)B?GIqS*nC}bk|2!IBTe8sCPEo(F&yJ8V^wilQ0OcNlPPuUuzA=_G&B+c
zJZEQ$#2#Lt(VIvsrG+_QnAGCHK1$oU@%a2zuI*LNsIA;a%SLODSVW&(TK`P6J9!qn
zk{V=9Yx^9}VuQ_LMxKquJNo|P$G86i0|TFIm+_V~6ahGw;W7pk0Xdh!j58>INpCAR
z62ALa=nt4-k;M&o5MV3M0dh;=+h9(EB&QrE|Nk(>rMUH?mMpzFv@La$#oAf?swGc<
zo#g44m-o+?pFT96obU#mI(`0m0(v0nFG#v+&B*}>tKszd_vv5Xq*Ru!ce+qYt?6>~
z+0%b>e%0V<1kebeCjjSHhwv?bkI}yTS8*)OxnvsZgTCAID@u?m&VHkd=tDUF`}5Zq
zSs0Zf*!VI2o4<YkeEILo%(g`kVBz$izhB}5)F8B1C+$g6JpJ>>>CcznW~#kXJdjeP
zwZ;re0h#7o8-$LJvQi9*Y)Cjmf*BGN)A#eMl}UX1h9oYMXa-1RlSih1doJHcnq|I6
z6=ksn(t`ERKYyL3JHP+ZBo1UC?`NKX6bnC3O0<Ed%Sc0+Om(<JRDb#ia!qnFn}Fcv
zRWxLYNgRo$$YF+6w{)qL5l$_1Y30otP#T5O>VO97474JU(We=kma(_<<g9RRp)U%G
zRen4O(&d7m=tnAhP48oWh?&Q9|8|ejN(<x6LEf~Yh|&wzFdKZqR{6fJ%Ctar+(?*`
zjx}CrGDe|{&k-6N-2pqV20w2)2{<nTMa4AZm{D?ui=UJ9(TUzmh}$e9;yGpq{K!)y
ziKNs}E}l+PRVYygO~L6ov++GASBjlTQhvnL6SJv$E4=x#2=iip@&tS@_Xi?f5T?vv
zWX=&uzr&rKgi%u^I)U0%>#`VbFTdB+Vf^0SWMvS_>1s=pcOA628O*dxp=U3fHB(iZ
zc)PYUnm`h|Dp}Pht)X;UW7lj&!^(Ktm&iBi^4jRE<;C6KYa(6pL;k5~^u*70GK*R|
zoLt~r9*!I2P~8@PvemTf%B7k@Xys>N5*1q&YYXj<a{1&?<*9jNL$nAoq3%sQYH6+w
zhbd1-4lbj#HHeCd`(SIFC5KQw7nB6(Q{Wmx=NTZ!M9B`95|Zi@C0;<kCkqfAh#?<$
z;<yH<ov5Ff5R}%en+r={=k&TrcaX=i-XK1K>rTz{v4%N+uVD)%mxEOMFb7SWY0lsZ
z7qz<ZO~c6!iJSCIm2h1T4^0UuNWIYBWBhr`k5L6$%`aBX(y1h=A9Qmma$fp9p|?<x
z^ARZnsowBPuW^@j;T|5>20d*UixtT;S3TWQeP-*HJ9{l9h{<d79r~2~#=M`Q<_VwU
zYWCAYBT5;6wkYP<GPTU9>m2Y=Dk@$PHRp;px8#w<in~56wQzEwml!?f!g*uYA2n6m
z;KM?*yYg?JY>F%?#}q9=%XXclaI0=>Moo6(yYhX^p3Qp3_jj5H6f4Q1ow7?5f7{$j
z+5e0Yl%>}7+?8>MPjy3CwOLpMM<+R$SC^Msn_uOBTT><G{13`;Kzp={S!7(b<+noB
zw`j?$MX12lU4nswtvxkIF^{Sqo9Xpl1-l*#yB-UJ$AVv>z2dR3^w?5@D*ZJIt633(
z0Bi@5%5BnwI=`kJk>eq2nBu|i$HS&3x9Xun*>-F0`_M3>Smk5TyBXHqw-J^SwjA1S
z1%@<#Rm%?D3=9G7Pn53rk)b2!p@bgoABiwnVH%`%<d|>;O970Yv7r1wJcAFz(+V4y
z0eCIIg#|y)IodOlbc{C@D>KSgvlYIaE8{M8yIu<jg5C(@)DA5174#0YB`=40m6iJ8
zs;3z(bnr8!`**~;Cnp(7C>hB$ZFgtQ;v*M-(lX|{b@<P)NU5gzn{TI+Eq`V!ITRKu
z;IKSv-EM!5+>!DgEM=%lpz?m2p;yj*5nN9i_&!uWH)&d;Km+q?S<`BDUADobWj>W(
z%8vftTZ4C=eCJ?o!lO$?UA{4>!}!@7^D~@oeExsU-d2;Yv1RJp?W>glWoJrz_C<<+
zYi;Hk4K{oPW|&}|J`nG5jRK-ijCMhI1u9q#Ms6a!wE*VQ=$f&NqJ-ErW$riDNa={N
z#u@6PVQul+!&>jqJOm>G6)V%KP4l{fEsxVO7cU6HY1<Abivagbzb1kPPv0yl+g8ie
zPFbWMwT9#RJ_U`bB>`I4Z4RZ6fz@JvHIZ}>)-7p?X+W=MEM@Pd<F?W54z`;qabq*R
z787)2@Zgudl{+>n6h6!ZmE72)N|i7ze;2-#PyZXZ=FPPtC#{h)SeHW?h^{${5$^4=
z+)*HEFtG|d$grkSTERT}j$Fu1!g<}@CgHp+V{f{tB%&FVH116VIyezlst_}O;|;vU
zK!-DQ^}-I$P(rZa47Ua3Q2VsXx=)cEj>(B;PRkhjG`h;&kj`5#WQE`H3~`6XJw*(v
zrOuC~Qg}cr9VEj3-3s$IZ#ZhJxi_ZV%UptRZQ5-BSGulKD150|%DQ&6vHNzuSoi8<
z=d^rEm?$1Sqb0nxvr&!MuQ!B$wWWln@F@*ptt>sP`60cER2#xk_C6yS#^O5f=$gX^
z&L4TSnbBc;sC}%Z7gRd0wvUbcvF&5DVdlr|Xe*tTAKM=yk%dA(b5UQOsGDlRk>z;W
zNgY89a~bOTl9csomYUw`{MPY(E-!a?^;0Qh{WNq`88@v@S(E%870ZTyD;?Ine%YVa
zePyGAtRm(m+x|TF#2Ug1jJdjH?{<~4^Ql~zy_vgR_DJuTB9u`T-(lo=uFX438f#k(
zXf8Y=>5lc45r(~<iK2rHCB}*IC$~(rq7!C-dDR5NJLutya@23&R8_>KR|}UEy#1HE
zP&jSsGXsr{7bw~a1?t#;Y#?J0sMRdTgOkR!6soKhR`q=+Sm`_Y)_2<Nq3M`;CYl(U
zqKX-I8M8{rFf+!5!mX_epqxmOELs&3g{2_=5v>YEr#_PW39X8W^5Z<`OEA3BT1d1{
zhT*4Gz5v5dl>SdJ{KUn>Vfec&Ntsr9_iRg!t9-?hBc<_8(WgLvzV{m4dO_FBa0kg<
zg?-^GfaG1y03bm5XFJ2V%2%9Wq!g`o3#5FnWx)vrW?sERSq4}c9tkU_-xt{MgJEUx
zfh2|%AFgr%Z%$mDFS}IboeJYSJn2BIqrNR>+v$Exjw^V_LwjZB>WO$I`tHy&6^Y8M
zAV@=mT$W{Nd?s6eKaur0$Ub3vZUWgk|CE!$<fp4>;qdm5e+te6mLe_Og7X|%b;f??
zRR+24r(Pl!3}rq292^J0_t^m4faA#Sbk7|HjybFt4*x)f-w_<}n);?w!VGZx$12y3
zAR!3nIuPefY?rsOG~!_dgGcY77aGksmJs*e&&uPv*eG{@O#i$ux8m${1T|3f8-J#!
zE^A}cX};scFh20U%paVOfZsa9?j7DN14sK?o81udsHM8bln?hf8!e0jR-KWTC0<=e
zuA_Og-MyO|TlIB!hcn3J^wmV&y-3FZlG7_<uTkG~hVry?hIXevF!wNd3N`6hV4O0S
z-L~~D^Gk+*as#k(`rCp=zj?~8VC=Sm;Rh6XR%l%3*$X`#)Ch0I`CISu<+<-(%g?Wo
zths?R^l!+p)$go1&_k$=xd-65R;8>zvUKfNKx;3>te>^dwV{pUv4f{v>j$MVV@(WQ
zZ}Qp?gG@v{+$3HXc#qPfc1kF8KUR;%t?B2matt6E6&o<|9OKwn%iOD)e1{FX{Qv&>
z@;|%}g;1CAmNXOrIg>Hj6PKmiG{S#<`s<fJKXF2I#L8I35MIVg>yrpGDV#;)EZW7k
zh>u1*q>|w+&ZTf(ZK#(CZ&c`mSSeqgG6YFuTp9x(5Kn7qo(KthstfhddAw5PS2yGG
zY=B$<zBX%tWYRbdG^--W(0rd;3KCwW#iub^r~Yxg&NGExI#Yfn9J=}WWq5ykO*ln3
z&*>B1@#iSFU!EN-X_Zo%IcY}u$q$)_iAqV7B4!ENvW_Z;x+&aD8$^`GCLUfeX<!^F
zt__ESjL${3TX<-pnbI?8{Mt&KV9Yex2*#brW1{Xr)UoH&A?kFiKk-e`pS0fe4gHC|
zHuWb^YIDy{chXo-kb&S>8k>LN)maqvs?bccDov(<Ux+qZAL`>t>SH7CWvNM_7@CQZ
zXVMJ1NO~UZOTayT8d}9BJ*R+}{+P^_;oAx*1#`lT2m#+#KB<|}ejFvvv^*V=VC$lg
zStyagHx`9O=x&70uhAqgs-Qx)DqryMNf@NG2Pw)^Hgh5w!BbMlj2D0Jl@Cn;_3TtC
zjn$1QvU#C}x8|5yqLOD7j;&GE1TBQr&-A9ha;@|rh1Az>Oi^4?Oue!o(!86dGeJqT
z`f&-Q@!6O~6S=J!0%hl6l*r>0JSjKqR&Xs1Q!rBzQt+7~;c@`4iY+6QZY&3Y&}hfe
z`<KIRYQPz$rI_KlLnVKb{faR_j7yj|Av6hJFT>NvFHgVDSfjjT7S#0h=Tms0Bru7K
z69i&}2IRNDPv4$CjbJJG;f&)Lro%t{_1pJpX!%T$oaab2H%9B8zZ_YiCc{0`w#abX
zs&~;BNp22VdDMy-WU;uo<5Et%QvvV%FOt)>d~V`B>zEgMoB@AeIXZNb0rnaxRfI?O
z>fu|w#uTLf)exMrV;*fm>^2LcqQiSF=~74}Vov&ps)R*9jL#~8yoyk}md`J%${}U&
z5;GCXj)q;zAcHal=iEbzOVbdy&WWk33R0GF`h)<=@MyBmZJ?bjyom|Ju=FFnv@$et
zxNB*Z3$3j~3!#6#!!WWqN|Qm$#%(1u*4$R=Ke{bKuyI?|&n#KC(toGBA;#!2>q!|8
z&7`^~SDkcH((-1=QwS4^GwEqPE9#Z;wy|2ZR9PS)gfURFyG0sby9CNCIs`~^oJG#)
z+h+sGLX9|oBz3%hX=ZD-gy-v)Cdv{0ogE<sMUuFDR9=5NZ;%}M7;O!+yk;}51VgXY
zkr`yIn05Lcyw?Ev9lX~t!wv5>Qu+q(rR|bSN<rX#UP)le%_Yo8daPFx4t>Z>GWR$o
zj`0=7mII{U-6=i5CMp?I&3Cqm{VZ>_i383Lw22FN-f0tiNL0Uilk##Et3eOIQ2<kB
zMs2d&uMdBBFo$ypEYvKF?Zcf|rmaP{CSWM8o>z0;gb1=0KT}J{JwrLFCV#2GTZ4!i
zQi;uw3RXiU3OrWKD3OW0^V}6OFONv0OWa=!Z(|&4P<CJHnrbwpfV1U~ZqU~R{DCf8
zdC$DR>9Lm*(_4E7%5-`w)af8#D~=Y>DTzv|mRf(;i#;jYu8@liNxmV@CCE2p>(cXF
zOJ4D9!6W<FvGz|`X1x6(Mb~zY<k3N1L{rD_<Cr`?MdF*a(cm(j9eMuP-1k&cj@`NM
zc6mt!9h@o3-9;9T$t-&Np~!-99&-72sj(;pOu#z?sXgF#h*A3qZiJ`<ng7@Im4Wd0
zDX)L@lf0q2(gVtIY~(UI=s#F-Maj_T0o9eEjJ{b*Y3na*9)W1zQEhpL3WoyEgqQCe
z%@4D@F`6H7eq1!agy((H`~WFeIAZj-y2l)Xt`vMdGOXf`NvZCNtI)ZPu1zS7dAAR)
z?biuAzD%(<0+1f_+W$<U83$o!gTXyED*%5VP=EU6U*Es|>FeJxZ=b&ZJB&cjh=Kv{
zrl08%B)VVe5$hkh!7y+Z!-)_?OQW#GG425uavob-zD`eJ;|ip9ZS#-N?3r>(h<1GR
z$~Jb*v$Sm)l>oIg=ShFGjTyH(OQfx!5}pDS8XF#Qd>$1ZBdyJI?uw)SF+F{oo<4sS
zzIus&LANb8OKinV4XlKgwYdpEhu9r`4Y9;q)WWsqmekTHm~lqoiWZJp4*09gS2<5D
z4avC)VPyyfDf%AUS6DDmo0vkzhG`!xn_^Z(1qI@4;%@!@^(PJdj>BG>?*!-o=#iB+
z(?J@~tZ@NWjJdR1%rdBZ?wzg2na+O_WLw{#kV(l^nt5rx4L}ivPg8?pBKFWw`)NOm
zfLSErHiyQrt8zwYgRv|W&aC8LnFwW=vbsIp63>mu==pLW>T*ko-%Ohy=Zve2Yd7;H
zY-MIm9wQ}>KAg^d_-^dp?RW`0b2tpU%Kr0q)+Et{qqYy6`eSEJ%Asp9Gx2}MGYE2a
z{7Z_K&H7~I!Uld;7^zr|)w)x4hW3p_@nditj-F|~dow-v)^k_CCMtoUbXbu+j`D^!
z<P%+4lKQgATH#S2_&7=2G;_@{tb}`e9wpIE!81LzQ#Z%u%HcBG_uN)gZ<q%=AojSu
zfuy55vvv%&Jg5PeaNbi#%PW8TYR96Y3Bk0f_4tKqi_);9gSL@1iGzAJO0RUx7+gMo
zR{GG6CcU4HSj#|a|6sug7V$XdBfg0(%?!TEc;=;;_hl2LD#bF=5HCXCOBt!kh@Fcp
zkBd}8If_LUt5v?;I4ep}H>8U(aNTCyD%=IK$5UwpM0gBMY&{~%iv@ox4oS=LMAtZ1
znWr94Js^3xy5Iv0J2=DCa?2Ug+_zV*FqT;i#Rq!+=S7ok7lT3&T#iVO8l9L>nV0m6
z7`U>Q=<RkjcTpi6$C#_yzSX?m?Tiu(_s3E?T0NbHhS#9m8D_up#InGM-s%}ND*PXe
zgW(8aq9!YA!njpdqLzP@qS~R{x3w+Qcd#wFUzuyCZE4`oYzx(tR6T`X=Z`1JXA~&1
ztwdpWX2u=Iv*z0p%lcw4S2D{B?iD^V;!DP-EY|0^*wT*AXfo?Q&}PZuLSnz(2uMP#
zH`0>zMGX--d6k)#P|Fa;s2Q(Q%WxWKJ1p^-uG<-YmwO_!-K>AxW|;=V^i2A<4Rz5Q
z%CX+%{KcsTGdB!{MP%fN%uttKNL-r?aWQc%IcUmT|2g&DsN8gD>P<T?2?cC{H(T8&
zNFnm=+ad>SAES_zvh3JAVM03>%08id<0SHnYSKgN=a&up?wL?_Bw>ArB?uwigW4l~
zh}fUtIL0VLd?J6=9OmKyp1vfyEF0-CW5CpQAZ$FOn3&xe0lBo1N&RvNEJc&=(9^h`
zNjXj<PP$P*uyRCK&CE=%P;Eg&>+;uyLhA{=t3zG4MlNY8gd=e3hM2>(9oLlQW9xM;
z=mnL$wNk+NP!wOgYLRaKsB)iX4ZU(xCkTT^>F%AYcW-~{Z5%~0QhEv@b+D^~RwDJe
z=DS92DfKOIn)^W?8wEO3y{F)`JYNx4tfH^8OG#FX==sQ2syH;RWJNEc(~ymv^0|}`
zyRLN=B!psKsdjs!R9757lE&+XN!A~m5a^ovH3ExIs(VW5WO+vi1%`1E?&hr*n}fLu
zb%|r~gwB7Oa;1^CdFO%B{uHzgX@84*wOkGKSOnZVSMGUkR-bI=Dd%a#!nMpKkIh+9
zVRGe?9)Mupbr%ZiH8(p13H3VM{3=~3p_wY1oXC{0%eITAE^;@NG^l0R79CpMwkb*S
z!)z~Kezct#s2BU15@5-SucJd>G(J9;b;{dwAFh8kV_;@}Y+w56>W!OpH9ldKrpM4m
zo9@nMysXtNJkixzc;l`hzm3AoKEFXwWdR9j1sLIX0u-})y2zEKe4#G2@?(?d^0sem
zF&nVP<tO-tt0tJfz5F+a%Qr9Z!R?${iEu>E41(BqAa}cLVFd$B)_cI|a$YT07pl4|
zjhsHNG@7j|jn_qm3ftHm+G1vVIlN3$FYwOoLQKk0AgU=*uLP{tGYseYIq%?=K;&`Z
z4}gXQqe(RW3(q3@XqRyXH58W-U@H`tG(;vSmtN>Jmw)FiJ%_(Oet(g~s1(tqpXtT_
zd;fU(_l2|VKntaJfB$$%?baw^g5GKT7-9GGpWU~YKRD4gmBb+e(jWZw^ZSmqd?md_
ziF%XH0StOm5|wFp&P8F;r=mBDS^<<D6h7&<ALRRd3Tly>amcvh)C^Lvs6qQm6yJ+>
z<%u**)PIFdJl8*2l%WMhX)}2%oS)64kfQoB@Y+TMFRYnKil)z!TI9!~)$OTRIrIBo
z+MK+aKW(hgM)TRUT;?rV8Xej}5C+Z5^p<p%beDKb{Alz@0OHZpw@`EmCRvNdNqT=d
zB*#u&(y*+4W|B3M)l}FW&2{+XWDXHsTS+lXo_~2MRGrpYMO8If&=i(_Pa?b}9%`Tt
zE}4yWP%ia+%kEjnB~!*a1_CMXIP@V?7soE|BpZ#XGp8}G&AzCg_I2nBL_52aANr^z
zY3!GdQ`ebK_aor7%8~f1vRF$N&=4?Vpc~>744C4cv(~RIcxQGo9z;}z+a;yK%8@;F
zJbxS~En*g#0?C!YnLA@~!t9UL8Ch0Yb{HEl){W<k)i}R4*CNBhI>VE#jUu4*XOnGC
zq${b95-G$eoN6HqgI2Cjb0)0|j#FJrBUYlKfcL&^!5h+V5K+%&!bNmB^%|jkobh$h
zW4OCuLs@8f;zTzWm^_~;v2#MX*-UE^dw(2{k}mO8^HJ<Fa?*gM#V1a`^8@rhX-U3S
z`C2?ZNJJe#;ImOg9imA7&Lmt|PNDj}ixr%p>tt4jmYkcGVTw_Z8u@HlS0|d$76mA5
zQ`_K*u~r1@In}(@p_b*5p3~@aHPXFD{$m{_kITK#Z@g`|YGs+tzxDm;qB`j?>3>?b
zAP6;fo@2GUji(|mEY@(Y?z5fNAiU=Kt)>H(TF5wSKm~1o2B?4z0{ny+6-C%9F2eRK
zI4EBuE#=T!%A!sBn$MfoyTGT-pwyOZ2*+DM;IQakLT70EoU^OYm~8;|W$HvePutQK
zN)|nXJ~)xHDSybKgO<b^qHUv=l7B{mQ*oxh+&?9mB&I&LJ~n^X1?JX%HFW@^KwQ5L
zU@TUc2<marNg5Gk<Vsrw)dZQYGigru7p!^_N+H;>R&z#E@#~z<QF$<S-z(hhCzDmg
zoS7!Hnr2Vv*)F5!SX-Opgx&~<)rYJ}-e+SFTH6_lY7UY@$@R!P|2Me+%~>>4I8%R}
z6IiHWdwiG^sHjDhGu|&bL3Z!Zt@rxOGNFqTYzy&A&R}#DF>20mat4rrc0x)KR%56s
zdM#p$QCxMA)&l+LoC$n4+H<-?EKkN-RAWB}92=#|^OS8KX6Wjq!3pg)NBbc4s*yKY
zF1s3n3&z2%6_GW76%j9pOvnJ{-hzJ!ODzOf69THKL{0EFBcF{veM6DLI9P*2v_Hgp
zKcSSU(1p#khO<+Y`Y_|`IY3(j2#_=0?geX9Sum;-&|ZYj7BuqR0MIi$1dPBHiqEv>
zAx>d}_p{#XndxeNqC(_1<7chNBwWpUU$>yF>9s60L*;!>@U|-It||}>kn?{PGr*ly
zU{X4!KSp)UDXkLTxMvujc5m1WYsOEtYvi4`tua(1ZQCvGc6~SR8Q-6VYQkXJzSGjQ
zi@>PnG<2~*#9}OZXBL5gR5?YT^@&#ek*c%mfy_|%f+W3F-I=CZ?R<f1f)Nm=opjPn
zekvb8kjVGE<gbDC48%_<?6!Z@ipGIZ!3~vaK0Y}Zs36XTRoF<o??GRJd0y@bklL!V
zL^<v!Z&;U^Xyo3lO0je<P-yh9wX{^&nrEI>7hGc%%DV7e0BW2U+AFU2rYpY<{s$`z
z%&FAc&OeM_E47eX^O@R4!8U%aLa>X_gi*=miuXgg(kxcX2{+iek3fG-QP@?rQ_prn
zGk6c%Y*?yaQ6WN`Cg^aO<*68j8t~BYOFilt53ypMlHASGm{Z4?3Zv<k4$WNWwzgZ$
zXGdUD;PjMhu!5c)l{L9iVct%8+VR@T;oL8hix$wCiLmI=G4SI@>!2E*Y+{|;kH?Lf
zA9A3|ZPFg(I{`rHzN3GnNPEXdjU}<tI=Nfxw@J>{yVDb&z1XF20P7t1Vqc<noXa^b
zQT;gbgoh;!EJHD-xt+{n<di+Kgnad8&)Af7KwDwG<6Jx;NjAUKSd&=I7t<=swQ}mZ
z{M4G&P3Uf$3bcwsh8bhqR@~b+C88Ht4$ZE_D2vVhm)wa#ug8Bn-SfI?<(Pic{TLes
z0{0o^*9lb60q0eP108~<6FKjubGt9=cVXzB4jXafrI4HX^+5!64ZFjD2WEnt8sEod
zqSRN~<%k$G)E(B=^(#T{e8v$wz!yrvGWVHj6IU%Nlk)%$UsFbSoq&4{!X|qUY%dV6
zR<IzVF+(x5eBghX?li^etPM*hEDYO<_ttF*gjzCA^OKic4$Rf4to&j|mTKA+0>0fp
zi^eU!xf-$Hg~P<nW>uEWaEDw~yUGak#g-GoxWZg96Q`VZm6Qoj$YY%HJdbhfsW2Qo
zo!bMyVIPlCKbKNyFOQA*#Tih!D73yJ1C9e8hYhByB(8tE8@HGd6$GK3{$DeqvR)Wz
zo|F-@vaQ{a5tRls3;fR+Q6VO>+Fg+mFJ(JM(Y!w{Qt6b1&9J(|UG~FS@2%CJeSnUu
zrKJ(p`HR{4ri-wHGaC#BqgxHyYG*lZDZzND^g@L;?J3mSk|;VlCvf#bY3^q_J+ACX
zoL@nP%DpnbyHQ#~cN65T-$@uQ2(VfG6}9lv-o%D0yxq=uca_(#?{}<iXfj`}Z!B{D
z1HO~oGna7%H536emtjK+6ag`p0bKzmmuT@cntwt3a6bO^%byQC6Z-MXVI<55$()<M
z{yh9XA>7Lpfe)NX@AUNVDGcKrgf+(EQz`&I{S!fc`N#KffBO0>Fr4tOFoREu2@T#&
zKhsyOahex$`jNjoCctVp0rN<%nha+k00C%X8TdLKl&}n>cFpLIFzT6dN{D8%1SXwD
zz<;tpGek55jAfM8`_T>(RdtETQz0XK3s~~p@D=CNtnf8_+V2F4vwoWn-=@Q-B2O>z
zA3lxcW=WjD6j)T4Mx-<(E%VkZL*nq3YrQhsGFk=&GtMX!oN&x?z`x3Bg;^SD7^We_
zMHpy^p5yq600v%@NRnD=rajIx%$lH-kbk9`BwPRf`ZG`aj+0)RXF~IUdcl-7(*w$v
z;cg(RXr-tdOIDhpGjKLic{)msPlX1i(bzQL-{<^1MZ=<Xmkg}Oi3EI_8u}8S4~5ZB
zn`MKZL~FG{I7HNfP^4x`R;jG`F5*C0v=Co>mpba;p|r=3NFq@deUAvj()uWZw|`RT
zsWf^j?c!MUX@dZ9hKb@We?8~-Cd@6PzG4mQX<pVs{OT#JRbX}6szg&*rF|Qz4R1dr
z`c$`$KIPRj*0`4IC^2~y{jxqU@{;E#uk9a`#}52Nc?9ZAS&wJ@J-52!N9B>{j*lQ$
z(NYzqpop3gHp?g}?69|46Y|~`4u2ufboK8*xdcmc^c1PuUC1LVI$%i$T+*0?ZQS*|
z#)aKTt3=s6H+1TLN7krY(~%6ns3UPl*COd4$`wnmW}8g??^6$!<MBfc$Q#lM!wr*K
z6w~y{V?|+MDj@)}p^^`aOm|9^1kaxlTFOCS7+xQ<`kw1nMqD<JMP(t#w117Wc|4g+
zSx+B*peq=ObV2ALjR^{s=bfAF)&=*<wk00pHZ^KjXmmsoG|Hz>Wv$J!Q5|Ik<6y*%
z97e8rYffeT-aI+ROXou35fTU|wG6qz!szE0)C~tqx}iKU=f$WS5;CQnoJ`_cisWMD
zRY`OPa7{C{SC~TD;#kNz*?-;m1fdy#8&?vYJK4^kAlDH!`$$T41M5N<WZf>ZZu5s0
z*W}Hrngw%EBAB2ce|O4neci60j1qd!P7#LQY+JEF=V+3#LSEaJ_Y_+^tr1;}Y{aQj
zRBnBKjdvZ**#OArtB*Ows(2f&;x_6+M-!WX_tA`$&IzVeQF&4UqJNR1LlZ+x<7!}_
zyx}m#zjlwLsGR6?8*Rd_C3!l^UWjWmI!@XLN7OoyPpWDje8}eFO;En@cp1K7B>H_O
z2rOQh#X1e6NO+NI95I1%C?B-7zfIo`pGLG<y)lCfX{3(1Bu{qPjeO1(ujgn5+!(AV
zTC)T+T?}-{4$dfhaerA3gMCGVpJP^@ftq7JRjuPsSzap8O8-R`x{=Q>P54HUp&@I;
zu2D7_D}1QZAM$9B=va}&VTR1%F3=OW6pH~1$3CVSJ{SgC#@+Nv$Fx(o(A&>=eLyJm
z&-NZ^G(yu#9PPuzbBXBDo|=5kg*P#f7O7;3Tvx(X4I~MHVSmP$;tlkek<K)4%Ra1X
zr$(MJVhyW;99DP*UnE_wv23sL#r2`}hLuLl%1LJZ8<^~Xcpr;ZR$!6<H(0<h<DD&5
z8Hmpt=<R3hliGyMJeE+sZp4Uzd655QBUaKguhk7ke9b@yj5i4{tcaXT<WudyBS~5~
z=DogZ&`Z(!T7MocLoy0AJ3gm+O`4`oOl5dUeZpHs_Ah8;JaEY)-SI{X7gM14jgB-j
z*5OEFaR#M8rfO+Slvb#?;;9;}S+0bM@Y_}*@n~aVrC0%ql;##~IW>n4@j#;R*&e0L
z648t$r<^fCwq}{Pwd0CJAswf<oXDF&qDyHqrh-hG;(rFWRI53eZ5(%><ag;V=}bDv
z*}uZY(wT7(BOl_T#H+N$Epct7)bp77p&gEr+`r1j+6VVc-OH=28oq@u>5!tiWH*<v
zca|&QJhMj3$`yxC@h;U0%87*=pZ!yx5%WCv8AG1LnaY&F8%<MG2d(bNK!}ze^Q8<V
z2g_jwdVh$Rb%^urG*}z~Got~P*SCrp3rp+wvZz^+`k2PXbcUoFsvoYFRr(HzD(3u%
z^rApzxKbTwyhfnLFfcR5Rt1{H-$U=?GuMTVPpxWZvmDEA=A2~NW4|PQ6Ru_LIAd!8
zu#zDM->jW!GUUCoUWp~15%!%+*J0Is$hIoW9)J2(t|1%yu>P7RW40{x2;Do@CAXRK
ztW>Ft5iPwD`{}5#hFNPj@+2_QaXFKK+Gw47XkON7f%Sqh?hOtX1Qy>`zd?i@5MK|l
z0yWrr@dg2Qz`P;A4l{NLup<qZ1MI^3cLrE=o|w(beFN-(bVGn0X6z7PM;a~%*a7<X
zM1Or<z|!MWy-_e7ux|*a!;Du3QxOh9Ko7xmCIhg{Yt!EFwB^km3f3bO{IYWO(0eFX
zkMWvv_0a!oG0!-LpN%{Px2?rY0`D2~NU6AC-rF6|7zd1dc=aCRHLu=7i|6(nEuGP}
zEkk!4!iO^$KPPM1e8LtBwCz#Q_gk-s&3}w;qgaGgOvu?r$GhA1o?Y*nN`W_+)VaHk
zb(cDI6_Y}}O{Xq!$!xee|3Tl8bWE~UX=e>oc#jUNbc51B*Q#4ayH?#gxi;ID722^(
z#UH0v*Xf`NO7;!7%cK9Ny}F@bkZPA^UC@{VXt%Ljm!NlT*KJ5x>f)=+-qf&bxPN<9
z!>-k+gkodAE-D48imUy)W1nQxiP^UD`I<C@Mz@e2fKE4l)a>GLNEkbG&csUt8(*2W
z?MAdcdGYmlsgjW;$n_hQjQH9$4}mrJJhVV_(L+$@64h5b=@c*kISO=JqMI=<Z<X>6
zow*$`t0v<ZRCk|_xvU`gw7QlkYk$|?vEgZZ2@_zIK6ck*4LT*ZCU+aJU`)K&9GEE=
ziAv<sKH_fs%;K+U7Q5%-oQs!S>PllP;b^ZK3nimLBuZ4<^L6ffu4BKMH>KaKV7uJa
z4G@a@nrDfD-$1A1k3-)$T}1lsA~Lu1CV1+~jvNPA2ak=aE;ZqjY^@)!H-8;9(9(q?
zxq;YEm*}JG#LPkLyw^h~pM&<|HqqIw!&gfj;;ocN<^qOO(<neXdun+h?M~9^`^U?r
zXpv)c(zT}Jq0`4;5;yE8KV6w+<I&@&1-6yNispDM>1zY!{H|VfclF}2uQhA)XA4`D
z%$CvhCi@4i?PF7+6il7#@PECT^mqMjC;Ae_U5l#`7b2E4FuRkNoB9nzjWDAmDo3j~
z`!g4^LU(i-YK`V(IE&ltP;H#;4nQ6KlH~Z1BYfWc_Lmnqz_&jUNU+6+<u2gHk3|8U
z36kt|#8f?nFqc~mvRCA${o#PIi*$|cDS33EX=3DK^u-0@LUhOwwSNU-x0e2JX=nVf
zZSUCcKzft&&?eTF-Gh0<N7or``rq+T&Y0hJJ?!b%Jv+9GV5zBLAYQsEiSiR9+iZK2
z{gi&Br#3IkL0LxifTocqK0g$#+thd=T|5NqX}tcUqgoiO{N^ZE9_??A%E?21{s&O>
zHu4H(Ze(+Ga%Ev{3YUBZH3^qd2Nx6pGM5qmJ13UfIs})s{4|t*pEXiA|MB+AlPrRM
zdWCRHO6RRW`XnW?mw=osa<s_VzM^zCN(cPSs?6mJexk${Wq~x7--E)(B^mBn$N>mv
ztA&g|E4M^e8krjTQiv~w@fw-OH_<eabz+AS!&=xQYkQ#?NvLa7BaY}%1yU~0z0ll9
zWlkZD#P_bugE|U-XPsn<qc+d7WhQt|Eo(;dC)$?e=6tN(=&6Wa<d@fx&T&j8U7&1X
zbQZl<&ZfzjXf0$-k~6!E&=7(?Nn!B`2F&RPI?;CYx9C<u&9>G;5C;6XwY~G;6Zdo-
z9Chw2HaLMJ1HpJktTXXkK`{s^bg4>}ZLeOc62pxbS|eP4a6~y7_O%#W)7ea>mf4i?
zZQ>xwVahhAnbm35B7^cT?bN4Dsq+~9!ZmS@TXd;!s=F21b>q7k8&}F2B)7=?Tye}a
zoAb?52Ig!wD08)C?mE-*7EiTfXAQ!m_fR|}9lR??z8)~3(YkZR21%Pm(*f~*5|^%e
zFD}&;amGS_-x{Su+Gn}eB5jkOhl<Lj`sIZMC|d$g$Mg2_rrGjZj*ccPB^u$Jd<<wm
z20zVK+XO0_!8EM_bWf*+osTeT+ic2KPCY>x;g?WP+yT4mpc$9^E$*BShRId!9jO3W
zpw2HIS7<TqG8Jnx|73->YAIf+iljrXlWV0YTY|`cGAF9Qsqy`-doBw_U#N$i=-3K<
zL+XPbRJ3uu7t~YI@r*LVI_Fv6PhxKLG#wV25fbOA2S_D?t+4<u%>p?#)stCc7oDT~
zaj(Q5DHa!VOZ@mceEahD^v^~p3tCudPJe!XiZ85D!eHW4#tR$l>BnEEZ%@CDWJ~sv
zLgE^K0r3z1m-1#e@>{`-i^L=+9j3Dm5h+ZqU{Bx|p>;bMH+^qNpD7`4DceOTyP-T!
z$T7n7`=!w|n0>(oq1`fFoq<%gxQ*J<pJaq?1dXtftEa3!$EVY*t^h9_=8s{;7O`e*
zoV5B<{c@@mQfvTy?$qyB42&=s+<M}-^m0mnEYc4POy~0ZFlSFS>kx-29%3hG^TWFJ
z)Q>3FuSnWC275~I)-@Gmm$ZabHz^^{%{rZPF5Wi^oh}0?z9u2$IyarR!W#45lVYt1
z%Ly>GS#S3b%2v3tP-+3nJ=z&Pjv&>6x>EDa!dlRIwUOm`2{KX4E=r(lKX&KSj?ZF$
ze?mx)adz#|3R;f}?AT6wEK|H6XWvF9)n@OJ%cw;dtu}JWsa!@I0UpNL53-SoQCuFi
zk%{30+{Q+#Y;Tl7b3Az8738`b0&$vvhBC`(Ka!|-vouk5i*Tl8LntzNs$VwQ`3ToB
zTE$Y@H7GzZzz>9HXyjhg2d&Hx_&m3NP~(TkE>d#<A@>?R76Xc^ao3E^mI3YAJ#ktk
z-YV9aa3PPg+Lu={U8Q?3rZG-ySO40=n_=gy$5;h4of!9Bsr&9gLjH;BM=`CjGUeR?
z)8etkV(Wi_v<87ENc$5Y?MQjYs4*ox0I=d}p%y=Ky@7qS48lt{t~dx-pRq=NdK?hY
zK?#ZI@*UC#oVhXbVOToHRtB*<SoqkM{S31x9i4s<YcItsFSLS<5=Y_D;}Vd;QUQYh
z$qC3HW_S8OWS~V@t?>X%GJ=s29JaErUVY$$VKVkr%2BBZWi5^Dz<{G4`c^q%h1?6v
zZEx0!cOSZ?(81tNgmz0Y^mlWAN90l@t8MlA=E{12*CIRnUb9T?Y32Wt#LRuDPe(d;
zAo?};YlueC-LxJLajllheX~oMUDwvGbnEkwR<9*L<u|P7NY@Mc<)*E<be{98nK_Mg
z*B5yOMTe!21XzjoohHU1c39O5>*CL}qh5QAGG7MyW?d`ol~AfxhQ8r{n4~LA(S`33
zYw3_bTWzQd-m<=A`!T}}WL?@>$EphM2bmhXtfI7Q8z$PIFdC(}x;b4c?3A<(CDqP!
z#&oxloa-qm0+SdzFv2e0ndy0q_cb!`dsRB+H)6t_F7DA-paD~W-i-yVTN$SChV0^5
zzO!_`$YU##+@RKWUORq&{Y}zthqrRSZwt<^>&pfnm5Y3au3<gh>M*Z|Qoq@Dlg0P>
zhOO0tb*|L!+h>@z7pfP;4!>+2{)hU8!g_O&ZsQwa>|$^35o@L3HyUK0s9GG@zj>WB
z<ISQ>>$KnLR}ffyZylm6;!iC7hdLzUnFS8jA%s>MbfJ-Bb%?@$a#}B~Bl?*^iK3`J
zT7O6^|6!yE*AF+r2TEGrY=955S;G8idvKt5r#(1Yl9$5zK+9`hk15`x(Lm6YcDx(m
z2p$&UVHt&zE>;S9FKz^?BY2F>^Y7gymRZ*U1mnK6KS|<lWCE|1^}-mv-Np=q%^fpR
zq^-H+noWz|S?zd#2cqAJXA05I4cu!j<(`!gNgbuzU<*g$Ra(>2jrh_LzVrjWlni;q
z7-r3l=&B;JgDy{24!03()kaoaQ?e@YtT?aZNs`xDck@fZvrvDa!;odpURzs}gfQzl
zWeuU*6OIOb#-Jb*23^j^HtaS-zL<&`?5S0)YY>K%t3Nn@fF&oENBy!$fyIa)22pme
z6Kvl*rXI82-HURGbm%up8ZdpysafKlfTWxXiv98iB;Vt7KGL_YHVZ7<CpryD6~B_4
zi@Xvi`O+h;+ChZM)wdH;{RdPp>%HK;wl%dKu+ezd5~;s<Am2-;&Gf%Yn)zPv=qH|M
zn+BwlAt7fr^XTf8d4*H#$6xc3YAA78F&0=*=}Vv~wChfSuiIDvKbL3LErPJww%qvb
zyL0r3AnRXRjhUBmBPA08IW(7HuO}3jJBtA*moNc21ee7HHJkw+w_*l0D*=BRcp{o|
zs(<2qT}x{P>>;$9CAl7+?-;1Y70?{e%@&_2eukYD1|_4<zN!oR7p1crxzsPJ4Xezt
zqk6*E7VmUAP8h@)B^IA@%;xo&=krxngT6?xL9&4>e7K9o=(v?lKm*4VHVAD}|I`|-
zI`nm=CCn)6Op%Js?ZHS)>1lskN<mW5J3nQK*|wiLc5^Uyf?F-O7Sl~DO8Ze(wd#ke
z6N2Zgxfez{?T}&{ykn@--jfe8Hw<+n#p^@eK<P}V8+z-@{V5nTgTav!nR0Jpmjule
z1_z2aa6sBQTDAMbI3Qy6gae|9XXRxJR3k60x`7fqAIT_{TbONQvgv=R%9^Ya->plD
zYc$;RoN<%m2`fD3M76|OczYTT{1on}r>`yxqx%ZZO|f&UZj_&9H<@OLk}aXvkprUL
zpqG=Zjf>V_iXT|W1IdaU0fAlep!Tzp*Z~6v2erOY(o)P83&sM4sO-;KFf1^Wqn$QY
zNGvi{Hu5pE`^4K&qE3HYL?`6BsmHvy$Nc%VqhNA>(L}XgeNqGrt6r!f3UZz$mTwwK
zj?EyCib3h#XcuUP3{jy98%rA@k+<Rm5MV(^a;gz26p_Mi+ko@7(E^YJ9&A64tQHY@
zvw2>VH!gt#j4^CeU~aF#u7n5*6omjcLIf`fZTsH^A}DWXjrM;6h_EL_o`Vs?tc#k$
z<#1x4D4tKkiJ|u!;Ka^(GgR1`_i~7!oSC)hA%a|o2qBOMAOP~`Km;M|xPRwW^Q>MP
zlGrVaGy}owfoo6boT9DIx<3cz%FHmUY4P8B8oaSJSE!mMcjTc{HG2Y$>Ys>4{|9FL
zW11617Hj1#?P-6Yc#T0R%7hn$;O()Xr?^PSF0eiWvln)xc#-*K1u8Hd-)@kE0v)LT
z8|1qg<TYt90lUV7@63YzkhWl(DBLKwl@wY~7j?pz{^*ml>pl29^(v{6DXG=|(J=vg
z;<T%@nq}#d%X?`a?VHl1%e_j<hs>9LNLc{o33y+7aiM>*<I8j2t=haH&DNG86Nivj
zC$gfsR&sT8kr3pJ#c^8Mu3}4VIaSj_a}Hh-%UtMea`Bpm>eERk31~*|&n20dc63K}
z6r_#@I!X1_3`~+xYppz{t8X}-Kd!dPH?izUEMarW*@P!;#)nnHD;mCkZ6))2du=70
z_=FP1Ab)?k%42eHfkSy#OLi4dlqAGz>^OTlEPDI;lp@N)?_J+z$A%Xbi?lSe)wr?N
zg*_1BA+s+4g)0R1W|9HaGe|kkw7K(FRn=le?LtAJl+<WP#`59mLQsg5tCU1RHf|A}
zxNm}8c2&jUTJF(S+|WUD9+WAV!4dGX>l~`k+A@FcNcUU-%G7fB$`lJP&=g<eHiQ+u
zlzt}yRMTVQR4UKBqcogpxu&X^;KR*KR=Qb&T<olCl<l<%(NAq)+S-BKKUQS-nJc;(
z(2ZQgDkTKTMf_&hb&@c%()Hko_v__B>iZfTq|gjc#7XEQOj~pmW;K9kpdnoEE|_);
z7ZQI3K8uCr327<k3v-^YbD%2DFOCfa5oVR65LV+TE|-cDpyjf-n!eZY!xmZB+Xmx+
z;WNL-R+hT4>1Z3r($6-~#?|0Ec(ZZu(L2iLioQgN;Az;$)}=-`o<{U%)1}zFgTJ;+
z?pu1p8TR`AdeK~2xwOcP6kS5{Iux0WpjLnUU5d<T_3Yd&t9c{E$E(YE1En+P&_nMV
zT)7%3p1E>0&VSYsUw&T-FzCQOwh-)$u-K0Ce!pJr3O?(K8%byh!dbiEQ(s*@;;UZ+
z#f<_ldO2IaRBfN{SUp``*+)vpeK{lWMiM%3BBW2=*f-aXP)pigJ~!}Tr{42=LEwMj
z?Q2q!9FkwG>78m~w^>s-$JRv&j*Sw$?kGl4AQ-~{5PGsUUPt<MqK+S@*qaDQPYV2>
zDKtX>g(l>I0q_I(!0q9~zrKC_)0e-&OauNMMo<(X90}e{Khjt0XNO1(e^1{X&Ln;{
zh4QI1YI158R?%}FR~o)d2Zff>AA*0Swe@>w^|Nvcu^Ajgmkk{={n}DE2n0(jPxrlT
zWTxs2aYcoT@GVfuW5ZV(&!fWE@M^zvR~q$?>F{+rd@9oPlK#=Vv6R}c2oop@SP88_
zxhWK%3G+IZhPR}V8?Eg^OC!$486|0HZnmVIaw@Zx#d1rtU~VE>Oo5yDeQbZ<5Wuk3
zB$6CZXYDD(69|C_O(2UkiMD?J@*|J?Mx#D9?}^SGpo{?j4#*?ZZU$Nrl5A9Cb9e5|
zf%Gxw1KH}Mb@%xc&%9=015sk()6}q-)_PYL`)Rj;0A@`)d0s&6@y6b?Z$*ew%@R!!
z;#R#Pua{N?{=L#d0v_>}+`xZDf$&`K6mm%XRG0Y4lK82U307D4P2Wyk<*7Z^F6hLk
zJL`*-<eG9NH{ZfZ<4sL3;|n=iQ}>mta>?KZ<p2{ZczUt#7l+GT`&+b3yBVeVne{h?
zvqnA6bekVETv%`mNsk*jNf-Hx&Sag`%?-c49K{?2u0ojv0Fdh>;lO_&mBM9HnKFI(
zFe3CgX*r^>Wt?bgZ@Chsm0RA9ONUPzsst*8#)Zgl=0f6+F5-`zUNS%%s=KUR$`I^_
z4FJEXo0NxYg^SUdmEv8jxF2lOa+$NRO#A_XQ|T90ZcC|<FYUYL8hm=%mWqzeNs|@*
zMf_8nfu5kDv*T)R=+A%QzGi%&3KD<rexLHnSbgUv!~=z(L&GMF22o1oMJ!y$N#DkR
z5nGEM`x=mLQtHoYmjmrPx*81Even4RsT;FH6<`ekRs67qfGV0v2<V?v-z}#qcTKLs
zSk4gzYvkGiwx`71*!E<-xF)Pnt`je7=HkTtMrZ2Q%+A)|W@dlCs}d9u<ArU0QVw+d
z(v1i0RVlDHzpha|5HS5ww?aL*6xBfkzM-ruZQujeZpLgcdA6p)somN-##_>=fGo8G
z_>^8xXT_~V_>mkA;uiY!AksoaH4IWa)F$iyCSgsuoYL^ixY5*T-Ka7CL6^NgaROBJ
zW@Q()e;lw@Ef0TXC+JaDh+wk(`XJuQ5Oum-L~pLh$G-J;1H0NBlYqfPxrvcZ{dIHi
zLHZN71Trt>MoEpWg)>t}@{!y?vs~O4<4pX6wG(`I@>p=z3%S|629*W65!0J|U;Sp-
z9H@~HNxzNn5LYxgd+I+hvL06YN>&Xk{{@mPuoDVpZe*8r1vL->H<$6{4HK6g5jDhr
z5580O@b}Mu-AQ6}<jQ#vNUw5*1SQ}oz|nx*S$}xYUhc~i;5;_scSx;?Ba56^>}xe|
z0p!uzgK|XCB466Yed(L%=U&IznB$DDquJ!FHBzf3F7Y>{-{&%?JY&z0xjj>l=YxTL
zrjexZ;T+VOj*J8mlwL>!IdoZ?^11$hT5D54)YmP*fEUKPk$TcA$Uu56>!+26HEDV-
z&7u>RB#)m!EB19<mP)50=Tx$GL+vjYdg)Ycr|6gsD57zyzw`PRw&k>tK{shKEt|3$
zT5IExXy=|MTAypZktc}NJLc3<3q;~s(MY0_*Me>8=~WxhItuMZrj>n@%C+@>!fCUh
zrq%@oAQZ&f%RC-zpao}Bf1ZiDieqr4Q7m1&^hT3eJdyHAL`A^ig7cTTsyUIM9SK_2
zH-%4I&5_J48@L7vc1^jYj;i47#N@8_iTce{*XE^lrQX&oIEOB!Jia}*UiP4~DQDAX
z#^RY-r!}0KjC~1hH2$f%QKo}`Z%UvpT9rcEfV!3f@Dq$R(;pagm#v>(E*fF82<?Pc
zEmWAmoZ`&%e(IW5B-5;Q-DSHj^FtQId6pY48nXh7UtwgbMFcXYWTwnpADJO?qip#N
znb_pHt=EaGyqe8SsX&Tt3Y;526&s^+)i7r~>Ukulh$2OFNrdU^Wq5jj{=EBr;qNv$
zVVs-3{Me-zV3aUuA{yUDJN@yG>Fe&}NR@OXg~TD{pOwYHW-Xrxj6{j5l+IZ@voD1<
zXkl{F2n=LRl}{{*P-4|;Fpg}oONt;7*4U9Ex@U$c`a~jv#9#fF*`Bq0V$j4IVPY65
zDzy$|Qlv$v8t1?2F}L%76|IP1Hu8mQ7b^ONq&ItL;aeVa(z}&c?GS3H1+SRDR(bVW
zSgjYlqRGosi*3cey4lTI_}sw?{4C~nEsGv6<W!sITeSE7>XrUAJP=H)*{$YKf;-d~
zW_`E9JU^2juG>ugUzW+Nh1(AKosC*Jv>_(p7RX$KseOj^5OBYLJY%gB;KR*%=6cxN
zlvt9Fct3KHLb~4f)L}UQR<k6t1c9k9P&7OBlA)Fe1s&WT{!7`@an3S5b$IH32A<xY
zOq4^Y8Gd0hvBn@L=8nl^pm>L5GElsdOvd@&FPUuPt0t3C((jl+tV#iXnDA2muu*s$
z7wo+;e^53W=Wg+T+SoBR^te!R#Ko%Z4iQ{y5hXOPj)`i`Lu!$?<~gbxa*3nn%7%tM
zu?i)v^hi{pOZL@V@!^cPJqR0p667-eLoir*y0$4rbEN7j=Oo>EYt<)giDu*y*EYof
zaf-KXdY~Zh`3^xZDO;#D6mD6$Cwr)_Y=rBDZDv1oc+iV~=w9|g_4LPC9ibNeU~8zC
zYgNUpV(9>e7p6=C^TM?5otSy0^E$s(wpr0}QJp!iOd4w`vbU{+7O}FSy{2dr5R~*S
zj(D)JA2uSY9KdfkuzP|b#N-0N+zN&=9N-%n%%RJc!A)Rj;9@%%>h@Q)xsdDtRH_v$
zxo<SDYk^yTa@!b5HhC*w?*)*I4Z>lN82}PQ5wJq~GXRo_$cYYj1V{tLI{>7C;uU~2
z&i{S@X&YY^APv2*yCwmo$G0IBYq4da-&>GMat^9EFE|HLdL}$|?$vL@8BeXQO*G^}
zN^ikACOGt-$oIjO?E9|}*s%iDRAohGbshM5E_T{~_IHoPW^*$49qaTX4^|($x}%&9
zxz=@>4#6<2sVat3&+htK6nmaeuVMjpF6mNck~R3eNupLL87Je>qSXvI+E;)h{As}P
z4NF1T7Ona}0l}CR#+;x7qIi>1!tWFIdzo@)9FL*UPXC>9vryVXL~2ec?-Esme@Q~S
z5C8su_Vq7cep3R6{U>$Byn<muVIcic9xWCz4gjX_^o$Zy+J{MD`GE49Ec%ynpxeOK
z@MYQomhTD^7i;~VTBCpfsWxhZb2js-W1WCCRUQ(>)>`NL&KoONeTuxM#zuNdRA_8^
z#PM}hdQ2bwT)X0^e@?rv)9xcl)jI7nER%(QaCG>$CRHUsd84>wYm;#TVzj<WC3-ZU
z3$1*;tx*dX(ncKUAUv9qnc8e^&9#NCyEY-MN})E@@3H-Y9EP=GOV6=bui-c&9A-2$
zQAN@%67Ijh{Gf5)aMTO?E_5E0HUg}h9*j;~M447bD;J<Tb+5f^Zm*jQzCqbK)0%^S
z4McFtj}v35tnHyM?&D^GU}BQSY!Jv>HcFR5(-A|2Gfik0YKOXBncVO8jnu#r-0>nB
z-&8+nCA^9wCtK<V@TP8xrz2Yr?5wP%Vwl5{(yMNP77BdRlc>~8%b|iBYls4CtTO!E
zu<m0W+gy%q!NvwG>)XZL;~f&>(J&}~M*9f$pqzMy`&O4^A=RFKQFW<|&|ZyP3_x+k
zk(nE;IdXT(;>iEX|KH|$l$#%Va(>P6Y9>!mBzALe7gw07J)O5*HO5V%Q~_~ZT+Yl6
zqPe4m4W}0Q{o+Q&dcnTUrM8|*tjp`d%<$f&y4Hv<!T-Xu3=KL@dB0M9pm>LWQhlI!
zrBok!|D1!Yl*_l5mKxOxy2l{9bt4<0sn*i1g#i^JB8s+?-0x1NYVPom7~hvmlCAy>
zvl~BKcB_?3-@sM*_G)&v#T}G6Dt^Maa?S14`(_xeqf{e@(q0+mg;jbSWd}1z!mC@y
z<-!xn_NdvVFHfBoJIb$5AAR0`n7fT>BaQIEESNSO>$IzGZKLAC9YVgxZrM7*Y%l$?
z(Wrba_rk3{1z2qK3%~7>C*4y3<LFLXgCoAasdw%g?mpJm8O$jyfkUH%Gg(3Dt=+TZ
z1K;S@WkhzHjx2XhBhOOAcI59FNMQ#PWj$5E<>}}5dkzmiBpGYE$VIb%3{n`)v8}a2
zK-Vt_gLC_yWm^w+Uy|hNx);>*uvY&Jnl>KGO2;ozTiqpHZg<RXF|v!)i%oLbQ%?27
z>RfGKO?Ry_R1lc4p4ywb(tgxLIAuI!Yi@h$SA5w~6)fhQ->zKVcwmmhwOnZWD_R)v
ziJi?8*rC)sRzkV-ZC6%*$oXm#e7ta^W;?>U<lh>>1v@Sp{TnoerOeYK*;a0y9j^J|
z?rI-99DdT+ddV@eBB{M?1zDTDAIcOjX~hPu9zcz80>_d^9ml@<N<efJIe6JQ_Ahg&
z1%G4EuCh?f)@WVa|3UBXv$LFUMxWPpErXC@9qaQV@-*yurxalemKcrITmJ*egdUZb
z@s>0c0yQz05xOxhm&njFD1X(OU2h$?5r*&cD|p%7@Q^bc4haMX@=>P+S|Eshk=F2q
z>j*{AKnfHENdJAG*>hHwbB<zJS_v>9=*;eN$ssv2?>obtR#Hh*+e}U3x24he?UqF1
zftRGx<4H}eYY38TE_XD!OazlOmPf|F**`;Qsn-27OI3PwZ9P?;OMf#@)g*?ok!t53
z(oEH7jMz)6b9vfpl0su7NmBPuHHk`NjP6O=@@^kUi%NYaF<tI8Bo7Ukk9Y|U=s-q^
zr_LJ#)sT2&j<`3NX_XpMWzH633`wdZdgh;6UD`R4E$s%<GNUmQnniy@kvw$f0X@aj
zNxNA?mr2i4V;<8|D}P-X1I1e{(cg5Ww!9%ogZ^4!V>GmCl{cDr+$*VBywS?CDJ>-4
za25hdQ=2j0R$8*6%%c@cR`*YPGMO_ZPu5hU=bFrVj8@rOo6!!%t+x^qz4Wp4H<Eem
zOk-<p^mwX^HL&LmbLmAxE7a{Z45EpoF0nIBH1&85sikiFr+=1upD`{L#t_kt4!9Mw
zl!3H7)t<a+WsJq-bb0DXOFO3Tv9(Aexk+fM6Djt$F$Fgkc4Cs=H59;-S*cJIQ+E1`
z7YwNzL+}=M7*8?3QA!LEUQw1sYnBG~7gA3{mqlxi&n%19MmWOK^Gw5+r#RRc%hRx#
zvP=VqnfnC_kAIOFOS+%)B1PkHnTAtLD{(K@+FC3S^OzO}+ngr*?Ff4tVOEQWj?q8+
z{PW${uW(VryT99g|IdG=T09Rl!mP5gyXU|D^waBS&;F=o^<0_KyOwG9)x}R2moI*P
z^Y$!hypVR^zPY+OyF5>_zx>N@SKqw2dUJJ_>i)*_i+}Sg#(2&))q*d^D9_ohYM9e<
z!B$jhy1$5s5Rrq6XU}%uUS7O?ad!1;_wCoucQ4O=yE<t6@_#>{G4g-TQ0V;X?ELDN
zaQVJ;@!i=k7r$P<J^N)J1>)8}&VKmu&6gLyy^3JhF>;^sn)fexGh$C!T77YTezA1q
zWk399uYZ=X7t>h6-a*iVVCx;bzSCC0R-vzgufkY`c@TP8g>|gH_S~1AUtdpoQE~U;
z*MDEd)!%<S|8LrTd2#u}+2t~|@_P5z-QRX!y{h|*@6O&{y{g*s!0@}x9QIb{$ihb8
zdby1LdiO=5`xm=!E?!>jzD|E?SW>CISuZZ4*nj=_^Rs6+*8M(8h+7``(yD~I<XF<%
zln|f8g+ENSwBgnJ!<A?Z<{0#<gpHD0zkMZe1VSMHiRwNE8!qkPTGYX45hBI2IdJAR
zs}l;<vy>BB5c}LQ1=mWfA(s2YmC!cw1Q1pw2y7XwxK#;aH~7{aB%MH%flC{##A=zf
zXMd?Z;5-9FC}k)yk5|IP3b>7FYPBh)=>{v|o;8pwcf1YkLM2qFnxO#4&f$#hK01&M
z?v=U8W;k{UE_clgZEQhUDch1u^W&{3)e9lgs|-M-m$rpb&XxmCAvQSxNZ6VJ!?Ipc
z;>1=I%+{M?j01g$SiZ$x)SBg;62NN~f`8&|O0W$>CRQ{$i~?lr8$>D=GzS>BQL}-}
zkiHzS-zK{0DWe>KB!-}Hup00`H*t-cAtbOsMuiP3_Blff>=Uwhq(xAIZZOXZy|gyq
zTiYoS_8HU=G{^>WKzFl<oY)VzjDr{6CM**Eu;EIm5*#2pDv}8S3C5Bkl*ju4Gk<f3
z_Cow|jl%ZyE#f7k2BSY{fyrgq>FrwdS__oD^fCxlK-qj3D1vYGn|!Oox8S&d!tYpF
zKW1eX!4i`D;g*6WYkds*wiRO=cEs3*-iNWx8;s3Yg5@j08Y{sXE5RDF0k1Nkk{RH6
z;5nus!78|(EWeJYaZw^q9uYFV=YQfvtHAXc*%^Fm=#dgCU^m9uOKFkIjVc?sC@aw*
z5lkz0(o~XhR)$o!hI7~x#R)*76So63(E+lS9ituiZTF{&5>PoI%{%B|%s=WEx9RYq
z7;$j7@M*87!o!UQtJ8M?LS)>brRvB)*Bx3y!E+^9f9y!|URL;kurk|GSbv#vudp)S
zg_T)12pN%jVISlX{6cU>%h0)%Z5>&N5l15Z5L|om%g=iTGW9MX@?NV0G!$sWv^%t{
z1_zs;%-%!Gf4}$b(Cf$TeU#hQy^n(1f*Vn9I|#B0x(aO-Y!&({_`^*{z2#XWF`1Z*
zKtPDKs!?8pmY&GsYtarNmVZi~G@vbggBYuYNEVerj3WxJO(jRw&6Y~hF)NKMg&dwh
zU!{|7_C9=zT&3n2b<t@g&cO^zjuNf`KZL#$3>`hl@rm<f`()4}wt*Il!=c5{92HMc
zJx5WOAJOBb&=w^=OuUibb)D0;z?%*ibV7L-vUjoskS0oxQC+F+B!A+V6!N?=*fs-m
zwm@FI0s_>pwP;v_+^WNHBV!|tsET#MX6qI-2os1>tzwWES+Ypug9Sj_#ZFcdqKrEo
zxby;n$C|e@iAYi4S&6>Mm-Ys#<+zw&b5_SFI*w|!q0`Oi9;N?3g#VijuCs$yvgut~
z65z7I96#Jr@iUC}Q-7%gV+%(RAoqd*m3I&zD+I_20kT4XtPmh81jq^j(jOlAL4H!t
zzLcoqup;V0adJ{M6V;E&wN7vDBpZXUY@|kI1Ivc6#Mzh;dOZ%_gwPu`WL&?UC`Wjl
zzSM#n5_cWLt{cRCsl=JOFGs+M;@mEm=}S3+yyis0HB}s5ReuOIg@CJDgk0PPGJmZ^
z{Kj$0ZA#$Q1B|vdF?1p8A{=e89|Zc2yMTP~4dPd3kcFswZBLYX>b`Ug5eY79m`IPq
zsg}oyTHT{;hAat|MZP5gdpdz&jeKjKphP4A#UrT1p(w6nW4;$}WqyE7Zx;4$1h?C1
z@JEL|KMH$l_kRg{dtDA;?{I7tr>qZ;Wad%e#uD-4ps<d$at&`FxE;HWN*%3PZ*dOf
z6Q;FArTZ}YgtD^qQLC`#%Z?MRKI$MUj7IwhK>9|5>l_`InN#o3av|7$mp(NjvfQOl
z=wGfj?$R=f&*q<u2iVa8aI^ay0Jr@??A`;~3hHeI^?$bJb!~NyHdf(Ka%{HJ1zYLD
zdRp9UgEL*}ppeCv9KgT;gnG6(tswxOIvseNgG@pwv^Hwsq@<Av=~&Lt)PTZM{`?pR
znXO@?fxibO*3+}wlz5zj%%Df=OG!=Rpd=19<CMD*@QOXnL1y5w20f}U6I?2c1I6}r
z>1wBp0)KMYM?()#d9WxOGLsHDSiah7S+LOR9tEc-QUV9gQpU9wD>u1Ki%0q7p<<a9
zq|6L>gEM0srMLLt4Yu+$Up%b4p*43>2=Fw=Csm=E>Jxo*QK3>T4)wE;3tOZuSf|%K
z<)hKo?s97Oeyc-X+@mrw?$A<$OBcICOVBJ297gU_$%|c&;L+=S;L-gZJoYvD?JG|9
z6({?OlYPa>J~pVKABrqp;%kYCO~-jK*Gdu~^@MMeA;gERqZDa)!Qr6i{~$M=ml3Ws
z6ah4s5h?~0ml^dqD3>rMHI{$wd6&0;eERE)L`2^`TwJ1XDR{Gmr4LrX{tZijU}1L)
zxY!Tyz^)F)96)z~iBm8Z3WPKIy>r4_HS7#K*o3>eH=Sv;5$Ly>f<A&c0M+)6Ft2S^
z9=+igpGQp|HFZtFHMCKk;k7qf8EGU#D04hgCoA*}vCOvkdq<pMTt0vD>Xi4R(YBxS
z6#{X~J_@%8TEZP=-LZVVcf1Rf_Z+_t0V(syoyS1e71tK}_Nkrun|C9sDkt@2kPqSg
zV4NTw$ZPLyMv+p{ij@{vBYgBt3*Ix%6xvDq3j~GyQsKfld8Ik7$g937bIadu{9C2B
za#*Rs4jNwC2<6h^l@os^X~gR$-S%vwMF904XJ?zWcF<J9dEk1b!c}UEo?WY^7OFhb
z;~ND<7L6Y(PBBTSx|dQ-Ru$H@;h~#Eq=0QqT%<5AOF{~tq+b32O1HoKt<$fghdC?y
z3k{%6ns`8(B+(M@lh)ivj3zD{8$?omhQc)BXX!BB2^H;1jvjxtU*lt{pWdV;4!G~>
zl}k=oAD+~!arSe1^|QQC0ltN}g8k=t(X>S;;mm61KjURjtk(r6oHfsKd6+HD_08K0
zgX__67EYV+rW5%zNuwHlQL*Ivar{eGC_UpS<@IEh7dnP{mGeJu!%~YF>}u!0e9DLB
ztt`gR;)C>t4{Cop(L9xMn78C#`<#{J+(4yg<u}Yek>4-16enb|GknT5YL4{Q_?A5~
z^ZOJnQ;*qXSEB_q|9>TKUURP@n=nJ1Y<^*s(A%rJY)L5$QmV5iNuA}1PDi1X<=Je(
zB=)QXb@ePSbTWycVv{D1+dL=i`nVb;84npv|0+=)cTRu5U141LP~&gl*<$7MQPxvu
z+_@DP$>!f`zFuS<YR=xg9p-YB4O--$dcSQ0%8LYyv{4DG8~>Hv_mOBd(+X`HCmxNj
z%3o%^EqIcbc1<oUN9jB&`O6(GqHDk|4PFV?C~)T#@cH^NSly<~nq7r%IW`RW)fFYl
zGis_P%rSrV7O)8&e}9huxB?=Ja_%y+Ovzs4P)b<HbYqcFS=~f9sNR@#Moc=Xq$rMP
zbWNlH_a@$IxZnZ~4A(;P<h>gd$_}RD7^L!LSDEafO!h#T&fJMfL1>>23k=hEmW|yM
zXo)Gooks1oCqPAU%n%goO5brT%E*&*D&ct9Y7T!@({Sfh_IPkSt?R{Ff!NkG_ik1?
zV7hxkClrZ-u)*ovaIi5Kayl|2yZsQw>r|icxDf@NpY7X^6%H%eM@Z*;P-?rO9hFkO
z6=97B2R73jeZ=xIY&*VI%6snA+g#V1CE30kC6dEydhUjzdkA3^bWlp0Jn%u7sMg^W
zWU7BGrwef{9>;U7qM7B*GcNGPgl_wx9f3QT+6DOCw*7W=9>0cRS2y51*#iaZzPdx<
z;ViIlW|cYcKoxtEYR9O{iKw(2EtVCfp4)uK8KDqQJM9D98T)mSv@Ll^ORe@ip&$Jx
zTDCTmUaZs2Bi+E+C%VB>BG0X)wURJ-eKLPkhhbMY*y&0)hNh7nFy4>}v;UP%9$~zv
zXPAgE8E<-wa~ZSbc0-Pv8eoF3$*fwe8>#M$Q~2qW^)NbX4JnXZVR%vmYi)rpGq36?
zdXW!L#^<&H%chFw<%xb6cF{cnF3~6MMIIhU+;!HLu8CuAHtVqa74}DYU_5M^th9eb
zCE@%kK%lAWXfFJX92@?{86H^wdg|X~@^PmJGpzewhllnjg{*70`VE=+Iw?v+G|lA<
zRTtim9BFFF!=BO^s7HPB6Pr#n)hGpnGjFqsA$Fuvm`EYD*8x8gbmN8%L?ML*%g<iI
zKv_WdUc#Kd_4<Z)9kOy&4oX~Cx&40(Hq_-+XI<X7t#<J1Wv_mIvOX?G7~}?TjhEcD
z)i-W#Wn@ea`6yA1BNS|0x{UQ9>|WhPU_qzMDprD=svP3$n+3RT<VBvQ!940UG<u3N
zsk^SITZJ<(o1P4FY^k0Nb(xE6ZF?37b?&nr{tnA-YJpBrG`b&aolOTJTh)K@W8Yiv
zctIV42&x}+o*2yfIA?Y;XAP?A-&g*hWhCtA=gQQqwKF4cW`WS$0OKd4H*xni<tAeu
zqItFQoHJ+=IO`{92R>AW-3HHAk6<rrv_H_`IqN8BV}w;|4=*cu<H^dJL12BdRqfOn
zhq$438JgXsVptxC$FvCl%G!UXibhLCBG(;-HhLfSknEs6w$V7}&g<P<aKEkj5t!Sf
zO4zc)PTH#<_p&`#{2SJ!Zh_=x@k{XmP_wJKk$hM|X|051;af7OvFmfIq-v6Klx3Si
zXU)p;JzXHsRGVtvW!bD(TK{N`+i^A>P07J0Dy2JF6$Chd6>pgSo^*faYT!>l2-heh
zA`H!NhrU)g?4%9c9(s0T?~nQbW0$V$5$kvjM@$s<Nrq;=wqOEFQyJvI`(3o69kzxM
zuWe0*B-PEw@6Y&R8=61<@M*QJvQk8l$LH@a|5;w-@^LB6>=!9aif;MeQig>x@NbbK
zcpFqZF#7{0zWevrFMoge{JRpc?7w-cT&w`cp5pSozB-qkd`!zXe2+poug#)FlE&M`
z!>CIzfflxb&&!L|0ie6cM!)6JBy6-=v&BXU69Y_a0N4=WkaujO1n(OkB&<4!LQ_K{
zzvW$UZGMIIv?{;mhriQQSoP24<;(K&8*<fo?*mMUQ5%<!A47k%c8jZE5GokOIZH!O
zD5sFDC$1zgc(*gk>1~5L-_EhXEFHmFD4XSi&>GD&!sw<U*o$_cA^M%icla>T3L{NP
zop1@(c_`0PR#AoAtsL%ufBuf^zQU@H?6>aonopgxOL<k`+Z<>z>{TxC?c^sooo4gG
zxn~gsg${qXbxnU80Jz^4hEApHwG;N+)69Uf(}C2^nK7S9F(x<-x}F*nc9J2Y!CPUh
z-R&MYby>aT?t#-3(x{flgr&ICdW!<*x4=j4>w&>s_c{xAj#z=+*i6eBlIv0+-;dn6
zSp$Lm*WMQ}AS4CSnRxcJyWZ-OG#p)|l$;&y)y~V-86<z?kwV8A72e<E!cr{5;MEKh
zFLl?bqA<N1j<Z<oY*&3e6Ys`34k;O`B+oU}aDk3pNma5(HhW$5NRRBXYyOr!cH64<
z_^)N0hB5P%1<fa}0V}N`81(ZQD(@5P{sq=#V{L$HCN<k|z4bt+JHhj7Ccc|lsRzQ4
zm6y4>V$6S4qQ1B9l3fR-`Xcy^JPPMe=8rhEw^k+8@ib-Pm@V_3H$ubfVNGx79E`Of
zhO@Bu`HHk;O)P{>7^i_pwtaVF|G1pxbPN)wb2^LWv{HQ=)uH5-6gdd923X~Oy|`&&
zH+O!@Z|~2Ye+NieRGgvAa?_4leU46&O&)bSRMvm_Rjcp}RB<=#*zE&+izutxa2BfQ
z{J*t_ojvUUaUA4r_E$4BYDTwEfNetUH!gzhNehw_5pp^@NsDz;hIaP*>CVKqbbveo
zXso1@>%aSe<woZRWk6kliNYsN22&I%DPGp@71q>n!}djb_Mn9w^6jX~a7XG$0Fe9&
z1$lqah&1z6ht+ECAq77zo>d2FV%fuv>Kx&j%UTHrH-c*Yll*A(Ja!7v?x}j+c&OJQ
zR!_8?VC5liH~s`GJ6!eey_tSZuIbXQMjKl$X;0YQcN?w|vQig$zjeCecj(=;hw>{-
zo1ERDg&M}&88S?rV{oNS)UM-9Y}<BroSE3RZQGe0+uE`1iEZ2F#I`e;Gw*l4I(4eL
z`bXbg-M^lvweD;23+m+|q0rBzJS!=$XwJqFb!gIi!YsN@k5BuC0(>tO8bc$le1jSP
zy!}8=t_BGsQDZ>za`J-E{E+@(KwhC}L8R<}9B?O44?DmdD7jm6|1msBE}mirIY@Ud
zb;r~z*<INxW}h(VRbU;cj@%F*$(~Lw`Tli3Jz-Y0`iYQaA2Y-9X#VKiQjjW0B2G_p
z^6M1JM)QqS+Jdx~J+w<BP&xWZQN`gUB!$0@|C`WZ%D$wZduEmeE=~2kUl{X_TVH*r
z#<E8K!y>&uP`iNNVq8WX&zMw>OIdug_EbLYg(M@G(pGB2{n)P?xk|z(PM5DqBS776
zB+YWHwdzO(9U6a6FGVj!Ixx!j@nWe|4s|9(*uI`IS6n!MR5D&SUwIi)FR^cGjYu&?
zBtvEeQ(`+gk6@?rANv)L!kR*>+#Rd~M)GCFDhM4L_yj4OqMW@_e;|wDCuk&}D^fXX
zE&n~(<q6x*TxaWL!yQrw-nWl8iPSb%&WXWq4+B{Oe<8<df-r{DFyE;Tx~=c11Fca$
z1j^Ra%>|1K^k<m1E~6B`k>pRtucGt#@m4f`kw=0br#7oAt;48S&_1K>zi?f!#GA-N
z=E+~ZfJUM$U2cW}J28t8ek5Cw@p|1210PPP5T5%rWu^9Toi;l}zZ|6V%*XY1I`-Jc
zU?oC^%Y*68x9XRp%Zc_=?aG3jxS!FyRM|UoZ}GQKt;n^Qhc&RhddRX3l9PlV|A-DS
zh(@L#dI&&?o_rI>Aaow>L*)<z*l&`^6Qk?H109heVj6U!XL+NIi^{Nur!gFy5K#k3
zgC_d|m-sSeyqV<)dVTL*ABh!|K}`Ynv&Eer;o`QZ=8GzES4&2buy*nAUorQU(Nvy?
zVLbS;uu>Tmq}<-_f>9$2a=aOtQ-b_AnBsO{)$N#Yx@WviTXmiw^rH+hMw+%-{FF02
zV3_I>mG<(HcqXI{;@k%5){K0Sk`B_=5)q~=wTsvnZD;g}*`&kKKZtjJ$|t<dwcpS)
zJ*O1YWTQo{vggU_p@B$eKL^JXnWI+p?+yL}NW?FFik9|qNw!>;F;*=zZ;qnJ0iAC8
z!34k-W<P4hBVxFDl#>}u#>#WkuhmHmpuhZqvUe9|Kgz=B;+9)wRf8RGfQBNVPJDv)
z2?<MzS68eo1kr+u<U|^qAXeSf9Q5dJwZU>S(o(yv7&-fbpmIzd<T~=Q&KN4V`S9^z
z_>)f6$r~OCP@8vOX2><uQMTRauMpoc>87WhQa5^kSPO4c@XumEMK2OuRi@qo{K=6D
zvBK8lf36#~n7zurqi~=u_$M<LleRL&1sDJHMNfDA*VtEDUe!IJK7;oZl#7jy^&VIT
z(liX$K{GUUWs0z$EQLsGk2h7pd9_rCD{=auzjH2>Z}2`ju8`7{6GG%x=22o2S3=*n
zB~rhu&`4yBvAdp-pOh?rqAt*(fJ~g}u&E(2I$f1Aqe_)tLXd|YKjR4vL5)8vLUI*_
z>WV!kodpltG>VqHx=HN$@YpG7cv<}U0;<!a`B|vSx-mxN=TgL4_mktAEz=(k8&WaL
zT~yK+zurUO>E5Ac^Jje&Pa|bWZ24r)RP}qj#{*U;=K|p>1nxN&V`PddfhFa%{V?pQ
z-4^STJPg_RPRxpF!vC@Xm|P#x9+>$MP3C6@UawGDet#*U&UAjiZr&?s_>mZ+iariG
z%tP1evt0|!NMD>iv0i4L!_0Cm=O9YB_)fSdu>ul+^hqd37duUq&wYmAD#n$H35uz$
zUSMPNzsN2cLiH{fgKUW90D9|1=FLfTM2eIm&$`lSD|6UmT5WMCSL+rkY4s?cojFz*
zLg%QEws={4SZEvP95*H{7-luu{tKrfoZxq*QbZEQ#em5bI7okNNG~Ziq=~samcLc9
zrxSG^nrY{=6|#(l=N|XoMS#L8n>0!d<T<9rPCW4zZWge^fOHw?1fDj*hmOqkyJX9^
z$Ga5V{rW`ebnJ#zG(G2L88vZ2n(0$T=a_Ika{gV}TYxEamsr<gm}T$MHd)#<k<=7W
zwXlQK<9<C^kQ*~7lPUcj#t+F{8(d&jR6Lo|Ew8s`qxYTS%ATm@TH@kq-&6py3rB``
zhFcSyPofW@->U{=3)li0$BkktqcZfg@SfOq(t6X21dPvXq^7tp!8YS5%4OBRZeBpi
zrfKjj!9i7m8LHCKl7wXOI}y*J_t6q}U1`i9>~a?QX&9z9C_mi7`K!m=2lmi&9R&KT
zPiWDbFXebX!o#I!c_c#i`Kl^nVS)Nx9kK+~+m%c?h<8*|fr|5Zpyph92x6KmoU>{@
zkW;F?;qI^bye&nnI_QC|$NNal`<O+LhZ<c3UZ+2$-!tZ4SR!?qB!d&|aF)|@6T%n%
z%C(SSHtL-w@b|x+arD%yu~taAw-e!t|I`#uzRI=J*(m%J7AO&T=1qyE+h5V%ON*B7
zsz<uYBUiOR2YM<(<aeTuif3^5TS5Y<73fd9^L-a^^GSO!%uwU+!Q%%HaMzEq%Z+bm
zZu^?Z(5kV1*E$iIiLRQ1Zv9Kj9Mh>a-dVXTq6&+a=NROm-7tn{0yWF>-_DiGM6`G)
z7>DuJWF7h*`cg3ME}x|8tv3hRXJ@{fL-Q32uF$N&5=AP5s+o4yyciw$7j|ThWzJ_N
zlBFpdc}14cOcF(#jI%%6+s@qSP&spR?TsR5d07wq>)D5=$0=jGvjRc+{rW&HkB@PD
z0J;3T?k)D>R{scem*a<WNk&Vpeyf^b9n&_WI{mWY<+$t#%iR08pA;ZIkGJyIcvWJ_
z-QX3Fc(6037*U<jVq8?hJqFrDyij)69Z+RPFY^AOU+YpWPfJ9yYG<{rm)nU9zKt42
zlK2Kr8anotN6Ne{Jq^hAT1QeraI59V6LiFVf=4d#J0yI!cVmQ9Y&TaL2mqS3+MMlN
zz}2}|rFE5(d`A52Qki)US(O>N|0^aJ)$PHt%+Dv%cr9408#>B;_8j7w>Gshdj=mNP
zGvxbhz5*W!xEV)#e%aK2{rDJN1cG<y&)b2VRsI3Z0*}G5Kv(>LA#mf?1_6K#JQU2|
ze9`|jEKmUpF+a#j5|9f-`BEHS1^R!zlSeAk70ULO;(V=soJ;}@cs7wk=AAxX3|Y?n
zS!L4m9Tl|q3p@V&`M8P)0iBlEarV;h4{m-qUgqpdISCi+6}zrz1T{*e<2w8YI_s6_
zM~V)jOrs=j;mJs_e<!mZI48=k&xd6J1B}ci^DvQ%GPMZ6hXR?^9uabPB+nrcay3pu
z8;_6e`qVpu%<vg}A$V~M{PajA{RZgYP2BAcLzjh<8gAP8PzJo4g<#m<?4GaG>0{Gk
z|D~{{=?jkmbICVQvSz4`3Wet`5$9chJkJ9mh!ueGRHj{IThOg69fy3+MpyQUD`u3Q
zpsA>N*b9!>xCbWQA9xer$JyRGIo{_}eXFm8MA11rEaE{}&k1>6nNt?yOhaZ&CnbHQ
zMj<e%=MfP>A)c!`kjbiKwI>cjUvN_vj~Eb9MU1CLkl-_}x#<h)rwubc8LDNKK)~}U
z9{>w0Yf}qd2f!wGiK58ko`i^m)acI}k%i(nDv3{~(gRUj#IPa`rSHH~I1e?Yasl1n
zZ;S<ciU_mHkMtT%t0HD!HNeo6d=|dZxF_b)z!`jVU$^EsG|T<+0n-<b#YsYTxkh|q
zI{CE;2TS0t1IJ~v-=`*1ub<MFxL9~;PUW-|4I?U}B?^6CVZkwQpLl{5=sZvmJS7x}
zK8GA-GzM&u<r`OKspMt|DuCd1Ypnamb>BrkN8@8(6}rtr4>1_Jh;Vqp97!=iew(cV
zA=-SH)}oJ2;%)%W1%IPwukG&vAsY6b*&4bzSk|OGUeq}-p05t-8@be9EE%IQX}$dp
zbPXl`Hi|GG!_)y0IJ+fpD%>eZGQUuCjf8(1F3^UCs=DC%5+(}H?$BG@ibkXy!+(*6
zn)UF*or-lxFH3*Tlr40no*ueW$bdF5_TS))-%WEAJHGRk2lAKa!(3vbrscnEr{#!K
ziPd?W%SBXlDp$e@mFW2G2}4uE0Kb{`M>;A(Tlw6`>Dt=0HfU=vdlv)#EJ~(S#X7NT
z0efmk3+1<MN-PfqDKm6Ri+({ct86626*l4`lRh4}S*DZGH`jW(+{WOS%l_~%uCUBM
zeHpC|51>!=Jm*M1`zw7lc=2M~e`ZHD4$l3+&7GaQ#!W5-dA*dq)<E&O`F%uMDoD}Y
zDCvmVtbvsbr*YZHgbz&P#x7myTJsLi1jz2a0YTYVAombHHIeVJfQKfUx=zoW?zb~R
z{1!)rK+<|z?)fd;YYHZ&n*P8C-zm1uvJdctgXg3;t~>SBl_&d?2Z!GR4|-UvKwVnf
zr%jg+%WSNM+7KHk?=Sub6BExw_Ro^p;X`;v<qwP6W8owZ8U-$a86%YYl4kUP8DOyN
zd~7L5p&h0bq$s*8;oU5saCy^%FW@Hhg6VtRjqmJu^Xsiw=v9rL=%7mM0ILV{4uhs(
zqdEn}W$p3fb}eIIN7B$)xQ7papMp?d5)pzUfrQl1G@F&0@Q{I|qIymL%J23jyT|Zb
zjhoMaMJK)vMd>IZMyi07*7wslE#S7pDV*HwG}q|nrp@w}VcR_Nty;@^wnv)-N3RIQ
znV_Pk^_EICZ<;ixh!W%EzM8c2S(qH`TM-_=DX)nvWYPWG;~({ei=1FP<@V9?Z~?wU
zu{f^<UvC9u_6ISOG5>oHiD_eqx?JY>jSWOA(T&F?i`cfJ=}a%?8Xb6Ba$p-etkngf
zsm3H}mm7j{d&YJsT`v>y<$+5>WsS^EM)bJq9%6VE)YaU4M_;0j#y*mTVhKGSHg>7^
zNtPX}mHRto)^w9=%5-_17DF>3uF0g1YCXhYMflK=V%bMp5~zVC5IsG99FK9@XKQb`
z!F>moqb>m^u&S!GX&-if6ZnhW(13-<QxgKqRDcl*@*+cu9{W)w#`N=AqY|$y(a-zW
zz;{(r4piy&zlRD*&pxd3PZ*Z(Y*)#GVXjif-;7%@jxd?AOCG-*)ERX1ibIe`gKpCG
z*@kjHqm&<rG#qGVbP%**N{;3bi4Qu;^s!HX`jOObmX~WP?)4BP3c#A((rV+{g~!}+
zeZKQT!UxzQQ)A}-E}TxFM@E40TziScsy<|KR)m(fYwW~?pwokFs{A*>={kumj^51o
zzf+&cOwH+r@->!h^nxdH4i1h*Z~7eVzhn*(8ZTf#GBll9xEyWFcT$ouK6fS;bF^Cu
zJ$#i?Lh+3!_5g#25#ZCBesyHsB?n(h6?oRR$i(}3iKldtmwSDt=#{~^fMv<S84L!v
zok4X?<f?!P7sgqrW~F7Q2F10S72u9H-mB+3|By|^tj)PUK?avTi{@p{?GDKiKmT>+
zd%N78SY0Ax!krb@4>gk{-aZp^g0L!ziGgQ3OtvBWLGz&>UBG6in031jJuNpvgu|}B
zX%{b>90KktxUIv(!=hE0@<0A|R1!AE#oDUA=lu8pzL=q>TQ(uUpB;A6DYZ|xP;x_P
zuMsB|T|wHorKngl9)NQ=VG7yxVZPr_ZYl#&sN%8o%nrI-B*pI|@TP`4$1y`Spr1G>
z%BO>X(X>>4MIa8y+(Kb)ViA^c)N2NoF^2D?Z}H?Hv;R=0W72!eW2&A0po{A{MUsGf
zZjo*^f?uJX`s+^N4YhHYj>DM9X=1aJC1*5s!Ky<+G&)W-Df6<74>7Bn(SymniR0C#
zu6YG_tUlsKow>H7Wrb&PwsnPPAl|CRIt!otONOWDH!zm$N)O3+iKsS=bttEb%<s$L
z?y@-@yE8Qr2l>^;cPG*~!WY5v;qO{RS$`z%ws7uQek2-6O@^Xhc|9d4ru$}doa&%%
z$u3wiEoab~jgmq+<<3Zmc|;{%r*|?Rz$a)0ZC{v(hR~wof;2k%X2m-UI~K*~xcY?X
z>oNUW0lZAv;Ls1^+TQ0I$yrJNYqhtu>akq`(V5P5pqtyHYA}^YHsCR?koRR#WI<%r
zyd16BhFIc0$`?U)y(Ap-F6(<Y*NFqPt7ylWEmxJd3!HfwcSzPRrOLK`EvJn!xNlm#
zE#N{USZ+_vMQ1;n8rJBzSFl=dtTc|;R7axN0d^iF6Fp<Bx}}JUf^WLC=da7#=7|nP
zCb05(bD1)X7W-u?8&)ojAUGj9pTH9q&43DedTNU`#)KUzum(zon==N{7%wCnE06SO
zXR1&#;&7DG#`M;6mic>`e*7$E%>R+uqct!m=AcozSy8a-vIUsH!DsN_M6%T7x&qL^
z0qusc5b_;6jU$*4z?cNehd+IuiLB$#7D^3z%PYVxgou1PI`N%Hx0i=b{^j&p2{ayU
zTl*3cXGqqd(^V5}Ooex1(XO!k8<l%y6B`Zg=M?bAMlY;Q72)>nFE`+tEw^ICe8>{r
z?>(3P`;`ZuLD!EgpbY=+MP;+vGR}3p3XpE#*sfSx(rm74Hl72j1c$v{=~3yeA;G0+
zt$-3#)3lCpN~)jSbE$A1ys-U|un&K7bUpF5BOU`$__M)X8Aa!%mF%0XoH};(f;UU`
z&0GHlzSAhn>d*%$Tj8WCrmF1<ex$2}Se4G_=|W=t;=ipWcm)jI?*9TxKwc^UWYB)!
zKSMDmZ}NUQ4QN0GkPKAV@A-RjHJc+Bbo>Eo)n_2}p-Ryle0kX(RY27#As@a>G&F}H
z;iq9kP8#{Ve~b4-8oU5@I)+~BYM;=tJaX!u1?{I$?dK5B*#wrZP2bqhDK<J}x>L5m
z&KnG~;n-O+Y~Bs81ZuX#EU*Mf5-$r3l$yQ|1utlMc~)Ohbphex_gB7534Xe1K+P`r
z;LB6N#*)refHz@eB0XK~?n~AyOnVTEBd_5u;1<$HVK!QRkkW&%nA?u2rnD5Pkgl0W
zMl?t57fP7WajJsy41a`M9M;ugi2mbWR~V0>@wU#3Rfj$xF;&dhnn~t~cZPplSOLd9
zK%>qAb2eF@E<omuU*FQGAPI%WD4hW8I6S&w#Yz?vEtl>b1{<*kG@aFv;QJ-^?E)rJ
z$R?%*M{O3;7a^KUD&bnONt*~<{5N?qA*03WWAo>iu?zGyP>^btO#F?m&wP|W4y}|9
za5Fj6c806Diqh^mqeC{`zh#g^`f(<b4mOv=hP^K5odMk`8K!C!B#}EJ24nlGzoW3-
z#zOzgFq0gRC$Xo=97!FFi22i{P>q6d<6(VL7ObcXhqmmogzk`*9|J$4tH<VF#GY$0
z>`}H**dXB=b7+`<b|agJ>$%xXf!u4<1>X=k|D-$N496QXot&r^v15z3c3^_7XsgJh
z)$Ji!+AYvrXm1W;aVF&jZ2lWl4PuQVx09QO)g>cJTeWl?2YG=!-9@HKbq;#Ac==|n
zz7)?u+K0>lq%2}7s3d6p%D`icYQvLV?}a@$2F!6Y1Mw^iT#wJV#F`hJsSbhOPHZKu
znP+9=6$4ZwaXjgAwpyy{&!{#j%fKRVW4VF%vI?wDOTOOfm8SSTjmShd<b+xmfKU(<
zdU8Z-px$!`fkXWJN@7GWyG`ucL#*Id@AsVt*1bikHq|NW??m(#aX)@}o>ARn=pymK
zMZaMImx31BicsoP%6jg7m&hZF=jW*@X9JE7DmK@Y6^^?}BIU}^ftjOx(CRdszG^s=
z>o_psBJh@uS<c%j50=+g>U<{{#k8hDLd*8XxkA#|pJmab`e+;dCcWPE%w?vCGia=~
z*4Nj7LW%WB#h`8v=^i=U?q+zg)X9c>ZTo%Fwg@ev*6`z+B3EQZqVo-{j?b*ad}`Xa
z-p3=R|6mC}q9&l}Veu`(y)xdAtixn8)Ec-<q?9kERiZ1sf~<`fWMk9i(uM49#)n?J
zQU=?VpF;ZmY4<GD=lSMQglQbWvb2<U9y&Ad%Xuy1bSdCCtuziN-+j=_zBgA;9Ss!V
zz1Nq4V#f}Z(cy!sXOpgf(H>NB@nnrs?xI@BiO?*K2HXCGFCn6emLybS&Ne-NpbFeo
ziuvg4sYvHZR{W_54RrNQjE)uGZNXVDLauegadg(W6f$_mu|l0uL<(A}$Y_acC2A4n
zb((XD_RnuBn8_KOql=%VtB?>XSEZT^tL8|fWjgz&ieV(18)8t!&a`^X`3sy=mB0BY
zqJm@HinuxF<t3;_b73j@7GHO!7ZRuxi{aOtR1IEfPO4s6(u1Z%GE54^qSUXlrNoC1
zfVN}GeRwY2e~jdTLzRjDs_v?4H|BF0acIap!Rvi!#N}l?$$1Ro?(%f$EbCvaZd;L5
zl-v!=q{mwj9-_fx`AHz-|AMcXzb(3}TFHa*sxKU38uy?VLVlr6Ju&2JGA%DgTr1}?
zaOz>sUNS_4flefICsQ-)on=()S)rUcFAH3!IQngkFuK?^hAq%qU=`;R7EI8-4(fi@
z&-Ys{dQ^{}(Ckii%=!zNZh4dQKSzFW&SbVE1`u&9fCDrKM1Y6l1+cTW;%NicIlx)}
zlUn~tJP?>1q^zWlCN}T_0xWXo4wi0Kq#QgTv}FJdC@~&@!7C<TGP7P;d$sh}&^DOt
zY`Y?OHqJhm1cfBs5%SnxD}LYHb7zPSO~N*gC0GyaOMK6a6a%enZ;lYO9S<M|UO7v5
zry6{3R7dQ5@Ej|Gg<>ef;i}!w{1N{Go^mQ!h5Bd6D>jwX*W$<<Ub%&RxlBOkg7SX0
z5#2k|!N<iK&m`?|iPB<Zh(oF)p|ka!sV5szv)ZT7vKR+bRy!q*3u~#!h|8niW<(?2
zsT(rOqfj?zr+&ZwtKM}P0lAqTsE-@&gVCD@h_wE21zsBJZ|9PU>)43%8C5x6JFu@N
zRbu@W4&>{lbyHByslLCRsI^gV`E5)Plta4Sf&|Ud<e%c{fqQb{&k?4*UZ&Bl*=mv3
z6Wn=$th1`4EMb{*$&avIlqp?_xIP1Uzm||?As#+cF{R^%k=$di<lZ`g3EmR*)kg9u
zu?p09s2jaSz(!sq2PvNc{sYF&lQCj~wD4GHTJ;hEIIu_wu!n5^X-pft#a)Ox|I-Pi
z>;jJgDoY0#K<uM|U^4(bz!UHvq$O;xcdx#EdC%7?<pGMcej@vZsU4k9`L_gqc0LCC
zG;+NcWu3Z8U`eF6{u6U2*B_q1pH!B=)A9^^eV>4JG7WmH1Jdqqn=Exbe+&$M_Wk&J
zBM*)Lx~CI-pM;J<hxyCG^26}!Vc9S<SX}~&73ZWm;nfw-uB$QN2O<-?eWF)ifi@wl
zz;Dn*27n$2H<CH|m*V7k_a_H}Kshx^HHOE2K%^9xooK;s;7?%T{_lDc%bC}+zt4e@
zx05twEXrbKM2BrnFvdzg1hrYa1{kCeq|oF(7>uxetzu^J%_1_~fuEeggp+CnsPxze
ziHgOgUssF?;nG&gzp2^Q`E$e~OoHl4*@&e5e}?A*B@}_pKHeyO?-JaiXyMm5wKcQt
zZ%BOm@{Uk>!hJE^jIY$p3=gaa@=I?|{<JtBUP9ukVHXc_Nf>}ePFt0xg>OPwlf`S;
zk*UJ8w(U|&W=NJzZB+wc0}z_O<q|Lk?|ji+E*aY0Lsc{}9hA=@eAVY46AD8`#btjd
zCjUE7@@eucs&Z(mH*zk#ZQqk|BtAl^VuG^JE23r3*;TD9iH{GYsp~~pF{Dj}aH(#Z
zbTOZW4DRpvHcXo8zgOR9c6;LYFP7qAHcRO)oH%m`5oLk@PT`-TZGXF|6>XZKh>$g4
z4(#&XX4ka8bm*$ueELF#in(ney^Q9<jwXc%@{Ad7=5F|)i84YbzMX#zhj5|9%B)gS
zkOB+8TUdw2oQD2VN@6?v2s|dXIXnJ|<GDe(sD=14pSPdp9FQ>(r~9mAe^c0z6uqp%
zQ{yWkO1JyFYvCnhzdEd$BcK22`N5u+;_~>${)&t98RC%Nt%`_api4v-d6|s@LtM!j
z7%tWPgcdt1POJS#pz&7Y^fazrjSP~*9;Jh_$Oc<KvW|MeEqirdM^1>)XIjDJ_QvT6
zkBa-n|H-)^;LOWd2o(%@;bA~ROhsj;$kNA{f$<!X#C{Y(SB~D(Ol?`-0dGpjQ84>?
z0N1#gcaZ78;((Cyfk#Wso9M=_NVGB^c#&hQ&|@S62O6@Uw+-UG$YCQVDrKNmk8)Lk
zCpdn$cL7!-8ss*znUwr{?X)w|nlxI{sPl5(9kDM$)0t({@bnXS=No-nt5RY^<M6;Q
zcy^{GgJjC4tN$!HKwvcnJK~ia#c3mz+e6JcTqtr0<_BQM8y1$3xKy*J1mN2N6=I<{
zI7tu^{^c2GDn=*~7nrQV3%(TIK5(&34g?fH=_e_><Cs~+yN1$O{$@vkt-boY0H!Mn
zCT9}#V<qwcO-N<s4sFS%tt`1`U8Y(`#-i2S0V}EFC5T_;jL2NC1v~`g9O|rn*DdC^
z7IkO)>QMLVn0@<r7|YMIvrZWmSZWL?u%JreBTQ@_dNz+LBBo#d``uQnq(@?GQ}3*9
zJ20L)U2GBnby{vn&;jw;3X<S84lY51#AA-dUDC2vBgbcmh1cknvUVFQ@vv4)aAt=S
zDEg4IEL{&ffS8QO(F%V+O~bz$i{*Gg9qy7(Tb7RDf;Ws}8_EymG^mUPjy~?j$Zc3r
zmDz#0FT+kIZL_*FrbbCW`dm5Y=4EI{t}29Z6!*?`P4UTvzQfu6smW_0n;$)#)&X!k
zK?39_q5jMvE1N2P;GtXGJfli66i8|FVA`x)czj{97V{%Nhp)Cyv}d8nI~vzlr!KgT
z-5B<!z#3{Z{Zjd6t6X*iw2XxlZ`~qp%&~i++~f%1-fhV!GtVS7BmF_~K;BT##K=OJ
zriQ50ydH~xRlYVRta0+0L>L~?Gh{bG6vgwG(>c?Qlg>k6&&hlZBit3LO0&*c%XGPj
zcWji|5j=E2Xs1^IR6oZ4b8&K%2~GczEuN%`*`(M>rTY(IctQ~&a7hlP!HOEc^#+eS
zO}#V(=zRba-mvkWPKB{3a5W`ovX`qrXQKIhpY_+rdxMQW*wwcDgC3(Rmzoo#wwLhS
z5^2VAQyg9uc}Ut&L=-V}PgYG_Rf<agB#0MbLr_?dR7ZLV9~SXerfvgr8LN4oD6<1$
zGEtfOe4T|MH8C{<V^WkWjz}9W){>T75?<;Yn^&s1%R_CNEZ_(m4utyP&ahy44}=(b
z;~>7^k)f6sbTqMkjur^$?%};F2cD1yL(B?m1eTHHoWg;-LQfj@g}l0ES(HKf5AXG+
zvwAMb+=QeBiq2RcWx=_B4W`PtPoX9_D54y-O)Ce8&|dMjp5_6pz5y3gx}SXB2=n*;
zW?d`OH8I4yHgMFOphh!ftjNNY+sFjPjYa;l1Ea9Y_eM0!eJNR22dOL7n>YENr@1HU
zMWk7?a10h!%65}C)3x0(U8GHk`u0mbE`I9BA62DhdM5MbbJI$A@TkKjX_U&1x`vh*
z{c$!-bF71T6rT1TO#mu55m~u`UC9)p6PR13Et`|@ACen_0YX+J@b`ks-PM)1`j@UW
zSDzJpC;d;R&jm6?=1$$|Xo=5cYJ7&ldO;qo1XZD`q0wiYF@b5Y>x0V8Okj*Tq2~&*
z6WYo?eP8)=wjT>T{8%o16Kj}-9U_>ii-F3r@o&c8i5AUHSb^)mk)dhpw!Mze5NvO!
ztnju*#)L*%*w|6GGayq$@LUFE`nYoT!k)jg3z+(n0Pq3X8y{l{F<1Tq7TQ$ec{78n
zr+F(vTjO7INri&LN1VLc6AiA}(i;rpWIr8HLYZCjZ4R7hu_)E-#o|h~{FKk!C5648
z(P}p~@+vMZ+kibcpueR`fphvEwR&uYB%PBp@^JT~726dZ46KmL;INmrIe7P?t>9HA
z`wRv<OqEijX{)*jur-h_8Cf6v)kO5+Of%Q<2VG^>{2%V#)-isWC}akX6zmt^WpCVX
z`@_~~u{n-x?>;W4P3we9kHEs^@u#F;b=MBUWu5H;gFxpzL;vKSjGP}soliS7g^L3A
zpxNT@7pfkJRa2R)O?ioN_5RhlBteDP=~6j-KPY<xreWl^>>jPkU~%f{7&{L<ICCqA
zyif{w>UcUk%00)kAattt@HMg3<1#EXpO@m9rf2Lkc2A;Fdx$_CguXfYX2m-e45kl7
zsRF?pZl+ap3`MOEMsdi~!EO3lV!Lzz!-*l3Dg%%w)t^c3=Ni8u%x!58gXxw^Par9K
zZ@At5;`MLp)Ca4#$&|grPdSR$_nEcUw;bch#h1&U4P!TivRfV(>6N2~-od=Ymm{zB
z{;kYafF5v2?@kO*qQkd;Kr}8;K@uqjBnLM)X#XCE1|(bq$N?$^p5Jemv5<IW>izut
zdbZz-a@7>0I_2d2{&f=W*mF=uay!YQn62%o_&2)s%Wpx3^+@37*X)@g%HK<74cg--
zK0F4z(U&HzP}zAN<9d*!areD27?(c6?mhz8uH$R%fd5(J5GW#AJFm+*CYYbydxnyS
z=8id*rA#R2r!er2vF>MT)47;ufe|j~L4-_RM<JRvx+!_23|f9QXyl<#8Mydl>DMi{
z`*1-;tAAkssX_O62EBTn!M$L`M`z7AXs(w0etTdU_;@zFp^mPGiEEhbHyKDnPqA8o
zxTv$L*86qUC}}(uUiX36ZPI7luVfZh#UPqxB$AaWV*!-#$r<7Z2%Fc;fC?SzQb(po
zH%^BRq$Ji~mZ$vvAi&WqLxHprU1k0Av{5<R8x|m)@J7r~li_rX%5asgV$&}pFkYLM
zk~k$)w8zrHRWOGWYRh=G+M`h;fpxZmU96y1=y*hn?4So%e#!;EJ^3oT`ju^@lNsgH
zw@1XPz6iX`XqV_`S&yjgDE5>q{9`@I%1ra&drkn&Ef-MvZ&q8G5L9i1ThW8Ai+w@d
zPUy|3?_oB{&I~X&wh_krQAnD4@E7#rqWLclqkS$omFgd#c`I+KL(>68EnDM@JZ5@r
z20juIcUZwcsiN}=QR7I5ndR~)39byGG4_Fia6mR@BzA!=WP&j{iC=+9F{Y3ud=W@U
zgTi+CYQbwl(lEb-t*M>K!|MYVQUb0N@r{LO+oO;s*saAI)1S0zSZ^yO<oSQ;m@ZiA
z5<Qh5W;-{thUI9*j&!0vdQ~I8=6CT~1x%Qr0$NJX336T1=xVu;90w`8j0QGFC_M(6
zUx0FB?$KI?amP{Gi)-gfO)CQ&4DJk{ygHk{VybeSI<p<30$AX*_K{`u!*a>g85giW
zM(1!1y@N6_9h((j;6zSLHv9#2f%-hu2z2{i4E3c>a?(B5d^K*TLbUx+#U%Kmh%L0A
zQHNDIm{6t~w~%y7{}A~sM*$A`w&>^U1i%<4g>zgAvR|LJxc<x%Dn_B;yQ`%w1R9AM
zZWs6)W948-_<pu?&G9REslG3Gsgj$=vCG?JFiu2^Yd=_Fx}51HuvyPPQq1OF=*IJJ
zRFi~vC0Cb3P)c~#U&PCY{-Gz_Osok4ui6TQCed8<jdX`&E$Jz1oAtM`m{T9x8i1K6
zM0(HyFw~}2?^qOl4>wmz(4FzME_07(Auv@5QD$xM73D5i_qu+qzS3mW1~9U(X?F^M
zpyB2raE8hp8>k1)1AGboWvOz;ImA9ur}`oDGJIj72jzqii#q=hyjoVkxKtTD>+?6?
z2rMHoHlKbQoYHv#Mr3uko9Jx_Vc_@(=ex|XmW>W|@#}hhzuhHrw|X(vpOqVX<KY8g
z4a$?l{DL#!_L+SsaY8K~y5{;`q&Z@Mt#{mt74CE@ca{LRTAx3oBSR&T=a8MqUD&*8
zUbLE=1D#QDwA+;LFRi%zI2P^)amECOiC?w&QWjl?w0~HVAZs;ic1Y8s)PdHY&+8P0
zZ{f0<XsA(|R!h^J2Jq;@n3KU#hsZ`YtbX3!xN^H(-o=f$;z}~?arr6dsoyqOiN%y;
z01^2qW~jnUmSLNw-(;*klD1>=vBeV~4v54SwYUJ2yWmYqVlGzy?vWez<<>MfiUMHz
zSa}7YtU?a40wnO_Aa$l6Xo30mw_D(xGJtPZ{|1{|!K{b0nsz}3!!Hqlv5H;TEl<*c
zC9Fn^)yum=-LukTUX87J0+%dub%g_r6M;!GjOc@BJH1mtAjkzp^(DCkcqNb{$SX$d
zLMVe}-wFcT1@MXEanlcOMUOe*f25cZ{-j<0WnY}Ozs98^4rqy01184UQv`CI{!51W
zujN6OnoU;{5!<duTz74hdz1D3Cu}vuBd5*AMUqdcnkkBJmSUgXB=kPKhXN%owYhAu
znxU`;D@)KKSwvfPehxaOt3C}1ewzy!KId~wt9&Tw!BLBG9fydgP+3LA2v>sFT6GW?
z7}IYrZf04dr0C2KAj_ZnN;83z1x8n{9d`-*3O7@Y-Ivwyj5J5J+YS3Vr=$|?BccQ$
zV=lTYYV-&-gY<SukZkNnK3K*qExb~aGiJUTta83FPXt7Y4JtJ4I4L}BwU#T2A0{;I
zH0l3#2&NShXN`bJwHQ^(Upb9|s11jhV7s*zeJAk;g0CgK79x!~)sq8`{&r(bK{tvt
zt1@Uc_J5lsAfgAuQ#cvVoDVA&tW>KEb|@AIqk(#is=WEmw{mT>>Cc7C5K_|{Hjlev
zWL*?%ExN^4t$Z4(v$Ew`52H$y(4v1#yaE<hGUDNxKVo#WVzrm5bU^6ZqkTw}FG2de
z*-qdGoR%N?_gBBEKgK?IMZR>a_x8ki9*bMzG^@*NsZED4fAU%WjapP&+OQ**Q{skp
z+(lKfuy8d|<)pRUihWLx<GYe`uATMP<R9?w4!}|*Z`xs$d2a9@zB<zTT0D{6zo*2z
zSVB?zz^@SafCoX7vEa}T$E-pb8SWikPAZrUj3irg8b;L*rTmt3dCxdB^oPk`+;_@3
zwmjE^)fJrh$b@E6?YQ2&rHCbdHS?^OmS}Ui5OH*NYa*r-AI>rcQURb=-=V<AkPt1}
z-!XN^#F@`p?bDr-Nb&t;;WdBKL&$G(CXGL5oM=G_EPl?@z6~T5sN6?AnR=4pdgq%6
z<k83yplz&$X;kOI>I^S#*M^{#|7di`U35>+5*1<8=z33yZArtcMHV#Y_|i*R;kw4M
z+*h{rn7j0owro-6@Xt~Wi_8qOx9T)iGir(XPDd6GlfhwDq)=cF@3?+-u0)W#z95oA
z1D&u=2_|LQX-2xv=5>B}^nk@emM^{t^sX&%!-rHMaL1P?ZQqj~ZnZwV?G0f6sNG^u
zDfug~UMe$(Wqy~0)NWhKcUY0$#<{`t=L;$P9-G6l+Y#fT(%OU6i_qWL7(Sid^}_@w
z*F{Q{c;R}Y5G6|(S)<vN*4Sr|n>8N{he^E&774FSKi3d>V4mI#uBVuW(DEu7*t|WU
zPc5t5fvvYF3OAp2`MGCMh<96d=8y$gZB22i*Gmr#SsJyzZ2wVEd*&^bQzzqtg{^7q
zW1U;eS26kG_?&HE7{3l7>dXOq0fqd>ZF0Ei2I+`ET6|Ytqbh6b&!Gr>n7d))TuKQO
z)OetERWwg7%ihd5R@W@A`@kj-kWKMn(GL{n&71!6oqXP#UW9(@n0N9K6T1|R5Drr)
zPGe6hJ*$ekZFfOm{yVz9Grr?-&f7fUO7i*S;oAtQQ%NH|_?N~tn(J5@7`kN64f~UG
z(t=u(xtY$J9OsjrwFV#1uy(|^o5#|si7mPN8<8o`R4x_;n=%xW71wWHpp}#98`tOx
zUz%hfuVKAF<aZjeI7~`{o(MV|_}U|kPwL(sR*SK=3R)cpD9Fw;G9E_ht%CNR5lgl=
zD6Je*ktC~c%zP)QGd*-zQ3i!<Gk?LIH%~g;mbcIkScb0vi?bY3%#4wc83I)kMl?O8
z8C3<lrI%LA|29EJOXPoWfHemiRIE%ZjvQ$t0rz)>dvX>60Mdq(<3+<#ID`@^;hCVC
zutb-0hEW%MY)U~@4XkNJmOX`nc$+~N5D(6bKaAqI#%Qv7_mqKlLsUPa+*u`D(qze*
z&32EJX7uM9p_?`CRDX?`3l=DQ7HnJIQ`A7YI3Ob?<4Twba=)kn=wh#~QR6-DR_EyP
zO2b5e)%Yo!T6WwbX)O?frLgY8@QcSoAPUV}vipMY)AD9QN6vMU!#}ooa-x+=@WT-d
zA>ylYukgv6*}?QNeT4j9Kq~G1^sqhQHmAE{)R@>cR8TUfSNnohQ4H%MuK$h%Z44-_
zv$^(0Ml7L8ng`quIH%{{(a5QBoSo%WZBv-9pG%%&k(gy%)r)E}c%2ykp{wBmB%{pE
zfT#=qNk-{VyY^b%$@)+IZCsfRoJ9#KZVw;54Rx11{?hn?nzVX0&`#Mgg^H4*{V6l)
zL7>DzCC`ztq+McDvotxwl~b4GD%k<$w~(Ca>!F5#VSf$OK9+vaWh<$CAadc2NIhdx
zq}io)bX;|s7D0k#jWtN6Crq{xTriZWenDfRPDq<owAq#V?Lkokqnr9`@VA4+syj+n
zR5`M%E)tUKjC3OJ;vX!0e16wi&o}^JM{k9X5OVI3rhq+rfy*eBdVN6&FU@=KXxh9-
z%DfKejDZH2P9RyK-_a<tVK?c1pI24kggt=KDenrDW_6d<O{d(>TXmJw1<*8XE;4=m
zuG<ijtr-_Mo7tMt<DUGEB&XfIcQ1g0jBl0R1S}jW9uo0t@kx5FNI8_q`jT4gH_fc%
zC0tgM*DslMQ5fPrZmS?;6coTdygG%%m&tvDwUq|$b<Zq*<0-i<zHB3UDeze;%B-kx
z2P$M`qNW-9bmLwLzDM=xZ80?Wk)bD=56DZ$V`UCzTJsPs$%*RA(5@4Da(-+iLTFxN
zQ}oDQwVeqVDDuxWIUfIhPVlf%kZe{}1wCsnC#>D!jTbLoN!_S2V4tG=sI?j9s_EX`
z(h>lka|2rcoAc?b%i~We-krY9So!%2_$ue4z{wA)%;IdRc?^c%=`A)hlFjO>7pdMC
zk6mZwEtzP$B`u@^wAvr44ttBorv4Jw2^u!YlFcfr7k6*=CpJ8Vpy)(kxQ7(7CoB@(
zER<SrtEz^dU}>a9ad2jNg)wjZm8)Gk7H^eDz{?r(0{gf*RpC)ZvT<NtjEB%30I(K|
z!Ky3LifHt!kURO^W>?ebFj!Q7_qes;;o~p(cd$*+o%oOcd@0-@HeYB;L|(T4_a4Iv
z>Q4iw0a<ndaDf=D<3pxpHLCG66q7i*wbhGKG@EZP{(XWn9b3a`hs|Xflmpk!<Kd*Y
zd*-dw@}h(`3jzrd2z>RK2wYN3%qe6D!u6cABf?|BEqQXVvR(q9e)tb0eE72NO!%J~
zd2J@YBc$0Bi6Yu!qqG>QrNm)mj#X1R6OKJl6phe9ZoqsUBK26v<hPAf7z=bpWxRi`
z7AO_>IRFyl!ZC+b30XjjX;z6y=JA6*yOMdI(L;voP+duatX7f4M(|Lu*s5$C-#1lv
zxTt+QW-O`Z?POufw??txre*eh16?LnLNlFM)@UkV(5+o!6o9n?^WlwF^vE8moNhP_
zK1!-s9!M@#>oMFmq-3X_Mpklq!G2A3*VrdQS5-I3C3MjdPD7fi_XqA<Ej=OJ<FP9k
zoZDe*J?Ha|gt*ktpHb?8dHN<&(tdhDY8&r-heLP_MrT`7lnB>Y4*%A}_Y4cd$dTW^
zDSSrMqi}=D<Co}2=cCAx#wPNRT2eKCqW)UbtpTBrNf=u!*(#Ku1qZ-lHXGMt6aA0~
z<Bl%BZLu~tIb%_8{B$p>xk9WzL*I-XgpzHnv~#CT^81dfD1D{JT1*SzKPgDRAT?cJ
z&%tNJ8TsAFgP55+Y~1*U-*YC`gW?&PonNJ=H|+3)=b<nA=v;rHvmKOzAV58^PC8&t
zE)M)b;PHbzyyjLesv-yOGq!M(EE}m7vnGeuF4JVb;pOe*K(RN6afBs49#nO=|F)QF
z=nwSz+%^8R@qV{+e&-qZbC7)3dAH+r!X}S{f(=_Wdb^>t*Er`xtA$gFgGmr(9jf*^
z{(2N}MiRHn4AhA;M9FhrvLk0pOEKbyG6eoz#Cf&EMrHjZEyN*3*OeM?5kf^2nz9Kb
zFK7Q&GKNT-E*f+qfo$}RRT&QJnntYP9$CK6#>As#4!=fgW&Pj!s$_I8N5>k0cZp#H
zohO5Gu?Xf4!!1rv$4L*MC1172oP%Px3^Ee?Mm=oOEp+!}3I#E8i83Pj<1g$r;tycy
z{_E%I>G%vds{<o)S`|L#M(@ZNVX(d}r79v6aTg(PjAxqU<YG9@2W+(PXP|LcZ?6uM
zLOg6PbFX*j`}4jHuo>5?!`aijYXIM~q}|Y_Y75#^2#H<4Z+pDESwA3ak7Ajn=hdVO
z*dgH6rX@&x1N8U&w7&CrI-D@x-UR}C!1524e|xzf_tyx9it_5_<`FZ_lzN@C{?roS
z-}-Cw)<RSm<XJr+baRvUdmP-4WV77k)l${XVe&@M;cmTke&B%;^GxJi<T|M1)@(7`
zLl^Z~U~VI*Z>fJAZ7hfkTY4ej_Lz)JN<s0Ni_zut-sC-adH)`_^l29M<rCODYf!C0
zV+q6?1)WHSe@K3e&JNx`zP&zLz4;xEaZ2DTD&W9QYSM&DMjL9JR2|Wc#Mbw8@ACJ5
zC`XqHDLF~z1Yk41C@bX=v4Cj8dj)hE6&xbdF!dsZQW=!UZ`nbpf|o%(30zx*SLe6O
z(MR)+`)~W_!*7dDZG<FzV?h37m%+)RJG<t;k5<mIU4;GC5n<~*q<+ZL5(0CW$h^l{
z2mTMo@~`&ZJ$%59Z}kCp(_iO%!@CB5pB@bj1%Lc`nf?3v)tTp5p)E^B4k;>wPVcA0
z807Da^nJ?wd%h#F;OqYMZl3?cy>drMKe+%ZJFyBpOloO*uvZXRVH&VJZ*Dr<Nw!_2
zjab+XXW20((@HxUS}}j;2RfaZ(qY#P$yo-Z-s>MoJnuOBm~19$J>#!fFy}>9rAZCk
za-Ue%1XI1mFQ1R)FCOICmW$ZERy6FyXB88DL(1NI-BU=jJ70e<Di3>qy%2F0XkV>I
zlloABI_Or>9V8A~ZhByyX<}|@$h&xUQI>p=rpKWo))@swG=&5a>?mCeMRX_3Fu5hW
znz(||I>%~ALV)NCRNMs9s}T`=c9TJFFW=ik<M72@_@Cuy5=D&{C&IGc0WASm!rYu3
zmKh23BqW8~hL1nl^}IG$z0ei-lK=^qYVcpTe+N^3P;O5S>3RU=@34PfPe;bzoB&h)
zeZ{LS-4}jU`WNnYR}-}Rgf@>YF)#(6k*SZTqh&1nO<spPRTU*EUsC?K;3eDMAZ+cH
zrZG1HCU&6PW_v;C>HApir96%PcEU~SQPPdK{_n%_=rXPiSd8BMfod%3RDc-Gmr9W5
ztQb~l%eJVJIX$qvIaHiMStRU!|A}80N!l89eou{({GT})3#M)j0$c++Mu$pSBT}Pe
zi~jRlVNL~EW4y3#a1+!Gol7jZ_2Km3?AO`m;3^ZAp!YKayRsUWvPwh1WGqjS?TA<F
zjbq$-yJg1+ZGu_?{S7YM41Co2WEPrq9Y953K;P6^Vgsm@q-RQ55D>e%4oJ+K(qnU_
znBC6*!6=`)0*<NEk2dlY>H`Ftly{D$@Eo~=Eg$0i!5e|<`Ux{kA*qcbdL*?YmKXNM
zXWR1;AgAlifK7L@$r?K(TbiiAit$(>--Er<jq8|H7n&PJeLc{@e}Ht~?Bq=2UXZMP
zZF|(CaSi-er#-Q)jB1gX`pcbp$&9MLpLgq~__or%I4CpwA1_$`PgU!GEJLEu%F5ps
zhLlOCc0Z&EsR?W+1+Q)ycB_zzxlO8`Pp4YpgWNU<Ky_T`=(m_70x0obS%L5)Gs<z#
zU(9GX>?Kf15gc`VH7>WwU_(W>91MIH(}8z&uMB)@#HAcTq+&nX*bXrFo;GamuG)e{
zBgPS_6ZJ@>3d|)C(e%DybSs4{GCStMq^XP9nntH4^<iB2k#=#PIOZ~QqEY2{Y;r|f
zWqu+hVE$5p($>mE3k|xuLWkZGz&*0Zl!_t;UDjd2&6#H22?)st(isLRg%lFup#cNa
zY@%RjDqw;_n5Lab9iTOqcb#FT-Ot|xO7g8<Z8n@q5kW!`Fx(PB&NC)iMB<fHYmXB6
zG0Ft=wVMhy;t1D8F`CmxHO1x<sO3JNM1T|g-yv{t2Q(u_a6lHU{d6-}-3;;;r>J`Q
z`_B|R3B>b1uB)O!5XHvmn;KA;^+2UaXtSVnJ7}uaY}y3wfjbuEa?iU~NG;62UV)92
zY7J6M=l{$fRBFt@q`}zxVRSdY;WQpvV}IF*$dlE_2~faGRG7^{wJr7rfR%S0aPFm6
zkosAOSB6NLc*niWB`LQQxX@vSlu6(x;XCe<@z@2E4tcUd26Jx<nEb+HvjQT;^ewaH
zm(zFsfnQ@!@-AMW4jLqFVJ{<YfPVIrEhp&8ix4v5?}3vF5g0%ej&Z4mj&PGeq=UUz
z1g_wsO3Hbhg0CpiCcevhms2X|ph41G$$Z5janTt_m*u-l-l`m`rf8vCQW6ZbPkjr5
z6VFnGx|fCdwx7@cJ<jN_lnhW9gw9LW8Evf)ed|zVJHWq*HrkIi`fCuNBKDK1S)vW5
z^7}03iD)2_0vFj%J4Swx@kX2)!5btZI51&HoW+E3CZceUbm)gr(Q`{@^R8zb>?cMj
z>RN%Cm!wz}Z&`d0kUF>gi+u8j1Ek>`vVH(vAedTz^WIvNPA*z#nI{mfL#X3qSAa~+
z%WPL9(5T>>PI-?|x-|Xotr${t=;ygWP6BUgA+jdbb>&Y|C#8G8Fow^;i5NZbo59~R
zSooj10>w=A^6SP}&{>ixkP{-9%g0dQFO-Tj`H32l?LfPS7ScgxeAAhMuSnBewx47*
z5-JR!oj{0cyF77mW=J6Lwh$K(4qrIxhnr9WWqoIBkNO(TXxsU%+Gf0qF2;n%{BPLV
zIjr2o`!z~>EpprGFER%Ta{tFxcmk$|SmI5IXuBfsmU$KqmZVkrfnE1YE*)o0evb`{
zMHK&|P#v$T{ZgN-u&jCULHUCY(GoUyGEd=7BYj$ZEzj~46CA5@gssr;uGc@FO0$on
z^WF=)>B|29vVpVwz41Atd3LL3Klhhun##w>uRX$UV{A_|klB3ib&(E{<YN!s^m_He
z9!abrSJjjuq1^)@hzBA@hM8AW7%@ZW6UD=$@8kal3GpVgtz&>J#{fDIxSpWFF#x9#
zQ1R8WcnSZnz)zG{YmyrP)Ft>D<SkR*K{s~*(^}0V>?guer;m3Rw^587h12&w1tXHb
zpJ(x&4d^e&!(mMDMJa9!6WQDlt27hu9p0tBp9${HRmGfA<^99c2;V!qKpNu!GN5nt
zE}1{<pA=fDtJ0=lI~^?>1%Rjc#}u!*wZdHMBhDW#!ribFDzf`6)w)*Y-MI|I^@igy
z99VD#@Yv`g(+CJQX$jloe003lfm6`DZIs?u!A=r4$5fSw<5pzBN>q6(PkQVc&eLYz
z;gzu`Q~a-wwUaAOZ-?puZd-u#Q5;WM+(&gwsqAaJ38@0RO$BY7%rl9VSrF0$pceQy
zsKw_4_I=g=f@S;wM$2B7st62)gPUQs(xCEr*AqVThoPg7d8|@>R?elRB}#Ddebz}&
z&A!oQAm+xI1g~1f+q^<!jqOVxSK$S6_qy-(ucd&U_5-fXOWfw?M)4Hy@tV}c$@T{M
zC2n(roge&N_i8Wd$4m>1x%}-5HyS{0HHoXOxi(NGb6oWVZum*-#@TVUI+1pAWWdVd
zvGx_!kNG2;hW`VcKx4lRtSrdJAdC!cJ{qsO43czJ(P5RY(*DE|qI6jJ>N@s>uU1yK
zRcKtZ5`rkzlb5xoHQ3HF?;o=b(=pn6l+nd@%C`z@UN{$w1Mpm`K6!4EkA2G^NwwkZ
ze+7M;$5tBHzN@Rhu+}BDv<F6NXQ)*z%9WM#tcIi#mU}0^jY-o7Ud9F|jBPfjsz!9t
z&wP5$dn)&8`?o#yvqv9!lKiH7l6rz$GegHh#|#Za1H`X#R(Os3MD1j8(VG_<4?R$k
z{F$>vTPLh`^XdcQC~};9SJv=zmuvP5e^wp(Tk5gU*~;f0YO|n-wMH4$OKZdU$HeB5
zI8JJqVK^_tZs)aAl+zq$+br)B<}xbV6v6D!X8m!@(d<TRVU3@;(pM=P?4nzt5&&_Y
ztpx~fvHuG4;%K=ew8_iYzMfe|MXDvdY#!qMr@h)_K<4%K`TN~(8xsCu%haRL5p)c8
z``4D61t2VnC?^Nt9e$U{%r$dOk3Uma9QD`j?#p)fu}alC<+Jt4GI4bHZzWX_2D}C?
zS!&WjNJJQhq!K;a#z8CRw>2u^LW*(?0$wixzqQpG&(y;BrY4+=W}qheoR^5rH6njm
z=D7!j$AEPkhG|CN{z-G?Y|}cSK;5bLaDvjVtX{CJL_41EuxS!p^5e!xD!koy=02Vl
z2!ODtx$6R%9T5|R!X1v9B|#kIo`?sngf@E4Jq<@zC^;r{<Gd+jsHmp%rnWR6)+5gP
zrY0~}BQ?iyP!~4E*U6WQ7+*OhVl;m<I|~$Z3Dc_(L{xU3g(9U&^OOK8t(j&=Tx2X*
z+TZqApGVlnUF1i*fF3#S)bo=u3z}DQ9n-8C+eg1g0*4V5o}m(Bo>|)Gu+<@T-Y3`k
zD$*Zt44Hb}7&PyXX$GtndY<c*Q0JjtYAc+BdDVLMU-D4h7py!~_cp~t{cV5a-6VF<
zNR5K47$Rc^DZDct#|cEo!JWhqpb$w%uQY)Oz|$%kYkfOuLE}McB6ev<RBIV<L~kB|
z1RLM6KwE{k5h?#;Si(7<utY#RNJz~GV{XI}fMC$OFad;h>;Ea1K>UiD@UH?U3?2j~
z#*<)zdh;_hrwY#k3AWsNn>K&WZbA}HhMz(bI-qXUNg!by`bFReRliB1bU;43DCjNs
z87)EDZbiZ)xSEK{E>rP4D@~zO=&7w$5SbGbff76fSLzBp^eQcOf-JEIueNcJL)!T!
zrl-OK-R9_)sAb9YGRBOcX1o`glD}mh=;^pga&v$^->cP@{hb~MMJ<0NwMc1`Vy#Ap
zZrf33M_5+_d%ZvODDBn5_m*;Q!F*0<ly{VPj4Zs5E*WR~_%J=;F8Xaxs=_GT<&ZbM
zz}bY7_}XDqi|5f_wl8-d=UJ=u<AzAP!se|ZOB!Kl`K8`sM~S;YI+rBEcYzbepzp+^
z1_M71UF6P=#{^S`8T)?;G1OJcO%rOC<xB$OxI(S+GLdCmQ4LN4LrYGo(LfeLzmf+d
z$?Ir@^>&#D1D*%VXpORhPxQ8Jh!<)Z%og;u{21u{mbe%%SY*(`Om|<E-6|YlGLI+~
z!j5yw#TJSfS}mC#&2AwH#g~kis&Qz(-UW&x>O6xRXz!mn_auMAl4IkobYn_+qw&Uc
zn5|Jxxq9n(AN7v8jC%K)T`d^yAV+%M(tStEqV24w95f_Cb`{aCAfEe8dR8x8qLrS3
z6B=eeIHcD`uh-scegcqr5_9{6ou5vs!|Uxr-+9B-NhE3($!M1>cJY_X98#7V>@?Z+
zu%RFP2T?;s7OQ`#;hyr3c-kP^fFUCj624BCW&0u+jQG&k4iQy8AUuoAEX&P?)8#c<
zSpm><J*FSpS7t}`wOh94yQp=S+1K6gw&QP@rxkC&?#8v<VeK;XV_!4e-aB@`S}3){
zd;~XmbW3F`owMercQjBzQ7F&FPwAmViEq52qQLcBdf|Vkt+}Yd=;cZzuXbBNB{WY1
z&F-KIVW)f00*}&`p{QA<>_;v0%W5c5Do+>dOzM4#qu$MXmZ#*ol{?Nry~+((y2-!|
zIq|5MJk|;p<VjzzpE-2{hO0&-a9Ul`t)TL=eJOkEuDMw&G|ny8K41OQR+}wQ4Hl^E
z2@8~vK`4K-!0{vxgJ<yV+z%&gu=J8ouVzyayilm{iRx=MqerezRpXdsKXG13J|EtC
ztDuE)VP7V>yfZV)*2wbL_Ncapg(yug;h`h%%clm?|899#m`+zfj#C{{Y?%H@+I4TO
zgh!E$y%!naf#ZR?S~<2n*L<x6X%=oz$l=vIS3-Z^0cPH=W>Ht!??auG%L;$mAay1t
z%tXnKGFgVeL=NixE}Fd!9QcK)t#B(bJt4TAfoUtGJ!@aQ^O;2TO`GArZ)(npU|?eN
z=kxX4@`|M`WpK{&Lckc`O*G10yV<bkf5PwxPlj_hI=oKl&>%5%0Cn0Thu-;@0M}3L
zq2_=0j9EH3YnG)t_qEM3Vx+a_^@?4plL#ioI9EfQbG8}1kipD^%2up>=oMR0b2iRp
zJ^4lB<YFHRx_fDL1>wMo-B@{5&$Nzkg3L^!akgJ|oW+J$EFBU?G47<LkG&CPJ^GGa
z4iKL<w^(_uf;?sA+WsErd_>1KOXa%vy@G!>$1}hDB;g0GJ6$;|KNFa(y9ngSy7lm@
z4}EiJqleH&+g+)23#2bwHqA7N4CK=0!6Vp?Fqgg`-2hbJ4JH*=xVcoWg)s3ZJP_jw
zp6Bv|mZ*GTiHkO7mjo&6@+3%%JPDHDUrd6;;m0R&mpLr!&Pm)pDsW1MyQ~<mBvXH6
z{>@S=m3(69*}k&k^w8=*Anid$z3h3dZ4E*iSeDQXJhZ@r<ad;L9p!hauo+S^VN3${
zYk_dJ_;gpYxz4}dj9?osb-j?A6>eUxnfm>HE~BNIB{n67uP1SWaNh4`8OwlPg;vD2
z(d;zb?BXQ7HYKL_T%WiRKCrlmu*OTfL##<Z<n(Lv_WWYoO($Br%)O9N2<(3D2E!bF
zoi$Vuo1v0%{riZ0HFAF%9FiG+2{!yLY`8AhV~o<2%6?5Z;U;$Z@NbN&BEpxEel!yS
zH<ywBE)$mz)ipR5QV*ao(4@_P9f#b8UANON`~Qbg4?onGUez@me>6Doj7|@ip99Jb
zC4W)Lpyk0Kf;Bi?eh=SYl+sq(Q5q>_j5H_LUV0v%40@S?G6Ur<;`rnUWkjas`6$ij
zYkeM{2#HjDLf(faJ~{rld_AaQ-1C|J;DLj)X3H|`2n2t!ETXVhio_xk6go!~=}*$$
zM{Qr?xL8T<WwBqMe-h@N3YHlx%_>CJbS&>ntSJ=qb8hjPUo$!jZ+eY0BeHN->`LQe
zpY-hWShSt({N6a5M~##9%R55u)AQ$Y&E;C&fp3+XGb3#PISf{NP@W2-qvC3|0?pp*
zK><nghRyIt?p%+4i|gcZQ#+%<$xw|nuL;}C*XXQX^sJ;Be=hfuKD4M9$8@FNn1tDs
zP+o1zwX4a^4t8VFCyA5?vaW|j71I3@O->iJIAw|K>{%^R2q*9qI!I9?-7Q7jsT}_|
zY5znil4qDQy?9FMvDs7D2GB<Lkf%brDbHpw+zmX7JojxJrT`Xu#FiTAwj5!Kq=oi}
zSh+1tmgBK8e}#2`5PHOv8mV2HDwa%vq<Bi31gOCH9ty01QudJ+Qx$?flm6F;(#TpM
zz}X%lqDI=~z2^X}d>sN3_uyM0CBBJ|N{-rOJ0q~}u}N#Bye3imw-UCCA140`WkZIt
zp7`N@^4&JR1|eV%x@({Zuw9?;V@Z*l1V9Zsm>vpLe@LH_z^wBVd&P0DOfsGz=%s)w
z+xsfSt<Hs*EX`J^(!WiiBbHG9-%BWCt++b>M<tXo9zvJ;phi;KWv<EIbM}zLd(x#n
zY#R(<dPMCCXvxPXk9#~b9Rj#6IeLwxOxa=4m$&&W<Ff|Y-<x=P1&BSCutv%}3m+~u
zN2*4rf5hs(8cSXS6nf;1HBydLC25+YAuelJ9Y+uJR}zc6&|L#5%}Y8s$JW9=)dDLu
zwiIY(s}Z#AaRS#!*Rk|7CvsTkr)S%;4y_O@*&duLq+AlLz79NShoaY@bPt!Uk@icR
zl50#~eA3#rmvU$|5cbHIYoyO5%v`S}HLfLzTm@Y+;u@*v{!6{<vHxnM`J|k3&Z{+i
zE#sz*momWI<6~`&RA)N7b;RBm$$JVqZy->Q9a;fhr+THFnHHA0?S}{gqr3c6vj)nA
z=V|pN==RYt_3xL%Uj!w%HkXlpG!p|iG&h%VQbQG&``9(O0ok{-*)?GY15Hxvm!W((
z69PFmmjM|K6qiBaHNFAtw&FES2LvnZ5u%q-=`|IX(Y7=umyGE(3YXA@1t@>Lnp=+?
zw-JTk{VV!0dEt_E;RP@Z_!1=n5<sy0kVMdfVlM(C#;zf)0q5W6E6((+lxfAXMOy-P
z**iUCk*un7PLZQ#-|{wUu3J%!M%$#;X!K1-kH*+s3yqSuS}hu7Ya`w%vJDy0(5>Ve
z-}kbWGTKM9B)W{wolk9k(`|n%K3cSVEA8|tL-*5n+g8T#X!T8s%=gA7Qe*g<H_;Kj
z@F*_M7*(*=<2^c%*25MCk|EJw9h+v4{#vFr=HKe3MdNQgs`i*62C1ET_Dy|S(95Px
zVjf*K?KLdz8cP!I@xw}qCFs7HSd822X3ER7+Gbv6FMVU;F$Sht>(PJ9$eWE4wlI)q
zV;-32xlL2*=BJK_&|)1%lfJoxd5s-%%d?ZRd9^tWK5Z=w7;MyFZ|IJl%BImOTP<x4
zAzRfP{bk*%S&o)%tX}kt8?Z<<T6JT3^V7DmG!m`8wZbd3#@0lm7rdb<D@+S_@GOjh
zKeQT)+n8T#b-rgC`!avu^Q~u%{+M7d;bpC~t;^}tzI6*fYh`TR!(E!>t;Ymum92Lz
zj3Ki1ar%@A=J3PmM<VfFbTIw5#XzU|h97BcQ8PZUIU}5Nc$=TvY0|VXylam!c#KV<
z7aTU1981Z(N*0~_fXil`(^Gsldk(u<mI#9R!l2n|i!n3J5u1P3EMnud0!DDK&#^J=
zQz@r4Fh(n<MZjrF@jBm)Rybv25`0OcThGyK^II5JVuePmcqu&sqP6IDL|nCqI>}u=
z`uO9Uum1D>?ailmclQrB&wlvlLp=TE+q-{nH=o_V`sVi4^GrI(FK+&H^XHq-pO@3a
zm$zR(Jjce=4VQl?nqrdDa}vC|iYJls;^xzh?w{TK@&2p(n=iKCF>zKyh-{oaehmNp
zC;VGuL}qaONm@>wjUgqaR<CSetpp#H-iAumo~@U4MtE<jkrSj?Dv>3}NZPK9P-ZTK
z)X>N@WRlz#+E67KV@tLaIn|q{;dW(2&m;uWB5_rmB~gEsg+^#kh679K>;@UkbISdC
zJK7lP;0du`MDdQA7aAcoIRSLp4U8&!Wdkt+Vh|Zw#Jd>zYM~8`?Zku%&ylRLiR^o2
z1h_yd8ANAopcGKo*BN0Wr%Gxo-jV`Tq~^qwuY-fUfRUYD80;81x-W1{#Ub%`tiUlr
z0W(j4@VkGlfEf%SAA8RLH~_X}Lz<^ZxAd5Dr07voCFhc`lG(LakjyZkWP?#;Sc{r1
zG@@3*K^~20Em`l^OJPJPBm)3&3&vo&(h4Uo05%a)XA73+xXE?iw&oxVdCI`GV1M2g
z8eu{}@LNU-gcjG-{CX)Aj;94i(*tIqWF`xZ&;WlARzfXV@JlO+aSn3k(8h)lsgN!)
zUFw<S%HXwRp_GY_SP_!+b-wEG7tkB#HSlE1zR-qTiA@%wfT9vW)>6L?QH9@j@&lA_
zfF4loh3im;CH3;S8JWL(JIGdGJxQcN4+W}<#>fk;=#mtQwOdgJFxut7B_u1v&SA*C
zNKt>Q3ypxx1c2#C2qJ7M*l@iaZmd9f0*(o-%2pPd;2=4YCgKuonkF4Y2HZW8HGu;O
zgIaZ6xfceI>m@M4B<Srywj?)UYY!%Fs3#o@ZNLw_`6XqCO>l;5BUl21Yom2^W(c-q
zp%u`GP`J6IP+DP>9HfYDJ%pp8VvsZ_<2Qe}-51XhwT-^Ny!qST{}K8m0r*>}g}WcV
z`|ia%-Xaf#6s|2r9pLfJ?dJ3Q@9tkc`~KzEx0}WnwwtFfA0BRB-EDGu{KJnAe|+}v
z^5J$Xr$0Qozk6VeCsRtnM`DyGJ!WTmdZ@9`Qce#o&>9JN`Xo?O^7#{srSS7Z=Rtql
z`8C`n=kseNdP7Z~UW1IO?!5*2mGd}tc2G|b-lqnOhfhAadHU-9>u0wQ&u^Z7@#N;K
z+aDiZys@EAU;gWsRk(Y&y?b~as{1sfFK=Jp|M2ST+t(-Q9)J4l?Kj`P{Otb6xzSN8
zX@iXN^yMqQiTXU-J00ZpsR}WN|LA}DtV6`w%`(KfA$vpjhV6~Z%XM$Gz0vo^I5)-|
z^~DGICt5OS?Ua5|P{HL=Q~{9w=^$A^M6D34Xk)EDI;`WvV?<gAAGAi-iCqf4X*`m+
z!KiB^VrYeDcjWO92<Hm2OQfZ<=)%WBj8;6CloTLxWy5Q}^6mXy(vp19*g$_4Ix4v|
zep?zA+9RiOpuKc7YfQU$XYX|66)C&d6xAT43=RR2!UV6I;veAn!6rcS(L}IYBVD3G
zS712oBye(ip;{F1JVnt2cE#^&Zb=S-e6eh16F>>q1x8dud<B8PEu&xvza_l|-_4~Q
ztOVF4C}C?BJ{dT=!U)hCe?)&zqQ`3#D-j;SF|ndwLnx?83UvO#^o^vuf?T4bAaDwO
ziD<wGB-jvgKeAK~q%iE7|E7UkxsxUqBnJgCbB?+hj7lO#8vfa%3YPK#m8iubQ-EhT
z?d|~j0xp4woB??;!V3aS-ep8@4!}b!3yMb^oAfOdA9}%O1k?e5zwv*<4&Z?iKmC?^
zOWE760rPL*>i=LqI8;3MUx2vn+}|6B8ztVKkGN4zzaZ||fp+YWInI#TPG36})6Wgr
z8@e}aZ}{G*d!y}*zBl%s%ieR@doFv=W$(G{J(s=bviDr}p3B~I*?X3~XW4s}y=T!6
zX8?%v1m}eTRB~)P)nR|hJi!369TZ}%*FqwcI36f%Wa%qP8@m-avQ8l%c~97RC*rO%
zzmEa*v3}L67xN`mx3di%P$;w3Q{up8c<Kj<AN`74LS8yiSK`EKr48hNB8+utVT-pw
z=j+YT>|jr51~vlyR+os_-3IH)=OH2_h=6fNL6v?~Q~({K+DCtkMv0@8g+@qEFkVqb
zy|Nifg3AX-6b<Mr*#(b?BcWc}LL2f(ObqbV<WYtNgVYYF0OP_MqArcvHUTmgZU@qE
zo(s&XxNc}wSn_Yh4tE=75dm2TL8~MS1`6CcAVpCL2Ro$a;e=`%3zLLd_(Z>~#H@6T
z1HWWHdJC|?l8S%uUl^e5ZjvBmXUBwXq~5N}OsxYJIrAf^faWE9<YEMl8OP-X(mo6$
z3Pv3q8Hp1WD9BN~21U%K>x@thADA^RpcCE$j2y&99|$S(=rF4SL9>Mpp;3X2+784t
zPX84w(}Q)WBx%<gN9LAQA!Z9>gSck_&gXYXYR#m}aczIZ7(;aoxG4%gYaCQ8G@=di
zBUG(>ibHGZ?XcQ~QX+Qq2<Sp*WHQA<BYaE|eF{&h?NCGGxE>@q*WTio#B}94D5|aL
z#Rg!MWON`GvJ$!GU>5FxZt;a~83zMZ^>mmDD;J4_f&<Y~;;O(<LIbUBctu{h8#yPB
z3k`4xjkSM3($@pkQo;5DHI+Cy1K!ZI>x>8q8Fj%uI7&xKUi-vw!ygNa)DTY&xm>s%
z@T1g1xty~loF(Z&%4UJ$#7u3FvLH9s7V?>5QPZ(UBMhU)kx$eZzi7uVvv-ol!dw-?
z)vaFAL{Y3dIL0JVSJ)-w3R;0@$kL_A;S-PCuFNL#*FlV0w_YMBJXp5CEY=y(<9OdL
z!2y7<>4&#ZIzEXw8l6CqrI&j2H5-2{E5zlppFSLT`>nmU(BQT0J>HBml>MgVg`Znd
zuy$2ZFn^DVLi6%=MZx7#QHYCWXTiM-ciDxz?jPy?kzPL9HNEbdUUyAzyQa5Y)7!4;
zZLb!XEfn`HB;18D*U8lmNZKedqrfB49=HA$GUgg({jNJv1Y{ubM_jdMoV$PKF;A%b
zQfXfmP8L`G#=PkkmnRc0b3l&(2lx@$zsCh9sS61+I2(H9!d9d5a-icXvQw=EJ%OlJ
z>Ateel2Ndagt@{AtQ;!Z;K~41D@Zw2h{Vc5OI&YylrJ82<!T3)%{;DGLD)xJ5bbpd
zQxpSrbc*{SKJN`g67Ukt@u+{O%L){YT`5X%5-JOfn0b24t5rR2hgP4N7)qYVF~5xw
zk`*o{7b7NN*LNFXU2_JN9Tq8~1nLT(u*r?5uFf-4;UgsJsjGh@8{!_-h}+#~8#FGk
zjPWiTO7RGYRwc?#Nc_A+)1LiOYn;GA%>ZYMv+ri9=Yx9-l|-2d5d(j-;_hEvp(kY~
zh&pdg1d)h~R#7yMdqzj}D2oi!sQC0MafN?@4GLAB6pdV+<CZ=-D$6!lhl9JhPHOU~
z{=u8O%78At#g450|3jBIU)br6H-Q!Q_jxmb<h|Y$lAl@m)o)_FMW{aDh^Ag2(bV=!
z2vvJLqKRwlCz9hkYw|OlHTlk(d}mF*v!=SUrn<AHy0fObv!=TDyx+TMwXQrQiC`>`
zxT%q^<^7Mx{|kOr#haIrel!yTH#wI<pB@vJ-S#yqe@k!UHW0q&SLo=VFvB5-FElU^
zd)Eu})*!cT4{gz1^stBSsXxENr%1}SWJh%ma%_qk&iBo59#r`7b>PFNhyB~b(^2AJ
zm{nlJ@b+Z@y20raI1Oqw7yzav3~%3tU!FM^nyx)v$hnks8TxMNui9-TSegMe1L!Wm
zZfg)|e}tyR`9PPA-^FpaML^{K36cji3Eb}2x1Sz(QjRmH{a62||Ml_h;qQlDRn09E
z7KcB+J@^|afk`U}0W$>P`=7&?ho5_?Wi3vR#1F-X?_Y;*<Ta7yIFZG*R;nQjI5CKW
z){aSkE0TpDI9CK1k8~aA^1z`1`E;~!QX+Dqe_izwJ(W_BltYxl(Ce2!g4)dpsa+&0
z$)vNrL}>LHXB(!K>}M+-5h<B2(nTq{OuUEm8+96oG~dKI9=&HFjGLYJ489MGPZm{{
zRMlls9t9!^!bB%Oo_-PyLuuBa%g_kuUg9~r3^HwCvc8M31N`IwsFFkYosj6r$khHk
ze=?~V8oMU4R^lA8cD8s)sBA)*5vUjwaG%D#u|FmqwaYt7sHkHz^7`rQ1`|RtBXFXH
zi(CfpNLm1mfmyW4gi^G(CsN60CsxSYk@QgD@=TuDCj*Ti0a-Irv!6o+AXB4@r#%v)
zg}oF83nXO+%ak)(o(kKF%x1YLL{(cSmjd}UDSyf6Ycbz+B{Bw={X%#eYU>YYR40G1
z*7gzu3Bu<$Gm*dB!gEDuoPjwPE~9hYd)H<>bhJ68?K2)aW<lO1<0&bbaj;^>>m;s_
z@j8h!8Lvll(w%(jI-8ibnnN;v2@vyS&oWJgm^E@urdx??Fx^U=W4c3h7;6ZX5@$hY
zu776y`jAY*nFB2PR65j=>s@gfH82?CzE~D-OPo$KOjS~(tWgrMUKE`X1v6uC&KcX5
z$mJ5z$1%gImr*Hv8Kd1IrLx>m4edlQmu+OY5q4O$sg-)Yl)5&?`g`!<v_lF>MsWi*
zBn)*fcK2L$245j%Nur&G+Lac%LF}LcSATr!j&r#!VU(0I6v7=awks{iC`UgEIMWuC
z>6zp~G40+)fiG#n+=GH<rNtEZYE9X5&~L}BNE;{afxB7hou<DHt3--#_8wT9m2R~0
zZ5SztSd{m`*sQc1V;T$DGiCS2mjefJE9;vm5ENrv>Bs&{3fh&HQ{Yl2gp{dFx_=*@
zy$v%373c6Cn46UX=F-zA>>*q0F;p$*C7OL@YN&zZ_XzCGLd)Zu&OOAkb~M#$LTFl4
z*GDf&N?lZiW29VADc#RerNk@MZ2k1&0#4a#KgU9Ft$to0`eM3+i!WD^CbU1VHWMi$
zMXRZ6EOUHf9(A3Og_Dfi3F=!<1b;x2W0Gh7e<uQhBLlr+o7xIop-XKA&NQj*<k7fG
z`q6jjD!?!SYMHJ=0Cf|tn|xbS5pnUuq_0jj3NyZehV&6IG$R!T={*8Or<Px*?%DDy
zgDzSZA>;+h^>$eCTm`IP0@plO=_Ib;v7N*jm+cX~+}UTBX`7w4o4vBv-hX}?1O~#}
z#OvJlwo@UXFs;PZ#{nTL$CDwSuMg$d#iEle!n+C~Jmuf#l@-+DlOzDDy~J1;tI@0C
zp%KYfrkwAUU;dC>>F3b>ghTY)Aq0)Wo#3UeJds)9qYqP^*x#KyNei>X-9y4`TfM4U
zDvs=XBuy=KXr8Q(wjzcyl7HTq(r@ux*I}bQ18)U=?3{}9Yi$}eo!YdvMqzK<_3_1h
z92L!+5q0}&Ec*cm2Bz$n(j-*9URB+1g%rteTW=|u=IN-#;AkHSjg8+K3h%T|z24@Z
z$BV9N4DCJ9tC$@!se<`iIEl0)&H21B-AzuJo-*Uya)5n^?~6i%1AmRw3-5xJo`o*C
zpf=o{hE9=1GiRh#?IRR+D7rew(iz3(uZvts+Roa_#a!vO|4W--mYvLuT)@l>D@e_s
zbFBkzK68+dTe<o1aSPh`pLh|rKM%KHCa|&Iu?s<;h(x7rHj4~Rxqwl7(L(wyV9}f3
zAD4t>>nIny9!H5yTz}4L>Sy8E?a<*k(=JnUBdI)Wl@WL`v5<I%*7gW;_6RzK>=C~$
z!@#RPH=i)>M`=xS_;e)7Pgq4b%Q)0}ZC;Zt$W|veADNNs!5U^#cHjIkv-^;q=&^^Y
zt<EDlM)p)$ozvHs&Rm?-SL>7dVl3cleLvIkmsO@3)nTvkp-^z!*iF?hn&L>Ev*Jk3
z`tcWZIS!IHzI1uT)lx4owncEK<mUfo$|0(eTPn%PzsC_{dQ5uNlIdVKAILyT-+WAZ
zc@x3u!!GLY{{@^s0u`5$el!ySIG2%~Iurpmmr)2UCznJ1HJg7l2%fIrfBgEL%*>Ua
zUw(Y|k9U<VAK7si>$R{FE<gVA?z`{h<tO^`>x&dFxXZt$p}$-d4e>hA?|-{|`|c0B
zoApjuFc<5Tkf_UMBjd!PG_*-1CXtvo{Mj>oZ2B8tA^>iJ==u=1Z4`ZI%f_bB#^jOD
z`MD#2B)Ytirs03;xuB@R)nJ^&r@?6FCDX)&pGUPytZnAiO@OWgXuY7>PXlPxC5T{^
zMGl}s0+5DUV+DFY0r>t;AL8mnqAykir@d)Nj1oawVnhm%3G??K3(5K_CJ6AVUFF|}
z#H^B7VvHo(ko=KUaDO4^1(?uNL80=nrbQ}Pp`2-`iZp+!mSpWQ$qX>$Pu7VjH5K;7
zu7e6uLte6wZTje<q&N-R-5yCR3JXyf1xJNN+aeMIM%!b;@*-&6t}+&u4%m{Au*^VM
zP6X(QKp&G8qk02bDNs7f>M5JQXU@Z?XD8)0S(#SC7=6mhv=w0eDJ$cv6u~*SYh|3|
zi(@9nNsfP;7^8Xvv7yn?gU7&TFbyqJM{yaoFD69_iG;b5!g#yT$PMM!PLQ_rR3lGS
zRDx3-R0)MbNjRkve3gTa_uO`hR&>xE)CxxRhK{$!@7ah=s9C^1j%G@#tn?(c4#st1
zSs+CxKqwunj`y#}kY|e=2fiU1;I35=y4_BMJ<NYlPZ?6g#bgoq89yxe-6FJFzA~l}
zv=;2ekS<u#g@f2qW@~~W<%Xls#sJkUax_47M1ufTu;o;r2rdO7ykii`I&3_O$ntR=
zHk3~))#>nlK;erVizl4oumUrpPsErahI&QBh|$@RictOGtT`Gn&<Ks8ry5y$ksqfz
zC_sM)<Y(cG6FSLT5n_kKK`An-SNfq>xOnA-i++g8n&MQR2H6wkpfa>1pABS)Q}M<r
zei+CDsLFzdLKUuXT^1Bfp{4j{Uror*IT`nX0#%mY-^(I?mxPK1jjrAMez0eJ6=C~1
zZcj1$6|o+uD0oaQi(1ELQhaJV049{6HB^5LD~WF^p}a8%HA9QwJDQPd5DEI0WTYEX
zgczVW$yhnCB^c`l1zHlc5=G8MHLTsGKua;VdUPp|X$NENf21Bd|8Dcplj=bs0JXaq
z`sqVBrXB{!1Uj05dXOI=Q*0=H7=&U%QmK~Yiw_SZpZ5o4!6@HA7RG5Cr~N(i)vABu
zz1sYu@=ic-XPRG<w@k_P?zBvaftc@=j_FIgofMKN)L?AYO&c0E3YqrocWcp8T?|;)
zj#xZTVXa@QXeIQ3l7&sZ*+^kBMGTph!c5%IoH41)yP2#3{Zu18=2=y0aprOUIyFbl
zEm;azfa!NvxJ)Wh=q3()wk|y@i<y7Na~rHWkT^>7=nb4xvfIp6<NXz)PQ_+#+UqpO
zbEyOC*r_R;y4FG&PbG5nixil&(U*Uc#w?$H^VO!o5JObMU@yfx*Rj0EzNJq_Q%z)C
z&q>z8h1SJX!Bhih$ULCLN6@z-HJi{zX&w{b6Hf{8{B)KbeV!8jD0O~<p^krRR+*k?
zu8SEnpA)=-&r_Yv^(kTH%P{(!%uin3FV{}?6?eI`h4i%H$T7)u8kk2vvFLGI-z(<5
z*5nrRv4rn3R$J+JkSkVJsN?VP?8bVN&LMP}fp)SgeG?O#sw3Re3uhCzIqjD9pai*&
zs*;Fb_ks~nQsBwA@-=I!vL1i&fK=F%b~hyZaTS7$L9(Ivj3S;=?_ocal3rlX9;<pb
zA}j<UrCZ<Bx_wd^Ezsivdt{MfV=+zBd36Rsg@mBRbp$tnpv*)oa~23nPCRz-!9Eby
zPT5HZ1i-qgMhEMf!_0%h_ko18G|ziEe#n(;$KNMZut%J`BIh0`wdsFW%i--_dx}-9
z2fx64+~OPZPZn+U0xkJ+Ki<6~i(K-4u7ft(ht|>V?HlexQ~jvjav%F$sv+NwoR9S@
zd_O21+F)^)U64CA0T;U%H)%#YK`jMn)z939!(#hYx*>(m{1mHifLSYyXPzYDqU2-i
zQ!2+@(gJebbLF-YPqBY+<%mM6ISk~#%&r%yG=?Q(*{76L0MnI(I;&=h`IIExa#&C4
z({oO#)4b0OFLBoAoDe*BY3)lcR^+0Ds86==WwWbhW!V+xzVa_K|1S(hy+Gw<PDFEI
zU?Vk2iAKPz>xBbWs<h*`5N_rHz7tklEVKbE#fw&7YZ5p${D*&XFZQY*=QK1HVVRbt
zi22ECDOkQ;(`+oZXlnzc%$1{Pl6rg!Y|XX+sNXUhnRw}oNY-k}nshh+W0a6y_o%Cd
zZsdZyld;PB%_9e9JLqoe4E+^)p~6D^_jR{u+|+dJR?!R9e)H|dW}#8O=VT+f(;+sh
zJiok^k(m`owS9k#WCHGS=CZF9)#Umn$MQ4>$a|!GLDNIO;QEbHK9VU(t6vRICBXiX
zUu^LUH{n-ZQ8*BSGHqA;uB{PP*YHiyE>BA6$$@$$9F$(|l9!zxz^d+fDnp!Jc0h7S
zOBJzYNd)hirEl3j^(vNn-c{9r+^4$Lnz{t(^j8VR4IY1bf$RFru9>;?5Voj~gd5#R
z>W-Z=O_ih6y%>(xcb{lfsDbJz*RQ(eo~#!Eo8&c0#sHVh^w)NxlToxXr=v~y>HJz6
zRa2sQowbxGt-G1LRC<;u<R_a6(oVu-<m0?HtaY<o6Uh1MGfJ~hF5geS8}tp@F7Vvv
zt|L$Qy}W-~NzaVRNq>HBdAz7zWK{u-o#JZo5d(XH+G*C@11fi5r)bM)%Y0cuEZJW6
z<H~L61&@Mh)P_9Sx6kFT>MxhGU>&oK-CcY_Y^wCA-=k;*MM+rbXFjG=GYTx_K8&9|
zWH<NAQHtiG5mMi=g9Gu_qIKQG43oR<V7iK+`yGGWw~s>^gFxBCbSk$pyi``EIbXu8
zH%9l_kSQf<jTVj-W@&t1YMD~-)snlESsU|j7Ji1G*5$vrAF9#onxa1N8PRWQZ{gAL
z231{QuWWyA&OTOsu+~D(x9RE$RNRkS?B8J*XVR0cR>@RN_*&rJ81l_*?fi^~lxInX
zv0HzYp_^rx{6Hhet$l3Cao(k>)EjbiMUGCh7hFn|E`0G7SNwa~cQ4;W-AH5uv3(z2
zMjf<2DAyhL5A}7tfNOCrwM>nyPdw3If6GaFA;Dc#K%unVyQdJ7Q0b1n<z%g}L8tz?
z^RIo&X>Xf3Jl*3hC$Qe@^RGD-1f8Gud=`Jf^<%AgPJ=4lZj5*Jrvs{=_m3Z)qu`)g
zlBd!6BV{hWqZ!=mkwk*nk_@hW$=DNE-^<2rJ<5E@V2H#Bo4W|}*`e7w+f>@OtBmh~
zsvw-{cJt%@1v8C8T07lLqwfyA)qqlH><}1T$n!#BI3*cFB+#w$)BeRNXGBbFILUuq
zFHT({J8}SL95W^EyrP5Fc?o2$wsC!aaOV+8EA4dWxW0Y#>)<sQQe<OVlIe>5bmQ#j
zfkX-G2plIKom9vBlCC8gm_dBAjnqGq+FbWnGuS#%48o9C=Bw;TRI4P-Q8zGVi&2F3
z9wOYnaiQ#V6E$k_!3!4(p(XkC0~LSF@Bgza3J9e3O0|TCMGx^haA=7Kq6aGs=$2$4
zdIbUlX(!s+N6m07hma#WcE3UydWbC_zxzOu&rZ16^oJY37heSnE8y+lAf?g;I`PV5
zs?134q{`&-rx%(_e(kz$kb65V2l4_@Xp{>mtS=rIxv%LEbHmV5?Z13S?p%L>HdDCr
z-Fvj_Dpx7u32=F{1@`%(@QDB|UxSoGJE5W5%6?#rMZAhx3Mctte=jD!NaAvQheLZY
zma-Jru9A&o05xSPPQe+d3k8RI;Orcb;<!^#J9;S1DX7WjymhCbrqTgWw@?rE5-3*&
z^$AeNdv99W6Hr$Q4z(V5cocu8*SA5srYK)OrPML~z?eZCM>^izcQGB{Nb5lw8SkI_
zg}Z^FgO69z8!DB8<Cn)0N=Es%3FUOf=*BwAu)J?xx=?2lO7ghYgj80Pmm~%%KpB^Y
zhGOX0qbTFeEKW4gG}3(6FESD*Q2EhvmZkaaApazV^Lhfq1#{@7siJ=X^IC5tf1{=l
zK6n|XmhLjELJ9GF#pgNnw<?lvtFcUCZZTc+sp?3JViz!0z*=VjZ^zV#lN{G0PIX9;
zx@8V(5~n(+O410Ar!<MJ^){-c+;raY=Bi}d%aG|{;5rHt-luxySdp0p#3@e6XSyxo
zkC}ST&)Hh{sx<RUZ@+(;Z66kl*8&xuk_;y~CmJ7b0i2QyTkB4uAzLxOt*xwfUt{yY
zQAK~E`znwB*JNO{Ff)$z(4{#LzwUM%G}#FGVLnq$rM8YPQ3Tpjw|F_bL)~AW#B{~7
zd;e`eCo1z`n;GS&^7bZoXp~iikneZ)%1my237$NgTL@D=*NcDfTf&Q%+ey%^@b+z1
zBgNzBAmWYA&Sp2*Gei;mS#k5rXn9O|1uK2OYp9i`nrlAht0tRYscX!YQ<-w1gS{43
z{m1&zD3B*Kc7BPEodC_0^U1&5|8EdtbAaG)X04g}hGoT9c>IEq3N_TL?NHuNMf!0=
zW*_N#abaE8_gF={v-JG5oZ-4G5O(Ds;UJV{u3flhew!N(Ti|Ig1%PkCyY1^~0D*x2
zci=U_qVdTK{i9sjvR6$9`UkM}pGE!`_Tlw`myv!n6Bal!GaxV^Z(?c+JUlN<Wo~D5
zXdp8&mq9lgCx69TNpBpv6~60N)X~PEY3@s+G$OS}2Ii)WPO*KMabP8Y9b^&=GWqxU
zxE9H-u41<&kAZfn*nE5o-%d829w$1zySjb2dUGe_WQbMF@af_6#P9{h|BPZ#tI04%
zv|!W2*XjK=rCj54i;tvK2z(6wws@-jPB4oZ3^N$MWq)kHGm>Eh)8c%FkBxuxyZufw
zDXE9RdQAA|{^P?BS2PiqpWi=R{c=^v5+F;sNn1fQWz(mxSMNX4>2LV*hlvv7%=F(d
z_3I=IBNEKf>C5!v)%_~7QWL@Dqzxxjip;Dh93)Z&CMAfJAYRpD2YhsVmv>nJUPOxH
z19p{;Gk=vzvkqz8?042Gk&p;X=ZCU@+bfBuvT?H1GrG9&e#eA0%g{n6LQRn7vh=hi
zSpID389Qm>jIgSeo+Zo)tV)S6DoF1Ydwq;O&r_B-NV&r<oSm@)3p)$95(KIjar1!t
z&7J0xBTjQI8xTwt8W5UD#UsRT8A2=~Q4Zpgfqy}&O$2#ZN1_!GqD69_u^6)mVi{pf
zwlLqVV{wd7)A91tIxh~YF8MjUaLS2hZdqc>yl_T|;`$t3IHyF~4$IFOL@k3Dq6BY2
zK2WaKX%kR>X<I0oKrCDG4_ry&AjcXcn_98|<QHJ+LPp~a<CXQ-K7+e#;Ct^Q7j2v|
zrGLCXxJgUcY8jhQW!EGFOyYapRPcm#Z<{HcqJy(|Rx_1QSSHRfQz?nWZk=PMKt}yy
z<7QeT88y=yX3$J)kU=vY=cSnGlEgO5bexxBrZvc*nbsgMQ#od)HPoP)Rv_QNOu5tk
zJARkUQpe9FrYW%N+HRYs0HZLDXEjX@>wlr{9McpWVpeCFCK#mp9Lub6j9O+5GiaGL
z$e?8ovr-ImNnRU<InGKk%o^n9bu8I1Ymh&^-f?3_Ev<t529_q2_DQcVI8?plU9>jE
zla-yy@&&QXK+_>9S1KBnXt<lx^Ac%GRa2+^DnY2*{A~@_6BfSnP>21HL>uCSOMk@V
zI)Xp1Ko#fLv5*TxdL}nK6ZTS4(6o9=K%}c!Kg@|qT><%b#`0&SBw>;mYtAW2XkEpk
znpvr0l~sgQV%-!Uc2?RfwasGAf0tQ1Ld`6(JoxHgCeZwD3WQ$;cdfu&7$yt~;J~nE
z4E~71z2o2XZ%lmiugA}S{`3<Z)qnK(A3v2&l5oY~?7zlWxMOLpgjnLMf<(0u6C)0L
zCnsfyk<g_AR|Y;!R}xMDhGnbKFMiZvP04u$EEqx6=_h6c)leeAFIWaMx-a!$CThkI
zI4V^5H@`}P^{;@>Q2y0_>i0YqP=BAUK2KNoNv2A3<EK~}7j_!|mShUOVt+=kTCRBr
z>M6NUG7H5S?Ei_;n%^=A-)}XlUCAvm6h3Gw*b2{iq~bddVXeGE9?H*t{Dd4jUNJI9
zWuhG5B+)#yQdp91l<@rbr>_a_5m4`{XIuv<5Vl|GX$Sv|co3|3%_`IMJUd$@OSTr=
zNU*1jsp<utS_CNXr;4P=+kb9m?tWMx2)<640fA()76gI)H&zruoQO!72b4v~GfGHu
zd*KVe=<VT(3`-#>jqu>mZT=?v;~+RWhcl%MY;yR*cz-|lv+_vo%3^7`XXZM2@Z8;s
zxLl!v=_WANxTGs@N4V68^Nm!wwI*2RFRYm&R8kuXOR&~^mSFxu1%FSvF#4BwCj&Bk
z9s)A&_vU$0oXk&*%yUGp`=6(jMnOw8S3O~AQ=W_cqMAvOpEV(Q3i})>C$l|WtOZ{3
z6SGAw`H9))qo4Ta)J{v$d{>3@fxJUHfx1T*!_Eqvlqh#~lE=$AhfFduW#lfI)y+cF
z`I0<kxwoE)HkN1v&VQ9uEQE76(ptx+2W^!a!H((6Tk9sY){(PbkG0n&^1X}O>yie1
zXLi@K{1B0tVCTwjxvUeYoE3#TQj91gMv85L(wRBXp%AJ}>L&bSwMFAoYi?YNB{u<}
z3QHgs8|#z28sx95RimUX6v8z~_9=3%tHMsI<X(L_kxPsmoPT`XWoS@{E^Y+_vBeRd
z6smWf)~B?dATuOXiImU!l-D9x7ehitQmnLXvG7=4e_k!jkzB(Yb&zzm{6`!N-P#7^
zkLy^BA;8pvuyvL!CvZ3|$m7_iS0GQ@bn$lXe|v)!rPN&CbPgT&W`VR@>!E`m!7=D{
z98adlEJ?uQ`+w4?#xl0)Uc+<;NjEgiw_jun@(5&{zk0u+VV1~V81FYU4D##g`|^6n
z9Q16*w0ZUGI!Y^jMb^T99Mhc&<U-xat&YBxQO8YFMG%|DZHbvD8*kD_Ht>a10tz~!
zR~-9hsRjsYHa6r;kh>A@v1%kkYQ>Ep9O7*t*VT$60e|4S<EW@rczj;(MLGvllkEbH
zs(e{j6~+;cL<|32#LU)}6q>@}Wg5tUQ@{E{L|s30uezt~46B-spAS%>O)t)cZlTzw
z7e7V*uVbFM-LoO<sz9FEy0*geiJZ~jIZ?3$GxKpPm#HW<IUg(~vCgCRx^6v|{%p2#
zKVP)X$$y+LrWJY2Fj2s+?h~p2ty>C<%`Y?MV~IUxDkr(2Iu9fSSr^h3s~l7be6a&i
z`rXU>E4&rL@pKqZmT){)+*|wat8OU+0e-(dm<V@%)|>Beb-r0891EH7xqIpfy@~rX
zx~Hi@koI+RPdQga3e!-yv8NotUUSnk^|U<&KYyj*X-dIz)DXHWaV1@gq$#Ej_0fg0
zy+9L1vB{S7g~EjhY>{!q%2`8GNiW}~zx9&g%-fMlw9{KMqO+YuUrJ{;$^?oM2aeG@
za%<8d{QuT{K80gIncIM*BQ^GVr<BIqdj1G!EIOPYAinjufr^SlM;oa1b4;N$QSnU0
zjeneC3eAZz9cRC7SCu_4>qpC>s+1kbkFSSRrUF?}8GrjQ99+5f+>Wm{8~4Jg>n`?H
zEs7bz>9&9upeKF2EzlouS)!c5ZZ0&Mgz)XXauCbkMrJ|-L*yn=X|Ra+DeT{@yY7qw
zVcDwr5UuX)MxT;;!@=rO%Szoecq)Gw+JA+2m5t8MKKYKWaaH*|k2kJfaN86}@{NlW
z7HSXBgn|rrw9=96*qIHs3?aK-eF}6{b2P10`p7v=GX24F)UVQnI%-R8b;=o?d*oH9
z6&3v|&R}nw24USrW!QU;B2JVWvbR)qJ=+t_WZ``W@1Q9c<8r09aZi10)YQwAsehPz
zYU)!@`hEV(N51$MgIqZ+Z!%sprib<96K+Iv+tVsl<xW!CA?<cNqrO?+%kAVT!uifA
zw9w1-qGgt_C_E4AiC}a_!ssd+5(Sxa`e*~UuRLwF^NV6>NzCPn)j+?vVsZ8q#cFAN
zs$#XP{Z5M25^;%QbsKozmDgb<lYjmi$i+(5K)<|_(WfX`OY2jWtX=JQQnHqaOO&kJ
z!1Zfn!%Bs^QC+T74fIo$N}LFNFBB?W><D)}T&ipwc|59tZnfuAwrx$rj#|`v9R=rT
zn$Y3bQvI$TU9U)&>*DF=S=t7ZbeC&e1O4JIEW#IumJ4)M-HKH6j;cd_zJD4q(!-_}
z3XYsaiJrGhwGdUBzu48S_ARv#g}Fsh*imqf7HUTfXGgnzmi$*~R9uHkS3QlaX(;8w
zkr_)QubMe~Q<__g4V$J+<dq}`3amLbNM-t6T~vh`ht#5)+jFuhE)P`)`>wwUEOV<V
zm7z_UwO%~IV2ESvLfdVjgMaO+jkedZ&Cd3E?D4m}J^Zs}tf%8bJM)Q(Wk+0HaQLw9
z8)8i9u$*B_LUKZ_JCgqMRX)r&eJ<CE2Kvcbaa`GqYDG)OQ?;U9?WMG$#d3*ObkaZf
zmwM|BH|7s+-2Cf`>Rd}QRUagnV-G5s%vVDXOx&{em_b&tX<jehO@GU<s;1bKIkI4P
z>pncqx~YB~a67M(E?UJowrWa7Dw}!Hd#91QvDnT%iWaB&@=)x0yA7A5*tOvj)zz6)
z!ZX|?4A!?4E0I#$(!*|UQQpq{cE>&KQB!%dix+$A%rhS~*@fM%;J4d?sNVhW=Pz?%
z4e*5?>{R9Rlz6B7&s4T#N8Y85&BoiO?PAaE*TN}~?vR}0)&NO+_@Uo5o+{l1${^N@
zaJK37s@?yD!n~}=r1ifkx?Ez?`Cnc1O7cu`KuL|odjsjNI#*YD{||C2)!&zqel!yS
zIF(^26aqOlm%*GjCzoXzHj{sFKZTRj>G5wbf4P$txp&VtEKz{g(rsbpJP3r}Fzayf
z6CUsJD{XFY*lH=^8NU&@MTFs0#w^aUE(6WLg0Rs$vkA;f;goK`hFxRb5qE?4$OcGk
zK+?*HZzI;?Q^g?vKGo;V^%>8nG-I{i6-qK}&nP%ANN2Rr$`J*XesF&U21rF%x2Mb_
z!+0j^_s<m*OY2S<kM{IjxdVy*GB@!7q)xlOuk$m71dGwzp+lm%LQ9wuCUL+zR@tg)
zmgdxOiS&_#D?Af!31-~K2^`lPCQIltEHXh>rCl{qJnS*@IX|V*4`;YR8i%U=&^W+l
zkm2gA3oF@cDYPRawzGfshGx8j)+zr`WS-|XIX<_!NjCOGPmc3*(CBQ`Lfa7ZA_d?l
z7;Bb4(9GAPza+yCYD<%~G6JGoQFB2{B}{H*+fKA~Lc+W`da;5&!cF+%*T?DU!^_?8
zD~{W!M6`bS{QWMy0HcIKLX{KIxaE((FJJCH&AW1^oH-z!vvz;5WrbX_fV&5wk&I@`
zve>h(4pe|#`hS_v!wB-2cGL#_T9|iiWLK2>Y@`m^HqE{)0sK52@+V1=<!OuO{8TvO
zsI85{FgN$UD0EiT;+QLN=9t&6c4Z*N(AZnw$<g|d1Ga#q^b>>w)&3F}VeQrGDw9#u
z<S||Ix78=&yX1d^3LIB0I`4O)JqCi(XlPzlSHkl=09v8y?y~ymz~?nDTAwO*87t__
zZ9X&w3Sia9lUw-=<I$?k7Vksx+{9(I#pQa^`~EId?GAYqup#O02%GldQGjK!N!a1y
zMSW3n)Aw7Qu&`vS8)ta+{Y*}_YfpJsBBgpJ0NhMI%Hw~0Z)PlCFQF2u;(4f=&4**h
zfY+!nVa2x1V1!s#y-XISD`8D#Rd`g{=$?wO^3lkH1!Ah@o@(zf-t$Xp%P(bk)fH5S
zzETdA&Lz9)loT`^sv@|DOTtL248vp=NeN%mEXzEx<sn-R8ESxA-xY4c19LUL1y%J~
z5x$c<EJlCSpaM55dRjSMU?EI`bSlAI#kye`N0-(NIe*t6ZD)sP=BlBER4|8^apXL_
zZ?twNFZ;^p>Cj+7XocFt-z`mmh^dw#b!TGths5rgcqqapv6G4t^Z7BGcMnYAc{WsL
zHo`;2=ZUGe>-gqFz2H(Af-$wR58Q~z^x`OL=;nVspQ*Oa$|oM}!xTK9j6`8KNr>It
zDFY=xOJ-dncNR%+DS>B7uw%{;l<3y1VE}NdEIZAbM+|PG@6U+r_D8gectm3`HzNNh
zozdEdbU2T&51r9nmmEnBhNsw_bu;<ov~tYz^N*z!1_N(z^_*6CCA?9uPAdRo=b+7y
zChmW*^_mm|;~!`2yt|g$DF#HLV1>(TQXGe=nDq0Gdnt}{LODIpChzNyrZUC%I;n0H
z&LpVv#IhrqNVVxdSvfF_WQ&XPth^zhR#6(4)jUpSv!gLJZJc(vmBv^jP}Aq#wcOX5
zCbe<JYqTai20QP#r!~O{<>LyzW>wNtuY`Z<XY!s+VpcZK$wq!zr`}*ZYBRYi>5`7=
z3JX4Pt!rhc@_s9JOh;aQ%ObxzVyeEb4!~7O<tDnQorpkrJG26)TSe|ixVEpim5a)X
z82p-xskYrCrA9SzRC7+1gTAbzkjM2R?W8L=;gYmM>cF|gV!94#yK-{AK7mRql=Xis
zdMJTYeC#KG%QqyI{GM3I_hAa%WPGEDAz+TCRv7<TA%oMxcsD~Yk+$6A%k4^_Q46nE
zu+2_uK(|mbw9@n;3Jm~4S@Aa)YP95eVhew=q}O3n&4T98spth2rp9uE=j8^u%7e?=
zz4B9e+agwiLuHM1>tu!j#HE!0cR+~0eb^v>wA4e2)Ua}AvvZfN<%tD|&NqHd2}1sD
zyWc5mDM4`lOrfT5E*CtKsI1p532<R%_`nE4sEN~2p)@G-)H+2izv{>{QV5FaF*8r5
zQ#-EJBq{Z9t%6A<ks3CcAP2=EoC5D1aTc#?WM{CU`I-x8Ezk+yO{D-QH29&qR%@$&
zI%vv?+Y`AZ*NX<tUDo+=DqzU;X6^~q-Oi|XHddur37k}{uOD6WuG8@#Q2A$CF9bKc
z&P7JyFjSj4cV)$)@?_hMH~JLGxbMk&R7FaqP*b~Vl9lOBy<j!*t2l`!F?uS!&+7De
zwVUQzvZc#1+}1Y(LbGaevsi5oIh1ODUE1yhGi9HP<+}o{IVBC8*G>0wYhT9_MD#v_
zo{FLw2Io5AXsA|0FoqhqS%qDMKUL1*aKUb)SXWj9LEm%l1?JFIF_+zR<wfD!=9<|@
z6Kx%8t;DaiL}Q3zD45<Zil%TSOF>tSjdO+UIL96pD2fow+q7~t66bZCiMw%s?31E9
ztzm|8MM>yzRh!qt!^P_MctouVt98v-M%>9rF7;4##4LlA;`R>Fb+7c?1)KBPhxSgm
z5D6xy(AhW2BR~m-ESO>MgfOqk8Ke}LGE9Ju+^;FbA5uT(hEUO6?CJI0Z`W_tMku`&
z=Z=w{@7-P1xC2=(s!fpFy_KwgOm5`fkgU#1);PEmVbr(HlkTnFHk~b>tsN#^1~%{L
zY?#$f2LK}~Oxb{6*J~Y?c9Vy>lPfyDTYZbOybZjgLL*cMr#Gi<2Y0nblBuOqTvS(b
zVo%D{cBQi}Y!?(JM?m%vf&dd{Y?@PE-&Wg~td7F^URtdEH7a<VYa8!>IdFB^fv*}w
z>)zZO9Q6<dzL`mmO%bb_O`)yXH@1rv4~1e1D<)4p)0;R*V}-{l6C{1`V&N-j1Z6}B
zW|PLVN6Qd9tkhB|5#x`k<?aPfe#McT&4o*z6$)$W7Cint3LLDkNy-y<E?#NvTFxjq
zq`?*wLmg&_N2Gf&DpUb~52$`cK+naTYofsqFDpe-sFqU2)Y|-gL2^GXnPwD+iD;Ss
zUed5iym!ud{5GI)I*R7*-9Nv6`RVhoO5lQjr4@2PW>^)&^1VJ95tYGH;ah!n0t0=F
zQ&MZtI9MQHBxrF$TSx<+mpjntybX%KkA6#|5q$<|%@*f=tWOhv8vr)ca7a7$(F*Sy
zA0(^@L`Is>NKa{(x;8!H`nD=PrVoE^Q*qUA%iWjd?i2Ad8vKXf&P%t&P9;F4?lNO(
z8`4fl4GJ1NqDLj$8RdlbsTis`)G5$Gc&YFgK_g<@2qU-+;a+WrHbkgR<2wwPXe}(M
zO-PYMYCr^OlUAyK%GSc#{`>Ryy6!8k`pCY^KKH3t3fAEdhYlE1OWZ3r<I&u<l9D>T
z&8Ca2wH;Fr!0f`tJJt<ya~0gkrcDurPYXk_(0bpA{dAZ?=x?lr0}Pr&BgPnzcp@2M
z%uc4*XsEN<pJG6GRmhlkjDFsE^l(GDW3vbBRqjenyA0QVXcZO9RhVi8;QARw=hS$%
z?cKzwBdptf?kbVm^vctv6?Uuenu0#9;066DjZi~k;8?7Xu__G*s$Tfre(IKbcSYJ#
z@2;vvz5lU{)9Oid&yJ!4e<7uCTQJ*|3nSmEa)JJOhCJ)*E9w<0A|6i75DpbAGR=ny
z7Q~%W0F8%#il*_#1&c(XUJ3E->lG$M;Q(gHhwBxyYBdJ#WwK>{b0O>1MQpP{mw8iA
zw{z1x=3r%L=v+=~LyA&<D1F#*KL3l^tP&nW^O4k}8J}J^AungM={6Hg0Ix^|R)0SR
z!suHH>A;`%vGeHvCF?Wkc8=5&yH;-b4U@pAg$-^FLf{I)(+#42#{LII!zY)ad^i&V
zIXIVLm?jjL7bP~of8g=A55L^Wg4~-6`pF9n&cYgI))_(jbIefFN0v|gPB?&xSYdoh
zxYCa4FZwvfpPKXhq|QoX2~L_FvAf1EizuvV{EGZBM4$0Qufe)k+{3Q!L9;B)GFWC^
z%G0O1-njfE!HF<1!^kVz`qQ{B(Wcw;bJ?*%qtyuU@1u?~e@kPoTYJtk(q1<6$XwbC
zoz`wbXHYraG{+<i&5n!;%(7?YUw)D}62|?paGmd)d1DHs$cgx5)RqF#X3p3B*q%`b
z7BV*=thZI>5ta$0TR~Q;bdh9W+_w3Q*mMTCp5%4(O}V)KRPd06|Cch$hXq#(O}4=>
znFsozgM2Pfe=e2Bh2J=6e!*5oDa&+DRN<6su=uGP^e59(HM@tID2GS^tn#OE?}_0B
zg#(R)ZI&!E&TWL@lJE<KZr5?olo2<de|{c|j<3oblyI|MY}!2QVwRp<vnJ72x(EKt
z4`()yAlC}zH??WkGHH#%D>Gs&pRTNOBGAyVD)Btyf4+KJ=d)Sxt@Hz@DW1OkEBXc?
zdi1%Ya1(CQoQU_&9~L4(X9XHPef)a&$3$-A-IRREofHr}PXC$GuuvLJn53{uIZJ;7
z(7)+lH1W+pKY#k^<L^q)vVX^^axy}r3V!-periEp5B2n={5HZmPe)&gK$>hOLnwGn
z6Y9b?fADd-GsIH`=-KF(I2wfUMr*d1m=4~>iFE@uRB?z9=h3p?7e3gqYFlI?DjM-u
zgi_YVpR~TNia+Cr|E^PM)!(PPPt)D|!d0}|#VKB9z`@eLHLfaQR8%O&**c`Hpx}U1
zwQwak((R2=PVYQ5nqdP(p%;!~3Ygk%jpjPSf9S451gk=*Lk+cIe1i`Itu)e<6e*HX
z6M)b<si-R4%^aS;fBah3eWq3K*xwB2UTF{p?R0Nt5RSw`uu>dH(b8I6Ql{70HCpQI
zRk3h+ry76p6E*;$E#FTJrA*uXOx*j!459E6(Hvk%MQh0jlB$rwHb!{0$A~(+n!&gG
ze_<Vvp>SbbmCb!Kg-bJC3|I=vE~8Ow&cgNK6uJeB>iiuys8|ET>Qs1GkatK=6r9vb
z0H!LugnOuNuE6avHGFMb1I1C<*|Bp<i;3P*Bqp*oMVgf&E7Cuvep)g3_srt&9f9f}
zkYowEh2}5^&(%#u3Y5(JtBF7ia&wx{e}WywobRbiF>Z6s4KvU}L(?vOVLIZQ6siUc
zp$!Gl=_|50QTFjQ6J}DoDo9RHR>Zc6E*m+<p!oOgRy^Z(W8H=WQ@6y}?m+ahK|28F
z4nlWWB7HUDO7WjEG1y&d;2+CyL3yDm5+LRI^*1^FjUN6&2FC?Qp+cPcwK!dJe`n+x
z{(AR*H($0&XmSWBt*~Lv|DMJ}>>k(%k7%o~V@9ou+N_&04dn2&S?tAgIj#XK*(|K*
zS_*B<e*jso=979=d(+ZvW=XVHa_2y8>_rYtLDcr{tZhy&LEnlTr6y-@UrLVXMZhqb
zpoB7RHWnqGFWFSfDLg+<tEZ-Fe|xPjocCpGHSG_J{WYt%oadT~zU&$pAWZ><BQ#}p
zaR5i*j#}WLL$39tT8_<lXw2p+j+=_3E;vR#i@dSYeEafU`)Ntenh<qCsGx5X7Ejrl
zm0ftVePJ1jGo2duuh{QXd`xgX<ZFWSmh^tHDsfU54%@=vOD@JWg^JI}e~>FoEW1j=
zxsY+rAmE(Ar~07J7NmmKAz6_0)fQ0hDhCwML8K7DSVz9GI|%DXthtRRmdu|Pu^KN>
z)Dq9>5#&SDk1pLOj(SSS5QX-Qq1jKd&<>e*?Q2EXZ1_N4KGSnvoW$jd=$>f<`kjv2
zfC0b%JsVK<#DD<(qxB>@e;W5HdeW2kntIajbWKmt3T^Qg{BBo*WIsp`DF57ro$bsk
z=$LYX(*4P(K>7b1r`LeX4JG~NXXdcG=~WH&u#cn6VqH^~bFXR62@?xG6#f{i7DeZ+
zjC3_PJEnR|s^@E*BGMQ1c>Qii)o;M@yST4=@8oUnb+_WGD43!2e-r{`!LpIU!{L%O
z9TD0Jl!TKmIY~VPSnz$y-xKkX275?Y!!9H&%H0tl7gqT9&F~$35MIJ9;k)G5m-{(C
zAVosh)Z%HrQ8Tzm?{URKpGrZ%%*QnSUL4>3GQi>cLoh@;wmKcuJg##+1&865EoQ>)
z_GB*9bfto8bprPtf3kRYB#T!Evv|-H@>+LIeaxZUFIZs!Cem!7Wq92j-g$u#7JEHE
zhmRq=S|ssTGI*uFA%U;C!ZQ<iP)cZRUPy#S$Y3TNf`}Io-#dMWblIjB`+r+VW+bL!
z!~pGzPc&MKut4EkBln_AFXQ7B7=a{e?VWH+ZHcF;ZD)nrM#(!Cc3n&F&kMow+9!5F
z7mWz3fle)wb)IkfuLg6oO~}RaD>eZI!5F!7EVyE^KbG&ltSdgg{dPH^@#*U!{p%4b
zjXK-64lOnT0|er6jNO-!el!yUH#anwAwM}JmkUiV3YYL(3MqfZnr&|z#}UWh`%~=8
zv_O};JF`zk5WsnG>LN)6%LURJLLkd*td<rHiAJ1n-`^a0v{aI|l$=xpw&!?zyR)<N
z@c++UuQ|Kmdd<Z}5fiFQrJB&3hG;@}SrnlJSB6_9x)$uIl3ZWk5*Nezp1PPqc!ZQ~
z%4EMWxpLWS4K9D?UQCKEmf_ikP=}wAi#0ZTpduApp>-E~vPpV&DI~jFk)kw~;IWjd
zDOpil!>~kiDa$af+g9vu3of-_+iTIKE;g<uC!utc#0gWIq)udgH#^bNS)&yvWq7tn
zCsj<-&a-uRw)1R@wnpdKmWMSO66*FA6&u=iTkK9ZYgB)yBDS}Vg|!H4bnXsjF|%s5
zyA9iG!>&llX<y$Jm!sLB1$8-@`5T+gvc6kf&Z(PJT`ucqo6DH)5faDPQ#D;U$B=E7
zSX{xpCRJC`ur3EsQXPI1S0vgj96-s#vn<ZU_8XfnB@YsJ*%oUgb)`qsAZ4U%vn1B&
z7ze4k8q9xwEk&ei8YsV|YENW!6$9Gc;A+-kmhNh4wg#$kc7-*BXL~kH@h$c@+!e>1
zHZvZ@Uz?d@Y}$r##kCSm7pbmQHC=F!*337Wuuw41Y0BUzwh!gT;8#t}!@KQVS!iut
zo1mC?HqC3dK;lMEgRKyX{bV}w$0U3`1RZ`%CN_T#-Fzj6n1+>$i55SNd1Dasw`O4z
z@E?0@5Qo(XPzyC%zSv~8QGU%P7Tfp`i#-kl)Wn<~eg66E`LA#1v&R<~%hl}o-OH6-
zeRqEGA2<7QdHHI7`67@9!pZC}v!}DKUc~jqv-#QTMU?0*(n@GPkPK21pNW(bi;wkW
z_Sk<h`|<4C<@4q2Yxh}H!E0iP8j=z(noIq|7yOw-iQJkmI8E%`cg~rp@noo7Erc|I
zzqcBa5LKbH_WQM9L%w6wT!QBmNiV^ZiMZ~bSlHE~#7xv@Qi9^!*>chZTGZ;hh4ZLI
zS0C`ZLt3<&BgcSOOCc~(N+vjv#G=HVrsRKv4q4hKGLUjmZO?&7lB_IU2!vddk~Jn!
zA{5`q!=lM|QgN&qWk}9BdD7;?D3KyNC!LN;q?UaoN)IU!Q@0ersg;z7&c@&gOrgzN
z;g*y=$4rz)ZyAMh=F>wdk+8q2Y?YA6;^etQO2o_^Drw4O?@&1voz61}JD)656yJaC
zvQ(mESBH?P<e-QUI;l#jG<gCYh=y(@Cwe6^Q_0@SBqsN=FHhW$OKc2LO&2_Z!?4QM
zlCNpk5J0VDNQ45}vz7O2sp{(&V6w*4&t^aV<DYIFc7=(fyLh))oE#Wc83hh#m3mQc
zF%rDvt%BcTWM(9R{$Zq>eYIRHFOPrUo}SH}*aL2Mbh=v2FE3nLUw-p`_3iQMbTxNz
z{lk;x#fmkah?PkynyvDLNRFJx`U1F!6=i+V4ZMc6GEXpMRIttZLhYj3#TQ@9jxLvH
z$Me;T+0oZeX3yvES0~pF^62#EIjUZ)<`=6gLl0q?;@SLa`R?*;eznGr_S1j&^H=Am
zUoPJd2W<ES*m=V9mpo~(eAvX=F|OA5cv$|?ivc0q1_yw=5$Y(kQRo|?g;9v3kVZj9
zK}TVnxeYU)OyZz9lO8}yF{=kTQ^>??+*3_MB1FLI0K1aFU{a<-7F0)Kvu%-ul0Z4r
zT`j2S<TXHgD7lBIImvWdNN0cRmaRlE%|O9?*1V&IrL3N`;;Q@)Eh}w8B`OFnd_a0I
z9AEaNZ!$IsyaVv4t1*IF^Mp!dTii%r_drQV7)tcAi*E27;YOv108v%?ggS5vhQc>4
zA!<MG8iKS9)q}S|XEK=tpPo<${L)zG=n;ZnLi+X;Qz#lh7yJX{jFo@7God9Q&E6<q
zip?7Z5&W{X-58~G2(=V`0@$RIO&X#Jzl60>P6<u;al$VtSbKt_w@xy_8=wG1wdsUP
zWa9wT@@|+0B?<so_CyDWn<NF1rD!BOmfWXfAjwi^Pg@3J-chAKFaz0Yv0)&GwD=eU
zu?pY%w!*`@_rQv414e%cU`&v#bBmE$)sv}iF_JU$q2~L56*s`c4Onpla14%>!sdcH
zKtQiADu*6o-du3vLF#6hb-`l%4d|n7=DUVItm|`k^kMz2E&6C9*61VF=o_rjLKuZO
z3TYH%6m%5wD3npCqtHfSoH>Tyk4LB-zJQdqcJhq%)e1}wj@y5_&v7@;Bz6dtrutF)
zwAEOrNOqMx1E(cieMpw~$urhS)apy!&HKh!LM$NSP$}4f+&4{{v98}%iAEumbWn*H
z61!Kp6L2z8u1M7rXaOGxIQFQ8s5LweYXK;w=#gn4Cs4VpQ-C6)_0m3=LdbnuK5`xv
zM5qxTBJ0LdG)#Zs6k;A|{o;)hnhX@x!<Hykd?<if$!%TY1ay*QOOj{JsFa+|DVUFd
z5?OCg0R)DCQ30qGXat^16O>_NepL6Z7TI88(!*MS;WTFsr3iqfVV5=yzqj_QbTS0l
zEuodzc?3B29x!k(K`Pe&6MF%nK0!fYMnk8=I0WW?pe}!dSiT}>i;=f%JPO7_5fMln
z2lXJXdA$#4RJ;%rTxElY>4B0}G}1aGDboP}4u!1;b-0&QWw^N8i?tj)@BsRvz6~$h
zqL)1_AB#{(TkjzUJR?{oyN=83{ne*?J~u~gz*$Vg=ueN7*nPa0M$Ta<Wk|25xCe_>
z>lem;2gHAM1HauBekIv~UmLaV2)|O>!mk!bXcf~0kSLH-TCrD*KHAGfW25#OQk$>3
zw<kcw<6ja!oI`R-leBb+DU?Xn+t!U{M0*(oKC7zxS~O5D5)3&%!T^KbbnBEpPB*R(
z*m`uY7$DzfBsIWL%cn<*p{$_y;rut<^)RWp?(ctx!{P=-xfT|8^>=MM{;uV>_<Pn5
z{+>ookwz{c<1aFXf{ZLfMwTHX%aD;}$jCBeWEnED3>jzEab_K7)^TPXXV!6M&A&hX
z?B$XIFbeg=0Jy!>uy^RxEDw;_*LA2-sK<Z-b{!}!af%v`OJlJN{G@6*76B7ivH*_U
zcC&v#DWC!9ekiRHfhaQtF)H0B95yGgmNmp`@T4V_%D(Sy)Y{sqrdnWU(g}w!1@glb
zr4Af%X#W@;NcOEpvsXM1WQqn9Cv2ruz^}cB7B4jo+=OEPFiPMEzRFeyP`0;JdQb;z
zSJ%)52L+G=ye4QC<KXBZ1B$4j4(+fC8xVi^k?6zd0E%M6=8uYJkV4zgJQyf*-;K~x
zTc>ap)u5nvOf`X53I);<^hO6DowrZA4(X6<%}P+Kz3=Q$D4n7y*wDeg(gJ=51^PC7
z@oHaV!PMez<Nk_=&X*T|^hQdiXjxbQT4$7H)qrIgkV^t20Q4W|3lW3d!#8vrEi`{x
z_dpJ6aUX9l5_~Y*=?8+qU`Dn#zJ2K3?`xNHIFuNT=o#(9Ff^JXQCc%CfFh(~39w1A
zCYtqF$rpRr>Bb!(Dn3rN`hbYMK^e==_EjM!Ap7l_pmA6RozWC>oa+=HB1jCBhOG|L
z?E9&MD%?*@lAEd+nxb@)7SW_*m1ckcwTF3+MeX$jmm6N+p+dJV5<Kz%CFKu6Ae=)A
zSvzr=Z(w(=x1pprMvb2#=xs(7wd!??k-6@rxO+yL)hWQY7-_X5_5a^z4R^)!CGOz)
z+<yz6FXaQC&m#`cbq{{t?{I9v#hYZ&`sl$Rfi(spU$F(Eu%0}+;lH!T+u47I^n)w@
z{n=Cd``PQ&>h0B^XR}{^`Nf~VeB;lSZ)UHSXIHa1qmGXs=cAuU?#0L1`C|FAU%q;C
z<<H(O=I{OM`Qq*Da{jNg)5YTS<zl`#e|dR&`RnY>>DlG-YWCI9kKaE%`sr^+&!2w(
z_`9E;{_y0R$H)Hl>dgXAfAm(r;r8hbb9g@4*S^tfHjoZWhH1rs0kqtG!#|&|UcY-e
z?00c`vG~*J#nt)jukS9--_9@nzpp82UR=T_xb(FFAO9j~-^g6jEk{}#@)|$?FMdCr
zM3<3%G!p?hmw}-v6qlW&G$?=VTgi?iHxj+qSL6>A%_X=b&=63iZZ+WBc=~4e<gtC3
zK8!E(`x7K1xaFc$NmZT!=AbT#;u4JYy-2m3{+Q(S_38E7(+?K|lNZ4%ucvRnPYS<~
z_^*;U=<MW`63(dU+xO|`Pf}`!uUCA5l*Zu8<NJVr?djR5fHM@%Q22kXQm1DR3P*5S
zp0Dr~@_T(bJ%a*}am6dGfYUGEeteQ79RL0D?de}ny;yE?!3ZbS^y~MhpMR0lKk&;R
zCn@~wr+=rZ-zP9in4t0fKc~Mx{cZQMvlGUJ$$2d#7{07|krGfi6jNFsFX%rtBPPE9
zuVhvJ3sD_9UE_yKP-}l@=>P(hTb$)-CyOss7@TR>bT)m~dCWK(qwj27=T6Vc>ZET=
zTe<udT=KMNyR-!fj90QQ-Y6Q<+EZMcj*)rmxNhsH5a#F&G;(EjRL8m%T@X0wQC2j#
zo79uDPT1CAS#Ne`9iOr#U)XAF+a$DRR|epBcDPt;jWiamb}oOjP5zc`h>G6zh&c$0
zSMKGG>eLi)kUys}+!J(Gb^1Pz%{d4)pR=wXK&_iWJrr|WqJajXN6Gj!$WmyONL!LH
zkaUKmJ5$*ElTnZsCpUY;`n{Vejj<$AP5@zHX?sbLdLjG!*6!8P$pciitdis-7qX{i
zS1*?K$(vHxr5%4m*~ifcG&ADZYr09zJhL!%o}AI>11z}26k}h=fqYuT$)|ToO~Qqx
zTgvk>ryErEes0$vbkOkz>HZ5{&wxPlX^B<xK}jdBE1$Rw%#Ay_Jf7U~!cO;=#N%h-
zLRd(D#Rph3`_?$*6aB#lePWRIkTh?yt<@Nu0ON`Bb02@goP1<JZT>Iio`5L;5r{?f
z<27{2N3|}wpK3~KntON=vCy?7#17c2?&JKgJ6D-Q?$@nJUZhuYV<2pN&SM7Eq+nLJ
zE!n$*ma~|hpDf}@@8&ifY=b42h^5#IJ*k6y=L1N^h_HHO<pE@m?jmHj3yn!$bHql&
zVVEPHy-R-y9pmuX`B|=n<g5kExE3*XP{p$~>tQV=bE17I`q|QC1+`mxGSZ3Jhw*&2
zd{k0Y!W)!g5Q%v#h{Mjwvi6YeoW$>nI+OE7n|-EOa*gfgnmONSyI#s&+pQ7=Ypqb$
z-9iv#%WaWcB$(sO8n?CDVJKJg0I?aSVdudu;81@ugi^&lC+S+RZCG)leFknZ;pk8^
zPE%N7%GXMdQq9dBT$w>mo(j>WjB?5Gaipe$oEo~qB`ei!W(S(to_SuCYDHJH$3j<>
zm<D~tj)Dn+$?|In;#vad5(-+=n2dKdnhI@}D-}Z+b&)<r4ShKooA0(;4`Ko~YGEH)
zF@}Fq74OAcq|z!$CkzBGSw<(-%4)M>SL2(c9rB#ClglT_b5nTteG6ja!;~fjGyhCw
z8Jk8;vzC)~Yl_TRHFa%-wjqY00DJ{w&GaV*;&Sw-ubUcG4$*!vqf~~yj`|tgc*&M-
zAyuwD`*GLxm^O>6Q^ILIO41e$_A&CeIlg~asxL#VWgo;0(AG<sQHL_+6OM;aM!1>8
zFj0n<CFD;p^?r0O`UYI<U+wCyl_d+iaOD$ZZ!I~&)rbYRAxDYCGP0V9_}l28O_cJ<
zfpp3e4xm{k&?{H9)q&hJ>euws=5bkIcYb$i{KO6oojvU_Q=>an*c-vS4Hbw}X<mQ$
zg+rJSN-a(|CVp2|x-k-yS5Ssb8qAUCC{8uotK)^5JWlOesL~2eS)`JUR_`h_Ika-s
zq;rZ%_(Bv0hTKuKw73`hVr8<;y2^EPb)vPQvf%-|U7?Z)&@dN`2Jp4;C*fU#`YM0g
z1U)bd>0#VeeWnCCyO?$2pRdEG73_awCxj!ycpI@#{DN_gQdU^R{eS*8{T|^?zb*I9
zk&)t<jsM`gl!?2Mp9mn$5^o}%v$jjbDuP7330N#`rTL!%mksC^<-AX<N%6~_QAY}v
zJ2yoV^ty|t@uBps7u3|}p$d3J6Ps$#XbH427(g_umu}KOuQ+g0RRak05YT@{7*|oe
zty-`QrVi@WgK%DrJ61SB%77>f|C2=QMp_Yr7kKe;N7c(y8J9{-ewh7M@N=`Y+A4$^
z2ZhzhDy&>w?M|%q5KmJ`TxIDrZ6u(p&S3Fgb6D)gE@khRm;6^M0pQ938Q2^X(F+$d
z@UTM_k$fVTg<P(k)OUJ`1Ob0mji+w82?BdB-8lPkaP*`q%ec!Lh*oFmCnW3wkmDau
zC&y0RFd>EfaH^q4VrD!E!;|*4lDR=M(wYhV7R%YNv2?`)HUJEH3;~$(2sD=RmkiwU
z`aR1<l1ZnmtB+49{#bx2Fzjle*rn#WETLwy43!oWxxL*myTNJvEUbSHU1(+Gb*xb5
zO<X{<F|WB)riDxi3EG$VZPMZmhNt)(PF-5QG$UH@NC1%R!w<d*2$SYt`X)vlEjJ;`
zzU~M8o%yVm1m!iY9-7|VL##`s!o&=!u2}abOhTZ~fWy>$edjgYd9B}IVnqt3XTroY
zJTM2HPn8)kH*sy>X3>B13G`sZ=>>OC=2cgY$L7r^nvvN!^C{0S9L1S6{+G|GX5W9v
zf%=@yc=tQWhKWy|f*W@&EUHQfnT-|B$UMn$PmJ!!PZg;H0B6!%;dy*_^J%5vwH?qr
zgE^<`waoT3luqieQ3hyVd}6dv$3540HOhvW&v!bJOHP?fN8x{xrfK!p4n_V>RvlDk
z?YT#J9yu|y@t&71`Xza!@#X#(J4Ld>-u-c}i;p0rf>9UudL3Y5wqe}$trXWqUIin=
z)jBu*cNuPCuN$R{5}Ecv1U0wK&C&*h{_#2Gan~p)FvsPbu1Dh%#ydhA?3avNog5w=
z=UlE`)Q9RM?HPaV&FqbM_OP@1T#Yp{I6~yZ3bP@xbqF~tCEh_)DX|>p^l!lQ8Kn;r
zt=v>%8s_?k2qf*=jqFrit-R36j;Cm@@48y)8n9vwf)#W9{og&_)idi;g+I3|ldste
zx8pW3SmTceh85A-5dd;vy|mA3IzZM3#!a#YC6Hw-6e)k#<qU|17h-a&o$>2Nf8t((
zW=wpH5EiGmxh5R)9uuiO=Eh;CN3<X~YsMcm)rX=F1eOTJ;<4}0O&oWAY&Y>{$3zWk
zqerRsf(W$mpvDz<<%GvIrL@!nEA~#8=ZJQ;ZJ2*tHJl=0y-H<TI!t~<!E{b2x47(9
zyVJ|`Bkq4ujz^clyPE*9NuwZqR7;4u(8H;lEus5;Qw1qheb(}bY{fgf8Kd~|J%43`
z&|W{V=RfTDfS&)b<CUKOF#Dr>{>Oynp8u$<zrN?Mzzd_ygL?ksZV&JItB7SR+|lzN
zclel||8VM+p8qiWeR}?5|6U=MHnV+XRSQNaKaXKMZv`}m{~*U-NbbBAR_jrs0ps7}
z)yyJ>6QFI;y9GSIDetp&6*{7EAN+J?IQ2bBG35nm1bBbn<Axc{t-=RCZRijE|HFor
z8ihe(?fed$5OK77)02jk|0Vh#pl#rSmyv!n69G7v@y;g{m-X{IDSsQwZX>zz?yu<F
zvOvV)`vDk2B#kvdZZq&L$RQYHCWo0IyE)|dQ^jYI-7U53#v%sVve?C9y{kx;`Sg9_
z)2FBN>(hs;(vxG}2q&l4Z<Bx*9DWN9gVsz=2xgU-UVl!1edb(Rc+T)>&Xt14!QUP}
z&GDdw2M7TW0^SO7JbySXU<AMtynx52zvcCK&_ZjDGtPO2IsWbSFHbxZ_u*>%6quHh
z+a=1Z(+t)xl(lu?3C9C`AaW$q7<eS-8b0+E9w5SSAXp>8Fqqx~EBYeT=><Lk84$-!
z982$wh~(W9m^G3`inG8japple^_+0Vbqd}de}=+QBUk4DXMg&HZ*_hfM_h$>iNnh2
zY+i)2WmR}=Uen~&GISiXl#aF%u5CgdHEB+qOJxjrIVf6rm(8Hc6jZKob(OwLvSOTx
zSTgpg3twe}wE8?#Wgg5(Yp+wpbrcrIFH{zVI($<`(QFWJ<e{z}AyNA?ZEMb?sqRsv
zs=ZL>GW(KwqJI*NuA+-=2>0CN5_wJUAJI{oL3I>I=2)vIe9U%YiH(GA{g7oX2B>1K
z4PR&!RDP%qqCJ9BKF7eYC`l<9peT}5mF~nlp?~n#vt0AnA{qBgAS&Ha294Jf%JwKl
z=BU_Mq})|3RO*t-*qrD{M~~0gCGnKZmH60&yebo^n13)~G66!I)LABHZkC~yWDbH+
z5B)|tAs<ohmkfQbTxRKLzvLn7R%~pELn{{ZE-Sn3Bgx{D#ySp5>^=7weTI(oo^Eqs
z%tj7wY8aB3*s5OwySsMNHxfZkHSTrM>SlzT$+BvPlV~kyJiMjfjk0}todpxaEeqfZ
zUw=ORI)8ycKTT1pPn@aX&Gg?1hhY)`TkGIcX(7V&8-RTH_xEpq{`z;pV8VZJ2A?#B
z_~56X`Kw|9Od?D_@^{M=P^>1w0#{Z~+Cv54xvvxJz}M+XOUHyVo{av$QK)o0)H-Cc
z5)3jY0+t2HP@w@aw$Z%a4?0+>s!K$oD;f9=v47;b_!Z`tS@;zn`n`a{tba~V-=?R}
zRh~h?4}4k&CD4L61=GSoIY~rXhNNZOVyT(M5gfSK$|z;DfimkYN=bU=pcX`h<kc#(
zWrQ$Xh5%Q=P=+dL!}y8-240g$l3f~Rf|!(MUOTi(k;R%s+y4IgGtc`Dvp$i}gy(>9
zLw_4P9gG$b6a?amRXo(nOGRl0FVH2qo{q{qx8f~2S1S=ydQK~sU__m4Knf&$o(PGF
z*F$6W^JW3TX2Tf0K|o`SB!Us1Y3Hk8R;u;FLxwJbA@K?_SgQBESj2#i3Nd5EFspiS
zvy&t43&FZVS5RPEGpuVh8>w&libP~S9DmJW%XM>wP<;h7+bzwjx;uGC^to@zgqpIJ
zOsJVdGNFG>Jy?$R2NLd!GkCz@tXkTkX$OH2G`FO&OU^i#x;v!xjGk2ky31X<<}SUx
ze&4jK)E1mTs!DUKilc<VX1vG1jse;$qcHX^jDGpFZqL#%?#6`(v;E$Dg<K2?et(9A
zY)f_{*96AHOfYiemPtLvmjrv|UeBacoKDK;<RsO3$i_3e&80$7gOX}As5DXxN~&T|
za{H#eLYuO#x>vr_<AF$JDv;~kv1A_T>PqJH?$T3PEx>l1cYLl$Mt9+O<`^36!1_)j
zm$r2)A&3QwD#njb!#4;^zs}Ig5Pz7JF)-naR82`J#U0jA({KNnzCC>&(PkrpG!=6d
z{xUq7b~o}RpS%SQwg%M=)GpMr1z5kqdd)(NY(Dfc1#o?Bhb>YJ)*F3gj!{EFrpyeZ
z&9c!*SNRpm?MA+ktq4e4=oxL2FK=9oW7en<8UrpK{>aPY>pK_`cEl)FHh)>LVX{ul
z_@)PHa{Dn0Klc_`xPxq^!}7K}>uWqbDl@re2K2I{TzZW6<T9*pTPiQn0p)<_wec`6
zkVt{&xQkn}V?9B^jUt}&y}XMx=^^AQv5__D4TI;stSNF1i3-iCe}ry@N}iKlPN>x)
z`LHfiPLH#2Yi)VV9w1PT-ha=lRXsu!Ok={##3R&As<e<S2tT9hvTE51%2?y<j+Q-O
zykpq|G~Q4co#SUV0p*gJzCpFHKWDhV9cj9)TPb1tXu~*bNs2jML&gCUat*S>3#kNm
z+T~ZIxErBT%sE9l0s84SD{nuNPu#YOD|!Bn09sI%(1^xtUT%>QQGafwWSf&8#q+%o
zM$*Acf<bgKsAnn#jDclx%je3IP(eWT2-Gke!mgr@W}FyUu;eYoMuPKw+h9*0u3Ain
z1?SK}XgvY}7$O;E=#PK;CGA?E7Fe7j87$pT`NJ>0_1-U%?o~)ifgeIuwqJ!bV0=&9
zG78TcM~rK6tHjM7x_|8qlGavg3&;)Pebh?fm=tCswKbCU(2wuE5?46_W-hd)`R*M&
zDG`{GVk1wm;ee<-lfcfIBNqiDNoG?~FwBZ#gGan%GInq%VPW^9_jITp;~oyxW4z~3
z14pOd+o3j5TOF#0-qE4n;6B_kgE4%diLY+=s>AK3bERZFc7NF>MWrkQADL0SvkMn*
z9B8`Z!(K6=X40&Bc|$43A@Oq49ekL~r5_5#rBjM~Cyvqv=9&=C*%XTdg;uJhi@%i?
zhL9cX+Uk;enp94l((!X1O^|ecF?TYmY16~PmGg!x+(<#Y7+t%5EXUjhBRxgyua&!#
zr>-?X-fqUIO@9GYE}HU7>wwZS@bHP249th|Knsi2U~Zph#j~oFFrST4YR7~$>xtv>
za1X98q#ZHKm2jWa%>cbax*0IuNjC%Z?_R%cqqd}*QMUh|_1iX{O8qwSbjm#Vz99$4
zYKxIBhgEM#209)Sk2Ky(1`cd%##t)p`psnVFGy{|c7Nv?%H&7J?#nm(SC4tEA*uLR
z53OF_aDuEc42{c*^4xZ>ewOA-p^&zZ-pFqvcqol!ck`>-?P~33ywcmelz6)St%J6D
z`Bsp)PjRV|4$HZsv72uHgbU2}J*6>ODM!22updeJN3MyKqlr+p4#N6Qr#{A%P%8{M
zDIF`Ta(_)^a!^pZzx1Nm>u67bt^HLgu|mp~&rRlCA5pmjulQoJRc><+C_>vYHLKL`
zM$+;*Ctp5vNmVJDVl%G8A*rTfS)m@h8J^jKMPFzgr<bm$@!D%A9f8#nWxU}(5ewz{
znvHbFu5@bRC6H^Knx(^8tJ-JKIw<Jqig#(X=6_viKiX2Pv4kY0C+a@0P|nvb-I^;M
z4?5{%)}>4P1@mf`3cb`JN9-uCO6;i7Q6huxtRi!+sSLLGZGy(s+u}*Z;j`DAiC$If
z<VZ4gr8K7<uYAAD`C=)`vDk!>Ers-j<I^YLjSU8#3Fk-5+4J{nF7R>hJJP?onAp}-
z0DsBdn3<H~nGX6>yaJcusC3tH>c)6#$DZ*eKT7l7-Gjf_DP#hTeciUIdBa}1(cDAH
z<rXdAX13j=6AfiQc7%@DIr(3TbIU7h5fVyQujZ}nyEf}66NZI$k^x@1xu6H+ciXh>
ztV!I}!lX5&p=sEf3~d;9>-M1n#&+e1>t4fO`sWTl{#%5d?Zj;j%2x`Ls30lfjg%CQ
z`I)S%XY*{mN?J92FPM{~*7fZgQhOU)?fOal(yv{mAaV(1OF2<Xty4V#Su}b4bZzGw
z`jU`1OFfoqZGry-K8E{Pmyv!n69PCnml1po6ql<C4kwq61vCejT2eNS0nnG}QZ{dY
zm~_O@^OZjWyVVG>J5h3sG~H4Xyy9boR)&-)TWD`XDG2L&`~0MT1l>{uSf1Y?(*|a8
zljhK*OL#qyX6ajzPbvSK#+6EN*ECLJ{3*;blTKMC$pI2|LJZUb^<*#pvW$s_p)@`9
zQd8X1OEbTYx(IT=#BZ+$zsAs|E9t|3OD0_O$Z((ZhS!T<rdgewmZN-rNHk%A(Q-d1
zny`w|^{8kv8PCIbgC^78!ngEBp0gdI8C`Obed*(i!4`}B=d8K#iY0||zxw&;Rktuj
zXbjX$Js>x^LJ-V>opE(t_iU*d8r##eYl?fGT|<kPWKJEgVBV+X25^^oh1bb{oULv1
z3Za-0*kUayGKXx!P~axWW?Lg&PalEdYNS+T-c|sa(xqjB7LrE1m6VM4qKJKYF@%MJ
zX))O-tQg~GT>Bd#G?UE1(Yv#H-eMz=Jj%(jWgfZQ3}VHG@El=0xe(S9mq26Wn_Tm`
z?F8izlmTfj8Lir`st%4ksd>SFmya+Um`<5e<5KTl5C+AZG0N6i358)F&kCXb_@|#d
zDxu&qieNN#jq-<|Y6sp=mG<%^MH_^+*fCFPC}vktA9B8Mdlb`}ffg<0wY(`su?G5)
zC~AG=p6O~WiaHNvXL=|Wb<UN%lZK@-Zj`GT5ba>_KDzU?pFNZqIHrz&iX@z5#?7F`
zha{3M)@E9N9X1Jx$Sj%}k7_#!h?zr!6gCOE;5o6;VdW^tjMhC@PEGL<k5!VgYTHs=
z^VpPkv=UvDmwFF%#lCQj-g(elGxxV|whAq?e9V}z;34p3FWsY|(9Bxb)221W9yYBh
z?%A}F=wX}A<N0GY2HN|7WE&)d((P6eB8@)3&WaC6VME$beE4l3z>h<KCIX~AHT-7?
z!wiH%N8~hs(VncNKhN{&-M`+x{Q2|WXvZGj{_R2efJ{&xX@_s=mbPY6svEwhd-Rz-
zp#yW2M&W>(N#q2O5yyeg!wD(F0jX>;`qhs*h8Be=kVD<Xc)&b=K{1qh@D!)fG~d^9
zaO7&7A_|e&@OPd{!u7X=pRxR{zsmazN?89GPG5%8r=(S-h4CPkN5W3i-;!2=&LxHp
zriEtn5P2-+oPb#?&X7;6jF$3FpcMHvlCp$lhSD@|mo(cba~>(=&O<~iQ^-RjYt8tI
z9vWF`WLlO?IohOuBokx?a#2XyjT)YQfBu&6-U#bs@g3=$K}d#l5;>!xh0e9770P0i
zMJ^Q)6L^Ns;RsR!V#f8BXIffe4H261(@?M!Wjjy8eLAcV%8wOUZeJl&I1pq;@5#a4
zWR5qp(O(vorhL?KqtZeQUsqa!KAg!lJH$#|S6hZ|Mm?W@zyjsbJiexD{L_$b<qB_g
zT1%&NH2G<W(ARuq8gX!Zo{mOp!tn9jjdwYPvlRx$9U_jJtdDCvj>V(a<%|N~#~O3F
zQ{P~-HJeB}3s|FOHko8OqI+p2`J0&O{#w^Db7Jdv!e*acS{9R2EiH@5#pAO0=TL*C
z?dx3ZUwJ`)TcTjJT<QuS`2D?1FUGC=n%weAhQgUa!w7Dk-$%Uc7C1T(X7}4K{a)fJ
zce}YkTeIJqMmhRB4hT9;2-D7VqvHV5rb?&rhDu(rX!`N%3*)qRHZzHX<@W8$%pq9?
zDrV5mH0Ea%m$Pu9E!YQ-yTqVy^$^f_UBub*x<omDd0oQ6@z*i0%g`$G5@@Q62nQ3J
z!qrPrCCb&c51Df$yePql--Ohbw<$RupS2+cC%Gq$Bw4FESY8=cGbO6cX9~)u3d)h&
z&A=SJ^Tu|bpb!QIX6-yWE+VakAJNW}S~3Sc%Tz7J``dX9rJZ(O1Kpvx-%#8s?zeLH
zZRbgUC77F<g*UYGBt=%|bWgq16!-Mf%>O^#&O0VrZs$EHTEBK)%l5H$UW3)Wo!4@8
ztew|#wcgH?(i^R3qq?EE=h-#1e@R4Pil4~^q={~NPb-i6L-w4fy<Q$2p|xwVHhOua
zGRDiU-^0B;$tiw=`+qa<`#19hrI66KZ5q6PIP$;X$ltk-SIBd)jaOpy@8SugnLyDo
zJ!&Y<s=n^-E?z@<r;FD>cj)3Z6nDCK?fhNRf+|S#NDIihXzGKa(fO6mu<}vSdZY#3
z$YQ4Cy$&B4Qo5OI^vestFn<(ox!T^*Lit(h8CQLqjzS@mqUURk4aE)(s)piDgQ|gl
zp6W42no*Ce8L~1RY!7mq^OXj=Pc<e=%HwW4xpCd=pGV;S&LssY$@Fxtnqm*<swwU{
zS0mAfAMIR+xE-CVhW`G}bxn6<{B<|GyR{p{LVI^_cdup);dndMU7nK2*~o6%pS;{t
zO{{Ke<kn4YqKq_(dSKm9iI63JHc2sm>UU4Q(Ml;Qg#{m5wfNWo$~8$nN>nv%ruI&(
z$bx*zGnKVZD4R-kov`HD*LhbZNbRm5cd0LIwX@}%@bi5&NkB!}oNQ~7{=DJTc1d#T
z_^n1l%2S!vc0Cf_9=~X9gU~Hyu@b*h{A{2*#LtG}PW)`3fB5EnjN2SPTiyPDz|Hw-
zJVMS~NdYt5mDLxi(j~&b>)h>1HRZgL)?BHEmd|(8yO9>xY+A_IwkOqC$d{J=82Gn)
zp8gc>6dY@ME)}%L{<^7JqvK&_#J!en(5dmM7LQ5?o2p0Jp-zTg@%Qb9`WIK~iPW+t
z4>d+9F%|+Fg5RSrYRI2;tD&`jv+Jhu<MZ+@)tDe2>yG*t5wvMFn46;jE6q`zo1-vm
zj)FMguMukU*VJz1nPT?YL=F4BdEsc83UO5P_Z38rXddbO55l#Y59vrZY0;5+Xa2k<
ztt{*5<?BJ&@cKZ9=#cQ5E2gkrxJ~daUQ+olUM{$H3T19&b98cLVQmVRoPIP20y#04
z0jn+(mkoa*D3=XfHj)8wx1d}$*a3fp5Hk!Ztz61ZQ365fkCiCMWV^CDY4J$FiX!hp
zHX)o-B&JReFB%3Tf^){#=?MAX=*VUS{rt;^xRjRY^(9Ks*7cH~KGe$F&XHOPlA*|8
zz3!MGF^IRgErXO4cn)#PZDqH~W>=}iaa0O(93{AsW>A=b$5CxgLF1O_tCfFnQ89MW
zp|ll&QX$UAEz4X<ywY#QVc5}F#<8UM`r}CMH36CYF5*%ojffmGUeaN<DtT>s;5p;b
zL7l0#aMt!z=Ne-1dgf!!d&A{wZLI*SIOMJTKSyIW`xdkEF#}ofYRjxbl$l@b-zCqj
z!d{O&H+Vr??<5VS5-9Oj(t3a7xlzb!!Q_@%ksvWFRHI*_3r@r?#AR4%6+}S0z^b#M
zBc%x1_cZhxVvo30Lu|z@Mp1UsW6|7-Rkzp-Nm+r}iET;1*=)75{BI?TGHLAS4kk5D
zf%urek?<$=-txtl=k!Kd;r&!46!eoy`Ey;cX;c~SRb(zb9?QOq(LaACMcnN0Y4W(B
zRo}p06pER|daqGQEhG&wObQm{ukRJW;|Zs<zlx(tUMB67za??!3|zZu)Xi5cle7c4
zIDRZUm7ISDSLK~PW!Q6d3iFxzeC(NFJ?@vFW2V(+?2E(yIL+79>f^d3#@eI8xoGL_
z8lwZ%2r?`*rpcLRJl=n!CB{+sbbNF8l*zlWS*$3SNphOJSxh7<7k)^ySX(a~xQ;P}
zLF^Dy7{o?QVY7=Fo@vBt`b?GR#f?}^b&Ct3W4nq$Y_zNR{x=&z>;@ce4X&0vDd0|0
z6p4$T1u5Nf?Wu8NN+w33Je!F!BPrF$nM^GAW*M&A&68b*Q&N9g26f(cIny!KXvD1Y
zU!(_{O+u|Sg@5=?xvHi*BUtR|S89ko{7Mb6<yY$2+dY<5F<}Z@`U)R|@jZsN9eoUg
z*zhrY|99nM_K{M23@7D1`xu66Cm+M0?({M7?UM98-#0RdEfr+@zl*luq=JkT$x5e7
zAZ9gX5F4+D@_T<ywkmh&xfewf1StQVygp*At2;XV8ej*XUjuBod`8a4W+sVLQY!=3
zczG~F1=Z8UYKScpWBb1^FTabF$$S_o@7l}Pxc2h$HPn`uFSaJNK*X5?S#+OzsAauN
zOD*f!@B5_5KGMoblN(9vchZETz3-$6*OuNVO*mH9hgE+~khBaJ(pg>vEUn_&eya+Y
zICpKo_2I?%!3z^ytU{JOR7D}q*IC-KKsnr__K)jl*G{=onPr%g89S{%j-9~#O;0OA
z9Wc@!W}4UTz_reTDynyNIHd$rr?p#$5vVW*N)^~n>jmGgt|1sLl+qmH)Gu}Ya3|}u
zcEO}op?rVecDl7BF1KLf)B=Z%7PxJ+z*~x&o6Z|*IB#5^8v5PXk0CmZu$%warXagS
zprgO>3yaCD=tZ)TNM#4?B^l?Z>1Hp?3W;(4lKin&yR21j0fd(e@=H{{v{z55D3S@k
z;kj-F7=P?oU2%xlpX+9j8t1y%><(EagV@L_Io*GK&vh%Q1Zm9`MIfZtN<B}%GKdYn
z@cp~Et(9!MCO#jw0Z)ma@9{1rzRJ}GIK-`dJRi#C`(H?WhkU%$5gc!QwvJ%4yPV!b
zX@xOvCkHV)C{POyzClIfQ2;LSZj>~^h_KDaQqZiG&P8`^5DI&$ZUwNVx%KXEsJJ!O
z=DdI2R+Iax?H=w;T3d7P4V9L0Z;#Gy@bH|<7GH0?Q0{cjMQ+d3P9U5(^n67$@**h1
zAvJD~OvY8Ox}v9`VoxEs3-?#?_KM|;))-a7cRVO;^p0yTpo}01_6+nj#8#kZ`@<~Z
z$%>t7J-zJu#)Rk^7ZFcLI={ek!$%+$%&33(>@&?v0CJpYJ6?GKDr~fK|I<4!yMn}z
zy7Lm0FyMP~derDs!+wWS-*ClchX2kRF7p-rFSx`eP{%TH4z9y1xH-7S3%L2zcXX*`
z2L<IuOU1h`Z8-Mc`l^78Bv}(`{-!(gXA+JmJyR2OMJPrK<CXsT&aUysVZ10%X|)kt
zxk|Izdtsl?uXq0eYJK!Amyv!n69GAw5PS_3m$!HhD3{D+HkJY0mt$o%dVk|1H$g$p
z{9_cqc*EjPoQ)A+u;z@ikZYp0B#eccI;^SU`SEi7{LTteD_a3<&+i$3&aLM6(GJ9D
z(~LpQvS=N#MLs>3*<R@5N{s(cW5&<JICu;&mCvI9)5O{eqt#$2h-;^bUrH(>!H_e`
z2(qq+b@H?&$35#95MJ8V{C`(kC+y-8RZPSGvMH{S8x;vuxRRST<woUrtFGjh>=lf}
z`z(ms3#<J}3!)(*VP$V+qV>X#9d|E%b_*(&Kq<GjV33_3wJE<MFF@k9B?GKAdslf$
z&+8_5Q40;eCwNPaE$~+Jzaw}nx%Dx4OX|MhJ!QcOyp;vt5xlhp&wqh;%aSedW~>1T
zJ)gL8pHM2n>r*-M>rT!D*pky2@$4%(ouK8YB;|D{rxIZOYJPh>zt_UkO8+fOrI<r|
z6oyA+N!j?2t<?^bS}_vuX<<=!jFBpicR4V{QHHBI?m4i8tXiX|w4H3&N^wgCCq`+T
z1x;s&-{UO63Iu&pT7N_kWPD8>BKjBw@snMyPR=7aiSX_?f{L2TON^u@PZ$XKW&9s6
zety3j#<Df%AEWZco8?kvI0cVs<^VZ~KSJs-p_ss>-Qa}w)~KxnNVD14&`WSli+PE@
zRI`SPLh?fdD%c_FH@!_JCKo>)ee)$*AemR2iF!2p%rXC^H-EkQM0XtE20OG;JuOi<
zoKn4zs*Q6dD+<(((<ufaKThTKS-t&v+OX@H*w0(9>Y4PmO-mQBBRg+F&LeA%qiupE
zY(Lqp7DQ_+?VYvLnwMwp*4hVJdLGvr^5%G2!FSID81gA}E2EB~Q^>OQrimZHeixz3
z_Dl4m5H61y!GBCMnA6)Om}yI)Y0kK@CvM%Af|S}?^lQT&Y;T#C?X?Qd!K`dQhOGUd
zmo_LXz`spPUq#9NQC#or&9i=C%(m_ke&^L^)!A{@a^>^q+WHNjf<m!Juj@DL;ECgw
zFHyhY^HE4|JH{q8M{|8l5`tt}$+lOoNij)9eNRlp@qYp_#w(6{F;R+kYrPW|#|LLG
z(uz8EiZY~~0Od9hkU7HI97%SJp*z~8<k+%HHGdbMwb-{Y5K}Mr1(ptEeN%KM(YAGL
zJ007$ZQHhOe6elYww;bUw$pLPHg3+j_ka4IYK+?Zsn)1nwdP!N&Ja~cWj0ss8<%=(
zzNWc`ahpaQbToYlvr2q6$VM(`JyY3LsF7i?(^M}(JZZ6jp<o@U)_f?`^1TSYKs`p=
zzg>g?1Ulee-Z1@xH;9-W<@`i>&CaA#Pd%4D6@(4iU^0#Ad8%I-RMi)M-?l=6(F#o)
z$rpt_a91U&(Lm{Q^mo7y5GIJyq9<qmX>UD9VC(GYW}1zQ8N~g@(#lIc-1^ER(O#V)
z3U&=7Z@~HVN<#6ian6X%0(ZU{P=c14**eJrh+(S46UM>V?!Ck`6c1GPlAXtdar&d}
zXwL{lD~9+vI2yx-_I(|T$nJ5%t>Fd*mD7j&J{F2ze}>|Wl!Lg@ZK>k&pxIN$#SY1d
z_2!1>QSHaL#nC?+2peHzojeFK54qs#ZZ^uubAt$!l}_cbW*)qjH7jCA6#AFy7t7)T
zX0UYW?CIe=YUHSA*jilqi1eK=*7|EXIo{gQ-NL`@MjGe)YfrATtH-`Be+C&mg39?T
zf8}5Jc2)bRog+UV!z)j0uc`MWOx737Gu(tCb=3uI!1&I?*8h>Nn!(XqSL{6Pn1v1F
zlWz(#Iwz!AeqDf%P-!~$>>>TCZzn(lWF380RBs~kod=(;6BT~)mj0WV)mO=a7>laV
zZlF~?btV*~N4#y4qk@;uKh^XVA3zH~M=MIs-Pzy;!705OchNeNH({K9@bh{$6wNSD
zW`Cp2ls<VCzoPzUlk~9uE+F@y8q=5ZZUi&%N7{V1cD#lx9Nj;(u*~RfX+9Sjpl!m(
zb2z%}6pavzft&gPtL>Vtp~2mDDm>uuY2h&6xh(6^b~+TO?)0Hc#~qTIdg$Awz`i<h
zrO+DRJu?`Jqw39eu;o*1el14^Mz(<mR;o}Q_sj0-<Wv<vPt2u6L242A>FpXr61RU=
zdg^0$pJF3K7Jzu>9SJk+uIjr1sER5q%)(Uf_iV09I$tr*C30W7-l!H){5u@S<U4))
zq^~zpIbU5}!IN@D6BoPTSL?B>kLR5Ar=qh^1JfE(u&!8-Iu5>j>E6RhAE&oV`Bjw>
zn1ZX~U!qbGsn9R$I3FwRv(}~i{^8oKrO=wAvSE5tvF%doYEvkMODWhYz+M5628M5V
zmtqHF)rvF6ucE(0r+s+wSBK^oKf1@q5c|P~&@JaD4={VGR{l{tp)JfdiyDr?ww4{9
zpthzR>_1m)H&9LMt2PvK_&aekDZgLgDtxXIbtrhxvQ&y0x$Wwlm*7LS;5blvdY%1$
zF0y>5dqaF7*I<qXD}7x6q$GL$!8}O6iAJ|Mjr(EZJ*ffwWMOSg0bxPc=x8W4`*U^Y
zE58F$e}U2>{*Ec2fjF1AJDx2wph_^UcY)q1$FqbNWe;p#82GyKQj#Ky*V%x#xN2C~
zSA`v2&A3y5Q=p&lEUVK*ok+38MBV(cR?=5AJx$IW;jU_horj;cUR?(uK}VB;!`~`)
z{zTgM!W8rM-A-mf=Sm>Nn`qs6^q@YzlqYuNclr^?&O8JC?>Qb1h8#!v?OM+>HcS?L
zfCJrbb`2W=N!;7nn0apH+vnZW4Pw(PlH~vU?l%TAqJnZTvLq`JkOM|ECgL_3;d)PK
z9zchQ#FSwb!7@scL3&%qYfmdr|5zd9ehZ{DnU$yI!b>p;o{E)?m!D8%2|b_9`Dxib
zRgd9CM<=8ddAwsvJON!wS_TX0kp>MMJ~>>CQHT=!VK)yNx5c>RUe{Vhl%3!m=N->O
zsU;{%h|SQsRxq#Ze*$zr1}8wJVP|ub)B~>i>l%VY#tU`RRD7O$3_i7571gjw&7RNg
z)bOED9q;0NfgT5Yzug(C9qiVaus3qP3Ql(n^jPkCn^Nc3hLBV#Y>|J3?gx<@!s@Ok
z%|=+$q~)%!)YRmVSNm>}X~$4n_SoPk#fmgWyqQ`!sB?i*aRS0cAOXzAH)G>OK;XOL
zPn?5n-V%XGm5=n-DRlk2`uPH4NO?|TEja?&U&>iWvwVog)+&nyzc7~db&{&>5Pe^_
z`XjIUX8z@+d$rIrkH|q9%GMwUHNh(HjWXTdFUf$(ItPME%03%fNb<yb#^eenFwNW<
zOvtZ~<W~I|^9O*UukJn3?yb){P7&MBPPCoDacHYSojRDz*@RWR3PboI31;0+7h2R`
ztl2+L0xS~XfnBox(myUWAu;S*ObDK)XVNyQSJ!paw&|eNYbZ`Wqg}J^bKSDP3Mlr#
z+w;-wR6hHS_#g)td*VR+=hl^Hre}HvFU5|K4QW9*fPo(0LlnYCK#r*YYDx>7Csx$g
zIq?9~mAksA=W1#~kjWrhfW|bl77$sv;bQrOWIGzLG+rK9dvCYUi9hkgDq8Pxk(d?1
zfzH@QPtuAUkKRyK<1VJ)#xv5C`bMV~m9b=}uI`=iLN^jb?709t@2x;c)V@<%eBDjh
zTzF_LOo!hpnbNq3r`oT)^_07$2_0Pw|K#y>7k*o8SJUXl=^C9M^YLB<!Ic_?jR5Z)
zx+_JU`zFE^!*20zbBhVaT_pRz6b%<kx{@3OYWt2O`#J>d3B<$g(RTmX-BtYmEwiQH
znt`E$F#d%8EJ0ADqk6Iz0LCYG!vX%E@36h_FVm=<I?5Bx?aP+MYPj3en*}i^yxP&_
znwkv)3<_TNhtt#5XmIv)^S~A3f~a2`;Gc`n=`^Gz*=kftmH#43XkkS`Q8b`>ZM7ri
z1IH)FyW1Uh`=-(uOHpNNqz1L^{^=~DNE8o*%1{T@{X&{OJqXZ|1AzRl9`_IZSZFLv
zMV*OJ(GN$z+x?@}^DC_un81DV#Z`BrGK0GQ>%qrG6G}>K$AtVwVU7ITEs(c+M4Oj)
znM8mfd|qE`>QS(-9WuzxgcB%;>cw-9B1fC3;ab@4NX^HP7J-N2bPIYkVPRtVU;*cl
z7o9U`OqQFT{<Bi|0&t$R9FIu0+$6mgXx$VQG%b0e5dtfdWhpKgU^ljm>i<@YYOYN-
z<QSQXSef^i>mC=>_|EW1jbe_G4I=&~;DoVdQdtopofqc@J6v5WYSOUL&~Z}vmzEhb
zV^Uyh!VEV{2r`kH>izdxo&ZT4vA8&LCPxl^%O{5Nns^w-FMv!$rZ-`3sJUiiY-=|U
z>~w&di{o>6dOcUje5oSjLv!g(2SezNte&j&J1dJyxTPCc`=xK!iYyIH&&&f8$!8@~
zI3JkPMbhAtiTYvvTE(V_YYZOk0KR(Y6A690rr=uPIZ@lEMJSZtp8dT6&s?pXVXpPz
zW_o=iK@C9o3vgeH3aRa-6Te4`AUc?FV&$}`K4TIX(qTjMV=u1dI)%y~L6k9%aW=ls
zC6AVdVW;YUaF~4=hZov%hX35ngg-_jjB<TfyNR@!tZPYOL`oZcS&w_27s7yoDFFsF
zG^TTuOtm;%iXQxlHB`}bC&A;ObJ*b+(BdI09)dFR3vlP-g3J_xCN-CmU_fNE8Rjd@
z55sn5FkP{~hnzdRZ+x4yYf}8s<(mw1J+_^n{fz10!h?z}JNt%x{Y#QUhW9`(!tcl=
z=eEviO*;<G{UHI7nrYT^6%1C$G;U|P%MMB#qbA4ac}Q_#?f^26i^Wi35xt};A_nHJ
zwDVKv5P&fSr4Cj*<@K({azV?^f@U(-u&96{di9y+Px`lIF@$aWa4nDmWI~$BbX%5C
zt3q*>6;*PDYebYK(3Gj*bW4_SU9f^;znRr&GpuNwl_!~;mK|^5BfTVY(GEJBv~hje
zTOgFMkQyosX8o2o5-N<>ZG4O<`fuAwq*5ul0YJyfClBr}u<X~u6E9I;ocKl@g+*#}
zVVElml+^e+XjQHd6mlBpa)LWVBFvu=P+)_s$#`Q`A}G+Nxjc&B(f-KNZXSkjpBvy!
z`biN|h(w2``zr;yAPd{UC`^^xW4M#f@PBPRHAie=2j=Wg0@DFw?`q8X&fM>V<*>#C
zW`G3xMSnJ1zVi3vk+gOGO-l@!GIJxL9SJH8kxoU#N&8ONo3qz$`VY^(nnsKQ+iF*#
zPMh|<{5EWhV=CL?*2>-wyjt&`hB+z%XrFFGW=b^G!KJEi;-Q2q<^2QRHi)6-b!B1w
z-e<bSV5dx<_SxEp)yd%9jtCq?F6sO`JiwHi&hL{b9j(dP<xq}22I~4xUC$4uF+K6S
z<Oj7?{Yp(u_P^1M(nds|D*m*+8O78R;oJF!i~Qe@uioF}vsUrpW@;-nA<W8{ObPQi
zI4RvVCw$uT&j@U&14x(VO%hWTuPD|?G|qAaZJ+m|qflt48!(L)G9FQae8`E0G=TAR
zcb^sMo=9{mlXmj4?~M)Qc5X07DNzMSl2{3nt-#^a^X<DL#1tZ;D&D%hve-HFN_-88
zB8op*2rB<5X)YPpf>YgR7m7B_BbYL!x!oT&z~6}0<Y8UkYXmJQ*C`k8wC6vM#zc3O
zeP#L871mVtQ-^=_&n8LOu>w6xmH?g_);o!ED4~mG)bsQvWfe<Nz0kAHm^A&{f41E8
z)lvd(ndi@|O`tU@Xu~m}v|prk_hMtfrHM?CDG~mO33!S{5O|L4*>RvSG#Usu2MCOs
zJu81W1Be7K)BL70K&44OQ^*h~kqJiE7>I%|^5sg0{f&;IJ{{MUTQ({#!2vr@$8Ph{
z6L>8ZmVD2eR^nS`@!TeZdRJfP?V;*3Z=I^Y3RSC8kJld_?9s){@vRESFu{+X9pv+9
ze?gNyyuVXko-}4gj?#CRKU1L7#jktuH{0kHRtsFWO?B<_-&fB#gcA9k`6eT|c>`$X
zFA?jduY=~h-*#<EyBFNRR{`;%468iUrP|c9W?%}NDgyW&+7d1|k_swkC2;D^JsylZ
z=~>D<gJ7F0DH|Z@4x`>9M6o^37I>-4AWl&(KDD1a`H|@5iaK7aOd{hr6&2JhTmL$m
zog<I`5b~9}X2?+-2&)cPq962n+@&!x%Z_@P9VEC<Lq#K7vXq2k;QbIsPJRwLi7Ux0
zGB(^0b+x*!B}!^p!xuhuRw0(sDk0#_Xk(}$W};G>65U|BuB>8MRq;?v8pxx$+ZcHz
ztRfl9AzF)YM#?H7WbiR4Z6wk|8;K5$aZ1^<NaoO(R4XBw?)BPlMOkG~EO%8aIjHVT
zj5WqwH!AS9HAUPd8vqqQlza`<@$8Da)dPy14@t2RA@I0mRBsD3xP^^70~iywdzb`|
zbQ3NK1ngID$y_qi&pVI}M{WI|@JHO@27(oJZ!SnAys{1$Wy*4@Hs-?_Ob;9aV7CuA
zgTypYOmu_Xl;ErZMupqdpRf|gpnAT*nEO_dD_?l?9H+jg7=X&%@)W0fBF}G)a$;tT
zO}W^Y+GNKYtaZ*QCwTH4ZnGTnoFzI54|*XarL6rVMxD0sn7u6BH1(*j$z`v!)EN}k
zaXH}5{o?~p4rf<d0@_~(KX~_;*?M6atFkNq#!^awkTuLWI$J*dxg`RY-TusZu<nXK
zwf5CC27p%k1Mru@{x%=e1<*G_V4?k?NbE>D#eeq4maDDxCB0dsGYH%IRwJSj0|1U~
zbK5LrYSzfQvd<8SWqaG4NMdbNHp1{5Z*%|BUyH5o|2AnPG-C9NDmtog-5}YP!+zd2
zjm%13DSc;R0A52HAt;iy<R9f*UT8=}@Qz`a09*U@iY9zBi|eks>d9Y!%<8z=*Ij^&
z5TodSthWv|TRB+hz9z>LK&gK_M)wQHa_zaY+d-zN*WntD;UOimP0Vl#`okoZ&S*yW
zGczyk*CUzGeZeq~f32%>Utp@&t$Cy`gy<^%>uXPoJ=0U!S|0a)AtbnTyxMtl_!rj&
z59l`V!&L#svTT?tY)CN&M3V`4z;uX?Q)g{yDy$gEU_zPF!hd4Oyp0Ch+Ldl5i?z7e
zm98c&Pca2i(Oe+Qhfid<_XRe!ZK#GI<`{-r5=yXDEvEm@v*Mz)c$P$TS7nD9cgY#;
zqV9R<s}6kZq89pzwiQ7r<5aH1wm^r>0+3CEn*Bku;-{u~wFzp0qmnvrRljJn9hHK0
z7VJH)(HqJ<O*hF)&LP4nme2hwJHt2C9Roe%E;*r9-djWjj0q9Om0UuAyISJ5LF3uQ
z2082cz+#P>yw7E#@rkZKGp`1Uvp5^h>(_<znjBX5!N{<U;_MeCurzCyBB{$?0N9)n
z;~vq4(h()bml2>L<uZ%GndBEQYW=nReAP`Yt;ygIc;2u%mF9~KPi5!+l(K!DUx3rl
z_qVkz>)0S9)95WB36EZc(^8Ca#;qo;UY!<$3CPGMl<{vAw+gD^^N2}KB_h)@SvSwA
z$+4Agdg~+~;cnu0?Q(|JiLFN3wuzy4h%ecv7s`ENcyjfE*OWJ-<yn3`V&w|;4|X}=
zXNFtHzABFRD4e+#_p>V6ec4`P5He{&i=ffoXQaC;rphuJi57np(c6tp8xrk5>26du
z9F^<@t>f(Mrx^d1i2QiIU&Z`}j5|Ga|L<Qo1{VhxAtRxKu{8`YFN3U^y@jhKAq#uD
zkUJtJhOSOLSu3)CzQKGvPt>SQw{PgSk~w5LM=U!&Kr;OJ!KqXCx&DXVezQ+n+^puf
zuGi;J$n1eq_u~8M2$f%i_IIWCOPiB5q&bxr-^cz%Yy7BGXK{REQbb=oKRgB&cXkqB
zunHN4wS1Q@V>zO%Xyg0U=obM|vv`Yp#kWS@p&s}yV5pH6z=+`ZXJ1d%&-r>2Eqe;P
z2Cc9F*DO(}#qhD}eDW9{p6THFk^-!YhpFH7LcxtQUc3{O=;irLhev?E_osS-K6MbV
zZ!;J!?84e8BX8Y4ZP=y?t2ZJ)L}40-;S$l5!#{^5irPkk<hlmF8FevF!in%Bn*=X*
zXyjP6u*S+kL1P1{LbufQffl;E=KcaH4EI$4c8q4(bLJPE*(A$Xno|~Zz>&f$A0qg2
ze|fb-d9_D1KgnML-rS$jL3y&6JI5U1(pPjZWItB=Y7%d&%b88(xyMordQQ-he}udr
zA!VtX)VBUAP&o!hb-W_}sri@hc;Ku}^we5g3otlBCm|jeDb!YO^A}IM1o)`y9)#Wi
z9o}>+`#sDdO^$8Bevp*`RhO0_4)6Ia@$x|`&3{*Ew0RJLrpUG!LgrVC_x1=if7&q(
zA9FMZx+^tziBG1fPrDbpgv1+HauntDRVAoQLks0-c(`a~40(x3>$qs?tsAqJoqdXf
zj~&KMA+*N8{7pQ!Ku<b8raMX`{fSKg6ObP$N8n@@^<IDYXuOo#+`KT3t@gREg32SO
zE@&w7z{^+=k3Jw=*<+az&$pad9%U`0H=NOB1^NpgOv7)^t6aT7xqS@kwyHi;EqZD1
z)HgKE*!Aoh&6DYle+8DCN7LHqt9RB-c<6RCy5Gf`US@JzfyJv?l<?jM`05q|j$bqK
zso1hBK8nIcjoAI@&>hk;R@-}}x}<kuBHMQ@rhJoXE*C1^O-JkwL)A|vyEIU7&@X;p
zU<Am_eQ_=iE1UJMOWVlv0i$|K@D&&qqaI%VGW`<#315z4t!}6*V3G+aGe{<~I67Ll
z7@0o90>(Z+d22cz9ik+E1HJn9!-?Rk^q$oL6>n9OmQJxoH&c5j=hDpALu{WK^L3}v
z>_)rQMM|O@IT_UXl}`8Dw>Zi8_}tgipIB&_Ylxu31^xva4iZf3*!S4+|H5#Gs<MED
zNfVj~#dCJ-o|b519JuykW0uNPUd^Grj?OxE_%TSL5i^k0JN`}PQNz^=h$-cxXbNGr
zaO?J53OG@oJu91mE;wR~!B6VmrXxy)zXv2JiO~@+2sa0vq%aBl0C~bgN(=QK&IwRk
zFn?1ix!nlTRq07e*I<$}35|EO3O>tlkysyuS4gjOF>vt#cW_S+)F_5E4uPhmH1~J(
z%agKlasDzcBW`R!QiuNo2xaY2^ac23e9S3=5$GU0i{VEqE|->;@hxMjr@GwUVs>rS
zqm<&UN)dK4m*tXKY?H>(%e+0YN9(2^!q+VIr*wK}kFZ{ix4?l-DDqTy9k%vSv3%<F
zJ_Q<~1T>0at#O3=o9m+KbNIV^oLWF=$xl&mWkHiJg?h=Do_vD=nx+CHXlur6^)E0@
zvO_A<gsTtPS1;5(z~goCEuw1SFuBP79<mOl#5FkIMX#?E>3lX}EFQ^i<!pwXrJGt8
z8_RW8c$G&8qsmvRqiW6C7{!d7UExS=6X_B|B6OJpjT}!9VDzLdx%9uhhThDoF~%^n
z<k|f$)hsU`8A+1^R4RYqmU7ci=J#Khv>*d1Y8NQ{Jdnt8Z|x38BfAioS!BXWyv{#l
zQ!Cl?If=~P-s0JolKY>bq_g1%%n{vc^?7&)2_okv^fii_Ol)4@V41_u{GkmxN<$R{
zAPtx^@BXDaJ_SklX6Y*I*UZb2W=nd7wH)W0O<vQJ%-MAKG1V_u;-ZP9?rqskG5mzo
z>;#<C`O;L^QHiN3chl*|N$f%xzKU(pf_xSlv9z$c$hGqW>(OQ(hfnQ%$?T%QfWE%@
z9mF;yEFdboydqY9<@-AI!RK&k7Ew>F*P-ausk0>}nrw}BWPew`;v5}(et^%%=^Gr3
zU?Z@gj*?(M`zoJb)gt!y!}2T+6GqL{Yq$@GZmJ;@<{ti8Od|S%?|?0OIzy>GP1$<&
z87m8Q1;2p44iNQ|FG%!5ZrT$kprTQT^H2o@YHl)TBPvS##Zo&P4eF&qc054!pLJzi
zFVZ;(r=6sHXW@#{%N*s`K^{Sw0|7()m<YJ#Sol%x$6et``1yRkfdYa%V)BE?&n##5
ziltqu`UmG#sK+u$D1@tmZg7j7|6RKYj%6R;LV_>!IPaW56<hgS(%GshAYeBRn>Zcf
zwAZWxqr1P4KHE8#T*zgbI86Tx=&96+5@P4nBFN85Ue!$X*rj0>h6HCFqG6VXh@_qu
z2zJw@0g<qGc@OGIt56OIdZ6p^#E?K(Hnfk{ry<^>OjnLM$WQ-9FCoel?A_J*E^hS6
z(STejsQzz`u)h+evI25GKrxIglp<1rFcCf-_>M(-0MuQ!Ggz!AqAr$`DxCATbJS5V
zn^jg8YL~B~kVOxq>tD4X`fuK2W2G`I$N}N9?eQHYAPjR)BeIF}1MfjMRMchamrGAG
z{J-&SJqPBS2|(>ItvYz=J+CorRTO19!n8>?w44J!qJ?xzVt*hKKq*>Oy;rV>*Ra^`
zzw(YA;Gy)5Cq4g@7EtCnTrxZ@cZ@KQqm*maOcO#pqg?SE7eNe8@e0As%9|!4_u~v?
zWvzeJ*!my-l)1uXjS+ECrl(?(7N@%vW!&^DFoV}cy@n-mZpKjhPKC+n)RmW{UVL)@
zEYOY?WfMc1si4030RfXCJtHGfxzbB!J6%{q62|j6Qrs{_iZN^xg#RkWk$RMNoR+L3
z3nv*yIKuQ7*24rNJjv_WP{>B`y02?MGbCma%tK&rWf@BTjtY&gs2u{CAqj6(XPup!
z(27VB9-g=2Irv$+U>l=1eJ^Hg>-97)lQrJy&+9ocU$oz+07Bh+2ae+s`2*--hTzJm
z@<84wCr=@1qL^kN`H9d5qs~ZYBp|QWLVtz%<;+rI!Pl_otnKs97>p$rSZr)ht0V!V
zHJ2NH`M%-f)Z#*_&Jj!Nb2>cE|2P)m7WfXE0o?-+oNTUw^%!1I<<J1%x2z|YB2ti5
z9omLWHaWvp0I1kw{0Zz!vsR|Y3?qSchELYiu<P&KA4{w8iD~cd=buq!r5$y4t?lPj
zHAg(UX`+M_fPF7Wmq~;3K>Td>G%w%ecA<Hdt9TrP`eu&7;2>K`UG;K2vSCDLE($P!
z^WGDLR0hLwXJ)%anb%b0z%fQeXp)pv0^eq`o%Iwp0YXhlqI>&(|5%QY4vwiMzP6s~
zlY~d{g0E{tGGBA*nO`EJPx^Z*I&*%HB2neSu|IqBSXB0P<|<x#UtSH}Ap-`FM`l-x
zljpJ3rTMx|av$CdXYQpEsxojfb;UCWm(fgOW+aLasf&^yG0>8bt*7nj+C4*%W%pgc
z<a1Md0nGao&!Xz4P3`?-JlCuSwk58JkiOAH{n3QEV5g64N3+8sDm$GL^)?UKEY=PI
zd3w_(Ksg||{OqJAhhVN}I9fB<NgeRv=-U2O^M9`bOXI4@`61v#)U~(_Y@Ae$iyfDw
zIVu_j-N#wcCkhgnuu|aD4|W1*+fiFcWcm}d00r3RI&SeW*c@fRT0%1#A+t+N9aT!m
z>oAEPRZ&Lv;M0_S2iQxM)L}iN05J{am?$g<Nwx(<_EBO}b<OleOG;x86F8CarbcXy
zxys;ODD@ACz$0M@F}S%@i^`bK&&sHh7Lc~5<;H{Teeq-WhmQ0Q_l5Q&6@#i{f{=V?
z0Od$UIn0k_Q{`{DZidcf)RfbU)9XTCnI~VJXvH)7grN@g9lOu4neLsT7sLqqa>WW2
z<$^9F_fFaZ%8sS5&Bs1{+3wbW{d4Et^mF(z|BL&wW;=jgeMkD@BQukV9sHf6pLJI$
z0PP8J>wRB-WvEz}*XJ_Yh)`Ws33vAmU<_o_QcW4h;Aa*50f~5Xb~m#s4=u(hMY=8w
zU923N^7~&$36_TSm=2XHa&CyOr5tHdl~K8TU@p`limXVtdM=+OQpkeSei~)@@(jHc
z{_$}0cHoo>@qJDY$XGf+aznc|IJlQn;p$)=*3oyX+V7`+?npS3Rgv|EIo)4?Fry)>
z<x~?14PAe?%TUmgVKb5yUU9tcoca&fqtgRij@_M&+xA{@ZD?o-SfjpjCx|0Yep<*-
zaJZkdRz5=^R}rEV@x<GW%i=r9c&*AsMNeWA>geXXW56w^5m$(kLNLhn&Np^g&ZD*`
z+_7k`t?h1z#Gshr>-P(w_Zc3T_rLuQC?`8ddSV?6L%OIKCIw(yTgd^p6W)Kee$M2*
z2w5sJB_KGkTgH!W4n!y?ye*8txl_!^a*vM_hw$_HvtnJ#r#a8Fr~Nz}Hi6b645dGD
zJidgy)*70sMO%A3#=tJnv05#8K>Vu&qV5YyBXbEHq0>~Oj)<dQg{q=Me%VmjK!^g;
zrp_cIW*QtmWeH%x3M*QRQyFOL3)78F$QTMny-WtDc-3q!dB$i1Z5nA*KrXpX$11yu
zV%YLK5cLrmDAziC1%#zDG7UuLR_hn|Oamx1NeZs?{qyv4X$)o^7|sA?WS2gyaOb9E
z0uU4mxRE=KCaBS4c}RI?wG5U{$vHGBR(MJ<*0YbNb~K>Z8bP<<@)_=YYJ{wYCY>KP
zP5Vt{VB8Bgx^{B=mpZ*od)UbAO07}2^HJQ0uLp{)B0O2^S4-$n-(IW=<-W~EhVaBE
z6-M{zi>ffwZVg282quU!bXFIey-=k<Dp)1k8d3+)3BNGW>lr)HLXKa}o_-=dH0foA
zjpF5X8WSLTxer2!hH)PwWM7i8bj2R{tFjIT+J*Ib;>U@QhtoZy@i1?*K~B6fo=tMS
z95r2Bhi<^m_b0~V00BL$qz^|d$m$N+(Q$E~1jc;j_*#Xr`LP{{b(kRCx&fKFXVuyO
zowmA@MVJ^7w7swaI6VwLRdLL132x#e7N`g15g7ne`G5>DF<fobNi9P#PpC!^Xr$KB
zkDM3vrWt|?l%lcg$I;Ul))+G9&+~C_@GX9*^`J;G!*l>|F97)Pba4WnX8h>*>gel!
zyZj7SLn1&5<GAL?Uf#CbyS-yDU>DG{+i_429K}gL--Nf-HNK@J!q}Q5!vJxZ8-=wc
zmjyijqo)t!@Hj`*sGfejIet94v41pU$-@6hPHIPw`LqCsYtLT3JWggAjCi{|TYS6U
z4qq-$wD?#NQ1AbE`wD+dw^>!Z13$uv=^q}St8sN5er|6+cPgrV)mY8_O}Kjz<6ipr
z?0y!f|LwiL{TqH5@x0qW-NWzBmF2_jWfO37z!r<?ph3XCT7n|;Sde51WxawnAwn*7
zsBRy24O-U!%U`=+2eE!QUlNF_vVsT87$Qy-(AR-w(0i@m(;yJ_8R%mq@snbxXtOnv
zq+$xTfbD)eIO^A08X-qriQinBirCgtX3~Zh)3>R79x(>jjP8nJi*iQVf_fofOaU0Q
zre}kJit@#5QfUzLTpRoX#ez;@iBzX2lwG&`z9lr|CBl&-OxQFA9{dw1s9Xzk{YO$V
zQbkzbib8OH8=Xo=nv|1R*<>ki$c5D(OiY;0NK3gfYkWZ6J*l+;N;*%F(THk~$%RmG
zCN)ul1*Dde8mqRSKtx=U+Ibg@7!wc&cAN(qSVN*KEv78NZ(5OOh_{y!R3mJP{D^2>
z`l}JMrcq+i8BCrd)!j8X=o*>dyMXaFF^Dg3Wiqz7ys4axtM!!X3A}j;zaKt*eLKR<
z=WpYTJKTM%_kgclydi%*x%(dt3iWC`nif`QR%l*mW@uD{HfI?6!w7B~%t$~|kJ((r
z{_pDfh^`m09{m<kiWlf$)i~(XE!?pCIVPDJD1a4Eu<*Q#u&`6UN+jii0X2oDXdsV~
zp`t?A_a7s{!l*o&%EDHmS{=p09CM-^AML0e0^{t#k%GaLn@#-6SS&eJ2zU`4MPiK!
ziXUHe>Zmh1s1{S?=?UJ598|ysW=-%z|IG~Qiy63=7bts>V4G0M94!>PLxmr@Brg0q
zv^z;Rjb}8Iv0}-25QLRGQIeJXi$)dJFUOV$F9s@B9p%a^D?2U&<v#_R<!tk>qs<Rw
zK0SS7FD|70>125mFR`aKqt9WVR-+Cn&)E5sakrmPg)Z&+gtno~(A)rMcIfg`w>EJ7
zI<~h)xZ*LP2Ey=F$#};}FJWjgn}_>!XfY<m-VIuqL#(=rM;Ylg|I%btPt@i{#~RJe
z$PG>sa11Xa>ug2YE%s=TnmB8qkv8@xks@7%0O4OX>g$C!Ni!8QMT#NmXlx}oFYN^m
zd_wnu{N!cY?u1CvFQtIjQpy-2bg&5FURHT6{G{ysb-(6)zlw%x#d&+PMgbs$TcOYm
z&G^4Go%M>JNY01ES(1@wq_#Hd0s&hv|EVYH_N}OguKZ#@PLyf`s5Tp-X^ZX1ti!J)
zGp|ZPc2$ecOT_`E;{1-;qVe&K2tyztnUJ@Fb!bJ?-||;uLjV$(8}o<ZSuQk-N@aX<
zY&1N}z4=)US7dl0^Vw=;e-0?|^)RuWz`d0ssNOk$w578p0p1u&Y49jr3#A;YMIynE
zDd=BN(wjiHWJ06OPes5%1Nn-awU}rz&6FtvKY|l3B9aEG{G#&7PuU}UF{6})JO>NK
z9J0kr-)4(0NI+jEG~va1W?Xv8?Ys6Ss70Vf1>Kx!^}MJ7JJS{3S<dNcnG}~~4Dl4|
z!6&s;UOmH1ALp$9x%<nC`;yKoSzlW@n@7e=hB{IF)D!&lkCm%FMftB(OiGvqq-?tR
z-wsJ#g2^-^+G&&+v1x9B<k(rndLXC>$e#yKR$F!q<A5h$L~=`=J^1vfD<CN=pTe&|
z<BJG}>G-nAHsi}B?J_1@OC3bvX@PAAzpI7s>VY=XkgUzWR_wOH;}0eA^Cq}`xW4ue
zh*c7tzV-?&$cSiVc*;kVQt6h?(V~ypS!;IR%z4fZS@|muDn+twqKkVs>S>m55=w<+
zS=h^BdH`cQI)E-~sc9?~U6wL$L>g@plD57zENh~V(lA})%$n?!Oj@##%g!c$tyBft
zC(AR4T;iH&x@A#w+aFj~rN&|r%>qrOi4@;BIL=w&chyD--shfM^xF%S1zOlkRNbd<
ze^rFFtIibDaIUqBjI<tms<Ur#9g+VNkEEyjv%&-aC|ju^P}6g=*=K=o+HG^#|H6Vf
zp#agOD^;*#p#T3#+H8!>=|u7T6zRnk>{UN`zTxrxov2nruRfWj3(2g1&GF<iWU`K$
z8PS)W+trhYna{hyjp!f_lA~4g`?cL*RT?IhN03R#h0t$5cX!BO{F+r9H?p&aWGC-E
z?VH4BNxzdM-Pi1wOJa9rZ^2bs{upCY<YdAlOfLfDpCcYVZG3}k&5@&li>XS$r56D@
z1!3c<kSwi`D$~qAFX0aW&#7I50WmfBhYd}f?V=DzZZP<pO)SIL5C%BvcOC&t<Rt>}
zBPpC(a7DW8epj)en5ZKmu!LICE5<L5yzHPVq~o%ok_k4K^RsYue;BL+)7?)8)JPXS
z6gzT?Ur7!`Be#OwvZL!+XMnJ*3TQwU6n;Wh3VbW^6`U|qT;<ryayO*aCGm1D8}=Xj
zZc;A%=`*Qt^A3j5w4_*UZPMV$jY)(^Hg;`r9yGb=&jC*TSC~!E&~@^n&_B-sLHe&O
z?x6gF=3T!$=41F}OdFCP`p94Jwu3RA{QqErUr%-EppaAf4jR<GHQhU0eJuhKTwt{K
zVr{r6?z(cXD-(i*DS?p(!AfO`vn;-s$GZEs?5=|&z5U*=7Pqgz3GKfF2>>pr2~GW<
zY&&oN%9vo1KV9Fqx5K%x@M7Qz4rH6e*^YAxr4Ig~pO1<Mt+7H+u|NVs@}>CWBpNZU
z3bBudJ<&Yf$ui%eU<UJVX<-1j!-8qstq*GQUM$!qHc7Li#bCjbyOd7^WA<WDo;Py`
zYsVNy6sSW~5->B87Im~5GQm3AsYli$^o>Sm{qgD1hK7_6Vyo?JPdAB0%&lZOvvFhP
zH0y4!3v}p(x)WIA=?2!%BMJ#q4pYtMq^#J8IU`RNHSI5Lz)mOq0Y82X{l3jb<RHZ{
z8=TtWvQ`ktDW;Ud^CQe*B-vL9=e)l}#pi6vIO6FWqb`ln@9c?{6wbcPBJhA{DN28t
z(FF!vsZ>rdBc1+pFgV`(EgIX|XF$9lNikSSrHD#7inS>DJI!3lBm(-1*s-nIq(few
zNfa@@)H`)F$D|G#NhcT}e}rpPyl*CGXR&-#0ehWv#hXy$>W=~idW2PN`p&rt%0~$T
zJp}rXg;|&rLi3#dcU(CnjIeHH+4JXGBMckz9zg7_|6075k}?Qpnv!d8(8^6JK*l=A
zVU!4D=g)?@wn2<W(fah83=PqMgPl#1Ln30_i8<?#23@`n6aF~h1&fp>iC|O^s$vmn
zl2yB*>U>`;xj#&@sr<e=Yqme)?hB0nq&jus890u7oaarGOR77%_Rh{xG+E_Ku3`C2
z79O!^#*k))oE2pxY8etbA~E>}jvffonVNT0eS`CNiFpcB+X4oYUUnbJ+7$1H`Xa(W
zm`K?<&M+4c4*xO00DT0w1t#Lp#uL`cjxh+4=zQfeG3xEzg0<8fX=GhZS^_cFwr4$4
z3o-ON+EWxP1?dwU@Ebp`PKhp)B%%}*!Az)m(45Hm^BA`RJ~q+`YAMuMV<u8sOp&`B
zI&+E%6p=4(OV$Z;NI)}E9mnh)hRIvjiUv;cVTmjY;$aV9+zdmaFIb^PEU7)QI>nE2
z|KyiW_SY(SbNW~~<?WGlV7y8oZBXiJ39S2<=&T*AZub_p8;sMysp!#a2a+4zFrh9h
zgA8&V7771zbxhoSMmCeNUeh?D{>nL>g!0JBfP*uA3QZdYA!ZZB;FGo@z7TzYWCRMS
z3-~$Qp)fwc$d9LYM7LHdRqk-<PgO#Y?NAEWx>P^8=|ul;$4gt1r3c>j761KYU%IW#
z)G3u@6U`7QvbO2j7?!ElqI6C6nSS2X$gMe9E~C~k>$LAn!go^-f;ZaTNL9zgoh53e
z6-*B#AzIU*NsyTFhIsKiKM28v?sg%PDVE2>8s1QVct^vOW}ogM@UQs+1xfcp2qxS-
zAreyxqiiLFIj9w+mTOpBg8K?}LrBZt#;<jmYrwypV64jOq!9L?nwZIi<4v-4;DSOp
zAtiX_emi$GY!On>(bHkJaZ9bPZ2;o}pAxx@EJ5w8YPB=5am77HRLgg+P&JV#uOt5>
zH-8BPOijeqIEO?Jw}c)PUC1A(xJE{qw_9PHY{oP96bsro4c8!W%}nGA(*UI<r>XK?
zTB^hj<pdKqx?axwkv|N?LZ;(z(SQDQR~ppEs$yTBJncJbjD64CSL!79#Ojx3iB<SV
zO@^HjBgcZFbt5D>C|l=3(+E5(X1;?10Ih)oAi!N-+yv<eDo?rRJCERE-OrhKhum3D
zy^eOi-K~<v^pyx+l=@FbP@G>#6Rot4TJ+Yu^(~o>`YeaEXm&RY2HAout2Gd*iKX5S
z#2FL&@_sz(B5&({9^1rRngMWNwqV^^+T=f);N4~3O1eLOuqq;nL6Xe2nN{bro}N&E
zl2&RyGj8o<Z8tcQX@ahVC3(ML(|Ak7L<#NMm^HGKvL`EAYPTb{RA*;iMeq5=cxgEX
zk27^zG^W`dHaljGr#DMrA{Z+&PGIWRwA&wBe^>}toO{k;Gv}Vlb%YGRu|L{H?K*^*
z&N1&NOXDLcoqhF|^W!QDTDVTw{K-9lG&`8iMJXoths(OVi;!$m7A5=IlHK;iOgo6)
zUk-&xSh!G5!;9@I5o0;5pCB+oD+<3fTQfi8uVm6ZP9KbyX07<46Aw#obBt*MHbnhW
z7e}cfsNbKA6MLF}^U*x!eB&Gc3%_F7pM&xccEIY@6gJ&ec6FIT&t|7rgD9>AfJ?M`
z(p#yICxS^<qu&=D#G$Y2&s$(Ox12KXDv-Q%QpI+cQ)lQMerUV3sc=|Bz9396Y*`xr
z-N)9)P)K+oVrdsv;if*}ey=TuNqckiGp3VO3s@=Nsa*%6&R@PY{Q9htUN^<eKg9@s
zKRYnL-3Zs56ny(m$3Z(i*I6n8bZY`Lw)gMsA4yT~aAPK?F5I%a%)P|d9$(#mfBbz1
zd|t&ngRK7y`t6H1kV3rX9}5KTmfkIF`%44v@pIVD$jgE3j=let^d{I7>C*dpbd!W0
zN;mtaRSVoM2W%zp&q$E3jMMcV;M@a^l=4FQ?>n=_KdX2#&06CT*I8u(AihTwq*^{G
zemL3hiTt~3Uz&<9*Y5W(_h%xfWz8$2fM(?T;NTGeU)^M6$zs!?ZC}afVd_IB)u0gY
z+fS$lJ?7t$9y7#R-Xya~0G$aP&5TBr+KG&(y1#H=O8@3LkFKS96phV(cuTkm)I$FN
z3^l5Q#aG?;*Of}qJFqIi0CV@=1;1l9C8n1zfj!x+v6@P~F+tl+CinjG^{NOB=aR*{
z?NI7Cv1ym?vf2wO?8A|!jJ5vKNN+2A5iZf0xp@SivE*`<5{u}tB*Xx#QR80DX0ME?
zIy;Ls0N@(1zV4J)w)_ra9h%Z>_1K1`#a+kK;jaI^*NDkUpWzOeab<RjjKm0S<7Sm9
zp!F?h_Nr0rIEs^Zw;v_5e}XK^y#8G5z=d!v{1n$K>LryGaOmmLYeM7u6i4<Sv~^^P
zG!@)NxP}p$Mg~-5sX<q%xO=JFfu!H4h;l$OdTnJ%MjSS2PZJ;Zu84v_IkHVC*$@L8
zJVLHpp<!C1;PwN+iQ>qPd~n;(25G3v5x^GFML?XRDYwMYsL*kvi&gXW2{rZX8IWq^
z{Sd$aear1+9Z!mTt*9pgEy>9=!fQ4ldwp9>*Vok4=sYx&6kFf+9rs`X;9U0wJ>6K=
z%WF2E9;ir)EOL`0lBFEb_;xy|_kNu@JSP6x!pW3(k5vFptsSqKv$Vv~ks02pt%da|
zp(l@A*$2yw^@*sZ^3Rjouzd8!<RW6Kc%lo|%|F{J=@56QN^-ZEy6#eE+aqRp0e)%X
z-O-an9*dQiVm1_@Y=0vyFV8PW@Y?sLOAdIg;ax@|ar(uhnxOjA5M5U&^z095lk^_s
z^B2372nYe9AbqbJ7k$_q0Rzg|M$RV5@>Unb6MX=-fG6^otV5lbA2%lLVVvwQK}dfn
zt*rED4SXC$82h{~NMT<+fBZMb*~sc@ae+IdyjEnixdvQM1YOO5-&sOc<W9S9jF*7S
zX}D5ki-|Ut*Aq&-=0b0&%R>!0MFt&s&5X=vS_6O!q)*RDHAocS(bt>2kmp*}OlO$7
zhAR|HrmO1PqO>UPaw|TK60P7ldM>Hz9rwoMKa3vM+vk0_&Ci3=<Fur<O|Y%)H@f@=
zC>xGdu@)oXxc^jP6L<qEd7l3%C*k>IE(~%5>XH|;qi|$Y)eO<zQmf04ZgvQsu6`(X
z8*ot@grluWA!&+ff8kn_F%cSeEkPMTf+~pwFUyG;f?UsM_i&($lxMaGJ5<UlUYH7(
zgeUZ<Jyw`XdzXN8)*2^5qs0L&#x`MONmky;y+U9)YvwpaqEnU!Q>$(#VzXLMI%Z6d
z*O1`uZVG->NMtgdCKYh?(3w$<diZa~0nm?kh-pa(BtJa7nfbwIeDz(9R!YIov@PpB
zTkQghvEBvOWgk^GaqZfPv58ZELL7M1XQUVF{e9jD;)D*$tT?Ek`zTN>m=~Y??rtYT
zisg!}kyx-Q;aY}jUXL4mdM(oX<TNe!inIxy6-sL+$itDw2Fq=qVNp*E|5)DS0l>}c
z^_6p)n~>zk*Zzmlq=L=Np7*{K#3QSwhka&)MnL1_sm&IZ<+DDPC*{K?W!i;_2J1C{
zXli-X$m()Q>Br1=^cvF7cf>|U<jfKok^5kglJ_3z#wY!cP0Z3JeTZOiSTfm%(D{9<
ztG{#WR++!i`ZRy6LL7c7&t<!u5AaQ#H+2#^4>PJJwI{Ay<>?RP*HNEBi9h?KpBKcB
zaw`-LKdT@@uyvEPT|xQ0kJA?hDs@Z~oinNLO|83Zv+)nZtC{F(?_D`dW1)P&jQq^z
zN!}?4wsZg#J^i)B!D*wNA>Y$A{b<-<l~vOCq#znB392e@H(dl^Y$&cf4(K(kVZ;X0
z1&nCkb7hG;xf`>VagH!4xX&#8HFW7Kv!`9-vJ=@R4@;y;@ZW^FZK>M!7N=0Q%_5+$
zz;kG`s-*72t-$M764c~ZhfyRwJPvr>iG+-jn}E?$Qap>9WEwreJXUqVsm>t}`U=T1
zg^PL_C&xTTQ^q1VtXwne1k?v>n{D(NRtk)H1_(Z$bPPKSr28{8@xFn3L>>+P<~Y$c
z<hq@b?)hgkC%+BZ_q(y4XVy8%Xknlbe(Jd<CxIw(d2+Z|C(CKNYq?X~GqAIIrI?+F
zn$f}j(3`8G#(Pbp!rax+;Nusb5fS^a*m`o)@qQKY4I~8NX-W-x5P(%8-r3v$A9s}l
zxmu0X<Xjns_!u8EYnD6*iH%^5Vgt%aVF@O|#V(^J9=NL4BtW#Os0YB$yNC(W`a}3W
zG&H4JC$!;cRH2e~kA2ZS7<lv3<!OGG2d6(#-YBe+D*S7eo;4Bj@lO@0kOJ_oU5Vq9
zn#wsB$P@j>yMwja2iS%QvrGyR|7O4m%$p(vHt|b_s_ZSW^c%iNbNU>_>KF8Q3w|61
zzKKS>f~jZw=P7yzQr{YAFi@-EX@@%`-tpVUG~u_jW8f-6WpR^<es-^loBAaG!DzQF
z%8Dqpsu<8Cr++lmM~BhL-!u-5ibi+PUdM9_DJR2*4PdN}D8PP6?_DRiPQ068U7A6#
z5p@)(&Cw?wS$$17VhGr~%L?kt4<XTjKg0PqLndf;bb~~Qur7iE16YqFPo(6}_CS&W
z%H9K&%N%u@-K5Zr7AQzR%NNWPkng`BkQ;yTgeRD>jsi5(bd&dMTq28i=Pkb@s73}}
zhSY)v1TQISTtH<@FFm?=S5y6%&o!D+iPGi~X1VXLfU<=aJ~YjCJ`70QpFvKEY2X~f
z3F`pR&U+!%a}hD8nH#c0hd@S|NrGLzgxe2hGMAS(Df??GtFijM2q>&Kt>!vjP-&Kf
znO<fUEA<NRyV-Du*!^~hG&F1dE6!kJ+p+6uG5yn1G(d$rqdw|JX&J<kTjr}Sc&(aj
z=StUt-&p;|+@v^=m%arSm(`Z`zdyh*D9MYDSde``=&|xP(^J(cK1dPHO-#Atq}ssA
z54+AZ!{r#41|=yV@yNDc>afPA?#DmIFwviShB^?p_!e}A+Qegb;kHa6BME%kbB3yF
z5T*Wj0PyKP2-LNUt`+rJgAPccK1Mx3Uk0+a^bu2nPhX4aDX4(x7^K`*TMFjL-z<e}
zm9_fp+?nIb1wVF}HJg6u48i`c+KhRr>Qt{$<^H`IhNi-Q%DCSYEKCl0apr7yHlXBu
zAIyC?xgg5MiNo(AZu>fi`!4BQA>C<pzP32x44|(W5`mRLz*BX3)zcX7_oN`4Tc0%c
zjl_?*SsrTP0|fupduKA3y7L;bYPMG05zg~^x^GX2R8bY|N>WjPyyi*0&E<PUVaFK<
zO{|<xBgbT4rUDGD%?<FRpqT)OJ7d+9Jz{c+OOD$|4k$Rl@n)8oUhpWk5_IC6tnW2P
zaXl_Oa4>_HJIy`+M7fRxJZC3^k3a{1gI2GQC;t~gODA?GLQS{n`9as5+If1}Q((Z(
z>g}ojvxW}Ll0NT2iI%=Y3r3MnI?i4QVyw_UHqH(U4K_}M|M<Uq*pivn(b5*!$ABB#
z#~0X9V8H<SdTIY%O9zhQL`{d-VV?wkNuS?g{|nj_Hr?L3%N_v_cF@=Eo-TCBj*)I$
zA&io4b;;fb*q;UXe>{=1>^XQiC!-$L&f8eViVbGKXis+&$xw((kxOqZt|CDP-w1;I
zSQ#1}ec-c_h{wtb)qrob<WYj~=}$>Fz-H2hwi2hm+P>wV*6c0aLPqx8r11^;*Ww<k
z4Cv^$|7$n&Q1z6yNM<yx@)k$txNnLm$+c2~$m`Gn>~%v<QzPUjZZD!%(x<jBuJf1V
zGaZ<ziR3Gp)e^fE^Z8!;7UdNzfQ&1MYZW%W45%$|x2g{{!9Fa_*C(pi;P&V5q6@C1
zfKc;oDg<()Pfxb55POxrsT_5X*L7rhRDUVTAE<lyJMwE`94+?ICZwcBP)Y;^w@S`!
zIKmAAlGE0s=<KABXl?P>LW2Ze0o@C0F0c$i=+b{F{o=OdY%7EcSV`J@eyAE80)%-L
z#WG#-Rj#Naimk=CKRA5yl+N;U%7g;)$$4L91iE!!U-ymnGf?nvoz3H2**)$I{H+ZW
zIclqRBBM|<6*O(H)SuR9Akde~MxV}M6<n%--jaNf87%GiZcy7}%sbLK+-U|rHWC3H
zA8io;JCQ{&Z*|7^r>{5DjNZhs`*<3Y$HQxPwFYF9H>z9amuH)zlEAM=RXXAJTs)vq
za|RsJ5U7jlaov7@Be0)_7Z~$YxL>)4{-RD=>|V~Tdpgy41m9;T(lWHgeuUm2O-~YV
zivYVUdyog-Bv-2qC(46ajk!cz8qyxE_Z&o@g0v(NTvz<%I!EQuER!xOJpw@vNqB6G
zLlsPuhN7=MTz3R}8N{J;wAHOWqf-g4B^U90XQLtOXLVP!RVAdMzc}gdwLAk`R~Iam
zAK{Q94-_Ynv9B|vX~Z&TpkV|LLtzB?%RKjxGv}(d_E6I24j98F0HeBJLzoV<!;^G#
zE(WK0gW0E$SCE?K3n1{=gw-&mcj46Kv93Id;pTp$^Ad(qX+|$w5vYj$m)^*J*Xu?D
zwpcGhsWNkY)K+fW`5KZXRdu(FvrQ|Kn^iKu<KTFoci+KmWn{#v4``vvMl}a4J3no}
zt~o{&-aI5_!!@tR6(2fsba!M8*CsKL{cgYYBH2>DUu)`aG?E-cMv1~hf$3IRGy5Ck
zOVrTB{q~*W&PfDgGdevaN(!PeL6VZo_XG}l_0*4bCJ;jz7W@U{hVE6^`km4jkLRVJ
zu%or+bj?aJZ#R69x6*!Ph}I1trhCwtHm%nCXo~yuyy?iJ^&-8CE^(6PW|l#2WU|W4
z61Bc~Vd1=|hlf9Xi1!%mV>$9ZsTVq$xxp)AecB8t;rYD=ca*y9UUj6^@3Cg!dQDgc
z$)c^ww%DircM<GXk(86JMOQNSS{u>_2~H2V^5E3{q&~uTmbtk9(u)G%ZK(8B;PJHN
zG>P;XS0EZs!_wYRB~8zIr1L)jkU($0tmG=sB}(g0-TA+>&yTav#Je_grLCD>M(}9n
zN^L(;BfFrv&2sDPAaNX?v=)4LE7R_Zc%dDZE0t`g(pC4%jHJs>y|mlHDtS_x_qbIr
zTAZx)!prJ4)O1+^8Fx~g>&+O`N`Ic3`FUf-GS{zX3Vzjc<(Ber`l94!M)PI<i&bJ?
zUG<iP*ED`$Iv$%k(D_^4wa_Iz>D@xgR18Qv&PSSzGyG}1R4H4n{?gPulq6WzX!}sN
z!+cI(*Q*)df)vO1z|~k8+i8@4;_U~>m8<uqnX1cd&-v}csxKKUJk~5ht2=LiUV2aR
zS&J(f=IDMzsAWHM(%+l!$L^*|JAuxQlI3OlZpf{^L0VdDt;0F+-l|1fK7V|?{Rh@&
z)0bf<0TTi_IhPS#J|zt>H8?L!Wo~D5Xdp5(mk~!cCx69SZEqXL5&o`Uu^-bfdpj>X
zyITYXoVZQ{w1r{2KwCj5REJ27xTGP`X!`H-%pOVeOiDUQv>G7R967f$w=?t1Gka{U
zYDl5ALLr;~<v_K$AVHX1(4aZHkVCHh6AOg|LQSEne}#{0?VoUHoX;TO%8DzOKq>i(
zwFcti>VGW-66+E{YSq7mo@#RC1RqMWF5=%x-2|kTT(fcDR18L%1cvCJX<!kz(sE$M
zQ$Vu7mei#ZxJcKlfot&xZY@Yk(v74bB)N-<r?%F`97Iw*bCDos>Yvsiz9>Y(ZPvP*
z1cf5qd18pZix9De?!1dwx8~YKtlLswj0~cVi+>e}wYW8^LDOcj2C;M}IRz!^)({q@
z+?`|&%BfppfvDXYBV=5=HO@grx-~Y4rCY<Esgk)hh6$q{B;1Uh8%;o=NM8XrYqcKS
zpl!4zFxy630<&$jB^q)XZDBWa8f}S%Ow^wYA;@I?$*^|jar+Ur%-YQ&{+VlkGDSj)
zbbo8g5KFhF60&Z*g*6=)V=R_O&oUHB&Tb7ml&g)m!H34<Z6#RBqgaE{####|Jg$&}
zY3}YI<d$pKMFm*lZVlgA$^F@|j+NG*EpxCs<}Yj(_a7o+S7^)}xLT<7CpHR2xO-+C
zijMAJk4qliLFFKaM%@b0x@XACQd566q<?U!+HXVUmfE^uq+Mw;rVFHT%S=P{oPvlN
zs+$zS4OMa%5vHm(u9ZWzF*i}<wTPabcWg-E^#~3dLAHW!g}fEag^ROqza8HG@8e>4
zzFe->!<$e4TKm<X-Y<U%!*{EzpB7i=37Y9*_~Y>7@clWB7q1uZ*5^_+Dsl>ujDL&@
zL-b=ov@$*ivRn+$1D1a?{IGhv8eW8NilP{ek);i|f`3~6-8206C?~LDtl08w62+t0
zBm!J=k%)vl@IhddkpWGbwSrwt<pe%7i%mcKfTE5C*>aN;fK3AWk%ONJG8NhJcrTbV
z!XnBoM@1^KZo-putSujSagCeauYZ<NIG(_|l0>gxkqAfuthDGLpA}os4!6LRO)_e=
zq(&-PQ#-*9AQwt;oT3P{^A-@%1hP=1n(vj5FT9|t*U0d<zyBkQS%!uRP+xxf@ZsVH
ztr8u|%pV@;ez#_(6d724w<a(>vg~Vy;rrEx)zzDim+ux~tn}g4<$AriT7QNS0fs;P
zy8hwK`f|MpBQ^{#SIaeQycEa9&M96GFTLRr8ce?cM?h}Mc;SH7IBhP_#=Qym3-6bW
z>gF4?a3&frig$J-|M-pi#YW9J1~z$fffnzbW>oXlc)I}Z@%B4@)73gyJ$p91x>~(^
zvsj-GuU@<y-Y$M!U+iVftAER%7ub^Jda+zzdmkJvy<S|eK3%<AT#xL<Kl=0Hr}vlN
zt$yv9WaJ>s7r6flcS4JFG`EdL{TvNP@xv6TN5ZA3DiE5l5U%Og?+0bF+moQ6Z-_`e
z7NN=3??-47{Asv82jz?G*^Ax%!sTp7zKGXI>1GFPgpBJA9~~oj-hUbC*z(%8Ew*i2
zY}>Zjwr#QPl*?AATcJPVhKL^a&g1BWejjLykPLigM+c1s+(OYdGnhEYO|)beXA`s=
zp6{0LFb&-L>+2u=|8M@iUVps){V*(+@xSkXdH>NjC$6r39{g(f%j)VsmmmK5Znazh
z&CVX%+62G{@UiF|ntuU8PKz?&P@by8qcuclw;La{6tp$I(TAp1F`|GLV}nsOpuQ*Y
zp)tz(vlFHkiTP<aLdar5Au9Nlk9&rF0V@Z^kILnV=vJc$Fi}HVc!w2mjp9>?q3s{G
zSOQ(g^o3!rVPKOAGJgWDLEu$B+n{-70HIJpFp7^@5!37wIe&rS19?b0FEAWM%X*_1
zUnMWu=@uqc_XUt86=Cry*Ge?zCrXKuq7P*P{^{od{+BT8VRVXB=zigNaXy|b+)2E?
zRNL?=P(7J2uamXixm;*w4}Jg_x{0UVztgrOo%R9LXgp}SjT=ohob8&Q4PYX{(sML5
zMJhH+73sz@YkyUY(UC}IsY;6mem6rEnH*VXZGf0h%w?8pMs?*`zA~B=i_TIl93{<q
zEFUZV_+^GF2yZl-wF6W+7MisK<h-fP@|8fXtjyZ1-b23%Kc5uvbTrzp^P9Q$E#UId
zX~@s-(qYkaeGt9YyrtLU8RVl+l6HqUs5U~VopqM#Cx3@EmNxI?q6d0|j(2sAiP~mn
zH;FzU&QQ%Y$HH}%Dp%B*-ln=k8c--wo2BX}I`dW)KZ=4Qb5yZvE@hUgr0A#jGj@PM
z*EzL0t1=MW4|C_JCdY(XtCEzMXKkyahur>DTh-DcTm9-ZV~P>`P%_~HH=hXHau{<n
z=V;DT+JEB_C+i{NWWF15T&;cqIl1jYj@ot<*Rqr9PTCFXIqf7m!Etx-Io(fnEu#8W
z9i#l+MNEef#n~_?*4ii==#IY<IRVoKyDh~Tc8b+Q>#f;e03s=moNofsGUBbtY7xq+
zVyE-Lj{=&%GQnR=p#GK)m~=P^Td6IJnQBsuJbz@APUu9T*x!`27MuD@LwsHjqX7&k
zIs3?w7&Oga*VGewk$vbuowTT!N!ZAd@4$)5t@)eC!kG>AF71Ra0Fbl44cl&n3Wn3e
zeIV1;lmv5`p~&D+nl{iZck%f>ja~1ofY>9w^N1=bMWp5b@fdrGsgvW=ImGSWC}?(w
zf`6tDL%||@C}=x$GulbKljH=nY?EDT$S&;Y25-~=gqfZeZ+sBT{;C=!4(g+dVq-hZ
z@Y)o^0#O4#gHROJ6Z!z4&HTB_6rZOjI_cSAzuU$JFx6Pm+5Nqns2nMEBJ7#}aDmZ@
zoBvW`TI!Z(apgU=E&Fk<j9i$dDix#lZGW0uWDq<^E_KeLjIlt|Ec<>yc7#+Pu}_q7
zNB#N};RJH_-=s__*<>P|fYCmHO>pieQh0C7+Po*5U{=i!!>pF}Fl+Negz+~FTWQDU
z%8t#I9h)mVHdl9eS9f?<cX(H2C+XvnPch;Nm0+x~_~EV}sG=d7U_>|qedNCtYE6{D
z`Y`Y`VPZaA%^Pr+t8VcZxCSTm2=+;#oJ+$f3DVy{wokydq=4Z0uR8n{h&b=9(W7UH
zfxt(&R;`Y_4d4qp24pEC&-o4A|4i1GQK&W)7BMg}ATS_rVrmLJJTFXTZfA68ATcqQ
zAuT>9f4x}AZsa%&z1LUx2ci)vQVUQhAmt<t%xxOE4dyiH?o%JS|Nk)5W?L#(xt2^0
zj#HLQJ}!@>tcM?i9=_~!xgXzmf4s3Dz$muP+2Q^B0J5UxFKDN{<-uxHM$6LA;qSfH
zXtk!JT%<J{xt#p(<<5sELtb`(jo0Cy_dknpf2+_Fxi#U;PLZQYxCk{O6vryea28LY
zswsDndxYX=ML`jwUNwS(Q&_cRfO3fppu>~gbW~-FHB`jc|D#G*0YxB;`6=sJxmO&V
zHLJ><QO@yXWH?ZeJ1@*gddDJJo^kb`he3oeI9!#?tc(j)vUn~db~#1{=NREwB)f$p
ze?#7%9hn^xLqO5Cgz5nMvXy3_7%qV?zzwYdM{AW~*8zWiv*KQG(F9S+6qSw{d&ZT7
zY87lu+<5!a{d#@h{n}0L!I;lsAn*zx4FCSz#RURZR+J*D05tsmHhkax)l2Od(l8|j
zLtD$-1Ou&N$pAu9CfH>z&yKvyi|f$NfBCjrZM;`#;5r=D?`hS~E>G$0mSKoHTLLss
z&RbG;#a|Hzy%a`QJa!6jBGM(xoK_h8l<u__P}ZMx)_O4!mUY=xNRe>31;sYJ$x{Ao
z8`ZaXRnJ|vKIy3*&}HZ3`Q*K^te7I%FEAIv!b{GFk^vJ&!^%tbH0YJ=D;STOf1BNB
ziwLnI@4B}&mi*vFQAl>xVlsBp&-kxm0~8M0d$BcduVWj_@7j<|$=-<fcoS9(R>4%L
zd#sXs>^%w{NiJ{L48J2ePo93=wyV`N`xj=O);1)Wa+M7wU(tcI26rF(>4f!7{zm46
zi;W4TF-eWT4aH*}#dE5<okp&$e{h^^<O9ZTc3LqGi2GXM5~Ha!|B9kxbXtE~vieMY
zsXO8?ocPX7HNArUxt=ZBpg=cXvdysdR_-=v%(=L_aw#8c*052*Omi65e5`}QJZYWO
zpfZY}G0MV6P9hV4=suGtk?Ln$L}%$Od-EjH;9d+OGNrNEn9}Z@OCUtuf1fm#pv><U
zXJRy7z15|%aGZ*OlsrAme^LTAPrlA+g4!rBu15neXPX-WC`-0~<FQgA_{RrEE0Cli
zi}L_-AllFL&vZFt(#zl@ChYSO0|5)!Hwq|NlBl^})AZIEd@>jX@+WVNxeR3-xuF#6
zKcJNbDKnDAozAWJ#8Vd8e_BN66V`g%GV2!;@kXF9$P8|<2c@N23(@!U<eCFg%Tuh!
z<p~rRx>oACKO>RTVh+djWQ|T!z-t{X6@Ka<i|U>ObY*%FdF8ynF+JjG3tafr^pI11
z{HAbodW7zDfalU9rAB{X+DH#}a!i*)dI#!%Y^)#wV+;>8R*>-=f9LNnAJJLcaWix5
zbmBbAF#*KX;rgV`8DgFLtj-y~86jRbqu<>a!3G7AORcgCeG#0Kfd#r39&L-}HC4%*
z+d6wi5|B;wkumbN)~SqSq0N)TrJ&d15l6Kk)uvTq+OHEUr^-`7UZz~vyN+M8FdLbr
z(}eJ3bt1)cZ4dRsBtl%l=A8@E#8x#jU<IQYjAQ_s$$6p>L>W9cHVGa61Ix){?Uzxg
zHWUFdm(g1`69O?dm$5WFDSu6m+cprr@2}7wScV@n`~VgP*4{WkZw>ZV^i(uG<<R{9
zLx<FmA}Lvty-Lu7EL$RnZ^$=~9MW?5XOP2}r}O*Mr>lX%36IL@;r-j7=!K+DCF!BX
z!6_xIQN#Q9;g1(7wWaG!7f5LgU5>tcy5s5DC{HDn<x8EO9jKS7tbe}bG6u(5I^^1X
zSKNZ9lFmCB?NMtA<NfsK`)^N@p%<Lm2()3a4g{geI0+8`-GQtkVZ?aHQ)(Tl6|@nH
zdl}!tIhM@mfDoY0<+ud!Z<kA@r))LZk;Nj_g;1<PsbNRk(H30UCZ4Q`UdoeBb;I;N
zVva3<%a$RqCyG%<41bpit-O}Dbh2Gzz0=2n#!qc^R&V*aT<btTiCd8iGY+5?N@C&a
zCE<%DKV7|AWVum70roXZtsAjnyfD_!YF!#$`P`dy#W$w*a_lko(5+((K^SNI){I-l
zsHGJc!Ujf@Mn<d5m@t+Am$7C}rP5rok103B)S4gKqX4aYM1S~}<7mVn&!eiy%c_6W
zg#y*g8Z$!*0%!ang+c2r-EEM1_(9sS{O0SI+x?%}MWdcxG432ej6FrRZ!}7ZVB^<#
z>wkTIfBJV$U>St=!(ZQ@;&V_atP6v(A{aaT_<Q*F^m}i#R`P^jT8r4?Cng}VmCPX)
zg<MyFaU7CjrGL@pUIj8)K2mQsPrg>V?3}Pl^{uM6cy+x*o=YL44UMDK=1CfxAQye}
zH2S)4#AN9xlF>AHtD`hcc_z%P;pcI3LSU(4*)l9`lD_L#CIP>nCWo;coKLEX_*CV}
zMeKN)tkG4OIBSlx1mEQpeZ;jCw)Bi@=WtnO$4wz)-hZJB$@Bl+<T<zBQSzqPSn`Nu
zq4R83`-XUWBY?!AtIgu_=}OVG)|*xduw@pC&7G(5OY^{W(u#XE0;D49s+mZD7XW5m
zB{-(i2XX*6PrkB}+=w<-5+o(`07XgOI54g}dyuDEbupe>NZ+Q0^OIURNYRZ6vsNxs
z|BWk(Nq?6%0-IH$GiYQLt8voNY}}1Yhb)9Sjv+5uMxlSUvV^pazcky7c65%$?f%~E
z>?r@*-O<@MyCHXMYjKA84w0uQQ#5;a7L{6Fsyh=W-I+%_CZZ2u)EekX*ojeVFRX>W
z^tQfN!hOCUBk7D&ZiW#ePP)u8)}TqH`bzRrO@9VxhUsHm{nStDq0DrXNSfg{ARIw#
z@RYIPfxK=F9zg5X;0CKZcYY12cOlH2ZVPX&KOUg5B_<NBI#kFM^2_L$`)#EUvKVMk
zWKbIr4q-7;2I<U^G;S>(!Q<BA7LhxTei4&*VJuW`ORp}^AK|g43}lYUE@R59Z%coa
z!GA!xkAfR*AHrP9pmnfEGPkvO1an)9Tg>e|`bEs$g|RTV>!Z~SZ+!mHf<g%`XJf3~
zx)f8|6P)U6eU!;Dtu}rG!>xqn_<bDXnmFCDkCPHUFEBW3miEXE7#iD9#&mg~_t}8o
zvA4Ogi&omyD0DkZJf6&qBNF%N0duUB-G7B#ELLB3ZQRJe?bLd%9f0Ql&^h#aEqe@x
zpkUmNsCQAzKHr_$v-Vy{$_2aZS!)qUY}b?L{=Orp-H}X3|FtjVfo2f;5qLbQwm`YB
z>sxC#r|(IoJ41RE_Fm$QoSSp!F6m<t>j0|jY~yiapcRhDd8bP|IRDsgoJLPQjDP!Z
zlzlVHXiT-8-i~hGp$oHvoW~EQMlT7m*Ks81@+MCVv!^TKtgRP;lG(d_-!}#uy>0+#
zD$#w*Xn*Qq0y!pBxaTmTp8&K9boWP2047w6kxSXJjspOkEoM{sgb2^;;*Of;+8*rB
z5_jck-7ftR;=bcNMbT8^{j{CfbbsFNo_kCp6!q(L6s6<Qt9LK+!^W7$V~*^=D&SMh
zzgCM*Q|m^yRq4d)PP?Y1rgc4}igGH%Ro^;Kz)|?y&PbzFz8tFB+K&|Ns=Tj`C?OP&
zw9cm)tP_w2<F(Y>7`}AovCv{V_8xmw^65oBJiU&8g0RLg(QQZv98*;`e{W?sUl+PQ
zH(D7j`9XK`4-#K0*SJIua9&r=m_$P4>?C`N+iA|@(@5(204H_L%|oAZ`d=drU2q>S
zs){1rn~#Vp<(#y7I#yJh$#psgM4sIOUze%#O=S}0df-?}z0q~R_X=z$dL*j<0c5F0
z3YSr+HWUIeGM9lxFBF%NN;xN&-3T|7e~-BVzlT%%=ksp|nU*evr|>o0>R*q~hyM<}
zAW92Ez9C};NNE1~dI&ENj3Db9rGy7;zWv*LKKx!&&Ah662b%+iO%fSXz@|tT;~F9@
zw6z&F%|c<g7z%ufcVi+n=+kxLXAA(k2y3{9G5jaZV|+%k4T#EPBxh6{HYU-re>tV;
zOwQPdn?PP+|MBfhQ$6|gv4>0<ClF071bhfrp@n`1k?=-O#O5W-r}mDd;$OxOf_Ezn
z!kYw$+q_B*T3DkttWvYG6~SbctLqAOn-r6k!syFWkSgfHJn+-wpe~jq%R4A7$~I?P
zi~g6*R#~Ay_iHsR;(g+#1IJdWfBc)qS`WX2zuEivE3UDBQW#`=7bJ-ov=r!rY!!`k
z7O*fbS(`2Bc5=7zca$_a$3!M}ZQ;?@GHtNrgt`2%!KRiF)`YwotXIQvE6!psUek;8
zLb*OHCbQ_4BBN>wv)&l23?N5cQwJ-j1ixI5RYC?=={`*IpyJ7=6u-P&e~Vwnk>Gc|
ziasRuKMgP|r~-24(0v;i=O-iA1w%q>KjCs3&<h69ML%0`S&X|(2~<sx=|q(z``0QF
zw}V%kQK-di?Uenv_Y2Av7ib1LD&?aK@IDJRTcSm~0qjK>GO#5@bKD$VfhjnqP*GBH
zHnM`{h-#=FJ;7=N|D@28e;~j-doMYl<AnV+fqgqv3cc`pB0S)Hsx&BIq9K_P;v_nJ
zbZm<!V}!Ex7|w&vO)(E}T?<0Dd>AckLwGrB?iz%%a7%M9Def^l@?0gVn&Oh(WwWpN
z@2vK&vO&%e3{t&r6O91$f-S~EubLH0wywB2?lVor*@?aQCgC`le{+Xzo8$7e#dO$t
zbTy62OX^5+Ad%3OY{?0F{Klx<&cy<{Gd?z-*x5<6<Hu}qj`&WV4K?av?#-u<i-lM&
zGtiZ3A{AmL<4;2cQ(J!89wLV^RQmQB?n~35GF{ei(56XMVpmPt^gghM)hbXq!+U~S
zkcX$fcGheeCX~eHf0XJvH^hUjbsci838Y_X6N*~1dtV!smO<@w(y6lF;Y5QrwWrb5
z$+TSi+SW70uzkYLZ@4=_JK_3GY~HMN;%8Lw>aDhR9C_L7E*4jZQb(&=T>FY!Qtf%e
zWX+GcB}D{%#H`rdaKcLh1gVB?j}&1MAiNjSmQug_>wTI+e+>z5wJZEjOB6boq?E$>
zm{#Al@7K6!0OnFM`JQH2J3qrQ=W5C?>r50+!E(mm?^pN8Y-I|lR(WuTq_o1+S_Mh*
z3)TzTaG}9VupIFyC!V7;Rk=66e&8Ypoh@dcQ3WS(VYxWbmP4Xbm#6}a6=+?_<@+j3
zKpe-(R$@g9e;X`*1W)!8Sn!j|rbAFV&3lf^9^nSdj(g|+Z76Yk@^_Awi~>g?zMRf=
z+~ny>?Vy|&XrPvLU%;)@ztpfXLb7@*&N;~}nHsB8D!SIxMzhcO3K$iDtQb~CfY_^L
z7O>U+8F=Jc#ZNe_rlg!rx_ov?XN$9ourTcSZ3>-wf3!uTQJLp@6`V~Rz~!bhms4)<
zl#%DVd^`&7N9JWmwBqU#oj78>N6BWRuwLpyVY66N68jPAno3e}%8vb_a{WuA*wDv-
z7;IIS0kcv=WQZb<V?m-uz5fH<PqT5~TN?q@eScsI=!<r2%%sycY=(4{p;IL56-G(5
zGF~~Jf2Fr6M&U!EYKtV|;qo@D^29IIOZ8y^IT!E-wyM`b^J9mv(uq};onG4)?!?wy
zc6VYcf<H+oR&fm1;BVNPswnR0#8#BfbYiQ~`#Z4}!84uMn*Hl`V(Yc8c4Di+JK9|e
zL1_AU8nFtvltXn(YcDC@pdSEXgG{ETykz~^fBeeP`f-;BC~1ZEs;0W&0xJB+1kz*;
zZ_pHWns-DnS@)TyFsD?jCm4>-@sFu(`Lq~`H#9)F-eL?c#FlN0_o4m=z!rm+<#gS&
ziY+SS2w~4-i}Ly?Bpnwgy3lqm!BVKnC0Ti@u4lKAscCfc#_m{4vB`UN>%Q#UhpTEo
ze{6zLbM4Og4!jjc;mdTkqrf*iy9fZp^NY7XySM^rE-(zyPmDCZ3DWxufKXZzx*HY%
zHAR1Qxm!^>Q|?xy_m{gB!87G<&HnaBE%V1KoxN>NxssEQpJ+c=^h)Z=|2$XC+EFfp
z(Sn?x;h+D_>oKRrQOPnXId0)yGaTuKI61z}i7!Br)vh@44JQE-Ug^MBV~R)^aqLLo
z;2>ezOecByus7JED`mL!thYLUeSALr5AP!$sh3fxHWUFdmr)!s6aqIim*MgkCx2T@
zZzQ)7zUx=aaTkGTvRK6jAPk@xOAGk68{KRlVkb5dWDoHr&c9Fb)lIVBqj7*7Mm^Kr
zWU(In>SK|Tr(aL<^y%U9_VCTC#*-C}l!epVFDJzhB>z>CZyGvTrG(My^!D}i{gaen
z_;cY8N~tw}EPr==rSqd!j^9uwe}B~Z(V}`9--T;=M^kDAzO>F)l_fZS(>g0df0x{!
zKL7ak-Gj_Sm%^LyHC*~%FK-Y3dKeS|Ey(zjQCa{i^z+w;@BkGFWBIlbmQKI@?expT
z=f2Y-s=G$3ImK3!#t=1D10#qW_o{`qwpdN`_%VFg68M^~#5(J6b@{$c_kZFu{*Kqw
zCRi9<dPedb3`#HQHb0f}63sFU09jjB$Pfrrt_&b_M36c{pFdeZyaoX(2u?;8$lVQp
ze*5**-TCrql;|a^ioHbjZTK`U8(TQxobFo$jpXaRHmC%e@*?@sf^qP{dXz026l6se
z9#_N}p`eqsss;)Ju4|w-27d?FWN9pJy)0|3r68#esZBSl-3r^IciZ9~ikI1=ydZ?q
zvt-;b5`qbyx$R|~59q>^c1mOe%uVD)tjiEVuAQ}XMtsWBbdY!kD8EfTX~OUGBUpOM
zbN7<Q2G&01cPF#|Y;>@*JcF&A8=gl*2p`+@8FfA@^AvF)X!*^nfq#={9l>C&50DoI
zsnO@d&%b}nZFuGhZ?$WX3k<g*#cfoaoUzT{Kdy?gz&C9xW<!c~!kA9VKXQ|K=YS)@
ziK2H-8wzJo(J`xL(VWV(>dmq$Tq`Rq$51Cc#5a39wTziy+LB`#^kmovfp~FTY)8<x
z?>*0_MFY6tFn`2h+J6Oy>G`l&7#rh5XXSk+LkWUBOgz0G?w5vyoM|}OPR4`<l2*-z
z#<5|{5woFXjI>9>9QjqL-7RKwUU2K1%totbBOS=m#39V=QY(h)H8eFOkWr{G$Bg-L
zOfn&A8p|AFZ}U0MBRn;I&*wPA`QbXB6EdX1_=j?KZ0>x?9DmQtHIBy_N;n3l;;C;l
za=(uD2Gw!aA|ZYo1(!Q@S|%fOz>iMKb+lf<k;n>dBW01BrbOR$$o-6zlii~scW%d&
zO%(;jlBi6beq#@&r0LLS=!Xe^a_zRgH;oPKL<I+oy=aDF*tSQy9A{1jg+tr-1ah7>
zS)N2>`nM;xI)BXEx(=0EU~P$}_u_A1ZO2Uo$hfK}$`_X2jTMM9$izdkScvp#{1e4C
zcWMij=#ea*^wj3TT(=q_4qsIRI{+_R;f|bknSb6f+`As%6L!oLm4v>>rMSys-^;(-
ziuks8Bo}j2uLGrk=xPlkdsPDD#5>Ho%KGF|eyQ^7_J5#7%kH%1K0uuVJzD)-5c|D`
z4~wo)!ok;6mVmJd!;Mr!iK4b_ZK3;3#9-+^wCC$u^=Y`v7Gw@5L&Q77SL1o2ByZ0>
z`SNLz)eSTU+ccHe4^+HtMh4Z-yj;O!qQ}zR{o|+v(GX92<0!-o<%LdM5anl4F||rb
z!86(AoPXOY_55l=Y0<_i(pgbhwbM0oOgCcFQI5&LDUeqr|C5@bW*%=O-UA_91E<N^
z*mzeBF)T98NZ@IQ;IOEm9wp3ic;?X6Z1-*iGexN|kxFU>O1(CBR+25SJAE%B0)-pn
zGLh_EsTM*Z5bVTmFZW7!$&s>kUUcAm0Z|((0)N_6%iX%>5TnE0&1t1B*h#&&)1($k
z+g{3h+y<e<fqjo_Zi59&&LzkINKf0YVwz@QQYMNswIC$kUZZC!VU+5ntQC3F+M;VP
zSE4WIg(R$TTQ%Fe)@!wF5wohirex4oUeAwXqB<MQ)mO7pea*dWPSWxw?n!Lc%~wXH
zuz#G;v|`<|#XqQ6GHO!O_#O!9<962fKg!cz`QdkgUU+x<cYO8hNwbvD?4PGUbDoMB
zXhXJSr<+t7_y8NVU&c<JN3{#anwXiD;uOQ8j~ar?QqJpAm}qOOTPdH57;9&XoHCRj
zZQJHgAJ+@R7=yiT@t0LywDLkTRvgeJ6n}(On1gc`TTn%#U?zP?6~+_qIw?P`*K5%+
z=GuNX1umWAf40=V3>}(;vlt8a;0vB!WRJ%43Mj@Abhc^z;q$5jidZ1|PGBjW79|>R
zaD_^ja+=KQ*=mVkhBqt9T%OXUJjLgH$CvW`@Fb*xVbgnN1oKI&JPpe!pW%@gtbYyy
zuJgy~vV@%6U0G0QW+$4Ux6)miN$25(O1D;wGPWN`VH``*WO&9A-5O0*>y4?+D^+Iq
zI_+-RJu{opMj#_=Bjoq>kTtxLFt*dG3t7Y_RQ2c_GW)~m|0O&b9xdCqv&j1x*uHp_
zz-iO;lDFcKv_OTo#$(lb1D+K@Fn?qxa2LAGY&x%PBtb9);zZIa7cDF@65Z96J0bTD
zeeXLrlK?`|aokL@nP};%C~7K%a@DrkI*A0uaAaYB&urn80_&2|bu5rXucU(l*NXNn
z*_;Uk-|9y3&b9WU4=hhIwqh?0wqwIt0cJAJ8<v#IG=vutO3z8C_V?Mc@_(r+NSW$*
zhAqTBlNZqK9&h$aGgnf-X0SmzUeO)&7&hA~Z^!S?!+3^meWfp0Z|A2)M|Di3T|A{2
zRNga<w7t&oG~p|*<tx=9B?EMXdsk~2`|@&}YZK{A%)t#a61JmOF)<CB3SG-TDrzMY
z1vI_1;jqI(D;17q(WDh^w0|NYnT_|F$?SEG5L-n^j+!A_a03IGYFjWz#}xRjnzUNR
zpUO5%v)K#s6D>o*JNU5L!l0q_y*O`y?W`w#P0m&<q<zhc4mUp%YDR+FBe9B?&Titn
zwUa^%qvH9fKuvmD_Y;Py<<z@7ErCFSteMWKb(vc8(p}BUY<0Z{Nq@tzjY|`69p7Bp
zMa(G7U^zjVD#jnt;=HyKj0@zhI!4iPnmB4ZUE+@Tin|yGW9-X-$R%UEHZ;539dW(n
z%$&Q()Rdf{rt4ic&?7;|D`y@Ud)djW6!EIi+u0Ht#^c;NZ`=)c>J6c;)VjsB6h|u>
ztPhh9n_0R#niQ;Vhkw$;wPds5ZavC)2WK{}%8||D%YnDHMLF)ljLVrKp~8E!j`z9!
z1{yzV-*Jys{6QNza%Zo~7ppjEP!O5E^@86T3l6{hXuDya*achlIxU={<J@Nmno(^O
zc96o!wn^A5mSaXm%<v0`Zzb3=eX44*lQ89!M$*skxt<5gSbx=-i$M}pP^Ll884?zx
zNF4#cOa@$7`%`uH!<D++6dsyTUp=k7es);tSf#B;H&#q&wh5a{LD%F529x#LE;vSH
zY+DWW(D^!btcSX|v#q>p>(uilRJJOwcvLH8y@U!R<9<-n<fnBD5D$n`=RpgYup;*J
z7FZBuLovV{6@O%T9)+n_xBAb<scRI*V;`g&C0~{U#xB%E`Wi^{a+i7Z>Y8d_AWOsS
zp&8Q<$4lIw{Sp4<sTtS+jA>9yz=%-eKzujiMztiG5R24igFHn{i2VgpJfs)Q_)_RZ
z-r>rltneZ&6(`Hk@@Bo*TL5mwOMldjAvbWFsjO1IGf8CF63m8?8B9$vp04>#`ejGZ
zZH_Gu?>D?&HBnP<Qpj$<H3gqxs;=51L%*;uID9YitSm9$K_~i-{~Fe8v3J*(w}<}%
z4_alH;oK+_1Tio%G?!767ZU+Dm+>nXD3={HI0=_~8#j@EPapsB`s1C<te5x7FOqOp
zAB@$4xyt-LOX^0|-kOAVcIX|Y(M(S>eXWpNG*c!BEefA?de$sQkV`Yi#2k~2G(MjV
z9K7%bjBYgvtVvST(YpWVLmINId{89hC=1gK^eShpJE9{5P6CISll*Pu*@q_`pO}08
zp8uq_X(k<iq<Nag#^-Yx5oFWnETe2?<(H|w5=UFoBRPTE^wR&dDLoo5v{j?{WRh_r
z!-c&nu%=?R$XI5bRiMmcQ4BmLF<F2umNjzYZ3fd?3*|?SWl$&DId>-W;j&cq8|nZU
zTVz%<9aWeqm>i19^XDjO0^tnc2849bLnwp5<r$xUH?^UHHMpsrscM5A#L^H;!V~sG
ze&*%%ut~_FP}=pQ{<}@cOJStGm5{f>df$)weS1MlmdN>0o%EE~(mAaiRIRa-uZt>Y
zBNFC*r{YDjxw(z1XN;@R76YYsRaO|eMNbVV$S`l{1W8vtbpY3M23kws9k(Q{vw;+j
zyHwAA$P39w+!Zm4^NFqjdo(6gPfjV1XwaQi<?|{~+0|y}l$_GGTLsN-wd3>tTQ;J#
ziz85q;ECE1KeKK&!2+lSBK4zw|LemiWlTlq3O((G@oq*?J$wRd@BbFw+&yd}9D+v~
z#k`1ca7w@^YC*VVj)b@9s2{?OypmX}TKhDA0=C8TD7br1<j*V=z&&mX#bI7pcdJm0
z7-rip3$;QBI4N-YRzirp$ooOtmPRcFjUdUY(Ayl6WsKmh92(7XxfS~^vuJ^xo>^p(
zw!^Vwcd18*ASZWc(|T23QmVzYL$(#(ei)u!Uhn>MH+mYh?DzTWfBkwFzR;Ze80&d|
z;TfSg3jg-+!?(Mi#%W6yC?t<b|41Hw`{gjSe9cL&mn@nzhNyRTMi6^d@=h44Nv{6V
zWEeSB0@bIVl|uY95iTuXn`Dy=yAVI4(gcPS&1))IsC+*2Dp}&27psDCLX_FtXE%$G
zCl5LLY0BTCW2WSMb;)VjB^#-x1c{%2AkP`*;oLnL_p-h(D%%=i5pPiTsv3%7`x$dW
z!_Gc1RsKa@yO!2OVL@p-&IEF~dSJvNnX&gwS2xhl2JI(QYMDG|J;*C^B%jS0O<NLt
znsQ<2a=(-u$lFyFqMFB<@I!f4ds9q|ht85&Q7`);iitRO6+0rTpOH>)M^rz5(<Map
zy>~!VKj9@r4dlNwq6S`R6H$HVy%B{VoVR--YM9Y9ReMQ%b+?~zleN9(Tx`bofVP{A
zA~`B^DLc%lTmTWjaM?H<yl1lRX&25Mz4d$AWtdUtxagZ=IZ%K;cE-SMZijl<P2<FT
zpt|A}fKx8TY!AKh<z1cu&<gf{H{+};t^u?X8tjY*Gi_r;?PoUuEBeDhcW-bLs8Cpf
z8Q*pjkr<GpW;zV%Yszq<(+b#;yCPLP&tJlhNF8RRyn$4945E|PQwbfdvlD7VQ7C5|
ztf7`yzgII2;7h+jl_Q?EPAIu!4mC`8A>xhXS741-D7%OFG33&335J$`gf??Awowkw
z34ob46UtRV`3&>ilK2BErAxtk*V3*+<l5cN2u%fp%7M+cMMb;wlltkrn_Gfj*ow~k
zm`#tg_1?RbvUxUazptfK+OWIRdPjiu6J7$?K>q&+V1pRuI)L?~RsnYE05)@F{I1-T
zuxCdi2oKZ0C`H%O+M>9BSxu3znlt7@j@9Ff!f(uN-L&giEAW^17Sc(N>xPboyUo^#
zt#DcvcZ*@_&oE=(5T`UUy~;jJP5N%oDlcvqAdXsqvHNT+jQWVoz7PAOK8|(cqN*h6
zLO!>yv&HM=2Qq(es&zu$Tao@zGoRjXPhej-9S!`!?P>m?_M{bmd+v~IN+od5s%V(d
zzj^zkpXkEwN8fpeNZrkMDNy&s-w~&wJJdSX{O2^4`~eT|NZ?gh!Q57}w$R?*g?OLq
z$<XB0A~eFn{z2wJz7#o{9Cx<+yvL$mphzD}QA(>9sP?JWbYzc$EFI8>>kkVR(N2!-
z%W;3NHaHip4iwIR>e&8+x3Wfe0^a{6pFMF-xj4k~5Y-R)(?bu+<%1z!TIjf$9g8#k
zZoGsH*`;s3R;RdnB#xT1B6970lyVm(YR%Nc^_h>!m#Hum(gYt9hvOB58aSyb#ncoF
zs&6R!HP;-#uiT07Y!i0zy--hQQw|2hodMFh8i>~q+~kFSZ#^Xh$1FIUof3w<27Td>
zaCnagf-CVadoA!<D7qXCZr1Q>G)UEPi3V@jLuY+BH8QTgEYt(6&756uksoL6S~l{H
zy!XDK9m<E9!DYny<a|}W0lcoQvg3?>Y>Uu;dehDoN-hsIez01~&g7_<#S-%>y>JdD
zz|J&HWq<O29lVApgENgliTbDO_~Segc##dW#1#Fsw7~0p+qaI`y$Y{s={_c=i+qbK
z@o2&oKqZe5uNnF2`<P(mZ6#Fnn5EoYvxU}Gmy?!51{g0?&~h(NzsU2#?+~6dgZ#?6
zT)mlrA(+ufC0)%hH(k*hwN!2a<SQP!)&G30mxQ%{YOY=_T&7E=(`1?3_8f;-l7*VP
zb}3l4V=_XzT&mzn8{vI>=vIMK@2?O{(J2_etv#D1yP6$X5)MX6b8o&1eQyQunjI!6
z+oHB(lhSq8nuPTiK&!Q;>!J}=D|dKv@6$yzqXYpGmsspw^D0n538h9Q)q9Q6Q?2_1
z@Pdth{T9`h8RDrMEO?bo!)o@b#*L%LHJoC}W)xgEz2}@}#$T+{%1W^-w@P(cIc^5w
z8GK7Gljp_p(ou^bmuXFRQu}M0u5bv=E!A_i=BRt+kwbuA_UN|3a!GE>&IL`}6fbSs
z)^fxBR@tgYKTVOMEqcnq#Wk8O>!{xIot~b58VatpRo1PImEf1<av}egfeT`RUa$O?
zAw+t|o`-XuL%5n(D)L$DR%w&3iT9MQ%4<V>0^9tKBmBxEd_HF@LwHWLrb_h;mTr`u
z<L5j@><;;dBIPF``TN)&(rq&&C)>+jvA1cizRx}C4A<Wsi?XHhk(jhIOr}7WaA|LU
z#3d^@Trd6(xa=tHhs$(&);7N%pJ~e5Aj$#2*y+br`Z&2n+oU_E`OkEga+gggm@#Pd
z0i{bNo?4JksfL=T8Z}VZ$0nt;4d3x>cM4ne!1=up2^Tw%OaESl*k-nrkWg22ixzv;
z-p3_7YUPCC<gI6-TERBUm@+Zj?SqVefN+WvB-`yfxMU_y5(b|I=?6`|+3i2q*1=Du
z*5aq8r%T|;J+JU+nQjI|e%jC6;JhF`>SrMjfl`gU%(mm!>nN_@wHqd%L{aT^o3fdH
z>hQfAs+k8;qnvQ=7bHTmJ6^QQ=rhv>;ey-5%V4-hm4sH19C3VQrf;-B&>Qs(#kiCA
z@_P3_X<y|vmr<xT6aq0emvL_k6PGL|H@klmJ$?H1<;Mq3q8B-(V%qr0SYRkkmNUo&
zI1>UT=kbm9;R!FKrSVrS@k%s5T%B0Nz<4IS8)*ZCnRtK3yCDiID0;z!u`>#W_=_Y<
zpa>UwHbvrA2hR)VlQ7}`=TiVzqe>`CR4tS1Nobj;&k4pEBnVuq5D}=vEVK5+OH_Xd
z<7k>4%xzTQ8O$u=jCDogEJadl;#G~3d0t4DG$(VvkTJ6vE_HsO$FHYH;4aU%MG8(d
zqV&;hGbb{VaVPi96dc(;HQ2fA+7Y=cX*Sp;U<49^d&(A-q=$aa8dp7tX4WG<MXEvH
zvYfW%bdByovSK-vB|Rd$PD^?bH2{Amoi8DRd?!c+=ZsXMO`b8MU-0a0l;zT`lyl6(
zo9J!nsr9xr-_w{YAs`#HMOw(><8pXWnmG%Tk(?>3Nyjr^2}Yn9I6I8S09a6HEoG{;
zDhlqYdUur7ZELEtCEaEylV~H9W0Wj}NpTI@ry!iyJzA>OZ#BGSbT}|$4x@k5jSlF<
z(M)QGYmYO)%PFm<e{lZ%?U&E%0b47k^e7Kk3(+F_6#2T$en1gKpF8=FqJV-jugyUe
z`qAG5g@J=Iv?j^3G9n>8hZPd9CjQISh@TRVcpXSqF$wl4NZsh~0#eUm4I~A$6d}1u
zo3x@!nGvuAQkNxP2bi`@D|desux|8s0j%e+3fOLCaVuaSuG%dMMrlYZ#SDhHESPJ;
ztV=;MUXs!+Ep`ZD7u;@9vR~_EIRqL`6$@N)#I}pMDKjmKbGm=bf|M*qqQtkR%+EO2
z?6Iowc81Uzv$z<<&jm{I`%V2N*0{or`ll+!=HgY{%^xL})_^g-XLo<x>RT=vU9oub
zcCC6OPl7Z|$r0LQ-Q*YgB;`|6|E9N8qolfJr5YlS3KYast!=0#U#i!@yX97?Qc0D%
z?rLyTye&V)2B=xjwUQx_T^UIkH%jNoTNB0ShPx<lvrL7%-9s!&J3}nZ)2%F<h<(Be
zT|rxOr#WrSoisW~^Pqp86{*cWtBh)_BGkDAsMXE}oas6nN@LIl!4jdOg;#3nSfR^i
z=C$oqs{bD9TMl?~t~J^vQDMASFj3qiOfbT<7o0g9TK)<{D=0HFLN;P)4N4Mi8YQr0
zA@5~CFb+ewYa20rCeV4;@EyrxXdMQ==2LqTCic;?7^6fBf{B0DB)Hne4*gx{Pj=tm
zlb}XUo_QnDLTv_+z_jS(eE0QznqlqY4245YKjRm~=KUrc1xzX3&-!nh4{+?Z+>DWU
z|0u0}_ro?nEU<(dr2O~KzqUDA!K87ws%AA4pl(vlh1FyiRHIpLGR#}$bUzE26~R_m
zBc#p5ck=yaHQ|3eqA)sPQ%z6?&V4&d8dhPreIA}Zy*&Iq3y2EulEd`dw}<cpU}`qj
zPXh1S*y*3&r>_q`jbwA_&N$9Ve>6|grhD^+wpKky)i`#dx?f<1X<W%59fDD|R^ema
zqsd*W9D0VrA^ukx=Qm%PjzDA0q5O=?r3QncGOS@N2FZW9+!(%u90@AkbVBD^E`{gZ
zSm%Nu774!D9>OqScqz6_c#=dj(2*B1?mphTU~?4Ss~f;xg&_mmQ_PzN41!SkPn5NL
z)0)TtbJ%+hlq6qn@oZ?o%s6_6)eZc~pgln$mf8ETa|AQ?H}9>WythLo0b`MxDKB7A
zRvE&~%tC)WBZRd;{S(=CZ(4-Nd$77>m<Q&V2e@7fLKI&#g53~a_S{_sXA$;}?mkd_
zi`#``6)s1LYko&=Kk~oZzC(<E;;BE*c!i4=L5;+Fslvk6Dy*pHL_A{~W^mekWtgVO
z=u#)%-J2~p$g*g`l&@R!xk0<!67UBi>eHL7FS&nW0t<+N>WcLxw>WAp!(FVOtiEuV
z^$P2U4gwqNnOe)o3Q8zu^-Qf;G?U18@B*w_<EH=Hyg-3v*4ZUDh}{Jb0j!bt2F&7;
zp2@dz(yOe+(rkus<bUm(-W;;93<tPJ_?Kv0S8FJd!gpt`^J~%qb+=2Ifly7zQk4{c
z(fNOuA!#gEs_5y;;cJWr*i!`G!N{AGUixKg1DGmj=@oO*wmw_7b#$rgW;?fzX>W~M
z`o=&S!$8VfQk*nAEuPjtqB9;ij4nI1E1G1!;BbAI`&6>Y<T7t`)wGg##<HMl@k4~I
z^OmL)UTKzmXs`UB(DSuyE`;ViVp@+2Ysi1SwRMt0#0WmLPEs<Di+Q_q8t(@hnggAb
z6qbR2Zg)%96V~1WukWa+TpzeCoWdSVtOFsrDfUsvSv^L%B}Y4jJ-iqNOee+5dD<Ih
z(K&Ba>448KJJR0fRVb~jIVMKgoNCSt%10T!afc5I>Q?8<*F{DJ*!W4>#bf<)rgDFm
z0}+ekgt|%fV;0!k-UP!?k%4jZ!N}ger7-Ajrp96V5JBfN*9(c=pVs9MNb<HZZAu?J
zr4uN0{HcMd9J93E05dP>j^EM;8JpS00m8G$@Xd=o+v`*|%=Z#SC~?#%%>1((J8Tuz
z&w`pyc=BgW+n>$|Obk-59bn((XqrzBzk5#2VH4%?>@E5D;xc@OWY!PsUr|Sm9_JlX
zwHiH8HqAc8bWAq&_Z_O0-pQ&H(Wz&cvRA|IEVr;P%^mczA1zFVVgZYh{{b{urhu1G
zs5TS=F*lb1{WcSqcN#D#mwfOn2$nB5i+@*|Pk+7q{-BHKS_K<#<5&CZ<IBUp5B)-r
zl(C_+#wpOy{r&ABzF;uQy3ir20Nj25r~C5oM@zNxYQ7U}9t1XRtuYhW&?;kmXDumm
zuEIvLP?iO<EW`+82IZ1PlYDb1?>Rr?Q@-bTf6nua%+msd2~QEx%R@6xibbOS<A3|t
zu6gqLQx6d`C`Q)`4Ikpwpe{&ir6UbAnd;<;P<?j}dI=ijpaR+JpqX%80~?)DY_X+i
zK5hew^2~^~u`8$vxDjK0x*0uUr$kq0=$&Wi^)sv%C4qJZ78y-N!W6m7?=>ZgR%Cox
zk4*{^U&m#qAyAgIq}r?IU#$k6n18?NH!IS8j$%Xub&=EZ_F0M$=qK#Qa0WjLn#Q|G
z!9HgFAH-jjy(qv;N+eXqV=HB(n5^YFZRlI8lN-|U4oC8An9v$Fq`gA0EmwV(6@`TU
z`-)P+0+Y%5iz^}4;%=#>xmQwe6<IgK=agoa9c1qJ(w6-xN|-pf{%!JAMSqvjc)?vv
zK`u{*vWd)|e*Uz&V&>?1X9{++-8P<L83?0UfSAEsAgn+n+Wm3|RgWBIh7se3@ApX;
ze2Q36T!6)iCb+cIOBMx54ACR#cd6_EG&`_{Dp*#|woG=a5$d4)jHWI$Prcx_8f`(3
zN0_F#AK^)+PKEa=&q{RM5P#~Rn!QBQ3(iSMDOpGFKZ_qv)bxmpUvhxk8AG(5AmOFS
z7{G%f+{hSmKujN|#E;37&n1gk4wW^FFp{%a#9m!L6#G06+lbHv=*l784UFrPQQ!h2
z-hsl5%V~hGh8WSoYXUL|Vt_3{(xV-I&Ph)q?go4Aci&|9UE^Lfi+}9C69rpcc%&KW
zV2#(?cyEGDMmfzlfV~Ps1$IiY`l=gpi)hZCT}w$q*-~BsPVxz3+tD+uZs1=wbV?9l
znSIcb&hU)=oi6>R1U5QjxKhrVh&H439efQdnC&xgW0JQJ^oqr!wG=<xPtl@e=yMs(
zGBj3Pp2Y1InyNYSYJd1hd?!!Fnp3R`B_Itx8ca+Jhaj0Yy!_67pxXOdwwtp1T^=PW
z#wE4PN308ngxPl701HWG6fFu!4h?p0qFchPOBOR;;9;}&NT*SG^oL7l7NpznxP?W`
zB#-mk4Z(1?+ny5I@Y>VC`<?cb-iFU^m>*QGppW@9ns0CRu755l2Q7+?ckKszQo%vH
zcRHWT%DL}8IY=9YOsyQt9+V_k=HOT5*ntO$b-F9zAaURtc|i=;yZkv@37)1>43^}u
zf<iRjyZaqar7rO=gqk^qcy2ky`6NBe-Z7HhFDQGZiC#-kbDW?gj?VE<fT(-}m|n4-
zM~ZbmK4$2cxqrDMgT|EN9{(*S1{zz6YaYCYkLl>-vVC#iG~9EoQk<H^%Z{sV*?p%v
z+WoGA0x1+^s{9uV_=DznO<o?$nj}Ap<z;~)^ILjzp~BRk<K;NYBV_Cd9n&8(7qKl(
z5E3I|O~)K)M@H9&*c>KnJ(tUrQ0pgoD2+%OeYs*Id4IetyAk~T+16SO2}{F~>8TV-
z{GegC7D~`MWyHrcg%TrY2#GyMw^m47<#4mLavh50JmF$upEjYK0(+(`OLV5gAd4B*
z`z~46{_`vZR#x)whU?J{8BGevwOqy;Ea{<!>uNtJ2R+uIrKZozczVf@br{=ZEd(YX
zT*DDgEq|$~{cTw047cI9faNg5EAOeDK`S!et7r`{MY~blxi_POcg<ZVJ%w>$Y()3|
zJ`a+Ym^0`AU#>DsCThj4G_&@Z;zkMDsi2|dSUVXAm^&F4yG4kKU?%mAy0<E!Gjaei
zGwQXz&$+mP&=)gl@Vu4hXf361o_{sXh}ABViGR^?wwrf&tF=|K_R8^`gLy)%oZi<}
zmVU*24fV<Wy~CrrP)2;93A|U6al40kwI#+p*wAF}%wg*VVx=V)YPhT(__?6YrFmmU
znNMlnT-ko!SUz$rIYBGfm7O3IwIs5y)d>O+r|f5)ARkMs|4SnXyf_zl?M6_`Wp^W}
zA%FOjG=jjA3c(zBHf$*FXaqHsHcE=k=>0XthTuj~v1NZ>o#FOoP;T$EvaZw(a|@>#
ztYc~9<8c}#&P}@`s|^js1F8)T#f@r1%YWw~tl#E6e=o~Dz(%rrPcx+v)=;}g9UUkM
zPcwX18gS(=Z7tu^xQ@MJw`HmdK3uumwSV&6?)J`;&>bj}a}tT?DB4_20a}(nIwG2*
zIjO`aY)`xEmq4HZ{P8uS+>YE)-G~WVN}sVmhNRkUvWD`v$J(9ikvT;E15V6X{_#4I
ze?j<O5mrkMfLb!k@H#Em0HHu$zobI&IMbweXO=^T91NYPND~>H<%v+g{`My4y3ET{
zoJ=o)>6BmI{Fs`mXT`9mq$>FgA^+ftDqi!~)0Ojzq_)_u<JtmiQ}mQG$A7(BleCvn
zs5TS<F_!_-0Th?lHa9naM3H&`iveaPy8(J@u(zU@7EMk$G^hUjQZL(*?0Gud7Fh4h
z*b+&7e25e+hp&SiK0O>>A0A&c4pul)7KYa^gW?C0|0~Hi4GmT)VYC`vzYV`WNePCp
z179em)_ht1?)Z~-AGC5jLYaK2-3N>6X&MX1GDcIH1-{JApDJ^Ia6HmFD+9elre}A*
zz5e_lv*=QI6TXIz_Sf_4!(R{MLO=^LelUOn0Ea)mJ%k6SNEqt}B4NE6zW+IVdHAKJ
zT6i^|5jLk7Hfan|1sfPa<OXB3(AE~%G%pl+fyfIXhKztOO3<Ls*OBk3G%f%`F8|JR
z;&a#qM8j=*PU!Q0Qz;J-n^6elC0iN{fk5So5Rr&WjQLNJNr(#MY2&}|Ux((-=T8%5
zgTkS0DXWj+YV^aHOCbXdphUI0!&Dz%yj%p)GI(TKF3TWRTvyfNop4Tfh&*4f8r`B8
z^$^t(R1wwjvaAkN0!`r^`9aoYQ8DJ<_BQP}6@9j5w6$`7ykyJ9hRGee0ro`SkgJTl
zL9=+6mV)FMm?Q@mBO%P)vlxd-N$A28RZ-2*E;;*ZK+JK3JPFA|%EE2o7J{1%TzvYa
zo;o-T27SmKm@+OH*-P^E2%jxf5S2b)F}o5taWyW4dvR5dF9rtBYXXB6%pB>Q$4dTp
z`@i3(IUK51mgZO`z&3M!dVi7)+aPFLPx(DhSw<Ed2~yA^%X%+}RvO(wMAmgyJ!(qx
zN`urvdQ<2fWNVJg3s6V`ovmh1IIY-*Q9_|>I3~@QqC4s>C1_A*x2Kn2IX5JK5Uecw
z2at0R!VF(dlTMMMyZd=`?hYBP6$cgDB6DBRZS=UT6O=}9P-!`ru)1b5>?OCG%ryCo
z$t&K67{;32%wFs@noEq9@dv4elO3bW_MXQV7V%UlWaYXs#{rvTlTI-$tR^rjHXQX=
z2%m|u3WWl9Ipr~G9#h<nc4KFMBi$U?WLXOyax$<f$FgB1+q?bgzy>@;L%W>h?mK{|
zN_5PknrZt83og)w<xC;5Xe(_P%Z@&3nRdybMxImXDOOOxir@DmmrMa&+Ki%X#}aTx
zTPgf-9-gZp=W?zj=i6M7I&#=&s<sOCKG0I?B5Oo%gmyF<O`vedQpuTr@wlbU|9t%b
zo>NET+WJB9ZQ#$-1w=}$;y=nsTy^JjE-F@1uAK6qMIk4L!;d*}C@#S(1=S>ag+u%{
zF)nxJ*&{K2+IL)z6MiL3lf5kKc82Y~C8%fnT2Qry$<X0A8P=G;`I&Q!xm=`RO3oK(
z20C0~Cgq;T`z+WTq800Z{c^BZVJN`X6zOnt`PEPYxxA<4GO8KfVRcP`krY}JRBD-h
zE)-h!eoA->kRfGBzB!oy785v(_3bsH0&)!3-QpE4$_m60y3>$kQPlHx7NtY(1Y25E
zO>v6?KbMdV<uwZ{Dy~QC=elAmcBs4nuQ#;An&3G*Y}nsf_m4Jzd3z}h26(J*C7TC9
z7QQ|CTCJm<7T&6s+AY50oB-$)w&puwV{{$cV@g+2w7x^voF_1r=0as_;O*FSAZRmX
zjM3llMrB_iw?{#)mz#p@eCb8`QCgv~Gl_&`Jl!~CY>)lPNbxi2I6B~lepVAa=VuN3
z|B#>6lHKcPHMM(x{H(#3PWoBHc5^?wNZF&GHOl@M{j8DOT;^vDniu(5-hH)a8g?u=
zc!^tP3rtxc*$uOWmZHBo|7<9o$v>OX_s64#;F);TvfsWHBM(ctSLK(_H~6C#miooo
zlXk6x^(bm*WedKFzb3(;bo||E{%D1DorMJ7kVW65hv{E`l%cva*LU#+wmbCY0L2@<
z`@nJ-6~4D^!5gR(SNOVQ{hoY<a@L@HE|FojgLl<udU9Kd6B4W#Z&-r+*Mz`Pv09Ql
zu_W_<Sdv!4fEtb9Ff)RerBjdl!msqgO58I1))emvgT@Obp{2MM2TK^PN5{CAPu$}1
zOF4am*=eAEoSpTVd8;XDH^j<2+#0+S7HK8#SvyK&LAB%Cye8iELl($TBz{XzPBbUR
zqdlVq*r_XQX6Vp(@U)0+eoF#Sw<-7SM_e<H8~3TS*W|5AV>U)is){6V+M2?Yg^tO3
z|1BuUHH+B4jOXQ{y-lHYNymJS>LzQu_8c{sqm}f3M(<&O0$#8?xB1_IsCv5S9So34
zIPWj=x-N!Z9~n3qHsD&W8@~b&no3IBmdma66!7LQeM@cSC5n*Md7QiIQoRx2q=la0
zRZyn%v1Yh{Wc!XuJz=dP+QQ<NUyM-J3-8|k#R!F5h^eW%#*>Xn#k+ZEr!})SbkNhC
zIb=hB*Dd$40jn{#7&dEzb8*3;wd1#;aV#d|8ichS*DIZ74MXj?{PomPFCDxn?}TNJ
zcX{`iUU?Wtl%{O_)iq_8aYe(shjN_^*GJj4RT5hY7a1Lg+)Is|_KhEAe_-^9s>shJ
z_qEM^qw_06cKwy1&8$?0wNfkjJwDl2&oPW^HSa8)3o)S<Hm-y8!p5*tD4a4ssxi#9
zsqWt^)w~%HzDZlyGbqzLjX$U>+%UPjDO}V0pY((+=SUk?gmZFtbc5$)&K$+fhu(h<
zH>Y;y5UyhUwjJO~%9HKi`N*pGdr2|EvEd&JsEh%ZQK&W*0Xdh^wlpP|uy;QUmu7Mb
zCx69SQI8w95q|ftAkdezfVt$%aE2@b15Q$>0UXq@Q=qNkhx1uUE_Ed1PEMTu`hLUJ
z?MarL=p@-E*r(YQmmG3tzWIjSoRV2cA>+G-%-p3GNQ+A~1#YcNITWTYQP5Jm#KBtb
zl7uGX6Zi$6F5<x~<(`Z9E&ERi7Z|2D$bSiUg?ihDfyMghO+wZ*ifB-C@3(2lO}Z-0
z0=Yik0|lbi)zU!B8H1s3VoUvJ4kVdFQxZr;p(WHOX`?E*S!5g^Q!bc=KaPit`t!L4
zs@zwnxmc+`Rs%yzA1bxL?9)J7ER}HY7aFjZt{{}>s$;(jaWyc+hJj1Nkla?|T7Qs4
z+C_q>y(w`JN_{Xw1I}aH22pxjXey;x7jqD8U2`fy8s;adxFp<fD#SWIMWH0}<;*Oh
zV1zCjjGV{8DnuRIwopp#-C`|@<xzzCb?g^@QJVEtU=8YaZEP75k0UU|aqZwF%B^=R
z6@HY|2V)n<`Qiz+RofJ_b?axWm47Aa4TOT_(fSzOIFD^Dn21{%dYP;`kYI&+2a3DK
z%?E>LTkW2dOM(NBhQi;PMne<|m2%gyaHw4Re2av->K98|bu<KAs@kX+Y*3qyoz>9b
z(mq;gp%L}A3iX7$rPSlaMi;gK6}qJ|q|uU^Aojq^MxmMY4h;TVM;&3A)_=I`sMg(o
zoSSPxqn7HA%cy1#hteZ7_r9$3XS>m`1U|Z(vX1I?&8qcjd1vUSuR%2r?r}Y=%Z=`R
z{`v6TZ*OM9!})x%9G<;>x%8{Y7xVvy;fuxPtJ&rA1cXU1hQANr3|~Ia<Hgh2`SQ7x
z91|2Ojbeot6yn$b1DYrbwSO1G!+`Fe4PP(5TMS=?&$1G*Lx_cnQe(vIpC90JoD#~h
zA^?g(355<6p$EJh->GcrATSA=B~z?Ks|A65ybg$lXzG#JVuH((qzHWwgSek*dm#*Q
z2@h(msRDsCSw3D1EN29WO<FLtP(<s8v>+8A4HPJ~l*Zih36x;XQGemQ*Gd4uWBnZ^
zuJF*sV*Z!ds-0kTX|aQ2i*RxvwnRpjDH(%{?r_LfGk}9DG6`FS<Lpvm_qd3okguIE
zN$gl{771Y~@Iqj)RE1CKi+dZ>7#>c&_qju%e?I-j|9|%5a{1=3!|?0uWjz1o-DNyq
z{5(JyUM<cCD0c7mntvfT^mq`3DnMW276qwaOd<?^JJqgSw_ksaKVQr*{&z7ykBiG6
zT(|qL&z7^R<uIGCY5CQ2rw$6mo(})|_kV!<!I0zTZ(qNDvAYGp49kTaQ!N&t)lJhd
ze7ShNxP12J?0gp39|*&fv*mJjIS({m{_WlJ>t{>o5%PHB(SKsT#2}B@5fixUqdaoz
z<0-LUcpf6@-930PJh@z)KbtL|4^O^&G<-LEw|ud-+E31Ym_dj6ayDOHb>Q!#V2!U9
zZ!gbhS1TIg_r9IIx;Xn{@va+1k>RYt@1LAq;#qG79lTjL;MIyrbrtSCAGwsPKT@e{
zsZ5eik`uqDFMp22Es4mb)Jp7x#lr>ch@w{63Kql%5+KMUGyWL^ISz`I><Eo4K?Q&f
zR)mZl8-j&Oi%j+Q9{Iz?^2fI?d#wEQ_UAYEU(J3QzCBz1_>VWsZ(y*?@8LA@r`N;F
z*Nd0KFF8_*dAPbfA1>yvU}p4vg`fZQ*zP|j+7)HDU4PYzi8y!i%mnu2%ZrJHY#b4W
z+bASXh8SQB90qqSNKPlnix|!6$V!P~EjnxXkPa-R;!|n^VlWq^=%cI)P6rM^lXXGH
z@Xeply5GN@_dX}%X4(WNFJ^9_rz7<)gxYL&jZm^rvyv?`0(y_80?Y#;+N&vSh!SFt
zmIQ((;D7gfG|i&O*QdhV55eqvXzbNaN%bHA&awr-UU?k^z`3pia9u%cPB&7!mU5b;
zyaRyu4-jPo=*kozDu~zHa)lEHNOt&>!$`*UoKU!9PCCtrO))3?gmb!@EywAgjPePg
zR6GHqlUi&jEU6UaOJ~*v^X>aA5d>xnO;st^s(+|(7&S1mY^}$oN|3f*-d?9lAHSBn
zOO;fxw(U4)$&{mceSe)HsDiue-A#(?{l~)Z<%rCGKw#HfT!SeWlmP&heVQWd(~fMi
zqbZ6L)N}i^%$Oqc$KcN!SiT0|8<2l%ET<h<P6uK+*$tNaPTLBtX~Jzf4!2P}#WmQL
z27hmoe0RKt??0|5!MIwZpoDb{2fYuBg3<SBQ5%EE<t>bBr3t7iYD#Ubf|HbV9AV5F
zCEFSt;5*m5%ZE9EGIDzl4~X~TjKn(_9z_86K238*ULpGE0lY!eHlkU<+}jv4xX?f}
z(w*Uv?;N&#Se$d&j&tV2;+)sxoF=A5Q-8$s6!DxViH?hX<YC`yoPr;unNzttA1%nk
zxmSz#j@~&#*dzSGn<B&dC>x^hc~e8Nu}yG<Yf{kFR((G;mlG7MsD7e(ngj4SFObA2
zo<O>+V$s`y0Cd}`S!(nH4vZ<RDBu8o3D*Ee0Oa;=a^>Xb0ltGv#0iR=U^m&Mmw$pP
zQ7iZL@F?-SS#M*C4lgW!L}1ohOjyO3e`+5cfHq+Ub4{z#^mm}iz<<*i{U3%V{8GH@
zd=O5FZsU~5VK^mgwWS}GT_e$iQ!?R{OgJSpNqiJyIlc7Lv8f`H7PZiwAr_^K%0cB6
z+ic_JWkzjQU@#_xiJzC5@43_+dw=~Z)Qk;_+bW_4u6Bx(pHT@X+zrdiD9n5(l@oIP
zjuS9bM5rlgLTDEO;mM~#PDZ)WEGBpkI*qv>MQc>%9eaHRh$Y&J0y>}%Q0xS!Pk_kc
zXSNOb!&k!Vnov==rtSOzf@6|3wFCvVz#b=Ahnix+I;>RiBd3zehjbuV!GDlAl3)c?
zYhLq*j*C+EqIw(6u8TT~DVL7{pPPtzgK_U6<^~jQAm#>^Zh*1BOMzs%x)8sxgWywX
z8$LB12A{fiHsWX3D_ATORF(-U3r~_vQkf*3B%7oOj!QeiX$mkFm48qWmx8VdEFSTN
zY4#)0PedM8N|Xv$^s^lhnt$3jlvY4&wu1l>U{swraFS~I6h{w}4u(9Ck)f5x_XH51
zRCb_2G8KQ9f-74EE>ho!Je3v&g(cr80a{gL`yd)dNMcr+|7GL~I|V4V6vd$^GcsZh
zLJ}+WmlP^*AF)-q4#2rJIE0@@i=Kj4r`mIpZ(InXg$pP_^7o3cAb+wWJqYZ^SPYZ^
zl_|kN^KrYEjiLdIrrpcthxB(xK`JN+i=1R%xQ%-j$T-=_mM38D)q?HVGTCoi_&^4d
zT@{dFlx?mN^)=qIEr~hzC;&zVVX2ndQMLu;AYR+F0rVk?@65K`^!{TMZ0~K_X#S%b
zXsyMx-3p^NqrIAHiGQfS_G#)cT=S>h+1cO>dohDsr+nk>lyCX4lyAD0@^wnvI;Cx$
z(zZ@jl%?YnIOuq6C?OPlTK$!>1YQkhpo~hK4^7~=DAtgvxZVIIE7C-}vtW|c{6NG`
z0F)(RhW-*R>LW1B(pulr@N=xZ<E1cUU&Ue|2`1PBb<$3hx_>QDq9SpCl;~#*h?H0X
zl;n5w_IrY`0hF!<#zlqHUy3-&wn)BwQ--4?srl{>4KOKK*4~FnqrD2Ajb_&eUT(;J
zT2@p>cXu*%!Xl^cK)kva;^oyVZ#Rfn*Y~n*!a198&L*6zC%`%86#YeeutMOP@3b1e
zt-m9bbjQ{SP!hPd#K6JoRf50$)`Q{yWcqqFmr<xT6qgZ|Iu)0AN;ks+{<l0!H!=zY
zIZ3&Mmr<xT6aq3ZmmzvD6qlS;H@bhGJA8iodB+p$oH^|u{jL3Xe%t-MD}qSHP`gfZ
z%cO<w_isD@0t7N`9LdKFiTC}F?#u3%mTDObRoDM~|JpSzza}WHASk)kh!vDlGep~I
z?U?ef0+bF(cwszHbg9E$a}O3j`zPrI$KgBQLKt!-J&^H@A__p@BK3q*d<K92e4)$6
zen|bm#Q#W8at|l<<EM!c;khGnZs4#Nk^>L&Hws+`d4wohrbtyoMC}5i`hQ1MKO0g0
z@m89L;FEJ?F_7IvIbawlla-^7hdpQ_?4K_UbxxngALNmG)k4(YUmRa_E{P0<%W(<_
z7PLv^1c4OQsI%^hnknU4iameGkqQIPT>c3>S1m(C$ngc2Ek#lDQ+gXkVM4dPQAUZ7
zG={2w2AQ<zHOiAVNwcU^n~eXxcEJTRjr`xCQ2`W63;VbfkbA?iTv=)+k*ub>c7#M6
zZtBgiMKR-)lp;Brq1OjN3pn1&b+GbDXj4)nVHo$sd({Si@h~2vngf5trgs#~j)D=!
z;8Qxb6`!-Q4NdT|z!b);mbH-WF+AJeO?@WBq`639sJ$S}T&9-JF#yvD^?K1h+Zrz(
zv4h|lIz_o6&&dpKOryjQvqH9MR7&OqtfbNAenrYeNiS{hgB8e+ARHBBCR`-P%~x+(
zH@mt7!h>XL7#AP?)%br4p_xLQ#7MPjSqOR(feCjVI#<O!X&f`cv=s47RCO|RNZ7(X
zm<PPahRr!O^&x4>YGw=6WRT>H%t<I9Bp(Kw%^zL-H{>8eqrF!Mevmr%OG1X!;V)CD
ztcsh9TUd7GL=SS*;*qluZ)6JF^~gTJyFwlE6Hco;zl~o-TqS>6E4{7KWmOLMqt=Dk
ztF)yc10%FIF^(ZledKJj^|Tp)F-&ONGA44&3=TAJJci<yE3=yk=U+8HL`JHZGCwKg
zyqY5InM9{Gg%xaR-@={M%o#JHpnJ6hxS>}KQ6!E_Sgd83RC9g{wl7y>1(HlUjWVE|
zOgT{<k8_2Sr7?dOoh;5im31HDys{+g<2=uq-@*$|Bj{mkD5Gtbhn#UM2Lq>=H6xvG
z>tHbopKfw6nzk@>BSYu0gde#BC@OVTF*wYHYKeQNVKl?ci!CiG3816ll&H7pFitQD
zwjF)0T<~elFG>W3pL`BlF0hEI1l>|61agF~a*kazuS$Qaq}Z4h?7+OT)Ps^=CQ8s7
zdu13tYn_Iai=(@~a?Mq2%q}Hc;X#(L%ZGU<npBfh^1o504YXwDCj|jd<#k}7ucJz+
zXi*`#0Oc5AnQ?iY<j8TlH&t3-P1Bq7I?AvjJ8?!k+oPS#`t#EE_4A@NN&rTDw4C|W
zATF^MRM>yJDqODe!-5X^qeZwDovOD-XRClE&`(oae(qS$sH@^DQBXd?33Vk3b?;}O
z(2Ot_wtdtIvqx8l?D^?mtcS#XAQ`HO%8YOoq^kE*K&nTW&21^{g?5unsR?_lv$^{K
z^Es|8cCP|f^?nLq^$6P)18ac2ztH4+q3TtNGb4X{TS+U3VKAP8<Yf>jmuqMxBn8So
zP9B7IApn<DloC`37%KkHRa~Ah^<mx2C#SU)7A-qR(@&|P3g{9TiXoftzP!W_sK%c9
zfZ(BmVfevs1kB6gAPSgLSbg;G@+iP5&)9Yt@v@i3y!+AL?7L6PgCaG0LmQCz=g+^E
zhc$nc-qfwq(KOA8TZFTUX%1uI(nq%QGaO~>WejT=7;jK<0u{cJ%ZyR!=%5DihnGD~
z!B!Yll=rx)zWe;LWbKLrIlgJEy{{r6Re8BOx^-nv2uD^r+TapaFeTSExYifo!)o2~
z8a%}qCJfxL2H%emp8A1v%WxGst;@Hgoce#5Qwn@@-AX?tZ6RBNkF5shJ)k&;8SnZe
z-9EnIA+@EPVm93HXej2&?52$lGI201CCL{zI-1^(Y_ueW)&%GFSxD`Q)ZPw%*M{63
z3&j*u)JS*%lf0${-=^~eTRB99NoU8HKe{U!DvQk6hWmSxN;GA+wDy|d>;1-F5n6w<
z^{Vp*re4uov-EoKXI7JU`9uMU>Mf9BxOoCIl2LN=8+y$bX>sb2EtbNIP*lndW<`0~
z=%x%)6H(`fR4=(=f(nU->WuOwH%z%P56|7!xkLMkI|dYzTCt`Vixl-TI1;PudW%i0
zfC}=kl=GMsC@{=eGi3&0*2L3Mj>~`Tn0YG|BmM?QhbFNjtK??HPvE0j*3H4vF9Hvi
zKJQOFeyJ9ruinzOYo)j2JqhWMw;X0q%0!4$ebz(tlyv2fd}z64b=FYa;5bS{apgEl
zJN{-{@e>;p$p+4Gxwuza?=AP+A6llhEvJ&yVK1E3#y^{f&pZk5>D&zuq-}p=;me1%
zuCVY;j&up7Dd-NR<L7(*btuXqE4P*=%$_znDkvn<tXm9EqZPeht~kiFwuWWBca$?p
z#S2Vz5@Ebgw7SEyx}&4p(*-lTI*U6NG_UaMjkOSI0yw)ZW{@Fek>;Zy$?|9xnWG?A
zf0!*Pn8E3=x{{(>cFT#T?N1V%89`m)h688q{rtB356Cz#DVI^GHWUFemq73=6qf>C
zH@<(D{qL{8Jjn$8<B8nl1bJ)p8g64zFuIIe^9jfji1Ct%Cw)9QwWYZs5(bOrtbjzE
zPe8_`fq&QY-*DtlmA(s7SReHBdQ6g)nk|785QMe+V4S!#EeqQ#&;)Y*dH_LKPy@)9
zlL;oa0WKd;7z7+=oHK^mhQga*#_;iZ`0Ib?*QbA;iYY(~GJZ133oz3k-=4yIR4mi_
z2|##-)Av88uTQ^@ho);9KO^I!-KAmdo!!9*;!8C^g0v}&>uh$>VCyi>vOFnBUJ7#R
zI+@wT@mp7|8&<{D#p0RofeRG)QoMvJ>ih;~nfPh`ChJG0tZ;Q7)_q|O6iIGoV%&cn
zPP844^fq#J)UDw#nZlLmXIqxD%mQ2ej1mT%@y1iB!Hf~Jtlk)WDI+bOdEvI0^x9&%
zEx*^aYb|U$3wJgQl6+^odRv=O!7*jkLbkqu*rQ-q9e2@7vP{%}Q!XeiG;?`4A?`Mb
zF~@pqy+&;2!|WlFW3Q1h8q4s6zhHmYM5$A@*rq;gT1KNsH1breeoXgJI2Z0ZGq<um
zVwMT#DczK;{0(}jPhG?>Hq|5?%O(v;{z1}^8_wJzl0avhoe~oJsfp6XOZ2Qyuwlz4
z^=W>RU3<xAc#ttGBaC#n;_h~@Vh^|i-uadTzJ&iU;67zO1Pxa-y5xiCwE%yu+x!~+
zot7Mo-Y0GKzD+;lctU8$^WT^S&ZbSyG2Y;KdXD5VS06Ik*|sUG067PoR*)a3Nl)KH
zl3Bd{a{d;9JSRQ4Ni$QuG@_zq-n+m+jY>E!6b9+>G!<j>z4X*UPyJ>!{gC$HCN(jr
zDHrslT)c%6RK~ERW3m9qNi%<$nNmQFfz{>J5vfg24&29@Oy|pq2p5jxlmJ_tHpo=F
zHc7@ZaL$wUE38$<*<5*bjz*)oV58!zP35|x#C>u!q7l-TIIE9)rD4}52Rq(NNzmuP
zr9^ei`{y*We^$F6N+rmptnv0tSiXl-<PhFVnhok2?Vu|SI}}%@803HUMN9JtT{=k@
zbV+|N3$CuA+aQ#7TKL+ChE|avw_KUDU7<a{$t_uqFQ>|}QnX)3X>E+qP&`?`%#Jbh
zZ2t^oa*5qKGB#WeJbo!FG5;3dhf~UTt=&SZ2IF_t6@$J&c4#B&&KzIK)s`i#ABwRm
zf3{_(1vE`<X%O;BIdOl`@qUTJ04EN;u(Pj8FGl8+NwTl!=DXbV;krx_+RRZ_;?BJ^
zLiS0*g%r554ZBe@l1`v8Yr3sBn9X2-$)O!<rpY+NpT<j=^<MpED@tUqK_@IwS873G
zA?5MNdUQ^M7!9Ft`Y+}%m&>lRg{D<PI_$B(*LVgz#DUCX<}-ibo$y5Wu3vAO+W=QK
z-uuVSZAzcBHDQ>^!TH#;M!+pc*9lmmjPB5X`0~>xpA53+xl!SF^jqKpE({)S_uJ5H
z+1hG9%5kr2BrZ8_5B|M8SNU);4{}`PjWA&E-I3koxXLJ2RsByLw?$tZm){%Vt#-8L
zxXO?KH@nk^zqfyV-2!)vcC=BilWl!cHiJ6@Ysj_q0(Ecq3otaZ5!K?VuIxD#A~VrF
zbS@l-j;7B{r_z|84R~CnS--yf$KB)XB`d>J8ii5`GdB;afj~H1oP1$W%_S*c*TM+P
z;jLXA-lh)xZC$sLT1fwZMzs@EXE-fd(@q<%8|ga7-HLx24sRym`9CqII$&yoP1DMy
zkaZbWJ3BXBU@QDvn;_-JZ6SG!MjAtA8g18&!PV8aJllgwj1z=bh|&kBtvcFG3OUR5
zjC%luIze1e)uuk8Y(|3lu9aGjKhTy&`x=S?x2O)?AUw_kwBDc?C>W;&O6@xFd5x~5
zDZ#d#8q8w*sL7Wv>2mVaqsR*1gnvY)H)q0Q#aHGDD`;VyZPXog|Cmmie`VDu)FFgT
z%A@C(=m&GdsTtn5+P{?Sk_|#NM`YJ%?)q%AZk+%%$b84LK9nPXZ>Z|WPp)JC1?M^P
zW|vW@HWUIfGnY|K1{0USWH-D4u(y3>H_rtFl{fyE!FdZ412Hf%m0>6om&$xNDSy>k
z%dXov4Bh7|`~y`*QBU9^K$Ck@V0Kg7-C$OOwyQ3t|GyaOX^VE^JlqtsaO}tu#m6B<
zGLgggK@OjHhxgsb6Y9YUZ@{VH{o4TaLeeKl8nk9`0Ky^+?>~mWUZhl(t^-|KN<_LG
zJ$t&F{Rae3Gr*P$_8**vmvOE-)_*x@@+uYT>T-v&gQuC!I~nO63cch0@Ap4;val{i
zu<<e8__w$B-M_mgh(cj(7_4yuBn<!j*u@tbjIb^YN{Ik^`1y7Cw)>M)t$VS1Qkw@-
zo3z%LPHoBxWBgz(3UscjO_Q1o38+aG`b3wa?<AC_G2_bjIOj+g=5a}aqkqjp3r*zP
z$ItHr8~O5ils8D)S#u8HHC{}Gq!uY;q@hfv+Kmv^kEb9P<qTeUk9-&GWaxG~r!hr_
z2%L_2yIF3fqyQq#<vz{i9#?o_?Mzb<uG(b#B%7FGxAGk)cB4lr3osX<bJo2oDwPuo
zu<6h2wSF1tN6Z+rPflYXI)4Se6vP~oh9@Hvz0+u<VhB|_)Hyu{+ngp%qt#aR@8k2C
zIdiKQYtQ6XL6gqiOm16NEo)1KATVgYN}if*x^kEJF?$7PwV20slQ-GA+E5wqRm2uu
z9t~#9%|^d`J}V(w!n3oS*UJ%rdCIaDv_#<oa>_B#nh-;!I&!OwD1Xt52uSkVY|pHj
zHuu=s8_FCvrKLUKK`rX^*rb{vp8TEk**ty4R*462)y7bd5MAcELPHf)(sNK*D;ZI=
z)dl}E!3m^LSA9QHJ(Rs<LG)>iWw#zW+6Lhz@<RvcZpaHRpE@}Xrw>ztUpMlloJ40L
zbUBHP)55j$TfY_voPV;BqM!=cu~0WiPvgFIWO~>bD-Z+U<Tz?C+hhUbXU<P=1t><3
z$YhHYM58LSUB-OEiL021J}^e?h-fQe5+kfsg=VCqs^_mOdsWj$k$<WiRDD(pRjob6
z8P7Ll1q4Kpic<twJQ=~!iD>LhqI<KL`}tOyJcg8X9p6d^5r1&TcT6w!gzw06JGCjA
zj~|JbZe*&%YG(~}3g9;AjIz+0*MwAfDbA7<yX2LR0Tf6*V73HkPjL%?_7v9u$|U-Y
z0KJOu22iHA!ee(7y8<Os58M2#y5ssWAWJU-X>m&++Xm#@h1neN+S5yd>2`o;6gL1o
zqqqWiF46A;{C{<PSHQE`HP31pLIHDw*m0C;L%?fih4fqET_pqTYkraVy<4_R4URTS
z=0;D$ga&U@#ZDl~(l5QaP^q*e6}w<$Do<5xJL8Qvi8O^X)8{FVDs~z@=qQ9*(CE7S
za=}QYz0r)XPp4K(Bo6(^VX9}{h~iuGeDsRK!?T)8=wQ9DBaVAjaXn>+=cVQ?fyy~y
z0r;+4H<WVX&o;s0H>sjXB3{>1B3xpaQ8RDFJ9Qs5Q`btxegeJ!R50l$7%n))gqG&)
zk+HPoRHoQ`VJ*BTb7KE9eK%uq*#I~ueh-)ZZ#PbVMt1;#YMEeuaeIL8@^F8E5&WMt
z2$ZLT*7*(7QATltbd*tCNk_RvSDZwTq@GICNVt;6+TLfg*8}_{avyHz(FNha&3M!~
ztpQx-HBfk(0Yn}j%620w62b}#n*h7~wzeU#dx~2CyQjDYY$nl@fPEE@4cPtc)imS|
zFk35s0)dP3WeSp(tAgeN{i;XNtSRzqwZg+DXeMjWwI!N+id&$$r?`e@Cef47d=-xk
z&He0)dHOv_R#b}2d9uqqRowTpA4M>rP}ZE;dkf@30D*eT1cOoB0J)6f3Uawb&q3~W
zJXYj#9bGK-^7vhe1qdZ)HU@V&7E}Eb9Pm{N4<Z@76zIbW56_oYaW_AI*ZtaMv15m_
zZhmh^jOsLUjpy{kLL}=t>o?6mqV{aHndTq*dzR85v~rB+jhkt;>)vm2&!4l(LC7@x
zc6an*Vqf@kR!Jh62oJ>eTw*Z>&P4WLg(H{hC$Ii7EKGjIb0hF>`okE%lXLwC=U`sb
z3T19&b98cLVQmU!Ze(wlV5l|)0x~t15qu33mq^JzDSui^Z`(!?zWZ0OHyfxoJF~Md
zL=a#*QGniR_?G5WB)v3f8ldRk@9c}aTvDWDD@hL&C6YTcJM+zxlEWW^9Nt~sKV7{#
zC>pGA8Y~V^p9bI?l0QN6MWY7`AdG_H>GSY@CnXv_?)e}omEyzlcgJVFzg6J)2{8G9
z{jDX~)qiL4p**8>T7|x>&S%IR96xEDm4O~}1Lgv1|KaJEE15+f<wPa4@q@7>Sehay
z%R|IxVyc=~>u}cl$X&B!N(RY@BF$F^2a7t;UV!UKBMYlB9dyV9jwsPdL8d1_4?zY#
z+w`4GkR-`D-n$DjVvzQZf8E!ejNkS6#CWw_34izLkg3UrYRd(SCL)cn89%P4^t^r;
zuh@o~%Xdwow7-Sj0<|^WDsKfT>~+I5))Yz-1<Vu*a!bIveYRK;(lqce%ZiBi0tFS-
zyS5=Ml~8kIE3>*G7E2K68AR@~<bGmSY7}$?vc(+?3hA(_HE9WG9k0p?tNX=gtJlei
zdVg01Mri~B^*lG!hO9PEFq1SO(E8$}4@H(JeeemNm6-h_m{$ggy>9*JAp;_<^*MS#
z^3<9Cq}}R0!qIpW>8*ML83A&p`66$W)ie-X){3$yEaJ@gCi!U6OqAVDtaP7uQ=&$_
zTP%B+Xj5lUre7Hq(M1kf11(C5?(Han&wrHl94n)yW`NdG0M<BBY(;rDV9QQjk|iFI
zOXK3v9@|(hPZnBn1=YO(EX$}+Xluqq;rB}K+P%6a`s;!vs@Mv&<zd~yEng(n^IG3I
zQ^pg0827NUamgy@%Ti%T=RpxEX_w`Zgv;@3Og4$6&s>?Kd|1b^BaZU3J!ZdL&wnaP
zkegW|YC&g&ulo>z7Nl{;Tj!0L8bPv0vR!?jg_EUSiifZ}F%u^`B>^QGH4`r)4y7<S
zR%fRm8_6_W@~<lUD)u}lD^ESwjhqc$3*@J}Rp(}vEa@iLYSOGbEuJt&3`kKUJuywT
zDN>xJ7iFwqcbV(uq@=4H2vqw2r+=T6Px<FA)%@|vyCB&z7Nrx$RF83uM)f+bNC_jM
zr|<(~T4+Y}eZv;IgqC7j$<1h{k8Q;_!x{@kU#GDX;u~jp1!}hP%sPQR(v$0{<g6wI
zYf8C63Z;!Rd^TR<@Rj?n<vWk>Hp@}%K>+A!MN+DylKYnXyWEFr3qtFWLVqyWXvF(u
zyl5<&jM!ElkwejtDH*&6M8ZfZR!)M}dWmWbtI2kP-yKXTR=7OSYPVKgHYV{V=p~%q
zb-&&|UHv`tMXhwi$?u=9!i|vBjP(Pe@LX$u{WyHO`n9K;72luZj38Q98WZ->TG#N_
z+@OiK$XT~o_lTpN>aW2evwt6OzF}75N5tUd%40<eas$Q<OfL<*3CAL~Fw8c%W$RXi
zc}2+gr9zANp=u!IYb_A(=eVSrDmHW#M#8QM9e2G3&fcM7`=Nmob{IU9m&-r{b)q`C
z5`EUoC95hWg`KIfI@N^HF=!|b=~}6tN)C+6$U9m<6i{=k=un?F;(sXxzfEJFQqb3?
zv|K64KL(QVYFo@D8WZ4_fyYwg5E~JhORTsG)IBy$(8z(CvejTO*<EeF?F~E~jrq#3
zm3m^|FywO?F3{2$A^+8ZV~G>mTdw^VMv)ZSTB154sO5NUJSn$X<d$2?tx+e1GKAA&
zL3|mYnF(G>d7kXtw154ScNog8Xr!?AO>6hkqAp4ebE!xB9MyG^oCC4+_L9jF@A5r2
zOO7$tO{&H9L`Y#O0iBRdB2;z{I(TxOv^_ubaTCQN=<k&IlsBAn>U4afv^9<K@iy-K
z+#CShEd-NyoZ)cM91uD^#m?JtQO(}+=zA*>AVXAn>?wsIK!3>dv0}}AMbjyqmG_ku
zVAV_`D|w$76b@#)voB0g%7EZru%$OmB4uKbyeWGP=UUHv(9rbVgBHM-TAJ7IZ_*Nj
z%S){A3Z_f`ytxH`{nx)Yw`i|Fe!bz(*M*8+q~eS)&@qOaRase&nr}^({KhC>WfbZ)
z1lO-ZEWBuO6@MsURVUu#&1wMR#@egiI%8;MH(s+TCj|#}rk5MT=*5$%=kvl9N^ZrN
za~G~+d_l7juaYOn51pN4a^*3{up>TRVi<BnX<j|NX(quBX{QKXWGNitx5~I&Y2%gV
zHc0symu6~Q`}&h2N&qNePw-_quN3Q%f(gHH*G>hpaeue<y`^mQjA`&^DEp`yvSK@m
zlgiW8F#U`a(@{pb+@R-F(LFuW-T94NM-gB-nCAr3^LMqI9$$r)-|l%0W=<O?%0>K?
zGD~`y+9#9(S(fJ~Y_wW%eLS<7uPL^82OCVniCr8=CnVKVEy;UKmRb$<BXte!l@?pY
z51b3ZY<~upWVrZRVNp?QTMB|L#i^G$Y?$q`3NGldQ;Pcjd<Qio9NC`IF$c9OG;jUW
zFL2Ntfjfdn{9ud^ogyil)}}UAmshj8DmK1!In*hwUVy|)Fvtdxj+Q_e0k|O%wtwTu
zWkHY<QnyqDyCfD>crRuyiNz_!mtewp0TO$P$A3UkFxl;>0|hm6_1QG8N$ShsNM<;c
zY#!}%IGWqf$O9e)xf6GaTpNdawR`%UPEnp8Arx6*ot??3610!}mr5B%qK1Bzd{{v9
zGu(|!UQX5*9a5x!Hng+2V-Q*L*4TnCiZRYRPL*UkSd3Cm;>BV5DlJS+EmvNEoOj(1
zn}6q=l4)xh05rqLs{_FhIvdM|?;@GyhOXV<_0PF`BU)a<P&P1K^3%-}jmz!P<%gQ#
z-xuWQ1@Ez9la6#z(;n(SE4ZwM-Ff3VbV5@n*SDI^0XJ=-g#JF$xd0unrvu$lPnVm>
z<;>S{yYxhkk!JvUF3*_AfhS>brk&!=?SC8h=T791_==xtWMv|!0s>|-y?7$$xs&vC
z(Sck;GR(N-`Y=s<5hBfW5vyT+kY_nc(nR{}*EhGElCS^y^5x^#)nQ!tL!|;US^Qyr
z5C`1v>rTe4F&w*W7^bz0wx)2NO@f_CPMpGVcWatPXJZK0MGB5Y{{b9dg2W1CZkP3k
zMGylpGB%gtz$FuxaKJYym);6L2Y>rlsJC6Trulv#fq{6FEznyH-;x}PBK9Rg(*SAz
zem_#Av>a)reP|sR=t29C$eH21=M!(98s7YRb@y;}vsJor%p2ik^YGXRJmB~g95=0L
zoDj?^(L8)<-oNHtT3mOyH0Mg;a`^4>ZEmlX@K_;izQpad)8ci%R&L4~Wq;aTGL+r<
zTVxI%E1h>d(gQaEE<)daeE98(XVEzeHvWuX^xwOOtG}<hgGj}U4UI91X&;(Dzg)!!
zS{P<sz+=mdw9U6q&EwVYlx)Y0V^8~Gc=PS4A$RVxE(3=yxwXbrx{N%GZ>$YW#Ye3!
z^;^yjPOZ1NHn^ZCDaa`chkx{|XI$=pamU42rc@Yk0cEI{8(6?dT=-*L#pPz}#i-og
zG9f8J&`F{i7)8T4N@n8G6(Tc{pZS)V4()nv<GtH!%cD~y3O4?enP#K|7fFI0pzm2_
zYsa*5M<{!*8j50Tib<)ES?9CrZI)d}A<IXSy}Gk&GA~k)5z3NRa({KJ182%<x_Y3y
z!}&|Dm;fb0QHW9g5-C9iIpFWJy*IsfyZ=5}`DnQ`f)mC#P%a%<P<Bi&wSDmaU6ut2
zA$yZFtt{1@3ICGLFl`4~h^Z`bkbQK(U`SmD`IzE{Nv$YVgbe41F<xwIRMUSXxvM+7
zzE$EtS*D#IdP{3LA%FI1YU)f;vav*UUc?HGOub~lRs_x6@$qRj^V$(Y`@+*Hq)AH_
zpAnrU{Z{n$#6XuDY8~oq)Of`-ol|j2%UP$|WF7m~)y#-9oa25-G05>_?w4F|Mxs(s
ziZte~l@i@RtRh9Q1-;N9CtI%<bVEY%9FIUH7EmUkxX&lDLx0i}5vPcCIW-2cZre7>
zlR0$G<++}ZJ4b7VZ7Jl*KHI%z{fH6}B-*tDPb!0D4t*?HEw`ltwU^2rOL907Pjs|c
zN*RWRb|9B&-C3wn`4qA`+F81bZ$>=$_z;6}n3D_jXYDy;@`+^IIky;#&{#ai#b`+$
zA<2A#=%lQ2M1R|ib*j=B0-V7R5SX)eQB_q=mV}{xHeXBoN-y-lCvLV{G=>G|U~H1H
zoYmE#R==#{R&3GZTi}RR)LVX9PYhPQUwZTlxQ&eBZR(ehh+6*~dvf39#o=gKCNddI
zLpNHQl81`2+j(I$v*qE`pt#7ow>jIK@tMv!zb!J+Y=6<D2#8j<;}f&QaXjjEE!3N>
zjZ?2nZgDG&jwju;a<x2j4mxt2)z&gN4;^QvWI^beX_kDa$#qW&%PB@X$z<pgYo(-J
z4dvK(C%0OJlK^G$mZZqjoA*)@373dFB)8)^xg-ZUtjGt?yjPTB3Fp#rrO@}iLmyKK
zrDqmlPJcD?cA%MVhjI)j9h}nbbaG>ku@?&}K8I04s9ka?%lYEqCQtZnjxTqs{=(G-
zr2BqTpdnLH^@VD;u4Ml1X@)iWl9YoK=>cP*mC)R7kXng+<#&&=ddw*kL*3{OM-RXH
zXML^A>W|OsYspvNKdmP@27$V*s{K#c=-lc9mwz(vMe9DYS8=52P)mEio=QSWSQdgJ
zrpig?W{Xgsvaj$=1Xo3QDPtKy%E~_}Wl4wA2>T-uI28=cFE5P1Q7&gAaD?CsLEsdU
zZA@ItO;Uv7f(RU;JcGcI)e9hSgyIYWNBLh4fjdyHjKGcP<$?$twb?QV9HF`Z0;f=O
z9Dg@20D)7Qr#W@Oh^;A}fx!_<GawvUeLf0D2+qK8RFOX(hf^{zFa68G;S_RR5KFpZ
zMQ{NWt|mAG!ja>=Bn<ALa~24$UOY7@o`k+3tQ-|+4trCn|E@PJ1ii&m2P6N7-sYh<
z=}nqUo(rqFIPA82paWsIYRiv;-D0!0Q-29gfZZf&iQ#%tQ&#I*zcRb<$qe-7nbAb9
zOCfIpSy95y&x5>0e96c@p^oUY!kA7)-`bK!xGj#pIc9~RTvSHfb__j_x@l(RKszOx
zOT?QuHI{c2>ZWlg3^X!@x<wBd$dpjGzBc<D8|rk)Y^*uz7|HU$G$z?Oc%Z;wqknc@
zbq@eRy~$wUUXRUTjiNpRf5qO32Ve17y`Mzu1di&ht+#w$UyHE%>)(jFtv-EQ<MS16
zS-blF^(+2Q$q;jjMJHbo{5XyHDNK@5)>~$=zBq9^MGkvE&Bz!gw4?gD2J_Y92N+jO
zBb%6PY#_RNHLZ;xFbt`B%g;0Tr+;VO5p9U485A?VO0m?za>gFGp+=}my_+I+$d0wA
z6e=7;8@h8(+SnTg#xdzErP4xt)BG)NqEOMD<H=z!|Dae<!&7WJ#AhFW$s;16=+5{`
zyz42$OvTkG82*H_ONG=ul=@J<8r~!v-5nEbeg|Q0WoiV=E7`0|ZF@{CrGKA-vTUl|
z<8L2)d7t5X{qFvw=}i$i%B0r66VVk7-Eqq^VCqkJ|L#Lq6!!AnX#@_0{9~J;$-K58
z<Zm@|$b_U-iVdhx&jR=hj!fe0g#mxc<!r#85PTtkzZYms{DP28O>jZDza~6`_ScIS
z0Q+l#Ggv?2emSWBFsU-qKOfY|1%ZBItz~fjn&<*Bze5HK;o>O&A#){i@qfc<5|<G^
zK@<ZpGB=mO5&{#KjPNNamwtmcmw#auMf<;Af7{8bx)jl+$8_UwFR#0Qcby^H2y(HZ
z=!MbL{QY&8{;-OK^RWq9;LmUWG@p0BbD`Z*5;MvV$>!UahJE=|JsBlDDV-x4cv3q-
z6dFfC7znPOv>&A;*ldru4!GbYtp+we9D^E_NKP0NxkP#{vxQB1wLe+i!G9O1+eva(
z1napZEeCpZqrDM^93PF)N~3dpuU<bmp)?n$)cEpI3ghjWBjGsf>jgC^mTyG8V=Yxj
z!{HKu!mA`*=hPs`R(&!}|C7}pd^uIW4XyRU8h`GPWDA}bG)m)0_1e&L(Nfuc+CQyg
zSHN)*i<$(fxFr^c64zkDMSr2B<`Sns(PENW#6V$P33zfdrDtEmVT;B2yPpC_3Bz%J
z@1%aL@#)@Hn5v@|He6__j!t+aYc^V+XrIAD{T9fuzCJ4$Ljqa*OfOsDtjwr>2Acpm
zo6mxyu3jIQg0*Ms)BO6t4J<!f|1h$H{>K(q7RSgbyz`DuPJb@(Cx27k#{y#8fVlEa
z)q{?7hSm})#}Wj+OsP>sUZKfye(JOwq^J_v&Qlf#t*6FGBd{|zb=6>8QHH@hxg(Vm
z`C0c%;?T(D1El4v<kd4sJRGfRNJQ@uYnCv*vjd|({Ct<DVu~G~f_>oBaLP}2GsQ~}
z%OFj_XKaAraCZ^&3x7Pv{`u6U<@55bt$JJ#UfHfhXwzWp?Cj+6#99kLve;I^-8h`~
zshzSfz8$YhHOvYq7=}tc9bKv{eNxZw71?Czq3BiCtSRW)l2K{l$?}p>YZ2kn`I?d6
zQ<YhdXcVD2rh_k+)sV<*LY`FP%;C(7m!92BYT6=QF*Ih}WPi_4{bqv|$Oau`DpYJ}
zqeKOE_a#wvU(>dYZI56lZ7cI@_IWs>B8<>aW0#&HMdE1H164l!aW{)C2`^yZGVLpP
z9*%JToXxI8sEtg`k$J1g3dLi;*{+2m!X%vX9Eu8ea{jyYFV^A&)aHtsv0ouS^$WLO
zPS4qjMZU_R_kUkbT$B-W5xPrn!33_$U#aF(wY(%ZSplvu&@GaTlUp{6OyWq9dwuh!
z@(IBtPc@IVdt;;5RL?kI6}m1aajJaAd1j(MTR+rzAA0Y0?;%I|YP~ErCxUUIUO394
zabter9V)$4U6#+1DST;9!n!jrOjk}9Va!Mp8Yz=WZGY>n6`i$B^phR!tW8_Jj(TdT
z8r;04HX#bHw`{4I#1-@ole9qZu+i)39VT#r-r<gXdwR!s6w&X9-AVm(AiX6NhY~kn
zcbLQlc86Q^p4gpvOIIcH$_TL)R(1s@p-2&zS`zuXeT0KCMDv408ImZqrrf@>9yU!g
z?8o=xBYzCebEva|kI+%!m<U^f$b39$Een`T%_dcuB|c)P)UEgkO(dZB7JNkR_a<_F
z-L2ZeCf(=!fs;->v;kwFv8FY2F=J3-0yK<t{t8|~;WLo=;d5>Mag2{maS|;*4X4OD
z1=aHC6s)4%aLO+Wt3)`KDW>jbZSiPc4O8sPEPsumJXTSjyR>5N%<(cnMCbRo_x^D{
zk5Py1{5AANc1FWfpLFe0%jM*htWV!+bc-S5DMRa2DHyKeBn;;_Dk@<YxV==9K+i!H
z%lA4r`Qh|yw$h#2VP9%`Nh&<0=k6cdHI7pX8PdW8#T1gXFo{G&4?CP>>BQ$Xa&(v%
z6n~@$`pU&o%t!6H3*!=8E~d~FlBjlcD+&z=FxtK)gO(D&%?UKN$ZGy<DDR!nXUckE
zl-aNYViGr`&zR%|`iza<fIedq7w9v-{_W_q3-zk>*@$0mNT2bTZ9|_isT<H|$^gA&
z-vNE5ltsC+CE?8^uAtADqy_qnjb2ZmF-L(5^ci>L+tX(#;)H!q{23B6(Uwf^P~Zma
z*-+pDdB)6nPu$r>cD6Ja^YUW9x(d`DhsF;g>t;mOT=cFdve+H%!GC2%4wpeV8509B
zFfy0%&L<O>`iVDLe?g)|a{hh)Y4f?E&AZ+G!|v5d$i@(>nBmRCW5e)(;+s+2)T%Lz
z5iMBr@TK{1M=96%+~XrD6#^fF-xgPOxE0Le4W`qF9d3<ecip@EnBNg9p5iW_PFI#N
zSiEVhp&lNvG2>w#K0f@sqlt7%l%>KplnAA*##KAYn{RkJfBxkCJMe1RD6d=)rCp<q
zB;Y9Xsz8b2_<%A63MKwaTToF;ODh5J831^u63i=*FTsZJz(oAdkGQh*EF*C7{fNkm
z5NFbF<zXEUUY;`3m5*mownuN2)|7U+(!>j;p!dy5dj%O!xj8T;y$p_#<Va5BojC2;
z_m_9ozn>m<e}C=<Vr7~b;Q#H*&OeY$5p7(<IdM!j-##^uyI*RkS<BTm7;I1?)mo`i
ziJB;78x5?GWUGThe&-4gpPzyhBEu3W+OJ?@#hv#(LvWI4KA?h~@OfJqFhoNM^)R@K
zP>VakU>zOfl~X|&47KJ}MP7pqoeFGIQsbNwhFcXRe+Yomdk-Z^<xRHI)e_;-5hwIO
zB5wy0>5S77DF-z9P(x3LTTQ(a1U-q)I4_~haU$&s!hni6>qZGXj`>C_{D~Nw97I_I
zb*er*A1<WL?x<qqxebh)<k<)}kgVof7BH0}47d6)ZyxvYh%uu_G;9F!E{@RsEEWwc
zT$e{afA4QO=Qkhbky%C<9u+6^5zWVs2;yq`e&63Rck{>Jz-ewiecR*vHEwBr^Vip}
z_&-Mj7ls&JQU7s28!#o(iV(G`Qb>GtLg1NGuy%yEdBtZsb)QlLO(uNma?9HIDg~ZW
zQODO15TT>_02{+i_)^JViaYO9&>as%Ip`{he-N)*kC1Ppg9fgSkyFm9@Ru=!UmfCn
zryo?2t%Ps`EXPiMAxNp#JC?#8vl{kW1H3L@=g0t+Kx)4fAZY)OCzu`%&jDgt-iyNK
z(wo`}VNZ%73$}z9gbJEv;x*@zHF*$j?hB#{GN<F^lS8p)Wc~I)b}M3g%8(UA#>O)a
z$~WQo0Dr&jYJ@D^_J3Dot;>dRPAKC{x>D{OkxQu`U#}!uuBa-EB2o{e;EVKJ2W0Pn
z!cdf-Wn+RW!lkI8<_e=toJus^S1jZEclx%K{eYo(h1Lev5;Yg9=TS+`@#HBR3shHm
ztaOYx*!OlGTR~L!>pfIMAJWv2raaUL-%izUUw`4DDFjylCmbOMh+2>`sn&mduucIv
z1m`4Jsd9FW2B$_$;tC^_2y~RGF(V8HxIj@(E14Q!7kDa}46R-!Y)e~Ij!<W}w8awQ
zH8z1diJ>j6AWj)Yw#ka4$|#DCPe)%5>X5ze&rBIdRKqE!^Pr#8WO85Ke4NKAcc@cP
zn19xizOc8Y>!Srw!Ft$$UKP!piSdTXYlsc#rt9ko;p+)52Ke<nestr?4lINoM?C8S
z*OwMGhTC%5!-EEc%`KVhSmX#pi%~Kn$Tf4P#lY24S;VfW+E!E&Es9ZSQP4Ko|HNGE
zoB|*e0m1P@C>~hLtCK_thfHK(233qT=zleP?B4v4`bsIsPxw}p6(_#imzFu+w{1WE
zl#~&Uw)pb#r}t87+*9khIx5TI)N~8aiQ%jw@~MWNLP}(1_Ad=Rg+g&}%w%Gil-T8i
z&eY&lm^U7D=DM%u?}9;RYVj&aTMjzC3mY+z$jqQKHG?&Grl>eJ7<3v(3^QvevwxOw
zu16`mX3~jbfCw!%opknLo;m3(j+{5?ERM93&Tx5Lm|mjKQ6=j$(@Q+zwD)9ojw!tl
zRucF=62)eT^7|i4ZFw?L@T0$&cxnH=j+g#ybzEM8YQkN`?yoSLSxyAQDUe#6nrA_9
z7Ul|LESy0m4BLBNsnxmkAz*ICv47fxQ+}uo-B;|=b(xT*@M4m&TYlH&#h4@W-4L){
zGX05+ZFpEfR9Gg0*rZcBOMHb;Ly@_xo27w6WvHv`(-YxpIw=cO=4zA%b>Jc02YIz6
zF)w80G04=x9=icRu2N&t>&qL}HU5c~&%V^BQlt;Y!r8vYZ2zq4hp_-V*MCkyBd(Q1
zem~bm#~MQBIchQU-BBv;T39h$;(dPCjlUzMroGdiIkw?+V5t^8l9c_~Z%bQ?qov8V
zwvnF(ea~B~MV4R#Oe?L`Cve~%K#>EAX6cot!>ZW4g`;~#n?8`xN(*<6NG=CZvMolM
zi$rL1V`Xbk$Sq;Z&0Jxp&wt)eq3RFS=@Tue7`H?){+tUcj&htC`@{>X8q3uSstVu>
zSx}+=KsMiS!dL>_a6MH5Uf52RFK@7%DgiF+rYh`Dx0*VSs<fFJ%H#<bQ<b*1*-Mo`
zH&{z?#)zQo1+1l*bp-XnhK^c6ykad?L0VW#Rc~LvmZ|_Qtfgu}K7apO3P~eW9WQ1r
z<t<22v!yMT5I0y$RS*}}QZ+{}X)Sdgr+F<^OM1oh@sd@PG{jglpmAXpl@?glsY_N-
zum-sF0RO@&DlM>T{M)Re7;?9g_WxDX@+yi6zY~1E9mTo(r;C?QeFo3mJ{3ovYyIS>
zQAS;{eex;KZo6a@Js-=9sW0f*_K(hY|3aX;G$n24?wqPU5PIu>0JTmH`<GFuHWUFd
zmr)QnCYK;THw%|3k~b=U?{6D768)aPg1~+0eK429nc<H^5g=_Yy#gsNh;zW*1xDaF
zYU0+F3|nq<|NXt;$}7o|8_80w0*TCBQQYBh-kUe%QlMO1aD>0rMPh4};8?V^#0kE)
zmO7=_TI5vQw^dxG+*;X{*jlB!l4`qWqZWQ*O>z_|wYwODrSE`$8-md)M|`ibSy5qu
zCNm=j44B&}k?>hV-vQ>-(k!WFan;s0xX7_Qu_n1FiQ3)hqQXjRvWvNy7=yT2QX4m@
zi?#MotB&ouCXQ%vZ8sB|?O`#pIi(GNVOUePhY>kWx%Gv2xHJcv8@WpBOYmUGeN&yZ
znFRBNu=IVi6Pa~?C{9x8?^Y+Jwn@M*8lw3z!cR)tIz+4o_mhYTP52^_brp#&*DPoi
zb4{qWUa_>E5zX92<>EAR8wx+r(i_9*TZ7uO;Dr==FQ8kZ+`Pcz68H0EbxA|>J|~w_
zNo}y;Qt3S^<WlWj&bV81Z*6fo^lpk3O-=1-8O_?ph2OG&4UVQ_Dt#nI*q6C?lOXIg
zv<cz6oT+~>2Ae%i1DdJbRdNL`XdM!o8trLV#zLiykRl?twk8}~n46I*uBbFeA=FFO
zHhD#C*FLh~`BMAP!;Ko?PQg^rGI&HxwNjM!EZn5j?1D)ZA<~-FRq59k!Ctj@ND^0b
zY|l!?AuthtT^6a;d<U4AZ8P7cwzk?PnOlVCV2^7q;u88h_>|DbYpb59Ju%y_3PwY$
z+hCX~7P#0brS>D*?b@%t8BDXNvpy7icv?VEtor2Zue&$@yBu}T#^cG<b>B>`K8&v3
z2Czyv>Hg7u-+hY<406)F9-U3!Qj8v<=iqv8?4i+rR|ExS>{F(bC*ASQ`>Boh@_hVr
z_iPybY<fN!cgNkoUVm@@zxp(tUS9vL>pq@OKi$0dXOmyLpKgA+Jo+&Dwfo_8`stsS
z)9=UA(bbQqXCwd9Mfd(<^1l0(d<>qt>#MWw$D45*^BUj&<)zeo^io;=`t<1&{9BsT
zl1b}-XGr};n<2=<Ga{OVe8QqLlP3#fC9hOYx@Qi<A9vqP-b}jZ?yDqNu%s!~0+=!Q
z4@T^2LttNxh&gyQ!XohvqOeUPt=NEN0Lvf}K)$3EB*}vj&)*ucM~wcu`}cqT3(^Nd
z$&GI=E>70>ppbn*hyo|xkT~z^x^E{JldI!@ORS1x`+@6Toldb%V;B3&zki;7cRW3v
zjvVzjUQEVQJmdwp2$lAf7Z#Sr3hlzOgi%7f0Hl#-LVsa+s9@7}p{0eaT|9l-y}Fv5
z9gn7OyI0R&bZ<tVrzZ<Oe|7qC1XB?}<LPxvHrmE`J-VLUT%C=sXUWR${V@7)e)`RS
z<a1jMu&@!)2}ZcWZ(B}Gty^2h>sh*NCOmoDXH1)ApD}M+G_;yiP96#}EeJ(J^9*HL
z)t4*>Lz!v<%5vHR$~b0^1w~5r6?s|+GX<v$t8f0l8{jDyB#;!LNi-wf7s-2o8T<<(
zWVn?<i_ij|k!U0okF>vP1R+l%TO^i$!a+*F3h%)PVJ|?qdW&>c@dRS!s9u#<Z3sCD
z9-JBeC~!uMvJCYq3qVc)a$1Y`rEWdmwG<B$`6KYY1o0*CTb`DfW*1@?c?N!@mB4uo
zEnrH~#9M!XWE8@l2_w#IT_g<ClMf*bQ;l=NsAV9BTk41*3@sX3acJ?-lA)!4yAeyy
zSm`<^mcoGD;dsTeH{|F(cm)_10g+fNgY{7Wc|vSMAlOV-vv<e=0v@^<pPUeUygQ_&
zM({L>0JxX4niThH#y~WM6d%V3TU&4?*o=IJEJp?73t~w^^+i`bVVo8yK|%zmScaZ2
z`!j-zJsgrkinjuTD+CU$Qf$nBwd7b#M;XJSKuvoz0wNW(2L<c{R5$X8uQ_c;vAY|g
zn0e&66&4e$4~&V;A)lcnO+n5OkGY4E4~BTC<)fFG;vDgolx)-B_hss0k0EVk#1_`|
zQDy2vZuvdUx{$-RPC-#tQ&5Z#OF?l#v(7T-Y}ezqH6$4tk__q4N<%Au53Mq^hC~*J
zL>6gHg4$LYX#rhAkJaE4avw|Z5Q<Ji@R8T(Gp;7SQCBBa(zY=b<L<y^vjS0ZYlXlH
zldAVx0oEYZTm>|OKS5*I1AT~isZA#r+kh+LLFEzpHLR9{v7$UjH7Y^DV+u`|_h&)C
zAGNRsd$K8aSQ6?Q#SW=|oRjTMrJAFN>?5CWB5!zt2KOkq;Q?5K(iK$_iuU+8NljS1
z!WFkhXfVxD?==EEK_Qe_k^fP;XrZ<1u{oEsWS<sxU;3h<JR+N4kET75gT`6c>9#Gl
zl0=upXjFP!F*;0!A|2sF_n<>(XaN^{+jA#Wn!)xrDR=;KB}8(6$O^E6xs-sa5OUoc
zVnG!s>S9DKi2AAuC}T~GlGt{yf^%VK(MoAIy#Z{rLxb6(CNLG{gDfgXEUM@osS$aT
zGI_aMuQyV$t&6--S1<zM{svuH5(@sfNmr1mH+Ap_2%aTbv!s3?RZ*Slnwon-{cNpq
zxUOoPS;@b(#*w&xpmBK6I6P<^9yAUQ8ixms!)bTLLJ5l)Y$ruMl1^JGYR58)q^WB%
zgXEY3K(ESNOUYpmxCe|;YSh6BHVndD_gMkCM5QDMCZuW*0cz)+pf7}yM&{d=LIwF!
z4gr2q+gR8KpL6tfR>uSZVuzA;{vl+ijy{!DpI(pJq&$9qMh|l)8)E`=)wcfO%+eo2
zaELxDZ|}^}*#CC^^ZBL0%1^F7c6QZ0zd5~lcQzT1-kpv=yt|xSoL;rd-BP-xJ+VW+
zSp*n!)o_rUp32e7%6G`0%a`iRg&uR>S8vD7n{);5OWLTbKoKNl=?%I<M0z_wd{o|F
zsCu_%cUiN4i7F4=MAb!hmx0M;U~<vj`J0n~DP2NU!L~N2b#jH2LB3tLS+!A_wFPf-
z@nr!u0%6RDK}}82Oq6zH0pWI*F|dGClr0A-Q(}qv2%l2#FoEQaWT32WI|-xeBi^K|
z9VOeF0f*Y5+8VgJ2Dth_;HE|3>Hu6FfU7&6WmK?#V#o`Jq^7|Rf|{7CT9k#n8*uFK
zg3`Qr64L<bK)|6kC8|4eK~A3Yd`eX8h~zK@p$3p!y2}Lm#l2_EYmZFp(%$pET?MLK
z*T{!;T?GP~HX4<3e|CMi&F`iC?&V`zwF_EPKi+rfyERdzyh>C_A4pVnfvD0zR7rND
zC;SzED;RF+DO$wnK>|NpR<Ap6*A<`pcRpv^b&AR;9)z4i1UE&xgXEMp%IQnpW^j&H
z-PKLHBG03^OMjM^C6=(Zd>*w3y!(*5HSqH43sv^s;e&ve7Z0cMaD<d;H{hl0Q<!HQ
z&Em0m2Z_^^EQglefX8-lU+3ik6H;nDh|pnIMhfK>F`ea0DKxKn>&W)Z(k4AQpgvOi
z)4ZQ_FR+#q-#z(pYsk&s&|HFC2@iu@Kkr+}4LhWHIMdB_C-9|=7@h9`rd+)+9VAp0
zIfC<S2k8F{=)o_SQK&W)0x~(5f$BXImundgD1X%&%Z?kl@jhRn+-887CRuy}f`A#>
zGLT!0xovU?vbM3ndV_3`e1D2>vbtMpNj(Pk(9<I}S*&*ztH^iXc6|5g;pO$={R!00
zGN*-=yVtKffgd>j6&weR-dQ1-0kM1izWZs<xit8^;G;MPz{leM4&VCl4ur!Q!sJIB
z-hWvo_HC{_mN}r)D(TDW{1$l!hcm6SJix=<gbmN(@2`J(;2Cw!yb1q?7yb3|_2J(S
zl@JL`8^1H!GQq{}pWh$C2PHH!*6*ZbUZ~ygUv^&~{z{>iy~sPV%?V^1H%9A$ZPGBU
z-Od<bU~OSrAYUOHw9apiZzNdT{Kf(OF@LZ#5y=#StABp~wj(2-KQ;O+?xYRfQJnDk
zKv;%s;EV?tQYWax2ts{-@_d#qC!#t~x=>)+WqNd>W!QCzQxX*nyrd9O`kbT4wNIQ<
zr?i7|S93-JO5DPPGan|LsVfvJ1)@u-L@DuO-hWO(7LBOrp0rE~=;5iCND~@_#eWfP
z(yw4H+Llp;qgvu4vZe%n9MkXmelA9(2dJ(<t@9EuyRVR-ml-<g-un}Z*&}_@s6k#v
zGVR<z%j4IlyhP3~u@mwi|M&C`^y5E%eR?PP<A1+Ao!!$5rU6C(-*Bb1kALG*=EYXc
zEe7tH^#tM1dC9!IKp5sdP<f#%3xBI7QpF9Vn-OZ-BC7pVC6r*uO1k)>e>j}OTR6iU
zYUBtsDOFY|5Bn@?6MiY8UaHK6v}rWCY<|%Mj|E0mDxThPMnt{9jq7FqtK&;>G|tIj
ztoS(4kc@aT`uWo$4?&{+aYe`t0Ys}2bx5d;{%Aq=7$3M(i#$YfPltC($bWsLC7JEZ
zp&DxAUZ7m|l&dltxvnZFeJAu|GX99T?o2hwt#<?_b669H5XiPu8NgHqi@KQ=TYGfP
z7!Nj5PUD$C#uBKbc8bD9m8le&;`X(0(ihG|XDNH?iWZ8MJd0zdt_WXjekwP`z)exC
zsOkgXH>}4MD>x3GGs|IPj(;mK41N4H9Itq1O-fuI_^EroS00^4%)<DByVOo&P^IoN
zW)CJY6Vt$qW{gXx9*rz01^eo5q}XSa(x~r8`!7{|s+vYWub8I-8b#IAOlM3~se+Sf
zzbaNUYFL%r7P<vhB7#eHzzP3i>_tWEXf>G1rKHXkt7G{-Vhj~y)qjh&IQsIISy9q{
z=eaVg0GO#_+FjMLNF}+_Y4*ai9(1a<q*_`th2f(_0^L!dnvMn7GGQO!`ueNvox@`q
zE&1%=cnTkMlAP2&tS_StK5x4Qj?2pNB9T0A&AFgrROx8B-)13IIc6PY<lqtIHE}4-
zZ%T3~j&nv&E=6XAnSZCmBPc;G@!t{4$(%alL)wh!e`L-WPI3LLb}4+z>bVB{I8D6K
zRuU0NEw`HP9w{rS1~hF<p)8B(DzP1Uzdx1S&0+%)*k>o~>)l*3{pPy1Q)h!Yhrfw=
ztMZ0*iVKcS1Q^4V%#oG4zH_;PGU+%=;N6)kXd&ZKXNFS+N`H2`?@!vD%^Ia<+N&WQ
ze`z>HSJ!FJwe=e?nhE?&>Nf%%mr`({fRjob@sG|`gJI-zF6t~4KXOq=bICmJStR+P
za~;`6q#z}ivP1l@ZCpnBLXO6)4g6%6W8+`b*HtB!a4Zy(0t8-io}_jl+p$k}iE;6n
zs#Icf1T|cA*?)uKGQ$jTWT}!|=JzFLD<|-yxPsY>Y{-ly5U2UY^|iNL6(f)(pRFzl
zO!mPlI(x#qD=J(Zn82KxKonxWx|J(#bpPTkj*YwZcg#4h$1B&-R^p{R^I+!qLi0cU
zG2?Q3B~x0=BFuyfp%fi=kj*+!V6^PKq+`G}ljw#B)PGXVtkpvdTkTK8dl>0;tJs0%
zOmI2XmLxeLmN<2erhwU;qq@&x(G`b-{)QGiLJVFiWPRBK#Z35in;l_ave|L=7B)LV
zykxV4x0@v9Df64n@NyA;MzNXz?qC9-fF*|LbxZ(*DyYRo2rCXx{N-xg6bgv@iGm27
zj7_*%9DhST;-*HxkIne2somfNz%tn?IYBnPBZyAmY{uR~dy!?x&5%1mkV0id2hrZ1
zBP4|6$^zURVzC!C!6E7Em1vyjO9VN~n$vg|4CyhR#uN%f8oNy{l@T1&IMKM!9z5^R
zTeztLJrZ}TFyY5s(4tBa2X2iTh<p5V?vyAszkkHy5-~3M)6?c5g)0WiZy8cXh<Es~
zSl7nDOdu}!u+pX_vxok|TY6jxq0#MDVq03LWcFH<213>7TuNavjmSGA33w*0Ur84P
zGlmIWZxl%Y&tKfy32r7w5}3!)JB9&-Eb7g91H%F*nSw1(R|Mh)hD9J=Ff2;Z^A9$x
zb${3v!y>acHmuuE8(;-fR&F2PTCS;Xk8d%yHR}=kndEA|{)5U54u?@woQbDJRItR6
zxi9DHvZWpcngJgg#if=U$uu*CQe?1W+FJv2SFIqtdAlXz4_fV8tvU5ea^X-}5<K!L
z{nB+V={)Lh&0plH^?1^~6W4V|a(S}vm4B7G`g3;+<wMttN-O*V*6mh`Yem;uUR>aJ
z!8~fF+Eg=j>j0`#($Mp4lmeYHN^wNy`D6;|U=JcnE%t<<>~&koQ-Wf+;}5#BCKb9i
z-yHbBoZO^+$qw|`*|3<j2(Nbd`H)R8n7OER*A=!9R~^t@<Wfz4;o4fb52*}=a(~zL
zFd0I<G^?+NoxAlE+C~D0dYJXr6Z##4xjeX$F-D?N`&eu6D*67$pMUxO<+uO*@~_{{
zUYH8S3=AZMfT$*`@2=T>f7NLDc3M`!<J3#JzKIk<1Meyx&z+w*cGpMZs+G#Te0{^C
z;yOzIYN!isBPE!Zl0seCh6zP~cz?S#4CZDICbwt>qG~W0{)jG&Fu<&JZ`_5UyWHJ{
zAprj%U6^oP(QVp=ArLq0!Vt(8D!OF$1{GZb@j^wH?*DdOn3Z_tF3b#H-mnWpWwuQh
zhCtn*3nL(0IsGGaVFY)~Z`y?+5cSIqCwj%4vJ}#ih2|4InSD=1A-mua;D1s<p*RS_
zNYZlA7wYXLU!HMrVqfdaSJ|`m%!&yYg_*Z!2bk&!Iy+Ph#K+CzgNxrbRJIY|234dH
z;DsU*A?JoJ+j70Bi2eSWXHkSzq!m3NjGaBACyYX>0c*88xz)``3m2jRR!R-iT&5kB
z+Wf)w`Zl2COo$04^TwtmrabC~N`&86i?iso_@=F<RA#If{<i7L1vC!j&ihqexwHCr
zsQK+V72LFAIj1>eOKnZ9k-YH~^?yY~mEM<8s5TS<GnX;>DifEyrZ>g`4wr+cH#dJO
zt&nIrODbS1fPL1)&RaI0*bS3}ndUd%@fGs3RzYmuZxkg6dCi;txy1w2cGWhViW~y5
z=b^Oa^}S2r`(%|rfTbkot}&-k)%c~NORlBMJMru7<HcVWN032e$ZYldmy7g(z$)is
zg{UG>_4kMB<HavhY8ydut&_11kamC0+A=^++2j&hWuV8mz@`b9W}GQunw-=5IW*7e
zd+EK_Sz__sSDiY=orghyBd3!mX#u=RACZ_0i&i?xK-)+y?$n7G^{AgQs1S_Q(&tK{
z=@eR+N|v0IqVL8++{rr826CKj1O?u8$mfiv&}NlwisQOYP@fZ&D<ZNUbX0%Hv{nJ%
z8p^!<c-1WNesqb`3dojA^wp#m$6C;pOFp$r7-JQ^2_^iVwdlZ0Ncq?RclpQXFCV`C
z^ZC!OSJOoY3Vo0`)!Ep#k+#|?XCe%)uR9zuW0>yVry}j16qtvbEY&tr=yfkm-wofO
z*!_aulJr`wBjmMhpGq>~P|JV9YC6?i34;ReLjsw_QpfS;<;sib-gIs3)jc<hsciO!
z1Kc(GwDs8AjDgNg@xE{NYnvijcG)`F6hV`6oL<vivFQ`_m(5M7xbu)Z<%tfGJEgVA
zsZ1T(Z<=ka_Uf+j6hZW@Rzv+)ueiN4?1t7UW8#sPyB@zbHy8FS2#$ZNRSMj;fgY}%
zPNh^!aOoFjnMn?|?QkCNS+EWFjo?;*eG-Ni*ivE*D(0f8C9mJW0@M?Vkc@hKp2^#t
zc>4&u<@!+@1w2~`i2$Sek>@GH9a!oQeUoa@?S6-DGFoL9adC1m7TQI;(-=2~OBo;j
z{Wjb3^l<>aRpgFPNK=1vtpWS$;DGUzyeP+vwhGBescCjf+f_FvSX1?1&0IaaGt8~F
z4tN+u^vBuKM6MoZN0FNHDOonws5VtBMb0lBcdS?;(cZ4FP{;E+c`4~sh4L20=NtL<
zh6>FsSO`2(2yKIqKvE8E`e2?UXfUOu=eFHl`-y#|!KKcv*Y|%p^ig>@LJ{0}JqN>M
zAjr2hn+k7t+aHMQ*hJVf+e(RBqV6K^xF{u_3VWM%!Xpv(V&rvAHP<OH7=6>d>Cm6r
zu%D0&i=69{L4X^xVXY&1vN6~U9A;^bH_df7=S(U=vovys&onk1MQ;2Ez1kKIx&k=1
zWxVPLNf~z(mQ;Ur8+IS!{z+z6@&#Q_y3C|<E3!@F6z#OJjhoSkTtfrRCgGtQb*J+1
zz8#A9=cXDB?HB>n2eaF_#}7RF6M4j=4aO}*9`Sjek7|1od4t0@A8zYtojiNblYHbA
zQ7G*^7s+S1PBA_g$!GYaVtwDC3f$&En<4-k%%9EyVrqYKY_l+MXATfCDCbCOS<3+u
zDB@9ENZG^znv`>9iUV96D#>7V+!hsHp5pe9E%LgwsACZMz%;rfDB;89zmo(a8qeXz
zatXFHC5<j2PGJY-l7~so5iPsI6P2-C$QH&q$8l1WOHx4$rM90YQQ$_PyfJHF;IWvs
zP`Tv2Mmv9Ln}IZWm5ZkNTEjzuivsRXlcXz8KMIZ$vnM_MHzPewR3VjLY1vlNGaw(j
zKuTRmdK%soS=+ft&qVN$^+%*9UoFpi&p~<`9eA!ZXC^(Rmd&K65cr*tp2)@%*&PqH
zLgJ33r;t2DdWyGqAU%b|8PZegKOgD2(5_5+j^KZCM$%KpY#-87NZo<-tnCfo1L>JY
zjnt`ktk_cG2GUbVnjt;K+uKP`A#jHDlt8{1(o?!IO?ryAcP2g2fO2+E(o;y>f%Fs-
zXGl-!(F-R%mpRuZg8=u1lb*|*Ye`S(+-gc;Jt4|_W{Bp<&J%>FB?WX;GebhOXUo>_
zP^f?Flhyyk=p-?kVD6wJZjGJ?M?U<f^YXXk0Vg@tSN8;!<qGy3Xwii|AFziGT>h9(
z(?sJ-JTHNA$q!fGhQS4%s_Z5Px860*$Kd8n#lZpoX$)>*=eZbM;1*z?jloUV;BmCw
znmN#(gTWOBo{7N~TAs+@3Mrcy+=yJ+O9_A6$w+M`a5)gp4-J~-Z*9Fr4p+?aw?;W#
zI6X_>8l|DsF~{C=GJvTf*RaKP<ZasOU7>55yT#}g#1#OfGp}Q-hAZA{q#5Ft4@9oS
zV2-#2j!bFT3~!qX1f5977eU)Pg@J@&oByk7PyuD}EhKHzzmVI-|4~)6YTgR|12zLM
z86KBWs5TS=GclI|t1c4(F_-b~1Sdu9Tgz@7#}(c4D>^$8FiUl-ZoO=oKtziHvWf-n
zoB#o0OOOTGi0y?xKc~8@d!}x`IZassyqFnIPuG1v`$)QZ*XZVNm&UF)S$`<q{p<1X
zSGtI<Rr2}I{2{-+e!Tkc)u4#6%K6kd&x$!V|M_^8e_(RThtv>0oe#|~KQwQzzLjJL
zW*TR@H>aCl-ZkRNn_|nPVN313bCoTPSI))8`=o6Csk0@2qqT$9{D9XzUa*rPp=HmZ
z{p-ZIzXHYs7w46=X~YGTVSirj-~ngw!WS1RE-xM;P0Ag;A|nZcPAjT|Qy9*%rV!7e
zkeN~WEq2T}b@#K+*Y0k;&Oy;q@cEl5G&3E9NGsR|d#@UsgJK(Y7<<+Xy<uyLtx+K}
z7hU<G$j({F{;ie0x^nKaNNth{%0^dm4R<H4Y+!l!$adTL8*LSVl7FB`G|AtfB-ms|
z{A2O={mHt+_t<)Qw_J?mgflkE#YrXWr|eSu2j5>8Sx^v4G#S^)Qe9byFB6<%Ke9qD
zWi1DD$__XTrE61+S?!czE2$!7v__8U-nY&+r?*z_>dJX+^?6{u;t<Et(s_<(&QeP{
zWyz)r)nyq=j8c4L!hco--K+iiiJlQT2tsq|87#`A7dKy!oqhSO=*{H7P#bC+a&c;Y
z<uYALaa)#)P1zM2x2f8N+eB&0HCykma3&-bNLrBmbRtKqQXp}&hM@HNu)g4uQ68Cg
zT$l-pmv|hVmb8%+=X_oQPHH5ap8Iaw_^ESn&f+!zxYX*>uYZ>vKi8=Dhr%}H`dE<0
zO+~O?kuuR_new1vw?=qZR1Z_m=KQl;J!caFJdQk<$KSzI{kZ1qj_T66BD>{e^vam*
zNY3emWXFnllruRa`;Ii!|NWiGVfyirkC`jGPG;IzZ{4~U?mfvq)N@4SrZ+7{M{<WG
z>w$VVjKNE>+JBYY!E}seCb$L^RWR;I)78P(>AMnt{d7V+iiZ~ujNC;A2RK?(;o!Ma
z;QEW3d=y7``bfYLs*n8TW?>mT#bm^cgUhIr$hJJ=p}2OjIrYn%q1KPr;eLy>)kx8o
z*50n&iZfZP%RcSU+e|&A9_G2c+3D7ELvW{4(Dz<+_kVq!l<S?qa}NTHN0)O!2>#ax
zpO^WNRS0+F_+>{p%ihUVO4H6Qor6aNc)6t;qZ0ATk}f{mg?y%SmkPvaGRBSTuj@MD
zJ}Y*)(3X>~v1Agh-{>rj+S~W{P8$^3y%d<f$uCieFIa8L+9kN0;&9VgWL#9tPEr|G
zm?_`QK7WPk!vbrOj;AH2WXlk`*vGH~dV|Vxewv}XkJahxf0tt8F{vZeY^)=LG4WGo
z=#)!cZ@Q+Nr34`dJ-rl>bTL{RPj_vmAUR9U@KpYxI!q{myU;(WR^n}uD1_-F5FVRD
zb>yF5f3svQqm!~R)Te&?>gJZbzy5uhA>2gTkAE2UhMUpMU?61S9@8`8$s-m6vJ*1C
z>ABb`j~dSP3&wD0AG&z-`}ZGz`03a8KmW9>Rz&AP5LDd0TsC6DOm31e;<6c1LWZ(5
zo^dy~CWPw`i~KAKBqb3Yzc2Dbi%}6as*ZeE%nEB9npc?4ifP+fR4g3?m#@nanidSV
zI)9<d%P!FgB`EOSlAkU56t#JBCgb~MOJLn%8Gt`7+dCL#$j#V$TCfmQ(rK4qn2e36
zDc3)~TP#C1k>W72Uc1euq3(p!1L1|%G&cn?s$<R}>7lJurzK~knGtR{?M|asU~&oY
zv#nree(PI7efmPlDLU9=M77<bW50e3wtsis^Y0D<>4v`HTN(Mgr{Y7aozuoRqbOR*
zmw0)SbBInF9qLuYJhaq-DTm}~CMX+ID{w&<IwhVq_Zd{@Xn2IqaGoXXCqZXyjLPfy
z%;=2daxpq12>zd-Gr3yB?Z-WYs|g;6&D4ZfpqcvS1CW`T;0iD!xIZ5*vrVc3%YTe*
z@{FjA@YXS)OilCvOvXC&jT5N;8DlcmTI@qdJ}nm%`(pP5R7Ox*fn~(qm*X;m;0iD!
z4Y?DUnSMBbSr8h7cF$8mI6S~ykZ9`pY4<R&d#Zkad@N!UPKP4;g5m`Tp`f%Di}l?H
z#9~cwEf$3#KO2-#s<8|cin|ZS34gOpsH_J@o1l0APADj@;Dl16&m1S*rd%Hk(zplV
zg!XmYqOx*`t)g7J_~?RzJ@hm41uEy|%44tIrgC16;y6Mte|o)NZtNa;50!F<X}Ob8
zY)5h*geDpvk$QiEiTlAs;{vJ)eO8dvMd0GOLL0%w>c}4jE*evhBm2?C41X5*ddPhS
zSK2c-?LiXRs9;D4`CLdM`P}vFCqfc)NA_m67SCJk95tMJ4tWCNpMV-d=Ojft8Z$Hu
zt|KmcF~fpy3Mv<2hJ=lay?Y{Ncz$}o3`1PP49OCH+yohNQYJ<z;$fH}C7=J5LZ!nR
zWJt=Ps#ZrfK!#2svFcMBA%8={2y5Ml;TmM<R0u+|HOSCmMiEEWAVZ^U%*M!o#6H9j
zTfVV1vlYZJ=6_u;0&Wd4gcT&0?TBH1kYAr&0Sw8vHM2VMVKF7FAr+IGOo<u1FryMy
z$kpS8$j3>D5ygUIo3ee|y#B{B7U#6W$__55mBDYq;v58uA~RM%LR~C;hSQ9VXV6O+
zicMv(HSp4*Gz%tc7-4YOep%L?V=+P#Re(1Hk4rJafT$H-n8yg)|I~wEdKyM3P9KC3
z3aXzBBNTkMVTAI~HjFT3{61}k2*r?vA?|-=jT91>QK&W)0W+6j=sXmcI^!-Vf9+dI
zj~mAozR$1d+-87Gsg_rpEfBCqngBi}%G?|S25d`^9XXB@Ao}_FYUw8H^+rzjNEYA&
zhhnqp?MuD-DA~R3WcQb=7Y|p@?hSQb1gpI69^P~cUr78_5)Ym2yi&p$)jhoLzIiUC
zc6hzOi=;FLFOPo*yxYUAQ2{HIf2%Kbxb>u-AJ^);evPqZmyZ4J`mQR6fR#RYndpI=
z0vGkw;oFC=uVfWnis;fm(;NTv>f!3&SA!zj2<u{JjS?0*e*gX|eL>0!=VPZR2qM${
z_G9<v>TjIvz)WIK`D11G+uM#kd0lN8C2T33v$nCNcEZ}wITwXVUlm)@e@iJXw5EH!
z?(l-0v|?KJ9P+=8jQb~GJaDm27!yZaKpEzh79Ow~FMP1R;qvT0s7bkl6H0S}pp!+l
za0<aWMpxn~6cRI4zYB#K#j>7UdUm*VG6hAZ;L;~uX(l=dku2C5_FgqMc|pcsVC-2l
z^oDIIW}`x8s=CsBm7S81fBl~<d-LSfWkKa6W0bXQ<Qi5-DU2t+dStuv`D<x}0wqRK
zsY(7CCBZ~J;vcHN?~c|TK8Nh(i{&CDC!BGnTogq#ZptpLfAINLl?4T%Mw7NhmgdPq
zd}(h5xser8Da#!6F+1Qel&-U4%t$LV$`GT9l#voSCVSsn(;fe0f4Q3{r?J)NfpLO-
z7)ML-oKka^TGA;=HdUy$Wvn7p>OB*-A=nqaOY+l^2;>=|x%Bh~Wzw<58?v)6zYV>a
z92jatYeT9|O|Mj@YbnlUx!RO$wQ(q+W+#GSOZcJ0pvF&yUk<$$iTZ}pOJiRYDHR=3
z7wHAtF$yhevJ13he;6`~_Qj*)kWpF*M7DZ+Br<|4p~ww+VJuoV7mvmBmv>w3qws2E
zMdhZuN-!dN#aGJ|xUY6QrC1A@+@3{`QQYEnock0~mr<N^dmnX7Bj)yW3e3k)Ra~5U
zN)7?4Cx?1^Tz383vGEbXaF%jIFfilBDp;>bJzpo8(y`Z^e;MQERc=p7ImA{~E|ofx
zCE|+8YMaWGMmb4ms+<O-FpkhPq77q2l_eMGP)0X<@>yP9>Bu<|)yUB76g8~&N+_J4
z+3IXD-We`$T^6HZkba-Jr?k6bq3r0j1eof4=}<N;Mj?5DB<rDj7{;`v=q+V<Fdbpo
z_Hc;tXjoxSf77*rmxWu3z5e8^XcP<2?g;^geD-iMtI6I`BmDKBZ_-w5q3JCiTU5K{
zPd5t}6~}lCIDg|YYG!bG{y?cMaXIzP>!I?GuWI@m4ylE4!%@rj7U_*z=I_bcT`zkm
zXyoHlyRCKAq1bo&7E0?NEHSyhx_9X!$tMX>G%Y1%fA5NDg>5Z67v2z2nmTE8L!>n-
z60KYivG;r#G}EVRbs?)jH_X-5aV^bh{@5PQxT%-v#VX|EXJX~OaH!gh%f^|wzi(hv
zkG+um@p>!AU@fWnS+?Ef#{x@2#V0nGb{cC(T#hMqo_Xp;1UJv!RP+zDW^Q}?6p&Ha
z<FHoqe||_(aEl<QQ6x_{6I6HMbo2FhwQM@mp`m*d+IS(%Gs(;{2t<0>^6Sk~B~TOv
zX<`y8SxuHR=a6dxs!Xj8G*xI&=B9Lr4lc2c4sWVJ-cRk&5~nGAb-LppV1Bb?DlILN
zUp1Tl`{m88a_;)WGCLK89&uAQOG;^tmW8?Pe{_@XxzOYeV};SJe6JT>sYwg?3e_@Q
zC01Ad@b3MOzx?OjzkXR(DIKb96Li;KESoPN2%>4tw`_t~PO+`swCy*AsZB;=l)@>7
za#n<4SFzo&3L1MfB~ma8gvzV_x@atfCETC9D21;#>eufU6)^B&2yI~gv}h!Qp6P7s
zfAtT`Vi^*KT$_$7V&TB!oS7DDnhB}|nKn*-U3LTdu{2{hm_DiyIwA&TnnGbL+>(M`
zqmqXLTwRwuX}W$|XI-JL$)~RW&9}1q0bhQd4qq6h|Cg7(?~HKJMRxz}{*Dts?(L9t
zhIgi^Qr9(1$!SB{AR@Jfe7jf)4Mrr=f2<JZ78{4ccV1(hJ5VO^oi&Wb*+i$sMyo<+
z^}K+qc7CXR$OJSzNwbY;<{(@f5!<~{a20IMb4E2^PK17j8!T@N$W?I8a_AX8)w!ua
zUqhdt(?(kuEEvvL(ejDerU81isaFxDaH0JhoEG)TMRy%Lv%nQXEJ{4?fUV(re{{M}
z3BjVwq31j?0?U|e24Guyp9%h&8kaPB$D2(?aYy`>QC`7c+3FqeS4MFKf93q24u3sU
zu8O}-*yRcFSMIZY@K;842mIA264IN`0DsjH4d<etr>tcxqqqTqWt3KsFSdF+^2G?Q
zAYa^&pB{zPD0CQcKOYoU+cfTme?4QdrMLlwWt7%pv0c4GEVcyKVv!m0cZ0%mHI`9W
zwt8n2R%4nqPVE_OEyW#BSVnOLh2<Lk&QaL&lpBJ98TYe9Vb4>!M`5|jt7yUHa;2MZ
z8gZJ*OR|{R#!>l*de9@jobNK?%lYnu_-edU`(vd)5a06<UyVRbL;S~te-1aHyr%-6
ziSjmEJ{8KVP%@+S8&F>5QghtbRqiCVj?tbX*;EhG35iP(V3+>?0h4(i0WKlD0R&F(
z+PIGd0xKPbRYzAeX1v6PB@K=@or4|rz(kGC03#E2Z%~w!AniGqOF_|s-!dYbFwuzl
zUHgfv=v$**&A=(ADS5b6fAM(>P98njZ)YGSo~CvA;HCR@>50Ct@Md|c8wbBktL!Yb
zx@Qqui%MV`IMsFladGaRLjIx^T6LIptzqEeOe)8ibE!4xEQC8@oxzjyXo|$-O^p`I
zFs`vS<58zOR$*L&X+)}K&v_`W@mhEtIk#mP*Z72(_-4~GjB5hqe~Ybv%P_8qPUrw0
zKK}&f0u-0hPoF|-@?{j4>XD%_-=YO@27x}7nN8c9TOG~y&&%AL6`op?e^};*GbX_Z
zr#pULOa<~T|2SbP7iM?_DMB<2k7d}8LPALt%bo%EE7`IQ_v<TO&_epXa6cr4P$<PM
z(HY#&kudtPuwT1*8Yk4RCHm1ozn1NJm>=82aIE@{L(l&KG(x|Z0mB6ombW(qm%>;%
z6PIgyHz^icOK%%D5WeeIY;QJDGaPdGY6Ag@W0$YMH!NRnp5JYr+-e-0@CKY3?%oVQ
zKS+89Ne`_V9DuMIhP(H}FHfaZmR`^4g;HuwFGqiS`ZU`s4W1~#mKSWV9KzEbt-7u>
z+T>L#)Yat^iUd!T&N~_Tku-4ebNk!fPaBsz!8a^_Zu9qsvp1m0En)cm{U-i^V1#vH
z0Q3S;4<BC-Z#F-3wncR*B(2GBMh+j}4(!RR5}6>$LTRlrt%X(?GTs^xs8qEOUyxP4
zSaqE(RnYisCn5c{6I`E=;8BXPLX&}9%Jt5FM$wDD7{?;xp{uvI0u*ORd~#4I#7F_s
zsxXg#z>314{N}ST44EEn3f`5K5n_phjo(z^8Tk<4a&VXZz36Neh!Rh5_CYsP$F>}k
zQ%Pqmw&HEk9ix)<E$QAo*$<hAl(LMol8s&?b&x_k<kS=29WP%=Er3#vaRlhfS5#nh
zP!svS#QS>Z-SN9mQ9fBOq~xSv9H|!u5wz`p`K65yeZMKXP(dhYQYP!tJek3laz>z?
zcp(<XBth-6g91i{@F<LFX@sIO#JEz*NJ%*+Yix}kc5g}V=E**`Y94?W2F(;LIdTAp
zJT+%i)T~#i&dOL5tgy<jn6ZpwEqkYAs72w?aZZOa)M=rO<<w_nXeq&r;ekY`1%f+&
z5=&^~D;DcgkW+Ckjzv><*9<t*fy5agG@cTgGOPo_1i%H2>b6p~iG4xT5^x*$lB(F6
zShQL=18ZWj<!GMYAB!!gB`mhoD`2r@xP(RK$P=+>>>)hn$e+IFDi?%<i5J24_ZF{3
zaKNc$1n$jhy^2BV1GHWfZ7s)hqHcqK0I4{~Lt(F|V-~Tn$6hcsqmV-!ca{VJEd4|2
zA?HJXsCm2x7>+V-1Osz!D2i1>N|zmVip{S+665+NZBfxU!d5XZ9f4#<Ts9fbGa2LY
zkUEUxOpeDSKSijvdnT)0an{9j-0T0<K26$QPS-g)UQJNVdaE6^7vMFgP9EfcJH<V=
zGA)I7xXx)PZ-3cCDey}P(c1}QtJYT|vphkQWhigQ!`mF1b6Fl;hjavYhK}IWsDoM4
zbuAD_7aohn+Ft`su<_&;Dbp$69R;1$Y-h1C=HmOyxRjP)e~F_d&@TDzvPEM5-hs9O
z?ddVqRBzKMg-UV8Y44X;<IZq@@R}v&(taOOJyM-)cULUDwwDVgq`eT8TQ!WPXc6gf
zgl&qfJEz=ib6om2W-1_6|G*4-c5CCUmy4!RH8vQPxh@w_epe|^y?{aj4p_%8pp+s^
z;KKbA`%HID2O5A<rckSNuT_K8IQSgz=HN8-ypWWOrbJrGSbCOdPv4q<ts+bZZV0@H
zFIL6RIn9Gma^kKXZjb7^YGc5-j0dJ(Ob=e;7Px3S#8G=P2C!sTyg^gJ#u%8b8e=gD
zYoszIzk9(2xxIo=mz!<E`!3Q|2==mxY5)~7tEdX@I{_^{G>zgHZ%ScwuDis52x!|S
zzY>bdDy#ads3%d-uKM(U<K-1td+|Aax{ELKf~oJb4km)(a@ImR0BO+eDpdtxcS$h9
zDb+4{Ri)^GT;{ZAS3$8NI9d{f54q8}_~YIC*B}0Q_t%G6vos>@2Q79kez?3+&W(EZ
zeX0$eEgcKqr_ZzQ(3Z5P9gjItnJn<nd4WEr3X8y(d3mv~(@e*ID#iJ(FM}2Uw)JK9
z#FRt}FKrvt1G)!C+b@G{!QZEXzj@bTyp)rn<GQ)<q6&g&+EsF1kVYzlu3hr>a&A|o
zs1VAL7oc<m3u9e4uHYYPMv!*iSaz9=qq+7qH=0^Aqd3>HVNVLI^FpKT6xcb<sZ$iz
zdcOKh&B0TdkQ1(d>FDe0iJ*2RsoVB-GMCH#_^ou|dAu@iV27+xMx~<?lM80PZ!(l>
zFj-VhoFDU1w;>Slp~igiQ%gPVvWx&@|C2GF*3|9?|G;BDF6HboA7l89jQMoDpzv#6
z)H04Mj`<kpg)twaUSZ6~I4+F&xcrA3^Bt*I8S{1g@_=K1J|44m#(a$H3S&MkQG|9#
z|47Gt@vtOU9n3S1XN>t6r-ji9qdtGM!Wb@$R=6X-{+LfWC1iir{5Z#a%1H9NTN8^d
zN8|60#g@|&7F+5Su-GzO!Xk6zBfz2#0<2#hi{`<w2rlkPD?+hhc*elFWwZpuhIj=W
zHvE=g*!u8)(F~iJ5%aU=mU!h+GXz86Saa0OIIb{iW*iqr&D^4UkDA+Cg38@vLe|04
zz;xPGSJEfKikZL|E~Ek%zuJ&`oX+eGseSzh(~))8H&zKWc17Q`43~Vv#6RS4<n)^1
zy-;f6Bj^w1R3IwPiF>ZE=^Fo=H#k0{IAbCnuf&X5_Ww`BPoIcG{C7_mzKKz4J*MV=
zP&s{IzHjT11M_Cd*BY3IrjFiYUQRVn23R#co|WI@TA2ap_fq6KlWVzm84Rb5#vKUh
zvDN+9$vB%<-`s8f1;DyJs+UoyHWZf-i7yqGTF5uXe*`Y{u>1DimsdKAu2u4KD<AE*
zH}9_gbu}-<Smk^gybFpn4S)G~B|k7Z<wF{bQHgl?{%^zktFKzBg;(=A!WIdzY44q@
zU^8Ah7YFaHvLVc{z1EtsKxc62e2+yreBlRNShUU08F0A_#5kU&=Ndq8*a@OCWhR54
zozKt4f8i5OiXqYb=lc&s^W^F643S|{Y(gtEeIZwaI*`;#3k~BYs?`%vy?ac02^zym
zCF_r%dB|%B6{k}G8*jidwN!N0j>Y{uspCa8OjOaNLQ8a-1#HnmC;G~(+b1K_tNpLj
zCeHJ#tsEu2dxUWT5N|wJ>$6%pTZwf2{$`qif8{RDaLN(YQ~nG|@+@zbXC-B$t7rZ2
zr<+^y{`ym&p|h|gcf^oxx(oAAJID1bQyB5$lmIM3v$Chc1t6&uWDV7npn`{zS)-WU
zH08U(B`!{hqES>&`91X^NL~eJYRUfC1>9<Qn$)27sSBAUqsVg$v~gY8EVSCjSkLM!
ze}lzP+Ze)i)`cJXB0`N~IFgXdBpz6$NVc*&3k8Z^5GSzp^(LL*<BSUM(;muh&1Cl1
z{~V`O-rjmr2tVUEXS{g(Z@$&TU-9FYa0@oY;XmWhj{|$a6ma{G;cK}5(W#$+rx0D*
zx@D!pvec-#L0Ox1iu!vHNKr@^F$o^vf7_ai>EjqJe6?*iOw+X4<t|KjU3p;wCY}Ig
z**IAfl%W>4zwL}Hgj;bx<SENeonbVsW0L^!2xsiY8N1iTy^90inK7Uir#uFoW17q=
z=p&jORzvZv*JBE#<<7h+&*V<ODR;!>&d@kkWH|~_B%951aCW;M3Is{vq>g4$f1ZUD
zr4tZclHg$G5G|oD2Q_}o>yu}a(D8^V<rRkmEc-#%MN)Aw=CY*>?w)CRDoi-ZH^~j%
z@vNgry}PAX$I&`dYoc|gQAg|YR2J>~u{$9VGKVW$3*E`V8mXGG9=Q{gFt}>czq7Oc
zu^$<T6(!rU>&I=23^{79CiTO%f3;u}YaIGf+adu)zK<F3+S;EJwXL}C5$uC0D^Xkg
zxHt7uR)V%q3Z_(J-G?hAsCR5Io&(|u!IDwh)q<@=?1bFd)=JRmzzL6v^PGj-ZshBy
z6bze<!CW98i>Gw9k*_P5wTLIJE$9_CP#l~hFW4V;IP28zr%G6s+2F8|e_W0Bcij|+
z1M-!up4NqCN?c@u1i%RscoA~xWQxNO0#0=theJm%#ewNM$W^B}%ghVW2+!{)&wu?@
zKbnJoLMrA$R7vM4fA8Z6f0r6x@?)0?JhTc{^hj3b8zzf8>@>reDjxle!RZ8Bq9!6q
z!obs+?(&3T4su@3yza9ne+=b$KPfOsyd3iFMyhj@0w?DqJ0}Gq|9y^~5YE}09K@xQ
z_<DT?iG)hbZGDE|8Vx@C1)8b;F#d75*ik)s%0-*y%0wyJI5~U!d4>E%&Oma?BP%u3
z)I$n~@ovw!JUOVckVC|>>4?k29=}ls+}X7diXPFaB{<fRc%K|uf0K!a!(vNOXGy1W
zlI5p#XHvkCf-j0CG=+{SWtMfkSHad%d3G0oeG-NOY)x?l)nua^DPLR|Bk>6EG7aUN
zp%=O4(c+Mi5;dcbu)Clzr>!CzT8c!A{0$CE$Rv*V=VH}}>OuBCevflfds)E{#woG-
zJh^h=ZvmHjfAIZHf0pG1WyatT;Ezy{QuD~c?5zVErs+IWv)U;G=hsk$$&7W`x%`Nf
zXa8rEyLxh{TiFg6nVgUQSX-u@BRYjtZ%rcElB0T&#WIi$NFTPd*oI<pdzV<KXB9cL
zlulV_ut=GFJNg9{nq6>1@Pr{02hjq_d28hoXPq7Ld|s8`e_e^Llft<Lm3WB~3XD1}
zDdDJqD-xBkQlyc*K%_*TLI=h9Ep?G`DotifU92g(`{(OoO=(RR>(LwNVoh*O7aJfi
z)J4f&m-TVl`O9N2l06mZ2u1SCu_fYU7N{MqMRDJ3ma6~>aN#Yrt)}>rU6&FCj$aMM
zQ`=s&u31sDf9;1QFxld{e8xF{$T_6GuWOZsIssnlWMZ>sv@>nOS<LzEtf3*eVht_(
zlm#vj<ow*4DQ&iAT8DkJD@90}4{_t+Nl&_Z3XK?*PGLhdYJzKM)T6g)VqfkMyjKZf
ztCc-0JA^>W=-7^40y&$a4FO4ZY&*$wu(~{&O7llre>s~R$fgK9;W%m0icKyBW62hm
z#h4ZITuIh#VmF^3duGvW23zHV8>~Z~Z8k-&(%MP3XDDvT^ofsD<nEDi6oNYq*!w1W
z1kZ<Sy8c71L@_VVknS~AGk$l|dF^Gdq!FhDss1@ROUP*nAz@P1rC*-ju9mKVwen$~
z4;4HLf7E_o>S^6aes#!dn4;ZuFN-?c${OZUS~$?R$950<r7>v@*jnV*kMddvg|R)g
zBgew-VK$mcsOJ^CjCYFI|3WV#iBIIcRn;h70%3F5ea^c!##Rl<>Rp==Y9xAN$By(l
z_0oNRaC);0bVe%Yt*`LB<oz5URbZ;tvDW27e>4x!s%px5n-5{hA%Io0KII<IPdB&L
zPYs`N{d2#!q>X~6waizVetJqOt63{uKZ@gPNY#|5eh<&!9IPQ5Bj)E)z3nf?!a}%#
za;1f5qvFuu_*0%j%d2On;4@oQg@OceV7%)$8R0keT%A*pCV`r5yUVt1+qP}nw)J<}
zwr$(CZ5v(ava6=fnTfa&_bDUwk`IZU>s!0H)7H_AC!ygR3rz-GQw;cJdx9H4=D<Be
zW9%3(x#qDi{YWn5eJAl(ZNp|^$)(zZe|L*XAODQ^_R+3b`#esXW$IDtUQs^h&S7H0
zo9r$mp4Z93oJ|>K*<QnDOGXl`XAt*M>EF4}i>MbRiFf(5`E_xVj^38ECB&DQ7bQri
zo8#o0h_K_Jo|Ld>aq2?UPF^ZtK|$xdxaosmpZ-{;-HG1xck1k#wB@r+=FdT{fqD@t
z2HlRm{(4aLUHC{XO=}qk&=;$)#mR53%x}l66(M_4(G{R~>dL)#sq?k^?1k(#`p7eP
z?_Nu26KaTCd0x86I+S|slb&u@9j#ZT0gucm`&q@)UJAgSil6JUcCHeQ#hQ$K-X}Ht
z55L%VuQw(m5i*mkAi=lWzMosZQ>|v{mt$%PGxcRXZ$wEq0-Lim3wX0`d#UD$^t1=j
zRV+5;vv9wmv7Dz;Z$<fk)KttrT?~*vYAVJ`GK$o)1@?SE(T$(i+rALlvOOMe^FvDv
zXw#&ZaF@xMVIF;8q1ub~^X@V*ALVu+39v24!))VKt>O~QbF};_{tegS!Qpdw^2NFf
z4O3_6hll;zSNcpE>1kj!!-l$;&$C4#urAB!cD&(-A)eQMcT9JvNZq{T?cw{yW@TyQ
zw+>8=k`@L4+f9u--|^l4hW-GSjJOJ#yI6!^=kw0&`RQPmb+rlV(Hvc2dqBcEni5S+
z(-6kS2=>VfqHLgC+>?ba45ZvA?S2q}9r`vWo3MM148}3>AlTQ9*ATIQBmyxRg~_R?
zNOrZs$hWCz?35`4X|}AZd)w+ofE8T<;`3Kz@-Y)2uPbk%ZL(FJH`uumMKJQLAmFIB
zjzv+ERX#Y`TwBTaWpSBR)BE-ILThvKI<oU)DX9ikc(>f8!%8!ph?pm(HBU(sVK$)-
zCBjwmjw_KsOVnW_GhqlaW;}cXBkNq!WfNyG%u`v)bG~I*h_M2fVDox)B^<%I$^k%>
z@5u?6fv8?0xkgl{TaoUu#zIplTf@O1<*M7`!r^b>#R#ZtfU3$YK}n&Ba=)KROC10y
z4FxWYffOttudir3cUKD|dYXOUZWI!jU%FvYbKjvIfk;6;FRip@Ejj`Bas0Y#xuUX1
z%v}x`YbgXGEUyIhbYI?p;M7pbmsFMG-kt^!8+9vA!WJxEm*1@{JpN|7%8V7h;_=uu
ziIg9H?2&w7S*f<k5(+5!Dbu`8b?>~G_}i2Yw!R?~{a7ZtF5s8WmOA?BmS1L(`2^I_
z`{Z<wI))pBLEQw`92eUC+_616^l4eco@o`GMyPtrUBlej*}`-^f(wq_TRy~mr`Qfa
z)^LBSD8<ZG@awEZ?~dEna=P@XP?G;iUOJprn7%Ju=xd_Tr<7OU)v}Ul9<Xx6+$=65
zmKTwCZBuJWyug|&QN;^1U!_J`lx2_J6?+rAxP6|tsqm}NiJ9v$upf<<=xk3vlxpC6
zdZjw@u!<~{N>+>87Q*O*KU1%Upl<;tOOgujmwf?}8!47r88y}AZA7LMaO-E=U+Sgq
zj;1);KkdvFne&Sl=jwDhi6^|v4U|@F>LoFg8UupBe?11GXgM?92bq@|{#{zhNn>8H
zxbbT9`-+5|K<O~)NCf2zsKgvKHQkb<{Z)zUJjjBlB(tyn^rryZmht~9XA=ro<vb2>
zV7Vn)fAx97^5t5VmoAg4m22T2IvI)QH*8YHiPdz)29Z>C*3o)Gacve|Zito7bqqIV
zb0_R8H!Jn+?0%5Oe%kM^n{LB_X?7M?qQOqU$}Cg1;dofVxQYBb9u;{?UENxv!=sML
zui4%hrSX(GrqBStD?^~#Hop%zMG@|h+n1Yf9-E?ZR&nF=Y%UU;z56A*VtiqDrlW1R
zygJ(N8YLXc%;5vF-@AT((z8sEBQh~fT!=*siz}2zRSYLS(ruXvkSP^D@dN?~!b1Mo
zuZS9dZA8CZ`k~ZL;^^u%g<UU+-L7Z;j<j*+g0ezC4YsYD*qelI{>+Cp(zRL|Ob-oI
zB*Im>>b2g497V_CF@>#YQDm7)AJ4Fct;~e)^{x<Td574(fAu)5WR)tpI)vWzT8dYb
zAT5OGsfj+kl85`eRbDZb!`yz!w6FMs(08q}mS!z%(Lm2~WXBr}sa<8KBHt{LH*F}n
zA6@D<qWhbhf7=L+w7>zf-iZnA#e1*{sCPL{BJ3r~_MmuIM_w$*lJ7~!-!OKi1lJm%
zD`qXDzD;>XzME;cu?fdzidw-dplo3%>B`kcUKWT%%Sbda4-Yv6AmD9L&~n+MZ;eT*
zOuH=LJ6MSjuqDn%WeUuf7s}+Cwe0!6-}U)>UtfD|4dF}&RQUie3r~VLUx{?Vu#v{6
zk^u@i(g}<VjP|HN{Rqv~T4{;IqpG-q6FR!srgY!)c&>lvSl>Z!Yi%8)!j7eOB2dI6
z<5hKvbY8d6%n$<{?w8t5j<)w<W|O<z_?TP37Fa3|Qj^npk)$($Rauw_Zc<gUP)*9D
zUf386;v&qH>N^1!8K=R|rn_0zJa^O}coYw1z^96$_*#`J@#q`;e-Tz&PC5)jbuQaT
zl$8Ev9EuGIBR4RgITPZcpge(oYZKw&4%kem6UJ&y%%_PD|8ocO8ro;6=iXPSJn#^j
zD?I1|@=}q{{q<1BQS%3{`QzRn5BpJy_AyF-2EuE}+6@cvbBq_z9I$sX!3_XBFVu*1
zNfNomzyst+kx5ICXCgEfoJY!!+C%(!&>rl%EXtw~_jr&Tj5{d;xj{K#A9WP@4=(B`
zYF%N=88$d~3nu%J(D9Z%&gGh!4j8x9AzRe#(Y|U_^B^6+qU`f<?umG3l<aaS<Z^f)
za<Fj9PA&l93qHpNT>jusV)N@o;}b1yzvA5ZT3z&>q~M))KRF6HVMNRy`GbWR^Zrx8
zp_NMMcL#e1Key(F(*R`&t>1+D;If2DWv+IR&V1w@2^%d^fy#x{kjUjlJixk&8et0e
zoaik^aa%TIJ(dR4D-i}IZg`Q^9H9q&_qD2P=0X5!-ea7mg<Vb%EkhKAq^kIgAGcn<
zUq^rut_GuokQ(SbvtF!%p5SKqC91|Scz=aPII4>1aGE$TzH-$cD#TDhf+0_5h)f35
zh1)D9v|FOD?{?~}_Fu^!CN;4GQB8;6H}n(vVjx~9y7>F?*S*bv56SDV7;pZeY^s;-
z(rdu@vj>t;o$b|E`#ZFivHByL2zgEnXnk%lbIJ~g|4dhx)BOTX_@8aZ%Vpbzfb*S8
zt%!L|OBqKL6R1|dtPWI8inSmA?I?SAo6BFH7n2<0)|UOUmjrOH`v4=<=!jZRjV&&Z
z$23ex!xT09eu^iKM%2hwBjk=Cj&^M2s!c#pACxZ>&F~n~fH4<Gc@rg&4I|{{Txxv1
z`x59?DQeniUm)pl1kBxl6cvTi@W&ilv7|#9W7SVziuE6wrUTkFEXr80G6D=JFKJ$v
z=TB>4;sMvkx^owM*y(b1`jNGoi&)ByIjzg#Ko{c(%BGaJW#WD2U}lM`z2Y+Cxkf||
z8<i3x4OxUc@`?5DU(;QVsfrbh$RLai%q*#jiRj3wj$7=bfZM^?+TGnVqPI+Rbn;_1
zHO^ctd=0@4K|2SBy`ebJR7-Hr7tq=@N7Zrjk=IrCgYEmj)4ld2-lKY)chkH4J4b;3
z<Wp7<z5fAazioDQKSUd73#^L;7e5_3UXO)|`8?Hm5?NT5cDU88!z%N;+<yvG^@P9B
zw1x|ol}iET`4eV`uvhwr!=L1B9&N*(rnX$|$CgdG%pUvE&;!izWXCaj@$oSIrOr@5
z;rj&EQ1>_*uS`_mk=v$O9ru^or`w*{!d1t6_uR@JW_%EBGwpwWc$5}Nj&bw<<6~p~
zkB=?>FCQyT?fp1jKS|U~NzH++3m9k3lG9r6yL$+~YT31KQVmK%#tA38VW(hVOHX|N
z0b~~~w)cH)x_w&ye?ayoH<E?|f;1!QuSq%I{{pfAqhKA3P3);OT!*w0+j6sJ<&f&5
zh@^Qe&JyokuFWWZt@cX!7~{%W;E0KO?&y>mXMo6maTMtv9HnMLBem;}>^iMJpwOF#
zUzoRYOwcJ~a?T*xRWGN9NLU&Y4aOWPpB`Z$PDdwGzg;6WU<>WkNW;z7eHwBTbX&xK
zGFdxU|5QeTYAzggT>jL#i3&8KlvSkobo(t3cnhM*yiOIS_O@L<q6rb^)MqDDr#c4U
z;CbC`dtW-!&6%kE>|DNDY);?lTcfFu0HLjdeUQgyLuK@6g(8zkI-xLV7kyn*bzPD%
zJ#NnKHBF{QCsmuS5D>@hXbZo-6Js%f<H-alnXR?F)IgksA=qREjz~mme5Q(b5t|wB
zVTaA$`@lQyQthATgJ@JPY*hZc)0_vm(R})yOHAC_Ystu!VlL5uR|gMFpck=Ta|u^s
z9L3ttI)jOtu)ge~iqSG)8$RSZG<TSuLV7EC*5zpkg5ktRKqfK{mqV=!59pdMBG|V~
z_=Qv<`Q@Mue(_KSzHB=*T2?VO2Lq!R*~yUeHrtucJSEqH!A*^^fL-(Kk6#MV(%hbu
znY9RR329mN;*|f#yJn16&><UzRDT5_XME81l|g>~89IZ$2v61r=7}GSWUuez*M`u*
zweOF!>mIeq$sdz^zr~rWtE3<?^E#5}b32hn@7GHFDYHnQ9NDexHq|pxqK?^gB^_H#
zk&Kvgs4=xMyK#svqrTmxGARexJxQfFR*#aTqZ(OL)Mo-aC9^OEqAV;<4mGrx2yK)f
zef!t<2Xu(dqsm0VkD;FuywDlGq<XZ`Z;Jf$GoE`ug4N4~gEg5#!t$jL8>QLUl{O;z
zDLE?ixp*d+K^BRDGBo4JJ}gBjSkN^h0&*lT<6@P6dnf)mbl2{tioXe9Kd`@ecf}{P
zx43!kra1j)4nd@(H<X+biSQ(->&6Rds?FjT+e&_70g(upc_vz)Q(o<xcULx1g9*Y)
z4cVLLJV;A4b^i2f<O;Zy%7}*Hr)Z+4%e#ae)<#<nm7G$0-Wp~Z^Hc6%ZYu2_@C(5q
z4MTzZC>jT608$vjdCUi(=?ums9m+NdFWyIZQVr-0{JKo_tBHn(zymE~{-e88iz$Q6
z#;I>WEhU(*`RZ^m;hwYgbnm3mUP(qQsaOZAS~VotXsb>tzN(zkDQB3C4)&+vLlqu}
zGx2VV@9)>X`1S|z_Xpo$3QDH@)qJe12)78O{nzpb>Sx`Ri}M12=s`hwodXpr?#)(I
zOC#8+e7MX&17xg;FkZCbAMPq5vn+_C_IONL2vN(~ZxRsF47j9o7!mrw&jA~-nX}~n
zFcimjwnp>fae1p|oG%<4F(5#S8X{eUOO0AtwTz`$<#=RPi&WbF;1!aE3l|l3w9FJw
zMduTDg&T_JC8Hf+w^6!-?(h~TPb?N*dWBH!bz$J_OYnF5*ZqeaKYXt9cWV9x554eF
z*4nt#I24%P`^i*E(OkqS9`OVG5i?mm&U^UU1M@(E$slfgQnMWtyJo|1gt)TtYZESr
zvY+|&sHbVsMnDVHH)_-JHo=^l{5y*eEF-iOLs1?)PGvg)th+!MO+mH+uA9gw_`JTL
z<CmA_R1(EKc`LneZIqe|f)$<d&k`eCBc8^X-Wu7g1_H?iycN-lvohcjbsFmKt|$R9
z<-<$UR{;HKz`_wT1f*$+#O~EnnYRT|fN}tauzW1@!qJb;;~>{Ki+!8P#V;dGM20ri
zVc=KllnvmYTp?*m6jV12nD#u<o2!f?pP`vovnmRv3fvkG9<{nLzCjL*G_zfl8R9#L
z0N+NCuUF3RoGlRFD&dC#!0)Z5Z4h6K0ZVr?B4+QG8<NjvJ%8dpUO9z%uDOAiYQ_^)
zlg&SB2FdB!@jwt{!E~&dK+bU@m1*=`C0f6={(J%Aw~N$zcXkY=czI}c&M8`dM$AhU
z5fB*7ZZb?j!A&XfO-GVjw0IG&v76<8$(b8uiVhv|RycJfoy!~!B_#m+9eKWxZF0k8
zTa>s0t#ZMk?$1)oi!(Huc;hSj1Xo6J&J`?2FV=M+d)WsYd+LYqZxSIJdN5u;{3b_e
z@vj1WM~jkK6PXRnc8`(Rw|m7)GJIsyFlk<Gim$%=D${aYw~di-)Y348FLKi`LB>!=
z%G`a;wvII^*4!ER4bbH}#!J50<cgB@yM~IAB{H?V@zOA*dnLMtMxo;9c6r20I%sK{
zAeGkYn;^lcqoo%S&gz(C?3iJ$@VV)bG0gx>uH0NahUbbWKXn<koZVeL!0pvPR<d+A
z|3zwQAxqKRR4s?P>uEBhvFhGnnX!sLKizRua#n+C`Fe#)RASZyXD}S4B3lbZCmU4C
z-{9JEQ6WDMWd|I>Q6~f(PQ(-~ogI*ld;O!^qWrCj#U5KB@80GmHB|TL*K30Squu}z
z)azXgDE7Thi7>)WHszZrU+%{48Jstbb{Jv9EWW@+3^Qmm!*=}Kf#yS3iGXXCX5Eof
zk5Ez{I;Zqb#hHH0;+)XUq)y=;*Bn0o?#w>DaYMphOX^vx?scr{G)TYKXpf%ELLL9M
zN>6lArKQ(65y2JeU0<$rd61BC$L_@Q(<wGdC!jyQviGvM6rz1zd6cbJdj0Ua$DiN1
z`LXMjv7sa@(xL$UsLvUP3CL5EPuMd71I!1X`ucutiJGpI0nxo9_*eQjde>jBtJnoi
z2H0*X*#wfE_`O@h;2|0s=x$Qs0%p&TH=XkC0IPO1&FR@AHhG84rj&N4%!YvVp9?I+
zt-gf@YaC3E5dlydln9ZAfnmONU*%<XNcOzWid#``6~kky-Vl$yBjy`124(`>qiDMI
zAcAQ<?DqKWu;%)B>~@M$(pRxo7d)`}vF3z47sbVg9>iEbLbmnz!8y&~N)4XNwJI$s
zn`3oKEhH!>LlRkRI>pbvgHtbOUJ5O@$M~Ke9Cb|k1i>Qphho1=OcKcaLw}PXj!P~M
zkIiXVc6RJednYwTAPyoW4J!nY!!r^*!$&0S`obBXS>c5u$mQZ5RG?Hlq6_8kL#DQA
zhgXmYHhs+T>&hnkspcAZH-wwV=O~p*a3pC^I5MReHH;>b_WW=M!=DulMHX87ktoPa
z;#f^GA5KW8z8;@Z2@wV@Q044*96+jZ>dB+}z8yCbb(xJMG-DSTN5lY`2^F<zWl+NA
z+H$|w!B1e#_lw!TK+&j7LE!=&`pr?v0@!8cDu^a#l2u4Utiz!~zn6c)D`#26AU1+|
z5W{@p{9Ux+2byT%4NKKUpKie1-SX@=Ws;e>!fy%nbHe~M3fjCV$r`SS-fC!QISNxN
zLw;XM7Ppv?dH--1*?RzDu%Rjv@|}KY#>Hsvx8rwT?bqjVZe>WlU_aC~%HSGC{zriB
zz2lI@5ZJrc)y<81VGF&4$-G=cn!eg$j6vXS57{dPYArO=Qu9rQ)ZgWQ#DUwULmZff
zhlMD9m2w0jgDA}bYz?Vtpx`y-D~%~ME(^$<e<h_?pi`o=X43&C`Q{w3fXxV!XN&2Y
zDVoqNW{+B><%f4x=j5TNk)}56d&2O5C?kW5?H5Z`Pf-X-PDMUc2WViKGVIlg1u;kr
z4eD;foqr^1-u~S$(Qv_-Uireu<n7S%QIxZ%o&Pe*u?&%n2(zF^J(ZhN!@PMJo7G}M
zMkJlm0t(kfhl>DuWiVrc$rSj|wT2&)`QpUtWSHwiENZe3rjWdz9jrP>Q*h~-fypf8
zxJew$>tg8e5_XPxw4<fawn`jtX=&ThXru7+k^WILw%N9v%BAvuL7zCNF<II5-1N$X
zoxTO^L-y;m^WnSYrD{rgJH{|zqpVPt?C+UZ7e}@);Y$HJBkYUwneK-}f7Mkt-cY|q
z{mNn7T!=G~o`om3r=3r$9tjsi=lz~TZ}?&t^e?W;WQGv8tFChx^(c6(o<#X-l<QHr
z&QhSf>(K&3jE{4M`s(ce{?wp2<k*+DhB#1p`UmNa%ZX{;yS3f@d)+g*%elBz<ack{
z=n0#{I-~{o$<Yxp%r<Vfg`fSr4q%@yT0f45`fxJ>mS|d$qZIk@z9M(Y9Dyp!Irq>Y
z&szl@UQJwVLQpM1Y>Rndzy96`#c*#SJ*BHmD#R*7Vi&2JU5}KWGC8{J&yzRDx`V4k
z)Y7s;#+x4%PPBV&?q<%Yfi>3i_OD0UQN&bpl9d8dh_X3yeHI0C7HstgHOHlD{aghm
z4lh@#a$c^3HYv&-(Hv<16COg56Gz`mQt!f3>UV8^U%e-|B<Q70(Xbs#bl>qVN9kXD
z_V3;_(4lWtb2d40-wJgp%OWks;jS@Aiti8`o~Vhn>KMaEvQvKgBK_U|M|-a8%M??c
z01n_t=aOkpv^aN*u^3mjW)q!0fm#;Dv<7PE*JU)c4k<p%PRQjL_B_NAlRgi>E}NIU
z7)CcSr|V_y4aT8p&cDqc&%EgHY`{^TkV(mLKCc`s*2RWF55m_76JSC2qkLdvp@J>8
zKqe0;BTaW9)<@|yLt&Cr$Ed2o2w@fX<pA)*S<7krYt)+2J$kUrTo@kGpA+BWp^mU4
zrCzXFWyK@+W?@xX_US{dEqd3-xF#$Qg)mqHllg)y4~^cP^Q>cy-j6@A)$W80(~!t{
zR-Z8sVMUE-$scWawvXh;x0R2du7YLx9Q`W@{F&k;dG(Z7YD^{8b?&f%;ca~@ssJDZ
z+uxC0fZW$_@}ap!7wo@W6vr=?@F0|QS~s}18e*GOykPc-NFc6PzAx;ISOBnFVq4F{
z79gPZGjQuqJjSC;0tw2sVx*Oo&ZDA5WltH2F@Ck)@QD|T;$1+uOzDEthG@4XKd}%#
zNKweIV&h_mU^g05`|UHySJtFm5Ch&D{Z{?S5BB$UIse7lLoF5vuku7!Wt<HO#{{We
zMBr6m+If~C0vNlF9we#t9*LZ^yHCYIO{*a&(xxw%#e$tKY>#JRp(a50$7GLYY;0;E
z9J1%!m{hVB5uKJ7$ofL4x3yed+c$gZGT5aCb;9*>wBA5CEgdwcr+oqAEr67qX|o|u
z1mN%$J&eIs7u{?s$#&(&;;rDg^r6we=EB@){hT&Fi~U)w$%3_hsk=#-%<=)0t4gwW
zneV7DTL^`GLOdokDmx|Eq%0+LYNt6q9=#s2yStjwE=ZgmTeEnSF)f3x9@_OxF$$h2
zGabcSM%0wCAZ}r`NZyZZc7U$Yh`dp;(^2!yUN<LiBV^p$C3D%*WzSlWBtefLdk|Tb
zs#^!l438%0(Ym3B3D)*~B@H8Y68oesE*i!Tw}Xg8!|?56hr#2MIuTR^T8OR9w=x|N
z{gXl`-ubsPBn=}GYmdin!|+H;ClZo*IfAfmK9}DF_1@{ppALi+ihzRqp#a9aXEcg<
zmJC(d`e?}^D_Pj(_%sn)2KalCJutHQxVSVZ>IzsHB4X&|YDqSM@XH|8`l25wS3;pc
zbTPH3gcC05jB$S?p=56R(=ktEHo#gDJ5B@{V@g=I5mqwktYk@o3oW6HhfO2b2(7S<
zImY5Xgw_B_aRPU_6X1<Q!f3SA^(Zcgr0T<KVU@s0GCwPfgy_jQz!+&V{ncj0?^sn@
zr%x5H-}?B{XsTftR@|dz18gpCDG}i$!<DEOkkTJ^i?If=A;c@j6XL|W40bU;T<%O|
z0sIt(<J5uJ^xJ#rd|MG0%o9xEayE#U2r9Xe+%7cY&<JWQ9dM<A;T*3crBvzTuS97q
z_lK7$d<0&5VLgwpLd{E%*XW6(1lEmfX8*2}yturjB}slpfR`(ZsedZ;5b0tb0yD9V
zu~}U%*WYUjjHPTeW;IzMwz?-ZACJ;PX8Vo15Ko)7Ce+J@gFX*$Tr>aSTQT<Ax@^=;
zKSqR?O+~x_51?JKs1OVL_;cErd%q9$95Gh!h{5uf!;44ghb%&DQT3dU^^h_3vC+^^
zQ-OMb2MJSJS8ce54Tm1$?V7`m01sPb?Vq1jI|4lI*H&R3ymECho|%{9kg!p=*_uEP
zn>M}Qw-Rf{hg-G3ZqE(@-km3rj~V-*VWS!IU~^RjdOokx>lwQv-0LHuyZE<fahkhO
z*B(=ykT781ptUc5QVmfk{+qP1{_jsoDRC-_)TmqbM8L`*&&TW6Ca`>w^GVb9h5cvk
z>T@Un4~kfbX~|9q5+;1Kw}<xy94gN8kW3+klC<;&Ag_<-chiq%lJztMf3)~qtK53W
zIlkbSK)Q-mA74BNGcSiln+U=&uH<`w6VCi7whFCFLjD^z6@Qn8Mo473$c0zLGpaNF
z&gD2(0ASZBbc80M#aJl=FZnMd=B_1Prr>&4Y$2COF3zPiCLTG-F?!MdZK4?B=Mmy7
z<N!A06N+dM&qs)Ig@1H#u1KaU0^T7)Jd>cF#uESX&t(xwJpPGP_O`XsGmao#kn!Ih
znkNA|VJDW!lzz*8N>;jH_n*l!QYNR>-;ZET5r8haiQlz)Bb6}3?)aK=9*T1J;(eP{
zOD9>n(F9bMW`3ORbbElMEpSr%gnp1I8?ZCguMNKYlS|pv%*Jop5AV2~^(OZ8#05v#
zvvwY1BTP9qS{^(e&7FUH%IN1c2DweP{xq^yB~i4F`7gU_2L=2dj0sA3f)4HE8qWB2
zb^x|~Z4=5k!nSSG@|(e?ZuqrRP|Y0q2y<qlR2k=bb1@sjoj6YcZZMGYz_tWWDYHwC
z3#M3jM83ixgo4U{g-Q@o`N5+yJHq1RX(d*qr`wMoESF6xqngd3I#p&zXt;f>^&B%r
ziojl}5HmywiZD4O__D{rkXk(<W#@u@l>qpw14`3p5S}iVsP#mHB~6#OxO^E(2XV;e
zU$c^B!tjkMT^WNnLn<jkDRE^{R4F>EshVe#1~?XaHON6^%q}$mZ{chL7>53<Kkn^v
z2w>?8fhMU*83VryjK$zlhf`eQgwPiJ`<;|C$E%x#mh_vBm>mr%`D#;XhB<30v;cB%
zroTK_Sh-DKjOOATCApNTI!n*A*dd2>4=Hn2;&zq{u<>_MWX7@>iyr7)F{`o6{JxF@
zccrwK?$D_0S(~)|-W(1%xHn6W7Z#X6DJsuvaF0_gN~PubQb9QJ&h9zW*yAx8MH=D)
z9a4!^tsp5Pj+JgIL~);Axyyu;jsZ^kLJsI}h_HyW_8YBeQRJgIzmm-_P5A;(=DclZ
z1t^E}ZZIJFamtus1F!KifT@F}BdCvrlfcdD*@z_N>}upiDKN-w-RJgg_tOMnN)Mui
ztp99OWl^27tQRH%!*h7&dUaMsF^5?G!Bcut_K`#I-SdnALqACB;k%FH+6EA$%6CqX
zwi0`_(}=A{9uX%Q$6WRVVpJ24pPP=aL2o6y>9=?&$n@#(k(a8my`zu3{KH_D)}uA0
z#S;h(nG<|bq9G4LKoCjpaKiGYa|pz73Pb?!QkuQS5gL`QuiR3p`j`5kF`q93OU=XT
zX-|Pn9eY@;>Z#eU*9PI)pa3{BMr5m}=*5^dwoQ?tij!kDJh?afZ3))DRx0hK8JsL|
z-4JH0Xv+masUKO>+mR%Xp@?YEQnIo|Ff{K+x5#i5BwdkjYzWX)q%_G{IX0$Al%h|w
zn8Kz$myah8Gki|}({Mo-67kEa7cQfsqERWpBd1<PHV$Cw&KTY9Rs-<wVxvOo-XSWI
zrXlDlr|yU`T~^W6X0Dof=sj6(Fe(H-x!K_wz_pSXzE;~ax{7feP2MMQ8vDzzaH!Qp
zA{z>#ZiWpDz*4Iw=9l%!ZsM}XPnzM9!LFA3rND)m_bLfpFeRuYGCiv70JW56g3>;L
zx+$$$?`$AXE$Z}$><Gwsva+A+ry<R8QRDF`sU1~+(u@;V1${SB?`JccXtbvhKvea_
zCP^^T&XXX?jC$wVaI(d6qt7yL_#`NPZby3J3vH1EH;&SkQIdvKEFLh3*)NI7ERj<f
z_9i?Z(nC77&NwM5)^RnNXrP}j3TdEQeZgt4S8(+%IEqVvG63j#X%S^x0e_jWLq3;a
z;ju8~A#Bfn&c)M*<1;<q-{V_fq3X_H3cs)s&rr*T<7?ycrfl5YL9Xkl1&w~Gg8yp?
z)XV7<X`HC;$RBs}me%CRLw<?DJx1GGbK0K63v!6D`&}!mqwHPh`<miUl?qdI{;OtM
z`FxB%`>b9aE*Ai*?YC)L$|!0l>D3lh;R1nWy1+ho%c|%!h)MT|!)@ChBrqkne=6&c
z9x53lL$UsQx^xTsBqzWoa;Sx+DxrP7fudBI#&g0A{}TeLZ)=<vw>da$^BZPfkn7$&
z?D}dG92~cgIWSN!uJd!1&_$X3@d1~aLBg@R)Sk5c*8{-*J%_LBIu*6Z(BSFguPnG@
zsGu?U?Nna<sl~2fAwfGN12$N6Pc7Np^%*Q}pYzUK{;uU*+E<RFkq<?i(Lg)ytKILM
z!%pGd=GetL7wyERQjuX1Ag5z~jTB2>g(lKC5ad)~KVK|B4V>L*Zb$*IpUlhEL>yBL
zOQ){&;wb<{t}h<59QQQI05lQpD_2>MzZ7I459uHVna-o5=%mAvPm|ek-=|^yPJfzQ
zknOk6pbwIO;x7evWqduBJe%^i)u4sgNs71cG{oZCB?rdAy=mYHu&3;iMB2~I&_vj*
z&S*02@GWzOb*ZDLdK2KAgM1SJx%4-B=*t8clxZr8IyypX-YYvu>OK<~a_T&f5Jl?l
zD|<SiU^j{T>+vcSJ6*EYB>>H0u%C-KEJMh=XmU;tnUGRkey+qmp2+36Jiip;VW6-9
ze(}e9e7y*_^y*$PC6YG#dT$*3_aI^blkVRI45?7m8WD&v5@RV<G}w30;Ij+H>Oh8{
zt7)x|lss|N3teRtwK<hD()mynw=?4;VGRo)U+E|YijmKcZYVYQUzT3cPFYMa)*MtF
zFzwKAuACmM?X(2qAN!HfDL<4oV;~krL*_cvL02^nbIiJmR){<cBBtPZyaE}8Wcy(}
zol^J`@1*J_gtlwsqBuk2!snM)6`HD=CRknhM@xT;hMKoh=Fi$=OW%XU;N;mLmhMUb
zn$m#<E?vu0ikr?EeYo)Y8{{Y}3$bxfibnJ5UFxN)1<vvlvVJDpY1Xgaz6ZHNVA5bW
zXbVO#thsD*l5|zvV5(UvaZfAbNgUY^Heb_VnUo}-9w`uRighG1&U713zwPpVC=mVb
z4vlU-jcC5j3s-Z1bzH@#X}N`uMHeBUI$krN#rq<miY#(**s1czcfYo;B-`b4&Wf5#
z{!t<piIYl=-z|b+w`j%7muzmllx<=p5hHI{bpBgC@$&FG^<%w(WA4JR%-a-%OZU?_
zT|IQ&I*qauhq>-bY_r0H{w@F6?1Dj|4t{yp2-creZz8Khj&(vIrM})L%sv;e-n!bF
z37|=1|1$vQ{<100ehXfBM(x=?KoAu)ILnXMv-1%iI1L(slJHc-PXW&l;I1Rrcltyn
zaipjgz2qdxB|eG2z%sTX@EM=+%Fz9(_raKkD3NF0xjT|p=247U;&LFFZG1>@BiFW!
zKPu}e+R)Db)4$m8Vzcm~a&rcFz_lV>(2%pRee|$5eI&&06%4`LSCCC->7>8Zl_XY3
z3WOHvrV1un!7~b18J4EtC*P1IY4}LDxXEdPy0@yBz%+?F+~63hGI_ZOzML6_SArH*
zWeY4@Lf}_`&j+UYXO5p+lu@kR`<`J<UoaVGp1?FyGzo&(ggaim3hxU5OgAa8jIDsA
zS)AwCtf6zDjL;P!c;`X_Pwfao9s15>L@Oo}_F2HOw*0Cmj}vVzgTK!WzrL2V{%Q7(
zD%dMcT}famUCJnG@rX!@9#|UXQ3ExE)mt?74tsHF57i>>a>QB42<4(>(du~v#?vBl
zfOi;1wEANZKqD&b)6WS2)n)1s{1X|kt%rBo2fDu`6fqI=xo(Y^bx>()qpV+>f-mDR
z;NMx$%ov70-vKM6C^|q;Q_ic6dm2{$+4UpRdh2nbnHxvSe*C*$t2JUoE-FJm_}JzD
zd+Qb~hHomAy*ymjNWxa(>l5e3Lu%(=*Nl4<ZduKkJG?`zeMk=*r<zh)Y=$1ayD3%1
zU;t7gQuu;#C>VW}uEIl(KtVRaE#1FO+mCjEHXy$T705#%ezOn1sNL4$FLq5ZkiqCs
z%>PEkK+No^io|dzsWhM*-2j76*VpG${v8jaZ06EWbpSTN*Qe+6xmrd<qNIr%5jkox
z=<9wo6X2hh#&%5-Wh|Hd_3`+yxydhHVM-&Xdk}h1$`9a&mG_e@aLE!GZ&s^7E}oO2
z^j_FtJl7XhJW!uOm4&%+**+tJ19eIKc(pT3DOiw^VpmcT3Eri@0c3dB7lt)qc+uU9
z8}xV!O`&&;Fu*b_pmrTrG__>-V_+`|Ia3Z;@$+;wPHAx}fu!N;IgATrglkAfSuBY9
z5S%9#bP{bVeR1=uQ#E%EpITUZA0Kqiow)hSC>19clw_$S_<=R#YikWty`u*aBpVEk
z$q`B{fCx@-(G9GQ0g`irPL(3K6lZy#Gn}wtq{fv$A$Z*6H-aays%a8gys(y)YnY#T
z7N6y^lr<x|KqO-(oZVrKo+$?G3pIclcV99}NDpshB+|KDWdgl%|IHHYPp0rSJ!p+A
zjpD%juZyp1q)HQTQfGK;G&LYd?3pb|78**l%&qNJI$fwz02r36;`On6cTD5$fSm}l
zP@G5NGGH_;SlnQ~Us`j0)Gq?=bzxlndnV!eA3Hi>=Q7V|P?Yy#L{9w+ym1g%nIg@K
znu12fkigFBAF2OjA@q>9{Sw{dNCYDv>;#EhUlCcN#770`cIJh9PwRu``;hY;csDWs
zna5%d;IS9D13Z{+#r7zk#B_z69Q@ix`Ce4kc~A?<Q<510C~!5@YYj(I;2mTV;Q=f6
zIOf<>kCPbM;4*4<on5pU0UgUhN?l`camVM$wFQrB+-6EEXc?%<ypDur5uCod7v-qY
zOEU%8w;tS|G^um;vq)ZYT;WpKS=?}ZJumLW+c#Hq1BCwyH@oM8G(N>3efVu~-*wA-
zqm!N&CI+>~It23d9&sF-r9kI4YqG`@-Pum&Yg$K&YstwwPCA9$_7)z)_}FX6?Y?PD
zu9Nl~aR!3?chgN%7O7;96=&y4<Ihg7X(uHO(TnjY1>M--ln54?Sj07E&@g|bF;z%(
zY`~2-DP3;)&)0U(KWhKP*I7xbVlXI69KE~dEFRwY0BTDL(Yo-2#k)H?#ek>X=}E|*
zg3e~n6!#vsg)t)2;}2qMCu2zDB#&a8xW%bSZUwh|AoAMTXqtiX-o7OUxvOuE0w^#*
z7y>64d;IdvM%nJClS@p@QBvCgu=9g|0boJ?^bEK-8pqe-(<_D#VQXw&+)P`vUA1>Z
zjI90X(Pk+q;sfsTNI8#qGb?QT-A%tYu{`D&mf(`ENEbq;SgtpA&JSI-pK68>bqx>x
zcaagx(TrtW2qRP{*`mY~=0vO|SC0(32;6Ki*%3!_jt0aQn-H=RIHzJxV8x&X2!J?_
zvw;G0bGYCGPIsGMhlG>;V37)4f^NxL5%?J4O<~?e3iWSupq^*_%U#CkWSx6_2*}O5
z3`_iq3><}XZKW4Udtl{Wa|9i~s-vA~PvgOsD`(F^6drCL-ubjwM6Czv6@~+&l8zJg
zdL=Bgml5?0f99_K<Wo`rVXdt!9H2PXpT;m<;aZ_{N%!h4jh1{?jO6t!WzlMyuwbvo
z=tECLQ&f;a2JZ&1!9J{!bBOn7pK3gHd$2);1Cx0`e~BH1Wn>hwjpkYJqZ-$540lD`
zLZsC_s%0|l_C8c_jG(A^G{{43QI{KN0)pPz-jA}4m)9h5rbHql$;YzB1VAXMexDFh
zoF%^FA|p;U9`y>=mmJ4YAS*}nN!*kyBbAvlg9KT04vCTW;k5e269+lr@jKWOdgvb0
zQ$p`=e;^+cO`?`O+#AHnFLt;$m}iGF^ZsM9SVGX2<TE`BYotiCq{6l9Zn`s6$ERD9
z$y7G6mA%T|lSNvqaGwQX5&(WN)~7ecF>uI$%*&sC*~N_FHpw!HdhZ4#$>SoYBkwtf
z52b+*-uWbkh<r5SWu>Cb#jIgt7fH%s;E_z(+o&G4Oo4=3Z<im-RCpSN;QXlrI^oN>
z0bM0MUByu^fYK#!cM|+&Eb#c0he8wQ#w-3KD^^f;WJlq+1qfA)8Zi3tG~z8uOvc2F
zTM~}`f=Gb37rICD2vO-&ZGkhw@@4x+7<IL(WqJ+O;Q{Q9szOD%l0KQa3vH&Z-Z{St
zbtmJjfhn{mv6R|Rl*qD%85y$VZO`QT{MqPFn=6SiL#3y(K+_OPjLh%a0q~rUv5zJn
zz7zQro^lL$L%aHCS-|9qpK{p>iT;x3L#8N4B(hf4Y8w3xQ8SEU)_w0$yBJV%LWb^@
zw1q4*;Z!?b2#jLBErQUg*SM2gT}M(v;u3^l=qpq!)qNdm+4LGUjgOd!GD&g7s#YVi
z@cg2HeQ1<=dlO2_-MH2p@Vr#Io2oscbw+YCBD2%ia@TNI0)T$M0nd>keM<zs*g&g|
zjwpNi2oGAh^2k1tKX9vJc>v6ie6tM{J$IEs>M%WaAT-^BCK11UU|@An4O_vzD2}=0
z0^H23*{_3n_PWKE9EFK^^jK5oIl84eRZVwDHxCvvuL&j_L;N{|rrMT;Q(IrkGWGWs
zLp>|@ZMB)Q3qV^T$nuEDYv@e;_-rc2`-m<S!X#>RPlpwtb1DLzNXpKb#>;;H2c>Xl
zlabxULiJf#7oVKDyM$}Y7TA=M0t~|&d^O!kc(%INzN!@^M$?N$;liCQ@X4<_l(2-0
z!RE*JCd=4T6l8eBjZ0|1LZo$5Q4&zHwj0%?_W;B92Qa_FQ|#a3b(wb$eG5$971wwh
z2$i75xz(W;WNq~I@`SouaPT&sQ5*yz(yT&uxHL!Y1*J>;Z^%Bvi05&H{i%J1D}c(3
z!U>P0aTSeX$px?eZ3nWQIW{|K7)dWl!%Fv_<aYAdgp@x4NA*243?cEV&(WB|4omV5
zCE%CJAmFJ8HN!5#$Sp$3d`&xBPECg3P{cS@=|N8*ymk28F8E|7J~RF>jTZq#b+UY;
z`H!788f#+dGLEj&v^Q;f%2Ex<UjA{jWqLzgFcVA|MnljvI%nUp&bbknvkDBP0T@(%
z9W{}=q4V(u9%nxKSDVo9P1$uW=tGEp!FJbD5r7?NaN2`RWmXj9Ey(tH+ne0Ojm@gN
zJernCRh%Tt<I>h4(>0~d7)DOKBmV#~6Q0-1M4SHeZv^*A#f}qs^z+zU`0#q-Iy%ud
ze&S)fKjJ}au2~m`$@o)$57Y>t1oj!qT4EHH2y}rX>3BO<dp(IderGBi4Yk_fPoaP*
zjsfs`7Zal_%Vj#UG|n9gJoo%cg&&)-{~oPz7Y|e(YTfB2$;B#d?n@5D`yYr-(B*GF
zkR&8X{Nj#Tm-v>JJ#Hiqube9ZBi(X@ja`PP2?YwkEFPKv>)*G7r)GO<Wj7i2hQLKx
z9!||joY?*c<wNtvN<>qqrP_41ZGAaTh}w(@me-QMS2goq+eXjZ%TJ_w{bWN-8QQwB
zmyM_=`C_VK0Ljk+Wiak9#44GPa@Tsh5R!#}afm<RDcMAge?oKB7pG*AGK3(Vu3R1T
z-p9COXT#~d$=srjArjf~`{?QB_b;sgv{Xl8(*N!5MbzaZ1!iF9Or7VZWdv+%Y1`|x
zBKob?63nGO2=d@Kg`^TN><zhE2?)YYf-Ycr#U0yTF3n>ta5;SY_7Xlj&0J4#ZA3e6
zK|NjL<m{V{y_)ekNzRnoWD)wHAL^=6f)Ap|Qptpfs{9L)W;@`)7A&XYBvuGg36M;$
zOp>KjotQ_7A*UkcDP5uC;{`Nt3@aFvC(m)S2L}YCX<M=+m8Bs8W68{~3i+6o-ULEp
z(C$=XM3d_`H3}j8+<nb`!eoCJD|thd!2}Z12uAXGGhUJiMqr8|T%cGj-*$v;ASqx`
ztSUi~2^@-`CP2Pb8L<g=)k5tAq0Gq`g`jAXs!Y(SP6H|GEX$;s=mSPivkmbpBw*?>
z1u(G5OMUz0Vi<lUBEphaUqJCG*~sQIbZ+$4#9A9ajzc7qB_QsOL>NM~)_cha4?)bO
zNHJywl~rai<iBwV9C2Cyew4N&$~==o^<|6q!Tqfjo{xzf5FTI%X3Hg2Q5lRG9<K#l
zmMpE*xF<2*91i)9sTE-JE9X?0*3_IKb_0!E0z*gON9RJ9%u{rM@%thzasxGs>R4rZ
zWKVhAwRx{^y`2kKBWA0MA#Zs~PdnJkOwnE1gVvU*K1j175sWagvG)2`=@_X>eB_&V
zHZ5dHToE!$U))BQadD)5MZj`m3rJm_h`#8b${|yD0hyEyHC=!zdSQ&%see|5DtJi?
zfl_29v6$XdZ9~uTnn|<&=7+c~I~9fS!ogBXHC0JT&2du}QFT!7ii{%Xk{OZ}rHy>g
zdyV&NE_~zl?c<}5VaLrWyS*8EEe=q4=h@c=?_a+@;F7%pZ^lloE~1-i1EY0L=`=_r
zd+4oD-6lr8EEm8jzCL|Zccc62_cdOvcCS`froQ{>_O3q9;`eTsE)Cz7-==GCmv-0Z
z%ZA=G;IH53%Vy{zd*g0rE*-l4_mko6r59u8>+13H)Yn^&_ov@a|6X>Lc<c7nw5mI|
z_SV60?#!?0#YTmr*;u^ny<PvOBooq-X3jLbzeec;cp`xMMdl|@<Jm-YEHg;5t~0-{
zQGVA?+n5Mird9mi-rhE8T0vPE8580z!yqe)^^?8PvJ?y6=XdsP`1|Mk%zOB{Ign-c
z?fXekn;y0v-ml}ugVFopV<WRNRGw^o`nl0XC>j#HtE2J}p4og}3n@Vi8~C%72zz7C
z3J)NzRC_>IkK>i#w$h=a?;J`fN0*O*uf*HGGsWi*Ge1{8?XHU)e$S2WEMG01hmX_O
zV|r$dAl6R3t(B<dkY^(|?r#rg)=phBM%s7Fr<3jD=X?3(X#@$x$`4SjHo@%*HE$a_
z&(5hDR|xW2t$^im3rIK8<7O2#rScP97+#;f+;_ls_J**l`w_l2L$>06gv1{N3+MoD
zLsn*emsnUW{u{rn-Yj?dDfTRlx(OI>62&M1&MhZwQ+Qlp^2|PV5DjIiKLqC8C2eQ4
zr)|t(afsD%;8Jv;9O$F_FM<d#?t790Gh#l(+ra-Yf;g!C-M|E8@<pU#;xB%rCsF)H
zR6_tHiB&5UoCMcB4)<uHV<2-XQgS+pLy_`J%%PVua5#fct}{Z(%_ED5yi;F3-g~8%
z06oaRZNPYS^gE9%D}qyrpyAksP($lMH2k_K*dTOKAU&k;VMZ1L>y6H#K}~yZ!Wj+6
z!a^BJ&$1HKO)y&d(KeFX?t`ly>d}yLSSx@Ys3r}##1|^?7b<D47>7Fg6qlPOA_YMk
znnjGHBu5m!83HLAPh)sD_c5BpHR2=rSp>e&ev#*NVvHpP+u^)8oVWz>vVb<t?dm9n
zVicR9C|_C>D`bh5fZ1)vi&!Fce+zjMs!&lPs7#3>?|0RYaq(e%$o>W!pg0f)%6)+3
zM<(<K_sQYE3qjDv-Cry37Q$&mIZ?PeB(1^fPR}IYJtI&mNWWCaja7I}i{j`LKEUI!
z|J*W%cf~Rdk==w}4_^L?0{)e_lHUbB4JjS{`Q0Gr8y0HFc4L(CWB7cm3DrD6HOsUN
zHD2l87GX;po*+*@N+R+yV3Xot8#Mv|_#_&DD!7ArDFc`dxErF^S*S+gmQwm$XSh-r
zE0U0qPFKS>`M^dG+yVu=`qfy1R1bo&CV2_`80@+)4^~2f9W%Y!9<5hyJ{8-JFYEg<
zbtW-&(#db{(;(VtZ^srz?UTc+F6WimQpx>kDQ!6I`d99K%Ueq8OFA%lVO{`#vnxQo
ze?#d5)dQ>Gfb2S&N4T4B80TSp(4ho!!$kPMD@<8aq!i|rfrlWmQ?mRa?XZM<V-8+y
zoDvsOgvahzS~-OnCX-4`L#~&e_RM??Q9S(+Za!{Df3Yd5y_fJq@!?DT2PS!E$MS4(
zGuW^3^oJ1_OsoVCBe1G(?Z+X=LvI9+rSv@oq1`5V_rqpCRJ9S;OCV;l)_t;ON3TQ4
zCY(Yu!ax~fNB`KE))qDJ(M|CRN7gF5<~9*4P_n1*j`MvXGQfZJcyDrsVHL(Sf_LtL
z7p5Q5Wu(K4INzLdbpz@|oHLsT-AhT-*m`>)BC>Oj`kEX9?njO1FF^pES34(3V<nPl
zLgUvDGCz~L`RS%AcDQA<?xjO1T~Ugnls&KCA(wE-55ICgvv>sj=wtaY+8RQUC;hWD
z-m=nSO$OjzXfjPuaS~M1hk74bv~hXtxZCD@&!wLJiG^;bf5876ne{NrWrv%Q8~uv#
z8saRw+>dr5u#fWHp|A$*P~46=D}>%~aDcl~${u!jt;Lc$ij!sj`O~$YlAxD|z#lHU
z4ox%B06fA&wK;}ort#3auUG+sl)i_jKW=CNZwEi3xaCZTDV{AI(iT|AXB;`2EfxZ#
zm*{g-L3~~uZ|Os;rcb@ugLcE)7uj{%g0F4E-s7>yZ-LKnT(6@7)tQra#vm`V@I16^
z%GV9>PLjXKmfLWMPoh10BB9fS$ONYvzkxo3?}0~u0L!>#i0xZVi`0u?<m2vUYV)Rs
z2jMxCXW|I}e#uGoBE0VPw<|mNe<9n?tnZH-{<k+Td1@XeM;c)NZgLb}o~e;ewC>jY
z<KTOTFXzh(iw%e(3O1XS0L7*E>lNk3+BC>DA9%x*VSBgNwU_^MEL=|UE<zz;1xq;}
zL=EGR+d8oCg(BaiCY4@I=txhh+}=fec=y&WR-(rkf>+j?xd|YwVId^dyO?C1QGV9z
zgKsemXvGUQ^93A&A}xCQ`b^&IYAW`iQ7P|)=(}~q<S#et?EE(2We`B;Wld|lb|<=b
z-^s4`WH;OM#XAz^r8xl33srC{^FBh+BdlJowJh{&KaH~93Z}<0pP;KjkJ3UNwN@Yb
zyS}Z{Q+Dbmt|+Fc$J^XCxLHRvLXLiBj>u|T(gstg2L-S{JR3UYIps33x;eji^}#eP
z&<}}U9UqJ8_NuNj6Jdq;YillqfCQ#e7!toNXKpfWe_Rkfx5J9!X1Sdw%cM8xd&#M>
z0HKg$lR}TPKSwB^WsW(f5vQ|`GwG%A;Tv)s4$0(@lEgOnS21tckhBqBPi&vjV;T70
z3iOSCYYu?3VWf!WdC)T3K?&Yq?bCq?3ckO52meo_o(Fu^<CG}JBNg^=G*<LL=YF10
z>q%5XrM8peEhLIM7nc?{7vbUknYNqu1z#zMQk*p3L6Q}e@)Xr$WS2lKQ01B@v5)Y*
zDG$(tt8Ri)q(Pi=Sid|E)}f=cxG^Mbo6)u{P7?si%X?^b2?b4YbB;yat5R0@%CAd>
zz)xroh3$6ZRHo^9z@K|h1|wLrtEH}BPwO>g0k5ja!6j00ZHQ{()|3us6Rs$p2}t-?
zgM?N(K~vD9LI%vAlOlNH&ff2{wd`D{HrykvHrd+@fq#MQzOp?f@IH3q_WnlaLYn#C
ztaJdK>kpP&VKT3sxhf}Y;Gvz&;k@l>OPR^mR>o%o5JfZeOk64<vg;FO5M3saOdLBT
zD|thXOrmKCaH*>9=x4IEvtTvuIz5s6JuSpb!eFF%@qu(_TumB$-qYpm^1iMM=j99r
z%lAu4VYEQG#(r44=UO<VV4+b6CVeJ2w<JK>Z+%}U1vG=uZ_38?X;e;1X{WYWYFrTk
z?m$Ue%(w3pPD1gxW(b*sqK?J_^}R7)EjYmk4bpB(F~I^V^ZOLLd=?-UrUJ&-q*cii
z$<lu>^Uk{4=@<V4p+H{0RMCELLYo;qNm5g438%F!Mb75j(=j_G2n^~o4Yj7&A13dE
z-+}#hf9f0^cZ$<pGSuue>%j*MRRs#gJ#vbRmYU6)!K$;M;{}t_L5^+vl7PwMZewwx
zj4x%LR6@8E4?e9G^auBhPL0?f009k95DIsKfTRgPCpeebX8meEUqSa#1?P*~$sUR%
zZru>n$!J-0T_)*R?v%|ex&KfiVGGsj@7DSVf0Um!FvwfF)Z1ytI#QZ$l|9j2Fp5Vk
z)dry`mShwclvFtsq^Qk~9<bDSFDoy1TIfJn<CepgK}(vMmUJrkQ#@FTT^PkBT*#y>
zg`0f+PY5<hYlHH4@&)Jw$&8Qg9f;3za$}6$^Y=QsMts}PA8AH#jU)~@7odC#z%~@!
zf2?)ya$7l)VU*>3%>)$+XQ09I0S62qz@CW_V$4k&f_BvbqNXep+wHUNM0SY_24`qt
za!4l{TupJy;Of!)dL<7>`kY6z8tSvrb7)ewY63|DgK(oDZcHEcht!k;3lHj?G&m_!
zg1C>Sw%mImSsZuNrx>lzvyn*d-~<=Ee^_jS_Txi}`6i)Z%o9x3gEFlunQrG2>bVtP
zG$d!pL&QcdoR4-<=(L{|oi16}q<=Yv6qKwP^39I)f(0qg>WcZVzx^gv!nB^B;%JJh
zr+oFzO3(Pt&LembsC-*8AIjvb@>?(8Q7#PV`5F*53erFswBoaR5$S%-Nddptf0|?-
zrB7(8{^Cq0HFZY)pr=D|e^87|Y6Zxu>k7SkPZ|MZk8<p(=BJ_Vr1Bid<TnU3=!6HF
z=VDC5?tJFrnrSir8qbWye(-9qc5G4XEslZ);dJ<<*9^dr^Y&)i=I=KT(-yvZp4@fY
zt^D}X#o%g+0=!JT#BGetX4?4;e~&tLSM4m1dX`7i9PcpuO2VD{+5KvQJNRk>Pz<L1
z!>=ajJNcb|@xUK_@nFq8p;UsrE!!YPZnK@xQOsGyS9fd0`cP9ELke>_)ViGItJ{c4
zJ)M@j_hDA{;U3xt2}XD{4JYa`Z)3GU&`l?X@+<B2ztV|OK?$WlyA#7#e>vQVVFdpp
zotRh_0X(r{QxiO~6H^o3=)}~6Pw2$d1UEV{Z1vCU#N5YKPS09QKCu(Sq;*awrY3qq
zCq@}Bz`=i@6QhvK%beMXVH6)I?<hJeymp+@l4<6WDjR*Il9C;7O>nW0l05?hl2vr?
zNh0Q(Dbs{`mo2*>oN7>Hf2P69j%@VFg?y#7(0FDcpHV!akk2S?6!Q6w9xmjsiGJyr
zen1)S+T5TF4BUl!5kd+x@LnB%tTK)yAhM%o5BBk@Zxo`^4aL_36dgsojiN6Y9mWuZ
z4j!&VW6Ah<6SZ^+?~{Ct%YAF8`}6Mod^!qqV>rh6oG%UQ7!EYHSDIRnhl{?7Vl(aB
z|EuUb8u8HRU$x`;3T^;lls^->TJ;id8`^iEh&$o`&KmA?1~1_rKQJ9d$S0nbQ*tL9
zl~ZyX=VY~!W8LnBA_4n;`*`{XEAQ>zmr<xT6aq6imr;`!69O_emw^#JDSwX~w-LVY
zujt%n0WZZWviOjt0ahcefZVd#Ljc={AeMvxF)RxRV*UFRU;SXSyQgPIa1J^2^h^(n
ztQWs3R<-Otb+UVZ^XBR1#e>qF6^_7S_w=y?{zLL7NPcLjvjD;<=$?M*et0D%8ou7}
zrKMDgFU#K@-|2X#!0`w$`G11roz?Ja7>lPeMp2?dU#RmP@(PYeT4!aLk5>c!5AKgY
zJ^lGcX4a+fCVUOI`q#Uso8NAFK~M=YzBAGZbh`V;FE`;24J3^9ET6C}-j@&E$D6;@
zvyRo|fYl_8A)2VkT44Y>W4ut|ucRiHM-ob@Sv{JW*AM*QAqp^~B!4FqFPAA<Pf}r2
zRuorg9^QX^`P5ZU9^Vf$dMT9E^^9zYvEqi&S{QV*^?l};%|JFuL6msXKL4-NGkr?m
zH6;lH9Wk9`lM!F?v@Er@LMwYssi!k;Nj)7MoGwZl?50N2$b5&ajVRr29#ONixtY{p
zn>?cS!YfmcZfVm<L4RI#Mj4^48%9=Jg;Y0uL{2A9#@a+9h*&>Jg20dM6RjTn`AxD_
z&{O!0r$IkaSCl*@dWK2Q0E&)}0TJTx;$W1Aj;1M7mhBUqiQX--k*3_i$+nNjDL37`
zGKOZLvD6vR3^>=<T=CAt+lqq**1{G9J#&Q@4;nfW-deWt7=Omt?AGWp?}zVh10_@F
zHicyhm{9!bb_JmaXLM3Ieyqqi!*}B)W~XfZ^Z34i?>t6jXeaAAR`u9hv{f`3HQ7dk
z9ORfpC`DJbS#a4?MiZA@lv1P{yIx!-c=~;@=`m@&9hH%c2uhhUeE(oVWVL~Hdj!(2
z#%m8)xHn`HYJUN<h>0jXKV6NO^?faZX2fA;g@)dWQ~jN*yyW=dd-Q<&k?!|{cF26R
zsTqT};i>?Zl_iA?b0DQzCr_ded)X2nQ3~Kk)hca>s`heQp(WT<Ge-Qh866&ucSc6N
zxa``!n)mC;abktvTu+`6iZ{je^rYC4(X8v~)a&U!K|w*k<z4s(C*xb2tmk1N%g#GZ
zdlGgel`X54eN73(?oz1B7{SQJnSwzE?(v*N0{GE*&cnwH<_4Q23@4yzO7i=cj3qc2
z0)FI|rX@Hb0oRwvB{&~{b5+eKQ!cne%?K=ZZOk85WnpFnKqdaaZtuV!e*Mqw9ooZ(
zFYA(N!&pfRlC{^=@!@pM?|DsQ4*&lAd22!TB_wDsxS$t@yw@s&{j{!q_M{MICnRoN
zEIy7nJEhp*%QHAt6t9EQ?EzMoQV;~PO%1*bH=Az#ukX|C=T(`1mEvT|S9bYyyOxzG
ztyyMOqgYNmUOezFS;Icw;owvs_;bD=gZtz-c3BdGMu~v@br!>W5jopjPh(PKiXW82
z;N;0;PHoyTtE4uXvo=({g!*lYyHToqwm_&NWmzG98x}54#z!`uosBkI<uvH89B{~<
z6&b-92I_jk=VirzC6LA$ObMQ|bN{x!H<hilP(VABy{Lww*y)V(+Rpx(g@btRsAfjF
zP_sxCj%xHXY9_(?LD_o7g_;TSDM<xHsMdGc-lsvk{qJ~0zPL_uCSSQu($Gs?d!u)m
zmuDoPmV`DlErb$gId7ER!3L{qJBvxE<XlCode(A;FUR+P3o1C8@lR2su9z$r>mW@{
zt;sZ`xjPVYbH_>$TDfcP*y)V4zhcJ;v~b$iGcD~<n{-``eoj?0L-2IM#iC{gh)R(<
zn>Jhltj{go&H*kUx7?3&AxpC`(;}Gmes+ozJ*Y=NV-N7@Q0URJe96Jp=k>D^q?QVV
z%QeUg3Xn~IOQhnSYnMQSeP&1M9a`aou4AWCY!yOiA;UUz%bYEg8!zIH50+%`>fF(F
zxFgHQUUhZuOoO}UBP*X>IG$cJ7pi7_4#F8PKw>@P62xVMttF};PM6y6MuQ>uaAvck
z78cE}E?%YtC9JX6L;h;Q4dkzvzgiP7A^&MvZi&Kw$ty;@YQinKr=y=!G^V!$O=(kI
z*zL<vuNGxJ>Q$q&&P$Oo`}AO(-CQYf3?1f7u1ayV#Z|)_GCH8=bB*>_>_ehs-1YGP
zKIHyBB${wM!vBe;);Z^jc(#aeb6M|lEbp1k%L&OJ)xAUxX*j%Lb3ywOfrOP;Z8<BG
zf9pwqR;ulKdlI~mu4d?4wI_jdvGDdoRoM&0NVe&<`;(!oV4a(oDZ>W!zTY1pCh9A?
z%iv)ye#fQa0Ul=OfpK@jIaO0&hL0KO<==~llwE_yjjsI>@qGT#V>r28o&6?Kf@POM
zKQ`@X^e#i?Kv>P#=!MpWhKizhlI=S1)*%3Yj0zozk}<my(vI73x7D@;<Q(4gOyhkH
zpT4Za`(FDP!uT77*ZF9<VY}FAb2`?@#;YCvzTTZ=_X!{{A;Hx~BXXu=S^5dd<3>M^
z=kdIXV$ysl>ANdS2x*0Ks@7Z|G3`iWX`XiQK+XGuW)`Siclk9E&s^9UH60O)kEry2
zyC8o*GYnPU6ypQO?$r=(V$(X=y5?q~<HWKA%}zA`*V{WS-G<GS30lBCHd7q(O+c3O
zZKf>mF4fd^L4N(F*sa%s9svlFa~QyaFQ-cGJA1m4ENC!1W@fSN8hi@pG{SG4PqcC8
z)m7@Y20|y@b1GGPtv_r35+6==<!u{(Zx;|BKS-*2|K2I3*fZ9Q)k70+Gz8t)V5?B0
zWm5Bq`tWrfYBWb$ZKfoc`nQEuW18<N7-GK)ejgSs??XtnE}sppdYH*~>KiVLo*(if
zX*)(!(st_j#<d&Fp?pf6gBVC6#oyL<7l6R<daaCMaN7NLZ3-=T3r#3qZ}h!FLts;;
zV@*h}s&LR|OYymsvvC_<m-Xkooe=6w0DiK@`pX?ExDKd2YjCC-2u4eM-B`lMcVXsB
z)cGvu%l`#A47Hchi8mDjGdY)WZwnKbI<7P*m+yW81($X!I4*xdmL-o!4xb+>%jR<<
zo7cPj)9(3EQDcSEV6l1n)Byh=`4c4HG`g_>!YF8-zBKP&NQs7zJs%{cQhZqc?)a=9
z?iDy*0VW^taBm4-bZhZY)+n8xLSLTFXUH5Jue8p}K##eB3+IOqPj7ZIi{8qKN@(L7
zV@a?yMNZ9f5kG%PffCqo$Aqi!OQiGs=wNc+S_vXk(^KK;aPVH}m?#nRko!JmzCL8?
z(^^H=s;8Pb9RiUeXdp`t)E@aR6jnubK>%Kxn6Q1k@CFSHm}n^!-!h7X)-G~+_USG?
zV(s)JS}4-PM}C%C-koRU1p@1yK#*1GsXyW)k59WlTN^;3u|gxpb%qu4>zCd853>2l
zKfY;LR_~g>;?gh7#VW(!esA9GeoFpnaFTbS*_Zg2vPst%;jMCd@Q6iW9si`4<19E;
z1Ct$`mpd&uXaRbcq%AluKe*DaD|}sb3MEiudo?PWT@`EB_)No;r}=2XM7Sc~UP0z7
zi!@<vTMs3xJfMScdR<)&!mlVo8PDMgA^gG*mk};F83;qm=IiICy7QMZE;u3uZyn=u
z=a*S7I5&U3!b~&J!9zCgyuN3Z%_-7)e}S@3s-Y-$NU_40AyqcrN0fPRr*A5&gwm#_
zh;kbBI*91JT0PR)1^hWF7#!&cV4A;oT%6U2e{m?2oy_KQ4suoWBeTY{Pegynev+Ph
zhz{XWJD__=IAp8nLWh!{`24a1<#K=V*4dB(?Z|(apeMJkh?9pJZCw_XYt?igk?-J6
zU**LunZG00CYXZBiXoyB*W03Rh(vO+ErJ$1-CU45XG0mdkwUd)C|4Be@(?n_cC<C6
zQz2Xeez<x~2<O~9BzQ7{8E0A=v3H%P6lJ#8u2*W8i-qWfF|<%FR0+M+a`k#_EHJ@N
z^yPnt&JEQhwBX}0?|~seb~@B)4}&{r+>r4IFB=&z%rk-%5w_$IW<^U_R$LWX#*`xm
zV)0MS8M=_VBJD^9>j9x=dxV*d%o;oPa)T==r_||964b1p<#hwQr|h%J^tF)^$<|K1
zwPVz5tF4gkt!~J37bErdNxEd{R*>~cx@3RoNOCSo2k(Mcm49+I(%i|}(4`EMuVs(v
z>*Cq>xq%^?euQ1x!0?gp?k@G+lgV32;W+1!)7;0K)|V-17}WJqQYeQ}owFm#Hh;IN
z#KrF>Pbz)B7p`CCGus*t`FC0`l~J7P-RyqX7g{0^2siFehD_SlvJL2ktx3YfLt=l}
z&kQbyfKEf{xQmzhzRqDno3$GRj4S<?7^R}P1}lWtS7J@^6j;burRIMFun=B1kXa}f
za>qW>iq5IHO}*6R$1A!6Brm)gNhDuQOQgf>V6^zS8gZaOUOS?^5^<nMLGA|Ph8eMU
zV{vWg`yuk$S{0*l8Awu`>XAk==XrmaJVg03ROxq)A*Ec?9a<~28R-bjbaxn%5JR_G
zy2Gjtk)-4JzXqx;#z8|6Gyw{(DO1W@ZZh-@%Ch2Khrs;H$};a&khZh?of;_4O>Gr1
z*RHIkDapcNZGp(z492tHc!V1z^_eep>tor7fqLB-mvfl+G@9Zn?uJ2slPQ00Xa|Y5
z94uB8<Fpog`2TptbRfZ_it?%HL>>=Rt5+Znlnhywttp<W?2<Ew#}WS<AP!=Gaj;ab
zj0Pt3GPKVoV+IWQ#Ei7CcFZ8PKvFf;GG<s&r&M^uHQ_ADaIxo|RMcEC+Ptel?iY2)
zvzrBMMN#js9v@beX2yrr>h*u)!;0X{_^?(aWz&WOb7`RA03w#AV;_@E<~QgA(F?-a
z`uJw5-o`XsL#KqdP_GVf4IVdZ1d+RSqiIWTt0)et>J{&vtS|sI)syd@lAjD0uX^>A
zUM5u(PgaIhM{52jkgmd{9<|gH*JVJ|p-n4@*BkY44=727?{@Xb772gipL8<y5MBKw
z8x!ozKFe^jL|WYMK1Lg)_KC@^7nA|9wf0tm>g@#OKFxu<m7r!jLAeEVt>w+W=U2zg
z%VP*}Pg|D9isA+@$}5U9FUqUc8@wp5D9*enulfI{UzFEdKmVe<TD>B<0>feN)c-Ha
zF9~H${G;A7l!p{I7(#zn6laFeHUHZUp_j>(hS1g88x5f=F6W2P_2TV@&=uuvhS1xK
zTZYiJxZ4e(Yndm9(Dlq~O$Owg3exo9CGW=LHt-S@3ZMaMiv!X$!pysokL4eFz47Mg
zKYMGSQ$eB+_ns-f*aSWSLi}g;rCVD!Pu^NBI(5a{AWD`*Ao@msfwj%n*7gCXKp!l(
zQF2@`if6Kp>q_`1zBUZHL&o7Ul~_;x()~l>^gp;sD5O6AjcKQp2h;w0yA`53KR)gL
z0io5@*_TnMHWUIgIhPT94HK92GdL=LOK%*x5x(nJbZ%=Pruito0AT=WtTB*V44o3}
zX%qXjhwLdof2;UvlFfdMX5`&Hv}UHeSuB3_CQE$!KJn?x!}IIIr;CEgGN*-=)9bfM
z-~*0-1;?94PgV$Kl$c(Bo&NpExiok^;{}{6g_p(e4nOtjqY@6^5GGyX^wENUc%0vr
z=lqV+ag_9Vbovx&28VB2XF1Ws*#tfl=hM%x|9Ieu^-mY=Cdah5MveG378r6@^Q|Rb
zID@9|$hXE>0eORSx^Hn-ApW$9PiX`^&2QpIxbg^iDkG34xJ1_kPCs3o*h1E5%?uP|
z6;E>@1;8~cnfM@t3x{bI;>Ra{?$gXdM5N&oFKJH%KCv2J_0>>U791HJtAGT{09?oD
z)5VY&EwPMKXrR@fBd?K<C@1C(iHbhHbiZC+AO81H;E;-G<0m5=Q;66vzdq0d5ICvz
zla$O0n11{3^zGrFJ=LTO{bx|?06sU!d;`8TOrz!)3C!SUR*wd_J=9P%QYauu?*)a(
z&rAV5YFokraQ5=&x9?MT=j)dOhm@QN<hf%|e5R|h0L~CK&N$JKI#O-!m(MjgJWF{b
zE~%!8iYxF|P4lO3AJPyx9AXfD(Cv`_f7qPGAeP*;Lu49aVK~G(#zi;Fzw#_<GR#}$
zTE%sQv7NM+OmH8E{AZgn=b7TFotIWNI39m5a)(d#9kP^QX|A)J=s5FR=VCl-l2Sr3
zYsZ2bFHT9pyk4kSSO?kv%Sq|SM%5bSD{dY#=jCp8#g~Lrc@o~GKw&pS0uX0H_MDA9
zOygcY8XiI{r*HDHnrVyyIK@<F5|71choHJ;oR;Hf?=*vY$Q%jbB`ZM^y(mwhij9}^
zHaHdmBbN#{I2{mXIVUt@@U8O@{8XQ--?%cDH#ayV86Lf-*qHB$1!hMTTcLOLx@b17
z`nk&aR9=^JH#jAK5@J`O=?M}9&wc^FR=%OUqr*F!Dkw4ilBS;}r1Bypmz!n+ngRK<
z8+bIkJSvVt!$)u9QDM>o>n!T~u;>J)6AKt~CU<R)tO!?>eU~E(?eVtUNs?n;82;##
zgf@4+ro?%iG3LY>Jm88-T?L<0$h#zuM@We-Pt7xgZ}Y@|`A(Jn3T>EDelg2=w!>&H
zhjUcn(9_7C;BCcLR*MwSl3EZ=q9b$qyN~xS*j6$r{S{!Z!jOS&DF#q;zzUSFE?5Mr
zIF+*<Wyu@9<^+>7g-LPu_5!;r3MQ_Yz(T*HD8z7niN&|_axs4u_A<KVw<~GOReYhX
z8WhGsz5r8yTxCYMwDQ5<FNv2M%0G*K#4J#t?U*owoUH=|#>FvHQ(Q9%s;5f12-VA_
zEYxb6{{+69JGE|QJ1~MtyHu7aINJFb1GTjzVr=B6ZZlYgC6|`>G}xA+e!kv74KRnU
zucNePpcV@R%<;RoDc;>+q{Zx8sw>7I&{<Os*P{tXNxn8^*$4LYVlEor&t+P8CrBf1
z*62BHB|X&xRib1%`urO}P=HZfsT~k*v1<hu%lD<IyNE_xnlZ8+vukTH`&fu0+H|~^
zTRAv20&0Vok~ugsL3#1wTN_%%WuzSd$UZ7(urK%7^;ZkOmYrJkZ*C@)eHxW86qgs`
zxB?c%nWM~FSO!Xug7Y?W>zt8ip&F5&Z|9c=Iyf+YSM-JiBcBEhDygt@a9J%6uP7y~
z54(=X7AKuz-iigdD_)3rBCoQiYy?19XiVkrwm3zpp*kwLoHcZBUw?jgHJ>@4^?a^{
zpBSvVw`0EacE6D=(MV?Ct|!NFN4IR5Z)p3->#T>eNb=ffwyCG5viGv3s&OZyAV;~k
z)sxbHB2?vn<e~dq5)XphrwfQlqXAnwP;H&CMVZ$N`04NIybEAF1gilf+97}Y&??_}
zs}TLVa(Ek^BpMqPAhf+0RCDxX&Ma5=#Fc15T{&~e=*%163um4BB_sw*QJoCmLv(sp
zZKaA{g;!phHjmA{a%Gvv#IR-Xz4|1i#@Vociu^Hs64C&ZeD<>qU~6cl{)p@M%mh}N
zMu>jaLGwemz1t0-^TN4g*H3TWRNQCerfoRttpc;)e&_*@b!4C_Vsg_{4DLDmo%4)K
zKHknAs+@3?_iyOrpdJ%+%cD*X>7nVF+Ueo|<{D~yJscX-B`4Y;`#l^IWIJT9hXZwg
zRd=)O_i#vb5K^Js&9N2^hsNGwd7y<u;-e1!R@mvZar>lEN;-yCJteS%u7gW;ruj5G
z<&ZEwl_0l%;Emam(EUtg!qWRe2MW5{K?h1yt6)0Nzgu?nLz%vPGDx+Aw~KXeuR&SD
zeK4x(WLcTdX=eGtU?@2thqKqw!6ll1`cmE@T`H!QcKkwOk~Bf@9noWrYSl=ORPy#a
zho691D))Mbh$)<@lmysDQn@2Ymb>~8miU`FSaSV*m}{*t(b96Di%mY6ot!X?HwEO}
zX|d8mls$TGJ1Tt|D?-XolR&FBDyUnsTnkdZlV+RyIP7E9pEGxaxRWsPd|~c?>2Qa7
zIgkMJ&h=1iH~J{V&yvMAgHle^gDkvbLfd5|iLOjEO<S=r7~L(UeRa5(%llapg@&2k
zPC0w-MpZpD<7T*p^mDI_u9K(44`qi*&DNV1eCW9B5ZMD;p?lTPkK*lmn+~CY-VU|1
zDYjN!3GK1#etNx~Rf9n9VeLA9(?KXM*Q$Ms(ri|sn`MYue6)RA4vC(sa^jp0i`=?O
zI!x{)thcEyO)<=zH;MI}c2(_FIX<Z>nl=ugqPA2mU6oX+KCcusx{u9x?-gb=(622W
zIj+WTAN#huQI(ub-w|4jxBNqZ&c)n$MQWoKwdbx8e@3Zw5F-;^!O?bq@uxd2i!FuQ
z?fsg%l6o`_#&S*Vg-a|CHn9nex6;cW(4Hv((4hUFwmq}s52~saWaoUZsPnC25{u2A
z`c-Y2M`eliH%N1^A@TZuwulY;-5vt%a0>N9+Q<YJ*iMLdZX@ewxw(z3Bls6-Bcl$j
zs;m6=bDNgn9h=Bn!aF@UWbMm$Xd!C}?sSlKv%g;h*)gtC|JV|fcWfW)q;;R}v6ko^
zn#U}9Fa_`4Ja%Nm6@BsV3W28&mmyp@6_<hL0Th?KHa;kqw>~(Y0q(a}KRC+)e+Ims
zX~-#X3vJFxa4cn$D7TX@2NlS+;43+j*(~^)b-q?sUI;DtI#8(nUcSyKw%CHNCl%b%
z&UT%z_X=@gXVzy|62=7OtZiFKY_FbXsot=xTRm1l+{fBQ^Kb@bSwo)vH7JYCy$>U{
z+EJD!C(tzrYfkVSV7Wp0^0dUVe-O0chH0sycuB`%g90f(DvBE&iv(arQcBetRUJ!O
zI9QsvyogL5H<a6FCyafvmV{A_KF}%U<v(vh%U|#^JChEI!!g+}N%sV#oLP9XZg2RZ
zv^l#FM@+JQR;{R?E-Nw1>TsxZFz>J7!%Y<T4c~;!TAh2E7>VTRt>=L<f4fiEgSDd{
zo$1(u0-7~L7u#v6$Oy7NE3Rb5&o3;cfka>@7^s4#llpQh=!d_5j8kO_uTOC@1?E%!
z_VH9;ZU)_)FvQ{#P+U{f4G`=7zZomffmkAW`R>LPV!d<cC@ZV6Ig*wzYPe&;6cmEL
zIrD9u*=B=>r6)xbZq9sLe`mJY5MywPZmxV=SGL(bWG(Fq!_Ap*>&)wHwvuPXNDaH%
zieiJVwxW2XtF1NqYrEQgvX-v48vVvy?LOH{hgywZ!+Lq<nRIwOT`nU%B`^K&R`+;@
zS(cU8W?|dYw>~K%3V6Xk9Omz>($Zkm{M+=|Q6{#xKKrlfv&o{wf6;6A+4WM{vqODm
zKR#7xXFoohQuX5v`*Jd%4Bz_lJ*Z?cs`u{{l^gB-QDjb(+KbHTlq(_=gR*$*OZOlX
zyt3{+1Jj%i%u%PkxSY(mA}qCVgIim+1D5IiHiPdOm-b{_PL<jT%-NLdArrk(F5Gs1
z_8`+qvqJoyk?BrHf96D~y~vzSxgs(xDeue;nLCha;)ce%&gK4eWKNXYi_Gbi`f=pt
z`)?@73?Ve!f2$}q*ng`i9@&4ZH9BkmZEG(qfu_l(7rG$jliR6Wc?+&K?#8A{lqs$N
z+WFFdzpdA=h_bWyRPQpMsVHkoZ)WUMl%=X9UZ4mk`;PGhf6cJ8pU3-g*ZPcIJ~pg?
z7b{`oO|D^<wX!T14jG}b{gQ99y*NZ!>fAn=%y7^qne-%2JtuJx*?l582JVgIqJz@D
zZSOKe=pfu0NUyw;F;#AFl9^7alg#=-Fv$|+<+g-eZ_^8A5~+LGuKi^!j+O`P^0_o{
zM$!28rzwHDe;7R(dp-2$BYH6fF*IKh$#+SD9>#f9=chDV=LS*KiZ$O{sD~FD&;Ge6
zo|IRHBf+O$uAXNOU?(2x1BnGeTC3_%AI9Q=j~wc&zS9A|jU1mNPs9z5TT*i)O)Zqu
zi_5>KPX@}IzIHwLj9euNd|i2Y`uw*cF=nHH3w^8Se{@F=J7@)RtD-U-Z(DQ9=4sUI
z?+{<89INAuaE`}CneysMrY<0r<q?DhN`ROZZAQqsDY*~da>%f_I>X8e=a~Q>{`h#1
zhm>(Fpqb9VkDmx}VCCHEM9&JvD$|K@SU4XPy16<({P6Q(@q*k8^Euzk4C*A!JMtAO
z;yd4Be*!0N?Y9{|9Oe<@t3_Vru@ODu`zx9%KbJ_>N44VXy(MivU$kXB^eD!&tDfBu
zGRRW0enm(rB8qUPz>-r+oC@8?jC+JHD{<o`jtnIc<Whhi3#G~whZLiDND5&cva24Z
zV3h`OWqbgfJ^a6eSY?gr@zIpWj|Z{JAVXm;f7SmFnp@fziFr_)yF7m7;k<K+0sBqF
z)=Oa&P7y}3wN&+US(5XX{0H~XTxrKLZOHXzUasX}=Rt~!Vf!?dOyQt9K%^|ubmHK@
zjkroNaGQ!%0%1zl#kySVxKYcPAW-9q%#Qq*`z+L@f8b-)g;q|WIA@frhuVIYlTaml
ze?5%hA(b7E*=b49)VDkp=av7bAa%ldU%(V?sMOkJ6xTfgjrtDv@>>CU#*rvc-Z9Ct
z)sdj5!IRLo?iPwCi^T$SL(p@6ATu;=soEZz-q-@>lVe+Mki;t%t_kl13wgOocu*>s
zNrn0KlgY2+GBR^M6F+5BLA%)B54MJk70tm{(|xkkZ@J~kll(0u>Yq5}jCPkHTsRd0
zF_!_WE)$o>dpszYdPg{&e}K;aeEt4TX6PSZw423^N@(Lx#*$!k18&AZFH&lzu92xz
zz(LRa4Z<7#d=9@4f1xlouDHmdvXV9e9hQ{>M~E@gB~o}mqXIpXSAmRyM`@kbi6g^%
zJB%LYQl3j0fBZ0~$YyK~=CC1HA*JcDMM{^vazNqoce}b%)F{BHe+jiq;#k7W0v>h}
zp6W~Wo!62tJknC(EyO*P9bV>lAHOk1O%+yOal_S|3zQ;XahU5f*Yz1p_-l>va~OpY
zPZ=*+Zd&Czwx*xv)l$%xVa>uxCJ*xb5LG;qP~+6Nbe=9RE!;GnK?WgA)Qd<cv^~=m
z*gQWfn=7VOD^_Hcf2$Oy+Ik*sI%b-LI{vPUxs(;-3>A?g<UNK>53^-7`=m-6=7~mV
zJE&4He*KhD%ZwpXO-3)g33eYI-*w-fU+?}mVH#;IG@SnR<6U@9poOvi1lkLdc>3u-
zr>}Q^?WtzRBH3_G`G=OLpME%XZ~m73rkA`yX^htG26<z=e|IBEu-yz|`;v9!Wz-~u
z#-oiZ#NWn+%Nt&JScNT7DL+}|GO{n}{b1Xp-;5q(irX0h*b`J)Ym1(#N~`fw8AsL^
zD#mG&P8tcLGv#lXX$CqR4w7uQ`Mt$#?U30)S1@~#4VkeW#mLm4?Y3A{#!dlb`4O=<
zZ)%Y_hyk^ue~hxhHH`LBC`<k53Geo=k5UQXfY?z4m|Y(^{40->^<Cu*i$^Kn-4HUb
z!bNPHER3DF$P2H`9A8@b@b~A8h5bR+Ce#T_^JYcA&}u<u!iA8w#1(L~j+mnCGc_f(
zKvF%`5F<kXl0~*!oyH##ck`y!tzrijSZnlDTax4ef0j74wx@t>&QaZFu`Ea8=Yba6
zQq)h^Tc{!7$o7<$EYvEYwCP8y=bNl_u=|$kf;E6AVQ|V#%JK~#+?RZ&q%w^!DOMa4
zrE!jKybAR(Q<_nwkmV((e*+T~aTHf-g9*17wjwL|d@&Jq7t-i+32plf+gi>(CgMm&
zoj&UMe|m1PkhPmAT6Z2AM>|&HiAKme)~==)ODv-h5ols(Rf@fbMTDHqnhE0xZ^x%l
zI6Gle&X-%o*=GJC{t&IE-lh}AHR2_v<os0Gzr3BICIK{2Lol-n!y{r+yreO2h(d}f
zdOxlkjz!f;B%c)O!&`P4<-B65laWsG>9$NXf5*OwUn@>6`8PL`Dn2b7N<v_sh~o;j
zC@CDL)*9J6&|GX@$F>YCI2NEWW4Hwzu<RThj!qwgjx#2McsM#UgYNd+hMW>s9(fw+
zDBhA^7%zaWy`#94U*sI8-_aZLi|}qfX^6I$@&>%|h-1C5V0Uf~d6c&>VeWzx3YdaZ
zf7w#Dz<{{Xl*%8rITbQPwO4W_Y3N4dVgWFKz8VvuW4oNNa4FK{2;GfNsnOegM6o)H
z!je9wa*_NBrebj{b#18joQqiwTZ!bgk!w{c&qJYQO)p>C5@X1-`M5|1I2$%cvu;$h
zwBJhC%2{_NdVPE$IBB*_8<LUMyIMmHf7ts!e@MkfqWLMD<r`?H{OLohY4bHe2CE#u
zX79%n!@(ld4Wk~;{bkG;7b~8`_|dB2d}`{-?@%$D^}_ioh9pOWHVPc0;TZ*tYMIGm
zxox1L{p<Gxe5xW&JX@fjUrZ@iD|-eV>)w~K&Rjg$wKZPFp^OpqMP5YVfi`Nqf88_I
za~DEzb}!0mVZEvy>BzV(uH}dTZzvkZs%*D7e3?l|-+mY_RfK+v<F(YDT%}YDJ1S#A
zAYWNs70^)%;AP%-{}`eedzNmMd7rzF7vTivdxA$4B*S5i^sS<@AW%*O9-Z<aE}h+k
zeuLONyJ|)6EVfcc?ZwvN-m2Mve~UMz$ID8wCqe@^%1S9M*t5*a>N4jz2YTzb)`}Rv
zn_OCldqh%G@qc{b0OWZIWJ76AD4NOXn%w=@9Fz^t&M+K9$uu*5{;;}+36%+KW?-IS
zX?6x51e9#o_<LDHoeZZ6%5~RYoIh~r3Z+NqQx<vu@3II-Z{Z!NmXn;2e@h>>&S`Bf
zcicmn;$~OzeB^MdUFDlqVC-~w6@Lvd40m{BTVS{fjJ<&XrLO`8;ieGUDW5)lD%e7*
zHiHedAa63Zq|Wz$SrtF>v4M{>jWB=su$qS`4n@{<Iv!k`BLA3LJ(9MMciB1jT5PBY
zI*7g5<V%J{9M<ql)lDkxe+}8r)R?vpy#nQlb#1=PsE-{QS5fnJnRPt?1*@nzu@w$_
z?$u2Wqi?8pBA-@01?L9waK7JDC~w)>x>-*#uf`-*%0ZOlgkQHNIll<S<hpe(-+p)i
zW9~nNhfR;++qdcQ^QvtSJ8ctOLycqIXi0hDC_oQ46P-7i=O__de;~Rm3?tN<5vF#r
zx(eqqz4u+08-+FtrRF{SJim<QG}m6mFE&WDdY)ek+F?~}V`^6UW<xzft%BUK>Y)K*
zvGt1WYChfA;D7dSSA!P@I;O7sbj{w`Ds>ewWW9AQE?iYxtypc1xk7D+oP`s0g-sc<
z^pWLusM6gxwqQXne=xn2x9zOq&G>EU%sa0j$Xw=t+hDu@?|OJb_G@mko5kyq>|}>M
zkuUo#IOCE2sc!0;y6U<?EY5j8EB#nn$LwPrZl<fb`DU}K7cZ{pYOK}O+fslEMG^ME
zJufE7=aM@M<(P?H>})|(Amj7>J4BH~p95Oz5Y4h&Eux<ve}*dLpzY&)_tq{^7ejJ~
zZzw9^iZv{Cih3c{gkaON*fxr$jrGl<J4G1!b65#?b`frhNpspL$AhFj^U33}c$8vq
zgM8ZLv2VzvidPGTL5F;@v;DHLp;OL>b7waxE~mzt+K^F>>b2$MI;UAW_!?6#-{#%U
zma|ySAXjr(f6fg&=O5jULvcM|sOn34=u}!vZGA{>VU~NaX)eSk9%=n765Sv%n@EVy
zKSUEGr95(NE!0qSfPKHK1AhIBw7a4TxdFX%yKBG7&F!uo!QV)`D}oAtJn(|Kqj+Gu
zYe#vb-L)HiK)Y*4aiiU}U;q8uU3cQj?XEL?dB=9we_m$Cw7Ygx4`_Eqi2|hk4Ya$0
zlLDMS&|-UvZ79U+?bY(31fZw8WUtA6c7t!~dv$z_R6WH-b4|uQ!-V7dUWXg?U#{;}
zb4EzG!GxRcB*K4ias2P4d}H4$YlE~Cag2ShH#c{LD*#kA;bVHhqKs1$LNCL^uGSCh
z97)&ne~0xv#*_sf7g#gjDFWIFw0LL>LNhKI-RE&i+bME&w@3|mXSYZ*rSBFwAQUKe
z;QGk!z@DN$DipMIH$p)>_=r%@Qrrp!9TR>pp<tB>hlGMo$_Islmg<qA05zxT`n_U-
z{v#IbjRio0l=(Gcfqq*or~vPc1@)BUV}VDZR^iBhg6jyHgCapoc0CZZV-JV}Ey0a2
z&@te55(QRSa6}O31bj>kXo(&e0vzzezgGk>e?)+t5x}yd{TlAyyvh9!Xm|Slar~~9
z@$UKjdiQ_%QgA$%AzU~Wmtp8UJeR6cIGKMGuT%Q;!^htrWFlP(Z^Fm$R)4#EJpA`z
zU<6QT<2&t)Kx4Wee|`u*kkZ0f-+2^<eD}-0yDty_sG$zHByWj-wCsNQ+Eq`!OQQ6W
zMJbKZx)3E8p|$Ibb^;l25+!|>QZv)^&aX4SSWKV_CO^G9H49NQFX(HCnnfkwPtSiw
z2C<XP-yr<PVkO3=ZNwlb(4PPFEEP|fr2(6F!a1z5fh=gvtAP~<J=Cl55;>gEr>FE)
z-vYIA`UEeV=?1F=$VTp!B8{uHAk)yz6l})pP<X3u#p|2^FpYM|sjTkc9;FsQ&8aCf
zO9e_V$hSHwE2GocjemOl#gnPlWUaSAR5;xS1kmCSG?yV<I2M<}SUD1xOQ8ZOmvUS<
zmVdwpn9sj{{_Rd?)TQty{0v|1-TUX=f9{5jC=}ZGlhIaygwt=o--SO2v@q76loB53
z^vAzW-|l{Hp_a9p??^TWOg3qZ))m>55n8*GF(}a5DcQ8*i#MrxlRA)%FFNyYlkVeh
z{*Us#<c(8#&bJijJ_m}1De<w2k0M<JF@FZb#~<HM&6BS`4J1*D_=#_kgtu@tRA$Oa
zAp;Dh6V&PnL;dpM<s2y#2GaJVI2RIBD=iUJ5%Y=vYM{t--t`>67G=7Kg0eyZ7Fa=s
zDG)nI9oRx&zKgU@cz*d{W$?U?GE~9TS^z(q!o2_Xkft!cb5o2IK)Rmt3-dV7GJk4*
z7712_de*=H`tS(G-2ZodLB|U!=oW;B)txybtigJgDhtu<6mLYJc{SzlWeSUe@D^Ro
zU#Wy+1vDsyQ!Y(`KNrk!b`G#)u_{+{{<*pt1y*P>HD=%QlzC}UY#KB}Z=^+jOr<0!
zbWm7&Ew)V;X`(1Q?Nc{Afou+fYJcmj-h&klnoYjk(W<E!=J>QKePp}>L{qR$bFwT)
zUF&v?p6x=zz5w_2d5uma*+Y2U&Pw>Mu1CO@sgba!36=}%F)NKT-FlR1V%OJ15?*&}
z_J+2zvX4eaZzjVp73?LIU-G^D8djgjU9;Cg=x7CH$Lu-Q?3VN*i+gsfk$=_Ew=kP2
zipqONJ|O*)5YPY6gdyt&QhN|cwlk3DN}!5QgHbTG&LViu#uJKw+NbG!m**F8jhEYG
zo&qN<*cQLI%DnKeD!3+r5?g3lQ8LabC+f!E_*SdmPS$CQf<umhcLBR%BzOglg$Dch
z!%l~s8u{}K#y@2-NVZ948-HRmVLv+GJ^)269NTk^z{Z<0&{wJFM^jkZHlrg+*j}wM
z4>VhC7T;!O(kvA<tI@7iQaBLGLj&#Q<8>Vehr;&Q2b-Q&9qsg{9SUBEIofWAC*ylS
zpMl82j<U6EsLL2##<=A%K}+PYu|?*wVc~#lFY~I=hnQE-Tx6c!$A7#U+QhZ?Y#4*P
z1x;g};24LzW6*k#2QwFC_LDid0yZr5?2ehm15CHtGBbMwnjI6Fxh_?P#fd}U784bt
zgCm4+G_h)j2~o{nAjHBZD?%&~>i{7dOpX)6N`c;AN{DLqF^MocBtj2_*vwwCs*`3{
zV)m+fJ^F~MH!~Mi-GA&-^#<CisxP{KX@sE;vE#D2E_a-08Z(3NDr&|F)NT<sy#u#x
zoj?v7we=lS_l3SlBplgA9L)|KX%yF^1sKJS(s}F(FcR4ZxQz--DmcUsovRzwldrid
zGhd1*RT&C}4?7gI=q<;JA6YTNo}np#lojHyapCf0QdR+N1b?FMRymI`Ub0@hiyt8c
z2h{?MAt{ZTgDL~J)9OmeD1*M|&^T8$5`%L(kQp3c1~{tJRkrE9No>ySwZ;R)UL->y
zYz?uGaSvC*f^`a2&IYoQ^ETD9F9;U69^KLIKqZcaAJDcC0cO`%(y}j7-TH2<3KZ%N
z_hTA#2(Mgq+kc4{dEwE_(WT8n^n1qTin^lJ(R{T$^LYg|YDcUa?^89D94rvFP+4VS
zA$F)Y@?Yae#JhSjm94A?$myl_)Jsd^9KaH#*7g)In=@2bX)GGgCfVFjV;hLUGNoKA
zS|D{?U#%4epoP1nugQEigRf9m;Go<&Y#}bHYf|o(tbddl9WJ$1N6DVujpIwDBaS&g
z7h4`{h`EdsBzCR>Fw*`CA|S_yjK~HNuF-I0=~3+eTP`=gjb4VT=`IatX^^&KN7C+5
z!w38?&~%;3QxnKjZI7B>F|X%b*<;~a2)yKLWO9hTBBB)&FAoF~LTZQfP2o)u3nr?^
zA-$nGh=0a9T_VU?(j=EZCOsT#TFsHJ6;`|31WcJu)wrZ}FXNQrCk7^+8<`|HY2j_e
zs<ynAQ<^El_kYQ+C5Yyy@K6Kwls`S(+!rX$HVJO&3u=h@t->MmYf=K!Kw7f?mJ>N`
z+R^xar3*+gU;}l*1%M}ER8P&Dx`6OI3FHrR0e?Zy5I2Z$l?w>c?Rr*U=>j->(57R@
z=mLUx9H#~|S-Vcqasj&)2@EkrCqF|{y`63@k&HriK%JfiV>Vkfn<r7oHPLj0A7ZsD
z@Ixh>?Z`BFFF6X%Ht`qn51C7EZ5n1AVJ~rZ6Y3T%#J9^KY7#)57&aX)<7mh8)uO~)
z5Pzot+tKm$QS8={qlqBbya%;|$?$npu~DU*@qQa8yBq~1`?e4u1?Stb;{aJg8$rku
zI9~S(=MuHcL$_4>yC%5I*eS1sH*|CC%x$o9oD@iXbL{L4cGm|t9P|q5><xpP2I4gV
z2B#;$>fS(H3NUhz)9&bk0HgGr&6L%kdVfcHO})$dm?7=koN}y?7CPcCSRqu&s%!~c
zAVAz`7-xnAdf0zeq|lSt;3gy><v1jSog^quvl)F^KjCp)+2#;-`qGrt#?gl3yH0X%
zlw5}*BdFWvbg6S@XjHjL`Ww(lAdfn}RgoUWbWN(~(_w{9uvxi1>52f3``wViI)6<*
zlp;?zu`@-krnD*YeB<bP7f)zboO(Gu37TlK<B6pm`lhh-VI}^SJ2mNmO4I{<d7BX5
zRqe7{ok8G8NIP~0y;Lt(yH~EM+XV2n1aehF;aWhLM{GCwDGpo7VvE^EqnzTnLbY?Y
z-oc_4;Gvy$+*Yn*=GhKWx1G8*E`KaVIjA_-&1pklan%|(2n7tSdZn*h=@6^4Y39pe
zF_+!bugkp*XgNVO9SeHV%ew9t#c!)efiUz}>5e)(!;j(wIh~g5c4P=`II#|81CEH2
zo0$7&W$%Mx7$X43K8~qS=;7VyY>1UlLWGVA2j)ju(U+0<!x9+`PTPw&8h_My%cyOG
zcxZqry^Z!6G=)XaxdqWpcyRFbK1j!YTK}N)6%bNbT-Wj4RGgl*NK9Dn0zMq^q7@L#
zUybms&sya7wJli=p0x;-2GT8*tIk?jj&_ZkJ!*kC9JSbW%z`9N!R<I@fsU#fyAD|>
zU15FyTRvm)zlTRO_wmMV9)GxS0(pgI?TzymUJF%IY~#3vk|aFUTiiHp0l*eipEz6R
z<f<ID*g9^3z~NLiXXCI18fKQR@Y*<QLBPzijkLrjF*I{WEs*-77OSyAc=>0jS%5Yg
z#69vpTC4_FKgjRk&dy_`8(-U>%P5Mr=vkF2R&9^MDAn!<h8239Sz$UllKxdGz=+33
z@{>$CGjVt{Mm)nok#Y2Tiy+SzoCP+7(dYHE51~G7=7_o8^~d+@xv^br{ZNXIEmPfU
z`x@nj^A?<WfV%Y&kqeHEgri6fTZG!zE%}?h^Zx-2`@4IWAzU~W0W+5Yt1c6lOWQLj
zmoNc21(z*nIGF*Mx5sBV&jEiH!`HHINv;^6Y>1Mfa88!Rvs9HF#;?G2b!VzmS^KRd
zr4@5MiBc0{4X3RsdCr#H(`|OD1w$cw8frt)!}(hL4ji{rW0)4ArMqUR(P=h~Z!lEh
zEVEX&6jv=ZnRNpKCPC-$znP$<*lox5Dmn4c=KPdBII1x}@fxcZbZURZ{ss`F0!EBn
z0O1z<mOjm)sNYJtiD*7=iji%ffs;0HGZRO&?$N>v|5sSLBCHfnDNn6ESb8PAx#aR3
zv($Xr#h-(a1tvx-AzmN;BSWx5`=<1!jCmI|=OHkAl#Oj%0?9>aVyG{IKuRzTg>mT^
zI@KplN;p;RqSc+oE;)Zr6d|1(okVh4%o#&!(s*8@v10iCFAp(CM_le=sEeq({OO^4
zED(mEG(BTMO;MeD#e$mdg;-EG?hy-WikD(R1MBZ47HophB^ESN?i34Zsy$-?*QA~H
zy<&m-BNpt81>_@{R_=ch3)EGyV9M~$SWtE89}Cb(W)M2Yf`)&hdOq%cH8P8)rL^Y#
zsTFyWwzYh@F$BbofQITy2oSWe)G$Z0YY6cFqe6bQ5a8`h$_pUe6atX6+|EUPV+f%1
zLFt(|#t`7u<2V{5WMyZHCU8+Dq|YH{`h$GB^R~I+xCR;eO(LAxnRjM!<8c>62*{+K
z0bg>p8~BsJteJnols2?$#kmarD*t{5Ywm3tVH^=JqGQ(0mHp%O6csXoGc^=?QDIP7
z>JO_Lb39K$4P;#7D&D^&;G>HO29>V!Tax55TxyklRNl~j=@Vqvvw-N|RuUv1d}rb~
z0E^_z(mDZ^U>0y`)9(7zEy8<Ca+$e99=5A{b7%T+2abO^uDf$*7r5(?+>onu*7i(p
zYKpfc7}_!6&Aq0$mSE&0r<u_e35LJ7oHYdFi-`l=_}0U}1r%oI<iJYGf5i;k6))s@
z@?PbVvUvc)+U&*Ch-<5tJUPEbF}d<@K>5<T>q-Dhc-;j6f~E9v1^_I)REATUu${Q!
z<Z8=Ha*TiN`SH}7s4#~!Jrj13Z{^>s@EPBomf6ygAwu8PT`W{6bpc%#!k&#t@@NR$
zRPM6lv#H<93*)#4(jU33i{o?;sAIj4dT8A|na|E;^R!31YF@5ApqWWax5>>?lJA!=
z<ej6UE<G7@)74T7si?0ScSWZVs`93O(cPyyAt`_T=(f8cQ|71A2FL4!>^Utv?RL?M
zB#U?0FnD?`r8&|(35ChghN2QXy1%1ECEz5p<Y(HV5@1f?g_@G)H`I@;#Q5xY$OrPt
zc4P(E8d_1c(TtTKC9T%!UiLf3qp7bJBd6RjCO$5t!#BKn2hIfDYNucHYFDBJ(hFq<
zVtIeZz9gedwP2uW;jM923ud1lLD^(k0q$IBMW#b}ss`U_n#3UksOjB*O74VsUF_l*
z!C!az2gx_}(y-V|OUA9Ld;Rj)hex6H{g2_-+<$ohbD!@0Ynus&A)9KM5D#0krNB%c
ztC=n2fpD>l;N(1Q-Q|be#SWVMt+chgg<*f_@v%&pv`Z;7UvjYr5MQ?G^55-2$TB6Z
zI;t*zzA$~MX*u6W?(@T@Q$t!98rkZbPY5AYO)B*Sd4U@IE?vWiF&g8v`Y-1XC$FXE
z!|1h(Yh8{WQc$YTZOaAjazmR=TWVl>f%0imiIOv+*KAw6mlPvPr##OoVw-dJdmn$+
znwB$P02xzpV!-eJrgXhzKq6OI#u3A^CN9HM%ha#Rk+N4&5si$(7EP6D9^(a7@qFun
zl>8)lk(L|G+2?jODjFtXK9|ifRO6cG;kPg1eSgKv(kShyt?Cr{dnA9~a8Qa-w=*P=
zNDt+)P`h0}Obcbdi`G!CEEN;=b7OzeK6wE(wNy;C!=Qsq^1n&Ep*z90L)u=Rn_PLB
zs-bnEajKJsP~sb(`2Dr>bc8}hRU@^s&GppFp-5_9p;p0h|I5R+m#{5*-nJ#!q**0@
zol2&7h)r%#00kIbv-9I7J4gY>DcE3XG?ffqCDdusos=Mz$R=S=TQ{mvREB@o&i%qP
zl@zod7WFj0Y)(^AxH5~=T=ZR(`dVj4KbhEb#p|XD_uYCQlJ>kMH1XQ`Nwt=n-L<8n
zekGQS-0TuetkqZb_DAuBE1FKA8zm}iN@BI<ynHP@v~U))k8C9UQstmUR`zk)tx@i*
z#Rsk!E!n@qaSCB6P+0SSdz^ow;jcro<8wu{IHSC9u40^-e|qJ)3csJLCA`<U3QCM5
z%#Kyix>%$8DyJ%hPc;q|5MOhuq8;VtQxy%t-^i&70gB17$g#tVi-uyya}^Ec3kNHj
z)*Vh(G!!o!t!T%8zq1uP`AUZ?7I1mT(-p1E`W&xlsCGDCAp|m%`Zs@YzQO@hqUW2&
zn&1_Bn5(F2LKljt>c*XFsA{4YN~jtje<u}Gn-KIUplYPtr+%s?+Npf1fnu|Is!@i1
z#ZwL8>uRSOBK=CI8p7{WIkhQ=YYL|tf&g+rjlZ|oO*I6!%cfRT<#^1Yv__3J#eWn{
z#ljy$E~$If4=&UR?0JbtE2Liq0b|lM9+x3pI2D&*4l@;(?UorRmtAo<pMS6&gy(;J
z`SXJ;s!I_ay`^9L)7zJae?JTpQCe6RFGed74P5^D%LDzvfECWii!xqFuP?v;eEItD
z7cO=f%<9Yd<JWH&_RU|5Ed!t}rE}KKY^kLPpf1iuq3KWAmS_^G71@QXh3|2e25n{0
z8s?hc8+!kauaMhWY-fD_Y=6-*lB|Pm<ds=xTj54K8-&JbpRJu+&`#pe`G=KRyV*CW
z-4E|kF4|H#VQsiL?S%26YHL2@h@ogxv{Nt}b=5o839Vxz(dXX6kW%$W(slOD)RNAK
zU_6(k<V=sDwbvL8j<?otQ`;ZcAxfdW;ZgyXb%;hd)mqaOKhg{JFn{PntFnJBz4s*!
zOtyCNdl=i>CRUya2&Umep+Rv=Vr$6I-{kxAlaui2E+c4zYhLROBS?X;P=Zd`EOGWt
zCk!>h1_D&$>S7FAXY^7QgWVO1+Iz&R#vNoUdCf2VcHA21WE+}51G5+sItC2lCL&4<
z4@)$?8&_`qreiXL8Gl8acAdo%Tl-98lo8rBhBxYd{V+iU-<T=KHw&C?<FD#(Fz>Wz
z!<=!%y~aclVjJwAZ<lC|0MmpaxW#}mm^X}6s}XY`6IE<iR#=GRxcHJIM5F~(KIg?Y
zKt?RBDn_Y{LAGFO!X#RnuNhjb&Y$_RO}JGsjAYSw{T!I_ZGR;VQU+|?aYKRzT*v3<
zXMYZRkXks5g0r;?r)X!R63*z;wTs_{+h7<9uk4B3hDoegdjq#&k~X*vYkd;8VNy4^
z4Y%h12)F4Q;S6rWU40I>VNy@xHryIrZo?zuJZ{4!-^OjYl=HX^m;8OWP2U-}a2qc1
zes04h_PNbk#D5&;Hr7Vro%=txVF}vc%9Q@}1pRUjeb63pBb=_?>p9wCXoV0@*Y5g#
z$c;22dN`BZFo|}WX$SB;Kr=8?IRGwqjp_s^$O-etcd;B?3PE#;D=bGL?*{D~Bb>@|
zG6I<2ljT??P>~HI+{JP*dQ^0cm+xdb7!A@l?ig9l?SHGitRI7nM}nG4K4--zCj&MG
zU6h?U^4|QSc0}M*-c-3Z$0cQd$25nwAjb{PAQJBF#RM&+HPcmpx1DC~vQjT&7s|H@
zou`hvve+~TUKoBJ!-XNRcV?#a;~zgzSL|YzyWm#9!Mw{KKAfEsi~|HQcVbRp67PXy
z7X=^%WPcLZHl1bT%V%^09MgBp7@LA3uTFK}<djD?3V!G06h;Ao8aM6=GbNoU!nfdv
z43J>#joxP%daIM8S&yGcME*4Fx#bkB?lgx%o(FoTHH-nFgXOIg&J`9v1aZZLYssuB
zeI8dY1OcJdFG0bwj?rO4Y=ZZ!RAi+<vUDo?rGJHyV&I%aQPs#Ik0y$4R1d7E8`vv}
zLr0+?oNl@y@Un6%u{TX3N0Y}y@x94o+Jz^NmFxB(b-=F?N`p(J4iPxV<kDZpGl{m2
zp<e$ZAv41QuoZu+J(C@m5xo>b=zxq)>N7n?MG+H9lWn<Nj-g~sWf*fUUfvy*RCv<I
zKYxQYG_`A1Bo9(B=3uRa#@#q()}~fMi3nz+RD%8Pl|>X7(H7@q#;XH8w=4x}I!Lp{
zXY%I5Yc>*mUO2nBeMeDqKcCMcknCeKWh=-e+ImiUyRLQH$Rv-EyGuWXff2ck1TI51
z&NDtvOehn3{hd9g=)f|-i5^oPgp%QVcz?`W1Tb~4$B;rO<raN*T>^_i-b$HRMGPeq
zHK|&?%oC0U#3|`hzNB}d>}qOWPk{A0{H3C^ESOsIt9A8|qP9%DI1c>=*|nwhb=-N&
zWU&dSwyvP>auQKdAq5J2mWg;l!E)&0w+4+_#Bd2|?oemsJ9Ai?ZHh_-)nffQ=zoB8
z6`4n84k~o#V5AaW0+&EKsIqm=-YPnU$sDF9Drr8h@OOKi%?E~0^)M>!eTQ%3D8J-_
z>ff11yOtPYR?=FS7oylP@@PAfSe@)z1#A?kJC{~)54F<=wTp-#NZA+V?bb>Gu5awk
zq!c!~8s`v2AeJ%jb%dwjrU($oxPNW;?04N@t&;h(-_3k2<k4~05fPyjE-ALk{M>Yp
ze@AB)nu_ST*vShU=PEz`_Tz^qaPILx>9>FUSN-DW-#&CTuyk5r8x^3cU>AEiD?GB(
zSufQe2-5-Bb|4A;$p$(-oO}53<Hw)+V<8Q3L>3NBDlu*rOxR2I=VdT(Wq*uE=VgeS
za&vtA_|_kbDwZG|otII#ozV;PdV;7PzxIc;4)XvD(SCf=-aRmYdHk_IigN;)$h^zO
ztcbW+Hmh&7i@rGA-_dEutP75UFra}H%=zH)r~W7b108NN9m0js?ug~Av_cy=VmT}6
zdNv-hoW+Rs7LHiXBF=iP=6{x}vAPVbbjfF&S^)HtOeaK}Cpr}=9P~m}LfSf}B&vMf
z7Kd_(u|&^us|wGikP{VZErA<8nW6POdDOozC35A4RQdb_pLV576>v2oHDhxe$k0lf
z3NaBn-m4~Ub&yhMgdcFw%IlNDbaC5gk$kG@*$4Yl6XBFr%`LT0zkeW8V(0p8kXAZx
zY<rEK_G1l6@#`#g=WiPS*Ir0vdpf7nMd@0dbDXBrV%Q`?1X3Ez${f1m2g<O`>M7MM
ztK(s0h2YkGI-~sA7IUiG)5&ThtZF^dC+-6dVDi@x8b89!kI?uL7X1iepC5tj=$gU9
z=*DD#Ej~-E7*?jx1%C+Ktw`HUe#4DQA*hE&>vEvh7}n9@=*X9+#?!E5Je|%)$`wIo
z#I<b#-%#7sGebJoOJ|4mPR!MyvOK7=lxvl_JkoGBQz1T%XO)?450Xq7^?m4fZ1+s3
zi7^^(r7iHGn>^e}uwIkoRc%{AvRdf1uT8W<;V!n(<`}Ff5Pznb(jHXN$YP+G))uO0
zo5~?ruWkS|ZkyxGwx(z(XIcmJvBpXy;@+L^wqdDRBLe6fVFU~kuSUkU*MRB(aW<^y
zhS$Jp0ns(RYT(<@ovHRBQO$CsR{WW-F3^f5;qTkH9Lb^y!Od`-;`ZeQqN89N_?qX3
z?iN1twOX+_$$uzr$P{iGlfzL^|2QI~+i9ksJ#rj@96~8BVa#>DUggxwL@Go6Jo)VQ
z`%4lB9LuH4m1NSi9AxHs-*zYCFcB%kuVm64igH_zuWpxUq<flvl?DVl_lpOrH1?D%
z^U4iA%5`!MTuWZ}opqPWJSOnIi@TxpAg`{ql7st8Vt)v6lvgK~#4w2qWhIBN)dHle
zQ6_n<P_jI|!CK!_y;3f*n8ejeNiTF%&G=!wAFtLOK95=QkY43lX{TP8lcIJ@cHyO!
zCtf3QpT#nQ8dpbp>VDr+(-KLQ(Ohb+GZ?Oxj@OugL9fB$lUur8qc2*jHz(5U^ulIP
z7US#8r+>DXMEdUtV@;+CHy1WDxk}#rG)0Bjxj+^)4FJqr%lDg0r1`kv?vf?|I<$7v
z>e}>x<Y4y22NLfO9=c$JGyj_b6L@F+L<yX^(0)PuyzLE-cGYm13tD-yt3K>EZ+t^&
z)ts_x)S6Q6A2k(HurclOm~{J!KEHoVMMXYl=RBzN-&5cW`x@LhtH!==B{k+m%7Xd!
zr>~H(SUWyg9?p39YDyZ%^m58mQ6BU9>Qi)ks;_V-@n70Ki};rzTsRe%aaS)Cm&<%O
zU4M$A`1tskL^od=-Tb<Hc-?(`LECs02)t=tKR3W1XnqCF4;?igKskiw^=tFzC#{X+
z>wzz;H6mZ0{|<bo{T)JJ3UK*?{hhb)si!rMg$AixWnx*K?~r#0OzDHyi68F<F<tw=
zUVq!^EV@?Fr9acHe)s&k`*(LZh(Sdz@|UN5I2wQEdf1Dw>0iEmX{slGWK{-WRchym
zMyf=u0+`0Rs4#uhsS<2;p!rRXI8so^q85&{oul>2qL<`|PFWug99_jxGkNzyy5l%{
z!W=)oM14}P#wnuEIIwKtvVncln%I}@3tx|XvG*AmZ8~Wvatbk442fNwOvgakBvyH8
zr5%4g8J?f54cb)9j2=mHAogB4fQlHOFW)sl8@6Lcl^J_tr=XTlzg)g)w91*<5S68S
z;-Esn?YkOAD2!&=t|q~lbppE@t$>ac>`(7%EH94&yP{DM!opqAEAX|3ZT7rNLbwE&
zM&;O*%eEX{4Ak0|DC0Zj_XH_m_IFMv<i~%Qsay#ozL=DrQ#L5n8Q=NNWE!3}Lk>m5
z#zsL9BMvJU>?DY-;OMpTh9o&sRNhTF$10ANLsk1pj@zM|_#Q$O#O{hCxi;RiKiU;A
zo{_oYJB8yNqE^IQ`)D+60dzUEc4%2~1UMb}9H_wAyH+c!(fWLu9F8W3UkOdhbUA-4
z+px{_6uZw_n#Vm9mej|;e&{eDv2B-SEB3nIpadkNSdbYdXvI}t_C5}*mgDHj8Gl6f
z9>axHZRG(*wI{mIC`XHY64@KHKQc$sm3~au`UoxY!)0v5A>&@nNkJKc54JltTsdMf
z8YST#S{sASa*k}+GE2k{-A07n`&xfj?S^^t<Z$U3<obTj`G)+4f~CB1EJG{orDB}d
zOUv@4YP#Hgu!H61!c#-qs_0=D@SF8CqU(b0zF(4XE!q2Eq{qr7En^nT<k?Fq;PuXq
z1pfJ#4{0ecv8R`KUIODKKYfr+1va44WCc?}#W5*TBo$PA&!>WlSRfTt9A|%0L9Nyw
zBo&-jf<!8)tt^uYDz1X5z<Sn9q_@fibjt+`bAffhk>{?F3usj?7z(^F7mSv?K`y|k
z3<SwsFy`1f-0GQ5#cgJ&Q&De90a~HAHOGk*P{uN~HmejsGpNwEQ1h4$z2B`tFB2G-
z`_kS^j2I+BZyFoAFr0u*p-F#+1+VbfakjvDH=W;-aEff2L(77=bY4U{jHECaap{7L
zbipcX<SiRdNVgr1(HW+coUm6Mw?JZs<ci}IBzja_;2oW4MO93G^%QrRtYwT_QO`qh
zk-B?gHWu7ks~F-!nNgE6_DE)Y>OnTiji&D`1WRBfP-Ol4OCr;CZkK;VR+YL_Vss_&
zs#5o&Iu|Y5Aq2@fckCzaL58l9eELLUovUw$f-bCYM@#DZcHS}P{u7KQa%C1pTZDs)
zb@mtOrivJ&7HC^HRrqv{KlU=g&hi}M9NHzBItU>s=Wn*Wt}TVCM$PdLa5Lt0jVV)P
zo5=W2=4Q<68uM*>661fZ3Ml#Sf|_I4G$N|F&5VdDYJm|^#c^gtRG<A#MnooT8W9<X
zDtMCp7RvvBL}Vy6m0o<5a4&<w7F47lF5QESbbOC7ic9w>Bb{MHwBo`s9boI35;E$j
zbbvPz(w-J_d+vbX6*v^k9Ti7`c~-@7W}a1_==;vImdRS?SrvcveBL-@7|D6oGMURf
ztG4g=nP-7_%Jc~Eelp$<InM%@Y)$Wfo;4J7;XG@!M1GzHWR(f-)@fb0(~>FKQaqpu
zH)CGcm@*YEaEeBEn>1h9m@*9k7*O^$oV>Cv+x|($p1{>9-3<B;3~Jw`(F}ntFWF~~
z849$SV}>(bW_5p@RZZqp!`Jot5*?oLvHeN!qNACB?U&js#ZL53NuuOtxNshd&+QDk
z{Q~WCIa>RJxBgw$R`+Ds?k1P)zjC`#Db|v5M-DWh6l<dj5o@L3I3CjhlQK$dC7TYu
zgp$=ZY=u{}z}|CjX5|rbKm4MM3_|Y9u`fc%vtD-ie*GC)N0UX&%F5&4nVkCxetBBm
zpI>+X0fX1%myur@69X|YG?#HtCli+nCPFEH)muw*BexO0>sK&0sVYK%M&nhMStU`D
zt8(*ZZ;ns#uFD4>;*%@CK8@GRpa(PLaHy^1K#?2*-Hm?XYoH|$zaHe_hr5^0cYk}=
zcyPiS<>2u7>p{^ElKv`553M;krG(Y$@cH+{Pft>UrPm9+P)e=o<><4gcXNEy%F_rz
zWy?z)A04WvaV(tb7;W+_2z7ROSA~P8k<L3A=@B<Y<J9x<=g)t?lSOnXjF;LUq|nBC
zOYb&FIQ&7s_J8^G1D9ZgI9h)_zuRaA7sYc&WiYPvl;z4i^9n_<agp)u2lnml^WDGi
zX3A4eNYd}We!q)9pfbX`KthP1@bJf%!`HhXIn`=Q_MY4_nMzt~OlvADjPVCcT(ucY
z)hcPKA+o?rHf2(>D~m%SH~dLX(>(b!F|CW7Ea#|R<E!rsUS(%!XT*OnKvUex6@wRJ
z$M1TOkz_vgj8OP2hHSBf@lHg?uevGDhrZe!t7EIFG{?u}TbQ+%U9<R?bB91OfW?NN
z2jQe`_}%~Xkf&JaPcZ_WZm0a`VU<gufJW0!`EH#{KoK~Xud7^08zKO-Q~oK9TxA3j
zg$u5o_2)VxMGh9!jFEr(xpFf7Y4ot21nZ(f5L&V+pB^6F{g>aqK0GSv?|=L9%etr*
z1%ThG6YGVNOq2guXGD3|c{|t1!)kh?FjK~IcTXWy9XsXKBdGd(NPRUU=j?o3Taeb2
z!VtH3lh$7Npf_qxBgK>TTf9l57ieKobA4yh#*m5h7I)I*3r2tGzE%77(UdrCfSG;r
zK2}o^W^2(_R<J=$O!yx6rTVnYS`!h0q8{y*+=66U=UHy~{`-eVaPEHH`_Ndz3W~2z
zz7HPuT+{B<2iI2+Y3YUR*f%AHm{5V6GA&@;M~w`$OEG5a4r&w%&m-ZxbuN^*fXlb;
zordfPh6(oTsxf~IB|Ff_jd3lMAv7WU9LSiIZq|{EkP$(oKc)bS=E<jmZh~AxH^v$P
z!3-hunp0T}i4+JFG@uw9t{l>*k#T)8MKt8PT89~zxa-RTP$A=!ikAlDv4G0F6$CUH
z!F}lW_-JJXBr$JW358~)0~H{h*Y{avBf&K1HYj^h4OM@!Eybi%$Y3;q()aBkv_MKm
z8Dt-T5@ga2`RJMMF6R$Y3q_R?qo`DuKTtKKW9Dc43%E^ME!rKv=iPa;Tu8x5#<)-}
zq;#WgS1wiL38O!zq5-q45XR5xH39GVmWvE6Q+0Nwipg;(Q-%wdqG^<gTg8W~SVz|b
zeW_fO38a5>J=bePfWE&KSt#R{fCZCfX`ZZvFvi0PIdk-6toOpPB%pMJa9K7(R6_e%
zu#u&cndNCxpV2-w{>g_A(<jpquIYv_2P!`g%51$<@?YhU$Dw>HK55<-$ajs=Q(GdL
zHJ8_KB|n7OzJ5>yS%+-98dt*}5bBmR-v_Q)*K~hTy=`>Yeu5!T!3sCrPZ&k>nsH{)
z4P0@gK`==vPIqZH$(uVHykj#_fvO{~ImN}zByT-jtyl~lU$Ath(Vl126QMZX6vs<r
zPGjttVkut^jygUy>VAh3v;jeaT&M(1o&bb2xe;C`$pH$&Qls?KRJ9{y(R*RWQ_Z~B
zjq!h%Vy{u>T$dH^RXv(0-Am3dSOjg<y7$-=F7Ssby{TX|akerfVK<-NuIeR`?5ZY1
z#K918w|C}%F>nEIdc54Y&;Ven7u8K~Fw%)-E+p_Jf^+01EEhT|>!t$V|FzU`v4ZSP
zaWqA>Q~vUB^XzO)I<&lDc5W$}m+cAB#=3vnaY{@1xEx5c(R&hd$?z?~g;Z3F3$93(
zkIk8cu$Wx0M^=a^9J(H|x;{nG+>ESddGrRp2td#}+|n0YihF#*hIEX}C@wig^+~p(
zD>mtL{qfSKze=6pzIST5uAV2(Ix`;cE4X2-v;@z^3S<8Pg)n$wEnX-W#3OC0LD*dA
zT5HrBsd@^O<a-_CR=*7(bhDg+s-GESWpaor&0=f=o{~Xem{FGfD>|(_1?tQ%V#vph
zuE3F=Cutu<F#2|$#0Z}8B+mZKz2}}IAMFG!`a=1DYGwq-N5gZM42n3a0(?}LxQaL`
z0dkk)ia0z0pO+wuI4l8|mraW}M1N)QtX>CCe}$APg$y%qE%CM}mo>sMZ+R-~^yB`T
z;L3C9EwbUM;+;Xx;WTY{Qo2FXFVDj)m~^=G*Ab6gVkYM6=;y-4&J`Dg^fP<MeqGY?
z6GIfPN}cR(t6lWe0`m@+S)QPG8T{Czb<907FC*QgwXr{|Zp`wm|I%0Clz*A=cwN<k
za`7(M&t|1Pdz~NE4JPK_c4n$NUayJ~JR2hrJ1f9)o|A4P9wWJzYD0`r{cLDIXB<o^
z_Y<&{RpLoT=&US_AMU8&X_&nlZNVkJ+#Zvgzp-PCP*QGIeP53k!YI<Jo{fp*9*dpN
zYKC@cg)`4r)wc-htnI5*qkrnAarnK)1nB`&N8DRMmyokQd2Rg#ha;qO`*+-{jD*?#
zeVWGN`l(&EL}?eUGlu!T%x~^^r>*M?dy?`L(~irxG1CFYbM@IOON>{nEGK2?sa%Mh
z#xnFmi^cO&Cx>hbrKC=_C{0xjPTksFPk<kgWALB}oP=-ElBSSQf`8H<+>*w*>~2Y8
z1piN3(iBkmZ|eU$cQL>y-moQ&Q9jd>#zx<uC5=%$(~`#de_Tu2g>sdaw2odrpkE4{
zN$uj?X8Zc3>VYbwI`&IleWRxo3e<MBSL%PG>4yq9qx=m6wk5dZQn@8`=1#dCy#I2e
zrFZ6TgXwVNg>tG*x?Y#c(_`UHkxV5*1R-n@wjan|i{KO+yu<e?y&5y#yGA4;M;A{k
zmXrT>aYOUW!pZyo<LGnO=*RkRHod!rXr0-y+6Jv?nf{m2QL?-F1wVG~b1pf~B&7CV
z9u@;ommyp@6_>%G3KIh|IX9O<X&)(nb7Qv=zUx=K+@wl`0nm-DqGHvEGfb6laddNh
zieq0gsl=E3{WSLF?Go(+Me-5oZhYt+5TAaT`1F^D*SCjHH^9j-tAvr$+s~7rA2|IL
zoE~a586lVkF};18et6+rYI?oW3pod%m!Z#=-qqz9gryNer<b@q8x${fti0xbF`(is
z>GJIKE)oYzBaJoO(<5%e`S;6@Z{I!eBsynK`;Y!@{Py+j;lB@MArhF<Zc-qaqBZ{G
z+k^iBg<{&cNuXtjF#Y<^>F0;Pj${in1ylX=>o3#r<okrB6@(?%TB#08;LHj+Y3&&J
zpCXn72?yzM@o3R!S|t)ziqOD+Yy6D<^Nn6Mj#D%aP5iD18Sc@cG5k89voG*WI%OHq
zvhOXOUXYFx>-b0UC|&o2r2BMpd=X2jnNoJrpcxvQrNuAA(I#n<bW(I&S9eEg2H-}D
z%8YA_)EIgp^PeDl_oPMEieN(nWyuFLNITQorHT~0CtGMZ&cNV|gNrhME1CApBpWh^
z5RsBIp=NT!*V70{iS%TKt-t|+6BI!*C+s~bqF^Y^j3RUt>uB;D1lu_J9oQ&j!tJm<
zMNmspoV$W;1e1IZ>^fR8)bbwCSYXbGd#=PuhW3tNlY3;&x`WUFS4>ir`8YHnfEgw4
z2n|Tdq`oKE6zrKXN1*|K#VAFXIrQWt&jhj@8l+;1EGzRR=jXih)0bz>MG9@%E#`3M
z!<G8F<vKa{kxYE~c%T(X8A*jD##hqCXCXPxA3@O6^610tIWWy-wShzY@oxMTnW83J
z*jPXOru-=?Yl|oGU*rGrOhzfkjMjI|WCn^glRKZmP{t}qR|blILM)oHFhKdOo~;N|
zR_I^etaLOlpDilgm?0)p6!DqQ9(`&XOfhAZc)NHD<~MK4@?(pWhQt)+;No&j8o_l*
zu)Gi%7vp6@x1Tn90K!6FLz;ZI#evTZ;ponRBbQ`nI|pVia1eIs`SoK}NVW#5$1FFk
z?q)ZPuMrIe@w;q)bR|HkBIH3b9iNF*Q7Rdr-BY=xBF0CBWicYj6O!yJOCY(cFSh`5
zfexv{HHs=+D%sKA6S`gjqHy+Dp!Ba$|KUt8KHZS)h^)!bh7TUmXzUj9_@^g7m8RhG
z6hl))KjjZk8=Zoxo0fy6$K<g{^L@=ljxOrx(l6f^zdP@LmE;x!?;?xJel7#7YZY0L
zpygPmoYAn<eL+Jtf^kb*S`(McGoVfU!tVHA^(Df#-Q7W@n6a=+Uo6n4r=}?DMMmAo
zPt}U12Z~?f$S}i{8DUiB6%@|4oI^}hY1}ocax{oCDp*nSlu}tMy;r1{h3tK1Pd9ee
zJA4o$L^4@_!_14M@2e(Wz#9}`d4oVxl4fSyllz`+BLot){J(&D$-FPmQiz4^lR{R|
z@;9Bp<?&U<XKz&JEvQQN9LYyLOMsUKsa6HNkNuo!1w3W;RE0zGTghZrx5*Jw4n_>T
z>+zgwF!hO5azv}BlOfZSfmB##2nw5m0Q#<pUKHeiOBN*P{B>UtL$vH{QD~b$khe#S
zTN(h?W!sBw#;~Sr6XLMW_x)P1p*w4{vaD>yJ**3z&lSZ?b@$9`S3z4Z76Wus;Xz{q
zVEQN<b+T{v?O18CKdLybQ?27qww_&6ac*YVk`R&UyL_p(Q@6FuNa=D2xYw3y+Lrue
z&+X=aeBhdC4$gC(a+Wo_UDd5^0mo9rU^g0Se$(JnxSM6}D;4{iKcsi3Ygqf7>BSXG
ztp_ern^i95*X$|7H<}e?eO=zeu#eZa(8+v{lt0l}3+JRgiL+{Djx*Yk!<rghY-SW`
zf)uj-T^%s7;N6Ysonxl|S~VDTCu>^>UPn!TOuygC=!0cqN_9pZ0H>)6H=Pr2(v;*J
zGuE~#F-`0C0`6g~&yQ0?jZ}0;sYPBCO~7OaSPS}?z^ugIgE5b?0_SjCV;}Rxv6v?<
z(U6O99f)~cGNhQdB*^KQ2fjrMGf6gMUYBV6E^&)f+_H9*@;-5^R)1zAZs}gHlsX-M
zufQTB&6~%oRgVf<Ga<Xaz<qHQ6hfuIfss@VL?*@Ow9U1&-}e!;s?hjdgl>(^AxI@v
zXd^tqGhm9Y=w5ocvaF3iiIIfo(TMa(@iCJKPr|GeYr<^a@+eWW%<h-llxd-4Y-P|d
zBejsRjI{4&NZK+$F+g`LnMi#^R}&n6#m^alKYP=qQV*7UZ7bR?Or>HaRi;=Ip$w?X
zt(8`C)7HZ$f?4KsT}j9`0|SAn#kjop#RX>=?a>UXgMmRHGoTkGIUN{KG3Dl)_Kpl}
zt7Q?lTfnO*8E(zeMduw?omf^c&`g57YZJ2zFtxCGcWpEVnA+SsRvWGEmJ$qqam3*B
zt)LGiz4FH_Z5Jo^rO)2#k<Ny${BWyB`bz@1bE#*1FNZeaQ$5ScPx2+7?c<zy)N8xL
zkJZFr=`d+ifO;^4%%T~&_06$khekH|-+|EYna>=k!%rPP$MfW2LE<($yRH7kFk{VW
z2o84o4(#fhPN75wdL~(^uFPeBztO!K3bO@i$N16yrmy<ihHUKy%ych}Y%+{tZsyYU
zB#mopA~W_P;*)R$wTOOdDwig6^y>Utq-nx%rC0g6qdib#kM=-q$g4g#74Pf)awsL4
z5@OuL9QjH`E1P7dEup#yc_@LRq0AaR<JcxbpXW7s)-`3w*Vu(z;u%4IHv1F$T&dv8
zl7<VV+V$l}o|ldjN*a}SBOMULQF)&+mWv$~SnhG@aQM#~im8{Rip)p*UupEv|4N+q
zBwOiBrPUAlOnRDi&ehIWq_&K5`86fOTRXD6yOgI118uICtrEgj-mW8(CBvq?&w0=-
zTA;Tdup=U1#;k1I?Z~Zv^c3T_jzW;;R`qq<?JN<^PaR2`rQMlsXXH9|JB?V%KLh2Z
z{Yq3i>nif+qGe%BJdXDkLvjVLUdHVn%%wxQ$Yb!A8QJ7#-5tySq(cBqis~WGdMzy{
zNAXYkGNGYPwWgzLq^17&@2AZZQbS1r%mvaz(LNR(&;XX}In@<^)T)&DdbdbdcEkA7
z{64f~Ze0k-V-LbfFxn*;fl@`wRk$(hi@N~he*%nMa_)O$hDi8v4wO&MfB2D4Khclh
z`D1lwr~iea--slI5mJ)rU(??xQxEyIF^kMI(RL5TCwzLAQa=7MzR9$w%W8nqQt(>V
z0;<}UW`JQ-cJ@<${!ZwnS<9j#C6t<*QB=99aHuJkD={iuuwl$uk5tuq5&M>B+G4zH
zL7+s)_B+Xb+Jr!nWm+*qU>j)#NZITPpPhVE3MT8>&8qT?a@K&&b%%BP=ft+9UBp@5
zLw}b|ie4OAx8j@Y_FjX#bBru9bkI*p>MV%ip590*d5Q9W0NdV4aqTh1BS3TZ&`-AC
z7DDOH;b4}{w>C=qhpOi-#qgc?TBoGjS%D}|BJOu5GBIHl`608=(9U3Vde_tyuiGw?
zn{e&UU&R#Zcp!PD*%0!5!o+sgBW~)5zx3^BsQ+gr0g0-3diLNE_4tlGmplbX5G%7F
zaVVsBwLFD?s!l0Q7Ay~q@A_$T`Tq7ELce_Ix#S`3^3@^NAh=|bm)lT-@!us40a}z7
zuIKa{7F3d3rp0V+qHk7;d{gS3hx}X0<qrTI^lJ{DPHeh5>HN*4Uhv01Z>uq<-4ME;
zvK>f#=R;7waEgm#^sJu(+gt>OfPIFGpaPh3n{En!ZgU~K;^XpnxJdusiP%q3+gvE*
zqEx52ggZEc25299p0@k)l#%hd>WV||oWh<o^`^0Cu7DkL;aW4O|3=`u%37s-xVdTr
zO)LFtycSnV9ATEOr(rq&%h7FiP}hWpYNs|Gp}poS7VQO|whh+YY20ZutzUi_+f1tw
zygkG$&^3~0-%$(B?y6Bb4#ehAqONr7s*qVM%~SZa>xVG&f2>LGR9O5UW}-&pmmyp@
z6_=5*F%y^mnK&zd$&Tf=@jhSij@<~FW|74$88^_ao;HwM26KyZ3gW~8e27oZ->0~W
zB}$Z@F%FW$>qU>n+E-HY_V=5-efQz@%ZG2?HQt=?2AsNm`QZlig`}S#>7g|@2OzA5
z+n1kifBq<?vh;bS4@#*ueH{Jm>9=`)(%@+Y*z$qrCx`HVab2sHx<;D}r9y=+zoB68
zw9<Je<9>t<XdNuifBEvq53-P6e>Zxy5#Hg=Iut}zq4!l92Gv%h5$N4!wKPP7(fNb%
z@5Ggh-?;c6JzeNS{1~DXUor%~muF1|NIV#oze#(_2+_W!f6a1jEpd>(?e=~zPhNT&
z{oORuUn**UC+}7H<g{(p#6Kl%4xjdgT2)}YKyz6IZ1n+qTm{chk&-!!?l)G$zN!3o
z)c{GbO`u8(W5dmA6wazD(CDG~8i|ZP6^h_65*bzkBhl&kOTRJ+B+!3e7RLIWK)<TY
zs)@pLbsdH#tgJX$@VG|75@8yae&39#LR0}SRg7(a2UEoER#Qp;!w$7kO5ePD*o$-8
zks0biYEj`4t9*ZcvND<5Ta<d)Yd>b9w89aZ&0Z`2t6`W|CJV?KNhV_pC$`D68j;(P
z*o-#}K^V|mdhpYf3qz4#!5|Svr!SgKTqF}!#VNzSO)4e4T;pu7@w`(gI<!SjiaWb`
zE8n|+<|3Vy*&^kbb5QA+bCzOn*}1E|Wn8vAk;S^+GbQkm1&?0pd6z<0(@u6yD0HxO
z=)2Isc1dfY4;HH=d`Df8bEU?uG&fqI<E8EDdcQrwv1^f4epuxPD=%60$scB+-;l)t
zsEeVWuV!>A7pY*J7FuIiRL*o<=CvbF1@^Un?Y6AD?ReYFla!Id8nvbPS49csLBJrZ
z%sFPuDje<lMDeqQ`$U$=fsan(pN$pJz!}8|;z&hHkoB)-gCTRCxPPmhLt6)!WKjN-
zZw>yfN-C`~Y*oXDCowi?;ngU;rNy3a-YIAW5v;Pv<C+vfc?S9L<6{9a`yfuh7UX+>
zDrg_4h`<SpoWzHJK2Bq<LV?Z>T&*!^ro-k>p2+P=sjE6%?$F_KNr!K951^cD!3^}C
z>ri<u$QKx76YahZsmBl)HXx@u)DncG)XAWJoa#_Z(jg99ofON)fiR@`ixHEp<~b^9
zkj45Ok^C4x%i5P>3kjw@J<P<LEZppW;m(MoNIuHgftPtzq5=wyet_ztW}*M6+LAD@
zw&qciniMx;U(P)YuzFV?<4VnpMJKMTeeJ!r^TwB4@eiJNR>uWBF_{^|Mj0XPsHC@!
zVOHo>12sWd1!D+;-!=qIO~;T!YO|%Sq|>RXlG<e$n0osk);dZrK*B)tHMJvusojN9
zjeB-}Rkin52)l-blvHgq#0!!DgCZadkWiqE--Rk2i=p4at#dC|8h9-ILW!Fxnj-H5
zy1wjyvWkQHMP<xxr=ELQTOw4#+KoL-TcP=Q04tqG9XUN2@<e_TA!=77yY9M<GHhFU
z6k+`uVcCDKcC&ZomkdA55*kl`3)^*{A73GWO*3GB>Y$6Hm*Sr<tIu}(UXpmhwbu({
z%w)SgPv|?nI*rMl-D(3W*n`d}L6zP#?^U~7!=Jj9H?%$Ok(yqL8FSa^*C!*?i6qxm
zGf7r_71sfUJA=5Zl5{)u(JIAMyHVYM)f0V>SAWVlQ_q}iGfk-@lpCXeSk=8(RK;M6
zMUfIVw`eh7>1PbA%2_8GWqwgTsLt>*H`~Exn-=(zfqXZeu>Ev^MT>Rvu^;^a)yKn8
zBvNx}sekGY4;U#+H!O*21GPb~AeXRZ#8btv8QkPm%rw|eX<a{8?#79E<zR}Y?p{nk
zVUW47l@zqSS2yY_=;O72(YK$eTpyo){_eQ42v#5!!5azbs!rRL=F1v1byc9`(u<rn
zu9lXBU?cRj#ld*G2(#}#4r>eyLVLd@zS?XNJ~qXa7=or+P4(6${xw>D4dJG|>IdP*
zb0}=n>LLYPGm$>cfV76?zUP+WUOU~muOeh?ose?yp;q+3ZnsN+F^GzOp9(o2Sepby
zi^JMp>zk$eMtj4dm2gOc15&UmlZ#(!qipv<sGB1i7>Ft$&g_9Gsf5*<gM2<|$Ad6Z
zF}50E2Spa8Sq}svseoKVZVKxq+_zHDx!uWuGXkje?00zho=gKC=Q9cv5;@|BKg?$o
zKDL)zL{iT2PzZ^Cb9x5#OGUKLk|xj42pk58Cz#Twfm+F^<Q`1Try<G-?etdsT0iSL
zyS*>SmD45`;P#>$Wu#OV%)ReOKxq7^yr^^43u#v$M5b_V1N!^ND~*6Alff<v496op
z0guNT`(yqq`LDpv^WQw;P@5G<;nb*E9&u20qrjtxqlNc>BmwOytm|tLhnz(^4ssT8
zG^xgNkQ#9uxsV8CUpzrmA05MoM#?v;L>ztM_#m|9l6d%c#o~7f^W?*52m}R%#o~`s
z7giLb;~1M^`Q1tMR1`JuB1}t{CDs$BPXaE15Mr%JoW}l|fJ-|>GP@0$Vox~hK`wGe
zzK0ez=CdDvHNglOZK22V?c<Yj?g38di%DG54l?b)=sSwF1z8Vr7Hebp00#1Z5o=AH
zt}#Y0hFYqP{6<9AB5lmHVU1GyB}b|_#=#W*@*!{i&y?#B<v-tA*co5og8+pL*NGj+
zAPUlOkhvYlFc?Cbrc*_7#Y|MZo?zWp26bRlc<i!&NGFjl_mrobUXB&i%i4sZ>W>NP
zO04_2hF6^){%9S9rIO+fA;s`zjK4=p+E9zG%~nc^Y{Kc>qL%>(lTyu|l`~`4LwU-B
zUAErHIbV#K`bo@M-rQqPh?G_-ecmA|L@%6ydqhPItT4*&8k9XYC_dF(Cz|Ek`BZ3c
z%s_E}n2NS_T77&sXzGJj3IJfcHJKem6^cUv^*}&>ET@JmPFh|aw`w^Xx8`qa1XJCh
zGJkV*(5mIppjCC3wWbNup`aFxi!O%pc?YAt7C>2$ft@rk(K>_TASXGP3MS+5E-WFY
z?F|;$rKO{@7wS%p+nPg~?wK&ZRYPjfduOtLx=xc#y=2&oqIw4|IbU;y9(dWV|5!@I
zzMVr<4^+Dz$Zk@-uy&({v(Tl;gN7G0ISxgRcrUD+%{=^LUiU)G`I>{A+n7?8lEwzp
zT>DA`C1cA$PI`WANrLV!EYw&SYN;5AJS}}xoV_`?3zMPlCzc9Z$;}ZNm$PjTb<)j$
zSe3J_qDIK6OM;72E;R`539LOs<}Bsuz>%qw290_-8Ga42K)FEyN2X8uG%7ttp~J~#
z73yv6#wdO~8R?kOTEBgqwd}FQuu3ry@?T4SYh@l3#ykF6RUh{I<rtMEF%*;S2WX=b
zb<#^Ab-XH)OEp8Cl1S14AuhHuECoV;;C(fxytEnbK5f<tFfh=jv$8uXk1~@I=mMwm
zq|~I$q;2#UClJFG<3|bPnC2cV@rtqa9LYsidMRFHh3o_aUMe^%E7!49R(d3Qs#OFE
z%*6Gm)oJ-mKC8^uKT`94!V{r!=<bPti1BS1bx%-2vhsLggJG%7?EH>GoVaLz7g{rG
zy2Bt}iamD1*A;T}HW?hXTDF~1$Ft6SR#|hsGNme6x(X)M1iHNYQ)+_Y7PSt~j^D5D
zs5I>`FpoW7YT4XlQ15x7)ZO;(o6*o881^@!Q(e>FqppQ8cF)T$-A!n_9y+GeBhyVv
zSp8#sH#6yU(9d4FmpZyCtDZN1$Ir@VyyF78d`i<<CTLVTc&YtO3T>>n^qYdW$|q&J
z-%L)+n$9gy6mtFCZmm1p8S(Dz#PY5fpb~PO{PkyLnDWW01GRMTom32AJ&%jeqK^5Q
z-7|dr8pn41ncZ`{BU!h?b=;9XXzDSVJU))UIqxs;twxShGLx<lEbAqIg{DU8nxoF>
z*LR!$K7IM{uc1;0O7`(o<*z?|h+kk#<S9+1c=+?*Za;kZ({|U*3v=x^bvV&l;t%>O
znGSpMxx613B=Ie+HGKR;b#<)uAif`H`p)<Gh?T#<cLD#}XQS|BnjG2YESVpwCR-n=
z_7rT@W}mNfF8tgsXAL%gcFDk`z{q(0k>tUibn?(q_*C0*J|M=$w_Oj7qccT@awp>l
z6NHtcLixESkIX1qupiCHv)hWR$?O^SSuUeKL9I!0&U1W`!uQvnbQi|AR~Ov&vJ8i<
zSem({7drxbam5d>y|lLtFTebeuTmCj*oeMiX!yd#$y0&biJEJFA(2y+@^&0se65Qd
zXsDqC*r;%ObqLk={ixk{qpn=Kpryh)yG6*}zidU+@RAz}t#Ee;n1r<difKI|dw#NM
z+{SN#&TpvIo!6U<Zn)WKr#Ck3dZV<AuiMj?O(g?!J{q<49fOm#cKcfG%4-df%leLL
zJB+S1!+SR3Xo4sj_1=rL89$?DtH|;e?I`)ZH_a&_{|8#k^46D8s5TS>GdDJuAwM}J
z3^*_^Ol59obZ8(lGM6BaJ}G}~j}*BP{?4zklrNVLxb3p5%8pJ*+z~EFoPt;)C0Xr<
zWtkjYxE-`RoVb6Vr`)|W(>t>;Z!jC79m?*qT~&VS#qN$FX?DRSPQ|6_BHGavYD96A
zvrz>MNn4*fW3WDUOrcT5N$RQSMA}%)P&7R-DxqEaX)p!F;;JMD#R`8)L8Z`-22;w$
zo+_4*(8Q_e)!4{_UrlZG3bvZtu!`TR+19G+h=MU>{8EZNFN36h-s*@`ZVUmWCYORe
zmncRFDDIa9seu}2c2x1)NT?r39vDd=VKgH#mU49|+H-0e=YXjbwNcUROWH{EN;72+
zFu3H}dNEg#U=yIT)Yg9q^`Si3T2Wn&5ECR!o>ORK%piwp*=6p9fvK`)$N^J>!j`0%
z$mK#!ZRnAUG_laZG4!NjGh@>WqnP@sh#dF3ouLRbHWhOv$3lIlk<zBc{&NjInVp2#
zNCGlKZv%)Rl*D$tg3{2|0*M*hb9KVpNNlT!^t}vHo3;R_lwyAq0F6?q?-Y>IgavRk
zHsI#WPN^T2$tm{L7R$wH=ElqD<zZL^X%GS_PD_Jg=HN<b=*bLs2yA-;#Y(9C3hGKs
zZ3PHgQUiU)j3sN^F=$e_Hw|N$_R=uBQX8x@Llf*qBjE>8dtNw43T;aP!<u^rAfs|W
zgG5&~G|7csEjfSMYS_<O1_n%4W6wY!rOlwWX6!kdji@wQn^*K08YF1aHBPSmSlhqS
zyF#&^n(~A1zwe%$T09^GkC)v)|MhP?5O2s*faCJx`}b#$9<66nurb0&@D-XYSY7wy
z^!@be<iq*J#3dWxx)<lO+2rcdQQ!RQr`b;@v-8=+#lC;z>GX1jMV>aFF1fAp)V7l`
zb!(W&lvCd@dm5JeMz)6b4Fg659?^L8sC#iWy*Qc7PP-RRo_4P$pJszHum1QjflR+l
zAnN68ayh$hL9q?%UQVv3AFnPZ*F&JRzMm&=-ktw2{d5}86!^JTI>QWC7;9{z{fY}q
zmAR$hd1HTD9?U=^n~bb=mw~pnNs(2X49jn3lkV~5<#d><3`RQbbBrn0=a?I+sd6@#
zrVVx757VnRldHB&IP3n_J?nl%0~($|u8Y}eB=VL5Sc_MSJxsnLW^rLpoX;L~Cm&zW
zY{KX7F8|v-o=<-<dpEu8PP%`*d}e>&y`9ZIT>pQi>wbARd;9UVznK2k{rd5@4-el=
ze(!!hpS}J2huJfj=ITG#IRESW?)CfW>+bjHDfrl3UtM(9li4*M{ONgwOFw*0LBpS+
z?~;MHREbjd(IWn!66N59_!ugY6oyI~<2sc_@~qb{_Kl_=GMz65QWBpm!PVlUl?bUo
zzkGkzJ$9J>r2A?5YT7+<-^G}bE0l>5!L%TxNrD(l4u?RvNdz0Q^%0C!0UO<fl~VE+
zUB0$2-yPpZUJi+3r8qRm8li6>Y{eSu&_Ll6A`ff;RS01vTH*;xC5kt*0c&vKVS?7!
zEr3*74Ink_7C>rV4j^@}U?Itq6@!CgN%emk^(dhN9`z8v$_u}0_^a_-r;rhzMi3f<
zYXOJP;pyunq*~VCNp`}M9WEAls=ODTQq10J{To0f>~xLsD3Lm}NFCz}qI9DeIb!Pr
zpA_|6bR$-c-VS;ju%Z_C7MQ*Qu183`yH5H{1idq89tz4i@CuaexVamYbpd5N)9!x*
zWgKEW+3{k4axB4XNk@S)3Iv04kuj9*Fn?!oT;aIETh!!n=)Eszl?~JP68N@;*pSZ0
z5UaaEET(0MmAxRAiG4Kv2Z-eqy@(tIVr2Q3uwS<z7Q-gda?IA_EK~9|mm|itoj`YI
z#KXZjrBv5te~hu)j-9Z%n#t{2YbSqfQdq)f%)@~)$Nh3TGy8UNc5|SV=!?>EP|9|o
zq)Pi8C~NfXI(5*f0<iHo3Z+<~P-4FAcKW5r9#<;~W_X;+Cak4;y9K_FIp8xT_p)zp
zRDW;nj<lv4p&hFcTCPUu=BCFkqF3E<%bSDLaySfCqPNT9oXEFZP#`;~tlWPblu(lj
z0tD3@0(NBS;$WxZNZ5z8R2-q)<#e3l2)2L=0pzz_i1J36i`G;~I2PM`jc<Bc`PF>j
zfVFyBQ%KQ9g;Y-ImUpDQ>>c|RQYlxs!P2o4DXSjmeh{h{8YSLBV+vmi=xMctO1XGS
zbpK1JJ@L9d5Z6>qNp=HbUao(d(tfoPPOd)jA}->X;R_!pE@rhuBh&4QK5caWF=iVo
z`anMAd-4+JbD$fS_lK$ta$W^prY%6XWZgm4|Mxdz7Iar_qPsaF(~=|RdA*cnzbkHq
zD=uD#i6T{`<J?GcveSz!S1LPYJ4{Vgmo8bwYb{4nNt2~Jl$N)GOfG*L1(^}s;^tIz
z6Hc6+eE0@s+J<pv)2Ow{d#J7bP}amIq22sYmfld?w;zKa^87-9c}~fk(m7S;RK1}j
z%&EBFWkd-0fAPkOfrX%YoL9lg&hDc50I!0nAhp^zP&mRzSUO4e0yn@Wdd<B0D!3$#
z%vzD<Vfz}G!kQD^wl9CAE2O~gq73p;+DA;Vu0!TS6|?$2S~vv+HDN(eBc)RG05VvS
zQKL&0rIhs7PlD87{Z;CU5VdkiFj*)uCJBh|aepD;{N6t9OZI|P$>Gl{LnyUZ-)Th%
z;+lk_#1bHDrCJWL8=@$;crFBV^!5$oT|&gHcDoRhh5|f9L>YfRfE0|QeYCF%1!oa<
zjHoL`VG{9S9Wqgz*CGgh5TfFED7hYDLxxX5sQPH%B*Hc^REHq#Vel0xG$;rB6M?*d
zMbJYCk&1`IxmvsuK!Mm3?PJaukg;?-OSGgW=}U*gC60L&UKGpMO;~Xh@tp2W!9M=b
zVw6ZnA6ks^?IM50Zk-fMSWSw<Cuv(J#j;$S>O3jxJSpltDXQ#$!AF%n3SXp4AP#$;
z4RZDgEQ+H<t8mW&Wfm2rBm_bH)B|`v+y`~$Tn3TiL*P4Q*en}tbH^BbNsCLJ*uvpa
zD)$HsKYWguD~I^%YY3=_;&!2S6Hg0*Jha<BhzuWW-cbi?{|^#d^>3FUTsRe%;4%gj
zmw?hYD3(GR1eX=DIG6!~x2~}`{|E$7Sd>tgvAQ@GmxAU26PMb$IKKhFx0<^+*9ipP
zqNWp<vAQ@Fmyw-16PKOEIL3dHmpT6W_S22ds*idiDPzOII##01sAo8mk#~dgO1e4u
z#QvDxjqqgAO2ld-n(=G=PCmSrJB_T1pA$vaIo_LST;wSKzeGMsr-xVT4?zWsmghD^
zON!7QEVFVv=yr>Dkz{cTTIGR<Tlj9^>f9<<&$1&zLRLAKxx(E&vO|C2evgQ^ABH!V
z8ChIoM8bf)uNnRG+oo_bUKy%MxRU9&WqPfw+a(fhyeMiSD~~w0qDV2TB+e<YRFnK7
z8`v?dP@$=*woQlul|)KS(la4?6`)xU<^402EyG#1+#!3kC7?vNtKF?*%Q?yUwwzP>
z^K3aM>DdxkA%~jAY_2<p5EL`-Tn~{@waSI_!fVfZ94QFRdhFd+k|8mhk0zKW8LE=Z
zc*{J=6f2!b|2eIipc83Mt6UPN9g>%u$2dHH4eVGZc;L;hB>yDAj-|jH5gxM+G<?SL
z+0Jw;y1R<=#AjJ#LZ+~}&<y2k(`8uPtl6pCpJCfI4J(ioRMGs<PyB3a65N&|wauX0
zCB-@wjHyXzTZ%&!$nKINDUIY(MZ%q2i;CFnlHy~Hs~vmFP3uwuC}((=lwcJ|uvH{~
zc2f##NEv;H6xK$W05!>WX=KpKxP4NfF|hAQkc;ioI})T;a<h!^>t-folCos{B3tBK
zqm0+&CC`zu-pacT8iy>n`ps4*EIBuNG}w72OnJa(;F&VaIvCEN-58YZ!oq=O&^6{8
z!ko7vcz>yLRgw^f`#I+_h_L*1(1HbjilaM@>CX*_eMrB<HJTyhF_k_QlzrSXHaa5}
z3YrBxy^g7wM6}^(m6r_N{cJCSjRuKtpPEk}-){c3=ot(cMTCdHeY=TYi2Dhi3I~kl
z=fm&+IDEbNYs)oTynRodRN7W;tt7B&TdgwDfdqJ1JvvZr(DzzXcJ?`K#wGQCQjpDI
z`jdROy7KwctY5K`ps8I9{1{)APey+ia;*~{7`RqfQX~YuX0qr8EJ5FzZ1mNp&aWFD
zy;mr`Rl`NUta(cZGZdF$ZxVVeku|U27#mKWZ7b+$Kc@J~EvU}>0MoXbh!Ze+H&;$@
z+{EY7PRKo#X18Cr!8wtV4!I_O`9)OP#^-`(gXWs*-#^^lgSEHc)4RL<(rG`WNf&#!
zf}zuaNanST-~v4yueTo&d{zXX3xObV6&!if_A4eBAqnD0S;U8;7Oj#5r>vBpImzEQ
zg^(0pPZh!3>a<S72_El}B6o(0-~&V>KVYZDA!X6jR{Kk*#R*C**SN%gcKgLfP|?Eo
zumiu0Rg6$mea|FdPN7fdO&hT#57P5Cs&<mJ!JQ0|6Eel^q}gg^##A3`t_|BVgEX_W
zGb$-~>6#l~|Iz_-(wjFF4;`c>80eJwPhmO9qR7iopNIRpFf<7`R+J^p0E|F$zuetp
zfZOA?Xcn*JMq9n<l`IVrV!d&<f4^<xLL*W%wR@e85HV%=r&1f{9b%~yW>Gmd4;a@G
zqal_K3l>%B`FKum&yJL2i7_(99@kyg<Gvj46C#0)!&EyRz}8ny#w#OcnZuu{bEX+W
z{Wu);Hd=(Sdj?xscfA)MYVnH3TlOah3Z()ymW+qpNp`_n1&8JCo{If+e}mb=3=<@}
z5occ!V;jM@hOt#TgXo2m7>Pz>*8mT?WGa|6i-4zSn6%+`@sY|G<7X%o?=-%FsczHo
zNJhoxs?qKO5PZs7q)AfxLQ)BqY=g=wexb>@>b@?8W7Is1+?M?JCkwe<>bqU+=ytMT
zsn_uYuc3-*ZqPJJ2ACSffAo#FkCW>UcJjqn%1)axHLkBYC!l1Pr}4;f;h!w1MUT&?
zn&&+^YWOv$q%6DG9m?KId$yG*r6O7D$8zpZJwbY38;bEnTCCr+p%74#OX&2WLpzoJ
znB2QnS3c+7biL6ky(x>A%Q{By{&BEXT2;<42u*ueB&7c)!R5-Ve`y3kZvqt;QeGze
z4f}R>YX{y6AZZz1cFu9U7eAWS1*<XhjC`<kjeUFfj7=UCklkVIlV-?<Z8#>Q;w%lm
z?(XwJIL4EthBNB*bEhD<SL$GM_ri8N&KYE9fn3WGU{yW}qK&w_mG7tWQ>;6F_OquL
zJM@xtXg!>(mmtx2e>S~hmwJEX^T({q<)<d?o`u<|M2-&GFb(I08fhbFnp&<BGFm^+
zPE#Zdr$58_=lsgtw@MsHUk}bLy=B<BSWMO}wY8?G*;=ByEMnP<?ZY)Aw&qyU-ku1x
ztUSAx)2Rp*uqRD)d-nwqnp1GY@I)Yt38FD1=dE=ERx_a>f5XYJ2`7waoomVMGAATS
z9cwO_+^?fgDJtV3-}9g&a02c{t!sK>JhJw9O;4;jW;pmRPpmohJh72`xhK{Pd!E=B
za;GOo^tkSg!_M!oxB~T*gM}5ru2&XMDmZBG*3of4UJa>WkQUMM6}_$Dc*(BYpg?2G
z@zl0wuWMJ-V{H5U+}+q&d&N^{9<S-=x)~|ly|mWaoW^0JXDlckq{r)5i-v|_CmLGu
zt78a*N)a^EnbAQXhi6kRd1XP=v31geub#4ojZaVg6|GS->{+9}`-)HM_9q2UsQp6!
z^;yUQw=d2(oB{)k4^o#wdp8vUHJ7pN9TS&uVFD>%)f>rfT*>XOuXwpxn7C9VYfX#>
zBv~>rw-`DENODSIpXQLk{6E6lUU%_ry*wpjz^(3nU0jNLv1mPfAN26$?&<CB;g!tb
zRj}Z3c>6X0|DgFVXnyGI-~p5)7~X!Df6+J~e;wE2b6Z2UKw)gq^&KjOz$<<5I^kn#
z@R)EO{_*znovy5pd?G39;^4ed%uVAn@nsm&Fedb#pIF9ivApxY{BD@0Vcdo}v<c#w
z-(!PHoaSHb0TPC^(lP`yn;|nspSgN?4KOD~B+Fa{$(?nDmhqJbte>9+_PVa{T`A4$
ze~UCJiW5p1+Kn>kq!K2ajJw|ErD05*#}V5UzdlOW+KS<6Rc5?gIdl_^KCSNG7^}#V
zgwn5#(?l7n&eMDTCAcOd-{cgUtu#&BDg(Q)XL~@XPwivIlK7n1r$eO!juwCgBNfYX
z6)W?UaaGz)?$C5^!>q418pbHYKHGdXe_?s5@=w?(QOMfKo}{8n&XnH1h;J`%cmJ7e
z!K9Gj<M7X4?$QqitaAJn(JHWR`1Rkzx4XYcwAEd;(hO7nSUvpueGpH+)@T$p?@&8u
zZNEdOEN|W+D)QqF&F3PkysPK3Q-m%~94X{aN5bvN*OORcog#`8D$hsx6}^pqf8VH7
zTb}YAoC1^x>BK3LyAW37DVRRDz|zbv8CzMO%p|#%)6hbbOW&~KOn5lR7UgcsdsEn2
zD`Wiy!d?_Z6Ksz$3l&eo;A`dGBqrOR%zO9bRAm8jni7~%&nJthSr108p2+Te`p97y
zINS(~0JHRwJ%}Ps=!awaY0@3Pe}^%oT*McLL&wTElP?ohBsV9QUO)W(rSif7p;{B#
z!b|sL#lFy64pS2;q+%dbz@vAt!Ze)cYNV|~LlP^0)tY@wH{EiMIR43ecTY}jtNMWr
zQ5jCH6Kg4&OUYzjQfqrkoXrKQn>^M!uWZ=SVS9uv=w0HU&ZuC#K(yqZf1Ya?&LsWf
z8@#iI-yYwB546}IWN4~VeW}uHY&GFgrny3REs9U3W@n7y7V~3WID|R#%ci%%QA;R|
zn^X}bF2Nj97HN#_X@z8z*R-b@_84m^c$FRYn6?bDUu|0{=xUj3YEe}ce@h-g%|(=z
z6qn*Pa-pXua=Fx4(>@jrfA;cROhC$8ob18R?txE&Y*Z5&>*woelU9DBUa<3|)2R&2
zg!C0;4VT75=h8J^@u_yY<?CYum*(EsBVDay%=BxKNVDg!lALZUL$8fKIqD?CgU=%J
zT(EuCc1VF8kOk^?4~;e}<s)!)w<lUZ6K=2AwC0i*{qCD~@3ZOfe=a%3c&=DE0`8Y%
z1jbvc4i^<@lo*%lj+$%ajzlbPErQk(7#-SFr>@SbpZfMg!|sBLcKP5nM!yN5^(sc#
zuNV^uH>xp7K30aC1nO~|u!JW%Gvm=q4nw&de`!Z(dxAVKdMfG2=q@cyB*amj8Q4)m
zJ;E&s^{cU+HXiwNfAd9O+N#Zd(lbimP4yactZnq=wG}$#vtOL>J)Z>8IkrTStLFsV
zJd>Q#ev&S)385`|q?Q$!eflvaJ{I{FtJ-kUxlp;Bk4HnjC?&7@$YE?Bt+|O%MyhnY
z9SoS<?b}myoyivFF0|Tq$_F#CkB3(SgH_S{bgJt!iFc;kf6#|N-j_bBG4mi?zd7hZ
z{&3%0<9zp%8K%IGIbd+Tl6obaAd0Qd{qpt8Vtc%GV~O(xnC?m;OR5Rl9UHwea@4CG
zvQZgJd#W;B<YX=|cMJ0E&sq@jy$$B)k6o8~so#$wId>47czkuOr%p`Xnk1R$<0-Ah
z(o0+4#P6CJf0?wtRpc<T-uV{qxy!v!iq!B#7}fF0t(WOd0?h~hr<?AdE=mj@6l%F2
z_&ihl-_lKt%7Yg~js|;H5r`3$k&KS#vN^Rb-kVm7!Dtj|n7fZ8a`N7&sy1CqnG^Na
zd#7yASuZVfm~nTE&3@|3!_jlt$GN6-$WgAllSHL^e+EB%t8tHy(6bOMG7mk--<hKd
z<!~0tu*mpq+ClyH=lhS~+=sue6SSo_APK_#>c*Vmm~DC}Y=q1#h)*Z+(1Uz!AYtHA
z#Z%}_B>>myi3(SyP~z_g_^EM5voH~I72Co5Yn@EwT985yaw_u$LXq9&B0}ZjIQAev
zL6Y_Zf39kYFffbM3-3>>HbF!z2m$ijs#Ymo2L~f<0ZoO1c_<6YqLXfzN{y7)Ts*Aq
z%vnXwTxjsb=RPE1OzVbDJ}C5N(a;xE7LZfFunl-vwG6Hb+3Oc9V?pxFlqcbZ-|tta
z5Ml|&LhR|To$@S5x*;!d@E8G0OtfNtfXH7te?O3}mtdm$t@(jaT1f(`%}?|(J9AIv
zJAOK83WvT-GJbXz0GSpU*U^Ahw{$XqaYn<m%zXOPwwq~v%o$_)cjHPrDb!vLC0`Wo
z+WbKrClbJQQXf~;%jSY{k>iptIeZ69)d0T&wL{tD=}M7H!1<UBSxAc74ZQ0wN!!Vt
zf0_=osBhv)?tC6g)7-bZyd>Q!0hn?28<5P`+H+zIoV7g0-|66z;O-p<mjtCBcyLK-
z^t{0(0rk5XTtdr%@z{NNm%*iEZ6{5Iy$6?AvV@bv8>EwCI)(oa=9UmB{rJvvOOndX
zb4vo@4>Gp|Ty3ZFXurDwfpN#VC4qTk5^hPX-eGP@VBDBnlJwtiZs|;yBH1`)f7dJ8
z(WN;dw=w@8OE8d%9^~qDPYEyk)he8ajo)*|T43B^jzM7Dm}8JT+INm2qOuUKm}B62
z%i6e~FZB2}TtN_j&mo31w6HO-;JM`4PY|crHpI{o-uqVTif~ZMxo!gsGYd6xoG$GK
z70<c&<D=<+0b=_zhzezHWOH<Kml@+O3zrbuI0Tmxi7ynF?RPCFf9+aJkL0)wzW1+i
zZaqj^BazfAX%x_1Gc`bN8T27Ql2ekI(;jxQ{~t>|ZOO8$>#@5z<lu6Z%c4k%{P>XM
z^nH@kmxtH4hmR+u$qJ{Hh3W0vr1%fXpGxvWqbIABFsP=tAEtjiO9_UrSH4I}M7}J4
zcYN1}CsdB#P$pmMfAC~UJ<spLvAlyijRIdr=ex=?IDXSQD+4{AO}+4C5C3}m>x0a!
zOW{rU8gBis&u<SuK2$*fg*JXNq=YhR`u7hH;SW+;80#l=EIUm<{dfBI@VB1oQcX^=
znxrvWH)_&GATo911%|(pn&h!#R^v2({Pcb5o_zhX;MWMPf8AuX6Uc02imX3LtVqoY
zCsqjpf5-G2j(k`7KUSk=b&^l-lS$81R;`9ZSd;j=Nkii6;nfA1LXrweo>G6_Vk42f
zfD`oTI~5iny&K4d_t8g^aVDKOizLGLQzp$r<teJUs*ZT6t>&e+g3#Vmb-_X-li{+F
zIVeMH(QH<ke@kw$WIR59nVAo~%V5~FvXze~Qj-?mS~h~ErLkRDLyd5cfBF!H@(>yi
zu{=bzL;mn#$BkxtbrKyvMs1woyYUjT-H!hG`lWct8c2%&F_3`8vM9P}tSCJyyU|1s
zMcOl)<bTg)4ZkTdU2QWly@jJ-m5`1)=3h=GV4)3SfAF-Lf<}&s|HNC;#HApoH45W=
z4{`v)t1}?i-}gW-LzLE@X;jc_m3fUOlT&Duoc}n(4?+GObR;v`3}NwEB%O6e%w_$E
zHrqT|U@aJ6`}Msb>QlHn1ZV9^Ap;H2iE8sCdVrTBbp;auxl-3`qn-bcl@1Qs!&OhE
zpvQ1ce>a1BF&Dl&S%W)fPhyjPV!Q}8sSgmbeX*=cF2X|fuR}EExHctpP?e9Vndgl5
z)d;fNuqZ(wU}#%PizLXHhQr5i8SShc5AlM?wL|{NEag!KhNC!UAV53n=f8Y-QpP;~
zeSbp3DClrCq57~Z8D|QNXzeH{6VY&pW7zfWe{wy3Ekig}V8cMySqnqIXV~EoKq_Ss
zkLr(wGaQcbY$|Mg&6xj`Y5YXOJB5uQp#;HM+{Cyg2(S}os5K06AfWkDm_P5X5hsnz
zSieTQT1J)u!S`zelt-!YC^jJvuJ;Q#rRZY(VOKH+U)$U4Sp!6Eg$W^lIcneV&=`hw
zf3s<LW0PR-X4LL1;^VGGa1JiW<JclL#9^3H*m+^+PGe9)?}hX$<;O%J!+wSSWqM_H
zfrL?r7p3Fp`z|9yT3A}4BPW|rWFpGF6r7w9GGV59p7W9P#1&MemTmo-*&b(28b~;$
zR|x&On`~(0K;Icgj-=YOsd9+`?2V3Uf0Z#ylXEHl4bOAJUE^+v%eJI33H8gbr4CZb
z7e`ZB!tylaWGVMQDAQ=F(xu9Kh_e9(7;5-(SSqR2)17QIM*N`HR_;)qMm|bDQF0AG
zrz=r*+2tjE5$}#QhYWvRkUO3W*CaU@!fLT{lGraPJ1l2=HPJ@1NSlti1S#{Me}Zq-
zC5naE-Bhg@Z^*e7KX*}etvK(jYE&SQJ*Q(j_<pv>#a!zG`tVeXYa8u$X~heqn4d=R
z8gE)GMe&kFwyWKUQuQ~g`Q+BJkl`rhRuY~;77AsI#C}vpmW7Y+L-n@NTCQK+C=mr~
zi%PTFLjqcsIu_N`TTvM0X0<0}f4^kqEz5^}RRL-dirtpEVz^h-CYx3#ISKXsL`At-
z1C37HijgYwc~%O=*lelP*7PqATOrzV1n!sfjBRne4Or3+3NDxU4D%{A7wz#MXP#&L
zNLq5mPRc6l8luM%FN>R(%@9e3ILEC8R}NPl9MCDdexAE%U6qCBa(Sl^e<L!S`#?=N
zbH}X&%-9w<%^-GZTY#(2kh`zj0$M0-&h!ORi5ow3uBV$PUvsZuT>O?^0apitZ0*U>
zrwq@hxk(19E4j-I@!LFcd9rMp<~D3U%W>9UX)(UD+*ZyA1wFyrdQOKYBh{SK@Gr<T
z105)p{CRz!Rkl=o_ZKMpe_VgNqZpM63k*D6+<Q?@#jajQ8RWPJ=(*PB>bQG*rMnAy
zKQa_5wx_7naDI?nPqKsd^Se3p6z$gEZjrfOz!#&T3Gw-Td=b|?XhwQz*C({^GcUJv
zva}$x(n8qAi8H|L9lT)umd%!pq0|Bhd#a^JUOQZu@1Y=Djal1De_FTL{~hXSPolur
z2&QiGS2T{-9Pj9_9mN8Am-r`#Jx8{uwB?@`!;V1z_6`1-0k|W$;RXdBdPOc1Y{=Q+
z&Yn{2J`YUd78BG~SfG&+){U?ojz;@%Y2zqXK&ySmJJ=#8y-u2Y+G0yFXM<PSVoT|q
zEw*o$Erys3m*fiPe}%s#mVo3$O7}BvVK#6%FIRfG)L7#N=$64=9*<FRU8rFtgB{8c
zTCxm<{oC?*-P|aJT`jND<EcYc#vavAeZ{n4zZmho^AdZqvulyB#|B7dUFVh3`#672
z6meLA?d`;0u?>F_`b5+ThPT=uyFh$iZ5N=OEzUW2cWBgce^H(kXL$D{Yi~Q&QG}Kt
zJovi{ZGIi0-LDd3fOCK5>y9y|qj*c+5%=eyr?{1P<XF>7)M5F4*Kl@L`P#N13U$ls
z)v#TCt&hT#kveRu+%X^P2%gEu`svRe`*~l;!_T$<t3H=HS}%ylmtIpm3P)@facQL?
za$E=MJ8GyUe|Szq-Sl@kLvh7H2qQcLR?qu55|#rE-P<E27xzRflr?ue9tV5}W&eLX
z?(*Ytpmf;C`JX#Sb<?r9rBLrY7S{~<y^h5x+aTmSj>W}ZvKPcfhZ&vNRpf<HbK|I<
z$4!}KXrOO(h|Juqc0zk=a5w?0T=`nkLP=1UJ3Ts!e~`B(klb_YK^q;?z0HL>zYJ$v
z?rOhjt0mW6rtabD625iX*rH<MxY%(jN1i^t<18Ro4oY<bEY1RgKWHqw2x|8KdGF@>
zqk-Y{2V3ZGb288x7L?kJtfJj(ReN1v%YECHp}_&aI)D}9yVs$dWIMjc+RvQH>+if+
zzW@L!0-C~?vAQ@Fmw}-v6qjS;IKu&Vw<qK{{|E#wc}`H5vAQ@F0XUb@%mx&fU9~JH
zmwW9vm4BHp|MdIYFAp-aKJtl5XyXTCNia8sPv(b+@5EF!^QgmJKgHcO1U~Fz8ljgk
zt8hkVzD_)1!f1J_k)QhU8*@z*=%u;gGwHlG>|;i1Chm%mIf<)m{%i{uVXZ#x@T9A~
z!sv4x4chp8cCcnWCM8Hys}<6D)w<0Uq5vk?tbbxlhE)rVYa>AFQR61d<wm(E=_KkB
zll25@=WvjM_;O0C;V0X3`Sjb@P1!(uA-!7cFZsgYWr1V9&n%h7h%KiMtQXd6i1vwQ
zJ0&w_!`3dirAD+?AU|pt%xe66)^1g-qE_WZ9LfO~*kRGMsvJz_J7jk$Zgono8XcUH
ztAFPw_96*DY2hrp4=7FoD)aIp^?C@B7UXO*JEn%@$aN5m&T%S%l3ZnU_7Wb&OfEao
z7>s2kYU7-ZN7+iRteW!}sLAolx1yafu%ek=9gZc~<H)%VOX5#Kf_+;og7`E`2d7yY
z=?(NUO8Kd+X%-pubANgoeNhTluW}r8N`D(?_-?$!6q2hWdtsUVS357X9=$%<pd{cD
zOEW)`!kh3h+}dxiZx4SjG7r-fh8^vX9}i)GKnr9207iIahM#{9-yeQ$sn!)Q3LuRM
zT-E{@X52A#rbElLk~KzhnEpjg&7OR#;zD_$5E_C-^@|{4*{p6RCl5%zB45S?et!`c
zhJ+*Aa`7cx>zr5>jNUB3>_sJ=>MH8OcU<+|7>$>5j0A%40*;Y<wEWF1hX&c-eV^qp
zOyZK&><-!IWsbh=|D_C}nKI2nr>7~m$FbJ{m0~%P=TvF=Ps+;$Q^B@C$EpqS&a0)2
z#W@k3J&BpX%e6pK4ur#2Ay^3otbYKoO(|NBN@5(Us5OOtu4_xml19e-{F-3pq^PW<
zwQR6i|4O1!Qdn!Q1(kZ$7uLZgb^FD*AEwHbXl!s`bNy4t&{1M?k-6;X_n{-3NfNZH
zKu%nPJb<j8(Sup&+#Enw-q$TGRe(2k0&jOK;7vfnIJe+uA9xeX&Q)xoy?=G!tpPcs
zs}BR-?fwn$zCQECOqZR!3ve5L2)1Rc6L6C!;o)C_x7o>rk=Tj1=d(BQHYuv`HYuI)
zHYsy}xAWW%yiG<|css4D@pgXF;O)dgi?{Qxr(>+0*Ed*u;yO~+<L*^_XW6@QZM}ZV
zq4i#Mi(l6TTgz-mOQfuGvwx^8-k_Z@G-aHsTS)u!h5hw~ut=PuQC-#oGa@5LEmL=F
zJ6N|u@ya{49VA)PEX;k{4gwGs)eZZOhT=iv(agSK=h5<U+<G)CKA`vL=eaZ=x3gdA
zKKkre+K*Oe8^>S0eaCbRR?fhl#uR7PV0P?B3J70?=Xh&)uB?us(|-xiQDZ87djJkN
zV-etjw$Ev_s0@>2QJ(jx)TOlgvaU)ackv>66bus^9;4PzrO*X=)#Jp$WXP#}r5U9G
z$@9ljAD+;KEy)_i3z`qE%R)KTN_NnE5C<(lyXb=6gS6rp)(fNW_Z}3Ep%#mdT92&d
z+<HtJD$t}FuT$E--hal3Fp43e>TEjzoj{JbX4OX9p(xZl_^`I4-M{m^zdt%tM6QQQ
z7wB<@Gq7nH?QBCrCu=7F^Pzo+l8URK@9ab7v;Xfy{@Z;BdPY!|{#|`Y!_*!7kZ=|O
z#@w?H0Z)RIZ~G7v&W>=u56S1ZTBeiSDCz9Ux7@3&pX`=i1%HY7K4!S+<2fD*K&o<5
z&oOW56&lLdbV1e%1JF|3>Vm?VH5c<%vRM~2FN?WvLdADKp2EajftA0X-`zJglso#S
zX2mty<66cjxMg8Od7k5hqibdRxhkeA9TIlcqAiq)+yW_8Cy|&0Vv=}=LfOF?K{QnN
z6l%Nx4)|upReuW24%YCn<&H`H^T~&Oa9<rM@8nDx%{I|FjP1T{!W%(Vg5N$QzsyEi
z<8$h}PM-sOzT(Qmf<>y6oK|7{IeYxg?LGZ!2<$q!z4zy&3J7U88^P@t4D2F~rk5J#
zmf!6(xIJN)#8#G-2;F`?M#r~{8VVEd8&a<)*3F-j7=KO^gYkBkEh-y7FR`@23tDL_
zP`CNkRwt<2`>CmZEv15Mn{6z|kTeU;?*5Fm(b{^}o@xo*5jFuC4(7Wa-vZd!p55p8
z7LXV}Z62RSVN?Z;%HhP9u~B+&MQL{4%b2-NH!F@ZraI(3HDBrM@5)+@&It^0+btD(
zZoAc8q)wVw$?1K-woqPE6daE!cUet{I)?hiu>D%TpV1`<V7NJF<dTHWC?#ZxIf27`
z-WL&@T=Q-miOW5^Xl`VC#XXRxhG+xP%AIq}KmF$`moYpi69h3ZG&q-0H8CR&F)=VO
zFHB`_XLM*FGccE7jwUC6#aq#CT(=Q@_g7q?FX=<L9L{jaDFOpdQl|kDG_X^kt)T~7
z-X=~R$&l7T`u9CU?o!(9uC0}>ECgQ7-R0#B$(b`}NQR=QO>v?qrB1>wm`oh4No5jh
zQYlkT<WiMMb8b?eN^QSZooxMUnTR+wu}+jUKXMVR*lOz%x^&%tJKiYL+8RD<>DqNl
z5lUQXni5fatq??|Nn*yADYoB@A&NrG_#z3*BNxk*vh)TOqG=OtN+s3SHcrXdnn)OG
zNli=>r^-##iG{K&!&C@9@Op)W`S724;w;g15q@9yN@e28T>;uwA1FjaFG*U1I7#9z
zVjv=2Oq1l&#y~oMN%42ppxb8#5_FfEn=YQDc3qfjbk~eQ-*{{yW-6yXE-V_Rpoxj6
zoGZquHBDJlGYg+twT+w6mr-tFnKY)REM<t=HRz<NbiHUG)Lw%jn)}={Pg-(u4Ft)T
zHKIeUeX#h{T3T0(Z>?Lv5u&o|f&-S6nz3+!!hHz%qCi=HYg?yMN;eVTT5}()z*OO2
zO{A%&+)Tu*Rqk^SZ?2hoTbZin<k}=pwe)p{`PHcRP$t7_HL*@cwG9$8XiTjwelzLg
z%`{o=69Q+nLe0p`lhwYaa2T^vU0aJ3>IP?jJM|%DpARm*d0ji5%idD^45*t^x_7F-
zWwDv!>xP$qTYJSfM17$Yy-_l2a|*qb``8qpq<wIRBSchh?>&=2ud0uctUHWWm#wMw
zI84-?&O<w;?&01+V(wi+6{3(IeeuQO>lYrY4k*tS|NPg#-TqMBmTUo9rnC2N-<~{v
zjHc7#tMj+#7ti0lJY7!gFHDOcUanTli?b>9$G`r6zWV0*>g8%VQUBuU`Pm9eo;JUN
zX|Cldd=BclKls$RgY}2(57Zxgk})~`!QCBeylw^*iE;A|&<B{)A270n;9Nfxe`nnR
zhRE)x{Q)S0wOn`59<)EWA37S~;ql|e4;SaB&zGwgiyyvzy7+PV`>M~x;>Z8JTVm`#
zEiom3XRGDe>aqb=ZhbskUY@_dI9*<@0nfkte);;%%P-G=f8j<e;;~pdK@S&r*PC%~
z_T=pBeBD>+COmr4q1<P)L-|I^L((B#wW~w2jbw31G^Dn0Cu2;_fW`CouU7u_-J7%j
zPKz(kFJ3P%n&t6i@wdgdi?3c#e|WY$UA+K*vW5mOg#tmHq!Pq&r&5SQCyOT&x_`d-
z=KRO=#n;nk1jrOD83Q>T7=b+b&yOEpo7vk~0Y?uGMoJVE=$cdmW6?EHKn(0!p_~BL
zsTSj;TrpF0cPkiU2RLqwkQ##ld=435mMar*u_6UKdfPr5_N#@XPV|StOVL1%1~~<P
z49&9-g4j+TZ4G1bWORSJdUJjTgZ}aCTmSdj&#TqD%fBoZKfPJ~{QgxqJ^yv_%llv7
zeg1m++v59|tDpb=ZuKo3{^CEtfbh%P#jCgHuNJ>ya52#0^5S%Hxm;c1hd+Ji0P{IM
zi(o{L?z*_e$Ad*J96Sdw=b*?N$o>I;ptduwie!~CD}WclhIC)c3~MVjG^(;T0#qav
z!0I6*=3ZO?TWD-#y|^huKj#NoTuFw-70GifE+RE0qyZlv78#r_7{D1*7mVc7EN*g-
zaz?`(z^}O0;1pOe)l}|%X_4$hV|44KMWzly`Y>@@d}&kFtEFum!DMg*P?C~=u2X}Z
z78@1;P8<pbc!Q>31OYtoCQVI4#Gm`7CR_Yri>GW+lP&(R#bEA0CUV>*6JEyLhD<nU
zU6BbJ)W9~>Ajdc)8qy5Z03~E7sJOD_fT9j@3ub`y7#+ULEhGEpVU>%83VO5&t6aee
zYI&HjGDcW&q&X0CJ*`r5@Io<v8O8_d3sMLMcK>kY+^Xa&xblpc^25Ltr{wdRd4_pS
zsCgK8V9*3aK;0a&IXpxp13GG29lCsr;|jVI(53rubGunbI25bEO}R<aq#jIs?`U`p
z%<c@0N24L_K|{JF8qzg1Tn}iy@3WR@pdk%3q*>4q#6}8Ws07BOC9EKS`M!_A?e?$?
z60TOka}{-x0XS9cj(A9+`p(7%58-7@4->Yv5lE{0!uAN_WHAU|jfg-EOdf_3XRmjp
zrL4OG5p#5EKUE7-I*amcGWZFw<`b2&q)={Pden3?0pr0PP}F0P$Jx-Ji$2C5ueZ3u
zL98#yO_~xWD~j9)4sOAJ*kS0~;(i~%PC;*LK>Nc^)b)L>tsSo|RQ3HxfWrDg{Gse@
zavcq^DeXe+dSpjKY?5mbO9Nv2texC8*cosYuLV0-4~qv0c2H^rdzLviH<&1h?_w6;
zapg%AHIr|4z@bsz|1r{vM34#8XKY{{_^`YkXloDr80*+-)<cMYi#koeZzi$j%v+&x
z*t54{>u?AjA6rMq3Gay$zEPayHH2Kn316ooX|to7lMbm2sop7el;FgAMcI`U&h{{r
zJ!|#gF`I+3tM9aeLq&tCcG`#%6N!T=s80FBxl+7xK(^Q<%j}gy$+xxQ2p&{Quxphr
z0UG6jDqSoGmKQ~T{PqTtA6tHe0Prp(Y4k$qMooPYjBnP|3uxgRHBBh5qTM?v9)`Xx
zgq#-#w+_V)5ssq~vcz44)OtIFEPNdyM~TLRQ}FmsM%gB#Y?Cx39+C`cAg&C=m4Uc2
zbjZ+=hmJgS<e?)E9eL=;Lq{Gu^3aiojy!be(4j+z4jnpwbm-8bLx<|!%T$pZ{Dj=k
z^8|h;sIiiK=sXW~B^PjSzm1WiN~W35^Za^>qF9a}IM2J86-N}y=Eo6ql`;LeC#6pa
zslb$rH?Q8WmK|6VW#r8Sn42MQ`XX!fys07jm6wW&vn1rr4A|SZ$fT0|zi2E-Tr-@x
z<S3yf$!-FF7-;~dK4=0GDY4`#<QHQy@(j;$sHcGedgwP$^_l`s4+@S-Lq5qQjH6o^
zlGwo+Re;}L%9xe9OC6Mv>n*7wD^dZ3)i{F@0=nQQ_X)*4Y(Sw^@mClwUv^1Utz>W>
z(O9gq)tqKHfU7Cs->`wLAl&qz4ea0^tEm-3rZW|Py}J=8tr|<-7=fGu=#zs+kibzR
z5@k`&1Vd7lIoKq$Au|Y|&u`IzRB~m%*B5HqjSEf!HgDjR@Cy`<L^PX6FhH;<N^WDA
zgo%h8>hA7Of*01p;7KM<?@_JH;1L+U7eyNzl1Bs*;Q<?*QzIfln<AJnxCLfnzIpJi
zhl-kiJfWb5LvRJ(X4@^!;kHt+c(uYGKu+Qe(V6@d8hm481jD98A26bTb+ig!5gk|y
zngOSzRuqhN7y+ULuhfG^)Iw0qioy(OSx_0-e38Mb!SAdlkn`l1J4;3;*#p=LhXhE0
z!C_7uEmY_8iR^om?OFy_8a73wvEA@2)$cifMSOVH;t2GfVKoHt=v!djhMYWE&_NqG
zWj{<c)T5|Al#t)%n+8u{grvdRYF=i?&OKNO6A~vYU~ql`%*1q#dmFXU!;KXz7TB#i
zU_~}VToy!zXa;AT1E>J~!LRCWY(P>75tZH{8&WBNDnM#xSb}70b)NWSkI%IcN{CB;
z<AX*ZGOD%1OUa-UBC}+!wM$UMBETa6;7DNXj8=9b!m=^YT1)ifQ$uMpheL?pz@v=~
zzPXCt58411sG1xgP-CMomXzlR&kT0W8yj*$!fKb*_uIgUQL-Y3LwznuDy7+$0uc$A
z##+D~P%3i-&9yy1iAq}yDHB8-%u_0V=+)h>2mq0*Uvl4E5dggd&wjw_35$#B2}7LY
z0tTvu=>Ka$JDx%wLcLItkqOCB_MOn(4Ev7Q?{g01J3gEJKKC^JlxHDdUc7nz)ADlh
zs{L(KEa~&_uHVCGTWV@II*riX8|Xj>@ZEBgrm4W$Q+lwIi=(&5>Yn=ym2bCCMOLpL
zBIwv!Dq~ZqjE$#q_3PMpDr4iRj*X`}HlFGn_a4^KqJE*%N-HmYmt2(tCi<5A^@yDu
zq(<bot7{JiJH#F~Sp2qtuCeI<KfRxq@rOke128o?mjQ_&6qgABIm3Tm*MEHc^+guZ
zr3g0u8GrHrzJ9#?`(;o>X<=-*SgVC}{_@Y?U*ZP{MpzdvU`5dO^2fg}zrOs9b8XG!
zfy^YWHD+QaWrZ=GxCf!*N7+nB0wXD8$UKr1ApN9KwC3_Ue#^=TBsRm3KYqKgl^@?n
zJuC{3j_(A&#A{-jfx>@jy3=wB-N`}*rwea%ea&m~+x3f;@$e9{7U?p(rB>Prl4FO|
z3h7|T*o-1N5t12=t+Hg<u>3!()kOB9VdRlc=9<)t%HE1vb7#L8AJwh25!#K$HCM#f
z@$l1D=9;|TK=glCjS4jya8=C6E6G<}u#-ed5D*<>D#^^jE$M%vlp-ev&LNpIRGE_`
z?@5{>_kvR|%IQ!9>~~q>vLI{l!Z^#hP_~&l!wp%1(i{@ICuIfD!WsP;Sz)@<maH(P
z_&Hf&v(!UYm{Q!$3Zcka!>ZgLwTl|{86szeN%Xafo3LA19gj$w9m?u>M1b89-Back
zv&wjc=KW-DGcJD!=SE}Saz0SjvwWazIG@MBD*>SRu6y#B)B<$JAezMvV%r*ngV-T^
zJ~1yVQ`HjH#PruMH3mlWuctomgqEI5R-9=$)gdKzGpi3QAH&YHtdY{Pb)9Vc(z3Q*
zSlJz33>@W>oaU-vrZBqKs*S6<xq4oqv3#_0q^aSkTv&hMTFUL3-bi>5pi>))GgPzI
zn0i}PAE_$oo>|MtXYVm@)g$v-DO_FXS=t3$7kUX$PtF|GBSH_T3TnRVuFzA0j-=yI
zV=v1+7`Muu@^+L_mrE5ZRj+fEHfeLa12700c+X|n>&cR$PC>_|Jx3U%!up@8URZsa
zJX5_knI(T)z-BIKwNzOgtdX*b=cinKJkh*Ob}!CZt6$O%TdP$FSFF7nEi=wOZ)Zbk
z)>_qB=ve3&ZK=q0+>Qb|qk{+#o&@L&S^?oR+B~+?20+IsehQ%5E_DTTjN-ou(A`&7
z3i3!PWaObt<~mvV5kXjxb1qjwAV1bnj<Xs_jR$`=z`A)S)>Nc?$IyH|hAe6Sljstt
zWWesudFVdp4N*TGde?0ob}mz851RpA;W$a>`Oc<TRjZ82uIZPJ8YJ+eALIK(hef*&
zw71p*eIK4IpbFKusZ|oKD*D_3M2cErulTv#0ja1_xEKsms(Ui;1gYpPWee<N+gIA@
zfP{bU6g*mK&?5!tvqlHhl?W3)V|2iF+E8g2#ZRd;+oi6RhEd$BG|G+<-rZAymuo9d
zsx(XKr7?irUvWw;*sioDMDJB+rp-z{dX#{M0vP&H0wC{az&xU#k>uK%@4Bm>p^ixV
zjzjHdvfS+~A<Ny)64J_yZkx_vz0(OGbZ>vv;#te{1Sg!=vpT`@S?r!CIH9%SyX@`4
z966W+&LwsJ+T8E_i#C&aLe^$!o}lgdd4kcMb<YhHIJ4se1w<5<>dNO16xgl@1`3?R
z&lo5$84eB<IG6tq6!<`aTj<<CfpI$u^R!bUP$gwB&r0EF-_tOUcA?g19_CF08A*R0
zM|qXdt919KTZe$)l=BuKIHlMK1ZTC14>4gaQS8$wqaMa}O(PowNX0202NJI}0@r%b
z1>;bAF2Dp<=64x{iZWGx#^}Kl3!q}9b>O?4m{`Oz5??}L9OKC143bUsA)1?{v*~2w
zG$Tzk*a?&{)h-A%fRI~+@+tqlO>2K~%(Z2#G}$8ZJ^`LJbudc(FAqvKV&;bpOzxw-
z`1%3zqqzgX5f#O=;zSP~l>|&Lb4)kcJt^@5ykkrbmr$UVfDq7gCM7!V68tAjO4v>t
z3M8Z0Q6P8CL{v<VW`lfZ(z{ZgP#ShR#ZbKEKrdavy|i8@GK|xyN%T0-hr)kBj`Hjo
zwNPMFx!!Ts8gt{v?P8~CG$*~2V-)2B^@Cj(M`2o1KAJ6a<$TxM*dV7=n^$}LE%rwg
z*A|dVHOpb4%hOa>$##qbQRQU#KK8b?jn{~6(nLOF(P$40En<d3-)XY5(?dE=-`OMj
zrZ$Qf$C;NPvTnULL-{*WWE+2bRZXs7i*>VXuxYk1So`>pWbk9>Qpb0$(~E%9nvPq1
z^O*Xo*ZfX#l+1Sh^907+S_qUhdaIDcO`@s1_%oB28gB39?l39heOb4YD5<NH*g0Nu
zl{D0~`X_a*iVQDtN7EWJzVz^frj>EtP?H(OlWOvyxsKBx^|Y3Y8t{LZai+T*sL8bf
zW}F_^YNa+Sw?gx+Ry7!TFj&1(OpZ(f7}pJx07kJ`&gUCO6`wJVA@b=hBREgM9lLR_
zvEz_&HBnQmf}{uMv;HtRjv2rkP6cwb8OZg0ZS?z2e!bpUfwjV+<y2d{#%vW1hsx>>
zIAlPmUWz?%$SLx<&VGMua7Y=`KEuf}x8r<ck3GY|?(la%!il4_PMP$u#riFea2Av~
zgN9mj&7-0GBI*V-H20JxI-;`Upy~utG8723#5o>BN<JpDj@u)V()KaUe;!%_?S*zv
znomG{4uS3y<`ZnE4Ooa#Jb6BGU%p4oC%8_>=M#+6SuFI}IYob&I&4349kC*+&OXnW
zQvlGp+3xjqRO2(w^z^l^Y%J9IZeFO%>W)H9;T}5xCgq+&O)y0T!#S)KYE-f*)J5hd
z&y|(Eg_;_=pgK+Vet820D#y7*Uq!SVzPyoj>$N#ksS5{Isa;d4i)_a#b@0N0dPJed
zAcE1H;qf{xku!fwWjcRxx@8xOsMO699#fuFHqS819pnAO%N=11uQbHqBf?DAyw6lt
z+EA_gmPh@=-*&=d{KjqBKLr@(1klJ=D2wguCpb{)rfVAI0^4$&><%pzH1GYm$=+ss
z0+e|8X5ckS=Vsl=$9E07ue+4FQS)|6gP6uz)L@(BCu}^Xl+hVB=*)d0I`hv%XJldR
zsiY7B@itX>>bFz1oBia+_q+2%>e+FOJv+V+lk~G)<>$W%75)jwi#C0i!5BFemi;yq
zm!Jwcz<+BzpZ<LR?MY_Sr3f~DjVJ!=)BDpuPlF@c2(+QI)(Gi*_t*EQcmos)>q4ij
z2*!3l{_eg${m!*c!{il(Nm`3og-Kh1=t+AJCf-V60x}rLA*Ap~a)A7ksOg$Z*Wr?!
z5kP8&k00MUcIC_G(GLs4gX1HqU*l1QW{_}(o`1AlLKoXqK9g~9rh@`+8awI2<@9W2
zoJfG=QGeiW4>W{6*4sVM=mh|BiD+f91t_|PEJl-l>vQr9-Nd_)o{t`a-)*;vbF3OG
zD8c7!rc4*EnWk6+K&GHA;Tuhgzo#z5$%qNlpMM?Q(7ns2I1_`4b5`p|kL5XgIG4(c
z<+ol8IVJ@J7Y<Nvm%$i06_?RlHWLCdH<uCk1SfyxTFs8zI1IkeQ}_i&6)EZu7zCKf
zb_(omi@7cKwAl94!}k3TOUaUBQMThul6HX}JV|WvS0q1D9zA>=^ziB7<?Z3atD(WG
zV8P??_GJKmp!qLozUl1X0hBW^ynP$~e9{^nUoU)-*2eJV`Fr3`dw4Vucm=p}!Qs&p
zJe_~m;<2tVcBH~ssmmu634vGo;C151(%_k|<KeHj-yd{QU90HQ*YqKOeSCZP_hB*w
zjj}EdP6vg7hQGf(qz436IUm_R<=MRN{|sLq{*YYfZVJF|(#~1ixCx!IHVn>1WzwV3
zO_q%yHpsGj^h;KJq9E4C+rinO42!80ywHClQ~8mSUPp5kG1>Iz``1C-`TXg`*^vt5
zC1?08U5)c53U63zCn-#)3vUTubmZ&fcu&7BTSZ=x@s$(N=!1OOcUYvAjYe|avWWKq
zG)jsTf@H^i8Fd+Tqw?FtqN^=dmi5W<WGzm198yG2%M#17&09_}PLU7OR>#X7!RCJi
zBN;9DK`Rb*t#o36c4XSzVWuBmqh5?<>y$B3O3A(nCtsz=sC!wYu4YGoTtuV;KJM^1
z-?MOvLCa-;6<BpKT4f=gT)d<+|Hcy8v3xIDB&lg5vn1C`saX!hwWJEsr6MiXjMchY
z%8~U=%H;rarEny7JSX#0z4SSaYUO|I6`r2h1E$!lCRo$DAaR}ClBkcB`{<yFPg?sj
z=iDaoQsieplpyFV#>ka6ptN|QG_3iI)l@n-$zj~c$w%Ro()1$zwH#)Xg~#gc6%V)~
z?e!F(jbAa-Z+U?@gQKO+hPY`tS*yVN&UeRobXa|Ogr`OLSk)#n>~Y#iZS;Q%mu2lN
zV<-VgE&xk;aUP*5lJamTfVS*OAK(Ijw#0CXcn5$MoOb}Uptu5{tub7Z&5Z$CaNPmW
zg5n%NqpMlAu%<voUPmK9Z<ecd=OC)6X>yKxFHj3cTmC4ZD=2o`y8x4G#&IVf$u<xv
zpl=2CSgwlUHn>NwiR2sL-kpEZi&A5OQtu1U3!PIDoa7+TYuA2nYHA=V4WZN2ZDT9%
zF%ooC*CLE@$cfPH_k{>!LTVhX<nmsGG2D-Wad(0kB2=jT1rS39*!kfO5F<G6fEYn>
z1;lV;&69S#F~kV2J0M0-{68RubmV$}J;Jnn?lcLuLT)e!whYr8cnyEhW%`(0&C0pF
zD`|Gy7!yt8PSwMm$3zo6_r*BF%SWd%$$PHoROE#81fHZR$%ZR1*4QSMe}UOjSp1IJ
zQgGNaTkgc{A2?(3j1v$#Z)|q}<cwmSJF4qv%%~i7dvocG$uW&SNG_!V=pvt|3In!@
z{Msv^4IC+4PWcMPnz4T*c6UF2TD1ra#|l>uA;pAQ41_T|@>QxyMFjHmgT&2vKbPrf
zcvN;lYBJW9!?<Idv-mVs=|{?SChw|Czi<|XHNW%PQXNj~DQ<Uqnp9*w5xQL3>2dqv
zq2ofODvy<?YL&f>V|EowR(&sRO&K@dxw7_92FYII{k4bkxk-Pwi~8l%Jl1%)LG)6>
zzCLEQqSm@y(P)iR5IRqkIkC%81lFT*SH|@jAl)osDJV##tr2NvjrsX}9N$Wf(N>zs
zQlS~wpn`f%?|rFjNjc-C8#XexxTR!niyupwds;st*n~cli_h%&z0jH<D4*(Q9|1cP
zhO~mz#<dJ44H|!6`8l%IFq>D7^3*n+&MY?GJ7prng!A`$(}<kLo-TRQh}@3)aE~{Q
zg7c0kQ&3zmWnMNT70bQjhG^G&LyO!7lBC!>oy<$h%IJZPn===#pZo-)9cPPz;x;2!
zu6^UEvTifix~O9xtC!neO=&r9o&uB@Z7Iz9*oLfiv_F4sYD3+m2D&1(0_1b8Q+}~F
z&zxKDV1oeHpD?0fIMr5adS<>zC#VL4q&iK8EA~LK=Xt@xbN#5%y*x?1b{?9}x;q=E
z7&y+Sz>1|AiLp%0ueF(3K$m7mR`!s_5-m=+k9F#9uy_Tl{nP9$qfGpDkH-oBmUst9
z>RzwyxUGL=#J}XV9e7d*_jqk5IPX9^L2+la10LkjnhVfQQ0xYVSQ;)^I-*!bgCrH>
z4c@K^MmrEtQ2aOq1cE#yR+}5oz56U&aNThhE+~otODG6NE92cygTdIQj6D$5`N)*&
zT2JCqt<$lqe$Y9(J4NGFx;f*58zH0(ir9Ir5$AsZlH0R%P}UOJTC=ll+foB@u?<pc
zA8V_irwwZ$)do0&vr+LJ($Zi&bK2^Z11Tui=z6s5>*iz-EVJPWa`obNp9j1BLD9o)
zmJnu#GQL@h6aQQ$z`S5>Uluf8IuZ$7o$TqKCU;h+ir4jpPx)kw{trrZ=Hm)wZe(+G
za%Gpv7&#7?QgR9tm+BZf!hd!C=jXqlcw+tS^Qo{Z71PEKhASp53}1en{ur?1$077S
zai+Y}!~cfR%!Jh72hNNT*5K1C{2Bhljvs#i_VsUH{v{X=`!99n12CzDcf*hPY7GOc
zxf{NxcgLJmpa#J_R|W>hnU$8?GO#p!8BTz@1SMtF?K`z9W*yg6Uw;sq83$606I4x^
zhd?p6=J~$YjhU-)iXc;AgWiZr8cVM@K98bT`qb~aD~|flaQZr&J|(ScL#pSDW+6)6
z;Ln^^ac*JJpeoJgCd2U$E@cz~JD9MlGmR5zQ=3GgP0*BN-dpR3GF1^cw^TysCM1<9
z<fi;=+ZUA32#PJmLVq>WD2Zi;4B^K~B+W+cj(>mok%oQ4K_9DUB=aPcX96ib8H<{B
zf@lRF#+9A(;nY3%PCiUtrX_yRcfkd($}`P*S3?A+{4`W71<{^H**@)6hd_yy+@X%l
z)QlR?t`#>$4GW<xO91$!uaX)Ji67MvRB^#4ss>)pD3zi(6@QvvgO;etV-uL9JW`@b
zqcMB|j1;sTyJLjYE)h1>M1jQEjhf6jZ&{(S2ho3zpYYX7<R0vV2>7s`AKr1LW>f)O
zVu)!qA@iVl{ekcd?Y$k9cn(1s9l|+@ixxtF(jlJCwEp^WnKgxtR&}f7qwGP)m!t=s
zbU`U3ubNG(`F}b|6N@r=&y+OnY~t<wBn3o;-%Ye)fGDCAcH0225u5z_OzAeewp_ND
zJsOhEVBRz>(_u1w7WO$jrOY!T05G0cwjAHG3sK*~-Agq7s(<B8fk%j<iQpnYlc?;A
zg}mq{3mS6K(}MO~Buj1l9r1{RJjP}Uff*V526frOdVgY>Mhl+eWV51DWAZ8Hqtl#W
zhB6vW%9#{8f&?(Ga~*V~vrWv2EihT@vmjc@%$F=qDu)SQwfuoub#w`uS<Kq4RHov=
z)ST_+HEDqKoFddc{7MPSgy>e`K_NpSJaf|aI^B~HqLn7C@zFBL4wL5evSwXg1&E_7
zeZz$EIDcx^T2s2L19-JoHm#M4f5o**h`r_~wQOh^U3UquH|)$KiV+?}MDbhrjX=|J
zMF3N21Z4!{J(tmRX21Eh#to@Zc&aLefEyOVcIvJ%5+d%a-602zAiOCGZ;?taqBO!I
zEeZw2lj@RE6;V?uYF8!$d4{BFFj=*U3zx>JF@NyuVr{FQE*;a*@b&U;^5P>R(CT>&
zx{M2qP@lx6$H(Rk7yXZ}OatezqST~(`XW>-2-b~ZRu6yvd-!_#)Dlf<ZJ*=XabvWe
zuAN~TQL&~#2)5QdX&Xz?$U6|y%)f#aO<F4JA*)qVv=kdNFTDYUMR}Uu(^YGi44HeK
zI)84aA}xk<SE%sho;h2-AD6}ebI4aiS^Pqud5Nqd&uT~9qEYfI*|D&oP>+s{E;}>*
zR29g|dx}VJCG*cJRTzxtzC=S9kr?Lg;|j$9<W);`O`!(EAV#QW#7znnG2ApfENC>T
zf7)55Fy5uI^e*?+KY8D0)M#20ogBcfuYV2KSPT@NG{L{)>d=yL2Mw{A6`@o?1t$;q
zBNCmo1PtQ&Y@o^OKAD8Ts=h=YZ*s$8#IlyQq#8o{?RqPiiv7N9h}pHKO5ZXBU9GB3
zXckwLqFB4CDHwwX=0(|u;Fq}svVp=%7BPZZKPp_G0iLURDwQd*n$Br0)pUyqUw?ZB
zi0~TPK~kCG%=>Cy6@(FKhiFP+g0wDGebQ_@B;@syu~v<{N>p0J{2Cou&2pcdXAun;
zsusGD>&dEcyS$CE#;sC_qLt}1)kon)JdX0x-e$uz4V3FsBkJM8btC7q>f-e1HAZ*y
z7gTf{r*7d;0KmTH01iEl&8VXCEq^q_4B>I9R<YR*ECx6Um;+eoK1e5~xn}aUyo^!G
z$d28eNCRsNQ3(gfk%ZJhYp|bTEJo2Tx)ECeToo2mQF*DHxGCoNaWJC5#-Oq($1UlX
zf`CLrY-R65_JU9BOgH51O3R0&n}+rk>87FZm!_L$?t@WOI*AL>>33)9>3=sRcH8uN
zhH!u@14~@NMq=_<;BwuLzP4Ug=-yPSJtinCszRjtIYGIZd`D1jMjQ*uH<ox(I|iT+
za4iikl2+}A&B$Es2L%hxmi)`OCp0at&Ur7qtFNvI+*ChEwL{FC0yjB_>P+DNfm<15
z$L$W>&hH4^RP&)Y4edjLTYoZ5I-XM49k>IRR|js7z}2>M9}e7x(e`{xWh-z;&g;M(
z&ssK3KKF#pU4``%P(k3ovx@_D2VmvmDHKg9&{35M=dk6Z3gZ5LvZ>4YDI-G)Nx*G&
z7Nxoqr`|fNUEO)vG^GX#tBq`aJU-T@($I$INMN`owQX*kE(;~l(SN&vrdnPiB6F3t
zlukMzovND@%$7H`5d%uu&acz4TRC|oGn5eBJpC11396G#(Kwte2Zy4(W#=^8ctQ&1
zJ#?BmQ0xNA2D6J&$ur@vISjhXft$6MEIJM6jH|2sMMihWJOGYBalb~S$Mf=C=XF|0
zrnv3QL2lZymJ$s~%o$QN#npeM(&&ArSCV*Z_QaBxA(z}w;%9IIch}noPS|#)Ww+d=
zU12qvg6gRH@j5^P>AbX*^YBz2*(!IX?c*4u3Y?d@oT9hUiOct{+K67rI{A`JMR6<7
z#MyT2UXPrn;L%E0pZ43aoR0k1IxCDt<78dVB3z-owsw$M$k6WeV`_f~i5Siq|8r`G
zX7U}iLo?#1)DHW(&b31`VzYLblWNtJ&lz!CQ!CIhBHts{O|=3YJqqYlaqn6|YRjyK
zy|n^FZ{MmF5EJQ+_MuuqF|yt%fxV?dWb)cl0Sg^3y15UR3Umv?bW}D<g~WMPDnPt#
z*3X{4xwSOl6c%Nt=>vbIfz-@_XeorN@Ji2=28r)WRThA&%7P^Ejb#D1SgJuM*Rird
zD?m6U=j?E+EXeLf*UP;B>+1r{dd|&FpL274-!^4oJ=s=N)5|ys&&1npywB+q)N-Iq
z!1_t^UQtKR#3VYb>L^?$SmjdVjYpITn4zsOEf<Sr!r_!v(3gLBK3*1Z$&Be#w5Kc(
zba~o#BX^bs2%3DqxYr^(VSxc+u=W-eXBOwZ8QEd&qHn#mB!vn$8t#8<siC~%)>1R#
zhHFET?58?wxzDwshU$)ML(PayT)1^9w#X6h60YCZIV56IobUn>A3W+FPP**V%yP{|
zoTT7}l774HR{np=eHc2vLtN0^hdFX75u-5{Lk*>S>o#3|Y-YJT{uW@F*?-1r`M#}s
zX>3FK;7&dFvLUd0qu$<vzOiQXs#VKj@U0niG`4o>Z413q%c}iTuY167!I|(a%ZE@n
zIk0O85D3@d|3A9@c`F(auAHv-=A-NjAl&W;;S>a?LS26c!FyZWiNJ+IlwkBvLE!Ws
zWHa;6L2wxtB>zK1Fsi)}Tv}d1+&K2oPCbS?=omeMz&TS+brcc;m!qu}_8x-36}o_E
zxmZNtyTL*0(dQB<uHU40#<h&@+@#mG6Y;<%y+!oYjJ_6)BQ8}E9t6gPK&Y?Gjc8nO
z#lYGJpz(i(@(wiKjJN@fdo%*!?hMAKM0a5E>45uOp(i@UyLMcmzXp;!fd=LujL4^x
z+yKa@gb0uC*5nK>5957MU*1I^=UIX47G=K!R<FrC2&mUntf2H|g-1WA$xJY8%FGD|
z6h&q&u3YPHpBPR556Vs8SPErsWOH<KWnpa!Wp0<w7&!=*t1c9msgE5fm)ayblmYOU
zZzVY=f7>+X=c|KN9T+JHXdGLyw4~C7*F3_87M&7k^f&@uf)sqW>3=jqmLzF?_Z+0O
zLEdNnbzas||EFE)6?sAA#<^MeB)Lta9IQY&Bb1(2Kql){=PT%gvaU^)yh>Z1gq6m&
zkjJp8Us!!vIoaAe>!h{{x}rRDlz&$P2wrH3f0OpRQFm952ttuw9U-qAuUW33OWUrT
zRJRV;m(|<J&-|`zBtcj;^<o89O@iE=lK|O6V2T~HKd(}P!+6q3W7$0K6r)B7=~U-K
ztM*Um!81Q5go!?c$uoGK5<%QwUipbP<w<?dc2%$v|Is$l_x58)D`8=h7=4!Xnr5&5
zf431k9D#mbck~v&KfiH;)ZDY^IQU)|MHNqDO`O+aJ88&BDEg)}ioEPft_=uJtMeyc
z&X?+JLMhU-1$05Fq_>0+Phc_zm3Tq&6%p%5Wgc@J<4z0USEHo2$2dt6HlW08rGR9O
zzg77fT#Gr0F~V_ADUdIx88CiwmYSdbfAMWQ*OV9`DL^e4a9w;o%&Jw=rn(77Q|?WA
zc#1Bt3ZiKUZA!RwiFVVu4J)41C~$PY1Lw9{sE}5Np4B}M5|R}<fMwqsU>$NekjeT6
zQZ+qRXJnG0AD58kRLD>85%f8{_c?J{Y+SaOzJyr;p$zmXv`e2h{lzCYB(%~pf4g5E
zDVsV1&!s+XLNGtsNy{}sr8Q@VZu;ebjlp+GI>0JHdfEVMnMp1VSfByT;#IwFdP$2O
zDMJ<Z>nP-?vCKBGXJWTY=Za4#DU8DEci9$EihV!_WIAW<IKr0iNiC#Z(rUF<wlj7V
z32WqS(@~uWT|e#t72Ay^i2?%jfAu3)-%ApXzK22ADaGVSfyTRHp9b(t`$Nxf=bZaj
z5?tu=`OTnLlM}e1gp^~ayo)*8U6&ki0G9|1mu5B|yR&*dPeIhedzwbQ-KEhG*LP}K
zJx(srS04A7buB!|0|R)iEU_@=W$gvpz>S>S<=?gHtZ<fqpatIazRRPvf1S#34R^U{
z>bs=NBD2cvZJ&p0Xrx&vhS8o?wWi2K1`Ta-0f6OFH?1LZ`~^k#9YlL}jLBRX^SD5E
zQag?^O~x6%8!uIg!PPf6P#|J{Ko&cEQzbM?c3IyH$eR{KT$)rSr)5)JpOfCACk1}g
zaHW>5M*C$82fYq_5+2u8e~X1Tlp?`3&us#a<iH0UhlCo}@q9JK)0oCkNaAtooN(*j
zL8-xtfxHP?3b${=udgo;|IJ|uR#<0GfB$$0FAx|X*!*1ddBD@pe@@>Yeji88W*F~r
zdnwvl8bj3E+QeK&idSg#I&J?I9as03=hbR}Q-eG$f1<;7PySfke=1ai9wuSoOSo1Y
zQ6UIqays--rFOd97OA;PTi5m`Ny&yp=ebT_(BnD#htHl4#hdc(#LXzUPf|QT1Q=((
zqaIUW1V0%Dh&(dE>~3jaTTLloh`%WX#9otMRkgTWx}JNc+%fB<&IRD#gq;}2T$3h+
zKHpWXX{UsfW=E^4e@&-H+b8e3VnlWTJBmv8m8WbQ;c4<iC?x`8&tB|U5O#^%p1M~?
za~O%7#~mj!L?;MM9_=h?J$WGsWZ>;vr)<^IZ>YtTY<3`HDfaZK8Vx#TVi)qfFqm4b
zE(~g6MmcBV*Hkd4K(MVHpzhDK0jBDB-LI6ylLLjlJyaoof2mG_lFjN4=Cr<cZN&+T
z6L0Xo(k=&fDK#5jc{yg;ChP)_c5KjmpFml*tERE{I}#{VB(!pO5U3arES`@}Kc50=
z@3>HemQf972&Eo4<Wu|Eqs(W$u#(=A!SymHdT?DxHGVy2Wm+>Uvx9IAEz|Z1C~i_*
zmL&079;Ugpe`6T>AReZX*Mx4{z^j^i9Wl{>v1o0=C}xmx8?%&ZQVzQu#mrMikK5ec
z-a+b{3p4!E=j{lEX={8qWXK~Of>52I-w@u+!ohl0j~bA4(2H&4XzZo$JIK>vzm)vF
zbCVIEVeEB7m}kxO)_}F0qxQfv^(llj*`^8hVv%_Ne=;1Js;-w_SWLBlW5aiZU&Ry{
z(fvaGsx03<zG(s5Zf7O<E3TcwXdn7i$X$b6A5@TrTFRJ@0z!pWc)h+0=tiAwox0s!
zVS?i@Q{&`21txgps$dfL7`K11s8nGa)6s38SsW58V02M14tnP)Yxu@Oqnbv!bcNjr
z&>Kfnf06`*0PM%Q_nQgLe38{UN_1}aRh!DY@OrmC)9rQn>dvH_io0G8w=PHF%Vj1U
zAk@rbAAjGG>XQOjWHygZz=vJw?V}U0Ye~CV%Tl&aICrdRGmTPa@#EIPW@2)k#$UO!
z{H|#VtP|4zLY4yJq@c~4W~t6xDN6;E#e*Fqf6uwV4g!5pdV=sdjMN{Lo**RR<ve>e
zCE?qpCC>K-V4QU?FQr5_ba?l)<h5|kM-wYSkLZUDg8XFTlNoes1!6d8$B&OvyLP`U
zk66G0>p@vL$xfpjLu@+(-uS^s{A4Arte(bKxz`}CX?p`e<<|909l8A&DCzYEaLn4w
zAaiegH0D4!3**V=*16U@;-x{M)jg+D<DKE-f8yKJpqIfIITZmlmr)4<6PHmfIX8dB
z1rP*`WXnKqG3FNJ6vRIH5dZ%nMY6~y*^64%cmm|0(WskCZG82ry5-^9K^{IoonN0m
zTn!wY@Ti<Vyna0>x*_SWl62AH!6_xIQHR&>hd+)|YD=FpeLzZM=;P>rPv3ZYHp<fo
zW%;8{&koda8mlj5jKMrhhccVL74LuGX{7T`#`Vtp!PB3wzdcD7{lnF(ExD}&LDChu
zX?qDI2a>!%s~JuIoBUf}=v(nUk^)H@2zfl%d`-psFMKbKcc-XYyahT>MV<doQ)viX
z<NYM`nvpeXq1-6!%hg6hxoEm@LPOu^vxrfgB<3aW7q5|_lVL=#(Khkx^YDM`)9cee
zPt`)&2(;l~jTO$h!(ZQ@;tfzJtP2ONMNn|~@%Q2D)9>S|Z2a*%#ZmHyv=*`RhqeOI
zA1oLl5sE*0TkbI=IgtEDQ}X>X=gP_hZCC&N_;whseED4U*AyWp=usl|F@Cxjt24+n
zVH|H}l90QuEb%9n_<DM_G6jFipi6QlWPLbD?$RNU?<l7B>tG}b4Q8eV2UPV%l#6^L
zFH1pH&)y1Vl*p1+cpvH>FDJm`=<k9B<wWKyG3Zl}r*n8xHqAR}eVpt@XOBlNL_@PG
z>g7HBQJJ%@2ej;((1DNZGPY<Z1)7<DJAAulF(bV&XvY=#NRM_!??r#0(XH=1<CtMU
zz~(%$okne|eJ*)klG&C!#g_GsDl4UV33S%%debA#juo~@1>-5i$PV7pw+&L~=wH0^
zWdr+*otzSsbt=|amb+lPozC2FN=o*iVd8Cb2eLgNSu;TbLtJv1=5V<;e^?Q1(u(^|
z*#ngk0PmDMPzB*s4g7!Jn0qa`lt8j0ej<lNRs?nwY%7OoDHLfp%=k~U2yHHc4|ix`
zWDpQ$4*4Zn1U!ACu6y7K;Mkp_k_Cng(&60}o)h&FR?l{N0M8Rs06kl=L5ZA|++OtD
z&PON+BFEiUS6PuckWbCb3X#8qj_?sN4#K7+Mrq**s~?lZ7|MTdZqaxZ(~=nrc1d=S
zJUz$bQ9wDS8F`(V`O>C2ETDcFlUbx2s-(J?jvmy+2TV^~Lr-%0Jq5L%z`;_$sqnu_
z;2*98&|AkCoQUbW=v@5gZXTw(f+nzOD65{bS(F=87&Ia5FEmmQ>A4(2TSg^a!`dpn
z<@1;n2^=HM)hmCm1&!@om1O_`-n-o`@4t`^VkAQ*&_0uyLNq2Vk}Wtwd2Pt!>)6Y>
zr7#q&82F+N*Z`pIMy3&)Z~#HkLAImkVoFHsxwT?0X}frGloO5;PiaG3%X!YL8P4|R
zh9-rs0ka^mwTzETMWN)S=B-u4G{jZ(C{qH=%)8D7J$QfVF!>?jk~A$Xm3;nWA8$)%
zd^X>+@6V@a!toNFw)~2AGA92T*6!AVl9;4yG1irJ3R<>tJpJzWR%MmQt`a_;S%X4r
zY;B>?JU=04D)uUR;mO8yujFz(P&w<VoE;J@RJ<<ddKKqx#(YVEQ3_M%w`ZSOr|wJC
zh0HVU+RT4V-sL>v+3GFoPZrBjzMi>~+;xxQ6aEFWU&u2q4QYc3C7L|J8Lsh3(CgXF
zct`BN!ZcMmHpNbiD{so~Y_VkXsFR1)oHh70tx@w=n%7x4Us;Z&W_F5-h1p5M3p0rh
zBl;5qIl39|(~6(!ib;p6m2V2ca~-o8d3A1!RP_ubRXK~V@`^n{ZDp~SiZVGQf5ncn
z{KDRYK~K5CJkWbElwE8HZ`yt6h~}xXTP$pd>*Y(+C&0q3d*+muzV_prX82r!8;oBI
zG<MMWomv>N{Edar`}8o3v{W)<7Z@wt|2}_k4`X!6dPKZcAKk?sfEyLt>VbGKw6^1n
zT|JOTIti(7i)yKpFi*lnN>sAwe>TNcNaW9&aczxFTG+03mGi<F99MwLLi&Lw>Ny>1
zrTIBZB_=&d+3Xykrdapm-uHG69Hk29W~*$5cC{}ueJ!1qMg96C(2%D!Si0uKnD-dU
z6)7}&uVW<B#~U;`!e>`Ea6Y5N%<r}A<96hYB;P;DT(&O6hk|x-CobvVe@v%9<3hNl
zQ&&*io;)_GV2^Hyb}9Z=h-Hda<tJu2(KhYOtURe#&6Ia*9NKOk(?RWEp;45_B#OV9
zed?$>2b!l(e9txKvuzy7CcY<`)=zX<4_dv)W=)o$Z3yrzY<F2}S2Ws8?TSs>h_S>U
zU34dqfJdSMdR*lv5C38ae*sZAJJB-DRY7*dhRK_nYO262cM6%<W%ydD#%x;yv=yH3
z4|6jfnc^kjn2&dDjoWs@+WK!gqV1ECUr2m4@h0~O@6}Qh!?173ds|~tV3enBU0AOb
z4n}yT2ch1Hm@11xfqf@ps+<y_r*?a1v;M7&$Q&!0*$M6MV2z46e}h$O+;xLBDsKgO
zVwiB>gEK4)+TKrdfvoTE5m+M#6@E(K%TK=vvR}ONCde>jss65K&^*AtmBGkH+Be9W
zGkMRmumW&lct|6+xvmUCQi>Yn7(!wl2$1rd9%tW?A<5Q)DxcX3^K-N+gVhvJZd_r7
zEh*)M(YVBx8e2ote<rIvp6!aqy7ZFTu>Ku-D60Pe)NF<B!CXo31?#6I*tS*|5@#g=
zHG<4Q8$tumzI~{;SaWp`)rR+pc04e-ncz2Bbhk!kwc9L=&aw{=_y(L7KEP}RXq(%k
zy^0;8&DF6qb(_5HJK8ZuY6!vvv#%`Yr%+Z;J?*zDe-srVQEn#A9z48(fK<W?0@!}t
zglW;=AK5<K+|R029E5V!9EUbN$4!d0#?IM1W=DfLk7^^6uk5SuiW4@@46irG=~2t=
z_cOqH5ny!8PT&Tj4bGSOH90RP21jx4$UNwUHKV>u4*_<*4Pb*niOVkoBz4<!RlbQN
z{|oh^dOw%J7&#RIH<y9xJrkFZN;xQ(pEfy=0s6NvH#yb^1aYhxESIsmI2D&s5I826
z4)r(-0yj68k#i>~e_L-`#}R(luUH^2>4Wy{-1brg0dlF+6f$bqF3{F63SFo~sUsQ^
z6}bKN{bmn!M29pYi;U<xh%}_;oSmKPH?vF0N)9OyewRR^ovMLz?o<m}#i^P?ChHW1
zqNP(DS|0xjZo)?=@dZ<Vx)I)JeBdNrta)T;AWL7Ufj8Iwf35}@3j1R@fvsxmBvx$J
zI)epHOr6YuId@@N32fBmZ!Wb;U#$f$Wqer*+*}Zl8ic{ygawppC|D9ymmnt!BI2*g
zgukmvCvy<%o6O0{W}PfS3fBWCYml1z0?>~%-zi8AnxsF5O(|IyhM~?;vpInjtDQFm
z)lo#E04<Ggf2cu=b;$_4t+*aDr(iVlLKo&cagu{A<;53fwF}8L7<L$15(8!9Q%;UD
zD&Ez`4MP`DPbaZD_TLv13VGCX77kfj7Z<zAt|2GEJNiP#61mp?2FanMHZqh@g!;l#
zLxH?~H4eqaX{k8Xk}swbO2xai`17@x%Z=|E6afm1e{IyVv`|y(JAnQwabF)Msw|yU
z1KGO&Xb#nQ++qo}DD|cASS_Ot8eX3IV+PM)?pw_6B)LC^HJd<RwPCTQqe<b&P2pdS
z@MtsY51Esrh(!<>Redc)Ra@%Ycicu7H`hbzj-9*RsCCVlGYWT=oB!0&MQn_x&eILD
z43+Lne|7YdjGy>Jcgl>Xh*N)m#!nLI?<v?5g0JtaKvc@3FG1Tx&~oQjr?s!;F+>gS
zgqXqC2*3(6y9Aud_#}I{lN@~U#qjb!Z)d|-mzVR!@Z#oV;h&B!F8>{dujf~1v#VDL
zi6*@s{ycm?eDjLN56@?(i&tbdCWNvP#R5;_e_UfFbytry>FeREfY-kmzMH?C58sB*
z$r#4wRuBXwDK~BT4`1Tnc2cND_=B3o;>pBPgr{h>h#+5=0l}IO*q#9hd5CK(8)VqL
zEDR9nc3I|V@EG`3WP}_x7aR1f(x41}eem8khOegApDr%umyqah&%gKoKmT#DczgZF
ze_{CHV)5h6Nj#nZJp6R?^V`F-*)PM><He7EeY^PnaxuF)KR%tspWX~7Z{{b%FBCaN
z8m_NShp(?L&VHC(4=4RMSUx){>F}sB_I0qWa+f9+DB1HVl5Q->Tc{5a*=0?=$l-+a
zM{q|3j>6zOGcea0Ju5E_%&kV1{QDaif0;ifkLRU<x#0V}U;9=&JnGvzc=*0xooICD
zxc)oQo*fs$9UhU;-PXQ+<={iHo9Dyd|M5>0|4>jvFK^zwdEGg}@Xh?q{OZM9EEB}}
z!tm^PfrT#vk01Z^Zt>lV#qnYmX#C>I{BnVJKk2|}`SK?~NW?1R2kWW0jUOE4e;5!0
zK79G*@a$@S`eL?tH9Y(F$?$UaZt;4>2+xjxm_Z=y`*Lx;%>Mq4=d<hi&DH7bdWo9+
zyQj0Wi{r27@4BrVdz6NVdUkw;XPr%QXX|zypVLu<gI6OU!-v-oH9vB-V98tS7e(rp
zM>IYxbaioZvzT3X!wc!*k<|Jtf5SJL43DZ6D=Oz>SbqzH=Ok+ecaZuJgFBGEGkEJC
zmPq=RDic*Fsx2u;(ObC)Y4qk{=^{X}5lMYFKJu5Fr_Uz8I6tqsot`7ZpAemsoT;3v
z%*w2soRu@>^K`E5yp4`Cf4jK&`QpE`tAAarIM`=LfWU|NQMb#ACIw~Xe=ggbN(7Sp
zWH+5NHeJ#VN<Dh4i1rCh4KhRXRt@FBaTO4N*rqxQh|cIbEA(aX#L~8}_3Y|K$EJcJ
zZ{8ZFh&WV5+DP(HAwB}lxyz>&H0M5_cz<~jE#~2oTHem9*<7u6+j6y{(cN>c&xBfE
zepTyzj_|=+zs7;iDSBgrf7BThviCh$l{gl-*r!}WNY`r!@xy8e<+~c<!;stBP$sHw
zs1hHG(Ov&&M+I?<PBWm*i)g9^aVyl;<5r6Gkwa_74PCz(ZJ#`}l_-TKAv~8;{jd}Y
zb+HKSSBhz3%{%E<uqgTyp&FNwhu!+PDoWna6Q_cIR*}H8CPErZe|POlcVUwMI|#&O
z-F?{}=Ds5azjfaU+EAJ(ny5Na6CY1}Jn=D4lx*+%XhI_#D4^qwrW`E6+pt8o9I31~
zs9Gf@%E;*nowTIZcHwnWBcVUff<7Ha<w~)+Od6SK-Rn3dshkS312=NdE87bpgdIbr
zP_PdNf|Q0%8Nd<Ae}J<+P!k`4-Kxdpv$b37)(GpDJr?hbZq{^5NUFd@NDz0J#2gc~
zJs0m}DBdJH0G?7Zz=-*jA4{T^1BfCcB8^i_R@U66R%*F6h1awyts3tpekw8qoG|(v
zw>VxBB=H81S`QzE*RubYnkXascMv>q-ogRzjLisHgb>gOe`FZFZa}`0kZa%txfK9A
zbcC9BAp-;bV*3c#O1FdEK#(ct_4+BOm_CZm2y`x7i0J*&#re0O+zQ78<tkEGlHa4i
zy*Jd^S+!GIuXfTNR_%0U!sBN;%Q_`f?UJc>iA>ZTpH0*R^kf2hvJ0ShI~!0$2HZ7M
z^kb21rRSvee@Lk))lhnYYy?;3sypygtfu|&E1|rhM}?oNA7zne#tmJf0tQN@DpsYU
zzihl+p(z}7wa}Cr+dc~ob2>cYqPxh(xkNvMgyH&cse%N=L0gVo<dNn8s>p^#Cn=C<
zV$<?LDJuF2QKSwuB9#q7$#!!9?!al~R44hl$NnsAf9>!nw=JgtP@d<*<(?!V%PT$N
zo;EN6N%$0C%lQUyc5L--nvF=0K$m;pnmI_yyxr!q#C|j5b`0BMp|ba_W#8KB;IY?p
zO~28sZeUxM{em2B<vy&Q?(yp$oYK&Y%e~g;W>N0PO2yDz@nfDreKaVswhl_HJPed%
zSpg+Fe?bhLAcjs5L#Gy8rxx6{JB2Vfe52K3i+14?2ms25eq6a-Y>Odrt(>t4f*sAp
z-Zv+}6QpGac)};@ew~JaD@y7Yg`)Vauu`I#ZK&kL0Bi7?MpFES)h^&fZRja@62br=
z{r1d<gb)^k83+IqIf=E>c7PF3BKp{An!j_6e~9zFo4E+gyqk+6UY7a=s#yG<2>R%L
zrw~N|s$5fWYb_NsWD{k9u}S3p*3s&3wT>$sw~09;joyRJa*f7(A678~>wP!@!K2?p
z`;?19*>J2-9(b%!%1TkNsXw);KeeepwW&X~sd(5_Jo5BizWZ(}D0wO<c`7Jnn&}qW
zf4-v?<N)k0JO1#a#!`Ez=9(hBbVEnWn#?$-R?$pWZ7J<w{^ZxypfMjBqu?PiAz>(_
z!1r3$?^$so2FRtPnpjXC+YW|c3|yly%&ibe)BVQPhlI$<&u-wEY%HllBfB`nK@acq
z`&%JQ5m(;pe?SOLg%iTm%<spbS|r=Se}KV{vl_6rqWNup$<HU+ZXWL|^pVc9rUt-M
z3sz?>QH$U1ZL#tuDN^&1WYwZvE7An*LKi8;TKZV7U!Yb$Tno^32Pz+v*dGLtDsr{1
zG_CMS)a1QaPQQW2ybFNj_X$1CNy4JyzkyLudC=q}t&cK$=HxNw7|kM*N&xzaYxsn8
zu<69|_YlCFuaYML1OkvBBX1UA*G|TahR&~J-@-G5yIS41lZpDBIJ50q3?Q;fcc({`
zq#uUwaaQT0%3sY5iri~%@E+R|p;bThulHePlE_8x!%D59FT0ljUL@^P7zX?gDbBYb
zm%$i06#+SyA%X}Kmyk+1xdF_#N=rE=1Amfc*RBEt#Z5EHa$r|M5nV_a-%7IX@<U(}
z7H@7QAeo9<*YrTg)oR$sU*ob65=Au=_`GO*wX&-s8bRfjUN|XL4~=t7F?o~Ele?g9
z0go;@)a)L1DPm<BY12pxc+Qpl+^3Di`#J|JLG6|jH#%4$mD5JQ!@&jIiZk9c^?wg@
zaH&RugByb$IF#_pj5MQQAJ`2_G9pNIM>|M6g;@dryLK>U%|(5>VR-A9^Fg>%=A+s}
zG?IQlYjlN{GmGYt?P0WRL^(ca0)|qO6@KId?D}>_NUGow7cw7YBn6Ci*!WhOM{Uo{
z9^LGfSyQ=_3)98iZ>VoM6UtiQ`+xq)gGynHK8w3Mkw;SZLkJqJ@W82shh+-1N8lrd
zj8l;?2y+aqrbVPJwg+q4g+>;HG*x!MH8vfW-tQhWL2jv2HGTcQjh>ABz~xd~hm{$x
z?)Mqhnfp~4`CBdI=}ou4AE?f<T8<R9`r$NwP@Q&ms5dX&Zr-og(NC)u?0<a%^I-Ib
z4avw2?Q5D-WVe+DDkrt`e$8eLGnAVLu6aCmYbp_ba^~!qN|1=FgH&=w>JV0k-XweG
zeDg#QsJr3XjFbdluPG?5Dx=2L)g`y{SDGQZkz&r=BY#bml*4g#?qtFx>a1i75<VBM
zophx9xUjNOOdvi&TXYNsjDHljw9VG;S2t%7+nUfm(;V)TR!}AYDon1EHqVwvYds`y
zyxyydqbN@s{yJ5N5w-emp~}sXrk!WEPEqG=uxZIeO-L)@`X<0h3UN3R-P+xGFZQeN
z!Mh)6Q7|1SZsFaa?~fKHIDHo0eZva&N@)ceMiMWq^^o#CmhSUdzJDK-@E_5D3Sk+*
z9sQ?|3&#G4<4Dxa%tIgdm;`@slwRPE>M_BPPsC>(L4wg#&4^n-y0qQ{nn;j@JnT;9
zsN~*gB87!3o5)C`Z6f2l-62He(?jQZ9Xn<95<!`AY6{c3jt^Yf6ej%+f&mYf*g6Gl
ztY`DtT%oKw>(mp7hkvw*v|If6cJOGsor4y6V)>;XGM7-dzDE!HLexwXo(Jx!N<WHe
zKrv0*uAuXwCww#{qD=YgsQCZ=#!pi*3OG_dXwSd^rAD<Xi~oM>{4QApAk)&V$dVh-
z_Lyj+aTNX>%8oShw_`y%9W5qQ`n@Cz`g7p<U6LbrOf8HEP=AB(qH{j{#^{PsGx{(L
z@X5;h`LU&2OKq<iSx^dOMvVuAUDA{r114GG$eg2AV(IuQOGiZxwTC^+`)i!5CMixZ
z3;p!p)?CmHNVpt_D_m{F@ATw&{2;hwR_Z~F4@QQth5{Q5Ox2KwgY<eu??**<L}0Xs
zMi)HB??=I1f`4S3D(as3l{3F*39bznG6e*#N<c{Qkel|yzle$~Xu=8LS)>DJ-O_1B
zw2!iQ2E9KF(BhLVE;yCGR5|v{o72H(w5mf|iYIo(8k_q`Q?AzX9U9(P$Ri~%yD(+`
zJSEDUYNOrMpeUDu9>{!jW<X@ln1N})tF52)l)tVydOL{UpBx=MM0%(C36Q+3esG{I
zHA!_-mT9laXHByCs>tTaux%Z%_l0wsgx7C9y3cUN&WAdy{{R;#nwLSsGZO<bFgKSW
zb|Dj&OH(<%0>f{Y3sgBze<53;EO~DStxygR1jBc+>v;6bvmoTnDTFM8Vuapuh*Cr;
zd@B0#Os9~?qTi#=QNQMX(z)^Tx%dENg{Bvt1ni?DM%PnxRT+(99<vJBEv|R2IFa^q
zhh-4cy!iC4fh81Z4+R20&YfI+&V8bcA{koXKHy@5$pM4DNILB*e*#vHjwG4v=%YYp
ztF#0`)n~N&Y#3Tw#!?vPpsZl}afI_ad9;M9s|iWfB6rbe)(YrC2EC5!`^*MXLP5Qx
zB;};_jh#NrA9aXf{<vD=0a+!}mv~`?#6z$u;8C2EXp)%r8oXEm#Gd8MLECKO$vYS9
z93v7@iL^1td5WTxf7uh7N90_J2wPIXFnUVQsB)J*N06^=!7JUX5^A0oY>jE3!Tr`0
zCQi<LI?cqlhM+UzVC<BCJa7#)a9cxIvkj_l1&6@V8S(VoUH$U1`R6F@$p?jU{`1==
z-atc&U671Z!Qk-yui@+Fw|>>)?f#sqdV*9l6Sa`CBJu-Ce|kBjdFW2gT<poy>YeQ{
zs^x6N^09DZ;r!&{QAj-N%2u=NulzNy4SD2LZOC_WZ8&~%wZaSnr8b>mZy?F5wOD(l
zo$CZv;L=emFcn|ACR#rqBqlCFgt2mk(Uh=Fao_qk)$~gVHLfN>$Yxp%wlG$EE`}Ff
zQ<a*yJ)D$Tf1?813Td8Ba}tp94hPbp+{kJ>P0j2z;=mv&^sbYX*e|kNY<AIR&};LX
zf4>TJ;{|l7#og*+b@kR!PQ>0TMSt?)IJ%#EF-40Vuk@^f_Qc>XWZ%_bZ*uu{6|qMU
z#RKf<4hXiAGJ<havs(qhZuNISFlNT?zlflN)Jfd|f5Ar5xbaLVa?VFf<$NglDR#jA
zO$b|Q9X$G8`Xz+ZcG8x;l5bFU<xazTsZ`k1xz~BSf_RJ4SXb9a_Z@mpQBJw(wZnp*
zQ}!vcSFx-1oDFlQp7Ts#oDB#7W!ZJg&3a19^EU@YC7je-+FpV*d?ZNRr@fT!M|nfH
zRy^sNe}?H#il1iW%M4{k$t>4PGEe90zT}`p;gN11+4=%IMK+x2efOuspv3;Yx>*yP
zbZgw}x|ZLrF{l?U9rV)MeZj1cz04{gtYUPyF)q)d$gGTke$}6`1A7OC2e68#nz(@B
z!AM5?usXw!o!wg)zE$oH3~!xN80bz6Z@ug~f1JIwBVNGpoupSVe6QO_GQ7ng{U`qr
z4BzQ`XNK>jFJyQttRg{sV_aTi_-@soafkK}3~vo8p58|-V0a^4E1+AG;oF^GV)%ud
zziaP}TjlRi0Srk7Nvhwc0vIOcX1(i5JL+31MDWUc=ry@eA*{p|1lvFHp%uc!tsue=
zT2Uc7UGH2WI_V25gb`NK7FLeiv$|_g0eYeTzW}-lQ((m!-cUQF9R`i7ZVqO_hP18U
z8RGQxaVNwtBlSB({XdBLcZ{oFNNVY5bbsv^$Bx3KMEnbI`$6!RQK&W)0yi?30sS@;
zm(^G~yMO%ipKrgOWENctZ^GAb>wkTDJN<L2iU102{9v>dN~+<nAE)qwlorPN0YG>~
z!_RNS_tS6vs%5Y49$|BeVUxya-M|J$Xzd1LP@uI1wxF111zAr?{6XP6-x2sbgMfSy
zrDt?W-{ZK~JeEO`41&BQ5EjK;HDy5ntN?}8lz$Y=VV{yzI!I#EZuD9V6=kydDzfRU
zOB`KC`m~0wju79Ur(^UY3iWNoqW;=Nr0#qPD=+W5rd@DLcF=?+i-eBBL~WemyYUi+
zp8?**H@|-^)?Y_jdI7Eg9gG*D$wns=#hZF_A9dO};kBvmBS%vN<$~L2OuylhfwiMs
zQGW(6?G#kyajzMwAdP(v2psjkvPDGjGA_$D2uiH^9XKz%hfbN7S5qv5EQY3l&?V`U
zWL<Lc{rJ|J6BWOHOnht3Cf<&*)PVaDWTRnj7V;XDaMDdfp2)8R1q*gC<kK3{kWX_b
z!IvCkIZOc9A&l##i(t9rUU7Fw66h~i0)H7KWhso@bWcPj>|BbHhO&;7_o(=6C#H(x
zddL(DU_=h4_Kx=$%XR;p+-CUh+A;V}m>puj<Nz7oIvBMK_{@BUlb6%gwQIr+>%0^h
zN(jd{l)NKhHBH1CHj05C(T%fRR|t~V17{Clw<o?FvRE-O&z^21Wk4+yr}0j&kAIis
zCWEhfnc_s@^$gWowCplEDXx@xn)tM?sJA8WCC@H?cAl>ra%?l+O}($KOL@2AP4=#3
zlDraJAKqUk$tt0}eNdT%BA^-*eO!jdc#p~7wiPXmMa*MeT(S%H%{0YeNp|*D*Lzh_
zLK29Ns-ldXFlx7oUJrazMMvy=;D2vWVM&r8YAP%`w??YnVN&h-3Y%A*W{R|0>#%ud
zCDV0QTpc?Cy)w4j!MFKH%gySJY+PrsM<yU0&U=#gM^jN>b$iX(-Ouoq6~&tM+Dx`7
znL`0w>3f^`ZAm=4grj~|!hE!8gI1}R*;(UQ$T>S<-sFhJGa_r))Po!|mw#R!+$Q71
zWSp!yszr?xyCU%tb3hMlv#!Q<9Ec`}L`E^IRg`yK;SC3(=a$R?CH8vlF^6O6s@8!(
zZ>NEgY3nYr;PVd99P|RA8H@Tp<%+~^LPo2Z8rspAXrv#X+aP4pml{F$BMxYZ@9G20
z0W}H0n=*&iEyfC@uP<b~2Y+jZj)7~HxECcrW%#ITA4CZP-2r{$Sc`~8-<-wT4JZIO
z*2nq(M_-MO12&t#a#WcYhSdVB9rHakLu{u2hh8eXdp9ebXN$yYhmu$$l!1PQost>o
z0y^E#DhGF1Ik+z#(dMYleB-Ve6O0#XY$(1XV*)uYnD8GvYrVHM>3^9Gk?oYaR|W;e
zM>iJWF&PvTD?DMhA3Riz2&a<PfPANns+iy!rrguErG&M1m&BuO%QSM@wv0~>w=Jzu
zTy8t*w%V5Oa^u`0I4^^{*VQy2jOk^pP0fl`ZEB_`$0%zVz17@Ih&KCOQ6bnkLi@P~
zY*z^KOw?hf^%WIDOMl1tac0`>ETWkiN$MxnmhF;n%{-m+vW2m>EHMVN)y~EECuyzG
zaqMc#HoaYH*%~=**_yuDoS!V6AB?<1%a-j4Tyr{fZbt@7XvIjo^58}6Dpr|ZIC-f6
zm(srmWA3HczPGFWeD+drw_7*&cE|Daj>ydIv=!ElF8kQNuS-JD73w3$H>JGb)gSf$
z=w!J#q~w*)<C3snf69R$$KJ}3#&4_Zy*P>kj?hOP@|@%I>;9joSHnIslm_`8_5%#1
zzcXh40;f+QRF_exHWZhzgD@2WHZhmMi32E?>Xrctmo8s9h=0rV^8EMv9}l{yUiAc2
zY{Ot3Dbc3tDSNVL0375KgIuBki?(!-yDs-Pxy!!H<4mGFD_xvtk=)sr)uNeb7-xne
zx2{dkU(=kZ>EqFxQ^*1ar9HPAW5~LYExSY3fWPL`BrWVrGn(dNcfRKc{PKtel-0v&
zZFN!pv|Eri%6}8J%131>%t54MymlEN!@5C7wjJbYRULh)QX%1x4br)c1u+EFGuX1q
zCFf1fres_6gI37aTe;hyVfZaBc>2lrZ)eS3cO_DW!hF|+PO%Z$s&D7&B_yp#@?fju
z=n1um#oQSnVr*u}adzC%dE%IfL3b?nNDh^p2O&?5;(wEQi->4g*kmMGTOnO_D%1jr
zqvl{QM;r5ni@JzPOcAX^K3mqi*eKWtRJ?s_K7D<E_^&fNdn<Py{`vI~ACO^XT^ImV
zFf{!BG5mb^vz1yb(cV-3Oy<_sGB3?7yRaWPfCNl+vL(>hpqwMT16t&W59q@wY>L)r
zQ~n4LTYo{sPpNb?19wV}34Kf2EHh_mGmmUdo1&cyuGQ|UwAr6$<f|*wBwDW|Tw7(w
zG@0R?r^y1v6#%ESxCS7xQx;m4mV(!FzZ2pD>JDYI8Y!Q=R$p0UW{K9fGxQx)%%6;)
z(AKPIXo~OZwP*4wu_EtU31g5E6nl?Z@lb&00)M{G5z_)(BIq0z0A8W+V6$sv9C(^y
zF=zGXjEK2tB4DqAuu3bn=#)_Q>v}mEqz+GgoI73AgxtD4j8z2H#)dn4boeh<%%Xa}
z^QIh;_?t4%mywA`W?6&bL=HJ0O1ENkk`!;?3PLAmB(>d*(985{gC-ibk$V=ok>8C3
z3xBugycNF_l47)X;&(w$d2^ZHcY~h!eYL3i1g<A3y(}%c9>X~*!mTY*nK|cr(tEgG
zv=S3G+Fix<Ck>E6-r@L{c;BOf&?#PdAKUh*fW7J++kwAa19+{R)Fq96JBk3-pulyG
z(-i@DqCGp+grOu)8*onim3q&lU5RfaWq(Ms^45ee^cF;+X4h*8pPXXHozxwKFO_C+
z9gZ`aFFDraGoB^Po6_@z?8o+|EjKz+&R#*#$fOAGMbI!RXilWFnfMR&hwJGJWKf6Z
zytg815h-o|y@(oRSK^CAy&wDp^tQJkg{nd>$gLi0DEW6im4kN*{F2JEtxi#SdVjJh
z=JU)sm1n72shkx6wAx)m<x3M~_&*dP+bBI`ziAFoDZK(&DLs#D@}upBxJ>OKC=-0E
z=N@VYDP?^pp`rFuBo2-hB6hvsw%#Aur1uIR@BJGIck2C{T|cS!Z>8=i1*Et`f)0&s
z9X}u|?QuuP->kS($G=`q2?Cm1&3_=Q<0j{k{NEv<r${yWg6-m9hX~pWrCXfcE~1SD
zG^?2&kHT+j%wVMImnw5q%ak45MVA^SnB`k+oS9l^dtU<6N?hH8&>P9+kug_y-RN3n
zCw-Qjqt$COd55BznOG$~PivdhbM8f(QRq5k+F<*Hvw{zb%!kj~lzx}a6@N=Vv3k2(
z3dh%}a8Bvdjn?PtQPm%K;T?ww>$oyW7^@MtoE1OWCY_Y$%aVB<^`*b?<LXgXr01ev
zl~wKAdAN_{SHIA#^c5Bfjf>7@ud}J@rD*CZWqyh$Xi4nBwYu^?3xm|z;Cw6PhW(eZ
zQqnVB(SR9(`L7T@qX9!dLw_#62!hci!^~xit8PdcYrM<?-0~2&>Z**ETFl+4sWS1`
zAvFDlVS(vYfJS@7+AKY$@{57uB#QJf{YkF582LUqP^{IgrB+e%)>Wm^on^V!DJZkl
z#Yl4WpijX87&}?);IPG{x{x7E-$N`b+g_rLuX92xF}IEIw0xbCQbYe?KQ@x$$=?_n
z@n#i#caSMcH^G4hm6Il{g=7DrQ6d4X=}vt-+ooWw&!BUm{a-GxwWGC*tv<dbLL~wG
z4@jV>XO~f^HWZhT4*?aImSs7=f63SR?;n48;feK$PmoM$chUwKag+I!L<NbWGJa0;
z=KrHRL<g_*zs{T>#wX!Y@hQ@$vT20PZ-zHu`lim{W%G03w@fvU3D48Kh|?Gw-$~E4
z^wM$q9>xfH0lF^|&53VxX{3LWXYjLyw<kF~&)1sQAgi6E7fDv$$ON8#f6(hMH!J4s
zKsv#s!(P7~y+~o+l28CM&JGl^)&j4qJfIZJDd;JL`BkAp@~`rtEaO(rQ#nUne@nXN
zX_Z8x9PQatAM8$P25K@pC<|37f7@8fbGpZGtLhN6Bw(SN`oM|+t%$a!Cy+S9o4<<r
z(i0i4z#nGg4rX@^Mi?BEe@I#4iZi6kN+nOfldFz*fBsmM476bmZO0hgT9rwNUg3Q_
ztX5(#c!Vly^{VuUqVl5L&SK)ZmL{#6%&SqsLMEvj@L4o5G75*X<A2F=q-Ch(WIETg
zkdT~GO|}y(36m*iVax%`%C63%vi#~!5F*oR+f{o5sR;taz;FX+f3!?l?XJ_Yi9o*s
zVr#}Kfpp1DKYN65GXlvh1cp0$#zB@r54axYJO+AtP+5sA)A9yI5cZ1^3L^TfBcmpR
zvv!#0ATnCRG{~Nc$P1(FEDU8k2F|o85a`Yd6A!`IY+3}&QCMZuTUX&kp98ZbW2Rp%
zETHl<-8dVdu3~olf9ADcjjA$)^gLw-=^IXOnu0UQUGh7+pi3|~tV}<Dew(<d;yXe7
zD`h&F`nUMBx(m^H3kP22#@#1L)-!}!rlJp(ZLHTFmb|BoDf-qs=4)lE*DB>Ws(_T>
z=U3s1z{M$|Gt2=e@ywho#%?Oyz7~-twp>xU<l?R1sZX-Nf0>@qcqf@hy~MbK;WT_7
zQ`2O%*Xt_S>QRiaQWPstMi+fW;THu|jtvn7!Px`42`1u9Nrw_U{_bSDa$M?cI#IXK
zL0qDE4M}De^;#lax+Bx9y5^%=ky#rsmN{cfG+0?JTdh{GUvr+6T{4-S3zlNeH9C@b
zt7CPJvC+8ee-a}snr;e+PU`unwcs$LL_^iFbKw_q6Qu>CG;_+p+*zs$m6Eu&LORz*
zof=1y8g`&mQo{bq04QQft$TiF;@_(duJ1t>xc}yjRp5G_Y!%hMrl-ztLrr;ylqq#y
zPFH54PMgYgW_C$(yPis02?h=;+ws)I_xajdDrOCyf0Q%~s0NJkU%2X$;e|2<6{F0e
z8y)87K9>2K{(HQ9Oy}47`+8>JsG_8#q@zqy%%sJW9uroVRhVDFxoo1TmhW8qw&Sp#
zev(y#801rzT}8;WY9N?$`rcJ*Mq7|iS~)}1RY(0PEBDDzd8B26z@*f@kkhDzItLw7
z-l*GAe`9sCoiNlYqKLCvKnTu^k;AM9O^ws1GxwTWGdh@>!KauTQ0V{yx*47Y6vuLo
z&o?{en1MmqEwj^_IaK|q&o(=-Oz2_K%X*nAeIaOu)tpaZ&;p}}8!60OSL6>$Wl$5?
z+n(>0R7UJhV=S`>J>~zqG{!f81y^^XR6BYRf1QLV*CC`LH<YfQ<?B+K6)is{rBM_U
z-5}q=ID-^fk>8urw4<Ap#`jM1O8M<7tkQ_w56<#S*B0jlb)gLOL*Ani(39X^(NTAs
zh^R~Wrh~MeI3VMoJMn_)XoV>+icS4!CMdC6Y?QL@X7Q?@F8ZC2{PvEBglL$t&~vsS
ze-DzWDDL`EUZ1Gp;Lee;OglMpK-zv!@XlAcQCtvc;y^(qBdEH3NiPeHi)@6FGgJwL
zcjT;QHhGx32demF^zh^TjGgHP_(Z|0vKC-7y#Rxhf$4tKC#M&Z@B*b6X1xr{z8iT2
zQppMTB2P1V$Z+9u0DXA1JEh8o*5waFe;-St$xg(GK2p!24-DuN?^HW=o;CHo-4XGv
zGlWAp73uE4DeIW!qL%?S(ou&@f$c{HoC<s%z$s#qN=7$wPz*M03J<2m{qB0n_P=i;
znox!*IgXq`G%5QozLesT=6E8whF;-~ddy_vO9&?U&H{<*2Z4j!8}*TN4D~Q-e*mz^
z#1?=}`rN9IIA_j^Vb;s;xP?IR%IZPdYU`t4-D)d#X}6KAl*;KPuC{Ftjbi|VTw-Hh
ztC++=mQDWzrbjc+$A6-o)thyVC8m4dr{g$t&PS<WsI0OgrXxZNdH4*Zl{&PbdVzF)
ztGf-=?!mP_Cy?WP^}IPLknhB8f5UvZ$c`L0ho)?0F4fVZMH%wu6s{kj+1gHy4T&ky
z^2fr|rR=+xI<lO~U{3o$>_2{V-`+o7{ym<&KqdiC|M>02-$$XyPTUFnwj$-7zyI^}
z>&stzswvWQN(N8qkK*x!OLOOEI#uU53CXoqsufa)CZEsQm60`S=Qcz(e?bsUicw|=
z|CNdJou5^~t8|Da{VdlmW~!-d>k^rfZ>t$Oen+s*MZBfUYL^hM)wX!fLtV*byjSS5
z8>@3K!_)M1;F{KUTi=_?CMXP#b}M^P4Oy`r#i|!jC0;&gXDQILoBsqUnmhI4kr_(L
zIJ-H3&-XT*W!B36?OazMf0!4zWP;LyUbfy0k5DvJ^FvQjsQW3Bs}9=d@3Wc#KpH)~
zt&Xn<02PrMpQ#zjM4YgjnxVY9{0S7BJGIqhTldFxn>~l*SGmiPQMs-SQBS_O(eH`0
z`s8QoLgweZ^r^zJ6-ehAuXB+0kZ%K}?xvgr_#`0hXW2f@wuvNEe>`aD`5$^xZ5bE8
zL5nRDMI)J1nksHW1mU`7%OM1^A{_nhDEWf1L!o21mL%?|K7a-Zu{w%;z6%WAGjPyP
z((7?di0JY}X%Z&vlm-Pu7-UyX)LgakMbFB1r2P*7O7BGVz30K!@A54AbbQL7Cv-B0
zk#Nz|GEz{^?$72nf1Z}TYa1QmzY-n6w}#bcqhmMA6Vb7uJli*`5%TWH-bqkeWb3}y
zj#|QFA-#ZYOXq$}ZmFf6zacA(>3a9_9WXa7#J*`$GTObJTdjAbj=l2u0<`<I@HvW}
zfW4DgEhL%AfF9N~NW;iv_RgD4@NrF}V@~P~O&_ADEJGYke^n)?l7|6jGgNV4f}2kc
zRsLr;0__)sDq%Sj){U_DKOU|GSq>O@Vxitq4BZbX?Pa{(4a6Pg#qvBB^WED+ljC^O
z+fiH!RS8G?`EAwiymjr*y0LZyM-<am?By$=&KzH&JDRS_Pj@`8P#tCnSG7Tg@6k6T
z4;y25s@S*uf5Et&qDYkF*76HfEWPQ0+{+i`0Nj{4(`#M8ZyY9Gw95`lX-(jzxwGQJ
z=&LI}7<yQ&TbeO-3dtw9*w=PM3Uks9&K5Q}FLC>F>aZQ}l?Wu~5OQ?NxOzTV-Di;J
z+}6xC8Fpsp$j>guoZ14(sJSS=<`4IZg#TTiY7S;Ee>+ywAxdZH`i|{Ag4I$zKcEOo
z=A?dRKxrvH0OUeDN?M?&xZuq>)#=_YJ2t5UJ)`$hAw3dyQg>n>w>g>Zwe^%=jK<Se
z{)oWwg3mWxL5o3*`6i41>8TQGFe7(YhChXsZf><6R@%4uwO-JW2L^6BFnru~9FXYt
ziJsFEe|&=Dv;;T&rhWO#-DVfB*=ssY{b`)06Jp(GI{W$dkjr#4uX{}U@{+@(=v6xI
z&vhR|snZjkrzZCV*Qv>EI8OcQm%GglKC{y_8v4_?Oe4U$!!#CidB|OwS=XJVes#C2
zG<T+kZ0$g^M<=?^_m7wV00QH)2?}LyWOH<KmsqGa4FWecmr?OI6qlaWASaiJaXB)7
zvFh>FE7E-WG4bh-r`M0CFK1AbVHSmv)5rHo&<~tm1*e-DCnE&YAf}I>)87ZqrKanZ
zF2y+jU4}keddK542um%5PM0`78zm04tvuy6fbl8m^3&;Eq!BE&G}dt6-)p)~$A3Ql
zeB#M;&Yboi{agR-{p0Dsr!o-<47Hn=J90T2f8T8xjMFpm1^M*r$JE{VntbUv`I2jm
z*!WUvhG-{^mH~B~e5u|zN1C-d({-YY{3Jy)%P(guhCQNY0Iu&5U*pqEdAssUFt@ow
z=9ZqcVpf^HC&?Q-N^1iQQ1`7VxEy4N@Xg~}&4lbb6k<HQ;|z>iIFcNvOuEPnbnzZy
ze|4<F<Y;m@62v$`if(a~>A4VyqsJ>h-#iOx95Z6!MS;U<NDIsv(YNN=dk9l~3F9_*
z=HAQ5QidrobzHi5o>ZE2`i}Gin&Y$Pey-qBEA8(_cX+qqdyHyuqj%41j4(p{E>F3q
z)T?)Ia=Y)#-xDJBKMX~1wQrWY%27ZXf650)ymbry2Dbob)}snEQ^qb~x@O#E%RFv`
zs=<21I*AO0+ub@GShvn!tK>yQskz*fyxJZgArbAE0H{D$zo4GF9xt&-Cb70PCfIsN
zkJ%wmLtZ%dTPr~_C+tRRB^X8)FT5oa8s6Fk>|=pxQG=c>`SjV(jGuOz33zM>+&B}`
z0DqK$M`8C+_PO7aH|VMb^5?>gkHZ)(=^3kbgmKp+&I98=yE}5C2->Ub`BF|!NwI7F
z2tX-9g0e~1<0XKSQUoU#<Li_ORPt`3G9TBFWkmIQ_-tz%uM05%H)e@Bw$r2hcBup0
zG36Ez=<-ZyEr1zZOv@*^A`nPcCh$p&V}CbyuD~&`oIb=Vt=7{f2on4lNsoo=-(xm_
z_uGAhk`ZhdK2l>J(Grj(V0s_%%rEkAza_?yZeK=gEUCaR83`jxf;IO`D*_%f6wG=2
zU-!t?QEL`CF{b1_q`;@h0aTV*ZuaESw#OahK-c4n9OzpAb~&)k+AIfpZ*7nRm48*o
zl#1^GlA#PZMX4VgAS>#Z$t?oR>-D9w7`NUpmD9VmZchldJ%(ZjYl^xGg$Ec$AagLs
zeaVTZRJZx}5ZQvaU9#&K2RjsA$W^FtJJR$>^vTIHUQH4KD)|>|3Hu%!l3Nv3*&G`^
z65z2|CB=qR|1Rb~-V&nDz|w8v4u7jd>L<#HmW_V-THv?bYs*_yNU}kzIRk~5<oY@=
z(wM{2Zk+<bj1>)*9<1PHMp5*Y$YkTHOGB41+dRnY!fXL#>9P<GR;x>wI?VghZY3zq
z3~^{X|Ch7&JC4TpPnM%Pw9J+%_SUULaGKDKi>*Y@<7#Cmv3n1@h8IXF<A2!iP-3Z<
z+ITz?jHs;#1hF2B%=UM0ly#F5=ZiXAI?2@13rPWGFJv<_pTU5lWuAtaDX9oQwAoQ3
zM3n4tVXHjDf|R1twXl7Oi!N5+8HlJ74cVCyAv}3=9A8pY!lF&~NVHaZZ_;HUQ{dw6
z6_O!mP_1BkE7ePuIbEh4&wo`4#T0?E>%|j!njaoz-q{O|wsDPz--;KMO3V3ztw_Pv
z3x{6YtJG;}^Mnf-W3Ksi-jIS)?_wm?t+BF5Udl904sHqCI)mkwt1C6P3^OUGCWA2t
zC@-Y8e|{o_QSfg_PhZXi70b{W4Wq*)Uj^${#OCz0ls*AR&!Fs~RexmT<aHHaWqfX#
z6#J&jhL|F`^xU*Q&1;yIMlen#ZfpE!<vK~#k>bBA*ZK6I?RJ*m=E-~ZzB)^;B`7c7
z`}V{9lM9pg2NR7XwPW2zFCCwi#Ex|`fZI~fx710kM^4Ij@IrOzIW4uA(>}!S@=%Z4
z>d8~frb?=GfajA#8h`&(x2!{ZRcn=MWh_g}Go^<xS9yFEhb!1nebrN*ggTcp>vDot
zTA4O+CyPuRIa^fjLzPzLNj)o8RYB?^(m-_-n4WU|Q@N_AC!L7<1<&BrA)d*HJaLK<
zl2C+R`PM}K)YX#bn03-~sys6u{VG=}Vh<_%dGME&r;vFv=zk~QOhv>a&T_2DXNj~J
ze-kX#pcF1njr?DoGCOfTEx|N%vb>s!ln}I(bwAgQ^l{073Q47@h1bv5O+m}7HT_dN
zaqudq`HWKe^BUtM33-le5FK4~Sq_42)K-JqTl7>>Vwr=gOz(vJRh>4s!46^eS(;69
z%1-8iEd^l>)qh`RPmR<<CTgSwYEl7pszZ(Iw|Vilx=$dTqlIC*Nqv@f62dXdRj-{8
znQ>xw?IdXZI&4FoJcWzBuJu*>kpGKq#Q$grM9MJd+<bMfjUG0P^=x-~IG>_ahNA+x
zOU-w4%OQ$Ni(TJ+xX^Yk-Uj=I_a6@mW0)4A8)RGnZ-3h1&a$?1$@{|xnfdg$vo!fS
z1a+f@l5!Sz3k&ziKpDIBeB%Oq@A&H%c(4KVcsTEhhkPvC4sz>hvHz<d@0W$&%S*c?
z_4Flts)sW9&W1x7|IX_UbQW03Y}iv@(+CcjbmIDv7m4nu&|X?0UrL{O@lmu~C;&ii
zqNd#P24Ql_Ps~I_zEeK6+w94<Qsmrf+a{_`f6~?ZFw5P;qAQhpweO;<ix+jPwu?r{
zB~qNv=j?4}fLOr*>fb^yOsUjhs@?QmL!qInM4N7qPn2%7TtVJ)8q~j&DES{<B0v(C
zQK&W)0XLU{p(zxX4ch@Im*7Y=1egDLIhB7g4}X9C>q!>WKfQRhhqrYg2s_4`wwFM7
zAmIxH(P;X&$?uw=qblDB3nXkH<oRIpn2O(<HE@22q+tXbm;aKdNk~X;F;yEvDue;K
z#}^wJ<07lV1DXieXO!ahdAqW`N{mg65Wz-X#GlW@+w<$`w^QYaHUceeYOHY1P2Yds
zPVoaM6xM}FYY`MozyD+Ue)=*(EzBN&lino$OKTB3{%b1`{ba!iiNO5tdAG(mdY&%N
zLoCDtZCC&L{l_$neEnSM*Q609=n*6JOME(=sw;>!VVobe7ziCKB1u{Tvxv+^%xU1f
zCc3K88<xoOlU8^isz&eb#|p*zo=1P8?WHyNpjDAdqzPUMj*{0Us&r!)P@t<k6CA_K
zqebo4ZBkn;9N<X7L0NC{#Yhy|FZfc-W8@xsF9O;sO)PYTf=A&<<P+q7*+TV8d6@dg
zU0Rh_0;L)EyR85Fjzg2f05?LFFN7`lLC&D{0Xb(Xqc#KEoi8(7hM~-p&U$~Fb)F$7
zNOhmADGzrEn(FN%LA0}Hg$g4%INNcRzbq)Jb8IW8!b3mk=u4H~MX%2Aliw+Ro<_5U
zYGSpHzJwlc&}1L882!2viCFU-xk*_gXyT7A@-eJ^g=Y~9JyN_yZHhoiVEPcbefZo^
zG>gJ`Ga@a|)uHId34fePn#_MC>UF<e^*)h5=S*PYk3K|%Q9zs@f!4T5+&7+ZGibe~
zUmK*(n6G|0B4P{_2}391Vo*sAOa71P3L2kW<exHK>5*&F6~j8|W-kKn{V57c(gu%z
zP>O=YduP?PDGD&(vs1h_9Z}v2rN_;8vJn+PSiL~^Ciw_?ODj1JWL1B?KEI-vq_w?*
zA{{6bC6yaiL~UXXf)a~#iM;S!gmAtZKVyQ>E*!uw8ayqRGC8pfwHryUAjmbPGC<J1
zThQcYZ=a*@S{1W|X33%80%v@l&A+PQQ{yg7l4<79`)Pq#^T~yLV$PI|M5j1=gbvW2
zbXSd#to%S84309a;AVe2*!5w`C}GH{533|zSOp@$VlU2!lZ%x@_T=^<l0*%dYw>Ut
zj!}=R`rMSfvBWlWf}E}8(KePQOG>UQezjwLkfB_v4>Huy&qBGI6&vXVRrL#g_FiW*
zo>TT~rCV^J34|px*xVm9<<Gc)uyo~{I9NlHPzwma*0_*D%GH1BsGoC8w7a-c`YM!l
zAHWreT5t<pD|}*%F4d0%HC1DF^<%UC0yM(fT1^{#0JxvChny-s<O)?44b7D()XH_~
zOZiL0Bh-o%YDIdJ6qR%rC}Iy3?EB$k$u>}4MXq>&-02kaV7I<@q?-%;dT%|}^^G@x
z3z|<7lu;jh_?&;^joFGp;T@0`lfOgG8J~)uTg(QegwGALQ`9V4ce85Dk_U*q?svL{
z6e($qeYF8{wT+T*L8+lNu;1m3jjQIQ)6gXLHqK5(S4r>FigWS(I9ox6+3HPR*L88m
zJh6Juy1wPbc`hbqJlN~&+3Pp-G!onPZ4}R7!K!0n0Dyn016(qCkv4Wr)|H8kD=P%s
zTF>U%wvblctZ71iENf)%+E{s&6EFs)*w82@x5L)1h1R&a1FVcqO0CA-j|T~DY)Z|#
z^#@vVL1p8=106fn=Jt&f9Z&>q2FgCz5-_nX7Y3Mf^N(fanu4|%Ni;IFmmOZLT`?SW
zc-`#ehA4l4?KT$4{W@vKl^YJdly&rhp7^Hkhqj#G*A5kg_lw<1aj_{mGH*+A9#$Q+
zZyg)%(kxa#7S|T-C4?QCY0~|%bTGPX9TvR2A|0fln4JCqp$**$*D)Q;IMxK-&R^s1
zsiCXoeV^XfB>QpjFPR%xwnqzIt=Bx~Rxe}KpE!S2oM#;!wb}SU)&)meob%y<E3ITa
zuXsKlv7Eq^AlUn$J$p1RCZ6V?J@&JY+9OCDS6*X}wQKjn#lgix07HrBuw&F2<20wr
zR;}t+{G!`bInC=hp1-d|(=xZU!#r2j(wFyrelW=TXSy$!kBN+zVj;Mzj4}jascIh}
z#9)7g2wh&;n@W2M#0kL6)|vvm6lL9FU8}5qX+|w0x;<?Sq#MZvEuz}jx5>_$iDq+>
zPxIHB>z#|4`98BOsYe$AJ&XCNNmZV17ueBL6|P2_5*_8_qt?i-9BW_`?)NL!LumkU
z)lwC(ia~p3I)_O<{P5kpT4#nnQ5ZAxrgDGP^$lj&m>r-3J$tn0*G2Uld#nT{HpYLO
zh|14XB9)|Kg2(B<QyLaZBg7t`Rm#Qli(u$al+)?czkhuH%eP;Zpk@DwQ{@CCI~Dx&
zmR~XU`dtXq&+={sfTiV11j_PYBH<YbG=Ue^fp60(whY0Lhnh!!#!-rrUScyh_4I!|
z<&N_P%@E^|Ao3{h_p=>jtmZ8eO%;v!7NL~2@s-x|s`wh8_MNBFs(((W@6+i^;VN3~
zl6uoFK6NYU&l*>0Z{>_?Im<(a<VQIuWFlySC8@3scGQ+JO+?$+JXBhQ;9QvMR+V5L
zX^7?_0>uz`h=0fN4JHhP(#RYJ1;&5wj?i)Mx05QI&D@>;{q3zR`$3C7H@_Looiafv
zf^<h3D@ffVtaOG$l3i`k=IK0}8wWKaRhPg!v1@3pYXCx9ewiAS5@~lcZC{SELuK!B
znH^&~GP6O)SZ1-ZVXIGFD2O|1wbDPiQ_8z0aB!_j!}T@@`yo!%Y`;=p%!Pk6lJAL*
zy+`ANmmZ8F4-wn~>gu86j4mh1>GM47@~3oFA=92aD!KHB$HIppX^!5$Kw#(3uTA7+
zX8Hy5nzZL8U92K+)()0;lWo+yDz5GK=ObY~riHbb7|#n=kBbox-XYLJ5^@8QL%zOZ
zhmFp!*x_tc*$)3T^|MN-yla1A<(>oK$m&wqocq%A+2b<LnooSvn4yRcU3_2M#VI4U
z=Qa1sm2CQYs~T7gB%RZ(T%NGPQS=&U_@LQU&Y%`5=ry0xH%qvT;rm1A+U&BsL)Dy7
zQ>oB*@AsQsMg;I9PVYOi=d<7O3_a`ojgHahO|CfB-t2r5*$d)}KYD-HMGNJE0K2$o
zVW|M4o%b(V<P&6+`7AlPux{jQxqPvHw#=6=v<?ED>soP}mtGH@H7g6kgBv0LLDxEB
zaSk$!yYJnCF!5T;f)jt5dmSc9O5qh08H0lGA6Hyq0CV+Adg^Fz<*AtQi{^TBT9Ycn
z;l$?p>o~(kTJB!O8PR{*>`kw`pLKt+#RBPs#qd#=S1-<7w%$$CEf}eIeW&7s<00;G
zgN?_QXnS~<MciC*c|R>$Q^>@TmSunZ8m~9HIOS0|m@hT@+8gd6zvmK+g5JcgY;0*C
znm#S?_0x0dTk@fwhdw;tqzM#5Ing>$@Nu_yEH@>2qsMXgt!sbK+8gHNdFbUM^M|pS
zuO%#!IOe2ll1JN)*2*OP4;newCU&znVdQY*4rUT_kn@$@9dyUuBP*G4tA&^19Iv+U
zb*;B<cV2HrFmA-*lB=yr*HuyS9jV&IPNPNvdcniyfaubFB^!-okibFbe~b$?-@TN5
zf+b|kh|K@-KD9JbR0#Ty+^42W(%73`r>^%U=31d%-f;7^jHdInea@+cBR+OGzVknU
z4Y!^t+5aEG8BP?J!5BFe0XdgZH8CTX9FjK<m(jR0D1TdBk6X79efO^*keBqqyySe5
zEP?<|T&D&c)bP4MTVo&AYc<}~*)?|M!1?z*!z*3MmeOcfvXut5cSl#e<Zw7Vb7n~A
zWVMA9a<V2N=Ps2Ra-}Y{z^%DdQ?P7Z%Av^kj6$>V8RKWGT|_rT{ri?fW=r>4E(H<>
zfQ0tK1%I6>noI2iR)|X8M@bM({q>UZda-dN4Y~Peh!!Zx_*w~+3-_V$lv}#zBAzOB
z^`Q}&N*7z8Qmqe>0zr{3=D?Yyixe1*yU2m18=xc!3{|_Rfh&(s4Q4UD=Srp^xpvK(
z1|dqJR7_vU7!u=aMsr9R66}VN@h!KpqY^~M@qeMQsCKOg9hr48<0-7&N8%u(xl5Ly
zsf|X`pq$627F1+>T7uf5Z4Fw-$>bI+jcrMS5i_*$g(VEuDB)|)ST4dM7%RRNe+5&;
zD_Zah^zLtGh-05n9?La~SXDQpH$bPQjPtU9br@eO0pE<{S2z)|&kM#Xs{JXfRcvev
zbAQ2(xr-KR;_8-DNbn^{HPnnzplg9-?gSZguZ#XMwMwHcpmOEf7fUEqX?-=ZkX2K+
zvE)##dsM+F)keKFg<AW5fOer`pxze!wsGd-?mRAnXlOjThFEAiE|@@3Os7AE_H}Hl
z-pn_emZA%-{c)cG4B1k5H{z~lcU~%ECx70MN1OC*A)~t#?@_JznEi=wTbKQ<n)|lu
zJE}?V(=d~qt<TcXTc+0gG4zAI+2`p7&1!v0-ZES3{rCm}Hgp@if8~b#{nb~)^IzW1
zhNovg%!Y5TF6Xn$`Ay#u3DELvcs9GadUt(3yIJgYfA;<C)y3I2SMOhV6N8(VMt`U1
z;TrGuSE)C9aCv#P=u7;;*I(azF(MQ`ET+1UYMP`xNi<13Nxh@fN$WmD7{0l>el@%9
z>ZH@*ufyZvTYP|dPNDz#`~^TQD!|54jU9B9VuKr7mI#+Qy*He^dpY+GpIlu2XLvAm
ze?Gssx*Sf1zdw8I|Gs=ZpTE8N^M5e>a4~=V?qxi``f2#_-A`}tznc9#e1A57{kOOC
z$1vgbzt7HR@y9pA%Qsgqho5t#6!UO%eLkGc=hqi6-_2(?`1*?{SmSyhKO6rE*1_L9
zPE0t*ggv~R7*IMY*CS30dl?g>+_p{Z1Fg2tGLIknEG1D)j@yFY;!s+nVt?ne=4dRZ
z!-IhCPloTVo?i_Q!<UR0Qfq0NQmTSQwhAroZe5~+@oBOyz&C2*V_KKQt;_Wg)>Vz<
z=)x7VM8t95alDffZNs*}ap0Kcl-SHsPe`^H_m(;^r^tw+wZ#T_luX!{K;srmX?I-T
zTdJ^J>;fwwMuVqdtr5LSM1L^4l?o3X<vKHaHvHqC{|W`+5XgV|?#-Li?JWSvu>zhg
z875qJX)4sBAtvt7lt?3xZ>?zn(!aU7K6wj&3nRD!zUT13%P^wu@R#@V?@s1t^H~@H
zd3c1_fI%L0#xvsZ@W{y%36CE<W+5Jp9}qsWraI#O@bvoX{A4zNF@HRL_-J@Odp|#2
zTdj{nPf2Yf^f=qPOXzVTx(Yp@qRT*HlT?;co1V4Rvng1$Ww3G%x|H%H(IoLC$t39{
z*_HryOHU=k?f_S@lB2-HjLnM(8e98JTg7o#ksyFPMp2JG9G#z-HQ@87$74|0sLrmI
zVO&#8sz=1*oEdnWIe#4hkAdn~R-^`6BbW~|5KP*00z9U?VO&D8(7?<ZD4!}!EPMbk
z6p5g$=q)yYVFoUP$p+0SmMv%qgpmw?=M3b|6&tu7Ob4k71TK>f$N>=a%VuEGtk4f%
zgICss#P4u+kb-$5CXmI7)B%|<{BKhOL^0%Z@XSV;Ac5EI<$vc$2&2);&qXbw<73j*
z=5!SkJAj`f{lM&yq9f3HejgjA%*73!?@Bv%w=IJ)tB0?JZQ0pRrQ_Q+$>D5E0*6If
zxwI~@B8t>cZ(Yjl`NxVeW7dh=AIrL#0MJeq7uL-qaDRI10s(s^WMN!%iP&$yFXK8H
zEhGP1kWsjz*?%@B{9&^N6Bf?^XqTpN=;(Ez9h!QY;Y4nSrsOGz+Ad9-M9cg+>GiD{
zHSs2+rhGR>O}b*#UgKOa+dT2!Jn`N<@!o8cR3@oTQu~9Ibuw$`gY6)EP$GC0$XJQU
z`vYg|+h21B_YJry+=`T<wN?f6a7$@F7SdAemB=LvX@4bx=pEC#;?)nNmj|-07x@|V
zvKp)l&Q;V;PLSGc!E$6W#MFwB6JBc4Y(E}SxaUw44=F7+J*stE@!C?_vm{B#SCXtz
z05H4b=4iz$lDXzXcnvV0ypC@r`eBYFc#ZFo0%ov{z8>0boNM?&$;1IfBk@tqt33tQ
z_7Mjl`hSf=*m{fKD1>cmD%iM^cWTOoV$z+Ol3^iLc4?X!c*yM&$NyW0eAdm7&wRI#
z&t)y-(^USVDX7yF)M*OpG!@KvdV}+RfnFGxvMR(Iz*^;qh4@n8^rhB4Pj6a|DboiL
zR`W5>7gF&UkgD!CF%NF0tBG;NfJY5(Lg~K$w0}dSffaD#`xVRQ+YJFBW<LN1cc^Kz
zO4hV7m1HIP&5EYRwtt0+Dw58;Qo?JqepTXEbU>V_Mt;f-`a}tvN!tOGNU(SED<24h
z2=!p5<Ok$_cp{rc>0033P()ydO5m}X_X~!GIm_zwGlP1t9P{)OtQ&5oN*@J7mYovf
z4}Y316XF~VbKIe+M?~8a5w|q;h)B9q(`*SumHvNDktA&ifOosV;FnKU0g&ksXE+9$
z??)&=NO*vd#0C;!##oRo6?VVp)2bu^ewx+1meh?a_CL-Wf9DIOd2^wp-EE<Cy*MFL
zNkXQQgiIv~nb?a=>_zHc{KO0PS=X*%sDC2<9TSLZAplX+c961Xiyp+0^OVwn@(Mse
z##~B-OKyf5gBcPj=#g+$ZFp(FR2f!fSxc4ufJ~0bCYpJfr<Js4s+<H6yLP`M=g9Us
z`QMgCUpygjKt4#ld@)b3zka(ENwS&*Fc)f@hY{MUvZ@*K1}v}Cx*t1mMD>f4%YU4~
zGxhQrZF_NfGe5hWUjSUkoWYsFN4!AOPQ2iQW~=JAUzOXTDT}jj+o>sW6tKx1n!59H
z{Uo<b)`H3bYvE-JFR(lEg0jvFbP6LnCBZr+!8RqqHr*Prsd{cxsIn<k*%Yd53RPD2
z=N`+Ap+(>i+-n^SKnEd@Y3unFvVWSw+zRXu3euV#Byj<sT+@OPPW&qIuWBLLz>@GM
zH5RI+?1c)dB{o}~{{oN@#pu|W%RmP?1m+U9CoBdEl6#U$;-?3GIXy-2$IaxlHD^zT
zh)&Fg1nm&lCy<z>@j`v{YLVxOmHm1uJT8~dqlDLLn_nEsUVx^42$)jqlz*GC;&*bi
z+v)&sOCX_6S&Cm|6~D*<Dpli_6C8tWfINtR*m*^fDW?Oh1=r|z?WrO2K)@ix#e@yu
z_lwN7+#K`k`3|xQ_hAK%Jw%}(lV4K{S)c@HH+qlUPd}{RSKe}^1{qSOT65J#1Y|9T
za7)GL{vTWfOqP4JYfenCEDA<DL(hK!Y6*&imr<xT6aqInm*F!w6qhKFIl=*5x9gBO
z&Ito6N6DALbvhIQFqZ*eGZdHQnmNLM@{cZm`|#(7xsYDzG@!_a$vRX_jnEnCLWV@f
zWGJqQapbDuMaR*3Cc2nNUHUtYS()wPD3vb^MbG)~JTp=8=H~U5)Qon@8a+}IB@i)^
zEf^zq%xlJXF#Lr4H!`pMIX{0BS~HFWo8T{r^lL#|2xam!aobx8XcQ5S#6?1Xk^o#5
zuwr?6GtCP5=Ye4QgGIg`{mZ*uF#rV>r+ljzm~Ap;tD?YfBiq7rVt-&;G-Ho_w4}hf
z75le>G-9(=TdrxZ%Nwn=spv-<jU<KwW76(KdR7{0jhb16W!=PRcr}r00Jc*YS4lKQ
z%{GO@xoH|*PsvR?t&m2<^X>J2NOx8t>J{lLa-&t8BH7GdX)r5{udnIG5YHReu|*Le
zQyqn$I56j^C$Dg&FVUZ(f#lmRn&>Ixj%y_KwZ)ch`A5mwx}<2A$7C*t4n}6<%HA^c
z($8d55s2ngvn7k0PcqOwJ;`)l%9rOluQRVQvhoBK`YLOuc}3D`g;3^yx+JW;!%Q-G
z-bWzoEkA9TP2T0rFGpG`W{jvbZ;zO<>~(R|C5&u_suEHyWAet<P&GHHM=A8dz+AHg
ztJLEPp?S?XrujQ8<Z!8RjdDy)=;ylu!hlmD;6UK>t!{PFJnS{kcd`yPG8`Z8hPSsL
z9)4f>L-LlNG=2Q~5I>-QCS_e<i&8=3^v6G@PY>@$tVN>ZGvzU9+h%KtmfOawke?_R
z<*`~5Y$(oJTKpex$b8*m@SfKcF3B=mz7K&hu83)#F-yN`maEV&GeeC+`{rern39n`
z_a-KWSC#LYqE)8)IzqYqRW@X-%_+`<e)_3~OyHH{!1n!yOqv~kwb8%IkUJ5kkS^(l
zJ~nDHO1qKRpJmkOwbEAqBBMs{6n07fo>A{g?iCCiNh{;v|I)zsHL}Y980#^es~uaI
zKC9U%@>s~APm$yIi8t!qly-|Z5CRYXcS^%b8{#<44ncE<8MhVqU-TIdZ~p!H)1N;6
zT`LCrPn^n{%4wf}4u1NYUjy-UCSm$g-aT)!M>}a1SkI7s059MT&k)Eq@^N}VE}IdI
z=_h`P!=Rk!P3_^Mjc{m!wn#T5bR?WIDocLp2Zge>M&hdy65k@00vlf$UZCP@{PpiL
zm7)GQJ$#xT-WSP|H7;exyoyf-!~eD<%WG=~x|gmDxe$GSC}$ao37%kiXEz5I+?FxJ
zOj~40X!b0@v5TooHOZHej%gWURDwZ<<T;M-2w`B9N2->RqA>aBwoFt}T_l^O+y4Ie
zwSawQpm*Ifqj}U2lx9qv^#?`hBCg~Ud1p5Zz&u@Mi%-<WCR?E_D~X5eEo?x-OTM2v
zk`il=jkouIXT`yr5?c@V#WBX(CI+Bbw}&c*9k(nT#25`N?+O>7r{*}3%t3|{3yf}t
zFHRf;zH{i=MSgFN(lT$O4Em93vaJl3;<Y;KM&m(v8)+P*3v50X?6@u$_mAl9(q$j@
zy#R=P?}+Bx<MCl8!vf$%?52=jTJn}EU?;-Kjhgs>1}VxeHg+RI=6Mpg%3Bw=3f2~b
zbrREjNVzB_usio%4~34QILtYxN2n;tm}<sM!~B-hOrWgZDTSPKx*08BzFGMO7mF!N
zxww-1QK)U*<2;gGg0b+8JVeFz94khsSkIeKO)$PKG6XDr1mmRUcQQvh9VyE&uG{1G
zUDSVnVH(m)sCY|HM&MA!qg1OVUDKCC<@caC1dTi%Wu3_@i(ymEdk(c9y<*Z<l6+~!
z13u?xmG1K`mLcKd?9(|~^t^0cirz*JO>3D_!!{pljV5BPp%KG9%_q-V<3^~sn;J8Z
zn#$82V`2;xrS@&>bA;0xR#4acE1Bs^N|uv<wNML6c)sLTan#J@Oi}bLyVe?+>qvhj
zfkg+edE6{8!JWVWK3R9jR(>Kl2!+)8cvvVP+pk>9<XJ}Doj_XP7MP^dnC91#xKtmy
zE2G=G;am7f^jw>luf1fuzk4OU*hv@+hP(sW-@mjrQ7V~lrCr39$4;_%)O6@|D@R*@
zFY)MLUPwX?9Og2=_HX4#%9T|}X5F*pCW)Sw;$x2kxKUi@jZ%2!2p)Zx<4?mX>45Dj
zz3VEc-Wr@VW-hsI)G{l2?J|ESqL+E%%imrApYA3!$d(VdqBL%J@;uNzA$z4eK7~x~
zj0_m1(XyY5cX{tdF5VLBoQwY}{r@9>x#**m*^gbF-I_7GjSw)%q?jM*r@_p{mhrFM
zPphis^RZz9JI-2kP_rKi`dK|QE@=6y|3y7B<A6&I>Mgtps0sCl`(p7hVP5sUjl+{|
zu{`rkTdd6N+G1su4c5>O&LPk2&vzSz;&?#*j$f%O4>}sZg*Rz$J>r~hLp;lW>UZxK
zK#1)Q7=&A+iMvxj1;q*O2<+|#2V`>qtG58lR@^yQLWrHM0Y1eiZ@T=x2)9JN=&9+c
zi<1uI6tjyAV(fiCvtH$JQp<I{#jT_Hc#5kHjpxMa%raMtrdN{Edo2~^iN0-~+1Dkm
zaUt<lB2>*}R>gh7Yr2T*dK$Zb)9te2{d}@yXnCc+J4>+Not3YW&1PsAI4Tia_M*(v
z0t357mQ90FO$IBP>WOe@8EVcVE7s8EmcHwH4STLZX}7#5acBiPrbysMz<xD**Ail7
zgY4^<NJu<@9OS>$!*xkC+mW}3Ft_skc=#?i1yY%RT{PoeTczQvslHu*H0BSR{kw{o
zX)ad=x+?-_R9{MSEhT-sGO%fWCyz}QN&jy1v?zt<1l7P%^YmWW)6OYQ8*1G1L+Nsj
zZYi9Of#IORmA@`-Ao&V&Vz)Xvbj7xyao^aAmg%{aO4yf5uHB&ON-*r5v$%izRF9ry
zP3-yJaz6v#_B3QFhIA=^10`$A839|7$)$6l{?;ze<UBz}ql|SYVZ3d+7aeb#wQC}0
z^fa|?h(_x}UvlA?|E4>XSmisEpK@!VX2#g*ms|tVf?-duy00bc=~ist@8=D`vT(GL
zg~sC|5wzk2d?i8g2<T5}CTa11>G%=-rc@WDIITIq8xxRgG{5G5a673!PLUkv*PO8r
zn9gKpo=ZTt33x6$<5>#biis_CT&U0TXkcXk#1(YF{+eYn$(Y#~dg_D|x;tL!%dTP?
zyXf}PsHTR_i7e)OhO7I5Mh$m6;KAt;DtB9JpDlE9%6y6vnw4oAyQPGcuWc8#anEm%
zL<zlFT3ElY^CJ0wBB%5j;>1oB_m8$WvOWIpt$VanG#fS3N}ImJz=+eg;Ot5-Pb7Y#
z(HKvb!}n%>$gToLBrDT`u^!DMi%o%D^8p-90;o^bH?Y%DqMz56d=Y4qw#^OBkfpJ=
zFp5mQc=J7qMx%hS#viDuG~y4&|AGOe^}4(d*fpy+ogTYf2=v(B@vg@@dsPLhIk*w>
zPK=CN1ZD)Po~HKEYhT!J-=AG?&`-T`r&G65|3Hf{aBk2JWL~91*yrE5LuhV8O77Z;
z3m%_08FUKKFX=qnI2@XJZ$CWz4@*^gm+_B06ap|fmw|#W6qj7a87h}Ura6)WIjniN
z*rqug1_T`1lSh|Ps5TS<IG4d#ITM#4tvNV<_XQAyNSbK_x!KSy$zjGAISA|^Gjqt_
zPm!$R_STkK1~LL#-OWd`78Z+N6|4F5W#ZF^)5Fv0hqKa?W8MfSr>Eyhzyl8d1&2m!
zCMN{5N=#4RroY{CE-kzs;H5cN3SJIAdw4f@H%fTuAuPVc-Hp@Yw)T~e)JK^-N(LH#
zjqd^}c<AZ8<8eNqCLZDC?*8%g*Aqw9ISV#^jBn$&pPo*?or)k*F=NAIjbnm~>GQW!
ze4vG4)`dw*7KEOD|8;skeH>>cTElx%n-`!qZmlt$+N5R1_{mzulygOGpaKm-8W7G%
z_<V+U5ZHkFbX3GWyh9&@Dy2U0nKAf(43wkoUDRl7=Aak;n02zq_3Fx!LS$f8i<U(L
zYo?95%?w0l5`&<9!|!eq<@otWVl4L7x5%Eum&v`7xJAU$GvM51Y*dN%b$p@_MZGV)
z#MXyymQ8h*UPY;e)~0H=`P4-}dQm@(ZgkTw^Xzdqcon5lBu#v3%k2HP<FxXBIz%g&
zt*>y*dPD2GzeX_a8E8s=+i1l>3+J>>56F#29j2$rE(~g~_D%CrOMc$zRc1?uogTt1
z>R;o$=gtnYjZ5Rb1S8k*^2(^`_h|C7)=wX3bIvSZ1_N7UrkOqca@~2JwSA?OXGW-T
z_PFvkc`|iRc8sgLpes?VG|^mtlSBp2=A@#!k}$1_RTZu=u9arW%5gT1Gw9H=n#@Jv
zOJ-9d^U*H4q}Gg5Z`@c(G~3NZTQ_VQGB?brV#lspQnFIC#mEHcdT9bM5vDCopxm9X
zA+38nkKC{ZxkYO_tE#1<UADVBpathKjlaRMyZ$CC3wvV=%y>16hhxZpi<t(@mN+js
zjgIXP<H5|tzzZ%J9NFWl>0p`g@x1<Rg2_Kji2|qvjNeTEono_yI<VF{cvo70q@beH
z5C8e{{Fl!^3kC!Jh$G;*2epadr*HYIhkstdG+*(%WLilWE?D5o>dC{F0ESH&SQ<W0
zCm<BN%ck4c*s7TGV2{mz;0Q9jmZ}nHnuv!qBefE}uWci7RY@e73LEiToCU|muP~lQ
z#jo+zey6T5>Yvl;c{+V8`V0#G;2oTnT~a3y6%G(#qf%<Jl8K;USmGW!*n~3{8apwu
zO@g#mg9-#-vfx}?4J1`0rItooYJ#a~25O@Bv3*7QLC~y^Y?Fq6nJ_s}j?>X11*$Zw
zwSNBm4TpV!K_8m;Jk5<zfeD!D##lh3Ce4-5BrNF?RKxDnoA&ShoicPt8{CV%uP4g1
zDck)uObRS~oEjFBs5e#Ck4M!5i{y<zpdMmsK`q=fu6$9;3W6mGv;vc^YH4v|NXHFt
z92}jzymxvPchIPRe8LFzARE<ocrNyoh#cT}+&WyP1&M+hdF5t%loQNB?_ty~?#=wy
zb7+f(k{USk7jmYg1I}i;kEfVXdPI()#0(A(|ExDRoU4RQIiVxOLxJOZ7++Cqi4KAo
zql4UX7@H#~5i|ZPI_>ZT*_{GcszZTY%TUHfv>&G?#1eUbBI!z-D9Gi;i>Mn(82cW3
zgnIBMVC4mIX&Xq!2i<5_0Ac**ZwF{d@mqRQ@HWlj${5kaLAmc^PdVwmc|x>{q_)Nk
zBrQ^}>+i~_wYAoMT<a=Rv(|zyub2pMq!er2i`Jr^Ww6*Ywc4x|8P>sTmR|N&scBr<
z2^=lq-1wt^#^#a25GCUH193LezsINN?AaQ=9ciufm0bj!cfU0gsLV>h3vH;t1v6TA
z`h54V_c%l@4*@hJL_g&D{U#L=p^EK?{2lbPO-KqRL%-JVe%U4@;ox%SB5@?3SS|Y@
zUuXz~WN>UoP}jv*3LI@z8UWvr-Nw70fBED6W>BVo4HM3dmfKk7eYcZaf3?cHvjCp;
zvW8aJ0N@&o9NDO5$%k!;hKdmeZkxp|iAO!nXnHr1V#)KSI0BAV?`$_yv8)y3;2Z-O
zJ|m9eU^Li>lOwZ19BK{C>1FF~%sab8brM@Ko1pDd{Fr)%ngPk4wT%v&b9GB8O4)2T
ztx~ps?Ne#MdNQtvO`H3WeJTT6hx>~)xr1P^R}+9@fVc%@D`;E*vK5tPAp7t1pEIDU
zVyL7H#0D|k30&p?d-2Jq7OrB1dQ(Hxni5{kw`%9XtIbeDIqmE$k{)yM+Q5RBgsmFD
zt!j=!`8Xc7HDN-D2O-8xz+E}HUX%=I{Lc!1s`UmUy+Mxqb1|2gnaLT?$BLF{{V&b;
zYk;>9L4T)wzgE#s<eOz!u8{9nV64HIG2gGeJeFg|o<}eU_H1(KNM$3}ul$(n$1z9p
zZ(fY~H!lz7`<HlCO%DEmepd4R7LB?`a!9ZL_R6&wQ7^4{lN#R3`Ug42ZpLqA{f%pX
zv;M}9ne68Ia@HSve2=WZ0L11yZCoqkub78Tulj7i*K#S_zn~+3G}~{P1gE90km{Gf
zv1@1Q()`9<n&uCxOY_q$mb)y)uN**RZC;V%k9fqn0#GGMl#vzf%xHHa%%gnlmuN`j
z`llfREts#w!)c~2Ce1AeJeD4CPT^F465q>nYg!v~mk}1wlD9+FWD&FCj7zv{h<vry
zZ*v<F<GOf$4r_J%uspOYwN*exk5p1Ug0yqnpd3lo?@}ftYfg+SPnnQ4@2m6OON;j?
zU*b)hFY)#z`I0=xTD~Mb?d&H}x%3=Mc`@mi=bFVUg>KD@Uc+t_;Kg{eI%xEN+I9m7
zQAxA3Zi^>i*1u^rLfIbF&yYL95xQITbMZy>4$%*Nx6X!QZ3^NXtKasDZxz@J&xDX8
z6EB7CA^>!U&})G132RHcgWbWhVhqRgKeV-`<<lWNuSAUh3ZCb%V3w!!><rT61+#oN
z^<1_mXy!+nuLT6yVz!TtOsPG8I&j>v;Or7bjarc*Fcs8XE4>tvKcRAjDS!j+kIN>n
zA=`}ac;&2g8#%{lSyDKqqsa6xX`*XsHK9UEAPqAPNs7`NTuIC?4!zqjD8est6b1I2
zv*XR0d<D0XD<+gX$gSiKR#(onbrn~V-H2^t*^=P9L?p!n(_yxx(achRwt&a#!$t8_
zGzKwS42-E~CgU8_&Ro-E`Wg33rtiHUv22*qV>yX0-taNYXWW7t)aU{$I=RRN(mecO
z|1Jw1p02kU=oIdI+}pQOL`l-W#_eg%Oz?9pv^UNu{o6S>8v9T&1kx<AVCiF<eu7X%
zwu<DG#k&2G%n;k!t9s0TZo)n5?>ZCv-_Ktj;kCHSUQ=)AEid24Di#cxk;mrTHcdj#
z*J_e#TpNz{MO#V6S{yOwq%q58!&5S<Zj!p)UWF$x%wC-rFKyBj-MAx9{D|FPx=?>4
zeG*cl5|onkDh{-y|H`-Yj8Y()++R~BI+x+qYXo7slMC4QwJ3~#bLOnNrcdc-+;b{@
z@9*zb4sl!D%Al<a{K_jFhIEE$JEJ}z{e^J2KmK}u(%HUjN*?CP%iPmDUi`wA|5*wE
z7b;1G$oFS&Qc4O4$=e_mm696Iq;STUa!bpZkfJH=EDc>LX=&|BN$cQJ3|a^@O~k{W
zr<7DV7F^JM)#ze><NswTskX4au2M-VtY%i4gO#Mz`>K_s4GRkw@I};13Q3L8uPP)J
z62e5-!^a9qg%(V?#X?eLaw-ph{&YbxX&!l`m{jHKk|V{Wig_nSe)oz=ErgO@j-w7;
ztY#Sw7L>y7v;IQnlmK@Y<|0J}T06kNtf;o06k8fF)z(gb)-fGaxo>J~DvkCRNqNz)
zRM!dcjfxvgTimc-u1MBPNM*PrHIL)ZM%IcE@rNn?P9%XIDzn<I{M=<BsXr7!xkkbH
za&s#|Ej`FGN>s9gI~K()<Ic_WF_uEDTpPggue6!b*2ieWVH2~><LKRTq{j6umm?XY
z-=-X?!8r$F>XASNz>&V79%=8l!d|o!%5zgJ?7DKeQ0;F2M~T^v{cz2kl20Bz?d)o{
z(%LSh*uP#u(cpwrQeU<7s(B8_ODPW)%NKFV)@Zb#YrFd+0`7Is`_Xc%p#%!-{C}vE
zjrW&Ps5TS=I5n5?6BrejPPsY90erU>x;X?21TlM?dzVqDHWUIlH<!Ur0uur@H<xj*
z0VtPo#5t6I<+p#m{=VTPdcGHVm1fSuWGpb0isapM-LGO&J0~a3057MsnttLhr=#Cq
zP%}=P)eL9`&JLWS>`JD+;AT1CcX<rEPS5dfJ83TJ-b-e+8###Qpg;SRv<|2ja9N{(
zItJthJI*z7)+hxaM@=&|&oi}2qj8o>7ohIJ7FwErMoglsU0PMS=@b51WWL)dVd59R
z;VAf4PA$q)me7&Jke@Hg!bdqGS6SvCX&%WD^byu08e7euphi2!y;%b_9jAmCOJ|Bs
z!d=L*$VHD_0&=9(19GG^S0X3PoI?)LvI#AfdYWH(m2XGl!ZWwBA|t6%H8kQgQnoVq
zmC9Rx6H@dcKgIc?lZ0?s<sv-<y<V~!fMni_QJ?266vm0#dFy};gCp9CUNidZc?<Wl
zKcGD}npi?LN-wlt1JxZTu1wNTME5S~D@F^Q#^kQ?`6gUfp<8mpyz(Qr=oh&e{UW#5
z`bEBkZlX2k7bA^~<TPLJXr3uN9HEkI86_)!vgWfwv`j^Dl=eX1UAKl4wXbys><T|s
zPfNuAE<MYqjIK_)y!+kYLv5TDEOcUbg?K4A?xKRtI+bakaqoE_w67)n3P3daHSpVU
zO0nG0+0OaNfpQZ_*p5Dce#^L4`Z^p7!;~NQ_le?r!Et)ZsLhCU4Jyz0W3qjo)&6>a
z;>pjPg|wQGp5lGoZ<ZO3opd?tAZ$}6;sf6WjA`!}5?TRcJ*N~i0?vHJRK%Dif62*L
zQid&$H~GkcTZ^sjmuQdfq<(n*$jMU>rW*Sm$Vb&DK?s7HM}&1tmUfaUyGCWub6U2v
zY-*_kJ&P+n307z3gyU!RBwa*IBSVCLigrCTIIt~3Q{`Z1aZo5{sc(xp+873EjY^>B
zR8eAwxZ?UsTi#X)3|u}@4Y^bo_7R5sJ5f>@RXNLuw%;|`)p6O<8tvuR$c}qXmutSa
zWfR5HJ(pPXIX~8XapbC+FV38;`4X)eOB0P;c@^Hl()4sWdgBZ?)`MxyQa$K@Tw4zo
z8A?4!BNPXH;neEM^J^-zBu_-CQ2W9o%5g4V*2*!hO2#)=(=lYG9HLBV;|$)7=Q3gN
z>@RaTl!|GCIYG!bYcEf?`J5&Ty7Om-x`E&o^St)Fc+<1lC)8%2D^JdF&^7JZ<h_p3
zZ!4aflKMFviWnqCJvi~OG^u!hysB%w7%UyC4i(pLVs|vzk;t~>YE4huD5k;MJ=g(n
zFT>N<*Uf*MQSYD)GhnB`e{8}JAT%@9Pf{{3VEXyb^nLTkNOfp)^0s3et;~%HC2(tH
z!?ecj8Kf(r^EPdEai4P0$MC6sL_BsQ-zsV(qDFZ+N+h16cRGhNiK_8`Cp%~mRQg4t
zL0%Kx)rAwAO`Fc@IWsPK*=$SHvaL0nJfD>e;DMRQH*}&N5_d2n7-2?PAE(E7N?TPf
zrK#RCg|ZF;<-#%-dUU>)+!%1joHXwQ@XM*T!acyq{c(j=T-PVZP3Y?ruhgEY&b2Ln
z-XU;c<-(kYI|2tPp6OM8;s%sb`JyFpS5C-YrY;O5L%v5SPzq5bR*D2lc{KU_L8<Dx
z#j^{HXh?nJFyq;U0p>6Ujxv2F->$e~AmuON+Xq=c-nS0|z0tQz7ZPQEC1fueG$dU5
zgxX_@-{6eeHfU4yLix6%rnjU5DN{eDqU1u0f{)G3Ya8hA;Pa(_vFJq19X?-z#d1SE
z(&^_WNqTW$Tiz}()KSwpoX-aUWtqot)(f~oAt1SD9%Wg_{fEC9B>1txlwZ1-QQS?T
zz<4GQQUkZSP+)@#6XS|^bYX9b1_H}P?%<b*27|009}Nb9-WUyp34ip;Hz4?;aWUsZ
zu?E5qOyV2Q5wSpjj=S$?cW8hv!57`zEYuT9WdU_LyZ_3TgGBwGjJj<5{GSTh+q`QS
z5dcgIe@7Szlw8w2eMA_ssjp6{gvhUS+>bA5!s~K_%qv6|hY63;D+LxG$ktZ^J7wOi
z$(v#g)}hM6Ai!Lx@mQ=oH7r;=(u{Ys;QyBgVPFbnZe(+p7p^o71Tio-H<wX0F(VH#
zFgQ3bOl59obZ8(lIWw0rqX;N}TW?&q5q{UNI6z<02Xi@`3pqt#AeTA~;Gl+`0&NXF
z$j&x#>PUvPHqw9JZ+KReoZYl*?TOVQjy=1}Lk{QieKS0yT&+%dqFgIYg?y^+J{9H3
zDYvP3iQm*FW(~>O#I=Rgd{UB-TqY}BsxnowF6q=XpM+$Sw?0d)6Vdj6XG~GVyNJKJ
zai0>3=scj&M3wtv@o=-gx<khtKS3-|w{*>&AsU}J7oAGU^8`?SUZLeY6>i!^3>EKl
zVacMcc2TBM=eaPfG@Z3EY=1kTPE6F*#ly@@u``ENRQhU;0p_z}u#k5hS}v1Np~*7~
zBvnY2Az=(8ouo+o3KL0xnN0xW3zz;h;b{>+2@yk8=gnXyW9lNlZ8EQf89dAfvBEDj
z4``EaC(H1bO8=Jfq~XyJWm2ytRH&0>q7yUN&@wC_ECrHxrQu;XDWnFe&GQi4UP9{z
z?;EVrMGO$>B0OnI-R^|xZT1FX0DRiF<&(MB{#Zuj%-e#&YBA}5Vwoy?>!MB7sH=(~
zHEg>||E~2}p+~Lr^NNR>cMV{1c%Ngf;uF^E%5a_<{L)22n9e3dur;fFGVG+a*@SRh
z^Lab0v(rLEOqbjh=Q=s_EW$uiBj_9Z<d*vqk|(b{d0IFdYjhDi@p)%`b}D^kiES6_
zOG`Y%^G3RgTIN}QVgl1HtIY9x0_Wzf9pdXgf%SdWkbSli{7CpPFNMXa%p=1y@Nc)d
zVjg5+*D|7nwtmyd-a5ya#<Kv3lTSZgefzIB>(yuH=NFr4_4&o+ck9dN8HkckSASZ4
zz4`(lu*B)=>H6j7IhB$W89>yMMigmDUIGg=LHE;>)w8#MFE%mZ<5%bZTzxjCf4O;e
zalU%C`peU=<Nr^7*lgZh{b9BG{?+D(w=dGmi=S3MzWwRVgYVYwR^Oa$e)#j7&DZCf
z_2oa#Uar%RuU9W#U%Xhoqf~OD)z#(8eS?1YSn7kvjeh^|;R*g=?Ju7P;yHHnboJN2
z{T&gC$fW6i{O#-4r~60XQwppy*z$m3s+JUPI$$Vlp9nZ~Gjv*gaq;@%^4S}xKXE)T
zt)851u%Yv*%pd>w^X98(o3qUtKImUOx;Wopkw@HJEq#?o;fNwUfAGLy07w7O{am^-
z4<D|cTwc6<w%$BnJ^Av{>f80to2|;-MNiJYUqc~(*z<gI)sZasIi9YsF5X_gTwiTL
zIllX5{oSjx&o6%Nff7j#G(W`*mv}Z}+q?R6P+e^qrmJxBd?uOL%S<xYk`1X2sSU}m
zrIv?8L+Y!Yz8}l)1>*j|rujn;>|&mPKa+|khl1R4l3+;Hg!DXJeKukGXREI+zP(s|
zIekKZ$_?}Y#05Sn(pYmr+}*P0U9^xaDRL7lCotZm1zKrPgQx5}N^oFU_*cPCtftM6
z*P)b_!lYuFxJ0#pl!xD?YNEQU1VIBp0)vJG2n294J6;QrMlxiPzyjWa25YK7JYi3J
z-2p3Ld<Nlyy}&F*k5_`Mq=v|ZtCJw5I9oD*m?o4f`mszw)lxxdGX6EdI%k3-2wGX<
zEO(5G!?7M^MCujQ7VB9&p=|nySpVw$YIAnJc?BHoFb}qQ@R-zZfd1|o2LSh|N|EPq
z*91~a4S8P>{nq$5H*P`3XH+BC#x0Z+V*Ngi>*c{?0NXEEwB2SjOHrHbqtdoZd@PiI
z4jhV-Byp_=4Fw!A$^Q+VwoJQp>V79WjV|bhP8$y*9K_lNv9>|1Z4j%EUw!=QH@~*y
z>6EnrBhWrn>c|ODPK+AZkX~76C#?`~g;2}`zjI+HY<U6T<dO`Py=ky1JPan2Py$?!
zg?5yvS&@206zxz!!&NOWf!>NmbtqAPvZjcIUsEDV_gr%+rDg=45olxyQWyAvZ?wR_
z(th0(B<LDS{Ffd0-)cPy<xCjMl!0>SNRkSFM3e&#T)ukocC)_fAcsDI_R`1Udy0xA
zqU(B3Dw5EG-Q%8|9^fy%+qMd0W^~?y_$*QD5Mz6B)`5~#?$Fs?t$%vAzDyT?m){>}
zXD-|mXp|pO5sKZPol)>dmnp|>3vE2MB{v;3p%4+h$8C{gTU_7g{WyF3&)xIw5u<ws
z<(t$)hAN}F1=$`jlwrE8hm52gu;#ycad1pMK)h2A#F5pV>H*T5dLRxMXC{yH#cRoi
zREN}tblqOZ(c&)me*h+WO>Q@T$3lkcR;=7(Pz#FLCT)RE0bH1BjU!!j2uNHMD@a(G
zL{Z|@x`)WYaxQT+PL|OpI4@966wfK6e2Bi-YH!Zbi}kix+Kvup_)*b`TB}T8D7K?`
z9J@*VY;x;glm}S;9(f#nDg0~oNQxm!&No=hgnpmz6N_t5Og)xjyrq19D;uNN)O*dw
z8Whv?xNW&5lAGfQ%Bb>1?r&Gbf^88p?HANZ<&c*7{b5^D^e6@n87Wj!<3mPvyu$Yf
z3?;yLW<O-8Yl2^XW1{_FZY{dYt%VQa)>i5-xV5xf+)9H-Y49iw9;LyfB@c;)w8SyB
zOY&%_NI?T-V&J}lz773<9SW{JeVkU}iWZMc<B|<*wEzjlxhS@yO(XbN<3JD{5P|`X
z{;m>6Fe^9+rHL8T1|1$pLlwc7Z`w&|k~rTZ4SA6Vk6rQ-Wrhc(#_2AYJx+TW{a1@u
z(iElDNAj=$EKD7)Wq_^p^P9jBiyCOR&VXaonp;)4#}%bgT|kL{I7Eu~m~nte%{$h`
zQ2?1hX1^S3M)>hvwfF$0nGZ>Bm`3AW#69w-qb%^-2NxZdi~P41?be7v9LOPb5{C?}
z-i!K=*i5Y+brzbj6CX-^$lDKbXg>FRAIuB7r{_|-8!zbXo~sPx=Yjk@ke|ztcu1p5
z<I$z@U>6?j!h>CStRZ6!e;I4YSVP7dGS-l>hKx0&9@Xm^dThP*TFA)Io!?^}Rx-DE
zb0NIMd!ZZ$*Ah2zdYVYQ7lySJtJ_C|VL)#ZEDjSR8K)9wB_5~g-~zHe0Enopl@ScJ
zc~!1jms0|Zx3~cYIOrrS2Dabj0=0^urHYnRO0%L%oL^t-P~erge`w>JYHxetowBF(
zE@Xvj@4SUtc9Z3F01MOrA*zXkloaOJqXW09-YibgwZbOIxW#l}vmTuVjEEk;zZx#I
z@3nX_LD64pl}03PyqpdwJ>w)I4vsZmM@vgsWA_ZALW!zzO#@`YZ9yv_H(}!b?lvxS
zaYRSnQv_QzHLp$Ge*^Y%uDX~K1+^#njVq!;&wJj=BPXQ=>}!_hXkdLgR}^ch0@zEj
zel&&b2Q=@&2c#vk0T|(^YbtqGTTTg7X)JLkGj7h{9~5o6r%0n(3~sOrumny*rM#RD
z#L+hwfHlw$_6V5Rht-x5Jn9IAcxy_$oG++ZyuiPP2SO*$f3oX}hSY4;l}$LB#45{$
zBsP{P>Z~nF62ik}*A<lvKwcOq88;vilFNAl0iG<b>|X1DMnZhAFVHM|aiCK+(55Y@
z11gfz%1BK@dSu^)Y6mtH^b6oBWGA3HFGt-CpjZ)@nE_Eg#jU2cdvVCnK!Q_5QD>uy
z0F0K!G^j8<f75k5Ht4bM{kt*xrMTA&bkYJ<Z`{z^0zK?3OLtF*EVm>o|I_bcYT4}z
zYQ9@vP;Xszj61s5)%K{~byV*<YH=O)zK$n!JYnMr8}+`8df!I9Z=>F~QSaMW!|L(p
zPurM^YaP*%UuU<qoZR<Zh+_|!S9L{jHAh{zTwq(!e|}toMpB`=namUo$ezOq0Y4U^
z41|fx9YE;i^4m}aag^qWc1QmfHyc24J4&Dw_DU4_jl5}LljU?k9V9eT9F=b>WjDt;
zqX^@;gWnv>GYa{|ROm;4x2zDC2zEWe<gQBI4+OOZ?=7c6v?av{*lUAhcsC@GBxut(
z?(d=!3PT07eJl5GzmztY@rOke0x&q2&><`omjc&0zJD`ahy7nKzwdZ*z4!@|p>~rt
zijg<PPqK-^+cQ*BvoFy_eIt3SR4gkhE1SJSppq*k-|iT`4|W-#wJ>p}=&};!^piYU
z)_(g+m|C%lsWVIqJ^Jb~R5W@d6zH#b8@x2B46#nOBcMClI4X}C!rsHJOm#Yebt=f<
z9Fyeh;D1nXMx@Sp2&M1s<gjuGP&4VUgn;~zHiiP8saNnRkb;~7A5ufddcfEYY_fJD
zF`Z({=+T<^LV;5@!+A`B4K*Gt<vlngATfcOS|Tqkb%i=4$8$tMr4T9OTe>zqLLl<<
z0W~PSr2s|yr52Nogt976lgC2)UXIDX6l~PKMSn<Smq$=6E=G$od=Db0G^WwlXg;7&
zj}dj>Blo_c&PbVb&CB#FNP1l6{>c!vMNTM+mMV$m5{xPi&Uani+dn7{)QTVTAd+!j
z4f8p-!W60;qTbcqD#BPwPdP+%9olvQX7DAM0%2AH;fJdLTG+e(admcfHs(0CFmgQt
zT7T@PQHPe#sr90-HvNX$g5~*S@U6rGrIMJ7M9q^RAh8;52~62CDLuw@qUF^%M=r$G
zcI&N8s2r2U4yD|(PLsXJlQxV;E)y1YS-v=<fk;d1G-a&Sbk~kUT*X(vIk%GM@e0MP
zl4ZQQK@1DOtnyakS9Hh<yGq`y`h<amI)5t29M-T!m`|!AAVge3{d0jwRbYg$s`~0V
zuEKuB4NZJqUWv`wRTpTr^_aBo9-eg$DbS~!VrxX}KbJ5#?XmE;uS4nSW%u7Arw}c%
zCQg6<*!eG%K&FkG0GJbM`uWfFefP)MYXW<`2fY&_%C$!9BdQr$nov_jUDZ1sntvBT
zNt%F9^<T)duI5{l4OTM&qjB*i@Y-tn5DRKpxO+P((d&p+bWbvL4jedU9DE8%Hu(DS
zl5BA3jgkxnFo8En=996u<fa-8c`>GN6k`hAMlqIJHew824oGBX|8y2)(NbaXWTn&(
z;Th#j2=Sbs3#%<O>6uyIdBGG|8h?6DO3%T-KfG;WU)3v}m4-zEQl?nR+9lNpd4wV&
z*Z7#+V4{lTBI`&Iz+>gxC2JP0Z3SnYxO}NBN`!cmtqHxv`=y^ewd;gVaX3`HRTlx(
z=!K$)Y@--VX*oB7S!C3U;!NH~g;BgZ4EdM&yY_l$B999-a;PF7v8TLX*?+#f3Q-n_
zyRK0oN&~a^mLWHI83cpmx9I)vbjkZ@5=Cl3!o4(!B2}&VW=(Pdu9s*O;F1YxMsLoD
z6!3dTN{>(6R-{CNy+)@<(g4UmhfW!M{dk=+IP^xHB8{VU=p7RLr2Cv+D$opu!zNPL
zb?i2gLd~X4BqJ`+EwSIMTYsvRkbJR}!}-d-NwmAvF2zP>-dcK%dHa!^W8Ug%X5Q${
zY35y1DpDJI!91~ZFtW1gluv`xp|G-Q70g-e#8!2dT5i${uPcJZVxj8jnvKMpjiKE=
zF{_)5@2XKG(t@Y2(I`@qb-1ZU>DG~bIjdMKCF%7D;z>pLoKGt-)qm%**|wcRtEW<?
zAJZLl3pl^ge=pra;b2l!-Lh%f-hy=q)e77f$r(^o<3=aX$r<TkS`6z~lQW=^f%qJ9
zX7Kgn<;>vF8|4fr5=G{Id4F<Ll(FFC;7?FFPLC?bnXM|vnPy+5?0m>Z6U4r-+0j=u
zLXEXKab7I7Cjj$s=zkXcU2tXeWHE}@PCa68`P8HFWsBBw&0fZuk^zZfzdAGBhSL=B
zRCs?lS8C>?ZjjXHj^zr|XQZ0w47)p~Vc>;^@)R3#R~A1ST=Tw0Q5-H4>$_<P5Tp^!
zH5x)V%Rt^#L!|c;-<OQbPS{2@&3w+klb92x;gL<pEuFu2I)85wS}>#Tr4fX-jDu*|
zs=|2@mTx$38H{f@ZyBEWrOsP~*C+5fh=1_)<B5N8=#9iLXtGe2T%XLZowxLc0;7j8
zcZ@DJ|M;cf>u-OUGQLf(?uC!cbM7?cl6TdzHxf$?^8}kJ{02alI@7Xf?!7GDUBbWn
z`>xM(X|JFvLmRhxa&rf&y2*CDVBD(S_cy%O4m3hE^*<$MIG16!3KIn}FgY+Wmq8IO
z6qj%dE+&^D-Z_<j=lZXg-&eXo*D7%Q8b8KgzrC#fUA2uERK#Ioatbs|e}7-a7Zymd
z3zIP_fSrE)GksnCF-onvhR<X+4`MdWOtdhYVMXL8W>n~0Gn<&e>|}#l9wQHD(^(VE
zQZ`<Uk<7(;+)!R821%P&R@AgVKfX=F$d~7N-@z;IaYO=tzs6@_%DTdc%Q2X7MZzh#
zNU+F1Z3PzzUMo1smER_hMciquyrZ}-?uFMe65=$Q#FgzFvi8x4td2BZ1>akiAO@Cu
zoUu`lBVy&Gw1i)R5!$afZ_|M?<h_aZl*c1sw)5WTm**M<`%r^Z3_L)+jUnTlWX#@>
zKQWyNL7_l@UE)btpvczg#71F9H&yK~89}yx+sHnOT%!s&=#=;^{>IsN#hcM!R^-~J
zVoWNAEi*~CnLL|2#m6q2!WK7Kw-+JF8Eq7jE2^|mR@v;RsMAuxGDlv~S#>Sl-emc8
z+8u4m+Y$`Kb>H-~+*J;xriIo9ii4B>h-tINHs+dt95>^sp}9OWS$hc$tpqfBIRQxQ
zSw1;vlh4P&H>b`>2`36xX-UMzx79*a(ePfBj}%4DDhEwMjKp@NHaRK@^KVo0Lw6LD
z=j^;+fHA7maa#noe4?MH-745a-pF!8PrK%JNuv)nj~+^M@q2UCnfyrTm9umwek4mu
z8+cWJ7CFn<$yZm<rZHN9U>*RTqD@Jv%ywMyhS~li+7vyCjr|n7IbeN%;v4{7$ecEy
z0%z|N^hdX3oPgmGOyX2aBd1z^EvM$0ui@0BWsg(S`o^hQTJc}jUspy}Z43cbCjX2#
z^V(2%u!p1VHgQE^&=m31@uP0*jxlhSA?r<l2^7%DAA7;zK^d5fv5u$pN91WMU4dhN
z1y5URCE5CW^7Jl?O+96qn#Q*IeVSY6)l=n8A0Ho7kS1xlnLBZ`C{z~lP8@9^mLado
zR%h8cXRF!NXVJ54ISj~7a&z?3xZrktlFn!@tFNGDN%RT<KLs@pSl^$T2S686GqRC?
z9A(}<*pJh)0Kh#UiIyphw46p>Ma!%QTBcAvmdP@%Ze&Bgkxg4`t3De1*ZVe&#TeMO
z*;t~oOXWI}c)Ha~g@GwiT<JZp>u4aY)!mGQ=yVfZv~ZI}dse7zu{%i2NOF92yZ*{0
zNp1YyBjCn#g-34KD}ui9=6z&y1HUMLM6!&s*RlL){?c7=yZ_-u03s?&d_{qQlI3-9
zH!U!>i#zUion1tv+yHZ4NMpJ#JAL3eUaSw87j?wj)DXsb1;*T|hA`eL4DD`aY0k54
zdEWiEgrM{{1A-I9LcK8`R(!NZUiv&G5<f+?FkpTEYGD9$VYOh;NOZ%EqWY+R!-EvB
zjz?|t?f2~ZMj&oT-_@w$vODtDLD$rY<K1|Y71oVJN@-EUvA++f-F#1sas25(Al8wZ
z!Z#g3`US$yd#O|RJ*h4O{iKDuRQrY`7yB2ug5n^Mau#k$aV=7Hx;sh->jFAMCw;LU
z0N#PmfP!+&cj7Z(R$`t$q0P2`yB&9NP#^&{XgkGSE+$+!gzqD97oZe9mOce{4Oriw
zy9PiPau-O!DTnle2tMjuCq}iS;Ql{p^NrY?+TDFg$TmsE59K{$#Hou0oclJr&0XHI
zM`~nNyEK2SA6`V~cJAHYkXQms+O=0OizE>A_NK%F<s!u0f>;*#F5s04A=e9+{0EP`
zcO92es5TS=IWd>fAuJP@-{U#Cf2>XI!m;`L^^cXz(53J;T!zzl_xtPWzf~OwXo1#G
z7D*_nrhk5{!ULmFSn?Bq@XAa-|DC?A{v4$iUBfw=%_%mUv=*^58(4wpCTq3OM8#}j
zU>bN}*<2=E?}JIWUFI{T0Sd#S>fg`r)6nwu%RDygc}y^BQ=h}pnXA4*e<Z9@2lyD5
zVO+?L;GZ;%2N|<r*CFM5NNw8Yw)_mkdp0=^VG}%VqPRSqC5s-7y5cM#p`EEw7gVVb
zJ+z*|VGAUHwNY!1u>jZjdKda_8&rVKK`<yEPNPnSK4|^=rMUVKs}H8MBi09+@(tsG
z_Y6dhfFc-`h08cGgS4Z=e=07JUKsDLh~T5nB{r%V1#_UXk9_oUxiXmS*LgOq1F~tO
ze2>|luC!Mv+|w{b^OAdDwtkKUl%;2hq}WTRBbcp`g}KFM1ldzD-awY9ocp^?j8)ub
zY5Rg<1F&wTyNI+%!7{>amMDwkD%Qr81r=g_TBE-9g%U&oam5#-e=@pjbUbogqtV4b
z&CDtD?`H3m85yjhiZxsfO)0p$c030)T4(ueJ+E?chAuuiHA~B=$P0_Mj><N;u&y18
zOJ-(5M2=gi60~gQ^fpb77(0|EL$rOf78su+(w4EN`LFXt(k{+-onH^lVa{DK&j76m
zR_Ou1>}_!%@ReH6f6LIgkcR8(iVJBLnNYz$iDeNSE1g-aGv`INyohz36$(b>r%2Zu
zVc`(&-I(GIjB7jyB>f>6cfk7oj5`3jl5q`jn7DgH^HFCzFUTBdiWIDMF=bioW6N12
z8hf53(Io2>iKZTfMAMyLAdjT$0B1`jm}N(4Eycw5;S`A`e_17v=W#)%aCaC|n2Q|~
z%av-uU8dW4Ty~J35`+@G%VcaU2+~k#t6Hp3={!AaaaCeLZ<JjZ(l5F=$je(VSu-t6
z+0xRE-M7hx>*Tr717&Jq9%V#l7j!7S%~6%&r=~<%TZ|K%#uz=dKZ@_bH&xcIlkDcj
ztBtOftg}xaf0c|FV>lDOB^fcmGg#qWlM$JVa2eiD$q0AVyX%q=L&6w00PT{H$0Q)y
zv0{2w2BN(dgk4fN+wd#{UX*#{>V439D)mUY$_Vo=jJYW}<ju*4MiOM<LnI#q*7r|7
z20&LPA6k3i=$1e}o^-IGJ&qbu-A`lEk?Y}eDtGs0f93ApN;($kVXkc3(88W5Yp_wN
z*|W6+9jYt^cN=9ww(&qF(V&F)HxP-IO3;0a+7KdPL-%vs$W?i*uj#A5dIJe;*o*EV
zKY>xehVsokiBW)9M)gN$%Zm+p5x01bJ~F-jlSQo)&glW~FSd^i+6XV{LlDb=_5F!u
z0CXj>e}Lj(2X&{2KI(99IP0^cwn#E?Kr*rAERu;mPm&DKo7=<Gl1#c=s%$f=PV1@K
zrqdkpQB$D-wc%>IcV-=JV2#_&sqcP%(vP9kq1aggcZ+o_NaQ6Ck8disl`Z4<^0B<B
zxJpoT3vnuIg*X1L%&D4WOQvai1xsq;l)wxCf1kjU3OK=2`bj*gv=!3%@%vP=-QB;i
z4%*K6!aCyn#h8)R2<h#IV8#LK`!nMJ=t^dEAR1mgKYri%zS@axtE^O+4BzseT4^n8
zFsk}pG}ls<5W48<E3%5vHT4>GHQk6sjyR&$x_KFGU}hX;bM%1}#8rg`a{&X5XI_N=
z1P6nMH<yuSK@<cqIWsYr5il<lm!fI{D3|8xIgfwMgtq)=`{!cJ(7$XP9K&rHm7puQ
zWq}a|2MJyj{6@=nlkT-hAjrQ73KBHv(|EM04<OBn`+bQBA#VhwTW``bK}@`I1baxi
z8nt2#9mv(j1_g0eGwBS=W@tFU6y-)`C~t#4!sWHQ`?_8Hx2RgtMv?V{HBKQ6|NdHp
zeguC~mi?f$@_@td@57J9U;Uz4-~BbZPBg)2Em30vZ55FlETXb*GC`xhle!xf%5wGN
z_s^m0`Sv<(kBu^9dx`KzXpP&{84i-0bQotYK*43QYgE@+Pj)eLyU3MXvQ@Dia&!@N
z`%G26_(1d(d%>vKx*C1a1&WgtaJ_>hpDll}7NPihm;Jw@(&U^D3YF#znWX5Q-fu@D
z&a<*$d#txpWl|$gpid}Id#{XO?EpnEkQX0o3?1B7HUW8~5HI!Og9c`j)wy3?yfqtF
z=0C-KzZ7FelZTWfl5mxkM_G~CtTa|bXO&V(bS7y)yfTl#tmv7m#PRiatlu_SN-=+9
zT_fsK&wH?BE?>;Q)bKo)9rqEOVqNZv8=UF1hb)E$6gcX2y5BmX|EPt~TEz_gYIML=
z<6doJCu%iObGA$|P6A6Zma}rTe$*-K_Wp7#^oozNRiImpI%A`n5ccN5OqWEG$hEsE
zUD~Ag7%{69#5$iED0S=AjQee9$7z3)w35K-!0ymy3Y=S0CutDYc@%e&mZZv)JsMq}
zt;jcVXlBGt@-Yte;;D=~SnrqkFXB*?0A})^f<rs3@6VwfpbI$^B|o9L)|(F++L5FK
z-AjpF%23o;cr^B$#iOiG@n~9$a*xf*uZ&REsGJ$c(<+a%wZ~0njjL_uvnziqCu#G_
zfM*WunmIXt<JXzftg1XS@p5$SRl3*3!nQ$Ic3*JQGKCJ8QR8k<?0A4hh0gR06?PhT
z-#POgxl*`Xfi+ifC2A*$8*j>$p}BlMnM%trJ=VjR0wx}1@}aW?i&Wl6In##vy~F%~
zO4E2P!N?DyOXHnls77to`DTAzzKJ$733dpM(xy{RLZQd}!D!P+#InavL7N@c_ovMc
z(1o;VBvm2fjlBAx!M&trw!O4@DLOAh-L1>eWAvFuG$m)zCC(FflVbN3oORUv*PT*z
z4+W>$GuzdX>j^V%8g!W;3#BFvEq#!CM`Du`lmfylh|K_~F(laA5nF#Owu9e>C%qWM
zsnp_{^T}__r@s6sS#QoN0$Z>TDgrf=5-NQVt7$38c$`vjqsy}uc}Wr2Nj{?p>}Gs4
zHq(;a*wLq8vkvR~vsnk|LN?QqpU~2k-h8SE?4*1g7NZ^cL3imzL7uy57|%9$ka-HH
z$FazF<r57?0n!zG5>9^>GValQqGzur*ZCe!aEetyD=pxLQ!=02p;7`0@8tZ0I0fWn
zYW?|9cvzWl<CXA$x)*+wR%E2!fqUEcb60W-NUCABpMq05tnbe$9iR(21tdoy*&BHC
zVP5fyaYxcet&W<9i>ar3*cVx3J!2i7zqH!pzzyr<my+DPx-%^geSqgBl^A;kEivEv
zpCQ%nxOsdRn`{3lqtE+iNQ3c$$8{Ys2=V+6><2(lmr<xT6ahJvVJH)qLhm`g0zn_Q
z0`NI91_QsfewRVQGZU9E#5n{3HJ2ejIV1u&IF~&6DlvcETU~Ek#}R$kuUMcj=|lGJ
zeC)>*fdMD3Qv(ia*bb0Z(1R?qu~J1eBr0(Jeb4NrmNaCvgd?*81&}<voV&Yo?(WRl
zxmT20wvYm)R`4#|5d|*N5eIGauM!H)$22sVkL6%%yAp`1by(31^>!+L%WOw2CD><}
z0#TrniX(s66g5Y%D>k13!z$Tf!+X;iwm_CUEG=-#wL=R`t#?{WU`^%7VGW#RJ__5K
zONX#+vn~7)O8sb~AUxYP4x+PdlOTC*TZ6QuK1K!7?m-4>#l0#)jJpT0v&xQI&_wg7
z6|fK0;J;Ftp#<gH9Y7x{GiwHeJhNsLwDNq^=`VlmZkdFf;E+QNnTa~gA)~bpOUSwO
z>tZ$JI(K0$WTVtOW?Vt;%o`9)Tf#4elY52s3afjCQ50_O6__a^b4L#SdM1N&McWKp
zDA_s$Kc!gi5C)A~hb7dc+F=cq>P$fkRZNc@TBx}@;G?W{w#8s-n{CNSwfQK<$y_?D
zp{aioep0yMQxphaGrKQ}b226;5WbewyeGm{EZLng%wEix8N#5|4hb&X>>{Ut;q*?K
zf(ds5nS*KO4h1v6Sce*{rTM74pUAwsK2BO&yC<?XcV6;r-&b1&iEZWxOgQvbs&(i3
zd=#I5+Ak+K0c`6r0`@kG^d5o9XD6^*<F9|?jKLH4u^E~olHQj2DkJUF&&1bnvCggd
zx$EmEc;|UU=Iey1PeB9g#6ooR%{SBQ-`_8$$CsDO)%5D-&C0($KfnBUn7&<Jy<J=#
zCxn-DGW~6OHhp(Y^Ou*4v(+)FMmT`D5Sy<@Yf-TxbCrn2bTT~-X#Lgn{qps4dJ=!W
z!nMjx3obE}70gb~e}06&dl`e)Se%yIG2|K%4epEq?qeM%u9ygoEt7s?2%qj1j0L*w
zl{v!AkVzt80%4v+pBb4f7S1O}A9*o79?hSv&X<=M&yO#k`Rl89tJVAKzf9B5=c{)&
zZ{peV*Xfs=U*CWIcJbTv!|CeX-`{_)o?WgMSN}RaTf|>3rZ*SMH`8wvIYpYTug<2^
z)#~c}&CP0Yji3MYoUj1Ce$I#x@ayQ0a%&}e0^uBS)dE|M9%fY}R;Ih$f)Nl=xfa2~
zMWC$d53wuFVr^xxD;N*A`u}DZ@At^AB#wgEcFQhOTSTJz^6YXcQ9b^w*+qZT(eh!D
zi{&281;v_NYBeG#e|d6A5g0thugN7vU?6!|>^?d6jO4VvoPPSpKat)8)^)hNxwts_
zv<327bk&%NqD!0|Od++94R{w*E;**GyO@$<01y!y(=dIvyjWhndJjT^_z7Wp0TLEh
zmx1T6fBmrf{?+PqwFopHc)EYQT%nVvo>ddnp7-(;*#jZHxA5rE^x|rH_G+;@o?bk8
zI=x<eSe<N5iWjFpFVHQn!{zF_3&PUdc)7S<-dvq6uGcEkkNvQCdw%-u@<X3H&~c#}
zpP+>+oa<)L7;e_*+4Wj8_aPh|&#Ku+Jd5Tvs%4-$(6Fn+uBCx!ARd3HdwOyotF3qh
zk+Im2;-i^;`Nf8X75koJz1X0LBTG4~P?ovHf-Ti8yID%Cc@KF4ate83!>%HRnjRRt
zE-l11>}rCnEPG&AkQgX$*p=*J>%iFMq7j>>!LCaUs{7D5ROWPXt8o}xlX76}l2B~4
zVVA`4xd(R9eU^>HNfLjP&;yh>h?*Py;1i;JQs;bRHp>j;=~!eZQ$@BD-N{t6LFepb
z%9Nv5I(9I%9F@xhQ$&{Nt|Bt~EJfs%w~ZodL*LlYH#YQ*4Si!WkPegw8XC%mhO%~l
ziimM->XsyujNHjzeraaIqO80rOqfNja$u4uS<JNA-2unk_OO4u(}J+DsY_@<4li_I
zil`d;E1S9mWFObO2Sp@|K)R_*h_u)atS-UG03MHqAj&0Lk$ngv!JbUE1W}4=dT4?u
zL722HL4>VSs(TPb(C%J+ABqSvSK7l<M1DK#zZZSBEk@A?IVEM=#Z(K$s%;ljfqGRb
zyO{dqO*=Je8Y_S6L(@WZw>so^wLVJ=(N=ZH@7>MX#c$o+qB2k&s10<hp8T%eEuw*V
zpnI$iaX|tpTUtmF=l77Ub8RtiC6wH}%y?KKPYtW6Y{*sUy)1jk)~a56Ya1<;z4m4Y
zriGB6OWy3TrS$l+2Q36Iy+S**5cU&(bLg<dkl#<lo^XGNAteOyJ=Be<m6*0RNo4)_
zVX<p1THBE888YRzhoZbhrp+D+Tck?$0ZJj-sW|kp*;)@(?1<7%rmEP}*e;erJ#g=0
zO6@_a9JyQZ@kd)eOXbK`#mArTS*sErnuUjE;h|Yr1{%7Ehc4owi+JcFzQ>ABvbdFp
z3st1YeQ19iXqv5ZPotQb4o%~5^asF3xhEye|9watq=~ItG>Xrh5A87&#Ou;FwKS>N
z!R4|KiDR)pCpXGH8EU;9TD=D!wQbdV5W2PQquzr(`&LPUsd;JS!0J8hF?UPjhzH%a
zmwHd)wpH)(sA~^U<jDW5_sHEMN@==_S+0s5U)_JjR5eEiZ#$R@x5#Cu0Wg7)oOdxb
z3{@Z8Ju+=~6_b|FQ%pH+DJB{EM~429p?_rP9~pHa85&E6#?qm&bZ9Ic8cRnXn(wt%
z^wDiI^&R3qZkzey+IpTB1E}fHo+L^+(zcEwY}NLmql!Nqy>;8HmPmXkI?5j{@&8H@
zNbg#PJ&?FM6@RjbAgu0g0kC__DtP(-Vpz_ib`MX(i*K_vNS4?#=CKLpFQRFoXyvY&
zE7&5{>fr%r<(#PA!z1~OPPKsvFYscXiPnf~jZZ}1Bs&k{KcagH%a>88HWUFlmw}-v
z6aqOnm*F-9CYMM5IhlVCzAg0YKR^EV#EJE9cPIAvHWnCim-)sCI3OPy`32-(N&K$T
zwL}V9&@<#hBRAmFcrfXn5Q)q99PerO8kPoknj+i^Q@T7=WOhrq5+t)ybWFb8O%RX`
z?AlkM0l(-a@tJr?5@6m0A%uTFx~uOWPycx;qL7Md<8MY;#)W^s{p*jX@BoBn#`>F-
z%nP{v`P=RHr@!~Jl9c=Z(O{zBxiMO=1TPKK+T9E&CcQ3#UkPnbVyGch3}0r8EK1)n
z*1S?wJ{VPu6|y{$iAW7d#c<E2n2jQS0<S1=(8O09T}FE1xR`o2raC6B#3g!=S6;^l
z4t%-NuQJ-uL^^-ESVH748{?8j#P6EESHt*rNAd4SM?24qEL%Bl0bOMh`Gggp>54QJ
zdCO8+NERaUx@YT!x)sb?*Xjn8-jNw7)9{k6j<uC!mu3djaA~UywOGwpugDQ3iM%Gu
z2{w7B70z9b^pa^itWa-QKFFR@pt37N1)eDmJq6_J8BBka;Tu`vLN4pQ$Y9A5iP9u~
z;vTT&Zjh!dSB+gpH(2xm<5-ez*a0(rwL66PIke{}iNSibXHKdr+r7P$f)7afk^)rV
zAg}yDm%@6cr5ov8rc0?DvoNlRR~|8X<lV1V9*C@!uDY-VZ@s@MyUCJ;IJy?QGEiDF
z<805NfPH@(#nE2$FC~_fv5_nOl+!$6KBwGkahijnfneU%d<V!Gr@5ef%fc2-&Hxlx
zqTy*>X@(KJK4~R@vkCQ1uDB?lX4qOm7!h(K{=V;+G=^j=`ZYaf%qRnXtb9x+tR5)w
zrO0b(<+zrBYyxzZKzWN(rj})7t5wyJ`?3-*DY<{hnOx9b_+NEWwpN<0$qq9QopyJF
z;@q+nXPEXBEl?^UjcDj&S`^p)`YdQ?WEUJqm>Tkm^)F=8OPR}5n@oHq3yb%mIt;#$
zqvR0N0is!S)AASOD|eC=8uixjbuV{qlgBQVk2Gq<Ik~$klVY2Ro8&=6yN)cg6KP7r
zPh5W?B>U*@7W%6heJ_!c4!Y{$=!vDfy4g=}bDI#3aa(uh$~W4VCRjsb;}qa2L>#82
z6aRAPA-B9E=S<3)#4Y5TmCfE!xxClzB)vx4(H|&xw8PYK+L=cczoOj9WI676Q1q@4
z)hkeHfGgC_Nvu5pKTc_N`!iG`!`)wh?4*B$4r?sg%S$yCW?b>oS#x8r$~3+%Ta@W)
z;0<r=ux59rr%tlk4gH=oBgDYBY^wbwRns9d%KL6q04xcEpt{+}TKD*XP_imF*xNnv
z5-|cu*YVzxqF0bvd>-x*wjlRQam}$@6$7PcC2Gi5)MPi!_6a!0q&<<YqR-HBlKy|m
zSpAm*CVRf#3#%ptk8i0aMXy)Iuf<unc4zX%X_wNwJG(T^b>wwZe37}_>|;xs<Dd{G
zUNVAHxGT{j$?8nBSh6QSw?cI)t`yHPSb;_R7rosNd2m`2`Q&DmZ4%#R!@U&gNtT>G
z@UFt0T$T7<MwhbjG7qVQWWaSVqhNm}amBw-+%vuA(#|g0c9p9WUXdd+y;f+qP&Cc7
zTDp`m#_w4Y?<b0^As#19Y{XX3jv;S-Y?UHc9#TY4p_<I1wbt}BXO;0PXCmt{WlfJ{
zmg&t?ndOR(DOV?Bvs7Okn=$w^H_H1b#b%J2IS)Pc6R{Z-@~viO5@GWk?jV1$7w$_<
zA=2%KX@|v|8CW_Kraf+AGq8jbQLaYl5ScGvv+NVVs<y(IEy5t3?jxQv5696VYriBT
z6anfNJ@;z5n?Ge?(n{w2Ag4y1m#oLBnBjC(<jUiGE{j!YmYBu%`aY4xf?$ddvOcuI
zh3s%i-XLd{sUwwI87Tvfy`O(L%~F%m)C$|*Y1-pP8B`P9UT3-l2RJhrw4M}nG;TZ1
zoCX{P_}@H+K$dSOL576H!bSr4NGIedH989y(7I0YPw8Eoe(4^8-3+8_K$4tHi`zCp
zBxks6EO|JVOrdzgLd<iSHs<un&gLu{7KYgdV{UtHI?G_S8T&S>@@Rjl70|n+jk-$l
zUNfR`GRxcadfKocl{Ld()lv;$w>I)8wN#bCoI&}|<;bb-)7w4lon8ip1b(Z3qD`*u
z{8}wXqVol5d#1%N;j^-}V8{46=P~r8Syy+Wg?l~|jReoUGCh@QgB_(~{<oAM4H_24
z!k!X;m_iNE8nY%jNC<zXJ5gHCtQh3^sXn<1P&7CH!#??;373(h8<=;~M;aT2(wPEy
zKfS;N%V_AOcwvtaq-V+wdfF%)Ln$pVfA0!MNU(U@fgOSJw!fQN50;+hPP)mBhy*>9
zaqmq@SLP1+0Z}Q^r?x_)0U9yUEX~qb&}<drqy^Dzx{in(S1^BLoPD?2k>O;E>*al=
z-Z7~*%JvjW4}Y~Dy{y5%=~@O`2{0%3tK#XXohrr2_DXHE<?{%+rPnn2wwk?@HsMNo
z%b1%%X_ZD#v`pHImw^V#F{K9ioj=`luH?ab8WX&kk`4~NgkE0L&3tPxsdYWui)bzI
zGfrMk@ambV$$)=>Yv_Ahx%Q6Yc`Fq<T&60aHIq{I^fy&C>v-$Fx0l|NR958nep6B(
z73G*PN-+o}R!2fjtE1XZ%}u~qC;AMJMR3sBp+2-n4F_jha4wB^O`UY$57Ha0kET1&
z#)%H-AGpf3Qf}uZu3~FqoO!9Bbld)iDwnANT7&-DuK<4#QCrT+L9S@iw+<{#cMsc;
z>D|1I9GrzZX~O~^4n#OOI?yno*_tXiLI_0mlgZLdIuX^J0ZSM{m_z4g^LsV@$ScvC
zG?5*YW^@x_zhAbIK?9Q*4Ld!40c56yV%`oGQ!<BYkpvsg?acsFusZq_C1M&an$8*b
zN1}Gmra6Di20{09KG8DSKA-I3`Q*upjsnp}jw%D<#Vd}Gk2PEup#^ZPlLkW9258|i
zws7dFtV<r-K|>jNHZ6$NyOCrFa&^G(y++Adkz*e_L|jI$s1aYyA>e|HgZ)x)N=k=i
zl>Y=cCAni-m@C1lHGUl<{a5TSi?zHE-2%P}tP+1#Fu=Oy=YUnhT4pV5HU$1McvaG_
z2aBzTpqbFgLKoXD;%e}UAf;YZUjSaY5ZJ^(=@Z~pX!#9V-omcDVHUSQH~Lf9l?X>n
zL9Vc?co=qyT|H3N^vJ{PsH)gPq!I@ik(#{8&!S$k94>S<%N@c+Ii{NE@hPW6Mk^~a
z^D2K=D9Sw_PdFusU6cf~Ht%h(rwdUMEd*vKa-d-8Scc@|+Gjb5ke+#G&-|ufBdZYs
zr-1=^lQ0}8h1~<CASTEgz0)9XG_FJ5=q{}0X%BsX+^||2-_~1OoS(2<nr#Pd8)n}>
ziYOSylwQVHqHW-r4bFZ|p;X4J_4H`lmZX3F;_q8<Rwmq#4DF(AR;Hf-Jd2kN#Ley^
zZdSEbWW72<+^lL5H><t~af=5td%;`{xM92(%JftoLfjIZvWK{>MYI<9nb2g_tUbS)
zwRPyRUa<0FRE-QZ;8*JZ1yqwZ9$;AW6frd~AHdWUm<Cf*!_KBVdx(tk?OsabAHaXq
z)ZmF#^(GsbngV`+sihG;rbaU#V`?Oq0XNI>X6DBNfMTl~+d8Xg^~%YS3!A`eX=HK9
z);tKWO-_U*+Is-$gytD|4FunS*JzFncr84g9O<#<i{Uj?nYFf4*@XwVzy=Qv9f2KC
zElYkAIBR|tHUB+Lz^f!oeonKCt0^3Z<yiObHFfJg3J@x0)f0XIWb3PQtp5KC3g=yy
zQS~kp1u-x=I5C%zN-h+a;*=^Tmp2PKlYg3Uu0P(sUy00m<x^>HY+%*_M{WwA8nJ+<
zD?xE(yjM(cz{_i+*;m9fU;XhKHR9!>MnG^UXM{{5WRRdCf>(HVNerXXnvt~DvA<XC
z!EhyDqyiyhi)wbXvP*cIH`3xdvLgTz4{;B@lLcyeA~V1?uWZUngxW^66$rCdWPijZ
z8I7&nK)bV}miYpgao{eM@zEN(Niq;3)*j8EjrA7qHVBoBwD{(bp^8fw%N;^w9h%l6
zWHmdv$LLhq`K`vgsu*;uThx*!Y&zR}SL|HmOtzu@keyM=1OxQ$ZKNI*W*c*bv{jkj
zeIU8dKDSXQDru!>cym<^q}^H?Ab&0F63>aBtovX|0d%__p5EQAzRY7yD6T!QyN4CM
z0BN{&fhom<1oriJcE7qAv1SdA_n3jA8HKgR^k%f&7?0FEx04Mz{XJHBXgIW&*e%I>
zEF!be1|u>AQxhRnFBgeg@o>1R@s)igUMBHKyhpxVi4;GJ?|CG~J#D6Ob$|6;_PZFz
zU3AY~XfX@SDVfHvqh7P$Pm-CVR;W8m3|!^h;m;cL#yIW(UJ&y}d(0Ye(wH~JGw5B6
z`I*aZG8uh1&Ovdf<iO;);@k+$HK<pLb0Y$`(vO8TDeaKPP7vJS3=dfTzfo{Yy;V6#
z`!H*Y@3J1ckX{Vg{7hWd{(lP^-DT%M2N;#211Tk%{|iG^UbCoARIv`;uzwjXOX{4g
zwT^-nqzn|ltMC8*bpONMFOnm`KLml^gUyx<o;{>TON9#16F#SBCpg+BFo|_cTQrv9
z!jY#GxVmu1R!E0~R;t<dnN~Fh1sb3MCYSY~bj3fdR1PA-vRcS}ZhsdGQxk<KQRRZ3
z2up^gM}#j>^hj6xT(%<AA8d8cRyS3fLE{glJSXguGLfis*pW91txEx+!Da4ogC*Zy
zTx`Y6YYL^cfGUs};K4aEM#ftuEh~+*WeHIg3}va@$MrqZ4}vCrB$qVYU_Qd6Ck^?d
z%qmT4o&LUi$Y7rk=zqGoCo~sQ2QCrQ1sLGuw1_LFCUTmE^pLlg-6b_`*4osrHwSL%
z$2T#Y3gwc3+VC2XA`3UHVKIt&G0A%KSoNThjCqWD#?sfQ1&g2&lUnv<%_4!02CZso
zvs%hqF%LG`ZMQI(#GNt626|F$e%$3DBVn6rWLd$C7NP5&Qh)GcyAkThVKq&Xq8USS
zX)Ns__^FiB<O-S~)OKpD;*KyCdO=xQ>83{1#_Rwmg;#Ys<>|=FRo)?bUK+P5c$al^
zyc#9994sgU2dnCsIob;?iHL`a1wuZuHYM2BZV5I%*@~$&X`-VTrT~l5EDW_7KKV}S
zn{ZOVQOIfE^?w)|<qUv{M`7v-sc-u3&o>!FFM?p3qcIdZkYC>%q@oP>1CgImsKa!$
zij#f^b$^(S#=%zTC26`oZjxG-LjK{SV87wZ_bgU%EaJ5Ff#FVx)sgDfPuSGO;NK@C
zaF3M<q~#)ynnzXXsQC6E$3&Fe;{>osWjQNp5Lm~|=zr4`r;s9NQ+&8PtjKIAhRU)E
z>vN0KVF#>LhqF4t9yYw-&cH0RcE#nF!)cuErqywVYeJ|!7+4knuoc1vyBFa8pa7Tv
zxKiqc0xV{mTwR(&t(ZL^L6w}9a^cfPP1lW@xhf3f=7CZ?mA!>hJk`}yivNngpEJEq
zm~)GL(SP*V(}=WpOwkUz0wG^8-D(qXK9(n)Zo5LaE;T{7lZbClWbKg&*Kq(pFnX>D
zSH}$MUunYC8ePMg_KbS$5cX^B1XHdGm}%v=JHd)rRDRP9x?K~Yb2{+l%*v1LlwUgE
z_Lsx>Tj`3@F3A>KO(tq;18H6Cx>IetG>k_cRex^iT+6E~S$p@y^EJA^t;H!2r8AzB
zDHpiK<o5hbqAXDhr_Ll(Q=BqM>5I1+O^agI5k_+z%q|YcArAxd=gKY$`$1*wD`gjj
zEtAo@`vSH!VLV!sC&)N5khA!2cxawKSw(sWPthX0oOaiK#&n%X=u*f`zw0F3w);$X
z+<#{ax2_4ncJg~pB7wb0QcuY&%4D0)%~p4x*%BLDJ5Bp4q(hM5vm{h1g_rGGi}r(`
zm}IM|bJObDE%CT%^e4(L@=r=U<j?elS?uy$zE59m3a`3eXWP&F$kIbr`|P@c@Pq<)
z7H^sPe?9TN>O8uMv&O#XC@a$9ykuYU$!1joa1bZ`<f=?_RA*9I*-aoPEGs?p=PD~w
zczmm0sjNsuuj=QpUYE~N<O!+@4ya8V{eLPdP}Q4LL#N{BrHOj%v73$7;{Y-%IZZkr
z6SC<KEzkLX9Kf}D$|v3Nj#2a<H@kB;mytj^6_+4HCKZ>?5<0#C?6-{*ItK^?s|+}o
zVJ86;0x>X`Vdy*)m%tl3y#j+jw|^Wug9HQ^z*#4kkw7{X0W+6D@GKLTFCjWAe@$=X
zHVnPzSNIP^6(v$1Krp~gnpvQ?7P%EY72Q4Uq5J<2r6fwWWqUH2uQUg*$F@X0l23Y~
zvU>PF=;7<b{`KLPgT=wCK;X^r`fULIp!r|W{L<0j0hF^ay#5&eeA3!DzV>{f*4Fan
z`FG$u?H(-zZUIg%*gblLr?IVhf39sTm0l*+-f{;SA#h6{yiWF>`QqLXcYnS9_Mip2
zR?(%O>DE8JygvN<&>S&Vk&A;PuRz1__m7A40fChBaWF<jz~SdV!?%aud#iP<?w-RI
zfWxMpBbtTHI7Ji&=d7~c*RW;Bf`lSDd*C}eh9-qYzQ;)$SKb?v`$5|Ie_e`b>*~h%
z<xD(_nP>TiHA5wK3K?bl=jZpKd-KQF7B=IIaxwOfz_00=Wuo=)pp0I+&e)JK65}Dh
zBfCPrxR~yU<xbKj;U@MC)3=IyEi0SMtqC_lS!g&?EC><<JwM@ddJ<F>1wW(9qY6(m
zj@X`3yZr=}p%6@p9ACkbe=QMRy#3>ArOOg)^vZ$jv0fu)Z$!y>QXUs0=TSi{>pY8-
zwN3mr)}WOO^#7v!w(H1=@jYm!^`iltCz{_-dP6${PiskOecYIqOfCm@Z)%Tf2XW0}
zvXuMocecH2mJ=n3#HtsI_1P3si}fgPsP%2*6=Ij31*71b;vW4Kf8tADnJmeE#)jf=
z%8MY$J_tg3-T|7Y#vZLWUN-Q2u;hZ{yNlYC*|PfO*o+azQM5|)<dWo(%mAvvaf_MI
zD;s%Ur$%LCi5UaI$+yXxEzhQ|7Sr&@(!K^3V4`bju7=Fr#LLRnWHM3znp_T1#b`Qb
zz8qX#=nVgYr^=M&e}b2Ow=}&|Qpq|MJl;uC39SN}E|FuXFNt$mV$M`ab~$#)j!C3J
zHV{`_`4EL99h467Qz&GI^^OYJjeMp;@^Yb&S$uQ)Zo0$MGS~%>q<V<dsh+%HjWQdm
zCoQTXNZ70Tl&?hx+mq01{FXI?FZQ)8wMKMB^iNuzq+D#}e`WW`xq6z|2xrSrjT&;)
z$S<PSVzu;;#yvnrc-RyobU~3Zz3r)ri`^<aLhKqlk}@mk)GARTO~pkvPok%1EbMxv
zr=9FEL>SYE>(lEf?;x;H8{dy0kvobl8pRV~f$jcNCxWs)41M5L59h0%te@z+=_KuD
zK`pXe0^9l=f7xb=jg1LczE(QQ6E?ecw4Hm@ReO#UXCp1}qa=L999C@S4dFq2&^fLP
zvN=|C^c@u*IZ=y+K{hXo#i0<N^72%PZwcN({Fqkv9*UW{coyFcvSH^ooID1}5Go&S
z-}C-_w+uwP1(3}6oTCNf`^d|i1!GFPV=OQBw^&}Rf6n=W&F&V<=a?(Y*VwXPDo281
zoAb1~ERH)SVXK?&p-$xTt?DG?Jvwg+&o7e@yfQXkwM?2F50o-#azb!9;alZ(VijdM
zSemz_b6;mwREnD<@XA#byQqA)*TJL>{Bg1mJ#td>zC4u0<pu|lyq-GtsRMqj%Hl~2
zapCDle^pt$&~S<VInE(Ftaq%kx{=SUvaCgA9epgjH(e!C(JVlw3apS}CZHwN$E33K
zdE?x%*vTTfu();ZDAYXR+)=DgDwaCl?E49|p0qr**1MfxT5!tSOAZzdPbNlqRmH4-
z9D7OkLdL|=FC$Oe-_V@!`A|rizg2DvQL{OBf6~F?+?hKaEb^pcvAWm6qTy|BXbGQ(
z`jR-8G>xHt|H0xM*$D@Wohu(DdB%eBpF{EHQ12*s^MTLQI|IDcvElaIed|=Q19G9J
zZIypbBif=fIT0(QT2Z{3ql1LkuW1;4nJC?UUB^h0gfB9hDg9Q+Jz3AEW$E-3Z;s=p
zf0_o7S2}%36$qYHG2m4-%?RQ4vklfww2VW?ftx$ADDSSXV}P71<L{(nfKE9*uKu(5
zUL4$asTZ@x^5&sKbWFgq-t``Qlzst#A~c^uz&fmVRIqO3GbJoqo(ym!(r&r{R#{#N
z=7VR|LcKWr)^DknjYp5>;S|-vc9>blT6;^iN~>PA5T>ey`XWtgo>i^$&1Ur`R-2Qm
za%@HU|AJJk;{7D1Wsa;JyarH_+~Ut|VOsXC(YY&ov&*G~EJM$VO%#aDk(&vy<BH$l
zJM_1|v%LHVkLhcdAxA0`1u-x=H8GbVKRG3rE{8x0m#LNlD1Y5sZBHD@5&q7v7%5*a
zAKcUR(p}xr$&&9JUo71zmT{yc79Rj_Y%V_5g2nRv`#jaN+kjao3oH<EK11znO?Opw
zy*yPjrdXARI1qmGz?2&$gJ^A(4qCWTX~=cDR)<PDl$hdBlMEJ<4YlI?WPZDpp>n&u
z(2n!tBC-0d8Glz<%}K+mk{X#n>a;eNqbyCFggY#^+fo{+)QL9_9G48{@;2d88Cb0P
zZmABOs9nu*;0(E)q=C&!BcWy`Y9#a|p04U3oSF<d4I;H|VKhi$f<Dky66B~19VJ!W
zsa$X^OA{*>kfq(@SIcC_g7qsmGUKjVr>$&Inp7~DSAVoISQ<1zASd&nHZ5l0t<_vH
zNDomO15P9<yO|4t!h)c1Em3p<4HOKCYXYifr*y#LO+Bk1{Y)uF+*5GGPU0q*+&2Pm
zrZ_2pLlvI(%mZvWT{RqaX}aKNHg3i&b;u<(UKlU?GEUN9M9oH(2V-rwRd|g|-cfb1
zG^qs_vVUxCF$fI{Ojm8NDs5xfuSEAD$O{ML(o{=B5svMmlQ=}EZ$nY2X9coA#kAvJ
z6G)evA|xu=oq!^4ElU;PgQd8-;KkMEf*@gNvc?<X`Vt$1n>tO~VjVb5vnU)7mMXqm
z)5+Iy+HJvZ2GH)1qLUbYA_)!$IbDJbwLrS}W`A`PR9_^+v><&Alctj|VelAVBDX*@
z-%^^vxNYMW!_P*$c0}3?$1bP4uG6lG)``b;%{=k=%2wJ6?!Fu+oqc7vVB6Qjh7lY)
z-+w>8`Tf&ue0F?1Uku|9^OJY8leZB$CGL%X8DEY+;sXTO8(+^37jKC<B=f*4L43Q!
z0e^>x=+jVSZ)e;+J6QO7KOY_cdwkYcKU^HmkH_8dZ?9ka-|s#w7N1W4G>-3&79Y+I
z!r}b4@z=B8K0SFi`#k=#zxeRiPm7nwi`mJ)_77*_*N@}D$N9ndnF7Z^<LSxaxW8DO
z937l3W~cc24?lCP@yXAOBm3jir#txnl7C*8q*|{eO&=yDr3S;HbA+-;PE;izO(I{#
zkaOJ|pAA@kcl>GoW<EY2zRO6MX0ak6lBNXrOpxN1LNGQ}Kg1#-B6W#cP|wg67LWOc
zLTrA*lr1!uT%ItQge3ZaM~p!6Bc`(Mh&NX(OK631saN*1wukgwb+uxWbG4qo^?$-^
z(B$>__kaA;lSnA2L7wQ2&pv+K`xcvlEukvZfMmM?r{)qgZnUV71!TKTSShDa;za?5
z@yGec`N{4l7-bN@VHjV*5VMoxfv1mu{<8RKcd@^i4K!VNF+W~llNaKd6XnQvd4Y-v
z2qM!5uks1|;KS3W<ExYT;qGklc7J^J{KfcY_GPhm!AD;0zn?({9L(|JwAInr)_6TT
zou8c?&Q6ywz_0x>dv~<|!~9EgrHq<jmG`j13GQ`Kc+$AJ-RTkxH4%2+PH@P@pU}{X
z(oUsL<xW|rN~h|I%CS?_slG?<d*r@HZhP$AGW%G_j%4r~h)_xl5$QhRgnwyG0?d*W
zSOm4sgp#V{+b2LFf^SxUDG6Rffe=H&`708lFkcY@^$%bR*sKWUh!{Yv-+~m%0RRK&
zE`^MAp_~Gvsw#Y3rQDAxP_>bZ5v~eTAce6}Fu0c{<YZ1)g>aA|IjcrR4krM@EeKIK
zI67Skfpr5M`&%*q@@uMv2!CBjW=^Jht9qn_1Gtd471!w@6Swz2Pw=zo79fWR;tYAs
z<1PW?OQj&le1d?(>E93JfV;xCfpbd6Z!}stVsm`1@cFdH*A3qo{vcDZ0h`(ueBJ$t
z)f+x>ur(y4fz}p!*hjt}U+)M|QxJ#lD+K7y9@CwJFPIX@le@FCD1SiU1LA8hx^Q-c
z@iv?txSC;I_jPZ_;cc<?M5rzI<?2gA7bF!wS1u3*-5(rX?#t03f}x(dpR1QS4iViv
z+#Ifct#sOOSgsw1q1#=D6fbbBkXi<=+JH?-g8&F^z>3-!d`5KF*w5QC;(B`^Sm-Jc
zw0bijSmXi_%pKt64u9~hgEZ?v&^n);^Of5}xK$2=^9CP8EK^u61V5uYPhr^slbufp
zaz?!_51GMo%V2iP_Awz{wXxw+uY`UQAmAYtE9#1sTPh}obnLZmHPR`lIW;|`VwD_V
zMY^40IP%M6y;G2;O%tx$n6_=(wr$(C?S9+Fw5M&`Hl}Ucw%vRFZ^d3GD<Y%L>gcJ8
ztbFp$`x-`-`Z0<wAG(a1qeCMJLG>ZljkTPe((#w<E;}0ryqA%)DwcI<7Z}J{N^e>U
zyF}V)c?33f3)=Q&{FX?_;4nf9Ml>S<o*fM(c0*Aow0`tD^QYmq9h7bQ48VL=`g{!Z
z`-SbGcrkXeYq0eGthonR@vb|PY3|{3=Bj-vUU{0o@IG_ve(cbH4=8;ER=5JqS%=_l
zM)0&IxLxc4NL=|W_WW;gTiwHcno_337#9@YU_j6Y7u9|}V}G{|W5LLRV00}r4K~zp
z0K)Yv6@6LIi%KTs*VTbif~D)@Fk7`vbk=h%!=82XOi=h9sz}X)lwj=?T5(+)buce}
zjZs}hgd1t#BT{U27M@UuN!>|+3>mv1tZ0A=paKjAy~fkpD}cB0ypub;`ZbVqW;EH4
zV9ph$3mAe4Pb0IXT}2XN9d#1lHIFS5X857IszDx8L%Qb4M<?zMbcyj05=I3VRdK>_
zh7bzYxDSn@X3Y-eMfAhhK{?F8#{}IV@cE&eU!yyQ6JP9u5FbW_=}Lz^2xD*OAW8@N
zq61o7A?1<u7MI52fjtL-l#Hp$m6i8#n&l)J`w2E`pVfYm=Mz>vce}rD3A{N!a!o31
z$A3VSl2|?dp|-+_m!N<lq65ok#*6n3)*^aS)1V4_ZHY;qt3qkCWpH^Fk$R;2O8zvW
zLeoj=iAnOIb@nm~F4LLJPo^!~Jw@!jWB^2(!c=MG&L*~3lGl<!W+#%Nb!?GnKfgQe
zA<?bmJ^qfjG_jsxmgx#f-l(MhapQE0JlGg4Z?@eIAVF{#iZYaLLZV?(WLWwT?Oq9M
zRIM4%8s`XZq!HcPwbct!(Jr_?@?oW^0LXml6v#W37OAY4)A42&Wq@`B9O?cWWdp4D
zaIHw2OMwBDhyzTLEW5XKWoU<d<Ki2oCAkEAXOHZ54~p%Qmcd~(M&u-&(4VGbGbMxW
zLiezf&FSMO+RjPtlRKQ$So4P4Tn?mTh3LB)z^tT;|C;FlnpANp(`_#pkVI=d>~)@7
zi|MB^zwF9M|Lt1b?mKNEbI_sL0kqrAy@nw_flQ7G?f+<v*l`B2dzoF(w2&<$0kj*S
zEQFJ22Quplq~Q5!*8@^mo)vJQ1ViM5t}!w7a9|?!hlK|?289!O(5)SI=<8uTOb7AS
z93|MZxKnt{vd>Q=dAt+jaza|dqcEo+5Vga_9BJD>FvTu#)&<Zs(d=_6EXw(qRYU$f
zx4!UD7ohHEaG<xC=|15XjppT8TFqt!cc%9ym8#yjk~+q|H0&B)5x`e%PbeRj7RAG=
z=Q+_H@m7#7As%dT;OsMWJN4XV@ImZLpsQb!jQHZK(~|wM!&m<Sjx#+7`9D8OnE&9R
zq{TCH_W(S(zMq~y0!c0mrblk>?6S@K@QyYl+d>o>ET-8eD-Kv7XwHUI-oX^ydT@6C
zt~MQ~HDYwyU=3E)aD$j%&|<q5?{`rDq`Dje?+l{}I^!|>9a8Y4fOth8c;foRSj)oh
zG15Ur_EB-y6C;d%b_5oq&_ePs{y4YHNzzx);sRo+FQV8TbPKCs#O1|}`o%kkHZE>2
zT3ZKu!z-A||I`~b46uMne|?+~^~{2JIvOP+YtAp!0{(%k1xB>yZV&k092*c?31ut@
ztkQuw>XlhwJem}0H*`>8R;cDpJ1Gk=;$CWpyUGaQaF+yn{bdJ}9BS3XlwtlUP}0#2
z4ggD`f;dQ;ite~bgYucDL#MMP?OLC5Th7@Yo{Y6?tzo<1u6YTuFR~ELko{=QnWt{-
z-kHmH0$;MPNFlwWKkc*RW4=kuQ-|lTmg9;iWa4_3A-~LyMHki%CQCcW!QJ3*hgB5X
z6>w4iTI=@iM%7GKVz5vi_VUT~gfy*;DFaTJB(T0%rvi1;19a6YSZmVozKGj@xZe5)
z2H-(toBJ}PW<3`ZhP7a)4{5Av%~^7^?5Sgr0aDshWu~`QSu`?EzS2sTzkftiY)fcv
z{<tSP6#sBLj44F5no1|q%U2-$P#;YszC-T_%kMAT(~yC=Ox_1JAG*}T-}T9dqX7yn
zf~G=(ZRL;+g2pSt+L%b#7?P2W_p$Gy3SYbI_@;md(VF!W^Y67U6X+t^h=EB>hXmWY
zn5j(;_Rm^Zu1IY`4g#15-|s;pZi+r+UdfA1?5UhutfI<&#>ktc#3tixeZJkbAh-gz
zO2EPPD!K{MMP62`W}-R;Ydku<qXC4TCmt86+`uc759+&9n7#Iaf9+A?7c!%KWje45
z43LW^AGxcO3zv{=-Pr(K-C;sCyS&~%B};g`hcHkA`Yo5|Y#AzTPD=P#FZOE1WK<I7
z&`lO~<i9DRoz+$KV)b(+L7Wqw(v;rHdW???z=0G#bWb(14B_frf9HX<EB@=tAO6)J
z`FA1J_IaL}iD)Nd?u$86R^BVm8RBWdGHlDMx}^$b3`t%Yq)T{R&N!$Q^u|5pB*x;H
zu0oRUL4eH!rm^+Pv>il(K66$piVBrW)ZnUZTBa9LL?y4N#Got1D>J8TyeOFwaL%AY
zO!(XA`4S2^v=j<iPqL7`=m5C$8O5+D&8TO2#KFS*miW;jVxNav7{67QbxU`aOo~rs
zqEY4Sz*nS$rKA0o+8v9kK}iI?{-!f8>$Gk3naP_}Iy}$2)(3~;&#3P*9A!~;*Hzzf
zw{R72ENeHg_w!|3@hg;My~a5kHOexkiWR$lCVp$T7cydhkr|csb_4R4Hw8t$q`sQs
ze1u2*s}{$HkHWEKk6>B<G_ZC3r6_Mn^z(PY(xM^j7szU!q41!(B^5<BkajK<pWwVg
zS87UsCaP#^Ac0`yE}V@#^x$JXjBrq<3QAytp?g$w?FcqhPOteIE7s5+^iZ!WYEYze
zZmL8D)kd(1KT=6Dr2{w!dce}zPvP)UcTTts7CNd5*%-b_II2nw76DM>GM05VmovI>
z9MDv2K(wReYb5rnA9Dip`5_$4W@W)&tgd$O^0Y?mg`Z|=hj^ZAihi#QxRuZvU1t1V
z%B9jIJBKF@<6|$%+fC8%#5fu2ughaiE=}(uDdtVt$HrgIL<JmO_$L-Vlo%{sbuL|Q
zC1M$}KL8mQUNN-qHD_crcJ8ZsOa`jIHy!aaehWT^u_s0%@&{}y|2oe8GDz;Pl-}X@
zV4j3wg1n{7LdTDCFye<Q12YBt`NMooj*2eMOwoywA3`tH-Ef%&qnn<_VaA|~t-`KF
z%6&F2&#G)^OAa_})nm#&`!M!y6l4b$UZ5xS5_}`XLshlcJ<F*66Va71apH@cQ<TAT
zRy$^#-hiXQzA!%RJWj9X<75LR;kD_hNw-|7X2s5AzDiDQ!KPJNSzHG+>s;f_LHFvg
z)j%l5za>>|!KxF=KdqmcTv*|}x-{Q|dcm@VySAOkv;}C9E~VhfDYMPxNa)`7In+Qi
zF3?)tlm8e<V+4c!<-|naQ@=qDWazh1sPuPp^C}P0dv-b1l!uL^Wt9^F1DY9@6{KI}
zM4jK3(}^XbJt058c7FRQ7-f4X8GFI@{c^Zf!Jr`S!Y7cRIqPn7l9+s#Cq)2@GJd|5
z9(!!&RSpo-`I{dNg&6iTki5l&@zQV5N#G`g7<zm>uzto_o=9`oShp){VP(C5$9a^A
zJD#H7Ya-eH`8~zXU$lVlCwP7^1ZOKgPHd)qdtts*wbb3Z(z&tqLLE9v%i`T_r7KF@
zFua0Ix-o92-pUTL;anXINmYNp$=cU7T>R<LdrWZUTLY>ywJBB0uac^>|EASUFyr29
z6U_5>U;Xo7fzNH-XI>DwksNnY5#>sz7hej4K-;PlXvKtcxc0X|v?q5Ok_H9}DAT|8
z!X6Y8DhTVpP6ReEIpUGFOgu3MV$TE3J3`f8GmW%8Zf+KU(t%+pNu9Ol2Rl&j?@tU;
z1#E2E07mSv)|<f_9jcT)ix~vzayEff0Ums-7pdR_G{bw)P6c!$o*K3A262`3V&AoF
ziBVMC+NSAo$L&CZZMcsu0*79EQ}gba6a%>dE!-HfeqA&FmqhYT{Wy4e>1fi~HG7Xk
zvC2t}-6u>y(dYv$R|PA<3hK9AQCpD)+M|_u@~+El?2*FgI<XFGs4S9R4%-!uamKGb
zJ3X6m+EYFoSOl->Y4tYW``0jXiZ>^{q6E{#Ac`iOqFK|dy*bbFFoLR$1&5xTxwiSB
z7y{TpLj)@CyfBf1A*P3=^|}|uXL+N5aVT2PWukbQ4^y1^GoL!W-xNJGsJ_~{lwiGb
z&(-fwJd|N&&~os2V~YIry)YzaVWESb$<sba2MK#7$qO;4bLw7cyplFF$^G)zD}_|}
zMOW}?y1d-P079cGf@EZT$Je`+Te}AxfHg)z!xZ}*P7#x+iUVWND4e)3Vq0wQ96SM&
zu=499JVvaecEfOyR<e23thfdf2%dbn_8dD|$7zUBvKjmQ(^Ka3X2zpve_2s*M@8E;
zQ@Cq#VOO;fu5u&u1`}z8ZB&*DE$;mR>lx~J;~I+*pehD^d`@<M#4))UFBq=T$r(|f
zwXQsYnN2-y34?gp+uYtf&Rf}tnua6C)~3x@xX5n$;bXFes-nwRoNX|l4SI|`S7dtX
zOpa4Z&RxiTI3R#on%9NZcr&p^+8jIAgN$I-hRNpwWQ|F==kNO&+rqOK+~nL*mJ|)Q
z9!~l&z)V86Qg;|-=~pj2okT1T#wBl#@c4N9WZ=>&)|J^~)!4()DAl>{s*Cxvwi@Yu
z=btczkkj^0eFYWVAk%Xz?^2lpO<9<+Z5b<~4eoPIF-4d9d(Oz3_8egcFk<>e6`P{p
z%06;b!`jr#YQ@vsMeVO|wtDVr_KJof&)vRq06g}mnimP6^{mBGeEx^C3zDM@$~Hxa
zoaQE#`WnxGVv6MD7TRCY%%pCr%Z-<I?N|EhCbmg&_(si8#hMm>r|C5*UL0N`REC`k
zF3Kfrs4v;l>mIai{t4b_-7TYQD@huJE^Ih6Orm+i)L43DDN=$SR|%Qb_zhszm(tlg
z0i@%e4USdK;HkzeB*L3lHj$hUz%?9s=*lU3>kT!DoH9tDFjM6lWZWODBTr!G<aPEk
zXq(U!3J+Gbko`POy=5stbp)0Gq~tmis5bErp3rU0z+RRLjKZ~$jiR`d5;wjjj+o;6
zE4``~^6d`ZlIW!vUjJ_CFq%)JPn0b?0P{n$6Z8!IXxXe2P;S}<?@fqH$Ie|<cB}Xb
z`c(6S{GA`?l@`}{205wwKA&Rk5=%J9a6tOq`-7T)d9HXIacZ>*E(Hd*T&pYpQ$6oM
zXG)N1JEmyhvg^<{g4#D;pEPKEB0rFN)D+QIhetDtL}U$shLfl;F(_^fA`gWEfIxJn
z<WaSW0=4y7;3C1=!2PPJaEFV{CFwtZ*QRY9<Hlf15F#$^Wv~CJlBYCtv5=?q_chkC
zsEu@KWE&GLI?gasq4n!*#vj88!WK&Ks_Px7<(FQeN+r8+5>MPBWV_l0ikXmPINk$7
zAMUI^;Xp7W#U_EKqs<OpcO{E7z)$|~7#6MarA7u{!$!Tlv1~6D0sbl$aDd_ruozyY
zLa?YxkHatG+E6lGE@kwE_{D*XICD{oc|PW$^vXG^FHm8h)(m-H1%hPC`)$=ay}_k{
zuL`9`KZ_d;l<a4JwrYou4$<~rgKFH74>fL=-f@SePvob?+AdvGog0f*oEnO2x}VOM
zN`dk!NWY=Z@6Iv;EZ!fmBZ`yO{}npuY5m9`C~4EFyp(DCg50TqkdM9EmmLCA<FARa
zwrQ9}?RnfCQzb@^^qas+M*-e@Iz$Kn<EV!zIvh*j-*1+%8H%1iR3dk~o_Q!EIzelo
zv8Y9^O34CwC!J8JudIiIu~!O=9RPJoo@R3p5W1!U*>tzL2b{8cDniBJ?@oU5arQg>
z!&@ox^IR~rp&NF9-aBHI>1PHua^XWy!5p;c3gf$PfuX|&&0rsvSzWK_2ed-}(BAX1
zua6(#^(Eg@BAD@z=RnC*WNSwNe*mE@hVX)?I#grz>$CLZ`Bp-P?M=}qeh<-(C^Dwp
zHeU+U9Vk)Sl2X0q$x}Ykma_%pA1K?T0VyYOcT^%pK%f`kF_l#7Am&adMCpDVhV!J;
zR>4TcBaBJX%wsqJgL9GW36SF@czLi;r0fmnn-BUlg~(E}aDSQX@k)QlPh>)8)^jIi
zw)M~e!9<wz@Ody-5TE3lRD`CI{S@V&sJ)m=E(^tE?6{Ny-ul7Mo%|z_GhTa~Cvqr0
zW*0@j;PQ{Yn%MlH>*Bz1haul5-3uXRbRrYE6ba6SK*>k~_q;qi!iAR=!T<s8eUi@m
z_1=;NEd+J3^%z3*IsGPGM#r?W+H|vMaJTkAM3X3|mzExy3_;jcl1hY|88B6tI}hB?
z1lb@hVoZwr&x)wQf}GN?3*S%Q;3%`+%g6fLmTUsxns_jxt%R!L&o&12?%*ytUs(Eg
zlB}ghN1id3J*u(77rRI>>cd5jhPB}0Bb~)O8zDxK%o`)|vH;+8#}3_stR+bb%Z1vK
zmE1A14)nJNJ`DFRyKB*pbn?3a9Zt*C3~Xh4fSyd`Xh1<s%dTf?7KE=J=b%(L8VO!I
zf~^n`+CbMx@QY#AQz+(@A6%bsg)ly?AbJ;wLj+SjU<FR35-1pheS#w&hTcN=$z*JO
zhOAz~@rZN6EP`sA&a8@uZ|+<whMnQWwy+vWvYI-~(eiw_E3nx`*rmHHQWvDU3;pz(
zhG|32lcT;wLA?l8*I0e$AGF~`y@be8pFs%Vshrpaa+Lo(pxqd7uWs{;nVrm<inLXa
z7Xxp@#`&7rJF9@@T0Iv%>TOHOG(H+Vz!bPcrFOLU7`g`uRYCB?iIP4HJ&?lL;{fp*
zl9z(Qm;Kfj^Bs&=0M0Y$WuV<A;Y6MKVzx2gS>h5(-sy}@!%!j5y1|p;GLki?wHOy*
zHa~wkYELtw%?2XL9yBaohcHmSUQgGKWamf?N=nb&8deT;NhxwLx4$^yrawOA>Ztw~
zzhRLUlT!S|qR<?nM>!-*!%oHv?2Q_R$a@_Rr~jr(bsnnHFBEMPi0wtvL=xyVbw7hu
z>)ShA{`t=z7Qb*U-lQCauGYAsZZlOtQBB`ql-?ZGmU0IZvHC8U<x&m^7K2a<;dH-|
z7;YuLwlxyM#J@xa!)jDb$hODyFMmyE^@^a|lXU>ax9Vr{O7cXdVhs;ACoZARNU7H@
z<0Q_vEys{`%`<y-bjSW|F1EaPPBOQ*!_VU<m3G89&bI}CKQhGU_jWY=oi71UBP2<6
zB#Epv#Kmel#K>cNch)Dw`{{^~wJ^$8KN-v0S!r+b|4}+UuBqN}h@y=Li?@j3)9)=|
z*wtSsOh2uh?vK6=6?D|M_xLM6w~-=0H`VPx$xeT|Sw74GZRR(!CARPV0<RO}iFfOF
z)<BLA6GzhIUA^MMbzOf^<E;cp^!Ky!#c+2R7xz(Lk$WE?HB<rBpFshivF&p1a$ln9
zMhRFw_bk*Xm>>96Tjjw33(}Mt;{maVj41P33rhL+7cM(^OnL3_2-2kb$RDc!{S(`k
zf#OvOolEUQ!V}8iVWHsc-|KTHjt<7`llhiy``|vKvE`b}Q~t4Vkpl;i8NdR9*8IH6
zF}ux*AS7zUH=V6YjCqbZiQY-skT>)#Om4J!`Hy2jTW{)sB$L0wC@&|!hAuM>K_l<g
z6^~$fwg!>Mca>;9XPXtPHQ1)zO0n(q&qm{19Mb;>RaA=~Z|lbfdcLrPWe{<MSefnK
zZ<aBnig8k%1;$BNXmlIki;3(lBJvtI`1_l0=kiq03SBva2UZk&90#0uDvk!6SbFVs
zM<&TtW79EL)bC?UvHNK>pi41JEJI})UYQ;+Q4-v<hNNTHeeIx0)V1x70+*d?@Kb<X
zc4coh6BjE@!wtnF6_nN8e_EFBS~{oXGM-Xc7^Yo+hvr+N`5<nqd1Ze)&jBGQce_rM
zMq;mwlOYpKNIz}pBww)QkR@Z?5gLUjqJGbz>Lho|PF-V_qvX{6$|7N>4iyI2Xq)+2
zV={@~3T{EFs};Aak7!kC(d?TjveSN6jym8zhi=i$9&Ppz7yAJ&)~@UNUjqD3_mt0o
zkD7*%2}hAuCizbgj3ETX`*U${qrjW5e~UVr>`q}vN@5G0a}OKlUwpUnrsXW*5p|sm
z>FBRZCwrzo+J?f%EBxD>+%u%Si=2{YlO>MJ3Y=0j^D*raale*$TIx2LIxTnhhAdXu
zs_3clK@wrmq>yj`*0r4V067<eRxS{aYUV+$#5kD5ptD{Apet<khmo^*h8*(!w@4Wq
zD9>%W0Tj$oS|7$<b7^1Wp4Mz`z)kT$r=&*D8&>Ev?+9t!c}|j0Nli&tsAH)tbygt6
zI|ScSkW&HA9FAe8w06JofXb+XXuKA#i>OXklGR3VU7na!BGgFY_LB3O*4v-TlM_bq
zv^gyg5_q2u01u80itS8hYmma+d}>-aHUl1WcuU$29(JuDBaIsZ(W_LcegOlVFElsP
z3jQ!wgfShk170<FwVXoN<lN}z_A5`bp%T~+-i#%f%4}hsA)$mU(g5Shq2Q`HkOU>v
zzp)zcKixH};cZ+Zvk;rFPpfRsrow*sWb?r<L(=}20HIjJL(Wp_1hhQ%ZfUZHjXH~2
z|HVu|bjpMzG|RL;@j=d9N++jkk1dd;mS=N`Uv$+k6RZ_ZL51mGCY1FDp2m`mbym<q
zm6&{0^KHM)vO>YZy#v4*J#d4>9p;b{3g|=Yx_(T}&57miX+!(+V72-OR0JqqktFx!
zJC#fo0elrhyesBC6}*o=e<yjxTN`;&_Dt<lKdCGd>7$_)0Ln3({9jLLesi1?Aq#k!
z0|n%I`#E(qaRki|unbf38WK%DL8*l4zgPEKm^dRC?J}OsF?JT$Q(S{t5;5(}vhwHE
z7i!&gMN8y%a2ab(G17`+Tb9oP^5qsU0(GY&06YFt?`JpL0y*D2WpxFzAvj*|F9ZtT
zp}&TkoG8{)PLFbXv;6_@5(rK5E+F5!HpSYXY%1e9$ztG#7rlD-C%6VNI3!vtIyO~_
zlx*ikqm99;9L@;yw*KqizwX!+{Hq&C>8PF=`AlCkdyl<Rgld0%!Qwg+IV9TttD!8c
z0KvFr_wdol){Z_eDut{fu6#oz*34g5unDHj5v*H<GOEWf#Bu%OW!~0}!gDRO6?uEk
z0x6RIFro-n!o?ga8Pi_3-bmIDq?mOFDxiXY6$N{tk6#fze3$;+3uhvDAIoz-r5`Zf
zMs%beV;5EzD7yXg(L*r=rEUp22_$NCfQQVKvOW*cR8Fggj6_bV?s`mFaG5wOQ&#*A
z=gYR~%q``$;I%c%H?7m!t#*XZqM4bcMD(ztr}z;@C<!T}ly!Sp?+iWbnfy#IUn!w+
zUN8p>F+(})%0k3RVhS1JXZ2Rwn;HG?rA^=O+UZxToLA-zXb+OzrZH(6DRcK$Kn(dc
zWua7!x93ah9IX_hD?;dgnMGtVFG~i%KNE(qO?SM%bHZjddF8gQE3EQ}%~@_>HPGiO
zwLKG4nU>_=boO`#GlUxuOJ!ALenS`sjSb5&0wOvLL~xKYZvho{&BJ+APESr4QGfM?
zl^(l^Olo}^rbN%-5a)JmqXx<W;ApNH_1SS-<>#?w*A`<5kl}Y#(rqr6-aYi&grQx`
zEVU_Dy7GOE$Ers4uz?KLt7ewQtLD#f`#{xUyHluE4$UO(4BP=s>A@wRUJYA*)k4f8
zHql~5GN|}D+X{q8$E#jCtM>RjrNJrb9EEE1GJu7ngf8P^&3Jedrm*k~z`EbD0$WJQ
z(0QaKg_KsKoOM%8@RlsAg{iK^LP}r*&9?uF$%eQ&Hy!(BojX?P667g@7NXR&73n_=
zvCK;vYOp@eV~y|UqUJ1Pg}j*H#q{2asr3G<DlY8d7+u{lMdgmJbaFVkt$36fqO83S
z6Z7W`L;_abA>-C5gqV#5V94`sP|Lj$<!FI+KG#X#ns}M*Mw(3OT>Zg~+>Us#+xC0$
zhaNR`mUDU8o8!z^jP*qMT8EmqW!tUvI%J)>a1Yq66#ia#t32~Av;&2DUQk~&;N72^
zRHzSs)&s*>*?y}~r(x)cSk~MKrX1Dins5`d1Q>k0_G3T#DBgYmFvY#`s4J>tvEZBW
z;jLlZ+I9=$)&f5td~Ul%O(cva)mke7c0c?JUx@jdy;I48%EPB8Iob*X<$SPD1k`Et
zxsIakhT=YPAP{(fH9AYQn>p(jbpH5;xXhaYUh7s9;iX#db9OPL&>o5;lOUJtDb_>h
zjNFh`gQEXImne1$AR6(w4stFw)V=uKHpJ@w)lUB+?)kMwk@<+NXJPc9#AnWg9&;C+
zbDdk3&KZ!Zt3Jgzwp7hq5tGbj*X58ZA<6NEf5Jx@QZgz;#pSp^S6yCgiKEBcvqYqE
z?hwcXT*Ko)mZWZ(Yw96C=>W;Ia)M7)P*v5>rTGe_1+NJWpg5}Am*)~yBn29IK0*Yp
zQ-nvNq&WJ``XbIH8n8+gUc{ltMar+oB_ae|9F$Xc15?>b{zWlxzB-?6PxqU;E+g9Y
zyE#j)vm_GZhdt&24=C*~{ankcaykje84c-EzMlakE{+7#(x+@D^*xZjv29vG^VgtJ
zl(6yAfzo~eKnAdy&zI?}SUk0Kz3z0Xo4++Z1k@90?SVhtl00sx;~|Hu-cj>5{aWM7
zv_+r+Z62>1lk8S5Na_#%{gAKVW8?Ak_xhJDU3Gfg*`cf50k~-tT-BtLh;OY6Zkt+h
zaOKW6zL=Ez#xP2ZOm{<?XN0<~<J*nfq2>&{k5`utfKe^#IyZKH4bLpXN`1(%xg_Il
z&SXi`9BqkO2QB)Md)b8tY<QIEI{7hY&Nxx+ok1d`HZyA;N6jdOb|qI~Y2dFvaV{O4
z3C=~70yv%|g%`)Gd1JF+Zh6H#5;!TWY_~jRfxV3^+gV$FNdVPQj>?Nq-GM$B<Z;!x
z25`?90PoeQHiS3lC|Gn@R{+5&A9fQ060c}DxOrYunrArptyob-opWzRcY<r#?o})f
z<bCbS^e!Xflz2qZwM-9hQ_==5Nf)+k%W0KER2#Zta2;f62y4W$YPWcz%@qHRzR$8H
z*ptBy#q2r*%GunI@TAlf_k#|9%5pY9wfV{&kWjJ3UK=G*%a+?=^6_?tsX2*?gv+R!
z**T0Ud58c%{;c80FN7p}!n}Z@v$sz*SCCpB7WQ={<-Rc1f8ylzz3rYlLf2B!y7Z9B
zUd6H|$qj2K&}j;Z5|<p$V!FJ}d>wWWnD}`j?Bp^kv?wIiz(rCqLk4bJ;#?Ynt2fsN
z5OQZ~ERQK3PqRl`g=}c~g5J5b4`pzLKGkS9@j>osC09e(%j_!u*#>Dp%V(hAAhH-H
zV^~eyXt`{havhfG9O$}9mAGfcFuWMqdQ}%q+$1A7{Khy#-g@N~PvQoCZYxc;ufvn$
zF>|y1Yrk#98*+~)VQU_)nc-pmU<Zs1NR_jGz|p#1Id*|ktDo%ktB%K6leeJ8HP!kN
z?7<;*3BMMon}CZ-I^2cPvlld+Zu|L5|8g~>OLh0V;iO;T+|_1GaWr@F6g+CChzY<&
z(6Vvm{_P_E5n%2C<C(v*CI2d1xuKj|xw*o#MVAG^*1=;cuybM$?3X{amEhM%I<Xb^
z6BjS#p0OF$_5p`?jXfE>zVU?Y@zU~P;B5T^N?{Y!@ZZ}Nor#H=kde^A*czIb_y31(
zv9P7(|Ica<N0s{zfZR|2`|cl+h%|ruR#jUQBR7i!q`usfn15uqRj3Tzsp)K^=WvP6
zxFlQJ0cO%*8`Rcnw+QIUd=<LBxx#mnM~E}e7zn8#o9K_u07$Rjo1N}JvEll5H#x1_
zzzf=s_l1vD?-kYaIQX8SKDHA>#0SXXG`MP=uz0Valc=&BAXgM#rpP_ED{dQT>eftk
zy+X(+z+(Np!OT>4%a@e*;<HQHC(|7{ygqEXZI1iwg$HS}IRQ&qXG&e$)femSZ#^E1
zh8D9?(rJ4{sfsZ8(rKFEHuRM&Cx&ogXba4*t7)(q>HMVAxW;>WhiFVc^?tsU?$Tc0
zqZ~ga#S<rBfE@|m0m#>qp@yUcZ&B;PB@AzoSFJCD7#Ad}_hvOJxPXw(*CRwl^wZgn
zKVMNRyE5(Kw5I2KnNOk|GV1%Qn+C&emu<{NE6E5pgm9%=9;$_@I7v@H?Iq?~m&Gj6
z+zUu72)Om4<(X%H`Kf8*qLM&lAR_CzZlmEV9_>&ez%P`-5;5d)BFYlDHF-gr@(A1}
zXk1aKM7>z~NK#GaRk!DW(k^1X$P&JMHvyV@vT{oW2iOV@CT?hJ;Yk?L@Twu_3i<bo
zQ%?^FT?`vbvXGuov6_!`Ln$FnwqyxZ#{ib9I{t8R(g(zMyY$eRJsYJF@8#{JIU{3p
z&rbO=KwGj9Lpt*ut4XgyO%GIn&onH`^>mzW=1i$v0k^C{y_iqvb)8aWa%=56)@k4(
zMW$39IJ-lpw2t!cM2D+-kz`Ek+5#2f4oh}RsdE0I4`-@QR%Geo*e{8ZM=3yg<wsAE
zoqGgFm6e^Ypmj>uS&`B0%W_*PQ{>A#bUXJWpkan(j15sta^1d2^Fg^%;N$szEx@#;
zKc<wu>7{0^*7>{b$WbgcHb#yo$X{jB33uitwUrqX?kh&@SUeRiaz9m_Mrk|_VVSM_
zd!iFgM^3V`APX9_<?K!Lf@43Fbmiu&CdYEN3mM%t*u9Vwd{R^e?zw$JbA>_;8ag%_
zfX1Gbr0=O=#K?`5a-Kz7T;u`yBSGu2@JtRZ*fZ%v;<wqfT+KbX0jcN}pD*<%e^e*I
zrr@u1K25)H{_j2JwVa;k(sb$kjt5Y;wY6y>(F1vX67_6-kIp0M!4Y}%eM<+8W;m^O
zv1P{VX3EV?@;!O9D}!`oo&&C`w#U*yKvfdgp%I=J*55>#S*FCzojc!THA=M!di*C*
z#R}Z%94jTi*sk!HLSh@897=g4{DOI(YoC)B9{Gh;luqU17(-3@HK$l`skxr?$&}{D
z=|O7ViX2fa6*wj=lQ1eN4kH&DZ1_X{4p~I`7MJxW!*sKOT_1w?OsY;JbVGt;Ku{Ac
z7%!j~2E<^TVr@Sr?uL*I>0HJ_f@9Eb|2IY7O|8!gX|={?(hxk^1;h0xTCt1$x(PJ3
zQHk}owX&eJMq{6xYng+211?m6`)G$yH!Tu>pZHy($Rq-uE)Qjqu&^>w1_45w5AS*e
zYbo`6Oyvpv17;4TJ8X|iU-MHLz|*i~`LDF+U6ca>)gIoV{y-~oevD)N*hc!}AWwH^
zpUaW{uGezSXI-D8CVs?rqT4Lym3#z3`Z|4EscowE5Td!{N=vI|+(HD(7YW8@m-56W
zCD#Y?a4Xd0N(rVmM=b?~!NVlJ$?{8~nN6d43HSGiD#DseTkwn8t?45dK$!)z&<C$A
zdyP5_^-X-DT5lHB<`A!?s4zW6!GE4Hj;(McX@vE-hq~-Gd^5}6{J>X3=s50Nk#L~l
z2O(PdAs-w5XIh#MHT5|FFwgY%&jbaID%1k%jzAq<u1CkXj3!#eEUy~K#ylir$u$fr
z&HZ^OC%Il?jEY)6o1z360Po#@#3UeS!=Soq56M~3RIZ{}9(Ifoa(I!gC2d)Ktd*S=
zoI~A~Wh$R^*Xos=;~e%&?q8w)X`54P0~&o|YvM?QJvFx(*uqQ{kv9Ue;rx!66Ul;p
z6nT*j_Wa0Ot5B%EIu>VkSzVRfZPWy@)lhz-<by=os=XomktHw-s4dJCb(yxQ!`4$$
zoni6w=e=;dhNN)~sLs>?TYDnByW^CA9@wqKy5YIlw=dKx%9o;1ifQ*K>UJ&}eDAi^
zEgU9J*pA!{hgOL*@zcFb=B2wIt4=xOAJnB`W>fjSVturBt`FgrKO*YcVI6*SS?2K?
zZ^VQa$#d8fSKj^xXdCu+5tg=3ZoqM?GUI*t)$B#WvlGEsjlVH;LQSFiaDlt&yn}_5
zFSE-w1N4MfE5{0$?SD`nK5e@88c8(~Zk)5!KOvYKd^H-X{u-Vcn>7YfGwF#(L-ckj
zkIj>iKY6E-B+5=d^O8U~-3?W7-y1Lt%xvZ!Yo05+coT#NVEkFJY-Pped#=XSA}dg_
zA*(&w_VA^wnJ$L4TQ#@*N@a(0o)`%OD|J*%P@JT)<tMc*g839_Yo$DgOpB#Yuj}$4
zpEzCUI5lcs&vT*CtBnS(lngUndm?vkR8ccNjH8o6kved2<MmKP@qoUp0EH|uv9Gr=
z(G&|Gsa`4rJWy`J<W6t2=~#WoS7am)w<-3^U;#B<KiH?cb4_)rkM%kHzNQ#S`#A_z
zyejB>^)3);Ic9oVD|=1sWUDYuEw#&a35{F2s}3ZI;;zUNSD3OqTHz1n>f*Cs*~sw$
znu|^wyze@dzRpUBBak19d92aRw9vKWEl`U8!;sMrfF;`-8Nk-|Ns1__FG{T4m89QH
zOLaO4OEf1#llaZ>Cv*#*SAh?B0l!woK%dw3)HMdg_~bN2?<mU))A?8gWolUVeVFz=
z|94Bw`*uFhOgQG|Q`cKAF5kJO^SQtFQ^G0IRf0Gi-+`YuZ1lV~0^yOU{U>4z?n;*E
zvj3(#ApcIJ8hXr2?YMq4ztZ5`sDQW2&QG!0sLew>J9+`stbp9mg07ZrJGhbwB-pE5
z&Go#m;vs&2dTN!bglAgrmk|ZKUzR0X@1aFRYO8RlJ3h9Z?#!DSOnE=ucT?wOdZ*1?
zwlZ`z>7=J0ER#V%^Nl|t6H0?p1=iVM7AtZvz_#s(PB{+Z^r*Jet~d2$Sm$@kg`~wT
z$8xe}`8rvIm|_-s$I4N_iNQ_!3_6TdpsOH)0<@jVkt}rDY3fHti^FtBB8~9~bm9Ik
z8x6Y2%i*W<La6<rr5RY;_6qSD{`QDI=<uV>1xpm|GW@Wcy<ZU*;+fO958BMbd$_P0
zz?i&Ce+TCuJ}zju3|H?T(&pXm{o<lk;YZ7}QqgI+Ea&^}qx5kg0B!ofgpBx4slfNs
z<3?%kv|?oHgGUcg3Kn^Gqr2&!iC6*X*mh?p+2J@QozexRO<^3i`|_WpVN1^e?E^wi
zU~By@)XnGlK$2|j*`-WGX*Jj0=ZL~z0KUQQqRrY@#{F?BUYI-J$rOnE!=QF;M=Rhp
zk=LR5N?(Fp*D$A!<aV2SLA;#91-v79%_dPb>&J%hrxHyAJBr+wOM{MuVpQFpBk=D5
zIu0Bqk~gepk=9LR2Z~8V`W@$sR-xzC&Ew5gyU{N==X+(?L7m`nW)E}gt8|w@09yfm
zW4tWXeT?3Gk--SQS=@{h|IIDPc0z4fi@DKfaO<lwwpRWx1x5dYbd`t}bpgYp<bhy|
zgK%xG63{U*GKZ&6X=zmENysnJ2XLWC#K&CIZ9>D4HJt+pWVb~BWHlMNW}t-v3-#(h
z{Z5~`4hy#O#JVFL>wQiG<w76$023f}7@*;eV~y=Av^nP~;%k|Yr`I|epWu5MA}@^q
z|8s<?aQ&8cjASA*MKp(H%)D;@O*sWPVoyM9oXMHU9gJF3jD}^-{gzDD@2)5_vyNqr
zGtb^7yfISY&A>IuPa<aY5LN5IeQ(1t7pjmfJZn`5QWI-OB>h-?6rh`J0QPX>va2;*
zJ51lQ&u)+_&EEOGLlZ4*EzW%u3J*28261yXbyPQudGou+pW<R6P>_e|C6mp0ZD_#@
zMj;~(R)J&7Uq~9dH%DR!&(1HLqeV_>e0XcWa8s^x^QC5t46(k*ul!x3T7z@N4ca9@
z+MV%^(-Cs$Z%_FpIYjpK047@P(i46X?T0nqM#gu%f0bpiylU4oKSE6pV<YiESC$X>
zoSx!zb?dgNe63n}+QKvp-}D2n`|n!OY+8-wy!1`9Qm#gbNqe`1x{)*ueX%Z<%kUb&
z%jCBLWyNU(Gop#PV5gUaKX`X!ZM!93Y*x>Hu$?>yB<e`|XZZBe0m5*9fOr*~M_XI_
z<$+dzlAN~1qS=|~PYte}ZZW*6TCPXZq$Kb*&zqz;_`G2AVoZek(7wPi{efSd?#a*&
z6N)91oc4#5uTMIV`;{sBUaT$-co2Z;;<FRlms#_ESA*M4r(_cbzoL~uT$zlUk47p@
zz-5yF9{W=>n2?-y<e@q!O)2R+)XeeG8jS8qgxTeNB}e`Bsno<`^cDceOvD~?r^<I4
zJbx%^LD)i~KCf{()W(3;2<&KD2EwrNr8vWBiTnUx&u6`XYKC|J6<xqgY5kg9D4@*D
zENMz~OjK$7#@um$Vbr3Y@8{<Q6w~F2?t7aq!dm)~M8g{9-UYZ$Ew`p5kVj7Jh!tW2
z3tunJ_NZu=t~mWk-`ini%hUjSUklv`ry!ra(U~J1(XgKrk0E8NY>lHIs?VP`@KIta
zd}!G8KuF>c8eSuyQ0AnIag*fGezm(MDCxS-BBsV6!+$E(U8P>r<H59|x;}R-f>&q~
z;)~?+u$Pv&nu{Td%)@fT6pNmT6}Kl;Gi<~@!(3H!j`u9)YDROehlLa>1sY^9YF$Rj
zbl^^!9AS3QgON3*z(9%=2H7f!;(LyPs-`^gKlu?mB_ocEQ5urNRIC>qd1$4k*^RG0
z1&LxrW3+z&)h(Fl-TZ~gl@3mI8=vO0F(IRa1J0vR<-RZu^9C}Nnr_E}2KCe!%#=Nv
z?%4+mlC<C8ozNnMoTv1wXR`JfqOPQp)zGdZ`aqsu0@FDuo`#j;no`lttwt(X#?m#c
zA{Y|CqO+dgnzJ$uiLO$Ga+)u;9c#(h$-~V$@)$`0jfWpM6q<ZSMxo=7XJrvYnU;W4
zzImTLIC{&`uh%GCeaW2I+O5+F;gfJ7qt=33Tu^*Am?A+4kY*Q8U?E?wVbh<ZrBnyN
zTgw%hG?dlRD1!ZAx}xgoqz$Fm{)>daN2U3SbdTM>GOeRzgMrGit)E_+@A}DLkd{;E
zzvB)_7BVovU$3^-u?d6)TCs%IH9vf2EDiY`7Lx4#=L_DgCJk~!s*Z-#n-V_AGHeFH
zuwQ6-Aqb6a{3VXEA_7xiK>1H`>Ha7E77j~p3rv2Zhl%Ngm@0NIS7Y^I&8@B+)%>eA
z9Q}cloSFwaZhuI!35D{yBRQ}Uuc34O5(5vwo?HoPYE6)bOt7QIyKl0dKEa32@ep4a
z2Nm5K3tQMScH+g>i<-xhj@*KVbmt@j+r6^Uz(#KOd7<zTWOG!}F{?@wOzyIcLdYar
zd-sWa6w`hutaPQd6p(|yT2GglKXC&P2t^=mW?VMl4Dg&z{`q!#dpJ8uIG^1W+^7VI
z$ptcLzhp2AyA;hzL+hCQsBhslB26|FGimV$O?*;h$h{n=iMOZPShzR(E4{Jl;#h^K
zpuit21tGNrO2iS?x-3e$xmdAD&(opjPj#+Iumc}_F@5wad9n*mDo0wrkazqJXB#HO
zpoA<mR*iHlPLRgm{(J2vS9pRQZPy8)C^Ikt{nm#>bwnp(6TV$t%uAHk<f}l`DgI?v
z=ND>kVees#Z8a|~2~pl4l@)$x^X-;A+K}YLLqF+{*7sd)B&*|1D>*jPk7kg8Uq>a+
zf;TyKb&)5na!RhUrRA{!$$o+F(2lj(-338kpT6G~266$eJt6v0HCp@7yWa*dnvjq<
ztSjp0W(WT)FDTTh9&#wLlFc9Pym%?I5?NLvzFDoXV@RLxw6uKw73~hD#qyx(<?#GZ
zQQV+OS}H5zLdq^uu40QUn<}T!Zsr6fX0N6{f2=KrXN$8SQ=gI$uPH*6T2}5X#WD8K
z98<9+nZK9O#Pft&u>8Ih{!i|%6eOkv=|0R~xTSE$RfDT>F1}Gh$jX=PH${TM6#?@o
za7<K`9;K@oU5x*lhE>?29)Ub;xk9YRDq}9_kpa`O8AwY@)zntwP?B$zy*2CALU*%x
zy|x6>@HDJCd0rC}jvd|5z3e1k!NaD^YU6MGhV^I|>n-ty)k#j#6u$sS)ma}pLY_ut
zI#7FIE!8b}LGS5|*QEUW;X&t`upUN)))6ZcerDbJz>Uo4pmc2k<2VqEduVfEPF{Fs
zhBV5r%IdO;0uE4}tE!4@U|XJLw<Lh@Sza2*`<AFLgj~g&cA8jMGGD!|uJ%iF8<6B-
z&pMmR_ef6YWJ_q6pw<89{bd&1&_jhp<#coGE7npxJmuh((RzVxk4<HN;K6<J<mGyq
zzNm-?XSw|6PwP9&3#I}Qv<Ad-(!r$g*=Y#qmPJ!`^`EEp%SPg+M8d2u0i0i_UQahT
z96kbaLiRvBwy+~aj$KZI>WJ-ub2HEZqf`oCbwCm8Ff0&#g+MES!c<)n2YCQ(0ntu*
zq~(gm@##<)F*YYwE^;?}OG^7uFTH=N_no?hyqw2|`q($M+_IOzL~N;&RSKx3$q_=w
zS?=-E0g|*tvFh);QPQym=`xue^q15-S_G$TFZ*nbgO(doYcttK*%MW>O+(hVO%Eez
zW$3M;cAs9E+OJE%LpgkD5^}j0L4hmg%YBP0adSydMwTiMTRZWvtk<;B+rWI1+ZMej
zDZyD6UtIrc3f0&AVJj!Kld6XMRJn3P!;M=;0WUO2d*Amm-a8%_3oryqsW|mqfyuv5
z4g!^6ELTz>T+QF&qs!$fc&>ke!OQ!wlI1cxJ8C7J@zDwR<@DkixtPqtiz+j-ZqRt^
zC|~+?Z<^vV_N9{K$~goh0QKgXm?{oUwiGWvUlBRZc%heGvau8%zHo}k4PJw>--z5M
zifjl_{7#LkHj>W^fBWFgyFcP7@=yS=uz)(#OW^&-vAVbY5aMj1C;dC{r)~>6W2Cyq
zo{t4WXB-01_(d+$DQqiD{B*7|E1jotSL5=mv}!lrjk3NDY}j^nV-4IIFO867j9D@@
zEA3qEMWSoC*&^QYWi+&yt()3@rH<-mHC+v^5Wd${{AQ#$#Q)FxCdh<-jL>$P{iLc<
z&7oCJ3!vn_V?UydbHmO)Z44sgYT6iEjZ?m`auCbn52pL$@_Bk>=Rbn@b}bWl73*)0
z*L`j^?B4QKFa2dSUV2O@QeVxL8#KZLvGD_hl3M;BiuD&0BMZlW=zuiF2`(5AcD6L7
z1#ZeT0~>B|z(05ZA;Vel^Y!HP<@JB`fCZp`^Z@$V<^M|$SpPrtfEvnZOUeZ1SZs8z
zF5T<Dmy=m6jWWrr!qB2rrF}(u1;2~krGnAno3)^#1R&XxZRwj7wSS;<JwNgIQ;!oh
zEt~NKmEo4(8)H@$NAhVAlO9t&>u>?Gl)qDz4h7kiFkA==>ouUVTzxA7y|~n&XsMq1
zDeI261+k2;3a#LEDL<sCNr#oCN8+K)uMG-U&Or+*!Rw!zQDJP$5fs@Do3e!9(oJz;
zjF<U!>}Jzq2en5-J2jO@)jo(}P}jB3h#>-}JO2bb;^RmnVri(eMzm}WBNqW;X@a$f
zoDJlQ^GN*?dr&L<&{Be#U;JrOqs-iuf{qs{pv<h6w?zh3%Kad1W9cO-YcSLZs+aRt
zSHDp=vgSGKz{=`pFg7$B9>Lh4#jIE0C-YIyEBlO%@J>09{FkJ2HG{7(-Io|$B#9xy
z^uj_IV#>q*DP_uqw5J)ur4Iq1`4p`z*99RNfBES`sply|@yfJ6I^a?@B^p^C4JUhO
zlV&#w^%iFhf%Ite$1H6auS@(w>y$L>Y7xKZV%rs9l?~8(6c9TT?-(@C@en`%-mXK~
zAWBRXfKI4>%*+bD(;G?z#V&+e2gkmXb8wyd=$5p9-*ZM>B+9IW_`m~Lk4)NAmC-%e
znmpP!i0V0%u1y*|RGicFQp!vClV>1XX2Y?kbpBp{>C~7A@T*zrk!@AUI7n^HYcB`i
z^x)`!efzWX@(biL4BUQw?q6N*y9rB5Nemi<=Y|p70t@hB>rcv~*WL}5@mnqY+4@N?
zWZI-xYAOugI|}E3xw!_coxK6V{;Hf%L@&I(-8#R1MCX3qY<NTue@qA<3^J5Pdk}07
zCA~Qe))M&p0AAgnonGHO@2(wgFa)w0%rSoW`O}9DX2Liu9ykUiqOyqL(l96n?~I+C
z-j4^3-6Lyzy~66|;qSB$72@&nc7I~cw*D-=v(3L6=@zDNVp{-EZSjL-4zFqGDv}HL
zoCI$e!U;}lj|Y=(tJOuJ{s9fO*+aju$nV~8*){CO68tjLGU(80mTG0174$oU&8iNE
zZ~x7i!_dR8lsJCxU0z!H!C>0CdZ@_M6G7U}sx}Nod3Fh@3_X^473z<Xc3kLjzzrfi
zG`<Z_yeLYZ3P1t|T_}y*o$x(BU(*z#2@L4tFd799f=eGBz`(uE>qiFb<L^-uQ2jT2
zJRdDR{h0*3xOBSka|FD&yE6Q=ajX9fJscRAW1@dJ_OxB*WO@@Yn5R7;U{EwpkY&6z
z=2|!?#P{X`&Qnz)_;B&_!(l#`Gl}568tu(Y47la@cAo$d>-+~uO=*cmQUy}`QsYUa
zNkvJE5<KhE{Y_%=OBr+^4Ovj42**Ipn&);14}{RCzjZ6v?cscbSK3zbEc%I=##ke(
z(9J}bpOSW~rbs-4X;9_bmzz>@Wr*4anhRyDkVJfi6bu$lGcBgzbAj#M5hZ|Il>&WM
zdD`%|RU!e<urslRtJrWn^&nEzOGHeWKrXHGjwA3emzUCmc?xq?R}87|&<I~@9(nqT
z6RU#Unrs8OB1V{uJQ~ZAZ!O+n8EC#*aEEyTI*4|~dSM5gN|}@ahWM#3Wn1dHoe}pu
zv-bArJYrL5m7v|Vzi`7^iFp^<_WB2bX)#nRV+;Z7iFO3}%qyo<cx~#yC@F^t<70%V
z*-!K;r<Rm0L1AF@TFV8`vF0R}sU?nCT}lfXg~_oQ(!l-HT~aC`!KYd);Br_v>bs8!
zK)2_M>2^L%wKlUon<!To)axX@dizcKGAN+4k-*9SZ!k-_!C?T}K<Ch+#aH&Q(*01^
zEj*|3K1uKBXKNizLoDd>>erMI13_`s666C$H`5F@Q>@(}VL_HJo(M?KBsYS{dF5J3
zAZ<XYAteg~NRd}xuFVw#<?#io1I?8Xhcs?`*mM!dftLdi!2_tORTsA8KzssUXQ3=m
z5YFX+9d1I?#-ITz!R*u`9_7bZW%?zjU|}>R;fTeeAEMG@amm^qY4jF-fP?uRDWjV_
zvV!MbR#KF~A`MCW`NtNG#8y>%2GC2%_RVy`oaBi$xHkSa=P~CheZGuwUML~CpWbAc
zfcO>n3p4!b``wDK@p0Y@)D0^2dI9Q(WHwqw?I!C?)=+?cp++x<Wk#eRCWn>O$Tb0i
z?ooR`<c%yy<~DgC-A??kmKI@<3(?tg(SPgzV&lKSxykDa#sf=&f}T~AO<|F+NUR7N
zdSoa5tscM&dX{iW7%ZDHp<Yw0N~{M#H+(>BfNU>-Z;*xG%cawyzC38r03+>Htf}w|
zfCpEDN(1j@Gi!JfHKp(T8?DMNFKPa{L`!^h+=^n2|CPwY+xD09`I)&rp1aLhD=Lis
zXe_g4hf)o(I1k}CET(hVq~wj<oxY4}?agmvWE*ODNzB<CY3YKz@>oxvDhQI=qjfu6
zyg(11jZN|J8Ce05LNnCwLhO-Pa}73Fz*-rW@-Ira!V>7i{SGM=GSa|;OxCF!N~`T6
zDrD56#Ah-Yyg`-VD@*H_K;GJ2UwAKnL?43D2bM#d(MG6o+aTtDtvB{17TXGoRn?g~
zDEex91b<t|NKGL3u&)XrlN~Zt`-F2l<#e4IsJWqx^jtjhHTpSPDo~^vDYIHkz%{~~
zWOt78ggh~8mfUY16Du^Fb0X|?F6IA^sdEgj1Zui<GO;t!#I|kQwr$%d&cx1%lZl;7
zY)ow1w(Xnu{qA?`R`rjrUHkXmUEQnK;|=w+Rs>Id=+$1P8LFR|qU`>wKnx@V5qS)g
z4jN*+0|~z%6#a_ELVObB^b;hF_Mv`9Z+^@=s+_Gha6MqkJn6WZnb|h&@{G%aOjAPw
zXVRvwq`NHK&chnT-xC_6uZ0G=5|#z9MgR_%W^<5<eE#J5=^@;*YCL*t*aDI)mx1Cu
z^mtvi1{-UaGW1ezABeqyI&ayS=AbfZuECYLCD4{QO1dCron*QuqBV(uz^F6dK|DKU
z^WO3YEvl@FY9r{tG4O(YT8)HW&cjhwA;zS;l9KJ4v?d?vA6{)aKzp9U?{x%d|Iy=M
z+MwxW<G)oq03K87(&&xjOK-8kBm_!q=NFWnZl7hyariOt*6%M^*Yxb(oMcj*Nimet
zwLfxm@}kqi_KFWQ`|Z*c6>0nkztIQ(D3xJx3c}Wn0=-0}?cyQpq7u}M_BVA!(%s4S
zTw>AR3H`9de?F0NH;FO1tGF7l!fc3vR$e#?-_Yp3x4c{<VIS|jVe9T@9rYn*xsg|v
zyUPBTL92F+K4oWZfoJzk?TUPx%ad>OT6fSfW0zabDVv;hn^lTcil=nbO9`e7RylQ^
zEB>wqGVSq<tRu|Q!Tg^G<tAICJKjwL*Ix0*0Ho(n;hbA{qN%ymT@9dR&2JvYl!i+u
zt;jS9voW()r}?%Z1yzvQj&76aXkXH+{m-}rf&{3k=M0NmXw}nuu=LJKo-U8wBu$!U
zP+Z|E^|d-&V16|&@=xc)h9y+u9e*BT;Dzm>4bL+QkwGf$)W0t{Xt|KG|Ltozb^JmF
zWd;&mKu`iMbVg!HI8Zv5G>)M%CdgVqg94Q0Dgs;O3hcua2A3%a2EPDlgjnk?Dk<d`
zB6xLbqogE3e}`1q1wNd1ORf}_mP%?kASlMNFw}GO0Rk|-JvRouH^nm)G~g`KS^8hR
zAI^um*d5iLphwa>W;YJuxv#fkFyS;Ie`IvKG?4*EZR%b7Y|<rqvqu~ERKJ7!QWmxF
zk2|*^0uK+CmH+}yC?97N5p=CF5Mi@stqFWkh51VQf-fok-D<D&3Befm@I}*<2+s}p
zvTYRCt@%DTFJol5bo4L=4pv&6=HAh{{V4WI*KA@kc4<e>$j>k4L3Oss8~P!T?;_o#
z=}7=yDV*`ly!h)Izvst`hZBH;7@abc-TMe-Uu9#-)~7EllAxt|rGV5(N5GuH7vIO{
zeQpA^mG67>Bx+1g%uE|g5UZ?>2Xh|gl4A5U)phCPbsWtG(x6bVB`_I4k}0(s#2#eZ
zM{@$Dn14HE0LTjC5$O7bE`<zVh!=|S1q;wP^2z!`&b1d9Hvg@BI+D0nm;$Pj^s>~9
z4;Q^|_gEQxvXMvz#}+O0M{G^}QEbQvOh^jLE6~*NF&`*Y_^jJ+B`epcQ*t-3Pjl(Y
zf}l4jgzn1Y(4(nLJjwvwP0={rR^)melP)X~iJg9tpTq@YORAPUNw*yyflA`b`~YaD
zI)okHrRM9)6TF-}Vl^j5GiMl~U_A`wdbn#9hhTeHu~_qo=byQvezZzfGcb$W!ffMq
z_5-+p!LH+4EH_Is$?SJMt1!JrpKv41N66Md=h$78_oifg2k)0aW%H{gd`6FC$w*b{
zbjr;THTA5DT&UMTr0#<hr`(Cg7J#x7RqEXb@~<Le3n>>pTLsZZ%fv-v<=y-V$5N?#
zXf`oa)}KGiUUehxyle00&Ay_1z_Zs<-82%Xnj1i2w-%s-)d1=$`~fH6r;3O_W^LvR
z%QB*|`st3%$F7opjr(9eybm8wXHkt$frXzl6{e40g|ecZ)U35?LP^Rr4*+D@y->{=
z=d~t;Vw(5QHzbkyR>Ri&YS56Tq!L?#?z7eVA7bByG^SM48*h}l%@Dd+)oByd)eoQl
z!m%Uh1-US*YIbD(W5%Poa(x=|E~#iSR*sHTqB#V%Znq4Gn>2!34uLn(fB`>*v0?eh
z@GK_FID>3vmDTs}3iI`Mpa9bB7e{Pq4R~dTC}?iiS#n6%j4o&r3(J?De{JHZ&~9=r
z?a-2PqzZnHGElZrA>lFQCS!y$7byPe3Pgha?V@-im;aR&BNzrR4lc4u;N0?qsh@pA
ztLDxi`PPoL=;0;rjVsW-!7J@jPzP!?DFod&zWm;n`GLYp43iOq@(7T#j%>CIo}@3m
z*{(F6WIC1(e-w?Z%#}=gLZv}W;&G>=xo}wAU||r&Mp=|?owjl^&kX#hlI^1bK{3x}
z6Vw)aY^4jO;IdGV4^+c09vSaP`c0rT&sKS|U#!jj`Eg+s2>&BZ-ML|I)v;Xrs+J1^
zFH*i1_RWUFE2O#;!wW#VLYmdpgIMfs%oD1ptas0QlJ3~3!ntSxec%|2%h7F@x+N~$
zVkVxEFqfcRzZeADOWH!rQg#+7rU^1c<Y9cjOibHhp4c~BG^W{D|7!!)XmSC>1jC9P
zN{Rc!Mul`Qn|`VUuR<$7TQbo{b<3_LLAU66huZ71PMjGTMIC@Rk=Nivo=?kJDMmua
z&!w2t8=f|2LQuwXX;JP2?<p>h(JOu4KA<3uA>t4pr9nT5Q)a<7h+B?`zC^HFFfFc_
zp+)A8vI<S;B<m!>GLMIienf_RdjnmVa6IJCJ6!NnGD<5%;q(<6e3pxx=}%nNW}u*V
zFd{@!Sn5y7dwsy+M&XnPQT*~aBUHH<p%6j(=Uh3-9(<K;9gh>IdQB0tEESuS_wt5K
zwscoV%uiU_fX$foU3&uszmimFzCu0+Q6w&Hyt#486_T5u;hUT<qQ^D8D=sWjVq;Y~
z5)1t7t}S^w9W6JR2jmh)VlI<z7qgqRVkv3yKNnmf5y%1J74GB-kr=U?jkC#A6@PQ*
zQRyiuIwb9PgKkbj;6`D7$)1VpDw=l}9woiweG!S5yW-JP8ohw~wB)J!v}{XyA^l=?
zSG@~15(>1t>G@M`hl-)AG>c;-WVVfpf=kpZPDKC?#g5DWr)NQ5$f{B>?@LQ@n{b{2
z?U_S3a5xd*^c4aA*_Y6>DN6tij`(@4#F>{!6m{0T6VP`ECBKr{68i7Y!f?VHPi-$_
zizK}OJ`OIHA3Qs!<4Zr}Rse!D!_3G#-tzWRraQ%yptODJrZD{P*oVgDR1cDyfAC=h
z@9Sw@WsKTO_t`}KC?v!4Oczy3QL-r(<@h9o-XrP&gnSK2yKb?FBje3-3FGV!BZdRD
zl+e4;w?mZi2za*Hu?|E#{W@S5`UaZ3k#9wAh&}94`qPC^i-fLuS@ZnXY-aPHR~&lE
z7pzZ2Rrn#H!$Nx{CbiA4<9Ji~B9LJoU9A^g1!O;Iwu!Z8#7Xi`vRZj5mXUc&v1d#`
z<|H^EW2Y@aI#)wx%Vagl^gc4DnXV+)McazxEu)ir*NA9dSE1x?jWU~A)|i6j9HrwH
z-8YlLGwcYLs`R2R1}jYpISC_<qC+fO_<>=QIHhd1U!LVu!)<}8`%7&V-EwoGa=}~%
zjuI2B*29{h+r&7d!C<RHCjhJ%8mu?VPNxZgtu7h#2I*SlaIv$OecTZ<HbIu#T&`c>
zJneZB+UHImB{?k1=ImEu6y%d%4#|dhBhPc%lxBAD&eNt)GJ@d6b;DGJLf=@hvcH`r
zopBR#fI@^VVLX-M(`fyfAwYqYq-l7{P+%w8`y(sWT%s8=z>IgX#f1f`jd!Z`XAv6!
zLM(P}BC!BHx+7%}h#fmu#D|3Lpp}7`;;^YE&)kFkW#)ob9;{Q?o^=d2^MGA}TJldw
zab*UB=LN!%c@J_>&%^Wy<Rokh84{M~1`L}AUzhKEmqCmI43R^7rLjdQp0!)+ocoI!
zY@=?=*!|&QKv5}%{2BWtVC)%hsR$h~-{qe7=S01~Q&>-o?s9ki>lX;bC1Uu$pH6?=
zPR&lkN<{r&-|tnkXg~@EdV#BbanSUf#wF{W^nH8Vy+d-z3Ivd&jzh`GU`cwI#s^C;
zdWPc6^DD<DbWN0XYTod!xAYQwkB!R}r)JJp_L%(NK0z^8hXhAGYyv2Y4r)Ivnh68x
zP?np_tIijBq~-7-><S>VyMKi&$$3_t0p@idL~IWsJd(sQ1~cw6`gOz$m|+f2j{O~|
z@8-%Wyh86naFI|lNvo#Vbl(hPJpxgB>*_rXOV{$#Kf@<4#o%81jUK;Pa0WjAxAX&z
z<wOHQCJIvlc>=hzf0)%YuLf{;z<`VEX8il#J3sXa(7rjRpikTsz_Up1Bv2n9Ac}hp
z5IwUDNirAYL^B$zj{y{0P-FH%yCrK2GqeR-V$a+Mdz<{b)D~ak!p9(|e*dLW^>=0a
z_B<do?N%$#jN}!xP%6$cow?rSTKouWBB}W|Jls18JeN|@OZm;|J33@cF$1u5UMPuV
z5$q@R#n0y*y&MUeV*{sVt;Yxhm<zE(fESszv&ssDp0)n3vz=)6q{ser)eKf3szEE~
z2Wd&i=fTjETp3!GuxUW`;$CH6;78l>-uAmn{4Dqgtll|=3!zmWnv%qagAxqN9b&a}
zCp&Bl20zyXF(WYj2#X6nEj|&(j(cCfhPysfH|jQw`Qd6BIcICDdQ@I(^kgCmkSOT_
zd8uz5sXpSBYN;H*#Q`qB1Aa&LE{fo^YI0(*)o$<~R2qW3SOU(R<LliPg+Odx8vCW_
z>ZRlw;M48(?W7ssGP?_WV^^DIfmxVEjF)(m*+y>fZ!xbX%Z26AwT!RnGsZ;cD{b|N
z<>ub9@v(lj*WZ?zAyc*q9?Cetyw7;rGaK`Uh1jNpDm}#P`O;!1+WZH#!rDPSBdo1u
zyfS_nn;d0%dxE368=-??e}3C(bec(!EK}Yx^#BXWYyTvCM0SGAYZ73icyF$hSV{uo
z5p$2%8!maAbEW2%814;H`tVH|T#*J@Jyc+5U_iGb7|oco*$uuK>SYmd?F*AY#6MA1
z(2<KZS{m!5rY&{@)!dR%A5(~fOv}0FS>^m^2^W1-LrrDNlo%)p<Pb0t%X~O^A^9qX
z`LJFrDl`;l;IAB8FwWu~T1MoGgT29(8E0ryIx^yTc_tIUIY?uPRrjdqpq<;iC_@3y
zFHGid8!>7dM`7m#8>R(NiH>9@Ep?Gg6+K^a1@%VFxBkp8hi$Jl$6AP}VdmrFPvh;o
zWABx95g}Li%7*A-pqMMiY0V*2C|JO9oji2yB)-P=;Z#CFh^d<hz#h5oa8E^T;=OaK
z4@#n<UrnstO(BRga#pL;Wwcp0{y^7Tk*}Bc&KA$MqtTM+2?7EbI3KvS1C$8QJ<ykm
zs&C>dTj6*{2$1p0!#!21gwczz6usnv>eQ?=Lp;FgO49LKbV^`}*a*KAviwGb1dc-G
zdR?)S)UIHP{?WgR!?0qh*@quTrVT(9ZwTW*YFk|@*T~vfBP838<qW1V9*9a^f6MZ+
z%=EpuOc5b}zbgYQ*GfB3m*&m+V)#EJce@WUS{vdL<6K9God)(M8}(R%y_Y)^6#p6O
zC&3z}9WINk+G$t$xUdFsLAlN32NjG4ZqGI&?KbDRli4;9Nvud?$;3IU7|Kjdt6xCQ
z39cjBG>6VoOKFNb=Q77tLfNZ8(c<tV6;-j-Kr2y+Q>y^pBR1SecrF@$C}Bg&%+Ljm
zj-$tA5$Bz1To$?$(j~1TR?zE}vutP)i6nzfBB`7Fxktd(Y7+?jL(SePV$iFWK;773
z(UcIyJF4!2IYy6Si78wC^7u~$p*#$=XEE=-BPKAi9GB%c&%j0BLf5y<vf$M@7dOJ~
zbjO*1ibxdz;<g%}#bY6<GL}1CYc1}tYDadGhBhsN>e~qu8q$7FAuwxYsRJb=aPvso
z)5kB{RGUc37mvW~-Y;04c%=d>Qa)F8>m_<>i70$KK}FV;tin_5gyb<D8mk)qf$veO
zzBKtzL4D~+p6rlw`%JMt9X-#9IiMH?nn90uq<aG>jRdsW5mm1)YTUaOvCC@T2^s)V
zDC^wb=ez7xR?rD>qS_06cMW|kNx16zxs7#Y`m;Z_3hdP_#SHAO%o`tvd{h|Lx*e|D
zP+0exQWJsEE$bJaIstk}zBlcn)Jk*mo7L-au+5$kMk3XeDD*}|o?Ckj)PUGvzJ}%0
z3tB)b_$b!V3=4Sagk-JQkcL>H*D<<#`*3mh?0FNFtw<~X%M)Uvw}5<WsO1LU!BXID
zkulMg38a1~u1wW1;|InF>Y=>e<z-RvONK4Pf_I*n$E(lpwlqZ70Dhgy<YcLnSG`M=
ziu>5kaMw7^q@ep`g??+c2p7U#(61fZZ=U+Nfd^qwKBXkfu&+XeJM(k8yHc1~yZh1m
zlcfgSaYaLg$P?&=H4$IF9^S#*>cY<(nAL?pwFP<?=ii&kY7F*rNy@|M;*5cs5Bae}
z+LM(>3UDX&YE)wP_P-MD=BBH<i3B!m#GmV1s3FRb5W-0kGkqXq|1jbmqmOeq-B|#p
zP`E)Rlt#^pU@~-8$7V)#7Q(%8ytPaOtVz!4e%aps5p@eJZakal<BEuGXUiG(Vx~>N
zFzj}8A5S!G5Fh(j^diPQ(tzi<(K-c@s#Ld5cZ+VaiT*sv#%w5z2;1sXAhW8t^gSMR
z==AV(`K5^Lg(iHOvoZa#w?Q;z9oz-LWsfZ1pVgCHK0Tr`r%DH-7~yr6xQ(fZq1`eV
zW7#-Q88dTB;$0K9{e{-fxWM+KUDL?s*MpHEUz?pn_J{52a^0w^KjE>TAHuozlRpW7
z!l!}^i%NNSBKmZg`@YStfC;R--ZlDactxjZZoIl0p<!tpS#+V53~U0bEWi!0#b=bo
zbGJi?db{Jb@i9d0`Rc=w<3TBB#lG)GeO66Jq6WoBP*KS65y+Sv1+OAAqFO4?f;x{d
zp?gd{RP4B)F;HjW^RxbZjQ~0r7bz0b>h%z`NdrFIbN2X~#Wy=~G9%+BRu411r7{FI
zQQXIeZkN7iZdvAN4qYU{vjhT&b#lzz8ib~+j**9Sl)<dR2U#U!;z?AsHcj1?f%7CQ
zIfn~z57f1Ks9vw((dQMb%QouiuG~v}<-Q<)nwMYw2|fcl8qTd|=ZFO7YZtN`r8RAT
z3j(Jt__A-OjE^<YGpbM;t1W1tirbheYh7T+KR|B)SK7(@lh3)0;?<wUz-EkSF8}DV
z5PXWwQDQq7A=Pu?{uwmZX*_@{ptDgH+7J1yDIixj<YEiW<r50he5LZgCCv{eF5t}p
z0_qP&uB4q(N}ys2cM1UdXlv8odsgh8oC14*J*$$0Y*v*^44=3)v2v7{WQZnr?h@X~
zEn0!9>ZNa>PYu3O9n(`EM@)P0{f~^+4rC-gIwnhvpiHuGHT{<Y`)DGW-IRWQGWqBm
z@s@r&{RLn%1)qzuYNMkWI+x7DD+LT;y56{zt;QIxxY(B%Ul&l@)b5NWdJw&phwXmA
zc#O6(p#`moz%>2#mY7w%r5(Ak1<Zmq<)E;PnpCWH)cW#_RLhADh4xHHx_|WI>~vkc
z@h#I9!EdGzfcZh6mYLZ9EvpPMvuHhCvN_5otHVC!a@H4PXqOx&j+}rLp!Ok)8rQV<
zoy(ua&%6*!&;%e8?m_DI=QBy6=XRL_G=}vLt;KF{!X{1(B0$#KyM0$~Ow=(xXv1&%
zUN&kELej>RP9D8-|CAr73!N6LXw?|2ra5AL;6D2C<iRf5sauw`I?NGLv<8}(&Dt{`
zfB`WmgjOp(_h;d$Idd%dc4t%j$=D&?_h;pMc&5L4iU2CkcT}xnK)AP9n`m%ee`Gvn
z{U7*UV0b9LGiy*YL{d1S?+?eSCGtYmoUAXNOZ-o4#+bVY-YtezARCR2Mz`4Eh&U2$
zsg@ddvRJ4pmp=~CliKHRn#RkVm$F@#xKAL=DhY&nPlP!Mi>^$`+7~ONh;}Ql5|%6(
zV61{?NdWr6d08xXuoC*gLZEi$e8#_C(6CnmUu@@K9Gg=4+4+_Zq{t*HROAC)%B%_}
zikv>s?aJ?T<<nQlGg1uij$_Os705O5=lHasSbbbmAcwf9E{vM$j!_%DwZCcM&Q=N;
z8dW1zKK$~yC}oYg=w(3^MDP3r$-3C@kdelxOn@3m{gf=*W$ELYQOFoB69d0GU?a`(
zKORNTp|rhPYP}6d>2{Qj&fV?GR`of@9<=z1KJ>$lScj3p!LIP+VDM#i413eIN^ztP
zeZ*DOoI~eb%d)0Zl*pT9e93%QsSH!ZbSp=K!SMY;2N%7W4%4}Z04`{!9jz%~R28JX
zC_p&2u<32V5Vv;_H4?}$*F5E*@MXIm6N6cCSF-ixABhAVGT7>v_(}wl{rRzXe>69B
zS3UrKXjASww_;@@7MPN)HDJsrOMb$xDK_-MmuC^+fpCt^4~1J7NLma9dWrh>4x=){
zWIk=}>%uxtQXqBszIut<CbI@Hm76zs4<PW_ta{0d&WOyflNm#~K&)A$Pj5E=;i{`N
zs5FL4(k@CPr$2b)3$!oTFY91dSz#v1P%ou5ieQnOVscx0b2)CT6b&R3ylrb$#H+LD
zO699Y3n9Sl8sRc)@FMh`5=vQ?&c3z_K83xx%+nBz<11^(=AVpFQWP?yyOt1>00iaX
zhFqcH!H``+;lBkurJZAUo--FsUAyf}J`+BCB41)Rw#nq5f<J-(Wp;)RaJ!V9Nh|5n
z!*6z_un|um8#vDx<2`HZ+r+_-B(g(x$@9HN_$w?lps#&Z=D&AJ4W#zN;mA|G<W)wl
zR(J3Os;0Q#RV0u_1GBB2p^M^e0VdS)2eEoy<?D?_zPNPA{9~v#IZ&s1bArc0B+Bli
z5F2~m@&}SL{}8C2oQtm39@0p=em=&aod%9;M(fg57^<n8MYX}>oljY)@Ept!T1aVZ
zosVjFt2rPRRo*Y!H=LBcREJ?FB;MRncV1)56gRLxBvQQ`7nNAq-?SB<0Cq2$->z}C
z<R)SwA5FW>8tsgmG?*u17Gtj+Q7Fm%;<TlWxeSeR9wcQ(`cLDI&||fw|5YkE$Xm#9
zBCU~4T*9P;J(1;q0NIuCiEkQ|cm~Yz_>fqs)kt*zsEL#bh%PFdEm3$b3yRjHP_n@!
zN38P`{gg8uuN+aXTAjFP1lWiBr>H24b$XCfeja(#SlPBmFWd>)FR6pF#TuouU{-dl
zas)oLldnxTqm_<tACig5`n|0QNssC~)&~$6`e24|L$X}47H(}!;`eLlc7Waz-t0be
zGw0uVPYY0#aWpaWFA98>dV_UpKtIQNgH`Eves!NBuv|-R6og@|0j7A<&a0l9P;9{D
zM}1Z4g2YQWY`01<giz0Dy_XS?66`h_=NtVT%7}@nBuu{7pYcmoM~{70>HO!)Z&;n8
znTy+AndeE8GsjT4PW_wiI)_seMZGIc2l15q2Ih`!y&I4GUN6IN<D739g~|;O$rI3q
z=duzbyCB@yK$As-0WVowW-%Ur_3HGgxjAgupJwdSK#*=K?C@&@s{PfObyy^I-ExL<
z6AH6a<q~T1TV^;RUv({;>1cZ1^<##{JBx?y8EZuggr{~mqzxRMBcn4V@!uRlI%!vn
zvT22)E%j~g%&qjWH$H{plVlH%7L+NRp1FjJ47^9gq6^oA0i0xw3mYHT`jyE<FDLyN
zT~g_Xf>?F+>k@LenA$c*8O!2ad(V9}<qVBax<`wmq3!QIoEU$`A;%##qrw6Z_&xPA
z<!vOJ|6Vg8OLvmr6Z66wrC}d=!u<m`sOx(pGTl>o;e&Yye5ZXu4e_{Up@KzwQs4JD
zw9tIzG?*CMrKl-=FYK4NsUt)*e&(5Bp3bwi*!1Wo<bg!Xxym!SQv-QHD#vR?{hyu$
zGw>%P87i<an|lDXrrA1&+vo?_mZ(DWWF9va47l$VwD<pf^xsl#q9ABgP_`5!SPGzQ
z`S)h*H0-ti$I#Q3K54!Fi>fvz4*lXELVN1mdyKHNU!rmVtX-4{n|mklJ!mDSNb$>i
zyj|2+b~Ao%e{2amKd9SMCpUdv)G=5yd}BZRXD+3_wc1p(YDNH#KqYco;>srIEh36W
zM~0Di1O|LoMn~qedj3gz1$lxv(dY}XJ)8~DE|}Piz&N&+MohOkmaU;0lS#S2dX!C7
zse|uL=c{SiFjBJ>*UCnfx3)a?PrVXlwxZ2>hERt7Hpd_56R%3C*Oo2kWZj!DIX$s7
zvw2+1+j@+<TvOTCz@3NbV54sCvWB=*r$i{#eBDi-t=iN+Wl0HAnR3r+N}15FgZ&k7
z`zyQf;CQRc?o{7~p+SFG!QogR9qwTP`wg|D2!Mx1G*zCcYjMc2S?iD=&q{u!hU+k%
zdWMD$@zR7mLwPMCtrk;<N4mNGZ$26=Z<Wa?vhch^7!B#Bpkt!vhE^?*NtlXlZcBAx
z_xEfX)&fv)#ypSwGf;tVZysxc8Z`tE6snkCBKW&RyJFMc>!UfJP@S1d6Jmmb#JJ<u
zURV;Jh=?MTo?RDWGV<uftbpR0uzO~YHO}yg;x)U{NzGSeFz%mPe{4#M1Igkq>E-=8
z>i}2tB<y7nIQ2~|8s8hu_loEImMtHwjiJt0doJ;g!vwUoKrt>N^*`+ey=)nP&a!-{
z4}JHuJIw~xOvIeWiTuQrA~yJ|#(efUk>Vkm4DWroUrkorAD^!c!O@VR#0;D#OoJ*A
zS&6@T$IJ91Xe=Ep<V!&5f^;-xV{kLX%sn@*t43$BJos!WB9`T&+<%xj9_=ArD<dng
z;JHy_kLIz9o7ymsZJT@ap&fi1jRiC!5WG+^_o>ZDF(eO_8)K}X;&Bel{Ob*xcKP4{
zcjiN}pytari_yTbROsnq_tDRjZAy9t0xslFb+E7$!5=?Ba(zmN^(+wQ(`rs}&tOVR
z6A=cU)x2uGVc<jAAIOV*7<P~*s;YOB1+Nz_`_wX*wCW1r-v{MF4jN+tm>Zu3aR+L(
zUFebh%l#In-An13_8ie!%_Zsex6(br-$;-~7i33lb~Yv8?4+)0!7^urST^{T<jm!q
z5v|DbUp*?bkkItFmmji`Y4kl_q%*TGDk882%iuAqDS@!^oWU#?A|HJ2+4C5~RWAdG
z3eu~N)P1GMUhR59OFbI^;KU*imTIr-vbuwZdDqC@dSiY1k$v{WJ8SL)n})F>3^&gL
z1>aP#WyapQ=r)v^`?-{&r`h2~maH#hB~Zo*vX-YCN9@TBQN;IQ>uj>0AheLrl>cH@
zj(<(^qqIL4&RNoWiE$0x1o!5MDz8iQ$^0Zl$|zXXtE58V@w@_nSu1QLET%FSPFx(4
zbOs|NuZt}eSZ^aFPoUayr=}9uymcChK%i?INYfKNx#Vo21#vvRA%0Grx*}$3p=`)y
zzkQ+q%|LMEIB9b+*?-8U4~qE_VX#%)-Q^E>nfgIB%}oT#@b!L2#xF?dAqWI163v_G
z{&aiUJ9zR^OMM2sXV;BLRn2Ofh&O5*QSu6xgu){EwCfoS!nw?bc2q@dLz_h6HlX&K
znIX^wWQE}X*u9c15N~P)0AR|l0(YRPxGY8K5^qLPq(abT7-*=|V$|JDnGR6F8wfa1
zf<?7?!-h@@TacUeh(3GHp~#f#^L~DLE+`_RxTxV|P|5(x`|H|v5|-=p_UEU?0};$3
zinTI#<)uroYHsCv_z%CfCy1xsgAML(IJqk_b1AJHZyQS<t(l%Kh?FN+m&6k(Os^h<
zrH)<eaNG&Do})J0WJ2ye|Mfjfg|s<XjHz2rx}r2b#PQcR=y-jbh#h!%Pn6S+sVQe!
zY&I|Aw<G{=(jcL5l&yFxK=fFuX9)2;cvB`AWAd%nr!~9-=wi_c%w*X**vL8Tx*i=T
zN?SBBFUs-7ELa3}%DQktoAguO7(>ZWWNeGS4d=UbTE7q@tYK5yyzj7IxN8I5sE=Lb
zy~=pUNEN3%Y4pFx?%YX9St1@@)T`B~8}q7eH3b006;|wWxVk(R=WNzw7+q%9>|5p1
zX?#vNE7a9z*TOel)(u+2s-67U%ZKaVN{Q!=3nDw}CGsOjXN**pJo+Q@LLc#d2bj~x
zHl3e+Dfa)}-XPq_%;?)rK^Fv4(bIQMak^yNvta*u)2vL8KN~V6hD2s^1i_42D_as|
zrUsymK4k?rmesL$fABsn7U;&3v2W;1Av<daWp9$dD^173nj0B9TtB4>jummxpZv%g
z6V7L&YIGuW+(_X=ozQt|X7KvQcjQ4g0U^b4B~Ur1W+O^=+(~zsi$QRCcB`06J-T|9
zzA;a2ZX&I+;EWhW-0;ka9$94@`o&juiV2|L=1|d&bQRc*Q8Lo}&9Q>XDzBoG7>53w
zBkqUAwb7AcLC`~3oa|~BUyg<tRI4U3s~n&zBY6l3QTm9SNogqWe+T`mNJb4rrb6W?
z%ac6SKnk|aMWp#Mdg#I5&T!IU*j-l|tJa6xD(D1#<UtNoEdL4rU);J+A8^$*kqUr&
z9NFq~hFoaFt|{GUtP2;Kv#7nTn8#i}DG=)#9E#V3r%76ELEh9suO|eGGZ}-&E>{$#
zA3Dp?aEu1~%q^~1g!(PwzsM9~8|i$l5ta<?jd&9Br6XROf=`Mf#+sl>&)Hj5F>S~1
zb9M~6++FT?db!}%CzQVX3#~U%HVm+fFXT7*{Bz|`$6f9%!>bOx!tiHSYjV;wc`z6`
zj;QS;)a)RKwXe(duwFcUf{m}LN1awr8+2z!q-@CpplenCm&ug4&DMtc!nviV_6Ysq
zrx?gv!H?3api@&?_v&E>Lam{G)syRs4aKuZoviqI19wm1_Z=9kYzGn8n+!mG3nO=|
zt9!UU*I~i-#(Z3w^jEXmeawSq(b7*_nkg_p4b!m<;O4~R66mAQbVxo6-0j6SS^c2r
z(q^5ud1>g}-vqPBP{_rl@;;hNW8iJ;*k!Cyw4Is-yoq<8dl?dLBrcAfU{Vr@YHnr$
zltpOjb0DdoZM41HL9Q-3J)NeXg#RWbsQvCk{q4D6Y~1iy7Ap_skEd(LCG(^;ROf@u
zBqZf8iO|jTosY^pEgTZwg-zM0T)Gqe?tRuJ?~`lPv9?@LYBY(#F9%nAl|u(^jV}mO
zHuly3+gV`&5`NR$;5oSd*Or8t0|-ihNeR%lb3pqq#Le<Q5Vzx#>gLN>c}d5RF{!Me
z07<EM0nB<(15SoIAnMDLSNYHyyJxn}xmi_!rnR|sTSdj$L0QHmS5hb%^C!bMDuXQ<
z1$$4Z7>!l~VTH4n3?{kq&XLec9tNKitQ5P@Ez+e{^^!~MkK>UEVxwXSqs+G<FSWb-
zCCMuTjYl0gJo{Ev8ut{~cC+Ori_o_A8+0PrL9!qHj9`crb_%R@jw~&*MMsLR>EcV#
zkGd4Y)W5_DUH(Mzjc|sTC%%h`Kg2UMzRh)JOJ>nHCw>SLlSdzr3I(r|^$TlOA%#(`
zXdA>(yFxs0TLor5L^Og+*&T)gU?UElbdVRTMQbA(CT+NW&(M-WP^}KJMNc0-Xd`6Q
zO3>|Tqng@d0W+{)I&odqz1cyv+GxdsovrxM?MvsobR}z~{n$RWce=z08m)3+#eb|x
z{#k&H%&SRtv@*Rjff+2D3T>4uM1l}Yv(_*+vJJRzOvb28<`0aUJ{(vC;KCM7DsIG@
z_x=?Vvw*1u8#ts3hDpt6XE)?3vLg5+)iY5os%c)7yipBD8R(jhN?F@icQa2Az^u5)
zHtk%R=tus~VS!)3tTtmj->{JR%CG7V`zd#2^=~aGO^LC)&EP7j%ep4)3~4}B;8OXR
z^a$GuR$H1}WzzoS>{D3<fLO_c$@xo4_#tPmC28KdJFYpA?S!>hGDRA=dPQp4o<fgw
z*<Nt(mKc=vG4U$OW$^C4c7%`h+fRwu`ReVeWT@s{PUzf||Ko$;3x0ruCOj$bYKwr^
zvo?aH1cgRd#wJp_0inV%ip^n28oSDN0Cp(`zJ4VLJr#?OP`@?|a1ws-c5o7f6fP0f
z^C_H`gIBE{HJ2oKcaSSk<?s8KTbE8a$AMQXr;>3b3IFAkcMMgN!+ifmxP5cYcW1Gt
z>~$a*YOa3nU?ytdbm9K|Ch6p0ckWboANciB+cWO)3;Eczky9)Cfsifk)7Po#-P5~)
zlPAvB|LSNUbFWVVFs4wWd|iqdXYCsBQ~st}N3?G8@p|9Xi+`~d`bmF+X=90b=$>sL
zI8o>%&sNb)(M{1#P6}H!ch9i0I*PgT`PutGLm^J#6R%ieY>p#pt}D!y@x$!Z>UBpg
zhara{R=tK($<;=qfU?ySGxm~5o06O_4IeW4)A-ft#hHr_fVibaLIts-0f+LN9K26}
z-_ud5k0Sw8AN42WPXQIQXELRjcqrzgv{gu=c%z2x8uWt1`1bQhVZvomB(2djTpt=m
z5F!Smt|3|+(^5GsP_0B#bJL6lFtHv$l!LhZ42j~NrZ0nYD5Y!P#4<0PFZn*hJp+PF
z7zq)Jgea#jKx2Fh@zSp;`eX<5)6%VjH&77Z^O^MZcDZ5wuyd|3cmL@5aJhHp+3@c6
zdazeRaM`W(&i>)s;{CLQRFS*wQ}QLu_`AC!SHIh!E60B<jrH*4{N7U7kdx2<!>N(8
zZ%=@CpnTTvJ)`DqlP;yb$+Za=YsbUxB#-2J!Cx*LP}IBOn#OGtB}nkb=63i4fRj3@
zylwBp?j}D$f!8$IYN&CZxyIdyi-%fmhnL2haf7$oW*%HcbA-3tMo91!L~0|tn2=~p
z#TPheuHk{n$xkMz3%(wdLD;As?(9=8FxBo-&Qa*S_FJFdB=ct=5x}7rvI8G)!md|3
zh2*#jpy}rV+R$ti%0N1?h1;t8gC@5(twx;QTq7ZBgngJy=j}&pvMWsMupl;fd>A5Y
z0ocT`APe_&^#7N}a@md>nIIM*w3z!Wu|vAvFdwZqPOx~V@U=bOjNsT34Cd(RoPLxH
zX0Cg`i4pDgo-*vF{IjRK*`O;2oP3m2T@b4G3g-WkRVDM*lZ?C&o)H^{PPt7ngz%+@
zL9ZfHoz<lg9-_bfd4KAM2sVHEMKYYHUleq$STh>Hs##9?%ii|)iqx&WDP>M3saV*U
zS8VyIUn+)9Cw3DkF~|xIGfTO`{$T&XqwsTNC(`aP_yLqfr2aqTEa-uIz{0$es8k(c
zKOu`qasybvKS>>Zp(-=R8tNM0AJvL!bhz6@;;4uPykQVk@WyJh_$(8bgz5y)E-mMu
z^4iorFkn?Ti1FAR-9)(6p^T?q@7iK%>cJ1H==*HHb=-}T1IEx{P{KeK{$axXKqkE<
zZ0}_uRf1F}0V)5Nk|O0^3&5(J_Qw-;wCN;s%0~Nwn_kyO#R`;87@K5Ez}ZzVkYfr8
zRNWzv6lBw|mD|WBUlgun`@XnlW&Fkq?g^5oE{-ppAD4B8e>XjIyKn2}?Fo;|3TB~P
z^OK<{KlVlzP_yJ7VUoA7PW*>?HH?i&yr@bX{ak4!K$D}^yiAE7VDos<9LUahlToUa
z!O$^u4&v>f|4BOaYcmI<?hfuI%9#XLhkbk5Sq0ap70d!GH#LxUj4q*w(RmL7nuRkM
zcI(d#>e1Wt-?cy)w5J66SN@z==?Rb4O4#A?m((<@R!VBUNVU!FJiozmiIO3QP5R08
z?WrY>&dcUQH53djK-X^_v1aPVv9RByTp7uBnk(S7>R*7@`<NQ>h*Sp*vJAaQlmlYv
z=!}Ng?-*8+*w7imahvb_js^RzB#E|IO5<QFuvO4erkF-4#yj#q>kY)Ok(}YI$9aH$
zt}o4g53D0tW#4u^)_CI?0h`!M_ae)Q%j&(M-%kH{N&$JD+@+r<#iFLBC~h<-LY;Qc
zAxQ|>lBJ7=lmYdyS)2At(p3<WlJt$rTtDzO%xKi`P|7dz`FiOF_H*lWVpif%J##T1
zkch2Twxhm1c5@9V_dlT#n)Q<RH<6%M2to6)`r=y&t^^r1)>MPcguco3?K+fJt0NL%
zhsz3uSpb<opH(ZpyKc&P&t=aA=$0r971I~okNW;ELl;q_U~EoZbFrOIJ#}QC#GxFo
zowIu=Lt=b8n6gbq3*t;?>{-h3N6khdW&L$2St|mwhjj|a*Nv8ew`vXBLoD_|81L_+
z5Za{$Ks^bDcEt5*NPElR8icN<!@B|+8|v@b8sPuE>`(?@IuY=a=4295n})NLEtObQ
z`_^ZTFgd@`s+G~+Qqw#zHgG#J1l!zPSz6kY+m4Q#QjPI8s#A=YW+sT#XS`1eLIKI0
zV-8o)lK2~S=lMfhq;>Z$Zn07UmVZxmgMV03J19F)36cw9M6?tqS6&u+K-x$Ed_m76
zQWPJX*1LIsbLh=Ky8aL(LwNzE{dz`hF{kp5|Jn4L>OY5Ei69fE5ePHNo$VJAlv4ST
z7e~+_sO$tYsDuq3Z>`Sfcyhz4FPXU@jxFOr`Sw~O@bwd`Ut{`*Y14=b?a0Zrl6+@#
z^!@s)H~tFkiu8;$SS5Of{Y(Id-{o6_i)9}50n3e<nFJm;uh8;P66@2=5@l6YuU48&
zp45t;m$B`f+Ulmu4^BY%HHq}Ei5NUhHuf5<#ciyCKf`*fZ2+`zGy_8m0j>_Qw=<U}
z7m%3cP}aW)HXe|hkqZ^k)^@o^!d=*x&TX(BWt;GqlO<{vEiz?@wOlkkHu&3YNR%04
zd4o7!9YT8u*p5KR^&(IYs~)}YbP)IMC^Y!%hoaSAoB!K8WCP9=GNJ+X8fhtjjNFvq
zK<{qu1OR62n}EK*U*gDp90in)J5E$k)tGy|cLZ(dvcBnC<L;IBXO$+HP}`t7vQDsn
z7r}%tVvt#~X^6{^pCg^`0y$qse^TZYl*I1^Lo}Qi1GsO8r{Um<@FbX<x?b;gsDF73
zY2^i{E9G3||1yzN9iMsMUk?sWa!XKfiy*KBumE-tRstU{7gV}}--wso{~%soKbP(w
zPMVdhBi}rukt56E2a(;%6vdHSNAfTbWdjv|4Q86O*e+5>*ALA`{0IY)H1^ET-+=E(
z`*cj+sazqe;@?93lW(3`w}JW2Db{|=g2h^*r461mq5)Qz4^OR(iWz0`jdwZ6z8$if
z1<XBdAFr%1oLy@S!cbUUbBHl{rxVLj!7T@TwT+W*BQ3Px`r8Q+E^lU0!U`*29X7e=
zfNp#fEpTS$Om3U@&oOZI52(4xqFvysSIKu_IF#&*aTci*qG6Q(RzCgRk8lEIw#NS}
zy9phY>;`Tto||Etx<{lU8QduPui@OL3y^Uw@Fh&Ug(%gguqe{HYi%oaSRv$|ke(aU
zlw9ASkv^R-$34RQvSKSx_73-VK^7rh;)Qw1=%}JRMe9;E8w&2XB--0ezi3l0gK1~{
zCi=NhqMoxadYE96BL)RA%D!LXBkf7sV)!>7Xy?RI#?phkv35CjV*T+4OP=(RE#TE(
z`iVf=X6syU6egx%!X%;Q7JbbYWYfCd-3oU`W2JJ%oYj6F`|dM~891Yhj5(uJLGg!;
zc)?vVB&44Cl6AaWb4nW~-mL#0%?kHo`HhSlTkb{sm^CD60px_`z^tO`+ZsFZH1xKU
zd3;&{I%(KWd?k(dq%A-B*Df?>IDoMd<%f=S2{(aCM?rG>14{+kilSGMV@w%eKEFYW
z`Jz1y4%%iwIt+<)EEbm@f8wuYgOU1Rxl}QPP@AC+GXS<|<pVewM4pvNXEaVKl}YZ%
z&s1*>iH+S{2Cze@ED3itlHu^)Bvr8bN}X#QyM!sXVMHYa39nUAx~0}ySb#+LLfKT+
z<ln5f0dfk$xMbcUM3kC+Cb?KBLqGE=qL<5z1Bvot99xliT0HTS+cc7vQC%jIlnQpg
z?f!^65L=a-;uYWK>B322@AKY-;9!WXGD3{N$vyOEY3ii}TuX`$x4pUMCaf2$%x6-D
zJWmAH)s_}o$$=uLB|)NJX5XNK`rcAJ;)#FAEL(Oe@46Hf@T5D@qdkjmzR~v_TRd}H
z8|E%1Eh`!I%L1Of6eP*PGm*>E#=IC(I=$*n6H|>^qwF*O_)vy8DKK-xV4bsv*@`f+
zzur0^aT*%jhyLJKY$VL{P}7-sHq%lvxrorvw+kJO#VW&u!p;RjVFMy*=}PgsoZnIL
zfMybUjE#0}Seq>DgG@M8C{5?h(I$_a>Q5d^y=F~U^YD<>qAU_ZOP)i}_|`I{TB!k<
zCK<{>6vZ=<0a3i~huICk<JQu&wP;9lu?r=1<cu?U!PRLSR>JR!i0#y$L&b>p4{~(I
z{o7UPDpz>1((TItgaGgAA6rlP4`R1CsUf9-zkZe9Zt&ah(Agr}sXL@XYCpFfFL~Ek
zHFr{__-d~tmoF?cgZi#kVv^hG(^jgD^R<apH|ft|=P=Nh#{9~gr;O<wcN=pG+E^SE
z>p0`wQu}M*v{&6f;jT*udo@q8yK2>D-qwz3QWBb4fTR~qLk=)O(=hH(9<WQ<V%ptr
zrD>B8?U-||xk!`IQi}W#<*@#Oshee1*czFtd2+w61V}gG4yKzJsV|HE$lpf%fF^a#
zxNID8V?*)2_#z<f{~nn-y?c{VQgBwq2@02-EK*$(llY8jTh23^^Sbdrfidgr2Uc3}
zE;d_neEg#Axdc22ujFfcML10ojkz3j+#V#Y${9iRarQ~<6=kq=d79CjS%u9dhMc=+
zMmdkZC60l}F*K%-{Ap}|lDUBW`)q$Ts|^~@=Ml^_v1Xl8r~Zhys>YLyu&vSAs(zd{
zj(@+a{Szi2iltPLYeC>A(MCGMg?1uXxkT@^mESdapAg_trkM8~uT5Eb_z(fq9d3Ki
zxem4b7u+r0CIIS`#c<Nr&!mICQ4>!1pH@ii+f<PGPKz9>5snh7e!zxx6Q;5BETvi5
z#?;ppNc-6S*Xqt2o+KJBEX@UuCK~*p-6~I=rNHRL7=TZ&^$q2f712HW=jq1?+CLk`
zIwWJ2=`4B9nC41e2P5^q_*?J!I<YDvhX5uH-W!q91s`*|w_L-;-uWS9o0|KSj#EKT
zI1kSn2*?og*;9y5S{(eVo>o~|?Hz<<9{hCwPl$--^#6Jh{9s}R{z)YKhPAP<1D#4Z
zDS(7S+{u96^DqAolM<Hu@T73M^>ay>wjiC<B~s4HZL(emMgPqE*t($5b#lteN^znG
z3VAmW7(D^{{D!wSHhGo%9vGMGv2F#gVHGx&?g+Mm0<a=Jw(VT`b(~4FJDqVyE|Fr;
zE=)dbDTvF9`CkkZYj<at-Fl4}*k)X=wwhc=Z2&;m)hpzWisev5C5XJsq3(^PjogdM
zm7jLs(p<lS^s4M<NG+{19`b^QqSMR>y&Cp%b<(kU0f8<zxd|Z9XsegzN-heV|7j}3
zVu9@+4Qegk%w7!Qu@k<E)nIJG?5nClmPCSuOl<kOe#FQw-9!+-DtrP+D6y+<v4pX6
zL=WIM$$T>|Iy$=c)!*g}-c8M<O0J5Jx}V{_RFFAFekGlgnv{^p<Fsf6*tg?tVI$O!
zwiYYZu&EB?b86?SrG~3wp|JyhnIW<tbMtH^<Jb?$FON<&hhP4sWQfK=3;`e0gRNzV
zW?ieT-q6A`8*-UbOA^h8Yz8q5G$dJ*VfO%`wu5yRR9u#_s=Br-`6f<Q3dof^HeL;N
zRYG;iAd$6_91k;f#=hzXcjQ`FBHh8Inv9Xh@)^cNlu#N&r;%bcGs%!65E--4&&YB_
z3`GM5!YbO8PczBp3p8w@n*Tul#dvv(t7M1)2;?FLIZj>}<%m$Uz5T<Bk)??ZT~q*S
zJAt&$*+IT%k-ix>wwWR!qeQHPS9t{FmJqu^Z_vMCDx-0JxEOcW5Bel2a?ZI_>N$x}
z_K56DvFwV32xBNh-^)OYMLqBf&s2IBDPqF#+<TG{XmlP|j$f{>3`NE-Y&TRvmjSD!
zwu{RR>l@isi8jge<Jgg>pQ{Q>+3*4M>zd}+C^~@DSWHqg$_^n3eEIJUEMn0K8G>HM
z_*?tu4kdw!k-0Z5-^_zz<qzn?6D|gI(qqoatF?VPIGM;fR~4ap)MTaUSNq2nI(_Gc
z#V!a|jGEf!L#KU{9EUb&O%m#vSDa7LHMpbaF{_G1_HaSDOg^m<3QaV2BRxP#8D){g
zmNj)PE~=M~3d6J_Yo{?%^I61LWOw21bi>Ox=LQr<xER&C13mYab*fBztT)AtR9K$K
zENIM@&#{Kq6scYdiR9-P0sSd;vWjTCu>V!Jy~E5XZ5l-?&EJx^t9UzTXVkQEcP@@x
zW3&`n?y<QhDI_S}U`>f@M`D0N^+>_{w_HbJraenOUg~eH8?65d!<rSLrdVQh394Ah
z&g-PrTp$g#vsRkJMFwdEq~`jAdmTwUYMZJY)ZsZBNr*{Nqs^of2Tvdt#ztn9u{!g=
zB2b(B!692@5?TWARfVX7I*-eq!BW|&NUCAq1lht`pMGc0V+NoS?J2;ifAy!+Yu7a=
zVFX&)2Z<OhUGV_lwF$H63N7El-(MN#JuG<zkDAHWp59LKJ~(J;)>F9o6Z|Ft;~t49
z4`dQL@wJP6z~lujb><A+W~DXvRPV}73^PJ?Mqk+8tOpbN84=-V-0d?~3{v=;5-VHX
z1MisZDD(1O-yzEM@0TGEg=tdn+DVM^VQ0d_1fDXYZG%+t>ubx`ku6HYwBP_2*PP_x
zrqbU2Za_vF?^7k>W%bqzxtSW4NMbN^(qSS&`T1VJbk6YUzx6y&D3}A|KMvUx!2a>?
zI)3*kPocZ>%Qarhjj98tJ51+oY3-YVCc-BAhj5R_L{ajrZL1sa9~c5@^aCE|&W!8U
z*KrrVzkLzn#(uyCX$%o|s}z>dzGmAmOnfqqzg>yR$r&33D<BhLC|d`uUW4LiG&oOO
zdYoSFk@y|vOb%9>`U(u<VH9pgn*E6nM}Us_PDwA9TPs@6%xC#&UJa{CAa;(my4o4<
z8=Dkmd+6`54N4gzX_u(^s%*zWz8m}7oGNo5+S|%8t+^u%Z$8MDg=2i&32ZgRq;$+<
zng@b<J2&$z>UCwRD?E*`xkQ{HL2w)OyIs`!al5ICd}YKVH7S~GLvEaGD(e+gGJp_7
zmWFCju4=g}U1`9)6Z;~8x7)fNa$<m0$yFQG=CDrw$9n(za%>jt#86Jpv&r90TA#*#
z*WFBns9^6TT2(@{ZyQ~wrgIi1Oo`xPWSqgq`|(5%r$e`exb5$$hdq`ui{0N-0kX#0
zQ8~0OQWu#K2{N}xSlTpNrM!Nn9e@zMsn{e{wdn*<<lm)F>DkCJgF$MHD2-yK`!C(&
z4j*h&PJd%XVYs=!cjOL7J4@O_eMy!x3Hu=w8`PL0f)u~mgeZ!f?XIuLaDt_atHv3B
z#deaH)g5b&XI*LCNUfgDLMe5pc3Ot(c<W(Ll((P$9&t-}v#GO*XKY%P<N&zpH)r2)
z9Snm}gm%`u1vCje?XW64djr~Txw@O0#d=Hgi+!#|PN?isGF$I#ar)v9A<M3V8Q~?Q
zTYP@XBH`zNiv12m%qTuNNtXTT`<%QSJlLRM`^fXA3`=D)_8+n=%33VYWl~sK*qY5<
zeYx2+A_yU1DDML84pb28AHYddf?4as6C!+UY`9k@mI+KI>fuE+ip~ynVZ`Mz67dj7
zFhf)D^HG}5w9_b#O&=r`4)c_3T_P3e22M#E4HVm31l<(hVX|fXkV01oa1KcYFC7>f
zDsyL<zI?F5@()V|su!8k04BIPgm;|vSGm`6T|@9}V7x6%rm;hS090p_!rJqC37p`^
ztUK#=mT08_^95SBj3k{5bk^oUfAg}bb9!~yV>j7p9x*cpD<OePjfd;qY@(Z5N^^~A
zu(b)J$-$6%!o|BF(0z!nuGl*QreoD!k;W4xW}xV1@6qAAqO%;-bM@QdHOSo@<M<la
zZn$Fa;Qs-IKzhGg#{hQ|8O(nj1NIQY4DNRjLdS!i8l#r;WStYThqy2sVTBTAuM9w0
z*3xfwuVQr~djsmSr(~+q3-plDkMw%Z1D8#UHKTE5TlXz67xcW7vwh|~k+pze1l;bh
z#9H<?n`bi@!-m*a$u;JZQl{{OoCGV(rJvuY$zJA?*X~cMl7?gL2JC-jP^ug3CGWjm
zt{g$8(a`$vMn)2#4bmhMI~fUDBY^ZrG7`n-E`x>!cX$*=`-gD{*ODOgcLER8N|0N_
z9ScV>a*Y~Q5evTcO{RH88kAT*(wPVeWBNA}Z{t}E&d6hj(hDnVkhZ{m)ZI5n>?H?r
z0`%(%UC)Kb8IUrH=lXwE@dr7OVvMO>E3^=x2xYP-qO}>bl=PhgTX)bJ3^5^SwP-#)
z^s}{alG@pstlhk~wOmndy7wxfV#f$XwW33tZ^}DiHT`(`)+qj*d~1aMOZnDl0koQL
z&FlyBt=+p2c(<Bw%R01OFyL0}Oa&3k&|mR!<&@nC{nw8^2K#@b-usWe6Bwc9ID$Jq
zr0jh|aNi*f#tM)0L2m~XI3X9_xo)4*O0sC@KKS(jct%1!xz6z{kn!=4s6DxcWgQ#|
zu<*ZU_+YhApq|3;0Y|y_AEEe-RS$CfM%oWT^5=Z8pXE2J7ic~(K!?E>n`hE|CG2#E
z=Y5>4-+_n3`X*V5uetfVR>9SelWz10&}~utet!Ho>29I;Y3;6mL2$$5<9x@no=2&@
z){PCr&zNrSi-2t1c2U=4la9%_fAG^}BW34{00wQU2UdID1C;-=%6Enpe*-nnmw^W-
z6a+CbGC7w)@GKLTPt!WTf7kfe+wU8hOqar&@N2mBf4#nK{@WBsGzc<&AZG*^GyMH+
z6JDSqVXPlK2!nq3{?G7r^G82w4U<z0lQf2?fk_)d<OXBBfbeR9iS3o7r9_4Y*#Trn
ziSKe7ohZGLZL>5OasqhDwcZR1?)kxfy(e0MU~^9hy|_DmN<A4&e@ZrxpPmKKiSMMk
z_j-CpuRM;K-XV@5g24dfC=yN44DQ5hmxeiO*eQO3V<ClPH-x6)djPw6^5rvw?d2NS
zAiY5EdZ)w8!NLd`PBSZhrficjhe8P8h2~7?Mr9Y`B{|#W?r;FW2(+#Q;I(aKq#zH2
z(IiOu(yGAQVQZ2Xf3C$iw%tw7BbIh*i`x>S2>VNFay``AXm0C!!>@!|do9rQgJzV$
z&qzdXF(vapWmwL(v@B>_7L*hie9AwftIZR3wZuEbgj8>Z!p4QUH3}!GZ(U`?MRQxr
z#mFt^4)u~RShvAT0iciAu*f%sJkL*<ZjIeH%C=_Ak4m=2e~6CD7P;0mzBbj`nIUDG
zJDzEWELfC3$^0pGZi#Z)GBG4^Gz`K#X7Xr_a2Ycv!SQ2Y7fehfo+SFa`sPc?f~Anj
z{LdU~%k-2|KQb1jlwZdrD<NW@Fi<F-fh<WG7Dk|6g0xt6^TNSALrS=5rYQ<KC0(?}
zP790XW^t$!f3ylKN~5gdv{sug(Z@ro4;U=zbm!~K-UQ`lgV$^&;q#<sg1eI|Yq2~-
zl1UU&>0Xj)Exf|xlB~a<A<-Bup<Q9>7W->}pCr-(`&HoAy*Vht6HsVbyIND5&f~E}
zjx!&1B(%~w;oTgw&?>eram+2FRt*+Y(>IMwm8v9De_FgGcF5juo4zos(Fvz9CKK(l
z53vQ#R+WmpFXJ?Mu3)PV;oR62tkjKX$0VwW)$0kIS@xwGqc}=h$Y@I>mdK?dfv=TV
ze(CCCiZK^|)T${^vf}K0UV<dYr-<1lXvCh`f*j9iJthz0#w?EM%U?9==vU60eO1C{
zZ?avfe-K8LV*k^4mzI^%h%Ks_qXO8C%s9_gskuEtAnSiMQ`tKh0<~kYV{ge2t#ljn
zdau=66H#1KY_(iFTK=^op%;$atNa5i0&w`C@(<n#bG9TtAJnm|kk{!rBrD9kF^;t{
zt<S?lU##zV%POq0onPuxxK^{2xld?qdLZlxe=T;3n}TSySkY>!2dk`6YD-^1r>D8*
zF6$4bQCyKrxvF8QBI_BS-|iA?mYL>iWuH&Rr|uUQ?5Iqt#>wAOJ6o%0HPuEF%b6T?
z?Ca<E*-M-~JW)r}0$fBU9B63Ys0OGid8Jh5oljlOITBM|XmTtHIRp7}Krm(I;_0y>
zf6KcOZF`druELAYn9`=S+ZP_Faux+h*Beu^#@@_p?G^8m%(-qu&T*WWJuA7Qh)fT`
zA$@-e0v=RAZ`K{T6+jB5)D_$cO7ULdrn@ij^&#n)HPxNDwRPlP+^Q`J@bDmR)lM;`
z%=zux)93f2`qAXtIkv*CSDkt#Mm`I^fAh|=6$neZAF)C28nV}1F>G}}_p6irdL&ji
zOBLa-_qspfYp${g?nfg0&ZL8t;UR?I8FNR%Khy0;6Mkpb9SL7^AyV_Figysc@+dej
z9=w^p>{H13<nO&ew4GD;B7a4W`Aj{C{1tBiP4xxzW*?j805u<*Iw5fMJI*_Zf1d;z
z-8*%3TMfcnwSeJ0-<S5|lab}~+KRk64c^5bWB3OTi5Gax2JzZnV086Kg5v6A4gCV*
zVqde^bXuEs{hj<kr(lb99MjpZs_i-`G@5hRn$X?z$(6C{%MWmBp_n-k&5tBvI+H&3
z)S@%yjw0qvw;wHHI<xL5V!RPbE4wQ+45tqZm77W!nZVwcGmGhaNwV$bee#5i2$(eX
zo~uqe#lVW>e|UK$m%(>769h3aGccDCd<_$q*x5S2e+o1YzrFl)C$s2McoV*ckMY;1
zm%G33>PA2dGJZ0E0syB!zTbrxs7M&=Cn90JntuE_eZBj6q*{0lpHXd2vD&0DM4j5e
z2qHHbqlLD%sEv0oi8qMP{K+aq-W2hs2>AJsHjS)SWK|I>0eK90K7C3m=GRg_MFmt&
z$osN1e;ERS$`wxJGZji)!gr*^8C_DLKqgWD{P;EvBVRtxT&xvFsgaBN7@qDx>I!Fs
zf^mV~6)V_A-nh*hXJyFe^$nvEXzG4dUJx|K+C-DI8phD)=a|Ohr%P+XDUi8X9}wJu
zx%Wb>s6z_2IVofy0hEY#BTV%DvzKc|1R#Z?e{rW#g)Z;VX4|2T>6TRjva$24L3N+<
z_TL~Wdg1k`j$pCD+DZ3zcS=gD1W0!SDJ?NU7zh`W3%XRSTV>Ca(i>qJGkaq4d3Mz3
ztb^?qSkO|CtY|735sie$Y`#y~JSnP-->AnH0<s>~5NZ>}Gx(85q4m6VWPDB3i!o6K
zf2jC7yp2-1G$WebJQ|I!{-Of}N>OsmR9vB4_#l?~xC(W3UObeRQ_@OZ{aof`sUoJd
zw3G8I{e+uNU#N$fzOL<bQh9u+CR|qS18kwz?PK0Otvj*N0Y&5iY{MZ23)gBtYRKnv
zQ8Sgu4bm9L-QM!D1z5vSrmKrq*bFh*e|EY%JgCPL`b#UYJ^eBo&t1>E#3R#!W*0F7
z2pXeflr%fq!ghfIXz5XM`IPtSvOIeJ9s|I7B+1@%NqCS&zIETyj)db#H^@f|@Jl+`
zIu~*R?VMj>-jSYdJ<2v(S<-d##Doh#&Or!R{5VZIVdP@;%jZpvhhPtO6lp!~e<V>`
z6&9C-O#?kd?@|NIu4@-8BNhjPL70jf_G-HZuQ;zT=WbgUjKgGO5eIPIW|-frAZ@#I
z)3a2-M1>y3%4Wo2#7XUVdzy?hF&Iinn_qpickIO!k=N=ulQfN^Q8vRdl5<S+0y$wZ
zYsCX`wj>Q!*=uPKA}+%W`&h=Je}Z0{_>KWJqG#l^IUsY@y-d%H+SzkiJSWIoqu0bk
z$I0!yVQmj8f)gEQ2+KL*rtS(U;{_nJ`F=GD1|SVtc-!Y4HQX1-&%rim2(ruPKpKsR
z=4oMhte2j1b>&tf2QE^@O@CuFgT+QZJXf?2+94dYQ?2pBVhptso;`FZf5rfWS9tYe
z47FoK@GZv9ihYY#^P#*UF+&MLdOT0G!vknjJkvEcN_loYnK09qWFZ}X9%)mmJjcx2
zi%-K7Kh<@R-+XLgaX}T^(ddq*&KZ))d1Xi@^RG66OesRUlg+fJu3t%=0?&{D0EHwA
zTdzAd>6~qTFIUZBr^w?De<=h%6<9g|JT|GXS>XM6_8eh^($y?GK+{Q|JBfp~AR_EG
z)r_sF*|8*QZY_;ZDcEt2L{~{3RG2@?)j3m6U<itJ@tAG&kLRTZU>Q7c<aEQxs0leS
z!V4nf7$3CO!qYXwL$BC`M9}z%O>&0Uaq01N2p|y34zk`V0)#S}e?R`80Kl~u)7Z4h
z(|Su~N3mTh-8s6vf5i5nJ&Y1vyMe^&^dpaulveM<%ljU=hcr-4rB!Vf=r$@{uASVd
zxv_Q}+qYF$){j<NSX#)Qi^$hy2VjhF61uj{Tx74cZ4$EQ5;;b5+oi>22G-cn<3kxO
zt&uc)Y4?@R92c`>e@iQ9Oa2v4%LsA`g8>~QX^oblcBoBbzl&=Mk<VL?6oyiAlfU68
zgKAuTE^q4?(cFT!Dg0~1rc&5%N9@{oYvT8O!j-1&D5-AE=1G<;vEhIADFzJ4r{6XM
zhepttVMSSv2ZeCk;0;>F9`);kHz?+&;KtGWjLc{=a-$W9e{K-;R-s#g7o^M>yBVuE
zdAN4yR@MsxzB|1Y;7i@TS0tO`?UFK%chmZ5X-zC@vi7K93*TOG!xUn7FUHmHmynr`
zOx|?Al&*n#g7N0!6{%Lf=~$c$Sz*=O2|TaganWU`Q_<fCPJ5S~T*7HC42BiA=`bu5
zf+l7hMIojBe>TY;*DLlGgD+Q*=P2g%$obmlnRDcUOEsUE*$mtwS(1rC<kpR}dnmx7
z!_B|$%&BQCum79DnR%P<Fe|2Q&g`Xa@h!IP4OmIt(yU0WB!O{FrBKth^y{3yYNB~2
z%*B?1vrA6xJzb(WH=b0SA#(uZ?4Gg0K<ElL$+JQBe>`Xv_wKLz2j$=WexKLT>N;y;
z*RoyrD=YkdmzlGi5BAKy<rUVhuJBq&hQ6K+$$d6l;d<(2`*)X3Xa*YeBOA4=0_2#F
z4lYvxa-pqU&*`6{>W$k7_7@l|drqNK5{Odg#I&Zx>$w>k&4~(*<Ss|B9QYwr19G)b
z7JB@)f4`k8-set~4Iq#YswXh#x=jsrj-$<W)rlSDJbS<2xp%l=tWe?RV;20~*ImN{
ztA&#Nij(#Y?0cby-!9s>Z*Xki){Jb_{r<&7z-u)e!$woocwteseV%KbOV>|@W|VRt
zxcobSZ7*nxgV6DE3+Ujzy)++d#ICv@di>YwE$(Wrf<uKdDEJJu!D&N2c@c^vw4{M~
zO_FS*eQN$0peb>Nx`WG_{{SqRPj;1YI~A8vataieU+p@=fB54+U;h3e3+t6nrG<5I
zaoz~zhWRAE4DxA0n)U7+U{gv1B7}LeK~^lXW{`1n#_u72x45WP_}H@Oj_N`XfGV3;
zMM11P9*kGiGWDBll+t5MoBDs@QA9WW@*(p_zgk~^WpNGn>J>$j!n!hhm}ymQnIbKX
zMQZj`xr}86e_+~WebNWi`LosvO?f^ZERZo6-&RlaW7BPIuV<M;x$=!03yvtF)K(v9
z5N?Zt$2L6)+gPM6TUm`$LIzd0=zU>~$my(7TWr}lQmHzTDlXfUGs0NETAR0rxKeu?
z+AHCBa59}ANe{JnBdUsBXW`q%s9Zp9%*+j`R%_Fne@@67sWq_a_@U)$%M2saW*)(J
zJ5UY@j*y*+<?&l4R({db3c4FrG?VwXuB_5-@cee>x#tEQ7bz~9H|xuernqTQRtiw#
zp(kA>{ivt4b@F<EI}EW2WodJWa=$^;R_9^>y|F@Ww*{HxtnQ}7B-ABCRvtxCZW2iY
zTsF8Me}y+IMOfK9P%^B<UD}+$u)y9a{~mhN$*m97+=`JLGvs8BetB8#h7I^B?egi{
z!ygM`$Mcc^<UtA@gI)f&q;BNXBF9oV0PpZKYW!RN#DVXA|MK~7pZ)~`r~Q|P%Ebs}
zA%^8!Zd!Em!Qu3;rR{{#Cg=r3l-e2OL1-Ote*jOc9iNs5gYas!Qr)nx)T@OLsIP7^
zSmA<Ex)JEQn1{3?_cFb&eJ69(NF<pm8`Ppz%3NyVe4a&3dipjG#aaJZ9zHJ*AB$G4
zQ=Wzp6<R4e{M*th(gu%!*=d%$bP!UZN6kXu0EfG}**QpC`V<Q73~5S<7`=~6B~=CH
zf1XZg?!u*b1@7X{xqn9qlc3mBEz}B&Ok1HKl6gYWY}Ri3{^?tp_5~+Bcb|Eg1L!CK
zS2{QqghzRBtrXVb&TfHI8P20QT!P~2d<7Jy!~s`-x~4^t<j19BDT#JyvVFW&9k?mh
z+8fj{M}bWZHd-NF)xu7M#sXlFi^1g_e>4st6pCkkYHjjQDn(d5<m8?)c0`x{xD`KU
zKv6+G_OTzWVY(=)`fF1><+%Z19+^jR*i*Su4U${D)q*3F_^-6d>$<|}JoCiZ(o2kf
zH-+M8^8rH&Dbgv|Ho44=-d~@0F-AuaTG@Fg62_evEpTU4hG>07Gm805yr-zpf4Hi&
zK7ZIV4WxpWXl8~|1Sn&V;n3DDKsuXOu|>8nndN9{Pzp3S(CLGljNBaDWbA_$jF@w`
zGoe}mkDS7&!E)q4VTH*9rPkBgP6|`PzLD|LS53IW_5}yi4)cb8Y0e`1m*3KV+`S+*
z!Hu@)q6yTaRM{{$fBCidA|-4%e{^v}aw9vDK1UhOiV*8YTi3Y?%T_X3AG+s&VtA*w
zvTlnI=bC#~>^yv@QOnL~u<u=7d1ZwMRE6FsYx?z+4Kx{$Z->eDDO>EE2AKntto>Ya
ze47v5x!7LLVK2su2qKuDjC{R6S0mN_2q_k}r<M}1>0nGj8l{itsfFhBf5$2w#IOln
z5^XE>r>F>n0)^25nB~uZFP|Si&Wn~En?IXbNaxTD2eWW!nIX9n=+6J3rddb052${5
z+LG&v%c{az+eh5W)U@{=o*FcYx=AEiqU)^Fv>b};D+yU_h=ks^=3`Cc!cERv84C1Z
zW@=Rpk>;^uQO%_S8lhWae|tA<lUf)&auavg>_GYMNSd4UpdDITT<L*g45RE^8Y%do
zx>5-Pnw>B|YcAEQst!3rD?J6JO6vxcGSc?g2OGxIIi4<0UD><QA}OMz+FgyPAt*bu
zKAG>UZMLDY`3|hTT43mW)$4)j8t~xeQkL}!;F;k7H$#MAe<~bge-r1R*qkrJgYm+c
zFfsa8_@G@BCitHO5ZYUj0)X2cL(^IYjI;?H!WI9DhchW%b(YI}_Ehv@D4pVe$p(j>
zE+|r(gkLcWALtoui|m$?G@1NlUUle679Hi5N|eD&8a>xUrP<s>={o;=%0~~524z>2
z<f|#P5~~vtp<vdKf7v{F%7ajHrOzKier=@i!Mqvrqq#b_Mdh!Aek~N}nW<FAt<`70
zstWtZD+SavDs;(FK2hZ;=>`b3GVLlQRg!4fr+qu@F&{~yz91p#7dNhcudOy{%6_%e
zmmSkh+hq}N6rJyByH10;<_nzW&@QGnd<rm=2r5=I+;oi>fBoYAjUK8o+L;1uUn*(L
zOsVqK`LJtOMj8Z=Pa0?5Qv%FIG@ap*LcQ2&@sdoHlLH?|Mth=@E<tj_f9mT4WtEF}
z2^Q(vm4wAd?jCaT)jq7a=4+#&_`tD9uc<eTy;Rh~!4HI9biER_{!2qI9Lf=^;fuD*
zJA&vX@%)a=f4@~YrZDHU+We$&j36Zdd~Z0m*5n;>dW;m%X{$Fwb}J&DuMH~qL-rPs
z-iquhy*4p1)g!xdYA(LJ`oaUSd+I8WXP(11@b&ifeK}dOPl9<pPHszUpr>w#B+9&3
zL=ufI!Q!r*uZSXw-0LV}3%VdW5xiR!Qo<cgE?}@ce-pYxgaJMZ)XuI5roVR{0<(Je
zoG6~VcTifM&(6743)<T>J9pLZYptNx+S6_R+TXs4r#D2jtU-UrGBzOU$JlOl;?cy%
z%P(5oSjGm8JjZxX@Eo1Tk>7i%#iA3xfN&Wg5wc=56N4Q<q7tmWqJ;0oOJ{^uE4Z|n
zXhyg@e_R5UH~sXc%a-r$bw`jmdL9H_>+KSiuu!ih%F8vc$-=>SD(#Q!Y*Z=Qjk6P4
z$7#Y<=<6NT?rVa_c9*)YUKW7^U0<z3eKWo|*N5MxT!NbZIyv-d!t0AWN(!(uGxx6T
zVAkZEnTY5(mQ1@BE1P@SE8!S{*}$a10uw*NMrbJwO^mt{%*tXKsunG;G8!;S_Ao6-
z7420?JD(|!%&sazuQ&r50T|S<%%vlsBl-&%h7WIoJPo%2`%&!tiO%W&s5JDK(W^le
z12Hl)mr*q_BbRRCIt`cMX9y^N-CA33RLK#3_pfLvFPjJYba!34I$9yIOV}W>3IZY}
zv3M|MSiAubvd1g(@AFk3&l$5b4ufZ~(Sk5F$35M3udnKK&Lvq$A>;oRGLcg`g`(Oi
z4b)nvY7i=&vY;xRYN593R7s&x-=$>y<o@M?<)n1rB=+Ujzf?G6E!IhYJgAyGiEmiz
zr<WQ^&V4h=p%7NJ&Jsxr%?~V`ia*0|na52Hq?sSZ4b11hsYocbGlcJkV$$~~4W(J<
zkZPcuJ1=AbCyi@cAToB6Qy|GBnFHx47?(h`j;9e6l==?Lfph8XfGI9AKGncn$EP*0
zroICXVjDS=0#duU;E!#8RJUNW#$DfRAWi%<g%R?2T1`R9$U4*_rS%=Kk8tT?La(Ac
zJ}p6N9iLK=X8j1|pt<xTU`{0(pSA!go1Z2cR8s$%RFES(5KF}%P=cY<EeT5$P_>H#
zzX8=XRzVtd|ESot@knAVtk_8k6_)n1Du?PScG5znI_@B$a_$^|z&EOJ{~A1?s&vwF
zsC7Kp3Y^ipLX3iy)VYNjjYh8q&&JyI2<w}UJXj58qqpJ9)<zznp_Zjv4T^2P%E=sH
zlfD+d+;ohGrXV3U)Iji0gb^tGx=J|!#qP7+g|Wn_GuYt1*3YMeom1+}l1rNK>%PRC
z9=*_AB2lh=J&sR*EUDArHK}r^@vT&w9ysmtUhbLt4r%;uKhashZcf(zxJ&PH!869_
zrT*AyNz!TGOW67@?!nUf1|GHuVLkZ%`{BiZKg@;)r>FD9@ciP{!aqGZIsH!<ewd%V
zo}C>gq?z<`_}lP!`0+50AD+#Q7l%YO5&{V}jVvBeITFEtJGI4R`f_*>u>JGlr}>Nd
z@GyL*4WR@*lfcnLQm*pn`}ngH6Ob-?LQ=MvU{ZuqbTg{d2>I8UsPKV`zgiW+h^(c%
zGXd$NM}d+=4>!p<Ht~p*2sLkD!kQ2~fD<R8tU2vL6V#$du%_yVhTJGtxPfMgh>_bd
z2O>M>#z<j*U<T1hm7;sJRQO0KYfO+;ZHdWqS<Wg-Z4a6#Eg~rh6w*A_pa`!<NJCMG
z*LYw<bY|EJnnT4)t=aB8AT38pbz%Y{M#|SSv7I6|pNl*Y@+5nSf&?vvqNiOc5oyp1
zO>8Z{JorS6;lZ@~@#18D3V#0a?6Lp*?(Jgn;ruUu!|=<=;_bz&cszeU{Ce^J!@bwD
z--e%$7H|LlVe$BMF*|#6bUcf{z8hY>o4*=<%dw=Ghx4=J;b^fqJ9%}nn4RO-e|nOU
z74G3@BZzfybGbO-00#t@f^pJ_5_gA!RS^hQb)#Tk>NTE95iUsLkg~yRYD~o;PKs65
zLs$WSB@8+RIBF;WfKq!%W(ZjZs6c9r?4G2^AHV=WT@ht#jfs+wYY8D_0CfovE|-0T
z@M!Lm5W7HVxe1slK$wL5vMHlFI0MT_HZ3PWyS#^F#snXmcp&yrZrAd_mP8s>8khhU
z)2us%Fgy$A)XCK|a;rAGOIOduU7gZyG3>Q}0eW4^RW$-8-l15q*FFC!E3qCPdUw(w
z$+O{~|N1xZZt#xe^y1ySmz!4r#YF=k7YOM<WQ(QL3{h>{EQQjeAu+6$hT+HgyZPDk
z50EmDe<2J{A=&KgG?ek<Uq3E>dcHVX%t9U?Jer>_aL6NoG_)h!PkH3n(fgtPVQhYX
z`uX?o4^PkL$IoYr!{O<}N5hNR$HmJlMt^$r%M3g~)2EB`-eIJ^$Fte_{Nn6*c7E9e
z`LmyAuTPGCn1AfPf$jy6A6Ve&(HXvrWhe`MS@)y!%NDS+aBw&_fzJNe11_mLQ8rO+
zNpYH}JW*w$Xrg$crZadtgZmjTcgrDv3Iif(pn;5l1LW8k&g7u1rga$LrN5xBNB~O-
z%{#iyJpdvybI`G50tpHUtz65*R{>>&GJ>QS(3$kEfdSA<^~Tt<oZJ#JWOh_U)?)9Q
zD^fzn=Y~?3+PXCgnk+#lOK7u+fs-lkj|mTHpD-9lG*3wGjn{2~+Nx&g{~I8GI+{I8
zb$=WPuoDV<(P&=ZR8eC$Xu%R&_Gy7aNk&>4?~e!e#4IZ&B*&E3-2vhg+LVTd+0Zco
zsoVZ|D2M~c7#1ODAn)SVIe<SR0~hpaXb|9%s2V$!ceMKv+MLQDgu?N6iDZ#%PzNE}
z9*`^mQdGxl7JWWgkoP3FZsw+cWmDcO1(q$)1f|-5ed=!D$tX%&H9)yR>Xh$XyCX#3
zlXl^-k|*5)oI~{X^>wQheJZw8t;4xozb(yn9K~006w-_8Mi9sAa7zJ>wpfN_gcQEj
z(rg8(CV$faPHi2)soo60)pP~G$pp-F0%ke^Go65$PQXkjV5YJ=2)hh_7X|dFR;GYa
zG;+w~Uq4`c8rH5v3~GAhZ!KhY)h@e-qk=EB%gDHr`Ri+!x#zvrNJuCq`(r8B2SZ3D
zlX6Mx#>w*Dl}lwDZBHnrAlozjQM60m0hQifQ?>r~Ks$CwId2&70DV_Zl(Y%8#w!Tg
z)4tZv%?qxq_<yBl!!l8S)cR7kS<3#_VY^+)n=h-|EG-CULzVfK3s|k|1+2;q3s}2S
z!0J?@>QtgC@1Zb}MuHC{xSdZs(U6Jv<>=2-UGu-%658OBxiH#7R8`Y`yf#AH!TseO
zL-$3{o|Dhj(TBA!ih&(i$Z>qnjW*oiYSMdQI0ceiwlC5{ci*aip8<Yf-F1gY{$gMH
z9gQ#d<HqZL+{(@TxYa9uY!m3R$)j!ZXq!NnP17)&reQWs+-#b-*#zv`1ngSg5Eoy7
zm87EJt=eKyj!^L}7M%C<*C%gIj*oobvF2#?zIex}x2NUicWEY=;vIqBqF`?>PLEeY
zRciO1RF-{|Q6J%dJ6K?j=683Z!ShKtGST`AquQ*(oeKFi2Se|;hcxEp$wA=Af@d$k
z;Bg%bC~Ar|sgj}W8=<oDjxedkl&%bOe1??x?u^R@<H|mb>+b{ItV{Fy`@jwJN=6qa
zTG0Tu%+P;o`!nv6g{65nvj7`NLJDIPi7c%ork!kCaseEFT)nUInU-X_JKJi4ZME*v
zwi$@Sa-D3OaWu>AM(&b_rE#y}Veg^aB&L;dGftP1@6Nat-A%3SccZLEpKC3yteAQR
z@tJzrW~mn^zw@@mQYsexV(1o2ONN}Z6|araW`1>qwgV`mv_2j4J9Re!WqxCD1<Kk4
zr`i%u`E|j6C6y<tOcYJjKAMorBp`tPhE{)>1pML$Q@Py~zhq*W1+HUa-$~x@X)&$l
z9O7sV=iXT!Xmsz%Csz`7uy2Y!WTomYomWwd@HwgtA-~~mUv6mda+8u2eppJ|O|v8v
zv#y4pKFU(P&Ee<iyAR{&NbWr;Wj7RG{BB`c4I7Jpckpxv!UzDK@;>gIsdxS*_(U-y
z@YrvfsNc$|X~TuhBIy5WljIk^G_!s>1oMm0TfB7nblGLO@+*>6w_1wqkj=JP$|jM@
zHY@$rvtLTzVyTA^)2-I4-;Ea#&ly?o@`Lz?Jbv)Y;Vs{2fJS9~faW)6x2SP`S^v`-
zM?A6y@IM+SEg6+@I~9}DR}`22DkmwIDfc>?0qvK?_c|kgXX5zz$-$}xv=>&w$XVG;
zb--@Mu2JdxMwb2VR)M)IZ&4C%v7yC;xA04CL2mKoP5JbD&?Qjo&{yv;*g4Xf!aI%z
zM>Rxi1an%*sxvQ@_$R(u+VCEpcu1Bi$1y9B0Rcqg$kAs|W<N`o;ZOMfn9vufox@2A
z#UDqSrr+6rAC|4ZJRj7Wyb%^~f$LFivc};zpFQazdsFzX^Y`9I4{5KQ7Sb-XS9!D`
zNkDzZHDIdmvL%WaKkaUz;>^$c<KA56i9sRRR%Og{=E(t%w<L(n$T`{&!;EkJI_Q#O
zK%5c#V(-$r#lI8B*rTz~JWiyc{K16c?7MZO(E4zH)MkW6D810^L<AsW-*z-oL7Iz?
zFU!y+wcrT*?dQ#J6Z_8d6dj0toxIc2f2Yt4!Lf}omY<{meDM35&;S1Z`sZ)|0Ko+R
z2^4ZtJQu*b>1Tc=;ek9%Kk~Z~Xb|-T!b>uWeNlkX+@TQGhHuji4<r;JHr;-NRua~+
znHyq%(84%{G!M`;6&WJO+M37v(Kc48>JU-rDn@t<TrzEVW%@!DUc;w-FI}1HpVQ6j
zbn~^^(-Z&Wr?JwknUe=)--gztrHLJb(;BEGJ$EpS<7#84X6eY<8pS366y96wr#e*C
zw6rwZmL_IZIFzO;*U)}I`$5rUkDT;`_7f?8!Nw0B3E8B{tn>G8KQq~PCVFn3iOntJ
zEjWQrw?;_;5P8LFgCSjuA-Q|$o#zbQ??~6V-bAW7-j$I{%x&k_z?4P!IyE9jRc~iq
zziu@TgQT_EV4ks9Y0LtK=d-G2IS^Wl2Dx`JODP*3N<fSORhH5r{Td4gTLYuafzfn-
z;a0aHxlb=^0rOCLYC<x^uv%oow#bPzN2N{j<xqJH(%F0~kk0rdNdGnUy>i^WodfSN
z*UY`Ovx4kgVEUbscNZ5xm8<bWsU@<rO|!Mzlpn2n-%W+4_gm6uEbSc79y8}$?!lM!
z;+wQ^R!v%IF6T_vVlVYj1|29#m`XW+J5cjjY~A~8NwuNiW|u%*a^a#0spt)3p;i&P
zMoEd%3G@q1;wZkEQ<$*~Qhb?<+fsPxf?nv;TcA{WMHy{76Q8O$D^SSB^7>MsI_d<}
zmn`B5cQ=Zw8NKB2p&;cc*UPX5QUoPWfwJ<cU_2BAa^Gr!0`3f!s!Oo5f#L#x2(`Kt
zsE&F7ila>NvlF0MC$wGR78d97gFxdzuO*E#To<$im~KJKla+GaQIb=gu$$I-6oNbH
zh45trd%ZQS3vL7NXQo}rJF8}DSX8!U@nz<i)|)f2T({`fX`(zQWuObENl%@o$-dd|
zFq;z3g3BL}cm^3(aql!+f>`%|GVWV2)6}?sb3Egsf~zOI6sV4R1*q*Y&5<CfmK;A4
zG(@4S8)dj2Xt5&flh9f5Y%{2a;i-GWXM@^NCs2c``&rJm3=C~*e;izo!iH^$k@H7_
zOCUVqD8uHmud~3mEF+~ba-sEmWEtQ_YZfD&lTj|v&pBZoEl`EL{_humR7ydN%H25t
z7c5@h^?Rq}p;_|QeRqCJ{#8lW?v9#7r_1hlyYPy9>Taexz3Wa^K)|b)^dolNQHRaB
zapby<;A>GQGeUv?L$t`%d?=vkX(7W!OQmhQ?WBaIg=f6O{7@QW!Ms7YZF6Rmej2$%
z)~o8x>-*KI5*9*cKK40(xpiJA-Ko8ju#JgS%g_?mqfFVg7NOUhpl5hI4{whzH@_`}
zC&QRv+_wuq#4TjY&5or~)9=4duQy*uuI;I3L{J*Tmg^J<wsbM(81QvlhJ_}KCzwfC
z6}I`xpXeuD%P-H}jujz{9EUBgIXAL)R*${6ozk$7`vg;+uWTrPDWr^koaF3UqAz>7
zW{I#`2C@rFm=qD>_FbtYp4RgIk|Q!FqZS+=nFDi^4}+6tB&=nsYXZ?RJcUEo`2rYj
zjfAM=+47>~lHTJe@A^g@hEsZFhR&zUspF<MCLU#9(e^#ZQ^xOmzrp&KOO6~@{6@JE
zr2xH@`zi;^L9bMQ(+AZaDroB=P6{4{!T}j+<4m0Q5|gT{zZ}VGHR2M}QlBocQO7JY
z@_3hofTPa5N)50SqXbtZoU0CO>&7yIpEqqn8^CDXx}|X8!Tq5Wi<TEW0D}VePjxpP
z{Z3OE8-`LPr}^SF=Vm&zC_EP)L#E?Jq=yrkZhUUYd}zIYK=kkCycf*C6xs&C!gBFY
zFwFCCpujBBqTI^w@$%7AzN$2oY`Qf;+Aj>`^8A28(sTXmHmLKK>`maf+rnt`CP`6c
zWfivW9~6Qz(a2FhzPZvMWqEyP7qVVnX;8S4!gFKbXylQIdFr4JS$}EflWDzE>TW*5
z%m;1hXcO9hgErkA>A9xRF3wo)d(B6t;l9`L`!KkEpm#jPH9Rx&Wz6*hWtx`|buCuP
zc3H=L=vOgyT~(v>>f!@29F<{<cD^5Ut<@MPDJ+ZwopIsmq-%H<gqM5c2%UM?U7(I;
zjzGV97Q6kTRFiS8Ukrrm6uf^!F#yiQ&7^A<16|aAP6=oGKElD|uZ;*+=>`6v<PXa8
zVo85c^5?86oX5AGU#ANE{8fTi!r(%RW0^iLQU{9bW%?A<@5*8#L|whVAm=kQ&Tmlc
zO?fm|7Uvq{4K;!BLh6NH`Y~N(G4iqsiRYQL_6#>5b{x&&YoH)oF>7~c^OdKxG@NC9
zLu~+m;Du$!ZI|5#T&lULx~vPf#wml7X9+Rf2y2hZm5^R_(){W5QzBt|u90x194>%&
zB!A=N{^{sI7%p_Wp6Y)m1`1&e7X!hkqmh?LsDNv4-!K=m4cmm*RAKgRaaXWbNo^I@
zU3gPnMKt2)-Zh<q1YzepDHo(u2V9H52V~P3Gj^^I;%dVGDd|PNFrN-Nj&tb$0pWv2
zYn5?37MFsdDHE5>2L>pW!aJLP_{YD#{Bk2R>r!|VK8CmP+n1M{|J>A#C=}ZGgHa@u
zqQl?6--I6sv@q5mv=-LG;m1D@-){ai&RW(Q-lNzYu-K$AT6bboMriF0#-KoJi`a-a
zr+Kr)qM#GM6u)?*n%^~VT8GapO0DylV}8!hB#WDD+PolMrvCZydKjL6{Q7m;w|Bxh
z965o{;nj(<P$*V8dNoo6^^soSF@KLrJ7pCQ`_wLd*7h-2LfkHs%Tr{avP!HXs}Dvi
zLEx&;o-W@svDM-@(ZZgkV3UwShAEUzlXg#d(oZKZ7j>+SK=k7(yK?f$4$Ai=jE<`{
zt5R&Nv-R|s<bFKesz3>U1er?k{APJ}^L&&zyG1@;n!Ri`gjrk~rO>iQLV3JgnmmI8
zHO{o*l7L>If{`M)T--2oOwUueV2`;sGJ4um{>5d{x$ukXcAEKLig>G8AqP|KwIGF6
zSx=dVtQ5_$A2F;(6h7fYQe^zi)+Zqd`uc5Ip(gwW@=@7`h0x-EVrVRDof%NIi97#v
zGV0)Y-^x&D9u5Ve>h^g1@4HN*mr1-Z%Jw9GVH>{NT8QH%8wmDPubG77?;?p78q~Ou
zmOBfv6*<#W{cn3A_7Ul3>%gWM*iwaOpSw^DsD#x6cmK^K%Vvn^1nHR>vg5KTn1y^h
z<)P#X+2o4*)F<bEnNT%2pJtA(mq*7@afVGJBg>lxdx`GJvLuwbQ1|3OSP7Tto?NH{
zXjS>kIw222nGK!j&awNTHPc5QfOF6Zj%lX+ryms*U7M`qQ`QfJaFJFh-#Tr_GE}eU
zkeoYayW|n9HyU}#rOKLY6uT`Ej8sr;gtN<5ur->Ku&Tj-HtdQ8t1;J-E`29Ag}7?#
zlQrHW#XYyF!!^NhC*YG4;B`$vZF&iFt19n3&RVUjPA|vE4L^;17P|avOcplVlk#Ci
z*_&%=bg>3o=dU}id{Q2httPEiXDR%ox^>GM&`=7jy>xc00WG&wTvd%!$3`0~>p3n1
zxSATZv1rGCDb<G-D|&^RtOEEMOeMgoM{7uluwQM`K&;P7N>8%ubYhDh!@9>E=`VUb
z8o@1Farj;_t&`j8(q&T2uObS&D_n3!kay#vrm@3fyTx~&1MAzK&&dt#FlXEV)UIwd
z)(}?}bK=wFQ1n!!vdx4YcNLm*4$>i-_(!DLnpL5Hy2lhVi?|1`jyA9){#M)Z3F$84
z_7Q&7&OLWJ!OnYC__SQy{kDq{Am_@N0hNL?aZWpVz{3=y+3%c0$8|Dl!yX^n8>v#c
zzxd0qU)K$-ST{IfWM^r~cXEC-RZg?EAcHd$dVE`REy^bHvy@R~C$KqdKoj?Ms$WV^
z6`lcqjFnUD+5=b-MHfgu_o8a6{s?M3H3gGx^aV{YU9nQ{IZ4`KQ%@&T6vn2iW0|m8
zay9C@5FO&an%~`s4jIFx>AQnNTmp!uc38lM-X`v(tz(~X<?|vmK-vncYDf7NfGJ@E
z!$e!nY;N=iil2us5wSXrn$mP9yo>j5*!4Ysj<dcuI%2#4jwiz^9|?~rwodK{h853%
zn+Um3m3TegG)(kCfYY$bha*m-c=jPDQY0MFJ3vmuOlLq&1I2F+Iid4xjuR@q*WpV&
z_I%Pj>?Wu@s#Xdsv+T#`y58F{f&kk{X}g`@zTVxVzWs4`uiWk5{;`eWaP0{r{U5h~
z5gby&aP!`g{Md$XG%J!Ft|R%j%@{kNvjfRKipBAEB-#P7IDHeF@VL%4kbx1JaihVd
zwxyUvEFP1BZDT;DGrnvH@Gx)9VgW7P9xSqEep}E-;;%p-;|xn}76xPqpsTYlgGC@1
z!F;BL47=*)oaY?(DZ>nBLL|X9!0O+B6e0<}qdBdtK_rIW0+ATK1d+tKHX#yg$9BBZ
zbP!3JzlBH=pKB0F&(Ap!3FpI^a4s+mbleUYn9XnoTH>wnoMT4Cd=Of~MZM?jKCG6M
zVMeqhDWg-59{cA`5g?acHyjMPzK3kp)gzzO_Ui8joG4%(Xo;0;fD_IU#D&X$FzdO{
ziN)BzjntQ8Cn%NRR%@K;cMhNsu=F^0jXJpiK|y|KVK&3>3qnCq!jruMLQx)t0`smA
z%IBtWP^l@;iM^Vy>iZ1cif*c(w?RD}6J@`Fp9Gf#@3buU0WMLj4WM_&OO!z&rz53a
zzHb$7L`L9k(1!|_qR5PlZQZSZQCy(#?gonY!9J>)#me&XCNVS`0r}qF&NXc}2MQiG
z6WSGRH3r-Pahy?q$r8ANSF*x7JLA<(Nis$9^Z4Z2mR0u=d~%*c_PEiknn;z(m-G1a
z#F2QWp$tx3#b>~D--}HOzQm=$Fk8S;u0j};;(Bqq5`_AZ=wT{iQs{VpgxVr-q@|Ri
zDSYzlX|Vj7lPA?w-WyYj16kawr45o$){NYSTY@pzsf5IY0wduvHB6qS+A#2+1I0QV
zuZ2}Jsf2kc%B9D9N<$iUXS395Vtg(V=nBDO{*}5`tH`Qpdh?S~m6}pC_08KTV_!AX
zwb^S4p`h=lRD9*xpt%o!^10tD$K0ig99V9q=~LWVqgU^$lvX|6lIZTFgq}ccrn^el
z6fPA;c&-6_^|F%&p*?h5-~Li?7(5u^ExI1e)Rx`PFbN6B0O8Kn6=gtJJ*=_|_$b4p
zIj(e6+lY^{!12(}7J|EbPI43Y<<@R-f#e*mh=~)ex@&b@N665B)Fz*}H7i4arR27T
z1SWxCa|k!T)+AixmbRfgZd4RYg=<0fS9cb47aBL$Q&dH|CEbT>VTcLPCY@yjJ2%O^
za1#1e_Q5_C8aUlk?Vx?i3k+d=SP0G&n<U{qULv-Z;!LqgjD=QXTDxLvS2<s7E!CM~
z8y4CDNm<AmFa`mC(t3dF=A*deG=6VXxPZ~M*xQPeoS)nQEEg@xpQX3}sAEs@x*6*b
zkn2FjBwT7>_QUTx^5hVt`$f*sn~wjS&{6H^jT>_hZnh9?#yCVTtfel-VHbMph&IOA
zl_5=xTOQKIxRoJojMLj<B5on9>BIXUwhdLY%Q+a+@=Ec4vbnh9Di^U|e8+qo`;^mt
zs{GeBzw9cWW>rK1eo-%hKr2^2L)78fJBGJc57!`^)H7)%oGdS2FUiWLD1)WaX1d`$
zFKx_FUd1BmIL#D`8NFO$870V0sNVbB`wk}%hJ$lE`%YaGe9uYSlk7S>*l_&=da7+_
z2R0<+xIQj_v>l{_XDL03Ylyy8sKk3g*S5N~{?=-(b}u!18bL!R5TCY(3;9WVd_}n)
zKc?sEVY=2&8XhXX!QZffL-S0gN2(`E)q}<5iJ9u=JYQ=(wl$hcGxlkVeM|0)TwBp+
z9-pSeo7xTa+?cYb@m6}j%h;Rv8nl{U%kfV+2+Wv&)=bB!;{Sw=jw7e8BDvrCo#{%$
zop3pdRP9k@8xY3Btf@0xIr@q*uEO382wqO>7iXNf3^TS+xz2Z_+GDaTy%w6D$%!MU
znqXz@q+lPWohFO#d(w&P?VfY;ljb16v^$<3z~z5;Y4*2#OfWle!RFw@4xu{DR)Kvt
z)4Od(y6@({Ahwewu%)R^k~PMy9rRhDLvrth|AD6m-U<U(oE?OtYU_W(?aOmB+?6*l
zky`*xY5l(5c%m?%{9j&f{tq#)SK^g%I~AAl&L<QBHkZ+lG$)r|3_F~EK0W;N<(C_s
zkFHg8>1+CszrMWO{P(63f~2gAC#O-7cBg-Rzex`aSmk^?d9S>Vryu`5{eJUzxoX)<
zd`_?hVA!;C)^@O=Q`Ux)bEL@o0-IstEE8)O3iHet^2LN&{<KWAP2U+xYx9_=eBbA>
zpUSg`gsAcJkZIXGOg~P4_|$_KCfnPSv#1P&dJsROs0<IQ|NZ!O5_f+6+N{KqKyo4Y
zoUV=~b%J$@D9Ra042<2NHUFNN|C~id;AO-7-TI0C2{xXtpSNi?pp47=2Bd`F>&fQR
z>tOoDw#e3srAsTFmOwi%+TGzrKR!pjDK1Vc8%b<(h*X?eUw+SjaCC>-lTSpgzzT@8
z{W5DZueXaS@kw>H1NQM^IqDaBZ$OLe^}AdXQ8py2{muatqOu07Vm@6AXur_40PT0d
zYM{;n30zd5QLZAPDaKO=Wtp-v>H=t(9RX0TG%{Kg6(PtO$tX?2kSy-wVwl;Wd~kZQ
zMW6DMki{ybC)f#pK|cSMZ`t*~*=<=*%DL)D_6_5_Gue3GoVMOs9LSdUsPmGmiFM9d
z4r-z;^L<4HI;&W@Bo$fM_h?6!7Vn&#m8Iq5GlwOSQAB6DS>7t(TkWjdA9r~QZcYir
zR<Pgb_FI|4s->c*(5rh03fZYyj|%)OC%6B}Q`q5e?GizMSrP~PhqiirwrE+Xd+2We
z7;Z!wSj%X&5!{W?CT*BZzk&7#a+4gGEc1(;^kx#Lm}DS*78Xh3f7*U7`ZCY8m`0|^
za3$Mml&casN!?K53_G?6f|xc2*n`1XP$a~shrv~X!QZ!pXFzC)O<>WoVl5VIW@-ge
zy&cTD??8EfsMfO8LEAd3j3gE=tdF;hnx8>-cfO3CoMW-GXhXncQ~3&HX|}biyT5+j
z*L&bF!+@mVvXFN>-ZePGJp`pfdt`2jeJNs)2ZkbS?3G4F+871BZ%k57IoJ>L@9y-)
zW<M1~m=h+ny5g(!kHoefA^A?M!8WyCiPy&9ivApbove{r4E|pDK3DJRMYht@$#?i!
zVK7n}QD_Tg|7q`{F21f*DK2*=bX>mb8S5(LJm%2<5VO-766l;`4v5VFnjmq_7!3Ce
zEJ7Yu9N7#p=O`iN!IQ<qMKXZz!4f^}n=Rx#c6H8~3s^Keba;1ssk}6k1sHQ+(@KHi
zfVM_|OoBVbzC4v*;l5dw@>Q_#hv-6Z%EYkR)iu)i)L7tj?=c;I9B~kGx-J)O9N#d=
zF;TgV+0Tn&O~z~jW_%34A}p0r3?_BL_&}y6zZKY!O?rV`sQ2W*wph@xe7<9BaUWsr
zns$Xfa*od#gk(o-4EBNF1yTXmVpKxW`Rt~Do@+Z(i?CR0&cuSIz;24H3vNnwIQN*G
z*yfXgBP&8~BFBQE^TgOV8MksYvW{lF=R9h$2D<S93^QTY1mEUXO;!<?gf-m*klu9m
zhHrAH(wok;rfrA1Yd4W^a@g9_`{~&gR<an|$vYI)aMBYtjC^H{k4a-Lt!jLT91@y;
zcdzl86f?Lh1UED8cusJW>V6kh^*EYS##l+jAVWsSD!^f%nDIVky{@L0GwTaVA{@Z8
z^qoW1TWTUHHXs%#iNxz_BE+6dbKWIZxuDv?p#|$2UYqI|R9F%HhRT^i!x`5K0Q)5a
zaQRUU2IUa;8C1fn^UlfsJ;asZzp1Q$V}Jt=Q?pF655{aHYc(O5QsC+d?Z6Y}aZ4#D
z+>4OtX>lV-Y3B?#tV&sJTJQ2VxNqFhV?9&YODBf=n=Z*IRCUFJ=}rj{U5Q~t+O$RQ
zg3^%*EpKF=b1Y<oC|I<`3arJ71G={YW!_0qgf5kUny$BW&+*t%$!?a88YhK+*Ou&d
z%4p;ide@!ysElS(=++Z>R7PW+BARVV^k-GkQli5e(`|BGMYAa_!hY7cipDCg5Zr2`
zaTP6jN4?^44rHVxP(bBltlHo3p7Xh(zSa?@*U|ZIn@Z<9Z5ha99e8RBMspd+SZK;q
zW(v_tMvzx>eWJF^L{TKncZ!aGLtU?xPTxvpo#U6paMG+zTQS=#vy?RI{A$6T78@l_
z;|fhAw63RevInthe)F^xJ8aSZZEa_!vb!{^=ul@|3x~tN0yGszci;1Dt~d_LjWD)Q
z+2+>FugL?8z1duGIb~Y0%@%ErYulnwMz77zRybK=Z;o5Kkax_K&Jy2$`+?=uN5xU-
zile2JwNSh+c9imHXv?-6newXKKuzB7k8QSHQ=Iv_w%D6CE)|?oX|_eO+PIx8EqJ!r
zPJ(~M1e%C*G53Z9nt%-~_;*d9yL6;xU9U>yJ~!D8E~=dk-_d&)@J0~!CuPcLobn;c
zOxarYOYrWQG6t(+^RD}Ut=%vc^!mo7$NloGnEdosPaJrZ6&pq_fShrKtaz)50KHLh
zd0UO2FKr?a3tg#+KpywnT^VA(GrBU^v?@N=+uK~kLe~Pc-{ty%3U~&97S(4DquG4F
zy_Z9*bVe_SnDNtlIUq)5nqJCbFGmSbwnM|spr}5@u~(9>0E(o4JXL|;O{Y@VgQbp=
z+h4|+p7+xWDu`Z1a(!~RlYw!1L?CuKPW6IQ5PU!DkHgN4!169Pg1ZlV?AyPU{la#)
zMVxwD5Q8$9!!EpQm7op*7U3`#!*AX8kko|%qY=QlawHlLK~}n)PMf<?nopRr+_@pp
zn{%>!N-b_+x*3&!k70}TDdt1o9w93Blk*f9x3IA4da~Unhny!i-cHAE(m|(8DnDqF
z+rW8{>A<m*eHCWylrur&xEEt9Rfm|ho170?JL^o;#6}1GMdXw*ZjxVIN}eJeTt=C$
zhP^jRPem%(rm@F3_N=@n_%Qi}a~Q`;9r;!_Wn9uCBISX9&*q&9n3eGR0id8Th+WR$
zoMj^%`dJ5@vuuDYt~P?Pg|jSJI`kHfddpLWv=+{94{0pi_K42H?G0%xoE^)Z;~2?G
zr@K>y0LPAv+@J~}yS;K{MYK4$RXqaxxuzcv+f{%;eSg@AzyVK0NleZP_A8g5IQ*5f
zj>^;4Ee!{M&E2M4Rf``>vw|5%HKVd;UjJuywHac&w-%Oh4dC22xo5ShSr*s8;xA5c
zW}6zMI{i|T{3(5EDKCgkxli<|5&8@I)aF^TPmOK(L7y676#Y=koVHIbuiq5c^Er3=
z)bc8?=~FX`f|KYRXfAnBj&n^h>#QC%j!QVjl-#&~R*xF62}Zfw!hyC&0HHu$zb#|e
z@kI8h4bk5;@4&t%x<9!`P44<GJ!)dnSN5pM*_ZXGiB<k0J!+C2ukTTlEBt@;sBwaA
zw0T2`J^d|Y)7kNtsIb>oDaz`J%Ui@5|FH>8e=fjLX-E2nHF!}M|MGJ4KW$<ey9#A)
zWOH<KWnpa!myHSz2LmxNGnKI=6_*7UJHmh9eQ`Ye?e*t7o<--(oA77&qQ72V@BVpL
z6#>b#@jJtTNp5z(eY*=cl+esrzq6KE?swn+zWZ|b7s|ElMZTlioIta2W3=wf28L<v
zcE(61tu1C#sGLS+6`Dew@dNllMKykFR9c6>(Ue-}HK%;t=C#lHx;X@k1~R47vTzYO
zEDg)FOf2I4z|Q4u@Nb^%w^tZDLkR<zYpItZM=BGSupv7Hmmz`(6PFEtAt;ycAv>0T
zjvpO9AAWoO@lIyZtDJyB8$TIKf~6^P2Dymkq`*m8UeU}~vlPVAXq|p1=B;_Bu9C%F
zHt%4GsQm82+5-fAW}cDW6xzty<R|QxVJ|<nau1xK?$?EvE4)<Lov5O$Zdg-+N96{*
z>RaA}#+di)DrK<X4|@ZX7v6xO*5l}ZlyD9@OTPQ}eIi*75_DE*YewU<>!Yg#TLIl9
z1#06Ae;O}g`pOPme)IEZKFO)Ya4c@tMkuMvNzv3$+>_kJ23QP~9MA@}u&xF)%Ivso
zK7?$z5ZKjl+5F)IpZ2Gdg098v#rY2^do!{PuOa&(kblfEk>e`*#^*Iyi@(T!vg+s_
zHhrQTTXod)ac@dkfPc_f_^Ck@NNd%D-UKs-kI%zzpPujjIk<%!48lm9e*gO}ydb58
zv3?>F4%zd6OkeJP8o6dajh`_$#bA)eus&Td7@@VBj3i-w4ZGeAm-ryrdo>NDt@0__
zs0u}F0;?zOk(G5-D4Q)wysiF!6YoH`@^cfZ#BRlrk@^^(<<RC!f#^qykatP<F;zl0
z{2sIIKIdkaZ4ln?v(hxj9wc(P?{&_O&<=!SKj(wx?&Uh86i&g&X+i0bE0|mc9-tG~
zZY5?BFBby`y+BaoDd_uvNZNL%u7;xF&?Y-_)St2iE|ArD&U6PB!4idkm>e}^yHz<o
z=UPj*kkgYea1aA+?&QOzvM0$2oH6`CJnb)*5<m*YOIMPTlpr|pI)@KGOPRCsTWPjR
z`XnQ)s@08~Dx&+G)aE86m3x(156Z1a_{rZliL8(1@;0{&G`AyNETv?T?Y&lu0Z9N?
z{Xts!l=Edc@iB7wU}P+RhPg_Y#1axokU!^%HLHN<3I+#O34AHqC7e=|e$$*$YR+6!
zsh^4kx;AT<-%?ZKO|@0iQ00}}P_?b&Ts04E?5k&vx0ZA2jaJ%KZfdT|`eUyvw~~D}
z&QZkvVfT~*<QnJ7r9Lv5q;`#69~s`mYF(|pXVXMQRRYJSn!34v<J3zXCy?W7WLkmW
zea#K0<?=4PMb}?Eu5>me8{{U(rjU3C@3-1oFiI$H652z)ltb(?g%yf=ieFjB?PIJG
z5}+gccl#I{!rF|-sLhU>>$K3Li*vVj5C_T#2a6*SX;`}p<qo7BG~BGXT4U1ImAZ$2
zrX1h9v<~<47(`lsHDeIPKD3NMnUk5jjdCZ+cJioh#mvE&`I+X%*nF>cWk;kL^WI_}
z$eG5O`;orpon4b%dTC723CB#*X&LXjueZ8+8}^sUCdz1+eatPe3n7v(d(mBHyA}_-
z_3K?}xU!i0-e+ZKF7_;u5`TWIWAJK<>BeK-4%*Qxc8%kIEM?%>kI!M>Toar8@74NU
zU^fbnzHN=8gh3vVC56Sb6;@O`UZ=?3&y}rsE5B)CGE?T*sw2F9Fd+^?$!k+&rF?As
zOog{Razpd!1%gOgG%Y-_ZYoE60AmP)S7F;%TXi2Tr)xa0rOY=$?$c;>q^c7~toAf=
zX{@*<13>_PDfFTYAF~X*<1&yI8V;E3Tcb9vFHMb2S3Nr!hqPyUV<v9$W!IT33Gb*H
zMUPjvJ+Vz!31hMqg{s%=#8kI-Z8UekTD_ra7o%#eB~L`v2IJolRpVPNlyVoKYTV)q
z>-8(6YF()xQI7VQMPzM8c}_&u*jh5UveWgIkTv9gUfF0jB5RsQB(JWHtc4JCN}Y<V
zo$iNZb8|MjhTJqe4VR;91EUkrwSnRl(6yqo(ZYGGI`7Wbxo323aIM&>Kjln_MOiBh
zN8rG11%O1&brb0d0OY{wMgZh+$`t^}f#Vec$WhkI0LUB2nE}Xw^F;vU7*E#&Ag?54
z1|SE2ZWjZP;g3hwekb^2FYL|W$KjYa^l=+~M9F1xMr*$V__3AvMEsG9JCDLIfgig=
z{wyT8cnFs~Pr;I!i$$XN>j+}$1(-pT6Cg_zid0W^46}6HNYKGS$3ROERvv`L%FGGQ
z#fj++w*3FXmI3?rGc^4MTdv&&?S%LA23uZ#0k#}$c_wT*aCsqYxg^7jVat)pw}&l(
z=S$gOx!)SL1PG&xd4nx4hAmr5o(NkG#=jwK35t6wix<F_VJdaxuL@gsSWkj2;Tm5~
zJ2I@0u*we#y%IL5q!(7xh)o*qmG#x3Nu@ozuKRjJS3b_Rbw!v<XR`u9DjplWw3kDF
zkOQL=AjpB@^&m)vmPeBLn}ttj>)bQC5d=Akd(2Mf!*2;L8qU+7Qq2jy_?kgEn!TI9
zhj+d~dq%f1Xn)G>4LX=-!=NQCD8ntt6#(F#(blERSZ+^%s_nY^AtAw0J#Lh0w2{!M
zc>#oHyYz^y7_k_{8Od)Ig0KsRsVr81LDvISiC5oh;97I#%KmLQ2hWFdkRyTERE+_S
zpg9=6MdjV(P*DD+{1G&V?}7m1x-+)FOorUEgR&jPQDg?m3+bUF+6HC<5Nw?pkJ-j$
zkOv84hKPv+!0bmm*M@%H#RK0TASz}a98#kS9o)CVVLf-OPFGF~i#RBfi`43AFy%pb
zr3a!}y7kzOu_<dycKAP$m~Z{lX@My0ebc$?<*8=@<k%Xd6HROI1&n@}vJ#k)vJ20V
z+~IQ`<j0}LlphaPHyUx!Ab^Wdukx$@Ka8QMtOPncIrzGlv$VTI56Ug>e=vqN-<M%}
zI~A7^FeMb1hblY7f0^xXZ~uDGMfDE{8z%KeqpXV?=Si`4z23})g=ouq(1Aap`#XQL
zRLlDyD&J`>rsu2$C4KV>EN>Hq%etJzuIgZ&g5atSX2;a~DwBDme<{+IG<)rSr(LDW
zPxs}GOHZ?&m9D%^Iurk{4{1eAE9&0Dn8xNkCAyw}zvEs{e=2KSS)XO~T&g}9!-7tL
zDRhh@Wf&EK4(?|$HJ<G-7)e$3KPxY2Wzim;M(nU6*J+=3%bT)4cXpZ1BI8%A0{9v5
zx6`wFk_ANurL^G$$QJy3k6m*nzMf-}`?`RB(&WhMh!hStR-6pqz~1V7y^lRtQ$7_7
zzlbjHQBe#>e^TAZ607yr+p|x~&!UJ16<=P#G;bj@npFuYvdcQAJ`9(Nb`_X;HT==W
zl3OD*$zbEw&`SlY2R3Z;x2=$itDOyD>IlY1o}VN`$rlWo3l`>$J))vy?BY0LCok`R
zH*dgTRKR01N>?HFx>K*$<Tctx=&;HzTd{4Ku+hp`f5%rsD;G6xzOo~qe)H+$X>+Sa
zj*gwk*xI@q)Y&b$9;g~GWL$IFDTLU213mKp470&<6MTssF!KIMLhRO4v2%d(u0B6(
z69R-ImtU@LK8I?_SdhaaqM~REn_l$1oyMq$yx7$4;Ocp?t=eR>w%Wgwp>A|q(Iw2%
z*GKW!f6s3Z|2^_I7Lq^5&F^0x(hCBsoR1stmDh3e^^eWxhyO^f<xu2z&PVdWYv-(O
z58f%uA>Vqd$eR9AH<;f7cz0e6=4=!D*2NNU+1rif5Jr)DC451Scv=4=+jsNi)5r{o
zkI_g*@Xz#_wp)%T#zaYxa;qFsq~Vmmb4GP2fBBcdDj7qG3h**GefCh3d$#;-tO(cF
z?73w$vFGFlm|7<m?8`FK=1F!CQBT@Q#Yw2UgwwPeYctMi%S1N#BY7!{FgaT|4JaXh
z_Zfn{9?+K2T7b4v)(2F;a|WP>Q$+AXYIrrEEu*ynZKaH(OqXk(1s1TMbpVo^gs6H?
ze`GyvA2;<C5E-okpyV{?xlB|BK~k7;O|@@p9H+BG8rMrD8FCsMW0(qAJ!13K83j^4
zVy};=s4%*hF3z70n*FL8QA)CBLsKi3RihNHK5->Eq>^vD@mo9@$9nov){+r=?Gn#&
z5+4qn=tM>Z9W_ZJa)`FjAn^G0sUt}#f3J$(gvQK_F)DJ{r4ZO4x<h=#LgH{zis7Yb
zS}A{45;G}<Z!+*rreGAtBIU2Dn4gqFss0dq`IV8G*UfUu>0C?ozj;A?ZM8d+15dM6
zt7T67O4wR10<?Y^xlbIOV7Ub+t;TK<UPiVNN<8;zg{`Fp%y2mEoKGd*#C0Bsf74v9
z+9}opZKxLYb3^fTqRJ(Y?5h2^JV`lRE>8M9eJEE-mqeCQwbU9{sINm-KcX39?rWv7
zh1wm;Xb#o2&4pa&o6rW%j@gg$Cm6m%*}on=ohf?IqC$>KV$bhWSfsUG{x@1;$Hp<Z
z-!>qfFnK;|M<|so)x}cVjx4D-e`*SSi?$sZuV6^!{%+efZ99R^nQc2kbwS%sp2oUu
zJ0b0kZ9C$i4B^gQJ0bL4nszPE3ww5g__~%|%j5^?*omXKykRF8d|&-K(w>WeTN-vI
z@Xd7LduZ4l3dwGWr`1A7Ir=#byLMA|YS<x1X!gspVaJ)hbN<~Mb}4Tre+@S^>?Q&4
z&TH2$Khdx=WEpR~y<sO9ozbup6!&Y`8AFN-soNWNg3(It5UK9k{UssqjQ)~I_Xq9m
z4LiYTEkIi->jNs_IRj8Yc1ShZH@!c*9MFc*S+}yKF#fcXnlo5k>lmYVS1Z{tTJ6+Z
zDeIqlD{t=9OF-Z(oeRYGf9OF9PG|HW1;sb-K_ZvCJ`BO3M6;Tt<#IFBT%X)1Y*I;K
zySKY?imxRi?ib5B)K+n55*ip-$XhtqOu)Blk8=E~(6m|Aw>3A-f1Nr1oZaejHOW!f
z1IX>_`P4eK35oMvvENh?khW9xe7B>8rm+d}jHYqf@3>jjo5EDnf4F;#Gr1J#?K8yT
z0BkCTKu?yVNXS)%KdMJl>a4pI<%}Ho*>tyPu}?kOrA6?HGv!8?$6q-$dYbEQBkzDK
z_=ae{xs7RdV%e-yFS(>@3(ajbsT~Tfl+Uvb<g`?Kkh@vvR3jU1B-c{ScdF5*f7v^N
z&BD_&gz)%}DlbeDe>-;d`HJg|eJ1|*Y}X#W=fKamB*(cfrapdo8e3Jrg#B<ZvavsX
zD1oa?E|D6NpI{y=uBB}>a_8vg^^bw662!nE)U?W-n4oafi(oD>GLAKhL*^w$W`L9n
z)Jj>Hpa2(^=EPo33^ovzx8_2-G#t$xufD)8P3jLJOxWeDe-BPKEjsCtx)BO3j8IG$
zs>eRGdonXZ=>{`(L~)YC6(bafDs@f7PL@U}E^_PvNj2yP)B3d|l=^gfOfne&Qj+D{
z7~i)i-gAWF?=nKczz>f$^)8*BZY^)Lp2V#a6bC9@lUCO=2Pi2+(+J;mfCAAfaPjW}
z%H#k=pmXK`e??GTFhG%~vF-pxNW0?z1=2qa0Nr_lBBj2|2t_cxaE2nOuN$HWE<eZ=
zMIOiHV-(5Z`<kPG<FehE8|Em$1qEvSVdf~q|0kGoIg4{2BJ8{aHC#k;BHHfzPNSCO
z@{#1E#W_y6-+j;mTrH4scN(-1&Yj4~K@4c4$oty|RWX9m87*Hyalb)KDoRH)?W@NI
zEuWS8$E}dn+sCcAGiC+>*x2LkGnkgoN?^7^RtHmHvka!^Ee)aaS}4a{RswP?Vs#iB
zRE|MB&tY>o_&>v_7cZA#dOH@Evh5ucm$@@LIDZyjfL4Gsk`~A*S?IQ#QxH2j_z)k0
z{QMMO%~v<OXEbmQ<I(i+QLHL{^;MB$e)@Ler(f@$-|xP>Njw>5l`wF6|9TSi1*d-n
zr-xddj1WvqaeDuL`u%}((DZty7jiBoy$t<r>0LeEOJS*n(CH-}?+uEF(iYCSja0E0
zxPRO`y^A!1rIyAT?#DZ)@$m6)?|;7INp#Mf_J8`f{?qIG-M{aag@9yAyOT!2gchg2
zf4}ozpioR3caoOYj;A00IDNhQYd>nOCM!ryTx+EoH6e{|)Jbc{r2i_YNz&qjunei$
z^3RWNr|!vbzZN->6fV1gxOnleX5~u<rGJ@}uJ7S3!rwxVK}cnpq?IKh`P0=xHUq5*
z$r-7fGwz1~6$fpe(4b%59ADLGURXPecOxW6QREDY7OCiO^v}{~NzR0}d{*GwG*WF@
z5y%;a`P6GvCJi?z6f{%Hl1~|i(&Qfee7Qm;XI4Nz(#-6V%nhRN;kQm$F-eTHL4T39
zjiP7x^mwni--hZbePfgEtp5g@l_vSuHo+!P<o)G=0^MS*rzAtNiXu3WU>>6Vyp4kG
z-r~FR@BubUX=7hZ5-StrO4b`B04|vW>?Fv0a7%8W)<BNf$sMHedNYf>>wcBxrBCSc
zr_cuC2!o#{izqW+tM55ouOuWx)PMbY({|okuSCs+>9yoN`<}qLmYp6df@@`yH`yRj
zmO1;E%}*fvluoui7cC0@crPk7DfG_G>tW(&NRHsukQEl)cNh@#wo%$m8XZ$PN6;ge
zZ7oAffGm}XON}EX6>2y+a(~d+1CxJLkND}IpJNweR=a3q#)!7dFR6>RtAAaLVqncl
zQv^0Tb@_vI^m!_qaUqa36a7JPH}_Jjy(|GA?|S{nD<Xhn&Pdy$vqJ`)ypMLB<k62l
z!CW?HDb}iXmakL20Y|1m^~+InB>=aq>+|3FGy)kdXqZNPz6XF$f0`N!2)hAQFqF_S
z0Dt;R?jqNeM=^|SO5jb)Gk+Xnm+zlzx6MtI)6^u5`md=;AjEgqcKJojRizZOqk=)t
z<=29GS?Xm;Z>|+?-NisFCQx1CVqhczoVmosz!>I)X=nL1UyB8%^~PE@+{JGfd$Ctq
z+M4<zw{(j5#hP1A{gGND8K(!1xg$tUL|Aoe`p@TSC=o<jQ1r+eLVt<KA-<q;OGaV_
z(1elrgfQ9DlqHndNDo4|45P+TCL?Uegso7D(%atlE5YY!C`AM!9IEe?2*(hAEbSRB
zYkpqjE(C)D<wGxf3cPXk&nXJJdd)t11-<Z0vq}&yL}byBF0qc**lQB^sg@-IOFrAJ
zDbP4uB6c2;Kd1L|CV!#eV?EErbJkBS{ot~GZXx2^_M>eKx?jaY7>U_Wh_0fLPH}Z1
zy5~Zf7P)F`n5@~_Dx`Z1y02-wt*GCeY%3yjM0*#Xs;T`#wDE{!_0EiRw@x4m?aTer
zx0yMGJrB;e$ZNB8AC-4=q0`HX7df$zN43Z}5YBz?lw=v{k$<+;&dzR@h6phwwSBMi
zJ`tzgpgazA)>y@%m}dEkM?f_wULok2DCyUx*D9j@68hPS1wwLcn*^h=iOF{K^PYFK
zzlvm%s2?sVXJ?rjrA?1GfpyDD7yD_Bu$rOA=5AJ9d1Ch7yn8$b;5_Y<z$k0(2CcHb
zDf}S;{z~svIDeXpVTOhIf*FGJkm4RCK2(`zRIGiB;UcTN+u9}kb|$Dz9)pHz!e$9b
zQ)hQT>_QFZa4A9a>z`+!C?t``(T#Ej9c9OzaPCBbYTAC5OMBu>Doq>T*O-V^hl<PU
zH^2S5?QYr01-YGQ+%0aQL}m1%si88HN~TCO9LRAq9DndC0Fa&~q^)35@DLUJ`x#uR
zYlZYU0cNn;OZ0<^Dbz(4gt8A(o;!6y%4fY+T7jS57S2JAsJug}?A*GKW0ff`tjD5q
zOp2i%3q-BNPCd4NpVenmr4A_zQPws-x9IbzG83UOmsk0m&t?ZHCRILP(-MR<U60)w
zsEDPMP=9W>X)USF7Kk4<dr4eZv;+?YM^^KZDln~paxFJ#V07<|BY}f;O9aP^Tk2s_
zJt97o5EJ+PqgcqA9+3)OO1ct@RXc$#s;`|GF4(u^9M)n#!n#v0Sw@{-$uZyBWQD>E
z*;Wom`k|3I)YSS?;KFxfue!b`n=GUW11s6x?thkD!kasUSbIDlg&<EV_HmEMA1VgJ
zJ5D-xGf5C2q}$E{uheHI151uq1yE5jI+X*=y1ujPhPLOetR+ejiH2P%8K+YI;#XL_
z@2ML1xyYKlL6vV8?@%>mS=nNFjEWZR>n(oXJX%rA8l;SXGXeR`^O>vl)|5q}HC^in
z(SKgbwi=S`P}`|GXowc6cI8-^ZlR~vb9^e~v#bE#RFu><vp|(46iJ#I!I(Mj97srp
zlt=fPSP7m!NbuL_L$M+tM0xGy?c5S^9OgMl-if?2CB$JVU#^wXQD}3+JX`psbm352
z{-`LgRfqktY2ElESbQki5@t(x$O2Y96o2ZF>vx!_$CB*mJYJvY$No9VNNp0@q)!$4
zrPi*Hl9;E?J&rpf1#VdDhAmY>KR-)l)0dPnzU8`-r|O=bds^--Eq8PPYx!rYo$&L%
zrx=RgTKB8)@_-wUY?u8$(~R=9SvQP{b4lnkJv3vl#?F1BxN+K({SYbBcHIHSMt_h*
zORV2DT^{OSVRjvI0kc>nj+7RNID)6I^s-XP$UujuldRBFEpXUN25)8S1l|qNxh`^;
zbII{iDP)hcN)ALO25I)-gK{8$mUa+6n${Ij=|YZ#wLm0grUkaz%ymf46!R^G_s}C2
z1(#3#QGls5s@F)#(le8zVVq1=0)GJ&ssBqA9OHU+sAB-@S6{q^<e#C$CM#<$aNDe}
z)}qe3!ipFzlp$(RHTq12z1%Jo4Fge&hX!TEt28N&(XCAI?ut;dzMxFCt?+e<2IDHe
zh_8=Q?W0mid6esh$N3EnVGR>jcb(EzoXwdWv344-n`sFQJBgKMLa57T1b>gXp8d2l
z<|6OBCI&=G$VwN^81g>R&zQ;h{0PAx7CR#^oRI*Uo~(EWE^xa!p$6u-xNt&k!Y!3<
zM!&?I8_e@Au)gqP##t$t=_<budqL!HOjkya&K?L3RK)46O_9@<I8e(g8>p_zrvADd
ztwcK}f)v}Eh>)pmo%GaHKYz&FACHQMel%32@E?(D9Stbn7_r2HBj{Ki0z;i{ruY$4
z-0J#GRo_16gw*zkc`iwEH9Hh(iq4n*)nVw?0YsY^9m4)&r2{en)6Vp?bggQC@W2nW
z2N`p-^1+nAS)VBBX3lb9VdGMeVBibNj!BcQwTSt}C`YSxvYjN~w|^1qH%WYbUFlvI
z8d9zH<#Bqa0yyKXA`e{MAoV+Xbb%^7IBRl`Mi(%@?jeyfMcdBuWv&Z<OUMjvL25AG
z$#AX;rI>SYsV<ZYtuN7quifg-FXUVolr?iXh$gDrFI#|5%-*P8CCf4PW)T#&il9va
zD-}tuI_mIejskUm)PEsJV^Z%NbsQ+mA2|L1X&C~PQvF845HBNlt@CXR`C4blmmh+p
zYu&vp8_5fqQF(l5TVI!vUw*<tU?E7Ay{x!?SOZA?hn4vX0P@S{mPC8auyX#oJ~xGG
zZAIUDfqQW?`+@buQ5>`%SjzzP|H*z54KqYwdRf;~U#WmdIDg#8e*PAU<)iPw`rJk`
z;;4P=`rz?K_Vec%n&UzHd62Rra$I9S`Fgpmh6w0j5TVyaH575ps6rlJye-CEcS;Dv
zgf%s(in;a8jFgtx>NpBxJ}=hm9#_Q?f4vEC2qPZjVK<KKpTveZvhkCg6h{OFSz+CQ
z!`QtaOm!n&v46*xk+fqV_d;9z;ZobSuh*oU^eQ0~!JUEnEPMm7g$EmpeUD}C5BzQ_
z%R+0FLVm<ZDfo6HU1KB<hG_bcwyZ?v5ghu~UHSg<+ZUYJCmxj{-PP!2*f=P-np*VI
zh$*$`TK_B>ltO+lD^5dx?==~JZ|LEYsHDPMcz8R~Rev6yMgrB;<hMLL?T5DYEs!Y7
zA;@E7``b*I@3ywbE$ozK;WAto|A3v0VOofOq-*RHE_<tfq%AvHO(}Wb_y%Tz@bf!;
zP>lG))UIvA?LTg;oNs*GSfKo`scv8^pxB6b)K<R0tlg-lk423I`;N(wv9FWixju0I
zK-3^SWq(Eu4!IFEq_AZ9L7JbHe9@j&+77aABwv1C!r*>|wciwx955y94FIVf=mU7v
z4m5|Ow(Skk=#UT*j5-<WJoR>1^kFk4DEhFKG7=qAPy(X1@r`ikm;?!iItgx#LY*vg
z5NaF$WDM#g)l&%Syk$h7o`+IDrCift{XdE@v=Ie{|0Uwc)%Mr-yZ-?eGB4AWaXS^4
zkq<8<1Tir%GMBhyJTZU8T3v5kw-J5!uOQHu^ufI3aAwFU0t5M|(*O!;#3|6$(1We0
ziBm^1<aOZw`<@|JcXy?gv(j$74wB4X^OD0E&fz)a&AAi~Ddb#|;J@Tn=0GgnlEBjc
zHfxYpx>ZBPzizc)G@e<=N%||~&0Fhc4!IaVauXd3^|$;D73+T{nl&A-;X$+hw4ev$
z?nyxp)+A_C4V0*NTL@;&q1K9r8Q%K)0#mAwB{>jT_e;=#Q#T2dlHp>(Q*C3c8mQ%A
zwm>GmZLv^NQ#UyU%%eBp97-9{fOmykpBa}>bi{)>l)7vnp|lY%*1#$Efw=~z5ixFo
zi;UL{K^{@S@LYc@dX$8>+DL~mOp!h{<fB=S0>)wEKA2jNoO;{JL1?^&9)-9E2f-7T
z@w5h6Mjs&*kE?fvr<QtWm0)u5(cp;K2!pnu%kZ}xlty|yGiCH!VRGuDAuB4Sx2-j3
z<=#U@4<@65g_>HQJ3>@xERn{cmNL@TVAkOo+VY5hOE7;myJ;}B_K+D~tF4=GwYD_+
zt)VT0ZWP)w=$c@TMS}Iw%9t@$Li05GQ<&1T!YBs>XLQy=tEu~7cxoftrol3eA!@MU
zIx-!MmAgs7BzYu#31aTrpEAs`_>f4`;^AYY%vvuUMx~jRZYDQdZ{X8P*~VkUHuXA!
z-7C-Q1aE)hZ<DG=lf5bs<Iy>sOqNC>v)kOqS904%8}A_%pALK7X6kL2c$o!nQ<wR7
z+w&{E&bfP1DLr~IU?hd_moRc|t)A&@)!AyZW!Dx$ac$+<qS-2Loj&^f^XZ5Gy*;14
zxVX4nOwZrHTKKP%*BAc{)0dZ5Kb~K`@P5+i^tXTM+v%$pdHneP{A}?8YbZJZY2xTB
zhb>1jtnJJZfj*tS2<ZO#^v&fDm($nbvs!^32sLF@s{+)he|~~Ln@(&W7kLSJIc_G3
z;+3TwtFHs0Xz)X7j%25gHl{IsG53D9czt;RfBy0PxBl;EKQ9(<-~DBpetNz5`TeVS
zcKLtn^vnBS-#-5F{I}`5my4hO{&w;0#p3+xKQGVD<1cTfS8py~O}}A`6!Y}%>TJ4r
z^A^wlbW+pf6Dyd@qnnetXGF1}B9<1wj3T2X01`2wz*#9Gx3?lnDN}AGCq}6h%WAYG
zqW4W}!r*GN6Ik#u0}bwSA_M)aBaBc|0D^z-9DC&z6%*O_nPpzNq-df?NG!?YladdT
zSi#h=`L&qExdAE#h=A=3E4Y1{ys}EHz}0IfB)Ly~rxO6bjKPr&-_%Kge|x2s4F69w
zr<KfNO63S?E&TX|50h4ozPn)IV-#|x6b+;RJSw&#pC+$h1F_Qr<4E8~xz?U%N+o}O
zg0$x9%V~wexp7p@M@TE%;}elpTEF(ql=^l#{qtY{4o-Nn7hb%7^X7DWi=1k7HZ7P-
z#@u#kis)DD$vZT~E+3t)?a-94m2%n1S15rfAgQ}Fl^mTlu9}AFtIIc+SI^%fwxLsk
zFg<&@K(sHy;0n`UKP<j^zIeGf4}*V3OiwQ_78vAd>HMfOl<BEc8saj3h;tgU@xuw1
zq)t$tJei(dU7kHZU%Z%}ef@O$;rzqmbZzB5d->Bj973`$7VkP~>ONq-y}NvWb$0%4
z$=dzdcjrI8e);9)hrU1+<gXf?qK7NI+Xcw*r7z}pOCi%;=oE`)VpV23Rc3!WRc1O>
zW;(?)o#L5J@l2=2%$Ex0+L|dF&tvdB2G3)#9fFEUUxiwIug2cpC|2y<+K_SHvS&jQ
z5jd}5-v)7I@O0ZBiJ&PXXc<Tgf!2bRf(F)6Onln~5^f0+M6!T2RBC((50=!BG<vXS
zA3&_K-KL}{9*4lnt%*?;rFwq=M0WTW$PJL)=n4ed4~UH4fGm<Nken?W_K*x+i5(FM
zIRuRZ4mTIGODCjBbPaTu6E%rpSICN@E|7Ry4z?novcV{2l?@jRx!!U?5C?6_jkTG2
zZt4MCAV4YhXvRJTH6Kb%(RD3S9-)x8tV?T<bjU$%uDKcd`~Y4!hXsGuIEnUCASXWv
zG`yl%9Jy{X1#GJoaZ>Ek3W;0!0M!BQh$UgO6VUCk`tLiOFcpwQl&Mwt3&ARS04Jc@
zTvjk>C}DkaA@Bt_!Qc^eJ0Duqa`VC@QFwuk>;POCJE36Dx^_Zba<n^rkm@ODz)6s&
zf`5qi;!K4sGOajMfjob(atzLt6SQkCau70%)#L}Ch-LCq9Tb>V@ciPsQMc(vU|ll7
zUjnN&fi><YFp{h1HGz@w+>RkIa%qTaFWE<XFzb|}`XS#nvj&(VsK=NOSIkH94Q?%Y
zjbZ{?RJRiBjb?+17Q#+YWS6Fd%|fm2(iE8Lr`E@!x6@NU4zz!9@HRcI$efMyxO)@d
znr;(cZ4V>9mDh-`&ZJalQmS<e0lmjDT&@6gT^++g=9nF2nF@qG7{q>MS8|TX`;5&X
zEudt_m`=@@PSgFSV+t->)<C4hl=v9aY0&4C$_)@v`#pia)?!xOU^0IAzeCdk{#&8c
zc4%tOkmV1@kGp>%!!^2%<N6K!xPdw9dtJJIr};^bF5l@t>Vq(+Z4uk;fnvMK8s^j)
zbLxyabr!;P7Q*!aLYNYGHTwxpLrVQ{69f;eUm(G>Zc(8M=l$Zf;!XTWD!kv}H5O5;
zt9lbCQb6!A4zC;F17c=RG%OrNR-_707?T`jJ_-^7QFea-s>*(7=Qn#|{iwp}fAfoL
zH8996=kXO%DMJ>UrD~0k<h>RNFqv$f*!gMAF?P(7vfpsD8<^a7BX;dJ1#_wIUjyvW
z6om)g&~|BxnsW5~w7|TM#hWO-4!<8k=|cgt<t@N$4+CafCxdN<<F+3K<XU6@(&YNa
zu9euzQLcY>$QLxH{XnnZyRH=|VHc6(6et(WN9jg^TKYZGwANy-&8q$4Z-=J+B5R|J
zKQ$HFsH=^N+bpsT#<VK3jq<y1#c|pqzwBZ1i_<z4YBSlfnH1Svg|=T6g1SI>tyZmS
zc7Q#`R_*5nq<fx!_~Bg2YpBAYBAJdcpQM-%l|6q^1xh5TuANUonrS`8e8}gs{e0@a
zgj>*m^Y7I#H$PH#-~c4~<-_e0<;;pdAl$XJB$k*~m(K%>Y$xVnx5b-EiJEAard%}Y
zmOI-mO)0r<$bSS7Zou7+JY)mm?t;jB!(A@7;Lg8hdl1~^wQE+tb{f#;7f;t#o2~21
zRsVnNZZF^?@WZ;&C|F>i_eY^0mL8vAvIi_Ih9)u4J!tBv#V>0NV!^iD7(wf5ZT?}2
zU+D_(DP;Yd7Ft8!43R>|Kwm5m1eFdU{D@H2{sESsKl^2%62{X&5QJ^pWnjZ9Wv#{r
zIHCsax8KJy%8_KX1>~Aym1ArHTtJlh=8=ECe?|LoV!7U8hF(AM72BmLCvcS1c4?ZC
z8{0UzI~eqCP{@Zs?~ZzK9}Y^lkK>pgb{xmFruumPY?#Mq!+RY3OI#CuLK)Kk7zgD=
z{X5Ne5Y$Zo1RKX({UerCz|rs%Agpw+Pn!F>rjMSa{2yaO1VajCZe(+Ga%Ev{3T2mV
zdOHaLGnY{b0uz_HsXHi_5KKD+e`L22zUx=?(LRXgQrrNc0jWJQkXwwoB{>Cg5FkFp
zK!E)G6v^Vwx3ttVhn8CJk+rcllArkHpBH}l<>~F~(@*aTE|xhhth{{v?IQ36$De}Z
zL!&QN2xgSHeEt3Mmlw{Z!Ot6hfODnrWAVGg-}?Hjgu@ZS&_`UKEr^#oe^%b|7^TxJ
z>GN#*Er^4|k=9ur=-r5(zW(j&pPo30&Y5;xxeI4X8)xvh@mya1z_;r+zy5-{I<eGM
zDyEIU7z+$F%DQUgswI9<GK~@f{td2^Vzidgidg}f0xHb`wby7RK&z=hO{Kpj>ZJv}
z%N%?fpEv~@rteXTFxFgye?h~PI6Mq($w(sgMr@MFww73$u(X5BukR+Tk`3$TuUfs3
z{#g=BEJ-3<<!BG)O)yFL^~><rm#<I%d74Zu3<Iq$fB*d{ya1t@vHl_@^C-$6|G512
z^yiUkUism7bmF+M+!(F<g*8lTcQHuUs=O+_Mei{uE+>8mOF9E`e<?qr=XFnhZT!PK
z=A0S?5kH5|(xD_el=5<z5F!cmLTdUu#?L!NTaY~a(|cOQ<crCNBq{WMeKtHif!p0`
zDXy7>6KaJR#MwNMkUA0Vp2X<%e6@&5pnvMg6hrvHYYwB1@s?*OKtEKY<op9B0Lw5_
zePa`7!z2d5V1hZ+e=D<9<ZR}83QFxTTri8}?4_>&FJxAm^WZgRaLPuMDFAY2mp_xI
zy{A~hHz@^eWL_s3xRJ5EBZqN?fdwOl;N2lm;AirV7$8|(q$x#KObvfi>Mg!Seg>~)
zHYIa#MSXV_=B1Na%7$K~S(0X%^;!cs%!0RU#83uOd@gt<e`^?d@wm9sK<6hOvxe*!
z@`r!@xMx6Q&a4rW0c{q@=GCT*nO8P`&pD-3|FgN3>c&s+n4R}L(l{(p)#LpYc$2(z
zsjs_EQmARvwRm^CjLP?BPg+;%IKrhA`5cp{$#e8XB`!A5>`EKrtdw1(c20&&f*+@_
z3dUqGJ>BV8J*4!^deLjNc6zeAr)TRH;V45$ubH<4zfd|<%n<sOxMZ)XItO=;YZg7g
z(&sjJ5~)s=x!I361qYBH-I|w5PdgibdSzYpZ3@Q8%amgl`c&j=rpz*wh|x`#Nz$dv
z>R#HE;ilB{fq0u_mFR0iHu0tOOSK#J=PynA5>#g2sXOKM9po}>!z_7v1=IK7Nm$<o
z4T%ofC)(u}o{Gk*X-q#$oVH7$#>xIBKh2yg#k>;q*DNPl1*Iw}J6I@=YF3?p<_~tO
z7U;*Wwaa#ltMRtUYq{baJ5k;3ajG08-w*bvp4xaG?D=4pV#oI|-7+s>%l8yS+O|eV
z<bU9M3jAHZmuK*|#?CqCL&+M+NB2svv=%cYbXxuJZz{c5S65RU;2ucyQ~rug{*Swi
zFcoXwo9~C;Xo|JVwtx#{<VQw-n(s0S&Ar0f+s*plJ|iI*mv&+#cI|*(#N@qZz&3|T
zGqC#aZmXjd#3D909=}m3(_;c62N)7idRo#t`1P!l35e#80arLT6mPklt~thr<JJ_}
z_mWzc;b$FM)t60iSD5U&*fc55loYovZT5Vc%EfI6=cWjrNk<1sNSp0{L}buwm0321
zNwe=#qL1{|?O~E>_Y1@B;1La`5@A~alWp1qhNV@~yQ15<_EK4i8NiO*>ISnUaZ}r{
z#&cqkc$n@WJcET_47$fd6-jIH1)tewec+VdU=0&qkDN|!jsO%Bd^!%un~!L-_YzAV
zI;K-jU=7gYF&%j98r8vnG2MwnloLmHxM5-7Ts>G>pTDZq#}1=oRS)N`U(#7@IDeJ*
zpYaH7&<cK%MXY5poN2Gd(aQII2teP_$aZ%^OUd8vmQ^BGfu<Z~h4nF`#1@24nuU@G
z{C?gl9N$iB24Xmyx`(1=!cMlF?gH+h2<XD-awA1t1<5~7RXYZMu&{bM!xw68eJhyc
zZnpmY9-5<td#0s78!i3lM@P#b%R#jCqrZ2w46+_X%V4)pMoUla5z#Ve^MGjSDN6B}
zXc^@^7cHapKO8N}J65;C2f@-0esHV|lI({{KlnRG${^`}pbR$qT%7d89uOvjDxZ&%
zo*;U}BZ6d*(wvHak<suU4-sdXsRx8dG18CzfSU=Wu@|%xU%OQMzEG2lJD4cXg}=rd
zFb9(?j|h8*eg9;xt+4l;DLV=Nz2Y8weUDp{QPz#X*MJ+E$qzml_(n40Cu*OGd}El$
z&EKfZCqmz9+$+ixUUe5$=a`hU(cNzFqeC+97S><w{k|c8j7DjJnd=@HP2B0?rW=cl
zS%Hw06wEib>!wMBuh-{rrbO7JjCP2h1|1PL=(Gvm6hAXVGYXK8#PIaEjcYl=Z=f6d
z!XMyxu!+EI-wWOVc@(nL@t*Eh@5w%j+zC<X(1H&@?hy7t^MrzkPS-ujM+9qBM6OZN
zcx+8vdy=_-4U!pR@>qbe8VuYbhho-K*b~bB1EC{q&j|<lNP96z5faxHB#C;9JLL}X
zlaJ4VkRdYP&)TMb=@b_(hZCuEyh59(85ERjKlz92F33IynTgG*KMByfK1(5HpTVbc
zKz+~?$lvs2#5!hW%Fd3N#sc*sPoQ3s%vBcHcf?kIn|<KmvvO`YZzoAd*jjy)<+SBV
zI{R(j=c~fr%IP$jMy0!(y<z~~#M_a(*7s9cym*uhXTjB5Z*&eW3(zWk+qG{VHC<`S
z4%l)l)bd@5bgLE4r|oI1ZEAV-lZQcDtYg-R(GI#PyOAs1c4-2;@N*~bp_5QjkJN)V
zg3e%n<p%<FaC6U;_uonzhu!Cq%iKDLXy9%!3|`l8wRj;ar;A28!RJiH`y~f6r;<`D
zVZ_wl2<^OmVJSJLbEkTQ8YPE{H#+9J+XJHWX{;gxub%RL7<*iBm?#uG(Z!wY(PJ%!
zZ6hp{We~cCK{7%o1YB25j)e2lDFMcJM-CT%rKsp~G9m39HvLay1Q<06#cxHyCez3S
ztIAM9Tb3=uY${C5)b_c>PuA^i8SSh4CcwvJ+B)`Ndk`TwIzUj;yqa4#fL_8|?Bn`Q
zFq?dd8Vz-d?u1`!9iuzUy`ZZ1>t}+h-WGRv@8fN8n+Ei21Lb{KCie*vGuTNTxMJRa
zR4Ibo*EH37#OeH~^|Vri6E9d`Sz$*JaJ7LG5Z1@r+R_VTpjQA0)-EkWMDrW<(tsQz
zm=I2MQFUHKnWzlJE@R%xn4z2O=_in+YHNIa-})Q8HqfZDm90eqELF((4>@6dl1!=v
zc_yj{Yn#BBd0zDIRt{)THfhvbfzFnHc}2sc4k(RpvD9fEoLqyY4;afTbZBfbW7XWi
zdCtHP-8SBOl?c`gvqs)$Ja!6YccKMP()O~o0QH5$k_=^g@Jk1Gxpir1*mT6%&FKhi
z;{yd?CFM)Jwp4s;o1l1$4pn^W6!_UqlPcKPw+4eq+Dj$c(nVCuw*%K~<bqUxe7X%+
z(IMmQWBKf#x*`MAWERt;%h>a}or|Rn?1+ryTU<6a0J|F|tC{rEsvQ+w$eY?8IJ1?|
zwKDO4fd(!c9izaWNohHCjL?<UHc>8sWo?u;sO|1wG+cu&AHiqTKy;%XS7J4G{#=fl
zN{X_Oc0+!WCn0F-?dvrIs`D{_(^1vcfP3c2{Ip09*;Q$e9K@t{PGN3<`!<08GskF0
zs4bApfg8CG!0U$~(F|uo=#i7^3I{E(JAX`gpr2L&#XC7wIWT_wje!V=JK;a^kGHt%
z;@*7cGkevZHu3szEhVwn{mSn7y+g~P-8vo49^Vf~>R-M-{U6sufE)^!VtP9im!S$j
z6ahDvL7XQim!(%bk$<My|M~XYNha2%@Fsi>xAxc9x6{9;B8bw0jPDFt0TQ~uexAYu
z1QN#jPATC5yI+5IKTf~5RP$cVJ*v$Ct4$h1G*O!}g2;8oXrZkwYGV~33rQ>pvC4?Q
zV;D=nb-24UNAmrW=Gina%)j#G1|=-AxoXO+0WF1sV=M(r*?)6-%8Zc?n-8<%>`NZY
zHbu52sW2mtgkO>I*I{iImQ0xu#CsX`9rlcrOd5eQLiJ9xrK~VMavw(3Y`2k;OkC!b
z*XXx{ig^p|SjU=-GkiB*s_s|N`r?!C-?CGz8vwMBp0Wj^94hITlH6|M<!v;QM}g90
z3R-5}4U%a!Sbrt_RzWK;Y9;RAM;nJRfL5D*2-rcPk&!i@Qs9MjEkn;&lU1*LPx_pM
zgVFerwi?EYs&5Qs;@CysMzS%&(%l1@3{z+G);3%4J^2zvDXWn=Wp0R27}RBTWuLe*
zQ+UlbN(w<HQ$C{4O-8;~p5b8djOkAB8m?tu^6a@5+kX|iDQcH@9=I-Rf$zrAT!*(y
zOGaqzTZ&uK8Wav~yW)Ncd0lF)gjFS0DaYP5QFbvC-mC%Rxs0inIzYabwUE7Jx~k<o
z`@8~SO(|aTayyq?Y8<GR(w3dH7n-Y6vmzjZ7&SR}r>ut0A1H+AoEUTHqNP*vTgm67
z88o;loqrAcuJYqhSZf+OrDoaXnspmgSd&4K7mfkSbnj_ot|osQ{Dhjdp^o|uzK`X;
zHGVY$?}_t057AtfL$m&+Z|s;NXy!8(1Ujtq2j%jm$T<~afG?*>53w;D-9iYO^BVhb
zvx`o}vqoX@QX>)@5YJx58tQt3B6q??p<)+NEq_x<v4feZSL$2?Kg#$aA^}smLYNDA
z?sJBBnC7$VY?>dqXerzBG6!(m24^)R&Ly&mipzWlDX65LBFU;5T3H9T4<i|#l_GOV
zCUV()n{p#}Eun0+98aij%8Ztgn$;}@`7+28wdU^T$mpHW#?9=s;lN}Lc_XpyV6KT&
zoPTEF%IS@ipXdwf)__CeZo!LA0Ajo6T0qt%O{Ibni$;oz)WRH!P%hZY&UW68P&<iG
zw;Q2$jL_VUUFvpi$F`3yWGK0*;tr(I!NF37+{YL<*I<dziN+RM)HYANr(Jz$)5-Zx
zAm9y6=a(yEKqtssqnD0^Ft%TlvyJTxSAPMa6<!l%?T0g(JGQsVjIfNWLn(uK+Sf54
za}m*IbiydqK9W<G81F(OK-CUw+XIDKHy)h4zyH-n2{!)coMdq}7pk)MM4#_fW^9<D
z3Zyl1m`UMCrt~B4!Tws40Nud2Ck3zzC;KND@@blVptAE!bslvQ>#dp2t{R#&yMNA<
z!-%XJA9+cB?h*%PN-HWcXP8T&nZWtl4pC~Vq>d&jPgBmwt<Cftr$~ua6O=ib=J7n!
zH?>CI4}TcWkc{veGw(TVV!Ifo>F{+?PjuQ!Lr1D~aXr3+r3vbEmEMyV6nxUunq}te
znq2%W=fw-?7gZmM=ITV3?dA}Xqkjr#9N!$Q+h|-e8VB3%veJNs+3a7XHc*e4u};xs
zKU7?(k;greo?%|N@o#v!S58!>&mn>Kc$;RWkTByZ`+*z$^??XOar^8MKX9xaL{LPZ
zUH1=1{Ve2HZQ)T^@Qpx^!T~BIIS!o90RjrS@gnu`oSN$0X9EKz8RygoJb&R8q^S<}
zmIVp3>_Ie;X4Wst?5sB$shO&}KPRu3b%w9k$0I{)>iClbVi~BKjVhn0Y|FWs$*mDi
z&U7fBcUE!UFs@w{GxtRE<B@97&syzV)8}3dwf1NY9cJ#=(Ck;>jnGwRdEXiej<^+2
z?Jf_@_F%nG@R4i}jL}HZ|9`T}8%Q-3crV*e1$tDr2gPkSepa>z4Z<q;B#85_jO|@~
zBha*R>Gksj{!IRnP>!+6VU=mxsO#kTCN(O{K(icgEx-M#{~IxFcrrL`_1eDQ`Z#FS
z(hCEPw(vPibO>YWaM0*%LnNM7UUKB#k6{ds-g0~&4?H9-YK~{!>@}alHXe>`IHhBd
zq{lAB_c8u&eKDB-FoK-a$G1Vc*L2?V*#q8)Q?G8`$mTmIq5lBMBf8<2VR}0i0Xdh!
zPXZH{iC#Or0kF45Upo^812tq8mjM9`69h3aH8GcQZwnKb3u-PYm#<_ymjU6oM`b%K
z3InPNNtZEB850CCF*P%nf$BXIm!NMuI0GmMP?y_pJ5GNrrM7fk=mIH?q07;4Pj`I&
zVw9&D%JQY2zc^6mHdnuvIR^7vI+VBhu6PAcGo5!b!h2<S;rZY1|9Fy&dc{*4fi|41
z13}mnJR@E}cmhEJi8KNP2MIoeI7$~`g*J}^tq*Tywms)3sPib&Y;+7kM<@eKTi%Hf
z!Sdx92LFEsSKdL!4Dne<?j(qD#!DQ}GzyL0>ys2fmp5ome-hQ((eK|#eqOEQr$Z1J
zVCGBYZo>18FYNICmG~BzGsv8Qo;z8N`Pp+qUKVhJaQ2i6Bnd018L6l$*4reuX9bD>
z1Q}mDslD|4LtQ1X<aw2%Zv0M9u|i?)*LOV2+Vw$@<GV^G(l+{LwU8tYCLv*kO`EFD
z_>7NHF3x)48HxMF8kQRI@-R(M!3r<kjIoKUm`<}L-@j(8HFLg5JXymFw`g%YXaRq`
zj6tNj=^b~)nuC;5S!H|8$tmLfn#+!0g1G0lY!FK7dG?Q3j#nwUpCIm8j<-T>mQPIl
zOALpir48XA^2L?StbSM?T$14AcjM#8jQcHVwne;D5BejPP<7XPVafXz;YE_dWN#V^
zOq3;!>PEg_u@DY}(<;)=o+Xw}R_d2eb36VCy}Z@tL+m>toR?N~J8FNIuG{8XCyb_d
zy-=-FUyQq)sgG&OD0ZE29_f4Ap`}<1#WmlVpxN9NJp^shDQBpq<SOg6C#lo>ux=h0
zr1@&4bAMgmyBLBYZP9Mi0M$Vr-k`|ca!@dBaxjeM4|}L}jlB$GP_6tP)VAU776Zk8
zr?ylr!?HSV6^5)rQ!sy=;P!Nrx5nOEziu7d19i$!a^uN$PY6A>hRpTj9eXLMY3ygD
z<}C)6duT?Po|I7RXYH$+@X|Tux7}?JL)p6NngT_iUJ0%m^7W;&CntMu>mqB5Hl}DG
zSBDG>;x9exc^Zjy*R8a-@|Lb$pE^&c%D4R2TaYSsX@WxuMsk0#@^NY;DG`%y=9#gR
z^0LaL4TvyR%Did1uKKcTQis&X1j=P$BS0nFV(M|E17(Gl%DHJZ{Wk8DR)+?M=>Tx4
zjXFVQ5k3rmv^4nhBzKrxL`B)2*@|s#oeqStYOv173Rao8(!7xF_6eJ-$zvaovGSNw
z-D#C{!a%0;1gd}A>y3N(%Jjy~7(~|s#&&_1=apK-@b7-lzCt4NdfeVTl_6A!&0H%g
z^fgDeJM!e4)^3dezge+yq<4f&vzd0JS((B0+|f5(e8sbyU0gd>xz$xAyJ%ONlq%u(
z8fT~x)1}H&SOjVNWaQdu4=R`DenH%q4>M2lyrYzMP%jhPu*9%%z@R&i5NWS&!gbuL
z*x|R&b~_ybf0r)Q)<Cnt&-cQ<75W?NjyT`iN_~8*Ivy^w!QIiSP6w;Jz~G|jwntty
zQb`f2RCPH#=(}>g%ifcz!fkcC*-xqK8!ud==n<{+k%ruA*pP9^$M3c>of6W+g#6EH
zv1=KR8o5tuvFj)pb%#P6J$!Tj=3G3D4R*#z#klbGf9k=ZnT7|PZv4<*fO3(7usWx0
z(BF*Lan>dq%OfQlOaJx92a>x&-$$1`XPL1*FvoJjdNWgeM0?EjxOFJkqAb~4HLt$v
z!#CbENrb-1R$5{GnduU>(ba&_pN8S{&5h4j++xiC5F#8Mo(;C{P0<L?>+hUy<pYiz
zO~XKe6amK8WUV5*a0)}4aI8|R7S@VaotOT2I~RZ4g=sbR;`&;tZueK*%UjSLgkVf-
zi*->yca#_WYmIlQPrLe!VPY3zRo@B7o}{Z}eO;9+MW~HjexuNMc{Z}Ef3MTNmNz)z
zOxv^OB&JNAV{j%>*REsRnAo;$PHfw@pEwgcPdKq{dt%$RHL=dT-+Rtir>cLf?p@VY
zz5n#Cb>Hi{IsM<!n;Pt>MkUruwrK4a94=vF%GCmNxO7<|hB{qqW_&N=K;#z*4BADz
z_GrA90f7s@4IWf?N`G$-J91-QJiT*YT{Bp`038<S;_HxaoVs(in2H_uK`|NyvfD3Q
z)0EU)iYHFZmJZRc+O1c2(HAE5bFI{e;%#HQ&;L~PAT_@AaY=eFQ~8^kQC~%Azx`>>
z^=P?Y4G)AHh=`?DO_0e7*X(66IZNT+gPXb7R2brtq!0Uaj7jUDN<<kkc3u6f?xmk+
z1sb3q=f0XN&x-msA8zgJPLPJWO}Y+Tv-Qxk-W6r!Y6`xXCwitVWcdoxXWGzKtbHN#
zP|cG6cPzkVUXTKTnTtu@+`-b#ikOuL2wI3k1=!O4|Lrd#deaD&(pW)3ev8&1kcgKU
zR<@`88ZG~|za)J8$NrK6QN5n(P65e5Z2iQ@<URVzD>g4~f~G<|o-K-nT?EDWi7F^}
zh1Q%G>B|OB(ae~t^x&l-M%OO)^^>+6zl%C0w@n%}=MgFx+pP&R2B4ATj*4t^O%@5Y
z?m>y=wSn7XYD^?gFQ(R&Hk(rHPM-mGGAW2d>|C?XM;(N@ON{qk^!8B&S+h74Q_gv9
z?+P!xL1G0r03NK8z)wsKt$db4vs_&EVwG7hZpyxtfbPy62-KEDrZBr19>Tj_$TdN0
zSGrlb)tT!-2w#!K2YB7-Onq_;+BP3k*<h+1++m6Q(6NEoY7!)(55G?@7R9*eqlcU-
z+!zq~*)bml!LU4MRHM_1$gYKUUycbcSs0-bY|oItXvC{g+hmmH3+`vR<P##>zaFT{
zgMsH&8A@s@s1=kna&NL1ReWB!7pvku35Hnlw1S7%#9jhB2(TR#)eX{mPcnYV>{vQV
zBgP^UjT9|T>$J+3#VZt<G!xVbyQ)Cy;!wlaDQ8$*Z!W-TY^mY=iGjD~HJ&enr%2J{
zP=(@ZrKEt3`Fr+`mcEl?o+=KsOm%$u&J5T&M7`%(V4T<9))f|yg!U|<@tj27?!LCs
zX3^~n{0W8i7myHz$o#@7lHsp+!{hF5%Oy|vSqGl7{(iz@l560}ywP&!n4ie_e$IY=
zakf1`6aSQss(K~RmZ>fM+wbk{d;&-a2jhJ7dVcf$QW_GCdCUCzizhr8*>Q(Zz(1<Q
z_;KiLW1$iI#(<zU1iIkF`0=W8%EI3R&CBoOp*E}50AOcezZ^6iVp&Rsk8nTvNhjhg
zIL-#$uRRQ{C2E-R1^QbNk?oquXf$1VBLMtH5r*|3WF&{-*s>U)+Oh*9SY!`<E;qh`
zqz8z7<5tYA=@H-)2&mug((`{mYFzTXFyu0gs=NvCZ5=KO&@!!|%6c3-_K*Kr7=Lo{
zTuL+@2Uxjjy&KQ<Z~Xe2s+>^-_@9~L^+)0n<wvk}T|_0bOwA}LC`cKaZsS}X!7TO2
zld0v(najG%Hs%!3%%ynAvKI4G<fsrGSWP#E!2Nr{OpjsiBWzLOlA(s-zkw&{(^D8W
z!R;>c<Fa4E8r(irN*|!?bPQmB=8~79k!!iO0XutH$$(nHtZpL5?d;!G2322AAGY7G
zha*$RTkk^E8bV!bA1{0E{DPm}j|&mGI|P2cU*?Zfl~eHm|JGML+khvgRn3pHtMkW=
zg_TM}YL}=cAAm;}ulGCB%(v4o;Q*&!P7hL5`xMQOXL_KGz7Fhyf7@ABv{2gE2hS5Z
z0BzcKbsS4*Q+Z=EU>$(^xAVW93roN~^zbZ92-FbSs*IHEp#3;v@@B0ns_au(`6%&t
z>QPFTBU_wIVxCOmEK-z4PV(AL)6vlS{6kj+X)@F=7j!6yBn`Mr?7J~?iepMX;Y&xo
z0vex$RXl8IG4T%o!hy}>E()p(?7;vYfRrZ~=QT69(WoZ44zjSbCy$s2Vu)1G36q}I
zYZNSvLa;%Pxa@l13KZ2XLiCqpc)S8g5ot9s9Vx@mVogw>>FW!#Giym2+yoeok?z=7
z<>4<RV?8|lY{Gu-f!Uf`h}8;o1krIyxzVEz5V#;qjuReTjNSga7aumb(}G|M04p^V
zb)g6g{3&=0dD25P0~}^t4$Mv))^6G~j!~OJ918yR;R%Es#{StUF&HYuHz)x?%+cl4
z1#;Ho`}NJ#c)xKaFiFNzT!o4ZY=-W39iG&!dWgreK@@ekK~Z(=NBi|kimaVC6A1Il
zDq-rsv;sX%?{&zWP7YAEc7?tefM8spcBgpXeUX;83VjZS9M#agOugIxmIrS0+|{0i
z`r`HZw;5d%rr!4PCoJTCfrc%@;pB6tonX_fN(Ycn>A%L2Qt6J{$fQ{021zxk@lwgb
zaL&M3>QjG(vMJ|zL>tg-g_5d@j#1pks<2ZH`+7nLp$U+l1p)&5!xHmQe=B=hvWa7~
zLy072S2#~o&#@L3QFW?nU5|}UELJC)pA41@{!SF>hmJ_!1(Q6cQOx!sUJ_}CN=`-%
zOq}*e*J8yuNR8sHht#BL3G_fTB}0R=QheYAFFEETq8=cF`HRE~H2|Ue*RZ+xu#?bJ
zg;|qYjS*vV*A!qB5nsA%;^72?zEixbI}6)Jt>2&v*4;GHq}#?q@Nz>jP*{W)>o-EH
z@rUXzZ#RS3T5SuJAg7E%qY0EJ@GfN@Yq#2)`I!MG<)hw_atlIKDTVWoc=@l(vdRmJ
ziL0)m7c@ta49k{eLw)O(BTeZr@7gr#nL@|?tA$0bjy}M|v(>pmXs~Z?N$9#_`xYO1
z=bu)GsR}$pOLc~;!3w{KH`+=YpmwSPqj|Di>R8^mnUXnc87r<@2112AlscsxUE>L_
zZ%uOX5|3n#=^~Hh5&{3^vNqoIEbkoEEr#SvP%UI7glpl&bRb%d#%<~!BSwx4Sf)Nr
z6ln50@q9oMsou2(#&hy~i(DdP2w6Y78S8ZaztQ!-c)#Y8)FiFyipGEaNWs#ARjWU>
z4F<uFCYIm|Zre#ErfNJIpw<E{xdH)=nxILALjt4v<FZJ>d8?#U!mn(UnYa!Rt6x#M
z%#CWMB>xr7z+XLG@ra5WPRWdrOmUuKfhr<6INT3-?%&6uAB@*dgg+}>+ptGxp~l~5
z+_$8^3@KgxI8-LhKKpuvN7@C~Pl0*Kq7`ZhSV~){b(~9dOm=vrE9*%=3;KIOD1Yb<
zH?B~>*MDm@#&JJw;$@6jSTH1fWrF^jKJDqd$jEH(G-CJ!&cbZE#g*wJ;0Mle(ONuR
zV$L6cqqR7d3&Kzc?M`oSIdL$1ry4_$V(Y!{ywk}-?AOY%?S`|_)k$LU)dD-2u`0tm
z`y*P_bs3Te`y<|tr_+;lrZMc}v)&JCeI}%jn!yna=z!XT0`9Wp-4&VJ3-Xu8>|I+~
z)dHHk6f;?GSzZd=!1LUjUfN69J${CvVBS^0q}&7nzEm+s>HB(_wM`F4oA_hwd6O}3
zcyJ4tL^z##TOF*Dx-W6~1!|=*qGX-nt?g5@aI<*~t>&q5+#N)b!H4VuR_Kj5xn&nu
zU?5n%YydhuQcQ>R?TrJ%D01=D{v|Cw$eQVIeU4B?Jl|f=Z3>1wSL*rVKnSg@;uK;)
zRe{@vJHK`UaeLZfs=ch%2_!-_j<oq>v|qj9X=PkR%mrwt(&mIR0~MsMRd>l_7SAPP
z{5%xQ$L(83yJj06?j<-KjKbVH8Z=l%PQ%NSnxwil#?%cG%tx>PrRZibUDLYE_$IRg
z&p;*7ojCD(WFP5K_C@lP=nx#$Rn|Qq0Br{Y^C2!FGdcG;{Yh6y%QBx1KeH#kcO?9H
z$wHSDj9@VO$yyfV;$KE}31rZn$l6dU`kt<8th447?GldXx`QIeaC4Uk$H(>&4@sG7
zCPc__qc@_WcM{U?q7bUHT^U}l|CFW%dkfxSiYJ(YIhrqS|8AVaZcgK{0d_XxNjF(R
zxC`PXVi^X6K5Vm!&_+=TX&=Hs)K=6ODD8C-)$0giv&V;ZSW+^L0Z9jldVUe%Am}&G
zddTC~0)c%j*kZjLk`#$)<<_ifT;pMe>ueIUTb8tl26p9MW{7Z*Xfb{$Sa6aj_~_9!
zJwlraAin(MjANyENGne!AQKTF0_&-V&rpq}MA7Spn2MnTCMH?Kaw4gVfGc5st>s~o
z_XmedZe9f)6VyJ9;yT%Q|5Hc<0R?X18UF`zJIW#S`>|T8$sPV*Ooy_7m4v`OUTgX#
zS}xIXu0h|9(NDzQQ)=qPX^~$?r_h>QL2bD%b2peT-;=^tVfKNNRNx5rz&sss!p`Rn
z|I4hot9go$&=;xrw%(NceiAGr@!oF`)rUMDV4^TD1}F#6xb`0imjxI!&qW19DW`@2
z=7#enD3*Ti)B^ZN<vTF8Lv#0*zMt>kZg<4YA)1H}pc}}Q*7tIAk@LYZVNC@dA_t_o
zKHfZEzHTjruGAD1>a&Yj#ACDjU`V-e4#JT1Wfa_`f&LME)&M1UEi?EkG-Av16q#(c
z;$-wph9jhbl`g`vfMJe3?4#Dg5az<7OP>5`@;9rZj#wWM8jLx~bI^p|=g(6V4N`Rn
zc3_`!KH*O|q`6Rq*ZdEF6+lLB;vj{t>sEq!-l{5HbD*+)MFYOtX;>f`v;tm-)lXqB
zezK91f&T)S3;@anN%D~uHGk435~bk?nnyO%)r41@h*&||@HOWdad>s-E(Oj`w_)|H
z_NIge{AD0KRFl`*im@g9Gq2WoREQ!p6odeC@F}@(D}5pE-@o`6ruD<c&7B`tpTuXD
z2KK;5@!UTVi6d!yA9l&38P!T(-N0zf^9xGtxTs5Z0dRj66OrP!T0{DiNPG354N^3k
zNf3uzS-1YH17~l1bPoC9)FhDk;WpzN6jr6jr-`T;%R*M>+2`h7#VY#qGRM;~@tM02
zeR}FNRUInb97Udm*EmsZKhJ7U%&a1T<xn74O&XJ5m2Xccra=RzxZ|Id8Ms+HTo@m(
z$pdD;2PibDu((ZgtDUK+3N6Mcc8HfRg*pJm7G8IaO2~wV;J#7xjI}-F!Og32o?80l
z!jN`)N9!ENj>I(Gr3QZYN5N)f%C2wvcEbxxPS3O6GN$@rzhTh~r!MAVPKqP)mnG1j
z3|I>B+k7cSNPQy0Y<q}=Sn3#0j9Ti_goaVe3E-GM)>hLpCH+^&p5UL)7*A(ESAlYI
zgtN~0If_~OF#S}q$sE43;Z`D^CCELDDex9SXJ)zk(^rgOQWVoer8}Sn$TU`Ea*$db
zPs%-bJN{y;l`$_h@=1m*DH2@=P{Xan9;a1sf~6};oFmjf?mzUV6NfKc9Fk9KG2nK)
z0?a?jk0)&ORJ;8^A0nBtAk>&wxGVW1Hm59!WBXN`gTDM1Y`aO5IsDSysIf$=1<9dk
ztwRwVN5uEn?qH5NNH?m+J|48c%Pdm{nRi-`jTZ(nDyRAw%gpco_!1N<s{(_c4vE3M
zShV~Gh!cW~=%4qxiw%h)#27J^@0h_V&4BvOx#Q@tA&&9&wNlnX&NuaY2jUarK#YKF
z3%#U7%M~9=GV8HN;)uiF&mD#H>Pr(J&K;<kjVfGnw^9SsL^qN=tQGTh;;);At<J`g
zvA&$9_&XKa^@C+wTAaWt+;=q=jmE6Ys?BLOo8ceK3(EpYKQCK3*3{A_Dd*IYkpN0@
zA5H8(g)4tzy5G${t2F@7<L%J1?!UYbp35cTEAkp{d?x9RZ?i3X6F~0$6gd)3(lTwL
z<xV+YPHyt@T}^(Z+oG^9+|X8K$RylBRQOyg5uBu34WyJ%Qg8=XY#0{<<mLyIEo=~2
zy*YPe)Lr85tD&yV7mbtyuW)<)H34o}Gp=SxBm7(EcPLWN(xfn&XW(WezilAv&uf!F
znUsXy@t-X;7drRajx3^^utL-iRE2Jr24kjmTk8$uIip7=8?;+A86BLf9A+9d{O1%%
zz9u`#HkQ8QR2^ds%@VYd7Ei;fPo@hcWXox!q1{kp<>DYo*1|=(*%rdDfB^pc;0De8
zKnwW-a|YB$GgWG3V{G?~JvvODNfpfFQ71cQLqgmtvm&1kSap$jdv=fVU>=UU`pC1`
zk`C&{vrI~tq3lquYDEhb!R_9d6$Lo+MMU#1PFlk0F9&nKe(-1`Bg`WoCKUUTMc>rt
z%t3;@x4e4;TcbbbLMYDN<p4Ki+Q6~uuo%=nh;p$=7e@2Hi<uXN_qo#$nW>`uu{xl=
zb5s{Tbg4Giu{dP;CGRA#3av#xG@B}DPGC|e7HopC>2S!;bk=P#VCgh>7P=dMXz8NB
z_vF1mVe^X)yW<}Z_g2dqgkP15Y3@c@Weo$L*NjR!Im26u*NP8erva=*><<fD9H)*#
z;BNpPXCt1I$+QLdMHAU~e}vzf4quCFzsL&l;o-Yin)q%mwHfIgHiaw38+{BEyfQ-@
zxj{7Rrh(i(3G94QolB7zXq{D88nVED2rS2T4|jQkLRXvEW{ATUCvLZKQ&)f${7Mtb
z(~$#qx{GaPH#*dY2>@P;A?02mpT3VpKMsk@q-kpOA%}Lm@mxJS1<^m$CoA>3dfP|&
zhNpLGV`&5J?e)C-NB(rBR(N#cEh?>pMn`AVME3;nZ`_vR-ZSST9(KG8A1T$j`U0E_
zjYOt?yJJIIl1V%fvch<a6MoSMoK3qy^NAlGn|`v+5dkfQIiQ=k>!^=SPp0nc8v1vk
zNO9elMV5{%?AY{rU40aLYBV;Jt{~MAdU{Juj!pdK#sxK`9(TVKK{9Q1DDZEj_G{(_
z;pi!monn)ZK^RY<ftP*S>IcDu#7$_1p3MN|(~o~JgZFdt>(Fy&EYE9o^sq*rLV3mW
z(p)dAv-GFmjGT6|m$36fM1%wz8S1$w7#AmLdr-;er=5fYDPoU=!>gW><u?gq2ylpA
z5~LS9KSIBuIT$z={^v&x#=-SJG7lIl3y>sUkQy-IK*xo=eM0jAHkU|!4g~>amwO1Z
z^(o;Y)Iqym(_ys(P?jyIajnle_}kz_$yNyem`Ez5LWbN+aJaXz@cMZ+ku@BGzD*4s
z@ob13Wgz`lNiNWIyqkiiQ91a26Y<?E4VxBV`IeuevoJIF{k!J)r7E=)Yy4i3AeW#5
zxD3EzERN-D#}1v(s*SR|U7Yu(I`G0Kyz%=u`;4F(D*$*u*eE(&C?&W_IAj`-ju}fE
zkiT-we)H~tPZ?qKlF=Z@gpjmKW#!TgUBsyzKb90LYue+6%V7})<<4PlxAFAG&+L<`
zW`PZ8_N=p;iD2mI+@uA-Nas&=O)q4{;sE%TQ$P(ej+TsdY0jwjXo6!Ch9_Fjt{*?o
z;!EM^t&HuswXQ;uzHT3G6z?Eu$*SO?K~OOWzdnv`FBZO!m9;G&I5G@X3Y*fF=POaG
zn1_kzZSc0$9`My}1zs$xuxYh+nIMzIA7^rV2}-Dh0xD5JX@hnx@|RX#@T*!MVF097
zoh_4QkwbV)IY>-vv>X@MJ8cBLbi9f;`Z^XYAB&SU<=qYyh?YujOk<~Z=xiCPEZkjH
z_`2Jo<`-0nV%-N<(l?E2MqLpw^q*avyx=`vZwGPUFbZ1c*bM=KMVi8k<OwVA{Ygqu
zQc3QH4vg{Y#=#iNUm^IWwi!k#Ab@fUP;2nI94k*h1(e^>`fWZd7gViy?rJ(&`ajh#
zc<T%7{7vHWaNIt|H8=Ex(taw2*eEWS#6G6ga7vH1w7pr(5JnID@E#49(iYLLweK`?
zL8B<bErk61c0{tK2=6_vw6NtJ+MwHu`9b_*9SacSW*diV0%&rxvDXYLZUCC$U!X5<
z;LI_OQWqNgI`d-^3YTi>?^ps_wS0BXJtJ8EDiCwLGKDs1F9Uv3XF<FhQH$Y+E_~0H
ztRk372lVgJ>tPRLKvSnzQ}V8dvhk?BC*}W3K7>)J=`+-Ops=O?bbR^l-|pBw#v*Cr
z_H3Ca&oOA;oq=<)3T#O94F+rixm==ytk|SG(?*Qls4#HgzBoGNarfBtO_E^vMczOS
zuWEuXAd)yutdt)Yol(K?L86k*^<%om?iueQ1NOp5=q{_DG8gf@r2|ye#=7eTfnE0S
zMEM4+aS`en0wMMG4^m+dRf1W+O?y;N*g_yUYa7d`WUkL~N{H&ybpRQBk35gR($yD}
zy80~o8lEhsTro>l`1EE*+X?GW?fA}^Vt^KITupQKlM6LMtSnD-=?q}lE$+$s);AJ;
zI@~fa`fzFF<@FK|!<bZid{qS>r2~s`Gt?Z|(BmaDp9RI_c6MU5P>`u3(mFE_94l(a
z1d>hYQVL|}kyM7O#(-u*36B*!+Oef$cl$1Og^}CJ>ZAHJx-N?u)80wc7OAW%D1WNn
zbMN8PD(|jGee^5ie1N3{@rm}|-i`<>y%kZZt_P;r{8G%0V=<j*o9`d3K(rIiF_)Bc
zhLeG^qYTb^0iBp@jh2)0>gzhs!Rd2yHlQ;VzV^w5hxsFE3&7CxA;4W67((CbEH|Kf
zwdwhFtLK|;krVEz_Ndw}aTBx|8bzz3ssGB@RJ&$sLooxNrPqvROKnskXjJ2Ej&y60
z2kyVIAdkq(pp6D;IOf{Oq7C1eO4*ige>XV<W<hr=hHNEBcJ-jXQI+vgzDCio?UgN>
zmFad8$M`C8OkH58HP6uGFOWAHTRS4z>%;ICEW9OB=_5wvwy-gY$7`e6tdblxdj<4M
z`RCkzR?Uu)H$8nV6aS|Y1o}uKa-v$_*c@{2q-EfYOVdxY{am%b&&b>|;4_#-30=CK
z)tL;GRG6v(Z^-FIK{=%o=?so4xttLv%I#{dkv|Yq$8eIBwmhbVBh7%~?}`E=fyvW;
zrw_4?FBd<0u`BM_7(IRN_~M6}g<681hH?QE1CsQ+?XlHjs1P&TM>6d+wwaLUR0DMn
zBiD%Bfvom#{O7mi_8-J6?2=ljuyD~Fb(^WDr;vL(JQup_;7N~(OGC}_zWzh5o1k$G
zW^Jb>L7q{D<7eq~iUbavg`q`>xC`WKw`S109TY!a`3jF^<;yKK)dPyuprmE|C(HmP
zDr=PZIw=wz_lte7=|T#2Vcs7W^`-(E*RKu*w*$Tcwa_<iRR-|oljqbP%EEfiyh|g3
zJs)6!qJ<rm9}T%f!Qw%oYhvEhq=RN9wwst&+x?-GNSiaO=A&nVlM+plBJJhzXR;Lm
zW<QZl<<}FSqwsQWxopd-eeswtVz2;SBR%x8cq=hr)oAM#+GP>RbVt=rghVnmlDo9+
zZ#W)YE5XEhPPFpyZ}ey-DTTzH+~cg0ZFz(W0~p?_5bpS}zS&y}9NU_4Qd@)sWS1kO
z%EbPOvZWk?**_qIjDrqC#u7!vaFwUjB;%hzEyU+&PQ4ceY|KKGFnUS|-*o`^uGbA{
zvEq>x8!Y&oVY9R{%|6cGBm7ZM4*ZF~E%dd|o>xJn8B#nR=%=cGvgun!Mf>A*gYTCz
z52lOJ&nST&Vw#LHIFN_8KNj?570MazPAR@A%@pU`Q0mu5nYZEXdXlB3-K8k~6t23g
zpYnm^&QJLbU1tl;){sJ-FO>%%SSri<r7+A#KzlZuDn<9!Bc!z6*CwQoq`sodT1WAs
zqCmi|G~vPQ;KRvc3E7s(ZIVHM3~%JwPM(~!SOKO(`^}wG)Xr<L7yEk6Zces)d=u5T
zW#^_BKLW!STFB{NlXA5Ac^lYH8OlQj)>?LsbWtv;)Kjigv%EE&^vnqW$5VVw%~C57
z3(A7VpKf^o_CNN?wxp0?&8jmh+z_^=Dx%Xofc{X}!19(Y7X&U0d};aDR2hDERLvOx
z7FYsuX_^q$YGpf>TMe_3n9MVT#Hp#uZBJ@-IlgV~s$FWr?as$XC~VBAk7SD;nPMv}
z{<%;87p-pY0Q@b?3k}A}{2y5PKL;ftKNV0boevR=q!MVE&c_ch-o3vYk(dQn_{Rmy
z($W9N1uJax`|y6f8t1X<=$WP9kt&*)^_0s!e~KEI#}_S?+cb#9Sb1ZVe<e_>$86i=
z^^YR&pU?V71$)m@0W4fS$LAV)sdS&hj;45E+mn;oH~f}nD0&0}txw~~84hz?B@{7~
z05|nQP~$pnPL!Ny4Q03Y*fKabUvIH>4JN2`i1N5|7sTC2hDn59@0uOb4JvmAvwtY7
z<^3qFo1&BwD7z0f27%fcd}IJ5)bJrHT_rPD>M8b7dYTYIH%(ZiPP^GdFFeG-dC+#n
zI-|ZwW|1A-;hh9#Ra`cvK8k26A(VnvsXu?Mlj+yU+%n`FEZ`2)co>fdeLF8MOLiUL
zYJb-9TC-AE7u07h5b}TDRlh$z9vt<W#yIsCEdv3culq#O`RSb3V10n>1Yqx%|LegK
zuxG_A=ZnoOZWrty*zA}wFGD4*2-k)8BY#fVK(uzY@Rpt{`8gDL2z;Y@dA^*kU4p0i
z$VjOm3iYauEPT+N&Ey$J*N6Y%MuNX{Fbf9Z7Qf?IYnEl}e&_UREH9T9fO56AIZmA1
z60QW7148?XkGqElZf>!)@}X-o<kgJ)sF8aN5*wtoP6fMW!rDZAwS+jXn}g4uIsWvs
zqf)OYE5twQYQQtKXQN;+MGC_VOF*@a@r2DSIH*w|jgGLz0+ohd)um3lkC0U%M?>ZU
zmy^{v*?D|^>KLa89Jrgw_XEI?4yMVT9u?dA7rRV`_HVTikehRaD%+xDMR6CqjOD`j
zK4sV>k9~U~`&s0&y5z6{P(($dD6-u37sce@36I|4erwk9>1N#%hshdP+FBmTq`FPo
z8&sW6o9Wvw!<FGPM^oa8FhXcqh)}Hy`1|T5D`;$A@36!B?t1&)S_P1sRU?!dZRnMe
z`cq}mhZ!FfW(}orjQuP9*I^)ypJyU7b5Y85DsN&a4FqM=3k0mM8X`65WTq&wK;ttM
z1n5}v!C_KK5<)Yhb=0<<%AMtu{%ec*52M**f$_hpj3#Nttge5&+-Ypmhhh2V+io^G
zIfy-E_~Q8DxioRC3&xepPGvV;=8;Y(DxzseL+i3!+EN6~5-Bpbx8;77B+$MHrZ28=
z(9!fw4*Tg@$;!0Rt{p^2;|#+&N5c&=hiA1-M+D)ret&k4Y25OxoQU`40p+v#hy*dh
zTx+2G=9P8e{2b!xGmpc|cLmL}C<gzib=YMskw&SMAJyPHs!EWwi)4G0U_3s(Hpm2<
zNFtagryO270n4-bNB|=9G7pNHdeY|9vt~=NhMWI*R*Hd(*$)XmCbf-xB%j<pp{xZ3
z<zi1N#Mr<KW&RmVI?j|s7RfO36O5+f8{YL?Nm-o1CY0`!*mX6nDTu#_%+W|~G=!)K
z1(u$iyYvdt!91F_pB?0^T$^m>^t`lX7qUvbKPbveUKLNn?*YeaJl%6io#o+jERKg@
z`e<>E2%Z-<`z<oAd{rcCn_*h~^pEzQliO$2<)6)pgOJ<x+T1E7;3y%Gn;|AEEK)-z
zE3&t=nz#1te{>PI&VBl>QaK{YSIT?T4X&k=Z~Jmp=G6VD6fUS5T$&c4;>?bu&23ws
z2z(BkOUa^W8vs#IeA2E<BVvA2ScYj)mHwuMNqVS4mqk|(=ga#P+Y+Y!jq?3w|EB%#
z#VQPJK58U>Pdx_GznpX;JFUyyC6uIF7=OY<odC($*8VnernbmoG9Ug-QIm<NkuTEr
zo+u_yZSIDVr5zNW-gh|FG+jd0xsxP`)LZm%3%_^{)ZylK*)<4YlW+N;+f+up@Uh5*
zJV>AO82+VJ;rtf4)MwGXzk~AG;nWtr((xl<h0rp&&RQdQSrilq_|1n2z#K>Lfie}K
zb|hCVdru?F=Ibk1i##REj!DD#DtSX{BEQGUgtr>AUz@L8Mfg(E%C3<<;;xiC=U5)s
z<$>c%LC1QacqaM9PDAa~@`rIfwbrA>N+bTVC(H>P;g#4NT5<qx0O~#MqlY1W9LfQB
zR1OV)9uCR{PFRta>g+BDm_)%@={!sjn@A3mU$4^oX+6U=O;BsK)x43NS<$pMJg@8r
zOY_sZ+y^0VYgC3X(&E_#qg(b;qgv}8r5iDIR0qMQcgnIXP2EVBBU1k$K^VQ<g(_}v
zrAT_Ri)aLlju!vdvww%J)OM`pxdXGEID@#sY{rUCOWmFYEh0-Iz|?dHH8H6_r69t6
z+WLtLshs2&3D1HY?(ou@^z!Kr2Uc!t8QNu0w$=jZ@B!Eiktb+HW>$BmSek=<<Hg;e
z*QXz`<(-dlV$i3uly;~)d2{$4!s@rD2@^G3!UO=R{su!Za@AAiH}O;V@ApXmmHW<L
zpY2xWCg1!z8LLeBfadsBPlj(SlN-$^<qioGM>WOE*f08esm$Y_Oy=12rTkXnJNG2@
zt<epHi9RFj2n|pwM0Zv){(WSL|Jq~!_RZJ!&vrN)t*Ty~{aE*5^kI*zK+q<g0*}#c
z^bN)HcIJGw)&oj+wed8<Z>Z;EvzUOSSvreser3;G8I0XLz&mQ__S;|^RfPiR^rOja
z&g4m0ZE!Y!#*C+m!u~@}3i@jo%4oKLOZ>15n&6LYSzQc1B!wx?Pb9Wnl3$_vN{r-t
z0RW26@Tb0hfuB|FPcI!e^x2Ckg8Ar|jehGTY>(o5_muY2WzcybDd=OnIjlocQ$34Q
ziSvd#Ql8m-|B{1vN0sMOuRSXcMAo^dBy?21BZ{}m*Xu-}Y`9&TCa)Z2^*wKdam7+s
z^6{1I@$tNPnz@$P9NOu=0phR3QGORlppG}l=u9L@6vuBj)ry^&i8k}!phhk;pa1Qj
zaIykLatYCa$@zSpfZf;Hm7U*al=^Uv1HF4ofVca%zc;Gp;`#s7jS%v_z1!#bQs%s%
z(E5N`9d5P389=zBV()WjT4RSNMiw|$7JIlADW^F^B7TfW&@xs%Dx1V4#J_RAnDdv{
zkFil=Vzm$01%#=~`=Y-Ad!8t&FGNWw2T1&eLZ#n>UYi2^0K=rZ1DV&RJmQX$&t^GN
zjLgVtKtKRdhvV0aPH`zc429)KUu-TgQgj&8K&$REN*bG7#|IC*mRF8)Ja9+yx&(6$
zUhorCr`RauUx>8;MzKEC&TUPuqXDBejtp%1QcSP#JB2mpdC`*PER#I7Rx{DmvW+#f
zUlf$4#|nBR;3^H_y^uKmv*xy;LW0>KQ0+4+qKXbK(G1P9$o76J+^+K33T0jGSf7=^
zz7-`8&I57=4nZ1id_mma4azslu~<P{KgJU7EG5-Trcg;Tw@}nO9QN<4gW7Vl=MTOd
zv?mz8a8<P0#2GtMHU$h+F^w6RTr_8LBS_%^=$ua(fB_*%SZ^Dq_<YXqBaY-IdOfog
z5O#r|&H&XRv-zI4G$L$u9Z9xf$RE*OD9*$&>!ONJ#7~MRYIV*5KhP-gv;5yjR@6~#
zT9}wUED(4PltPvh_dm2Bf3P3C$(`&>B#}A|%*v~%f3A>bwdp!`T@D66+YjMGl$B$)
za1PxQ1MDSN?(NZNcZ*}NFy&X$hckAV>?j0g>Iv?iFO|<D4so-CF#aaBn<cZT^On3=
zKviy>63W2QnC?lGLfA?W6v{_~O3fH1vNo5Rf)0;mjtJ<}p)GGvG()SX1z>ZkwQSoG
zc2ez4ydu^Z{G1CCUg><p;tG^CstY+Am(e+F0nAV;NU}fj34=mygAwSGCplAwyTuvJ
z8DL9s+O@chi5Mhj7CF;W(Xv0L_})>SEc^i_R6{JYnn{=}uwp#O3$O^0F{Xi$4pXd2
z*lj2=2M2R^V6EZ(ZiiQR=pK2|Uxtvo9{~-tfOIlgt`9)AcTQZGG&R^TIud{+&LWat
z0I1gEdG3PmBjJr$H2<>=gc4NCc=0s$B;fqHEH>j;{+tY8jSDs<un1?*4?P+b7n_IK
zU95dZUJ3j`p<TpkoV!8-?HB!=^1EUIyd8|wA9*#m)<L_0shya$&~>^hZ??k*7Iv73
z3u=g?HsUuWZl$X(shYg_bngCgD{n(OJRnyl*M_8?HnIR#c=(pnmI0L7Zc}5R!nmtU
z2*;MgAIQ4JI}<<97VES1^m}-~O(v7eG{Ggi?a_~aM;=uf<S%tRszw$}bE9Q`2-~Rk
zOt7d9NW?RJld9b&@)LcD3ZQn>^=6=nL%Q#q%MDa^q~PFqmod<Vx-0A}C8QIC7jU~j
zqV*_Dyt;r{r-2jrl~P?3Xquq$31(h0BuldG)Dsq9qQE!sN+`*nYvvBv$0FN?9g%Qr
z1)DNvK&TctDI=O6m<9FeljpOA1i0vePKt_=%~0c7v`pwMuUbE41RGCLyCZf~dJ}5Z
z43(8SmhNi?t<aJf4}WT<97$260LHsi@K?EL#)_R3K-17e8Lktf(JZ&{{P6`lGn;}e
z`7_7Zpp&<0XI+h4Or!Q@OyO-7KR*|*>lP#C!s-jiE^=P8!6@jd?zU=rH(`)ypi)eR
z=H#6tj>4O}c2ub$URc?h{f#1H?Wm-oHCCk&CdOkhpR~{{;YZhOyLlt=0E(a4*>`V(
z2lf>q>VZGI3+!C!a$1%fY{s>8Tiv^URyaJUL#CTL*PeU9A2J%EmLEqAVZvFe^d(|%
z`B482Q2vOlX$0BRIX8tT(KR_sXMzSMMyNQ|t_&o*{|q3o5YhsK%8{Pkje<ewALKCl
z=v#lP4_d9q&^oF++~508185I%2fvK@281g1g~yZ;EHR<QQlb65hJxl{17++2QS*0*
z|M+X0j)Rtp|Inujy>PYf5c5cbb(BLFBRPBw0pc@kHzW3!OQAnfTPf;u1*Z8xR&;XN
zz9E=Wb$J+4=ISj}0vTw3DmjZc%lvgb`0U_V%Heckn5Sc_Nlk4R0^A4nD<>WoVS9>?
z5Z+)EFi#Ujd8!ZhR}6>C!9O|<b=;aJ#~r(u@c~~RHJZR3Lp8pz^K+6&ya^obs7y8)
z2K|*v!<5<J;)Z!SdHz-i9RULgoKnaUkwbDyDC*=^JQ-GtqkChCU}N~IKczvM4Zu2^
zTtWG^zVz~VbAQWJ0IKNnfXf14&XKKebXgq>I%jGpRAXv3_2oFVji0sX0Xi_%rE|gT
zRCR*)W*ea%_2=f?Mot`8)%P0rf)TgKVUEIN2zx<8h#;0B<F(ja<&}eB&}y9o3w}88
zS7fw)gRVt6b8f<Tp0$X93SQYcdY4*6BZZ`7DcKXBIjjCRY&9z0#)qr`_1p;!2Tx3+
zjiJjy2~5lR)2VN5)qfhoo*0l9U67EWzC%0^J@y-<0dkn-zjgsmpfWTzI#8jRuLo4H
z6|sfy5)vYkW6|n=R>3*gxPc@>JY>KVNI^tk$3OH4K=;NmYXtG@x_2M)sThBE;ck@d
zCcHoxK(7yq4WERxigg%rU{&3@2xnG&`t>f*r75_U+(ff%+o>BU?q{ehMHnw7c&Pu&
zg;wkbOs+ezMJYFuOtmt*wB3#iA#1d%nb`9D_F0~Ip1APQ6GN$p?YsT7n6V&ev2)qH
zBvuS2K+b_YV3dUe?HexMu|KVu6YXFv6Lbet+Epb2gN<W8T)bE&m6e>*)mSBhd|q-i
zU5YUlyH{G$zAiglZJ~@5`G;$kTuV^@T!|92g-zqI=VW{N?@xPsUZWCvJ^f7MP`jV*
zgD||)Kkt@Z(DQd0ipvYVQKEhtE{V)@li=u>0kTo5XuY#h<cT@gQ48gcRuLkP%6M6e
zNJe0$NEfG+*R@KB^|h*$=+RqY@wKcM>w|$CNi+fQgIw)c^#uoD{b1?8TFO&FKgh&}
zL`x}K|A-hzNY3XK*lz}_&lGlqK{NaeRg_qPNAA>WDlwbuSNS1T*RM5NoXVOWBWnUy
z0;osSI547UM$L;{+LSD_rgm<!KFTk0#!}yP^v<T#Qyk|GMjG1Z4n~ADaNAEV&0jpW
zfJ3fiNy|@yrLGb;A29vnl}FoSWGfhFNjt?r+e1`b!B*EvYd!At&;pN9sYlWgIiP#6
zx_zz`9fyF!Dv8e#2tK@+t-8gSxM5S#2{1}Y96`iOZJXPTX*qUArv6DsLBz9ssWNuO
z?7&oP!H(zcM@xI<pB1=P*4ax_18W6TaXD~|HZU?@pe$;!uQ-fZ!t3mGE|8|m4d8vh
z&b>KloT;tBs$82ucGUQC^YDI*H+~v2dG+Y=!$h^&zgQ!(+?nk0lQ<ZRejPej2dE&P
z)0J?Duckoe5|UOyOnvw~y$0?XPWKr2wWj-X_^H5Qnli2)#K1u2ytd}<%_z$<%aqk?
z)4klwM&-YxDWVZu%fZSjvx#61#>KM?4F2_EP&qW3TgeLXm8f7bLKm)tus=`3DscL#
zkU9lyl9W?TCv`P4+-XJ|z%Yr{2Xv<9rKHpgz?<Y4?WH?1jB3{MSFnZySl%ZLzd4CV
zaVC8q_qIA$I(c??^1ZsMcu+KUL6*&!(SN+7EF)Nk)H-TKtO27vqmL$@Pp}{IKaJA=
zVcV*YCn`x|;{h+`kB<{euQr0SFZ*lzHjpGU{B)9`yuka(#-B;bhyM}@0SOu#;G(!B
z-0y>yxl_NHu(6p=Dm5EV3<a|MxxURG-<E>~guVli!fG@gDwo{lxYvihSb7cov<xO|
zS7b(a{P_UB-d{uez?Kw3ysDBF9OK9xpT&xw`={Vno_JE23I?&St0$}_2l%w^QG66R
z<2myR+NH9k=ZLADF`MRx07fz{+}a!73y1GeP-vZZst*RQ`>@LdewiZ-+607Qc!!{5
zzAwS|E@VJ`hn`-)^Y_O-{`dPBa^=$d#m<byj$uu}%PplIeL&{t%U<bospE;zOpk!!
z=WXe8rDGk+MDHNmK>Uq`<2J+gWI##atZ?JS(GY?`Lzmx!F-Pw(V5OHwjJT@wMU!>&
z`*2ksXYXrB8PtP7F9sb_@)&#QvfNHeLTuU-MnssD#@8pE7y|}6!bcz0G!$gLGPD$k
zmWnC@`bbZ;JABZbpcW#wgwL?s0m|)Ii!}xkuR_ZVN=7Jg$|7PlyoioQS*L47%yg0$
zm^APs)Po#`Q#@P;P>;c)0Y#6po<FpAJssY3>g9HIEgcA&xO=uq5cvF8-ZrOiO(0Ne
zw^dDmmx}lSrf^?OZ2v>r0je)zkx3sm5(BScr*J4_7sY65*Ib^qPxyr$wokjvDWT`y
z<C;T$JBTMH0s*`&)0~%JRA7UXx9=hGy!C1~T`7i+9<&b@K#xYX5M%=iGLQ*^UyvsH
zsR=gwg@gc>KwE)ngqLbseOK0fJgaJ!ti#-^bOfs!ZPRZ+XC2%vdA&{Thfu%ZnF;BE
zLnM%WCt!NEm!>|FvW^&S&e2fXQaLtUgUEgC2hU9#Sa>a7?YS!H2Wu5-a!jn@pg<W>
zSD*$TcGj;9xM0ZhfgT88x8YcmtWLX*(3x_`*SKm^<6~TklWRUu`ES_)?zrvvoc9ZP
z>|OHWDx?7aw2J@bX-sky+&sj4Pb{YEmTp~mBvT^_N&$aCR)TUjf~)!vN5|30US|v|
zkpG)WJ`3dvU6um_CVc_MpfVzlzV42*36-n@K-1vEhfLIbs@!5f+dnwH-U9n;oA~+8
z&e!`9iv`{^zJ(*bMc!E3F%3h!1;f7t*FY7u{!SjFXf@4zNd+>OL^$6k@O#?^zxmYh
zyIIp@Lm$y7S8SElsO8$K{L%<hq#E-IMMH8Ny&|H{2QcwI`AxTl9qY^&UyZ&va?ElA
z=y1GGf1LjHpG)4nJ4A)o@XW8+Q%QCvyx4MZzR~z>t?sX#R!7Xd`66SX2X6C3nZD!|
zs|`@OGsXtpO!F~G4oV`Ej5-|-68+inKJ9kdp&~rE`cV3hi2=MhA5sn};~R-)L}{jI
z(S0Zb1+v;=MMmL*tz#CbM@0)-BLRTIeCYjNIq}?Jc4%X?^$i=B;wt%xDUeHH1|_ki
zu3G1pOoV*>k+zq4mMx`Oi2Z`W%*JK_HuW&94^){NK6O`OTplXzhTs{3xP*p!SklUP
z;??P}8fp7NI;+p*VL*LbhXKP36Ka#xA$h5#q-qtNp9ZcvYXk2-@zRfduVp|d+YgyN
z92ts2evre+<}}!K6$GuzE1anyrAcO!LYC$hd6JEwEpjRn5nfS;e%cXOq!OHg0;=0)
zwEfTWD&N$)0t$6;^OWhbMI}r{BiKEhHamKfKsL3D016BmW*PdmiI8zK^iv}zAFn^L
zNZZjZ>CflFu<ccpvFeb*k=+0ZE)mZdo<8tEfsMTYk-sA63Zoif9zSL!SZPRMj<oZl
z2y|(^xstq-DT6vi_wGaa1YukJ##Jx9g#svtEF9@BG`S&^IrzfG%ctU-c9bAABbsa}
z>g@U*!7t~6OjxT}q{v<9ImwJ4)J_;Xln1ivw&^wths3V9AQox{r`7;726Ftn0b~#Y
z4;wfapOTkT*#0k{%L?1}$;Qgr&@H8tbut8eYwW9I-Smh{`7Ok0TEsUWQumc8M4lVF
zUx;XzTwu3`bp2(=8*}%96c}=-^*eH?LLe0uj}Ewo7a|0Qe0CQ)f^xG2KwOVHS;%|Z
zBNPpd8;3KJD%e1=zr+AY+CjbewbZXuFhP6l+klWW14zH|EU@ib@7S=~o%@dbCe`Kg
z`SPsb4M)!{rN#0^nrk(nh(44%F1tuS15BwIJN1ukb>Tbwk82VuyTCWCPNp=a&~QD|
zgiO4;JA7$8)I-ztWzhs{+%sWrOa)Yshqgxd#r`1OIK!2ab2z}+!cvL){h9WZ>YtTA
zziPvDkH!0L->X5bQQd{i9$K7ve9uJ7%oc0m;aavKyr3W(a#uI0N%n!APjq+80coEK
zv90{eI-7x!-8#}SxO*r^@O4TM+j%r~5H>Qs(pR6@ccH+IsX+vGk(R>XK)ocDAeoJ;
zrAhB7U#Rd<U;^MQ;f~$)LerECaX-URYUfQve$O-W!|z%WR%RQG_C1w*5`$==V>aRN
zDJK&{{&yXf`Iy2?6^EHZT-l5tuQU3zN)4#_7@2}?j~-*%L5#x8uedBEV!zSW0ybEs
zO#Uf1oQFi28sirc<xaYpejN!017#{R!EN<r7}vWHZVjCZm8L_Q8vK(Op0sA9(c+*F
z>s%z^!BN<N5oq_4?-N)#PvKg_U*9v9nyozl%r=@CE-<CJIxa_ITB?#}JktDu+r2rF
zpKrqS8=Sw6{De>@F5n5i!4yF(C;sn&Jk^*Q9q6aYM+a2u;!6O?d4LAIK0LNk>1z?a
zh!?JgPxGn_O!bvsj-iy-eURo)eSlr_Y_YfurT0uAM6uO4_iXBaDGWeL;5ppylyYwf
z0_Ws}Ojn1nDp-N`zlX9dAKB;A2~J1t;6C5}8fIIPj>xRXb0OO<_Zz7THzx3=s9$h$
z=-7=Z)uOdPeF+0@u5AM`cxq^HI(wN4f`<x<x=ax6yGfGv{WL^7nCV!`H>JQI#}|P7
ze6bep)?t?d9y*D7+jVCFQJwTsz5ZJm%55r5Ze;>JXxOXgAZ%RQD!7n)>r8GXwIPkp
zdzP~!yA4W=KRk3Hq9uwY$hs-ow1^?4_m@*APC6l6mumrb)?HR}b&nd?oK};(2A`Hv
zGvG|(<7z&`r1;q7A3mo*40C1bmpR-=a<?HV1gz*GH4a#@b59tF<9#`h3`w|Yj4!4!
z39&c@yBHU?bYe@gjw8BQ|8eq0p__0d9P>F%)_{(+ZJpt6kkUR}){#GY{gOFD8!P84
zz?D(}b^xFsib62&{bEG*BP1}@J>OM<hoB=D;JK3<P+N{wxcIE5lXTSFJn1^2J9Np1
zBq>3aq0>XnrXwVg?A5w-tEYQ7cZ?&D(~DrOb{}{>9DjJ!nCMu@RA4%Wr!@i50Pbh-
zXz-VHL8P1XZkVniWvAY(WHr#4eoWgf&oN~9(ilJ!Lvh4GlrX?uSVa9Jbs$A6dME6_
zI!+WCuGuR>3Uf53?;s#+lBRjJX#G8hb8VEelDm+;{;jSJBQdiCd|ySLc6o_TM+{O9
z?5C!rYNLr*k{SeE9S*w}Uer(1zLH>G33KPx31_NbIwZQTpt-F%AkBHHkUQ0pP}uy>
zGUfon?2uUOw}w2k*76>xstl!5i0h^cP&*}JP#xu+IdJLqR*=Uiv|sQw+HxgC6K#z1
z8({IL0n?^Y#=19kBkd6o2o;zXHqEFDmH}>?=?4*vACr=339vq`*%@WVLaWt^Fs^;A
z1_E52bv?`%FyNwFTY2!w*TZSoP*ht#Gv)xRTZe1zo@CB~1dpH6WSdvz(@DhX@DAp^
zpGe{^X>nhiE29c2cA|8YXLi=9^JtRKnTHurR(3EdX<czY$ICc|!z|)2m{sXM_oLMP
z>%Yx4ePa}iZ!1|1J)4JyX-Bkb`RiGTXU5}SpW}uD=_`#+u{od;^<3g-hG0vii4Otw
zt#Y=Muac^H-GK~chkTVZ%>7X_s~t2(8jO8|SzxSS>o5KS{igIRXVBOEzj24Aht@)V
zGg3p>=-hH2W_+IeAl;{Uv0<!-6~oknR6-&T=b&z>1SUJFN&HJKQm5l|!jKtcvVV!>
zu2|3i&jW>F<sxNvV}*X@`73GyD}b|XHgb=LlUl}{k8HL=J9g=xhq=AJ*d2Mz!F4HP
zYPEkd()_SKj;x<7?+Ax19ZrQ+VkzsDahb9ys#x{e4A<F>S~h70lX>sEbCgIv$^e-&
z+*3qrE2!FEx%K(WnzcC{<>X2nA^1RIIqRm|XpmohrkHuCL6b{<Y9o;PRe&Ej=|_cm
zzOzyka)V=<%_gy}d?2FyLvd4Ld_t5uVip>7gb1j!e3)MRHs0UP=zq}FTsV&Oy%Hu|
znhWOWu+VraMY$hvO8$MEpc<V*IP$*`Zt(ypKVh<})o*N{wgU{1HCao0T4&BKKqHj&
z@GOhyiFlmaSL~tpJ-8Hv8h~)ywFX3ldUfPC-OCuRQu#`jVxJts>pcEb{EdEws?V^>
zo2NaFbmp`YteXZwyyqW7_XH5ErF5umZs~OEZ{>=aBY)d3{)`<>4GubtzXI6qwH3iY
zdYei}sUk=Es=hp!`~dIZ7q7KMJ#+aon(NdYmEcNDdk-yOxSLOc7r;Qd+QW*vVks&)
znigi((qBQ5kw=<rMIv>#S_<D>(o~sZkxqhRAsuUp^~Ne-i@GWJEObZq4=rAcqPI+a
z`7)*CW#3EDz%=fTuRraKe1Sf86s|h6fGxE;uV63v{(w}KZZV;slZ8Y#vM<F%WG-H3
z{_rDLL*{maN#Ki#4u?eRi#D>1*V{cYLeow#E+IwsFcrQ3RPQ5G_0>Azs0{cBW}@oO
zfC^N+QNbG{=C=_0V|4WY>+SppZ3W!!rKJYy4DzKY2OZ!2lSz<S8v{H&oFW}fn~kpF
zcDlM2C!tv=ratHnEmOb9YK0SaSg`+WTIKevP{xAxYJSLnRJ{s}nWFs6op0*nHm1Zb
z$yvZ0Aw+|<FKiY3!&eImP&WJk9<3l~{6Kb3IVxWlT=`MKs&OIKQ`hX;q|@x&R5-OF
zt_sQ+V_*@@u4Mwqq=+KIpXO3CCHJB;Lyr4_38#c2Y#Gnbdp87AD~pg`j03W<v^#ea
zQi3PlAEgp^HTF7;+k?-<qV0P(1sSaw*^CvqKp($Gs@Avcf;CYGh)YjOc2d;>x~ksy
z!dxti71H=6fOnMDrpHQhMv+`gF=pYl;SJFj9$ICNzbQ*A@lCI0v+>gZqUJx%G8MT}
z=EgC86@R`__G{RXf9~E_MJaZo0sXXKUWIK`3Kg?TIUfrkDp!zXOqmBG_D#*O^UPSl
zg>t8Vlw^yl!raRN<SLtu($7E!?x0kafkZ4~2`u6Mp;2a<=t`(bZ%)|sd^qB9FJGg@
z7Op$?pb;MQY`cL&3BZS#M{q&(H9`(D@mWpya_yj}C=GlHm)vc5x>zhy_qPF=Ucj+L
zH7dGC7f#IumZq85E!E9C!aL&q9{`3xdB0iJX#HYV_UK~dUxZ!Cv0Z;>YF$VZtH?uY
zfx$sK3a5!u_02GpW;Bm9TQ3AkK{PQ(mV&gQXobxfDmdK~V;`%3z$OSr=sCyqeC6VF
zcjgjs$*q+Wb_Di;nF76Kw(YseL^~^;V)P-oJL@VPo1R}@E_5khmnLZJh|0)IE`FDr
zA`M1>tnf!aWl*0D66AkG=<3JyR;+J)`&CY(_L?Z#QZIm^0i)=S;QKIFDKiP-hJTo8
z20D0ZJJ5^zzNT#D8EZPbL)oip$cpVKj!AS(paERWWZBW(^9lxMyaOy_V()}=s{i_b
z!BqQeFau09FcWnJ7X+gM&5TN8nv3qmNJis`i$eAYElt4)g=&A}Oew0mk=7)|otzTL
zrfG6h*tJ$!JK@f}mrcx@w<T{e<8jhWZZKmbKvP^(gwoXtQop=hhD-db1O&I3P^&__
z>|}I2*~MKrWrK_PE&B_lo6o#u@RL`W>q2JOU1pwVF?x3n+c3gg%f8cPCu{VCw_ktz
zl!~L)%Tp{*f$4vgUq0=1=xd#sG9boto`Serv-I25(Nmw^A6XPO5}@ryg8!FG%MDwG
zAF)}XV4Ci)Z&qA8OGD2My-`bB=nabPHyrDRO$;czuu<9L;ah=B;BH5=&N{x)jHzID
zjL~GbWk9(LO>&xL$ksCuW6ys8szz^x)-W@aA9HBI4()$A47NMU5wva+E=HX17t@=x
zTZKQho!7jN(ie_qa_n9^Ne%lDUfud2b#&(OAT@XO`_<haM<tT`&tR28LCQ=;{-}*N
z<Yvd%pS16W-1~Sv*-%+dCwtYOFyiC(13~%N?Ag3!?*k<_iEi2@t{nIvK+cTKfgO+H
zLY~c|CWC)^Y}$qTwj2jcDJ<=x<dq+qy$@*lDS#qVa!rP(x|-?Y*WB(11~p@fnJ&jF
zvn_#XuBPm)J(aw-(@60+9<t=U#JMhRCYvXLcIGBOuV`t9LL$!8lB-q{{FZw`T(vCD
zjQK%TtArtVBOKa&W$Qyvm%wsA?DY=7*55f#{|5t{$a$CXhCmboF*cVW>?#$Plgc|O
zmtgBL27l*Q=nsVD4BtRtz;WXMd)p$n#hw=1o_g5+|HBTc8B(HT%XXaRAls6tZ=Rf&
ztcM?i9$p_be?PxHd^x%}fKkqe;D@*G1JE5!pP++^-VI)(GMaAv9R7OJ8ok!`nGUP9
zGjs&{9qHTc9*vE(0%!$<-QRD2Ft}G3tqp77&VL1EgB5T?k2(Ugj=nV=NWTMLJ2D~P
zpp&wQdX;UPqCyCQJTgTG&|!Cv^rf>jjzQ_%vSgWXkuKkRnG=l4rQppf!C+NLSVbY$
zT@u~!?*pIj8zMdIAmA#Z^U9>Ec@tRkk)KX_^=b4fhW_De1ziIU8F;~#+q4;@8hch)
z6MvhaJMbw(<L(gwo<=r{_CmgbPwaQ2He`)>Vj8cOYsZnH;;Fi+zy*1#u$u(cMljEo
zRdndrqlk-)oSN_R-94*Qc88gG06nRXw9=R?q$r=|Fxd7J9)xOp+9uY57nut^Q>ddz
zQoStH%_7>lzyU1Nwfm|M-()G&9pfK)b$>SWVFxyo0H6>M!wGK#UHo8ScXC&Pt=@Eo
zB?GazLjD!bD!Z;t%~;hH_M=*+XptSN+#{!h?0q>z6l;qvO(ooEeY6R$_C*n^qUxjo
ztD9rmg`(p_^N&m``-nBE(%}_23TD9zvslkqX}!Y6yLK9}%bmt;p;e5RSBXsdZh!U7
zm~kAdiV;QWEO+$=koA6|{5jXMXD3K!y|O+#;jM<8BWuD7kA)so^v3a?Ig}j>H*g?S
z4^z$fC(*^K%}ojlf)Sgac1ZG?uLVIXimjBM^Ve&4_VV`d-+}}NP=<EUKR+Mx4QuE$
z;sD@sQXhW(JA8lmvxk*JKxukLxqtjYzkdA~x+~wfNGQ2Rd+*#Vc!Q>hZU;lD0bR{~
zj%_sY#UdFK{c?=3ie@OW(5(*5tVUeAZYB}&g@5mB!qJ2wf;bj!c7;R}%7m^Ug`Tn|
zzoPOd^B`Vg*mjdh>@{#Os-E#I&pFw-9G1qb)z6TlX+^?GM&h#*(sPWt#($8TR`gC`
zYQB2lgl=RXmppipC9PK8xE}GnId9Z7?>4cF9&zw6-+v;LHOd6HC|24Y9(~1THOWWO
z-t5Z6SE`rJkx|by(==gKO#R}H>8r?7#q6Ew5j%INt39Zs!-Cn%tHEn{Lv59Ha3OD~
zqY6~KZ<>f0BmT7L>UKv*i+?KE1%Gx-rWZLjQp}|S))-V2``;a+uCRMFlt@xf`IOxw
z^;J}G_k2z^oNVv%{k8FYCtcSKh(JD3`>^VD7X}hW!IYG<_t-vYl!w>>p3)*)d%pOC
zHV<{D#Z;GZM$M)V&@CQoxW0a)1Gj3nZ1#El6wxEKykwhmoAb!BQh%94X_P;Ttg|aC
zL99d1YP4RmWs^nnf_Cf5NKPfYK!cZ>F+!`di|k%1Xr^<$AEEUsQ7P5Y?6SZ$IzdPE
znzm~?0c3tT8=uevMo`+o0^wL|iAiWG?K+A-TrIz$edH*FZpnXp0Lh_C$AvzM9vpjN
zye^0Y@;YqQ-C?cXu76nP#}#Oug4iNbteTv(le1|};cba$^R9;;uUP4s<I9ny4co(%
zKs$lz*dDD!rd{b>bNVU!tDA;!XFExSR0&7wBZxmbqH#uQ=<e#b*OSsAMG^E1elClk
z<FM{j#kOH5qn$EKO-SE7N=sN{XZO4thXG4%%gW1Qes6m#x_|9FLav;cCGlv2BUJ5p
z;jHZ~6$qAkmri26-+0PVK7^&7sz*pXHhm{>a?XJm$?9=h9M!0clKy%2fJUZ!X;(c^
z^kB!%>eO^&)=Mp?8?vI#N_0Or<SVlxjdzJ{NF$%v%^=Ejh~rS1CXBo;&@Y-Vv`(*|
zFN_$r7EKeWi+?V4<G*n?<z`Fr#eAvqQZU&oxm;UV8mJe7+TWd@(cC`aNEB`2rw%xB
ziD44lX~5B;-P}FAzdENaPkq8YTmBC56Iy=os2qGm%ikfNZ~42$cWC)L#Eq7}XZB-T
z{>ylimVeVkcW?PSiZ`_U;6OR|87+TDd86g;`TYGYe}4`IF1@$q2Z+?-raSih9pXmQ
z->v@OuKz;bxb5#&uU7+;-d}vswv<)hY}MJo4xX+a59ACT@PnoUbMSW#2j*bEWi~J;
zd56)!QXyAP20DO`91Ls-|Ngl^=X$q~1(H|6!(FBV-YSE3@rRAW?2YrO&3b4JVQwhe
zJMQs?yIB08p|+%FUfA-A_HKJxn9plFSt~VInEct|xcdHmS|j#<Jy;ADMDozJF-yM6
zV1mx^Y5p?3HYYH9x!$|Dp;knDGno92y_TPF;J^LMa}RYH{{xRxcT$&;KspoxHaM37
zt1c6l3*<p4m)_AkoB;@zd(u0pf7L@2*brdSiue=I;`A^3%zE_emj}TdqgK|15p!Iv
z`=EXM6^4Rzx+<^M*Rjj2arJ4@#%il*<p3SbB3`!<NL#ClTR9R~<s&sYzAnKRwUYy7
z6BJO3YX)=GHTf7dG?)Z{3Rl|F*<>jkyjIurONYrbSHl@I0{dhh_pmLOe>Y@bMfBA@
z1<on7eo*}Rk}Q>dHtjCFvAZfbT1A%(mJVNre{b&(|IGLUvQx#$vK@oR>G!`M(g`vg
zF+2|^tC;EEKc+t){x?dk9x)zsh|adrjw8>)HsF-ySY<t*1Shsx1I38GEd7L7Pj4s7
zAZgb^W`-e@jYzO`;1WzKe;lNz`73X=#Ynf!OfcDhhMCMS>9g1hS}|~1X`MH2k-8Z9
zno>a2n@I3xMo;4`5U-hcNG{tL0}!t$uJnJ~#*-~<bZGl7&HYln{!*>tyRdZvN002z
zZk*xXIEuBFvtSL?f?<t7hw$U=hY?1J_N^lDaa|!16IvnClKivae}7C*1Qksfr8#}t
zQb0x#_)!w*yIBb+s)%=6xi<>*YURkGm-N!hCEgGzGiNFNx)mhbtUHuPBBc(>AocWe
z&TU<eN$EJI)K8Rt#mzX^&Q#LDHy<KBhcaP8Jp>#zJ{U81(^RyK1FWQqp3qSTFY7BS
zYX2G{bUoe)Izn+Wf3|{B;|6a}XeD+zK|SYlm}e~1&pj4z%X8`(<s2h;etlu-COmfJ
zaGsB?439G}v7w`rrk|VFxAel|ITW2TKCJB^37&+9xOsI8f|@mJ1l5F*_O|oFY_ddM
zkH2*EwOlDk&-XspYqK@`IWm?7nFHQgJ#&SC4+=ud`xxp>f8jc#Bdt8*c=wH5wY!f!
z(=4;nWnQB%EH(QIN2Idm^&D~~HI#ZH?Zi*Et#*2c?-Y1~t7A4Xvu6-wsP1kQD&Gpb
z(*zM2HqV4%Ac!1_J>DRQ2M8f0`bjqkVHdb_&at{=pPY&oBi9H6LuDh5Gg2}Ck@O?f
ziCFg#X6p7_e|iJCf`2<tmqf3nRAV-)FXM={_v6)gbL-y^N3k3`LWH>3lqGB)!cwGY
zQqkWs|2^4h;8+a9yaVHzxjHEaqmn2#=BlYqAyxchq@Of~oREjr<}KRi->&veE4Q0k
z0Xr~r>_Hgs_<LAa2-_h1{3hY3q}Yl3O=6`%0yPf4e|})IL}q?4O5fWowGtVwKDt@3
zGMbzraQ>ZYwQQ@zL7vaEZW|QQ8ljPOpOEN)8F^fxg4_PX^<iZH9In1&XuyDqt$ZUi
z>@*b=!;FE4*>aqvQ?0=U&w6a0`TAv_KS&BARu-2@24;Wk8mcP~sy!v-m3a=*Kdz^g
zNwOVFe~m)qNjj0P3nZ^uWZ{}uc@Y}SQEs`$@6>*R5OarmU>A@|T)KPvDU~p(bQZ@v
zjp<C@Q-GM$Rl!=~HrEnPH}pv=jt|%8gW4ONlrek{cc64bY`2K#+#9NzlGxL3^-og8
za9Vq0BZ4z0v}XpzRP&hf=haC(%dXT}h25OPe@*?Gv$~8dBt{N%ag<n?_x1w4&z<}c
zl5B=tV!qW<Od`N@N8FFEl~Th=(`p_gpH1xuNTFNd|EZ=4*tlrM+b5cz6Quy-YROH!
zX|f|;T7h-+X+|lmUR$Gdw=H549T-(Q+pNc^D40=7-!&-e=fFYo@Mx=|HuwUI`b?JM
ze*}{GY|Wi}OG5;Ow1Rsh{_JJ}i^?-A7}s?hR&>bj=T~<gizs{gkY*>eE1Y-w3qZhJ
zW-NnS)Sw$}2A3apEQMUziu)Y~&&h1mrnaEuAz1I-eSVo8XC~Xbo($}SyhDd&AJ{M`
z7J&Y_TD3^%Y`w=7h>4&ZCO+7#NDnAXe;0YbL9@aH4q>=zRsd@i?Wda+pbghHZFA|;
zNXxC%?I3clv?-PSa+1Buv{p8QMtj|pzVbgoKk8$P<s|GZ8nzRNJb@aP@VWj-BSA|o
zb~GS1c#6)8i;s6l8cw6>(_=d~7fYXJj?H%pbKElArEfDFA|%oo7?jmDdX*O}e@YyX
z{=X>NS1WschX|bcM&TWI`34j+H4|{PpwDnN%Ylnv3nBwm-x`Z<aJW&zMR2hFJ3F=A
z(KWkI@i<`p4cuJ+$4hpCz{Q^#XOPTk!>Tmi3zC$!9>x_e*VyB?DE5uz6ve^*<Hf!~
zP9pY)7yHnqMe4Q{NatyOLx!_<e+wi2NEEt31=Hgy7;yMb4FS%A2=tv^M=_eItM`kj
zehT(I%%9p*;Z^Y8d>ey9tDTwAPCB?xQ$b!pSx$MSDwpQ<@`fMMQjiw_4(kp+DsTKD
z-Opzy%`7~Q+I%KCnZV)N4Fa$A6OtdB(S0g&7=0`-C9<)9_>dzKb`}voLXPu`cP(mt
zqK#EwtXV2>&1&h<(s22mY$gP8Df|;Q15P()R5l^C2e6U14r+nAV_FGOo-^OETsj;%
zy8Z|5*dR=okw7{W12-@+mv7oT)qe+f`pd^ZJ<x<~!Hf%{v|F^11n@Ivz(7iTE9|BK
zTS$+M1fKHQe=wr_?h;tYFbChjOoAE=g_+7fdxvjbW?#c@9*?)rmd!ZDnIuAq5g(-^
zE@21bVWwT?{02jb5o+8eVC6A2o+9@YaaKIytVG89#6710BhT`5pVRj>?tc+5GLbSS
zE<AC@cJ%6&XYnau@6j&|BhJ_r{Vd@|Z_+NDfQ+p;c!2kkLKegdwlryVI_tnQllX=)
zb=}2BN=qzXQ)HsF1ggq;sB_}s`V4eCn%5Pc=2;P+%zqg$TgkfSgj!zc(}QF)^OPyv
z<rkBVI1oMAukfnHk|%wUQGa^E*mISrTQbclv679@7@N~5C1$%~X=I_R;h~rKa*C$R
zNV*nuqSa-O=Z_7(=WM0QS=u}@*Q>NPKVxO~HEB)U?cB^Ym49{qfwK`@63O()hkPzQ
z<>_o(gjw|X9-YG=TXFr6&KV+&Sgm}a&M8V@^~YVeWQ7}|nJTj4X@9>Yc&cd8wiak&
z92<8DmjB$G{5hQQ9@flY2f4y$Ws`%Ht~ANqp}CwpNQYffe!!cFd<x0pd-#l{Q?bo?
zZ%7LKo1%Jz%<?K_r<}<r#V8S$jVG0*G-z7&Lc!k?r@hPfU+;%s?>;{Kx)KHjqh{cM
zK7D)e4<u7W8@FK8GJmoB_Sfa}!-sL#HUJ6?r&=r36QGF#ZOU4h%woI(($}bd8!ny_
zi7m>1zkOMTk)N7iMtgBLBNR^Op^p&=HS`Ln6&^tHl0MhFapFGYlPf5Y7|PicFYZkj
zbt@{)F~IYIB7b*39+DBsSG-{xI7l2jj&hjogUw(%vw5P8TYs=o2_l~guT~NOC3i4*
zt+SjtitI2qL$A@Ho=RlU`BaZkkBEez8?`76Ic4&gDB|IBVmWZ9gr)xi%e#?Q9IYmd
zkAP@0hQWA<XjCL`42^DFUchSOgwlETyR&zvM@{*07jl;jIZmV<K@RG54~IZLan{XP
z#CSqC3C0E+gnt(+wmMCmR$_y4m6ahn<k@H3O}@~MV85I6Q@>RC|7hB=8^MlmpB<f<
zNde_hdd1=FiGxhq4%$tFt2A&fex?|(nf(Te7I`FwVHT~%w|10A3Z_qfdcPeNWOu?L
zGD>#Fl_yWBl9wGR0@`rTh<EiYhnC21WE20E3>hEf&41poCRd(=$e;=5&MgKYmMCYm
zm;y04DrGvm*9elzp`!nN-gn)3M+HBii|ZC`$Q+XE9gNeAVwO}=3m#{xZ1nbn)Pji+
zgljp$WH-!%S8`Z2tiH+T4O}!tYK6)MQ90_@)#;Q-=bCcw%%~Cp_W4L~S}BGM5OiMC
ztqWPeDu0Jz!{M804%FbDy->|*N}a^i9JkMD!Y|qQ9oJ-~DJG3MTA1y!AwNaTX~Qwu
z;K-}XOs*bL-zX8@>@eqi%=ha8MSQ;mZG=G(&Z_0#OBjaj#y@Y(wKS~Q0+@$~xBvL^
z`H!D|VFb(m>8B!aBwR6Pmv8B(fDmwK)YtS|6MvwFw1E*vg_kQ!7}#XW!ZPq_d4Qt>
zbOa-H^s66*nXoVv&0@h2AR9k1189aa4t~cvn&5qH2MbmOL_oTr;eYvE(pvu$>+>rA
z(|_9Ec`8=@ZF%^-JbXx8Ra$F<Db{kKyk#|3${APT?-8=Hma};X#vUm?97f`bGv4OU
zM}J#->r_O0S_)8TiGk-4OMzBs&La)oc?f&u9r92?Z5ZF+!$2!WijXRpX}Cs74EO>O
zWr@3y!}ae^-_p7-SoOa7j{7`%mn0GIKT3q;4k%%-kg{y&Tz-0<-I`0USt3Q15lZ%6
zt#%C?0NjVAK_}AnSPA=Zm>JknsUysIiGOQfGfp5IFsd?c!?vgCdG=6J&SkMCVJ`te
z8OfP|DCNLu24{Wn=SE14am@SCGBTDCOf}3o$LLM>P^!A5k#lR>H~luV+Xam4Bx$Fu
zvCoB#RhBjpvgT-JLRJMP3fbS6ep+$_kBukTQ;qC55Yzb5Pd?#u;?K2Q7;N$ek$(d}
z6s-v8hpLg7FJww4M)$1Sa#fQl8b4b$MP7fBaU?nW8Zg-WR1}9@=I)dv)Be})mn<t;
zEvA+rKxHU7;7F?uh5%{y?+XDQ4fAry-ASmYswatRdt&-&yQt*TDk;Vgt8)0jdBii<
z93j5bA@-|ht{}|T@$Srj`1&R-QGe4VaM=<9+b#Js#N^E$pdwtzPLL+XRlnpVdeOc_
zVf%!N3HU)C5>Cz%pwm7S<{;O)hw`6Yav1zCj7GoYHaWr)3pMIi{p-!HdWHCibxWQ~
zt%Sl!BOdV?u#Zgqg<NnItI@p-P`VJ?4-owQWq0F}3O@c0G*sYSU%lZj*MC)O38RYj
zkU4JM212Z;88KnBLSpp0(an)@`#atN8o%~ty`+(W569?ml1OP+X9jc2gHlVl9M(A0
zz+Y3ICX>b)AMSwyCytmC+wB+2Zh60y*(jbzjv0hxn10+f4AK{NmAOY>eFhKbc#pnH
zqlq2(dqZEPRYXv?(pR+)$bTtf`Cxv&?&>S}Mzk`osIQ6&VmNy}eGLWLgDR_>h4Bz;
zmF-Kp*^DzVM-Err-&Ls@Gf@D0Nc=p%>h)uyz2H7ZHPhZ<tmyFyF170$7<T;u<cn(C
z?Q=t2pr=F#74E0F=10_pgZsW5+CkY@7tWBV>z<9b-5d95IUfM+yMMRP0D`Q2<m6@l
zS`d!H*cJa@!DVe;55kiRZbu9zu4?n*8!<UkE6VwETkx7n=O#}64oJJTRK|GH8ut`=
z5#nD%n?|9k4=Ju>Lu~3&1R11z&n4NL5pL0u)JzaDX)c3DtHEJowBzkHbI4JAgiT8n
zPt{W`NamIZpW(zO*?)zXEJr3s`+T#aitr058yoSsakbB>3yxr1b#AV!E}4QZN{^Hr
zcFMaJ#@_^q9*G-)I(M*byr=fI>I)>U7Si0U`tL3v!C@_oz88={9!&>|_HB%ioN(*L
zllYD<Sx7_|1NGh>?aEvYHA`*da>)BgitvZ7^x~DmWQpWR^M4ihzC)NSl|h)SSLbrA
zdVR=?O6!|7dkk*{ofd;&p`cI?dMxM+?Rq_seJ850xv$U;0#4~J9K|>>Y(tZbnWy6r
za4~4oAKpAl?tY+ejudY=_+f;3jqDQ6$a(SUA^d-$RZbQ0S0V#4ua#gTh~iF<(0E%t
zO5}^`{ch5r{D0lYYS-lDr~VVsDw2j;7>AD!Zk(4Ar;N%Q%3RmB<m(+$&LLx`Ro8u(
zd#hY6nAP%Y-gVxG7@@cKchL;4dlgF`K)YI1^AMg+vf_Dhtt~Z|9c<tK>-$USbA~6{
zy)J!o&CCo+*YIoeX0u#PoAFk3uNb>^#0nNbyT16kU4PguPjd*C%HRGkm(P~=-L~?D
zyf0APzz6D|l<Ni7;di)n>@sfGd>PSU%qxv)6(1x_<z0MRROvc3Lz(|N%E^dn*Fmng
zc%V`jYjAbU-B*-dojp=0G4!BZ>y^xw#Bf1q)1lZjAIW5#iq;Jw>iA^$Vx>Cr_F{#K
zCjAdEpMR4Uww>a%E!f9jFVn@O)zpmiTnn`3`k_+SaA8P&DxWgWqU~<)!k9XuZno#0
z9E(kVNn%9Pes_rT(=$hq&baOpspo9TyB_b|>#k|v#g2CNS9`JDEp>zFJu7X;H$m=%
zTCHxwWtra{*x22)-eA|`-0t8Q`OjRuHfduZJAZV(Ww`q0m8G_$ylUIZYxuR2Pi<SB
z(&IY-QZ;%xvEgag?7#JKpHQI5vQD?LLDNN~?>gs(q7kzSYmSeZz2Bl&Z%Pzwr}0jV
zrLtYR&QBgGifi7fu|wjgns!oIpLzWkPNz4}%9S=Bd3Sn5_&;DBNA@JBt_gPm@mFKr
zn}0n$&)>z1$hun1Ax=2W%N3@-MD_Ec{udssDDfwk;J1{k#=58Xy3#Ds`h~JflJiaO
zTzX##oVCaa!MTxYk#`;G_+8POFCV+w87bCXaI|T5xHP?XS^jv!@7gsf=+=tOnkDM{
z7q)-v<t1$=TyO#b9jWSj`^;1G7UZ?H&VL8mN4PcE*^Cfwy^E2#EhSM#Y&UckF6S7i
zd(z2AZSl+<h&S3N4z*4&Zitxlc}}}gradZ^gyzM&?&*q*StI4`4w-A)WAoBrEL|q9
zx`mjp*g&8tBUaYqmYZ)uFo>X8ui$qDD3$t@@~#hZ3sdW3=^|lF-9)rsAl%8+vnZX`
zr*Ec)yZC8I1eaYW_C9|70+HX}_;HfhxbAhrWB4`wQU3!B0)Lm`<3SVxF*lcy|1J}k
z{O3Eu0rR)N=sWBR1T)8CUYC(TIursnG?yShIVJ%zmmFX#F_+jhJP&_(Ih+}iQv?C>
zQMUmU)QD4{t)T}=-o#35t+84e?!WID-qqcewQ^R{T8<INbMMY2hr{8SGbD)$TS$SZ
z<PvJ`LJc`dms$|2U9zCs)i{M(xJwkQc0Km*Qdi^_h%*IB#lPf|p5Uiufg~}$tS(6j
zmK%N+{{bIqAnlD5!P|eL>Z12oOBWd(HXb`bk5;=WLLuXI7jwvL?TL#Va-nXQNXRYs
zE<{7NGJa-(srqNRQsCST&n-bjmsXV_ng1^6hnY+Nsz9v0t&#(4%HA2w<3inXMO$_3
zSE10Pn~TOIl~F<9O&f=%K_vCYv;<L%#Kjt<bi<0m^DXs`)q;N*TkkCeP2(gL3EEu<
zBPqann7v|n$?mb_)1;DGSH<r#PF`t2SLR~owC1s`0+GAr7_hbiKPI?QE#dcsfvAlZ
zdRNv)%^XU~eT=0-9H#_cEP1pW#vw$ySPMky##FqOF=}cHrC9ZWH8f_qcY*mVtBFEK
zHKp#4st{#t*TsJ^ZW;znW87e7DphwbJ_(hHx>!S%Rl6-<X1VtU-KyPmgRW7Z6Qj`5
zXr$%PsC63<2aQLM!MjbQn~qs*TE`YjXmy-QtD&{gQPx7U>b+I?(7KTc5n}LGH;l0Y
z`d1hhM9p1Ph&B!aON4o}#vryn9xO=H(m%tRvA&w1i?@F;vUdhu>uaVG>`%#BmwiFz
zJid2RSjYG5^J~3x4_!p6blGEQbW!ywguhdl{Y}c;Wp9DBJb5vKM+!f#(Bn#KlVn#C
z!tzR@Ny;uAJ^Av>;ivz-oef`~pU;=W^UIT^KOLT({}zUC=8KoJ#S8yB9S#2&z8}7Q
zLF2=Zv(ta&3lWM~xrk1RKI_u(FXE>piY!ON*8#0RAO1f7X+C@xz7R}CX2!S>2$Hgt
z_UEtg=eiBsyM-eHt-7V0BbE-9k4C@?P%K9v8`coIw|eyC+AxN%r{+(WXY+Fy^sgVk
z_rG7fS}xyS{B;<9Ia|KEJc+0CH^Z-&Z{8lfoV|Y=emGvf`sdr__vg#m;^*VjS^V|&
zaPoS7GQ6XhQ>5WyaXK6?my5HL%jN6>KmX~l=7YnQFs3IT4(g7MWwqEGf|R2mfY>Zz
z)dHwECaRy@v*<G>^nc}9VTypT&*xc72Zz?Sd3N_fWvj7zI$I5jlf=yJ5eN0|cyV@o
z^7?;jOb3Ugdk7XxcJ!1;twdx5<eC)46PlGMRBi-IZGgLK)~Xot3P8mk6fC4I%0{pt
zYe#$3U^$(iUo02LXXne&v$-7{@`h`_bFkzNmczgQ^Iu2Y*oreR3g?%vUmxA3R-vYz
zrlX~uR3&qtiyc&fTmZ?uiz;c2=~}B{_;!E(dcJu67A^_gF@1Q3H8WeBhde(1?fvra
z&zHx`Sr{>Ycsf5{qLZi2ITBFXdwJ>^E(?zjo^CYt1oG8a!?VTw^!aT0VtDr5)8VJt
z`{mI_j66I3Wd<8CEpU2gYW*`r#>M<{aXPzLu{VGB!|dhR@i+7LJ(QdXRm(?cVS#^d
zol4pGX%D-L6$kAm^d!eTr7XFUWJ+U_Cu!1?wVpma$vO7=asmyD)SMO~^58bGu4rbT
ze0gDDG3WL20+R-eea6)#<={~HVOT6pVy)}c-JMz=Gj+e2zdZZ-?DW{%U1e|JaedgZ
zK(}LdE_#(>6vGt$tiuwJ5={?Go#cN6Z#`Wo0MX@Rrp~kTmp~x2Jq`+}f!WHmm-K1T
zs{v3}6aj$0*ifs2#7Ar8lQS(4x=d@w!WPi|&&RV!*jWjEkUOPLISIw+wzRpSah0`z
z%`xq!DvhI8SUae)AQvJN?4X(nM9Az9oi+>KOq+%7mo|%Tq)nYtq)sVPrxbswQ;Jlb
zq^X+JshZTOn$)S9ENKr(6_m;g3`J7qhI~L7Q|2WcmR)^>(ouRLwJsfzRF&FiTzCr3
z3g7Mn(9lQj#iK;uHBfe+mR>w!YH#?~caa2cHUB?j)mDCxZ<*lRx+W}4@NIj{df?~F
z^V4+#<O0INE$_j#ge-yNkez?601t9)7%`!eOo-l_o(2EojJ#QsZ^@d1+i&sg9c}Qs
zLHfq(Xhh#Duv={aw5H@t7}K0HC$si&8yICybzXo%ihSpmEgBFwlg7FwYqqMyqD|n3
zmX!+e9*i*5?;A$=?h^>ttwwy3*$~V|<+TxBd{b?=8nN4S+@bic75#s#fs8<`bNh@y
zRuY&Q-1`B8D0!#q`<mK{Ml{()O@(5~c^B1e87t>8X~FQc@BEHCvcpra5M<oE#e?`m
zy&rq9E$pH0!XEs@<2rk&vcVood3cV`xkNwf$>4zsn!$+yQ~(KtDj`HaTpA=5g)?hY
z<RP(~Xagn2xop2jLpOgkJ{mta)s!j<0NzDaT8k_@b?$p}thA?CyB0*(!t1suymLe~
zx*1W`?iW$5ZbVd_*kzsAWo>(4k`3&_kHuIcY*GYS_0udt<xO-W$S!6b0RIyGsPthp
zM-+&=L38l$qn7&L%+60=y*Xa|cCjI@s<`X>pZD9<<%mFECl-Iu3~Tm>7@z3Dp|%e)
zOiMdum}_cR2vI>$>rSc#s<xYI*68P|JE$^(Rjcfx>eE5j%HO>uuWRVCuL2ij>n5Ps
z{Q$*}7}tQ(rqZ-crKwE^oi>%fHkH3NmA^KXzc!V>HUX4P0A<?%>JC+A_H)pU1Imm^
zl=g>MLltYt9<YDi={GYPIIkaFH$^dx62!oHN$)<q<Hh^4cX7V>WjH>$7;;M4w|J!P
zqhj+NStMreMX{3+i7k1>f<Cqs!?{|k8ULY*P0}jWeUxkpC}3S10Y~|rhTDv&MO(>|
zz=W{wjr+Fr18q<^wPLcaY)~^e=5hUi6$KCj5p3hReMNuCs{Jw$(lKp0NkOq_Ks#W$
zqGZ~>>2p`UfB;}EeGaN$_;FxO)Quu3k5fMjwde<1lM#M|D*09;K1D!KLPF;pGAhOd
zkn{=`9)?X|Q1VC%1Ef168%#C;D7od^Y>+CN1$BaDt$rITm3vV*j3K6kWQuGK+J^31
z>=&$RiFSV<4)^;2=K0+3NPWD1+0<AUCJf)`E~>tbUQ*df)i(oaCsi&!89S+xmZ<sD
z@3`My2>GQyl#Z(h^#@u#V9jS*J@_p$BfA%)$v2Dcn(tS1`+c&B(bROsX#7IiN}@DL
zG)Z}qc#>q2bdt&>)k&KAps9~M^^vDO@*b{-WmJDi8F*iWfGh=2%Q1uI7NtmAYBSP?
zbd6YNNts$<GTQ^ja9v$eL&PJDeVfoP1s1o)wisd6FP8%98vwUS<vkcd!m;sqWv!M9
zfRpIA{uI=SwSBBi-3!hF>lOH_zLY)RQ0mOhZ?=f&O|bWM$1TBU)G|WPXb(xx_k)-w
zBMN_>{ZjL7Mi8J`8EL9YB0rzN2=<Wwk!t+VOUYnG5KIhrs&0VJyk+N_aw9I9Kt!qt
zKw}S{5I>nIS2ifnv)?Da#Re~!KvPTwaSWuhQuaZ{3V{%E?;UN2Tx$eE^^LHyg@bTo
zAQWJunStqRl0EF<d#5c=Hn3r>qIA2>22xDlHzmJ=T~OkI*lXU$i3+H3q2PP@DM|2a
zt=VsNC(w4giBl|Y1tzbEB~VlF(|sUBh9_WzA0RjtLN2fU+5ArJM-gIVr2hvA+JHir
zkw7{W0yj06VM7WOml6Lv$$#PSj}L#pkww_IA8vZ#Yy=^_A3lD)`C~xB=OIzwNTFk}
z!~cfVEI?Tk2PqtYclaJP{tbU3;hX<_{`9wx|3ogs=l>F^9E?yFVi>;W#}I_`5Y6x<
zKhuZ`MIJyzl-dkYiKui`wPS7gINTWJ1!x;atzW1$3KNV{qs|(H4u75qd4SebWgw2V
zmB;%sZe*$25NS?T40<9ir7t~Vf9^$(^cp`iDfar;aPw)nc~`V*o$^EyWM&*4{w--0
zgaMBxG}0_Lc_p;7*05Hhm)5Sdj^|q&m2klsjWZ0wOTbTcv^r;Q8O@uUkQZ~{CVubR
zSM<=$6<ey2T7mXC_<x9+(qy7gw3{{DK7agL`hCV;&!hLu=N6O}#=7Cwd9-0etkTA{
zVR7RY>O#!>7Wb3|DNV$8mG%{nv$VgaJaE8w!$_VyvD-GYcbl~VIAOg-8!Xj=79<FO
zs)ZH!;u)BL+nu#Q=wlImEFzNihuatxwi?7Gi!XCL+!9_H#D5-hkFP+KmC6v+K_!YX
zZcX$))0@OqGuLB%;_qYOM4z8nibRoCC?g)4S_rgt)Be&nfz?9m`)5=@fF%WJgh=mv
zrvO!_+E0@JyO>82ExpuOJ>x!yqz6^>sc9x<4zj#I7S?ux;%f_LjKN0J$xBW|8AE=B
zyiOX6MUIt!T7Nb9gzw;zSh0}WV(6VsXou)wc?`=-gE}w|Wt$w!Gc7MMM#pzIUjMR+
zt_%eEbCLnU`;NcLq9v7t%EhcMLL=#njmL*Fo}W<m1~?w)m>&H}6D)d0CwZ4WdQ!u9
zu2}?|rGX^ML*XLx@+WI)ld#pIiz1)OfJunmth_F%vVZ6`w_3UCmW?!!=uHzkYb6CH
zDTYy{o6lWS?BjAAvpn1;t5=Aj#zH4Qn$cbgQW3Y6f>%?}p$2%pf}pXOu;Xou_M_Pf
ziw$S7GO}E*ODls281+r3nrsx(zS!O9Knv;Yt5}`=jufjgx-_<*Zi?@d=OQT?B}|I=
zZNZS%Vt<PCgarf<g^pTc--|u;sP_rPEEEm$nVYOF6-Eo~Bo;~(j^+V!*<Iv#gS~|;
zHpT9)s2^Z8S()-#38q?Ll!K{%4^y&|@!XD5o>M^}CrHjFugWG`odsrF$*UT}pOVH|
zGxrSWm2Ms5&Sj7mJv!zl<4Ba&Tmv~J%PI_<KYy~Nlvrq$yTR}tv!sr@yg$<`3Kd-6
zi~MPk=e|KkZVh+5_v%xq_PHBOOSBP~$+cg&#<5z$fE~g$L>OVEo~i%o#KJmOZB^ub
zLqMx5-IV5hc**^a<9^Ao;iO?hH~}qXKhKO=Q*PXjLso;*i7prpOM*eLH8Qs&FJP4^
zc7M?2SYqFSYt!s`&$4XHH2?G3pwF_&H=Zf5Ts3`)3zFtdxss|l)G2~gr`swyG@V|M
zaI7X8+6s+u+>Ms>77Az0`Y+mYZH_hY$y+}<+9>q)+&!TO_+w^1$|^x;QHb93MLxGt
zgL0VkpcnY__u<pcyPj>?is&FxhD`sg%zqd&X?b4`^rFOhOXsYeKp%P-as`&&>xMOt
zBLq2SlTpw?Kw7prLxcdQZO`z!RYSO>+3>$K$w|xmb|j!Ao%y&<03|2VS7Pu&K`(7R
zjyzsIdOCv&9m6gRsKnX`rBy$AX!_s<sLwgPxmYTX2By?Tsk{>&V~M#=QeejUB7dcB
zZ;K=@vcV~vC7HA&7q@JGS}LK*V@y{omH|`?t02<(?<O16S_v6?qozz;BJN0DRdl}X
zUy^##GSVOK5)*NC70m4<uUd-H7xi>%N+j82T3fdCx&3T>KytI=hVD4N!p?=D1^C#H
zw6b*pLtaM*-c#yp_O3K3?d~|PZGV%3GrhFyNy)|X=pan+TjeoIcr%lJTYh=DY{6p4
zY`4l5qL>>cWr+2VsDPXrrM&DYY~&-ieIN&;eUNIVs4FqY(>a6gxiaXy+yOSgs_^gv
z;BqLku}i>327CGO*Eo32CkW-FZ_WPohKUkRX3-MMYowdoRnBovnAYmfU4KRI#4-bT
zJ)Ud%J;rtzV`7evzR10T%YeI?=Xmpk0AMiy@it7o0-(jhM3`Ux`fCn|jalx3KGgbY
zm%sdasUfa<6*f#`F=6-BtK5&`oreS6>X3>xeNq?XG==&M7GJZWFmwYKPRl*t^ema3
zEjfSlJ;xMi63XrwQcM`yhkwf>27+?aOSTp;ST7u|=9Kt~cmdU%j%TI}S6m=?L)Z1a
z636#6s^L+!T2pfIg0Avm$z{|d+$FE^662O2R9p;mlR2Idvfe8DcJ@qy<Fe0v+4W|r
zeH7a39qqE?c#VlzO+_a?$2A)%7TV9KGZM4W7tW34Q12akV|ilCN`H&BfJT=`h*1k4
z&Bm##h+&h7J@vV8P+pE1T05cjd`WaRZuB)v@^R4E^{RJr7mQDf471_Q>ERJ-w4W(c
z9(RcjwZlXMCMi_C%y^iFZc7WoSI5TjA?Y|?%QknyOqbw-Bf(`cYIL|ol5J>pfV&Q`
zmcuz(J-W8pQ&mlX^?$7AoU?D4#sbQo3k%1S+87p2y7Xb;4w$Yj+UlO?ksZf9Fn!AL
z9Hw_O;!xDDIrUQ@eGM#^GskVx^ASWxtvItP+ght^+FP?UswPQSLHfYOBt%Z-6N$WD
zrA?jL817yIbIJ(o;{<YLi>T9Ray)8PcWkvleBoa6Ro`w1Du0)^<0+QNtu0>n4R2m}
zyaLEEi<0O)K?~OkMjrM7D4i4Ci$lJNx$c9KTOjgKYX}xl@)hDwjFg2Rm6lwORUOA2
zu&U#D2CMoRby#)9We%6*T_0WAD7=juC#IYc7+Uqz`@vYPef$P6rh*jSw8+u^bTFnI
zE^aVCCyw1W<A12U3OyHDmRg5oPa7h?t1^J1aT`Qv#}Yr6#-G|&Oxts!<(GSorQ7&(
zl~+nB5zY3B%cmmCW8P~oKV08blcl7k-)qcjvb^a_cf7r%CM!_oWykznNtTuAF0kvB
z>Q%BheMwdlsaCP%r`IUSazaKkQ`L7{L9dM#5MZW$Du3GvwYVotlsH>y_VipzUY0<X
z!lhD_q)I(y`F09sY%`|vTr0pzy;qb|`qqreF37x8paqcv0GJ%tPwM$rN~o5kX!l;r
zrBgg--F)My0By9Rxxm97OH5EpsWz3#$<Z7_3xqSrTAH*H%GeqQj7sBn`<02K-vEVP
z>-(YIUVrt9CVHWIF;{B*z+=R&N7uy%ulU&AHsy}$`_9xc`js1^%@vO67Z)6EU0QR7
z5i#b&eaRV6IgW|nq~r|fW)b#0x9>TA`=YZ>f-_ZT9rX@nXC22gb!YwfuUB}sQLa*X
zwxE|+EIsSjY@gb*j_VG^XJD+5*8Bj)XY|_!(0_i<ojv9Fip6K$NN0-AI_e#Y&pM80
ziqCp!_b)!%C{HOq>*d)ke@Rz;aJNq()k`)XC^V07vViKvlU=;OZ^ckA+2i#(X?jI;
z^;IEIX#m8n^*X2P8NNZCPDC%@ez@u)y5*IyS;-}tG>_Bsz0@auBSpJ_fus6?>Y)}R
z+X?O+Bj^7E=?&qRq0urE12Hi-mjRAA6qik;EGn0x3Ot*C$~`=te|i4rozB8)6>R((
zU;VF7&v*a1>qWPYl?ZtH^~+uSg9#LQJQ)?7KYjh}^!e^zeW%Mq?YEX?Ru+7*R<sU_
zwOBMXP%LEPFB?mfehQ5jxwK+=JQ+HDjbqPBfBI3z_pmrVTSX#TH-5CZk%z^i9A$aP
zvcx$8`Odk2JU%ijyjZ`Pg*6KxkqQuc35@h&Flz2TOvfv>h$9)urSW?hM~`+IT3l-F
zEqy*_Yu@N3CWx0H^j<<<bmVJpqNPdtFgndzQ}R=GnxMsG+FE0pNdrBMVG$LADJJ8O
zS@la+R7`--O~T*?^B!4Cu$sQd8gx0J6i4~W#~^lpQzBy59t|l9zV8uQdB)mKS@G1a
z_PQJ1E{}JLiTBDA1ibuA`h9LMMth9TD-R_{9voEX_s6V*IY*4s;-gmM>x_-TE6ri2
za->m_eoJE$q0)Y@v56uE&5RvajKPSI;&ea+lP4iZzj3$A<|IHp?Ul`w0$B?Bz_4j@
zOQ=PE*|5q~COD%Kr+Ulw9h|^?%4*n9KmbirqbV7Hd>IqidWATPU<_&SXvio?6b?dM
zCU@ZiN5|5(?^U$dy140eD7%VdTC@xPNE~Z>JOans9<Sk8Tl+0=Y@a*@$2trD-EnM}
zauvtk%-9vj_Q_LltaH9>V9o|(J27UMu(?}*Fif}-?sdwFc%7X>W|Yv$lreWY$*VR!
z0Jw}%Y;j4t*8rC{3VEh;16)hY>>5uuRxGcv&MjnnJyecfnS14~yVkM6KT97E+0vJu
z!=_%h812ZA>pVs)K23Quhgt`6E1!GG@0W673Z&S~c1~b|crbnwQfwF%gmB<u()DP6
z9&ZP24hncC8|uwPq^td!Ma)SF*uKZLMI)k%gPA3}MWb+=>V=$r^s=>h42wm(eoR7A
zv>rbOm70>NN@cpm#gOMVO8Zp0A|p<Bkx`h(B)3d`73v1}yy}}eC_*~YH`^W$AUa)-
zSBOrpQBN#(V&B|X5--*ZzA_Mv;orx9!)OO77kH=Yg`{+~uhD<~_yYQo2t_X!CFcfO
zUhX5<2-tS5#8NkpA14`R!}Ip=f;GZVX2U-%MAYr!9ycUpn|LqmAg;PJ@u8&&;W8mM
zl|?<qeP;5x4L4na3-f79jG@cDlI%mPgrHNvN;6wcg|Ze4d*QfK(B@E6jRLHHxp_3l
zI(JJ9Y+zx_DEs||VMw$(<68s;0YPY#diUM?*i@SEauXr%!bO@U|9F2irn43P>fM2u
z-t~9@rguGF!Sr6E!u8$W4dE1|Ujxg;3JG^gkK~71U2QBGnjF@{ZY=E0E~QMoiN7hu
zT}~RuBaESXP06>5*%a<x0&|9cl|kP?uFmv4gMrc0QLDK_yU83>G#RCGYP=?{{G|LT
zm-}Y&x_?UWmhHidfZWLH6i&>fjIHD{+Uy8~qL8NPR?gS>>oG972O|@duE69Q5Ku;t
z5Qihds_pRzuxfj}23DO$9az;+hRKxSH_mFw?$?>HQ!%uxgj@BskHfKlG3)#da13H1
z2>*w1jDk!q@&9q`1{_;Xya{{r$^>U=De*>Dt8qbWqv3o{TryrmZi~0ie9wtC%77c0
zK4MVb^PwYd=N4tRB+O-g2jHp64cdh)GL~BcZn`zFxt2NKm!u2VP%pJ7`O2g26rd1q
z6K48cyHkMKzh@|xe~;~d#~slAYX$^Xgu-5^1w(M2;FK;M0`QE9`&AR?6{A5Q-J)lk
zj7?Bne!V_8ceyYIStCb|wevvr7-vQo<jsp0TI`-g_Bb=kb1-P&bNde8C*hI;^f1@P
zpmYemc2`|{=OR6Z!E!2M774e9fj6SSY<j9gt=tkSlWOCZIvkpRQfPEZk^LI$J&b%N
zM{?b^nM?%m#b@<fj!!-)u}^2Chf8&15>X6)Xg#%%z}*Ken6^!bLO&?fgBD>U*2Q8K
z;M(P*?me8TtdM*Z&=ib8g5mt2xckh=FST$z4QLM)Q#bmO=!Mm{Q*I6_knIJB%Z&4l
zv>Sz~hMB6z9VTOcKv{WaBCREv?LaGgao*@o42kc$uv0dQTX&+FY`fR*Iclv?!WA^i
zF#y3Rhu_30HzPmhshscS*~1Yg>(u<%IONsqGA2{Y!SF!)$0O&uHjb^L7A~Kl)Ez$0
z5BaATAp0FQ8RZ?Hz=3kEB=R4%=@uHnJP9_7aW$Fd2T~e;Zrb|Y=bW&0Zn23Gf1<9*
zzeR=IYGoi}6aloY{<}Sp1tqG_27)q>S#Bk4if7`X*%r@v1+t}STK(~UlR|k?9$jbn
z_O1Y)w+2NGiuysfjgPX#6VQ#uM*s!Frf)Dl3K91nZZtl|n8!{$+#ZNzF(ho8{JgD>
zxV-D9>K?*>yC1fN6F?NeZX3e3a7LO4OIWkQ3b8N9$8PYD>1{;JNfA@{Ro$cJJ1S{k
zxXtrQ|D=`1zZ0~!0u}B*kWHFohul6y(^hF`RuOIOs(ePYrnNAX?fsBJ<)Ckrw}Xe>
z;urEV#CtnRe;eZ3Jl5?mj3mtJ&N$ni=}`fW8w8;*yy&NSteT#U-sxuEIoV!rd#Gha
z<F${!-a%V=a@qy)!-7n6|157|`hNlzvt3%3kw7{W12;G^mr+tK6PE}RJi~u(_xbVH
zjV)mNagWfci9s44!~jP@cxQEF_4Xms?D|BF#86pVPObIPs3@C2wUGd@Qs>sN#>kB|
zf-5?8&36Eq&VE{e=Gi#!YA^fH<?v89Dn>;p1>sVJR}(&CYXo7{58d{7I}W4Kh|mLv
zQy)D>pDcGMXYYI(gq)pGpBjHJ$plV4GQ|cfkZD07J|h}=t!d>s8L>KCTgKh4<*r+&
zy6*ejI;CHy@8;TgXt{^yCG0Zu<hGdj@<<Pq3g}Qwq~FADu6_35V4E-3y;Avf<i|*Q
z7>mSmL5&3+#&NlQ$Z0|0edk^!1Gw)((=EG@F8LhiJbbLaeR$mbJ=uRhO>B?=&7a>k
z`G$}hA5z0?5=ry@OY^k(t$u5jg?3m%S{_^PRYyE#Z&V|^MI(GDA(Oyf8r5!(<ALfn
zKb8)ZM<|=apO)gInE|!g1w+BrU<q%eUuLPyNP^`2WO=ijpO<<*rhqh~Fu_Fgc%m~2
z#i_=#it}e<b9r+~S7Lv|QDMSlA^8vyoq<)pFXYiYIgeJ<p7*G%tl?@*TugCh6jU2(
z5bYk=lG8mPTgC!HeSf<41rokWVaTO0gKlke+2`H1EmXIM{RP1CwCaxgtm!Q=*WI&S
z32)!j7vuXedyeEne-<mUC`mec7JF`xG@iZ8oy9}zH%~WRFhPHtITIx#t}^)pODI-H
z9H|;dG`9M8*^wuMrB57D??~sZg7Gqe$FdABL_;|EM^^vbEH)(3qgUW6@T0>FgjjOQ
z)|r^%Il*U}Oq6(L@R=N%20kbFOgIh6pBa2cX|h_u<7G$SulNNz%{y|*IHB4KS+X1m
z#bK${Hov>slJ|f6-97>nw4yjpvp1iM30AyR`Vjs6veIwOcWhfHZIPi;Qo^99Emu}a
zYJ!cgTUjd*t@BtKN!Qo9zM`J2WfI7^lrGd2W9yZ=qB6RYuBg0Bbj8fxwMUMWy_?q+
zrwAwl)~Eg1pZi(PWNTlmgFY!_8iolC_Z`$pDx!rZeQkd#jzP2?&G6`zf#Vfav;tR`
zGI40UmWU&pYw>X8<R_B1F5+3CXUJRBSHl#4W_f#C?(NxlP*Xa{k}7*HlD0YGQpMz2
z($-}Wi|+czNLz<AXSY9Ar$N0qqHPOb$=Y_jy-eN;`ehC^;3avhVkk{k_=4)T=+zq1
zF?Aaa8oPf9o-3)_POq+m=1`^4+E1YUXm!iUfvM>VWL!M)kYO>_&R<^D&MDgXOhtR*
z*kvD0@N}V?9a^tcvm=w$YIbC`PR-Vi)~VUbiglHm9a()np=O~)IaGl^qGsnMnWeV3
zsG!;t(iDQ0h|k%wOndA3%q_$D*ZyXB#+EUQnJ$0mlmvEUI84y7+-y076#<cCv?k=&
zRYH?RjGzI$meS1XUip=Z+{pJc^(9+IKSURH^<!(qVNF|K8C|Kbue?ke`x9IC0$P(o
zFv+^>YfVxlPk0?Nj(tB<Vdl|$zL1P#>y;X_GP)8OD=#nfW5-LVOd@J^(d8;L;V^gq
z(|;DBE~LscH0CfOYZYdt`k*TR1Byhw@RyN5Iurpnmyz2v6PJJ(JjDT?w?-K}I{|;%
zRlVQe3wIGNVzHGwtW!4l%KE#{?|Kdq9!n+;yVEo#zX5)+x-gM*$`I97|Ey0Wq_Fha
zL^Mp~maV)J0FOX$zdM7+b{(D%ip11bf4^IcB?_VpMyrXRY~<v+^BPpvMuiw!sCu1A
zPUoGKJuYMN%0=1fGG;j1?{XPqW0CB285{6_pl9&dy}M!WH!$K@=@#hVJRE#0X~@Cv
ziJ9U&Gi2zJVp96c``qT;OJ$$?NupU#bI4^1srGHsTr+u|H1ugMU)&Ger?)2)j5COL
zcR9gaAfhCg+8PU9T>!zOaA;%HZ>@6ZunU3HS%@nbXZka8*7|7$nlLu7X=fP=z*OXa
z(X!8S{<kF}O5rmUIwK)ZL>?h%?RWqie<n^rmWOqdV#TNM^Rw_9%b2G^&_>ioC@0?}
z1e?ioIWav0rxQILf>vTq6xW4dZH*#zO9(bb2ZdlW%Pa&}e6ukewZhgn0ho-FAh=7^
z+#_CmGAS^}NJ>OTU*(f|vRrgv(Q1}|og#0h>m1i}*RMHeCf~1pI7f%Irjlu>N>f|r
zri#rhMIP(?(q&WK-S;&#=}V)!Nka}hrRp-|QMqcD7>{Zkp;7a4doM$2nbLx@iR=%h
zJ`Inh{a(3gf`_m)uOv53%$jU)^^DhRNj(qA<3$xGsK9b+%&USkC~Swu;6(+0XD=Ze
zL>KgCWMsG&%u#By7m%BP850#F#2RF;EjA6jg;5WS#r2brkx{j8LA<=^Jm2K^Ff-xl
zj}@K<)w?HurSP0uZxo)D$!6hMS#1-Z?L^yzXJy61M0QzqB7ZumO+F1K5QZvx6YA_5
zEGmq&97cZN-sJmrp&AW}3~C&IwM(cHN|`ug{z;)4vQXXOVrHA=*gh9CBN_WW9%h4l
zBzC8Vk+*V{w!_0n849KFvWG>MJj~83+c<c_^}4DsIwD_(!}Ki2^iDwDHj_3Wo~dhD
z(mM-(rKqT_H|pBP=pa$i%+iXAOTMd)fLIqW8?`N4#>05LwoUmW@>_?0V-4b&8W)h1
zP`EZ6E9;FKw=p^@9P3$D!*Sx9jAlI?UqaiG7lq{c3~ieVjTq#=qHQz9{g*qrutO$0
z=o#9!=H)1DODHxxs_WO$wgkNnQWo>hBzE_+dZf5*KyOgDjl}`-w#BrnZ(BIW3fyhb
z28G*Vdu@peK7$E|1>~lGgRsW%zY(|sO~Uco537&HTfeJLEVh4EofvKWs#;I8@uTX*
zXq&X1TCD4yLD`B?+RQ(OUe>l2O_h)KN!vt6Xz19WY<;u1r-iN8vd_6m*SfIBxk=Zu
z-|5^WYa_`H=O$Ha*eyDD`L#t7wT^4&_HQlP(-)`vzjz+G%?f3gWE3b612Hi;mw|#W
z6qleLJi~u4eg^jN`SI5qos4~bvSHFvBxPM}oHvT?=`9tlmNi)($+Bj{-%VgGMy#~Z
zy*0jlDtca@0xXViodRu#@zV(HNR&Ri&Li@<cn_k5uVd$h)}Du?{df>34Mt3d15qod
zeQks&Y$r&W5u(6%@8DhG!eL5eRq&?az$zbI8ts2x*AAa?h_T89UrB2`F3tcfDi5`g
zc)>!aR-p|68?6jqfh;vY*^u$mj|-+bROG1EGN4Y71@p0Mt?}s|yO>$G4~5Jd6T+a`
z9{>e##-y+x)39Ek=toc-6t2P1hN!Ilm*H3$Asn;zfZYgb(!;b?;b@&zX#9n6ETuKZ
zEpUIVgsud~O3NY~v)Byon}?%TOn(Y&3t1N&-(A$$SvvXVW9IfANk*$kp~xxY#5ozb
zv`AQi1K50igYEa-Tw6@~ZqDsIdsv8u*!{lK)cK^uWrxXQh9v0LqQPx0B|V#l%G9(j
zPVO`P@30-R*I?guug7dyr2<({#?(GIo@9SP1D{fWHKOdHb3-@91Ts!i5-YxasJ^{_
zy!mSsD-($0X_`NN+{g<9RyiM;k1B#`e*W2fz4@&kwTDV|Zx{a-x7s=3aGqlv(>k!W
zm@ocgCvNp1KYQQhB7JJ9DbAOVvP!p)_-9ThpBUYo|Flppo@5Y4l8KKDh0|=VLC${`
zj%Qs-%$n`?+vGVPIULyp)AsntdB~6K*@W%gwh;pb9c;9X1O^Q0+BUJlPS};cFOX;&
znQKd(vBVyAPlw_1iV&TO{HW5_j;a%LPYD$|-J#!V^}F9b<+VG_mj065+U5P!U9Gjp
za};^O?Pc?oGxqa3w4<~ZgfvRNEM$KeSy^2TMg=y^uC5H($Z<hurdHQn`qj7F&XL|v
zidbHIIKYL-?>vPhHKWKHWhy;?FY9qyfiojL(Ce|&QceXuvJUOCb>1r`Pbu}~b`ej(
zDjjOS7b_F5LQ4}wjmlLzKP{Sx+A7)?Ed)*y(eI2Ff>Ff#ryb#vJ)`EB><fPvat;FC
zWmQso#XMpuaMR(MG0y<IuJZQhW!nsA45(a(y%sYm!sj<mtEI@tUL-V<f+!S=F*?qI
zD6aO($!H?-%WX;C83^{&2T}<BQtD^YTdALmY^HuLw2}Iio{iKmrO<4nelE3<`i0*w
zp?<D&BlWjOYoPwFXbsfg9<G0Z`n$q4P`{MgO8u#Dtx!J~*+~6dWFz%+k)Haz-&mqP
zfbvL{2ez`GqF3nMh3wCzH?qGF*~<PxXdC-$J#c%6-oJYLc3Gd0lkr<fukbk{pY^Ho
z<Ssrl)rhyUh0+y5Gfsfqo&X>O#mL+l0OUgf;GO^=FGT^S0<g##A|-z%h)QaW&xYC>
z!(_PT_{@tdRS{m{GlS5%%5&J^>FcsG7Zq>6OG&+~ELEx}BRhk@bQB`b2VQrs)g1fd
zMt6L`H|ynWCMlnotAWU`RLeY{B&n8a2{05o%Ck@C>w$9i2|)LKf#)GVJLTu5bA4oZ
zDzb5LI%tK>W@H&VhEso?wtVS_!=2M|vH~riS|^$1`lb**LOp6aC$H6jNo*o}wN3<)
zSHMJine%X^?TIZ10#rVl3v0Zk^v0G$C3K}4Z>6QK@k)WK4M1HMw_RPVC5}mONVT%Q
zAXXLb*4BIEzy(Vey110y=;BJ~N-nOnj9h#=*N6Mo$8)+hXx@MF!KH33q_?`Y6uOpM
zOD!X}ic3u@Kbfus)pwO<T)LJZWh?A7UB5|&6qNHXlqgGOeJ>0z>xH$aEQgn(t&rZ*
z3kzs#sSIvQ5VCP3Np4x6>rT{_ls<8^2`9ESuwh#hrJ&+=T7uf~3D``|+zr-X-<Fdn
zRPIZjgNSID?O}gp`tx-OXM0zCUNB%L0`Xi^BO#*QHMPC{_?f8*9>O)JW;BnCt5bV-
zUgx`MZ8G2R5ytotoQL#B=S{TnRWl(2?Eu((Yu>#tH{4>=D2Zs~P5+S466ohQ`wv+F
zL$SGceLBw?pZojPYsnsbP&z)Pk1fX>@Bh<%Y?p>NeJ_9Lx%=3q53be62IM%yOZBnk
z?jz&}veJpq(#Ib9*O&A-BjX(X(=FVzQz@1GZ#taLDb{mgoih_&srY~}g-VGHRZh#l
zhC*nq$_a|6U8`AZaWD-cRYEHzHtBFk2`d-+vaF_UaKu7lx%v&!@J=ff5-WW#4V?Np
zs)jeGMcaRCb~gvSYU>II(^cBJN@!4ICZQnzR5)QHb+!E5#_!AJLgJKE=~%DO8u>?#
zume2OpOB)>F7Tk7I!$fcN@rYp6gY#OX`i{EtVm4wY$mRrH$fEZxSkV8pNRTwd0LJ#
zZw*P8RQ!HBRTAK(>mNgMK}HH?Ze(+Ga%Ev{3T2m_KspElI5U?4{WcSqawa^-0n)cR
zCp;(z12mLOmtmPb6ag`pa8D-|myj+z#DA6T@2|gXbT;<+Kz`Cvu!>x09HSC@I!mKg
z5=|s4k!VKwj$X82w31A^i`>peJ<kVkX4faDjHZ755GB*P#90?5y(nHGW}Tms=dBGd
zS>tY?bsB%rP9eBjE4l2E%_D1VEr`f}EHLYRu%tX=BRi=$xV+k4Q-$ye1dtSZu0*6Y
z>}EYM1d2^96d@op54ZzmovkH>BD3loD;;VfeTcl|h+M>cpZ1Js3z3=Uc0PFJf?er1
zXi~wrNylApP39kEjw`npFg#BL1VmgdYL}5fIursoG?&q~G$jH#IhXqr7%_j%T5E4y
zw-NpBUqPT>(gNlsXNHd{f&h8cX#fW`>=bBg2!Skb93_g@kk&@pU*9w2F15<8n$>Eh
zN|3a?clj7{X3m@;8Doj+f{U>b{xi2ybd+-|kz*w*ad3&+JB}{p)Jo)XsI9~<r`Ss3
zvPx@PT;cKrB~e#P>p6+j+Q)yu|BzZYWYufyhwy@_i-W-=wXp?nQ*E6J-j>+pkm6!R
z*VZkDM!z-|T2P3ky=4@0>rASn7}~%Pov6!SFi$S6m{38M&D2rq{h~4D+GNP6Tl>Jd
zI8sp?Eu)!s(X%=Z+}EI>Sem>gIF7yFEGTwyD&BH#b!&Dkr4=(4JEeavBs<Qv3yGp~
zw9(L|gr(@oNz~SF$xiUoip5DP{cUyVwhuv0v}*@zFWURXdZs{4bi7ER%TUy%7{x|Y
zyrp6@t^vyD+9qF%OCr5lbt(1Lsc2SW>sL`%yJ9t(eYCnTqD)<}kUC4ByjGY^2yHY)
zQQG8TJ~_3~YD6;|(u#j*R{M+<)~2dWvgisdrWMJRuGk7qmlV3_;;@>%ZgN;5vyzJM
zAZV{p9O~CrtPaD3R)ooNGng91)g-YM6N=4X%oJ*ltr=Fi<~E@UNowsCU<4IQZKK8D
zG=@Hl>Xf7n0l!mjZc(u?nnG&^-_y*kE}TzG7YE*=8rn3?TS9;AU6{KBTK0k21ek{n
z2-kyi^+}nEL+;(MwbeclTU%{*Q`y>TU*~4Rfm{8iXKHV1b4Bj!K(?OJz@oOADa(^j
zKb^k&_xsuO`T6;LF}?cmW?@gyPtSjL(--rLx3i1a0Z0>$roT?#Okci^{o&>8WbvA)
z_#9!JF?$yCB=vs^KNCt`B_2(mJ9PhQ`g;D|eEQ0L!juRLjYSKg!k8r2KYoUvy9rT(
z&#)~C;%%Zr>=lS55uY_}2qA>^VAiAofADyvy9*ITJYnW=f1j|<q~yr}rBokjLxxMV
zIh~*X$-}<N-A9)gJ=6;TOx|2Or2uL0oRb%pjp1X8q&a_U003o?B1CCO@zraC@rv$W
zPXGAlzsw9h9K~6DIsb5Wc63v<f@%rg4&6f?J5)`HyZ|wFs0!@DXW6AHOY#I4xT@-=
zFXw0Ti&yWDPiBs-fty|&FBY?lb4UI0FTXCneziDW%v|g*p3Tn}801;=J%Fx_@+>vS
z(*984QzU=f9{`^*g(l2rpG_|=<|nUai`UbOubxfc&3;`Rttt4$@w*uWI$zAr7ndz^
zh1SQ*+2#Dh#mVe)8Tstpx3jmW$6w5UZK}_JRT%3LdbmJaRLi9g-)!b`8L68HPhR(6
zZFA|ddR0oJlvgDg@8s%TvC`3<%#vX_7OfFCe6W9T2F)O7Wmu`+)++b+OoEYR3uB()
zkW_=WK$tlK9X35PF%vOFKs*3As1`W@L@9ei{gn`+>dmNb5~3=QQ8V0^BH9qw!>vOM
zo}mj&6YF5>5<Ftjh6srFa440OB7P~sA`h?*0fPdb%pSmxBzmj_U;t*ZWfktn9w;P_
zpb3AD$sTbKQiunE0?{O|<$;7~VZx}JM$cphfiR&NLInu1WhqF(r@&cDq^ucmT+4O9
z79Bu{ez#%uT8#m~1)tzdJ5(hWZ#?e*g{spt<1o@e3lCxsWiPzVc@uB5?1#4{uHkK}
zEA&jdD&;Ur@$QI<)CvFs^n&Y{&6*QOqu_rL=A@f^OW?pD_(2jIoz|2O;6NZ72@ZsW
z&l^wYSU@`P%!aI+LPS#`;tBGw1m{bvhd>g7!vI4F77R}AkmkUvE!hKEp}fV7kr2St
z6HB020lcaQz|;))&Bof<TPg(F;d_?F0JtrnT9KA2=qDT)+&iuF-{yEZ?L@n-Ra<|u
zo?;<JcdD9Dm3ONOLZ3k%cBo3mbj!T}#O*QRH=qJ49F;oFdS;2X#??2;zKMI`NR68~
zQglBYsd)`YG7Zca46a<22A3;?x0S)$N*Sek0K8cmcrL&j+e_h|=HwcE*plEC+W|nL
zY?(LM1E67xlWiXsYISjqt0c@g!=Hb^f1q&0W~z-jBC$Psa%~#Z=cD_URY9J=zx>Aj
zKKXI6cz^llX?l0M`0>LVe=`4N`su?j@1MS%eVo2MUi|pC_ls|!(2E~nqW-6|>CM^v
z&Ga#PKwF$HFHWW}E>7RRn_ZgTFBccbr{|05&GOaR;@#r=#r*qI>)_+@8QOpU;klHj
z&$AJPCkKKS$%-k{Sb2e)ao${c+fYE@=8#Gav2doshXXi(W>d(J0x<wOzHj|=QOjS4
z{@HR%PLEjsnx8(8Q4VMT2vj!b5nBk54E^3Jw%lC^W@HU%=pQ(1abaI8+}#2q(iO7Z
z<D~@vJ3Y9><FJ4q&Q9Nde|dlU?!2kKn(D)v>RY6&;rDp?T8*)XBE=JxU8>p!Q`i#B
zw^T(yh114+kI*|s+iMZ})}dFoP<Y*s!rLzSRp_lFeAYqc>mc(w$h^i;qQiw*uu~rp
zw^weK1a240LxX)TeGhYqI8xwqx$KEJyZbsaY;RlsOIMeK{6gbnboGD83SUp<fG|&4
zyuh7^t5=vZ6+ztJ%%;S)^-2j~V<{-yIpzbxycTcgVnc{vsokdHAt=%*_)-bU&KsiM
z4dhFL60T?-5;|W3KT6G4XdcV*8<f}G@-|IdHnuTmyx*Z}v=lFtM~L63X2ufJ@}cdn
zEfX;;1yS59!>W-tV7`C$u5J(Wv8=(ouCggRiCmRvl&-#&QA(qfN2!cb9VI<r)<vLt
zh7C)1%=j#O`{TYbDZ@$g6)>ej@SD0PFzpTsiPg7L%_S(<rZF=;Mo=IjfI<!j3<nI}
zn_#l57|c<j`t-1@pH!fVJm?4q08o?%rX#koWTO+j0Y>4dB1nJ5E$6yL5=j&QK`Jq4
z35VG1w2jt^#1{!~F52`U(Gbn^mi*^lqBEOEBpEVDfQ8l@Ww;kK5U3u5UCAI8lWty!
zlG;rOyJCa{K0=3dM~aYvK5>mAuorzKXz&q|&M!{?Gizv2o<3*3&JXND%>T{;*K3T^
z86d83-leKBsHA`SVNYjv<Ow@b13cKZuT?H~OP;`6cmnUq6L^g$*cG5<T#qBd$5lF5
zX?#nAMbKao<Mh3b)Au?~-|IMiuQXT$9dO4sRRC4H6UQFb;gW(&1bNte0-o!$oyx3)
zfY;KNlivF!FbVO)?nc^4x80dTdMkNP#M&c#<Z9=;7Dj&@69`-emkH6!mgMeUqB(Q)
z6{xe6!LCW!vE@xN+)psK8(}1A2%&&)(3_it*e#adVrwB4Pw8qKQ8I&0J(6w2%ah}?
z<7?p~9)L+2T?#^tK4rU@ipb4&#h_NtTdt%@<q#r_E{9?_VDu2jw49B`UI_t}`eetp
zHwkec9BzLQ$p~Z6D;`O9MXx@P3?kkxCoAS1+Er#>#R6l7ALKoiaU1$1KB}(Kr{Eup
z^`9M|{Jh*-skXOLf5V-acG+9GUTwLz(k`0uE>#f)0h0GZp!T6jH!-7|>Da9~)AEkb
zHO|y8{rw(ivNMG1oGH;7XQIK6c<>`0?1{&>GQL5T!JcHWCmHNX<nF9!1%|exQW1~s
zt_p{~qq2zru|!ZJ&RjU!m9LNva7VYDWkT625&AXda)Ssrv624*ie9v2mytj^6qm8J
zGZd9qA1RmGKRla%AIJ93=U+EE8~d<lKWQl%#Wr--krI1~rN(Umqd_F5vZ$t=44Ial
zs3Ip%7-Wbs<nj~SD}2~{m{r79A=rbWkfh1RxzmhMAIg39NlSS~Px(3I&mWU)gob$=
z(TcZwgL-qbLDN1$L(2J)LA^5GB7Vq*QC8kBQE>fO;d)qqo?+!3b=IIl<UHG>=IU*&
zkiExFE0W<2o9>4^8Q=Z#`81Hkio9zy9b+T<A7+4+^-##;DTC>UfF~d)Fy~JxWUX#O
z>jc1#l?84MO-g5D1wAW_vC=FtMVMA%r||-nVe742ZP1v2oP6?S;HDCV+A9fwGIMln
zze=B@-5$b!XXNO-wZ&r!jx$8~525myT)`%N#|-l!j5xzjMN@KFisKT*$ql<f<Uz_H
z2X_?Da-M3SrjLN^^_W3=HI$~@A9uB(wB}xXFjet<6sxUTw6%;6;y-LK+U;YS0+-VZ
zQw07g!IU7#?hQ6#9xtDpZ=aqw|BQUe3@RG%*Y{0-d_W*!=R$|50#Ntk@9t&uTRUsn
zkIl7D>3vLUYnki(Wfe;fWp;|l9}{H@@@o;E$X~}Tr(>qyrE64GwxA;29sl%Opt>^%
zeaSImIJXOc{mGn$4ixIqlKTw86Y-b7pG)1x819$M;kYZ}l+~+~v6o<(gk#Fd=PDAT
z6zM>JH}UAAiqsg?1ETXj*=GQBzUKYTwsa6B0g0}>*<z~c3-vyY9^Bm^`|l32Qum<L
zy-;e#wx6Z9zz?8kyn3cCQgF&?bD}Oj>?K<SaT{8is&zx@U%tIdAUZcG!cqmq|CPue
zVt$r7p@>OQqvzjwJyt8QG|~e_4^K;3<#l6!9nPEPVt>N5$nWpw0tuvkn$Ut?6$r-d
zoKn=1Jx#6iWfl)f6w3RRup^*y%xiE*@O?5`hr4tUN{2PD$IVC~tb#o!6mmvM=YMD}
zc4bEnJIncIS%MgRJCSVES@khZi>0m9Q%Mi5^z%?QJk6%^Lt2>3goin1)BR9W|Aw1?
z@BD()avZNA%qPy^l_7g><+(;&-hvyD*g$Ocjau+mSCGDP%-7LHEjSS>&D0tSrTv~i
z6B0}A4#)0$vPYKFpdXzE7o7yjc%|L_l-=5%eoOe=#GB*!DOKC^^-e-WJAbm?&7bXR
z>AuxTecJi%>lA(|no`GgI-pKsBy|gaEVPC#Im^i@&_fGIIm?|3K{>@L?QcL!7$-+f
zYwsr-8-p&SC1jH-baVq+qSkbWIkC_ZjiD=OiFS$Cqa_TILT@ghC5$HN9L)cLmN*bp
zDJcy&syC)3YK-2BmO$$iBHW0UK&kK)#<au<7V!RPiDm%{Xo*JRXQL(1$iahu%uPJH
zs3J86{S>qWf;7gAucRf0(hF&cN@OW5Q3)-hC0aepXo*T{87)zXETbiA&|g7IR5};Y
z66;_uw8RSB3oWq@_e@KyfW6QXYhcd_1*H5rdj(penP?#`(MnxTOH5|MH=-p_dbsh*
z&(RV=jy`a1<yWL7B&kV%+2~t;`k%o?6(pqMcquI*5ecJ+Tu4iR)c*lpe{EVK=3Ibq
zKXtb;=t5!wjO<}_uOudF>2<`!?0@<%B_<lfRuB`7<U(QswCFYV?THERmF(;LiN@NV
zOMfjvN{+m@__4%L(G@e|wMmKE&NV-ks26B!B=c3yss90m2znk0Wo~4bl|VWW0ysC9
z;cF)o0yQz0vF1N0m%Bwgn1B5K&(kkEnT3^t9Mg1CXfn?5r}0vCU-|9yn@_K-or6j3
zlon)sXDkZVp=d`eDY0~XoFtYO`8&EwiWVP3S~!X2grFZXHpKejQ__`m`mR_ng8Z&<
z@M(YI6i`bz#do|P?LI(=hfHGfmLe!*p=~C?L{`{@y}?9bZ!kk-gMaDBW0o{n1%~6|
zdOjksEPHCHibxTgIUmx_3uRP;(cu_u4K|>~+7SlepA$?8f&{qmCSV>uUYl<(r`<oh
zGh2cY#$fl?mtA;)K*CtxDJ48$_x11YZTD+?YH_XRGsZ>q9BB+u?Kwsexz2=h2eY4H
zd@aDqHWkLa{E}G65PvO)^ch$gnV~`!uKyDwqFVVh_^l!*5UG(L_(1r1($SR#p=7IY
zh<~IB=gZ%f&rt-!T2wlNFgS2jnEVhD;mB(1q~j$}Cc&7p@TEjR&aifH7ke(ojM6Bq
zI}<1}kU2<>8336-bFh;sIk;pW=9MEY<e4CFq&;Tg4<o@3J%26jRg~EcR6&h!N|eBj
zplGZ*<3gZGIEx(_C9KIRJxQuIKOQ+#c($727#>>)NuaZ1;-5Z$NDx{(NkLo?*!p9N
zKUnrmG{c5aV9n*<WiCbvFxYZ|!iA=Z3}?bxG_@?3H9`SZ6u*5~@*r>|v~SR^ssgF~
z5=$jH(sjl0ZGRC8P7lJlZFno-%|sgjR}eCn+XlE&B&6SfcZfgqehc0ug=i;;3r^`J
z4kZRtOsS8f*q=#8?6~0CgE%x7*^}iIug|ESOG1d@k1^He5XZ&j-6w*{+}EG6nBd2p
z;tsp0xbh$v#xouKa;o%BXf&5J&sBj7pHrb6^5Zn=RDUMVRu`H^4eT09wfy>)_Br=k
zSJ#YKE0!x~lQO4)K~zE#wu0uy>hZ|!%aR|f&_AnQz3HC#L1T21tW>OS7reWi&*^2{
zG5`2dTMHDU!sWJImjYNy?8PF7)(OMDJXhsrz{ypvrjGqh>dh{<Typ>>he~bMyid)7
z#1#t+w10vnD=Y`KWk8?>B;54Ycp-**jDr{QO+ob;g+^Uq$X963TruW=HSUNyH1za-
z+_4cT+N-Z0qrS_nb4RYT+Tx5m-Xbmbu3K?>hiNf5q~TVo5p<0!b4#UyuQV!}TwAse
zCR#7l&I$1P8wL^DSmDXP<3Yr^3HuN^zZPC|Cx2WXMtChehgXf$_ZmiM&4B~{PQwVT
z`K9lB{q>jcODLzRK^G<x{x*||JZNZ?SsG0^HmQX(^=~$t2=MI7W%!59CbR@r&)&_t
z7pjyO1f~`0|1qCfQ7;wWw>?}tq!?4LoKjRIYsVB7(YiTB%V*u7qM}+isi;WSjVfx;
zuYZ_TR6JJ=E4IO2rWG4-FXM`BxaWDr2H4BMVhij!u_$7#^w4Pz(UhZdh|`=tTpxvc
zRO6|NZ51!>tHuQDB2w`!mhREP)|^&%u(96w{EWKSAyKCqvfQ<f2lSqM#zvSN3lB!d
zpVVHKleFbUG}kJmF{e>XWuX+%B7LAyet(~urFr8Xj=^=9rwl(B>gE~~nuh0P!*kFX
z4%IstzT$L4oNj$}X#^c5$Xmb>t~owr(<|Xvb2U!=)Jixm!BRX0JXb5F?%*jkr*mj>
zzg6@qv!mOji1`@}2-GbMH}?+i&mE|1A^0A_vOGfKWno6Y>D_ve8G120WY_o6ms>YN
zed90__!>+0+F@o*%@?8zKV)z>0xi*s=?oYpw9zf)%0VMY!#(I7291tuv84CqEwxN#
z)Ckv&8Vw4q@Je5)h(noJ;VL!F22ZI8%NVaIIxnZ)e^T0xftQg$Iurpom9Zrgm(5E&
z$bZ|LUmkzh=xnT35+KuQ#o~g)(?w0)7rZ_E=HmzKE|`P6BUa*K=e$+e!?-(vr2^})
zoEBJX@!f{PirPO;tO_*wC+*L|H-i1y?XXwj;X7epV)DCDLmcK0nv&NBKlSf0-};;6
zLVKG<T%Ix-YO>Kji`YSIRFmc=_a_tVLw|lZEIBQq_g*Iz+u=mS4u0sDd>BqT%`dGR
zP3wKM6iXa~^2SxZx_wAzZhZ0@OcI+8;4T}}yJB#>iZ1y+Jw8<5?jJXQZ)Rs|Vt=N;
zf8L}Q)`)UGc0?)~+kN??d)oY1uR4ZU_3ZCQA3@qVkurju5(%9nQEYfR|CzJT^?&qm
zNC%ybvHM|~PS+>qC*C=oeP&_h^rOEb#mL89Ad_<{3fCJkSkb&c82aQRj?>o?i?+Je
zHnEP%#+h@+h^Xwr4CRGp_MCJ_*4rjd5u)Zq+Z@12%*?iAgB-9ce{Wo}!;yOo=p-6t
za0^~DOq-ZM)+(}G8`c|B8=&(ugMXU1G1rVJ3tbsW5}u|_^s*0mL$|vFx3`n5+)xj#
zVJC&LewE&WkXISe{B9Rc3m3z%sNijfB%~~ld)jI#X1{%K)I}U0Ia?f+6=j1jq_3Xd
z3{Ld^pokbutwer?`B`c~k)&8OrPM!dMb0Q}eWu8~xhA5k@7nr2@KanV{C}yfFOV`$
z)^02~Ab(u7F@&FIKCrQ@NDfOOqKRrL%}w)2i$ZBsO8xn{X&;$1SccTq_iZ%{$O@&z
zuWcpl6Pgqf?IR@%iP{0bHGM`%{?@Ga83NDgC7&%oiw8mnuC8HG7Xi1LS|h7q;+m^S
z5lPX-l@ZIDm9xAiVpDG91AliuT}AJY#0H~ku~7~O$3pSO0BZ3wZ`ur~zUfH@LJ8S5
zcRwXqci>7M@nlN0I;`>}Dj(8xrY8d{c84zj%EDGE2F|-!rs8?eam}{RQYPm!zn4*u
zq<RX9=_}0ig<XB+6LNFJWBMHv{P7eEEQOnwGVFqf#tgeJjJmmLzJC!ywOh#ySd!~8
zk`O4>iTDIYDc#qPv(Z*N>IlKam6EAfzROKtqbo}dLF-uMzCXS+9kZrBVRG2Q-g8J<
zq~l<xCbuuBwqp}}L-g)=aEewY)C(PTy+`rQwN&r?sduz?tmyhGF4u`;A-${Grn>KT
zcM<HiX*7}V0iw2g(SOhFdw@gsh`xF^c{LjhZ$~EJuC92e2ICmISce#QyX%k6n!<?J
zve%XQ1TlUFF;l;G3}&niN{6oyXDY*b|L)iNc8`HM4m373ZkEqPno=6A`YBLT30(+l
zDlKclW<Zj4D543q9KQgHlnWj8YDSS#szi~plN@`?9UV1|rhkhP5BHz5t}?sL(5|oT
z$M*Ztd@bc}vV`|WeWkbM95w>y$jm!!jOEl|pe(PXCi3m&Iady%2qoe9oNFrFwD#W?
z=gRMQ7WQ-8d0cCV3Yn(z(WSUnMz2Z4uf(-7<9TkvS@!R6+81P41wDe5()=dBYEX7m
z+kH`X6-N<bWq;2Ce~F4TCDqcF3o)zK_XW~Pb|F=2OO;)(&8o6ax%72HvR(V2N%vu9
z&^N)Rraw2J`#zIR-IqR*O)W%DW>X8P)7aEn(Gk+odF5<n9t~W@(@{26*BjGqWdS&h
zQmYUdc^g!ZPDPJIUXr5JQDDHu(A!-7)#Q~-Ood0Rj(@ji>BgibH`eaghYKVe8a57L
z<p~m-K4q>5mBg5mp|y%E&b^MwcBqhbxY%YMDo4YGQ1((hCtOngabZ=sB+AvfxF%dE
zK#Q!27(!_yZXPi^CNf7$U-ao_q_A-+XtVMFQCXtCku{N$7*|KKR#D>cV_Xv{DHEE=
zXG9987=P38j7VVsnOPGjeN|eAb30BL(Wu=5w6wVp5*%`H7h_t95s-U*$bX*xD}$cN
zu@dfxcm7Q3qm(|8WvzrRM153RYU*QtE~{t`6RpzZpHFjC3!O-oR#LA{mey83k1VZs
z^Ri^ANomW5%af%>Qdb_o9;B!Y8c9+*k0dpz3u2>qCQ_8rCz7O<(1nnq((=+sk^XTh
z#1}w{a-kCm(Msx@B}8lQoJNRNQWqpdqf>i-TU$?udXK`-{twyEZAn5Vah4n^V@eu}
z``aBNKI1=73s;>=idIDB{p03e=<`Ekmx04P6_*iUD-)L!Q#>($@2~K27sV<fsZS(b
zU^dAV=&ixtl3ZFexip6c2+*J3CtIGi<QdzV?QGJ6$1@t!q#mCrlG5GhPItdt-9KEt
z*o*1B3YI)~51%^19~z&eG3adPiIfxS9=>$%Uun$_*F7$ywFoYc?*VsvcPkX|1mOw7
zDAC=mH-r(vh~OE2wzz9tjPE{=?X$dtd~iXhH@wD$F}lC|`0(qM&c<pLUHY1C^{=-N
zSASm}OlMKn#m-r;NK^O6m#g%LA*-B^9kYsLy03qBpRRtZ-#Y55?mbX}0F-vl+5#wc
z%G%I5CrZ2@fl8CMn61GC4JKs$0GxfD$1y>h0I|b&<A!H{41l$S`!36dyM$rFE<K)J
zWqkfSz2fcQ1D$Z(<$KiH=>a=sVCCWC>*uZ-dH2f!Hg--$EA>POZ=^ScCk_%mDo?(a
zIBBbEwn=nUCeFYGw~pB(B+6T!*^{8<0alw~?(H+#HEG<p+h(0&ux6uePJ<>ivn|<R
zkJy!eAICC(+j@+-nWppgJT40MSOwu(Y1*56D$uYoijCBU^~UIlabft~-zkIT#-P)i
zbl>a3h)?oyaC$Gh&ji>#I^66Ivi6RX?fvw!6vq0u^cIBt7L)b7ex%chg+XFf@TPMb
zgyfj#bgymk^ozaYF2a56OyMeHl?lF(zW#JGfC%b;922pmj8q~YV1AZ5p~zY#M5WX}
z%ZeN+Ons(^Mn(3x6ntwc^Y~_o7UU6Yt*anB!Oc5kUY+7+&&y2FrJ1oHm;t9Y2*F%z
zkJZfny1Auz{Su$fUw>P~lMzIsxRZFESTiYz=ovWun&9PGde|E(|Gi(HqIb$zZlNOw
zMzCyu#xfwFde=fnj-;Y*WFyA_YQ<VgFIy)XQcSg{m#o9uQdH@AxvV7RC=^POZ;P{$
z6xhYK&PD_qYg3F7ED&Q_V~k*d2y=^5#YBZPa_-fS%Rtzm4BO*D2Eq842;x$ZGEQH=
z8%5v-kV`_5?uoacL2@c92JF4aPm8Nn0#fLIw8NFFH75!$_DCKVHf)=cta88RVN>ci
zT9-FH({E*zH$6WIcgy*Jb!i2eA1@!M4C@o5Fc{HaJiK|x16FrpD{1xGlRNPVkn&Dk
zO2hq&Ga$=Ii)*1F@yMAQ+YmTql(>Q*>FzVl+333k(;O>tQfoQXSE+`gvwc^BvqDgR
zGD;(Cj~4hl;zY*&wRV#J`$SVeqv@eASfi?mX@B_fm|EJ^>zSsMW7-%t5!HD3mFY~P
zmeQyQw-B`wx)4z-Ej3YvCPf-$=yp}s9m_)U&=^2qkN{gyk_Csmi%33tOOO6DDdG;?
zG67{77tv-OW{)~9H%biS4VtI(JJ^1IKg_j{TYi}HSi_7oO_j%3_@%La@fMsM64~8^
zNwl@L%@EI4%)tC3*4WXS{aGtRd_vq*nC$j$tj<!9uqkJsuGX<~Zte33l0Z25dD=V@
zQOa1i7=_LQH2I=v=fvTjRP?_m-z5y+pU~#1CbFT;Q%PK)%~RX7+~%q6xon$%XQOSo
z%~RX<+-)97*^~7bpy}d{h_x3&t^EJd%Gt0tOUku$IJ2EIl|B@XC;BC+ox@U!d;p7<
z+c^n#iKWy^-zXJ8A=Zl4`bM6Vca?c7eIxIcv3{m#rEe6FTE^IlLo_I&Z7KC_xuJq;
zE>P{^Vh24mR!%f8IBj_?*fVy26lp}BMJuY9xYi;`<aMDDc{Z$47^{G{%hp*zE#i(d
z{RHqR+FhLCKIZ6t8FSR)(S|0_QHw_%T44_sH8?lHj-<qR*9v<KiYPP^KDDNtm}7*0
z3{EFXfHAZoU1EQBXq!lH>{J&bn>*Eo)V5A_t%z=J+579CS0WtzkRsfFRQUJhhBabU
zlp4vV5P~3Nn!se8T3(G<GU4fHTd-vEDyEHnhUFre8F5%@p|uzSLXvSU$g{zkrZ7W(
zV-T@XQLs(0JJ=|;6?d1~noY2Kqr5h)up`8YnOV0ILsGm7j*j8JIfiJINaozruBr6a
z7%~;v9z&*58)HbNXxnLjS0Tki*jbraV5RGc>6H`-#cCq|ROIt=q(XjVQbkkvt0_`Z
zvXfhUHiQsv^cgc)X<duYm^~`6MwhV|MUiWlEu9_oCx5!4)AHE?Dk|lB;H}cuY&knf
z(JAV=ueK&g9_gYq`6-AZmR3InOgDxmKSf67$E}WvQHe;4qf&N%<c%D6lygT9c~Hkj
z38kYZXKKq0@pSs>E_Psp*?~mn`&9Oo@+Th3u+b`*a}Q-o>5YdnmC%KH-IbQvq0EYZ
z#xCI4-K)BkYx^)W(i1ux)<SwKtfkPUU@f)GU`<e(+~vtoiqEWkBBUB&C!7swA-xsS
zQs`2UmRe?~GH)h-4CUF$%UUEvUX4WU8IA_~luN!(Oe}Rj@%WEYRuRrU{wt+760H)t
z5YZ|vv$eZ_geYAyc7b!E^ThQ_ND8Cnb73u{x58QqT?*Dx%M8|CW_8U(yBXq>(;Szc
zHmH#vPfmxH`x3{4Mp7kb`l1lt=!#P2LY^p<{Fleu8##_8)J)i%32Py|5!O=XLa>%f
z7GWI|Q1=-1H^4gH-#%RZ3z~rtm6w6TJQV>lm+|Ed69P0bmr>I?D3>i;Je_~S`_(rO
zUmkcC_KSzZC>%+qjUNmb%xW6mJ|2D^u<?F~yAPa6@AUBd5SN)Ct^B~55yBe0dx?L;
zFWC6{r}yvv{`OnJK=y0c$_K?r3-5-<^k^;f-U>N<NY4&uE3NDxn1|GAaKuP?Y$$}|
z!rS4XpgkptX4^i5RmrU5dbWR83C)aCBqs*b(kMebv8)z;ALffis)<Fs(<nxG3a4b+
z@Cf=s6&}ND{#>?#>OaHb-Eer5wW<x-uqAK_CwYT^J*|Sbuz=7^vt`ME6XeR5zy=en
z+FPt{o7V`JHYzBOf`Y6cMpom6%gXH1WeG=>6UtI~Z`T);&|DNNBSL>Q)3}KT_I!Y+
z$eK;soj$*P%w+FDbU%Af#~g*k5jg2+EHv#zI4gQBEZg(iyuIwsUMmIX8eXU2b;?Dr
zDx7-W)f@trZ-$vj@oLBEZg2Lg14hwW?NCP%s-p(rT5>b0VIg>x8z{VouX6B0U{O9^
z|MeX%)}vIGStDA?5axdbLY4SO3VjNpOq--ojn68q<!AEve3uj+*0s$X`?3-l=7nuJ
zMoO4X*;RQ-X-48yUbD_qOQufsqi(~k*7+F;SCA+CN>~q}Tcr0OeF9w4k@i_nK{sC{
zof^I>r<4|scyhaV>P<cwX-27#cp^UvPyTYUfW9sglzsdrK2d*2P4SyxH<=Qa6FQva
z(n@$0>G|z=vd$;F(L(v;hycoxwCET$lHIRCDsD<ZL2ZenqVe3B#0gdTbKp)k(*WIg
z@|BLpQ+gVqq>-y*I6Ls5j#jK*)k-?Or<L^0HmW39hVxo2ME%{)IPyu4!HNMv{B|8d
zHXz8gC$C^&O-_Hs4HMGsk~7%oX1AOwc%(QH^EMeEDKgYi@%mW}!Jv_i&c?+Wzr${A
zF0+{@CyA95T-P~5q>Xu7>R1GIR=>Jm4v+<~gur8#1Hdj<$aiw1Ta>`R94t{VN!@ZU
z?t;SUGVx1iCVk*WF`$O;&4MFjN-9sL3;6Mb4%pKPJ41g!;ZT5bJ3l$NY(A|x`Q3D*
z;`I51O5&|Rkx$LZYLr!937feI-VaYX?auWb?tNX5cO_Ja9NB-?ii#<LcZ!OsLP$l$
zkMV!hnhxCp&?z<QvQWu;P6@l+Z@!wB4!Ku2&YFc-IxOHjZ`V#iUN^e9nOR(C{kiJ0
zZRezoVj6!%`^-Z|d)y>%iRiNZbgBHzQZ9Ry&fncedqy)~PZ8`K#|wL7V@8?A#_SpU
zHC=yIb1f{_!>&f<tXdsh-vu9ySd)-mtIIlW7WAN%2s#aK_stU!_47#^sL8=rHdr=%
z5xXHY1Bew~sNt9Ihj)iJ?Wx&l?fZlbZVWO~g$#cKaJoTDh2BC<rB<uNy&9a#ARfA4
zAC(_JUXPU#f^w*g%YPB_)yfR&Xm_yUTLDwu+J!11-Jri|DHg>w=gVYJ-Rct#O`I5a
zTfvB*sF?QGRp-biLywGruunEA9D`X-WZUWnmYt`M$gRMnysyH_li|1Oz{10BSZcVH
z@*{tG?A@%|aZ_Z8i8zje<~3+?`w-M(Uv$@`;W43%wDUCP>qHeLaI*r)<`XhT7SJxF
zX3nZXSyXWd(CJP$TsNf$zVfgM8C7?Nnki>1>Q_JfaFMi#TBJvVFvy8&nm-CjpN%%m
z3cbL_qV|Mcq^l#7K~)q;6yjcL`d`sH7j%E`Czts*z`TAPhaR2Fn(DWURUA3PIPhKT
zyvjt#I{9+8$CS8aMp7%P=BuI-lCb8sFX{rLP?-o<lYF+$1wKh|*RFm0wj$&zZyRHY
zb-^Th1Eo$hAhxxj1QUM2V0~neLh9ou%=iW)pw=6EmLdRJW5ijCFi2POR1fPHvWS23
z_5?qqFOFpG%(eO~oo9~Zx#;u2tUpg32(;3R1t;!uksL)cHR$`>#ZefO-fVgLJUxw(
zNA@I;0hwfK3I4gN9VrdM$hj7gDwON?tPqFJg*hLr=$UnudQP_l<x{P}g>h-gshXG*
zZl2R)J~z4G=ToUZ&oiYS+GtwGyhDEox^^+`EDa=c+L;qSSJo%gIL<e`R7V@R8q8UJ
z%=0QYx1H=?CQzkP2-2h-<9^zl&W1UWl#M32zS$Teq6g8f2FH;>s!J-#i=5-l?Oetv
zy_?O7Hn0_DMpV^nokkL^`82~UjsN<jI-f6{Llp2YR1Y@XMoi=}NvhIH9k_p4iAkfG
zaP&l%sHxF4t=|AXj;PiUr9$M>O~;VbB*JUMrJatIsn>j36bvD=`qXR=FmHaEiNhB3
zB+@@5p8kk&CDKH;3=OZx;T!O0pNKznov1Tj3V$`XGVr3#z-tfwI85r*Q}I_(hX#HR
z{%WG<!e7nh1b=$=_rPjKxCeimSL4)t9iAUEwkzmDiN>@tA2Eg_ZJFod|9K3jTepcv
zM$h>)W4Kxe&ozc460ntip)nkx7H~A#RWHPzKWz-x2)JVm*HApe7_OCReGJ#i^SooY
zeX^A?Tr1m4kKu%;Dy%#ie!J3M!|g9Ih6^(&t>qKPaC7RtW4MZD-xz<c&t`QD2Tw*@
z>w7I~*<ldZ+&Oc)N{rc@uHTlIIbB6X8>2nNysM4!>bL)1Y=JZM;<X|Bv%Q7cj8F`S
zTr#HY571!D@cTQcSy|J}v`n8wF^E%>%x7ACMiRAGP|2JsX{1VjlR$*`_{Ynyzv07o
z`0~pk{t|KcJ#Kvj^qGIh&hVSNe<2sfEVm>wszt~pEXR8W(@#C-o=IRS?DMFoTISM^
z%)#4=n)lL(GM{OHiFS~(Fz}LSa-FXY`FR>X{jo7`G=P9O%h3q_>`)X6gIu8f9!DbW
z;c{}(krp5jwC`0PYYVbrtK$vHXB%7~Fm2;(6lM?3oHV;|hQnTzR@*qU4p{J?7H8$m
z9?sS~TLIM2^m~L)K2$-U%#5Iy%I-C{j~uE99)266pV-va)H{bN4bd|-wGEd)r>WgS
zrW5hkoX<LPfxk#nF9gn-*_i{E*$;vK2Z4XuXqSP*JQbF!E)<u+WIV`!^7OaoKOJ~N
zHh&ikrDE3Lqzy8{BVqW&RRi_rV#Sa?R1B@@pgFhzy~-5@%!nQc31KDTwG5QCFmZ<T
zu~NzDS3+dD`ojil8kp3wzooFwvqwa>BC!tXTY<-b{x)%sQt@y9yZ;;NJ+@^lQ<g+Q
zcuA$Lrf=;yM68Fu{7{L103=h|O&T3D;4-Ev-<EDiy|}ekGYyU*S2e9jDyBRQPl+ZO
zpWz(oP!)aX;(km#@%-`X7JfT>UYxup;$s5QX-mbHfV4><n1veGGP?u&^F9biN--bN
z9Ou$i9(#<u2*;Wpvy9fJ;z=i&v##UQ<HeheFdh%0P@!l3IkGB$vLPv@)1L4C>uvY+
z{qy0Mg*=r5Mg}sy|9tR2pioR3HvupwaQgM9>BHe&e`;r?ymm1`&$-r0HF^%3DP`$4
z%Mgl-Su*-dxf}$vWHhaPA3c}mi9g7v`M;q0%}S;6Kz>6LJsr}5{FwM|sN!V$42hW?
z#8?Tu9^Q_VwG2RiFK`lLd`P&s;QDo@a5HL>oHPZCD^ZiuF)(g{o5w*7@Ji1c_B2+~
zRnMFCpFA)q^I{sgY5G7+Q9}1f1)=a{>z8yXPv6OJDV?KocBFJyav0uadZYvFB@ysA
z3mYE`tU5vPo_2<!wQx$nz@}5O!#=mFv|<-Iek*C1L$?}#zbQI65+7(B#qS#W)z3e_
zj-1pmgi(T&5n7o4WRUSSN>jjBE7ur1Y37Wys*m5`K{P{!5d^kx(QZl1UrFz`$+4V~
zlC@C3z23}UtY(}{q2+lK>nN_JOYP`V;F9t|;-Tk%GSOFm-oz6qd|2+rY#*}_KrrP=
zL;BWN@esg&G9s_%A?Mq;NrFN*=HUu%gp)yi6K-fJddpS$8Z(4aq!q5ijc>{qv`2U;
zU(D&@Y4bvh46wDxS7MugmOM7775<HHaVn&>4agr}9|53HH)f+0lzscq)v-}+-Y3ba
z`=uM$SmJ;nWycZ*W*>UDR39m+CPV8k8fE%A#gXiPXjO`fE`74w_C=LTxwnW}ukQVF
z=V8~>Pc!XZ-;CRyGH9|0?_Da09bM3d5{8EUUKLgpomADL=@=I%oMT?f^h4Y-CGN#y
zE$zu{jr+#YGOW0n?ICujlq5#pa@tcqt-wKjT8Nu@ud0e9kul6TT~n3BDkjL1viM$i
zkvn34n@{0?nMl@eryy|BTIp!{-;-Y!UPZOm#@Bc#g!3AFc=OMXAO86MM~aox$A5VU
zssKo<g|pM=_~^(3N(IxW^hk9oX*y1VIZ{@f$ikf^fn~7#{C+ynEes{Vyy=r)l+0LS
z#SH074vQe?!>grqeK;;J=DR-Ciy5K$f<R4w!NEWIQxdFyB=`*FAN{9(&RYrU@6+MK
zba+=d=_LL0RkGoF#7MeS7%6II<|QLZn-$84J#Ek=!Qo;<W?BQHR?|Kt|LB~d1iD;R
zD9tMswRwq-$|vNd{N1mwP~Q<6`H}J`@(7cqbErNcT@;C>k*)LJ?>{H7j|B9#`i{qc
z90eu8f=)W3uU<?zE0mTb(ve?H+w<=1SRYSPJ<OII>~79?y(Q(Fvv;g5t2>lXcsErj
z22zhRt?#a74+%*rbp?C+6uM#yB}~3AklC^l*D?h<M(sYZCHINiJhW~nyP{G0zfAl6
z3Ho7nUWMJJK0ln5N{ax;GaO?1Yf)K$okTeKX}Dv(Zq0YjiBJt+7dTjTmUN3zpt!30
z1mD9@6%0>D@o-I}>2$QlH$kP%cO<p!B&ZBk$|dey=Dt>lJ$ezv;)<SYzEY_Ah@@Q~
zy~6&Pn5xb^r>J?3jFpisIz=*>K9EIT&YD#x-<R2Jsz#>iD$s)WsMSf#ISTiG(CXye
z_bgX8-Oq-=oD9Tkx_Z56zsY15)CYFJFTRD%YsQ}&ab#OLk0nY4l@!G$`BN;%8w(Ms
zVQ`6UK7RH$VoD~0n62*0x!fTQLaFq;Fx?Bts+^FV^HZZdR&Oa!yfZ6KIV+rAi!z>a
zh5eYMa^4y=m?rgJZASp4H7^H$<YFX;i0QanRP(OTQLr^NodIg*-%y%TGO3&T%g)WC
z-fcF?{UwS4teJ&!uX+QxW)^IRUSDslw<hE7=&GiV|3J;vSaVch{%2J+@N?%Xe1BJj
z+SVj1ZEy{QKX&+gJ)sOQbunYK$vWa0S<4mdqNSy0q0|kuO2X-k&>qWwS}2N$YK^G$
zJ^tG1;80gm?9R%vrk|ojWhkWrHigDEgPmA*UPtN8oJtJRn>kmB>CHc;c2}xCj&u33
zm8PZLk}`THY~61iygxchw3SUp`J{%%c0X4d|Mo<ETm9S)sr3kYRTovFqryY4bW!iq
ztB4)f_NHaC!nM&n)8Y4jc*h}EE9p~-P5LTl07y9ym4O{XcYYh|+aQh#x;?V*^EP(Z
zQh>4qJzI)739wZ~Q(UzVSZEgJQkc7IGTcz*_^p{ti_~`RzH2gIB@<v?z+~E+uI}&2
z{d`nB)a|}euW$LbbI+`-W5($o>3f@%a}3oUXP_x;syq-kN!ufTv1ouc;hn(?Xh3t*
z*YEJ&E_I_qx3@#|-cq-Uf$vytvjZNJviC-<u2+FOjI{t#wrR}&FxCP9lrG`7Jl3Lm
z=YGoN=8KKCSZ2t%x4imVqb<T~pp!2&+9D{it>y^Sy#i_U`|2x9w>ajguxjz%8g8M2
zOgq;B+;g@iPOEf(2(3-FXl6*sI%K}qWJ{b?xe7Chq<vEMn7#B|i%=AexPDQ6+)nJ!
zzj2m50DVA$zheJ4vn@gj#+~`X*_Imm+S!%{=DOLI@im*XEkYuJLOuJ??*Pj*+%)~t
z;ITU$k*8l8DA*a8L8`c0sl58__w7*t?^CSkBze=sj6g<eNhjD~e}vae%ux71hht%`
zOr=PmjLM88vpSVR2w5X~zL$xal~?viQ1T|egObelidOi(f}AD`vhqfy0Hc}o;i!h%
zn<Nm7vcXQi{IWACL-&}@+IEH`$o2c|xMjl5qzq-ry5rlGlQKx6)eG+JP0A?Z9^Aq`
zDJhLMa=^aqimw|zf0Bywh+eC<87h^bilsA>-?s@QrBYIS2}7w)rE=g_hC0cNvNti*
zXTyEq$Z5qQo^$V<+*ch?3a|bbH@)zL(u(OS(sFXT9df0^wKj7^0m=uC3Tk(-$(O6D
z1=+D(BbnAMgx54{_xuqhyh_NPsB7&=P<eY3ZPGXPg=j%`Qp^aDloP&yI7t<?Lp3C>
zdS!zG2J>6-6;FO;o1VJ)Efm;ggmK^Y{8oC^A5YA^1=qksdunWA7MSHM*T`HQsjI}M
zdwJWKS6_irh5KK2Ug+kRfx|o&m*S@Z5|?mqJj8!I{_^<U1x>_$`*=}^<ANyd24yY5
zwBg0Oi(dy!ydT2s3rd8uYWRKd!vF#&S{vMjWXug8G3Bd&y?On&7ta|1z`s2N9VFq3
zIXk?IkB$)|6d&HEM@gs^N)C)TkShnLh_et{2FuSch6`XJQgA*_dg})TF_x-vreum}
zE4hCd7q5mA`mkIc%(&jxgPEWaLExs);Gg^|3D!RXK12COf7Q=<Dxm%`T)ZAG9*QKL
zz(3rzp?Xe`K*|j4r4nWyGJ;sCq=?jGf`$rb2OBoi7zei+^CaWMIb+;VP!&$|NX2a)
z!cuvKJe2SK_yYGGqfs8w92HS+uvmv{Z;O8{i8P9J{Qcrx0(%3X`|2K-IWXY}!%7DT
zm*`;D!9$_ZIS);<)8yo!;^rt6?h+2F;BboJ_bJ_Pf<tAzbB@)n2?8F5ioU?-VPy1S
zuWX<prQ{CTWSXWjQO?2VZ4t^s*vbq@LxgojDhN07Hc7&V8YMBROq5ZO74vWK8-agE
zaX)@!#?PVYy-Cpi9V;v7^d?K&)a!?})RFrj7~p7mBg-$H*TjYS(jJ-i^KbnO(a+E=
z)mgL3j4LNyhY^+G6BW`{+}KR3C=~hmVZDTp(Rm6@p-}FoFs1}0JT@Qt1C@7~jzCMa
z&Q6m;!MlrntE@?TJaDP94*2m9_`840GzOm2j4)qDZn@OgwUlv*M8_#|5-2v!>Nd^_
z7tfk5YmF9@@nz|zyVRBaL_bWhjdhq-H?pCjti!W@y>`I0yKW7(ARo!8!E7UjQ#9s8
zjEt-Xwbia+vT$ZtE>1%4@K)1WMW>7_FhP@w{EDY}RU@rSQ=FE1<=k{a3gUlDipKi%
z<m-lDgen>Zs)*~>?4mT1`YK1L1ud*fYp7Fxs!c7fs*)soGfOy=iH&Q8)hnf^RdzdA
zbQ#S}9l~tq&Z90kPh(yelCl|HBuZ7%?kU;J%|v#Ek3y}f`3z8_{ARW3gFY{gE0cAF
z;J~MWHMpKku6P&qz^xgz5^8@zq-JY5GKMcyP%#N#gUpORyGM$!Y%WD`s)^;cr>|6|
zzWQXbTAj)$+JJu_<ZefLC9marWY;a2e;wEsRV<+C{!LM9UGOgDIH5ThkUaz5R8Rt@
zB0XlLY$jfmwQ9Mw*PR*ZQ7wTkJI0w3jiUM9REEfkl4_%<<6L;#nc;t&x#aww&1K0%
zSzgb_{Md4Y+!cQ9kg#J>R!gl#N6vf+G}be*1bRQTvr>O@s2Y^DN-noEsJf^vy_U!-
zN%1Rqh!=shm0v~$1$LG!zO1h}VR9^rRO-!mG@0y~qKZ#^5}RUN5>2iM);{kriLiM~
z<`T6wD({Z4<(!+`@0)){d!y<Uu|^Jx166j5TiQC+Q$!FZQIc5GBANsc;o9t<x$tv*
z@Cff^B{P|l#@(@GD7`Bna1^^7tzt`0O72e<HkC$)sJEC_OT|pIwYf4zq)MVGURu*j
zb~FlN!;vWNf{)ezk#So!MJ1N|oF>i0n~LPOJW5(05B`K6)e(QT-UDoUl+=uH$G(6|
zwV0;nrMp5QDO7Biu0fzFX9ttgeN%_)+gN95y3g;dBgW~T=%;o(WFo1yB*U9dswx-d
zDr;LT77KuWg!&Q&P$Sx{f4?2+V`{igW2-YNJcgGi(791WGu6@g)GCMfavxq+g~4x<
zq4y50_N$*jI>LYN2?`WD+VAT*$|J$EZ=T&xHen@agi6*V?`hRC8(z{LGiW4+nVQeS
z&uz*Nsi@;KeVt5+RHpngt=^;Zy8@o?-nE#CFLB8b+pBp02LiTGj7Ztb;O0)}Qr9Q1
zxYu+_dlh!9UkRd**A;H}-4klsUnjZ}8XdMWz8>{2!XAGz8g9;sLzVjR)$sb_p<lI&
z6tq4&2<A(*R;ta88s)JCHpItq>t-s%y%L<-FOh!RF#iem+Kg0o%3)=~@tzuXx0v=g
zr|1Dx53$@4o==l>#0fp2iOtiaBw=ns&c1!Yvq^J0VA)3+?2$M-r7hL66S~s#ilfY1
z-&P?eM7)20vZ<{)D}!tthB|C=uB5AXgJaLSV|y+2vsDIbpziYSWP+#*RbvL&l;-{@
zpe<3XSE@zv5DVf<cI6?O_)w-vgmUh~!%x4^;U(UEH-y8q;rB4}-5{ZE5A}BV0ihxU
zfpyGQBCf@@{*qrsS~Z`Ai05Ls`|dgs``-*A1}}emVVg)JC?ncNP(+~!^0<mcARIW=
zVR^`C=6MmYG9y_F^OtorFcCv#3-WxO8DTMa7B8p|xYP?0Q?nq-)77#6GaPmukw#m=
zMGN(xzhA97OL9%F!K>=3<FbYGR3#IWj4-NtZ2zzRbfI_zaSA<jFOr>=h=1p|c4IsV
z@il)pv18mY&0ZKAH&gRo*|@J#PCfETP=ssu_twn9pS&8|@^ksHY84W-219KyXBA_p
z%$q*C4R~pHJhKv%!Ne`EjyX+JAWdu$d6zP~bS#LNbv=1fqg9kWY(4I?!(9`MbnANt
z@M|%lK(4p*lp}6W=x!<8l2+nlMOq^tYW9DWp{!AzOhg(=l3U|kpdQq6D*cVh7!g!U
z(?jLW!Jv=ZB7?f$F{?%^3?~xBMZIn@+J*RU1wnn>K>T$*U^)=5IwU`?hJT4%rX;W^
zPIM}zHsZhZ4HTb8vfzHG_zj(?qB7*_{5Xj=ov(I+_D138&T1td=D6$1jo5U>5^sMd
zZYzA3A?mY%>MWo>(z_XJI%TRhk(H<^QwR9%aV<@9OT!t9cUqM&U;cBwd_mK05!@S)
zAFmFKx{aLdcX))g^T>>yFMK3H%?%KT4o=tQ)pQDsx2wPAO9TI@SI<CI8Z6aD2<V)*
zQ2b}A=sGfi3g6fs8KoOll_iFo7HWT^s*-{@Ct68ulvRp`?Ux42Mun&>VWuN&qe9dG
z5gnFNQ!St|fMFA<DYmN;!!Qw4PqN>>6Xn(17780V!SH$7FfJe()UX}4eBf6~8`~ab
zj_sK3PxoVJaO3HID4KXGhUyc1HaR$*`4*y_3(J}P#_jEPCefleyRh536cT@!M3zN1
zV2Y8IR&Sh!!rI!kUJs6A`d)7xwUjV4zNgvqR9e&oHLn}Z9#e++F9kiQW@3*gGcC}B
zc33u=J!TLnsAyr<U71;bSgcx*^{~XyYFok2VTM!R;@Aa|&1V+sqUYBjo86w@|Dv{I
zXQSKWf&)`$x^vCm;(h__kvf0fld33Nrxk(M_Yn=)*Xu%oQ3x{;%0}wp2GZyD9369`
z;ZO1`9s7?f&Af^i<mM<*#-3xYe$!4#*6a6c)GAu{CCgIQw?v7Fve-t>S4ff_UcRxz
zfr!#E_w4FmE<gPC{I^#xf4Dj*d->|chxaePzd}8E`4NB3<;z##A^m?oVK%?~@cz|$
zT@0IWoyqsTa7l;jJ+aq^>zrr3y?~7qE;@r;?=O_^yj6!~y}wWlaqYbDskkDp51$ds
z-h)GixB&P6qqz*Lxb#KfYBCcr6KS3{nZd+0XJ;j|irKx%O!y-hepWK8vFuG|HO$Ur
zR)g$JX8V~b$!u0+>&btt20=1Y`;u7=^%n8LU2bJG0CLq?@L4GhZlg5o3Eoa=+W$|4
z7Aytljf8_ToH2Hm9()aL%~^UtZ_J(Q+1kpV&Tr_NSv>^i%+Ac?cQ6>A2z4X%pi2$^
zd8dZjNIe1}R(GzmC--~|PdoDtA_vo^H?ltOuqN-!awv@<#@J#rwGzG);<}3|s|t_~
zM%L+ZBUw=&Q9C4C$qE5Iq>tT}j0TJJex8x6e=g>~Lom{Qw(l;(NtpgZXOk_mPyCcp
z!#b$>0-yafi2lT%{j_v`rq6!Xh4G1BdR5;%{}0v{+?8<x6az6hHJ4E}F(a1<0zD3w
zqOL0^f8|?ik6X79{qA3JfqqFpn3tRfA4L!#k2($DphlbmZH<K?uN8ZfjkM8f4L85O
zXLu!Q7m^!Edv_tAfg{b;<zq<BoHIjmL&%&v-vxY^E-;%UzN08r3Z44;R;Y_iRpKtK
z-s02yWTILjHZh?+e1<BbNec4B%HkbqEgSJpe?+SoyP#S{`yo;-Aow+DsqIQf-q(In
zi~`rP8atwxtC%{{5UQ9>ZMNPbca%~c3{<5YY6}r9<e|5SA?8wBNU38V>mW0A98wk0
zJyRU2p@8~2V)V(8V--;##caKyhWJ4Tj9Y@#juUl4sg{K%BEAkNNhc~p+0sc&wcU&s
zf0D22WVDd#bY@hRRBB@d1&XTmRwC9U)=?zwbeQ#=xl>NHi!7b$Fyf2?ahU3iYPk$z
z>Y}Hix7@`Lt2zZMEQ1c{Gm50zn?Nj89aN4wSV%4qlQmW_tT^af0;=U<CJVYK!}MX6
zlOH-RrAt&hF0iU3DOEARpnQ2^B6TU%f7vM>W|r&h6wH10eyFBh4yyvO%T(>GVD4eG
zR7HYVSE!%{%+98y5PprhDhU(A%=_9K#H>R#bWw7x7PLcou43#8>+oypisaf@$X!WP
z2T#%!vpW+}&4z6vsudY}3lOu}Wc|Uzxb0`N5KpDb0gZYIwHLB7Dno<#^eGShe^?po
z7&2SKUhBZwK0~fUsNTt;DqdF!K4$&0^`dF0YptbfiDfY3VJ^*v$d}p<JUtMR)k-#Y
zqB_82&P(iv^5)nqRey6-(%O50-v#&}J^B3e{>T4ZP5Uorv-z@rdGlszPtPxAzjXaq
z^Tpfg;*|&N__O|R{dfJ>ufp)~f5Y^A`3m;v7)}Svbpbt*S7$b$i5&TS)_>Wd`Ir5-
z^B?E^H{EARZ3+l*!@?3MW%~0M`0sv7h&U-7*1`o?Fp0XP@G*^!!<GubI|klFPK=64
zyrvy0?CurI*@bnlL{5PgO6<(ol$_atvUgm<*^@hi=)Y{OpD!=wGwAb=e;>ZH|37=b
zTwY!OrSIQeEZ^U}ap&`2`=4)qy?Xj~`l0`RvV8ydtL1mI<#h4W<b3LWzU<#z&foMO
z0;bo6{(5oVf3>)H`)+#Oe?6aFFBg-G*|LAL`sw;|a{db{{pmS}^7J`#!BC%!Gu>E0
z7hz&RcW3m%LM{c`ns3SMe|-S5MGrt>4<8$C$tj|5xD+LrAKVfc?n2}(9ViuGJpRGw
zya4%Ayw{qOuuBFEt3@aEjzmsl(SN#`)$T5SpVp<P60!6_$6NZSP69VaOYD^NAV>S=
zmEoC1Ge7kI{MWxba6<zA&TcL*&vuspzB-Ev6iWbl-KQwQA{oDLe~NaPugm%3<rSv2
zW2?~hFD6S&^sJ-d@vk43-@aT<meVc_KRlbymgwY}VY3G%STD~i%mV4_1Hym+ix?gZ
zqy?-pJcvD{;lU`@SYvq5t};%2Fm5$&!NV6{^e-0k^Ow`*tNz6|&-x#yAD3shR_(>)
z-P9I+Ih`%9Yiv{Xf5!4$&u<pz)9Y2lWWRkseS0zaYW}gVX9U}(0{46|S)gu>n}+VI
zAzZIQysE;JS3{6z;~!$Ym88<7_EtA<Q<LgqoQ+~@R2<m6xp?H_5KV1a8<;GD{}X{S
z5!ESTxO=%c(8_@yn~TT54l(a(x;T@-#f3(2aUi4<1W1yCf3v_-0CkAuBzvIy4k!0W
zFg!jvb3$Yb_mt7#oh3SRto7Qnszb<dtP=Y#K7-Rk+AKc92xB_c#aG4RS4O-4;!BNX
za&jJ^Bv=LwhE&mkySL{Zun!wia69+~*d{#M+D8PCWy9>SM#}LZ4Js|faElfYECAwm
z5zU=q1A9b-f3lW9?@^Q~I7`GHm1yqJ$t?!C!z~BXNy^)FlHzf6lIa$mXjzj(Y%~#b
zliJ(ln-rQvP2wiW4vIMd-jSe`O#l~Dal+w59X!Lq#m5oRT^Yt*;u_uIKTh*@+zb-*
z$o#d~Cxc?scn&TBvk_zSa0$in4shR3hXb(Pz%AdxfA)g#HJ>Q9OR%NqI9}{-=~{s0
zlOnDH)DSE|H6NKW9jQa`F4}J3I&H(!$3l-3Tu>HQN@`jNNG}q7Qj3Rlf>6Yz5;m*h
zkyvIqGP*r<B9eF6HnUbri1$vAzkEuGKP20f{wSflRig1Ms9p0uC4qwu3-Ay$dpNV+
zq2_~Wf3}2en$6`nnk~1pw$$>n+^90QtSxWewyZ5TI?nn3hEid^PoIa(W8ms~3)I{2
zjnBxLU1`9y?C>H<q6EyrvT6!34#Yl^5`wjewg-xYMHwW^wgr-oRN@d)9guuOs=(H8
zBBXkKb$PLTjV7-yrypNWf4iAnF6Y$@($nWue@3(c3JOZuY=aZ>$VBClw!v3Q1TeZ7
zpRJStUw~@DkxJ}u14SMCjsFLaq{z!Dn!xYV#r%400PHsgP?@Lry;8<I#nx12u3z>k
ziipNh(tV18oduM1daQ>xY)8KS7z!-GZ&9Ef10F+xcAB}Nz|sh<wEVKH^ULCWlR}fI
ze@WaVX_7W6j<Bt{rVDr}3%pb^g05UsFuwyT0MP6+*=9OOD<TF8Bhp1g$jhJ$&IpOc
zj<6l=$O9!H4@e2&cu-%ULbxAYh$Q5G+OR?tI2q$9Va0p&yI4T!$?V<bw6R4TlPxgo
z8g2Fkvw>&KN|qgmrcrYw@rb;E357!%f6oHQj#E;L7YOkJ4E=F~7V_D)bvLXT;?AJu
zVOdjx6YG4n7%+4fg?Hqngj`tIYswy~#BtKWwHAZ`B56Z~0BwM7+f$KaCcF@^lieDT
zn8#BhCC8~@f^C=V*!bTM5{}dWU$ID2;S-y`N#HG5w+J<~-j67E@=?3*^1t)Zf7>P2
zbSnU2+a5R(zD>3%5OnomPQqyy6WcYbaMYe}A>L-<Rrd!o@o?BVFtJ_QSutm5Y#JJy
zhS;P{xiJn7VKo7fNwi#u!LR^*MHi4A0`hH-7K7KKh>?-Qc`u~*J37>Co2p@RB_sn}
zl}>}*F0A5*$>o~SL`GyZj^ez*f0x?%8KPVS#6foKBt_y71XxRosb|s=;fqRfz{d`d
zQG*ON$JT8G1N9XI9dJkwYGYgY`&|l2(Ft#E0#U~IspLQN^VbV}JV@3XJd%Ig&RtD7
zPzjEf4w~?a#zgyMUU!SFZiVCk(AIs5BI|O<z8?9o?lz>~MgH56ewPL|f63MT@Ljey
z8|_x$QTQ&mGOd-StILpUDL1LT<@T0a&c!X~;+Au9V+n37!Hp$`Ng5~nTJ2S?aVlmH
z^C?H-)B#(h2<z;qq3w~Xx&TT#35+m&czqiqM#**9Zcb2#_{<?z!;E9Z!LD%yaAbO%
zV<$^2G7JvD<LtIIM|PR)f1WptVb?haoH$0LA|iITT8MPSJ1v&<m3DcKHM*FOyrp{R
zJYe!mXeuDbatRUFr&M?ll3twv!r`)A@ki!oM{x!Ax$Ca(!yEt>Bv<y=3WE}i^n$Sx
zpH|{QGQRZ<RwQ`HLN+&8@m(ItEizJF_~u%NHW9mZ*e~Zi<PvRge@1xga7k<zAe>IT
zw4A@5&1dKH*@x-J*T3CN-d14D{PcMUqZ}Loe(VB2q9^<F3v<p~|5yjB;`JaCsIiXt
zKgxDx$CkD2#8KL(C>dPd*|L71#CX40Db|O$BXGLM!EA{;cBS>mUFmo`R^k>HNsDsY
z72DFT*cNHIuxx6y6G%#1B&98q(iTZ+?eOfsGzAT5mx04P6#+MwahNU?m#haTCzl_G
zJe+^Wqfh^O_{V`~Vat1zP-^C^8jMv8a%2hbct^!ZJ4XX&3XhXo4!`4<`RI@L(1jC=
zx=@-aXFE>QUV^+L?NfyZx>1+();j4*`TSt6##|Nlj&JS4_bI3;kZPN!P@1NHNGdkG
z+0>fR$ZBSU?gdRu(+ba^4sx22@z>Nrw`qU-VXILUHvJvO!k*M!9^cxhc8dlf$#jrF
zdFF&0C9o$!0fR;J>Nc~qK$o&NnHhvsMHmwqu3Ek@l7e|*Mn+{@muhQ4$v#`wWllE9
z5zPtx=yURdDBC$mkt`p@%z~^~mNb)6_ku*s=o8V+f}VpTe&6VG<~ne4<6p&f8cu(6
z*U5@2J`ff5*SJnA=u5fI6?GKM*w&;0wK2*RRpq_L@u;$-j>^-hp5`9qUD`>e>?kDV
z6-Pkp%6(RnNLk5frtRcD#rw~31tgB(nyPV((JOLVN&7hob2A~nlA^2;X#J*36DOES
z&16#B{wFwt8G5RA2PL?~a1*7DPEUWy7vtO}P>g$v_7d{0*y%i(TomvD?$QBbQX6OR
zHl9-g;NmY^K3S}y07688ycb=NlI<K*nOlw@xj4{dnG&*w@d($u3`v>3={$wRg3ajF
zs&L9#V=X-9fF@c<YcoaJxKzca<mBjCV_=C(#?r;4GK7|Kw}l3&4712P?f!pnFC!8{
zCerFAMf{+G#+2R!eTLV4_w~cW;me`76V${=n&I8&L-;`nL@Mi%5|9bh@cS>r`@`FQ
z)vjjZb-@==!Hv<nMTKD+0YQ1ZE1S*~xJP?a_~n-rt@$M?J`@EdA~VH@>Awi&X5_~S
za73I5$lD1gE~_byWKhp+JF9=<n#apz2(IR5GKpn*jA0(hp_gYDmLDyf<en+-cFHD&
zWLDEmHkE^NFIWEFaGYf{UJFQAm?`r_=xrw?PmGlAIYmwL?5UTG2%g`ij+?@KNncX{
z_qnRPq$gVRwkla}$Ao~?B}%d8OA0%sb$JTDv#K)V<4Q57P#QufE5(1@n+xK#hds!s
zr2?&{{^sMyZ<7`cEz;;Pms*<t6_RB(Iz&vp!p6D*-eIL{BN7}8Qx$op^9l4D+U8>X
zd^7(9khizvklVRzss8wO6GzT44s5qNZ?X?EO=jwrm=c#bS3`%;KeuwHs3a8gLGcLS
z+7(HU$aH!#fLT%|0VRJFY^G2k6n>{G%p`yX5}Z8*N{VdeqaaeFe#k76f}1qDAr1Nw
z+d4s7_2gzc(VIpm4V`E;GrBF{vWrwG<Wz?g9x0H6`@ke6%<WXjz+VQq)cp|s3Smxq
zsQJG`7#5-aH^$gdiYXy{2+@Z(|Nij)yLUf>%kbepfdUB}s1|<+&Y#n#vDh?JGki+l
z7TMl7s|TF)5bJ<HVt^ke6vA=f-EaUSrX(yfjeZKFi0CdrWFih_gHt3Y2Gh_eLtL?p
z7Ji@V!6MbfBJODvBYcHbGHv(-eW41U;i<pNR8ajo9NrIyx1&D2#J^C|T5guq38unI
zQLD5JX)G-{kR5-e#{?5f-lf52n|TbW-5CWH2$T`tqTCu;jXNzP)o#lWmMSZhq4Ms>
zS9afRH0dLUQl`V<j5#3!B1KkdQtR~hyU&^I1BmXccUtBsq(@EVhokh0Ij_T7u|8Zm
z73S$>c1hB=C#kC|m+}0P(q3?=!nvfP+QGU-A;R0C5;1>n^*HVI?OyeSTTAYCsHe}S
zm0H5_2*I|fWhE1o9oWE;R*$&eZ5?sFuMd;)t2s=j-&Lj>Zv&cj&*MBh)fdq$xu&RA
zO7oxXRs(e-B%kS^zYtulz0>c?SM^VLsMgqYvysWd&$>EcmLe{`zaMW5_FVobjWmVr
zb{kN7x@vzZiqxi<Y}KGW`a7U#&y2G5se68%b2^fkZa7kTINj=mU&m{&>24>}sGY~-
z05zoeo3UIeJWW*qpSNx&zK&y|NO!;Eh;CLaw*uLPqC9?r{X`RBpXHv)^5+7!(u9Li
zZ?eU6iEhWY&PG?jI07Zdp903UGe(I7(<Rz@JZ*nGze>XOnSkJ(bp8Gxd0xAIJ1bcz
zPdcvrGM>q9E3jOevzh5mUG>RoJex(`sZnnx!(Ua=D7)S?W3Q7`a_qXbYBaKxET_b@
zkSwRnmy+dgLpv*%w8yHZ-OBo<0v0Yt$jSv4?h27J-K@#7tHA=8FmmD*j{Y5|XbP9J
zh^&9^l<lq#3R!5;h}A>)c2}gO=5)KggAk2GPxQ<kig4(bM)q75d!4IYh-Yg-pfqMU
zb-a9okECQ?y#oANMo;YD5%<i}SEH|?=Hz#wuO)gZ^tD`O=-X_oOz3Q*b7>>SO0w^>
zv>;6d{AglTQpK2D+CJ&ydow-$ghTY^;t7Af7-2L3_IGZ@%u0l0h^PCOM3hjfQ^8Ah
zV-SE4809N<V^C@_sh(bC^$nS*`uOM9He?h=d?(*#KgKd)Wk>L=?U>YH=_Fd|#^9hw
zyjSl_HDhvnWtU{3(1JyJ$zHn^;~gXaOvU|rN(}eZ)biTjigB}6jQzW`VvNIX=_`Lb
zF%|LNModkzuMe}xWU~!pi78|2E=)H8{wJ+}Oq-?_qH29CU)Yx}O-)2s`Z|hv*H~k0
zTdWA?IkWJSRxmEM*?7&YZFlqMX7?!ce!Mj;Odi!dBEhtv!4Y%b@9iB)%pR;_&Z(($
zq=dQFIRaz|5y?8HceRbAh~W;X@YzPMM-D5!7#wI7qH#^JZCXMxDxGgA{`hmdC1r3h
zZYXv%q>v^!yrt9gm41}<@vfpn_my^(^T_uth*CSMFB(vk{2!mP+?NrqG!p|cF*%nJ
zDh3k)F_%C$8Yq`TkUX3LrMKXaJo*EF0_&cg!l#7Ihobxp4v!;)vR?j29G=aA!-w<u
zx{!*In!R6&uN#aWF?t!oZot;yl%<5iX3KMjlvy+L?DTWUIrbY{v(~7&3Cf|8X(XPL
znxp&@u@jpXs5ugWi<2E(ikfSyjPQ}Rv5rdOB<L^B%_$1%I65D6>-j#_m?*G+8!gYv
z!*6eLQkmuA{<2)0Jlws^#KRgBPlS|9Z~yw|4O;DSuy~|yi@K-u$TSvMXF?!kij2hw
zSZddBy9FAKSA}Y$ZjARwaPa^aL^BrMVB(Q*vAW9Nr}1zUtEN)8!NO5$g%IZ~T;kxU
z$#mt7F#`|t?05+491_eTEgs{4?)u+#hAV;_Ocvt0z@vlGbzBQ$-%tgRrpoLD2&Dua
z>g{Z|Ej(nz9}jH<?^>agTo4d_+{h=pApg5=Uef|FqibQa-sNEJ+ElQ<eI``H_9UY?
z*Ev;K)>|BGNa2W$1=brJsWI)Sa`1Y4gBe+Y8gQ-mHyF?t<LjJpWOsv1#Dm<%i*+~1
zU>nX~`qr(rd}AkrP~UiaE>EnbkL+YLLKotRjh3xWMirdq-@s?KEP_GCb(e2hxT@24
zA=K4MPHavLBk0(F<pY^imodgY6qhkbLKK(HmORD-A+EBwmX|!y1_b<#!v2>r#yu1Q
zFfx}xpB@vJ1t~8mm%N@lpMT)+tKTlaKats3DT0lc@l^hLetGie$;Nc01=~<rc7h~T
zZ>~?`14AsV3zbqLkf}bbt9K_qm7})0igOB-Cj?4b%iIf;vVz%H)@q@hYXcRXHgL?~
z%y5_q*RFA%xw($d<F)F(<C~zy-<sli6=N7*!@t3+F{Wv_^E~8f9)I?EGmMP!MJ%ms
zVkr!)-u(ISzAEm#da((k@>&E(rDKE<>DyjrHWI-K2RIcHqc+;vHt85_;(<-H)gAy;
z;HA=>P6#<(yKJ6mOicIPeA67XklEKX=YY`xYT~pqyv45kdwVFet=oW{Q9GR8?Ho}~
zDAHdAYLgpd^+-7-tbeqn!6{+Tz$h%{A{!_3nQ{<qTbtr9bCz`Bj*If4u`J16WGhea
zfnZKIvb<8!rxczng?nj~WP#74n5^f#53`XuLz;zmfQ>|1lAQfIliYjz$ptVHkXmI;
zFO7zU@vbLb{|b8+O)Rb2E)|5f)R($@x>%C6^}ktkL?Q4h{eM6oP3yB(1XRtQ)YZ+;
zizT~y{pR-O?T?G4^7V)G_w72KU46b-I#u7^tS1dp>`@WI%m^r}gbw|v*Kem4S<F}$
zMqM9+EsCfKMS44RCe4*ey@8Nop6g&>r1+2wYZ-)`@N>I9O?aDinc|nwC_{bQWb}H!
zl*URbw|T6lO@Hu&P#37qE>EXcBgD*R)Mc#h6D5=#5Qw7NDWPfJ=Q3$*t~VxC5z6((
zbt>L;iAM1dSSB7%i7pIIYng9KbUL!3O=;d)7+G7PdAu-=1TPoHzBF2)Ne_sXK*v)E
zHIp?jvZyy+#l;_o*Fbz~3I=yyO_&yI55iMu*>uLS`+ucR$#W5)9q7fx#aJ6F+=^5L
zYFfs))2z7&VK;#Arj6SLV-t)Z^hb{d#)ZMjQ1k$*?E%_#-tPrtV}tP09th*U^qw#-
zgpLH`LdyWg)S2;e5~@<5OlS|?8m+O#*#@Z^%eFr0bl)44#efCZt3AVFyfsmo@EXN}
zobLLwNq_ZO#;4BRcxM%Fu^L;$wquX?Vhs8qOnANLK>3xuNXbY@&zmVgS$o^Mp66lC
zcg>(}*;%!{?g2B%XiAXihw}8owq19uGHjkTwx!nH-!rATG+%oMLa!f)=IdA$!Ui3K
zR2-UzCWYz?F?}zTvW0IGL?0Ow!G}WNbMpL#xqlZ1jj?<dAkL*R$71;nZ{N={bfjMC
z(z1u<_!<-mqkc7*-Ld*Qp|viJ{@?I?_=dIF9mw;$();uLp42R!U+UYB=l7%rB=9?s
zz);`3i+3F!<6K|eeOxRx@5}iuVhI!GMMI%cap1%~sU4i(84;8k4Z{e8G3`}QMXKj<
zet(a`;b%IiL0_FzhooDR6=uTG5rqmZW3W5&y(j7SD7Eul_D1N^<p?mJ0LH4>0B}3<
zYzWHH?vA-XO`+nBiNRz!)&@Y@0dn?n)$hXSPuv8En{P<tf$XU-y(fDrgpS0X3N1U?
z)2v+2+6Yia;bA!bMu1YWs6IA0_l<5h1ApkS%>dXWEcrv>+>_oH&V8X{!MU&H3vULH
zvsfL%Cu{}~Gl-7;iZ=sPE32}3z8TVVn~nH=J_9v>Z{IxQgQi7XcUhP5ITphw^1^`&
zJPgAp1wIKtwbN@5;6tO^BxZuTr}Jq(pB+Z!F%fMDr4v&h?H#HLx~tOd&q(Ga3x8$w
zz@`z>9J{G7Ov91*Jjd;CwXbb#Q^Q-@_Rx&pgC1yqbc9Z?j9Ow2slt~1dAb|$bpK#H
zeb={m6gL$P{S<#MJmX7zN9eE2*#6l;(5>)6m+CfzjkgLp;79iytfxP!l1Gs>eKjxE
zx7X{thTUDg`2p6Ftk<t^BGID`(SI9b$4U^pevs#p<A4%3@LcX9$6m1UJ*iEgqbE<N
z=_&d@FU`E{YUR^H)ywX*8g0|J<gUaw(>hH4?`Iu}kJCDsgdf8VTHLK7amD2#XFKu)
zT%*{|M5wkN^_XWMH15b6?q}u|kXWyOM_y(9`tIuQ#ZudP{d}=xU*BZp-a7vAG2Je=
zCq_PLI1iqtOVpb7!{utqL=#XN^wtz@j{-{#fl!1_@nl|^v<TeCPJ_>~2||<g9~^j@
zmk~Ze6a+FbGclI|*#{Gsp&mRZm)oU0sek!({o|YO&g5w9<-HB7mXa3M#hvp;z@Euc
z)@q^2;_eoj4gNL(S`1rhRr=KU;VSF7ybr2+f9r&jcJ{;#OQyT<s`&rLjCc0qbvs@k
zS2=%XlQ#yt?ylKMgCREyW7fk$S_<i?G{(Adp?Z8#y7{1t1M-*5mGjcH6~UVw0)H%g
zbmQG_r7ax4;TU6u3BHg9JgO^DmV{TOkhx>;B(;KU2(**J;1#shyN@vB`0DG5X^w<;
zy3{hGj}fNw7}_z0Q+(JW#!MMonsB93+`?%fMOv8<W`q2-BF5uzcRN9Vtze}3BcyUT
zyU=Pw6xRNqptUfD-D>SM(bd9`d4G9Lv|48c8UIYQ=F)IX4_XVMuY}e@ONrKHd}f!i
zB{5wuV>Y0a$>5GrJH#yk_XxEI-)(qd^1O*c%jfb$E-fi`c|I5VTAt6f9C=<HT$kK8
z!ha<SV}AA$f+ZtjJiu{E1G~8EQVY%N-*nt`XHboF=0G*L!|k{SDu1<L6o2g9>Z+=5
zlo)s@PCLEL?l`cRU2x}tBBoF82JcLNne4AP-u*tk2E8{K9bDdfzl$O|;zcdt6@?6D
zXRH^|r%Gl2FYx(!k^=v%NdaCFp~<9v`{w*YIk*3OURCG;t`LtDs!5?;m-0fPkjkW}
zZewu577Kgz@ZtT#r;q0s<bU_?e)=oym2v-oxBc74yYmaL_8%S}s)kSq*b`~RU%d3?
zI#!L+GKxU9a9mX*LM$lSQX`p3Dy&LT5JACV-U=!GBxUBfVtdE!bg-F}k-~9sI7pPx
zzkjSN(oXo`cE(|E)@hg8|Ff<IA}#_w#tBl1H}>x))P!{r?19SXB7fgDd~KB=S2*r-
z-7*`52A>!dh{hL|ZLHs+NKz(u^q=aBaZxzQu|XCYN+RmAZW$858lB5L_Cb{tN_)Kx
z)%8T3p}G{84XDCMuS<;wsL~cxC4+7$^;_LGAV^}z$wdBK*MfaCx<boSeK}Sm5Mg^c
z8-OsvL;yl#Sy0%Sd4JvsjjhON4SIB4nFlBsUJ>d<Gj4*d%!7!=W^NE$@}Vp9fNRPH
zT6qWJjizv0Bkv-*Pe>?axei@v7k$dqrI+hzXCzj_TipO2)0K8cX+h+3naAq#4N1W}
z)pGUY&RB<=(Wu!_z?N{c)l@JB7CL=Onoh*PfpX^A#E^-rE`MSuG|ri{p@;xF#6w5a
zWGWojCL0@)7Rn+mq(p6)guZX=dW4!UKu-{s{4&VLUy=Aq>`ELEjz;00i8Qw?FC2oH
zt+<gopd0VlNhBoc!m@S0Iwo8?@~dhPfkMsaAccz2xz1e+P(ctfndI4*Z_fUEb`)u}
zx(bE)Pw&ru_<vFEe#VF2p%Fn~cAtl@zDJukoKwl&ue-mWeV-2Dwi(|8Y=aT1Yg+-~
z!^)*VV^eVuAj%}3a^P?kiozvBeit1u8Y2qWGG`iefaEk@<x-t=z^f1T*>tcJ@1%oV
zV@FBLbpm=Z;YsywteS6y05}I|h2D~udq0k@>a1u{Tz{Zm)CM^+p7%B@TK2llMVj;g
z4|q0=%X8aD57_ijma=Hf*db3q)XEUWJG_L2R1y#kvI69L5^%aHigq;GeHn<-*xFpA
zm4gsbg_fsA<%9sUmL-xJHG9xQ-Ob$TtnXvLs6=2^Tgudnhau^CpIK%`U-h7rK!tOZ
zQw_R6)qfC4<x-t=L5s5U8FaA}?`taxjR(5e$Vq}C8s4Mr%KM36X!tOR{G+af+iVRb
z5>A=gvR2|N!Ta@6V$OTD?f_#1TFmA8$HQMfPt(&QgN;=o^DWmD+Mt6pv-j$!C)(>F
z>os{n<8z@F)dUnentftRU*ufl{BW+V34saW3x9FVbW_D!cwZCg4Hgk|kp|ZUlUQIW
z(%>49fMzX)`*ppEK=G<e$anGpaVb*fN{a>%ZE~k1<wOG%(1^+STpGy4`)HuhI4`YE
z1B4RPJ3562C_3T2K7|Je-Lkdj1RlsF`gmaJfM@W41R$;J<$*D;h&yx>PsyuHq%W^B
zk$<Oc>LBz9Uy3+Jq9d@E@%9v=Bah8GsJ`Qhkp&9p_pv~R_)}P5D^=<&a6(SbEMS35
zQqN|ArFbU`<QjWf09}Cf<|GzC7htE-sFMW@KNxeTu)tE{+lJHdT@zMZUO)bH0G`RE
zTD7<lgjO;WY1QHo(S??$#$iMpj@rxRG=G?oPuo>)+X++%`Y;k0LSe34=%vEw(LynX
zqzJCdg?!`~Ow2{<E;I@vs?f4Vg_AYt*c?3ZMaa-d#LvNAgQ_odReh=NKqUCIB~lL=
z8WesscqMm#gA9==WQxS^BSXW@joPG)jo9(GQ45{GO*0#!fso2Qn+;3xMrtdxmwzFb
znjgjTlg@7%&r-1WYPVu^RjE$T7c;2_pMZiq6^1tWguxwii58uxlorKmCeLGgTa`l6
zNuNm#PwcPjg^ofUU^jJbY+$Gwl+SfF9%XB6$Tk!qdVPN3_Yc1!!gBxi)4TtjUud^~
zczJ$d!~Su6#{J<xPs3$?woUwOyMHfVVgQRDA)=8lmK$1Iczyy=eb|B^dKkx>XD#N5
zhv&^(q#23Dc%PhSOv)LI;U3qu5RgW0@|JNhwE=hLP(EB*O4QL#eEJ`99UMQU_t$k*
z-U$e0f_2YRP$0L$@$!BiRk=BYU>N<p-55g|_uFXzh3C@L)if^3L7RiY!+$Mz+ruxj
z<mQOhe5Y(sD@h%OwdQZR<-FD~P@Be{fCL?QgR~AAB|$He(&*FbC$;gX5xpu(p;XAG
zNsmWoS7-9ZF=2%o$Vamrpqkfzn}(@|)%i43HLZRgbsE!P^7=5GnGF-AvBR6ikYF~8
zIT}69ZklOQEL|&SJ<|kax_{+31gWPXNPQSSnyjPsv8Eb2hw3h<!Y?=XGa2yYr!`k;
zI*(nGg*ZC)Grzqm{>mz-IezF3Y<lR#A8-NYxP2uN(QJQw*!_Z{skmz~`X`qo;R9Nv
z<74a$E$BRKfVMC_Dr!uYp61l+DSPb8Br16j707UMbc>O@!#@M!vwthE54a@lu|?^^
zR&~EQT#9OLs3D5Qtw#PZ<adfi9!OBWv`DR6!)NtgmDY5@{bb8D#}!RVmt_d0@Z~D_
z*8l@OmPU<#R}yF)l}*BuB%E<y4TJTmW&*RK)GCY^A|S_Pz8{syg6-Y<{^3_gAXAwz
z7(7N722T6ia<@6`U4Q8_nYjCC7aFnCn6E~4r}g%D_0|+-8uDETcMM_j*eyE5r4W}3
zHdX(A|NDy<FB-F$T1Yxl+Zx4;!U07~q1GfODncJJiS<d$w~suFSx#bZ6g|NX4^T^=
zsYXiy++|VP|J-QG{BZoU;iHWWOmWMtb1-C^YOA#%ZUs*S1b<Jyc+@OQq2HThl^WcE
z`Y{|$@@$+V>GaE{g^sbQ_%y<<-Jn@_8szu^DisTBJ-+h8smM~|Z!k}X{fE>PKQQ*f
zp);zV1!$h>qGPOG){iWP^ZIxK+6?EkHm9MDPbz{xLfc5;##fPM(8OvPPD2_GyQ+|B
z9K-J#h%+Gwu79hSh^vRvL@m&1uOSZnHc?&mjLItXN1ToerS&*~>;dz*2~%1ZGQ+^p
zQs~{|G1Do@OxMS94#*3hP79CIW|$=G$DQLT8BM3*&M_BAU*Ha1<Z=LN{+*i>I?&uC
ztu~ncsKGpJ^iyYl)GWF~tHl-6^4BCoA#i93(X%*~zB(Ptj6;IVH}jbp2d<tJ%c9S%
z|M)RIJP*a%S8FME^z)+a=7@$--2A4dvzbfH-|Fv|&>Z~%2nscamodgY6ap|bmr;`!
z6PHYgCn%Ttu{;$5&`g&RvOFt)^!D-edeLTb!ZSHFef%;J9!UHpi9ySgBNA5A^znK6
z=^H7P#pewlBc;~(IQ;hb=F^KNj~Rp+gi)f?i!+1~jS-DGit#P+QTXl3*rlwuSnsWu
zaff$&0zT){&mZ4E$-?aGh5aI@N(;7Ovd#$bQ#p-P8I%~iGbkJUZ#<|kn8+nlerjAA
zjdWfwo;LSqE3VtCb6l7=i9*HS#^jBoQqs6CEY!K2m5|tZFT~H6>asj3e@*hB?iF<y
z7RI|_wNznK0AeK^jf<iZm@-+R*n2feVes*q_4F$kxgGssgEgy!wtA#8`&{%Z>P^SR
zbV+5^EcVu<s(KT(<~r<}GC&YBGvrRBq8Fy!Bh|{0i~i|u{4<*KkdJa8!QTFVopfL#
zs+0Mo@(zW8S*%?zW$Oh}f5aDhX}|iQmkx`*l9vutMlW5o%j|_^?AvTdD?6oJm9rvd
zRV~D26;@RH5Lv4kI!d>)hT<BwL}E!~agGV)xoLb7zfJsoIaRMJ0kS8$G7bW$k)a2h
zSRqXZX?Mk@t2wJlQ!1zt61fVC@<~J(3MhEd@FR<F8>GsdE<X98REeybER+|)QCCfF
z3+7^M7I47GYvf!j-^}}4mNnHSmt{_Jx~n^zV%AKlB<C*mWb>YzWI=rtux)a;sfNsB
zW3JB`FmGAY`c0zS&2rk12!RIK&zEJiJYj$BnyZA7)Adhs+-{_CSfva`LeCyuR!fJ{
z4yMtAIBP81`}wg5Zyxrq79M$VB$OOgd&q<D=1Y?n4$18ZUjXX@#u)f2dG&^;^8ZG-
zEc@UdUNHt7OZVk2w_VP?6)^=4RFZ~eiOox)SaBd#_<VX%q*AJgzoG|sASTo*bxMCh
zpZ4=*zczhm%vg3vtI94JeLk$@XtF{&Kg#+z?3;o(SE)ml)({d6(W3IJlGVe+HyTvP
z?uoCti0*nE8XZd8282|Q6~~L{+@SFrEOtmG3|L^*jY>*{-oyryGLX-;X~;!2^|0CJ
z{s(`|JERd?JNJaBe2xBjw1Mgw$`^l6;i|fM=6kodssH)hI*^Tgy_F|J{H^Igdi1-r
zmN(~IkZe+E@r|ZTZ>u^lq03Ir_aPJ5QETah!_AfF^<vc&kn@x^T{KE9l$|snKAYV4
zJ^$-dS;5OI0(^t}C4WLjE~`-VjR>A}AN3IkpfW;7fPJ~+eb(EA(4{#HX$OC3&e^@5
z|B;q-2t=z8fkK*RF%Jmjr=R8Y7d-qvg3I8if2XOR!MiX4zy3b`2sDy##)?|%o~@Ao
zsS^4LWz}ilK@+VF*u@HD6V40h)28Y*dTo6d2?Nb4)6r1dG)#a3?jyo1Bs!s%NHx?-
z^(rx-4f4RXou2c+9J|%ShlqdU&exk<5j)Bn#VHNf2<l_Eo;WPKQEc`ZwZ7%<Ic#c8
zA)%3`e#vi}&Z=O2;M{Hnf`Y56-QptH)AGQY+|)H%La?SsUDpG)q}{hU?^X+xyt<nU
ztZ)vIji$*~)9)k$0J$i3yQ0YjYr&<n*zKy{KD}su{{5#mxgzZfYaxGhAYl~uc3AOV
z$;dMrOIUEJNzdkp3>}ffGIXnU0t-1s^Q1hsE9ca^8A1zB;VvRbDonbkvX>(?^0mH;
z3et*&Pj^)GHWiKmVn<s>qyO$uc2N$_Ky!tik0e}&G&+x;bIQ5PF!cbV!vqE9)tt0?
z6u22dN+W`zS2VoN4!eJL&V!Z_8F~<}LfXt=d(hg5Rl&{<QvnK;y#nX3Ube3a=u3h7
zlrsX!u#cjZ(*|h(>(z}M6|6v+>YM)XEJ7-NA3@z*AdX2F0R{PG;Z`_xPAGS)@xDx%
z0{Jedm$?Z#w@%}3orV|PY~`G61xUsOAB^59qCFgrnkpKbDVTqzWu(q)%CY5cF^K33
z{1E)p(lKNRuYu0HB?mGonx<HD<M9I>%JB+7FCddrn7fmbNqb;OUU(Dd#ZwTq1q@LF
zls5x1T&Wgy(**a{K`C@$T$%>(u|Yb4<(~2~T9~&)nPwVG+iC&PYOqLOV^xS4geM9L
z#aJKFzY0P4;IMxL*^XRWW7;XP{8EtMc`SCD-qSJC_7%%YLGx}0?sv)^I+1u!e22hM
z^GuNna*Yg!sAz&+?^3Hq@$feERCbE0A)%9+8!9XIbryBk++|-6<gERseI2la9bv|{
zeOpEj+qYeG(7x@GJ!FRFT}3JgMtuyKDdB>-i-`8Zc;9~!*#l-M1?+nwE7oOT%W_3`
z)Vfv+V`$U5Y}4(@k!HPGw`Jt8b=yS;t=ld+Xk7*v(7Kmqg|BMfUK*A=PGQlmgnNt;
z1DIL&ywNsqvGY^9r}DaO93}xM0F!&P^t@iQ;+E2;c?$C$Qnl)yExQJ~+zznlf&`x@
zz+>&rkrsb^cENVZZVz5vuw5jNxnR5GSPzrFyEET%6j~aM(~FU&y-1|y4su4M`NB>D
znHpv5^<BCTjq7<hWXU55NxxMbJI}irX^$k!oZW-?d2M$b-}2JDz*I*If-Nh@?x4s#
zA8XcY_tG(xJ$yMrek(_`=6V)haeC2bf0kd@n@)c_<+YYFQjT2n`q>WWMa)j|)idJx
zFlO~7fZ-LHW&a|JXa&4&cim|sQ(z-H2Sy(KK8TAKC3d9pz?TKKPQ5JCNI8apGV|S+
z8cnH`xtj7}uT1!edU;n1Uvz;}Z*Q(|i+tGyjsbJp-zdx+d?6t|%gWWmSwvPT)cvsg
zFdu*4JW+e<RF~$;$2NVjxw`n?%S0twQCTd9`&2SF-rOX>^Yt95u&`Rqq)LUiA-!`*
zp`=ISbi?e`65Hq3EDRd6mfQ{~)F|)hz5}S*;r3#M!Fb)_2L7n3Vf!sZA74ySU+?`_
zlbYVIbyXjGK{pV&^99%X4fOS2911Q$3T2msH7^hYF)}ch!J!Hh12Ql)mqENWD3^-6
zJfD97*W>*+&!64MY;3uoEYwDL$Afh&U`Mv_QJPBXqob$p{MZQgu93#vlJp`fMhQo)
z5W`?YqgtVi_iB)Wad~a@@G}gVkAA*lnpHwu-D){O7vUk?AD|h~8H@qh@iSmGX0(Xy
zC=<t-qlKgU((`m5J;T1i%J{B9N(*DdU^#yXPkb`S-IIzgPj8P}i$xC3WcOy&8TGNt
z-`EWA!pw^zO(M`a=q59r&B#n<?1~NRPV`P-tkn|nGUi^5z+??4yMg&%DzIi@hy`Q4
z-y4RNv{YeJ7}i493d35<EDVolh~gD90zaM|WZsTrR(4nO;uJHiCXjGp@9M6l5lDYq
zU~Po11Z$(Eg;krpQe7R1F~(y;O`lJzIghKn7TkRa;`pzcKIZp%!5I;x+3ST$daD;|
zp=)`e)-v+~ouM1y%NeG4Bw7`!+;)Oi7>^PMTePv>;%<Xfqov(%KA+^2oCl3T@x`37
zb*4BnJC^Otp>6p#$Z<^7;yu~^m?M8ikN>NuJWqEH9S-C3oFk5A8F!4dhe?{GI^vSi
zr7UrA<TzFWJ;8b`rj4(hAR7rn5iw<5M5~{6<$vu^#-wksc{oN2LM|6J!*50fQPDW{
z9DKn>tmEy|=D)YkH~*gbkOK<^Z~y(>P5gq%2<yV26lxC-KYurTfAhEYt<8VxlvBEv
z;#pd2OkFvxFsR-@0=4a2qG#alMYi^H&gn9;w~+4-`Tls$^@8Pa{*yw#8kvMV7O!A7
z?F8t!Sa_8qq2z4J{eU<x&llJPAVlzCW)pSo(b5RTdS=h2weWJ=*a;VN>o(hvQzfgJ
zZP5nPu`B<-vbg#Ydx_YwWx;>;s?oW+^p#>^q;Bj;I%Qt!h0YX4UGyV6C{l0Exf7}|
zk`#x`a*yMc?0h)8mmf~DQ+}DHFqX1TvY>~?$k($jGo@Fqym!h3PW8yNmNUOv{bcP?
zq)qG44vE?4;N;L1;<88{N}wU&P0Jp#B31JINKs#pQy49@pXr_v<~V=p$2?JW<`t*f
z`;-^qaTfweK}d_-T5+L!Vbl*==66;N0V`cu?(w^c@ZmqV<N5>l=t%&gU}}l4v6hRF
z?+V2jb6B)m3#&#pbogr#L4zWK)4nc;Xq{u`W|Vj`VH5zJKw`gnpG0PlAY#alUEtxz
z{8s1tZ*_?{R3e{Hv80btkV2Y&3gE+U>$V3c2(*xqt5TJp8b)1Ml~gdO@!~0{blKDB
z!>cG#2k6r-)zjh>xoV{X1`e#1H5Q3(Ck9ghF6k!u2ILCl4BR6lnoO)>PN!EZFDz4J
zPG5Z^hkxM9&mw6Ke)ugt`fl(bQ30QR8omIFO8jWaTqOR0yrp7uF&4mo!><?2^}@L5
z2Q+HrYd<VNi3s)NW>A(8Sz575D63lU)hTN&#*_(YC4CiR|6BJSmX!6nA7lNctDF?h
zE-qc?x<VctE$KN6$#pQi6>1jB^0*EFdTA9ycyl6B0g$N|CL(l$iQ&B<5h3vlXI2x@
z>5=V;NXIJYUY3Y-%=AiszaU67!?2X@Izh6~S}_ZtIQL#vaT8-(3ma__1fCo37`f_F
zG$eS~H$qabqk@zkq<aY}9vLL4g{9M1f|w>*TlVqo9a(#!6ljzJ8`;cuU7UIr+tc;y
zsxD5!y)du7o0t%WDRL_}(W;vm;jScALTY_Vs28>tczb(i+~L1}|N7_q@819N<F`NE
z-YI!_y1jGifT7h8H4=>`g(gIE-*jmQJfh2dZm;_rGIyH&Z1a25#&--Z!}LcsNJb!2
zw1%Vx(dU;J-g@dUF0xCF=3dj@z42Ol;}K0GbfxXXM$7B<Kau(;pDnWg^+44ci8y?K
zqZ$qNNJ0GxBDOYvq@3oceWy8*Q3Sg;Vr%J*r#X$#l_Iv$(mPZ%rURLI^^+attdMT+
z$xbc3@l2r+x)#K>mMiuDuW`Ph3}G7lJI)srcx|S&f70Uxnp7*9erS-=s-7I7O~sG<
zjufmxM__9>u63e7fd$L(fld^ForqCXTs*^b5127|BtP_jP7(H;9-!eSr(0v!P7fGN
zQ`=VJ6;BV8LyJjnW>bHM9$hjxInC_3(&<5TF4*rj+q6a_IGe}^KRk&0fL5#5jl^|s
zL9uJst>&%g1>+B*rsoB@3*K;EkY#tD7hGgG@lHPM-141DAAY&8=NpTRszs)IEqz6)
znEF;q>-8OfTI-R?;v_YSUTzj?wTXqxMRakk$E<G~5K27D8pOkw>v}wp<C&iItw91?
z^VIliVQa9zBPoN>Y~i=K=ju-2^Xl{zdUQYtDH5t|bQLe7%ymVyaJ59Vl-Bp{v<Fw=
zT2Fz|+V0LJJ7Lkj6Eg@|p3wdJY$!mBK^_4H))g&hcZ4i(X;us>7pM}grsbFRW3@6f
z(9Qamq;(){G+QTbeTOxkN!VsVmf&6Q5lJG5&DHiYJbcUU56ITF;@`P-fy}QJ@BLNZ
zRCr^LQp;THH+_|o#a}&b_0{Dcn3X#S?f(IIKo#egfx|o&0W_B(KRF`}F)=VNOl59o
zbZ8(lIF(@vCx6XbYmZyE5&gcuf<V8dA58Hv53&dX<WaW)9MrH=pslem)@#MtbS-Un
z<<$B2J;N(elC^U7aaY|S(ax3Tl0(jL&KZ)4vF74}i?I^9m~)fZCE+HC%T=4Cj`N^b
zoCr0kI;~<72Uk;VQp9ITO~hx8d}JcpC2B7T6j)p90)GXG?OQGox%H&@qEJ&%d|SDR
zOfCg!3iu+$I`k$k<vjGJE|FyG4Ynt)LvK(hc@UvchZdx`977w8RF~7Bpuy!lC}?z<
z2L;6xD4D3}II9l2>T-!{y=jKX&74vk1#NR!v!hs=sL4?pMxoVFhL-6DN1S34qa&53
zi<zS`%zr6f<}gfZA;(FB^;Uyp3T?zSLZsG0O%O}7tiVxFb}@y378DZ)G2*o{c&)Hw
zn->$sNf>4l(<K;v6WIw#ZM{K3!Z09pqPdNtA_9uEQQ&2bXxw@WP7~!OMyJ_)43V9R
z4HKe3<l$R|h-OSI4s#lhOiU0<Yav=_X?rb^D}SLhF-BM7&_ztHB>2X}><We2MCdNN
zi0q1%)<qOoYHSLyq()KeEhWTkXf31+k%ktEMRGC`;f@8+W)kZH0oKH*4lz75#;z(i
zbCNiCe`pM9Aq*N8n)r>65EjM6I=qO_kg4@zksesu_hdSvgO15MjN9Tf)Y!(=T+lQj
zn}5GzY15xAaC00iH{UR~VP<o$)YeW5<C0oL41vSV<iV0M*kGefm=R2=DJB`rG&zT6
zf`uxks}0=3Ahi&Ij(uw4qc6Xl{P^3&bn^9VHeXC$et5I6r)Ougf4j*y^UJr>%U1zO
zCmc`yI{AL`?W;IEyqKOYUZudkVBRDYFMqc9Bzvq1iN$j&$CIxeT7NnD%lyar<U998
z<cL)*6$X+JDW$}J{0jfv%LpVZk9goDUg7&h?9I`f$rqwsBQzp0z*5c8YzVCPw*e``
z+Y(>ekW27b2YYRR&ycGGD-bIt<^ZRTHV?$r+uBN{c=X~(>@V-mLr@{)8qFnR?SExX
ziAI=+T09|*t*pozeTFyoTah#>=7F_fxn@b4$D0bu-co#0@>o}v!_AmocPKJ(!oL~5
zpaO$rgqbMhlflW-h&~g0j8J6hRAZ(h!h$)Ja*Yw!_ls2F;%e+Ec~(QT>Z>Iq^jDFf
z2(d|-kB_#-G5NYTf4VrE&tTBsUw?dW|9$cEVsUZxr^)2q+2ZF9Z~W=}{p6Pq?=POb
zo&Gxc;bigi-!2y4&lc0mpH5Dv{+IK~oAddb$*<9;;N#@#@^tde<=NYJ)2qq1^V!wn
z^5kr`n7mnjbu~YIy_mnA&1a|c*{{=&umAJm<SiQc!?P&z<Qc#hV?Vlg6o0@HU`-P8
zh84)|-q6g#0sP>gs}hMZtpicvq8JYwhl2%~)ki$-vEZ}I@w-<BLKSVlnEd0P|1$XV
z2v)Qe-R#5p`SCtAg&G^h@(hGqt2%_n`TX+b1tQ30d%;bfpDYklGna<PKYv{O<>lgJ
zF?DhH;^}<0KqpTd%z{N(FMm&skpfzVhX%uxhX>;UMkm^Xp(AL-@&WEJ9yC0#{Sue&
zAXE&Z@$l7GljoQ7)0fl5tI6~4o=$$8eq0=HIOO@syQ#(ZVmez~HPXnfjTh6a`G?EX
z>G0iP%Ma7HXD8pxKQ>nhz+Y=RMhlmC*Pxz<?wgOVmW;ZYaP(^6)PFYqfl{v|>5{G_
z>Ay<<Ro1_jF4dK!p-atz<5A)r*-lYDtRh}Y^q?^Z4*8`KtJJ&Ch?}L}J5B^B5-3}r
zhe%#{ml_Gub7oOg4i&5!WkVft9tIoGCf5F56K<CNZVo`0M$bmcmjJ{JAe8br0Lbvh
z5C}ljU`L-kqcQ^KfPdp)kRZavYokFJGx$FkmX{z0V#LI;^^Sn@RjeL0kK<VcR@5CU
zP=cb^LGP$oQPiP|7=sCh=p%}oOoqC(tpLlah;0~?uYsNla6$~G!del1g~lij1WR8A
z%y0V=a7aXwP6duh*H|IQM3E8-K9*E-;xeKUK*BZ!IRgpBrGJQWA?=n2YP?)hk&L8d
zAC>R55+m|Jwi0l;5@SVqQXdx;yto?pAH^rnMAj>@u?eNlCV)}b@r@0Q72jY49&r37
zs?nk$KLj(`A|qQgWs7obF_R7cfnVB#EFTavDK6hhmS%`|c+RakOR2k@rRd!_OU)ac
zMN?-Moo$rPHh)U&Qr9cFOEdx(S*|<D5m1cvSH#<9DV7wFNe)y1ER6uM@JOzG@iQ_K
z2|xJaXG&gd(`W_51QtN_C>MX^jN(tz7{%Xm+70uVjff_1t1_!#_-QxsmqKFblQ@;w
zcJ{{eeGW!Ifj+r-+r7;(;O{OF6ULrZ14t`}iq#e6UVpv^yY>l{@5=XpRshc9aQ$q4
z`Tq6U?5DHY+2Xeq2W#wSEZ4`7J-An*;)z=7^%}3?y$a$L$(VmK-l3&#m%OFisN~&J
zTR{*K57(QhCNM<XFFt5>ydQ(MkY>Ax-mII|PI@AL9=C1@NNAlt1*CS~@cV$2jA(8G
zQaemp1AkKK#c}DyaqWe5?Z4Lbuf24xy>zZiFk4N-@)Tt}xYF%xA>}aHLOA%eK@+he
zyjFTd>0&;c_DmroW(u+tVSDDVBp^r#Bd4-k9Y_@=t&A!H6&q(P<PkvvEM1bWL84~M
z4A<0*sNM^J3TwLL6pHA0rzlyf2U^*gY}-Ay!hf|=J<z_5V)|?BMr72j0pxI>5vW*E
z+D3RJu|^RJ_gerNVStR$7l4(j8Xgl~8dsYCcY3*61v!j~m&h|2#Q=(5&}4hIvz*%j
z=KYO`3Yr}IQjCdxp*ya`o4dZUj~j<25rj=1%VA0L{ME(u^z7vP_36pgv=_=;w+iKU
zi+|v=h4Plhih2kt0s<wufodl88R&Qe)k;36`rtM7XEyk^$j#Pn|7K0S!H#ZCRF++B
z5tUf(MpP1YqS7u;EQv)-UCLdeE^(J+JOaX*eKO{}+$>|tK1V(5W|^HB+qIL9$tc`Z
z^uaR$k!_Z(k4@S3%Hm_r1a@LtXha~e1AkM_=704yEDtYIJ#5Vkp6z*!1Gaa1-MPug
z-bc8@6YJfQSRyDue$4K#0D>4kMS^&fbJ|bTmglsDp;x|%YE)3vp#R4tzkf@0V|(=0
zc|FJ7yuR$@)_FbW&AhIwe4TpcPSPbEbI&t)F)}9!RfSzZuoJ0k?qn5Y?NS0N3V+m)
zfeD0r%uP@99<}ejG5#jrJ%OdHNH`J#b55WQb&N|{-3Ss)LF$DSvuy+GgV)os*d3^i
zQx|S`E*?cHc{iQEpPtODptzi#BX)a9LHbE1$>H}n<%Y)kL`C6vel(Xowz%Ndk;>L~
z*5(eDUE!=tHg~UVLb97pi0+n6sDEr^lT=ojBX>radtS*sujHO}^7@UwiA8-Ai#qG1
zH16q(2HMC(6%}4ScqX!Ip&LoS4utDrH&^WF1&nEZMU_Euaz5-8)uj#n7;29VV<$sS
zmKfF*wq1?Q!jCQ+Zf&omF|LH<$ZZ#kD0-{%D2mTSt+Eg~6e3f$(~sJ%V1L$z%=@f(
zn6psSa7M1}x_F7Lz7)H|ZZ`u}?iho&z7)l6B34FF3~eIz8Sx+!K`1L2n6fegPDaH|
zdyV+LdztNjO}Gi)-%?v1QngF#bl{Vxz38mfa#OtqkneuSAqBbSNJ;K^q$HadOnp~|
zI-+t%RPKn%eIJMW8{AQwJ0NOvM{Vw?%^kIeYqet?Ga(b_H7xHrW!jsL{s)>q2kDnF
z#yu1PFqaWt9~75J$v!BTXU{yNe}@m}-QVy2aK*E+FAn5JEtQIq@tv`n!JcBNvO6Jz
zK^mqKRO>wss3Hqh+Tsog956=Pe8)tEFAh%3Dk8(Ak-eh)mY{=pa+g$kqYi)DtRHsh
zLxLgxzRjAWU0L&ZFeHK?NVS^d+i52Q$Wfb4+LA8-@cRnD-Va60S=|{Wf0^bbv^Sg@
zxF=B1#I&=GG?#wX13{8m(Fz3!NF=pPl5?`-Oyh70-hP51r&nLkm}VqXMzvac&IMkH
z#M>A!lP$3M>=Se$fDuAI2vUGI+`rw4%$#1-rDYLcDe^1qAjUdh->3NJSQnf*BZs8`
z_Y@gta5kRH<oxKJ*MmqFe@1f0NEd&p7~`@L=0s_`$^kMXRUy`s=_a8UC>?&w+q`fJ
z(I_LMi>8EfP4kFazXDBuWPlpAD`lW22CZFXA_XEZvIOjLv0t#r5O(IgZ?M=OLS_;+
zXB5&w>Fsfai$Y&-0=D7wL-Xs^-POmFkP#uTdfB~we-&<MLCjd+e@V%_(A}r|?zgMI
zv~MkpY0fc=#}&biA*xfBVT80vWbqC+nX<&WNaEV7ga}^8Lss+oyw7LQ#_)`d=H2n1
z$n5IMfYDM5rloBS7Kh1fCP_2lTY*Wd&G{xd#XNFXYmyi@2kep3oY=F9>Sx=ARK~1b
zX_`Xb$HskPS}=h|f6U6iXO_}9KbQ7T0?Cz=UyqFQdnY$wlxej_(0N|Rhe5)3$Gk(P
z6g4FgLRU@(A?3x|;gF8#!TuObz3*j*JG3vp?&4db1unlu54NIAm4zK#d&Lx&lRG#{
z1&`QBqsMmeq4ehII)RAIU`zzr%8{%^enmKnlpB?b`!TW)e-A88`_tynMFjga)|GN~
z!m^S%7d*qRCsY!7C?ezT*szC*);6W)6=78q1x4_TyIK*J)R-b?fcl9d;`8Rg6RW9~
z`gvZF_sqIeMdEsi=!jE8ifSVzxtoiZ!~!)S2B9iY)CXD5`X`lU!8^}Xk@@j@D}*t<
zm&S}mKpGM9e+F@Ed0B9`pb3c>D;eCyN-jWdg@&r1W{D^*95beMjAl76UD}~r&fD48
zNiUfMpDHOnA2?uV9GwKN)QxsGr%(xC+b&Hx3#5`jUg>iZfRfnOP#jGrfN@32F((@C
zKYf}75Um*#PM^8T6cQ9NX)X2RydrYUYdf;awo=N?fA8k?q1p7_o#z=+jxzRK&%kz6
zMYYsV^NKv7O*Cib4+R<ueUd+NF|pKrYXd+e^_%*54}!+qoiyI7nr<CvLEpG$sFlV7
z$>mN+4GgfW+HDO_r1KZH+qE>x>ox6mBlJw|cBAE_-99H)RUduU_$SaebPmrx{!w9H
zpl@Bie>PgbaP*_l0$6|P=%<!mIr?dYo{5W%mX?d5JB<NoA)_?2i8X%ScT<CV>|&d)
zd(zzkv6f8a3ibksD{0uh3gTMmxgf5!{Qpna(8d8M@de{G>G1}w{+GvV;l9q9OQ!}k
zr<O}bX$mz9YLeRT=NhGjMnZG_jYetGST>lXf1Thq2ih^IoQ}*Q-r-ic9Y34c%cf|+
zUs%=HIyXeqC?l0=1it1F&7!3hwzX(#hL$F7jWy3SLJM(4k+;gSJa20Z`_=<AsU+iG
zU&g`tjl^LiIZ^QSXJxx>%=0JnGW&(|vK6HdnsX-A^u9bXlF9q>P<qAe2g^zsQA<62
ze|NnN(~*D2vt616*^qYBiDEH7ufqxwo@yz%FsG8t@EV`ZA3wai-YVLBkDl0V?w@Y2
zw^nXGJl_BHdP{cm5jWw;n@{QX{FDPGS?L@6$H!9>Z7mnrRp`tZC{9c$KOci%=KE;{
z#eOCR6T=TEMwY0g9v&wly@M&rZvvKffAjF>@m&P>pZLi$v};Sx5K1Dtkm5t>^%Sq9
z3cL%{HlekVCkhv+Z51}>+Q}tqBdcj_hn2`{;J*3r;puBo+oI{ProsIjYD-T{33(>=
zVQ?tgh!yM?D2*aXifQP_87(~Hu7UpHaT3Xv(fGMZ<iCj3HjLlRg&#YH(#OH{f5gWQ
zRIk34SRfdfY<wfK|Hr?clIw5YJ_S;_&3_`P-oL-kHy+;p37BxddGnNb_V779c%KD6
zMzq$kP$`4G+cY>$%>O5bsHiV$Cw@u8(1*uGZ(QpQ+zSE!oYvF*(|`@G%5TCicvk8j
z0sHu19=9)rMk*TQ%RzlJ5h&tVe=*gI`h_x~+n_NFxuez{QSIBAy5w;$>p5YzZ%`gh
zHLrv2D?hDE(aQ%jM=>nD`uBzc%g@I*>_k-i(yxUHQVAPpdyV_KJU$5!ma6Q_#?;Ef
zK56nCE>@wonG0l(&;jAe%IAf?HR9jD`&c%HtJG?Au}#JuTb}Qw3!bIDe~voMv6iOW
z{>ik4iq=IthjKkXWGQ4z0bAHWH?z^x+%O)t%Ol7O!P5wy33RW*(l|lIEK_D1rP?nQ
z{*bUH{l`A7eH#2Jdf8X#1$MU+@m*z>)^rXC^eA`Y6q$xrHeJNJ{;w<?Kd@`0FL9GW
z=sSqHuZbEKd(E5}e&A~#e~5`RO{6!!U-)XM5GmD#FU&c#;WL!bCZ45Q`WzJ+emxZb
zyzuQS^m%qDAm_jT5iWW;6!;bB%_}juh%#Xyqn1CaDYM=U=`@nF?I7odxGjIaLo4AF
zRe}qMq4x1kRuXcRsuY?tG!a&-bCrae4~{{7>r^E%h|Uz)5Yp#BSs+nHhjJ6zy!2~W
zgPkM(pPQk?B5gudt6<nrXn{9eGvu#IVqG+ASJp@K3{6HbWz9Hf<0?>0Dzc+G#uJiq
zs8Q4~ABid-<@>{Qw7(wvrt#|T>c1`1gZP&*#yu35&mk-qmw(tiH39@km$ukEHGlME
zg(I*yJ$;;jzmWVdNM1BFSpZ=aOi!Pt-`+@xhMyBZT1ut(vHZ8=Hys`nINkxg19%nS
z@L)CYisBW;dyx1o`9c2M=Cw<vH%#x0lYzsTpCdo&@W<0HcQOlGB||01_{mr;`1~T7
z=2J9FMGz#hVTI4ruE-TW5gjl{M1L%Semp1&>#n1*>03Iss}xrYr|?OiI^HHz0Y4`1
zrzDR)=XG1@!|DDyx5yz-$@D}C3&(Rft?rch&ds-UiduVpF>dn08#KDLD>^Nlr^#q7
ztWMf{CGP{tVhbAKK&LG~-zTotu5$HC7*LZ@Mp&D<!u<o^TZT6xzWOP1Fn_qlK_Lmt
zhjkqO-8@uj5XwSNg5?xiCvogMrYHHGh1@=fm<oMQrM9nQ*hyjiHtV9HXM-Rsot2MR
zC4+6rBuhrNr^09HyQRFPZ^e)>h^V&pO(5hbU!(79%3~_<>#}WY50FK)jrJZ8!0%9`
z=0)o1aa%~z!ZR>-BvnYaet$17Q0hMRpo_h_rH#E~=6VIGGaix$;bleYgoOdo4i9L|
zfHY|(B1_okGFDwOh6Fr0*Ig!|UqsFu2$w0SL9Ju@yWs4aED>W7)QPzS$OtGLkuGpQ
z&e)n{?4*YQ8^f9@J0za~0VD&BwK+S<`DUGtY@slPBJ7Fu=fjrsXn#b`k*>4Ajlj_8
zgeB9H+zbq@tS}mTs_mfF+6a)UCwcq!KIkJL1%SSR6xw=**(-;xyQFp`Bwu`R$xsq#
z8I-CZ8`X9$P@F=y&J<0Ihc-&8%d==+Q{jIranidvUB=H?T+%-yiKpGsZL(X(q9Hh)
zl})jbxl*~I&kh%z+angrLbV+-ei1VMxgnEI*$f$90zj7l@N)nZ4qeFjMUX`bU6+a5
zJR*O(qsZda>d__RH`Um(+H;3g74j;nR8?Y8uF3T0)|qN`splFr@v~Pez_>R%F+j#2
za+?`|WX!f%TdFv;%f>GE(yoS8amCEv!6tu&0+2fXP4=LyOQgoRY;wnX!j&B6P}s!*
zQJF3)Srt#pz7yJvck2y;BH^r_D6}wOGPi$|q;B`s%W4wZ*@0T3-@X9x1z;pfu(7RM
zgMy32kuvs7c9+$)<F|B`(jEY7Fi=r{Za}nMm_9n_9`kEMg6uam|8PJA<Qgpvbh};V
z*W4bUu`e`#Pc8Ad$NU4)Gt56whPFDJUkX+*7Z*|}GS2YZc!|?zzQ+1DAC|4M2E%{V
zt?S}y2@I>s%wquM4VO5x?VH9kKv!t#V}yorz<JIwx}4y1CNc%kMn5@3D<2!{LYg+*
zPRV)iX+AFE<~g#YCNGBVDmVD*W|N%<93~AblC(+6{f@1H3pO0pcf8b$5Tcdzn=Fg!
z;9hbq7R(V|?}ty1Pj~+;%&aI>YI=YA^V41ULIVlIVa3&{=T_qLU(?6CUq`Ol-s3w?
zhN8<#V~Bb<Fx+!GZdI5)&6A;p{$Hy6q~ulokILt@$HZ2KL6XmJ{?5p(L(Njxe`3^h
zD^0W@1F5$#UJNc*p$6HkmMuT3;+m}H*)pl1Y@mfquqqELOmG`8inFh<eYSsSRtW`q
zuV~h?k1oAN=gHhCSN^<_Ox7LwRuWU_Bffu6uib<UG)izf83onlV3pf+*9cK6dUL&`
z3<w-^{v|zS$XOgfA8Ya1rd(g^_rzD`e1BQ%q;95UB}ssFS5oqrg8SW#sb*Q0=Hgq(
zh^M7qH2p5}Su2uCq3vc~sUm;kuBvoT=h~xgWpOq&ck?hx9HcQ4W2;dAf^F*uL~%HB
zs=aCF0i_M8cU~U{m%ElBcKoDutAh=<en@^@cmJPyC22So+^CJvZ6@79r`x=~S*T^-
z3Il6IRV~%DaHV738nIWH8{q|w<7C^_$f?6^Tfi89rcO1aKyqwz23UW7oFbKgcd`0m
z1Kv3GORpB^m!DhCYc<!h;8MP+<;ygIX2juJ!&gPk=vwhdqJ5}2!~D8^m1q0XsD_>W
zGoLHG!*A9<qf&w)^~cXWgP>p7hof}DYItEI)>H3n#0H|K5&J5^mYcTWltr-ZvF}wl
zJ(L`*R<ShqROb;-`*MFT(zn5;4`qrU-QltlSlclIP7AH@#nXmefdYoB-XA|qIx)+%
z0$(<*0F46tOM|4R-Z`uqh`v#f45y4iqBiS^Ti!o=^X>rnGVkYp-nm*fC>qP;TP5;g
z&o6mC{ubGcg62}*eo=v5i;#0^7TOtj>cQ%2<GabXvj%DY>7p<BpB3lIo6Ay@r{iUY
z-8L;(3MYp`g1vs}w)!_))&JZZtW|A&5SI7j)7^hZyd#%^3JwzlF)}bYmr)4<69O<e
zm%*GjD3>PSJTZUkvP+~$kw?^_uI2RGB&Sbzk1u!MJZqku@P?e4UVfbj0}?+;;-xi{
zBNA5A^zwQ7>7$g&;`4|POQ|(J4!?Um&HV>W9ufoz!ds&I4~_|MG~Q^)D1)cON8xvu
z-#ur23w>|Bj59pp^MDWE|Lf)ZJDHvR=Gpowh+viF+EcgJ;5_RDe?JoRn>Rmxx~nfe
z14wA+I6aqZl^)Wu>9-;D8ryEC>V4U#fQDri%LM?vJK(o2R)fx+A?ROrP3w@Z-p72?
ztW(q0`?eexy<hq7gFUqk)U)~(Z~v7v?|{mqeXg+8Ak$mt<*rZPDu226%LS=A@>hp#
zO9JM7ah(T*+eB$Zf0mKKnt5|EpMHmDu*6XXAgmH`7(fP{mTINdF@{V2QtPi2%0$8o
znB(JP8=v66dV6q4reoNz(!AmqJ-GY~q*T-Ft#%biZ^tBda>UeIqb2^R!ABdx#M`i@
zJZ~K2A2cVvO!Mf>bPn@d=GQm|bodiCT1-G|Jb}SyUa^onf3tvLx4e_aWR53I3N7;z
z;74h?^y~zJO}YtbwQ)EdE`ZRQIqH=L_7&kQLi62gmVI@M2IalrTe849RQY1VR3}*c
zf`i@b5D~xXv&Q{7i0F)`Qc+h=u2|+dA{5*5uLIW+(Q1RXj3A;@`hzRif}YOG8!Ojq
z8juxgKu@i7e{X0wl#CpT|DFv;@hIFpT5d0{?UdQe<ITcS3r2U>hU~ri5kfnTqVr|i
z5prmL>}mD69`BO<UE2Q+t%?h%39et%QL8dH5}m>y*r{mbO<7vGw9^j6tl*UD{duQW
z#Rh?lj{4@b&_H1C+m~)t;$ahP_~M8?K2Fgj+d#c@f1{lY%g}W)O!AG_nP@3@+L+ne
zm9}8^CWmS9ZS-<6{wI!GrR{Y`=sxKu9Mq<_k{h9|i)vCVByq!h2BR_8{3RNz|5`aD
zeSw8To7Y)CY4a!q!M0Z#>}Uv`C3$+cjs_H2^IvW|#;-p+*WTFf#Q(}r{7P=yC=WI}
zt<F0!mp$k_3j^`6#FsIAMic@vF_%!27Zd_CGM6F$IVXQwNpIvf7QXve=x85`=HpJp
zDZs9{8W<ozCh%c^B!@xZ407lWrqT2N=OaZ@B(>BMyU_<(Wr>gP@$HNF@SlMXA8wvr
zZtjm#4VF1gRtztH41^aPe~IIv(Ss#oM$+)|arotdb762j;ZmGSiOb@*!(H!pk{nhL
zRuG;N?RS4x5uQjqkys-%?i`oEZ=0SS6TAVuGmeKH4!9n1>HTjn|G43a*!!b)i<An<
zwDE(nih-U?skD0%fglA^HBc*Tsenpbpi&lZG~og<%7!-)+5OScCK7EJab1v@yytkN
zuv`3nNW1&=9{6=y8)3{ym%{N?T)PjP3M2+DMRI=+iLl3txMf)h<(RiUZ2^bYdl5Hw
zi4Qi>6NL0b!`^aIshP8CfY6!sChhjbdtt2bNol5>?V!2vn+G5*n57;lF=kUNMlfk)
z4+(9G#4_!mWdmnY8)tAgo{Ql#j<EdZ=ZXzk;IISbT;4e_E($pbH44%#Gcs9_c9xM*
zg*Ja)iQ;Lr1}!_av6I>;pHwBO4^gipMRkf~X*ASGC?%q80#TW<<Fl|<>a;d%Z`v{|
z_T?HW6*2FrJNx4?SUK5bomvHk9>(L3M@wRZB!i4_`_TRO{BrZ(nT)jZ%yTvT{_!Tf
zP^6i$eh`9rQp4xh;g6et_PZuw_4nuzqpp8(W3+BvfMFUgNCz`33Nh#1DaPU1^XNEv
zaDG1BA0yA>PiTeVi4_XS%b%#Z&67%Fg;03nJNoFjYUx5kI97H5t3;d2DKdt6-6P0S
z?y0fF1n=2<o*G|FnIy@q7UxlhxW*unp0YrJUZBdHw3=mATy>6F%LQkMu2Xy){YQV<
zQ)uTzI4LeC!S=|3+0&mloVWB?V`i9g%s#BSi%R7ydBtF7D=w%e(8bv2x~Q@e;4}^*
zR8Lq`0Js!EVb40&MTd(2#lz$k<2^gqB)miec8Dv(!47Z6PATArv_HvCnOJ7hiEO(2
zDqje+^GcaQJoh6jlNA(Dca9V;7J+~Ah=YE1n6jIFGq!2irkDB@+eG$DBC*C*c6b==
zQ1UAEDM2x^ZDh?^M=u7*;Rg&g`PCT_HoW7<9Qd*(2CW+tW{GG(Dz6Dog7${=S&0xK
z!@HPJhePq=rF!a+Com+u#XV)U$lBt@-@q~kZX*UnO(Q~dt~0TazlmnI+b@522}Cb~
z7-CM@7UWm>#{@;oP<G7Rda{qBlub?;tzCq62El@QoSNTl9uBLy)h*P&?luv5&xq!T
zC_}jcG0sZYqzZx~^>6_Z!i<t56y0ay<RT0q9solF^XYCg??+(jIiEZNoI_H$%0jbx
zWzTSo?k+u@pyHYmU24Z#_H=*yQJJ5z6V9qWVJ-U}6UN7~TZLoED}~%TdM4GZuJYy#
zrJwdYBTdfBUCNT5W~WGSDG`DQ+>)?`GK@Ruq~WM0WYJ`?sZ~Ntrr@U(iWpR|tyUSc
zC~K{lhAfpr$|e)Ye$!qA>_s7)#Zj0Ove*lphZ3W6kWZU7BRq;M<gb7A4$)>RbzbHR
z+!52<&iEoX{epx-8y@f(D6}b1cGKXbAn3Acp+>Ww(n<sfRH$%|%sF|c@e2bBQm%#i
z^4$rhkF{_Ob1Bo|sL0M~4NR{Yzs<rLdRrT1jG7(753hfHeEstIZYS*RpTGZjw<GRv
zU;q6HKW-0qJ4?5JeR?8I+io{mRPb!3>3=Mg!#k<mb^R3>3PS+6zPj4(GoT#BvzNB*
zJhCLMs?+a;COfI;eM!)xE*|m<Nv)<fyZfZ!&*QsLCrTgbIi~%WjP5)h613xfM-Njo
zJ5ruc@I&Rs4%1GTyzV?3U@WFox-&Hht!cZ^zbLg$W!Hzv_Dethkd7JGG*wngAM#Kq
z4|)1Sk1hIK-scobrNb6Om?|@$51OCU#l3c2`{9sTRN!Y7pje{T)m-SO6r(zhC$G;G
z)uL%|5NIHoE|)j&JSzd`mt*fdEPwN)9)aM@en#41c>;pcjc~SZ^;M%F#|*^W?X)!u
zf>%_+&M1AuVNjay^fawagOK2$!tLpKuW`^g;W>{s2?Z90X0(0nfe@hz(p4?~egmN%
zQ=iMNnc7+D^XMWk7zTyh7ZJQ8>YE=1LAsJovGVT~F;i;%-oqehJ|6WVcz=gs5ci|k
zIH7@VY(|&v4Rs<uOnGC=kBdqOLc(#~0BB`sLtND(rISOOm03y}+H|{R21umn2*sC7
zSH|q1-tUa*O6X>wCT*ZjO;-#vVvdMHr_$bje_~)_p@i6kz)NA;fyqjPpA_~%ge77+
z)uy*UoxlQKU^C4``o8G)&wuUN4?$DCZy>&>?KU&QwN!RPew!_3gl=yAh0;RJ7QfBt
zpvY`<LwP1$$e#uD*LM6^CQl0_^lm}Mo1-o5vD)=4i9@9MSZed2Zp=C_JZKoHal3rb
z6_O!wPcxgY+XK@CZK|v4x--|7CC$pkb}owDRK}-$M4Z}fP6V7`(toREyId&~a1<r`
zldSkKiyT2uEhv?`5Vk6cZB<H#lG)B!Fg~gJK(Ue+sU)?f46ZvNJi|2;bT6w3y5VwR
zf?lnA%=leDn{R1UC9hm%vqT0B6R6m6VsI{{W6r4tqMDnppwS4NkWCHrzfJEVH6+Y!
zC#N%=3z7wRO5f2~e1FRdMyr@2*=xpm!AClW41v}#u5bkj3F^QGXgl;MPK->8=6(~0
z;)c-*=T~r0_%g!uN1>D6Pl@1ldWlVKpiMH7DWO8?$lOUM%7l4^TIfme6Ez^;Cevud
zE$Zpa9}?Kv=LdH2^8@Q<P%EDwEV-XB*7ei358u?63Hr=06ATt72hT4z{|CX@x<Z#R
z#yu1PF_%#X7ZjI1@;u1_w6_TJJZT67{ZOYImodgY6aq0dmjV4Y6PKR*JjZ|A_?Ek0
zpMTuw47NMJBDAE~W8*w3!I2T(YbvD_5HmDdp&V?<Hs9pV@amT*@dcYvUmz)DUuoLg
z67d@O9<)ZW1;t(x`$+B)wZ6iT4|$eM`y9w`;`sY6x~UUOk3jnvJ+>}(x-jZc(7=;&
zhH9X#()JBUa0f6caAD3J(3yXlZHo3Gmaa+m)pD5J5t)|Q_0W1YF)p(Wp%t;F{2uS=
zUCX%2Wpl23WfQTYV9&v(Ler2<ZtpVOW!HTlHkCk6#inU@#LyD9CU!2diAiq@Gg@UZ
zoFmMIrtzYYLz}WB+D19#TxC)8!Tqw>0V|8Ookx$I^(usU8$W04)N+5(LPy)m>k?ec
zqSumRCubKlw5Ue!eCaI)U=>9cnV=j~hlE^af7YH7s?=I9coc#PR8+ODYT<pxZrk5o
zY{>8J!yM06Q|2!9;_KV=+-4D%A91}j?bZ*<uj;88FRbk=tJ(_N{0i<_kWt$F9PIWm
zC>PvlCAKw~2H-p7$%TK1$cNS_)~<5pb{E%p<D*l>34~F>L+j1fQ<;~+#n8mlQ}yfd
zdGpsG<m9Zv0L^b-H}M7<saifr2B=`L`SyGBviYfgYgH|DzaL6e*Uk|Y=?qR0v1kj*
zJAUnl7A$hh0@22BG!|2V*jM$q@WsWUtL_SQv;Rn0R9xw04={f#09PAl<c;R@Xtqeq
zRNydh$03YvH74mUmG=WkW*m-`buoSbXHizIhvvO7kC$R&$p)|kY`M@6aOK|<c{fhx
zg9}lYCl;k_`io)iWC;Orv9-1aor0!w&FE^E&gWbr$#q+iu4ap_TzF!2j-Zq5E*rY-
zHB#GGS-OMHu1<eSuBEpm1l()Osy^#L&UC#8$(9Q2p%(Mw&Osw}(}BaWexfh^;oqCq
ziION|BIJ~omZ}i>A&baTHzwjh8R1bO@{3s2NsVKZXgSZN)XOwBw#oqy6zOS2ESaY(
zi~gBb6r?u^{;);kevqY$?EoIoBs?&&Ln$Nt;m?~%q8opSkqH$kiZwsCxCfP5$!}S#
zRuBUam2qYbF(ii-U>I@ZttgjodOC98<0?&SC;~nxL|RD^5Mo(a!w}#gA+ReM0zvu=
z8lVHbxs~EI?+|OO3NTI;=`$t@xl{pJNcP?Kc#nH)k#h@|DxUAv-06iv9#uxzdQ}TW
z1v+Wq+^T=u-oU}<o_o4PBRVN)XV@JUiY|DfRN_3*2w>V7OXo?^IcW&=DxaK@UNJ+{
zJ{R|n8gbT4g31|W|I-;FyB*+9y?uVy75kW_evqZF{qk(4YEl$$T~(%*pc6^^%~l`O
ze5P8KhDGxkQnl1{!MR!iJyEjGxZ@bvx|`<UM23GJW0Yixl|y{TD9Kt@!MOjyC}|`T
zsZcDcr^gs2m99M1D9LJ2NajqVB&%6@Shpcx6o}U06KmTUMoE=%3r0z`JB~g|sv%8}
zl4>}o9VIQqmPbi7wsVh?;%3f~yBOv!>N_gX*D*?3&;iArxe4&baZ+D;;W()fSvF28
zL>7OKlM1P2<D^>AvT;%&wQQVp(4xt4QY8hrXq;3?{XH{IVp&^kMT&LfB#E+=;}yh!
zxNi%zh8RGgJmZKPmyVMJrCq}iklGp65+pXeTt{GLniPy8<dp=0!1DBfAGCOw<fL&q
zz?A9Dsw~#Bmhj1@NuWcj8rQW}l%U@^O)`JNA_wRAvq7opg7>Hu&=c)ZXWa39_o#D;
znbuB6r=KDjL`A*7_+3SYR-oTGMKWMypLULcZ>i~$GqePHstlcW$K}j?k1<gS($;GG
zj)9VJpFLgwJSqJn7}~SJH`M_NQ|gTimn@wNRkNDM=goh%QY4oOWo~41baG{3Z3>q>
z#ytrFFgBCn7$cVsK|KutIF~WlE+>D@TWxO}#}WSSUvWUcq#wN8nVo$r0s~G`rvV(q
zuv4I|VF<Fs*i2<orA~p{U*BhTk4q9HZAv-`DiCY$IJY}DJM+vlvsYp(xw9Ry_4wCG
z8Bxa}mJxSi>#wd0E|igUG2}AxUA9z4_{9zn<3S2lGGRt*pXd+|cyW3+_>O-<EV3#V
zZ7M@N7JXSNFowCvM&CifGFi+WlP|+yJ7%hnK~?9A4iR@lT`l5nbae*qI8}9`>p17~
zSd5+Hx>`&aCa7lxGdSnC+!YM5I)=%2&c`wX#9T;q2-(O<S4-5n9Lj1*?Uk#5>x8+i
zmSQI^6eUx_FxPpd6LCcVN=bhpodH(tByyQ76NXiX%$>L7bTt^pla$@b(s@p0UQFpd
z*BNZ*gZE`f7-lI$%n)6DEMl1S<%#fE4ypt0cR`Jz0-+0()siqu7o_YvS>FX;Wi13^
z@e8uC3(3`mQx|ga8?w2JR0YU(c%PyJQ5U7S4iUqNitiBKJ4PqE8pVGuraA+*H|9YA
zlVzzh!1r=00#GV3mAhiQ<hTr(Ff1ED%w6(aR)b;S16+pErI>@R#=cA06(u=znL?c*
zcA4u8*qkhN2A46cFYn;>p2}kgk6evPaPXYUwvcYRwNzJD$0AR<o-SoMl~o+6!%<xZ
ziXp1%I<TY-Wlu{m!nuFyE8-4*C^7<h>{5}@kitV%2LF;2Unjx-L#M3kYWz^D^Q(rG
z>->@ilj>N1Whq`$;=;xHTKa3?GH3Cr5bB%jooLIxYs7OLp4FLSich}%cJk|gFQ=2|
z^ZBBmyt#hg>($HI{NLT=yT$T+x_oPaJNDh=Z<C)U-@m1L@p^xH*1x4d#=#AO^+re|
zT#N!zATdGmyUFto)4!SgxcGH3`Jwv;R*e+JS#@iQE+F{-{0#paMGK#;35a~a2(-|E
zNhbQVqXjeP2rq*kEQ))=BPtP{!2&QvGFjt3u|cHlC?OPe>ue0z#+*E$c#jhM*Tu$m
zv|5+hU`>FE#b1AHt!I<r*xpBj$=^NMdXLHTHvL&YTg;)*-(UZv|9$hZ?=P?ZGMRjs
z^&hX_o3q8I$?w;nE}x!H|Cs!I+JF4zvj1t`PnW-)o=wf~7nAoFi}#a1$gnjuxmuo0
zzFW@DKTNMC-!JA@{ql4+?<envTOX#2Pt(&m9{bbF`1F6JPk7ZQcQm-TXd+@4cx)nK
zz)LcMke5hCg6tX`S;8gc9q%1G<QE@D*mYRj0rN$&Nty!%0nmxh2AjI;{$kl#morj0
zfCuS@M|6MZJ={Mbp|a7O<$^bk`GE{ED1iiR0`ej2ZUDC!F__(s4oDt}Gk_Cl?}K<C
z_J}$l1j~Qg-Wk$ONe%{~+iu74F$kh!qJsDkgNzttiH{=&m(z>sykDO5i<9Z+%T*|H
z9w8K48X%)2MqfLF#vl+u*mn0mXXgVj509}!t_-$W{fD<o6&>b&J^9B!|Ai^QNvNA&
zUtGM~F9BWyd<EDe0*{kqNV>Es+QD8Ii{+b3n6-Z^4|J1Pr#%ck@3>z6^>hE@oBp()
zc2sY?Sj>AY@}lN`DXY9F>00WNFO*a(ttc1DX)@d4!s`VM7w{7cj#Mw;MDWCV5z8;J
zUXVTmu2scgWbwm==KoB?Iv@ZJ@id49rQc<pXU`_DmW#7D)Bf$`)ekQwzfM2*?>4sm
z)#-nSsd`NhAGs>Lx@z=#dbPM-o=vX?uC4ceo}SN6zgv7R4iS_fW}wWg(<Po&F<W&h
zo^&--1Vx1>Z);7U2COB)FpBM7T;Jn1Dh7Qwj*D_e)<{}D7!RulmyKAz>!bom2w)Np
zD5d{oe<x@FV@G!41`Q}8fQ#;NK*M~o{B(bEbvZpdnO~nNKwMs(H-HEu0tD(TLkN-;
zGJ}B`-jg0c&p_b7e=Rczl*BM{o<w2RQWW9^H1UCRlvBTR%?%we2}RYtyP>0KO8GIk
zAqW77x_{(u$durSZX;C{Y>$2e+k?^O!M<d%C00{FfH+KfjHn}zWESruTWlqWt<-;U
zt8B5E2JV9VX%F(JTOxnjME=xJJ~fn24dv55lt13H1^bk=s2x08p~iWB9G-pkpKI8t
zl|MWNr21sEL4-QrfJJFz3uuOIkMLL8tvy24?gk2T^%#%3Gs=jJ8}Xz~L>KWlY5Na6
zcsHq{Iw6aKBjA}3vNmAG(}A28UF(0$P70~{W|_T!K=NOspJ^8+gv5vs_W=`IA-^5v
z2Vvq6wA>gBa^3}l9B%~%+A&>&L2eL`8w6z6M$$(9Zm@t_B)WB5M3C&+J(7~_<8=9H
z*3UK?BV&*)0RkYnz-tf?%$-MI5XuM_Q6cT?N<=ToIiyb*8xfRq8nYoF0vvxv)x}~G
zoJ6S~;{Uc<+@Kg|GV6V323%TtB<Wz<pPVoHldH4Si`o1`quMfNEzZVDY-T@Di+|DT
z1?zG;ieznx4Up(HBON8muo_!`@94P&m|NxbHm+_$>7knbrcg?>%d51fe=8`h@34Z>
zFo2~V0<A{Ajn==$HcD-jS0jJ2ZA5LvZFDzwri5Eh&|3A&CMABn(FhwneOc+rodeP(
z184N8yg*G#_sN4Oc#g*Mo`AIQwn!N)DLCtcjdXV~c0d6=R$D0{AdAWIAtjEhI&Y|h
z$DU`Gn+|$Lq-SG?43SXzr({#gKn&{zrNWp!0Hgz<jb1ru!d`s5kL`bGxiOOw5TVuu
zuw*EvhuM#Hw>%m>R%K+2z<PS-_wwc=H{>P>TEQT1q-br4JdzZBKJ9DkA*ZJ=u|MMh
z0b#R&eT5h$eSN{^hh19cun8QJAZ9|2dR{XEE5oD+Lf@2-%to`uAtmky32ZXjoLDJ=
zR|5hZ)ZvTAG|DLS<A{GRS8SUS>mHDpoWe1pg-SaASb9KPM*&;JWsA~mb%(ZT(CxcJ
zG|axmVw}Gni;)c$L)=&gH`c+fjT*Dz#%#DT8;;|h5C|%Ts2?LCB{(E%LD-6ap!jV0
zDLDM)b!LbT`6@qbW^jU}=yU7VJJix{Pq+S_kF_0_&dTcB5+{G<l<~ITmfRpeC0n>b
z%TFm7pZw#>PhNSd9i^d&e54;2S$G;ekE0kZSxBjR+&w%G!U5qCIob^51|7(E-*Zmb
z<H$`3{kDVjTdD&}Nc6!)1fQVb$y&p1SFbc6-dC#EZLtCKBeHdP*ot%mm%oJa+Xn&%
zPmIUv0g`vSknex#7|5sLjCu|EK5s&PD?_=Jp{|vot|bK55`t?9!L@`SO-`DeG&xzz
zwZtJUaY!qPrRgF~7jH*MD>bY_geU8_X+u~AfD4L>hYhQh616kK2bK6zSl!tW-kYo+
z=d277yzzPW<dgODQPxxfmk<Ph3?mw1P{JX%!`=s#kSKqi1zwWbM+!k~7|#&GwFG70
z61+5c!2X_NXi3gm3Zk_qBhThM!l4PfsO`|T5?MWg4=Z6Erv&ch#8Utx$#=`GffD}4
zbwh}~v8a3>MQ)wV&u3?+{bG4?ie>)*o}Vn{C;yu+7pqRKqwx~1XoEbv77$Vz<ovM6
zuu=lMYa)O7gG$^dGU)k4a?}jrQU(clV2E5Pp<kDJWSc`u9A}6de3;>5J+)kYm=VO-
z$8t`!4v60_W|vp95A&f(>Onr8ACY_}r-$}y@|j_s{LticP-5i*hm^R_SRmkIT(bqz
z)2lu0$@{fK)nNq*hjbV_67)NUygs-?af8eh9(OY!lqxLJ_V-Epe>(X}TWUk1wWTN;
zNk@sgLgS8+%pT~T@BNF{?c=qBCCk?j(f$t~gDRInI2jWJF)}eXm+|Ed6PL#!J1CdB
z4LzX&3b$MiJ;?_IHe8g9modgY6aq0hml1po6PHpJJ;Z;HD;$1)`qNG(V)K2ZP#NK!
z?yS=SbR-HNq^Y3<8}GT3LgVtvsO~p>IUW7+J!Ii&Ru<X_?OlUu;w92ccn(r($i;w9
z{Wv`9_zWJkALbd%({6Y-4;7_Lnqc-q>6XtPgP~!C3-HLwi4eAVw25|^c*MKmfiL4S
zRk(Z_kN1B#=lj7{ten;9q7`Nx8_2+!Tj;pLbaNfo*srDI4IWz^KNkAMxV}%{&7xl4
z9*v)sGAm(h=&aJh=#c<-AdazPlx)V}5EZn0m+*5TSKH%T)Mmy*Rya2yYB-CO$5q)(
zTxJ@~<1jvI+GV;D<^@e6UJ}&MVDA0I95}|qUWR{ql4tY}`mT~+YyT#&2rAsb{%cIP
zw*LnEmF$0u$L|XJ=HXk<lmDXFWt<c2>@9RxVVd`An{;-Z^ceSY9nzKc7j+p@{H6nC
z6V4$%hx{e?Ur@6%!jfqTy|kLW5e(<w5C<P8_lsBLLq8<3(C^-|j78!^`W9-hFn!Hx
zuL*y>mfF|F!S^Y8wNcny-UOhaVCWVmP+_`2)Xu#3CUCBrH4DS0CFas<HY~`6EtYvP
zX*r7*C9SYdwOHgs5@1+OKNLC8=uc?0tEq!xEwrRr`PqxfPbnqHZp!ZhNuU%6Bg%s&
zWvsWj+aP(Mc=pSWMdFhZIK$93jWOKVqO5-~DX#hX)F~CPMq&l7c7f6DAJXQohWFNH
zdJM^RbdpDJTUA;(M9KRim!qLh9;0Q7pgoALMLC~u^h=KM`1MJs(u72Z8s6`ZvB)Nu
zGH#>a4Ownve_$;cS+G%-@%FxX`~GS7>xquEH9`?}A3yKn2Th3fa8PJ2APoQZx9)#=
z_n}?2iKMbTkm&7JT5C+5(pzCrRcJ@B9zW`HF<D!Z+^N07>$>S*Z=cmjX_vtZ?^R0$
z9T!vhSP)okTZl#4?5k;FXj`*8p^4_{f!|w*xTWVTgMB@1-V0`XE^SE6^@)l(?N%pb
zrS}yFnYX&Fs!=N|5CFg2#8#b?yc2&IYq9h~mdP8BMey}gv@kZGvgPmvT*;+lpz15+
z2zxy2`w$W-y)V+c`#ig<MH3{(hHr8Ruw$~=WPcmAswNgJm$II>N6TG+KG^EHOs0gs
z`|q1UkO<>*AV!H0&7}n4H+y#O?m0m^Hw$S|ZmI@SW@Pj9IVBX=5Z|mLLHU2-TJmU}
zBhSKds6k%dtXfmR_AhNC^=gkXd|2a*Wu%i|BY7r^sj8KHl7^lI;aCGSSpK}q8Y@gS
ztwBk7JgkX57Wu60g4pLJfW(#gRkjDeM-=r75|}5<N+Cl7^3%H2v4vb(X4cJ)9i}#J
zct=04OWK1pR;{E{l*XptMo51+k1#64E&taJzWu)I9USpi1#i0lbp0@L52HiO?ucB>
z@&c02K4sm#{nyL$A3y#=0$~4+Q;|17+1OsY&%>+H0?|iN_bI(wtj*8{QX7O#R9Dzp
zLTpWBz;WPXx6{~lwPG%&eTt(>IFH~~0LN^_YcM||D26f({fc=s<NJS94rZz*6g^KP
z8SyQyl3?R2;4@TwjaT`ervmEl-R`;DeMnL@HtH9dzKok>+}rRsCsp7^*rHOhc_^NQ
zxd6-RnZN`awK_Lgaht{%Oj|wJO9&3h-iT^sIFA%e=OHYWSI9&8-i|K_p<xsw)21K{
zh&Ccx9L$rFWC!U^&mVt3C$JX)y)W*=GJ8@%5LVhN4~3G6YxUXb!ohZ$0Kh;$zn*93
z)FB;P4;4l4O&L?(5KHfB76Hl+T|rXM+J55g!@1&sQVas=wm8~7y%2+m(MyeDSjeC<
z2bSSE{z@4X%V;L1#nR`5Zn$EmguvL3j!(z>pldi08NZ&F0O#3{^~}6~o*rlPxmcK}
zm}o1ru{3w}d}dt36vZ#Ke5Regz-Mws_*l|4@k#1I6<xe}%5wJD=bOoZ8(k;M$XZyS
zGUc7N=<M{FY(M{TKCkUZd;w(+flkYZITxDDb;gAze;8copItpG?Q;8Kn_Kh!yeYC2
z&Z)B5G;>z5SK6O6w9aaOW+|zJl<vx9NrR;~aG%adR4x>9IkH&{u!BA#qN^W))|hVH
zxN5Lpsd2T%<9Us%EjwmoP#wAJcV{D8#?h4xy20{0JFy(|J9Ue0UXAJI23=#nmO-!c
zc;+Ggx&4@UcqiR0T5>g}uec@G5PGGS+=e*j&fRlL6vcDJ{sK9FR2b7w0luKWKqwMf
zhp)K5aOy>e9Q<<Avk$kI4Zq53<W8-8uGNDea;Q7?jd~CftE1^hkt@Bi*GYF!%w<{?
zF~*{XZi!uMRp=lBqH)XK1zQzJU>x@3mZpVv0P5i2=!Bl@v?^3oZVZWM(WX>P_D@K=
zb*sYrzTCOCA+DEy;}C_k+|Y7<Kf}$|eRV%$sh8oe+sm+8DAr%t$|!)(>|~S>`7oL7
zd<R2eZ*w()(z*4F*f9oMpjWgCVzWV^sxImLU*-(M@K)vPh#8Opz)vhN^I#iy|5!g+
zmn4*>Miyr@br><<8Lj{~PS<0>u4;BR4%gWTYKU8J=|ty$AYD`SveEVc4sL93>Uu1-
zFn|WZYtH5tSkWG3k*y%U{dS`pVTIQ<_n+SEDct>aqa|U54xxhkL&oH1Av_>Jgd^d?
zP#)a<u#N*Mrd46qh2~8&9&`YOb>nYAC(E?+03a%&C?Ng<&1Aa#ED*6k8c_kQccc0r
z4=M8sCsRlnb0mdXCtLcdQ7rLoV`I%$R~)bXA1cczFqbjLJrn{mF_*zl0u+~c8Zal9
za34LGe?vGP{&@WPMkZn3pUKTiib|02ov~W5JXzAfm6#Lp>cpHiUu(x)bdZvuf2wV~
z8~D6GJ6K(xi~vc4B~EFTxB?%C{x|RnRtc-C597UfjXjM{)`)VQ9)gjgAcsPjdAFdF
zLK<v^A(IqU>A=L)_Rru$Wy(P{MZ#I#v4e&6e<rPVl++l0!7!SHcDA9G%Ci<aE7zav
zaDGmTC&TGvI6AItEWOZ*S}r=S(#t83oWo8E&4*K@y6-Ic^zP?nT@bvG&QEm7a1D@z
zHK$(&Ro-0Mxq*v)01T)c%$Aq60k3v~*WlD$m?h^0(Hp^RBTU$kY&XRFB#a&f+ksg!
ze<V=b1<XDNCNjQjP7_RAiT4pJ3syrXly3EWcHAo~yx-Pw5fru6H*^|&r@c_pyoT>8
zPE);1N(?&rEBP^`H|^7LDmH*maTfXwU=n0cDddK|^|9YeM(%U7C1HSCL7&^rj@=vL
z!MZs@CMMmhm`#%7!lGX5p^i{TfaM63e>Z_ooBo_uS3@9eGBnxE05kC7uqurn!zxoA
zlPyDiVM2AGso1Jx0LxG7cf*xT(xvSjuH$bRM?;)8kR$j=uZJ^JAS9tCN})%9#c01F
zBitA7n?eGer3myfQ~0SAv>K{*a;L;GZ5vAAdZee?q8UH1>O+zs!0Fb9@~}are>g~D
zOEp?D%A#BQM(bl-xy6w7#D$e_3#(|8a<|*0xlOuSO6x|%4R@ktivZGFe|VZvA(LJ6
zo4SoFfx?WQUCaHv&xQ0j(d4wAq^%Lwn3gI8z`mFqWG;4ZOb4goNsG&5_o4FcZM3a{
zGAPtUl#M!38HNBy{DvyDA>=8Ae;#)FUAcwyVGzsoPt(<5Y#yQu?Vu!8?sX3k-Eu5K
zSTaH6;q%-2zA(tR+M%z3psj(<gTp2ObxHt@P%o(?D2L1x8ONTi!X?<Ot{S7kAh1my
z?dftHnOVi{j%ul#%h`-=svPGmMaCKaHC|$$(Pp2lCHZWUP`n^idv@CSf6W$?sms+`
z3+G*#7YA1yQ@Nx1>$ZlD98z5EIi*}$MbX|VhoZWLzM0-*zF4V1Qb-OMg!@#il*^Pm
z)6|s-&d2rDP35#8)!OH9HW^2ubs-(Y%eVsuS92p3{K`L{nyZhGH~&td&E7EF!T<U7
zCOpu<rIGbq2@4Op`~G+Lf92+vcGafxl;vb2qDC4MdX5q`hT9Ow%d5C<_W8uX*IKvf
zjmFYRp8E<CbblVI&wL*5Q*ochQcGyrjep`+sqTCp0gcY_Rxqt{a9-@E*k{6X=i5-s
zgJozuFOf-c|7Yz)B$yY6TLdn@C*tg@pDq^7T1KARE1Lu1P)%ghe|3E@$d#^F_T4;~
zx8hxp+R}>ycUMzi%d)3nrXhIDB$QNcMB3aM9GZOHC!;d&jYUeot}HUQp0;QklLhXj
zgx0V)4P&i6I}Ag;Iqf)0>f0~xdK)tQ=ST>IRt*XK&hjU+1x|*WKQ8bpvj5gOafbAq
zQ!5f?1ep=dQdtxRe_c(5>#{kFL)ePs*LB%AWZ6cWEZ0K6=ZygG8b_5?{gQYnIrdv~
ziIWge5H=*Z=_Ia41jma@oP@~PiuclGPC})G1$&8;P;&{{s<iGT)LuBQs%N>ZcjXGb
z0Hp>m9rC6x$A9b7Op@FiwhIkS!~%P=9PHJEYOKLmSz67te@9EK%sR`EAN)MaAzaM|
zzg&~YcjY7`y0MdLgIPbPX#;2og(ihU8K0Wm+2^@(|9)Hz)DepLOaJ7Q`p%~IPfw`r
z0kq>)at(YcLw>o&OqsYfK~ew$U+k+6dL_`<w~<U6VV?L^2aOY+{FPsIYU)(CZtplX
zp)5w3i#$W`f9_ktzC7NA5#;{F(AD2l_92D)1`F<uAf*=OiQiICBFD<Bky25oPG%oc
zYRU_cQWN$K)7A5Z*3)2}0Qr=8LQv3J`-bNUsK+Wh4&@UJ^GA;?Nm01xF9BHEBR#{M
zmC6iNe<ZnL{Gb(=Yn(`HbH3F00gxdAuQh%Mzt*d&e~GO8aji3=(dYfK<&5d5;+heJ
z=e;@6Z*>9@zCEgmMV@m4(GL|`nU+QnUZC|8d0uD)k%kI8Wpjpm1UrkK*Bn83j-Ke>
z6nNM3X6Q>zBliAj#rzQcb*2@2{(r{^h1U)&3gV$bU6==O^XGYMmlPX4pi!J{Z7tvo
zT&f0Sf3Q@sqZa#8PPkMFwB4C=KcQ-)-Tt*P0^pEsS}yA|3JhsOHHudgiUnTW&MWCI
zR7qKGkSlJt-+p{|2gYp89N4Ak^pb81BM0J}XfD7(_$H|owK&*aNaVf0YC6RxH(b%R
zrh|8P?2X$WR!yk15~h}L0~OqIa^SuU6^clJDR0a$8A&WiXgRM^jf`*W5^`!mS?l!E
zx)nSK%dbwY&{jBHLH!R3{hF77H7^qbF)}ij(MtsrmrW-<E0=pMISGGgBQVX{xFC*z
z(Zn*4Ta39SImK~K4iKOG`6;s5Qa5|oNaNUt9%)*tW-YH?6<M;ozjt=`?(Xn-_xeb@
z^Ck#*+&zBWfqr21FIc_lWaj~lBiKED+Wq#%T6B6H^kQp?^z!;X=&S5s5d{4P=r=&`
z0_<OT2E9Xihx9ur`f7jmLVfScy<?GgDj!_1=?M>d-RpJO|MB?Coh{0?uO|&jOz^yO
zo{j2JG(1@8NxeQg+(i=|I_}xWVmSDo^ox%#lj!TC_oE^LhN%+y-HI(z6AOtL0_aP9
znP13(u~r5PLA#KQOL^nCQV)yBM3B8VR$S%699%v&R|*tA03CloUo}u7drU#i@AaPe
z{hTR?U?9egK0_2^J@utN%=!$;8k8{UvvCI?IpZ+a4iGh3aNaOh*V(OL&=fEsRABh<
z<I7jz+yeU0Im66b7xY%q7q_4fy)zbQ=*dL{VsJ((Z!RDZqnJoxfIyc5pw|gyEfV#;
zYyQvg6Dr*z3W<NC!I6Zr@bVQRzW8C*hstpVD;*cmNM6|mX%dYJ)Uk9#GUC*Ks;*Z*
z7H3e`8PXWcLHUvs-)8sdr|k3H=J^V-x#trYAKi4i>lMq@uGt9$`DVpokoJig=Ht|q
zf&9paqV#=7@d=a!G}hI6=LOQ*R4!~DoU=Dop`Uw|U{HTIw8C)+lXYXlhi%>HNn4Dw
z27sXPPzyDlX_igo6>w(_>lH|7_gNJg-hDr#nnOd5>Qzo9H<eY`8t%#|6Xq3QNgxCM
z1YAjxJ)G+XzFCH-5+sEBL5tp1gDurk4G6F3UZ+oodX?I}RNUCYy+vjEPI7VnoI5Fy
z!jaw|_osil>Q;Fjwctl`1Fz==uVyy-U|8AhLbEg4T6DEckQ%?(5=Ucpmn^V=?`^|1
z&aQF?TdiAh;0}QO8qO6bMr-=XJmoS>^m+=_eTj`v#Onvddy3i;`HD6U1pLjBZ>4vB
zV5b+%emZ$#k$gqHp^?3ieEZTX$+r@^5&2fD49I`?vc{&~15KLRx>rkpbv)tTtJvFy
zqPNmGEsU~Vp@G)c7CMJ@)Q7<k*LGezrbtS#anZ)Ke-&<G$|arKG@aluYxK+7QfhN9
zE;xO4(PE?F=qK+xDO8ArH7~vNyP;v1IGTiIZ_kQZChXdEO5;OgWA3Z`IZme(mqJn8
zu2g?~(c_nf+V4B8&`kk2Lq3b}8T#T2d<y%WSn=UhBokt;L!SAO%ZYdqd9sVT!KBL7
z_s4YH0lpL(;iSMr<L$<i1;bj-5+wBHUG?AF$Gd-rkN`HQTem+y-K7s0#5k>lL`n*Z
z|G#!0?|!YHS}}|59u4KRu+E9}E17Qdw3>giw&Tp_j$jb0O9PbyGF|8{3Dmi_DhX+i
zO%Kv0rZ$kf^PilEy*r%*Bt(M_YWv|hUG-Xzp+4)})sDic)3dHu3B(zqSS`@PsMf|Q
z$waEPS(TQXHAAc2nC^gAX@*Kn6tUK<)Qo{v%V{}|^`#b#0vkg4CV;)jK%^u9Qrmx&
zA8ZTblv<qr@3N)1_Dr{}2>H|@(^j%CH^0U9Y{-2_;ro+pw@Hkr?MZ%0)*zIp@W>8e
zxsfCVvxQ4ff4kBzKF=SoiWsk1(-X<(Z|Ur=dp0({qK}$%*cbmYzhVeR8;ick!hvqo
zov@?ID_1?xWiU?r+o2m~k0G<a`$B&U=eeStkNTxvXW=|ov|f>`?eCrEPH3o`X)T{y
z5P{wskIR6VI3l!Q1ULGTtSCWNTK-eckLQnd3AI{9H|fxE8%Z1OMBf`QNZK_E4`Oib
zBi*R|dmXRUN4gAPLZ#yZAQ=+0U?U(|OA1@8osr2dZ+m7zsbAA1h*h}f)#QJuIIaTr
z)4Vf8ZxW3;W9@!Z_hQC+cPR^Zq}Q@=SL6Z~?usu={emc*ab(5Wg+d8KS88GE7qw6W
z3>I3LxhRczsnlDTxhVL-K^3&C7Kt?Tw7k)0s5ndqcRi_vD&e%NaXjg>2?uy<Jk_XO
zRV9MD4rmKvF)HsuTt@5!&uM@Ah7dNok{}ver^&iGBM}=XZtH>+jL;Uir>`Q8gl@Se
z37d*@mP(Ss1UL>5f8COlD#Re?*Ow&qAoC{lGm~*NjsE!)?}QQW{LQ9}slf^Ko$aFf
zHa&3ynHosJOAo>O(kl<aE1?@5f>)~y4uxl+kk4mkCzj1o&caFG%9(#ri&YfLEc@ia
zW%ZzygUd?lMuW?0m9Iazti67l!R3W!KmXveQgr3u5|vCa$n6K0sO60WzbihkY<l8Q
z6CxRvxscEM(kuDA61oweSF1dk&o3Ll)HAe|d(-qIM^KWz{VcS<(sR#3tJQBb3th1B
zE6qZ4Hbi;CEEEV0d-#7J&q7ho_tH<VnNJ0X-*!*l&mK=!ByH2LtdqzkL>X8&cC~+Q
zHNcE%vkUe7646B9tIsb%oB5QsopD^>zn40<iA#5)_w=D-Ize{<YNcm}j+))Np_DN*
zuQYYU5OfRg20Bd)_adelK&{Sg#*U=quWiUm)rg`&1zN4!a}OpSofty8IUwhM?nPEp
z1}k0G&kOxvUKpNjURY&9`>Is)AOBNheU~xDJrn^lmqCIc6PG|QJu81rkKDHPzQ01p
z*~2pABOgC@*8|LKb`12?q<bh3=TJDc;j}5x07Z)a`o2ez63LM$*|CirwAzvM_`W|$
zIlLd_@cQQba`Vkao52Z>%E9pRZcy|>(qAR%p~b-|C9GD%%P+$Zk5YoA>r9uCQfs;#
z{qO0H$2+Y&eWB<JMNfZ~I^H>>=!vE$n!W*~yQB-~f0v$J61*jNZ@uIdp6EL9{rKa{
zpKfF#Hs2=-H45*{V4V>}N22g7v4Z;MVqpj(Sg@wz&!8~gRnUk3-ZChG;35cW5sa@v
zAcHJpv07+su$*u2P#!+~>(94$N{7SW-u~m3WJeF)r?s40S_^-a7a*}->y1R&r2qhg
zbfJcN`#`9zJ^a>3s0pD8R|r*tS~eUoMxYOkTBISRqN0lwkHE&__^HWeL$)z~b3Q`g
zH)Irs%}Ca;@%d@IBTnmF=C9%xCclkq>-Y=U7!gKz#PpSnraE$je$f>`3$(<MAt8a!
z9BE9r!ni80$g6*$PxP2X@l$%Av8*^;#-d4K)=pS>g)?J1-T|7&nUfm#agt&Oe_A0v
zvD11gzaOO?HH7J_Fy47BUQ=EO%8r^#(stFV^RJ{UmVLad@!5zMA6uHR;Zx~zqG?>W
zlt5eulFgW``Ep$orf-uq7|ZKV*h03Ef<H`h^T(b;8DxL`K&7WSaOlB`;IzjY#0iI%
zK%_u7#?r%wt``}r$Zd6N4Ws-mb8t%XdMbIt{AQMwDRn(+rI2l(*U2V93Z)X^f9dk7
zAPDOyNK&-2%GSIt#~BXfjwM=RdDRca#tL98CCIyeC`Ki~))4>fvw}u|hGkaPkT`*+
z5m*Yg_k4eI<8Au&Fy8*tC&Z)BL9Gz7EFkZNlZ}AhcTG+OEv&C;f1HBjl}ShtV5VRQ
z;piX-0S<(AbXqo869alSOCv*@4$9E%QUJ60Yxg*Zbv*GC4s-^KfS(*0pYq$jdq;=p
z=1lL%x+)eA|LbZ9f)YW}D=Ep-P14ZM{X`i&rIUYFkWpY-u6TH2-jrG3*z+w-j^doH
zdFdx`O=cggY6**Egnbh3N>>LbLa5i@3ZRlr6C3gN1E`WzE!mcfm`Iq`=cC5Q$r`eq
z6-lV(llh8vxse<k3V_->4t+B)UJEL=m&DCEEs=3e7^J!;XZV~8FN^L<d<Y175`#_d
zme_x4>x5jLO*y_Bzp2tQfEA8|?8h3<{sLEILCI?J>QgRQ#%Jo=4c`Sq=LV!=T?f0%
zdCA==W+O69s`1_B@Ctz=gsHK>?HxEbrJ)sv<bKLFko^)~D=VzQ8swpm1;+BB0#RL-
zG*}_AA?!W1Du@T)r_TUca6?}|J!#{GZb37k&k`3=%B--oOqey)ec2l1ekmLOl=5(2
z>pB==gI=b3B_K#cvVj_;UqQvlMftb3$9|WWGd&-F#PWvXPKNS}P)-T2i8ksBed?YL
zf9=!eq%f4Ukb;-NCW=xqya+;aZb4uJOQ2c}YmiAbW?=a!U8ihMk&b#=q^8=CD)y+x
zlWI`cLfZ;1RxF5l!`7t|3RDf%k9*{RlufZQfuA?R7YfwAM)PV6e6sI6+6s=#CE(7u
zQiAM%j4LI`&bU&7?2IcV$Z}liy9T9xN2wnvz$-AjT!Ye^He{R>il$0v!!<kw_7(&-
z%UPpz4YJ`HWV9@KLtTW+-t=Lg$x6y^_M<G8?usnnqb#<-k_=YuGS~ucr@a;sJI%F#
z*lDc=#7<)^AiB2NA5U?Gn?+xZr^Zr|+n2F_c+CCYID(6%sT;>dX((5ewCNk0N={vj
zY|`+Dd%O9oJg&-JW*+9vUFG!kJRmD6xx_-b91lWTE#WMBnOqDP9p^^>sF4yS*0S2@
z0yJWrH8kR#i`|vnO3{E38KeqP?aQs6k_-7MaM8Xqg?WkwgCT$56dNIa+lU|JbB4Ho
z(Nr+`eMqPq_RLC&Wh)v`;tnmAjS^02D%@?2=8Xi?_aQQV7cmu3Z#T+mlC#t`$VU92
zJZvdaRUjMe7!{}c6$o^b4%rJ;qFCF^4}C$CJsa<%&8W~MVxGb+fl$3m?RN$8O7+q3
zW=SwL%)|Zt{liavxluDIb)%hrjjgGF<x|n77Yww&3t%Ysz%(7xIU1#IHP5XmvKOQE
zaqOC0*{k02Oqn**wh0tbN0)K&xtM>Q#yHHFcz+F^#(g#Zp5ezC%~9p~@2MbWWi+K_
zoN+hA!~mIQ`cu%v6oTy(t=2Z`SihkC>PFp)VA4ii4b20`Zn{FCJlO)Jp-M!5OI5ii
z&68!*<;h-$Ux|?Gr{8x-vk>dQ2FtX^$@`dUzYg(v)T5P+VEP8Bb)YgT%o+KpV4e~d
zyJegdjk_|-BAsoRCh!E^{y|K$CMH$|cmyc6d~jlK(=>UWqi}5x`LozcGUFQ4m96NF
z$u&I9TWOKEb7LzjQlo6dZ9{{9{WJRmXrAnD^Z0Es-GlL6u^H9w_piEm)*Xd^N}XRr
zcNL~<zU*AXP8+Z|o|WJ_)P#DI8bS)EYIFeG&S5(tVMWbUJz9k=p`TmLo-!cCgAH-;
zu?{q8Y@S4tW5EhvyH#$fzLeo>sJz1TM^<@F=$BG?UmS3mBQ9T4cWAnQyivF0CFM`a
zjjFeFx4bzJsa$;X{p+g@4xmcdwA&j^RkoM`(m5yKe4?tDHRX_Vicx911)Ae15kn{v
zpr7;p6vy4D@KceJ)Oeg~yP2K8&D*QdaQwJr>X5h7?Cu;(OOd=yBa_p`=3ba<jPV@B
z(|p21WpfLbWz4tN&9CQwmz#fIX-;`6_bE5M{pE&V7)1t6Ee(JO$_&5$eRy~CUAt-%
zFJyV1En2X&7MpVkmU?nOs6ZAE)p@pBsGls+xP_ezTwX?B&xS6H9OggKu&XD9L4#(-
zuLxC_)glJClk@6|wM8<gYVV_Hq9DWb?v^ICOpgmxd%Go3?-<;Fd@OC63<a#(NgMBS
zA%)A9wp3<a(UpF8oM_oHw`yrVTf%!OpWh6-#WVR9_3;hzynP%UIY)3g&-5@I+gh62
zxnjmPPSui<Ph}G@!LA%#bDKxkHY<3g*)Z@EPl1-uJ5PaDAkd$krc<CbjfK*aR4T>y
zx7@uK)Oc1jY#s%FRRJh@*P@!HzAJdBl4ome@NAp_m4@PP%WZw|^LqNL(?QQ{Md^U2
z)$Gjj7*x(X+~Q3~gr1MuKF%MT$MFo~)MCi;Y-nq+Y<hObsmVlgI#;;n227lZ%jKh`
zAzAvR^#zwEM^o9DYbITEMz$m%N=PSiQ=c0N!rEtd9%WTH(h3Op#im$1F`hevuN&QJ
z+`n^X#hWK!K5=AK0q>kx)i7UZU{xczs^VAGwb-oXI~8~T2U+{<m%*|E69h3bGB=lj
zMK2VWjX6Cx0V$WoIXyRj9huH4&)~3o{Mdo~pye-Ud1$$F0Ll_{kDt5W?zKiM*MnR}
zYa+RveD-qZ%^iW4H$dJ1c?z((a|YxI$rF-ykmauBLiz0Svt6dQLhr5DafR1%2@jyn
zACEuZ=q&6Y8IocfI_r!Q{zWq0YHftd$ZE_R#QXR;t?+W^VOB?f)(U*P5Qa>&01P{2
zG__Plgl5|dtdPCOPAemqC#LSJs4#u{{@t=@)+%t&av9)@5P^ISy~0KyW`UG3D&N0j
zLjK+K&!U9!vx%Q~8@SI{*$U+)wb42i$WBmC49})EI<Jg1PgOg5WmG#|(IC*jhw`1H
zuF$KU$<iy>zs7HWduO)#C^(F^N{0y?;`F`mu%ESRaa|VcY97>c8^9PlgQ>RJHcq0#
zxce+(G^(qtw&*^E5qXt<+DC6vnllJKe|Np|kwe5f;v6G0MtF?c!F^njZ8by?Sb5(v
zPYfk6i{m9_2XzFyH}m06vX~Q>0Fs16d9?@4)!S!@Im60-c-L^6cfQ$}%Fb3Cujux0
zT!5ihWLiD58=o)0NWwTA*6~eAq}eI31XGRr0iBBNI0#q7E7-2)bZy%;+%IIi13DHR
zN^A|ugr8<R4=Q-LhV3d&SGQfw{Zh7D*Ri!-iodLP)N^5%I@*>}RcLb!Q;r!;EouAR
zplDUBdcEy`8yzbP+{k%un|oF$oA5dqvHE3`_u>pl*^8fRZmDl|2-mQ8#p!!CcrECq
zEWQp0?)FtryrLedbup{DX(gDD7deQnw{o{ZV=v|GFYl*K6C*1ZhWwKxzzlXS3Y$H?
zZ3tP5$!%4d50%_Qt`w;MK9|i*$!4~Hmhh-b%--Z{+fiW%edRiszEu63mm0CD-u2~q
zS`xFA1^pzot&86&X)j%Rf43LGY}1mF{0a53Ay8i12FmR&_Q2qBL%=37c-^DfBbEJB
z+Ky;)xxH(?y?wm-XJT?=y+ZA}KR@5ZAB+^|)&<%3Q~=YL;XFMWe?}T2<7wP+OaOh7
zA$l9m+G5QR`)Ja;KCzUjk}kg6|4C6)JsCw0dVvsas~qeWyB4vT?DJs)$C8ZpWSOKl
zQ_f99f_ZUdM-bc>t+N_cr;FyCc%DC2HYtfb(nPjg-c97P`><SyyLo3m)GHflR|&#*
z<GJfCwm?iot<cG6e>&%kvdidv&q}Z@qq>r_g;!2IF*~O*!)_$$cC<Nf2ikEbABI6t
zS|v%~M_YZfcOU5V4H%`D8jhDci(L@QqBV1s$cXsGx7!{<*|p{<m{c&nqQKkL(Vr#R
zU|lWXf}><4lCx|QVzt9$>q-RAh=w2)BUMPe_+4Nv`rtf~e>~K+K3`tl-hs6*KD=6m
z0!bcXs7QakU6s&S<)9T1R^dYdyq532jPKUb5ZDd-6Vb3SK(I!mP*7)8aOAlTDX}g@
z3RDcXC;7C_#93t<+%n?d)_Hk>kD=!Ep^%8-W0OcVdO_;X5K#cdSWCBnTQIUE<C2l)
zU4b+x2<C*Se;|-@Q`0d~v6ur_?N62!xM?)N?HwY%_-WNdBuJywT4Uk^(Lh1+5S$VX
zMx(N)1%uIE+@b-kGytZghk2ch)LKaH1m8A<3c)S-_ACl%XVI<qAg`irj2GzqtYix~
zuu~XX?-}hVWoIg_<8B=qFN$()i%;tUNY7?{o0g6of8>)7%IW<mZoZwmBjYDC{G;ag
z<Ul$5gViLB(3}q%Qdda(!<xPpIam#land9$mjmmAiJJQG*&fDStsy$|kN_+yk}z$!
zOojv}+4`s<$lz6Sm+I&FC#TIqN1mKER2MoqZMgjRC#NmympM6oqT1I#Ic)?zHsNYk
zqoa?De;u5nmu<<rA1m=TLO=7s6(vsv7oM1SYw9BtZ$osU#M^K=D)FvYuBQ&>QBr@L
z*Is@wUvs+lnOejBLT73Rbi{4euMGjd>f~Ko73DBIWAcvDWk~k?0VnS(>F$jGc0z&D
z%(cepQe$<9U8&(Kjn$>1Q3mm8-SryE{W->kf5zr)l=RV!>?>1ql-eyw(6`#2Z)%Rg
zDktvK$^@%Z{3Z0yJT^DtSwzRmCRxu(hHquN_Sih`X~^zsp1EELQAqQRMyUVASX>gU
z3TC+;n;%V;LSRgNczOyZNT~!h)t6?pxxmoILrt16Tpv4%9Tn(W$wSk44`S;Umvast
z0g;z>Ks`wo+Q&|cwCz51MZQfEEO*(N{{=0YfPa@U#yu35eGNSgmn%Ox83QpfF_#V{
zHZhmjG&~J|^dY?D%y38+L4bVJX@CR`>=bBg2!bT99VPOvv04X7e|-<hd%03qw~?jY
zg$#Q=yO)<d9Fk|wko4pu8D~8HkH-6IDLxZb%NzEgT4Y3Gwb+Ez%87}-l?x^bS4&Nf
zQtPXq`mC6xicsr`9-lcCGow0*<t=ikC8SmXevP?*wTWVxTBF&JQ(tRv2JhCgF&W}k
zAfkYTTv`J-3c2D+-cY1k79FA#w3=)Hp<+ma4pOZ*1;Z3tSvAb5){9RLV}WEiYAcWp
zryNRa*%@)vI^>P`kgExIQNsYJB2=A`#YkuvayAlk>pW^mwHM@2LyFb8S&d|03wUpW
z5B#ium}&ww4DeB44)rw*iD;`qOb9tw69pnuHSt=~)><@p-d;;aNooN!q*7f%UW^8c
zwY31V!XeBI)dq*X(O8F+SdFICD$6Fi)(amoFRCBlDHEF?0N6yKT9%xNA=I*PZA2BN
zH6jxe)z&a%sSP*5bEyqC;pN1n)l^Kl)&$yrOkpgLOp0R@*#KDW5VOhNHM|0o%Pb*y
zF1J}u>`bQiTJR>fkuUQ#ORgqIlS8Vlal-ue&CDjJQ1J?C$gMH~YUjOF8^1V`>dUe`
z9LM^m;HeAEn}T*-tn;qUK9^E`H<U7&+V8H|T6<m=j2|6>`R_fuRu7)~(7IU`D%NR#
z^{SLIu`W9jw2lee>MXT6u62euEjQTZlU?mgb+B;7Q%C_X)$(;@w60>(QZjL^zs|6%
z{Oq&Oh97@_zZiZvd%GCES}iw=<>tCBQwG_*7+x%{S0C0Fi|dgR%5UE<USFPlx%%yu
z!&e2b#7rlsVU2f7F#s#YzF00-V_mU-{P6kb&t5go!k3Y=CX2eo-I8vFZfUVjI_lcQ
z@bbg?rhNPEa`}rHzFe(eFV+>CI~o2qd^>#g%C|2s78jdW?AZ#&BLr_1^N=-K0c!>I
z7E}LZ_`;z2m%}%!A6LWI<`YT|uJ%q;vFe#w<j<eu&!a$i5sNWKyOy!&D2l~@JSaqC
zL9sg+R4`&eBSn|Ja1fUtkAad{aBc@7K`r=i7KFs09Ai?!lE~_Ei2HMdQ|sAp9pQr&
zzl(ul25;amM$v(Qebl{S;OuPD8l7xe=Hi_7VFv<M0ldI)1>?miZXa#{hgk%r2?D!p
za|tnbL%83Ryt`G9gYi?xi7Wzt50*f(H8=#axh#noEGIpAcB>b|7rpw6&E;wdK>v90
zZTa`fPn*sA>%R=c+sn;QAI|N?>fP}3hj;HkeZBZ~`2KA3)8F53z6G_c-<(}6?9W%j
z`PJ%t_!a97-5ajg7sHq9%hzug*M)(uH|w*@<z_e^f4W@0zPvcwtk$Q0*Q;y1{HO2Y
zr{6K5!_Qiezb>p&wb6^=pa1$d2w34nML6?j`Qhs7WUg35u~6540z+<#8o;<(tzW(e
zTGRLl?6v{^rD;Sp{PnlZH!nA5n}ul{HarJ)qLJr?f?cZAH9SwgvU2-U=r@yVUy6jw
z2u|%uC{G%<J};~eY95P!3B;5SjXj4S)~kz`i_NRyhp(RxKQ4aToZJ}l9Rkm3m%y_=
zh`@{LyFuWw({buno~+pEJ@%)uKTZ8<>g1m~`KM0)sgr-|<e#)#9YN}Cq~1pEZRFlY
z?rr4WMwZ71laV?!dK(eW5Lj)3X3*rV6TeHLiJe1a3+iG~Gms*Gd*BX^7PWQ|5X80w
z_&E>|1kgp030a{Sl%cc2-bhqIWLhHH<Ov0p#C&8=j3AiAI!v!Q5RiXSfD(Z%Ai_Fm
zW!jr3LiV<(t;rM6@-_<}*ApEC0!xNRW<fZnlpq^f0`wMZ=KVPWF+C&EnPPh=x+5JY
zv`-pm!oVTg5(H*{Fdzp&yTXX+33G!LB7T2<2+&zbP`V!wb=AisJ7NUEAPa`bnF9ft
zmk7)th@=6NQ#9?*5yBv-1fd0jLVg&E><HmVD+98>1eys1#@d!-XAT5HTSC}MoRH{o
zOaV*l@hF(7&BGCecxdkMe^A_ysqQcN0>Sq#b;%bf+TfpmhT^+kyt_QztWM7_muKtW
zC&?$em3)t#fyl?LgY25PnG({EsK@2z4r1iySr7;0X7ZQ@n0MU~7@sId_QQl>^26O2
z4zmzPKlqHCFK@Y-zzQxNIY><)rXV#3B5FuV;!;2+9y9y+C@KK^)!cic0?2i%OizOg
z{<B!GPFKr+MNj<P|0I4n_KYt2J9EApC8nG&fNo>DS5c(<NX6yg>EW%s?^dpOD?PlG
zINnGPhvx{Sj1@Y*gXQGhQ~^S5**|PwFuBV<b*rF0i4Vyy$<t1LS@xbElwW$ikzeL1
z<14%6lO^5SKGkl;Zl%ZPbjT&GLz<|R!3A8-C2SdgGeusNj!R2WURauWn?nE!1k90h
zhzZ1$L(G9VWDX%}Rk!<tA*f}4Wa&I&VDO}1q0L~BC;`e=Fy}+`s^auf^en~w<&<Ki
zXNeAqJv^1oxy|D3VqMf)Vd|SiEqBSakBLf(*|zOAG&?536Y2Ky!@G0z`eL>Gb+Nui
zyHl@!DcufF6Vv{|zU1v<E%1nvE$3JFDC%Rd<tXt0G3r)W`uP3KkB%yQM^xblMissh
zRm!R77$ba-7Cv>WKPA^K-!1A^k4V%b67`5gJt9$WgL)gyi;vuIRHW@1jbbiwIa5z{
zk*F^$xqDoa5XYhlIm+T3hEDfE$yzdIDR&EhY&oJNRg1`H0^$Q!>5&dhVmY--9fV*h
zl%PEa;vgqz(`b}rx*Q?E!M3dK0OgQ%cT)v=JtL}$!+>!1PW=7?q{DyPZP$Pa;WQmt
zC^QUxM`aGfz5`_PR!a^y1;~`_Q;CmfS68R6FW<ab6q4$p(eGm?W1M>g?gs;5SRqP(
zpzR@6!X0a{oE5r^1%M^M%$TjZKPI4nMJT>A0jaa#yW=v!1Y%->IS>bB0uTCui~AZt
z5{9Lr#~P1uGXo-75Dh0X@R$>j3dbdb3B=@ySrB_>17EInlzTywZVX_9l%60RoURv_
zZ|;)IRpzn(gME#gCAwxrG%WX#?orf#W4V<3UI);O4>{?+AuT^zGln}fV|buu4B>`m
z<gVhlt2pjTQMyu;t`wy!Md?aWx>9VrGy74140;PdJ82F^TDVJgB^{2Q9JmfDsW9&T
zwvMw}x7|WVzHNc+$XmxFhH+<N7Q-QT&|?DMr0vcz)CcpCPxU4c)7Hcsi2IR$?E@X5
zY)ksq9T7puo{#K^5k%iYp8;{tgWmgx9k2oDlzZ4AQ5&;7jj(fmxjY4{|8lu}+fUQt
z9yMrjdKqyaEQ3)aP^DeX!z4Mml%sdhX+f2@F0WB4+;6WW>;|)%2WvWbFq_|szrB5b
zzukA0a@R2JY{fq=8b;ZThU2w=w~vm7?F!cz4W;jHNRNxs<D&Gq7<#$T%Y|Mp^m6?o
zF!hVT)Gq>4zX(kIA~5xfz@)v6ejAPYZ8Yk)(Wu`>qjq=sWH0$nzwR~WJF_>f4w>&v
zAf|k04#Z*do$<DpCmVxqZ+lU>zk1|Mqk|Y#z$}PEo=;?C;WrLy6iOsqlOFrB*9c;A
z#0-eT9$Cd~;|*OHAVqQtNA|=BV!B5?2jX5Q=?^`+W6L>gPpl$idCHNGtVR(1$Z7_}
zA$4I8#Au=$M|Y5Bj*oqI*Fg09e={Ivb>V*jP%H|wmodgY6aq3immz`(6PJ?5Kq!}B
zOg*CkP`3n4J%j~+>+aNLML-Ik&e;w+gqvY((9n9(167U*F;F{42b37NT(H<*LCEFP
z9|oiuq+w9f4$E}(6poj~wR_nKWgLTLJRg{}a_SYP2t;(;)<4aeaQ#A?(nYDjWe=4@
z@9T*d1yI<;7E3Q)>9Z(Hc+XlXDV>YS)gI+Z9Ggb)mek3ASmu!n_fk&>Z?DBRSyaey
zU!Tc#I4})tD_`Ye;Q-VZ4NPpCzTJoGZE_@r?-GaNs44t~I7Es3V}~l0M0q=(qS}+7
zIi)CWDkmGtUU=22eLwcBk~AZseqWxoBP#AjeV$v+Y~4Q3iUiV6$Ai6ReHB|QUbw-u
ztdxTj7dp>>C%}=nr+tQpS})YQ9rjmt=On|4#5RxLRb6RH@2W%8N$OiJJnvm-CV8X2
z!rrx*F8MM!jZxLyz*luBWHl0R-4puS+l**U@CzH_2*V;;<>Q_#lHv<i_~OXsxpbfb
zb+$9On{c`jDO`N>!@G+Un$TS}+Vm$hU7oq<4Duj<4fcykNi|=SJ9*0Lnoo?VI({uG
zt$Z~o;j%1r?dZ~WKUUtUPI49t?HSE-LKK%IJ0I~(nj~h;TKB|PyJ3kM(!ve5vU+@;
z%H0r-(uBx)o?Q_|2@&+dp5ped`}Owa=AV~)iprIPfWQ8H6J96*HAlQ8Vjdyj*S{S;
z-hAJGuiAMLUJp`5z2+v)0Mzl$FpZc?NZL_`4CCErNjh!PqOfO^;Lpi36w^SSc$oi-
zrqDd8OaMmAl+rx|;(0axg+m46mn^T*Xs)Iy#scRrXcCL`gsu!Jxw-{I+qj-KWtk0r
zeIIRMUP;jfZ8<_;&}HV-XvMgeHFhf<6Bc=YpHp8|J;y1TQgPEd!pdoSH2z_G?6NBu
zn_zxETc7gE$BLQ933g?W&u$L#Z6=07Io!=^$-Q!tpA&`}`u0cAG9G$TP%I-_s9*Bb
z({JO^F9U=qq<e(p-aPh^3Ena}HN(~i4#L@GE0@nu!r>$YNd?0cm+}*L_e6C2hw)K=
zp~l{cP8c7C@(W3W_kLA=>*cI8z86L_lST6o$fT}3fAe&T14^lNo@Pc>mrzWecWQ@6
zMXs{Q+(9mD%eiq8jpuRpNCEH8+ES3h{Yy{i3e>+w{(nOE_dcOp@6*l+U4cL@)HM^j
zoihBFCUnEeg}rJ(*N32Y)qHM^W8ZjxZVj<(I=2ScH=Mh~#&|Zj#_?*Sxr<~BCv$5u
z_6_D%5bJZfHXIB7PmkqR9PXUT4MUmws-fIgieG3Zw`OGDNN!8kt54+C1nnEhZ6v#E
z9=B!TD~;pUjP0Aoy_EE84&x$A^y16T;?5}cj^egN?VrT$I~8wVZvF#OMHy(9F~&U=
z0Wg=r_!<-eH<uv`0VbCasVD@0<Tet%`&abQMqq|4zQ;}jZO`-o3*?Zj4+~^+%4C8B
zNPwMFvcG=6q9}=EOO$Ll!5rka9-mk&*5j+9<l&ctJiNbpc)I%ftj)oTV3nuC)2D;N
z4T-->;-a$$uat0F9iDza{BSQNIUEl-jFejA@c2LA*B)=R3V1@{35BbFN*!;#QMjUU
zMdKN=_?0*a|NDIHobe9$;DY2gJmSC@${v4w`q!1r!hU^rbciB4a;B7&0zrI#aMlPv
z2zvd$8%6@54^Bj{gRK!>m<WItr0jtB^|ujX7YB)6kT=yJbaPAIUw_I#@0Wm@_2+7!
z9EYN1gcHu#3USvTgOre4URMa!osK3aq^+g;<>uB}dmZ}184^m?!u{GG$^@Ys1ofw5
zJ)Shn(^;0v8KJGK5N~d7m2<<7AJ0DC4X6FZ*3k$T^`gkj0vKta7OD{c?%JSH62_+*
zBkT^BE?7Mm3cF*y;Nilu_RCS1Pgp$|EG!y9XFfxwg>lOesJVsE#wbx(_NEdtEsZ&_
zew3y<JQ!g#0t#tB+^(ewSeKJnJtP4zm$+CxWC1Xj9a%j+e=|)uGmpc4i|4I9QhEZU
z&+s|0#*I&>J1W2N#8))q?fed7_#HSN$G6)!U*)Ix|AaW__qby>RKw~m5^g!S>%TS}
z1$GDkvw|j1E$a^~SJ>!V{Wbb}$nvu`9wyzHS!d8u*FeAr*a@qdsJ>vZDPuw_SP~=7
zBJ5N3xLmx<f9cM<qqg_cB8<yB8O`*3ek;EvnM3F<m|?x!TILP?8^ljfZfTlUO_Qqx
zqimyA=R5qY90g4RUU&z*jpowGx@_fnTGLx-7*#7jt}olQ@2;)Fqz&0>j4Ac5IBifO
zf`hhVQ90G@Os6cG8?K>N>xXcU8W83T4i(0D+;<tRe@IwFeG6TsTib+4t0+ux^(gyl
zEUkfP<l88+bbtm!4cVx&448T=pW2N2<Ql>D)ma?LRU_Yq{y@QfJ0<E;6CEWY2&_z-
z#5WS;hyKFC$`nrt3mjrYj(u2$XfO#^L%vb{VNil`Nn2s=-n@Bp_px7yq1^V0Y9(w{
zVsNI`e^j?BF%<Q9yJ6-O!>$~H!pJhI;@VQ0T@?aKLrcvlA0(NLN(?p!B4k0LtlL+K
z!PpSj5jIg?UDTq>bkjKzcseD7x@JJ0AQ;sUwNbn|L9!VIM)0ve5Vv<sY)~77JA$J(
z8rUG*Q41qiG_b}a0t!l9L;8eUOD&vZ!89Ooe;?rvh@(`S%?9CCItf%6=SD|N6MJUI
zI%zFbn>12^`WBp2#;uIPY{#nSIFPo`H%i`Wuzhj2g6xaC6=Y}JEq!!f++Fn1e%!4f
zJL7JJc}Co=l&~-E)<(ChE-_f8kV|Q9O2?xw>yD9JCMNcncy~{Bg*c1(R&Txjqc4a9
zf7f!Y90R4An4TNCU*?8B6KR$X)l6CYfDL9;dWyzodLr9lJS!P%zM?YMl;g9hdKm9g
zqV@e?Aybd5n)_dy)wy$6H>YRPuu%^oi)(8x^c9M#P0n>?(t*2}H8W>l@#wiu8nvfp
zopXU|-%dK2s!;W|UmwEJm`s9wCpubLf5TcUkKI|-kj<7UW{2p6xB>DN?Vdc%izPLj
zGlgrnCrfeu@GWEtPH`C$Q!m><K5g8tETy<%<5N>Rv<(sFNlTD0B!Ut$2dCv>2;x$M
zse{u>>Og(K4qhr@V@4(negVtfn|GP6MO|CLhASaKnPh8-)KS$7x>AKWyxmYee@fGu
zMKbCHPaP4vGSAJsbIE0LZ8$b(pn<}nxOFqt=|OehkJ1&Ga>sCbe|B#2bNxA$$ts6T
zKQx*c!@Gs_IP4N?ePj>jO~k1k!O&A7*mCqCgM!%TR<Ja9$Q8fNLfSF<OmSk|ew?PM
zZ8ARQBR+?}d-)*S&B_%e8D2Kuf1o*ob<>R{-d1Wrxt)=Q6=VZApA-g1fGHk6Ly*g>
z-*-fF2-2|AW|V2imUm^<OV+bdOh2`2wrf3E=hP_6L7eWl?a)dA2G_i3Q^Tkkq;63;
zqraYtolv}x^(d{1b<DYOWDVIeb;{@Y{$gA1sP3yK=$6*Wc@ed{(0Us1f7PbVB`H$>
zOQ+4>8t}B!@i>zvq}^<V!?a!7bHnj&2qsSMQiqYuUN6)!wZ}FBtMtN3n!DvkMd7|!
zs#9JYCNaieMAxEq-H)(bty=%(1=w1yG}gq>M*cQVbThsZ5*<0U5?^1%8*?nKM_M+^
z&VB=px*g~Cu+Q}<B>LQpfBworoirRgZqM;;(Hy4X-r#1n*0+NAxOPY(n*NEUD+juq
zmCJNytOw0M9Z<-lRszed<$Ucm4!pK;LmB*)l{I<CB`ahM11q!Iz1$j=U~abl&Ffu_
z^d&ENE$H91;(c|{d5*Sx=(3VtpPLlW3959-d&wv<q;R~il%!m|fAjtOi@h<DF_py4
z-dIy)EXM!*;h_qsz>F{FHf4~99HH`F9Toe{V-z#vCgZfl)uy&#$xlT>D%tH~(`Ftf
zSzB(n47;bPsWas||5|xDy^_DQd_i6|T8SWCqsr5HYEdyovF3j0lCd_6%|OWwE?#o{
zyl>9lJzf3xV#$?Nf5}IG`u&P;ptw(aBt#-AbNKx8;nUT3?OU5bBg^knlj%$6tgSha
z6Bay-3A=ZCGLMV3c<DkM$4%>Hj<mKAvp31}5Lia9LS1o~|0FM}t`y!wSSTXbPEMUy
z*%G&eVu&qbRYv!0nJ{JH{e?(sSsY{pMB58-c6rtFMe`us5H9SMjT?KUFJ#Mf?n177
z-k08AJr{rcIwHl4{T<w?Ps<Y>zm-T`dMZ9j;K`gOY*|6XsWctBy)x68bCDa0+`Um@
z@F(4IwSTj+SCEOwjJ^Kc?F}nyg^#)fe1PCL{1F!ZQ8)dsKu}Ocx&U`~GlE93T}k-I
z-SE-ZpZ&pNOx{!v*8jkPP%x7!A=KRnAWjM@R;p!xX-8>o9^`@u^%Akt)XgmsU4OOt
z9dE4Uo4%A=87x|%q*+ng+Sg%7l-|}#sJ@w_d6#4I1DT(BIBHq23?*x9x_)YRf|7?6
z#5eDrjPKTSb1ix0X%2%tZWB<)U$B7x15J)smti_869X|aGncXL9TS%VVLi_Qq_@&x
zJ<19Lw(0$s@xDP60x~m~kq-eAms`deDVGqbC<K>{7%c~v{A@juf4YcY+5UR{<C)B0
zyZauYX5pP_tTO^QGQtbVC6yrdo*F3(F0WZPU+`wQ`uYmG@M=^ShK2F2;&hND;#Emb
z6EEn-?9!b%>B@wBVJ~B^iL&FFeSF?VHAPZ(c?+d&{3@x~%0j3qqmg4_71ux$)AWT0
zPzQMyWW$m=qG_X;e{9ng<uDgnEz`0|0i9?M9Gb1SxZ5Bpp&Whly5|z41xI8i2uGnq
z)*_&+Y|cjV1=rf`Ed^mD_hoBwXBncLtkW}PleX_V4k5dNO>3<_IKRVw<C4kG4=QT6
z*q-ULJG3O`;5qE7y}ny-KS}4A#P<+qRJ8@-bNob<E=mzCe-?xKfHa|W9nw~1Q_g1;
z2_Jl=y7OI(Y~`X7fsjqK$7Rx9(oFWtV52GG?W%fub$#~Nz{DI#ATFDCpU&b7qu}{2
zG(;jO(|oyU-k<$cf3->^vY!Q8SC`f@7m-g^Fne(G!cdU~OWl93;*-rrb>CO~AzI`v
zC5|p?pz7xEf2WO-Vx<=y<dh&e5JusP>g8xOkk=wOH4uLaqrEdGt+a4V159cZ4=Rj+
z=#%D{4V$}Tv&cBCj>Tq;@NoMYTg1R4xbpM3F2`bGpZh0@R2LYQkj>ry=Bl$f0%Zou
zF*V56NHQmdu58GB&!E|M9Y&Xg!7FE$m~JD;mUg$@e?+9M(C591d*1D4-cp;_{DBf{
z>77if+Sl)@AWsBJ*Nd=rBA77Ii#*#o*94f8tQqSs`qZmmfMuO@X=$7)2yInK_3rt4
zqpV&1v)*XCY9HL{TmH@jcc+E1X24x_+<BDBOQB*Y_Nizdun&UKofIUO>y1;ZKmY#x
z_3F!de?xwC^Y-ga`gINTI;?(w`*{{WIVT|HP<uTYCcM!*fvQzmt*Eq2dml4<|C4#o
z#a7GCk78ApC2hOjrN(`K+<I283rt<&{@XhFCuI*d=ZBOL`R3&FY=@#WmjsYjqn2{_
zGT#4ev}swY%*yq`cRt6Mz`~)DJn0-$fov17f6p4$UDS8EWihf1il+|Kkv)HMD5+7v
zXC+Wf0#&yndf+ZzAq8uzYc7K>{$94Nlr-<O#-IZW`^p*HMQS4^pQHA%<f7}kf$Q!Y
zkVW@JhpPCgyKf1y?HBrms}Jf3_m${Q_z}*?1*72+IrPkVnL8mq6#ecjwKwR+apzw0
ze?pr|$_GF8#wrmOC46~sl-4PzOZB6}jL8&`?WVmKL54>#T*>JP4>u~HCmL=ns3Xp-
z*yLU@tE9^VQ-h6TP3~}pwf-?oUT}KKCNF`W%H;FvxJyYm;cN>{o3MPJ*_MvkQa|O{
zR;$jUG_;>wo>v`RW#if2qw6VkajIbze+4C^d#qs<1qUypLmxG)vv<>E=g+MX?lXWQ
zgOyMNoEt!)=Hh%>Q#{}RimVlZ2OK7bJlzPb#&e<pR345Uhpiiq)z}_;07X^_<=w-%
z-+a3)E!t&UbA8x}($qA`pLUu=+TuQa=A4=frRv_N-j9GLhB>T7Fuo*N80J_Fe=n=J
zv=bY@di&cS>kY}(&HUWSqn}n(eqnj)g;eTLAPcIm^Na!u?ahGE!f;4}5!XhjheLBv
zwL*S35KwAirA~bR?WEHU5|7iKOA;p-g@PmrmAlW??X2xIl4^w6XF^cFbF25WYS9`U
zu2v=Gyz;ft$z*+!=kr`lJerjhfAf<?bZbIWh3mt5<Gf$_d3eaq)`ybvW1fiKE+{?-
z4^kCKAQdIjZtjSlaMZkv1G;U;nuzX1@@9bxLK78QSvVYVX<p~2fGNaTLGnV)>p{mV
z1sWINP2niViz&x|0yV_fz$Oe!?Z0;vOkO0%Q}ete`fUSBb^xP5T)#+E;V4~8G=Io&
zJhWtjkJ2;<JbEjowIHvK(lpNXT$3zmTA}3vf|zq<ZX@^EXaX5Yp=?akD3>O)`FK0e
zhGHA!N;XRdRZu|1=5{e5IA=>PFBTF)5EQ5+`8E#>CJf$I^3P`k^k}er$>q3IP>^(7
z=EsmK{*M=>Pc}RcNKRfoa(h%Vy(i}^+y<c?edj2=AU<*uUecUs5MDAF3`0ijv@q!D
z?l=<u2cdAG)0Z*EJrw~pm%$PO69P9jmmv!QCYShgJ(~e|my~op(SJ&SulY3Oj+_$g
zg)8K)&~)yX>u}iS?0Cv!A_o)m)#oqe65nlTGeA9reSRf}#D3~FkJH=I?5%QSYeQc<
z&A}>88-F<*iN~zzT<k;A6TysMTH$30utBT~4xcV+?kt>q@@2G@GacE;N_6b4N}Ipc
zm;u)&Xd{zm&bJPf<A2QHC+97hVO-=bwyT3sTdJL;w-K=}*BWbrax4Y09!`|j>Y)`q
zc*75*-_v7%w~bB6@+mXP0~yPv)!UK~Uk^PTm5#CYj%R(yn`<@d8FiU3VSjSAb<1V|
zW|4)ev$nxdqm{T&8&9-cDf@(qRzA>p!whxmka<wDjhQEg8h>Upf0!P^+1<jIjz=zy
z+(1X;ce*I{44p*sG8fctNi`d{$~Zh82biWN#hMQ#5~*h{RQm)re0;9HeR{t6=ZpwU
z6y*{1zrNq32Zq3nJ_3I#l4*YY-F&_IwO+L<J$l?2wz<{LSzC5NPFe5=BB8W7HHMSH
z+!M*3B-;e^JbyW9?OKBd?ov$rdZuk^&?>F%syb&iH#6zG2`8YBIA`Fxfx%*CqGNZp
z);^93>f<0gZUeIv?sbf(GvggcTW?%%*IPx1DRMH?ZrmSHek1b3sbHM95T$hIWAf@_
zj;WiLb_~ueb>Duy%OX520<d~liu`-GsKvALrHQxS?td0573h%7xmP3dvYb7eiA(gL
z0=Y0QOh!qh>A1j^Ns*`oUKAp<7)oYnu`(UX87=65*;R(UyL->fw|X%N2&ZCgr?RlM
zAvVcgZ!Se>A>C#ooj|==K`2g}TuQGZ6pa8HwL)YKp+<SMeTB#xL~#}gXI||=XjGou
zg>hjD=zosfbX;bz1KP$(suWp6D74T_DCZE0T3jb=8$vCCS$5^1FGQ%3!ggrQA4d%>
zNSi;5ipxw%v%1Y@xe$uvqI&C<(Nasb7#LcQvf_mPJeD{pkR1B9>uFrE#?|W7u}t*M
zv%Rpj2&JhlBePs#Q4JW!?AjmGj|P@?`iNwl>wlDGodMN6F>NK)H@?fWD8VK);dFwU
zK9l8BQ#pp(G=dVak2BZqa4N0+X{v4ErSCiX&Stf*5D>nNa)mKQsp))-vEjV1;jGYh
zuP>i&jl18ClTRVE<i@=H<db7)%h7dVS809W<dacSot=YS4W07s@#L<j(JjLf3KKE!
z?SFig5d&_9aAxflfxdeb*e$t$p3}(eO~z&bq;lHq7@O784aTY7IQdDZ8#10m4nv_~
zM)cyz1}S6e7`)YFgQ7s2yV7K%ZGcB($jLx%jpF&}_o9&oPDUz3^Ex)Iur6WGoN4^u
zj5A203g*(oVky0MoH3T(IL;_U{^&(q<9}525^O!r5VU7xuv@wx3U42lm7q1&uJmf<
z7)2pzu3UMytR+H3DlLC5AGrWRCMiXh&s?m7N*kq0(DrjLWBcQ}FvV;bWqMbN)8r$-
zmdWT(r<g71K*z}x!}KX74y}aLu-Cpnr_Gq=Y`VDz80^6peh)4<Ktf~>#5r`orGIRP
zI7e)IHhVk~Ey@e+aYaY~KC=hnphfsMh48IoFJhy>c$YF+A-#9(HJ08u_9{fym<)$+
zkDY<O)_K}QWdRj6R$A7mEU2g}U)HEBs48}=OeP8;FGW_EjCA{3Tb-djR2CH3cwvs&
zxUK`(p5rtkb74C8b4=#jY1(9p=YOcobeq5QYL&{c!8q&z273<zuzP^P0dB;w2jm>l
zzB;o9<c!YLg<T*Q99C@H1@gp9wnEPunK7k9>K>F2oo)IZQM;GVNUMc5&PXe%pJ+x}
zS++K>w1By$GpTL!%JiFovu`%8#KslA=HaDOxN}h1f8r?mAK_ALatdW`WS2F@Jrb7z
zt1c4(GncVH7$=wiHa7*AEqgr*14r=%mpXerIDb)RnbX3`?(wM;c){^sa10vVSs|EF
zqI>+_{q%uzX>gr!5$8(bviR+A*LO!H9KI0vLSR&gyQ3w65rq+jZ=}YZ<C6Go)7T}!
z8^Al`c$ncH7jOuF_sip-4?Gc@?*oNOGiRwYmKf+r6rQ=Rp_tUp$&NF^<&;+4?|3sE
z{Xp>nvT$Nn7Ni+D+h7`b3G|BoH-Sg`hzmRs_}1W)@Lv0VOyaI_*P?uX7sRAxKde8^
zOl|tl31X*9f7iNK1NoORd_5F@?Pld571PFdMsucx9OPi{gbb?BRMDxYLR%=6?1zLf
zh{vWkCL&xEct)foUYY?B^(;jx*o2D;){@|Zp~ZLi31g%uVVpO2$zkSxf-wqC-%0r)
zsWfAYJU}-eE5S`nlY6Fy#nYPR#JsAnqyBq+hNWebg{&J`UWBc@Go3Gguo$GopmP&y
z|M!bw**xNu(HhJW)~gl<%-+FmK}t#+XK**3%U*`FZ+<MaoNQV<*ajagELc&*>=MCd
zu+@p3B$up>fg3WV<OjPvC%aMovrIl=uvwja1!YXVNb42Ti|bUF`IQPCv~wv?d<xU|
zO5L|6Rj@Sf8-7n0lh;E?a6FpA#P>bRJq^j!y$SLQw}<BG)8pZ{sa~XqQ&ZiS?}zY$
zgl5Kim^<@=y5IlWJsm!`vo<>uFSpgWdblxK*PhKV4ezSqmx_HoG(2qvUzY58^n<~I
zqj$UXS6n}*Vh218u<+d?`|bpVvEG~V<Ss7FE|UGgwtpvUnSgc7`}g-3p*s&fNJE$U
zeLXjSnQoZ@8u4yb1kNxru7>*hZq*XKVxG)I=7oRMGRb{jYZ8GlVVAs3Yf@q*IKM}0
zQb7e))FA&^6^tYSEH%X6*O8E61`O{o4XL3sqj#Ey)X<sQWgHTgE|l779MU)@rD&n}
z26^yQ!qF~a!wO+mEh0(lOnJjBZ{)yN5vdq|o6cGL=xG+LKKm&XlDM1wR0SO^KBQ=t
zHvANlC0Rpn`YF$Nkj)&(lAj{xv7c5TSNJK1I9JtBTb_!TZrQoQQ#nRl)KFWVD(K8j
zq`t>fDMo^B)SFZS64iV3CKcR10<;Qb%Tqyb)L_|gRmeKY?J<tt=j^l%X%NG6+G87k
z(rD(r+i4rpVk37Gi`sHj*f9~x?2@(Tj8E5y)T>M(He=d08rpJH8o{mEW$h{cBnAJ=
zhX^r7bLX71<otWRW&raLuANBw^#D~{tWVWK#l<$H8Fc`fL((zl7)C=F;8%$ZL4?tA
zf-i*ez@TyQ`@5szA!rpm{E{)hSub9HkViZ;5hA~=)1rxmh^P_RX&qHit$RCxDWqXu
zZ3%2Oi&VrMi5cPKY=TnOVAm;usZ3^3b{#kNKqYrei&9O1oKWYSf!&w+(X-_mmZ%Sb
z_swzzF)Wlgni*31y;VX)$jnH~GT)r3f*ci1t65i74v9-<y?Ld4wwXQFa9pZ?x~QCG
z<~(*Yi&WM#X~i29uxm`OS+gq(5?5B0mDbFXdHt%gg2nRJDs|M8=lZd@t$wTo#Z>FM
z`f>gBT>aP=(!5VB>s*6*liIP0$xlmZYlo|Jj#bS;@8`Q~vLVF{u$mToA6dT`k4hLO
z#Ws(6w1`;4GP8$WTbZ5oNSWM!wH8u;{YO7xXsLa}6Ndc;7^ZAb3$E3nQ_S$Iby$5p
zqr)6M<Qk?fjzY#Pa!HSUTvXLvsr-ImS&i`s)%Dtld!~B(aw|+S2S46Q;s*OrriiMh
zCohYh#9QRZ6J0dkd~w6WMkhlkZJE-(IM%3v7dA13ltS1BdNa>44nZ(~wNQeOCL#EX
zDM&aZ3C6Y9S_R3=xDkd+l?ErSQiLQ><6A^=$!;WK<l^b&42gj}CHksvu#ecDAoA~B
z8rCom61Qr&g1%bAHPmZsxW+co@F}^THuTZAX&Yh<74Zh^HR?7pb}Y=$8#`IftLpyr
zI5{m0M=F9`I4%6G1K|&UU8EkuRUTW`zq&9if+USGHXI9+Ku`cdAKw4t>(dWk{vsFv
z{uv;6M-aa7uKV6UVs0l%>+W0sY&iozK#GJ5DP<Q5M>ZTRX1HPCOLw4fPNAeMKKm9%
z5qY^#1sZ|F38y2Zr+NNyWS}5Um`5|ZZ{=VnYQzvZ8mI_QVU`3d8y*3lp~7Rh%IEwk
zp#Ih!p1Q;5L7HCSFC4C0Zsvpuq{3oZ7XlrIJfO*!R)Re<e{^UoU{>g0k0Ob*7wh=@
z%l8EK6+jQgy`SbNlxG4n9i>EEV&X}7Jjq`0aq08?bzV}ZJEe=PEyf4qxolWY8FOCl
z<??DK1qq+Kg2c$`apd*$wc>%2w5BV>(}rE6SVD$pxg5l@6x1aSI22p-NR<Um1V@_N
z+@EFYgKN!A4;eP4Qc?dsEwluLpiOI?mllIPLw`jtcKC&KBJ2-w&oKRXs1?NfC4~;5
zv+;g)KUt1uL_Bn_s1WL$Wn)qxuK36mo<tFr5A75-r3CS9Rbq_LYnG%GRWK=>?`Ik{
z@nYe;wakQ#WL@^J7|xT?OmWvzP!xu<%!^Wv;A5v5Raz<1or!%VACe1)SM^j?Jj^)D
z!hbQ%ILp!V_0ICw`2RVbY{wh~9g8Sv?0H(OyNPg2;l(pL--PtcO7TbSeDn4D+-&q=
zyXWwk-K#1bPODZvuhl8%6^jANn>5OK*o+;{u;)d5i+snH(3#e{XSLGBGoQDiczrif
z6k7~ydj4;F?+OR_Uos-;v<hWzWOH<KWtTa|Jq!XeH<RHQBbN?AJr9>BU@Inn)mz(+
z8@UmE_g4_e%jRL(B#XBS76$BfyfLtgH9QWG#26Tw8QGS1R&q2whBsfIQ|xX^veeF8
zBw+&hQr*LD9u|w|RFOT_x#*Qq7XJt3C>AR!O5MGrf}vt%75ksEO85&4ZP?gi;a@|(
zm;wK$R7|udYafY;8x*<*4knj>1{tpjSsrkVULqGWqwCbYA>+Ml3zY1XGrnNSq?}C!
zO!fe6axsD5`_fyE$_42MW>UaeI%caJ`>r><jB~l0vm)c9H!?wyD<H|RB8uH(UXhQb
zH{hedr8gEIQRxk1q1?B{%b2^ith{wxUS^bc(H7GxPuz9kln*WzlN5S?FvU#Db8)#0
z%QJPzJj`s+8#$|jqtd{66{xs`;G^JeC?*;NUplr<g_vD2Nrl`uz+;^s2sffBo+KMp
zbf%b&FxQkyZw6-9fLO)Y4G0g%L|qq2#oUd}B1D)wIEWG3In@P~s00OJuql+uBK#~x
zUq<Y*!faFN4Ue%YZON#AY^by)tFkTeNaR$e&O2~H4x!xjXb^M3;L#v;!-QjU3f&vB
z%DF_U#2_2vq&NJGtu5X`=&~zOLKGAebt8_7eF=4VLvifh;BhyVSSOFW97`7-{~BvL
zJb1)lG^IZyD<k^SA>L}Dz_-bJu>sc`SIn{uD8OeXx$??DVyDu7sSMw63F+cYYr1I-
z_^IuY2i&oOUu;0IK4$4w0?T%5fMlj@?uHkoJ$K`a;#1oVunXmGgt5FWmYFPTz%$o7
zu^VD>K<Zi+gv|NT4?i@o|M)a*p3LWqRy9u-%ZusqjRDn|)8=o@i{?jMpu5xN)%3i5
zgAR2_h*9Tt#QL^>!D)sY0$FXdKYi4kT%Wb_!k=gJ-<l`&_2=zuF>g+qzrT7Rf4}?C
zwx6#4(lqa9?T70#eZKg;`FQ>Nr|&POpPOGM?T24KwJ+xFbop*_KGh#Do3qQsS@YRy
zXS8jumgmjW<?Q19^s4!BF~4e;li9p&&iY&Dv*r2qti@x0fBHFn|1;5_A3uJCKO1A+
z0J?kNw!7#C9<3H{BCNCor=zS{C?|OB9i28$6kdPQ{IqzzXr8I>YyzDH5LU7bdN!7m
z0gQX22pTa7`ZGFXxkCc9N6cdxY8%oH6qc}Nqfnst$(xYw4+3M-2-yn4o1_i+WDEpW
zPsExMhp_d3aG2qJc%qvR7mXH@ALawHr#WxWM>fuYv}4HLWGo@f;~qeGOg@X8>4X&v
z&D9x%0d3O8yDbbJoM>$H0bK4xyEyM*HVEDaAq*g(Hd;t$3`7D2Lt;UMF$Q44Ozr@2
zb0%=k>gZf$0w}cx!8{NqSX|HBx0Ckm<#ckjA_TjCB7|o$5DdepSIs~E`7g+!N|5gP
z_2uR1?iS#Box!3|cP*LG4VpS&#Shk+DxkewEKfdRP^vtjnwOIn!<;MEUH<h;`_oA~
zX{X9|H=Zx%EjoEFjBm0py*w{+>bowVi;7AvQ!Yd{AeQ|_kPF*gNY2J~Xoe{?W@v`s
z^>_<^E*?K_UM?5sC)4&#^YYpA=JoVTd%6}TFDLJ(;_`MnZ?8%s)xG1@^lEXvJfB|m
zNt@jJWqL82JY9S#(G(%&i4mADCrdmlde-&WMc!4PWflmJ-gJ4UjK51W2g_B9s^zN{
zsuio1#5&!N2xO<h?kf>sr-|K>NN3Q1kTp_&uaySGjD=4})_@*F)qoKYhtYtZewlUR
zT<I4c*KGL6^gDoHgOG6JAPf>|!~zO}fSIJov_D56ERpKD<h4ba0G8Ltdbc1r;6&=0
zq}VVXX?WU0RATjMDH(eVNOHAD`7wq8aflSsdJsWKHloCc_fw(3x)f7`3Sm*0@Q_4*
z=xlnqn7_XoG$?$nL7CTN1A&YzewPq!H0#v`YtSMkJ%GA=Lr~UbV^dJpr3MrM3XJ_l
zlx)FnNey05nx!pCyR8V3iCZEhMfqJsNcI+1BE(f(xQYu`ap6?0TrKV`GNxL#TCQ5O
zM`7|=%k~f4rzLtoxgu0s%P)9L5kU)o9O_@T#@l_?CZxrgtJ(xqKz!`lqyw?4O?HDg
zs4hic)@3aVlG84RBkR%tqSvKS5Qo$y)XXu3l`tVHQKlSOn0gRZnMOd|OquTJ2-zhu
zYYOd+rXAVRVI9$fppdX$-6#mOLC2o3F0-*%sHkl2r_quU#nqmXvv5lE4@sndFJ@Qm
zY<^xV#b`EFTDG3QT`8`&r~(Z=50MWc(Bqo?jf(FMnjB@HeYi;Ff_J0{_mOVrwxzot
z*{!A9`IU6bo_jB~w#u)q@@uJDzFMJLaev}$B_|~?kP)H8I!FNva@WF>Lk5N_L6_j@
z%z2MxqxBZ4!D{Q!1`DA$^48&h0Ai5mF%XB8XQ|;4bAviVA2ZQ$#d(0xYx5X{14=W3
zLe}dbNcb=B*vB}i9`z6gT^@n(Em<zS+f&_m{q+A`6pa(`eK^-|@7c*A+zmLFy#I=G
zeOTw*R2ei^s}@AA@~ey0s!zE3gsV?b<#4JTPL;!HA0<@3Ad>C&^7|Ek5i^*{+aoWD
z9>kEMje*#EiWa4QT-(f#Aw0Cr{2@hqyPVG64S8BJdpzRvNsg4{1Vy&+9Ii9&@9;iU
z+NBIuLI%PJTQYm&WpKVuex**{TAhPGt~^x`y*`bCIG{jLj#@r_8P+3l&}@$^Q3HrU
zqsBlSvUVgyW#knxQDV)1>Bz)1fFKZ*eD^R0V&8|1<dY6z#Y=MPZo-4&rT5Dz5{Kp6
z_QPa8(2`4gq@^gSax7<wCL>RZno`_8a`KrYs_bMUM8O4!=w_Zo{J$z#+TLQ0@c-rk
z=4ey-K*EN@pLK4gSEFa0o5|WnfxMoDt>s#`PNG8CPND+cEr}X`&R$CrMYSNMnlMpK
zn5ZU9zTWezdo_vjHHivle`yjSBf;zs`V@&V>-M<%R6+FmGz#L71(bzG+EoP=9bvT|
zyM*dN4Ei(%V&A83frTOM>MJl#IvMvcXt|x7T~7D<4OscK#bDST@{epF3o&~YkR*q^
zBQ25xe^6v#gTG{d;Pw4w10;K?z(D9*&O1aPd3yr|r4~XzaRBXQAMI`-zO^5)=u0Z%
zw0srmO9`(W{1L-lG2CJw*)v=aSNq8DKV$62kGfyaJ_6Y6^Rx!tyX$%BZT6oUdM<yp
zhaCJ(PI@<+^`9J}67iOA)D$wz5Q`f$btw8Q)4{h&Hso7>zWrsJckPa>Yx$<iuWyw&
zq`-05bGW5qC%diKk?*G1X?@!x)Xj-dHzz{X-&i*>Vr>^|yI41(W7Yjwbw5^(k5%Ji
z-s`r9H@SS-gvlG*a5yNPL?egQQg35@J5e>HxD*b~C*W=rHN2)`Sn3agP5ym7xmZrx
z5i+@dMH#h!eAzDF&M^6|%H6U5e4t5pNLf%(B6X1+PdOAZ@(V4_%30ZQ5=pWUE=Ma6
z@I)HNU~Ta15;+$Dg|o?Nk7CLZ7^3o4lpV8!)nGCoXEsQ7cFNaV$fV_voU>-sFp^n^
zxGm`}L=>4BxI;wxV~pQNMvcy+fQViP(7Z!BFWEg>GHY^o-TS;NhG2AJ`L2w>5w?70
z9u3`bqvN)-Aq!J}D<?ys4I{rodN03^gY-IDzT<4%oDeM%0+Qdl>Io5HuX7F$f)cMT
zC+EM_tp4lNDw&bZjsF6N3=x;XcQz9QF)}kbm(h196qkE?J|~y(k3F4#frmJsez^b7
zfhS?}y^~NG=ACY=(+qMX32(S5sNP(JX#xvYT3zU;4b$FL)LetFhJ5;6oaN2MPE%XS
zcqn<ldVHNo<TQz6+}V=kzpod$ILTZo;;T=IV#aEoGpWZ~0Z}7x0d*sjW;3`1$DCJJ
z;v$7+Du`)ZpvojkU<z7)6eMC1N-4p`sA(u?xG2e_tpxk#^-)UsiZ2=ga5Blwlofg(
zo3TdrP~+h(_F4}usG}ulg*d(9p%7P=)T<sE80VoRS@TfM9G4}@nu|&%{79=k9vYYv
z_DWo&@Jt#zjmwPhNX4Y8pj`7%a1k?5&UvV1qBhtz4>bajWtR<qppmf<Pbn;RpCL5o
zw9pZdjX6n4u=)h27n3s?9<5nUq}*wdy>U!)RZ~SDpnX8Nx2-3%!sMMrTk;JW5$Z;_
z&qNc&jlPrLj9x^usUdx&4U6hXV*sVDB@=p4ABo*S1~<}2w{mH>%+m!JW6V=6^Yj`#
z#fVBID&rRoFA$l3jOpx^kTucp30cBTN8Cih&aJycn%`g-2~u|Rht^8bf-dm_9~lx)
z(f<~`K&CPJ`&~;**#n$XBu9zDoCgBUz@!kM{74{U6JwRGBv2HN5}0>Ih^_;=af(@?
zX97C5VB@`PIMcX1?Noo@m+|PwyLli}@Gm*_=@X?$qWxQclfmAV07u{w=+huaAT!S9
z{lwHWHj6An`due~A%p1FHuQK&ixQ-kSDW3kV;YTYfZIVJ5z@aMa5|@{X{~EF^3E7R
zHhHKLNITR&$U{qkBC{?ACjeZJ>zaomedl)0Lq}d#yd6y(mGfONUiGss$X4daaz4YF
zkP?Fkb+xpAfB%}GYKYcEylyji#HO8c!GR(hN9$y?RyYHp@<2uz>n-j!AlHhxP5$z6
zVA<s68rhaJwupP<KtJglsQ}d2+_aKIv^!tCNxq`mXE{jXbmox4M9Zm#v0W&6m9z`y
zJaFxeMx3|EDVLow5@qi1YtEb2ClF>dRN`>HbcqUo#VDiS4l|B!ALhu`2v@q@RZs8l
z55JG?B~9e^@N}RbH1=uh0$iU3>;ixM)I1(O)U#IgF&`H9aV2qUX|Ep!LCXvRu|~|%
zI;;i3?MP;Sjt1bDcA>@lxp(hL>DS`53Wg!)c|K8>gEF^FOIve|-p7-b^CriK8EDR6
z{V0QfC0C)v5QF2Ty`t3TLCiRJv<UqohyKv}bqvP5UBFq-QAk<e9uRR@aMxC(CH3B(
z$&h+u&t#Hhx{MM?n7FJ+z6WN@jGSemq*<=ksEReUGGHT?%ViI@vc9BRDb<2vLE91L
z+v{a1u*k8=US~C`WVsJ}tR_*Ca3@NV|E)BCSx~}OxcsurL|Y`rW)~BqsZOp;#+N-q
zQXfUBJuWCM2EE$pf)bAFjoagdv2rwGrxQvzA-t-jl9`d7BtaOoymX;gA2SKce~c<$
zY%z)>@jy+i00s*7+|F1MOmVy_#MJ88+G$f%zbkWXTq~&ecG`y28#`^2By(|1vw$ss
zyCB&R*F4f{RZ^`*HqA_}&W6b5ndYLTT8nICGe1IoPh?YY7=5MPB$A(iyl98rs7Du+
zElJiQ8|<blE-RtUpdRvjoI@f<)oQ12NQ?4;>pi|9tzjYfoxUL*Hhw?ysI|b09Tut2
zF02D*yj~+~ku9`4@{nXLvSH`3gf=pNQ@ts$!B3@gyB0@IdlhF++Ar_WRujpvsY6>)
zt#xRx@4M0|qmd(iR6T%$!NqS+N6RTE7M%W)vEo|J;x!p}1R}qz^nf*^gsTGVywU=O
z0{-OyQz#6BS_5pcu2q_OEhfOp#mct!wSudRW?4A^P}nk++>O=?8~|w8F;HHAe>JWU
zhEX%7nqgyOUI|K@^jB-l%K)oe@xo<v#q|rj?-Kc;G4chv?<J?N*=Zd1M>e*~sRBZF
z@%3BfR01mT;yzPy&HZq3T|ZoMB<Wmj?1z_cPxZq`haba}j~D5ME9Ou^)RKPjAx_rg
zlz~o8vQ~j93Ft;Zr)>V&)K)NmC%k-BZPPnR?6cX)FNxNwNh)f%l^dO=eEXk%h*C@Y
zZ$DVvY*Q2?LW-?6E#986X--LU-!&>^lIPuy(voSr&e{qUs&TfKVkLK{B2{5+R8)fZ
z5()l%{)8NxK3l1gs*NrNIHl9ExEbiNUsbF1{r%{a#GBYYd3ZYf-r&7|yC%L2eN;fL
z_E$s0f`qr$IyxoOLIl$7@aDHqkKaE0C>RL-BnsYW=qpgvKDCb)N=j`^^SOOS3IN?g
zgS-%191xe(lE&|ut{Zr04w`zcO3J}&pJ|l7(V)t-C#HiPuaN`i`TLPhf;eFw&Fnr8
z2eVK;hp15}1wGL$iIyIJL7%DUk*?u$eic+dH;2dO@S#I9DEy<dM#t?Gm|!ZL5<@_{
zAymc4L6Y=%!O~gR<ls=vG$x^~)qnuWaR|<lG19!9rg>zbZ63l@d4@cc@AdeM_LZV7
z9;r}fCN#z~Ck*+cNRYO$_J2P-C9+Q-x*OcvX^ui+8DXZQ(6A(b^ub*50B<G-Ot0s+
z^Qt?alo_Oe-4_b&*%Ga)%z0{cRg*%556wWtsOquT^}|+pU?gMo26*Z)+`|$q7QX6G
znTzpB0m)-7(xEbFnuycUaF-9|ZF%-)xE&UrsT^(J*E0v6R4^N3(q0hha^ZYloZWYl
z435zA31jzN$>@B4ucNYq!q*PxE9K664(&oc<y`p{<8u0XJB02S-52y@4Q6F2y0|lC
zMz$LF%m_cXgHX0XUy8>s&uu$S<wd+-+B<znp>ErAt@w^_cb@j|m0ppE=Z1w8LfvF+
zY}&^a@3q2H%uF(FIXiJFLx{f?ZH!30<Vi~Df~K{2nyI&E#Ixno&M`~-$TId@vOzWW
zfh~1|wg%aXuO%2XpEBly;!mWHP$GJTRT9!|q$~X^g^=8X*z|$PY$8(5IUOncu-8V)
zU*rE@@>S0<CqTzO1L{BXnE%?dJw@OC4N!fD=KlwxA3evHF~&U<0yHp}(IG4omy4M_
z!hg1ZKK;7US=o0x8<sS6-YDqPYbtST4ekb-TC8jyyJwT0v8oAJ(fIprWcUJlnfK`}
z#$3DpnunZqN{k)WO*1_7Gvswz&n&EiFZ`hEmXp^yxip$)u_4Pt&O+YGST@aW;Z+km
zk14hxaHEyM6~s#O4L&>^{hYiSXv#X~sDE^fwRJpeLyCCv`jW@HZ#a*&cKK~Gw*5Q}
zyUxVvWhyvKhef<q&eVDXja@9KtsP=1K{?AqEOoXf$21`~dA?OKMS=3WkkyTQGX`r5
zQ(9}!a$3yqLwD_`?WQc9H0vA+N^EWC``v&*frjn28G|f4PpjyX^V989_1nj%&40g}
z!J90~I$%dLcBC|aechxNh7cJa0WB5DG~fPizHEN0XU);A?`@`PW2c>i0pr-QQx>Vg
zY1p<;RI~o7fcvrAvQB3kjc)co+HF&eV9N{(2d9XubqO(}4+T<(?_*FQNUi8{_Bll%
zW<)h%zto5(wP8tNsNJ><#nUx%PJcx!M{49ru|5c7ku6S36_e1K%>fy;){8=&I}J$7
zl>#%0xO(O^119RWS>EdD_x7GM%P7AOb!Dg)F=zc<cco|hKF>=@vq5&KDSIi5b=5`Z
z>arW<y2tzt&4XF23f?r<L?x&bolOtgR!Q1%6XAabEXN)(R+-=n>6`y<JAWr44^N3$
zQPG4_gtG_~nG$jEyRBCFe7hDZU9EDzcWE0E&cU8BEox;$>6n}Ew|8Q1e!jgU=Wl*2
z+KC;2!YifxAAntNs-KiXVs|>0QhL!;M&8uH)us*zYVkJoAO-qD-IcfzZhSXh;%JY=
z5!%9{BZa*ZcGDP{_JRPEzJG?*FSm=U81xBR>ToSZ2%VGiHr5f+dU+v$(KDA)KNce-
zE+}6?0Y22b)^7)Si($N{B946@iv7>`BahEyK&urhLNyUug2f$UJ(9lEQf=tnJbc>p
zosw4|dFWfF$8qu-C%%0*GSTkNwy@SJol;cgn8)oMI+Ij~bpK>OEPvX_9)ZvYE*_f7
znS)6mn6#t)5I)J4+rDY+4+*BVi?VHgJzRNyWORkI(-jI<HFLk^+ZI^*GSU~S)ly;5
z45WQCT`DedXz9yJxzhg7`8evg!@h;+?GWzD<9rje?W<_tovj>3*mxMgwL)Dvx!hR|
zXPv0ss@Xj#Ozs!laewFQ9%e=q$v7<#hw%ZsZ(MzUur9GYI#93R8N19#yat}jncMh}
zTO%_%&K-egO)3!0;Q0WP#n;_Yl8R$~Tp=qcO!&0d@ghflX|r6d>8tG?N#Qzh^rd%C
zY8rs#RkLZ|6L*2j&>Q#6anioj5NZj~MXW<Q)yhT9ou-p-#(#&*yeR}Fh|pG!Q}JOD
znFL_wHpJpsW#|oZ8-;88y5hPNlS}=a_s>WM;;ce*R{7&v9EdQX?V}gsKvK^fV;EvQ
z>)2Dhy5G9z^Cdk!Q7O5=n6M#b*w+%5q)amBQ0p|uP#__py|3MTS<*ywR3;v)lTbR1
z9O(4ov<XL6o`3AsptszhGcu?4wGx+SShja`KgF{6>#{8DBF^gmOqOLVeI(0Lh#beV
z6e2y#;#`B?)JrJSl}F(NX4!dValoxl&~sf}#OX^aXPm~(cpUa)sp|I?I>su%;}nN5
z8KnFfhxjn%5Jr`VETuz!TBdJ=T5`Se#@nH9gcVoOT7RQ&j8<MswD0Mg87>C-+m;gR
zR17MAj59EpYZ{C~HLATmL&b2~c<%X3OgYbyL`*64hIp7#G14s+vj}lG-E!@QBUu+l
zJ@n3Bjdfwtu<EtzU8%$~37Yf=Eo1x(5H#gVZ-Ahw6uJ^Yvu5@ThK77Tf5cqd4dhq_
zzCL%80&7g{Uw2njN<5S1A=GAjB*`=W_AOI9rJk=&@D%D?iQZW@`FwI`&NhS8?9qP|
z>*jvp971Sq=b^O?o}a-F`Jgn%w@44!{6A#e{r_2yNs(>;?`7=Iqk;_D?ntHS$1*_`
zJM-~r^FL+;0hb}#Jrx2pGnXNP2onM~GM52e0VaRlTiuV`wh@27ze0~^1ZFt=7?R0l
zf!nJuXdVL8^-GZCscixnHP8mN(?7pIN~9!G5~a2GE=C{L+SQ4g;e7sPIF{Y7o$S87
zIlsC2@==@432(`%?#(Y9;SUo3lEh86og)&gseALT`|(an#rT}@F;Z%ckHdG5t3AEc
z<nez5;R(WBqSH%f2zNB@Xgs4Vt`Z-G?=IbYOzRozJ$o77@PN-bUQa*2`R+z0W~B%m
zzs5`b>(!f^f8UI(D=jRCj;#|Usr%czoA?JqR<H}5QX-J)e)~uF%gtZvw`N+^wHK5p
z2ud<rTL?<Au-13XT4?98ptMaJSA)<jHf(=Y5*iNzhyt<r7;M<!x4Fj$!mzIf!Ll3#
zAPh*2|NQoASKWF2?YMUZwdtU0ZU|zgKNaE}S12h6TdG!U^2V3*WP%8q7d+8AKh_8e
z3vmXJH#3&--Nf&Yk*&*F@?I!)n5CeES!h@#7Hrn8O>Cv#>k>??t{Gjju%#<zdscr=
zDAJ#9VLcC<jvrVA+^VR>H{eCm>MZo(cYmcsopTU~zs$L(g*z^;4+D4o+Q;F$y-)9@
z$l18oo>UAQ@y&S^?)7i!Iq>N%Cee9F&(NlgleWS;)7jtzdro^gOItks<s;%9OqCUz
zLL$aO{jOMj`}XaNL5i$a*H{u#sd9hKfAf-cGM0rSyFtdm{-77_%izOGDEO`<q2&;y
zZ&kmddcp=fYBN2t$p?q!c7KAoXHMMeDlY}>u3V+sGV(?QUuxy{Kkr}cS*L@Wl$snY
z>71-_w4YOpx`J25NLf<<`o&9PzJ1@6kkUbT?-vrXO$osl5rwMl%I0nrGMIl~?l$|(
z;PLHUI@m5%%KV|dJI{i=TI|lXcL&9A@R#fk7TYJ}YIpK6$BZ#P^wDHJw)A&6MTJ>3
z&g<g+GD=KwW)@5Q7(2EkX#j+@59^X#mx%$?AX{ntn=E|4x<o6;q+<|sS}%NWgEgZ?
zh!8|-L=c}$8J4UxPeF&*0*HTMi`)_;2yg?`)Gs~lPy>LHQ4rvfbs5?=1Ytd>aJi%%
z91N1JHLhHe-4dC5bx9aP-LxV)3!Rz*cVNx&LgKc~orhb6O0!?JBUj4emc5VwI3tP}
zGFzXWXLdW!YCWoeS11?ml9j7u*)*(*nOFHf>X{(|sv!d6u1t4F3ekV&UH(goA$fJq
z=E<XGu{2SidtZ73^2#~pUODZcr*~iPV9sZc1;U(*^wyRgX=pfDE@Zf4N7^CDXjd+I
z|L1mK0y^w-S*fZb%*Y9VWPoMQb8#=y&^ov?#s~*+S<xch1Zf74+BjRUilI|^mpP!U
z=$?<q8QAh8kC|hhl(l~Vv=0TuE6_zYYG4=1X7str2#0{XW`rrwCnHSHWh2}$#A!{L
z2HM6qhW)|tnz_s&ZfjrcYRAl!L#;`rAok2m1sThVN_PFeBFC>z+TM*t*XFLHgnD_)
z;%=BZozHI>E1ZS?w8xHVUfN7FjVThXPTJ#{v`6Wh%nc7u$?1PD4>P%MAP?j(B};B@
z+c;ZEi=^htC2gFoVn@SKMdfW9-B5_Rvt0Ad3&cdX2jNh97*`8;hH=~sxl%J!pvV&u
zmzA%2qB0O*02T6r+E#Jqw~DjEdRKw#24s*1z?Dmh{{$NCyev`Ihf23KsNV)_F_hK>
zscB3`dBI!}%%Oh*73V6%L(Hx$_I0z$Gblh6=vxhpt%nB0?H5h6GpIjfg;HvdvVj3C
zwjfGu357AMXr2KMn|wS*8q;f?QcQKbxo&*FL`};4OpT0++L%#v`lwV2xD+AVEd+x=
zAFSJGAxHpF4<AKf$H!}{hd$Ke?(0UoC$$?&N>k@Fc1M3Hb;SJCZp!=6YVgeMj@gOs
z%3j{s|JVp&#$f}hW@vEc)X)0#Z7Za~AbEG&8U4C4$V8*iJQInU(6LTI>TbdwQ7UTw
zn6E_9_7G#WU^97$JCX8tGlA!BX@oe&E!<?3{hq5Bn4P2Na+Qi&FUJwa)j@sVRgh>A
zkT+aqVeWs$(}T{a0DSEhaE2&lII4RFCh24XjDCVN^J;1Wi?fcw5_0B*rgNuZOm(@2
zlc$$;W!30dZW^AA{ku*FWJQ<b`TZkbKl-o%kaZQMKQj6-tO@K_3|%|=I4h7|cehQm
zGAS2Q_f;=CT(mzOLkaWR{$hPxTho>$6&Bf!Erx%zfHG=|WZESsK!M_gVI%<9q^G5D
zVM)(kWcgkS3q1M^mS0f%94x<N=*pI#<FAelw}?A8ww<98S3RmEj+}dY3w@Xs`C-_0
z?mRq9DWuH5!;$A)YQwQmdB(TrOa#1ZL%o?1Fr26CKQ@<n@%B2AA%i2^n}dn0e7fIN
zbrgTr?6nN(VyKeF)VIXgR|$G5o;aBW$7>e9nRtiia_a1*dp@Qs8H7bRsHk7ryfrF!
zrR?4w;|zi|seVm;6FpU_^*q2PkjfULNjRyGT1@fug%&eeihoK=F-VVk!D($NJavG^
zE`Uu(<sp4$JxCR1ZEWrXMRitl$dF9mN2`B$yS5!!)Wq8CN#Lj*Av#IzL08*&ejo)0
zsRpiA|1z6h$fbeqM?Gv&*DkB{Rl7^GJ<2;FORzHnnR9~RIt??-IyX6jnlKGK^ap5E
z7av3Dzn`}s_rpo9@>c7d!ASD$kLha}5+P8v_lPHwuf<x<nkdkRsWQleA(xII#nFEk
zPtP@i6iaC7_eYR;Pl8t&L8`bpFoIOAc!Ck6ny2OnQZ3G>89_S8(lCNlOX#^rkorEj
z(w}Z&huRSpwf~C|Bnk)tWIl2PX}<dC2vV`;*a*@(ndS%*1>h{*m_VYS1sOUtffUc*
zB$aF&m_S0IsH^?%Odti>Pem0=b|!z2;vp6JP$_a_CTINsN+AVFbCEKYa>D46>g|V}
zDJ3OXU8YPG&kGn!5CGK?RVfn>ptn`Eu^T4XLZzzT+hS)gY_%jEq*-ZKhJ_5*Q6Sy9
za9Aj<s)mKe7R$!e0c;U#D#L1;Z|sgpU0KCAETmV@g$U@-=-&5iJ{KCqSfGEAcq*o&
zR-s&JBN+MdXW9K7fBP=t%;3A<`zPN4843&i%kH1uUx7?|%&{>zl;6pK&I9FttxzTd
zQ@2Az4`D$X7M)3J(4dm?#%b92PQ*7`4HIph?P<`e?Cp!TY&AA!T`yZsXJMnm0m8-6
zMbgz!2Fi%uMEzJI+|r6r^w586yskdRm$nsR6ad6^^^sq>qK>WU6(^K&E56)3+qH4J
ztqRsVXS!^J3;rUX^6?mg_kGF*gh+E`C5|s`D-qF!Ib2tZ<2zT>xN-I!BUZ3=?3c1B
zV2kpA)g^vq<G@Dm@NK`cl7$N4x+<Ao+Ej@p1>ZMURmt?u!=7bnnT~&pEIFGCYXYN;
zH{lIV@sSx%f_ny(D8-UABC|WULCEe}SBpQiai;?gW8V*+N7#1+?m5@bPZU(raL8Bj
z;)cuk>+9k2p=r=cn!!BhjC{W{P6U9<R5CJ~HQ9hbgyJuT_}QxCLmgy+Ajv=3PzUMw
z-v;b6m4+3yPaNtXvVt|-a}0Hqls<!@j*6it8tT~F_GdBFfnw0}$Io;e@)A_Gu@Ch;
zPEr!PaGE1uk>>sf`&;FymodgY6ah4sL4qI?mqx2S%70~UzdZhUrL(Z*b0?u;<sD7d
zkrFwwgm;>2DuwJlPFj&%-msZ|l_!UHKYbKic<9=KSdn)Pr@bywR*>&OYZ6^Z^kvbH
z<r(w*4u$-j=6iHMf2a8U;WnBnlCrIYUvmzU69uppC@#t{uW$mjk*x~aZmMk;YDbmj
z9=Z9Qq<^Z*y73%7yLWu!DE1N(B9~$7t$f;`ahAIK<kKP)&?vTHvJ{lVP$cETEpsP(
zz_kr`i$Rg*rLM_kQHQ&%lbQd<`&kcO=ttSblx-626T^PzayPsBeR209a=MYRUnZdl
zvk7x_8BWMaAj^+xZiQ9Sc`cd@feHZ{-QPb%NPmKhYm3xO(V+LU`m;<q(1MM&j2|DH
zzdk%({qw*_98d|MroVi<iVp}_Sr;Zm6@aE+|2BQS`eVClvjgbm_LX8+TgzNQ7Oi6T
z!p$qf-S)L^#4QWTy$|ztXLVMb^tAsc#c6eC5uYg0WdgR+fVdZa=*A<WRv;=h6sxk?
zr+>>NSP8}+$OL_Pm=uHqa}MM7=^}Zr5RVo~0*MYhj9COO9_BqF6KdD-(GR-SL!rd8
zMcGx|%Oh-vQ3$dZ%dLi6+VOP;W!@~}-{~{&O>E22Ty`<xm609yt0-ZjeaO4z`*M;*
z$#%4Si{+FmRB|t^k|gkXm9ksj?|1r@7=JGZI2)$|g_Q&R?SpK~5~>z0^$z7F^_yRQ
zxt^_Pu}j0CG-6HjpF+}SBZr^^USZ?>6#Zg+z4;_335iw2G?U<%iw%d${3O9VR>x~4
zQB_m@>v|YR-YPBeUE6%;?>S;~Bs$k)irPp**w_$<a-#OiLha}`|Nd=g5tIV~9e*5H
zgs*SVxh+Sd@yfK0{(a0);>oon8;(Yar-(+D6O?V`>4R4!*vZ-2Tr1ffM@bHRq#p`u
zcL+^*?xWk6{Kwb(VWCSoQH9ez(idpnpDk;qHdZ1^r0fP^KZJ~vox6>Pcf~uVXKCZ*
zkeAFZTG~6vA(Tj}EW>y?r=smh4u2ODVwUmwi=O@}4}XjaWbo7P>D6zN@+k(%Ca1qo
zf0C3VB@@>6nde1KP*B|bj}G%7scDMYpp5voWiIj*=0><c&AT6B`!$U+-q-Q7VqQ5B
z#h<l#zZ;sD042Gut^3=df~12&9ZqZA-J%=TjhUijI;nN{Yj&*LM<+>U7Jm>r=p3Ph
zkR`u0{gbEfMlz&~J3vX={(h1(ckEC`sb|b4UMOaPCx%N|bIYG{va;sW`g6Bt$DjRu
zpR7^d+xDQ*8}@Zxfv0W6IF{X!PHf<^=$3`O97{f&*223rdlv4;zT_N^n-uB~*EeA8
z?o3{Rfb3-Ksr3!R5TUH)cz=ntUyaaQM!bEHN27cmB)_3$IgyUee3X`CyG?PAY}}NT
zmwfaHuBN=>K2%M8u95@6D{*wgj5-F%SwhK9N*<Cbf!kru<z$PKM@t@WHZnWs6c|&J
zj8k~@vKOLjx#W#hGRk4PD<SAzLp+W@BN?*i1Iry5e`E#RNK5B-dVg3qhElEuxqfpC
zcaG~-BFUrzT&P}cxb33G<SX|G=Q?$>QY0w~oW~2adTUOvpw-(5bf#ACmfELj^=?5R
zCa~o85OBAE{`-YROV`Uq9JuG<z)0~dr}u?$&~Qrz<XgnjQK3a2c^d~t+=a3Uo*xG_
zr&oZ3MxZm{U`uVzs(-rLbqNbYjLd3ZZX#j~NVUQ*n24A-0MzgrCn77sNB3z-{$YOn
zf{942`?)3}DGQRl;W85ul%2?et6KP7BW_u6KM~OnyD`*rh9jbHXvG?iM!d3;q%{<$
z9F4G?Kg0(zt;``T3j9LXuqT|2$U))^9VwGUqze8(ri%|pkbg)@G4t|ZJL-^axcz?&
zM~*n|(=d=|eL#Yv0m+hjbJ$^|V)ix3qC{F@WX0^Z7W2)Y$72YGY0_6~_LMOMNO9#v
zo{H(`v77-FW2i~~JFNa7IOQDFBwxnWrc{4M*Q)Isw$vp7Of#m6<kNWa0J6?pqg$>!
zyes(=1}Q6BgMS}R65QD)3F9e(kwncxKUi1V+{UMYO-1r<7y*WYq;<X@sViL4{I4PT
z?s{t?Vm<WY?AZPgYElwN&qe%#`dGv-Nwx=%1@TL&BN1O|_AJCNnI4JwCCQP9Uy>Y&
z_$A4ah+mQv#P^%N1@T%a^<kTO9ra+Kwfd`qy=kqz4S#ztrCCL?4STuz;kpic-$u?h
z>|KCfupJ9~h(Ca__)OR@sW*&RU|*5E8(T3U7)(vG1$tS>aiiChKre7)fST#&jb9tc
zQQ{ii#-K;dvQN|$8<>~MPy)Rs!MUgWb<7Jn6&Dqy9y1e^5H=3}c?|qG`_OGfy(Zv?
zbT&RAW`7MiDaISaZ2(^iIY8)#><SmlspPLnHh}Lq{raODa4W#b^$}RXA=l~HU1^?_
z_Tadk$&zIFb|K@&1rs|dFHK98JZjgYqO`QggSjg`o%>}$wJfIgt&!c`)D;}66My@n
z<4)K}Mz=RsM|0>8<s09R$5+m$gR)ua$SDkvm49>@&NhVs^h)RP>G=!ehT99LFd+Gk
z^A~77SB(0-=Ru3GbR&1&*1O{x&0;G<jg}|xGso>YY-xz^vv%!-fakv3ZpsRWzfNu%
zgIk%0muDtiz^4_XOZ&8DekPxeHN=N~qdQ(F_M-#k!zHj;F}k!LYvyP2<5<J^pO%<%
zz7s>UW~S`iV;`2!506*>1Lc46dzUfBJrn^nmjM|K6aq9bmmv!QCYNNiJ(hpq`PbJU
zo_Hem{Y}|LN`+)fyGa{lL{FwvnllJb5E7}1j>Z@I&{4|JQE2Fi0{Rd!XwnlM4t#&J
zuz8Tu47e%>xwsm&-#hLY&LbAohxp91p2xyzftGaIiUOQwj)uC^!YS%3#3`MM@zh8^
z3GG88w6dF2%o?0XmCQJucDsLYwLBXYLu<O(L2+JYMESD-uUh3Q`CiY@kmQa2|64P&
zlXa$(31zLAI79kaspRxG(K3$y^l4QuLJ&lS8>r1t^H73yJ^4B4(sQizRQNc<sR780
z$Ys?@AqXg<Cnx{{HX$Zyg|s-ZqBo%;B|!wBU&U`l_#SST!~=mR^0R+UiPnrzcPAl6
zFvLZ^b)iK>m~=o2N_ZF~kQ3ak+=WB}WdJ86c)V1X>Y5;aT=!^!(b$469>$v`-ta!&
zWcsZWPhZBH(!kKXC2z&>fFmAZmA4NW%bBl3^n>F}iDCC*{$=`xG<HqDU%YQKr!3Fx
z>4Fq55ykU}y5^nA=!}2u_ct;NjyWwfHYUuG#qc+{qU_zz-)-GB;T!{pJu?7F8OoA_
zC6P;|t)_47xQGf_e6ppK<oA?QOi7tf@?Z)ANs91xH5rw`pQK<y8N~#5wOY$nyCnZ2
z<NqQFzxXy4t#q);I{6UjJ!`boFO@rlEyF%<Ga@Q)9k)F0w<Ui`ixc2mUSDoHgUAOa
z3QsjsDz(wI@;aigPu+hXU!VSZ8gwEJbLghu{(SNe5ERqKkxOI_aQgfA>GRW%{jNzl
z{dWik0|L0#O0@{kOi}cu5Sb8j<g+TYUSc%%F$Rm8Cq|J^^PhmrX5^;|J#=j)#V+0H
za9fRb1eXjb6zqRitkS0Mrb$RI;Xxu<jXgc2B`|W3-zG5cr^x^z?l4V+Pb6#9%P7&H
za+IXzI~VTddL+9Hvt6sgEddVh$JBQ8Z2YdW-#wUgo~~WG{{H`L_5;Uy*}(g|>n%_)
z=ePI?6u$JXBCIW8*1XCali)^pw3=a;I@Cz+dMAcQE17>$VkL%Ob1~Vq)~nzBlUxkh
z=A|V{l5w%^67k)$Fy`Iow35PsrIp$;DZ7<?`!FM!6zz&XHa#F%PGN}Y<WY1HWBlT}
zM`x{*7Amyo;3)XCMk}*kE8W>`r%v=CxMhhi%S~8IaVxY-zK;^;o5GMB$w=B>7%4XE
zFC8jJfjWN{SKj3ZVn06s#TnVZZX@z-n;KJwY?ka-<pg0(Q<x$Riml}(`~PcwYwrla
zR&q$C825Lpr3~aIKZ0gk4#TRP<ZTx$5FrG8CAV6Pr=BI+A%8wpr$GNrka!!*+U=fi
zQ;6T>oLxtYx@*%UMYF=Jn6yV`7R^E5utp%z5dD9mqo*nnJft$BG;>_<F};^fZNr#C
zGjGV=zV0T;dT`+$j%ULtO6Ye-lWk5aUNn?@UgAiu$H7pK^D}g|XBfS|4(Z8^71M8w
zbt#XW4y`+-WN4mS9ajRGDlLlS>RgJurLx?ukenT)u84zvy*Xw$rSmvez4+8+Q|ROb
zm=k|;SoKoXCoY@kqOY+D=y%vL*eEDzwgZ$x9VA@m!uegnL9Hv9-Pg5|B(LBB1@Z&L
zF}k5lSKbS=lbf+poZC)zJBPCggt_{cL)ujUhlD!My6#jy>bgmGV7KUu7lLEnqTSCt
z=rpCER0^AZM4;HU)T`}6W!a6pkTMn9HS>QZoH`}_FQklKGI7Qc`3t8j`KRbAuh(;0
z5&)k-V87Pi%FH|?1+sCMaN}`K%u4!FNj{#b#v7GKKeV;rxO>e_(4FmUQ7~Upo)J$E
zprNWy0$-yES7|Cz%?N;6XUXXI`WC@j$-bGgj(xpAe;bt6mb4_x*Q%auzTl=lI1Dhr
zEa;wp`#uiveK>_l#!22XV((bjLo*43h&>dpjj<64Fsc&NU?LD7#=f+4$(<b7|8IS}
z-9~^GOuIqo#o3x6-?=srSTQ`wOb);{&S%frDqgD6gstlQDs_c4&&3Qo`~Kdi&2dCy
zudkGZiOLG1n$^6SVxeV5nvT>5MRAgow6*SkHlJ-nAG%d`BJ-Ffv1;)by*%a}bxC9g
z8H;4iQ%qKVy>wj|D3tAJChlK4d$z1w=64Y=+Ayir&;uzJd@fbZ)~t>OC@x%$2DZHB
zJ+{!CDhv;ii%Pz!(#|5MNXdD1H?Z^^Ys-Hdt$&;~lnk$xoVyZ{0MYb2#X=0%jW%<C
zm*rNT_9;o+sm662Sxv<sXHt<<6sCLG4j$!zj|@d+Q+u6b=2a*r=ut<3zoF9FBQ{XV
z3du>r_`oi$ph2@IAz9@T8C~W7s)v>1+<?4Gi4bzTtL~iD!DJOw9cqx~D<e&4^bQSt
zwQ%jCFm-1L;#_PMn@8)A4H{EiR`tezNDsy+Us9~1j6`8Whu#b0<C=X5_WaK=%i|&y
zAE=35{$C23x$;8aF_kYb=%fY)R_z`9c<e0zt{HMOfJDvV6k-jNWvfj0R|zp>x=mg>
zBEHEY52$RR8Kc`~44UoY^L|{=G>104_8Ye51vj+T{fE{(SPE0M{vWpH#ifdW5P6OE
zf@V*ROx`h+(X-F4^3rmIPH!&i=yz7B+$msL1_nB@y;YarH(O|zuhiuSrJ1lzGO;ZG
zdxal`yD9&Nyf8otVU8?q$H>+T1C%xxFazt4dtty=n$QixU1`osh+VvH_Jdy-kZ72(
z<|s`{FmfY<jK1~@10X2&1oKUQ;kn%v)RKW$UQ6ilcMn#&p#O(=4;+1!mzwroA8saB
zit19!Mp?uI$J5>`i+)xqp4zH-YEqY2RCT5Ojwvact{Zd&dtx*fv73>9!M5t1rW+-?
zjq0k`to<Z~OX0Y3xup(uDV*lF8Nz?Qh%ifF1Jbeeqs#2CDODdR0{N3!Rv4hNWTBC4
z8sDtiCtO)c&*mmc_cpxGGv+qKqC_FLc$yoaV}z_vOKl628*W{Jv^0gT_Y{WPyIwc_
zQc3!`ymhFCFjLJXi)aJR4o0(`zbT;j7dx0FB$qM9Jrn{oIG0iJHx!rcy*<N!@a4C+
zpB{J;_W6c(Q&Nx&wOh1NjO59Zdag*EkY-2XtmtoLiHq`5qKI#m2~Rzr=bIJ7`B5{$
zu^@5rBjeokk1LWGP5f3Wa`5PQE<u)Sqzkk*(gwM%FCHW>UI0`_a6DM?brSNm5_H>S
zpd}Lsy`ZQQ2r_*j32ov#@nZvjb5UC&vqmjsd}f?Zvt4*=6wUyxg@rRkA1hHVf6<%u
z)1N=?1FVoKYdTJMSMjdk!E_Ngj})4_9-c<I^j=k{(QoB{-y*;s0^NDk4v@-_0?-Q-
zEG*&-UqNjkGdZr0)SmLQY*t;r4{BpVP3klauYAfz<f3BLP#Xh1OexZT_|9Du=irvN
z;+oU3O4&tOTXZ%r0_^+Bey=%5e88H|3s|&{A^r~nwi7fsIKBWRuMNVP?7@ctiy9eu
z@HYt9meY3-u$@5vQNSLly%~%%kr>rJ6^w_Gcm!SX8s(Ve-y+SooYwH^v{tL6Px267
zafD=5CoGSROUs)mVObD=Pmtk<2$|+{-eizzfxVz#+nX$Tk<U_NIDB!0Dyo<v$ysVw
zuW|B4P|Ufz|KfcmD767k9B)<nDToe}TH!(|o@;ea@ifCFm0%)dbV_kK&hNZm;=3{b
zG(&izbK$A*o+87#5ua9xGbYv)Hd$*^kH>47+r9|xCAlOCc&wg(Vu*$E)}zpCttgcX
z3Hw^#SD`fj=1NpB;&=VC<X)n*1U%ogT^w^7JkXVINy6SP3LUd53iAk0wY-@^2})@$
zQmCgiCv48Cfi#pfsi1@r8-mBb(-319%8f=^Y=>A12}46TlO{UcqYy(jk|RSd&ajcx
zq6nutHju}dO*TM((<|4q`C%*>)oE5hJLI3SWKuItrQNn<J8URT3Taw4e|~y;N6CO6
zAz{Ej?D)ws5>_$|<`fc+oZX!Q!da$_+7WD<i{dm9rQsfk#*LL^2FNCoCJ=hcDJkD7
zB2|(^0_bgm+rG~kClMMpzmh%15o2hSl<lZ7PCz}Z6LwF3xNSCMK#8_x;~fU96Gsjs
zy@=TIg98WERBRb{2J_;V4Vy{7tl7-@<viZSFE<4{@XHSGO}_*~6NpoO37RRr3X!p2
z)@<hevK}?*m)pXP{jz2=o5*rXuld0kjFQq%fl;#hJQyXbZwoih>OB}ULF&ASvgIcW
zzs(9j3CFm9!?pyFjgJ<Rx(2|fnc7bI_8ptTSeKGz*#6{V3pbk}AurH(Kczgoh88LR
zmAjr&zLoyymdm*HC~V5HEXRJM?@Pu)`I)5{t#||@r9jJavgk^&OhLh{A)@-Iw@FYj
zt?ZCQS(bfdj{?YjVMszAZp2s=S~-jyGon62W?w&lC@#@3L8+VSWhShfD=vW?R1$2Y
zu*0e160D$tRG(sF=^=Jp;IM@C&EE+wKHrfqcWe$8NGjo#GcB8gWT&G=njr_tPCEul
zp96prmIHf=jng5=Z3A#%O{6wMsx#I^@^3X)?PO!lGbzTbxg4!s&L`A{$;8q{LE~II
z>HubcihXWXBid1?L>58W0l??%e?qWq|4q!smQgy?qh~Cm6f+k4m2_yDn#;Ut>QR%X
zsbxE;|NN5Sx71Dw=i73rPYGuvqavx1?Wleyi;9-V5oXjR5D!T?3g!Zhs&*|@7-_-H
z?JsCa8Ni%kLn)B(Z@C<2Km>V;i=(zGEUYzu8#$*80OfUeMj7B?+jlAjW@S!r$;vcr
z<QyA8PED!c3>$8#J#llDF{KTSwkgQE!Z=gZ1)13<sSC0llByuvqo@k99g?~r+asw8
zvK^ATAe*c7sXd|Q0KQGBzq_jnh8C3MdEKrm2*`W4>%%vCn-43)k-{ZW*t?A3j8tcT
z%F-ZQ^Hcc3o$4e(+FMT2swAbUjxt{FZ;LePhy%;Jf~6@>AT~JZuiYcHa-NpCT&tb@
z?Gp;$<$;qc^uM?BE}GZ(VZ@GKhes1g8AhY%72>yTr}LXDq8wnf9j9r!$-HH62gYqX
z(8i_8;egjEzbSWDWD|1Gbio^kLm9_^st2ynL|t{T>Mmit3{q?6M;x3q1AUv_GsFJe
zuce-lYK?P^B6_S~=)fQ}C2-qw#>@Crp6qYEmvr)`FbWp%F8yJRSbG$;Z>2G+!ZD+9
z7S)EjPEYavfNqZ5R!FMq*eQ5lj=CskwYRdli!wx?ibX4W0dX9nfDra3d$rDggFg~Y
za$rKOWD|1ge7G2jlcRu|HB>`=-cU(<d@bjPbuers$7WT+FEb^T|BCwNPKP5~TB=Bd
zU1`3u<xCK5B!RRzgoQxuRv&P%0a)FPo5Qs`0p^URkapqJjB2atTl<13+y~fvGrA&?
zmL6doL($&gTVG{do%vIv?u?dyTh};+k1KJ%#+j~f!ZlxrtZxD2TT*dhYxwQnEy#SN
z40k6opUt|g-0NA57D)BHtECK;Q@=c<zjv-&C5<W_hquKNZ{JGT<{CzEs!J?*f}5CF
z_~bRSK;^13)26E%jd-!xd9AK-iO<$>7<G<%bi%OF=k->SrN3*3>V4V}_Li+LcbA4f
zygmF2Em0AmmodgY76CSwAwM}Imn^?Q43}VZ1SprE#yt>!ayT54MPR^A>ePUP1a?}q
zH4H<NFSZ;-C!EfPn_u5=$X(uDq;t>G=_<gIX76UnZ->K~Z)RyvAqC@%$6q!aRFGn_
zsE~^x@hX*ILaI_qhEr8b*@%l(Dkg<g?{KDMsuHxl=Lah42i(5LovNfF8y|fgDPt7j
zHYI+iRCSAgH{?85ks!vVn;{m}WlT;~RF^$;*`+#LG8BEOw=wX?3a3Cf<u;fYD!$H!
zn}g$~n@#WydMPGwMaIsV5II)Sn;;b#J2H4a_drD$rf}1RWjsvbePVBz#t|~a*zgvJ
zrBu8Qw&rRVr)JMpm!J?UNDjfUlqz~MoJn*x49O*bwkKv29gLx3G?B<xQA`ZgP=I10
z*L7r{O%&g5D<-Ds>TJ%$+|~v+OEWbS8S$)^WC}*=+CXF@v5gR*E7fKkpeqdn!5LDh
z7$Ou(dkQ>va_uQ7ngrLXB0hyFtC&pUl6AJgCZ$lJ2k2(06XT9l+8V>dvZpq2Ho18w
z2RJK#H-BIUzoiNX3y@P&fR%GDO$??u3OXCJDZUvBzYC3c@hQBZiV<QP0p=>YOb{iT
zB26KYDQTPyx>>4G;l-tlvjMg9M0GmoIcoE%?c8WrT}vlz?V{?W5tHiVrlSL6p!$f?
zc2zZkVCg*7&BTJc{OGd+q&~JKOnPrhRbTgiug?{{EBg5eSy#drSMRN(xn}z6J3ciG
z`n&C678&D!j;U6OuF0w+p*;BJo8h~cs^Sn>o)7>0*T40_tgdLx!A69J^ABfduO2@9
zoGP5q66RW|n&I2|*?jTh;^<^%G-y3H!%s)c<!o_os9pZ`)AIWl%cJGY`1Zk*`S}um
zlRSZU!Ju4cd6Jvwv<o%6x>dcvR`LZVsuv2PkRLBdFB;4z`tZfqdmcU<ep<{=Ud)y+
zho8QCGW<OIv>c!8=l@*H0PyV$xX+if^W|lYE^UnGv&;F1#mVe)3~Kt>53@I?M~~;9
zUaE><-={*aFv0@&sv0-^kIv8M<5;18L3r@8t%yEyi_DW$CMmlv`7TkHLYKHpu}h-T
ztNu(C!0_V3@lvmzou2<{hR5^8o7tiwcdv%O4Nr$}U;1|Oe0H*Y>6tBS@Nu?aWJFcV
zKpeBP*if&AM+T$67`~tXJRiO@Uj>Skb1Km}EE^({<Uc=rxMtUzU?38!p&&YcOpYJ}
zx3-9E(X(|VH^gxLwAe?+PmfMyBXTBZZ)@a?WD#+2ovn|r9&9|t@TiY|vOJxi1JB=|
zKh>YF-Yu6Gmwy?Cx2MZ@ACB$G{I}u#hu<!~elz<x{BX2<_v6L#Dg12l3-*V7e>NPS
z&5wtV-jcI^xLlkJj~AzJ-p($6haZpM1J~)t*>F65@)6hb#p_=_oYzTDf7b#2^elb-
zjP#iw+|uyl;*%sictCzGwqQq6N*2aH)Z&v(UWo+33B&{nwjgGa_tv>|1%yu?TPg_n
z5#T4VjdXhuA+Ysf4MfToDRess1oA>D5Q3~nR&_b;ZyWd=t>$nfS(_1mGkl5Gi5zG-
z7|ySRL5kIUj!A6NbeM{DIa~n&KBFPdA*)A+2;#l$VL6gmA<e<jA`K#q+T=7Jz%W;1
z5@?(hQ8E!FMp%Qxt>OWqPmxtPNQ8)ZpQ6F#+0n_bQyd7giUT!e=dzZv;Xp{$iLPe$
z4Yiip!7?m?YTZF*|ML8Q{W&%Oc7qBeju%>Zgmjaf&rYu7d?l1O$@yC9ta5(tIluJ0
zU3%Uw(xsJN=LTis23qAzlJ63ADcnB$6JfVIb@_mJhw@1`9*im2ae>R#kc0XLrJFCv
zmk3E($lQLnk6B2wK88lYs1YY<Ict-I!z2KT*z~?*2+mop8Mb17xSROjG9f&`qJZd`
z5Vfk!AS4IPgcAt(HC7{U1+iQF-#z$ZEQ`T<^bJ~>r2Kz~zyCN|oE{yYt-|jwi`kvf
zFONsm`fqZ4UxG!oeo)7sFoXbeYdc^IhQt|knZ01ivDTS=*`MQMJb*~u@!@|KHNjyW
ze*!T%{x%T%cKjlLh%~YLJ)*Lf+4sWl&rUC1U!J}_ug=f#e#x7pLtB0mn@IjTzOk+`
z@eKjbEhv?DQAJrQSd1N1L&B&;yQqR!i@yuHv%yw2c+3Xf*&srjq-LE6fx!`Ze|x$U
zuC_R|L3ks&6V_TB+6o;>i4R@kF2ycMmr|E<mpbvGPJF0;6CdhxP@jW%Zw(Lz^T}~V
zJ~E>SIBX0aAq3FytnKM75J;QrAZu7owKvKg+5-mz;qx$dD2RI6su#A}eFF1D^>d&Q
zOo=V(j|oFSDD-gtZ5ZyP^z4Q}y{n!(83L5v#%r2UcVdXqCxbX!^reOeE7zukvoV)z
za<bEAbru+ZF6gfk_mPva5tI-BO3^nj;D0VrZ4k;eML1`8q8E+=EefubJv1W%axuJp
z1p>7=eFj6~sDTHuD<*gj>brdyLK%^gl44y{hLmk-f6_IbAxb4D!oPs6$vOG7kA4V)
zjIUup?bj-VEg*I?#21iqCl=4NBIO>0E#C`LzPOx!pS)hqUte6lnYcN-J#usHK}%V2
zb6CXZ19S5c!;~1dVfb8Ps1%5HY7K3wjik!4&4up#rO#(tLo43%M|xDhk!!dTA!Z^X
zUBj<72$}?Z<ED9v)<8FUg@iqKm0v}7Gkz6ijbBk`P}CU|bp{nW(+XWLbiJ_B>r5+j
zrp29qX>n&-+?f`4rp5bXT7n=B)<5ki$=*6X?2Z~C1Wq-_ZG|9+;vRfcPYO^86mR#U
z2|3FO^4Anb1!%DL_a=<u=PvpiAn{hx8n;FkOIZ%ftw#*gl5fSZ(~{qi&ei2!*#S>#
zL0}JRhY7=!OSWOyt;qNS1`%BCs&obsk_EeeyAQ6fp1)Xg{owaVA5gkPhhV3#_E-;q
zIJg}jhCO3lEk^Ai0)a>9!J;i7upcs^gkTVZ2rPSfpEw#KM~nT`?NC4`I2_gtBLs|!
zqRMZB;2nXp!|h|r&SMu+3VXAH6sw1gCM%?@<=tVeFhaodSv=iV2%jk+Do2d|e>MDn
z)--w-Pmd0Px9p&*l@8NAC7kPgaf5JfdrSGnOj62KCMoG=Ofu0LljP12xiduW43RrS
z<jxRd|2Fn-WB*n<f0WK2rSnJW{84xxFDspu02&3~@LuDYM^O^>Fr>|g$)c=vP?ipX
zGt!$6TOiy?gG$?i8A~=nKuoiG^7ndw2zT<nW{PGW+gV@R<YEyufhbu-t*aPxVr-HW
z0%{dcDc_DuM1*cFi8_sT4u9{(tCDs?W9^vM`tgig^T(2{msRD&+Lbw2Z+7I5BZmHh
zd<%x%*8Ge6!yu6$5m^72DxJyR*WLuO)M2`TA!n;++eZwDde-ZCu`L)p7a^B_bPpCT
zr02!=(H0HZ`dUYbO|`zl8tCagv8em6U?|{m>IVT)BO(pD5cUy;5PQXU3_{V0q^%$V
z@@mrD2qMZMO3Y;+i=jmY3=tT0burOt&ssMK_RDomI-tW^fY@aeHzn%-skl`nB6vUN
zRY8wvXiL5q1^w@{#r(2&MClKI?`tLR)Ct^B8&8UtU|rZnldu{tr+f2$_2y^8xgE{k
zq+{&}_m|VLyx!Q8PRFFvvDDjnsehYR-*!6IPLuDaB{y=3-jh2VqnbK$6QM$g7utqE
zaz=etiUdg_SX$a&g{nLXRd2UJ)OSq>K7})3VA7XuWh;hlZ~XoXZ^OF*ESE9HJrn^o
zm+<8c6qnkpEGU1~TFb86HW1z4uh0*KA&0{kaADxui3@bs;O>g9iYBWpn*U#PNDe8H
zk}UZFH(AKCDT?ztGn^S(4&Mej{PuEufBAUUI5^>raxlDq9TYu~^r<8bS~EDMgw<+z
z|33WrDkWIDj&z}vTGQp|yQjN3e9+3%3`H{(jVg8c;81_jh^7%ubHLDD(gpP0@z|N)
zxAeZZUd9zpeA&Za@4vrDW>$(|<JWlWe|>&``S+!=4q6x+25X#9QVoB9e~AxJ8DU)*
z07OtY{P<`1`tnD=YN6HNg`_2fi68Xs$G4#y`NGN!l9VZ}HKtJ}SYeDGtkpt0SComQ
zL?opcQjmWkg&4zQ>31jmHUpKKMd8u))Di_!t05bpZ&v?!_G*!DZ-s(>A-KqPwsRUI
zVo1O7Z6vruj5Q?!1`)KKDUr})RHSqXEmhDJv=_)g5E}GUQbRZfrMAsQvtX{~GHMQ>
z3b#7{hVt8vQpC5T-jYObBjvIVF6uT(UuHyZfLDL(IXx9Bl`lxVlwTpws}xvbJWplG
ztJGh}mIdQIkiwBJuZ_;uosE8}aGO~{J1+oxBh-))LxNg^R~9xsUp3P~y<S3SCX=jF
ztRVURiAk8wZI;nwPz&p4XrhQ1-=1NPB5bg>@vf-vbBw6&g3XOBF*ovQd=a^f76}p(
z&-^rwY_X<3H4;_cv=QFnV0;kZJ@a;yrt`7;ZLZJ=^sb|n4d(dre20DLp3XPnP+WHz
z)I6=1AJjcCf3NKwEURI`=_I+OmGNm=tS71!JELTr`Mgq`gA`zOxQxfhlsw(0cqCMy
zeum;3N<W9<V=kFM&5P|-IFz`-Q6KI7hKP~hs%IFYq4aBb0PKuP_ZUDCJ}KFMf&m&z
zKes3BRs0nmv6WF{P{PR(pEg4yi9X+SIpZ;j-$Z9sfBg<<UEQyLM{Tku37|>_?kFX2
zW7;xHo_6Gskh5jZ!lqmEX_8yA9n$l2&E%$a0<x}7UA@rda5?ckd_d5z3pVyS(v5jx
z;^4@MLSDP@>R9)-(_pPgrdel1RH+}v_Y2qyrV?$tn*Cx-WDu+UxVF$f?9g5G1eES|
z)MC=zf4RJGw$K)2Z^`^#Uk5a{Vwq0m{p~!vu3I%<-(?-vRU@UhwgzNf?5yi}UE~=P
z3nncT>$Dnpo9Op=f!k&8DI8|oEn3-vph?wtil(r{md^8MNUaFltA()wDsjvoGKJ%2
z)q^?+r=GEUx0L2_d~dvVjPvXTw``aQC>%Uvf5U7k-7K@#lh~9z_`uHJGEREYp4ijI
zdP{d3BxL3npG^B`1-udju~flZ804&m0lWw(ZlY=rq~u2A;-C@U)3yUltFl5NOkHKE
z3z9|e8X@=9(0R#PM&DdRuK|y?RN<xXbn|wuNdSa{i5r?J4L&;$$h@gs-5REvQGm96
ze`}bE8n5BH)^KY3uC!0N?Msd!##-rrZ*9axw2jh8T5$wZ1c7eMrDDp?T~&dxL0dkO
z`!pj{w=v`{9{bgDuG%>#nSvX@RK%xMn~@9MU}edHgYK#58=BnC2}?1!TgiOBUUILp
zyLXbG*LR#i;m6H%bz2rIWi+ikcRMg+e^4u7R1IP!i@!!XO);KU=vI2PU3Ki}sGdry
z;r*xq&O3M6mybib6{~&k!C(+d&NQNVHy-oBeS8)yJf6>h!h&Dq^EzLFcp8YdUn%0w
z%B4`evr2^E1e)6Y)#iZsI@S_-t3BkLP}O=1Yo6ECyES{fkW0GPwfb}Z=KjE3f8XD1
z@y)oxCU4#D_^)!9zE7)2w=VO=3LCW(%BbEplU-rWs)!*SymoI+H$L9DKc4CFC0ux!
zR+?6a8z&kyv8rhT$&ML0JmMC++glkj7?K9onyY=5+s;nDmh61$&uyxWj%-yotETnz
zl(R=c8ZGtcWOTiT>Q5-O3VSLqf3x0=;&-ZkFVHq)Lf4$K`QCSuOui(%vAbay0oQZc
z=)3l8+783mujz6lwr9NC5zAM_1h?aCHTtVOPyKYuY8OuaOrw+27#XToT_3Z`sqSjD
z>|#oZ$*zgXW_h0uYCQj3tGa>m3-orce>Jm5T=*IHxi_)ex#)Xz$-B7we^j4`R{rtL
zEvs@TC6uQRJ1d7^grQoc)5EO&JS{dh3%OnKka83OYG#3HM)rvR0(c|zW2M_2@zGK_
zXlJ_3-9|gZm*3+ZBU+ZT*crZ+C;K+t=GM>MMbEz6I}cvaNJk=`PRven`1Ga*<e2H5
zYx%I+GB+dynx4YWO<xhTf4v!7!ag_cl40dMt-iMQSAbNXn!fg@deXVA+COL8<sVDq
z-<QhBzQ2Jf@*uK7-Ag`=UHXDrQ2S~IV};hCUvV_!^;I6%G~J48vD@j`KI><%|3&9&
zen4uWwA;xKNGEidasQZ>OFobbk7>CS#aS&thesGL|5vYDaLsO9e`|KLA89wB%$>lx
zj~^hPL>Xr4*s)t(!;|qTEbZjU0#(>^p40?0n!j7FrlEUY@bk8M@~9qD;_KLS!<qRZ
zjK@l)Fe2Ma<#vS^9UEFy#B{4JwV6sFB0j*+x=gg!j~}`DJ{@VA?i?fIsVj_I2Qja>
zT3S0}%V*-toua7ZU8T;<WPIYVxu3Dn*SbOu?v};7w$N{~_qP8t$})O7pfv?>m_(rS
z{NEHSyWI6>Cab9iZ@PitA^jakD9+I~V1A(!6vIp2a5AESz+_vIZ#)?34l5`j`VTp=
ztxT6O#yu1PHJ4%NJQJ6?W<e=`?O98Y+cprs`&Z}>EW^hPA3$Ir)~*Be)?jZ%Peqec
z4$c2RbVv><iIQw-tvrDCK+C4d;c%YcjAZwvlikPD<@NOLYM^t%qjI`?{oE<~L(->`
zG-<JON(pOJ_xiQ_<GqyH(zwt7DUG4w=)0#Uo}Y~Jv_jDeMYBqspByNEnlUtEXpKgC
zN*bEJJD$7peoOnk^)gcEdAReRufLrnGb=@~@oRjvzg}KX|C}o8+6c6vvj&Coru*yb
zDgFTpg>|6=5Z>tS+uz;i)9)?Sa#r(9g7S(4C9OrA2ufRl=sRnTFwPZ00SO2k56@4!
zIFK-qo~UTV?=+9d>wBJm=e6m-{v2h9CX<O<<R!_aG5~a4hn+<&K>4bvC~cOY09sl<
zzJ2MMnNJ@FYD7hPXsM~^I2wJBjHQr~hQ>^FGDB3~U4vW{nw0{BX-_ff!sWJ;(^kdr
zoUlr@l<;N$&Amz%-`@N>yLsIaB&mERtLPhm3+m<HQ8Q_-XS4W!kaZhXE3Z@C7UKdX
zz?K7KzyESZ3a@lS=xD8)AdFVcAttZ~6mdk$#%;>A<g4Hsf>pPV*k2Gzw&F<U=h^(O
z+fA^T!Qos65LWvoF84W!jzOF#G-h;tm_to5H5|a7(ITk!qR}6$WFD8I^lI)Z34n5#
zp~zSm!!y8pj-y3?)p$->3suZtE(PwA$&v~5#gEHuG+zkySM=XYx+w3*$@Syz<(*p-
z!DxRu3&#AwN;KJWZu*XAtwvH=yGhamD{v-p6Rgkxt;9z;XiZljy&yYpcaGC1A&E?Z
z0^tp9Be|g*fJ$Usj(s`*PS28KfuDU530)nsIq;0D!kpTF@rvOQ;_ws+(9&gONeCB-
z%jpA9fOpO6<UdKx{X~g!q@oKebufa$rx|Vtt~G)*drg>PGP`Wder^IZ8}7KkEt8~x
zQ}$H(-Z&fCMb$`4a<U?@p<MIIct(Vspph+LKbI4ELy=XOm}I2@4G?2}A_{}1n({_R
z&(d)l^-y?!MBfIec(2hd`@z+#oGK=ruDYd*v(^it(@jDrE@%1>!^>7Drc6L^s(Nue
zr-VE?&&+#E0)Rk?Mztk@Q^HY=bGszS`7nJTqpiuzCoU6KaycV8(uD}VUJ#yHC7f&G
zMYY-nJJ1SaLQAn$SZ@G{jEWkHN-c*kM+FbscC}uA&qsUt(b5QI+(v$M%phUL{V|1D
zF2efTg;)`YB%{#g2!(hn2(&i@*Yv?Kv{`+mKC~2yjBp!2S_AD5Gk!M7eLrv3ia<ek
zBX8vKno@47s%MrBW*$vRn*N3yAIJHIiZkbCf(jh5^?+1J$8B3=Ej*Z5VxwxQE)|&^
zNWcewA_KE3lcYL)q>D{v#`elV4jh3xS;3ql9u(u3Wh{1nvQn+qic!c*)!Ar%<e4##
zGPUNPTu>~OygP2lbu<uWQ&77-gz=J}S44SX<1t+Fj{^!D7=)oMkdaGnm4QvcrU08n
z61l4oc=RK?3<FFmVjBQ$7Lg8_<f1iIn<!F$JK!Em>n3RJ<)(eBVyOM&iX!7y-XE^0
zG?%ji4yij~FSgA>s$%oY_m*-ENw^@oWlTesV?0e$-tsnuEXw~`uuK1Z`nXyGR*0oF
z`F@QzSi*g|hn;l=+LklO*bRVj{(MwYRWi4WQ7u6P2@MAyuC^dtvsi@<X=X1TH8G5T
zSli-JosU1dMQsS602~5LxYO5ATtg<@p>9?WO38i+_y)w98-y(%&+9!~25HTXxAtsn
z4{ZlhiA^rOhi4tD7Gn?8pSDzHg|Jb71zWY}D|TyHiq0$}qq7s5Fl4W#0GS~4e%?*A
zCSx|p?;%VU--Cdj*~uv2EP%1M^(jq%^d|KC`jiVr`l3Eo`j&TfGwBzOdW1C;*9L+u
z&6Iw3v*l(_(NNT)*w{#iqr{p_qjfkxlj+dlSdpC^f_Z_d><u$*X(d7!ynxfp&^N6h
zC#62NZ;hU6QnF<l-W2pQo4u7_k>EL<+n!&e03(w-UU|R%%BaR|?s{rUVqUp_ma&Z;
zA0y*S4);s+ZMCNxXuGj*J6%2WZT;9`s!`N#Zu63%7+i5kHoj3mZ-krE7^Nn7nRHdN
z!IfV{I+FkazU}?ZJHwgZ^OJgiPZu+Xn)IyUp$_c0aWm>;^BwTR_E0v$XUF4RO>U_*
zK7e-#dk6|6KN0UvIrxrvHy)^esT%t+cvmY@tAzvcZbR`7-fdPNh<95)?&IBN$4|h!
z+j%Z{_c;9(yt_?*1@E>po4gO+j&moCfPQdP%#EH3f!_ihm9vzK9z~9u-gF;`7~KRh
zy2B8oasm(<q1*$Ev0vd;vuOqUMGU4!t%)Cv{o3;z?Dx2Z=04%nNH<|WK<9*}sfNLg
z7*J7&a1ee845$=|8n(W>pRWEN1O7KLV6egrF9Y5H0m*w5x}S&uCk*TWfDI@k(DyHn
z@bcZ4F~&U<0X3IV2Nx6qF*lci5k4uG_u)O40q(bt;yro+f6gYDX00Qpc%LM}I16Nx
z8szr6;5lcR1?FaZUF3D$Vy7aC`hb{sE1(xq3BPT@DTcLUkI@`8S`BTRS`d&h&MoB8
zn_3X-&KYijUJ5!3s|5CJ!iG@R%T>LCt(>y`V77nM&S8iNd^p81D|+?f>X*-Fbp!2%
z^lGtw4Yonze|rSK80@l<&x6D|d}#P0*io<y--i<?_zfr59);C@7N}%)l~Wd9=5uzw
zYEH+Z3PVrT)v+1dP7Oe5;Vexoj{sc^;6>^)Rf87f>@Yf(b1f(yFa`xS3Q{p>9aGRC
zrK5?2U*;(rwo6+R^>W#GY#r;HNyVe90O(W9^p9l1f5>cbd7dulxOxSOsU|l}5)@;I
zs~+pB7x$sgFM7{4^!drUuHHF#)_qs7?dP{3+OO5no+k#|>ul>zq_A^0w=CAc6*J0O
z>_;7Jx4+?x`x6Xd(KxxIy2_wKVMdNdp-LGg7$N0YIW9PziPeR~r26WeE%l2VD?WQy
zD}<tEf2j~mjok}(%4|u464uV*T6Of76u$xrvFPoADl7QTEnJF)t&rc0|0qd#NrME>
zr<&uHY{nI2i8X`m@Vt^;k!5ye%{#Fxa!QcKJF=_ci5J+_F#3+{YLw$LyBcQv0Cu%s
z=EAP-w||9Q?c2Y?u13sGCDtd5u4Prlc`0Xbe=>&}yv-Veh-|Cyi&f@<_hueMu8*6S
zm<MW1Nqws}1?{-X#Kp6HP7nn$OqxY~-U6vh3FBoQN+Y*JVW`H;C}TwZSmIyMS3NZe
z+D5j-oJJIGmB-g2CS0K_+1A<<ui#Z&aR}wqsM!TxMeMWy<D@HS6=}sjtQVWUO{*9e
ze_<ArF0rbNa%NQxp;B~p_xq5xja&5y#t&&PqXK49$R0P$I>V?49G`z2qZ-#A7}f0o
zFL~s;tL!T0oDr6Z=^*G%hQ;DAyKz4S!%|%Npxw)`=A-{-SpRK?rHtn4^XDR1bD6#)
zzfzouEZoAcxNsNF-<4m@%i~nWJ-1RWe+-#>M7mBnfYt({_#Sp8kZYCU%Z1*jWA+Av
zejR(`9G`MwcC1oqwcz%4CWlVPME5>S4m~iVs#|y)wrSfl1B))O-fO*zU~t|yhDJfR
zNRA3rSdBB@Ow8W!UWU`7cYcS=H@%mU95DQ13!@&G>)O-SJKLSi1+|G3{knH1e?6<3
zC)v9NaTmy>8|B+?syo9nM5;Qja_ryOq{N%v+H4j<Yi}#I-=XlvgI%n?^&aC)OyiWN
z7GQnH^n!{VSJ@lV3vzAh0Ny#hTxmb&3v7li06tUVopJ~`6bSmAatJubfe5d6z>+KO
zlPIJT!1gvsvy~{q$23eOhZ#3?fAC66Nw$V=botYpQ^<a8N;(*3+|b)oY%geFA;Wra
zPX%jWn_F3bjzy!tfeF1e#Tq8s`YL>w`Edwa$k(&am(FsZZ)o;`RsyLR)Zr1)6oPfu
z-@~Zc+kQ$^D$mUsCyj{?)Dz6ou|eAtLD?~x<Ivy9)G<9h22$NgqkC!Yf7l`6(pG>}
z14LtkM0c^hC=<^gv&GP^<gG&2i}bKq)~Mjxp%|Bat|*>y^t20R_9Q#H{V{eGQ!RAt
z$CpYds?$*J=x|5FQ&lBfi(%PDsRVOjgj~5WnK;*R&2?H&!&NT$Ywx)K{v_zxzHb?7
zx^ky9M0B&a*!nm%<nosoe^lC6Z~IQh)o$^{Z&0*cVzC~e&(}_zT7ntPpK_-67|upw
z*AGq&kL*=!lc~v+kS%^bv~5B%Q5K%6CFOauZ(0oWxRvHf-_I>LLiaYr)Mus6_ZEK0
z4kTIQL8mv=pMj;ITRg}%(<_BC*-^+yD-5iwue-W(b#?*|atALze`nKBXKU+?wS8Y1
zEVhWsRoLemtHzXTl&(e9HaIwfc^aBft_xH;8`AI*Ss`j>@sX4{zqy69nBu&%)wjEj
z_e>sKg)UIJFXdAj*xlW@N(+Le4_ZTc?CRD#R2;2+HZ{v-Eo@f=Gj-gVvfDll>7w?p
z6_k$keAn{JvNcQdZ=R_g49eb(t1Bfu*<q3Il=Z=D7HYoKnOMy{Ny^!KjP9e$t{2}W
zoD2Q+;nTxopRn|qk3N!*y45vH^4F$+{_y&0zkK)FP(@r#0}cS*-ag#5!GwI#s$uc(
z*vl{^R98P<Ze@&TX6gR|6{i^8modgY6ah7tL4qI?12Ql&li{c)f4y3{jvF};&F3rn
z2TZeyEFQo>fJe3l*ldl-78@<rjvTE2KiJ~gP4=bkksb6%!zSxu@zp`e;rk$m*Qb}a
zr;j(K2P+(b#o_JS0Q^GoUywX#Xs`gnC>Y*;41YaKiH4sSezcTI@niYB<4?MLP~dn5
z@C@KlfXfG~fkzaNe<+@V#GjHM<nK0*-ST?F>z#2jP`KvDU;cjk<4I;=Ki(W{NouSX
zEKiX%aMCO#F(>_!xROf$BU2~H@0dHWWZI|s+N3dnByoJl^R>MMW%Q@K#H1;@nUddU
z^6@(+l7wju*{IZ-+0l97J$A}`x&`xq4IAJur1+Cu=3%lLf7i?dqUodIgg3!<;p4UY
z@5|fMe^1qds00~57-fXlcKGMVQ+T0)gt2}=6dv^O^WWjy)1N)nygB`6^sKmF(ioy<
zzt9LGHyEu1qAvS&+>pG%FJWC)Rx|nf=jZpK8~OS=>g%OYR`(LX=kPQ>H?DBPIn}QS
zry$Cm<%#m^e{oZANhjC(V$@M4-wIzpk1i>1mk&m!BXn3vCB4wz-Jzruy!TSbKm#dJ
zZAQL^!{_DFX7ZWmX{P5AV1celN$r-7@tl$(!3OOq(U;?zD=(1!w%7c&y1hzUVV#|6
zQNZeYTm~cA$zdh~Eb}*7I7vs{OJ%4E-}!5N<aO!ne^=YDfPDb~hEnvZ>Q5J~pGO`M
zbLbLbo*3ShkKZ!D@;>L04)4mmaDaXI&+i`hx+a5iotZ77Zv}-bJu~9p%wC1Ai!ffK
zjH4NimqUwEN=QGqi`m3*rP=JH`&CK0#zu5O$o8$FYJ%f@{x_xbAZ*Vit~VJjSKvoi
z=xi~je|pcFX1q{n1Is+CW;*YH$sQwn%g4RJsV8KNSPk*nKgp9&%2kWzJq}zdg%uCi
zD73Xbg|m^8I3`+{MM^&OmicZ*$~z(>SriWa`yyjhY*e4+wlt6d;p~)fiU*GhWigD!
z4WEayMR|y=17mjSM}gW95&}Vi@F_6oJZ)9@e<YAiq_=XWh@C4!ay;>b<LUopLiT{(
z3sn3A9VdzP_`M7AwP%PkcsJz7B(#BiScH6NAV1C7hWsr6Km|fkFD&})T$<Ba$!ES>
z1#)Bya+|GYLra($X&e8Mi~x?}-Ku*cR2(xEm-g^bTN3b_upcECf_{Z(ag0suhX(tr
ze@pCOKUxsd>7MTa>@PJ{N1<Xrli-NJB>@>{V(u$3Wp|5jzP|473t6EiIc(YaQfiuo
zteDUBMsch80BAF<!uMT?jjUJCohxrkC6Cu_mn(t*tLWi3SJ`;y*;&TfL~=Kn?Id-E
zBfjhOpA_bbC8q$L(|B+;zfhKuo`L59f0cMH<4C50F8Xyd@-^oq%LQJEIm&UyW5?8C
zGhOJ)%Y6YVw@?Ua+jFDi<s`TO!abpLS<d0=4>co>GClW=3KR`SbnRBWqXI=nJ8fnb
zSgAm{UXBTEZ?BaoLg)Jt_sOI9m{~f!e@lIbQ)y8+?rsEJmE&3D-SQ|YQA%|Ze@7(e
zFj#;er${Bs%tp5&iE+Y!n#s}Ul&kx5e`37h@;k;tuZ3hh>Vym!AHkYxT5xZ?TF(<*
zlNimmD`dnIH%X=a2ZW0JHb2)s^$)Xfl9-JDl?*qJOp0~V*&VKF9G}NDC-c0f<R>~w
zX;$Q9sK!LSCX2SC#;H+|twB>hf7muL9^rOMSz*-7o{w|8;&aLU&bgfqy){j}kRh8z
zrJl#^bADP=u9~)Ja$?uaLBLs;b~E=nA4PU?2RN2ttr?;VxoKHtcO5#vzomz+^;q5E
zl56d6MZ#Il2Ofp<C9i#v)CKwn;GX8F<!r}k=76=nQ2D`J4cJKqv<Nz5e=ayl8$&^K
zPf5SL@rmS4pCp&_x!!tg|KL!E;}=SxQdDU=7}wTC_QG#*G#gU7-{EIdMvco<+;w9G
zvsqZ<6l}5~&M9v|Aky#9zTvc~+MOZ1mGpRJ-Qyv0K$6}bsHFRX6~V+F&I1e97JV)y
znA`_IizZudU9hRxP^}Wvf2vbXJoF_8HG97B3_j>B%^KTW{L;g3fJ2QY-6M(}P+lm6
zp5jVzQ5w75?DIioDWWX4k|DI9&`rxLWshr17+Yc;wFh7ALP)c`>bab_+64w#1JF}C
zakUE$;-T&}zir&yr@Q{%@+Yr5Hm(Um?*fePn$w50(+mNsC-mrTe<dRnr}!P0Q`@Qn
ze)p8lY-@3lPV<dzP3@gtiFY7K{)I;`!b-bDDgnyQ_WUs^+M;)n5(wQ5#2M#m==>_*
zYK47=Yc+)4<5{iR`;OIW@BqJR1=w+`X7bUWfvnsbhAVI+nB5v4iL#tp2II7p9_>UW
zmbY9YB4f^QZXLHXe=R<cCh|FS^U2lT<}C?%q0#oV7i3?R&V}?HdJ6TPG$Te#N2mRx
z^0S%LjQmu(SJO^b>Bza|8{G9)@8WIlLQj6ZtE6KRzhD(*cW+%-nttVardIdq{9SXn
z;W2J5?e=A}1}8XM?pGbSYz|j^sOEmTec7y)i*5eiU(Q@Mf3I9Y=JVp1(fBI}JlWt(
zmyU4;PAV;PSLd%($Q*UY>(#zM?u*=Jatub%@tb>VS}ddL(I`@?Q9o__qY+1&j$74v
z<;cx%5|)qIxXMnTAUeJqw=wi~@|tSh4c;`SBID0cn;CxWB&J*T#$Eb)kITvb0U_R8
zSqf!tWOH<Km#NV*3j;ASHJ1Sy4HcKr@;$==m$#JjJ){T(oxNymmodgY6aqCjml1po
z69F@q5&t_Uf9+e%jw2}&zSmRu1)6~X0+33U%C7QAbDL4!R(o3Q%xMlg?|;|;0x}_g
zop$y1NZXRjHYS50e!mE4dHUldPd~q0K3{&iX>)SITjk*N`TI%H50d^WNrPxlPAP#}
zoj(6O{o_?iKzd#1Wu(-aUXDI{dbj6qT6vnGXojLuf2GdfoKZBQX++Z;u=FnJ1@zhR
z*o~h@dfua#k;0W<m-9bAe|eG2tP}y`$N1KN`}lnM&r4+;w6GXXXq`|}o&NRbOZ>qo
zE6{}#fC$Q*{`&Xn_m^LLs%5SIT}UD!Z2Uu?{`%w8jr_*K43dN?5v^^62`H@fCqyl@
zb48eje}rgAD3%0dNg&qpSo+>=ew%^18XDn^>sL!FkXSA0!1#vxPdBe-`SvIj^n_rQ
zZTdaXJL%M_Va&vZA&*5HvT%;35x?3H`N?D=$!4l$@-${5nKz#Mk(ZITY+fkyeuRnQ
zD1VZanvs!|v~j}uiIfykRPT*+>Jv!fw(4$if6VTI`Q6s)l1&*fLgR#gQv~nUoTuu2
z!TE~^ffJzgzMaMZ;HGxQVIhq|{f2BJnqR>BcXh_wAhl{habIs`hFP)=tTWOxV~Xa&
z^P`<z#vG?#Dz|3UgzGHi=S<;k!hvNvZY+m(tz`4f9OLpz11(p41u_K+u2H)aeaRre
ze^+(I#joY@v<>K7PK~#A{#IO4-j+o(R(yIMSzJklXN7T&%=Aq)ifw_qqA+ZX#8t|m
zUMEyr1rNMN;c!r!7`HYQj2(8uQmPYCZ{yb@E+#(jIYufWT|-;x3+vL{vTj*><mR4|
zAUUPB=#kzrNXSuO_Ltv&Uffi%02`E`f5=v2e%TSn*JQDNHoBFBZyoIbOra1Hc2YG9
zHzx1ou`#k;M<!HO2t!PWV`>t`Q{W$5g{SNieV(gv91fyb&&4#B^2M0z^r+<b_rm$A
zR<#SlOW(=n?N)tL)2^&}9S6RVn(xFdR%_u<KLfWUOtO&M8IxE;nPyP;M7Bo-e_M5u
z;`#@}KX-SY2JiyWbbPL)Y0w1wobDbtmZc-IpnsRVF$OKh6?eWLbVQ*m+i&K58FOfx
zEUwd1LDAR4u&3id&982OCc(_=DLuI%tVJB3ME8-$<aN&Wcft-UweS?7%X;=~Ofe=`
zg4sF8e6<?ul~r-Dr+CyD=B+&=f7yMJ1P~~aK+jN3T_)oI@Ve9oH%2aogA2xA%?AGA
z86`zu3&o_i_zZK~q~eg%JU_dUu=A{C95<wD-is}Km=0E~?PgWEHlowuC(0e-%#s$Y
zM&ay43k#^qcq@XbC|J7zM<!W2^*a6rBVfY6jg#9LJ~3|2&m=7uw0o(ee=Vqy0<8BC
z-I`urTb<o~GP;5lfaK;7C;O^FWiD&ar0g5-;yqjDN|xW5ez)R-c2x2Zy6PB@XcbcD
zU=HH57+L(uovu|)Yc7;j*BEN0wCO@w?dvWyYidr5;43Iie1!(>x~90n?Sg8Inf=HS
zm&%a?jtIsIMKHDFeupDMe{9j{$&<C98j4Ck!iM+kbF~uCUIaanx}D^=b#O=_-8$P-
zQ6)1G##+gB#-flBnX$~`-bx~nj3sX)_rDWMYD+P}PI_QL3p8N^OXgNY=|>FBi88XU
z=Vb`Znu;Bw*Vs}DUGqLc?Q2Gnq;%9)LvWnJtsi<3C{=+fw2W5lf7oy9J*61i5A>42
z1TrM-SX;xWWHT80y*vQxz*%7sdJg7c9Z)&rl<Ouh$NT$1CY8S?_<h`HrBgEQxY=Ke
z-^Cdj5d_@H88`eo&i=RkzRsVP+I`!;J8@K{gavq}r*;(A1hZd-+?4K!gPiYb1S1e8
zn|s95q!T(!)?ae9e|UUiaKi2Ld<~2e2HakLC{-hAt>@=!ybZh;0JxX80K!aF^>rPI
z`?3S?o{}oN5!k%@W@OESV`D;3L?0NH!q8|_A)cY_@->4ff|@8awk>k&qQ#knQ78nS
zXdQwF`U&n(ys;&?pkdmVAP_%U-Vf?3;ii5+=oQ>G+Q%2@f7n~|b7JWm9Z#?Ws?{ih
z+kUkN4Was4OSSZ`<TVvajWKYT);<Uv+B>y%LsjJPCwHaCBQq&PJCeGwa}{OutK*H{
zp_;rs+UPB=pSw19V8dnou#|6O)LHhexd%RPOWRaYc8UFfR?vxfick=9*JsE)D;+<<
z>+VyKGJ7MFe?7uC*{1`y&UfWV)m8#?HdDs&FVEA@t~1v40+1MKqQJE;{D$MSde&lQ
z4$zG6&Im_Z*lM_HN##g;%zMiM18+B5jaUvbFT-I|KI&?I?KhTt8@al_mX+GZAQp2R
zpZ=M-W~42hnXule@C4qh>o)v6h_UZ9iBcv)&qS7`e{EW&SWg@BA!1^1bP&!nOe3q#
zb!!uN(8i?MY{v)ghBt?k`F^%%MIIv1*$tFGdtytRuXgW|QW~Y@ZiO#Fgd1IZN62)-
z$k78nVpMoAlLJ0lIoR*glmU{g#5+p8yY3$`uILt-*yheFNGcsmx5Vy2?n{w9_IlF&
zp?s=$f8=P=DFL#V=33i_8!yw%94$t3$<#5hR9;a!CHtO{<i=s%q1#nt>RNY;Efwip
zl>lGZds<b}5!|Rsy4laH(J99(vg>{P02q2cO!)bt8a)<}1i*IkR!qKA)sIr7mlkeu
z?R+(Z>mc><z@c>1VbE^r6p+)KA1+aq+np7~e_?Z&f#X|`^S9&{q0f*`7X&dJgV?#~
zH{4G%%{RWrmBAld+Z&(wy&ubU$_7Dsyw7ze2*NbqCD*Cog$DJ+zGg@9@nrOPOM%-R
z9cPcW6u42*Q7tWnhmI?9BJ3zH1s$(NaB3p==eLbCfB;lG`6{=0=Ir75Ixd%So~vUW
zfAJOQB^{@>(eZ|miuD%4{~JO-R6|HPB)d0%yM~ZLo8r;myCL)y*M1d@?=l~_^{b>O
z6zG27Tfb?{@hizblVq%7VAJ##`!QDKG}p*rbh{VA+3MHz|I~7{7eII#>DTXb*Vj49
zqq_O;M~y`HiZkbi;rHQMY{j3M#2osRHTPeeEkLB@f9BfkLl}w+p>`2{|NojQaHS>G
z)F#&wt?G@n;z+(mtNuF59vdJN`!AS*K^~Ve#yu1RG&nbxAwM}Vm%(H=D1Tb*Z=1If
z|DL~sz`k@J%!j-?-jPF5pf5{Tz(s?k8@2^T;W%pI&XEFHDZ2gj`yC(Yu^l;xVkvc6
zr1nWZlE-`f9mz~$8dAvkZ6R~+R1Ku;RHY!Tb&5mb@jD5{N~cQDj8WN)g6cR+3znR(
zBnq|V2Tp-Ir9O47U|-MU^M7ies9`8K>m=9;^(7mHtYsuI!dhRTNyv@*46Ou8MLVg1
za;*=Xv_PE4Vc1CFafyaKsdd3ppqf%AC2-0b8bb@5N^#0SVWj^#2hP$tu}>!RwF==|
zuIf9>AcZ?`rodYI0+|DwZv><SNoiy#L9&eBwjkUEFUdf*QJ5M;cz@I#7^Ew?Ksv$d
zz8Ztn(it!h4A)75jcV(>Pz%)FIteRf?W<9N>8wje8j4Tuq=llRI8;Naqv#l1TGwOY
z6qHgI5^kY9>VaWt(LMu2h0&>#64da+$r`kkahsT-jeFzn7Nh71<fuuQ#&T}H8b%m?
z`nQn8xI0D>3MSko6Mqh-+J}h*v+kBfFhk=mgJ`H}J<LMQZ5&oZ<xwTjTrH*Btq90b
zGOSq@3@|*tRO@rWHr4VdO#)fE1r>AE)^}++w3Iq27@J2ufvxelOZaVrv3xNEMk}L3
z!641Jud5VV9W98Tt+hUbV#<_9d3`U5)-3pIlF_FO<CA3<*?*OgtS{@{l4?T0WoXZ7
z>EgN1r&{}7+zBvER}4Zs=R8{7I^6@Z7}m=|U0i4cEIqIb0vjyjmTPi)^!u8fuKB^|
zpASF&_w{V}az0<Ih9_^Ht^CvR`TRd&_-e6yK3hIb$TaC>_}lQ^@b%L?et0}PTRqJp
z5qg9Tqe4ey5`S}cUjUOpY`@jW`DFMqVDcx!w~HSa!#CkGvQ~}ok25EMp-lh$1^(Mh
z1ga7VirC1&g`&W{-XVjSAjUHCojZf7q}yb;eOYKP<ju~BV%NFVD4r;=Y;)~HoR-?j
zgB#lzzMQ5%Tb(cF5a{oZzw`e;d%0S@zWmEDyf|OIeDf@xEna^Ozr1<%`sn%W*Wvrq
z)yp4VufChFX3L*X&t~zLi{aVD;@R+Pj+9~^E|+J+SIhI~FJ_m+KTenDr_U~C!?X1l
zFK5eF=d1J6Ifnh|*p7~svDF9TUcY(jsob~sc=+eP{*8D>9Op29b8&HUI|m##R`+c!
z5plK)r-0LT;v|2I8D6^!Yv$-#@`^PKUoS2e%O|g)p};;M43AD%(AhlD`0=m5t-gJ-
zI$g~|9=~|Fn6I$NL(hGPxW39m&lVoT{X^|_kH-(lAe30wAIK}&DA_~nf3fukL}knC
z30&4jeZ4Qf7#=MbXHRCUr^BOf9u7awep{Vf^SMW-FJ^z(1k5~NUG_RMuKRd)xp=cY
zn_aGpnt%8G?D_fWSBu}eYbvT2JoyAOEHM^rEaPUn8;|44D8hrMV;%SXk7ayKwKP$A
zqG+OcqGY1VMCnA?L``e7%^F_2L7nUs>8OR#Euw@-N0~(K(W<C$8mX48NH;}QQ51Yp
z6z%Hr@1K8dHVLmPl1!jx(NF<Pg9vVhw?qK98tNyYh9r@Mq^pOM!EzLoD=0ULBLU<D
zt=lrFmyTSLrkYDb{%v+R8H$?%Nw-pTe5uH})V5>*GO5L@Gb$922w6Uy1d@^V0oHOf
z*FwpW#kM2>YM`tMg^&d37+ISRCxZJU@&Gr27eIdt>{{SkEJ4^k&=h<z0eS(D9#Du3
zJs=t-duc#9Bv-_C{mWc|-c?`(;5-A?k#aC0;O9{RHBjKALoK*}1RUG;2hoZlhOlrc
zDs*x<8Hl3-TcgNFpfJb;&){uqRBDZ(F>`c{62NUQ9Dr^iHb-u293UuHfFbvQ1IF^d
zsri4Lu4?|>F;<i4T@0KiBBP>9q6lw>ow3R7a0jxZn$ZfRmVwSuR@(mNg|9?*-&)Hc
z$Tf<J{SFy2IvCiX0p=Tef!ff6>}B9qB3+?8PdAC~*SfiuqhreHCaClN{R%TFFq7Vf
z>~6A+(8_%nf7yi<4uM4SF+iAicivKOgRp;V@a%&DRj%8BD(fzQs;rUT6`<N+ItNsK
zlCXw2nI?)SN+zn@0)Rdsy}%KAtGP)p<X&(0>wuIQ4#;+Z43kTZsrgw=ir(}B$wZV0
ze6A5OydB<!DgkPkqLwE?f@=GPHyUEJt~M!{6zDRy`&F>e(Q)Aes<;16hDN<QSVVs>
zr`Yv2Xf)(Z6lS=(D)J|?x$T`ix!p{ka=*GQiiX(U%AMKX(-2?hn9`;%17+^ON(C)U
z+l3PeR*+FXLV~@KQ*UJ3eG=@t1-_ABx!f(m%Jl?GQ(C1dt<scMsZ5kklucBfsHqd>
zG*MIP=4lO{)?nGcW8H+ErWTp6vYLNpWbbt6tajI44K=&))Gs2z0mI$3U)(+Rp<Q2o
z6Z-`{LTmY+fpI!I?j7Kv605jH3oKkMnxE`b@l&cT?G|j^#)C%%I)J?oW?xq5*=S>5
z$LR2;k8fX=6g5xwMfZ)8a?kd4WZVvdOKOSDJDtK%E`o;}@Cf-m8+!ch&K7?G))f7Q
zveb+og!Tg^=xzMoXoE>$=*0J|TI?L4PwgOWWr?VFUV-32eso`BDSmsH{0!rk<1|DL
zBQXkKA&{a<Ilw`ZOQC)<mTI)-4&3}i39tzL<(&-tp+1HL={Il9V(Mqh8rh+K@&=cG
zCoQmO^d`kxL{>olLZyOIYKwm<-*SrNXPUV+Kg()<p~;2!@^pj~u;%3H+M?zA?rywD
zP?fhJ;`$cgeeLxPwh__+a731}3o9edh{`Uk<fn~Q{y0dn&^A&m>264|*fpdm6N-}w
z#mR&eWkQNFAw`*xqSXD`^P=$ItL{-yiw%9%eL3(?i`T33SLgqmExUg!H_E7otcNIw
zYSAB$usD1~iE7Y*H|i*mHQjRk;7((}B>)?g3bN+`5D&9oI{<o+_7iu}9KCBm8j6kE
z?<2$-c2P?)!KD<fUCOSI7HiRM8LW~*4oaz2q;MrW9Jcq1cgk0X-hL^bXghD0$`y$N
zRl}s=H+~&X22^i<u0nq>ioap-W+&HeMYK|pw^%JVP6#|9-V-Fj=(Sx#N#1YV2TDSg
zyId_#&*!VR4&nRx_w^!D^83yiu5oNKy;j+U4Y(iz-|xgqsI$5gE3x$B_CHLS(Y9S?
z)b3Vh?D~O(Oh*Z2I$kK#cRC#}=ybfGQ(@AnFzHm7bSg}$dli2nMYn*Wa|NOhmqedT
z91tBHALgu>t6w&w-Hs&gndlzGYpx*qZ+8u?)s&H$p<ZZ68135ung*0X^T&K^As8)8
zYUQ0mNad6=pt>d_NZbxm2GJ)qvPlTZ=v?I8LLk(ulB9}SnXGX>z(W9R%x}Ub8T<)=
zzsLMt72zF}N0fiS;~kKr#sfY(EJGyN_G<vEioPYIg;T{iAi@I(;eOcPcdmutOwr$W
zzDtOY;}QgOxJzt=sN-w6cL|Xbq0tf)VndV|1C{JojX9&qDOdO4aIy}EK9wvadUR~J
zp8S2lZHA&p=4-bZgwiOV%ob|{;z)1zty*{LTW_$fafg4S?N)YS<(wk(PP6*i58oS{
z7#{i)YN7R$K7Y6&WOo7@cFWbW-SKL<(lwxA6Kt>vHkj-M5ek|Wx(ZHzw}vYADR}PX
z@bjB_Uk+XCgq!6Mr;O-F(3W_ig1OmBe?}tcTU**b--CZxHERBbMVlf_>VC)Lna~E=
zwQS))+U_kmTU?&cUtG+7y7jOCdhbudt|t$B@4cvPUtyxp15PWuu;y-~sk`_;*DOCS
zmodgY6ah7tk=ryA0x~w20ed|sf9+aJZ(F+&-s@MKKVXJKa`*rOp|%nk=q`bGMHd%A
zlP=<-_jb|0-x<CR$@7pbJH}mfAxnoQIm3B=GdyzoG0Exi@bGl_aMosW!Xr2}J$;{m
z{vqivNE);_IRIfbOi$0#-#$vIEPWp6W2DrYK8}8S`o`m(22V49W&n)>e;n_e0UFUX
zqG^so`j+%j^xNgJb6#(0y|-TS4yW=t{{88%2br0zoGC5PhRHf3h<@eF9Kq4Iar~j(
z0@F6cr#bV7M--0`7f2s-I^IE<e@`Fc0&SheqaO4F@rscCn|6)=J@IIHLZ17;b0WVV
z&K}kzstgJ(^@8up&&UhQe=tcLGZ#F@)Z%<o_~5Zjd0m%UYhfIWl;dtbOGG9r%t0JG
z$MZBNM74W59i#|;p%t@ny7*i5bmRDW_ap|DFh-43*w(CJ?dw?=MORsVQJW?`M{azT
zU3*HFsIIzP*ql{DdQ~Nu^;E4RS>P#J(w?*cwQyj@{q(*b0ew1sf4_=OAN@oSJK9fD
zXtdtaw+&KFe~|aA|L5!DHh<;3#t5ZtXE7?D8?!9^x|inoVHIi0n|eZ-EVuX!#{!l&
zKdWlZSvyQ2WJ}Ldjvmq3)G<2oO-_^NhHUOx^%ROAYmv7_*z(bRpEH&;?1%H;pKY-c
z(lwy@*qU6U%3xWZfBf}0JbivT{JyYD^iD`L)3@gX|HA+Z>%ydz2xMh{{A2ol_%iOA
z6*hjyoDi)-T8r3Qg%yZ?!k~rLHGnPTca899>~nHgc}{D<M7h^TOuu>i_)W{KToX@b
zeA8=+jL-=)dktr>7{im#njEV}W3f3x#N0MSh-Ng#9MjFse`1IcP8SiSj8bz=zrOAm
zsp3&ZJM@jmoQhc&zGGwybtl#C7WXJS?q-=^7C%r2LJ#<g_<;~eYk%tau`r!uMIL!b
zyR_XJhZts*a!Z^cU*x>o%js9e8S)=tP`^{0G2RQOLhG^HKKc?VM+I%iq%0dl+|)nO
z4Z4x9v#~1ee*`!^G7LlvK6U2Ju0zWu@1Zd5+Koh~3UW;>3?MB@?1f?}@i~SHQ^DdK
zHrJIvmJ$tXpjN_j3C*tx$qeVbhyRD<jgU+}IEqX=!hR}d`rjuK^y3tzO$yM#<MiK&
zhiMT7Piq~AYy%M3&<-E|{p0&zzWoM*7W{`-kOPy+4MFhJb9%I%unH<S{Y;lI6+RUT
z+jJm0G+U`=+E15H6+RdNvzKTUJ{$&4hwszjOP7ijK3GL>&w1GoTJ+d_r)}<lKmyQC
zcZSdKoH7F&%u3+MQ_b9w(Dz4;Ib6g@4|9q(_qmEo6iZlY-KyD;NPn3cw&L;b=JS2I
zmnRlJvVY_zqI{SEKWA2e{pK9STW{P*SK(aCs_U9<%4nhF;so-<n2EVF#pbZ^x+YOA
zp_fvgYl>xtj7}ufhGf&OELx~DS^ymB_!}j~=L&u)sq3U;fD)fwl@2%YOhiU1q>w&H
zsaVx1BegU-VzQ=kOY+vMYhe?Ct=nUc$E`iu4u9i8mhGv-udZor_Lx;=+^o=h#T5l(
zHchovtO*vQJH@t=M&QU{ZM2`(ir!_#%Y|~azjJENCi%?Wu(p4z^09RTgG-3}R9(L>
zG&Vw>Wn=^60b&R)LF9r-dP_x=T8O}1uABk4RhyL977w*szFu@B;J+K7DOpgVJdk|j
z{C|iMh3Jt=srk1-Py;$vfY6L;5)k@V{Ci8+Xx=#&HFr%_vj=xd4a8D(?iF1ntqsw$
zf8wr^&RfUEMr1EFE%a?&BJHT0FvGcjRhLKwOCF7H(Iv`_5tOM#63wa``I<XP>n)Sg
zNun$eL0SW+c3KO4!{#yrm72{UqNlcUuNriz`2rNtK#dzMZDP+LRyyplrG6}TvS4;P
zmvI<AGXl5hx270Az5xMwmt`70U4I{&xg}z#Y`AG&Xd0`oYpga`BWZ41fuy<nu`0Wh
z4Xtw}8WGp-yn$IciL`9k$>Eukt20+>Eu)LA<tz^@W)SA83vwBuefN%U;w(w54-~i}
z(^dUyh}O{b18Q*8wrv!-;jI;xpLRo0lh$-~-s+kPO%n|>&#h9$tWDRv0DqCz4s}Sg
z$IMfCSJ>qDt1DlT^10c2w2+-M#rfqfceM1x+u|yt>}p#ZN^{HA3ZM9MM|oaRdfK`5
z9c{os@hR=LeQH(P5U8@#U3OPpqFcUP5@wl|MsK}>EN7o*Uh=<GiVWAe^GhcU-K($F
zHhVcG;Pb5zgm;Md9if+Bb3c^G5L%CkGKmUTZ=MOg6i|`muM|v2=sg5;0k_^FEGlob
zGVdU)j?l}6Rj&3vWxt&-Ki^PT%}ve!07cezA(t`6Jrn^pmtp8U6qhI*KEQv-&*|5X
zf4#_raDEpofrPU-8H<8(Nm#y1YF7zO&gn@C<i`QQ^cR0wkN)`1{BUZMA4me7?RHw|
zIR}bo8i^m|-^euL8$V1RnfZGdCw`Mp<1~LmcM7aR28EcGe&bKNT4Lw#ydusaNb?yK
zxp@wnG%u{t5hSUG7sK<`Ajp6C$q@6xm-7ZQ<H`8(`9^pF{CghJuQRd{Q8UvLW@H{j
zq~5}KSszg@Yh5BZ?DL-Ij7f_0q_uG$UEnsD#_w+l)AJVmymN#SPvWHGw}sNTv}2fg
z33QhA)a)Y@c8UIl?n31`)2z}jvJIBla7d+nZM{D15|8|nBuA!q77c&bVCWMJDljn0
z!Roxh2@fWpca+z2C7+SoHqpE#*(%AGU^ancO#(8`@Y{H)XFc2evZreqH?0IJ%(zAV
z>U>)1OkE@0z4bCuiAtE|a^yiUg5n02hTr<!vNCSqE8%oo3O*}sr{g-~sXBeYR+&|n
zZe81iS!t;frSutfH9dcVjiiqh8i<*P9ShzrCZDJFCxX{KJ*&>vo50}k`riHe<>Tdl
zE1@FL%ziul_UB9ZfQp2%JWRqeDu4Yx{qgee9%^UdW$mJ(m`G!YS}`$#$W3Imkh+}m
znx)tBlf~n`RT5-+EKOt-Z?B6J^ocDfr{zx+^=9NpC6LHR1U-Ks^({P|uVVg5A!YPW
zGtg!v>W!D13?U-`e7{TZ(R5DM!*$+Zz*-AITYE_K%k6|)D~wVr$}cO*%L-?N(kmj0
zqrDvk?}Xc6zS`VgOAv}3e1GPcOkl6j>9_X#s>DN5Ubdx+U9OT)c6z&8x>`V=C0*?<
zk4RSw=@IGbC_jIeu8xQ&NLPo#fpm3e{t42Rts{7$usBh92Zg{PVc=N{>Dl7N(1ERZ
zw@aWrggm4d->rN|F}j_%oXJtJOA<M#FMDwk6BCf=F_y3}rZpiDLIX=6H4I+ZZtzG4
z`hbifqAXlUD`-osx`k*PfzC+RQ9BxZU4;^9G)L#%oZf$x+J28X8`I2lVbwt#zqKc=
z?a7GMsI0KYta$yl%Tr{<Q2eRqY04(bA!n*Smj+TObP)SqUUq39tp%;*mvZnP4|`uq
zlLnw?K-wS1P6uMjuEEd>TC#zwJuR_8Ftr28X&ROAv=`60xUkG|IdnC3>f@TU(A98K
z`a~|r_{o0`K3*EooI(-dimqEr16$vtk}X-rb_#H0?%3A~3%W;ebUb7mqgYuE?J?#1
zk>x2p%K33PChckC@Y1^Z@OH)02Y=q9kaIKKR|>6E%rCmt>=zjUKPU*ysO4n3m@tz`
zDTWglc0WUAoD&w^irF{H%sACNl}baEZ61XYdN+S{I_2^W4Kce)97%!DKt1hollN80
zWm80v*v`1JvMHF@*Q&5%mtx7B>DN|bDPy@C^2Sdrc%XK*f?iSZ`1!&i-cI=zr4~>O
zauBUpyZV;Iy_Nx=n1OCs?a5zr^*4pfL$VSQ;E><$i_0*_#3oRivG)uzhu>piAKF}0
z!n1!dYH6?PFThw+v^^zIxWC}S$097tSXXf2D8Vbdt>BV^b&_+~9_6_i`H_k;gD^g&
z9Y83&EoY9q2~HQvA0w~?+ss|G{cdzePCDm>9iYR!Y{v@TkYQ=`u?XAkzLUmkquHwX
ze#HZMtmfbV&58q~Jl1+))r#?Be|0lHAiaM;@2<oL*r!-}?)e>y;YbO<T2{-o5@gr=
zjNFxvS+L`(8#U2u(n|c3!AU8~vmzK}9EWPK!sFHmW_UuU+--Uw)Zpk&NVb3-l5KaS
zFz3iKD!0=InF-nqSAd<c_0)#59AM6!n%wTQRq|Hlm%8b7Ex$@kK9AHDfgF>c=kb5i
zXR~F`Yx+qz!X?N5F)%U}H>lpH^sUs>L=J$n16c9%c_NqPt@L2-5IKe-2&qRhW+4DM
zs@v4mly}r^(*4pw=<v9ec-?RFj%*I+526a#<&uK8q4!vx)Avit7|%^&e;|>gif+0C
zpbblh$1k`8(C$<YG{&%yElXm)NlkzID}qs+8}<hUr>*M{!*k25AuaGru1B;xH91u=
z6|QPn>#;$NY%BV4O1=5J$Y;AzjQ^)nUvIQe*s;I4i&D*5ht^*~YO~YLHaM5<%B7aV
zk5hJI+bVLTAHCeHS$Rj^-rK=YmbcbCa4|t~{H!hfLsQ~yi&g@%+l<<^H<W*d4pjoX
zIp|pSMK>9k8$!buU1U(427-gX#jgzSJ}7VkHP#s(*xP%E>=z`54jr^-_Ygzu<)Po@
zt`_y&SneR*)1r22*7%O{ofb7mSzuSZdL*|`UFRN$QIFaR@OtaypOo82KWX7we30W@
zy}}-@bfGuZK<28(m3TSe?0$bGBlpGMyEjCXwgL@3e*dTg^hq~Fx?LW)A<{v5;D$&~
z`QZ(bo`}b9i1aAz-w^4~{PS*zD8=uk>+rx0kq*))-4N+^iTadX;pGXHwP3K(Tze@b
zXTAHK5c1K4wV*dJAL8ahXs$W{<qn3enwq%2pOI=@x&7(t3R13u4VqnIW22wfl>6N7
zwwn(CD5T%w2^`#PMI#uQZs9mu+P%TtaLL_P2|Q4_C-8_Da-_N9(b46Q&>f|2k;+l_
zpkbKrTqEmP+j+mEP&l^<m%eNpDRhxB2j2e&6*V*<modgY6aqFgmx1a%6qoYn11Fa-
zIw%B}3nf07f57F>x8I&*X1F{nmj;D*&{+q9c*$J8NSv&yvjGKo$7UxPfMkNJlo<I?
zjz)=D3B(!ZUy-;5SEK}&e>M53`mug}oDt`zQIZg^)6x20+1Zar@LK4OS3zj6io7Py
zRlf=|&q!LAx4K5<z1_;*m{+U(>$Pa1&It4$WSM-Hf7~WpH}zetgI-wY_gZH%d$3O2
z3okk%s^<uwxy=CTw`?EC94BsXMSlA{{1Qij^pxmg9wV9HwdgO-j#8qo*tlkGRtqFc
zXN~kIaZL=^=RvQ@#<wh)9B_og`a6*sf7Rh)u<VVMf1RcKnt8M%8;1#&!hnLwtSn2<
zhaGbVe@&i!vhvKjV1?7BSU2(%aTj+;Hv2NebyHKly$Vvt7!<_3@tqU~t+({FLF(=o
zVXghk_iy9_y=wYEZ3NoTS$cwm!gG&>tR>87@oh|JFKVR`&V`C;S+v&cDtg3pcB41F
zj!t0NA&hY8qpZC*9W-}I@clAAlKry+MehMYe?PB>z@=!og@QzG>Z_vsOVNZhF5=Ks
z4_|I}gjzJu>&UAX4mPSeKEAR4Uf-Vned6Xrf*u;({q^%H{s0Qezl07zcpClnclYDz
zcTSbX8#xY`2`tiD#3@*;K=d6t5>dM`SdQCZtc>2ot2Q85y|7b9Dsb4L)xz28Q&S{%
ze_5MAQ|q5}P9`JYM`%VZqz6t-eT|=KoTjry%VngYv8Os2NvAHzrXnq!)L{57W{$S9
z+oDOF6ILls319D*+^cksx#YK*<aI?LKjUV)GTVOmUf#5(cxmTWX~$%3v<S-25CPd2
zXMS)L36oc9`e`X+l0i{&g1K@KkaQ-if6tTG`?YSQ@Je$^XXDKTVYK4A9KjoTIb#B+
zbOdkZ3^2)8@MeQha;B@AZ(omNp7F4d>v{6<v+XxwwUKOC*>d^S3=5@8@%ze(K(;Tc
zz)LDGiZ|}V_a&#tjr8T7lzdlu<u}u-85b~4*<s-jPK%C#E3l(rVaMrnNiW9xf6YR9
zJ1H{W35uv(yQ7e@m;{6%Ak0>dBV-2G=NtKt70UuqGB6aSn+f$O-5To}rlnEU=Jc>H
zpez?R;H;2Z?<{Vnv;PmJ=K>}_qV(L07D`*(kiN%u1D3NjLCvxB(gY=F>sfjkD0Z2R
z^s<jn=?`XiJQ?|(Q_h9Rl_M=sf1I>6<BUG2l)K3)B&Sw)1+Xc>M`B2hN>hSMA%rq#
zDiqjcz3lYv&XlP7P{!@!MpNl#0%!f4biOaG-2Lf%(U1zlDSb+iPhp!8TryD3Y1!;E
zEu^JTgvovUXbrSJ%+&D?IW}d>HiW%=SqgAyW_sF4cGCCE^P)~mE_poPe>JftG)?Z3
zbJatzB!f%kcDW7{p{ZurXw?VB$w)68npu57L&KlGZ!;blgE#v>6P$1X&g`zlo^?K_
zwMhv5^pw+v{J9RTzb*6PYie^Ol1)7i>23`V?Vxdfhs)Sl6+Y15ilEXSKe)kVc6eW<
zT8VHHsCwbzq!$j|z0EEce+>XkCm+=8((&B}r5`LCL8KF+PH%P@MIxGd+S__QqU1}d
zQx+OpMrxs*IjNz=sJ&A|i|z72LyJ**prOSj-)?AeDURnJB^~2Dtmk75%b%5e7$+#h
z+RG0p0eZ&$dn6w&hSc49@<s$V!^oKpzI+BX6+SWK*978z6au+Vf5Qo#ef)C1!FD-x
z{QC5pQ95$`@<Ha3uN=QVQRhNDkogBMd_(z5r>{S&+~yDW_ZHj7RHl=%@r3p<qx4Sg
zW46l*ypDA<b?hKH{b)Dya8K}|eo&onGp8nhbngAQQ7--E?5^r^hxCrVC+Hxw_1q5G
z>n&Li40+>#d{sVWe={3vBH{gdcW}FdT<n}OaJAkQ1mx%#E=7#5Gx+@S5yGmLG&&Ke
z1mTiYKQ+9WUcy0Yu{2vN)3(X#%hIX&<x5Es`^Kb|Dp$POtt^@k$Ol;9;$(xY!UcSK
zz06h?C)u|!*Y~zm)q>&_-tnG;R<uUq17%=L677fIbJ$%5f3|&7BDt%Vw{gdT5<=iQ
z_@J_n#1p8Hmwna>O4QD-`eH$2hbs=N72OWgER=W(g0QFESzy%Osl;QuJW%2>N)MEH
zT=MM_k4y1*iN|?ZFYy|dKe@!)$xklvxIW&m#M{YNF7ddY&M)zl@xqO6GRX2WGE06t
z@J0xP2gfNlf6gPXy4Z@xUwUvB)?a!wwx6n^zs#JJ`&!?DHkC>N-PI_|-V7bRDKaIx
zrq29IZo1%vdm@a3)oUWtxsO(n3l*eNHZ?sTuOJzvHSOkVkgb>muOrnT={U%ZccLp-
zSkAhhvYPsn=)o$+Y6IR&vTCb@As|f&U{3U=<>5-HW4U^HMQOsFZ>V$Y?urIF{d{2p
zt6tyHAmtyHFKj!~#ob9Wn!*E9EtJb2)$**EFHM1RL!IB0_uB<~6FKAi?T3nKqL_Tf
z-FCoOqG+~m=RFXLJY>cXxinOsl&k*-z*AtjmodgY6aqFhmjQ_&6_+9{KE8j4PyhP<
z>x(V`*Y^meVdWhS){zoe3Y719K5HCJtjN1=bi#oIC-MRVv$T*v%JR@zp4j_Vd#&`c
zw6OHBNv(0GUoIZjsE9&62)^GjQNop(>-0x7im1m-%S6pWFSZemDZC9nqOm=W5^6J)
z?|j0@5%L*}hVdLH+2kTUw)TG{>QLP=?Qza9v~jVYF7cVJ+331LxjN32)w#TN?uELB
z&wXsi8G;jq>8dyJZ$U%bXc&32kQa+edmMRS8wEwmlcXtG(D*?sS<G8`+n{myL*#G0
z=iAp!nE~NE+b~$;m4)f1iIs&IWJhopVqR3&f*0jzVy-I_Gr6J~M$&)B-+0bKG9@>K
z!qkE;Ko;Ca*i(|Fw=%`%A#vG6%a7ci<jxg|32mup6{u{R*0>~u;I-eSZ83#7P1WW7
zQK8;dDdSk!2pwJCE>Wz&B|~rmjuvRD2_uJtjdYH$uidZD?=Sy*>Dk8~6=@p&{rgMo
zAYf%(5WQ9bX!zql!*72t|L9lktel>j=_D-LTIRX1SjFsRHxU*;vU-)a_Yn3pnu|E~
zb7s1<;5W37B<!=6Z%rl<1wKn1oFH()$EpycaSEL2CPcTBl)DI3H5paNwh)Y#O58$L
zeauJ|Q5So4D>5b$9k@s4V-p!8Zc1S9xm11O`F04%DJPbxg8+ZlcV{tQge$uzJ37)K
zd*ej9&*BBnYqDRK-;5lp0<gS8+mBDlyI0GZ4cRHh3=F$o6c}P-zv8{ZurBRMC=OZ@
z6o=r&igy6@t`fbOI7WUeJ?HGo(9@=?nvI}ho6mJczG`B}xuFmmx=|@8sG|}AeIbe|
zz1+M3HC(vDL7ji&-NR;=xr|Ps(6DFj^$>}TRgM>2C|a-(N6|r%-#+H6%y95J{e|@x
zjD3?>YAS6tipX;klh2hjV{0Q+)+ZEQ#HiKsc*7=Z`=WIxN<bl-v$jWTG%0KTdg-i_
z!c&&}D3?3_{h;%`W*v;gvqrlP?ifZ&8c+#X7?UE8M#O&_$X<#JW+v_FkAq%H$Ca+?
zPiFeQ{;bLwO2DqUVwT_vHNwcEQB$QVZu!>{EiChj7M6L_@VE8friCb>E=YLVL4PDs
zVR6)EF|%5&40Xi4((z_I_uw`XBnPvgIxn2teq<>2oE`V8Ky^zJVC^|NZIzkCC=s{!
zmdd0+@-lx>{u={H(jb<>_JL$D%2K!zND4MX`QQ>#%v!#c;4<jF;F3Hklw7p3_9HWp
zYP2ATB6Qln5w5b3O4!w{IQ8H>D&*;l)lef9*~dUp?YAAn_L*(F)whSJ6hxG458Gah
zDq>^90-FcoI+PTJEG}SJ5A4td<-ql$w~hYoai)I`ih@4~j{%b$a!-%z=JsZ|N=SPU
zE|EG63(npbKNeW42BmLJv1&{5-0NCJs1?8hcJiaaoK?IIsPIBxMJ^4xXX<OW;vQv=
zb5R_QiiDUij(WYS5lC5FAoqx}X{StB?7yXKu@Fjj0qygY4G9#<$_L>Qk8Oo(Jw2YU
zY!ZL^tKi@$T#{uXF6i`}l0=uJ6?=b^e<<jSV=X#_oY9}P?CGzO<j*YyA(YS*TbT_)
ztc4%aw78n(1@mL!A{Q^_!KO2@o;Z+X#^xy%g{q751vhvscqs|w7u);BuWD?@-*@Un
zsT&=G3r(!9`8BfirH&^Ef~0GTi2L0h6oh|Kf99*!3YI>nY-uJ8)kssMUKn$#Dg=^9
zwvSwrp0=zx5EQr1Ye{9)tNW|GIib9}7`dkQY00CJ;WldufiKgTv*K(jE~{d6eO0SF
zkA4S3KtwIH!+@^JkYNTBza33*^GLelz8-;D+$$U0hq=lI!&pVAI4o1R((^ptbU1&m
zS2_2SKjUzsDtPX5zIvQBIbWe=9nMfLgO(OIW5TsUt6@?ea2D2<gE;hjwZ>@ck=A=V
zVM_DethdK3*EMoxf37>tD}wfQdF;4`eVLmN^>(WiULrPek{!m+JMK7kjwG(q;sT0H
zO`cUr?Rd}t=DpUhnmMc^XhtoZ)JcEbr2gCZv8Xe1QLwQnX=dD=mme4MLlV&NwcvhQ
z!Gf_uL;Hn$E51zpaks+{Ii9*wuy>050+4&p_KY@Sj>F<?Pj(LcQ%W*I66MS-C7B~B
zYadbcyDX&hSQ6<Wokw@8-4eEMnfF-g(U&M8IS#nvVA6p?y&&jb)u&WlR?vUBHxUiu
zLzW<H#JItxQESw^a-4TZQOuA?yfOW5TRD13OJdo`IeuPR(k*x;E$K!-GewWJ8<Fl~
z@E}qO$a`}9fE2w{A`il4tqKyf^>97kjiz+{^WG9})$R5hJ|8%cMn^&<vvmYm{9C<v
z1f>|rlSt0a&N6IwHEBGJVH|(`<5u+3{OJZ2+9?RH**@Dm@a$J8V=pc2O>u#3=ImBy
zhak>Q>|N*VC`j;Qeu}e0j>j`kEN9LaeB2Q|UP>rsjyYhDRuW2mV!miCBs_FVkv-si
z;mvwrt{!$Ye}1XRAS$$Q461sCXGId8&)s@j@l@`b7*(O!;BG1t-CutIilt}^@c#<X
zU#0?tM&yn0Z&d+82vdjqH?08uh`YVuX{uj$T<!%+$~*rTz1*9YcICl(2d*hUUX{De
z;tNUo1HQw5a)}qiiOuWYwVnI(eh@cJnt#%rUNDqS_$D`MmNY#=(COXR#@eDI`f^7W
zl=GSnz?BQI55+{~FsFap#Vr4QaWxPsH#@qOW{JkR+N1FTTGf=%QZDY6+>?Zapg4q9
ztIZYE%-1m|nEsVw=SJ5o^6#sRezxMQYS+~Rx9^03a>4Z=U%66vFuUE<Z;>s_)NNYt
zj3qiu4E>k~jQLc_zjp78N#rC``Ymp|GDce9?AO@1dMf(`3#&taH_F^8=0E42!ozBq
z&(o8G8HUq8(PW-ZGE(Yq818aelmaC_?}6N}r^ah{!FQ8><ey<kiDca1FZ;LH^FIL}
z+ZvZ4+C3EkHkV<qClr(CS16a{Ha?jFnYVs7KJW&AFTo!2ZxlMBNaLzY-KC7@P@)mx
zd3YWc+`;p>{fe%-Q(ix1S|d~0y@TWWV2Z7^Wv1yeeWW9*cBDI+N|7gsNrBp^eIE&R
zdvOjldrNOl40OYaCq%ko#hp+$%>Dtf?if8K*p2e|G12ZIO*z~RvwyC52P&9`J<&h(
z3-0-7hko`Ch<C^6De-O;$B&732WiUjZlvj_j(4L%&xv=#s$206z0lU5i06L6Jv<NF
zujr!d9UX{w;KX#u{XOB1-J~(l{)4eLILt!GTdw&Y|3}Db2RhfFcJ3j!{~<Ddb7lKK
z`}O&WmodgY6ahAu5OppUm-#0sCzmjj2n3hFMK1)Gus=SO0r;0iKt5!DkmRzYJ#eI)
zbwVDmw<u-fqaM360e@3ySn_$@$|h?t$zF9Xia&mR5&wODfBEO7SAqpat~)smsyqGl
z<0U>u!<_i;M5KIYPCx%XeS7&`(k$AQ-^o0a!nJcmBZXV1h{DMsC{0)uzAWXvBJQ=K
zFG&jh!n`1PwluIxchviTMYQc=s|tB&|L5oTlX&v=%Rr5oN<c}?yv5JR^0i{gYNaC$
zOPT8E$=B!|ot_nz(~3HftH9Krgr7|*d9R!?k`jI1zw=<S$BLcbCOhwEbXtL*$jTOH
zqhi5HwK!O`T~If_*7`QDTWgd7*UQQ6-yO13f+TdbTFEJZZo=+=BWeW&<Rqm>)LIQh
z?0iM7!6{=Wo2Yr>x}SwEs4ydZGrPRyU}ZQ$iqZG13fvG_<}9g(C!-aMq*B!n>;|+7
zQ9ffzN1GCK3apW<teDb#1YblGB{fjYN4%R#pUx@cS4pO9MCXJf$kbd_vo~{4Fvc{@
zpRM`ShWd7Hvx72!KhDLj|8GP;&YDjV{diY6h4Qg^be}&XSFio(JK%s^zpFI2<n?)+
z!wONFW1y6yTSo)zl(zG7Rx8Dcoj2$x!yIAXMV}mkbZE^b3G!OEm4w-?!Hia=Nld0y
zQkrv%D(uIRG=#+NZ4fAfv__z;<~4y@@~VR%H_}I+4Bpm%G#=k2STT`if)G-mnZtrM
z?wE$YRA7+;%m=3{oGyW6N#Tf{A|usP)UqZHq@Zl@Vq!InD1tk{6WP=ZC+9H%Q>~gq
z;#zcopg42(F*r*6QbeSS4-a0m#zs%ka^y4EM)hczEYv6utiRS35l5M^f5vzkf{oFd
zKLH4Gyq!FM{kl@K1XRSCNAM_Z&%O>Ywy$MMz0x}z4xHVI#L?i<$VIEi7QrkbWZ#cy
z!-Jg{<M~4p+Dy1>X&fd_a9pTjcTb#P8aJ)}_}Yl}TJ|v***%dWCr^uEB2&t7tH@?Z
zGDqzgW04nqP3*XEUt?_zR%vH%`Q1Dy-n3f+r<}Hba@7h(_7;oCNEtceh;orD7SHRh
zmx~$Mq(4+()ac3AQbI0Zo`TV^^IRQuz1rSNpt}vRz?y9^x&U%P@fj@ZI%Qbcg5m;}
ztuS7UUaEj-e_d12@-}{~_6oWQ_v^<c`*}j6&~D}l*}zzRQtO8u9L%2@RLyX)<M0T>
z$W_ySnBJ0El)`yD_(-eZP+-^<HgRJu^7KxM;$WL#9D}%(BWF2|&T{ln7;OsY$15^`
z7y+fux&nvokg!<03#X)0n(-Wnj37E!zE*Sf*;$UW*D=9+20I?T9SI$xfj+JXE}qkY
z^+s7n#)9gc5^6SasJyk4GXw?)!=JRX(UVqx_%+iJDDU9-)A4hK$Kyb<{d9clY(w$1
z$=}xWGAf=9*z*)Z4qbs+6dBdK;qZjj7BttQI{IF4$y7P_XVuQVSt~9oyfULSM>WmW
z?YRl=lPwkCL}@+!A6qJm`8bo@aJ;d-8eczP>4Ngc30eI|?G`W`TK%Nq;+$emj_$XA
z0F$+y9o0AskZ7O$v%lXx5Scx%2O^G?a8NHsQ2Rx_7_rI|y%<61iC&D{`Tbsul>G6%
z7(h%b)!pe5;VUk`Jy27j3g{)}89hrlBxXDhha?N@aA;=vu_$soJqh=7ahL_60w|6X
zZB{6~=>1NUL5iTbCd?}S@r-ALSsg)t@US4wYV70Ddm26q%^t8;A-MyZJ%ZAjG)v<~
z%ov5hM;bLYjN-gqY>3=%(5%P)ZEcbrEfPyzEN;XGgo!Vnj(Rp{<0rEVG8uPwRyswi
zJsj&9EX+w5)|q2JQ_f9z%W!#hMdH-N@KMg3kEP39B#v2A)2J0!R%J2=(T%HrUA^?q
zj9Nq3wslrIZsqg^-`X;vUvJXQo7|F_mkTje3R7)wPz70Hpl9XGjeQR50_T(AeC|5a
z-_Yf|!fBjooMEUEwGHUrc4%#Y_s+@0a#GPX3SKnLM;WAaNIM!l)Z9@|L~B|kw@)+H
z7-hO37Tlm1%OXY;+se`neP6DB;dbCI*<yaarMsL)PEP%}&ICtoNW{9L>)R~txS@I3
z(j27f=6cw995ewFtHtD|OIv^?Z_6{PpY&9BX>TG!H^^`dGK#T?J3>tY?Mt2G3*LC<
z?1O)z;5ET3{1vApPZzu}N<qUrOQX}07FoC4Vu>^Di3-u3m>)HOLMBarVax7lC|==c
z-Qdwm)^MJ(8e}EQ(55qZcqtp9EBc`;u_)ZVlI^b!CinaNa6xKq;Czpg(Ur2yf>&Iu
z$Qe(SWtR#R^0pqYyCBM)0k_=FGPniKQOx{AQ;(MlfH^uy9_C90h$jMQ=hnYlcwL_W
z-^Q}nHKoz|EvduqBM>luO2o<P@;<_1->AHwHMZeI&OhQ(qhNJUj9d??(1F~DOvwn?
zBw(tix8){8dAr;sYv!jEaE(uB<Atv>cSn_!XH@9d%x}Pn(X*GdjApp$MtFU41<T*N
z*EYUbMeUWWneuOcRb=DQgbRY%Er>6gtpzc0;l5g@Q%h&zz8Zpm`PBYrhl;cQ*J`ZM
zRs}_wLb9c66-=G`3vPf|4v?U|+}2)?y_HS8W%D1#4<Y|x4~=Iq7zE!D6|;b=0E=Q0
zj=J=s6_)~wT2|du2)|O=<fc;T3ohmZCy3Eu!}^Jb|1@|1vGwI;H!ZgsC(bvzC5bJy
z|Am7Cmjw`l5Y?6~V-NBQKu(-Fb_h)*-LuQp3z5(%#-2jI`wmXfS>?SGL<Huj-wcfZ
z1>1Vo)0Z*EJrtLr(K{8Fk8>v|mljDroB`ptvq?UK2?LptFqZ+t1rr4^GB!Ckmr+wb
z)qnVx-`;<Dl3C!NZVvXOHI@X+RHO~uh^cAjMwi4{tMnVM{2%f&mO!&~<kNgIX$&Aq
z9iQ=hvM*sn^qQBLG)3iQby^@1zTXxyDw7QCs#q$Jk0cikg?!xi%sL8rQ2JrNueYE@
zuwmycsSe82*Tg5uk(4RC3Hl8Geja|ky?;Ob@2P4PmC%N{C?l-3r~mx<6dn+0VXQwP
z3J-Mp>%XVppZ+yc%?4Zj7m`T`9scmQzy3H4Bfn;^^pf{2jnTTZGa8|_I~h_!YASXv
z3J(Se3Xa;$cGR}>pd$iQZia*-J93L0s#>@Fb1k<s0&pZW_lYfs;IX$9aPpSchJQej
zx0e!Qz)OtGCZ@2R@+a=H8)<fl)(Z=(T>>+KAKmWi3Xg(TM!;Po%yyN<y>FD6ODJsw
zvd@h!;Zk_hX#OoPohHE&0MZEOe47L{=Ja8j9mO>A@{G<EH+c;+!-!uyP8r3Jc<Qgl
zC&G8OZ%PE)i4QpuXf+C$oMShGpnviLHI6Hz3odG#vW-f|O}=`cuhGJi8CUW)pV3n<
zK~@P-{dP;rj?Z05x>a>e2t^f_nXkKhQCbVXw(-9Dl;Py(^tdYezRLyV(t1KDKdBV8
z-VjHtCq#ksCUSIa`pBY?#ak+`-ka8`YNiG)>0_Rx!EAYwt^rsBV~>3aHGl0%ctr=z
zzz{)L_QPRSRAjA%L0t)(wCc(uSzY<padm+QL1xAINCZgEKAdldmAN>t^kF&YNm%J)
zB9#K67_AMfzNXMblopCjQ)uSth2z7o1gVCRR172aIoFMKZOHS+_(^MOtTi<!(zn_9
zisLQ(hP>TXsWz=FJnkU7hJSWkziS*~80Ed)GR_^BA#w~v)TTBwgbh_@(3Y#(-N-Bb
zTn<iz91c;ZuQ7ie3Rq6)tl38cwqLb`p?%a&DR7V2Db74#CsbThNslrfDXI(>hgGw%
zCMy6`rQO;r>3|C&oVUwenz59-eLFcV6muFl-m?=!Jgv;QYDlbhIe!O5n?>zsxpiwJ
zTVnkBV(~l$mY}q7mQI?F7EbES^&<6v;ZQq=0m$;lDXv-(F|I~`-8nFlW(aEsPRsH1
zx^?(lpZr?LLD&_TPU3qwv+O&RP($Cb?~c=__Z<$-so<vra+{8^2Ha7&O^%&UFDWIb
zugI(fv|J#5<&?1AvwtF4ic9=f=ayuL*)3HBQ$EY6>--vw#h~qz0_igs-yogUu5+-Y
znb-DLDqf&2rQ2|A_tWuPc+HArdiqGxOw{it&D6!=J)Tq7C(<vj;FrEhK<|o7pI%p1
zwNi8mpEcg6kzfJ2vWXnQvI5yQVQk;SNhR42CyBJCJ6vt9S$_(VJTJYYs>^G(!U|sa
z4e;O4SY*oJB<lLs$`v`^8rq{5?d&Pe&z4+8Lz1m@wlasMD0Y69=Drr~^Srn1eH1T<
zzLHB-^a6ksdz&7hoiJ3B?KCcvSF}C=Zsh~6nUkbw{IlKvGV@-XCN471H0?Pdm9q{+
z+{;KVS^G|@BY#DV*zGr@{GQXpDZheE-^NYUN;Elsc{L<(LV5EYs_8g=hidly@nh<e
zC`f+?XzDpVt}e>sKXUZeW6XD|OF1Jnd<WL*IsFsfHP`5)>P(jr$Kaacgra$9*x%&D
z6FO$tAnlwRQc{IgddE^MF^g#WGMv(TN!-wtZagpNrGKRu!`(7a*4A>E=ysA~DSnr1
zu@(J#QY(Fke70_O`r4edZ>8u#S7xvj$ZOua8Os{xLX`Gg&x*E_k~$4R=4;#a#|l}e
z!CIDaUhI_)(NYhNr79OEhVEFn;r{iqinXL1hnv#I8UAm)#1xF1Pxb<`h^CN_#I|eN
zVlh^^{eRo8+WDT5bs@Z`|HFI6>fI}~CXyMI^jTrQ)vZ!1px<+I*NFZ_tMfj-`smdo
z60k(8t4FPc_3+r$W4u<JT+iieMRXvR_S#&PGd=Q`OCJo%3x(m1O$w$}qhaqV?D)-H
zxt|=U!;d-Yv?@f)xMC?ukIRs5Hb)U)o4w~?<$u~c%$`@k$1gKc3Si?NlLwJ_y<e$g
zH_jfHiH|)*hh^eAY5Yp2kKL};`@;p3a9mlIMSCGVR#*42bN8_EH45oyq_*dPhSRoJ
z<BHEXU~%a|hKUDSD-6Jj|DDWQl6Y-ZT-KT`IfR@5RXe8gx|R42Ho`EZT(jIs`WL0o
zO@IAWrZl?Jw`rmZGwj`#PnngQIOSKRU;4bIkx={5iFVUmNe!WhLMr;{5K%g6L%#=m
zZsw>~D%0Y-G*30Ht<u6#cK;f<P->Cp0ll=;Q=fa;Q!V>d&t~a$Px<W>fRztO&I3ya
ze!-@;wu5S0f>J{<wfgEPpA(`KmssjyG=EA9F0n&q*Y?+fUott_UInc0QHM^=o@>GZ
z*WA6qC*&}Jm8zDcHpMoaz!+1ED?f@>KskY&=?AnzUN|Pci&hSxlaRZVcV7ulay0;B
z1S$XJ@WcyXL_Sd8g(oRH4=X-X9Wvq=Wtia;5(eEk`%CdjLLA5BL3F~HmGobYPk%Dv
zI4Bd(@>;EQ^tJeeF`OU=eJFoO<DiXDEayOR!wbU!vb=c4C)Y@|8`*gatw8aCiz@wq
zq7|$!!rXmjuz2pEiMLP11Imy?aHWljt;5T|WHxv8AX*BqvP}>Z5DAU=C5Q>6V>D=|
zhzTp&Kzpw&;UUC?U8zLd(bwDdFMm%*#H8dA#5C+-6ER^|lyQof@GfGiJZeG5!-y$q
z!=&q>urHjUCbTVTn(%x8HF-`>5=YR4rUgxCwm?&!(}AX5_$|c5S+Q}0M7c#wwn0pZ
zLw!-tjGN7J?YL&9{|UsTd}wqGTI!SR95X2aoI?7=@uFDTd|kh>v1{49+h)1?%gXQP
zHeJi+`N8YA_+qJbO?Nb_x496FSBKRw)Ko5d+w?BSMV#yelAE4zc_C(L+}cmxWzO$S
z-R8aNcv~Ms?7(YN!#k4u{=@~6%l4p{gAT>{pZEmdFNW=F9md^%0}K;ImvL1$6az9f
zIG0g1F(a1^GeHiQQ9n5+f8AP9j~llUe)q2+(3kYVTn>li3=6|R64z;f3u4%5(bhN^
zcW1>`dX^5lwc+&F_ZzOXr?X^OJ}JI)AL1KrxZL4z=9_PZq?o0+;3(#r9dUtFLU1ve
zlqEZ6HA$1p(#C0a8ir@Z)fAf~j!4@pivPuwn^+yCRPNd2cr{CVf2)!s3AKrMO;d1{
zg(1oC5Tk0OCgMRYsfo$O5X96Yagh?67}>>`xebUevgGzLJk3qT5{iqmToqo^Qk$rb
zLkLZ*j$@)WZ@^-u%}eAshxQsp$A#N##EwN`6O-c_(mfNiOMwT?s!LF+X%-;XHUrj@
z*k&Ms)`@i(5M4?!f4A2P#M1O-3@9Nr5l@Ah7!_hOWQq7h9hBtYGAz}^>~f-}WE6<G
zO@`TX4sC6*Kup8D5u!GIvF<FfO@{r>%EP>Ps?CxG%Ys>kPpE^ro4!~}No_L=7F1*~
z6lO1yno(1RsKdO$6^(5`#?v}jrvxWquw)X4RW?f|brNOJf2=x5)F#Uqn};3B<fIIX
z&(Vp7rVjRB0}o@x>EI6qCeUcsAyW!V8^%_MqV~0%oTlby1)6Bi?Zf!ImeRz6d1csn
z_@Jt^u_e12h8-&Ke`?`jZ&WVBQ+2gDImGI!#I`m`u4-;}C<L)KePNASWBUX`t=7RL
z_Ms433>9-0e~4{5+q3|8X=Cs{7TuO=TM^RWf)+yo|Jpaz#?&Eb5Whq;m=O^SS(^qH
zv=j#)P%^nRCCzmywXGL(A*j%Fv+yQE1CyIiQ+wCegILb@9z1Z}_w)1l;@QQ?*~Brx
z<*DnQob>%<F>`VF`PcXT56}9Oe&T3&@OVD!A3W$De={Zym>mzh#|<<PtL=v=7ZVRZ
z%)beC!;gm#yC;kJ*|SOiynFKKarevQeSiG$A@t~e`Tb&oFTa{Vp;<qf^_LAON}J>9
z<Z}LYaW=VJqKiHIY4UP<^4<LX^MIcg`BLLCW?0}|8x}(w_U&voZ*v(^HWltYAHdSq
zGJxeuf4X{BMoC6VN2#u)8b+zv=6GC5`>gA}n=f8Y7EQ%)-2JWlvHKoBV4Y*E>8yt)
z(Psn+V$&0lgL3p0e-Tq2@Ns;vd-nFUw+WA?v;TD8j?<s@)A_7>*8TnIkM`d;ulxSu
z@-JQYYTCbkd+N{TZ@S;!zPY&na`LYG>7;-CfAdBEBh0$^75nUeJMT`<=cnB}@-cYo
zE*EFrcZ=!ESCdP#{$;;7na=t(5pVqIsFeFhLVtz<*LLhP*O^&;URx*7&xhMy=jrm%
zujkW?7njplv$kBhe?(hXdrj5tOQ+#STY3!aGh&}L_zI}xLO#J=kdnLx80cGv>7R8!
zf6RZGcaPjR7Q7W`6cX}GLMlPhgc>)uIcN5UD@&7eu6Rs$!{q1vtNukle=)TQ-kqF}
zme&JV9+?W323LXMjZkp$u@RQwg=E9<3U(Dlkg#EMWYR50zo7ypdyo`>gs+B~l6<AQ
z@qLZ~tcr*MPT2V>^>8Xwa{;J;PZJmne=?I--Qs&77y-Nsye_5+f)lA6Oa~DUdgo#l
zFUXN9J}7UhK%id&^d?J<2}!+0;bBy;&;x~DsUUDtEt@JJ#d0l`qgFMCKnRa2JbFIR
zZ_6$J{MWxNfqRe;kRfujx98`_9~&kZ22z#CgC^`SGBfN2nz+MI&g7-6fB#@<fBjk}
zzn0_oNb7}kBdr^;zEN6_yq4B$91}7xB`Ygdl8lm$(khu-NwJb>l;S9j^HCb5yd~-1
z&@<tS1Y!Z|>oZtsG0>`;AVDFQ6hr2MT|qWQ8B_Pl4k&&+TT!rL9ufW=<aX1;<)nY{
z=Hz0P4VVv+hY?R!HzbK3q!>X1e-N-@P081y8<x&V0mex{?HF)~C^!HPMK$C>HJOtK
z`ysMSfGQF`kn)BKRyKld218>M^cb4<+UsO{Yjyf^uTkmxbVs<?+gbbE^uLp~>%jl}
zN0IU&Xh0EfeF!G}MP-RX(`V}mL;$RANkD)&ftM)E6mrJq@4xR1?qT&ee+SqCppuY@
ztWM@o;Scg<lvqLy3IK;dc)xb9sf=X6{+7HY4Q3=hq(Tx=g$EKRIh+zGaIBERv?u^d
zBS*Xq?;^}o1B&kuS*eQF11iDEgX6e}_#o?4;W5hY4JCv?$62+q01$+=#>44A71Rvg
z$S4Qk5iu&LS=wd@4pOrue4Z=dLVsm1Fn}$+H9b~f5Cax}8@T`T>2J_$`fjoWgB9w>
za25Q<d@P1n%m-X{3*lNHF%mA=mt)*zXd<{n$~z5399!FEhoMvv`k+%^4LReqNzORl
zjGS?~B4^eaT@t4_k}eMW^-b_MR#{^&>ZR&k@#4rJFS1|D#QFtly}ecaUVjvX?KaZ(
z`odB`fR?b|(ju}St?eH2J)-K)>|V<Ek8=3{gqU|J{jLvNB2mPL^G-v#1fvDJ42?lS
zXXDocFDY)qOA0rGm$VKqJXUQy)_FWa4%hvF#;OPF;ExU3<io8*;r4V1wycLB%Mm4l
zh|8!9>~x1E8`!v<Y}s)OD}MnalUP;SF$AJvUsj^lEls1HtyvdBzEwM4PG_&qCocwb
zznEY2(>Ewn7Au~T@_r~-U}{<`z-#MH-sXZwX~24`#N-1QO!O%(wl-o!KvY2qLA1kA
z(ukpdJZoA*vTGRk*|3&3Ppa(H@<y;$)=sMI%ykK2X#}%0@~1pDt$*?ee|c<LWj;WI
zYFR#KEzADxP(J6HeWfp4KW8I9aF<p6oJD-q{6*{M%+|F*)o4&<2l&=}-7MIYLRC%*
zPfm*Tqlpiv0*F$ze*Ut0&M|x0*3S<B7?d=i$*fjJr^EHqGfE)D@~VrF!R1u9WC%g6
zTB3x6WSpus5z2l|EPrfyDp!0sVH;}UTL8>|oGhj%r{^EmtyK5R2NFE%Rb4LDv%4d!
zixG@Ce~=3DPKP1ahb{Ag!i_8MG*lu2f2W~vFdvHi5vk)^w)hx5{_NBtyqP+Lc1ayY
z$_J@KWG!`+@yw!(XBH)o(s(XX#uQV=6jSN}Qima=p)`R!vwxj{BL5?EMDZk>4M)+U
z(GUVIluOE~93pd|l(Iay(jl=Tjn$7Tk?agqBuK%2BDK&VPC%`}Uuqqdu(Lh8F!wPf
zDAf{!D6&UQ75O7>HN|IuCc*M`4K(o_k#|R6r7)@&*|(Pfh9GP0Okz(;cVvaN-6lE2
zbrM@&yTRB_Gk>CEsy?;Ln0CCrd6FO{AedNDD<{r&te*~WjAe~;YG?R@fM$?&#osEH
zxjwH*Y$z`AQ`N-Uh^y?8b{Uqyl|q$WhJq>jsQK&ZrRbK-p*PGNx}G^?EGuNp2r?EO
z8V-;ffWuUV^Jf7*N0b)@+f*3zr|KWyX^5ge(G?^{wSQ|lw+e}$Xf)(K+pSchr?n(N
zD5lEoCV`bOu*-7)h~xi9Gv$LM@QKTX4UjbM#G`c!ZlNTyi`_d6CA$Kp<?A8GRW>1R
zc?IfoA<phbUqRe>pSO(ndCPd8x74wlGF~YvS67OLE6B%(DdnrL`RaCJj&RxX1#c(j
zkbIyrULa4*AWq<$U-WunV)M#6HWN}pNNu|f@Ui=e{{wOYfajM{dOj5aF_!_WE)xPX
zG?zgc0w$M0Xg-%~Ue15M|Mny^!%}z?eul69->>ga|2$RpQ3-AQ$$%F~>h#x-r|^P6
z3uC!i3d{2T`}gVF)9*diRyht>Ino%dXUah%w00+Byinm)QjTUpG)PG-sOE=W`C&on
zg&&uHv+(?iCFSp+jY|}_L1{i92m}LV%YK(pdOj7GkYqL$mydBiDSyju<hb$fujn5b
z4y#y2Qh?Eb-7`HvZW;I%<P>E0w1@2fA5wgZqSUnd;XcSMQzYyCB=zCXgFgIvdH;C%
zel~P)$}>1TeEe|$`Jm-5Xc@G8Z~)30IDGtd_}8n}XytjAhqN|E9w+Z!e);&#fR`B{
zGeAZGj^7-Cj2Ia)GJgkIeziO(?=Fv>^Li`my(S7f$lhS`^M3sI$3I?lM)>>L!w6gJ
zNC{2_+W?JZHx_90o*=VH|Kmx1A%7DLSg@lh%_p12020>mo6IM73<S|zUSiV}T1=V0
z$9GKIvQOZ`))8a?AJBnF;3Y8_O$r=!guR^uyTAp3Oh-1aqksLQC?XTpWCI_=&#&&k
zpC2#(d#T(+qu5RdYpn7P4*&V<C43-YW!-c@R1?tQ@BbeDc=>0)YZg@h8#OJ;PFu^o
zkR4jZ><^ZV0`f(6_c2uj{~q?pgp+`(fBydS(2e~5RcS{pSO}v#Y~V+D7P?qh1PdYC
zuLu?rX~?)mT7PK96)(`&>S&}V=<q4(9sihBRiBUFtd65aY0rrvVTL|kO}G*C*_bN`
zH;hU|(XWWao)Qre+bpagkRh&HdUVk}6%`f32Qby(ZQi$Gd}N}_$`XxTYzKqQMAr~p
zz$uK-Q0X#`$lhb%ojl&KNjzPR{=V$05Ukh$uP`r;@qd(jipiX+*kpAIoC_Eg_eI5h
z^^S6zZ#a2ju<xNq9u-ftjknQW+4ah`M92Qjkn*&mKJxvvluMOKppNqHrzJ%4#fi_e
zkI^skdp09JBEHW4Iy{Z<Ef7yF#5pA7y>wVt<|civb3c(&T1WG`5s!;j^TY~x;R=IJ
zL_BboIe(ppyau+)@@lsl4TPb_i^dwZ-pa3?G$wuC{4&t#inntLZB-4M*<7Nlt?JRc
zk`lj}B|zwB$8XMFBW`Un(wx19&Edw*p=IgP)0QnZc-fJNIt<N2L>7jYe0Wm0XPCq`
zpB3&|egWJ<S#Fj=D}W@Ap$u-%4ro%7u|4vm?0<m0B5`=#ipOLJDJ#&K`ZhbbA^uyj
zdxjmP1%DvRwT_W5?YbC=o?_Bl(l>TeI{2iH9h@bL)vL|WZpjP86DTb)xi}I&QjH}g
z7ZC3r3YO^iqj8BX54|%_iqQ3YZ{U6rJ0J1`=y#wtmY=fwqQ<bpx3`Mr>C-rZD45b3
z4u21^LrM3orGC}eG9Q;0IirloxR+d)7vaFD<M;i3I!?dI4tJ$mL|);vQpGO}zvNJX
z=R4vrF^GS|XUAX8`HY>Bt$Es%zv0#Sk1$`Xz;#4DDVP$Fa@Ia6m=Z!P#fs!_X6Vh4
z5u~v2-t?G^BBSV>nXq5+tjKsvR5700^nVmjxe{=YqKCZUj+!9U?y}YE-i)SQk-Tq#
z$vMrTAC16WE9in#9I+MGL35>nCAAZ5L`_$srqi#pb7)|-RJpDua|_%n_cnGZ^h2%2
z%G^(DuGD_yNQGFr+2CAO{lxVuxvu#FI%>;G4<uH}+-hzcdKQz+b7_iJ>WUN+KYwf*
zElxV(@5-7;ksrr-KxMS)@wu^HL}D?>)ox!HundRF^qPS6{Vd*Y&;T})`QU7z=d36{
zQcLQ(o?9zW5eoU&x$ljqajIC*0Z_(0JD%Q0cGqeg*Q?+~=slfHh&6&C$L>ed5xg~~
zq9cX+SuqtQStoD4K$il8V>fi$K7X$&NKGJqh3qKo62vN|TQTR^*W<vXU|OL7!+LlF
ze4mL#jtB1kd_bcZi3DzKf|)to4DDE$C1qc~`hmJIF35}!i9PfhM7%C(M|#_?Rpcvd
zgtko|=NAE;a%fEJ$7NtytLIQwA(;}I%U;9IObD8jZQAoOB&OYSQW1AQjeqGbSV13t
zA<f98>q<|=plF&y``f%Sdg1rES<og(HKaqvvWD>y-}2Ey6YwKd%ap;_@2d!LtT0nU
zykfCoP1zn-^h%w=6Spbt7)IfKmXz+2x1lwO%{zh>%qLlsRL5Q<+d(f$N2F?+H3z5M
z63TCjWUY~3Ng#2GGJ5Y0zkm2>Lx~v<B;#0({$WMX@^VPANasr;t1I+iWmvItb5dTU
zWZOg(lHOzp8#<h^9azo+?xyBwRF*0mGb@_--Y)<zC2>>e<uezAMGn(s`|vs|T~Idh
zhb5DUSz=g=gFHIzJY)8_UX`Q>8nLIS97z`=88x8^KXb_`593|ECVwLcO>9>PotC9#
z1WXI&<RUCHvMW>4=%pf+n}V|xS9wL#e<1o25s)>KVVkcJ!Bz^qH_?z`<$NtcD*1&;
zdfD48ESeo_8(<>3CBUX2K=tihgHSYndOL2!BA-EW6XjEUDdqPqQR-p?cH8hJLG)95
zMo+#)w4NVlMJt)>1Ao*6EGl-;q0qASvWDzCP}kwm3_Ae&{J{GHva!Hk0w^86+Qw+Q
zdiPM{tU@W_zm6IYQlI-?9wpnfcT-?Y=qB=pz;i13C`t@T=?kU@^so?cterRBdEDR>
z`yL-CmpU%AO^SsYi63*;i#|2$BHQ#LgTL(}87YI**X>raL4QGKT(_q~#rTJF9l99#
zooo5+<Hl0UgBTH+dNU!kORJP}0})~Crr~-=w#t~P-|+#ZJcwz8o>8|m>g^MoNh{;1
zpM6gmpKrc(J6@{ebN9Z-c6{&Qy4HqM;LKipa$BRv=kZ;eJ6J4=Bi?V?Nm99y*NfOU
z*~59h-xmMBjepf9s2=WRwegbuI=7{IebLE3GcLE^7@%;yQm+&prwjq-d`fIWgWF}s
zB<f=P%Gg5xA;?&tns$@Kf%#ABcef+02^V)GrbNzdona&2)1$q3ti?Z<U`*~7ZwsYm
z-?3C#GY1nNtnTh%T+fF|k}vR!Udm~%f7A6<FyZa<V=A*U>yw*R_n<3Qz&!0mcjs-S
zuSeU7G41B@jswFh5#G~om#pCb)9uP<mr;5?7MEhLCli-?dOpT~a=o4Z_2btUSrnEc
z*!VSm@L%sgUjFma)K^-d4JRwTP~M#W{ntypVE~17fvggajQacc(;qMY;8NRIs87qE
zzyCb3mfs419tc2bE#d&6vI5bctPMiPTNyxnmD0S?CjnqC=Py=9h*AsW%Mcuh@s5i9
z2{0tUjLUb-wccreP`+g-0dm{en|{AjpHRNXdnS(#A-^aq8N`wd$bV2~vxgQk`|CBx
zU7L*)M!RaW`A#DiZK9=70|mpVn<%0cS_dw1u{}!>CMs9|B%OyA(z!{b_W@FIi7@KZ
zdOJ;T&+^{Ul@W_I5$o6T-t5ErD5(XI<oondG%a<Qw2z2?COzU6-|>ci^0MYp)Nrky
z_yn>lSejMlPD^MY5q$VSWGOxgcQ3G7qoAlZIa(+a1{c?j;3il0kE=w^Co$ZsQ*=I<
zbc&NX>ErXY3E6wmkK$mNz6!H`L(9&DCSMW&YE6UnaB|nAHYNxkG?z-tgp}3o4hTXD
z1wuG72#pedj|f7e#9KjV?EMiz=q`O)5Ml^liZ|H*hahy9B`pYj7_v18-KS0sLd?-V
zC<xuBQw>54qW?h<nl(5g2#r#QAat7mPC*9@@y<r}Tm~h@9lyt97tZO4%uzGVcW)C&
z6ejY}Yj=KY2F?t=P+)YPX6vGp&5)rlFU(H5@f?7E3==X6U*cW`JkI$zQ&H#NsuEdO
zi7b5<_st_^qB^?EIdR}N$#R$YSPR(K{g4vxVW)<9OCn>GN!0p#qt(&Qt(-?IYLtQ5
z`+AMJ3Ap5PXVl*cc91Q~!<FG-nMRLhD)hYY<mD_=QAz}wefyaT`@AD7K{8ePC3HW9
zbq{oZewF3TuSfdFW4Zxhl;cXn_js+ZP&B!Nw)}}_))lL0#)V0TjTO>b5hy~ejL%m+
zRr$z!Ssfq8M}rm))gAae6NWXpqOvgA6l2_5b_W(mww#c-lu&NMGA63pwGdpmKuVY|
zT{Q76HE#XaW?a)@qrOq59eKV8wB%N6@SJLY5NR}5qvSbNHd0)pSb(5X%2h^H=)H>X
zgf6e^LNT@OMQZXE^oHw_Jj3MZ6%pQAV)f=^9HQPWM3g0MmFHTlhJh00&1;FL>vlyh
zMlXX2+z1HULD>jeqbIdOi9#4(h_0_JERKS2?anH2?QYTo^hGW87h00)-DGP0RhPVf
z)M{I+?iRTBGuWDuw{fDwq+n<(W9oHlm71tQhIDdMTEv45Yu1}2y=mrSi!OZ<7O$<v
zpVKmim+a7WK^~k(RY%X0&=hAc*tlN$qJf??P(^-v(-GZ@&s!NYk%azP7DTq3N^2#&
z^BfWF_Pt0uVI}gXV#?83Xq!H-%Udyj*Jg9;vw`?_+E9BGZ`wrcT77Fk2hR$2f>QO^
zBBS{JG{#}!s`{j*@C{?QYP#pJ6um^l=v7Ih=1*N_<7$XI*OKU3?sY4Q(b|GwxpI}S
zRz{wr7iM0$YFW#_UWp&=nN8e6`lTWsCDhDU78mXlvj=&g4xSdri3e&=8r|Z5fp=6*
z5;xp0oEZ8v)*e_mVG<9Zgy12iCQP!D>xm4l9oiY*`x!b|3MNeAje-du-!E~u&3xCz
z$e$?W7;+a6_v%$>@(M2VF2d7!ymCmROej2=$7_i~8O^085_ySh7)g3Rtf*)t<@d%P
zIX9TZt*on5_K!%XD5Ol|Zp<8iMeNJSsrNH8hgM>6noHdD(46{lZ4WGcIuJ_wd-bV4
z%--5z`EzDz-}dt7E=PPKo1<kWrW_2{Cu?fT6ITZ=b(n+eN!tFBtF0H>n`z12<*&PZ
zut`Sp2ZTPEt&EHUOUHBTo6Q`3NP%k49rmZCsU)Ovx%wPde&m{@emAdwaJwQQbGzN7
z%TaPo=rYSb#6$NfH&Jc0lOmCFYw1`z;-++@nH-f14J_AH1KxVTMqo`xw)vCoDMq$S
zzdOnH^g<@rBFbjMPzyeqh|>rG`9R0<@K#6t86J*C_~5v5ck%G`njUDR#Lt{G7y-WS
z8eSx_3fBR+5{WW2!Z=cYI{U%_Og5PA0Ea6FF>F~NYd_V$fCTCwuIB2ooNmfYHT+!+
z)r-OCTg9#dor<}Iu^Z6cf@6I8F*tA!k{nIo5J)zoQOu*P=0=!86O|rkew;A!ZPyQ#
zSAs&uC-~tw((`_(bGyp>E=T&8B~EM21|*YNL+`Y8M1g*9Z3EPQL^>3I@<#0kEWz%P
zKUHgZd244_g>!Yx)Ncu7Ujo%wsj5BHP2E+*h?YZcb=k|EUS#<_GqPP<RiI20JRc=3
zy4lW6TPhmNo+BJNQUFAlY=6L!Z{d`-A#x;H+k?cq2YG$a;WvmWQqr08=a`)=8-<65
z082{{I5yH_$h~%dJN0ISMduKJ>QhmpnP6P_4Ah2^9z*TA>_*2+yB?;D+Rr1ky-HCA
zN8mBV<vPPu0F+9%DA?^Sl)2th-W=146~;0I`fR3;0QJ`_$#;7mm)5$m(^Bqsdj@NA
z9i!j%Fi{~*lT{@JDAAgw0Gm;r=T_a8{qct{AZVeqf8y|e1qn_k06zEd#dAqSOB!y}
zX*Q>U0TBW)+q;7pF>YbZzxw6tXHQ;4_F~a|(#Z>mw&_pXG-2E7m+o}a12r;A<{&WD
z;kteNq8xu1A9$u{=z?$*>X-%_H9D_FaKMqiMIdhOqGuU7du=7UIe~rRFak&vuB(if
z7oMV=s|)pitmOGiy0>iiQ}4R}nw_RTZ^{9Rd>>yI7?M=ol{d14T0(ILA9;u#0>Mod
zf9{b7jh;M2b1ksU9Cpm%5GL$%2E#`Dkn>k}L11ETsV9ZvfA64=2N~jj`?R7bJ7SU2
z3Xl7BFj@hZ-O{$~)+4QTd&wPF6{954(&90U``Kq^X^a&lY}BE-mG2YJCCzO0NZD-#
z*6G+1(Oj2Yr)j$-=#g+4PKtWswN-&rK}7mvgFxAigQkDig56Aie4yI6?35Rq{YYOH
zyGt0_O|NAjsDTaD|Buwnd^U@k^^@qre*vn!B@>rXdOj5bHaC~Cz62DP+dVWVmuG}N
zmw);5{OiZRUN|9~KMR)9Gvmz3SYU`t#!~3GfBR;612h9?$2Z^&FT_I&`a}qT`@?&>
z^fl0DU(;Q!(iNXx@iMep(><5D0UlS%0I%FBQ1hMmJV3Mw8Ujh5-zzTK6pAYCifQ9d
zMswyt6$$yX(yz^~Ch#=8AMbF&&gjECXMY%Qp*J$Gd|kx@H8%1myl?Xk@jp}ca|x61
zR^*H$J|rA%fM4DWd<uBk`A*)2J|)QwT4+HQQHoc3BV%n^Y$IPBN$}F{ROycUWZF%~
zKY9|^cX%S_WzhY!7G5iVi{7{_jf8!K<s!X{QdW+6SM@bO6VYbUE>So2rNTiB(to{{
zGuFQn20%35&PA&$hcaA#Nq2-uPs3;AosZPNHFCJLIEj`wE<?*Y811R(PI<bCRiTvj
zD_ZoPG>Kme9N9t5x{+--;x^i_zIN~{Url2NmhhQZTpM|lK1E5J`c}lj9OQ5GTTE)_
z<cTxD%PFl+f1xLv@BaLOKHP~#AAg>8M_TlCvi0GaFB!+;@2*uF12)ljOM1tq_ms%*
zpT!6+ixmk|K&C}zD;Ev2bJA<UNIK8Bm<Zw5b*IQZ)6(UYYBY1@KjZ$=YSXxZta>&-
z-8b(7ry54Ud`)=LI;PRB50uFF>{Jt^Ae>+TRVWLJm%r;JNkkzwwx5t%7Jr2Dz>Kyd
zQpx`{nsg=q>#{#;iH|-${a3ssJt4mpqoOFRg|eR13mR>VwqC>uzPz=rZlkhjZ<)3w
zf=0+!5@oj6<V#0?Um|+v@2l_1(aMpkr1d&_oM&0NlpGbsj8xS4N|#)ZiQ(C{F+?Rn
zB9B<Bl9DMQCxo)&oY^x4^M8^WWpye4m5`EH*;zxIP(?*ZYivz804h^-FOZc~Va2~7
z5XMdQFWLG?*(In2b)mZly-r$vC~Y984R6~X#R#S_g&1&uu187U#2)u3)-VebQjtVR
zk(t=m`id+@ioEhQvt1s@Z$ysGs(KT+3UvvDSH_e5<j6f{n^i>Qg?|>+#5g8I^(?0G
zb>V&4=DkvCQpQM&oHAxx)?AZ<)0)OC=uRU#!MDUm+_4##Y6|CX3zE#M29HN{6eB2j
zPChWzg~|uMse*Se-KJ(D$)>E6xSeI-+Fcf~X(2Y%Pis7_Z(fo`<^HB&X>OR%Zjv@9
zSCn@_8Vrvq6JORAoqynDr;=;sD`kk+Wn;{mnWJ#z*Co~G$qtfH-Uc%e3H+Wf61R-@
zl<@C`lrE4q#N#(f3c#=1?g2ADuf@oki4`T!M9~RZW5AZ!Kzimz$eTjTct1@r_e*7m
zZ$>(XPX^-Ff*AmlT(!w^B;7r)5&_Pvm^QkbkH@Np(7F<K+<%pr@XI^$Wus`)(lFy|
zl%%RA;xNWzv&>dZl7)u?0VuP9V@;p3GCQFf<>^UGjMJLI;cw<Vu%7dTr@rQDiBFy{
zTBZO7UW`<ZGL@^j)u#St9^wwnIN#NAQwxDYQ?mP9fKz#&0?A^Y^c2O$9M{8)=fH(%
zQl@GCQSLO>$bY#EFB%`(tuYZv3HV|zaxf{kOk$NVAexa1ii!5J$-t*EO<7=0!Xc)$
z(VHT+k~Q(n3eYwo4!Q?0hADIlBS%V(^fY?GNoLi)lIW(0fVLyJ$BUYD-_MR>an>WU
z#h$5H%~f06Bc;k(v=)UVd?$u^6SfK;KZgHqA20u#+<&C<SOFM$`t8q`@BoBn#`=>L
zh=w@*_510Mmw%5?^EiNUe1+-o2fszZba#Hzj3c@*Zj9DF$%bi+0-%u}YO0(MSzDC&
z4B4QDV^uRit`a9!LBKQriRgECel~LpYcZIK5#rjE!-+V(87X_K^#{Lv9tZE3b85oj
zdlmkGhkxXjDKD4!5E8>9?r%LAI+t=#H*M47<RDPy(HxJHqcGNipA_af6*RS2mfH~e
zM&zNvkd8~8$m{MzFSM-f!LZz{?IFCxa4IYBg%2~lLz)hN=Bg$ph9L{jTE1_js}>6A
zBP{AM80KtoN|8%l&v3HLBb=(XrI^xY+(q@}%YPxe_ncvLn;_l#1koVz9_>hJ=7oNW
zcOVLSh-=;<`;vb3POA19e2U#4!Y3x`FkydePVsr%G+@Gd6gBM_s`w<T&9TVQnMfL+
zta1Hui&r5E@d^0{Ek4Wuw40<AE&hF3&m~|HZ=R{=J;Xgd?^l0?o*%-e==p%p*XsE(
z)PJ&`52Dh~(Djm@eyOf|hTBiPeN@*m<AD+AlB(xGB^2|%j_LX1AuuPYTrsX{2h8<l
zN(p;QdGcPbLzyJ~2RjwHI(WN{)j{o8w2<3jbE)Lp;|g1yiX+_v6_3V@L*TBmHQhKV
zqbJtK!>KRc5g%8#l{DRHPi3e^53k*@YJX!AL3lJiLATRSH_jdV;a8cR7W|G({p0x#
z!hhl9hve-UpT{Wg#(9rYI*ipPRfy+j^fX^;Cf540*)A3uy|)<wN!Zpz9ZDuSzEx+T
zGb3y}*A%t4U5}7OHeLJfx8^I;&vsk}i({<KnIR^$7X8^wHTA;z_xugnN8f6v9Dnq7
zcVA&Rhlx8YZE7V2L-*3A=RMbEgQX$X=&cfuJGFI7-3_-fdEOZh-rEhqXv#pEu^R%O
zA!6Izkl5*1sY@t`NbfOB`F0*T+gl!Zs6`t|;UPAbeXpc)Z(@2&LNol1TqXsm0e|tn
z-g&TsNj_2hK<+6WhIT6^999x4=YN<)IS#8Oof4fqaBdweo^Tm?doBm;`nJ@VjRYi&
z99-3>@@)uJ{Qzg9K_awaO8Mah)g3;i|5x37)AZ~V@LWv!GS=pkZcU5`m4Feosd+ai
zdO!=TcqEA}?CW<~R*GW{32Fom5GPZ7JU^AjFk7c=pUh9U^Ejy_6MU&fKYt#~a-Bbi
z?Grxtd1!&21w-mrd1!%3!tFofq1WrP`#pE|nmt`xo;%C~P#=aj9`M|`Z9C$y?70<|
za8}E4^%qs?3QZm<K8joDsOX8Cdr@WU=xuOGWaygb*;aF>^MT2m3XEwIN-4@sPWM8U
zkgmb#A9pVz#8bcPUG|CDGk?I)G*a~kF9(Rb$D;`dBSG7Zoaneyij>=>2ms<}M<}bw
z!>8RrnKloWzk1X{Yh>PpP(8kBKXLFGLzWU|BJQcaZs$y5n3POH3Mbb<s`MKhE1L#J
z|4X4<L#JQEP`|*gj!^SK(>n_WFPQ-`Mk6e!9LA`>V|prNoPdA9^nY}K{DkT00C8t}
zI#TwW>FJ2W{`7R@!;u5T#cuB!jfP~->rdf|HUnnD^~(<z52RkZvx=AI$9%5zDV~`7
z=75L^hO_pI$CJY@4~!>ANVjLX*AzDnXOmBDkfV8dabJHN_i>@O&I6yFiL&4_p|SYn
z$d9ftL`&%Mc-fS=W(pG8_=(>B|GKpdmmxP869h3bH#e8zYbO(zACEr4e@e$c-+w#k
z;_&BFfF*C8C&fN>-X?3rS}a?mm)5ey_HXmbS19+`1Iyk~Y@a9Bt^u@O=X?A-`C|%*
zo{NZUAIT!VeAon4BHi;jkhydlZ&d&g0X&~lTEM5o$D>8Q+R(b4T+FVgFig9~DbLAL
z(WQK)%UAK&+xy|4L!UK+f6BVpICRQ;-~9FSkZurI<$P?6Q4y&5^>_3A@VnGn3L>w$
zqjHV3bJkWhGEP|=8b_$KrLIPoC65dFlJ1oNiB<MLe*I{~%D1mWjNU7k5?fn^FX>bD
zRxc5p0wV=dqvlp(LHjD(VsP`k#d2+Qxqit#klH7S>C^F()8?s6e}%s~E)!5ie@R?g
zaf!9kNr$nm)@tQjnyaX%YG$HQG%8sd`jmqdXX$G(pd&@7qoO#{WZ$k1908TLqx!&U
zwxGz3Y}_bSX$#BERwgM|lZMHft1#cz`gLm3`W(z`sol6~r$UImAKgS3f9RC6TIRmn
zy4sb{?EtP?Nn;cHf4@c&b0cNlyarL<xsBI&P*I0bOhZ%8?s(k5!Hy)eudmBCqg653
zvE-sXjz#%2*R@o?XL4FGa4FHc;PQxQ({|$Ioo@?ZRO)O)P|>=+*In7EJFs!1>5*{8
z1{r+u!Elyuhj#1A4#)QIt<V;tiqVK*9t3R`K_UC|7-+M|e_%s>Z(X0r?JyU)KRrng
zmS8Rho&%H;x52q2lY47F&PC=Xcu8P;a1OOqhDOQQ0(1pzWGid{y3|x)N0Bc89a_b(
zHY({y0iD&r?7{Yabgkn4w_rz#ce7Dkmw=s921k+?fn89Tpg>BlfE`nvh*Ii~U`Gt+
zmTB2`up@`cf0*#-v)fBC<946j>Hr=DaAcSax$q!>BjnwUhO^s6$=iK)6IR#HZbIO{
zdUjin>jq~xVPwbIO=#M2c9VgxpWS3dKMK;2>4+ZoD0|Frtn!B9yWQ^C;tF)*Pqeth
zgXQ!ZFI49UOaZIY;{F3ctAHRBP}w7?G&8xg{j|IHf7~PQbBCknn3|*c2&6?M#;>WI
ztN<+%3y>7daoeNG05GtPo?B#e=&H4j$rwp_veL*3Bu6eYFV#YKL}c_q0r?$>jFzwF
z;m1!L9_dw@dcO0$$<tJ;vhK#kvT(L^Fe26mfj4K6ulOMFP6vuGY~Q=mT#+Lv?Ui62
z-@4RXe-Dc$DCbD{d23YUvk`a{MpFBt5k}TMpUb!*E@S{{L{Cnep{OD{S)Q&0{%mZ0
zA|Je_69IvzO{A*Og4HCSzc-5^3W5H6oY&1HKzIPt)**0JWpd0s@?7l=dF6w@&>L$Q
zU8L)`er2+UpP0y_;QIp~;<1IgYS0&s+pePkf71}Y4EdE#I@dA{z|&hzFZ0@wsA>Aq
zy;Y+EIXz370d12uc#s)dLv7}rmd8a%-4V8pH%u<BIY?g(JFHG=uo8BhR-S!Ygq;gu
zhTEUdVE(Q+ljH9V6NkmMF&GuNft+tgBCt*&Myc_}Py~*67xB`{jj@QqF7@EZ$}|{t
ze@AZd^a>hHR)f*pX!-5TRaM}vIiOUubo%egR!i)_DsY?zo?Gby%p|X*x_<hGXk|1i
zK!2e>hIu^3k@vTM9Wq&e@ZA5yKaRxB^{*A(dI-CosjX~yP10At*0kX@Nk4=l^3&RT
zFB!y+#xLIvNf+iCt+b`6`Py5WxO}W~e`06p07g5-EVi?<i!Xt%!0H3|QqPL`cWU#I
zE%Um08oiu|B|ElaR!~___6nt3U{{N|-~R%~N1k0P^ILX-S+*63t74?o9oZHzqlakk
z$+lcjR*zoLdr%%>H`xLcdAE&^CtJV=GiPqdwQ7ZTqFS|)J25R`WSMBy4{zXEf9YkN
z8#HX^SwhhTJWEL2fM-b`mU)(V`VKtnGJiUrCH;8_&r1Ivp>aRWs`cE7Wz|M@kgQtK
z4vr<lUZq$fg8d9jIIv8xM6q}CtJ$s!R{66p4c*qRs`cEdtAw-LyQ)^SqpL*Nt6e1`
z*xyycf#t5MA71M!bi>z#Qyo;85{W0}hyOfTB<b?XH0obAP0>}CQF=ZV0Xdf;f(R6s
z5SBi~f5zkQpZ|D~S>P(2(gJNbSZ4&uP^8oP2#(%JoXmyZK#z-(rB>o;NOt3ktZ3u6
z(kHc$Y$lX%24`9(jy~o7w$j9-igQe@Brp2Rnc7uuH&p@?CJtD&Wla5K^KUT86{StQ
z4WG*>XX+dK=rp^iv7fFUmh7sH7Qs2LITiA#e`Rmb#tSD6A4MWn*+6ddwIERW!D4os
zueCxU4t!zq3PI}D{vd@$>n*))km~Rk`O5Myzy7=`U-c-o0wfkrSghi+HEh*z@kuFy
zjUVHSfBX1+`Oo0A=(W&ff`9(`5`Qp&!aCwCg(I2%`q$x)mtQ#5cEP3zNNY(i-2|*a
ze<THKEsPI!0-7%e`mC~BWj9N|h3u?q&Fznn^&!t`{*^I#6gskzQUd2lp+7QVm2G9H
zd;ODa)V2KDtZuS_)`3%m_xLc~lRl-8k%q#FYS)r19^|6JtP$Qg&P82fgkoIwWnVRm
z)6oY9E^;}pIVUUt?_J_SG)DDC`Yb7We;T=L4T{?sU3{Wo@O~>8S|*wkS}?+3wthqn
z`Iq}euz?n%8f1VK<z-d(z14aUzgHe5wEzl3x5=X@h3YU!A5&gjefT~F?KSzL&eG~F
zE3^dL_jxtOw(GL}C9MR|EY5z#{ym{*FT&z8Ox!V{6>hr=t*3+^_fNbmLN@d3f4b*;
zN*ZdI!~>2WyzmRkUvsqVmVNJt(NZYqXd-{t75zwQSrp011bq+|O`#~uOqO!fDv~T|
z54*(l|BTgAvNR39tZ8#?CB6mG%{(iPQwuZRL{=`^$PfD6K~~-f?U9fAMr4H=$*0^M
z$cpbpkdTtn-TF#a$iC7BOx41jf5IWm{vX=_c<@a@b#|Po%PN98&E`yT>b$9N+T~D6
z_iPfUYjsb8;HclSLV%&O>_*i*cS24>AuTw~At#}X0A?F<^7jo8)R1o}vkyTB1x9T7
z7;+5cm(Fkro}ILEbY@(hj2u<mc^f-$Sw@aZNdZKF>wX(KzFG-iLx;*Lf6`aDLDO}F
z1EJ*L9nXP7DNe(qDaz0`S9;t_Q3e7_QBzC&Hlvm#lGSCKJVuK$s!bD)rHAr9)!rtj
znJcv@xtoj<Kzua@Z!R%(BTVM+D!05pG+FKA9<tR>rwv<a6wR4dpQcF&adycu=U*|?
z?xYy><@p%AFkq7{tqHN|e|dMb;B$@Ywe}p7=x~GT({7ojqElV#R`s({jR`W>T@z%~
z(@UL|J<*t?E_kT+)eI#S1X8eB>NUx_p%s~B%(*ujGjp8`(#TL;d5zX4s+*~}hMQY9
zxH1A<MY0)bZQ&}=S_I?x$Q&zILy7j3S`8nS6ZasMv=j^Ht)Xuaf6BSyR%rEE(Pgm$
z_(i;YZ!pTJ*H+Y9F|%sCA$f=@9@(Mm4!hopm9wYcQ2}j7^!WUi(6m%9T`wpQV)bN0
z?LmYJ+0JsMCT)<Av}3)c4oBzS_Vx>-or`Y$(rTS-b(!susM-b(cl(Ma5UzQ<a!LT;
zApJ+g2+0E^+lH_Ce}C4QMG=C!RcBrAbjomBE&T7-8|jO%37>&H7rhlga;x5uASlAE
zdh2>8jI_T3{s(-`(LMpW^mr{g!|Qg8&CydU6wY)-qd5lFS^A6>vBtqJg61rD8RF{@
zi%W*AUWhxDh%)-)sPS#CwahUyQRy`gT)HgPG@m&q4Cg<re_HYba>i;_({aB(J7pFH
zizpfvVM*6o6r<HwQkj;ma(l|YQVj)b76_7D{#r(Maus!_ku?e0#(COsUO86Dikk(N
zUfUAJr4M@X0M*zC0rjcKUVzs)BiH~-6LeIo&<x%Njayb0>z4Xhbk2rXfr1#$hi9j5
zG1|g&YK6jQf3N6wE;?(<IkqhXZuv)9u!VLkFeHF|=?Z;0;3YnCkf&d!OJ$2|D{)i8
zla2i^`@y&NzqG-m(T6XGt@*Q^ZL+c$W3UZsK8<llu+uJLFwAVK(>!q&|L)8iT*w~n
zH9LT5)2`5a6>AOjc#V3hK@6=M)kcL*0QF|me%>+Ye|&JXlhHMTa0<C3Yi7bA!d!<{
zVagd5;Re|(-%igywArEwFm@xrq%FSm-1-4hMP>?4sr59EpU1=+4OA+Tzbf^mQ&ZXP
zdT4bt@~s`0D4m$+*cl-=z1}&$ZdaUi6Q#D<<4j}enM$k|$Dlg(kAVO%-!#c@d8zhU
zv#LQufBQSd?Im$%lo!eyKI-8L_WSV_?WB)ouZjgNJ?~vz!_Tzqza%ScP~%jYACTa1
zE`n?p#*cnCul8>5wpPy@JhsJOZLsdDEwU!g?Yk^W7^}E$b}Sfgz@X!Ubh`R96){WM
z^9x5O^L3X%ZsG^CowGq)R)EcgRy93k0VVe@fBA~GojxzO5)exEDr3swV4_o&LRk)w
zKbPxV@j0>L_VW*kl$HV{Ja9;)WDw5zAACqu0aaN;rzwY|3HZ!X_dq5^-e?MYtk`3F
z!@KRERygo%{y*q^3BXaV8_t&~^Enxwe!c|G3m5vR_GR6-ia&DP-j$t}uRVc{*Cg=q
ze~UJZ&<z)D*oYr<u_QA0&a#VxUzu-D=$kH<FfI>VEMb%$xLCqxzI(BRPvY^5C0vB<
zizVFVI~Pk-*p4!B`?cm>21Nk(__ZZQ=)1g&iTi+e^^_j*E;jRh?_!g9f_E_ycD$?a
zec!buqW-b$N!o~j6$gCzDykBafIFl1fAp)UgqVWnT)yclD(A9t71cw+Q@nSnd^AYx
zD;~o1x#J;y@AXR!i<j6UNC_c4<r16kGVbyPfrV4C&2vWD8)864y7AXlnH+hI83xu`
zagIaRX1BapYTVe_jRm<dS8dRDzf2|4r`2&h1qI}jn`2Yn+1w5B+L&^Lw*g-7Hdb|l
z6X3X9SS}2>?%;O0o@p}V@~m{0$``4+ST@kUzZCGD|Mn0>?0Lt3J+L=z3;qjs_LpXt
zF~&U<12{1^m+{d+6qlr*KEeUbmuH|pQh$?SOYTQc!h&xMdOZg>fej1ck6OKC`H)nO
zrkwIT5SJi2Pk0mj7rwqUzdpY|{rgnCh)QVV585l9ln#IYcnTK?v@n)+F~aJ@&wmcz
zp8jZs=Jjd6V}Qjalg4O0Su!+2Yj-fD1du69cAYPIo!_FZRXR)6tgTTfz}0{toqs(-
zd|UnV^ZTJ0`TAufiI+#Ug5Xp5Ok%S>pp8(<w@-vfkFm&4>Qft6Lr{%lG%lvh-=D(u
z7=Ph0dA`Du>=Oim=ysWcLON=ns7ZXfTn9tKk?nMV`T30(AhDYtQ+&{;94q74_h}6M
zdT!{?F^7UXL-fGs`S1ZLkTar$Pk$6}xzWXqQ)x4fmx3=#A%hImN$O<eYrsG+XA{au
z!I|665^zpoGt0K_i)m|>P}+2}$oU0ykZ`%VTjyMeID5H2y%WxLmM#PlN(LB)(B^Jk
zg2LvYaOrnWtQMxU_d-FfgYtBIHWGuJ*Hl|-4RPTLB#54Y-&)BQ>TbbcqkjZiRYP7K
zBkUK%D3wdsf)IaUE-DLEQag?i-O3(C^`cgmV~hGHUDoO?VHOpPgHhcbCtcMxRN4gL
zg*TKM9sHL^N;N*`QZQTz08qS(+Bpm{!XKx!O0m<8eqCk{=(O;<vu=R7ROWY-^R3lY
zqy<>p&J<(lRD3XA(Z(Yt`+rrAN8^Y-f{D~5RR~vZ8i&e=5%3Mlv1>S$orxgBFdH2=
zdu7FWHDK8^#gI8puaZ`}P^!jBoL4JkPJ<Uz^HpU|7EGzUPO%Rt_@MC7T#~kUqL#p!
zfDh)0x~e^Jnq<;C7c_x_H#$Fv3KG|A!j1|e3J-L1F34CctgIlr?0>Jy_b%fesX#Wf
z+`J4Y&eF^h8l^z{qAhN&sp%y-BcKx$iHlAG19RbemFvyKJsb52J4qj?RisE5L)BD$
zjD1|uB|*S3AsBas2+*J#0=|scE|t_NS4Mchh~b<a;5uFQrWn}R@w8889a(#XmjTc+
z$_A)jH0C@M>M#leP=A+a1`t~_Ofh6h*J_5eYL5aXm}NH)rQQnE8y3AHRAqgVwl^!P
zd|7DO>7ab^LxxJqT>d;j={(f0K<uvCOPZmqTjEMr?9D6>3G1xsmduQaP)?0ZgvXR_
zp9hll_972x>qVzlt_)pHGsIiz$M6*c{4`;4^_MI4#|&YW*?&pL^k}a0EvfRl%m-<g
z11GEq#+r=rvDU{rXT!K|mIlOBN<rt(hFk6p4mHKI&6vkfxUil5jx>-YH~4S?3v8*G
za>QJ~JbDYSi1LG|8q0V+b>ps)7>K7vJ&3z0vO;l5d<=IBc+b}+P32V%y^h;?1~u%o
zxG}dYTM5l9SATFj<irFXY9##+$sNaO^y~-8UCh}`qhbuql0Dc-_J9OOR;%0Iq(m>g
zr0(fOvbW6ipfbJ0@G7giis^B`>vgNv7SrSOQ=Ki3WqO8Hu(peon=!qCaPMS##>Tz6
z)+u+OdQJ&TrdzX{QoU=)uJ#xjfJUzcuBT+E4C*di?|*fTbGiF)E>s{ivzOMc$8au=
zJmGxj)jM)7bXM4lw%%IOg`6?^S8s>W>MNUdNOxLbYsZ6V7OJpf>OnLMmEnx`OIu{L
z@!IPeYL(yNG&A3N+sv~UEw!^;L#;Tb756-?rxvMI>$8W_D_)n$PV}$bJ8N{A`elPs
zfv^Tz34b4>RJYhX3s@{|tJmi4nG@i<ryjh0<^-~OZG8MXQl-5B<2t9@WB-g1RH|;#
ze*OL#ckv3lxL9CS;G8g{vq%lALhDDY3Z0x`RcUn0s`AJRtICq@z^b@yvdk%4yRKzb
zYV<Z+9!rYTc^}E*a!5>)8f7xrW#{VEWU0<vIDd$f-<LiU<d^x1UPcR%mT%{l<n~OV
z95Ig@FWn?GRs&@={4AG0HwE)T%?<3?hBYlXD}%djP8KPPTN|%Y<~_pIdLfyBHGa4@
zT>JAx$BOy`Kiv#l=BbYQ#km8BN+lSwZ)N^8TZeNQ${cHyP&p2<4@{A`Lp2+&=%U(?
zuzzUB^f%^wjw;Umw2^<>*xnyEtjjG&X;G)gGj7+Wl0fe88k@RfC!LMKzFj^>rtZX~
zx|!kHP6rf6ukSh?Fv2Ud-s$Y<Y=Dpn#s@l^(vx&u16)S`&xPa%@kM0md&~LMaOA0s
zDL*}jSPllIMgktwNPsi~)Q201h}C9D@PC1J<BojEcTui-FoW%j*+5QM&Yi{%I;}Sz
z@hx*_eK|-qYL4pKl_P7-esiyF8JS((%REx0R?n2Es4daCG!vL>^GJiyzWI19?aOQ<
zqANZuGi(r?v8~gt$hZ=^-KohGl%C>ox7mg==Ev%xLkcg?Oe<tq!1`90%Bj3awSV-S
z!{cZI@$k!x?LF1&u^Q&Be3@msvf0(i-Uw)JV}Phte<PlmwKxmSLp7X}mTu_x+CzSc
zu&TL!BicOt(zE@BnFi2JI$5UZ(zR|_pu+M0eS3l1Zvn!o&MtgL#{#lDC+(HqY_H6H
z_-aWRhXQetuWj}J^E_=ISpX2gIy6odxC1psbb=7>V(CX1sZw4TYV|7RAydeWmU88U
zi@W;~Tobk@=z`Y(?KiW&{{Yll^4XVBdOj5aFqhG`G$fasdOHl4I-No&m!M}U1(zTO
zJ_UccKwr`a^Kw2&4vWA*KI$}pf*NsJv^4}nlDCdhSu03u!^yAj8D5FI&aPZpySCH-
z(av7aC5OYAIcLZv&XtBZ5I%VzktWH&IX6iMskKQqWY#9tAt!@WVjOZU?O7U%`Y46s
zU|E`!hgxLM_3%APYOi1j%G29;AT>7;f0KW-_anwDX?hB^YNd_E12eU!1migQU<t1=
z*GWMK7O6fPC5XmtggUTto0uV{OfCvu%cY4_1}-+4YXjG-De54kZCs9nur*z#K~ij{
zGY?WE`D`qMSg!4<4jN+{409?;8;tXyu%wA)P>szpp+VE6APjL*fcrJaHke?roF;!=
z3<eMHX-RlkWHP}ZPKLo5TB(_v>X0HgQ6N%kV_|2K*d)@B5;q%_hm=#(RT)yLZ7|g#
zRc2qBrXk0q?P(q|S(_Ms$->PrMNDYhilPXLT-wuwsIb^xON0QK?8QXbTN<ju(n8eM
zi!j1=wHU!X@OyIzLJyv3W5KboXc~W(2aGojD?`aT>Bgay(w2tJ#Tus9cz^}XFvOx!
zX{*dSn8X$%h(Jp`jgTPbW)9S_mF+1!YZjY`tudPd-tCW*pRm>{Q)m?#YPwQT#C(}v
zt3$0-_po666XK#dU5d+-queo(sEuLK<!Lwk{m8MI<hvGS+At&|MV-becq@OVIK7L}
zVwrS2mSffEi!{%*Sf=ODB%YjV7@t&|(>%Q>S)0ojMrB&R7Xz{>B~cC_UyTEsX-X4o
zf1~&~#_kl-y7%Rm<4?by&&IFL&gP5p*_)$<f7?Gf`_C|bJ->K4yLcW^Z{orDm+|}Y
zo98rr`Eho<cn+fi%u-4rsi%Kq2_8bo;2^0Sj9(2H{n_~4{HOW&?eGPmTvfycd>Au=
z5lKaVe2o8gVFb!v$S6kLh**PCUN?fh?jXw9P^(Hs#gN?@VG=UZK?EK_%u1DjDzI)u
zOj$Xips)&3JyWe4jMzLc_IRM{d6`4PPK!|RElnIMDvb!7Ky<M8-ZFp2ulnf6i<9{o
z4Eo!T-~0bx{JdD4U;b$vU!5#|esdI#=dZ_K-n>44^m6ug{NZr%^Iy*w-=8gJ7ymvy
zo`qjd$D`Bv(fAf~#z5oc#qs#{#mUQ8v&-@ChZiS@N2jy#a<+K!`tTgef7maN_QjpN
zH!bkn=Z-)bYx;5g$3K7ng*q~Lu|9iqdU~)~16dIaiJy@ow@?$V!H}m*)nWW*emcK+
zb`JXo@fU{i>EQw^IvaTU`sa6x@189V7qfw;2T$f_3rzAvJdH82&+?>IaFyvx#a1}N
z;LGF3<I{`z@w3_D`S|qPC*x1EcZ-9S{d{`(Y6ctN!?VR@!>oTc3wH5x{^sI%c6kMV
z{_Ka@%ag;e=kMCKXH-aJ_W?$@z`G9KWm<6yk;^OOZYJzKpTOJApRoH{N?pobDqU(T
zIM^-$t^zS>3Dt{Q$;bd;Qz{{|tz`fZUnr}wWuuCC&Ae3tB;g<q)mIsSqzV@0J~QB`
zJ!lC)Ssv{JA(nqRu(hM5Sa94V%OL`>#0*7O)=Lez3JGFCQSpQ$2mozYYO;hzS*~kM
zt|4KEKfk9Z2fvt~FHT;c{5HF|t~Hc5`T6!MWU3)Li@sVR=qsRw9?A+4M=hF2wp%9l
zu~GD^Rbm@KuOH4LOY}urd>=QwUs;y>jB=khfyeu56DofsY-`n9X<}gXAGgp1fTQf(
zN|gx-YV^rL$%!p--5#MNt|JuN-4IGv5sEtUkaekEOB}mIUE(fvbmNY0+|i9Yx^dYa
z(-^!qg2m+satBOvYN;?KlLip)wNeY67y)YNPGB*@qS|(t1`I{FwZt^zkdr=~6czFm
zmAnI3!;XIiKe<3xFnwx83ZfN<&!cUi@ySRn)+A`qfq}9ugVsVJmL;rdDR^EVinEm;
z?Q7Yt;7igFbrrdz_iYbliK1u|>Mmjlcpd$)z$loXh%w5!H^P-d(M_?0dYN+tN7AfX
zkmG6lmQVo=j!xHBfDH$)%b(W@%7neww&zh9C=`E#`9Vf*R;^hfTAeVas0o}LnV9nS
zMlcB`ElVSiTkK834MtEdsws+^MTR|8qwT;7i<p(W72YK#Kb5+{3d1I#ORp&9lu}N0
z2Q~=WBcY`yRL!f*&7SyBP&B4-h4eS~(W@FC=%Y8OA#j49Cf!0+8-29fsEVQ@aJrQ$
zdl7&Btd0D81nQ$XQ_Aa{De2ufQ>rVRNjg8$&X2V7BklZ1J3rFSkF@h6?fghP2h+~M
zw9k?H9EmpYsE?sAp6PXk!brFmP(3~T?bjC@VW$FEchZ>W6cSJdMJq72hLbayiNVwj
zCm;y1VhDOG?GOu4VPyyP{+iAJZA8d#Rsw%Gsb52cy6Gg+kOZb2D~KFQzQyK@ZLJWj
z8FBCDpyn4qN#F}G4oyKfoeX~(E3OV@2@n{yeO3VtVhW^nz1~;NRx{^>wwO`%ZD%HM
zsR2U{j9tkYyJBFsdEF05#2W{o0<zy5h@!i|9N=6+1B!+}qwFQh0VBgGY)ak`n5BRC
zbq4+d{m8%?br&b(n4N%5tT}jX5|A|Wo4u5S3T@y3i4;bF*t39$WY)4nD}+&pQi}<+
zRDSE|E#5jE62G(~vMLT85oxtmmx09uT6T0)EueL|hkAJRIW=<W(SAz1Al@i}ez%3i
z`H3V7V2lk+L&^0s6+l6g9CPkgU{`;G@CU5;G_)HV?3e7|`T$Xcm!bsuRbLFH@W;8L
zk)xDub?o(?+BJzwt^c@%DtLDA+w1=ux|Pd1-OA~1bStmYt<(us>VztHQj|OK%3Uun
z^>laFDX4lu96+!$kwDx9Futi*`2^VI$KVK(RM7;mb8&Qz@S~rGf!2MlL=u1RE6_cw
zgjoF)6^Vrkni4WAuR8)=S`u_?xd>tpKx(_dt_s9NAShrv=60Tt3QCJt@4Dg2B_al*
z-vF&ZQorP*>4gD_c^5DaxHdnSnv77?+AR5IBQn}nQWgcmvj6m92O**6XBXWO8I>r5
z>fU4n`>%bn-*HDUVu|Q@w?fZ{a$~W73ZB_7oK}xOTXD@FxZQr>$Nt6EYwA_8{OFo$
zv{F+`ub-vY&(iB>>GiYdb}gp?%ygyYME#QVpjuAxppin`9p@^T4l3O8#2Srf)d#jx
zi&&}YmIv1bOmewU|CLQrdRPms?HpmtD_JgRNBoDE*Ei3_&@SYASn12z(VJI)ZDmOe
z?Ln*DTdWN1AVgc<JrJ6MWOlzR^WcOLzpLG)+=^<yr*`wP>ISuBXnE7ER2^_*x{aoQ
z=4XA_zj>+~Zj|kgH_BC4p{1X2S_c&CfMWe5)RubvB-F}QaFdg^)09q1G6nFVdkt(b
zu1EV>wk$?TH0UajY6J_u--TI!#=@%Z;;bU&K!^f#cp-vvfTagdb2bfT>y1TKQ2K(#
z2W^o_RpPt-j7qWvt~{_Gg+~7+cn##GzQxfUO8w!*yOXzJe(`ENJh~h?Mh;PQ+q+_B
z^G0)tL;`c~dG`w-mf}ths<6p5)xkfI|58yqQDMB(t?Qj_v5G#XLK7Z;sOG=9-mZ<Y
zcJ_GfS_B~1`^9*5*P^NW9TM9Ayj!#lmxh8)eFM5}P-VZ~tMwscmeKkVlmE2%E|_J!
z+78u@MYUs5?YDBZ-^$hNTRDrdOVlOql5|PCl)99=RJvqc>T^(^gZdoQ=b%0ZO-<T=
z0c23pZVF{?WOH<KWnpcXJH|Z<mvB!f6_=b}FDsY%vp$&t9G8={KK*~%xy<M#p<oE6
zm83qUd?bm44Cm*=i{>G?8_@X4-vR&3Ys{jOEw<d5NSlkSqUA!h$>GH{v(^=R)ge@b
zFr!I&sY9qUlqE0V+<we%uMYu}10@AROxAoyg|JBZHD61BOXhs=O?7FK_ELrdC<Y)~
z30Y6F07TNoOT^b$RgQl}$$5ijBfl|hS;Bdg67mIuB6)ntoQu#%DmfBWYZ&0=f@OUO
z_JD@ME94O>t<2Qo6iT2h6+B8rI;>POCR2Cy%klz&RH1CAxO5qaRNsBPs03E3|5A_f
zC#wQ8T^B|#wrM$!iSh|<bcYEhEX-z>=hRXTnUh#A<asersWyKGbFhjwvj&1_c0pk^
z1R#NdldWQ&H%Mz!wF^P))uGi51nk>i%2rE&DPz21r_CS)r<eq6HC+=urDAl^l6Ce=
zRa}myId}~|8p#}mbfZ)iK?FuA=n%6#fmZpmtcfCoa2_$%kIgdBNG3%OVkU&A6l+&N
z^`tZ3D<%FMV^n`7dX^qnW)+YnK`fCNGgj+VP!Ypwb5Sin%I%lE?Bg|!sN54J*X~JU
zUYFgIwAtQ0V97lIEefrB)|L(Es8=Zy6~(>g8fgv1Oy?R;R%qpTP@D?S&^7~NK3><Y
z5!@1=v#ln52t5FWk`zWTTHm)vA%;Uq+?%r&5E-Cs6)b-=D0D)&O1<2<E@yAmFHulS
zO8`oxtqzQyVkN*kduI*AikP`89k$XVGYs13ZbfaoUPUXZ(NFJsF~RgP!Cs4gSl2tX
z^+{bX5&=;BSzWK<f*C98e7_&(ZM)t^wFh;*%`Csr^)`~;Y1fOI;z_LTdZmzUizjfY
z>lJEk*Gqrcx0>A8^BQVs8VPmJ3zMEVHM~Ld8Cw>EzC2SZ{epz0hBqRzw&9h`aoJk6
zqT%JvbZ9O$yrcKFZFo)R&}|ytxE$Qp?d68I#pH1fFVecI^*d~M8^tzhc)K3&QWNp8
z9#0~T5xTYT?j|A?zZ*C+75zEqKOoGI2q+1BjIDoL^s`87pyBQ}X2wWozoy*JjFC%*
z3f|d%&}Bx~ybQWV>aH_qO~LG3gT<4Wv(%ai2Tx+o(f~v1m{;x>t!LKJ%~$<!AG1cs
zoYbxITXSn=8KP=uxrbXDMG~Nu^iH|8l$uG7t9O({^IEB1gd1+1meDobS}<WdU2bC6
zRLg(0XeBJ~C`EQXXN{OmjKhoAk6TLVxwK+*bRlDMfum2bS(dQC(bM>K932Hkw(HQn
z!qJiVsGaf20!OcL+m1qLWDax}qyO8kR7o1}`+N^2>BJPZu<cCSk@S1}6km-cLaUf3
z=@|O7j-)%jZ+~?pJ&vhKdc1Opq&xl~lCFQLmZ;9cXOeWnSP0W<P^v|={K}3?Lh!<l
zOByi^-Y(PyFzK-7m`fgsZ0Qj%+qI<@LbtkX+e(y|?zg;TH;#&Ar`T`UN`ONmE4RF4
z|8OQkdErwtAg3ucei%1rR~<ZYv+s4pCQz<pzLP(wgVV*uuFS4t{d{g?1HJkeS&M%H
zCQPer)p<yoMupdRg@WGSYJiXw7WPT41_;M=^*Kd{jPC^k*S(*+3QHZEsR&I~f+mRZ
zG4t9onX=AQ@qyDk4W&X)=rAPCkTbkmxRSh_KRm^mohz5NdogQsi9KeCRb1*xoj6Wy
z3?pS7<wc&HzFw}WlGjh<_IM$mGuVGxF;#s!Dw4FK-NjHZQqVBXCm-yM4NiwE&RuO^
znFkVKYdj~olw=yKvQ67Ojs6+gQ{G0H_Ycm5a?^+wtEFPrV6Yxseo=Ges%cxd;EU9B
zTGPyQ{cDj43RA7FC1JLWDz5Z3;PJLiW?H|u{_8e6!HC>G!Nkc#VU-Hr`j`=-yndnL
zR(mj?X;t_SpjUX7modgY6aqLmm%$PO6qi4{KE44RxBk06KLrH)#pH6AF~&U<0ysIB
z!A}Admq9lgCzq1JKA3-dI{p3n`-3e4Ta$eLm~Z;qx7UaN9wzaTjPt4U77bb7{qyr7
z|G?mk52*{m1e>~F|90OW{?JrMUh3W(D}orydhc9iEcC{?*m+NeLYRytim@olPjbm;
z``jxt&0>);xXwE?3~`*rVkRdSf2n6P4r30=xQw}hqBSfer$2wcespT&&);UQiH*mo
zxx&|cDPxNPt+AN~x`ApnB2+)WBzwssXvqKt%>|46E>xVB?26*9q-off%T{DC>?b2~
zF;X#9e-{I~H2gLhm<NONOEIY5Y*3VyLJ5@sN859-M8$$q(amRr%9ZtV>kOL-VNe9!
zw?g!afd~dhqhWuA_o7G08fW1xm-OOeiHlJcSDv5ZWuI#XtJQKG?HOd@+HkqaImsmQ
znqu}2uC&z_FF`Vnaq`m0!OyF7OLp-hb0SUkV4H(a1}thWn1(p*HRr(+ka9UvoZr@G
z(KlmWQ=IK)4s(o3ft59=EtZQP<U&|cZlbQ9rF}fsB^!T@i!?p;rTFzy*;@nzrJ}mP
zCdG-P6#l+w_sKIyW9;L+hXPI@fT-86{rm|ZOZCdnJdGoLj23V8_p`Wm=PPKSj6J<L
zQl3ZTGY#Ky{O-%Wtct06KFxBCQjT-xiaU+=vx>#gMbTY+C|4v$gKwMBIf5EHlGrhb
z$CMrMX<mQHSr2|zx#F@{NOFZ`(A0WupGUh+w(@6Ic2iaz^_-p^z8$79TW<n3GH$Ch
ztE|D^8GzOrL++hp@i5II7>unicV+2S8afVAup5pr#-}V2iXu5VAAO%uB~A8vUHs*W
zZqs^qf|YqfUvAVb7M*+}yKW&IMH6w#KDh;$*86|+&<^w~H=N_cvT`$r5+__%bIyub
zVdPWq)~9*V=SMQxKAgVRg}8S>Db6)O@0D0GJVq(Xg7Ru*o?dd_0wJp<Vz`_M-Iub0
z<C$MJwk_-=tqV8Rws=`c?(nX~*(BW%PyX3#xW&$fQ<T_ockXNy*LUh{6qj9{jiRxD
z&^Uj}3XC7w+9(O$rMFQ$4rp!^#e=&W4qg&L`0(yV$?;CzjpA}ZccUmC-QASjmJvbp
z@et(8ga&<Bi+qQ&kdZHyu^IUeBYTnWKzT3n9p$Q|6squ8qu=t&!RS|dT1r9NaxeOI
z;$>_+qtS1%*8=)2zL=hkvvm%`iHODlH4%TbqSEd`!I31ez2^3g7%}4$OMC-Hq~xWe
zk9Wq1Rni_4Pi^c)JSbR<Tu_3a;XzC%M~}q_pqyKH?S+RVAT%VW;UTj`A|8$pNmNZS
zU95DC(J~es41H-mZ$WcW42zoih4QB1pPJF8mZ?~5Dz>E%zAev9<yMw#f<}kD%QAnr
z6`RVeqO=<Xec5a=0+McJiw6mF!h%No?Jo+N5SBt=>4PG;;=Lw-s|iQP?_1Iw6229a
zTcWojbTdZW5yWFiu61GtI3Z{s?tdZFL^4aM4e&vsTr*!2%k_+-gZWK)4vFTP%Prwt
zQ|goPW(;^kLXM)g-{3{cK?km}z)pYMx~uCVkZyi%6!<i>C4J1l)ntXo2Gfnt^-XI*
zcnjb*B@+Zb*8(gorbU=2TCjlVn#emsag>FJ6mEdx7@QGtKNyPhhF7x%ir0eQ35p{G
zlcb@&4-`k1Je+VT`*ZVJ_Q>q7n6JqGYQ{qL|Ku$^UTnTL)r9iaZ@rXbR#tyAw!-6C
zg9|ac4QmY9cD)8fl9!PiJL@pmQ7Z($VDejJ@e=a3k$$w+^iF1e6LksRplF0d9~9s<
z^EClp&$v&3ua<HOa3$qV2z^|jk72u76Ur~$<w>o{0f9b-?Rr@PNttP+-x}xwVKMLr
z1$xbVO`z8^?i1*%rJMr2p0R%iTOSzYeIvMD38ExVraej=u?|u+kvE9Td2f(gaj!3f
zeLVmq0l<M4nt-byUSdh`p{B3_U4loMsR^v<A0$r*8g5u`Yak$b<Ha$GN)FS~-#isV
z*PBGwBtwh_W4OU6dBOPvja3%6Si-!;a#ga(Cb^dG`uFl=b2H!?-FSaIUNA{>UIjsI
zXL{8qo_!l^eco@Lr;oD@rG5nH8PDj?8%*=I+Tx~>a|T*lZ155XHj*PWg?H>ebIkKC
z`#|7WFfiY_`JA>(S01IPH}+5GXX`Ejsf}Ap<gJM*DO=pBpA|Sn0?qF8{}O=Gu_t#P
zFZ<na?t{~KxbwOj77u?e0kC26Z{+N42crdNa?q5oL9dK6o!2w2yh3fXu(BHGe@9tk
z&Ksc{WE2rebG`=M$`K<qV@DHy$pR_wyT|5iJM3DodrnT(0$j60f0(C!{DwyDk})F4
zqDu!K^skA&(>ec|$;P90AhUC6sfjF{7YZWWs>4G1F3I$N*Z%@29+#o+J{6ZS`6?8b
zNXb6O0WY@@%05U41V5)O;+HYTJrn^rm!TdmBn~k#F)=SpWo~D5Xdp5;mr)2UCzs07
zJ`aCfayY}`6hVM|)M)?(HR2R#YZ!$rZ*8ZJR*=?#^Y43xcSXt3Zo0BoyJ+BGd9UY^
z!{Kn|oFSb_O+pG8zcu7idP_r*+*>Vho8DQd84}b|3d+`7Iar<EQOF6x-gNPjdy~;)
zl_PIT$kKai314WQni_Ir?M-}aI`v?Ia_)aU<P?aQdou?r+GR5ZlGIf|KU8^=<v^;%
zN6RgcwMoE2N!0tLgl6>~Dl{)Sr`{}RO6g5P)B1$6ghHkFz%7)b;!WXDxb-ICvuf(2
z5#BcHvV_m9_PJ3FoUH3h8hB+AXu_rT3E>OoX|$38*Qp1H@yaBH0-H{6OOTxV+zNjO
zA?`X92~t)9HHfS_(;!My4=qTmePGN*%=*Bbu#}SkELY3a8B?gNy$3F#*3^77fvk2r
zRIFx&eS1?v^;JL<ldnxC>!7nXIf0-vo0bK0sn+_y%0V-w-fTgs^d^>9VLu-Yny5@(
z#x-c=sWZrG>#|}&ZStL(0vxQ%VupV$Pj;vjV8SkoWkBDPEQeMnJ5&ixjeRt@uBDZ%
z2Fpyn*@6}616K>i)<1(`OjEZRCTltkEdF-#GeHl&VlL~=$zfnFv)nD|{iGz_5WWs6
zPi~-4Ad&R`)Md)lq575*>iWC+Qk`Vn{Zjv1wDHC}?QJLr7wO+NsGp1Uo#%hOWtrp+
zy}?j@%5Ko?zUyPTOzT^tHRzCSGD{<G_s!A#t8CrHtnYC=dAm!}<h<sZl$4J?|9tl2
zuNU*#m*?k;<?Pw}mrH*-d3*k!F#Bq8`D%XoJR$0&7qh?2zMFmhJWmfl%+Hq3vG&oh
z&{9*Z>^rVRI3AojMJn=Q_GN!S_s?eEE`D6hz6qakH&bz=sBj{#?T=sJzmJtzKQ43+
zc{MI(5d{k-%#q*)StuIzt4NAk?ZweY^O$`(x<6aKU7SOqzy0u?|Nq&W<?`a{PqW$U
zx63#0U&gb=&$FN2|9o-$YW~aY`_tu{zg{fAJ73N(|9yHkk3YSey?lSScscteM@liz
zt}f4JUtPX^^?H6ado_Rg{xu~2a3a}`PlSA8N57rfJ<VG%EyNOpp&Sj1%~GOoYvvql
zZFkHI$6}+ZSam^+N~x@|?mczjYO%GUg0vJpcx_SvsiiV`M6b0_HF%vpT!~UF7DQ8q
z-4k{I6GnHf$lz~xm0*8_QZhk6kVGjli9N{AIKl9~lU9BPKZXnN0sQQ0zWmqErx)GR
z^!Oz6F5Ik2EX~s_b~EBS*NWYYq)8mHi2HG~MiFuNS_QEZ89{E73XkJvh=);-3JaDy
zxdAdo50w6k-23=`hCGH{rNhq%X2{rOhxM}zN29Wnc&11T9X)?W11N;~q6A9nA!R>q
zrtl}n1}hDu!S#_?H|g+5en!om9cetK<W4S;94nY$-{rf8z@PxIu!I250rKra1#K|_
zduJE~IT@&!yDqsQgSWeeK$b70faD1@A(_-tcd#0Xu*=I(La6VvjY?F*QV?s-{)>&W
z7Czj2(-*iuuu^}e=p+DWKRT$_o?`BLL?Q4X764wLV!%_N{2e5?rZItYS%y1eV^oeE
zVCD#zxF+uS&H%tlK+fijr|rNcu7i*owgPBdLT&{HZWf8w><kb*&=?2B*wQH_7;6_h
zLnET-wGO4Um=TaS=>UEPZdkKJ$)aFKRUZ_8D_p@j%_@KV0{Upy14mz~$0vL%&42UU
zId8>Oewh9JAOCc=5uG0?<#7J~-Mbf?#56jisvsy^hzal>ynHLsj9+l7Eku38gClGs
zN*<Zh?Owy|>&3gp<+BS|FBJbE%$}YuVbAkG)8n5%EWdrWJYCL1o?bjzoG&rR6Avng
zoPCrhPMd$=FVll_CG4Z=!MP+tygz*L#q8<j;_TUc`F!^Dn<uj$=O30YZh6+z)7Nt-
zgylS6UUe4SKZ8GAE#6<A&97F3+TVRYfA#kCtHp=zfZ+3RZ$O%-r<eG)6VgeS?yOg<
z;zd{C==oH>@VQLYi|dw*Rvj%Jtu|VA-70Cc@@Rk2XpJ$-7^93a${3@JG0GUDj4^nO
z!D9>_V@wu*@vw{m;P2L{iF?onn~VqmFlqr(D|^8ZgM1ni255axAfWCQ(b~$sC=nSW
zRkWc6rxZ)p&04T2vXv5H6Pl@6(p}&wZ0LNWg&=dM!oQ=%RuTK&We8}2E#c&}U>}y&
zpp}0=8%S%}7nm_1s5+F8B%MrbQsOaC-Ehqc0~lZ}gH>VQiK9Jgp9rY#QI8R?5|mr@
zm=?LofitM&<CB_qz+XK2xH%b6(aIEke93(mRN8O!VM)Cpa~*xyOK^KF$TS*CFjdg0
zyip0`+A_0(WvCz#r?Lw!$8P;Lln}3uAw_>~wuB~u;s`l_&N+kS%fqTdPFHSKAH5oN
zaJX3c_(XT&>;S`70Kvh<04=3T%nr$6^^*2edj$8v*cj8xeSo-xl?Ek%PM}Fr_T}sF
zCs3WC1^hTVB-*6Kqd0r68oeHdB%{?&&k*37`WYAk{T_VXD^W$tt;!SvfE+rqE1ZA5
z(GHv)>lOP!DaG6{yW}+iJQq0JUE^leM!5mSAa^5q;?E9Nc%=g995QksuQ_^cBm?K=
zyWCeRFq}dWQE*gXD&?K&0QjVBDvEiUGx(ay8Xp5$0AV*ue#8q8$v2MPX&tvg>ol(=
zsJA43kD#s*^P?U>t8GL*ZIf=*EBSx%1~U3Kq7}Sb>i-WxRk=%0wR{>u)pm=Z@}P`7
zC?gNbD1#`<Ac}Pm#X5*$9YnDXGFnH^b@W_E&vlT|I>xB>aFR)kz;5tF&4>mDr|>Z3
zo(Z5~YszuhEX|}e%N3+X5(v)D_F~PhM96+dgY;Vw55fM|=%ALE0r;UIvIKu8k)7!v
zCed$LTn7Ya->uy}4*`dPXj0e$2?|+QcR}v38lA%U(Qcg<!*59VPStf4d-kXd2%=S6
z5vE<#C2tO#VSYG22>>rmQApPTk)}0UZGJ8fF5IliZs3sml@Tz@<Too2m|464hvlHV
zhh#`gIUzG-MV3NX&AS02pp<_O^?SuS5MnE5&<{g2wEd{ym1zJy&@Mny@<`lbbvEL+
z5!M{jZKuRNa6e=5_2!h$kL6hpaZ304^7P`(bVirx_$2o_#?cN|I3pGLX<aAtSZTjO
zQ~*I`4GusS=G;=E#D06ICCd*oZ^4H5-)r+0ZZRnz*1T@rT^g7CYKMQ{<yje*{JlQ9
zZK+g0P`-ij>?qR?YFo<3CnP(xZ3aOD93HmK2-aF3-*KHSXuWaVTr^TWtmBq+d{TaU
zdZfJ7>Cs1GD=5tp{ch+Mq6MiE$Y>i;D%|fiZ6ONq8%woEN8Pu8_D2x>-l*H?F6y@Q
zX{g(79}DTg&^j=*4h(<Y26Ao#Ik$nF+d$53Am_FN<Xp(>&El6@WAR%B?9`C~*1RrU
zKXC`%yAUr?=|%}ZO!v#Bo0NDQI*-Xj@)MDj7QznJ+D0w5<Ma0^#rq|$d{c_AC1QdH
z4MvPeFT1Es7DNGd5bF2;ydWq*#C~t0_(cKRlb-?lO4gx7ho667v740m-FnF<;m};L
zwDnf~IYlNt=n~g<e9~Fp(Zgzu9(!-s8f!T$Mk$yVF+Xf#*<*oUJii7A+F~mQo>>;m
zjOaHI!Seqq-QAM7smV#dy}5;``>y{xi!DUSD01O#MEj+!ve!!^_XPx&_zobjM4to*
z{6hZ?Ah40>Z6qvu8;RaVqPOcZoTW5cc}?=aN3!t(5#1ykKmJ!bY_cH*^hE}q9Q(f&
z?ZJ+hq3%8v0Wp{1z$FuxL*70xf9EOu0;?58>JJD4B-zOVy>&6SqNk#}r#*Dv|4>Sn
zY)f`x&qT=-=pk_;%hV^5`bbKa-H%RoUw5b1-RBEZXNA*XRrmV61HK{o8zeUx-B|!(
z5W3f&-5+}?mEq5cKO`k0e=PrYeAS0f2#$LI_W*7MIDE1M+=ASK+(&7?e@gx+{%up+
zB|UHWc_)d21eOh$uBXGFufOeN0{r=+-7U1nl3+|3Z9qqimK;l`fG&(rELW#H0gJ}|
zzw!r?+((Bt#;S+D{nLQQ7jC&s<&j_PA2ADW0_<@4TK#)@-TkwhfkzSqru*yXF5Dml
zflhUb?|R++`n&tS`@NQ$fAk`S<VX51ExTVox~k<{GPakjK52~Bh5A6UPN_4Rjjab`
z(-)qMOH)4glT%O`#KEFOlI%9RmKanN)I~x4QF2lh1)TXS!Y^7F{vg>H0Y=r5l`L=#
zz1bm*g}V3PhUd4B@2nu4=?Y+Ze9yg{o1yhdqnBO}YSE4OW8REUf1wlP>_&Wwl;Bmm
zq@U7G+;CcP_cvhlPOz+vHbiNJ*0!My5xhVJ8+C*Vg|_;MHe}usXt(+p1|L_8J{I(Y
ziFPAp2YafC?zoT4`qS8rUlCu?GF=0iFTyJKY8a6YpNv$iG`mc(M#7@!6?0yuztylH
zHHQUtJ1oR-W}ZX!J=$od`VVA!?UXQN8Zvz)u_4n}l2@3%>fM0pD~T&iU!#3I({IC<
znSMb(Phk3OHq1=l&W4%kx9KP|{qvWt;65t>?3dZ#J}Q5n6HCY&b}Idm!32=qv1KeN
z1hxz%h2XVFQgm(+ND6^#aioO$gfOxN-4I1)x_L$r*<!sRhP1NY5JI--XNVxrvfdCt
zHlfS$V}|Z(_<(+o=AO_YIas-|ED3XkOwVzwMBM5!rE`9~4*{2l(>WRDA_X_-T&G>C
z<C7?z$s2#Y|3FG7qR7PUIK7hCkkl*5E397iZb0jm#1&qz(LSEox8cjozM!8cQ2RC;
zW^Qk1!%Xhmbd=fsc{a@Sz7b#I_XT`c34WlqyGWjpBZ@WuoBH7Pp$P`BhS5JE`Gd^?
z1NRe+3`Yz_tIM`<5*-6TZJg=Cew$RuCQEk;enfxv9hjjhf*BT}>i~8?U(YpDaP|OH
zi;X5RgUIu!(qO11Q;!qYD*U5z`sseTHHRE~Pih)_>E;CZM;yuTWB>l{slP@nE^kbR
zUTV=SaRf?Zj<{iT$tY!<J4GZ4_Jz5JmVU%2jYbqSA~+^(MkE7*<EYJe)mfv_NHZ67
zT%v!Psd47!NUBW2!pcd!vX(HRaY7k=BVjV>hz}1Q5+5MhhbK;`94EH;fr%567S8&H
ziBl!9Vd7LtUP+v)-VG9`O5#f5RHJ=-;<OE4PMj9>^Mu4{n+<c~)Xs)EaoVP%oH#wt
zhB<NCh%Y5h3;3=kP6NgD?RO+kh$I$I(ae7&Upn{0MOX5Z4Cp-O74OZRR1R*(QT<~8
zn&wg0&+b&tvgT@eWa+LvGpEDBp^quV>ryiz_DoZas)X-j-tC-DSuYgR&=B(uVEjG_
z><2wA3>@EyAsYHDD<Q>z`QTOyfZ5<N$<;&bzY3k1(|R>~+D=h3wtEvzCA}c)W-2ne
z{jLq_{g}16ef7Q#*fP^wWL~G4r>Pz{W_-Q2U9YR4Y54Naw{nPij%FE!M8RR=DVML~
zKGc6Hoia=^w}ze&A}^3@=n1vNJP-4(>H!S!Dz&xLs}-P(FQ!(#@2UWMBCaG}rR~06
ztkFKQ0wfOaY9qc}FD~fkp%ox<c1Ig+h*AlZ*wBUuI<ClVI?DCp^K6JA*?_G+mg~g@
zd=Gw!5}Xv^{8L||Y`}2Z31wR*rIo}6v95oTxDxB?8JVPE><<=5<;p3o;(~$us-xF_
zqlgD&1p;oWrp}XX-MlyK%l+jTC5J@!5VxWibKSn{QotM_;4sx{tGahw7+iR|-yB`_
zJQiYuGrmPNOd^I&g@%h!dQdWu_c~;B;mKnBFDD#893or3&s#}su^dZ@Ys<0hJ>N}S
zDVJpUO;s?tvP^sYHn<n0uZX{B`8MC+tq@5)A2GhPlF8Ti|D!E3nLxOCbBnefCF}cs
zAMr|PXql6i0yR!7l-7JK0CcieDjwwNm(JxrLI%3~7u9xJm(htg6az9jGM9nsJrkEY
zkS!>eXy!hde?YpPkAJ@Z_QVszoH^}p{iXl(_WtzGQ_sGVoM{1vzkWRVFAxL+D-S|B
zhFl(g{(boN^m|XX9H!tfz5V=t=vKZa4O&4O<XS7$Xizu?8V*|1)qfQ=$Vm&zagHRl
z^7o`2Jkyo5qD1^&ydFPk?scc$I{CL*D(6d50U1>*f6IUbtIwwkl6LVmovZqDz3%yC
z|Ji65f|;Ftj<T+tZ67~3>l`#gWAnMB;$6~Ys!}l}+{}^~rSMV(t@`^*vas%@g8>#5
z$W$@t9v2}s0eP?Hz%+$JEtp7x@hC_uMA&gkZlZPzq?}|LVKUclQz=Vn%VMU8V;n#k
z&WvnBe_1AjzeYVF8JxC77k5j#8xYcp(wEU~NbBTqV!EH(3=>DevMKRkcHK+jnz&cz
zeRj?ILdz^SgTqh(X#s65ebBm`qz^Kv6plIBnB{Qv-m!xxPPG%M&S++>TlL_j`ttdr
z?6uO1*4EwzYO);~DwpWXt+%L3eyDg{74Y(we^XH4W2?xc_iI<;AIwo`=Ur9$s&4$9
zI!dpuhJo-6V!ne^Rf7AUVx3nIKTCH6e9jsF7&--2_D2c=PIAhe<S415P<C=@GF(vX
zy?p>6FllCOfS`n~%~l)0={NCB@#qfCN^S!jUv2}Uhakm*FeMg3>9teTc|}Okc+F~K
ze}UA~fSOS>Q^_HfMm3@5F~0@Yo0Zy=wP227wWBIqPcFO3V^HXRcWLj=_n&WpD7l`Z
z5X<LV-&oAFDqWS!Z2X*?3{|<O(s%PG0ht7OBYzSsLu;1w6DmfsmFJV6QWVelyq`x0
z6vud@4S`Dr90uADD4xh&Z`2X^d@jW=f3zWJ&8Ujp${+bSm_;A+`hhgggi*-bQ&zg8
zJv{60{%S%kGn1lJIElVm$;}kI=tH^A%{;HGP}@mvGsBVfVPt-*>FvoZ3TObbExon8
zq72hylxDALDsQxOTX(0m+H*r>w^=zfVoCMPcn=WiEr`&p)PaqfIjTwxTcx3^e~uI)
zvZE3^{&O&et|!cua3nJd*`1Nqz27~6l9MmG;~b=;IKV|ZiW^SSOZ1XVZI(<QJDMpN
zALLXD>D!ar7?r~3k(r`o?(`F5kemfJhFO_9X-xAC>SOK+Lt}mN{<16a5t9t%6r>&j
z;!~*%C-G{~nbMnU$FGtI-!_tFe>EIrL<W-@P){*+T{S@sjdpCDkG)oM%lgoCp_t%~
zDv@BQ)!fl<VQ!f-9;Wx0kIutO0ia_FM>e}1#a!h0>F!)wV!fp}IGdKx>rTnBL33SD
z=A`3hrH**%m@{*Jb#4kgCYw19ZXgQO;EfG=UvLXo`+EK+$Wr5x$yNoYf7G)S^VmoW
zLNtlUOc`?qo_Jlq<S{qLF2&kx#Em&3!xlcW^RA>1Tjhj07o1a!W1{3GKHoGj`$ii}
zJeDM)D^b#rEYUq@2cl1kx5C83g6q!@(=@YM-5qviR9NNOd3ewDgx-#za{BfXytLyv
zrjjK!Jf|JOO|nc`b>7{ge_UCSF+7UwWqiL=exOq4yWqL}b&R$^<yAd^q7m1ryY)~c
z#<*%yu~!S_D9Ie$t%SNIm#0EbFAF*6&sM*FS=EG}2ZS@d+ub}Ur804S1X$A=eP&Aa
zX7G=9FqYnI&2yQ_X5nI~ez}6idJPxG3(Qm4HB+IFWx9Y+CPPvWe};DcWwKpq&UAsA
z>NF#46K>J&J!q9Dfe4Qk6t3`FQ>y{4Lf<C3r%G}$e=HRAE#Y|vrDZNEIyoFdD!l9i
z&S+u=E_&cha!O5Le+h6#fx@hD1AB>%VsL=~B0n%U#7Ur|xCWYe_(98l_nx`MSmxwR
zqE*l=LTvr{whEf+e<48lY##&5q;FW=jeJR|n31@ommVz34^J2oPmV<52VR<$D?O^D
zDlk`BNT!!`V9FxGw|wR%*YX*TgXpP-CmyF=BauYIoLp-C!LH@?IQOiJwD5ppf6p(a
zrtKpEpq}CwfV-o*0dRM_2LRk1#SMVF=l0_P?%Vjw0QZ7^f7Sr*jjru3AU8qR*KK*v
zRG87Pj^f%>7@7-JO@+RtcwNWV67bwp-;owP)Q&iZh-EdKCXgmggHlzx*ReGzKBGb-
zoN25=$^Kh!&+tDWRe<h1M}9oW>}MGv^}sCCQ5=9Gb`&>I#9pFoKZxUq2bOb-dmUAr
z`j}mwC9I?_f2DaGv>QDu9YzEU7n_W@bqT8~EI?aLN4k}!wgE>rnH|LuZq-uUa;tWC
zm}*s}ciXzwZO<<re0U+5LD6$GD78vwDG|CE^>Gg_1eH3N{sK9|H(H8y;Ced<^+0nI
zM~jZ~YIB238`1CH)!f8(rK7m!qt5?3%g(s{$mS-te=E1~S3GvEpI_44Xr`3eu;qQt
zO?)ojQ5@CWr03*4^&KTh8louwf6Wb745z;ekH|DF#Q{)9M{xt{=ynf)Iy#CQP)C3M
z$3Y#p@s**D1-*P6)X|%44Ajw49RPJWWa#vN0O~N3sp;S41H!Z;I0oowiEiK=?dAb6
zM@w)6Eam7}{dkb$Chl_6FxSm5ggCBR@ezchB{&%1=!l<}9y(;+yX)Kg)4vXjj+Zg>
z3KIk|GC4Dsf$BXI0y#C4;W{Rl^6oyNfA^=qfBx%<6TzGXiy!fA{PyAV>DSXpzEOf1
z;Oxtf6aJv#tMYOY+B2=y^4mX`ucuEV+On8YOa1rTw`Calg(T<&NRV4=OeaCXL9BYQ
zHZX-hkp%VAgXNNBX+Rc6aIVkrviJ*13?-#K{Q^)LBo0F$9(c#!J`@~*aF_<df8}5K
zGil5uNPf=0AS}+)F0G_r2RU<KBK<QwvEGdU>+fCzzsR?COe>f9)-SMPT#2{^N~P#l
zok~)3ut76oEVgt`SeQoQ^1n!PHv%*lj7BjP{D@KX^i*iaEa-dWDC-XhF~|{qc4-VH
zpJi)o8t1QBzj4GQ3_MeR-Pl$Ve^!=h=UaIs{U?i2nKs;`=VVPL(}b&s1RpKTBll@f
zMws&x5}K8jhAKM<no$-upBoZX5<zC}NV5`}8GEEvbqJ3zst(~bo9ckgds5B0s;V_B
z{2bMGCYJrw)Dra$tw+$mNPRcbP%pJ);I9XQxLGeKvR9(t;&&wc1>rZjf0e;AfE|%%
z8Q*UeCP-$%p&w@!1IBjCq85a?-9D^k!pae|<l6?~G_y{O(D&G<NB6ao)yzNx1b**V
z)DFh59e_|?H`6B-1ZD4b(wFUZLq2_Aqa8BP9X*#On*GhRDsGuE?ussN*?&qHG{zC*
zK^qYAG^X52_jd#z<Iz6!f1fU`F(xo)j|@Nq#sebIfbmKQ8lVknDW(Rl#+;}NN&1+?
zq25cz6ausOw}c}jm|V%`sNIlv?|}Wlc%<O`q5Y9o(qqg!gFoLa+F~U=#-l<#_sns-
z!k%RdN{vFSvh7A%6Oe*w8&;TX>uEBj(L;!mV565&87^9KXp`UKe=j7#nq<2<Vwlk8
zx|fciMSGWz(T0P}T{#3QL2E~f^3gf^LmYjNBkr-L@(qlwyD%pA=1Im(J!gNGvPYTZ
zk?`o^PoIQ+?0ss2h&~lGt6pfnN`95*!>4p`DI=OB73nF|X_BXFe1&e#54O&Xv(IH_
z)=l=@w)yM|RHz&$fBZ8^3H2;z(Z1;P-K+7Jw5FY7E>zD<m?F=f<gc1Y)zYgwZHbzj
zNB^iv%A(7bu2~VFV@lSTdrmr*-sTKdmW&2Q-D8z_w(hc0^fv32tYKQZai!|lFVhJ*
zfw!y4*$B>=)J`m}6-P_<PQlTPVgO{;gh#hq1V_yy&6X|Pe;$<*Mw2-i%?^2M&92P2
zu=x@_IYA|;JkJySb2VP1tpwaxfDCd)g?nt%qayQ)-ztpDDOFvWnD$J8(;an(v<i7e
zQc)Ws%6h0bV9>~}m+*up@uQF!!^~?jRDz5W_Z<zb{t8T9Gh@Y?3_qtzr1F6_j1C<q
zlH{n=s-kmof1@Y``H9dgdF|UMN<*YKVKv(&XG%QjB2ClC$yI6UDv@>Kn}6|5P>A8X
z1Y++crjYRRDyRLc(O*7JWuUcU0Yd%2g`7~E3rEoseGLLJnVfN$l1S5(vLYj#-d~DY
z^|{ox*kM%`f*J`HFC>y1i_6MrrUBhhT=sCp3YSaEe>=*?T0)L0)x@xFIu%GD7X&e2
zoYfta1un!0eb1>x2PX82qv@$6WvAo%zR4mstCF|VMHP;nbOpl#oSZl0GS(Z+2ciYc
zfKCmN<3?J#kqhUDocoSC3$%i)d6mGt-0eC`(4lHTYI#?zT^XV<%-9;upKlkm2(aDM
zh-qekf7x7n=(`Ls9iaw{Gev97d>DbtX_!vL*TNTn-gL_1aV}@BndRmPex+ho_z~lb
z-z@JQdSrNSMz&_WbP`enL{dsq+aY(bDzycVZg;Yp?SGK;OYMmu)hZXSeBKvRov8ql
z*0mVamEmvKo03dOJ7S&f6SC?7`hfD$fbmNCe`uukCl-i7Z`z2R^yNathJ0oP;*FBy
z%`{rgxKKNFD4hgPEDzKn94HvP_R7cywl?FRs6vI=h1zRk3QXLDO%p@$D7sJ(FC&nf
z8tTHEn-&u^U)#9z<ofnN6GO+az-<5K-PS|jvwx|<G+><VUkCGrT~RyD-rulaITOL{
zf261E?m;~_ofdW-6~yC}uIbzJHU**AII|M!eR-QgQ!y)5&FtF==A=I`kr*&uNhAj7
ztJfr^`|rw?i51_LGgnV_NATTayvBDA{YUolg~apbKoBSQU6im93CDu}tjM3^*092P
z>mW^Y;Rrn*pr!LHn1Bc9Ul#!UxI}0`e*lirEdN{LFyur=<NmWkIE!c%{GkL+@BZ`c
z>tDY7Lok^3Uj&6yIP6dXZuLj{3P*I<aY`)T={v-qAYpC60#_FMm`nz|;g%<sfiKHR
zgKC8`z8U?Fqk(C!f&V%=p8XvH<^`IeOao#pqj|mW?O>*=E)jVuWZ*Z%LUZvef6ULb
z@GHLB?->+k{cAaWT~41!s>Vtet2{g*4i^5lqzWfd;Ti$`s6)!J-~vQul1gx3@{`e)
z-!h2)%*InSCTjvdmDLK*d8FYx4*{->ArIyEaeN|&fmax*@fmUtm}i)94oy^$bd!YZ
z=Py5K-Zz-_v3XB;9?+X;@ZJwfe`;XKBCc2fnlv8>JVS4{jFJOPD{bwt0SPSmX=zAG
zyggLrKJ6BW5X^w)t_x&z8dVSw4w(_U2x8Jr$UJZvYauz7`ldf8o`45idVEi{7evKo
z`gJSDl!9oAGdh&QwafF`)`d2uQP9?*UJGnfluoeGiTd(;E|xc4x>_+Pe^68crqxz4
zq=m~Re}dg80k3iP4T>tA+;b}(Jlo-xsc=sC_V>ZMI8FM@+8^_S9j%_)t8CjP)EnH<
zE=lv_#brDvl)mDgW(2KrZu_e1{w!W-R%+dY>^hj#404bCd5W49ZNS~w(#_`C+R}A<
zrY-&Nr3a(X`_RPROU@`qe`?G?Yf9N&5h=}1oiBvukuCG3Uy@0t9!X54H=2n74t-)+
zI`y`9bxWl(5C_B9E!9Gf<*lZcocrclsGiIt8YaSO=8PLDyFF9f)$Nd&Lufy))~M+y
zz9e3Z23s2Wdecp!UA!p)8M<2W7X7%2ULfnz7meF`ON6xB+?~j#f2>o(tubYS-SWf<
zZ?>EkTaFTR$AKW%w-+w87y&UeTyEL1y1WClS8(vwD}5C2+VdSfrOmug8irGw&btlV
zQ!u(MBeeGp@L;ha!;AYE3yhHH510r0ozmqr3%jTqtd#4ER7oir4h@*!+Cv17fLA5n
z_I%3qduq|OU%sj)e<8gBY7V@rHekG`bxAem0pqNqM)j`&`kq1n21&W%XjTZoZRf&j
z{o`Agxjnd#E-4LjQg6xSXS6O;lVw!dot;Y$2Mj#b^;n12<y-DCV;5NHjcQp^Tzn{I
zQ&5+9Ldp)mc`R&T^2kRV284GuuPjy_+<<iYa6_s+_rP~Of7_6@UXO6PPt$`C1F1jr
z$p#Gz$$ix2#~{|S`I&92pqLAKpZ4%TuDCi*d->UiUGh<Qn?Q=+PkRc(ir8se`9X^{
zb+e9DlpCE$@9|?;i^NoAd>&(bEHLiz>Al?2lUFrQ*YE3z-0-HQlCs0Rj-{--xw(t?
za}z%PyyEote|u&l-f3zw$OBwwkMV$}`+)IE(|zRVkA4(YK#HM)`ExyrD!5|W9ho_g
z7|%`l;SDC$V_Q9(_VygF1p8ryGi?L9u^ZJICRa2klAvpL1Lk}WM|?Q!5gm_(vT&gC
zdW2g=owyax_11CPf~(DWsf#wg!=dv3ZCW|93T19&m!<AL5dk!p!A}Ac0ys05;Wh*&
zm&^J-Fn=MjBTFPjQshTEa`-yP;g`et?eOV}G+5y@ST($T8Gs*1{szf|Mh_N17=+>N
z+wl8QN@e(S<_}4U$REr1j<5Rkgy47v@C@KlfYXyD;1T2z<T*<7Rq{viz0G6Syx#D7
zCy4?DrVTt1bhtnL@%HOMW`U*fCftTg|LgVb@PGHA7vD%Hl%wI#Z-?-J5Cl5KyAqZ+
z_5H8m%i*`4YT+e$N&3=q`2Kb1Mm}c+ddUiu#%MiJAV}6K4aRtZ;Zaf`y-2Bf!^9h>
z3x9aC3MQNUbak*uktCZ!_Y^g&mt6k5rr+8o%37y;c~0Ckt6uUIUK7U+TYPtaWCfw!
zbANwRig7=@4+{HK!7Z}v6oeeVqU@6qmc39mMu1U0WhEzamD!ervCz*hu6wzkzYCOR
zx&m0{@A=3%H<Q+9jb4+j!|!0vXk(e}ZG*n<mQSG*<m^WI6e+<cdr3b*9<tdNui5Vw
z<-A`;?*!A@XhW1nXl-}2Ap)N|1sipQ3V(&R`U7pqkyxPJ>SGvuTrK)o&=2NN9mBj$
z=Pv3FQ#hBS{xlBtE6O(+Y0tnIFVG0Gu5hH&laVUhX52rWlYUyAM7^2xo33G1OfHzU
z7d-DdM+!<S<LVIgAUrF4LOi)NaWe85=>?DNcAD4enSXbwX@^(}B(&Kxluign-G8{O
z!H^ZG%M3-U^O)P^jirS7>8jl=D50$p=<8YZCmqs#vS;3+o!Wb}9FlEMuMZ>c2>AS3
z&r&`!?*SUEIo287MugVu0)36?H}gS#98c&e!kZ?7Js}1e&T1k9kNV@7eNQTAlU$mc
zJv31Bc|KD$$Gp#A8CzxspT1OT<9}Vu_GW%U0^H&deoU|`VS@3DIv!(Vp*GI&)p)7$
zSzUV4#uUZwOW5m7y<|`qq5`D*Wtox_cPFJ#cO1jq%px)VsiKF5ZoGdjItl*b8x1Av
z#{)65egJ-!^FhZeNLbFbuM$W94|al^K1J-&o^i#5YU>dbtF?Y@@~q-B?tj=uzQ}dv
zxx_gQCa`bW5f&FK#|fzk^e3yb5@Yzx!7zjX1e+C~R8XJBnb{EW06-`9II<|)NI5E7
zXV;ifi)u3ES5A;oo!2bUbK(FmOTp;<nt&J$vX^uv0eK$84=D+>)Jzxei1U?0ysM1I
zkv7rh*xz%xWLL)qa@cLzgnyw0f$1ssVW6guZ2@Ap5`-LanA)?%>L}`4^50PgWIcA&
zSIZD5yqd{!mu~NX*9tg<`JUTa381vFYNq+ML=Zs6RPOubvr1^Cw#vtW*>be&?TAfT
zJeJKJBrgE)2g?LhLOR+gCo&l3*gv!JzgQ+X6o7WCd|bGqhnbx#5P!RuKT9q^>y7+L
z60i45epWhBu9hmXOu<^Vgy#XaK2WZZMYgAzmn-+2!{zQ(iAA>oE3uDwRiZAo<q?b-
zdlf3BTs}gjoMB6zo|iQ7Ne5xls!N-fBY>R%_wr<fpB28}0%pqaAoZK1u|Au#xrN8r
z^ul9L>_>5eGMbAg+kXQE(4Ku#eF#z@uHTUTf5*($Q5*}B(F@diq5A_3ua4VVrt^#|
zxXkEhKbSE<1)Fq4*AfO;p<t$-rGz2RSri+h2IQ|O%4Fp8ykpi;FHH*Fg^BHyi&-Ud
z6<5T5eSn}BKs2CNG4eN})}pmJ;kiNx&-t8<?SlvSYby+gVt;guM^H`p=gb#RLC%I#
zzTwS*ygtr5;53Uf-Z`SJ@OF>uoFj_{8u^Cq(<8eXsS;OywFy)#ncaGfBe5bYZH#_*
z7+x&A=G9PJZ1F&pnQEkePFItAOndACt<bf3B=uSyneF6z*^*!8m8Ry;a;#^&AYa>M
zqra1Gkl7`puzyS4?H)T-hL5ysce+-KN4t$)Y1iyV^5O7;poOu%dgkL48QEF&P9&@c
zlcSq={i1zeWGSbFlWNCY+f!TxHL2=Y>*&<FdRw-B2=6KjjK^@&j2pW?ZU;20kUgn(
z{|hjUjiKiW;N0ARiU$-XKo!;@>a=6qz~@qAl@WJJa(|thcM8}7W7~FWYkxQ!I_AY3
zAmX}hkT{l*Y@Mm9rcw7c4CdoP+I0kSW9=qqLp*4@g_t?x2F_evT0<Ry+EiR<75aDH
z0H}#=eGPiC9r!>SkoSs;ekQtCfZ^eu5vQZ5>&u?qkUa|EOwe&!MfT}cc0c<;WG|ht
z()3hUp?`g>1G>q3A^ZI1yraG&vR6_GcG(@FeJ~1;`)SY~96xdXr$Bo!UZ{^p_Mn|m
z6}jw$6oITj!2OW@G}3<%vWL(E)$~6=_MkMwSpA<u_5j3nw%HNcgJkq!?VZrRU$rl`
zcfcj^o^7;_;N!^t-Aw%us<$k45w;Ps{hz>1SP9=>8BzTMM@@T|@sB(d12Q=^mjSCT
z69F-k5g90#Z#gLkmrepdC4WOpEjC#!*7~YieEGic<>le@_VDE*)nb@MFk*T8wg5fg
z^a-2>H7*8#X$i~QkL8ai&V{DyM3>@RO1cc)Tl&VsqXbJcfMx)V0vsNV0veGtB5965
z`sQ>Ax;JU;lGbZlZ@E&Q0?`JV2<WfFpKrfC@Wk-Qoi>UQ51Bhp5r1hnWy7j37Yoy~
z38YoyV(~SxO6*C9_sU-<Pml-`>&ml6&(OttMeM6HT_*lU(`-H$tN_ELf^m1h5``C{
z@xO=%g^8Yudmv;Hdrtfe#DnkxrXAC|I4=po^L32;qseQ_jIv*#nCG3#h{-1}BYS5)
zCyok_%2m$83L96z*MF!`=Cs$T|9TmIJ-<Etv(hx}&J5V)uOAQofs(+a6=a^wkahn2
zd-?Y8`$#nz4QHH|#1G`l&+p4H@|rxsaZ(u98nLq$;G{*hXr#TY%L-FuE@&HZ&|w91
z)|%{ZO;K_3DI6)@W`Ie2%JZH-!jZa>*Tz?@WMJI58^GMNtbbuju1>pe@Wj|D8QsKT
z_~r1Zc`#H=RQxJY4Ot~Bo;p9L6QW$|M>t_O(w;DDh~#Qz2?jUXYhvw+y%t-#gDv5h
z^+6zsQZ{^?!VZt9g44+7Drv-OS3z!xsUT$@=lV#g_{SLp%N!KJwLv5YApbk))<%Zu
z4DKn0@xWx7ihs}1Y{NZPQa%nvGZ0TPd4vt<DNPQ><88qkz2D3yiC|QiwL_}%&Xv~t
zxtP{ubNR^eDa1JX&ia|VUAB|wG<uGL0L%4LI;<a7DJq=nm@*$`5uIkv;cYCoz8ZK*
zI$I0h+fWc@5<C{cPwZ+ITEAS-E|#G)TF%e|qG3iC(|^yXLHIPIg2iVUxpa~&$6{|O
znbJNxC<UcsQtPshYaKc7CEP*rr<^)#d<w>@cawJ6M-Uo9S@$`hM=aQI+=rJy`B4hn
zrg-69C+jX~N#@a-(sQKDtBV(!GWXaxnMo%NPYTL9m${Qj8#!}@GqMs9HIcapXW)hZ
zAu@%hwtvzAc@mAx<0)N%DxjZ6nMahUP$!faII}7K3$&ErDVr@ms%KAGjW*t*2wEAj
z(vobUOyG+2X2eq;>}gEur^>Baaa-h|lV4#k*MbJ@G_#_xOraRLYb4}Xtt^f^rkyRG
zzYYD8W?RS@wOn%jCLT_EDF@8i;zg48iOO4)Pk&ziw`gEw@!K4Ca-M*dCqi=0xeLZ6
za%HOfc$y#}w`ySic+puf-bv*ZC(U5!W>hL9L+ut~oneDY0huPKWylV>yMjuG?|?uz
z@S@NRDF7-B3@av+GxT)otw<4{hBS-3t>WN$K8vELu0Tai74?4nsT~6@M(Xf&1+K~p
zl7F8D7+3VRv}H)|=1&MkBW|@DpYkT8Agp9(yan8zYTS_N63PwtHvSHCn&;k+d`-<S
zbB$2UFwn({RswzAe!ph2ocmoSnj!0PxnQCI8>n_8uP@v6lhq8wxQWc{S0sK*=_Dvr
zNw=aFa3>^6Ioz+XXKqH!(5dF{HykRgoPSw~PmI7Rmzm+NR-bvKrRW4xc6GSTgH0_|
z3c-Y4v9)DkK4(_&>)~x^2IM?{EZFESpYN#y<ZJHtInHs-Ohvn;BAg@84aAlrw50T)
zWXYcr5244DGi68$l<h-a_st8drf4ZY3CW{H2n}8fxIosVqD-eNq99uoXt8gz?0;ow
zp*16^ji<RWB4*yV<K>nVrBY05HSX)(NpWbfbR7x&xZTruY7SSY+2F4&0ePunJ``_v
z&#auHK{GpZlH;+~RYB*v*iEm^l`Wzp_nslY0c{pcx5nbTIL=5$l9}yuC2MRqD+y&Q
zJyJ~myE>l&W(-Vl_b%7I8>a2J`hNplQ%z0)q(EE0P!LuWRh4ztVXk;1mhIZ8x->qs
zKVP~ADdc*ki4$F$^@<k=OED6r=l(x<0SK}sHF9^YlE{`on2O|kLsr?;ArdMRc10=b
zP>8PfDJ?af`~R&jhF15wv2B}pooks(?2vTvyKBC8xgqLYw^e_I86j5aT4fXe4w`=s
zEMa=m6eTu^nhF??tSEKHzR>K+>&tcdQwEhvxBa40tz*|&*?O}g2(>9K|KnQl9gu%B
zgi$y*;D!S=^`SHeh+Uj>vLt(dK`4|X=rY!W!q*Jy)9XQj+CRLjKDq}>3E|(tl<h7#
zP`05nxP~h(kxd|(K{lBoSlZMaI)#5$Sc@x$T)5v!n%c8OW{q0Fksw9JlqZg}a$}O4
zkn)}xR{zSlD+Exr1mljO=N<)|qsn<6-?~0fi9B1;Z|K>TS9e!%3D$T!#M<s9;4Qn&
zYS=9uS_mDqJ9XHt-q)dyZ>xpd&rwva8S=w{a<8jC*N7%Z#1|*!$=LZmHIje#&z@)c
z={;dzdfI>6d!^T{-k09Rcloa%G|Q-9AF1w2E>gY6YImiAlwG;HS1R4=U8$6w{Ear%
zQ{I!zZf<r5shX~~o%r)-N>fs@mG=s>U3^`ZGW=`gCRuibuZc1d{&YhFg8QiNl})#L
zUpO8&U2b!~p6agqsVW^&Aa8&8FAP)={F(CN1Ul^9!M<C)FE31gK|0=MOg-g2nFZ;z
z4Wi>c4wqUFar9VQfH6sQ|Lf6PcAQq`7ln1zu}M0%`O|@(r{~%X+|ZggEPii}=@#$n
zOsBt)d23Wx^k#@S4#anZfNtKZpVOZlY*bolS(|Q!Oqn=UrU3-!x3eipc$r$QCQWjy
zU<k*Gd8?Yz+t-`i=oeIxS}`iW-%&)4-&7*~3k*;#m*Lzf69X|aIhSFWCKH#m3O~gG
z-M4xRKLrH?{<zPV!NM^U1u-%?I5n36*#{Gstq?!Ne<;7w`5!NTdEklQd@WQ;#jL@D
zHpqyUMCFyM2I{-Jg;^d^F|?+G=8AAD!i_h&0)2>9B$^d{(BJ<|--#!*>3RANpX0~#
zslxkb&pvt<iEVWB!T39Z1R|FdAhc!DiJmz4d_JL&vt(u!1cI*tL%I`0PVb7(Q5`|#
zkiQqWf5Zstw8kwl0_BZMUb(+pC@~X7cxluyC{0<$oqns*8fd~bDI{DIJ)W1KJc!a)
zJU!QQ3iH|tN_&&SB-6BGI>Nldq*Y10UOYuq@_Z5~o+d1WobGu1Be~_>TNHhIuTph<
z?jZsrCMok?V+re;0<F#>n5d2B6llOCG&|z(f2)_fmrCK+g1w;!RuxlVMqZ<wHw*wZ
z18h7K55~8uCD4L_#vZzo%_?r=7)sGYOSKQTBPi|$+%B14@S$dQ18=M>gl2B!ieEOM
z90M_MlQkM?3C=*IU0{=am91afRB{H%rUFNo!clE$I$2xqYFxHh`hGh@Fb)bMt3E_>
ze<t4QK`L0TFtQiSKW|c;h|7LtQdTn9^iK9A5`+KPT1w0=b3SPvy^sGE_UyA?@~!L9
z{Y%oCs^ZtWEg?AaW36<e6XQ>#@&W<w9F=^dEiHNP1|(!SaX#gUn3YY2p?x$1KDRc=
zt4OXQtV`=fOji#P*ch}5mI+jiG1q~*e~M`WvZ_kt^R5xZOr)FcE=0QgM$LJ=v#GxP
zX5MOHi^)u9MZZabAg&v%aRu)X2x3wuZO#Y+jwa66>jxUkwq13Ls08DFH@33>SmTY!
z)VBR9rHL5miQ)HsQH5(J+=p?XW11{dR^ijG>$*=yWJc+Z4kA9g<9&fjXu~KKfB#^1
zWHKo-lI+a!d6gZh3BT>>$WEORA<`Z=lS<P|oN33!;Wu*M^H094ICUYJ((a%vFlVP&
zmbX;SJf=Ku2q~%JQJN|qWlL<Y`1c#E-Z-MJ5Hbc==TM&4D9<3<dvsw>%OVbWDa7nt
z+|&0lrCT*2tC$!Vx!iTOA!*K_e=s=g$>A2AQ0nRxK$jo8U(YWO|9uz)E|BUJbNJWS
z2mb&;F>M^>7R&)2e*5>~mxrJFRl91A*X3wXFQ-IWwd&PO5idk7EswQwdLPGqY<SM0
zwDIoQpWYMOIUwHCD!SM+XGn<4pJ0;BoiCH&2x8gwl*N3YEndlp37Gp8e}sx_Hg|$~
zJ3hz1ASojsuymD6i*1{OC25n}WSSoJ*m#E_5UzM+Az*0ktns6e!N5+gj)H%(bIIJe
zkO3z6U>N*1XU%1;h5S63G&&?6TwPhz`D^h8c5=!k&&}cbQkULMpylcMxg<b2bFTUe
zb#D9(y(JpjR>r);TOvhUf1%|6+gs|?Qj)PX^%k#5N;hPZEkobRYa%5l&dEQ>YohQG
z!M(z3!c^=9lg#Q8N$y1iulJT@IKHpd(e%;&GWg`4?-|NJ!dptiz{nok;4R<5P4dS*
z23N(dT!*@$X}at>Ddp9JCf?;brF9IrdX4MM?F5tC1AKBBzM{+xe~_t$3oGP2Sl4Zm
zR1>Xc#-uHWI8fR+o%ed2DLwY;d4A1?;#~9f`kYJJKm!BG6AcQX_!t&QY~Es@fpUo~
zRNeflOHb8`dV&iSivp*j3eqo3#dBLe($l6jjs1GLctzk(wRmHDKww-UH3aSr#;Jc6
zV=%ARxAT0$WQeg7f0`TWXCbm7nyD#JLzZ8;vni_HrD;2yOB6QJf*MDlOf=HzP2Lpg
z#!X}9HLn~KP-1&&E<oLXl?(C<jRNPA)0{fGa)TcIl>5tK5hpH4BvI3JspU>Nr&ZsI
zud>srj~$vlqO4%5y7~SFSyUA0+@8@hPZPc}VM_U84A#;9e=~o{f@iglhMc-+XIjy=
zgNwRUm4Binx?rCORJ$+I>U-<AXcckzB*zZ*4lSiOXS9@_%xEd))v4K4cb!XyJyBZ+
z3*Cg60Y{v-<UVyC2qn)8>fk8w35{*&kPInxk(9z9O#S<7DkU?tmIuw8k+w?{$P<&>
z<`hyfj)#uFe?KM&Jr6+vTDA}(K^Qd)ft(rVT97Xp2_De*Mx^L-3oA8&Q0^&X7I6HU
zLByOKsN827TqB3$KOfhaM-I&TpasB`h(YW4vWkUVH$|=%NtV|_=g()+()IcYsco4`
z$N6$fa%xG_eM*T4r)ugUfTjvtVo++qxDZrEP+A9_e-!Pk6qRKJbr6=MDH6!vR;B1&
zI&dOWYsyLE=05(KEyB*z1<VQAf_#E?{Td|siAEa7%Q|J2^lQBUHPc$Qlr3QKZ7L2e
zmfzR0kZ?H_xbs-lyseT=fNrt;Oa`!Fdy48}*FxP^5g}*>wuLC06=xwR?^d=$^4DD!
z-r4D3f8lDd@EjKe#RV-(vfqq3O73%BW0V$W(kr1_X@hK{jahGmE%M@~V}Ql-P3It5
zH4E~rDZBm{BxO5rEJ0eQ^AIse=im`xpqh51b`4(ICU~4q6^J?+zP1UmMAeT$C(1-M
zbDCdye~`}*buCh#aj>;4`!qXCIEB4in!{Zhe_6Hz{aNFYz7EA{XkJ3>*hS6Z)=YW}
z_^PuQc&|^MG|j*K+qTJBN(OYrU0q$8@<w+!Vd-4VhBVjeNGTjSeAUau*McD#SRYc6
z*)T)yc#!C>>5dOJ&J3va#wdn1-w3r(H&W%mjDn_agfh$(-Kh|twaPLe6?JWrvd54y
zf7@X+EcWL2a<QI}Vky55wAL^=Q=gg%4Sa1%wv4jf<c!M3!O7&y=^P1t^%7S`+Rpi|
zo#UiA7ldtnTCH3(kn7XQzy~c1GG20e{SZBCpW>gX4p+^%#%Tz><Qmg>Jy$hLW9kh@
zncjMZqwI$1P&&0IX`6}`Go9sJRkl<qe`|>^Q2xGN_j-@VpUhF%>tW+O4^o98ecskW
z#gRVLhasvOv{;*{T=*q+q|(gXw<oO?z7a|vE0v7x%i-?I!~w0PJk0jAmGhvt)IxLZ
z>?7M(aH}qTZj)9+n=YR&i&N?D<YL<iofn*@+J}#3zpy#LNoK3#TZx?X_D2W{f27CM
zBN}A2JCP~DY)cRVh&W{j7Td7&5iAN}n4p4ii^n(H!?`&B*V}{iWi+auIN;wz+w^$O
zt_D%8I?C7v8C_%+(Yc%|y-j{VbUsqnQ8(-SQu?iXL`TIv)sBM!Rfv{$Rj43!Gm2ac
zii>FrXvthPj%+f%`Z!~8fpx=Ie|P7ux)WYg_PNoH?Qy-i(KT}rv};3dx4>-E0y9Gk
zEH%LQx54}|Rz+F|feg5;+h9Og4X!pyJhbEFVkp5>!gg(s`92sYrBcS)fNV6vxMP0X
zH6ZIns<19}jXJiX5hj+5ne7WY3L6yLYRKcyq}0_OgK&;1*NX9YdpsE^f8v-&CqXmR
zFrb57t%Huep>?3g<`^eGh7^uS4O+oMbL`}htG)6P%#cv(1^38H69cA&+UQDg3<s`(
z;yWn4t1IPuhg8-zSa#~8V+S5j!kFWyb$k@+eJ`~EUhkz!X+kgc6<i?*rWCeVeqH6t
zpHMJ8$fwBCdixM1QvB=WfAPcP3A8?b?=L1^%C8^d<?HwR47fGJ!GO(T?~5Xgk+Lso
z5$!@ri|D$NHjQ-hfo5MTP!-ZoFmdP9E}@<a6LF2%x#0jaSJfsOUeW4G80&dt6wF3_
zHgUVRK1&cyT|Hk@PYT#*OlilADMvN~czw+EJ}nT=sKeA4NYM{=R%IOd?DU_%fRw^A
zl&TZVM*T<`CPgIK0sSe;M`^v!>lMAXT9D}<H{`0#rER?{%|)UI2E+SC1p-p0;0C??
z0p*}F<Z^}St>`}cKbg-j5B~!iyxqc=u^c}Ymmzi`6PMW>Kgt34x1=3EJPQO9;4(j#
zq3%8v0x>a{AwM}K3o<WEWo~D5Xdp5-mr*}CDS!WOTe%JYy?=#({c;;nJy8@zX*d+<
zm*v)>y#`GOY%78xP2488Bn}+s+Wq(YP|taBW2fiDPO}4Io9~5^NQwNAlti)A?1Cfw
zs*98xRa|0jRCSW=T@AryJQ+o<Ff~eC&5f_%JA~L?f;}X`m)KV+kL;y7qFJK?Qd1+X
zKY!9jTH9=LRFi>g8#LmDghmoZkXRat#@aU{yeNHVgwD{pjYb%_N*gV*i=i}LG>%kz
zl3a{!^cuxQPCai1S(*$HbCF`x2j4MeF%okE85=L7GbOUoVg;GFy$voFZSq1{4*d`;
zjw!XyID?!qsN+aILj;){10-RtO-;-pOMhr2SCBE-Xc&ldZ9=BxSYzuU1(4c$5RkQR
zh6PB0`pz;)>6@ubq-`D&7CMC{x<r?9Y;T1@=BAq@mo)VA3kJ2eP#a$}5voh2nM%dt
zrJ9>GS)BxJq+};VZOR01nR^+F6Ol$jZBflO7)(hDwVBQdq_lA{^-}s}gC&ZoYJWYz
zFl%W8m*jG&jjSMJuTL%@OW#>_S^A|=Sg~@ONF|p`v)+n{gKalnWvA4e8O%y!ZlbF=
z<=W&e*{Q^SU<D~nvQnJp&?GArWN!|bBgGgtS|z8mwL(8STPrA_O=Fl!aU~Qo5-$qX
ze&7sW>(#`H6>0ui6}&n13RPFM^naZNS77=!8v3ar^^!#gFKLoNLzrdbg?qt#+WcZ_
zu{4brc86YihU%)>yGa(zx3Q*1L#jZA1Kiv_Bv|E-KX&i_dp7Exo}P}Uu6s6~9E>LK
z15hRGc7N?&cF*yEhP&O{(f;&3#pp%N#?%|Yz^%PGC#=h--ACQd#fPbN_kZf>^grFx
zq4WLe(fG96>HhxqrG0)loKDZq|J-$-j;4nfAN>CKr2FOK<m~am=-2M`-t_Rz+4SY<
zbTs+6w?Fc~9Csg%#~-?1$;aTSJD==#&n8C)pGN21r^(*gAvpf<iUU8!SN-FYCy(%V
zN4?;*UieBc7WLtQ6^jDJIDaft>?^UBYf1e>(=ti%W+PFmC&0Qu>j+bexp;WxZuiuo
z`<?Em@w;*N!u?<<4_ggDS2jGsszZu3g`kjqfm>r>Plh8Zm|_zOQ;oU4E`)MqxF<!0
z?*LPQbn6R&u;I;z061QNoJ`qkwIIWu@`4QTT>zoQ9zjH?Cd>LVB!A7YH5z1)0vojG
zwp<1Re1(h|pagWyl7$BuUc9$3%;@}W_m6-6%k^^%v_p5N7stoDH?hHs4Wldj0tr@N
zMR4(gpt1rhr|45zjg?@Cz_k@vu?U{SYOI`LS@Qz8?)mt5JlQ#e`?=P@b=!MWIOM5|
z{o`N0On=&$?oCII`hOQ&<I^by+2UjwQEH=X8Q90v54k0l6(K&fhuYGfVh>NAbla2h
z{?2InzT1AW)x8^ineNUR&Gz1>5d;DroKDYMnAX<E+tK;>VzNK#-${G-dUSBK_iX&7
znNvm}&>DBq!vt+DiIZNIW{2mq^s9;R=zX7nnJ#_)1#0)!_<xu&>xiUW#WTWaB!d4S
zElD21?e_Ti`SWN0>0&beaMUpJV7z~E0!{WBd-c*yMjuc1PVw{k-yg@5lf9`woSq!t
z4cw7L!VuwDaJY=ciY4Q@{lrvnFT5eTpF{MmCSZ3ZTr)_@gFzx)S>}mz9D^f->I3rA
zdobCXtn@2mzJImJ#<Xa%G2YQ+%bdwZT4JcTMOXkpMA`#%0KOraEnbkABK`46_*N+2
z7A&rYuyF~(V#r)p;Y76a%=-HQEnZ>)d`Ao5uPmTmf)<B?+JsdgE+IotNm{r?+2Sqn
zPXcJuMXMlb)eCZR<Ua|CZj7>lEGV;A6ey9|2m6f*Q-94?xk5PpKB*XLV~=!d5CWC1
z&!~8B65?B_ShHv5-@-gzl2VlFRaRn+8EXUq@7p=Pg2y>exwUhoxab@y+|4<X%sB_w
zJH>G(#1~}zsNj9Ey$vl588NRZ=C$gakCV~C7=_Nk#aZ{aoi{)4>>rL!_E5rHi;Ia?
z0@1J7W`7`Bif2>|_ptgm#l!VhUuend*SoF0%&&gRmw=dbpvnfI(aInwD`qAL?Gkf8
z^Rf6$vH%F@qh=U?LN!^#5<I#^;T^<B!2eKqBf(Uz-?8%EIyxTpncVq2-D>buAHT}H
z-qr#DzirXZ6wClU(*G0vqV1jJYmO@6x4?XkV}I~bC6BdRiB-V5Vy{SxrRXZGjFKee
z^!o`Bc?n3!9YG@N97xCj5;A~<l)IIGGaJSl_>!)+$D}48@&g(1^JiQ3X$C9`2c`8K
zGy#Gv!5sE9Df-0mfwX(IjlW*ExIXw$&7SPoWu{vU5rF<s@cR1AmVKIbM8F=3*TE-;
zm4C8?ya)h!$(GX+HnB|rS&7wlIfQi;R)Rw)s=E6{pIc*0kwuItyb8wDdL3hGn!}h}
zE|Db<)tU~~n$keUff}}=G!2xjiApH%?I7_Im4H&L4+TrFw|DF_Z`^AhN+d;gp63o*
ztYUqr)Zyn>`0JI6WXj1V4koZ-*$!)iZGU~M%fSj#)O&znm9mg`W)v$dS)+zZi6w;%
z?8g&hx{*0hB~P|ZNYKARCRkTm%zYEV(NzPgc%&yqi1G$1awOc*4oqh<a6%fK=td##
zWg<{x&4NZRghYUvozUDMgr)$DjwN#?wy(N@4H5=E+opb2AF6E<6j+8hMS)V{;(sQM
z64nBPE0gWf1lhp$NUFu#5$j9@vF&4E`wb#|%N|;|m;w1W^65E_t9-ipq;(ZmJA)0H
zzQrmeKHG%^APbAE!Wt`FGSDil#Qo;NDy+5@Tfg&0PP<w{vxjV#OG21EWEUg#iTIY<
z23Vgm<3GtBF55(O5HH(U7R7nirhh$Lwy|;IWp8%KI6K7I(&pKBsT10Llxr~@@ANJ)
zoM-aPVz>@<f(~_p4sl(FxL$_1UWT|{hPYm3poaeH&|e+;tF!)iHu<xg371qHC?2RZ
zP%=<?{mqV{Ft?HwnP80E_QW#6Ye@8cFvbp|vqij_5Xq8VSk+BJ+zZww%zuPkPP!CA
zxOhgneUlJTYte{5#!?`5ljhBZ-z-2pxDp`%2rw#Vph{Uh56221FcH8pic-#)xlP1h
z_7Uj^P%79F0jeZKUm&`VX~~g1BBd~eYOA6OK&u<Ln_#yzSZh0#&k3f0=L&GLq=<QO
zS02s<n}94i!!{FF*?<ssD}O^w2$(>w?H-I>&Jv^q%uA9X4ahy116BhybWz^W@dUyt
z%&{WxP{$~)GC4^V<ycyBd`A#5)huLJr4n)8L@pFOybL*zU3iKZoLVl9deG))aUxJ3
zW%z?~z%*od5UsT@3Z<Z|T-7!RoQOIPl>G0!*nZl+`i+Tf`9E(W|9`hxO-OeQaG<Tg
z@sJ~NwNul@q!%eKoCFRJJ0nSH6AL|TL<4RLU?+y^vmF*|@mS6!z7L0ozZVU5@e~oM
zluh7I#WD{<Fxj0bx={$&TCrHfa1TKW1P@$q7amq-+ZR{Y->Q!>JzS}IK0fX@sL3w(
zZNSn{D8Zjoy8_2CV;-g>21jj!87D1Ez_l_XxsjSHa*~j8Z)^VplcTdwm$4i_6#+7r
zf$BXImvap%D3>TSKcIhb4}W?4(+y7okJ4#_jAY2td1;WOLmO88@@mCWH&9F=OsfUw
zieyvtjwBbJ(r1znNhYNd)g6l&!Gs=(DpFo~0;CN63k=YtvDYB#opjzzs4IJABFij-
z(q&fqCC>1pkhyrODpAhgZ4ev6N}K8M`$WSlXEF_RX}+lI1Al**WN>Lz<q`FmN9flV
zzwY5KjC)#-?W{mCxT#R9TAX)4rsQYIMn<0ApBw8G*@N4r(9yuSh2ke!EfsvEvVS~v
zAKX4PqeT9jzR!9jABK^}GQ!;Fl`7M8`CGJpS!L7grz!^&kHuG`O{Vdf?2U1im&9*6
z1lewdQ}SGtSf+otddyO*1^=M;7=2NTN#mRd;MD7uA}i`=sxBe2RHPkjY&-cn`hlZb
zc{IuQ+Sxu)D_Zw5q_`tq%V<!X2`Ia^llYWuErNYx3Q+baX=*?Qq9ztycXX60$8(Yp
zZ5KNsZbI5)v>~P4IFm|SP4C)q*nX#Exc=qa*Ru&-bIpIi;uT5n4npitJEV<29=!;{
z#~y!*bbWG{@*#SbY`{;;WKN41`gGh&mbj<fm1+BI5#{lblp{6nk|Jr^YxUa|xq?=y
zv-zOaHs^%I?XgB+knpMelJh&y`lwmfCjP6IH3Qcy-|bH`YTtORqo`VKo|Bi`T<^-^
zImV50$RvNb;+@0O!tXk>@L?RiAkT-G5zlG23AH)qG-zxi8n|Oz!=@B6G`2|Qwe>pY
z)mm<vIU}v=R~v?YHCjfHBdq@N-`liAFP9J&a>n&bAT2R!y~Hru-u;qqnG0^noE5U$
zr~tq`74kCUT*oE<%1dBr2{4S3S8Iqb-?LnHDVKi)CcIn<NP;96&r8gbiPTJsfy-|c
z&c~u5XXtt-6*KQJLd^u_w2_gxJ%V3KC<3Z(j0T^JDkP<sEqSQ=NHM`_r9>vAZjqZv
zyQ}2=WBWG*zWnp{F#05CfRdH~(-=|6mw)6X>u{IEgxj9WZ*%~kvA<ykuwV84ST>3Y
zO^1KUQnnqJDa|Iy&P9f%L(2@3orw&snbb2QacnpvY42TYH!GO}-LLx3vE8g=0B*;p
zOD@upXdrto$EK$INO@~m^2ggdFy@OmWQLTBGz^CL!|k2WI=-3w?Z;(1t=})xPIJ>v
zD`$y6=~2vO+jv^Y`*dFVQ=uYE)uC5Rh{}I7)y0Gq3xdD+JSDcT1wCR$spX8%_1kNQ
zduODa&$h>wLO%8I<w0W6HqdqrHasP{REu>nSs-X(E26BtdjwwVmL!$!v76UGv<UfB
zWa%o$##X1j3NhO(dnw3gdb=$n0tsD!-%b^OE12+t$xVjQi$Nk&A}GyLZCvrPb{T(Z
zrmY}!vSF?O?2_WCP{bo@(Q;+3sW`{`wEi+x$AwDe6s^$dFj|Bs><SfAwHUF|iPfn8
zCV7K|pJIb1BW4&PC-^_bXt0yLxGF+znKZ@6UFC>HJ6HLH&!_bL@woYtc@StqghfMW
zwA1V_+avsHx!G!_N7hHpCg8}7oh5(!ShESxF$)vbi@Hrw+f%8935h6cfoe#kmz7G!
z3Ue9_)#X}CNm?rX4lUzZAsbnvAG%6$&k7Vxg!N$Ln(78lN1!L*-@XptUfyp0JtZND
zNQ&C+uRm`54=4ozaXLyR9p`TQ{cqd%n?H|KlY}0sZ|RTX(SF^?x72NQoD_fITB{B^
zkPS#XDhCKi+NKTh+_)|s1|T}qR1V=&Wt>O8HC<D};##VS%XB0T#t$x^WhO8t2x$z|
zE(~mPm=`er337KMjogHKnW0K(-~@c9oj^EQOj?%C^XYp2Kxm3eih-gKYyJd_sJyz~
zE=2Jq$#y=N&PEbf%yD-`B<+9tIY&)TQSEHmE9_b#sHeQ+S7DIq1u@S48Sy5-@9u%o
zp1X6n4Ps5^i3mCOz!>Gz`X_eu+L6{w$)%Dol5397hdZj$K9*#O!rYjz3AE&xYlMx(
zw50mUxtfw?miB!6sUA!p;qQw+DHr@@1F*3<>KlOb8T|eAg29=8;8A}X{d-OR&yhh1
zMxPflT;5ddWmq45ABPX+1Qu<G|77$atYDz+z_ebNn0z@E!5`1%%&I7}yO(fN>44p6
zUIUAZy3#_eh2{+2)dX=bx<-p&IR7$TTdiB;jYOs~cc)`4+h)nk`pgKW=eYs`fLV%X
z-(cZoOMz*swbSC3EY^P%hxK#`@pE+r_{&{xF6E6{UdB7k94koFFxKCTd%?E0%Z^C&
zV%7Lr=pHmB6ro2e$TRxEsk!e8Y40*r51MKTH*(8b`#Vmn+)&*mi?^0heiWN6o#u>P
z4Y}wgG@Cp6qP_D@Hr73@Zo{N|@8rj}V%JM~`}^|4hZk~0D#w3>f*IbXoOrcwB0lkR
z6GqvsJvk6yvZDmC^JCgOu)fQ3SS1r^Cb>tNct#SW3{Hpqw5scMOlc?|f1&@9@AE|0
z&C&I~F7NT;-#+C{U7MYxe8wM_ra&nf)Jhkw0~2Sn903Q}VAxg*quM^AY<hd~m+`rj
zFnaE}6u6(L>+*lCP=Tp<IgTYN;|1Ah(5AQ5aa?Hvk4lUbeCAP3U2B=Ep6<vNC^y{s
zNs*3A?P4|gt9??!ep>EF|FD#ttof62KUy8`BL{BTRw2h&H+GkeT4&i+ry00pQejud
zc3oQ2-7wfi_q?@wc>-a|3>}b${mgLv(?4d;JBRy$V3U8H6zAr63sAACj2&0Kx3wlD
zXBy1dTGN~v;`{$-tqDuK<YzWS28ubtuW5iFom)7`w$}hrhH2WY!|V?<wuGS(!gT}1
zolZw?z>NIgg*2{j%A~g1C@bzX;VeUI#@Qd}z6kADeXQWK;Ff>Zb$Bf5GrE5Q{kDV8
z4EBNIeB6J%xBE9x9)kb6a83c*QyjWkb+%;=Ce7Xvn!-vZjGOr4`!#WbX=^9hUe>|R
zITOx|v&+FD-nZdWV{??jtYrYfP){~cZOilej{6vO;@uW4>0|JbI!+&h?jJ}Wl7J1b
z{mAFM|8ozPe49Izo!>s=diJCFDQKSLBp)O>{qukMQ|?-w;OEK?;@cf4=J1k!ec6uP
z!vgsVS51!u^u4pIA8;sKpe-2}pl+QXcOjZ1xGQk9#1$orZ(;(#NMSGc0c9-ZHba#>
zW^nh!*PB5|2q>+a93#fGC3Sx6<jmt5-G6OpyQ?{7+sAIl4s?1ajQKyHj!JL}Wo~41
zbaI!t96t^LGnY|K1{0T9K0m_(?02^vKR?k11M1k1m(j8?69X|bFqc7sAQhJ!Mn5!v
zB8sFQK(N4ivo6qEgS{0!6-`b#H2?ol>M<{A>>1m$adVKzmNk5d<VSHvci%eQ{dPFM
z9X_6rI;T8?Gu_+Q4&)ClpP*&Za_0b)Md;qXcYl7;+E^J!8KgC$3@6{cT>0q(f|nH_
zD?nxeo<2ANnL(LBSz}nPT85GDuFRc(%KcXE_nIhD5ZWLMLDs|Z>94on54s4fRj~1E
zywty5-wyvCYVj><Wwh!3{(gvm5Q0K)Itt1Ic0c~<z8?OlsSdmtyYb`4x2~G`QVi%p
z45+PTo*59dSd}_!gTnY%Hefc53{e_FrJR;q^9Sq9a5f_hKc0fV$c>yLbUinJ`m8kz
zClWK$iwrTXff?7c@{^X0=bEDIiDubA5<RF;##fn>4I`Z?Z!laVH(Eu`Z?e+TW_>)d
zUvy(*l-Hr-;FWcurypIAINK*_Auh@;+lWJ+W1IM{1&EkejdX3f*TryfaR;am42EpE
zc<uKiefN1^cnBe0m2Oy><ebo4dM-O)*o&KMMKV10yU@&j8TP-?&!ACSc4V&Kst$SI
zc(8+9%{h&pVs@A)HFwKpv!l4Bmw=v$1)q!BS7rxtteBzB4gr;=u$CRxgEeJ`%kj$W
zm0b1<*EzxUmnld;9e>{D9rI|^$K~9kPgiSToKb8{t-<F-4UDtOIae*#$LqOzP_iF2
zA-BxcvyxHU4X@`azU)HnXZp2DsoM#nHVQRvQv<b?_^OSUw*HtQ;|2#rBf*P1Y9!Eu
z=r?L4VaB269<|XgaTIOt2XYpvtT^e8nPE<`4W^sHn++>#aet^01@f#mSHK0fvIvEU
z4*Xp1ivJv&D7{WvRuD@8Am#KJ_k*+>VE1m`TEdZdfj5p8Nmo5dk$#MmJ9twt%6ZrF
zp2uTf{7edqw1fBamjvj|SmNQp-;Q3w#pb;odXGtSc2gmExm+W%${3!WXOd?&MOZG}
zTvNC`eHf~irhj;IjTe>Giv?UN4`n644CP$!1$MElJ6oz*tsFPH4RV**X{aq*Dnf>*
zhT5$)Rw~#8Kp$6Qk-;R%eyOyg_>NUp#c8F?s#d>Coz>eueH*u=pINB0YN;-}Wm`k!
zf}^6cBh-=x70B=DA9IR3LM;Y~5g&dLs6{!XX>2n-Z+}02-^JbrIl{t2&=#dY1TYUn
zTO_A|@xf?|0BY=g#ReeHx+BTw{N+9ivJwF+Qu=x2CZ`NgbyJFq)4-db43rH4T?`N%
zV3e-UQ2zSP425Fmv7k5=XB(X?ak&*(jW;TYjuo(iRRQf!0aoLKf&e=v&WhqY0&B%-
z1z4-q|9_JPMy<o;4o0>5juu8u>2eRFw%zSbjGFKLU5ss-^lgmV-*@R_?5I>}WNcKb
zbTaDg8Cn_QP_IjwX=BITu;im)rU$JnU}p_j49fXogjcD9@^NaGzsKJDHqiSCU3G}h
zApmRnw^nZ>M&-oQ>Xmxx3#x7Et%H%&Oto;g0)G^MlOW>;pfo9~{m(%t!GFVhcF}5z
z_Id;&2{>wGGkO_;*eNSMRjqyxWDG2amWI<AWE9?#He6$mRR>)T`(<=Y&U_0%_JC2T
zat0tJCZ%4hSKk1T1FTeh>Gqr+9a%!LmhE*pr$bO^pkAE}U$RIx_Vt?eXi>>=4qHfh
zYJZ$#iY<CLcMh)vyj(Ap6m#Bhby<wX95zy!)BF&IebrCdJ#VlU8Bk;zPEfKq;G%ql
zBq7z9c=}x7`i4P_QxnDqf_icYIcCGB`X#={F^R`~K*`6L6sx7;O)Z2W_v8n-7w~E;
z>i4XvJQFUJRm=+UZIE979bc4)=KyqeM}K6fDDK$w0qYcj`$d{Qz*Z6ahdf^aVnrc5
zq~ik;zDy5m_<%vhexcd5vG%>c+2frGlUHn}%@a>m!zBq(njrb*n79f*6;Gh-&nY7P
zPy(DSs$CFpss*5xlY}S1N>kkwVM&<xPk3xEJJEhG)`+>ZaIDO=F_hOcYCZwXl7DLN
zgxgL5#w+7$!|)Up-C&m5Nc~p(!uq9D9)9WwDZ%za^?>sNcr-&W_A94dv`o2itF0ww
zd?C@A=weGiH@aOiQu7+5Rg!l{*V=rT9iAp;|IZ0wJfIdoPxOZ4r!5A=GR#=IC3kRb
zo-S88hy4f7-DdR)Wo~41baG{3ZI`<oKMR+_7Yq{vGB7unAqxQ}5Zzi!kK;BBzR$1l
zA6TtOQI`}=KQw<TnyhddEKYB~Prx@Me}d#eqbCa>jDqRy&*|r9DbetC<cp+KiZ9E*
z9e?Y?Ck2jY0M7s(1vq@N1U#a6MDZNd{4M!H{%!NvYkuDF^G*^43QQY#BJg}T9e#QH
z@kwTerSK;F3}5}fuWwI(J@xEss|86r{rcxqxIqepa>#$TNf_y-zki#4fBLB>TNji3
zO#0Ar`umTm8~K(s=p}1V8l&|>gCJS0G#TTC3b&F5>5G(_B_@`fPJFSn2n#7czdBf@
zNRszL4-|vOb)vw+PV+}(c`?E)S87@b<y}t=<${+w+ko(_$Zoa*mWKCJ!z@w!?UH2r
zcD|UJ06%}M1hC2%@=lV5K_|R6*?eKrQQ=dE-!|};=VhM5m}J}L;W#%wn|DgS-Y$oC
zLXsck__8X8k|>l^{cL%>d0i!q(0bG#b)IwsMbN<*2j0G}OM*(U+x4@f)+uQ{Y5I1_
zh=ij-GHKWOATL1Tzz6TJJmu^4@X5#+qe@&l-OhiNQv%dhdsWV|*A4cHjSU8yi(2y8
zK3{ifi%<*WhLu3>wgW;hyfUL@1!+UfUX_>#h4F(Q?C>b|bAfV*STRprSl=k&?VyKb
z1DOY{ruifCuo!8hW{K+B@jjXw2W5_TVgHWFmb6O$*Y=jm-kLNP_@86xA(-TPI89S=
zV>o{+OOs6`UAF&Y(wgkE>-B_pHQOm8$ht>nyCuajI`7{@_7Dxe`XGg9fSa;f-m#IF
z98dI1ek$8T7xi!BR&yL<{{a%(G~qlm8yV0^)scy84{`ZmX72u_rDC{*D{HG+A@N&g
zBdex<_uATX_#ic|U5q)pna=-Bx}xmKmF|CD-!TAoGzXiy{zf0v<`RUhPH-89Yhov8
zV0Z&fW_=e3P85FG%H!cvG|7v6mM64pG#E_&CA$}D$H19GGG9&=?G2;fszVR5p!6KR
zeJ{&aXQOqQm!V``i|7JyQ2olqDQ_(*h&T@s_C<wfCt<wzJ!%S#Bv3st$Sw)K1FC-q
z>x9RCp)()}0;ICO7jxABE@Ji^jZ=lx<oo;coDO&$9=ZifiVw_l<k>(q{YUZ~FTlu3
ze@LPu#)zcMha@^;L`l|uSfV4cjE3CG`kNBnrX;7`YLY_~E^xpJ!c*0NDyb-AW{|}K
zLA`u!EHxxAa@j)SPPt~MY#!G&6<>c#wu&JKn(B$6Z>3MXhE#17$_T!AeXMC;GyU}h
z2%6;t>gjKcwr9EaR#+OS#+`E6YE+|x1WAfC0V+spJedlVNTWrI5eh8s1aI{9pUZPj
zbKM%BJvDVveYR4xg^w8R8clAabA7*4ACgc$r@m#p^OB$N^OGT0C}{%C;!%Gyx2AC1
z@*g=(@G9)iU_&8_Nw>^hppIpqT?JgH2BM`%Qto74<GQfXz6%5_Hk+Q)2Q2j;m56P+
zigX5iW&Iq};|y7o^))Z%Oe3^43oqVc84vYft{I)L1&h*$*jYX=e+A4`b+(o7C2V$i
zC-uR^UbbA@`)v91E~}Xjc<z6t{tnPwg#ypnL2n!JLCH#IZ3|=j;qvBgLh=!7bIIN@
zx);f#)tn7okL1a7oL1ij$t$kvjfV%~$eyCl2>ICH-y3jy?)!ICaeo&PERTTcDc;ea
zF11BR@l1ajoMntRXukEVJM@Ks!A|;3fZf6&Sb;DI2loKZY<leUR^gX8Qa>qw=pe$c
z$GKWZ@g2+NZlx=Ab2s~Xc)p%^mqm8%Cht8Pb6$))a^`3ZJ=>`tW=C#L`FPLoVuLe@
zPzVDbbn#%To1NMWSVHY-Efa0*@*x?a&6s{)GC*wrV+N=R?6m-O1}~<T#Agc<%Zbja
zeqH#|Yxg4PLTT6ysi%^xzXB_NcL4L;7P1Gd#qChcX6U{*tMfOY>9acWU9Z1OR!1bE
z%p<dUN3n+Gm%K%b9>M-Fvs~rey8t<ZcC+8(T>#X=;;_=WcLBorq;-S#`*#60^;6dR
ziuGJ>Ppm1?qZ8%K>j1%6RX@3M)P}ts=jh?h4{ZkOXU@^J6$S<a+_9{GF#;I7_sf3W
z|6+n=OoYRRpNrpY<ZeZ&r@Ie>5fpQVA9WbK(VlNT4BnH$LLdR%cg|4Zy`3{oxzcpx
z?{1v2J=&gV?Ss9hK6mE5^gw0Sb93m1N<HW5UeDkPWPP^jcBPTFQ_yx-$+jJXrZ3|g
zgE}>xD^n@vZd~A{?cCIV`&T<ab2@JLyj-xCS?Br^e1mIi8Dfny{B1n@NOGvPJMEhu
zj^ev3ojLh~($7sZ=%wy+!a+8ljGlNqAfFn3nU7Aoc%YRLR%5qF&Mkvshv4OBsE=*`
z@Q{F+?mDQ;F52lrf3BP^m_d_e{>5UeP`1T>kEK{kMb``y&)IN)=A*Hjv2z%?Q)ka!
z3eekVyL-W1cy-eed|=B4ND1#BSx5F1y|m%w+Pm5|Y_5#OJIZ&pZP+C(^c06a#0?a4
zwc1gX>xc+^cn`z%vme#AK@eoN@=}0izi-9C%nC2JqY6+;NTrun@jvg;003h|`$*qf
zQhdiAO}EmO9!)nk`(N9m>7}~dqv@sUiWSgz)lNMxbg6dgCO2<`_^xS_g6%zx8>{(i
zj!ENrM)~HH_77H_@Lrd(96uG8VW>hAm&U_9DSzEt%Z}?d65aP#=m#z@i!8o?h5_5%
z9U!|2dN-KWV3Ji9lmA~#@hLteTb9)~$t=`%OH`5d<f$r}Prprk`swBJ`SSfk;$)ar
zV8rzK>jd<G(_i3pQ>)1UFfC#F{C)c8E9XMfb)gG6my#|+-z|Nr(>Dp0MgWZfx(jgn
zW`7Xqj-)%1#t229oGwA%O}h7y-q-ZL<%pgF(FU0)$<x0+e}3U<!JIkmzxt2<zxU6V
z|GxCg*OD{A#q{srU;G1<z@!xuI%YvmfBa|q_2nNu)iM{SSK=SVr$2t1x;wul0Xj|s
z<XS7$2oN|)6(_A7lm1Z@Aii;~h+!ng@qbDeF)N^9^81H{MT*Et6>6a9xFJ>ei*r(q
zqKjW6MTUN@`A=Y>xkD@z>W1!s?x_pjDc})`mN_x37u0@=)-aTVO!-?wv5a<#Ih4a_
zk`Iz;DKbN)pOTpniWxEL7<O|ggk#Rw17#+pWoXSnkuS<jpk~euR^a92tyRHu@_#ad
zDIt!Om(k1^b4Olfv<1c~nDrKsIZ<+&|Acv)JGCWd?^c#clGosir*E1=wze}}HcBsi
zs^t6Hj}q~{5Ea4*Z<{8`+uJDQQR<xT`lpwC7qC3%iGFw<-$pa@_dsB~8`Y#{NOpQ)
zRBI_p$lT9Vv-~`DrFKNP6pz}O0DqRFhm{PB>qiIO*`(8J^o$_ytXGlNEvtJN$AtST
zc@WR~Ie9a|Bs~7{xKmljMvcI7OcYKs<A%{;HP^U$Z;O8G-LVuYBvWRTFfOh|3fiM(
zKbWRXuHzRH{}Z!ojd$YR?7~Ys`S%ZH7cq_y%yKs==Yp!U>G9c^izB*hc7L;hN3|1f
zL63XxDP8HX!h3j2QHaufUW+;fOtfxt9(Nf(0y6?rkFGzyWmA}sRJ9D7L<}HEVCf<%
z2Q+!s+gz7Xw%tPToU|pBB{W86M3<F0=ni*(DdA&~oAme;+#t8R;ln%rR(B3AdV|e2
z&Mhk#r6a~fztHTbj{;konSUYbKM~j}MRCWu1Gj*l;yHtj$3wl6v@P47qStu5mP5YQ
z47BLq+6AVOX#+ha!EbIg-DnT8<on}uJ2Egk@{Tt_axZ)S_&&rr17`+GPk=j)4AmgT
zXTUHB?+d+x-<S7F0Sxs~YRSlcn_7fKCZsq*Ehre};RDYr_&@<gsYs((CMrg`JNhf5
zH8RTgk6<~a85`A{l7fSHj%aKn%L$Dgz_PQ90MUWYC%Isf%e&6^c+RBIGL3VT&v@>B
zssw+>R7nX+QOTDnSwA3u5;!IW4C_!UX;Mt8y3ti0biSMa+|D_W5QDWNc8=U-44-|h
zWDMT>g1%tPAyTfhjM>Rnm^HGW7v6g5-prT;*W`xSQ;F7JvLbr|ouaw7Gx7x?k`V=a
z)OGZT|L=w5s076|*+X)l<vR64AUUBG66}#xnT}!!%jcvb+lY21Z<y(3S%t~5`rD2E
zg7BB?y4_4G;V&b*Uiz4lXR{=$brWnuUNo2PSw9wkq)*C}9;ln;r6@y*{a}DQruA7J
zxMAA&{m{r~cVbZZTA7QC+%5=phc3whq=T&j^m!oaQG$NA$k*<|H!5;nXliDqrgVut
zzF_FSKbRb{+51yH3RA_&XXUda7lU6b+lY<6l&7wsL*W0CN}ShQ-NmiDr=YyW&J8nP
z>${$RzM&3#Wa2sVTi4~11+4CpN-<41xud@^+^z4L5UsJLn(=a=_)%7$+hjb6+RjzO
zI3nvi-Z2S{WF&_PoBOBmnyny@3<1t9Qwn!zuZG*T&$5?eLv==!ao)Zvndq-oCFy~*
zWCj$htfo)xxJX{H^rZ3*L`gQ`2Agmvsa~spTJX8(3bK`mAS;lURHO2nH-0R7Hinu_
zKZ$R+TZ1WHi_7R039NaNgAs#ZlzTTAD&<;HW)TC{US8!~32O!Y6fBtS)Ra3q$y}x@
zOC$S;iJWuqJ;E~*qKGTHjBCglVX0o5+IpHbVDiN@zf7ywgWiW4`n@bOQ4BU!<KZ=b
zZ<)tu8&B^nddnOy_hL~=pCF;fqLk8TbUy@(O2-*IlJ(9Lgm3e_`A)qHC3xCRa$9Me
zaBo7S4Ph%}RgG42<zVhr?Z71h3=`1vaHm9_v1a>H+Q;UdgP;>_+miAu(9q}Oe&u`W
z%BiN^QC+JmXF;~3ca*j=?C&T&b1)@;)x3L7)YFyfj^bL8HCCZ}_3W)H$13$UwdBm^
zd9~yUb<grjku_DJ58^pisc-9V9;_&Ytg$@!jJDyx*ug&2Qe9bT@tJn?Gd|N&T1%|6
zKXknBR8IDVj^`DF<dzw(58%1P)0WZ`d0ugro7&suG~F8CZad8}dPz>R$rOu!)2yiF
z{#Lz;53-}tr$D*nXeuR%>HvU%nh85M8*Ir3AMIkSB8H52q+(L27Bc*ww>Bh5GO{D<
zjUB}=Y;ANi-Dz!fqyMt4jh?I9t&N_muEq!Isgh&A7ImxS*bS~uft{;f$Be|EIq&5U
zjRHTYcD3nLmr$0%$Wb$XlMecSe3Mt9<68A*G8LZ_JJIK$4(d#O-O2L1eb!!Q#BoHb
z>7*56|15JE-sI`?ZKbN<^0BL22@2FzTBYZTmr*$rkN+jfgi4#6<yc9<F4O?RlOoj|
zgzJuxDH<O%OdE}wH=11Qz$B^AJ(HC#psIi!bruekw`<d{#s5|#tRn?~sZwu9|N9{8
ze~01EQHH{5wziMNLkV2*9Nf0^ZbI=;Yv>kJH5AWt$tIZVr4&W3a}i}$bzK-)D5yLt
z#1emyU=a>NlUtmkl$9C-Jtr}xfxNQxs>;~c^rQ*#Z$WoKFW0g~4VEVyqwT-`g-%mB
zZBBZc*QN}3h}9t>c(Q?iejSHd-FA~h!s|x%D+#bFd2x)Fhw&~nX$zsBaqH<**|hnF
zdhZ&aQaZ}l28n;=DO>FehjuTY2k93TVFW$!4F}qC_9*~{_-FA6j<B4t{VJNBDEE@C
zt+)HVg_6wUz1;hjKb^N%GIyeZz+5js!RribYFOZhiu~QB%T}3xBUED>I1x$0HX9>8
zrE@)>SgjY|V>hQtCL5XOJE}rJqU!&|6Dhdd#afE+-X5?z;qp-@9c3HJ$gAy#%)N(&
zUjQ^%N)MYV*7Vb+mn^O6`6Q%NEZ)?=60psgoaHWv8sgyUZU+{czH_0~t`6s)Rf*iy
zoKH2?s1LcktKTSpokSBCXTr+!b04*laGoh6)ikfpNqXv#igJP~k8mx#@R%1D3IVf}
z<9p?ow$}){5;d%T7jnrbx>6#0DQgt9mQv3&K9P-_3Z5v%jUuPwzSm}MlX|t08ZSPP
z<+-SHP5a$^VsywO%Gs@Am2Kaz!s5QY*x>Qw_ex())zbEVt3D=DhEO&K%cCM?tUj!@
z%2rT_Lk7n5e6ExnKu8A45n7~Ak*n|*-q%tOkiQg!0!25|V|#dYv_w=L+F&suUq9u<
zIXF{SS-ZErXHn3n<v&Ce%}m?jsjiylYW{hc%y#6;>9po?dw6XVa@W3yr22&xWcTQ=
z{*g(6ee+v?+EBFo5zTKWn1Ungq8-7FIP;;7Q7m+}6RkFk;!lj)!C!I^Gq$_iiPn1K
zskzdza937KM=nbpHdP_NfH#TAEShb+NpuYOB5!wNRUN1zul(BZ_8zIwOV6ALBw(hL
z`+}4yAF){Mdu#z-b9bdKh$!2h9qcx)$3);5rO1;PA!n_6dbu_8wnbALP9pZ76#M=M
z28;`rp)3Lv12Zu&ml1po69P3cm*MgkD3{=1KbQf*mw917sDIe;XG`iqYld8mR6tsN
z;7*n%(@9V2jkIn^(ImZEGEN-ISacj6VCanLFb)f4nL<5c-OXW0d!Do?OJde4-;eez
z-WsO!x(ObvA1d>2mmZrywSu;^^UhiRGcWG!p;C@1CnoMT^umv`!nb}eg;?Js0H%am
zu`qu0Ca)zh<A2;ZTYMx*y=3jC|E*Sg`MM;?8+Mr_6Gph1kwxNdwH)3SO6OOb^*M}6
zwof|ad*1UnljRVi9QQ3jKCXvJq%@9cKJhr%I?s`hL~m=i^Gapfy~4?TvH_L}F=5Oc
zA?tRSJ=4gfFcT)-`kq0PmPQ4hTOW{)15WZ071Hmfet&BvqrtzV|1L{{?(J8W1k&)(
z6X=Sn*@Dx_bS<6APD-UDnIc=5GONI=(ynpJ>0d(*DdL)o#NK`RAOwvkTxP@i-Yzua
z<yRkks74VYzN!f6@$*f_Jt#$<)2szd!&uSI&F++F7bdBuS4X9N2o<<!vqLEOAdERo
zKpmMV8h<_R#o$Q2AnsPk)wM?<iOb(3!^kxQZYIL8A`h&v*-IYhA+vZ&8qd{JG008J
z9%Elpil8so8bIA)i4=}p*+dv5KN~gA2jJ#ykkk;82c3bc+_AOOrZ`SF#3pwi_AoEn
z1d5ya-l(LxVZVjCh$7k5Dy;c+o9P885O;*e34fbSdQ;7#l$$na+qrW|$qkgSIK?I2
z9!QjLjY-x`2%MxljuNf=RQj!QFVb<Ga9YKTfsy9D!s(N+LK)ToQ{iY2qrT>Do<z)4
zO}pn~DPs)!+Y{Qbs}a0t8$9c<V<aY18hun|ZHeD>yVZs9Q6O){Wx$EHnTa7T>x|dU
z7=H{TGQr*Y3gzfG%UW)^g|vd??h-dTD0)|9v?}J*#k6s4w0J8g=GWV?t&c6p2n|pJ
z4?prA-Hu_kqS#1`5CZ>{8CM<@zCv!eurzUMG$2jFk)s;Cl-|tU$^6M;{`SOXzV8l?
zl+dKy<~EN=yR!4V%`XL%(SUh`U$@M+?0?Mnu=aEYfVP_HJv%^a<7XLU)b+BNvTuT7
z)&e$DL!E)QlN1}vjD(GteLE8)T~%tFeMlxI$<Dz9IcMxndqe5JejWb(_<Z@_v^+p>
zn(*|WKVN(YD8Upt2g+6$YIFMQ_tPIQ{~W<4Q_LBsIq?H|^etV>Z|SVhaasx28h^1{
zi6HkwSp%XDI5USo(%xL_PRlU|4_u!i0!gaE3F6LE8sh7zeuU+8E$t~AxkW)bF<_E>
z{oK$ArmY=E?>LitlrAY9b7sOM#QHg%0w$H5R&keKq%D_TThK5~kpu!VkTnDB{CI>m
z>RFyiw19?h<c)C1988$Dc*8jp&VNjF?G9+Rq8ZOSdX{ZYNJYwn1>=usF9_i@Bg?yx
z3_%wQozF)JSp5jR=~`OeWL>041?_j7gT88Bg4QMb4&N%@nlu3#-vMo60FFR$zn$l=
zpE=dIOnh{^`4|t#-)VIM@9$*E0QpCzawA^*5-xuhrQHUuKCbx^9!=gt+(R>k248<d
zN#bm93+Fy3Sn;p<pL>@`ASDyFY~<aun3xylgx?j>EeDz@)3r^e7cW7!bK5<p^jhY>
zcbRflv~&Ton*V;!5h(#DR!<x(D_;7{x|oKK{pp&fP*v#>Lss8M!x-hrW|AI_VT@*u
zko=4&Mmfi9wZR8OF$2VMhq*lN`A~l`<?IM?!}Dc7G_GjZ=11>#TkO;MZIt|I6V#8#
zLmi|Ll~0Tx96%lI$QrqEv>Xy-etnz{i;SMVboi^$iRjh@KS6@>GAQIu5){SuBy_iE
zP)#=V3_X-N3ci~jFP%8kv*TPd#6-)Q$=K;4l%@!HB_e;HP4ak0wQ7L#EM9-_Ir80V
zzPU~FvRx>5g2&x&Nlh_@7}xgP5&OP_Vkc?mCXCM-bJ}_Oo{YH>%qa15j5!EO6^^v!
z9YDN1Zo<7v!ARLwzL_m}H~0w~D>MP%W&ji_W&GKTTE}jHX(L7^@~4v$rD&I}=+N_;
zP|igoIayGvheGsFN+n%#O|pNwCgEZck+*tZN+jYzEqQCkrM$zoJtBKgUZt~y@koIk
zARZ8W3=r=GALELiydRu{kr;ezgyAViy?iN$me!D}R<HM-wjGmzT~eLddsDRrsU4ek
zn|{LqsgOjU*UN9LJjf@WBc?+ycF+CY-ahA~6g8TWuxFi<P!#StIc0wuoQxi~KgB!w
zLq2(^viAU2jyRt189mS~Pi1Jj(8J(oJXLZ!J)odxH#`+-DyWH}&wDCKd9(Z*dMZi6
ztr}0IzRgn!%9h|zPc=Y1z*7wn?|7<lMTZ%@XzZ50nBjhQJ$f^kQ{0qvcYf}Gt<{Qw
zJ@N!(gt%sy9#>L<l3{;6-~HPtsi1fo>sfILCzqtKONt+s6gQY5=D-wFl>Z0B6@qIL
z5o}9Y)(clC?}hd@T!EX4R_40$Lkd`YZfZwLe{zMlG-M=38gfNOTeMV6!!;E>foTtY
zBiHg>_;tQ-3Wv$4Z;mQ^k?Pq>&r&cv{~p3-4UD9Hka}$&{M&yR;k^KCbSTzK@2j3C
zvpsEa*`&7A6x$Mmj#=+%!An#B7@5>;j^L{$0uETk+Z00Ax;W2rG$rspPPoAvNYztt
zo2981KG(y+KJn#57nz_APpEPQ)t09(d98M{Y^d}O$9dnzBk2S))HLzvC#t`)K#_&%
zX=swLyz-@%M-_i;rm3_ne60$a>JvjxnP2a&N}X@}MnryHBx0TFe$n=IzoWd9Mdc13
zt1UmjLQu4zg0iUu%kQ(G9#{gsV%mApfMXR>g0kZ_I*q)Pg(hgeEHq)>^TrhFoo^0!
zWwFYociB3N-;}0qazU4x#KY_{WnTVWqkc2L!FG^xGX)-cl$vevBEp2~%LZjWV6Qj0
zYnT1$oAUPm1K#ChoR_g2KNOc>=sXkxGclJD4M8WD{%Sv;0hza+Yd>`ce_ynpT!S$%
zCB=d_ueW2&m#{mTtt8`qDYiG<6vyaF`6HPPcEJR-!R#427s>m5pI0rI0{hJ8#D2BM
zgGwnkr^M|-tKKSScrEZfW)<B}i*VIf85XqaSAPdrh0bReRZ@ATAuj0{xOq}*Hgqj?
zTrU<hVcBc6LNp?&`p(w{f6Q|Rs~+egJB*iIWC#4qqv?IbQZ(H&>*@>ddnk6287cQt
z>@0K6T}$!4#<J&9E_H2V89$&{N^)rf<DryaN{z%}nZ+!Mq<dEFYbVNhCX{aMYgK?N
zn8D^D9_m$TMt-|cqNSC<^+-{2WIh*~X=e&8!Ll%|#)_a7;mB7jfB7yBN7QADk>{oh
zkC&?3=~P>de1(&*S8(*w^-2RdplH3=NtDQRXGrjP$CyUX#FKS<LVT76&gRheK`hC1
zm@k)PI^Y9JG9AW^l1#t;J9H0no^Fu#(N*m2fUf^<_aLPZJNJQU!u#`=V%2?gQ5-su
zyZ2}x#MZ`MCGKk=f6PsZ9nQ9Wa999W@wz|w+it(c<KlHY?7E=>l5Z9|dRGid2|`=w
zdn8?37J}_Du1UHPQ40Fy-yz!03I9&2NLxFyEILqK*$-bi*WR}HrEYXmHog*H?YzS4
z_850n0Yyp!PSjyQwg73Uu6E+>PVB3oTcxxYaK=jeI{4|Ze>9gydVGU?MB@_e&BR{C
zD-y4GiX$7etBFZ?!Hlx~s^$Jow*4cii^7P#-6AH17#_Q1PIC{4d|E%Hd5x+G=?D%h
z;II(4GI(3a@-INdF22XwgoTV@!1wq&C?oAt{;gE$60*cklF30g!wSbsz<aJaEc^Db
zJl9o{e@1-~f2~I#Nd-$Mt9v*MEY}*(B2{*`9F%^bk-lbpZOb05TiqO+ic1(`qa+X7
z+>Vei>EKkk=s9zoi0ziI6gz0$QzBMQQhKJud<*~5#foqm(rN4#lg{ka{|vd&fkRNG
zw*$Lq{l(_PNO9~I=_rqDrD!AHR$z1izMoPnJ5xHtf5qBlswN^7e=YE*i`|PLh)jcj
z?5K#92<cVBue9@~HFQq$sBER*W<Db~T*F0K^L5cf_}capR+vncX1c|EU)C7tTE^1&
zfPwCkj}&|?Ujm}+xsz<gladC@=f||x@wVh#_W@@A)^tje2dd+k(s-M*jpj)dEZKqp
zgH>$ye^<tOn<9NKvlSVB(}ZWy!gPF6D(gb&RxC}yg)^!?8*Cq*za<^L5V`ajj#Xgs
zR8PsReUa%d%-X&$5|9T7Ixe<Frk2Oc)gBSUR6SPyQVcX_G6k|R1t}1Ht)H~0pJr1r
z8-p5?F!hr;ZHBa$TZ>}>!npb9JJ&B=XpKM?fB(|{R3Ofk3oo5cSuiMoLr7nxTv5?I
zuWC=FW2(LPK&Zs6i)MW;$x834wohE;rK+#J(P#^kOlSU7iT=D(3OA~`=6MboJaVVv
z0|i%Fh^M)Ib-dGtQ&n3vu4d3|=H2w=m97hinqn(%RL)LsJG&wy+Aupe%T*G3$w>(*
ze;LR`KRY*)yQleVtKSz;5yS1<6`I8Bb|rai;Rp-5UN7!OM%@R^>SnD3HnGk<-vCOz
zu=YIr8mdlIUg!n4d9J(OvrqC@c*OB<0^ZrOH$JNVb-<h<HZb<rb;?`A!m?3b2L^lm
zC4XK)aKpGe^3N-D7;kC%#IEfQ(^~NxZfn&Y^Z^~)9mb80?SB1@21NReKygo}+geDl
zV%O8EFWwt>&<FoSI`#*dBmX=@i*TINmh7g>8Y$aEJ#Wk+N{*9EgwXtAP2Jpr*CHGA
z!rxN^Cs}J3Nrgzd88c+0F8$#b8tvY}MEwspx<53Ru^c}Z0XCN*KRG3rE{8x2mjY-C
zCYLWHD+_->zoZ|+C1-|1P7wr1;wnwyq=uaWZ4Dui<%_LEz7tO8#QE?0hI?9{a;I27
zP6A@U5_iQVhr>5--f%Z_N*q$i_$~O!y;TCK_Lc@V^HxqlsrSn{Xc_NOsEI;OxU5q8
zEe^G{-jdKbKk`9pXw?T##KEk3i}6@ay_xYduLpnhCwaV!8&Y$B3e#%p!xL`Qu>;{o
z)BYI2R!i)25&mY<Z6^$7ZR}uy68FJ4p;`Mfaf5Om`x2U?4cr0=R)(hFZlT_UW2k6f
z1Pi=l)n{Wks&V%UYar7;uB5<eT&Co}nfny5nse<_6x`+97gXS2q23g9kTO2j0vGG{
ziUohJBtBc=z^#mr37X^DaD$Dzkp&Z5?H!bY<lINl8ic4fOGC4CBe7V)XuU%UQo8FC
zRw&x~+O!0z)J+E#q>irE9HfojX}D|BAA_mI2S>A}fU&wuuzLlo?@bCt+Sue!G<Qz|
z6s1b?*<f{PwU1jhr~}WNHE8B>+y?xOcz}NsRXSP|CekwEp#hd!`@{gCw$x{XkF<>)
z41HVcvtiw8<Q*J?HT9hb9z)IJG8>T1BVb^4U2jX1P>b~`40r1Y4a42_wpa<(+Blnr
zmb3Y6fI<U~`rU@07`=|LY9;rHQwmLchnzBErFY1PtY#h@Xd-S>cXh_PZS*g~E*O9J
zJ<iF3iFgELcn-VL2gAcQ22RHQTkVi@R9MwME;iCE^$~oZk^|b?@NLT4=ksR}Kcj>2
zGfDSQ_Mimd_|a#d4d4HEHXFW}&ljuV`MZ}Zzj|^q|2Yg_E|#xm%NGd*lMaV}4Brf2
zy~yLmv)S?L1rQjmd6ZdXV68}sBt3t=Vy^9Q_#$BV=fl^F?-#@4@QErHDNPbEiU7ij
z^{=1f-<^!0BC%o<NowTkp-EBTFDf+>@wO4QSz%QH2EhwibGo|`TvE(NlM#|zWJKdt
zM&xEv%n>G{gkB;$jMzOdb}r=gyqPE#gd}B#=Mi*;qJcRnTdYzKA6;6<@Wp>L{PF5!
zF^56_e)f(3|K!bTb$0%@VR(JAdh_mOJYKvVetP%z?9*4XUxsgwR&V}ww)$qink|1k
zI-bR!PKTGLi<iSMITF@AoG*`ugOk(Q!D9LLXmvh(_hdL<oTA%bo^bi}iTEx*8W;EY
zg=dt8^*kH?^WXmggc=(Ns$ze~=I>5V53lN0(G8SlEdo5(>e@>5D7{8kN(H-jEdz4J
zsEFt5bS>zbH@k-6tHtSJ`TPt{44$^1h2iPZ3T~YT8ZZC;e)aY9)zNAe@_6H5F<)Vl
z1NNW>gZV56Aj?d`<AoD<MvBD6=bsNxmy6@)v(<~?>Enap``P=|;Rb)Io*uoP!3J#P
ze0AOv=Q!ZA+4<t#@_2T>PNshE+u5s=qc0clJNVSKg3%#HSmIgCZZ=NefqK5y1l@#3
zFGf}1=8vM_qGcB?N>i&Z^YF&F@328~NE{W}lrqIydb;sAxH;};^Z4t@&nIWT=(t$E
z9{g(fZZSVSna_@vKb(Ip7q6E`Z{MEGUw;7e?dbw$7nm;s;sj}P)riE6JYFLRVVf?%
z=h6)N!gB+XbSZLo(z<XWl6-7H6XZ>Foxqf%6?q^lNF@SYq%8_m0#k)+5R<h<<hm^*
zAVU(8n-qmAA<x6@?luTXEUpFtBj1Krk=Kwyz^-u4jcn*_Ks<l30g}NkZrkW_!57Mk
ztVw^=wd4|&4GPSV<F3~g5Cb_~V}5A~V7yLGtUemV(lxr0IJMS?B*EdpMURcVGaNwF
zQ%Y%l0hy1YVZ5l_yt@92D~NLO4D^@L`Wb+dT>g(2UZCXlJWldP>(N|TmN5<7n1<Q5
zX?Q*3jcHgmX_$Zcf@oQ$R+(D5gM4pFWeAx9wl9ql#wc45zYx#GAVlg84uM#Q3M3GB
z*QkVQV-FkxxuZTSm`owOnGNWJQG|5BWy^}@B_i;GvZ$t3VRzXBi9iDu2o8oScE_Vk
z0(*FYh)`(TMxZiMlwDpEqL4rxQfLpH4uRwaDXQIA0Ox<s?cj7$b%DBV1bGqIP;7z&
zxl|P6`7)RB^~4%)3m|St%1{{h9s3)RcQ-XuXgdr`x?LC+-3UXO)R0VSNbK&xN1US<
zy9LS8t!K%oh0JU5h>CN2w^4&v<JPzoriKhw^FA<FqXOv2gwPMtF^4jiPz~bt4#)vN
zD8Wn-npA&tBict92yD75p!(SlT}Y!Jm6#xt?<fMBAm$`1qXwfSNT!!jU-HK?_(qQ@
zgJ-Om<vLwS8Ftn`Pi!b}$A;$H#fH|6*eFv(l&L<@w1?P8Eut)j00ILdClHVU7XkQD
zcBBNgX0<f%F(?$BfZRPC3nb`C^D_yF09g>GbxMDLO|~5x1nJF1Ac-4S<depHt78E?
z4jn7KS_Lf>uCe<06I_rX?C)|a@XvRrCr6J@e*6*OTmE{qeC2JF`oAmjm9CXFE_GYW
zP_I4`U8gG}sbRnFBZw}=&ECrGR=0zwmD>f8<&7Y!Q$bzx)S{_1slYm^z&feGs#9x9
zQ+0m|$~pyQo91YFAG#4-iGsmXSM_7*es<W9x*D<u-(P45#Ri8s%uv7uF^vWxEIpOl
zJ;%oFb8JEAq9O%xn4fo)9958G!DdMITi&Y3E!DCJiaKRHya#>DzC-1M4Pa3OYP-^g
z^I3NpYRb3=D$P*Q)nD!)v2*icO(OZpC#rwul-q-NfE0A<=i6%=T<MEnUtz=j9GI{o
zsesiltdJtnp>UNET53@^qXP0~sJz<#j*dLZHh<sX=dz#=%I=W|GQuI?2aOj-U@Nda
zcBK);nNZd66FXuk`}QJ!1(67T(F&vk6NW(b7nBveq7P*R$g6_KGeT6%LzM8FJcWOu
zDBn^4q6BCygGXUvqKvKCqY+?MKgU8P8Ku9SN)TY%MnGaE$qi}?Dahb6{hbDH`Y{c@
z(c^;#$Lu>a_>QiGRO7Eq|19z$uLs<e5B|RJHuAyWmVS^AZITLYk_v5-3O3!d5BDbd
zV3T~XNj}&lAM8Rt<dmjXo?0}ucxr#i)GBwE8QD@afKjW15$OY!F=Ry0hIB9ao|6_=
z#G${YSkEr%NvmEt`~7e=)IS7{s(S*W8J7**01krS(T5=@Kh4+VFaF^OSMH**`YV6v
z7^>UcK)JSB68uG`ro8+8=#3L6ybl(+Fhbx7M9Nh*n3m!h8+9S#2TWD=fe3$NEF8N*
z7K?v&fs_U8D;xG`*YR9&{ZdQG#lOU%w#WBAnjxwB%i{_G=QdTKv_4R_@WVm>DkP=o
zUo81KvbZObMdfy*Y;lJo#1E?xm)XYCQY3K{8Ym8I_i1SZQ3=Q|vqE2LQr$;zP>I^h
zS>Vz?PXb2#qZbBi(2fXIh68^Zb_)+GkcnKjhy5Ty-h`ME)MIW0OKinIn`#b@2SO%z
zvH6!Yi6i^#nF<<)qVNO7WDi4^UKskPPHPiHz0$hc1Qf%>CQw{f7#3LCN3DZcXf+t{
z03DV5GYkKY2T5bsYuT0!9>`V*l}<_o{OkdgNSv<bMjQMnzcjhphI@b63C~cF*3~*)
z7tm8J7Ml5Y9XE4>OF;it?2pf+?)?b%#)>cRP<;88ijOu9jB=hdSe`Uko-|mVG+5q;
z1_Ol{)&4?k{A#lzq_CkiIqy9Xpr?|xposTR71v)ocu`L~%9DF4-eHA1`oaqF6JojA
zimD2^0`1sJss5!H?;}jM-SPgFuyB8I<7qETKgglmn(*<HH}W`6(!W(hx(Hp4!3YA;
z(!RdfdAc|~Ie+s53_9}XE}u&0Tt8A#ZuYoTCbsEAvi}3g`A@8uu^c}Xm!JqY6qiba
zKgEBBD7p9leE;=L62nqNm%gTt`PawyxBtD(>^mccmiqMHzuwXfFiM!9PY^{==JfaP
zr$63)o2hc;1cl0x&RIJs$9mzcIXM@FNw<n}02u~U4&3mCf&mI?QD{H`ttx^DKoafk
z?>|q|m4E%xWr(Qy5aw0X&*?R(#YLPh6AgcD3)SHYQvK;3<(e$G8A3%ruVP~1JIQ2|
z96HG|th%Kgt*!7HW{UQMH1G*tbrjJrG-9%Ti{3Mz>N!5GO6x7yEEZ{TBdsYLXjCDb
zHZ&<k-aju&G{VP)l5{MllUIYHdKJ@&dP*dtjj(<plf1K&cYE0o4M<_*!raxU=01NK
zM1XaJ0y+@4@*ON4l;jC+w27qSq<^$Md=bA{sa6*CW`2h&JGw@Pf<6qjVuLVv#_Msd
zX5**BBqbx>SE^U|5mgdhC+?zg0?;rkWjvxbkZdHYe$FTxuM@?=B42kUPKhY}OjXO(
z9YpsnK)J=}N2lWYGGAH-;pJkb<l=wi$0~1Hd7;%pKX@X)IjNJKef$_sT@d(M;}q7<
zf%Z;#@7Fn1KaNwQG4z=wqw9y&Xd%PGfUP5Ms%C~0vuMpy-k2MFfjJ*USmDklcap;X
zI+rOD!oo}#AH*T&PStp`7Th1DJt)603YzosJ+j-Bs>HNf-XwC;0O_hE-C2K%{FX=1
zkdLm;GeaE>kv2TJ3X~&Dkz&}Tt|aS~GN0Bq=H0-V&ArjCUE3VP*Hh-2X!N^eXqL;+
z<0q|RDGSK5+?w332dTU!xjIXG0WzM6YH_cX+NDa~r~qFr(}8Y%g=V{ZW5Gdnjv>A)
zf3735Y<0K#S~>8Pad07B%8`GTEve>g2{vHRu(6$toC_gcXU~lP{449jbd6QLQtp&F
zNolclY3EWJU3WUqV7g9PG~Scc*0=SpRYRSFwtCgtrDj8A&%2%nN}lnEGtxl^`kiNd
zWS{ZPs>pT*s)dym2WQnN-v*uVy(srDY-KahYu1Y2flkk@T-Wt_mK=YI<Y~VskS<O#
zc!NPK&il8M*KYX)E0yqAG)$Dr9=U4QHSZrZME+vct}r}+G|iV?nf)8u52${D^z7JK
z4drC%=%%w)u2{CeWk27IJul|bxh}u-B5m2#iv}xVoCT4W2a|$O5awBFTZyF$%M+^r
z<Tz-A<1TY`J87g$mFj<`RINEAsC38YJiDl)kI%}vcURGP3H5F|mpnG-ZE<7>k~_H0
zXLS&CC6KGSc1KY-AGS@m@%6YzPB%NQ8<W4e?QLUoAOL!J*V%1i#qJl!n2uPk<=aut
zzlNh+?oaZJ2bHA}n$uZ20McQh<;cN~Yu-3tnO<|3<kr~%${~NJP-AYSy0sQo3OAPR
zgJxl&uf_st)>A#|wz_Bfj8!+D1$Ntz5xzpE8MeZ2tF0FM2u@hlDi~eEPg#-Ng(Y{a
z&Rsj%Ilv~bqjvqU1I#Hm36N@*sO$L)Y@^D+><`+vdrwH4o3w+KKtEG8*?ko)0xU-)
z7GD!uq%>+>yM2GS=$W!HLrNPpDpiseEBN?@b>v9zYpTeB=2Sy|1{0}u=SaV%F1P38
z4n`;+M;hkiby<TKV5a@psY_MFU<hU$f1p6p5<Rt5)4F!O!ldf?(`!;qy^wXQvL%4;
z)9V@#+UVzGH9<h|pI;L_XpC-_b$!hO5Dr79%J}8wy+(gmGK&nm0KzOXt96fX5Yfj$
zoRz*`e<$ap^cp9+Ro%piHK+cTIT^03fM-@|JUXq#pl+1?hMwMywOYd?R&#lRaGTpG
zFyaC+e}&H|j5E0oh{5bQgV(59p=3NkR06r(F#jWbqDs&&+MR=Zh));s(|skMHkAUN
z^Vv_==lOpHFH{H{XY3Ha#tE%^P-%dAIXg!H?o^<pc9*<;US4#{+CMnZhs0JOj8gxI
z*hiqc1Sv8zGCdRBg9nrbTn6UZm0Rm~t5^jIkks?+EPI8ce5+4tZaaf0y_VSqjrL^F
zIyv{S(10s$&l03lYUTZ^JViR^XG%FwXLV|Wlv;m%w4VEC4TjD^*FzYp&+s%fw^<RP
zu$a4S0=3#Lf!Y#Yo`q?N!blI&H$9~T87;W;S#~2prO>Ip*?^8{E(gOH;?e5f#tjFD
ziL1F{wLFxjQL?UG>(^_W%XYKSZwSi0jsaV%*QSxlwrSi>2%q)3zIOE{e7EX4bVe5z
z0Rexrwh^$sbvu^#2Yw-nXu?coszSE?0tob*^X#lSJA>jHEW<lA0K1<-IoJc0C)9_J
zR*f2g$8_()aie;*A!smg<vCZD1OAtOo{2)$aK#uZKIQm`+Mvj?R2k@j36)y_vHIVA
z-wQHTT8n*tfIFtrUP-gC4#zcL5xpY$0Jwiw59OOq{4nl`=0u;@Y`n}}eQboC{teHE
z>4=fAj_JJ?YnB+@=o^@i@vFEHuszP6a1R$6H8ixu`W%ktY{;*IFuvy{{$06vuFzG3
z63O*gIi@N0=jIMp1X=^E7K{&q9)!DApwKP&bp@TT<9JtSOYLz4&D&8PBnjLcVT6C!
zTZkixHi~A2HKM`;rXyA%zNQjaxn9;Qqp`--)96JTyh{c*0yfn?^yt1|FSs4?Ls0v)
z4E}80)CJh9x<QZoIqlzp-VaNsYyk^dLt%Gsyu?ra?A{Ecb11AKgfw?Jy&7zGp=B;*
zNh8;^o4<34yghT{R>IQ&c%UJRX{CP-&uVBGU4z&WxmX9hw<tzH{;59r*2#DrYU<o6
zNgq=C-t#RX*M{{}N{fJ%sn)5047@ynk9Y0(e$_Aq!`iTDJ3%;eufRLr>2Bm+GmovF
zBwNaQ6Ja+Uq+p{Z2HRj~|MZaRMw)1~uu;uar?!?>=k>Of_HNKijZf)Lc$k0wR~nx8
ze?8ahqEV&iLZ_|7lNs+<o{6Dx&lX$#nFqyEEMIMqsBM&&Wp=e5CQ~X~)udNmij-BK
z@F_~JrXnhW0JBwburRUy@IeV^+cmCAz4F*GiumqzRTHa?SP%8GaF$cHR(phde9awK
z)-hgI{591w%w+6pxMC0MqB?&$3ZRB@vb)QnYdK*}U6RLW%An<!yMszw;ewyLgC=AY
z`=4kIDr-at`i1u3Gmp*AEyH;gp2;k8i_J!yb3!>iQ%lckC-0eJWP%R!9|>&voxjr`
z3P2Z_G5yJ){K25#&)2aa%faOJ28gvN{X&@=&%$a4LMpQm<KyrXq!WLot=<zgil)(n
z8<y-L$H(^d(74c^6C(P9jcZ!vlg2et%h)vit0WuJVIwQXrOTxwMpn&cWN^#K>Sj;k
z956EK8BUb;9;gmh;A@_7sIyr9(y+4jMkCosEsR`fX)x>~ErDpon2h(d*31}UFDb!s
zNWjl2)&i%XL~lkf<cMeDv1q1O$3GuApF{$uC#PWl8LpP{j8A^CM~X*I`m<K;38PkA
zZ{#mNeGKi1A<LxZ(DC$T6tQJ?1a?1qsA(VT*{P+T$Sl`u&&{-I^Fh4vS?(<!zy5K(
zEsSdJrqd^;P5%eb4rG_nXfzW8F*7lj5s5Dp0Wg<A;RPp`pd~W|m$;5UFMn@Qhm`g3
z+n|SUU(O#d-&`#YUIhYghL2wd;0K!jg62U-g9lK~!tn9?@Z+1-#_{LOAJp1f{&@aB
z@Hd@aEd-tcJOg+X;PmPdc*OFE<vE7<Tl2^8|Gtb}%X-J_gGNjgm^Ro->GadbcQ3jK
ztW|XB-}EJay??y?@gl{i=wy{OH2nPgOL~A7lnrJGM!7ir`OEO@%lDG##7gsu{D<`L
z=Wl};`Jvb^YPO$tj;OI8Xf`Si&P8R@qijF?sx`5`$a>?2Kdh?(p`-K60qg*Y*T}dT
zVaDHF1FUjIo@+^RHEfs5k3SuMCM&aSqRpN~pLwOoAl8ryZ+%^-f?lEZODZ)!Qe2=Z
zF^45qu6b;XqY^)33^FGhOI%X=eM+G>{w$=n1h7t_4-X3v^4y+r<(dW9VQfed#68Aq
zgwrmnyJm|rpdxn<;n93yomdSeZI#nz3WiG#oz*>aE18gMoSjr6NQuLLzvVz7%3A{2
z;*<jEa(Yc>OutJI)!FIkxr8;H%s4C5r-?>CN==B?ghe7y$xTm8$7iu-9L1_ZGa4ES
zAnDf4PgTq06+b=gv>KCX<C|(vjGT#Z@^s~Y8CPdZ_{M5qmeP$@UT_(;@r}iNUPGo?
zQ?(N7Cta?|q799OeY2*2<CCI)vTLF-CdNAE8eh(~i^Yf6!p9T|(_!&-EjDi-5@<N@
z1uG8;0N$XSdH;!Hf{ark#0QRvTStJPINN$@=hLgx&?WC;#j1gxj-h5`Jcc&3Dv5~L
zg3(SQG=U(_JWN#UpOINJvUlcr<ns{(Gx$2kgFNPfRp6souz!Akt>#(YdMf$}8;vuu
zB(Zwh7W{IDew}zN^%aG50x`uajBzqjSdQ$uC^D`KL2}6{d6xTDuoX*SM5SRmB(PAa
zwxsjpR`5YZRzXmHwiQA;kWg~ju);lCLT*E3Z)%!Sbd3vKrM#O(WXeEBYhc|IQ-+|p
zpEAto*Tu0N$_iJ1{68KNtNR9t!K|NP0EwN)$^LM(w3Pp@Xlc15Fonyb%Zx|bJ(f3M
z(s3dU^4_N-r`JtsqcSj|SBeEcU5m>OnzN)JLH%z^L4wPt%;}jDbyj&V7C0`eVTgdP
zvxtc^3aRJ`Vpq0q%1e;=3Zli;pE`i+jy8cA#GBdY2$;8j(I_3~8Z(|Oy-n__;4RF6
z9?ugj?%_r;QP0ge1q71YDuoSsdDjw`LxKXf#m5ZY_oPYtm0~9iz-AWPlFFb}h_R&v
z_H1Sd-9q;<r^kKiyivj6zO9P1J-XSFl-aYtan*iq8)O+yAsl(uSydzZtXH>j?=fqX
z=b`1=Cw2yZ(*I*|o;w5kZn92!*Qzl(#W=I5r;M|2i+Us(`+b^|g2C=PE)efwq3>OK
z%<YX)caytvxg$3}w<haLr5X3W9uz%n<8QrdA4DERY(>vmd(vN2h$C3Il;5%1&xrat
zTXeK$t;=c7u`6dn7kIqEf_FzluUY-XuF&m>GhSJLKEs3rn>D<MxsqV-h&j)E&~}>M
zrNO@zZd{u4`6Gp{0apxkN5FMif7!4(7;yX@362I_v-&_caQ_VC`o=KQ5L^y4hIT`W
zQHKrK=~8?Cm+W^N7j<%}RFKM<7v}0~85jCoV`vPmTVYhYl((QvbsZPMZ0e(~YgXUc
zZ@mG3Ss37@z1dQJeh#qAhl_u~XT6Ge>PD_0(s)wd+OC6n0~8umj=_^yJp!BUCd{TR
z6jcO4O+qm~k-RSnf?*i{3M=q7kIFuCUYP;YG-r9%Wm;3yq0s#@cbyj$COe<Ny^XCE
ze(E~Odit|v7U?ltr|PyXH|+W&LBkmBsy`logT<*8YZ<I}luoY*dZo-Lk_&CK_r}JB
z+t#okx!@kP*IF;00Sjk?(8A;WC<ezxnw2Y=g$CS@ST5^`6RqS)%f9=vQ%`Xz|FS`|
zdADGY7}-(3GsV#mCtn!z*4@%{$I5(T(jC>dr9!Ta(Xzsn%tp<PKv>*JH2O`m@N;*6
z1nJvy6(sKnj8kOXKXON4__Q+iiTyr7uxaQ$-ZLtf$1PE_antO1IXF4_u9o8Fg>)*;
z&D_ei@9nH|O<**8Zf^C;K3f7BkC&QrIe7y%mrK@13~ORwFc78SpdY?LnbNg3%4*wD
z#8qEe%^^-+&OV?Cmo+IE@8t_oS<^XxOket|>f&06j6cc#|6D5s-*iNFo0xsb5u8!j
zq>?XfI3l}E^4)QSXV*o1iX*0aG&|zS4hT_kW)<>o*ug|TZDTk4i2bW41>gM*DEaQ#
z-vx!19<~3C0uzoYYZzX8F0zDB;MR@o(oLL|vybrghL2{`@m)7EV`zp_zT>-pV`M&d
z(7U5XUU~KAA??sH<aSbut<cRE4Px?VoGIAizB2{i9iy^ZjCTKHk=qJ&OWAhzXCn7L
zzBO_M-^=FvVPJ_9%3L<+?U||27J%vqC~W8e<<b!B1~x3Auw1hyV?nV1Ryl;F-thj3
zm(+Z_t(Wgq&2J%Hnm3fLxypfOdTH*~>YK4o;AOk3YenAkM%ns1X{B{3>$@aTuE;{I
z5ksx0Lg}TES-zvz$aATBP1^cJLl7tKt1k;owVHerPZygslC0Ereo+fva!+JY(|wt5
zUdLWl!Vji}#pRo8y1xK5W)<z1u^c}X7BMw2ATS_rVrmLJJTFXTZfA68AT%+T5e-2n
zmlKyiqXFT!x0gR+41XJ$?_k$AlxtF+rirEIXiVD~mRj0*;~cN?v6+zRW}DFLA_~c~
zd!>GQ%K}Sdk8*42*2Nz84ij>^+YZ2JoF5lKw6n}l*^XtcK|c3Vf5-6E!s5jhVfOs&
z`K`LF<AunHwf*X1Q}s7rXf_y$5N3=%PGTW7miTOB%(MWG&VSfP@ud#r;hE;1>Y*eN
zcD#03T772n<;-xAZy4s@RdbMYBDv;<4g8qawZIn9yRp|w#wi0?1@?(r->d1|{rran
zC=B*y^pg7{^wP3Y$7B%E1r|t?uCdL^n4}w4VN<_#VS|0W#mwUJ#P5xn*zifFjlZwg
zb#flOTd#Mmr+<3g3dt=Y!7sb6On&E?fNyDiyl%Pl>V^<crj+78B2jL!HE)e&S;$>h
zvb&1|w6=g3<+xhPD#mlF@th8p4ZN|O%d|aa_thTxEe2A;p1;k4$}&E&4NpkC{+nx0
z`L+132h}IV{A?<)H=XXz=9qL51b6K8fkwRwL2~34#DC0f)55Ua85AfaYr4Sio6<z6
z)tOOE*Prt!<E{ht_+~f9r>n)J5vAFh(LyYW!sX<;jj*?1SW$L??ID-D)o*-1-%pVO
zJfX8d0?RA+vI47iOGa^+;QJqrVQEL02JF#3DmKVEk1N9%FJ~~ut4UzEL%6qc;{$9a
z-E?2}T7L)pNeo$&Gl#S&nxEP@wRKLBol6eLFaynZ>C5?P%X<Pw1W!QTSRMCe+}GJY
z`KI-_j~~_9DkYYUoepYu#!fNaOzCd)ttD5bTRfaZ8o7U}{tN(*=EUL0TFUts_@?g7
z!!iSHT!5#BrT!nT6sG(MYwa+R@>2<pC1lXe+kXpEuSmN|ET`fr6~$C%<9^I{fH+q&
zla22HwnTiz(=aYncgz8es`*!DESG*?B{s)iFY=z1wKK71-Adfx5zP)xw3tf0?!pVZ
zZpJVA*j13x+6^#MdqtwyZnJRNt(5xO0s^Hj<#AfKp>S)T7*w?O=mA_);nrW)r%RQs
zPhnISr7h1ZyRd8=Gh7Zglz2RDvsd}&xwt4g>>ufV2jSd>mzXP}rO9+gedIr{(tgW<
z<MnB>k^NXI67@gm#9BJI2gn|DUAhlqj+Ja=C)KSit2fYf;yGl&uKxmBe<V+ru^c}X
z12Hx?mx`x9*ncUV{_*jzPZ9~H??$CILb<g0U<J^ERK7|#jCyl*aml06B0$qPTI#cu
z4rsJMhXwl7+tai4nKB1dIalAh1aB|!0ZwcL>TP(zH^7ez*qeT5{%n?ygW2sXj`jI>
zdn4<bx&QQr#+0|M)gzr(q$VPDZLQMsS}DoD^Q^AwCV$K4mhMgTBA7VKCF`xw_DXAI
zooBAIHRwPVP*$_Z!@E=IQJ&8@&eJ%XkOPmBYHx}h70p;=d{O3AwUxCZ{uELL@>H_$
zwbQTHLPkw|G#Ov&XnQ_AgFU{m7JAEcoaw6;k$dOW^;>McVlPW?3SUSY$-K;`kzQZ@
z!^+H^zJF9+kp|EDg{sj)G0r2esJp0^x|3QcrDmRuLun0mQh%fN4lTlhMZ}#T$s~Ov
zzKLL1=tA!c6U-PbyiqTp*kS~&SM0<oi9IcA%MK|^&nzI_#D4B3L2EZ<$41$07hM{U
zlS1j{1T=T~1p3bO6oW=?t@JP4S?GNWe-~;gYJWy`lOeS=Y<05QlR|Dbmufs^#?Zq_
z)&tHe*$j%mDg6RpopZxxK*{2$2s$E{zD$mHC7uH}Uzs&KO1R3wb33ry6dR`$U_8FL
za88N~JX<4QdMndF<!`A;6GgG>wI+)dM@Q7fs!gD(2BcRC|0k_q_ViJ}sxvgtpkQ*H
z!hcO&hpe}bvDIrzK)h6*bS%<h;jYVflI<|N{V6U6S%fk5Fj^Z`TyCu|r#TQ!mcmfL
zq!!7R+RTrI&Kj(4q;;)DlmH5p7EX2O(!7Ku-=fn|KqVmLk`+n2UldT4Gr}1&&q}m4
zzhDu@q+;KcY!bqb6kl2tE}%z%_#k;pVSk*D9=?5&diw*Q(E82aI-Jq~_M#7aUQY#>
z-Q<_0XS4L|tTXt2Thb?oW^LspK}dRDxp;bo3SnHoL>-OREUPRO#nUFO;!BC7w1i$&
zeyGioK`cU#`sOk`#m*VUz4}Hu1g*BhUjU)d1wI(9eF2bjv}SbKT?;{r>ln?*9)F`2
z$)}c!iGxIz-M!Q7vXLpk!>s^ag(L*%<imijW{o5Jm1{ks-UeOv>{2qF&aqxQh8rwL
z360J(Jp~qnZQ~=O6W){LTTuql+IYackZsT?iK8fL*mq;hH8Skv)@-NB6yajQ00!r+
z)QmbHy)CsKlmHuR@q$^7Rhr^Mi+}35)gT^d+9qIY=jhscQ|>QMOF()-tiv;VS$`VK
z^K)&PGEwv6mXX>_3+t?Nkd(3CWSxFzowykt@ltFQU5D``1B@If)f>5^;@57I_S!~V
z7(!xLbf43>tco;>yw}=yNr3K}_GgqA+BP&SH;gNPw-0-Go%y`8Q+h*4c7I&YFn%R|
zzhzOOZd7~JqOm+$HJ;q1?;qHfnxcH;g{UiU9P<h}@UoLq*!gLr@G@+y(e{fUTshBs
zPI4w#uS6DAB7yVF7GZSN!I^eJYkROkC}}x*=$42!?$Z(8u5a&3tmm;9GM(nktVp$a
z&VAOpr=zvXj-&J`_Rq1Q9e=Zg^%3_;?aeEe9?+8X%Ci!)-&iH;YSue*PQkWKLo$Zw
zn(eLjZBfSaSdQ-|8$tdl_BEN4)l;Oa$gRexvS*stUTWDmp=|ehUKCJloe3$|ycMS?
zjO7)B{iq-$S({?ssdm2D3o?gjj720h_^4``7=F!G!uI=DR`&31mVXC!%ecZ_D;ax|
zWPaB2aBs5Hp$fuz(-5`6Ea;fd3hkEJkx@I`rrgBa;I~w!TI85I618x&-_a@;{q&CF
z`;tG}?`Y?R3x1xp+Nj%m)Ztt)?_kVsuJfHybvHPTs<!1Gwyh?A?$vRRTUY1`^~uqG
zozJ#k3~RNWu%D1-oqvUL>FcChn(6L7y06}%!_so46JEw4`pj84@EkKo_Yy|}#UUR5
zTLkjdW3cs*m_%Bcy|y}D5p1nI&eaz3?U0tUQ$c@+iQ{pRRa>FcPBs<~Hd<_0A??)U
zc3grtBBijCKC9^vrf*(<Z4v0%FKdIyvn0I?vctL{KD|yjB!6%XmDtpEPWEh;_G6O~
z-f3<>O$XlbJ(F*>x68N-%e6$~{fw?Wx84P|O?l6jrQu3XGpy{nB*;oC1rP3(Zc#MY
ze;QGL-#0LS_HKnbJ+BZVuQ#z1@6V9LE&GLm!WcE{<|kvkjZOr+*Z5tVZTwX|EUhRi
z;Zr|~`l&1p>wg(JM`EM6yvtJ=!LZ4=hikKNyQ<?JqCnlLht}D5cF`znk?sZ+&95qF
zOTWSlTrsij^^jr(=gCe@Kbu+0Ym5en<u06?JGlmcGg#0VD`MY4Aa+%(+IPNj1bKto
zjwCzkyMb^MI~kge9C`gY3YS8L6Z?v>wOZKNkCof`4S($n4XfETJ7GDydGU=pH?0HN
zNGA0iUQFvto?CaP6o>U|T07^b<hWn%s*fu0qtzZSRhSRG2dorWU4<P5yPcLaLpotx
zZ=v5>BiAlQy-8u=6Em1B_3>RRuZ?E}8|{wc$71Hb-t9%QW}c0z<!c0AE819bO08vQ
zyG^EwAAegjyFEVXFfj@x)31rOwemko>BNKNqZ<fgg!W+~?tJxp<8}iMRh1{7b&$W_
zPfzbYKK(viDRDk+#wihk**<-H$}g+}MTN9!Et0a^AOGAwKYf^Ioo=Z}e$A%HTzc=^
zcvb~g1@3KJ6!N=`YPmXaGxMTDMw2e{^7{rI-+ynVH%wqsbld;7+^n+Edk->(W~)@n
zKJoPCKVLro<<oCUz<~ee5g-ci6P4ojtv)&j!;&`JSA32_s{lMHk)-i<^Z5WBNNEFG
z!>8>D782HI&ANTftww|hdSrvM0-}XqZ2{IqG?W>smDc;p8;PpEMB%BCk)QG`I5t1R
zcz+s|AM=MlQ&$-EukGpc_Vl62Ga2}o588tCD{;c2Dr6wmMM_O3h!m3bm3rvlvr(=Z
z2R5-y1GhfM7p3z)0t~k;)i|XV#w|6$R5SxM(f8bbgZqimDi1vRgp|4E3ngA<i&QAm
zD%Sq{(>EOU1qQum-*uX2Wm38O^$Y<?1b^rAT*cFFUG}QosW-XW`DFTuHoDoE18L=z
zgH4kH2_H5_Vqx{U^ZMaY@qm(?a}S7TPApw48sK=H;^>+m_SPu~#{v+~DLU*z66T;4
zve4mC%&&*KdVQKthxkr<QDMQX_H11~K{3;HuuNy2*UdX&-EYYtlB8P@?bL&Mjeisl
zSI;E73dm0jHPt9~wUGJLP&bsrZ;bF$7d>S3%XG5^wO3U=90hkE#1F>%LEBs~>?Pkl
zd}?6gGC+(A+j=`*=w(l4ai>wuf}~tMlyp?sm5qN<arVHGyVETExI1O_<A0a`f9Ar)
zGiB&!mZCdcSX^zvgbrZTc8EqdL4O_b-8T=r8?y-Vw1J=${(O*!U3WW=WANwmCFF@_
zj{U+(xl%bVmC4hzQTuM}2b2kG?Vp7Gv;?;FyoKVMvT~W^a)9}6hIeF01tns?#5ja^
zHC)@<Z`^n>+m6K$e%RpRa&+271Rg*aSJCD>y}j%m*ul~F90eU7Loc_UkAEB^&j&pu
zX$OGF_Dg3MUb^5HW)BMw`c5gP9i6r48StXR+qKsl0LTF#p8>;B{^mXT8&Su)Qpgxb
zcd5KxY+4gt)vX^&&BGDIJBk@6HRou|f)f8cpx;}(eP_|j<BGaULPs?-k?#@*Nj2{#
zc)APg)$VLqJ$h@`tA~V?dw<FYPHf6uxJ1dh+)ZgKP)!VHig?02-&*nQ`uI+@D&~m7
zG&>g4nGO28<z*EiJ83shI#=l*t2AlQn}eu3PBvxoot;aQ>vBJ8=fmjXuPBZTu=*d`
zdm7bEC9jui_P@xcb+W?gncaFnUYjvot_fP7&i}w|;X73~1{>$(c7I2e5qA5={-7p@
z!Q)M~xu(z0)VU(*#D;^PSP@2`tFKI~@9N)3J+h4Ee{E8y%n~m#@4+eV#k`bv|7H?H
zHV+dt{kewXM_~Y-u%(0}eo!mFyL5W{_{j<duST+dOl;)@d^Cx*ES$=J!ZO)Yzk)ws
zwO9qo;DzpxnL#GJ%YQL>3BOMhcJc042vSQcB>2ck*kgixNNCzAVXU7PsCRPdj+8sD
zDuagrjnfQkuUfg(v3+ZZe<99_p$)rA(;pL+{jKJI=`~#TwY*0UiS=s1@qsD?V6WWF
z<vQjbo}n5GrZ#ibF9ao4Tqy=-*dH=<?jExM>K(>i_<Y$AtrXu|0rDQs=2#AxqBz@P
z{~x93(RY`z96uBRF_+LGEEJapwLigs=hL5Wza3;^INuAGnuN318H<8>NnD;KP2T!)
zbucR<l0avxw<Yn~$UG3GJM)apBiVH40Sdgm^q2PiIGTPAsWj*H6vS8I@wFe0<bra1
zuOuTQqf16mmV^PS8HAV96DrN$@SNx}eGE&RY=+3}IWU$-B72ej6PRPXD|JkNB0+c~
z%(UT&?>UWJyEMa;r(r$6yv}Cz!Nf?7Ny!Q6&0dpE-tJAJhxM}^imArtT%8+Dswr9g
znXgO#dYlqSUACEY%HUv9nt0Jk!LczwtMZ%Tm6@(zA#!hvYXli!#iTNsJ((~#MmC>d
zlG=4rXfn?5-FT_GpKPZ2C*M|ojHk39<2$3Zu+A0ZnIuV;F7jWw1G%*<X2RVTiXr3B
zxz`y{)^F0xC9MWgZ@n!S_)XNwdEt>V^^oJT1%e*4Lxtb!l7nqztsHL2F2K(<3|!zf
zIvu80qXANUg{fPG$xHV}j3nv2>Z@8h;U!e^Jzt9_YI_0-DVr{X0~W`BVDAoIz_+i}
zuh+N3zlYlDL`oP(-CsWr;Q;~%iuIjR!UJ|c|L(pYey^!k+kzbLYtd%X7^10<8A0qU
zd@&8U*CVD%EE=3=YjJPQ60>a9m$V-C)IJLB*xbonrnlE92aqc9i)DP`H)E$cJw+#q
zj#JhsbhG56EuoRV$v>EXH;s}*hCDOQTz1%a+;QZ@Wy8TCZ%U!aImN0(zMLlA=aOc0
zC68#$xs@|drQaB1HeD*)@0QK0JTq@mlqeK5E2dCr?9Q{D!D9Eki}<`5uV)=}!dj}S
zM5%Jt!O?cs<nHzTNot`?Lmaa+uCk^SC;gy?>g}ZH%+fo9Z82Sc0dF<@s9IfKgPWPj
z=C>+^)*!6Zw-l^j2gz_9$7!nsgT0#K5X5FKWT@9h{O$90Y?=+)W^9-6i9OB)vL2-c
zJ4y@B4+}cvO#=B*?c9$GGSWs^QuPWS2t?2s;fLa90}!d90U9G1eEcG4@Q#ePj0sQx
z!}F<NBTS(DiF6KsHW;d+abWL-2&lZ!Xg>)eAXg$r;rq4^Hu$*)(7nyBNe3vq;Ul{1
zWaQi6R4UZsXllm#v^wjCAt3`xSG}QU@O(&<M#8ZbOkVEi)Phjb-NUK%!h)^e_$9bZ
zdroEKi?;MB_9b#;ZP-EpL-?DT(p0mnaR!F<p&DLrlqBMR)?rPL{~ndb!E*uKEc?JL
zuDC+C(5hxY7)7WYZ+RpMy^26Q@3f$5>U19n(jl>+tEq&R7I_~V0Ldj5;B5rLmWo`D
zgSN&Av(r4wQF062qT+-IjZ*76pf{T-Lgx`mP^JGhv(06A5-S!Qo+Y^1#HX`@LBL{6
zls44>lycR7<HqxKGTczZ+*J=v+E}-sgywiA3wtsFqVQJN6rxLKz9yB_sO5*_p3YFD
zH=UlfN-nnz^$14N#76nO$5wI5ah=0^NG-RnGx5W#Ic<<(hN@RCmR{&JWlK}&m^ZPR
zTcMHvYw79j!a<6$wrgMz@*wDS7_<TQXa-qQ=^Lqk^Byz+#7Ws<O<Bo#=xi9UdtJ|-
z#VqH7P8un|v52K7SMb0cTr(1p#@3XUxJEV&SLN00a^7*-p*6$htgk69wav5Es`dS-
z8GlKeDh(dk?g&Av_v5{UGizRP0hJ=Z-$=G(NiL;8nrY_TVNEU_6x7rft;qqdo24IJ
z0Yub)eYdU=1S(rIw}h)1ZW-6ey~oC!U%WOAB|E<^ah>Sv$vX_kTZTZS#un2sxQn7u
zjG-Fdfc8^N<xZ`d*LP6t%b}Xbg@AIJAwp~9%VJO|Zv;E}xU*5t=v4C5;L$JBqs?)f
zvi0iHkJ#hYyovdmts7NsH9ImC7{vjh?DF$}K4NO>3x1xWqMBWD>1ASuh2F#mw5}8}
zUyQXLcZQ}k3Tf4-DpuyW<*0Yl8JYst-+-DvxlyJ!oLS%7jZTfSsbC*<cm~!nBJ`is
z83RW#0E;W^*51AbO*jJa9_oWL1+a2()<4Zfq5_o)r2gCT!P#(VmS*0Rw#=-`tF&f+
zDz6eh3kPMkIw(uBJ@}wZvDbjB^;zSb4H9oEt2}dO>%~G-MLT!@IT&*$RKxmXPQlP>
z&Pr-dF&mD-FbL`!wFWg7ejcB2_*G%3368WEHzA}szV!giwjGTH6zzjvrf<(3jYX$f
zKNbT^g3vX*O#3*Sj>X*avDnVPJdOWjTeI|cCwns;^r!Bl6ZdT%q**gw`aATTdiN@R
zQ}Yt7dOYJ84eEw13QU~oFr{Qg&t+uhZ)6{MSO?%4{OWfc*7<O-?*AuebsLlJVw*km
zY8j7CWZHR5R{;bDApZe7YmD8Ou^c}X0x~d{ac>I~0x&X{0bKzmm+rhjngNHGiM>DW
zf0YnB8w88#p~jcthp{Qt?!muQr{6zY3ONkYulQ1yO-sJEkyVD<n?4yTvsw(&F=dz;
za859_5KT*x8M17hj>s!dJQltVQ=aT0JAN5yi<Hoy5;Wx7oby4mBokV%td-m>(Ks&>
zdQjR+ky3&VHkpf)cTT)j@iZttr}Jboe{vI34!&Y21e5%juYerD*}uEdZ(kP$vz&u5
zf?(<%2VQYm8;t@_QkB(&c4{Q+$_TU(YHIFcAZ1RZmh}sJQ<jsPXCFL`nI^l5Pupe=
zndAN$UdEa1>g7qAH#*u762n&8R>ER4-=2g$xHLEICb-XpX>Xt!*$r1FU`8Vse^zYO
zZ)2g%8Cec5Q#OMPvtx>YAs5iGQJF+bnbZ4mePS|&$x348dftdIIQbA6ygOe=V-$me
z;94FFgX${fyxOVi?p(TDuS5u)^2W(Z#6}eWFIOU0R35}TT<~i-Z_AzpWRwn8Qa1YR
ze#Et!qKNYCv-QKF?U9`n&MGcif2t6N?Mx%Fj=UhW#It3~a6*^>i9mM0zuYR+;|(fv
z3NZ?T#RV5NHeWqX{Q6j;gGTS`b6-;E_Oi=p1tdiWd%iL>X-AY|Et!yzowxN1oP2J!
z;hfT%5$A6|uJq0Y?`HjCYA`x&<j#hz2XtQ1mU9BfxtO)2X_lPT(j<1bYkE#;O@Gs3
zU7_i;b&h*V*(HkBr(5Ggwas;(N6Fbzqhmo?(@A|x6jgf(N`7J8$Yp|)Z{LvRQi%$&
zGX#i!X$!%?MP_GO;ik9rb&{gKonq<Xu2i;<%0p-5lm_&1+QZ(F?rYa3WrT%}4V9;k
ze{-6bWH_xy;U>4lv~$~Tr3|U0<bQ+b`Qz*!bms$ETgM9>BJW~BW8h&V$0bzi%6WR`
z@a&TdJwdCLv{wYSS*@m&lBjca9z@Z<P~a69%LY&$$3EGVZS`|WJS9=(1Qln6-XgD+
zGI!uHglYNED*VH+Qm)VaxOq=Eo!ThcQeGzuA&e=MD>s;w`kC3Xj{%HTIe*oCZdI_$
z&2MCvwq6MJdTPCaiF?A@5rCjySgrxEwL<2ktc>{}sLmpC@cOBXZPVk!p*q3QDpV)l
zUxDf-^4tJW8D}_n;}#x0D5E<)H`lIduKM-s9Uv6O1cps8gaT)jI3c%+vOq#vm|H<<
z)Y`?U0ZyGdkeofOcE)!8fqxJRgun%P$93&qbK0S$j{hpm#9&~%_mcGpL5rUTLD-5Z
zEOhYenW|EXm-3B33@&16f%VH7Mp}6rKxj4cV5jwlG+f@=G7ppf=t>W>Jj=RnwjQw=
z8lfbbY0u8^nq%pOLT4E}v)*xLJy+Jj5z5Mu<s~MR)~4gbJluq|4}U=?H_GTPy~pD`
zA1NvaBv&q0?ySoBkpNo;@_TR@>>W|ODg0I(Zgle=9`jv@4EW&Ljp)>7RK>W}`o(aa
zhg>BJ*6DJBUQ86;DT^JoE8#b^R(VJ73BMU{SX`3Th11G2Pxr!a$W@9Z{|@k*pu7UV
ziMKbxZwQbZn)~rLL4RolQ4nuG0D%(}S0FgK{>Nf)lGn>1oOpZ5lA0yM_zu-=%0ah4
zG`VCIY%P_JRWb9oZ&GnlnxM3R85k~hob8l%xEcNvl<q-j3dAKfcn3mLz$>YnFL$6w
z8J3bPF8FhSoaFMyMV~eyDhluC(%dPT^2uz81J4d(%zH;wK7Y_adN|sUH9i~ZqDr#0
z(qb)u&VtH^V?L=K0emFpW56darbm~q0j96SeIRmysDBskbFDD*y=wRne7JmAKyy>6
zLC80OfKqWw_GQ{j+>ZGG*Na@`d<^EpHMs)V><|w<FawAjW|I>$wTJanYy0UZ{Aq(Q
zes3T8IzJ6~D1S9d^O@2z*m`Kdzg+u$W>a@KZEemR9}x!~9@`-zSDd_YOTFIqRqDEP
zBpNbSBvxGL41rxbgz+ops{ol-Py-v3=lBp*vQ#zM*owoQtM@NAZ5kD){f_#G?p!`?
z#mU~fZnhiwxii9+QxvBO?6fjzgTqy*9~w>;W=*;HrDtSmeVM5Qk!q6E#91C0Dz(K3
z67=fPfabzf5(CpR(*r#?Ndx{2wEBQETY)Qc@SRgXF1UzIf}wNCw!J*Q?nozp&}=9>
zx)ah&!2hl~;Ep`E99>+-dpnow#!o<EkLD8Kf2J{ERhO|GKNJBnm(g1`69X|gHkTm_
z0VbDDygLqm7H+_xftilmKyESS7UUGfKKT&;{~<++)KXe{9Xkh2_fS-^HdejImhHZE
zw)_0_`u6nkMBD`v9R%#&zIGrtto#KlgWh!kz<7f0?R)p<v$g2udX<Z<C6X)1_b8w4
z@PQ!843HThqX35wfk8${Mo8wMlTRxb%J)#mPW5?zFV9EIoF#~CpepHj`0MTWCtDS^
zCi(m|f9U@{y*>T=G^mecoKKx|(Qtt7@9$6f27@y`B)JX7F?T=y>ApVwp~bev)MxQl
z1o4*j-np5#&>LB*^PUWa(7eTB35R-#)hre{`E(#x{x0Lr)p^G{E@Sq!SUpH*F($aq
zJ2VV`u}PN)i?kwLe8{KXSBB-W;m?n6of`S_d6ewgL=IXq{Fbko=Z7bPSm?Zdf)={S
zz-iqI2E%@EQIZE^Z+dpm)<a~CvqKtY=jCNwxZY&v4N=k(S4_4>637~xWuWV&&PGJ)
z$5XPa3DF!oUxk{b$Ce1Q6*to>dIPv{!l$`^tGwJz)Okpf+c=F(*>T6~UDaYemk6+t
z8Nqk}JsV$Yj>kUFUCod@ma7N9)^Qjty2tT|(Rj}bEmEJ8B}c0kkaDHEC;*tt2s<q`
z=c`WG76luPi}c7S%q5<S!hCkFv&HMazuv?1Qd`qv-8=oxQm^B}njOEpB=#de(C<or
z;u*iI**K}{E$dyX760@=zw31$<98)>&+k&AM9L%mK9_jT@AKIiQ#kHhOY?Qk`BHQr
z;d4PnywLG(uS;jdprx*Py|zWUOt5}$W<@4W8dBf^nb;PWHQ=*}2^*-wl>O5}O@{*8
ziove+<|4itx)j90p!Z79912tc8=82RtH(bHDz6?TxgDv86{Of&87&%!+6Pnc=J4<V
zQ!4sYCrkT?m=j&DNV0TEw#Dd|-p4;vf9wv;1Mc;rSh27Sv`)Ko+S12(`4+_%FD_tg
zh+Z61e>J9y%@4A7)6f#8fWLXZNH1h?(5Z}s=SwNj@I00nw*Oi|Z;kXg+^JGa@oid+
z7`x7D@=Kw?Np{!?G(C>`)t2+|B=vKr9CAv9Oio)*i$hLvns~>rmEH3+mN^;ee_xu$
zQWUMuu-FS&D@zYW!bkE-VTU%dmW+12hgS@tX0LL}*P(YUtnl7o7%*bnOV?;kPMj39
zq%Ey$4_349K|Yn5Dv4Xjr+H1SpT<_4leugxH>TjS<_zZz*XW8;(!njP2VT_`k>&Y?
zWm}PS&Jc_x#@`;vxukQERO{Y+e|odFhX!PJm6s_P!NXd!jkB4CAJyzzqFeF5mJ~MD
zl_p4D&%R*?qGJPT5eQJ{GS>*~iMw{7TQ|g`P=Lky_E>`V!CSm<gJX<FklaF>^$m^*
z4KH@XIVOKYw27<<i?k`nu(=D;s&<q&4biHv_Q-`4t!OFAu#mjy3+b}he=E3pX9s-^
z?GciR;HOZ&Y}ASqcw{7ttzVD1NqTs6ep<u_MDW_jJNNM^Xq!@cZj`<vsq8m`v0O+C
zg((1uV~qyb1W~&@L@Ft{pcH>O`3AHw9UMK+4_7w;&ed>)1c2JOI8{e92_Ao_PCgY=
zcM<1S*}dWTp+8VitL`0%fA^=e>Vb_DaVa8jTTg2p_j1MBS*UvAc6Kt^+E?hEgt`)6
zp*LA)kZIx3wxL?ZYx=$1ah5?uQUC_myepgGnL`iY$0oVS`$au;TFo=3Bghg1p|I+q
zR2UCp<5hNlWe=HL&ShLl*)g(-^7LHDg|v!5wv}MUccfJrZg^pCf2xjg?k03Mt-?em
zr0@>3N=ZILtJLf(X%$8^k>Uerm6CJ@jH}rXpjArZ8Cs>+|GBhE>$*>?)a<QD1~4(C
z2hb`d>0Bh$y8j<qr8GN3tJLf}XqA?<Ppj1I+i8^!#Qn5NJ8+*?X|o?qtF$(|v`Xjj
zyU{AG)n&A5%~u=Se_9Gw!rRdC9FqP5=7#8GZ_yRZZ4)+BCI>=@Jk#ur3qLZo9LnB)
zpqEb*I+jry4`<Fa;db24!`ts$`@b$VJ)#6K;V)v8bjAp3Z)Chw8Kv+Rk@aMK_oo;j
z<1s25C*+evl*>QHxZ|kDt!xvdY666x#5Tc!p)}^tO^$3{f1Dw!+XxDXxiR?}c@q>5
z&-#^`fs=6i))K?L>}PC<MmVfq=|0Q%c|k^GuvoGEdRn#j)AgFgqm|<?la*aX^=ie!
zJ8^m(avWz*G2MLja+Oj;YwaA4Q_jjcUsERA<MCBCNr2hiaXeNs#QmaLhV>oO{B}io
z@3g)*N)<rnJO8t|yfn2ru87P-<=-lzOvnD!)&8~B@qQV1<yF+(+846_dG3<wx!Vey
z9p17E3sllRrSkm;)bYyQm$4i_6az9cF_(PIKhl43`@dd)-`NV-n#ie{UVZQ7b@%UX
zgpba6jstr!U}5<C`!0PT@P;K0h$aFJKmHlM?*8bxHd);}D=UC1%QAbvP!_WBJ`Bvx
zIFV6VRXG@|!s>#P(g;i6fzl58Zex7u=XRY>xX*gTqVuBfX-KB~ti!CUv#Ysi=cXtW
zgW!J}^bAq$Od;7-ltRb-Lqb6c&Z|Yev@c%eQ*8G4iq1nFr;8#*>3!+vwluDx&h;))
z>Zm#*(|wXOj%<ji$xhV5eUBO0h&AI2g>!xyHz|^6nxCkLvDdyg+^Kym^VFY{c0V1}
z&Tr|_MuTDWfCQp}wson;4E2nuuDi3tV4Qy%vUg_=CPuj8&&KMU^5gCB<J-_tzEmHN
z_AMXxfsKA^E2QU1P0hpz-KK34a%&RYC@OfA;#n6MPox|s{Y}PPNV&K-l@nmkFvX$K
z8=uM^%FLx~zvRKqTe4qPk0367#ul3Qaq8jcS!_F0<6hVIRb?|7A-O}@7uC=dyX1d(
zf|(My_LDI$D05SB)X<7OM_A$+M~j9=&nNPs8Wg5g<#rhN_Pw|;omj*-;zHd*CtBGq
zaJQ(P1Nhaz?cFf_<V7whbLBS&@y<vX;|)COqa49j6&|Y#*M4#9=DrDf9>rbIeJO{w
zue&Y>Ni5uyY~i`v-*@14)`YYj2~B^%IO0~vwlC|sIxs<C&K2x>uBeez-!rZ`6qMs>
zHFrHQqIYxd`S_x<QSLnwD5lea7Wld1-$+6TMm_nwhLT6FmmnI`gclN8n2zhYvz`$k
z3+~72^N;JfdvCyp^W0myyIu;Bffz4Jx!(q>&}fE@0Zu&PP<`%Dyn_zw*R6ly-l2O_
zQtP`Z6)h2Qy>x4hW4gZ150UOd)WdJN?|hB)T%Pe1c;3XzoCHI}H%p}T5?KrdxkX~q
z_7-c@RLNtwS>gt3B%jm!xx{(mnWR3gm*U9qa!X3hvL5TwawhiHje8OwGz(I)Dl4$Q
zIch3cV4VB%Wz0-3-kdNlTx);C(-grcTc)p7pgJcYZj<AujIqy8_u~b4k)oENSCdA&
zq?P2cB!*Y>$E(TXrTe{;Xo3x$ZXS=^?#js13Vlrk^Ri%Z@4fq~syNjs_>(FxNMgW+
zn-lo6wm4^qF>$}c7VoRTAm{|44Jt5pS-JS1R$?GJV+lU06vG;_A@+an4Ui(a{FF)z
zh+ch>Ke!UpOIcls=@|YZl^CF)Bg2N3n2zHHm6(p>l}b!6|Kln#x8W+4m~QNLf&6`?
z^!qC^;KX>Ea=F7VP>sphw=Tw{P&oVNXrp2bSWRMOaow})6`)6j!15LctC<Eg;un`Q
zz|~9zjKq^V2*uf*gwG7zEpJJ)^qBh(2N|80F-{p112Ho)mk@jn6aqFhm*MgkCznmC
zIt+giB~cHcV4&>o3Xolb+67q!nOV&u^Z$#GdR;y4%N}=vEUIkF5=Bwe;URS`FTY;o
z<<D=oueU#aYIAYIBRF;W`sD)jg{1!=snO!a0SK$%^7Y&0@9$D7OMf@|Gg4|ze~#XJ
zdd2HE4W4=c^#E!GxPEg6s6|tYralVkRnmW-qW3Phedh6&#(QZDOCZ^xvlMgv$Jbxp
zWL8*;VEHY-jGumdz5V;Gs;{&_8!py>01KCY{`SUS7(iiNxM)u+HkaT3b@}D(ucOpD
z7wXIM@AqFXL(AW?0X>ibrL~Bi0ksq!)WzB$G=G&02;U{GQc86B6@~>KPdd$jUTA+(
zy_tb5EzSKk(v1^FyGc6he%sreWlm39P0bpuxAba*RC(^TZ%Bu1)7dG@u98zKr;AQV
z4u<4ZBu*kJXXvx>oo=IBV<0t6C=#@6cIB2_6m|Nnq8;;hWrmRYr<vH<X$)j7=+%5)
z+4D^v%3Y#(OkJZ0Zu`i6>>yrcQM!L=Gf^aWi(g?}-`4S^vlzXuU+2dsA2V=#ki!-A
zd$t**l|&Wp%PJsy&!dOVI4gp)<Dm;dSY;0m-SE11z?2n;{(=}tMm=E9*|s$-`9gB+
zI+LMwIWndRLdts3xFe4+SvqZcGMAxY^<Q-KuBAO>)(CIhD47o#d|!89r}=-lj3e}<
z3pISMuY$mj`qfoLTRg5gDj~fYrLAbr$C`#qmZ307?ONi24{~*nErCSP<18Fe@ScGP
z^U6QFSvS4LTS2Hc2;4H2A_=~tL_8<?7_)n!eSX7x$*(qBagu(P1<4*VRWD@AR6w08
zBAf8IWhl;gJ;-^rUC;`x!!Un|3VSek&Qt`m+L=M{mZ>BPaFtT&AJQxRoXyI5dWkPT
zgxP2NSzcl`;$gh$y<|FKTedU72t`gg*v^^f?3dwXkot1_1p!2ChFM-<KP5G}n;Rvs
z*)Io6<27l!c<6mC!q-_hQiBra4FiL(dx6(T-?xKr;U0x$5l1U6KJ|a)mdJAOYGW!O
zOGpEKQw^DNbz~RH)95~&XP&gqD3t|)2FA5CL2=6p#6N`z^T@LDxW@gtgA&%@Bw;<u
z>@hv(+$g0`&&XnZ#QbcrB>M{XJ<Inwkotw^V&#cV7O}Ms%00-x?X-?{dbG6;13|cX
zC!owt&b#M}iq04=bf|xvE7m?V_PjSdpS{oc_r&J+cyNjHN9RJL@Dv1$On8}exk+3*
z%kyZ1!gx1~j=GTy%C!nErKDTl+t0g}g@8g4@g5|{{qg0D@_xtDi{)I#Po}=xM=R=u
z?a4N+%#!PxA6JbTA(Dsnb3_CvDbq(Hv5hXP9*vDhyG@VY^<RGj1mn~++UIuQuTCgC
z@0ZE?FRRC^g0gB?veauVmSpd0?R4wbd}#H8xW&uaW;_Dq5I>W5X~G;IRHv;QF*Wbr
zp3Q(#JV_b&An(9t;HB{67%mhUr87UxQG9b=j&y^BmOX8vwIqzf)#azn+68I?8QSi(
zX&d6PIkh483hRG!lWVB=E)bR?f-Cj3=<eBt;x{3~SlzZ7>^A|L92}&XGzbTQ#R;t@
zmziP1l@g54+Ug~fUe-};jrV<BJpnmo8^f#`mnzC(AW5WdT<vB+jEC}GXv98LASG4`
zChhhC*dQmHVa4%RGyEJzG*?j&o(v=X=zK)}a44Z&jfH==UV5f_WKYBb=;_@NeBo-)
zSX$SD2AXb;&7hGqSle~PQ#9y+^e6#gvcv%Z>Odq%`pJH6&`7JQ$j}E3*9Hxof(DJv
zL>ZeCn%3Laz*<`5n$Oj4k+|(cF=``_&%&^leN?IbNl9f`iq->dE+?}MopvGm#!lx~
z?ISqBzoUN?Y@Bs}w-kUNVmXd_C}1f?^g#JR#`)Mm>50TqgL>Y6V$7hFT!uBZPlOGQ
zHG`#pcGU19NEl~*FluN@V%9AEil`y=WRt_F;e@lW;(fbOgX_Zey6NGl(T#4r^aYWF
z{7I`}h9+`^it6E1<Or*gBkV?weva+P5jG=7ctwBYpv1xG(Go}L2BjH%aXFZ;3moBW
zwt;>O999;C6nIsm<dXqpHS5q$KwDcmV)=*yDWr7cjslBS!tnv<=K9HDg%X3nWVM^~
zAgCxW9N2!$$ts5iT4R9j(kEZ`bi451tad1ZUgPzXfYXoO1R8+l?;+3$A{aZ)G6f^A
z+RA^2*99ihyCW!yy(TDHS~C=x7K+l0W*;bYpk%{L+CNwyv?zj|Y_JM%S*ug^!QFsI
z)X-WzX?_54MhODPNz)eNl2y-;<G7X%<Ie2n8hR-rxTu!O>Pofc`8o;(161P_M-U<v
zMUY9c$9O4U%Jb3aWSlCcC{j7t$t)*pDTaUeur+J*p5eo}T)#GV1@(1a>|Vr+HN?;~
zYH&b`wU5DR+~=RmikZ_|jpMA3uwt#4%#VwG2`P3&Xsn-<J0Qg_2orF{_4i1z;(I>-
zY(>5w-IY=pA4@FaN8$`FcV6Yel7cciv@9cX3&_u_;ExpMY`VDnW@Y^5H9;`d@$7%F
zjLN(f<WNNK;6%7f{l{?Ew5~AkEZNlAZ1eb3hUY9{GScA{p0QfEX`5(k4jf4eaDE(>
z$!XE}v$>MYS;C2nJ>+7ugh`{%C5SGP7Q!R%yBeeM@w#x3I$e}!r{&&StzS#HTqTMc
zeU@5+e{XQX7Q$Rxi}vR;9Yr)Cpq_tp`G^<y33=r{-tz+v#EAz98??T=J}GsOLY0ai
zXVfbzJ6jb?@{Us%I;d%vhBY6=2gi>ko?Yjw(icMnQ~GIt*cVUk57#I{fKkE`98*-W
z;aN0p2Tv6Wb$bAin}<G`J{C~C{R|N1=T51q)Bb;o!Pu?7Ed9V=_#SE40y2O30fn#>
zvByCz9<gRK9t~@(5h8xC^4Unl-|GCRNV}Fg!^e^&XP&r<mr4q!m%>UZfyjn~!9z(9
z1YWorC9CFiEpBGMA7deNF^*Whyt7!SZVh7C)<%z@8()><q{E4lu<~$H^H-spc9ceP
zo-)jO9^Lr1hO5fsYR_dl{D6M|Sid>{@s1nH8qn$4+UIx?LVVPH9DQe><CCOl4N5K^
zxUY+sf-=8R+R4IIJPT@hez#_3O-W=aPO1E`SZu+kid@^?{&0bdngzhK;i?~f1eW|%
zwI;?ijH)PqfH>9*?O+^T>)hXyO-h`U_p<k7@azff7csFp<f;X|)cb$Q-uvuUD<ibE
zqwYKhK}Rmnm1I-77#m7N<k;Z!N<N+YHFbd0uQFWQdP(WO&32CqUwl`CYTorHF`uRv
z;HMoui2w+iMol(t5uoni{^>^w;%;8o0ng*dlfgZkcK4&toE!(^p#Gw?l6_e2I^>cH
zU|Y@bgM%&U0)?IO{d`#QTAoj>bwIxat10oS-!}vm+vK<Tc}mm!dnG<gA<H(u)_MK!
ztw3`q34h@CRsdYM!~f^sTcvI>kc2#>+wHpRN7Z=#5maxFCx4eU{8sA6*V}&p?b*?n
z5ef?v1Tix)GM7;`F(a2Q%Q+1JHkT36Dk*<aZydP|e)q2s(3fk0Xe5fHq(fjJcWJHx
za&3^Lz_o@S9D9>Eb#{&QI>`O|e$=eT_U*bI?{4C|2iu%6N1{mT<44WJ6r;M}Vu~dA
ztxd{~s5YrM($XY$P&Fw<$DEs#oN#+K1t*zJ;_9@v|BAR0@}50gUF9}j!W^Mc6ES}k
z#Wr6u{iOmX<D~*_XzT~YB$=BX3>!+Xz~m@IX<~AOsckZ57{62yN4!)}N3t9NtJB&h
zmgu0`?g|Dg+8d6+YHaHg3|4V%kHHnSe@xYU(PqoZrKPVlKrGEg8eL*(rYpH5%e&Am
zp;@NzW=(BexR--ABeg=TZLl01+X8=0gui2vHXAv{;xKp>NVy*tV(R@=Aae6&tQbK0
zFd;Cky?ZkhN*~ayPC{=@&Q4lrAQmS{`pGJyTiOT(;nmniz)~bjn;1@(s{P&MH1>nx
z3583VSRktD21G?Ua1)`sOrU~f;Trbrt~!<Gk$|wN<mTyu$NGs8X{Wh)gaUtbi?k^!
zA>C42d`&KgK7y59rrvZoDsyi-f|o_2tqmEPb8mVrF4x|4xW0tu2MWm*Y7rz@jT|wb
z1<9_Y6x&^7cY<f7Vywc^2FxaDX#WnbNFdBb5kslvHAE=%c}xLf3B7J{rC$RD?38{D
zgaoWqnladSRYEU{tGQ)tKp1~(Uh1l=C7K_jV1$B<&5B6XQv0XK)!Ooc2vdfTQ<L#K
zSnN!tnB1#d^a66u25DmRHJKL!FI6yD&=mI`KX%=Zv$NU!_4(29#3f+KE7!d|S}Z2>
zsiXe#uU{5Fy<QwGCNB0jp3SC<$B(;bw(5W&<7W2^DH{dOXcx9+h*N)mAu!HR1Q+}J
z-OKsx`1NFQ(7k;6ta~&0vN+t|haKIUU(Y9)`TYb2O&62t;-UeUwlQ8!E@qeW<H^Mm
z%<SIJlate<A7)<;0xlI!%9Rc=!W_?HcDSsc(FT8d&_kp>_shTjZL9YMK#(cA>E+qk
z;azGOYQUkSp5XOs)t!I2?Utu*`EVPzC0cb`3M1SGHU5p;QXKpN$hz&Zubw^kUb%$T
zm2<hr%DX4i>1;WcEM0NXmq9bXFN0T-4pJVZGDyP{90n;45)D!sBp#$;4j$&<Hpk(x
z_GW1J`tt3<u3nr@|KqwJX7iKDysa=Cc7N-hcRwD){^Hf-cyWIK1o0IEY0N&=R_J^&
z$foS!^26?l!|<=WpJs1n-Bb4+*MfSHB?IyTinPS~=l%T~0ocfhLYNHoCq-WYV@ZOy
zN-rF}NM13*m_|WO2VSDAqKMVJ11plM7rM5BHF>KMcUh5D71M`O@LwfDz1!Xh6b7&1
z+6bxSwWK?ZU{Zer_XySk+7XzlZf^u!>W!z=-~*Wh5^MsosTH?*#fBoF37P^)Kq+A`
zhr9zDa`q8L%l=lBR~6WV-2x8dv((j`KrpE01OurS<rG5L-ik=cgD4;u#8>bnY&xG1
z4$jynVRd_ehXE##4T^%Egux`@4t!Cg&!~J?RwzZE@b`Z%E68YyAT`)X4PbN^nc0^}
za9S?jm>d|TH`2v|nwaCND?mWln2=Po06+^7-T_o(q(Qp2f&+3BV0)JpUne)=iflNA
z&6<Gu07Z`t+p+@`dQEPEWGgZmj2E;N$x(I?4<LdyR8*0osF3I%Ji#*w9tmr9X=*@G
zKqrBl6|H}c(i$Vy&x>O0;#CCTX=Cn`urZ4@Oh6_?f;fA0jfZ=;*3mr~hCg1M&ZaQv
zpRb<V&vzddi}Q=`yYA!V+3C^A>AQD+Hh&M3_O?wCAVn|^%|5nb!az>g55FJ^_k!`p
zceswCiMP~xnuwTtK)@JPCZvw&7*)!_ONd{cd~SckH#M)aE|tw#S57xrmk!KH2j-*$
zbJBr1DX-y1Ba%A?gOOHt3B3^kc?;;G2K+60|Kxsuy7+MU)*sJ4b)U{I-=0n}&*#bf
zV)7ITkAOk)AJ2A@^@43^rw>SXG|R9axd(uZAP{Sa!^NbT{1+72uo)15b6~?KXEP^j
zxrKi~S?q`@wyLy^uRdOWI)8LB`P}_{wD|Ck^Tl(d*8CmP8_w$9p3UBNpMiG4$L?Z&
z-2HtvJv*IFj^@`A^#1fBKY9^)2fk$xAUlmI#?7#;qxyhLcsSNoK+$5Xn30SH`zJ6H
z?V#=%s%P}@D0jdFg0B-mxFitgxQ>PdDiwc7D3k=$DHOd8YN~fPaVr{E$fE!$Rvn|t
z0w1Mus%WO|z-WwWkXCC(s1jh`j(olXO5Om*?RwL)<}jo1W+++14V277-<XHKF%NxX
z9(vO}^rm^}P0N6=WkA?6AZ$q+;p!DuXtH)@fC^0&8;c$UDr~xo1{3IF0TfbqfE0fT
zq$pS?8>_cNa!Xir$xaAlS<3204N^4;7-0fk)<U4GJK(3Kct#>gKovu#NHWS+U=@&A
z%Z8oo_$V`gvJekpW>nKe^3Y+Xh%i&xL6`yWo~xbugAbwtvQf)yC;!05^leTGl4h?p
zUD?14R*1{HY(PBV06<mN96)hk#V&siZje|2Cp$W64Ut+yR9wo6hG<P=icTqrqL)Q?
zoJ)#vOXg9)*5n5B!1P%kB+ow0P9|q9&x9Vm&|PF%B9t;z=E&;pP%wdvlD#4T;i^?u
z!6c#Z1xd&<;7_P5+Y!4M(c#%UvZZ^YmH(O_dK<@pZ5;L5!HoixafkV3^)P>5fz>6Y
zQK}f3n8&CBe?5aFj8G-S*Us6-sHWtrj#?I61h)FSZWAf2zLv0eZJQ7vdyRNafa*Y4
z>d+F_ffUt&6xD$g)qxb%ffUuc{Q*1MSskWU0uxZAMfCtcFdz1%@|vS<)Rj^=Muvwa
z2Xs?gZDO%{5c05z1^r+R<$-@O5|D;X+?KX=Ek+Hh2{j1q7)p;FOIcWrd4>3mYO%e*
zsMv!kXV1~<VDK8FOR$aT>(@ah0LV2^6BjS^&@m>|L5@2i)oA6Cy|fNK1A7>>wtaoj
zD9@TL1mLd%UWDG4GWrYJ#XFIVRs}Xhbleg_=nJw9HXsDpuJ`M}H;I37MPotNRL4#;
z(KbAEEl$afSi<dEY#(?0@56<g8biUX34E4Dsg}^XtP!e&#>>8R8lxKRm8XnR1=N%%
zf0)<*)`70n*+`q!*_f~EtbO;=*GdXykgkpr?4y?}DemG`wfMZ;dhKhmYzQytZq-hD
zHr;~v1tav<)q3fr6>NW}ot*EqV%%xqJ$h?<HwzaJy|wX*_`pz~3t%9u#rk_xF|-#c
z-st84_a#1bHy2~Y+7fNIHbd`JZliaq53;n``DA{yn9Z*cT-G6YOk=vK+iFW}28wdk
zF{(fX)bBDz6{OsUQS%!#(;o<$DXxL$x*0UnjW=cX3E(y-wS+sAO*c~3UMrX1MapL5
z9N{<pzkLb$e>letjUfY(EtUCO_Q<zV=l<DJR<mXKN#wn=rT>m^fB(N|((9M8O$`$T
zF*7nWF_&?<Efkl0sxc;)*XBQ&0ll|z=RZ9L0~oVYmk~Ze6#_CfmjV4Y6PMfVKf(fv
zKewOmKiUKY)HZ16m$4i_6ag}q0ggBnm!|VS#R2}8Z1g{Ie+TBhry<-U$haPJ)NEH9
zt(A~*A<5RXu4rV;WJQ@hbj&_^n!T5-`qX=mrA!>9EaAJWBvtLN)Iz)SG)BR8b}>&o
zTFbmInrGjl-R^q>VnRFBBhskHB6h74TAI>JH&N-&iptrt$)2u>!t85YezN^j6Vr=z
zkUXJs)Ql11e>BfP&Ni5}7;RKMIC^c0J3af+`i)hz7@s3zT1$B>`?w#i9x>-_Z%y^M
zAF1AnAXP7wJ@xT<;Vc`Xa1i+Na`-VI9eo;7PCCHhG1%e1LuwYJtcimZj>tQFMvcGW
zFLZqO?{8oK@#PN^5c>~zm4gw=QVhe(cxw?f#vs*xf6w<&Eh?m75)t{)KnI~B=i(wb
zHGCNk26@_OrK)b<xm61v;G;^ItZ>07Jr7Vd6&(^{X)WV@uN#Y2^AKrHlNx--C|Q<o
zAz!5UmaqE0bcNJEhr`$5@Ht!6I_0?wkHSdN;cv~V7@K@$pqIHxdl7v^T1%5a2M4dN
zHa4@(e{BkrcF41al!(#$I7~xL6D}>)q)QW^3PNcrpMCp=5xR+@Wr|QMER?o_BM#w7
z*|MW`=kH%$ve-8WeX2f3m^;!@5RmSiHp2VBScxCNvupeycQ3t57NeX&{8VSkNTr6?
ztE&USke`Q2r8Kp>d2XN2H3ytxtvSOSJ+Y|Fe;^58++>Cgnp9?hRX{Y;q^DfobWV;T
zC%0U<l6g#DyvAg1WUNx9A=#zM@%O#rFe7Y}lswjmo=aWQ^f<{+t{nT3epHJ-b#V-H
zLAKSHsrNX}hn9#WbT?8(7kyI%yAl@XvZitylN_u(zO&0-ndm^zRM^~13e)ve3Rqj3
ze~UM>?+Ya;TD3&xEFxk{OQ_2<qm8XupSeWpcIK3$sCcs}w<l=rHRTi(PD|Tc;8Zz<
z-K+`=yA=wRG}|8+^~nX*BhU%L>9jx~E~Kd0G^{0@d2z2KoH=`p3FlwwKk87xo(0)M
zTjXr6@)3*xWoYA<U+c7zdBj~sBvM*ge>cZOJ3L_*7AJ$cIc$<c-IQe)JEXjSvlb=R
z%q+7C2@V!r!pi5QRgzN~q3oty4o9kTJf+1=_tkm|Gwq{%UvJ2ann`9M*a=$I>uH2s
zE~(v6P-sLbwDWy`oR+JK50<&?_L7|g2MmMF$seD(I|S2@1v><mFe*8fe@G+=e;mPk
z;i1xBe-2*{pL^8oul{om3hA7+ZBRJGC8(SNS%`YtUR`D5RRgo#<zt%?4W<Zw4GCej
zxnP$BlW?gm|I#{XTB`LyFcmm+KMH-|YdyY$1KaKe^og3MchyccWo_~%BENaJvK78)
zk``N)V0yHbNpb>Vmckgyj;Kuwe^B2Yqg;&=#G*xXbffSgjLocs>t%|LV>9Yz+>NI0
z0JlsPkjQD?`yX?$b61!P`x6~|#J!ksMg-$~2__X=+H&04&_wkswQUj4(YBRX(>5=B
z;~c_Zss{3kb*8b4VqdmMw+=y6S*@raMdh%&B=kx!w@B@&)<}6!()pK<f1?GRT+8Is
zXd7g{q;pN!_MLmGxJPSPe?AfF5czbkS{X(6F|FWEHL=K^E<Red;^yg~;j>0sjQ*O_
zL4|0AmObVfr-KU22hQmww@wF_CG5=8*Yi&RHP{sbqo2wv4gh7uKGye$6Ro(MK%OFq
zeEL$C<;w}8jvC4G|A`-#e>uqZ7;*B2N{|WeG^N?)jNppKbSC&olu=721K}r8M&2m#
zzGsgzo;3TK%{Qp}CXLkKK<bT2WHZKT3fXjjBLb-n5`v){5y%ehjR<5n%IgVa2e@S_
zP!tu^4GCn2cm^fwBz6d-CZ;<2h6J)hJV)DBVuwI#83doUzn(yLe@JgcAiGgsK_EM%
zHzttED-pH2A%W}=e=>pWN52Ju>@Y7Nke%cvfjq0A-JVopo2<28%<mPfQhO9{+ataq
ztK^pyC|+{xxcQhUvrMHTKPH=2gvn@r_EG&jghyZ!VbVS6_aExDTG7y_wdW}CmI>v#
zKq03eAC$YFr5QPAf9XwY6y5n1nAsXXnIi8ckLG=otwqBJWTBfxMRK)&AZyB#3SD%(
zj@1%2KBQgGYO#Aqt-Lj<rNDe5oReh>ecNEA<WZ64aF;nbW+#?Xb4i)ZrMU&(tL0wv
zD(wXHLX%A*0(OtIGm)z{^y{#{f>vkM+7Wh%HKm2qN${dif6TE;eMQH#c+N-)s(g>*
zqKTK}g>B;*umj)LB0HOX%H4k4{;a*Elnh>;^j^;X6jn;?^|&#sQp%*KW!)$`+t9d0
zQaQ!0l!>TFZehQV)r)?Yb4-PF5y238YtU-2DUU25!|JSdd{i>^<l{%?e4EUfP=gKW
z1#`ZJ9g-vwe?#+1W%I22OlFA9AQJZi?RDjXqCq@`9(87X)$4mMquY!=)sOCVul?5{
zQ83J6yg7Pu5j>GPxSsN>r2I;1Yk~2s=rSW<qkv6A+%ZDzUSg0ZC?Qj@xHid^I<rXw
zNsAa44u3-$7))JXuQ*$T9ku$qo;_bzZpL_7tvHvK7n}VU!)8YDyP~y;&ZT7?R{|CL
z9}EUTmthnr69X|bGMDkrCli;N@+~Ttxcont0fe_u{XYi*1cJ*b8n;~jKQ9FYmk=OS
zm$4i_6aq3imjQ_&JD1=CK$?Fg?EifG?L{Vm^Rq~)QFv!|))_&vBq@&)yHTId9v0_l
zP#Euq(NaT|k!3R~&D$h-&r=)A&wbU2o+0T3r}y~V?RomjM`6|!O+AHFVH5(svO*pG
zLK4G}%8kz|HpBplv2n2#EcZI*{`o79LzQ?hD?m#nBpG?&jWjvbOq+k6G$CU$wdoz%
z#LU4bIhbQe{cw7p#zX~}q{GW(2dk}kX0|zc;k3Wk=H2LJo1atLA=%V!8WMsM4*f{^
zLky|V2p9Co`&vVq^VDVz$#{0nV$q<F7|m#5?l)RDdUc_4TMRV1Uz9q3gY6XwR_HLw
z`ay<tUKoxBJ@2axX;XijO#Pb`iTN8$tMVOIr1AppH*Pn2(YPghoYMP3)a5KQWtD9s
zE0Wpxk?WR`bBK$p#dlwL?SyT^dMUJ<&$5{j8&EHVQG*jtwvy@`jh&1Lu#~51WsaXt
zn@Z4uJY9{3Oa8i{2tRq#<Tjfe$cM?eIC7AO?NO8yX$6d<&=!B0Z1wLLtVUDhx4G;x
z^+{S{#R7DWI&*7K!s~FaTX&<E-TDDHmR1C!K5*G`;|Q;0aw}6UY20<utZH~W7NRR2
zGEsoMe9G96INLVtY}x{Y6L_~R`q7)Vh{@ky)PGL;1LddXin$+eXAddB;Y=D>3SK3=
zuSN%n!dd94fj)n#x^gC;ao2@FO*T|kjnb(Ut9U)fu$Q7s^vhGpS+^u2^h!`zyVFV9
z!OK3<;tDeb@mR#vVrUy*P)TJ+Mjr$(v~W@ltRvZ!`uQ3Cr9}^Biks0guEK>I;Q?BT
zRd-vZ9lUInrp%01IHnRczuW0ABEc(Apq2u8Ep?1P#!7z_R4l;2ddW)Ws<l%sC1~kW
zJ|!rE)?0eoK;H0+OtAdrdlP|`7HGo`37I;jYEDjEGK91xO3h+;+EPGe1lSG&m3bW}
z+r;^}3U^wsGoj7<2{u8UE9Dnxt>LR9bJrWf&8lm74rJp*4(9?-xbjEd%D__v*coMr
zU|BKz_%?sMy}rHtJ5#tp6r|qWUq4^?g8>xQksT;F2iX1kd-vnz_mS$bT_W54p6GMZ
zTEuRhDJzh;m|z42sLs4DE%%bD_n)kU+d_9;&pIwE<F!y&-3WSERvgyQu<}oI)UKs<
zP-_U$yOA<{jjOiXMIz^>;gV@6oTzp!-#H!yxu}09ilDSaMlSj?lv&k$@WOjN)5vRe
z>#P+kmv!kVnp5?kN3N&(+w>U(Pz)HTKV6?L#=3#a1Nu}_Q_<jNT6s+xxL;O90ZfOP
zMpS5aP*Na+!26fi$*UcKHa${JwoZbfoJ5TBb}d`F&1>3%(7Gm)ylv+OD=vkQ)uf)I
z6Z3!YC0^(@D{A$n*^<D$nsx*8muS}#Ah}hyd3D-A%zew~9BuYWeW?*A&neQhg|{hG
zy9BY69db|3+%dl&mpP58mG`-gb!ng5_{F)DmdBD^nO9m7QMbMUXHelmeAXD8XBUb!
za!b}k{bIQthFFvB4MtdH%c_zq+)7a}y+wa~bWAOWjq-e+2l8lBymoazrt4qlGsz+g
z5;L7tH`%*If2~h;3I2AFwTga|)-{z(()kBKm!>$2O<D87pv$&cv_gtVvaq`fboCQ?
zFVLk>Q(6im=%Trh4e{xyOB<S$byuRU>y<eo1+A^H(s!BQ%F_E5<lwCUH8NWO4i$d_
zaZvH&hzH=MR%9>}T7^#4>+O?4OUW~g6rT-)5_lcB==OwdX&+?8t&#x-CA{{kvgFHz
zH#A4|FyD~AT~i4f_XO8AR}=2*JF9bVW!-`Scn*l8r5r}K4<a~2diLX7NPH>J%PJ>`
z>gu8TGj!flyrlHLHy?23^;MEW3jBZXVFBfZwPq&8gGN9vjKPuWh81|5#X0pvD-0Ab
zSYhZLqn$|sKfxNEOP^=>Jy0;1$@sAAgD}dCRJXM5L3q&5)Ot_xlGgj)OJJu1md}bf
zBj4wQoPp|+-cv^8vjLCh9m-3ELHL00QeN^gEtu-9RM3$lFBFdoN=d2mTP}Y%@<~N%
zofkZ*Kv*ti62BJgpNF1I?a)c4T8VOrlUlAQMU$krpyrh`7g}|eN{&H~Ja;}ePG`gk
zSmtwID^)CKx6f#Ko5ib<GI?Lt!y(VSEB?2nHCi)n*%o$FD2DAEWCjYSsWO}dn`+DH
zAgd_8x(5e<)25gxjQ~A!&g*|S<TvMl1UPyBYOdI&dRzgUp;Qh{mAo|rFcb`!M?d<4
z$-#Jv0yCDU<|J{29Zks-z8^*c%#isiYloE8?VQ59mD~_0!s_rq$qmj5>G6Lmx#LhK
z_q8X3`I=C=L(_&m4uzTIar<nT?7$UUKgXsTS7DF(OLIrij*<2L;+%ih9!oS9=SqnH
z=IOOL7%rxKyTWeQYgTHCWYLWK8jSARpjS!?AejdNpIn({v_5coWP=_le1?(W@_9#z
z`1ipFAkeEFwzrqt>v32Pv^Oq0+5RvbyX!`my`jB$H1~%>$2EJ|wz>a>J;YeuH!Tu2
z`{XI0#LGT5jX2FNZft)6n+AO3(NS-<Kh@q&>SX8O+-QrH6Tj8WPVm(d&>g4%EM~=^
zDc`uoLkVfx$<8*<xAPcfA-*rylb1;|41E%i`g#$m@Wn0ZMp9SgQU!)Gp&kZyyM-67
zjA<J$oX<JOTU}8<)xm~?8>t&Om2{c6)2n7(x=LL29V(q^0bYN&DMX=3t8tEl$LQCz
z?&Nl=vPjiTSW?)9+wqJ!T)0Mx(?uza^aKh<V)5_@LOwh?)xc48hrK2b1uqr-mtVHl
zJNZ~B=2^M_6q)3Ey-{A@Lk{B>R}$BkwXR=HYUf!1J#~jMhvt6?bjLI}9P+fsH~DrV
z)+3)oYCW}~aJok)7sFAisJSKj>}!!L9vrR*(B5x5na&Nk3QOU*?2Vn4kFj!V4ivdi
z;OBzos!X$>JrD5m^sXq_ni%k|2f^c$0Nj5Yy`q;9R~HijF_)ou4i%SRKq)Jid|@yQ
ze<~I?pkbix9uJUP27L>13Nmw=L+1YvDQ=QT?Y2A1!EH+-i?!FQQp@gpC%Z3C=l7>i
zS4Eu_PJ_kn{aXk8K=NOZeADR80tlm^d;ih>_97)3{+#(kQYytC%fB6e>(jFW$18wW
z0N({TJzE04qxg>EHK_Sp@(201&G)W(f4|}VP7(zQOdC|D@_PFH{g)@18UA#2u*R*i
zB$%hdZ8(jXnr2FLPA6-XuJOX(ke@LJnz<vN){{y1fS;?d2Gd%)o`OJ>w}vH=yiQ94
z>2j?z3BpYNoR8M|+T=BtNHU!eWnHp}0woN(M#k4`&>Yxc8-LO&&~lkQlcl>1e-c1C
zLke$#O2gNe=IZnN(_c@$Vo?cgd}ovq*4plmA5Y-{ffmO44pDfZ?&qJ~x2IoQs@Z1k
zcXYaFMrn-Jl^M|ptyx4;LTV~zyiKoz(300sPy{7ir=Lj=#3%QEetz$oJ72%_sz8?6
zx2(Zi_*6dHuSa4ElO4e)17no^e~vf;<n{Dy<k+>MNG|azFUNVcURbb4xOGk-nsx{7
zQt_Nm^%1*}Uy5T%Bb@WScGM~6fZE(SE*4^+EKBcNN>fd)Mi*G-NBKo6Hpp5ct@j?(
z&SB6#f1J`PS!Htf>q_SGPAI3^bC2kKL(@nct8CHcggz6na+G6El{U`sf4A`xlN_^8
z)?~>xW$}gArVn3}`7K#?uxXzgj^-mRwZKU_vLGzis?~I~BmHP&gd%uc`iZg*{>ma)
z-D$1a^4NQt9o4Gi*@?@p&|bAvS0Wya7r+K>b}ZzlSqX?U9I9S|N*}jNP*dF_LA?X_
z+GRm1r35jSSkEKJBqQO-f3}ooBnvUd+og<IBvGyI#OU*K)_XaFaiq4V1f&-Ezsw=2
zDHUe~U`u34Rb2nIG;@JMZUnDzgqaH5Iv-b~rwD+WQy;8Glu7`@tLBseF78re0t*d|
zR*NOwFvQl>gVr0$J)>peg4JqDYoVjOPzddg4=?v*>tu6Fxo4oFe^g89aJkn4{34ND
z?K!DDw%ViQZp!KQ+=o{CRC}8f0>UKnMb$A72l;YX4tFk5f%z(HW6dbsB=M<JeNHRF
zVt2y{2i%lG;MR;RQGS`!(FIOJg&KH|wac$_%Q;SYv(9a9>7cHL!6NGOo>Z4QMLUos
zxvBv2J{F0Fsn|v<e?M}PNPgVf1!KN8nI%V?;Vh1}0%D=0N2X5^%8tx~__ACqwHd(g
z;6SP6uaptKWu`M~t<!$j$NCAQeq5niB#P(tPs=H`56*SNN^Xrcv~tftt3@{q9j$2z
zWUM?<FC%O6LubdWyu#AV2gy>vnW8hkqqEe5?(N_}QP#L;e*!rNJ7*fjmeF>q=q-h<
z((+~0vZSj#uh0X9@cfd`a@j&@Z3?0BRj}+^wn(v*>E+Q8UX0djKN_lBGmVmFrn1-}
zm`7ZUz35!Ruz&hm(o}fbb3)~LItkRd$UX9@T*r&9D?Q8`TBYW&#1U?KA}Gz#tlOP)
z?R~1!6QH(Of4LkdN-!<u4{YaRcO{RsbM078dG6Mn`j^IXWJYOh<d`L2m$Xy&bZJzy
zQM4o%TM3PP|IkQIX(uFiJ^#!|PH_)|esJWQbeh%-<G2`x@N(lgj-eEpma)}=oD{&)
zH9Iy2a`!d0Xbb}*b%aZ<|75V;vAe0g+=0#&I_`b+e|XqGZ{)u6m(1-RHxRSPMO96B
zIS!EryQ4tU5<GxSk=tEkXhLCWHkg%zZy9b~Gft(|l7Y?&#(%4FZ)78ewbr#102Vqx
zu27bR;(pg*$?j*V=T`@)<7~6~k%)4N2*(wBhF%4`You84NXDCwYu`NAUYclOZ*1bi
z3}hoZf0YR-c%u3RShivA;API~o#w7r9{~FQFSA+>7_VDt)edQIH7Oak22AvOqfIRc
zPqf#j)#{Yki<FlPv@&e(+k!{fbC*p=<E6*gRq?Gx;{F}<B}2&#4cr{IAQ1gT-y1gm
z0056<g;5{+J_al?5_T;ZHBwmS>E0JIh?=1;e>?_!&Q7ush>x#FL8a@s<{C%UgyQgw
z)<Nurnv>RMw4O~J!n@vqo4hT~>*H)&KKg}CLv_asn`Xz}`U#(Br2>y!m4^T;c>a>T
z0f%0D+;>Os@T18|-<96@uK^7okA5$a|DB*YCHzy3&Qj8K-@6HdINWL|ExerwehWj1
zf4Zx`2qj43f1<r;Sh>5qsOjC%Tr>neq_<cTdRJ>v6Id9F0<zONH-2-rF^~tvK1RCT
z`G82yT|CP03(K@mXXUvK%Mj-`({i==MlFFc&Tm}v{7x`duYOQJRc52nIj%jp(gn7;
zts3g8>3-#c_;z}RmqX5{q!rddE9(*We?uR0P(z`96JZ<uRqsgb$XhDurh(QZo^of_
z_j)P^ZM6aD9d`%H?R24Pal^SSmQE9JFfR>nWfGpaZ&5`heS9O3r5(*<&;9Pzb}4&}
zKIEe_v*7ZsBkrSnEPBjltM;;7cf8$cuWIpu_G^#E?r5$r`3=>Qw)*^<wI(-XA)dHW
zG{T<K+td3xur!%J=Rxu%V3PjrYCGBHX60{m(ioS){RI;QF*7qVmjSCT6PKt~Jt>!V
z6+oN<vfq~!7C_p6(o?srtX&YG!e9xNHTFKsZ)aUl6rHrb*1~&#sr47VlZ454F+XLN
zgeAgtEAJkYC6BFzerijrAcY&WmGaj)`ERu{1`(4?r|=t%3{C`ZE;aI^x5Us0dB_O~
zbFlg!7RDDDIQF}%kpn_6e<?VJsAu|qMjtnaHCbBvLHnnFE&kzksAy=E2ok2x-yi;%
zXxf)4%}pxO!Q=Gz6q_OH!&>X;sSQAg>Uwzh``0i3`uqzB8t_*fAtysBgy5&|^DCkV
z0A~7@Up@UC98N$}YD=Cb1y~Z0hw=O8>49buMk`f2eT$7+I8P;{jsz6e8!BUkc+-@p
zPvp3f=la%vHZntXf+S7hA-+W_d2D>8@jNQN#!vgsU1`++riU-n!^fh@py?kyt&?_5
zku)hdf<TQjxk-B=71*o{>R`!)tBngWbDJV*tsxZv**rKGrZQELlv^5Uxrs<I1#aSJ
z-@YMzC+Mt?*&R_JF>T4Ib@WBZDxK9@zkmLohkd1gK`+g7Ci4iI${kU9L{Ifjnn)|5
zPBQ5#I-9$5Zw}(;oyMmcxG6w6P@hvM@yr~kZD@5wiG`0-!(tNkvC8`KQuUCL5b*-_
z^trlG3mPgEwy0$TjI#vRl0gd?*KP+kq%=cGRobVFQ(LULfr5z;7~d285hyr~#f&r0
zAYj6OMi>S$Wh_QA$;tRD79{DIV@7DP(8yF;_L-lMix>3;y=QJq-&B-rMd@4wsASy@
zWqMC2QxB%=%2M&>L)P%2;VQPhFWNg-iVQlT0SDVYH`Mnhy~8*ww6{G4ct4_J_1xem
z-B3Q`$l2{P<utkg5kh`zaKo?<u>}Ws5ps)v$zJD|6grr-mLY01u4ITB?sJCtE&Y{(
z=FzbRX{4kZ>qwAlc9hj$KDBuuSN+GP?58}i5w;8xl%awFi)9>#DrT4V>nJfV>roov
zK(+43YKy@WT{D^3RYcUn;<B{0tXdiab#X-aF#wf7YQJfnz1r8@4N&ujaZMlgcGa{k
zLD**Ae+Y!Ojq8MZJ{FHsVTYQtBUBjHuIJc0M6gk3@$u5Vk;nWo05)TtKuDI0KO_u8
zm0XcM3F`d$>-6Q}V^1|3rGHk-OKZu~7M8c<e4aux0wUMq%j*<LhTiAz?e>J9?ZuIk
z$Yy<*SjzZMa_dD)<IaRvs3r7F#=Nnyi=lK>f72kA8rbW3o2$ZHYmB{~Q?EsEa1EzH
z(ei3w)V3QUwb&)bo48SHb~s$kwkDX3cnRAv*F*#&Q@NU{E?QE_3UbZYv<*VyV8)sU
zx|!Bod$Ry=)M&Q=c(_Yjf4uz8se3L0P*$ipJJt1Q+A5U2pYul6PRpmZNZ6+px2kK~
ze*|jE+c1~8$kwIzI!C9<uBPx&1S~q;WcS`&d1@+45{FTTvbx@!&p@hr{m8SycbXpx
zEk{7QqYBK+2JmOUfVeGY`pr1ro=Fg+?p)lIu$2cY$bG&}30B$D3HyQst%DPa&{vPB
z8Lb?z4thW(q&GcfLfESXKBkDE``&use`78LISCC?cuPs5;4|I?xrS?vq$Jl<+Tz-f
zZ5;N!D@wjWYlLBS=qYY(wXxQ^`LEtAidwy)Zb}NoL3%Ljl_*!)uJ@KD9SRkEO(vD@
zRo0AW6VOxJv?fKk9;P-~QM`b>uy#f_U53$&RgMV*T&-(LdPcdPxvi>i;*QiHfA=jm
zHbDe!2TE>zURsDi=sVvKhO|^AXpDm`psV3mP_&AW#K0S@?@B9LLIxasK^_>uskeoB
zhGji7uz}&Ma8$K=FFr0$d9-VG5a1?j>zMu2FjImnajl_O`O_3P5pb<3Ya_a?Nm;Iq
zZoECZm(<QgcQ^IV72Qkb&PDgKfA?FWyQ6eEx|f~b65X9G?hxIb0?$NuC-%LgyJO}|
zba!6-1Eag+>%!<>QPxIuqZZm<iOa7Fd|K&&RNX{daqpb40Db36mg@U8AiOQMA*(C6
z2MFZEwxqZP0vUuZdtbISyS1;fx^Xefar+8g(b@`}9yDe17P-u4t*S0Kf6BUiefd)2
zzN7eoD9CL~(c05k^^fdEEq8okN@})HNL>Wr<@_h_(Kjuh#@xH)T-uVYu-}EK*Inkx
zH*#UoiHcL|;ysPOx0eexV`3?!+19-b0*(16vcqzwbLpY$egFJG&`6-iJLZRu;+g!=
z_5O+TL$9f~=ZBu+>HMHPfBA&6Z_N)Yn<((qLX5ZM16I|NRawIQbV5+$I(=iW|Au3t
zbH_jmfkI7RaSW8U-vVWLCGFa6Po?5sQw@Z2my<qz_ZZ>Q(|mEWc>1yP{JJxjCL?zV
z&g!+vfPBf;y%xpiwOPlZA^3cH^N<O8va_w5%Iy}e@s}~4d$u<te_Qv%lFCn?b23L3
z=;>EEa`IA+QqK!G;-0#md!lMI>3W*!q;($C?{u-pA4(729-l{s)2C_e930f8#WjbG
zX1P4X=8{x@e%sJ2cMl48vY9ESl;`1w$mVG`r8Y%2>TT6u99t<+DqGBaIO`EeDbdo(
zw3~%5h=E)tQ1+j=f3AcICc0~6QFG@u)|rwrf?~Kxch>K=q`uqDh{Bn)1T*f&*|`HK
zKLfvR*7h>vX3Q8>uPxG>=HlNUOvvt*tLfu|X_c6oY2EQ|Gbqxn9&~$|51+r?>V*t$
z&|$l>c12$v^q#K@96HS$bB0OsbB0T{(9P{4*XzGrQJwX6e@QNCY?pJ<YysQkI=dQ!
zfffv?mUup2CW8J7%JPy;Ifct-g3FY`t?|E8BvzjreCX?vC!Z%WW&RaToK&_~`CX%O
zHUW;|HQ&p1!#QZnbLD0spDnHVBsTH7Z%;#!FMYV-G_^?E-V3gGklpf9TQizVM@<T=
z=5tmkqZSADB27&xvFTTCAB`EN^l8QBYQ#6l*qDi_?pVIwJ^DY;B4cBhu^c}X0x~m~
z(Y7=t0y#OCoDTspf6ZE3Z(PR>e)q2!pfBlz<_w4AaHa?h<Wi>rl-7t+psk??S>7g2
z6|Eqx4CmkX;~eeTl_kR~t>Rt47w#_449OusUd|9_rjP=0&JuF&goZ4oQ!PkwDwP&A
zsei{QsMP*z3EIdhNx`yp%0lH%F^5{)Gp8g_8kc4)akDn#e_ye_vp-Op^yk6Ws4p)J
zQu||q-&*V23xrENG8B-t3&1U8EbP1CiR@xHiEmj)X)4Ip`ePtV=1zj4-teZt$@*d{
zf%CYVYGCFrnHG>X@>*cYsqa=nmcE!$U`<`Og2zhkyIBrw+{uD1Y?QZ9kTlw}<RE$E
z#X2I|7b^zKe~;pn8ek{&-EazGqw!%mvDUYgAQdu}=Bw6G#DXsy^;NMWj;`-jH7J!%
z8VAkeNtp!Y@gxik+PY?j)2%dqYoR0_$r=ic`dSNxM|CQPqU2^uErWy(?Cw&ub$e7u
zSGo}hwkxIPPGT*~U9p-#maZ6dDlK)B!zC>BWpgbce|22U+}XMawFZ;c1*<KXrEvug
z*1EU^ud@bQm_hKGl0EZ%3rOux;n|v+uLa3N&Ez{U!F6H6&Y+=69+$;qFit1o<!0TE
z4ePd>eOqYM{hU(@4Sw{@NeYdt^>1^4LsKU?w4%~U39VWu3BRS(eTZ78Mf>;J={mAw
zuS7lCe>f+PtgZ_CH;GH?d*Z1iGKzxLl8);q=dm;@=}}qxp}-`dTem5@35Z%>pYSgU
zjXc<ww8ta6l2mG6$0Lah(eP@-kQ9Er0TEEg&%XR}`04kH+3?l*`Fs(EujiL<W|yxL
zeoV*1--hppZ}0)C9uGgxP8Y9Wzo-zTP>S7Ce>KOJyVJ!|%JH+|==$}-7d(7>{=ea?
zY5CK|+xhu$H2m|&@BQx=zbqCPSAQ9XpWiNixqcl_=f4fVUjKG+@MiXI_~B&n%Rep_
z-=8mLm;XIEoyA|zhSz8F*TXxC8T$@bm#4$u=jUf{&u1r>!~9})d9s*aUV-mVhcMK^
ze<9B2&(EJf!=Fd$he(Nr&|m6DY!UIl6a9XhznPtlx|tpvX3~{z-}X~AKg;1?|Nal$
zq{NDpU~uE}>$9`tkFgop005~*vyE6Ox5$9K6>Av2nV-!sk1n7>;O<z%%aa9^IuCjL
z_}BM~?~WEHi&>!Yix>0r1vYunb58Bve>l8wkn(iWKR8?{f~bE`cWQtyK0JRuyu6&B
z9?cf7hL_*I7=D_)UmUL;@XM2*XAlTBJzre)<k7dpv0lxuFHdJzH+jgv`(gIx?a9~k
z_uUT)ISA!rtZ<2EJ+^ritozK>P1@`t^dMqh1`$u;#5z%BqU@d!yX8Ok`y-7Xe@1^S
z920Vsnnq9S$gfd6u~&_(^?3LyVELoryZKM^;oI;<=1i>MK*}+J9hu=QkB<WtivoyT
z#sN}CY$@+SoQuf8p{Q&Z=OPrFC&3&2L~4{~`=DQ`(9iOA`oVQ$;gx>K6A@Up5BjwX
z{hDow(})XRJ3p!v>t39nCObGpe>L|3R!f?Y6}WpJOs{b)u~(%;FFQ716-lwAjaCIh
zMMudESjhwe(B}mI2jji+UA$L%9NueN$NRcObXietNom@ykWv8v*{YB#1OTf0;J^%k
zRpIRd6m3Q{UFoNkBf7GE(612mv$Q3^YKclK{n!&W?S*~~08~=h5@1`2f4rt&PB~U;
zd!Szt*H7&OfYrQFF?@^drWi(TQ_ur$z>20G#qdU~tQx)R+<=uDumJtqKBH_YcTu*;
zCMcVYK0sN^YbaaN5>X3HRGui_3S|-IXy?|-So3ynAIex`c(H9=#sXEnEttw!s{a2T
z=vTc=uexOx$;gsEx|&d^f2635_d!1|)2r3(^s5|6R_<9_EV3K#^zCj2DHWbc?ZaV1
z(N7hyYaA0^mKL>?O<08>7w1h_QM^X3bt6`vUTh{SI*2Ca4+r73(Z+U&T=QK-uIX`z
zTy+hRZJKk~H0Q8s&SBG>!)&6a`q<j`vyVROs3nvvW6-x0{XIote}OaQJF^c$F`%dl
z%p_B5z=*ol?8Ap+ZnV{m6PcWN=V6%;mQ}Vb%K@lW*UEALYArj~Eh``gd3no{yc+5_
zUX|o1t0>8?bt?_Jnck<{0|SE@fx!SZ&geBV6qZ<8>v#yo?*xXGQiLScs%j;pQ{HYU
zNL6E78wyfG^^ta+f3Lh9(DoGO&F~-KPQ!ox22UzNqraJnx!$!dWh%Ad*J3a-p*gZP
znYSmIN*lE#x7e2Rd5iwmX#-Y8V`j)S8?drTthf4%kF%EC+xHDv$-VG1%!r;WakhIB
zXM1GgY_gU(>y#Snlp5=l8f)9Gm643Bw1(W$kd=0g+>YB@f7bZifGVp&0{|t8?HW`{
z@$PKf5en)RUTa<w#l*X=BPuj&Z9B~h$NoxwSs=A&w42%c?A^)PwRdG>uFc&aKb*Zc
z`c&)E?!ncbRCN<pFEsp;#|Es_IO?sb(B6%Dn9?15<eL|>55-5nva!O)HZ>7#LdrHF
zWt)()Er;xWe_vxs(M08m;)#+cc;$+`&)#;RM9#&pT1C&q#IY*<;k6Yijaa3v;2l22
zx|5i?IJvx<EzzCm2`a0c5-KadBn4&Rv>Ca}uaHU2kwyM+<6^BSZDm~ai95C|)q_LM
zPheUye{WO*XN`WfD`%g025%KoE0477Lys4I0QogLe<V1+p~MB7`)4$lR_|U`sAyO+
zYB@paoT}lJ^a+G8!{&Z*P-DLoC>j}fgc1#bW%rct(xg@)5}>A9%7var3N{=Aa-v_~
zVnuF5(VDGBY<EWrslYfYqNw<D=G5ADvkONp_10`zb127@+m5{*&IxyCenNl8sYOPD
zz43Rre_w1hKgi!bNBi}{4SKZOY&SjH-KOh4HbDZRBzv9l^W7^`Ozm+QKd)Vx@(Yj4
z3|%H_neqLq<C2>AXyT(4U)?^tqeO^axe8FC_}yPL`H0J+1%0l3CWxv41HY-u6#ETj
zHKakvlu%-PE`X|@dZiCF*>6-MC3|((hpqLoe*pE+!$7g)7t<Aug!=7j)mVYuK(Huu
zk0~l8meiEEFpeaZdrk3xYAd6vtE*9=azgE=`_RIu__)xveQ%bctX=;?LX3^~p+zG?
zCU}aPazyLqm#d>7sK^jp{@L-C?q-65AXeDN67nIXmg*Bk6B9uQn}qONA>x<hKOzK?
zADU3X?uAh$H9SkV^HKiPJ+XC;saN%5UyTd@4{QE_mthnr69h3cGc=b`lNS>KIF}I(
zK_{2RDL|NiAoDKs`R||qdg2*i&b$df!`Jxq^!fDrX$0SB!9Yx3ew@O%lE9=Blfai=
zPk;Pl`g-~}Zd#AxxFr42eEQ?tG>m-8^z$6)$BogtqaSc2s+x@POoms9e(F8v8u3+#
zx0>;TxC-cSe0OoMx})OA2|ezJMCBSOs*qYL?wodiV=BN|{zUM*5d^>PYzzY<#+?Q4
z;HtB&V#dNayK;7Z{`t-_B^R^;SU=yRL^`)PA92?<Q#Dh!K8JVa*>DNrnYZpBTtX@)
zq*$X%&mVk#%CHz;1WM=xOE>cA!=`)*?U~hn!84yv5J|QI1;7VgIwWJldA1Z<)L+g~
z>3Ly)t(H^!6T$9AC~J!vXvZY#*Fsnl<8ayrp{}76AiHKBDze&LGjT@n$hgCTl;I7(
zA;0(eolfrqoMEl-XB~`Ud5#~QpToQ&2tqrwODKT&ak5Z#G5V>pZW|jenDGARU=x0y
zgH6ANOeOoeXA#mf;BEvo9*kZ<X&(uOzYH&bCk3__fOUBj(ReIEQAbTfGVYf~ML>_P
zj3cs~7lq5$GgTsno^Qq;X$@0i!JQSE7sWL^g|R~<RYEXr)q<oM4P`5uG7?ZgA&Jjt
z*=i)qOejP{`QVAHVReMkH0M=i>V1Bf49zH&co36H707}$iHyp6tsk1ip_R#@g!0#a
zG$%nT!RDE9pBjdQIR*X*yG^^t^>ZG#We+z{i;P&Zc1|$pqEn8pAH{JPb#<g@nr%2o
zI?N3+^>9PM(+tD2kpnlAJoHv}vAfC&)+l7DOf~A5Ob?kqB3m6>-Rwe=tY8XSK<dv@
z-xKZ&c^{!qCl-G2EWMWm6TjTf?iyf!T8I&@!*+KgeA_jm3PY2{sXg7J$qTz}?t)&~
zXglu0AFi)5NEI;Ss>szy{cWy>8LG?>i7!zz(rDKAs%t9i{Dsastd(hYmFXlWMXqna
zc?F6A%KXUrt5~G8sl1E$XA38@l`3nsJtPU5OXe{U9C>&N6wsB{g&ra&ATa`elj|CW
z{8hTfYC6kOugk$f50;i^Jc|ueP2!9D&03$GZF<^|JdXK;6+3<_P77y3TssivC#$%`
z2;a-asS&+3HWU{M#H|EKnu8jgKNU*N>yCyH^)bDL_nDN>%)T}E%qTOa-_pgL;4RW`
zEfEq0H%z~W9dDR^4?FLq-^1*G8>HXEjyvi1c>d2zzmLIH((g664&^o5ab}2CUqN4V
ziWr^igLueLoRV)qUKzv={86wh26$GjiQP}AZb+}$0<u@uHk8mPxC_N;#f?`LIJ7V<
zIFxGzju(q!O1Vqc^K^^)N!&7(P*%z)FTfv=V?tI#yW^@2haqjl-1V)03q$C$MvtW=
zB8YN>9_#lBCw`dXY}`P4r>Fl;VHiTlV2rU-NP2<}A&E}!{{8LiU%vbX3>W+-tU!l{
zVh-L-KhmQ{5&@{`dwz8I+M{6r<}vyU={x!XG{j|DdHyn;6j}w4+O@&&VbC+>F!^sQ
zqldM@JeSSMP(D6vR|m^~S>Nek0jM1iv8gbGr?5+28y<0eSrs0`hd!67xayzN>Fad*
zSSg7yjtp{TEH`U}1Pp7f1WL>@gn<OQZ&QikEsD~{C`Ggm3bY7DNkq*t%K?AY(<-uM
zgdkjoaIdmM87kl7_yz$6ToXx>L~5ozu0ylJV*>*+Srcj7?=L@p^15%h>I3;s``iJ#
zpAbLZDQlU;Y!CN}k<&(1G*2(Hi$QzH`BTYGU{S3oC$Fj3uqlG@aUvilT<;pP9}g1<
zF#SgRvrnMWqY=U=!GQZpn2nBD!9go@F16GlR4O$@>ts8k+)1rUjIQ<JVC3xKekak1
z`=;cXR4Yw6XHklON~0waOI6gA;%^j<IO-#a5!a1icQb=Q9pb?%4ZA#$IObBJE5)Ho
zx2t6Ytvedkk;7%boFbc46d?S7y4Rf3)C3JJ<kk0aznXEKO(=f=@6+lcH(AQIX<0I`
zTEscu#awNrfR_5JDyH<ZKsI;fOj9JIj$E3&sIEX6%L9FXagr3!XxX<ESkn3kt<EDw
z-aJY9b&};Frd%@&O$FBsLsRl(82&Z&)3W!zBkuc}i7U>YV=6^Tq6hPNA)l=Y<wWTW
zvmmyLm#UiBC<C4pGfbnqlBO%mgb|v#(_&4YJ9bLaXqMZ4U*$kPDpKTkMQiH1D|(qB
z@f0UhMqI;xr7sLoSi7c0Jx@8((Mj`chE)(u+8U8{$2EO#jgB@ecYWd$N%8VX$O~F&
zvBTeXHf*hohS?Icq;B&>ZCueydp|Y|XgMuv-b@Oht)(zG`yVrxS|m%Yl&zXi-|(q+
zrP;QdmCP`^bOJ^y+(5^6>t;$ZuNNoRQmaW65lKpaBK{k#2B`xM%|7&0+H445p2;z8
z99F-?YTypsjvI}LKo)l?#eJ(FhkCcVCso_{HijZ`yjEyS1wM{Ofv3GRDVM8UcGB@(
zRIa%>i71G&3gTlaLWbXU4R7Wv3Vz+Yc84<Tq>D4t`SPALge_5lWPV&eHA6tou+4PU
z(x<6^Z`XR9KEp5%p<Ttk1_i71$c34OXrYig(}IC)h=he__Ab+k(ZJ_@RFjYZp|E_X
z=a!6UqNQ+rFvr7LIan%ZT~!L<P#J?P_jrdZVex#iA53!1EGAUL4wWaCNE+=tWPXjX
z(~-unip@-^8{|LF&a_i{Yh~u-HCiZ{>Y4+8CbaYsO<?XDEhHd6OOg3U-9hcd$NEl<
z@^_*VDVPUIoAX)pui80qOxkgAfizOdT|Zu0Cahl^KUcRTb`urQl1XF6Jw{p{>~?Q<
zJOR^IP%k8u2o+se;7LnlIgq>Fs|AddmoUuR?hDWnI#_XECDd56lQke%wUe#GM(j#|
z4rAu7$G1Gq7d;>DeNd}f`-8ZDORHUbF0H7sD*-0<!$YUexa#!dRnm<<2GY?kSCkYz
z-gy~pc}sdCzev)aDMD^FC^saU9Sz7mmAAX8faaemWk!qz@hB0?dKafJq3(oTl>->~
zSyNLS(EA+dS4ew?R<F1T>Fp-z%^Q(_UZVkVYD86B$W;wPR8-Zg8_y<xsG-SPwu`s1
zcgWwbIH&-=uL<;&{bdyZ&~z>kCO6$`BKa^x5x#G;J(nrMP@4^QpDF!{J)L*~H+}(~
zaLg&UKvgcU1PT%m4;Rx8vw-8btHBnyeHQd9Zs8dAt02FQu6WM$k_bC@o5VbS+--b(
z5wpUL1X3`LO}*>P8dki@haBh9|6hl!#h}nF&|k$brxZSa=4S51CdMg^c$q5#V;M9i
zKI}8GTW||U^ruodbWRu|3`AH(FC{kt+&zVSp~C`&sTs8YczN9CUBBX04(m^)fR1^w
zM1O;A&sV^O_U$F}Auab}J?pc7@>*_Z)z~(Q^OYgxbRZSRna(Pu%`Pvra;#0Y`OTe|
zlO4T|NE;8S*`;GEX+yBuKPk{o0RfGc8`E*R19DONOPz%(wsB2OS2Iern`#%x=H6YL
ztvk|_l2P7O5*=0qdGZeee7(m2!95#8OA1XD>~U4tF0Xjzl^V}kl&bbh^O)_@3MQj=
zJfLYesS01Mu}E})U|t94-nF9W4D-#R%`&Al3=`}Tkc|?J^9%#tnhrkJ%zLf2V!DPA
z9JO{s#e_6r{|6*wArzO<KtL6jVVEWqmp(N>%76EVKfnI*#1q1tc|#xRt^fA!_4LcB
zM_*aPthLjpZzp=7Brxg3B)DWky6M-yPM=R7da(0m9M{A@noqxenYxjWiG!Xa2e~m?
zHyi|xY*mvno=JKXIjG)pu8}f@l&fcaA+-V;D&M>~ScIrJ%0l-LXP!PsC`=K4`A<N*
z8Gk{vXNX!d>0J*IT71fE%Z!$1`y(~lCTbE@^LKh?uE)!3=9RpF7jnbmQjimnOajDa
zq=Jb2Vc=~KQ6TeBExE?$%0_(E%OdqcXlBKkJo!_Yea0j0kh7+ikw$(<NIZE=W6i!e
z@Ujd|S*D~ddmfXzRmdmkp`LTsXJr>UaeucS`S^Za_pM_dr*u0A(hNx2K@>kGZs=Y5
zFSuitn=@_@-=1gQ1>Ayt7PD<eYOU~&aplh-qBk-evTh^Bz$4ZqLoE~bmS%~2(s2A{
zqAdptP=TiUX|j=Q3u~4o4O7;QmU9br&APM9+S%ffXh{^SB0Cx*@2{THIE;}*$A5t$
ziSNLDv~Uf6H1}Jmm$=`~CVxRbB3KPnR-=lK)js>#F%sywaIGyQY)B$umbg_KpqMtJ
zh0y@LB#UM;g*$<m$6^2!4rTGs*a-HwWbe`cVI0S(2p5fJc?<Ot&)eD3XL~NulG4e}
zT{%(m8LFSg2%+P|(DOj~7y$;nO@C#_+0M76q`hDUW23QB>Y!ebQa4*J(i^h`qBgvq
zi<yX34m=Q(2OT9nqYfV>M+N6Fm2zsd;997c1lP`%i>@Q$MATj~zjEXR!5tA+$DE<3
z&QZZxj*-6}Ew~oyCBe0`*97;2Ev{qK;lt}n!IjjU&k12OvX!!AVQbH<nST%9fRj!s
zyW5whX92g6j_|dEH1icuX7ILFM^dDu20GejrXbjE`c8967BfG2dVgERHey*JQxQTd
z#~d#y({m>YWM%misBA`RgGhxEEiH4()hY8F=03GJ_labn&lehLjQCJW6G6?Fs}qT3
zIjn;U<Z}eK%&urT702|usDJ4w>8$rdY1k$p_-q{uOtnSD6d}^@+K@FoR;{O~2_f&^
zS0n@q$%;t}!<_Tug%uIKMC*=osSp|kwU5?HaC_=C;(~)}xZN9L#h36bU=oleWDK8u
z<YP*+GxJ>WtDQ$u=ysU5R=c(o3FxMv5!&y5|1&McC2DzzV2D|}<bS817rgCeee_as
zRC9%U!@%Nb2XU>qcg=xq3<D#2$bzpN*vb31v-c{LewXmOz4#^sFolzQe0|zM6jKFH
z#rl24Hx0CJ2Q*mC(-8NsykZ>0c~v`lhTD8}cF@rUBy(TqMDGH61h#$bbq;nz9@zy%
zWW*XWS2@9q%<NzH<A1Yx5GW_nQK)P6{IcVWBD=zOULwHotDI#`*iL8JEa{!)(iPLY
z*RvZd=`2$aZmfQpaN_$Z%tipxJ3aktqG4QwlTEnnkP5()iJjj3<ICqieEJKbKYjTp
zt>BYFn+5NtZ}F?4J&cvp*Z3XdbpaZqIA9*Lcr{tU031%p3xDT<Pt!@E$U$mXjeez3
z&y-U_RExtko27|)gEBPZKoIBAyx-S)Fk@A=2sF)T&^LmT*3ws8pI6aWdez_aR9yA9
z>GXLzeVF;_CH@OHNZc&36G$u)YL?DJ9(BT?WD`5QMMYg3tq7e51?Ij~s!)P%I6%?;
zRRTQ|n@1{w^M4S+$`JBUe(%Ru7|_ugS(5T1rahGz7%4q!B_y^+)|StozNK|vaMk<j
zJ;J#IA(_%<x>E|zQ>zJ9xIRU&OBc{-dY+yA%j1L8_r>uA|B=1fWAadO=OrWSfFcSX
zrV7PK>s=xC!)^wl*U(z;U{D`06vm)|qdth4G2tLyXMc<k7@3Ld9PdMPV=H(5)-&OH
zz`?z`bhye=YDfj;JK)Zz++x0opvTW?uVpHP#!a|7qvGm%?<PeONZ)P03-F>e^jXd|
z@<Ptez@3h05Kd&3EDXu_1OauL#XT>xaw&kY!hn<fxy+H$+33gjt9(>c8sL5qAIsUe
z%ukJ>C4bP_*zQ)9HjdIWYd1!PUfA&!=8zEXp;dro-uoVtAaE5T#C+|k>@f0$MsGq<
z)jsm2O8JtCqRbjP=gek}6;*V#iyP8PbaQ9i&=&(@<Q6!!tCkm=h9nm{_X_Y-GRzSz
zFY&*&oBk#1QTC||Sp~pF=eFrB9iFY$_B&)RAb-K$S9i9=hx^JC+j&MPdgkt_z@&tO
z%C>9gr@NfeJiDl88rdVQ3XgUB>G9zO+pt2dinrFag@d^|0+$Zva>&Mm`R`LZEyd7x
zRS><S$4wfH!&K=B$`a(h;$on~)`b}2Lh$IAU7GuN_n7jlUU6p!i9D4Qs7j4tS;*<R
zNPp?^(vzkoiW>1`kF1mrMh2uGtV@-4YS5EOG$A4Ki)}Wu&yj}Vq8>6iItftMpOQyW
zPnq=j?)xPi_=#mZjB13hlz13<(wUAU!>&elJ4XeEDRhn<CT*B8?#RcqL5ed?5S1~m
zJI%8@lU159m$jJ7cDqCe2J)%udZ>)$xqrlkm4?Wf{xj3lbja41Ez8#tjZY&_@;0&a
z%6?EV8Z$}g`$56syZA2cpc1VO)X0iw2XR0<Xb15^J7~ZEZ`TgGjn^&hpdG{;+Ce)A
zEA5~XXG0a9%k}G$wfnf0*3gjRh^8In##`Kc)!aiga}jd2I|8n2G?LVQZS+W-JAa4;
z=f#_67}+)`ykL@#?o{ibUT6>KX74`>j@|n!gWlVmaiHA$k7`X5<e-%us4+dnJsLBN
z8jEqBb0ewgAnvr;gN-pldxp`-NIE)*hwb<%h03w9YS#iR^sVaF2sug1G{9&N)k3}G
zq1xI1VGT@9Gkl#f>abJz0S!z-7=KKX#efE;4$>`JPkA`z7d@nzuYnh2yrQHE8(F|s
zLy`zZ0y(fDNpR$YRR3>C`fq4R0xwZ%&Hr3O(ju)+Fbnoa(z?+!-1a=pv~IRNT}8n|
zJ`&G}xcN|tAeS8U4ry2j*M$S65m~rcN?Z!zOS>4Xwu_1D+r?^=+SvioEPweaFWKIu
zQs%w8)Jt~mQVD7AF7=W<-5i(d%z)8_C_k)w>V8k;&vO0M%MIljm$CzReNkcaenWs_
z;JZCkvZ`G6Q#MCmu|7^CVzUI?%*b~KZj)R|0`aORckop)U2rMWZL)P!9BJRalZm%J
zZMyw%r(<a}8hPg7W3<%79e<wqXKJ3&O@mE04YN<kMfvQlU*@~_GyM~qNwHbVy|u1t
z6hc;gSrye3;eVI1DsYK;mA#^@x+m7|W6XY+YpxE~V(pk7b&!vwXLM0S{X#9owe&2N
zR=e5TS2!_VzkC}xpZ%#SdR;9#rZZk(Mpo?cZPC%5lGLW#E;vwWy?+H;M;DZS1KUHt
zs!tTQo3yFvKBaS#f%P4r9Ejbhce$kAv0z#4D78`=dQxuqwPzKW-6AlH?Ej6$Rc~=X
zST&ddUE(=4Z?JP!L*0HKmj7RdMGUV{AUx)M3nl)(Y32SUcDXnBAJAT@n2@co)^t}o
zrkrheHdS=~+$K5g|6X(v>r@9Ym7I$0au24GQ`zMLrhWRIF?VUa`kChutVxTd-Gw!3
zru5KaT~{?ka^uyH?hjywMnG}y24Y9dOp&sv*31t6Hz>+DCUDmpWZmDs9A<Uz7{&b?
zs(apEm(f5#6_=rsAQYGC7&$1HB0)f(e^tcar$4{^@lF!KQY4Q*@oRqi_;UBl-Aul>
zN+_?EFF)?^g*8f;s23A~W4rwN*X8To=Xup$I8lLcr1#DZ!m$#B0;~HZ48AJDk>^6J
zkT6X+{11Np`fZs;e*Oe2C%L(@3&K0Ico&5Ma%g1-qO%|l2V!%vJUcas<}%ibe<cQC
zLGKe5o8SY&x1bc~5##GIu!JmP3Kav=uHZWysg*`AnWhKR)WKBKOsAj`$C+>tCz%@<
zlh0_P0S(feyg>+i3TU#~h@beaMHy?N)mJ(1QWFhmEX>r%#|uzSM783<1*RfnMhXml
zG-*AsqalAl8$D%;(MmoP%B?Tff0Ct<vMw1e(aLI?rOQ~6Q)C)+eR}Ly$<UELm8NW;
zx=PX(Y2rtHAm6yrfnC(#u`35wUL@m>9hj_>O|;o7kNiwdzT=8Y<{;r>1O>-x!XJwG
zvT2<QlX`T;OJQSLvBpFW9~SB3|1#ObNbhFOfCI;QMK#hh2bXE~zNwXWfAZ;ex{?#g
zt}v+&%@8Kqw6wxHwUUh?Lq{`uDssv+ve8QVCbLJn!GLe2SNka7a|^w}ix9#)>Qcuf
z*5<aON}Y@dX*Jtii~<-ZFZloMq?=%c*DLX9d2{~Qb4CMER+*K!FJ`Q*xBzix6=lHs
z(AOc#{IbBrPfMN+3xIdIf8}2b4g(NO_CCP3u}URm;qIq@eEa%`FMm-20RD*(5Tt~5
z3cT`1d2|-02Njp^^z4OC4*XDw1ceVOQMi;~0t428FUy_P0XPS~+vs;3O~OX2bqAb7
zW59`wpc`TuaK$>x_`dgpOjSdXd8%aK6RtwA_z3U}g^&2@pLr@kfBkK_`?}nHuJTL<
z{=v5o(r<}V39AB>pPfi~$an}CP+_t-QULGY8SNRZgIf6LAxK#rMSw~Z6++l5vpmu<
z%tNpiufRk6J&*4QVB)oigd#O472urECZbg;5^E7{`~Bqy!F>a$Pu+J}=U!<8^6<+&
z1V$l~#<g<J10UKde@kdO&*rf5cruPl4o9Jb%dAr9Fy<{Mn-&EUJ}(`Kh1dJW?B}Zm
z1SuL~-W39x<7X!r1+E``70gb(#yn^*qOIE@Lun}s1GX?Ap9e}xmZGaWY4t!=QA!Wv
zXpYgG-eOsUW?5WKb(NNTtm}tT(PARRH(0S*ExU@XKxV2!e@~U;nzoMSY!T~|AM0kW
z?&xjm`N0)?in1KyKqDn<Oe(w@XCeDsXJXDu6-VP+(!0<VTm!e6o?NU{=y*~dVV_8@
zlBBeTEwU-5JL#A#f^&@LTAMz)uhUt6R9o-JudBQ#`OzxmOgV?0scLSk+G}g~san`_
z!e;JUPS_l(e>maamvI{9%lj^29)%N{ZbQuFAbU-0lyezfpUFa@`p(S;_M_@$IpkbZ
z2p;dhm@*H+Lbdh=N?M}@nU4gYOZhdtiU6->`1j}R#IA4ybJ{8qpiG&Fvyz$Pa79nN
zkt~e23YOo4(<q*It`yMQ>iU-4qOOUq;mAmH1H(3>e{I<N&_S@7ab=!V#Z`kP8-6#t
zZ;d>*O=;t(1Vt(mQ`;wO&`^1}umvB&2Ah&iLbhg5Rp}J&vo48~jMo+kYn!H^mt~;4
z05mu|Ksa1eK8=A;I}U{5LLk)M4)RCsGW=2dfiQHN`>qxss$6Q-0`HQurk-1(BTTxb
zxSS|lf2JdcaRnAz*ytw7b5A};W*aK%19cP1N!>KsE6GyMIhS!=L^{LAZ9_0nJ<c7k
zs=SH?V|rGA8MDG5rko*MHK+Py`4?vljvfUP-};Bvrp?Y=AcqXaqq7ke=xdCKsF;k1
zMp<hVK-!xc`#ixI&Z%xK-@O<N!B!}!#Rhq6e|s1jO@zGV;|N0ygMB;+YQ9OW#}U{h
z*u&!1Vt}zj^K%cKg$%9J#Fq7k_^gJlMw)2kGPK6ub0=`$D+$Z5-mps4?+lB}R9B&9
zqkU{K9`T}+QZce}50!-SwI!_2LpjB!=7*Nz<IspS-A;shR6AW7g?&O0l}WI5Z*A6m
zf5NcSBQsirHAF593Wam(jzi1zv2%pM)x(_Y%KSdwM|R0Ll$~{CEHzyVQ90C2jbSc4
zGhRs!IA@mF&$=yRabx{h?g6&=HdVR1%W@wfbd^9&1II;Q=Cif>t3KHFpQ@D3gOx>Y
zAZ@p^<7Q&%=S_M&4(hLJg~KvJ#k6XLf7>W5#G3y`FWf+ULA}0TFML{Y!yj($J$n!F
zMb9fw&B!_%TbDu2v)|H_r^B^r#XY-EO86L8_%CiGZtpMW##g+OE^QzSnY9$(v?Z%&
z3G;~r6x^^=G4COLh#yw@dBr{Hv-T(a{{iVMjUIU&Sy~05mD>`_=p69;kx220f4y0|
zLdr05&|B9QLa$v30nZ6C2HnaCmGE%?DhJm=3#awGDxN1{wMkX{rZGVBfC~IAM?mi(
z71qH>&0D2nAu*yIjH@0=g#)E@LZ;PSgBsr*w$b^&%;_-Fy9fnyovcvCubr+qz}q@7
zrEif{v9t7Jx1*0^g}S|TS#r?we~~^)8Q^m-af^Vf&dPk-#njU0JQBT4MJARB$?Bcd
z3T_c0u4+8k3h_t@=Hgy@i_nvLxR%xxp+{0nqqMj&kfoU>Lgwbrxe=J(Rny4C>hm!Q
ztOfIdUm6+PKF~a=ZU=fL-<t}8IHpzFQV@Gc>E(`l<GE-?ei6$R65fpRf0ya^a*99@
z(jDaPfXAhMDW@PTslo7m`}^OO-KDkLAR2;I7(sr+Ajeob>0OJZ2Ps%?iE_IICr3T6
zg!W^X4%ML_n!&as8CF!jqbA#(H-mnti}LC{xO174zdzJ_ac>UOP6{K{4#k79RBCN8
zr%b6Ps%)cD`?phNk}$T_f9;Q5mGs1PZ0&9f-RoU6zj;5G7<qgPuR6j`%Us~~ioNMR
zhqq_LTT<bPH^m1ceswlH4TqHOZNV0bF01YTv*G`2Hk4ji@7^yP-j=gE^SCyf^dMK}
ztd7TPvGgEw&U($=R@}yRL7%$XxMztGFXZ6GnrJwHtwviO+&H;(e<Bz&V|bsF%VK7i
zk7X~30Z%X$yqbut9qseZ@%wqc5?j1txFnnAuHA4MqxEX})XR6r@~%H%iJKL}aX(r^
z1?3WK>cDO!S0PW*ZEYUcs-qt12rUhV@9XQ4{cFidSiIvjVgPRV>h!A9`Nl(f*4%LE
zy1!)Hu3N<Zd|<rNf6Ui2&V%lcF(-A<7h*kWkVDk3)zbj|)*5oxSELa>%(NDrMK$Xz
z>i5uDUS&9|iK@RzP~OLu2+EZue&epY#d~j3n9nn!?k+ZUPnCwjW@4@CyR(hn=WC4U
zuopaPqtc_%o#S8DI#i92F6Zk*{drM;GPAC1@h@i}Wz7x0f5Y1Vz1Pr5-E{WUj(3T#
zGOPZ{!$jA3^0D&z<5Ms`n`(Z$_T(cB`)56Lj&++CC&-ZUwN5+$*IiWhRoneNR4jkf
z^dA1KumqU!W^eYZ2%JY==goAdhdtV_PG5cKJ=(`x{5HzjZ++<fwmtc}_G`S<`z<f^
ztFu-g_dRjTZHGe?+J^l@sp6#G$k#Y&3|meb{Vh%!eeXhX#c&L_B|+i^%`_KC(Z6;@
z=5J2u1%3$57yr&FhMTTL(Vwe4Z>rRLq1%n-^%*&lU#^eRm7jJUCUkbzuFkAa<>#6*
zukg5eIkUT@`EMT-n5vi2KtL9k5qu33m&8m!Uk5a;E|K+Ob(bnlKpuZG(a>N4gwZfO
zd>(#&C8aX_y5$#2sWrbWzju74{f!34BY;N$Qvvoj7J(_6DVoP9;;ZDB;`cVw4tc)e
z`A#AR4$K>PATS+h|L=#Nb}}p6YiBG9)}d$@I1)=kmu3k?&Z$8P<d@T=hd-Flocivb
z{or8H4~kgAq98T_@A!XpNZ0W`D*?doG122^{2qgH>^@`{P<ZKm@KBk`4ocTInN4f`
zv9)5y3gk8Fl#GH^SQpE8_`2MMRYD)Rt0+AvBBP{ZUQk}Mzh@HS%E>7b^j$J7tne5e
zcWF$#z8Rw>Nn5r@8!MF2N9;MJr?yrYR5Lv>>u1yZF^XM^#0`IRMx%tCsSIy&AbJzl
zFkJ4tyB{BRzt7^D6bkZgc>j49KH$X_nnM<}73gsI<Co#X?rqODM~M`Y=cIomQ^Yrv
zclp8hk`0l@5H$f0{7?`MhCrZF#S&yUkei}CM6-e}JmaFwIilngLOsVZ72IVXVbHJq
zi)uHNch%!hT4;ZEPsb5pIav?~1wZAz{Ik#S<$8G%9`#DS*meQl&OCiKXPinHOt?+-
zFcxjhFPIlj7FVB-kh9^TJdeaJ!=vRmnNaj^xUrsknEpi@8cMw^MQ3n=kmdoHf*~aA
zkQo}M2j@b{p8BNoQzB;?{dk1+>%2xVtyOfDImSA%td@T^uvR{KIr=7t3i!0=%ESJ~
zNH}6!xJOEB6+LPyfj-fK{V~X(3c$No>18dVvPBW*5|jxGHf^Mw+Bjz#PI2VysYTE}
zGo>k*qT+{XcbzMmB7K#vGeu8%&9~l)LMeOGcZ!rT)Os$`OdT3g*&ybWh_F=!5%E<S
z77U^iN#lPa?#I0xR@=R_K;(@u$}KHWc|uafkV1A`6gh==u{TIr;o;enzD|E5yZnMM
zaQ?NHT${0glMXb2_mA)YnKpC8<ta+AxOU2SWy(n@wf9y^X{ID8mt?u4c{OJ}awLrL
zW2TOQ6xKt%nj*INm-AR&sSRtUr{rwng37`X<(q%NX%u%^d6kvINl9g%u9VTj(x?@^
zuBwOIqQ}Apmi5MdlX2yf?H-TT#;4dyT-bVb#@4fS$tepj-x)927GqwNhm$UT^v;-)
z?&TZKV&X;M<EVbw8K3RWsQHnwxZTcp(f(D3$J(8xp0aZPs%%^<rJmC7UrQ%cKJxYl
zUL=3V*LdLlec*xjLmbTv_?$<>|A$Bxpa?y)GXT~EwwnJ!9^3u!+s6<8djB6bV)*#K
zFqBWtLJ{yO@pHaaT>N`!hEMq(h4ZX72rsokKEVqofe#>M-bi>q?2tnMG`WiUDNubl
zkMr4WnjDYnHVn-CD@xHIFr27a+^3p2Dm8yfNG235!d>8!=Z0IJUuK2d@T%|2P@eVQ
zVfSI!z0E!)qipPi$;8+j{#)`XpHjhUsQp?<$_RzZnBo*~`7u30t?iwL3gZl)oKZPp
zCGfjZ+Gum3RCX5<^C}z)X;iJFAMrzHD^sTSfglfq)`A}cj!($WO%9&kzyF-4edK>h
z_tiDAxdFu-$P72=mC!QAdhEmSQ?=N5OvB6Q(#k5?&X3KoB5CRI>X@*|w?m~;GVNxx
z?d@uJaJsNoukc6T^{D({`9bDvRQzzH5nW8kqf7GTtgcnb>#lS;WA1fjzZtuol!~m}
z2~9=hsXih&F7UMPg?@oPqOzO<6!w3T+g>hh-{)4|3F&Dj_e}XEaaD&}#~#njpIu{Y
zDR8G4Sqj`K3RC3%I<&*`?s8LYFDoz)E(v&NG1o^GmISRp6=D%JDwW4H-NC8idnBTO
z#Az6N1-C>LoaeMYlkf$eFl`N=DBTg;PNEBrkJ((w^g3S$1qm@L=RxvVNI`#j-U0dJ
zuLvn1IL>Vwwv1bb8vqI|!!vk7C;fVG!YZu_n$QkDpDNY@31g3YV<16$!)EIZqoU(@
z1t38?;(`fZfP}9JBxsAAx#*=`@?6^`+Gs(_{N;dzI`CGuNSZ0@8Zou^)=H_TY!4)8
zQi6=V7)W?pdGw?_jUdc)c7A^*=m8teXBz*vKo2x{Vf+htaCPuNODF8~J-wIR@VV`V
z%3EQr`bywI9Ug0UlX}X!J*&2HrIcpMBf$d&T6lL=@PLpHRZgk`9*_~n8a|bm7CRr-
zL+q$=;z(%FgOS>x2h-#c02l>Sq@u%L4nELM@FCA&iwP`Kzft%ANhp8to8SkEyin4v
z#1G2oC*cQ`$fY*yWZXpn0ueq5Kv0TVp;`$bsL<*1w*Dr2+)NHtza}2gPa^|d1S7{}
z$E{E2C3TnTgYH5eU+Sqe4V9z)SRMMs)JrY;$4yY}Z<I^DgVcrX`#xmG5q2saGi9NC
zpT?f+iz-f5a;Dw2z8!zgI|hO3rG?)cB#m|Bjp!r42UG0M6(vi0wY*MMB=D$Ma^vWY
zrL{YRPkrmxwgku#nB#S+F?9#}R!znISU{?DWXZVC>#DzaD4q9GoWw}Dn`2(Ce5qJh
zrA6*a;)sz=y?xh3Naip*cC5i~sz>w)ev^~HY5b<z;aA`{;EjI>yTaA+n<YVOeif2y
z*>doiH)QuE)*gl3$WU(e%3X-vlyR6#&oQo+GDQ<XCo_dc0TO18E61oUvIDQ4GQEVy
zq9%qbGp%Q!Nr+!#zX)i;RfoVWzh~bv+yFFb8J+=6I_cK~O;%}DK$FhNXP?itfJsc#
zv4yZ5U;@NRbJKqx=5!pd0GI?xWb_MQ@-+by;Ny~!_|hVJt}P+}C(QEilfG%dqz=55
zZJ}n$y5>{uy|q&6Dcb`kp@=ZfTnd;xq8fTiZstPCD*MbZ5=X+CZwMlJgoze~gRPK}
z`0T6d+}3t8PPbJ1D_6pQ0!usD-3n!HWOH<KWnpa!Wp0<=KtKoqIG0hA7ZaD$RY1!D
z?YEy+Ko$XiPI8t6Ne{|ad-i8jz+i-7Bc>E-3$;`u#eyoQq$BkF<v)S#Zsc==Bt%I%
zJ5mIg#~nqdln(N^{>4*zJzq{ShjJrdBs*=g8Q<Q93UiP^18)f<{o(WP*?=BiJbpet
zYc5`*nFna~FTBM~xI|ceOv=zMTar*w8Qn+#%Xz~Ql71UtlvcRNUf?(}1<re;O}9l@
zK=B3x5kZNUabq461Tix+IhS#mE))Sdli{c)f4y4EjvKoX-Rmp*2gJ)_@d;cEGLmNo
z$j-*>f-C|fb{2M!d$Y*zr}%z!x4LDuJ%C1PiY!(YPd!CGeVzF9cDlTuetShZ8D<q2
zF};78fWF}LA2<zaH5mY=5vKR=)8BsQTxfb-=p{Kvq?e)3mfqFV8Nt#Fpcz1;08eKl
ze}P7jMv&$RMem$mf<Bux_L|mfT5q|Oo&wPZnh4fYdiwkOUr#(S%$d{vqkkKpUfxfC
zoJRJwk&Kv}K7T*?FC+p(D<;L9vD2UboW7hsjzpKYYBi?pEsbd@7)X~j=C!0qa(Sh9
zpqHti<#Uj3>kPDzOn@mANiT|YDs5VAe_?U8P$c@eSk2hE&4L038fH{T2qpZHn`N2_
z3_;jSU^R_M=U4jqPoVt_0)!%gBmqInMm>HGlpTa$S)OK80~J5Vf0L*Y00VUHg$!!N
zz}gWqLK?;mZh?l=n)pxg>CdmzF!Cuu$Z^s(*IKDg+ra4v<fJ(n9T!>K@<XGee{0l}
zfX*-y2<8+eC-YS!#kszcuB7KP{|oGQBgB58tTiD+jFbiM_$sr>4z+=Cc5w&S`}eno
zNu22lV0nK>v*=LA{cS~D)^2$YW_CAULb%{u_VFc<><Yz_UPe5tm&A*98Bjt+Ub>M_
zZ>#tc$}ywd?95!+K~x(AH*>yge;5<iu{mhTL2*t-%1M=wS>LvfEz?lO(Lu@sO=F;_
zF84!GI`a54-quV(8KxDEvmY@9LnEL^ifg8jOmWntOmOor?|fYXgQ}X9Drwrrvt4i_
zKH&2#0!dl_1rfTDR)l=K+A>2Ro+-tPL!UR85;?FYhSVi8>Qjo$)g7*Se-n&SGt149
zVC*SA0t$r;ogFEzfg*#lpB>YT=n<9=oCaQ>&Sd+KaZ9-c<riuvw?Y7u<n%^uZJ+}o
zea|a|y?d$>1ia)}Xgo|I{R`Z8BMtY+1w%g5p5Mq*9L<TVMX3-4<93FsVoe9&m9jW%
z-H~|hDIS5>p5hi>2Z;tJe<pR#BqNqX-m22PCLy0=J2<imZ&AT<0N9OY#+aR=3Ja9S
z!%Wp5l4+{QHYPDUU_ao90hz=0TuXL-Nt<DO7HwgZ7?KI?GWiehyG1#Rpxax=vE;_p
z%Plqn$-SKHiy{#LZ6N8+{!w+@I=`Gd{iLuZ9XT2}vb-259^eKBe~KG!V3_^O!`41-
zbJ!ZSB@+A+BBXB0V3#22;u!^m=&?+W))B1f8jTB+0z?E~P(pGFZDKw|zT+YdX_|4T
zj)YiG@d$|Z6t^JO&;Iog+r@2x*r2T)5W_4HRO#k8c|a2M*dW;NlLSF|Ff5&WXg;RA
z%5yDJZkInFiDPLQf0!dP{DGocXb-O9>BqylM2{jni|3cLGf(dyVplDl^$%2A@L>p7
z_wcG1aKf(8Oej|i@-c&cV27j)E*Z+w)b_Xur#iEc;vG#qjfbR*d!rRh3fx<>Eb5q7
zy!g*Gm59#2%O)N^4m94Ss2j%-m1Ko1net2Q;Md&QXEK`Kf9mBI^SRv%M^p?~I^n+k
zWv%dzhh7%5gtO64Z(CjS`8GGZgfXS-df2wAs_l7ix4e62vaAiE=xmhj+3cJ0GE>zl
z9mu;~*PHVT{hZvkzvgzle-eonQr@lI?<sy<%GQm%CXvmSb8k0Zc(+NObm+#D5mA);
z6}@-_+R7f8f5{9K56EN&iW`~Cxc~7zaBzw_i97is*%>%9)AXh_ZJAOVNxN$9!C;CU
zm=U@(Xne>v+0)@`a^QZS=O(bfB<G%i^IYpa&)3X5-p>~m3C!r)4tq_3Jq7ykER(`Q
zqa=qn?%EzlkYMKFvuybKIG@Hb%WcPJTj`rr?e7lDf4-!7n70`1gtpAqUvH~sR=K=8
zfCnh9$MB_gK*`7s5|TWDU6678^z-I94m3JhjP?GT$o7%Ak(xOaEpBqbBF!!f%Xp6#
zyPb_tNE!nYXk;M6je<WWn)o(_*+hUmv6}ujg<%Frf4*g`r6CdwmA2Z8dHU@?U%&k2
z^FM&mf3pAjsT9>HMFH~T-{YgA?3C&lF?~zVnrWj^Prw{mg`AXS-Y`U2SOz{%Cn*f=
zPAN|v{pLp<lU54REDo{M_=$OgW+?OEDVEW^-?w%!b5*wpG8H!blc$o_`bS!ySNTW(
zYM=8|TJ^8#^kq7IOj=c1lYb*LSV4i5!l|)Re=cd2QdTest!DEOY9^giCTqnR!uZN)
zD{mR}QA$&Elfp7XXveASR%Fg2jof*NXk`j{D8G;6D|#4YrI8xqnR2vFQL*PdL?LN6
zYPf#>{5`GvN~_+Q_ekdq0-02LIty+YQW;LPZryzVp+H{0Nz?P}Y@s<EW3aWn%aU~s
z1ZcaDQ-8xwl<Qmt`?y;f<UGim?oh^<msBOtVNlKPiV{{zlr$YId0DJK1LxG)sPrTj
za!Bt!(S;;MH*V~8;j++5;bO^%g&cRk^S+M&Bn`uiO3dby46V~`xT9}%_zk9f(ly-*
zr9>E2&rObbIjnFK&X5OjBj?A|Zn17thtzJ~ZAVBaW-5Men0(EuCZ!7)Rc_nE=&CrQ
z*Q&U9AlJg>473U28MMn4-Ykb-FyY2nmQ1*sV>ID^PyMtMX3tHW-Erv-e18k|y_Z2}
zKoftrnic?#A+MOkr*P3RP}5C?bwvBZGNEy{<Kg;v%0C+Xa0kOON3MKeH1#kJa|)U&
z4eu0(U0MsdCL_P@;@j7L0#?{QqaBU5qQ`$LM}v*?v93{~FCP@o>-=bF-h?&XFWs#V
zE)>Ueto_|s`lXq=@)-Q}OZjG_uu;0qtcQO#Lq5G@EvNdHQ|J{Dl`Qwxuk;b6PiHAG
zCl|SYt-D55eY!OF=P(^~*Vqbf_mZ_$)@9eJ(){PXm=Btc>+`SLwmEm*XlUX7YN3*T
zdC7htXt`+k*4@^mi682yX+z;>e!bOKgzA@c)U0F%@xX>!Pw>FItv%t5tF8Uy18y$1
z_5?RBwhp`hynC&?xXRaB=Q{a`w_1nOI^<GoPjm53>!PI$?_`Mo2L`O}%9pVmKNOcC
z=rj}pF*%p<?gS^7(P%)Le}eVb;g9EE?|5REGpGHZ{;R!ve7^hpu4P{eWKP2L=eIlm
zKnh^O;)K#LZp8Hcuj$L(Z!Ouf7^ipQhvL)suTwMfIU(pcA;`5>szMMr!73-M9TWah
zL{L6(u4reHb}mnJ(QXl{U4DGBuu4&Kl7(t1s$+bcNzYgQ2<%oPf5h%g$uW=}Z^$b?
z%4}trR-&C<WO=%LeM2LlTZ#bd>l?|mfeuMRPMUNPAIilsrDkNMN=Ospob;A%xx(~L
zkix;MD5vT0n6|Mk^-YnJdP!kMioB+H3-n4b;cQFs9Q0ZA^=zdhe9D}Il?I*;A2f&C
zxJ57wLG9!g5t!7je?x0SF$ARNd{z{%5%UhzG2BJ&$!~B5=_7DojV!nq9GT!9dVZlm
zQLXMmfbxpay>%L=pQxAuH&Q<viu!t&R8!h=N%iazE~%!t<&v6xH!ex)Xcr2i&3JnW
zS|}sZVZ5`U;F1<*bg|KB%joL1;^Qj#*xJO@3n*K~=xQnRe}{E*bb=Yzi8L7#nHJtA
zG65715O#`8pkT^$ykl|#K|DHUH1e1n0(j9}&X<Oee6)P2Rs73aUZt>;T5v>z!xI;b
zz#nP;HO1gaJ$~Yf@RSU@rg+W`Xw5)lJNqU#V41^iN9Wvt6%349+kXj%9#INF>h#*7
zQjy+pxs!8Le~uZ|aV}C&4_qgGDEGliF6YZxMhNWWtAxykFOk+yq1y8noaxEe9y5|b
z>$3GGzOw@^mP{E!vKJC`;Jq4o!GZ*7hqb+SW`u%^-(@+%Db``)_2Y`rDY<1@uh>jy
z#}`rc$rPGSZZCT=DI|k_W2aYfWXe);YUMyj)=*Spf6f}$WA(y_v&kx-2<DHluWDrJ
zZ1O%R_b^x$R(Zwno$nv~YQ<IPsBqaFa)#?mbMJ7%m8tkr+eLK3z3$CJJ;ph~v~V^d
zaIv7LS2#R0>U7Co6$9=j<sA=Yw^^M;QKz&o{+LsPg{q|yU*d_!_(bv>cuBYm!>qIg
zcb)V{e??sil8q*v>i`q6xe14~HsRsKa_Mox3<=~ec`h`~|9w6giX?Zb$?Yn_C36OL
zsfg@U<XE2Z7iQIpl;=H#Orux7TZJttPvA=KEl4p-I=L4g(K=CmO7GiPF64k`<8YhC
zzJb`JLB`#DEY(5m`9!E77H1&7JvI=Yl$wn^e}ic#&Pi_EXE0xWbqtWgF$8;^($8sQ
zd~%I~`MBm~mE+j@k!!)~<zr(>CTzs)-cAP)!S1|&vi_{t+|M45JXShAF1cMOobx<I
z3Iq3TPJM{aPKBDxRKoF8-r}dc?Xt$MqP(naTY5>vsH{bwHPK~77IQcHxhPf#7FyCa
zf2BLHHgoVK<TEj2&XtR2xIA5MEMr~TQhK)r?FMpmy!5ahP%=f8LVH-RsU&moCWrO3
z9wqoN&}%dlR}P>rE>Tm_+fY7NBBzt8X7(*5vi0YZ9aiT`Wal$#+pYf!g(f;m8|+TL
zkO~zPxQ0uc*{jE!3&&h;w=z2$k$34Be_F?SrS*wjKEfVYY>~SmlM^FNqo#O6AFH9X
zacJDk9&l*fP~12)ZjpSqeJnva2`xQR<Tb?seXNG!Mjxx0{pNkFeca|gR*Tk;?qfCl
z?CfK;Z;k6?QHI6<N9On~#q0Z64W%uYRL>sal4^=uE~(k~yXs@LTN&5KYAA2?f3cd`
z!~0nKM3(zlt%AK?AFJhSUmvT*cKq=#I+EhZT)iRat~UOf(1z{TlLwf7O>o2V>&e%5
zuDpvFVhuS}-n9$Vl93{|7>+M4atPeTmEf8wWm>K_>-9^E^+P@sxzfOM<EYuxMu#W$
zh3c1H!@P}P=tj1w8;Wn#-9b&Ne<Q?yqPv3%7zKcL(cSUq0aEE7(%nIB=!yNmy|bbt
z;K(}~_Y<{uLM+wzbGLUuQx?p{J8AEL1ZM2WqluQHyV~K?wwXo!uW98d`%){%7e9;E
z<=*B5pOvu0*kjz9_jb6mqD-mU=D#UuhO<j4)<NME0q^%0u(VuyBUq6ef7a{qzbn%k
zT1GEGbFFKz(Vy|r*q?FNi$-laV64vle&K$B3ND|9GCs`+@9FujAmx_9v{0yV%y%XQ
zHZf~(hk2IS%Rw;Dl3D)?m2#+8=A7y`l1i;hN>cqnKJQvdt%9-C`zmR_LG=}dQRJ;g
zrHZ7r4=R%c&PI%h`D2bKe+9dR<YRh#K)j!~nCq1A_R4SU7W%R1?Izq+rEi5|?Cmb}
zOvSEl;z%s~-{Np{84XMNJoji9`c`{_Tv-LWc9?w&s{EDz4PFiOPOb6LzLbEAO?eQb
z?Z+(OGV_;;Dl}|)3Kh;$%sPURYuN!{5ZU;{@9|(@_*HSv*8BP?7wl~hIN=wH&b8p2
zx}rXRKR)0618Th%m%+j@69X|bG?%gM9TS(Ha6rg^Re1jW^FL3rFwD8L4h8YhxC_r<
z=-bc?hB(q7=?brOrRYO^7@8L`Un5;3%^^Q%B)T;6`CH@|LwlFT;z#_%JMmefr1Q6u
ziMwAuyjs*@KoOkhTtX;3n9N(6!<2cVP71&1^0j#_vn*Rm8!t2}K1-CY@>wNMGh(Na
zc+d8KvXa(x(tw}tDtN6LYubvIPP%*@#+EtJ=Xa8ZuSJSsK2@x=sp(bv$7E9Fjco5m
z!&HiCj!H%MvPaz~ozL)QwUAyn9bYqLKu8kQ|BRjZ!Ix^;{;7cp18H=68qrTT@)I{9
zgb+~Q^_+TMAA(Rod`W@y@nGQc53QK_-Jd^yR?(^<%I3I?ag`=ZT#Fpn?z(WJR9@M0
zzVm9wkUdAX-SeRMW%Komehv6wgHR4R!_o0z2;Z}FssPH$y#+v>u^K?i$d}%nJ;chL
zsFqM+-{v;_t!m8TMrL3$QYhUPbmKx6Uf&v2ziEdR9MG=P_;_UsILmz~ORve>p1f3l
zeKALno@T?hEB~yb%Zw(*veXkS2qNa%D0ZModaBfI&t+TW`Ob#Mvn<VVY*EmKm(b){
ze7m>$DEGuka(P~N3!drgTBU2+u9(P6`>Z93J_sKIM@B<+v~1)>H|rl+M@Ac=5SFMG
zTrZ*%yhwz~=!$OD2hwQ&<myUjOL*{qqsAkC%UfXz-8q@)ZCmcF*fu|nTtxJhE8B;a
z$2d(V1xYy5#(fq6X(;LPoN8UsSzApv*W$05%*L}-B%AhR(O%X4NAj18_SI5Y1#78Q
zwexOjl{HN)kvvj>vEI_R4N_-qFFyI%rFxdE;Bm%}Ef|311kqVRNhW_Q%T39DiJlUg
ztUI@ewu;D#pd|gvfTS(Wz7^?%jqJs*AMEP=^XY#x)nfqR6kh)L`zby^Pyz^%N*Q6S
zyZrO#<*%n-IoUp?Pgs@K8Z)G$iZp5PVuBZTitjZezZPauUZy|EM8lolG#W^`o)Ii~
znps4XaD-(x`)+n|J6F~UW%V|HSA+^jZxS+;kcHrIrzc2j9F=7RHI(ITCd0>t*8-ug
zp{D$AnM0}-KzjDvV$MAOX}-o%qHnpn<!i2@Vl@^6m6hr~tAf-5BpkCUNGptSTsKh=
zEcz(&LUl&(`F~QyN&J#|z~MoDsYtS#>rj7m0pE6sV{vDv^Ngs;(S>7wJR@2W3^#~=
zDfsbSiskdYP5vvlGO}gjpSl2sJ1OV|x#>u}6V|HrijxJq6}JSI+?(RmVCGQ2;0crn
z(s9aKu7r%T%;$W3)jxWm@EthCH3vzf5N29^fQQtS@HveK`O-S1E(X`)%aqDh+|@)F
z7J{L9FMQfip{7l7y#1YjE!@=*Lc!h1X(fayIM<b+LhM2v!=E9!;_@1)megf7w<X;n
zoPiD`$-!`nYv@3v)PFYp1Hxwo?J>40Xy<A<)(qIK!&4b?0(Of23<~GXS<s2m#3@dd
zL#IrQAiTC`Ag3aBq_}3H>IK>S5AZ9J@~7Yq@}(sTG-^pN#!uycMBf>O#m?C}+#<{%
zLntBcei*gnXk_|M96@TmA*b4^r#twK!IkPO&aWN%KZkfU5NotKkLM(Bhal6kw}(R+
zl@>wygM6WKsZQs7y-lGpB14<LE0n|IyxBjtj!Vk9<WC4R-nujW$w+a}pRoBK=uZyu
z)%?jgI__4*7EJPg8`T&!Z{}@_m#$THzq4os8{_CcXQ8PKpdiYrZU99`gZaMLiN)d>
zLA3&h&2VQ_tvSU=bxFXQlu<zvNF^I1OUZ(PlG$$HQiEovIoG;0l=7%BE%vE-bXIdw
zX(X1A&va*VF*el9jQvWmrSdo}-?DAPx~ln(>$gN-`aW!b*G0SD(h2r;E7I<D;nLp8
z7kg|L1?X9e4SWy=!gh=8Na>u~aNE^0Qn1i+YE6cc_o&V<P0ByCa4-m3^E1kN3p2Tc
z)WT0!x8LNcv+rdytGCHO+*`*0YxxUMS#%R>`n|4*%&7sBn2kvLLTGVX7_Dz%*efAn
zCgM`5+gdDtH}n{bCqr`TnoQU=8EeK+vxXGF+-7KIRPIR)N`=EHuB8Ufi$EcRjm|q1
z1&Ds~1qp!HB>*-Z&h$euL3i~V-Y!vTH*rdGtL$cj?*Lv6nT+C2uEj>*u#J8DO7_A@
zdyv1nikI0->miiSb@e#NqMgY}!jwtZ3_9(-=(5j$r#pxFHDMdN$Jy4fP&y-=w>!WP
z4CO7rDXu2CIUQ!B9|?xod5-oy#-C9!Up&BG>r1Z#04Py|E@y7reFXqrdKrxm{Z5P-
z(+v5#lof;bB&{EJ28TWWrI#F~rgwRt5~4>!YpXw&+`+TKQE@BWSt~H8GuRs`?eRB`
zHmj9?2pB`@l|F;Pk<uQ6<7n*95WuuzR+y$Q<KPUYM@oB4kE44RXu`Cww!$>Ttyw&S
z>5<YN)8pv0$lnezthGq#>}e5}QQV2JZ1k|%@28#GrNFZyP&_yr?3n_M6!!#T^ZSBz
zI4n@A{!3M3WG!F2J3nW65)l_}I)f^0P1R$6iWg4_)(@q3^hAH}JoX#w$z1CJpEfd#
zBkEC<{4~xm9FKE-VHszf1Y1t>G!=i&`A6p*rMtmqTeD^$D|*jIFTryehMH1xufo42
z)HvIoP>i<)h^{nzw2f7Ki#V;|x&g42lA#o*+x4>*Pf*J(*UHV~NVQ#(Yx^gy)9BiN
zoz48HUq9REX2)CZ*xj$1!JM3S$Y}>uY8k3-o|Sh&G@uB6Dis?J@29LF3ip8i3Zr~a
zd0m_}qqyqY^T{SI*@JsK$_cH^jK_^uNjYD@_4Kbbcf3jE=nWjl6L(?JvMbG<*=mtm
zDCxf5b6b}Eobx(`$LMzz`WsgC-kAx1M)9sjQ$BseD6Te|a^cSBf50A8?r<Dqt1cBb
zI!*6bzxz7%zpy)vKkNhQ?gTxgLiQ2pDE?A;DU=Tn@Wd+qLa93(eV!*~luz))jN*<b
z=8|o$_(i{=&<WrNl)vn&OVY=PZaBVLx3M=4y6Q!NVt>`(8Ff!NAjK$dNmF5eF}`sI
zEEo?vuwe5)Vrx8axE$lFHg%WyQhfs39jhtTx(@6&6lVIp55AY9rSiGu$6ch{*oIY<
zAwV!E<nKz6BEZh`cRj4d8d=*Zq3)zms)2lWg0|5}p2|!=q9}J^^uY)%?=L^ln8xuK
zpU0}@N}xNPFxIy9XGc@T(EYf7h*cJ5s;s*Tq_3^&539Ou*U`g&!3q)tIybk%EkC%M
zs;o}(+w4`()cZb@nyR=3UBWrbAktCl^cpN^Vbqxk+eq+)d~GDSldg?}&&ZFzyrpr7
z@3N!8tgyA?dw&|{l`v*M3tK#xz`ox<pZ*X20Np|gWo~41baG{3Z3<<Vy&OLX0yHz1
zQ3(PQmv4JOD}Qb*xeeX>EBXhnMi1%*G#aoyc7W_M@Gi(I$joXMng3sek|>E%RdrRn
z>t@k*Tcx5%ijPA|mX|*+^76Nrx38C9J`G&F2v&K0`TG5$@PWjiO5&ii7q66XMqR%C
zy!`VjrFQsv!w*Pl41PR*5BS^Oei#*ShQb*NN0qw$@PD9i#Ndd*IojfHi64#MJ&k?R
z`wrg^5+G4P+6pHs99OTme|`P^MH0hOM3=s%U-MrdUoZcCnc27A3XuNt?>}GC129UM
zpz$#XC&T5h|6G25`NvGQEhh0DVbc%%_SYYmY2-K3V3epq>72EL29-p$!o|5LOnOu_
z2vJIll7Cr+-%&FbH39GVTc@!>+Fw3{TBQt8C|FQ#x_vKHQ1DCtk=Y+cko`u{DG^ak
z6Gb)h(`P%OoL)%gND5@cqC}vJbZJb!WlD{L#_d6ab2bQrO0N`>E|PAtULKbMK!i9G
zRllasx**oz%rL)Ust|Q+B>n!C*h>SS>jHrnV1In2lm)R#HQx|MqT|__;yr1qS&VO-
zVHOh)J7%$5Kbb(?T=*<*pEL>|Wo*`Fbsggb_)say_jR3XE3Ar5=Zp0om%G}<NLkAb
zxk*do>@~UN+qM)6gzcX2P7$Jd#2fijGDS${`M6LXN0gfu+LFl!5w)ErKVwP}^GFC2
z!GA4FidIB0vrg{Dm$i-0hV{cqHMrvT7D$6xLAUkv{+eUN8rECmrs{1!v211);?+hi
z9IB`qy(K<8Wfyxqn^-TLH8aJ-7WjlL<mNUc+km!m&5r(F?r&1ra1Xy&Fndm^BL$;R
z3i(-odF?%PWRuH>9_m6*8bBeg{hh7k_J0FcyOKTGo6eS!v|4#+_C(s%HXvR>Z^BP7
zxJ_BZi_zcO{w{e%2xQTV9_BT<=MjzF)OAo=-<oyhY3~IpaAqhQRZ|c^`j*v0%;d!h
zJYh_)iX>;uEPl3k@@S(qIB}g*R**F`B&xQrgkn`%iE>=wYeg{$^nywtR>zxxx_`L{
zj&+DxOE=}N<V-*`3~fNkgGk+`nXj}Zu{uQOpt<(K9F^f<y}2;aC>o}%vvpOZtMqj{
zd}?FT(^oz>X_WY2BZH1r-3NFZ-h9vcC|W=V&}2J6feTjoQ;$nC?P&_$HArWD$~rTL
z@yK>+QybYjyA;#j<-kaxt0Bk4?*Z$V6n{WOe@bzQ0Pv&LAXb7xB5PAMh>W{<zZT4O
z-bVhUECxQ}Lp8uqLJY8)VOmx69cVhOwo(?zBorlML~@zxFmf+56QrG1(Ll!ZehgW7
ztg^kDsmO1j$6$o=alwzABz^*m5I&uEkz6T<lJM|q3mZWPRt#aD*9({^AG5w1ZXezc
z5u_?q^z*8Bic>9v@Y2n1JW-eEe?TN8A-^vueWJcul9p0@n*B`uMg`$SajSr<{yNUi
zpoVbg{p#r%moR`pIswm@Yk)u_fA2KzPphWESb@=Qr<&eHGa7_;o@c7H98#fs?hHK9
z?5y!>Ty?WKMYv*|{Y-SL<~LH@Y)=tZnAYE~{vXeAQA8gO@}&*VGI~uf$9f*Dk9HS3
z<ak!%mfmTtRdd3kz{v13+nW^_k1Or;Y^T{zB}g*Lk>ZvK)pR=P<=gGX8OI!y<(Tg@
z)r_}8vWzzyD8rSKjVf}Woc5O{fk00IBA0f7KwAWkrt~A1)qy}be_xq@r%2>1wV}(m
z8tJl?#$r+GO0mGLwvt_Y;#)&j*Cu}1zrGGxu49R(y<gkx#<?qtmF>;F2$Uj(Dc1i^
zT=$jV2Tpn6;#suID|bHZb(E$TIux5Te?v@a>$ugpRGRuWKTS7ZY9BIFkHdw4CJV*r
z#@o2gC|#bTYAolPf2D}IY9kocOk9K(H=L%Qa(HB9#!2IXN6(IUz2DH}01{edTyVm%
z0BaDeZ>d1Fd^_ao;SkBs8Jr&sWM=685(4R!>__y6?00J?LM#D(w@r;3BoKg2U1}3D
zlKy{OS<(Rmmj4-7mL|$4+)bG%?%Yk8C7T*jC)^#;5(HSaf8gs*YqazoX4M?KgjC`G
z5N#i4+NZVsBTNS?@N?pZ>KFu?+QAJqQe)v{*0lB<S~<1-rm;{8%AMGLn<ze2u$KEk
z6UEJfwUoE>`ahxPLGEyft=fj0X74ASu{zzqJ~YFbGmNB0TL<{t|CH-5T07xlKklfv
zCBg`9wIAEoe+R4ee5m8Lkg5)WE)F!@JRXbF3}(YL!aMA(krCOJ)fRfbE&=>{LSQvy
zudW+Mp3oKqR#}yI>XnglQR9u*9_(^KsNIDsj}6^8cofN(1(-Bk9~BQ~S>A>xeT}&l
zc0%Q&5kmMlZQJ*cz>XDmF$#}m<+y0qPS<;A6^Sx$f2v`8m#&{Exr&2TbRx*b%`w}$
zE*;NPeG!J<2O>PPa#^P%JipW5=pCF~e^Bpb=t8mjr`kQu>z7VW>vmu$QC`h@Il9I9
zqyk#k?8x?FSuBiQthiCzrnYCwiyY&qV)ana^o)86#H6RSYNENMb^5Cz{5Bifw<)}F
zKdm)?EV{Hrv)kq5g8NAm+d4-LfSYAogwzin=}LWcY+rJ1v3dag+3)O(CMKnmcbBw;
zKqUcpm)e9tB!7D6t>U48pMn7JBVxB~Pp}rYr+aFqH6B)V06|1{<hZzF_du3kyDi6X
zkPY*_x|-l~gTz!CYQ+{h<@>3!QMUEwpx(P6%OONtI|B{f-k)^IIFj7K?#w?g?IH%#
zez9){n<)Fc*>CF|ex?6^%}#u8MrRPg>i84;vPxUT$bWub={z+xqgPjYWhJJbs5%|f
z;jNklDzHNhp39<;j2qUA$(i-aa-_*~SJG9bqFPKX@ujz|>$?XMBUf~RJnG6@xs9_0
zXs<2TvG&R_v`N|IU%w`rjl%cKRLgiRn~W4yJA6YJ3G7Dk0vzs3UqB<jzE6?b^}T8I
z8O-F3nty%-IKsr0W4~e7H;(+4`Dv$}kvbauRy)x#EDG9NRChp|Ojn~;4`m4j$T8QJ
z(cwT#GALKxsXGa^T5nHCUZMi`!Iv>Y#s!ZLc=4w91|x#jGv$VKK(f+qLPPffBO*Xy
z5h|}y-oF#E3B(Z@tx(-M3<<mtPJ3oz&%vi<K7WNpfeQ;sH*Yx{t0)2Es$Edc*fsx$
zgfR@Ujo5OCgB&%I{OlNkcGzwn!bB@n82@8WZU7g$gwZR;3U(+kEKhZpF!FvAp5IUa
zeiwm-ukPlUfHZey7@+Gr5BB%ZN!Ei@-C*!TpjsrV<(+l<{qgnkKh6B8`3hxjWOH<K
zWtXaaI12<bGcq`rAwM}KmoA4u3zxvI3MhY?U5{J05r*&mD|ne+l;YQr!y+)?#B~y&
z1q|B-+8VxaHbU&gSs80NP5*tL;gOD4kv7>~A8!hKr8yEs4u|vc&J1Iy*##GKPQ{fx
z2yt2m!E5S9MP6Gns))-$jgsI>928@ujX82~EtEk;r@7rTD!Ja`_>Ae}?rUCC9>0Ii
z!t3tmEGS_xk(-OvULYCPb~C_7pJph*r7+D<qD!$wTdX9P@@_zOspPSd(p-`{c2d-(
zI&GkINE!2%0@=p~EAwg^1CUJW(-M-|b{I@bahW$;O%MiCj8ChByRn2c$K&*=x?KC%
zMNKYio@VHfbud|*D<KUg*<pzjzkGjJVwrg8t|T3MLq;X1F+;78bqK04gQT&VGN2A|
zW|l9tjfFY4qH%1qWmkF`16pws+E}7-d}%kJLKd;rRGgGBW>9q^ahi-zC68@tc2cKf
z(dI;_b5rKkG9=j{d$$1v$<-L9u^n<!w9cogRa=c=wS+M*$53TALv^*bArgOt$YNvj
zT_I(fAwzZ>P}o52ISo=w3kL~PEod2JaSb)OXCwpikZ%dqwbJ_d8Dr~^1Hy-tFi2FI
ziH%IrH5~>~FqYPb985-SzfE<DS`V^2O=6NXBT*BPk$fuC!VJ@@V+qbwTO3THZK^}<
z3p%O23=#$DL8gV9>zKeG)pdWfS|hvTT-8=X#kx!bu)*G@0cM0MLr_>@?>Z(kD-1h^
zO>aWl4dFH9X$*E21Nt%OXmlNun|wnJ`QGQBuO2=(FA(q}Pgei@*T4Cqc|`?NN#0cE
z{Qa9ZFYe#}&7m>*gwU=rG<Jr@>kZA~J6iu?s9Sxxd9%5A`u647+L?dnd*W7)Uv9VS
zi*uKz$G`rv{p#uV<#z4j^ueRe`Ibc<4RN7Nw#p+#7WkLx0XvW8D#wEfUr+@e?%!WM
zzSx{SU2mVS9zT4vdba*$yJPv;f8VZI`TI2top0CY+jqke)iKAD^}Egci?j8+opZ9!
z{=WY1^~*0dzdW~X)|h|Ly7Yn>F8D3OQXAZ^9-N<Vrnw|cEZloO`DokA<fHdgoT)TZ
zd8W!t$xPLmYBQxXHLp?THOjn3nb(-4_F|U8V0rcQ{kL0t`rGUCAKmJU&Bb@?i?P@6
zV)gavo7I=k<Mi-keYSm$?s3B1JdCfrr(8Y0&DGF-4*A9Efn$IAr>n0v&o--v?lU~A
z#)g?$ATTG6sQ+~T{-v|H4Hqc4H(#*l0t5Hvw60<zN-L3X3A(uGuDG-%m?+*Np=k5w
zMsPS@%u)B|@p5;>1rvI~Q6XO6FpiWnqJfLnJs$api)u#^Q)(fk9$`ioU;-cRBR+G_
zgoNU|T+KwZC>nnw0V!b1!~`PR$8(WXykI^ecp*dr4DjY_ncSCpH<AFBur^SoB_9bI
z7@aSGovhKfV;*7yl~h7&;|CxX!g5-I3B+zNOVQ9^5UL=UWT?v5Q<_=3Mh!&Egn>4d
zl$SUTsovsms>R2#di4=Jky`eI$73!6zO)fYrD;GgV3dEn02eLzj&-I27lt7dF2lGH
zxaj?eivr*V+yx+s1Y$Q@x`ZguNMX@bJcq)85TLaWW7T%TKESc6JrSL|5@2x^vV?tf
zTm|47br@h+5?&Kvj`em6YQGi*xH-g%afX(F_rRAW&Vzta3F7FcfgdxJZcwtLtMynn
zQ^PnC{{eshp%6&g66euT3g9)H^JbiAq{T_J4>xu~GaH!_3duOWZs6D@%%e5mEd@a_
zg~ZRo$q<adL~F<H3OGuBgNq30Gm$j(nTdhJid@2B7Nx5J?<2T&Fl>Tr*n^K)>NSQE
zz<m7_xb6~qyBg%(7}wQ!jO(P|4%dx@eUIy8#%q5v<29M_n#_1jX1pf)@#u?O1E@hZ
znS#fGku-8}OiUu;RqNeYFd-8~C4W|rcfxxp25g#vKIeEx28@wN`@jTviuZga6R0nF
zF0Z7|jZ6>0l@nZ?UKq`Eald&yht)?im?D$#8vgCZ0B)_9{^H)jMph5z>Cd*WH|KEl
zk0*cM*xzTbw%fPw{<2zq|9boC{kQ&X^V90=)sO#S%;N{ozJ9uS^Yi+`|L_JT?|m#z
zMOkpLdyY*+rOV9mP5jL56*sBR>4Vsc5&)xK$waO-W?)IzKmviumUQUs>W8EgLp}6!
zci@vOBe(Lv^ib@fE#R&6*jv?lJ8w08u=9UbwJebN#p0V)C^*8((WPuas#uMXr0W>N
z<eow^QAn^V2Y}{R@v63K#jAs1yC8%SDn|FI%L@nRe^6_<>ilnf<n6?fw_6^$UM@V;
zIT*FM0MO>3)s|3*WcjaE0wd@FFk94ouOs32p`7x;Jt`+H8aqrul|`Uj^TahPyMlk`
z$#qIZEr|@MWj$33C}L<Xl#-J{Yl7nvS@=#CC94krOUpKZ*$%XbOpg7Vl&Z!4Qs|@`
zY@#p61mb1!HY^k|SiRl{Ic7pNlAG~b6);MkUfQ9Bwsr?j_G*j!G)`7el=}3Rr9Rb5
z7T0Er(*+7`$$)m5jSGYJ)=jv|xITZ4LMc$8Sp!9VwWdq7y0Xh%EnT{~$JgJ>Q2s!r
zy0IR0Io6||ucJrSuR72p>(hMDBc02+eJ<zrxt!b6a-KVr_Z69HA=P?j);9>p`7dC(
zsPrZstPS+$-x7`+=!a|~P%vSX9Eulf!E|z(tarmiOwA_{Ml@?9C|b=*ts#F2GdUbH
zQ8Ot-szk2Z8;OMWB`N_`{b+sendlvx&sQ>mCyE%m;Jd9>)yeY8p`rD1wc|b-F;@Ys
z@DSuWLt}wM#PkU)_W)*HWyUu())Lc^v08`mR>m6TgRxqpb23cpbM7gbsk%TQ&}z}!
z`qdl{5CdotN+XYp;PK-+5ov!FwvD7L251fOWTtylJ25m>t$j;S0_sWd)@oOPhUUAe
ztY_e*5HTwF4wWoc0T2R6B(XfU5fEks+=+yeaq%#+fD}sWU`*Y!P($JUN)}=Zt<}`n
zKC1(;Qi=y=YQ6JoWI%_*b_N*4w}ZCr|92sdy&oH3x}^c8a@hdeY;1ozQ~L;NcL4TO
zU%(VI;LEN|2*BE|rnKN=bHoBb|8ezAdntHIU~~lzVFqyt9R!PY+(0F>NaG4g_Lw<5
zkq2o%k?+Kc#F7C+ySvcZGX151?bR&aZbDdpDdrAO!h5+L57(%@iB&m~w-xy;J~mtc
zaT6Q!(#l8G-hO!h)7yVPf4Ba5_2rwl%aq-s9ZXdc)8xmc7crql5>u<$wwnlPc!3oE
zz+~*k_CDQ7IJEca?!wLOy<E2Uv@fmiaQp5?&Yqerf4*)}=Ia(^zHU+G>lS6cZc*mz
z79}qbwuf@oQK9c*O;76!CnrGe6RCJ%^{LONd4jhKv_7pQBG!L}2PqH$ov5|i%w<>&
z`$X3<Wg%46s#cmIiEAw^;4G{}-|EO<t6c-M24JydqXHh%iE9!dj9y3qa;$+IvgfeG
zWsQLAo|z!)fr)%26FK2nw2{dMY@<dp>m|;kkEX?*39=pEeoGP@fROHJ3n?p{R9&Qf
z)5*IcPMdsd+}nQ?k6_3>K>8<iLeR<5l-Sa02i4`eL^}W&qt=P81Pq6ieY3|1s!1m*
zhlV19rj9SgJBr3d8n87Jy=8Qs?H$nU$CcHPeXA$#YG60u>XtbOv0zT<T|yM%?M5j&
zV68-H<brmqQN%jbC!&OsbD$X@&d#aSMcQejATaKf7H>@cmo1LiWxZ*ViDUC1py5r6
zpc&wM>|+V@;PlmQJA)=hFOMX{E)iMd0b$=bU``k0eivOI44cq3PSY0>@oNmtYUSs-
z74bhV6IElEu^c}X0W_CEf*=%^$&x@Sf6Z>&I1s-3Df9wi$(i9FP#CD4IKbYv$ZfHw
z#kQv&w(oz~Aw`NJC0Vkqq&>)TXlnRz_|2~(yYHRsUJs|Y!>0>UXNA+sLihHqQ~ZPE
zPbGQK=*}u7465$!NB7rfDZ%h{;)|q2<jeAR$9H{vLgjdd;u(rZl{!9IQapk@e}X&*
zG~Xp(z~5~eyQJq0Kkp<_kifDPPgFdvPRGCB{y4}4SPE~#*Klk9dU-qicWB{jZv}$x
z{`qkTKM*Q`4mt<I`0nSw-M7P^t<bS<oMPQbW3;Yxqa-VgI%B-R@KdiFI!mb)-+6Du
z*fr~kc(1%|-WcDp--{p?GkpB~f8I4CUth~MGz#>tT?u5Rm0`9~!b{uE9&~=beSJ&9
zSk+48vz=YvI^kK_tzP*2b<hI?(dVQS+mFf9NMEKOmzU#{k%2jJME4S8MlS<Q9X@12
zUyU&OPZuw*o||^8RWso837@WZwRrl#)M8_4q=cP-AEU>eP-`oU#tG`kf3|IiEuX70
zt%OH-;ZP{+TZv1&a8?jb_5LNDyY01Cop#aU{U`yQAZItC1OmYhtDPM@RMd51XTTB|
z1DVyT{1(wu@dR<c?O7k#Pwtp$9h_&Jo`@gsswhLA_aRRvD}^Jl_fnonI6n~*VDHH9
zM*1E6t|jcj?^@y-zndL=e;B_vp&a;K&wiBOFphfX$Zs+uhyIe?V($>c-Wz_mEJGv6
zx<QLgbO_O-V%{y$EXAjKg{5ekPqo3b8s&MTFBv0WNxI;LlRfB!vl<!WOo{+4wDLoM
zF_|(P$I{g#M?3flMM4)3?ebo4UT6|?B?=>9Rq|Xaww9x6Q!!c(e=arWqyt|whg{-u
z$Wx!w`P`UUBBw^mK$}@A$9K4%T5i!_Oh2!uF*?u$pz`Tit$Cy8Y{VP=ycbNq+axEG
zOs*lgi0CHH>5zk6oN7ByT+@iGNnA2?c9FAB0dSH_!(rD>b@mA^VLZKSnr=+wLbH;3
zF7%qD%6($pJn>M-f9l;;i(w&Tc`oamo2>83=~501c66HN_8<jbc<VOe0we)A(IljD
zA+Rtq?b73cj4cPKJMgi=N>p3%IW72_;G5ypR?PX?_<XV${SpJ|Bycta&yGN{gK?!2
zIfp<2!8k$&aw=kjM@*H~BrNZiYO=QwyvfH>%~r_1sb(wvfA&<fk+4Im*+^VTHQOD0
zRI0fRWllAl*}q+?N!Ifb`*%z=n-x}4%@)sxrJ7`%Q0hZ7O=4SATJ4!?X350S*bmvu
zP-WYc#psE(-;>Z1awI@p!N1l*B!vRiN3;-|l~!7a&Fmdoh>gUR7GkTH->rp65(Mf;
zTZoO6cW)uKf2i+>d$oi;;$AIrE$%ft_^`OQ3FQ#?>e;`(h1e1gIsHd(_bc$S2@c?{
zSm7n7r(5G&u?a{vL2_4X=CX&`<?d`wXxBNRy$z-8v^jw!xzZY_klvdy6`znoZvxW^
zhBGL6dN~W$X#f$IxzYncE{nm@_Cl@??7=j`Dd!`ge@j{|x`E}@Br(C&*bT%3vwgjR
z*;S9r8#~&cPb5#QjYulQeR6jz==nAgnD7d-!TjYC-Z&~7(A%TZ;l*QbWBt;u82RST
z5TMYUy@6q>c{Ox~l;!})!yvq(mCjIezxEOmjL^&pupBAPE|my+fuLH6OCFhcw^mQq
z-@k(pe_CpRbQ^Iw&i-IdEj4BOdLg*hQYQ@VYPN2M&*Nw2xUtr+%mhIzK;qaZa2S3<
zLz(n4Is4Qyhz@b&-3MDM?Y;DL+@rs0!=Ttx*a?a`4duW^FC#y+U}Pc1VE@xd%^smg
zJfJiw-e9^Q#g)E=lQls}1D7zOH^iy>+2q1Ue`^k9fEgsl-6}jmX~2|cQi7EJDyG%{
zcSA2B`nURtL3%txDnWfX$FVOAPdpDR*U>uh1-R6p5}tyl&e?kn8ZO=Llvc!j(WpEe
z7?fw~DzO+$YBAPiJRAkar9=+hG{anTpAS`}g(lnjQ1XH|wW+%<$K`J|veI2oepNXV
ze<eYBPJO&=5WH1*poY{&Ao^yFk>MAE*p{95KCcq@SWM2IFiJ2(>eU99bxp6T*>KA2
z;08r6Dj_SimZ;~Gg?j=}BYCNW_EMmwcJ?Dhh~!cUZ7Xr92np8dVY2=^wD#FZjZ<53
zvGIB|b9$LZ=7fbkL1Gi=9!oWHg<oK`e;_l_LMSET3Q(+r4@|0UhK4eDVr-?3GKd~G
z*V^SH;;u%jrd6^7%^U-qvZX({`Lwt6Jm>dR@|3G33}bS<3J-J9>P*zKgoJzUm#?qO
z6+(94!DDM*y{3cGfE!gdRUvhy=5U^CYj<;jlhcLp`>4c<=|D;_z#p}%5{^>tF-D*-
zn~EaKC#cM?s);u3v)$t{`}k^`^BEqe4DrAEs%<WpQFRT~)HNLQ{|8?6joO#796uBS
zG&h$~2?7%VIG3TS1SgkJX9Wb8YnnisfAm}}umAe|`;#mLOA%~*8~@Fp-anuI_teM-
z6vl>&HBKn0F8}`HDSn}q5!QtZfCx%o{`}A7_osi%tJb}y|4CS0k+7t-#tgy&D~$0M
zYf+$cRalWhqomXe{SWj%ekLIp5{M=yb!0yD_vvNI=Q!p)GrvjRO~qBtAqW#ff1`y2
zuCI}Xv}WOCJsX4YXjiiz!^th00TdaVns?P@bwj2EDw<MS)5lNnQoh6dN;@u@GVSUJ
z19f&DU*rkKdmw>odU-T>`7?@l_4Ma&H{}Gu39HvquI9bO1CT#Q8ni2vePokQl)g7@
zRGrq<>BI;}hi_WltXA8VDtb$%f4W}9#Z{U_ZPer|r1!qa)Kw{~GQL<&mFg`zwq>O)
zkaiz$Qbfg-6Z6`5prLZvl}$S(D(X=;*+gmhZdJ*COp(D-2!10uSYPdro|q**300a<
zp4uYoN_3fT+&n+oYo1@E-{viu=|oeJc66*-I)0|(%9gK68hK9A_$c(efAJB`0`i{M
zvTrClF00h<6CYS4YYR=fpE&b_ow_`CDQc<b(&x|GR|zmXTi7Yh^dvd$Ta-5fz^=4+
z-}xX?L~!#WI#jhcabdO2SBo(|$+YaslqAI#A2!xet`OH0ZmX!F9l^qQ5s9{4kzn;{
z)2eF5m3irt=GEs%qp9YVf7RH&v5e9WPUZOngw;Ja_fln!cTuV<rB!;@zQ#bW<i4cM
z#O`@cXlzUGCZKTGjS#z`*ZmR~%d65+-Rb-$e~T)g(k2EUL6qh`H;G|CVpAjn+I}YE
zzOw+t45FD=sh*`!#n?(B*Ex11Bkr`vT|@kYcbaXU&&`%SIbjTCfB(7*Op(nsnoLs1
zu7@E+6397B^7v*PWTX}z*1ELWUuWvg<n@b;@~2G$^-4JB+iI#>9L+U;!ckRXX#_HP
zGh9$+ElJP&wrPYRJUmlyuWf=Lsm`l3{yPsUehQp=*P{`SRngJpkr6_0mtW@3GERG}
za1yg^`SCDpi8@dgf3f0s#+FAsN<Co9Thfgwkpfsrw|bw>{a`ntq4DbO+S4=9-)HU1
zCHq68?MvpYME;cxTgpoSIfiBak<NhQ`GPGH_^3(gDSJpGlW(XtA4k7&|KiJOjC-dw
z3);-S#j^~hjP;!wHf9<9VYMfn)8=ePG!;Er3mQw{Sqd8@e`H(Q{AS0VnhXsHqt)nD
z3XNt6aouJpsog0K7@%Z0_`y4JSik{Ikwmmyz`JvZwxc9FP?n#G=9#Yfd{jNdGTK`W
z@y!?elA{e&=Op4+2*2UHtR+RciSV}%TaC7|z!NcY_UXH340l#N$iWClkantiD2;-W
zvHQv%Qb-w%e=h!^cgX^WCqGMRB*?we2#JbF8$v+mI&nN2-i<;FnXvSFYqdGeme<jd
z5bwqrN=AdQGVIONz+jGSu4HpH;{{nIJ5ib;fsBvJg2m!8*15f)req|jAy;2m{^i(e
zU)%{TVGL+Hg`VMc^mN~fU?_s%A?**?NfAy`r{tbuf4lD>e#|-dcaccv*?vVA`@1J!
zvqJy|5I(pwD*6JONP$0~|BBka$fk>Vco=bA+4MMWX9uaHGpc`*j~br*?6ox&yzn09
zT|rKT=3}pS%jE&ZDpK8wFP0aWtysHXTaGnB*^X-|Z~a<n7->n+z-#*3CBO=MWXYzn
z6$HD5f9!{CPZ0;zyyH2qyB}8(h0=Cj`4sqMbqHLQIP;SZw%eF>+^28gWlUqYfks_p
zOl{R2^^!in)-TH56~xH`Jr^d$l@DmUwiJEhehYe2eks5^Uf-xmzo~v<VRGqk{Lm*g
zKs?B$fov8C)OK1B%kVUyS+ZwRzwTS6reH-}e@!cFd%sw4RW}38zMaOZAVGv_9|gPA
zf%F2^invV|JP_AJwoeqn7KeB*QC6kAbeeTS&AVL+@|M!A1XM(u4e71yFdXE_w`6qy
zElZE5%ER@eRm+!oOf`Qtj;Xwu+p~2hnv~ui*Y;ZuIn>srTGGU!+&3KU6aiI#Y@qwA
ze*`TQp}hO{CP5SMuq&M`Ty8l@YM)rWplM?B$a*7a72e}Z`tS;`nspeuo_CKirB{LQ
z^0caX&)pF#_q#w8o-CyiGflPbk;}kCaPTigC-kbH=`L%skMa~5B4i0g4;HUM^;GX>
z#$BaHw&ZN@FuPYJ!^_&xQM(#lmZ|A-e?8S014F>c8@^HQ20?T0Z@doLq^0VZ*^R3c
zFqKto-lnF=KF|Q9w2m;>M0Tw19drUg2Lcjqo|&ukc$W26WKzSE#{Mv-7n}BR^B%)U
z&e*J|#(m`?ofZhISy}4teyL+1V*#@zb&Qff?!#&Y1Q7Lx^z*EQux}eFfk2S@e+!wz
z;W1D~1n5xtM9mu|g>IZ@-bvfDO<t9<5_$(Oz4wtSjt9^fq)^~z3Zj-tVA~yi2LCH0
zt&OXpZP!67r>FSAa-(1=T`k+5Xnr4~$nPPQa?+9vYc0L87My4GTsu|>_cF?ZR0=~u
zP&Z}Fo2Rz4B^5&qYuQw}n{_Pte-`O!uLa?QnTNU05U+Y^v(ddPpKnXFUC1~oR72s|
zS-m0DO&}qW(%xR4S@n`xUwCK;IBujpTUo}8<Y8h~A&${Ay}L%bU1L6yA803xSxIJe
z@3T>|mIUXXMX*~&*(2H`k2HD;$7bf?ix;bIALDo@$<Mo;1Y^ouVSR57e}`Ao+SROW
zn)32ei>OVBN_RnHhm{*N!aGfHmvZrXFs{;U-X4pot)|p=7|bsKp32W3X#Y+qV`g$@
zg@$ht9C#@tMdNYvRLk)kz`MD@xzWrUQD13!vWDIPUW+Y#I}`Gy%Jn|7!q`Qg_Nxu0
z*<z}B#m=+Ev-UNN9t;c;e{$X`nU!yL%3g)9huqQxEhMTHrGJg>q134kJngPz$5cx0
z8;x=TlwHw(P*EZGWP_P!-F1Ns1`OjlLOjR_v<TYWNwJnnXjt*dYAX-HJ2D7uj*KDJ
z5REQTuN+Hs8y(79#U5fs`+1iKUJuch4(ka%YAQEZ^bI?zx_kN~e{2@pg$BNnujW?J
z2`z74dw~y@rr2NN1tie1f_bM0@&eKdgZ_}0jn08-ZK89$P+S4R%-_h1B~%&iWu@Jl
zIhju>0%AUVBOB2x24#-;z#mK%P_i3!9#A?;T_WE2K0XN103PZCEo*))llNy?eTu^M
z7LYyZ;&RP(qM($ve;(%MTm;wdZa%+ehbE>_VRf92a1#I}UZ-*SFnYvIs&{f~g37A}
zyY0g9moMX$=!ZE2j4@QNc=(GD6;N=E7n#Q0`|pHC5Wo4Gq3Im(`>yTAZj+>{X%fE`
zEBAjHv%btlGI^&Fxg_t$?*BvTI4a%bnRW7@U!NVNrq!yIX{3|m=3g0=)as_)wlad8
zZ*23q8?E4@<lOtz_+p)5bL}md;eLQURSI^b33EG}&J2{F-3)gG9D7m|K7NgT?DLQH
zmv)U(F(hYrb7?T(H{NI(iuya3RY{~j#>+3||NH0D{{i%d(j=F$96uBRG?%gV0~42l
zr9fGKt=3#}IPaNnoXGCaPIf=-j&Hk<XHuONPJ_kn?Mnyz2gyG{@}kk51rUayd;8k`
zx|b3SKSzF)ltlbk{@w9gA3hK`-T}M=copFA!7AVt;uYdOsQE4VLH^yQwR1Xe_`H)!
z1qw_Xcq91mKK%Cf^G+s)rSK;F8eZ+E*SFn&U%SFSlF-I?TA|R?{r+_q))i=BtnYxV
zP)>E<{^-8!erc(eyNu;EpbmtT-M6rzTKSw5=ztX{jnTSNAR3{y>x}V&!k>}?Suoav
z)T{)>N>D7kI#)s#3L1rEE*lE+7<ZRliY2wWGujDa0rEZPWT1<?cp=><P_sB%_CKP3
zfYr+9ffv$7AX9H{k!&_2aX%#!Km1t(zg7Qt<)-D#5-7F`*+EWzPLFZhP#m~*J!hJ^
z)qIDh#I+4Og41etT4j3PqbxxNUCx_+e>`uNr6MDgbz`YY5S)@^Du|GOM9Hd^kz^cQ
zLVIeda<EPEV3csmw!1^XY1L~X%NGKFIoOZ2oiy+Om&kGESMrF+5v4EmgfBtg$ttSX
zbkw-+c~!@>4qNf;oIo^dx%S8z<bvsQV$$@_Qy3+7YU}JuMkyo_g%_Njh5*QF)qZe?
zdjroq5i(Ze^Bi~Jq}k5}94-bN+!%1Eg$BQMz|~Z*1zf$GK8>`Dy)@f6EmvuO^0<;a
z47FUD3+^Cs1u0zzyS$9T?v#APZOOSK;VrD?JO}3j=QQXxay-TPgXN_0tXs-S;v#ss
zPO0|3asrAQn=NZ(P4HT%ZYzu=!H2e#rlGi5#aDiyk>!r=mECVyQqA?kDB3tFHB_%8
zrDiux%G<a0LE8A7hLnI;G{9DW3u=7OfUTld`QuY9=QwF{P21RVHPvgDt9R36{PjEq
zX~_@P43b8@=%kyH#2+_RqnXXp;iDF_QdrbmTdbyf&0_WLIg7n>tTl>yJon#dIT>Jd
zVlhMPYK58pc=mD<UuSD;FRUpAOpTh@z7kb8&Z6~fx0>1`%55l>=!p7%FbN%FYFx6O
zWO|o|aoWq$Y(Xn-x8;Cuvv*IAGuSZXc$qnLCJZw?7s@Eh<X=7)ZCwtB@^l4AfGoqI
z2zxMjfKEAi2J@%!lJX=Z770R}mbUZQ$86Tm6l5vZPCW-eKUT{6RDGO``@x(5!`1*Y
zl^epYBuq0kIEjMH&CAe#D`B}etOa^Ghq)HmAeudZ4&$wHT;(is>sU*%lR-PxLKGgZ
z4Yiv0iUF1~D99E^j)R`8;;OKW%W+H%I*l0O=j{Hs>6|~m_Bc8eEW-5m{X0w}jl3Is
zG?D>RgikYY-w4ZwZ%$y6e=Ko!$MAVdo;V{2^1dedlYB*vDYMOgPHD40u&s=(z`aaf
zmTSHo)Sf!?G_X}FyXeGRGVO^1x)h5v)UT4af-zqT#gqq>13KpDJk8x7clz?Fsyr#=
zrn3;C=S*uG;W^Hh;*^NJ8k;kDJ`)6=@c<w#oMjJ?$Z?)M;3#Jjw9-O3+w6}1qM@i8
zQ}leQOX`4Hi$WKF%W7q$p(JC>e5rW%H}<AlR4N)q+T4FPoCB=kH)yYa2%EAy<G)F(
z<c#o+R%*50&Dr2!?fzQJTXaxty{rcb0thbeTD~V2sT@N3xLh@i7Sgqs<oM}3rY5zV
z+8DB1nVz_tV6j_nqUo=Qt&(e$>Yr2)3vG4@f168K5T!SNZ9QVPG07e>KFr<pq7guI
zHwVp)66Bz4kc)}~!!@59sG0zo<y*P=Mgz^E+f5(f4QC37ZTs<15^$`YE_vdRq8EKW
zw5+;iX=9fsE?LI=yXB5;?o7tRd)!`0FPytDKitMt_Hw*p^;EFK5FdTrNSynM%U0Oe
z^h=whW>HmtI#rso*8{12UqVnvRnagTNqu=f<Fj3P#mN28eGhv^_<4!HrW+4~Vzi|t
z#CiHH-qdC}QyEUq_CkxP=*)T^CtEK~drF>{@8URBA-J2nmU9YOEr9XO<y~2S>&g&u
z<EiBZohx3ZByn!J;JLK!sUaU(a<N;iyer!+S~CBCGk}p_sLCJn*9mUaRoeRd>l_iB
z_esGR$LHBXs&VD^#Is)49*yRLr|fA5i2pL7z_0D0G4m`m(F4$!TU(*D{{WIeZNCwr
z@$A&%kQ<LQxR?zO;39yok?~n*8N&#V@H^pUG)6e-zYAWzIzD;|e&!pwLG-Pp3-~$M
z+cNBGPF%&$e>ceY5PnADf--z<^dAGzjr8jfbaUti2pzt(2IW?Qy?XaQ2|_oVKN&)U
z2f>+brB>_R&xOzzcRd3_H<!FKgl@QbCkPGPgc^r;V(9wxS`giwx&uWw6js9M=F|^?
zqgzRy1ElYk{AozK^?~O?Y0w}ny8of{Lr}WW;z}spWSm+FrQ5sKLFvo8?t{{;zpR7O
z&8Z(6N&|8JvGi|&(tyC-yuJZSr++grK@lVTU{ByGcUxn&T)XoBmT-dTy}r$+70o4%
z^Ki^?14&p1-PtOVlY1rVTIAlKX-4aR15D{V-j}f)KNJHsFfo?_Z!8m+TCG4Tm-!GU
z3x6q63ovNF>j1eW@GZzG$mBGK%>N%kt))^e?ryh}L%X|F5=n8>BV|4PJ?ZJ&-Q(Nc
zmsdlRQ=Y-$^!9xMd7<SmXc@FTIRIr1Om9D?-@j^&Rz8pNA+3#(kCWeCzWH=-z{?Df
z86cwor+Y^rBSuDy%t4lKEgzKMu8h5w^?z2@drcH62yKvw5aRFCA8)_h=_0UJ!N#BQ
zYkYcnyZh&^iH}CH4U=W3K*RLskGuGSfR%M&LR0~0`uW%N{qEP1YUMS2CuMmMWoc`f
z7s^7bnEhm}QO3EdthiB<P1Cb{WRoO{VA(i5$#=+OQK2mJ7|J^;?;*cuS<f%0q<=?w
zZFSbiro_9ptja*AtbA8nQnZ<qU-Vc66l<`ljH^Yth@C`f+h}j&8|>#M@pYl}nx840
z5+{DDJkw?q-?}zEQQ5ki*QV!(#sL$x{^?M~je5w^Cu;qys(jmKCZ6>vdTE{8B&yFo
z?-(F^c9O=;x|A5m))L0Xi@6M2Z-3?62949tXu!=kzkRE&wr@)vBtEMtxq4CYn7BdN
zdLKGpUW48yh|23Qu_vWXbB?&Bq(CYZpNemCQA65#;;<`fNg-^lLeqjm7ox@fF-W4P
z#Akf!0-<iD2$8o`biJ@i3XWZ~yt{|37@SGAFf3c$jxZxu(H@sueR#Fi*?)~Y0aNn-
zIl`KET$t0yvm$H9oF=Rt-#n$UVa3ib9Bm^PSe-J|Rm$E<ctf@$5vu)3akChac&TzD
z1r>}P<%SrI&x?R)D=bRXg>eNI+4~(=Ka9^NEHh5p6X#Q|g<{EMB~Y4fNen5rVkBiD
z^tz?$*_d-^(q<fgxmXk6B!Aljj9lP(rHPnzWVwzskTccA$Zsuuto_7W1voO<R?d)*
zwjB>6;2^tFPF2A5qM`&!SV`)aX(xe?aGs<DiqMS|i_@-*4Mq;%Iw_sBQ7I3L{Bn_f
zEh$fKAbe}{IunnkfyoEV3&AVz%|LJn_?7N7DyJI=O`&BLP~~dnEq^J0miCrhZlS(G
zBV)o$!m-X`CYcu_M>3bhPXUx8Y!2Jf*%2T=6NnO9!5Y*9Q5iZoUWi?Va9LMngootY
z!^JY=eGF50&(O23HR0sk*W>jycVyUCK(4?V&`D0juQMEMu*X*}iEn=(j)zT`nKGg|
z-TPcr`T4-;Xl3Yngn!D&2t)xMZO{s{r!9ze3G6)tN{m5rr5m}YgZ60|`_)%SaOr$3
zH$K&_wQO;p3b#q30^J30dj!Fx<(v%sKcQd@g^bT1b1t&qwjze0NV4a0Q=f%*jd-l_
zVZ^)4$_2PN;5ANQ;GFkTA?rJ++O$+b?>;1sv45rJuYtUS<bP*C-eJX@11^`r#hL_H
zGQc)}7nGAMu)%*Up|Lg)URsEa=)lNX&b}3x440z;5+^q!p<9CiNqI&J?19Sye8!HK
zI}Zrd&9mSroR89r-A!~Z`A1)2@&eRH3G}wfQ7hn;yshech;m#mgK<hg)<_l1whRTZ
z53}3eKMtA<NgH5U{uIz;B-sXbJ!ldW93&tF96(bZIRZ`H8kZrmKqv=?)1Wv0*wL3z
zvOpYvj~6%0UbY%N03;OaCW4hWJapdXyWQDci1wb}kwZ0ZbXlw1Vmn@{47C5jcG!4D
zyi!~_p~hgtz!LJXNYbvguDj4kpsmCCt%=i6duy^&Yg(^&&^6QhpeYUM609}A%3LVc
zaEnHLR~y#=?<kcTC%57K+0{qeIn2rA)(v=nsCA3v(StOIPEvlDk58ZLt+ccL(!n2w
zaihjTu;p9@b(a!|rc0lRX5X~wlV+N^W!3V=^+33d_FiKKuiEe;%(JOU)>$sc#`}`g
ztBvwhqpUU?eP{m##qJQvexP$LY66s2DbL(-@W4Ku&-%dI<q#GiVrgZ|n>3@(%q_7u
z_l~0k`t|sJr23$sPr`6Lggf?h@0Gx$k;|A<NQWV(5a@SuN?vnr2FdhjyFFg{+k9vi
z-fq5^p|e05f8{8zeIVcFoV&x0&w39x^Y`XAMrtm%-@$=S>f+ByE403au2ttdevTSa
z-8gVwWFkMGcZ*9^e*#;2W3?gBPJv>lvM%&z%+6KuVdg5JYd(25GG%urmAq!>ymnQa
zLdrTLMa9si07(fHyH7`LiZj|0D7a5|_Xp2&>#HUOe}r}qcV#mVYmoDNfN*aptoQHb
zcXwlzE6L3Q9W^S@uKypB*<H%Y*Ugm>Kmqw(7k}>V4Dpzfm-^F(+XTx#7aNAF?gbOu
z`9040&(uP!Yxs>y+GEI)-5u_7?9bq*^;W7xDTbbg?rP3si?*U)3ufH%Xv&~yKAuLw
zi$BMfe~SS#*We=X%l1rUW#-Q5vFEQ-b!)_}lE8Musk=%kG*MR?SI5;+In#B1>wnIq
z@8`uo0Bsr{NRhEUc-?aoZ6sm44>$XF!TzZ=2mGa1{JTO2R1E&3GVjBw=GvQ0v2B^8
zTxAZoxHPGmrFCG^EkKqj3_5)0R%mn&s})Oy59GAk@{RL4<#NRR3x+TnmqEAz6aq6f
zml0qq6PIwcK*fJN|LEb**Wd1BW>^Yu!fm*$pI%<?{=J*oS6Yzqla?S*)AZNZyYK>m
zgt0tqogg=T`+NFw_xqY^4UPi_M;b%afMcW}uP0-?(BV}AXBO@$h@S)`78Lmz&*?fp
zhj%8=<8@T#vhYY@{zJeHUBZuadPS6$#c`R(Y#NW`Pf34AVNJNwOT=$x+*)A(CPPkW
zmbNg?6y&)-6D<R1XZ?pgAniF?p^ZT6-VKUMJ7KmQ#0W?c9__+RxQ|OunOqoq+%j0S
zrbrO@!XOhI&IDvDnkDghYm+foSsyP6!97A!3nNwSMO4X0W}gbOF<Y9*D`8OYF|Ldt
zHf(b2wGMylneft|NnThp%ZSg(6pLnpeJS>2BO~}S@m&CCBjq<LNF>YMTHi$>WEKN2
z7F2eiOi!Lx*gPe)^=99Mm-6{uC1=T`V&GVyEX+Sv#uh9^B~L4^6Nx|O4OCX~Q+B;z
zgtFdnyPJF(gRD82qB{uk=>Vvh`L&r>zl4WEF-3oo1z75UD4+WL^3^$CyYQ^0Q+7?A
zp6h@RO*!nExxwtK(VhL`!UHCI?i?5o`C!DTMByzQ0~XEoz+z>M#eyUMSXs8+0eCUL
zQ+ev`s~$65*Lp-4)5T!3=M+rBx{xP>;bnYE?pou79)$#TONAysO$xmLPEWBh+;!PV
zB(Hxqtv$P%xok|UA7s&KAqj5^RioTlzL9?0$kq>QF|@V5@#%~(5xawBh1bOj+kiFL
z2=>#@_jw8Smsnl`?UE1owS^5!qhfIe1n7f#W#bKluv}_`Xk6w(2tVYNK~00$!h{9a
zG)019+))bGU!j~(<PS!pF5j0U)J1Ft@2`KtVi-rj3Vl=8h}@>^JrGD848e0Kh$Y(W
zWu;1sQ=gZ5vWwxa=S7mEs3DEe2sh7|bGe!1$+Cp6ueiil1+PoI#}5fus?_*i^&-oG
zn&~ETK*d02Gy;wlgMQKZ0AcM(Xs2#plN3{!;kZ<jl-PTjg-xg*C(@%?G;w(-ksf~q
zBa+-&+<VMzDGLtdvXR0a3RE|ZHmxbnAJ}`=*e2TAH`)^HSPHJ9#L*O7FZpnPWS81g
z4(`${Ik-!uo};_eORg&?v{8bjc~5Fh+M<DtA20TI6Qm{(g!X@2Xz#{{Ig!6DBlY6L
zUe?tp1v&Rag@{9epl1B2^VCbObRK_j!q^|a6#uSXJ1q0HWPD*cLMS$IE@zF&-@8Y|
zK_gN-9@b=>;j8gdP2a-A?wilEvbG2A`NUe;CW26ruGnnF)2x0k`xKRMk|`nyvwVsH
z9LJ1IHs=A)xk;dcjmn#TGq=aYOI+!l6c3CeY#T`T3#_Hw>&e@1ifEQIBHDk*`*;q`
z3$XDZ{Q)TSFzSOtfSw6m?p-*^9N@Zj0Xk$tz2Cc7U-x<$*$=>TXqwAw>b4}x7iu|(
zZ@!7sXG-~MeUnal3fcBBa_|R7F))@hb)Qj2o;Lr9uGoxxu97ygI`x_mu$b`N&+^2@
znD*-xOU%5U1>*&9U|912{r!Kf4)J6odREW|$ozaM2@kR>SY&HozPQYr0d5H$0eH3b
z4FF#8Z~@?U_9cLI4qUyiSaC0-g{ylv*Gr;_z1OZO?V*|BZx_4^%+$_WvoHYDFo<Y-
z`>(42Rceu!?crLnM&fL?w6w0wU{^`$Icr;7?+DLhpuOXndc7h9xpIGyFc*xJn%||h
z^+q0JPIs{3v9Ae*UU;qm-pb!gx2XvT!YjIkT@|-$DGr@9HJVqmuRBNCek6q`DbF#^
zLEBvo#T)Z11LM436r~yYJQrcZe-!{7QgqXCk0DvldG%05?;sXjD{3wKyParIfo<2E
zYFD)i<@{VavER#lwG~{f^ffm_^96y*u4kuUk@%k*wkm%V^m>U;V^pLUEccIsO3Te9
zRNC1+$v5a!$n{XkA(x|?^qY(ojh<vO|0qv0vgZpq7djSm`IjOrm5JT^SD!&)%Z2lQ
zcmHUYfi^G`1Tix;GM8ZvGZdGfO&}<jS-e1@0os@Qyg*iefZgS*l@v}ozG~5gPJCM}
zA+GW)W47orK1tug@~!Gv=4X(88ZDN&k*?45p3~pUcU5W76MTuIb<HXHs%(X)OMl1J
z3x_IivUB@w<>y;|=Bn(|@~QPxp^|r}doJrPUE$!f4&D~p(Jq6pT20=giyb0O3^8gV
z4GfuvD!!V3Iq9UW6kCh77dgs^v8#qD;+9wuL<syv3NY4N`m{mn%O4a$^C!Q4-}d+-
zJEBk+s_aE)KHJ9DT@R6(^N1x%Ufb%s#doqjVoq2z9rH=!g;+R*>;ZYdN~Qc!^n`C$
zRotNrWdkl#r|BTEjimL|=@Q8?F7~1hG!4qY_WS&QW{ai4U6qMx&b(%q@r~(%jrxlp
z-`TfsKOX)!_>`fHWJtUJ{{107KxKq=;i9z&3NL^B=knXbKRDHFpy5+U89gNaqQ4e*
z=|+Cd=S7gTZ)vSDy{C{Q#$Qa(!ejOFMSErv$tfSV9&<*3%FQBiah%)2p{li`o;`e&
zn>0v&$(a$l&tHZlrJkz4NP%%1<=3*iZlu|rc1mbD=u%N`QGc68nf4%pQ^PjOY5RJ+
z48aOVF3x38Q~tzA1RrMnmpqpvIcCRr;xFO(ZMD$}%WwG|AVmVe;CH@e%HGFjp|E*!
zmnbuq!pgT*MvCRFDr3#f8&yWl=iS^$!6<lt(W|yFpTv2VZcnf!MR8utVVkwb(%euP
z<o(w(Y+1h~UXou`>9lHWku>9mI4x;<dTJW*!l7vcGgT4UbF}evrjqy4j|T(Edxp`y
zGY*3AWL?gwJ6*Pq77g3lM_qNaC_>+&<UHRbRiWlZLdq^(osc;W+GUw=<K%KQ7zT}h
zH#NscX&eFyh`HrYukp6LLR~d}hjuO1LBy8t`6t$FwCAPoxW8tjSpvQPqMGfTnYoSg
znfso=8mjN&NgJl$GC5KyD5)Cj)!3GTvMGjgvZPjJd3J9_h!%v_{v#^FrP|x5nsT*=
zHqaq`Lj>Pkhm=Z>abk4{MTMiK3NEUDRuJ7tM==ORA;c)&=Wa7!OD=e~yUl#9t0Z~t
zMdz1xf$vAYWMY^*bz&&ru8&r#Nx!*wZrorkjA)U>(OTtax&zd79@xRg@^uyhCu=qY
z5HzD0_c?1gGysIs1<p{5dh;{v3Z*hHA1b@>P>@UL^{kCLK!>QdR9xO8I3P=Zyp@kO
zU?c;=8pT*eKhYtfZmWMRBhhW%TI-b=I?ssWWh-q~6+ni=VSlfUVm68da%<<*fnx4R
z%it-YQK!JvKpWElrHu8d3^%HQleoXsl6{EQ^FBk&$R=Evk7+*-HBn<lD*@gN$Q0Tq
zo<sHnz<3(e1HBoA4T*d_Dv?uv=y0zj)pPEG3g-6`pNzdl9feHA^ck_OSB}DmI!vmJ
zhTVNjD0mY)wv8rTMoc@h5mPqp)5-%6rksI;-6G;{7S~*=v8t=?v~u;7daTFP{=N~M
z&|pTZ?iW9ZBG#+jD&E@;*6g~*)}k4+-ILQuEBxR&``L-qC5#?z>WW8y*-qx9v)l;{
zQ6z&vsEl7$;C$Bf9t2Vl!Eut&I}Y}voDx9~vUISq8^D03P4A>FK>0oJfFI$3QxNwJ
zS^`wUYt6M34z&bEQN?80!VLfdf)W_kmoBYdoJ@|IUL3sSc8I?~u-!h7UN8Y7=XqjF
zDo|S~Jzu$`M$Gz6(qamKzSIuJQB_~mtaHkEwO4B=r`8iZub8IeG_Octdk)!OUfX$R
zUi;+PVXe=OBrv1Jjy%z+*+G)lmc(h-5x%NPR)aS=DsdaYdZyp>Lg;lHyj$aSjp!<c
z@{{PDB;GXdxoOf~?ubEcop9*+zD}en?dVvMz*qi6Km=@TPTuo>5-VmLy&e;PT|lJA
zz4p4%&GTP^KuWH?%6>BV0TPAO=99n=P~?MBe**ZyIDG*4!6<$r__6O=+bD<fv=z$*
z{}Pk~sL=<|qgnP*4vGjF%&>#^pd8vES=oIF%)xk{#5wq?>J9gT8q_!&=Q!Qy7hpN6
z0rm*h0DBJAARBsrXE@h0u!ahBU=8wq1g>E%InQ91uL3p<9n@}+=97^PZK-FU+_||I
z+2HhUM>ZIRKNYgUclUM329x9`BO8NNPazxKs!u>RxaXWhHu&D|K{m9B8Q;yI4IS`y
zXd@QkjOIyf0~`+0G!c}g*#S3nfZ-OPx%pNkHo0WwS<kV5h2kwh2Av7<%{$u}9P0#h
zpf+gGNsJdNei>Ymcj{A({Q+<V#4|m>9de`e0;z=Nwrt>vq*o4C<WbY@2&)w)?Q~x8
zTE3Z$9GmC$aK$tlzzgk$48A}a;BJ93z;P&J8JV5{Fo1I4nBT2)x>Eu=H~~7aV2kPu
zmYC4lg(bj$9&(yw%~b;}ECDwHsA!aMvC8Ugbt224rG7P@^!>^uqP%J&s@G@>w51U<
zXtdm2%C#P4-Kl?3dr2*FTT>oY<nbL_tbzNLXL2fP$~-XRnZPGV@%3SMt*DL)QgMDQ
zkW%laz1&?D9(7NeyhFLErW1q)PyL;6bnSgR-P|gF2Zb!+t0%bT7GnKgsdoTfOJk|f
z3rO_Z-kvKq!46u7o1yt}eF10N{;;lpt6Yxn_f7p-#kc!v*70rK2CT0Pc0tWdYpuV_
z=!Q4_L})Q~x2WF(Eh=ZJ2LCE(@jmQly$?6>wRU=npM!TYw0|1v_0ZsjI$RAr?O&X^
z7dKRYl<Ylz7H+7}2&KD#ct7s8LcLSnXQXV8)h3EIyRF+}Y{i|*jpdKR*&I-QMy@a|
z@BY65@bQ%)3?@+QZgv0vU;~nuZ#gk4??UnvRm8!34U)%ruNRD9r~@-9V6$yyMu~L3
z+uxvA5o3#fFr%@zetEXZ<615!6*kcqk3a=~O3v{;G2doV*KQd}jki&Qf^Jm{G?4eG
zmK3$`)=bN0A$dYG#rXndE2(clF%_V2<oW%=8v)B$3<7DjClWJ@F?I*9G_w33qLKap
zuS|1K#6r9r=QS7Rli*4~xBP=AQ)jNx@-EM#mEeuQf!cYjGIr-+fxATmv}BPcjcQhZ
z_}mQdhboOW@nGe9Tan6H1xo55`P$nQTbCUVK+JFo#MCm@4VJMz(Z}hC**WA3Z#!T&
zrLt~$5j#qCgbv3S^^UDOI5D^1o1l;mxDbtJq0?*2xp(tN(jz8Q_JX71AoiexmdAgs
zYG8<3R9_#x_YDk<4mM%va}cy4UqL&6X!j7K0A|V&N=cgB0-*ELlK?bTu4qS(-bbK;
zy6nJhogtlBn7a6WRBdi2b>a>7oJMZJo~ina|I^qqItP;k6lg44cqXDL6<EjDeNaFf
zCjz*MP)7kxnE)MMcyB~Yh^P;jqK)WV%>+nF^LV$`C<}bcbN4d=DlHUfu85N>nZT)#
zGr`fTP#c>;;B0REXQBYqP}sS(i;g!Mg8XzXf_T$PP4WK$34GL-VYdns12Ho-myr(v
z6ql06K*xU&DV_iM^~;Ma0{?XJuqv(6!q{-K4h6|nrHvd6NQo1@tNg$D%o3S6M$(|k
z-(i%5IGQp?(q~ka#yKveC$mQ)eRqgWB1=!vE3I>DH#Du!(GXjoq@6J4tdek%U}UJ1
z?lZM@ZD|ZbqrO*LaXfK#Pt{gtlSoER2ef672K|3hTT_~~uo<g_q^HYMkq3U1l|i&>
zH;h(0N84%1<|hn^jaZfrWz5zYX+fcVsuVH)->7yLh=uB`$okaIzQ3Z<Z+WEc&D8Tu
zD%N!+acxP0My;*5u1Sr&(6xC}FVl{O%sEpeDvC*5HHdE38op-^pg&aMdC5-{yI$-d
zqELV2tEwGWNv5oQ1?qri6*Qx5bBv!g8P{DXMdO^>Gu^t&h%8oMXU!!pHv2N6OtelZ
zkX~LJefo>M+j8``ZGWds5Kb$;a-}o(WNEJIj(O1aGo!p0G8j%#ULGfNPy!XK%G{lw
zsPnczs-Bws|NfcFE}|W~GR0(*M|I5v3cG*th$NEskOt9f0gf*(9j;{H4^t$DAgnac
zQ)Ja*S^ZLxy?0qE?XbqIHG<gkg_Kh+1+74PQiwLzTY9%as;sWnZ+<&~50n*%q51l%
z249w+L(M3=@q74)W(ccPkB{ovZ$7B{G^|k|dA@3Vvb_H&x-+}6yYe+&2`^`#Cu@Hs
z$$WLkMRm~#>q<i>r;j=>vAak`UZc6gI#irtd~|!6BkOU2=N7NbKH33II`wRg?-yGm
zk0#SJ5i0gZ*UTo^xOegGXZGpy>*dcy^e|{h1U&ugkC*rbf)Un*lTso8p8opx>Gzjk
zIn`nu{5|36v~QAZ+l&@yg&|WiMhSly@$~JA9JIJD+O?BlzzQAI1LfF^ED{8Q<s4h2
z_0Zg8{G{GS{8TuAt&!EwXrWoWZ*(qYbZRrpHym$ePFpCam#j&}b>thBg6ztt9uZ{T
z%a~kg&f%esTsrN4R!N#xX35{X<N2|!#{<45z}qISG7<&O_*qRAsWyD*3;lnJ<P68W
zZ6%iJS5|~wq{FP0LVHDD`>1;eb_=}RYq_A+p2Jj-rtK_hNh3-d`n<r`UHDjxF0k-{
zov3_@H8zwCyQd2Yw;}SalJEPI0)^1v`7GnQ)Nv#171Ka($wl|x<Jzt?`>@}0(`cLs
z6lj^7CZ`g#-FMS_ZW)QXGbewvC?JH#xWk;FjqakvzOr_dr}<ApG{XoXndUvUBUGum
zfbb)}2BgwvP&~7=S~jCqUZ&e+Fww>qN~!^u^QU~;(o-w^$1|O)Kgo}O<}u1PUaf@V
zDCw3}XmrAPcY`JgT?$2(#Ek<B8s@;cd_H6#({AkEsL9o%plCnraQJ`v1RcA;Y3C$=
zA68l-ZPW_sXQ2WFvQo~lloa+PRR)Bn6VD{Yj9+!pf3&tVR`9Q~`_cr$CXN&3VMxva
ziUr?0-2QH%E|AG8&BeM;^R%IqZFxp?0|FpWdV_jVTA>vh2`eQf(~fThMDsIj*6I?@
zuj#ukB$lUp;NWVycFunqhPgbGpDw_1*M>Z0+1B#a-Rt#Z6*KE2XU;VnE{S)l+8y!@
zgD2#<bw_Q9Nb~G#^-%=^D894h=WKXTqgyo!9Xy>~-wy47SYicI)<TTRh_Q@pBF6Xx
zH<Uc*O4V+6lhF!?o=cm&wO^@W91QC@UL=oF0r$62!o11Fij#j|IV#L_1sGZifM&5-
zQ$CqxOBM>}=NVOZo_8ZiH4ftI^lVf&uWWZ!;GLi`&ZQ=kGlWi{9vaQL@C<ztlgz6x
z(aP<D|FU@Qk7-8H4rsTK`A$t}a^pb(r~2J0h$f6DH9iSmYeG;O;C~lhGXW34Yc}Hn
z@S4l853g7F-vEEFho^6c(y^EY<_4{4Z3F?bM;pSp>SP3WA!|*T$T;;BXuU%F$=I54
zobT(HgWGjpmfTBpxFanMS;}<htGZr*sbKH;{-gB<?B8u0@-d!ssM%Ft<LmA#x2rk+
z3Z#y=24=x+jlxD?m4WcH<`gynG$~3rW7YsP>1i#j3;=(6erBEpN9##>?gn^v3_Pn4
z4uWSJgr}ImR`v++oSkSo#?gS%O~WIS?aTRTNLo?wMXrmAr1K8!An6DVHgR;mk={ei
zTQ|`6pk@_&CYWUZH<&rLf*8In1u>WHSYG$D16)S0<3bZ#s|+$#KNv4pLnD;v(Jb+@
z+rZ24U&MdQ8U-rPWxi7r$`O<$q{pZptanGJN5O1S0<4$Eu3hi;dcYMuthxaMg3P7`
ziACt^obr`@)>`N=JLN0TM1w$b<$TsNF9QR~x!_n^X3W$q%?zzXW2;f^*YNwTRReUJ
z8=I;VwZ)-Lh1+GIxV{Ii0c#&*SVQ4JKUS~YF8Y6zXWOAg*u4-WMz#&04g}P_!?aP9
z1eO^e(j4o<q~r*_W3wNL7L`=#s4xz+nz=V1l@VuW<nB^^DN35&S@uy{w=!g{b`_JY
zcg<li<bqwzbVG@_zoXsSF9=##>%X~Q<LtZ-a^5FBJ!bDx2LP8f?VWOtq2oMqrz;{V
z!(M;bO1IsUJKx<%S8A!dFFkj|?z|$%)is#oa!$RBy-nrsO>L2dStGT$)tq8S85!!V
zyX}E;y15XwxXjq@c;y&*D+kN&^qS3*cMA5qK4zTVZhMk#V%MqebGP{;>{rWF&0?IB
z{c6ScK8@UMzlq9VzgqS7tB>`0WZ<kY@ScAYJ#?1LpBkqvn}cneqnWVQp`sVridNfp
zwQ9MBvJ=?4pX*HS8lD5%N~yPsX9vL+K#WCrBiQ1zo)5Md$KIOevmSRV<*cixd9#aM
zb@{Z}>z}y>nJDiJ6we>3oqlHz!6hJ2VH_9Y09^83P}=fMPc}@RudYi=W^!hieS?2=
znsS=J2qnJ+3%>+Iw7Aq>c4+-|1@Tl6U96z1NbpWNXs7t2^Dgf^NoEqP)c1zec{y{}
zehj5`itF-;hYKl71!WjR_FIaFwS1#rvy`D>ce`8nSo!AMEMTS2=1TpXybK{au8ar%
zc%U8?O5&9Zw<-W@KU(!x4_$n`zlnbZ>`}JVETZLeegdEf!7Vr*)<!vTt?#lPj)!Um
zB?{#&lQH4s)VfX4GWYblF2R=UNY>7I)uL34P|Clts1xR28s$FWI;Y~KGG`@Pfm}rz
z@t#WI@T0IHq#gPJRNwsCi`U&p=khH7(bahA;n#jVRq)Q-Zkr`4&Kt(dlA#odl}S+~
z3u>8U*7wh^m%jmd^V6x9u^c}Xmw)O#6_*sFK`56i(m<VmGCyhm+xt&XRs?SEE~WD(
zMcRc(hGh{c->hGZ`r(-1<Q~ZzN>PoT_Lio-6<LeSwXN*+P;^jdKdZTvcs9_p!YAG#
z;>$D-u?{K8Jg22Yu*SK(l8&`WEDZ&jg(dR9@7GCM&*J(h?N;8O%4+=fo020-?k(yS
z`${ygRA&o+ZAW#DIJWp8Qjok;|K_#rzRcEJVQQ&sB^fU}^}G##*h6)u;#TBBQAFN5
zUxs4c7*ABb=Y=4a)+Vr>R6@sTVtd?FI(IS7CanU0L9v!L73(n=Oe@7!RUAaNNv@S^
z1~isOHdB-xB~dnt95l;C%{h+K!)kk`=w9?U7aT8tn@p$Ij<Z?2Ti|8=TyngeS?j1*
zwB702F1FX%KNk-uz97A=XsbG7X)WQUyK<h!NUIkIOl`f=TG#S>@hlm05IJP#rJj9r
zV3yDxCVx#CYgiG_Y&uM{clQ0%<8|5-yfuKc?^F*cXTgeXsLT=PmhvBn<*U3fAIoy2
z>KIvn)Aw=hG|4B|?8ooizlBh>%e^o!)~fqKk>vpH4QJ2mdqmOjv;s)WEI|<6I(bk+
zfaq-6BZ$qa=BIIUSRMDW?UhaL(Z;fVwW|a*+2^524t>{E*KwX<TC<H1D{Y;&ZeN#7
z{Yt`o60j7Ib%*JsRX}nx7*%j8?@t|wH%}UWB$KjUY+<%kownkqVLiCC<B;e>3j7Y4
zPwBsms;9sCWdm!`86S9MAmd%=D~I7Q`4;~d=&c!v&tyXC2f>XY@NJ2K*xh^S#yx!1
zy-sM~L+Ei21Ekd^@ru&*%c9`EJJ+^(95@p&sXe|P3$E7S!Jy}Bm`M^Wf<9P-^>QeG
zDlD%b)u+$zPyak=(|FIsCG39x`>C`OcoSmYA({-d`}dFCpHDyQNfj-#Bd@G}U_<bJ
z=?B5^LOUOb*-iL5;4(qQ<i(S!TdX*WDu`F_f2sp7T7H?`43Xn3%d7;1PnE)*a!kZz
zLLfcjKB898<CGaYr9DGFn=LULR^Cj1X%+U2atW0~02M_M@X9DEJugQ)WOt$vwR}1N
zb%t^#icv7Y@kY(U-0m5j8!@lQeOO?t$U1PLCjDB5)^{Yb)HFpya4ASEeI;%b+;=Gt
zeo&A@`uI`5nzJ>G8?+BI%aAd7LY&^ijyAX=xuD3$ntfIK$zxEbdnIs0m_zq}I!O?I
zZ;O7*{FHfL9<^V3T-mtyYv>(1m^`=OTX8ywp9FuswveI;L|Xm_reY4gK_6EdTV@PR
z6Mmr=VJU+?mWkNbPF-?_jj-COq`%evG?W3-l*)evA7SG#S|q)4)o-4oSC2Rz&U(do
zng(0r+#l?cV>kb_-hi7k9*M+%P3vC#X)1KIGd%&b$TczJ@nrhm;~)ug9o-~Z@WxjG
z@Xu{b%O)BnVxLHyA2DX}lbm-GmaSShh!_%|eCzdIdtv4vJV!@RT*UWFID|PGm)9E2
zS}QH;&J4WkE6$lz%T<ccS8a7d)UpZD>g#+Gr&HOjCH9FgO6Pu@jl#-*u70%nrqv!t
znmIo(bQ5X5nfwuxzf79^L8vIPN&c=NdlWdDK|JMaF5S$~fLP?{^{h?*I&I+Cot*Cc
z9q~;9L!8&^-E8np%GD}r1h%pq49Sk11GnNfck)Y4Z%szGLc+_{B_L}eqMl*t6CaUl
zl|Wb$ZzaeaP@)1D2_$}h5a#pV(aWIrIEE#Ltxc$fQoHu6^>3rX&Sm34Pa<bwWi>#9
zxZ*FB#SV3tQ<3i+OO<aAV-%)llcB0YVVwH|3l1)qf&<=BaB#6C<L(<39K4@DcR0Kl
zrS?+Q+H@VuA)mS4a(^FGY;Z0a-qQzF8yuC?e0G8_(&-raZwIY^oJ?XRD@nG(ye$+8
z^@zL63QVDbg_o;~_P6ZPvHdW>4ZGxw!NjNs%xG}C2!$QEwW3e61IO$jrq%i-$Enq2
zT`!l<53k=i%XvA+yQYu9Q5E;?7q7a8J@nti<HUIvtyUBlKYz&?4RVtnCAAPi5OveC
z<j@1QwEQX$*Mc*DYai}NN}eyXT$^8?0bdRbKw%~SomqQGW^KGPV2l|IJL?*ZzX*V{
zN=j^>r2+h`x^rb0Bys)005sgirr{#ypi?iMV678iUA!3h)D;QVm_}b>R_xLVg4Xbz
z#wBFF=2x@MX?=XrH0q2C_nYjKZV{acH0Mg!Rj3N94k77((RWs;I$PT&(tMUIf?j)V
z?h{VE;EgxdsxYxZIMVAqKc<H1d^YI3UhQ0c>TLR;j`0JTFM4mX$46ETIGHfCW^rGs
zY=F+0_+79zCCL@6O^x`bmqBG0G~*HK!rJO0hz8u4t|I8=^bZ!6W0sdk)Qa>ckfCMF
zzR!L|bk6O62S6^R)x`=eE>`^J#R{E9%uP&*z>RBNen4T|r+O!>Ru*jCyyRgz;s{nU
zKlYLbQZO{9%jemKy8+%pFfO5<`;ote<AlHitafvO!^C?*%~?)F(5-g*AW(~({$S9n
zeaD~%?9(x&M}k`6Qph3vF`&j-2d|R8axGlVx?Kx@SDo+2)mF^wrPnX1gS%)Tu7umH
zgR@~0=n*&9!5>->k!r;CyK)}ujr>_Zxm|`pCfRF!#05@X3(zU8e(Q<|02qvP$Bl!U
z;na=8<BF}ty<^Za+<)0n@0(G6<8bN10Srfgl?=CDI1mS5r)N2n&E^uG^8f0>A^Ms!
z+yGDr>dm0^A3l%|m$6L^69q9dG&DGuAwM}KmmiWh3IR8l!94~if7M#iZd|t!efL)o
z=u7$_UUJBpA&bC364$8#1r6*3Xln?8B(LptWUaAU8A-psXLxrh$-7Ec-nHdESe|<|
zB8TM6;W^}uDd*&Zizz2{Fkx_XQERi*QE67LE``|MCC4;c*)emo*l{f`MoQU99)BxN
zN`4SiLVHTu-UX-9f7;XHYG{93$i)!HX3zwM(X)BAxCB45=4M)Cp#>@Uvyz&x;6>H&
zz|8C@25N8F5vA6yh@&K>nHUj|1HlWDG`3qEskFv4II5v_tPx{Q%~#2#D6KI~E>SJk
z8-l{LwceCn;-!PQB(?Qcf=gv|C}^yxHI^8S)z~^#g~_ese}d@`wT-Qk%QOzNB$$b<
zw~}4X<9ru(7|Gafb@<ofjpFd9t=)*m>Np_$4g+ouHNxaJpdv6^Z;ZiWL8X~^n@FrT
z{Eky;qo`PCiETkD7P{~lq}ZBEGu26qqi=CStqaC7NGiFR30Xm7Z<Zi3_Qp;SMC+}>
zWNL3=O6!<0e@CKiCK*{FrHz6k(kZls<w!7FLUBw^k$=s%z-$x6NIOlT#ekL894#Rk
z*`~RL&KS2UjqSp>aP!4j(62VLx<aD^eiln>GovfqDuW3Qg<5Y34n@Y^;825FZ&?3Q
zbMs9JW^F-|pa8~Er0QzqHq4w|rIx8o<my-fjG3!re+4j>w@Mft*i~({#%QNj9eWci
zr_sa8BTyP!hGR(D+RIjQ8PifTB7&LQP|Y#qmP>4F93-|Svmpl^moKx~wZ)zB-r9fg
z_1D9r=Y}8!Nb+>}*T4T`PQF+Prs5?RcYb+xcJ%NeT;1^9>})oF_U`y};tYD8y5Yy;
z#bPo)e|I!K{{6Sb_s<r`i;0Wli^sF`1v+_b)f$je*2`lH)YiH^pu8eAB*OzYD601G
z@Zs>|e0KV5vUooH_~`NQ)8x0s7{~C_@9!q)`PBpgoi8Tmi;D(4Er#Lg<YIO?Kb>4G
zVawkAFnKXO{&x1;b9)ray-G)DVUEAGx~{Fqf8m?+^Vza39Yc8Vd_+(@9QB@?n;X8J
z&0kFBt;ujS{9|}Be1`{&<_IyIE}lms&%m4*vu7(TCSS>bGWx)D^k8^)d9ttu52xq9
z4&U_VPZ!hK`S5J``RNn;_topg;@!pHhT+w8@%r+_pU&P6Z!X`yJ9sg9Km2gKc>T|J
ze~TxWhWSgZi+^)AoSe;0hWF7^@Nu}9pAJ7Srf1W|baFAgoxPZx!RIfBh0?(xVEC^X
z<T$G>5jgflPy_#XZT;6gR;hx7@`P2}#x=n;ZRd*li<rMHTv0b(Q59Rbmg)u7^$D{2
zMk3!x@m;c7>v~qJeq2@;x|Y@4f?Mv`e@jb3w_>-bTd7;QTih+_mUgT6A-xaneQ57P
zdmq~S(B6mkK2$bT;*B!eT$XWsXk|%NyZ|eN#AEwQ5`3_2FyVDbN5eM`tv?&SpZzo&
z9=We#WI(lq8YCsLeN!W9=Gu~dOb}QL4}RfTtB=5!9HTG51}?>Gq4g+$ND+Gge<&Y;
zmIOg+Z-`=hPy!(NZ0WlMC1&+C@x6f3<@xmeWPWjc_VW1b?Bw|L*9K5}a2V?qI(__9
z8;pR)P)t4;@2k}-ijnZXlno26K~+d=D1lIFnP?9S4m#+KVy&X!l0CBd)*vj`QlwI1
z;Yy@jboU@4z|WXB0WOpMk1RBfe*of*QdkS3CIzVny9+r*&{|B#9x^{^%lK6cIRwu&
z-xkBCDZdB|h$d77K>&P6F%d?Y6bVXsEefn&2oXgmU{jPrt!4Kh(2#FDGK<$}Dvf>P
z?JPctex4w5DBk4YDhR;6&qft*3qrxOK*q7v0Fgv;kUaz;*%#y4%OF61e?4gatwCs!
zLbYijVN3}2nLXr0&c1Pd6eb8dR2;C%V#4dV0g(e~)Qm;b5<pXc7-<hVfgETge-#C0
z272SRAXK;rVGSV?!0l7q14xiaE67y{*lj$N<Lwgx#T-j02~sL2=rBrqh{902j>2?|
zLa|%_!E)q~3@nL_^Z`1hf9_#55_V@APgYtG;5a^+Ouua^2<QYA7^-BWOGr`PLv<*7
zsj)}FhT4CWm`KY4VaS3^0#+!~9<m}CP8g4^;0x5UqL^r>YFo%>0;>|d8e%SkU;%=f
zmcJ#4n50B4*u@z60F{D8=}|O+px1S;Kp@CQ&7<q`7}_5yE^p;Ze~hs6C)XU;WX25v
za%as^%DUz#@^PA@Ue_F@v-{H7ec_$dHG?@<I}TeB*@#4~`JUCaP&_!45;pPI?Z=j)
z_-qGM%ds_3B)#Xc)q}&vh<6xRH1x1^?MEW&6W4nl7!%eke{djD^v@1Nu6gu|2bopM
z+qkC4prW#kE6Uz)f7N}>-cCP1;~B?YA;u2!*C1wz9e0HoJNv$ZnA#zxc8IC{0HO9n
zirNn;YS{^Z1goNa1weuY1rxfLT%Mi0L1<9Zzno&OkEiF8dDqgkTgd{4e6*{Z<uD-$
zJ?Q_v4D<c*e0qFxHo5p<n4qB;XxYc90hpDSoFU3GC?8}hf0Q9OKFT%uD)IAZ<&p37
z7>sgc$fD010|%_L-MvAZ*=^xco0&|2T3rhQ%on=Qu1#tM_o`_h0&$gul*lWnFvjL1
zaDXz@YTCd)qe6vfP!waR1svff@3Wh;EqLEew<WY!2H(hAxt7#QWee9#$rn(i|CeKF
zY;bG%y=wQ)fA}5q6@J@w;Vx)Q8jumS!xWirydRW$IiE~^=}@)8VW~F>hTWR}F~PX*
z(J$v%p#f<t*Ua8-96xnx*)=8f=JLTrGU@I-f!3u6)5l2>yGOptv$)~}c9FcYv|FWH
ztBVV}L|$2OH^z`5Tl3XcDjD)Mr~BDT5!h`0%j~6Xf29|TIX2V9n5&cz4%0ri(wK@b
z1!_GpFistS*;9&+k(iq6rURuKGq$jRV&IVMhJ6obIW_r|N{_;5n|)DnYZOqt*haF3
z!mJEjNxNOR#OgI%X*pyu)<f=P(Xq~#r;AOCZbXrH6s;1xeL>PitEdklePIJlRSphw
z(0$~$e_fpg`(}yTi7}y^H5yq-o^+idfg9xrouUdL0Ah!jcgm~?Mw002`vSXJDMVk`
zgy)zr;UpjIexzFwF<KY7s7^3+-Nx0<gwj@xCaxa#Hm<Q6yOS?kW#<hnEp5fpc-j5c
z?wPOeo<w~wOx?=e>U(RtvidHYbPt#KT-r&ne<gu^GDoHQ0AV1Ew3H2K4(BSV7EOxZ
zD`??94#g{ar`)5+g&OqT8pS8s*>Wvj?AyC#6ar;tJKL>M?0IJs;9kmC_#+AUlfN|l
zAp!o-<`dbwPIy*tUt|`$uNBAG;6%s)hAFMXz1>b``-i;&qLixEy<Jvng4lr^25MCx
ze+cv@ZTOhL?!_eIrxa4nk@G&T2}|(yjp#Cj958y2yjwzuD9!?a=K|P7@(20gCdNLc
z{1X|J*0l|A3n{4ExH77wl77xhm^;HY=XJPVe${wqxaMnzqW10Q5`1%q;oQYe?qVl*
zvBO>L@Qz|9B0;ELJ07)My%%seJwAOsI6Z$gd$*Xro&L9<86F&}>;V-B@M|~J*Z@!h
z*^wO-fqD=&rn~`!qhRYGLxq&7=vb2f16`+XhnKM&KNJBqm!Xm%69F-nXbLEo`Q<<k
zf05J%G#aqm_5itM;9HPWkeSmQGXH-FwN^>BxHq>0<WR1*N}@=Le2<dk@MDm}*WJ_G
z?#nCEV1?7l!tnNeQ2asiUnRNG=)o!_465Pn=kUi~N-+F9@k3G~@?-h8<F`IMpmN+p
zaSz3<N*x|7DQ-b-LGA;Z-;y8T-!`|sf9COq$2&<BB(QA79hDEy;m^0<b}}n0g*V}6
zc=i9iyzTzkRrLXdHhwS$g|u$?>*p?fAf<(|eo#huG{djIhwr=Jd#O|BIK|G9#%SF*
z2aM3#v2-smd`ix##vOxkNpV3ur2hH!W9VAGeI0cs*4gS_Dm!xOg0VL*5cZulfBTr&
zpW5U%`<YlJ9Ujd8vwGYuy`OWx(D0PnLtnCouyo-Qy`F6epcTr_lvB^)S+oX-7lxhP
zjey6jo*LB*YLiLcPO26?x#5^~U0Tr9P%p89S`hkas1qf**#|iu<`PH>3(!kw;~}i%
zr3(}nL+h->vKhhY*7Q0=xhf>5f8A&4Pf=n{SJK(apwG(zX+QC}4ZTd1a(IaG;5Yq%
z3>F-hGaY6twP^?<2^&JYm6G#dwra`=rDm%Z^p8)l-xm9$wWICL3TNGH{dz2(i~p5%
z(FdrRVWzTsD@)!A?RfP$7*QK%_-(v|ls}7KzI|PfvtqN9!d35@Y-P!he;8TIn2|wU
zJ0+~dX`Du%<><gkEW$DS(M~eZPhuv=HM%i&d3{xJU}?%BuT*kS5K}__G<@0)8A%K{
z)EV~pYRck(KWN-%C#kY@c@9Wzb-~W$7)@r0V*rF#w9PU7`vlI>8aQ64tleeKEMR{=
zM|lzkW?F0^yF{VT_9H5~f0~R)InL;@jlWg>uR(gvm_b-G2<({G)pr~^B%#AhlC(e3
zzpvaIOilyFO&*K;G1x|FaipoTvBj9m6FCP%S;a)zAYGAN&@IbdLAU(XcQ(r^rJ(1r
z>Qs`2MjRORz|lWj+(5u7YG~&BtKwi26tv_FG_IdfKNggVNVV*{e}w%@e50ii<gh{C
z#y9U&3R-wh>`n@*WNpaa@UErxF`;vOKi5)wvkZ`$q0sa*L?8RZSOp~it=9!z^n4(j
zk@FlGcm|)H4bfzj08byZ4IpE=aKG^`pcT5k326IXM+ws+CeipM^?%^L-Z7iyzY+wt
z8*ta^pAdyv;HT58fB#OxQuY4<r4R6|j2Hdvx2pT!ulWPc%#1&~<d=dIzn8L7@)PGR
zqW)i%{OG-qN?%d*qxCT<Kd0zl<;j((HsRQ*)A>&IJQ6Wk;GODu)H2SxT2EYFtLu;^
zzMo+OhPgfyj`#1~h?~{w=bs>*u6sE_y0PxK9JiNo41+4e2HRNm=>t`l1?oT{e+xVa
z>7chY$M0&sdn25jc_{2M2rwTaM*uGV2km$kjvIa)@{kx$v(^o0MdgLHG+UJh)YcL0
z{m4qEcH3@zFLa0?qR-qhAH_7lOoK>QFb(fg1aMvbNleq0u$xiG2pPQ6DtsrJ0j3$F
z)SYMs7#5v#YxUX16cdu2dg+@Kf5VZ<<l2q+6pBfkl+$%DyObO2j{Tj>6$}GmOB+7j
zfMLoWrKBZI%!uaIo^8g>cxYMvO-_DRkR7>>;XP63jdwVRYbEw!Kj>pA$bitGXUq2A
zq97OUVHW)IQ4FM*fsBST80gC@L)^wKwz4FYYc4!uc&5$o1CraHu@PU<f4<Dcq~gqH
zU9PmIU44cciUHysL2J~6A(25#$3x=d2p5-9q5{P{e6O1n1y*Q9%RzCn=hkp~JH$P|
zyt{JhRWfi+NLHtvq}(&0kMTp{7wjhi_i{kJ&j{`1X$ete<OnPQ5DhuDM=jw#A@$ai
zdyRMxkq(H8W<JkcrPuT@e?0@`dx*$!kTW9L2$`3n2IW|%%8xTs!-@Hbr}jEFk_vlK
zKK>c|*y@AHeN6gZ^v%|EGwAl-q>m7%grRZ!<VR;lcezFKF6z}SVrVVc85hDf2Y@44
zPp_O73Wf2U%=Sx^q`4^*)z;!-J2|}7S^6EW$w?rbvytiF3{w9B1RB(Ysh6=FKNJBq
zmjTiN6ap|Zli{c+mwN0#mI23?5$!-8e`DJD%f-Rwj%0*(bwef}F>iEAtdLIMReOlb
z2m3T1O&SA8Y!^cCatIQNG?k(7*rcg=X^soyd?c#0Eva+_|4DM~+-b`&1u~oKYG$1Y
zO&y!ZOV9?`pl^R5J*E?))J!8qmBt_)A%!<VP2u*f`}Xww@ZUoft<r*w9}IXVe@GAi
z{CWrv5J(v72c?7u9De^h{CN1Y-!<Doe~*S76)BBjc3MR$BZ#fW@Evqf<jW*G=Ubk_
zoG9%ugn290X%YuA0rlVSpF=nD{aclYQph8|qY=J_Yi9p-feu3InX1lpzMleQ;j;ic
z3Sj+o^ulYia{62j$-SddKV|!We>yxG8SRrt=$Hi?Jmnb7{%*Q8TZ=MFy34gnzt$ue
z!Ma8Ayk^D~c&@-?wq^3Xz+#of@qjdD&H{&e8*I=BYZtqaT(zvLv{)^`Ep|zO#||AG
z88LhcFdNL7&I!NJmFEF^NY!r{=Q6P5l?krujwU~c`kZDvMi4%zsP&9}f9dQ)gqYw-
zv<B3H<9Lc1^9g2CiW;5Zi3=$<E*?IET66zxl=Qqa?GbEb7&1oKBnHSLiDTjLr@E%P
z$3fHB3M~W^^|isDb0Sz?c!p79L8+NaDlH+p#*<9i0c2!vVYJo@j<T(}q3eU#!NY_E
ziyaIICa0P`#~$OyWI+lde=@e3hsl-$>3K%6F&9SEs@*5XmDaPJ>ioqHCEiBSUFy7J
z6KYSJwxo;`&a0lmw+kDE!$T2HWb(XihE55M*b_C0rr1Oy1gt{XVi!5fBLxd~3+X1q
zL}BTi*w`<Q#O5xEwBooal4^(?l2Dq3Yes~LXroGkIuVyG(`3i1e{+j1<@r~IP-aBB
zef_rBVFkxqS~4AxCRymHNNBS2+m^}*Hq#Eble2}qGzg1+N|X}-#Z{&`l`_B6$rY+{
z^f*>&rssB|u^ByPDm#cF)6K3}aV7Bq4&9C+n4}Y`*y%}9S0!8vvn0^j`Qpke$wt5D
zEZK85Xp=SQa!&8_e;m^!?Qke?Rs+?#(JBs9s5>&4;YX`5*GuaJC1D1`Lu1F5EYK(H
z$(clq#*KB4{4Ivu4Rz-=RL*nNL?;mK>ZvlTIzTe$W-S%UXJH|`2l;Eel{6}qOp6+>
z*Drd*Ng-7Qf2|>Bax`-_J1i!~ON4bVwXyl?$@OzVSOP?pe^qwXI368j+j=AlD_ws=
zupd-<Of`82K=a)Tbd+%Ds{uC<HUTF2CQZ+I$Yd$SiQO{5mUFP(yJ}q5G;o#5+xjsi
z^h>qVtX)q&Ri7Z=^lI>274fmNy&J7)&W*<Yr0Zd)Ddt+Ov<Z7!ISW3~#tIUg_63Q#
zo(hD!0*$WFf7u1kGGdyLNDVe{dXK6pA_Y^R9T!%bj5B;UUaC~xtv>m_p~f7=5IKrj
zXpuY;SGq5zI@8_;5`y7O-?i;7Dm|W{2y~?xVmmmRwQqK{mO@p55<aW9bOVua#(MFH
z-0j#-Q#r1A2_1hQ)Kt=(!3K2rn5L5EG64woqWro{f74jZr)@Q$luF#+#|iCPG(BNk
ztt2IRCiN>`(@V<Dsh)-_%_MCFSiN@st!`2%d{Q?ldVciy`>Ew*!LlLm0Fgj$zn7Sn
zBoyj5lvwD(N@;dVtn|EIBC*%>xOC8~uW6Nv4G+6%_H{;|&aTt~%gQjUZtAPedY!F^
zY)|5(l8kw~{eLfwuNm0wDjiYiV9;|~W|1=rWv%3mxfI#e)|=ZU$i;d$Q$mqKmw)$)
zh$ZT5RnrkNdc4|G(h7vcmgzP|#r9RLZ=*_b<dUUpuE=px%O1j))1;H<EJnAo1H=Ad
zds2E^pxhgawxFASh0S@7>Lu1nE8+7@VDlx{BXyULTz@el#fhqS9S=cks%@jXsl-mr
zg6#$=adP8TziBS8qStPh9@<j3_#KlwEw#{2-;vz)36^qN80kML!75u~(;TZL=<O2E
za)|@bG0<dCq0yR}_)UHC(v(RdXE^So-oq(kwNh2ewv-Yqzba}ej;*SJfmxwtoBrNB
z3r%hNT7Qz#{C!2+xHsdX91F&in+|E$T2S@xK{^Mn?>rJ^>D0HvMP=EV3@#+HJ4o(E
zE@M=G$~yuD(<cbcD}Vw*E5+*1s>IHX`M1APuuTJU;U;Oz(vZ82DVn7v*;eMfeo$2g
zoD(Tl(Z@gx5F3Qj?aP4s<=YE4Kv9bN@(teiyMO439if{q{8N?Gt$5WN|MU{Sr{PMK
zMkv6q9sjN+2e6@NGyR^24)GU4J$$C_;p|lcykR;Sew?9uT6To@TU^FYE|`AjMdlBs
zS}2w{bzGbGd%L=|*L!^x_F<}j+P?eLuXJPQFXsvRnqKhjqD^}xke2-4_p=A&Kts;1
zNkB36u<hcVMCokick`E~2X5XqRCz1zYF93Pjc23JeomZP^>#nCL7NJ3Gr4sPJY0p$
zc-foVcENyy&Hr)6L(`YB96uBSH9412BMuaojPyV<3Z$%s+JHi$v%*!E$n-!ge>}e(
z-n}XqEOS~|IlO%y1is+-zu-7%^k9WxMv39=+wjjP&ZWW6Gk$<`rSN0%x5ICJIx69C
zhQJvDM};^YEeIS@IHGWl)cDQuBk{M*W3PF=!S#*<1PBOQ;6&l^^L+Z(+dmFG6U>=6
z;b-{jpI+V$e;o?>QZa4(V6<g|e~aPMw?lXVLNjCiASLqxhQI$kd_H{ap%Sj{I||DQ
z6c#r|>qc18Fs<ESjAF`K5>{YVBeQ_a0x}7KS!4>v^jqg~je_AQR9JkM1i%1MmH+<!
zGIS%qeki<4u9#4^r$)RFpH7@8BF=b#A$5Y<j3Cr^ubwaQEQ4Wy>9GX<e{`T*ApIuu
zF(x@^(O_SbrUX<HzUm<J*VEB(iNEwruGh=RY6eE`kWpFT6!TUr={Oqncs&i?FpJ*T
z(;yb{BO>OV-(_R3*^XY)7$@46%zQNfkOxW2q)<h_5FFh;B}{;>s>fofIRpfqYnh&e
zdX_Vbrg<{kiY+5P!_<}4e}ox^W<;Bk*#;EZRfa4wZc}3bAGy!gHse!z$w1nk&lMY6
zla>+&4O#Y3vMy&b<(^4bR2WbM_jgFP6l)@rA(q0nd*xYi6oLVqREVfRB9uHlL78Xr
z>HzJWRT`;~Po#)bQuaCH<*lfLQ)XD_%*B`}gXco?NZ4hjCDPW;e{qDp$I)=EPYG8v
zfQH*jei_o3_?sZm>ImCrelmY#LST!ObTI7=9FU>d$z0<aSLw|xs$YA{6h_yq|FDYq
zkDQ%PUL@l=f>gojMsgCngO7}L&ERoc#Q@@ZDu|B&4Y}Mh_7M85lQTZX=xjt~EwSXf
zTRKkeYSJ2V`#8J2e@${~Ld7`QaD4mP;2X0~l%Q18v*=}$El$#&2Lgf**>3Enyg^x}
z)lA-yLtUi6T$DLr*f(?AiIOUuWeuJy$+*=Zq^)Pl%f%>^eBMl+7JXnd+%`IgTB|vZ
zWZ9;<d!^s5UEQ38(#%;HwAPF}$UHrBy*wKe3~;ua9rIA)e<{kZqq&lZpxBaqAEr23
z`Ke8tV~`s(n;6+{Sl4m%?iKUMC9g$Vr5MnVm;l=S&#z`Fqna-^z=W{f>MC~$Ml$6M
z#F~=HOyxzkPZT57jOSD*Xgxhwe%}(ml8BIQrC^f#y}E5?rwA&t7mt+&a{Nr5c%7af
zq)`q`T6wERf1B0cL!-Uqz)YIdFDc>Yuc^v2{3?y2o|sa4r$*b^5g%Duq-o4C#Wb<w
z@(z8s22nB?NN@GYcJ*!zF6R9W!zaqU<~%3qrTpP4PPJhG{#H(#)!)TdcWL})?m;rm
z#jV^otM}`<tB4}kdP=0Q9wY9r<oe}pmJ%q`kT*JXe@p%vkh2d%WbY6x)al{BLm0+I
z%o$@Wek%~dhbs5*?$0lufBE#AV7TBvVFe#BGHK!6@GU(W6b_d5dia{2(e6=Heh|zf
zX&5LZ<2JMcXk8ijG#n5Q11un_X<x&r!bFtoYHsZCI82)h{Ar|65CPICq5E175>XR|
z$Wfspe>?>!d2M*a^|UHHhEM%WQ*qVr!{PIA_*kUr75)pY7t76(FcGP+O4I@^LmpRy
z&D7(BkRvV+)@<f6iQF1M5eUh=w>Bsv!!4ZBNX0D;At-`CL-gH`uW;Wnn&gpvj%mzM
z9kU|ZqsSsnVx4|}`j(e{!9^dc?*!&3lxG4Ve;t!<1E|=@Tk-TfJ5MHGJ)e|*n*1<s
zhox-J$#ud@(V%v&7A6w#aj58vd>$u8Kkk(cYDjC?A)9_fon)enW2lxQl(p?TO;@o~
z;ui}BW0uNpW2YWWe%ie$n_Wdh#2brA$mM#X#8ia{H8)(LfeOP+QTgR3I<)Hw0HeFo
zf14t3#4(jxM_e7GGoFN!rO8!DU8``DsLoaE%Qhr$_I*fXW~l}<^E#@n2T8q-Gq9x^
zN+PSOJyKS{#BE#u=8;H=%y_x2O@h~4HEXC@$Y)&xPE~r$&0Mphud7;YW(gB*ow`7z
z9t*F`S?Ap{cD6)#qsK~%iGnpPrnjo?e^h~Px9w=AJep~wG4m*D9-~om+w}?JM_2u`
zm+ejuX*Rog-Fa9!f=CVUoJonqJ0C;iJ#Zs=X~AF7EmF9Ja@kYSwYe7pW)+<lBIXqS
zQpEgYXlLb==2+!4TQQC@AjFNiS6yU)Ygz#FW*xOTd>+wL!|NoXOOiY0o^@%0e@%@O
zS+~xLX`N@ytFb9r&V(`b`h!_!u`VQc;Q#~kw(aB&b0(=F903_jT1`n#_N923;-b-N
zMJXNt!<-Cjw~|hUb26`GSy%>|9IuQvIo`B*%H0S-HAFQ0L`fb;1xOpF1S~Ed!Qj6u
zb3JL81kBso6mDto|1#IZFahw8e^d{>2yMXZ%=KtCxxA>ls%~#`SD{>pRfV-{k~?eF
zUCEvFXy<Bga!0fK<gN)g%SV0`vQi>6b$y#vYdOHw_op&n)jKmZryuTVMp4y`E7{5n
zemTpjLU_SlbtS<rpB>cwKzM)Q1g?bU!t3*e^8w8L`CJ2CgP@jAl)o&wf79l<f#(R_
z_0TWxzLohb@4neSw|C!a)7f%G+M<jO`%Tl3U)ejwdo&LNPrYDJ0pQk-ZGeuUB2<%i
zaaXm4Q$<HJJL4)Tu`V<AH;T8ERQQLf<}7BpaG<uC^kKdu1k|wLvY80XwK!%;)vCx(
zErFng=jL!j7IW`X-nzYBf8|clIQ-JV^OQ+#6gy<R*Iot^4mV3o?z+oB;2GxN+n0fM
zT>g<#Gv)N9EE~w&WEQcGHPyA3M}!mcepZK7^5q_SJoANq_C1$Z@+G4V;>zU}$!$Dq
z?D4xvxn-B2=jr_XD!TIJp`CQaR+(9n%QY9bCOuY{ZQ>m&&C_kFe@JsPvT?Nvb1}@>
zW;|4GeCZ06Qkr26H-k{^zM9`nrvUUTi#CWkCgz|eaOnPS!tSeIm_ITs*m{VIbYI~s
z9?JyjdZ-*${iWDzx8nIepOFYH?DQ?vHQSVq8LTV1*%uE4#x@HITV1+3C+_oj#Wj{1
z>H_umTq#uCGw?G#f64W2c%LtLChbh>5Af3x`7suPDMxry9=gxwGZf*&0g-<|Yz`@$
zlA?#WVrS&V-R$*zGhbG%PL}8<ShBI}cD{?(6Sq^wknSjepCx6CPf*=&+L!nVtO2kL
z-%kFfo@YsvY}M~nQ**uCt(#rPtWvL9I_c(C!ZdE&;2Ia2e^}j(z(PZvu)V)6VKcE)
z)8}gjD;gBzz4@9zaMqjV!e_Sg^$zLvzG$$V;<Cr~;LF&;+R(7;u}odueUT_<i5(Pv
zLj>}UIj3gy@Q&l#V=wK{y%y(v=?NRkFvHdvTQl<O#+7sx{$OBcGu%)=yz3g}NLYJN
znzwbTl4or8fBS>8Ek|*(Y7WdP`=hM7#kO&MzF26x;+|cOthJ+IWtYo2V2}SoHl@0F
zeuFPa4NUkR@grTqNDLG+3@aB*_ogFWF=yQmvZEH+4l`;OJiv-N3bu`?oh)}YfgWK?
zS=Ai@540s4u1)qF-f2oM1W*5i8zU{!3w@Dx@&no;J}tltU6Bs^_iKvmqbm1A<}$ge
zC9)bYPUUYGbX_$pR0O>pp>2929#|l?058->?c|@NJnFwOjs71N=%jF$u^c}X0XCN*
zb|Dj&mFYDpf3;l6jvT!WzRy$C3yiFV+JM0zGwm4zx!Le7$RRLd=U@l<{~YrAp%!Yb
zs;*vqXm^)NqC|?zM@nP9{8;$%>EZS3!-sbTi)Bs=E0?d|7lAK0{wp{(8ogK{m{DT+
z`fK^y6X(+4=M_J|xl;JC_}k&PetuNKVGn^l1hxwCfBa}cV2i>Qg?*&PZ;l^{zin=N
z&*Ke_cN`#4K-vO33Xh-D^WVSz^?~OFe|Xn!b+uA4ZTw;^FkDkvEmGFFT0n|8IoC)P
za7Ms(@$vZ)OqdUFeOjis`h@F%zz3Z0NvBWf54d9OQt$8%G|}ot2>kYWM3{B6cX$)&
z>HPbYf4^-St$E^{c@vlqU!R7jx33R>Jd7ehn#^*)eEanfUVzZdSid;Q@a3QXT)sbi
z9*MTv-~`H-8>4kEU&Azt-e}3J@+99V)4-j~T%Qt`g@@4LHS+;Ts{Hfkk7a22@~N_7
zJQG}wvx$#!*IIqxfho>-pdodl+O>QMI^+2&e^1sjBg8mM(Cn~_depoP$#~9`g^>zB
zUCy<4%sDl!hYuV_)-+E{Mv7g_4#yI0TKP$?VYi~DyQrW2Y$(G_Tj$xjmM`0qkxDWz
zCY|AE>4wYT7eux@o9x1Hkyit+)Drr^OV%b)wJK^_p%irwvW2~8YlCd7Z#XC1O@_r1
ze?Tznh^epuO-m#rK4tdAF-0=K9N2;IF$*KFbg|A!^bUTQa>dL_cL_*f98i4v74s|0
zgjM4T#QN;m3FIgtN#0vj|EX#JvquhXlDsoX?zLEH*#bZpYV~3iXHJ%wSG+Q0>S46G
z+`g@1N}G}TVBe*;6EXe{y>F1U@W708KO@BrVHa=I_b#?*v)++3vyAvV+3_t}p@9j1
zaG_E=1~3Wguv#?zqJh1Wa|!RC3BSzypZMpemS@5S#rxmjm#P9mCx2VCMmlB;j1;%j
zhN9Cjzew{3TnjRJjnEc<oBR%+rO(*)l-up?J5G^>JLWkQi=(h&kL_}%qP=Av&7uiV
zYq5inZQw6zf{t<ZV&9Uz)iCRpZ_Aiz=zYuHXYCs(ZtZ=>LF4?c25xc#<@rjh07T!o
zQ3uxro_S@~<PuyI`G4B5Ho1cZwlPC=XQgX#TXo~-kVxv}4D-(1BMd3e6#D6b;#L^S
zZgA+`r)*h$ZED*OOP3i!<I2}ECbVmZYJ<(_SaEeB`M|s-^1u3~!sX2rR%^wq@SS^>
z)!D}o<Blc8Ifd8g7?rJiry8dMRT4GzE%JEJVd1rghXK`Ya(_c}3&Es<R&NTW*Q7&n
z4;z5i?TT-Jd`Q~GXr@V}Criy;H(3o8i!CNA$(GOcGKmr>TRrmm@*c_=!al{oAatlQ
z#Qa_n8`tIbfYu-}Gz_`D*35dbpWKh%2a<ho5_z-d^=40)tt3k^V6<+q0v&SI`Hh)w
zaKr-rhmhk$Q-5xKLp33z>v0C#!O0^LT!xApm`NyacR@!3)ida5<mQ%K>pidedKn}8
z7zF&Zstns>jLb?GYlp|=je5<j9f=0lPXJ%Mt!u48Ay=G6(K)*;?Ml?WHdE{;xOe$0
zL$-brMP_$d@q+MIPb{izTjY6dili?=B@)m!{Mr<;pMP&M=L_hWP#+x$#!9pZ)#!CA
zv6_LA!=S9_lQ%_h2Tn`{!Zl~TS(K+ZAPkeD@(m%{i0nDFewOjKY|GBEpiI45KzS1C
zRZ_$aJ!D&igNmP&P~puBmFJ>zP=8g+-P#bEYN=_N`Ne$Q$sm9HIHeU86^GUxq$JUo
zYdH#``hUi>XFZd-L^sxSH-frSF`y1JR3VJ;vQ$so(x0y(F+!^(iY?KbvN=utb7f4V
zNN>oPM0>VT;L;#|8e^)8lv4^@K;@)r9B^}rTLd6rvlgV|dOo(>pWtLO+diZu+A(t}
z<2#AC7XF`!LkR~)z7S81tSN<<fUAI}{#Z~cK7TDyGf+pBcY68X5}HL&NMno*P#y&O
z9Mso`5C8e`{V(7CAs7z$Zy14K48<h8TYjZSZ5VeJ<I2zc>=>wU+#oQySD{RS^pp>Q
zCl4$Q-<AiQ2)j$!bo&`vQ7;|WP2UIs9Y|FtXqspZks`Iyd_UVpT2-AQ@>G@)o&uFT
zHh(<gcp4QR!>4_wt~lyn%ft8O;d7CvM*~wKR9SA;#EFw)2o$YIsYy9>`<7OZ9ZcBB
zwZ@Lk+=f2h89)&Tgf}QYzm%y8r_|DLOHD|MC{Po9kL?@W4~!;xB&F0$hw#IYk)V+x
zi!_O~{{HqW5Bq_GJ~iKo%wtHe1yXucPJiL<%Rnn){f!(kcc<RuOkbXre#$1Q8%Pi%
zcXkwMW^$&jfel1)37?l{iILUE%IoK&;=xU_Rvi$}7^xev07?OqVG_$mpd}8p$8c1u
z?=6Tyj}1_~aC}nPL*`$HqXbcEL$_|y+;s9)dUe6MwDRQT#nXCp8OAzV<yv%U_<!1o
zF5)Sl^84VE<th2>JP-MF*Gl6+LJi6yxK^X&`OBG6o<;Z2k7}gzlI10+RW*^Zm0*h)
zo`rcz(Hmfh&0VY{ZOOVtv&-hJ@}=2Hpcw8}<xfZ>MnXq`YPZ}X5ve%D-@^*z$3!7w
zBb9K`YUym&fjwzs!o|>2j_ekgVSo2w2K9g`oYMu2)T(Pjl|)I@+Bv#o-`+#r18asz
z0nUuP-Vtm9s;t%1)(wgIns=?9Y{k1)SDC!)zn8vO#z2n^2-*u11et`xOr`q6X^XNb
zTBK(W7t_it98*(K&zC8oaCh5QOzfBy(?n+v0CHK;47!)2WM3gmWyFh%WPgX0MtaXA
zpBQLNg-pcO^Fl($aZRT9pEFpyGvZ3W`mW8rE=oM^uXKReuX-t;z(!fQGL6%Sf_^Qn
zhmiliujCTF7mTZDZcuDWJyhlp)D0t}w}vTqi|f^U#dc-)Qs?yczNxd~=pDqw`+XI^
zl;wYe;)4NnernphQ>hAXdVg|@VAM>TH(qEsUwdAu@MN{Nl3Cg7Q+M>VSs^UtpiR%{
zaG$*%rU>8*q-?-fuF-Tx99x<TQu&$}zJX(s3-@lq70`Ziy;<1Xw;>BiWF(dV542I=
zR3RMzKscCr2djIqe}E$qIr(YxGFYvp8P%~|3PCPI#F7-y+X|rf*MIGUK93uF@Bm~F
z4F`lU@C4mOf=xTohAA&r1hP@x&e|<G-Nw68Gc+O*tL3a2tF~q^fE(Rn&%P>YZzvp<
z-lr~5=F5bT5AtY}z?MLF4n9dx%t$rDr{OXWox;vGyL;kFbD%<ly9}r{_Xbunk$j))
zF2{fQ#?%>`<iMPPfq#_%MC8*bvFEhYBJGrnB~*ZYn?rvlxERRyCw1d_Taw=K^RE>D
zrX#SK<W@)x*N~&J`K|;Y@_E<xWz%w;at>unbeUStCX}lRN)f_~Eh?LW@~ha)Ut(OT
zcXvjZxAC>iOUY#H+vBRvkac#zt#X*WUfIkr`k)Qf2%b~PNq;sc@@@RBc)5@JI&5^j
ztU6FiNUv}K_d|9FYru$czT4D|bggQ$xem333r%7YYs4?kJkf!1Q4OFcN>IXf3cel*
z5YE?23T1J;t+UwXB28!!PeXl;{8Q~sF^=5wX(vNk1y_5yRe>}VXwF!46NE~UrL%%u
zp({MO?GUJarGLGVqpz-hviXZnVh#pK;eyq;slHI+O8HwG*!nw~6duZ3;sGu@H#+1&
zg>g)}YHJpKQFK?O3buaq)mOmm9usRDalzpiuVuzA%`CL`Bc-$sb0s}8dLuklGTd;R
zsKO9{TLUQP5T`1;f@WKM=FX=73Vo4qYIXMC3vj_&rhfs(nU9)=M$`hTEs8~RS52_f
z_4XXt%apztjP0+g9W{mKj$h+)WW61*Q3Oed?UEznjPXv{jaSrXDkhvD0o^NnZ)Yu~
z4&IbJvg~AcbS}l#0fEhj9lxRWtW>zyr#so+C7e^BDMgw~!@1@d&f9ejx#e9d@5V9(
z<{Iu5*?%wOOkVez6vjyZTu{LEz^16bl#DmG?T^P9a%ZP5^3I9td%uy1)!Fqmmm}|R
z(MCDO5v~o~O<B=<gS$=R0N{dpHZVI)RIUu(ffetW1$ma0m*(~}^mmc3U<-#Z1O7Mi
zrL4vo&a&kiM4OF8HI?Oa_O3W)Mbq~D4x<OlfPbIgrj@JW6@?$-NpoNVk>)Vd#)*Z%
z-r@c})e24o&+KZk&`rg{?(f4*lPlP@Oro3`!|<2m<zTej5a=P0Y`p>uQi7Mp3pG{#
z=E(IbXt`6eowATA^D)i2{46ouNoV8k983~kzp=2nyUw!aEMx`6Z^r%7<I0VjP<o45
zUMea*-K$9}PTnD6b1Yz7ae?j}J`Jv}o<#2b&kGcxhQCO7`}**IaNTu7m$4i_6aqFe
zm*K!A6qnJ31t*s{4M8yiR<@UG4M9AAefD%V#}67jjQ|<}bQR$E!6DEUO;<FHQHIWv
z4n?0`zIM*fTYBD0#3+Ge0}X_Scs~C1`ujm<g{25KevBvn?fLca&tXtsX<=;WP%9A-
zy1&02;td2NtP7n&C#2Ke_s`vz!yjDiFqx!h`aec?-@kV3%1icOkmNyWtud8<2bC4Z
z_|DoObiCDj5TB$pWHFNE_(%s?tDv&-<JrTkL}P{4flDN}N<%(H@+l_Cy)=PQ&+;d!
zUtJ;f2gx2%w3%cjE4_x%K??1Vj~-(Cc=h~BY604Di2$?nE5fD@YBE3BZd~?i_U`aK
z-m};1g@hLh#zws`h@kD9U+VaO(D&!83t=<^N||1l>dMQfMSLk|1loxgRLC=^9$V1^
z6fnxkO&HVC2!)zUr4*Ku@?MfJq3FJz9;>-Qd59GEhPhEoG^gux!J_c!xTK|AaGLyN
z`RE<F0Nx5vT;KxahE!yZ#0<B1kQydrFqqM?XtJYyA6hd5Q~7v|6})AC#FY^SLd%FV
ziT3u0*OJyEUXN}N@mk<o#F^PF;%NJ58i<9jF9Rw1m<uEH*QSwA5l_LtejQO>i{M~l
zN?g%lW;Nd~ncORapF~?e%{ocTATUhOW=_ef4_3myh-qyE^*|d#Mkp1^O>|f1vM%*U
z?4|IO6BS};uGh;sYI`Psc7vJq|L|0Vl%4=E*80}<p4NFQRm<$JB}OPG|LN<qu;g_u
zaj6QFj;S8KrwRlVR!w}gSO&cCUeDP-q;m8q6@K&hMuBgU4a$K~a;xS;-@-iCzI)T4
zg)?fSUbKB7C{N-acR6W($i?%Vi|J$B^^BfOs5^~4SJGdVOuq1cdgu!*j?g#mYSU18
z0t>8w^(u+lQ42ot)!UT=spIuw^3)wFhoOA49vfB;Cei4zJz8EIhqkD<%jj<WlkBRl
zTxNH%!%5*(wKpg;rgQoJl$PR>o3>xj3;s$g7kA+?!_!fu_<j#y3TEvTkLLNUeL~;u
z!xMwj6}D<AJT){(#suN49k8SaEM>dobNr-goaHOAW!2C99hA50Qv6Qv_UH${+l(L6
zV<|;8d8+7^a$N26Lzx>4?F%obZWGK@NH)jw0%`H{x9bo=odQkg^n1rbiOM+j)W=%2
zmh;zdtB$1+;x7W3MN!umIpYPQVK%_ir1KgcpiVAOzA~d-YRk(q|B#o#0@|UKjtio`
zSok)DfFk@a+5a9hm$4i_6ahAuQF00sm0t=em+*8e1eYolK`sNJEs2*~6hSpy{@&qF
zaV{nPEZ#f3>W6zN9A*g25EvEW;od5N5s49rIi&H*@ke-X^VmyXZ?N8Rr2+-e7MLg=
z!~2K7e){7b&l|40BgwS!ld+0n4|R9y;a*YLg~ocVABe(xm+TZl9e?-y1>a~LzIn{w
zg!v-O_nB{;Ytmo%R@s)CrX&#~?Ho;<Df~I9<@7JexE%fX9%tml+8L3TOb9({q3ZY)
zPGOW1e;%JtonE)BKq}6pGb4vt3z4SF@&@U^Ob40s=fF?<tcRD#_el4q>aytGl{Jwn
z1%3A%gT^DLFkZD?Mt=m)=kk(gnwDmBCJXB_#5hhPOC7nxrh-jJd9qCeT2k0Bs>-{_
z&n5nm=S$wt(`*yTUpk6;hq|%2xQddEW!gbgzOHYylt#|eQuNRu_4B8N;2MSn)BF7W
z?&k?M@_vej0-KfI>FGaH7#6`mjWHIlQVHRM{oeiX&#zzp@PGLy!LZ=pVFjNQY+86X
zeNP{?fljTp)3^NDF(r)vbLJ5S)MP+UdQ2z_>%iyf4va9JsG0UHj7nx546j)%G&4?-
zoEbDjWe<^J87=(2wS#3>Gm9uw-A4EnsN}WbBi5Hy;bZt~KbNUk_4n!S%XIgl>eEa7
z!>h5}tf>=Bg?~jbZd6)^q+`O{vU^N0ILFn&PRu+;kk%-uKoIj#pPwqLDrp&Mq%A`r
zRVb99@;Z)hNIwXg^pR{*GmT9f20?;FimcM4*7^PC?|Io*EPCI(CNlRzdL}^W-XNxF
z9%;q=ha+9fe|dVDU1ICwgN$)c<ya?XpHh!GiX_e@wtu$%H4w!XK1|IPqp0_@tRK!*
z4-Uy%d4_t%MB1o@YbLp@YT1AVMFL?%s;X)!LFkkMxJ*hyun(cF83_SWnu`)gSy2LI
z3~jENCCV(lNZ57lPO`}}+Ax~-dS<f8^6Vy?1c#Ij;h9c3d?&soyt`I2Ip7C#+Byjk
zXP)eM5`X;hV3kDZ)8ncnM};+NlrAU9Rc%ITQerwSPQmqwwOCKKwZfgUW&^Fq=2T<i
ztckAWnv*4esY%flz?J3SfK0*wR@rd}^pq$>=p++ts^>4$DJfdp!>+c;1JJ33xJ8)>
zGn|Pky38)KveT%}D}7OAq1N)KDCA`kv2KAyLx1K_0V(#f-Vlo_?U~h!qsu5}FWSzL
zZ6(^yp)*C>e@y+X42kY%OthCZ3V^a2d9+qda&bag6D{WJvEH6pRW1SAXt@Z&;oP~|
zK%vq4g6(j7Df;Axp_JPT@Ml^PX;MlQF)Z>$I8Ib&>rM{OvyKc+o_RxnS6$^ojJWD_
z^nd2d+q4A8Z*>Xak{*`)1&rSAGSVm#$n~rKyU$2lW{DPzL}8>?R~f|uhU`|+{@$#N
z#@#LX)7w4Ul8$2CKiF1xZnUd;oHCO_UZQ!$4RENuQ1eP~Cgeq$54hd|g)TLJ4^aB%
zcl-Qg1hM_%Uv|$-0fC!^Mezf<V25&G^MC6;8x%Ce{UxTM3@RMqE-?*d32{`6rV(>6
z_hU;WH(zQ#Uf+i1uux@2o!(3R=Jz-gV-F<PB#@@`sA1B(W}syV`hywpUE#Gf$Ud|p
z{95Ng1H&s{;Z*@tq3S2!_cl^&O{3AvZ1<`7Tmyb1Ky)*-3~`ja%Sa*9---Oq4u4!1
zPYada*B3jkBc-oadKzoco%6Y4aTG>-y}kKqKO6!McxU@1@GaU8y~Z=G&A93)#v_iR
zhV2&pvd_l>HhJoo?1vW#u*i+*m;AcV#cM=BxsXQe`!N?x2A}DTl)tNJz8o-bpyH|d
z{M16o5&^Qer$qiTpz)2kWi9h+d4Gln@zEi4d9!XT8zMuuptBJ9)NXzZU*e4LMSzwA
z-Gi-Y2}8GL5h7CMyjQ!_;35^%s<tI}T>UH%iAoTQ<_1R>n0EJrJf%XSA1Ii8zFNbV
z+_<Wd{}$zl09fLD!DTE*=%wjGm<B?tD*aJeLeL`*kfP_5q3X?A{Zhiv4u8dY%t*Vd
zBh%O`M3X$U2kpo;cno=4JSOG?Sw!YFi|v_CK<8#5&Vj{B<(YAMN7I|^3_`0=i<M9N
z{0wHxg@??@CfkkNN`Lc}2YD(mz!Y_*WG!0{)@x->10PILce&PzjR3n6ofqX>EtHQ+
zSLHneI2=n?ok2%xdLm@1*jSVCP0VqJ-IvP43FgUM)+9(7YO4!mxG0T|QyHOr`Pj8T
za^qHRM$M*N-z<n9A{G;QOFN`uhB+{6q%$HtDy6N=XXQVnR+u<U7NNBt;L%9+R20LX
z%e$9V8$lm`{HyI@jQgy^=t$`Rmo1m0kccHLT#pQ!%f7X+bD&Lc=gad`dySU^ap`WZ
zd2`?Y%x#t8YomYjd8R7fnEmy%eOb8T!Of0Vt{F>*U(}|Io1@e6!qRmO&#sA?JLOdt
zcd*OTEdToDsJs2;=wY-(*`j2{V&PGS8ZFwJ0N0>@8$j3K6L@()e0u(L_w&Mtga?!{
zq45~T<IN-tkQ9=Vg?s&N`f~SSq?)Z5SufuHg#R>8uGo!y%+G*$4k5WQs0unE4T>dr
zqJ-z#8LB-soy$`amPYr9zP+TVO}L*ODQ>jYF(>S^{1><1jeKmL2!TU}zSTgP`1<oB
zs0So}W(~9F7;`ay;u58p9Vrgo>3KF{9LmFyf8*A*PhUQ_CUn8RRJh3&=k5y-%L8Yw
z=w(d0`Z@0fCYlWjNi6QA8;OyA-M+}86fg@dgh(AZ(Cp|qFfQy_HYPdpX}HVoslwM|
zwhkFtl)WFj$Dwa<7aIyU^o^dPes{Sqp!jirD60obTfSgXT42^N`wCwW&M-kE!3~qE
zt~e-CWplIn=FQJ3z9_bRsB04&o?@!M;fYH~XQaE~iH{V`t51BOba3Ls?CVc_Ab4=%
zqu9fVpPQ+dC;w1dQ2g>dLQY1=Hj1JW>J+jQ=@QCNFA(|SwE5Gjk|g)BJv7V?;9@R+
zcdj@wDK!IDKJ&<6rHt%i@B8hOa5R_i%UYnGh1^1~>c8!sO1UsVe!6+Q54UiZZt46z
zLH~N+?+6{aemD6Bp5M_sa{O-cb=7`T>0bffG7vdS3ituDV$}C<0Nqfqz%tEwyOKnV
zj2L%yv+l=A!Tc)!RupcrN9`TBw~`ir;Cfi|TR68O=H^FBZsFZM#q?PEQU0pPCCR13
zBcbKLC|5VqI%APTq8L&geBZpsaW$D-)lgX)hko9J@(GZOry-qQAS>X#3z3KL>T17I
z$ZAPhuDQ|GSXpU(VK4GIeWUp6_mcHa-k|`ut@lOySzMBqz3Kd{-!*zDypcG6aYjd;
zfB9+FUguc{%KJwl*AYSjXzE<DSK9~~!{%prcTGfXnuv-fQs(h9R_JcUOH9g3Q@r}%
zm13d^`3Hp(l|wIgL+@5j?J@P5e>~?Z>Hl+rQZf>A!)T%6uvZeDisgoiug?=C1S5aT
zqJO05E5<qHn{;&poD*7!h8FvO--K%pkyVbff0M{^eJ{y5Mr6j#UfWB`5Nx1Y4Ei<R
z8lH}ue@WFn`tAlpU2o}Psti9g0gQJhYKT+jz4p!CC3)#E$7jLjJ(=bqdtI*`dwp%R
z#q+NQ!=ty}clbj!75piM9ZAVr===vpcW)O9b3kFG#Ts{&xt5z2YN)q=H?aMa`y;x4
zsE*kI{5<tl!q$wqD(}X{Az=_{t(W)lrQXg+`yjL`KQfT@hX}8_h6qoek{Yiekb4~J
z?)_Vc;WgekY=pz?C7zfTuB#Bs33TbK;;7YMk+c!>Hujftos2&O%|C)I?{#_qP=E8<
zq`phazMuo42OR|XAf9%An=}1_YkEWBk!D5X4mJwU@a!qhd&zw)XsD#e=j9JW6pYO3
z$q5|4?|yJe^P}@6vy@x@Qo%O9@|tYeGZjypiY|9N{eol#tKwM1;8B>?$mSo~yiS08
z=hd>=*?wuiw9}K>u0@0m=S6JzAFx!HstRRpWOH<KWnpa!Wo~4bzZ^dX0yZ?4ag{R^
zmv3YsCzq`tL7;!p2hach{Pjf^hNTEL{uzIb|9yPE{P(4@uZ=((PF4k>!JPi{_e*>M
z3Was_SPDn-{`1G_&zIjus;zRoBIQVH5j*7oQFnN<HV6}6W#u3V3M34=_;+Sp08Obs
z{`u=PJo){Xg=gXlgBhu+pW~-f!!jcXCG|KXDcq$Q6UTq(g&{G^NQfLm`TbtFe}wbG
zM^Al)b1BFuQpiX{W2V|YA*w%JgIv|RbHXY$u3|#wJIatWPw>JXHd}Vo(g<bTND)7Z
z1|H{C)Ali8(n#DksOWP!#-El#;bb){*KG?<07=8PjcZ}i4e10fd2Mj{Z(y|A{GNv9
zIVOM7Ja>OL7Ans@H&4cb)?Qc(<0{{S1q7kONDx1=-%84Jd28_&-!g--0-UdA2#xCF
zR=XB;Gkc!5-~HPzNMRg|D?WSM1uwie<0?@riT?<#>#$lPi!v{eR_mU$GH5Fh{mE(+
zNcv~3c?RpA|F0(9h+!oNpbr9|a!#=-#hJ8*NdSM*`nA~%Wo^uB7r@NatrbC<iMnmv
z>UG(MaqA~6w0MXSFk}p*{|ju>dwq_<6IEeiT|{j`gfNK+(5Y`r1nrG5Wc^Wu)m^kL
z0(_SU#w!6@O(NK`u*z0Rj9}60g)NlA)i=Oozw|WzXOar|d39_|A&z^m<3e>4%B9hD
z-dleY6bU0LVWhrw{I>T3j5!KcK*5})xeh8zsGyw3*hW_H%Di(pzO+cpavGC3j3Yz+
zZ{dTfZJL9%$#V-ARkA!Ed*@l+CNR+mY&1P_um?6nF57uGX{O5FgKI+v`$+OB^mHqb
zMw$9G2S0fw&QoUE>&|VUjyKa{1^ZMhG(mrALXjn}X#7TkNg4O56*G?ateEGq6;s}R
z(280H;j|e!JD|5+H%7tN*@5b8xmZ(xHYt%htE6SOBs8Ojvj^^_UEA-AcUWe(x&>S!
zHM$jj@sr6q{z2;`$u8$_#e4FcV-_Ohi$FTu$IhNLkg6mD`hYc7F-sYgt=ro$)0cl1
zvQg<29g={_e7-Z4W$;whvPrKgcV^XMr~W2<gX`39?*%ytuC*ORxg#R&dMOomP)B>z
zZB-Ndm?Fx)e?`iq^rU|-xG&LQx#(u7*P!GYS8MY)&A_jMrdZC!V#<YO;czkFp4{Lm
z!M1APZwo{^yBU_E4aI~Ctw#5)mA-#7X(%&>?U^a3@vafAgbi&>w{Aw`9;LmFUe>y>
zO!|&sYJj#VjC$Is=QrU<86?#^9ies4OX0QM&_{$Hwrv@EEx3VJ!S(GU>VTjHW!`J*
zcv5arfvFrts(TY{1ys9H%`(pZzNo1@QppNE)!B9l><(Ic*xN3UGzk_)jvs$@C+k)4
znRoF@A-suKAQJ~Ssq)ea+7)2eWe=wWLWY<?gtLJq`J#E<&aTvAy4&qrD5ff^l-GS=
z61QOVni}VlB&#;6N<SymRM1KkR9bYYz<Z<hKIo|O#@!WXv5U-I^sa`ty)IMg)oj7_
zgbf{+U<F37^saJ380}~(S2utBrWQeu3K>yd_hsx3T-tukjrxP?4h19iDu`W#Vc)kA
z{(vY2iBl5g+Z1UANJ*$xg56eP)thz>d%X$UG%8ldF@-Dd)1D{aqsDr<_EH#LZ-s|T
z@eUiUZSp_X+Y?wLtgU^abT**YX8;P)+c?9#mF^CEruQ3NooKs}NT+`oeKgP5FL{g-
zA&d*j>qd3<wm7EZBw6!zaZ0Tvr}wJy|Chl<#<hN)Y~JmHHWgoTQk;>wTlaQ&vTfyb
zs8-!|wO$KvCyhRk5}{H;lhYotr}mLbd97Re-FO?Q-GsYwYGbQ*Ad>p_9HF?c-xwZ4
zb1raM%G(+vCu}bv`u>0Lqk|WqM;;Tzv2!Y@wjhpj4dOU8Z}PY!Ef-XM+E(UH#rra@
zE#B+cO}IiBD+JlZ%WJX00>TLuCORPOgU3D;sPNMf_`}Tmfv>z58tq6aS?01x)iy)G
zs@IjXBaZlQam0|w^PF8Q^1QuMs~G(~;t(LqJUg^`h9-aFRQG?h7mn4qvM*HL`?vcJ
z$&!cn{t0$TEggre*sztOmEhg#@J4}B&_@Y~w?E(raXSpkbZXWwyi#snrmDR(@)9F`
zP4AXBY*U^$bqY*dnsRX4>@VC1Zfhm%=)I=GmdCv{9;oKL184hqb3#UU+o@6?i2p~+
z4Jw+{>#hcNy$gS1jKXV>P(kLFBE9JlPE`nl3qEq?)}$OTrNv#!(zJYco43`ku}`;M
z%XVsaak}PTxf;n<RH}W0QwOw9fZQrS>uKE)6eum6nzX!iruXEE6Qvet%((hSnG*{p
z+f>DG=bRTrc|9LAaShLPXUkdMCvmi0({!TIM=u^*%b<V!U-j?hjVuoAy40CgyK=mX
zesOV0XSU-RDnE0Bl-KJxY*CvVC)^pYBf^5pG*ULJ?V>@FPXIe=I_VMf%K5s$qph9f
z5bo>*V>G|Ui78hq%2u!mG7bseOKpu!0^t@XnpNjEjo?;+;chFGB7^YSjC_1iL&wN?
z8*kD=f~S8`rLBccTZ?7lZB%7%9Zr3cYI4)K;(ja?ckJ79E&bY2Y29{Sd(%SMNrHrl
zAAe^nzqe<RbQL{R(K+-jdQ!?L)s1bKM+}JBO{hEyE2kCzzX~)8J|fI&NpMP?PyOCp
z#j&icDW8}6VD(kz?$l4k6bG{Fcu~tv7JVSOcAkHd>lqsZzKf7Ups0eD6OMwD{ktJl
zTM93A*Nz2lhekB3_TbXXX@7KQD|N5lI&R984yaaBnL%4>jn5o3_A`12B&8o`Cr~M~
zb|mc8>-b;t*dMJ_Z4<fC!bD51(&`S6Js&%xZ@hIW6b8CC&YfbX$ipb>gZ<+OX=8EH
zyKaBHGH%>fxFM5VZ?NooFh;8h!h^dL1qw(r^k|sfbt9Wt0nm1%AvaOl$bD(fwaP&b
zRl=`1!lqkaL*s>Mq8{2Y$YX0vQ%y9=dTa`YO}Z=bau^GpY(pz)k$5*cj%eh0N1Sbz
za;+@@Z9!EoBTshMWY(Zp&njITYGTt1Tl9aKC1I@gc8#Y3%Z-?@9O)Mtm(pDZJ9<69
zqb)wDH-j$Rl*2JM0sA9ly6sy{@Iq2~HfZ<Yt^x%jD789UKgjdg;!JmA$78yrrO>uH
z@y>jRQKe9X=EC-)=gCu7<IF*1X>8nRJZYhW85vS+KFS1k#E7kJS@Umw^QiLBF|B|0
zQYfugPW^7(Er?arFuOg+Ja+$KDFxHFdv5gYN)0!x+Iz*WbyCast_3m_+T)S%*6Cnu
zggQ8Bq~EpI^)6Sh`foO(`q@QlVt{lm$1iAXSdRIiC>OkYl#g!qd>ZW+J+XcxkG-`b
z&rx~z@vT8f?>G34-wg0x$L}4`#~Xh%`7ZzT^!EcYb}K!;@27*<b4?eyM<lBgu7Vs@
zhx9`E@D=LlR(e<+?Wq+kZjjXg*!Uy<u0|+-^R3pm<ttxzs!N6FLe&?jt6%B+&~t5W
zpxJt<v0lv3y&KP`caN#Z+4LQ-FHldn(&OqGFXk;5w4R|3E1P@&KU5vvN>6{hvzF1a
z^X09BUr18Ml3?z^KtIgxE-Y<cBy7S3-VW|Eq}!##N?#n@mC{1$2V-hCxybxbXb+b1
z8}%R;-Q@g5LRh@^W^jigz%%#vaZZ_;*5*OZyUDAZ^TBX!&Xfy8PvAUWN!IWMo}ydm
z*?#lj?cULk^6!t&m;V8&KJx(+m$4i_6ahAup$b0~m-=2gD3^&WL7f4!w<j$@sRaad
zVgiGgu^c}X12r`_mmxnnBbN>{K@OLCN;f8d)mvR}T(=Q@_pi7>UeX8iay~dDi@-oW
zsx*a-8ny$pHH1KxR<ROUE6A(B?yv6|-rcK}ch#=7D_MnYX|&=cXNEIp&XCt6NplxF
z!uQg#3{vehCzn#}a<nE<S7ng!8_KoIYTOZ1!G}@^qEY)jK9#60TJA_mA3@C0QR4Q0
zX~GjWpS5;WQhUr|$FXn|37T+jB6qAl3ZisuwJB)Yabam<>LjLNwc6pS^2pWUPK4TI
zIYLa$noJODX(9%w!X_nyRE7y4reR&5z*1WbOG>sTX6>ZNFm38ICT^x#r&O9fId_VW
zn^-zc(k9F9PiyLy*lDS4EY?iKZL*SoAU5x5q%QH`JxmFY4eDfwHTr6$bV+2GtU}B_
znJQ*bYJS32BptS*a+lIzTg7UY+hi33GWh_Dn5?uV4BkpL4FghVF;`y=su~Y2tGP2B
zjK>G&o8V&Y%xaq<#?A`YCQ@hB!vi8CglcM{beXv{QM)Y7sRnmtt@Ox6>#{X}mlz@P
zavDT}SO@Q5vLf0Z%MkORx^*RzG_iJtcn}4mw1p9)MXYq$e^oGxFGdNXw%;=*NY)fX
z1{BLpB<)JprX%B1$W27h#Mst_rWNtsMCmHe2PmSdg}1@hYLa|4_@!DKpq#pzX_&#q
z!3|ib)<Io`{xJ@!38s%Utmn>uk1T_pK1OnT)EyF2Y|io-B0ljoJzrzCri%KrL*$uY
zk78}PZoWx)pndLNk=nx;tcWxo*1}g(X*=sVL7qu2lOsWF-uA7c0$acNrhoPEeA0h=
zb~c@L{dd!gqshhV2#ksQ{a^c^`tR`r9^CI=Ob%zSkpaQML5l_P;7J^RD*UFELy`SQ
z{maWYGoSD2$=UDyx8v-Gvy<sr|FZx2#ZUh4>$kJn`TIZj{jVppx0i3i;q+bq+vU6S
z$48St`ez5Tw|_sM{d6{)TpS-9PQq`e{hQP2oBj_990T?5FAn>kXD6p8vy;jD{_x=N
z?aA4%)AQNMyOV!UE-?ImtEa%G$4{B;Pfwma!rxn3DOG~+|H4YC72j<)X8!H++4Iq~
z<KtrW@R&<^Lv&1Xq;jk>$&>u%C?6>w$78L>b+COj^_P<+AAR*yNFL)$-OgxE8j4;S
zjXDuT?cwY#8O)^2+nGt0gJa0tO3E6j&;zov5UebDJ1b#Tc7&LJTUoGv&JQ(y**s~p
zUYY#EYiAt-D_`{g_~*YID?<V9Wx)E`<>~4EO~W#Vp)3-^-i?L|0E|I^u7-B~_tVqq
z#mjT3*R=<_{`tWS3ZHd6{P>p-vmam14rUX0zJ0JaJ)2>Xy$129<i5&YBU{>Lc=Duw
zelb0KIhnogpZ~Cb*T0&4nC&m^?(>6RCs-C<Ih(z29H+E7UQFIkFE0)!@8`V7KYKPg
zIyv}m`k@7Z^MPD>A2VFw*+vToBU`|<=ki>IN3RD~=Nb>BZZ26IB^o6jB^f0hrR387
zZKD%(Bp6uFB0Rw;0#$vm@mUh$v(k1csilONY^ii=NGU&mAmphAGb3$VJ9DV<3Sc#A
z&{!XA?Tit%Gp;2J*EC-50UCa29uO)3-Q1xZhw2@M1VjHJjPKaz{@V_-zwCdUzMA$w
zbYBB`BKSSm6h%FyiWl%L>US-H`6|B9+=a{!DZaF#vL)GttHDjsX?08*82y#gQbkM1
z?5?*v_cg(P*J+t6#T4R$;z|j*jP`xITs8^d(pNBD0~(?)V&uE&;qRYVm;qMu31zT6
z7I>8Zb!=g2#DW5#PY#tf8tULi$gvHEGKPBJy1`IF^b)$aO@?OdjMh7*EB&4y{vcg$
z#a)W5a+gB4<1RHWahE*ukDNxyMk$Zd$Z_(>aZ22Oim|B1V1AUcsA`Hpu=GW8;rs$6
ztmS|SKYnUjc7Uh^L}l!p7cl}g^Qu)VstAL%sIDL?Wj|jKBm`onpuktNtsqsRbM~d4
zJ9;(!LXbK=|JWR?8O@mHyaNX_GS@)snNmP%!?6ZhfD*>3Mbe6cl}y?H#45_ADX~q(
zs-c~K4J}oyT`}%kc(*?}JDPm><=w&gz_1&^-bok>*rp$fameCkdI`hNaD!x8@p7#q
zIa;MGM3EK!(N=D**wlE{*btV-zlY*x1u01MNzE+0QJZ~j*u2dyh@mPQ4Mp^Wd2KRO
z9dSrE8VdR+w$V_D6hMkL8S3OtsBSW}rl3WC?xGqm72xeo66%_2TtBB8FL|Ky&GhPS
zc!*ylrC7EmB^K+78~kqKHl)OdT#=GGQjt1RkvdY5I#Q9kpdx+&vyf<%c$8$6bd=I4
z*>(pnz$I6Ifln#bKqOzb$R+}dNxLbsK%~T=YiPm!Mok1ga6vB)HbZOrLJ}swT>^)H
z$}G7enRN~Rf2I~n6u{9%(G)sYv>0*%!|?<5a!)#tc(bNxL$Jf>ki8wrh%%>Dg?Y6@
zrV>SvYZzDyB9nKZgdag9LKQBALc#C^!Vn?Mm4cOf(xHG6_)Ro6#V^*>yFDNZb2~y<
zQG&5_X5R|G9xrHkZ3jwt99F-XBc~yMou(KZ<3RF4jCwyxs99(?mxB&Ku27wD-=Krv
zNvhvz!h^XKNxX~gK!mrXAOasnAf%Ah0QePMMqQ%=!*-R3o-Tr;8E4sn4hVs^YSRI}
z2<B%wH|UUyUzd|8m=TgRsg)fBL{==Sa}}KOv@78oRM4!6L3U7>QYQ4FaTno#pcdK%
zcvB&vA0U76<0dayD#jX1QiMbmx@t$98rZx>{|^><ec1f?1+6u*exsq@Gl7QPBj374
zK|Y&xP+nynly1p7>J{tow<fmX0YH_|Ui?_}DacB-FT(@QCuawzvyZ=AOim9z9QZo(
z(_*yV2>R>YioZni-%t3KM~sYr9zpF_YIhJf*9htsJRQik8Y+CtP?5MCDkF{Lk$Cb*
zJb5IZJQ7dVHNo}S7|URm6^$Rb5d0&kGAJbI84-E4-Y*jay$)#YM<(X?!xHm-Q85!V
zv?#dVt!8b^BLixZAF-`F`MH~-C)|4Fg%+4Tu*r>D+)3GkYk~D3g1H8NkUW3Et-Yt<
zccWEqNrOJ;-4ExZOYeS=Z_GnNz5Mv8eF}MO8<kdGgL?_Qd-V=t?NIa!CEx4CLciw~
zs1B&q@A*rDl@`D8W0(P@wyVHOtFS8dcCe~TFH-m$e{;B$5kh5zP#Hm2#@=0Qdsrn^
z2Q-Hepfsa>uUHEC5n^3`ZR%3YiP|e1g{=y_n|A<ba}|Ie0QI4<L{j;=Hm%8Fi!l-U
z2bKcXC?!?dnGR@dgJM@Wqp8*5kQ*(<Dx$)0$FNkKL+4Dj>`sR)A?ej#5c*B2t49qv
zC4{;+xXM8ESZs%WNV&Bz416<xr`SN}>S}#su=g4>%1aV-*=UD<>&|pAR9YbUVk-cx
z!mBMlVjvSKBM?e5H2FDR?;|w+-ySjA>Pf5F?e0Boc-LzIHF<=P#c`s);Jc7UAW9<;
zr4fkI2t;WFqGY2q&YZ4h9uF!l?hLS;29C#2(AWwPOIPn>kfBA9BVuScDW+BocNCz`
zJ<vRkDwqY9)pHML*j-gXi;$s~k*l)1B+)8<*$1!y`HJ8_(WzlxL%#ez8<fqAtL3D-
zIR9DbfD9FE_d*BfUc9;!#6_wnDmWlE1sqg%r~@dm|6di>)Cjp|zqXS#N&vDzPf~F&
zq==0fl=u($A88|(u^c}X12#A@myt>?6qnxd0VbDYK0%rT$GFqBKS9X`1Ol`b=9jS?
zKNJHtIX9OfrXUlS_eDX&0>_`XyhcI00)H&~ObSWwRSbCgcx`m{{lx&2Ky1J0FW<JM
zAioFY`I?`@;w-+`rBwupoK<|~XjSb!v#NsRk*fE~<Q3@&=r>g~bICkKpU&^3xmXI)
zOle8mmU(g*m4z-A**gDWlrZ~kSv3r~ZcF>5Vu7%*_||S&)xV21Ql5w%4A|RYpHd+0
zZnA&(-Lg75;VY_bbBe5CbxN(2mQHkT3TTjAJ!uq}`ZXL4*i5m6W}%dmAvpjvmzYx$
z44l-Ie(4j|u6a~I+S;0h<oAcT)&#E%u2S6RMH5y~*7G-Gd&(5qRdtJ&&JG3Y7Ox0i
z1i~?F#TcNNbgW9bD3<7v%DLG1T*3favtoZU6dWbLuvqN?7TV(9L;zvbhOVn8j_H3u
zXVve4@Ep!t%miad2Rc-*FNv|0Ae6j(<`I}nTxyP$6jq)_3BxeAXf7%*>7euPnS<I=
ziY3z3Hl7`xcJY^~^Q2ftO)W2_1n1GE);kFkA9<edHPP#5v&82at~3+<I^CrnoBw}W
z7Vol;`(gjFu&E^92!ofk3|X|SUTS`Fvvqa>0hMnbq2x$=0o_z@vu3Y7Pgat)lxo2>
zz^ZKM<Nj&hSr56^Ex=&5O9i2M#vCJlO@iVD<!Q?TsiE%cuL>PnO|Y8{L6VKYx1G>Z
zq^Y^_giS`aSy<{GjX>`>h1S47=OTYuy`*(X&QTTn)h}2fm%Pmya(x!K%q{a;otOj*
znd9c5+vS?uWgaPTnYEvDapJAZ&9x2!usDy_wkLS$8cTEF?z$=k0n#JJRKrhhzu-DA
zoP>vpbm<&8ndt;w*1RJck?MglZdJrZ-k3e3c*LcTAGe5~swi`Aanx;p=&gU2?93^^
z%XRnCMrb#;q3rGMHBz|O7PZKlngZd}*Mqqpmd7fC==8g9%w5TpA<*^*IMa`dRt2#S
zQnl%B*XBUrZpU8Kl)<TwmX#W-15ZbF)n?3Fo$A932~^J}lwQ+7s}kvL-_>)58n4+y
zI;xDbLHz9o87bUtk<j6i9PfWgO{(jA`KFS6t5`u$f-`oz3kzCM`$@(&C#fp7@+e!S
z4#Db^y)Hq47TO!y63np4&XT3TwR3NR8?2nrYwLvTcDfxDi6A9yXY|-(=)D?4=URTv
zM5l&kR;+YiTO-HTfz4gqC@9Zle7!WQX-eNFHmv3bUuUixFAHd`1(JW+*ku8wL;$?)
zuE3E7YPwIb-$T_pBo?}YxUKB_&qmAf09%dqO%FBfbIY0$dPM8%6Afu)yg)PyWj?YY
zttl9dT%1Dg^<Gr#+r*t&X>ysNRmK{ri$+-2Xq%4$wpL-yxwh+#D4@>kKxJLct^(Hj
zw3m}Lz~=_zm0g}TP6U6ixDFq8`Q=nQ;|E%ENEJM|g+T$TiuB_<#^@C}s9PxDy~7^r
z?q<|c7goVg&e0qDi-&Z^G0t&(r3W<Ju6D*DUJ?Ml@)y?iwC{pXly&k~B4oZ(S-;Qs
zW|BuryH8h@MmfK#v@0s8N`^<2_5;@dm8O99cCh~HB|msznBRYJejWJ>u;6p&f|(+0
zZ$0ogzH?=-R|6SW53THE5uw2>lXC7$#M+(Pbal&(W6{<KD~B=mWqI3<RR4@NjOBW_
zw-eos_A`vda+7m27wWh-Rs5~IX62W|1Kk?xX;YeI+JFWchf_kkrGs+RwSPoI)!2j;
z3Cj&xknPug(@}q0%KTxWt90}dbWS&UzxUSd3}?REUjryBTkY72Do<a#v(%=G;cza`
zTfN0GtWypzw(J4R8qHl7Jf&E>dZ(>7SJyM+NUoePw47Ya^w_Vi7`F5u%+QVKFF44q
zGjX1`+{Rs@h<)EYz1bhxnxVH>%`=>GCf?sWi4I3^*N}g|c9iOBs`txAP2c=i_L?|j
zHU2V;0uBT{`vgb<qlI?rdx4a@fa9&bhf&6_0w@4MNU;ABKmlch3i>|}P#B*F01A8K
zKLSt~R}Td!K?_i7Z8No%)KKX|Ni9PXtMN8S;vF)k4j__vy>oo20+P%vFMz022S|!U
zdne<0MhAaN_Ox&Xlvn|`%BT&QRRVh{|0gNgpDCXQ?tIt(HvLVTHUq1C__B38w{iv`
z3wY_y`qOOy)%Sc-x(;3N+n!2Yv+M9Yq)AZRI^LqQ?%!z!U2Pv|3UsR&PX(mxh69q&
z$YpuhU9=?2_5bj-@0d;r&p7--%wbVzg*QMB{!Iqe^B-*8dzUeMMic@wH<!^{HWLCd
zIhR44CnuLeP(ho2`0|f`KmTzjGsB<GUM_O0lyK4y#-d=JMs5M5L|&bEMdIHSe)GzR
z`QC=HxWcEigDpK6GQxlwDF-<sl|b#7WFnpJ)oG%REE}=h=+pebO&AND1^MF~o{#6p
zpsMh*47tQ`6yltOh$O;ub%kKd2G$Y<SnLEloP#0222=ZgN7DRFF406Y@iwl@MvmqY
z-UJJUucz+r%k$lTcaxp8Aj5W2Qdo(@KR@om0|XMr`hl(J0f(Re4&Uzn?5SoK=)a>6
zN83qbh+3~Og2?e|MtC=x?q#}UdmsCG(PWEL&yV*;#-}RV-lr#+ZJuVRhy0{cabPDd
zLHT$6)+wuh8(F~e@$>u8jeLD78rQ5-WuVgmd=6I=TIB`Jrm@|84VkIz?wDO14@>qh
z!Mc)KvJ;S3Y;`JY)U??PRrJ1IJ*@|ZS-4=+>HWg7D|0W$&G>FK-e!7GMhkGNmu0_E
zw__Bn2tz&fsQpd8G@DN=n7#SRoM9yoni<uY2#dvku(k{-&_`$_-mhI`DU_Xg%5qR`
za*l+;;7pUDNMQ^^eHi^DSm{y4Z0xqVb5>ZS8*oS9g!F#V&P$#gXw9O@aom>r6pb4e
zmi)S*N!{aJ9Tvts<Wv|Il$qtxXs0IS&8V3fHCl1B==?Y(j}mNYeavB8lwXDODa=|5
z^f=3Zj)t^#8Y#S#^;N~gp_V<$dCW6hS_N&9g_^l~fyZ;jnW>P08|lPQI8+7xxanFX
zjR3A(ol$8*1(DF_bT4t!yNZVxC#;#gD^rVps*+GDd3{CzqbLkc?jb9`%KUj2b$L6%
z!0o(Uki(2e0-?2Es|lH}3pF8eux8ZEAB~cKx`)Z}gK^_L*9Agl%uGEx)G{?1ZG`#B
zxmgQqStjR9I_@1b&&uJJDeS0Vl@*e0Q=L}ixLPqt;LB;!2|Ub3zdo&JJq?24|IAVq
zAdWHwjb>mRxKR5OLJRm;GA08^H)CSWskCNXsb@~rIQmY7VpAT*_ZPEL<6KZXPISwE
zI)}L49irpu+yE>CHBep{$HkaCGzJw#2ROm<qQsTn21#%c>nSK2opkC|#()@m96DW0
zJFE}kSnD12E|g81L4zjXq2%-}xzUsGXo}o)wcH(rCAnXia+M@O*e|dpdKSAlj-3~b
z&Vj0YQx*F>(&UTz8+(g=Pt88`6=5rXgqGv`vSD37nZ>T=db>6R%9g`vsqMa|=T0s9
z>mampAe2%akE3Ux@@sWBdq$oy<@I~ivW6li`AQzWP${cf4K?zaVgsGZ{9U##v|7ql
zAF#T*>7oNm1)G_gF^%(^{mqgWmVFCWj&@x2GzHI-!g#4tO<a7kQ5VNzgJ`&aKI#UD
zX9N;CF$KZT{6B9sg|5d|<?5Zcnl1ph<?WJ0@`_H9wIvyIw6ea_?vwT$dy?L1_epyj
zMP1n%*r(s6?WmpUsM(I9Jo}KJoL;eiqm2^Q+M8hY3JgWAQx~49AT)S+Udt}7a2>=e
zE96~`JbIYz4eSDc3r*iGP`Gw~MsQcz+u#-Z#Fwldc2N9$!J?#P<BaN>obE#33xX{5
ztF&&ry_Ld^0Xp4NSm_Kjmk~|yyWRZd+s)omAJT4iOuQkA(8>z0=oKhJOD7DN4?+?3
zsRd@6lqi<BMiH&7?}Q?h<qRX;JD~_=o$%~}l|x*w-y5L_ZLU6s;*b}Ah}4U*a1Iq>
zgkP63CW=J08Tnd{foIB6#t6^(q+3Pj4U3p=-BK$#r{uKo7J5q4vE3??y5eXKw6LUh
z=%hc)>*X`AKyeB&`t4c?RJ5lADzky&T#e!qJsjVQcc)y87Q8QKx;(QKCe<H1<mdf)
z%gU*pCt*>|WWRJ`&e=wPnf|T1f6ZJRCLNCb_!`H%ifejz`GjhuvxY(^j&6@a(XvL2
zn`z|jP-yX5&B<5c8^KVptOtqxDp$Zza9#kYp27+YRa`q%b=v+LhW@HBw71lUz|f9~
zH-w?!`C)ged>sr0E>&bcYBMYRRc)&LIfAL(+boslP<(dJ|E5}h!79wD8OEl+K2!_8
z6l8+*+{Vs8CUO1paC49e$hczqvtRA8R5UiCz?bQk@4uuBXXN!LQU`E7;}g!Sh)HqB
zW*qcysGOr_F`4i6;{JC!A}Rp@<-bPl&a{qO7@)a>n?h%-iPb*RwbFw@j(*!yzA=Ud
zuIZd6`c;-SvI|RpgLlF?-BaxRV5UFrfaSCcPUC^7e!c!$%5SB=c9mP`uhU2$N0B*_
ziOtA!GqT0=82?~|r0F!JczLItm-HD97;^h50B2C3OER-kzIvFs^IyJkHt3_uhB&XE
z(ciADziD-5f7S}GD9-&imz@6rWkpX33T19&b98cLVQmVRyc|CX0XLUXatafdZ@@Px
zmoSqE1eX|CL6(1)bRGYG|Kp&uz@N@h&q|A^f{u;zq*$h2S^%vr&rUpp_38h1v`^;%
zYdSmYR4}$5CCN<GjX<YPervkl_{w+4&(bWcBgO80$}=Y^OqPK7eL5eLS<=6yDRhEL
zfeG__+)n)zr>w-h<Jg+Ynjn;TIVS^yPcDo{Yxr(M9@l>ja#-xz*m<qlRdmVn>Gmdm
zeR)6pdl>9*lyz)>1ZB}Q|NK0p2Lx6*9~<^g1ZsYLYrY@;l%r<X%6p5=)=%0wYin;i
zWi6|StbgyNZL_f(PZAH@cFw924_^NK`q7A%uWy64k(5I*KYUJCg;uTDBwFb-0@`lO
zs^x1z^hAG0)J6d#xtb%bHv8J4kAhek^R`0N3b?7Z)t7EnCiV5}#rmW13cR_m{>0E>
zm1_Na$QG#Xvw?FeS|e?c^}azy4+^!w!4MJl$rg-?Ug)VgDjLW-=BQKf{F3~3d~!Mk
z5&M##_OU0VCs=%Zv2Pg`ytafqaGv;fct0v_qws%IALTE{mI<H&N^HUW+`%dB#TrSO
ziEOvF+cj>7Xq<kV(fVgQ&V;$*v4zZ-W1ZzFU^uGvI@@ELy-I9pJ#uVwBI7g%-!_jx
zpDRgZkL26VljRMUQNSam{(MJ#9uh{Ihj0|>3QGDJTUoFlub?SN8M-)`XE~MLDU!zJ
z-a~&h9ltY*QPnK-<Fs};r+1D|@Z5W@JbB5zW1eZSUh;B}Ftj0HLK9z{_DqMeN$c0Q
z#q3WeC?~xtvr4U9VH8`KAx1S2$<Z^eC&=;~UjfGvg?+HfTud6@%AUrpQUzd*jSI3@
zYwbc8Y>7j~dnV)9R;?RI={!8FOU}FfyNrKnSYA16dLu=)7)TB5V(i?YPN>|(h=mKW
zT+FMqGS&sxrU{K5%x0ggK&?31MkYpMthG}2vD{6vykx4DK=I1`pD@*h#j(XyYb74U
zRIRqkGd+F`Q?-CfkMOx^xCO5IdIeWySbCBYd@x&s7*q&G&SG-R?j4si**@M%tVw^I
zSndN@8*-d?6N_BKtusdzBi+O*$Oo>V+-0m{E86af(f)aM0q>0NqRQ^2pe)Hb`#$p<
z@`4wv#dsD?5!+B$X11=Mx2-Xbv1H{4cMR<0KpCVPB0CW(9_2Sh_O`3f7gjWHHKKDt
zqyq`z9~9AvEe8}I6wwKcq5(_G23vpJb;uQ5QvIs9-t9pMIqtLzA!ZpPjuVUqFJ`<Z
z@)^%$h=ww*D@8h^4&_r7aNH=w(70@>%aZv*W~oS-ltSwS)G*>f2v4u;%Y=NpTUmFy
z7Wc-YLgysz!KrB6I_`Ji+*pp8WIhPz=YDsh)T-?cRFB(bWL<|Sw2DXC4XJ;5w(@vL
z5nJ3z>9;UtLHVi9-F`!*@5u6$!*Huxy{EREjTvovDqR_MJrgskOz^VzLLF>UF0^GI
zKAj&7RbBCd9&zbY_DaP<mg-9{c4o@^<6TDTe&=kx<})t8rNHxGUe2D0aG)GzfSJEw
z3#NUC#;tv{tT#|b<Ah->s<(f0u*^O(CB?E?HEc;iAvaRq*UBZMFmXp3hFttOYi^)n
ziP3N{{Y~|?9cc4bEp0j6vr#H}A4I$`(Q6+bM7)r>r5~-1%0`<>$dy2P5!p(c(g_Xq
z+%sfLmffPq4IAS<6_YZ|Z;d14*F=<WqZ`~k*&rVQ80QYpu8`SVt^<E?huv->srd*+
z+WNY~BBQNhRyV=f?Hdn9M=oS12)^*Z$Y7#-6K%^e<_+?t6$W<Ij_Cl!j%dZp?#&Y8
zh)J)#hh6g#)Qxt{2Tx*@6h3a+&_VV(x_Ik`PM>*4`Jg8*+gm7Udrl)tS=tXRK98eP
zuKqVE;#{0<Q<>gisWE>lAA%I*{-<0sB&niRZR+)_MkENz5YEX?*>E-v1!=_^7fyvZ
zX@i|WWPuGsL`LmU&BUXz&*Lh^Ur-hgiBVXrKKof^&J_r0$5~C{{5a90UZ%?ghiznx
z?47<cA~e|^G_}Xhnr>{5725id_Ar4LB?m;I(RR4)&i2QA)9HT=oRNY+Gq$@I@di&V
zVY;4p1FsEM@^>+B$inKp&2a$}XU-D<gaQwu;(%@1d<+v$_c*|5-Da~H<60grX5PWW
zMGuqZqcLPc_d9*o*vn3Nk(ZB0u!Gj{5y`RtV&RJK$a><st<FvKm{&bbymiHKSI=Li
z-|LUbc7B+2BddRkhINAdqzWdi03>AX_}d$-7y>L3Fu0k$K*e65>_p*&wLm2kPWK#F
zvyJ7Y<DE3z#=#lKFsF^^1jV9eveAUk!?|OZO-qY-KJS%Y>&`37TGxKuLs+{RXZ$*h
zHpjXGtQ<P-*$7_c<(_LfUS#zoTVL&XUVZ^Zr#mKdF5C<*JIj=JPF4Q_hgE2oVLB=k
z0x>g}fn_-qm)eChDSypbOK;;g5Wf3Y=npKzA!qmi0t0rs4$xbJ+=`xx-97E0`~MFe
zK1Gp|V>?b$Ac}2@8qV{ZkCdE#OmceO9bb13CykR8j=-wv_4@>TL-K!+JZNaL0K#aP
zUVl!1K1!(!Uq`-BO0D^_{N3><?Z0SnJOg+J@F>9ki$&lO&3_}B=P2S&$(Q2qHjkb1
z^M;>y5-~_%*}xOwz}IpA*X!>)nH7FGIXI`Su_#!lqHS0WSsL+bcuXskPQTR?e?q>;
z8W3xTKFvpy#sHF*aSqSN(>@p!zSf~|+oY+KAB*?GQio>lF^yq<j%1z-QDl5H468U4
zY~G5DR$54>I)7Uar{F=bAx3^rI>>a%!eC8NH_4CDiGCH{1lNX-=kC|n*WEw6>R?`!
zjGv4)!dg51{c{&?AdoQDPf7_7IQ{xIec%1jOU<$CKV$O6a7tr{7U5Jz5V^@9n@U*1
zd7FQG=kkr5R04Uq`tR3|sT+BDuJ$WVFzr>Z5&RaeMSqG)!ETa51{sQy)Wyh4NMkS8
z>{jef->(r$jGj{BqFXD6QH~Dr82$G(6(j#idD?$5GG=q*G5Him)noSUu_+B#??fqA
zdr(U+&<{GnY4Biu0vtQd4E`t(@^rlpT5|RdD!}^kngruqKRX~FQc9|jc+ojHSDE(7
zY2|lJI)6%zFUf1<TJ_bSZXj|gL1~0@zPf|<5jBhKSoO4zSe%ldalyvGl`?T*L>C}~
z*_McvQ1GwJ3KpQxn6!?+3cLU`gKp+5k|{28C!v3;SuYffW31yq^5rz?6yRa>WxF>h
zZG{Sn*XdT%D1rCez_FeslKbNzd=_>d%DLX18GjrQS|>T5Yd?nUsRUBzu*MW~X{F{)
zOt>0e8iFk|j-0ZQt+BR9tWWrozEysc#LcQHV66^IyxIy4FUz@qZQ2myf>CNnn>E{m
zoW6j_qwso0MK{}pxww;mN<&@tnY$$>SVnsWm`<OTv_x<TD`)Z&A`&QA>0|>4HKc}2
z*MBQEiqMd-dhkjm2<#s1*cwg7nb^yuO17@vIcIP2zvY%Cq3!C?p3-T|(C2!ZM`W0G
zxKE;af%gV6l<5XFXkUV)mxXc`1wA)Lw4G56pVM{^6d-=Cs@n-1F;XMkg}`M>Lo$(@
zm=lYUmrU~3$xb18q*w)C6A#b!*N$qvt$#pW&7g3>%gfd1dU6HM2vEJC>ysigsg@4B
zZO$fhB*#fagEPw(pm*FY!~>ldq}GW?w0^JD<6J6jU!j`Ji1nMvRm<H49-9iyO*fH|
ziotmWcwF;womDzfqCb7NN~4Q8Sg_sK{aBf2mlc=F|HDubEeP`V9jJ%|LRtM;sDCKO
zRuvOIhdH)X?748>{GMk?(h1M~Vrkz;0VD!%t3m&d0Ft-0LN(m-V*#>T?RJ3d1l^V&
z8$fx*#@!g<nTK`SoY{6N>xG5E5uXh3nv{?@c=fITuN`vWXtsEtKs5vVRj@lm^*ksi
z;#%dkJJkx>VOML)=E@bB>+|t)l7Czq)7)TY>L2IZdCl+{j!v>xO->i7cxW5Qd$d|<
zRuOkWAEA06>sy{}J+jffhNyFefLAr@wSC`jql1#S7`k6c>CS_y*qgs^phLaquG{En
z%@u^b7IoX0;(cgp%FX1YHd41FGOpQ<)1l4rSS2(@;sYs!M#pjS@1PJ`S%2OH(kD@f
z+Y%y)T+fN^Gjtx5zO%Qb6tZcmm3AgPyp>rf&!Mz*A7)`J&taMZwUaDHUM@OcS3Wey
zk1ojm482=b9&-udd>%EQP7i1P1F&6B2`mXTgB}3oTG!rjaVaxs|MZTdTxA#-UFH00
z0G}Rs++qjS0MFf^nys{)Nq?PYx0AOE&!CimVbKdv^x?!=e9pnB_-4hk@TM^)HuguM
zw^cL*z-1WeD%z>=#^-!N->HBD8`S>4F5u;=`E}ad?b$gDbL<#qs%sfkh*aM~YV)P^
zVz^Q&*JM|v;R*K-O_|7%fisY-=?2~jt<)V}DFGRw*%zI}4G;m&aDVc;pRI}&t}Xa%
zaAA@8-ORkF*tX7boQ%GwbxVnFS+|_6PIH?N*DV%V!<ykeEm}%^%cAA%h$EQMDc6!N
z@A@vV#!x)0yQ4KriJdj?c_m#vuv}rgbT2z1U@D|7_ELL)Wt0O#fy{(2X}KT4s{t0(
zdIPWaOf&tby|2h`pdhgonIU-lR3nPK;dO@AN{}{7lP^82>u4@F-G2cYYh#x|4JH!=
zF*7$cmr+dy69PCimqENWD3`6dEed~zW)*i}AYi0%2gohP+=83}+b195|39S2=H83C
zXQViX9!YGnHrDc396o#>`0(55@^<<(3ON{N6&M`ez70SxIQ<1qgIWy+fN254+mGST
zXU<X6bD>9aE(AS>zFYcK=O+P{W&q6q8U;8%83{BZXhhH)RP@Q|LHcgu*cgA;Yg%u)
zl%4|72AT*j^jyz>z5RaTQDDxT_Fw%+|LgPH>7P>;9|cp|4O$rn9EQJtocs$3ifQ8p
zMCL#aKfey&PJi@N6R-Z0aU#K#|3lw?ejmD#FVUcmlR>%GN;L*W&6Ki()`AIRl0hXY
zE6Fg5jHJjYO2x78yHozQ0GWTAmdr}ia|;}<^pJU_ZK!`5EiB4yH3RGk`8^Dx8c!&G
zx6^O$7yh@x>9?{;ooStPlTVNUkYw_&KTqzY`6L&UUaj;eXxofb+M=}#QTIXuITyS(
zTBa2wB6HTw)Y>ijhTOMzY0XE3A!*1jdmr)Kz)Q3h40#_{0y;$m8is!$ipFO!eizgk
z@p(z}^Ap<q#`$03dD6UCA&pA!Taf2?=eUH8rV$xw5U0SgxK8HK%n$T*TmuUca*gAy
z1ai^^o*Ozc@-xLv`RLy||MsP<3EXUqkjO#4ENk>W2|GK^C<e=VEqF89L~f&TErFT1
z290PIOg=gD0ZM~NxhQ{Sg`~6RJbFH7kbXlbhc7Kx&M<*0yHHvh6N)Eo(FZ9#RwywA
z!EE$PPC9dEm7EzNdQK=N1FVIp#AM$1y>-$y-7SX+f~C4Q;~XfUuW=?iOb$w0Hay5+
zQzguAU-F2K@~Wi`&Gx2TVacjwC!w^}^r;=k$%>0_wk(<U%s7809A<rVwe4!MU{?@&
zB%qX|jOFsK^GfGStO#QIjg{kP`?ylr{1`y=oaC}&s0JZs(oi<bd$+zSFhRGqT^Asj
zM%(H5GU~MBN!!3Kr-~_&<y||7O%wvN+};P9XhU--twSz4%}Bx~D>Xi16NQ#>`-pJM
zoSd8t-E!iW7%qP~?@jqxbLeVK;4{yKhY^ED3LV$W;XQ477Ntj{nAA9v{<`D<JL|D@
zz)D==uFN83^2;j6+%5MKm@?!%tzh>7e8*C;RTLk(6J#rg#%=pFk}q$WoM4u-x%~O_
zhq@H!3)IG{o*j&9;Zv;?eOz1baW}&HK4B~s5PZb{AdG)IPW;m11xW@mY9PZhNhvMK
zGgi>bCdgJ?EJGq$!>5oC7OA6SQmMKwg%jrWa8dxJnGmk0xRMO$pv#mKGybn7Koo9E
zPFO>d1Kd&amf|hRJN{4s$+R~wRsQjN=1N&!(~9td66T&-wG?k@)$y;j5@k<nps_<M
z;V8C@y`z6t9mN}3_53?86Ik}F*C1vELFrV3+f}q4fk1_9;$#N`UCrYWkqS>)mb~K5
z%BI8K7@w-0B0pC+a7ETcnR{E|FT%R{coCCxCby{6Ca^_H)P^>_Rj2~C-e}=`nLvlG
z!Xi4=djT$LOI~u9=G%^{hi0ltscCY^h!KMFQXzjs_8X4_OD?Edn_e1tU2R37@32bU
zulPWQL6kJ&W-_dRS^~3L%O(D;ElTC~_Np??A=x5?K42e9XwASq?L)cL2)WZfsa4qX
z$LK>b6E}eq>xrm*E+8)Z!NURys@Pi8-S`ls5w$8?P72c5uDujTZpxiz`iNOi3z^<b
z$NYcBR+Tok(PmO@Wn-V;fs0=2WYv&qucd`PVfJzy34GJJecV3(JL2SKoJu3HI(Lkv
zp(mYrE;v=1gZfEjnYCu7-ww~h?Y-I3HT#m!o4hmu7ku@B{V3!%;Hv#-;Bff?I)`a1
z`|<;tIia`9&yI2f1g8pi{>9a%6=p}VX%~Ooq1JQ~Y?qoj_Xk#*{kDmP=6<$`b!L}s
zQkiKdV>+uxUwU`d$;~uj4(mY#>BO?bl0N*?br-7Gzo<vH?!Fu)!<lOwzpwORZfEhy
zkMACZF#|Z@Khk?Q&iu!Gpxy(WnM~yk8}%MR_SZ?Eo~v%pR0Y$#9VlR0zf|8V8ZLj&
zn{U-x4e)^$az?9NwA6Y#&@l_MRZA#iW{Yi<VHjFmwe^jQy*5H`pbW*pC*Mwkc=fu5
zGA3<mlu<0c3uQ<K%Dr2^k0TALP!!rw|I0&loxODjXnk#KSMQa(G@Q4+iqiqvs-DXd
z8j3e`PZCAlS8DJ4>fWW9jW&0KNeh3u*bmJ!MM1k2#Ta|5^yk<j-5p#fuZlOB>9QE|
z<$OlG(mWIa0yyR>{P_R`XsPljGwVN)P)>aq06Iz=PW(>*U@OcH0N5_L0|4wK*am=e
z?hgcj`)v~e;Qee90pKp%Bmme>-kt6i554ndu!Ydzw7v(OJEwd9i6uC8%2YI|-+4K1
z7OF)42-XU*wJ~&1ma>=nBO7CY(FwGdVEfj$oxHn#&g^xo_uhR{X8RYdsAsB|u^c}X
z0ya07(Y7=smria$4wr^+3nqWvTU~EkHxhl%uOP55-G}g!^T8n(MS*^7x&>TpkhH+=
z21emHYU{3R8M0iY|9;Q#+E;R<c2!Ghg*I(Gx}HmNhBI@{kZR_l98$>mDfpYaq=AZd
zsRU`YOBRf!OEuI|y3|5z&8317X#91--%|QJK87ZH0#Wc(wa+7XS6hFdNT5)-k5w{6
z>w{IO)kIyy?^^DnVT4xt6o&Ps)?c#{h^W727D$TvY*qtV%6$Y1)Y$rLP-xblN)4QH
zUruF++Lu!W;@!9{aOOTP9xxSOgJDt3Hm=M9i;B-yF__jqMa641_rY2U+@ufIa*$-*
zC5?h`8WlK*()d(@bRU00tU+q+X2P}7i2K9|Vz<rB(v-&66r#F8hB-C6iy+p3VY`Tl
zl&y480{q{)h}X<r53vNSd{nSd(w#yLg=Czqg~EO526q+hKC_$y_M^X+HI#OzPy$xh
zy(0~x_7ExvoHWJ|qQbI_>&4StyKQ*Gh>Fif#e=M`fpfqbyO@75m{Na@Qn1oD%kaQz
z?FvMpX6gpPU6sqIE1_z$E+S;>okD|H`o38TwWYpq1y9=;o(PcI#_%)_E%op;P7rOJ
zO#?!*#{+C>G<p)NZrm+4rqHB^oFSi@mgcr0JDT;VHpU2zyU%>|lzUj4zmjNNvHv#d
zxMUB*<Z0*AU}1l$*1jgssnn?Nh%X|w(O<qXIVJL05%ZAmmRMb`<2ZOLXYP*jku#M(
zZ^8KZ(Pj5XmYxLWQI^%Z=M_dEK}Xs3$NVxcAcQ1-^5KWWS3h6RhL5kV=F8#b?dzp~
zI=Q&|Zx}wAFW$@+uM%obIvxHod@+3bDvuvt%+8muvPge~FGW)n1Tk@riY*bXQCm43
zJ`NcD<?#9ZtNHL*_#h(yl{HDhvXO;MCV%}1{~Tly%+YP5BGC(!2uH-+#v-VyAUX(Z
z!BVKU(zdK&`?T<J$oJC{Mfb5}jy(>PVnggoZqZsjee%GJ;p1ue^X0|-3Jd!B#TWkH
z2j4H3*EfHE8;0*LmfzpLj_30qhaYZ#yngy-_S5j?+4B2;UN676TFw^Vo}JI)50}I1
z%lYf!ryME8JlrhKhp(3xmlw;6+0F3d*>Z95_M7>)-<;2{ZkCI)i>u`g-ucUkJw2%g
zJD)t>P--bE(&gSz7#$Pykr|5QdwPOw*=Fl+4Xl6WVGJNs*oo*Ci-;YdnM5S@MqrsL
z@*Anf0t7-`F1&wm3sQ<4D$Eegz@yq?11{xiv9N3`LMkeGW5Le^fNdq0y`no8!u->o
z=Tf6tKDIrlj6J8eUwjwjaW0do1!<h^*euQB(Op-Te)h@{kzm}5;a~s$kAqKi#^;cG
zb$fq#dHPFg2xd?|NhvzI?V!rF#%eo_imDnbSi=sgxd1c3Tvw`L_;h|bU%b4Aqk_}j
z7h!mQwuI}h0*xR4{&xBK%jMZ}7V`Mu+5BpWNuD{YO+X@_<yplZlkoTf3=j0S`iGA`
z8lEra=Pzf=SHttqo(*5k-Y!qq;^z6;cQb#i0h<N4ccR#5LA2b=Zx`pYn>)_w&%T_!
zxj6e|{<eo#ady<`6eBF~Zhuv~S`WpWJCfX&@Z{A<lYRLkOTL%NB-teOb)W84Nx(8A
z7>*)?Gv;7~xjN;v8i9hFSe=_GIKS2`){yh9?qAIqrL)0vAh*$XhLLklqP3(5LiB&$
ztC_OI&cW|;Tm_J(bZkk^gd`W)ubiRbiJ-HSt-j%+M_14U%miF@KU=|PQN>oac1p-c
zXKP8Y)#RPxUDH$)QC+vy?KO@2vgz&b4r<_Qu{@}*zo?2Vi&}P?98aSP9d*rawpp{A
zKB{JWBe<&Bb*jp>PExy<TAHLhNi=^+JV~;DMMql~Q7d{?tA5O);E3u~4LeC&D!Mcx
zGVEYgM*rJ1?^R71h+<7w6{aB}(DwZiS$A`Nwz!#nGbUfpHUU#A?Er0n)QrNJgM*i!
zODI4KxD+_U&3V($Z~;q`;*gmtp))^3lgnUK=sg-?s{>_=m;g({PUKD8%te1GDMAp9
zuvD(-b9t|=CbS<Vt@VD``keSkwz@fl?1><jqWFQ@J%U(?Egy+}xjsE1IYdW$XDv9N
zg7g#NCpdX^N|-mJNJ)jY7^s)D3hK(TR}>vjhqdWr0uNz+2bx>isjqsVcHb=ZL#Z7!
zJ+=J&Xa`jf(wyX&ao4hJ;I4lk8b1nmb6w%CACHgN>8IoOl1@^YB%7o<N%#2d=i>KL
z-XDR5lQ&O5+!}onr5r1GD7vP|X)~j069EddSwUi@mhyfHB|!R?R|pK4igJ917S)eu
znjN5k0m>p3Z)O?{_LFChBIqMgO6nGO`!)+@t`%rli={)H$c=l>LeYPV!Tp|lf5n;K
z`<^qQ5Fs-r*2Ro@+mY=#CG<L!>LK`-QA|dG#2uY`(@xlQLa>+0zwfS7MnhFrnGQmY
zfPC+po?D4lR*6mtP+H2dZF)wcQ~uMO!~S%(xHx-#IlFO7m#}lT7+VYFrzcX<L$=`8
zy1(AYuWPItd5wO9xr=|QF+5UsQFUPOaNR+b2@BY1e!mE*X77W3tTB!UT;w;?587<|
z{SNq1^n>#n{m2uApov0Io+O&2iKdVof?#;xB|*_z3)rVrieOkJ`K{2FE2zq9SOQ7|
zx=>TGJ<m(QW9nMlr$xbQj_t~ixc5hUc7Vv|2XzPTM&UWOdt85_0$Vxld6ipRR9S-)
zz)O;&gA?(6)wXYE&(U=43bSIL_2~As5Nx&dXMxV^#r)0f`O;0jnN2nVHo@NCflb<}
z$$qHz>)5$X^%t;ni>epBym=+MO;yl_sM0Q~<hQ4FC$l_a(*v`P4`+ooaaPjfa8}wH
z&hmt$JRvDh4L^TR4L?r}KTi!mPYpk}16-JL0oW11R=~AUBauyFu!v*T&EDRUX~LgU
zG&e3OSv$b7ugM6!4U>|8xe1^I)13g+_Wt`0HBi1LD*^!i!A*~>cp_t;o@m+g!c_bS
zxUP<X#gBm7(Ns^rCq2lwniz46obkJNmb)+QfeRpMV%~oeQc`Nks9aea{%B4q?RS15
z7PHdYy)?(7d^CO*^ssV>a|{29mZ1?O#}fs~0>ShztWv@Awxxi;y)`Kdb3-4aS@yf0
z%h|vNR=JxAL@4XgI9lQ8V(*Xv$^Wm1UeoxM%kXV#YJMi)c2N~jr_^>D)xT%q{Bj3X
z<WQ`+{hoj4_rtUC3&ZVED6ef3<<%aC@`cw>UZ);fryg1-<kt!LbwYmC16-Kjl|F3<
z1RJkWQ3gkZueRmJ+{~|03;tCopbVxx;k05RU4v7JR?E?(Bpl@5JSqpQe*T43YPsJS
zbL#(x3V86AZzq%t&P1%)N|Pj+sqBd}75HXrI3p^tN<AK5i~G9ld1^t)zX`Cq4Ir@7
zR{j{)vY6f6Ufx^RR#(>l1Jv|aRF|<FKNJBsm*K206PLeuLCJr;pa1^)*9S=qH|~@c
z*2TejBanxlJB!qDA!(6<MamXeGU5EFtUb}U%VZ?tH(b%Yl;_JY?o=q5xIWWFS^@C0
zqbIJ!0knCIid&{qc}6Ohs8~xpU1{9UspYx!yD2h4Ug44v`|uKAX<xPVLMz9SJH`cm
zR(#cN#br%0+K7LMn#5Mj$hnT%(*n|4k{GXrbb-^=w4a&#D@kC>{w8jd*4E#Wo|L`c
zl%Q#wQ$hySB(P_d^xKdmS$kdxyQI(HPsB?`tBxF$4}ujE0$)`f+$puKom=uMuY{9o
zChzQ|6b_(JlL7~I7^KkF1&8l0B9TAw<Szg6`^R<JE8>5~2=K0PP-PD!vuh|;3n5Y0
z(#Q11zdgS`{4p~jHfmvD`277LjT^AS`8eo+Oqt=&e-2+BK5?SG_00TB=d7Lb?}W9u
zcM7eiH0_i<5o+}$b%^Fy-Nz)?Gb-nWY?&HQ8|kVP2A698WRxc(zmL{ZNi6_R2;&BR
zORq)ys*-=zH;&I}*86%p56a+3obdq5>#HD~GfV4Fu_zuSpY!K3CST5vPC`9-zMhcV
zIgU!0a06(_orodc5i|{2c&iSAhW=MbIoq_u2E)^_CWVwm(MEFwIUsO=$Jf~2ahTgB
zgNVk!8%}KlU5I!jgXS~dqmot084oKeiIySqXAgg4J9C-%dM1w12ua*H{72cs@nj^M
z$tai1j0##r4>L@v*`J#GD$$zJ($WfLW(eEo_vx@wLOD2yvQCB*W?HuA)R*vO`m1FW
zCI+<kQMlmmaXLvdMsY_nzJ4TAR<%bJDK)u=J;{^S;s5z1s|-utYA?94P78<O#(_{8
z?Kpo6WevzvL4UL$b+m@8^ld)tU;xahWggxV0HbAOvUWCX4J8c@;{wHJE(NhOboRn=
zbkD2k-(<75!*J9;H{*YTW?r%Jot@@QdsM<+n$d;5vz^}zd#gTLw9Zcd%&-^zf%1Ie
zji~q0xRmi~60GI}d8sGL0kmphq;LAvpe=v70M1c$j&09h%eIu~ZFdyYu^}oqY)QvO
z%USBEHJu7|G%9e$H9i`Tdd}liZL&<`8kwId9hk^M%Cew<hg3yJ;boYc^t4NFaUR=h
zg3)x;d{K}i>>xMUR%lB+CKsG-x4kS18_aB7+LX=pFacNE9fIjip6EPg!P`bu<9>fh
z`wr-|aYE}Gk<&J1c!AG4>UR*QQ#sIiG^=MC1+L0%Bp(aOy&e6&>(A?xg@-n8LP;)^
zp9s3y;ZPp6Z8T@BNJS-y%5^T`P1edI2eI1JLK%b&f%9{~_o|fFB+Z}B{DEZ$i?L!b
zx0D?|ykNG!QARZ>QnzE?5YHx_w#$Eq+W1o!cU?(GuHE<jBiGh<4l#`F#LRa%k`_=Y
zqB^lYvLRhr<>>s_8G^q2)jOSWG>-a{=}b-n=gZp3$SQ8+oPXPP92FH41Q^v@#OC@g
z#q#CYx69ycbrQBK-?%cCt>^@ub34eBEh~@a&wT|*9WoD_rcS1MnPGxGy|jPrK*h8A
zv<+Z8ST|PEFw8?8SG<&lSyF_ZQW}3hP+Bv9k+T}D`J2F`65AxiYu>C@=xX}>hH4jM
za-qEOuJm6gXlGCy2!Z!X0Se7UL{rNsm)cs6=;G~|cyzzu4`4EuHv&HC_>A|Y>{O!T
zJglVbR1o3#k{Zn%p*xa-yfc4I*8v1H&!p`+UzydSvZd4_dvWOP@Z6^_U3}3mlW4#k
z$*5L*zLCgmU`&(lth#&Sg^s`{)pVoykI-4(3zum5yf=B?P&2>?+5l%MFSk=RNb#y$
zlBM&Iq4>?&ewfkQjF(4w)NNmkVWb7#E^*89*wMmuhVa@nbhVV=jNpHoNJnU<u1=#7
z8*AZxl9wt#(e%1$cB4pe7CpD;bhm}uN#{Ir8fFybSl!9hr^0D}<=36Je8N&eAeMAC
zeX>>FK&iWJNo|CJ*kjDjSX2B^V@l*B1eJ^zz5=E%#i-3jHzHY$u`(=3Rj(;K^Hsx>
zvBG!6duA_5inwRHjIw{iM#GuuL%Ra+2wW?w$!UOCo;`B=P8;?#txHfCdZenLEeI9k
z9aIcZ?zDBBVylxZDbxH8`EJUgq!Zq8c6dEt&&_46bzQ_&wg+yFKvpcLiUOk&&Lo)`
zj6eByDW6<>3O@fBkow0VFM)XJ80_%BAq_(U$2qhgIE)58ChUJ6KK$qFm%n`e2MC<-
z-!ubbSHu&L+V}F+q5pR&VfaShkvDMB4<I7qfC7L(*_>_@+ra1H0T&WRV@NoTeoLcT
z_#kcfY>t?NQF>gU8>&2HigmQE_pKkST#ZX)nHn4ETcSd9(^s5dXQi*{)qk&#;;erS
z4_}6dPYutg@xOl*w0h~5;KW72Yv>_uLs4iSFf-5G;{z^b5sQPpm^7wPT9>0UjDn-b
zOrz0llGc$<+B$@!Dup^!-}(58^i0r#N72b-?*@x*7~fDMfgmlgj-Nk&r)ght(#P(-
zka+|hGl|D^8t_Y^l~bdYVgo&1pRNb{>8$N(H0JGUu%v(W#ch|B>#6L|7bA)+d>T3y
zv#5`wte=j;le0y403IHekFY2VL85jImKF0`CBSgLtLpSM5Z9Wsa^YTKL>6jglFXW;
z{*C6&!&oA8V<#Dv2AcmOZ{+%pvdxegGcWCdGy5ZI1=<TQ=POZgS~zYsaXk%ws^RYg
z)Ph2P@k4)7R}`;}s-3{z2UXPJQL*gkUV9{24#>A!MfA84dfyGUxV79_*xqrYt+Vyc
z+!0%m-)~MwCP0L#nuIMiiQ&|ZR%*GSWjp^+GN(jLtoPJBO4?mrb&2NI^3~Q+Qzy-?
z4vzKZEbbH?;Yl^%_F!Z^;%K7nBsI#Wj+7Zgt-62hI%TJ;GFCQNFq)l#S}>ZU1uRDM
z-}&FS6tqXm6b(I&;m}x1TMc!U)k1oGL%o|<Qd21^$X``SZ5Bo^bf{iOvO5-0g@}=a
zV#55S>M2AF-{5CePn85rNkX6P9s{sg53|s3ldX?Dx6V8LhK8)vV#Qdycg>V4vuu@n
z)J%Ud1~SHS%J=I>q-L*ud{_q~oA9pkX(@<u7S_tArQ?c2T`&8u^x1bN{YqxOt3`|e
zQkCu|j=dHd&(CqtQdujgnT-0<6>e`>KlM3kb<6sx&(X}6ua);{MT5eV0?%_{)5&+#
zmvez_wBv{BMd_}x@A<LUyOk;+q<$sM*}8vEhsPc!DHeBzV<ImLt^c|+{C6nOV4Nql
z-I4n3&8NpIFhQk3F8#Ib_I0=ZpLE$M#YpV^y-3PO6c;{gU|XB@lHCJyqfyf${2EQO
zsU=8-pRZ}{*kc5k;<#U<Y6{RwqtRnrWL;x7QY&{^H64|cQD3}Fj%vpHpW4+>({+Dn
zzu0EG)-y39Y$^>+IbDy3#<(i)=4CezN&ZL8cLeH5TPgk2ebg92+0kwE&La1eG+b2u
zR?F-erI#@c?=#nNp1E}QS7sxh#a|=u-!S!FGC%(`;iv6W7pCVBZn2F_rs5ux^>XNs
zxFB^CSv9KykxZsEEbd({P+KG2+_isstsj_57xWz2F4Gz(xAC4fV2<amt)jXM_&eMc
zrKS;p88XZUv>q^BcFU-yNr&8tjk)=ct-=sJ+6!KgJkSZ0qDSNA;rCLk+3;O#`E9F?
z1+JwVPmoT?=ofF?WEVa8`Tex=X0PEmcTy|53s*ju9jnBRLa{=iFI{(Hc{YDtu+gmU
z6xQGN@gWYM#*<!aW<+%vxlN1eL4brAZe5LP=VWV4bHI5{VWLaE*CCGdTE76vMAnN?
zI{N3je)b2tJdb`^)Sqbjmb89z$@#I1X>wf^bVsHg4x4sZH0_64?S5bjrguVVd&?hw
zviIlLhyMeFA-En2Wo~41baI!t96t>LH#wJq>OB*e6NN!p1E?~Smv4nZGDZB3_$kh%
z#Lwd1;i?aNNe(>-JqWEthrLyV7Ks*#K0@Ql@e{bWx$T_C8;o~csXzg=2_5Me?hpTc
z{`Zb&fjRRg+=fe+=7m8*e^&^vprktCo^ZA4UcnFw%4%HsF)CImD#&Z~V6<ZrVibx#
zQ`DIP?p3%yCrW|K$TgXYDUhuH{rNI<Ek8evT)kpOxSlJ03crcr%Lu_GBWe2)BFhyt
z(>BYilE&d#G!}^OKPT7W=X4DQ;pb(GD9aQ$7K<Ej*Af(*><Mise@WUfDSbae9_K<9
zN05do>q<|4x?M;(sw2@;+LBVj`2_VN<ngdKTm(}+!r;g3a$!PTcoSA+GXL!|I*Ku&
zm+`zVqvuRoHIuPKf!pm=9k7#uR@~LXO=ngW?ZOP!Ic8Tp23O27EyaY@Xj#Gyq*GQ_
zJx!VcEz%4G*N;mefAF*+Vp{Z4zVwKkbI_Udu!31X<8fQi$sy%bJOLfKpz1L<g&XMi
zxzNKeE#cx}$GQsEY$mn9>gv@g6UksB>0p1UjWf6!k1!rxWAV+;PwR%ulcOD+09}jW
zOg4L29??+CZk&^$Hn?(X-8h=UrfCi?9n2CVUY-VRn+52{f2bJRh*A>NWvb4HV7oYj
zr76!`=&G~lw1Kwt)ox@JXQ|F)CD~7=Hrvx91=G@WIzOIGaa-XJT0UgfG{*KQl1OcJ
zHg8PZ8q6IcaF@1)V4l==+U~Dcq;ZW%N~9-_akHJ5j2vx0S$OZ4%Y`w_YuPKakU+wU
z!EhdE2%V@-e_G;N@_fl;!fK|B=(z~L<p&e)Ym*_lXUfkeV=<(>E;ws$K)H?0muDHy
zuJl8ml&}Vi075$_!X)75B<vo|IBVSsOHMN5R4-OdjX6GC@S1rl>#V(Oli)2@n~r(P
z9h+H5SOdt|OP#r{a&rn3E_!ad+H;kg4<TFj+-^)$e{uC)k&qDSQ0Yv0g)&C--~s?l
z`xr})HBYRGEj=;r#hb+56cxjm25|2Ty1fdsD<dU{sE2Eu9cdyIZ_&j~b$Zt%qY5PN
zw*d0GRH#B|tfD?@1t6zu#E*`91H|s3qae?OR~L0O85O9kSJpSyRX+cUp01C#Da~w!
z-1QW1mm-NlF@Me$y-#^Hp4urXJuWS^OlMMom1Vb$^2?%df{AyC9OOb3(UHF<BBWA9
zm2IBb?lN0lRrEAs_?b#ZGZMb8ff7|tiGy(F3d81JlQG&gu4-$alwR;;%TqC5p!t~6
zd#2DJ_jcm-rxCq3{HC&)5%13SKWY_F<?!2xIif`@K!4%I!&qHZny0Ku15mO|&Eko$
z#ubTe17Ak#Oh7JgppTt_*liArGp$jiHzjJj48r0ds=9@zsA+L!RjgRv6w3rja?3X8
zE)-X0Jaf+;lqq%|lqpo1Yjs%OoF)=d36qIaLpMB$%2Us1@Laf;kj6)jwagghX+u_O
zF{A8>SAU=};eN^-T#>6Li{oq#)GSQRW>ygs<R+UBu7SsOQ=@2}qD9d>$Bxe0xsNK5
z#$D<8nOe3;e!9W0^N0n%&Jhf?X>R}t{Hs!B`p1CW^EAW+gxn~-)5G6GY(|{ZDwI@)
z$c8-%FYzyA%iT}EzkK`G*Ixh-hnK%XSCqHNpjd>`=SO<fib=>WhVS_q8B<HC2Vx#b
z0odE2Y_)Y*e*8M@Fu4#;h}z|QXp_u3u4}(oOgkmyNVsV#yd$;JS~z`g8;dj}gQ!at
zAv^^td2D#Z_%bRyhF_N`i$P6)+PAM5p;Hts<8C!@EJAMFs)k!qwrsNQc>ncB9`=Gk
zPt7%v*@yjP0;RnY#2ma9wBqIoW_Afk<nE<+$?B9>q-|nI;#7cIyQ(@OVEJWeM2f2I
zN8P?$YYt3`wc-kM^jm$S8Ki@Aptve#Siw_81CB*+#<PMh<w;K0hWrA5^B?P|0J?aa
zXmov7N853yS=j_3WGq{iozw)Y<q#@ugT>Y9!5=S-a?B~ykBVs^$yY=DcEK#L$brp!
zI`6X+m-fW2+O=9ODvU*iKT|xJT|JDLxKwqGA@XF3Pg%Kh5qO$$vtt!EY_U`+NzM~A
zE++ZcjCo&~#w6z!*-mmNDAnAXO&s<<lnC0$R{F6|gLn%HP<DGm2X%+HdZyL;PxQEv
zNHqSpnv6k90RcX@=8ZwP0R$7=2ZWcg96uBRIG2&4Llc*{sXHl`^N&FjSM03L0{Yes
z-v)CUOwXwg)Bk@MN+Bt<#Oh4D=8%=nh@ePOTpk%x$;*E(^77;9?fvPO8`8xJr<H}v
z`=1xZHzfb7BwsZ8VwDmG)#d%K%Rio_1jC;<mt2rRDSvK>-Y~tBL}3ITTk(U6`6&JJ
z&-dS+WD>YY2Pm}h7h_4V3|TssB91^jqGk>{eCwAemu7ho;#~@V>F4Sh=0>+b7rFg%
zb81m14um4?xk{FkUitg>^36z<#+BDJ*0O#<p<v$qSS&55JOiA5L;h{jx0*Q6b^4oL
z^UUJaw|_W~G(%z<4T;928S9{fnO@t?StIM3SG1Ru{O0@+tgu8q6|*aRmE7h_lGk|D
zklE%-e33?~*Tgq!s9yh8U)oQ)i8{I_J?TT%LCOdVgLT9aw{3yI35!dnP?lb%W%Sc;
zN)AR8gvLA#ii??-{yHsrFiT#I<>}fqK08T#LVp$o53XiFNC##4Yr_DvKuf=}twGg>
z3D|Q^S3a(2P`a1qfRX`>z>#v2bzM&KDMY9==j?cK?%Q%YUe5)N+l~WX3L|l#qB=MP
z=}aFk1{=vwMOu}wtjsgdOIAU~G;__g5a+PIvN=p`hZWK&tR3dL<KDak0u3y#qA3#O
zkZ!Y23>tsAHgizvKF>jAT};m}dV-chK^y*E@W&|)M85>gLFX;&b33&dC&<`->>XQC
zR*O-ftPG4U&Dm~~u~))!p6|xMY(zO{z~rh3lS|;H`|d6yaHul?J=ZQuZ#tI`=3|Bt
z_5%S*&2f65jMBR<tEz19lZj-ElU*SFs>AAZWmkVY0UtzDiRebK#iF~gX>-Ce{B691
z%ioN7mhb%WV|_idks=69vsa~QxFnP*QFee<3HTWW!b@KaLRP-?HW@+$C;<Rot2~f7
zBo&aEy#w7GKk4?Oo~zZq&YFpiL`hrW19Q5GE9BGXnufm?7t7#l4<s=7GT^#7zTS${
zfPa6{_`DVQtHtvss4#r}=zjhB{`B7=d}wQgw(9b)zn;PkQtZRlbAA$D(dF-dU;cdh
zy&pBLQU4wBd$a>-jMl9kFc-#`nlWp@Lsc3%uh*>9DcEUD6DMX>{)yJwT=`=*Dv&4?
zLz_+-^*wwVGYl5a<T8u_dK$I45^dkhr8$2?s8z`HTwp(iXYOm-1mZ>FTiwVL;)gj<
zpVEoukWc!6P6~xF!fWg)N=<b=FN5{Ms&QS^GJl7{+d&JEqT5hkq~F{6FO|}=$*-oq
z>`jdKy)Hli>|Fizz_vv85T_0G3EEbF>+%G)AS=V9=pO}IY-cLs`pIE0Y4W1g49S1l
z$=H#qE()p&H^Z0`_|_n4sOHB})2eA#&1NloBHPd~LWIdsYi^D-;YwYLl2iCbYcZy1
z9xC{=QXx9ave;3U?HPEk;`vrb70V<$fa;-(&>)0OfQuY~y<XB(){Qp1`X<BH<6#(u
z8eQe|o`8DNcl%6@QuscvLBe#C-!2zIU5#3JZ~LNeeFZy&w23?|2U-rZJ&%`@ltC#0
z(wDuIK^uS4Z>|gig_5a&%7tU5HYv7T`_5It!CW#y>J^qUbb^V~h}5iYtVnjsOD~if
zR;G$dn_g#8K=8Vj`V;ZGR*VMJ9eAC*<AUgC;dNcN3@8of62_#IH|>tSY5~JUz6g|)
z0S04V1<Hx5eX9z9I;OnjcEQc)G|*F=_@kAyoHu{Krt)mWmU@bNz-B4}dKp&0CfhZq
z>VdzR!B=$TRAvLU1x)VgU}fVWsI*P`{CeTcEEqnJJE2I{nDEX&P?4d0wHgj8k`po)
zPv+r@ya%-tqHqAUEZUX!^EX875ew281c?U`B!ChC*vnHn*?Q&=8N&6HRjYNAZ$S56
zE%JZW+k81tSz|W%X6>zDkg8HhPfoOjE~g!v62^Wj7c~bLImk3ycuFT9vbd?I6OC>=
zWs?y|!-7O;VZ#_P>)B#T*Tz&9BXD_E&&|*%ba@$9RCry5&;%j!+9uYU(j=__O)~XO
z$jqRIP!D4sd6%-{klqVRb{O%1>qsx0QUibG0oQ@havms4xpGD$Ku%~<57Y=a#}{cu
zBcpUx!L#5t`EJwLuAk0NkeS;L@$X;zR&>*9riM0K_5|s#6J+i{ezM!i`Ref)NN<jQ
zz!q3YK>^n+*ur)=)5<f#f}Zjdfi_7lAl(_HeHu)n<*I>WbO$Dt;R@4igUJKn5gmUB
zAovO3vEzIIJa!Zh0FS*4+u*V1zmkElc8;opsjF@uRvpFX8cV{*v1E7^1%xLQo1x$1
z6pO*Els>3fM9Ib3=jb)2_@LXnpk>$q!k;y02~PHJt(`gnEvKiC<VTKd(Op}44na!>
zrONh{N1-KWMCEaj?+&zNbm4++t@3}(+B*O(rzbPCOmrWBmMjZ^8CLZmv^;Krhhe7X
z*sgndBFa=X%1l;uZLDSmWhPy%qs%2-iKt_UtJ`{U?++}YOu!Q;)9~CR^&IZOR2=6N
zsD4EGx<kclZ-<#q^MRuU9&z!jhcS+LVQBr~pl50%IEYfs^up|M8VyYa!w@dxoJLWy
z96-Ao9qlDwSwp!K^%lyF^t&jx<x9=z^$WUd8L`_JHLOd%lD$gS^Ol%FFab%IF_=Lk
ze>=_xfOkjn0N~xrunoL>{wv$bzPf9BT#>nJ`#42r=zdI*xoi7qMcxCg1GMzY{NHqK
zJ9$oYZ9B@7UE3wT$GW!be1BfowqMmAT7Ssm&US4(A%1$-7Lap>QEOdWFzQU#HuNaR
zF&CFC<&Uh;?7nY1he<f<g6laPHESfHe>4vA-GTSRNd~~B8Sl9^hdV<wP#$n?&oxE}
zuM0XjyJE-zhVAAPR}8I>Nc>|fhTgE2`~z1kZdOu$WyMDA@fFyp2dd|rO?c(bdZn=F
zji6%{S8YSu?v>m1ZgyWqOda&Tiuh39{9_0{BzR5z9FHUwk6fn@+WABTAJ82se}Bg=
z73Hk3&i<@j<=3hZ#JsyRP2z2GWc|HDOt7;6UJ%)CQeU)VMREEj{Y4vAl(zBAM6-wN
zj2_c5ne)4(b3xNduvr2*;T-l9_mn_sUwJRXN(sbE;rnOr-Xq-<KryV`?nLv7ZN<$e
zdw0*y9y(1FUQmPW#|Ay`Pu<yhf2W*`l-&qTdhPdLkD-5SU-!zHojVTtp)~COZjgCr
zAF>L^1D9u7o!+WFY)ViZl`yu`->SWm^)a_wPHj;a-`ky>RRhaaNBj=#X9f~>5O+4m
z9h|QOy1_8ocmDBUc9mbN15>)fMkV$qjvSx1%sjqhD<`2KDJ&|q+Pa!Ye~(Ck%0^x2
zw)k(m6z2h>=j3MR9mo-a`U3|_`Kba(O!~O$Sf<HObns>y;47gSY_{U<okK;gT3kOT
zvh#VNc}}yciu|j`gPR$u$x_pm{xtr0&LZ=(3$53D)Om`MS)(&e-cg)l!z^oz7S;04
z`VGDkehqu&^IDo=M{zi#XXG4&9C8VfjO(hHTZXF}aEln;TRE(ksl<f&D8ej6ca&+{
zR;^|p&FiCu?JeUr6H{EYTJv=Ob%x5d!<M*IA6$?BUjgGI)X)9S{qp}3nASD3<I7vk
zF+S~^L$*D@Vf^$zvlpI@m$4i_6ahGwvF#lbmtLDe!~v+cBb-6R1_XQ$@;;Zb96uBS
zI5U?~2?7(B6r(}Fe@+q7@t5}>o+J}oek+yA2=Da4IxUD6rt&0Bx2vw!MriMb)rkj9
zxU`QvIC`RqVuSe!JcByv{7j$YGkOPf(Jo`_4q9Q%5hK47oU{)4_m%hZDuc(ghUBzJ
zYvIgzA%ne^vu<_IS(0s}nyCOA?<I4gRpCPAE3RU$=&gCqf7~1B?^srqbKYZ{_b&x{
zk_6+`6CH)JDDP13Lcz{bj~a;oUkW#o2J=$*8>Bz+y7*RgZ!5HpzGGo%iA1vuV3X(~
zXkCP7Hy|-7t`}8Q{Y>TB3eeR-D~;Iz8xbU&)lCbnkv8M=iak?QQsA|Q*sPTpj;i6R
zx=e!b{l&v#e*m&h7~=;>EYWCS$pH;>!S^aKujM`FH5f9N)nKDs&POOj0h6*E3(*0Q
zVtP{KWiI3@+%+%TiWPPux9le?WQf)GCJ2^eQen#tK9?z3^SO+(Hx^_eIgSG9|DJ*Z
zh50&Fqp4~`C5#ORE6HdAuE*GIY)sjvfw&+5&4?+<e?<4OH#ws%2EVOq)B@}V)T#s1
zYP@J9T|R5P*I+isx_;PQ|0`AyC33z&8p@zh05iJ7Cb!Bs%7im*O}3rvB@`>GREBC<
zn=}O{K;q06R-9B}hOG9(=oU}fIial{u|n~2iYZIh#CB`E6TyTUaD2A%M33amcpV)T
zqLYH^f2IoYyl>eQXB4P$r#WroQ&u>s$6XYvyZoFiR5kKg3%@AP&7hG@Zqh(On?TQK
zxa@c`2!(Ej>vbEKvw}FA;Ucd=fGNptV)kA*@FR>p3&buKW($dOj+Lw2mJnK^8te^P
zqPjTMi6;9cD#cZ{JeT)jh_DEFT_og{9cGp9f3cyYt$7o~l52AAWO-i^rL^*5X4lHL
zG)<CE$v*H^B<|Oed8Gzeoa!x_Yc;-^TgFn^O?A^Li`UA;%4@m9?j7>GsORZP#8lLi
zpCm(=?re7_XRBpi(?bQnZ%m$Co~o^Zmn#e(gGnXLu=A0s=mmUU6r!Ao&U`>D+G{}x
ze_-IPRJVc%R{lUC)>v=p*9OW_e^VX1{N|@G>*1r=l9mE!XR#&QlsBf7Y0LF8x7h}Q
z@Sv+6WaDV3oEnp3x+(ZfMJDS-PX!@BZm)|z$kl5^bWI;?^V0oD!F#!fm3q6XkW?qT
zFfjFUsTD9a1Sa)htESVai`v<!8~XV&f84#jKm9fHYweLNTOWS?^C`a2V1y-F5JezT
z{{HRo`_qpj)M2_H+wn-!ytI~_rPI7k-HE0I;-~S*?Sjk>!eLViP2lUFWRu;KpBl>v
zWT%v%2MXXdUj5L7?3lL82!rb71hspTEIi0XJ_AAbYQ&=a#9Pxq6^v_ALnJ?;e@6o}
z^LDSu({yxgv{`KnKNQxajPaaf(6S}%)r6X=b!%-qRjYHNFu8T?hbYE#hk&UL(Ay!G
zSCwvqcVANpOwgO*gOL>m%Lm#jxCuF_HVLXsWzJLZnmWCTr6~ErRf8Ss3hrTNybs;c
zBCm`w!FTPNS{>ylW3}a(9-dw5e^ePM=B|Wqau@G0IwM_P8I|2>_H-j@=%58kKia^R
zee0funb-vw9HBoj`DvwvQ~HMFcY92at(?X&rh^qh+3lEqO$-MV0RCIX@ScK4#qb`>
zXJdE|Z7YT=&F-ah%T%UzwG^>!$jz0?Le$-l=eHqxP;6Juh+y>pvHd?Bf7=HexN~eD
zpg%CStJp#z+z{LM&J&e#gni}gt(`5lHdL*y&bIfxrA*xYY<fGNznW&wu{N<aa?)YK
zq9(hl_T6)_OPiZ{R7YQPlBR7tX-+s_m|io*Ie6Wkb`>2^U*_od+ccYIB}-^zk)O)}
z<j(9Ph!AjK<|{Rs7@JJ_f3=!Sj8jVAdEs>e%!LVlpfK@}EM>ra_rfHXKLZvuH-g=w
zge+T;F^rOWyP0#0-#chK<NPw?d@P7YF3Oh<=i6ye(T-+%Upbydim2e!xcWXvrMaQL
zcZr0tIe{h#97%xBoz`b3ls86IWbS!l-FElQsj6vD*~Sx2$D`kQf95IB_3I!*+nw0m
zir@CtQ-gQ&l+9bReLo7de2bBwp4Otq_8gV!jd&J1KrYqjty#F&<zlE`=8*2Gs)CUY
z+&<eh2qNH2E)T4hkUC-O^(V%_2wEunZ%fRz+1*LVBsr2>$HQ03$KWUiex+;-UQyPv
zyV<xM5u8Rp`9MY{e_zBA`9eaTk}DY_&dB{tTH+)-(y)@2IFXL{KYjS$4g%7Om!`2Y
z3>}+>5~d_rn=n^T$ZnAUb&O;JaD^5=Qx+*3gfwBq5<j0z{NSu%yoQH0vsSSPQ!}li
z955upjm#`04BeA0e&YN|j|g=80h+zK($lfpnmfX$$`R0%e_vqwKE)jKM|$P!B>s}i
zKTjErzcA4C1bxx0>s!uze7Bs5%c=bp`IO;sdeCpfWF@~%XZiJPe)e_1HpRP>sJ!aj
z-pz3=PMq3r1<E?T^VNV&nS9jKN+?OA$QtH24R&9oPT^z<H#Iem5y#5yGQ$Mn>D55o
z%4czaAsvmjf9LJK`}DYR34?bVOaW0TgBtPQP$cd-6G|#Wse~CYJh0pX$}mdXM;<&-
zo3APUJm%aX_G=2hO&!tWzM_oiH{79$7%12-A_nYtJ<iQ7?pvNaSotl>3(}2rZeKZ{
zsxC;O#*ZR>P;n7G#?MUsj<tox*ldg<`)F;U$-h+?4uEZz7UM5Pk$(WQ8D#sHu^c}X
z0XUaIf*=!@`l~^~0)Riay{thY0e^jDkRsUlG2ZyMxA&)irn&+m$Jo>Fe?P?^5R9-c
zoPZ2g+ta^)oc?_JnL{0i&BTDT)|j3FD~us~NpLy_=8QO5<(I50dEw4m#EbGY{7GW%
zR(=_UMQ;TtjvT(kYd($!6jn{h9enUG35xQD4pSOn7Qc$w1?>d?LVVDe#eb8;=!yt&
zaPav2)cJ?_FysI7c7C?9K7p_0iGnZ@p$ecD#tT9$-(3prM+K#W7an)$DtZm=mNG?!
z476KG{uJb#IbzfX2970W;>1p5?Z&Wev#@0Rs61H{Ot49YArWq5W+7qdR@V5@PJreJ
zQW_YARu&o-$u*uBnwYArMt?u&`itdKi7Lh`0jR$hB>(e*gzM=J1<4Sca=_UgpFUok
zKuk6hqgyCWj77@F=9^U~4F3bl6Nce|^@%}>C(0kFPk3KtloM!hoN=3_T(ZF#oxiK=
zkZ7=C2qWqFVwiqpX}2Nk)g_+^1vn)9MWK5m)P#}Oo2}-g!rADTb$?@TDmqxaVE{q3
z3oqW35&>|%c;ojS#T*7=(ZMmxtp(ljBF;Hs0XWvRDlV6OY{MstIEb~0*Ao&Jia4n$
zDLejjYS7G3FV$dJ?O-Y9&IibI4fS#PQz;s)64t|!6g|Ypn4<EWqq&yv<9tni<*C3|
z#;L8WEF|%QHAmxTD1T`>MFll9yQ31c?5GKG5#K_(9#OpQvJ<JIAsfVRmR%VB?S^Y-
zW{J};&!Imsvmk|2W9=h5Umb$b59)hS8eyFM+M%+?x+7Hf8}1P*83}N=9<GJg9>d4N
z>#oK~#RwS!$M~tJOC_yv-hGd{OdkOr8&4>F3Z=p)Dpjp@8h@01n@Tl-uHy<>?rxaO
zn1a4H2sZLrEo4bOBvsDmS1M$ciFY21trgDPy^PfqNUR%Xk{^<6&$hhu#Q+jFM7at|
zIA{%NhIBl%8YQGh4(-k(GO*-aR6sB+>uvh3Fj|MJ46?i(mqeAT2Qm!vC`nX#_gLaE
z!d$B@?Bm=90e=*q7$i1#owb~pCPJ6wlI^|bh|tsWp}8v%O);!7O!t;$piu3QWBGtA
z>p-1BGH=I6^JUo1IltYNbETvJGm)>ZzI4oo6oR0Vc*=d2W%H#^&}&dYm@~WwRaY7`
zIfYtcDLcofYEm7YN$ZKC=i7IUE>&g+C9pNe9L+C}(0`5!g|===d%b3PDCYuf6OS<^
z4=QRI{cx5`W$m+-Y3lr1i?(|DQ_qI7a*oHE8V7rfclNm21u-ldih^WTWaesT%*ibW
zgV@yhkp@61Bjj*+tn8@r(1lW03TDgC8On+|E0}2-Qu!TQ1S`wFQH-s7Yk58cTfMMW
z;I?{(pnojMTRP6sV!c$(cI)rMiDQXi#rlxty20dI>$O<IIxDmWPEmxd-PE$H6vq-{
zlAh+-BBwPuLgOOqBymzL&`(ZFm*h(*h?|MU^zvDiN3%H!wKAw4!da=8mVFo7C+&r^
zibL30k1WzfwTF(D(?eYtsX?IjTMu<uYd=ZWrhm1@wp@PXOYDSAo7Otp`mJd$_e*@$
zw_U#Y#6*vk3Nx)~7Cpy@G%Vqqx%HEVI9G;=JO+N;==~oP`eEO2O5b>_Nvk!f9j|R#
z)t=wc2gR4&4C>5F8XEMnZ4}y;l@?DZEw!TvoZA*XF11!2#}9a6tJm@OUA_XI?n5wR
z0e@qU%NQ#BJcl#1ev1TH6?1pLuPL|8fYvcPFO)P4gnZlZMU_td*`{^nfOjG&WSe09
zm=>u2vjz@gBBQs+#Uj6~drEp+ZHlNHrA5nM_fi%6R&EiX)={evXH$IBLS(a6$l>)a
z?H01rK=oubjqjq7T-u5pDg!L}7Tbt!VSg-G&HOs@`r#MfjU7vLU5JKj64lgsY4T;x
zcQ*#R#|$i=lFbIGRqbJX-_CkY;L;GCbRDOVfJ(U0%o`{5`J#2B5lB4;gRteYCJs{g
zCLGW%qO+P}s^L5DGXYCtssVUXIH_!Q@iF<91M_^y?siRHP*WRE$8Idjzkj7|bAN{^
zJFdG<!<ydXk^i7}b6o8`fQ|5JKzQfbrd?WBLTOI!oyMMH@vbQy<YSdSQs1I;%OIQ#
zT!3a%fFuqPlidU4x7%q(^6Te?#gc%kB*`YCdt?iT9ZEgWp=2zb+96H)7m9-()kD~*
z6Pio)yN>u&zw;|QmHhHJr#fyXdw*Qe0=eK+;5&TKMS=Ur2)vEGXM6s6fzNrNr}|st
zW1R2<n!9ealb+g1&xl65sR}J=2mRD<{C1jnbP+M9X=jCUVUMO~!r^j*6a-*kz!zp#
zTb)aMySE3F*;=;=O?5BjxPIf_eu>Ib^lA?4#w(Ommd@hb*KjLG&8P(iKxoelK+^cP
zv@P|FzKuUF>!U>kAsrX#=(uCiTa2Vy8##5G|IM}cU%9fUY_v6KMc}lf+YY+M2?JcR
zU%OpckGU3wRu71|`EVlthi0>%*U{TA^*z4sJ@sP|y+8UW_qX?_{{Tt`;O&>O96uBS
zI5n412?7(BstOJ$mx;AO4u9eXGz_%;+5vJ)(6_;y29r7EF!}$(6jzBQnWVb)g3Q6~
zwxlZ7zLDaW|6KUx=a<{Z%TEsl7t5R$R$e}SzX-g+@vq=`(ddg6f*B<)AAel_@yfY0
z__^T+I9Cck7XNqnt*_sdaQKA4Cj?#<;`+^kz$*%`D11h0{O0(P_<z66*B%+(AiU!M
z0R!R|_@Kabgunjt<Chnn2`=B2O6i$#=3*=`q@_@q*Iz^a^l)OAPtXjU-C&Rr`l)oZ
zWL8ihfV%9{<&0tqG=<d2u14`_6b*i*Z)5^cG$8YDm41ik!b<V}J>L@u6vXxY8YC{h
zr`!6z0>j6^=g07Wj(_*~<ND3;0uernLcOPJsKWfT96T-k`sr!hccQH)axePUj&X0Q
zw%l=jw-_{14W<aTuX|&d^MYcSQNFcI%MJDCDZd&(Px&pYKi&|8SUyM943Y`0tD60B
zHMba|$avx%-iB&MS#2=PsM#<Oo3;O9A8%Rx3;g}w6(=aPd4Ie_g@Tm9tuTF+fi`em
z@dAcWyht%g-(TNyCPC~Pgz2wm9YfEhkjiHj>4YW9IsE@f+{k-U8DtgfTi@J@!78M)
zN=HVibd61;nF?s(B5f|Yd9)Fk=v3K0MzmLnJabRQv_aS48Y5v{HF1o29QbQqA9AR<
z)vG1&nDhkNSbyJ@90RU4EA@3P96rq@g^Iy%LZaR?ZMRE?!lb*b45bo=bpjdLjfT)&
zR$|NxdfZu7_m@S_(tZPp{#az5Y&`!ep#;4QuZY(YHR^p^JhHjsif$uBuq&5Cnp*6u
zhBLc{Z6bzAaYcbC-hVG^R(heXQzJrN)2?jR(C~;*%73ppNV$BMuZW$mf^3OFybZj?
zULn~j^#JV^^U5^-+701hZMRw*-mgkm4#Qphx_C#>)5=r6DIKI1JUH8N#)dqJ2v@k~
znU2RARu_r83TG=_JquG@#E46{<!=KhEO9qLNg-u%a6;V{gY5Z-QkDrRH~7xVsyt@Z
zCFg})w}1Rf?!V5U$uq}!&l#pLC{VIV-jn0Jr-i}{1{BjoVA2wEi`uASi3=`y-L8c@
z!Mv;eU$n;R6D5mWje3nDl(y8;`F@BJH6lbCjn<oOd)J4BI~FvZSCq<`s0nD(uz{>H
zNwh+tNAG6Tg>Td4+YI?m{W3_%mRBQHT74aX(|?tj2wY*?2;9hc6=AWkT$k3g7}we|
zV^uB2VN;CLT82W@Dzd*0YR0OX$AXWI)*%pO9e=g<^+=U=O%l8H=|gX_zFJJ2cMRww
z5QV;7UkbAo42s46U95I=u#2AcEOmRIc8>m{J7ba!=`Fy3as}D)A?Fso&88=E10kMq
z-+!3`P<AN9h##@4mmZl$vDCIvTSI>OJXp)Tah(=-!vlfHJjiU>O`QY_J=Gg_`hr<`
z8_Bp_rY>(|S_uU8%x;fzweEGQPkBl&U~jat-rJa70x?1Fp110g>>GPJff>J0b05=B
zAZFw}>9b$v39vcBdPo41yv-tN%ol;0;eTt=clCttGJHU}3ujn>qh2_2g6-v>NbCNc
z-#UY-TvsU<Di^)3tKy2KOf|J0+{h$P=9LM3YRxK&bTcX!zTtPta<z`o$;WDc?$UTJ
zOysNHAF&{=?b&^4!-VM6)vjN5S%D4)g!s2dExCe9)a$TRN@~HnI_LnuekQ-(KYw2S
zyQyWRa}2D${OgaGa07(K^2A?^VR-S+e_wuo`IRCi`wo1)aS|2GjR`ZMwSo=P+F`$f
zZMn0<)-beZx3=^KBo`@C63Zv)S`3+Uz(?z!Xo$^~-&W?LWs-P{Ef0&E%@s6W1m^*V
z)Cp^IC7QM8yIdP|1uVqLD$-9U#($^s^bj!;%z1EhND@V^!l2%Nlpu<@n?>5e=^Wqa
zAp&Dbi-^o;cg==lwJAh1e$r_MT=L}+p%ulpSWh>f%xt@gDyDPPWH26EpxBD+_+zyX
zc|R->iDAQU<guT>q~vjCOr9RqYI4vkmK=M8YA-R9Zpkj6UGl@z5;n@GhJTkMo<vWz
zdOd(VY;0J*I3X#f9b;P{JX2Coh%czOI54H*Nr9pbQ_9n)zoy_aVrqv(X&{isQOT5#
z8v&N%#&4(<k5ipF<ao(9_7DrQGGf(r^flwq(qoEN8=X`)*h|@?%?qiU{H*jf9zK9v
z0;NJM4+KOWn|M=4Q-fW8pMU5jYgDUDUiU>S>DF1sGKFEAP_T>|%jC2ju-BopT@}o4
zh4W^F?#;|F&)dB*zRN?h-Blva43Va~`|I1efqQpr&smIb8(Sz?`HiN}-QDb=rD0QP
zRJ*As=EkAxab4Y}NyEF`FvJOZDdX>J4ek#B`{Pt_*sfbjvFaa)kbmJYj0OsoBj!dL
z(#06GjtX)axF)ZQcJQ`}7Y$c-y^*+#jCC_I&3gfHuy;?p3~4aB+P}x7)zh2c%$mk~
zOLTzSf}IQ@MrRs?GYqurj8R8t8VCmdJ~*l*n6r*TO`TPeOsWmV{Z3|S&U=Rwkmk7C
za-ThA(4I+HpJ1pvGk<+&4DA@U>KifWp%@fkiGfJ}PsFgnGr3V!t;+(_%cup+*iJO|
z(Q*TMj(|RYaAi+AUpR{-LXI%KZHb}Z5)6FKZ(FD%e%nHxp^vgldXm8oeHIwUj+TmG
z#%F;U{)0pi>25cRf3VL^+bCdtFMC^4LmO=|@bAN8u)=u?HGgTNMpM93LdN%VLPop)
zf5d9h=0I&kIb*fhHgGHc&Qgi|)45YN9c<o>+3xk&aq7WlC6t{6!%XT4Cu;2m^6w_G
zFC}K>6a#BEJbvQjBo?$%0>w0Of&`>wN^BG~cSr*icEe(S+5RC72sx1NuRC{W2GF#`
zmhzJiZ)zo&6n|#d3<rk|i}MKA3_hP84agfezTMDzJ}F4Y!3V<vj5k_pN<`BtJ!u%V
z!rE9-$n*T~oIGn+T7Fw?3zow=vxUlZ9_H8)f8}iyf?(=plosZ|yC^N-DK(NjwMN><
zaZo4Ny}8x)*YP(-S_GeXVWdTH;C&IY@(BuBsIRY!oNV!aKn&@hZ?P<AZkfV;;%$!>
z!6#n!APDB)^=JXlP*I{j?ZKb;3dc+Vi?ul@M|nD4Zz$(5EmGkO4)3e5-^+J(Ie1S4
z!<Ys|08h^JU^Hg<J$38BOC*{pt@nug`g<Dmbi(_`%l`n^?#P$XXfzWAF*7$dG?yVi
zIU@}*FgGtuWo~D5Xdp5*m0=1ef7M%EZ(KJHeb28DurJ+*=!&E$QHvr#KQ`S0E;dM7
zV0QzfaNKF@uIn+z<D&idd#F3UGma<QsmD&Hhz)b@@Jb{_k%yE*QdL6=B(<c#%$=gZ
z_;redBnia1N|0RqnbsiIoT?=#t5e3INT;-fnrnYoLo4lxQ+SWsIEjD7f4P$uD64f6
z!xiguSFjoNp(;dZd7Oy@rxk-hOzM)<99VK6p;}<A$!Dt#(x*O!L8UTEZLZTwYz<tc
z57sc4l|Dkl2(5O~GPEvTUeh3K<6s3Apst$11gd>DvmnK!KW#9~C>fN^N&0M!LgvyH
zY8rBG-LQBgtMn;a0a^P9f31dGRwFP-^${$A0q<fqBRp+s9IS!t=CqWs4!%qvDQM>L
zH4a*Id@4b$j=VV(O4LaUg}WUrHITkcm?B{*Q|q&luj<I4Ae&EYjHfDn1TLYJ+)0v9
zD)re2Q?#zUFJz^zyc?#Deq%6V)$VFW7O;$^&(;DA+Zk8_<miLJe^)SI=Y^|*A6}KQ
zOvY@X2}UulqcAC$kS~=9zR%M2&3IqCLxL|@;jS6pHPca@7OYukz<M+ub%1_t^ml`W
zn$|9XNNB!ZPUg@WTb}~Um}s=L8za#;oJ*Zp(vz0WS-6fv<I5y{l}(*Jk0y3;l9q8I
zr;CirfHT!`ak4+hf4rUcT}6E_nJY-r_lo(3plsrL`V*>YoJ)Kcx{eu}0c7dyiUBy$
zC_-|2w2SXdroJJ}g@V01jSWsvD)r~iucNzpz+f%%<l~QruYbFo4WC?G%$H&KbiVj*
zws@ToWYY2QkKxPVGkgH^@$hnXx_pg+V|RbD=<Z9(u^W~(e@f!xC&R1jH%lMz=<MRZ
z!za`5r^~bX#qetQ=H-|E`_Yff<>l4ihT(^^<&W2I;_3Y7;iv1LFQ0xl`(^m*WclMi
zFPC3lEN6@FPflm?r}N>>`TWiB3q?+mhO5Qt@Xhk<{A_tPyBZd=tLyXS6}bL#M7caY
zf>3{b_Us9Me=z;$uRYEg6nZ)Q>)-$Jm5PS-sn{78*XQTQ?_#T9)1VdXZ5K`vjk)c{
z$`Ui&bIlrt&*taz#j8uG8^m7-!;6z83~&*6{P_1@mtVYEo-Airj{e~J{9=hop1b2E
zL_MG7xd(1d<CM?cb2!WR!Ci!~`uOnd+3;d9KYcY@f4&}GeExj+diLvbC30V_7bic=
zAQ0?-vApU*xb`t#&aURyi__WFDlq%AuV&w!oqRg~wfjtgXCfFLV}u3X1zWL^t^3#2
zD*JR1p1d9tkn1ugpc`uatWK0oRJ);Snkbqmo+z29Zkpr!x_ZeHlV_cktC!rOmiFOY
zy((PYe`xLM_t6f%603**m3B%RIe&2Fa;2w7dAoAAMBA<@jUT$E!X*;?LQ0C+vokcW
z37DX!$ZE&KCjrC18oro+Js&;~AHgze3M7EdBESU*$O<v`lmdY|*5a_tVn!knNwFXb
zi&_tyccjRP5F#=$QccC%NGe;T*nOc{DK^hbf2)Otk};S1L00<v>x(}6%8^rsBU5{T
zwI<_O9Q756)Eq?S*c@<FjsPcjtc=8)5|ykaRYsm{%^qN>8^;bOS3&?3#GdotAq4!4
z67X`%g*i0^4y*f1kp-r347ifQH+FJ_@jgv4!38*9!n!c!D4vzD7lH8JlPORXMe-`e
ze`6aUwP=mxFbkjsyKKwegB{raB}4!Q7T;^<RgJ{!{9zP<YY$-GF{P(RSn9X1QY{l+
z!OqzRlQuYR5KV3x+islLk{OG!11oEaiYn_7VA3sQbPF@>1Cv_53zHh(4<<F+fJvj+
zs}lEynqZPmFv%vEWD`tk6HIEf7Z5^Le~j3fYpBpDqP{u=Pv!<smhw;;1ZGsk3`Qlz
z#lxnDPmdw*Ic#QFSO8TJv|#Z9p(v8J_msv^;xWANMhFp!6;<-PgkZRAMX+QcN!coG
z1uTCwbgLp-@+3q?<>cG!T|%IgR`_X32;hocm0I2_b~`-dHE0q9uUhB;Tpqv>e*l@=
ze&`=-6|Jtzcov}1%7*|7sa2pt0N?#IB+1cZ`_!n&KqBc7(l2op*RZ3D%>>+Jr)I@1
zwpFvjx_j=z>KLeU-G$YiM8y6yY1q=;ZiTl*?~{hThr3S0?Ix*u6ZeLqiQ<WpiQ2RG
z0MmP;XzD#kjew*d#Cv~%VDrVdfAjCZJ%tM|7bj;IQzruH7d6J7od~M2A79+S3xs;u
z?;v7P@>oP9`)QcH4ykJmb84~FL(tF+p~!5{Y=X4k5yF%$2~%e|1pQimdeqXM9?FQu
z&}p4538ge0LbhyP$~CGF0D!%c^Y$3Pe)3(I6V@<?mc3nd7xbk9%K&{^f6khUA8|~0
z%kzmJhyDL*a+BjG;}l6zkRx_rRVtBnCsuF4agjey9%H_n$2i?Dk8xS&F+cuW<tNG$
zr4v;qs!o(mRNF73QASm<k<sAW0Q3hD<=M}UDD%D@^>!=Dd*kgeqP%z49C5ZiF}2hj
zsclq6mSSm#&`LIif~NKme{m2|qheEVsZmu!hO(?2fY&3L{h>&B-+n#SywR_(aHB)$
zwlqcQ*|a{OkpUiKlCm~(^3##_=thQtW6`xIlX+X=5IngRM=Yh?t_}S<#}2jOZMIcT
zC0DWS!m9o^?Zhf7F(=!F6)_Q!pWbKs{lgho+TDyRb(f55tLfJne>dlijLTCd<tdZ$
zR87iMP0CbF%2Z9tl&xjIYEq?Pm#y)uW{oNb(d|hIF@+NIgCJNqn%}+B1a&LaMv%!U
zx75q6M~^9}*JIh}8D=18;=?#-74JADu?MuWU|S38WM%kuZ2I{5QMANHd80Keey(&7
zW-i^COWn6QEBL!{e_{_uO4SZPL#_7oNKGE%yhSahRuR9g-_l@Y3+dNcQP-K=b)#Hs
zLeTOHaWF>J{cc4Np}gI={z51x4+5)m?|(C={1yTb`@Mo%w%lUF<}AQhJFp52pGw|^
zm4yL?`O!1_Mh(4Rde<!lwWj-}_kI!kCcVp4Jj+x(=jmBqe?OaGK%QVg-d0fGa~Bm=
zKfjrFQSqQ?`LTDAx25|v@ac+NjzWiVGMxNqn5jHqtWvkC-(g=_St~`g_SjZ-xYpL)
zRmI=TbPz_C;zrI7nGiC3C{cltIr`mdFAb3-wq4e#{@)0!ic;gEx#y)l@Eezyb)6FY
zBGp0Ma;OPbf3CD=drG<;H=6XYci}^5(pOT~@9mOTcPn{yOC1XrYH24{@or1vgDy^W
zOL40AEKbYDv8Iieu<u<~idpnKNRvl31$6r`N9>o2`Oov^?A!U}Y;oe#jGhIwTHbc6
z-^EeeF76v5ew&UE{s-i&dtVA=Ze(+Ga%Ev{3T19&m%ki81_3yip^_jJmvhNMDu20-
zYq=54`z!PVkHaF18z2Z!>$M&*nFpO2j*J;c6L|hVnBsc5-9;@u!+~1T6^pg7Sj!=s
zpZ;^=ryp<cpKrh1C7ukk3XC{?{&52Ofzw~$^iZpl0bp9f>GRLiKhK;CO|N%)A?H%k
z%g|>_@9OeRf~65aBY>U)T)r6udVeD6iKH<?(L1M?pwA{fyC-^0^p+#~8E9^xfk1rV
zdimGquWvjFT%;3{DeX?$AR`&Fbbe+9oTH`Ll%Xd`b19u(#D_%Sisn|tA8+&?DEw$I
zdw8Prw=m{C&0&V8wouW$c#rRX{_qZ&#v~ebi=2thYxK+A!p!c%I%cgLG=DRRYQ9>+
z8HO;}W6{UtZ(c5!bW-tI_zaDXNdq)V8`bDa<57Dnl<^|#mS)A?!qFcT%BagX?-#9Y
zi-a^Yzi-+IwPg~a*EUPRz2AI~FYli;+c(WU$L8s+{=RA5;%!<mZbjjD&7Ke}B*b}l
zbwP0kc9<v#zAXY#fb0p&`hP<7^=;n9GQm|p2mjb+Yz1>t3{=}I1lBMou%`mP&C3a<
z6+Fs|d|%^W8~%NJ$8f>Obw*ajn0X6qZ&!g#i@`NOY%>>*Nu>uW2vmdHp<+t_zNLVL
ze6}6qQ@sQ}KVr{-<QT|h$H(osg<_VAmd0<q;LB=f!5mYjXXoSowtvxlCUUx^JG@Me
z0Zv$|ZI;q9<vKckD|XJ?FMLyV@(GF0KS5W1uM@}e#OC!36gf@F^b#?k=8<S>+9jow
zut~q_9GY&KXOR~R9|~7+UMRzCIUVRpdX#RK5GQ(t*?E-$-mAn|rG@OHc%S>cl(b-~
z5AEv-#`CrnZ~VAORev<}znb&1BSN}cb_b;&Z#8tqf75KFQ-}sCYn{r6obDs7$`-eU
z&J?+}{7Z814|{F-b%%=5-ar%=Br`_~@_lj0t-T?oVIagXpG>YNlG>tjLvbwUhfK^b
zaL!3soa>Y<%EqsgbljI=pKVzIt!C1PgHG(nld=NI1Y$qafq#&sgkmU%rC8#EUDER4
z;Q|;{Y*}nwZ;4!=U9^l@Qv@e;%kMr&lz<QumaFQv*%k)-SiH1Oc0wMBm`EOtJ>gKN
zBx!<1c9A!2zM{3>)LQRq9ji_hnB1;}DX@L`jZcn*-$L7l-^RjKc!lpP9gJhz>XQn|
zP?r(_OBpawP=69aYX*w8hF#@D30o|chT3P-5@as!VoG^b^$)0)peSy2_i%~!df%4L
zaOyT?;imV+>nO5((|zRoRf>n%_iZR|6rQ(zM}0)E(R^EWrJ3TT3rW52rtBTXj_eYX
z=VMANYD}fC)CZ~6P^H`fyRT12v8NAI#nOH0OLH+B)_-ePpKTq4VaAn!3~pbCjw0zG
zl5aOf5-#$T{V+ta<MyWXgVGkx*L_(#ihgfSWRW7qN<YT#>|csyTk;~spSo#z-<7wg
zI3VwAzpB^xH(F#yEb;TUFMmgI#NG$@?9)0?`{}3{qG-4PAq7C&QL807qSe8js}Ag3
zMgz!)xqokjX;nDsLH*=kB<lQ^-+rt&l&Pdbw9F`7*HnRHu26L;1qYKPv=b$9rOGCw
zZe1fI#eyE@ELW-s6?a%^$3xtVC}&yqd#Hs=XLegUFo2Dc+IJc|&9Tt|1Zfqx-b>Nt
zmqng;i>0_UR95?Sptm30w~x=a{|=?BRKSeE(|^DJeDgmbD5j0aL})cmfBk;?<Lw{)
zq{;sMdjO=vB6F>kY8RQ3rm`pWJ(JbDonf$<Z$aAX8%bo6N;;B6m<a;M8Pd@5CpcvD
z<hROgl*ezdpDk{s#?ibJ6FK*D2o+~-o&@)He5$#oEkVV2$3?iq>2cMj&ZAehpE==0
z4}V!A+toN)Vf(Wzk?q}bkt4jmtGRXzD5`WbE#bCUVbG8$Id=^J<^AruB+f_dme;nO
z_2nLvv~9V_>^+Zqjn5m{&z5b+@6|ES?O}@uiXLvT$2^k>!Kv^Z)?}|}6FW^~PPUIo
z?F8l2QSAJhiZ3~}SU4<x?|3QjNo5&V6@Lo(u&+^WupqjLCa$&+I3@*j1W&AO*}ts>
zD`g6GNm9CQ@N;XMF{TzJil9KmqoNl-zW!R!ZtR7H`%Aa6Fu(-wJs}ij8DUig`8u+b
zg`*TL`}HX9Kg-yAXJk?+3J#?&m+BJEbTL&yu@+0103_VR>4Hpc9~{s=3F3A<iho1K
z-S(u5Y?fKtFzLf}E5QOZX56%e*OwFFs%E}s%^LvS=_6R4GcyoBokP#LggFyM#B}HU
z%`v!r^L*ACife-})aSEp+N>kxF7R7C&3D1jmT+(k4XyFIa!3==yVNrT{5)6kx9&`T
zf&gx#snnNnDz7?;_s;YwSQ>>r)qlz%q2J0VbzKsbHI?F=P6NuWh))Yz8Etm00!giJ
z7+DT*ZiQAY^Dk!v0rk5@ei1WG$0SmhI|zj)hv8dwgc;{<jxfvkMl0Fxfg!&5Q82}j
zGkrm(BiCDX=K-dNRh1-4tjYCtYbV57>)5h|BdqF3by#(azU(-BN>;=0Q-9-^O;9d{
z<i4~xepz-XS7y3Xv0kMXI?tgAYO1Pfj%V=ARhwG_upc_;*mjJS<A?-hsB04(yuROC
zKm>jsh~ZNVI`;&KL7|s%WBw08jB;+67e(gJaY|7ot}M~}F7_cMg;y$4TD|{A*ax7$
zBH()L!&;`bc{%vu8y<h%<bMG8v2N>`p@aRStyeJ<!5!)LBFLnj>?a{6Et9XdyB6ZM
zYSsUk90DegJ=yQ^uD4~8Zf31Dz1`qj?7Uwo6LkfMJtLSS(YXN{rK+%$uOLd`EHeN<
z8ByxEJb@^6GaW#b{AD?#+S`DKHtlk(`iNZ<=GSq3v0Vk145aC~^ndvVuaEVsQZNUD
z>}z&4)ZMz$exUn;4^q03{X{Q!!H9EoT)rBujA0s}pG~D$@{F7*4t6u|pO8-PcpO@8
z1cuZ%4Z74@>N&76Au9a}ufUB)G1THU7Ffd}E^r=@$icI<BEKxhRCxWeXSWmmEYEJa
zJnPx*Y_EcybHM5>vVY<U+~Yu^AHC}dYqwJxPlz7yJSLst<w2!QP&${IppaQp<INc%
zonoi(b?mfc&X(hg!PDYCEthA|)9f<sj8DO*AOxcd;uw5-`6jG73Y|)3of+nT5IS8C
zWzlKBh^ks@+^Ep(gv(e_RB{Ga2;1YAY3%$qWg5Q+Af!<(_kS8Wy>6qQ{!RwSUYjqJ
zb?aAO$yqE+?WK#vjaBp}UR-EIDU<eK3i6f^TxQs=zi0IEsK&ALT&@B8(r^a}7Q&D$
zx|d}IA$5`o4Ksk0!tX)IIE_sA-{ZFxKqHUs+U$6n+#B4TE3g!n8_Q*hATNzu0LSfr
z9iy_u=uXiW`hT57MGSgNwijM&d0YD`z1*R%td6y1)$3N}Ape!sn>i5Gzmn@VI*kjP
zK*igBNLGe+?=Wp!3Tz&PP#FqTytWg>8_#lb<9UKg*Eg28&j6x+^%=kf4Bk2efNTw_
zEw=v04w)3~1uVO)W}dveqL|-5KHvTZF;?!M3T19&bC(Ch1q}l+GdP#=&L<NBF*ucB
z3MZHF&@%*=Skyrg79M+m+%o7}kW-M!X%3nHKcu)yq!!DUf7C%40a=%s)ImspoW~m;
z?<8WFf#(M92yfv!{QmySlgt8_dzDgoVVoI^MZvNZDdP;#jQn(Vu&5`JKxg|I<SC9`
zwGJs^p&x|fva8D(Ei_U;lGr4~DkGbMD9uDR{f{a?cd0G(E5E}tJo0bgd+dJ*7ALlx
z<SAshpZGc-o{bDo;B&fy-{?4hS@e|J*j2f=n)FtAiLm7SsrFe=lLb^s3FQ}8HmPS&
z5zkrG()-x?Y{mM;>*U#p)#-G-)7uH;E}u7Ju5o!Y@Qxk_`QP)nVEwckq|g*RD~0h=
z4L{h8%TK;eF+dA4elXeyEnPLRnHZ6>1g;5>9_tKP0g}oL0jsK8-V$zqMOWC?KGlUj
z>u6FNK*G4#=NNR4u8Q6(O)d!<y3&eKWB<=p__gq46cxmVYL$3B2VeyopxYlvzhr~J
zxK5m3rM8kk!kb{DaQo8zcY1&NduDrbAT%%j$M>i30D*+Deo#tyz~RTA!?&kj`&m0k
zm(!XQGf5gl)aJbrM2=T~v%>Ja#nTrJm-2#gEfL7m)ju)DnwDv7Ssmi8Uok~?TbVd6
zdAJO7C`z+7Enfp1dAX<;TQ9_4w@Rq7NI8=22opum<t$tRz}bolL7=iTOP;Q_kYXq`
zyhv}M^M_hU5eo4AA{ufu%VemOi2Dp+J%jno01*t2OF3|{B;`bZAim2{LOMiBg9XWx
zgpnvk_%Pe0gN#tnPnzR+RTRTAGqvbO)zrQoX;;N7N;C4p$PsrS@AUv9o#15b0R{}K
zYPJDp51;@FME^)YX;-`#ptP$z0#G_jo+gz4cs;m~<Jo3<bS$s-Hi|9e``+4*lq1&M
zY=uZg+*r!7l=8QK55NJQC>F&1IFPG!Asd`9c|)}vjrD2ZgTDT^NVUpSM9^=TwRL61
zKwGvrL6XfC<^pxCnWS;1Bwf-)R;?N-5jRm!8(<rnnGI_kWoorXLNQY@KVix)(al!R
zk&YErD<|GWSahDN(agzrL|Ej|_H2?nA}nfRXt8%fSPngZq1@G!Z5Es6dc^f?Zi~GH
zAC0))3Fmab$~_S`V`FdG8xhy?&e*7#wN?-}v#753>V|K3W|p)}ds}FCs~Md0E|Xph
z3}y4WvRl>m769ixVIUm;1NaR9-r4Zy1MpVBI|8`o72Q4IJJ{Tzx)XeB!}J^czX;!*
z1b2e(Zj}drz;|!ikAd&rTHEm5^<IN-jmev_-FB}nk7-PsqY1w4T8=&l%C$z0fuD?v
z(?w(cyzp*~uxf7Co$$Udj8f=qsl5;C37jC*y|DL$dd6I%=?bv_bY_U%ID=Cj1SmT_
zgBzXSAya)X&CW#kGC4Rn<}jSSe%^kYRAjJPSQ;sR-fD2gHO^VW*GUC`t--b`(cu|h
zs#GQXQ`)HXq-ph|&LzRdjr~H<m&A~{#rM)}q|It1)?N!<L7aG_&d#)UaV=Xw!%asl
z$iK}IcST1&cKE#g`my+)jL=T~TjwGs8|j#hQBJ|O4ZaoBS|am4>=b*@S&34NB|r~4
zD^YxZcV11==z3%3*PUZ*Ud>CS7NBh}%>s+7Bf@tFX!;Sia3h7>;%73fa3ke}G1&LM
z$OzZClNkxectV=WZuZ2q$1SfO!7E`MTu*nn>=%T|J1%<NNoLx}Si=*=F>W~RItj<s
z^3iuc+sc6_GiYUIlbOseVxZ5M99|aETTUE*+t1q;fo*hQ0)Pyg-3;Maz5sCj)L?~S
zDKC`sOS<haj7Ewt3wsTY4Oes%4tksv)BIL^EL(OgTU#=7Y6+0$g57f9%Tr<Y%PFwE
zd$&x^$lqXJbeeHre^4M&@VA}KmVsrKo?(kNVa*wIBV{_)RE*^l^qZQ#>}e=N>&<<C
z6y?~k{9{8*a#6?ALDP(cRZEn5>{>z-1}|&NNh(J&tX&<S0FING>n}q!Urv)w5I$?&
zqTMKk)w*A);6NlMdfa^!>Qnu@s~9GN(vpiyc&rG7r@3I20os=WQ{zU7&A;fZg{|m}
zDQ>IWjH^l?MtH1Dc&y3SQX7LpE9{wnjSI(eH@|{{@OjU;0V(Fmh4e$cy+D2XkEb$f
z#b;&HV`r!lq3&4g`53ZnF3SZ%k<gxYz)x0F$qJjdD6=Z?72IT|OE`4ZW(xZNmv=?I
zUZ9s}Xa0z``Oq>-zoGTgo+oObcYo!$H&|lS8vXQpw{?%q?U2KiYXCt-_p-nu1oL{;
z9K>SwXw9}_U|BP+x@M|ZIy2uOs{NfI;~!`#{VtcW96uBRIF}%LFBO+ALMSJf5UD5x
zm(1HimwzDpz~%h!_kTR;!f@jbjba;4){zn(DtB=w^6`q?I2j?KMY0FW6UIzSA$yM}
ztw=6!*qr_r>dVpJzl(M}tZIjY(!q@Tr(I-!HsumYS)^u{pAnY3NJbaALVgy<$(H$0
z=)`e4M+K%KH&eux+0e=$xREh2MqK^d8~asps((7!D$v}F9YuM`N)#!QNKu(l#cT`3
z4fm-OjaTVWF_0_5PX@BpNart6R=H=oTZ*^JGLLj`Q)<<yi1gCA<R$7U-aDliVWSPh
z<V|szJk}7kB8*!N#q;&0j$rX*ec4r%5kZ{z-5^jDgl$t&<bkqpRb}O9OP+<hDl;Ax
zYJa~l#X{_ZGSf_EcU3}is5F{!|KgHbiBQH%%o(=c%H0NyIU-k|>_ugDI+SQOVHTBG
zB9n{D`SrX~vL1(^0&uUZ<Tk~9qux?{yLG)$)mzC&QNJ<Ur8$>mKd45oLq0PKaBmvP
zB(FOCzBiIp4*Veh*SF@07|v{-3wu!ZDt{u>J)32Njp~iJ@56tu?@#|*By+ZyV*5XT
zJ;etEtgH(sL}7(afB$j%^XZ=>)ndQ#o(fxX5^W>NnUh$>k_N@6R51AX)Sa`(BGD9`
zkUmX+l9$g$exH01(JO0dWE8%|Yv#Q%GGw{dsfuN)vyl`tL9eL;jgzP}<1R5#^M9qJ
zd_R<)b;Tl+g0Rr*Yc9(%H%g)?X@kGkI#2?}*OZy(->fX7gQBpoD!eU})mDHP{n&-z
zEiK;8gVD;l1q-mgy<y3U3+u+dig=3Kq#T;!tado@x7BNLcBMgR!Ddi7J9SQcPQ6#9
zopa?m*BH1`l2HMU1S|9Gn*(jK<bQ+!k3ew0rKpNIg&bv%YuS9%5HI+=t8+Nki&0RP
zU`?*Bw5PHG=B4bF+E-W0d*Uc%qE_&m1MC-l@ouDrPR)MTW@4^&3yzBmFI)&nv`#;*
z`5yNFOiLmwbc-E7v4x>vRh)W%3qwxoobx}cg@LYw>d6(-{BOlka56*+1I#3;{JVc+
zzm^s+$J!bfEkJ`#eo$LuJV|;pV_V~#aymS`ts!E?VG4@7?678p+}2#RLu+*!V<c*D
z;*W-uHLayicoJ*=Yq;;0dJ9&fk=#gmr9tW|jYQ(O;sZ4wyp6?gWMRGLLl8G~w6^_R
z75Q8w#nfva({Jla6(6(Cw$c)M7ruW(p^~s9u?|%C6l#M4NmIj$4=EIK`WkrH&}dQr
zeDa|t%67p(c@ECi*ZTM_32fN`q+r6lshEoNP4cr<MzHHZan}YV4CX67Vgt!d7GS}{
zmJN!<=CkiLLf_%S>$<izD8~)oHwtgXPG1XL>*K061(%Y~!d|k!9jloLQvH9(*`u)<
zfg*I{F7aG7@r_d&ZM$8UVH3|qI3B`cH$Ci-7-yYDNIT3%_T+JpPNqhChI#9F6n_R(
zv>-MHY$=mQQgQ2z^yLFUO8SaEnIoO2zTV|dAEBj^a`uUUHBBr+pguj=_f%R~!p+TR
zR6J2=#u=B!Q3;TA%va}sZ4`eO{9fs#`*j6A+O$jEFYj^Xuu>|tfapQ1D}`<IxKSGw
zcqCBHN%u|Rdod0$E80|SV6oSFYqNbmN(!tUsl@ju#z)A96A<3yMCu^%g2pFh#;YMg
zk~#c<cXy8vBz+p=hK08rA+S-7V1+<CWrEu4^Xwgje?BPkI_HR_G>?B0ANN+2cAa=Z
zz}CQOP?~#&=l08o2RYC`6_gqg){*-FC^ed-g3+%IO0S*gUO7ZDN_!k+{Xt+8<!;cA
zVAEQui4Vt`C~%5yhniR4$*nehu^k}~4~9Ae)dNsxSn&X;GxD$tb%x159_k>2V#EhR
zoq^&3s57j10Mr?I*oS{Q!|YFlIzn3F+Wdl0XW;CUpw6(%Cqtcax5q=B(VicIIs>r{
zsDl`k)$|as1EJ6OS6&)@F&yUo)@~1Iy|lqJY;i*?1zt2ILvsQ#kyUI1dY^+3loxf(
z2`pZ?Y&3EZ5A1nl{~(5wOfW%?^50ZXGoswAsjh0eds|eHSB8HPOrihwh8BC;{7Hv#
zFn{m1jncGo{(e)DM#V5{x&sQ2cuhp{%QYA+GJf{5nVb91Z91#9zS5>f4xX<k*Q2o|
z#fAKYBC{bM_R`D<LNYM!a#S6u#D#KoM+;LQkvs4ERDk3Zh$sVd(I?UC5wJTrau(sR
zd)J!R%+5RNxrl#*{&>LQ>TcXuYKbV)2Z4D!XJX<}vd7l(^1a%A$OTYgg^}yvTzFDO
zny%45kE3%&P)3mE3WW}yuf1Z9O*dm$u4uUUe(0V*I~cdPrQifoK>*%J;nxK35PQ~e
z3wXcwX)qW_Z^c)>?L-e<Jz98rf7a-+rqBlBCj_6)Dt&*A<a(gijs-gHr6~9P;T~h_
zsVWV7><#8bD~;;zi!i&!J#nW?=+1pTws)qDBDGaj6Ja``DiVY#@g}(ozNp{c=DZl5
z``5N@H$Ee5_GAb3w2)^WoYnpil(=;BsJ@r=-O0<h{V=T5bW9RI>yyi54Bve;vToAd
z4$!m0g&+@ip8E`kF|Y4W{|CNzr&yP<96uBSIWd=UZwnKb3*<q}fBpE+m%rWVBCuA$
z#*guB{_o?<-GA?z_-GW{aIowYXgK`qx4ZZS0W0gm0Z|2@!|(q&{CxNKdDXht^q!RE
zL6oJfWge7;Rx$g7wMH4|s<P-&UUE2=!;&1DL;)=C9LGTZmvI*5Xi=zqmA;E(aSlD_
zgDa9P+MnNlIZPv;e?Ki}4x}uuq<|0cHK?)9Ky4HlH_vEBI^8S-ih{#R0@&UiwE|J(
z#_ppgvYU{&1AWY6`gZN0wJNZ=LOWWC<7lPh8pwIoVMJE_;W_9nPe8|tEUwNJEiLl1
ziSiG-o%1OSsAy$O{+JhwUL#7lw>m}`Ki#Yj3066ey}?;be{^!-H*^+bf<kw#yBLy?
zHY)>~ASxBaU@k5~InbipL|4Jn5^O0LCD?S8bt~@AB4ULU^vaoKe?BJEs)3$@n~w<#
zSuha2jo(uI<Q}?bOqDUoQruv@s|Rs8n0Jkkk!JJY2fmD)%9`6JK?G&c%xk8g3xk}a
z%bt@_fL#Szf5~W8Ctdt^kvV(wuDnJIT+>Pg&rQ^bvrlCMJ|a9^6xkS3(fh|Gvqtx2
zAy(pi^F*pnw!f=I<-%&)Y_|Pv9Vy-rnDU7+pSb8F9wX(M$EK6|%_3*jT`^EL^YHAs
zY!&q$qRTJ3&fnBE$d;TJ-6t$49^I)R_bRBsD(4%`e@HWRkqq<dX%!b~hAYm>upc@O
zgu>9gb^*(?VeW#`-fc>BVLvanS#P(U*w4l$prD(bSE-n&#2Us6WEr;Jro^N%qwMB4
zpFf=!x2T?jps=W^%4<8bb2jEGW$klD78Sbq)q(>#JS#_}P;~M<Icvr0$7W`~xDpi-
zkpDm4e=n|ApRVCh0H9x^I#4L#_FQEdzytcxBo!CZF-X?tRcW5us4b|e^Q&crnw@NW
zQWtr<Z@H~Q)e5kL7hG_^s(k?&Wa{&C-6x(i=IRtyZpn>Yb0cRBr5su`29TNhVud3-
z^oHmKEaSHDUDu`L$L>H>jENN=@{$EMn%?75e;;b#mnB!(`~qyoc%VrqkI&6xFdW%D
zbRgG5R}r_x-i@^f3t<9@!EaN0u!+DC2E4l24qHY@wz+JF!b!I6a}!Y)*halN&|Vo6
z@dn&HBi5VA^E(8)MUEf{EN(T+0dH0;NEE8z!%FTcDrrgM*Axy?mPdv%E&4^2mx2>j
zfB35@FX0TVZeVEpSyTv?B(Q1E8v`FfD6WCcRcw^dJoHl4k#Uz}P9IP0`AU%Uv(!HA
z##ejFcQm-BhGgdW1T7q6SwCA0*?WxcDKBrNgiTJi8r@0thJ&&aAhR7?>6Wn@OX<Y6
zOG~t1f;pSqT1tqIn$66Xr`IXQjTKwYf2~7BbhUFPhcP{Wsk?kd)<o6|YqB@6ra4a&
z30-vNJLT!5rc*~tyF4*?r|pwrUJ@4umn7Lu9~-%2*e0i!^-g=yO_D<m<D|lGl-5ZY
zy2WqRQY7%o2RC^4&1T!OS=Rfj&34v1QZo~7?*)}yIGy*6+jB+-MPa4YFV48Re~&e-
z`b8<%c#^YR-DN73&brdNY_#3Z4v22uEKKWioXfp37|c8?L1gi@TrdqHC+NG-W^xMF
znwgT=HPH}#be?BYt06~)9dWZpNlRdQ3c}DahtuLlXqP2R9k-wN3WtvUvD3va9s{Ys
zurf}jID3j@@iDj7<OR@W^oq#Nf3v<~jx{zi@NiRSLLhUeHZ}wba3MgAD~LOi-|aWF
zUc+wlsO}>tu<-xV07;{jF4T2~Hu)?UbKlTtpko^tUur@aZIuq+i`-lyvgnoqInAeq
z+}N&ne5R=YCf&1*XYAOAaYx|PAU!Qj(bv)qv>`Ru-+*qE6v;X#%M_c#f5_)TNw$+`
zr6lOALhyYL`n<BI)O}fiwwop8$<4d*?Kf%QtpxqN(Cd6;M4c;+bM~T;BbxF`nSQuP
zICJlPC5J#kJ6>+ekb16g!E81BM|IaxKey!Dv)G`-cSSI7f=#bs(*eXt5U>DUc)M~q
zUVDpPCjFR;*gr9hP0v`ke|6#0gguI0RoYU#pc8+Y1pJDr7BSt2y}p`7O|p>tPJo}t
zTe+{I4;!~oh4`7_wkw;7bEj~=>bfHSFLj+qoqVkX0alO%_#4%u;7FWmp*F8l{Fc_U
zeJ?6LQ)K;$U#1kmoelucJ8k9Mh@NJ=N#b}7oC8TxVc3TAt$W-8f3qv+Zh(!$mDh@;
z);g!vHM*>#Hz9m|)=q-PAxmjpuG@BR%i~?QosV6wgsGv(ek`)kh=B7d{YCiM+SzvG
z+;-R8uBXfCFw}KsR(Wni;yVBGr80R;Ev{J$V?sHbNJ9M46VJ2TP2!nqsXClsNc_c3
zb-0ArU>-kCrS7S7e<ku=qdfmduSe1*iDsq2ZWmOpxoTz#+wSVFch^%bsfL@_`f~)|
z^@$cs(o`N<JP2>M>0-TOeO{P{)X)OiQog6-y!-e2JVWyQ<_xcl!C}TfM40!x+XNKk
z8n^l5FLerUO^Ls;G4aQP-IO19B{LvCWz8%ZyuU~1-u<=AWQFbB&?y`a^-wID@<`R8
zEUtnxp69u828&Aj%$wi+S&wEnCm1L{Pg-W+kNesnR^-EQwqJkU)wy9%8b@{fvA>CA
zq+noDpx!sFv9ZTU!=yj=w~_P(#%`Tu&i}$M{s$>t^OwQ00uux=GdVMtVM7WOm-Gcb
zD3@@nJqv$EV478YBgz1hY#G?S>~4J9>?w&CS;RoHFp&KG6rasUH+x1J*~mc}O*e~G
ztSWx>AjOwo7ruPBx&L_c_EE^iFss1e^6}FG^n%lG;Pg<d#Q-oZVEOoY`T3S})bw*t
zKaz7H=*Q6CmOj<#jQ~p{fJOj41vtGi66lGbCxU;*prTJsKS+O@^z1RwYofPYN<Rb5
z4KxrQ{NJZvK7Mz@lfXqf6ijKiXd@ZPkfl?1oJ*2U(u_)-5O4k8VTQMl8W+c$MlG9l
zD$c+*6TSNH>qxPq8D*Imc}tR3>BP&zCZ#qEh@zL8)~Ybg0Q6%{Q~pjc+~6-GwanCH
zyp4bLB5<ZU8D4!qT`(eNMw=%r7|{_kH7r;s6U`TjQZNGpU+-8F#cg!HXzuKML46pE
z588j{{*76s&EK>icX|V1Ofo@`C`v|3%p+x#7v9a)>PW5hN(;@5vQe$S{q}n1dgwHR
zmLhxJHWR^t87Fs`7_&)3_U2BvG=Iu$>?VH+*;@lHS}L+TMv{L8lJ_5P{#`a9?SH0>
z3gc*!<*%P_e*T3of1{V*Eu0Bwm;Z#JpBG6WLh|wDpUa<bevG)Yt&}opE$^rK5^1Qy
zj^r<7*Qxe}^dh=W|9^b_2DJ8{Q$uLdki?eu`?-*sP=j9eLc$ztGR}cvMz|~BTs0I?
zQR_a=X|L9nceU*~LuD<JttI`J8t*|G2)j*N5^5-+Pq#epL0AC+Eti$>K_>wcm%Z>o
z7=QcBd|Rnd>UG=#_5G@;01IVRYj8?DqlObilu<R%#3wohiVRnet5#N>ZQ3qsbHjE4
za(S_=ZM6T;%$qm|pP8=1ly%4)7UO)6X@*p3$h%rXKeh(rpf|e>@DeF_HR+XfHC%TU
z&XL|=*XLo|=gXTCT5L=<q6GKVEIC*{RDbEhMp%i})pljIlUh;!&J3=pl{>}IiAnaH
z02P^n?$i8P2f_X^8NREfOcso9Wb-ZQDQNx9^GYN9v-`r=t^N_$mEVLN=GM8(Us(f?
z=GOU5ootldzf)!S>Y37-S~rKKkfc%wTax2aNKkD;HDW{cC6_{ThKgbY=z$^3>whOg
z8rK^A)jSsvDh7&6bfG<rnM-v+b(*rB1`l*#Btvtt4lJi&IZ&aTi|2s#YWOneL%9l2
zF42(uI*rb>0T{?(Q&!t6gDGMUrt!T>;IO*<!?<1J#>N^x;<r-KZbI=)Jqid;)Sy(*
z)AiC$W!Bn$y#Z2lE_}ZYTg1<|#(y$}sp=^+ug3}YIW!X-_%l$g>Njakap=yVfYL?T
zQqA2N6lJSE>F5~E-5F@Nf$513ycuZcXxpvn%Un9zHqv#B{xIj_h-6UQflmN!8>(4)
z3x_=rJu!z0(^TqxW!2P7yMZxve&yYMWftDsl60)eif-yV(S|DdZmrx4Cx6T`ncFo?
z9aV;P;Mxh|l8O>vhgV6l4~}R=HzOK(n8ub-DJ_ZL>i$J~GR%9FBBfCqDN^FNlUkz1
z-E{<A>vb)`J1VOto|rYVEqP8odMn27Ywsl{Q`6Ex3SXiQHhH1aL2=6*>Ym$9Cuq}m
zEJbUAf{IcN-km!UK`LG8M1M7&I{NNi6`{ucY4$0Pd~`E)o}|BW-94Q03@)s11Sk)J
zoN31~cE2{?%(|#qi`w@yT@%5wC#xd$=80N5?~~Fol(rPtyhS;Yz0u^NcU@cHkc%Fj
z)BU~=cY@pfR@~Fv^AvlcQx6s1Ln+|4>$`g_y)-I0l(?;S{eYO2<$ov+do$6;hwfX`
z|H)>df{<BIz5(Q4NT5R7xCI0<#l`ab-<MA}Klalm1NEPwHx#^pYpqo41v>PZq(8v1
zSJ?Ap{4Vd2BAm`h?&?fQyf6EjLM>m`U%{K3CqGp_N5uoTvY!p!mE*R-OEv8cCwdBh
z8XZu(Mqjzw5a+x2Qh&LT?B>7@Qt(ZgQ38`_2Aw5LpGWDmdzv^EPtw;26?fF(GQBbB
z$L~1q1DgNbG_G(|YiV(CJbjU&DLJ&#^fS*4ZG;EEo<giu^B}dQUEg)B?9-%GnhD|h
z8K-WU*4I*$%MK5h%U}%C0D}y$&EG-*?{4(bwh(&lO0NB^^nbkVQS=ZajHVT<k`_~7
z)f|U|s~99T4*>l>+E4TC8I`0I!oDDA3Ho%}T?w?h+0Fp1j^Z1ERxiaIw0fO=JJ6Z|
zmJq&{egV8{Cf+cy&&-IL^KfDrCvBdDoXzpgXq_RhVRE=<U90u_D773^Zl?#u16p;J
zZMhGvNG`A7nt#yR&Ug;AwiMTZ)=q{4Xl+M7%}nP(D*)pTp8>7giR96d78*+)+uZcq
zu3~(juCHI?lf!BWW$6wky`*E4)TEzYCZo(<(|+mXyzurM8h?H|%re}b+mXQu1_}ll
z&a09IWq{U<E3j3MG<SiXrQy2nE;aJ}s-6U7X3)xfJ%9e*g%1)PRk)xhaqR%j`xNBu
z1WM1ds#J!78|BY+IhIZm?CW;0;<m(!Q}t%;6MaB|3w@c6I`-C^qtf-B)v!~wX6Q=t
z6^^HNtJc!z8#*fX_}Fv07r16lU@uL+50ujcAHPG|{%~;FlY%A=v+Z;mJr-b<#2ot>
zke>l1tbZNlfro=X8Pj&4YHs0Vl*~<yy4Mr?#gpFB-u<x4yHD4;O@)S#z(Bv|b8k}}
zr--L~E$KK*JSRmw?~)c}?UG}^s26O-`D_a%KBE<f{^&u#H)_Ql+=|<euMo<kR`{>f
zX9EINe~{Pp+0f^`^6G6iu>NFqqvOykEP|lAe}8fv(s6pBLZsvJ6)KUAqX)~B>Rz7K
zqg0{k`Len~B`MMPimDQRA1R!u>gm;(5Q<{1JVRycrFq(DLR@G4T{ZOZjM~LFWdyEm
zLUjuNNwj+@ZF0G4Y>1V&{n@Hn?*~+8RQuIUj@PnCu<sZ96^LO-5ZA3Zfy;j7A(;Xg
zsDJSIve_9Kv5=BU@1x!9XI4d;GBX@(SXE@@UvY|Dm+MW^J_4==8RuE4KS}WyFfY_w
z@r>X{Me0IjpQCMGe}ERFwhvBdsqPq*{DS!p5V)Yz7vw*ehkHtt+{Uq*uS@hi+@BNP
z!9zWsAKo)62Yh(A5=MjbDUIl*nP4<>|9>Af*i&5PFyJ{DZT)Tj|2u?+Bpa{j=13#B
zpr~rv^&a6fgv6sIxe_@0fBbUL&3!I3>(H!dXs5t@xw<4$Xd(x1<vW07D&{TMm+9+e
zL770u|3_e_<vAoHgnuM%XkHMJshAKbF9GE4lR)@?gy^u|3T19&b98cLVQmU!ZkN6s
zKL-IhmvL7w6qk=gE-06a_(7Tj#FyTeDEUDre@xuv`Ql)8b!`M+W%Zy7QCfkHNiVI^
zZ{*cA<afNI4!ip_KkORD0Fn)Kk>;`w<_eE>D7-dl>g~!5=f@@2Kgqh!<3JGz@1fEC
za0&hb8?0dWpo81=DXE`SOKub11kZ)rYxC>;w)=ZmEsRQN<2!?rE$zBLzV5;c0xgX7
ze;uOmK;5@LyDz(6+Ep_&+Iz%y^rtjN>ybav2+cq;Xax|x8!cznntf^HoRSr(3vrb-
z)m{l+rtTeZ1jVL$fTRcqneRU(iO7~{qeiwk3lsJtHg#}JnXMBf%CCZ7wTgZuFD&U+
z;u^`P3^&v7tc#^|M*JWeN@P%pjN=vjf1%*@+TD9iqP<FFU9$IL*i=0+#NNmT6{@S6
z6G--zooMb?B@UoPPwiJZBqlhp3~|J|_F=D!%+Y%1qDjy*daLS1w9Z<_vENj?LqJ+U
z_ryo}_bl2MeVyp#UGC<!E~5ZibiHEAaT62i!+;}?6#}x~w}woF#5hzy9idS2e;y;>
zy8M+H^6k8{v=c~uGCW0KtvW#gQ;crEyaFLmPAI0q=BudJ2n-SMW*F4hKw(ihG!uCq
z6U=B(7{YdUd8uB^_RoPS1#0Jzxs~{GN~`W$VA{#(udmDW<+vn3JLkI=m$xN=rXX3s
zt|sC6RkQ07t|+9v>I}$YMA7z7e+c-}R~f0|xDxpw2wl7EgM6$xS<80w0$qei7*)ff
z9mBRf4aADi!!WHDfeDi*7@<<b{HgM@>GNi9O%9qhD=G#K^3Y~hj5^G)tMYyBj<BN!
zj<b|H4=i--4816PQ>D_OuU3OHR|p!M=Fq}+mq9omGG=q|B)WDZD$Ar*f1YmeeqP(o
zmYfp{aQ~u_`uv(Da>IHn(kS-0sfpw;tQx|1MV?9%hc*et(aWG3Q^j*t$Mat4qSJ`d
zr<F>QwT_sK3^^P{zfT4DAd$7TAPmxDcvdCZ*(H0*nm8w<{LZ4D9N#szMLig1NV8tl
z-%!Mp^PJJ#h-e@iB4zm`e>BOi{}Z#wXoNFJU6~n7I%-!&Ib!~vY`IAEY9dEdMyYdf
zR5GiPr7fm=S_sl?NM{#H;wEr(vPF8=1E1#0;slbyXJZ2Tyvz(FjHFhNTiY;P@bney
zA(Ix9Lc442Sb1CJSbC)mv%VC@OH8`WzFE<)`YBX%vStu%di^5Be-A>d#a2FeAn=UH
zAMW5`(T9swl?k1IY3#ApVT5AdLT?{*7@?Ryv{m;<K`JT|*TjQ0c$u_ZA0r&aAwz1=
zNzUhvVYZr~k6{HccKyfBG_W>dpF&DZ{v=_d>*uuw=sn{|wL5G`K#B>7t#=0jv8mZ|
z@VyYwsUA&yu)VuRe?Vh<Y(+rG1SAO`L_o*{q`-exgZv?eL5>Nqc7|b0Z_bFKQ-1)`
zpN#IGlFvuyf`fjhXOj7|L3Lg~2}R%%&rYPgA6IBNp?TM?%;jMz)J`L*7yGK2o?-QD
zH1TOE6_@*v8&6D?=TgT_EcBqGhl(O)h4tqDp6}o5`A~*#e~32f`H<HE2^;i$&B|AL
zzE+Oydp>X=dSrjME#Fv;kEfcp{!LBCZH=H}hys~LmJQ1=FvJzUzoQHT(!8ttFO*^L
z?+v+pL4%Erwkpj)hHmXal^O8zj<ySzwqv}rP2_m8{b}$I-6b}vZu~zbwy%<|*>pp~
zV!l%S(fJL03Dw>?bp8RF(O;L5Um6paVE#b_m*Fx76PJKLJ}8%;{z0Ds?Uz6QL0x|p
z9c)?3fD$pPN%r$KX;k@Svv{)RSlCJuGaFe9UB=K_S<fQ6r0VqZW4d~KfBD}_Rk$|7
zx_EFV2>gHe&z~=80+bcb$Ai`)DmeW0-^1@O|D36kE$83tjM==>Ico>=YA37>2L~X4
zjmw(eR~>o)8Xi1}2S~7f{q@IT8u@?uqe<w!a7vAPs^8LQaLKwpIMl?rJVi-z1j~lE
z&A+k_-sLdaLF8r4Ks78jf|#wOq>#bYu&}2bXh7P_TC}Gez81q)tK=>@aPk%OJ|CJG
zmf5$wDqlrjRYxiOATaHv?dm>nCsTPV&W|&i?t!>MvjoR_V$jhD<IJpqCmDZ+I3w@c
z4A1~L=w}{Mf;1r>O9HHE=XKEKLPutyqdHo(RjjwG7jOo>8y23y-e8){_Kn6bG>&8P
zx*T(IR}1WmHe$wT1tnWRbE>LVg`(e5QcxEDw`!L<d!^Z&=If|yzDy%_xQn!(I4McK
zXtm5*%I|rVZ_)o~qO_R8G}wPRQ93C+>S;ew3YpTAO<$6-)-dvul4q2A$<ul#jIFgX
z)jA5^l?uD00w{;KGnJHFbxaO%`sSpjYG0GpZN@jb_87R<Ex68A`#z3KLY@fRU!xzc
z7}}(&w@|cFtMx(~FHp#N^<!0S<oOwDXEQr1OpL$kEa=uYnwd)G`Hg>_CAU5CIb266
zjE4o+_dAQ02D6f1P?$7q-g<ZE<`yJLpe^%jNrGXd?=DzF7Tbl|FG$Bxq!fP!_Mc)}
z{Dn{xFmPAsAKqz^VK4x!);B5tN(%#6Hna>ijN{4JW8Z$HLdvCxdwH5IrG?J|s`dBD
zjM~K7-r$v1eJR!Qm??jh(Lj!%3$;wK%E%?6@-me&qSd!Tc?7W3<WNvViA#N%ol068
zZHJFWi5j-TEz~_p+N>QI+qaI7xOYP6`A_AV?m2Vw*mTQ@t5D{%P2VUe*%~jPMvMzD
zx0zUg73h{d7w&+`U0YN8md-4BV;{CUC?BbVvd5Jp*2eKNDxrTM%)5@oPmRTI%b`&x
z!lCO=v@++HqSE9b*=0O;HQb|Q_wunfL|MbVy629@N^&bx6}l&(!G1(xpFDB(Q2yKO
zf1MU!>`Yr#e$!DrclO0<mwGa0S<k(D4c1kyU!%3En+X)C_vrvq%Mtyj=m1#~&MiIY
zTBWIdX00V5=+u8>;3pWoh3d&P`f~>q;aKq?bgZd!UqhwaxUUPDD(i;7E7WR$nTrZI
z;5H_pG}!FnIRv#7r7p6r&fFufbt3rjfYIftPRa8To=_yf*34S>cR`V|Jx^Rio?z0w
zgRC{Pe!d840*QGO^cNvbfUXc#{DhcM$6tn|G|C)j)^>j&DXp=&l#4r_1AcrEk}{ZV
zowWBtQpLX-5WpZ)4F>81X_la(b~U#Fbo5`g0SrWp6z&9@oD&kKCfS|^HW}sa0Gn*U
zBBo~^Hrgmmt#}A+c==RPn=ZEX2kZJ<06E!yJQ4>0rdf5rc=IamXZ^ap;j-+eZ`%9Z
z^`O^vps0U82P}QGgNQbCZG!O7qHqHJ9@BNzXq6(|2K|sW4lX$PM9>d=8xg}|$G;co
z_iOb^<1ZR{u2)I|#%1^t#19jOh~9k=@tdf8I^s8T{V?J;5%>bcZ`$z&;y0~+PH$2A
zzt0B-jl>$K{0D)-iOQz~gEQBU1A{Y#&jto{v;u!k{-wyE#{OP*;OFy@L5;Dk;0EDa
zJ-K#Ts0Li>=6dWjtMK18^j8$h&|h~|m7OWw3HuQbQ!yht9qiZbGc_qRGZnNp+ChFF
z4Ga^M2-?g#eIPJQ#KRz-%V!EuW!{G!KOEZZ6;=t|fzJCd=6y*DL;yciX2gx`u@8NI
zRJwn1)LY}AHeXXv-mY4kDUPbxw40K<Rcz+%HlX`(b@f(&(+j@Vt84WK9fgtY6%4`_
zDLOlBk-T)m7RkEQI`$c2``rhkq<JT%z=sJMam`~mue4p&M)Obi&QJd;)ebl0cI=Hb
zjYg7KyUJJA^f1zC+9Z$1!{VenF!w)?1hao?mDKizrBzTlSoDcc=L<usM#(VJ@d+A)
z!O7XirpwfEoexmLgoT9iajyixM&87k(xM!%_w~<5I~C^P^!nHRaOd|c8YBwPTDrBD
z=#)d7`35l67y9sn{+%A*<Wu8G8GUTnrDb3OsR!&~yfdrRK?|@jKn}6_R{qa*?%aPU
z?!5G;&?~xD$#k{WK-gdvr#!2R4hFfnYdUX+dd}Q^sfAgP=oC(`u(11B2^n{2^tHQZ
zrEctx&gj)*|2w4mroH3GLxgM8m$xphvB^&ty*+jbPg9GMXZ-Le$}z@6`{XB<Wgpkw
z*_+$d`!3@{tBOR(F;x;hF7cOPbkKj?g1V|4ch7}kcsOVWd3@Kb+M2!@1?tzc)a5}`
zljoK!eu*y5z7>t|@=33>R-m3<-_@5MFk!Vg0=x8a1X&A6QmPLFvbB@%pwb5I6mYBT
ztoZt^&c+3ByXqz<XO88uanGgS;Iekqj-QdNs?|H8Taq*9lMP2!k(~-_k3N69V^d~*
zZr`K2-1Pj`js~WxwBm<cT3y#7@p0u+@2I%ux_9@cXqcMY#$PW3y>pXJnOEJL(p7mq
zVP`paoa@|$u{GB}iVY(-%Xsfx=hDTMA0Khg(zC1OSoQ7RjvpA%#6HhgeQY9*Ywm#^
zk<sm4<L(S&`=dW6R2aPYZbL8+20V|>#^MgJcDs75(NWXU><D28T%Dory#&1vsj}(G
zh5WJ?qTjdom;VC_?agYJu^c}X0XdhEZzU6#<p)B>e-tRa{`vmf3nzg;UbVX=t+9%s
zOhsChR3kTiM(Qd^r)?Oie<|dyP57Qil`8j5dS0QF+S6PeCAz0KqYVSDvm`-J>Jlhq
zJq6n8?N#beiEiK#>06Q+!b}_P>Ul<4O{Ad$4VCA*GB-b|X>DO7?1Zv~rYBlu-}ETq
zIou}>e==R9f%=(Re>e*ABJyg=yfEU6jH<{t=r_rkIK&x1{5k2slJ*@eX<u>83^%>@
zMP98)Rv0PX-bW=R^Tzj-(5j<7;=N#!;p0p9>+SvJ-<QE09FyFi{`&qB#+3vnoj7T$
znNx83@%QQ5%kMo^vU$#g#xc#)pC4gDGxC)je>tjz8>4k&kic<u>SS<1%K5wyjt*zI
z2+H9{UaxSMWMCX-f#@wLjTMYQ|5xIc9ftkD@1KJOWva(6=k8i+bmENjyf2(~su}qj
z5X|$N+av_jGu{Tfj~-%^H<Mqrn7Fd>iX(FCo}%nDe#18oSr11kKxXc0qBY5kKM^?|
ze_BsV@h8i#oo5C=^L}JRod}!G+hmsp9UgjW?Oq%wv$4!dbD+*Uiorvpo+)MoZ;jM#
z^4g%8;@LVyn=oHdEJ~glDHIr*z5X~oD1=vVMGq!N)>h+N_pYTDWGxxcl;n?}pG!uR
z)}eyH`A=L&GlJ_VoUCIWp}phOp-h?We=JjSSmmI{?fmg62Hy4*bIj}b3$B>bgI=;{
zWPC~a34Na`yFnfCOQc*@BrsTF7}iqZ{`mt|RP48<xUwQpILi<0E91_t6hA-D1}@Ju
zMh+opk?{%+FvYkSvNP@4sK+mNFzkW-vxxLUY+S9dhFNR(ESHyP`^4VJ;8)UZf0S0@
zPHHa)X$gx<_Dq%*w90qLgdXVJz$M0dI4ssnk2RGtJ(p}OG0ty`@rpZU%pfnH;vO}=
zl`&CZUOUe_4r@6u@8nTf+fs{_qfYo5Do1fn7%S`!jKvW!1BVri&4}F2eg=`k-aFS*
zTtg(UnOAbq?(-O{P!5Et12MLve|Q^Xd-nf3jNP`{jIleGdnm?klQv^)FYV94*pACj
z!q|?>6Byf3TEf`uTMnf&rDvc!!wofUv{Fra4dZ9(Z%|FFr(Eam&f+>LhFaD5tWE^D
zNU<5Yp<96(vD^$C{M@6;2n2qihqtU}Asth0kb5PLDdQP%Geolcy^?{Ff3ggm8yh3h
z?p<B6)IG8^M^U#f#yeAkQK=7UX6}LY)lTYR^i_En6IBe271WgJLD#{XIm{B)cM{em
zD~s63TTc#L0S@nFC`2D2{6RTEM=fY4)?kn<FXjg4$D@}q42&44ujK=Z89T`GO!TBG
zuA{h`52Wm&=fBiPcmy%Xf1fwWAV7o$<-&H1##+*kha5A*sL)K<1;4C%Vv$YogsJvX
zEvjgx=ln>Stf$5^%C9)b)$^#^L;j<u^_|H=gJKd&w=Lsfl4wV6g(#;~Pj$fZ2PkUG
zjFvscaOM(Lj0!-^zHR&MlWQ@MOD44_F;E|)hnDIQdT1#=qKEcLe}3I<=+X-&y;`vJ
z>ink91t~^c>9<J<aXe28W7IpBgpzU0m#QgSe-lGuRJqRaEOzX!#Wxnu!WLu4us#Qf
zadSSjl>I95SkYBAHuqe2S6UtS6=cs@Q#bpWZd$_0j^HY&S|&gb?3bF1w9U^LTquIA
zSui-6Au+=u)F4(Re>Y>9)ubVMp{-jRotpu<m_bPmS|~5i<zt@<ndg<#c#=TLB0=+9
zI%p*x&73;ns^rfplp0c$UAoquX;=tjx<1yO!xtU>cCt)O(7L0+;!M+AOx$x(<(SrU
z6L(e5I}$8lp4*<of#pbU>GT|OQ)JqfW;)5h`ksn>JEW!Fe^I^DQtxH(RSHk-op{hs
zDUR$8$q#8A<krPT$q~0uj=_%X_;N%1%~uOMDJ|1}Y*AW?A_D!1k~bp@g`EVA0bov4
zbiO*vdw_EG7ThN*+T_M{$6jVGt?EX}9<H<=;RJIFuary;R^P;i8heVwO)?yJ>iRfV
zr-FCDEq5Eqf8_G<cAawAD*d*Lg>}<jDe}m1FJj|GE|j`E_QC>l%G|ZB=Zk2}?3i;U
ze%#sLZHKxkneHe*V#Uv_i+o6+6kfPyU1Dy@-d#g%BX>mFw}qY;m3jlcsMB^;)A_+o
zIkS36USpHhL6Nr^`8tM+j=GR!>SFzUyL5jU$~CZ2e`I)DK#9FEr5E_P*Jy;(46i}q
zkGFj#$D)kOp3)H>K(;Gge8a4=D&@CX71gdCg(bL*V_pBbSlV{jE0)%BTn?pi<8@6j
zk8J2E_Z&9%j#I7GKYSToZAU(l8`HNJUuG-8tQu8yeVi{jV$XItu&s$hiCIr&dKhhT
zA}@qve;jJCS&9}Z+uI#ybKcxCJJjCDqFV+V%`F>i`|wfbrkIf8-qN4I$ZVkvgYuT~
zd0&JBys2ku=PKdP?cki?*?rNDS|lB(XQQ1_s07(l<=3Vs-0bLasZjPPZW#BXDZZh+
z>-@-;cOfRcJ^)f4Ntm)Xa0;tsbtGlVcfrcIf6ezO-*>2cZ$v}DSMPp8eHdPekg|EP
zr-$obR5Jo#>=N)RYsQt@Z?V@lSB>!WYo%)ksZZ<B8?@K?-JkUq>CF??<>usZO-;D1
z=;5^Z@h0MM3Z5FI^;R|~Y3V%rY4u}9ua#PQnYnb8I%U&8iqsudz0%#yqj}-YvH8ls
zD*xmy&&OF=;o;2a+)y<{QT8uiA7YnQG41qS_*(p*lJ4{E{pCNzSJ(cRu^c}X0XUaX
zH8CWYntD480yQ(2k#i>~e^GB6Hxho&uOM(Qy$5s284k%g1Od|K(hK0?f;a{4E-(Ve
zQ5*MME3oAt_wVl;Qd-H8+sc*$sX!vnu14exIWymU+>KnNxZon!T=1DqD&#0NDJMs{
zHc4EnrAg{aj7=}NBJH;&x}w~^&93%QYIIuYkx94?m6{@yVoFW4f6qR65f?@KUFcTo
zgBC}j+fk|`Ds6<yFfNGGl+8@mb`y4{+9wl7LelOAh<ygdSS8pD8ZbbOeKK{Vp<7!W
zhxuUYn52!M!7)CY7##~WAC(-J6l^v)&6=BGus;QDgDDn^sQ*k|O42Pa5H&U%f=j~P
zoCvX$CX$0w&3BBGe{#0j6l<M|>LU{|m<DaSm`zI`Oq@ieCT1txoYzpCNa|BmC#4UD
zA4_dUYGjuq^(hiWX*NW0Fd?+rGQ`vs3dGufR$b=io5ldK?TX??maf3$a_&=5bh)$*
zP{Lhv&_s46a92pKP);@*tS+owH@lLlD->61Y7TL9HFPhqe?~Q8FvaMQ`u9{J);0i)
zuBlI!)K!U1WZcc6O@^gMD0XwjRf}m8v%><lHRa-9a-X*%)>2ncr&9a4!D(%wqX~g-
zKQIx$2ngjSThvEo%$J%>eZo{sE~b;3%+yD$Cg(P1!kv&pb4IX-LAaTf5gZ|>CQFk`
zTM&tsVLiSrf7mck+d{#Nk)+LQ`x{By_=1rmTL1=2`^Y^2&G4A8x`@h_sC~J`Wa-XY
ztjC<1#i$N37Azpto0#>{<Hv6JZhAhwJo<QYI&qBJ^2!Y_PiC{p<%Oeu`S;JW?~i6D
zvx$rS#`EdL?D6B_d9w(~jGM!A6p(1~*DjEWFroH~e`Y>H4i`_J3@<OIr$>|7o8jfN
z=fmsC=h^X-C$M99{lAYB%>45N23^c17qhEYp!yiECRfwz%hSo#ylC0IA13e5PQIOf
zeiLws+(1pEV~lW#-<sOomZujN(>_+AZo;ECy;hnpy;edx9&KnpaKpFL%lDJZHh4H5
z{xQ55f4;*7oIHj#r?WSa$P4xpQQ@;igOInVPLg~~di-cOx_&pa0T0hE{yRJ!hd-U2
zO)rL{;m21m?Eg0(X0wl1e;bCM&t@O4-}%$&ufs3bzkWP;KlwELa5DSw&yTYgSdhz~
zutWZr^Wok3^xg0&dI~-cSC^;5kF&G$v)S3?e`=V1oLpj`p!Ju-a&VZ#U*YY_^4`%k
z%GtAJ*xWVLbdPenbJvhB@c^GGE&Y4;rh}qw`M>`CA4?T)ky23AF0RkdkAF+8_=c9)
zMZni4s+7fBSy-y>c@9-t&!HObmqRtL<WS*74wW&Rii}bor7}u&t6a-0*rjDaqPF@H
zfA%0Ca}We%rm|ha3H%EN@^k+L{-HbCiGM#`U$oJ$nuio4QVyT8orft$UuC6Q=IU8=
zKRm4FA@Oz|B1e4I<sJf%eblrU9&$1dS-12s3Zinkhv<k9wfnJ$(xO!IZ64k*?{;@%
zzv%9=`eKcuU<1JhNP<t+T;=3d%kl8ke_{Be;rr?9>F~^bQ$@2-P(@=V){3iHv4SNv
zj#y$0>=s~J(O95O=HwYTwt*@bkvHPDfvO>SykWqeW0>C32kQO$K%<l^7^Vwss&$mc
zdm2V5juMU1h<`fbpWfC@-Zk>1`T!1`OGM<Uk$P_&IM;y4(>q1py<j$~k06F4f6S8j
z%)FBXznaXh7X2*KL!@<4uV9cu_zQ|osiI#eU@y|0))yE22~-$0t|V}?5)y=4R`;G;
zEHQdTQ$>_LHLK9kXuhu%cK~9*Qn5pvMXM3WgKqA2j~QK5id&)+vCm7tm@!|9JMrrj
zxHP*wIlEYfaI6my!YKxi9CcBnf45@?Nx@S}jsg%WrS;$mC8vN-u3TzmA$o`qX7YBN
z5DOX};ABfAH24t>P^#;RL`6Zt@sA^iR7)(|RjFurY^Qzmod)WoI_||rw>ZvUkW<}^
zoa#E`%(1jiWCPWtXctsBQMI;>l5C<H0SFYnKIL8M5N`AicWoCmu2Wt+fB(E!yAXBR
zF4#eC2YNflT}XM9$|%)Q(yb`3D6+~90~c5vlFw3i66T*yF3(Qholln4>>&V|0fiZf
zV2Bh%#%d<;v?N-qBUw0*a7L$t?L$SWWjiFy1uMKlNecjEN0f9#LS&pRZ`u+Ivo&tZ
zMvkrR-*qDw+u<EQ1Vuw2f1p8*j<{OyW{+_^OKeaiHLgPtC9t_rP*ze+At2PY3LV1V
zc7ne+AVp6Jinx>YeSdZ}JGnTWyq!(oVy$mWQ7iBEfa>>I=&Hs700bD1+PH~o${B5d
zZlcO!gq1f^MPIC?Y@$j{m3PK%hv)Xx#h!Y#3wu^G#}IrY_D#3>e|fZvI3FOhwQuv^
zWFrLKwL3!}hjr+qdOzsnl_w^1v*%9-7ZQ)s2#hoWBaOgF<pEf6dl~^&9Fwm=UrX|W
zWN+*gNw-d!QpUtWQUeum<d}K;t~bRDR6<6mMVmd=RS-D^E~7__`My@%1+FsOr5kV+
z_o(JX>wWbBiaQlhf8k=U3cz7jdj#+8ku#ysr*g}FGv{16fj;L9G~iTeyXwvnMNgL1
zoeK=fd$ZC)2Zwpfv)st39Wjk03e8vES?}}q{d68g)*|HXT+8e!=4Ioo$j`EK&&Up_
zzo3RX2aZ3WhBwt#&^GfXs>ImZuk+mU2H@QQ$s2%o6AF<~fA(HU|E67(TcGg4C}B$L
zxR~gExR_QcA&(fyBL?z_fjnX$kK~X?a>$F5W_v?rA;s;nk}5hYj!sZ387R&oUO^_o
z2SEAPN++XMCdM~4l^7HVTen9_ir`UT#8$Th7%bK7#jZ1<#H=^xsT4`}DhvrouNIW1
z?SK;HI_e6qe@pCllmE}A*I>=h&Q9kHhzuh6As|xh0MHs~LQ4d`7m(sv!TZQ3zNHy4
z*29qt6Bf9JyzLnrAlF-FjRX^GoRaLtD%+z#k%w62c_7=91GdMXp*=9Lb56~-_MIhE
zJFizkiPV^QJC=$1?o(O5^$-DNT4*m8xrBqm8Xqu}f3e2k=iqzEbFi2Fd<lBbg7s(Y
z6~0{Hst1S4U-S!eK>8gHzoIdoB3EF5@<yuQ6vo~*P(_*XEE}m>v4~+KTccqF)vwX3
z+-N9PdX+mj6e8;lh4B3viX2uN3K?4g87a5W0}$v1FoIZt!4h~hV*ye{Ag*JDz2YkP
zPodUef3JSimb?`$NPDe$eDjAeZQ6?@18yE1^6k&&?og+dy}jE$@<pIA*ZlxPHr~Oj
z&sJONN!o~b+)!IoleCGNy$xFnZ=xDepK>i5siLh7VY5+DbnNx$Kg*jg=qT39N{*Y9
z742^_udGL=Kf;L>jo-kD{{vb=eA^0TZe(+Gm$(=T3<NPVIX9O9t1c6lPa8tS0q(aH
z974(o1Z#a0lb5j^KNJBum*K206PF4mLM?wBxe>kBSL6>ghf8n+8U=Tix&`>Afo_Hm
z1Hoz^q`}Pi;NSP-nj|x`_697Vl$}g6HU{GbgXGiu#HY`@!`Iy}Z;B?%oEBD2U%yWR
zA2|LO9F0a#RtRR4n7;m;{_)JYH259xBhHn=&*E=~r`|s);m|{%hd`?k`zK2REed}v
z3Vo!;ljA4xx6N&DdAz}R$B6<3&=%+@j`;cgKfnHc$FsnrbjA{c3`IKi!Z|@Y0!0n}
z%6qhGv}*7s(9<?Q1)eYaC$K^fqS_a-YRE^BOZ!X-WBD8U%GftjiBfIZPIAMv@S}A6
z<;*C@jJ4w^ZHjC`rof$E_ewt}*`j}($11(e`css#v>uaA+7*3+MIxm@@vYJ?c4_Ps
zD5YMBPqvrOUykkgV5Z1fw0$Ebmo014Gx&xvF8fB5EsS;R^n>JK7WPs+7g~X*>onOv
z#jmqAI#Hb80u_2m5;XIfzXW?_iDhHS#5v5{!uO11+{rRe7s)K_Fdw<4SnhvVRuP%R
zKlYPJ=BQ2l8to`+U$;+G4(S=kW(g=;9~NKLfi!%rE}vfMEP4?#3b<&EqP!GwY?%_u
zYMO14O%~a$irms)xX-8cpdb|!8p}>{Ch<&K9k^catt7Kpfqm~7jJ6HAEmhU?rJ1;X
z(nPikZylm&ANz|;naX0EZC`(>6pp$NWch2BY5p=>i+K}F5FVe0x7V+`KNp74<e2w<
z`u1}dJ`jQ$Su+_Pzy9^F>HF@>NVUZkClKYhF<N)HqBOHcOxk-UM8%bg3&$)hp?m%B
zulH$a`L&ry$f+fb6tl-Krd<c^{DOZsq1Ve`JhL#tg))Q)uNOAZB@%y$b+_k-ehn8*
zoP*ofI27C~BAA^K0?}@Dq_s?Vsu8Sl!x+tu*@0fyM;pVW(sguPIID~1YHK>foS>0|
z<m=Y-nklKq-fDHBp#3$=V-*;S{9lx(YiZ<>QZvC9@&r?l^F*?Qy7c$9Rl@jHeWbUH
z<ifsr4zH10am~DPm&$+L&HpjWEH~v;MleZw;Rk%_Z)MGE%_Krdk>j`)m@=2mytG6o
zX@Fcy=sI>tyOtfKfO-c2T^s7phxrZ+T$6##U|cANdGal&ratR%ybTzg9%=^mC|_Co
zC4$)PNR}a2BcIT9X3s%x-}_;ZTYdIWzcl$*&6E>0S>UzWNacTi5k7@F#1&Ef#m}*s
zOtX)W^K6&=JW6-`2uOx}G1L!`EDV;pxf_xRVK)Sx$3gNy@eCx7{2v3!7s`+!dFU-S
zAdw#)3V2e5blfhu_khQl@fga(c&QV`uI1MdHa%aPV@ZZ7yRe3ww`(n@El)w742W6I
z)azZ|0!V;Y+Y5hZ3ekd*SbD23JCtVQs`H`@1%@H}4jVwj#eIIRg6FDBeN1VrQ@@Zr
z`j=`#3eV3^SyIZNdJkNZMylt`{{vjEglV_7!Og&yN+!`Cmn9R6ID;iHYFB48%*D07
zak5Ag=Fs>B$5e3Xfp#t|%pTW(k3Mgg_*M%rCoMsQMk#;KS>n8}ozz})j|5JSTsDde
z1m7c}TbPg9^mE@w`PPftLRPZJSYk1wrDjO4<)8Fcz_)NyrAXMcm9*?_QHW`ZU@0N0
zU#YK1Tny;hM9NC+0I34H5==SFqR0M6-45u2W9|UI0vWoo+dISSK~(`pz;vLzwKF4P
z&M!LqxvYN`O2WCWbhRip;D8yapTKux7lNxZ>minY4>k5K49q!G+@0P)-;g~4*Ri1*
zsBan>m`n><?#bS;vGd%H6t^sk?Vm7m!T-+e&0%8^a>C1GaM&`dg$+(+ZyPhJoM(z2
zk^^IFWP}?jZf>Ly5D;vK-qC`5_R&yNjmf}aqtJh|p%rjaJyfZV;5nUo?sqyyiYb=~
zke5c^d8_nMFr3OUfy^bjV=n6q8DuK;>|>&IXokxrP?cTu=CsvhbgKq;Tk=;>hBPBR
zSH_Neq=nDTAq4je4CVR2sYGf*jxTu5XjB=~7a(|ZfYu#<l-#cJnGI0K)xri}r@P1#
z_*#FSkT|TGziOomz887P4W}EX$&O2Xy;7dDx8^voeG@BMoXp*%U<kLvEJa0i5=O{z
z-K4hjuAJO+@aB*S(x)b;JmWq}u`E_Di@bG*O2}-+A%IxM5hsw>{5io-KTpwF;lIi|
zJ^goz&DcxgUnpQ20yg8LHpnWm{&&Cp@&14PZ{L1{VW#*0gsw=Au*DVLO+V8s5^)50
zQqzz8jx}z{+9`mSG87Ifus=YDGO#v$n|6fQB$Red>yOat89728Jr9m4nMKFSplK>H
z1d6q_@cYp=7O7?yQKu?KcnehW*zk(+WmI?#zxKU!#i)NzyYJKPOSY;u1{n?_zYBk5
zS>zH>gx0JokU&`_TFaIu<(Tj`D*=(ch4*Z<w))ma$uO0{e$p|^1^%kss?MdQ(YrK(
zR-sUu%5!Yru*0A$TIPBa(_tHq-FR|Qk*%97JpcXeXCC*CQJ<S<qVp8ckhGbw+*Pc=
zfmTe*wOd_JxqIo|0^%`7)r8%NO9_AO?dlo`fZdm=5hv;OG%NPyVq+jJwpLwWj1gmJ
zBZSSk1*u|$ji8ddV;8_InXS#C_j)DhdvDIxVrwrdo5WNN&#eV4%-pYVqjGLzwvSoU
zm=Sg|rFssZH<)kB8h_U7Wo>*5dQHnx|6?cr3`t(3T<A%{!q4e~O|Gk)!XJMv9NcL!
zBeG8=%e>`T-YT_>C?&2*xFTAe&$%v5xK3AckQ*y8>JzPrG@zzOd{L69MFpkv-6a+7
zF|A*8=uRN0q#^csJ}DS-*OleoEUdNVJ}=$0+<%|?ZW)rF8ce?6-j_t~?9Cjc)O_ct
zu8ESw9hYU2xKRl<vEvOSWy*hvOFcWwS#Y$C8LbwR&U!K)df4I4i`%&KlIU#~B~3S5
zvs5n6u_?0@(^AuyC36aC%4uKJ<E6H!xinPnJ87%BD>kKXO+ND(QJ3VJ{zA>V)YR)R
zmUgQir?jp-)M_bD0{MA!Q{l42HaM^(JB~hcaI)!{NWdJ@#gSTb8!mtPSxt7cXV~Ea
zQ!Hd7mWxt(=5772t<jP{U45o-@xU#NbMhL;=_`OvGecs{+*+QN#Um9Jeqo&B43k%w
zS|N=Sd_gU^F-a9`czZ=%CEE@FSfX|XLU((WG?bWZbL^R`r1#!L%Eb{W47kt(t0M!&
zGv$$y|3m5{7mAPzBtw7iYcC=dMZ$FxuCKX>lt&6r%+IOEr5BOPXoqyz*^ju0lq;bF
zhpkF%u1F8PA9)dp^RNLMIegRrbU=&0MI<YkmJ9pda1lv5>?D-9RPJjpB85W<<raRh
zauF#ndF&gpad-D!L@L3FusibgB2qecYIK-yd)^fLOaMLlpwK5m=1=BZfuV1pcqa6X
z{2vnfE)*e$zM=Pxp)br<<9b(wK5tu3yfX4RfWGk;g8eOluV-*3?)9BNAnf%FJ~-+P
z%x({Qmq~qN%o{vlJLFl-=0xsfz#E?kOaB938Hg5_u^c}Xmyv!h6aq6amq9lgCzsDK
zLZ5$gz5Mgn-(CbU{M(1|N8DPcH{_{ui<}zV&0LYXj@%g+N6KFschW9?+6bSikuziC
z@5IGDSy^K@5T@*|+y!hn(}v*Aw3%<jbtX=t(<mSlN3U_)x(3Bypv+0^;E^;6&IF=n
z^0@e@v##+;rOS_*p98JD){*YGWq&5JEC7EGD*?zLH{s<ZfTHZENXW+Wph%Srn-5{6
zh(;AxO@dc`O9V*mn)sKNPM_qL|MHh4Rr5wqkVm7kJR@&OhBD5GW*N;ro*<cL8qM-1
z8s;M9nM{m+C#^;~OLNZpP166M&y8eHX1z({QEp<>`0>i9(_d)l^XTtC_hoPNW}SZ>
zWsrDMq}N+MXZrp@=hTB^?dU;>&1pGNcVuHXI-{@hy>-&o<yVvsoU1zqnYH<<le3Ap
z943<VJjn)KL3xybGS*xCwn4}&pXC#jQRUjhCTy7OqSI)$Ngd_!QMgPY&6K}HoP(zX
z)5)srqBU$9Cdo&t;#N+kt{3CaorHhLgOjI!6l<}gpN{hDeTR*bZ%#fT?Iyw9KYO-$
z84Qa+aR=!`*|+#g8@<kFpeK?nJjSP69W7BMS!LW!oM|Up=IAh3iHJ`;vz>T;5Y3Ac
zMdpK1c7~!V^Rz;xoWz&?TfF6`jnO*)b!0EZGg<o}tF}WZ@}zHek}P$-lzM-n`-rrG
zp0t&==!$;7t*ry}CT$-rD)))cSEb}(zKJsw_xL=Y$$*}FX7lm;*6bls6LsyioILSO
z>wAiYrxLr_J}i5t$YN6zFSf#VEPj?^y(&>lhq~jH@78%`-br_-<z87w?_^9ZF^Yvz
z4PQu@+W1WiH&Sdz?9>b0;XZ#T1m?pZDx9l{)-BCT$V~}bxG1}1+7*`A#!>qHULemH
zbQnY_C%9EAWgd{4K6>a=e>vtvW}J7NE<27{I^TgK#_LTvT|;lUDAQoVt1_LV+(@=Y
zFJM^GpOxGw)4b~DHJ}u_i9gfMn&em45Dy)EX+V?#L(3p#-rumm!oz=rQKQD7o<=qM
zNuiMME|}gZ^+&hGHRoxU8Z;S|%6sM!W*HTY!?EXRgFpgCt%>WMRc@OXWnrrUXBIJ3
zv~TV@S`}(?mwX%#>MfQ=Vr_IzA{z14<tPVccPD{<6vqjy7O68`G4)m6$FQ_?!N6TG
za0=Dwx%w!UHus&SaCv`Cv)#l=<4yHQp}BK5t6T@I(8$i&ApC!@x)y2++KsyUCakVC
z)=WlD)K^%&T0oLtarirq671%k8tLD2ln}5aUvM>QW2g{V?P_+ytVO9?RV{wMGAn7N
z{*@>fv=O>Q&{<olr90jTRZ?}X8NRqyE&{Y_^<LA$Xt|ks48?z;kt*E?&<}MWjMNA#
zs*oq}!?XQ(+#|RGXj;nok-idrPEgE{B?IX^2P7UW+i`s&DKWl^X~zh^D-6Ez{GNiW
zu1KOPN7q&eYo<I~VD2|SVwX%!+?KP4<7S8On}Tt1>P*NQmsb@imaCg_Zd?gJ=zzh-
z1@Y_W@YnmVm;ZlG4y7@fISfz#`tv0|&<Zw%u?;ohgE{^6@6#VIe;=vxPGXXcu-2H)
zb^rsT(I+Ed3tk;)GBd5ksr{3*(T%V&AC%lc+L5w^rLaJw9+jE+7}6<mt5KsuHZ<@?
z+*yZgfI8t*;;tLv+=1CZgpra|Ipjbe)IW9Zs1Y=wfii!<^At&vBgL?mq_SujFv$oL
zECT=Hi|;ITeu5L@HjB}4l-iC?t=+&*l3;#^9KrWQs<q+Y3+~!zd>HOTck@tm*C_WO
z9jhyx9Ftuvra_-q=(#*8Ni4%8y8=|T6<2_mUX|xy7>-Z+zVY;ErgUn~4Z#<fmDBxd
zZLJSWl~8}utF1F?Yur4Z%am1{LMaZ?w73Jv*5v79Q-0R$D*wH<stcUIB?THRqP?FO
z?F8spA(n(7-YU*nVJ3~4A~@(UlcvKKETkEl2G)4kfd&SWfsd3QOMPHHVjnxMF3G%^
zTE%beCgjG7+=}X|Ia*d;Q&B5!4`VWo-$LDUTl;^I+bSP05m-*Eaz%Jt!q_{L0J}|#
zR0F=Y0_UJ_kqM8`7#J9)jf$y;MNw)igJZGAo2($&WlN@d$X!wIEf;)6xwP_u5>HYy
zuC?drEV}Z`uDO_qS;MZ4Qp4^m*AA%7%?ULsJndDYu%*c1P#fahH+w+N4InMoodVX2
zUL}7<qY$v8Uk$80iwv#Ckf}#tr~m^L2Bt$~h$2xLAf<njAiI&@A3~1`Gd*VciZx96
z44CKsW+RGH#$X8QuC^uxP$;a~Pd-$%Cxy#Z1x|6gL-^EE{E<IXD>%))z`36;ONM%H
zy|9dUYPSz4>D;I_P8-Ob=ZNB-mT1wHQk#DT-R{AApUeD>O_+=v51KG#hMI+QUl(y5
z-+63^5bdrZQqe5CWTAPbRag3~W`Vw+kXwA$BG^uDtV2xPtF~_+6)V42l4?Cc(QkQC
zfSqLdLPA4+WJ`pz@|vS-u{<MRqw)*X7J9xSpLiMF-tI_EBS#;VPnN7(5?br<YD<5j
z0L(iC&Ha&4n{RIfq?^V8@EzTvLNP*Ln=GjbB$t=S&zxUkFn7-1je#PTRVMyZYSeBd
z^)Z4tinh}E%|qxQr*Dv@`dRC)w|ZzNj%!RrdZPS&(KgPztlhho?_=nfHl{aNcU)R-
z{Y?2<j@eVrL3drkQo%c{Xd&+NJUD-m-b)yH*eRi1e18O1`;WP4k!+SyHCV3*$!{X&
zMDs*r0L!-EN9sqCwSY6-xMDBynfDxWt(Z?%+(MfzXtii~TzSD=QcAMMO10i16jwB&
z+#s-#1wfGsYa`#<EH$hvavZiBs~}psl&WwwRC}g5Urvv5sGxG9sZSjLYjuBpyMP{!
zIr6QIv-G+I0`p1C1S90<Leo>S$GTKVGt2&1o*mSXd`;mMYC2tc+K{?6yKwIS5iBl}
zpVj+XpNk9svae;Mi(5uoYgubZeO(`*Vs2UJqq7%Sq?QSweWI0Hb{7(>Dc7}fkL_R`
zaVb9c)V>n&z1qNqk>Xy&_ilgFN{_RXoIIsj)qrh&8#Je3?%DO7F4HWO`bBV_Q?AyV
z4<`tK`O)uZ+&T8rLXVp9;X8yy33s5lCoD=L<Nl$%f6+RG9F3ROUj>kFy9cvjlj=Lq
z)e*dRtZuTrHsh`R8A@r&_1#$WUT#D}0hs8#*^T(f0os%xW3OP~A+&$KaNz+W{B{rF
zOsVDOGz>&nIh}kOjNfE?d$W|kPecdMZd0JXDLM?)Y9HvpLt{^2uk3G)fiIYtG+2O`
zF!HuzW71;ojPmMdKLX{e7CxUL(&dy{pe0G~_)&HMe#OzgMJ-}me|c<9v{HDDNghTt
zbG^X}ZrimRM3Y*}u6KWTU>E<{QtA^~r11#aT4c{BG$Rcn_yv|youpwY>D<s@J|HG;
z_a)v*b4D|g-X2}hn>p$mM(uU99l04hqrW;5+Di-#wQ-a<Q_FL{LmQwG1=ypiBJ(1y
z9MRry(Cte=--enXv^UR;@9Ab6pFQ=U7SWSV7p=nCmsah;TFZZlCX4<`<XiA02(@X~
z<H`Az-??LuE8$`FSD*u3qYo5!I|h`Y`q}!%aN<hJJ00&Y#QGpN!(K#dq28`=Of_NM
z6(_gxKLPEB=dNrc=*vo-zS!gU;{Hyt)^%%-xf<pw_F{c0$jo#SJC^Z%u!Q;c{41fw
z!)Dqm53EKjBKc3zuuJI;Fn{ke)4^M^yIFaTw4cpy<B`zh+kM{K6K^WzH)c2)DRy2{
zN@W3``3*$@CwHeL;8WY>Dk0E2WzSt7U)~T7-|xR({s&L<**=%C96uBRIhWuv1{47}
zmofJ#Czs$zGz6FZJVKcR!sq#ym_0&Df9^D6EdasGf4{#^!^kgRD#9R<&vsl9x!BTQ
zp+ST=0s}G*JU@sAAmc55T7?gN$8l>J?!@CbCQ+a`oM6EpNM?XC0hEn=Jv=%|BlbDJ
z>DwsP4g|{gLQ1s@0-q6DADNUelS5?CcN!y|gl8UCOdhj=VCbD-<cXd{uAIqOe~8xx
zVNfUTDBga>Gve)mNrZ{Ug~-<Ab&G@~2(aoxWVH;!OSfPaVMY@A%-FF4CdG^Mn8_5W
zN*z&_By$%^Vh|GIg_Ibng^^}W>>4pLYBnpvI{7SBzi%lJcmc-WQ-S&w&nQq2+$&Iz
zFi$W}fDdKDx`wSy)K*v(??I^bf5KTauDC_0QzA;g*#83M)A(sw`MTe;$9&0jBo}dw
z_E{ygy^s=Z5lX^>lE{wC&e$d2ZD&y=D%g)_5s{llVp=m=6PE&Et7Cay!es@<gAzS{
zhLo8JNR~u&=l>EDx{+3%ypoMtI51@;SB{eGu0uI$WP3(sxN@42QkveSf6pgbhD<6O
z%Uid8842}FRG-Tpo@b|DGd{F)!snHr%=2TJm+>);9!dUW9h9_!X>qVemZ@v0&&ODL
z^!%yyy7p(@1`N^y+|kVpEAG3Q0r(O|Y@T21eiWAyUf(m$8df|*U=Mr;0<W?l8-;gn
zX+ac}&l=9U^inM8*y?==e~Mx^3PmT`!U#_rmL73n5;D>wvKY!KA(@75WQ_(0BFJ(g
zX){{mx)`j%138SdZ&3s1gmMdfw$uRS56X?uaQw3ZG95+q;X-V+0ZeWdw3#MEE*h(v
z@VSw+$>$;eW?4SvRQWg!MjK(hU1$IWS8_cZ!(haidrI9v6iQwye|1AzXBSeZy-VtQ
z;_whCmA&4w4Sqt{I~#MlQ%;FfKA{F1<{2e9!biy_hB&e`<?6{98y#`2P85&Qovn24
zY0fBfbg5Lvj>io@`8<1W&q0X!l<Q90&YN{7ZtsDv;I(!lgt+AOCa$L%k<7lTJE>vj
zXNeC^Y%!;@?!X`Le=RR-e6t>>WPe;V0BXcqqm&%Gr1fLNud<(hTEtVPuMp&U$*Y+Z
z72WLZl@-xk>+HM9=QJuM^|z<I{ibI!-T}iu?KI$$$Yr>CEvz0zaW1Pcrv6r<@z=HF
z+;$`SX{Xfk+FJOU*4+{tq5Ps(E3wh4e`)1H-mz52PL&4@e?_eqc6*sbwMNTcEbYzG
zua=+rnc_2;_iZfMRR3L0U(5V59hBl7GKR%ZvJH<Bo0OT4*@n9j<m{){DECU95UG4u
zGQxE{KfG(%MM)7XOc0Df-&$-d=jnd09L%FHZ_6^U!c1HPm4Sbo5b(>Cpm728G<N#$
zl!k@U79#mIf2F*mWexr#gdaZr`~CZ0zWs`hZhHR@fyxPkEX>ox^pjt;FkbuV2fqSl
z3<~pks!te)6glBE_%bcOf14fvS*wf<b?^rbMgalPbj!?Cy!Hf^7jK4mds?mz@?JmM
z!GciPU`9XVKyS1P&!tzKpJ&l4J?(o2#aaKH9==Zxe_xrO)@e^5LCy<DhyT|2sW80u
zMzuW6LwH=HF5(yFN%W{=8>20~6%=z*Dm;KDudRTNE8eZpoJShE^AN6;SI9#JwPAdN
z4+E_@Qja=eBL%wX&y#X0=I+el`u*Eap7)NkzBHfnI*$s2Yrs(T5l}S%3Bt8f`@)9R
z<_?2rf9QNbpQ=x`niKJPr#0Ie7QrpQP7O*kZI6|>uNN}}t<~B~7&4xA4I{J`-d9F!
zbh$DO4Bey{$wj{$jD4CNzEM7>$jQwewCaCp!ne5-2D=5uJTC|GJO?;L&yKyB2L);l
z4BNMU=QKM9`Po6+ZuV&op>Pg~nfqJ|{umQee?55QkLOwITs5c>TH0}zdMPyf!iC*+
z90;YxW#sY9B!>zY$61u)@a7zg7g5SiXr~MtR)n+5EOIbTEq0~|B%SrT<`61UlXRC3
zaKuCySA!-7?xVI&MROI0D?KSvvxlgbf#RA)SKC>$=xUv_MgME+!T8L5Y!2NE8qFr6
zf1`*cf-}d14cWkEBg`?AW92vQsmXFU)8SFy!5o`t*5@@H=gsr#Lt>JDZ=%P#axu0#
z(pHnvw9#gs<|sPJ25OCWSS9OM-EGfd%upPY=L{Zvjp1#RYfuKtGi)<TYaK78rf?@?
zQ91XyR)s3RW=_wpvBafSf~S^_X78@#f1C9oV8CMf<|p*ls?2kf9p`-YqgMn}j6&-1
zh5(iPj8pqJ1TKkelj?7T)Ew+$2tx5(G4XvhXpMLCegYM`V6k%B08BfWxuZF%FgFK3
zR8Zygz<?LtmGnS;UKA3_3@YyhHO}Zyr(AI2lzPAubi=7rGnNYUscpC0jnQj;f0*^Y
zStbvW4$^ElNUvx}u&DZ#_ZY_=(qYCe)o`wFk2q{q!$CN&7W+3HgSHZu-seiyhbtXQ
zlfhkvmR4S9W!KnZbuA^DGtC=1DCFz)`me1B03Xr+eumyEjtmN!eTM=Of*l&P3#_CO
zQQz|-%dp~p|7_g;{+gnK5z5C)f6)qbF9JsB);)r5tUFhs^`*VPwVu0<q|E-~cPn(z
zc$U<P_4A0?)!naQ<vnF(n;&M}qDU-*5r++lMQSmJ#r}VQ&*5I%Wq9RY+X`Pk>~yZb
z-ZbJ$9jT3;7P$?jx8+DYy`r~=ZpK{5GGh=bEChcqW8(`!+AAUL*5ad~f8|LIqqZgo
zx#dJ<d+Ar)JW&JsA-mXrv(vau0Ywu=x4z;gDj2VR)lIKD```hZ@?W|lm#z%EnG)(5
z=BvMusnqqFUFsRsUdhW9SFh&VR3qjv_a~Lvy=*$nA}!m#OQlUfvg^gB?ch;sicny2
znd<hUyjD)=dOP+HD#~kVe=$XOR~6;YrT@weN^|hFW!Yc9vfNC9poY{G6Wx_U0YF?S
z6wI)>P%y)k)Y>#n*S=An8&n}s-8>byjWxMRHH`l9WwThMEk>!xp@D9*C?@PR<@uU-
zd+j?c<gs0)LzjVmc))<l7a^6q$QY^B&z$0`M9it>o|(@>X`}MFe=RF8?6GOAsSftJ
z+wFC@<(*fV;Y%&HN^NMkuN4?5jZ(hGV>?$LxNgc<ALcS!<5u(UxtZs7%Y-Tb=DTri
z|MXW6s`(2D`W|-6gp~CJyI76vqt;r{Qc|fT+rZVO5{>>5lZ~-f;zTs=`ah;%_o574
zN>%3tvZeOd6)rgUe;sz;jfLk^am4rkq;PS%HhdfCIddy(AuZk(Pr`I3B>k7R?MAxW
zSa{mR#lA%Gsv!_B?L|k;M%RnAIMVI@dY#KP;xTA7kp7A(?pj@5Dj+M}GIf`Y*!+5%
z#<w?}py}F&w&Tz@df>R|PsE|#DtBFvHdaKmv>t6o%+-8xPwR7YU4e_rs>xE5m(sOE
zmWiqf4quk9e_d2=F2$}AzT3m`7`+)k;91z@$8H?h^9;`g`TCb?U2%V&(VQeUHfiMX
z5cM+fZCWV*0~UoHmoYpi69h3dFff;)FCG+^0cSWTmu^TxHUh^Nm!?QUV}H=MfQsUb
z&-1_Dem?Ll@ax-Q6woTee6Z8!?}y(fsQ7V;wGW)B;LY^EDHb!K;NQfVv)VxyOd0F#
z^aoV@@Y~lf|M>iG!JyfnSd~wjNh5-vzUNm3_(<cXZ~5Iam_h&<SOCfOq&*W{Lj`JB
z3O-K<Egg(*d{g=jO9RthOJ~_MR)Q(-up$l66p;+rV=c|&eQOJgRAq=LbP)sJu$SBx
zU!gs<!dHCScd81l{yQDMOovYyRb#DnP{n(x418MWwO&G1FfAO=s6m#Bq-8)Sp$V1X
zxG^iG?R;xh21}_G%+NCj?3e6GLLGl#8I@)ZZ~=;njBY}>e*gSExBCjMJ~p4JpQDh9
zX=A6Ob%1j0Dk`t>i^`j8=gL`kQCBtSlu=QqXpAM@>f6I$pQc8ZG^t~q)u-J^fHg5@
z4-zER0s`8>mx92G`Jx%5%4EG@_Bx=Ya`0vl3(zy;$#xk5zx(_+D%Gc4a*2P~QQa4K
zOh%ND#r7|yoi24<>cQCbonMkwW5$COEY=9X4qT%8to-sz#Y2LP&}ja;3C8KDJeZ(~
zYE5rZtaBCwpE%hfrfloUmpVB#`LV9)%rOb0v}7R+vcMz>XzbTiSkl<~wSy;u{BBs3
z56r6(7D@PrW+hvwig38UMNxlr5(|Ejf~J5EUjzX56sDXWhJ`I~ZS(5sxuRfIS(1YN
zI`!4^4tQ*~Kr#v*dT>@P?dX<!*cxTQChz3(3>av{K$(nEAasaO@M+63ts0*^j)A2J
zAXRI9PML0L9QQ1gd}>t#e~OGDZW@F<L9bmdyEGzfJ1+*7r3fIyvq69N!ZcGUiYrr@
zv4yiZn!I8H+#q|>wsV$Q6<LpwLcGElTk6g6C7G%S6}e64<XGiWOXe!;jfAZZcYEBm
zK?_vUnOSM8FUT=9Ak{5Zlm?f|hNq?J?p&&w$p{j>2h|UZjH|}NJ=Gpb535L;z|T-3
zjB$BvQM?ZqFE;666hnXig{IA$ML8wFaS@{dJU0tRIuY{7c^QPr&{d4YQz}NYLTenG
zo#RHa#bF6H)k!f3`Es%Cnj}J>t0g0o!$9$H?cw9&@CIh+_s$R*8^FB`U`wN|!5=i3
zAnU>upj3jJ{GaK|;nUb_UeWP0nxF_gw-(&9ZdbNo(fnj2dRBj`4Jzj7z-pZGWHh=4
z^k3w3SJLnqygz0fEG_Wb$>REmpYZ93h<gV*96V}B>a{C@UO!v{Uz~@LjC(zff{R}l
zqZ@4%5gWk!mt#-nHT7os?Fe|w#H>TfA2<1+ffRl=La6&HX^{vR3+@6!N(~bUvxR%j
zzUr{N93#21(UyPR8ppoe4w(hVBB-&~N0Ww@x*w(Nwz|q*M!P<TRbmAmJ(6KGx$M6{
zPy;dC3;~STi@WA;##Mq92{Tm1q6AVxjBaKti*~lIY_<Aiduf-6UiB#{4L=tTd*qEq
z_mJD=-tXqk(Cc5sn_-l{m^Z^{S9mklegtnuRJVDvxw3!n=Z%ATYCncq#eL(n4)Dkt
z=M}pYJ%ZbD;ED<5#?cq}RgnX4O53qTbsuk+A=~9qhPBo3;H>D`_>I>Tv7%Uc!39=}
z*pZVEJ*u>rH%~=@*CDMKg<GLE#6--w2O_DEOG@oT65=4{{Cs@b<Xz869*p2HkGjTP
z3*My<Wv?+jH;$HX^z@v3_~f%cmd8n-dpWjPOq}amyobfgG2zrqh7}fT!L)R<@&6ty
zhF~9-moZL4FdJ)f%5m5w+xGjw1&&~9X&B(q8@d-U{Oku&q?de7LN9+5y?@tqjb>t;
zt%7(Cx|#7g$HQ8Kfpq1Njh~;7mdl)Errc9X!{IMtDMDNd&9}1SG3w`oOZglR;3%zP
zQff4(#+S3n(ZNE=eroAE%k1j5G-aSmipP_q*e2bF{9DNo2z8PvJCLOt(n5wDJkuCt
zPvb2m<GX~>`gscCr2>CwuG90pg-|cAtyD{!0;N5RbX0rD3(^LAZ4EK$3$iG%tfmA^
zC%$Mh=B<Ho8r+M%^$<vt<(mZ?nS-EunWw$OpK@OJx!~J5v0#}4bsMAx+k6DQ9WqH{
z`X1#@wVLyWToUJHGACr+MiN)BR^?=Mk;*#P#5mO)IGx9BwOM~q-NJ3k8Mo}s2THlu
zK{~;Q-7&AJG0YZYGEXtDZq|iTm<+VxUP^VXXEgY`QmU6aQYxrDl!~=FN~MxwnzF2|
zK?q&VJ+umXiB_roNUK<yXjP9MR;iWx*hwu3QHjv+;TGoioVu4=UF&<}@EO<o<CU!_
zYnw|B3c+0(q(px%DXE$Ro46H9UAMWd(+b6XJlz|59c4zBIlkK(5yrA}!Xn<nk-3cA
zZnrrf1-fD640TPX=yGI!o>R-E(44TwUn8T*r4o4!D;{b%a}CsU<n&9EV>Z=x<hGd&
zE8LX<oE_n3+-*6NkGsLu6Iwo9CA0JjKHZ&U?pr@1Fu#9F?0s?9HYqq>P70&+NOGz4
zK=M?whva~$FmmkmR+6vJFNWMZrwFLB;KR}aA@A#?i1)c<h4)gK4c_Or%e-IYu@fxF
zvcIqDL_Lh&BGq|;`C<q2!N&;xUgEcnx5D>1pBIDV6b;oltwcrZXR8`EeLDye_k16R
zQ8^IObX9)_op;+h$gb|B(Q))`U{S6y&{=kTi>KSyE+1&OQ(>HBmfwKoCgEX=!Fz&z
zXC(wHS2FZ#xj{yyjaQL_muBv9mnbvpgRbtEHbhLNjrU~8vpc8qbY)~&H(+5w%L^Q~
z$=Egb1e##>%9<?h3|T+GEAJdc-dtA@-EAj`q1}JeX2dtD>umjI8K~7*71mH<*2OXI
zeRFMKmxfx+gm6RanHesA_ocamu}v{4z?OPF_FTjKMUEv1mqu&WNbn7ByCfA>Sa=Jh
zFooj#-i&D5cn0XABa5FotHSj*xDs8&agaUOUK{vFVJpBn6XjCct>E^p=U#*DLzH5?
zb=7~kc*%#ZE!88{Rm6?JyT;?ld+A(9`GF97_|WQJh`ZMFh-pZ>Q9^}uR?#^LziuhC
zgpp<=?qaiTl9!8ja?Pfk-92F480y=A1_w<Bb1%?c>$|yp3*f}%e&ojhu6N0M1^8=u
zS)ncXEyYon<OS*RCT7#)c4l12z`*fn?v8&(C!QR-wDaQyZ=FI<(YOqk1_xg}0IInX
zBxq?CoFQzJGTtU{yXY&Vo{H0NtJAN$K)4TaOFK#}xl%$2DBt$q1}Tl~KKHE}?|V_w
zK7XlamaokDis@)1k}Z-unbn`S<kc16goK)+oUyKYTen`Mw0Jr?UZYg7Jp-9GuTfX-
zxHcIy{yop`Ta)iNq0Hy35dWI$O~3PoADylCovvK)b9g)0{_y)Y=^V8meC1?4U25p7
zzU!V%x;fK#xOPDypM(yA-jxAt{u*V^?Oq`L{{jC?TTGY1?L-s-FqhF=HWQbjQ$ke)
z&?5~+m-<se9Dn?Q<6psX(7?e8!Hg1zufGoeI&v-zJ{Np6=StyY@w>xMIK3<3aE8Dc
z0!M{7y<07CMB#|SITG-b<0J9A&10W=y}|X4YaJly{HXAp{{8jq8_xuD=1uq-KInfR
zUvK|?tK>_?!1x1r%anw}fBt$4KWG8WSbuPm;g5g+et-Dm?KcXvm5mc98*U6hCmW5x
zEgX#TOogA4Y&7x#C>xFZYUB&>e{I8f;(=*gUjO|4=YfoT`(+{DT4scx7$hgsKqffj
zVLkE~A$B7O@uyGES9x%T2{|A@v4g<p+98?>!aL@iqEOUFTwP}w&cNd=a7>|$%dDab
z*9XL-5PzSc5K-k?ss~~Avx6#w%>0J$xK)5Es!;~|n#XNY)wq>9Z7cs==T2Lum3<|5
zJ;Xik2!5X^jvQSx^^iCSimF8Zou~h2Tq(n}b+6>Ahq%X8Kl|1_yOlc$41|9rcRj>C
z?)uruSuA@5cAfg-euPFJR~PefqBIP^9Sm2@aDQ25^1|Ml+e5)P^vpPn>^KuW3NDSz
zum19_x`#1_6enzhxC?qHd+74snZ-p?IprV(AeeWZ=ZL!-<a0@T1}a-?qXoKP$hMF$
zg<VO3EB+p}=p>G5I82X(wIGV3nMmX;VoH!qSW4{qGekCFgYx<lsN(`(XSh*uvr-p@
zxqmm#A$v@5L*tzH%t*5x=;-HG;L8dHh*kRP@3)19n{`>$Ul5WhRd+SmY7oO>1@~K_
z(adQ}A&Z~GGqJ>Ew`9dLo*)HQO2Whuc-%It*U<N(NSRV}<R6<(0k}YjH4Ny+VMa3u
z5Y7(>)-!>|PM~iJHn}*mLax$-P(D_<!G9@Ci|YQ1i+@Rml{TJWSmEfoEK5WOMzF--
z>=q9YB?334ToPTP@{&S?k1T=5X_X<<*Mme<@mAGK-A1R9%p;V6&QuT7z+HCqTs8VB
z9*kig)W8E|d7n#`ej5+YF)0U$1dnF;-5Q5Rjd8(KXe6g!W_zanAitUH)W>=TP=D;m
zBD5t0zqC@xcGP3|UUt-Dd4(PI5MOCWgw`kRh|>9E?1=F2q#gCpp0%SM+MXQ|P9C%)
z<Jgk+oR<<Su9r<3Wb${FADQQsh;HxZyXsX-!;9@qO3nljK11Io(l3{!tm4|If2tK)
zFfA>?MLQgkXyA$b2)`TI;zfJr6@QUeEd@T)9+ARhN~;x2a@WI}=?GW`TI~DUtT8JI
za~O|Q`nHg{Ut^@k5l<<4wayd|)nh5XIcmM>tDI1tN9*9ZDd7gc{gh^T$@D<b)Inr-
z($5B;#cdVh6ZuahqZ{cQ6ELY2g%r3H*X9wqwV6LTtkV_LNsZ9fOMI>I(|_Go;!j(?
zrIr*yk(Z5xQ*6s7*A|Ldb1X)5dJlzj_)I14O39qt!pX_z0(#Oea2kt@5&ui8{aERP
z-LB*vFc&PjC!~qmGZv(G=s(4<a=o;j;a5^H>kNe=3U1u*jqOw3#E3UkTtb~cQk1K$
z7fm}5xLw*Eo7UmEvDBu-&wss;U6oQnlKOx(Zfry(nl5}bM;ljbKs&Z%^;S(grNbJg
zGz^fgMMZ7oIU9`xho*#H1|O#&oBc4l)%gk@ND7Ki5%UPDDK{$g9qCFmdM^g9qNatt
z`?<}II}^Q>y|3nJ@`$J-(Jz<(z?FCzbC)t2bME?ItL1?Ue@Ymy1AlJUV`(VzjW*UZ
zr3lrxSZTzZ^;e4i8FbddPV{fkw{CWHGTLvuE>IYTBwe}Fub65p=F0{H*c!B|t<e%?
zw};tIi*tl)3>S?IBQ2k0M_kvsn@hKK;ma-%ZE4%qov)}feB4nf3oU$8qCuzaT@o#F
zk@;7%__>;Yh)ubpGJi5-1jX5FZIS(Qys)9Aovy=R{o6W=+Gx1HRcE`|cj;`ua;G=2
ztFr{!b2@8szcWBlKT1QzjT#DA)w(;x_W}h~m?2qDasE>Zss!e1;?bbqt$tkX1mieF
zphZqX4)}3WaICpsk&bE#f7yJuookdi94*DkH0^`H)4<YW-hV%IOJI=C-d+e3Pc(0I
z7TxUIoJA<9fUbN&3uGz=h1I8W6k~QMo4cVxyJ=jxUiwe*o6vHkk}dVu6t}@d(+{hj
zI960VoG1BJ$zQWeLx(9-jb@tB7GDnf&8)_)N(~dYn&*?0F^yJvUJC`#(3RHIxq4G-
zp?VFi8p7jp3V)ZYxEemvO_juZ9y=GUN*9Wv#Gz{m&m;`WFbnd|ovyaTG5J<*O7qRu
zK<Q4E>Ne>p1;&GN^hAURDlI<}_GiuuGcE20MoM2sXU}!bj&h|MtW1{iZPQjtGGztD
zm>q3$(QT#(Cohu=eAhgjpz}+vrE>U^;E!1yaz)#8&42J$*JL7TZc*tj-k(#xAi)+z
zz7#$y6usL6&{i-X)T!rAsz!cmGTn9q5R_8_g<6cYR`Q|Y!AV~-DQf778K>HwM3)0O
zI4<%i*H0kti7VOLj_c)^IU7fD-RpX#VvcJHC72UOCuU@zRQ3C_t>K<IJy7)>gO|b(
z1bmSB4S&H)X&mHgpw%uxWJh`mLy#4cLI^%VbD5lR+3hNU(~>sU(~vMruY?-qOxqtw
zkW_z`(z>-^w?>+D116?pwYpF6bKd|=tBJKEVLJO5wAyiIrCUvft!Xb+l?!pX4v3hn
z+3i8>hkQ)bypoiXdvzXDNC=kGLy5JJ=WQhwX@9QXsdRB{>a>)rRoy(j?8yReDLLvD
ztOCV0mStL5u2x=2hV^#Jl5p2f!^+ya&Trp`DiZ9{^95cf0W(TzsW(@3$2exK?0o6A
z&A%%US!41Q*)30KxEOHU*9i_MI8qIi28)8|1;btJHvNI-)Ftk<sk#%Xm$cWHMv5&*
z?ti*opDz|u=GLBH`fuqUZF)6KrnQrUG@kRo?<nAOu!G;^eO^*q$$YRkLQ&;T6iHo{
zE-Gqw;k&i8BtzXs&vktR!(7u>>bfCzNX*mI`QyF*hpoJ?c4(vLuf5u2E5s9=x3$`#
zm%`kJa<Z{k^*r;YtGnBg&DvO!QS7|AE`NI|_SLpqmQM1JeKlUX&1iSZZK*MXo%ZJT
zN``**t51mM1+|L>HN{>$rmh;IV}?9Ezb$iN{S>#Xr~Wy?cRACzq|Y^;6E*hOM6PQG
zmFm9kG(FC1E^sy{S`kC*&Ewo8$5VfR2`$`_j}7UpTD}UmrgdleqvX6OzTL=UQ-9%<
z9WuJzF9Tb?+LRK#>c7#=<?AT(rMVT4aTgpVyNL2kklL6LF;_X7rt(_+AI?%mpAdpj
z7a4b5@na+ZN>i>H%SdTd8kn5-#Mov{qazKx*01)|w;7GmBwrLoy3qYfwXObw(plYW
z5neM{><gA^R=e7kEN70SV_p5T$bXm@udkXa&!pzcT3ff%eBs6(d{f~Suh-_;jprn-
zoEvCdZy?hXYT-jIYume6+O_TLHg68xmD?(>IGHsvo~aWW79Ct7+30(3JC)_3+LRQ-
zji>h_-czfdLYCN|mvfS|YmdDrU^G)w)7Jz<Xew~`_zQ9Q7K#W*B1<y$s(;Jp819ww
z6#C>oPk9G`%zo|`$^BQa@}($3#U!o36rTyJ2AWrt?;aXJGvRL~0cU9qgODfp^+N|P
zUosk$KwGAS`iBHBe7FlSP%(Y~BE7&!X?{}hXjcA43d)U3GW)?-YJ@Ufp0V;C$>i3e
zkbqGOwoP~Xo`w*bX=T34dw&|~JraUpX8h^p8!}ma+<lj?yqH5EaTvEl%ceyknt%%F
zx5-&txt>Zfe73Ykfj~=Z8Tqzix;>=W0thsSb(+_dBaHX&PR3IRZg;OKQVp2jp%z>*
z-Mpdn`nRSDh8<n$CqJVrJ)|JO-M4Vtwkcl{AJw+?A>Ub?s}kl67H8{WD%trScuazt
zmvVmeMfTtxdlFuG?>pamCOnR<q1{X0`GR^6Xd+Q3F!=fZi%Z{*8KVh76B{UrM?d+2
zyU#^dJ)n5?xm<PD&3Ct<%UDGjNdc|jdbXUI+umv>pO_o}4=BWz1ed|>L=*uqm%&&$
z6PFHPLco7?e4l^+_~k`1!%`Se<0dJzvEI_#2C1fhsdc^O+gDnlht(A-EsPD5btq^Z
zWrc>gHN+v2lZH4(`i*|Z#V8|dP=HZ|LR=#f|Ay9xZ+X2&=ill=Z{W5vL&*K}Oi^-K
zuBkn8<IYix@w4GBad={o$fD}V<?fN$EyXGu6*_<L{CbV@2NyLQ&PJb@JY=pC`3uVh
zl5UecM6gjL@%451_5Shl&&xv8-~>7|{qgrpY(Ox=x-cmv0^s!TpVRl3Uq`A%^TRut
zZc?nY)|k0stuV$<mS%NLE%(Weboa)KLn2rIBomu``PLbN)6wk48R6&ntecL9Y&rvv
zk+**f^X9X_qj|_Tep8#gp1)WbErM((_uv)a^)TlOg9!Qv=gN42xZu1x_T@Ma4up~m
zeH9}c=XXY+RcrIlWs$)pnk~NPZ%uM&n=vQ%i@jbCbFR?|=iLG6k(CG)7V=iB{6u}P
zIU3X4gl_gW<Gs+K3Mi5f9qL?JU=ev7dB1;=r-oU{+im?So9o2ZEa-lWQGMU4h`5Y!
zPPk$&3`^^1tc)UN-N5vQg>Pb;^P~?Lh@j!YfrV_&N}Up_7oJVSyQBg}X)=jXE^!Qo
z@`ZlEslqO|q<J>cb5{x|y&(7+7Z7$=onTastG;J86=w^N$s8$=7|<h;S6*|H+dh96
z1fBs#E_OV0ixVT2P+kp$Vje<-q6b{gpj8n~Rb&dgScSFlB4-6^Q~5%69m__ajeEuD
zZ8rMSqc<Q*lt90!hE#E!sUa9IY*zc~Q$W~FInK!8ami6;CY*(lQi?j9kVG_!?}(vo
zMOgwd?JIDSo)T%Hw2rIqtc#N$!;*hUZdAu4nO-~4!sb>bX2+Na8@i)d<y1{x^#a`B
zw!2a<RkDY-mKtWQmDaS9OTW$pGi59R4xEPU!<_RfZ|Im-V9|~%NkF9rxQ5$$4j`Of
zP)j<s@>JpRKE8|muu9=RNxn9ok)_&k@Y+eRjwwExq|Wxxq(6X7`TBF!Lt1}D4pG29
z9M_S2?Ui{hA&!&tWLejBt3pfOVyiC7ii^E^wk~Z(%hz@_53Mc5qw>fN>ANER75Zi!
zzZdlFsYxvJn%ea+Yv}Y>SB0mRE7vl|XP`yrSAjM$hV$iB6Ry((GhCFr!t`9&l`_LZ
zg4^702FMJywqv-fW;y|6s(XJ9S&cxW@Bj~hE*PNHTaA>qpo~f<8Qd0>A>5I9lwPwj
zMsl!GEsx<A-;A|Tj>g2TgQK>u38*jK?3_tqio!+9=CzI)3b2m1RQ}YFovGg%+2wjX
zx)mrHZn$5`{M<}B?N>_<n9i60G&&t$z2scae;d(YU|rwR2*6ok*sp&yM&k587mA6`
zYUlG1owZh-dMk|9z+N+byJ?N+sgPH`T4PzOuQX;-YUPlifJP}A6;kwgZFDZM7GL-F
zZLAW;!N|ibhhN%7UyS{9HGa>x(FWlFYjGASy|L6J%lz7Qg^uomC~Jhr#V!a0N?+-{
zzYQW}EJIEFBe^=AeffXJTO&d4ZH*umu4u6p7&j}HCcBMI7fUFHZlwIcwg5OMEaBon
zabsJcY^aR$FE_>x@esW5Uf;=sl3IXX$l(9nctSEMN2}JPS849qc%nL6dK^dWZ9F}q
z!+^q)kB<Cp=&(q8-upovD)O&zCkIMN0miIJr`HsgvgFcbEy#b^N^LrHzb}l2Y0u?c
zU|8!eP|60Fy3edkn=UD)|5nImlDsiE!|Gxi$_PqX7v+7l%J<_MwS+Z6k5u<8j?fgS
zOdUq_@mTxge0VEhL3Gowq|gr6SZJb{Z`4=4T>46)3i(q`3MeSf9dXk6h({c;!1KTZ
z?5k{;|3NTimEC{GLQ&bg_vS|FSgLWHBSf~U?gjC7>Zlk^k=MFs874c(oam;Yvs&hz
z`+$*rf3Q<5`z$-QDzOHX1!oj0T`M?7>Z2&jC@1?J>lhdwg%S?g1LZxxV51mDJg6}x
zogf7)_}B;$@j#R?WQB*rSn`>(l6?9x@=C<i2EX15rR;x`Wu~b-QK;{fnU+Cd@MG`K
z+jT!Sht&6Syc72w^l@X1NFl@07%^{qf?i!%%-y|NEbJ3O0u>5P5HQZK_vhD+GR|>O
z6wyvXpbeEb<k2@k5h)qd_~f)<pn3}y=e_HmRM`ns?e{X(;<QZ7H)m=|n9AU6JN=nF
z?-?0i)2n~OE=pLAe&k)r;UdHBI&Q0oBa;xphlO;)yzCw>E!;EX!Uiae2C&a86AR|g
zDI3!=i$3!0&zO;5wBKTBG3__*S@z4t=s@*GF*@{quKk|NR`!$aw}SPY`DYq1c<Nyn
zDspWX=Mm#A1|U&{;1?$5+TEVPXg}j3iYi<=N2q_aJ^eUgPXWop0roe2j&G-IQ9R^x
zY)+q7eo4)hQ&bJ8o041K1$+KAFs0+E&I7l=6j<RM{)k&(-8GJ+y$%tvUg_vgfNtCf
zqnt7xK+G5LgRX>umjcL3f6SGzSc1k2XhqA11^KyMx?K1Ulke6!dE0Dc_r~q8J^f&L
zV>^HAq{s=U2QL}LQ!Id*OlosZcR1INjQdLt%PRrf+NeGFJShL_wnNILf=lV`Ru7uD
zuOqBppxii;4_;WK$CV#fz^+qjs{|D@<1A@SIJ+#eVxt?UHlQgGgs^!XqF!aG_nOB0
zUb)BCRfwzuGrQ(-8P80XNA2~zH;?E}?)ZO8J@~KM>dp|?^q1dTQiu@a?e^JKXC1pI
z!NyX>2qQf#r$v@iao_1it#ewjCiF33AG9g?r<!#!`};Ych<k>L*D6iEc2ifW>cz5W
zoXVv<h;q7tpF^i|8=j!4GKPiT=62Fw!fIC-s*kuX#o>&(oQZap@_(ci!HMA118Fja
zqg>hBACOfj-Q|SkM%Jaz6K~uhv*|Y4jNdMaD4Zv3Y*%tp+fYW8Hzf@J0oJV2(U-yP
zL=*xrGnYZ|EE57SH<uAV1t*umX+n>Gkj23JFPDFIrWkDW<l<wT#($qKyI;H6e6*U~
zq#Sz<2GrM|yZC~Dweyog)Dw{U{Y`z}{X1%{XN|M$lr>-#Xk*mxxS^B06+i<BpvF1o
z4p4NOIVk6>w%*Tx$|YHVApwEq(kwHb8ky?=n6)|5f&8e=h-+AeIptlO7rW|z@`@RH
zjdH2R<#f!}P<;Nhe?ju{hJASx?e;5ah>{o*aTklz8RXfQ`_(zgleGo8m0gGeG}34n
zAIq-o+hANh*>P7o)7?JLvSt%j=lSfF<e_l;Y$i?MMh!6~1gh7Z0t-$>@1a_kW#%mS
zIcN}N-edY7cONR=34JL3Mh*dg#o;6)ShhJ#SJJo7S8s1g)~#T(w?fpE8M3LiK3BQ#
zLiieJ1VOUMuN;=(*+U8~0RSri9073EF*pa?sSmUivQ$6}*F!g~LN~25`Lv`h-us;H
zvPv1LD+2OtvoE`H0!w8~x}aF@Ypsl~j%czi14}6ME3n0~SXgg)m{*p6;8=SpW0tJ<
z)fS4UoaVjFy&Fj==g-#ARP2S4)Ljy%0MBV5P~0dmJb7)A=K>RQ(MrBB83{V~<?qqD
z&r<iv)2i-^RTIBc_dll2vnTE0=`ue_`@%ul+(c`YC!?01WN;^WYZL}ZZM|1m{I)th
zI1S_B*JC!nOD$-XHMyLBIRnp`kIv~y>epT7qX&&<^h9JMsV<!Rs%28!Y+C30QLu8k
zG-J=b&-*;uM43^WK;!OboLtDkpz)fQYmD`)>9ACsKkHKr+Bv$@TC|b6t+2k^tJ@hb
z0d0ebExX{NrzZg<S-tt@M*Ex#kdzt@zRMacnZu<kg>8Qn=FKvHqqA*1%zf-wC`5@c
zJS{h_bRyR3xSk4~0LW~;iAH!W;^&+DRVEZBJxw@V=5U^yq7^eZvVK05${!ZrKU(!A
zHYGLfN3i`nDm;0oQPh73MQW1FCC5l-e>jukw;H+8F!(!;?Q9Zi4RX8jDsG{>3)5(_
z=mDFA3d6KA>AW(3DW;Z+?n`{N6x>#ll5R3v9;t0%IAtalB?W4_4Bs7BJ}&S4T9O>o
z01KJPZ2ZMqYRIeGAg|WiN~vZEUuF9P<;&Vn?OA=RDa3Qu7zuLz_Z~rRv^4~|w=J*Z
zZGv2_8%c0p-;&@gxkZArEgcD-_uWBw`@CDcQa&e+OUaok>1?IRd1a=_Ta=%qZ2?d1
z*Wp5;{j(JMPqZWg$s>HEC5^Kb-Twh=q^p<lzCjcMG%%Nu4*?Um8*W1A1q&jJpx8hD
z1Fhhc_?N-$L=*xrG?!767ZU+8lkr0*f8|=sZtb=Xo#!k314k7l>S5qu1KCbeptAy*
z6&)3~9d*$D|DbwM5+%#t&S}sVxC3X$FNq=-7ay`*ep}@7{nP#Z>AN?Lixb{}Q_K6W
z3xpdo`~?{<TC+F+VKpr8zb`+1l~UPo+{1xVY8?(2{`cY6+`ee=;Ry&&K)3>MfBWJP
z;ydM;HUFx&o5pZmt2{<~-_g_YPMk^GaMi3bUI(NjyaDd^^ZO4^l9nOGYU980XFPj-
zfBNHTT$Gj|P@CnK-=E?Q)F8B1i*{ZZzx?^X<=3Yl$5FY;JOq_VYmMnthBBziEY_|<
z$6Hxt7#4(^ltzZ%uv7!#uyjqufAiZHE2E;)6&ue~JbR`qYk9BU2}7B!o@p8{KI5X|
zugsa>3r32HtSbN+kbsLt7xd%G3Z#uknNQO^aV&hJG0VA6E9l2G0_3__WrTsXj0v)i
z_gYs{uc<iJTJ?h6>UD-+(t^!%8<U?xI@hGLa-P;~BA}X{;yCiH(|x#Le>APi5888l
z{>M7VcHnkFmF<#bY4dlwDhL7;#uil>5}}G^3rwQmX{dX=Qw6M(lq+#jarvf*P*uYr
zRRlH9q`VGv<&IHx<^v@&2yf#Ey`)V&-5c%wcn?Y5;tI$#Q4Lcx5u}?cRB921tS%WL
zRLlmn3e3{0GJes5Y49qpf7MJlg}Palu&DpPp>Cepv-bx7$F$VR2~&F-AEJ^+ZqKA2
zll1`VqTGsPL8ja7mcT?JCzn}CeA~p_nPV%;YbwDPn)(1HP70^YMVKI|&mBAd-kr!L
z4jnK7OcNZ6fqa#^yeApt3?DHo?u3d=bCwFux~8n#F^+xMh%lG`f7Ajw&KdN=%Y!(l
zNZz#Sor^TkusA-ova-Va^;@lsIMRBOhoi5M*mH66q}yk*#g^Z|`(a*T=lFoHy>sL{
z!?iMcB26x7vck%w+jiP2b95*;L-k?2G(p#E<=YY=Dx*BtsX9*TK-_We00d8TfP3C-
zPLd+%TQd<h^K#m7e?46^<4p{%8GyO*9m@kwqL4(fI`}+#c?`fg?WA=K?<VFQoZT1(
z8{f{;*aU+5UU9&Z>|$nZL#M>rH6@XaC-7KhxS|MC{F{til~4ln+7{Z6g4jUEjI?zX
z#Z}cyHj8gAbgb^bq3g`O=#-GFIjAuPh#%AxVfg}_yJ)W&e=GUD+m(dSJa+(A0}}<S
z!Am}(v}Pk>i1%M}`W2TXH=fee7(#Z>F>9VKPvhaB7f=&AP=pI?$yIWnOBu|q0QoF@
z678a1g#yocHv@!cv&}^g01u5>+11NI;c}`|(ucn8!JS)m0q`RUOnti>v%sxYZ*wRs
zFO>PF^7v9+e^unNebnn33AP3K9Knujxe`U${R3#3No~SvkvT$o*FTIZX66peNgpQS
zL`BjHjc`CkLkbjoQ-<bN1MQLJXZF<|1*(fRVVR><;FH?0V42cWhIAZjxI{Gt!9**G
zyi`7LsMQSBy#UOvlAe8O>2MqiWJ~yBuu3!~Z&l#Se`@P^(v4LHfn-KQ+G&BdMv|hC
zG@E8KHCn3kKcJ;93QQ;`I>_r7xodieC2i?)@-F=ZPE5d)#)ouvX#Rv=IR&~lAPgS$
z^_Ydaf;{K;Q&6i_SoHzYdfayCvKE$!oW@@}#@Rj;sf%M`A$k_eKN#nyS^_Vu`{tx_
z55$WMe>pGr^}c#w056`;3u@zek_n|RcV}6;ZGfCq0|iXJbGb7Io2pSd;Q&5QkF~S6
z(X-F9f|$_|k8rF;VT3SWX-6Ebv$eCAY-Gi3#`w2BLEYdMyNz%2uvpr~!n|iP-PQQ+
z&G^L&vpTD1yFej#7E0HCThnfDZ;hT(>HHtpe>vTo>O4vIXeUYq^76W7o4T&J>^xPx
zQx7@sC(W{&Ou4r&@U`WG`RUp=p~#y@#-sH@T`%m`3n%Mk5=Z;f+{|C|{nOloVW|4y
z+J_yQ#oSc&+7b{g%i&6Z+-`soHsh{PF@VPKTOCkyagF`TD5Xqf%bi>=7BTAcvI&#k
ze<MiK-EMqW;EXM5#57V{ThvT0Ml<qO4DO>^md&wlRZF<C3y~DmtD-Xw58e|qRmPoG
z4k3_d-+DcfE2JI9oht<X7B*O5u!OqQmp*kebWW^VjU!XP?NA=Cg_gUwbg4jJ`mKr9
zbv%{DC?h=Di7aGo%H7T+tC&?vtg>#Uf9aL=F1B3{<&=id3vhlM@hFJI?eCE9m%OJ@
ziDxc)uU9Rr*F7hX(hD-QrINHTb?YJLJkw4{z(Yo|?zD3+$EhY_AP#UH-yXBuBT`H2
zEV((Y!Kk)LS?rRTgpbp5uc`woz|Y9G^4W(!6R;?>!tpDC2n40zG+lFWC&AW@v$4&M
zy&K!MZQJJTZg!K6onLI*ww;Y_+t!!&UVZ;e-I}@8)je~oyXW>f_pAtFWSQv=qXLt!
z!jN)j_*^q-Usmz=+f#IP<_*l?JN|*4_~0(As2_jt&EdIPM^>`LTQYV*^1xDgp0old
zwlkF9wO(P!kC#g^cHlWW0i}7Hf;)ofcZLl_&n?^-r(x6U=vU~e@s_l#0<Xxz=dhMr
znNS?v@jFi%pY)9zM5oYfB2j;I4+9!a=aBv!L#{3bh1OI6=7{zO;vkB<m!XA@HBui(
zZAw7?vix+TvhMkK+B1%xJmPijtMAVgPla-4dM5?;`Z4yJ={K0j)%}5KjUO5xQiB0C
zUK*zpsr7FM)N2}P4Q~dv&ZFeyr_Z#p0fsgcNvT9_{Ay*Y5W5+YNwA9kYygGgrcGwG
zCK8{NRt4*50`c2An123>GvsO8Uabc_latqn{+wFr4-w&VfcgdMppE_cZA8dFlQ(o$
zBsq*Em{c+}Uut(rqD>&g`QzszGF`LC!^qBsr3U4!jiBY=ZrW0W{fc`%!9KFqdTm3o
zu_1iZ+d=C00EVCTe$T7M8$1vnRumV$+N^Fb(Tfe0{1{1Iy`t5TN4hJ9eA2UCIQKDK
zGvn+{x9<NWzdnK)-pB>KA|><~ZW~a5|9{q1_!lV_I19+R@DsJG?JtrFFiG8H0={og
z6i2u!P^s?)E6+!Jcee-D{xndG5G6+sj)k`n!f=;9(}%y^zJF@Y-0t0u`shR^Yof7Q
zmopPt$ViYn*KmB1s)V<gE;xBL%MW1sYs^<2Qa+IHD=^9uj3yH_JkU<s#1kBA|0??y
zT+V20E|-i`Jr*X=F8@dl%m+wJErU2D<KZT*TISH2-4*;-ir7}_eKY65TDDsxzJ1)-
zrSk926s+YFM`{%DR!CW;*gMdl)b%#d&1!S|prz(929;+#>m>#p@#}#QjKjIKeol~x
z5?VQYp0mJ6xA0QG0af(kg?HjixpDpJ8m}w@>0&w?^Y~x9b(nX6WZyd%O6{itK(9nv
zlyI=~ZY5lq0FGJy2vq{l|8H^`M^BLw+Q3&H(<K_r0b;hG=A!yGMi?a>e7gc(Ql(NI
zu%6M5YfaLc;Z+@a+FD_sl@Lq%o;K=*pY80mM0YWxC5aQHt<Ibt=#%|(v%RjMXiF#4
z7-`gMQw(kWZ)z04@b4AcYR*kYA4qn?iCS@U>ugr;g4NLVjg~pszx|Qs;#zvh6Ju{n
zZ*ONqj)8r`<llu6!j#a~SU<2z2txM3R#Ii)1e~j!e5fC-TxzAxn+}myz$?&*Q*wpB
zwejJn@g(1a!#xGiZCh9c<A}IZ9qF&6?uRzB5?E#A-wZPW-_=(stq1LiI#O))2+fZb
zeXXyTPjN4JKA0;|pWvP!ws|f|2d~cR2^vgQaoH2BE-m3t@Py@eN`Tu3ky~y42mO>3
zO(pk;?|Xs*p@lVaM=-(}W#nW;MuVT!!U9+Ni1fz-zuaNwtiK0Vdfw|H5uWK-IQjwv
zD2n=RRFLKXco9$B>3NCD{W0<Jl2;P`O*uODaRo0<qY1vZM6bBOK4vJdLcNN#*=wJ@
z_9(713R%$Vz8oj6YrdF@%$>KDpH+g=B~0v8b9e-6`yfL1w`*%2!^~!ZxYlpoKlG9e
zdggPA&FIrJuyy}B4cYRzHQ&<|r+J1r$CoKhyh(KdgpfuTx6nyni)9>b4#xLkTJo&p
z9<OmFzKE3OPOPyaT&2~yM-p>?Oe(w}b;Y0N#3q{*uW~XQnl0IJDk!+sGxy|r6KNg9
zw+rGgdfwosOU`iXjxW7`C;D(>4Y=-svNTrf{DBI^ksEZZw4fJk^3{mWm_oR*aUlmP
zC{~dOn4rnju33KGmlvQb{GlnLb9ygoZAP2PHZ=jAqk?HoR=Vw5eDC~;;;b)iC(yix
z)=GJrT6(zy5iwUd+6^MvqQDdE9fA88;rkr0q})*(-0uI1oYonVst_#7f*e55^I8h!
zNE$)WWnyLl!05+Rq<obO7GwW7XIt!m%_GVKo;xEq8lKwOc|7exh>A>@47}a+Ys0kr
zs+?zh^i=;&iv(*a)UQG%)SV=@7e=((#!$!9q@AAEjF7pc%<1UQV}8Y(yDJ)!j_kJ1
ze~UTJTA0qnA~q`a`IVh)QQ}9p?yCnBmr`@al$Hybj;naj^~GJ3L+UezV1Y}((a~T6
z8h3EVS}c08$K(lBzEtHSt`X853bY=JrJWw^|Bm^lHp&N%<JCy=g-|p%uQ_a89r}?w
znUKqm0skD(E$V=ur@5<OP~I0!jNgJIA@8z!1pei2*bS%R{_zMF3I1+_PV&(B7?E2#
z4|*}XerVobk{;eXMLBN&(Cv>d7klFYl$pf|u0F}<m%SuV#!lNvj~v$;>@f-7=*izR
zO?-_hsd<Lc{OZRu50Mjl1`L}p>8ld&E&$=BUG9|}Vut&v1lzmFi@DNCh(+p#{S5gn
zqA2#zRUWKHzp&NZ3-b$*Re{F^;t(Cyz^UL1DLw{W_P=(}_1dV3My5#V;W?hvuxDTb
zUT6*`z6<fJdSwr6wKe)DaQ);mI}<Vf9wPpkv)?llqb+zL;34*rK;7c5yBJ)x)D)xx
zb78Z&Q|IX)<$broI=E`~{}a4%B{e39Bus=31pQlt3Y-(<6edE`sP<cA0}4d<TLck=
z5G?}NC>0{o3l7STCH=3jA_NC_<ELLV6v#151flWMWfUBo1N2ppn*xc0`<tA(gQc4l
zDGMii+!hZNXfg&49uUqloaRj$ZQL6D`h`hhdqBvA_H=i-l#~$DyR;~QQdp;OytC7C
zsNR5qJI1`VS!~G1gMAnAU@!S={W4YMAN1vdF3n6@>hRHcl%B_=Vw$w+CtdQPUCVBG
z2v{KXw<fLN4H?&n8s^jhWPy}eI0V*Jgo?KlRO`GsF;(p&Ch$~ZNA-35kBCW?ruY0`
z*stb$WbvwJGhe2;v9Yt;LbXewrjz(YAu8+tsF8++CJi=4yMM}-Oh@lvCR(?JCHGYc
z{Q*}L9g(3TN5t-*RzcTDjrRT&$FQR!8d9So2XBs-Mq!EfPN<r=gh&|pp2HW2+~Z?N
z6-LU;9fulICjc11KvQ#pyrsE_M!tXhu9-acBX7(o%_;E?8g3(Igkq6Ud0vis>(^uo
z-WY-EJfcj@d{91~hg*zA8|eZao-j775md~$Ob?V8^Q0{oog6Vtpz5V|6|^X#csd15
z@t7^v_JB{KMSWKhAKutr`Zq*k3Ry_&-;ViP@Vw@8L_q*FvND#CI2R)3fFuZ(8!?m_
z#YN-M>}JFcYAKQ$O-;K6f$T8;?8^nLDSRO05i6uMnw{~F%)l`3nD3sXac>IndBH~D
z3DZB(z%k&fU>19~Vd4)qf5wT@WD6;g=CNSpNepHLYbaw0_V_6u=qU}&TFL%4PCS&`
z?e{{MHs1iesLAqse&FRW_VNvF^9{+T)ZlD}<#B^A>cO3ip+rcT-9QcT35gIx^9gY*
z2M$Jw-`TOGj7=1@2Sm8jI@6Vq_{ytJ4)nyY-~b$gY3%I<*e!fq8R4meQKhB`aPU;U
zy~jzeGUTY&NR<;@x(-}sa9dHS)QTk}mT~XeYi>XZNhXP=<B4GEc^vh>T%Q+f9ntHo
zAd#H{V$K^x3T-oYok?m?Z2-Sxc1UJ>#bj`M?r<j2AKfTExM4TuUPMz?0`rE#ovXa`
zQ7kNE9%xu^Mrc@JcO0Hw@VTYEm!Jy+q`}UxV=ETkN=xn7V~ua6amU({pwyX`4L3Q8
zVo2a6C~pg_TG}Sk(}l%mtvmx>4$ye}?#)}A5he+>3ME0h%ZPYLNci#=M@=8JC<$kq
zmWODs0dWo{Vt_vYSej8RE15(J_YOZgqARE&Tut)kyHuS8yib0+tOB8d<NP4s8J#Nf
z#n1&KAR}S<`7S%*i%^yDy#?EdKHKvJ3;{?5$3DacZ?Hk`{0<A*o6>#$DbABWzcz;Y
zR)D7y(l}9@q;buVSIQS=BYXsHjve0~%s@dOGrk^6F_=Xd6Ar!yGi<#k@k*{8*Z_Xl
zm12cdm3W)Tu*!rJg!4>yl@l9Yq4|pleIgs73Ei)%A@cah<T(xacer*0_W|5yTLT3*
zfZCkVqF<BWx0krE<Hpgjqb={ZYM-mq@#I%~HF-1@_3uizn~`2WTNp92>!|TIS=9)S
z{#htXO%DXuCQ8frI1q;?B6iH%NOxImcg8z7mLFXN-be#XKr11VZ|=GmDRr&zow6oQ
zyR+kWLKNf}w@&iygE`MCGyW90Cm?&>qUT%b#g}~%ta>yl$yFX&>#1Q<w7l!9fV8)k
z1wSuVdxhw%TEm~Xr;iW0!@dmGk&<(M^O6dasU_Weoj1dW<h2Yq@mn<03Zxwa8bYhN
zY0e+7V3&8_?L~6rA7wlQA7zZ8J7QZ<w8-$9Zyx^QE`q;64ZqRK<jd$9!2!Oo5!l+f
zk-2Z`Q(m10>2WU-Cx{IK`z^WG^VTSmd*!R4x!CNKj5abdx@nD&Lf{QV7XVQTC+-3d
zeY21co2x=<@aW~1UjSS*{sM3cE6*W;cc$Wu6Atz7yquGsUVG_1*nl-v=*_o?M}uB^
zu_7i4y*OwMw2OrDhi3R>5@0J{UYTPXYJL8;v%=nhz0q{rw5ElnWlgQYlm{o{M_v>X
z(YYcgVSg_x(JnXOHy@+=oy5H^Uc;9%mjh<)=lFoTAE7XqV&Ekljq;R$Q}bI<zOu@<
z6Gp?!GtWvXq7IpN;%i}Ya6JLJ%_Omfk;7cLH|XVTTn_Y}VlgrNEuagQd7X6XqQiZ*
z<JO2LJ5XZOAv$mB8ZGV59A|2`S-Fcet)<!}b5YW(kAsb`Ysm7Z78NRK_aDc?6uf0Q
zQqu=nQVdOWsT1*K2K1pK==#U`Bwoq{$)B6O`5c9t-bZEcHr)CCJbvbGEm37QkjvS!
zja4A<a9yZCS%u}gCBQYig~n1}NQ-B03L|dC*45QqJFq;fFy9)<M?n7kBw1sySM!kJ
zqPq?Tx4V<TD()+UFUfFI_@9?Y94!v#d{2a48`Ai@+dD*AJFSa_dxNL40MNQwM7cee
z5-dcszE*^|DNYYlS;!(UN@SU>|A@X0Mr%HMwQmEJM?E_;0>D>~z*ql#p2{b1C7yKi
zjri~r0~JsH<h9W&$zw;{VIxi@v~Vcm0ym@^>-y0~x`T#jy+o?LZXQtI_8raJcrGKD
zK{HIGHgEuFreTwqPOWS#GE&YGX-ho44;znMn?A*jO%vOYli{hSq}&mp-g`Re(BE{9
zI6Jl)IqQ*F2Bb}UnaY@`WtRV_=L&Cc+@23w+GaEl|7`S(+i;6p*`pw_J4#6<^rx*_
z+uzC#(Fn5o80qS$kFgqM#w>md7i%RF{kQ+~=geWA+Ve+Md{IH5;m6X2burY&o$s&V
z1`4diG1hdX^{@$%@?-Be=fz4zfa^H4VW2^lpO|B99PkD4{PbH;9n{{4C<sjV<5Z^h
zS&`4)nM#Me;uT)1;NSsT%f{N~5ANCo2^3O?88+TaKe><JCfC#2cwb&h7YgIT*0oA|
zbB7Pxc!=*{T(?+O&{}U%SHPXE(62_Rg)iag;cwb9^}|w+wH=>ph>-=@y6_CEDc(aR
z)u`ni0k%^?A{2>wr8#X?!`IX|{kK(Qr;0B;#F59>$3jo$S4+dLTD*dnqeI$d7Rtkq
zAteUkFL5_pUP|d{qSsYVZjUuQ$KxksE&d7Oqc_B2f!6D^j<`;us_p4*1SrhL>HUS1
zA;BqrhINm>`rN<OU2tFO05*m|lg>jS_m~?C@a_%~Q+Q{@<!F)Fms@xc_i*<xN8kWn
zrFL~_Q@$aZdgP{NST&^{*@nzr*PEj^H!>(m`IcC^{Agt!?LZg;KV)`OYt-H+(72--
z@W*VlsPj4QF&Z0R+)it)T#`3bOS}2JXVP(2kM;0CU_HBekJB96l+5mU%2cOgfJQzU
zn9q_7yA&PcFsO-ZOh!IDNy0j-K-9AT^7V#EFgfrnS}8$W>0LHvSJVe?nB*(fV_iX{
zXTaIJO!2+0TL={n9MlY*W)u8;Um@6NKJCo%k!)to>RyhwYjw*jlzsKyS<>Ry%$9Y}
zvg$RVbf{ZI#GIB?1~3jVS^q)$ddQ#-z{~S?=IEPdGcj^Z6)1Pzjy_apKujQf&r?sy
zN<(S3i5;`Pl*YzEXsSs}xpYCf&VR`=Bu8K!P25rE0GklW0wFR{Gu9zEmvh+zQhG(x
zCctLntVNQ-Ju)U=k6Q}VJUr|VqW@Ld&oDMZ82}wY7)xHaK+ww%O(D^2VdhXN-Pg5d
z&=@gFGi^X?%FBovutL4Rpiv!sSvG!9$vp5&kn_#{;F)g({Qz(|$Ir6`kyy|4WDaz)
ze`yXn|M@dlZ{a}F<O5|khZ6L4ng74XKvxsg7@&qU5nXWB|7@D68@ba(Xc0i!ACMT0
z?nNTV5TNRNNK6o@L_`UMMhAn@_*^2gB?vk!5kYS}I~Li20L8Abp@X*1MB+cU-9N{m
zF&ab9MQ{y4Zs#JHjScdm5>TI+mp5fmSJ2hoCmK0WMHEE|<fI~s4&sxiqHP?{5#5A<
z_|65Y%oqIu`s(o&vr)7_bW0!fy+9PbkpoIB91=tUC58r)C>KX<+<_H~V*vHQieWT{
zu!+eNf;`wh%_Ai-2p!Ozk{D*=M~RpZHR!bD(~OuFlOX_kPk%ZbpNYZCf)>s`O;~Pm
ziSM9WE^*98V{7pOVvwA*I94O!UvWH2(3ijB7>!cp;?o=;_VQ1YaaTN&5QM)gj#)n`
z;X{znDuLOkPbe`)36dg|z-kl_k@$lMq7(Tv7ZoJb;s5vVKnWtH6@>ebMp0MK7i!>4
zPurD%vmxN2@^DZ(-|kd*POXHYQifsW%AxjANIHBBPTWR{uKHkM;{uG!j8j^h;;e)h
z1w;HFI6l`4#1DCT`aoN=+)GY214~j4g<`~fs$s-_zJ7C5$nJRMm7*jtDl7s<)seuC
zDIJ{eY|Apfutu>Fk=<+~!;82*a)Ce#b4#>L0kajW6*$D#pU7GFc;ON5#90jm)jvr~
z^X%;UR3OwLXWx;3HaJf}f*=T(pb|1)2KvXcW24m))nzyY=!V2=ppw12Yp7^US^p~M
zj;d{lWuT@OMUd&}a7=-kbvV(SU>Gqqg*PZv{~j&oZL(OI#i=lm#xvDV{)Ge3xQ6xc
zrHb-Ki{vSc^$M(v;rHyaoWttB!_W#UD&oVl!c5lHX(e}MOv8K3&4UNe$H<k{;%756
zQp!8g#UM>tp_p|E89BWpU!vF8HRpBbDpXNWfh8+18Q^1t!x*Q_TO*$4QXb7}NkR9p
zb;rOCOXUj0*uzzZ^to64W`G7Br77B$z>KI}k14*A8_QELlgm4yf~mxev^9z)QHjAZ
zM^OEtq>)ZF8eZF+Sk6&Kmc>7l;;>9>&u!0x!!{^_i*O;fnEKi0)@`ay5#?D$A#|HS
zoqsHLd;_1trh;abWllbvl2K<D56inwiE=^Z`eJlSWO}`dTNN&*e8~YY)c#9^H|k*K
zK7^nF57+({4ZZw}*a)5h;V(57gq#H#EqineVt)oat&uj_i3Haf5q-!;QM$t>Wi%K5
z;<!zXvzwDxsq3XrUkv-M8M6q%{0QOnr!*I&_D`}69u|x-#ji<1TzJ9eKgTk&PI&pV
znsK5f9;tsjeQS8eoqEmy&cjKst<*mhP=4*t8I1VaJk4sa9A{Tf<4=U+<QkjaHjr>V
z@~f^S-L|XIhy@EkEK@2;&;;&$dHrG`+9tG?g;m#nfO^Ch$v$n0q?aYFm4Lqp<+^#~
zom_|e>*OCK5leLsQ#cBRfYx735vZ@fvdu#T#U?c~c%DaUc#4^O0a`dHCT;wm&1ndm
z(K=3$zrJJr?G69wc0w6JQ2CEc6+H|*+!XJ(SY$YJoT~)7O%XJL{MNrIDJ84id#iTJ
znHHgI?vrzI79)<v3HzVvZNq3aISUy<^CKSg3qJ88yB=Ztu(}!h+3pW+6T%DmX9Bi2
zVGNK+zZqI*%+oM4N&r`tHAE$AX0=(Cs7Eoa+GYw-Tc*yur8D=t9hamyeVI~TpK;6x
z+H)De7}~MM_mLqL{zrNJk~CURh*`D3Tt`^ld~d>EOA1q$0wH1Sl6q@7%Mks<RbKqN
zma@eQNykS?-I?wwM+$v<mxC|*<OXt)@V1-tUt=pJM9*!%QUXg%u-W-w_sS*|&@#lz
zpPUPJC*R-VQnw6Ee2U6Ktx`N!F&3DNd1d)UA&?;FshNhoGBSj(_9nB(=tSQB4b9{D
zg95*og`UYB<nK{uYj{xDDa3OERm4J&xlE7^q|tR`tIe}b!w)X=jl_jur#+ML;&Zi-
zFhSTZcsYozwE)SGF2=df>sl;@?s_9Op3B2V!<pj<?2AvVop|88LU^*QmGU4l`1~y@
zyZ?!A_kq3H#xUQuO2E0lV(=nL%d2zVTfi6o6g-u4@dQhWRM@y9WAbUm4$Ii{!oPwI
zQU3aqeY2V}cbo5zw!YtS-=B$B%PLhwR3iKjL?@eOFMy4w1Ht~(N=?x#FCJ|^!ih<X
z+|qux#MVhzkm%LzRA3{DLO_7WwT%yV5{>u6+~u|pUsw3mrpfvu-_4=bbH3+-3CLBG
zdU<N51CF4*qe<{{V};#JPK;QQfh>uWiSrkd{4Xwl4%W68Gj7zxxgyg*uSDoV*w!BJ
zEZtT11Ryy+<VGAdAR31PwhOB43)~5=3=ZQHVtP6T8S-5JU05hYIR5t+Gm6S`=4NVF
zgfKJk((3B&T-BK^3FT&%V2{Bs14W4FX20D99lW~3i`AbO5B%V}>t;tkI9Gmv@vcXX
zzU?eIn)GW&M!eQ@rR(9b*YzN#TD2FO;aCU92haov@DsgL7>rE>vwm594WxXBL)6!g
z2nOYOKOwxg&eZbDEYMNg&!MCfuNKX?x5Lj``!n)l)NMIuz&$w17tJ6lvO3%x0lF&>
zV~i$s5LcbaXGA&AYrQy*27ZrETbG7`o(q+*aZkf;EbYE*M465Q?9lay=DiqgdIM~m
zfQA%oy3SX|-udW(Vr#q^E^8)J9z&GAI^vltUejj+6^D@Hyhbs16<&$@f{E0lxSe3?
zz*0(1-m4nCQ|X3Z{I5esgP)a_3CxF$a?f6Dr#?FGfind>3EJTVp(V<`BW3Q1B%5U3
zbR7%jLECDqi}{Q_9=kHU>n$pj{`<eyfa%LZKPHGR$G~^H<!!dyNO}J0tq8=DqQ1o{
zd#<jh!P^3@{mX+L|23j*=FOt9fvS_PM1rZ?ss#(%yQvZC44usfZh5XqH%Bd?T;&7{
zskaQZscciDp^K8ZbQWF@+new^E}|^-QBz7PL)5xKngG9Gz{}>9PDxufQ$PO+pt7n1
z%IR!o$gk2!jQyCXsnhQm56!!_w#g;SCsB)WP`a^T4lOl_UyWjrK(<70dF!brqy}XS
zb4t3EhZ!#0szk_e9k)=4eu>GLY=jA}r$R#y+)>RV_jgVcdQfw#7))MAB}C%ygEL>6
zMw08}(oato`l_arM-u+%ml-1j2+wg(p#>S;!Xvvm6x4FBff*6|qes({$D{~|#4&x0
zbkyjr(V8q$w*U6;7^@8$ijT=U3?0%+QPsK7{}YYpmcD7K#gsK);HL{7aD*N;^^Z}`
zCYIwVp!aiXYRPjlfk%xj=|d;eefGn4pZ;y2rY#}siJ0Gig>mTl<&N4BAnD|LZ{G{t
zSPYzIe-NmycwP5Z!p$W(l?k|LzJ`Ho6n{xnrJT4Cq*->HzV&IE3DzS#zg=2D;kVME
zrOj$*5RO{R{;L>CM*chg!j+~#jx36fajidoNn&qal}&?dP`BGfOZR(xgOYWlo1;E`
zcoYXURFoMyYb!466FO59uu44;I1@3U*_?mJH0azR8ufG?Ef5dSyvt*UD}#BnzN>*Y
zY$S7tK8|^`JRuqSdR^D0E-n58QM2HP11?1Dh*HH;H4LoUR&O0X9?dk7)B5h6{vRwB
zw8B?}jDttNlcCb0d|#<WDGXj8?n%lNoZ1L@Jze`H12<=NIi(vxARo`9xe~$yi5svw
zmHp8|0(ucCrwk2Ze^_OKHNS~(mvD!-u?AyRr#LU_{{hpG<nlaSp_xcYfI#4%tBeQ#
zO~Q0WU5J{ose5YHwnRK%ejRf3Z0T%jP;oG7qu%8F!xlr7q^M%A{(@iEaioDVznD?d
zW;Tc<e!8K^lRT9cSQ@bqyr%F^JtJ8KJg?8ECQ%fK{gqrs&5Ik{ZjO#>X0^>w+S%Ie
zbqy*`&-d<qRTf+P6`wb6j9|z=MAIf<EPJ;=*PeW@rp_%|-JskXbxY@jEx}-fW1<MA
z`IhweBdhk6vv14oG9C9CZWetGVW<zSkcGJs@5@~|Gy2NFsD5-S{#XzOcFCIcor2_=
zLTjT|cewqqPvOFOM&!It`gzXW*R<NBr@N(XKekSMlnxi19ELtSCjlOUaJaSC=WCZL
zTzP6fNA74#14WIf03C^dd!UwbvElh-Dmb`PVEi4wp^2Uh@BFyR_MS)SO1If|c~U=j
z^}HgrK*t8)36uO)F}&{)f40_|3#z;jZHU_`ueOrnl1%s=TB|!8)Z;f(hz%dLyuGI0
z0FQg?0e%riekH_*8P#WcCB}y}xUG4W>P4l{=Xxd5hS?dWxn1u?z8>ni-Tp`_5UOZG
z3bRu>3F%Cy5FGZC3&T=%Q9mv0W()T%J8Md??@tqe`|q*ej!O&Ca=R8c1PBrA*9wOP
z6&^q`6HgqN=&_MA_3zau{Vh&1K>h{;A@J9YlOIPjQ*s;rV^VT^!-UErv(SWTKNo@2
z;BWt?4ub$jv+Wh=QP&U6P-46EaWX}^Cz8tB`D5$iLeL_vNj@w|3CYCnIUq2PfbZN}
zHd7oh1?44Hs9;CmkETRhOMVsP_j|p3G{>HHgb(4-`aP@R-x)n+kK^(^RsQ3hFa2FQ
zdoDq=lxWZE`f;~A7gjs6yg7U1XT%ChwCEW+e`@tiWSaVH86uzEA8C3%8YZSj=-zNY
zfH^7VWaIeW=YZj5YcG|W`rmG8dB~8Ag4i`+)Oo;nGUFjGFqXU{p4gzC%A25?7GYtz
z00P35l_v{!j>OUwaROP%Xjo*JUmxy_7~XN;pxYHHEw^2qIC|@btynNsUzuvX=@Y_A
zuh`>x+6b}IZng|X>vF%D2oR46icZ#UDKSXwxYhxR{XkIdQA&pA_@iWJyZ`h3-&^3c
zft#UpX|w8uk$rS!j?hubRry~;Nl}dh$Yp`z`D!OD?%_ge#NN1=J7gLmZ{tXA3x+oA
zYh!DlLPymcWP`2wLYePJe{SJq@p9*Tlm9__N~`ydyyJS%riLXG^~0p*ebBaw3zm7w
zMs%(LnIIAs&Y734dzc7}=HiP#tqd?G{qpLjM>VgI9uzlf$M%8I+Bvz6`2^SL)wDc@
zQzR$Q4g;@MkfzsniGa-Gmh+WYesJe#<3^<NoPZ~N^n`VHc$g#?j;KXq5g6a^WnFgJ
zFVo(SEX1i3Cz6}MOOwMv?4beW!j1WP^BaU$3>fyCIq?wxg;PP<qUHGZlnBfs!b}wV
z{!!gMWaxwyHc)ZW8xOMVy)n*5P&eLweRdPjJCEzPlY%Qn-sdN;2aMN^R_`clZTEcN
zug-#l6>GlPOH<Hfd1AC9+gooW%|3X?-f*g2HuTrB%>Q1hfAsV7xZU2h#~yH`7ob3F
zXD&aoSN%xuIdrs>tZ}V?s|IW>KF)R*EG08*P<0>3kX8OMi}U)lD*~2Ig{i+@l#x#s
zbbFk230=ULdae1i7$huv`)_Ud#TJttuUGYLd#A!0C)a_;r`@}$;ZK%S7ES4qy@zc|
zp6&tnMmch6Y9T=di4NSq)PzVW(aOF;_IpqH?a2Us0Rf-br40)U1mK(~wDCUFfG@GZ
zop3GgtuEdF`-b^i$%@0+?6Hl_*}dhBPKjQPAo2J<0vVh~1`^$0AI9~l37pJ>r&1OV
zlA3hUS1R>R3H2u|ZLz-45>8(weO?2{*oYG|by=OY7>+aHM&F=S($YPAZPR5lbKO<{
z_*MQn``9^KvvU0yXrL`8<vx3vKhR#KzV(FbbVoTIa+GQbDauCmj<H1fT)4j0e}kgs
zT(JR*-=O}iMRCl-V}?L1Vl}~Pk=kRPqcqj*t`S{)B~@fK(9YAaR{D<Ms5p-5R^a)}
zPLRX;+3|O&wW|CJO;fAkVjjMyeBb)TPS@ITtMGh<FNVU2EYMP%p7LuY(5}(OK}FxQ
zU{mL*ll+ut4ip%+MR>;0IgX@u7kGWdewY7f?<R0joYE=n+5xfGfBIVA1t0$)eOT&S
z0<mZB0CDW1D^QYsag?AKeZSwA5;~(Ju}3e<p#hUt=9*R_M58%!Yv0VnjY~|c3wq;*
zo7X_>C?Wej0tgu?h?J=1%p%T8#(0N?O2Rw7xT4GNQRoa<;}!0pY09JWD<a4KEQuT9
zh@eD&IG`8~<*V+A+eg#zGFeA=jm<IpL89Jnt-T(%n2NzT3j4cD$wBl8p_0!8BvkUl
z)Fp%3ml%#uCEOQ`dJhSBZeg8atju6<>4*c*^2>r30&e~qcU|X}aF(wc&v%OqC^h_q
z`s8+W?Lth3@KLig7Ees6u%IIEh|x>r(_`t6K!NG9SL)}NCQFZ}S$Uw4<0%F2VK9^p
z$&Vv0S7qY%4lNmn>qlPFX+kr9F8?e1^A3Bz`NyI5DOe_k+*I>zDIr@XstcGJ$I2Po
z&L%^UfH@itmI?f%<$~jgNdK<6-v!)DHiae0Y5T)G!~-EaG){>z3)>J`$9>iN>6Qvw
zKR-dI%pXidMHrLUK8XvpZ3b6gU@JM*ogA?s`3A4Z4^Vc0yT(`kQG8KS-_kZPL|_sL
zmb7u7&(<IVbN8kxb^#P15QyE(Nu4lZN5Mu3fJM2Vz1&bRT%til#}gU~O>H(zmA8pX
z(2H=n`;zVm+2oQkW%+uN`wfYskX&0m-FsL<NU%(mU`ZGa3cI0Q&}!5bhV`-MlhWI>
zDO}l<_XE3Lzx`?dP%0sTxRzWPIaG)(h{#7%g}bu;^F55SQo!)i54Z|+`3GmqdERUR
z(niVZl@Vzp7j_{ze=Kw%^`_!RTCfsLf_^!TGB83#4_KF+<Nd*+F2wJSbS#kUOZZGp
z5}FSjDao^YqWMAVF1+CwOieUD<oa0^_T^!zaJ*gI^}*b0ikZKV5)OnE<IN@M%p&5L
zn<5ID2yUU<HC;m(DL_op-kpFcXvxI^xZZ28M&C&Bid-3dBh7PXXXKv>t>B%C+?!Ge
zbm{*Vo2P>j7q4)G@%7+W#eix_>zytp;c$F<|ExP@b1K|OG5RfHt4}%;(oU$WJGFB#
z5spZkxfH8x9oz90;!jdt21zDSPKNZmB0h`Wgj6Q4{`T4Jqf%g1O2E+K&fYf$Fc2%Z
z|LxFO|9S{^Mu$)}mW(Ki(?a0S5oT8M4ja?m{Y{7@Skvgb-RO+IBxuTy18D@qmD$%~
z%FKNI#g@gTSvoCw2HCQ!JT1>><a@}8!nT*(Dc11i7wW{m1}Q?rkeyPEVMwlGU$8i?
zMrw~WwGNM2(#l16HCFB7N9FNg0L-f|D3EK5AS+1lht;EOF%rO7yww;=@XnRHrJ0K#
zpqjgdFc##ve1DMVfPH8y6a-$-BX|8YThjH(6crDM_oB$}<(evUDhCM_frS+#Ma`y{
z&qb1`3kug9NOltrNv}a_5#eF!O4epNr9*d-vwzjCNChit3Fl@ARjw=nl!78m7HK{&
z5HsctPXpgRR&=4I#_W0inc9y-f|ay2>}!oeFX(Wjnfp4g)%ht}F}7+mf?%C`e(Byz
ze8wRIq`H$Xr_?<2N92>Id#{o0+A9^cF#)xnWqGNtsvSI?Y(;_LsQdhN^CFMM3iC?j
zk_R)<5S%0l$OtSU)RLW<-x9^qi^Ugouy|FbOR{F1cIiETT|FC=vRh41B7nLWEV?;d
z2L9RS<7*mk%}YQC{JH2qunsH;Hm?6-F+i^l5{REU<SR{RDj;2Fw8Xp_DSbinl*DQz
z(Ttyz?Y9sWoHPat0v=o57if8EH_hPP3-ZkJ{F%n%S^v{vbNx(vTkTlNr>_DB%*b&2
zrZptCs6=KyW!$^bw1Z?igF0>u8Gcsbro;bwX$1{T3A3NBLcPb<Os&gIWB;|9b38;p
zwYL(xj-R5wXaf2^hfcf&e+YT_I}r**xr1|!ca&n>QIQ1QxJmkA`RyptLE~SMRdPHz
z%p4kas+wbhNhBJOiOYiFhP!8WtMc8xEAqsu5i-#GHobHYpIyeREz(f;yAqz#1<M`_
z4`g|ESUHG?_p%RS)SMa<1`mx9XUN3V*91$FA@P_v*#NUQikHhR>VhVvbnYdRiz&rw
zG2|wtp!D@|=6Mb3e6#^Md&ad90?fo^*#XvrxbN4BHr)34t_4(sBi3|B=M9vlV{7BX
zD$eu`(gX9WLk_MsgvYk+sd27Q1mvyqOLVe36l$Xyp=;x{-C{Ka5o!(Psp^)oY;mWF
z*wW?C4FC_L%v`wI5QY5URWqY(#=OgZd?GoOoPQptoPs#5<6u@=cB(n&e*95P9(5%l
zzl4sHiwpm+!+lU|g5x`8O1W*RSWw@6$7$mJFiQB|T2S#4+rM)66RW%>&iV2c{;56>
z)$^vUa{X-nO*z|qg&92pNBMO&P*Q`&!<Uf88i3rvv|vfKlBIDmp~fO289777;DAf_
z!8UD?v%z(&{2-Ms=xoxR-byQyH-a-t-t^o&(P>l~cI7*elFdC+o&Qh%$`m8dtfp|A
zV7)_LeoU|3x5{mUEoFK<&H=$qc4y7rSIEZHQ&-nM(1SBU_vGB7?;_wP^#OR>=oD=G
z0M1{4fJg3+_uH*O#r$FC*S)uk_q)+|pzWh~5O{s%yh?N6I3OFMbb$~-sD${H>Ec2C
zoj&^1kyOH;@PG^WxL@I9f$LGy3zTe;0e`Lg7dyhJWY>EspC-7+VgO1iV=k8<W$aQB
z()jtu5px4Kzy;OmmWQ@IU79Jx_4n-#0AXIhJ}PIMm^crWPb8qQDIPMw5KVuI9*20v
z4guo&Mfb%8u7tESBu>-;Rmgd#bbjgQ2cdBy3~q^0w_@pT4Yn=^81;{w@vjR>Lq*J<
z*Zknwf8et6{(_ZF$I|F@$-kPX_lZ&|Lt)dDFJ)n7z(5Ki;KZG3fMwJw^tPuD0~6B+
zUz7uF#gpVt#ig?e{~#QRge7iaGSvm1QBuK!W@sQdzGM|7A7O$ql3J4{*GvDTTu4e1
z3B|$~^l{^=?WIA^ZHk8>oH$+Xs$_PqT=I^;Fo*dD77Pc~4i@+0WS}Gj$^z4p%!@aU
z=H!q@_b>}?6PoV$xPl15>gS(xc)-XlM7o7q0*q2@O?S@&0;1@!Q1y~INzUF$o|<Qr
z-%SHv6#^p#Y&Ppw0<r2c4kC^F>P!x`qV|w>oVBd?ADE_!MxWN6jNeH8eeF;=_Sg0S
zvG2UJ8>-5P$j&MVdT6&SX|tRNYiQeU?QI?tih`Ckh@!&WI_yz&@wo{<)Bt0MavA9d
zhZ7h(4bR*#xiYdqvZ}o3lqT+g624M{ez^hVIyDFt&L5zYxIsCHcf15})<f+4A9N;L
z)S|NDUn8zD2==HO9_loc7K`|7b93?|R1_Q#+kQYoL8^^1s4k~jc`uoguqI=1*lSeV
zs4`o4+gYTdYu;)78sSX_LIFXzt{PjpBFv1)pzg0Y;%XMnG^AiygW>eh1fq_F&JiI(
zr8L3Qf2L}{%64VWXzkP;YSe`%HK0j!dlr%yaLt8ROO}W0OwwuR6ya5mA$}-eQj`{X
z@nV!xnijFSbQ+XY!8>bzi-&2$PT^MDN+5LY2g&>Ehf(iiu`!gW)dEF7ER=nP!em^r
z)$4U;$*z@OiG0aO?rT2TMN|d^`MuQ}SHo7t?&()$ea2|q>VI8LzEzd*K8aK5VWpUY
zwzO!Ay<UGm)j;6>umz8f*L)lnrD9<g5r-MnAY(L)&(rYv6-#ffIym3udZA-9VJnAy
z8(uE<OJ$<Og~nCZ8yc7gkvl%Sys9a4o-8>>z@b;zLPIS;gXvm9>uPe9ApOuef|Jp&
z9PUR+)?yUp?B1+bSuKHaVzA`Yg0)^|ykvIU-pLvC5tsGk3Kx~-nsz>=T@{|*5^;Eh
z^SJ?e<{BESR5Oo;lA+cL#~?PLl7;U?Q#VpQ7LNojO6F=N^#MgT6cXOz;U!@v8s4Z}
z1aKue%tdvWQd!+tkvSzBjjUUt)6K;YG$db`KE^_e^S*$qLBOkZ{$#-0{^rZTZ$ego
zhXD1v=tq@sh;G8CN*YKco3O01t0r`zEInAR%)`}Ko%Bz_!Ea3$L|BZ<EF6PTI&gss
z)q*1Zvz27A1W}+mszFidgh$vp5RkytDPbCUGf8K#o3+OT4V@Ce4B+lDgwSP={amxo
z3AvzRr&|w!^0!PUpz!gId&N?iVeZ9?7Q%8yulLK_^ygUXtdI!R-u&xCVmTR>Mnjq)
znahnYOow5{JSnj`Tuz;a@34!*UERQtE5SIbt5H`5rU>92((@x6X@40OSLor#f3WI_
z=u)U#p|hm6v8}Xdv><P>X`M^#N;W0CaMtBrrk7n=K?}RLN3bYBsQssjnHj~MU9bU?
z#vT81U9Oqv3^bB%_+ZcHdR3|cPm6DQH-=8v_xgFKVlZc3FRFEANo(^SQ3hdCe7%G0
z3bnaweSrM%Dgu1b_U<Z9x=_{WIRqPl^2c+i{hFzQ)Ims@@(=2UrI0UaMc%O{*lN(r
z9mIWpUW?)ih=S=v+Ek4Si>>7z`~+K?)&nAs3aGUOhvGoWuLO3@6QKTV{?~b;yRPf>
z(qCrQKbG?CW71eRBI6kngb`mlb#%v;mpv}ftbm*O^c!`B$RaN1<BXy2nvP-pV1+_?
zNnFn|b<di+t$39*tHdrjKN&!jwqV8)7!_0))_~>OeuPb7wPW^_hNT7$*VZF%47-Q~
z`L8In$ohLmPx41xG!v@1`b~+-A;u_f<_9bH`=p^PY6tD<8YL%c8xM#|3PLtToyx_&
zQ9x8-k(at>IeV@~V^n8;qdQZ&^;<*h%QT6<dwEz|9b>^dP1l4{V066);?>WpLiyQn
z-kX@t*Dss&X+CRc9C;zAwKP7^v#fy;MJqbIQ_FB>RwdMsI9$**IG*m4kSeveryPrM
z&h!ai+<mbNVAu$z#8<BRe6~Z1hE#R2H2_{y&(tq>7(2+;<lny@HMma=vaZ#uITlwD
znWCf@?RlP!m#qcajd97#?rPrk6&Y%yhBdL$c}naVJTT1LBR?T5(V$sW4>7s)<g1aL
z-g@1Y;mnim<k052zNGhDG?yBk(<K?k1(LWPkBt(6FD{#wnd50-6VN#fRBMo_$pfd5
z_V1xw?@4iu?rr5ejB~oC>+3<bXAF7E%U+uCUwnxpL6z5&QV-CDhY1lUWf9*O4&8As
zdGWM<vScaZ&33vpI=6*jU{N0ReZ%^ZK_rqXchWK0g}eB5o5iF7D`$jGWKOY3i9GE)
zYUFl^VYRNm)E}oaF|EJ9??iNTm8OAZb4we#ZiD8uWBWc~ufeVztl`7235e^3&P^4!
zQHYUqJ@v?VAk%On_S~w81ig(-TKEP331~V4OPZra%5QWELtPkLX1~H6!4jIG{zPwv
z3-4k{3ZoQ%fydRZUxvD{?z27qrPg*V^5B=}*8x__y0k*Rce6CpWR>-&sm%#6s&X0+
z@SX7!aChc2?M8YnY6|e3^b^qP;%^vgkq~!ob@qMrTXz6yAWmvJe9N5lT`U<bM|PjE
zoIFS{Tspr$d_u>P{ByyD*vcPMfk2>_w)sjFjMsQDMcviSbdTNGI0Y(#LUi+$@h48X
z-<mokPhAR5oeehDhgl_7UZ@68AG#I)?MX@cTl^$sKk(OF_cd;#<2KY?=%}Rk;M3*F
z)LEDiF^KfOkH*{zdhturH@`S`X_&DgxmYJQ*L23zS8)3^qagMkN}fQScbIX^Q#mwR
zJe;Y3$cn+I$e-y8SuU|Wgb$4x>bY97igIzj(z#H3T%5h#bTIo3hPwrrsq|%E`j%<-
z`UFxG^m}E_|9Ia~6a;t_65g*~elr7aHN&NogP)HU3;!GhcdlKQ{p#u<uJga=-Vh||
zd?ys?csc3?CLVTQI^#YQ`93p89|{5+V}<Aw2~KfN4%FwrZT_J<7i>=QRQg)_G~7tP
zE0Vmw9Qg@;>?zd-(*g7W=0!ROxX0k7?h9A3mccpHq7ks7Yj>RkuA=F47(qPT0z@+B
zp^1Z`5(lsWq@Lmk{5!A*85rd0#B&D9groGgpQKZQH@WpS{MTf588NtBSk}GPckPU`
zSFw+4p%1?W!Kj0v9O`GFFZBGG+TWvydVUjjKC6}eZwAqE2-eCBWMg4#&OvQxhT>|*
z$k_)<WwtQLxDg%$nwgE6<5PIH5}XW)i;MIB;p@y?{}n3${I$C)_Sg{DHYyKMvZS~B
zJSf=EMNIynioAk_X_J194g$Y=N?R=;ZnbVrE@;2Mx^kX&y+}=CakL(#{VcLsA>AOy
zhh3LIr+eX_+t<_(A?7@;o8RD`P_EvSz@cD$GR>KIdn^>gbAD5Rr<o}4u7@VuWlOY4
zUY)H+oXltneAUX*V<N{PvBV$#ZKRTwV^Tq$j`~`mi~7o>hdPEtVh$zS8y9X=zL|zr
zMx-dA8E8$SPnk!zOqYC^e^_dUr+BRaZ-SYQh0n}GLCPoW_zyzypPn^2%o-LkvsI;(
zv=*sWm_sN`n2j$jK0`Y;-*9x5(N7p!j9*h==CjIdfK<K)1L-#nEz&0RUq2Ppd9lAZ
z2&uD4={?I%kvfwoqB4X-Q;9*!sV)^2c1UY7m@yn{s1k+M&4mjouiN}|%YjCaqQ+&g
z8IgtRmg>cT(qy9!--I#g`C<_b3NnK6w*HbBgFJPzNsmetKM2uHl!<UOk6s8;TEia*
zr-loQ0SvrsFOH)Nkh&?!lQ6aMy0QGAQ?uBIs5aNU@^ZU5N?7sjir!7w7A4#zL<r{a
zkQ%D5aFJYk8jRV7xFf{-MFz<1^@KvFNL9zRNRxhySVd25g-wA4qP`1%FMRlvQ>zlH
zGQzoh^cqWtO^kG78%nPh3XB~%utH5ogb)352%M}q3cfzSfCL310;#!xPQOP7h7;eX
z8^=t+w<mMKu#1<;%=b0L!={&`8DNLij(gSr6*p@rs|vCr%i{WFEHhSkGB#C__+Z2;
zeCOiF@x){BhTzBB9(_Lv;JY!$*^jdEafhPRF{Rr4T(5_`XYoh7!tlF1YrUuUkX?oq
zki)8QVfWn)`evey^q(&`di@4MKjv{5#_=YH`*Jf8k8)AIP5rhK)vJN*!g-yJNrz(m
zxsJNFvYbr}Z2Z2m29t>l!e^zmCG!M*Igj%P8V2HgMjE?c*ZEnj%#|TjGzTxmaUc_i
zJnq}!Q~ulHJ7(9&1i4#m@yt!%{D%8^K<+>SCB)#M;doW<VEID1WZ~(TEPJ}Guh3Oa
zqx!8)&VpZ*NC3`)JB}tsJQH+!$jJaDM*LyGem^I3<5b#{S;M5?!n4o5Sq_NC=PI?e
ztQ~8nxvsP?9jkmAJFY?R{2+PJkGJ1(i|EFxP|Qh{54PWRd|bMa&>$@C+K+JpxDmLb
z+_WT+6F{=CXG^V{!i?H(t)m@dY{j{CW)0Z|bEsHSEbunwvhe2TF(A!ZZ?xcrE#2Ky
zyzV-VHp~@Oa3U$S2*kJlUZVo625Ky?5eqt0Xca#dNNmi`dCmGM=3a<vtal)9A!{tv
zjy#l&(Xiv^S77H^1mmAw_KmayMkrfU^zOW`P0eQJiN#Ya?(S6VrfMClo4hf_a};C_
zfh3$Dv9WLL&wEc4u$sfw8u!7p8y!${3>vxpHnttpJUFg$5s}^$=kS*<6mx{Cxu(Qt
zW5_8R+j5N;{OlLAhv}B}G?(zYc~8T2!NW<tHf~y!hSu)v?5cBCCCC1NnkI?*jdFxQ
zNjIg%@RD$XZ~d47`|^o2qs=*J(wb%6C@NL9e$0p+|MXG)RlR)9Xbw>$w>twpedYML
z!N#F<%AP#5ab1>)<A+r#HD+h?2oJ!j7_sjya^+avmcd=Co=)r)61$jNb{^NKbPO4~
z0an^?OC~DaNVD^e%upGazQ)ih@bnMH63kMIQqhRFpxJ}pOXq6t>+_?qLA_mk_w}}D
zd_!SzeUSvog6sGH%l(JpB)L#eLMbcv5VBoI1Q_z<{<Wrm{h`7?G9oUgON&BdV5R*B
z8zbco-kfTlGDcps;sK+if0FyjS>EEpSw11sxn65n?10ASE@2s1)@4m65Tx))_nvD`
zo+#(b@=C94N}5pr3`Lyw@*GW!_IdjF<fM7AnW};pYx7R$ovoCydu!+BOzh-C%>RTj
zrH(K8O!A0m&O)+A`mh_5Pn1gA1dd+lk>2NUand!W!a@wDu|_hq=5fS0kXSbq!g{AN
z2#gd;n>zUwfP{*|1o>M+F~P2b4lUo&0#~=Tq(dUb>yldT46#ApY_IfhO-Vb_Zf!q$
z*00Rm`8Wo54!x)`KN_@%qQ<Y9K(LscJO*b`<L_D*bLQqzHASNqu4Q-N`CjpLInmjC
zMJN1Gnv<qCe2K8`?FU!K-gX&-F2&~=P0NI-<IEcXkhyvJoF42Zs0>^CCN}<&%GIIw
z``d!uz|a&f%5D3s4V#_7;M9SM$&JVjU0pJ<<jiN6vOl|Y<jmHN5vu9W&d)KkW91`k
zW#At<e$ccIG;hYg{d)t21upsOspgO3kft)bunj=^Ro3AnrA|8EHml|=G*|GX{;vV{
zg{RQ4BSQIaIbJh6bGM6fQMpIyO0Op-|84i0w8&G)fGBz#oLjIcwY|l><VBE*r-|)<
zc}NcC|E{WWOfZNb9^WrCfPeB@i76-IpNkvJ7wD12>1z^1$ozk$Dl$h1WBtvzG>l_%
zvLu}<4Y}l!yd7VZRcxi7PtVx7%kd<6Y!oEp8r32Z{U;@f&uFf<p}Y?QZ3SdVV6`?z
zm18?bOK)@QJZ3|6t^L?yFOIpE|2??4jmItgYnzTtW2>%$l7|Q<2P~~f^J#5K>!tr_
zW4jJF_SE3g$=2YO$ku3yhxw(J(phS+s&S8Q7mP?_d#cZ82hnX5qMHwC#f#yw;7HFw
zvfvJr!Ho#3u@Avk(};c+@YjdX4p&AVa#MqsCJm#vCJj4@7J|vsID)~8<{9N=A&toU
z$v`Y0Sf)f74bE+e1VA~SqYwR%D`9|SDXn1}EE{1V6-_V~V*0${qHezQwL-jsDnC%S
z92UjfJ@!xRu)e1vOtd5WkFuFx{2^8*<4W!w)f|%l%FkQI@rO*#@hOOe=tsB73O`?{
ztvL;T@~^L9sa3|@!iNmtquVh95%p`?5y}g1(FRfq>Ju!K043aSo`aH?67b>mQln!Z
z{C%`ftyLByT9sfkp(S8d3Wh=5BBx@o%8E3M*E;&PON3KQ!D;Ttu2n4*OBSIw5^1;p
z!|tqV!de`7shj7FeNovFi_xC~*)1J5eG;_vbh9p<RG6A3Yf~1ZGi^xi)p};UD&!^C
zSa4mDR{z&WU@Pj|#)n_rh<HZ(hTtm=+FI_00O0>}^78i73cNnwU%V~NqI7n=QGZZ+
zzFi*0X%3?#%o>#DZL6cIsjy>ps3eEbQccS-m_CZ}f7LK>4E4lHs}Fd;zFYA;IR#BV
zgIi830w3h)g1fu22~*{2SNX)pmc`wZ{E6VQW0e@di`Rv6Wk*&lF<ra8TBv47`>hR~
zU%m@fM*IKq^^L)qeBqa|CQc@{ZQFTcPHfxOJF#thl1yydnb@{%WApp(R&DK<{dVeh
zJzd>>Z#7Pz=bVagsJ2e1%B2O~<5qXdV@Y%OY3D9z3%cYCN4p<Kq2JzBk@f6*>$hhE
zr`*jWt=LDa95gw1;*1Mi>YJq=$Lu22P@PdrG{v6DsHzd5^tgG!ozgKh63sTpeRK5j
zcFb+d)4?S`Z4zB9-T#Kfi+&XT!)-f1A(=J(f585UQ_(+(R1TNXxdT9I6oc(?LWxW~
z1NOHsYbL*LgiHO*1a8~PTrq$X_rD>&4ZIVx!b}dA&V`8~jGy&w3E%K)Vn{6+S~qIR
zg_&;y{kV?~q0mN^ZzHJ?2q$~MuIm95z2{8p&To#EzM=7F7~h)qEZjXN$-o~!pJq6j
z9Dtec_)+d&$&kL?x!As6SU*R;wws{3s4Vd@dhzE~lO_XheEZtO!MXq%9|6(K!Gopc
z*I;1R@gO{+0#tLUN1!MhS6c7R=(gI=N17X~%-(#P-(LTF5D1Ffwz-Y`UsX>?6(^Ti
zU?4W0hv*V6UO{g!JYq-w83N+WwgD9M=JW$psD%SlAh_-E|Ecx=KCxb69K7-EIL`kc
zVv;+RGNq8nE7Kt9?0CFiYHwH0=zcUaf3mjU1adsmR#X-fbaQ%Uq$dBZnlGK=zr7cc
z8=#2SiKQ0*$&q}oRs7_hT(0*AGeWu=_ov6t2%14vi^5&Mo~a?VpWv$;LfQ{BJJ5Wb
zg=-uxxSq*?X6%iod~N;>j1#p&ZBCc+k~NYIE0?nny9XERBJF6q)*@as(rF{1A5{cp
zo<NbEoB~PKBNZ0{J@1Iyzjk^5z<B=AvuzIJUc#N<J?1u=eLZQ}o?G=cfuFh6bmJkT
z^tgdENZs@PZUL{s-~QAJ{&3T|?`y)3qD@Z6ie<yK)B_+rspIAVe`i{(<Y%MleijFf
zj?7APyv<O_i%O(eBp#0{lgUVMJT1|zBmz5+$#GZyr*_ngG*dsJ_<;Hd()lsXGLRel
zZ~=pG_b7`x#;j?ioL+$3Yn<AMW4LhP2ID6Vf%AFr_@_l<H|5VMzE#5Xj~Odqc|@~d
z#G|AcaQ;jJQZKkbn)&ByDOVf%MeO)z>G!m~6<PpI9wlu>m(~cy-$?5Pt#E;Do1aww
zl`D1rcUcsj`LyuuJ@Zmz3|pHCqO*}yPpj$u$A4P74u-yJ{vBBPA6Ozb)haQ0=-1G;
z^%rdFr2zAeUCZ}EYEYCBr7az=lH#doSHY2un<bHS6XbzHa%O)-Hs*Ux@e+s+lRES5
zBIw?_aL=;aldevab10;e>ay|_$h4Uo-E*+m7cQ*5*T-?cIUjyeKU$oB(v6dPd~_#|
zyvM42d6(t}9&)dJso({s-h>FDr;7VXP#|-2v;W`kLsoXCL_<n4pl|lSs!Ohf?4t*C
z?_)fdlmT9U3GfhvSR91@$MNgDX+ZN22YsDv+vl6kF7f!+vPx&4vU-(w^7vkZfr#=x
z?5>z<hlc7g!J9ML;iZ)~b*>jqVDM1)k^1;+^FlTHGpAp170~)?EEssF9zIMs>2=Ed
zdZpZLb+Glk;GuU3)XzRTcwg|;6X>1{g#PpDn#0F@GC>&|^>GHz$7=QgB*1ytE9J};
z{t|r16TIKc!f`*wr<|IOm7#szGvRox7FE{<yow6i@%vfInKK??TJucaLRU&zkL!Cm
z!AO41W{6Pj$XLTERez)TwT3@&P9NV0QTWF~cbk+o9FRPev6F6|^h?+p?pOCxf{4tE
z3@uz`Gb26Ky`Fea^+-lz)`-tTghvFHmhAiww$^GaIkFG!UBZThw0rMMwz51kI?3By
zsjHd&7gpeb8^o7pm-N{lv=)f{!PMhItloQxg+tbcpc!y-x}Wg*EvS_KwN1j<%1`=0
zx~i-B6e!KGz*-j9t_QSkXkzgLpYgu2wnod2W%~@tot*179~J)k=e-77nRsO@>=LsI
z^BFQ!8`GV5oCO7mWAd-$sConOA_LcZS%!^21K^h=aQRb?U05H}d2qUK)O+V>`M&!d
zU+hTl-KSP++?7woC8{`U0%)yZMZktaZv3w#7dX}Q=n`|(Q0)>X_~E{^h2Xa6$lbNv
zb`Sk2;ObX!g4STFRkbwPDpMPB-rd+JrEt4tROpk2xI^`-op`;)6DschY~R_OV}85k
zWnY=CyZ3H|Y5m-<|Mfulwf1dwz13~`qhRtC__Ejiv~yVf^<L8b{;1FmbR2lk1-@K-
z0ktr$34u>*UvIjCbnnAk?-Ro+_X%gz8Jn@WsDinuU4rD5C0T+#a6uipEEG+G2Q4{$
zYVXg~FL%^8FX&A!0_30j)JGpl)<$nCeLm7>f_~EFxeB9OIkT%?(x0O$yGJUMR|-;R
z)PnhTA9<JTwI9@xZ4Z?5mESwqD$~XUx}GVNY73TWcdIlroTlU|Hv1?y`<2*f&eMNa
zycF4e$Zyn@YWs<7yi0A|X1AFtaef80y(G3B@Ved=*nPsf-ch>Vxt|`2Z``Au{!utR
zyqcGbY&)Txm&=`>*ec7!aGg^q#T6{e>{8*RKOM?d<n&YK^ewT|nUAiX$L9nNT|q=g
z<{6S&e0jW4r1Yg-0blR*KOdghtGhqux*4$Fhrh0;zWzYJ|Ke)K3>LqunR)km)-ZeS
zs;eo*n<U(H_e16Op>)y^6F7?{yp`HZP0xxPGJn4j6*%$rD|u=?_Pr+**ni&Ny5s#K
zh)fxdwR|aCuCly9(hTDF11tj-8r>$WfI~&rK+-eUHBM;;_9p9EIeT3n-sSU*kk<3n
z%~4ZZ?`yJ4(b{6_y(<2;Ei^`dr_Q6m%sp3@s;8u>C%TR1@~(H`)~i?3vNqj0-_857
zqptGS;sZDGX8!HK>n-yd^!2K(CTO4O!*(a_Hd}SuT%>1Jo2`dcWoz<-(4e<@%dKCl
zvla)TV(+TvTa7EqW-rlU-w~KW+HGg~Z2~6ysdupwUBEiySR&d#?<6SfO&5_nS()Q+
z%1)&sxEE14jDht0dRC{Ec;?fDCjh0VA1d(NJxD>pfBk-zPd7I>^R6=1=~;ITo4%ZK
ztTcJ~B#pjCl8LrlRo5A-Opuo-vQ6Qo8NfYx@QX0;53a||n&5o}1aArsS;st#Nd9BS
zCqKE#PX==1-OnNy9O}tnJo6#)cOe)a+>j)|tp3^j7ccUoO|Y@!5941|?Aun3PHL;o
z_@kvgf>=qLSMia~6He&+@Rtt#={5jaW=YYa74-tdJCUNS(xX`vy~_mi5Z;S$jCHV=
z)~u*BdJ2yT^fR>9n3u@x&|;G4Z0stE_cTV>@&k}y*fpOzOsk-OhJP4&i9{KWJ89aw
zjNXb+%NWkKbo~aLo;j8E)D4BQPrPpQ3OdQ@zM^pf*|C#tXwdTs+$#R3vN+5N#OfcO
z6bA(*K3HvbrB@63+91{@=T2chIs)Xt<Abs~m|9`W8-gWZIHnh7tobH5?&BT75>8~A
zYBg|F*XI=T-Wf%y{*0Vcj_iR%`QZJBnS@SOlJnU5w0Bim(HU%OrSBMIYK(_iQADMM
zO6Q#D{0lVu5ZNYD?E-^5Kfm53w;D5PN!gitS@@Y>d`0OKmpl`GLaFb()JTcBLy>q;
zLIg(IKgWg~q?o|8MuhbT!O!Pr;ODU5*C&uP{o3#8%L67+xjd}{bCS7$n27#5Xad%P
zDFmF-gxC9DvTi9SE;_zzr99>CK)qVpmLnRVgiGR?D(tASE^TGs*jCuo+!-^vebV%b
z`sKkkdU%pnYw>*Y&yfZ60Gmt-0IsHaq`B%T2t7?tM+e3^IH7@m*OC83EUlp?Wf&OU
zQ7j%WtnFWOe6yJsL5JGus&nrL3%UGGb@UNH@vq{f*GNPP--9l+BZm>RL=H3Z_V<uj
z-c)N&b-Or+T)Q(wte-=~u5S$cFZ_fM05vfNc8P%au#wV*c26|DrR)9lo~7oTon=+B
zCBG6GI-&Hn+k_Bsoiz?5x-9aT2Nb9aF$3Npd_b@??-Jw~C~Xlw0(g5%Vu$T=GJ$0c
z7phf~TZFQPCMQ;?#T3!5lh@7+CDYQTS}f!F9TpO$kBJ&IEjTo2mr$JHpHnz-2zvhQ
z=eO}U(9dcwAYG64eLKGD<W2~s_$;qsANsxxFz=YML}XjABIR8dPM*{#rUaCDiWI7-
zwML?Ya>~u@xe>*)Lgg8H4$Exis&Leh5Tk{OBIRWJBlg8Z@!G5T0Pn~nHm^AYPv$bl
zk9eYjK?^adiV@#v=@}jAlaP-pI-A5ZwO)-MYpoNx_%i!mTq{4prl}9l^-e9hInev|
zfh1h;l~~Y1O3a*iY@>v^x(_rNY7e8#rynftW@Sr2)!W;h!zwq)7c17dnU_M-X26N?
z2$#yYhI>n*o0t+WBqB*(CZ{tft%^>W{3id3e*cN%*cYa>o(Erb4D#+kaf=^H!1`IA
zyLzC5nTcX=n9DAn4P7hBj@#&s<^rEcRVs@2fUNW-W(n*N8n6^o5CKkz!D|?3Lo0!)
z#?Bj})<fVM^T*uO!H{v&P?5`n_B7&5<=eyziczyL!lKo3P$}MpQZ6x(sMVFh?!ee)
z<GGAZdh1xd(ut2Lu5COt^uMxI!jhG-Z2V26sKh33NEf1eVn#OD!(-n8>l7fd99Vk7
zFuMnxLOz`F2Uz%^oB;`twj~BSe}}1((%Kx&{sEI#?&l%U)GnjMr;7CR<I`tFCxXoq
zF|aoEcbt*@ts1-Vk~&V~yRGK3gtJ6!HE=ZG)0bKb)>*aDU&>ilh!q)Uh@EjQP)PF6
z5}UGNPKZEDmml*IV?1s*HBN%bh03C)78w(uNN3^C_)KM3bpkvgs(AWg3Q-$7nbd$R
zG80C-e|-a<&08B3R61?lkeAG-x0i4wm5_u+6gRhGfMP>01Qi%liD<#}YIB_72X21@
zMFgTm&&@!>#PHxDU@}rp=F$B@Eh;^O61B=gj}q+NV>sLz$9_)j2fG?hG*+xgKdjXb
z{B28mTU?=@2M?@~>+^*PVsSOo#0BN={EIb7hohF%D*w^DPdgOOK|C1iIFXlp1HOG?
zNL&~58_HzGB7@}5WRnRsS_LPPOde4ej$qs>2Z=1F!K?=w=CK0=L>1Q|-m6cgqK=x&
zbEH~jBExTGI@3kC;G5w3X$+Pcw!m5FBCab!z^PTbh$-+j9kIbS(xE#qP@DWS$gHT<
zc5@^{hZWu6kj%{z4-}`V&*|^ho(O4aJA4h7!dY9|)eQv3;-TB91Gq8K3VlgC0X4;p
z2s;*RJ_+QdGX=FqP(8#!&>Neas2TN#cZbvOZ{I-*WHK@MBj;?=8!m%_fT%{XC?V0d
z;8r@LVlyB_DtmKSGBsFOf;HY6`d;M^G8ku&)q8!<Xv%D_88tKQcm~Tsbc=&e|G1DQ
zsk5-G7&~y~+ZR94vd}07kw?_P-DEpgfo36`22yaE1daeCO`|q*D(!+mTX89)$4(sf
zO^UAbpN7;sUcJmp<yKOnsrzrxHc`)tC1QJRODaH}S?Tw_j-bD1zc&D)_*V|G;LVC&
zvr8mcV)McphuMnUZvJAXsm!#WQK<;;a#<a)!B205yJAiS(#w&*IhLH6OV(dScG;}e
zZ#AID&1l=`lB4(*m=hmpWi*xZ>l1mt4!&FVs^!+m-~AxkNnW*#A!c(-LPuni7=Hgm
zS$P5?1;{F%pyNk?+ktk<I99@L6UXzffPBm3Xm=%a>M8Ol&Aq|N5J}*}`OY~DElziz
zBI1cQ((0ofelNfJJEXa2F6E;>PW!EOWSKLE0eL9muzqrRK^TWXp1R0gC37X3&i2x@
z!W*<>z*K~6u%e1u&Cm#L*pMt<^r(lcRw@GT{x!v3zCG+g=?y?&dQcrP#VvtBSlpw=
zL)WG<QQ8WulYub9%)u~OCM3-J_WUVkxmSr-|Cg$WRt6y)(xGJz*;e#dcZU1tA#A^X
z3I<aWF;PC6|E1>qq(s$fP@{wno*ap~w(l+^7HausCXwn4^J^nY-wLyFo`r}tCYdY{
z@|^bg=Ld!FK4!&F^aegV-Wp95H=fq+t*0OE1+@d0!W?eD<<?%f84`8F#6na-wJ}ts
zwFuam^Gd<K5;I8=!Xvju(?F6n0}mlV1VIUecGwwe=!xOT636DVM-hj$bu3LmNX`jH
zQBDes43w^J*ywiVE?W%rRbiS3DgmOv9aedoQ^<X2%~R2o4ymB0wcLUtqhfDhqjU|p
z#C^bH-$DxzE)L9*Rt+1i#k>@aj8@0%zUKtJeC1v=3B+5Nvlq+s8Pus&Uy+?&%2yW4
zTJMo)I8zLvGD?<FUbfFX(%vS9lOoz(_RiYBo=I$jXH*31Cn*b)CqI@;*+(02xEQG>
zJFuvAWFFbHn4Z%=rOzHNM%R%RdEnT2rRUY3EU7nMgFRIMFXjga0Fw~9rXp^g@?sJK
z0(*%Y;}@-3>P@BIIKhur->7~eNOLXF;$Y{rmdMCKKm1_TS|mgJ@`e@*JSEaqTtyF|
zw<i;m<VCKxO<QyDIxrq39+yjCtqn(9Ss0}*dH!sM|3sZ23Dn;dD&JUDo(g)liLN0?
z*$axKu7#JGYy$neND5?B?N^fBQIl9qBq7}s%P688#xgho`ybDiI!(*WI(54bFp0%$
zAoE<%IwXj}q<2*u@3)b^XBmIq6crMI1OCz@Bm=wk|1DCF3C*D`tmb`(_H|KF2YyL;
z=m^TCJX}0dNeZ$*>t+2B7&I#PiCv>72K+_9sIV~P@+raVh0VTEmj3A;hgJjAy@WHI
zSQ37MZcO{bXsv^55st+ru||4;NX*}EUoCX+QzfP&EbwWK&?zW)|97yI59l?F8(%8R
zZBYR+lQa_V115{-5&@_$&~!x9xC1@~{K%0U^!vsy)Qe5rvu-%75h}ue=?IFl&NZ8+
zEIpWTim~`?cJt=Wmji+&I$T)g95ss0-NnSUY1^PHkXn=SUe*%ZF#MQtus<TI1)Yp1
zckARI988akq1dTTWHQdYB(q$UM)!Ajg0gN*ei&Z<&19<I2Yt~4IA~7Y5ZGki`8f7N
zlYvC763&rgFr>O_J1O~Go6QJZoM_ZAT=|SbJDoOwyugSFgw`bH40sbXe|rPc&@4Y>
zm>d5c@KU^S>DbQZpV1g7;Bd6z2t8^chtSPHfEF(1O{Y2iDs1*p(&_TE^r8^iUXz&2
z0R)GwF(&#al#P}TaFL5ZjMHU2*k1lV_)ps;Ph8S-s7vX6G(?24$G=qMcxGr}eEvMN
zA}KZKvlo2HV5PDV(t9?=Uo??&MNs!p{Gs2XD_CgIP0c+J3?*jDx`2Ebw2ok<t85S0
z)evs3h^4;OJ+gM`RUvb7(A_(sP0yDL*jy<Xgigw7w`C|^pq;y>6BZ<*1bx6IKDI`m
z#NN#`9U@{XL_L2o7yDuhVrdYQeH_OO`uXG!g}cLFBpq9yapXom{!WIxk*5*UHC<`m
zMUGNZ%&FiBULB6q-;rN|9tE+3<N@T8Lffd$I(w_+Z^gCY4=UQ2UdJDSBFA713I4=k
z;k<TIDs?l1KrFCLY#%O@pS4lm9XnzudeK;NO$bgJ98}mFtN5RiP}`s!L(i5M1WR3j
z9m^dE9>!1$eK^<(Fk4uJ-IcQ#^i*<7vff5QfWnMYSpQ<Ihs_M-F6a>q0Om&DO=>It
z$L>-_14MCB#bIh9^Hl95H%8(-ngILkN_?1`_`P=)5T<5}v5{s%6Nvp2+*zlT1qH|*
zCJWOc=3R&p9LKQgQ7b9gED`DtnJ?(KtzzfPK$2`?nRD6+F}f_~qhP&5yy$V`#vVsB
zeJ%9^AB!04KFBnWJZBZg1FhYX%jW!-9vR?yjE!Nr(mZ@@5VXm4oTXQRSUnsR1<^R2
zUn7A5cn~3^kU`2L-VsrH!TAGqJP}^t!YLlx0;uzj#c)g{_m<Z}Z1TpY{w~+e3Cuo6
zaC3S(U%CiL42CHo`#v=2Oop9JP<~yR6o~JzsiN8OLQYQ;B*PO|5gaX3lAa%6`g^dY
z8j6xA_SAIqf?Q#x4RqWX?m;|w;*M!Kgd+REU+5hhALztI;bP<VSo=&d{s~X_BlmF(
z6m}XS(rJDtZVmv?T>=<4pEXYJWfomZWKPQ(i4?9yDQ6f0h&n%q)dsY?6Sr|Y<IC6^
zp?UQ$%-{^g3Jl?c7T|PI9!UaXuDj!Jz9XktcUYqPVmfmjZ#UMBQ5;w<+@8-mJrhkJ
z2BIflsxC=GY*;F_q|F6jk?dw{iEs};WcuZZ4z`hpOC$B424l)$te7q$SW8Cn&Y0Kl
zdCc&i>B*>=U{?*icH53UHF*&mpC!6uJ9-7k)k!}Yj>2f)Am}aBuWpFmxVIe;i4!wf
zb;y3yNiktc`Coz{N+{#Xwy5RR<L>YPu{|15-Z@CxL4O*x#ZQ8I46Z8lFS#@3Bkv5r
zq(Cj2WU?<*Qj-;i^$<`Q_sx^CIVIMpyrG5jo{|l6ZO80C6c`@)6gap{nuHRDl1erw
z%`B4Wij_e)Asm1ezkCWibri5v3?pZw!(KWP#q=16e4ywy`0J<o%H9U6Mq?=fBf~k6
zk;VR9AuY>h5gD;QFPvV74!wcmvzLnEWO<tEFf?Vbjy(5dxd!7ccnwj^yL=HLyK@=j
z1w}`=ikf&wQ{qYaM-|&aYt&Hr{7L8-4SH%P?1T74Cqc0SF6(V(p0NI-sOq1aISJ!%
z9&d?hmLdA?BPQla<*wm%62wUxFk3W4PzC4;A%gDTXUqSi8PM+n_80UxX>Yl=ZU~eM
z`|N(41H`r?_lfC5kzK>g2rVu2AJ5UMPG3NR+b+rTpx0?jDDJxR7+yjFychzYP$MJF
zu|oNsvw#{~N&rk{I&y`yR_!M2IRz`+*F`4b)B??stQ_$TRgU`hp?IVLs3V1i&oTjP
znHtYh)Lj8fqDiKS?~)*kZBfo3N4^mlH4s?}t|isb^Uu9n5M_qUQK7GEAikgI&VEJ^
zEd!OOKUU|OeOE~OA1~;@d{@q~mu!jv>aR&x2-o{?Mby6uFx-k%99QZLiWEhZKZp|y
zirKR~oWe~i9fgqf^m>^ofMG=Rgz#zMx%C76d4=0s(_M_H7$P*u)9IyHVWgU4bj%Vh
z5<?<`ZKq^p5YR|hu;>@@WK@=@`*_*?c2WOuV{rN%Qj6l5rs0=O=nr#1yWE)@g(`Xw
z)7iUQ{y0XjO;oea;t%k@T?cNWS=bo&=OkKit6;s@SR)N+6yy=9kOtfuq3$izOcGrZ
z+xTVs`jIja*p0tI$7cz8vqIt{;nIrTq%sytkOILTBY%uip?n?#z$&{7EvMhCvr;7w
z#R5XeW(}{zTZe;FtOh>^NPRKD`>+It$YBAt=)5})tR%dv{GfRGt9(q~lua1Ny0BYc
z%5pi%_XG$tD0K(kuV6zAz-tpJ^-?2C>hmrGIOkqaQqODxJ8d7UUf6L(3mTa8MOQeF
zfbEsTdz|Q5&=GOhzZ|dPhTDf6cV#Mm`ap3(-1H5ZI-kJ!8?yE7Yt?1+i*hWLjIOc8
zkc}ay`3oJ$lTd}?iv~}dL94S_pOQneAr=fZgE%pEb?4G9E01(7_!Sl?<;GN>^=v(@
zEspG?#ZMaut@NWBbNLhj1AYe0Ph&Gw0G+A`x9>qFkgZ33g~Dq%G_C_$3Zr1t8-m1;
z8?^rUcHd?6;zDOK6$~H29UeFQur}N6S7hZ<C1P3f#$#q25=CpWVAc?XsaS9Vz->Cf
zK3Ggiv^bMZ&nC>a=v;rf*gLk$q1H_U5K-o0M=V=@i)jn?;F3yR;u+MqHBoe`1BH@k
zdC8F{5dpUZ7@KrFc|2Yk;;X`-3x{l(jXl51M4ennNPl^-FSok#Wmi<!ZGU`*qEj1i
z?ZJvFXwD;=(Un_qHI6)u4~!=yYAFQ0z5D<6Pg6X{t}F{#ai<fmazs_D<qLpp<DY0Y
z59Z?rg^aH?RtiGF7d+*cKLYX20gm*~O2p>vP5fatPD8V!@({GDArnY7RVxsEw;_Qw
z<)QC?sc-3#UA+#C^c(mHL2^FH5HSMjC4vWb<}cq$c2^QYB&ZkbLHrxVjZFPv-$?>m
zDrGPN5gou;x6aI%aMx&#>KIhUOn5c|&AW&Mc0+_3XcEwHe#|non^EX&1Vk}PpJ%sS
zOO~nQU`Y%WhH3`qiK({!6&aDPib9mxp94u*LiE5N1;l?E4X5NSiT}~S(Hcj^sY+gk
zy4-X_L;Ju4h?$4z#gFo#92z?49q?b1>J<O+(?{x2U?U)8MA$&hWIT@h*k|aMgs`JT
z3JmCyD~n?c2mHS`=mAx!1K=F;N``zAcM5dkIB-0?P@GkZ^&^&dnJ6vT7vCmL2|7-P
z+VKi$#bcloNh~6pnCf3KeB=~{7ltKGt$^~oXq!F=$P}tr2IVHWWLNz49W<8zK7lcM
zLgeq7`4zY)t}-joeW9udd<)dqCl9f7c1Am)>iEPLm$(k;pr^OoI^aNBr{i)M4v(ll
zjKPJ`5(;W?_@%WDoH-f(kat#+Ag^NB7l*liY9`1>5<Df+=Rl<Up;7?y)ls12wLyG=
z2;A*-sB^4~229X>%wLEl63ehSpQyw|N-0Thc{iPu`D}KA5bKH1B&sV1hJAL1p~0~t
z1BmG$JmS8FAwy={Twn`2rVN(h(%N~&zA*kwZ12itg$(?*H{`(nK%_rSL?GT-d||>O
zQdm(h05bFs>5)Y;_xZpCOyG`55c1@amZL6S*laG;z~iCnzf)%fbCbv*9m*cvC`*(^
zLctM0bbw`m1c6kjsUhlZ!3`oH4~myK&$0b(nASn+wm7jS6d1XU2PsOsDiQ;aNHY1?
zpSqN-F%zssjBFIO)rh0Z=7o-uY7ZNl(9i9{!=33sXULTNIVMoJ?<J4oFVdc4bp1yd
zG4HCe6QbX#jOh=XoniWle6F{KwK?FH0{)L2Kd>v1ca#{dq0Sha60uO601VxC)8O{q
zUmTX=wSnk$Ho#ookU^)re3{4{0z|Casy3+(v(kPvz*VmnN#+<1uKmobaF6*kK2tyR
zv;=m42FMQViYtU5x4ADF5|ZizH9+s1|Jy*~HtNTBQBZ|o#99HpsY)#6s>CS}f=*u6
z?;5)u`vAvR7=9%Q*`1L{86O6R8H_^?jb)A^33Q-TOC2edzy;3n`^P^EdCEIdh_J&h
zwIoiKEyZJ?OGm{q-T&mKK>TfAr8MkdARaw1)foEgFX);~xgdYP5=S#02u|Oc0#sPG
z1-Gt1hHDHBu}TMSkNpH0SG_B~i4;g3Ly-5NjgWiL4L)nd>JZ?tbG%-nOqu<j@o7)o
zmCqk#*R&6wqYHTXvg2Er66#WwqZ8~H0C9vwM#L37y7OBir(Af2eqsknlBOtbxm-!h
zqrLv-4Mb+d9^+k5w=eE$ZS-_dQUVp(RKir&pjL>oTzy|s!C_V?dY0koJ8`l1Bq+(H
zxV+&erPz&6nTVR5@onKWW;e@ql1EPzA)&xQ84xd-@dot#B2sSih||j|yL1^`_C@9m
zXVuUf2!Ys+aYWqZAC4ItfwC*|<oXkA)b011LaK-*9$a|6hZ^2DJul>Jpt=m(I>=^j
zWUK=yR3Uj63P~@I*j{&tb5nJ^D9Zp*laI_vCgl`ekH%FlPEd!c8$H*LmedjPO$re9
ziTM;kN&$1@eo9a<>{hM&ub4H68H8sszk`5hZnBcOgWWCYiDwTP9WO6(S+caL*KuIF
zP^{>2x{(i8&YIH3aU$e7s~$`WS_EOj*>Sy>b-<IdRZLse*HJi0+H{dDB76(oLpg(e
zaLKugXz|J326zVp*#7RuNrU^Vm7LDEW0jylrvO0!R99jkarN)Q{xEMRSj*Xl<`Iq?
zS|Fxmlp#NPmLx(VlVZKIG$4<I1CezqRVGcf<IXS~Wy3?B6>gEu9u8+=sz5jlfFwxy
zzY#wRm|Jfk6;f0Pym6bRB0A%Jgh}ZB*)I=3e2cLw(LESp5~D!HU()90|8+t>XUr5A
z$pV#Biq-}dtTjj^mLD6$h-@H24+i(1!BJZi_MzgWOu#*h!A~93oW69~$>p(NnPe12
z(ISFJl0^y;g#DI8EO!P9YHm#XcJ-b9thMp)y-we=%oPoxwVQCi@ED}Rd!kC#-Vw{{
z$KhQ3YDf&KOdy5rZXsU7qR`FQCq8jxI0gQ+Xl0wI+a|$J*t&Llm@{L&wL`@0DV0YZ
z(&pCv5F%N`8#)2lf-c-J)Dz_aE81a$dO3pvUA3x;Ud<;*&TfF6-l@3$BI2l>qvVua
zOf*4%p5mOaNA)Z%Pk;#rQAz3_7%?C$h}XoZ1&v?XotDn)ixorKlrhc)*@}tI%Ru5N
zsFe|0+)R0lJ-krEYzZBqQK(l>Q(<F5@^|_mSQDAh9qepB=H1-NsGiuMNoFQDop;$w
zmkGB|R|5UvHDgVCdkfhdkT6&Z-%fX^KP*$KbQYBHXcJn6^Xlp}EJ#IIJhY5xfiNxB
zvAn@G36M<U0_OKDamK2p%#u3dbwKoUrcBy;_&Kv3RPsvd=(dj0OB_s6cQOO_nAXHW
zBc8>zS}8<g89wYRZp;#S+^h_i0+lR{9gJJ&>=L$If@~v&1*;x`%NXs)GFsEWyyo$%
zZczh=G%+pk>&S*!KS-1z>txQ2v!VGF7k#*fV|9o-nB7td&n_TiyN%sRe1L>$oQz|w
zDRPO$pBdZU*iGL#v0dlevo?QkN8ZK7s^$P4kwaj+Vc?HclL}g#g&qL%Kc%hF=LH_}
zFL-!23RN>kAJ8eh1o7MzjeY0Jd@-YF^0ehzzd}>!%nH(rIPef96UI0~nYA2$n2t4{
z#8?iG?ow?7$=d;v;iU==H9%H@G0&~Zx3f+p=6LOgUK=}BC1*m&r9_(*%VV-V%p<{7
zXkf)EQS4cbO9IscNh%uG*EY`4a}1y~ugtaZ*B%>sQBb5Yc!nOBi$S!sa@AT{9u&s(
z!LL`Rp5}#Bum#kSX0RjEgwbr&h9I4Mw;BU6wBB$isM>XR-T~^s7GR<#-q61{7dWVk
zmGmAC_V}D{Gec;mUx~OSimIlWT#D&-Hm!)PDNNh`@L+G?5~!~Q^ty7_wLG>;Y__7b
z)x+k~7zA`IYB4;h#J9Vd(z`yC_>w}|pA0i&)(DCH`H3RJg5lR2R-O`5@GAw*Yh|KR
zBl&G`2Fz(Db>UvWy@0cyc)omQ7?MN_Nad^v7I2UeA9T)ZBMXE1GSZ;T0qN4oiylNs
zv`XSTI=Arvj$HZYtg&(ML}BHATZo=<=PYHFfcj^vKC1M3cc&OM#`A5hg6UofP|yo&
zEKC}18fedcRpFYmIX1-kVDQ5MZN#=DC6*S%b|{!NUX|m#Fkl&?;E@kEF|jB?k2v?m
z7BM1s=#N1tkeiUCu|ggodFrw=RbYx8OVGHEhxbukfJ?D8S%;;9%wIRG2XSjWACb@E
z4<V0Tn|c1oudSGmvx)Wwp*fjJr$y9F(4}Nve~%75XOY?bfI3=*SD`+%ky+zbW#Ti`
zL_U_uwZGRQH$a%OS*6lEVmPjr3J+W<XV7j7%4_Lor)SAKJU8Ri#Cw(qv_6ZZx!JaC
zFr*XGw*5TPkh@x0V`ea|9uf`g)hwIYdP8eG7xVbvSNYm8p;BHUrerFZVgTk)t{Stn
zehBWH*jJ{-_%a5ZV(v?_?Wijc7aK68Y{Skh4K@sdf54>{Tx$AU%v#ggZv1J{<Iwm`
zo+FC7UnT1chD8s(7_0+gB5?Gf77-$+gDdB9nn9EeJOSaT{qr}FZxEC_V9c9#Q&hxX
zXv}L<)UpMQ<tgWfxM6!1u?HErwsclUJqB>?`4al#P-NXGd$JxuGs`vbE1J$NB%R2@
z+hzZ}_<@tI?v3gMHR|i&KVr08`FM8bRp88aE=k(mT3Qct;^62Q1|kpGB6n($&~6lu
zsX?XT`#OImB>tKD^^9Y=AQx2xO3s^K9WTpi9(vb>me$Kosgb1X)V66ySI{q19?Mhb
z1JgBb7{&5F@k|l_$WWKjh4$x?ffA;^FW;@DXcDL+nlZ2qB@iB_glOH|v*2-ufnH7d
z`EVFo1Sl$WWicELI91E}$Wl@YVZ}!+Wu}=&z>1Y)WcLA(QaKSQ`WJru2Dl*Y)$!G|
z!C|+Prn;Vf0C++n7+nXBnAWa|CX<cZ+PNo)4;{3bLg6SCkOe6*0(*(2W)ee74n!}l
zoPcgxju)C?qA%kCc4N8Bk)`*4z=KI)Ja-gn{#q~|Apa;u5kgmcCH>V6-PoG)tG5b$
z_IVG-%^n*=PHldq1u*9-Z^fVKk6w7rlv+M*sN5N!9eE0RWAG4T@o6#)^_>Q1!b}$I
zIC+|xQvldTUju96X*?h$G~te}jr&(x;13j2ec~>(WQcA-Tul=y>0D(%%CU%`L91gQ
zNv$KpqUTI`Cv{!JceI!rYQ@elFCfOjjVJpn>bJ{QQ4+g*3uuS>4(KQ>R*(?nI#)jY
z{uSweB}>)jVtkKXNcrO^4EsLEarWLg34D<|>Iwx~Xh<B3&X-yyUBP{f8xqC@bO?AH
z2ZPYC$g~>pDD~6X4bWr!N|Y#5FN<2l1vbCNawIVx9tFEj)gm?NKX{6BNg6ucnl!Jm
z-0K+*ySl0UIE{Z%!#f}I+7yyFz`8(9=B3I>>+}>Kn&!cKS?zvh6u`6at=X5~j<uL8
z?aL%e6Ppk0LQR^Qg|IumH{~ODvI5Er;5#Pp`z?R0s$C>%+e!f>c%}jF6b=<dFCDxI
zL$bB3hK^6$Si`n<QDg$<3wmG<K25Wo_LRtQ_2M)&&kQ}seW*)Z{(=2GU+|=?5DhNl
zH}ZIO%KaKAsJmnXHS1mjAfs(dLnzg9mrg+`Y5kX?zjMhE-G%U0O``)*EK>*SCAfjx
zzC@{m8}kl7rwso{>V(V7<-L@X5A2^NY;=?F=8Ec2@;;i;Wl9UU3+CS@y!RX5>!oKT
zU)jgafQ#<*{dL@TpUlEu_kAv8=H|{SJlJXV$r;t%)%g`R$hf&ZKXBrraJd9fL2WM;
z(*yId9W!KcC!G*VX#d1N%<O^}WTFBJxc=27!MB~ZYn@pBB?6bF%4O2NG#Q_>e${A5
zA?Z#GAqiE#(b|Sb*70%e<(v-_9u3HZ9Zsl6WONkU<@Chj+BjmO=~gUPnW@8XSGr4k
zn?2qGJ7CWBa@C-zYQgQ%q$PbG9zBENGxvBz@Zx{8=M*f0^voJOjdZE^dawXS$!JYi
zU-U#m7@QAQvROVcg;J*^!#fmHO~Bz=9-OtzXigU|rd3frE4?x)F}3b)tnVtfox6y_
z)hsbFgU#CCdp*rz7g*J)4$tv}S>5Ydh%PsH#GcU^&d+@X@)RFN9G=~{K=igqd)_BH
z&UesF5njrp<{CMjvQXPo)tmrV`%7Tn83SG*?T4wjHBi9t?iT><9+r+pxP*6XywkS&
z;K^NfMI&pJJGyN&#ym<X<Y#>-?kf&=Cu}x*ybS4DZg{zcl|%cjvaaHo{5NTUu3t~{
zXGd-L+)4_RJ`u%hiXxT-hc<xC^`fM|$uLH9o?40Z_FcS<w4|q@x|~4z`Y5ns!=e%o
z<bAA|Q7Nr)8XFN4M{Y@Wu*-dJ0UpozdZDLY=FX0BWiYwk4=FShI+oMn-8X*ugHCjO
ziJ@t*XXwKkG3*CK68sq{bK>TO(I`A?W;It{Nn|Ii7Rrid&UK^kQ~O<q(8%dCw)ci}
zxPrmlY~BgD<)y9^3~nHd!NM5xMHM^mlxqPNC7ld6#6<AA154!y-UbTS$TsS&(wK#-
zQVG_C9c7SyBk$EHzUbP6X(E#4joZ1t{pKKKLIhZj;p*;jjW2tx=%3lxdqf-2v8Az6
z>TLhY@uSdEBzMlS88G6iC@xEF)UQO9CdHY;J_8)=mBFNv_G93z<&Hs1MJd*I6I95=
zs7LD>!-Tn>yZPTC`D_?3<Xu1fkb`yCHp#X?GEWY69l6}ydmA$z#KiY#FmeO#ECkxH
zcxngmN)-?^%V<a<-IKUdr1lPG8d8LJCgs^(YEG{znDhk6NWSAXdEP-sGMoLmPMa(l
z>_4DlL+m1nPrN{a9hy;7<v_c3)_#e61Hzf17BNT*=qKE4EF5n6N21DlY2_cap#d?Z
zqQ)*qSt|XSApLu)F-hy&g__?6|KKur@nv$JrH_F<ifIOEP~Fgv2>#o4@%zaAS#8eJ
z=LS1w84+BijRcqIGOvcY4gY8)Rt>FetKu@`&z9%tiK~I{OWZM2Z1Mf_IW}C3H69Gy
zGhrmdVOuE{dlosWW{XMM`>vdY`7SU@Conf{nJE?QJmPnGk}<T~OTxBIOW~(#oNUyH
zzsCU<P+|eEnY?WA8t2jh4|a6kb&jY;893j)J9c#t3`>9xpSVsQ>@xZdF=sHi7XWn;
zRsq-Gt`86pzBKpgbtfvV(Wc!8cRJ%@<6@ENUwmGmr-&mp%CEQJ<*GH&+J56NkC#Ug
zZ)2cQgRx{2XS68sWLxK!CYO4HE>^&i%<>~OHCC&NI*8}sTD4Xw-bW30;iZN0=rB%5
zQMpFa#^OPEFit?>2|(OGEd^R$gkLX#TC8La4F{Yklrl{}1cwGNBApXlrkOD<lr+Z1
zUi37}uWCA;AsDZ7g;K|m*-QHLPK?qU92w?0!^)95e8w|brV#b|AyV%0lT|^tCGkM5
zrAWqy<b5YYnlJnws2nq!C;TTW(zzN+^0HXI#y)PWGc+Jo1*Q*;%b}&A)gMO16EJO1
zvWlEYG{8D4@x~z#tbR>~28WmG(<(s-44oOi2PYPc9joZIDVJXJRMQm%LHsozwCCu#
z8pLKpY<lm4<T_80;LWb_Cl)qcCpk&fqbVwGe*Q;r*sl_!Rk_m0@*yhlt`JL|(5t`O
zK7CtI2V-I^%d=JGK`tKJ8eUqH<-0VIyG=TFZ^kq7ej_+Xv=F;0X>Bu;MoEKzfb21~
z%7A{i)V`f{t!8sN3Z3|Zay717)tGXG(a<ap-mC8?oKrv$)~goMaC+X2<AX`|Tz>#<
z5t#%?s-mAmbM{$Y;XpP}b~@6GZ1}h(W68mNl%#4~`uA!wTV#X0SaBpNFCfqn@zsh(
zE(~o%=pTC<YfC$E$Qe!wOd+p3(6wH(fZy{MXC9=Li=36KRfD(OfiXh6dc`?`^adcP
zPR^Yngt4U<N0VfJBt~|;4K*UtxrrF>1{#Ps>xyfO@XyX+&8ICuDmaSh6h3nLBV0rf
zPEM+VP%~jZXu^>$N6ALYUy7iUGS!O2Es`+B0K@O`x7I?fesf}`v?d5BaClk5Jgb%R
z(JxNIHBXaTiZUNK%EZE&2tH6mIJ1b}z+{O0apZ}=Jo#-pDpOotIPR7YVd;XU<GN>0
zfM?A$S6-$#j3&gA^Gs51BQ!SWD1ynuN}N4MlSdCRa?Bx;s-LHY|4Z)dge~+R7DOLu
zl}Ic@6Tlkl#kr*~8&&d&FjmiY`DId=@}aMD?F46S{U2GiVFYMmJ5Qixm7l;-YE!!e
z6|nK)!Qq0-H50oY67)QfCE8}`54wcys4i7mE)w>Q*W#9)Ul`0JN(4$3sTckSd-6#`
zpRaffx?J}&yd*ZBV^X7~s?F>amLS$o47<yk{Sr5I8tR$c&h{x%9QR;9DfY+)d~(N8
zf7kb(qj6|ALc3R*Ch6h??V81f)m*#G5ukb#ZF@MH+=)6SZDnvD=(>kX)!E~Qj7FBT
zMtNE9;7~6D(T1uK9M-G%Tzn1JVaC}{uCO^>iIgsRsxsPnIy6pqfUiUwjXb9MjHA}n
zOio!g+HVy5qUv!NZ5>-vhr-FFxfvMNj4G@VMhFd!g;VezJd?h7i>!^kmZ}k<VqhKo
z4ggeX;L;&t)P<nMBe6j=qWhSDJwT4UVy&!?GBW-1*SuEDWjmVbFZJ>G?o+=NA$xE6
z!*$pQ6V%w3Ea6s?Fo^Uc4>t@5TM~~?#~9yCYf|Q)Oe?VD$_%@xTa7OBu(|`5jNf*i
z+jaEdngz*5>WU}A{Qe&}zN=}!PJtxRNM_sTVV5oQ_I@(zZ}P8-w}Km$TNH=Y3{<k8
zip**rb=NxjQTfc&ZbU!DXY^&Hnu-s{Tgaz@#Isf5v;2xT@a(OmnVDAPIDtl^h!(s{
z+3m50PShv@<IWzjM}KC9n&NC_!6{hWXQ{O5k(h7XHubMI55yYV{!B&9$piga8Rj$Q
z3OOi@UO6Q<Y+3HTRmw~UJVR2KeYFs>DG!d19e-Erf(5fqXm4H`YN@Gss7gKJOZgnJ
zsVh<hubTs_pwbaf^{Y54(Q_fBmZSNC-5~oY+7mK_v*Gv$U#xl+d2O{=TOy8k+@}38
zuRcqIb;EP-XZt-6o+;M1W`Ih2C;11xl=_h{KA(j#b(md9jUoPd^JKi*)lpNq<KuX<
z0xoFR83j#za0J(W6OJ~X=2a;(UNgux6ZnC<zex=xU#;7i{#F^T@2G+oUz%7U5n7*K
z&`^jFEnA{X7U;GOI|8-oa@6}jJLp;?;)o%BmG%@9zlM1zDBwyRFaS*?gl8O$2Ut_=
zGi!){+3bfKSjH&78P}qZV)tNkIy>y;?Hp<ExeN;M5L)yg(_>uANY6x*#hfY#I61dP
z8}pTzcSQ5BmATxY<S!ieT)T|1l~9`SD4NMvbmOR~Q-#>08u}?=d$+K%FRnAI%OjC$
zUCbf`tJOQY7+j$tKXa9vIqpwr|B2dvXVI)`{c07A^1Jq|n(=2**jQMHhQ>-8A>^q{
zTzf4=$_Qj|c4;7OLzh@?%|=R%>Xu*yUL;4jUq{!wPn~JAU@kSx2xDxC9A)xzuQ9Wd
z2KjVsb8Q$V5$zhH=wpQ8RhCv=`Jf9(psWKt`LhxMJ6`VD82oO#L2-rXnfH3ym?gM)
z@muvd6IEza`A391Hlq2XEGW>R!oPt3LQIgsL%8jqqEB3`iR;6FJ(F`Gc9>N_i0)Zj
z3f^)_JooSsv-0>XBI&*jwJ9~x?0ygbXC7EfyoT98nsROwTc^^6qp$d0!(0yyyPs&0
z{Q>@&FOMT5sMgLJ3D~KS?NYx|PwIH?rqxz$M|I2sE4O49*c(qdZujv(C;ut7nIbgD
zGb$L0_Oqg$-{i1OB~2Pr+-V|yTXQU;Y-cc1zU1TR^4JKnj8l}D-tq#Nh=Q}>Kg!X>
zwD;1|A<`XgOFYQ0<gzW4ck79+>Q^G5{28!CTrrpj`RpcTbt4Qexw;@}`etmt<)9;I
zIgE@RbnBiV*s^sEcZifFI3&AgmsKpsM^xe2?;@M5(%m_QB4$MsvO1eU5b6C%RYlkO
zYMMs#?BO$EVEYp>;W2&W!Br9#oxqcIZu}usQqS@wmwNiXMPb&`x=v)}#OWlSC1PQ2
z`F2JURsylG0lL3w^Uttys@%_sWCGjgNW_TciiO1`Bx;hLLU143ii=ZTX7}hMzxSCm
z(s%15^(({IiyhGIaw}CJZD#Qwl?*gl5h8IEpV?l1q}9K?tYm`WY${HZoIWw2U`rR_
zj0ADY<W2ed$+0gHPE59|5E#>8XA7ay9f&j&22x>0=2Sg3jPR^h2y^*afDYbNwdclp
zY=vSbwz$MS#+iJYjWr9e!0GXpDTmZV#$W6GG4^rej^QrrH~<A;Mp4j=x~D*P#Z!5O
z)I6yj+nD_<9;NK|ef(y2V7%`2&m*QAowOW+P#+v~E~G!WarU@36ZF>rCwn{A4Ys4U
zH|oR}42%5v5E+UJKJE8?DteZDOQz-bn{t*xYeGo8JPAr=Bxl5S$7m!un;CF5e%qZR
zcAb0#s#GY#mE0Aeh6tSd_u^RqsV$1>&W=6R{U(fl5MVofT02Wqyi7Z%!JGeei?M>P
zFGj~NTJktNgY6BbU=ivANNOpRV5*R52?a$R+7u?;>g;_Q$htuy*@i;N2+RzemHo#{
zW3coE3gTgm?B1m|4-HM8tWm=O{%qJO#6TJAPGlc^8?cT(H0=)=7_2a*|NCcGeI_6|
z*eM@U;?QPSTHxyCIBYVg`L{hasHBqA0Ka2*Dgyx_p`_CiKBSPjjhEpdOij;utNtxg
zS|<)BtKBkHtrPdAg%xGMO#o->6Dz^4Gke`MqpnSG&=3>qLylr@7Cu0g6y3UU&Tjlt
zW3P1H<O$fI(l)6?-kP)Jz_j~2`ja8Oy9)IzgMTH+2M*@lzUFd#{K180{G?YG{-6L&
zFZKY8xG@a}muKWzrgG!reaX}Pjg90_(bU|+Rx9aT_pZ*0?D%pk#L<Iks>b5Q{IR0E
zenmV!hAWX%G%wC+EWc?r`hj=VnkR!LqDy`ujYBHe!1pTzsd<UnRl^OLditwd#ncUP
zWAd~7cxGO)3gtyRA^#9XRZp^biIBSOOdIZ4FSYgZXpKul?`-vw`{Uxq!lgA&){Q$X
z18mxH=3m=cTQaNjdu!Lyv|>mAVG(yOtFf^|jE)_vEqd}|W~OocZ!Dd&-eFF6!yB(-
zeS1HcNV70u&*vZHMZzYTx@QLnZSj$^pr-)a`5d`EHX~;{H#ypjV+Jx)U+EMiiaG3I
z0WcgV{Yr-vul<&l1xxRis6{WL)l}j^2>`GS`@8w$<3!)@>w>YH=i$jQ96k(tL?3fQ
z;@q=7y`7K2Yb|zI{Q2EOPb{|NDDqvd`LpFXCilS+1vu(5R~_m0AdoZ4*1tD<-7m<{
zhT6@~u-iC_gwwlL&OOu?R@f|HFImzA9gRGEHeVI_SvumJ!@%g-HPYK6seHEv`wZ+b
z7G!kp`Q`C8_i;1$++neNn}%m5a}_&^aL7kM%i*R$D5arJ&v#~w1dGn6CwPAHkzvu2
zaXx8%?@Aq;nx`{4O#AFM3g7ef2>iTI=>B@i{gUGQJi95sdoD@Y3LU?=o8A&NTo)hP
zTZ{AUyzx}$1M&me<))VMxAQ;o;DAfIXZ2LE#7%a-CKt?yZ-=dZQJ$nFwbxjY1m)Gn
zqsAAmbl&H_urQ?p#LvbRUgy5P6UIBP72b4R(Vm_s4;;YL<6EJP(33q6pRW(U1cLrQ
zPM*b1*KA|7$p|%ScXh*QWjcCg=W45G+P`9%Ra(20S2p8}J)sB#ZQ;AnDuA)Q>LT7h
zPkb$0P`LHe=brC0<jM#G4dJ^`FHt#<Xb(c>xhK_U5q6wA2w9Ha6CfUWxWX{?!ZD$5
z`e<)<iS#^*<C?iUXDJbOgP(eZJFcZW*#6sYekfJ_l<q!zrYn$llDsXD`{H^ckn>`?
zv$MqZ$n<Baw>Ffy8!?6-7=@I0_m=+dw-u{U8!~)3{Q5ds`&QeD7_*@;e73#(!MTF(
zhc7XELs87e)x_1AVE<eeb1&WHQ?|FXawxpUd6VBOy!I-vCY=9j+jO$@eD=_o`N?4>
zc?yQ`wj2g^e`jkme|DexRQv-!=#_&b0i)0P#gj_e|K*bf73(`KJ?9qLVIE0PV$4HJ
z2Y;z4emnk?Ma0G;JtmhF`tC3@(Oz^xbZcv#81>f<Bk2mM0%F-xM=Gl=t|PhSho8c2
z#~CM8FW?s7i9JeCCMZz4x6E(<+RI#NY;F7jB|!5Q1Y3_M{5E}(e`ibOLDhlaIY{4y
z;5o`zG!Y`?h&}QKjHBEd`={tinDNTO!ZY7N-{QSLY9z-7ITb?M5gpf#(nT$g^ipMS
zH7CmPr)!iGNgOHLUUXEndTxXu!d`Sx^u%<eo@>+YmMo^(n=h2KJIVu2(i8gzK86$L
zS8PXoX}X8~)d_)TSBd^w((M}H8sLi^PGHGMI!cPk`#vQI{Oa%iIB0g9eW_VOlHYbM
z5CC@Nc3KmyuJ(Sl@(XCkRS+Pa(U_m6mHOzqD*EI!oLSwVsdc}cS#<-Pnb*7DPYC_q
z9;5@`Is$HW2m;P#cUa<T^l|;WxL-V7eLOvl{ug0y9Ti9O1OQ^e-QC@Fad(G6AUGko
z26u))2=4Cg7918$aMy+4B)A6;ew*KS_wMe!d++|Kt*Wk>uBxu8>7L!~j`8)RS5#lj
z(XWqq+BR9#98U+;n>KtIhRlN9tR<&_>SD9$`;2n}?R6sm7%pM)J@Y`qyJgHa501yq
zp}T{$z6||-TcEMmVc<YN_NL3Pbd3BFQeA7iTpwqe|IE`Q(!<v^|KwP_a(dCK4Nr8t
zX{cJBYg$Zt78SQ?+;ed@?Bb1?zl0Rzd*V+#!XUZfZN)8CX9^h<rKqb3b}t7E+Ofwz
zI(b@Riv_I?Ex}Ook#R(iPgyCj`sROj`qq9#zYm#}Jo(x2T~A+(iph%6{4__<K4=>0
zJQlIqX5-#2VDmoU)6T)7U+b{DB2x19g>k*N^3UU{=$@nP7363!%f69(qwbA;h_=hi
zR=G#mmDft#?@P#zuj!4SK4cB(I=cCabZ1t0=w3c~(2kpbMX~E<2ar`J6D@Y$KZS;#
zZ+7!`OvVyZ#W;w{>U+-_Z+;4lh(0EhiRWlimTvdx6%l#O6~(z^Ho%OpE{@pw{?Nal
zM_F@@_{lY};|d&Q!+kT{^6|;dtbe3pYFHy|mR?K}WK&|bhqM~wk(~oj_EL0*T}{Om
z%ZW->7!BPmP^uKg5X%C_6sJa%v?YGsk6iUnpTnss#3}S#&V2=9j6-u#i*o~gBT}me
z1N@&Bj{~ckk7Yl_hZ-u%uef76(YKsEin7W(2%9dGQ@;G&Wy?<%e_q=Jt#B;_B07~o
zZUbAlbvCMNC%Qy>3@(AKHEm-EWzMQX>~JXb8$4px_|3R72o|QDlO>UOp7~0R>EqHn
z@Z~NNa6OxT<@)dttc_*MBhduTk-0j8^jb?owv*AHztcFAD^*<vc`hW2*-D@eJ;;6)
z!linY{n{QVBjRP%M@m_PutTE&HY3~Q_BEx*b&*-FM>LC#-vkuQ*0I14O#R~^fybks
z-3fsmlYF=iW|=^%^V{*^>5Yt4v#$3Xt#uB?vhzK(9`YM)_*x30qo;;9ErA>oQk_{s
zla`+TT>M66!5y5wbhi{(j54cbxN2WaOp&;p_F=Rj>u=7$Lvxgu*m4A8J&^B-h2LK5
z?pI#o&XDJZW&y{5^*ajcn`ZQ2?Eq$rPQm~u!1%f)mVH*0?eVIsnSx24tJc}_cAs%v
z<Lw<Pi)d23=?dpYQhKX4%!CIZaSir*SYgW+$+meuiB*@@TM%&6{DO@25lNTes7KM9
zbqu}D0s8|(rs+~uXX4oswaj{C{#uKNjSZ&5df0mq4sfK<>qy)DoR!|X{d7`#>fyE7
zochbn<fEOFjS054Z6^g9C2F<YAE*T@L6_fU3Qymn6&iVaAYh|9E#34{`x!Vdy*FH~
zL_$EVE}!$9R9!9nbTQ|Jp#@c4mXOOyQDRVawNx6)N-7dQdTBZ;y?Gfb<No#BBnWy#
zHroz3_|wWAh-T|Scfi@C)0&-^c!PJ5+Hq>=`f%+vRV}H|xSCo}*qOfXi9uWQcFwgn
z6M0oPi?7+cauwRcBZ%wNwf>p6QyH@&ayX76tI5IZdb@szTr>-2Y9M^fBmeZ{zCgjm
ze2x67qz1as-V5P|=cIG_v(()$j}@hv?q*<}Z??PnS?(@TsM*Z@^9ZIGSaPV_U4g0+
z@zK9CT2*3SY-^A$ebK7D+djZw&?^xK6VSFRZt!=!5S~2En%@=#p@sWrJ6CW9+gj$G
zFHw(mG?;$EYiK$>7tjXX$^0olothMJ74SvRW^9`;by5-i8B2A$y8=*wNFT3)umJxJ
zGlCNHPn~;bW}*fs#aLGPDi<x!zU&+)^jm)iweC86?8pjQ_A8wHI1sp;l5Jn44A9B)
zykUjxx>j2b93Ib;_zZ9N8=BWMvHKJ}<~HgYnu+iH4OVWnlaP|XT6u=KV64=SYt%14
zN!%v4Ja;>{H2G79^nU1lq*C~xxCX@4>D(QXflc@@yyg_=3(n8EF&<vTY!kfV5ahHI
zz|cUj6TS7iZONU7(dIcdK8jPbo8SJvZaA+&SfH&Izc1XFe!C=6*7j?1>uu}$@qWC~
zknZiKubtKE<xZ(n#RlbgD+wlzzulv=C706`$`@#ir67@tcX8CE#&!;p0>KKdg1??1
z4ei;QQ}MCFJA#~HDChaz4R*JRf`T>Y)_5MK&s~TSE8^>gG3298G=qr?7)#-^HQqL^
z`UrbqGd>&;I@QzQ=-OT7-+4z-*yz=OyvOsk-{$)QQo8Kf@r$N=t-h5Zy)$jupbr6H
zSDmhx-z=pus{(Cx%f9fo0RA}QLP#&AY5a3{Ttwb#O7+LW$2@&S>HeVl?R>I@#_E9Z
zszPD*t_~X+cQKzOl$WA;XCVaHV(~a!^w^_4cdr}wn$Ax={Pv1IALqMj0)lRb$`G|X
z;aQ30S^}pridHn(Ln{K~(V<B11xh7G=bO+9!n<MVFRca!FA4^LW9JA@gnhR1u4y~C
zb@RCNRK*h?-jUInJ7VS8>~_cz{cLN<*+Xbl>}*JmU&x;9&rNybJ@J7T$HU5tDB<dK
zqshkx@sm43Glx|H!qx4U&OK;Mym#D$X{6(Kf+wBbs&G*|;`%<&6Zv5KX+cn%3M0--
z2fD?t$0<fy)|+!6znj~RKjz3Ozc6y<t}O#O@O9mmvYG4WWCjS)TgqOLkna5sr!BZn
z_+4PAc(L{E&%6FzjOvv`u)@yM&zNPOchyewXlI7&ciy_*`>XX0&F*OQa+sxcPSboM
zD>DU>kSx7-51yyjElz$347u5a?w$A#jE+l5^{4e#b~YQp=CK016;Pkvnmo8}F$I2|
zQeB;@lMU^BN0FSqc<W%gV(s0*S5<e~OLkAOkCvVFzu>6srK9j_RhfGEIbAQ=-fe`>
zBfR~Q$-}J<8qoXf@#;z<xZqB7WBk<HK8uN|&gJQPwg{PJDjQnID4Xj$jBN3}m;C?|
z&(6J4;I!uW_VR;ds`Fus5hadC%I~r$@~P8oJb{q7Ip-L;8SRx=96?DXR~wrd?Wj&U
zcFwp}?2JyIvi2sH1Er;Znb+P_X=F8!8vreIqx)y@*_x~Q=4Cn$UeH>Ys&+H?kUhMh
z=JN`z&L*A|f~qsMcW+nT;R~^r21-7;76Z+{Dv;w;I)!n+Yum+lD*(0hr|cVR7cNHR
zLae8YN{$-y%cXW?9jasvSxsqPo(o;gf$?j}!ly+gzIIOk{5GDnS9(e|*VSw?@NisB
zCgRpxlH$!88uxai5?w^aw+VjTfM)kwZ%FAWNWsIUP~74K5w2Nd1zIO^?(Q3yn%4^=
z3W}S<Ka!`eu=|zET*rEiQq51b^zhi3I6w18oxVsC4wm@*+cpg!o>z!JU42d=l1hjl
zWXQs$?qVtT0eUbBvz!Po^c6b~4>yRHhY#A8elw?Fl|I|bPeLrn%cWrF=I%i)#Q%S9
zP{Yf|n=ajr%8>qTUI78%=x|#zHfsn^Lh6j3o<R75i0)v$%=<)aVms4HN`$eN4heko
z_k0q$PRGXP4bjP{#Ia?AlC@qEIwO^~AII+Y1M~6bEzjhAHIM?joxaen2i>~|=^91;
znU^q(9?O53<zjmxp;{L)t6myDfc$L=m~fly{XQLqD^hye|6mDVwg5hd6NPbDcVjCH
z_c-ESn8)Yu#QyqQ0r#ps^j4<uy{mq9&mXc{FYWbF@R#*)CRqt=EBv`8dTI@0{2Dl&
zcDcDWFH_xy%wL?I1zbcYS(j?R{cuXf?XR@$O$+q<K#_*&`L3v7%~UcC=OL}_X0=V+
z|Mk5Uacz5E-D@v!X)r`lJ@@29<;gkucXjF9)`;>L{{4RE$JFDGOYSvpA6X*Z6rbOB
z{7q|%*Q<;6i|q+|tlKB1s-0@xbq^eU8FLSO_cG}6YP-F1L*5YdIOT3iRsA43Z6J{q
z%d2(vkTSVC<zDAjd?L2bDrj*mfn;p8FT6k8xWcCS7)(<MY#W~~94!WN0!K@W>vcD)
zY*z*wi^EltV=WJ}Hk-S3Ii25WE;{Qse@zVQbvRGC*i^3(=k8_&##DC&JwAxH4ja~N
zQU-YY4l>W+_l`ewyUBO3wUu_v3U>YSj_JLB?0x+u`FcH^4?M4LzU-E7zCLy6K6JjD
zPrZKlTTitKSf(~+T4FRZD=)H)9}Xo=vh549G|}@x4?EjWfEX)CExskG9%J~eQiL|D
zkS5(C+=O1Gt}2gT0TO>#FD1|^Hd>Gc$)VD1g4p+t6=c<>{dtE7aopoA$TDCpe%q2#
zBO@?wRtyDc^;(nly4@?x<A()V={e3?Y1MF$u3f-G!Jv*x_bP8Q%w9sLqnh<*MdQ!3
z<kL16@U+{u^ToRgc)3u#m_sCMXGjM9>}yg4{iJRz7k~n143Jd%%{vhDxK$rwgUI)K
z-k;P8K}%3G%tJFq`$ZiSa(iT3zFAuP_+V+udsGqLua~a0H-eczdXgx|N>EpiA_U3+
zX=HkWBYsoUG-SfP!H#85h`QC4G%;wS>{f!PmL0t$sg1g+(vU_ld{ydN8+wJF3NIuh
zW%o$BC8^t2swJuSEZE+v`q4wj=hvMjlVmQpj4DA6x6G*c)E69;Ggro#Z>Ac&#IRcX
z)#wT^YWJ+6IKenQW9DD7dUwCz*uSI!cnTHwMjFJh+cQL&(k*#W0u^5icdOB>AWH8*
z6^L7Z-Yqh#Lij2c-LANypxi;LcGtyDt5*C739>SB6Ke0os3WX9j<Hg>B7>=n2+3lC
z<W%`KR_L^>35)6)g?>zuZV?EfC($)pcZH%5D~3!RFOT_qZ?m_#T?-#Vfjp#&0u1OO
z6hF-clITM*(g1oA$C;*DIMBgze<*0g3+1P|Wha8;w-xiG?}&7=p`XUwg0!4U;sR5(
z8XFJNj|7ZxO#)_<(Z}j=lhDWF7IKij%@XHbg*MWCPI1*<9<h>IHgFSR7EQ)OH!ps}
z?4KmPoS7vAUMe*tghtaGP~E5S(HakiEH=~d83sf;uk$WAb|~gwL?K8Zew$6YNsj;V
zrk6=n=ny8#&`aYhVw7QH4MmLkC?9v8t{Qqz6epu<6pTE^IOt83VTSOyIRe=zx0E&=
zJXqTfK~n)HiZo~!EG3}%yav6VGX;V?meh+aze~%_j&WY;2LmYf>nsVv#1U3a^H%Jo
zU4`1`^$PdR{!yw$-m>-efkr&@y-5Y3Ks79cwp!s+idG+t5NRSASKn;J!aI!MVgZ~{
z-U`ArX!ItU5RF)wuFg_At1vy~o$QnyIN1z|!++fEn)}PoJu-pbhA6V`dvneCK{rGk
zejTEc<{0KB0o1RR{kpx%GFYT;Vn0ssO`Ant<_^~8>#v0+ZrOVtoe9=<U`)5ts->6X
zMUHDkgC(C#G9v4nWj7EAE;eYihf!4hr!<2g)<04iHqLso3YNTf)CQ_KdMjezY)wZg
z9C<4LBq@v{-bGqHZ-w4fFx)5p;|VARJ#z<unD!kVmTiUy(2I7km@haobdW;fvL|e@
z_4p<Rw|#4(13aB2@f*PkhR+aKVgVzWM|db#3>8jcGS3N%Sit`b1XFB1t?(u$OZ*Ng
zSi5QZiBtDD9xT<87|wyN&exCSb9EVW9r!tJZv+A_VhHb4ML9DY5R6<<EbLYfpn32s
zzU)aUy6@<rc<{S(TftCFX7<w0%;t!vqd}}MTTnbw(8E9msz;TKP-Z1pE>yPc{UCx!
zmR_6&DiPGxTP1ix_J&7-OrI8-^Bmxtb{f3l6xCI!J>V24$%QlfwCCrIEukDu9cvF6
z0dF;^wBG~xAN(-_KtI4%B)V4%DC%1k(4DrWTa?^GP{;-7ZtC&i=R?f@jXqs38sx`Y
zk%HnR87sU4TLscfrkUB81r4T|#9f#n1d+ySMjcp+<x~q6yn~TC1ZBpVM5}g@&vERs
zIVc{M`j!S-X-8`nhJ+lDni58d&v9ud_|ZjpL7HA}I=~h)o5220X4N6MxW>6gbOoP9
zYz1cMkm9Hu)i8caNnP6tE@oFm=0;ph`jLkxG-fZ;`wya;XlyZ8i;ou=Vv?#svE|y}
z;u>qJAAObr`XmP<m+Sx>g@NkFBdBs>C+5s%Ri(7SaWxFK+@IrWN;4;*>@*YPgAqUA
z3?0hfuSfH+6z}BQin&>f35M4*ONB<95wtEJvv{D$y>-v9En;eQ&1Jmz_;Eg};T**|
zq40wf&uLSEbc2aBHa5?S`^Shg>h|hLRKa4JJ`7avNd$p^a7__Lad~1gREQN2++xn|
z2!9jTc&~&6v2JDgl+tZo&<3Da8JvCK;Wx-b)+i_z$SZMydMr&*wScY(A2v%$(Sb5U
zP17UAU!Z{Q>M1NZZsO_ys%qZeJ#-u}oS?hQ?)i4Xi!?F|Q|Lw`5J4R;KW-(}qNEf6
zHJ^S#JTlI!9{?YA(LG)f^W+9=6Lt_G#Te@B>2jK<>~miu!~mCU@%O>nc}!Y(++*%B
zK5z=YcEY*RX^PB!%v{UIu*G8D1Bk?_B1CjCDnTEDwM|viqz!6jgHoaJATU*Wd99nY
zFR=~Zm^+wm8de8h=<+`X^m*3kwDTanoF>tZLtd)q0;9OW&J=1tDLz_YINl1C%gd^S
zCFcNYQs<#%Ux6LUA>ZH&raNB_fKjYQfg~pnQ7kDohNeT)(NEM4jySbE0g-rDTtEsw
z=92t|zmJ(pKkYj*oo6;8af_w0a3AxuwkEVGF}Mj<;a&-O5vDkx<YFEMYQXUUDBZsD
zJyc>I4?f1QF@<8N6v9zj1w-y@3LFbS71tqzcE9|uu*b27+VgeN_a$mhL;aZfKzRMq
zQns-P6Sr2E10AXtXc<vW%YoS%xk?#?iA!TohlSMaL!b7gXAr2&-l+QFm(AqdHGJ-|
ztG|{|zv^f5tU+QBoo0}iJ&|G)%JQK$_@*wy6sNK8FR`d@0?6EB;%#<)%y%cy+L5Q9
zhXEai1gP3;O|Y=V_HYDH>l6lez{Kfmcp=}EA&qG2RcVVnj&EB=?xkBs#CiH=Rnn{!
z40zEBzGso{L931!IefrESAh+&rC7Z@nV(I34XFs*w7ak@Tm11@Bc{Qm75a#qLue*Z
zh0vhdA@HBNSvA7WoL1cgzW30si^`M<k|1IvJbe`v%K|XPw3=Gz+^mKnu!zJ2!W&k>
z+Ov&?u*F5Ytndo*<LZderZYF_Rgz<2_Tup)!P=0<1B*sW*n;Ab3r={ED6vuQf?^+Y
z+=5~T)jcRZY>wc8AO9Wsz;@Kl1SnXk=rFDT>9Cnqs+jKd3mTb$l90!yt~)3WnBDZx
zr4$o8J&{DB*04lpw138BOXe<|^+_aA<ScwH{=vSF-1Of59+R9PPXQk(4pr0Ib*thJ
zg6>v{z;qmi5UW_&0dyfF=KB?1F|1rb5h*UJOKH1L8;4;Z2~xD@p~W#*0PI3(lq``?
zx;vD^JlwTY45jcQW8;sQ9{9eSY48*uUd?z3wq|kDavHTWvSQ$=IX;9SVSZ`;!bc>*
zqGrDEIZ=-*O)3YTp5w!)FJhS8gxsdKQB=2(`J77n>CXyFW$`FMR)aaW96{FmEN8KR
zG|GGN^-&wLNo_Ix$i0zIz&hDvB2S)RR<T4%iSQ-DppoMTKZ=>E(y;gE{EBDwxsP{s
zUAqGvSEpnC8eQ|`zTuJHtOo9n=emu%5@kB0<o1`&Mj6#|3`UcsI-_ZOWW-2mNtH*T
zDy@!%+O6DySJDhuarjiajgwD)n?H058&cKQyW(ptodZ9z9d=yT0+u^-ZAl&TFcJ24
z&L_z#xBO~<OOkxohiuk===_YLe1K{j&NMlDRi44G2<X~5ey9_foW+~FRd(Mo&(+K-
zp3X_yJFv?O<q6EBTb^qbb}Q!1qQfe6<{ut`|DrvbXkJg2O1``qqq2v=<gCkkKI2)O
zL<g7_C-L3YD=^pqXTM--E?-!@S?`Gxb5*7kRai6cR=_%=b{)j5ncE&`x^r&f=el#3
z%tyMsnssZHB{w-tsm;xQ5DfyGM+=hDFvxW;1?D6JeKyz-l)Nv06$rb*%!Ka!5i>=Y
z{I$Fqx;KJHhLzPnUq^_QMK`S}W*g^|qTwzVVuigMra@c`>}B{H^J|r~5)omfcv}z1
zxAb2g>xzo?Vz#JK-okGivG^X_!?+P#XDZv`OB@~IOQ>L>>ZpDnHc_$0Ns$v)e$UKq
z@SdgS{zl=$k!O+CsMvI*3`4XPO)9%x8AX`NUP~@B)M|+c83qRs66{Z&1gX0&9f~JG
z)>>`66HFPP<*g1q4KhT9GA2~3wwbM0!(pe8?;nl*kA9*-fv8;NBrV#qtdM=NXQ)3y
zC^US*{bFBg^N3sZ>q!mw?Pwn2Uus>pz`dnP%lIde)QP4_Y8W<O7DnX0G2)IqUAOld
z$_zGoNigi5*wQe1mzh!jnhzPE%@Z7108Gt94L$5nw_{68xNk}DpP(WqV=}Cn`}R<b
zy^|aEy)VbLWKM>}zetEifi!mFRuPQ1(Cf0L?R}80&~tO4_+me3L^wyD*C&r%mD;t;
z3DsU=2pg)c3kI@G3oeWGKYT8j_=~*(l}t;LAQD31N)|#1In=Z`fZz@6&+Pzp>|hAn
zuVZSMYJD$GxNpOZf1&8Tr8Jc7uqPJ!V@(n_*q6{m?G9UJG>hml6sBfR+yScK)$ahP
zNY@QJa-Y2^EGpbLLtJ~P9ft0rOfc2;R=4F?n|LD7nH_hcVQmM1@YdhMMb5ruPa;IA
zc5#HoE>;ZS!=i1;Y72RjxB}gv)zfgB#|O8vwk%CTIML84e$FoTg*jJc&zyKmWs;|f
z`fas1S*TL&?HUOV0O?v{i|9QX$q4;hy~mJ7{Zi%2=YtJEE}$30DIWHO32U+nzh$}_
z9||&K==~&gLtn%`c?jGK9^KujD;c^y0sa2Iw!pLC9;B!K=hLxEK7yOsOMLsBr<>mD
zIF{PwHH5)<@fnUhJ=h`nIXW0k-z@s~+|(1Tx$)o9r+UfP`puyGtNa%VR-t@apO8Q_
z^FSQFs=rQzm+z)3vaADndJZ;eDb1+3d-Pf?6-b|rBA$-n<yM^E^hyv1XkCBb39Z5+
z)&ZVw4|^oc&0Wc>#t(ik2ohW57~Zl?{k(zY;BASZaOJN(#T0#@VKW@SGT?#nt*rFI
zzZB6;FC1~3iN6@}8i|q~Ei-Uk_o^Zl6*~+kLZw1Z+0b}Z$}}2UvPtczK$IG<_7_U&
z49!koIVbq#h{WKP(L@o|@GM6*iRoi18wdm{il2X!Hp}}I_IC?S^u;$bD4qhL+jtz1
zm*J|tVk0+fwysy&{|<P!H&Gg%vF$BG=IZ<Q3p2Zl@?A%q6r(UTCK#-BoV>W3H76Yu
z<rJaldyiXL=lX*&nzkM~s&H?w90p9yTmHu_hi8GMMVD@#M~v@15|QGHi!W(xz;s!u
zqxlPJBH&^fe|zP#?PRnX(K8`??|(}fPI{v3Fnca+%BO#`7JIo`Na0)oJS>u(-)U7l
z%>YQ2|L$V2u-3jtVYn}j)Q$xP0vnlt@YYTn_^TQDQ<m>y9v3-8H6$CqavSo=H+&~i
z-%WRve_fKE7MB$B*mu{&#R4{ik9P|t%>8kF4(Obw5wK1Zk|Qj__4+d9gj+zXt-?Nq
zPOEA+%&V<-M-;72^H`ox-nE1O!wi;L%J>0}ixHWjmd0yq!)b@2q|k6)@HITl&!z-d
z7+!x?Hzf8R2^!sA?~tHCnJ#A}k8d`Ni<rCE9sZEArhR6nu2v>i;O+*_ERVY`fEe{Q
zg@Lp7Dz9y=e7{Gc(CJK^)+~Pp<Vw79%g=~e;W2IXd&ctwMV;K(Y1;nxmc73{9+J)G
zRknZA!5!1gJbpfbLRL#X-Pd7=kbHj1bID;cIN$bHj6|EN`p^ebO__Xj31Nuu8B;iz
zB>%b}Z$Ei<vsHi}C>SO?CnXcd=%`#LUeGI%ldTur`xWan?dxN+s?%V|vYIFI8(*Q*
zs4|@Cqoiw~j6dP&JG9Wv+KOTVg+3!;A=3V>THAU<Vzhcg)a|ueW<w5iuePFXn-2q>
z1o}?Ix6Zrf0<`yfp6u*$m&cJo6i^)MIi((mPOV+`Q8?gco2_?)Lbv_yRE@rOV<rn{
zw9U&_NA9xLvjrpZ$*-uDCduWr03$KLAu0tG3SF7QTMu`g96qA?bzQW!;5@el*+S%A
zk0T%OL;pl^pw3FxHzYAN+erHG&%XK<Z3Z6z;<GMN*PGy}k7%>Nad@lh_2!Ug1=*?a
zn6%~!V4cG*wTdpE5r%VLW)^$Ui}40EOsQv`ZT+9~uedd^N85Z4mLqpo#jeBTE+;W-
zW0iWASp-*x_Mle}o&3Rcaz{?XN4&(^kfz`Kt;I)sPx$=jOKW7|uBP*Y7<yr{s3aC0
zv+I-nmvDXC%um4T;!F^wOP7<wz{|79bs_-$;1=ZgNBpJMN5|YoZ2)>Py%=xLmkQ5m
z|G_}F=`RY3sJ{vjw^<7lgd7?0-zLQIRNShj^dakZf|TPV%hcZ#Ppj15V~xpk#E7-9
z@%OjubYc>uqT!0zRmJ>rxH1>$e2)P?rNX%^m?gbU9&61#iA1;ZL1fMr5jg!wI$3}N
zi_Uv9!(<V0-xKpk{w&2og1dhv2sJQq@}l*kK)Ph-JPT7HO$6WPgRy~ICALDU2>#vG
z*Ve*yJ{JzMmB~M2^@UVPZRQwGN?kJc2T!qgTo})Px{_^~5y(8;3Fgs7j{d0R6bACR
zM3Rtt?-H=dY9pO9WgcO5H3A>)Und?pG4Tvc33)tah92?CG{8RheJ7VBOq0|{mw#Ut
z>hhUZ%0oU<BIPrhh1pb=XT=Ujm*4V=Ws3@EjfK{RcL-1WWoe-LXkBuzU8ekUp!K!q
z5}z*GY5#SnsL-z*=3_q1FEKUmyjbdc0rQjQ=*PTHnNqr9yxLN@4=FH=zkR_R&GKVg
z7<*njT#-&Kn`oLK1H@HMGO*sxx5@u)YVxRTCQ`IpohfmR!U>hEUHzxpRXGVi|HNbk
zlH`7!+WZ!uAa9n&bo<@3hZ`dDhZVF>OgWXJ?pGOh=Y7w5b=_j#lpUChYc<;|?ClP_
z-`<YKQr1b+oW%;Uo4Wz&ea8}&I_{&Ps$E#ZpBOeuJ6DjeDhZhu2x)#tXJ(9iP(Noa
z7qCLEH1@u%9GuJg?U;Iu0fsXm5%fMBVEs76xuHeE1$i?*)|ZcnKqCH$1eZ0`6dYwh
z(lwiUKGr9X+*p;<94;ro#*|NbT%F@%Csyw^S2)|H^0%g;OQjDW)7j+@y~aRKZI1~~
z4HmSttwiX$6HelC%c9ZM;8Vgcfcw*`s)xf0b55i39@Z!Qi32|xpI7QqFt!wqwyoM8
z1+$|@TUTnQ%pH@~H$C{>8#(9nm|q0g9VRQJu!-r$JzUgoY;Yggs4tVoy%FS`OKm&2
za}aM?p|pcvMj@bN-U7W3?Yz*}Nn=KG0@-I3&OvCGv~(ZeY?-*JYx0d~>55Kn8B%0P
ze@$4iL%f;R;2W7Jf{OT6nsUpg@_gZu(RYB7Rf`B)PM)@)R5xw3V50Qe=q?@3y$>}z
zff2E>)J(vVF-)QnBn}3<<DI~D!LNmx;1TsDGh#qSJVqpVq_#2}o>tHVwhoHgwV_6=
z87^Jtx^WP_6uW+gQdv@eP$L2&06nRl#TH?xzI6|9f4~!)tS~OeM@%qX>gjcCFfQ{b
z9&fgsOptV7HWpj%V7Hu#?fo)798@6ot$JK^LRBm*CWUb!<NaXq$vFK6dM%j<G5}?N
z$Wkvgu7dAdwVu7sy}YmMD8Si)W*i!|TlPEEj1-4h^Y+R=u-F~y0{ql&^O-D5m;n{U
zU*}OApuPrc7{MlHPpLV~e9~D1*|t~S&!Cqb+95hANb@mQz(WO=3oP2rE^S_71$V_A
zrmb+(dCFlgK*<Ga@V^<BhXEbT?Ss58hy$x?LtrR8(V`$8R=Wy!_1KsT_Q^(Lw3q6d
z#Jc(7YWyHphr+_`0wt}I)}T!eH=lia!-h3u->5^~JO7@P+fC<SGdM9@<_R&TYsbRY
ze5Z`<F9vbP`hI24$cJ`p<e$NbMkCBq417fUE`1~Fg#f&QGce0fM$8G`n`cfx!lMh9
zqafvR8_{T2h#}`?=BG6sJc*6a_tpW?ZX^`u9*2s2c98433+p!9;g_`qnJc$9pC%#5
zv=M`FrI~S^=|i>X*OZX8Ek!n!P_?VZq|vpr_Tb$jicUi!^s(*el;vjYRf{6#Hf#I7
z{paCB1|SNwrO}U3%Zx#kmG^Z%2$8r$|I{WpakTzArT-JEGa0vAy}g!&rF0V1lBGH~
z;<~Cwc)Gi)XIyT#C5Ai1n6+xJw!RU|bKfyL#7x6xX+9;bMvVXEPQ@$Ug}np9@6*${
z9qv%t4XAfL1(o4{pQ<wkU_`VXV~#>OKz5Kk^~Q!JOAYJ=XU85qAM>2ER9APUH|~Y~
zrVD?6#d(&hroUR^lZiVIsfN@?K2nYB#Mw_rObY##`NSCeS$TYnr|H3o{!3WsF>9Vx
zGr@_S&I(&!kz%6-%y&|a4OyUiR?5IDoS787heC0F@yrr@C-e4m5e8eSu0X)H3Gw$S
zIAtN&C*AlMi|+sz&)d6Q4-cvzu1>Ui5rY}e2&lAJBf(XL=MA<?zWjW$ZnfPD{o>pn
zF8$J+H4gpHBSU~!8|7!8`L&3Fr#(m7ZA4FQ>!<DLA*Dz8F;3U-lS*YHUQJ0IIT*_m
zmK(Y~!g*CAqgV8Sqt-0*>i~XsrCK7v^4QV4=DeN2q{BIDj%1gf%yAioFO#Na`|_Yr
zgD+#s?hHzHe3@Ygg(KFWI@!PxGzL7P#;jP?kPb@_S$bA-7dh$58oJG#cH*uhgTC;6
z(E{1Xd|_~0j7{-@KE0V@V(|w{2_rY{=q?KRED@5KG~;oG^o7-KZot<Qw{o+EFeykE
z#m}u@e?&=LqpI+4<cV$Zm~kGuU9pP3Oj@(H)kVIVUGXR*eq(8D+1U(8FnMfrx}-L7
zhZ<}acVZ`dC+q5Mq!f|DZqL?{@!68bsUoLKJ>(TsY&bup`a1#i4qCLXotml-eR@e2
z!R&A@X)T-$iIdHX0K^H_+q3h|+W7cr^5Z-x@@BZqdG17cvjoBfpMBP`%<d?%;lkL;
z`7NNS;pVQkw2Xj=9$L&sel41)7W}fL{E-`1xEamvWMciAU0I7Zwc+R_YoVCk;B;WF
zxCp}Fb>3?&d{Mpy$&+S(ce1^cC2aeZg|(bW|8>YIH~f1T5aS%)kY3|l79?XWn0|O=
z?d|@#GW(Z$v3mCIWQ#ZTe70aM7D<YS8sm0$9hu9bpgy1IWPt@QtKGR2qL4U$r@_|s
z#EEstd~J3%hQ>$fY#a@&CTG2NB1>U`^u${O+WXEu8P1M`_Wr$Ow9|AY0-_N5jzRtD
z8t1Y0gqia)Aa?$_B3p#B*`F>jN1+fDW?UhftiZY8qMTN=YNE^}=$V_A;-<<;l{CQO
zcNU8F!Q#!CH6}`If`zk{1uav0vJ}<ZNp<lfX>dp7u&1)H;7tLN-Tt0Hi!tdizj*96
zYq=|QsF#wZ8XD~UOa^uWl6pR;cr@XwYo3)?xm`*EC_iY#*8@xjJ?z=@(wK%HV^ir{
zvgIXVPz<7iV3HaS>~QB6Cvokmd6g@Ffl$8vg;zGTpWU%cO$uXryR0rQB_q;ViJY9`
z)uh6tX4q+9<)vibY0z25l&H7#q+BNO?-^PTt~|30kpXuayopM@bhfvRZT8lg9!Pj_
z*2@Hl6JMN<B3<5Ysf%l>kakrvIVgn*t`_E%%+KxpRH|h7cv&_$U%w&PlBV11;>U53
zrxC?rK*yxM5pc>Y$xj3d$$^%JL;8J@g%ku6-UFATF4LAW`lGq2G<?5qyBF!V-1mP<
z=H@Mw%gY;T)GL>)PF9-z{JwhW9AiW(I{E?Ku6CYEd!>_!{QlcofyGc~*(?{n>~PHQ
z$*+iWq%HXS!FkW0+bKEf{9&P8=MRugV+Rpz&g=;ZmS?7^^4T`gapVt;fdn>FP%g^$
z=fB~4n;b{<+C6x9-;^C?<0}622;};`pyMu0H!`{@K$P@!Xi%blE!eV29k!YB^93lo
z$hNnAd#`>9U1bWkgi@F0+aZ1DPM0cDKeZ7#O4!iU>~M^d6ff(wd*Q!ruY9@A7%6+4
zKZRspYPRXStVoKt_g21oST`TOdY}%N{Pm0TnTI(7?Wn&wSV2p`)QZn_QIT3_>#&3M
z#9AvuTnOyVrkrUJr1|aXL@3u)_X21R9Ebc=CSq3lYJ<s%qUgBNN2EMk+3?;lPYycA
zu}a1`dI?b%syGhHR++K7_nk)pSHtm_2Y*p#J3tB>?e(tAArF9a791=<{O-#v>vFuT
z8%oN);SmZxG}pmYfLwEjYA~BEgw7G0U5AE0B!p;@9tx7msP|SPx8(7M`T+ze*#eL1
zs?4)-L;{5q#!8y)eqAhXS<<8|cBnQZ2C#1lG~~Cqz(0y^XskX!9uiu$RtL*C@pt;^
z1naAvu_nV}sY`>F3`cw`c<v|tE09Q_|N5Qc7p(rp5V&IvT2Rdcb^a=?jYAJlN7v5Y
z{WG_yRMxL*M7N9?4MLN#ya9gnR8kF2SW2+ZEji!HuD5oPV5{l~qflB=wr65{_Fh=a
zvL8KL$}I>)Oh&z55VKTZhkN#gJH|C{C&7(|`8=8QeqnqxrQ&LIGVLG8c;BK!>riAm
z(2|MgDbcZv=jkwPtyP26Jra2DRWz-aklgFQ^=0Itd!)Bw3B}nOkniSjY+^ftxMhFP
zc0pKWkABx#-I2*ZSm54dXD=>TT(+mFr=AJxv?t%~AI5ZvS(}R<1NLY#Aihbsv9?&~
z#>s{rC1(Mm`<QK*E8US1J`n>wMX8AoGze#D>$t*}SgOsmCKl{7`o+3Q02F!aG^8Mo
z+G0R<^Y_-;`2>{$7}@wqAd>LC!NEnIaM)A;NJ`L~%$F;12_4~cl2thJt_{Oa@MA&5
zvZWLOc*TxD?u}xc0bCD=r;HAomj1@=X4e=i>NJKk&W+R>dx~VduJn(^hah)esVvH9
zMzj))!)HOfohDsMPDu6T2(q0YV^&Q9sN|T2ZzqW;F?j%Z7o+@jgki}|Fw-F-XUKIj
ze+XqVd<dmha@53%ceJV5k2>-sk<FJtIMS3$U9(aDaCwB@4lJAJm{dk)HNtGCW1Lx&
zbc;%xvl3N)qf8H-h>ghGftB+DQp&P_DPP*?NQ&V!g5Q<?x_;Cn)gVR7WKyr+pnEJi
zH1*-t$~ptWn9z2iGjk@H)}q7F93y`5W%ukwCg$`2r9ZXK@_i0-!wH3R9+#SFD)XX`
zBf54X_e@5mhYw67Z0W|(WsoI0jeDtqtptB)P+e=}m(skFzl-%UebwW2mQBXc>AS_R
zs^w(k_t`kG^&5d|UaD7Zs|g(WAcafs$uR6?1~4FNs^)%rhny(DDYd9NpK6%bvV?-L
z%r*JOJ^^;#xYJ3-a{a3eqZ-;Wr3OfCKD5_zfM1+lN>6P*`eh`>PGp)%v`k({csZ+O
zlWzHMsYMdcxLAT?W|`5-NQRxrB$GqgAuo3(A4f9@&Y!OXIim#~I~E#n8_2cs$wG$5
zR4IU}xUwLpkWZ7n=?YS^!Bx_FX|~DLTbrgyO%4dZo9=37r}Hwh8&uslFDK~^>P!aT
zTFMkU?A7`!rDN5Wj;87PmV#igeHn;kVQ4!|x78FhmKe+$-N1Q*jC|!pD4L16^EWAP
zNJOBsE9D{=Cqj+J^TL~Ql7(00fVx^9TL94$xXe1Hlxy_%)Bd9nbX628Dm)b((l+T{
zx8>xX%cexP=?j@n$=Sb|3kbO^zxNeS|BHW)Gix5@f_bf)oQz}wx82h*bKH5O4Emh&
zr2g?@vVb#}!&k;+8-K^hqnk5_y`b)R)e`Pz<sDy^jvns_@)0TLmOPuEQr<*C9WcIQ
ziR5X5#UxvypE;GxuyDhSmMLRWjAMyr;x-v>jXW7XaGAcCymOhpry1f)lI&X_(`wQf
z0j9K~Nv=5ls+$xEHhoc+&|a!QWRW)!EJ=E|;2F*mDGy!wiQF&UU<)0}cPleVeVt`U
zf55vK8__7zUNYhv6*%HLtC4m80Q~n@;O!732^iHrQZF(n@w(nKq$AbvvP}wE$bP_D
zUKhZWp?!OPBx&h;&KvP3kLGk+S^3E1k%YJBn8S;n(Zr1<ZQ>l9oPJ$WXXG@hb!3vw
zWw6zmm8|v$?3t!pT1RFys&@pwN?BqhIHef5lW2w{UImU|YaG2@QU&%#Ou;V^t4XgW
zSih3{7jAfk=Q6;ijEampOXE9A)?m}Q09(XNluN=&+H7YfhodXMRbxXDoUi{DXe7Hg
zCzyv?HnYyD6PJEJ^K{GWYywBTz%RCbjR(j-srM_7JrPR8J-f?eC(4iGQ&eMWx?y-4
z&p4@(svM|4^H*GP1Ee@~$6l~Wg<wROk#O)%7S@FMk-l7)Ungm#wuy^AonY1RseRWw
zwL$`xw<cWrbgEUy9QO1oo*3MYb-Lnunqb|famJC1Bk*Z(Nu?^1c6x+9mS&=e=R3V|
zj4=jr;iGzQR~}GE%7cjmjohNctWwvJ-!lmd+d4rf4y^-pYuB!-y6_)LwMXFDoRcb5
z&do&ZWl!BL`M-Ykuy%STf@R5e?;fl-c*@&Cl3gL5M54<2CT*q3-jJIR%(wMmXRn_p
z)r|9OxZl94_)+9F$I*7;v2L|kIjm7-;jS`6*rt{G(NSECT(n#%(E=zw0zxuwdQ0oK
zJ2sx{<kC@kz>+ztfuTt#3%$LkxRz3#9&p+sDUhagoUHUx){?O0*#Uhu`qp4{AZ#I{
zZV8<D9dctRnA8sb8aT3jM0!h;dTdW}35%R4s{xxh1G(uDbOVDoD5MSbmpbW2PLBL^
zll0m3o;kmafTKgn>#JjxZ(|p{)5|u!tyKk9dw7k2E7;EE(jeqMGVpO|O?_b}i_#OA
z^m7$i%J@FRc?y^Y#)X*(*T8&x=J6WiJ^$m-^=}54uJJHA4w#J@J1*r9vikj_Xq`Lf
zbwMBg3uURDo7BNi!6SvoJ6&!%`}g$G7$y=-W{wA!!Ym(Uxzwp091m_j(8t9xVu!Ca
z0EJ6C$r7MzHw501uSuK<a-bBq&ti?^Bn;vmCH(g#BxTTb<!5U8RA@HlS$~3NB({`_
z*YeEjv=TR&-?Ol=wQO5{vNg!#Q8?2&T}L%giHzY&1b0edY3~ixCQ^-+<r$Mh%RG#A
zYmLk08N7m>d<=0|71R|Uj{JP+(h{)wk?`NzlAE#8K{j465t-kECQoqcs3ml#48tU!
zN!;l`MVQ2X8zv1VQP!qrr_w}&2@y^Xvm3~EY*DR)2#)Q-!_W>doRPd2@}M1jbh}BD
zZ(RDS;h|(s2Ik5lZ!LVxQT|8<p3YQjJ=U>*33Z#KvYug3e7o}XEkY6i@fzKAf0L<5
z#nzJ_*46lekNZt5tjqR^Wrv_nljf}KNR!4{&X>?5@ba?9`^nnlKIh@5(BB9_EesCx
zO(_O9Q@16#3DGmPfv&;Yh>Y}hyUa&SL7Xd8L5?KBcDux`;9~>&Uj1I<!N)zK)T_O8
z*@v-gW8DoO_~<d;-O*CO&D7V8IbA=Gd~A?|qHOolkd5F<kF6HN#eHt<4=+>gc+a@<
zKv%y~^bUc7AwAb`UqXX>;V0@IACs^#-}QbP@p}IQ)_qT`{Wp&;rmsq`WnsEOKWI?(
zgMQ3ndhV9Ogo7{N2Fse<;Q*=Ij+ZlTzM3)C{bUMXq5E9MsMTTuF2tGP0p{RJ$S({!
zuZpRyOuxjd71GJCOD(pXUCj!7zpOY6y*D65@Altu7}B3GJCy`pe)A{dnFe=8_*xk8
zEhDv3{eJ0)7pu;&xr!R`#TAk3(u9f5KnhkL5bLe<y~01n=D}(hf*ydID*=5^Bk{mA
z?QhTFyLM$TE<Jt?VDo5h3gDdgd(-aWG}2n+QTtbkNZv(q5M1F)SL8en35a9)WJdVn
zmPceDxi)i(g%?|$T`|$_UEpyf4I?!x<$-+&?ef<qP2?sf*RrzwJ7fpe`q$(pI3}i1
zMWRoj)IBNf-o_N;@9q|zfwj56x=a#xU!`!|>5TDs?!E(=5go79U3U1_*EtXB*~ogA
z_cwP1g84Gng7K7fxSQhAtb_iZ?|0TJENS-;b7?Kkrb%e`Zc!(*JK4Yy9m50i6O%CS
z2K>4DWYu(U#U_w=mgw^<?Ifu<ElTpH)qeI$){~lq5)ZzpO$Lo_S3a%%wn=o<aTRA%
z>c!$A2;>9gd>{7s#2)WS)AD5MIK{)n%pdVD59V}6YJY3ew9!h0t!-bF!Oqm`tTep}
zihOUfzCN`1Nn(MBTNOc~IHM+FGfF$e{>cO9BQEn|$!b_eoHLPySjZvv6^>fC-A2Xe
zd@t#Dj|laHxoOo?Y6bN?=cS0sF;8*6^63<t62ePB<sMu3%9lMsDlRPfxWSC@Qo19O
zeAp^~X-LXB#i2vvzG<!JcvF3+=9QV&=6M)f_*jx3SbP<Ba0*;{6?Ss^xU?@UZ6&Zh
znNw8yy|6dewb!F+x~6hAKxOn>ka(%3{<$VgCghog0n(D9kaetRXLM}y^x@NpujvCE
zPh>qHAfQZbGLU^CKhT9`$oAuSvsua1xlI0SyhpAJ>oz)L*yn_WaWi>Gnlay|lQie8
zI2LwLTwrJq{)mx+{NQ(j`Eq<zOl|B3j7qYeUnz6na34>Wg`&i{2JY8j+9iy}*QD9~
zRYr@S5J_h%p+3CABaqaT3c|&gTxyeujLZi*u_aL=^P@TQ<p!o_PMBaXt}stD<JNDM
z_nd~mQ$}lGC;KVAO~$ViD=qnwe5r&?1U09#MCj5>*{-BIX--Lxp)|4TNSpNYsJr}r
zvQHO*vMM;|HYJIDUz?9M`}Qdh0W|HIqiJ!rAC{i~Wv$-i%aQ(dddlst{_90La1Y#g
zm!|8d$C*992%MQ+y;$9G^#X%!X7y7Ty<^YS8@*?|C!GENPw>Rqf6n{=1%7kVo$>w$
z8W20E<*By86Az?L_P;`IH(#E5UmxZ%6eW7|?e47~+mDI7pMText=$08E;o;@cx3-|
zw(?xP**o;?u+e++vU^wG8RM3&ea1VM>)zqc={>bO)v+d=F1qiwZ+xC}oa^50V0O-X
zWTMbLUg1r0^XmwUt^W^je|RV9Ze{}Vsr4>BGO6^^uUb#F_by%ef1tHKo#VFNB+s11
zn^wSeXG~D~>^bj)i9+xA!T-sP>5Oqr=Q-mYG8t+gC-x@U=KQ#31!eBjj*f#*Ct-(O
zhVs0__Dl+W^$*UxkJir@IyWo5dv(r($~T@hx*v1>?nH0?-wpfnqkmpSV!RIcj@LCI
zo%vPXlJA}$UV%e0%;neq>E4%HHunNfqraQ4(`G?0za*coS2cmZSLeWAuJKEyiyyp_
zPxEHLV?P(*7y15kEK$ZN2!8Wk6L?9_2mVs^zC3Y=hrRxte|@^g>V>m-x{2?7IWv@e
zS;cyN*p45fdi}E<U#<7=9zieA!&l`$?>)b(eMN(TZ=I$1zy96;ePM;5NDY8YxOD9N
ztf+bTd4)h+`bIp|eAGN9TyI@mJpRYyqvoc5%cbmW?eag8|36aBNQ{q<n@>o92LzIp
z7Ellt<mHnCDe&;f2+8yD$;<Lf%SllG|BIM#p{rRr+j!Vg^Fi+<@V|q6>2vLjEscEZ
zeLsg-P)5O36=dvZmStNXI2VwBvn$cEDoMyhm`H_51vsSDxw8rwnUX1s$YX7P6U)0Q
zMI<7GHWZuUVBomI@#r%wMd0v9DPXnbz209w-TE!9bXMIy*1YRCZSm4d*bXQ9F5Lq7
z5&e}m`=pgXHxd%^J0v6&kePyOsD2h0Ku17A61H}aFexI2lZiB8)Ki~@{Sje8qbJWp
zE))sDtu7&krHyRCt}Y;cQxv&QucyI-IDmUhqo=@wIe>S}pr^tEOOM;BS@Wd`jUI=L
zK~HTKBMH8t7TH1`Uq_0yf!87oU%XGXCg(luAb)~JP11V_d8!d=z$^7V<)B)Ed`;YY
z&OyfnS;vg`nDXy-aL8!N7)5NiW`9s3;fxnmFzL!oL>&I{<N2&l(pv6Cg6~N?7BPr3
zuQRV)Lrtj3o?C>gCC{FiVVagiG-B`LEyBYk&mfVo@@JjZLWDtIiaK`c^Cyc+{Q3|P
zi|qjqkz;9D>Q*|zB!C4J;_zJ>heN_bWzh3e=zFFH`Y82+%=bh*r44V^=t0n%TCmxo
z4J_V>lt_slP?rtImiwTsF_sM{NO(&X5F4D)5UMtydEj*_s2$w|n_Bp-*+zLw7dXh(
zwc;kX9tk~vm)2fR9H3>{fQKKrCVm@)jgB)f?D>`)k!~O>5~%E4Ln<EFaVH++!4?|S
z9)B~i0AcK-Zy+0xVbS=EG_aEjKIG3p!Hl{cDKwA5g8zx)ar-TTn$<(=;XV#qttRn3
zwtQS8o=ObULV2JsqUh&x!tN9m$FH>0LBq;1MC-%bXMe`LX(I`6{G}3`TJhBX@!U6l
z#hee|`b#+iNwmu=C?!0MpfKWk?{KwqdffS*ZTtpHrbyyYJ*s_sq<*iYNE+)+X8Dsf
z6tVS;LFkWI_9-SdE^w=9Kr)G@4apiY-;3kPPiKm~cq3wQ$cmI))=)g=f6j1Q?0ARg
zbtTJA$U6qp<_0pw|3v3m`&rm@j-cUx5<=OQ4`l*OpRz2baVoLuy$()983t4$?pfxR
zy~iTf8NVirdfeusT8IGM_=hK=uuH!w8Swh|DMTJM3YvyZW+ETKP|?F`SE9jz1GIm#
zhEV5gL_rY|i_n1s#eXa*n|J^vG-%W5AKPKzA#xVU8`=@9=C&rW<%0+_0PQR*l85<v
z7Wl^(jKsikC|0uBL~&#gTb5Ik7dWE^DW5RPi)(F>(32ndh@|9MGVzd~JCs!r{Zy|>
z$jJMup~1zo5&Yn+wGerO_mTT(URCSuL!>7(Ahb~^sLS?i(z!^!Mu^idN@>%+L02WW
z+&Y`DO#V_lU=az?22%RA2NvQzRv2FC8^kPH@FA|0*vUvLnIh}xJU3zE`0&$x8{`Kk
zvjCplpg6021PL-DNc{wN5F7j8Byu%|Fc5-IK^>Ek{GMR&(-wUhY(bI*LIi!9YE6g*
zegxwXQ*e)46c2S8@p(?575=uA7>@+Ob?RF+n8<;YVB9E9AVE%zsWYR9j2`b}$r)Qw
z1{|E`BqUj?CZPz)C-S0hR$qz-qXaLR{~kY(I`lnagIH81bu@9YliB~WRc)U?yomKO
z;_$vQ64OvhEtfC4C{JrE6HIyBaRAL$hJl3-e9T93DL7ZS|K;isQ)7=ul!_EVgX+L_
zV+zovI&c)}%ffn*9@&C3&(Z;BA8GeVwbX|I!9&d+nXzo%0E;w2dC@W|c8iq9)3Nlo
zNV~LJBcrw>+~mN*_mm;RPstgV*-8xr%{cR0G;da<k$Adc2X(sMIOEMza|E6RYgk+g
zr&$mkFqVDShbE|$ji%u|HL>~y=O4V}?Kz+o>l$yKOo4vT-#qv|E)R<&^4AC}T=)N_
zXJf>F*;$8&RxE@#6gf)c*@D8Y>=^u>d5{HrW;nMdsM1*k!#aXR@jE_zuJ~W+f+nUO
zv4|*a)|d=zHTJ0Y*wArOVgH0#_s~)g{{?4T{?UdRdg?8YV1r#04s{Z7Bk(|_6$3Xr
za3J);^1pFn8kSW4S5ZIe-fC+78>YX1_o&Jb;FJe^yf5OlJwh!l+9G1#s=-zcBzy<*
zi;xgfytyvVPN)hq5a%KfTQGF4KcidN4SZ#`4q>-YI;ASfz;KJ;X}?Ee-U|`42;lg$
zS)74|8o{IW_mPD#1faEuI6E(XkFFANgL9?B{>Ey+XX(#o>3dAL-1We*&ab`c77P`%
zL;55D)}Z~T$P80O!9mhbnZ)p6z0to3v{cpho0EUP(^8C)UID@_#BHgEs84&<8*q&2
z6qs%|i+m{7sE0ZedP6PnY#E0-OL}uH(At$_>ZM;mQMCoJ!~m84Gi#MA5|2d3z02DL
z_;p5*v|9LFGP>B~j&g3gb(G)d^#KplYkm(mzsQTo2_Upcqb8p?_h)^2LCH0A1eg8~
z%HBFGis=6z-2j#bk#1>e5F{iP5k)|xyFpqSk(>ooQa~hBY7wNnC6`#bL_!e+>5}fe
zgP+g$dw=)2|J?gL^SpLu&Y5%0PMmkW-ZMPBnedK^h}A{aH@B<;p6yF$Bq+0Y?E87I
zaaba|7T6D7OGi??O#N^gqTaBO!O_WW0|+R2ry@FzA1vwj0;8+8;rW0`4s97nw&d_n
z?G~yBxHG2b*8=gTww9nKF*JBEeWd3|0T5S0;Im)<g-9z&Hwr#Zy~iT32Nc!|yZZmG
z;3hd=E-l<DVw&?mB5|HQq9wr!hVOM)#GSAJK5FFMkH)#S?q6C(y2PoeF?|lVBE<Is
z4ma31oB#8$xvUAHuP>rQTu<%_oUX%*V(;HYTqFNVS*d^?vS~?#j%FxoQu%h=j{u)H
zddUEFo{W%lL&}3!k>NW#1EuZj*aY%<|15d%EIptT`5kfL=O`T!T$!?G+-QCF{WdRD
zw}U=>_1SIT<Zy?$9W9+XK<MVYb%lBE0%0C!4&xWCM#8o59n$rQ)glB*$4i=BZf$3r
ziSV6#Y^jb3A~>x7q<%1v0?Re57VX2;GV%lFLh(ICkqxUy{xaXUytkCn9s!=NBWWg~
z^f>{yln>JH_rJnnX|nREK6U^<R&e-9bg4#A{si1F;&#h7%B>Ip<#&>yi+EoXh|c(c
zm!KSZghW@;=xncc3N0d9$wlM_PSPmDz~_oa=8Ig`rUy$jg?3OoPfW18Rc~qiBo26b
z?|(-!sJIi#fK%;f@M(mlHElm8F=_>6hF$QW*<X^+_t|MK&eib6Zl8+COP?I<F+GwZ
zVr(7_Q+_^~4A{9*zjXXkI(y~?62;Ol3}>Gu1~HucdO3B+?8J~mBwgbdJTkSq60^#G
z#wTsSFmw`qCx}{7e=#E9k#kD6a?K#jfj?u}+Mc2GC|WIplY(9IIVe6C_#<eyf}W=q
z-XXq6bnLMzLD%%_;ccsBL*y@2^TjM9hUF9D${|><IS}*cQ_YRY{m4qBPrksYC)Xpp
z{k%%zpkE8q1<%9Tns)f3Xd@iiTkQ51?$jUCnVuJ6LPXcSDd?C7VWy3T$;?P@sVwdc
z5wo;!z!2@9!If&e=m@+0u{#gH-SJzYo8P3n*xe9COQes%PG=37h7}kf7k@YC(aCAk
zkp1FHTL2j=-9WJN$nH7nx92!?<&Iq~RCclG=CJ%=J+x`Ah=v%lAkSBYF7j<s34&dI
z8I)~b|EL}@n5h1FutdFKH}k>mP?cg2K{}g={dfA%KliVs@oDk#Zn08IB5Cmv>f}&1
z1<G(d5mxGs;Foa_ieMrRsDv`!ZC0w6D4OhGFJKl<$9mHy*jWRTa1lhr+xUeWw6qL0
zp1zKJjmD=8rO+VkMM~vC648V@p{i&+6FhySKpo*4Qq~><N`yL67Zlepu@UCUFySPq
zo<V9c&)pDSA%epsen_D@qH{FaeyG0%ggE4uF!4i7pfEuvG96d^JW9QXXNia9!~!6c
zfqb{X!;x|0KtSOp4`=U1uz!p~^pIjWLl@jIuR|8_F^kaVT*+grx!II(_K&L29(NQG
zq=x{Naq-?1#d3D?Vs3UloLvfcyFI7^vF#b;K|lNG+=gW~*T4C_Wc*31oD{?Moj?!4
z*{+j?01XJC!VMlgPBdjgu!kD~4{$b5J$kCtWKSnzsqmDszyK~7%7>;l!SzNni=iI6
zk(eMPkT>n|1ZKJHai61^yimrockAHN3N$NtvS^wq+@N8^X{)R-1h18x7LR>ajuzM2
zf>90iSQtA-ZMIk>DoawpZ+~3rw@8K*uda`O1*@zNx(3H<(0c`dc>j7;@EYi^0q+`i
zLqYfr0*`Ri{}T!~h75*7Csw8U!o#Smc9^L&E$)5~F+0bjx?ywoSBP2AV<njn*igg&
zg`b2P1O+HFF~c;@!pdy_yTh-|b{nKAANE@f@|OSubCb@u<U^e#C_g9#|KuX_vo1em
z?WyGOP1+b5HykeX&<SR!lvs(Ug0$OeI#ScwzOw}3@wsyfuI$0xzi${Sr#CXWf4V4a
z*=yi}ZFcZK>}PA|7wwPVG^G4+U_5-rTqqLK&o;n+cR&TtLoNzYc?Xqg5!GpoU-24Y
z47@CR4KwDsh6Nv0DYE^n*lZ`Vsvm9P0?|L>rGi}!ip)-^nVtW%{tv7=vM@WZ(qBU_
zGYEdn5i@Ml#v;`FAAd=w3$BTXI%ob&Dp?khabD$Rrf_+eCm{2qEffT{WaEVZzPK5o
z=7(WzZw2sT1vr&}p>#D@5klv8Spptay3ZBN`Mxm|6B<L=M`GTGhA>_Q-oV6SZcD=D
z@#LXb@F2n<yusF?Rw2w)h<S*4uwAPX#8Lr+6vl^UVxVnENw*Mua)QrDv2bWPD``h4
zP8_}oD?y!ts4_Gk18-y{$%eCZLq4(+*@QC25$z)7cmT-%pT`WQ9JGf!VVm{;y@TQE
z`TJ<TPAqD@BUTlvAI6dlFCmaG!7(8SG9eKRrU%^UOrX`Xym{~gG*m3aX;!)p%*?xE
z50y~hvxomh!>3@itxUrN^+<^oXw$6VcBtGi*=8_Q82=4Mm==1}%0mmEw}1(oJr_xE
z0hb+@dThfgHTx3vR*rFFii=44|LlVQrM_NP{=59Q29WZv9R2U8`59{K<^L}5dFriX
z<3@9rorm>f778<gMzZPeybIq6P5QaXn&J%${9B6Y%)&Z*G>N2{MqJ1ua6`#4AL_(y
zmzW4Xui#kPYLZg5x5CLSsgDw@v{~bOAf4k?AD2%GhO=aC8Xd)q=NmMDA7(smC}GUb
z;mc@QCKPkoBeEY^hz-j_X4|dIbF(tTKT+P4^_|R2Rzk4tj-xn$?BxFieq`T{;`<8#
z?XE1vjc>OOSIGUn!eym9(o;@g`>$meX842~E<OG#l>Yd>K``bR^or|2!bk6br%zVs
z{ek@i+4@%ICac)yGX3Y{M||R$?>D~L9j;J;Z8R0y&d=xyq}jyERX!^H(O`Rc&a69Z
zdblEZU1(dG{H8DjEXnwpHT3To4u7w=O{A5{GzCn7w%5nVPiEzAeA7N$Vey}~Exk;b
z{N{dm+RZxcRT^ggaZ*H$I|+yZU$(lD?<}74=<m0z!xdToc3W4!Sl$&qf3HGWou5SA
z)mr(4r{m0lPOcOdbHCx0M?s3!-!0w{aNc6=-k3k<x$*ls>yMDqa}_0SeDSkvZBR>t
z|6#?$u?aj^9D5kTLry}WQ1r9qqSO(zev~0Dal~3k-fakr+sD9*u1R2kXaq^o4g1N8
zSO}5PAQbIyK2rIcL`I0b#RJ2lU)>2xD^}Bh!pzcwR2mYs8AuC|ba|Pd>4ksgG>mL4
zw`6sGip!cr&PR6|)RcrsYmm8slQyTDE%3BK9EB>nox+YLTFDm=wjGAKgy;>EmKeG{
z$R!=dg9+Ud5Fg$f0VH6G4wC`1UVclT!g14^h8A;s{wu){ehOi@+(Gxd*70O#sk?|$
zzwaON#^ib({T~8Nvd*bBJ%DNJdfLYfRr1<7(W@D_{oWgOp->A(s9CesfT~evNS|C0
zoAf4LD7lvA<G+eTm+BRc?$2qIe3+<#seKo0nJ3LXANS`HZ6U62^I+UQA}=J86yN1d
zOV)IMbKyu1V*hc!YZ`~Bx$u2u6t+PS1vJ<<MAbBaU*Rrl$fuS$R}z0XYZ^U`{b3rk
zar)(zVxno)v{L^xF;P16II}?`cnk>d8}b^B@rFS{Wr+E<k@&+!)4>J7Nnmpgjio*x
z2`gA!*qB>&A*h2n^1u8OS}>+2k~&pfJv))am*a^lYA~o(E&cJlY*t=MLtt{#Y~Dxm
z3xeB6)Ai^LZp-S>D2|ekrWcKm&kdK_>5!|nc*lDf0kdXD5$B^ATwvC}0NZ1lI)!Zk
zN2ydfZANstl01rI=(FPr90_9TO%o=W%4UHg&6^xkN^Pb^zp(YDl+)P4X?LPa=bXtW
zkXOH+js6Pk4ya?bV5NP9(qJuRHrUp`diaEql~_^%$D-(0TFvk!P2po$c+D)!ej#WF
zC@_ZC#?7{9DN!g_&6)?PC?&t1I*qB(LLuq|vMp81b>fIef+y|+m)p|?%9U}o8IF86
zkNIs@kxD_2FP_~o_GRt6pZ-%v>25dlD;j^nxVz;bzXn1XqNU+{Yo<s?i69T>8MQ}a
zsR5`QB_k7GpY!Najm;9~mB(cSA<O%a@5xF-8Bo4q@M>z09cuA>gucQ}d)&9w^jEjW
z*05Em6i|7>o#Y2b?|k9<iw@N<`52pqW|*oOo&vPr3p#oy+>3Ve6FkB|wnLeT1kR%A
zk3t$O$cck3g<+eankdD4lk?D(%HheOd`UdeU-bcYA@;D6Ns3Z9x81(KUzS_@G*xZJ
zQ`BLoCmY;2NQK{7!uLbJpzYkax2*o^bxf0axiUHKX8<}bvNPD@b&q>*J+;5oQ?zif
z$60DwJ@vKRj^f`a6SZPb(dxmTDH-m{wbcG4Rad+IbTXOmqlpOTo!*bC%Qk{(@9DhM
zZHJ{kN-W#Nq<VJQWyVT-+l*T8roQia$-LAh{gHjyW;4Yzclf^f0gYpriCTiEsODhL
zz9hHydMcoX@)U&*_E1VJtABZYnW^eOisD6SsX2R!It_f)cFJ@;&--Mr<;0GpT56a2
zShzIx)iRS$UiYS$wR+0?kIvOCIs&ABtBh98u9TllVwi?4Gl9aAzEN>*?TJ*i2cDZ=
zMk`AQ-EX+$F(2&))69r2rnCz%x43epKKd?At=0m=LS*@{gBM=v2!Z_+_Q*`S??b6-
ze4d+AwZmJf`LKx(u1izz-0EuXSZ2zf*E@<t8`|Cw@Zgyp8w$GdK(iVeXl9sY#e;wM
z(Sx{1KZ)nh_!uSoZ(O<Pz6$1<6?@d1WOaj;q9erK(*{tW5@CgMz-W|Vr}i2DnFNKp
z0f_q&Pf&17Jl=?cOgEksE6qZ%R-8l3r&5PlP+5&cits=U(J-A5RdhciXhpc_qUmu=
zREsqm{;Q?l82ym7-ZxncYO8$#I)!#P-&Dss><3;o+s6D+6nBA51P==CWf)IFE7Of#
zpi9!V#3%(}oKs|!6qgftlL6VeqzT#^|J7yDw<p>ya6pHs1#xsJfiV1ZV5gCiDTo5p
z(PiGGFUZ<jgq0l(GYUNmh%lfb#xXWH3yH+9V0MhbJ}p>*)Ri;4#K;rr^%KO|GAWF|
zvy~-9WBn4O(oX=#NA(lm?9#q}jgeaa2GJdVKd>XnX~7gW4!GMhsg5@-9lO2a6}C%D
zEj{bsA)Ai8HQ~pFNj-4(Jde_+;;D|&T7M037~dNxzc?#U%07VEimbG{TQ_wun)|+U
z+;MFpULk@Jm_Ew(CN_Tb=|IIZph4@c6U3}|eoWe|IO{hE6Dji?W`8vtnas!Gr{~9C
z?P8_YUslHea_a+2$B7B%uF!0M)9=eS98BLYeY0vBXluJN`@4w}M2W5Y<>XbAoF5u}
zSWln4HRUId$?@he>B>BS67|Vx-fT2_^v(N$QH!f3=a8&Bv2j<KH(t8;f>B0Und*#|
zfLZa=(8pX$h{}WEwvxSI)(R70b*nI`+?r7SqN{Q%fRx<IQr1V8xVp@ax0(vt#=jA(
zs6z}IZg4h%jkg(EN|fO>Wv-1+u`kJ;EPt8nh>+GHDcnMB;_xW=H~L5%lL@s@G#nws
zaReR;^{=n~-W!v=)-Gq^xU55M@+WNKJqRp%xJ~|Ew!sY^1Sj?77hkN30$m=Aiu!WQ
zP8ZPZ`C5-1f2Q&L&rF&gmQw@%Pm0nbZ`N7;k@w4|mk3#(#mmE$P`GDbvwan)_18#&
z!5&(xz^Z9^m2V_Va)CP^BQ<XtxoS>tIxDy$Vz(B_yI&g}G%I2ETmJSq0<Trp)2vt9
z88U@XTxu*Y`T=T<pu#AFLx#zdYLEvCU%XFcl<g6@+zw_a%%sd6km~Tx?AC2X3gc(K
z=w!RxAkbwU;eAuw2@8|TIHJ>i0821>i&s}ePkS2K?e)&1Ir2%N`%=KSdsT?GVlGIw
zp;E>eho{^r$r#%dZ(A|j1n;L}KNM-A?V!GGzNs`aTerv<Tdyl-JOJlp?MekocM3#G
zimiS)-og~>niylhk2z{URR&HH)HlFhKmX<>EYkb5N0L&2$<cK-QmP!|gWS~Dbfogs
zEB&-5bfx!J#F^@gUMY0c_&J_(Wu#o8ME}VqT)g>l%x2l^_l3t5lP-f~vdQVBDE-20
zciq3_4%L0Bz)_I6LfshiuoEx{&}nZl<G-wy%4IVvK1|u0lhA#TF$VR#T={CkHnm2M
z$r-!h3BFlR*J65&hqf{t;i~0KsAlls>Pq{a{FaT=dazP7wWAY#KPrs+>uC~e)Jtw_
zb)6UZ!ci~xo>PS?{PO~k+!vU89_mN|{fvTl1k*-Ip6QRRg+;2@04A)z+0vP`lwis@
zJ*>Dd72=g125CRh3TChg5sOm$t*a!x9ZVcYRKYJ=i>wB<G)k0ARMB}o@2&y?PgQr4
zA@+KTLQu_8@KkqiZsk4Z%)@(zrXDp|xVX~!`^*hj3{^;d(lMl?iCAm(c7s<1QxDs}
z6JpkH;2PARF`fXB2#uW_rRLqQ*;BB;*QT|Xy0HT2IoLlcbBPKFEPJsBS{&`UFf3m1
zCG%apO~VEAa}}y~<kgMu8EyNeq95<z9#`rgid`Lx)^rN&{>_+hxifo<idTKxxKB2U
zt^mKH`zhm1V^Z8!>Jy)*`Qes*KdPd|`beYBe`n;uz5+?*4pI~r(;ocD{Kc0jvA)Nw
zlWb49s~AWH*e5)QlkeOV=>1tlxNrEvctMCw=RxH&F=KVOK}9rYpJdfWIgkI>$y*bm
z9s*hOAS^@q4F{|8N?Nq;IU5P$DA|;y6!W7v!5&R;RrnqARQy>dMDGeDhtQ+s`t%j5
zl?ECC&A(+1bOT8u?sTO|46Wi(r_)N}4lr-(gSLr6aIkeX(UyG!)Ca%hHE>G=W^jtB
zy)_GH?+Vl!SKZ#}!>^5g(pM`Wy_6L>U2Z?b{`}U2N3v2wOVMYuXR%n0>b!8bJ|6)t
z>=(8+0rn};PXc6yYPxm2o1rLbLl(Uo6G<X~vo9>#NNwjxH~4+ueCi-`%|4ByVVG?n
z*@-9G`BT-0lG1hg*CCgWc}igw)J?zPjnoYL#swsZbIalO7N{&~)W`EoUN*=T{r&^!
zZwB)WTtc2s@CNg}WXTr+cXEX2%;N1(@6FvFt4C5O&q5ZReNi+vAvRII*eBxZa9|aK
z5JeI;=%V=0DF&q7xV~t78^eAUuXUai)MLhUXAQhafu*N~I>s;+1x}vY4|csenoz+|
z-YDORUzCM-$w-p_c%xtw7^*14zF}sPg}mfqRPJwErdh5!7ZC<^aR<<QgJ+LmnkBWb
z{?N7P=tDGx2lcs=p_&&n1vrS9Z|Ia0tt9NDg2>A@orE-tS@wfL5&5Q-7T@3~&7_(E
znq`3kF2&2l336KNu;)k3lTU&X79qDqJCcZBk*Ab@$au=fwa#ob#oE$yY@+^<Mg1`T
zvMkbD^)!XoL*TVh(O&x`$$HYhxh{f+NBZZ8xmK3I-5p;?3OEG>XcM?KLE#}_Ijtp%
zdCS>o6j}KrVP@3jHsnu|uronB2#;Ixe?Q}6Mvqdc2>cnQs1eHS)k?zJ%A1FcETFAW
z%Br+AucTC;Q5zqsh_~)=W*Zh%8C--(_~jFtfEuEXb-6%>u95(zkaN`75r?PRi>WHa
zFY0g5S9<m-iV=W6bBMxB;9W82zM*5~k4(E!(l1F~moA1Ir56Sw4~*K%I2n!BDt^%D
z6~Y3OhV2LflQxYursV&3OW1BT9oxLtqI0h!U;8~Y%2ZvCI4tQQ|6#r#)fPNI<caE!
zDlyygr-kf-mfuGtfg7YSJjM=N{*p>$xdl1{s8nfvrtu^iP!y7_N*FJ6qya^sOK&0Q
z;t#PWcW@U9*HB8pHtixNBjKm6p0Hv0Po}IS+uVykXruf`3p-uV@^{wF2g@-C8ETe8
zRq;Ptz2#<6LxHa&4Oz)-3<7%Vt^S#oY8v>v0uDcOxp2`ql!hbsfJ_4sq59Zjz(d3F
zF-u@9Z!c=_;cLzd&wQ&ybZ>m~_8G4kUNZmc73dPQ-W<A$0j*0u@ZFY21qMz%rNAbg
z7)sg`xNCy-b7Pu<NGN1+w&ZA6KSAS~fKHRKcwnDxtS0yux;-8^5yiR$|HmJ?g7D$T
zij=<a2{!ZMro}np=ez{Kk2y$kPFnv{x+sVZRZdy*rfEMw+AZkS3{@V7JF;xMdORO6
z=pFO1=R=QHmfAXrP=5I0b)ll9W<m_@u3jliXypR6O;DM4v;O9Ao2*wi3W)mgHQYAp
zHtV34nMSEmk#R@8Qlrs>N+mJl2mMMHe9QyX+5*oug{l!R0Kv7=o#3P4PvcfKZqILg
zrGDphW1adA|Fb^=lkf>@c*$5m9S+XJj0$7>nsC3Ulce2xK(_%lxGGz-?ONWg^?RE_
zcSS_QwR}|TcYG%8nt&ek+_)nLOQ`BQ!L930fc^~zsffruZ+t!0H|w`b`w~#5`j!ED
z%K!PNmF(6IF@PKOvN~KV>PgaW1gCgs3sw(h5;Xv6r{01j7;Tg#3~?2RENufVAC!uW
zALL`1Reu74t_HQ?icvwTC$T|Ta+ezu)M1bjek<B5yl4{Ln~!e_?=;x)e*Vp7{e$h3
zz}(CnC9MS+yp<K_6>Rm+^V{-s^ybcaq2J|@&YfE3?Z6m%|B#NCE%>`u`}Z+HSO?00
z95JT=Q-*0KeKN@Dj9~pF|49<GD<1RXX2XQKAug&1&zx$-W})E31(UW7K!+7OqZ6qN
z?%opnuB9sXW!fV+NxvImxY4>S9#j@d@+|Gu@O{;$T^Vs(-`B%dn9qcnU(ggf!H#Zp
z3C3#&fRJOXD*BD<Q$^?1?txiSY$2L-<r_Vc-5#dTe{iVkmGI9c=O~2)!*=B)V#aG2
z^8$Aw4JE!hTo<@(_MnZBeeLL~@njZ}7d)+yG$tSdz%j+8GSU6;`SZ2!9a*|k@<r}s
zs^^_pD6L3$Qbr(A@ZfQvFkvS~@&n{Oz;B%iCpKPl9dcTSe!%n4CzG>E*S9v<fJqB}
z_;T43rSst1YhjcvQD)NVQfVF=qhv6AcIJNemrLDLN6-(6>A+Pp!F0SfP&F0Mx~XR3
z28}kwwBVi_V_IIL{8+BaCM~D$S9}Tj^O#E0H8@e0dW8SrPKF+_A&n^Xg`mYW0Amxd
zN$FzWqKR{Z63|ptJi3Lf!>uv}oAkg)GZ7ioqP&UlYc;IW+hSLEl<UB)fR$tIifR-w
zk$%5+VP%5`k%lYufck%6g4*plbb?ygJ(F%Fo<qtUocAFP)N!`ZFVw=$nPV~oX0qgK
z0_fkcNWo);TwkCbFs=4cp!xoO*b1z2+V3f^Q0+J8C)RIBD?HR+s3QWt+VPVB>quLk
zZ~D2w7~2G_nSXW|cFM2z!LG!pnx`5AsYZD(fAD0sNtf4>kd)XX?Q|kCAJkl93>z$8
zkm-;CEDMzKJx%U)W$s}xEB77*d-6*?%HE^hYn*ow2OgH}8GLeU8x+}?A}L#y`6jak
zsuY}_FMOXb3(_MTi43#Y4jCr~wFk1kZj>+ku{ujXy)z`QT<4h}-Bs`T=Ifr?Ucl+=
z0%V(Kqx6u-&@#*IH-5|A-X&NojWwAWnT89{_2SuC?J000G-8>@+srb({kgb8ljrEY
zM@@Tt5a8XQ(8p!OBgYWSW!e{J@a?t6-!d)AIszqW^J6FCdv7i?t?+$f+?K)DXm)$Y
zh%MK>h|ymjSbCeHYn}FZ;CE}NCtjl`hs>6kt6%dc;p{Z)WvS3{kWQs4S3+2HudBmz
zUusKYE0I}uc1Ftjhg4e{*~i*d{{juIePrE>K{Vhk7#Oz>zSAphh>30;JPVBNgI)b+
zY=ZZ&vV>5DrL^PJNW74Nqn_kKTGIOiXADQWUMOL5+|;Fu)hk^kobW2KE*c#$PR~mH
z!1xn3j|~qrkG4^gs%Jm$>8yisuEaf`jN+H7XXidT>u@vGU*|QE4y;Tzp~R|Z(W+D*
z0al`gP)1c?I!gB)(JueR6LqR6P@#q6CS5MO>M9ZN;J#Ly8@^e9v$*8H>S<RDGlB9;
z?&tLelm^@f{XM*SA~P(MJe@|0u`p@9%c6{8JPo^M`IARu<DsD4WDgra<4)leqe-el
zN!>uT!w1Q|0-Hu7rf}ab`F{^*&I4#n=C7*ZjXRuW*r)pP>ehv+a?^b&h4G*5{K$wQ
zU#uvjgKy1B%$T>RJ{mtOa6J3>)XqX_@6I61vO2X{Eu69Pjup|7WQM*GlSJnY_;}&I
zHweo%VSfZa2fvri891tC`OZAz{$~D(VQ1m7i>{6tiSj&do2kW|(yJ3WU?Nl2ZI_L-
z18&n4ZeuckV8#JYJ~_Npe7my`6A_XWqPe|v7Hg;Tj&OSmY_f6btkg9R_Ebm|5^1)8
z;@`VCkJXJTq6zoagb+e6O&30G>2K-Y>}bTaM9E-8^=TH9ZXDW<eg2IrZ8i!qX~?UI
zUU)d>Sj^Yan8$hBX8k_kWc8;Hcg=Lv@S)2``mEpJD*87#Ix(Y%Cnb&|JFuHVCD8MR
zItRLgui%{w{z3cN=oHS#TT8eUW~_g}np5S`CFa@!>SkW?p0e&bUy+qWX)hA$>zwG?
z&GoH|8n)q$^|Uq7tIH#@u;CrpSHVvHd0a#Rrap=wX(=8MgVF^R-lsbD5h1a~D1O3$
zf}*0M!pzIK+<vw!dZNi3D@;zB$_jfB!ZWxjX}rVQDVly+a~O{6p0qO|!pjwm)A!Oi
zHZlT(s}=l9lypw_9q8Jj-Ztn2xC+`p-}XQAiC`C5^Xz;fV^k8cU}S6mHd2XoMl`)C
z*9r>o;Cw+dQW`H6vyUYOQ)u88v~s#p=@{f(Q90^^u}C8gf+{wnv}ZBP`Hy9XdmY|_
z8b)pZvZ=m+>X)W$0x?gD#EQFa-$hye=0E1EBbgM~({<G?f=^ZS=T_=&;Q8?H<u6l=
z5!CbVAwKf{#+6O-Z7i@7qQwd7a}08cm2(5gw)-|zA7W@}D~d{L^Ls$S@TOJl1I~Ps
z_`a(6x5|t^xzidoIs8sbk9E;>dkS+;hc|Er)bHoLwD4s)xuZKf_L$PmVBDnM%Kk-&
zEw}v(itW;l0Xq5C!uOIWH*NxmgZgKI^`nPXLLO<X69nUFby5v7)9-0#p+CGrq_}Sb
zNeA^hf8UsJ^LU56=a%$}e`W|YXHGiDk$DKLe=NQI#+y6OL(9T&t3aE$RGoY?o0fIr
zasIQwmG%OqlA8(A`DI11KD=+Fxa(eoJmF5}UpN1xlQ|LP&8^SBzBi*&G+xpC@v{I}
zud)cZVxV&W`=D^kv~pwlBGziccTw9JaQNE9)fY5cL|EKCVf7Q(H^9CacKq3LHWh(v
zdw&nzUP04R@IwF<!Hw@T;vnUa6DintyL6bkB<&cglC}?Tl>R1k`N4C<TFQ4u%bG)G
zwK_FfE{n}0&0;zteKUMf=KKRMt&w`%`UAR-j+?g+Fdk+j#at^dMcbu=TQLM70RI-t
z%Q=FlZ%`Vu5KGpn-I+ayG5T=48%akWud)m{!eQ>L&<=y&`iuo{ms8~|_^9g?J}(CH
z8B`0#;b<vTEM6=p>%4=Fpt^R0c8#!&tnh5(-<9%O{fG($=I*NV@U8;Tfz3)p8N#Lk
z+ZC8;Dt>`QLlQ!@`i3r3d3%unRbJV7q<^Gt2>t&?0!h#}OyC3P$=!ANFYxR5w`?7b
zQSi7&Bd!sD%Y1|56sg!ZzfZ*}C~t<0#^)!2J&qoTeN+QVmo{Ug;=DW~Kj}4nh7E2P
z_ZejUo}3iuuU2py+66A09Qz&LK8*<xK_`$HUJ|wBO}YyM&>T$JV#_C7G73_gj;h6$
z&&a<EV~V^XMYGml97CQUT9dB+7*Z+nU+wr;QIRqxUAfDK!cFcWv*Edmu9qg|_$W)n
ziYNK=lNP^nUaKP3wRbh-o2-L`ZEtiF-0tWNpCOfA-BYd#Ns-WbPzQf?ZyRO9x8kUe
z+s4HD2l|Eycv(DKTbc(WARB>xVwxR;nlZ6XK~<P+j9LRe(JzAs@Zd^%q4W_{kU&z6
zH8g$P1^fZ%MS}NB@zjLq=dNh*c%!rs_&z8hMBNLGwzUK^G#ot&wjc&|mbR}7kq>H)
z2>}oS5GFDM=6%QjX%!F8DI<DFA%&`ORd*6l<O}`$J9zx0BNbk}=}Tt+<SD-tEnbt}
z_Tkm2%MUN|xZy4zKG!=}365KfT3P8VDonG(P*Nr}7&wZ`PzN(&Hl}EU)|t`um*HM+
z8Rn}~j9(&(3G}+ws@vLp9LLm?78t!C)N_o)2>uo+TBw1Sr!ZmJ=n|`wEonWH5g(`t
z0nSrnC_;tEeiB-yrS%027v*$EehPKLNPh~w%}UZ?+?tzQ<;tr(t90+9A4^B`bcZOC
z>tDc*66MAB!}mQY7|B+!VC7?O8I)PQ9f?JL6vTaoX4FBcXy7^`GpPbDjHT1lU*5=d
zpq)Yldzv?CJy7}|3dsb3`Hxe`?I&ULkr~rh0<jQVx7T*PLP){1oEO8;O;n1W+o{#x
z7erOCv}ERCB0OX+mCP^n3m*lR6-2NQ{NqywV<Y?*#KfTrd|v0FhqN%GOYaUs>z4cv
zdf{o1qhRX5w2HQcCu0?Y8#UuE0_|VZX?E09n`p&RUH&YO1b$Q%KQnc=zt=%1)MQN6
zUCl&3*swG=B2{W_tXkIz5txe9|7BD{HBiG|T=qwz$`v;Jsd%NXSl3XNwTI9c$l>a)
ze(UrLGzQ%48NzDZuVD{0LFk`8?!r^e<Tg?%q!X%8Z_fGA;ZMA5u0sPk8gtj|`TW>_
zU<7cUAA8YE_1phrOtjN4T;*xpPTy~)`>m9uD)(!Jpy$7ui8UM>MhG93u0ix6SB_IH
zU7p{vGKcP#JfEMuhOxsZ8m_PJh)mM7*D(C>`4HIWyUk$IjtqWo@~gA@`NCu~Ntp{c
zs0+wMjdAfdZl<&uxBWbXUPGs&8ms~|BZl3L!?w!Xk{eAMNJ7jWY%MaHEOm5lzp6?X
zvT|Wxx;y^ddipN>=L=-l%;QmLnP*Dk+@yx@ncV5|$$VDi9lhg8&1!#T{_aQ$I$aZ4
zOV%HRGajlS?~;%ad6i0B+vZb~KTcq!kU$l*?L!p>QjzGgAvEEYZNQ55`<DQ^8YPS5
zgX9LQk(2Lf%j?|}fNv}@uC!KHxC##YHJ>0s7wQAor%6;nx6B%-U`e}wZ{z(`OZw({
z8&UNAmo|ZccWv${VarI#v_VmAE`1qTmxRZ9MkB~|#PjQac}i#klk&DK{WXmVjJHPL
zQ?WD~`Bb3u?(X2|695Y`mt2Bv`Obr0A3g_d_s|90$}fD{(ulO#8mjalrMFus@UBak
zkrRg17j1LT=r#mhlYLBuiT6#eO(?vh;_aXM4z&cr1UVR<sE5C;O|2Zi;)X=wl9}+o
zob@YOmatY*&voP+P~<yp_gR?`7#U#0FX_4gQDqJ@)C49O#%N0oYTB`F_#dBgS5P4|
zUVBTM#99~SAy1w`$!Mbd>I{n74<vgcc@Mk{vF%tZrjo9FJ)PuY+BfA^VNq1~O!&Y3
z@k@148el3Iz$;=K=w%kv+i(NR*30!m9%ghItxF9>x&ROS?WOFg=bXPsJ-k1@=_WgF
zG^D&&2TVnt%9nQj*f$_jbP~7Qx5Sc8%=hVuT~6(kMJ&o{-K`cD8!1DDMd8Yr@X!9S
z0!KvjvJM>YLEX1V+n~I_Z(-i@GM-Y=`hNMfP@}Shu&6I!&BZ@fx5;Q!ed`Q*+p#d_
zk}f8_{kEC!nZjX-vL^ia@jW_~(VgT|x}Sl#H;dq&tv^=2WHf#L*YyvM)2W&^^rd@w
zkQ~&fZslUiehu$~!!~zmzZ&$lVU4ICli=^Z?2R-#*ixXxg-e%{tsLsp2G~hrboowy
zhJShm=MIOwV<qf>hEd7NqF;Pt#e20QMyZMeO3W$n=0oYsLd|C3J9_)c?<N5W5{~Ux
zUY&inZG)ZT;E_nV?y}EEoldnTaRiZ|iyJo)jfor&%6st)i6};vXw(q4f-#bR95JyF
zUcSj=8m3v^Vc4GoWM7E>ofN}ml|n!QT6{81SYr6i?stBmF~<A1m51a{38|w<hj&}X
zNLBwCs8M6XK;-3LhrfluLDMha*vQ#Rdpv9d3fsW28O+Rg+Q})t32!=gEafWtHZ4rt
zi9t~@A@Z4syd{WXNz*PLVDYw7k?PIaTx1v=b|=|c=b~%s%Zafkgx`>;!fJjQn_Vr;
z7c}?y`JYK6SUB!GJ;noRU5oGZr$vAI=#z@DhUI6->Xk}rGGP0ez|Y7#SNc>&9ol3O
zyiTi-uyHr*0c-4Qv+8-`;#&jspNjwV89`a6og}N?Ae_GHnOqY54VH5HQ29p6DX$_4
zQGgW3#EbJAwue>|@-<5r1irH+aTt%YEvE3PJT5IASL=!wPm>;L(zC-#nB`0xC<Og0
zCUIElc<co5<F<^Gx>=J*jMrnH@NFJ@|Bd1Wn|;IJR9BNTHr;dbY21su+G$#E%D6;!
z*y;&U!;$dz$`{At!!wA)_?x1a7TRyCE$)pJIXpge=VLw#=jFnPeSne~0w&CAhC87u
z0xcyk$cm2dm}`kJ+A|&CqjkSA>79Sp8uOTly;}@4GCwS(eZ~j1*Y%>Yko6qD>&1=0
z@!hxmn{)7O-nZ6AcsuVgy+!7~f?T5=zTd^2a<iGHyc+LKFuY4e9xLjDHFNqc3UBwI
zs_B)#Gw*%vj?D0TNQFf+mI+HmG1tF0j7g=soZ^4AVr!`A<B?Bw<(qJr2Q88w42dvI
z0gNQlu~@^k;ep6En+qu7p+=>*hL)*`WYa|fE4GV@5_~N6hPS>A_=UNQ-}9=(WZsjS
zZT+QR%@FmbayK)@`^I$9^A+{Ys=1>}lWJ2m!#>K)jb;iCzorU^4BKH@)<&A!uS2>Q
zMZ@aY&zh-tP=Rsz{LY&e6l%~w4lZD)|Bt7TN-!PY?|@`da+pCd@J#`$$d_-Rg=-gn
zJJ2O(|0_+J-uah@x@C=W+NV=$tDbiPhni^P{xHnw3ls<k4h=kzJF<$W0+xLX+~8!$
zU@Rq{+@+$feMG?cp!sOX(NCo6q@BSI*-QzS(Sh81^PleVAN@C#ifv;NQAR*}Go}5a
zUzNpNT8i(>O>6+)WriLTuI`7bW!sVT6av6u59+BR_q{VLf<9r(u4*_mi*WCBDvDoq
zs7glS`_WYYgy$XHp@EHiM^<kWaY$CTmD?HRj@X)ouZ9NrWhFe1r=sdwWn?7w4yM$`
zJcViu9zTdURN`1#|N8OIvTXs7?yvcHvz%fAG4`su@7GFw|CHw))uE=m%3y}5qauzA
zMaM9?{$@|lELI86y{Y~I&pQ%B15+~p8dmm<l#625i$62{+gHp;eNbE+`u=;59=mmJ
z===B2Ib*-f-)>Hw>${zWy{(cq*XK}{I9i-XFtjiB>LBBUw6CdmDi;ueD5%dP6B4Up
z3q{_fiJA|EA#!=Vd-FUZ#d5>FdOvs;7%@seo05E68_(RdqRl9+)b(8rrJNhhn=8jh
z_Y;giOE2Cf%AI2PLm8&-z5>FxWw9X@=H{nwHP;fpE$c|gSp-Op{UsF!SEVeFQc7>M
zr(=;4LOhl6++^MWWO@&B^Qp+3PAxYwN_#p15~#jjyihw9@suu5ok0>jiEgGHNiI6*
z2J+cXhm=JAoB|PvH2bWxw7Id{);$iniH$pI)x@H?#4@=}dKDfew0F})yHh`7Z!e{a
zMyDPc_MEMa;W((Dok)fX<>q-vJ94r}>j|)>^(NouGrR)0v2{`yj?>gZgC6+{WO<n0
zr}iO_p7;NuQ-<R}6`Bo-xYWsER!IJE_zo+SBNWAHeo=p@jPvdPWK<%u4SH&4asw-o
zyxj;~G&yCkl?JQ;S&|sg@~9y(#6tsG*YhyN>=E;FLE4Zd^l236-t(=%GXbDudLGhw
za1$_+WU8UDE2?@|4*nbVoA8pNk)#MG?w6;+yzPIuV|ZY0OC2s4#nf=(lxLVwy!($H
zj9VMLshi25IaL-Pl~LeMQwR4$jsJ+8^-<gO-G0E?jeEjMwSD6(!YKgw&K!vJDIS&@
z4!pf-l)r}zn<(Z3@Pad9e|&XzSB>nk<a=^_PyCxH-)1NQ<}D9bE=Ez<_Decv^?SGP
zD1QyLH7;uV7nC+%o{`jFyn;@9v300iowiUZHnE%Bi-gp@SSTCulRduiRVEVl()(4>
zqhlXRr!UVz*WUX_*P;(15~~GKValApA<g^`cC=#4fK8rOm3u9@iteGgcV&&(mY>#E
z4!@Pyg;+7)TDsB3bZbhJ6WO2T%*VtttT}v$-xM2nkVkLvLz4*;&dwpjTRqZ~G6@%n
z?R}}-q!J)iP(3n3t0~aT_^0oiZs(bE&Gr`lRm^`O+&|@dKM7yps*g7$Ri>?nJr$5~
zgxSCFq5;3f%`PLkRy8sc7<-8P#J(@pmdf5Q)z*<^iMsKK*8SCuN1=MBqk(^OQa(mQ
zp8?@VJVB3FbpB>13i?;SmU~q1PteS`n$Ag9%>U7;?RHvyWjOppVo8urxdl1Q3y8-@
zgYn5}YYx1XK{k%kJN+sWG1ab&_{)F`u697u(}ng2eDAAw^R9^%Fs)$3O9g2+!&N)^
z97Xr}RPek|T)C`R^4+uv25Gxq|H~p@#Mpf?KP6(S!m)u<uJohXbM~r(5(H%5@@kNH
zN%Ix$r)&*3gAqy#?+=428f383fTNsG*VJ;{DF}()=?qE1a{worI_;Wki+G}Ux+j$X
zn86+|4c)GiF`$7JaAR0`K#UUPYf{#EF0=x_Kr;}-$C)-7$#O90#MhkmWEY<#@H$&@
zXklQ;`$CaWrmTWCM>pG<5Rap~DzRyEhH5@WJmn8W;I3jZKi@;px`~iWnOZ1#Qe9Wm
zb76P)CW;8?Ff#lTe>kjxQcN+Zh~Na_BQ(CKXzr0?m{^L9(;X6v!5lJ+A6zHX58|HS
z{QMuF{Lx=k!S{4s<G3P1D8=wxRI`_Q#90p#-}rb?LbbT)i=jVjH&=*r(vCS{d_}$z
zGXv%s-@0$|<gIotQRg+9=5+}B-4w`AV8*$sqLrpVL~7>2rzcZCM*UR~tUD=w#PN+k
zZJ*tIGC5C*iGMQE^mEYIRV<B^TjGYzP4>H6Mn7-5jxq-c;|p6=&h+l*wj?~EXaNUu
z1bOIe)v?5ZNgY5yB<?;c-gu^8#c<^9WKB8Ctnjty#GqUDO2zy}Z=mnoswQ4YFu+eh
zEYs38)28ioAIXDXQQ_MBr2HR;n-RaZLV_~?E`0|?kFW6Ywf5+hi9yU2ZxokLvj0bo
zbYO*U-&20V++9}D;_{stj5OMW+{bVaf3ngb*e%|;xXS6|1Tzhw#bUY;L1;7Y=x8u6
z-<d%MDF2fcKpVKA6%a?OmPw$3LJDguo}S0PaDok>hOpw-#A`esjOnoBvkc}@=CoPr
ze{SQd^24G*7x_8_9)^_`_D`<U(;oO2&GgCvIqk8*?!r(m1r~W$^~!^|*o`XVTW5SW
zfRZ5zF`aq3KJHI*aNClV(Vj$l+Ua3LM!5$I4Ipyoq~+xrVT5@E?q~_?1fhdTYbcd?
zL`B6Gc4gbXoMNCk;yiM|!3f+dV*rAVp}X-<vW!yMMu>#o2z04vB}^1_K!hinpoI81
zd^uSMo|>V-FEemu1gE@IR-YqmE@JH_4EYb}Kpq(7!)yRt0X@Rvji35!FM*YI-95Ef
zb*6$MbT8rE;!${@{7AB9zhP@?qMqB&-u58VM+(_g{e})u6LE8ePgGR~-)^_2g2QNy
zr{cec5}6Y8D4|hNm7(uO8EG$KB2f+{$G1{BGI)pSPh~|q{Zf&|AmP$=^94m}+hh6<
zi4A>s`@CfO4L`y@0F|Qf>>**eVS3__=(oQ1fdSd6k*Rh8@F!(z9=~RnP5Zgzuvj=4
zUcHZG3`Y>4p0ZCxPqhzp%T9%W^}1zAi_K`Q7j;X=cJ?`Cud{005>xzR{25R82wLLZ
zgxtxFD308;-D26l_pZl!p-kcsyHFn)r|;jpVPDY1Ktd=XMyPU%aH^dIBRe(yYo4cO
zH${GY2ju2!w9&-4)PSirvmW9{`-g5fU1PJxd;Php8IpPHZr6pSy!5a0t_?Hx@vpP2
z?aE9>XKaG3TX53YmM97UlZ_ThS8$&EqwHg1A52;NYA30YoznZ&{=8atO6^zsK$Yy&
zec)HSK&k8$`LFqv^4)SfuLVtEyoFAV;e*Yv^W6vUr9f{uEN(_>_Z~MHv%yJeVgb`?
z`g_jrdY7&A*fJYIybhSn2X-P)BLzCTrMBV#LCv=}Y)HeFgss#)Ni-}-ahLuJg9zon
zseQfX;%NBnun8APKcxJiSSd9Hd|)s1%yy6Q8lo~daoV!41;zHT6{)}yA#^ppfjG{x
zggUNIibDR;+_LgHV?@SxrP~zJ`VNbLbQU{%`xFWCa9NBqbYn<&=T3$huT+M064s<4
zv1Rpak_30DPTn*{oxr94UBdsgkQ&6qvE#YmQ}U}<UoI0q+X^r;_h(`l3HuLF?#P_w
z?tMnle%V=~kllNZi(7szq{1ZI&q&D(oZBje$U5sN<$Dmsv5CO?k!~xxESC5z8g|xL
z6v+RRtbDiSib%XQ^+F>^m=k(NTONDkKhQWD7-(qu@&~u;*@8y~X~Wa|N8axei#D?`
zK?#YjL4a0RS59E|&qtp?=t{n@5BlDMvm$vGjW|=T99l>UjBku(IO8dch5ZLOf@bU%
zGX{pOg*7&;2j$?{X0F;5ti`mP(MA>)1gBHNe{0Me^R3N6lTB8O_sXU7(y3fF<~!wF
zHnM(#b>>^;0)dR?mRQcSXn+_T77P2Kh!qHrcY`pQLV(euhjD>8KOemb><qmi{mtra
z<99A6y20N551MUQ&(2H^o<EGbIMciI_FldS>j;5}Zp4ZblkH|{ztYVHG^x%-fAueZ
zT5`i+mzTiVltKX+wFPLry-r-gb!TPffIO)qwCp2^R{#i=ahc<YU~o*V+~;{;1^QvJ
z2Q1w<hA88tnpHXucN<wFC=ZSjnl=HYrGZz1Qf1w?J+3TW4enLI&4ZmF`}#(M=?n3B
z^%@G1+F|sq;^@A{e+gU9(Z6}w%nGOfrKZtCD*4E~kN@w~t)O8Ss#ELC+OGg&KZMi`
zjI<3<GF(rn!jJNO!I0baL@H&uko9$lck0M%;r2E}nQF=)D`dUitOUnD<?ICi#?Y&H
z3JjuP52ui}K>Hq!;r9(3979EN!3)2y`|GCKF^qbe+f?E)P=mHF=0eGDn@pGvOIAl%
zjohSuG`yiUnK<yc@uN=(un?jNZZbJI@c1~<uv8OS;lV^HQO8v6^CWq?JmR`L0{@Bl
zj|o!kr^J5BQ?qPk$x#?HL74wn^$pW@c*(loSH_LS_!3aB<P~gs6st)xZ1+UpD=xUN
z)l3lKKcUQ%oZ2)6{ndwSE<_X1WKwldB5tIlm=Y7hX~K5sVF~=`yB=OplgavlNAX18
z--;%u=X&mYTh$07H_QKXq!wGBTtru`3Yli1$zIphOGpf`e}~q2?8|fVosKKG$du3N
zBbJRylbxCz@>PD<=^kfXC*_FTx;usx1{0)MiFv>e+-Eu^I^-9HC(#p8C?x2r0crVc
zq$`esp7L5*t9W0Wzy3E@$^*&eW{BNO@un0~hM$intijK74lJ2?X%>P5Fo|_Km)MU+
zI0Xv0-4&5D^kX}qt7aJ5i@ZDgP&c^r7gRV|GLb+mzZzt&Rz=R7<dC{ijRxbnk*o;b
z)ctQjb_(x_-p;3I027+Y&5-|q+8*x+`9C>*_2M0nI%}$@f7f^d(Jb364WpK5Aqp_p
zX*Px?Uk#cz!5pf$lN0OY1XBySjhS4`|Big99DJ)V`o!l+G}qal4e%SvhBCy?R271m
zP~M8)Icm0_$<d>+GDiLLtuoBlkktQv??kx}IR0RgV&}2hNQ$D~(+udg8lRDg$l%`D
z;|P+2vv$jzq;W6rF&F-L85$Fh+r22mHZb2Q<I+Dr{6DqQJBHrRoWKfS7I{h!MUVLg
zeB@4^Iud_wcC+uGYu0ZVzr+YjgI9$h>}PKWJ2kJ2ELAIju`j)|WF}jJ{{*fMZx&$Q
z`g+Kf@@4S|KdoPe!!bu!hEKMiU6NwL!R971?0QaQq%yjaeWO3H;4B0*Pw)nv%P}Ea
z?5sNikQsefS@3Rnv6j3eRg2%|!QA!#TVLYgNH_odJ!bW+Yhl;mB&|lJ>z-~faaXt7
z2J7<v!U<jBhL*W(=i7Mz6s+5`21H!(C~FVjqM4}G!>&JiA9wZszTxaL@QCK@`gA*I
zb&}S--<r|s&mBt1hr!mAl3deWHm8zfym#ikcq)Jm%j2#xYaA(C4N0!hE@jVXMAu6<
zwfI=i;)?B<L!~I|D#`KwXEKrxGpzAmTgym}7fEMJ*@CU?blC&${fi~@j)g@p(@9@b
zcjbA22iGH*Bjg{xxvm`V5e@6&aZ?+N?O78sl;tYyS~#!!$?Fu97i!fZl6LVyWbVVC
zKj1<;5YuxaA|=n2+!d(zb2xLFmNMqX0<CAv<D(r)mZHQ3_d@53`pG|?B3YPnnGa>X
z^1%J5l~*v}Np}RSRZprehj4(D7y%{=yt-x}c_qKX`A_a(Fxdm+#SUvGA<%37Wc@K*
z<X`e2`rg(;Fk75u;?vPJ$O#%kX_Xt3OaSt_c@WhoC!N3I=pVf1$G%Lc#b=Xz3gSFa
zZ`9bvF9y`oEpzz5Y=!RNQ^y&QpHK$_uIbzeCe#^T^$gyWd6wyW8H56heR=Q988=hp
zD#c&5U#?qFpVlYBAR!XpB1GxzKM8+ou=vIs{o{(mu8F1VV}Dps4sNkr+0LOHu5k+_
z74-^w5~-+Bm$*riik8-+3X=E=g?I_&UA)J?Ehxqx#_2Y{zGBaDjoTvuJen2iCE|;B
zqht$&!H;5(Kna~!F4Z`^+N~3}cs1_r)+o$Lbf3JTp@exfYZ5CD;pI=yM!PN;FCr?c
zgg*Xe{mx({6#mLZ2U=0zTH8xP(c(@w+Xwc;>dR2Ack+#nNy$f;b(taCUvWar7?S^k
zwS}Pxc~xp((fd@gnn4Pf0SoJ&P_F(S{;RY7EY!s17%DZhg6M^bI<vN4x6{MG;T!7w
z+{<`T9#9LL$fo}{Vy+u$a+TlOlf7iC!_I#ytSiyu<Mk`QD3^}zgYMIyCDBEAc*jEl
z_x(Xbok11UwF^Ex)9rv?-dR+^J@wH?uLu22ELhC70%)SJb^A|=$hJeo;z&i2m{3;$
z>3@tQbbadA4LYPfpp|SU*B2O!-$F%;cUv~sc(7=kgkXM+>-5d*nC*EM#v7d8{DvVe
zAu5G3Z^sls9<du-ER=Vc^jL7S9@i!G9{Q$Nu*)zU56NO*-J4Oe+I?!rG=ngGZUK-J
zLoY0CK?h-gywBTU0`ir5nxV91cQml{FEKmL^HP7+`Z3K6yT_oBkX(}E&R4Et-lE6V
zJeYbxxpD6uS*@JhqJni~b-+VA5}KJT<wBLfQY(&x1M`5`{a7i{fp|f=B(I(FH&I*f
zf?8rL3yO}MVnHKY<FzhWBMt}_HwlOp^HHGU@$Ob(j8&FduRdmWkS2}*9?TiO!*;Rp
zH9s=c)AcH9pNA5CJvR7??dlb{Zc&yOf#0?((EKPn8fz86-#KS}Le1h-d2CRMJ;KK|
zh{kroz`}*pF|7dpiaCF;o&w(Q2lf+C<}#J2SY4xlqLDf4L~4Mgu<Dq3b%M^sT8Wy4
zuj1GM?304O`O&E-=WTzlt^z?o7jXcedYPTCwbFh)n3ovG+j2=3JLLtDTi;N+rdQ^F
z-%uBI&sh_I#}^+nXF}F9j}3IOU8y27w+I<OV=^zp(ZqO^D=XT?<Kg#%6)7bj1S@hQ
z6BVCHTXX>HcB2MTY&&7~^w|e!<?~w?M{#n2%7<h<Tb4z-B(EJMsw6u?6-5KPl|7Z0
zWjI*Evd`kYSmHMoM*@}W$sQlwI+5iF4YS#f8)eQh92<4=vY`N!=g8n$DpKZDy*G0|
z&I{v7GK3v_bVJ;fwMcs|Z~c_#m=0UiJ7B2w=D7gc@##qUBq`aLK4Wf4$#X1+JwbJ_
z#2+i}`v0)d6*3)xZqJgFk<!tT(ovGqk&%|K#%-{yb+LfG2az=tuPx3DH``o+`$&gx
zt#E{agPbEffHQrR%kE;CYyAV}rEy$p2zkGmaYz`pEc82g>a_r>izukCu$AuBDFqg2
z3%>aP1quO!jP0vrujL*01E0W^i@j#q-;2{pjWPdMV_z8-W!JTRD+1CY(xo6>(k&ta
zO2^O$GDwX`Dh-!}^nfTSARs;DAl*42-3)@HbW7)a7kEF<d*AQz{`mf!$F*avwWs!8
zd#!WNw(wl1wkh5ET|e==-u}kpB5m0bEye3U?(^Q(VIIzlZ6^1wyng)#GFXEbJ77qI
z8qdTI@0Q`c85?rz&QYUB<iCkX7FN%DWVXhhI`<|Fcs;Zd8WdkT%$x9o>wk2@VhjO<
z=PHjp^~3h{LkrM>8))BT^!uOefsTZ02QxQCGAC!gpnsdJw_Kxpf)k_>%7kefzx$)>
zV%hbp@lSp_MHTi0Me0t?M?4W(lO()wFcrdr(13H(%+Rz|vkIkSFKNU%$s8P`A91NQ
z?gzi!5mIC3Jrob~7=yJDIFdR_Tv!0cO*+PMXRkISHs`%M&-V)5!fChF$E%Hk)7@Mh
zC@GV^YWPV9$8C~+nUU>0pJZ&+og2PK={mgPJ6%>2wEx2G9)%e{(@{3Z;W0oK#_bCU
zN~hgzIK)_5P2M}275;FsEs@h$Tp6U7IZ26e=B~K*S<pCuYC|hAu<WJ2a$p@G*78Fr
zpSGo%pU&KP6;+&Ktxp|zD90Qq4AuV|NaUgdbzZnxb)ILN23tAYF_6@j>gUBg5qnQt
z8Ddlkm+U1qA;j|BG{mn#6*CcW4TA8hw|J2-udSGYuwA&ID*Ci&YtB&zrk@#DW>?AJ
zA)FuIq9jO1i9|gEz9<RiQ?#gZ9JVi+V@U|*%e5Ga(v3;cggNL(1-^CCcMcS`0zt>(
zh@miJG6myd%W{!iBo4~kc66;XJ(+Z@imWH(>L?y0lzF@D&Z2uS;<){lgUrGOf=y)g
z+~TN8@-~GT-d9c%t$MEqYJzSguZk%qkl0<?KF4YLtpM-|Zft6M_Y~?sK}kz5vnx_~
z=jWd<fdby#<Ka4V(7D(<|Bky6$8k(SQYGlmzh5;R+&|}<aNBR2n7dbT+$hzcN4IH+
zhk3JV2tO<)lNc_eEu>!N^d&p5p~G44Oz+O}bGoD9s{~1-5zdjPUmbX-(E6RE$6A)l
zSj8UbKbd>ouH|!j*cRz<oj=e!N!)>EC!lj6luQh7VN2_R_iw27uW|9u6mcMpbHvtj
zByx4+T{skN^{<$fu$#GUHz(3Ne~n>||I1z2LAP0eTd5}p@JkZv>9HVHGUxj~*h$*y
z75`VuL+y*Eh6s7xM+UgjZ$T)v10F@;#AbR8$pkM%J3PX}o!frjOvFCf$ht)c6Dg)M
zLK1aK+;Ro4p1C0U16_@a(K6=+NcXpN8tvEq7pT$|;}X)fII=IkgAy-yy{jj@f}F<2
zf#?P*zJqrsggJfx0^q95e)4XRkiFxb-yf*teG3k&SJn@!hhzGnZ{GKp$So|<<)(KK
z+-S2$)Hqj=A(NVnRdILMSF-=3uV*C*IhBU5>vj6dsNWT*bG}T5WmFnr*UG4Z4VQf@
z2IpL-UQWqV+hv`}oWQ))Z3w*$ZS#kq=E(p>!w?qM7YbV9vZLf}LMg*DHN+N0_Su;C
z{ZXA;e;Uw7*=<`QkC?*&L))_<=w;=NGPHGGTOc2gNTJP+B>4j;8{~jW*!1_w)XX*<
z0UjT=AFdNx7qq~A2372EEkoO<LvyZ<{VF6=?(_T~Y>>VxVMNY`w(KKNAU%PiK^w>X
z69sKcDVi#)4TpzE#QI7bT+Z-p)2*{Yv3;7uKT0_aZV3*Kn$XbDwto<MDYk0@eE>eC
zQ_xnGqBSzx0y%knvI(!udt*gY(59B6`@gnrsbzQ(rs~+<1&7f2B4-dP7jtRrC|Q6i
zvlI=mgK~VJ90d+LkBB~ON7CYflFd#HJ9urCGHCAv#mXZhVuOrQR(^Q<cuMs+^7Ogp
z@2I<Gm-9~ST^fFj$Zz|Ub|)?*e%=nl=nI~5vx^FizFd7b`cZQ%5hz*cEy}N(Z@74$
z2OHHC)3&kuQs0s(kzP2BKq(x);5z`!ag|7|=;(c4VasR*s&tY_1&@@71=2y!;_x9~
z@kerG2|BeeJbhS33*_l(`h&q)_1!12_d{a--^7BrtW}S<8Ym_tBMBZv4k@ao5{)@x
zMLGyC0DF{Bcw*&y$;u$Al_i=l)gl77y^?;*>`z1CR>74i86~Q{<Gk)XI!l0XoLEoQ
zul?p|OmlgqIY+GQo!;!r7AvthF$~zS8jUHd*Vjj=rSdwCI)8uVE71bh21a8Br^Dob
zt+g)SUV0KI=33oq(8tmn0ZJM!fxAykRqHWWe#Prv$GiYwFyTa%9&IcAENVf#?)0Bt
zpc)d)xe&i4vKoQghijEWkow*LUPlp7eOfKA>aom4Ao`0kG>Q26Z;3+TF(NkQII(un
zLK{mjz{?*63;C9=Y!Wu{z2{P&W3%)3wa{ox3WlpZ`#LxTM~OEfpAgC>dG9_YDuX)a
z-YnnFmWvakul}{~bpo7c>gzFIsy4(<B6|+Z(XHKPlZ~q##bOiv^H#(_%8w4sh07e&
z-dq|=Ix+rzqQ+@dCb}scI`FE@;Wnr~RXJ(DcIks*txtTX?pD~s@Z5en<e>9jmUK`O
z#><*T*GxFp%#3s591GcTo4bvvHb_`M>@9oqZ-~p$%X}mQEfSBbm?%-(0BDPYM)9CS
zli5)RkkPXq$1UD<ibJ~!mlu3WTI0ukC??MjP4y6$6$3AO8RD4-5i{&1Vi9z|+2ycc
zj_fhd>nV*m6fYND8ag?S-ZWBGYXt&B$Lnm4_}PwlOPX<=cHZ|GmI-^mGF;A^1Azp&
z>*G!OM>M<=Y!c<JxT1igbd2}98MhPGp#xWpcb5!&>u1|*sa9qw6xkEmFSAv<GX?zd
zcT1W&&U`q-@_++E-Y3zsvYe+%Qom9c31Ph;tfEr8adloe7;v5z_O&{l2;}hjcR7+O
z306{fKFA^TpFrM=@t*G0f1QZGj7k+DLbRnS5TXv1QgQvO0e-hzl3L>=n0b9yMj@yM
zGtP*uhMRvie-7;*_(~7?Ad5}?2LfeRNMPU{<Aj!4;~iaY;nz7GZu&9i`#-{sd8MI@
zQ}*%?h~ngFjW6e+Lk9^F<De~AYic0TYFDZNp%MUED3VLWQO_*dF+%zh02*Pk<W!Kb
zCZk(tsR+W=rc^t^sh~YvzU&gd*ft!7N3uBNFEwOwgdMl5nt9%N1KT#7r{_hjad-p4
zR=Ksirfr9@P;rXvl7^$rqbULICRqMc3)7#}uKBHo`7Iu{jS9ApNkz<Yr<4~~1d*E*
zw(T_0FRVCC&l4{{X$2&3(pq;~%t*cdY{AIBwHeppmZu|?qQf0`I;DkEA4!VdIjRWe
zAT7vO?ij|J+g&l~@9>Aea>nSHgnH&_`r`<F#p|(dq8`rtebU!XQ;^f{Zv3d0+_-5R
z;&Z~-<`m~Zz0@DLR{1_JxAyge*c#u&%0SP`<W5GH);iF;B5{-fVt{$qR(g_KaJT5r
zhGX1$&&Zt{>O0+B4;z#>H}nIrbI)Qv#U8pWE4Md8F4{5o>xVx^wB??q^tQ&~52ih&
zKN&>g(+nV)Vo1&=^q+W3O!;?93T4YVOWetoGnLrO1mRAGIaj%?jzlQ<JpeFcZ3hs9
zoQ?z}Q*KQH0;I}OODJg_tLH~7T!}$4nSDeOTwiJ(_np1vcDi{on0A*Qv~&e>^cEUn
z#PH?*xCqYSSqDx`6}n5}O*&{Tx){{6s(svWdQquoDp8s-f5%Y!xOKmB{>Gd1`8S-H
zNYjaE;<rgLif8)O18KDMN<)t1k2gO6K0$tOjN?rYf%a<HKpG3Z;b7V;`opyA^ss?@
zn-Rx@XT5AtxS~cN@qU8g`skTt3*(vXWS)>V0g|13l}`VMB+IW{mfSNwfU7?GB6=_T
zjX4bai6qM=hogFkqf)7Mp7yaGAfY?-kz<Qi`*zZxKZkVuAe_TH;p5)0TCgXeG^t&Z
z^y4#r*oiq;HU9z}s#LFSQq0k$j_1SK>wp|medeKbqCLf)^-_rC=&D+2TiY+?_F42y
z_6-@95snqm&gu)!{8w;Xy+96&_>ULSu7Mo()Y{ej78NPKvxGr0c>HfLEWgV^`An6g
zW!R>}Yn8k9nhL1d4bX<?7VoC-SQ`qE%Y+{$aokhqXo>F5uC-8+_lwgqn~pOl+PbUl
z@gS2$yMo^|qqEyw>JYP2sVAFt=cSTtmJn2lIqOU1FFR5SW)Tlj^4~^tu}rGcEzTD3
zxbrW1wRrH0*Og-hwwzu=Kd=cX)@#;1PPr;WUs<=joorN{YzAIWHa(wgIm8rXBX1}p
z3A>Qj|GSiqnn{biU>0G{Y_^%_Sg)zXlh<Dk89OFDNPBlK_7`S=8}TTN3mZ{@B$_~C
zNg@eqG{bMNH-WnqAG$`_mku}E9q$-v&HnIp^T`woXGX6!?7aQb#IG*eYa3kZ)R9iO
zz@B;?h{<^@we$Iy@&I$NN<53Z?SFL73r8kAC3LKw;%$y17R^9_Wob8`6cnC<TG|#o
z5<O}|D{-mK+*g|w>h0zefK{0`zw%<h5Supg&2u4Rlq<vk=k+DEdyz5fpt#&AmK?gn
z^ta@tBJ%mRBn&-9F9yPO^g}4t0lC`N3I|&7$9Q;a4Dqm0PU<tzi}^{0VLs!(c=-iu
zWLKu$$60Mtbm^DRCzVZ6=7kGk&8?-%5BSp`;7eyzNacqu^6Kg279BNredt9}Z^t>3
z>pArFjlNSb-h1@NB^fotrpRcnkRYy^CX72B<5qv4^8Ah|YkoV9Z7aZf$4s#N1y{M5
zq*5^fm-ehS2~=CMn9T9v8h&e>XE7x>If0WBHQi$l;`y5qOkw6?5tp9)0y-H#<pP+e
z{`EnTny!T^vIXk6@s^a~bdWS+r$x+<J0fj_Nu7=^GVU8rS&p3aAJ+xDK+N~Q-pg*~
zyN(S&0zY?~FK=G4GN7x{FRwCC1hTv<p5%IeMPyO1h-zpF*Jyh<v7J>Wh(ADLzR7-8
zWOIi1%Zvgi;klO7rnD6JhBcdaZ@oFK(W&y}7|xaqy$~YzIWx&WCG7hdb#{SH;Ve&}
zFar*qmQ4?p^8F~wjnr%QOaA~~O$<<MgjZ8m=y~z2ld_&yB=gfxiSdF@Kf{Q~G&iiL
z&jfp{7AxXtWx{IEIMN%$cfJ{s)d-h+iV7h;FXr2UvqZEPe{{8xK}}4V(L{|-sS#Vv
zVRrMD7YD^C+77-kir&Ox=pmRW$p6{-1B6v{LBu8G|HNL_IgElFbpnb|PxAhCyPnO_
z_09C&I}@;{>xRbry|-vJn(bKDb^i$ou3ToWuQ|{98fYWa;(zY)@a_ed8`0MY=bu72
zz4oV3Oh3j8l7E5_^7AvgpzY_$7-&iKR#|_ZUEOK~qiIn!dG6^$q5M-ALSKj>QLoT$
z0a%T&N;UpD=H(O8y|m`s_)7sdmW8{Oi5ms04YD2@f*=o|+_g89N!z?EO|8|)8)kwE
zscJ%T5hZO*;6`$-n8C5V6N1Dj+1g)uyT#Ob#q>tn@`d*(A~0||hgXy?=6o(0p+A1>
zXHc=A{gZ=*4>Ciwr^&;*w|Fb!+JOQsIu&xg?D;N2;>^&guAK!<IyLf?3u7<$8|_hY
zKA<xVH<ka-Hz?`ch(I*;zfR5)<wGkt_%#Zvype(-K00%qQM28;qc%_2J4e?$f1S6N
zZ+&<A8rU@1j`hc-k!ll^Y9E_rKJxIwGC|FoGy@h_0SMZ{c^tbr3cGpY%{hk6`KB`k
z8N06J<>rdHiS3Oe#iRD48xKSJT8G?lutoR|eetltKN<c5(#icpAr|^}aKo3{ZLZ#)
zHw**+IPOjSJQ1u96`muGHXFi680JW!n}YL8wLc>ir|8EL?zom2W$vnv0ZR>7GJcG0
zRy*cYsK|M??N3J>;0tL4s~=)`^k<fm^?hTC^;FQ}Z0kv6F2#6~=m2`#W?~Vu$j`-!
zs_1mqbxks#2)xtgRuT2HX9y}E3clnzM`+$w&(<8~yuw^;dDp>3F_Pn!0Dt5D>&cxu
ziRH}zZ-?fGVrJQi#Y2E9dWCiUBN_NJcWDUm@0fc~o*im>e7Hln{)Wt@HH2Oa$8$WO
z&X$QpRy8{PuPD`E+s7}-m=bS|2Y9iqYLH0&3`n+XRucu41+&>*YE(@MNVaM26*9B>
ztg(f+{{;leq1GtKjuGG7%8wD&br_jT8QvBt*}*$J&qg3X`b9It12dBNVq6t87VG-8
zKlNbs93F30D8`Gq{Om6e;T`mZaqWvZFM;2zV*Jww{y6t!+_?X2!d`I1nu+Va03#px
z0*t?b<1u8+h#QG!idF7!bVVCI=xWQO!$BAOH9CSG9r7Fv_eTfq=R#fdY>TK^0JMMI
zws_4@=G@SlcS6&>=(#^GcHn%+y-0!RbZ`x5KBGMgI-GAJ$a%y<Jagl-W-_PVg*&%i
z<fZo|EEb|I1{V|ya~6vvs0{w+?C)Kg(aAGSL*heL;fU+%NK)Vysu*9?)@0&%sDRCo
z(dq4Kj-?K46B-WEPJ70KzDyI6<N)?$s~KW&I)0?3<DrwO=j}nBp(pDh{8f+HW9$Xb
zUaG4*Lh7x==<^_={*P1S(7n(@&Zk9wd;bfR8Y<XzOuTPT48cu)E|0)Xp0CAd7BTH3
z<ckdVu7Fc+s~+NRZv@yrQuc`a@f<pPuJOSHdqw9GM*K~rmq+}k+<HfDITg)#MrL+e
zHIUJcWSP8P)d@2xSR09(LP3t9&$z$dtrOTvjzoIPd8r%LAqMj}{{>t*Q_8u&ja=Gc
zCxdx>b^n69brq>Bb%5bu9!nhvs&ypWw-AH0I+|f7Ee|sT-s|5;Ss1#_EuVpr((xI@
z#i$qtE;#%)HVnbe{mqA(n<CN=8_E6ciBb02lFo0sa<f~0)!T5DHx+fh4D2tZUDH}1
z!9Z;-*j$=azH&s)L5>MzNDpZ)xHkUbBLJ7Kff)X76l=rqkmE0n#%Ix)ou)S7U?`Iu
zq3Lfp+*EWn{7ykAjhe^$J;{w+I`IE`muC1n-dv+}B+u4bpfs9u<ZIeqTM2#G-seaI
z?!QGH2_hYMbS|c<^v6;%SFR?M7R8<6z~v@Esl(hh80+h<_%$ALr;6^dBQijpM`bhC
zJq?$S>AXGnw>ls-su4!18OHW-=cTqzizteo^955yc-52ut7}(nFrHcF$>VK4-OSF0
ziy!l=AMmCx>*Gau%@xYK31kon69SgbUA63iGd(S_?nr4Jo=P&>ek`&o3zMf9`EF4z
zXXi=_la9%7+XTv8SzaSD;8jjz*wNFa_reW^D6iMlyJZ&L;NPN{8ll!-B~VsD$`;S{
zRY^>&=faNi+1$RBTA19NqO(h&^b=J!T9ki{_;HW`ws-IAp0FzD4f|7iyf7>e&;235
zx7XPEiW=;%f{NwWF+xq+Qfn7mm?SCboHhuSyAnZ#){Ib)*VKS^nT1MMEI$vZ?QDQ0
znzBMmM}#&{Isj3taqC-=Mfa;{cP=+F28$_$Np)=<d4tI`TuVWO1U28%H^>M^Z1dLH
zgjj+ubGKeJf0gFq#vtVFK?RhQ|1&X%xk-AsJ6wQOs<U?A=LK<kICX@8y?Kd!Q(`Ln
zuFGpg^@Set8o<M^u7x60q!y&Jm_p?Mb0V3Viv_XqYwWwa+l%))<6GFxLDAe$0ejah
z1TKlWr_T*-8Vi-gXnuPQl!{zJ#bc2wO%z0)TASQLMJJkH6clHO4`<&!a6@?1GmB*6
zvI=y@SFxMheGUH}EMT8$KbO`$5eV6bn})v6!~h}}+!_Sms4Y|!qWP&c7hj(-KI^bo
z&qmmvyv$D{Tx9L6)xjB}wCH{q&3~$~SWr1_qO8M+@#4;>Wo7S-f5~qC5mXl};_|&F
zca<y!zK~%r@KtX72$J%m-A9tqwQu3b1TtfQFxlPRS-{V4<!m+O-!dZ`H8()eOx_|W
z0-6kU#MeiiFqCyOx|bVOkqq#Lp9tad7tAi{RHqLZ$0IS;+**Fm#exYVUI2)b@yJ%D
zAcw!&9q1OsJ|m#rDRSd<B{@&XVVX~M8h)E2pPsShT_>_`#T(8<@cHml*gQ6NcBHYo
zB{TCKBD~*)>@Y}rtn$X2h*Eomu|C2RD4Mz59hqDgQe&;GgV|kzMj2H`;6CtlIWR$Z
zR0wQ(gR>XoG+Jrg%5Af+=)Ilvvym`5ESX_40vXkjl;e>jCp?PGRUY8B>1U=~EAd|w
z%+n0B``oG{EdI#gO3zr?4rzf^gG7Y#{L=FHyQ3OigtXSicr>D`o_{k2)~5^HtXC|R
z_AD0E5#^4MzyFhY<jeU7Sl<-H8^*L5V-d-)$F3EbQlW^D4$|5maYc4$Mt>!o+a!Ih
z&Z{$h`j}hmolUhT>~6&u1mrx*KE(4?-A5#O(_wIVHaSslwYM0BO(4VG(=t$=P3!Gg
zNaf-EUfERXZ0&uv3#QHOb|)axaF6||`q_9suJr3$uOs(7Gp3n0833ij)|5ekt*Rsh
zW<1+myL!MpXoJ)>@<dE$=E(Q89o`Ge0?E>hCA=?>c`Ts}+%;}}Es%3*)6=wc&Pc8O
zaFVSc%VDS`$9?H^Qb8Ji5I7&0N2Va#`#QWV%@7y4=E%bO?#fYfK%qm>;{o##byHvK
z?x(G86y$uj4T*Abn~q1S_gS}gn>_VUAhBHkE?A?z-+=Mgmm{%+SSVmyW(cR_3TEb>
zjL||#Pxr!OyO*278M<%!puR{k*HUA?#3vNB{c492@gxicC*(W=ml;dQ3x!3&iw6F5
zLcyuOEbmJ*pHtTW=iv}5A*jNM;M^C|!)|%@mT+%?b@1aw_?}gU4A^6(?<oMb4yHt1
zgjZU%N|P~X%ZGnW&Q%{7lA-Q1k+o#_+7uLWq)G4Pa+JL%`<Tw|c|e=KUuLk^(lqnM
z-nqV*0Pq<`Q!iTo=EY`{r+yRyb6~zxH{wvd*(8zQ4&c}ZG|e;Kxz|gi{|Tnc9Wnyt
z&L~5N#1}b<(rtKxo1dA%65I*xp6VW?A-~HJ28CZi%;Yb}-eyK8%ix&fqK+7M;t_GU
z<N5loop|(NDjhZ?^|nrq0uf!DmxB~YI-oGhm6vt`_dO4wTmD00SnWcL#&UoAo4n}&
z%Lv_Z75cY$QqVk{gwEb;o9fbAAjQ4d0uvfckLy3e@EaHtJp$N*TW$-878&J+Oe(2v
z>YgYpvRZQqM;YZ-jVr05z_OqH{c}rMg!mds`JgC&#2ON^o68u+NB;eeZKkC&mp9qa
zji)DYYt>)6Czf0{FqV=ok0uu_0WNRmU!uG4N)f7wPxTygN<Cw`u=;YX-Lja^sRhF~
zUT0d`ae04w59uk@g=LiuisEYZY@8!w<~71ZodiXVRuYd?cnU<3azc-#?4KND35-Zd
zQLafuHb>iKW*KpLJ9CamVU#9GgR(MDSc$Kj`c10Tf@_7z^LvTP1$#Zm0A%Q52>H2N
zHQ`ddmS6=vZ|Inbn|**4tV~NgFXppb!5%zAHCh+;ciC{mVc`0cm#$hpUD7)iHylJq
z&Q|k|oW$vRc$|LfNBP&t7t(TpPXjdhZ4$Qf#x~{jlHf=9m5Ezb*xMU9u#bKR*T)Hh
zh0!JQgi^l(jJ<W=+)C8$IW!3!Ie0;!G)rX{WnL+-IQz&h3ad)KkdzD2dN$oHvlnUp
ze5{z9{xK1Z)GBZ|{hDK;GMTZCE+hsP=n4I4Ne+kW&I|R<aQDtiF@=T&e-h)J$1j|t
z`Qbp%tHb4#YZ#?G5^9Xl{iRj*RnZ}%dKqtX{w3lVs7eW0ntNQ%G*(UChMcnckV#5L
z?8?%r;}iCRy{nK#AHT>s?orK+ewKHIPq$?e#|tUpL{A1uw0kfoF-KZb&lB%9753bE
z6qiwR|0l}|@Ao$as=^KD1Ny4l&uIo0Y!i8{Cf<B6;lkp0^Q>6e@A>WZk1yhG)FfpL
zm4WZv)I5NWepHiGH>87o?Oz@4qgk2?EpH<7m`1#gA+`u?*Z@ljZ`9k6eC3zVnbvL_
z#C23Z8NI6d*^&U+RrEC126g+4$p!!E@lhr2W%d2}zm{hmzlSSZ8Vk&Sq>kj`-^%`s
z*-1w}y|I6HgrU1F4sa=U+zFp(Le(YBJJH5XRK7&92=XTUEfe9(=t6k5ePrTxj+ze$
z*2!J8ua_3#OWzskW!?S}NdPm}nD?Wh9Ii}7nIp14LTbKB>wcQ3tZeU^QR;S#f=foh
zCi5hDa~x&i|Llt`d{xzDZp;A8Gtu^C{@7QXA9y0-)4WM3Hy~mD%J?N%O*vlKfsz;D
zEddpU*-G&qOMQY>#I<HUS>*O@nJ2kh;aCN4e|GO`)RU-`Qv#=Jw<($<1bz5MC=T1`
zN$1ZYr*OtdKD~R3P>;*3tJ>mt8AF87`j6n}=m_#$kwZ$*|FC+!?iAZu*n<v?5exsx
zu7Gy{W`)eaQ2$*4u>dh1m?EOUfI3n$l-WV8yHHFS#+Kw#ns*htQhW46=RGT(dYP)E
zx%Nnod~f>~w=A{P@wb1X|2S5T#Jv_j$g}yMKXyh)pD&=7-NmTw_5PlEHfNm>uWP7*
zSCb0HQ=5}H;0?7edU?MQtb6!nyW!-zpxU-}tE3&v_oI3rbtIGiKHEI3>O&pQz4K$t
z(=JW^x^d@qwzZ`E1ji4x0T-0){Q(!HGG_zC`6(Ne%aJ$Yk!14X*oq8U^5E7T>+DC8
z*grV>)qGv>OSMyzuW$UPs!0uiFW-S!5$_)RoL+te6VyR>Z~Opt8n|?p2&zTq$vzzM
zSt{3;t55dZdZS>FE!Zi+@^d^~j88x<>?152BAE;yU1{6AW1dYV$(JfJLj0yp0ehO`
zs8_5&Gd8(oJ=!9hThd~jqgHM9?bLf~$*p<c7Do|bo&=lkZ$go~Pyj)Hw&q6o%oWjB
zDwv!_WcT2;gTKwOT6?X4$+``{T9~6@_z*OSV!o6ro4Ld)_dqcncPMIg7jnvmBhIzg
ziW4AOhkZM#+*FOg-2=C_b?ACB#d7JRY-|ahej8sk0(%c_#Cc{)c>9N~Qm^a<NuFt&
zn)A{-S(y?%B{uie34k|0>Lp&ACYMX9OEx(gZVw%Ft;ppb^c0|-k|gzE_Tg5rr6h{R
z(wS^|$>+&1V53D}e{x8eA|mDK;Lyh}v3sQ%*==GxD4Wx*(sS1z7m}|<lVgudt+Nxg
zHtfANtZIU9%6n(ANb{`NJX2TpFuWA1w=e$GG1e>YEj^+J=(g!Hor^KO!X#6i?GX|1
zU50+ZjFkX)`=z&N8~7o6N&CD=%|+6~Qh2HboK!tdRT7K{w4M*e=-UyLLut#+ab2L4
zptA@@cP`li>}_5r--wU{SRFIBpp9{KVB}BR?Ojk-*|#749;OW7Md%t4Ex88(CL8_R
z@fScU(@y#)(aG~!<LO&2N$5|B;q>cZpm0~re-WZLe48Mp)uwdjI`7==9}XfTG29XF
z`L1Tx-dR?=t8gI0tOr7Jk5$BhL^CF*QJZSr?J0KPM?S{v>N8{RGdgvxpIu`+2#$^q
zeuV!50YZ)z;KL<;&<Lc+s@|0}=UOW!O)TzEWR=;|2s1)1tZBsD)1lV-zaf|S+1sZ&
zfg;(}i78?2BWd^OX|z74`S&NK@f`d(vTrjp<!t$|7j(x491i5gQ`tcL537I7UD*^M
zd<HxP)&esm?r)}j>-SG<$=#G)7n8#m$jFxSkjRje+sW3H`p`@i;_}ocaE@Fn>|Pnw
zp#ME-s@sPjDuY_3?~kOB<!+AiwVGRh{}4;aG1&i%&#^TJSI*<X%Q33Z=W;rM4D)zz
z%-O4VCSe30if=_3L1ruwr|7u3skUIhNu{?k$3WOl-A1r)`CQA6r@nuge$eImhz?+~
z4RjN+ZR5><`nE{Hyldia1rg5fkK)@wGt!=Ncn6J`D3L#5_M?oldy(8O+12U*Q+ZMo
z>7C24`rTgI!YAa2*<?VDXI0}WPSO(`Q89OeDN?&K*gSW$Muwqp23#z~3V|8U!nDh@
zHMngC9UsS%3+c;XT}vk9@Ei<t?%ayprs^G@au~O}`_iA^ifB4X*volzyT@Q&*a|+D
zB+TSI8r(e|N4mTVIb;T`)A{1eFH|vn%5%!OG^ygR-Yu?!51Hbp0w`0XPpfmk2G_gL
zxWdhU^%z`gLnQ&;)R-H0zBI*u!+G6Xf9H#4%r_jTUK%&?Qzt#d>0>KfP9O6v1+R}3
z!(GReE>O-*@(QzrZ&~Yk64Vv7t$Y@H0zSTB0~$RBEj(7Hx|!9cm}z7o1m2S*Y~nmB
z)@_h3WCfp15*~CKE$%jG5wL={BncBckDirX+Bd~N1}u^e$&0>0h$@o2#T#1GE?`oD
zAK-&Y!b;Ag>^+y_rijqA&)@hmVciS)plOs@FHMm6sa81Oe86e;t;^A%_Pc1t_+A+|
z<U5|1FrsORHsj74r>sBJ<@2AF@Y!cq7bS%X->}bb$VXn3pia7cZXHiKek{*6)2}xN
z;$qF;Yqu|@yxdfKhTz+k6iP`T**Z!>(cnFJb-yt`>aTA&hh09!0wd==aJReI{10A?
z_5%j#HyLPu+N4L$vYj;c+9frv;~pGQ)%Srdau^FZUi|qBWrQii7GWS|H;u`IS8h%D
zW?0_>Z+FAnBU8muc-=I_4_=8i=BF(H-*5=JX)t`JV3NDx!pl~v3K}5ek8erD885u~
zVzlv{0Tz}e{6@U+?H8j`L<aLWEa4|{VBx^O1k&KnEt07R4v&WI%7pK=ZwlADF9+_F
zo<IB7H%9i4{MWuM_IJ`yUA-2wixNhmtdnHm7Of61Gl8Xy2Cp1IY$+40XB_mf|DN0p
z%bu|);m{WcSX-owZjlf4K`X(UiG;ZE=XX!OK4mmk+dl<F8031wnc~52qd5WXXIMMv
z%UIRbmII_#%@iE_v*lha+oNFobDp(t)V0ssF-BhYJb!Xo<>mysnJ0lVk?%#*ZYCmW
zByRzk;~?Ko0rH0A;OiY+r0__}Q6#d?8y$&-4g}LE=^bGHH~pOoLZ(4L<bxc81$+Cw
zfAFZluvE6=b@JScA0ucl0z=J^zXOqAcI}A*oA&w7EOev1akE8IGtp!#&m@X|vhC{U
zRjQ22`Z5PY00)4X*YVA<$vndJ&IgG9LN(zOyMo>emR4!bO-;whP){MHUG5Q}``NR0
zh53TJzrDY#0^)6l88V#bnk$j3sAb)b*AP>MHfhB5^jBKvG9i?X5Ltz{=8!Rk7&FLr
zx*sF?S&m`2<=qv{XOW0TYe&S^=_!&tH)0j&&+!ZA9Lw2L%>1K_v%?zui3kEWU4T*g
z^D1V+7aeUFj%Wj=N5{FKX3uFyf;Hh(F&Ci32z2j$>!nGe+>6r-p>nnn&3$T}51Mr}
zptTuAY#Gkofv0Q62u#}YW)32JwLeE7oU`TM0JK;~&x(IZ<r#^j3k8#-%Ws3q=Fu+A
zkGY$n+)Y~CEl&B3N?@kUZJH_>@H{%ku!)r6=cV0E@6&4W^O{7GCTaxqpd-q1FJ!!=
zD8~H?^n<_%Gc@`Iqf5pKX0J?f0;{2xczusx!`BEm>MFkcTP*NA_rgCE-AGC&4^ZH0
z1pbpA_Zi$Y0t;-=^nL3`%ud#AcuB1sNhs>1w!ae<OgXfb?-n7LCCmecWxC#U>^~DH
z7+*dlE$RoeVa~=X`<Q7UVnZ55CM8dZxFxga#UFVH-bwAZ_fNpk<iEdh3Y2A-GM`Xp
z1nOMGOH38XopWi+4W%7v?Vph6n#!J#XZ!)?6XCA_?OrB;>+<;DwWizWo&60G!nP;n
zZbJJj(;y|c0r%(fn7@zoCGf1s5H}ri$pNgw>#M;9H%m<uPjoU(drs<9&Z`HCXn`vP
zRTL3k0_S<pErSNby*&8_hCNEH7T>rXD#O7dmU9pS&OdJI`CfztYeZ+6+MLAv4JYS=
zTX8-Od#B_X7+YA!1>+8t$4HUhdFXHv?Z1>`2v6o@h})NbiDZ*H)A0WR;PR2OX<CU$
z2U*n1(UG*bRO`Pttwxjrruox^oRtK#ui$?qaGiGzzF;JgpZ&VQ?o>;|P&J;#e**8r
z1w>;Kez$vM{6s8`-I<-OI`iV5g<8+QdVxZa7i#Q)Grj=Zxwu8VWHb!}CDapNAUnVr
z#-B~`e*muY6v<+M0NXCuolBhvSjt0M^dt-=Rf#>mlX$2roK5Ro&Ho<&y9z<oCY`gj
zTBePt616^yOUG+}6ZcLSio1khad`YC09P3AxyU2R=^!ZVB{j9M^yPHw`V-m3A;!}z
zFuWEErH(bn)`gPmwg?xvj=lh6hnldXy{&fLQ7An{IyjACUpQ6HbD|}|Ec}kA8JO^s
z5y=Pygzrpnp%~WN9@<_}hyOE~F$YO?YDCZse<h9)`#>`VVDzQzu91rcdkFnEueYN&
z0e<CKj8WvqZNf>4I<TYlbhlDLhve;K=NyIqr0QCx;jYFIFE9i)caJwCu<v;!e5%FU
z_)VS7FH*S1wICm7;`2qilODL@WTaNha&x6QY5r`#y`2<nM@*RaWT;Y)nl7~mfVX4!
zcFsI?=#835Vr~*w^mC55u|$X0D}Fo4`M%f^*g!fpc@#Rlz1BcB5hlZ~SN%mySKD`|
z(qsJ9mfP#z%pLLe5a)T8{+qZToQZLlP_9{l-*J(--Up%s)r}9ApeIRTE6%)SeN|r{
zE`^>XnYULtmO1<qSqgUSC7n+Km{H=2;m#3MOLX+T@V*9%vp5pyu1;{zs|dN?|FrP;
zia$tlZu03Jn4PRt5Ti?Sz9TMbl0tK_tO(UMoJ)1GGd>{}TB7sq6}L@tMvE;KJNBB*
zC$*q1hlVrIOZ`OB>ej|&rC0lsAZB+3ont3<J=ys$@uk0<fZl7ryevtfQMkFYd6)-w
z9B}{6svYarmSH!*{OlXZalA^e&B>7Yb>=rK*fvXK3|Af`nq%c5abEFH7nk%H6%!e0
zM)-M#6;m6{%>C{bv27<_Z{=M_`Q)p^Cq&!R$<uFm7L(B0);AY`%h^qETdVcUkq}pF
zehPDI0+0;%XO|P7OVFKngGrT$#D?|geZ?0aPGcalenR$nz;7t$8T|e173JcNv2IFL
zph!}SyYbbkX5938vN2&8y^TaK>ljz?wI1}&U8Y;0@B9*TclNg0XbyM&B5@>8-y!VV
zNQxDt_n3OK?7o|_G#zQz-5_XOLvaWLdQU&^`UovOgPqM5Rx%&hW%6gbPpK@s3#Kjg
z1bhKZv8PPG&@}kG<<3`SRp?8+A#D%ct9>@IyVc%^Rc%(dTU*t;yC~6@h$oGyw~KF)
zvLD^8HR#*5l<rIHkiuLn4x=w}(Y1c@pdi?i(7(Ht_kLw$*Ak~Mky^^W8Q87G>)U0O
z>Py@POIHO;vA)E8sY`8%Qud#BYi)XWv&8!ntH7$69)~Xg#sTIv&d`^NgMeD@*M9HW
z-CFZK?%Ul;2@r^1l*0JP!OmW~=Vys8KKX~zzb*9coSk-VUyKhQt)P=t{l16-J?>L4
zSKL1(E~(mm5v_8c`nC9<LXi8#`LW?ym^`vDHhvPcUJ-R@KVCp__nynTynZ-~eyD|h
z1i5~Qjy_2H#(9M#UL*C~P<T?`r!@~=kM35g%I+r`dYF=p;=35CH2=pA@sdwIb4%%p
z=LQ&nbZL-7a!Ct^tv`j!oEz;3b!vogN<QbVC8GrC-dCL0e-TGSjN6Hfo6aUW`clHJ
z92nv>!-Cx2DaNZ8=Sx4X3JKo1q=;<J;uU70`T%X)y@!RfRYXT7H8T^~)4WG6O%(+7
z(pui1Y1;YPoIHU3!s{D@_HjO*85nM1`m-%$@btB-ikc4-V;QLF3Gzcuxf+4eSI=su
z88;*~qG>P=p4@YyV3A{E0!iK3Ur4yo+0T1^kes{I=W}<oAMWzgdJOKeZ!;F`B5j?5
zq2DePjMe_&f^DPs*`>@{FARjD#;D<=3TI?75PPYap_EIj#;<hj5yQruU6G1vR|D6m
z#s`<zu(7Nwi33jd%Q0~M6DlVx#(BG349dGop2u#K*ixF*43<`8WC>4`M&?CqtfC7e
zHckc$pKj=pI{!d@8~hHANH4>M9u`T=@$^(%%HJ?vDc2Uz;$a<ug224HmTPm`5aL<(
zQcL+yRh?kz^ix&Lj5ad(M8!ip{?1{Bp$eic!;4KXGwTe1FAQG_?pq8dHfPMRk-yYp
zW*xB|+!W=Qeg=;p4u008Zl|uguY=^kP8bgUPH$PBD^El-^FDsBo<H~cCqsc1IMfL+
zg-Ps;px?!+z`kdksBBhzj&sw}Vb!tF5@jV2`~0~V1I^Fk{Dcs&A&Kuc+=#GwJ7=Ng
z`)oe}k^;hT+;_LfCjz;ApFq(2G^M~QFOp09shXc!UCx=6OKWKVqnmNv=i$6(b)WP1
zKI~Z2@aK|#`WtMfQ-;%kXZBK`rrcP|?6lVY0J+_76`04K^v)wCvE2MmtAe=?{{X)#
zgxgg{*n4kjK2}6kA(6pNmrZ*pn!~?Som)XkHHAHlj}l6%3oe%#S->diA+OHNO?z_P
zc`!SDNd;}SRiHIvm?yg;BZMsEu9%)F*t!Fk%&V__Tv9_r{Yg<0(=|b$??`e0Wh6G&
zgPn9|L(>Q1u)y3Kv6<*s3NYy={!8=HUA$)S#kw_Wc&`dWnppH$crKTI|Gs_oh@Wm!
z<Fh0`#>H09VI&vI%&M%cJ!=(DYL^#EzuaUakBx54hven1#5K^ZO`)C}w*cl+EiJ=)
zuP}Eh-2D{4y!6XZj6DJGZ`)hTqGD$xJp<XxPkU-EEWG%%Gy+KzB0IdEEHanSMQwGI
z4DSVPwKsb__=?N?H<Zx%F-GkLXDGIcm2e23uN1Xa`++)w?ZVgnQWb=(o^&ujW{6r1
zhA7UF^F31qcFL>(&R?YWB5%59Kn&zXzOsPrq`!kdC<aI|{aw#6KD;+#AGy(Xv33r7
zqO9=wZZzp}M2G8>SZ1PJ@pFg#dz(>iPx&K^-*GGQ-m^UTc-*0L6ZWK@`8WN*I_akx
zcq;>Ef7G{J_mOdQ$)4g38PEs`sezKmlTzm08gIZX`KS9YHQ1CsZC-9;hqLQ_?k53W
zqQV$5Tpj$izlkMge<ELCx(H;<frL5Q*o;LL@a7Of74IrLu2>1=;No*j1N{oJD~5j)
zLo3sSw(~I(J#4bxC`-li`Z>oz_999qL>^rGu>S1mq=>T@cT>WI0VgqNqi|7^?HL_;
z&SL{u{l$Cz!rKcu?+1oQN7AqH<f!$K6ty+Otsf`dC9i<lPJR0f=To1%BsuZU3aD7r
zZAuwx?~VcXwtHaaAz-Y}c{5}TjH%2@@MbJFOi;cmo&6dV?)ir|w>kTeVdfsgEdLj`
zaP}Vf@Y#+Sz~%B>Xq_5@V5@f>GN%)HJfu^ax%hoLSlgOXF3N)<Yw;o_bc;uJuaRq0
zY36YsleK<iU{!O)%*nJLfe!)ib>6+FsJC_4aTgTEYeo(Q_GyZ<9_4bP9*|D*dh9U_
zh4Hr0vL*-R8HI;#_TrZKEp@pTu&rMUTkkrhhfq7vR)?*3PT@hEvB-Fa(m#4onyZ-T
zvDdfX@Lyq?KRtCC-hVNMfj4XsLFBNFF-FTFO-Dil`(W+OHy8B3tTtPldo(yGaVU-P
zjF$&)kL^<!lNmdl+c#l1U|*UKSg)S)Tk{|&T;r)Mte6>`e%$wc_zWO0f_Bl2rknAu
z@KipjoVh*g_bQyA@PwywwPFS_?HA@rAU&qNn;6+%jXy{>=G(qob4}X8(cWy7=^H{|
z?c72)_1j>|e&OKmePN8pdfK<Bi=p9<rVOU5sC>}`1CDM)V?O?W;e(Z4#-hxTQ+^id
zx6TK7gx~mqvO?DwgOxrdK=%Yx6myYZY1Jl8D|vcW)0rX_qU!!BVFYfF+q>G7&(IBm
zQctJXfi>)+3&S<q?{6^M^Dq7Q-JM<{E}jW}onE>;mUTYg+O>D5&s=Q#Bs=mySk+PY
zaJJ-oSy?=s_-LrR5xP)o(G9>_^JlEEWoSJsAXb?YT$|&9xfU&<`f?rL9)HpKJdRc+
zf2k{L3BuwxffVw?d}$xu!Ea;QU7KzMDLR4U$CX?Ut8eEa@xg7t&aQ`5{7CTlQoYq3
zR3z&ai|{g({A8O%yK5Fa?dIgB9q;{1R~Y7ilkF%{R$EC$0Hn11{j;)d|2(V(h*0o0
z>$c9odF%?>Kz=;GQZcS4lH^n~ls<Ri*H`q^)j`B!GR3Gi@c1FIdvqEdHb?}ARV9VM
zbyIJbiTPbpb#_{f9IsPqq!`-y2#Yw}&URoR0$I5aI2RR*z*r;cz*H-iu?i0f=7Gpt
z;4<oXxp{WPyClAw5%xwZ0o<sW=mq^&>6ERpbMUkjGfD``O|!#ZcBUn=seg@#R=bd2
zm^N(Z_1Gf~^N$Rl_Tn?zA@;U5u;W-h6wVD9)+7FBkv2>5f$(PzD%2>KM+)WwjGvC<
zyW3wrWZmWWjd-oflk$U-R>zlr1&9psw58p9vHqY{kad&a5iF{dnkNkZUYUjyJq1)w
z%m*1iUBn~8@!a$AB_3wYgQ|vtksdeQ^FJ8YA10b>m(Xbk@>@`8$ACu1xlrqj=2m7!
zRG^Gqcv}?5y+}NFPzjq>SAP>ahpPR~OKu#`(4gTo{*}mQpnBks{&S~%p0YEvC-RyF
zmFLaZG@h$uk0y7-%Fa|2CqvT=(kaxn)@YcYK_v9GR_;E{OlOEmh&DvYFt}dDNFL`m
z!&=2YT=7B4<@KWseZFy*ms}lkK~hc>(}R5CKQp7Otr}0mZUTCW;VODe5zPcVJf&;W
zeCbF?9?#bjH`cBTo_4t0AYLa<qXrK@{T)6gkI7w7_u(s_4B$*^wn=I-_|G1Is5@@9
z=pl_!NC<Y}U4^B(h)>{o%*rsf$^^rQXXW}EAeZwRgIWE<7;nd|q2cOZ>hrouO3qAW
zK4XlhCF<BG;Q0gfL2r&SJ4s5{oJT46d(Rrw-MqWB*DhEbxA2!`QyBJEF5!Ev`lNTB
z)*3QR-wkUoRo(mK$tc_I!-PW)Ia5to!Fid!e$<JGoSiq#0NZZ;M(+L_Wsn=^jX#)8
z>diTl{nYzpFwU3iG-DbppK7m_6-iU-tZ2tP+y_=Zxu+=X**y6JRHqSHql&Mf!q_W$
z0VezZsh&ASXs?N$<>~9op4`a*t?#84iI>%*sSnE^rFt7)TCvhvo6ZaM)XCgR3Wt(f
zEZ2N6toiL&G`Pp1It?E_37LNW!BA$L%5_qu;HTAGZ0w%Rvy|AqTtGg>a4%PWT24x-
zVA?>cVbj$|VAL(i@KglFvqkIO_W+}yc|7a)=A_(4k5fja-=AcELC*Ww;Tep6QmP@Q
zNxh9pS$mBH;IMP&J9dzVoewm8ZaDmQIOoZwy~b!7YQ(vnw&;~3&RD2g0E|jBOs1S`
zPS)@yS5}JKar6Yy0z)BrfNPVKX~foRCAL6{b{_0w2W5#JIhNhSi++;U!}6?81PzjQ
zuhrS&DCR~Z-juArGMY+IGZCk)aLN4SDDik9sAlLpAIfCM%u(XlYT%Sac`z4I6lvr+
zm}`IJHTfe^&0EoG@<*H+5$5USj}$fU<CL}ntd-$u%lXl!H-N_}AF4n_;^&#S<C=zq
zcP^7%8%Fk$+ke+GTRct;uwA=eamgi~yUok&!)GjiwpC5!h~8C@_<8*OIQ5}8W82Es
zo)F>(zC*!dBNn<`n>+=gaIc}o?t${FkOs){F(NRWE&WpdX}fhoYDYx9wuj7qXh@{5
zs>j)G7f70rTKfC>e;x&|c~5|cBmSI?@RoYl;PQi^Nv!g<`u3fvjvp%VzhM__7*#!N
zNS(mNc^*iWlpCuvypDO`y>S1xUYA+xC%wN9Q8YfVEE;W+4=Dl<QmCw)B5o#{9rWrM
z&9<!PU7hvD1_Y1l1n|c4E)PzK0rMl5M<?Qc?u^{a>C9^pyFNzNATV^42a^6TkDQ6$
zz$k4O{r|(Nio`i>Hf4q&K|{$523HOHo;tpT-Am^;P8MdbFQ4q8G%$SUH)c+6+?*^d
U@r6W$1x3Zh@L5>o)fMpnAB6aiE&u=k

delta 2101458
zcmV)VK(D{Qnw|pgoSp)(UMH9FR~{#`g(r*we>EYXw}T}#aVv5G)--}RzH$u{O57Y3
z3+lx(%?fD23{qI?vL#Gs;7#usqRM%kcoV0~iWiV7Cf<?Sj+EI)tGFbbsU}j8kQ)50
z=Cl65;1@lZ(QsNeCQV9sxo6I^C56(fl}<~h$E*pDwRo%rT1~OT^O(iYIJ0>(Pz(r{
ze{GGz2$WzRq!lx~G@$j>W$m&KNkxQ$YH}T=N2}D}Zu@4aC?|r-CR8#(ReO3W(<$Z^
zNF>yp8Z~Vwz0rm$Cfvc;2c_*c6dFndvL;F@p+sBGztt?0v+%1onPf(gSy=V&nF%Il
z{-TxtzPjp_^XvpYX4M;<=f9aMrHL?Ye}U48DsO?il}ZFLiE^_ObNV<N_9xHXVz7%#
zq@X|)n(C8u?rwO8&@}LL+DE0NBmixdG7Fv@G0Fq-e&h}o-If-r2%@Sk-o@ibglq@V
zLLi{qqBT^e`Xog{PEUZR&L6X3)h3Q1vFBGKbwZx_dfDPQ?h<1_;KaVt!t9hne+)8(
z7ow@2neQQ3fa+q1GY?Am<jjpBhGK>tj>p(q<1YCdd*2BvZ1OZNcnYB?7il<q0}8I^
zX@Y7M1N9sK{HldhDBQ&^H)F;$X~eN`5o>oAy-tc1QZ~^nAHAY2(>RdaLTBxd*}G-*
z2PRE3(X5B~Qy!t${%UwJXt7Ymf41o_fJmvzPux{2;k1KAPo@2mvB?v(S3&Q%q0X3!
zSsCT2D$<C`+h&2Rt2D(`^5k>g-k)GJUvvHQAq{}Ht-)>D6esrF>I?ZfwP;SweHNZy
z&ZyVwZ@^e*JvnfbM+nN<<}3s;S9AiBPF%2@+Up{6MX1u{=osT1pZ}N_f69#f-wbH`
zhzSuBR{EGo=NX|34#VGptb)7E6`vbow5bN{>+N%ef$vAHLQ7GijH%GVRHLyPth+PH
z#sP={Q4Qb0__}}!#<<#Ms8ACPP(dXVRO0A{UnHQ6*Q}t|Jg%<X>$FNJy2oFeX{#o1
zXoxndnpy|s!Pm>+LLWdAf7<~ANwlt2OC`YY2#{BZy&qlBZcelZu((WSLbY_BSBn?R
zDfAov48@D+kL~pvV4XoDY7Nr~v6d>uq96U1l}NYo3SO`YI+*<8a0THL^)Xtb^H(OH
zCwN6}7Rx-EA8iqz_L0QogD+x>MN%!Q^Vs%aw6%k&tiGxryIKQ%e}VP($O&*xjH*@2
ze6IrQaspjbo<*b7K7cVsD6cgtYx?lzhCW&$#hZ_4xJU*QC!YDmX<0l4<2}XWQt8lY
zWm!3;j>x06YM)wtVGN;CS14u8AS{DI<;dvIo9G+Msf{9B*OTsWY|*3be&1veFeWO9
zQo{$&o5n82boHGYf8>3<avB|r@8{!b@8fM>+=N2HuvVmHlZQxQdsp(aT*#onE!URZ
zB%VXUb?@Tch1P;`!iB6pL`UZqHPUyL^;}C92q_4sNfw5fEy%+Fvw6KkcAy*%@9tH(
zmL3qu2!{mO@aXhFIbf&y`O*XXv_%PF2sffa3-d_s97n)be-xzc$}Y3BQEG>nn~NQJ
zjAjNWq*(ElKD%<-9jJYP+MaI`RvkswzCYBdfltIl?70l&+ouB~<t#cM&*F&e(|jiw
z5|QmJFx2^h^%ew~*@<H#ToTHg^hHm0y$CZ9t&dFJ(WBt^!YDuH|528PeLPnx2=H2#
zQna5J6-A%Te?M|gkg;l@t}V~(&7QnByWNoDh7nM;BBfja4?bw_hhk>-G=0*Lu7SLC
za<KzO>D#)b5lt<}-r-q`vz3$Y$lSZUjn35B43jceOlq}Caesb3#>xp-tvF_ObntM!
z-JXFoq(!WZy{*<(aI0z6tU4v%J9Z*{mG1Teiz%IhC$g5Sln$8oHk-^AXRheN|I>Mo
zO7QpsQI5aiDow=<`rZF3nzlQAlc_GGe{k%cbec9%P+2ScsZKg!vIHU9?kx3U#l7~*
zssP^TB3vQgcQ2Py>f}G>??Ue4#4>~SHC6$HRk29iH;?AEL$L{uWe&)#%>d5DC}+@l
z`3e_fpHfJs1<_fL`l*#y%5p^zwhL<Q^r=FE0P;}$($_^<7PF;H=%s?T)pfi4e*~SL
zd-X~0qfd$v!z-Xvp80kyCNaPYnTMKlB`@E5pA{&^>QeKpNr8Fi^?Tcxu9=1m4KZ0O
z__#0)gJT-FYHxgf)kG`@kv{8cb~h0jXhPHSI#w3qq6@I<sdv3EW3gQ>y_(xEM(rd&
zi!7U6GPO)dLbKNNv93tE+Lsl5e~f~;#ZJ3xhGQ{@&cs(3j;CG1h>QcNHA*fEMS{&|
zntQ@~iU6BoM7c`RV)TCBzk;KEHg{u;6P7iWkF~vM9T16cVUurbMd-J>!f|$M%gAAD
zV-AHv?Og7)o#58f{N+dJ_}shHjm~(x%RJNsjgq2e9%>{U;;J*%bkB#KKZLvGf@VQz
z--zKCzP3Tgh_IH@^0Lk!7Y`}GSAy`=jOz0=OgYP1H2(XID~9aya$(xVf2rjk-#=dd
z1KerMm+@&Q6SEpIzzLV}R~{#m4MYcjhVT6peu3t2sp0}O1lZ2>067MI3vx|Px#ay1
zp(IkIxOfxW_Du8OY1x*g;^^~V#i|lbcmM3@?#IjF<K^4CfSn>5nc};Tzjw^OLG8bc
z+7~766k|jRw)^;Z_vag>Tv95yeK7k__PGw*LJqH>CFOQMI8usc`0L~M7wS%bMhQo)
z`Hm8>4=qsa{<H7oum17lWwAnVqO4(30kP-sRxq#+VIOSV9$vWy?TkYlI2#SHhJoD;
z)wn&p3Z}+aVcZ(`MQR^*AfVj7?<-@y0bD2oQf3CQ;Z+JKY~j`KVn5#1pAdSy#!NX3
zj`OwK0+{54+PzAE;R2YD4x0CWk?t8B3!F#FzJtOccXZF~0Z^VsD>-#X;aSjvNnJLh
zg_JN})-JiyE;u;NvTwoOK{RbIh$*+1US1gi)3crdW=Puo@i=2Sy;DI^8=@&@rgn60
z{6qo~Zih%{GI{6%L@MZ$7?>O{I|g`AdFNMv=SnA-7}4=~CZ%cs8JC2Av$9g@0GS#X
z(l;kNI5-(vDC0~V*vcnrCq%+;pQbFaqEqoql*#RsF(6!)rOd2CEzn-R^g4F;0TzwL
zYz)elSC8j0OU&q4JQHSWJ7$y<L1huMeZXu#0K^{i^66q6V)6NzRX#U6u*WFEhc?hM
zDl@Lb^2x^&nl5+^uQO_Yn-Q~9Kq8P85jzCLa7N5$BtTt*7nqRKn4Bq{KYboGsfbXJ
z8QIaeN5rqU!=*G4rmSDbGF<O&`gm9XJI<`n^8mNk-l-i#E_0eDmN5b#21I5~zw8fi
zm$8aX(b8tfq$B`k1)H`lXq=4+UG9+?<FfCA@X$^9302Q5FeFreKz5=I$P*tg(X#z^
z%n%4@S;Y(jW<Io!)A(|g#aOw@GIZP!qfK_&e#xINJ%*FYTnpH@chsk<RtTKfJ!9U+
zuuo0;Da?SAd_jo&Cuu_;P*vp&xgu2Q^!R=taKk_<<v0J$5tN|9^+>iAzRanSpN;_%
zV><tjjRCMpeFB+(CP_L5^2%2nXZq5F3{mBZzM$3Srf->_B`213JnlD1#&(lrDv6Y3
z1&hjGuI#ncOpn3S<4np$`gNYE#GDSn{h``EIR*nnu_9AZHOvf3p7isJPBFI^J<X6y
z9Vch+m;6~W$Jk_24BGz1vSR6VFNj*#TBAaJMwHzMIb#`r6mF!^_moV8%FB+iGzx?Q
zNo<LVt*R0&(8jYyXG8v4$Q-sKqbO0v6p=9(gz5gmPNBMr(s|k?@bbN~%fFf!k02S>
z3~(IxkeTb8k6ac+4<(+@@CYzANMqT5!B?HkZtq0{)I<XmVG5Nij_drczp(S=-V%Mg
z*<NZ!cT10do$bjcvpoPwv@R=Es9Eu2UNlHL%QnK5n9*VQCEybF>WZ>*9rM->8WfDt
zxMQsZUB&6Gj^&Ocqj_W$2Tjhb)qHYsd&!W$qT7>Ay3c@fVpLg~*0m@fdM_?dr5AH*
z(AKAw$es=v%TC+1=9yCn(#gg4HZ)+2P!3hxlA4r%jub(y+Rj>J)>C6T0*`lx+f$#y
z7(%74B8EZH%%IfSRMmuA>f9ij=xe8kwG=2PkXzN+ctl}cHpznutq5bHN|;>v%s5b*
zvkY?=*~HE<6}DR=eAjoDn%1%S>@ZB-R7%f~Qb-uqMN6&2Fkh#k!8b0zw5r*pCiP2y
zzko}BxqzzySW*y9>sqA)Sm8#c@uc70H7oW=XhRgw4$Q^Hz}31hxG)5?Nsl~1?t0+H
zJT9Mla!;#ZGU$%lJydXeCPN@29O^1)t_a=YY{M#iiz`S;7#H<WaDR8z^7&E;VY%P9
zoULG-a3Ql{b$l*~)fypXMTMKX{_O)&!Zq7}K!785u;=1@m)S^9*wy7{aI~s<Go;|w
z7bS~l;L;7rFHj^@XFDR(llbq=zIjWeU`Rx=R0HNMkuX<3r+H4{jvKgW*e8=VsUe;2
z$0lu5lU{W}1%YI_2F&{*eQ(jypYCw-Js7{GejK)a&z68PF|R{%kAQM}hX^-}fGVqh
z)_N6?@n+K1O<Uy*UlbmKxlYyUkSQ2-H#4?F@j1;$x=CQkm9g<&S+Uf{4(>*(m73Ur
z`fSzGO=_~^^+Q=!Eqz1tb}M~K%<7l_^Ym$ZrIjNRnXJTs<>tuP_0@y&E4#K`xSpEP
zp?Pe$uA9_YIS3+?H5f2o=Hpjt)=r0i{B@eOb(8*nE*O!ftVhwZ9oH2hl7o&#h%234
zwHN_<*01#$(qaTVQH;=>2(Gh%cg1R%z|&3A51d(^psZ)1g>r03LLD#S5$+SHE#yIM
z%!DdCL3eaNqvLFlkgz-`T~@)Q-SXs(B_XX^n=g9j7qw}IIp=&NdV2PlnA(AV`Rv)F
zX;SfL2JL&SiWuC4*qJuVWTrXwIQeyJj$zH3T5jL1gqxb)5orU}_G}l^g6L#L3@kQP
zUfQ?2r4%%`RHim{S{_lRH%(GrOfh136)*WYcOfKa?T9>csopfZ34x&@M%KkFG+SO)
zrf<?EJ~v&argb3h57_qf7z0gzXgvZ8J(o+rgme#=!$O&6%&sg8`A<~VF}>k8S;RPl
zWg;D02kJaf+w&7lNJ6o$SY{Q|FYSqk`DcUt&!eRG7f<Y%XoX4Ls$ygm;W{_F!*Xj5
z)-=gJFb<?HOZ)t`mEhW9HwI}MY)OHo$@|C=;^DO8Tbb5DxxY$nUqXX_F-};rghmY5
zAp<PdnbH3+>z|nM&I*Us(k#&(>G8*!i&ip<&d#l(wQrNpqLi>??TnbJW1z}4+e{Fs
z^-4ccYyT6nCj<~FRW{b%648<*tqB)pdBS|D0}(J;5}r$GS@4iYw`lnmu5vq9aHsVi
z!s+*Y@Gbt~Fu$Wu-&OQWDj8Lmt1Xai&y*@jxH479;8`KVlaM8&>X1DaG6_WRx{^iN
z_@voo0uAEM`Bu@B|BaosWgV3J1@e0r$gjW2{_ppXm;V6>Z&pK-u{so!(L)uLKrSbf
z4MYup;czK#K*NBg83)KQ=v$C$a>^zDe+Wg96jh|<MYMYc!-KnR%a)4vW3g769{xGd
z!*6fH$J^JlfP*GRGR+Ske;=4TK;2(P-IGclG-E^wcKG=B@aH?FTu~~xyD@ju?y*_!
z3w8Vg#%SwaC++dCkKf;DxDX{=8p98ifV&xgA;sZ8ccgswkKf*s1cDRogi;!aJAagb
zfxAg}W8uHMD(8&9PXbU1P9XOzh0)#g^q7gk2`t(Cob=BfzX)dBJHWj|hNDu=N2Pl^
zKzKC^&#BdLqKMpy=03SXM0Wh51XS+HyN}9W{(Qo)>7M%**lGI8=Wk*UP>K`kUUdn7
zD4l@f5f2BH6b^}V2AEm!m~ipf1i<9j3NT}dp|TLU1NPuM5KL-6Op+4DFKZ{Wz$60-
zd;!UW!&<>29V}KCB6V1vd_G}lg6jBX2JCLIzMiGoT8Nb}#2?B)b4wr7{^HX*EQeW3
zd1OpLNCk8fR#kK)*Mw?QTqMpD`@nF2F4z@u-f%@YThAL6<_)+xV_=#Q9Ng*+Vf9{C
zZF6MmJs<Fq^Y^<4%_S#LO49Lu+4+z*w)?d)1Hu)rQY8~rVj?PNU&wT<U&<%DQMD#b
zX(BWf;&n7$vB;16v7(#^s;XG21uHfy{|JL$<j1JIqLnzIrJsNn60xK3Y_tY{zf$c*
zOIk{VWL32EvJQ^OgB@qoyatPyeamn=CCp~%X4Imd8!zJoKrDzttp4?Xf&-J;m5;%8
z&8C#v5MkGPO(&ajiu1sP%(tMw)Urf6K7yv&v?uWe;!t#Rdr~Sz0M_$J7xrX|5b3nS
zL%MfSU7MiPw?B+cpA3uF`G1{%Pc}U4lNFFjAYfD#GI<kZ@@CWXQQ$&I5{udac}_@j
zXag2-;*Qj~Qnzu%Zm~OCaJ81SWA!|&%zmr{B~r^OSGshE&CH4SVa8}cdNzhU{Bi?9
zkxx6H*^`*R%9FYo@y4Qe(Asz`4WMF%pb$%B$}A3x2`qRAB`{^Owa&bMfUl$_FX&7W
zu9ZUTRU(DccnkoK`xgWf`6CTbqKYV0*H88Z`bnxRkEbrLeTEakO(AaANlnf4CJ|t;
zKy=*=JS{}gKuNT^Z3vKztFQ^%?fm%!0}`oI^BMVtMszt%pk-ZQ4bQ|2Wr%%M3(SxG
z70l)m#;qn!i522^boSVP=coNRX-c%URh-5<aC+eyh(+!QwO#`oU?mMP&SI^`LfuW2
z6>~N7*sjs>MzuUDF`begxqi=0{KZ5Aov?gx*$gfq9!5v4jSDz5&m#aRV$52PnWZ)*
zFKg24a;Dj)znVwD7np`JdXih`5kLU3Vm)_Er7oqGRjboXDWRr+IPXuOm-=+NFo_-`
zuON+cVzsT3wj>jH2$rz-e2ESO4q+mxg-Cq^kwOztst}+9@go7Yzs?yLBaVwIVC$@c
zuWg2?iMIi}MdgJe0xUoW;&Y1##wNKcVJ)G`RI$n`R_xIW6RgI$v`N(}R7&Rst{M!<
zM%AI3MO<}Uu-F}co|cZw*d)9nj3$hUDq2}?VF-Af5-{GWU}9JR;p}}C7KwqE2hs8t
zYMxI>SIEx{QL^MyZ)OzmaDVNT!uXKyd(AZwS`sGL+R82^PHO4L@QEaSSGhbs`NXPt
zid!O-Ae^pehOL%2T^UXm^^u{I$#*lw)qmWIWMe714t;fhDEo233Zf(<988s{td{8%
zwA#4VwK2?N>e0uXLtALpJfP$T!}(^#2^O?N(%$KaSnaQ87Qzy4>xwFKIz;A3jEf4I
zN(F*d_?TvnzAYA`L-Ql}wh^=azO8U8us}`Wwsb;tR&9B%YNrR#LYM9+{77i+@8}gw
z5`o&oZQ+D}2zdl}&+-~PGp1rmlFl<C9=%iah~2Z_ox>uME3>%-B9S^G&jYAIfP@NX
zEZ22YR@D@&pQL^0($*tG!c4@fUOi_~QoGL?gB2{S5dNLREfKUUn^$0QJ6z9$Yk$cc
zw~T<UYo}9|pPKC}ACyp)roysEOlSKIjC0;LEso}Y>Yf8wtPas@VsB<n=X=QZ7Y=Z3
ziDY%nHI#h_ECVV}>qjl7%aW(8Twu)JtyUGiJogs)*}Dr{p<OI?hovcAZO~GhNLANc
zOR3XbDiE`svwbwTMQ!hV;_M?TptwlaGe1v6YuJyNltiH}eOoVkqQNgbXPPVR@3hxX
z4B1bAEKWz^^Rk~NA+Q@V5JaJV{Twn_Z5}p<MS~6Y9ot@U!AnhgYO%fi|8s3iF^{C(
z38|n!w6w)=I#l;V)xIyE3q};`%RkkVR(7eOV>AOSn(d+tN2H8sT5?x@W!!KqAbTYR
zhD9}LYpT{MFwQK|sKftMsmwddRb^_NM-Ya8c3w4~UHB-ksdgkvpEaL93-<^9xSM?%
zb@!jjj9AQejZ0A;oT3lop^EBvRnUUA`itI>E?<*%bL%|oHW|j*Ba@|1)&(^M)Im~t
zRQ_YUS5GYU4M|Ug?PqS&6(B4TAd3(9Go9u($S^D2vARxLvRyprpDC=@Zfnq>i?9fP
zU3YxGF4`YMW=0T$I`>ay=e|ijJ8_2Rbx!6crx{<WBl2@N&Cn#C%rqmGS2<f(-1!wD
zUvm`NicIZI-On{fNTANbXR&3QyERN&Vo|&QP__x-&9;Z=s(5(f_S-TmD}tq@Ehs{<
zZ13RhdA;exV8KihuC)`xz!18&7lzV*hsmR$t4|{&_=cN1@a%eX^0gPVvF-RASVz+K
zxzpML-5>O1LJ^8O(ucB(=tV>&+cxfXp1BlPB2joC9k%=J&iy;xG7hA!p41ay@+v6b
z5%2tj-%t=!u!adVw2s#XN+F@#04FwgI@EyDrFj|K%ji0(yo64)uN|F3pgaYCNAd>a
z8)_A>P#uNWp%u1oYEWqGzH7oz&-rlKb9oT0-M`-w3@IdRN2DoK50w;O^Hz)@9Bl+J
zE;*Q6_S-CXZ!zfAr%ks6Nnp|sUbXg0v?3gJqK|j&o&2)xN0Fp`9zCa4Q8tN=D{#sW
zs#wXgR8g-i!w1m1O;>N~X(VwH$E&rPUN4jJ|M!o#{{h})z<-y~96S}XF-<!Nf2X{j
zcWtA0DG*ykHXV_TJZqzU4s3SL_zbRQ^Gttwmj^CdmaXpPlZ{4`9=(5Lqk(*5(`je$
z4I#m*lXmQ*fnBd)q(N#L-;6mh(!gG>;G|W|bDCKPP8!&EI+)ZqW{?GWY(I3;KwkG-
zX&%!Zd1<tn_j+lCE+RXe*9Tr2f7o|ATKN*p38HA@us>s_eZfAi;inZ$c{}ves5R*d
zjv7?fn9v+JYNX$3v0yWAGllH<ltWLA^l}AP?J2B0a@9z`(`UhU-mPD<NWSs=jIZ`3
zecA4<DRS4wH-#ZUqpcfnS8&%lhMq629=L0yulxNqWOVZd-2;D(>>Cvie|m!m9`sm#
zgTB*&!$$gzk~c==+to9|LY$Km6wua<-za(ABHv-2ZRg9&5Wul)x@;7@rm^z<%O~=#
zd~an*zSElkTdto|nZ^^#b<5>TqmeJ+yU~c)?D`(#>1Cg9XY2oCI9_bJUm1=UFYkXe
z9QESvO~aAzj9=`k;s#c6e*>$yHC8@9<JOqDHD+#&nOkG#)|k09W{zx*Y>sS>Y>r$V
zxjJ%n<m$-Pk*gzDN3M=s9l1JkbL8g8&5@fUH%D%c+#I<%a&zS7$Zpz7cjWHK-I2Q^
zcSr7y>{hS1)hllGid((nR<F3#D{l3QTfK6zdiNTRbwpY}M}{L7PL^K5al}r>{|$8L
zID&qs?N&gCti>3<%CX}JcH5VwSG7r-k*Z;)KIb@oPSf=Yek0=;X~%w}ZuOaYv^=xL
zJYR%6^c(dlPya96o)+krVT?Z%lgd{XmwkI3CX)tO6MrDd9s@ZBa|?1!PPyd&51}aT
zq|}?*nMrhz-J^{x>DBhCD0=$yL{C3n#`l-6Zvsw=Xk?0?-v2l;dw|-18MQklo)lw5
z3U+$`>-5_zrCd@fxZRlDl>J+$-$I_hfR>co^UzQV)A{%JUtXvmj1rDo^AjatH!V<{
z{<cTjy??*{c!{$RoG3dnsesrq!YmlrZL*sknb}df2JI(@c;?Lh20JmZr=gmEpT7vE
z=Djfgn$JahDe~oic$+8de~0TZoN2FSa=v1;n^;ERegu@u=B{>kD?|=raH9y<tr+b5
zMGENr!kdrcf9`(mKB4C~%j|P!Xg9Sy>qlTEIe(${tRAdXgq5&E&jVkEi@F&u^WX7G
zch3kr<*!l82=h^R+lE&q&H=AS;H4jj7gEA_-)iXuFAt2_N;#O}S-L6F00JgPp21B8
zz&HpA4-;Ij4mQsnLkF)eWTYZOK{LezGEcc*1+DRBXnlP{z2&B*(8L|`?{OYQSVSmI
zUw;=5ebIU?V5G~L!TJ<u58*62E@0zt{;a{<sM)^Rv+2|Y9dbpe(#>S&$0G02xYg(-
zXdst_GgUqouYr&>PzF;Qjqe?Cf_rRs0QlyQN^iU;9}elhCfb<5&cB)Jp*hbn?Q$Q}
z&VX>qyTHN}kirrRIBil71)(yJ7`%zc6n`1G{i+Wz&57W;neqX!hp!OMx<*);$6-b7
zyP~oStIw3@RAL-Zd&F8j4lC<^h(Okbm3lI)T!0G6S|QEynk8`Ef>J&814@rTX?Pq;
zN)uuF7D^Wp6-Kw@U;NmBCqQ5&N8yGTZJHqu2z(GnSb=MN99&WofWDnFJ0#59oPSZw
zS7?7U5784Jvz0K{Q#`=)GoUp+4lM)%x^K6J<O0pv(AO-PEt!<)1}(r?C(8>-^1{ye
zkbQQKsB54_TNw7z^$an%!b*e#D)+iVwF`TUz==K5L)4gz6)bN*=jdk7*y1G;zL;?a
zP;&?Yhc03_DPVJ`qHUnYzb)g>#D9(u5p@mC(?y;TK%`W)qz9F-D~tpssIZHpDd8vt
z<})IO`-%-^p}}9F*NFo(L>tvi@US~Sk5*8N#n>YUWSXoHR<Mx8oU<7_{CgIm4~4H1
znT0aJGoBZ3M?E%uSd}G>Q^I1jgmROrs^)qCWLHSt6;HVWgeGW|6uklj7=K0-PkjRR
zr3)Lfqj8Xe+4;<SW1RuL<1rbebOE1!1%VMkSv{JQOFK?~@4^ZwLzFDzMu>$K?&Cv{
z6)hXR`zqE$F{B<Kh=RKpL%^mmnw8lVG{ulAVCbqTrG)C{exE=M4nwgL%G`&z24RR6
zezJ=&TtgVcWj>yO<RxpA6Mu_a^=trQKr}Nr5b-Y{%sZPQh8DX!8?K2-Z+Uh!#!)~V
zQ|=Dmz5PXxs^DH!a6&v_%Vb$Tt+KB5YDaWN<!qeXc)#3R&WKr`K}!pc$r*v?;s@_r
zv8)k-41(iTRerY@GDqgQ{pJ2XmA&BHstFI#gYNF!C=x4S`Dofo|9?F7F}&fJM^8mo
z4MiK)H4l|t^-$U6hiYFU?`PsR17`q@nEyk2lv$Vipj1Og7(u9EUFcv=4JywzyA-=5
zTrhIdMeWOlhb<xKb;oR1z>pJYXLop@pLRbV9|fez-Tj!5k}$)QA6rr+pELzN&?I|O
zcl>!jR%9WYzN#8A$^kZ$5nUG|<zwoBh6kGLwtdWOgt;yag=`b3F>oiX8E9I0)<oC+
zYp;_^T_S&P>Pj`h;1MkM*A=FTP^GS@R#ys<yj#3pEY3MW#+tznFZfY8?ZJD2*ZckH
zmi|ylfR*~AOP`}NqAo;mOSY5Kc95B6eOarjVTZeGLzlFQ8z3M{t<pDpAj}#l>V9Wf
zdq`~UA>p**hYZman<*cf@mNd5BjDQBo;h;Bs`-D^q%{Nq5R*@0bqqg48asvLcT-#H
zVFr(3vA?z`O0X=fnr`!dBi91<R<42~2%Sa6YAx@$pedXcWr5({L>P0`#t3G^I3a6U
z4lTMSYtv6*;p{awr7LkYBt7wTqa$@xMktDHt*CA~c(CqY2I3%zgmU(+3&{hrP68Uc
zcv*iS($s6&jB=&?p|7r*)q(G4`;dFVbwM}gjv-y@##lsPj(*h{MgNwy2&jv(Ry{xv
zW&aFjPA*j0-Zi#*5+(A>FK?5IGbGebqZ;eL_vi30<|(WoGic9cB`dfvgPc>pNHf-`
z#^r+y&M2!?(?V7D>cpzpc4I{~c9c@@ZYh757DQKVZ&R`qQ%R78!V2nM*Y>8xs0Run
zncXc0Q;e9h#cFCTMj=_|FK?4BERLwgJoNVCWou@a{ym2Fg-h(BD^;n&`t71=xF}}_
zl5LYvs|P!P^N|TP;|aC>i#QB4$;$V#{HlODGiqK_)gtszxCGeO%`b#=)c`?V_OE~4
zFd+%e5@@;lC4jhQ!*uKSP$u=60mm$vZwtfCou1&uLzh4;?Ekw<!0fU%jd38X3m+_f
zJI#3=Z*7Zgxx4ry!K#K}(NO?*0)qlkXeuENn|d3NhP&vb21smC(jJu38voh2f?L{w
ziEoUCq*}As!PDx}pvLS7zl0>~Yngx9%g{=PSD>nr=)l!&xEgm^M`YhWVKyV^;Y#gl
zqOSZHNmet8QGtr>gIi>GEy@9?YuJc!moM9BZC^p(I_V<Kh((Yott5-xjj9_v-TrC3
ztqzxJrUyv&%qgZ_S}T<#T=h*-TF=kl%5RN?CXnT8Vk;|96hTz={g1Trv$+wzDKV(w
zGqL{v@9!`F0Gh{ymto5p6O)Hv7?(@hEhdx2V8eg?ro%S;)u*O@;X~C5*dH=Rf$)Dd
zo4ViRa0$wy4VT~#6OWgHd7t*u?858)tCk%0{HIN7I1^rz$Y)5$qe||EcOZq2!EZLc
ztM+jy!=M=#la%{cqp`9_pFT@%yz`#6@;=xWaFGD1xC$4IaESy24=@BM$kc)GHyjZ!
z25vH6`9qIKXTZc1Fmpd(+Cofh6~N30nA>yAQZZwtYJf!sx&WLcpNMf0r~&f8;}v<#
z-jhFKE&;ofZekmMIdU0p(~}7WT9_(gZbVEEjeTg?RlQ-)ZT+-+>!m)1eUpmq49q%|
z<iI*!OsgUB`|=m2{AF%Xm+QjVYj4mbXRy&(7^_?u11@iVI}F7#;$pMXTouY?fP9}E
zHy6sp7POn+H?$DSil-Vsnhrxh5RUh&nx+iU7=JcRa~pVn3($-KL|FxG6wN9%lfiGn
z1YBr1qqfN!X9-|-X-rDvl#_HNaj&SN<ja&(ZNPlBFGEpr+>IC^psnY!iENewIuQsl
z>K8mrF*-M$3B;^$KGKk;+FwqmeeJ)m_S4PU<!bFmuVwX&F*BBa3_WH@V}3RRZIlw3
z8&gx}X-HRp>y5xXTE17lpbyXjh-rRZ+ZyK8dqp##F^lkx?LScSFQ(=Vs=u^Q7yxDq
z$GjM~f$?n3<zW*ilY{|d>e*{k!@LaC0C{|qso%_GLWjVzO4*nA<wd*($D_8K{lsp%
zU&o;Jzp0+hY`t2pI+QD%IdIl30=BRDk9J?x3`P-uvAfY?0qk5AEw_-!8K1#%N&EKP
z@?Tm3a~P9dw9)4xy9y#>kS2@iHE_0X>if<779<N6u?o{+Ru9wKu)%kEPaW69BEQmg
zm4Vt9{~TSH*<>{gq5*5`IqNuLW=9)h`gkvT!Iwm&0AUy7KWsH<<9&#@esj+cHG<Rg
zM91TQLRga?j?w(FG1;rKGTVG*fN&{%lQO>M^P!g(j^{TMFQ4~CmvYoFF;5>;DbMQp
z?RsBX1Z!MBUSG+Z7%(U$2!&OwXoeLfYd)GUJamg|#!E8#$^Z?PXL9;7XotH&)6Rh~
zwhG$l^eMOJGh`_R`(xVd%BK>eUcTbRlz|(6Fpp}+;AHLMHftBOWsp)=q2tf0<~5;H
z25L||zR<z69-B5<Dyx}5Qa!&-yub7E*omg@N8j%Q1+@&+(0HbTnxW&H^v(i0qd|#!
zew(<0?A5i~Wx4`wRl=lGVEkz1s7+TtyI#oD^*7mv*HpG92@2<_&yGuL^4dK&Ssb7i
z2KRYbSd*9B#Ko%VVaOHTU53Qzt+kVyWiA97pPK%cVT?Z%gWzR{;AH`a;AH~0;AH~|
zzL&A$9VnJ}4hxqYBm^XXvM9m81Bg{Ahrb3ne1ExoynOkFdT_!UaBBGYF#s({`U{dS
zS~EBRVG)LppTloorBs$a7y4)^5$WUT+|$vVUJ*Rq0dxi6^y)P5{pY;yPGG<0`>%8a
zUgKvRU$yZL$LEo<IXV2=s4GWzI;WWeocT9SzkmGlBC{A$1k0y?e3*ZG|9JV&%Umcc
zy-;2cfBbym1&tukE0R?>67}zY4L@G~F{e6)%mYCttu<x>8A!rW57q{Oyp$j_l0HgG
zO~PmrMt@7p3r&yJKY#x<%&z?Uz448f7=$@xB_}Y{yR{ZR%x}g88+754w5hLIQy137
z<Lm{p5%1@}NLH|a%3d%cXk73|ycuHBxWo^`7LRX<U&9x!k>Uby;{}qI3g9+B!IV_~
zeoYd`>yXx1db5VgH6vghi3&J?kQsr=l`#T^6YYdXR~5wa1fr@F|B(wi*QRiDLxDk%
z66e%z^84-dY9;SB!}@zqyOe$=uPEju|2oV58TrT%weu5~y(I)40kfCeB?K>jD<s92
zZ$Yj(CndGe2Ijw?xSnu)%e^UZLV;P}GO@L2g_teg=1;Z~Myl+|D67uj+2g-I>~&N?
z;le^sQm^8#f)`SU#od+b$dP=Q3k1VZE$E(S5$XYo3WaeCoaNi9T!>0Yzd*rU9+yOu
zA%%s7Dp&-Xhp6&CHw%1VT}#M+yJTUcx7<ExVQe?du0?U6BaDu#S%S6%Evk8z07?ne
zHBBfa8aO`wo~QFAP2)j;w@mZ5-{tTpE&f6`vnTv-y!B`O9Xb4a_$TKf9l~VSOxQUQ
zh{_$pa!?{rVvxu-5(Y!u8V{4$$eAT)g5f?{QfV-Ax_5Ie5Q^L>`454Asv3dPHQ>Fa
zfWYAbaU48Wjid=MF@*}vZW%h*7)!hNAZQHbSJsX5zY?Ps`IR2U>=@t1zC8MEHO^Z2
z=^A;@`8j@`q8Yqxgmkay9YnPepx_`#wI}yhlNG$1+NIz}aAYWfR^iH@jR5VIq=OJ@
z0ahJ^5K_uI$>}?}VTrzfA?I%}yQ^Abv^CDRt~tZ-1zeS@MyA0Zt{6H0OSZca?A@Sr
zz@}iXBpw8|qTD&u+Fav39MU?Byg)q1>=|_1!KhAk>*sR4luV@%)7kQMF0}ws0U6i)
zh+Mv(%m1&5dR$_1T&IMzS)zBzk3dL01e>{A-9TCE+wyT<+h_%UyS3G^0#HI}eUKGU
z25Klx8CgNkF=0_txMvkfWfQ}A+$wNXzMEuNm^BtT_<w77w<bComZS`C4$?5yZpPv5
zPEL*1vz4iTM4vipj6c(e*Rm?rR{VC$c^^MAotVcRf}f^Jk&1;$#ahhtGRlh1Te?sc
zYul<<v&6fmo*DChYpQ!A_BiT;%6xpQebh~Q*pF974=yE*$LPT;VMiUmytiWn2n&26
zV=EWf-s)KPFxxF#2Q!<p!ecOVAZ+Vx;tUn?dj+rJ(@Q>XXU^HjK2q_k*LhwQcY?21
zR(ev+PuK235Aof|^bj{ftIyQBk5brH-7{6C_7wN?F2aU?$&5x7yTl{Qh>`MKCNZYt
z%W9?e;PgaGjHWs7i&zoRetLSMWe`TSMpj3N*n8|X^4D1D!HW*bTQp~NJ{bm;w@4aV
zX<L%n3b-%BhkB)Lti*{jp0zJD+Yl0nv3a(*`fF{&F8OnGkApdXX6#J;ynt$hkkTYK
zvZaXTI%@!buOp*0?H#%J<zDHQ(27Z#D!4=^6}hZerbw!BNXX~c{C#5JI;>z_u5>Wj
zy8%#Zq1{m%fRc0Jo}}0L-bg9^mhG{rL+ZV-xWGZtG2p)bC>e`DXP)05MW6!TyU)ao
zhe>SYjNPd^D1AU_g#kMnBIE#nKl%E6vbA5T`)H|uKn0wAUMIHj`)s<EUr*M0P&-n2
zstbd^q1;hzk^<8<#qDOeg9A6}&C`Jh!ddb-6p<{}oT0QgieHR2?rN*Cv=r*A0H;^2
zKu^rR%%{3&?G~#87dP*NItVUHby(|SkBS5{9unLh3apz~fv%b1TCq#dsEt#X{UCI}
z<u*Ql;z8gV5Kh5l?Bd%nJT@NAi`I7@v5dwl%aUx>V{ae0?3cQOCb#^ZJ_t=M<FU9n
z-r4UhDn%qO37fSSoNhQmxb?n~&Y981F3f(>JCLdrwVxj3o=ms$%PnYgP}fL-7rxcG
zEy^j|rU^Q0aq?b?pE#_WNpztXQ{}s%gNG7-+@qEoM|k^j>EJC>;6G-|aVqXx;frmK
z(Dq0#6rhD^EKqqM6`RgBH=PZ~f^=P%SfSbrqUbilyc>@DFhfXb#UU_5pmI^QR|PY8
z+u+qczD+v{wUDQ+D!)yu611h~A0g#6*aqo-(L0d(n0g$8s5ex@`p1XQ<rU519(aX+
z?9@*)LX!qsz&-${pV)|c&)+v@j=PcQ0PIrwaSX!F2r0vk@xGbn+}wyh9^u_kbRf1V
zCpiXUtA#Z4N+u3&^-Xloc@@_x8O*3+2UXmN>Tc;fsOlj66KC(0-zU?p{9;N=4#{~y
zD*<h$KdS7{?Oj^^e!HX{m2?~H{{8YZbZ}BBoj7K<AP5*u#up`0b@AgY-%m0=+ng8W
zb+6#%G_=MSQXn5qenElHeIBF82i?+_G%N%pTSSDzMkIGj+(A;NMB*4EWuPtI{)fkm
z`t@cs<uIQ{Qa)glyCuGlWJvkKAwUK%)O0Z3s{ebvKBFP*>h<jwrh}0Z@>8?y2MKH&
z!~_uje(tn&ls!NG59vr$m$Anf6Sv1K1iJi}o%AFpf6ZD+uN=1#zR$1d+-x9^;=TZb
zL1sp81o$TB-3%WBIdTqqL986|_fy=8MRxa0zZDiR>TWKzek>M?k59i(eEPIIJ?uVQ
zl%5>(MmRY=e4PYzaQI(vXtZW>LNKeu^zb<Svgce{_?_UVIadmP4&FUH&EZZ74<iJ$
z2ywV`e_F)x$I|Z(!tTrXGdxA?i*sCTwP}Ql?}@TGIJIxorGt^q=}1AqY;ijL`tZ|^
z7c~St1$o{-?Ect|ikQHeG1HgFU2NA%Fy-as1JhPd-~Ty%-F-#^l`v;MzQO;W7ARV0
znU^k-{N-0Z{RW*s!C)2q^uILp5!vwR+w^x}e{hSiXFL&BVDoT|FrOp~J`}Qzgi(QU
z9iC2Nn`9<f7OtzzCoqc>a->Tk=LJBHTMPc6kOQbmqb6u)!TJih2m%ctFaQ#hz!d?J
z+jM{*0LH}kOYR@XM{uQYfFO|LfL?BZ#63JjnXdBi_^#s<`cxby;k~lD<azvZxU*ao
zf22kVCgS=2H3f}QW?@a4>dW*v1lVbl$)t8!zRyG<lf_DXsAwX&r0aFP;Is6~xc+?l
zit7&@Cp`mDfIAuIq57?=oYXC<H74u5kx4W`(!q~CLTAH@P(nfc!F-{zXsO?qwo?g5
zwv2PRKT?d}O`2y>^`vc)>Y|S4Rr6`hf4F&l)K|zKP^4R6DnK%o-La4|ZwaXKbsZPi
zF|~l2B|9s{c)%U)MXnFI{e^%!mvO4JyH@~{luRi3)@Z6B=T$;fCaGTmQTm+`1sj)e
z>RE`QzBlE0d^JSr@@;b4c2qTkCf@2b4ElC~o#?gX9?qE*`bF-khP1hFt09v(Rv#`J
z5;GQ@Gk(l>v<cn5`NLm-EK~II6bVyoKjkk!ZsO9mm#s|5jEy`<Xx++$PT9zX#I~(m
z=#-6I$fDlLg-+SXg<x|l7dj<#SxVl!Lr^^yydZy<dn^Sbe;D;?N<(*vx&iX38E^;x
zfTgqAiOAYmZ>R4t@WX$8|N6Ht{}2qO{WlJUb19sF1jzQ7KaB<R!su`1(=#n>P?KN*
z_GArMU`lepQJsekU#6Xw4wxHHTEE3sIP!WeC7GNI;9wv!AWcPtNDvz<yKl6y3e`9y
zjZljC6{(cDfAJIM*IDs1{^)ld3bXz-?Y>UC&&8(3TI-_Ey;KIC)_GVo3pRyS1}jof
zyVfQJn}ddkUX2nQ>`W`IR9|eA4D6!8R<O-D*dn4^3OCfbwh+B*6VR#@YEylV?JIT|
zbcGg#s9^?zs)p5WQV~ED>t+kL_b-oSUSC0}bnb*ye<|qP8GwegBGO7nc!*6)RiiSz
zj;85n%>7i^pZkRcp=yl)u=_j_ak5@_O|j3LjRC9GL2$lpjB&jaBftxH292;%I;%P;
z*rYKljMcJ~57(sDAMTV?B`4CSmT(U}Bn>yV_T)?&zjS=l`<8md6uM^F>FF9>p-!^q
z<|RQ%e+8MDBH#dKtr`>=h3b!l%h`JzV-^Z3F%Q!pbK-v~vMd`_S<m7U7-Weu$eAIN
z!q|Gwd<~ja7wMV-Yf4I5&?#uzb!VibOmjxj(UAks@v5N_F$G_>f>LMOl2`g>y;{cF
z?u~5aO%5T|uJcu|NDtv!yHW6D9$Blq8V_r^e@5KluT+iAk*oo8kV#fwDI{e|+KtlY
zIx|gwmf}cTu}g8J#e^J3{yp`>%1(brw)z_VJWz13I2LN=+C$C}9IpH#y*U+FYQHBA
z^VLaLEseCTVX!|T?c>4tJz2?XA&HMO`SHdK$e2ORVU`KSFE%X_S$uOT-nF<K%4uJF
zfAF@Vy$OtVh)lM5Uq=gBIW9q*YicDb46zg?(gDAYg~z!pSF0#GZ4CwuaYM9Ba#Z|g
z$rSM&o@DFyh~7N^Qe$zx1IhKY+ABZidGUy?6mnQtn~`fga4M&1=xbUOZ(s$+;d&yP
zCySmDZ!>%uSUB2tE7@0@5x=IRZomIff6HKnrw@X4lq%YhN@LP22U?I9OF^!S3rm4s
z%%wnU@WPAww^AdPLR3ri*SYF~JS(!2jP<DBY|w8QRLewXRu#ihBSvM1QTZgirO`7)
z$!my;cXe8YeZH3i4Hr?NAr);k(|$`0d&D=9#20%<zS@v%<bFVe;uuL9HGErJe>;gj
z^=;DaW6=#g7Z%fk)apsLWNT}0yMoFjA(81=<dr2RWO^yeXDKn#%Xy?nD(8&y#=H5v
zYK~m)n-rZ6Gwg(7mBdSy_lmbnp;`hlUn$kKo#d$Uie)^&A;>6<64XnMdMSO)@F;Hk
z?)HA@4IH!eC<!+=W(5XgX7bl;e>!IybS1a!<XrS@dB<v2q$FeuPnNCPHr}|m6j!ch
zMYA=Rr6;kEj2$s6O4fjQP1HC=Ou<x|&rd_l%k#?~lrR4Ikg?J&NVIJslW6JS5S}k|
z3$DVtE~&)OBG5EV$#X7Xi;u&f2(-LVz{ycTil*?c@>E<i8x4+h;H7?ae}DTBK`rxb
zN=u$QYO9B3ZJ6^4jy9ZyP#u1}*h62_gSDeP7LGWx0W#6=vE;Sb^Yf>5h#3uI28QGM
zOnYE*W_4b`zwI#Orb2h?rQ`$2jrY9*Tg?DZaPT{tp<~z<PFZYnPm)Ul_iC6$C1*H;
zG}<Y*6=&V(=nQ4*=14AKf7R+%{yh9~&a@QoX!wrdp9O!?Yo`7CS)t?iHt3TU4sq6O
zL|;F89ep)v=IzcHBlfvJhCCCIQU&I9ct>A!3=xs_BoNh;mSm|bX!QGGn|me_!F_?h
zClot|Z-hlIn88iwCKkKVpj6rIOL{8l0gsckmG}9^Gw~-}VA8qwe{(?3@Xrzvg@v>S
z^Vjz<L&xxCI84d~*k>fSOKG-H3ZA(8i-Hz$0+WHMYsH>an`-?dflQM`*JDi&?>RQt
z^Wv>8SN%;%tI&{`a<vMRy<Sg)I1m4FOAt+F$+f4a;l5z~93SH;NDJnL9W|7fwp^EF
zT3ND46;=xOVWe6sf59)27xv>I<V;@Iwm__P%39{Xa%LRBD?0Gv`K|d%U%etw@#GqB
zen-7KYptP8b6C+%^s{I8K%=d-aLh@+V(SsfL4$y}hsbg#0!o(RNH$_@5km>7w3H%n
zHCjGw*WR&gQRohZ+dQ5R4k`5IteklD5bq=lZd8lknq&cIe|DIJ%h1gWqD9)JvT0tR
zS1-ksWzwpr$*6P;vZG?xLrg0bx)7tVIv-%C-lGez@(Z3pcpZg2-T*9TVrG3>?Z`Dt
z6pxcyD}rVYj}DZjpissR6)*yvbGMS~Y>WwutHDDqt=m22wQ_!5{`CeDLrN-|db;(@
zN<Fj;o2hg)GZMTaXo=~Ii%EV@vVTQ3MMp0B+>9q8o)(pcijL`cXtJTr<;ul)C-<{U
zKyn%Qe_qN+m%;ln6Svtj1wmVv>WeNWf4y5tlj}APzW1-tA9xHw5?rdfT-EJvzpC73
zrf;d7GLu~99VVCj{eb%-$&zJz4&Edh1U`Tu7K%@QOnmzO^8WSm(?zPuFpFTs^!57$
zbc547aJr~*G5}0Vn7)2Yzr1oTG`-&Gr8t+8UWUF~`oz;W36@3x{Q@|BGYb6pe+#|N
zbRXil@V_!PP(XSKy7Lz9NcWmXae8SQsZFFy*MKg<+v(S@pI>+q0q7H&&(E)y|GkVt
zkZ?>0oPPW9;%_JkOj<F47R<Qm&;Lxnzx<<6$doy={u{l2??_t4iKNz=K8+{&%ddR;
zH{JZ121{qBzrxTTemG|i-Spq-e_uqOx1d`f(7D!#9q8bg(SnFOCjG4lIwy$KX9Qjm
zn2MlM^c{7C2?;($ickcwx_H>=n*XihCt5%D<n`f^MwxD(1LZKx|BYirwvoF-6QyHZ
zxjRJPfE1C@_lMCf{63g&;a4dJME}7i08F4H;)W?{iV1w2zG*Jf3Ji*cf28zt`b^4J
z){peM_T~{x<C07u>)hJ_WP02i5=P?*|42rJCez_Mu0DR=X&;Ai*}z*$1`eLI<fLZG
z*o?RLOaYZ{jdaM)7Z~{}@Q^}PmTTOej;>H(a*G~lk({8YZc>_^Fdv3v17;zZkm5-&
z3uFkcY$pl4eefy^x`Q{se<c!(I>G22piMi|IbM^n6Vk(?Z$Ro7wUJMP6cj^U4A;Zq
z4fiAM2fA7Ku5%Bry8fU8x<htqJA}ioODKa&yKnRI<oRd%t#%@O7>NypIS~Q%B!szS
z1jp?V`%%H$+U&3Y9}pjg;yT1=V^H!Gh|vkAl`Ifrjb_TmHi!2uf8CYj8w98xR+(Ux
z8?efKupSqG16E^+h<<p~^Juz-UmXLJr$jxZXJ!P>qaOK|R#FIsivj3<ePIfM{V3Wn
z_UZMoJR3M+3@DznT+n}3r{6|;y_0X_eA~r*x_U2)?7mUn)^D!CANd7D&;$J+d>Clz
z0-{@S0qu-ejsvm*f3%7r#dCmGmRW9Epfj94jSBwB8{B&_$#ZXpo%a&SOm)Gy*#&l@
zcv#vE6lI7ao&%z!G+dR~T4Y$T#@G?AyFuhP0C`m44IG7!6`qu&pwP_P5<UCyx90xq
zPVw{&6}V&Oc37M(m^jn05eEw9Y$m(J+KnW(I7_u6BCD>Gf6etHJ`f%&NIS}F%36wh
zsv|eCvg5TGOJv<r;3>@nlk^;_K!Tk2Yjx;8wNE3$B0Gs-6*eiHLIN*5eS9B&Etbw`
z7A9jd-<QsqlRB5q2C91un&3x8)HTzhh(8xL=5|%oPRuBfB9i6p&p)Q&%0ItXnP@hk
zWi-qx_)t+yf1{0J+Q2wketJIB+}uxs{*9-(yDK#JPZ!4*5wJe&m?PkFoYV2NhNIw$
zDTFzWck?Gecx^5z2s+mGC}5R+ffVk|K*-MNh8ZeF#~cM2BgKzJ6${z}e7--7E~a${
zd`d8*U<p6v_Nf+Wqvtbh=2?3R=iakv(u+66y8YNmf36v#nJo`EtPKh11)FIjr)G9h
zD^if_ku&J{k^{S?3m`27H$J#B3QN)2Te|?sJCXqzXO3(Z*Y`*NFU{RO`{T_dc5>#o
zU3^aZrzqNRvIL4D<KRstz~)4zMm90#k*+Vr<WvzqV-YYXhasmip`u-U!d3`$Sl|sD
zxntTrf4t1<odzvo@4by19qOj2hk3d#D-mEu$|uRn1*^rmz1iq-mNrDCecc%K98skM
zx>=gB*?(@zdYGpzS;_lZyvIr1Ok$s3_GHoOhXwIyC(F<oEti6LsmPXeEfTz)Lik<!
zM>%u#72r1t&Q%P~{WqIqzFZ;Jvp$_y@m3lk1tke*YL}O}lm#7s{pOcGZRvjL@+;3R
zVnsY&MY%jji2X}C6rFoVcPzyW8$|lwxk-vE3bW>UFMau?%D1Zkb3ly0x(X*ByuN1o
z0e{i}m(o08^|lSrsFL!jd_|`YL!=Vqu}xji<S`kuL?dIP3mQ2=Y{m!fR77yNT4!58
zFL9Ohoo0ONt)+7BtuX&HfADZg8-Ab&@;#y2%Tp{<=Dv2BsqKTSZkD_b-wXFmAU*4$
zcgX7|+0H?RUo##-Z}Gy(*S06WYyq5nl`2vE0C6(MS@yrx!f%&OR3qkeIk)q5yX5B!
zBXCfcrcJKvJuRxXy=5av0LLl`S}`fq7=8N`5OAZCuqP7>etIR>e@@i~VJx$?`MksL
zZY61;90N6@jI(-ts$+bn*;G=RuKq4~VNv}c<}Tyq1h3V#`CJb^dlU2d6735@d!WlE
z&=GRQCEr{$K^gYVfjWH5vT6#GT~>_1NVU6itsJ6(If`(GS(fAaeX2<XmJyZDguorV
zQWU1`3=x0UBPd){ms6Gn9DfP1JM`UCf0n-cSstP90hI^pyQlmpefLza>02RlwMSlp
z0ogc_w9Vo9Kyfr_q@U%{dN;nhuE$yK#gaBxa5LUV9P>6H6(tm9FHV)R8QMZoT8Xoo
z;0=gMO<Np{l-A14N-@fEhQTGOg%mfZ5FwQO78iU|`Xe%Fi^RlgcYmd2H46<g6znoI
zrN`Drtf{lSyuGCHO*E9QthlFqcddX^I0(^Z4r)w$!lw4MBQJ#qKU$3k%)}K`VJvl0
zQb(30Cn#Y@cV!7dE*R8?v5+~$YC0EK9rI`ec@g$hu3EvO9u5_Nb2)yLfv*~^kd`Ss
z;}>CZNqI*KwGX_IKYtwEtOm+!ODv%~j9xlI?UBvLt^LV#|6S_a5#^Q{uE!Y%I~#f~
zBYR6nw*BWvHK!FH29|8qc44W?SwwE7oqO{Nrt`j}oW`VXUlyPp6S7?uIsYmr091~5
zYqN5uzu3R)iLG*5&y$AU$Z_R$)7{gYO8V{Tt+b~6BMX4T;D6u%cq^O0nh@WkAGAyu
z(J5&AnB#KWH16;N#ceIgbj;doix*e@pe#n9?=Z}mdONO5e*z2LjN*zmHI34C934cx
zLutw|3EiqHhd-bmqb^moQqjBG?;fb`SYD4wJaRL3oc|6Tr6%l1aYaYTMEb$_e!$j}
zY4nufvc8QP!GBQ9M86*4uX}wTcVewIZR{Vxp*+FkQ!pWG$@5?!G*Iadc=D!fy>LJ*
z?{M=c39`JWQZQQ*`38k&^$j9liJL4kdX4{(qIs(B99^P+bVOW^s{s_H#rj?~^i**1
z>6kr`KL+o#QU}qlz^y5%@*oZWIR13b3}!BwGG-B=+<!#5q+L3D6eFYR!dnoyyRwj*
z53_K*>0yL(R75=l$Z;91bcSY}?b4-LSLaNouy$6Zm_aFRMoU<AX^s{c*e9dgPZGf0
ze1fhA_6nr2*fUJ=fMa9{APWr99Xv&rN)H2#2QZhivyt#a&yeM$pr7Q9C_M+T1Hq50
zK~q0?%$IAK1sqUv^{}F3V0sC-sW#G2ah3B`HKrnKJB8xqIZD3evtnx1F<AZ8>@lnc
z?Blgep)_;Dq+&V7UO(g)L&X)occ1m^oaagpZTS5Bdifgx)ZWvVam)o2x9gb&xDkJ;
zmFLeqvR{|aQJZ+y=DL?oCkmI|y`5`pB6mzl`9}o!RuDI;tnm#Qe2>nKwsFJZm{go0
zFCvvwk6ja_b@b@`tgfBrRj07;>E@)r?~OuqD@`Ogjvd?4(<j%MBLc4nF*PVtG!5r(
z{Mxz8yn)$=t<=1HbI=f6G0<Y=-GG1UFZ_7aAi|lc?e3AKp>wK}7+m`8n(Am3kl!9{
ztYJ#7DA|fO4w<7mdZhb#2==ncrB5-^bdl$h=rnTiVJrR#bao>RoxU-Md_||<c809-
zF$2t$Gd3+-Zf0*b>`_waLNetM)49(Ovjj;|j4S89`B)}PChYv(Qay;~@b!N^OcUT$
za-h*e^d`Eof3MpHRy!-10{2RU{tr*O)r}DPFBivWS`Si%dT@ZOVnW!`vSl}*C|g_a
zC>|{J@<~|~I8Aj+N#WO(ikDHpQc>`73;0NHdkpUqLWV~%(Sc}%5T9=2?*Xi(Zxpp>
z)NO*!FXIti9m{oO*PV$;z1M#{KP@`d_HeqIc>v?8sfYQxQS@^;EBxplsrsZi*PSZm
zJz_nKjfG71F=X)=tzP)D;he@yzSB|m7^a!#yG!2*oCvAVhv`>9*@r6h>*035OXD$G
zx=#b$81u65JIYApt*YjgO3(RIXs4LREw;Rm%5ZqpTon6_+%M=wx{-gANwSfx<#)G}
zVH>A(<L2X}`=SW?@Y<>QKyb@$T?xtTk8>m2&&@uLpF%5lOlZ>;aH_o+)Li8`4{9O@
zx(jn7t!hI}&Yz{;!dV8qI$2roB82-0YT)yEi{3QH&S^j8sUN17k{(dqiDr8VN|_z^
zp(o0#wt0=8#$%%yIMILUM@|@C5<)rA<9vNhRwmZHMVjqEul+Es8kuaGA=%B@;xcry
zL|YyC%5}6H*OMKL9`E!iZ1&d*jJy)2%OW&PQ;9vUx=eEsp=3B8`@D>*K$gH1V!Z=Z
zv*B`4NeryHj&;m+1qh~Gb*Hww4R$SeD^TbY>rKf?8TsZKCiQ>72FP1AOcL}kxi+fb
zj6N$t4^fCYdM%jaW}MwxL-0SJS~w)L>rY_(c|?E3C*^)`Wq{<GA(|T*AiW+4SP9?;
z+IG8@S8o*iTafi0U%j(ejti36ZXLdQBLgd2K;>q}25K^R5OP4Bnl5o+u$A>f1g2r6
zd8??Mv;a2m^l*Plq@^AK-*~T+(kZ7*$ABwWo2ZrqbPLr!#^P<4j;Q+9?_ilvv7Nb<
zWy7V#U5Xs1WMO2PKljnvA6|P)mB#~`r|U;>LNejp0P5NaoS<NeW?0l;6nD*`V64E(
z(YkDIE1vr^B^pXyzx1sX#%1JE*_ZFA#y9OLQ8QxOyOLbH+fw1)%_!&;XSWwotqW@8
z#a92&+}1zb!-xRnJ@nhiL~u-tQDiEfK+Z22GYP8G_IwdP-JeyH^T4)<PnQtZg?xL9
zRJ5mn^GZFM2k-ycA%DKS-u)j<lN~0Pam)o2w+pZZj~oFtl|e2hm%+>hHGhdC5Ii`P
zJpFl+r;n%C&!>O9Ydl%u2&_7N{x|{Oko*afFB&>o0AVzoK7XEmeUVZbK3@4iDYfRq
z@^{B)IzMZ0+yeL)!1>uCg!b=zT=_nvcJY6;ZL~mqD86%vceH=Wtt20YTN<0}Vm{3I
zx6famWOf1g49)rN^Xb1&y?+y=oB)TX-+w-Z8>m5Or%pg19Q*dyzfV7&{#i{F(^9zb
zjXys|Hl2`c(-^~N8*Kjin>_u&H-F*A+Pl;LV$+|YxfGuK=|87`v3=R1?toF3#t@BA
zXS5Y*HtvNEw-V~|+xgi@mGMVLff!F@?D3k<;aWJm&@v!T+>#hq<bNkXF5kVTYXrrf
z<Y!3~{+IvZ@;_{`BTd3xUK8#OC?l*%dm+((0hblLi3ERtG~QlVqu97(OQzr*;xeKV
z@RA8#d><j(U>?zpkqQA)7#@~yg5IQwpeRp@0JO?eKn;^b+l?^qhGPY0L;-C77?{xt
z)V3y*4Bj?)(SiB<;27Nv!wS4w3-E9dyxIt~u10Z~{UM6KrTLeN&IKJCdHKfQb-2F~
z;@wcJK&+HdS{($jB0=Eqhd8vC!p;RFe}&Mji`*35OTMc)Pbk1~p15Yn{YKPwd$)p`
zWFvqFp(eFpcUMyDrf^jY86n5$aS35JKDTivm&nwk?3VF9+HM`MO=RQ{W=W3uJX7Hb
zv(%TIphA4S)rn=sX)AHe)QZfu=epmY6`V|B6Nk(vc2rnZ8MtoQ^my&G4Nr6<e_m~d
zS;p-7nJc3)fr%5Qr29a#eg2*&5Z_9_H<Fh9fUE#biB}v0G=;Nifi7^8&Urrs+-nWV
z>o^QYtAfg5YG%^a#`M46+Z7ZsQHNttL?@gan03L0hC(Lc^<@+{Qjh)KuArzRJ2(bJ
zZH2>uKSQ4Lqq=&U5zo(r;KwM*f9=cOZjaW00zr0#`zhgFF*`T1J58*OEUPg=ez|S7
zC^WJ&+)FeY1T6GUsHU{PoGFm*@IW=#@Xb$|d6K7+&{BTIHo8mrnO-GJ%aRKx|CeMv
zNMopUl*Z1npp7#yW%wmu6oQb-^uEZ;%D=e4auWoiR8hx^u3!;;3_SX^e`JGt)#j^i
z8y0SZ{FD*Crj|T&(p=$ThQQ&|NB3)1*ux+wINJ+*nv)@6(e*m3HAW@>MNs(QZ-4!H
z>aKjx;pOFg&XBPWw8VU>>Z3K7XrxjuW*)_U@KfuoP$?EK8J6{-E{+pX@citD8B3yV
zLnJE<6UW{{DK1QkjbX@%fBaM+asnk&CB%vI2AOKr(uG>_tx(SQ^j*HrfP(>r@)5xq
zUHSeo^k`8S3w@6Oouph^JK>oE_Tz&*KV&X*-t#!PWRwE(M;yoKN;3{6Q6NpmA*q(E
zvJy#}9h*|b|HO%f_hm`v%jhYq1Sz!<J!Ors)=toS4RO%O$zKTVe=-G0^YX3R$jo`X
z9oKvo*tC71K?(5F?E^^XJGT#V=2E=kIlpHSyuI!tSn;J*|4suNM<tC=sE(;=QF;!{
zD$3ObOlH}wP~1yTf`@5VmA6uY*}Rj5xiouSIWF4?-+nA^W2`+1<@7NavlZro72=%r
zV_gxxzm<7+1KN$1f5Uz^R^V|cw8DenalA?bv>_O?Y~4nSYhYLlFVh@<H$uBTSwYBZ
zp$#2`5U&S;N3=w#txU}WB4_vbS8y?km%@VtIUcYuz6x@cNt>pE&-FN7AAr$rUsf<8
z5X{C8!ibndXlfZ^WQX6<f%pJic8`Ar7i1dc+(Ec-g=cM3f7uP3FP`lKaM?Zn6<kzW
zvyZ_=E8)P7=F~Og@-~Bs&zM~ELay&v_^-L0+lc3GzgO^txIP~PrvfJ0t}Py8dKznD
zG!%s-XBN34cDoE(0T}adyWzNO1bHtQYZ#Lt_a$<Dw~1}lL@=w}-zBLjr-YV+q-IB=
z(Je^|!uGjxf4zkjAu9+4#+m+KgOC*#fGjAz55#Mbs?gV8T`FjWzLE6T#CX0EWBjb(
zrH%7ca+gu`R%${n62`S{AH}w@qz0v3?e)?LgmI=$5Yn1Cv+4mFikh47hNaJWJAt<X
z*Fij6US%zFD6za6w1G(g878n6Z)mcnS6{1^VUnJje?fCa#hlkFtG-(_0W(n|Q$kVX
z!rTxTrkJLz-%PH4|5$tkN@?Mkly|-<)9k0@x0;HwHEg7&=hSLwwJnuBuuiR8c$Q!U
z=n2B{Jt|cZaY>Jk2%BL-q_DH?CI^)5&OEwXns75Y`d10F)ke`9p5YK#s(q0W!wX(#
zt6rM@f2pjv?Ny7FLe4(A*KvXj>FKswd~=mdGE@-h?XXHNJ-~b9N^N+6&|Bs;byluS
zN^jqIQ~cdOG#9p-Cr&RZsXn}P;3g2=jQ6Zp&gqzg${85gTcH>!9-EZw1|j|q-?R*Q
zP;Z1?hsO`9Z>`_%@0S;Gj;;x&-xm7yf{9t*f4bM*t2%whI6@$CDYS7M=mp4$|AfN}
zNHzBIo^x0R!yCo@en6Q>I?$Y^rn#|?ba&i_(PG!f(uB%8ALF)@95u~PJLVI6BOt87
zZC^{udfq*_dt7lo&CRpxTs|$tbKyM7u*z~(<2WlgMr^HE1SZ~;S?CS6sl4MbL{1(`
ze<;NH0OzHHFRaE}^lv=kpo8c*V!S<K2rp4`7|02!FJ4<t(-I%$bx0>{(}!f<zI|lJ
z+hTahwNrRk*EpG9h%N-U#6qlX?Xv$oc93+B|F;Hj7){BD7%OOa&pQj_vk_x7<MJTr
z9B~G+eeX+r)sFe<Gmu){39zGtRo$gjf7BIsY1v(-+Wukw^28txlCJxyHh<RiGU)T%
z*^Cf0@d>psYA7YlvQH?*<Dy^1>Dcw&%ITR(HgdZ5#%^$k*<RTDEe_dbhZL9ljvklA
zJQMM*`#Jd=tv4I|2rnza@L=4sAF&pA)ZFqyf$w{ts@3N>iCV>46X6E8n4?=We>@()
z>=rM7U>_E6%ww6ob=}V?%t(Gin0XSeslYFbZiH2B_5HXzypWs$#+}c}!AL~wPB<O<
zs7|~tI^noRHV=CgK5mPSJD&(P6s3Yi$#y<NTZkoIap=0}5VzOzQBQ@rg|fEme5<m=
zt2!7t5l&O(Ud7Y#+aVh`+y4PS0urs4!4(V>mu}Vt7?%>H6epLh)&(nnppq(+2?7rw
z2!Kd_`s>6`Kfhc)Uw*m?c{0o@FnId>b^^M==`V1)sMW~;FfHKp`Tg|k8|SF$bD@vq
zTnPFY`fcf3y}t^uGy>>9fcIA;!H>Vu=Sue>jtkF~v4Mi357M1;xFg(K8pY|OX{0uh
zu1p{E{@dp-FFc7b@~e%1WW-hG*YhWdUr9VDo#&+~xg=Wh<6S8FU)p@vyrr=wkM;Nb
zSVfYnYoS1N73L?DNHj<?FST_3Zumdm=Y_VX1Zno0rx+<$4gz)mY@(O0=y)|eW8?gR
zPmJ{7OGJe^?Sb)MKM((Xe7^jD3YSz2q(1%r{l(vq0+_IP0?nL%(5JuuJbiol&q%d3
z9t&hVTx+E|;{i?<B~MxblZaUv9v07Z5W-~0ye{Z-iNAwD=Om0Ia^=t8f1QRaUmNvn
zV2+z{MnZ-o%o_|D3YpgU#E<Y`4U=LbH7v~UXkl49nO{Z)XG1B9Ho^Sv<q6R-XC|`@
znA}BCSSIZhY2hV*_20A7B8^5hKE!YDubSfy`{vg<`_Akye0LU)#xX@eaL3Skd;j$}
ze)@xM{X*n~v!{Q;!0&!A*%BG_|4#oU<6rTS=q$666b?v>$+Tu3`1bO6KSCGwX|`ur
zBbifRnA0*_;TRAJTtUAh%M+oMqxFpDQBu?5rgg(PQy_|eqt!_gm89}c?wnk<eSbx5
zJqYNMjxhm2kAN#)eGUN6yjf>{1YnJOP3Fr95^EVKSIk$=%%8xsk|`q|1fEsQC^JqU
zwkVl4CRLn&+%nG1=AsbW#pcp)llI-ZKThN|njx1(UbQT*L8jqhON5r2#Bur{XW9v9
z6Mhv+g+1PXCAsJpoRe%TGHFS~s}A+DOqw&TAErag(3o-hAWzuJOX#43))3k-f<gfV
z&4Uh-v|e-97&!!Kb<xj~XZz_!LRqGqA)#iXHid?PEm50~v+~B~%`3oeQ(Pi*L=^p%
z7X-xlBMBo%UmW<_iW(KS%ve>nn=jF%Lrf<K0J#=_g`sStusC*%!V-lPh1LHuJ^2<D
zk!S^%TWO`Jti*+ljKOW?=R7{})*$1#xD`4%mNTtH6-B=jq?xV@QmIy&z`LnWS6(vL
zI&obbtt@nP-78h=ie;r0uIU(^)8!DIN16e7v#ME|fJqIbX8C#JDe@&aJ`|Ol!`c9<
zEmzKejW?`7<tw5Iwu7jhW60(wY0I$KuS{2h^>epI3+VMqw|+N4ZBv4l^K|t*dW39@
zDIv(-{@QZXJfBP|&R>!&SMDbyN+i)G5z=)jPwR1OqEV8J-h-JH57ykpx@#4qwGq=C
z`QDq9NR$jXj!6kwpds`6(WFFD;EzzYB54|bg{qj5NgSiNVroZ{l3_;V3keB54|*Ie
zKGM7tf^Y`uPnfs6@^zZp43xZ{dNc6X=EHRg?!Zkrr+EEdUeGUnuIG4Z%}sMC=GL0p
zGKWer#i!cjy+g=4*Mb5>)jEXWdT>tv_v=|YN}?yTEa_PprYTq*M(5qoySI=GLlY8z
z+WUEOia9x%WhEglni}|;<AeJNmH)#NAKjI&K0<MPZWZQ)bWS5ZUrXzl!bxT%OepML
z$2sqWw}-590gefr@UxKi3lx-1sCAS}zB{pFMTrhMye*mcZCv*kI0OUZMlJ?Br%2mG
zHvq*tx+`C&MUauAxKD=Bl2eIdpc+<x6(6Iy)*8Y*Hu;^)yzPjOE&tK@$vdpbdj%L-
z&SkyPh-hO88)ZK_1hl*<mfSJDvCLXAng2GU-aDvz^-4XV>gJ{LE^DH+g@qs~CJ?qr
zjwv?Gp>x{i)R(lJ<cj(iW)PGg8dqc1^#YZ|-h7HX&nD5${$2vA+eU2*=o^}U@7aLK
zZz+O{F<|nZ(pr}zrc9!0(uVmg{hrbkj(jg|uq!EZo+p=Gwg_8t=AN)%7W8^9j?59`
zV>wF+$7oUBPhux$mhe(rWinEb8tJZvoe*kM!fFZ<YU1=Imp(;nS+3T4qDz7@4pX}7
z4@&%Rk+-TUQfC|mtiw`mpk>H^2|4>Px`p2d(=Gf`{@{drl6HM92+e|aO)aD-+XNX~
znmf$#1`-sn9s+n32*(Vi`a)4$ql;q}D%W*V@%t9j*e2e`n#B4$lA`&}?o(-Y8?B>K
zZGh#;BRmN$^3a02?zA!;ij^%4W~*>*#4<In3|(~|`3%-I=de0nN0rWhY3)v+4<oXH
zH2QCR64HLJD&AM{#w=4!MxMJ=)HP>;yc|>I8jqblxN=o}<v?}|%)>%&fC+{H7f%8c
zJtgRp`H#9o$yCD&;qpwxwcryW*=ewN*VlXl*`{)xs{)(w9)@89UamvR|M1Q4!E`gf
zNU=3o{hZCupMy#xrkfvs>b6RrwAd)VF1PjYI@8Nd0?RuJ*YOOc#MB%1aU7QoylspJ
zp0fPV8m5{Gp=gd>^O?$RYlom5vfm>7_TU?Fd?w>pXlTmutJS79pKH7(U?;SPA=rS{
z#0cX#&>GEDQ^GU#%zhhEX~SvC9lP$e>5w8qBGqp~mz7PuQ3?8gK;xS3$=rFhQx%6H
z+fap$;mdPWp_M=2ELNV+E>lV`+fzH&`4ZRkSCzvhZilEc9lrYSOFE3b#G~i1!F=T7
z=baEAhGGL^72=lXAgm<QwTQAGVX6AKJ)VG$2$zna>8C?j>_qq|5*q-^7|T2dVJVod
zMWmb!g(yZYqI=GNCjm3rPz{iFqIOuk4cO2zDV`Lpqd;bG!wK)QR$TB0V}2BRE^;r1
z%Uvh&_aS3F49^Da5JRm;$DNO+TlnQMRC!L^=}Wa7XNoT8e6$@8yO*jBO}4$WBy(Jz
zb)3M(xZ^3y37lhA_N{td%39V`TRyj4)UE3(<GxqPay`F)pX$`UWl{^5jSu>R$Jna2
z59s57YyfRzRPh|3tv^MPEzlXx(6y4=t}ewSue}*g3qAAYoV{-7xU?H6nix4e2SsC;
z!y1FD<|HMz4Rkf~r=xgS+6@$Sj1HcIqSnmma*pQHkXnaXE5IrVu7`>54}*`3v;`Gs
z+Wp9x(Y+LZ_MMV>s&YNNS~?{gkK=Z{j00$x1T}fN#kO5qr8{e(&9T1d5;Tsb`)jr2
zw(muh>FY7m*IDulGnwOQQ)+5aMA5V;mSgyj?p2*z;<L1n4qbk0&58b~LP;D^-DqhP
zN-@J>e9B9MJqNm02SS_E$K;iwq0w`$3(szM;kikFA74W^_v6uB2emC_tAUszhFaeh
znZ_mcAeP?E;&nw0p-TJCf-R#6o$nT!8rEf@%L{xWK+kuBxwyX5jbd22o3(AOl+nF4
zMQ3Si%`x4T<(H1;3UAxZp79A~>mo_&i?9Z#JoDCpW15>mj(PU>G^+eq<4-A%F1GE0
zuR;%8Itq(Y8`0Hu3m(^28!$B1%CL_#c&9G&3yjXI#hDK+xG?ngCEFlqmFoL67W;I~
z)He;DNV%p=U45w>*H&toZo7PQ6mltJZ|x~lM`6$%;%vJ=ZL$9X|6`q7mvPJm6t^+q
z1-~4ZF@gvuw-^2e{s@13fpk22aEUKWjCrmc>A&}qmfe7C08KrwI0k3n1eB?9PH6f=
zHn;H+@qC=zRQ2@n!Nr+vneHgGce3jk+CUtR!O$J{poLb;su&{6>6o(Dy&J<$;<0<O
z>lm8A3=YB2y6~925SmlPD|@I?fNde7{rjl9=d%HmAXbY<iar(hO6V(cp;Wz2t9s84
zjqm>j@M36VmvPJm6o<J72DiBf2f*ExK`thjpXCQH1I&!Nm(k@1A%EY3(@l+o5rSzY
zhL7*V&tEy0nm$+h0Ov~4$I#!Fe(~j52}>gcT?z5>Y(V(&@ASFRa}mdd_sZBn0qG;@
z$t65d?rR#w>7!|+Hjysf6WY^%`Q_usC!RzQ^ed|0w~wdapL!v*kPNgP{{8*QKL90|
zvU1RdA(-Kh{|>)C{eQDi$doy={xACeHb>HEW~K2YfBuCJ|DlIJ(qQH6@MjqMy-dT0
z{|*153APAZ#sgtRfV(xqe2~mpm&lGJP&vlI9VfAqGhvwpSSO3XjKiL8t#4M4zHzM)
z8+{`^lc)x|&m1jP(l<pbt|00WkRBoF8w3rD_tC~{BrQV!{C|zE9ewB-&^xH(bIo78
zI`ZoVNgc0=TKHWbA4p2S127Kt?+ManpubT@GGVAp9C}Vg{A8}fvr8I}fh4@e+wwKO
z61IYh9Kvf9Y|nG>lE_F>M=Zka_lQpx1c_Vl_Y;R+$S}qv?o8s`v|Qd=m<#FKC}5Z-
z3&|7t_cZk_a)0dKMTjIjh{HQ{&ELbGFS;Tvqpe38#5i}wG+~-gSL=z14^jzLt3_6u
zw}@O24byd5?;>WcBHjXyY)^UiHp-)kohs2~qlJsJ9-T6fT%PsiS@WQ$UO`k`ZJ{<k
ze&ugbYdmD)lJ5H@yRNPY0OMGeBrP*!uHK?I2z$WK;XC47L@RR%>FF&Z#`A2qJW76I
zshhbys5s6N{7aBOu2D0g@;mZ1tA7&eGV1y$fFA{UW9d=T!;XOJT9+s22S0yY`86Ep
zKDHEvkCx)P1(K2pB|mF{=C(vs%ec2y-UES$FYiFEo;hRC_8t__HD>Ye|M(%j0`sp>
zT!gxP<!?Xinq}PRdnc*h*f~d~uX`t{U)ecJwTOFXsb7h+%!N=*ELwuI^HJ1BbcPhg
z>rcgvM@j4qm5HogC%Y6X_b~|hj)Vm*GWryk?dS(=f6uq>YBmW&rI`TPCt-@V142&1
zI;us#IK9Hqf6-rm{5EuVeoCt0_<U}`!yKrADW$V&c$1s}uXpfbh9XH0(z%XGetQRt
zi~JEXiYD#(i=J*5Z{P^qiG)6!2+M~lXbhe(@e7`@{1y0Z?)>y^TtNejvhbd|5M3so
zM_cA3W=?TRXkX$6;3-E69$NOQmd4x?ui6k(LJgu|+uW&n&8vipDPxLjR=c<Pyi5s!
zmTHiJS8}BEW{hgWJT{IBqpvZjD_XOE6(sExu%_o>VHG4uW?;hcb^v>ZGfS(W#9q$G
zK+~tUKI#X*0)MBfQy?8+^S5+u_aS^9i4BBJj6@!TuocWoaU5d6Hvaa9L63*{G!z>U
z>lmIq24X|gYT2hXO)JwB-b;F_X40B`=niftY^TNBfDLu1X+8!vbj)$JVOn?Tt6VS%
zeWG)Bbv=#G2JWzVDt#)hyf@v#ukx{oc}!d>m0@l?G=Hz^CztBX=xw=fP^HZ~t1IW_
z*}zH0XyP&ZNzy~9XAdA+=UHb`wu!3sxXUPZex^Or1~B0SCvPB=8q==h2J+`$u}N<y
z(5C^}09u5&;xRZ2O%F|tb4D|q6X&z+YthJCtA%r?hI3_^V4?N`c~;;J9C?f$9s?vt
zMiWgs)PJx#25!pTr-4*EK|C$(I*PPwLV)lP6oms;9XpD;xUyDDMG&q<iSMtco|k9?
zDeF5V)MrkA4wKk-;N{8c^$_chjbg%|%8#+pSUo9+v#RY|ooNBHbxYUU;pEvPT;D?@
zCnva8lI#3STBbs45|)}{?jx%jagVf435m3tWq%`00Zssl(EAzzBQ*<+KH)2f(cGz9
zf3%=r*YvY+3>(pY9c!6=M+)r>!}a9F(n$=d54m*9Tu0`!CBQ0j4!uk7hXu%eG(8`e
z5<gt#`FG^u9ZcT4-KVi5FP;Df=ey^8M~&-=p!__0zLSM4M9<y<?bMnHEvLNAw@RhO
z&3|U!d|Cn@SJ2))FMy<VH)JJx3I{r>Qq<nB7FP_a<}r7?buh0^OR4L7n(W|qbg&7z
zJ<yzJUyKcj9o41wMc*N1CbIL=xYw6#DqBTHIW0R)7uyu;JF7Zim=qHp<$PP>vLaBr
zZzgDuh74!^1EFb4l&x+z@2cDYcNepwD1Ww6QcTwRm|0QYKdFhts6NkMY60x|+rKSS
z+uww>b-$7RE9R(@1E;;uCu+OU1_<Qwv{n;Exy@MtCmD_ned|uUsL)~O?F7@-c9U@|
zu5Zb8L#W7iuZ^DDob$xVqRw$6R=7zmx{KV*#S!_GO^xNu&`xNN_u)Dl!t-|e1#3H}
zxMYgL-;UR7G!iE;%Nu%%SuSy?32?_ccpas(vZJ5=U8^N}fNAnLKF*$Dnp7*Ix2}PT
zk{0DAV&6_-NNZ1tCLYy0@DYTtb3EKH{-fX&(QWqiZco(b7>3;BsdnN1moljT1Nl`3
z>6h_~1{Ak+@CR&Mmm(5DDYpqt2*U$^ho;zjmk$(g1LQoHAEjrVLCbMKc7Qf9u6Pd4
z#tAF!6z76w$Wx=)4}7~+MqXPME(<p6Wf{11f_QYXI~eK|IXnkMZH1+py<mYwk6BDx
z>OKr&XXJ5wwL2(M3J#uwA{j9zCm~xH3oG~PsG7jHD9Qd>>hTrrAti+d^_9+l&%+}6
znq5Jxx*uL`HM{0rwT_-6@!iU5$K$@KWyM5+os}O`S8=a~l18QCasAFh3QlJWrC|SN
zW0X~n-lSYw3!5BS1Q7Z5efHOCUs9Q{@ov$V<fmSjw&YA|&)MK@)HEokg_1SAjPUuC
zVn;WZE0eLbs3z6)_q-TlarMW4Uw;m>PyYUWTo2na!-2|dMu1tXHjg$SJiCwCXnje#
zMICQ>yAJ5$K^W)e(owGJ!gC1lOI?<s^rLzp0-ICci3y?>78iRKLXs!$8*=brt?kv(
zfnXEcw{NxdxL~Cvp@_CvSq^&%B=q%h29Os(i<!3As8zIMv(aT$XX#XbSlQ09$I6*b
zC$h2&o65>|^cpKqzzEqI^3|)G<>VRhv3(1yWQiYZ<~qyYb52S*VW<YbV_CmAq*CCI
z8mLJX<ht!_^k$3aL8s<RLYO`Qby4Gq{EDkW%bv#L+*R>Wa~y7b*;RoNNP1ccX?^dS
z-#y@_@=<SR+cXT>skNhjYle7h>oM}un!Dh{nL$cwfv(0f@yJ|Dw`L6HC5>I3qmIuy
z;Ze_L41q0ftgslR_8VW&0VU@U8WuWi9AW8VoHw8UoIn8ippvt5=`$=IlE}%1_X~Z;
z67c#eHK!_^3CzAFZzc8Ucks!4$=fQ^(6b@LF)ROgoIwiXYjQJxNyTK$*`o}b0GN-i
z^{CSH&f3e{0Y>vD(5(+U0%I5YlqS2L)=gnOeLkQgq8Hw}P90H^K(*K_t;SPFr3zf^
z$^o6SYwXl12Z3{)Hp04KsOAjJRvaDM*U3A4*E?=@>)tq@0B+cS0@iS^M#f16JC~Gv
z-bpf%*epm^mN<%k9gBtQv}G?!Rz1X`P&4$VZt5G>B7U>6Jj_c~ah^Krh+2ySppX!8
z(0AX7KnyKVs!8|~O1=^5v!D#I0SgUpwE;GDS37COr{v_7EU3j}#^KvIL>Z;;NMQE@
zxrtmTSvDZlUvtfM6*^uwqIm?ndkY6EJi8V|$ss}|e6AmVtx}l3AuXPTdpN0TKE2Kj
zwMgNQw-k()hCBu;yZTY)VR%ES#-}BUMIQrBIYug>x4;vc(m-k*-U5~A*2ip8Z==F4
z5x|^`N;~~7Dqvh0R>Hbl>vYr5bhy!5zlM(1v4N@YA_Lw`YrbGMkpZvx24^9|_LJ0n
z`uf7gw4e-sr7;_j(f$j3=|jYkuzuu<-A2TLKxs}!q@8{j5pt@BQYIPwibyhY85Pxm
z3OjH&zS#tB8x<4!uV?W@>Dcs}O;0%{t=72s!18%&Qy|u{d`uLnuXD%VN6B#5;i}sK
zzl{>7a{FE1N4Ue$$ZYz(r4~dw!GsBzpwpip`71epJgKkay3!otS+w^Z$dQY0gBIl8
zw~-4`62Z9^dN%z@eyaf7gAzJfPT_rD7LpuqM~+>#k>VgoIejux?eu5+GKtjBE7xJS
zP;;;~^6I*cR;WX&lZ|;h`$@(ejWPYSf^U;|YL+#D`TeIGhi@Sw51wtM*yxM^Jr=55
z78-1S0-Mb8FA=3yVB(_YMM)|p$C}>c1C%Rhdj(Gus7THKV3Y!dnN~DQRkF#XY7Ns)
zpdP<RnW<6Pd0*R=s;VnZK(o^A;0Nrgm^<1y!*}B)ru2CANmI@N;h2l<0`YdI)t)yX
zzKeC+$fmuJ&d-FF0@zoy<xs_Qk~P!TPRhGJ(56LWNqW0Sx^OqOuR7Z(%3O(c+yz&Z
z<5hdLSak!{7SV{a0BfQ7GUvlsYOJd`6Pp~1ZUg6KQQ?0DFyWUWUp*7IkyZ!^HJ5QQ
z0TTi-IG6FyJt==nZ{s!)zWZ0`n4k#5dGOV418icuKyO{-w%t?F-Anh-0tJfx`VJqG
zMaq$_$nl~FIgUln4Cg%`sJkzn?ta=lK5y>##GO;#fWz+ja|iN5%O`00(VETyC`-^i
zf9*cL(;BTjkMgkAMDjTK?&Ys}xFhhg0^~n{hdaj**YAJjd6M^#*QI^K+C;&~gYxEm
zdPCt|S*7K%veLRtmkVuv`1JhqMwcUK6>R((|Fr+@o;QDP+5=e#3Yog!zHZ_RCQ#(D
zvsyXPz+b<2pEtj>R0m#Kw&V|^yT874%`YE{N`n@aYHN+DR0>-3%$>DCQGAtE%3H0C
z9E?R(EUJHEIVeJNNb7_HPzRA$VqE<qjPLgzj&fyLOQz+DqF)m-H)9T**%V29CBZ<=
zc@=O%O^Zv&Ms389?BecBMeiT(td5Rgw*7w?xfhN`+(JKPPRdYs_(*50e)%wxqSvJH
zio`@k3bSprcG_GgZADhG^Vhaj%{AJJZPsqon2CSiU9lS<G-=#%n|Y_0CTE$kS+)m9
zHT56q9Mvx)dmw^h8zvn1s62>9Ih@GTTW<+pVugeged&o+h8=8G5RYkj?}}D^)=ARM
z<H$D`$mGS!k}w+%5EG$pJY@mZK+1)Q&fo<N2d^|w>|yn~XcM~`2nXvXoAnW$b>2t&
zF6)1fb~$mg?$_yE|8010M(@2FR>pl+&n5fv>;2yNxe{i?|3|rs+ym%XWrnmTdSMZZ
z*hRJ`u{>25jg>@b24%z=$063WNFxR(Kc*yFDxE3e8v`t*-RNjT9gN|Z-uki<=4J3w
zto%d<{r<!*RDrqOB}t=~x(@&<4`v<Z6-j^2l4w%8=rzYB8DVvzv(i<57t2D?(o@+r
z=q*u);eFB1NUxQq<0-hes`L}Fx{w~Pt!<t<m8`Q9E@ULAaVRvglZoz|q|;ejqoR4g
zx7Y=7QnXyBcc+536}6jxZu1riQ%OyZd|eR?DcQh5ViNNP=r_2dJfx99Pc@J4ZfAeR
za<YY&iRFy=QY{zuOw3Uxo)n146_a13)0;nyq&*v&)AR~B?&hdL7_}DT9G`P)K<$+U
zsCoT6D;|R6*V;R7e!RUy=S~SJU1#8ll*Jnt_q1f3at7L-7Niu>b$VKZUNOLeKIv2s
z)KHvuDvK*KSgSy6-%xI<^1;Pf`Zj-PSJSbMDRl~BzFm=Nec#DJYOVOrb^FeS>&h9S
zp?1cMkUD}_cb^Tv$)LD|V(w5|irW<wC4ERPZ7G`L{afj4TQ6AGzp-a!oH_8$1+upT
zm)<x1&C^7&3zde2?YPwEE{|sYBaioKZ^-@UM2AbxVU%=k9XlxsaG3h)K68IHb&E>o
z$_aIIg!%>ITq4h1J<KbnYooj(_Kko5mhf^O-Vg}jRBGvh1+Jk;dU-*U_6v9-+lnz0
zq>-?q;)Y_v-Y7mR!Yk{Mg0z#GPf}iQF_HovyuEU4HC!(Q-859sh_VZT8PcEP<WOs|
z49iLkKB2mzUOQ9<aiXT(@#=qC5SnK7D*_O`%s*4>Y-d54i7gkiC5?*oB^5|9W?-T~
zINll)1eq%)3QU>(X;mJl=o;u=DDB0!Pe4ec_oM96@7Nv(n$jx10N6xW=i+rp!I;`v
zNCBZ$WWy4qpyF{JQm{m<j1*LIore@uOwL9Ms$FLx1&zfPAO#h#vygv+YDYl|^jSzj
zBjsz60+7kVc>fKN0@6l&$X~FLYA9Y1DQGC4Knj}GS3(M6?IQ<QZgU#0D`$v?+M6Q<
z4ZjN@1uez(NI^?+9w}(OU>#C`R`dk;zefsWx|FO-Z-f*e1;v4{gcKk+Wt_WqoY!2}
z#(7Qc8-WF=t(0&7hG2gIcvkL$6|SPU0w}2HTmUAh=v@#b07uF(Tmd8i8E`$)m7}WR
zdLfXYp?YnQpkaR@kf5cy9wcbhG!GIqt6vWiv~1@=f@XCdtzVt-qThMT>&9x(OULSS
z$7!!jaB;Soj?)llG=0Em?5O*ADqYDqa@;6de1DlBdV0LxkuHA$oajPi<vG_sJgQOm
z%MW*G_+5WF;dSz%F(>8gS0x(59_HcO*}V9CUZpQf8R>n=BR*ftIII|`m+K*e9*@t%
z<g5&ycBIFfkF(GNIVoY{%A*Z`+}=UB`9VIdyZL)|{HB1?2v)m4&5qp^oRVpww)1e=
zgmcfEwV1TF@2`LOj4jb$Ug@n=){L_5LCfQdEVXqKZC|b34i|_p<#8Nt?M<%lYUK`B
zpV|YI8#7N?sM2u12>y`7slhSH0sDCY$zhjr_;?3q(Btz7XiDY}U2oL$tZekB>by5!
zgcai8S1+&<3p3Rl=RFa;jQNdl3tSN-!<%ONr`vhBN$L`-t2TUpTcy_WclW&c59F>u
zo0m~?0uz_Nkq8u*1CBH)mk|Cg1b-Vh48HHLnB(qY6_FzKXch+c5eMk4!Jd-tsc3R(
z4h>MCKfjbl^N=(i+iS1W!`Pk~=_5sc6eZER`_}31<JH6S)w`{s&MD8}uzUX6fxOW2
z30fXn?i_%!2D<0(-IqJ9(aQB87in#bTuy#_`R2{70WU2;?f^Eojv%(*%767J@1bv(
z`i*T90?UQ+=03eK;V!q*a#?9<U5`r{n>R1dzh3D)gm+u^Yx-M9O0*pH7X%jlNzg>F
zW)VZ8Bl(2-Zfmg%Dp<5sppjES+XC(S-<#Zp7rU@kqI=l|Z*I}K>#w<|bBjHlR!&A$
z^wjIZ94lu++2xxEL%ja7E`Kb5a#hDKaI<QR5ER$K-mgkQZRBBXjF;3$%bV!T`1P^6
z`|y1A_f_LkEh<~z{qg-Oz90k2c$CM=2?u`s*?qnGtwzmk(Mn5~{)2V*<6Bof`ILhq
zXd#uhmbn-?Xki+4mO*(xaEhLUM?UlDvF8>-#ZnK8fTSh5xB^VXlz)&Q7l|2xdoV>&
z%i$v_Upx`zM__9ei2F)l)P+_Pdq_!?RgF!fg*=eZ(K1B%kTu$G&b_vTAkFCrV~WX~
zz=P=^3QAIoC!Y>5cqtWSupIKVz>{R+VtQx|=T*Q4En10}the3V3VEZ4ALjPsY_iFI
z!;=;JA%Y?s8vBXne191E(2zodHG;9d#&HD64z?c&%M{LTkkET6J&)fuj{0bpV(>~+
zGln$Piv%!>RyUD3(CDp?j}jvY2BDG9>+%gM?HlICxX_PT2R7aGk#s1Y%p-M(108L_
z;~el%aeiq!+n2oAp#=04(T^K2csIB$W?^^NJ2(*^hP<4pzkluB$~d%R-e4qeB8Tbk
zakw8ylSne*jD0Iof!^H0-J#?fQPx8=OC1O~VA1{fCQ5iH+RnyAA<N`V|4z@QBAV%!
z^@XR+tMV5~{t7<S`3pD*Juj2M#DKD1fHP*W3h|67tU}(C!>ZO7N@5a#>HD%ttde?S
z606V-Nn$5Wcz>*|@?rOcce+=|lM2pX*ejLP6TMQQ9pIIx;V#b2$kRM{g0Cp9+aZj;
zdOE+a7!_!^w6993(N{%Avi+=|{6@3v<d%Krog&kW<%s3r5lbfwi5HbD*fuldQK*DU
zHtr27zaCNc6xbMAKP?&%B?USTb?Zdw6Ua-TfN(a^<bU9Zi;(<_hhIYN{+D;tTXTN8
z%i0m~(g~xIdM>(`sBcU6Y6Nf|x|gUxC)2$$$sW3wtxu)9Y$VX~8r^3khn*yU&4$A`
zBY`{S33ETJx5rIsrf7>`jBKiWjWB8&-CSzAQx~zS386unul-5qCw!lGIMnP$b_PS<
zk@P)izJG@pkkfc6hxFr=|NDGD+a*nU;Hq_u_|!1eV-iaYHrWmqg3{;PBP=I7e`Tn#
zTs+OCfXDgM1li;mbC5r2RVxBkY#XiObYjRho`uG<`b|B%LEFbx977WiR#MMwUr=+P
z*exHx10b<Qpc+>!xajYlJmB>R3QnY$M#|uE_J5t12F{wrIY1)WliPKi1iG|%N)PwB
zO{@%%-@DUKep(&Dpo?&acv|0yozUsU^N4k=(7^%c4}GhIW5>Hm=0UxAh_QV1Mh_EG
z=B~|W9D20HT=LEl%ymeoJHD5jq}riH(#TISlyLfZAEfZ?YzF!yKl)?&{GW_fHqOa2
ze}9aHiRCH-nV`tpvddpLw-B!1%cphM|E!iA;u_M`{a<cYXBu|Oa;RC;pN++Fjy9T|
zm`%Q!pO~$AY<Z^8Mv>&&rEFE4ebF|1%olH;Mo*_1EBYs&Pwgz@*tn#kLh`<vmTXGN
zl*%4h(+a3O+Op@KnwHE$Uc0i(K{YJ|nSX&~k}7(xvkgHAiq*)DU|U+B*8N{r#}|*^
zV@<K>@;ewB1kqwIGD2%CdwpIX8^={@<>+5tS9?Pv>s#^T&#S$m7m+5M@#wNbJmb-2
zg}mqJvTA*yqe~=-X^R)$iB?iiJS3~oE?R%9?RR|rtwLz(Z#$Fv>uPSEm0Y#d{(o;9
zJ8ND;rLMO0D{5}qDrVQnk2OEDUUx&I*tOn1t?q_K7Q5P<weE%*Wg}((&vm!875GA}
zE%$2bZSG9<Hur}0HcQGNHRU!G9!e&=M9z}nQsB>-vns?h=Bx^NPtK}ZUnpnAzhfCN
z7tUFg)Dv@7g?3=MZSboko@_CJ>VHQwf3ljlSUGHLh|6cayK|S!!QM5ac>O#KvPCr0
zVZ$4g0|7{OFZR-`DW#suniB0Nl>;H}B%FK0a-b}QnZt!$RSG>*b)*D+)8a@e>P)4;
z3X~Qq1;>;}SP0M<Wk-3$#-g*S8lPAlapLOLN+8XT)eou$E-oCZZ&{f92LzR3F1(jf
zasm^Vkd8SO0Wp`+PXZ@@-C0X-BR34b=U2=h*r`Z~dH{ie*c%t<twnA{Pepf6J+%M-
zQ0h4nXW~cfb-O({u{9dLMUfxX^mS6xk5Buzryq{mOm6lB4yU&-6YvAYe?jr0rO5%z
zR>So6=k)8dQfT?J=Z{fJYyLR??)g9MK56j01Naxf?vpbR_aFIxbKvKY_oesR-b8`;
zgZ$(rJ<;L0>{9%(ywkc&*A&`z_uJb~PpYV(=3wL3_)q?NeS7-rNh)Lw%wS>q{m-ZP
zz-XAY$4Ps(ftmjPWBT&+v!vR1DSjpYAvOK|brLJT6q5$UCRNrF4JHMZcv$kYX2Qj!
z=Ce}7iWxSQVN)4@Run-fX&rL_JwfD^7!NDV_=ls1O|FLbYLZ-$?Q7VX#HJ^94#n^6
zOrk77lq9XYL6Sz*yV>ZKE{a2g2_&k9shOgeI_^GM6$66ynci%o&FY_|(XjH%!V59X
z!TFU2F{G5R(-l+=&K#qXD)^66B$#n!X!Uv1fJk9>6y$Dy;$TpRVj|Mh@+Yw#Ryx)#
zv`lo1Yt;79>=@Nj?VGRP=yUnz9ki$V)afd#=W>SSoPkbPDKwvKv(LdmIj~Ext|@3e
zEAJ(RCTVo|DCxVhuzht1D@v})!kNXWvO+I0WaCN<(aVlSeAeDDamNnR_eMGCYs%q{
z?}b9+SZ$Gij#n=Sp0tiu94iD;C$>bj$+21pM(-c5u$;DP)@Eh7nA1(b>-?z$%t7mm
z{Ao@~Wq@X~E2&~)(Z;Sk3oFm6Z|d0{+I-5Q4)Q!u;8+CT626Rq8VBi~yC4q0uq}d-
zbeEGo`up8DFlS={70w!0jH=VgPo5=lR&S*Naxi;;Skgc=xGr~sUG6%QT$suMEbb^b
zNBSOjySBrYC(TL8Wbw8MZI9bQw&bBQtdGvbSQDJ|Y@lUPG`mme!x<GYsHVaW+;gwS
z4)J+WWucW-sVh~DLW$KN5^A6?YLrbW_-rL$+B%&fUQ<!Ba8-i3u(bJTVW4^l7D^mI
zP$=eqL)7z#!JH(`d@9%cye5uAG2gyPZVMM^Z@MDY>rT^i)xK$&>A}JiRK7f`!Lx*I
zXP0`@q4iR~r1`Ozdr2tSPGq+aH`kNKVmsV5W=lFTFY#cz&Wfc>*6kdvD59PTLpwLU
z90z_srJ7o0RS#%PnuI=GbxF!E$t~I7(B3eAIl5AW9S075n(tF!d9Us3Vy&#y^O%2~
zjkUUSm28`YrY39Od#jh$P-pYPt<&X#cuyBxN)vd<k(l|fW1+VYl>=|ffv+b=qpLRb
zO~!qZBJ}!o34av{eA$sREw*f?k$7UYe2u6jzT8%PfEOpUycPQMl23Py*Y;NLVHh%h
zZvLe^UNtRNOP8)DS@1AeI~TcEwr|>Tg&4GK-vBM28T*j_&3Hct3Rm=R=Xx_iaos^(
z-7|r;+~uOAEc&Ik+ZpaBLGIRDo!mY)DD1!2BGY-3p0y`gF8cAo{jlH~`{C1B;vLJe
z#c)OKl0KO2`}%=iINmlv@+*bShq_jOq}k!h3-I{{in(8x6zz5+tj&yEW6CH{)mo`d
z=(76_?Xls)8E|Q7Y_1!scT#RrV{?|hrNdJ6<$PUknAh^`#NDE6MZo*w<*>5sHF~i8
zioMS^18bHxKgLon=#9FaI@cel?20<whdSVl#;vH^4eGb&7IqFdFyX<uMR2`;6V@iE
zUV*h$gQ=#`uyVqV^(;#}9$M>v?vATCo1hTepV%zL|1oasy>i3p){eW>7FaQ;#)Gj=
zzcr})R5J!-3d(|##Xd0ai5Fuow{k(2x;~wMV}~QOCN?Z(tV#Iz+5j<?U?^oRwsI~C
z3%m?2XAjk>6v@OsVo@uqGp=HP&2wcnDSNmB$fr%r64cCn*n!4cMEX`lc8y(Eg6gFp
zY$d5I>sx7=-nXp*dk+<fSOxGQm4Xd+K8<$7#*6!YK_M5yOF3UsM`@3`qImO?XoYr7
zIo!~i=cS;yx&T&s)_lZ70gU`K&_n89w&qL^PpJdbdu7yt=wlP=Xv7bH<kNw`RyzGi
zd1?L3A3;206*0SPS8$%4sHH^e`lbbam{;-Dm~R*0IwF)x-_Bge=mva``*!iFW5#eY
zI&+S_?JiTHAG`kFFS8b4!)4YyKU`+bBX_*a&}nX~7R2{Z;0*#fSs?6?-HHKM&?Bd#
z_TDAHSEeU{FG?U}2jr!Hcg*651c_6dP(WS5Etsk7tIyM&x-Y=8?cmr7#!~}>I$*20
zzX6*C=%Vl}05W*ra_`=ti^VqDti;Ev4yvt`bG1=vH7*qxUMTwo#xiabY(oot`!5Y>
ztxtL{SOKBt8mGdAxkl)9&Nu&sQ>Ub5dt<Dcs%;Zz-Oq|fUQ~F0rG23TEb~*f2Gu3l
zR{`uKrIQrSMt<IP7%INgQhHen4zOaGouu7%fe$+f@!S|H0t`yI|I!$WE*e82cgz-`
zA46qS`J@WyPmH0U8$)GGS`X;`pa!;aWUm1h4T>yv0+rYERA7Y}3%YJKyD1YTDAK)j
z$?nkO<?f0+ol|sWO&6|{j%}lpj&0kvZQIU{ZSB~$ZQJPBNyi;KIsN_TjB`_Kj#@W$
zRdd$F^RB0^)&ovDnF1@5kQ!@CY4j0OoK|-BoP9U&EGSgnw2?<udo%Le@7he?Gnq{O
z%1^dVgT2W>Y=yG0WQ-pb6Kjo4tqU*tvYfR%oN;`3Gg~#YZgC`3{U&56jhDxwo!PtW
z<I)Y#l*Ma5bQDtT{%_%D(M9h1$*K!-=u`e6v|YIe@cwW6{bUGY;0AFBLhANuKd9K#
zyPc>=aC<pB(49~4Yda3bSbsC<;D$v!;(OuUfmIyD%$xHSFCK+=7IJSu%GA`GV%&!G
zO>tZX`_}uP#b}7=%5ubpYCv?KoB*sp2`6S4*vdB5C$7uN&={9-EL+-wJ0k1$;yF5g
zA-p60phscoB)o0RI>c+@>McNuaX13ddwhdp|3Sg~U%c{zg9}K!#g7ik3S?scK?&%H
zBjG@vIi-1nN}Kwn#TzgHl4RZ-z`{Vzf{&vwf%p6SXS*kRyt+)yD*UA&)&#fam<fiC
z@0-oRQd`;K`_fvIYO)dLK*ZVS7&z;|838APbv8W65^Cvl-jc&yHk@1?Ne}v|^5WFI
zhfLnfomfx_hF~9velIqJ|7V6iMJ6D;oqj&GB@k>;W`3G@TJkwt{U2lJ-*t6Vh5C=-
z^8U`Cz-w2qrhzviSyf%PrjXb>iAqsx2`+_s0Ehin`%(Sacb42Nx~NR>A$_JXS+6Wu
z>g=Res!U#|DjHw*%P7T55<T|~*<v9@3v11rOIqTs*i0M6b;*p6;a;eXQUY+|#%o1?
z&>O@~eQlS9<H2LOHGbO(F0FS%7FT__Qo=S-`qkAI&rP8-*PKwD^lWckOBp3<dvwvv
zF`hZXv`0)I<~bkaMB~hxFUOr{-ZiEP?|uHSI7Z@T8?sPnly6s!EKw`2HH}Kvy&K|_
z_@e8@x414DKW4gcv~DZqPY<9`SO^*@66;ehx(r91xR<q9sOOMHFT>~3+6nzUiCCJT
zDciVstQ^Epp+&zs@g(2urnl%wfS;GsmNu`J)8}*6)u)4C0k^q24=|EJX-39lTdT8Z
zj+U%QXrGcoQ>IQPxpG`3c!bCn8)?3X;1Ym5sI!2Z5)H$-6blnfbO4}GV<_O$N3g}g
zTX`emQ2rC$xM5ZK!7#a|aSSI&ZdJMaToKY-Ib?#^4&H3Hr(Ae;iY<7+H6)9XlF(lw
zqxTKBflNy?l&FGO$<4}g%H=09S<SFPKF*S##!5GjSz)+}<=fb2zBe8|Um7494nMWf
z!$hQWpaa%eL=76q0R=D}Gcpm>PXhP|KD~c^1gL&*23#uP#+v?kh^5@FMZ_?n)A571
zUSb-?@8gidJ;+KfF6jho;X%|%J__>f=9B95`)*Uz@l#Dwg)wp`lZBT4AL%5Mz8E~o
zULGO?wi5j=@Q@gIA=L}pnOlXrl-5W+S81x3G0Bh^HMPtR2zkwxCR;HUW(-W5w46on
zsV-*x3RgD8|NOeLmWjllnsv4P`Y3Zi8^M>V60@TBSZddY#574~Y-!gui`rrvnLkjw
zvOb5^??5l?2+F&#(J4sKx`s`;_>lv2YvV*$Cx9&yk%^G<?44-P%4%O4Y-;Z>@f+L%
zuHsXsrg75)%n%McUH1+YO<Y=yXsC3Y>;2dCmbWES^P6(k_7oh&f_uC@btQPOv2mAF
zj-Nu5lHOSJ;~lo^+Mdh$BsM8wr5Klf3gmr+B=wUX$LDK>F9`tT1#<nbI!en=V$h^~
zMF^Yfa9(q-UGVd|YjGQ-`18TV*r;mK2RrrvvHY_DrxOw}K@UhspE!IB3vJ`MMEFWB
zvh@0JQ<7thX$PSj^CO?s2I!QA`6l#kB}+UBi80?fWOv5pFMq3g5(4gLmu6W_4oBaf
zydo6%RD8~`!@1qU?YVVj(Eb?`x0Vz`Z$~b}vBbEvS2{;qaqBp)MJ49rqsx2OaOert
zx)w3OqQmrk&O=E^xBD~|qbzKjVsWZ*#oi{y$Hh(Ys!Zv~du2Eg<uA62&Rv{T_bTMd
z3faBJvQ8{vE0Sr;Z4pZ3P8)jx2$2$%=>Tt{Fn9ck!Y-y}lXWA@@c5n&K5To-kd5}Z
z5(OFv9>XAQ=|jlccOzj`i{KK`5D{w*J2$g`A;TG}imHo&p_B$&L20KW;2x_pk-J!z
zy0`x|94p(V7bIQhK^!uhR*1;qgwQCg^)s7Ik<Q^Cx(X{Uj;CGa#xj<mT|m&`EFJ7w
z$s2Bpe{{<b24IMIHdZNDKe6Hqn+cWcILKOSn+v-)<A+%L#MNFmH<6h_q0{~}3|0kj
zrO(kT9#?pof53us=8d<gx12p_X;ORJWT1h{s%@iKT(qv|TZzRIUeqGb!85JRJB=c7
zLB%pVH@?{zyj7=v>X+yfUZe5s6rFIl_SIy23-zdsU3Cx37v5yoZsJ;_w?heq4W0Ef
zr#lM_!4}r%US=@II=fbUKC0N<(PaQwnQE^W=r%r3@t0_$Dyz&H{?w#%!zZ(4P);$)
zhuo&elnS#7TJxtk<y!i@---cz{1)H95{~ec!Cl$XFM<}b17<3;ia_nGeY&nlWoMS0
zH%x(LFFQ{IZI9>BPOs#bG6OOBPG#g-mXWdlXiN#k|6!Ex3TZ1|^7nb8#&!b8L<_33
z3=<xK>nkQ~#Q65v-^*7ftbN5f4LB2eLfR8o|CfkN{t*(MDc>Kbg9G}dZ>Q)rRuiMZ
zMFJ@?_-fn_W0lcZS|&;HLcT7|(q@=<Ut7hvufyjGwsu_RJh<!fI)4$g=Z}0}L(YA9
zbti}nCdFxYX+fb!<L0o~vIS62qHihi+&IN3s3^F;aV9*+0Nl*(H}xz^Z(pRQZ9)ke
z(HL6JwSr(VmTdPsA+)2TDZ8R4@qFvPApS+2ro=wF2{?ccCm1{Ki-7qj9N;<qokhgA
zwL{8Xw5Mibe_%~{TF#@zp`;~O!hI$%y!Xj<L)s+G`!p(}|GHT=EJ}Jo=niHM(WKtG
zzwr@kjN}5A@?vcm2RDu7Q+A#FuN@GSM~n)1tHy{1#>~kGY>=Zy`@zfs9Qge&+n_BI
zPxAi|%FTlRd2iq0Mi41N2xw$5_7lt_bW<$$deBRiyBif>_6c(j8}nv<OV%aQi&wRV
zO7E8kvx6mrJbDI&JbPuiFfDMBQs5lg{ExD{{<DJ7|3>4CykEIL563&^A3EkGoao8?
z;6Lrp@DzTMX>Nc%VD+B?ij?-A|2?To+#=owI~q9@1>c~xxxz8d;;gXB;!ATjL_4V$
zTFU^vA6JnS7@RH}-{*aAeYpm?{yygz%D5R2Jf9aA$p(t3>CMKd`PiH+pI<j;_e%<V
z>x1yetlux6Q_Fm>T<gm})>CX^fB!~L0QE>-jg#s_x25hNk;{_;ur$^j<hZ}XV=XT*
zOBp{e&zO|c45hHboj)-xm;B?<LAX6*pZ~xK(oqyZDtF<=1rWtAXHkfLF3sFF0J%bw
zGg*PME{P-(pZ#3kG)khBQP5(ZnDb66VdKHs{VcG$Srma6u&wA<slQz?(8VPeN*JZ{
zyTL<PSe0NO((Fh8m5R+tflhp4@(T*XztoZ`ljJ|#-^x4Uy?%D+^K>7T1?B#7I$Rq+
zQ|pah6{Sd5XO;<E3xAK69gOK0_-#O-aZ|VVvyonTcKC>9m{RC3PO*jeC5Hsg=0@&a
z*lwXhV}<8JiC3eMkdTwv@M2JUdv?=SY<E?f&=6}N$O2Ol!0q;hY@(uD@KqwA$xGQp
z!Uj_NA!Quo0cUeo^F%T@w=GWd?-QJRclM9+P`)LIkyj@2BoMSOYbJ<!w|q?FyPsjd
zU{UNZ>^-E+W8rTcE1cnn2Vo}3ED!W*(RV<tg^u3*Z(|B=_tH|ZD7(f`st$BIFAR;q
z73)qEDc^n<1JD;#be&p6giUVF6=h0fJ0oz+GLB2<k{lv4(iZucmTs?2PTu5<OU*j5
zSZ1yrFj;D39bz)##=%TV%RcOkOONj&($6Tbhh@8bXv|3iMb<I4AQS#_3@5gMwWt$3
z3gP!mVrGZamlSPic@z}<BiL15`*Ea6r&&_=ou|!d0g6h9+%6$MISi-qKR$mmtYEtR
zs6|ig<-8q9VfMfzPr{|yMA={Scp$5|JBs20lPhtbaAN_;_sx7p!Jbj8ZJOl-Kje7)
zNq9#~R|xU@h(j~?EWCM6Ltmj5(~!~2=^vSOk<(#aT>TslbG4Q*@J&wrOE1I5HEO^s
zE;~sz3J|JxodL_Tb#|q3&g9T?9<c`7+Mn$tJa8US4tDEmx^`b(YgS{kx3nm`oyXKv
z5Bn<KNmn^Sq~KIz*gE4sGS$%7I-lgKMYpR_6zkja^N9nd%PEcaiLLO39fddW!sL|1
z&GB9vWAYC;$q!4~hDOtwSQ4YbF%)aamoYPB0oeLAR!stboRi1&OhEd~IMxyHyUi>W
zbj28MrH(FVG2}r*_R=rDlw}p-1~oPcZ6d;zH>}NrKkU!_{7~X99F>UR%6uBnyGyjA
z$C$o8R{q(MEMu1A)3Ge^3z!;c+dzThEDJO6XgOU?YSCyN(O5<`@jiKlO)*0_7j;*l
z3V8ZW{LHOjJ!pHR?QYh8PEB9yHJIcP3O!)0?k<CZdjGnUqU$8%;RC9f_=sD|YPCUd
z)Oe;-{6{94%<aA_X4g&LInBk$KS>6IXD-MYUF9{2>$hg#2~{HZsaMwVVv+)>b=7Ce
zq*J1G@`ALU6SX0xYh6LPEnm_E55qK`1K^N)n{%|{{G$9vw}HM)3%ZyVgG1{pmsY8?
z$%6k0=jn7ngXIEmG#qzsh=M$m$_YzOo$8<rB{7y~V%kILhEsuuNex>wRopbTr9yCb
zM9f-24F<M(h+cym^fZF0%-F@*&5!kSdebpIw#LI`@0*Un8fo0^Ov&2Q!cy)?7Qizq
z(k=Mv>1kGC(xlTW$zGc&hLt>#tvl_s5!M&trfds03tFW#9vqs~4h+Lx+^g3W>Rl>}
zHlki3)kxRk#{{dnSSsY|nrY@JI{!eb>1($CT%_1;>QS-sw5{Min-qQcqVV;32Vr<g
zg*i0@DOGdTF!0#wfj^M{sdKI<1hCdCjSl%IYWPd7@C3cY_(tx)p9OD+)V{ANA?VtO
zXKqXV?pxo4h7AUhk1cU0AcW(eNGhLaTE-+|RK2-*{}MtrNU#U{QAO1aWfZvdxA^)u
z{o|qO^F7uTA(f9z!&a5s;sA9A>F==Vu;X5@5c{BOa!@4I8nQpTk3hwm9e_Tcu(L@<
ze|Xf6a*8I!hiGbilDOWreX%1koytAD)X6#hT6trQt;i46A-ro8w!eICAQ;!KXXzVR
zG>^_FRwv+i_Sdc`NBXV<fcHBT{6BXYe9rtvdyA_^$8VGhbB{jPBd*T&=wx87XxJ0z
zb9UpQhn{Wd#6xXe9XA6@p@3BHvye^&4!ZJJCvD3WMD#?^f|^p77a!~X7#zZdG^-$A
zm!o&rkZD$K;THDvr{PmOa#ay9VIOdQ+e+aEd2_3H-{*Z2OTU-sZ{`_boXyE!H<bqv
zZp&@wmpdakhA*}FLd{EGG{<vTYS%~o-RnN3xj%x1RwF%sEZzIS2mly4L-9O+=BJgk
zt>P{Vh7>F9+%`kQiO@uROpyqWQfJ9YJX88mdB2(vkOZY0a)^BX=5J_i?r_oZDtgRZ
z>2Bw`ka5ehS2hRwSm8K}pw*?`fnS`kaIj|xgHF3wLJIXawA%aK_kAA4nJ|Jlxil!9
zKXNzRJU}C+BmcGDaRgA@%#i-{rqt5u(5(ORz*%^pJbjaLVJZb^S#<(yRWzlL<ymy>
zAA3ILZBKq|K4CS8Pqv1(E=8z(<t0(uhkqZJC!rm~k=W^gag8O^Vu#CiYNiwVZzEFI
zW+=8Um0Kp$zweBy0W>61t$tnhdTuNlD63jZqqcyyt|aDD#sS1Pra5)R&7V@rj}d>P
zMc*}z#S(L*5E>$J%$zvHWkqf7_hq%;-jtHj;S{Upi4XPpwguxOhh*!Qh5xefT)814
zN2IV>0jvwbE9V%CTQ^t8%uy-kRMF3|R!H%JJh?`})Zmf%+d(~^y}urxeUn|O_rjXl
zo4U9<n;F@`ngR8ke?a}<WCALU@uLHss#z%jJ=!u3$E_GU?;4g?MU3I$4(7UGy}6u%
z0pOPBVgamG%`EAKO_tjE8Gzk*bg9G?(TQ9lxYsvnUZmU`-ioO*B(%KVj!s^K9<L<p
z#%ZfYV{d-}6hg-jvbZd9g|rA9sipg}vofC9&;hrBf9POT;K(;(-rJXY(h%$*N8vj_
zr;4%~F!&g8%32rgYG!wo?*zl^6`@evCsV~qADlWm>#KHfbp!DKdWtHe>-Ji-FYarI
z5}K+8Boq@V^$bDS!@e*44i1Xj2=_^RuT^U{9G%2i>)%JMa42DGFN!}dPQ8tza%G=S
zbKvDGd%)@2-G!Vw_lWIofplN}#UBW`T2@K4Tud%kDQaNYovv<MmD;Ty5z#cdVV=u%
z=)O<4q_Q$HkDuRJYqfsbK40dA!TsZd@`O)^bL5|+@yC(l=7D(J_EMTAdZSrhG+SfG
z82nJ)R5x0^zANnvu<?7lYm@>`gBZt30H-3|ncu0$v8P8H)_mXo%9{S82g3^>0QAwi
zL{1>-3blv<23dSy*8D+bS7@5Sm9>zstbU$YEG_hs<?-(>!`=dMTZ-6XJe6DTHwhw(
zqKwPj{*AtWo%>4^sO8G;jgGMLx|AtV>XBt$FXK5QWH)@2!7*6QT@IlY#?!K~cH@8w
zVVvVdlVeXAVwj09$*SV(4!i&W$Wd)LLZOjV&1jD2INAm`Y&vl&%h2y@$$Yb@9>_z<
zPrTYlI4}$8R2JJvy(Bg}6p7R$M3Fq4qk1&wS6L>GCp0ZJ#?G6<+8-Z`PN?G=bIPo5
z=C~~4zEdwc8r8}CoQrvBxrEo3Zb_dGrNHE60D_zyIf}U4h)8iN7oLF6DNgr!MNJ4x
zHjAJK*}L1Tz`6mSm_M`)d-fbo3Yb<bTjWoN1l)7uZ|2cI$2PoYK=jM(-KiU#D_dNq
zD>V?Al@)&pS*{y1vX{Zrcsb0G<=Fd({n>54u(bqM^7@o?RaHi7y~RfaTXoA)z$^c2
z<kOw8s1Z@?D@vm>jq3ymriXjcse6S~#)(iM6K2KCKV|ah8O_>_hbZUO++UqrZXbXH
zOIy37r<P;EqeX94AsF=QZ0)+U@HX*FimHws>{}gYf0hLMO;$Ny=QM3k$Z}L7OyZb_
z@>%t{vzdo|I$E=qR|(!tc}AA{XlqqYJU1US&r)TmbSB?l>(T&VuS|l~m*z!dO(P>&
z{1MgCHA?jVA$XUg8AGLDdMi~q5}(q#$*s#Oal}-a$IuFuEyI?dXjyMWS|b_WOj{uh
z4wW32dkki@%Q{<2yf(3_ocO2kPpQl^h*t?jZ2k$h^m!FteaTRggQOfp(XF$MBQ7#u
z>D%DgpG$d<NfrZE@7mh(T;$v~xA82n6DY)z&i+y&e{#|frJL3?B#@-5Rxx+0z{VQ!
z3hJ}Ft6tJsV0Q}!bY$eE29?f*xz<a*+R=^_AgnnnOv;N~(*?sVD}S-FHiq=MJajQI
zHj-j%c!(1^=>NiUCb4N5ZHhq0B;LLL6a+`5oY>&26Nm?hNnYxX1ILlEjTiJ})8}QJ
z+rQ>CzCX`Xev0_BK2iy+HbhC2WIW!b*iS^`Z6oqdE4xbJ%=$I+o*5y^K|R-GyO89K
z)sM@`aOJ)V%Sfa%;3eHoQ?S8Omi`VqI@E+U=T~SYLouPk>}B9iy0JeOlAwfwFX+09
z>2e9M1OWxuD{GbZXp&*}K=FEQPQK!ChX=c&PcK_NzY-CnG&|S0L&?Jc+tLg1s3?f*
zq~wM5le0s*$+=*><nFd-YSMl$(Bb&-jY@fs6ayJhD%wpE(d0z#K3jprY)a>ev)(w&
zrApVfi77_E?1zW#vPkLz*9T3vBzgY+QKas$3ylPDjdTl80)A^dE@USCG$)G_YTsO!
zi~^_aT#Fywa~VsyK!0a^;!rg;YkTB|&dEG7{j*F;AyOuGs`}5UgUPx7CG*eN-OaZ8
z1XJ(K{utSLO*$?0HI66Lt~2)<x{=hsgHNJRzq**P=}4aQ5>(5PBbd96?436CN~_%=
z2hdhU`B$dZyj67nspq<#VfBiNw(+pI9u~-3++NzA&DV>m$1f;8<e91pbT!(@@>VKj
z&)r~u?dH$&Rc|c()$1ir)lF5izYSw}hhoN6ST%sc#(DL(G;<+yu__16#FIli9!dBg
zh-%ZMmUp8q_zj5vNjURtl&h_;AeG@>4`5TRaz+M*o48qLBLx?rTH~DV=Qn}T*I$jO
z1Xjh>`@glVP#u^k`<kVl_`VhF+fk1-yf68=_pGnj5Jg|KyQY=O%${$VKstN%g&*M{
z77dPZY`4|r)qVBE!NR-9|2YFHXzT|j4U_JXpaZN*4ZfLr`rc&bTKw5=pe3D50Io|E
z$OACVgE1pGx6b`UZ5+HbwSwATA8$tI*B^>M>u@Ng_?~;#xT9D|-1<5>h{E+h?ws|6
zhvF(%E{qrFKrdV6SD6^pxo^klkm0VN6zTJ_98t+Z<9BL;(TH>k;Vn)QxWK9?XZODm
z(sl|rnF5a$1n$pnJYVQAWn9g_0nsu77)-uzYY%IC3{Ww8`1UaWnr4BZCj13_PW%me
znwDpW6XVfZ*~{=I4JrQ9nDp0fa~{}e!{C`d&gCLg1%KUT!5!qGQqPX{UW5WAB?4ZF
zm`RIY4paUvrG{(fzLq!_nZ?I8p<ncPXGxrZp?!+A_a+QUDL`s0KHw@`14K@S=aQh4
z-TsW;?|*k*;jeNVA-lDn4*o<e$3JOIz1&A}(E+%4d0E0@eDaTs=HE+fUni+>wkNUH
zvf&p5`((}mjHT_bKO8Fw6XyfE9s=;)nRrwDI#g^C$|~2g*AVq*9Zue8v(Uq?&CtEX
zwn^qllYGXc{)%5mbHajv$zz>!j!u^F)Q*8vyH5|z+w{$EK-EnyR?@SyGDIRa+}_8V
zF|$5cTWX8<ficsGI2*vjZn!jwi^BV3hi4}zcon322zR`Tmc9tcT_-I5-+B!vGjLx6
z0__J2BM?*th611|6Tiv%UyI2LQVey}G8{I1!h|gB!>X8lG#}kYuJ`Y+?;+*JBPhEO
zG>WSMuW?o>KXiel)$MqBk36J0;7}Qh@~K+=o}9<)4(RUu2TUeb&J}NfGxYYc#q)F1
z>ges`B$id|IPi2gu<vRLP1q1-4*<fTkV`a?_sV_i2l#V*=MrqYYWT|RdzyY@=u5Ah
z$uO2Q`|Mruhrjp3<<)uK3{G1N=wHh4dh_Z0azn&n>ZUrY{CqPf(kTQD8+_Jp6N=br
zu-kbPUZ;G8UI!8bbf4Y8(a*;bJ<pDYU8*5un9@YKiNogb^Y?P*E-+SbYuSzdrOxd8
z@wO0|3h3jH-^?!$z;utvEnt?^nadgp)rjF@%1tX0#uu8=ze`#z%K*w%9JDlC=B|87
zJbAY=KXw+1=%c)yhr3UtNqvp&s1HV>a-V)}zTM@KIG}}%4e;<2t)fTH`*2B8$emX5
zoHdd`dI^Y83*_WFU;I|fTcEl#LRo0XU1%351i0|%pkV0NjKK1vE44Ee3Ql&43o1a=
zCQPeQC}T4})FO>we&AwMRbxJZ!27-sNXyX$Al!qy6eQG!n{h^<1?>BA__8zhKxz-v
z5iSNTBDcCm45<xk>Uyz|z7r-<MZn16MZj3StIs!@bVI@7#2-DpSm`Y+jts5C`n;Us
z10LDlTG**?wmh|??0D|Gr4HOEgr~qh4qe6AJLj{Jks<7O<mJBMQlV)CIdrhkATdVk
z=U?1YvqPF!o)0xhM1&x&o=1U_q3w!y?tMNO<OQz-`)=}qIsuLfo?K=yTf!zJ-MQ!<
zCd!?cia{~>Ot2(yPCU;$BFyFjWO4Itfcs=5YUU+qJUV~ainIJsS34%yDvPidzx#~M
zcoVI*7RQ~-GUB3kq_i|p^-tC#)gylu$y3Y`bix(BaB|YJ8Z$9>sI>A};aIqzYYoH0
zO@^5K@|lH~GI1B)Q$BfZXV95{l^npR_uN&yeYBrjJ$<VbOa&9QG(o%0(Qu|P001X)
zLM1R73EZ+D7KSt3ig^4&Xq#!6D;U*L?G>JOt{a&w6LZp|)t>`N2~<J~Ztl}^O7`?i
zQGjcnje@?q3GYK!%WgHy8Ds7kah#twxrmW)q`!?a#qrCMt!RAVFbb6nvzNE>(>KDP
z&qD58xpB=ahdJt8+!?}S(yOsk0i)SnZ{XTy$XHT;!XUkGLx{!wEaNV4Btx<zU!PQK
z+_Wt*LxVRpk0c6wO=JJ*Xo^EoJJX@#qN$X5kh#U>$d`AZ=u<$A1}o`V+YrHa<6-03
z^I<Z1=dteAn!={)!gj&c{*!*Zf{ZPpgo~}2B`Ia1l018BCAw)ZB-rpQ0VrE+W9qvJ
z5-=}akJ;GYqIk}*+3&LfXET!BF;<<ZX+u9)#TlfRgsJP}vHih-Ff(|%<i&N)IBc)*
zdaAm?xDH<GdG$hCU^a4M!E*`Sjk6z+{Aig!`z?e`PMT<L{w!W+Ws66(uS&)usB%@x
zAO1XIHM#}mqGq)}D&KN42K=Pv()QRnVU@iiUHIR>|9b*5cVAkHeby6iIGVkEV&7{0
z+;g~U0?tR3#(H-wdEQY*4wzWvy0|Xk&d^Bl<Jfm-a;~|lsc!EZ&{ONZ@D2)(J+uDe
zSzSt7d>Sx)9D6EK<kS>DH0~jhKo5Tg*VnyVZ@5p!COmH9{fW<cn9x|IR$+=_JiSnu
z{2+bAZkwW%j}xscfLOpV*Z;0fP)^{&G$i^DcFzC67s>r|5CDq5{00jGhO(d|B8}pA
zF?>~*ba=b!KGKaokr-YKub=3yLQC^}H+;<*Vdvfg`Stkz@XjyI>pgaaNE`6-p5SF+
zZ8u=^LB)MUR*MIFAR#M^-91PKdBv(}xdeRT(luvJ51UI7n@|Y3L;*V>tw^(iAp-MF
zWp3E}iyjc{MgT!?^cvnW41!#2=|vNm#ZHtuqphy1wzfIEiC;&zk9JFqwZRzT=lgTd
zGh^bM&{jPk`DY)uV`0P6$zc-Oz8_b7@C@Jf{tSjob+YHgOWx%XWHp5XrE1Dq_3~IX
zy^@i+W~-!4k*Ii!cvKR<<nR`YZV!EMrSc>@Il{)Es{jYSi?dW!)~do>8afzGRfQwJ
zo`O~+WI8x)8fUomm%I;>|KN>bBNnBK)+?Dgy)1Vpj_`A5QQ1LQ^5KS07YJF_S*}CV
zeEz+^+4o@`!neDcA(XVYqC1eaPbh|GNf?4Ti{_|vw)@j`cX}FMknzgw1%f=3$;h6b
zn(bv3$PN&Hng4MShmiR@HV_j5PlV*F)UiFiM{xhbPX@&>qbJ}O_V#U=I0bDSxtFhz
zMGV?|NP`L|IH)v6sLt0fW95(Dx~;5jZs~X#?e!+eY7!+OBqCB9Lv|G#B9~u1AUvVE
zDHQcpBrZaI2<G<c@RE1wh?~@X;a9<_Ox29!18abbD~>WK^IUL-!Ettb#$MV<nLB9c
z+R!-nUuJ4|RI5h1iL8ScD{0CDFk~}1v@3=@L&{xQv@7#n!6n);V{|Kmhn6KWB})0m
z&OLdIu+V6G9fOp}qha#t_uKNs4FS#{$(&g;v>iEZ!Z@^X#v}-jtk*h8LnEdwQz=WE
zf!lzF;@S_ng9<}RE2Q;Ql2Mq|czK{z@sVc)RgrI<ru$~cyBg?;VnVZ909mIjL`FEe
zt@$E@R2#PCpAfNaV?F}QDcovUk(vg-G=4}6Uke_YjGD}nyqScD_;4cN6Zj-9alxTa
z%9F8x!KN3bMQBluy2x>w<ejMo0>Nmi<Q)J7R_D*Z{3of|Pq7^%^3AKpQVaHIGJ~CJ
zpIB|n=A3IVAL3b4ttGoHF`Kp5Ca;bhfz#0Ez6}mnTAb0Hh5tSl-eHjtU_NsJ)>&@+
zW@vDbFZ6BW_6)u#>Y=Kt`ypZEe~5yvAG~bz@_7qVb;{vNP5&yYMSP~LC}VB|0kNS+
z&`nL!frngGlMmddL(F)$<&W0Nj&k^P8f95D5-~a#H+ozrnsdJkXPQ~nV-DHd4OaSz
z@WawhyEG8U65`|#QgtfDYef)7TPg?G%n@Uq_2KO2nv`Ma9vN0kA!Cv~md$xrtsyL|
zRcPqujml>1IX2gHS5~*+JYB^#0mnRqNLAY~S2_8a%eNUtr=hPsx1JX0PPpb{JR@ES
zc(7(z-3&PPTma;}fi`S~QzW`7m7(HO$490LqrKNR#u=zQ$ioe$NZ8y9^Z+2xp`?d#
z$IaCMc!o<O%;3=#$_jk0l0L{Xfz(_R+Sz08z7&_p;kr;>`>R@|x&Yk@uq9^&cDV||
zb%>pJNM`A6vx*rIY{CPpD@A8;+kHv#opLr58J7dS8lq$Orh&PcU1ampJRN`EbZ?t!
zW_DT-k3#MA>g72XU6xkWnsNW0)Y5y3?8=r3EY!+MU3vyL6zv^oH6YKcP*mb;BW26K
zcn1v=EE&Ef`}(urm)O7p=By-j&$=>~OaH;E-zSV7S}7qhh~+I&fh1u%Ob7!aJy3kP
zv9B?`g1)tDb{&eGN0z*(k%A<uhm90UL8`-U7gE}s0hdxZLn(oNzfPC*6v7t+9I3fZ
z*BqQ*qA>-~9II)+$>yJh#Sy>Z0jke}HW7SWG{mLFQLcnsL7IpFRl)HWk_{M$b0SJO
z-M9nUS<1Tv_h5;Iq#G#PTS5d0W(2!ap{e3ElQt+U&s)y|Nu*;UVLpBEq%6=5vf;~X
zZ1Q*%#8<wdJIF%i$?2}B&brkpDnGoyo!uiCU?JzyD@;wPL#^W%`{Zo<^Fy;ipfg_a
zkbAf8mV!_CB@`LJ$R&<#?pAxb#f@Dr#aJ)-Qw^&S5qwbf$fBg1{<<L{Ew|>Fi<~db
zfg_ie(D^)?4YbrpRiHLysj<!}%{cokt3jpzc*mv2d-h=pR({QTys9I(ADO0S6*k;R
zEUyKI%JXCw$aD0|VFe)~TWvGxwEA#TlB&T~oH-?5%`*0<YqJp72_p42S#)}lzhMwL
zYANx8ZnFBf&0m4C^;(esxh*ixe{;wvg@o2+6GPB+pe(=vT^=-`qYFJH@UH_GIN*>I
zzUP$Y1FQs-Oc_p*GM`$_%Fv!=+>WtK$RqEK@$);pz}5=|W9(MhPx10Mx)|SWVhZZ^
zho$TN<AXb&Ux%h`H!S!L>)H*%U$~jP{MUk_Q7}r2f{riSFaOmIPJ4$fMH`zAuU;hI
z5+DU!AXz{kXg9A{2q|FS{p5Lm0?_gaxvmc8#PV^{;%9Q`lD4>oFT=fiHd6imaX0rE
zpFFEFPu=O7MSr(B_V}<XB|mPhXj$?4d`i_L2sa72r1=v)^kl8=<o!&c!VOjfBus`!
zLbV?vlWBOgqukbeWkBqo_V-Z&G11y~n?^3Q{#|Nc^4xQ8R<^J@&wt@RAc6wmM{s)l
z3d70OxBwwr<f^!kkV1HAk}z>H3MVqL)3(Bn!wAF01no0ImkjS|md`B3;)>3bjH;Tl
z2Xi4ORP-C;XA4A5RLgM<-4^RmF~fzOqmWaTZT}{_XVG}N^xNU_gw=gE)`UlN@I_6-
z24&S%&`|YadiGE2=mr;SKq)`V^NDVak+OOzX}VWug~$j4?Bv~##}Dpre*aVwhTE!?
zFH-;1VA_8Mf2KkF`r`dX7~SdT2P!-5Um3}tvvpmFO_2u9{z&&S{eIyFt5gq0dqBN8
zEJJ!88n&5RGMw?$@C5l#p^Fr9hEN>`C~*kQ8Ou?oKw1*?%GWLcy+1sC;-Tb0tv?OQ
zrG>F1%5dcm$6y-`J`ju*;+d7`WsiCFhAKopSM~&lzb35pY+rWp+<3qFLYkb133}lN
zTwX(!NxFCS)5cdUpp`rIl3a2d@I*f0`_W^aErEZn(Lysg3GGH@Vmv|gg%5w|3*%bh
z9-cbER&XHp&~xzs6!R3<4il}&ST#KVET=Ap&*DZ?sKg?wF6&#9Vh4*|+@D_K^zMvR
zEdO$v5K2s?b|62Vu%m*ft~Y}UW3)lXreK>3Wkl3V^PmuZvxgp9XB1DX!>(k+Jnb3d
z#1O7N(9DiC0WU86j5Vsp{|l{-2@90elIIY^o6f#0^9#=c;Ho|R<uTFZI?(-blSo-|
zHKw^zLo?si`;wdO;zCAg$C0J6;aS%Gtb91Fj2B0-A>5+f?DcMBtT<f|db_q#CXd3$
zUDW%O!s4AXyx#53MbD2q)2Cb5>lS02IFe3c+>9<#x|3~}%@fmblv;%ey8E1Untkxr
zy0}H2wF1Wsu;E8z41oo2$kA#!8lbL6V&GEQ7~P00n8IQKg(aazh{q^_se6u1#7{NJ
z8n)~b<}x`fT86&0Z%12nNw2p@yo)znEIsl4HQ&cWl#Zt_LYk0v*-w+l2Oa8i%b3R}
zNn{!G<T8ItuP3nqtB032yA7=;*G|y6m;{zjgYjb<5MG;-8yr1lOb;SZ;EI)rM7bS0
z8e8&DQ$eMZ1!UTuLHe1G%W&SdeL26KNetnFycyrnU<Xn)6KO{2^DOG$TuU78hn!`0
zMTJTf1x-oCl!Ai%5ED)xLzn>0%uZRn2+N&5*j2D)<1TeXuQ1xv;o3Ww1PhNC(xD0O
ze$c!Yfcm}eTaC_vdbY#^UMI$hRkb@<tTuseFp=j~m6Mx<ZSLymUI9@!GlZbc{f28;
zTmS2!>co_ToiE~Ypg)LyiSMf&s!Q+T@eq7AaTN0~8*7YfE}ldI_RXnV?T@ARSDObg
z3y!P6OMy6pih`A1QGPatd}7n>aT@<=mJtj|QSOzGyyS$Bs&A}#xt;kKPh>J{l0H&K
zLSH5y8foT*E374Ji4i}t#f&^A>y7U3k@jF_d3GdRCh*J0utkF~|Ly8z)d1({@zDG2
z=NsGx#Q%RQ{AEN3iYFsd0w>;}a{+hQqP>^5Yrq>MZDGo12Yp1HH9J?>7CnCepD#Zp
z5>sCt60dHGz)#L<-U#PY;Llzkh#ue1=lwD2wcMzhGULmNrZl6HdT_A?o-U{=H=$5$
zAI(I&WWAg8FBaP)Gu~|+{s(wya!$s$Bu^_`*IH#>Y4PGo-cz?2BCrg;8~}2_eNn_A
zs)EX7gsD1e*Co!`P7^O6>3~!jxkBgZ6ufzr(!cg;3BJtq{>4<d^%?R~9Rc++DIBNk
z0bZ(c){CA*>f91F)n!=!zzMab3il;KfQm)VT8wx`nKp9N>mq(n%~gcso+@^<D`rnr
zSCjdmiO8Ay4?I(sKK&eN6X3`a?ibxTvkWax<eDb<e$Y)(6>6w}(n03~lgTuG$3e&f
zY&j;`LufD^(5oZ>XgQIRl+1-UDRBNtL->LLmiR)cgeh8e%?`)x?Vu+-RaQO6t-auS
z0sUJgk;WI1tkQ>e^Va`i$*mjecKoJjSb-*~%%q6`SNj;TK99~W1rVrxQH@qN#m!yk
zgc;9l44ho={k@hEd;Bpr;#pNFl<`_@Wspe|aeBZkjg%_N_*j!Ra8D0Zmw&?6uR*G?
zo~2b1v%r5+n-Nspc^LqD0DAH;giRE=3TxgkoO_&mN*gyBt)(-tRZf2Zy=^2;{n1W6
zf4We6<5{W_ADc{u0F*4p9629-#DA3|JF?zIvy*D?+*pwz+POZw^RIKXB?@DBr<1c5
zMO>6Sxtv`on(1SfYVCjET~r#}@XwqX2fibp54{+aAEd&xay@#<qse!k(k~|Ln-9i1
zZ5>D{(z_c8^e2wHm69o9JJ$^`hEdV->mBp*s#<>z;d#?s1Gd@mR}mUgFBj7#C$Q27
z>1E@vAcw#cRRiiFzm|A$wtO-TG5?MuSZ~p8e7gEvn=LRPXE&Putb$J?Xsiy3`7}{a
zSJf(Fxtk-<$H8bPiV^S4WSO!)%5ZERx*V)<!HftbOg$-X>tE6o;C{BU4_V{Z#I5<@
zX)`W72wp&-0lb3{tqn8c!npJ$w9);3aT|l?qVLATU}-y$Oxz?r2@l6)`b#xps52~L
z*QM>~Os(0)%YdQmJ)<)V>SY#-N4qVB(7yaao6w(Z|0sV8sh_EI$e!I0uvWVA9Ox9y
zLlYOqp*R@(e*eP?TjxOE>`+!Dwl`n3_xO@zk&-9L3UC_p6ebMQck%7Fc8f@0)<;lh
zvHWPemUC*3FSr+F#X&SngMOzp@;ec%M(_$O{&gI$wO=5?Omqh!2q2S&(j)9)b}7z~
zmq`le-lcUy+s<%0y2261kzI1_d)Ur>v&DJHDqj6{<JSTGsg*BK!S*@HMxxWv_1JMP
z@D>G%1ia(rL<Nrhg+>I@L|_`WZGS@}sQ?F5snCIcSs<u@R4On90OXrB0N;muX)?^5
zyP%xn08Y5bXRMIfI7eKJ^Ip>zWoMD(5ReXN!DE3}_CrU3%*|IaR;28g0Wtr#`{U=Y
zg%_#F;#A8uPK%VMQj)h%W2G3%u{g09OpDF9qr;UlSyS?3&a!4vbIbMDcQdZ)R5J?F
z{C?ZiAt9zv<=Hkf09P)0Qcm&^FBL`oqUOY_*}{r?f5ia)fi;}CR;6ywC5TNm=SEr-
zAH57y^^dftI_L)RKsqPyFF&majEh)4(Wr*<h<v9c3=|%^iN3jSpI(H8RcdA<yh5oh
z+JTtsFu6)IP9!y*Kpc^9S(B6ySG>t?HKB0j={hW6*eiY-00UmAz@6+^jXZ;vpIv-(
zGijLNFd#cgSR=_-H|^ISq<*qUt+YW@OqIm$9mpTCs4zRQKXwy5vnMptXpIAYPz9yT
z0qTtVVFxKA$m-=_i_l#Rq_t=)l{ptwm7hXgIv>G{7}J$!^A?xWM2bo_{z!K*Ooa+g
z5}~GU)TCtn2F#WSV=1doB5;bOAtf0%8|#d?=uD!MZYq}&y*4K?mj_TMwWK5-oVJsC
zEhRmGN-&5ixB}Uuy>-e+oycJ|PRYVKJp8NbQC-}eU}kGV<|Q{=LT(i|U?WVP*K}c7
z^m=@C&rXfC|B{vH(=}v@*sUj}q|>WjKury-=@-<Y1K#VC=ntH7S8X81M3Q1T36pGT
zKGj@XQ@(<Z9!JR55k<PwxKY*dE{T6B86UM!N$rp(9k&E4Zm@;;UqP>>9mz9ys&w%W
z9|xJ@2x-xq2G5he|G4g)OV68f7It3ik~I1iN{jO+TEQdI9{5k8BztqpV+)EkJ1xR9
zN<zV}3Xm5$$DNsGvOIF!bcG|?I(~!Q(tb79IlSrqyqH$N%N+Vsd%`(X%I>+-k-I7x
z|4*(Pe`6D-J3D%BwQbm<IRF7)Vf)%jr$%qz;#%N}TfXP{)mzipM^_U$+h|nmLS)?e
zFf*dyXlWG4{OuW)Jod}ZcVjd?IqhI+wP>Ec0g#Lvkka-kZ0k@v(MRbivPh(#{WyvG
zY<U;l<hpb*x6ZOn&?%-Cz2bBj+orKn_8#8EK*EM%#=rQx``Z_@Pw+M?^`H8fNm|VN
zo5H>4Q|n2lx;0&N6EDg3=L;?py24Hg2NWS`Xp4&~D;~&tJv1JCKC-TRN?Jb;L7<5_
z5rA?Y+y05fpE=qH_gUwKbj)y>`r9&<F1H)c8Fo@SyqXh-k50p#i=nqtevFg`_4LV+
zkCC!JQEVkEHkdOh_25zh-+(`mq%GKZ#)({9eo?v8$&DK`n*M}oRT~xlI=O9?m_16s
zD??-mJQq{iJnPG!KxaJa4Ji5mZix}Iij|-y%|~1hJGz$_SCzC6W2nX)6FNIf)V84k
z!|?+gIU+LR|J83;ferZ7XrTW%Q}WD|04*7pLw5N808>aIG6~>6S#aitrMs7U+_E-j
zb2qN_GtW#JKHniIlM2hj>3E>DFs1fiL5YahKQB$6Umd;qe7S9wxb{)9&-DSWf@n$p
zLW}t$v!Kvb1wMWky?oLYw0absF)7+h$A9|DcnE*8dTN6{fIIp+LVA8?)aIT5j5M)!
zX>3R!ie(n2Rk9cMnQD?;wY06a1%abP^Pyumbq0vc6$v*Cb>Zr+1l5en9)Mng9i%~_
zi2GwCgj3)yQn5Nd>zRCVp1%*+pWJW05B=5HZ|1u2>C1ux%$X?vscKV>&UkabG*mxj
z?c~KUHFi_Z#yC-2c2mj2@Ka9!#Xqp&Zo2DfOVn7A7k!6X5o*|Vst!tgl-}qvoZ4Ki
zGp}{yl6m9kD<#D(2fY#}l0E7`<HXBThu1G)&s=sXI;_Of0?!c=g0RmDXujY@rvbo2
z;5ULwBU?6K!p@@haO95QllT(KK?n&0S&v2u`-ArD>tmCRcyAmuh@BKb*piBdE6_^I
zF6M86UbThse7{u$t2^x&<ae3pZ#w<_m&e{?=z3axL6n%>i84rJ%=`&gfV|;RAir`p
z4t|j1D0=Yb1Eh-bSx4Poj^WJ4VUpULDd=DjdOi;_vz8&VvJR2VuO3JL5Z((k)MX|q
zp>G6pD|LKqaIS`uR_YtT>x&^-IEFK(Z^t%3#ow@0%Mokj=rHWyzOc7AK45QS<&uI~
zy(CYX{Rl_O!?R;ug<6qb@hO;hMX_s8rL2p0<?lO)LJN8u-%9o9fQ(OzQa)#ge>O90
zwqn2<by2{%N;;Jb-z9nyG-Y@m<+`aUjQfr<j57pfnpgRl(IF}TbZp{0g=?(LCCEfd
z?=^=~r9g3!kerfb#Hr?|Akb_0$nzvcX-7lKgD3c14)g?Cb8!t6^Oc-F-85prT>L%B
zJvuN<UUQjlWh$`McA%1R8S|p)2|xQI`-9f($$HhFS$G4$4b{<1p|}bS!_om|=W*z3
zcUq8*G24+g#g_;;nyAAyN7wi@QIub$M<C-FSsbGz%QO9_OQrQe4n`EuS$JMUy28uA
z)LcCZ_37P<<(CEIFV&XKMm<rhsoI>5y>X`&6!ljv#jT7&=*EGNP;@gDzh$+`a~14o
zxBe^myR1bxYm`9=3n0t_P4ZNw7<s?IRbAW$+@T)L^3oN6$&MDKk!~P0L7W?YVs&=(
zQiH^zABNbuDCGVerL;J&L&QFT$dday0|G|fb!`!Lc=c|f;PZ3ejJ2ZlWPb!IzkOR4
zDp$GWU3cyV3AZN1)w}U*d(15co6hww-N+Piky#6ndf#YpCTmPQfM<}z0SYh5CTol+
zC<_PVee)2|z5m`g%>m^!)J9cp{!OLNMW$Ghb|;>l(zrYBtPPz~qCZ&ziJHUuhgXyR
zgq!T8-43C{Y51z}U+las9hDQZY)g&Ia8<LpAs*n%i(t1YxnSGh)g><Y4}&;LJ{syS
zxz%dR!Dx<8i~4@+Su<8-zmAcoVtV2%rXPz#_9Y5<krnTq?UYAAi=@dRK*KY7WyQZ6
z4=`2_VRxe9lGGJ9;J$3ZpL+~6=V>W1>HqOuVHICu;WazI6Tmi4Q2EZB9)Mb&_eg}1
zN*w@SDr+Qb!QXh_kjG0>QK35#G3T&r_KYRxUoetg`W-E^3yZh@x%8>jlDP)`ZmG2=
zpA!V28*D7HmA(G?AHf?+Xx&dBl7{SXHcxHS6y^(5r!xx$Ms=hjDv5>E{)xgPpVMC>
z;2}1Ug@ib!`qzLOP->UmH*?8~&Om`T*N|hw^hd4nRo0C2#Fp<Ys|TnNLnKXuf~iy1
zCGOKEQP;s#l*V8xK{n#p{)L!8+CF@6QUgFGj!ofXd!&Vo#y@hPv->rxw|!n+C!S1H
zq0(57qJufDG(pZJlC3Q!3|G2rHaz>KoxK00U=)evxct~Avjo!O%iK~Mff?S8ie2o)
zz})e2(<GBXo84?gnx(AJD5*fg-^iPt#@-1d=j{EML~WssTe{J-NK|Q7CDFxES(cuL
zg#t&!imX}Qnw#j=a3VdC%1!Z#UTWC(pnqOUVnNDcvhWpA#c$^0XAaoC<o|aeO3`ez
z!1*6w0&U<%Fv;{lk+6TWn=9IO4tN}>U)Kf$@jdVT#|iko_^@DB&ess_BEg(c53`K$
zT05N9wsu5yi4VKGEGfTidX$?tvXh@g4v9S^XQx>limi)cCn@pll<6mStCM1KluT=6
zJ;es5WEE#2k$0$XCMn8%iC&M;PX`Y%*m-{>svgj?{APe_;-Kc)2iz7rr}*L1Fr%AM
z?=c$ilZ)9;qzQ>>O(2h`WDLR1fxb6|s>;urk7L;t{8bCfuQ!->hLjMipd9_pg|U*3
zeq5;2pO6nh2~URs`>jcyO{~4Cd!nfY<s@XJf+c3YFq>SC{a2yB(M751H)ZwmAin!7
zZJC&IoF>006g6Hp3V<(0I|c_w75xk>AEanH&^Sqts2#3U@i!T|4jCLZMsImGNF>)K
zaB2fRg(J*KJSsm50TqO01d@uzA$Z#CN2V_2K#L}Vv`?RIqF^e<3rPJf8%}k^h=2bU
zm~VFC05Gk~@H7P?etOcP9a6pCn8aLKGzv6nT1&A*@h>43EWki@dOq0D{6vwTd_?gB
zS``xFyPq6VmHlpoCeFB5bOb_-*Yal!jVeeBFMFjRCteD~h%+`aoctD<5SHfhA^ct>
ziK#%lAGA0rNG>tD5V2N*AZV(fSR%*{nyE_GUMz8R1*&UYsjnEi7<1!~onX&E9WpEo
zQA{nv{q__PB>-ylS+b}ilb>+<0E8K$<(MX`@TKT2j-n7KDi;=cU?3b>L)Ocmv7c2~
z7LKCgrhT>kpK|~d3n2<@wI62w02dd3-|keJAUMsAYQAhcfii_+Fr@V*$RaK~=gM<0
zmnB+x0ol?P=SD%bI&wQ5hLcJ9aKQKSm+yxBhW3K<0iZaejq?y(JxI%s>sa(xRzlk(
zh`kVeG!S%FMBix&G$yp3qY88?|I*u6w?~_GX{nR1O23D*VrMyTBbDC4|65U{^a8hV
zcGG)jM`|O=EPCxx&!sPY)niJQT;;p6&OA9UQ^CSnGC+l>c(%bmFf|E#(rKmXwVMjh
ztBd#R51_hOS*gBw5ksk&#~@lMO|9%1SQGQbmDQ)QqA+t%sh@H<?e!RcZua;QlNP-8
z*vwb-r<v@9n#XQf%5R>u0l8W{CR<|;y;Y#b-+qU61?~~d^e)j?FnirLMk4cN>h?B_
zZWMYSJ^_k1myZ{e3r^SzN1{${a#s?%5m{Ie3BWC>%%Ua3O;)!4h!7~f50)fA9e6~G
zb`T7Ium?RlhZtAgOIFE_%`X9>+%h{}{m$hVmnnkXAE^y)n8#br?X(=ph?=Z?4zW_?
zC_>z1j=@P$06%CQplUmPS`M!qDj`;KBT*tiME$e+awwtt>|N7QdgB&Z{-3|v<Oj9w
z0xTIo&Fgj4gZ(K$uD=lKXNCOH>QiVFY($_}&H$O}147{?7tFb!V#Nvv#yOG5{w0M>
z0KP)v$+I7Jf^6$Ed03zcb#9U~W=Jrd;CSI!GdNR}olczn$KLIHs6BQma0zk0dA%s;
zO+Xs@2C_RVdF*t}(mNoWwe%u#zu*GK39xc0+Euc?uD@bLAXZg*&37j0%pe#Y$e&K9
zXw-p*Ek~dC)2%b)eBJxqipS=p{Q2AeQ~B$CUvJaD{ar`R@8d<uV3#3tDkGP@ZAFcx
zr!N#Yc`%nvE&A>1%2Z!(B{^&T<L%(E+Rwv@tE;!YC!PJi%bjWAW@8iY0ahX&50Lh`
zvDaIpM3*07{s1$qE}=_}eJwNb4VpsLknntTf)M6buD$9Eh2L-GX;c2h#rIWsA|Ui}
zW$)gL3=k2nMJt6GGq@fxh%rI6mlZ*^54Q=Go83S#B2A(^5>T_}MUNZi<4nQ>8<TF@
zzHJ_;r&r;(_M-DVZv}W<FqM~q0J6f*BFC^fpD*O3i{P`vUwk%_q8$xf1`zj}`-8ds
zIl={gw|;gh6FO^1SNUw5YSUk>b*7#E9{`O&a=-t_dqn{V_1swP5DpYs!_}Bb=U{uc
z^t-y0zQ1b9rgJ^d`>yAC%k|9#ojrfIyYzt|CKDcmD6gR-K%@r{<x#dYWb!6ErXjO`
z-k^a-ab}%Oc`L@H48@vQ^O%0@8ZF8FWocjDF6yhr?0x-yA*%rbtge3>*=pJ3bf&Nt
z)Kk*(YI(4r?S0!UW|Nz{3N2JUSyXQ))7j0e{r%#mnU+qgBjby+@)*89A$7S{Ee{(V
z^8iK{gASeMXcwnNk7>Le!s)=t_S-gpTpnM%UA+`}FN63Mt=Pj;)1kpez87p{hX>n(
zWyyUxGi@@gagSD~oCa{(geybjZ`>GczToY)e)pwt^9gYLGeok~$0d)!#qN8Ccyez)
zSoK~9nyee8`koGSlVP1<;Ap|70i0wAf8+NotH1to`5Tfx*1x<hKL+l!HyB`U<>y~A
zyR$vBJ2S@Yj{a;O=lUIo^P^PfJs#(HtH*hthHwUOa{7)*aQ9VdBkw(OOe#Mn=dbWg
z-+O+SpNmJp<L$Td%x+ESn6UF3(Xjgd6?LTl>rwJN%65Bh9%<XpL2zyPe-&1o{{xnG
z+W41Iasm?qI5n5S7AF${G?!r+LMWGqmkFDHWncce|8bE?1u24!KjRnwce`KyyKsdp
zDgmnb_va;EHyVWYs!?<^VDsx+^ZoKCXF5g81ED3YMJ&()Nf4UG+8}hi)uV-klp);=
zA%M33-F*`h<~@DWe{|`$>va=`*bX1Rel+aLUtdSH99;p+cUH1MX{7cdDCE?Ge1tK7
z>a@r%80EcgmPvt%$gxexDdt2{;DmxoBuUU==tYW9VT8wT7sV(+p7^LN#)?o&-pN8s
z5=`=ayA0L~T9piDUIwQGDpgV6UY@K}&&f!xH;&mZjh0GisaG6k6UXTMP)lrjr>4lK
zq%f<ij$v1c5@9BF;{#QEp%<DAl;f*^{k-@pcV3IHlKMBrSGmbre3jH^$Jb-B<oKGA
zr9B>BOoo;CV$`R{*M|<XE55jX562gSdyKF7LbuivSfzy0YEaFM9v(fri@XS8o9lcR
zf7i~z=!$zzv}$xc%sgU5z?^LBm0h7077Q=#8m+KaAFk~f0<j-+kKUe3fbtxFNy0cS
z<&tPTG61923E>W@=%#kN93}`bl|~Ce3k0hUzS7_JxpfGalw)IFlh-nPXHC2p<cn2z
zWQVY?o^Vds5UPA{>k5&du$B_ika&|T8q7suJ6Ku$f^ikjGCqm%LnemBl_^6DB1p#=
zB8(>OLZ$7mnb&!v1ZvAYmud-rMMUBJ(8EUqbyka@+(xT)7QK#DYDYQ9^^fg%Ao+oJ
z!caP59O3O_SW4>7Purxi9eaZ%yo?cEsWrUJG>1jg;eNcz9nZk4<hX`cN!^9Z&=f4=
zq%|d8{g7Wp^xt;+lo2!o^4$($qy@OZIRftIm~mtrV{M*CUP&ViR$9q_9o?Yx!by9g
zHo0Y-)8;m>jS3>TO8@8$lcuh!sMJRUIcBeRo=Q@!_NUfA9ghGk=e$(^w52`*IQ571
z&n(tJ6Q~|2SJl9<7-hx<YD(%F7#3@wp@taXH`KtA@d-7sWVo{imQ0pVL?-}0vj(yq
zPsr(v<4R8F8e=&RFX2RgBN~LytbJ_9rEWN5P*UsLI6iJW3Fx0Su_npgBPW+#?a#T6
z8ad}U#|0ZJ4?=3FtcAPJIaKh1a>~m7+Xi7P5Fgbj54l{U*GVc!wo7&5k$m&zJ;=G4
zP8%W%n(r2K>eQQud3b&@)D&U1GJ<(KE*^FDG%Qw}+i^JFV30+B@hJekuRvhuCYwwi
z8!P>=$tk3DsGS)4`ni;#Gzz6ZtLbFBEheYFq2#ieWi6BuYBK$OpheeQq{G5|?}YT5
z8&D?6fijhjB^kq|43>IK4g`+*c?mk8W(06AbX`Vi`yKj5V?T7IBqu;uGGr}$1zkzK
z&}5r<9*LGU*+;T}Q|7EasiFPO<eAOMxtt(hYS=5zr)JN5h{kp~kUevk%{eZ#d9nDQ
z=G53{&L=D3oLZFIwBL+oYFbZ3GoT1YKZqvzE!7}bp$W(F%YJTypiyp$p9zme3N<TM
zBj>RF<#K9y+tNdZ26)`h2p_1s@d6c`Id{jl#wyNvY3@#cZ{8R+GCAgE?KE9#;{@7b
zR+yeSXE00=GSi8CoDk+)lDrQk94mFPPrJHLySxvd>+?Peg*9e+s(Ew<OH&o&d}LP5
zVOMfo3p=J6PL3^XU%QvE2Wr$jo_W}jF)2A8Iqk@plpHr6cPPThsY))6OA218iTKp2
zpYoUrN;o?y9AI@V6^tY29(Yne5r<kY{R-!`*K`U)1c1M#n|tZUzi;=;|0DTy@t09@
z0uur_HkZL%I}?{{oe9JNJC`t?34njL3D<k}pwS2!Twy%wA5VOjEqXOHYdnvT@9cP1
zE>ay?s=5!4%#Y&HFh!Bd)f?if?PN?=A!tGSMG9@Kw=~)yVfju<o3DIcRSBqsS9X*T
z3X*QlQ&byXl3cA_)>B;&(ueU>@{k|q*P{qzp(DQq^CKg3Wqo78&#{J*-h6*W)bhS0
z;{l@gCR_7m&!8)0d+};!FWQsbLEp;Vp|xdgD2;GV^}CfmL9y8>)~K9t;5V>NF8P{e
zgA<CzEz8C?S$3vsEj!P&w8!47lSENV=R%D|7m9BxJI#7~^@qH}wEJWHOn$D5?Z%Z9
zSq)27ESPz(7~S!H83bsk7y*CB6h0}&ku+ZV8?r3KLoge$Dj}6uH6wy7QJT%sufCnv
zSXgb<7tGtJ&>*|+l*~2jByoua?aBwAYgU`16|A>sx@?oL#^+%7jjV+`+ai<SW7f1L
z-W_va<>*|`C{|JU_YLo^KX1K#lgI2sqlO_k0j~N^&KM06w^^e^2it#WDbJW$d`bEU
zV?0MsTw;TsU#-7ta4YhcmBFrv?;PtYN9MiuTC-Wz-HC6FRG7IzU-dUdlrn%utnvBW
zaa5xWCH;Mi=lvX7BaQFg&LyH$Y+Yk>c13W=*D+r}r>!dUswGCWL=f5ds~@>foIW13
z*%y6Q%Lu(y8M&R8kn4Yyq}j(G&YId=K~>tI6#c&pBe|(xnJ&iHIK{c8hj4JP-ze$k
zDa|Mp=)%bFmZ@1KP^q6%oE10oGs7HI6q+!3UCWP&i)y~dCS}&Lq85IZ6&!2?A`Tyi
zH!AdBZUqM=j8(a&rz&16p*;|4!UwZ_|8w~kYxn-t(m;q<@(X_xf9PK_qHg9h_enug
zW|Wp9be9>Gq;!c3XfGTcT;k?Neae?|)pUEBw#2v^PlHQ7G4({6A-d1@U7jO8y2ko>
zZhRZxNS>z6cp|Xz#ojLd*HiV@v#CDw!USjatoBge3+Zahr0wc-C5y+lsjbsfT%n3c
z>Eq)(OE<c?SNne`8-(?`?S$*y4Pvuw+&Etw^?XCAnp)`G#>wg~{*~JO)hyj+KMZ-#
zcHJ^K><!9eR;*c8-c$m8?r5tQx%N~>_SwF{YMz5|3}){QMr0?K&Z@hRLga$V+(j8f
z$-~Bw^5v^s{-DKAWU`L>ihmR7Gj)vA@&}~<e=omLfGK}{|HOy@v~bi0Dny|uK@}sG
zeW6cczGjNC<patB!zplCOD~`hl2eVrQWYBt;m$)_SjH*Jl%-eDJmrFJsbW!LQ|(zI
zkjaT>ju-_2W%SQNOj-fnEec{xSlhDlX`r?~Fq&JQ4Xyg`pc=D<9dC-kBz=$Z{7uFH
zw9~pyN6deext)^3bO5<;&f`~GDYlUXphO1mK~}!~6Cd#@K-n3>V8g_r!O0jvbVZlx
z+E%O<$LBl``?YBp-JNg2_JlfHTsN#mB(HF76}jnau$p{JXzT7NVKDDQ*M|!%tur@d
zD1BQ=`4*G58Ee}$KKZl;E55lsCDXj2a^k(4pJ#8TvSl7aaoU2xP3Wt(K=DbLus9_G
zl>YuC&74?+aAZpzF$I#E0NRg_Ym#cx&!lyC&8Lc+bj5P$u7<76igc|@5+{#cvT)s-
z>TB~Zk*(HiJ5IV!esg)^{jy8QeCE)cu92Q5u3Mk7{{u=P$vu}*asm?rI5jetA&3@g
zw-%%cvrd=eL@N>kGB}qpi8?2jkkT0hmk9F-qkj<PzyJ36w>O?uD1Ym|rDDQ_i*=g8
z?()8G+@M)?w35t$;&^o0u>l>VkGs)2euK}wTrqOGqJ4Wshx!dr(YN`&P`5Y(G&MHf
zjaPj$&X{)@(b68t<nu~(B+Taz7!h<Fe=pL~K5f}+y>ibh#t)J$;En)@K{VGc&%3vh
z(tpZ)`ow+JJ=#fboS`3euPVz>b##$svR-;ZAnhq)cb)NVaZBquD!<H1cRWUnV2Ugu
z;`S$xS=ii4#=R^y*CDT+ywedYX5Y3GVoS@E5wrez&vL>SY_{ru#ap#x+J>Xvibkin
zGPn1%=VYCNi7lEW-F(K~l{=pJOuk}@#(%EGJ+xS7i@G{TPZnst&GHUdF5_*Aq!#mH
z&uoF{vtqIY_e<34m6h0mtc3g^XE-Y~3$pq_%tfZinxj-tyVc*z)hV{xF-FF3HQGf+
zsCs3R>X6jpT{zxtJ1!w@8*Or2rWmwyoMfrjIhIuuQ}upEeOeXogWeIg%CL<FZGRr{
z1q1i3$uc%a1eRZXWRiGoSJHlgSG=N-R`({>7Fjlr!w9Zgn@bj~tx>O=S-$v7^+@D<
ztxu&L-`D+rnKzJrD_$)>n)vjao@%^djiX-Q%_O{Tm(K|&xy@s-rF$FvF>H!C)4Ujf
zkGx%uQ^2}%u1TC}<|9LN#fg3fuz!-Ns39wCbheLmm91mi>5k5MrD^rLQtQlHMme<a
zq@kO`P=nD$cJ`9sCv?(o0w%f?oGq%hYPv-QMg>SmM++A`dd)iP4jyIx{IK!@)2*}`
zm42SHpf!^H<*%Re6m_vIIOPqVVwu$*rg)~6nx_0aPuVCiLnJY(Kw_uD1Ao^R+GMWA
zS!V}bG3}!<;|7zYoZ8l1Zmy?R9e>IBJrI@F%mWe##0imN%S~j8CrHke&__K1jXv2E
z8rfg|yW<H4wgXdyXI_M3o*<a<cA-Hvms>4Ka~y&R<w{xF6b70W2yV4BMg`57={YmT
zW7M8BJ?G4Ean$r^^qT4Ylz(JnmFS<T5_@V&E+Bv%TK&uay3QfFXC{P;4uMH-k2r^f
z>P(A;j<bQ>^9GUzDt~S8&Ubt?^O1JZK&qDBH(JypUYZS13ZSsYNwoc&)0(+AzH2*)
ztSFz$^O8c_;=9K^A{t9xv^LE0S{$X8TXbEI%vd&2Y3E^%z7(_fZGXc}?Ja3eY};Z@
zIzf@G3TM|ciuY&EMflq6^}Qz)v6OXZ+r(5F7DDxy*pTjFif2rv@M;b-IGf|vE>|Xd
zA?zVjwJYv0y@Wt8R=^a9%iy0%ZUqr;nR2y#O-=O<B6vn^gPJ;0Uz_Ifo_+3KqG+R_
zUltCi(p#(Y{GlqZkAJ%N44PM@)gJ0WF()DrP*ucR2NGv+j7v|k%Lfsg+y(6s$AIq;
zzJj4%hiA5?gk;t(%vxFl-7iX%j1^VDd+hU$Qh<eh3cg6<Q95<j_0CK=sCLxq8G@9m
zZLdC>j=ptOUq_OLX1ZEVW%8KZ)8}u+NqJ5Zk9?Ll-7QVmZGRq>{(cmRnx0rrE4Brf
zLZ_g8+X5jNOA+DMomp@ZN+mKYCbI^Wl8ztqNkHCfIj>FMXg8hCOSLRgZq3=r4PmGe
zZdZhOdupYcva>dt^Vxxva^=iow%m-f-c2U`^1OJTg%EVP$N15&V;~1xq0p4st)bet
zF$KpzzmRX=KY!o;vph}FOi6wD=a08|T`R$qmymj34iA?<|8@EO?eA1-`|QDsON>lw
za1S9k>|CcF$~e^)7qzP^xx%7`eNsY$nEKD3e_Y6wUmJokiYZ_*vJw;$t8`=q=;i2W
zwK#Tx3%6Vx-lr!U@&C#`y}wv6J1DQniW_t(MLgSDY=2K?mvw6iU~)Wu&s&3fo-^<d
z0Sq;54P9xWD@Kx#%#Eqgsl)uZfVUU}ON>zNn|?lGkNboMEdEn;Kv#Z^6-wZ1jwgX1
zv=MX%^$|ur9m0>G4j6RP)g=THgs6&-*}`0TtEX3(YbGEpFBTFX+itA`6VT<888L;b
z?=OUu1Apw)VtxI(MkVZs#E$`qmI=E=-LK>0V7{<)wC2xa1N0W8IbfToqy~`Y{Sxu~
zxMaYWig(6Fd44>4hIbN)2fP!{h&g7&JIVSdcqfr~f_D;$w(p&I^ef((tu1|kuFIcj
zZ7Ol!+GIv7rUtA{*B`ewl{ju~DzTr7$mneCg?}a8d<dIbfnkO$tS@aW0O#X4k6O3+
zBw&m66jY&}N*Sm^q&6&Qg{qMKmaA8V5Q$GzAzv*QxhyP=*K}3pc5FmXsi8kd)z<Oz
zziv@WB=g(4M+p_pdAuY)r7o+hbIG9fI9Qi$dS84r6B2C-?T5>q$z6yvl458=BBNhP
z)_*x|5J~*|M%vKeT4JH|izMn56D}Op**m788$ChiFoh&tcb!T~F;8`VjY@KFbxdU`
z9FxMFpt6*PIk^B2!hNJ-l<_gGcS~*Z>wA>(AuZ%MHCZyc(n>Z4t}OP;ZOMP}aHbw5
zHLr!`W2I$1FvY5etmwm8v5KV$(~ng=dVeHVv826NB{~nr>M@;5td8qkVwJ4B6RSk$
zu~<E%a*fqtm20eaReml`HIr)LIMZ{v+}&+tAR}Uoo)b6HK(T3-KD4p4g+fA{5jWCF
zX5?PnJSEXXVWmVDP_@}=#fKw`tCPG`#gb0QgQ&)H^;md~qpG0zHg4e?gOs9TtAD5b
zDza*bQdG>^W}$gQDf#hodK_w=q7@q;zRqG5By6i&FQkG7k8lbSv_~me-$6z>q;DaV
zVSNjq;8l0%1nWD-CP&n*WHPF5<&vJdFHs2=c$`Vj2$f1CcoU}2Bg0r08tH|~ODwXD
z<pB~o!#693909aa$S9zdL6D%c2!G@(k58}Rg+<0NClC_kaBBUTj_SFOsSlZZ6l6S;
zPM*a<SekHs62hZLSct64`!qy!zK-$GF`Ww$jq6;PsH^jHPkA$45s7U=*~pPnq+bG2
zL|<ce8Q-jNOE20WhQcVrh)OM))8b4-Gih}I-D+N7hV>_frF51#=N9|eD1RObt&9We
z12YR>A7Q-U8P}l;XovDdU`n19_bMI+tAm&-6@TP+M%Y(6!IDN<uQWGAn-f^d#TVF?
zPB6Dky1dF%UO4OWly*#s(|RQ=ZR=rI;VC?Ngr~^5J3K{n9^|QGIv1WA*SYXiSLc^_
zib#BdrwF2hJVo|l`aCs^Xn*CY^AOEE)d6&er%nmW%2Q+?uFq5BfM%Y;g3jQn4xk5k
z>Xf)IJaq`u%2UIbR-VF=&f=-lT)%WtG%KD%dc>oa&Ds6+7dA2>#(vf4Ju1oXvyjn-
zzx*ysf8Cczkx1N6=atfkRxchX=J(g=J~mDi3$5YNuVkM>XQv@rj(-X}fNrGo!e|80
z7t?upyi4}6UU5`b8wWJA85T5>&dYgW2T--&PZ3aGJO<8Obqv$^d!3+9z_cD0Vo9Uv
zyc`9d=K7U8X%LS$GUZ^9xD|S(VE(-aS!evEBOwj+Eauf$-SUfDBL0q5S@NnI9Tg-q
z!nePG?!CDW{&f7kv@7)mfiErk!FBhAIrzx=_5Sobocz=2<u=~Mx9(2Xe-%Rie*b*?
zKm3{oHJ4Fx0uuo^mr?646PF(nK`ED%{t2popT}Q5e}5;_5)^-{>igs8yMN#Hi!j;<
zlyv&**Sm0oPziK!0u;Qw(;t67{d)JOLLpO@!i8V__tP3lYlH(ENdEMTJpGMte$Ru^
zyVHN-(65D!JpJ?ZBQw||?1YSjkrnPogyji@cRrDwNoe#!l0TosLCFjV<Hl5#6khj#
zWFyR+Vwg!|v~FO=@B)oYTMN&`6U-DtM(F?~{@C=tPv7}}N?_CJ+os<pd?t`o>uG>d
zG<E5cb~L5EFkW3{!*fX-ge{XL50}8U_m5~*?M_Z;Zw;;m4+FqX0(J1v?CNYJN&JV{
zikdA<Y)Ng?y>vZXqc?Vc3eF!#u#L%oOe5n=m}cwA)_Z}>ND>65$P#&nKokN_l!y|&
zmwa{1m8ZOw`Qa{yMRMhe&ud6D++u88^~|gM${Mig<APSo-8|EjD?V$kr4{rRqzs(Z
zL*^B)LYP%Xza*L5q=%Vu(Mr)sTsbsy>(#Z*SIo42Ez6aL`w`7GGOcT`p-}06C?KOp
zg7zTDmXy}&&qj?Tjj3dTmn2oPN|Y+&GtKCA7?a>CcS}NJiF6?;>MU=ViPOsSs`-9K
zqJD%RRaj@b8pdRdCJBPAz^SCE=uK3uBCP0XLM}L$Xk>mF=fww2PgmBc(36(v<0I-U
zjh#IMT+mEdPqg^kY|+-2ioBVB>(gYxVBKo<+uon%XaA*h=T-KfDaEmRmOuWUDZkbH
zYTcIAn6zG)wX%f>u1<{984+k7m6;qR!z&6;wW^yt&q)B29XyLO1=Vtul1XzIcJnAX
zL%9U$WZt8JdYZkX@=B7oN<mL0G&SMCZHB{|TD!^|aKgL@)t5T-)zm6~3XqdQtAX{^
zuditiYwpZ|fKC`YC&{<RNs8zwDYY3md9IJZP?Op?y@F~8-2d=ynqlqg4AjDcYo~md
zrm!q)Q^*J)YNz}<=AyZ%<5b&BQSXLJVIKf#+bQ3Uxfsb+ZmYk07<1t&(<sw(d9RGQ
z{}8_H^i}@*cKZ4{){r-ULQ8C^et9=4N<eaj?y$FgPNrb{)K~=-R|leux)hdHQh6HJ
zrOocDI}$R<&BaI>T@@oU#o&wSKT{=`x2Z2A75hReG>S(ytEf_6p@wy69(o2vICUm@
z&Q|1hLwJj>=|sqIH^_CqMk|z^^7H*ah6v;!!rr8NZ({S=`=8l=07ER2(?Y3k@{eN?
zeef$<7jc7&*If0&D&qe&WTv@Zb_UMThnC#Cbl$~!P(xSHiuHgCSI|!xnp5QiQ`#GP
zcMhmvxUaWIG^!Nx#V|#)X0g-gy0-)OgVZwlS;9`>;hu^y+0Oc5s24>fY=Cez<)KdW
zwypG4>!`Vw%(Yv8-4Cqy0&cSxaBX!pS|}JVkQdS{bPP1$f*2jG+h8MnIwd4ya;TeS
zDG~Ki*S7N=Q24CsUdt^y1Giaj&05_~^3n<G<{B?|MOamsSt7)>KBv7}#vKjVh}3tF
z6UEY1WSvy~j@!4l<%~+r^6r1%u^kCO+iRpN%tkzslc)559n&1#85JuSL&dd1u!0rf
zkJKX=q7t+$QrO(#1RO3UHLS+50Xa5sQ@F}&RFsU<85+(N#y4{OK9<998{)e)e;C@5
zJ)Qk<VM|=1^eIJ}+$>Dk%dw}$cJ{4^*vV%njK=Z0>W{;o0z&Qig;Hz?X3VD|L}TMO
zYR!w3T=igo@?pc*m;5yHwJCboIVz+(Q(Tf&OW_95>Zxkv)LP?1We=+~F^NaJZ)BHE
zQtsiizOziPP3yIcbXBy-5hBW$R=*Mb$U5PCiPQM>s9kPjK__{lb#1m)xznQp%n?)o
zrI-?P-!!K*(QUsz@rju<jw>?c-4E05RwTK6-UX?D+TbF^IDRQBkejetEBF5$dvomc
zfTvhN5{~3Z=D@5?ulpvaf$nYDXsnLOv2-VstY?e$=%LzSvSlU}>ZaW@ZFZ%~HbrBj
zI_nh)%jU&%;nCD8Lyd<z!Mjf5zm3aM1UAD`eJ`8aip25LQ`uJ3rrP+j&Hjk3-7u&A
zwj_&x;55al-cwXGuJ86dwJE!DSJ&^8bbBkl+nj?o?9fv$>Fl5^*K1O`S}l};kTVaR
z2spp_vf&vd2*)>;9*0$-=gM4j1A`=3mM($0Rff*9lel<*Ax(0Uzl>{77wDXb8y82|
z;$l6&g9l-)Z>M~>Q%p!5b8)SSouW7tEL`P(@)XJC1RiPiP>~?paoGtgimP=)>s&U<
zH!=droX|5j*`QsirUc$J6PJuWERD;!;5SFiWR<Y&gy=wUN89zH>vm!;)KB@a!^H}#
zv7M69z>#znKay)V)lPvO$m!5OHZw|-%BBdq!B*v73dfg_yB_lX2o`{g6;RYzX%7p3
zlD&qhBLmz^bEFl<c-+}>;a-@Ay969igm&#DHRgjTG<1C2PEqYSZ1{G{Z)3TLy|kY)
z-Wx@&IT+}6%6M;-J3j`Q*;xKYVa{B*V``?1_eMEU2xU6KcG{5eK!fbKd{}Ek#=D{D
zIbmXF-!Egdvg5HUpyAfZ9Re$`_c#uJP(d}u9Rh0z^)}$U<^t1Z-Hyp53Dr%MI|1Jd
z6~aP2WyfNQG2V-};}*%0>KR|yL_BQIJ4kV>hwaoIUnd{8=`K*83EqnvE89RkuT1K*
z^vA8JU&=Q|ypN4{Gi3$wcCI_dh;I!sJ%Ga1MR*#6zdm*Utv;!wm{thf-IGdxiku*>
zHqxU?9oZv~*>DQcyj0}mQ`KVi*&1vfy-zD4CRP6Y@#WNA`SYiFG1>|f5^2i~WcTD&
zK9n&WtDql^AJ6B@QIC}ADRobbokx+ID9{Uv=5mUTJ2+feB0kKdFQ&@2;}B(UO;tfa
zSWoo%a)Ml%5*Na2V)|8Z0I<V<afR0dx7(x1@CGd;`Ym{`>~5bAZHx~~b(}V#t+<x5
z{b>An9lTp=iq?5s5+>xJPA76Y>^D+f=wzOAhg)c%gV{VfRJ(X{DKa!WA<XsA>?qm|
zScY?Sy8W_<yGv3#CO(T|IE&Xiikb&&&bxzU!G+7><g5X9Fmo2gZ~}yX>ZrN6=i<-J
z*M$-UO1z=DLf`Kc+z{BY$A_buLf~S}u4-y4=5iaaV1Alz^E~SONJAu??-^``+8mhM
z(YUg2C7-8f*n?={sq^p~&rby<=e8Wv?#BXx;qg<g7V^P-OuO8OpSDWnOOnl47DC~P
zp`T=HuGFea#tFMs%FNGy7U7|d>xFJZ?N1A8@_BsxI7dSLa#`3&8t(XB6AhCUdWx@C
zP;|mdv!J+9&P-{m8{Ebrs2(}H#`GK^A?@j9)8~_t*Y<hOdGmPaS{=pX=Ev%^X<q02
zWbi~wznl*c<hID!g=9~co0AOJnWm>0wl?MKM$Yu{xuI^xvDPYoSh%vX3<Bx!CtJC>
z!g7A|=H;lt1FjLn(|XI(o?7<JHW9cBwXmRa1NqQ!BdeB_=ge@1wKnC?fvV+lMZhT7
z?<l@rL2+u1dy0LNdWOemt1%K`K7|h+fAI8e)IW4o&j|Vp|DHOZ7=7f`h|0BUC*PTf
z&y{IU<@I{`Qia-oQ@$<Y!!eol3oTreBE5?*AoUb4M|`f5dx|~cFV}Uhnpbp9{tymw
z-MgT0H4YF#NJj^7uqj)pgUK!^?Cx&V{#Ip~<CddPF3BYp-WFx}SyWYcq7-&8TSkY*
z92Tp;WzHnW9a4ESEI-pi_E^r$RKP@C<pa^3!aka0FP6T4tae<ni1n)GG_LBlE0CoG
zkLqDni1jV!LvP<I&F1iU%eQZpRYECq`?qhEvBHD9=FQEsNgR|+b7gA$|L67F@Gb>t
zJ8#USP1*4JEgpN7_Hz05){U&Is|<1`g@X`es@ADnyh%GcC0b!Dx#1wQI3i;8qq+Ko
zh`mZkC~eFm%4@7=Z{u(RR2*JA${YJ6xl!mw=K^vVjrwoK>Hh%fK`W-0QE~zk0ysFA
zF>@Xim*8C*CzlDJ3JZTRISj)`@dXSFSnBQp_LfcG7JJ%cb4VsY_L9G!B1KZgl5Dq4
z%|Y$49`ztUR((~JN1pz9lBaJk)5ptS-VL0*2v&K0`uP2%@Pfp@O5#apPhKhEj5>XM
zKmGDmO6~BN@Bk@|!NcRf1Af}`7o!5+q41kh=Pw>qy8j)I8Lxk<<-Yt}yEjX)cxb$G
z$yZGLTJDl~IK0#OMJ}FG5s%CH*N>lGWN|^^ry`$kA20uX;ZE2ng|(;OzQ3gNV3aUH
zpF$Mi;Pl5oPQSnWwF#(#r3mR?`1jk6MCY9d+9#2}{3=g>$BRGV%_fG^|MIQhn;Log
z_vvq_;GW<hWEOvHaJ=I!!SbX<46%q^MOYJsg!s6KPg>@*aN)L8B}KGcY(`m7h%)IM
zP6I_5qDxzYvaN_Hyr3-4GtNICV&HxGwB@_S3)Y{%IGONeV!6wuvv?qqEZ%D05Jho$
zI-LVpu4l>kiMLDjx*&;o-Jz$A=d+IA8BeQ(*Xrm2jlzFBjl!*vGPVItiw6fznNPa>
zv~72M2AR$KF+KwbX`goaY18hwtfL6h{2-T&5i+W$eSX@uJ3gx*0>IHeTVbU~qqR^*
zH=ol=HV^v!w0(E{PT3*YALKX2O6ARv-%s0j+ix44h_|!h0LG;hT0!&sL7$(u?Vis@
zL=#35^}~N6KF3XxSbtmWHEzO5P@WzePRNwBmTr$A+LUSHnz}=oo)n7U{3XHk>%#7G
z_D;F@vgGKTlW{fy`6SFs4o~Oc9a$eF&X>=m44$4W89qy~%M*zUz`_Ub@mPzIQafMG
zOuV(~wXX8f%pgL~B(i1ID&^B%mtuQ+4lXM<Y4(4?YA^Vvp?JcqM|z5JIHXy!E;#&j
zQR>qlX_9xZ{QRv=y9OnKa@@njyt5X+Z&R8ic+GzGe7Z>Wh*1O{lZwoK*xmplD!s9{
zj`^jG3}J4|`1vwJdEP=BQCFVJKAyFzxC!+cHdj#alB3QBF9xdu8O_%3ZnUnU4Wt#`
ztNwp!WhPyD?Cw~D^2oNrm9fqX@D{Q$KNw@LA{#!1?By<s?Bxu-_d;+B#iKM!GhMIw
z2~{-JC+(`et%v0aOP^0Q--1yL9Nlau+v;6<S}Ox9n<OZf`K>e(W0IuM?(k=&>BDsa
z=kI|`!R_Dz(F>4zGX+UefD#}YnA30vQwDz*V!g-miCdlw=LPuF*~qVTDWA2jFrQ06
z2~}$U{4(L~np^a%uX{nY_=-6<Vl}3*HYUDm8wC2_`TD2-{Q1`KAeQkFK03{6Po`w`
zcP8-BH!U(UV(}Pyi~e=fB9m4cp5i85)<JEt#4N)jakwDouv~#w{Lke3YZuS?bcTO^
zD$11xgnY%6Gjpk4FS`@#*Rxed*6BK^*)pWQ$O$z!<-y!QH&_AS?@o@|tg$7w){5NK
z2+YQ!<<&eEWo@&wYb=lD9ul284;L%4{2NmC#(BA7$`!L%n<E&k-nQ7~TE3}EbkSTw
zL$+t4*}rC<6+LZ)dWKOm)SmW8;@W@QEN}&BB;u+obD7s09KvzJioWcZyPkm|mT`Jt
zv$nnL;=qvcdpo!6-1w6tU76@I$Hk9t?AvaSFQ;DEH2DB4TxDvLk4bHGTFOLR(jWX+
zW-;BFp9>R?a=%bPCT`&b03yWJbGb(KiqflPk8x<Y;?nf%A~Cl`Yw2^}?7M$9Bjtz<
zF*T~Sbz<{+{gl~&Ob=N|FHm~|aadY%jflz?UJmm{r)o(Z=}s?Em*teone=aSpzTTv
zsr8Yz_Yzw{Ksk<Gn(cY%)?{Jp9PKT{ru#M{GT!3FRu$@aHn!2pO(^LC)XI2PoaDe!
z8o{zJ#KLWcmYV!gb9>~SqA`Cev2|jJZ<|UkT#t;(<~~JzLY4t=T^8KQNiB=4gK|4B
z-^qw35Z1D(Ne$%_o<w5xTqF({0j%bMI2Dx9-VVe!yFtq`F&1kixJM=zyCJQ<`X?{S
z?hL6(>yly8wcM@O6ebk~8(SU<b`RK=z!(D8R%eIX7A?kK^NekBDz<;QSmT$ri*4PR
zTfs~t1aK`G&_ajStL*Tn8yCGF$jUhP($d~e#ca<vb8r~;0r9JR<XW%qCVk2mjlmeP
z&V;7=!n_rey+PYl`IAlV&UA21+~8!=($cN+PZMd>S}m6yvYGI_w<UZ!nJS%hZ@MaN
z=fNY`)G`P!9k)bHr3ruNOTRWeufc%OMsGBD(h0tlcB9?w4nL&H1%by#lYCyH*TU+p
z?U!P|Dn6LfHl3NK#{r?GFwRjx=$cMk1%!rmxpW(>j)Eg^^uri9v=ZPpRw-9ppUUB%
z*8on$w5oPA&;bqw90R@tC2+xPG!1|Q3vWC&5}_P*3EARH@+p4~14JSnlD-3o+v&iU
zcGUuh44Qd?(on1b0akGvvV(LLPXmBMis8dXgC{$qQ_(c#?TP1O1qg)JMw1+mK1^b@
zSCwp&Mr+=Qa}Ps7#~=Vg2Zf1JPtqb#6bh-WV5UQ$z-(wxfZ}8rhQgkPnE)=92;qmJ
zz%`9SflHjp?wWrqw$D2Y=5Z(lFCfH@N6>w__qq#1`cw2ebnq|-Bn~UQ25whd!=W7R
zc?~wbceGL92s&nxH^Oj~=bA?HMih2CHW#+_8msxk;P6gZvxkk$gekM1x{Q(D;d?zM
zOmrr6Fn}=_)F{y_aD=4za!g~2YLMs^9S|e7Ljxp`8p3}z%`g(271pZ`3a&QkIlH8}
zqn3T7^f(}g6n4q1YUrp1!aU3_xdxg;JYi@s%r3d6adyci&a60_k7AbEf^bUT0SD>Q
z^c?f3l4Gf|nK2C$t|pIxaujoN@EZ+ohscJpa<mkWT+>K7>V)@xcXWWz%$h#nke~rZ
zxITYq-|v5BEpHqgPBGD-=(+v|1v@@Lm8CEZ914|Y#sRhxg}0lr97-Q0A2Eq@v9;P>
zzYljNw9&$?`$AX3!|U20ubuwmLPKd)F-uw#X}-dyA8%E04UVj>N-d_bZbZ%gOw)L)
zic73y0$9%`?$DvfL4hgUo4#EI?8TJ%hm@ZsHuQh`8x*ROVLUc44UTBB_4h1e)S<=(
zrs;s#z$DHD#uw6*8cz{Q=Db8<P+R(*y$Ay9h8euCufz0>s?^v%qWtXb>fqrIDL<J4
zx6_;$Y>-wO$TW?YpG@FH5=Z4e(-GiMF}=Xix2ay~TO{NSP*vWcnF+TD9XR567qfta
z*|mQGhaCk6S3pCDBQyhs!ND~RgM$m4NCJtZe=}Tks1QW3@eZq?Tb+u%HR*mt5mLGe
z%(@ODZj5@`^pg<SXmBh<*DyT~&nQzg0Fm_Lgt(PhiH0S`X>9}hVVrm`BH#3=ImFf<
z7uuD25W1n4zW4#mv$LW@!{FXj;hnvF-phZC?Kc^tth3iM&N+K|69K_YIu^ql9?Cfp
zLGA8+)pt(c*EdJ&n=k|lYI$P3Kfz4tHnQhIEG+#AuECM^a9w7Xo&5a7>@iK_{Ru8{
z+T&h3NPZj<7)E30dR97iuiMp`ztFP_W{z!UDlfFM+@d3a0a|z*RIY({D{5}`<I{i0
zG|A!iD3>_z6xQaDPytp*y=_{B^`)bnJ<5w`Bt*!i0y$vQgW1raP>X``ZE~)`6DZW-
zghTd5xu)@b4=!;cTQ;Wn5)Od^CRgdY#+44sd!XxnXYF7fzsXU-&|A@g;W9Qw6%0&+
zBb@z0jxY`ers)7MFo{q`D6%~~dkBA*)J7=TzjP<p|Du9!jPEqH)iH4BZRp?-4%rdq
z8XVhR_Tz78R5&nA2f%?zoav9o$#(9%{{n4!g;gO{Q_$L9sU$2fmu~KqN?<wDUUq%r
zeNKbAwBwQzpy&(EcAHt61nYAnpG$au+!kbC%IQ5@KKt{jD>gQgGi+mP!t5}N7qRJ@
z9c5ql06S_OzHi69@u)<pe|isT<>mJ%1uoF0X1?(Q4gUG|@$!F<*2LSFQE~zk0XUa|
zMl%(c=@tslf4Q80{r2;VtSVgl{`~Fbzb~YW)>1^PPtI#0!Jhv4{pFWm<>_BI`145$
z6T|6$dFuBxUEy4N`p@YfFMmHmNLvt523*u~ga8o&)Yu|2c0IpZ=_FpZsE^h48-V}>
z2B1vK)6*EjrXwT-`tbP`z}R}J&Po6b3aS`HkPc;0e<FZ8Qcyz_R;C+^hC3+PC|p!e
zLJ6GkX;Ei{_JL3`At}ODmlqjPS8u|>ur^K|46A76_2KKNp~4XLM^HP`SVrM}#8nYa
z4WzMxtpE*i@WL4XB*cvqTA6wO<bjZYS(5kpbY(p&li<u4p&P<y^ttyotn!psgGT<A
z{q8kQe^z)|q0uUnd>xO(+kBZWrQs#3Tc+Fe8~?Uir=PM8`#ROoO88kbLK_@S5Yfek
zqdmaUSYd*G7LHQ5yy&pMem`e;GH8y#N)aoY+)_}^{+rNt1Zkk?iNK_ImsD{TgiHPS
z45y|cj!~#k+6<*3D4~8Xb*inm`F$w07dXE`f9VdO)JA}QmPvvXfZj9hpWuK?@zWeO
zp#Yem?RJ>b2ZD}a5DXbqGqPfr1e&jx2-Bs0!w4D;shUyJifcC0UC@*j(sujv2cRJ!
z;GbrHK`1?MpOSAjy=q)J&6F6&eHX791O`+AhA@MwY3P`d@&swD2Ioi2R?6(vHe<bZ
zf7!=O8lj|Vn1$<@*@%%<`RDm;kapfb@r-6@HM4^6o@)6x3_zN*GjHUzTD~z03-#kh
zuO%>7=CxrP00TRcSYu^q;cW)gMkN%q^r_+jpur34o@j*D!@NHMnk)}^w!IADF44G8
zXoaZ73><8Mb1b>|>ZWc#+*uYGm#g8<fBGm?>{hrOfR;#{bkDOwFa5l~sUa@WyfJHV
zCnnO;VI*h9tPwbjlyTR6XZ9=|z}A^BY}!gry$_@F0_~u|csYO(S)Q}cqK*cTt*d$e
z9T+Qz&hEU-@U^7tjN$M(a^cQlTurcVQ;G>&4nvl>a5ru&Zv&yVLWgdN>j8uoe+#bo
zEK39}jCb??88E1^kSYzgn?PIy4w~&~aM$HInUYA{+FWGQ==V(eK=^?128=Me+g#e;
z9DrsJXLy?3d4XJGmSM;4J`oKpwzsu8kj}j^V!XDjIl#9tlXTTLcZi;y=*?xvWh;C}
z;koMpzJvP1A}{mdi8iNZ338-kf0QLhxqZrl)fk726b&O}h8bjJB}d3@hh}jLvpC-v
z?*=Y~1t4+|4%b1MVUdyr^E7KyD1(}1&=nx?otEU^(0d8lb=TQ-OS=Vs8L%^u{yJSD
zx2<&u2lB#Pm%`y?A4rL=<3_tXC^9V4wBVoBk1=R>SVT`UBqz<W&Xoc;e_LS9wKg(N
z_`Y%CSPBngLfiw=?pvf|F>$Bx09JSqgWPvwh*qF71DFTJ2Ovd@&!T-85EO>KNrasc
z7|%Yp5G*rM*Ub{#rktyxk#a6a)z#d)j464UvWu_N)@@3*EE!QXxSwGl7-8*bst08X
zAeW2oX{KsKRI>=$b+^v3e-f1f+v6sSmAW9H6D!05amLYZt&>+Jv1Npfb=o;m?Ucn8
zKd*$1qjersCEx;<{?CFB$+}Xr2(6s})gGx~>TAvkV|%M&n0h$D=(nkVwH6GG^sQAB
zSk9rg))Xc;!FGodR{T<z71V&=L4g9+3DoL&_80Ul0{U`pG*de;fAyX#R?FFjVWc3v
z&E6Ox;0lu0_#y?torc5Ai_CRv1TP#i7>?#aodN*jX$@|LY*Eh==xT7;`0QCzpOqdB
zFz95;*+_h_&gRfjGM@TdcESX<&|O$DZR%~|I&|nw-i(HLP@h0bvjER)45s$NB!kX<
z8P<b=gbY<FV?JrTe-Vk}4BxWtslt>?mTh+i>MC(+c>A61OodPzEc3ec+2d*j@LPrW
zymmj@-Yi41DyT_>x3_l6yp+tNi8FXRtu5ZO^1YQfyRDgQ@6s&l+S9t9-fcK~3TaaN
z$w#hScBa<DLlq!YhLY6{#6y%>K35)6nCPu1o>;Zuq4us#e~o<*QZmT#ap9~nrwDrP
z_dZoQAv-_LNL3(3O?oa+R&?E9_)2T9lgfOWOO;dU8PPqekNcwaMYa8N@S!$cj_O&D
z+$v$ADNAmKy@=8Iro+PJ-Xh_+pE|~40!i9X0Fq_IhU?kR7dX--3k`SXj9;G1&=VN8
z_I;0Di)~=xe~^ftcED{TRkO%o2HW_{-~M`+IXQ{BwsjF8G6jmjKn2JCfwXxmJ#fmY
zn2dz{`@KAFuA>~zc7@1LgR_j>`DD7zPHq=TrtLn>l#52!Ou3BG4skke-)2~@RgOWQ
zN+7{AO0gq-eVKlZ&v@w=loXg@Q_M{RklGVSflR=Ye}mH>|33YOVRl-Tza@bcx);(p
z49WbEan*D`mWcUTVLjw(P3fW6ssHoG@2Ba?*Dqa8EdPx&^GaF{akI_F`YfQm&HEDE
zpDNqETswlC1U@Vy)+;FrPrFuvyOA+q9d0(*$gv!22%<)=_sZe(;|Pu`O$dHyMM8O8
zv34{zfBKRp$|0BO&)v=Anv5&mvW0OrkG|LK;N!A+yrSH^n^OC-c@WO4jcj*?oMtN@
zkxy%6l6rFrO7F^-9x-UM!LC`~UN}(rt=RW&xX}(Pe7^}diKAiRTEh*k&}}p1?E?P2
zvt$5qSpU%`^p47{Ww9X7%i>+?M&C$_h<tT^e_<$f-7y(_e2F_HLsK)>+@Q&cQqSui
zMr<WCE~>rRl;6O3t7CQ4y&qq#rYkkrMuY2Ny>8&;JcbSUp_BaR5_g&|F00%_#yk5+
zKHB4AAvJ=1@1&=fZNM_-v$6-R`IQ|i9tJi814D6)Xi|%lxC{_eY4tw61QaTBDzEef
zf7Xb;wEM=_1@B>$T7fWH%`YTX(tEf97M^J0^9pjgaTiwne!eNIu1SR)v1O0wrTU>6
z1F)J%XFCkrD~VPY<yo5Qgh;gv<2Z`XJU*y4>*$cmjQNoCKzgrebCu?n(UtGeh`k+T
zr|bu9o?~@Pna8WKHm4E$R_CQrzmGbve+6C7y>@(mq8Xah7!6uWq3KbmIkX}mvYu;w
zR0WovsI<bNRE~#+hJjIUw!LMYSdxJZ8)4shjnm#PD_6^UIYDgKix!1|&3XxDc_>Nl
z25lvIv|@3|-C$-AN;_e-JFZwHk{P_F>OYw<4OhP2Rn@3a`c!lfmOc}szP;JZe^r%>
z$h#W0H{HBaSVj|7eND?rHM|>Ls^~td-=_QQ;eD3hmt3^`h19$Lm0M%<P|Eghzg!;&
zh2Tza0V^{<4%!Q=MSAYwRnt2HW}SH>U{-t%0%paij-*ypy<%9Wn;KQFjcR^(3L^Eo
zV|~ghIn%TTU^*A}3g1dCx3;FKe=|%ka)0Dhj6=58OPA%UjGNOqMxuqxekKd+hYToF
zAEaYGo%>us`5N47M@8wat#9WwP3Gi0U^{!^-rDIpm}W&iasBXVV0mYvZIQ!ly?a+4
z%jJ=mp3Imx(A#5{*?DS|tef)oA<2<GfpIs3sZ2GkP5;PR>JC@Fre=ARe}`dP8Ufx9
z9mmkUd#CN$+wfNs0>u2xXI}Zw8@E5&Hc0KkfF<mt;SV&AOlsn2HLtjzKX#iF20#Sx
z^9pkVt)|qdg;imoKX*d@Ko2O0h(vAN@>+YUwgD&kn8PI)cu3zhnRU?Nj=MKzwGUfR
z!uGf8KVoqY!!~Yt1GY)Ne~-Xse;77<6*gq*ZZ8bTOMipDQ-}xpc+_X)8(o8{;>xG;
zFqCA<a((>Bv<yh;M=8|46k2(q)hGqPgFVbBq7UP)`r$~F-r<Mf$K~w45}@;Yw0!C#
z3V;(aj<A7xTmhSUc1QAGDTB8EC-ncTdg!GJWo~41baG{3Z3<;>mtJxL1_C%amjOQo
z69O?emytaICx2W^ljF7#zR$1FarZDV9^H6UCQ~J8N3QZIarWl&DT%jS$sXcM^7GRm
z-T+049BB?5i6Gc$^w&=a7=HM6;D;ZdPG6sXxF|RnW>FY<`1;#H&;w3?1*b-h2O|X2
zN*um^Km78-xzu!>=mO4_qRY^COP~1qtc0Z=g6@QPeSbC}eE$nwXL>F|zv;c)ccOrF
zNqTaeo+$ScdvUrn_0%TNr8eQ{I=}w<^`|Eur@)!h{;U7!@7}&X{pYElkQR<fP=~*N
zfASAN38t((XwD2t;E#VEetY_BPqh}86(laMHDV(!N!00mtsPVTQ4kl<)KP>BX!=N#
z1)40Ha(|CJOo#!I%76a&cIaAu{;|x(03)&Vqe<R~Cik|A3F8Ld2u2{GA_<lI@0WOQ
zg0T7J^;vTdV)KMJ=I6~EOB76E&jNp&WmuwP21kZT_s|PZUtxGVTZ&7jxElESGRqe!
zF|$H$;%lF{6e=1)-^!O3LABQbp7`|2c$@>dd4F0sACw#|Q)C%w9J5`4l_skbJ*7^8
zH&Fy&*7XDh?kDIWaJHv-1p+}3IBxphh&ybXT7<rJTw|H@hSfm=K^>F-@$mb5gy?fj
zREbsp^y+oPs{%h)noJ{dhCKY4rqi_4(@8o5hJiA9I@;@<d<p)4(mNa3Fj*r_WKc(j
zAb<Ao6SL`k{{&^eixYu3(P<LX`>6-K1*%!!dF)lWWQt#{+Hjs`$K!mP7Mkx57snR^
z18F=MTNI;=iYXAZqh1X)XqZMdN3oTohdAP;U1FJVn)q*8miSM>qa95q#WA<>?2tJa
zc&@yUskG}I()J!2;h-fl6fT=X(XEBtA%C0JH=<NNXR`=2lom<RXAx2{N|;g>ktme$
z>@=LE?=vKwWaPy>$zOiuhkww+pQy9<LI0h)e$T`3!@m!I^J&ToVM`Js!j^ml1(OBB
z{2-Yl_Zi3*5~v*G;I@;vkuzhMBN~fj2~vkW-KA7a_*6p?&-7_?H1zqT_X^4734hT>
z$efN<Zj-!B(8w~0MjNttcF$u?K1d4h-i6#&5;rO<WuSJM>@{WWt>T6v1VJ#PRE9FS
zIVvQiFgJvX&<lA<%q9oy-#GcFHs$I%_w|&{;hj1}r1zlhDH9buX0^FCE1%v1$bMG*
zxcnAn800Jax29zClI5UF64xlT$$vf0>CRTcGd_dJ!DLti2TD@1ta=7yilfIeoT>!{
zC8zYD<?I+2r>h)L#DIRZ>kzh3m^is?;idzq5?J6cx73ljq(QZkh^~NBQx7Rpq%m#Q
zoRnBnRa3?5#GUok@Wd~1E)oY{Vjhze`kVmGy<PNYmnc`8s%no0xcV*`OMh(Pu}_?X
zryyc_xwV=GO`d89gW#x6Qu~YP7{!qW5xS_l*;g!6P}ais)UFlFilVU8L+?u+6-s_7
zBLWb$-(xrFC2Z%@`V4uP^^l9cBbRgJLad#e+dC|HE%%d1xfmbj<;xHCHd|kU(@fGE
zGj;5Il#azrYK!&~a{ui(#D841<D$bM&yI5@oEdno)&NwhICzus-iO<?R&pb^8pvGL
z9JeWKJd|5Swn!Ptjod1)4{BAkUmnFQ8KiRJR%X3%fnF7D;`R>3PBIsZOhQ@qadE`k
z%rh>W^hEB!UR*#KC=ZqHEudUwUu*IzgAfGlU-*d&Fg?X<3ov0RwwGHd3J-ruDm)C=
zca)|S(#QSVgh?3~lLAL!wqYWs2SB2iwyb&8R<*mZmz=7`v91TnELq8}!k(NJb3)#~
zum@$Bk?@}Ka5IS;Ig^?h`g}-X4~mQhWm)5;i&Ajs()VO`+mp7iEU-TzqU)_*9E(3$
znQ0X@_ybw3MYL7SwU;^fKP-RvD`)8|>k<3KzXGd3s`~YvVU?CiN}vSs?n>HAC@L=a
z3%4hI+gCBR3W~%p+{nv2Dfq9qNq)D4RwU0UN`tw#LHW#_L{EbVRh+Z-zp}>9W0J;u
z5CJRn2grW6jFx4uoFeCMA0vC^s46n$2P=q|w|+Z`KqVh{3VzNv2YG*OhJ<i18{n&O
zzzsO;m)8momI9G<55mDx+Ad5JRFcrK#UaKUZyG^E17bGqQRpp3tYM-Tt>7vqTWn*$
z>{c)_g1Ni9B`RqtrB{ZsK(vi%=pfYuhmHspYsqk^7(P#>87rig{MyI5t=%GAK}HKI
z^~Hm%MN{hSToW-&g#&+uz_b-iePeVbLD%lY&dekywry)-+Y{Tib7I@JZ95a&wr%Un
zd*5~McYjrPt?pj6>QC+7kB-?7cq2hhkTlHyX}T+=ISH?6*Oy1~dqZ|qhFqXtz#5_>
ztqt)Vj@g;Mr$@#)LEx4MbYtvrG(Vr3niMrM#M@fK90%Be?SG~;z|?KzY57V8Ry;7o
z$S<6@jNH4WhFR0({g@e@tW3nox8ZKuE)o0JpyC`BP&3U^3TzEzczADFZqyj0S{{jc
z`~%5VcFaFYkS%xj=!`o7ada_p5K!QIpMZjoEcR`>BSr*;7qDTCQLRfRu=<XEPHZCg
z0cuc|Akf^$9euLH;l4<FQ}er^j3d{VX5$Xp1c!wev3-M4yEiIW>5XC8j_qSrX)QA}
zu7Bn-s*;625!iqYM=1X8ECg;Zu==m{?l><O63C(P+a=Jhxg}XIfwti2r>{KcSdF%i
z%;8d-Yz3iya59!8s8(Rt=;FO;#R%h+0W2V$cM$w(%oTzQ>-K3#*91Y(BB-rWV|G}2
zi3WPm7(3n0QYP`u48GhfahYb`#ig<PqfWbbS_<aF0Eo|=W@Z))oich=)>UtCm<pO=
z0b*AwhE0@%vter<G;<_^^@;!d80H4SgJb^|T#+FbDRrihg_nOHUHQ<9eli{+9~P=5
z=5TjLRYf6^mdC$Tv(d!<h`8)L<^sV6@AhF}E2vvE?@?^1pFgK$rDtnR`BDVR-c0Ht
zb#i=64cs`Uj!q0p=qO|bs?z1qR1|iCjcg$4@FNK%eW63ahG75%UsSo(<AWwsG+~{G
zn%=o05AepTmySa+abT3fvmZuTB1CM09@5$?axRiTV2c<HzoKhpNK)<UZpt=Ji>lT$
zHNX+eP|_y3^&;GdZr22)w>AB8X9ZY&o2DzivGrG>;qedc+9es6F>eUP-x9t<%cX$v
zq>BMJ;kl-kzUw=|`e#N(@G`p_cq)VDJ8`AJznZGj)4IYxjui70C3_;_>f1~$AzoB6
zT=r`Hn0^}d8`wU21i>@2gj&59m7$qORes<O36(Z&(C+r3YtK<mnJFQKe6p`zm0IeT
z0(<B8j@h%B%4cRT{vwuR)ow<PU@rF9Ui6ILP8LoOTd!3dj8M9C`_m|O&|zEt{`^-(
z%~i`_O&)fH{GlcW(Zoyh<F5S~ZAx+fckFarEBn*YZNEzKk_o2Q<ve)U>-@j@;*HqM
zHzn>t?`2%!Oc|!*qVBD*hsQGd?AfuaQ((uKrpys5(}Ham!Fh2UCg!COu!yK5_iiMD
zxv5)`Wxz8BdLWSr$*98&cOf*oA=zRVWL@$hixw)rNH+BuX#pe3-C%{dUXr6#JrXvy
zs?;bU`t8`97o&kpnJ}|ekwQT2P+}ssxhd6ik~kleJ=#QUcX;mcu!T-1<mET{KcIbL
zxcrHBq!gEG?e)b`DR%b-bfS-0vazW90eE6ghNjtCccP3OMLw6Z{Gd`9QWn_$<IIn_
z_{o6X^zSYUdO!F)!<F3CR&(jV*IJ}FgX4tB&mW})$a=xAOgvGC(n_E)#W;Q{L7L<T
z9c#oFF4oW;$vGh9RvBjL2OQmvra-}1>CtbAa@u)`aeR<T2_`G=$CZPpAPzG|P~CD@
zZfXL3A~-Zb=JWkq_UD6}IOaBX{ObK~%2J$Kamf)vxFT_b`6FLytGy8Qwb^(EJNRGH
z_kGH??0>uS;O1dRRVGG#y+-e4d+4s(LPCFg11|qQd!<5juT~tfe<vTfi~x(<4Q@79
z?#hL`JWq;4-TRjqPvbNbd}0|)qw;k61@ep*!<_lLxN{cANA>s;GJg)4J|F4(_v*P;
zQyVa+({pL@$EZ6%DtrI~N7b!P=frW77{n&URZKYYDXZHl@wL7M@}FVFpmFuJ&U_~d
zZ+D#6lWwLS4&*Wt#wLlpYgQ~ZH3au_I;fqoITGV!d59@upHvhFC~t8)N|jrVn8{D7
z+_($K#LBy9CkD2zOW2|MeEj*i2UFSEg$x##LU54G-Vgd7ITVM)R_F-D#!K9A9C|%L
z_hAOHnlfvDHH=^PQf}hQiod|9XxJP64=|9tbB+N>zM_Mj0S8a~eVY7_9<~s;jY}>A
z9+gBftAN@f!@^CH>b+$U-S6;w>t)kGmhy=h#HyrRKnr|N9CRUG0W=kv9zFpBu>ZCf
zLMikKhIY@UduS};8x{L~gw(%xb#=_GdtkbMc)4l<ojgD>>DB}pPFixGyVi70e<Dvd
z6n17kdz-`xxF_>eGbx~xLRbSYA^`9W7Lxr?uh$A{j2J@4hamC!%RdH_NTE0}#pxY!
zB6(8qB53kX(s411<QtWiq-9?|H=d}%!N36V+Sg#KVW-tFS-GD<yJ4?qv)cQOum1X$
z0E=H&G;lWz<!osf22J*6BN_8T=f3?=3C92?YxA)87($dK_?Ze*dvf5#2LEvAP8_pv
zR0vY(@BN$XY3GN6BTXa>ST?yiZI4?HecXf>Ra;stt^PAWAGDGgXQZi~QhmPUF3*}!
zj>E*mJDu%_mQBCMm3y-_#&}hhMj-QPVA3T?V608gOrk@Nk45+SVpD&*g*#<@WVsu2
z3<-_T4t+qh7G36^-8xX|cXlm4XmWskEycSJ6=MHWk&66p=US_;7XuoLiGU<54OS(s
z<EHCig-(87cJ%xHnND+kBxLp}?G&iaQ<&fH=L<Ku`;)z0dA!d_nfMhP$Fzk8lMU+V
z0(v%tHLPX=ioOcUEzz^qw$=N8e;q<xmCQT%^<p5BJ;DEC6x;)Dmx1B%lD2E2;qykz
zJQEH$oP^|xW_QNL%0)NM%Z2iTK-{K173*vcQhr=Y$bq!8Y#>;k9t_GcDyw?Fk~S8?
z_{In%R5lh+==&E@-p7rl@nF8WN196Z3(V!y*;6%X3@A{Y`k{(P+u^&tx~I|&rNVd;
zaPS3q%JOz<opS?042g0N&v?Tt$z_HGmfN0qy*|hExx$Jrhtr+CHLOK{Ig73|)SH4!
zNRX1U{t~Iz$m;AI{-t!DCrysWS-Ckt3n5pji`IU8oSRX*a<1Z@5$rAsZ?NxWJ2R7+
z{0b((;8VmqHZCt_yx5Y&!7CmTi3s0@%@!n=Y~W+To&5&nY#MTg4pzi96*!K_B6rl#
zR6pJLg=1C#ik_5k8$t3!0=*&1Zc?uwq$5b(Kk(<dR?cq8f}%f<9;aD2Cv9g1BP^__
zkXU-g<2C7W0ABZ(NXZdeUxSJs!MSkC5&RFdwfo#EA~J8S(n9p7+VU!t2u=y4+4qTO
zvO~8s&DsLsZ;6o|)h2TCSLVI-G*HRPnBs1S+&T|~W~5dYq|TzGfPtFoib6FXa|;3;
z#H1(@;u@#Q1A<WZiGt`o%oKm)f9Fro9dOKtLLE0oo8-4c$2yCz?()$1FbJGnzy1Ci
zN7MV<-#ChFGo6DpLa@uzQV((d1ZlCpU~z|cop<yC@@pVsF*(9V@&+(Hxxt|yi-ar%
z3_w~`zdIFuMMVgXs_k7m_%IJZWC$fq;})R9p|IIWG!=9c^bzXWl#?AUG^6nQ|MA^w
zTV%WwK)PkhE4$^cS`%Sl^-(^=2N<V~5u&XAUS+ZpRgscF@6QNKDz6eNG!y0oBQ#6W
zdkPr@(qUjwX^?hkfeEO;!RZR9$iH}Dt~fFj6;ZfRb3|>eFs`w=t>a0o55}U{V@Kmm
zg{eK#1kR($S?Q>FjtNpn52~{mZV_e$wvF`uTKycDDUg$gAvXAeJTY2MGE|$z6sc=4
z4vkYsNwwDggeIvrqX;;LTxU)ieQ%K<S`ZNeK61psFKojEQj*Gzpit2nLeJ`}q>Z%$
zK1)~N+w*?uM|ux2O7n1Wj59l46uyY-{A!)BK)q@CM3yL5y36W8i>M`d<nE8rL!rcQ
zHg^JDj`M?*5&U@-n7S@`!vE-s(|c$C<sTmQWAU=EBPjIMG)N0fE@OteCIe)EBmw*#
zD0QBxa|33TrpzMu*g8$SFjJ^Glbli-|Cmq{I8#K;l1VQ_QPgaQp=UR14ydoYJb1+?
zqmiB-7h=T{1>rqw^cGoeXyt|_x5juf5Zy3NXulY<)%R69Z2DV)(ICGjAYUfv#JMqS
z6CcUHeDjz~#dAfsVfdjl_S?C{U%5R6NR!jAv1AXUW6ptYZ8FD5P8H43R~3LkqOK_I
zWZafCdIjwit4$RVSHGi*$ot?IfIL9@ut@~Yr)VGmHvt2BlFLJzjlWzs(!7)4E1EGM
z8Q+wKWe)e|)xK0v85~+=#n0WvKwPn<mU&HmzjJY#w0m}r0VfVJI6YoiTD#y4Bp7?h
zdgNRee8X~^t3G(^F*aF6x1^Y~A|lL})!@8z;;o51_HSCt3DjB(F*C4TQ6C+UwQmjO
z<?iDBNlb+Jy8rPKbQ+XT@KfW^n+NHU?N}W*@Yotg*0QxKu=(S{1k2jck!KqvN?Liu
zB!e*BPRm_y$1a&;u40WSE!6f6cz6y~5ZZYDcW7fE4aDzPQ?bj;Qfr=A)w+=4r`Y!H
zW_pf|yr+w(shQNWzFClIHy?h&q=6$TLs}rhh1zR7(`i0_Fk04O^VZGS)BP<K<vE+}
z7f#qGXG1^TSKIr9zYj{EA$1o0-!!svlD>Ds)Oy-j9Pu3cna#(z<>6EvAVyxSb!<i4
zc-$4MV>5OtK<d>Nl0oceFNd`&Jd=q=M#irL3aG@RnYLxQhvg?h?&3BhGmQhqU$l#g
zcRePWZ*@`%6>;7~;)$ZvP`qsmqL@ZAx_>62t5VWd0R&V<I{)stw^H|OTI{f1vQ3EM
z)_X!{Cjxi~B!Tp721(4mz>#w7SrR@vkw~SQ4?*hqBehgX{wZ)Ti$7Fm-=c4*Vu?0&
zqzEt92{MnSo3e6bEeBA=w!d8RvA&vIAmvlqzSyZ82GdQSAl~FO@E0d^W~Y)hof^Ul
zC|as7b8FI-0D;X_2{n?YSvV?XA6(GIy;vg#eF|XIci86Jjt`^=@I&mMK+t-C7ItyX
zdKOAE^RgS&jfHsApJc45>bs?9=^q?84ZrtvIOSUkr)=}gb9|@(^zto5bGOYfKi!ES
zbl$Q&<ZpCU$ocYg=R{@6my`TbihpRqS`+<LLz@_}W28zjr1V}J1n4r!cU$kv9%;)w
zw_EN1L=HR)=<^DM0w$7wi{8KS_}Y0wys}#MwL(xVw^owSGg_4O-tb~SuS~VHG#I8C
zjC)HDU)haBPRp9BI9SL(9!lpT+EwB}FG_17x-q3HEH}Fqq&B5P%eHb7!5G}&_<Ciy
zCG3^eW%foljwut~y6udKyB*!VTH!R6u)1xB`=p*i36-Qupw-z>g)XR}2p>wjil{}J
zMtIvGCb3u~V6v$;wr+4x1Xg2ZSxn7<?413_&m_Izh>-30KEy)bP$+&VTu2;ObjNMI
zY2V=g-oQKN=;yI#F8CpG!*gEKANMHoL1lq*UIV6mhxnkxNGmkKaa>{l|LJ0&58^r~
zXM(Om+4l$k^DMG2;=6%Q6rr;fcG(yv)(zi|B^DNig{5UdO=4tZ0?VUc)KG6P;c?93
zp06w%2eHYH+@N5NARZ*nEB*0wFn{nLc+Pvh>D*{v#LE;))-sLh+@Jo4Ug(a;ihRL;
zApSdmbH2;ufp$W3-UBIe1hYXo@BeqO`GGTQqpyC4e>`36gZL537yS9Z>sh<zJF#5D
zEH(ds>v8sr3hwQN-Tx&8ZD&~S&K7Fbe)q>=4(4U}($fQ8x4%*hNh(HbGY&A4{jTQ)
zCHxJ^kqm;xF?_&=VZS|>17)WCj`{?lIQ@SXN4(uPv`7(OaRq_<iy3|BZhPjH>8dZ6
zHOqc!i3cx8<;fx}?&A@Xm3?-&>o1ATNi-(O#=d^@x#(0Wi8Tt55;)cN|Gfmv!jM_O
z{QQ%<bkek#A_Y*Vn`3o<ol5`mj97VrcbO{>5{MVUEj=Kw9%>#S0B&wl=$6K;+;GMe
z-gUx$1I+QQU{ew&jPXp1v<s!91~2w;Nx#@umAT1rgcSCp4U-iFcl%=8|HSViz-PdZ
zoWUM6VcQsH#N$09n2~AZXK{5@VLWBghdrFUCGK0K8MyB$^4Uq)SO$IfP+H1TjnI^l
zMuZPEnV3jr0zV)yfQ_*M5it4vY8hw-S$I**-qJdp%X4soh6O7taM^3{bteP+=&gO;
zi_7|bBnY&K7Ca#b;9{7+W=a;XE>A>)Nut%qMebESj@1*W)Q<j4IQI=T0T6P0PR!d~
zU)@NqVk@wPs#|ZeeAYkSq<T&U>BK9<Dy;pm6l&K10j$FTz#l|X)q70h@x;tyP79~V
zIx?~a!$g(l7&y$dq%YP_zE<CHMEWc6T;p=Ia8k?Gkb`w{oG|iQ{0`Y;JXxMhbc9+W
z0yFreyDPi<TJbGV_9mw-)2s!m$rDxJP<rqDYG_wEc1;L~Oqe7Mm5rrqFS|4l?5(>a
ze+Yae8vRDXfDk8X8EwyoJuVbjw29>w7e9Y^FXCfas}TMO)E-2T;;dnimJ}lu&d*8Y
zJ6}TaYB*>9E95IrNT4Xu#yTa@7A0S-&Lo0!+argDjsvi{e$tfWqLuoKgdW!uQR_jS
z20qLe!V!L1>`xRazHLtN=IE^{1X!5}WD9c)-5Gsg1`5P<$8rY7!+#?bbK}qH_sQq;
z6Yp-ROuU9-jp0nc-~+2T-d-x*Up-&)AbkPljpga;Qw2+(pWtdsNK_nWei+(pN?JWE
z5t2sU^hy4L?5b%rzmg67@2Ox}fZw(BW9GE`fB}gZB+j4k64tRJ`vyN-WXpls(MH9^
zZ0ms}U}8pb16%r3^?{o*TPrAEB=iGN{OFGLheB4u<dWQ44DS7B4@%a}<6;vORMoZA
zh!etqpo?pe$F8~`6AbrB-&B!fTNJBK^|N<(PJ-+-yRsbL&SWMf`8^DV21Fn_Et-V{
zl*Yk9XVONwHD33F%{9hFF-I9PUGw@2nWr@XD4duy8oc*jXhEhi`ovIa9&B#>GS#Ks
zY`?#F=!j+{dGn!9xhv$In4XbiWzes5f-K3j;p>&htf+9qYKAD^K+6M(qXwA?#%5gf
z7Rp=6TquEm%xs3(>yXY2Q;L4<y8X$#iA^%R=2^i-5xG<>fjOhF!6dWfeDk-@DM?`t
z$i|RekWS>SmEYRV`S$5=w#~58gfbC4V7SH+w3*!#LufmE5s(kYhE9&JQ#kTh((cgQ
z;_HjUCsl0!pKn4297y?C-5|%;IIzC<23ND!2p#@TV#YNeOS<i@NHjy>nXFB7uI-g<
zS!O(#tX5=G9D4fMD-)4ZyODWX28)&+xZ*1&%ZVjXdf3=He9StZsyXZjhZZ7YzUM^4
zeh`M1?iT+An}0syx&nWA7HHUY4{-<LqYqNhk;9$jXQle(L$o%RH+xv4$<RD3S;a8}
zYy@9C&1s%J9(LPr$!_W>PDB%S>mZ`cb{b|XvCVuwyKT$`q!!j>8pqOYAz@bi<cO0V
zAf@94&Jo`Ob8)_sXGFmOpd9qcc2~@R<XU4s^5imp*j!+5<R2a2%j+JTtj6{@)~f{G
zh0c|Z&gb2%^)UKKi&HYt+ani|#8;e*RyQU2@OG#=c}x3qQ-Q_-Biss~u7|z*oga*p
zhL<FXlP}5MpdqUUz<#T?1IryhpZGV7Yhn0DV1%5_Gk!XtPL0R{rB24fJl(QQZaz-@
zN<9o4L;$ECTP6+o4gGoZI$-M;n1W?0XpOR5W~PF+MS^)x>y><EchAmP5kJJj5ExGJ
z1zsej?v1%b6`_xvx={6G<?PZ2TNG$2de-bLJY5wsH=p)7LNBduE6nTZ`X@McKgi5#
zjUzF7ecrupmZYFCMoMEt9^+5<9}0ArvSO*ftYN^~$g$nmM12`2&0{4S-dhDX$c}!b
z!`NivrB%n_ekDlWDe3Z*-<OuwoZ;%QQ~O1t`4!{Z*6tJU>$q<2+dOvmlgCyZ&_pCm
zlO}Dqk7Xm8`2zw(Hq|pp6e#Rz<K8uvPo~L(i`-GviLcO964)0<O=%Pm7wi0jMn)X<
ziIPB-#Bw1gZ&869D*Q9%Ld+$ZMfxk~N*`#}A(Auhf~#i6yz`p#R&TNuwN;A9f^LvQ
z`sJO;7UN^O=$3k(D%j5M{hI7vs!5Z-D2x}20a6+x2^42JDY+(}$Fl&CTunG3Q;I8R
z{0dl6wNMf7^gI#35_r{iL|?eO2ojctx((0{p>ND$;mV<WnSjvnHJ};UJWswxNQY!I
z`6ZFki&XE9f%EOzSi2L|&2l-F%GyR8r;tz7N@y&-rb;g%sij((HhxEx#gYOh{RM+J
z@cYy;D9tqS?wMCR@={WWdH$|WwVWwLts^P7a!$>yA#lSg&@L=#ywVgm`aii!6@W+^
zd-E|1o}!2D^XL)Tq>Q0@kWxCIV_%<blmn)?e2#N34Mv$h3KW<h!_<3?rtH~jhP%j{
ztww{_#C^Ws<iKwXg6}vtA8QC+7{_6ZT8Uvqjt{>0o}kBw^sI9M_&_<tWi&x){Mu}R
z68=`5qk3heDet}+y3w*EAG08O0*p;`lMd4n@)q+V0+JE_>Az7UAiAC@F?rCG1A!^J
zbHC(#J@jcKuSCEbtX+Nl^xi%5n2YfU7T25}RxsDn@=7XY?v`aTe0i+4o1cC;3v^mC
z$S*Qf)KW()V6NyALM%{@RoE2Fp}BnvZIim2RfkjcGfcqHHhS)0i()rN1(Mgvhzi1*
z(OaX|!BI2}(JOg!7EP}+Sd1gm-H3_7&u&xZ79!GTvr~<)7hW#L@s)s|?l__M`*`jo
z)NS9L<$(WlE}y(nQVSQiLs7DuziA*N%FXCyp2Cr`=g(=$@h~gp*7JfHi@H^uW6ViV
za?*SjQcoIVpc!Zm+>JC60ggG}r|kDR#A((<x#dnBk`@+B^2DbnG*Pz3r;|6scU(BX
ze7b!aaSRpADVymz61OvxsHF$>{hQ^BSbpKDTzc*h3ZPZNJWFbPAcdPBCsnWxxU^bI
ztDl!DRT>b`2?VD-06rbZC5<|??q9p^{JH~~P;MsMR9hSxjg8uw11eY<It$4&U>8L6
z&A+Z}j_Jdt1U<*yNyy}@i66C#U`HjG^T(KEPu6Y^l-TA;4W%_7mEvNm*hs!shxIW}
z+f2>3@{I?u8w-)ii4@sZ|2fmdqLc&~C3DTlRAExQ!bM$5Lzqku$Zlwcc_UE%4WP7}
zn{3LVQj2I_P7%Cw1YR?gFZ5>dbZ7=MNggM)Zx3V`e{%$FYV+1i6f)&NoYF@?62w`;
z|Bk71S(<NZ{~B?YIbm3>YkK~}QDiVfMIsBD__KlWlIR`Tnbj;+XrqU_`GUH(rBPQX
zf52*seNy>kI^BnBc;WEklKc`TRXl7uEj1}aDy(bO>4H)b0Q}?3U5A>F25bUInXC{&
zOkhoo)e>Whww`DoDwy@imi##y`wAQBdX*(Bo_)<IGlAXnq?J0lev^8oBbobU`uPsm
z{SL<k{MSod@P#1Qo#^&_u$2syqtW%Rfg8}Ihz+B+2zY^*V1ZPNJr@&@oSF{&Bcp#b
z+QSt3+9`nw0(@vIY*)YfkwAe_wpYaIdd9*rFpREMxSV#$;VSDwVIMwz1mYfTAU(@v
z?MBEIm>TTT*(?37w9b>Bo&b_#J?90mIgQWp{6uOA5gG%rQ?z!Lc!r{^#BXU^+;BvL
zAT<J(@xjSi_^zNr2D>#z`Ccbcd>BF#Zw%pkj)0pa1JxKbjqg}FL8<5=8U#ubFaIqR
zTHCHyIKmlju~704lH|h_2sq>Y3dU&i^=FcDV;`!6K1yw@`l0tWH1_iNaD-ANs{Qsy
z?DF3mJw29>MuVaPH-;_$QxQCp%Q(MQNl2RN8GeMz?^&(t3i^G*%`+s#$V5STz7|Dh
zF-rgF2;e5vM0%HsbPV0jPX38UO}rztD1pXtomtAb(se1Re5(hUOz=yqFS3Zt>S(+z
zBS0e29pV@x<CG~Y`gHgY-s{qbxxI_*EX2reHmsnQyOf;--;+epJWqi&j_-tx$-E|J
zQQV57-HNC@*Z2m)JiP|vS?vj#+#KXwumT*R2(YQ-fUvBNh;he;We^2qLa(DP*00?b
z#(CdErO0D81!CSBCW_uNY?a5!Xdb%uP3%DaAEJXZ7z0ZFmcuVa2JwlI6^wtIWpjH^
zp6;2m(mE$wqLg~kA#QD59T6#6gh}nv?v4#pH2N5*I%XBTRVKl$?Y?Od@{b%P;(-{e
ze}S+LOeGVV3a}8IOQ*^$oZfgCfAb4G63i`*k-JUgW*2G*8aJZy9aGY=$V1UXvi*^~
z!fvcfvqekS)25ZR4Ha2gW{R(of3!{bSaz}Q{F^39#EGjSj7)(6ByeJ3_Hu{K>RDUo
z{4+@KMwGxWt<P~x=cgd}EhkQ4E)!nd4GsL&5A9oVKHuq@gamgwxLUfKZ5SC-s?VR!
z_Yr`LbGXkbux;@7gSySFbE-&V@l4;<O+rE3NMG(JAde_$O*v7uC27?bb)t5+kkSkz
zu3>o;^veA)bjI${sz=}8{n?ShopGa18#%w&;oq<;>Hd@IJ$Kg$g7GQ(e1^h3_ysVf
zW>;)xj#IC@Iq%3$+e$Y_f7o7P0XYGo8*xSJfmtNvs`Jz>!vITc7(FS&K>gTVw{_^<
zWW~W|<c^&jOBb$Jj(Q5lmfy6XU`Zm21OcNAFLonOK?;sOaWfH`Rh23IhpKYx0L{^*
zmybtzpJ8TSj0W$BcE4*%JQg&6#0hYc9~z*z=M~ecU)-W9)2GDipR(>$xD_|%F~yx&
zo=ha*UL-{N1~4qS6^jo-_<fcS{k<uW0*+n_FY&*2<4SL$#x(U5Rd{X1U-JgJFfhu`
z1Ivg40@O4jCXXPLMbr$E6vft(b%GA@^@o>nhR*42mKYoe`gkB1q8UV;gn>h}h7V<B
zT4hy?NjW2&`DK0*qy1;@l2$+KC0?{MJrx^o=H@b?aydr(oOG#$$7~T95mRvKU?;wR
zV&|u&zp?{43{N>1JJH*!>MD@iZ91tiDODZMWC6HB2lQWydJGIce;`oDQq^dbUvGXX
zt%3flw2qYj7hxr7G!*G)REfpcT4&?2=4?S}a;x~eUoY#sqD!0l0y`_m`)_pie@fPy
z0T5(RMwb6@>`0RdrD2PJz6Jr_ub&6$Tv^Td*OX~zcVip8SA)_%OHy{t?-zGku0JH`
z5B3SM>BlZ^$9|3twetdygwmMMx^}jH;z$p%&oU2gFF>Ykc}*pE@Acc**4$dZ$YQ^8
zz#olUbiW(C5X~M{&S&zKIv6xu7SQ_I>BY2|P;yHspbZs^_8<5HiE7_zMxo^L$p2K`
zK;cvR#4U#W0^*54yg+#-Kq~<B;^aWE-=E<Du;>NTJFx`kbrp6+?7<$g(E|1;LD!z)
zB_-VFfXgLeW*!+a7X9<-nNcRhPdMUM|CmK7;(47w>z${>6I5@I6u#H$a(UBk1u9}|
z401_Klm;P_iyeNT!*m_<=QQKi)URNRj^bG_J`c!wM63Fyk>VU<&XU1YB}MyJ)J9=u
zb%!nm5^L0fsE|bczVId0P}L%h=Wnwmk}WZk0%LVoh0sRW7-X|+q3#_od)O5OBVo`C
z3afl<*Tg^a5l?yTZAc&KE}cXZHFo9N8foOh@Pr~i=7Iu%rD8Ed)c*z@{70Jn^=D0)
zae3}{FcsmPQ8O8^SBC~}f;5<6J2}VsI}!cTg&V;@L<vE)F+fRDlYCqB$#*08v8c4{
z<&`7WB_d)M#p`d_!yjf1B9b)f8A4&)B<VpxgSI1fmb63vO(3m_2(%~YcZ6EZXfuM9
z;p;8oRUbY9n|Wn-ZO0nE@Zy$yy}8)Aax{X4#@C1|SM^G_bBHTzWYkIA$QAYB49_@<
z99i2~mWpcQMiI$bQ5=+Gk*5?JNbFs2Vy$i!laUp)xJ{%C2!JwfC6jYyY&7ATCyX}9
z7l_8BM%Fg|sR!eu>ZZryxv}Bl$35ygF6MBNf9pDdg6J7FN|mA1-sFKI{2_)G>9BqG
zfsh0RSoG04LDLg)>-=I^kc*wa1y-3XB=iZ!!3p98!^utezr&E`8T6W)NRf#2gi&(N
z!UaILQ-a%+VkT=iSs(IcX}47>tdmX<Rbt{cn|oog9XbGN{L0lmr#LVF0+ukzKm}O_
zl%6wy?&+MY^l6rg?ve4<zsSB$Y=%yZ-7dEa^f^CR=_$srTR`L6VT+nleoLd;%s#b-
z$`8v&_DI)dV>pW5+Lh)qLQ-%1&7XLhY!-AW4{?c!CxAn<0ncO}Z&j#=<sD(rN1YL-
z8%FrqY95(s{?A;)V8}@Z1Wahi$pW4cg%_L+=*1<(Hf9e+6nlHObm`y0y3t1`TY9ht
zpu$dR-UTjT&H<?EA2r{6sg#e6XRHK4HyQR^Bmg5FM0FO6-b|r9lO=0c=$2%=j-WkI
zeS5$8=5a4jr}!kSjiHEjiRYRQO|0N}gmi?yVkUX@Z?WC~a1eyg>|=O=wfM^oJvABt
z0hfjYqhV;P7KO~awLY-NSDBkRe;DBkiQ#Cgvyi7Lr1nEyiMl4mQ~kfQx_BY0;)x2~
zl<Le@)2-q0(F16x_Wfw9u);el+MoGaeIi*Ww-;P#lXXBF=E%0Z$#o@vjqCBDJEOx&
zomNh`!cO<Jw^V-uC=?q@QcJAClnJ_k89Ko-BEu-%#CHZsQgaj%xAhpgA?Z{|-wNMZ
zSp_+rnVAd_s^M28tsM!SgnloN#!VNpO3777O{Q?fhIE>F@wiDDBZNNvlM~uu&a{mr
zV*|r`b?te}7K!?~5|#S7Oy~338I55s_iWB)GiD>8o{(m!@oAfTZ7{Q7^PK?@PUuLn
z!{PE+fE#MTO;!3Yh^b+N!G|l#0Mm$rgh`<dR)4Hx#dSaO289Mnz@OrAabXk4m=gSd
zWt^4V9{n?gzu%Q0Qs+~$dK^&o8A5J}`{Uh$la^R1qVJXZn=_xLpIoI0xT?L18lCE&
zewcWR93ItCGTUB_zAg{9O;g+fHIov|j)+(Hz4HkpdZ?;X_Y2DvmB`6k>fF#?(#@Xz
z9(KSZr)FWxO4kFLxvP$7ws1%Q)p`iBon5ijGW8u|)&z#9m8I={l7Dn45#<euB+8-K
z_Q@RYEN4TVi`U*L7?C5pSmWFzWZn|0jE43>6EL)16UAEE(kV`5YF9^rX9htNVI+tb
zL1AQSv5Y~!v?z5F;fsIw>o7QkA(4P5g3_U?Gf(TTBf2t(kPA7GVi~?qV`^&F%=!>A
zkm#yyqd$;?nZO(~?L@7iL2qvoVI8S#IHQJ3(SHrMKPb}<WT%^hr;?#(@UQj@Z-(TT
zU^W#(`kxrna&|4!78)i4sVb*?9659}Zb}0-w&QPOa5b!G&vN?SbGlpsuet(1QQ-+!
zk*X<-YXn0s7_A@=n8ZW(A{@|JV2Kp5l<MG4jBzAy&6^Mw%G(29Npup2Omt8Nqqt~J
zCCq{<nBlH6F%#7EaE6u5LpzLQhVnP}XEb{aQdv^vA9eF{rSy`4DN>87*`^<lnkC{E
z6j~BxF1QMt8-^SvER8)JEFx}{2{Hzp(oq$}eh;)Ho+~uWB(m5G5)~ReZ81rEjCZMn
zt*Se9(<G(WR<x4unRx>^cOs1{l;Nb><6!ZNtTZaA*K;crB(PRuHOxWGIx1r}>)K+c
z>2@;*d)oTUG-p}Bc9z_O;C+6cOmz{|Hjk?A!CQxFD~&710?6(^tOIAIwx$*<I+<=f
z1c_zKdXwd|bBx{ROW0|If)7U(iC2~+evE8xB}QFMX#W29Np{)_km`+gZ+WDI(x%7U
zmNIhV%bqRdQ>T(e&PsI+4tqzCS~e9l8Up;{PptJe{%wswisn{hjn-$%_%0gcDXqJ#
zk@kQ1US&q!D=#YH;w`MmGTkZ^{lZkqQA}!7BXs6OjWZ~ifc6qD8_$0quj3DIhhD{d
zN_|$iAMZ+eHq!RC-B<PoL+Q(M|HguMmhJ=fVdBIM2GPPu{yd~@lHrylZIT@2>Ht2s
zo0qvyFFn(NYf)KK>5JHK-hT@o*##Z->}U1H!nM5}M)CLb3s9qpwY;IyqbMI&HV-hc
z=kU4EdWZZzIJ<nA^6$mujSJzg;)1V6$G2%3YI}N5HiMoUFSEo5Y#q)vOR2O76?G!W
zF}&h~7HO^RUWz|9FO@TKc9cCKQeBxIO!LIz+k9_;k*^Fq$~`8+o^5M~5A!BX&Nt+|
zW;uI99@HEj-j`ny?XF_=5g8M4*rMg$v&GeTxPc5$ygZhb!FF*N`6_3j!8D~K*MBH<
z#4A}LxjAMVzQh|-vJ)=#W8+Jmp@vQa6mz%c3gdSbaO6Y8=cNW{G<r#1D8R=5&Pndj
z5!*Kc(GM%_+p_%}r{x0B#-W87zswY^N8N)?p2?fhE%~FZa_fdPnB^7yNk$0O6;AQV
zLX+&Y*8Wu=i0sieiw7V}nJSkrGc*XbbEA4HY?pLC&tzF##|Oq;x<A}qdeuF;nQBiQ
zyKs=cm>EkvD!?iaEhh@w1iAaR6-q^d59KNY(KHO|8cq$dOsCYc=Qi}UZUyRoKbX$K
z$1gpr6dL~pc&`?iU01V9G&E%#{_=h<pW#}&Qn4}(C8439bC6}na?%;OdFO*StEKR;
zTDvWZUnqDI)U1A#5C2E;s?JyvxmU%^%@&F3ElxK!5-_d!h0xOgjRBGrIn7qmkGL5N
z+;W&(!JYn@iyR-tuQ2n*8@tBD)J&GCkIKZ@i!jGp>2k%&Q|K5)Vf{+ME2m^XN1@Gl
z8AM?P$e;tH-G5}9#vH*>Ah!IfKt)OyklR5YUh_^`M?@1~v<kxEv4M0V;Q+hfGJWz>
zFntJkCZv`@$%h|T3KY<g8_^DwgVeMJIzT0j;I}$2R<Et+(zyQHL2|j9hg{;%ZVz~B
zp3{(c`g%Rx(&4czYeWg5jB{X+yj*av;#{PUajs=G?zJ04)IoIOD^04_acNLwB%9b(
z5RYf=_J<c`n%>y_jn>urlw+#-yjA0pKsza?cNL@4*zx5i%7nA`_*$9YZG<WgY|!|6
z*^%%2D4y6?%5{iWZ-V>qPG>-9iLoGicM>J(4`<>r$!=okou?@gs}EQ^gkgHs2F`&h
zzhlP2e&r8zbjV~Not0c)8T<KSzMnu88Ig}^HU|m)sObD^@#k~bD9Idl`2cMjbnaVP
zlgv?-1GCZhV8hh%yGVV#A?QLUYX}ZR5O?cdjM~Q@1avG&T4P<wNtRfP*xwiG#a2(z
zxqLxPUgcH(S2oX)OfU?JoLmFMqev!Ihb;x(@6p-tu3s|;X6ab@{8mlXDZu4>2aSw;
zCg+jW1q-~=Zp6lRP&YR<uJZAnK`30~-72ItpB~3Bs=d%p<rmi?7L=<UOP)6Z(fJXi
z=<Qo0konAHoW!dnMX_*8wg@$sp~k?5+TujrV1%@mrGcDZ)YlRi8oM2d7CxdsQsx8X
z<bj|oxnFrA81OBg4eKJHfJ-4)_)J?F;&>z>te>|3VtQ!P=vOr#m~NB2P86Z9dO-9k
ztc7pw1lKha=S}PaZ0mv$^1C3`wBqf1m#ddX5WSs|_NkKPRp`MMky2Sa#<OxCzzmw~
z>-r(%|LMTHw%Kyo(%!WG<hVinaW4aGdP}Zm`-GD7qo9JK5Hrcd@P1pnKAXUEuWPI)
zref5dUtaCTe-lt(2EZ3kPkxJ+VCE~{UoP&Zd0uMm4f+@kGNnXAr?_~;=LqE57f%mv
z?U+b%bQ(qz0?Y?HL-J+=DpTa{tu?hoSGf30$Pyl$lp2i%@WOtau+V6WDvSda;-Wer
zu_+4o$Y=WwJ55A?RRdb}Fc)GijqUJ68Q?@?B*~hcb8MsXQvw{Ydo9CpgxN*dag<#m
zd`ubmETRB%qBB#Q+62TyVsfIlsS`)+VQK<6!X6<3(`H45JaPh13K67^_`QbT7xR7`
z)~w`@y>RM=>g`CfgE0oPVL((C#aVtpJ5ff`w204KWSJuKGX+HX6oc1(u~)5ZP&tNB
z54WiJ&Dx-(t~{llt!3cbj5&o>??Th2e`uVP*oLAyD2ayJ8mEzoM@2bSlFo(HREAj7
zz!_OwVi}fdrNzR-UQy27xE;lZBT_VaavX=8?PZG$y%iZdUFvTMV_=5m_!kFBSgVq5
zIg%f+O5qR_X#|IS7CHAVHhG*jf4S)qrGm3KXlcJJEC215b_}TUs{dIaWV{cant|%_
zE*@r=S&RIMBvUxwr_p>`dr<5a>Z7YtDni&Kz{_FEyQk@L-Tks+BY8E0cA^e~D#A{|
zmP;^?k1mVPBI})@23!VorX%f2@*1(Jq=DvdcRComY7ei(Hq1|^&yIU)<6W`*Wb)cP
z$PI^Wcz2AtU<?zN<_ya|Oq2kLmw2q~EOBZ136sWA9G!sxduVmrF@hgNZEMU(Tu)4W
zlK6dksL6|C1}Q#0lhH45rz(o8FJ_X5A}j@Zt4QFKyNL_i17E1z>*t%A6cb*|jw$MI
z6N^);XgY;y8zZeoZOUl>B^5I!Fx@qm^`*2}dRg<-Wj*Fi70(XNGlgU%Zdo$NYJLjN
zmHtBv?ita^ufh@FmyR&Ty7rUb+!*$<Uy$$IWGF%7eLN^Cpazo|c~!fN&JD9`0{o<(
zjM@&eYpRSd29B8Dp4jI^7!P8@0t)5sIg!ZEmX!>Wee40!t!g>{ByZ?t)#}Z){AEdA
z+X8qX_IEPU1YA)fMn!kn7n)!GD!9ENd+jY+4Y54th=W17?;>U~L-)T!j=04`o=|-v
zZ*=O5gPIHvUyR4?>uUlsmKA%e98!!3>=U`T4p`#{fG3`8?L)osZw8vJ{y3hZ55UXq
zH#P06ZLY8-uIH@nHvcPi+5GAs<&mipKHV&5;ho8<XarOH3NY~L9a?8SI|l+vT}n(d
z6b#wAlsY={&Y;uO*va*I!t2;PKOuF>ep%}y>`8d8{tTmo^T{%Q)IZJivZDiFqY#x%
zD`ODL0SkJp;`v4s=!?*?9S?5T?U7PXC24xWxM8Yi<AfbnYVl}3Kz?m7QNPo}HCe(4
zk+dN`W0%*%9e_^{2LL8ud=mSFex0nO1lspWQRwx73zYOKOI1j+BgCL~)CjHC;%_jb
zSzsOw-!r?^<@&s*qT_BIO|jdmpR-6!vddYL1BL$V#!y6bZ3OL5du0(&a4<FO{u;^l
z-s#sl8Nw^6UD^i*7U_)9dp-|civol6{cA{c!U&G}j~1ct_!&*K2)W|;4$uqpljGUr
znX1&l?NTWzqWZJWPZ@r9lgG^+C48v9F|uZ-VfA;0HH!=zEqbi1r8VZsiE5HWONwWd
z0{w(g<u2uWORMOzG1Cw$rWs~yH#5M6r6hE`^u=P2=B%mNrN3>OGSj1O1<cs5)*UMy
znC&{*;sD%p$2^`p%jug(k$?#~7?%z<g;{xkZQ`Sj)UMQI0K&*ldkfKlgSeIl(NP~r
z5NC@Ir4gs_K~FL($GBA=1A=?8aus+iP~+3`3!Ke|C^HtYf&-s1r7AUy7o+N-H4wZ7
zH=_3{cF<2yUfCvkwA1X7b@nV?EH1k3bsay6*1RA1T>8{w^To|}eTTvyd=-#QLTA44
z@Z9_Ge35ko=?zJKx9~m5ASL;?|L3R1>-YOu&t?+rk#1z~%n-SlUv%d#Pu#%fYa+tO
zGVlX^q$6Setcse;5Fhps%nQEFe;7bHH+KXvh1Zb0c~vE^&L5*JAX-)*yoG7sktIL&
z7YW*jC`TrtAfES5H{!+-o*w*mpOvZ{O72HRe}!<x$k}iS?1K_NrpN9b<jj*<`(!y8
zg*YaO8(?)=cUNxh{sJe5Sl<9D32@*i&!)#Y^{aT{8TgXaNy}KAkx*k8!d?Uv1`3E7
z!kEkic9>Z?q~EreXX~#%OK+9&Av>-M$#fTJ1qi{d9-YWQ<DCp(8D#uvDN@e$$sat7
zJa9Ebz`Dp<2v_8B@o7p)U{+!u2xxNRsnhJK72qk3bg@_aStyjyJ#hr|30G~1fHnC%
z{cZEr;H@?BOgu2C`1hot{~|dWsMEb992xpBGQJ-06$)sF6RlmS$g^2b6i_gmn{+&@
z0x%%(Onume$-^r0=p2ZNOUk=QemUcaWOt}rxF_{_QIV;?O-7IiDhwXl$Z;qcPzr`^
z9cJEe-k&@!!#A=(#i#?1PmXaa)rbgAo(o9ANvJ08G*_C}mD06wm=_9bb=hA~*E?1c
z^@juSU<n87*5lCd?@!O1Z#0^g;Dy$S0^iNfw=Vu<we{KGI>ee^sV=yV9}0}|ONS<v
zkhK`Gt1ED?3L=eTn4^L)k?WH{L#s2%^vvU5FmOpn^^6#EVyOU4Jmtq7=P1tlEVv_>
zC7QmCQd-WBslgBtz<A@kHY@EAG!<gD)d@U1+R@a}E{}r2K=iWfu&h61%6JL*d>)bh
z-B6pSekxXuRyvBH1Qk195uN&&$^cmzog(`Vt(+?^+32WES1@DqR59wq*NPt+Ly?dQ
zrREOwR>l*z0xW=0CyQM4y6r6I%#_|CivCT+$yTRgqGl{dIgJ&1g{&jLjmYraA#x19
zFejgh`!Tv{TCRRwE*piHwNHr~@Gr#L!y+jms82i9mYsK*(*|0#Ka5%*`5_k~FlpOR
zxYeF;OSF1-IzIiCfhvbL^cC(FJlUrX$NS^?-13K*W(x^eTv8Ez4DUu@>fv=Jm-=%B
z{~Zg`QvRiBhlSi}$Ve$^JxG5r*sU@)fIb?v))TRT*c`44(T8JGuid)eXV?%*rZG74
z&UYp)Es>$j83!{Wi3zVQ<G|0;h+D^4M^!^d$^|g>0>?%B1sCE30T4iTT8mPk!uMjt
za{mtSSakySd#^oDArCjuPt+*4Q8}a#jrXI=2BPRjluLQxHp5=cJ*|A<+UmU<C(jju
zhh!o+RT!kbmCGeD)1y*b1G~crID|++@rk1we%rrYwY#5?OM}zAPYcVzc}FC9bPFy!
zzH9_Z<~d2Z?v=CX^CT<ALET1zdQwP$fng95EQwGHGRqU&XR#!cy&ExEF?<F{6*FJ(
zW~P8&i`$jeW9gE5#D4onT^#bl_i=~+-x3->4GoZtVEi5b-rBrj3~L4h=4AttoSOj+
z0O3gP+2#Z!+uOoUg5eQ<HVfLp%EE$KrIID{qceRs?m3cohzXIwnCRL6+rj@&1{#p4
z`M)~(Z-xu%$Zo$1WK07USSz*uIRRf>JPs7=wqOSK?VX}${W9IeB`$N?rE%<lyY~R?
z$XvI~?S22;!`<9lJ1ZF(x1VD1RCk6CuGlMFoV7YP#}7T0j6YxAuWM=&Mkfx1(Xk2^
zuk5c|b8KB*bK7R%(<nV>TqgLysP4HVktu>`Nk=QoG?0`0;}H^`nH^0L@z=kYW?EUv
zmmrb8*RRo2K<_vsxT~w<r|0HL=-s+oycd>#-2_?mlh(WIVlFeuHoM<Q3VlOSYqNZ^
zr%4XhkoIVW+{rDJjZ;HyAn2$>BGTJwP*<b{fPGd}jntj@7Vn!2*WwTC8yiwJnsO6q
z$Xt5PxJNc}E$|3eK`(4FEcSEfbEZ7LlZt<_oZ1yry*wp^QtYx(Hn!qvW6Yh^KNw;1
zt{Ee|(E9ndmBBNCCc0bh*M@{DJ%}?p*nMU%t&YR2#`0H$G<+D9N=#}syI|V1^SdEf
z!1=-N-0`_TgqKw&;ur($%S~sQeHSvp^7-kcWpo;~d)E+euJzO_Lud7#sUh1~WGLQc
zMybT1(#Gx)PkI7kxO&zX7x8;=b<`z^bV)G5;I0`7N4qz({!iK=oxihDYT=b}1bk%h
zHh4-ZL~4><e*A$j1$e3hqTWj@Y&GW8#sejn88Z81fo8#&z{ovt#L8mX)PkA6T_iN{
z_7;|&=Ie&##Z|jX3_K{s;Q|@8TU$(L<#->C_c1Zo@WF^{Le&yy2eO)K@90}&EIiZk
zLK0~{rP&;bN^!A2XT8sDA1k7r#6)^9D`9&mu~%OpUqo>xxGDHp#(LBjy!z@p6pl`G
zn!!=!m?&x<lW78+E2KYXaLd9eP&2kkB{(Cx%l?YKOC-ywE)%UlJ)#hG_w7~dL_J~v
z_nl;ZBbSGth_SGpl%A29t;);v#LgF<y&`&n;DNUFW@6(tRThWr(y*KjG>X$-)Xx7q
zHeI&=NI>b_eeB_B<$m!d#lR-|q-(Gps>3U#ipR_Q;jsfuU~azd&Gdf#p!h@oog;l(
zz%w$I@%emol3d(9UY|5f8koRf_K){oXB7W@G8$qPLVP!Qk&yAmU&;a>l0Xm+PCDRh
z=l)}(|I~67X^(!2l|un_hG8w38;g#w89c=`(ZLTGVIyd!vd5jDHfJbJkB~7CAt!Y`
zIsTD;E~^NvvwzjNjGih9^B<@Hd@VUoc#7I~BC|{3ue4U(U7PR!2uD6rPLdpc5DTHF
zg+8hYgw{A$swOV`CE~)yK|ZS*c-f`hc%)YlA)XV_BwU@@RnRSf=l=I^0Ev#@6fPUn
zup`eoGOrE-DhjMgA9XurHF&Pl-n7iGbA#q17%_REC+g&@+hzzbB95ubEnjMgS(U*3
zv6dZ8GR_=xqd=?YvE=TorryG1*BEBLAnUNB+K=4eq&+Y+OSvvpv_k~ZbyC@S3}dMU
z`Z7fdG|=szQwi-|W{89z&(FUvoB;s}vHP_!6v|4G2sDeuD4hr8fdt!41oOr51Zaph
z=c9<glH%$Hv*Q6rq|JjOe=vnT`*k+tdNa~?1#M+ELN7KcX2LlV&+O?NYMr1P)Z2rQ
zw$WyT18*Y}r4?y5AnSEEfm|**{ye?by3h#*9>qm>k>BLqh03(r5$waSgt4+@Jz+L`
zG^3OuLk)5&#9W{Q8J4#E;8HwzXHg@gv7iOezu@(+8fb{Psr+8+tv(qX-Mn*8d)s84
zihKfDyeJ^mQZz|fBS$?Z?1do;VIdCBS?;51VLRO$%6jUG9oFP=qU`Cnr5CyY*VE6b
zvku>MuDdYOCBI#?<*v*@l&)q^QFpyyOe4;#W!{kzG9QvzahAWGkIqTuS%_%nqtv*-
z+z&^1`6F}%i&n+G*dfs!i98EYf|t<-78jKouA`hv#D-G<w0ar39M|xic03Je{LNi8
zT_CusIOq<JUs!}t=Xs?*F|_b=Vyvh8ieNTmf=`HJ%dOh>mcn3GzAJYhgPNW+0s`bY
zmHUJ(3<q@aJ)tiGNF}cF9iMjzYY#aPWk|*K(@ND!*UCM-N9=I{auT*6EF{)*SsBX}
zFdOK0_B)-R<Yj-Xg~-rIeG^wA%DZSw#!a(5jUoI-beV2$BgzYcADt_7k5}36)lk^h
zkS^n&f)SRu963wtQX6<9NxzOn3`Un)M%g}(B>^#0R($DAhNYl!@XO%&A(jCUFP@x|
ztl_}J>(fm&uSB_agg>wQLIS`t5oJC=s+zfoMOnN4C{8n_j`e((`>vYl#R=j4qxmw!
z>vBXx{za^iVr|ctgE9;x0L7#hFkRQ4LI+}^X{n<a5EhV~?Ea651S9YuVT$*{qo|=z
zlr0>lCS!v8>-FOg5K|0YWda*$t%j6X?(?y+QRphhGb%AWSyzAMynis7!d1`x^A)xO
zHU`H*%a|&YGI0X={yR1S;$cX4fXn7OXzs<vUSxc;c%d1tWn|+1vn&I*ySv|fQ)v)7
z?47LAkN$(kd6#8UrF7(7C6f=fbY#5b6OIvYj_1p1i7`JOQdk61M%N8^u|<%oMK;ri
zA(M-U3Il4mO`NU0Y4)=2scTTlr#2nHMkC_8<bOnrr~Sy2*4OSRbsCI|13susUJ9OX
z9KHD$s!n^E^J`5<6vG!}t10T+jCk6+03f4!e!T(g_TZ?)#nZzT1L@^8<Pk!eN=tm|
zn|iSm%m;veNZIGe;{^coI+^0UBFnbzInhaM@y$vV+Z23N=JW_o9&W+q{)eV>49_Ik
z+I2X=#I~)8?POw1Y}+>8*w!1{wr$(CZO)v0d++nBS9PsI|LCiGRn>ZKtWfW)9%+;3
zE5STygZ*oe4U6(XI+@L@hm=p{JC#p<qr=j1*$}7Nh**?ViRbuk1Tl9zuCWr44TyuY
zjC~<S#0`qLCG~gJ<Gj)1WN2fysKZhAuG!4p1%DGo)smo^%zS|QSE>P6uz8L!n`I_^
zqH&L9YKRo+Ov>AWzrwJ@0QScl5j*4_MK{9c5a=G>)G}wi1lU#f&4E6AfV8edgprpr
zT8wxEo^CtrH%La>KRL;a;O8d9+i0_quv;lXG=;7xDWO`Pe!xe%YLasQV8*82h&eJF
zpCVpqUl7^}))+uYUl7feg6mIEk>|F26tdi(>$eY}oBmcoIba|`<%LvoivHu5NHQ4g
zSaz30HPsUDW4+RtIzm(`zE72<i+I&QT2Q6@mp4|Cmh83%)qy79cTa#+UtH~T=gCiN
zuA}AjEYM>Io4-rjXj5KFtE+d~oDY26MRUHu8l9l-RUrW1L%5mnaLIIw)*03r^DKTr
zM%z2RUDc?`aCu2rpZQnXi{ty;4l9ws*yu%-vr?6clVB?lgwE^3efG;~Gp|w6Bf4aI
zbbU#)YH`}t<!0i{MgPVmi?%F}!qezqni(Zt)We2m%Zvk}=G=yDldQc^TJYh~k)ITd
z2$#|F%09p+jJmo1LbByDtU(ECwchQjIqtc<*-AvNO6@=fL~psnxQvF}ou!EZ)L*aB
zw#SY7iQLAyGKm1I7I}evt;{h%WjkoJ^aWAdvq`>UJ-9Q)2bm*iP=c_EICKj;ox<-=
zBq$GkwKH|23c_hwGXD2jPzD(CXh!G-(R*Fad_@3dRnA|snAI$QQb?Kz+jOXw8S&ib
zWW7kWeeS?QK5X({%5ER*#A#S@LF3?w33e#uUiwEmJH0K@q5#Z8(1pR`91MadnIXj$
zrD#T3dn;5{tGxS-#ETI%Vv%9nw1ef2&q(^`X1ac|L6YP{oMQ5B32>MG)sRrrMZv$d
z(LDeQLn}xmDX?-mm)<*@2w6L3BV29gkX0*$ARX9KNdYJao7t|$G7e^+G65J#<Kjk_
zBE5vojVkuGo0f59n#MActY1B()O@$^&qq9BG;e_=f}#wSS{3e=$t*`enwDXN!Q*e;
zyrQZed515HPS(%)yS_ZV;pw+kpuX{Y*#y8PJ(+*5=V|383dyu`dPMJGGjd$mjNP?B
zkDI5r#XpPZ^dHI29Q|LEFuMb0_`jwduryg+Dyx-|K5f>a2gRG!?F^suj=nYCP8@%g
zA0{?d>=-_pp!~6h^M#YE*DqD>be*9%8&o0#jeY2^Lx%8Mq&zgbtcE-632qKM)`bb-
ze(3dhjE_AN<-2Iv_vaszy<}}Kzh(GM{{uZ{5@`GXVF%M5<agx2#t7W|W>+BLjXBb`
z+*Q}?&fO3}2pY)(6~bUy0YX`HPMYc-@9!6qyWlYZ$dL@OoJyogKTe-7*Y~r@Hdc3?
zJH<@!U6Z?x0r{_cxc)t8|DGUsk7lW$yBqhd+ct@V{Z3|f;aPV{Z^v-?-W&_Gk%?Ui
zBUJ59N`^hpx3>#K`ON0tG>$#1S1dr6kI(zdLMw3r1gco3@7psPECCe~Tqdu-2u>*w
zF&q{b03D0c_DWAY!HHnR{n=;N^UoFM`ty}f_}Lr1-W3iBKcF(+8WyHJ9(28EM-#4a
zO}D9%#C@|zA%(iuBe@l<UT%#oe_@MWftgAwrGrFDEucoPFw6?lbi!#pvHh6GzS@Wt
zHbRVemLgW=Tn;+HJfy#r;f&={ra<t)4bNl=>5|uEUi(ST`Pf4j+kx`OIIuMwmQ+ls
z`;(cKcK7EQjaQrVBL=c-T&4m><IAKa)jV9I)NYlxM~Y^qmlWjWqLw>K+ci|$u8CYj
z39D=N!%^9lYUbqjK$HG68jv^wmI$D%;(4Dw$1qEfq*Fo|o1mC$s$uT3c<Az5oM&xn
zQ-meyUgOO^MMKZ}_mT8IJ~Yp`oimCO{iytYIz{zwc5lm)N&_MatDiy*7?l~QQM6nv
z7aC)^FpT=tyEL?Xi(Y+Q4Gtc$S=Tj_*)2*(XQlssKGJ3$9E>})B8BQ0LG=U9Tkcsm
zq9Hbrr{{c=I&N5}C&xx4GDuVeZ#7{aoPX^Rp(!8B4*uL~bQ}8YJ^IZ<5}m+~x}U=T
z6aqpuos`jg2#nT9z=k}i%-s(XqDYpM!uTDy>A*5>#`Y~WUdcWNQ&`vMU^<YMCg>y)
zzfiNhIqUS206~%Av#=-lBiR(7mKblQhKGtKH~!Xbj*^|cY<x^o86deCNE;U!Pv=F}
z_ge(E{G^nIQHJH3k6HozW*z#yDD=;eKoq4)CrVB<FEMR2t0lwh99Fjh1m4@{_a}1}
zjT8HrzLkF;wYTw^bBHl|9<}82K>lB$<nnYiDccXVZoQKkz3if<ox5Lviv(y+)b_MN
z73J-vE(^G#<nxtJ6CXQGmZGWYF6;?7j)U!aG}cuorAi~=aR;S$7atMxi7dx2VSu*Y
zD79&juD1D@g+61=Ag9%sW?pc-FuK)}d8#w_L|(Ngb(SS~lzHyL>OS<Dn#0~%nuYM6
zoPh^8r4|Mwu%aep^iWm6i)=y8R2W_urp#iSCEH$6dF0t98Eak^jXM(D+xlu6E+iF9
zw#{&b6t(O#MF%=xA*HC=hI+VEnlwQH)P!;l@XupahD!yb(JGNJ5@az<=A<JLJ$|A&
zUHv9o{aou~{mTA$IN(s%LZ-Q8h#8ZLFZ;+A;oKdb@gyLvbJ-h!RF`R}fRc1rPJt%t
zj!dr4ZhY$Yy}+)DuCKz;m!EEw&i%)AASW=9*e&OF;18&v<q4VTB<wG!)VD!rZ$}Oc
zu;T$G!3+cY(b@oO7xo9HiRcPfp&;0(Drz5@gHn<}ewoY)tJoIjM5WF4OF%e8y4R2S
z>2TvoHQkR?*-jLIg)4?V8i<o{3?yD*&)vmSh8HYJ-EhO7@I0LGW1Oxax|sQu-pWYv
zqe7Afbv{LYu6Y8->O0YtAEl0#?XPSGNbs9pN_EthlFF(Ev2XI5o^B;&kYh+vqcH6I
zP4`jHN~V5>D2JTjI1QS#5JfcXs;x<ClS_Zx8MVFK;w`uVg2~>lP{JFf+6U%ZP7$r(
z+o?i8`y|q*Dc^{nkqEl66+fBeZKzX)H3-Gz3x^UUk6t9knAPwYf2njxv7L_z;xa7L
znsj)a2%{@TTm8D5wmN1?#u1B62eTV_n-{_GEDgDyqu<{GxpQ$lavhU0xxz1F(i_vM
zmq5XzXmSk#2rs4U@TjK#$i*gbt2{QrRf6RrcCU4koIMQ0x?ml+&@Id~)8q@lAQ<pD
zBBYY<B?m{y`MxkIEj$O_Ly!#qFdSbE&5EU>o!hv;KOv3W=KKgC*!S^Y_D$Q(gs~f0
zO0gDDOBINOv5oZK(o`1r-9tW4LNi|Nhng&eGW`Oe?SoGls}A`FZy8P*(QylM>ElEM
z@J59Y`CC*7#D^d$Bl1fL2V1>SqiqjkJSc>2N)Zs+ze_RoFsE)iIN8TS!d@ryyp%#?
zlA+E={Q>5a;h6}=7TioT>4s0YX@gItT49O2;zowSa)t|ETu*D$vF;)U1<Sb7HzS^U
z+YtjSXGbStd?Il)C9nnskV~%RA`=y&4yuq7bootC+uU7bbF^pnp@m1l`k>#IN6)n9
zX>OHRa~imq^v|!p$SB1oQxoDE_Tlo+yCW~SKKT>P4oZ?3dOy$1eBW{TKS7}a-Aeuq
z=IxAge?HbM=!vX%fnv$pLPW8uTG_wB*!Yt^Kr{)@?k3J{!|d^Qf@C4M8RFAf)~}Fm
zHVq>ogJ}xxhWG>5Vquk#$lGtNk|Wc^@*>N)Yf=brfxL0BoB#}OgdOXMoEH{3JLYN1
z;$Smch!!Wf$V)PInUnq^_PHruw#ASa{tnyW@v17?FA`&M8_zf~v-nOe2?Q1=5eYam
z#|bDr_T?X`_eC+(8KghUU80c{qs3MJQrxDAw?`Oq`>$veQFvyhA0xGI+wdYT7zB2Y
zvy?p+oL!``-2)&rnc<tJu``FRj<7BMaPt{GXz)Tu3dEQ9J1p?3D3*RMthBA9vhaa>
zt~lj)3En4pxCh6<Uz1%MkXV0fPGWIANDiuiJ@7C6#sLaX<#v#)$7km!n_h*<sBEFW
zVXS+5`RdT*9qeWW@yo=)3mb*t#1!&}&h{S60i?}4XaGaN?BimoI=T9Fl-K1%LQk+A
zSCm&qVs>acMfDGTuu$je;q<UM@-y4#P3orVmWK6@GYKC2WS(&|{qk$}%ndERI4^Ab
z)IVBqAToXY2{Ko~_Il4)xe+2`2dPD3LUY%Lsm~as;c3S-7Vr(-+3QcCUo*My?2T0z
zjZe#2odDCzkzZN&^F4n2E1JsJR*S%D9#1<K!K&@t>|W=k;sy2>ju8y#>KUKtVpgyt
ziq6de0`X>>Am@^1zro?GMOdJC>E<w{6$zH2@?0Zt6dMliw9yFs;9msWjKA1n&vi*p
z*4-Il8;?R5%9NMKrl5T5>El`)Xtivj>inZK_5ibXDi|IuFT|%di@uZgvjl5<D#nfc
z^kO(}HY6uYij{{fgFSm$Hh6FmHw>3e&hxJJW$>NNN?b%=tJtZciyHsGr@9-}dZ+76
zC=(GLTfcd9Czg(_!ifz^0kJ&e-nCaa6|fO$2rP2ZaWTyqSi$aPk|tC6Of38V=Iuy|
z4EeRs<xe@Vi7uK`vOCI-vTiVCs?24Q(iytup&xZKCQG3Hh$H5NbXB@k%+Ad)9wbh`
z+`kQn?R>rwPz`0)1L+fCH851u0Eay%(}n`Q0(&A(%^7<+Hq#6Gz|2HgDZ#>oNg5{M
z2($W@Yi(B}d5364T%Otz?_U0flB4s(a+9pB1?!Ux#_X|mU5;o<x5)UARM00sVBskA
z@to#kpaGB~3Dz9&r={aSl$!5hIU31RI@ybeKfvPr{R1T5*;;^kh8+r>hm|-9wK7W(
z71e|fsRTjL>B}6Yp!i3rf-gT3qMt$T5-S0Sj#2rki%$&LU|YP`f|%)p_<+q(>&^7o
zFN!&xtZkGcz3%Lw#+&$92T4r)*$H9id0`!_vj>AK*<eQ|#e%i#hZ<$twqGm4JqO}M
zh<|SOWq&+gucR9mdO{<r-E2<#;4N&DaS6sD_C~Lgq5=7nVX4768BKte$*?x);xG&T
zv|kAg!2^2_R~HNaKxXn+1OESiwU7=S8t@w-Oa*Atnn<{yK-!v7JB4D%%kxla(>D2t
zQA*ZHSSzX@w@U1gV1N5!4GRU4OfjAjHo_HJczl@s-kY`B9eA-@iEX*LMu8fr1S1WJ
z6@bdc%=`;hB3AVSytgrE?_lG0BHStOtZhCpQ(7`M?<R$R=`9jMn$OP{G_wZ_V&|XS
zOf+D7VD(qjgbz66@7P=p;H_A@(&f<E+@exOJVO)clmEpx3XX8wT^r0y?xW^!!PnE@
zyXP6L4kFR9&D)I@j-4o!)J8)jMHUdY&ySJYlZ{yVxw1yOxW?9HDV2F)<7IeF9*VCF
z1@%VeYI6$-VPQ?)1jbjs+8qSt6v=ctghzmtctU!>5U4qpRH7M46BW<&XpMsyN@_&T
zz0YA#n9tlMEyd$*0TnjN)#u4hPe}1m$<4!KfR<^i?a7HAZPDSjekxI2^QJDL<KWlg
z_`kjQg7RR__v`D|Oqw7tigxO`=mlyOruIVlJo<tXBW>Q9?72y-g*(i~u4s;n)P6vU
zpSAHDYr$dNWPp2s0yUXZu~M)*sGTCCtwRil^M?C_ZBQ2u|Abv(0b@kh&MaO42k4Xp
z4#J&$u4a==DT#P`N#qZHI6p*u%Vpa5&Ux_<*TpJC_5j1;8_7N7XnMYb_a<zOfXcpd
zlYWb0GBl;#dip<wBKSI}hIlO7vf_ZZ5~9+{OQ~L>DQJ2XLQqUYyT$#|m6I$D5Rt3(
zw-?R^Q3_*9%$uoE)Dc69ehwyQH@sM(!fXtT{?^3F)@nEX+vXkgR#T}o+iR~ASaLK4
zP4ujEwM>^oiw;sra%zW*={R0?;y!8Ix(Gbv=uUcWU}Fd~SLOJ*-v$VCksjcR4Smi=
z?jc|>A*(q3Z`jPDFw=&%X-`FAl#Xo6Ax(-h70;S!_AKuW$5Z{AQA-)*mdBK@oseQ<
z;uXnKX;3ZxF7n{OYV#r8HcIdoP1Q7QtMOu8E{mtJ_L|M2*Xt*@rV~+<@sT7U1bzW}
zMQLog$uLJUYSOyNu`cDcPa^=D!J@$H{UoJRvZ!T9BXP^Kn>VZ_nLILH$%K*?4wOjx
zg8fTH$sZP6d`+LN0(#dXK6J7~hIbMQNQUkRA&W$RR;*REGGsdePFRv<Q<muAL+sR0
z@>?3*D2-IM89cJgCz?cJquL6sC5mb35)m&6EUMR4&hKrM7|*F#ZymtqIqQ!Xo6fba
z-29+-d_wpl+vWFsE<rUC<%nM-jl!m<-wRzVYQ_bcvQ0tY$@^fu!8+LW2kIrn0csn`
z12lbdc&oOBk0oFk&bdX9`dA8j{6g|V&5KGlI#=sFT5I^O&zv@*%tTfOxto<`)Qhs1
z-uqX?ywo~b8`K3=*B3w)wqyIn!r3PnQhX`TB!!UP*6+u$xwG_<hdD5L=g+o5pX4Q5
zFy<oMeK|QKXd{&6kF{G*Sbm{UXPQV-$Y^cSLwlykK?s7~Z&?GL0kC_n)-GHF^8uRC
zIz|vjmz<Rt&l&CCo1BL6U(oCrK{)R^pvF)`>d=;4!wQ4^v4Mb$(CkwzBv)&(W(k(Q
zA9-m_*F$xRulLabDedAcih-O}styparnQ9LURfo6h71SmK}msS8_@`;MdBCEKlr6R
z*<5~gg|RGJDzxLPdDjH^XyLd1LLqq8l?<RXXvSmEaNIn8x>)O~d3cp@4Pvm+_}NQu
z1GVAM0@NhsHpT-CF5-3F<yr)c9HgZiM<hA6_N$@9aH7J@IhODTGQ&U`_t`G!6i1&y
z#mwwW|06?IHYOhCJp^3W!kwa0<TRK$%v(@QPFmoxfSfThC{R7DF=l-$EsuB!?#1nM
z$A2EF@qfzw0_08;sQtri_nla{-MiT{1FXOJ)sh!H29X06{Js-3iBd@;amFxEq`1Y1
zj0D1?GadRm7b+9Tlew4Xb$I3=dhnA(GZO^MY?YItF&hF(Ct};ks41U9ESAnvK;;=J
z6^tm9N#FEsc}EC(2zz9rW>d%l+740T?9cPQ-${dIX>0i3F{~JjfiAk6W0*pUbyVUF
z`O)K*4cP!hT4SBLd9GsWU%UnZ9g@}$-0~PrT>CCnUQijzuC+-@Q7KMtD1{WK2jg=m
zo-UIWmvi(`0KBlyO~=-pzQQoGWoqb%Y!Le3R2rZ-Ohn}s(NPUgP`DR|v>HW1paN11
z{oWTnv)5QvBABkV2D(z2`AyfKy}5{e9i$S@ODcc}ezyX$6B045X{E2vCrmSP@4Dq@
zT914@Y2V%=Wg)EVf`qI^aV1~Mdr(v3N61PEzX3X`Fh8oBYvSrCw#``ZFlGrqnXV*E
zKUT4_SZ04TFd5?5(24(=vFI#Yj^8n{vykUck>p5)(hD<Vd<#DPUh`4b!Mr<>34PjO
z4g`QPXxZQ;uFzW&IQYqQVNBY&M>}NrsOM7YoMb63cXwy8je)aq=&eqGcvuOMSSlbn
zEyYA%*FNeNdwfA4xmY!3O{4A6dB8qLh%pB~SAZCGoG2RGZLdCl$2<(qB8Grc|7h4&
zm0qQpY^|7d=2g|1uJX>#hnKPf&L;sto(J$HTYf7>rRs#!RVHrzbbw*`+4qD;RFp>b
zk~G1@PBWzyz=0h=beCN!<<b^r?I*7<!~0g<EDR%O=$CkdELXhmLTCJe{Mrq}170La
z>IkVqvFjo2$KE#w!Hr7sGnZ{~k@>PqRPMfYjb0DTusNWQ-I1^&0xMD_sHC+~R|z16
z8gnSa!!Z0la+z@sf}RSwRXrOz)#;PDH%3b%@weN4TGCZ&c;(Mj_6tN0MjH6sU52r5
zctV@_&ptANhIRuiS}8N+-g-H{$+87>F)x_8+6GJui>;*74i~j~&3bmdP{tqUb3`>H
z{!c>{G2Jhnl|0hVC2mE!l$0n!wsL^P4n)iPj#-sh_ehezmfm@d)o=Z)R;E$6J(p#h
z+L9a^#|-|_%c*5LiSUnTvWr&BbvD$erU%&6lh%r)Y?D5YLHW5Z>0;FHHu4+7DrBBH
zoNk5+<nd8Ftqr4U8Vslkx6lIGuirY1IBVD-?*qBoH&IUjR#ZTab*e{<)&XGS5-Ouf
z$zw)TBr%oxzGh-jbGxK!L2-FfbtjpMEqnxMnATS^u^Nt^<oOL*hT_tkG|sj+b=*;M
zyuBe-F<-Xg@17=+9?~j~;1J`tErKO5#GxplTBGyrF&?uUx5Y~x0Ft$o>C-k-UU~B4
zWsFe+knXyI9wqywKb0r=MUWhK>L0KF8ora&z6fNn_;mi~4-Jfq9hk>Sg9-$l{TAx@
zM)sBi>`r<Bf6pXE(FYG3|DE-$SG>%W00f#~EznS*IltaF5&?o#l#TWz92(&zg#S2w
zyxyHQtcFqOn5qAT0=`~-Ew4XOWX~Zol+CM@O%$nDr=925WlssGC{(O5hD^!X6AfAB
ziWF~45NSKQ@{au{nmJy%U$7=ao#b&teB34h<8q_JVN?c{QMp)Rbq{EIR1G$pWSqIm
z8wH3oI~9oF|1xv{tE*!&LC2$YD}0_>UKtCaHx^>I#Il<#bpoRT3ll6y$6oxCU2r-n
z@v|FiGa0jwI8fsaf<FO3DjlI9c$Ux@7IY|dY_GzybAqoZMAHBb@#M|T9J_Dqt}GJ(
z89+yq+0@>K%Zhu##QrZR;jaLL)z=Y={^dzQy($j6L~}%tt8yM?&VvM$-G|<>OLNal
z7a?+2PEmN>uDB3;lN!(it{XUepYBETi6v%ta&b%FKuMzMe_=TJn_k3g=b7HfGo$Ud
zW2DP{skUYEo%7`#so^hbrl!+Z*@y>#v^vS64%82`#=O3{LF~mNZ#bsiDZ|#=FE$*~
zt(Plf&GCRBZt$c7Dj7bgp|@ZVpJt_ogZ{$hx}8P@_t0qugkouEF3w}x@t%2SsgUsl
z>=2n&RAY0PA9wLbPCOmi(Bmaf$x+j+bEpI!7D68TBaaR&He)rmRc7fyE$l~twMSHk
zmS|QjK#JnCG38Ozux|dYL8x--GPxVYou;`CggzyX*ugk71Zo*M`S2KWLE=Lbx;a-<
z-n{BVi>r*9@csq=n~5YX%HkkdF=*~f{~T7JHB2-nT&5LX?-$+hnzUbApR_;DJG?*l
zzxm*`M`2Hi7NfW?Ci@EV-g+58diDy7r+BbE3qLJdUTXg>%7<cE4ep@yq^5%+m4ay=
zXK>YwZgweJIJwM`iE`62bqTpn`5;6xJ9W-<z1_sB(RC)c%)r>4CQSM`h?1hLVEg^Z
zgFufc@)A1eSSQdD`ITAJV7M}~^0)6_<_PbhhBYd4@Za8#4@{oZ0bH2?Yvk(A!g$8u
zJc-rhB0F2mYq52-n_25yje5A@Su<EEn*L$qMJ&O*--C};`4m<Z;3BB;WvpI*<?ETV
z9}nTyBLw_SXtkbWjn0oP)t;l_QYbEDtoU>0VYF3w%bak(P-<!=@kNI+s})Z4L|o#|
zh}y&#h5E-L{JAa1GRA-aHx!-6y(}`+f<ZcG%3?SR-HagpiB`?nJ~kHVxmdv11nQ9n
z3R0KY3IZWDOhSNY-eu*|B*p%EWv$Gwz4Am|MIQNkB0Yl!eTYKlHU=A!{S3gCmg{l%
zf;*}~RukjUB->GG!pMZfFr~luG38d8rbTrF#n?xClMq4R<!K23^E-{*E{!M_l}+wD
zoL`vfiBT$LO?0Y;N9ZA-u5Dr6rzj0Nu8ijRMro#d*+m8<NgDL6noJCTEOIWW%L97E
z3P!ILA#f;-uEHDlROrn9RqWg&_mfSR*r=_3W9@n3_magXJBYu@f{6x8WT~g|il}Cd
zrcH}J$n3>^nh=0tDg+JO(%-%W47H~g=*h0FK<(g_pwOZvV+oG*N_|F-<!Z3E8!S;2
zwKrNKT3Qs2WHFXBH@IeN_n~*rLczAFD&w|!z-WbjF|JZnmZpUDqXlu@R8cVTb>+X&
z2i13lJi@HaL#49|0=0ror}FGV9g#CRC5TkP44fl7?L<KINJwvDM_q5`9fm+}Q18IW
zVJV$8Hotu^rC;bT0~A-(JxFoXlR+HF?oL0ZnVozxrHo{1k?MFfH{Jl5B%I{GjAC~l
z-S|137swZ=X4Lmz6OqojIals}u<j~;nvXI6lCxxJnX9q>Y`_Fn{vJjb(QKZzj5!Pb
zsS|NT!(0w%-L3MMHol&cEWJFFQ-AA{wongnA1k6ylS|m=4ZgA`ox*z4qS>F9SRP0;
zVw_?@vPPt{=drMTwzI|3uCr50eyu=<vv(l2ob+1D3GWcvYLb-1J=sfHl(Qv%7*d_r
zUVt-0e<cAs--~qk>Kog-+q5A?!kI73uiC-r#>WISWvCxp<VJ9?T}Pcat%9K^6QhVs
zxh2!TR%@0k%Xc69@F=|dREJrUS+4Xi)InI^T6Y&*(8mtO46s*hG*6}P-n`i>39+}p
zLQZL2rx}KurjwLX-FZz*-CGXm3Op=f!s2{03Q=0CM!k+6!X_0(gPPAPebQ<b5A)_1
z?zsWc`{vvE$I{q$VC<{fY;Zi>CvrsuiowZ{>kqLG=L?fLKF!A8nmrULl<-ct)pRt`
z-hu&2zXby(_R@;S*{m>~gsfR;+=1B<y*k&8L%W}c%N%aj*P;q^Znu}?*V_G&JW0|5
zG-7(*{F#^zKm|pGr6M4BgO1Bag-ZpSjyGUTaDX0)ES<H?lA5dWed4D=dNQ!#{@tb$
z#wQqD`srO?;Zz#4V3PDGu;kTziu6f@yuQk>RFqu>sfd_eHJEBOWSU<QU6pEMyq|x3
zmcEy^Le|;7K&v%h8HK=6n!0|DzG#%9#r}ygo>hyP4mKT)2E$++x^;@4|4KO8aRxvy
z^Nk<5%8#mWAsWfAEzpEIa`Pa}u+#Rl^tSF!Cc(wmim-OFHi(Kw=Rv|OGA1lrrA@Ly
z_aZb=uT!JSpnaAq*Q*1+-HiRE$)S3QG+apVe-5EE&r*(BL%QMC6|1!I-|po1>fa^X
zid9^2+i!<>R)gW2nWBb7C4fiS52ZNp^+#WAO@`&9vra;J(ZXvM<mh|OF&XxCO{r!!
z1>dO*#9`$p7L`Tvx$zg4c`F5Iy3?VHhsSa6kqz&E%YV35)(0?&)iV|Rr4?4z^4S-+
zfX|~0E}7sNx+k#ntTiyWJPaQ&Ch(rrQVHpd^NLMBP_(>G6tf`Ki~!U<6{Q}1w$5Nt
zy4^{dLkhs(#qc9gp35X96jgSW=$QSX-)nYu)-&0beQcOfGua_dMYQ5q`+Iz4oXF@u
zIfl1jQVUqJ=hH@H|H|mUqgrckQO_g|271fh%nG28cr?X1vtW1w@YI+%Ooi7T(PBXK
z-d8n4L^mY{QQnBH82}cN=eX8@)yc;?cr(5~;`&;E=A?E{CBh7&MPvd$_(@sGF<k2y
zf-*G(=787*@LVCe!6NaQ!)iqMPJ3)G2zF-qS`L3wpX9Rl)xphUtS<BwoPhS#=GdyX
z{e;lg4MrfyX~;Tkb8@1f2SoEzWBt!AL&p=C8XgdwuL^sUc0k^8jsaD&e22LH!+Ay2
zxna*!-CWN`K4~pi;s+_nh8f4ZHN!5n^|jE1ZPE)Wjb*jdA@Q{-PEnt0-N#QDM||45
zuwaO+Ox82o)R(O(R+EJDu;8H9_c<Iw^+Abu*k@=rR7RWn`p9S>>qDBG;jKk$HHzYc
zg&Lh^aIcMyXTS?fR{M*WOo?q*=-DD)u!qHnOi9Y>fHTgei?~T~10fGn*41kG+2!!G
zOi7=v8SvW-m6ItcXvSyB`ueV5UA^$&{IBUO6Jzw>7BN#(sb`lXF)}4A9b4a4$amv{
zW}N@IZQ2aWm<hNVa^rXX0{o{1b5=};!=&@G2xV~t{`ZLg*Ri$kAoIdnbXMB5shin<
zyG3XwVE!gCRx;u*CaTK~RA*ZL-I6J}{vuc?)4DO>c9A!D>tk9dBY$ShLU=a$?$3y_
zctQVT%;9^kTdQf`a}AW@Vb1!R4L@^U>~1LeemVeCR?Iq(X<;okD^)JAOZV`9L&>G?
ztbO0aPm>!t<TT~($YWVJOVy4|oA$jNbmm@N`EQHq<F)Vk`Cg7X<Cy>Z=e)~(u**8h
zdj#z+H?wTWF|XD<&k4^W&jq$oa$>8iwTH_O6G;lTMd;#I-xKugMDs=mXC+RTci6yk
zZkxLRyRt{;EseiCittL^mMPrv53!PflrMgGkqZ9IBWn5<;?L?cr=5PKQhH%kq4(hY
z0`*&~*!({uSZgoM0*Bq1DZvfy9*Ag4sf7tm2OY2A(V1<re^%lTUw_{o1v2`3>i!b~
zfOE1j0YTZp$bbZ@h_Ju}e{4Fy#whaNY`r?d8baH)hK8CUVnK$og<N`uFs0q~UPiQF
z2j`UEBKfE<J1=QP#*wDhU7XL@61&7)9Xvb_lPm-1HuqxKCW#N@$bUUc=*Hj)(K1MC
z4dsx)s4+z;A&x|n#1-D~L?vi`PztN2&9j`=k|b8<o1-CnU>_5CU}GKvbf-zkf8!au
z&N+2gq7o-tkPbsp8j%i*F(ibd8HY<mDAutV(gdR0^hcxwOyV(t7Lgc8GF(uo-&Dzf
zX!&3x)fqA}JM?!^ZOBV5yCO&zw&LZWJTi|t!wfjyiLghMVZ~J$YIP>>(Q9o6H?0*r
z&d|D=By!O_n<TCYjR&y-cA?}9h1emeF<~A|Fg_du&?dV4pfxB>v5x3)+eYHN<1-y&
z1|<@c9#%2-1onf;>&N0cSEF6Dsifmn7@XoW?}stQL{M6cNsh{}euptl44b@cW5-we
zA69OHqiF|J)CZ$6!J#~&gb3BgwMtY5FfljC(jaZhK&#}^CF)-R^fg?ETh$m-xGMe#
zrqmsQBpJY{<HAldvlEJ9Cn(K9Cow~zmvvwg59LE?Yl7;T29Obr#;7=>Twdm$G`P=T
z?*ieL1yKgon}2K$Dbv}fa$%b73MCn^dsU#IzZlFLle^VV#6>(oOFF7n_RE2~X$*Kt
zOwK^x_IH|46{_<Bj$jgip!R@sL6AXXqj6^vhcvl+S_}xB;HglANUUEC5H*fr*K5Ce
zlxz4w`a#RSs^7l^{{3bU%Q~!EsDHdP<mU7Iix*Muxh9kGw)bYPO^e~9(uLn-!<R$7
zq4W2u-p6M(W?fUk?I1U;NqrxrTRf(teiYQ2tUqM}5(WMb;OH8RfC5o@fB)^_{AGv#
zuR^TJX9kPEw4>?U;Z-L#;qq1J`Bg>WxsT79u#vN3&m~;d_P?9X5DQ8+y6mQ>N*(B(
z*ZScO&pGYMk*E9v+P0^f_XXoeE2lT1Vnj0k25~Hb5ef6OqjFk3kCx+#+~5Ph%uxG9
z{BCUb-RdeR0MDu?<Ayb#Grt#82HTdX!C3@i@Z8XS|EY9d(BtWN6tj#<5Br1sQ)i`q
zOnHx*T{*C|cR5CXuJublYi`uuRn=|LtWk1T$c_!0+fx-^*W22p8z5&|vJ_ddHgZ*%
zrV_f;v(pv-Pr%1w<6%gz;@o)Nx9zobWJ=@7Ll=+^ke9nyXp0g9*>(6DdHl87m)_Cb
z#8QWWp@j{LutXHG1#k0f-IO0&7IVp&xkCKm{K1$#J^H03Dse?jo#_XDaN3v^r3yXd
z6vgFrzx<Ra$3tR7)FquN7lraS_0rg^+9t^w+Bq3M-j=LiHW<ChJ_1K~LwEsa9@J&G
ztTqlEkh<Yj`{!GgGHRn_KfPhzs19j0urS6AlZt9Mm2g5n9!f5}F;*#m3uKrr7<~|o
zRC0gd!41)lFDKuM%p$FwQa5|nt&{$%WDkK3ip(-8#qvl}U6<&FzKQ~bnltAbKhHP>
zS?WVM&bWqF0&b!o`M?J@)T{tp4->&TdwrM^pw6ipQVlyY9fANmBDs}0DSo0yad2?=
zXb8$f*^ro>1^MU!t^(K8zIZ!lWp|a6iu2-sS-T_MiC9L~;;)>5>pqRW90YH-+$i^M
z{jrN)eX>mWdowMBfBNH<GsA{xec^mFbj9=VkLTp|2gArSOvlK@3SL(OaG-Z$?dr$`
z0N_a=?6c|cIv3*_tYG2l@ZQhV9^}JIpQ=qiY6Ob37%qR}oBkfT{)3VYVeRIs*LlXz
zb-&t$9k)=X!a{VSRD>OOQ0B!79+P-IjnrY<+g4g2EnkDJj@@qg!<F567H;zVS&K!0
z?=P=SH@z)8s3V2D8M>BQYq=qjnn%|TpwfbVTaZt{wqzFEEl1eaB%S`_ShuEYSg4~I
z7Jl2B`Rc)x)T}lmt+=Ks3QpYScKNOJS4d0G+$O<P_k>q`n{W8q+CPiaf3?B?YWjVS
z)>?^umeHy`QD@Q@0vlah23l{1pK>4d;JoUzV*mVdZ3(to;aoKstM&ch)CNW^0qzx<
zD|J%fCSItq{hddz%MC^6t#kDI)YIF_F+0+^s4%a*S+z0|&x7DMkfQNZK<0`=iPL!`
zK9<1~Hs-o{xUM2x3F`S5_{~pT8av3>*J>adFfX}G@WIPKJp1prm>}BEkb<h+-40`?
zr%NDT9+pBXo~6GUHvhgEEo7>G06=>J^;G-vej32wa8rE3XE$z8iV%*<HOq;|F0+Wq
zPSc86j@!jDK;nDEF1&aa_sGgoo&+R2!u?WP+Bh!c(rrw<vk)-w7)U#5?_WxztfQFF
zGWYB^y~Kz}>1uwwrF`ltt^wAz02{lwd#zGd*}k%H3ws(q3VR|~&UU_@0cGPe(*u<^
zLYE(LMc0OspbAigN#Yv_C+^{S$$B@)B0sfz{-l1@dTw>vm8+eX9CD7Ipxwvbt>e^x
zLOZi4xi<Q1&x*y^bcwe+kF*5+S>0v(;AJ%YPTZN3Xs#H7jGVZ&qf+OqA~$T9yL^?*
z_W#Dw#!4NACXe2zgnx3Q0Ul5ywl_;|^26MQ?=9q4+!ch)*Z8}ptxQhKpVrrAg>GD2
znQSOWWZw)N?wk1ev!bUl3+~QxaUU{9w-D1(jb`uLMoeX;HBJV#%uyOrP@e=YIxW>U
zLx~|nyQnw|tu|};okx9>K@5J6??J`gm+#zRAJXaJ#}DW|@`U+E0}KSi(BN?&JEn0%
zbpsVi{vy!Z&6(%<#}m@>njM-jL}M6<Zx=AZJ&Ut3{!<V5eD&V;&RjP0qA)SWha;sf
zNclBb_>myr0Zl%o)Q#!o{dLvJXI-p!;eo8T(%{LDeKYmFiHQjda4v4gbB?RQ2l$l%
zDGSz0U^d+y*36>pq0P8H1qe_(G5@zl$pP$314jksHNnyVrV>ag;krMl-ytW%2w_Uo
z9aT1r<W1}{UYnar$U1;BMW(+zzMyf;d#`%Z)|3&PH~jeH;`qZu_C7!c9xT1I`S2RC
zEA6SOaAJ0TjDh#c!)!`&s#8N+aN+5N{E7vMXtim>2B~V(_<QvlozavMV<z#J^2pRC
z8twsc)0LM2P~xFbJsAx>DtkAICEf^-Y4F3nlc`5#c#+iAIG<WMC_yJfC+jhJ+9p)1
zsnCTHM!}i9{g|ME!=ev%wfv)5>p2iB5NcO314&F8XupRkkOL&5hbgi4`@BaF5pcgn
zm0#Y5|2J6n%LE(iFOKg4WHl%cD6)D(-xs+)R?_V>fN~WZtO-;9^9zaNIx30C4O=DP
zc6C6E!zp14E({5m!{|k^6dO0#?Do*~u6E=sw}Ou?RIAbFqPf~UQa28G!yW2HZ0pPy
z!lv6b^yEp{;}0UimDjrr;fGg^*W`y+5F!uz7Q06zj^N00yen4xV*u1j`?PM;h-ymS
z@$R23AU(W86Jd#YoHttxY&<V2%3wRlVptNv2HOPt;GsvdKAwNeZe6Q4>iypLNa<cm
z{x0nGY4b~~m<WV8`G#5|;WnKa(xtQ|2$2T{M~u?I@19tP>9NqyAngA2&gazA*C*SB
zQCGsLK0+BEwPej4GC0oG4C=sunap|yA0ze(kOJq8xIu~|bS|<+Kwl5(3${^^9G|w`
z$LuDu*~*e25PBXT^~bi@6aO62HXApmXE5nR1fQu`*}e+krd95q@trvk^qF!qj{5C?
z+K3k>_gBpW3>`+f8Q}$M-V)ZOaI!quOm<+}XzDJw*LYysfBYFYIu>4<GCvZYLrM%#
zy>t4yO|c)?sY{Y2GR+BU;Mg_AEg{%v^DtyVpgD=a%)K9O<A=C@W}Wfpc^JQt4nyJm
z(;`og>6xwyni#;Pq-CE_Txf9_FYS7xnpUM&W!|oPU)LgG#9t!1|6sZ1Y38bX9KCef
zJQ<4-?@9ZSC2KYGHPY2sGL5oUym$&gvbPSrPKb7Gz@S)Ss$)o9D2n1pvoG2EM31SV
z@n^uF&Zot)YeS&AR6`hSVbIC6)-oI2kBXGy0F6hN8=5xh{D4cytVUC*`ni<nTdvcR
zfd_H$Po$v}!J3lUd?m7QHxZAGysV}v=TKkmbobhD&JQCZv;%hd?)OMfhbk*Td&2y;
z*;v?h4PujFPB8(O)y_(2`mck|AzRl*+$mNaLzn{(*1Ez~nap4~HI($@q=E$)(et#K
z<SrKl9JY%+X)*>6sZzG*iFq-YK$9~Y$D&J3rjbp%QDyFyU058RfZB9&6Hn3J;*uX^
zC<*UdKYZs;qJkqsi~FC3HV94uQ6_78(Z6=0>|g4JE#?IJx#<lVW4RCDS0)#Tv+82O
z{58>Mc*ruMoC1u7JsNy8A5lL9k%n&lf>s|H_Xqk0NCJ($Ts~eNTs&!b($Qz`;~V<z
zW!Lha*e5E7$kEmng>^c~k2##IW-7BM%){J@R?zjJeDD|qNOt7vYiMBsrZATo+q|#0
z6F<FX7j{&NKHmm0mp!VUI#=m;G*TyC=hHRb8aMHjYE$G1IMB<eXC!x{jvE9|y`|*D
zRle4W&wLz*wDpvpm<)C(tyPQ&9kE#fl4&TGUO!WN?o;2Jb0cuG$lgAY`*yYWT;0NK
zl5<Z2s8{|y`4DyHIHs<k{>X?-Ki-)8UBUMBFM#y#&s>cbATl#F8aNjR6A(F{z_LZ9
z1D1ytn2-X72FeNCV~3ysP8`GL0K$KM@qR~s4?knYUmX{!on*YM4$;1bX+o4u-*>oH
z7z}}7I>@$1chvK_@fe?&L}-fD_VUq`$}6wBL0v2Erj4dzCfRdm&;`>cwxvj}qC=9Y
zEh<v;G@gbirbYL5Fq%0ar*{3u(9)WcuVEB-7<WJW8=DEgioDm!d$pmx0KWZ^OF-Y9
zxdtnFsIMy9wDtW<*Qd);$eS(<Mmq`Q(~*_)n%EiOzwX8SIgHf9=8u&ZOwYVR{drJJ
zu7=V>z*ot~!BGy|^1OPcV4qVU3IU9+{o;=$Z0fW|1{i)kuzae11Clvd;L*Ro@5?Ik
zzHi*K^zbNDmXp2WVsUxN0K@5(-g;ti!d&B5wge-(Yg%JUtk}b-?4-lOKrlvfZwjcE
zS$-)gaqL8Im!#Yb;uRLMwqdcxGwO5&EjUV>C1_%SXO?dynN#(4%{|ozqvdh3HDkAY
zHs;JgPzxLDo0zex1AiDVnQV`^)M-aDa#2zMlwaSn7a>v+x|=8>fQg#kCT*^MxLrt-
z;$46sZw5J{@o>`%CSSyN(9FBVSm@`Ec48YbxOJ8aANVa=mET|`U94ZgRDRF<pO8c|
zCEj41Xg}YREP?%V(F3cEW}r7%)fTndH^uUf1C7n}$9TujSc7<9itZG7d;RznltRol
zxEeF5nGla3(bMDefNEFks<?*kJ__3g1P#l_!F5)dm^=9ia}NsU?8-aOoGGUS{wVA1
zr*YX1NG-QH3faW9$!|{T#gB|wucx(sdu~X2zAu{8do&jH>svI9s|wVq(%9L;1-8dP
zo!~;EUpieAa%%A|RsHuMY6hY?E6b^b7c`7Y9JBU&1ZoEO0H^A?Z7#JvG91+c?5>*D
z35uo%8Kab2^^D5e>J4Mh^p~)Y>xMBdI>g$schL!U4IzOsiV7Wsb>rwG2lh36-w0XN
z4&Gi0tEm6f^;5(nTrHLc_^a1-0+$>Q%W`TAC5E?`uvV@VipyUW*KyBo>rUH7x<$Gi
zvdZ%w&H&weYauy_OO5?_EBVx_OO16@bQ$GmK?buJn9WYFAEN%%aHy%{q~kljCtE#t
zM>EheaDVkczOq~PwsN*=l&foqhT*u0ejFF+2v71r^u4!<^W@IXhmv`7L_{a;dGMxO
zM4BtubXDImG?SH*2=gSSgW4us!kNeqCpI#<nE|w`zY<*E>(CGXu8QyE{PaBgsMamh
zfaZ5L3o(j)q-1%#9aH0suxizaN42Hbj&}WN#P)9HESZb-qk3|c#m8|rsxK@(1uKd~
z`c2&P*DP<aO|DBo2-721>n6_nq@b7Qir@yedHx{BtnH6<oZ5E(XO<E&^A*7yOu02-
zJq$pumQ?l65@n@%Wag5eA~W!*KjulCFKJrlXslNpSZEBDI*ju%jD6D<k(!8JCNh6a
zkN@TMLTKq7u_*uShvjP9&*b`dR6I+aOIy3X)NC4zBe@@%;~OQt_=plHJxu%e?toKE
z<3TF4Yb0MuahfS_cT6`Ne0Vqy`Ha%Wa*P%*Hl=HUvrc!q?g3PD?C4C`x=fx<bk1?J
z_%!L8{$i*pbn_%Xf~2)9ysNupC~?km!SdX5cLXY>>lt9Gwr<Itb#73n@nk6yhuQF#
zY1b~0Z|(7@dMB8w_iQGj-H6pyA^eduSTBiL1a;}FGZ?R84a$~M&<pWiNlBbyEXf6w
zobdkSi{0jU*x)Tt@%Wu>_3X5wtOpx(HBxf%+Z?BH+_#l>zWjl4oVvL<=}WU3=4L__
zH9BYd3PeL51BAs$UkSlwfrznZ%e%O@Up*o!HzGb+Lge+RuLo@EhyQRL<{ZZD4dFSh
zZR$x|ClDj}a$rxcM&IN-sYrIeuD2X8>gNnj#GiHs=a(3)`&%8^;M8bg!Pih=T?hl}
z;Lgn0qs}R_^@!_}yhk#NA6uYozu^l$FyLO|j`#|)VkbXEJ93~5IX<#I>uFcT0mz^p
z7P430)sLj%hL1l=K^)Sf2tE{DV7CZ%Hb-BUsfWHo_#S}SNE$nc#>E~^K+Xy%NG~xF
zfh7Ysv6^%Ym~u?+sAKK<9n;4cBcFt@8R>4jz@qqAwcdK%dDMG;fR?3&chGK~(NSX}
zQI$5IZ%a0JJALtgrO55ABgHPvn^^^|mMr?M&bjCa{6s5I=j;ek%eh|X55n0R4RZwa
zYLq!Pnx-6xvsB0CJOfhzBw&(e?y@I797zXJ=Ourqh>MqDRjh`L9LYXxJ8x*SGzpms
zlBYtEsreWq<8)`Ld`^+pm7ZBLS0VUIz&%C{vX<ChluB2drDJ}9H>n-g1DWg3(@bQD
zaZqMqjYI)+w8v0fTyjzhar-FX2aVOWO}M4QQ-Z3jc3EV_xjANfBVgf9ymS|Pp|!+(
z?>H5j`PA(&O5!b9CoGcb14oD4y5_lz>_E8VA(*%O%~8QaVHUkUlHK9al2D&`KdI6P
zdRbRDy~7CFUEi>^K?T<v#g_k`t#}(|*d_}a-(}wiuWU5)VXvPVsXb(SAM~B^g(_dc
z_;pRn1q+N_^cNMmH9&C`uR{c@X9P8HSQwD{uwxi-c%Rj-0nv+1&>`GALmL@~+^o5<
zQIkb2Zu;v=&ONuU1<u6!k<wN9r44=aqG`ofixM(rW*k$9rg#!*Q%y0LQAP*hatgsY
zJi22(W3s_qf_k1W3Qx+V^B4R2SJM88rD>hTs1@VBpXDZvJK#8&m;JCPvUQQi^EFP5
zb6s$AVP-wrkklk)9qKldxW+y+Jnv#==fNak80vBD&rRD92++s9Q?!8?1q9$+?P00Z
z#=pgTeIQQ_1#9cun%zqLQS#8VB<0X%US#H_y-%BN{LXU{+n;gf!?-O|qiRlnE}33~
z5B`%S-7a#F4)Ey0E_KZ~M4S01;_j$}UYv6;jqNlp*+dZzf977%pn29CeHKl2EN@(q
zw1lV{cg^Kho%S+;S^Looq2tsyhhe=faL>$)s5r8DUkkn}fP5Q10(yw!sJd%{>;<YN
zdXsN582wuDdGso=ReH*;W+G+qG{njl+cfxbAetkK65wjWtg$I(1GfH*D?TFM)i3)f
zF5E`xfEwvb?sO4>rYuMHl5Fv?THS@7Z24(xIwO5+M44-C@TT;cJt`w|Y+)!_$}zsn
zSfgs(`$2cQ=YQAdj@qR@9pEjeK8T|6b4BboqX{}@Vp)GUyk(<eQ&IeTQG55cC{8<~
zM4(GI6Cf4OWSqHz-UMc!YxxXf!83JZltk-zU_~1k7{Y3Z$JUFv17!C(%(a6IxltP7
z=^`%gz5TMZXsLT;WJ&R`e*Zk6=55Y2d4Y<z+u*FwP94I$`yLq5bNUAYiCPh@A!`@v
zMyto`s_dBd*C*~8*$f}7iLJ4dv!jWD&9?>Abp-qIEuj)0!-WR4uVkeHX57It0zlaP
zR!;E;I(oRby99Uvyr1_W^(>P#jQJn0%b+gy?>{qfn$F=s7NECAUatRh%G#q?+%ZPt
zf2XXG`{to@^*~O<N&8{}1tlfEVF&$QCf2#Rr^t&BXr8lK^Hg%N#qsN+?TbHHSxO&P
zV&xHeZ9q>^%V$<%LjeTr#TkFZ7#Z4kdIVlw!+U2VW-l_&eLvTHyk49Bj3#k4{Kft9
zA&hT~LY1sy^TCmSm{8mP63PPl-@{S>+<wpU@h0de{CY$ROGwRyc9-m>IoX(wP4{XY
z_ys0~aydRs0LHTWvJV9#f1T|6lr4>3&dKWw`#0Tpq4Bc{=Vs*m;($CBulyF*;dh>q
zb~_J|wn4OFR5#%uj!AezcNUv(5=uOML|l?-G#fkLKPW49k=A}3YHXx0BXv@EK%zOU
zm9pWOR#Ny+YY(X|ci$wi!!UeMb}W&0XnOYkc8?F*S`udaTHLMr?X0tuC)hrSh^|;r
z#=N*N+Tz|Lqh6>zPbrjB#9=Eg5KlVbUg|2R)R|bf?e+|;oUil`L=?<88TbggJ|w!4
z)neney|$&mu^c|$!1;r(xA?|n0Kr6T<>3gqOYwo*KN6QA{<W|NQ(jg3V^N4hAx<Y_
z2yuAl5G4ae`M;ffK-9HZ`8^4baAzr(K6k+lw)|F)mK6nMc2D}y+|o>aO~X4876uxF
zMYGOpvG|spkt-fup?Oit11vInk4M#%2WZLLsZqM=#{D-1Nqf{g+aQYvfa|ipqRc%y
zlp>=U%UV&<mV}{=3D@o)8SR%GN3?#*pb`TKk*MNAeX3CVKBxP((Lm_WM*d^w5<0d8
z)l(;#VoMQyph<Cl5_pWSePe(?W@CCRL{VJddv-E*jVL9~A0@bl5Vj=rHEYXx2h?%T
z-PFfd(zH*)Y}AJ3nHvoZfaz%bw#ocCwIfYil)>La?vSs@-{GYPv(mi6zZJ!ythp1%
zcvwru<JLRPW_lPml!W}v8c*BewVp|KGTdJi>?x6-1s6s>4p$r(;YX7KG*+w9M6wF5
zVU{r1#4c45N_k9!*M{2<du}zKhTKMDHC8X_$9_l}(64GV3lXt^09q(;Q6wwQkjm)K
z58Hnvr>H}?S<GfgvQ|$*bMro@A_*mUHU}~a@0gA>*I1a;tjAMcU+?!<k>Ma_%ZsLM
zO=20Cr8$ZnBw!~}NjPlm?~fUiIFqlfc|60Q)!>+@Mw6!9Sgo{2X{sgvMXSWtk~Bz6
zGE_hEU^FdAp6{e`1bjYbs-kG+uKzY4Qkoa$+x>q4vOrD0GYqC%oYME?f~EIclzyby
zC-b_n5<D$wz%te@pdH^Cs6@Ojm{_Dn&{MpY_4HNbQCnu{;7%w}2#+e`<5yY2`*SpS
zrrtLh+r69zzYcG~raN2Vinm00*8z84g)n`SGa(&k%1kHbZ~JY{sbFY-rdyk*m0OnU
zO*BeaST8X~p9X(Mn%#C##FPv2@CNr_HBXxuv~8zA5ssmz$*fyFCp5^k!-G60G>+bJ
z`<&aUCv}F?;7mcuvRx5-;w$flKY`<;me216AVFTVp%XRPg5V=A-9%A&mtaZ|?4-z)
z2P>xd48i+-Mj^~1!-WWc0<_6#M5)Qi(fd1%DCu;rxk1<6m77U4u8ceg?c6a=Bch@O
znT5RaEOlDvp{n8zPnAv?3q4uS$T~0G4+ho?mEiHcse6MB`b;d+Kf`G}g%?Rk<yr4F
zWvd^iOFBU4+Pg>=RpqcTlFWhCTakH@2-XQ?|M@ax<^2A9J=wQ^`wFoC9x|kjg{tiw
zUpYwvf}+{GqOv#LvR-<}*n=u6NcNj+cj4rlA-5Wx#zxL<Uh#D#g!6fUgK8$okpWCA
zj1!}FC51Y#FgAvs+jq5smbGLo4yt?zCm1L(%1h?T8m~f&S!TXykmd6=%h`KvSZ<^A
zQDb`TC*B&--HgJ23cAKH$JOpyWN*4<y_!5ssgKbj!jLDAa;`;$t|NhDd@8P2rO2zU
z>Zmk%RW!MdwMgD7eVcJv!COU_;uz^ujtMhV1+elkc=kTlBn>r;MNPQsBA1Th=8JLR
zqE`x{f+&jN;uo4mm^Z_*0<%<%7IzTL(lUfnM;8gaHh86fg>l6^z$FrlI>AWLqOG)>
z?}J82H*b9fQj9LaF_5BRsLao}{+2EyG~cl@v2Bk#R@)(5V+}BF;hym9W)f`?sX*$v
zPPR3jd@ZFipBOBiwiqQGkWT@J3<pKKJ=L^Sm5TY4l1mXg=2FmxT-9Xx_vBI}0ZQmi
zC|*pagv`r-WbpmD6ab%ip4Hz-5o+PKrBX;u&{D<zsT5HDT*-T6^=1-{D<fV4KgbmX
z>4;Lf%o8*!uWoDFeA{e@oT7q9H<qcK)=;m}Gs!uNY-fXRH^JW<lCdj#dHc>+5Jkm6
z?(XdQj3f!M7v{ma4)!YBHfX<~x_yl+dl!;{wFm8gT^J@Sb<<aN`j)+KUgO%{fq}cL
zx_dK;Mztp3C*4q@_!_F4+3Ctz4(^1qbzM<Li5Ky@kQyXD^bD7Q@_SFCU%3Wra*LFu
zZ&7>XnS$PM^=e{8v9vU}p=m}}UGvvV_1TyvrC4E7m917%W;#EXY|?BtbIIwA>*!;Y
z(T+o0!Ly{Hm+>&CYuY;Jl!M&O{-G$|j>$*i=^Ph#;Z~Tvd@D>2Zc{3u=AOFD4VNHF
zV-wTUOgV7VaqHEX6~+&@v1Cf4=MveF#Ypq_XCtcGV3ySMNvi2T8IQw$mr-&869PFo
zmoe%m6qmt3ASaitx;qGe_gCl#hUQY-fI$P(J<|q!i!(PPrzG|zKE#*&eu`v~5=&CA
za*%_%r=`-Xg|(6+hi`)%KJE_hyALM=gBQUnuZQ=qgTe<AKb6Fr&JJEF;fxyIzYkxY
zrPL0u170AdF?f0WJK(q7KN%H}Lg6o^_D>#EqCeqv#OJEe<@XwYZAP$oX?*gMpP2Ye
zR1z<Tl+G7iyr+<}a{J%jKkX!`AVqZP&-BHAy}j@L-f@MTGeY_>{QiBH9)MB81U)Dx
zqH)8IKZmcoUpVU2TtT6^q;uAG=28-E2#6kqNso%TY?KmEa6mIyR3Gq?sDwrBEGlkM
zaY!_1B8#R3U1;ooghsvXpPWp_OuzC|M`#Jv7jp<eLE6WUZv(sY`D3$$)=GejTvjca
zVQ)c<BBD(hq6!^iqc$M4(VSWyNs}ehEBscI@8zXcR)F<<X|)mP#dT>b!l-A+>;nG1
z0BNj<*;&5)CWk-p*(anmc<Y}``Mq(G!(YR%$#pK^26Uc(gsX**`W&zvG_sBaJ3?3!
z1v<^`5LZfu-VuyB4~yO&Eyq0@9!h9y5Znr-TQ*V06wi3Ylr!)$<N~9B-9k6j@0Rg`
z?kb1?@X)H<3UQ?$BCG>@k0Wx2&&y=P5QL51nWdy}gTi=jk!j*y`De;fHTO_f%lBUC
z8<!AXFl9r3qRzQ3WvthdwZKu1J5pW@i87`3jA;nd<tNDvE|n&WncVu=qzohhNd<3v
zW1kBj?SS}<auQ`KS91T9Ys@twYiEfvC+PZAPD}xl`-x7qw{l9~Yc=IXLL`JFpgbY9
z&*I8ynlE@+)cBJ{g{wrZl;9IZ<E9A7x911q(@6z?(B4>Fl_m$1UX>~}^974-iiTEq
zW%W?RiN5-n9xHLLb!k~3d$}fAY_(|hc3&39Nc~Jt4kYPS9J0k?b~Cao5LKqO`{II%
zIQoV=rKVNU&toS5<o9-b!w^?DT6bK(M&+raAl+Id6!%i*(1i9=g`u(vi)N@$;Kqc}
zxmS6AuP#sv(yUm;aKuM-cMgQ2tK!S9fSDLui|PtsCx(6@*+Rj_zLV)-io+9aa%<Cv
zz-qV^nX(%hoqO2&E|6p7u#uxBi2DoadhEE5oa2SW8)kYO%eoQw{+F$sBsofUS4~_!
zOXN)+W}j<P`zSqYfQvM;;>0Ffn3RA+$15&>tkvS$YmdF&m<BF}B(B)&QrmfXjxb=c
z{LY$9<*_O|_3$8Jr4<sZd)1^{sheAVg}tKEj-#@czfpt4na^@+!@b4bE{DU$LVAvj
zQL`wAb3!>ivxS>v2tOw2l+5Iz_Y3rgfZYQ`5Tjd|xBu&82H4VrV&bl>lD4U*X0%Cv
z+%`zb#KEQyNsPNY$oRu4$|Wzv+AFXd=ireR-*y76xw6JC5(od8gpW%XqQDTDYi`xe
zBkY>+SmlM0+yV;jo{^1o!s2ks0ACl^)}$<;xZf^&8Tn={GOZgRf2M-d%JL~YNY1x1
zH#Cd!5{ytmT@x@BxYc<Q^1lt3Ufv6T5z`ZI9Wo64&Y0mcZWc6i<$p%h^jx+G8-{ox
zZhGpU5IE?{i-F^{&{{Xf2Wr+}Y((L7d)c&JIBPiiY{$+Ud{xm$M(pl<o|jgJn6x%4
z6;RscppuA<eGo@lNi5g^=(OoT_Bn2ApB{Y+`}By(K6)kLU9t~DTCtDC_EC<18Q)Jj
zkW*`WZ?lcb+e8fcA={KDjv-3g$iFlgPyv|j`K()z_KP{c6$|3{hrM}Aa9em6S+Gam
z!h$_w$vMldFpd?V-0Y1N8|Z+;h?`}hW81sA+Xg-I+5l1Y>#3<ZzinWkIP8(~#xbm(
zFeo2HOqHJTzhmu#7hqtKe;hY|tk+_m@juR2jK)1s$VdCxS;+nMz9C})1sCq@N18`5
zlnvxF4sE06$eT7K3L-YpGb=pMkhDl(h|3L0t4uxhh7HM5{F`WT3F25~@7ORXtHY31
zHw-|8IAeRoo>@|ObrjDEmyT<q^Y%job}m-yD|M|MI-{OU5wq0@AjV^V&MkIP$f<ka
zwM`WA*PgiSu1>e>Cx-#a{y=Mx&hFoCH>m^xR&fboC0hpJrQ;IH+qq&A)yi_NkDO~;
z;Wb>T1|5R;#OcJql@o(IO$koK$gQL7xe1qBaw>+T9pXERVRL<Ej`ccNW=>|2<EHh^
zcq43Vb4NW50e@Y6t0+!?3unwUnqhkObD~s5Co6T6{GP2?>Wvu3waiL=AEUlDwSw(y
z2`tu<l42LiNi|M7T=TGH2BZCGhP;--YBB1m&+Y?L{z^Mnmlrb|pw`-ZVzh=nQwPq@
zGmIJ}57q00e@ODWy}~88XSY{YBp)mD22+<wb<56N<bkQnVC3Z`lbaqf&-61Zw;p4`
z&MPxo>fzJBDscRB|JTI7-`;os0v+@AVwV9I4HN+|mw}Hp69G1t@$LjCm$v~6oB<EF
zR00dg1AiaR_i+!=(tm^8`A${l-KbZz$<D_R|8H%2VL6|pvVvF9G<~dMy<4UgtVDth
z@*u1s3@fTw^+v3B%d~<OQAiF^kHN~@19GZjH5;+sEz=5CUc5l|L0Exza;Pg-w-M{z
zGOb{hQULFx9)neqP^KLK{YI>J%e0I&oTK?v`+pd$;E}g+wqiYP!+O6=D_CtxJpVIR
z9WeqQVzqKzR-=u0@0M!?uSu!lF?cPX1$bN0W$}%8@0M!?uTJUVF?h8XhB~+#!_@s^
zt-w_&Lp%nqa>7{b@NNuM_sg|{mr|B^3|_LrLT}&WM!a{+wT2gjBJ(rHgLjkIb|4C5
zaeq1S-Z~LYPhQb9+6_)<Z^QbUUvT+Qiqfq<k&gwZT)$Mc`dEr*RIA654;(H?AyfcE
zeENuo*;#9>W~^La8iBUPJCfsqLMlDb>Xli#kBldrV>`RF;&`k4g|l;{bGcr;0X}u}
zHm?_|-e=IQV^paG`+O&szG7b)t4Fs8seiXtmVLQXIFD~JM)yN7ttoGVde|$pvCe$}
zWpddjUP#PK%RW~4_VCoMFeioWP`<`;T~{u*Qc>H_u5B{ls;o+Fd~K5WEM7-sCoxJ8
zs<D&A3u%=MirE>ulCEvHE&815S#tR<A?z0{H}%Yt8tEyqiOAfz@Sjwl{^iRmC>Pf>
zGySzohE9KO!^I8aty}f~09&Sd)t6Cn0uur`IhPQk9&5L*2n+T?mvPnw6qjT31t*u^
zOADNT@b2lKe*OCKg(neyxGFn1hC*;X$G<knh|A1><StROr&s#OJtHI*5}U*eS2Qyc
z+enWk`l-XOHvXzi9O++F#5U4H{P3gc(&KaGY6||TBIXhMIq0(v0&OWmE5U+{Q4kM2
zzNfi#2Gq~)ISqH>PN_J;&x=<-%J@;|#7TL756_7gNOzS+m?W7gGX##)ynggKs)Nx3
zX$Q`v(pJ-5J1&O5NlEihe){&3B@_tEg6<{srW0l{^E`Pq2IQ}eK@_USF>YN?l}z()
zel0QwV#066{G5=vCJVugpGg`Ddb0&lo!$#Fo`iQBZ%J{4f2Y$GCVR0kXD`x{+)hq^
z<!;bgvNq(383XOxAm2sti8-th3^T$mU_G&9i44^v3^_QQATNH4kOy#aG|osbLRct6
zqRf-?M%eRHFQ_X}+f!89vg>BGHt)>`4C2$h?Ti|cj2tZ@ymhX;fH&f~cBJCV_J%8d
z)!3jd8Vo_!GUbexv$2Wp+{mKR3@avo7%pj+97!-s7nJ2~{WE<&-xL#-wM-lHkeFCW
z-SS=iuIEuL8pvL99V0;%Co8Rva-TrI^|BN0A1O*X?On~^zIER|zh3^DxR#QZffmDW
z-(UO#l)$7F6cY$T9RB`e`19pIJ=NsPoN?}hpZ`OjLL_W{`6<;4j+08b)=D*hDghqm
zP+I%InN_lsE=u+eL2OD*5*2#XQ4=4LVyF#df@quVO`erv(%C#qT?k}@QDz_CNa;eV
zo}x%b>1!PSS?b5}jucF#dnE5IML848<YowqHC%9pbM1B~sJth~FwLRLa+-@YZ72(I
z&J*h5ZzV1%%4R^xJF7t*s@<-CmHFDlr>dh^06Ejkse$i`U}!Zn%T*nM*i1uy6ukBC
zlvMb6l9=Ncl(94UB0dpzZ5~e<Tq>c<T>O2FqSRHcv188R6njzXT-sl9c~4I+`KhYo
z<+<SkQ!#If#3=>GZ(fdIV=6`1Y=$OYkv=9%1E^&I;tpiG5<j7(;by^qA%gX&nY?-_
z_1Z-9)R-7m8Y^$>!IsSE#Nqx?RlS!a10;@x%=pS`eyDOu!Z~Ihp(@UTRga}pMQ>U$
zqzV;HQ=Lq^2ux$JlBMY$mg|)84wP>;YHi69)2JZr-Ss?jV1fc(rx70oYORR59x3R4
zz9v>O)I=RsA0yP|ex5adMK(9#S;&N0)(s?$T)OBKynL2#B?t{lEFI<2H{Do~NHk0d
z>?;xg#gsPdB}phZ7Lrv;e@Gn%mcmo<+EB;VkfBeg`A-12`K1CjR~{{4!Dq1oN|n@N
zwqxali5OM*rHunw0_uLzj8><UhWy)NES`$yQk%oy=fo|uki{5(k`3elrL<^%sYTA5
zNU}Q+jV-yH@2Z+vt@eU|ju9?BwMBlfIsbdR1-xBhPBBO5_uRuoYqwAp+=jf|Q3W}_
z)&Rm=ZZGm3ssP6n-5;cimfDIcZU)~<2FCR>bTHkfPGH*F3B~K86AJE^6QF-1n3K9$
zA_0a8P&?+oiK=ja%$ccF^Ql8fq=Y0Udb`7(xUZ5`V^7-!@T^TyTJxPvM{(P|B_GR^
z_-iQhKM!aqI$O)Y5|mk;?dXHCrl_$K{&PW%mZ3GXbE`w{CaA4ce4<WRa17-rgS$t`
zZUiNAu`%~e6fK#ngb778FO5}<t|c@n8s6g|(iv)gb?qR3x$2DziK%l?-2$Uoa1x@X
zWKZ>xfj5Q8ZiWRf$T_kbebf?CZtUmgs~$Kvw>s>B{oHqkXB5mJOsKkfiBL1=Ch{-T
zcal2%YK?MI->vn1xr5sbu2p?`Ae442`T#9pUUZ?D5fh;%i#0c}kb-2-!@~SoQQiDI
zL0nvi+t06m<eGBwUVf!F0uYDzy)2uQAuWJiGO4ndZ>nlhjf>!Z?7n<`RHw<TS~izd
zn&U$|dVcw^T|10c|0t$Ye%Z;jrP6Go3308qdN0+&Nxg9+$0a`hya=(NYmFg^XI*bt
zPRN7nt>I(_s^}V9%G(&lUaDoM+>+)Uv@n0_O4Ox)pZk<|PMXeiuGI&+Lv^2BOQ3oI
zUB8mxLxQCyJgp>T0%IDmVKz&jXI3W!FiNGK<}bhU;dgrYqhEEQ0{5RV^n00y5C0o}
z_DdBTge~KqFwmXLHNt!#FCoSw*^z{jj&YLnW)eF&^LurO>tx9^oa*VWF30#qBN^hI
zlw${f{88%s+OFKV>l)n_KBmk5zBkC5bn9V2@PA>KBCdHKa`175V@KHc!c5bY_v_?6
zqi!1?DOPh>z3XNpoMznWhtbXaKALXfm%=#+^Bgl5mWmd@1v59@e!AYLx#JvfAYqV!
zwvR@FdP_~w*BFLMO?Liu?T%%KuRiG#yQf!w6o5?xC{4i@T!+QlfJ;jzxO);@T2l&Z
z4=g0dSJnE+bi=2Fka+{k*o}rLs`Gi57$1GC6W5*S9!6mUU4KGIJqcXDIwEb|loVUk
z?H*#XM>p=>`kkl<U46DA{W6=ps#6+UU>+5E152c6v|ha20VX0-rqrF@Jf}H);@JIv
z`KWSKFGDF@8v@W&em@8IObr@pOr5Icz0u+@RvS8yvEX@lT6{3w!Y`>;8u_HOC<N0Q
zrfE^=ne9kZALnf1+oJ9>y<h>WsfO!#M4&VsTlR4nmkqpmtQ4N(qlCK|kGIJB5C9(!
zxUFXnsi93B-bKq%Z5`8xKRFPNa@*K{0*2Hg#-rWR?1cF+9P2QHi)F}jVCI~eYA-l}
zw+~*l;UGhmnIUy1BElPBBxt3rvbBE(`yf3o`Ua#nmMTwy6fE5>*6U$ghp`gOZtMx;
zVWQJ38LDj|{!C|gpgPA>U16c*bNtS)w~gQ1iyen>Lp3HAI?qv!vCIl}yJ}E>O;h`0
zRKwHput%%sFpL}O(Xn)Tj(W6Txzc^Z6)G-VR82k(Bxf2fk(&;;6Sl+RZNR1^)8a{}
zDa{<W8zFp`sf&du)OTRFS?x~fk3zJ8JjS~1(M9Z|=@x!vtb3kP#DYZT+|1HWy+Zi8
zyrfZe^g9=Rj*GJalZbW8b396aVA#CRacHqOY!*9hQ^D)e&$vH|YI$2Kh4a9W?w8~$
z{`7HrF}RovXMpzseHf4ppn0rjo&z-abEKjMditM=;WTG9$)*39yVU0yW&#%OMDeh+
z>nOTdcRU9~E*JsX{3Af#waPoB&5p!j5O+rM!@{m(Xk&5k6bv0O7!RW}`8K9L6TZB{
zt1$fATbCJk8rxxkHgG~_jlE}g?p6x>_9Ka^UZ=XW9XZ+e?C$>uBOi2^5i$r9m*Fu#
z6#_Fbm$92dCx5kE$!;XM5xwgx>IcMOa3|0ZppsMu=H_wV26M7KeQ_T=7yrJ4YjVkz
znRQ52RVFhSe6cb}ir#+Q=<RPW_wO&iJh-@NVkBsG`~K$*@BziYK=G!MHw{1(huim`
zx8J`}$`oFAyac73<E8Pt!B6@2%E4d@z+V7wuUdd>|9`;i1D``^=ike1Jq3vu!zb_l
z3Af*3D~gxGmP!Y@cn`2=uy22S|Mi8&Ay8tK`|3X0Utix}{`=AnNO4LSWw(F+d~pv1
z2f_`z2}=y{+h6~_{rU3umTK;c;wwbDKltshAGhYtZxK>U5mKs@log~v5od8z$`bA#
z1xSUZR6imBQdR@|<zZl!3d3=rtXe7`;q4tPF7n3}LIFKx<P93og(H{#_+F|zzpe0(
zNQ3B50CBj7Oi^;et?4L!R+p<+3psy}nxgwnA5oH;<4_^~@d#9RN(7W5pt@chgNVYb
zqRteh|Hhno(e#bVS&VQ@1g!4pJ<@xQ_Yx7IBhUM##+W5mo3;;td5*XsjQ1nF$9|Pf
ziq2%{x4N^U^^6-LOgC1RC9}&anFQC*GGI<koWMV$s;H=5t{WTgvG`p?^)m(Gey?h8
zmuOfEDFN7*l2{8*0Z*6WSPLe9KuXi+Y%(J?BRKdrnL%f%bjoECfvG#=do%cbCOx8@
zXixC>Kj`gW`0Q6~>%!~*e9ND?6MFm4?H?{z+MtV~@pKV6%hm|f8za`(NOmG2xed}i
zN@5>p(pb!J`8rt?sFf{UpX)7TIpE>}zahWGGMKWQt%&Tb2Y6}Q?w6o{z4x2*AL(sW
zz?KTFpNME11vHh0CD6yw*+5+z)KT#y)D3!F%!d-{au@0%v}`c@IEcQQMPCgdKa9{8
z@{%Yc&sgGN68n6QBP)y=aXzh+^GYJk&7jB7iqFk#-C)3E=geZz2f8O)!+EJNF$qhl
zf^RAOL@qrQF|?`BO1#H^q=HVt0K;8uol=k7`y;jr3bQp+$JdCqS?S%(T*N7b4D;x3
zA$Qa<9DD9)=r{D`TB&dCJAULYx%hhHGKaC;*7^}Mh*>Uj8=tuO{&GK#vNJAk7S25G
z>8_s$EmGxXN^?Bw4%1B0_Z32Hyu<YyC>z;!jXfwPu;eV1>s($U_=uOz*(*M^Q8XTc
z_lkWP%N_ZS7oAPN900C290+)*-rtv}S_>f>BkFKUY~zGA{TAZwReHT}l^CywN!w}s
z{EV04S_>%xd{37RTMI#dy%%X@DM2-x>PnnsE6^=f=UwE-X`Ykz39E=P+4C?SF?%oV
zG4fgZHSZthCp?N{+SC$GbCIJZKaOd*7NX6PLAZJ1M$K<E+Xz{{(1J*&%Mxj-W*w<w
zF^-}#9kuy63~)w(skSx$I^J)r`ZbimjMw9YAQ<^ug69i-Ek!+lPBBi)TZ-=s+8D}f
zZR-KBwVRpsVB7JL7*0U<Xv=Ym7BIOp7g9{+#Pv~6tpyPhE<l?BbH;q(Kw{OSrFcI<
zF;qdXQ|(mGT&I{@3x=-o60k!Bv?Zj*2O=%e(q8D8m@)*f4XZyy5u>)&dc)V{);KMt
zyaPK<2)nR?F%ScPZxvSLcvxwuZG@G&^_j3zQ#zp!DXcUE?fJ0MP`n_lG!*4}SZP|H
z2rCUGaXzeEC%zO`8jAPxVP%!aF07nJJB5{okU1Y#R@G)GtXxMig_V6^r?7G!>=;%q
zv(X$@u7f6pmCI};hm}<o>%vOidMT_(MijMqf(ggH4Oqf|6>ljmg_O9?*7mE4!a`i+
z3?Z&qH9X>Zqc5#{SqQ|S@qJKj>vn}OAtums$y~$lSN+F>V@-Jp!5H}J#(Zw`!F_oU
z+|A!v1;3fLIOGcDqpVDBZgZe+oyo6xLS7`R?o?_MIYIT&oLAKXn<}TN9#vKf7C;Hx
zNE&L3{8kHpkt6v`3SV`9XF{x7JQuv9__?|zqiOAzr60o~?6Se6)#IImNdd1PjG(2k
zSW!znqjJ=sKNXeAZI}4xcMrv!v8K4KcAOR?d6X+gY3<wBqwmA@qa1Q;&Rlt&E=WO2
zB58+?yie8dxYjkt;qj>mr;KpQ4{*pqy0Y66hhA5I#<hZ5j7%U)t;jpQAdXr`rzFZ9
zA*=h!8mc=q<Ez-ccO?4_`D{B1dcuaiQ)mcY;1pVl>rSC*-8+R=9pycNIOYpuvbB@T
zm~1tzD^)zQafK6YdsU>T1erFttV?n-GBgKZo#LJ_ETiyKU2Rx3)t!d*Ic`|ZD9#&J
zP4EhrNnQ(ge|{{h6q=*uW=W;es($v!ag|n@Y3+PpdOIr9s8%(bOCn94uWrra-dDdF
zgf$x-?vqEG^4haZ8%glm%eAqjmxSLR!|S<sijizryv(KyD;@=BELKl-mknj|u6F0y
zhSY=LgjSUW|Kg+2p<>5Ksy3uR;~9u^fBDgN9(KC7e?^Dk{j!+6$!@nS!%xe{w#zz`
z<&0gwUzTGBz{QzedWT8u<BVBFK+EUs++@~bt}>N#@?K3|s+CQSFIh57?XpeB1aOCE
zaQVR8mMOmD(8j#y3H!W?E@a#QI&SlUBuv}JV$QZBIQV?03ugOwqK->^LzA2n1$7zS
z%<si?e>1;Yppxn*X};DNBjkeS>mpG5^AbBHH_Y({5{jpD2Uj9tiB{;36(&t4XUcs3
zj7Ze~{%qdct6(51r;rwjH{lu{)&^XX5iXvDmXt&lJHU#qowBGo_M11WSzoVT3A#xS
zS#jj|BX~^b{n%qCxT7F!U@O3{r91~*A&4sWe?ibf6;-_@II5#94o6PM28!LRMM<>_
zXVWBk6U^aJZ=lIB(`D>QXmXs67t3vOF06NkUG2J^<|F|50EX=jqfT}|hxbfY>a$K2
zCkS$<42H4VkO71o=<=|5G2P5Be!6vp=Y&Q4EKW5{VUeks-Nl6)n>1-%_W9`LeepPA
ze=3`(c&PDXr+y6MvW_>{kSRRJMk#hXEy?k6LqARF^j7HL3hLP5Of>)d2NmLnTif&h
z9wHun_OXRc|1N^-eK3#1u>rFQX~<Jx24jgSRwN^M``{JEpFAi+GsMXX3-EdvAK|P_
z4%hA(?1Xf9=o^sgkgGfgQf-Kx)-U*ue=}SW=Kk>XD=Mm~A$Di`>~xB+y?pKDYx+$e
z@c<8m^F*ud#s0b2ID{LbQ6bZLj%bv|7u9jmi2nafTqK(J{q1*)W*EW^(a4ZCJx4TB
z5nI%n<y3B~rVO=M(#ak^H`(HQq1_4V@bEWab*lpsE>E+rrknZYArpE^ngv`Ee>zo(
zi|RvJ*ZNwXbr-YFowbSa@oeD4LI(33b76wObjEgZ_EMGOOqFA*$?0Jx`3i>C!Q7uj
zg%~~V@o=mTM?n3=C!ORMJ8>SyV*_UzQkdu94BYSMEPzgMR+GVd*2wl)>&#)*xvECm
zQ@e@c@Mzajv>|DE4v3TyMtOC@Og*i%q?Nv^j6Mt^-LH<r1763`gbd>;I9l|PBIoKZ
z*)9(vFV`WZ0Na@MK8%J3vw@KyR_QZKN=GT|TSP;uTA3^U?L+(dfB$$oe3t<i4HN-0
zm+@W=6PKuB3(9}P>&x%oetnX03Q`2ixBOH8_wnuNKTq|9EJ~rI>i+og#4j|0K(9I-
zga_>Y`d9bo)8A{Vc`Qj?gyBDQ_t*EXdh#`jG)NMuwAPqHq##Moy0bP2<X3}8b&%4K
z0E{Un_RHPFEEPrbAgpRCx1skA78m(LOQL|5GVunL&^3RPOMm)NiYH%J_$Nt&ZlnPE
z@CccrL4m<H6yFPqV{0d@!Me2}HPob_h7{b;uO`Jcq}-a2(KaH)?_qc@NfoI}|6IGF
zFZYoH4h>qy6OT5JhH=DONs<4T_lTrJnkvLUy#$IUQwdlrz&4{WgCvD7R<bFY{!X!Z
z$#f^ptc-tXP6C2!=-pF4=6xy=?IO?nQBx|2;Cwv>z&uB&g~E1*Z*g23lA<d!j9WZe
z(RzhmXy43LWn^|)r82<dEDMa)gaLn#s$!#h@h~AiQ}JF$_4C8?R@Gh@Pi_JqQ$$+*
zJ&X!^M-zZX8gdpD3`T^yZLh=WI<CtaoB%hY%XxpM_YLjUj3|4Uio$|rY9}<hHOuT}
z2@`EQVwoU<boD5Vr+M3LkKnntZm3*pk0f^mrXFL-9u){pCRTHFfjufCoNDHJ)E<*n
zFG{R2wBr)9UeL8yVonL8R4Xyogk;HZp(O*YZweYYE?6>K1J;cTQ@$X;NC6N98k$^x
z8l`_y+WE*XGyI(>jlZ)oo>5*H<LVgC(Nk$tjF%{k<N1C#PxmunqeOBShDrYRyX^i+
zFMp-R9C!S;Z~7w-BfEch|KK2UgRmzFF=4U|@74&*PKn?{Bs-EY7zBln$4Ttt%=kc-
z=-0^-q`}m5`yhBB`~zMnu`>aMgh}|f6{>&W2ek-O69x`%*gea3X_#}aJq66Yb-o*)
z;dDuzKA-DVwAHJ<w&_kr=w~$8+6Ygf=Lz`p2yGxwmg2}^e-83q336x|d9w@oTY)@l
z%j;?6x6wSJCI27fPb0K}yd`%Bp9H)EAwzciV0Qtp2NkKnw+F#L2hz1J;yUopg0z2u
zy|J7J`XulzhsGg;?{<S98u0Z6mR|_|CxE}HfV;szjnW4C+6W)iC!x<tpV9LDZtz>T
zkDuuFFLmWdlxjElr$O2PAGHXGp9DVb#c02n(GR=OH;vza{Y0~W6#3JE#BStIBeaD)
zh@kZ&+mFK}_U*Wstkb7<T=mwjS5klDJKFyRXL0VFLkA=BZbZcd5Xe25yO%Gh`GkTu
z?D{Pw^JA;Dk$1Tj4vB~U?c4NeO5*gR-DBTDJ+@A1v~vQaPW+=AIO%!hGnD=;hVzE@
zc}O?(9LUJcJ<S51>pne~yxxAipw4@Bek;?_g?UjBGn~Y2?8j$4WSA&ZA8LQaD$mjz
zMwpJtZXSRa27aO@Qqr%fUEcAr`?zKPSY9E$=TUnE%nP{Zp?A-hiz)Uyx8yf`x&)B=
zaT2Ddjq&nB&VyT!V#M~vvA`{ALP`}TtmFS$C)#|r)uWqqM#kUwi5}b{pOD66oY1!E
z$Uz>%vzT0Mbio<{IuyE`IJ<w3l=^X$o@$Fv=W^-5BW9cFDa*{fMviT6r(+vEnWOJB
zIjO~*h78!~s7XURKOnZKn?s+;QIsG!(6d7@*4sY8DAkaUntd}(jY$fjYB{kjPq9+3
zlg!i<Sv#LVGhSFS(uykEvbjCZv6HmjYoNs_umlW5IBt~_8H{l@SFV3Z6woM5%3&To
zTm=<`le$j4p+e!jT|Zr2jU<M|q1KACl*!A<!%0!#q$n)_C7~)OdWc{=MJNgO0WQv`
z^3zmEG~N~G>gfpd(C{cM1=cmS`6Mr(5}<6|8b4+aMzkV;Ks_Suc%WQS0(V~AT2uVn
z5cCTty~DY1JCPJ6PVawaMIZwOR5;u7urMO)B}p(3AA8pnA2>*lhgBEyNPbI$Z79CZ
z0~W<jvl1bg9e|yYS4F*2o-^{QDJ~r1MOn17n^QgX6rZELQuQj58&-2E18M3N2X<Wz
zD5#arZK%N|s*(6BN30xG!yi{&j9HBX{4IE$xHO=lwb+ay#6o{70#y~uhdI4`l(g0t
zkD~3u(t(IVX}DnNs3=t$bH)fNf_mKunu+XPI#4O0yt>qWOOh8<Ow|<K<F2rx`1sPX
zPO%X?P{gV5vzCsM`UOjEP4RVo>1bv*t9p_)5>{WTdL_v%tGSYaH1&#ubu~~v5FPUq
zmyT-0E&d#rD7k;jmkwo}(B5w>9m-f?)VTA7)BC83($dnAW?hYICz+5~=Qb>4lftnh
zCN&g6R--KGYQn8-T^ETZiD@IoBkQ*A^Yr>w5?+$uV1#8`oGU(kTb==GTI!*Yy&p?l
z6BapTu&>Nj3hRcNH^|jM#gnh?&R22JVN{XCh{afA_H2Kez*vyUP)1x)Tjcki4;<cd
z*m4N40K$@GRpULf1aR@B#)jhiYDsR@x-Uz-pD6ZIoP4=%GO2oS=VW5_;LKzQ`3YER
zlKOl^<!HJ#V4-r`Y5lb&uA;cDb|@z^xlhljQk@_7&o1zIDoAN1oQ9e+POHlW6lkAe
zWFsB(mTP}^(+S%ghsSptD3w5o2RKA9oD5vzu-BCbv_gV#sHa+yEBzrI+eD|Nrxn4g
z-xX9;cQ~X<x+fh}w!m`yh2<|>V8!kIbRB;^?)z#Tf7QA`@`%n`5!jDaNXmzDy2=FM
zOxP<)tA&IX_w#xQK?>*0#d=v${posHjlZ!n)ZBlLz6c)HmRi2oXiGi6b#1BV_kgxI
z{&LM;tt}<h9qZf2<QudD(vw<f1I;WRz+(t)DuNFrBC-Nk6gL#OY<*&nl|F>1$9+IF
z(N)(J^{T6`T8{ymT67ZChinHp^hVINSJt%WZY|)XcSzC6y12*4gscx5;auc1k#~pa
zcZz=jwG<|_Hk(1xxARizbKIe|&xdRjofqD>vYtOOMUUhFT{FM=jV@TCZT#X!cdZ;F
z6;^C|yU9lTjBhGg<oH)DeP$JXK-r`JHk!YgKCw-RA4b(YEQ}vF74`T#g{Wc;64z4j
zJ=zmyY5Pxk_VJs|@&v!PWtE~u(TA|4QmKE6Coi+H{KELLB4*zP@hqlMtx7hRAWh3z
z*;>VK;d05;w6N)I^l0PG@KL6dtKY|UboCX}KH@sbUZucC?FxdjDZ`3K;kYpe_FE35
z_q01d4o?H15y2O^xxVLpqC=Akv6Qb%3r){K%$}KBJ^JviaGkb0{b?P=uc(>=A98;a
z<SNd6SqBmYI(Pk-b-*bzA08=fJxpRBXQa&PlZvf_5@tM;mbtnbr%DxaQD(Gb7~5rw
zj)B|`zrfMJT!AF+_~?D!@4V#N)1=FGNEDs7_<<yhg`jblyJKnSqn&0c+h4nV7VdQ=
zq8LiK>%Id?%w9}3f9C20!YApxqg@#Cg2uba^4ynV??db~w~gzoe*Xs=6I!*G0T&Gv
zm(uww69O_dmq9`vCzr-;3!Q(1t3Td;{ru^LrzrrPp*nwjzWnjhPXaA6iEjAM_ZR;F
zB`|5l08TNj_3-EahTmTPRY+vgoLT>i{{Ps}w2Tu?dXdh?)BNREKKz#+{zQ$Xv%_DZ
z>383pGpF3}|KZ<6pVz2cAk?|mN;Rm1V@3-i?wItCBI+I^B=i)aujqe_gsL)cubO*|
z(ccyEiGuh91)vA*-8T~cmx-`db84Vc(lJgD9iP?)TuL$eeG**C5+z;WG6^n%js7^^
zA~1*^oOJC0gA8E+){HFzl)7M7Q1pUY#QqkA)C;S)??m@73LEGmGKd0x0=j4zN>k9~
zl`!gv8D9LoB9Ja&?=gSo8M}b*HMbKPb&<r`jXWc~_n~=s)*EO_$-u!$XiCk{*o>wZ
zP&vV2qT0Tpj4c96Nx}g#mcq4008OR&b9mcc7pRMCY{PH9yPZ5ZjMWAYgkVC7r_n9r
zeKOrLUh<Q|$Y;w85ml0iBU8n^@nVidQoqRX$V{xv*eFQrtj~Y+f&i@g8m^-!=C_^V
zaTt~jxH%C4JV`~lWCTbtcru}KOYX!Qy!)2xkQ!Rw;We~`Ksx#lp;CzCfzG}~j3JZU
z>$MNVu@1AcF%CHeX5|FaC`)LWz}p8eTF|X)h6r7NQ70G)T4^hH?VrIuNRLl_15y)X
zm2)7q6s=n<lIwr*@RIu;bsBZ0`HiCqS2)@S|AYk9wh(`$5S*X}IX!$A{A0Z4L=0~G
z+auq}n!^xoutvu~=N#4;&Wuoxvj)}Fvp<J5<^<LphH!&5Dn?D`utxc_?z62BY~dnX
zF6pV9%eLfRXm`STc={W#qGAf%)1$0s(=FqbArLwx$Wl+zFpG8?WywoYKra^`%UO4o
zbn2aKJvyBYl!T98PRd`D(9Bv_lw~n@DQd&(oG<4X+2&Co`3jb|zC6@=y!u>{mBMr+
z_3^n%P@kPi%5j%1aSL02PP_NPbbLA+Fj>Ql+%UU>Ih)DO4DC@8TO-SLYQxle^ptDY
zydyj<UQ>YVd!kc+HSJg;GHb-7Ku#uzBD!24?`W2uIK9J^|3|<6{C()|{G8;=@p<_o
zG!zbqBxhq?40An8HRb+ADHSG6pr?|{6x%~w9G}#L1xzW2!bv@UHqOwK%l|?z&7GfD
z#!u7%Cn#i9d7y_vQOO`UJ;nEK`6RbCOj8ilwVp2q#Uy(!01Don*Ono^lgdRCZb{25
zlvqO5g_eq<zvj8ZZX`_=(!YQ}bEif?YX*AK3s>0~g(zjL&y-3vSF;$=oESJWp?5F!
z)oG>{$Go3wLOAk&SGMm1V17gtKl);3Kccim#<a_VPG&Up+uT`^dO;M)8ugS#;mj_r
z6v~IzD}Ze#!BNzv)rmP#dkY@OC$=9oM`ntdk8(wlU6u|eXz*FCNJoKN-!{l_J|354
z1jE40$a0zKeM5Tvo9Hb}M`1xT{aKf|NnRhutk}(#(gJ3G;2n_b8$UsziV<ijFDMix
z8w^4Bt)HM!V327uh2t@W0_)2wI2qrlQPVo!!zPGCO@P{om=M5FyD4HmHOZ<7&r?OA
zEu&+nSP7%aoteJ%sEqT%51P5L{y*=Q9(PCCA-B}S0-&rC4UgLUq8fvhWC+}Z$`b;e
zJbO;&Yb?uuKN^T7Mec~)ItBj1Fg4J`&3q60hL}}xI?-{QqZ)<y<yx!O0k5tolfk}6
zpEHUj&z}5m`$jc}s8$p?A5bBm$;ruRaDF6N;o8=t$&uHoB1IvPlrp+;H6n#^IaPc=
zY|A^kOD-j`d1e$xeMz#u9YyHVfNkL}6rHaG*vX84Zfj0bTH@chlJG3^lw4m&y)#4U
z{Q!=tDdsz?F3>VZl=~jcn5VV9td4ujMRvW+=-QX&;-WUp&E@@Ra}gXFD%jx3*xE9~
zryP8qYmi3zheegTuxw|JF1295&6@m%;2`ZtdB*rtjX_dA0%M7PVf^!Dj$IoKy=j#O
z`#Oq$(oB(-Ke*2@m1Za$GwVAjmTE%lWiaVCm00zaGT6f;c8<*NzYr%a^GHsC6O6_i
zx2geoX?65?eoY(sH1nE<am|88us~taBHvZj-D<y9b=U1Hl^#cIofE=u(RltQ*T@Os
zZRohk71S<HU|zs3cwJ85Mp>@{`#3j_qqf0+4GU!8-AUXaui?f$w4ir$0~>DG6S%P#
z_rv&X;O@6>sFPr)D3y<!YlHIdgIzbUn=ipWE@<gznWV#@ZSX>qJJj$Muk|pAeU93b
zRa!h_@g~TPvV&tN1VwZ+-(?8lgXxn0$FD9DSKZsI=T=^ww=tg$h0+gy9T>Rv;fJ6$
zGk49F`c#S!yBotd8D&GBsdh-*#uSxe7vC!Hlqv7U?cyo;>K9sV+~SmaNEM@!wOx|*
z(F3=%bqgW_1Eb+vm%(W_69Y3aG?%d)DifD*b1o>CDtHSmf5;NadB15heV@$q<#2vG
zJl~w3;+znoo!-7pKnI3?!O&*FDFV#F!Swb%{qbUq4fLGp@y0kuPo(ceZ#+IZNCW}2
z0ysWJ4+Z{4&xOvR!ZDu0G6kf^(#eTWTz(NKLr);c5G!5U1L6#q<DYN89*n3kzF*%C
z{~SmvR>MV7f5fF3F-{KLOxV-c_rvoq?|{?w@5673wqps4IAfr9#@U3VmdM9w>Fr57
zPqHIvhj_HSmx_i_ts&B1k5Aru*8Awo+xvn;PZT<|_P6@rrO`mLIa=Axg*rV^Z5S$V
z9|h|5=uE*%D?5#TwHW7MtZH+vbc+VC(xAwNL07)<e?|8*@~8hIf79PBaxnx7mm;rW
zyB1CFG<10mF*gm}C`Hi=itmU|Oij1=P^Jod5xszUf0#wt%)$Cm7IS)lW#^dOh=p;H
zlsU_)NXRx4FBF^b`znxa-!`n=S=DY@RcW6Dx6nA>^Upo@ReT#dp5kwKrPw!WxhOHk
z_#8EYf5TQEqiFDsZ!zSn4><Br-nWj!Wz=<vpd$zQI*6W-laf_Vih5d{O$OKuG#lqg
zPF6P+CZsOF3%IkDFpG7kDk5JWDRQe4RH#+@q<eRIurJL)&EY-nDg;86h_wz1)o_zw
zEWA-%a$8QhxEn5Gmm|0qKgdWWOMF-&WDu_9f847wD8ghl$fXy)nq{?GD1I{Lh1+f=
zY4VRTzv90c0DyB0OP)T=2d#{vz4Qvl#;uhB(?oqM$KbZQGNd%ys9o7L?a!^c5+$?P
zo4Kdjr=-e1IiFs-e1R?$^|abc<UG}HhA^GQTMhLsH`(SWP%U`weX?5to1P)hDWd9m
ze>!4n-6_W*)?8+pjmX<{zLk?aMqyC+2Qg9kU|LdCHD-MFS}7pD?d)1TZ`Pw-+e&dt
zu2sRGlw0d}*V}bE$*qppN~`c%3KS+TQLKfq?<N&5FVlcd0=-0`=`Io0ri`t80-8;-
zjwY?NvoYGhL6yfBT}m<X)~#b>H?n-5f6>{2$D@XfGz8~Su4cHJ%3bZO>1?lcmj_v=
zaMB&>ic5nUqf3)z_~{)2+W}4+Do0I*e!yB)chKEaPji(<9}JxeD@D4a(+>zMDV8B{
zHz^=}Iw|&sW1^DbMT@DKb0|}Ad3_mHd|kMUKS;Alp<ETpfp;*wWG8C#DdY6xf3N9V
zxjC$<wrdF8F=!9V4}JakJ`Fp+%UvTI3N#Y};wsR<Q1p4awBYQl*7uF-)6e$Lw*;Hi
zTp+<=q&5U*HPK_SAIfa~>`J@(e`H=$vr@OqEwOyl&2YoDN_)Snl;-LDR#WlDcQM^E
z+?$|<u$$-@9@|7$VYX|P!0i{we+FeJi*!!>1m4{QSS6BXCz7T<TbB&1u;7%IZTl!!
zZ*x(Kc|32G%Lf^l7>@JVN``Jtx0Mq}s2Dj-E6SH@3m)M<;DwKQZ8ez~XoAw2jmf3u
z-ckBW;IMG3ju@l>Yq5rLCJ6mp$E#6ffqCFm!PQwMbQ7_x@>2S^{Mj11e>|jLrJ-?_
z-DQF!BLql;5z6@JUMo{`s{<^*%N~ss!SG)?{Kd`Jw=Y&8>#FNPbq8C{$dzsrYvqEr
zYtDl_3-)<2NR{k^SlaS_pz5+*+2}s9+y24QDDDQN+H>Q^kOlo1vh4uyeujJ}$K~M|
z;>l|xQvV0d#=Ow=R6H$zC*-J)_T;oCQ=r(9<#kt13t>(UKPjgl@42Q**t+=rAASD<
z7JWOqmjM?I6aq6jmm!E2E0=H-4KtUJEg}Syj*<ohGcYwVxAlDsIDrB+F_$5T7Av=K
z6b+>Umoc6J6ahJxF<%cUmpuXupMM}FkZ=F_`t611B~X5iRSdam`PKSO1YQ)*uaJti
z;qSnM4XvthgN$&EoG6?7ex{k2H(|2y>sRva`t|a^kp-!l<nHv}zhA-wC4osNP6D5L
zefsB*)1NQ@qEy=raey1bjnTT>5a75=>SVNFTFJ6N9Tw;k(~<dW_?{V8IDfSM_~)+^
zY5BbopW?X9no<{hUD(PqPy;V=5Fj?3lftM=+T!pxZtHITyk;I}A~T|*t?`FEFz|NZ
z58JiO3UEmq9)Eu%esyTLM2dsd6)^2oEA==cP!jq3C9OENLwXigzrI6trgcTQcPvf#
z?eybn9%j~YjaA)!?x!>F_<y(3%^J*e>W{N908P~&4D-_5R)5q%i)|oJz<Ab(uI2ZK
zj)-=KNx#@Ed2g{dm7Gw!xuu_PKMW6>po5xU!(6*nU2w&ehG7-`QQSolM36Hf2x@|$
zWdig-@#A@l0+=#>pm<q7-$FAh#;vL1S0DSut^`A$b)?ueFrkkWyMNY9b9JQHwP#j8
zEOvvw6_4Vo*hLWgb-&mh9&CB2*qvvi5ttFXm(2P@#qJ&B>QS*vE-;=L#EurM5xvNU
zEqERt!ldhjchTrKjThB#LiQi;=sO$_=IuTEmiSGXY+RwX=<J8y3s-eyG?}%Z%F%+Y
z&s#rLsmE5xhSz2@u78#l$2OA!PXl3;e1Fq1yi3M?YsJC?MrA-(%-`>f!h<k~Df*Jn
zZfS+x?2=Ot{C=s+%{_{5(zX-vB|klP{Y4I4N&s?R1fkwrL56LJp6}8K)~s(u|19#e
zd^_joyliyhB&eVCCU1D`Vf*l@C=`a5c8SX4+Ri0RsUb>Hs(%!oKvp4_)g!eQp5wO?
zYAK)BS3;m1p5YPwp@e*G<Wy&LZL4d;r<xhGYMtiO9fw&d>y9pg7SNSc>FGLKc`9j3
z;l5}~33B%aVoC)sEomU;g3{iL!+0=SI{YYcSl3eGFwbwsVc?lq!tZ96S~Is9&4q*s
z&mE;wBC7i=Zh!J8(U9V#1Le5adyV=g|0RCxXPIzBpd~s0H&FbTplF%Va-eu26fum+
zlGT}F*V0n7I0qO?k(=tM7A+{jnp>=8uym<kw1&y)^t?>$d(kNFm{^jzy?Ut>x{QIy
z$GlpW?In~CR!d_Tm_Z0LYFW0NP@GUpt1$0_f#QT(I)BR)I&(sCQOn}if~&cumL(dA
zMn4<%VRv@$m6S<&%cRgfTY(8_$}}ybk%@9XO!Cyt(;SE7b5wm?ZISKBoDVO#aG{hz
z64k^ZHN;B1HJ)xwh+@S@l2+HYD8J3~Iyci|4-vI7qu5*5pBc~{JD2uZ7lfHsQ(8B^
z#Of)CyMM7k!=^_qb0u6Hc`Ev<RU2%!Gc@ywCAb2rx%crGDnhIA@Y)t@c`RUG$6PQR
zH_lP24`ME6Ig)Wi#bCM|^`p;NLn1Y8x=V^k_}=MRWQdzU3{<Yu{7Ll2Qoum@Sl1$@
zVB8H-m~bsJL~3%-hY8mrF%!{V4iq<1@1W$$4uAT%AJvn=xob0=7e0k`$l%Pu-;sFj
zjgDGcwYi%4$VwT-lyX(ST=fTgV>p^iD3qT<dvlNSGjNB((_t52MJCUbvbl3IxWbtw
zL%jtJfb`6uOCdDx01BXhIicvN;>m@x1T0A{t?LlZ?E*_G)g~U_qFQp}6pxqDhIstk
zw|^ULH$$c+noyu#)0@Pai<`CakFCiSu8?Q(+R;%|q7fU(FMdLFl2>5**F$kk#%F@(
zKx5xH@jNCw=n^bu`l-(vBBH`?SwoKzpBNF56+Ge|#<cXZ6h*1B=54<<Y^1i7`FZzZ
zsePG-t?>FhVj+1jl7b=@=v<BZIgo`65r3hANyvgyx%9?r#?YpyYjuL-QpAi3kXdF@
zL<7vD$DtG_QpB9bQi?Okg<~Ec5{eCSF(ic0;S_CL8FC}3Yn1tXDmm|yC3^dCil)#G
zO5}PN+CeFQ586RyIRNdT6z9<nN^u_TAQbPZ@|a}LCfG8>gF+sFYi!04rANs4Vt;N|
zQ3l{ESR$dfBS(EAISxU;GWy`;m{6Thj+1Zt_9VHy$0YpkPLfk9OR3D`uQC|Sk-dX+
z<eYz!S!QzNoTE~TGdZ$nz{NmuBS+4eEmg}Wa%7AWVi-t{&18-o%&TyBB);@^%#B)5
zSF_&)DuEBoya{Xr4Py2#>t=5wz<>C;CJ<wIt&YYA3DraK!F7XpunFw>f7S#NDwo4d
zASL`gCa|C72op#s9$*3q#XBaDsO5wSB**3;6L?eNg9+SN%>gNi!0zrlU_!|~uJn6g
zLV?k=UrM9>o9*mHhrQ(57^PYiN0nNdh@$x18Y|vAv_*B*Ice0*N|#kCv47=3TieXp
zp-^@1Hd)A&$7G@OSNR>IY>JU}t|wBweFfEVUkk<7XRRZ{d%{hrL9-*Z6-=`BVCu-?
zEp%|0X%)3LczM&ghTI7wnYX$tDlhdr23|)wYr7P~MO4m+roO9IjhXh9UhY$Lz);V$
z;L~+h(c+mcg({)cFZ;)^JAZNIx!fD23K8SBrDxS0o#%oEEj8`DaYxJ=(vkXf@;CJr
zXY;bMtG-Hf`}$;kV{|6b((c6e8%~^wF|lpiwr#(`#I`Z9ZQHhuiEZ7S^L_Wvy}x?x
zUA3!gtzNyly6`;Bf3n3w7;)uE!Ij2lvWRqD&fKX_80MV%|4nnZY2#2l*s98dqqnJ`
zOD~3sxtgf-H|T3SvvI46ES8?w)}Xf-C-=)2|E5)v{Qxwe=L-}Cr3nMID3kVK2O!6`
z6$wR6Idb_U6*GS$=L0600p5f$*!%f_nBTZ)SngIv$GOUq2Q7UJk}qDl%8cV6@>GTF
zNQH{wXs7Ahu(%`n3GnB+P-10Ps?3+EyO$a^w8x3^96w%uugg*pR8)yfe9$nRrYpEe
zItZdOW(rWs5`K9s_pAe-r<^OCI-L1KG^z#9zy>+<g2Qdl*Iwc1nXQl}NL(lCA7Fpz
z>uOH@Nq-hlzvW@*+g>k|RHvPdn7~lDR@m@C(H+RF8tJ(Fv$QCh;;xyK@^r~iJho?1
z)vkjo`~%Mvy#-sV*Jb-%8$}<5uxLoNPk=&?9WnW*WKzzXIfoX|ls;U8+Ey<#+DaEM
zs%yU2J}ceukEYu#>$ZYW>Tg2SXxH=$$NZQosp&g2-2UmFZSy8Jk8!sTNLaZIoCSEk
zL{x^s15v`0%-FjRrQ>$5#I*+xXS%>XX9mxG)?5S4b%Rg*0ah`A-svk>e-Fx_4>c#a
zj<*$C%zv&n?K!DHu3!B22exK^o_aT@AhNC6Kfv2E<8%aVFmw<VA{qO`mu3%78D};i
z;CJ|)If6O~i9YkAz<LieKmB)ZfLORlT3}BTT2Hh!jV#hDI2$8X6-c@Q*&=ZIA*!>w
z(k_&YQkA61a3t&FsfH9xxIJQi7=0ZyL4nkFO$ymIRn>w(uH`<lS4_IRw{G0V!k&a5
zavP+of(Pvbm!CR@7PRwoX#U%!?@3EW|4v47VPV1K(-M~<sZpy^ywV*vFZnzN6sJ{k
zbIxtOhD}z_9Ew_iG9Z><`&A$rSfvuR+SX@QI%;P!A?>pdM>cB<QIaa{6msENnkv*<
z<@np0L^JSOM#_u7ceZF9cCo6Ws@iOAT0hElsG$mN4L1dio<X9KC%*s|321@4sKLp9
z5~SktI)RvNWul$49UFbRUTBHg+H01F4_qq{X4Q+6oBjm#3GZ*;fhq7l8|v_X`Zicn
zh|)39Qox8IDL@;wQf?98X@UDI@q7vk<neg`E}xw9`@`Mq&+C`T{X}WR*p2&78&Im6
zd$?rXA$rmuXgtzad_~;^l`e5rU^?cNPtK&TT=P8KKDV)@kTYmK^TmJYmx1|b;)gS*
z-_zo4=tMJzB?qm^&$la98g0ajk)^wHT<lX%#lWpE_nJ;P#jVttSfDb^6%9drM{27~
zforV2VA}Q^-cdYrN`Y$KLhdoLmQ~R}YwydW2zO@}-+MC}m`?5-_s8450UbZmkQo~X
z0Rn%juWwct&pV~Ta(=wO=-?1mpQ&=soct~T49hv%OQSF-Vd0<Y=t&+|*8@JG01Y2s
zx7^;O9<`1kgI~L9CN@v?ORO}!?$Tp*bMu+Fhj4$WGAa*=z*a*FD-t{jkPVr|M0yo~
z$&BK}DBdKEQaghyS{*h5?luS>0cVI!-ok(1555d$8~`Vb`@P_fM$;xdatJs(iG!&M
z83A7wG7b=81qp;C{@%wAfuZ|>v=6>~#*C(o^sNxO@?Ry)P?O9ZPe#UG$|*Pwi<)mO
z<YC$te8O)fVBGg&oW1wK*f`*fh_UwpIc*1&0TGQm(L8S)?wiN0MHrW?g;<}wMVK2N
zO<Q3|);<g?1g47OyAO=m@OyW_A>dfwkOrrmgsg@C+r$4_{g2W9|Ho);BpUbMZ_K~|
zxE@AnI~I_%O-ny%TObSnQ4_v?@goL(B$F4MI~IA9W=h)fYz9C%tTJRJtrBWv0VC8R
za~%|Ac$5}7X8(L?o@8DNDei?9`HsCekg1ei=G$@CfHRQ)b^eB4)}9ADZO1cI+TQ<}
zTh`v+ox2-u!L0~1VANM4*=U=I(bbl4g0HE$-9zX?z}J`hQO!%Ycb<n<Lz2b@NNw%2
zHJHVetqpp}PDdce7^eM%VYm1x^759QX>KxLeK&y?`Wr0NhFqRlQj9*=3O@xC1)d&g
zH|35mmCPgz*>H2B)bLQy+u@{lME<nm*8A241TtNPczFgo54uR;Wv{r-T-?21C^F_C
z)m*fD;_%0fR&uGwQ<7Gv1srmsK8?CU?%pc+d1!Sdt+J`&c2!K=KaiZ`Vnw#&BPViG
z*rSz+?RO-Ua1FbI*?F@29S~Oj*4hKMOrpfOVij1J$Pq?b8?ULHxhXC_*>egie);N_
zsNr0kstwEb&J5-6_AiA>pza4Jwi{7N!PEi9Gn#pC;}l-mZ0DEL_BJw@t4;t>lU^dr
zbiYq4jy-m~X?pn?u<UD^ZdKeB!+09rJUE-l-R_+2$DPenIzP@M$9rc;O}c^hmZOJ;
z6b*4ndYA@dVNRRon^duGTYnnR%-O`Q*i5##aISQIm-WI&X{1)yS@<G4RwRh45{aK3
zvFf{IO_X#<r)nh;k)R6OWj>^j@LY;0Rtvq;-nQl&cXMqXk9(F?k&ANS4~HC;MZw8c
zX_7WK>DE8AT!m)FjPh^c#RUS*!(;l%54>*=vXguAWr@2P{hGG*G|5k$`JB-YOpUF?
z^w#eLd)ph%6Esp8aso1GG3Ah=#Y~yDk~%r+Q&kZQ5T#9*>lfT;Z%^bxxIzL!sT&Ab
zartlbc@c*HX6zLjH4|W*kJI^kpTrtwd?`j|>DpoDL+yjU#G9S@H<bV<w<hq*Z@P0V
z=uFTyjv6kxZT_xDk%FtI2DMxdWsh;l6VWk%oq4ODiu%?M=`C+&w{Bu3cfbi<?osoT
z5$w8pi{vr98uYaxQ$WIa&AUI9s`|T!m1Hf|eb^S430*Km9;V4ExVw=6(p5BA?R*b0
zyb4H7&J+@$A-|u~??)i`7~U+Ju^l0Lm^0<^f|D-n#iRDJ%SHSrLSohu27!am$a&A?
zL?XY`2t3aQseo$jYpcbY8oX^M^pFYG+dWV>j^dQo#A7TriWUnB%^6kXw?MPdXDoP@
zxd@Q}Im+<&mI9x^aR(deaV>Zj_^i~7#6gV2OqePF)fv@+8~ClW$fO;;l(UrNmDYnO
zncESq<gY>%n^f9^XtO>BjlOaMGXO#b$kk`QLXPQA%U$hUZE&-?%rny5$mAR3bfT<I
zv@D~tMb$M(IozZ%U^}7T@a_Bz$1{?tG^v8}Dt>JmqjHe;kzCawbxXLFuJTmh#en~D
z&f;qU(fW<O&R*!QUJN*+XqAa~D7OPj1}}jx-V{u(EjhrLyuUD$W;*(3#`+{%EI$s;
zkq!4~6mj))CRr`pNv!9$1qmwX*_%1HOQ84X9}H2cWcm`M!<4?1hWI$~3Z83QyHexy
ztI;5~H~x#AHP-nB>l)r|wx0Kf=U+phl=OfgcAD1+`g1vj9~~ac@^KW=ro;U#vROAz
z_tkFQ*B8W)b!xvAJnvOB!ThZ6b+y${a;&aNw5MmIm~gEyH|@kG4wbM{iO`qd5yzXJ
zw*X08Hc$w*WIW_o@tteOyk+r{8(-vPzxG6iZk(GT%G<}o$1YIrHKpf2UtDw$rj!+Q
zcmd$>l`pI=AuOmwPo$aIxwmt=R&ke!r)7rj=}ALMm+zqO8X{|kj1OU~$a|D8wYonG
zUnC2G>iXvC<7@&9GIk(SJ7(aT!NYG&9|~(w;!v^r^^D`jM%f1I#@NO!E10!tO(&;)
zhBAm#VCbX_;lyK%=4Ar`SG((m`8rh%Mlp~lqiML<#m=NLR8#Vw^8<Dns^gu>Q$t-R
zT=7G{?Sofgz2)Y$0NxEsYT(U30kMqgx5bNLp77SM##d&<cK$vHs8C~V!x~gE)KVB(
zt2vIisjNk=qoKxW!$<l^=G%(K4K}s|inmIDh7Va`Kjpd=3<>(l0>!;5QdINV`#EsM
zNjyHImGh&z(EgPQ4+gpk+!4wHoFGfU|6uU_#a5Rw)@V_P8wEWE-GCPs&@d)_Qxk~Q
z?-wG5*)D-;qF}{c5;3ppxx>+!-sH3c6XJn{ik7wVsk`6`0zszlta0}e!Lr?OClTcn
zeq?9WMgSa6&pEmiD9A*M4kv?ya|ep0{Hb(^H9R#Thk1mdDQuRS$V@Z2*3?T}bW3|y
ze%c!svp-ki4lv*Be(#V~8gp?XVZ8$LU`eRbpJJHYEHl^?d;YslfYr`Bk>9ZNZ}hAX
zE^D+h{|B}|?c+HkgEYADArQ?-|EDsy-!Gi7lW6~`qe&1G%iz+qq9apF+hX9+=0ZW~
z=p7c_Q)3^!Cn2nwAr?fyahw4*3T7+rOj+LclyS?-GOW<Y4~58Y8J1w|w`{g4I5dJi
zMt6ieIKror9ss_XQxzx=O&qP)C8S;}?xgffn5BjhZC6o^8hjNLssmN<F<m&abrJi=
zH%rmsVLL}X!<5i}%>oxw#w~`qu;R(|DWl-o24~`lF?s8+k}x0s$F#FfnFC&M`aNkM
zWlU-1g5|XXidpKYeQh;*@CqrxvcW&+T(T*!enx*8hP^Sp9OXZ_W{leBRzZI+;$zO}
z7yOrGUYuX@e_9!+Mz7{qI^g#Ye)qb2LjStrpzYK!+yaqf7{5K#zT5!0x&Os^96tye
z<gk46lZU|OEdyfoVwd$V56h^%T<o+|r4DYkJn)~Xt^88R6N-->R^szDB=MuB+lP7P
zly0@1+9${=M5*n6->npd2o{?EFW1Y;+I)ZwumAcDp7DQEy+xoYf-2u(uchjDRZRgC
zArIuS`5qHN5)=q32&1&LEUZg8Fc7{CUzDKL88Axpr7{M8am+?pmF}NHqmTox#j&=)
zcF5LTI_yPU5qTTqMs7>Wur{Ot!siogoJhM~fKaYu>sT{1^;}w%1*+Q?S2BB{V2=4p
zMnAZi??+SGjsepDZ!x}!0Mcv}U)J9fbUS`2fjVl5-yayZQ=%OAey1L0W%`h)$yhSN
z70zYMMC@rL%N%A49W!VjkG*1he`$b-ZW;8|A<F8QZWv-QsTx2W{EK+0;PZqU%qG2N
zu0;h>vwneo#u#liV5v*&ES_?GEq~-ip{1p6PADrJ-CQVt$*T4z<I}SSu${$`XGSf-
zQfS+f^c`40^|Kt*W~kEOdMh)(T-!6zrD8@bZE(E{haYOPz_kGqIT^wvXZc5_Mhb}r
zeKe(+<KcQ4Cx~*$HC}{#BLER;hL@h-ThfLBXzk}#joN{VxRYvuzoknrA5MT|g)_=B
zFFM0z*i-g2i6cRjFE*_t4m(^xk2l!-0Vvd8djyIS98IFVVXs2<x|&HrtP74Z+2nK2
zZ)I6zfbyML=Nm{pZ7q4X+eG>Y2v$R{%P3cd7!Fh~7U^$+$q9{Irv@mZBr+*^9cl^p
z^c>H{|5-~apDHurlOQv>8<l|^)c!}pV0!ul)>2VvI<|nLr;<52H)&8{MaCc_M`snc
zE(gS7o;t(Ve*sOF@x0fh)OX;?Li?p*o5V*jeh~+xOjJl1JwJIB>6n_%IXG(=UyX_x
z;|G0>doF48v*GO>P73VN$YhSz$fsB&XAk2E@r-F`CWjNIDI4?`j#G$0<_2k`7~C%r
zq2v{#t;6wNP305vcD=&q4&3TVtI6F@jR3Oj3qGI{=}k!)Y3zAA9%{G_cF0g7xp;P`
zYIGW0=K<ytqMiqg#6lsb37K4*((G)}Q%cA4KsFCZPtuQgnD2&17fk9iscdy>8A}q;
zIe>9KyqO?^xUfH#1LcDdp7?r$ZM~duhuMVIAfB(=@paBa5D#e*pZHKY{qSZM$$>&p
zf7?(ww_Jsw7We3#kpsy2p$gq`IVlE@X9QHI1%_SkjQX#&*%*8CYHt+8!tILppj6zs
ze34<vcc54noZfR4J~pq7bd1s8a}jyK_yEGmaDP>NaLs+TMTOOIpH@e~cyb&_h=kcb
zGlp@Vgla{HiY5F7qi$g!F>*~k*nnguY1^^rmIbDLF+UxxnU}34i3AX!F(ggcGwH{e
z`lD}8|0s|}{OOBEpolc2a+j1SIkuO2zD(8D9RG7sA%*y{u!wkx`3T=NRgWy-e~obG
z?a(Fpf~bVWfk;$Xv4V$!{(vMVEb^v;qA>eeV>agQq$8X-dkEt#B~wd9ED4km0Lz<r
z#*jLo>x3Tc2{l{nmjx3OD5YEOzevk%c<lmqC~7k$5%G#1U&$T!WG!%+!mjfCdGNQy
zff`TT`@Rh0C+7KZKIRssP4oXO=!v_9B!B6N<n0$@58<7aC|u*2lxQT@J^VFD%QQ5}
zLD|SNECor5&Qt|=j_~8JA#nCDccFdmuzW4_ZXMEJkDr5r{mx$RT=!V%UZyrz<1b`F
z!@k<js|ap_Cu)*Pp}#H8EDm332I4|D=f_6WHE%rshz6Hu;O1J;*akli7;r2x`+RQm
z<lKC~h_G=;-W?2x#2J`R(Vd2vVSa3cO}yMen(jufZIOwxmjp|*(E|e!0g?q*vjgIl
zsOk7th7?J6>8a6<DRE2Zi-n>g_8g8qAm5O1Hy1w!HvkIZcP}F9y`XN+52Y8v7LK20
z+dB$zXg-0*a?^J*9QsciY$4V<uBqWl{M&1qAxV?m&M;#W3aJN@9Fx(L3C)h+cQ)Ir
zFxI><U+QhR=~l<Vi9mTDX7v%Me4{H1Yx9Jxp?<shq5?Iqx**mxSytu&ufnzWKqP#L
z6|H}$m7Mwb6mYfDq_~p(smC#AV?rpY@d1jQ`3ICg!GBDK4~Cnip@@j1fF5j<j!8`y
z{&AWpU7P*Ah(@WIEL(b2y7ea;Qfk+(p8J9SkzUWF^$^EW56=eNq}HV!AGgY#f&%<U
zGWM}<lY)DlTl4grqahJI{)ZRj5~Q&UhoGIfu?-H_Q7io4_kT5<du$pyUO6A+qhtID
zYewr5bpi9rT_?;Ix5&oBSd2&*%No*SJpp%}gUH;$K{^tA21xHvc=$)_|4qT9^l~!+
zQqEJbC{k*!p}<pa2@%pjd$?1gh!EC+V!+bzLdujX?d-9`R&Cue(qySdw7&j2Tm8^X
zY+7DuA>kpFVEf2eoIXYW#;|t}@;@OWn||`_rmgRU(1peuhtD;jvmS87EPv`(V|}pj
zlU=}sY9@$%k%Ivt{Na(i9^XF`C0Vv5Sw_YWvhYabZT^2_GF5%P#bP}pk-I=F5H{t}
zT96!7?tR{Gu_vC;A+ytReKec?2nU)UcK^J5^F1<V<KD}VX(wF&{<mM|$IAOc^vX*L
zyh?^+CgaL%Npc9g%v=oIG-r^(rc=DUkU~aj_MArSqC3nBEe`JvLa`VBduPPe;bHy~
zgoL_+`TY{B<_7BuU|p2i2Gp*CLNhWrXTO$YVTQzcIbu=~*RoS<<7v%rAveHDcpd(s
zBQ0TR;?SKo@!!Ygt{qmSTJyQm_s2uURI|I#Xc3HRv3_xn(}j(EM=6Dus&nmtzoTy6
zai-<irwY%4Q!$5qkrf+(;>B{C-3~<4{*;s$LEI3qrAm0)K!r^s<^~PeFwV-p`dN2+
z?)_)fG?DadpzlX&Tiov)*zN#|SzQ-E7#5z+7csBEXm@fMu+L6!CvZ#;0G2J9cX=9?
zJ7*kU7osR`x}z@z9%VZA@5wGKH|{>sCn9TG+K}8#lOAws<VjqDZWBEEip52+wP0^{
z&N;OmyfIb_tVA_160$9*QGy67<9BE4E<YExhV6YAxyt)tYE9`YbjH*~<rDkM37G1Y
z^QmaK*8T#GSKdmt?sz@Naa6<Zu7vA-HWm1jb#!+K*{@e`02z71ZGYxQ`5DGch114=
z`Fsw{1#kJGUP(okjbfEBGpQ+kYR!ScYh*WTOli&($N<-tGV>JGPh~oh5Q3c(PH^5)
zN?%}Hz=@~g*L-Ic$AJvzi*{5TIw5sGpL)$y4#E~P4Ip$$H^At{WH^<&#)b2VQ}(eZ
zPjO^3ueA8-c*_R4RxFS`NF#U4fwu%5jBTJ+gGlBqQ(vhMtFKb;kYbvD!QzEqdYnzw
zIY<!)+%A+#kFxlkd%>7bRc_|hlem|7EESS2-Mwlvt(8u?C$64hCaGR1hu27zc9&?K
zZGbPZ@C^<+^w0bj!<rovmZ_sb*0``h5FJW%+|x(<#?YZ`S}7hM>O)P!&D7#p!-61D
zYIVXJ&y=O1uO7=L&pm4*wHww-=a}l;QWtL?$Tz~=S8*4AnaN8&mG`7H*d(#EDk)Qx
zM?%hU6P*vUP(t7U2p^iXH{00Q8|}b5tLqbbh#pWL;W-wsU-2fz<{xFN-78ztd-$i{
zbm|5E61&wpQ?^vOGcd{+q*T8gewW<A2B^Wm5fR_%7dg?1-AW2YX75Nei;!U}&(V?t
z=4NYhb>AABB*jT7d1Na!dS!`)p@PT6e>J)f8IEi>M!i2E9=NEauM^J8J9K8F|B8r4
zaPc#w|9AdF)C;hm5}J*W{ttW>F4=ld@dE8YUF`+ps@!8rLSD_cNA!hok|LYCy0#W>
z_J;Q0elTj}Lq_AQRR=IQtC)X~Rftgq>?o{Co~KUM96LRbivK1~aBFOaINOuUsE}ip
zqpJY=G()=b(!6#;G`FYH@(k7mEW%PL{2qgCOJQb(vredfROl&s6n>b(jBZ2&M-MXf
zmD64BnG1I9S#$+8+erBs`?_91Fc<9_fbF}nqJ<ju=SWj=r0tE4sKSU;MBASMDxt|S
z4PH|QMIzhOcMdYW7#vC4IzZ`qz<&^P9Qk+;7m0}}3^)%1QTM6~efhR%C&53mJDLGb
zyNbt#mY)TrB@Q%A0lVdyGcifl)J|w6ST`zM4Y9>1A?1rlldho3_o*Vw2zV;Cx05<-
z0=lc|W7OMQHekcQ2CJhLmo2z}bKPeYMQkrVS9B74lX501;tDuQoA$XeSP1R;9cx0u
zq9S%|8gwhU{&(oq?fyJa(pu})sT&T{*-FZUV-P^%wH(UF)VYSle>YM;XGnH-K7yHC
z`Z1LLl7*8`LWsj->Uy#W9^OJ~ffuHU>Z}t(oN)Zf+Hd16p|s0PBXQ~k3g!<;?*_2)
z*>H;}M2&xzw$jL!x}rJPN(PiUGEb(wjHQTwtk9AwQjQA+_;EB=!a5~ISf<SR(3Bcz
z4q_5pe$vYywH8};>)o1X)ePJ!E5k_7M%La_?uv72-X!DPsk7M3s6d##tC0t595L|P
z&cfRt)&~Yn#r2-u-b@MtKTZ0YNP*I5PPZ)aH3eOqr-}ucnt%(gT)NecWd0P<l|}H^
z=WixyMi`=)4!%5dJM&0T{Dbp+?3X&~wHG^CsbzT#2s=;BQ_EolircxR<_6WQHL2Ht
zxptx#@0C%u^1>_q=z%3HNKH?<eO@|p@6ld1r$cS*Evk?oyKeWu&H-fo)E|v<@F4Jw
zTHyj-IySK-+(b_~oP6Cx_!?4w`)MsV=Wki6PTGE}xLBrNe)Vx@M3+{z6{$ZwOB>aV
zUyZNkzFCPq*v(u5jN6tcG!S81IZr(uHG4mbWbLZtxx|>yJytf#t#$Tod=QbX@S;^;
z_R?^dzcy)7Phm|`z%UU|Q+kvs0iew6DRDd?WWY%cDZ71E<gPchYotHpyYPBFeSY}^
zB2P;;wh=syo1M%QZ7H80h?Mck;!?HNP#6Axr4fz?WA5Y=pXm~xEuNn1-t2`@Wj(=U
zBU0=@pM&<gbKPfgc7GIzApfrZ&N>t}9NvrsCJG+xKCj+cKO#YAQ~Cix&Ixb@3ixiZ
zo`F1(&K7!wGPv+alXi$F2*ELFk~cNd7Mmj}%le`3PkMU17s{-EvvTxYfmaO*qmk@F
z*K=6p&=R+{H|uAY{nOPu1M_<AdZ$h3a^_}*j{!ZBCO$60D_oOIwUPs6KUt?zJ7ukS
zasHJtx!0`kSbXuP4|STmv$kE1-;3|V9s&*At81MCO=@l!WG||Xy}bfLMNv=9-Ms~%
zQ9CgJ1yBIKPvq@9f+Ov6nn5k&dsFY%8y#lqociMc!)Z#Zs5Ap;Kzjl>XZ%pc9w+Ro
zGZWK!|7^boUHZiiiA5{?{w18U_5u=ZJE-T>>`l}>WgRszK5=#I82w{)o<9rI56CK2
zf{P#t-?jhDyzidpI)`Ij%AiG2X#^4pUh<hyDSeonP(xAK^gwm3LYgtn43ZoYaXfye
z{EodmbvEpIgJ;)D70oz7=s=oM4K9Y}vcad0ZX^}IFIFZz36>+~A<d%c)~d#$@lBW*
zg%78@_>FW#%OIi>e33^xOy=r-22@8MUZfB6ALm%c7CuI}_1;(^=(p#Qj!aGtyM#~R
zShg~#`-o%*nx9u_20(dAY`LtFRa$PfpwLDEul^(tb+OINt+<8cYR5zRJC_};$0LkZ
zn)IhjE=Axd<30;4%VRm*rJC3yhG-zhq((g-B*u;>2keA0F)4E@xlWqP0;zRubLNxv
z@+<l{DsgWwpf9HWJQHo_^b1?Q>j>@)c?zIO5M1+BU^bZfC3M#6nzwFL>;7Z!mg<}l
z2WgDm9M;VJ8|`0Z`izhi)QDQ)EA&Vf&0*b)Ye6*Ph0R{sv}BDGzoQTUm$eauGs;yJ
zccQBcFACYm<rz2oo2Uh@6Tbvqg0be`;6<r6SK`<<R}!0cbMftp4u(cpx+orViU2JF
z02u5gGOyTd(3VI<KCU>)sU74TsQ}|syn9pR_`?XGxH<Zp5YY(0*u*va6WOd0{NgIA
zg=6`PZSSb9E2xFzL1Q3In>UfMFY<BlI6&OJDs%!j9(eAmEE-eKd%>~gH-a@g>*A}y
z|4Ymn{v!;@2F3TVlEfa>Gjq%ROdJY(2B>Rp^j-k*m+i)u#ogMmH`uMTmj&yRt!8N_
zr1R#kEn`Y$zc>lrmaVc^NqnZUYRKC|$A|B2FWJ29OmBO_Bk7b=H}AMfm0az_y7Zb_
zrX$$kb-ZQuH&-LO3(By@HvKHvwL&v0$3S8+J&Yu))aA<|eR>-v->sad@l>Jk8gPwm
z07ufz25bfV-(nL{BRJfky(lrb@y^mfyz^k#CDpVXN9L@gJ;Btf8<AzNJIoL2l8bhc
zkyrfFG4Os$EToqg`oWu(xTPdd%eRok)V74P$2ow7R;=F7D$i{@eh4vUqYtlM(5U+j
zYIOAqK`$@79Ijf?aMki{4XrHcGvLI9=RsA;ZihTd;#Bcp)tEvII8rqF@mh?Gs*1v(
z4bsL>c0DiJES(b*%@0deS+xM!VsT$D1oXR-J50|C@3k}T<nsI7s9X^4eq&xQ6dH+z
z1s|1J#RR&iTzYmB)tmwkUtXt10OsbDNU~lR)eTqo`@q1BRS#d-ePg)79#HP)>|k3*
zDw2ml&3J<RpdZJGv;kh4v5u#f9;@XLeqBSmC=H2YkcR1S1X+6Ag(YzS>4hR78B!J<
zGsg?sf$qWn9e9AH=!8Na`Hy~j{$~%@9$hvEAS>W#v<RCuyt#_SYEuK3r6JJEYrKGa
zf+w7zkskb@pa5OxUfgb8_t!x~`J~oU`e7HywNJ+p`%>2~)7sAH1#kOW3mqfNT{ls=
z=g0q)zu8h4Xo+c>S?Ljqf2Q<NvZJM3OCbyakxA`~x4V2jb|oNksg;+Xs=t{8?<LLY
z(p*fRKaL}Sgb?%w4)oz5E4$TvAq2m_XaI~RQx#t#+c^jl<1OT+EIMGLdcggDg8OSs
z9GXnLPt4%Y`+g3mF@!JdPmy~Mgf<A!EUQXoFiwi17NWgqm@kR|#Jzij*zw`_M3ioz
zWIH4iDenH<=HDGi`*8W=M~NOvsl@w>o>3jFEyxW62sXcjc#v5jqWv5_<pdb5>?PD?
zO~B6{nBJb=1V6NN2_rcB+c>EQ)g46Y77$!wA%8hg#1q&{-G~G?5I6N+gdOvFVUlcc
zT=60doy_-j_{$pKpDnZKl^wE}Y4vHq^8j-%32TXD66pAuGAm;z<zB8nA8nK(_0?X6
zu#pGFnL_ig6vwqMA@L-^T)0V$AW=#TB>)HIYxTQb_#i+;DL#0VVUrCYAVI)KiIVfm
zcHn^{k#aq1H{w*|U)dr*8~j(m7{v#d;&)j1)QwK;ND(zzwo&|be+;AV0*3@paqI3@
zBIBSose<e?rO3WU;YrqW>Ysm&Z*XE$dexc(?&mF9$MeD4H|0rU56wSaA%amzI#UfY
zK7m`>e~UxKVP!;bzw*yY6brnYJ5(8=*b~@{d}T+x3tCgK`4UViCNgM4P0L|9vuDI^
z6pv9pvuvx3>r!?aO&^JAPA&$3+E<X3ELI8H3`V&=KL;T4d4J8LE}_CRnXA=gYcO&a
zT_o;yh-?ad3RIyWm?~RqiWt@%>92Ce5YyZ~TjRx6wsPOpaec}CGHza`ADy~`jb?J}
zry@g)UEaak&Ms(4x+=(U^tdoxwOT(ZDN|~oi)Y=CD{mk?CS~qyz&5i0nzaNwyyWDh
z)!ouDc)=&iOt93SJ0xW{q@X|FZa_&<W|TU^fMhK*YqZ{Rk+dbqSBtdUh4RWFlo*M|
zUcz?4X?bRK5RSQ$s-8`m$_i|+Zl1^{cJ^kAMIfC0%4b=MM_;3g!c9mKe^ix^>8SH|
zR8eE|7_tyHBIUZKyt--xzLw0}aAVjIKW7@EKKwB_ysrbBJpWxhvLZSUxz+XaGH3k$
zsM(Dw1-GV{s8~X2)tax!hDmy_u0J8UDzAX1B;J|xW>Txn?RiL<P6|s$&V1`Ms&yXl
zL5v*Tt*Yq_XrucHMCUA#T$TXk9mVg<wr7i^O|O_1PX^Vc9@Z%VGDkRdOw~HH?EkS^
zDEIBAAx|+`uyBIgkg8EQQ*1gq5ahADDq;>%W>=!Fw*}M#vQ68YoY=Cmw9u2I7QKbX
z)nsa7?MrRgA#n#)hmE-=pTh+kMb7Iph|NqqCd<rObJ#s$HmN2SZ1j~h1QJ>9=;o$P
z=^?}qqso`GM6;-Y9(0Fv($wJ3;W~*N8nqE3Qu4u!^&!HSN#)S<)yw;1+^r_q`i1Te
zb{tBS?bmOm1DPhZo{L+G8|Tg=YyWOnKeQ^qrE<BaTDwpq?-L0nl6ImkC6GCuc3Zyk
z^oUsPF1_Viv-ZbNxDJygQKbg|P+V7&(ubW`GVq(D;(A&Fb*^!U)-b%3OcNtBmFmfF
zd6BzhY$ageN7oewjQzS`c7Sl5%$-53Cez83W)13+G7zM~IAe&>NmqkA0{ofVWxu9a
zcY9R#s2fZ}>9u$%(G5yEQVykGH3Q^{iFw=KKAc!gK>LO)IQ9pW@{J=;=C!K<J#8wR
zWZ)?-!gS_9mkD+`ffs;(0R)nYWy6l$v4N2zWD^|b@T<h?x<|eMK7;9%-A_Ggr+})n
zS^sYM=wIO1mClRnql|PjAv-KzXAHtu4}lr=q)F@tH8aFYf|HtRJ&}HAh-a3x40P5+
z5c3^0{h8Av6`Ez=Ghsd=pRkad8>M9^s|+rQalcK0r-VL6RRlDqhBHy)=)Y@)xS<ly
z*RzeXvvj$OQxgoy^pfmN^{}dv)OL;9pG<oo7-=WHO=i)HvvOk&Iv*`zk{F0`5lUeV
z{r2`VxL80nP$R_;U{Et=$8NznMH5z(w343u1T=bkTC#;~jglu{{1BW0e5#XxR%1nh
zui#rC>vNYD3HUwSg99uPR&cH<U^`mrm!#sf0NeOAp~(b06X@uZ5Q(w*h_&1#U6~?N
zv)^+#E3A2YwuW<)wje2lJTpaA!|PO%Ic$=V=p_H3gGr38>!!K6P!G27lRIb|j&mdV
ziUMz%2lyBrGXy){A~inIVb!UlTZd!=Hv2sAP8OwK=3i0cDr%}=^#?}QvZ$1UNxy*e
z1U6Sjyw2ZjD6j5JRfgAz18ohGrP|iRYX_HYq2sc=^tBcKQ5S0)4<^VZM!3Qb2Bcm@
zVaAakE^;2U%Hj@kY@Fr}rOC&NxvulrWZhN{MsNJ`KhvS>pWv0kX_H_>coCUds-GNy
z3Pk;t3AoE|E$tE;6luSXKd+gXefyDxwHR5jIS#0<lOoAHd3TFot##}buDWh!nl1%(
zmPJmxfby_JKWX3k!|iq`Lqpx?R&f9TS0QwwLe}S~VnFnG(_bcbvWbM~K|;lV+@C+!
z@^1}hS(S^*|B&OE3uy8Mr&xmhtGL4h5;&3QhCH-IV?q!m(vOhuiYP?Kv1?)M;}N+1
z)E$0U>x)38sk~gJcTH%qiMcn9!#Ds#`qZY>dEL>FIaX7U*=<55U#NC5@qS9fBP_Wt
z4U$6lvv|2T85b=sf3;4c{@9?X((;w5t;?(poy>GBW`+=%9~`0W_HH9$_TK*qEV68j
z<{-AA2xU}Feq+)oQY~}8NSEhkV;WRkZ;dz17sr;9&TKEz=53Fk!i6p`SE2>^D^=;C
zRF{?YnA!m6>`}a_I1e-PeWH=4EZEX=X>qc~8)r9jEux5Wtq9qI>X7xKu582`6c26u
znQywb(-*XCMam}2$o3q=SDg*WS_}qKS9g^!q9^0fI_p#N8_xt|?=t<j{irhspb^n6
zE5nYzuDj@1F5@Y0f{J@sini5$ok2aS(5@VzK)H+5v*1T+-FvdU+K+Q7!;2lTAmv1H
z1F($|mZXWx<1+7=xj;zE!g~yA<0q%ZWLFz6l4WIu_MGxUfX)Q+5;Z9>QQh7G>aA}W
zWol}Ouw;M36e$b5%X41}eN*Zx;sIT(G&PMgAXW#sC9Q-4TW=KEZDm%1=xh}UudQ|D
zrX1kJ^p~`SQAoo@>a3`YhNntw3AJmB*lPF_0k1B3S2VC9;nE#A@x+qk?(SuM!!71k
znDzXXspI4Gza)bxC;s$s;EWt>^eG^$pubY=Z3$sgpeAT2fs2|pc0a!-^Dp&!V>*$B
z4`chAg$xK~PPP5Gm_wcTui#1K&nwF{lf7@LT)sYd`WM&YS)&V8HlBHZ;ann)rtLD>
zX|Ll3FpeD|qiel1QZe}ED@BV#SFZs2s9X<K6nV!p50pm)vcb`163HR!CqbYUVY~?|
z##B>{l<U~DfyN@nCV$TX8fvPB@_(<rM66-;<PWkisi<Iy#^DtcBL^W@11z`6<P}_}
zpQo`jeh04di>d`1o0!BU8hQj<Oo7Y)X_i*a77%HYrKs5J<ogk1fj=G?(9%1q4T@gN
zO@RU}wObiFZem-jxl3|rO$A6clM^%;_dAsM1xKSL4+Kv#rZUJeH{s}`-{a0i@u;h?
zpot#FTE!Up;{z+g&PbxAkHe*B3NA0=5V3D)RvQAAgOmq(K1fv8A(QL)w{8#c$5GZp
z10N80YZ9!GJ%%qmrbRAl)Zs!S+CS4SUMHr!#T7ROf=4v2r9MC&lQ76jpC3A`PKUY6
zQKBDE0vwlsxD6W1tt&Z;f?W!eKNlGEqvf6%qVd*841kx%P*y-#10yyxiBU160z8WU
zY6o$jQT-E8G}|KRAukfZPZb}w0w%Q!FRb%}l#>py)kfj9BTbA62D8(NX|xk+1YL*!
zP7<t4QEX_qia8(e7f^>RyWq!1s+0_7SNH3?JpTz(1{>5|X)_q89d9-RYno^SVt}i%
zs#<c3;VLtlf-Pa<x3}|lRjx*UZ45Q+tM>7J^+iEZjHdh$$0C$@V#vtB!~G6Q?kx%i
zPn}uj$&oQ}G`yT&ubBi>=E-H5?#Zjtg!k#b-oa}tDpOhFW6zwyy0Q;VE|FJo?g;K2
z4-BzN0%;JYF@NF%-9yH9<noh|AQt1gX3Rb8X|&aa{AFj}nl6kCV#4Lut))2?w&$Xp
z%*$P|HDzWt?Bb-@>Eoi@S#6h5$guV0pd(Zofzu|N;kmXUzGHdltF}FXaaaFZgb;N5
zFX!RK{l5SFQP>#f`Y!c?@@Sd?O}nz=31}MqWQrlZ`+%nIHknIZ6uiXub$`?RDqYmk
zDj4a{e;>=!147;Lo~PlsP2=EIul&1cUPFVTwJ=Rqk{HCtv)RI+EE<#GC-#RCcc=0q
zQvHhu&+mwg`O~Y#jMGDy#y~gut3MTss(Us~IMLDK)d=xW@u^e7#fxx3i!tTP^}wh4
z)1+5KN`XCB<Cvag&_dzM-ytE?!^x^0Bt2YxL#q$1ngp<vmPk+DnF%^`sPx%CkgRwg
zklD*fgxw6=?{UH1;rEcuMvFqi6N7V!t8?=f3uzIPh5v=N@afm2nJSZx=%p(NvF9|1
zq(5dh?!RLX2qz-%h(6j&`)h~0g#xi@AQI-q6Qv6}-%a0>^CFp@{=w~5x`>ZZ29Ajb
zh!5xqgidhgP5djkUNS%KB7_J5Zx)*>Fa|@V_qG61YO)D^4)lTp&;j+)gF%OWuXNUU
z8g@R3o<B5V<?E|FzL7Gawq%02QVw$Qa^Dt+JO}yQ9vc_#8gYYv72!6A*#oV#$32O?
zE)>ST!bI1kExFX;4S229E0n+6t0p84J5jSFw`(G03!PT@%xF;vvJgkWN?=`L`=%@!
z$lHr5=61YcuC3rVHTv$?K4HvSlU<i)yH#@&A!m{3<n2Yf+*S@C@mzH|&Q)*bP8x-^
z44Qm&37~TLJX84<QhZj-oq?e2t`~(7G8Ag@%<LLkoIL7aiVu6XH}$668K7uio=>8m
zK2KKj+S~7yWpqQ6)e$kT)~+fajMdY}`{S>UboCC_?BpNo;Nl!B@H1MWFf^vjHCdU`
z5>oAL<(ZRtX_LeGqcSYma%Zz0ZJN^QHz67vFE_=nLf`Y0ZP^QKfk0A`oC*jV+)A>O
zI37oAOFP*DlGRS>*OTJ2BX`yp<!QK=w(O<5n^k)Su>83e%*B+shjMXGFNXB&7(Ut4
z!HZ7oz1M*ls5gYFs?0;17I_A6tL7cy(cHjdJJtKkbZ(lby|6~1Jqa}pDQ=6O8;^`;
z;`clMP#P<5*(Dcv`GN8?TSn5l!<ji~EA6=CEJY8pC_TSX27iVeh2%h9Eb+v{T$tHP
zbS%ym+POZ}QqyN$fXt^6^*H-Nq?Ur2;)6>s>~a2W+0$=K;bJFX>BS&jn04wagqzdZ
zkP4qtBR77iCOiqOac+=7*4@`MM5B%MNY4l$?<ISx9Sy7*^#;aFJ_gGrX&sZsJ(hM>
z7)<(KTK=Z&mN`0B^L7bv9bBFYcr2^V7<@T*taXzvj%7|45J99*l4D}9kaS>!>+^e1
z{^ISz)q`WbxAs*Sg|-TfJ;NWk7uP@Tx?6GlyWedRQJ--)&~^8Gk!3cBVUS^8-xoyc
z{01T<73`gQGX-o2-jR^go8SQyu(pf6RP5j4Z{}aAd2J!qc76>te{vB`6?ZKe#4g3?
zQ1WmsrV#Y1svt27Kj;&LNmIi|@8}<n-xTHZ1o^OL3G@DftPe@}v(c=9Yh2$T;|II$
zZRD{fV|*TR-=ZQO^YcMov(-Oxvrzdh$iW=@z>$VvA{nTHyiZu?HeC&?Ikt)}z%{-~
z9Z&@PxXP<4K2wtgZNb8KIK<+;+w~hGPy9;Zx><<v3TqNO=%kI`912lu$t-j<>t*Fp
zxn6J5j_sBS>k#`j@xe=;LFBvX+iUT`>Ea`M`kJ}ma!7f#f0C^+uC_R}*lC7G!ON$1
z1%Bb}V<sx#)#b~YvBi8h>%QjQl08j#lwV$cbH4{7;PyEVCQb{C0^qh0X@wh}gljvm
zX>PwoCrG<8)UdGDGx(XB^X<PQl@Cba{U`T>2FgI6Vz5MqmO_b%K#@YDhmZwy2flsm
zFJoql?@U;H%>$!=K40!H)8pc?qEq0Q^8|c)?{NR26_(&M1u5?S<oJduZ$`1?koxH<
zI_~Gd`*DcrEa&?xS(7l?u`=mY9X9ishP)C_=!>d0kwPWfNdlb8Ft}rAd+v+uQDPlM
zI~Qe$&}@bsCWO^dgeGa|DjF#NOe<&*-;@Fbh1pCXn1sQ=#$wC<;fUM=5*nR2x`r)W
z{mVN}vX?Xgb_g%LWZ)`<5|yw+m_RJ{M?ZK8_JG~^0JOg1G!{;fv~XOMcpd2BpsPfl
zw;XZDklSY~uzCcs6*v^b)Jw!9i;oQIIGM=EM1<TmCh7t=W|;sQQ2_L%`%IQ7?f2X)
zA+8PZswWCm9!*~(|8uvTu`gBhN0gD)H2l{sWl_G-B;hPhux*j{d-~0K<71iywG^_8
z)taK1t?_B#y)xsCi8q|GNvbNl;3oae?{p#h7-0eZ!o9ds|A5O;$PgG-Ftee0k|8$-
zwx3!^ge@*tm(|1QY7{`zfymJx)I&$9Os-_{qS!vqFouM+BUcZtY|PUK*Bf4kuA3}j
z*RMw$cph}RPqZu+i_%5~2B<Q%lqfN9ygTXE<3M1<%;?VYVYefwC0Q8C%O8N{JEC1D
zM+dIMJ_QyukDx9J_zcxAbZk{9D!QmfS`qDBJM%6~k~Y*}>~i36s4dM&>l6x)P>JqV
z@IK9lt~e@t5N|a8fnE3A3BmoXnr7sb>=%L*Fxlw&{_5n4*JK66dQ=cv?VIln#0ZQ&
zXT-KbN}n1!C^}A#2XRk!qQXS>#O50mOg74%O$nb+F8WWsen5Ou#dibPfr^t!G;z~&
z|H0@?p~|XL_&*>IdRn=?M1%;Vk|)wV{J-^|I?;Uyt686b;Orvrko+NfF@F3B1rbXd
zXp6my%2_Rsw|t~6i93r8SM#MmOl{<|*3j`K(R%m=FU+=F%+!<5^#c_YEVD-qsndB)
zLbb|(A3GAiq@-?!vXgKBb4YZQKZ1NVG7&=AXbDP!7Z|mMKgth5=u4%J2jQUST?7b=
z=kg%qjS1_)<3T?8w~iil{}H2p)&WD1FR1H!$AfUES<W_Aew*Mil=MnaD+wzoi^DkF
zi;C2M?+^-MKgQ}2P$M0*1Mr%JCj6D;iER<1NxKGH=MfbN!dWTxVh_KYQ2@Q4Dv)VZ
z7uB^$1mui?cb7|5#-S%g<4}64xV-t3Dcs>-1#`BXQHMqvF=eg|=tkm_ZKAFDM!sqo
zFr~AeQ^=)+j$>OEO1hg}f+!^EE8ft3iE}KD4WG{bd^XWB`SU9<zCJ<n5WRC85wEdB
z>u*Q_CCsaRxMU_04zYW=#cuTsy?G&;G=&7|6;PcrO+^Cyj+YW{BNJD&e?l(be^M2S
zr`7ZyM1+r%FwB-hOhu5)EQtv7h5b<3WRafml$+vcf=f7R__Xwn*V@_B2JGJ^hCnGI
zO;Rwq3Gst7&HCWIX6WE!m6<TqvflBhJ(%_}rt-D}0Jt!Nvzm`7@?ul_4*~7F%e;{I
zHlRznWmq`=se_tg&E#EyDX5wH)krcUBbCNrixm<6z3sd*HdN9SXY8A~=aH+`!ptI|
zE`K~e@<E}UEmeivX(7DDwEbbDb=o9zVj?$v?M&Y-+fgUDw#CNEV=|4X%g*)q>R*Rc
zF0<5v>3Bt$V;(wc#!s+=BvC4H2?lHJNT5__rqmcyl>_4*FrPPC*RlnPyvoHg(@=H~
zzB?;BaXB+-E8!qrq7xZcF-4tiYg~1^Zhn_@K27A9$>u5d@A!|4?CA4Sf05*E`>kS8
zdW9j90lQ-}!crw}LFNpwEQ3ZnD_(uN*kgwLhL}kR$20fDTk?i!$tAgWe0q$3{=iSY
zV$hG-;SkV|PpAoSPd_dae*v6wj~cU6hsXwF<<=OiO7*~|NZx#k<zWc(SXlI%9*15>
z`pt4f<TrOWuLujj&FVmKO6(qJ|L&`ERQQROQp!(935EKCr@{^^RkDS~y2&=ds`O<6
zW9^9(Nk~)6dx~?U{``2f2XZ0m0N~|`b#R+bKr9BFftc?3!C<Jk$-T6_*&XxWSZtX3
zG&uPnq@1aFWD2=}vcusIP`|5;J!lKs-)O?<H&)zgQIht!F`4xDvCF9Rdu3M!2;~u&
zoCH`>6Ijm<H1q*Pul&>3<w7&lNSIQgtr13~Sy&w3#mTemL=qn(Jwl9Od!XwtPJX=m
z3nkP7P$eY~&TM=60F8c63Nyv|3I(15A{N?m7W(1ImNsoYj+}U74_NY&rg$CT^;|OM
z(+&Be6IjC}HD}flv%`aXI$tJiGkR7Gh9mee1eZ9koO(R`k`&L%{XaSxNCriX`7tYN
zkr?W5Ss${^h6#I~Jqjpb2QXu5*h-BgQCK@}8O(X+x%Llln3;dxcTY1b&-}?aYAytt
z>J$!5Fr~$mPBO9}6K#&Yh9LBJcBph5HAG+|3_FQoj_0Hf9KB{@<y2u(h)#;Ra-AvV
z*OXF6zXYQ6UymvEMKdVX5eC+aIZvEo8_lpQ=``|9z>g3Ed3Hdi7qB6F*dj!AKQ)+K
z2ONZ2kX7O6(Ct5(c$%f*%dZjziV}y$od|y~e?8DhdL>d7A$Sjw5I{QUA>Zm3q9C=S
z)bOA_)>Dp)tP5JgiCh_rw8fFxy4=POMLk7^`7jc;`ktn$GYBN}WjD~1@rtkOqDjc3
z=>NfzyA&NBYsfbOnxu#hmt{Cmt`}%~h^k(h8>iNLXm60WJ1rYGu>#0!;WB6I`Orn{
zx(${+Q|&wyjvP?*G+JiA^Is(PCRgbkDQL=rq-<qEc2+yDS;xO8L$VMW-%bY)7(w;c
ztSM~;IurJ{<U5=0rdOgPZ6uNoCo`;wH!JZt+UJ$FwA;G`a%r;KU#@E|$d0TSQXEXz
zws6w7QP$ev<lkbR;E{sQewWE^jmfxVn6=@9^+*~~YOQ_Evq*=|v5Dk-fmAEr1O6M>
zq!5mP0#b6!5k^6?@tZ9z5Vj$~gZsXlf&M4z`L^5*g$BZ$!VmyWo|0vckPCc3<x3&B
z>gFAh4(L<*hTgwI`tmgotFaDBE_z#^4tee~Cv7N>ZusG`N$~a07x+B><1c@5->}~E
zXB+sO#_p&{>^5-Z^2+^%n(S<+{i~pq4~n_&P3=jIa0To<NZOAD3%}{dc(m{}Q6Qc`
z4WeMWJ?=mfSN*Sw?clNk=nY_ijO7vfg0$wib}9BkggO{&#E%_~$;6W&?o~O`75}+<
zt+l_2fi(h~zXJ0KtCYF?Sp26-bjr8Yu)0a9BkM${W>*j*x1SArATRw4I^ge)o_{i7
zLE7k79FfMfv#;|Up-d?1)j0>s6zttJzF5sha<KO@GZ~qP%rvDsfVlv7%3QFN+LQ#x
zE@RmL;=DBF7a>DuOi9vq;CBA)N{48i9+O{%`I*!mOx0`nLtI|QnB%?3pCHLDMNU<;
zdhBmV&UJ^e_QrI0{i-QsAU!3!FM3`%3i8Jb?FebO#7}|rway2C%0k6WB>Vv=5i@eU
zUd$)&n}flg>6qG@5z8&m<^WUR1y8!wzhr^fZ4`%SF|u9x;4s4a)jc6Q4H(v}kt~5z
zw%3K<N?`UYJMmZ;?dUfAHq6}%+h9oQod&zVN9(xFj{#ZfBtPV*p8$RT$3wx#$=h#~
zsbe;&2x?Fm#V+03?T&#L#5%{dWl1C_xX_phQHrZqQcE;Ai2w;;8P0eP@TH()Q$RF4
zH@QtzbbTamx8EU)Uvy?!ACV}q-k|)u-FzgMdwIlJqmE+BoHJY|>RfD<;5XfFDCvR}
zti-r(HVwC4i3;)0l2J&MP9Nr_lG{;!UD13AN!%=11>5x)%0^S;6aa*$Q2YIjQSXXj
z9;Qrj4E8HJiMA+Ekr>Ni>?Cw574A>N(2+)d*km7!@acwlb34V)dO5NmvaVL|1J&e?
zzZYUsu(=V!;_=1e6WHj{ywvhHohGabeJ6N%;jNBm7)c_I%|E0K0}j~lC@a?iEn-at
z+EOjvYZ@l>UPDy?tF1BK1ki!bRtMu}7n^wbyB>H4Cb3lD>J+o1{fS^ZlP~__(1^?$
zvo@|`2!`LpMALj!;g2&mHUJZ9NbG!%QY~~m1D5>U_9QJxRE;nb7DcIa^+k}tZ;?22
zC6(fOoTI5l+gVv<WZwzU2#N&BF8DP4l1h$3QXxsj3J<L;DD|K^5J%7Hj1Ka_J`cWi
z8wJSva`*yZmb^eSGO7qw=_8qMi<^<!?7Gi=H0J6>o!F_;KOfS99f-R~c#~#Y5DnDH
z7B^O<{{wkIhQEi*MEJjE>IRlgchA%ts8?j_N&d$(T?u-FYw=~*G0Su%RF73VJL}sa
zyvV{JjTUY1zow`U0VP-_nQMt5WH-p9T2^nT$;(NbmO*F(v40s9<<}*>8g2wG7E4fT
z&Qin0)JqdmYu8};4f)Pyj;A%jUxj+rcEUDT7Z|F>&=8X;S6sc@v~oqL`;gK%5Oy*E
z1p<*aG>du<?Q($c<46AeT735`-t?nSRSzN5xuO~7r4*uHF)5%@WEo-YBuHmH7}P>&
znsp5EGVrzoyMM+xp-kfVY7$ZPWbZORzw|s!iR_xIX>!|)K?js8!3`K@1?XYdprEgc
zYzuj%_*UAcLa>>gsCeG7=Brw|8a~g{Oh9a5wf<a7SF;yHbpMvF=IB$*E$zg^Ag<h*
zG033#V|u#U#N?XB1Oy*BYHo5@a+0~a5;S!feMz7G<bP%h5-AE<i)pM<6~+{vv7<8B
z`%dThJmota)mkVdP<Q4i3ziPREv(X19F0CsUrF=nz3tE(;{&@7q8U-gn88;neURqU
zd)uKIO0I(cAet2tI&V+Q`Sjj)XvUM|O-ECQL0sv&;||r`PDR*vK87u;m(nkC0A)7&
z$vnEbp?_R{8v62k`5iGhV9dg?OD}#xh@%*wBR@1eZ3^78(QjW@YeYpR9euW#hTi-}
z8*iJoF(qHAa%1X)yt={6LksWni9Ao50OalC*p5k<hdHG-?&p}<ZihF|`OVX?QjUdv
zY+~ejtD&jvg~v0Hvc?R1oY7Fad-?tTsZGJK$baJjPEEBLuvPeNg`YQ0A#w^g(?ZRu
zRw#NNF1pafJf!|UH+4&rFZ20GrfF>Cj(rTjBCGsnBCDL@H?q=W>v%&}lU9<d3kuip
zDq|^~*uKf^+R|f7Y8m>9)W%Y1q@QO;sgoUTw-^x)t#75nlQwyuM8%+Cw>wV8YIuqk
z)PE>4lXI`LHIz~wq1(^8KuF!K^fy;gv00&mt-7{Vb#$V}>8q9oa?|@5W+R0}oN6ZP
zr6TRQzcmx}C*E_3zqLKSy%5iMe2a0d1Pcrgt7rS$DK?{?uN0Lel`z`PeKwvnY5Ts_
z*0@u;UeoYvcY=&x80V+my=^;P5nvDyVt)&5qpNpVVjdZ{3oV9{$en1W6&+|HzOAZh
z=?*q4N@v=fIEN?bft^uxQ;V=~`n6P<^APj>g%;wM8z91)A#-UVE_aN{kZ@+k^)o$L
zj^*?bf#-`4E$Wwv3PXDNE(k#PCN~|~x)|cDaBjv1T5h1PUJmg=%<cFVkKh1Mg@1c_
zgc{lveSUQp<`$pbQge{GsD_-hGb(Q2u0=g$E9O#W=9juxlR`;YXE5tjmem0F`8!-n
zcrK&E>3Gin#@6C${7R=dk``)?!+t$Bf{{!$5N96n22IGRNP|H)d*0E6L7!_V^4zIZ
zIR$H=tsLYit3;4)hWkVmbuwK;Y=80EWxNjQrJHfCzYKET0*n<FmU-Xv#0KgWcJP{y
zHC_iBc%DR+8}9p+m|pMb$bG*Oacc;--1o~j2pfoN_x*C5Yy@?vg%L$?c$->U!R%(+
zJfFOzLr6O#T)r+6u1O1aGs>rFc%m+soRy;-1K9Kq)rJez@vfNpz>}-gntvi*Ke?J*
z$$Z(qWO>5TWY<yeVXeS&UWBM|_3Zndal!57e=)e%e=VSR(0u;`F7Ji1bW30OD1__w
z<>D0r{V}Io9?BD{n;^0O(zoebkDdN^rFuh;OY^B}_E)2B9IYbR&QZW=4(u=%u3p`A
zpm`L+b^7gM<-_+l2WX%F0|e9=x!{)p7Y!2vF*27Snimt7Sb7aAe@k=iCKBHJSLhF1
z0yNJmm#eN7J5|}ujCZN5GLvlPE|N|D{xk>>G<qC8oXJ8yJ_t4aqPzL}h!4LFeE8w%
z_2cQgcTj_6P75oCk6#CYH#q(k90!datPsqA7(TuYKRt6U4PLK!Db4}#viQHlPyO-*
zgu@vEzX|d3#VQfzf4}23;(amAi_ew0k%Goc;+?PY4*2IZi{oW*rm+)UoF^b$FF$|$
z_{5V4-@R)$IHtWdu*AQyiXoRZ{~A}*$csjvbl@K^g*<9AeUJa+Jr%#l)%LVxv}Q&X
zEd{M8)RJ-&uS{y^<iMH2%YlaB59Gcd{b!KJiN%$qV&153e<hx@HwLYrmT%VPvYax=
zDJ5M!P*g5bL5tewyiT@n<RZ;cwv5`qWB#T(j_Y|zi+Mz(c__8gY3<@ofr|2llWnQ<
z>VdQOM2SypiRW2&lPpCtQJzy})kX&`AMu+Hvjd+KjqrcuJ+ErwewqBln_!6W@dNqx
z_VM)l6LmYJ7#W@Wmv2wuh62F=J<0<b$q#@0XZZE$AD0w-4IqF2iH_Zjpi>vhV30vU
zDNEisVVSKIbDmCyjtv}mouX3EZ-^t|e~cZ6+4C6doTDpd>rnQ*W?n(ZU+`t9e2%Bj
z@3X(+z!)iZ3YQYCVf)>MpE7+PE6+(j8l=}tS{aiDkR)dU2ElmJKr>QnAajCIUvO6p
z1bcqGVnEN(h?IZYD!(Tj$^jGB(-laLIOPL`;`^$bO0JmFf<B>x555Im9hOQ;!IbHw
zBI_Tc=$W^@RO&oMP)vgCD1Lu|qGQV0j-p-Gx{?e_3_4r!<J#5&Gw~6F8!Rbs^U$To
z%7xK-JnXHI<kOa`M`EJ)d@1@$YnZj28su2gW1Fy-W4M2Gp7s0qN<`pz8<WaS5uv5F
z5fQ!^p7Zey$}KUQNOMrR+%d~dC&hW~YBU!{o<_SeX*1;wRo5|oUBem}ct<JjYdGZ9
zXwBC=DpL;16<4GY?vB@FwJt`EO4d>Xin(5kUZv)`P1_ZDEi+g;6RHo$`!>-PZ$;ME
zDt<&~(tLk%t0-49vbBLz^d=mo`tc$EO0kttm}~vxF-{rgvHl|zS1Own5vidLA1#6z
z4zmiuO++XBg;WK?GmxTlk+((RL1Jy)`BPnUsju_dq}>y%m7O-0NeiVm*r^fG4fm?L
zZUv2{wNPb+6q~~n!C;Z<Yl@>|6Ir{N$Ez?rW8;4*o$|0&H=|~KaIbdutjVRbP@(Y~
zQkXxNPeoiZB@8_kNnlopd@9NmW`}ZdIfwr*ki^Q|gAXKs`k4>^#+yIl<Pd28j8ngr
zW%%&l;a}m{wL{q9zDL67n-+`54Z?iDwecJ@k)26s@C?tH$4NZMSx9nTZj(jl^P1A#
zJX(K-$b!2|;7O-B5C}T~7y;sGgS1fej?8dSKWY|7E81aa6&IW3Zyc#;_OuH2RI8ce
zaLMRLNj&K1NY-$5#?R@7ie9jd%_)PC0MurJ`I;&Lm#PyxpP~!Orb=er!|AY~VjA41
z;&d9%CEY@gz;wq%-G%Hh-f7*5T*FluBrAW&PDPlDLUw61L3Yy<O^G&ErD;eZV?zn1
zD*<%|!fs69T`j*V+|;07qO$_}rTGbLQxAcDX;ugNwSZsNONIi=h$=iaTC_7CO=`|g
zhX-5jfTrekK(GTqRRgi+5nn^26(HFs^a||MDvib&E0C;8`jE9f8mM(ghLrj{O-O${
z9_X^){R`+^4e27DdzloP#pKC_5^4UPXWot@PZ4B!UrRON@2v<sVMl7ogtPRa2B&Un
zz%BD{aR(T;G0oA1#!c=3ldz-`#l@;!o2|SCgfA%9w+Gf_-c9yBCIYPa3Hz^`>?;n;
zU`^IORJlhL$ucZ;6krFguuka!c7cDx2J4rr2z!v5gZ>1BolxFJ*vaI-0AVL9?jY=x
zzw-z?IJAZ}e*=UatYJdfoA5M3?Gq4ovce4rJEgdfu#+clK-j4jt|RPJs_O_lrFb1-
zrxdplb~5`s!hXQu31O#8-;1zQ<DWv<>CU?l5sYR+-_>&?)c$gWol0{KVW)pmT+pso
zO!&L{7lhhp#M^cSVXntpLh*rk+diRB#91scu_)SK`dFyNBev{}uolL0U?wKJA&%s5
zN?hxptEb>d*t_8A3PM?bUn{94_&4e#0pJ0~iJLn~gzBkI()byDrCblR{2l2lr7jSm
zaHX%L1NQ7G#kIatPV2;Om%x9pb4{jjOaj;OeWA$|j#A2Y-t%^zV%UD#+{vR6)2Vio
z$1>Y;2DaN}7dv5ql{d)G?1Y5|BA8n`VQJ=A&y@G{%%reP3Qf6P+d28X($2oGXC}45
z1#az`g)%$zA8zlNNej&JOFgs0Bp&3<NDKm>)H9RDgk)Iv%=A*vY`K32_W$jqomb3`
z!ZH6ieW{ann8btLk7SL$uamaj*OJmOZRgY04A@KYT;kJu&k9d9Tv94~JvGdsN*Vf8
z`2{$@1FfQIR?H)XSDNv|EJ^WjQC6q{Xi$rtfCLMJt(qJ0_vnN|qp`Fqkw>JS!^HPH
z$yzq^Fx8-C-fJ59+B|<5&z@7aBWPg9N{ZuISeS!WQg{MWxT~m@Djh{$iZ*gTB)r3c
zDJS)5RNbeU^s?Wl*^)+c6HoTNG%3{J0Wk+1)x7z!yBCCFX09-@{iHL`Rn|6>9i_9W
zd6w?`$^$bRJYxcBzE0Z9`M*>z>Esc0C!p;x=V#NLyaUp@kDY&8GQW*XNBTrUGb|~I
zigsKGq+}Yprvl5RQ&_r%MIfDpHqX^-6{Pz(3oseORBl$On$LzS<>x$ugax9M!oyl0
zc{WRE%2iK~<dqP!*wd6;GGk^3e%u%WT_3pVD2zK{7nqiwKDFMrD@Enr&oP9>0ApbO
zam0&|xgAV!vAusiD6pKg^P3Ex_&yX}iTuzauDp)xt;%7;X}qJp^uAMZj)k`4+B;9>
z{(B9`th`$^&S$M0h0@)Wt4d~>IP&BUFL!3Te9o~sGwavo_o$XmMU_)^mIsHnQ+M_F
zqtxTvR9<`oYf+l>O<%!%gs^GT7mqU-1+VseF@@v7&o_VZV!=UA^n3~B4q?3j>Q8-Q
zx`0n!A!q#y-vP0!3Q4~jB~;<<#P3h1Uz8F`7myO+EK{KFq?r05lp?Gk@SF*6yV=Ke
zlRGKf>z+X_?$b>M^tUNeK>r5aBmn<LNfN?(qqa3xZ2V!}Bs}&8-5AI4x>Gl)?Rq0c
zrx{+4>Shya*pkTwm|qI#^b^y?{{W$by0(`A7Y!2uF_+Oj7!(3DHJ6b+0VbF54GaXA
zj)V=Ie?ljj{cmr->|_>!@J-G4^V{yvT`z<qC5#KhxAz@CFq(u8YH(gy<A=ZgJ$&DN
ztrQAnDFXk(|JOB=-ie?*ll=8JIs6L`f5BiK!|=Z}^t~=4ho8egu)sOOLC8c{L*elh
zVL2!fL(F6+5>`hcO}w1MM$ViPqY{r(B`Koqf5|2@3j{Nh&RIJ%GnkjM8nms5AfC;P
zU{*2W8wTJc8xos?fmv8Qu`mS#qp~nW`wVlzNCQK&Y8pdvqyl@D^fa1GBi+nQs@|>@
z@QmIp39sm*6=#K&9>lGHuBze;pFTb*`#Vz7EW7l5K1XBMHpN^m!5PgMDZEpkq#0v`
zf3#IGdXr+zg^IaafHR6wQ3R<!NikXr8R?YC=-V}OwFGB0LqP;%K1nl53DWAq!*r=;
zu9o1OW-P+EyD4y7N@8QI1X*{Q?6Fwc2jvZj#Res8jU*PZ!UP5!9xdZ?grEXJxQFA0
z#SU86#21re4EOsw+Ekc>vBWVcjIoEme{F8h=h&D>73#d2O}3KlaUqJbULY9tqCDnF
z>+&mSL<kG{(~4Do=8okLS;CjfW!Ax{tI}g)ZGnz?oISUDMP=p#Yp<@7^mbB0?Gb9p
ztuUL|vPG#UC58bB4*=H@i6aM0s7C(il6~UqwQH>cMgz(rnjSEEgh6283>Z7Ae_PxP
zPKY&wwGne_XkxQT%J#*aGSddr>c78!3|-IPUnjn`1B294R<gs&c{Cy;j&L_vFW5lG
z;|TXzpA69XH$eHt>_sFIA@nN}t+N+_1&v<t$9AT(ksR`A<L{?z*W7VP6c-9N@Nd#p
zg}Xe1DXH8(H0PMcXDq#4)5;-hf53T&N_rqrA2l%RX*Ezd(Qwu<*|I@8+pC#S?aBdI
zZeSGH#GcwweqRcTX2h(Yd7X>*5HtlvzmQ;VVwX|G@x`um6uSiPB8tfK;*#Lg4FtVV
zWET|O>b59FNHb$U8vDsxkWD6H5y>kln@y(e!P;M5qFnPHN*kd7PWla)e}YuY(=i{U
zj1zIGqgzZ`VVMwGlinCqS52Ek9}>cj>K4<^5+qexIVv?tFzY#`;bP+RwVQBuR=e2%
z=LAwIgG(G&oH}KZc`cND#B(Zu?&!l}{Y6;K3SH|z9QUH+*ba&lJ>!m(*`Icxiv^?=
z<Y^YjO-}8@<P0y0ivaY>fAy4)1)dT`=cAJpkIflb3;UbuPzKOnN7us&>i}QEb#*Ai
zm?XN7uR|FZgp=WxbmC$X8#!YaVdCc$qTsqj)QzV<5A$<@Y8B>xY=P=}$<JsMLBYk(
z$*GMQebO?g-X`+3teufq+XUr5M`Y%m;?*Kw&D$B7l@Z>5Qnk$}e_mvz+O|#Pt64iI
zGQ4BZH#xCONo@4nAZvs9u$l<+-zbbyya&sg2*@C%GIe`G-X6Bh!~TvvT)x~X0XKaL
zWaIwll;NLB$LnH_1WiCem&(DR)CfvF!G^i3tl7^?sU)Q8jcMnZhspVttgpD#<bJ2R
zOPDgdH2NXiA|B|#f4a6VxCFqZ8ezohlNOB`5`kPVAzyN(Rs1?ntZMT#CH6gP^RP6S
zPe-+Ylw0@>MnqBzetlE5p?PQ;_2<_sJ_Xi?srr*_W2&#s_sX2E0R_k%^-}JcwvJhw
z0Z8)zX|vg+AJxgs*&w0wDjd=E!jg>^6c&ce;--7bgn#pff2ZH2m@thfMYOTCq;T#S
zq*3+gOmkFy<|kBgEaw^x@dOuL38HTDv}WG=WGb<k`QD`hxNHtt;YK&d)OTf39g3+?
zHn?qTY^S}=kuA0x4i^d6tQz@PG&nQyV*J(c6YrU~Y6*7DEio_lP+IC~oR?BWy(eo=
zSR-knqFOb_e@lF7n2a<f&hGe$H*wmM;w6$#o_uOqTem9SS5;Mi?XIuJDyM`QLvWim
z>5#I<lhxQQ<$x4F*DC@gGX;Pb&eyEqSZFy@9Vq!_N3FAgTdNqAXM`0`^D2hyt#cdS
z2K;K<Q%nWnm+>tQ_55*sBWv_Fek9*>irTi`CsUOte{y>mWp9!MbQs_&V3L0h82CD*
z(Yu7;EFfI!7No4<@}48|9H67dAFE~e<v%~Z|Lxm9NnqN4c_{1}2`B;oy^qZZJm>jx
z`RCXsNxURDBN3%`20a?VEIO)Fr=D-a&L}TPb7f)b{>j}S8~_#D`RPdcG_<TT^;J%o
zXzeX)fBtMc3v1INNwUfbx0p(v%T3HLv$&Z=ZQ5lhX8k_wz7M;vS+~|H&qIP72}g&&
zHQkax)YsJNw{!)~g&w7gUZRJ69qDc5t)2YBh%-8+KOT6R%cOa=N&C5{k-Kz(R-sV3
z%6H#?L=T;;=$R_1!bTe4YxMY13uWz&8lHZCe|yjK<`t|$=MI_>#=7B7X|9hZKU{K-
zwKO0PFQdCU^%zoefSV^W?K$g+V9BpT!_p|*-6Y)C?Ftc5sSWz*q2r_nzJB3*$D)TL
z=}-{?xu{j{XgU=2me4(p3eDx!Y;F$QqYqQCIzD*mbZDzbKfLIj-E%+z@Q^*P%}ZCj
ze?of&H{CllNl$D*z&Zd|e7FMjE0y;NnJT|MI7+4pR!Xf}K`v-D*W}Vw9du-fmeI7#
z4SRC=0<6bpqbAi-lgAPlF5?oCh!t%$W&Wdyi?#?7wx694WT~Uu0=p{zrrq)S3PWv~
zm&sn5adGy#L2sE9ov=;4y@7wq+ne`)f0oK+isxV=@4b=f-Oz`<_4kDvbKj!w=7vem
z>#>Zp^XSaGd7i`7%tv<$bFZxWs`!4SHn*RBdRSCmfS@mA@`X_$dYEO+WsGaDMQU$0
z_}nu2xj!jFSl?Xa{``hTfl-@~yl><xZ-dbNmFT@NV#h12g5D)uk;G#MG~_-Pf4wUw
zi~(3f{s8_z0#RCr4AJ!zZ}gM-2BDl?C)L`e;i`g;N6scltwzog1;@bo$1oq2VeVan
zb4$pF*NWchwaOon5?ax$E@s9n!Q{rx_BXCJR~w^-B({ws<I7#XJGF_Cyz99@)F;G?
zq;`7U%k-ku*U2t;dF)J8qPvxB2ln>=U&E6-mjM?I69F-o;o2q?0XLVyiytSKZjKG4
ze?Z6k(=YEoJjgr*DT0k(<Bk7%dw=--fe%PoFA#nApRW(`fzcqeSDm488g$(s|LcBz
z_>q$xI+N5!VEmwOfBe?5lAp3ngCv<sYY{7%f+Rtk&e|Y!eALU-1SyeZgCrwE@<o!V
zA(^3yWz}duUpy>QHbH=yQ;t+=whYOWf5az|<Yz8@xl#FheD;N|*IZ^x&va#O5M#Zk
zIH8F#=!+$hmamDyfe}SgCQY*Fg(lf$MpEaIxXYbi`e6wZ1y??jJgX9t=Rg@XQS_X$
zl9hml+7?8Rj@Q;RKKfBa<%_Bs6;i!j&6|>jlm~ck0E@gS>4lSa#+&LVXqbMPfA3%N
zMB*TADXb$VKzA_cVO27fTbU8#gwbx0TkI+1>BqjO_}ySpk%nBY06U`=Wr$0%y&RAw
z4I?<-188RU*HFyzW)v???NN%HfepL`zFa>V6-01;N8eXbHA;|wol%W05~2%U9u*=v
zDl?nAGDkGv?9LQsotl>X^nKH@e<^DO&CpE824Atui797=bF-PhMd!5@hNORz&ezoL
zq4RYQXXremxI^db7WjHx=QsO4r1OmGPMv3SccRWS9XqJ=jK3RnUeSmlR64J+j^~x6
z0vVyL_m1WOET_1l_n)kUyvimUdnu*ru?H9}v;aonun@LTNix;T1i`rLe?f%vBGA0n
z9nQ3J8|5`2o{zrE@DYWP`-rKS!1r`G6PeV4<a#ECEN-r*oTj=p3z&b$^0!>4$w|VG
zu{df_YJ$0m%M+gOTuY1*K6DtAuttA`mIM@mdSX4Es3rw4vK|W7j`CwV(nkKha6E-+
z-i%147YGiDcfeG7>t~c(e|S3^7OTBSVQ^|3zMr(YhV;S5R7nPRgD*{$p)ojx<63r}
zRBFmwQmJcyj8tBmQo$gc9b+>+qM;JmrCyVzW(hgnuR7@?NvJAWaIvhH=ttDDd1i)_
z<G}2)WLY)RYO}PC<X>u1lbL7YFwG%Oc`hYu=9yPzJ_gseth(1#e@V4WSxY6f)U#eM
zr;SU;1O%wXalPJeT!d}3_Oo8&JmR9==%}b>F8^h9HePr+25EM*hU8D<T4WCxx*@0q
zoO;X9y%ykRyzDY`U{Ff6#*kDT9yiv`(E%)Bw1O*dKsZ_InsA620BZ`R!oC#?ov@^*
zjKW#-mH`-}PI71le?B?*Vl7hsG3F%?1VEr--jES&WrM&81vq2a_G!?RK#dpX-?2Qa
zPE4vM7IP9dmD!w>HT_-tq%yHMK31X6xE4~W>T6FX<!DovwVCBq8Y(Ni5BFL&3g)VN
zAqSDw83sqMbB3}kjSTDNTB*cVzY3R8RL>=;#%m<3?z0}+f2`R!E?Cm6xUC?7kTV~{
zc}52$Bt6aUjuFLmHdwq{MermO9KdZwe9Z!j2~-VkF~FRGs0p?VP#dAN;_RMgf5rHc
z9hSY2vTu?S&;9<%qRo*ggM)k9T;0nD+T1O+MurpSyq{)kCdB-XS~fO@jkdEi?V!7#
zXhQ%(qM2rGe~M6wc_(b_mf=U9NJix>dsax%i<$AbV$j+P<7cC~#i9{nK8_J=S#(bN
zafa_>(&;DkBuaA~H#1?;q>WQT2E4OLvtE`=S`9H&M2Ka}Pd+CtD^K1=fQrvk!={C%
z#AA`;mt+;C<aB2No6e%pfiHPqa@lPVG$gP$6VG{Ge}iTjw5>K^VN60DN%&>LNK@_H
z4S1VTI3pR7qaH>X8ZlCcsUbh(<Mvcl!4HhGg5#SF#pSV)nd1~Q%3u`b+szvyGah@j
zi?$45b#vV4KyvfbXmy$|K9E;u6lWRJE4DLo9LFi@-srL5-AfK|S5GcX-AiM|2*P3D
zy`|*Se}Tn{ybCJDsWt3Ec^bP0Ui-*qvkwAiBD-<ltz@M*RkIAsq!oxW)#90(qJYZN
zs($Z~lJz3<vY*t(O#;8?^YVu|{@^|D+!HLWg0+6BReV^q_Df^{MqDE6M!{+~=8~N7
z8L?XlO)Gj0pf|XkY29S{KD&u>hQ{sPx4aRbf9_y<n1pFv#dfpQE?>5`l{Rb*q&({|
z#zhUf9rlpfYXc`{CK$yN&`QbkG0JQO<_9P^A19!dV%|07ThK~D0yRaj$0=rYUKQ1K
z1hX$;Eba#@n!I^DsCgb(VH8gQE1cqe!3yi`MzF%Ey{&>3R{J<uVT4WsE1arb1uJ~$
zf6EJ;_C^?`?+jL0FCPe2c+b~?72f(huria#x>2y&jk$DM7>g&u6-IFbu&~;HCuG^j
z_%IoD?SD9I(UyFlJT+UBk7sETE*iH%xe0;_m6+uA)fTBK`7|}j>$=H#R-EMAd`pFr
zl*c0#Cv3GaRyX!qSnZ=~q?-Hdy&Aohe|HpS`{Jh@Xsgt5Hg_Z#tH#)N<8*)8>Ude*
zwV(9>J|dDktw+UER-H&}yr<AZ;UeK(`QN|iWKY2_^{{rXbshy>yzxu>v=$cayMFo8
zah5gj>%_A_`Bcs!NtE5UD<tuHToN(OtY@mYa(pOenvI<`xP93umM5KXFb0J<fBI!C
zJ9Th;4au*GsbC4xsH1C0p+W$~YxD81ge|{92Xu$YeeoSVAhcm=nPBheQN&=ZDX-`i
z3Zsa9>dw>mh1dQ}^k*TG{P`Ez{g)p8K$YVgWq+osUx#jF_y6wS@g2Yo!k&bVX|K~M
zSR*VuB?7^Xp6o~hbr7_?Tu)*re`lnv&Z>2?q$L44-SiG)ORXYm)WkC+-Y0vj_dME}
z-~ELCKDHQM|Iib?8--G7*C6i0a@mkKSm7!7JXzimcspX6+(dxMUc&B^8E}`A89QZt
zSSIVThCp78mg04jA(IEs>0Y7ZGSI>^rgN6#@Dhnx1}F1NW_z(LzU3Z5e}Ad$s0*%J
zc)ccXyiWm{Vl$h13KLN4dwWrN9?Jm}T{iJeZvWPw6--*C23H@+0u=+@<9XD0<-F7E
zd_gwt8o7F*98I1V;Hu4nbH~%4%h(*!d_2j~n5jz{+{)*RA+@Gh$D!xSVn}<ZgZkHZ
z-{vM}xrw4Cwr=vxw>4hae=w{Y3}D@`2y6vq@3UycJgEZnfs2MYYtc9^mkoKlcvjfm
zXVCy>N$&VhTQrW#WJA_g5!Ci~(rrt@;l|ticaD-z)%U})+L!|*`sx<6d@Y52@d2P0
zUL6S!cpQ>+g&e-82-#V^^rtGIR3cq;#n-1iW5&kVQ7iJ9&8@#`e*tM|$||@L^+<zD
zao-cz$5!lDYLypFrtqf!s}=$P`5RW;K3V<9IUf4l$gY(n;kAF^VvW$s@aQnc<^ezC
z4_rJ591ldC`2Q`8@J6hTOmBu*eVQX4_etS#J1O5tH99Fq&xRwW8cvSN!~f?pCEeOJ
z$tM`6RfnSmzUQLP8lN6M!dP)(8t-`Zo4nJitEmotCtdyv)e^y%QNc750W+6CT`U!s
z`9?Y@mui^}IRyX$An<{gsF@9Ee?d2ee*W$Crzak!z?swjtN+M<dwG5O&r?1j;>U!w
z%|E_8`3Il`Q&u)gNM@yJ{`^<-`_tcZwyC|WAbWAG5l8lt#GXv3wPVUZ>g=UNr!4~_
z>wo<DN0Z(8`Nu&~rAZ@^bCNup_Xdst%(|SR(B7aUrp08?GDQ{$BY+%<f8=l=N0K)<
zklXOm{qCQE|9<Ugkht^v^C!*4+(<{C<Y=G}K^vaw&E}HIGO#AsH8WRF4VpQk%BT@L
zU68^QNKbz4LYCjsS<F+lvI^^gVW3Peg9uS}BU_8zw}%EQ(%R;nXXYxjV2+!dS`~V9
zGw}=1FIT@P<;!+IaA1HUf0ri2v>&)-hU-E%9~Uab$}Pl-|NJ4u2lm$HOdC={s75dz
zbQ|6Ic_N-^X@Rl5-P90;t#5B3BC)dy*crQOCUn8g+YqO0JH(HVmpzsZwi?S5usb10
zN_MVf_`|fTgke%(PVEDqY?Vw;1{0z<<|K&>g-e|Tx-+Zi>=R3ye_+C5j1tjIJIeq`
z6gk%lXX!32c`8r6HTkKRaG!5?xKqqHD0H6okK&Rku5#)w*fJfs7l!^v^v388>G1jP
z_*8vUQZY1`O92-=gCpJMs-yxE<_cZmc3eMjOmJCp_R+C<q0H(}Db<6BOv-_!sC+UZ
zPC3UJUmMPpDI!Cqe=BZ-Y^!J?nnL#d5ZX`<Sven=1smWC2QNk($sht4>QR=LA*+Wv
z0y?QNZa&kKxk2cR;d<`yCAKXp*LYXfW|T=#bSRrBt{I}qCovOX0)5nH#AoLck8qXE
zgltVE9t&Zv=^2?MuJMfVxrxM=C7GMxNFFhkS*vn}GhU7he}6>W1Kt@p%N3(Q?hT@c
zNNulJkNy3ZE}9D(AJJwz?DZK1RIWjhqxSo}2d~*2)77YpG*Kflw<0>pG|eRqdK+4b
zG%3j<+-5t^07o;<ZHBtd_+iRdvzk1&a7cutrNi);{2-7*88}_riB>Re;VZ;VfVr6h
zXNHXw40JKYe+W;z1YYOFU@Iy(Mq-+Gi>k1_MUEIa<386m`s82=U26PRR<G?o1IY{&
zy3Vzcp`FjD`88q(7rjXjJM56&S#m83PmwY9n^kQvx+<?q4UAw<sovT;%Dcvym4%KO
zJ(Q^EjRa?4z$QKWV;MM;IJe0XvS~(rdPj9&;abNKe+mO`L?8iJ<y=mHAOVFHoX^D|
zz9bbsJ%T}$WZV{JodY2O1!mO!j_HwJkd_CSDf3k>T!n_|GJq%3@gZ<dT$mu)`N9&r
zodDc2ttZ^tEi7Ng#q?GQPy?oIG3%M!1B5sEu(_#*EwpYoCYVPg+1RMAZ)<<FlNj9@
zRV>S#e^0+-DLzfeX>1Ua<t520*VuV&$c$k`0TNnGTXSQod%UxhYr<|$J{MD0LY!q3
zwZ-V`mRTj;biEH^gH`xodNqd?lMzqHj-tc^{Z3+wIAu}`FwA8S(g|@hmN1KS{N^(h
z`i$z-?v3z_YmrwERIh+*#{xE1FDV^!#ui+Te^;;gMyB9=YW0c$ZN<tP{Dw%gE23d;
zSKKE_EMobD>Ls;hbWNGB7IsoQLYfqDeSZ0hyTbW;6UQ+`?$_1w6?cHMk^Fc-$FwCi
zv)1i!i15jgszlD=aezI(Gs+|$a7=s+lyO|x*p2TWk<hGDX&;&HZtThKBZft(-mqaD
ze+$5vq3T*QjD5&Gv}p3so2WixjD&>(Q%sSHh%;iYbvk)Nw)gE=>eh;)fJr3_E+<r!
zxG9ixKDnZV0x*_xsH6m!XB=lrO580NQ=d^%jASU~l>Mqp3NpHixO_xO>AKRfphFcU
z^qH9x{8=L5&nqoglkuoMZ`lGe&cxqme_?5S$(P&uJeV9`x#c?Tf9VGvk_x-BcJj&q
z&PTkQ>lPxFRg5D%s?`izTbY!$)r64v%^(df=QV}_f#;IVrfi#!zK)C3j>a$M!NMYo
zWi6+k@g-#<?l3)z)E@r3?cUVRIL>So$GVzP$GQrboHa7FuXuNA9y*sg#d~LDe}=R1
zja0W7WCRFi>-O?wupGw9K>ikL*m{dEmdxQ7Xf!J(j9KBlc1#nLBg8Xm`5~CEHCZrM
z5ASXRB#ilYI!4-;Wy3rFBViKL!&ue1IEm?COuZB*a^;N<zZS0|PJ5>~inqc!k~&}<
zPBN_=3T%&|%`Mpdr$~T6@u(+gf9-e;`dp^dOgNx98hiMS=YiE2=fVjCffyZk4N-s*
zN{IbSQ*dS}_4|`Q{w(#*r5U*x#X#eTOW5sfJ**N7$)JA{;eD5uF6W`=?2q<MHf}NK
zgW;#nfD8H*$>-!Db3*b;LVm_v`VvXe5zd%L4ui>UE6&pQiF51V7kJC_f7jo5^G|yC
z6A}Ap`)^D7rI*8-e>cDQL+mSzEjht6rWF8si7{^o-BDKW*bXd^l8xV--pry>GewbR
z%`#hx8izUGrGxsyS6@ZUX{j1a|FGfz@>(H$>-J4T3*t}WlLT=KpMIn38ty~5hvB;k
zbTK&$1{_6V@0#5<T=tE$f11oFm-&LCR!--9H~2`(_WWJ+(KH^9h)ic`6q1s+x@AMx
zOOW8YONPV~`;Cd?eWoi2>8W*@b|o<-xg=I>=W}@<4)?VkbKn*KzM$2{h0dl$BwgzA
z?sr`n_`k5w*@jTNPs*dB{G~s87@uQ&E~lUCmFFgVKCa(Xw@stSe|vx?KRoE8dQr8@
zis<Xg^V`bvu;Nx}pn?Wh<^mO_gb8y_(YOklO9jm}(|O8uoKUE<h3hai-F#%NMi*9?
z1=CPjpm7y6*9x6)>noz0dz%6#jMg3|Q7-Li&&>|)%3n_qWp?nHA=<nxG-`Ch^Yh*B
zsZx_+(B|nEs2-#XfBCO}`z{QD=^>P<QICiG<+~~=NzBzsNr%))Nu5+FWi+IdGF2NV
z%kF{lxLkG(GU?P~Adxhdp?&X&Xuk<pT%)?se<&i7+AuBU2O@e?h)AMt&t$z-M1&PI
zx&je>6|}Pe57as$l7P%`m_$TafiAuZ+L0A#N9;vA3YY<We}8BvU8J=LAsteQLy~x^
zm68srl#)29R!TZ#E85v#`j$(o>~P*=(2i8hSp7}Fod|Dih~Uf6iuMo)Gsx4@RzjZL
zc3k$4FkSpV7V?Du%<%zv-W25F{;w$Q$0pF*+XNz6%yIW_Hh~U99$_4F?-B7NYqS<6
zq(kbkNE%L+e^N$6Dy0lQsuhxq*ob!y)V<|;syn3j5WFL_WzLCj2Jc(~j)~hjam&Yr
z?=E1UTcC5p|NdI_&u%-e_(yip|4_Umkc6p^EuOc9cLb124Se98H;8wX{}(*?fp>a?
zNMs`UzaZ}h@T7~h7A2%ZDxpXcPqk9gA(c`RC)G+xe}`;^Jo`)Ea!C~()O!r%5x|5q
z-vs1|TR9Qh*!b^ElkJ~)-whc-Tq+^YZaXgfM|a`>Sj58>lgfS|o;QVfoTQ)b1Mr*#
zc<kArXddub!G``~5iKpgG?U*LBzoi-AO9y=TD(6*lq}*(5Rs0kLn=eKRf<Rk)JaIZ
zQz;@H5U~*n?XHN$dGGA=e`*2ThnE2t4HE$|mqFJT6qmE44bcI6mxZMbs(+x1c&g|8
zoUc=U_ng~d(orSAnO@KYAtBr#!yVmdEws^>`GOJN*p}}B9ib<VcGMBe+vSV3L9&DA
zq9c<vF=~-DfJw8c6E)hT6@DJ+GMD)5CHD~0#cK&k>Sz+Zd>wm|m3~iYZAB`VR$}4E
zx{p1D%2s^Rxz08gM`Y6~XMbLiugaqdX(l{+fTw7PE}IA4sa3q>DGk)TBohHy1}a~Q
zxsC6!yw~gR?aq7Gyk{dE)uz4BZm{447+175HuXD0d*12#wDbPF^FFV*T^i`1!JWB4
zhbUo1oNF|pgXXz|=G2UNO?Ac!aDJgWtVOqNS-aMS9b&=SR4&kj4u6{GjxnG2S45h7
zbOCEZ(<o4dBDDK^hq&_h9mLW*uxE&KbUD!Q*v@>#d3^s06evV+4t?u>f#N~EkpKF(
z?{X7ZZ=$J*oj3W*cU`?yHrLxr-K4XZN=d!F)J;+^Yq4=X>|P%pFNa-@jLy6hBvK&I
z{vi-iK|kkw_Zb1%3V#LR<U=E!>qJCXi}12)&?eJ=UqmF0AaQ*|M1Lwogh2}n=1+@=
zu!F{SAfn#|?OcEdW*-rur{vtjDk8!Tbn!LOPS2FjJ7eBOJG9<})A&b5JN7EHQ!mm^
zgfMH;heK+6>g}a&(%DO;q~2cYCI`{Z4WaLHNX3Nn-UaO-27lq~2zQ``J6IViSs&;X
zni(XTpiBv<5&HN(SaSVb_he`U_;8u87Qq*`YePH8VsB)nWN3uG$hA}vWy7~EBSvkL
zS$l!OS+iDTMP-GZaMLM!oZ^+v6B6Rzz*SM;nkZ(h6djMBC?@=roy+AKF9dR)8fx5z
zifgp)F`9_7*?$Q~$0m~xO(s^TFnh=v=cO7iA))fCXI_b0oY&I_Hr(~OjicLH7)oUW
z4>l^BZl9X}Rx8GutV_}BredrqgOqNp7&mN3!;;iSApfCXzx_HiPd;a(2y)#8kWC^C
z;E>Stn7MOoigHjngPF^tc8QzL1!z*R<m|V!n;pxp-hUEZ*4nnxQWVS)Z;IR`ak5LW
zAnX*maSE@TKSTyH>k?Rw<xdpv?8)a(6LVU{yk_Ujm1?#S3b7+1(-33E0?D1qgg#Ek
zN8*i8F0><o^^tfES<nz~36ADy!?;6Z)bX(yOV+JgPFg<A1MLBh0D>HD!!Y5u^%<AR
zfELb-u77Y$;&Zwl&FncVgH7)_nP-e?wodw1TZYM5pf9lQu(-7G6meWT)<#?-HCPeI
zKQz?X7&xyC-Xx2daGb?9)w(H=n{sYS<zx;IrLggO{Q{bE-k7VhN(AXzYO@$enh9_D
zF}&t&K>c!D4-yC*s<yX{)vXC~IB6GSt5Sk!Gk-GQKTc^umbMhu(IAu=)ach|Pd=~M
zvnFS*rqF68D=e-wS|pAEbzQDjD7)YmD+z0|o$M!OPFF}a%WB?V<E;*HBp@G_3_q@`
zN;#nww$yC+IrS@`7AM4Ui;`rZh-+&lL01)OU0IHtwo@(F4rNt!Nv1yP{5VC1ltLd?
zcz-a$1z3!ET|Y)@VWerPOPW3#!#)(+!a+tGfezbVA`W6nU>D{wK%IrfSmAv8_!vz|
zz2v;WV?CPK96!0#h|%Z0+^`f4I};Ta95-}LnYVEZ%P{qdJJ2_yAj-wtzF%)b8z!Jw
z&IdNZ25{hzixxL{5DJlmciha)>+IHZAb*)*_OR{hN#3Ar%W>a(jJ2*)d8&JT$|@|n
z$)6Z$1)^KbV#CH1vf;PX_x2}@a*c0$IqFXs5aD91_wXlqr7t+&*Pn27(hDiv;!hlL
z@@BztL)X~@G}I&XT?e*?5~5)I9m-*c0WkX(FzpBy-s*b27Z}^lv0jyS;V*i6Tz^Q>
zQ7Yo@T2aWrwBZ~*)p<h?Yr5JTN!mrrm4ny>a}!sU_F-u_DFNkJH*H9MMYG9xci#U!
zjYca0ww>TwX?M3KlmvzBf>rH$7op*Jrd*(K!+k4z%q5bBx<_lv5O!Q+oS(yDtXtV+
z&gvWLu`ooc$;#B2Aw>P;l4g;20DoG>u^z?~3td{(76hT6<?!|vQJXw&EGxPxIj?2-
zDIq#yA-Q7EaNMac7(<o<v-(0jLVUj7LgyT}Dufj30xM@uT|gF#b*vw_UeI>s$cvqH
z4P28QpoErMcT{%G&K|?urc(8H->gP}3e~=84Z{vd0;Ct#nodywduHWZ>VI-7W9jwQ
zfk|EnS_@Tez?({rWpDYl$SF_EnUqV!KXGESC!bf$ZQcvpRL>q!RaMa2`zTeOP^m5L
z=Tw`A^S~8H994n!9LkeRC^y#*+BvyC^j-3bruC_ILRBW$FKfdXX*n!h2eC#ANmMQM
z?Mc80)p#FxI>-P89QBE5<bNV5+E~|aWkf|+=M=``);aqyKP8FlJ}fTtX%XMz#`qfy
z=i8xf`7s0W<YGhJ<Hz(4j7VEkGe>nd6g#U>wo1}I(2AIjHSZLp)Be*&V7Z?GYVth>
z!ZO59QeI@G2Qrep$4NRACR}Ufg$bYlr+?q=%=N9{uCJ36Qv3?8-G6IFa&_Jx{WvLa
zUsI5i!CiBF;qrkE-Wdq;rUM&?S@exe1xlaseNN+ON%t!G|7yW^eh4{!c*h^x`IHdE
z59{(_RQ|2A<;G_^@2&aiJwJ<^D9Nh%oQKYyF|w_`=hL0%m9Liijxs%`SL%JVsR9L$
zsaWxUqRq&HHTN;!?SB>8yjtu#>U37fy$=qqR;Q!*slRD+sm-gUzN1Zx@etJeXwwqz
zs7giO)H$m2YO(LBlam>|C8Qr_(T8HV+~Y5^z2hNFr7@m5D3&)roU(S}GE7HA7z`F9
z9s#GH4(*iR=n9w1*+kcwO+}fopT6t6@0rb&gE?E{VQ56NY;MORpLbdrp`~qGVRHhs
zmGhG|j|XJ^t|vq-g-RP12Sn*5;i0c_KGahxEa4rFhc=9N(y6T#0eVeKQ+?1^C9;kL
z`@+dlN45m`4u{_6jCX3%bK-YC@;2{m=fqnxvLgrHR?WA$)9UyiAnw2#mjM?I69O?d
zmm!)L69GAwVHrXvmt3z6G68OvkFO0s1HXlDm(i~cFMlSy{Nt}5KfO?#0wqrSfBjeg
z>;2>9-!J`ygwlj_H~sDVi@zZSBf|1Ya6ydK)9?S7etY?AFE$UR4P0YlVuaG^ci*77
z@(pcj8Q7F+tyE)E4l00|wRS}KTWC|5m2#9yn6-jg%5(f*`v1XxvYm03rZr)DkVe5f
z(2**zsed%Sca%aeQ3{dJJ^u+)Xs&!KBY*_LoaskkqVvEl5-l-Q4-!2}=<F;xXGCh(
z_fV(XXTu$F#%>t<CaIDWfchZSqm>P4&ApT&ips%v$wx3@>hVMgR|vEE3XK2d;^-X4
zxRwMwG++!cr9dBL8hw_K%8e$*44$Xwq~Fyr7=JLYED;b)55jya?HQPFM|cG0JBb@G
z-}X@F{g}Tx_AbnKQa570^W6h6-(hSg<~#lFz`Qg>3e#YoBgVfXO10N7;y=X00`3j?
zfazZ13hrMu_4U_R(GX*&bd?7r<#`P{3r!>-21f|7==*<-y(Eu1hgiL&SoNddj7?@N
zaev_0Zo~nX0RW0dC<KYkl@(kt>#Ze+Ik}?0(u2*nj5t`01Gbr~M^HARukkSV*EXRs
zXS!B$@b;1)6#n+t-?7c2CPxnsf7yd5bWzyFpv&>Q2a1%$LT+M`DE%2Qaf&!8x&a@E
z5>6$Nigg3{+55uejn@HW(Pd~EDhPyG7JsC0>N29O#=uR{A|c<fUiv}>VO9`k23<VI
z_Ke)~ZvMW%Hnvh3o6q;ju5D}ihLI-_CMhm-{5@~4HJAn@OL_ZroFxz-8>b^ECxVK>
zy{1yxE=7xSoM-@X>Ld#5i4_WM66T2$F(Ou%@Fz^9xiXVUyHR|y;PXL<)gfdDg?~1t
zxjM`v7Vi#`7qklIRkfb!%39<ZP88>Z-nZ53zV~SbL+6`q3kz3b*{GwM2MJSN(V=x3
z_l{bIK(eqLw=xHl&$Vt_c2b6D@JA=%QRxXwZ6}?|Ws-}@r0a2SieiRvrh2JMDGG`9
ziVi;2gbfVbrVwv<-*0-b``xM;aeox700hr))q5pm7mGp4_R|yuk8;H5QnV8<Ft?O9
zB`vYkl!Z8^?M$k6l(tB2JIPC>EhW3Q^%gRuYT}pj^-xx@oKyrTv+2O1=M%gDA}DQH
zsy)sst>kM~X<L7eRn8C1a9k5>-8~tumGB5(Af4(Yvjtx`pH1wb$NiC#vwvGFnX~;%
zd01W;E{B=P{St*FT<~5ZeZPCVVbC#R)z@|A7PgBpyFyuZq7I7PEF8nE5jaqy8@38k
z-E0Ml7<wTb&ltL<EUOZqQMw|avgVTuJ1~&{mForzglWgtQX+4Su!WG|P+*~4c8&LJ
z;#x;kZ>~GHvl6rKgvy0wfPW@R^^od|Lc$#^ft+E_qW!os?nVi;Um{>k8BM-Dj=1Bt
zVn6am;+`@@*Zb6hdk0xhTa1H6*2AG|J?~5D{3c^w_l3t(fgA2#%6AeMIBH5cUF#6~
z7SjYpFPp7ff;%<mijh&3C320<=vwn`tB7yPXB*c4-gEUlXR&lQJ%5ov1*noWhc5ew
zx6zdZgI3}L=t^2zb&|Uh+&`b_^>_eX!9edM-$PfB3Um^g-Y*d$y=to41jek=tGJ!5
zNT{sTsRz=PPT~XjN-y!bgrytpPR7znyFyvI*84e2KkQ>jOD}buwe*HuD0T47gnlw@
z=|=fP+R~4F8*S-(KYxz43>H}rimvs7FBObt%p++_EAbj_X<PqKY0D<XvoqS(|L?R#
zIKr8|vh}y;(Wm4VjO+GpDkdq+5K)69?S2NE;kHw@@wrqRWy+4#2%uo&^4$C0&E=E<
zYGs#TBrK8Ev~Bs7?Kr;r+&E6MMa$!!2RDct+ynJP$g+^-Sbr5#?-a>+s=;sjFGlnO
zFpXLXs3iXE-jg=X{gWT4+rmmBx!4QvBt--HpzYrK3878#X=r*r$IGLbP-h*suTzM{
zK0%w9N=#8snQf4ur^+F7CmtDmv+8^HNTrVp3v=s^RDvGCaCoFLi739XVLS`}&m7Yf
zz*GQnAoG`B>3{ThxcL(_7S2xp2~EG}ap?4~>1Tgt^NL^#-T}douPpF{6~T1k#6i-G
zVv7g`Nz^|uzFEXZ%^+1+C)dRSoSf?AuAMn$zV->w8Q4OJm8U!x#{bgtrkZR6n{3A2
z^Sxq6NLEg$C)p7?axOl_j&B?7xO=|qcI2L@8~q$Ra(@SeM9WXHqugf4{qwzIM<C9S
z#IK|apzFZTI=MI4(^^;$mIe9zN-A+_-IWGBg}as1s&^d0#Sh17mR5>0=MTLer^w(F
zbKv!QGSxR%6(}3elXApE&`8I34rHpG47G7sD+67OXWNWVMeu=d=oLP=bAZaR0;N+v
zh+0Sn@PAqZ&IA4?b-jLi?0kZ**QYej$Ir#@DupAr0*PU)<zFuTaT$QwDBFWY0ut*<
z{rN9HU>7;xg+mwJcKOo}EvdQ|So;HDJ+QPO>`CcK1L3vikCf*3&i0B86(>wS%Zjj;
zhV|heD#`Dj?-e`B$giGbM_9zm&jF*_XvW>sy?<gxST-|v7dyXM#I^+q7lycXwz{++
z=~q6%<B(<@1PAwP!tu2`{|LubuM=VEh%dn%U;Row)p1+ub=pnKZxi5eU-E>shf-kj
zfuD2LQ>Rv;<har&|97d_L_(9wy+lEtKjoX4gQ8fJ*55Q_#u3IdBg`c+X%^lB%ximt
z7=Nzc9lH2d#<zM)NKgI3sbgC-&$^isKRAM|EK==16SyI)_qftlhqqHGk2?sncbcbG
zj98}o8S+czu^FAb!^3+bm-!tgkVXN-oz%1+b&tMQLM`TlzRBiqXqk$fMEjXTVMfws
zi>7=N=6K|3mov67hrw48@e-2afEQt5-hU9^!@L`7Eul?PMT}u>ZzfKB^}RE{i)Wya
z&D(lFvIc)^v4~{83CY}reWGKu>Z#fzs@Zs!=SPU%NqwW_Jf6*OyHxklQxP#-S)aX#
z^hr>prz^wNZ$B>j28tCilJ^#<&-<&*IjemiK#620WcS_n+q}Dl-_3WULhCQ+)PFAG
z*%i8iudPGXw=jn-Q-jQXs8VtqrIJ}DiLHuzF<Hp>qfm!5s;%6ZU!t({pp=d{YU}qF
z@`}SWftdBlAuY<)w4JMLS=5x^54-Z`I$5`=ATAUmq;v0_w#DqG=h4v^YOhc&LpVh%
zw#C65hEhR`O;kzmv9y&`WzpmDihr$+a^yZS*xFxdaXfF8I;f*P@{BJP2}^is%Nx%l
zEtOkl=e*LC+-g&v#d}N1O88e5O#9Mg<4;wP8?<n7fsYoug@Uk66ofStgk=hHw@v18
zl3qK`CCW6KriUlhCQ2!Ep%i9+DOpZXtC@!@{u=u_|G1QCSbOu}6)KuDu767pSix^*
z?BO4Am6u7QFS#yuPrqg#;b7^(7Iie*nMXywrRhrmjzDq057#cmUdj-stbrc1s4_&+
z>&gqVm9)^>LitWxEa&zai1%{TC~8un@2!!Aaaw*Qq+`j+8Ify)c*M3rH_#vx%PO14
zlP#O<*~TXN)+u0k<*b>iOv2G+whw=rd8Yy}_$u&>CO|0LSGiYl*LIkLoshhOUHr8M
zJ8ZE5c0#(I@x;Cz>`((eXi*m?nb*fD$-SX$6OU^(urals0dA&|lUGhWublYm%1InH
zACao(GXrJ3n@UdpwFG4$s~0DG_i9N*A-Y>oUh2UL%B#9ELK)dnuID$vIeiGff4uw`
zU5G%JAzwWc12Zu<mtoc}6PF&l4atA-CbO5{zJ7V3QG}LXr6obGGQaPXf{E0`<&|5B
zcLBcwT)5*9aBi@aO5{?aG!jRHaw7#kl`_2}ZJ-1wi9{4Yhj;w@c#nPtz6b|n_(X}s
z%LowDU&!h5?vL+?y<uzEb83jCyp=_M2F|6MQAJK!{Qp*PWasa2i{l}Vcaneeix(W+
zkP6H#We~`ayVWp+A|p6%ar{)=eFMknHB1&5&Y)*tJ)9~+M<Y`v3SjbhJyQze<|j@-
z>F8fxS)`ms%{)m+R1;sB;;rPw^KOz15vtofL*&<7n0BMDiO;AC+S76<E>ux*LUGX}
zT6uOEcoKnLL@pSFM<ptzM^JyxsKoI!ahXP?CtfNR6*Y`tD!!5FB=s=m{9aNB>y^|$
z^5bb%W3TCL^KEUeJeIqsnK$VX+ERGIC1KpQh;dzTj-TAd)sayu6RB=!tx`m3(Naa{
z6y5?e83I--vF7-c0K&OQZ;51ZNn6D3+?fhf#PDn}qS3&BVow7jqJ@7Qc&%6cb@40u
z8L7S%I`Wq17I7X&`Qs8-%Z-?AioHkIW|eak?&Ev&?c?j^&xHYi(gc{A{`vF8J&=MC
zf$Ja);w7iQ{yqKv^7oc%^hA_Ue1(Afga3IKTHW~(qnD)!KUE5#f}c@82Us9F2d|9a
zj6&`Gyml^$359&KhFE_SSpq<Prn?*)Ni^hHJgdn?@(riG!M3CrxuK=Vqx!jc>;4z&
zdAw^uwB^s_y{6cA2-z90t5wfj0q3<`W(j0vW^XC11p}N-a~@|_!c4rYW>*}{G8^k>
zp_0kzLusI8uHp={lzG%RC6mOqrpXu(wAy{!^mezgZb=iEC5C^hY<MwfrTj{!TuD0R
zN`Ji0Rd;?A`CT%?nlh*B*`Sof*mkg3uz@$OJ0@`ldlu7VW7-*GtXZ&!p2SMA;E%`U
z*0b^W&-bX>?#3<hhO0yg6BEnkTHPsmON)Y8oRM>CEaptAia0qF*nrj&gtw*UaD+8{
zQOy<g%6lD0wc&r9Ef)SBs+>w9(IGZenOK$ACF2^`sFino9i&AIMD2XeIyBUF*5PjS
zjlitmwlg$1jjU%&@pIFV(^@re!0m2nIOEQZ^{2jzD94n5FBXBRbz!(ARxiYVM3syp
zRy<3SDq;r=`hD1->!}`Hgrp-iQMqf*jj#_afkt0mq=kPm+3}c1zg?9J?Djf<Ai|oK
zVt*k7ofC)wJI|2R^I{H08LWfIG~o=|HwHX>HlRl4k8y1fl5oKn(!jQdmz-$H7Sas(
z7+{1iU|}1s4J_v6b}{YyW9q1YP{vDj%%zrCEArdSm_a7BY4Y+}B{fkXTB?1ObaOfd
zdPE)TIX8dme#x!ZgNHPc5vlA#)CV*%x%)<t>$)IGZyQMlv@q#&qY4jZO@Lbix==9h
zCB(LEDMRpDp!$G<gCv+GKU=|Tss|Mu9#imoPWHB|;Pv1~6uc4j2@2i_vJ*IjA=2oD
zG$Trm@q3BO!%7}~@|r~3O0GEp&KJl&pyW~zYUO{kmAs~UP{~!g!w2;)B@4~$9~(Jx
zObF`HBN~z%YE9|UWHaeqLv=?xlH+XTztv5HwDN+ZVnCL=Qxc$ZXB{_YKtI6rT$@pI
zsW>-bioRCrHu-0?XeU7EF4ePNa_0icv5&k}uyX~&_hoVmYuxZ_F&0ZYUe3%5)&La^
zE_8pJ4wW&pW2h-*$FNpiWBAF6&(-3cl0=}9w1stn@~Kq3O9W;}jGxWm|4WL(?aZvN
z)&Ke%o&JRnf5E|SKk{Ec^k<%iPJc}QaMi&MVS|7mVPg<YvqzXt7~*h09LUZjKv+Vh
zy`RKE&a@^V?LJvF8VfDm-3qAN0tn>U^?85ia?yM8prW+5Y!=B-SRS`ol$=GXR3n2m
zEL8_Br3gHaY-ablWHYyD!|vF^Icp1sSY|iYw&+nn*^lV0+c3u+B&5W2gFhPy%###)
z9I2i6O7&})dxntu|GG>jyx%uDS$RdrC@ZZ`Rhuqa!%}t75`t(ZPeM!JL&~aVa14Kn
zP?T`0W<3uL_!vgl^($dF>me^r=iEYp$epR-p9j$$Mxlc)$DJu+a}K&(A@jU!_GD#L
zWm7vVWb7V>V<<VpEt=`^&Z@^QI&VERho$bIiS9`6hwKN?WVm5B^Ipc8_QxERJd)Rd
z&01S+!xOTtE>~=dQaA|du;?8~X^4N?Pmgj>rknWX#uB5=Ibn`lL6}m@Fvp`grSnhu
z=v;AHsc=}En~OxxZy|PL2TkR~+cEtZ$EAa}4w3&9ljDXHPUo{iV(#M1Hqr|PLa{m)
zxHA&@<fRJnfI~c)X~O=_#|l25)D-$4%)@YWU{*e@I0t2o+AU3aqy)@Sbftgs$K3u;
z=#HY$0T)6Rat^wR5<a&!MZXhPRtV(mDg0?qNll#{x~R)y!C9xmbGUpp<RgAR7)HjS
z=s+w(Dsm3Q(h?!=M>yAA^2yaTghhd{*g;qvM0gm94#Fa&C+8q63=wljE^TXk3teu)
z`xc*ym2C~rXLAq2HY{ETHh+K6_M`-yb4?`U9nX45U-^bdXg4}fcL>AybZ}=Ol{yD^
zrih%|4hQmvZbo)=Ycnoa2QL~@q37bo4dEJT1Cd2%;ku$$GOCs7c)^3!ig9uFVRFZ5
zTrP74Ow39GLKO{|ZJ1yQ)kv@y7pDW02}#K*Fd0EW>-qG?4opvqspEg*bYN0Gu{gy?
zG0ll$N3!mx`KDoUIw(O%4o*=f#fXBOJ%>XyJ5g`LJ-K6G_qeBVkvdSxkPw^#6;Prp
zpx^;Ne$?`Y1?m7LLIghrDCFIp1}dRiDH%UI@gEnd0~HU^``Hz`XVcC6G8abSoC+Od
znwa@FYDFXJt@tI`$98`Qrf5JMY*UTP)4_>`IC@U>p^6wK%ir?V1^pa>%9^yP>XT~T
zuTac|(52LsAa_p28&Enc6c328YvYs;R)59;*$3K&Kzs_$?q@9&s-H|2zX8iXFTG}k
zywys$?0l{Fw~lV-hGpG{(1a*`3WP=yE9X-!P#m^NC;F=c1H(lS9`xhGvhG8uUGzQ$
zKrM(hIni|W*10#e*NO(oY>f~-o<R+YvyYL|L^Jlt!I^t096KuHf7Q!JP6{oKBsHJ^
z1NF~dmvOu@69O|amr;l}6PNN8DJhrT4mS#abZ#+_rTHkn0AYX`?P>wv;^=1h6hAK6
zLww2KPmwILNj9gOGb0(vVKfq(tT(GZ7F&Gzec{X3r_0;Z_pb^T%bXTgE^ogr0v~Yv
zDL4)qy;vcbQDS-fw*2(Wxit8>;0HKY3O^QqclfPOpOkPoL*QRRoIY6)Vg4(AUhuhp
zNb};lGB;At_>uVJ9G@uvOtUzC49+w*(ZzWJ2gUjH^V<(koK)!LH5M33Rm%%VO^eJ^
zqcj@%X%|K;jsT7T>H#jy!;hsge-b9HI9U}W_&nw;6%*Po1{~8`6Xy%p<Jn3wi>7X8
zM?1nL$dIMd^7JWK5KwRUoTLM@N~e8)Jv%;;X+%5=b%8V%^(cC+w4-)pkw`huzZ5%p
zvYaL=K(f516;ke7Lwe@8oV7+Xdt?nw)(DcE)6-xHR4dJdq$8|4n%c8Ps|n4sEu+oI
z;?W8~mZumwvH<QJ<TY!7X7)}zXjR1kC#hb`EVX*9eHt}utQ1FrZmyrFb2Pnwrl*Bp
zWfju?`&aE23}tVP8s!h|a6hhLZmMu32Q>jrf8&<>q#%l-X@w&8nr0Yczn{Le(adNl
z`~^`t7Xwf@<L20<rz~z)(bMDSgH@hc?)z2R)Y~?hKAJxU2$R}5xo`&faZ0P@4-A>@
z=#O8KuM<PQMl$JC&o{ag?Ff>8tUu)mn!oYGDMq7hqWYR4q!DPojX*i0GH(Kb@b$I(
z>&x5Izc)!<qc=Qq`scT&@Boxx3XLNSj+*6<UzgvW{@zoqOJ1}ZHzwrY%4z~Fi^6FD
zCQHQ|2O!A}d|7`)H#H+aR{I<{)=}LNmwcQQz%v28Bp>2FjNwa46>-CVQyC1b_{+#S
zMbh6<&h>lfCO|c`RK1%FZY8kUO_@Tbj6R6$9R5D=yihFc10?y=&wTj>AO3)om3Pa3
z;?!@24PXAX{3FD{0bz%c5D24xJnRwXi)7w;BD<2%xXVN8zL~_FWS}*3-tUvuz^vCj
z-I<sdm|1XRa>Y2h>gGIuodDc0Lw~{e_A$(2<VvbPp=u)uf!|0uhM&xc%T!CRR^d|y
z0=W=p2>{7*EMkTWDxsR>Dp@|tbX}mdT0Za33Swu`L#2xRoE3SkP=yLkl4U&14}e9I
z?5n(S6$Ic+K-Y!+S3lJk`YrGY`*MXFj!C<Km+0L!v)(L%$Rap@5<_i7cqt<ABDC4(
zQ<OTBQmcV$2^u~j3Ir6HJ{MkoV+_u}l&6=0oT4$VI<`x2Z_$2rji%8H+2>3YEj^nU
zUv-R$_w&$+PB2l>Qf+3q$DIcz)`j|N;`6SFp<Z>>%zDDih74*R__!u?v*Y}opN)&6
zjm*|Yjb%nov>c0nw<-8xv}<lL(YaT3<JQKtqg<*E&T{_~33P5pjts65EjhcUK3Z_+
zM$O~hhP@oPOT9xxn<k>oC=pZgc%;6lDr{cp`~nDLn70-Wa~(QB>o&#ZuYdcFmcV$4
z#3iO(@|W+XxK!D1mP@^4l1sHcXSvi%Cb?8GIm@M9lDLe20Aho*GxU4_BOYK+#H->F
zP<$(exnn?}YK6Gx7|@?LCVHSC@i;9CIF8%PR-Gv+o^um{GA5p`;xo}4_Tl{w_*&sP
z+V+w?4%_EW^U5$SpywVl{h(XP*X|T`?s9#qx{cJUiqe?FUaRtk+0vCh>aDf2V;?$o
zPf2U6KvLR&!=7iEg&vpBYsE}-swfboU0h-4m>v3eX+6e4J>FDUO2=cbOFTInr+a-u
z+K&f!xmMf?#^fht_mK9*=k&R+!XMT4-If@ru%|XY3GzzN(So#hE|pwW=e2yFvlU}6
zn3{UzAx>RQiZL2pd#VkJ$Wh`jKU#eh%}wt;No=TpGaLKwJH7WfyL6zzv@+31=|1Ni
zQb$&JrcRkzYUQ>jAtncbys|2nBkM@rPCqyEbxIv^Lu-{A#g(CnyRE3$>#F9oQ=0l&
zkKBwdTqC8N&`eohn}&A{dNX#imCVW7$K!r_yu;efKv+$=v9lSdnLW%vO6uvmhdjGB
zK%I$y<RB9%2~{Q{Zcr%Qn28)BqOD&4UztcFU;ww#Ap~B^D~|5}CNfd7Zk>t9u5Bh7
zh9pRa7_h`-wN0e+&)pW;x%Wa_4DP*%24x+Zv*A%^9n;mgly$DB9O>K*k((r_Q_m{F
zDD}|%D)mUO2dRhj>{8EFeFsT}=I$qz%AN;*nWIOMN*HEV5UGiXrOfI5a2ry-i%(NV
z&*Ftq*C&D3bsMaODwz#L2_~pbk}g+Q{;Lyk4d>5m^%!0Y(sjQvjsR%h^od`|>p4A%
zMw`N8s@!_tTy^i1T3-eIULpekrV)m;&RLJj`x(Rt=D>D~n@d*G^tyj{r3*I-&h?>x
zvuQcFZD%CM90z7rOw+7**}uA!h|eta6e=v=+)%q&7dAKSyDls<yZ1Nr0G_0|x;llO
zNrn`1dq!4hD@-XHTdgsW8!~JEuGn$ch6+!qN^K~UKD#m*Tp`@COjUDsP9{KYMZn)3
z<#EoE&N6^9OZt=21-O-SZH3sQZoEx@VE5kgO6{4I`h#wHr7;X;+rQ?;jP`{T<B#5k
zN}&-s_Sztv42_JtE<>f#=tK4Xm!T4&2{`>o-{)o$lafinv^4u<(ZgQ%bSExBr3B^#
zT+Oq0w{=gspIdcfo9-^f)<Ku=e&r90DjXI;`;fbDB{3P{NLDHLC-wORW;?fkzAf|8
z!rfAeY=z4#jLy4O_XRS|n<<c1&lX5qQt^H<k=C%#Pi)oGm)$xconYK`Yhv|}Xn(A7
zLed3|T&-@xqF13KPPdD@1vP*YcU`ImEhLF-W$GFzXY5~`lTH}>V3BiXppTPria2P3
z3FdLvbRNl7HOcd`p5Gsr+=tA6VSCBV6wK9_CbqkyM!D@Sa6>2|tVS%mD!col;{FSW
zyQ&J~FwJcE#bgefTjR|;7yFw6(~;G3qYIsJ=q<j{#XxhAtY5>u_By9XpzIS!Rjz#N
z*EnNf+pfh=$M1UqmjH*@@S|R<6B-apP64j;2FQDWF2NPk&c7$>5{MFiLciwSQJ0XY
zp?v5Sw3|sxO2$3fUVjYg5<<oA9(8Tr#1ZfICe9q(x?k%fYnpHi{E(k-B{3P=JRJJ%
zp9HrMmPlLPc#p}?t%W7RD<wx*L#ZBKrIE=#Db4QxKOphSw=+GDU4W&r<OtO%M6a>!
z))R}-;|A-dt^ky*>zZm;-xN{G_JAm=GzEwvWylv3@h`C&&Q<MiTL2$%EFdP)m;o^z
zH?`U32b@P}!vIrq3|6jnuRaQpNFM%!3ObSA2ajmMfQydw2f~r&r4;(c-1%Q0fcqww
z0T&Gu0Wp`sg#{FsV9yQAe=e`TfBgI;(-MAsCpS1j-Wt8czp*Ho%gn#_uY~+Wo`IMm
zUdQhkzB8}rct*6`gG~6vPe|XnC)xD$({H*)e)e!QM8e>NBgK)=N}HEN_mbw5)7*ih
zIF}pS<U5Zb)r_cVQ0S-gO2U}@c3khP3Njg2waK+o+=!~@TE1<Mf6}raLAG#k9kE1S
z52?rvM4V;fuEsX$koG3((AYEbZMQk_R@k##d(Vx2#(BXmAWAN;g=X>!t+uSNtk<M<
zLhA3*dM`C4Z1f<HiSm2za0%GGP4X<<QR)}pHyRNr`TAGGH*HDAEJ;Qcy_sn@YGIto
z*ky}~BNB3Ylp$u@e~)rrwL+Hvr_s)ZH^J1y<EQS|=Z~kaGgi=}K=uXy`tlTBpa!9x
z8bDit4u`+~GyL)NOHVdA6DcH*5fuLLx4-@zx|V+?pXnuQk~FL`jje-ZO~Sz_5X!lN
zhEFI!lJFeQ$(uP?q>Nqw&M5C;h$_8Heu2{Oz=6Dsx){dEe`~^efBCCx4<3UJ%P<C7
z`NCnIJ$olQzGkw*knSnUORIs1!uf&Nh$k-}k;bIKF0Bp2BJ!WOwx)&GR%jX{z^IO>
zk`3FXGjA}wMQw7>qa8i@(kh%22*m|O_a8*|mQ8jPX=7SpB71&`o`gr;&$zNiCE2D3
zX8c{!B2D7!e=xr*>UW%f<-A}^>)BV!?0YBJ3U7z^ZgzFhD421OeOz%rCnyHBQ*K64
zb`<w5dZOrrb90p_yTc;3+H-t}rXRJU-<F*)sos4>LAcvY)p84%p7N@#I^h-V^KsZ<
zS+3}HA(Z{B{&oi2&bVi=ZSUS-kMsL5gYDSgG1yz-e|?<6wj9h1_KbNYgY6L5F)wEY
zn^-?;MH@MSB@S}Cy%*Eoi)Q*AVapM!DRK2FiApKJ+kVznjda3Fb03d~&5|EvsZ5e|
z1V6Pz<3v7vu4(dXmG<2L3aAiLb_}dx!qFU>8N0{XOL)}Fb6t$-UvtVASH2mFCUP>7
zv{MqAf1rLg5?&4oKyy9-Z9ecm-*JnSZN+)ki0r4_4MMO&0rgz0N`Zt`*!NzGoAu`j
z01P#X&*%BGN{~|hjDXAhStG2qGfd?w_)2l$><)u}t@Po6;I~}tf#3H29^kh#0&Roe
zVY&c*N`NR7%>0zDJRr47X(eMIZ1J4f#S&#pe_BdxfuB%4mosPNs7O+uIV&wJ2X-lO
z&J?EkT%6AF_XU#1a=_&R$=`mL!@v3E&)iviH~cSl{Zgip!+(cgLuS51m?MKo7y}j?
zbmhRFn5|)?I#D3)V{CDj!a=@>(c*k|lPt^AQ%^V66;?7~Rw>1qFj9Ew@6Z0TBM<H#
zf8Q!yauqd`d+pROxL2%5zEw5SOKxSKmfy>%5`dR!izx;&!P;nNk=M=0?n#WT%B#Cn
z=`4wZ#=~UE>syScx6%WJeq_JxJ=TT(dgrK1kg0%|BW^0iy&XlV#^v&JRQNeho|6UW
zS)$8!9hY++rENdz);gl*ghg`((AN=-fAHFDmk~EjdGzI)(>d}}gnHRD*VD;jDi?OM
ztYyWi(AM+xfZ}2??wdCy9dMVb7lRH(x7+q4Z4^5aeEgoIiM5tjc!wuBP2wPDnk8)A
z4p|T$Tu=A-?xcwcD4goFvpQ_rPgo7po)BI{HKCkdsB>k{G-9F{PY7XrP81gke@!!9
z0Q1@9Z1nM}&Ws;bovl;})Va)jZ<TEr>V&dvAHmqn12W)cplAuGJJY=ROtM<4x$iT9
ze`h{4=h9SI<ilWOn-<>MkTSHUf*haJ()s&8en>-bd5Gd6rXBLPAKH+ACv>{Q{CC_6
zdtvAAa(HJ+9Q5%_7X5dMoU{Z^f4fWRTidG8x_w*?oc$7~M;(Qo1M84!hQrowR5-Bi
z4_T3NP+nRt<uX7#<i+@=Tp&;-N~<mlM2oSb4UIEI$5i^3>jpcA^U4dwndG_Esvsy%
zde&FlM5=}VEQM?CTg9H4Z&zKfNC)vp4yVibYZKR2S29w|tddqn;<RO?f4;guDsNVh
z4)>>(<knVdljN1H4)6J?{Hs;<bpd(bF}va@N|Buv3-~REkoShi%6hh6btV*zAp0B6
zdfD$u+t0WbDyJfkGER`5iug0I=RmXQZav)UW&_wfh`xch7^`ioeDnq1?>Yg8{2cb{
zelgV*l_OUOIDZpb%^(^1f4rYZt%T%6qUZ5?_=<xL1zWT;3U%JC_SHXL4O^|;N6~F2
zV)@C2J~Q)B&5R|%uTR49M&V+I>T7fWEuG3+_;@(1*s9fD>4xo)(HVE+JpMWJ@jNlI
zwIlPkuYw6^BV<)g?$QvTc(3eCo}r|2D~UvYbs|duQo7$`52gY<e^hM-^q?=e9LT4k
z=c4pB3dk|=QBqAPy+|r(WxK(I>pmvPkT)2+SyUqkrD;=QxUY?$CI}&i6aemo=51<M
zxa8)TC{HU(qN`J-D_du|sOK<dZ9e0}u)R#a7`Pp5*KO<)1=)A=C>v_?E~q{3b+J+L
zwz`l*zthDq@Y`M1f2GPE!Gk{|)h&sD5nB7`r8Gfzkk&GgHG&<AHTWQnK+QV<^z+k&
zVFaKiQ)lpvHM2qkH+wDHivZ7tk?aQZ&S7TBH&5fEDE2dDrr^A8($%(Zkf02uSM5!A
zEVc+`xRT<ee=L;U{2?UYff5oTE~vMqc8QD|(i}=#Unoz_f0kr{(pJ3_Y`JHU+*_L3
zCMwruD4i8J)uGJ>l1|64_hCqKV9`xap8}9-EUS!erq($0#@B!b)skLyO-38a5Y^2D
zxX^5^&%NuZoYCD>O)=w0Eb_}N_+1SYw*NZ-h=kJhnMb*@U3+z`^R4g?0`BquAOIY*
zxaneyetV@ce@E@}+$w)<Gxp4VchI(Qa+A50DM)xPHG$ltJItFn*vDslcWYj0kRx{N
z(@G<SW!S#Q<!t~IsLeOHyxmhih~{3_h36B?P>J!@^&G*c*$ua;3gZMi+mBEO*Pv;t
zH9zpOIY=$o=RfeWc}H>ivbk66wcUd+o3})FE}L(0MUb^(>wcC4LFB)|J=~V?&NbXU
zE{?2(=WVZSU2ZG2q%u%VZk79Ssm;t#*`=%!qhJG{+N5aWa&djzDSEX%{vWy#IK`I%
z7Y!2uF_+;o1{Ifx*$vJCF1J_O4YmmcU7z6nml5O*6_;_17!#Kh0t_gB?OV%oEVm8a
z>nr95L=yz~s#2*+BVD=5yExjpyh?0uau@L?zaQ|~e02BBX!^Pmmlv%Wu}KgBLEr%7
z$kT5pdHVk0{PFPJJJQJtr@^Yz$FC>g7n1*i<VK@U7C;z;)5q7-Pft=R!_S!?k`j?0
z%ikS;>gPuU$31}m19*N{v;?95J3p`d9#X$JuJ(--G(U>pyu>%?pK>qBkKvxiCc4}Y
zxE=8M=Z_yAWEMdRZ^GB`G5+`E<Kf>Aqd>->kk+04@%16RAOwL<oxli>=Jfk7r(YkJ
zH{}f@3M3%}@$mZR_uo#JWaSMa0iKtK<qaNxZDB32bT4~a3oWO$feqHQ7gkS-4txc*
z5t|@|?_ZcN=2yNyKN=ZKJ+wv+AxmKs*KtH%_o#H0ZNUc$S(-I-+xqOy8oiK#U1fzL
z6U_RjpXKSF{PG7T7>wXQvFq2+9fW6*{&o60+xXt*CoLSX`5;?0V$}g8X@bR&dwx`Z
zAXRAgS;$~jF1<ofI=s@)V1sVL3Wm+Blw4J2Af~XqI_Fomta;L!#;noTJd7s-9fy_q
zBM)VW%Ze@vcX^TYvx*DmvN#JV=tyzX(+0V+k25(nlMCzGuasi#%d}EVXm6!RqEKuL
zGB4H2azpS2c-(R;c76t&AnVI=V{61?i{lD<3i2wys5{EdU4h%c)>XOCWudZYdhCYM
zz!lxJA-FK1(QtvM?ImMvn|*DMOSY$OHUQ^X-O>o&Ce<I&3Ik=ir4kzbLCdXGWObN7
z*fR6*s9e-|inaNH7<V5zE~|}~2j>kRf1KgOOP(_=(Wpo5Nr0_t)Z1!gPbIXYW7XJC
zAGN4^E$VG8>Wf;`6052xH?sx}7+IQ4ZZ_k?qkRNaRmpyMUV1P!e2&;j8zT&Byu<?)
zy<Y}%F}~&a7uZ^&cyhN|g`RfP$Je)$mro<$Z7o(_27DU%Q`&C?_%y>Z$tHCXfACey
zybbhu!N6~vWfS#TVKh!WeJj*Qf;5b?t)RY}r~=~{&?8kz>tKx?G#Nih{EdL5;aOnP
z?-r0WPVh7Kz762~Z-GoL**98!EXdT7;ki+Nfy^U8W>_g}&3zz~oEJvHu^^K?Pb=MD
zAah)@plwp>V35fM6XVnsAd{?(e-7(VbTrKWe+x2+y}9%L8k9*!3j{n^kNy0ADwz4*
zyLP8&{L-w|AW3=QY^_123S+4V@^)qBa%nQUQJI~%%b~FGDo7Ej+U)*aX^BC=H4(<t
zzF`OZ+K+0x%#_cI(*~J*M;C})0)H6^m(jQYk?^*JaGikNrJ^iJCuH4I6j;Q0&5T14
zvZx>d^&HbEIk^nBUzfM&4L1Rzm+$BeMgduuHt7v3f9EFITxL-9eRUxj1hj6%m0rF3
zkelv9cJ4z~_aQs?@r?4Rbs-nrN8<Cn*>wdG>Pj2iq<eBOg+_2cuW3wLT74dz<&Lx^
z!y2LPpebK?OCvz8+k$lhYE>aBQ5Q>_T$-A9P-bVTL=+fb7?rKtYxi2oCTh=^S}i-Q
zB=U7!e=BNB!w;%k->>D{*7AY7xP03i%1Fl7`jDXDntbJ{j9>R!A;BzK=<=dY8%3@r
z7poeo=6y$v2UWmVQ;jEJH?Q|=JW|Mz>Fe-E<^n+1@?)+8dAU~yLbSr*lErSs#)^10
zs!C{>aTuwow?r|XiRSaqE7-osvogZizG_mye^WLNZLi~-I69&pDXMIM;cPd2N3&Cj
zh$zqr?XAHz(8Ewy_~kGPClD8TZ`1JitH4yK(`c*(TwFr%BjwGWj8&hPO<3cM3%%s}
z>}7o3g6Y3@3+0xAwj#u-qgpsC7XXy%A|*oH#$(*9jcdtplO2b4yj`TAQ#kO`nCxk4
ze{HZ2lHt5)*4{m`tb>ZMBc^%c4Fgnz9pu?;6N{*+ebzO|4!3BotYFLgv=5eW5WdUz
zRVS>pVbrL)0(ldrqD@g}1(X5~k!k{5{tp@oZgr$>M>ex8lF_QXl~M$$s%v@8mXkwo
zrk`<4rq@;>n-nW*u<V~@?0^F10AKaFe+_1xAfQpd*{g2taP;hrq@PADVSR5-3P-Fo
zjn?~f5_qAl8ApW*E?G4~2<$6++O+$j52C~ByQ*DUuxB^UCVZF2rS)62St4Q7a)7m$
z|FsYaH$2KCDnDCf@=h%fi@iFC6)AS$VQFn}eUUQT4fnv2Ql?$Gyaur<f6-R=e{IxQ
z4rsGJo4E@FigU&q^yMrpz?hz5l35j9ots+qxk{X3&aQojf-Ja^I?z5(I-Eq$QoiOq
zCuwn7t$j|MipT2{&3#TdgXB1{#z{H_@cGcOWOA$4oX_UA{5djb_EygUN(*45GK|hQ
z*EaD8_eO1|Ig4h~4r*e;Mk1ite{We@BJ#_#c28ZgYlsF&cBN=Fr8;f>C((J$o?tAh
z?MbO5$8!TCmr`FC<%Q>_VN{=Y?^9r8>Xo^5UPPf7&km#dqTnR?x>baWWs?LIW4tu>
zsj|<ZeixCss9jXy-ruCAD-R8$W6`D@W9O4BqCO*R7C)MTzcHvawbfM!f1k}W<%exB
zD1oD^pti?!nU|eVCzY^%*;;M^AvByrk?F(el0UcLE9d#9Im9bEb`8AMT@G!ULR8_v
zj#=_DXUVXe3c+UOV_Ip~K-n&UHwDmJFVK=eMbmW(-d|eXf1d)J6yW?dXry9ubNjuD
zd{bhWZR=izF#?Pld*he+f27EUizU3zYnmFb9Mq!Tr!Ro?L<c@UC3#1%ISE=-fMqgZ
zM$zJ}1l4Za^X&q9Ej;0I0+wqjgMjU6;=Zs-g;?V@g)P@<+LNE?vdi8xOEP;4-;UuJ
zC%VTe!XrC7`n)2LQyEmSz3Lb8x&xGg&T>41<CKEI+(wnSi>jb-e_tIYn|PgpV{ro7
zF>p{vdx4(n5pJNT7|(O@3V+}jOzBC=BOF5AzlX`1e{3*{dV@zmC&<}tale&IC_DHt
zRUiW(l=HG*as)aA?49NQBRm0ePGyhOg$H41-!8832!l>mTtY!MWaG7vf^6h^uW$me
z+?T;mIE&!64UUwyf3Qt)SbgtWo6Greft<mSU5Cl)y~9aD&W_m)*9>UAYJZc&`P5{f
zyd`n^A=S_sCpMC>n@1S8KpZwS=f)W~u8meeH1M}}ruo`maotilXM=>MyMXGJYog_t
zs$KG!4W8E)r@2mQEQ`tp4fO!3Yw1ifb0SfYt$C@!Yl_3WJwI?8C~orne*d8vLlMGB
zhMFnA^wIs&fmz;mUI`8EwrnS49$8uPc4G{F)cRrKEnCbo_nq$mG9|bVv%i-S?+qdW
zZ<jRh4Leo)P6$i+KtJFUzi23Fz{QzwBATlO1JQB^6?bt6k2TjFoqXB_fF+Jp&`ouZ
zxY?t&<nGx1#pib9-Zr2@0v2u8wD;Sej3_<0>uY`n^5x^<zc+5~mw_2869Y3cF_*Cr
z1QP)>m+?ytCzloQ4V{1GkN2NHzkA_n37k3YPyJ8(-^b_6UoY)~kUOSP4ZnPS@i&wN
zCaoC2a3(bjfB!c8{_=fGcASjUGYtJ7I{W*Np}F!y!qjoXlxvMxVG5jpR)f}#Nq;M1
zst6Y(e3YqH_RZD8EEOS{2HR5Mm-sZ(hIj!vz{p~Cg;?zPLCb$6Ei}uRpjRYk6p=6V
z=Okw&2`G|)x}@v=nyz>rDQjSko0c*s%^@OzvNCbyiZp~@6YuNrE@JDDIVIZ}6{cBE
zneRI>CS@Um(8@3+O{Ru2p>j%6^VCI}Ma8xF-ylsL$I#kdHTSj<aI_HtkA&w*enh?$
zKa7A7jyYpno|S(~JmXy}7fLfn__b6n$N58~VXCLxNrf}z$>l#`mg>q6H~80<(b3M*
zd5!fMMv6u-6rC=>ms!mhMl)kfJKFod(Ly)OZ5_i7GX60?wX=-G-6`#9^R$v{hS)8F
zw1|$~KuREoC7>5g3L4TM^X?>@qtr~1if2$i3dOvi51N0FLK?Qz4y+86W1v(^@qp3+
zN^)biqd2d0fQDk4?C|k!h?g#C*5z#ADm&SRt80hesFU0>!*%<FIu&8hPN5})6cPXT
z?(zVyLMul4)bR@Qd}g;%kM^!Q@O9o{y2$8gDe?uk;UqJz)1VL}FDZ1$38N5P&K*9T
zy$Eu-;2wWvFEzDC*h@|E5%$thT(Or1pQqVNgB+~eOZ~)uX)g_eEA~>44)#(6^gZ@c
zQ(Uu`2B&TIawy@<$ulU1`e-g+5jm2|b0Ww^naX2MP`{*9IHmIeC+Dv%*hsRdnj$~X
zdKz?sY0~sU4;GUA?nz4<Oshs4<kJRY=?U7Z88LrueCo8ienKf`-GJzr(fYqMCQedv
zrk#1on7CpJq2tpfD9#B)lXe?<Zx#f(oi_wV=rxlq*_j`OJ%n5BY)h6%nhc`jZ$9VE
z5>}R3ZvGDm*E6q4xE>wMPbK~?X%C5MA88*W?c~;*nU}<!_Bu0oK60m;oOe6Rww>yz
zC=Gv=<2%*7QRyf@bf+q%VBB^n97uvBC67|tj^ezmC8c%fI?a$U?{p>oD%*spHvFK_
z%v!fL3Meg-RZ#`&veIIqC<}M4ItTO%#qVtSPf~)Vx$Vn;hQg}O<-bf~l7bEMSvv1&
zDoa*F!X=tN{mh47>E?Gd*hkNQ!qBhdJbZunbNJEM7gq?=PTmtXq=*#qZxH5#V2%)v
zWG50RT_^@@CUGZcl>ey+<1$$ix5JihsE;mWL*OHmB%UcHm}+=aN|4j?P%@BX$$+kk
zJA;yck~~;?f`Upqiv@_0(rIP1D(5aY3cgA`ddLaDN)<=Z?8b0KHnh7@_$1M2%DsPz
zvR2By`uuV%dWMjDCm?$nU%^Asp4!2q0rAyLL&PafDGCOgE5xHtf4Nut8eQw|<0+pB
zUTJJoWLp)-ZroPnAqhGuo+J-R(X8G$*H{wP*1(9%y;$t#c?AnW8$TnTgoTgY8agbn
zz=Ff1BYKxdEwykDvbzyjK^FYhT0MUWS<uW)o8nn5E0KLnyt0OGMt3&~E9i3C?;}15
zT^`G3_v-C_bnzg%yHU7}u5>X0dJ4MIDN6U$qdnw)o>#E2G2?g+7M6C$Qz>ay(@HG#
zwg|eP_Z3WZOgf%}iIM(h@%nJLZ{N@7%K9+|`RA-(3kH4dqp^=A?{gn6yZL`#!9~S@
z{~TNp7%019E5~rz&HoB6GKTBt;G#H#spe7G%Y+4P#9=qTD>#T4Sf7J~a7@T872-Nw
zQbi3Y7e(=MCt$l5wgMQ$Sos{lz%pUPCdBYzi0xk48e+hl!y_fD%_Q!t(s{Dxc7n20
zrEAN9LAYoK%g`Au=lUjvK?Z-Ku3@Epe<+ybTEI1k>{TlJ2J-X~S`ET~N%-ejBS=?P
z-WQ~9ytSTAftu2i@g1Qj<a|q}dCP^1+@Kjs8M1g^m`ibAs!^HIpuOYO`FDW%Za0r#
zNT8&qa$mfwQc0W%v|c()c_B-bu^irAor>-q1VWRu?jZ0iR#FjD*W-Wq-pF>{j!QUu
z+tOq=^(0}9E(vI!J4t(CLq4>A(6=S{4q4BYFSvznS>jUK)PH-4*HYzvkt^SsCPYI#
zzZ}@wMk{CsY>jK{=)U4WrhBuC_NHr$ZwQ1`jND`i-Le|;*tK!{l|eUykPJ|7*MMif
zoB+y~Nt-)RB{Kl~-c5f{X2Ohx$G@8ZtHbM~1CD+(i90zX*Q=#^P5~T@Wlo6gw&_hR
z9ul*+PC9oD1#`&t^cj^Cx>UCu|HbBl#eQhkAouMbhob&w5_jr9kyYw(QU7*i1zJq;
zcBF}m<kjBF|FCRO*GcM0;$F&egh&+p2NM2VPLy<!R6uf*W~6_d8wYrDo;K;$NV!Zf
z62-<1PJE{5mB*>^dUnO<cgf$TMit=o8{M%9KfDylbmF_;cL8~ZQ_-x!<A)^D`FZ5&
z9nD+jRaUDXc{xq(9qMDmLDh5M)S{IrVJ9b?<E>1zntsbHehPnT3R^1X<FCSSO76}g
zwesYQ{p9aJD~Eq@<tB3Rne|k#ppK%SE6swDzl9(h^k958d!cKgO_%9I5ue=&8%8GF
za%hY_w9VSJi|rb(;fZFoSiUg3q*Yw<9Q>_#jb=vvXjYMC_3Tvqx)%A7#)wqV>Uef7
zM%z(`@;vKWyDO*C<ea0cDx<%0%z`|eQUO<Nwy!w&i5!2D%+z?4H7rZJ*kEkxfeir@
z>-iVDU%~n>0~%g?K6W%~#je_l?oz832#;+wG<>bIvOQnOGo55sLEqB19mmuvS`%DY
zMDE(D2j8`fsrPxeju2ZPBNbQpWW7&%sVS{qOxJ#XpdZZwqYSN?Ab;{!>fGALQ|Sww
z2KIT)u!1l7eQTc0P9NKiD2tUE&j#&DALP}OE@_X}v`H4t-~&crXmauRhOhRQoa(<E
zNz|7yDl!uUGcqwbmr?2}6PI}T4aa}J|MK>6r!xy6cbME?-gf`&=1j0g<z480|F(+_
zhCq?W4xCd!w)_5P_jUJcAu*8F%Eu@9eZ4@_I;%XoNb~a--TfhrAEh$|-~F4qek<eX
z?(gn58K6ekE1d|FARcZJ)*Y(gLnb?tU<yi`u%5)inORiU-=+#$1)h^ldggyY^h{gJ
ze4=NfU&ti1Sp^}U^(;!?h~BVhlGRa@lI&@?;#-ur#^rvZmn<ztdU1Rw|BdJ=z4Tui
zJ$#{GhNq;dBc-QGZ!;YmreARSB5CCWzY|R?l4*soA{hqd5H=DGv(h>g(L|^PIuqIc
zAeZmqW;Xy61;Drvkmi|J5HWvRIg5`*gcWNi#EsHC021|%#e(ru4+ckIFas2|>6rBr
zvR?Y4(;t><-%Que+WBdCLVcpKzG&?-aWunL5eEk4HEkpgBBev|QRi3+*4{zEWhoZT
zbFN?kUU_3S!U9oz8lGXn6D%~0I%1nF>QoC`kZnewf^787o;N~P%prd`x~XMTWFO<N
zywj`EZARfXx*V(uw<DfA(A8QYn8kYAjxOJaZaWGUbbTKHZGtZQAg;W4taVxlGq17`
zr)GXCIJrI=*#;-)6$ZNsC%zM>W_~I-*%YH}1Cw=%$UoZQHcz>N1oz>=Hc+rt<ibt6
zHHE^ii^tmMUctolVZwhlm>5>3gdim}ZPAG4X;#>$-j{E~J{e_fEgZe(%9i2MJpT$V
z*n8n^a6ytgR+h?rxHQkdf{X6`=Qg+iDmHXsy^L7!N*tQ!Tf@OBwDz9b)M^sT%5jmb
zwoT=@dEIY8Rn$0{;h1s8>SkX77%H1FYA_-GbMJ@Bz|h9L$5ekmin+PY(FOK(0uDn@
zmkTK|&X{~l<M%w&7>>E_Ahh)H%9y!tqG-n2TApnmt7Jtz;~G6FD<Ye5jbT#LwSAoz
zIoEuBUjs=MUL0${<nl<D@g+t@;yPv{nUL}2>&V1dEtREm#RWd(r6S>6R2#Xy&i-Aj
z6&oT22IjS5l-7U1L#cR1b;(*6#bwip3?y7A0cK)~GCPM}gz28^YGN3*e(v8V5xQJ&
z69)AmF?cqDtTO{{8p2Lti9$&V=f-e0w#RpskuEd6r~prr3^1G(;Kd<0fVTR(i3{G~
z<LmDxZWge)$HuKDv211(ASvM{RbC3Lz|(z;J1)EairarW$tc-<vQ=4?qPo!)^+P*V
zz4nZ%9i|HW-m9S9J&LiK#Im&z$vXXiIX_~aZV8sswi}GIxRb%{=-keGusP=IUC&^V
zib1DbI6sU<@`>vzu5_?Tu{}A_tV}209`d_Gp<ZZ2irUShOU&DiPrXxVbsdCOIditl
zXjK?D8JmAS2gM>c*lTt@+3wXF5s$F#=z;?9kF@O=g0co4Z`(0gWti?+!mCLvoSAq?
z6U;WY9ix5lCG$GbeOR7Vw9m1|yCgvRuB>7;iG_+NWa&292O3s1lC>2NETJ;kkc?L`
zW0TB*^8#jfIVN`1Y@wLc!8}BYFU9P-VPt=9(xZQYecqEE46^GpnZdO45Ihz`RK_|^
z9Dpy*;aYa{l#Q6FWH$pfKe!=3--br^QAo-}S(B`Z?@{V!Z$G_W&4MU6CHD7JNGW!K
zK10c<Oi!m=leiJ<h?3u&61(6@?Zk2)AI6GdtJs|JQZ#DfHYUzGN`eoI_+w4VOA>NA
zK&yXWh2WLJvy_Z{a*X#nRur#ofol>~Mt9>+g8byor`K`bgbL8kDFe(-E)Bh{umkhn
z(V^G&AFiZ;2oj*NZISC5E}4=coQ-ykOLo$W=DqLXQX6H1J2{2Hr#5K2gx1Oe0(%zf
zNoyFZ1nYtxa6$y>@j(w*uYmX-_qv+I!kK@GQh>AD1U+CyV=evcC(7C}tjNY#h2S=l
zjraxDW}72r?K7_+LL9y9Hk$w--YmN0pYTjsJ8l(upphVIqmuxDPO*27-r{bba|H_!
zXS~tA9e@NT=zr`ZWz7IoaMhTCvu%LY5)0Yz<DV#NN8vWQ0VJ;9=nz6M%E6C+oUASx
zg4@XYKK9uJSx?IG!%y2b&$x0Vv=1~k0m7kj0k=B((>(LDv}1Pv4_HNRxs_oK6_+tq
z4JDUwZXgK)GBB6HKp-cV-rp?<f39D_k7++}$>ESg7J&gLuF?Q4AXqNYRuF=-yiKmY
z+zEH9!uj`khI=PzvvTfLhGM|d>@Jra&O9^o3|GPBVw?%C1mgo4a%6mpdV_HxDnvtE
zRB(ohH^IALm_&taVhAcY6Jr&;NjaTCh$fTMJIUmjRS;8Ly+fujLGW11e~CWxY`jlZ
z^2U?XReV;B7!qL|2YuB?NbE5`=JrwQQ@oLUoljKvdJj^rYd}3;xXzp)iK>lB$-ANw
z_C~BTQ!x}ow}6FkOZN~wr24K}hD#ln48@=`GoGfZGh-I1x&$$lU7aPvSfcv~rB?{!
z@BlCOc7-Xi1mmqxO{iLge>w@{Y4p07nmtvUMf^zpxCBX}-h&o7=)}6`QkRHWv84Jm
z8HQc42CUEIbmmwf)i;t;7pi748V;n2CNbEpt|53Dvp!8^IC<S<g3~1@mDq6MI!iIp
zWq(HSTJ*lYkRT<U4#tF`>U3fvQ+*mCReNIxvG<?gjieUwLXdRNe?DVA$+`{vfF^1x
z03{Y*XD%i=u1|wWzHTG4N#u2w>>$N@FPp?pm51vvag_kj7<86GkgCVCn3Sri0Qw}N
zn=DCzTqQg$)qIGNRMh?&C8Fqul#GzCPctNSu`k64tEL3sNraVx+Fw$MFjX`nR9{sI
zrVfgC$qHe>(ny7?f6kemI+n0bcIs2YkgV%K3J9-IRVk39-orS?HopG4dH2h0+q|C7
z7hQ9C_rBAszfb2MP4mrS`Jr82ImEoXZvNJM-+X)J`-`{jO?QRZuncv4NH%Ib6|(9k
zz|H~;*Uf8#=`Wk_7Vj3#8}n7kLV{ovSBe9CUHsE4{8Js?e_SDD=-h9cfBpMEu$LOj
z%<pEi>n8^w71*4ryYV~kFfx>E6>!@}dhu4n^kJlFzFo`~%gfuzO>2U_V490b*R{*J
zq5kr(pStfZyGhp?-`_Z2%sZ@c9$*g#!09UISrZnA{sN)qHI>whqA3c_!^Nvt&BbzY
zbJ=!R&BdGZf975LsatDYU3D?}v4ukOuAO%)O_BZNZ`;-4Zh6zL`mCw9{?UGzPQF=u
zsxH*DCk0-)m@M%oa<ueoSJ$ul3|*H$yQ)cAw^5TcgzF=L5^O<$F&7))Ed;qOSRHx8
zmRLqWK_)L5#Rg$h7|99A>>epcp_Ov2A{L>9#OiH9e?XuTs8S;(kYU6{hhz<0Cvk{v
z_ZY8Q*&d*itR{dlXosVEfFW<%0u*c?>6t9j`?Evw5&8zW+!u;IjG*ZK?ofQNg{RNw
zf@e3u-n%a2UCSW>Nc0eZNc|^?JHbR0oh8&+g?vyd5(@x0hJ1G9pjC<)l@MGha2g=q
z=Uos#e;C!{m@?Z8j_^fsELxG)zBvasK^VLB6Ywn*p!JiePyEr?r#wOd_}>922;mV5
z(mD+jZ-f)!q96~0|JoRoyC)j}zLdx&7e=^<3=3(&2A2s11;?Kz9SayOv(z+9<So^j
zAU2MZaZ1R<nHZ>yB~T9T6e{#)0427{!O>LMe+Z%l`ictS$Kk=bq>_SwU_qBifd=70
z^7{nl@qkTaq>Sx?Q*Iy0nXEGI?IXcu5%%;P(Kg^Lr^=s9qsU=*NaS#Qh@3vW`{rsb
zlqVwRX_R0u#;;L={?39n1aK8KKUy~HwA%n!kmj$?2B2$R-_L&2O&4>Z=bvxC*Z*Jr
zf7EritG_hOkJIj_yZ82H@pJRvW&2?{{n&n7>f7)J%y0iU1E|h^3+)~BR91m}u~C#g
zP(TML+TC@F<#aMzHM8lwoh+N1$!zw1a`SQZ8Qt~_JM=BqB|Scf790;2^;UT}A{&a2
ztJ12i8d@sW58{peF6D^F44+*9<&zyoe<I!5L3F-s{>9@O#OV<J#oJODKaqAqwA$)F
zgRuto8CVWJL;gtX3++AE3Go;+#xaeJ2hrF_TM+sM;Pg2Q?QIkp8zt^w6g}V>8YOMP
z)}sOUeSUCOPms2cM83CRu+MJeoq%^#Sxn<#PHA_8Ka@pr4`ngYx*Ddi5%`$0f9NGy
z)>GegS@bzsP~n$X7B`A+%HrONj>_Wf_J>tFUroE||EJwAC#sE+z>1+QIy8sQEr6Oj
z*HhD+WXy|M0_bo6B?mVpBp5G)8kuC58kxq~oXA;m@-lMOQQ2I^=VL+##gROUqZ71Y
z<h1okCh37=u3t^8MtxF3#D0Hge@obsS!I00lo;_K;SY1#na9mSc87Q~KWY|a(<4YZ
zL3f?d$$~USbkiB`CH#(DOcZq2G4JQr!?i#r!A!_^1q~+p7O*qt(Qhk_miUww?4m<~
zh#^OE9K(=p=bhMge&US2pv5ia%eJGZVm*0$wwL7))|2y=*vxs?!+|G8e`<Y@{Ol&)
zpNH>j?-ED$F5wV+7v2S~gHh`bMIewt=#O9c$c@?<a{ho+soyQ{Q)Stt%2IlP`cFlN
z5=bN+tKUC21d=`S6!Xg?kQ+re1afahM+EZc>Fp06?k2O7vB$u%9<x(ICk|+_e%gYK
zNuYWupF3IkGw?n*=qE?NZlA0k54b;6(T@zHWcAd@;M{LcGG8?1p!FxLK^&Ds;>h!T
z98s^zkpwc9+7+W@!|(MkSa1#Z2P=WcC7zPk;tqcbnh7q6!@%v{?T|Bvqc-1nvemyc
z`JzWZM`XtF7d!HDh(~E%b$$h0K3uz(5#$XNmqErh6_<;%K_`E;TTgG~whq6~r|=hu
zR+LCd27*D8Y!aZi?aQs`q1blY!{*_&J@o58N}^=h;~9IL@d93tWr-qxqWI5>?*8a>
z_j-7IJA8Vv*m)HQyy@P)b-*t){|}lQ9d#Z+ISbv}d-wCb*2eMm$QNpDEnl902fows
z&O+cGz)t|jJCA=5`@it@#P1>ZOXF(aM1lA+{N^FOvHiXD(tJ7Y>3pWk{eauS9e;WI
z@t{Qn;5#(;=eNUOhp7-wqYAz6zP=yg3$&nYFr6m_5W2sA@4g+rR0@T%Rw4f4|6fNW
zgVK?tb9_%EfBr>xzwyf-xwGv<_iyU@9=mJRQ{R2>{=<Ltbqjg`23<QxGy)w+k@ify
zS2n)Npwq+g&S`V6g9rJg!Dbj5CP$)^?Uly<1AH#XPYj+8JIPOhDY+s)M%8%E%a1xD
z01*pav50D6%e4}h&HOLmf~bJ$?}H0^h1Q?oLN~-^Gye;?SgW9iop7;Ep${|rWx#^2
z#9=eP3pjrmql`5>;b8a<ebo;(&dA1>*k}NHK!v{(%Z{uM6x;Y2A7}QIMv@UAz4@7*
zRwA}pvIWH00~v33LJUxmy9zM68DN_QTL6sjCg5edM)jePonh@-5=)#FKg=8kx<t14
zW<4Q1_0DN+DRe<m@130sUC49jQjMhzd0jE2)%o<&>)FA7CMeHhjR#{d%gjoEW~?Yt
zIpVTo5M>Fu5A}me#J=%2(lKexi*$Fw%jFo1$g-kfl|&0+);S77_$6Pgdw15@`$1>P
z$mJSMeTcpAUKoB%sFapjwnse63~?-(I*yJe$rhw<r14m}5q{C<7O;<|yicPagv&V7
zbD0GN`uwJU&TF#Vxh0K`OQ2gM>gaF;e=iG%xmViRB8?t;l2G-v<!SUJOPrS5$b4fw
zbmdf#sM05OVVP0J3r3T<AZf$N>Lc%n!Kg>>%M5Y0(c0+DkLj&onq`P<99z>&SbvpO
zqKr`tCs1$|3Zh|2R%+j3nZ2*n<}G2Jby$CCk-l|*%RHRWA8F$84({i)F=rbH6nBO{
zg6r?IS~e4hqO60O<Z0_v%{GvhDa>w|Z~}21gP&zCmzP#r#a0ZW)5Q&(lyX)P3}*$=
zT1hG<v!z7ImXxaSZOTUJ%j}K1<}ESEoZXe75Dl-wXumDl?(b;HPg;L^EnfB1U<Iul
z_$jA<`&!T95zbjr6Ff)y{Tx$iu|Sw8OqtvUp9L1qLn01*d!0PZ=9(CK<#OQI=9-4z
z_Ed7s3Dq_63ZqZhk#?##)h{JxZNt2l`)4t4jGBHNOjFT<2%&x&YnjY|Juz3I4T(lN
zYKndj`>5#r!c6l*uS(z%jUsDLK()#PT${ju3rS}cbG;=Nxg35W^*llNNceR%iItpT
zP`$o+Zpu+Cn*}7p3DXd6K`!e=eX&kTJTEy6OGTrn+-B2O%a^6b#ttqr?s<Dg<&109
ztwiNUUt1>G4$TteyqgSsNExmru~Pq;tiMclZ7hDtM$SyESy5=Zm4WS=bsx5m$hf0_
zsR>;*@Hte?s(`v1{!2b?)bi8oz_Mjlnc9RZtJk;$DZP=spQUN<+j+kf*p?XkvcYA|
zf4*RJmog6PjdD~o4)a)Jvl7|nyyZ+7z@8R3Uzt#uS&XxMkXGJL31rkbenBIFto*v@
z%1f^r?}!OWp+EK|iYp@4{wd;XBC`a4uCLT4!L_u4q(oSgo+h}ip>!BR-?WU06F*6!
zs+kiCX+)A9rimN2>Bpc5t>l?2b275Vi6jB`lZ4_;DBn;j@x|lO>XPg+J<t~e`nhP9
zEjq_Z{U{tWjkJ330X1H4!?T<$u`!<#=Ez*RREGMiV}*b1lGY!RX(P66sab)4<$t^_
zyh9Js?vV>FT7$!<|NZgphp)ea;$i>9u56^(!vPz{_xwo28mqNuvzVU)d%l3o0{al_
zuw&7Ht%R=wYs1&>fE)^f<*0gS{T^F;g#nFe1}D?#xnrH6X(~KKinX=O_q}Z_Tn$qs
znJOFcDN>QK@sY=uQSmWe?Q`jW%A@}54&S=N7eSSrb3VG9VAv4y0{ehChbkxpZ=prD
zG+FkX26!exrT459XIfi%Ya`o?;E*GdSjy4Bzv^sN=F-y0U7DCyp-`I2ecFBohl#A*
zGT1dmoExB19+T?@K{q2j{Qmkb<Ho^pp>qcu`XWcKyf0aIU7}S=VzU;1Id?C;OORGJ
zm6a_z_UrBDnh@aWzH|*cS+2W5urI5@;P{aUSAa3~w+#rIlEWz=oN+2e24F2P)$$lT
zT1pn^m@{ZU6e29)oF^E{XgAB8y;THu%G+&Yhg=T3U!OU9&rD5uO5VyPk5lY%^dv+L
zQpt;^7AkcUU(v++4&>c`L)$y~uosbIv<?%Wdn@cwBkQ!E$~kZ@xGv<{;^&q?MHiIQ
zc(T$@xtJ!BK26G;8J)A<ckE${p3AOiqo*?kH>n7@94C|_%V-616eCr%p^_A8)FW%0
zTO@-^c-f(}JM?JFsHsP2kS!(Z+!O1sQ7vLLSJ9bpHd44`)CTi^F=x~U1C@>1f7Ac#
zk{;dNMK1)nCEpo1k<jLDh;3HnLdJuFV8?~yu^#MKv2nbo%&F8qI$*c0Vq-W4V3}@T
zvEf)n>0m#Sq+d;9C1(cNNYFkN8^h5Mf*E~Q%AX;mj50YtASIRSzADEN`ni^4!idtt
zLT_y6hSL)dt4ce6+c8{N8?XwT+i7jsJ|NY-YfD9p@>hoXA<7qIsI71;A+nkC1w^z_
zDaqO6OTA$ujXjR6nJHgj#IlvoVc6u2P`=$%t^T}L|4z@>jNjLDCT+nHi<Yl1K5obD
z@_Ine%7i<e(SjZo*`+gDSD#ng&hr8mfrG-Z)4446vWonF<#SmvN-J2nO*XaF!Y#;d
zMqmNicvx|#<2euufDkS!SqtoK99Da5Y%RE3L0G^x9)$Eeotgs&%2B^sYqx_-9|U(Z
z3Jd6B4uJMSH<qxspTuiD%1V?rOS6EI&7sIXC^7S~Vzl`H-UggD%d>!!$pOkfIHhB0
z=H}jZGtUcuSj1zMmUhAdI3F~2;S@B?c-wWDZ03Cd6P<&EeK66Eb7Z{5Ukq&L^RoPX
zcICT>AJ434+Vk%PHgkU&l#pHUE}#TdJ-R*rSYR{vmqGE-bM69(x2g{}{X2n;ye^#J
z_UT;HM-FgbNnz#r?f5;w+~ml}<o<usGV)pqWo~4b732*Nm!a+{6qkJx4#a=p_TP?Q
zcDfuvtHALwR`dVfkGubN^8+y%1*ZG!a~Ipug0kLpMmhs^KmYE2?EaV=oucJI(9+CA
zCuo5d2(e=h%EqpY7Rk9t4nl$w3JP+BT)#=}<3Cbfa(f($F1MqgreD)J&sHF>tZb{S
z5NA-Ffq53j6}ha4xsLs@t+Id8<wJwiFsUzNs!?I^XP{ao^yxwEEsQtk({f&(FPJo>
zw1H^y>o?tfm(DLT*vcxu($LRg8QuNc{TXkqV%W>QMZ=)HwYLoG&Zyu+F*}+d8<fVd
z-o(P4#VdAr+pLTUG`D+`nV<|=F^&q@41OFY{>a%!4HbX&nMcVx&FFu>(bwTEm*};a
zc(spiPK8R5ZuScOqI8VZ+dr{R0-@uRi4Km1h9gENWE;?LhN%L7WS*zHh_TkhLSK-Y
zJ-7J6xzk;j%kzbatzv??UjwUP`EIc!5zAowH7QGSWs8U&3f(AE6KNuOk!M0r<RYqf
z%NfZ5wSA<Tk!3|>AFqFAWUaD%7im|UShzD|6l&f^%}AmH2kfbn@&a25#Cn;OwC}NU
zb_fc=-E}K#O)Mm$m^I)ZAteivNA#=!CqZZLqSB`n2qqpJBZsu<kC+&`=i<vlZ@SDv
ziSxBc{F42AtrT^jsER2TuI$#4icND1d1l!W(cZ*(l~gE<YD9moPi9lhd&#lwd=7@9
zWVFwHdW@QVq)fQPF>p!2>ZvUvD-&R&+!iGmka3p*&T@SvXGKZ5i3g!iqD~_X;?cLu
zO%0Ngau&m!Q!oaxb#jj4MG!WSbBvKk-zDd(O)T7*mV6vHl5><rtR3Xs_OD6MXNe~Q
zZ~Kf9d!OM<s=<GcID4Xf&ognzG4=b9ZTb^ShstQ2t_{-}lx8dnH|-eSz8f1*Y45hV
z8ylRmZg{0l*}X3rSb2>_&N`Tpp;2+Gh_vs#f{3^BlJ6gl2r1G|jI)GhK9RFkIqJab
zx261>vF)s%4*k>geX3{q@`R`zx5|AuBX6H!qx+EfjWmA*rgJO>YwuvdWhoZTd#+%?
z;#jZ|6r4ZA%%PwO6dFc~aFe7?N!S8xD*zQ-NsC~ajewObQ}%z8V^d%s6R*6}tHEtX
zp@Od1H^FX%E=q#9oVD%X(w*QogHQq2<ostFa3vFE0}q$S<~>)ifShk^gN4zG+1_N@
z6boJ%KAV5{Ucp4?jAI*2bWlzY$@h>-v~Z$%pSRC%LXPm8oZkem98$$UuBp!=T-x_v
z!Nm{1yKI1qcglIL_{x2_H1EHHi_6h^8(bVKKlu)FxDZchB@WH|t>C~Zu5JSZlk$A>
z^?rG7lir}|^h;p+n)bR7u-1cB@S+?lw*ib$1sx<-0Y(o4todXW#Lz3g@7L7TCYF7p
zi)PW<hx}+ZtN#a<NT~#u5#$XNmvFxa6qhNRHz>EQ7Y>~X3{)71KfZnXKi~-69hVX0
z4HN-3m(f@%6qkP?4$gmqbl&-&AAft2L{j~3bZ8@#jVI?pAhtxutF*(ew%TankrCtS
zOeG-I5Xvs0Z3y}#1>h$bYfgV5ob~B%46>ENdbdJ88;Re5e6dKPMG8T>^aN?M$SLsY
zN#d+UEUYbS0p$Qc^ZFefT4|vr%%^1tfxb0^hG)ThIwwgX!OnkgRau$eUg&REu>oj!
zDT0a;f&?-BZY#6|31ZvzTHeEZp%-G=+fT=xI%!5;6)T?StRLwFOS?0FJYl9ec7EZW
zKjvPgp+X4_Q&*uh2#GG2wV>{L=@j9*B#A|noE8$bqDnv#M||f)<t@CRcH72LqG?zW
z2&ZH*Hd#j28RCCfhWkQP0-fcGHCjkEqu+^-npXd;CpIb?_tP6snwjKB=P^FfPMCLP
z3S8H=!tM5k<d7i2rFemf`mCMx#G7=0Vbh?@oUybgYwCc#8+XUsB(Rk>TDj8Q-UtVM
zARVRlUa0ykgFZ!?TYc8B`cXbB1ehV~>K`Og7?>rJBmsYt57Skv99a&@yISXw_cB(*
zxSX5ey9&G6CRBS*Hf1BTeMj>0b=6h>S5{T4uAbtxof1d~U&(vDZ@go{?v+k>8P3)X
z(h&~>Ur+L*Co4B+E;L^|L)hx73PMBERb6S}I;zUw{3N6JNhRiIyHPiLE){O@T*+&J
zHe8cQ{&aul@K^WEV@#p93?q9PAasgM#jtC0mJ;_h4f_@yT$jleYql{6qi?n5aW%JQ
z#K28)GDlvfheJeB_NwMrPP&~x&-<I_bbM{Lwm{G<L3SKV9*1d9jSQC#wZH8$agz<u
z`z0DQ3Y$jOQ<@ECi8ct1N8t7pE!&Z0x7cY$x{H5C>OmSay*KcAXNxvCVO(gQu3`}j
zZ0+?%2K4m6CuOd0awLgXff@TVNh}oJ6j)V{l8q4b!(6cKIk3EOf8{pqaZFDQ;a$1r
zDvs)AtL&Cdr&7*Br*UcZHItreub#VK&78Q5mfxWzKVG4<=bz^rTd8Aq4yxJskXtFg
zwMBowWKi7FhwMqUJWLaR$Xdxy{qk<ZDGJnR>2{T$1rX6Roh)@-s=F|TBX^P+FlKdQ
zNhMmSj_J^x2!+p!Pn%q)T+6W0^<N3<#IzTd>=hUOQWJOP#9BgGozf6oI<rekg)fjn
zQW)!k!*3U*E@{>F4NI*wC=ryKDbzJ&`aOS6NdWMc>DnevK4P&EXQ(wWeFxuw%AoXu
zZ{_(53$VJTV_ado8|P|idCB3pl25V%(wCri`eFjo9nDJjz^x-~18LD5)wo-kSKTyv
z$7@htcpVOSy=XF?p*%c7lUvc`I<I(IQq8>uS2qU%o))nrjs=nfo82aVS(dVgIF^5Q
zJuDgQ;Ws6O`ceb>ocfKt9Cfay4wW}mZ#OQKf({j~O&^=bNZ7oQO+}X!BYpik{q^(X
z=|77i0Q!DR&rkpQ^C`UmqlCc%%*h-P{`&p&$J5_usI|2vq)euoe&BD}ZigqoQAaOI
zRCDQ^wS!YtSj|~DLD0ftK3b#BfpdSOL&@Y;Xa+MDd^JJSZv|rR))`{ab+oVI?3Hjs
z9a!t8p5NVsHK1!Vc7&%d43_Dz9)UN?)q#2><|9PIE4Au2<m2#%C|vJeJp%s`qH#U1
zY);Bext@C*Jtm;jhDXvfLNuP`=zixS8eGi2g0Rs$vrByp)_)PdclBRRugrf?Z;=LA
zqCvB+8#Lz^dS7G;2)86-CLpU_w{&RXU~%)Tt2elMf3?y&2r;WOcPuyO{Seo}<+oEF
zi=JA>nQmjx0MAFfXT844|3dY3{?7@9Ns%I|`TUm5-5GdP@IptJ1CI*${r?F(#(EKL
zH@@3ha%_ZAC|DtXQh9*xCh~u|>EK3$AunS-r;(dz9|oL=(H^(;Nl1`5RDWc~4Nrz<
zs}X4QGh{9IZ6c>c#H@2#Z8p~niHgGzcRPMsqNBq~aXRs!>6I>^;<DJZ<Hsu7zV&kF
zh~b!TV6<oqcfDf(n%Vz*7ioHzh~M40&?#v39x25dTs6%eo^SzGOtgO$()SX`u&y&t
z*zKp~gK|zNw?I3@7n5oy75UBYV6Q+Z6&{8C2<-;!FW<vnmR>5yzV=$)Hsww{rof|u
zV*x%vyiS3UJy$MIb>4TsLFp)>4>JY|#{2AhAYv{8!7mW^E<|VBTsSc3*BxhDi{Suf
zXrR7Mjg7)1q?sD~b%=j12>dKE|M_)@-eKLgay-o?nDqHN<Y<JC3qR%CK39-h;L5Ew
z-!Ds}Y>Krv>XFe#SZ`N7y8ql{<>rAZlu{@e;ED?M)mYM+RXx7IKr9Yyd?+OVB7nb@
zl9U%oeL%_VRCUy%4=7v-(Gso()dfDLm>Tq_CDI*vQsP9MfvSI*wRgl8N+KtS+ORgP
zTk66dM*tzKc{gw1S{5;OH!?MU;dOB7%AD2n4JphWP9<1@pM_L;m}hD%LoridcD&P&
zmMg<qd9*7%yA*O#S)bphXR(ZkX}{MF48%H^2W!U&^by)ILcCWy#@(ge$qcNvP7gI<
zrv_O&zN!fh1cra7g{)6BfkNH9qe>Tw)KjIQEolVRTY0Qc_u)Ttpa~f8&6QTl{hBZW
zeS{{A5bxE5argD?m9!$VJLEvFxM@VugwZ#YbQSCFFL((g-7Ki)ghnk%&EhG>T}_Sk
zm#Si;60`-ENk_ksWX!|?U9icTy&cSQkg7ruX6nv;UsZocppQ`15#qh7I_^GH!PoEV
zn^pqcN<Vz6Z%q`&BRuzDQH($zA&L>=y`mU*ABy7gC{axG??v&gj(ioLcKJT6QfApC
zNX?(r-KWm2M~xK?gfKJ+%T&@$ZAOBkVyblhhh4_6T@|FxVNm+`s-QN@TZ7)R#=NYn
zYEp7AymEi-dO#VQ8!A=7?KmLErYSTZyX#|xfe-qdJgi)Itliam4Pf~|pMD)L!}@iB
zxtw+G)2Mc^wyaa=hPLxUnnG&`>-?*snVs4<qq)(3HJw|puGnNbHBZmfU}|!=r~U5(
z-HTXOEwC=<$Tkz$1QnaYYd5)mV9atgZx)>S<#~U-;{blOZdpd+w|t6G*d7zjIscyq
zO_eu6wi`Y-jwEOee7Z{8>1sdEgcYnN{+BmqR+KUWR&JAqReUC~oCDL%W0R;ICvyp_
zX?TO4x^`>Tr?g)B441BRW|__n4wo*4j#YIO?f)ynIEyzAb`bPFC0SwBiR(+l*Dw!j
z#@T<(aZkswjZ1hoT^U;Rn)M-gbXu0*_hjWsRvA4X;f?!$L16YB-Saw{+$Xh)cwswB
zl9l$tyoaywnjnFUHZGTqvCV#11}!w4Z*c{2xFls4+YAnjgOKgYwDU>vwau8LpRmZV
z!@yTQf*bukbYG`#Cw6az>jlmc-D<&YoE0z|C!<VvB<_kMnJ%S|&)xA`4g6LMzx%{f
z=n(qoz;2r8eluPZPWt}&@$_GI_&ej55#$XNm$BhE6qm75Gboq2EDorD!0Y1e^|#Mo
zUU(W3rQcY^kjhLyApIKAHPWp0s|++Kyr%HGc#lXE1Z|k$x^Uq38fgt<1Q|vgC{v1H
z_McOWco#>lY`Eep6#Q%_XvigP`y(JJ<ecUM4u3>Smt-e_Gq6A@=X&EQVBSr}8zii5
zfsRajpdj)P&w-dgU8GolBx+S2duXln=qfK5zJErkawQ-_8uNV^uX+?DGxQs|lXxAj
z^{oQ`^xMEcbsJMr<X%Zao77fiepfWcv1D4)XdX$vW>XTH3*@OgC<)CS6HZPfvCN@a
zy;X_{<;Rx0%Dt$M!O~(MFvvC9hZ^;pQA~cS(r;0;*kviCUTl7UCi)50+c1CRE4klb
z3reco=H6?WF#Y$TeYty2pGEHpm^Z;0gwxOD+sEh2e<qBQ+%s!k_xq2Ra6^G$fZSJ7
zG9^s+*T1^&FaMyh$=z_q@foJWKm7F9pB=gKEqPhbQPSKPtt&4pIEr6&Mhd1;8Ir7&
zA8D$goUi=dPY)-5W|5Ug=}(aZ>MdL2NY24RWs>zH_57gLi+}hmzt{REk&URp2FIf9
zpsiUNDsUqIiwvr+d>i>}Fvxj9ktDB1d<YS3z<CN7l?C!KA3oBcwD-)Q5@hz>ng>OY
z&JT(J3e^s%fq{1uL^;nIgFcO0+B(U5!a(Pk(XKF1CH;bb6jju5hGuGO<TDJ-BdQUj
z$8p;#Vl6Yi_}OSx$=mg@yM2T;G8y57`nokt=;*CCWMI+LmS}RAK8f+kazlF*VwOHB
zWbm1kyt!Z&&hI&SDvpLTW03{{ohUxGy7FyCD8?{byq<(T(R`k}md7Yi0|rLwIR<B8
zZ%2itkRZ8#VoD3rVe!K`yAg-AeK}I}?nG;5T}kxC8UNIW9*TIzL80zwm>@oujUtP!
zej;cad>l7O>zJ4N$`LtW!cMvVK0g{~nSu#NgBG@gS+AK_6K?X4=Fy-PFyZY6e<m7E
z`VU6Qk2@n_nG|}5ZX80Do_Wj(FO+yPvv88~N;4yWh|pJ+@$A~svjGwG6_d*<T|W`B
z%@&jMf*GuwX#Xb!kky_bzAq?gQt7&Nqsf(sNzhK1T#Blz?3vgI>Aun8vTbat#vxQC
zG2V#->P#&Q0LwWjdgt2oF6Uk~+7-czAil3>b9L)hSD&qzpnaV13ks7jvB957;3WNz
z6|X&iRVv9IR_dPDHEur}v9D6bbP_a0Ig@+}l%dj9_9;^im(ts+s`qi$X*G?h5&5mI
zwUPm#%ks`F7a}D;B@Zf(oR)T6Ipq1gg$JxhJ`N?1Nz)a)kk(RuDdU(B6T?3f!+Pzj
zf%$rlRX1)a_*0zO)oad$NaI`fUqKQL4q7mOh2kNgCxVT{EJ#w`ECrhs9ta~F;tVJz
zK~f;mJ`8Y7xrw6b6EVv$k5PD)(WNn_lA2Q{4q|TOk(CS`*M&T?Sa#YC9+k!pT_8-f
zw?Si_XIfpSu*Icd>=pVcKBY$9aw>J}3ueJm{<n{e;t&WKx8~e$XTRCtNbB3^WsXUI
z?#{NmyKDobmyh#CK-5RATLIxNZ!-lC%zCvkn%o+=WWZt4_ab*QT_f7s6EFxRD4%dX
zzP^z071N#s42(gmpK9j?^o3+qx+(Id6KH@&4qj|%%Ml%bQ2eTibuW|*5E2gBmQKf~
z#0>+M@KpgwGrW|;D$p2w;eu>fNettE^;~TACfDTmk;yOK^E=9gjJU5q2<m)BtOOQw
zLN{c@!L0P~ZlldH<;stsoqAJ&@-vbmo~ce#g+USD*SSylBa&O0WoFQ3@V1^1I>k_e
zctN5-5SPl?#D3DcA#cl!r2Vc>tu1o0Eq6?{$2tuSNeKl!pLON5g9?5<?Vw_RYDjld
zg1@uhl;P9Xx8BQ@n)+nA)91?9*+9-X_St~+j$wqK+CWpb17QJ#kQ<{JT6+d^qOr@&
zoym3%(3bKN(i;YlGcFvkfN&0`MlwI*VEHgn<%OUP4toQ4LC^-{nX>-kpiNNcgEnY~
z>AW}|v_V;>jl3vm6SND0HbK09AZSy>^`K4pvmUg;FvFpKae9uSHR2UPyGGm!+I8zQ
zL7Viu5ww>$c`j(L($EsL3E+i6n+Sd*XcIA=2-^Gl%|V-p>BRJ_aSV?O6+X{!>{IwW
zdvRXjrNbi7=_{2j=Ngx5TgfDLNxfter?m}2#Du-whQ)y3mX)nZOb|SO>|`vb^OUfN
zd`43914}b4lno1ciqy;~`l@ZxN+6~is`rf*V=<U5t;<^IfulCc;f)y#%fs?A>AV=W
zyw|eOXHM~7<Y0AWE=<+E^aPeg;Dc-qASe2zm&_VF4lQ}@oF2Wzn_IlC<aFt~?w7%R
zF<i`LJdv8HDSAv_zOuG|rY)^MAoTKmFKF}ZHGTSa)rg~X#Zs<0E*=_0<OFlv&8iq5
z-UO_P1B`f<JZj>O%7Z+8YZ=~>%|>~g<vqGEH|&7eeg~fj^`kd5uHO+N6}L~4K~3aQ
z8IY%MXNE~-@tSusU10~>hq7Yb2ht@O<e~dzD@LSp%)_uoO%J<&RdjYI^aa6ouXI<3
z1d9nUw606ZD8ewV^%Hh*%x`HCcow7c?4B8{sk~5h%Zw1nRwOxXnaKzP4_)le;pZu&
za1z6Pfc*70-u;31e!;e3E!KZy%O9m1-u=7#XV{Xnf$NZ|h-*DlUakT24oe_Zv>tYV
z(BR`-*I|gAlwqNNpaIrlQ474LxN%F-lG+F+(e((6-g%ftDckt&;}}1DlV4}I2BbUL
zZ$!f|)n=sBB5sxtX>W}keeO5OrT*gc$qb@`=6L;9dLrBR`m0M@4AV0vZ(|!%n(xMq
zqb{AH-$dF|6E}B=yk(VaA{zu(89-5qX4!4%1m^PEs>^VH3%wWP7J9MRHW<d9qr=K$
z1`XpGtGv!EyR}}<Wh0|6Ix*U$I^I*#zZEZfbsy_+Y?C(fmdqt@#l07s**6cGZR$PU
zF{8+QAGS=2@$rty9#+>uHE+``YMUzAEQ$>g16FI=JW0fWA!8^7pDyB&6Zu<2jUrY~
zn6C%K+RXlc4Y5iruf2VeSTWJ^PXE8e8h%<Jvi@xlP#;&uA}E#ZZP>t_quH!I?mb3!
zYe#mgcM7;!5E}v(fN8nAK9pRHTj=Fj-@7O2BgZ6qm^%pBP}C<o{K!Dk#YIh(Q%&9R
z0|7!@dBlorpE`1%v$Cv?W~Vx~ieX*4zE4Llm2TL7q@%HT@tTRz9Lx4j&n)cOWV1{d
zJ#s9_DQLX&MwMdv+Y(9i$9;V>lh%3f@P?5WTb2zo=+L?9S0>K#euc}U-<U{pa--j>
z*e#K48Tz=RZf4`UbWES*KE*ODFyX~>Eu&2i2Q8zSNs~zka{eDZ+oH<~Wo~41baG{3
zZ3<<V9pnuN0XLUn)-DqPHIqTvCzp^n4y}Lm^2_^APdqgV=nmEW>-*Dxp4vtUtr!@X
ze*N*}6Ho$^R!j)UEaLRXzoy@w{$506(VSWTjXu9EWLm~arq-J7#>@Qq7e4)y5`Uu1
z(%I?1A?pvHoinH0^zZ2(B%jwPSRfR*)=Di<Angk!Nw{OuCq)#z))nbU(H%(#czJ*M
zqPf?nBHb#&gOtOOKG1(xK2<!&GU)QJkcKgTltR}zN>%5QOQ&~v$vUN~RN|bTo0tZ$
z^?LD@y_x(*FmvuQ$}(RO=?aDgx=K1m&HDfn9fTFJhTDS>R+c2`oe@s>{1r%d5k?F8
z`2kdiwckM%kpaoy7gbbDSWChULur39s@w5^hwv!qzs_>-R2Y+F__tJOUsB3z%OTII
zkYZE0jQId)!^-V|CM5$0JA)>WanK}OGc@+E1?>(zm|-are7cO+aZnDcyaS~WOh|E1
z{}YNSaG$aN{W+iwE4Krhw^~N-Cb!ggXSt=moP+=#WV*RzgxrE;W(Jj<SDSw=vuDp+
zOv1#dklp!bA*-or>K}_l6Osg$>y_ld4_lAxNIc}H=p%R*abg_^j=ma~c!fgYdr4Z9
zRO$QN8&W?Bd_AS#*Kfa(t2AML6?L0!93IB*J|~ooKJ_6sL^;7U%FGGsues@vy(e3H
zw#&=&gMUrGRYoBEf*ZMHMxcL}i>e`gHlD&Kc%*RRGVr-M`&7zu^}w(;4qb;5COYBA
zP{L9qZLv@SKQ$#@KR^lGR4Lq-62s7SD50YVe+(rIXGWB;9@M9#M6Sx|fx?&ztE(ZV
z%MR$~^6_ys0&Qf-Fp?dHs2DCB!w}_|(Pdfy^rvKqeRR-Bo=nN6wzGdjY-fjjgdM{u
zcG!V20y%~qXqn|@dI8L*WXCCGbtB`)07H%%KZcR)Fhn9#diPLjXR)QW!UsEWOeCd*
zX4aN)UEZAPgsZ7ev530TEJ+Tpsvv%CCNdGrY~;bP20EIB7&7cmgq`J<`ts<{?#cwA
zIRj-#I5KC-vQr32bqs%+%Y==M!Dk3=xsaIi*%|E{`M2Rb+`jvIqp&?VeTVcQblx(Q
zxd_>vbols!&D-p&Z!kJHFP)e3t_@QTF?JIl8dJSt6CcKL>+r!u7w;H8Sjo65L8mbF
zq-UAhsV9_DYUn-Z#GZ3IMS6c_r}ggS$F3HyD(Kwp4Ox7o`9^<_U>vy)BXo4@j$wom
zj5}2?==UeNi_LYjMGcHtv9kZyj2K3)!w6-V#e<ffQ3C@xFFm`2Ywr1&MuAJ5q)ubT
z)gnb%*x166BRJ6_b{r=jAjB|m9YSFAl#d~VA}7q4t097?dHv)}1-IPJ>tHo$*3HBw
zCqgw=-rvua6}NxYGnc|knaY{DTySl21bLpxu$%bUnE4Ic_%MuHhYwOQCl8_n)m?zC
zeah@R7E5ud{fWHUOUlJoBb6=@?|1#pYJAw84-vQ{C^5i?QQSIo5HWn(JC@vAZmBQC
zux0o9?ku;|mq)kynEu8!lL#ZTAX^rhQ;&3&?bfY@E>?furgC{%*WQ|0dHG_j@Qc@i
zmkAZSu@4L@^FBz@MNjt-kfc*#&*-rry;ehP$}wBqKNX~LWp+Ta(fK?EB#R7%906(1
zEt5k)34vok4J)(*l!;E^F+dr~P^%%J4%p*50vO5XOTadU2;++F0H&hbb_`%zFvay}
z=~?0}rRaZZcs=L6l8ZNQgfn@c1O0DU?HxGLlI9PJ_@n}+(0sQ3w?C{&?&F<wSo<AR
zrC|q!Z4x<C)($PErD>bm{|FeruIXQ-Df@t@ZgYIxL)o}4H>%1GKq5Nh$3P?;)52{y
z^r0(PQPkUlm))>l4Go8>L_VhFM>MTvL;h-FOh13n8HRPz(KSS$dhaN3Z@HzuJi3?1
zaDy9WU_N-~580g(vL>%vL!OV3RcGQgdpyok&$F>%2)K&%0%zRywsU}uj2PC$eGLoe
z{lbIdY~eIBs)TU$?t!4gY}i%?s<5S5SJR$pU2V<8gqbjrj!t4a2I-%jqa2S;NWa9n
zcaMLx*f{2m>!+i48{NBO^llY1u0&;VYk14u`#P2y8YO4NVN|H^CPF1;+ob3;Am%fu
za;uVumC;eMiB8usN;b%>EDOJC!$trUD^_|wl~CCYT}_t4z(V}S758xkbri2HQ>x>l
z3+8MVyHd4BMf7Tx4<=hC@4K1#urhB1x2J!V14pQaR8MELTszw>BoTWa0A=cY>NI!8
z=rQ_J*Y<h#>GA7}(8;eW&rWH06mc<7qIs_T!p1l?CdkvFcM@FyGEj>XvY2bC^XYTH
zEaTs+(%F;cnpMuF?Ces_WrGJ=oz<!%o4B^+bGm$@t2Md){<hk>EI+t+rKZiQS}}jj
zw!9^`aIW_~grGqvY>2vx5OfhLZ}uCjJU`-i8qe<gjaR8&DSDXO+0<KBtAjYN90Cu>
zUm?u}6C73zv%Y?3Xa!!Os9wTRx!N5{d2>ta_%m&Fxpt4J_*^pWD#fqqo_@z1H`=c6
zGLcvODom-5c&2<Si}HEuK5pVlL#lt%VR)h2x8~RCKOtyjQegW>2tW?L5<348LL;?|
zWSq*7Wt_sdKYpK@C*RW`<oLRwqCrsMM23XHyyvsCE>Zv0!|lVLjsadvwzr0bWcv90
z+tWSm6P)*4lg4Od0`1s5rJSq8&^J1N1YInieE&8Jk`z5r&6k1%U$pHZfk=O*-Zk<d
z9lUG*JfT>Nrd7%^VP;cM2mLuJ<&QU2$zZ!0=>Y3>QHE)9U7Fk__Rae@(c|-YU94na
z+*MsLE#&HXrlF}9wVi=8ZOL2wK*J;y9I*0UX^?uj)|tyosSR7MG`P&IS~MZ;l!e=^
z@Fbo04Tk=b4tqwEGR}pi<Xe9&(ZC!xZEks0kJys1`&-tA`QiErat;l7`OO<|a8-ct
zqhVXUdwxN(V7C-56sbqGqz{rrHLrGA?zt(ad|RdD{Ri4mONk0l#^}Y9?<+$A`B^w`
z2NQDL`czxO)&bNE^p&)m_03aVlE3C(;<*U`)flkSGPLHZUE>3;K+S)gM}H}C4u{q$
z<BxKYZ{KR<14wa-{;PdjA4F;?t&bGIcqeOW4j;oq07LCssXaU-hW|!r+r2CID4wJ1
zs=Xg*tI^Dmq_(Yizpd6W?`t(p9jBh7$Z_5hCgi4F4Xf-k)I!jWz+~SW561$9tM0Z<
z3gE4jqgUv)AeUd(oaPG*N^>Xo`G4DeZb6q3<P8*;k(L4#mqNf4D1RHflH|5A->={g
zEQH1zRaaZJGk2>hWtKNnDY7jkzQ9WS`)-g8l4!m`@*HqD!vVX|?*{<c{MpFn=a<Ll
z%TFKLG+qP?UNxV8G=P6d`~-;)oozgTa2lG=@6B&tq*M;C2VO=>t?}~sJ>cEGy=n+J
z0wP}U_UaAbFAaF9?tiX6@Osks@b;SiZ{f7>`Y#00c5kTn_8%SdD2bQDk;Z77@ZdPW
z=kR>{{qvU>NhC=TUHY1C{MWb7m;bzQ0u2&~2%3L<e@TBB4MGRiC>KPG(ERnU=8u=Z
zaj;|l1wj5u=d3OGXN(B3aV`p-{^{6fp_{BQII92n>(7Qg`5yJ>VV+hC9Rr61ijjQ)
zq5~17pFp(r5IsHt$05qEmm)(BPb#aQqfr7~#!eMd7Q{q@G#s}sWi8S-VU2-$LTnV?
z#trfHqn!~FCR!_>mzzTlD1Qy_I7K6mPf<z%(Q~W|Q&chtFC81L<$E1_vSpRH&j?$A
zFmA%biLGv1)=wW%&czX>qKH0B<{Ud`X-iB?(MTDG(acStWFDTFUp62Pc?fm~__{cM
zjK+9P_<X{-dnK$=6Nn4`CL0#B6{yq05QUwvY$>k9AW+<VbVXp%Qhy08^`^2OfCy3q
zm@sj{X=^~Qq<|Z9h1b?f3+t~mWUVk8++s~XSSumJM1Z6Ty?ad{F{yUO3#0wkDD8wn
zJI|VxY+IAZeRkfsy?@1fNIyEyoSwHylkc9VBx(r*W&svrk0C*~*-{D92kCTHF0YYO
zKp!mwizjnQWr+nRh=1N1tAU_gO629+tCOnzV_w^J5GG8)eB}3R>_SjG)j-U6<dCpT
zJQ!i!-Lw}-l5ASB1qz-pA53w^uUO#tL+mLr_1w{JfZMB%8@`+L-BmlR0xQ6RztX}-
zK6oD^mc=KvKnm}cUx_8@E<kcCttWVOPN_peV&FD)7;3`_R)4TqqBDGF(wbleUL4|5
z4olj>$eov@jY>Ey)9#fd!_t}}F+E{hGDDn9JR)I^<_Fm!L%U&!oa38(Se6|sn!~tv
zv+PPNNp`9C({_(SVlYsAGzmAwLo;G-OJT^O*e^_mcQDsYTzIKPvmV9hw>0%3OfNIq
zBb!uTziA5Yoqq-$?~HVH9s>&8JviTq1o|4Fcv1k20)cXfD(5R~s+0DQLCC=0UJvXf
z_h8N4CPoGMI=6UT!fDwV_B&n9Pg@=NIGzk+=BV{#4ocQ1!%1O`pYXtRR3(DI&C50`
zaB6}$cd-JOK(A2X65>$>F5kX7s#EKOVL21kyACC1<3WN~q|Qb;`Ult_)#dBL=5_gk
zDAr>AHMVD@uu)GG>5T26YRW17W832eJDMo<YAiu7p8PsvpC*VPhnm#&QhidFe?|^H
z0f?8qMh<8J6PG4O4srp)mzhTn+kc)KKzMn2_N4Le^o+{45w(Th^n9!)c!QCc^(hx`
zVx&{T9$na)S>d~m*>w4xOs{6n2GdOt8gAmb1&qUfC~Q|2<yu|<-8Efiba%wV#J>mK
z4SkP8DzM4NT6#Cwh?OsrdlMfqpK&@KX_HNs#3uYTq?cKT<B0E`em_Tg{eK+x>~8Y7
zni(5RR+t4wbrX|S6t0ix*DPJN&ccis8c7I|DRdUflZ4?$PULYwxvLqp!Fa60jn<dJ
z?dN+j+<v}k3-Edtazars)$y*cj(=R;zEcOPX<z4QT9;0l$Spio(TO1SIL(<>_<h*T
zmSjhv;X!UHyD-N>aCUDybAOzAYB#tTx-{e-E(R-tF>~>D+A285&7EAVrgnpizDq3b
z;i5Mp1Uo)D^Z%&Q``GcIxR7Iqr)T-O9|iK9Tt3Ii(oR`b^K?U2E(j;%nZoTbh<&wc
z0&CX6F4d~MC~noth9)EyIz4wHOChKj`o@Gd)EzOj24~AP*}Q(&mVf-Ilai7;6ArG?
zb#LUBRz9ZF<C6Va@78%T1?UgsaGuQMvtMZj-KtG@GHhsD^do90g~OD>K<S(66Sjid
zJR_5zxXJsfTN@?(W_0umNzfswhjqcF<((*1*DRX-r)XYkp}pfgL=JWQnQi*?bB!fM
z0~P6{UDwhY(QAj(0e`H$XTS-ofX|{|Z9I7X&KaSn$L|0$e-q65@w+_@x7MH*I(^hV
zg9rjfi=C0!0nERuNyl&b@L#tw$02WZ<_uxft1C+wPz8*ZJtxbV)n^PD<b~zt+=*tR
z_1v=RqdW51np3|#TfG;CVO-i2((QBM)7w}bZj0arhPx<~9)BN@%AMd=YUH@GBy+ES
zW2w1jQv=u_lvl-4@2G}vXJN>AJL8|1pA+CF&^9+Sw7Pp{_6DnY&{xgAgtlY$^?R`R
z%;D4)pS-n-K?S|AxuZM0$>}{IhqMMZEMcCNLkVq%9KP1efzA9PIh0Vh<iO}XAqRx0
zW4V2HJHMab%zx3Hzh)4(Ws!V#EB8mSKzGv4!3d+>#api!)IHv+dg~R(^=(}v%F#7`
zhPGYp#7mo`INl1o>W)!!fyr}bMLVb&RCZRA@1W+SpHM%kg&7uZuTsAf>b>g6aD81@
zzjAa9-mHEM?}qv@yw7U&+ea>K<+=cER=E<=(h+pp%YVf2R#+RIs@h2}9p5>RH=q0E
z^nRS%s0=n0J`<>V_3SUFz23~D5*DuB-&+E`)yNsDbsH}SpD^(f(2~lNg)^+x2ELC}
zvTug;YqfbZQqO67K4Zy%E~JWN=#1|9;t~N{d*>cz%neufq;1xzv#l3)|MvOvKXN1H
zX$oa-WS90;Fb)GVGB}qZnimt7JX9+vm$ys~qkn^sI{f<a!wW}|I14s@jUVH$FCQ;|
zzl;JJEtyf%Z{J?xd#wahUQR+eroEp2`D6O}^5aOgh0hCsk6UX@2cHD$fT*>Bso0d^
zGa!WlrZvD|00sl_=}Z33PgxM=6ig)W&p&b2?#)l%R~f8efxB@Qju7fN(>pw{K-%JL
zaew9z-yL2pk4un#Y2?G_-r_yB%0#P!mu8lL8m#ba<_nE8tMkJnmuKgvBtoX)60ecb
z*oB|tlUSNTZ9G%3K2L_apC$5XP9(JCE3gR<Y6Tl^q*7tK2FK~P>^jnQ&z8NPCCPr?
zF8e@bUH0>B*_CD5d7`IW9%+)op{yE)lz%@-*6vL!tJchikti6n-pE*&Q4)+J!R}2*
zFfbrGQj`jpWngs#&1s-4#?fV@$vcG>I>J|WnZ$RJW~l8Z&6;-Ae6!I`+l-y`hQTuv
zl_%|3G$X+?u4iyZPmz3p&9*;zj`;4J#Zea9<GYYOkaWx1Aq_Jk?5rI<(LJ=Irx&<Y
zJNnVx_UCm|UDt#ao4Z{TtY+YuXP5C#4oH6?T?`@5@mUmC2@?aZfk;v4DE4JDX9_Hk
zQhr<jKCI{lf>A@X8%5O^U;&WAFRv!)D_0C0<IW1(6Wv2&dx~3CwjX^v{P~RHNx`(8
z%fC(Yqygi^XEJ|M^@`~mkbGSCiX$GRy<hoEkj!eeuksB<cTm28;zs2gM&FJ<cTj)6
zQGQassCaQ%j`;3)Q~3reJ1O6Y<mZ)dl=iyv@wmDigqwI-`R)x+|Lzj_twh^^XR(m5
z(^?wT0t#_rpe#1eT4oWP!OBlSjyYM<-84fCv3otcS`;8E+B45xN9BYo$#eLY!ZmWx
z8lAVw`dH$zrJ96*PtklX);tu1-8z4`S<ypErXZ=uq6d+vfcZ3^hwlq@6yX?r0Lfo|
z<<sxb_5<{VnWleJ&u?WKKK(WQ9E&O|bRjiJbTw=qd5bWgBnx0)GTBPPsF1RY%_Q#R
z46G|Wx5>hi2Q$*WRX7rYNuz)`?8{K<ZVJQrTx&>W8_MG=i``lm;UDJmw)B4*wzmNj
z{1oYRFwCnO`&hKdvv9Vcku2bSh&;~u3L;>Fj5m8D60N=i;^0DE^;M<j#J`mK$$8=3
z`@B9Q^qn|8EYAu~U_`C7Pr}J57DV*}U@x6s!iwHU?Ge1<)%5qSVy5oI>|xnfFf)N!
z>z;&}1>@HgsXNRXY+@YrPC9=j9>eEhIacrim+H8G5<W(-&{USsML)!5`5)$UMSkU&
zdzR_ZF=?;G<@ZmB|0tU)(#z~DpW}-ZM1xIzd8dPw>!~OHK4F`~cN%}6FjFwtX77al
zVT@L=7ui)m2YUf|UQ^gyIBEIcaeP?b6&!hX@Xx^!CJ|Q5EZ7o`V>W*p@b{^~?bNiL
z5ZCzK-HGADvaVq$nD_cKxx!`=_hm<sto0A2M`Ie6mfQF#FfApwONaK#Y9?)Q==n%K
zsrYM5zwx9!9S|9`UV$nIba)g`+C7$$jcRNh9XGu%9XF-*U%OMFv`i;~ZE@*@R4m;i
zD4DyfoW>s6qSB24b3}hG$~b*&@^Z}l3!gsa>NEpZiOS&_)?Faz<;us^G1l8yg$i8e
z`oLmKTl)Z4>4hC7l!U5V<|!^t@+X%oXCt7Spl}0Ks-Jy^U&tzgkTZS>^CKd2LG93l
zpQA&_Q}%d>sKs$f#)C{}2bOxcx7k{0)R$zvX8NK$$Q{&N?hb#gYucu{Va|zuv`{C~
z^5p`k77(FGe+BEb@os6hH0j*ynr7>}G&}cdHM{Oa>U-=DeUn6&Y}s>fZR-(mnu$i5
zI(yj<d4y@l#}l_?9Te$qY)Q9q3`*7P{`g(ox%=VM1lh(@8|=%g3i0;O(us9KDz{Pv
zO`}BVZTeLx6!d>=qpO}UZ>>UuY#oVpy3U4bLc8#RbWPPcXwhZ%+%M~7Ilf1pI*kan
zJ%uCgY=6)nf5)3w5CgOoBSE~|Rx5~q-ya8e7YEAhS9$dPNh_|f42Ie9HcEqpsrAjo
zhb`8@Hmj81m?is?)+3DeTzlYBH|Y7{_RIN(<)@ZGuYiB=Z0ia~W~;syZ&udsgtMM=
zZI^NbT)#y`oo7<m>vqN6Z&wyD>}eF6o~=S?D$}xD9l`!Z%i>&L4(Vndc@FY;F=52s
zs_Ahv@^Ft>hAhENZ7aPR{`867>UJmS7J-(c;h*-q+IR+zZCKV)^tbK71(~MEj(tOw
zmhC~;4Mu-)3_9AGT3HKmVo3{h=e}`^_)djMU5}{7u5;{VD@qE&WZpvdt-7oPY`*U4
zpe4MuJEPtAdA~ZW|B}Z^>q(be%B|Drtu$U*K&!BLbpXQfc)~FV-g<Wr&dFJ|<hdl|
z2twAlmCl$5lyg0Is$MnuBC<+f@cXeW-8~)?!u)>}lk5>M`?{=2P3-k(uG^WHVPE%F
zvy<^q6k?Eb^W9t_Eo>7zo>O@V8|wDe+<Ol9QX+NbR1r_=M1Xy2Yvu1s#;|to7w?up
z(?CtWXNu`&GZoN2aS}>Adq{?x)5XPT<vzYOST1Eoin6nk`Hoh}X!x+=JILdpQTsjR
z=_!B4dt-%~YA!hJz&qHPn7Qr9drvXyj-mJ${h_+!;+LeH;n-S>gJ=u^g=V>!NvTLK
z7A2L9SMKzFP#MES{YQg6pz0|q)tE82uZn8TlI`A)_nxA%3t@V*tZ9jNlt(FuOVpn7
z^b|*VqAXX}XUV>=x#|#a_?UW&^q_egcQ${!=D`}4g8;&X@qjuZL?}`8<Rt~1<yQ`w
zo5;?KF=eMDrY#w@jyKbx3;v<K^t9A7mAw@`WxVTMJcaTpENEsZQjl$E6Xq(TsQkw|
zqi&5(D~rGYDUVW=s4wwN5+yD%gi807Q)24h>Xl%{X-vE%Q{Uo04_v5K9S>H$HAh=B
zO$uY&Vw`L>Q#J03<kb<4pliruO{)>GM6V^RE>TtujB9Qw{X!Bzaa@V?()<?hMWd(#
z521wKB|I;&nyj@^!JE8JG}%y%>RQ#qym-8~ho2C^{Rgjl@i~_f<P8)RIW{0LAa7!7
z3Oqb7Ol59obZ8(mHJ7nP1}B%-Rt~R!;Q9NvZytEQQjfo}z!1xVzq&o^fF&S&>BDat
z;Q(R<ui;a_#XD(J)&(3i-V|Qione<2`KQgNKshDuL7cVBtHO!*dcF-$$~w>);_}>N
zXn3C1&r)3T38t>VbHTUgE!egAw!52Kn}P#=8(@8E&yG;-#?4K}S3gI-5e!9t*=sWe
z(j5Yf=g7IZ%2rHNetOTCWXf}-qT=eBZy_^>?E1z%r@&{b>y}oD2T@j5GGZj-P8Jzy
zxmKV=Y~o$xZA8K(NxaTj7YWMWHdJfYO?aBl)>5b!QPvkOOwUu@IX=r@k>(<ww|f2t
zH}Z-hH;!Ud_z%l@SxTpO7|<Gjt;di{08D_dmI#W(MxH~JR$cJS6ezqWnptQQY38Ut
z1i1+`O%f=JjR-h%-pFI171g&zqYI@2mEpS5JegN4yBl6cDUd`1W!5+HP11g%p?5-)
z%8e|aknWrA*lJkaPxKz{i4Kc4%ASXj9z+K1Vj1FX1-XM*-AM0pgcG}eGL#G@neb5c
z*EBqGi`~fjt5-TC^Pa|z7c6G7D1b>*Eru%gP<1Cp{z4-`((7XH3wF1Rt*uX?{I<5z
z%bH4!_pgYEXWmEytp}IVG0qLa_3>X%Y0KE@OQo5@N8R<nHU{{0F(FCR^tGr$VW%D#
zJ9%x%QAJBs0ud*pE2A=hfr@YlYSuEWstkH~($t_Qb#4rL#>MmwM;8IL3(RDy53>Wk
z5;Ia!cKm0P9pEi9-2Gqd7)mdfOrd0S`a=B@@pp?o-?91bm)rbzAOrDL>(|Bjl;l~8
z2xU#>qm%3EiH)+?0>w5qs(8C?&%O3>cIx+gEzS^<z0P8TvcbH6RZbHb0L45P$TP#N
z7F2AE8E-+cL2)5lzC+SHZ35DYIp^rEWG+6vJ^cG&jSw*F?d30DAHMybU;c^@zqxRx
zyu182F8z8zvE^u%`Q@LNKRtX-*~aXajybO{+T%|vj@T;D9*w@}g?VDIwF9Osv<AD{
zk!+}en~r_J?Sy@Q`N<?$N>6)X$>&2h`^DKTPh5*nR%yG0bm3XBS|EM3Q5Dv`g}lB)
z8wv=Hiqq(+4q7+Uo{ZX#_T*J^^;dQY4U)SaKG_1}WA?Z{<CgKx(OiPHzCMy>C<azC
z5UyI%Ax5a3cFFI`Upo8v$ex$F2I})$X8FB-(?wO?x2~Rl;W8y(7RwE$h)1@pcF0D5
zkd0~kt`W4YER_oqlDsydBq0_LC<QaD>P2%Xo&E~XxXaAt!AZ%Cvu<Vmr8+L^Dry9g
zJ-1hFSWX!rdUDT%!Egp>kD9cvgzTCc_$A%#O@GhgvCF%?zUI5?Lf@Ej<C;mLB=Ws&
z>xRKBVSY`2`0kAxyb(v16(_qI+oa0(t}I~BxydRaIjKw-ka7smd->cVgJilAU)2i}
z&$%M6ZP~u$q-(}BAzUI;OQQXJI_}kr!0UxkT~@V_cn}+T8mYw_-bUC|g)$4_B0gOf
z8=i}g<BFFEj#XZCh!)l25TUYQwFC1eyWY{($TGu!O;N4(t|ik3yW)q)cChXy2B}zG
zSMEBA9H;e?`HeCSz@xhtHDZo+!fhIKJCZ!M(o{3bTpfs>iucu@A#aJgB+u2!gUdF3
z=K!ieRlimiadfqoDa`(@=nh3q>?qP-YZ9eI!}lVO#WYb_t+zKzJ{)(OsYw7Z7%-|~
zt=5{x9n^|Of2ikJJF0+K+i0|$XmmO?X5)8q5~PYKNi~ymw2ObPhP=6RKEpS4wo(;t
zVVliV5rvO=F9<yzf7*=R6^e*)frP-cPeroFAD%X$NY2m;Qz*ZGdMAvjJFOk=RG7*t
zy79wPxcj5n28la_v^~g8Sei4fscbyTZ75o!XEEoXe}pfAZc*D%3@{IR3}u|MDC053
zGQFHvPty88B6)pcvdH4J4q&$;?&9>@RrN2|1{+nPONph&WvIsrm{4x9Q}axC+a91$
zORj6h_ca9~OT}8%PEGg=X0&wmY0~s9Nf6dBEj)q!`17VDge@GCR<h;CO?Aa+m}$)3
z0X?r3e}by(q#@@FQ($|d(4a$Isd+9Ul3Xn8nx;x9Q=OeC37~A;Z!L0`K!`X!)s(|C
z)HRlaB2aXR_qt~{77gd{7n;eFxE7W&Y+7AYx&c-&E!Dsz*XQ?oo9%)da15~YcKxo8
z8MP*{EJbQl9MQ_4NVY)tL<_TG!+nS5L@8B}e?u`2DN2Q6RnbMX4>Sru`^?cofb`9{
zIg;p~S0wklSq@hURZbB6cbuwJ9_bOT49rP&gp@LPIO}^VZJFwHgFRE7C|Kk--};<j
zC~!=3MJcd-o#;wIuCCM8Ehiq;Xoi6&f%Wr_L)8o<g|%ZwgG2Sgj2Se6Vio2(>WmmD
zf9sf+_7oH+nFOFvc7g-vm=SBqX}S}c>#bhtrS;f^uCGO2qAm^n0nwhl@}72Wc`xbm
zQehZc2`eJaHVbX5=H5H$_^K@K#zsz}cp*=$XR5@ze=ISc8~R-2yuzqUp}Y1u+g)Nj
zwYkNh1mu8biH#>v&^==wNw$0to6)C0f7W|$2+{Up4%~ulp<8xh4lkM4h?(UKlsqEF
z!sBI(r%-lmCdZ{loq>YU43pheez$F9F+q~xdMQ(Fh@2$pKxs9t+-+i8(!_S6Z(=W!
zdsfcNX|(Z`1qO!Z#RKE6BiW~?Ero&5Ch^fQd6%Pte>}CS3#huGU2{UEKv))Bf9i3K
zDm#uV6kwVP(w|Z&KwmE;oKYy0Rc5R{qfnqM`ua1Hj%Jua@QTOvDddX6O@NF!4W5Cw
zl-1h_J!pj7>-mRmD8}Md^eJhvVIbk0xRfm?<ONeJH@3gDkLeZTCK%lykz%Wli8-1T
zT^n=K<1c3(7b1`LO`Vb=jV4xTe>fvWnj7Z8oq~ckR&<_xCxp>LW8SZNv#f8l>X=7C
zRvf9ZgHchmehrS)Gd2F?;*g&dXbA#GP$?8Y6ebH1yfFc?Zj%fqCvtMV)%O!^JuxQ8
zj`S>Mvj<!1E|p-?cSPAXk5ZvlfjvXVI7|d$Ovf1%jDbT`(?Zz}1uBCve^+zHt@=qz
zzvtFhwbMi_nOvzw3)i;lf%Yzt-sz0{#>_`w%Of>$(&sqP99+s4Zrt2E-dP_8>k{=<
z)x=*7rbJ2SDxCZ=H)5r4@ssXK>+tED!i_r86?f7RiriKCn5ii3mG;W|(XVu}zAf{N
zp_&+RQoQfk+!{{HVP{d+e{Oz)yfa0b*KbadB4-H|rZORkOvpN^+ci|<&yQWR^+qtR
z(o->PJ9p_DJ+0d+jykyvN0_Eb)ik++*aL@>o2tG(^Q=TRY87$lY8h#)TbHx+=7s5H
z6J$E!GOKIS=^ki^JUT@msZgbH*?Ni|;&L<#$XKXFgyUIl(>RJre=;|5a>B_r&#7ZC
z`Y;8W+`4DK^cZY4Wy%{Jg>71TiE}pKlAF=*Uxgl8^e|P1|6;^aC$|k=OyI$WZ>IPX
zqX>t)Bt~*l=5X;ho3Wq7kk_Hhkr+W3=-W6GC^o*|A`M5^lLo{d|Ge$kFep`4#quC1
zeIzKw3=at)>6V=&e@6PyMG-CW_M9|^pC}<@$BvIimJ0+gF=sGdA1ZBO^a+?Iwm^33
zXDDC7&49RB2UK&Qy#MWX3r~2=i50=?!zNRd0*o^J8EHXx7u%~Rq$M;<705HvqC)y=
zhQiXeMV&c?I%pP@`~>E>BNU7X9$&UG4=mO*IiJ;i1+!e7e-S9mJrxSl8L7p@#|n7{
z2xh6k4F&RLr|OHe<y1Ak<7zv_4fS5>J2*dlITZ9tp_sO2mS~A<lL}1oBh&%F6WWvZ
zaTXAALsSD1UYruga1Qr#qQ`kRNh+Pl9q)E<^nAy^4`=aCc3F6s!00Ht(`5<PPQs9#
z_thl(Np#Qwe{)Lp$1giQAM}X9kWy?Wxr6Y!TA;xk9|$psw!3eab%nXjJ2B}=-CbHC
zqP&?D*H>t=4bLV}BJ<nau8YU>suElcPLXW`+mJhvvC3v*xP-of(;jafq0IYtai<0C
zRb$=^@3N-cjhy<jydBnXV;Mpna^8h#!wuiMx)F6+e=GI9$*5g3X22K9RVBBpxz4gn
zk#`fDpqos7k_>;zq{}pUH3fDyCE8R3y1;wYv9E7gruSJ?o?^2b)0FK@U>WLPDi$@X
z8_&g3Ja*)oPMYoo?qMbXt-nwPO@urfyN8#ZFcKB{rIG`&dx$G7SJ6+D_c``rcW+`S
zDxm4If9}NY-o{Z^AvNRvP(c}`j3FQBmYvvLp=0!(Qg5&QjaxJT-LljE7M5}4af@b3
zyQ4#YN*fAgClHbb?>_-Djgm1&quk$#-j!t@I=ntS?Ts@4J{n(T_-2&Sm{Nt?WBlSx
zma*rZ=oABOM}d6)=}>frh6E|r{MQ4&AUe~Se*!yl(&OiyFw2D4vxEIBM_Elxry8i2
z@c`2-4x7!Wxocy#=}Xz$@}roNlu$eRreu88jBNRFJ2FBc4Y5u~+(~5QBm-IBIFmlr
zOR9@$$H4FHe=yL!-mo6MAG!s*`Ko0*Fg<hwWq`c={PgzlKLWG@(+Xv7WOH<KWnpa!
zmwe3>370T)9u$|rATTI@ty(*e<Tet{=T|872F+!WRb&y^YoNQmvw$<-O<=%w<QERO
z4Q><r=esD8YQEH-St-zHS|VBXef1z&qMrVk^z`ZJ`ttPt+2Z6C6L>Sd{62xa(DE0w
zJaja90Oc%9FQ2C$KWJ^7TvxeJYis55@;}R+&L;~jEg;AR=aWZ&kY5($GRxgO%XLft
z^Z69t&*RK}{f9InEj`R-eq%#!rR8$c646qWL-{)Y{PNwCPLSR|15H5%XB?JzL^)-M
zCr7Bjn2ZMn3yL)Sb{^&Zvopiq)+sWs>ix}M?@||eu7bBrRTnf$S~?QB)-65G?;Z+4
zRspdV@_m*vHj`n0uXLcv8CFjF42xVeg53RVjAaDRLp^m+NFo(M>7=L4dh$?Djmktl
zEi3I8*ZAg#+{KQCpb9KJffEb~+JE0Y)=S`Os=NFP9y4gPG7TGdGmSMfB$@^3;uS!n
znH#&3HG<pLQhpLA&3OFSx<=6~sOTjo(sQqmFHiq}y6Y5w1qdse6$PGt`TX?b&wBb*
zUVb-eW#!SIbJNcgD(_>r|4sjS`hJP$Eo$<ZMg_n{{FlcbY$Eg0LtM=8t323RMILJ*
ze?2BsXpy+Cy9k-|m`oBf#VM*@r2j!VVUyfM5;DQRPB%dgzUuJ`)x08Wt4>)l$v@Ge
zUR-gIYROrD1#-`+R$>=%iDAU^qb#tFfZ`91*Ac>bPL(7HjwH#-U?0+WG3{}Fg0&=h
z8dEb$swT+4<}tV$$Iib@ev6Wf@i~qwzs0Zhv$R~^UP@{6C@w#NP9sS9D?Xjajcq^H
zQWx#!H}hz#&X=TjNN-mCXYF{=aK>^eHEgiz)CLNF-D1Dnszq2Bo1b#0MW~q}o%_Y9
zZ+;T)JY~%mW+6{)_I`<;(fXaWB@Z#!HRmTyE6JQVxyC8h%^EBd?;|ai$yg@Vx#g+}
zu5*bNw~zB}ynSkZRhcZSDkCb1N=an#Ia*Mb&1AHdxsj(YzfHeO_G(5g*C-6)_S!kB
z5|qq;a1skRiJcUL)$M0@4oOjzrt6<Ce@xAje}B5`YOHd=?Px|<>x;FukPosVXj@mI
z6hkn%D<+y3M|rbu9+i_s-$={mXm#_lkTCQ$y6J1}YBDRb<Qfb?EEV!d_T$?TNNSSV
z4_TH&cl{;oAT3p97mLdkS7G`ZmKq_PeOImpYl^P$)t9em4xoSe?8CS}6bE=p9Q#92
zc4lP!X70+cct2r{q?Ryu3}KBFDkS(r2m?^$xFgIJ!GgP>tb{d^d<sKINk3kGKVglO
zOqfflW<OzCE2l$6m@Pu`mj0*xgf)`5$ds2P<p9DODcSItw@!NrvmBJ8;pCVXMVH<A
zz7dia57{7i$SQxe5kiH4>mf{u;az}=Dl8(z9z#_lrCdkspSV`88Yx(ibx5YS&MR+C
z7l)T-!E)IhhpnVC)4Ozq-Ela=lOvrEA&i{z9=lA>#gwu$LuSGnNoC@)d6~DLumPTO
zEo{fz`LqeQNQz+%X4HLUD7UmiAGtl-^;F(yMuKlQiXnfpU14Zj$B#XRmQOcH(>Lir
zUUHXxtqu-U(7wZdNuQ4V*Xh7kfe)|MfxSYDoepmKNH%{syNxZEN?F!ear|}hq;I2o
zTlT<>dsTrmUV8Cl7?dY^^eI7F3xm?8_?EEs(a9+g!x$!I5qfi<Pu6O#*gu~Pd*6|9
z?mm7PWW|4$W{qRH{(;7TLH07Ze*WlDJ87WdAk>m7y!<sXa#3Qxmy~mDEiDUA;351|
z6EbFPaxcs_&9*rbd`6qMaP$Zu1;skDCd(^ZWT}AyiC>9Ojs7Tp<5Gn>@#7ld34$}*
z)xDFXT->h9b&!Qn=kF^@+J-DiSO;@3<VO6s)>nT)?u&FQ<Z=C@gDg4Qn>)>J8?x+x
z!5(I1FS)~a;<qbyCvb?{rM8nqq^>MIhHK)du{*_n8?FHbe6)od@#A{iYT<FA?5!4t
z>ibYCAPgaE$s^|IIBvv`t7er`_vNlGj{RfEI!>eat#sIg48be$oP(XZ5x-sgs+_tn
zT6KR;eHp?9mes4iQ+(Nm8>eb;$HHyJZ&$81;kpDDwdra_W60jAfOm?o+mO9g*1^G$
z8}Zw<rxo(JXw*S=YZ1AF>`;YJ1kyHSSp%a!+RBai?dnlw<-Y8*LaypH4P;q3gOR1b
zRod1x+ilo!jYDv3`f61e7lEpb-Is4VE8l+xsVhN_=dZZlOmDBeWIk3H+`*6=6~@J!
z;UslW%U$Zajp3Rg8_c@uy}jxJLyGX|SRSY_F2@Yh*S(IiTDdBKjA4Vw_O+F_S6%>C
z5!1m|9;nb2RqA=UuY`0~Zpv@Xy*&>StCD`LWW-Z3$mjXg30;@1PduBmoI}Z{hn803
zrSZ(qa;7Go1>$s|VQo>KCQ482dbVXb?UT<9!Tm4<u2N>1(lVmQG<rJuw7i4dYCh*<
z@|l?p*@Kfz3CXP*i<2HMpVr}xlSobNq1g5R4Z5k)(wA|~6chn4mx2626qkTHCnuLc
zj|ByPxDDO+SLg?tLl6WXRkvJayKhg`?2`0uYF3lXY;qTwO@2QhMM@MUSrTn!!Is;G
zK=J?t4*^Mbe|EC_`R(@g_R}Zo&I(6hRrmEr2mFKNzaaTbL!AW>2BG`<(f#^ON@cj+
zxM?X7xmo_~_)3>|1jiEKxxwY#YT*A6xT*essy?~h(`R>i55Gq}@m~8aX{7!Rb&bEE
z&r*_`VM(o#LO+z6x%~F^%bQG7lERztF<k2ZzJI;_=dGTgF({;U-S0o%!Veljpi><f
z;n8$|{j2-q?QgZ#5<UkEpEQPO0v|}mRI_$3F#OcRM?7H7a1p~suK%AH4oZ^{nr|O}
z{`~c4SAFyMpU3fQEu61AP;!FCAWSIGyINZFFf?04`@YFr8iw)QpA4NU>W4A;hKKRf
zr<aQpw^|qrbx*;1TBnIq-z(iFV_k!ieL;Y?wMKbM;}7FzZSei_ZX^`5;%-`iRhmuD
zc7BZPA&o?W<jfgBFWO3x7nWw0mG|R+GcitZoAt{M2dbJ)yTz3Q!E5@37Ds6W(L;;l
z_rqwQLP3@`tZ+;thY*}8{HPR03`zEAMun4a9zctzgmg2<i4g<>7>9?x3VDbodImYP
zJ3VSIK&qWik0C(OY|+Wj<7PIH;`$@2jM7E`%oZE_?YQ8ULO<{oitFOmqFso8g*e+<
zI2<T&d2h)oMvz_X&++Gl?Fx;sGV1*GH`)D@fBc1oLva6Z6#X&uBfEch{|KAchGNI=
z5ES$B0D7cYc1n2Ob9%L-3PLY<RUTKdb7V?!I53aZ@_y^7*86aS299gIconhZ<LI6}
ztMEQDi?pLG$1-l;d0z;7Co>y=Qlm`dKzwDgG&o){#fh^F;5E;313jK7@%H&*^umtD
z;Q}~pxFTF|!?eSM>%F=LY?$qVlaa!eqB@C1O2_RQ#4qnm{ameCrxrw9zHKMU&1h_(
z%p}F2do{|;1Pq5ziNY&0pqz0q1rEvMz(vSxjjm5HWpUf1iwx|Ns4wh)b{m8CCc4ci
zJVuwC=RoSu@nP~Dl8jySVVyo-)3rZu4F0?1l(%$^K0G2U7f{~mv+c9rK-D_IhrV+_
zb<A9(TMZ0(h3C~C{c|}mWIy!jmg_t?WO0|YnIhcWcY@ps#s<d5@)Cw~;sRNSW3S^v
z4yD(^6=^vxP!1bOEXIg`d0fbjO{Mmoz&0bWfh-x}<XP@(n40@l@1+FpEurQfdvCeD
z+0vuqEvCH17Ggt0*$H$rAR9pIo*|v%!r_!Krk7@nz$LUXq0JE$l*A>YxYT?RQdS5A
z7Q~^vb>}wMj>rbm*t4l~TpFE-IldL2zE*J)2M47(66AnkN2Bn6sY}(=1Pd{9AQCsD
zuz{}XS=2e`D$8d&Tt#<)s#x2!?8~|^fKBdyTPG5>(tadjG$XQsH1v$>9HhbU!A-9s
zeJfZ+hR-d>=9)kEPP!N|azLTNetsv)&1h_(EPIM|PF$8GlsAji;##2Wd%mzlEljaN
z=A(yOkH=YS-u-%iinkBpW+WaXtbNa~PC;0EEfDSE&UdYj%@2GyljwU2wm6LzHEX#g
zu^cIOBHWI|2Er~}pUF8L)vzzird_x~OO~xp=k2JmX(CoVEVfWn(;?UGa!rjt#rjyx
zIbojw)~3%j#m_cZ&|=L)rJ_>S6S7nHtc&QP;9FgWy7a$)LDmy9T&Bog9jvNpWrTHj
zQdLbO3Q2yfsvWPM@hU`vwWhW5SxEG=?dpF0Xo^srz-*F~2FxJl&(UNJl|FVxjk8$;
zERU6!s&Q^MR++}Cd*=w_U9U#IvYcj>Af+~tKBO@foN7E)%4v>wGNzN2bSN1<pW#Q%
zUMg+v=6n)=BiESVXZ!Gc?duc#=mSOd&jj1UDt3;{F{L{xzmLuf(qkgZ^wc9Rh0o;v
zaxAHl9diDD8fRlvrPt>Cn<u`3r{(=b!a4jKtz+l*i@b%3)G-$6B3I>2>Xy0s&T~uq
z{5EhiOv&^~DJ3*k7+i!6i?lHoIXhkC^|8oqT&r_`_|AiM`}8(o3$<H^C&9+CG(1nT
zMw3l}hpQ@1wm5m*i{r&@f9}0~Zd;fIB~Mjb53AU>{{*%0SnUg`FSAWy$NXPrCk3IM
zlVOvBu}|`;AUoV^N@-et3?=J&lIAMjT>$J!G*|g>E*MfBsGDhSrD=oa*8QGDYb}2R
zrtWQjjn<~;$|~v78MY|x^K?&1pCjn2=$<l4SPieId*LxMB77#o99FS&WSSWSoA4CA
zrwD{|s^okB|7SVcXSjhJ1`csj3R`KuQQ>MS=2ozgm6RWr^{^t_WMZaUFNy1-SdeR$
zI}@wzXl$UY0_8X<eXW$|z@t}r;5vM`B<_oU`bvJr!>*&93kA0#bGWmP+m6Zx+6;^Z
zcM{sX|M4<jMSI!%R?%LDZxL-@teD$S#o32+Ga_3^^H!stN&p;Iv2zn-wS+IDoaDL?
ze{DE!%_%3?P(T_I@9Lvx3i15{4DT02V(5z#zw9MqzM$klG0)mY9K%se>6%D*FN}JB
zo;QNF0>@%x&I#+BEk6EGEui56Lalj?NA|BKRKhqLMo-zoD<L7N9pS{7f;l|Al=G~5
zT+NV@@ET|18~Pvfo$>mSC?yRPat3&SXRJw*8A+&C>Hm9?a7xMfNLZIp$wbvyo+^(J
zh9}1--%N;p8i}HKD9i@J<#UpzJen1M1M;y(&$O+M_cN2LK%^QT%^&PH?eB?4^9LJO
z8b8{z&gnIjTBBLcI!i4yK`pK2@%(u#AI}U6&!x~=n^iLd11?V{a27i=&ogcc3=f)#
zo2ulNA60$po(nAx-)yQhZo4Yx<qi2;^=*7ZrNKbe0Ip9M2ZT0$rt#ykai-FL4aow9
zPH1QM_M2Xls-!~%c*Zd#3T<ua7{d`C#<W9>WqLFq*%jyD69N;9O6!g|W>We<Ch1$-
z3s~8b)BfS}xy6fGC{#18x_snt<<+uqNS?vbZZ6Nf4cv>ZHC4L9SPkXhv^`W^J+Wp}
zyss3lES)}LOl4alPjauy3h)w3@0u#@OE@cmpwD1cQ(tCE4g?iN=?%&i-@|%LlIjjD
zJ8j9s;%%R4HBX59$j$OfA9BTk^xx~QFZrp72}N6?==GIP0RIC!<43-iam^GI0Wp_R
zhc6VD@M1P6mq3pN1b-_j4&B#R<PQ`LLcAg>D#G@?z7f6anc8*qs$ZMAi|Nh${s<5x
z`EVspJMSVZ?Sz!!An8bglEdGF9KOArKVE*iY8;&KMmZQh{u&fDNcyWJJ+x+UN(rmg
z@bP{4^-W5!be-u!DYd4{(PvM0^Zu%pryeTMrQTm1D*C1sU4M}8aHZ?QpZ)u5Y)^91
zyY7u8n*0WwQi~3ym!!*5Peg6v?-;(^`)?mVzet{vqPuGDUq4>{`!b&tdMScc!|&f;
zVgqU=w1+{3Af&<JpFfAcUVf<}ifAc({6v4ht;n>_3J)&I{Pj0E{6URBQ)eCg@W0gc
zd;J|b{Ac(#{eQrgU@v46tZ4<pXM*JbBKW{!M-fKTgfQGMqH|;xg!P|GRfM#ei%ojx
z73rO{)|iRjk-mY^B-_vu`J#97%loU95Ost!(2#y0X_FycqqUUA#MY=!s`iVTW_0eG
z?yM)xj3&BKvD!PS4Xlh34%E}+*5N)`ZXGX5;Vc}qxPORbXgAiDJPKmIo4B^NbKQ0i
zcmoIs!b|;ZAVM0f)mq3<y=(M+NuO<=d(v@NbQ@Xo6DlgVASj%uy~0koc28>qF6D(s
zJP0lTp)^#uTrq#TNRRN6fhp>fzDK}RfzYfgSix>|9#<wIo5?3pSA0ei5wfDD_V_Gh
z3WQz<vws_*4df#cd2<l*WVy=eio9$T)n+{L6rWXP4ZJQ9e_3!%*q=*3C@ZpX=V>ka
z(A+)i&(So&itss|KY*r`f(*AhK(kJ3MbnZiuzY#G+<2F!53Bui+Q7<Zu6{PRJzH)a
zuQNi*a16H@=LH12BsX9?Ngj&56E?nRw4I4f_<uED7Ujuvwv^k1A#LOXx?ei=V>d1v
zc-utxhnSMFPH0`@#F#-OP%caf(J@`O+`qPJ=KqUN$83Qym;}Q-P%Kk32XIQGJN^G|
zG&VpsIg~gCWn+bQp=N=MT_3vWmHlz(?uKClUd*A#G4P@hIEFRR^rH@cVHa;37Z4_n
zB7c-M;Ps#u0@Y6=!pNqI?=d-#Ucu!nikDyMvmO)aO@wzNv4OD8A;vKXYZPeJh5~m1
z4>y3H$B+r*h`VpFKj&)~?z{2Xz#Vch@^toovfMge8Hi}jLF}zS7}P7N>52I_>c`jH
zDOr<|bSAF*=kqyE7;+$Tj6F#ujIR;wGJkrlHsu;J_LJDfj#fqy$}8E4T{j@h1{95l
zM5$ALk(#mrbppK~kPV=H3_%V`6)2Aq7HX{|(q!USGsZd73RnWWQNv|2yq@yLK(V)Y
zuz$u|NTSgAP^t575q;I&C{}$g_EvRwUPkUHx+4WK&QI$coHgen`&iUmhPt_wUVqpr
zKB?Sh*J~sBjnpGdet{J$KS^)e(QZ?SXKU*vx756y3@yS*w{Np=>m_-xLIEsF@|3Y^
zTPpH1sXbOi3u=(WL0T&YbcMe2ppjmyfByM<m_7N|w{hGgQGuOHR%*5zQF>GSOhcUY
zPHynrBxq~X>)O%K=!FL}hmXQ`<A43W4Goka0L=3@G<4i%+43HrTlw80I!ET|?he|~
z(1F0R)srVy<1$(@R9{GPtnOxyuER7OEn$fHnp3Hy#W0$73N&5H{=3U)`)9s^sB?m-
z?AeGKVWgu$N2{!*mJ}~=e30S>&Ch%jgkLz0aLwn?AB69p&jwDG+}}a5nt!%IAdW6d
z^3)Yi{lrjt{(R=onuCHf-%I{-N+DN@HA)oLSzy{d%MI|15g^q;N@l!3UrQeqnUYms
z_!(!^388|V=l1q_FI2ndw1pLUbp;PO-+M*$@l6y<5dEMKOM?;??KFX@q0ng(Y}p)a
z70-;Wv`E)_ok{4V%C}P7iho-s@;V8oma=@!XjWyVVwvX?7?*s`f=^DA5b@FmWfZ(Y
zvD0aZxb$)*bW`KV>f2$=HBz2Y#%$-ID20vrOvcXO4}vCjF$V>II_|dE8ciEJM#71B
z%M@XA%6A8e&Wk`H)NBMG*MmgLs!%V-y9icG;CT0VKsmD9#dzdtGJl@CB!UqR7GsuQ
zR%Zwr1!$P})_DI+5QIb3`@1HFU<+WM5_tm#VeCS8figIy_C?;{q>$b}Kk^3R--oj6
zI|X!(il{E2IwtA{6DW%r>D#On!WG}K34q%8@Z7Y!<mtO8eE+<*@FI-OLll0uh)%^N
zvCjM`3J0KM+E)snQh%zePb#tb_p}cc&A!lit?>Kyy$Nl)S_jE=H?NmGM0_eBxlel#
z@lpIJh;6!v&&;Qrrg}f&(G!aBXTN-l@R%Ow<Ge+9=6;$WlW}q}o<|6eDReIJ2JtXG
z&S&@(<{>{Ow0tV_#P#!V0P`s8DBOO;W!)>Gb6my>>kr}{MSt8AY*^GMO4_(~l?<T0
z40F|~zfQ4#UR!u+VS|2XA@3E@sk$T<!}GYu_`wSkoYiaGqZBoI)1HTGldG6J#KK6=
zj<UwYE4nG{IX*wfJ8HRBUwf|A{^R}2YIQ%V%NCQ;P=y+69FzS9Sx48hVrYJ&9HPlR
z%d88lUTez@R)0-gsu~WKuB4%|q~G;q=cyD+V{7O*E5$O(j4&5g`Q7%+=p4y98b3R?
z$nSodDV9`UzBL71J5uX7pDJ%L!o@C(?2clj+r&HR#as`(r}V6D>ZU3P@);h4w{dX%
zjjPsmV<@%gc`YOUFGz6ZM7}D0j2`zVTeD~ZM5mi-On?5>vPel4&h4)zLmT^@xZkxg
zB8Hqt;Idhbb88P-^8XCf*7}wm<&9%t1-#V$=V_0!N6RucDl07Gk~ieWZ{u9cPC<Ed
zb)>R6I(lARaAyUK$GR#+cB(yPfs1}YK-NUJp@|U8W;~r$1d~>4N7|jt0dD@xwK&Rm
zfHyw#lxcrvRs-Xwjq9rJxbC;qUZuu|Qw@X=8V!VC-<oYyRo*VZnRfcvLQvd!oVIP(
z<~lt9n&;e_eg4fJdf=JdPY;WKWkR80bORdzHYL+?^x#~dy4;dWtG?@u)pQ@Za6jKE
z*!e$!(fd1>am^GImtk-X6qf;j4$J{?x6XhLkqLhV^%kSqXo@h6`4pG8*L)5$WP{Nn
zr901Iv{1~H?iuK2UqSb;<dZ_)dL^8;M&=tJ;<A&8hl$-_BA%sDpV=397{tD<kN{S!
zgI(GR8TVZFd}Xbf=XUmUSF!g+`{A3uCbggsv#E8^yli<B@+yih)fXy7_d?~~Pbna@
zDd9<;zV}Rxjn$8101JljJ!wxoy*9BtNiy$8j-~hW?Q3nU<WDok*FZC`T=y}wYj(OD
zu3smwu(oh+E4+-jjQh`YMhp8tsF}r1mvPM$6ag}qF)cR~mr#oi%K-|P9E=WQe*r(9
zQUIZ!UVi-;*quLryB*MZ5j1iRw5;ScL3<~_EclHYf%lQ`$S2y(d}!%~la^y98(HH+
zD;;IbfZ;r{k;J4%G0Ag6FdaaXKFNkHdSM7dIly%kW)dP~EcA|;1&t%NbT~hLPZWrN
zx~@}zC@MGcmN7!c1n=8bg=B-le{;-CRZ=)ZKUMi9VXW@VF<K>pbR46M&ShtEe9Qb3
z-%5VQJYQd|jE<WORA=MxGblY`mbk87XN7Z1zTVd*ZL9$2D@o5R0OmX-UG8uax}z+o
zw2=G`x2rL#Zw<FILb<@9i&cVK%BBPsjRIAeBzReeQXtTF^k!wWP}(xYe+_3=bSgvf
zu?9|t#GN_%+j(nze2tcP3z11y**+ioeVBc}<j?AcDf>TR(|vrL@|N)6m@E#sxQRgn
zxhA5AH+;`fOo$ne{ouViWMB7!7|oS`bm$ETwGyDY6dTS3<RW%yeskYyc#SlIWbzER
zsWCtDn;Lp<&xsG0a~Bo|e_D;A7__ipj+Y?KIE|i)L=RQY4L84w`C-i1kNs}+IW^N$
z;=^Z5ZhQHmg9sE^8IO%JNE<J}I}RN=&9)+ij6^WL$V|JsL$trV2f0=hYE*(!14kZe
z|H>@2Dobs>pliwEarSa@UO7Cf%`V&I-v>cH#)v4`9TRjg!i6QqfAyfCgBK)n4!v$|
zqbHO=nCcug-cPn%Dke8&1(j0q=#4FBp_0RC{EV|ChR)dy=dPN=S6#p+;`17|zb|1B
z4g)_z%a@9w*3i9tiS$RU<O2IfXDZgfc|6RYkyaqOwbOTWCZniN(hDgZ@TS-gqr(Np
zb#F4!s0{~xf-^OGe=&>}X>Ymx+^HXP!lL2W5A|bKD7)Y{^<xh*VylzNZ8gW(2s+?}
zms7umm%ehUvG<TujWNne-wlRS-QVcvTT(3fp3!Sa(a<~iCv^S{bBE5KVZ759lpdJ>
zq9Frqh4)k6IaQ9XOcM6XpvK%o1~o>RM~fRK9K8N*SrG_re-^T7Nl>8(w{l$0-2Zwj
z=(`Lz#!*B~@Q7UVkg!G=kAQD^NO!q8WlLDf39UF+Em=vR>Rmk&7M7XTx1FO>5SFMg
zYDcC(d9K4SDt~wQ%AgxEfdE1(T}2_EXL&iPg4-R-8TKPG$g!MZY)!&)6ym4DS>NsQ
zXF1E3+3>JEe<=gfg+ySJvLm(KgxP5!!`|9+(kH{xfxkRISYYm2dnyP*!9ofLYR{4h
zIL3y>6pO8)`xKwc*-+6I<rb3ft&-X?3(DH5F|`&mac&qEWIsE^X^mmhfyepj?9BZ;
zM$!#x{mHYIo{mA!I>v=IkaCoBJtuX=s%y-xNnNqne;T^f@af9Dl&RgmarfBVo7T8H
z7WC%2Lp|ffhaVR1tA`!-HtX0kw&Aqp-LVaW-g)lQoO9ZZ!Fde#N;l4(H6U~KBloNZ
zD5_H_#JRES+~h?9gO-=Rsm%nc(r&E|D=Z$Rwt|Ify{W{T0!=B_Y`rq<w)Y;3xr%mk
z>phlfe-*5;wZlf9dhEe93a6P&T3hfuw65j%mW$nzz|3qCh_}MHu?7ILc>}82!p)Wm
z(SFxwk^41JQ*4FM+@&z+6dun{SqNl3)byGQfw8q?(A5xg18va4!9p_fjg?G?R}7rp
zeJh5=p}VaZR>awV#lY?F_=<sJyuV`L9@l#+f4V)<#BgWa(lN`mo^S#+>I;*-G-t9A
zK0ZP<>)@)*yU9njvvI+c2DS&A9KspZBYkU#n<39Jwz`lg9WWdkXGtW5*=w_(Hy#9N
z^Mt7)!`w=Od0NJxr?P7M^p;^t(_7N8#_=&eHW=DOq3C8w!Mx>%1l=JDWjwV;!H!et
zfAj$4r9I%vL7wZiy%IFe2(b91Ee6BhVLHz+7TL6)2b+py`*1ACaW^DpsjrmwN&5_j
zy@$kVjKw2Ts1LKW;PQ-fN&~m{VvUD%hKh%B=Uxtu&(PSwc`|Y4;WnSNnS*l)%TIoJ
zxQpSxXjlLhItcCwdkBWP!ybZROxnNOe+=!}K@@p2TI(~>)N*?N;ORj_IIRv(cn*qT
ze()R=gGR>3_6G-xM8^S`ubzV2)uqn+l<85!%dDo`-lyc%Iy-wKhnpxV<2iJR8asv&
zTtj$w7{Tf87=q--89CkcQIv_1dSMvyGS}q6rI3lG2G~z3fb=aCVh06)q_RPVEQKKA
z|5ccHt<$IU8UKajxFX6^lN;&B;0>^<k<vD<{R;(uzkZzl2d2v_6qj+$6chn6m*A8G
z6_-ep4#t1W{Kx&5lddN5^X-1xnqs_BLHp^?_tV#tijZktUT~T>e?6g@G|5j~{POcV
zr0Mma)2|}e-uSK;zU-yD?GrE?V^r1xv$ZoB=@44)X9CtC><)qOh=eB~0?vAb$`>h|
zSwZx1{VK99mNP&j+?m9|UO+^bdL)|ZfpC1Jr~!YcqE1^y_42{4faZnmE+0kx^=8BR
zI8j?>)~`GR8|K8+z^k`)6gI8A^Pta#K^s_1g#{Vtmhwc|dWJ}aBDtPQ`(u?fgb-^L
z_`Dg&3O^ge?h4gKrk-E5cFpF<Hq)1_I^jO|*6Fo041P>IjZ)y&$Y4Y$(yoN$>a895
zK&!V5l@9a*0|g?Wmyx0!69O|cmr#o}Etk*&0}HoMm=4*om(uVs5&<%o5mrGcmt;y0
ztO1Xg5latY1M~x#w{%MnYzhQwtqFLS5#$XN0Wy~{iAE%szIrnY0Wz1tKp-cVP*x8F
ze;mmX{?4!HkC!h!UHwws9i1%s?Bil7$sus0OAtcKGQgf;hwKirPf_yQ^HlH7!gzsw
z;YGG3pk}7JtGeo`w;4NYNg3rV1!c=$RumivT-oI1PE<~Qtj)?>A)Hge6`r61TOn9+
z7G^w8Htm3iR4$L4Rb-58Z4)S8?m-B}e<T#{SuRvi<P3rGxaUczAj%Zk-hhfC)03hY
z2qLf*r6^LpQ<QTS$}3A&WOy7%F0l(v;yB_#R@g}HI}l!W3j`_!p{&A>$ind<IYUCg
zZamWjt2mT>vC-_TIG)W`o?>%n%39B$icUeymgQstzi>XGprX%Z1q{N0A%nr3e_$;!
zol9IchCbP_Q23R-Y{FQTvzU?L2G|T#Bshc$Ujh|mEo|gn6jfqS-gDf6Exq!gxFA|k
z#RafF9?H%M_hRvMWKda*k&N;=iOPvoFy)ZgG#D=TKv6Mw79^N}V<e%9NwHN3#b`<3
z5k)Q(56PFEvtlFI!jah&q-CLsf7omllN03=07bEg2#Otvh#VD*`#^b6MHsYT7B35Q
zIJ!b%mnbf=Fh3`g<(^m30Cu2Q#bQ4A2^5zVPDN|r4F_gy0s}dmAbM9Q3rc+AOj0pm
z8SY^hc19?yh-GJ(IYwWef-l>$Hh>am1O6(RXduISm1uU4K*5Hfl1j3vf5Q@$d|8X=
zO1|g<brX}UfXD<PO2Q>2Di0BINqp~cFB>vDhb5ei+MmTUh@+h!f2=<Kaa~t$$K$50
z4sK3bx%%s9{7F?mHIvhNa%jM5j;g;@AF6kUG`rZZhwY)uti@13w{`*-OB&~sDAAPd
zQT0|~`-AG2=3`U6S3e-2f4s?1kH}=WEs+1y8~l5c5<UbiQ8?Ej7_Sp7{*VrVU6ug$
zfMRe7C|#)VUhI>^EY)xS{1-3)=aU-WTwWd(235UlE}O~0^<Y>lSL|5r4cfMzjFp{T
z{`{r=<)9t3wW8UL-Dcck>s<hl3?e`_-$m?@<R)Cac~k98n&Cm+e;!tQ?{}+@^_TYO
zjz#tc=QYHR+j`tiB^#Qu#eO|)ZYINeI!m8&>u>exXz)|>rC5pun{xWS!2~OjakwZn
zv+#75tBZI6h3!CLxg+A!(Sb14NQJ?&4pK!@&`=^c_=)3^Ux>3sg3K|+Y><d^;P7;C
zP_()f3i4V~7Sfw8e}QVfCg)yB3vZlufK#W1X9h<3VJ(dJ$-97@=aGU#K%n<fii+Ez
zlnnMlIp{-mnJa<|X%|kAa`&>Ju69w{i01Nlv(#vlwt{<Hg1ZZAz#X2O!+UNE$KQ5b
z`@D2eAJ;$VZCnSwjcae_;q0j+Y)80`@EsBE3Hkf64Sy=?e<G~l53L6xEupTW%hC}&
zj!#zml#h0Lu~NNVY(H#A%^0Bn$NmTT`Qf5%ucv>ms(P$Hk3NmA<=DEJoL6#H{k<7q
zj>h$1a(q2$&L@MbtI_xzF|qT+&_wTqbR^9g>4YA_N2gKIi8PAaQeaN&qP2GV844FA
zqq8)45wVGGe+ETKlXRyk!g4PyJUA`gYW6xHAjE5#1<O{me^4|(r)V6zBt;G|1Ns;!
z;2x12o=-AU`GeTeXpWE&kd^rn(XN+#eOf2^_PA!rmyst{bD=5HXpbJ@4Ay8y;_=Ym
zKf33FlQN)X*WpC_B-6=iJ)IQtbr>x<ZIExCC*k%ge@id>qQ{^-@85PYk64*5XBUhq
zie?SaZj@dz#Q76xUPHGAb0f<CoDHeW%*)R%MjUKC7s?{1Ovf8%d3H-#e%_aDnB_U%
z$?{#(Hapa8u$&WlPOy1~e$Ho&of+ZcjaVmZnXOsUSk8}x4?jcXLUu|JZljiQAMGok
zkIBwOf0;lY5uWc6zPhTRbZQwU5EWjKfHNAITAk4tUN$dk=cD%G=0p#htLki0pH4=f
z>Q56{4a@`N`uEHGYS|ikCZ{tx$xcyENk_YdqU}xFOh$vtX*KRDZ8aTTU0>G6^_PMC
zoo=sW0AdK!EHZvB^C)~LFE!vxN6*zE>^IYTe|&;y{B(SFGcIdy)s@u5_Td)rYlcNK
zW(RwwH8?%Vq~X`#Bu3%m7imCy$+6dvKcNSY%OfYJZF+ECHMKmO9qyd=NnLsLOGl4(
zo1;g!0(*BD7Io7z8UgxlX0^CH!o;mcm}GB@TjZ^#66dTmCzeo~5Ne)cc@1EZCPT2B
zf2W+!pFivKn@MvrDmnbL8Qxq$`9WdN+EiVVto}W`J{}$q^+kJi+0Q?5t-dkd(?I!v
zM9D7>gkIF#LB*y(NNEWOiMIhlzVnVMbpvba2G-QQwaR8IjLd7iLl-TRUsvNDS`F{g
zw^!pkMY|fmv7-0Y`0HlUo;8=FW^47(e@kP2LmwT1QuzV^vzE^-DQJpmt5~yoH2tI2
z=P=vwLf5A^)qno?-|67usy?mc2{Y$vJ9WGF73jq=@L2gigCP*90fWD_?lshvFi&ws
zLAb-Pn79&VGo;kjoHk&LkA4kKqyik<e>b@$^HOffew&J%!)rAoy|MC<xwJQ9f0pT!
z?;AxCS!VVXr#JF$Xh>(4ySltpSC{h&B<FixLr2z;UIcahw*WoU_pE)_raf!-J#;(k
zf90WvTJ4Sf5We<<e0AL15Ar2*b$ik0nx}vybtQf;DZr6gw%9tzb>zOX*b3y{Wmp&s
zjU_q%noaq}_F4L7;wY?dpGEuSe<&~7FZqVORgaZ1Kck0d>$R6}(IIO>gt7D5`0!k_
z|IcS<qu~grI6j-e5Tkn1tG15#wUYf4GJ#!sIp{2HgKKMF_gmLfrbcJ;@~PuCXmxap
z*6xG0v;36NfSl9h^I|k0Z*@%W+Zzo!MZ3{pV@2;94dhi?-5w9&t4VX&JkznDY@Kno
z&oxh>DE{KDSA72kQ2zGo$?q^3mVCIP@*l6vLvRHLK>F@6)VifM)H%NmYCHY@54eXn
zq?d8c6chn8m%-pV6PGYw57YrYw*_Di?g<1*!r-Qtam^GIm+pHH442L1HwXeXHJ4Fp
z7bt(ra_lw^-TN!_17-+<;G^o6tL(P>R%JIC?^0Q1CfUqgOg8iT13o2@QcEqXvrs=o
z3M3ElA^^$Z&p{5~pWZ&7e!8L_tZ)QY4WEAuzzvc=LGn#Qg9Q)<Vfg$p{Q6BwW%zpI
zOG}B!m*wA%zv<-}!Eq1pe8J1J)xiG{_)>rATV46O<lpY)IkZPQ@mc#VQ>65SdW|jU
zOE1Zn;htI}g>kqa@N58Ie*66SN#-R1f5YVa<MZjiPwk=%8G%v{zyEj&4H`k9Q-eam
z%N_pu_wdKlKPrVnSqc|^@#ps?lGX^PY#{mTZ*uq-H~!3>(YxV)vFndIj2!+m{F8qN
z*dpwNjD(Rlz~3S)2PM4one0e{&<m;k{UmnI%qU^pZK@!Jr<QDFI|s~m(iox{+cCQ-
zLZ(f^Gx5xJ`t{}6NX5V8TlmKpGaW(xW>(P5SlWbMfvq$%A(bA?BxF+$QsG{qMX&j8
zXniYpLDzPf>Bsi;wC3OPl2~qn&WwM}D(*)C+Q6N>Fdn|d5%-eVDFjSbdv!zzT4~#%
znvAkSX;pXx^Db#uR8iO~x@0cUc|R|SBf@`uEs7`ZvQ9h`d(QZUFH>oTR(es9%Upf_
zlBg2~s+pb~xir!XwC#+}NYzZoO8p~O>r6Y-2yLuf6Zd%&w}QXEH~$uu9w~n%n7Lb}
z=UF!}M`WefP~9W2%nYOlBT4WcRDrb_ew##=-twe1KUP2f`g3UR{PKOAH!1<jw3OjJ
zTr-_7nV0!-8E7a<RA+aX>Q7fM7w50JFH$=S`n+LfGgRF6>8PuS7$d-_mZH2?wPNqb
z!aCs<JsJyVg~o0#Rb{|k>t}zY1bEx6|8n=L6)QX|I}T8?vG}<+usVPaqr`oD<0Ffq
zynh~A33g{X^jGFV!GyJ)$0^)DywNPUxH0^F6rJF!qo|h`WR+m$(TTV`wyBN;U)|Vh
zd2Cis*0(IFK5&Ahu{;Rl1A)gA!%y!_#V9K)@{Fj~m7#F2240*akQIN1eU^r5WIQ=~
z4N;w3iz0~5@suB;e3DF(8dov9b3<&E)It-q<EL5Q%_FN~oyDse1?V^zDiE4=Q%muE
znG+NVtGd05R$oq0?M)B=SIIY+D&b??oB6TjAi1iRP#QsG+b7Db$RE}V$o`M(*zOBz
z0ome8*DY>EEg<qC3SEE4c2l)*Kr|il9O!J(oKuy_8>zhlknCfUO&&+tqGjP-e9{2h
zd|R3F9KkNWNkjc)!3M(Up*N#^L9i?84aH@_2JYph?Jyh#dl7MBZ$`Rxe{&SCQ8gSC
z`r>^xW8M&Y@W3ZO*FF(v=JL$!afhI&;yLxNI0Qu%Hx%`4Ros8<!2!<-wNR)wR?_-m
z?LI6h`lHBZSLZ}ztr6DRy~sM&iwHfkGcG>CbN~fKaN;@<yg&sz*NF-RSz27@<&cab
z>q_$2jlb+xqV@u0x6c(>c^p|Q1zLS3vQ`VD++Ji^S2>3#$R_e>5IyXaP4px>{Sg9n
zCrPGeonsuFh^v2AUU)P+apiL{pK4bbwlPx0nL#g<_T3TqfLo;a#OC2_w`flF??183
z$A8p`Wj@LoCzgid#))O#d;f`r<e4XyhJ$+N<u2t=PBje@E6O2oF{rlp39e8*vfP}<
zeH8PtrlGi^Zj0GzIQf9OEhecMcg27O3GM~RBy|KIOCf*a8n?S0R@Yt$Ph(gevYL4<
z!RnYcsiE)KhLeza4zknxG!#!=aK2YGv(Pxqg5{<>oqvCHbYsxIpP?(sAT>z{*&4Rj
z^@*%ZgSsQvCjg3w-y3w-e7Mrqb(uzP_4MVSJ7Xs-j4{TWTOpI$S<~^CMGL0M3rl0v
zVgsfq%_e`h$vgv{>{7_Y%twELZ`}lzvDlT>RSd57E!5z>ki3FJSzXN=g=Egp>T1RA
z#8z(bb`I}u5<6!G$R3V=Oj=ii;Vr6XeED9C8%`O^m$Ys%kWl-@Tm(*8eP1f}_^LM0
zk}3rI&mvo=doI~BTXyH8xp>IlqSt(;^DBF6$HaeXR1Rj}z8#<O1`>{mpgJ1~R+dIF
zapV(QHK3H)IqSv9;<oyJW^SB9i&mg#Twk1zO|%ZLY6C69hkJ_;Ldz;)%-lWBMIF_E
z!BSej#zU6*t6(?kj%%(gGO3~{)+2>`mt?pCpT_oU%KwJCCbd()97kvad2&Lhr2PYs
zHz0qk92r$k7`2p0WKU{RRzn$oIX##Z-!9)-)aDZ}Y75Q7tKL9UTRwNvgU}>)#`KtX
zre&*EIE{)gYw~?Jamd0^k9J~pcugBvVKUWco3^v**7@>joA5#9XvpU*r>M7W%1qBI
zxx8qii$>YG?K)4(+vujjkf8)f?=M0-j>~@r-jK}wV`K?}|6Mb2p)Rx(`IRGuzHaG%
zR&R{~C`w%?5e&|xqHG9?Xks=3xlk4N8i2!aY``p|FYy?Zr4fFfM=1qx+vsWs=I=xI
zC<?d1RX({M4?$Oji1--%4a?Snhu`OwWs#@-OjG!>2YElZg+X<VOBgBvL-}+59n6=v
zaStJXTOSf^hTZ4O<$JFOVLuGd2JALDBWGLhv+36PhB%YckFnm$ppaHAoJ?O!WoL>h
zt{urHCz74#*5m8hz)2@Z<T2tzO28OS?7i1%9YwXpZsxm~QOejud1m=A%X%va&m~_j
zUEx7)$~tW)(1!up02-4A@))2s2pFTPiY8H5((*3DEC5#DQgCCAZ>dJTQ&+(%?6k4R
zSG<L!7S4Vt)Ok0FebL=O*4!q0;_C`ZDxuYAt^A8s-{$d+`0Btlhsx{A3G-}5dBq@Z
zyIsz&*av1bj25W6m$7mW6n|BP09S$Bc-LoF$$rSa>5lRo>M>VYv+VCnF9}RC9Vr!8
zWcVX*WtQ|1TB+vbj(jfhc}{k7jXD*^w?o=obguYro~nn)FBB=}f;=!@s4*2q<Ev9t
zZ0F9+$i)eBg4pcE0pv;<-<<FYVBk;ypXZuoHM3|7Nj2*!eufzAiGROw;N6$PkaJ2+
z68gTH%v9ysQhdKzAky^4qxF+~2TX^{%GhO_X1#OCe_v{oA=g&mBoo@Z@^?=<zsA#0
zUSpTC#Ch+c1tv-PcH*JjfaYc4*F1TuneqD(50(vD>RrKy#Z-0_KO)81onIE@VNNkd
zIXiC<Cf;BykD{9eGJh@2<sIhE=xYeByk$3g7t|gHYEHca>I?d@0N2ca%AscD(9M^t
zQsq=bZ3Qym&RC;pCmEp1>T~KF2B^X?e`=+=8)PV(uxFa7>^yiS2*&!ohI6(|QzpU)
zm@^Gie$#PEb`8cLDq(Vyt}8}q#RqG?*~j}ooZi=)3T19&bC-6_6b}J5m%&si6PLU)
zE-06bbPuI};OVc|-*>uhNvr76*L0hIy?@<(-OU?xIw%ux|M!<&dVv;{4TkrvJa7Bk
zKl@L+kF(Ssz5ooLc8+L(&#-hx)VZi^dR5>f-UTuY#2X}r4(pwGGsv677rq!yboIF@
z?}%a33_D%Q9K4(odhq^qaM~QpfFw8zp(^B7kWt8gI<=2)pZC*~KRz^23_2BgkCWO)
zPZ)d4Q=xujiNlj#B9S7EvqVkBsu_#~#I<fX>FtaDnub-I#2A#uxRRJ4Dmz{ofr7HU
zt2Z_R>0fMwQSG-d!YDBrA%#`amt48-fgV6{;I1S*03E$T>)#yo_=`aw)qV@mM~NBq
zD0-HEjiybv2)KUV^CyW9j9{HYAKuIeLFW8oBaCXlg%L)H*$4@MOWk0o5#S$0Kvc9A
zbj>hqVsCh9h%!>(Nl;~6L&ot(Ap-Ksk)5Sa5Vg_HDR3)Y^ySCb_51|fED5&30*}gh
zSgrfxE&;(R@bTGoLy8Pst-BE+pqcdWNI*q@T|TP{Zk2nvGmV%@go&isDVRfi-6_ut
zr`PRErWR%H*ByIVeu(wXFW>$4g-5>pn}|jo_i?qggRFH}7Rv;L!agX9-q|hwW*pNS
z-!_=IR|?LQPTBjIO(o7IuAkr7^m{<}E=M$M8;TsZQ3VN!%boZEYn12t!{E#vvo5xO
zu1+-*>iQe1o|F;}otHex6o{Xu<dgZr0CPZ$zZL97_C_a@pReb~nxbwsm*udRw-_bk
ziDarN0z(_)>KN!yLs_RhOYE0S;;4x!=h5djzh>2C<?VH-t|OW@j+EE#waN3U26!%#
zjcFQO`|%lNtSPTb%sUEQexe*J2kW#TE-Uyke}crI)~je54{M`sXH}@X%`88!8l_)Y
z@}WH0-c??^((RQe8k6#kpLMZP<Amf~s+4ViHpt|L>&GO8(cE-s@qT#P_1nl+e6q!3
z6wkQU9b?tG+Nb#J>A*MfYphno**6N(Wu%m~v`T%#u>r@i%J(>rp3ro=E*_p|o)^wA
ze|u)tv5Td;a<^mArS7o$SQ-I(vb`4_sB)yz9Sl-S%!4}?)FcL-OWu?SZjL11Jg<zr
zHpy#SObCPC_I&M_k1fsJh8=9yp>V18tFwxAlLn>y@DLlg{#bIs@!dsjN}b$%)7MMp
zh`=Kz-7GUU^vBL@5}Gd+9)Y&HT2+-gf0qc`yvW&H=%xjm%XFnB%l0)vdO8+TX|RW?
z7+!$kHFL_K%MuhqVsndldzcoiORYJr4ApuG2Il?Zc#!nd4^sqkS@j-Q63+{%%ewit
zAL@clQqkjNhFr2O!n_99gdATXvF;Sa`bGHn*dg8B)$G6!QR#&xEp&cE6iX3$e{`1*
zc;fu&Cb3Cpo_IiGr69SM{F~$Vsi_qTY$ai%#6wr&bCL`#RcBHZ6&i+cixh$5asiDL
zl_RTyY*O}EPM~rH^Bo_N(7-S=f!afrirq%^e4;!-`DWcxjZtzzdxMvy>SqmZ{Rk45
zs%L0}2G=&~slM*9Jq0dgcL}(ve_bn6Hl0nE-37H0$Y(ijXr?`PW(-qZ2{n{p$eU6n
z&p(|l4T1{EW5*q)lgh?3H*`wQk}<t)r$ir>4sn*sY44}?4X7fRmA-=cxgsOy3Mlyb
zxz0i8SSIx$<;D;kY=w{0Ew{)z+V#E%wC<-?Zfm$_9A<OKn}_8GffKR`e|NCXiQtiE
zGj9xJ?s_&p<=JIa@7d*vnK(|z?(&XqeY|evlRT5U4nX1orL-*bjigbiSwT7};V~_L
zn0Ug*BdO0^HS?GbzwcxQE>UsmvuFH}+!OdA`QvixOM2a5&GA=c`{SZ->ae=iOncvz
z-JH)Gl5GUGnLRfS>%7{@fB$}773n<>Zu8%W>#+vPXuq<;Mr|o;Kd3F6u|=eUSy?%C
zhZv#rz#_fd>%ztRe!>sAn`hB+VEO|^sUF5uBDxmVLt7k2arH1Dl~}S;7R{0n+t|>x
zH)W(<jGv|V?Tubqv2mr<EXs$XLQT`8auN{ces6nG>gNt89+Fk^e?08t;Vj*~75qDi
zMT)8a+6{gA(lyl)5Sv~m_elTOH}7V5GRoj&&e|{EpF=r2kCUveJpnB)-XCVPw466)
z<siu~$s$A84qrOA-EzQSm0rl~LtdC`q*>+h&c?T)u0mg(O?L#{OJ1G)<W=mN3(1>j
z+`f#oHBRe!*%Pk+Hpp7+u*z9G0X7|p+@DPB9ym{4@RC|rFqWqEGGM|zZutuu8mP2b
zzdN2A{!Lg;b^ik{_)CVDam^GI0X3H~Eh1{SS$q$rikGqBI1>XjGBB4RumvZVI|mS~
ze{%eO_siSwZe(WZo3}S>7MwOBlA7K>-uy9P<F`}VeItd5!A}31(y~wn|4mYOM;?z5
z#=799PuTeEKOcVk>-%>~0PMfKRZb+dRWVGD^G9n%kUqre_xw2sWr256B7$m~99C__
zh61=Oyq|7JdtkSrTK%0@qo6=qSHMOIe-k`5E(ofnLIXc>SuOZ}uNMnd7Zi!7g$(}0
zPsy<S5%2|yKk}>oT($!0<8<@ebn_-l)jH=qwuGRK#k=!?XiKU}kn%9JO0s3icoBUh
zSELfXw038;=66}t0;(hkO9$blBGyaBw(wk5D!$7So>iVumdg8jeMb%}Ua_(mf2kEV
z0?r8+Bb%to(w!xI`S<(B4EF(0-&XJQb8Z!^oUEH}orjc(AC=tOhE4a*yYuGk-R#W-
z549%wRKc<3TP-}0y_qUnW~Od0T)i2V1mcagG$4sW)rJI&6%_3vf}IFgSRjx0Mg;d+
z3><De2OEo38}d7XleN5K4q=4ge?$*u=LOEC@5;XbkRk7Mg2K}f2PL_@&fgn;()<iw
z=O;5p-b;R=CC)2MgLN4rKS3?H@QLE`sc;nLKnV+lGeVN=xxrY&-ng;)!$q_Pq5^em
zQQ149f{qFK=FYgxnqA1%g+Z0Ymj#fZ9-7#_pbJKL<5%cFe@YVeK1-&Se^{84j*Yy7
zlT5{bkx#g2WhO(<Pc3p)ne>;n<g8vkxk7Z-!K)1OGCRx4tg;bi>J$4}3`NHUW2Pr3
zQ0A#nm+-yJ_EvayC0)+XC~*4qY@;br^8X%|C)N^}S7d(Pt7YY7UQt<j^J<a6{K;i&
zqHKN5zS+iBl1<+hUKjYxe>3@{wY~Jc668*dxx|Y(#{7F~x8)+>wyFXK${r`3kTP6F
zg+_u&rhestvyB!HTuEnH$v#{nO>&=yE19|FQ<9o8_qwv0MfzkLO<rN6c(L4My*lgr
zV!@S$7x&%ZMEg7CazVyrAMaPMlLntXct0WGI?egyB2ldgIdTG3e}bGz0)6t2r!$1p
z+YnGZN-*o73o|FDCm)|4B0*^reP}`ca*TyE%!*o&KQoB3+aOUGY6*id9E*rwMG!j(
zoe|o&7U9pRU!UGu`{e6m{<MV<BWgSa^VFkK)ZR`OL~COTI^)I!Swr;%X^`bXngyh(
z<p^i~<K}9c5?&{NfBoqfIsFUYeK(~_9ECXVe4MOM0VN~Kt{-pS%+9)xIYbX)H69RW
z)Sr%3$mor3_<J$8H0%jZcU0&=yjjZ>Dm=vLNyuo|I7iJUpw4C0w>U{2>oF3H%(YO@
z4#_kUMeIuT<DxOj!N0?rOs${ZYTvbP2bIIG49Yu2W7f@;e>PN)`rQOup&PR98fwhC
z4GE>GYu#u$g12ARx;45et4>*W#ouRGcPrCT>;62PAqpAn8Yis#){F#2t6(+~)u{@j
zW`pF_g8VsVGd0L!GxjjG)F3~jU^u4DawP+xD7?);uQx0W+sMGwB)FIH$kfp6(wM2`
zXHsNG%vSz*f6ODYnPs(~<Iw}ku7vsF*uDao(k;k|=!rO}wO*4sm~;u6M{n+faCSK9
z5;9Q(Qn}Evi8WBYu;hhyexhKZ@*Et#-;GPC9fcx&WGo8hF{+Zmz{Jn=s6~oMA3K$k
zL-p4(xD_LI!~>%VYAsXSvow#2iLNw3$C^Or8PIaBe{m^fF|D6`b<AW$@Q<#`<z50s
z#tv@RbdX!OptR=B9STO&P)iy-mW|=DpERxUY$UbIzOMBQXElDNQ>wL|Plut#@TBkB
zsXC#?pb$~4jhe|18%*-PB*8DpWO-)I<e$fiMEQ(1cBMyk%pzj2S{dtD$|Q?)D+faw
ziff(~e_Ql%ez^WDA+3g*L)s<ja5sez=6HCTX;J$j3?wTwkO%rPPH1TSL<l2p#o|~9
z1G$<5N(FPG&Ow#L-B^Qh@$O6bQK=Aar1pI$J;I-m;A)Vq^ti1iuK9y?Yvp!gt3m0J
zhTSbxC$<^|T=5EO9b0!$nwTov##nYjmNH?)e_GIV81F`7R(fcyyyHsT7&w~$Ok**Y
z&rR00vh+4P7;OMQYdaXlh!ktC%%D^G8&qaVCmqw<L5a!?YrAn}8z<5cR6(DKlaxpK
zAJ#aVCbh2qNP<3tl9h4{|KqqOE<vvb!Q*LY5VeiSOpeDY+#1y&SVXQOqtjeIf&8u0
ze_bHrak8kflw@%{3njcuyBKkr99FYlwaj|`)%4Mq^O3L`{QEqWR+qW<JO^E%?yK3j
z-<(zIdiwo2GuG`HXgN<P6Y7l}A%F97V`6*kW~kbIBhW?DJ3%v`Dr8O9AjhqNi70ru
z_}Id-ZBK?PSGINTDP8Wu-yLEh$?j{<e_$bWwh~^~VEJ@tf~2g7VQqosoPzwe)s>0)
zzJ)FJLFFtj>)T%RUrpbsvGn6=NkWnlr3ms>8A*OT<WU<AG^znvjuk88&qt8+Q9xl_
z-seNec(({bzUq`Bfr)5I{QBuFX~e)P=x<ZY#kfY_-%zW}y~#q%!J#@pFqISEe>Wmk
zd4J=S^8Rw%MSX5a^I4%;bJIq0iO`2Fi2A|Wt4VlAEfv&@>inb-f1sNARBt&W9#Bm*
z0()CZp&A{eu5;s_YJwnhMKDYHbtY6^j<IK5)HZZN>S(A0TA>x4t<aUQD8{*Pi%MM;
zF&&{-P$eYOs4tx0LkB06Z;6Dwf6>3Oyp`#?LX75$*iF;EK1(YMl@rx`g#MXspwI*A
zD^AuLLC1({mH_YDrnIAu3Z%}g57ps+QOm2EgE}lJeK}ubh^{YK1e(y2%f4W}tDB8|
z-FWt-mU6O{e!i(IlrIyEq)ulw4b9JoiY%}y$j7Y%XFVUMpDd#;P$gAXe=EK^z;-!C
z-(RkkT%KX7<Kra-x=Vu83oTrV&HL>Y?c1eT4vh_mbwQb73w@y&+I$RNj=(c)J_6sP
zVB`oq!%~hdbh)!qj62c3aIUj~q=WF<d|qWkzh9nNb`&Q`=3gD1@3;V(9b+|WcPB{u
z`bc1R?xIkZ!R{N7)!c;}e_ynhJi%ARNVn!LWgi;Lh#+~;O2gWNYCGxulfLv!ipUPR
zanZ+(3wj)j1mAG<{<u9dPJ|=`5N32Mx4kBeP@Ae2%a2cQEtf<_FGoaS6Q5H;MxCbI
zmWG&eTmrAqA-I82S_<!NWCkMS%TWz;$HO>-o-g`Mqu8CfLLL-be~4=2x`W4Rq0I`a
zGfJ{3Cs!8!yn_UbA+#V`dU@Psr$#E$TH=jkpf8cpCyS%fB7Z;TZZ*h`yCF04__*tC
zhV>f0th@2`5k;rmt>$RZ-Ha5`uBMjzGmjuci)Gl|U;#nLgqNLY$loXx6AeE0H*#v!
z-)ipEn7_4FajU(5f6_m+AQT7<<nuA7t3jlWh-e*$&_Zdy7D?+6VBJA#3T=s2?P+bt
zA!?A$Q6`L2xfTMALjRr;&vzTBXC2E%txvWE`E+b)1f3_^e409rSQ=Sj!ipVOnn#n`
zXKS&XR(xkHr-mG~oEk)t^v;@ioE;(wBh6Z8^7!@*kBqcie~CuPLk*DP^ZVFpK>6d*
zl-9zzxK=@H*?63ElS0)ky<geFp=CE9Lzdk@{Ll%dT6k>%g3uOnu=%y)2~j{gEtao3
zHp;1}h4=^?bylh;{~U^j4S{3T$4YP~Y`TFQv*`vz`d~Z3(8C^RrK*i!sIevRG1fBN
zH+sx!YN#=*f7!#c9jTC=4ElZ*WID2M^s;kuh<He(8bmMV-(Gp{edzd7hE;H~6=Odg
zACI77C8O4y^{xy#os6{x>RbkPUZ5IpAUdhRqi>gb8_ABR@}WEFR^XqymrMAxror;d
zF=4@?)LDrv4XC6X_0*ccvASj3SdaRIJ(YHxw1yh$e-j#ru|A;zamq%0%uwLW5Uw~n
zc5@SbY9AY@CAA^vZ61dJ#;`d!AukhB*-MB<q$pGY9akiL=;_Vcgxb)i(`8zPuGFjH
zkOTUOCB?OMxf(0SqG63_C>3yg<J&n6<nvBqWT%cT2o!02=ms1gSJL2K!K>lySc^oD
z`P^xde=ON?TqKL*)gTU9k)h)hlz#ZIGLWuQ%vE9o)pd%wMpO<xp3%cjffyG}3uQW9
z!6RC|)^h6>YA(ijSY_f+R$!lS3=+~o&Y8Nh#*z&{o6x;9EQNHwh2~N&0=#qR%f=c{
z(YhGx2Cd5b3$5E$1nn4hEbCx}Sxf)fxWmI&e>Y=H`^fZe<XJUzg{M5Tro3CdagQAA
z8p!r;9!A&=Hy1V7^RBB92k9CFR<F7?%+hF}j&HM*!qyFz)Uki1pW?Ab%;lz`IX>*y
zelGTEkmnsNaSpsSi0t%wCcu`i?0@aXj*C0P)ATqr(UKN@Kh$AVI-EH%32d;875X*A
zf4gp#&%?`xuHHpV)HA9@@>Tb0x%B!JUC1kj^GZ$Eajl7ZBc0bbSpMFLbuJAJ8<2EL
zduvjHc4{Q%p~7nkeK!{K#uZzN;w^*QDI5@kL$c9f#qxz=Vox5|tU<0L@}~&nYZdmC
zW{tN(t5cdad^=WeX|^Ao^(hHNfGB%Pe*!4&6<6G~1bTxmjUemKNPwhDKd0H1z;?ux
zNZpd5d`cf9v~ipY@r=-jf5B_j8KLnX+K6ryK##RaeAj^gV#*1jVJjC1CxvEWx>DB&
zty3(*q0KsquCt7sLm&?-$Bs?-1|4*#_@}+@%;9d0V=To~0LLAS=AV2R*HYEXe=rUi
zrscesVbHt=eTFP6)1AhKtX_h#>9-d(l3sox5jeY65Zxfer|87aG0wdQ1Q!?M+^63%
z#g=>Boi!<nXSnC{ARBhM-=-`CVI6?95;~3@knkNh)8gT4J<jW1Bre_cxNR=UyYZO0
zUm0tzE5O0tx`IGwrd8-*Dg&oRe;$XrmFSCDQ^*-;5r&kDf|j^VtU>y=UE)m*Rm2xM
zjVVd6MQ1T0GFY{Ode;5fEJe3QWK{kIG5_HYH)vTAO@wr(My0K2Mn<JAoui}D8ft7*
zT0;zuN^1}mH8P~9L!P+Okyc+Z)vG}O#Z_2?uQM3Z^KUB@ELkZU9Sorkf6>VbE2vIW
zZM1M}&0gnvX=vgrCFOkahwfa+qT4}KHs$ikuS339<0w-h8KsWhV-uSO#5H`n@DS{v
zGt{(>%`i7;9UEcRfMZd%hR{j}r^DEhbqzRtVqF9E>@*)(BQZu}hs_jyy|oRH>DJ^?
z+m(9Px0w`DvYXGT@(r62Ta&9HipjYsyVo8ha&isJw;smL&G8AWAL$f|xwUi-gx(rp
zIP%s&LxHyj7>m0#1j3H;n&ME{T`@Ekbz5vhLDy^HtvwlYYalxma<3&<{w!Ga=fVCD
ztaEm+mvPM$6qms~GZd4l<tUf*7Z9v}&;I9+KkQ@`iGJUI+>N5R#rP;ZiuCDq_xnzO
z_K-XDzwiT*|Grbml8E>(e{$MTmv?`E)pB|M=kBHTera<((qxZpwhc9F2tr4_7^g%4
z-P9cSUgNJpbs1EIMHO4S*SRf_x{9jtLG`HT{iD(5sVK@|i#<|5;|g_BxJA`}T2~$<
z#rE1;m6`hJsfRh%xqW%enR=w9rZG(@PZ}fl#4+t@SeDtWjP}AtAbFD+rMc4BM(NvQ
zW3McY?-H2j+S}E3&3k9f+1le!G@HkCG*;GvZ+){)_*;@a>)=$!Pb2w(=L1Q>`RBZj
zGUidkxHf4-TY1Yo$Z7?h&q%U=)frqJWt8M{ep7j_{4>5?ou^7^&VxkM|JeB+zK3q7
z*<O5`G079N9lSd`y+je6c4iG`jIszbA=|^lpWhV{?^j7w#H|7&$zQ*FG&VfwZ2gt|
z3a~jVz>G5@uku)BPl`wuqnjqvN_c7^J(WGQDVx?HB2GRq`noBb1}}7f@Uv{2vzREH
zgtur=TQB5H@@{(;#HSd{Nb-ed35!EXA=r_`2@eRn$_QmlA=zFV?TxSs*48GY(Yadi
zA*~TU$xe;GzVo&uLSG|5s<%i4V1=}>F44NG(OU@_;;ONN-c&iD{=kKi^!x4EToekX
z-K4zPR5}DNV6r*d5OH6BTmf!>RUJox*!`BC5F3Pq#Ca6jRz2xWCX8~z%5cHNsRk1O
zvzO51AM^e^b!;vu)^@O~XsIe%v!2wRkyEcCXKyVcU*MJm;X?Ccgv|v->kNnLZJ)Q%
z0xcz?<lvPs-W8?&dSf4hFbHJ>2^yF7{vicwH;_<AmywW@Kk-F>-`+nG7kQfdFKKS0
zAO3q|Pe>u7UuQdMYTI0W5E%4b=lY7NUXdau^fa>=e5upIU@*IYMH8(d+QUy6i{7_3
zoT>_W3j~psWC)D%i9(G8e%odgoIo!!&Wr0}(!@Vu)O_F4j9k|V?J9|t!Xai)>XA2a
z(FO!JJyNOdg-6GK9Uy+AULeQ}ER$Zwg|@<DoDW6`8v;q9^?eJ?sou8GjPe^6n$MdS
zS~)<jjl6H6SGbNhHlPC`vFY2`29HB}HBS8fcEnb(7SNr<sMc}@u=+{CQTMnxd4SZ4
zEl@F;B7Wb#=n~`Cw`n|{ajoYdo$j6-q|ccnc1KxV@jvB%E9D?pIjUTk_mvy+FHMT#
zwlDK1nK@{A$f-uM+cA=n=piub<C=f?gWv#GeZR!60oPXrVbH)b$@!-5&lo(eZ;I}~
z$$PYtef<7WrCLEsYO{h(cf(l;;{YX4EG|Q`K%$w>WjK%%pS2(WB)TN%u80<i&5@J%
zQ%kKSODzn4CEm0o%uI8dot6~*-uqe!$@Lalm$Zz2>c(lRQ6(eDV^>F*)mhSZ3YJgx
z>yf_Xaa?=Lqp@v8S}b45tjm~uQF_tgcHp(0KuMvs&p;WXxho{n0h10hl2lv^w<979
zxg7@hocjKVNP8{NuZ?pfBC9(>A}P^{z$)9?-inBS0C1OHXZmdck;O37T@m7T{18A`
zA7(i=<A;bDQk+)oX8d3z_MR1Hw_yg62_65eru*EM(U|h7IL>#U+tS7hBb%SphW3~*
zNk&<vV{ud&tqjAXw&z^i=No&@*1e^Rizs{EldCIuM12pnA0gg^0AxD}SSQpvG6#}=
zpNF!4;lNc`bIadQlq6Z)Rs8ao_!Xw2S7cscK-o9SN^5gik$KHOXt2WAnj~J;$GKF*
zi?drVGxoK_-MwlBxg}P+;W|2Bcmbj_+}9Kh=6>7BY8a$?_~B;3NsoHCRn>A{c;`lv
z+p#zzFmA$$$ja@hJZ1z5fH>85kb@prsF)^y0ugN(gPbd@VYnWYz2HsVNYOEC330Tb
zToNf%IQJIvDD<D0nsDTS`dGI6%O1K1%T!FT8(Ro|*xOJ`B|3UmFs3rHj)B5fNBu%A
zA=kfrw>pjqCa`_*+>x-(BYU|#f9Z48-BI6$V7J=8Tq|vleAQig+=G0&m|IneWIs}W
zNVvK-|2Il^k+R&iD~*o0bTF*a!*AP85o|7|PYbK^$%mh^EGAOTahhnSo=sIXL<*l-
zRn9W-bzS8xvbr!g9|vC@27@IaTg&fqDlk{`D#RVe{sqS4-~vKBE#lug%tp#OJOk|h
zWdoq&CO{9ID{}pD73O(>U2;%%igX@-93l0TQ|jS)fH7(yKK1R9twAT(t6Km<T;?&Z
zg!jeS#CA?JyfKidT+$g^$>vt=0lh+mW~J@S>is?+2^SlM6-&|4TxP!5YWX@QtAy6q
z4Z+POw9B;Z+`wImVAZzU!30?$eh*ZCY<mL?LZx%F8VY8aySApeV{5AS4I`(27@^W1
zYD+zW(UC84<;3+P)S_m?j~<U4HBCobzEDR#-v6<5#8K!6cVv8%1iEuH)(=XVW!)sL
z%?D#-l9$n#V_PzIxjQ(}h8r{O90z+=@(I@?KI3~aX^0;+&lj4n^6cO?3>$hCVB0o-
z9xDoCl_F;^l(}^Ibd74`b-1p7aeoFGJ6(S0aq_rgyzaqax7Axb*R;QcwjmZG;M{|H
znqvL|-0z6>RXXs~7gbT(Yi;j&_8vG`cjw{P?FlBR=T22N&0M`k&>{(54GP2C1vK;c
zt>;#2yLshe2zQgJUuU6Am#t7)zkF>(U>69B6kuI&`0Jw7%fG)YSHApz*squXM%o7R
zlW8S*nKs*MC<sBeXnf1a)-J9X(1_s|%R3d`vT&=Lum$ZG7nAJbi~w!&y`%*v{;=fP
zQ3((+@Az{`%}A5(6*-?IH>nhX%Gka8*O$+K`}B(vNbsLTAs2(0mWp9{EwAV*0_FgA
z`C8sRW-eaGMTsZ@JrZVrOwv32PIRr|)3QsqbQGjAeQW)iTB9(*Xw8U0r8DHzv5wzc
z8V6~>)=JZT<&DHtpCL=sxJYklluDakk)Ej1YkK%Qbw#RQm)+-O_tLZ(0skfY@zSj+
zQwfZKL2-q)CiLfEd?0N^FOw41TbZD(QDau;44`O1c&YHOHgpqzmRgv=)P!Tv2-HNs
zbNdwqOtKc06r;454Xec>V=CWXQ&p`+wf*l;ua)i#Qhi{*bQ4Yk=B%##Au(K`n{Zc>
z*4mwV_Y||lb~RvWr~Rtc>}ck-tAEp=$imCQSS+$Wc41%cR*+J{T62ejW+dr~@tAv|
zxoqDMq~0+KADUW!boFLBx+iuN)H6tSOvPF>y^$Yp6{f+*(zp$y70*sfyqS|j$NawI
zbpY77JXm=SEK;j!Sm%^dqmm~2LB<NTQ>}9?WnH-pbfttIeZJW<Dp?+G5JjPto`?#^
zeN}kH!jR0z*4p?RqrdqisIJFK>RRXODc2B{RO1OgXZgf`;#vV!%PDM!!q}OGg`WDg
zV5)>6#OQETR%uTxOt1GCS+05G+|-<E_(&mA;gV94IA!sL8|Ve)%dE7G!87baM_Kuq
zozhYf_SER=p`K&s!|jyU%^5Bgm!OxqF9K99SquKko9xGf{FW0d32$DMtczb^rd+X*
zn>XAPAPL4?79RaCopNm79#cl<BVsnmcw$BYY1j8A&*_qf)o<cC^+ojNJa1n%Q$gd7
zy~^@@UN)Sg@cNzWk)&~#F@42S^zOZ6yWyj{H&}B~-Q5%y^;`abzV-e;*$%?3mvPM$
z6aqIhmm!E2D~G8g5QnKF5r?TG5{IcH6St`&6nAu&k&HMLmtovCCx6vk%Z?kn5#7&M
z^bZWpSMdQ129e~Ef$SX2F394-h@FKU<lZdu`zf-EY_imnX4I68JAg)NHhD<an?-iZ
z>DwfyPY?UohYv?%Cocjhucz0qlfoMk|0{`w4koXZa7In9Kc>GwNvR!v_V}?<8iOB?
zzXN{5^P^D#Yp94H^?&^6t->p#@T2E%eZ<d!euwABbU)+dw*H9_VQ!&!`HqRDmc)<4
zn${Xf{a`(Xqf!3(AFqFXkTfMJqDw#17r%OWefa%>Pf)oiq7KuS9}nq<HA<MECs0zz
z;HN+SIemTj%wcP1659w!|M1rz-zHY_H==QrNMq?7Ku2TjBY%LM91g*xTQiLT=?OS9
zi?qUFNW1H1_~97TDw=UPa>dc~#~Crx&&)_~81~gMLl6duLA~U$Vh@1b0CErT0{q*?
zZ6!(=a70bB2b3@y2$9e&N|Lp7??64Kx^CS3tQXe!k@M>j^ZW{$S<r-rGE$(icO^LX
zld)<LpkhImsDC9+>Ze(U60I`LV5XTprn^b^AQBZ}o?pHTMBT7Z^Bux|m8!}LScJy4
ztb02_i=cO4Kw2Vf8cz9_N7<FM^s`ok$Z)h|vwjC27?j|^0JEVoWUNf{QR$m?_73Jp
z5YZae+06UQ-QCQ^%-QsNp9xHoA@{`RB-DrTe3nLFM}MOj&CqtzgS~;<Q6R@T+9rN`
z;lOaj%xEA6qfV*aikEsjqmgbxMbCR#v^?vrqfcn`7uKv~k&i3^alWfGd7MBfYM#Lz
zBm<%r4#jqiWH2<4x7q*(#Dsz?IPdlb)8mwg(sLxG*ZcL@Ubz$EwTke5%fZkZ(;-Q4
zTy_GD$$!W~a-0K-WK>QB<o(VI<xhAGE6&R{jz2jt6jUf7k)=^4<5tI=OVeH7szws<
z?7B(Fle106u&PLp5f=wv6l$TtqC)w~7o=JeNs0Cax2WTIMS>{28|6!TG$W6jOe`DQ
z029j)7yn70vKUCFc0ZohHi#hoC|WbLy#(uSr+;;iHiFhYVx{f=O6)R#&IZ?HHjJWk
zkA@3N?>5MEk2ZqNJz|X$4~<M8?oV^=L<n&d%{w$BZw|`Yum(is46%|lMQEO`ca*;f
zB=U#ZoX=6*c+x9$#KS_N4$Qcl3E$z2pnC_Hr8+i}eJ81wRz$Bx(UzgvZQij*8(~R$
zM1QpbvBaeev2Y3+MKLSOn(dsk(F%-%2XT(0CC<UkoMUJMIL8pXxTe^DAOm#=S_dOA
z_!vcNh6bCogrSX~b&uH5dPnh1xDTDZ7Ba#pIx{rbCeuCI2s-zO1FR88ES!tvn!}<g
z*<H)@dps9$G{}QdTx4j8i?^==7}^Lf_J4>S&4GnY`-bl0lg3CIq5P=T*D2a2E7hZo
zpnH#)>0GQ-BuiE3pDdOgPj3^2R-;y7r)Zn3RgX4;-aX<7YsC=Tg~p-bUddPNW=*OZ
zbO(;=AT*3x<DH^yvRpmd2#)rMUF>S<d*4{HeY4QJlLb>=j9TY$EMGScZ=ijPHGhKg
zE#3%C>hM}UGW2a5i|832!{*$yiI5KioaUxW9{R9suk@R%EtjqDYx3>tfs$r|6;_Xu
zHIA12yl%GH-=7R%h<dw>FvPm?7H{8hL#fjWI5OLCJ2w3Ev?JonUI5=<lZVn)>H3TI
zH`MsO0ppFZZq$0CM>AS)mx6wL=YMwF3WwQj$X8m@`?AfM%H{t3W1`(K)(gu?@y%FP
zyX=GAB4rF|k+NfBL3Y<9t)s&DZCm3sRv)$Y>CudI+a;<W-!+M1v$>N*^}Sq^sJ`|V
ziDF2dM0E!Oewj3Z5@1KI=onhwgWV`iJ>m#2)gx9f6~=g~9)00>-H$GZz<*rB`NLa|
zf?V`m<1=N(V)Za;?b)N56c%okjDCFA><pXDo$O5C%Z9kS+9HN%&sZ*ukAPEoljUMa
zi{*-By-p6Z<?|}0v1|M8>__pP$M%%Q&#-Vh3a3Y{xO+6C2KdW<eAncd&E`&W-1jo?
z5MGwNzQq>FV@O@B4<DNYz<;5Yv!CTJRS<5NyKKSq@wi=<y3t(|rG6@RlBBMaYk^l+
zd5aA7K<=_3Qgocm#*@QpzB$wR@-vN=^unuAo8~=Qdi-Fsp!MUsCTVOocM`R}mus@t
z*WMy*3~7-zXe93C<Y?rtaE3ZM;bGL4K10jevQ4fS;(&0UA$FTMW`A?I&(NE2-!G1;
zu`;fCipHEA=UoE*t?5$jtq8`hq)TJ-G}8@fQp7r5ICM)GDe0SSCZ*@fIe1$GMbkb@
zB-h4N4juz-C0kod1Jy0ZhIeHNmF^nZ=$Iuv3xyNqNv1Yr$WP8$dU{UYNAH{w!H-G<
za<q~R_P6U@k2pf_dVj>aWw76_b3O8(C>g9li>O9Lj|{CQTQ*2Rk2pdKdc<x6Fl?8E
z9{o?8Y|$3!1*4KJJ=&6N*&rGH_^!lzd^Y!q_gF6jk}VuDTwL<9+8ZKBj`XXNEjPp&
z8YO%s%t#NGLvgxLT=n{5t2$=aPNP0x+&p~TKLKT5=#DOTTYt69CAtg6moO|e%6<oB
zmQlzsYVpR<K;3$roB<6uPtFi?3f-(6+jzfRso;fj=KOX)G&be()0I%3$K1DZ;59MP
zr()9R<osA#?cpR2KL3E+U5*+k?}V144i9s*H&dLv3{ce7JZ$$aTO5R!j*sF*TD`%h
zivDH9u4Ki^GJiw~&Dz2(xt?M>>l|neBH_~8EvWS6_2jJ<&I+S#da1?l6W04Q<;vI|
zH3+8vO}Q9{jDKko&U!=;23&AyUJoDs^X=<jzWk;HHv4a?DkqEjqfq~T%$EUh4eC)2
zzSDION~Yv}^2Wy5DJDUR>GTsdtOZ}D2dh!lOsZZ>zkjFFC~UA=H;wcAI6PLY9W+H{
z3<<H8mj1rCg=JJr7kN%~80jjZP}_8g?Mth4nLh1xsfw-sH9dTt9zGLQ!8zx#N(kB{
zi{pH-c8#h^SmljsDO-wMVx#xIP$ha4hcl(^d~2Z=$VxjOhVvLinyL0y=Uhsfd6y!@
zDhQ>he1G%uGjw2c#gfI43y5e8oNzY3cSR9(X9_RBzx<$f->}uE=6l9@RN7!n<nV_@
zjqo~Q<=3rXRI~O<8>(JvH?OqGA&y?25l%+c*zhzb0O;3lsgn#dl^&ZkLMVz`?v~U<
zUJ}F_ktCHavsqJOHc*__=c$?YOuxs4fS=z?bbqA;fPDkedHlH$+XF6yL#6eZ*9Vm;
zpc(YKI%{K|wR&ovjehA~bMpxKH9(2xC{imlr_wVtpYNWbIo=J$vCPi!oSdWuo{pN2
zJ>%@jc|)Jx$$eoIi1zdB%VHKoPp<JC?Tn4W8PEGKLBl+c`i%a*N%<XUuPRhpG(iHD
zn13dPwFDKJ?$9Q&Txk`W*qJa?<9ye7F-tg;1qtml-0OUr%hL;_tTs7y_-yRc68y|B
zifZRIzjms|`*Y|~>H*1Hxf7OTO-U=xHRpwTMVqChT+z}6N*c2X&*mD1H&j0mJ@<^>
zhtq^0^`1i-W%f0}*o?ccdCuObaGj=WTz|osH<;x>CXyQ(a%-(=LXIX>BFj86%CIIf
zYRjWit?h*Y&v!2vNQxLw)zd@~6d!m>q3UwFdv{U4(BNYTz}gp%>nwG7`emorH(@7n
zou<~jZ(?p&($8rjKy{!HBjZ|<!II_<Eg|ZOzuj|8#3YWLBZP@YMD5QJB3vbeO@FTr
zLZ@(v63G_R2x<4*(u$ex7rpcCtGZeW)#PYsERABJe5``BCZptXT0`s>)hc<mHt(N|
zF)3r@#-qNRgK`^GKTCMpL#WtxQJ&e*_>v+Z^{nsJIxg;DHXFX#VQGoB7?h=6btq~4
zALsYtwd;fGLJqY>eP>u}{O6Y00VS-gOW!*Bl+LK4cOf`g!hp5dWQjB8*yP{&e_h7V
z_x}O==^FEwam^GIm*JT<6aqFkmjRjuCzm;-6cB$b*?K^3^Wa;MLomop4l_aC%OSs?
zB3az3yR~#n24qQWvRHc+$)fo5W8%|qPlwN^FDIoZ$Gj0vPM_Z=0S`F*7aSU`nVb;J
zDlvWjn*R2}xwP;)z)N$k6ucaK_V8|ApOx^?Lj-t<*Jq~%d{Y8mGQZ0cypH(Uzdpz3
z2~K|+>s}C|$s^=jdZI$<#o=Y4r`F2Ed7z&FY=nRP`{!SuIFiI!u<>Jjqu)M0pZ<8F
z1Ohk~rf<KVVzX9)DK96fEEB>^fBtj&{`8dsty4(@ZHDK6f5r|&%kRixFMz|`T4M$d
z3l418leK}V_$YIj1C5r;$PM7D0bXgK(f}k!fZ+7f*lt|x8M$ac3i=SAl(8Ckt&cc9
z_ODR2$t|%*g8EjMYo!z!e~$QF^s^G!Gkguwn6L^JelOV>-;8Gp1cH1UF@tw`7Ow=E
z=_sl4F9~pH`MsOff<{<LXO(;zN2i%}g7)SIocFTg%&66lY31fx5>+B;hDVw$a@(N%
z9LgK)$sA+HB1`DZrI^7fJd#CfO^Z^86S21qw19PK=a*@rnww}ae;sWzExAlngpvHk
z+ae7!{G?$<go@vByVSU1fXbZG=tqs|D0Iad5*6=Rp2_E_Y9lBe?V1GxCFN<+|H8X!
z*U{%DE2!2`eSscq`yN(lf@D^!YLzpbZ?`<q76mSpKwh&vt(gzBvsz-^VZlLge19j7
z<Sn+p4qTE=a@F3Bf7!n#S-9eP^Hd8`xRn!%OQyJ@#6h%O;)LRs8RKTGxPs)YRpRkp
zRdWq3VNOwbEmEe1V=-MJqxiA2RqJ3fp(epN;xM!<*ea02oUCzJD%5%1$yQ_hJk_YW
zsDnm<*Y^}+Qn+Os&Ny>kyh*`%XaStRoC06WC47tyGa72Fe}5}}f%xvrl3LAPB&1JK
z1|~3PYlIW(<{4&oK0x&Dyq%BJCsnm5C=^m+waiAVK5lnEUV*aBW)gZkp+#5(;`=h8
z0TOWUXOu=4&u<P)(s~9j3atgE!ntbP4pWOEKt)y7AwLC^!?B_!ctU61ljMA+JlZqQ
zOH!kx+$N!-e>elQT@upZE|i%;d_Sv-W9O{`l7eYFgVqV?)a7JvmxNTZK<%1uToZ<Z
zcU`kG-Q|rjcN!oaHssHO1)zt_BKUS8VMb0N0@dPpHY+D1YRuT+6b4AEzN(n;&>Q0O
zCUpndGZE&T+^+62MqxAK%Zj>7?wRA3_8#>8hPjKre_zE(DFX}6KUi0#)eIup4VopL
zWJbzW&AP$<y<za0K^pA8t{Ar3{!O}G{DZX2`_S|aT|596ZilN43>+wdzG_LvF>4gX
z<y_d0Ioy_@5)2Dk&8gmDWoWkq;+mDAO%h5sMimcPubM$?T2)qeCAw(aY1eTn><ige
z2C?_qe>A|Ik30f+tsWhChDq)bcSe9$gq>l)o~Y9UqNZ}qGGZ2(dVq?Hl_-W5F_5t*
zfQ<~iL;M;6UI<@9=e^PEoJ3~9>yheSW7oCXSwq*dc5aAVBcK+zLfhO+iR)q5!d6{a
zI}96CCDA?gTT{0tx1b`ym17Jg?7o%80PQX-e+>fh{7Pf&ymzHBLK~NYX;<jjc9@%Z
zHmopoh`K2s?C8lXckuKS;&HamL0vr=vBTGokS?t-#@_OD+U1}r-TOT>g?QdYlioXh
z^mUvpPWrLBTD<i2npNENnr7YZrzs?R-suWF=hI>6hr8d27cm?B*yx}&=sP$1(Y|4v
zf8`#&fk3>R29gnbe8T|gqHh>_#}koVzJbtv2j4&-UhoZL@4dd^Do$44@K9Z?zTqr~
z>4xr_R^7fvq>#qSf#MrJici>_qTO_!TD$~!K^}%$3GIb_!BP;BcAWF{QHal43Jr1a
z15c~eM_!YP_o^{nXI!(o0T3UT$^D_(e|St<j{DR={<uIs>j1$;L4H|6cSHGh_kcO&
z2ytJQn(|Hxa4kvAJ36HIMhmT;dxVzG`FD8l;)@AIMfPg5rb=m;lQZi1c*=es720)T
zh^L-l)DTPRIYX%}(&LWJ`YE7dR(J{}p2UixxJ#u|AD^>2ENMLR%EwBqKPKq-f7_IL
z0{|5uTlwFVnwgMBhj^MRgtHN>g1<=m)0hAJ`2Lq~{{Sa8{rGR}$|ntC62VWu^4F+K
z-Yci4pQQnMD9OVVl}uTk!@4rirHpMY-=-%G#tL72-~2N+2d2G-^sF17O37U6T4w3n
z3Mvt5Z7!4k>>CSiZI(n^MJ0ZVe^792{0igCsQ5L}+I=rwVbs5-r|;9#SLCj-(#0;G
zXTZV2-x7BrYzh~HMdNU3652519OyoBCO8NKYt22ywNXOC5)byZjKN;$cxaFFH348t
zM|o)iSOr2w3g2n_3MUA)(9(9GnF;YtG9V^ut3U>4Cb!Sue&N6$FzmPPN<9O5j<%zX
zot||pu@oZ0O67vaFBW8C_tLvK!cy)NQh{AGxiZs|u@N9J<=3f0DHHeHQTO$FmI!9R
zp<Lz&4H%6R(nV{Rl@lwM#j6w{e?eBqhker0<}H^k;;8;?JeNXc_<|hCgjRH!G04)%
zC6f(07F$J`p@W^#DuacDhmeZN0S;Vdl7c*$Ys#a`1ueDcX2$K9dNsXMU9p(dws6pt
zP?lRs`QYg+t>DnXDrYl|nVsm9wXKQWImX3=T3AJ~NJk$QH^GYB&8w6ze|<qN>c=rs
zZ(h!o?zTg!RJL@25mt0M*2;6U9w9e7c!l$&It^3sE)B=0fMr>{=dl&LPAjo5k?Lv*
zVAsET9ilH5Oy1N4s4N2Xdn+nSS|VxbVb{{7N!Uu#q}JL>*h*_DekyjH`0caJ<2c-N
zPGb%7S;}ss1N-r29dYxse~nK4j!mX$KhG+N-bRi+?DkE(kTO?Caj$-=mf_VN$D#k4
z)~s=NO{%gl){cF%U8mw<-*gGR)=cIaY=^gvX<0Unfg@HHO+}7q<R(5*%`;_A#wV(i
z1@`M`iO&H)H9u3pz7<>ej9kpJE_Pdc@@ff%K6oO!#eFVtZ1duie`7m*9%gf$Mrv9r
zSzuRMN+hcEHY%G>je2E_LtmuKtJY;U-o2Z#j8i$az%`_2HfC9>@iK(Ay;PGtvJbhI
z3m#)@!g|?u;#1pMNSp!V%HdV<m(FKK%FM7?=?X@MSRLZ&U(-H;Dg5TBbLcBGUK$JD
zYxK@uu9m}BXh#ZRe||Ls#}eOxI94uonTMaph#0Tp!MfoQmOM1;PQ3z=wUKlB7bPk+
zYxV7=LSBEoD)T~9mqB%M>Y~P=G9(bZj$1ORoDB6rv56p$I`s3l)N9FMC6a4mHc#G~
z0X3EELSAzL8LLxs#JZ?sO9W-@r|Yb=63_Pu9;LFLo-Mh9e}?n1DZyGI#w{#b{8A!@
zZl&8XFnwB^9U8e`aVaSaunZNB%z7d55^aZuCSU4&@?{*;fzF1GOf+X8L$9Zn&s<z_
z-NuufFmoqd!%KBroms_Hu5;F1LehBMCnT+J(ins}x*l!ER59b<ap<fZ`(U8vNyd`K
z$e`1`fj$*me@GT!wN3b_6Q3{Hdm9&N&SP>&sm~UyH)<sI4c$$%wzNtqRJuuM%oWa0
z8#l+Ie)9`~0jYZ%Xca9{%or}v_R)oew~^w}!&I+x7=4tMj)wor4YDwt$<j-vblb19
zog8he$SWA<E14i95nfiVyCiJ@om1+iyYc1vC;A#|e^NuTmc|S=saBJ<48Mfy1jh=z
zv8C{JtS~QF1$-X7;wAr&-bFdYA!DyP&1m!1Xmu=@s|6}|aB)%MxG7R3ek?J=6>gNG
z8k_IblK5RDl2xJiGBvWU@v}wP%8MM^hy{?Ci~3AQy6aN%bJ0kNz__F=G4;Y<7jQf(
zjy=Cke_}MVe9q(#atYEdrzO{Czh}~^VhXa$MqA=h4$$5^F%|X70ZM2wzXi6i>DTS9
zEpGbN{ky#9_wV%o+<oeHnVQ4!pqO-q-lP^D-sS#_)D2xcFL>?_46e9k9dC3?xlooS
z1B$6uHP!6s*o0WD7*9fpt@J(VZMc<tlJT5Kf7*72T(Rg*eeQgb2`d)J%q!>0;3(2&
z9L1COcTUSa$mDz5iEwD@scp6C+s@_lNjsTc^1GdXN#&XLj?Pd_F8L*Y)ft#HU_g8(
zQ9P@g$z8-;%&f?L?-eKleNDD_KumGzWt%X0jBK%jgv@?V*<y_dnPH1;@idn}T9YkT
ze~_EmHQDlgMW+P%CD{^O2H8eT9xK~%_D9HeoMDe_M@TEO6+y=1*e(8g>kSn2*JS1k
z{+i-B`0KQ0dZfQ5vp>LJlNomSYXWIa?vTI0B5+ymSKr&F#Xcj{)f*Q|G9fL+VTTVP
zkS_QTO8e+T#DjeZnf(Digv_wRhY(0>f4UTH)P}YAu=7#^fqqH01eZa!;?ePe%>D@3
zjx+3$?FebdufYiIVTmQKT7!gc9dhrrhfofdd~*BRqwMwezZ6o}0<P$cymM?Bv3^8s
z8E4oNTSiE0{y<u=MZG<zSdv4H@eh;3i1j1nFwU??4kM&BIS9k7Gxv`11JsK#e}2Sa
zQMN7S-d(oEFOOI}!Y_|A?D5MZq#b@aDCW$<Pi@Gv8$rqGy%p5U59Ot@=#?Mg%k>TQ
z>2a(DN^DTwKH$P8JI;EhvZa;k*XoNBD0Dm0x?3pmlDW__sJb6?dlrLMGXh-ukuOYn
z#?H}Z`W>leH~n75@+x?FjkKKYJuc@-wKd(ko8u4-d0{r!TCqBwve*gcRhY5ZZD#QZ
z@B)j+&I60xZWg;m7Q2VDcx4-!ynlQ?{U6Z@8u^!T%@h@vaqt!s0W+5|i8?2L#a+ve
z<+u&q*H_#hXpP<vpiw~A;|`Erg4zXH1({jRBJ=-?NVa5~M}AAwS=_EFUz#SVLyDqk
zHGlm6;E!Kl-o9Rbc`E&I%p2k4<LkGFh+pvdui){bH4i5Qvr0U^{&@WJJLl5I&s+Rx
z&XtNE7k~Hhr+NLRgpU*=#E*D?{pPfYe<=|^a{MWu@$;U4_pjg5_XD5dUiUq7G~*re
z$MB5`g^I_Ijg(p|lb#dlgiX(T{nyvuUU*iCvtZM&=?DJx`StSO7gW${rlfxS`;V7&
zT`R$qmk%opvq3!m{LkaJm%k&|rTM&w=Hu2HL(C_5w5fhr8<<L8jpoyTQGFBDH_>c5
zn$1KdOw>$Ah4pj(K7FTk`hHZ7>Bx5a=g;3CaOEGr&U(3cMOb{Y<it!fwP!(^ih4s%
zQ{>e7yZ7<hKx;3{DwSv2JA8VC&zh`KOBPX8$+~K{dphv9*8#PkmsL?pGVZGZ7kX6{
zr7Tm*BQ<}%ezRPD&&u$B<rVcAufB~?F}8e@q@Aw1e9!yv%nFLeGZpa*P#k_{GbED?
zrb#A>{%3qPoxL@@6BOx_kDy4Y<EV{8rN$u~hS+7<?D#i-8+91I*9DIAaGQZ{7}|x2
zLuQb1FdOtIc&XBX2`*9D4dg4MB1|enMzmgju5rwY>aiMP=mmy<R*Yd3IDV`vVl5LQ
zRAZc6?|mD&+A*!%M4d-{7@j=3!|&Pgr|V%x^oZme2-eIxff8&C(}ZA3ozOXX)|1LM
zAhvj($=_#>8&J)2z;Y9U6Ukt#IRJ!=#eCCs*CyuYEb73S;p0DpfykAmfyxj`!%Q65
zqOvVsCd2rwK_0t*I%^ayk^{@hJIz8qm+6OVg+eEfOxmjHN_KtimNbKPCYt#v<(UHy
zpQSQ{&P>25SM3}##-Reou`ZfOsE1%YTaH{=vz*Y(%&}2rwclvOE5kIKZXJQiy^R8<
zHIOoj=7{$;PB8l8`W((ZbKIgH{B}f<!Z9ffFl-1>QVOPjZ9s<W5lTiNJBh<Wr(>~-
zglt>6v}U^Nx<!LxE}#hf<9>|hk}0k#i;hwYzaFY%1{~8`l%j-)8$joZMK9`&bjtM@
z>M0zwLLD^efOX2T2iSq*0op0`HMq|fZG$_EWCy&XRWsn-KHx!)`;(&aaqS$}1ofo&
zHY3v%KQL2&Fg~6#TPK~!GLF0H?2Ewl#I$2a1f?b1U7ySzgc>#YWcCT**yNLe1?MY{
z8$PKtlS(3Wy&{ah=#0K@F{>FD^=N&rOXDOnvXW+a_Vs0t^oGSx_3RBx<j8=Q$g8}B
zsZ<4;u=QiWoP0i)@d#4e8@^%!bGCPe23#-rifi|O70{@IuTYLXd<8fj_zI<7@fFkN
zD==CIzCx*oXE*o?Jo~^`z!H~z<wi+iEKZ!WQK=WK+30$ONu~9<^e1hkyo!$Q{tq_N
zsnDsb(b3|zpJ*+a4WaTAv#Pn)mW7q5(-0N4XP&#>QIT>zjEagso3nY;q&=Gljvcaj
z<hY-I%>#9n%_pnfiiXC-Ot<2{6_HBpq|{54+$dnmm{W5lDvPYwl&2U9GPEtw6}QY7
zSIHoE*A=x4M%(u6L^<ZQ@9l?_>v49{l9d!c1Wh{lA?4V^4}s%>A5!XtA6~Xm-ulAl
zt5l1wN;ubiI?f2D(rnYGJ&PZtu5;6yl>pCw;_COLERz>fYIP%vC?0z5xl%FX5M7Iz
z=w3|GI&yPoWhglpAgrwtqvAR%W>nSa-p^8|{0wlro~0-kBc!jRPeXK+HgXKTqa)>d
zB06@+Qb3aq(UEfO5gmc!L3E_lJ)$E#zhiWS5kH8Il)A2RwMEBMvP{vDQny6MbI&b*
z(UDMhi;m}HDA5rR)+su|=yr&XfZK!U2sUVlj&Wpz>Kz;j(-ZqKrad-NJvxL&!mvkV
zqzn%NBO$K$VwB;|Hmu(00DbN+-Uu;WH??7pYZjd74L-{CL{;@iJNiJA4)CKKd*BZ^
z9srP1uhb8I+(4g5`(-G|`jUB?c0mw-(>8%R%bat_wIP5{#R$;K>=z>dr1v)%wO5Rr
zl?_IfK~J|1r;wZ6AuKyx#vT~b8RzrW_ng2yN$-s=ojc=S({H8QDCCb{xU#6bkpy3C
zDgVe%HH(t$@bu*-w0U1Nd?l`Yl6ulglrodf{X2TA??~n#kruA0$yk=wtM}%AEmrL$
z<DvH!D{wtgAeC+1Jwm3QXH<qaM<n3b6ywIIg&enH+_=94)GHkija?0S0IWb$zb$|R
zwjosxO(>O}oB5_%Dd#096eyA_E1eaTtKu8&)p>mZGIZQY)mFr>d+($I*KDVWN;>A9
ze8i~2gmcTT;uWV^y;FraE6;Tv$(5A1e;rDs#$T<0Yl}xscAy*^JW5212JR}3>(~{k
zr`MvLa6MmHRYw0Xw1(2=aU!kyf=A99DaNEDrKTyyt-U_sQ}?f~G~BRWhzYh<NImrZ
zlurK}o}a@uVY=z~Iqd`y8qHAN?Ko!!nKmdZi6H>SRLi!@<pVSNBub{3xHme{f3WLf
zRq-zxT6I{rYi^m+z1L2_^;BxH(EN640m^hpEhxtxsReL6NG&LJomzAVcBRz>CEI{;
zUZ<dB?ZK{;84)}myIY|^-5Bi3t5=lkR<tXxUV#kFO)$fIH7CuG--6jS!&<OwiW(};
zH0jmGzEK0Xo?DQ#^k9!s1C?oye;UBCL)1Wy`%wd^o3=y7xfGwj(IPPxW>&_rbvHM!
zX;2KgTO2_%+??d|ynW6DC5Qp_TpPT%Uyw50+Ce^cqY|p$wg4;}A(ZQ8WE^7{h|n_v
zdCj!!z2N{{H`ag3sp#hB#`;g$IMs5yT>quIRpa`Pa_zDH1CAZmf5>rvfBgs42kSo-
z!;aCU)30%Vw5<Pt^TGOWMGnJnzbA^^HY}F<P+dT3ni7JyU7sOOi89T8l)75r@b-0A
z-cdri*7u(#nCKYTyzi7LinNU!faz8^oII6^&{`eH@5BSwEu`c(ijn%Q&mto4U;+27
z89J8<fV#sXqV#W|9Qz4Ne{GV+342Od&Y7bmEMSgpC$6T$nE6@RV<TS)pJ+PItAy#+
zw8|ZGLfn{EWr)@`ayhLE+dQpOu05tz;MieWMUMN^Do`Ixs|l4ZsD=|FL|jb^X^~Tl
z-qAy>G8IR;8+I#A?oG_a=D=+$+>XH+s9OuuA-x9bPs$Lzhlx;0f81OU9)maKx>*s<
zSpmpU$F+OmehlCz&5>IHfV#DpE*Hd3$}pCU5fxCI?vNvuN*0Q>pCiebe0rt-f~^$b
ze2_}7$<fxaQMOW$^Zr%}P#+iT?;|-^TF%K=8nAZdyYAV!5=^;ndUnonK!(-?E|*)*
z*`btMfch`k>nW!!f1Q+TtbZJcb#kvqS{0qsbZGLCI=YwfD@{I|?Jy#@;+S#}kk035
z<z}Ah1c76?{qhWa@1VPwbVgU0cgR9HvWGlr^SHbc%7DwqX?NI^FBR6js5rk@`R^HE
zZOUJg7Wx!u1iKZ2f#b(+&JNUB!^+AsT5XO$Hx{wQsD#nhf4Oesq8lFiaeq=Gj0A7b
z$0<WhbT4>pt^!ep+~T=%Wp!lU>fW{CujQgGEEa5@blrXCPPU=Xub)F_kcb%5g_KK&
z(czpk=X`&K!}3(U;S@5n-kqoA;`FWOt)ts=84QGoeCtJ_Q{(b__-9~BuB>eB09S4n
zXA>?@qZ5n6f7YJ!iErH8$JppiDKipQ)fiK!#h7q(dR*@JL}sC5@*x-TLM@UXu{i2&
z(fm0Iwvq0Q%PQb{<*>7RYloclp*DOo0+uxl+%d`V-n526{Xh5CFqBw+s?p>PVw637
z6l4GOAWBWURNwEvT##{Q4U^jR#u?>$ywg2*3AW)3f3!J)Gjhi)M<dQa|9zYh;;uLY
zF%ED>O+lJB+i_+zliv2m8RdGSqr&#V8EA6?XXK7qjz*k;{`)v1#9eU)VjSQMOd-)6
zO*qpr^Oe3g&M4O_CnvVy4753cGkV7)M<dQa|9zYh;;uLYF%ED>sA)4*JI<73%7)%J
zqg;1ze|EX_RkCsNgzcjC*k}dv^guM&ZEcNlL92rLb6VwX^);;;{R75r-#-X(SN{Mp
z4*Ua*wrtjE_m8wKB6_<AVR}5<i%zpW)#XGGUfwWE(TFjs{655%v|SNKg*X5hF$3AP
zMtpUs!H6ic5zC{qh9hI`d}8_8noKSbBFCRxf0HR8m`c%Fl*#QQz?B>LPa#TM#7B0o
z_v~W%{nPfMObAQ|xl)rkEyd5T$#~C<Qk`QOW%|=9GN8+GOrtk+s>guxpAB!94{?w3
z>u{(|*?$8;_g~o{#9g9q>YEDcq!f80uL2gl-+c36m+@eg-GQ6W5Oqui80C%nRZFhn
zOI)08dY4Km)1Ows0$q;%lHAa%eg(=m`z7Vq(Jw*JeZM5cJ^k{W6xlB;V8LtBO@6sM
zaOjs5I_vemg<r1yw{H0J^Xuh*GEPnBmvPM$6_>%_Iun;cz!cGc0dfN`&%b^C{KS(;
zoOu&&!x#PU<MZjiPgFo91@j7~-+w%X2TB5yPE0~s#;u(G{MYoy)88rCGMKU0266(%
z`SfS#(6s!LbnF0i%#G2y(Xmp@`N<g1WO$NPjFKu8OrsJss!E4@RI^5b6)Hl#rFPp0
zUWj^}Fy|;$j+2Ie1R_<1)D%vd!U^HG=WmANR7j5F$Pa9N$BWOjnL?uFBg)sb{IZNI
zlwlsVL#5&2TIEz|&#b2MFMPb6<HMom(O{|kAH+<KQ1nbIN)TkQbG+PX!YC@!+IFnI
zuc}Le4kko;iw`-Q#{(*piZkuyy?iLofQuCalQ`3lnvt|HjyWd>(^q;s%cO_SgakDs
z{Cf*m;%Dr()JL_$fy~uD+EN|*DA)cPA0^V<;iK&vm+rz89)I}8qh?fmba(nXA8mTK
zd|czBxMVA7ulCWF>W+`jZ1bQm9X>MKO-pUFLNhBA89{Pb_e9~Dw;exMpaIcaj|PXN
z%7}5FR7Np$(^aV+#1=Adk7G+q@JehUtPZfH>3u1-+{Z`8md-Kebr-I|mfh*=u!R^~
z&Bry^A`MgMeSa5YOG|Yeqr*ipI+5m%^}l&z;p0IcCGtP+qb<QJeUz{|@X@CCr9OHe
zAK6Dc1B<=^7scqi)7SZEYeEelYd$(&a(=pbzM5AHOzka7j|;2@pZUw~6r5F2ix?xq
zD2oThbekZ!84%(XfI^cot(-@QA92j4rA8sEkRKp}1b?uQZbq6oSip}ouu6y&?`_T>
z7F<eCUKY%XZEqRVO2!eGPl`jOVXFuuzpRN8O-0wD<d&VuLPKDZODb9moLjmt-r2tV
zdPGvZ7gWy54DX*L=_Iyv%)Eq&Ivh?7ZJSGmb5YxN6vdtn+z=?4Lk!hkFr(y<ve221
zPL!LLJ%5kMBUx0YSaN)m%`4RZlur0m8}r+oGlVvH&n6lwgK=8(wDa`+UYl&Bkv-AO
z7x2m`U^bEy$k53Z7ERM&fgVb<9w=yu*y$+6aiO5r$=oWc(pr?27bEZZmuO?E;QEZ0
ziQlBXh}gHqI>NI}uCgnQErte2gp<WWM4Ib)oPPpVDI|+1?-v9PKB&vP^%GTD#9}M4
zB}p6rb-rJpp{z4ua7`?0O{5^JIc{u{&+5NWUoOPI9T$~Hb($$lf-kF6mOk+E+WC%H
zW;*+kx|JTPphTvUEj$D61%8Fmm6y#Ds}^T2%k9QQ&C7I4Bd-qeJ-)0Oo2vv7DjJ4)
zhJWvO%;te$R%6`4J9&4abmJIa^o3j6Nip*dO5A@5ZEv$M!6xGK(B|iT9AS^!UYZ@{
zLn<%UZ7<kUROCx-ceT1JTe=^qD}Pmf+kP8da9yag=hq4sb&lWJU}WN>4li!=J%2qM
zGUU}<S;4s1bzmn}^%FIBBL<obalyhF>wgPY@)EY`b@Kts1n7>7Te~`BUJalWODR}a
zcnLEX{Pms{FS(R2@<j-D2_Ua@CkVz%`3~N&UzA#rG;(U^-A7f0mhe$#C&?HsG|!GX
zo?}Uz3#TX#U19Ec21zJ_-pC*kE|sOdV}^D+Z|54HSX<k<6`OVm$gz>Sd+W9uwSOq-
z(TbN?Asfot#x3ACRNt#R!_sLcMYl05nFYxE!kNG_Auf*Q@P1{lnQK(43?zU3jZgo?
zhd*QI5aInVcKuO?;nRPne}s+a0bz&Xbs&r}I|kW1g!v?x#|3aEyO7YRHU!V(B<|x(
zhDZ)|ze^Tp&}&L}Z{r+-M@uBGkbj{9&R+qU@rZ#<WhA5kufb%KX)S^`29#BLUiN`b
zonw~UM`rsliTeyZkk#CsF>vZhar(8}JDVagJ5sVl9O3jxT#H4TU|X5POE@{=NOX`|
zbP58^QUa(@dvroZO3iqM(72A-oHyHj3IcO*07@YhSGM00h=%l~klxIPXn%C0hI}li
z5qoaY&jP-0ik_*3|3rBwdsyzTl5bV%_U3qNDN*v$<(;y~lCVMN8qM3YjehyQ2JhqJ
zK1N6Tr=YIUv(S<Tr+vo7lxVdZm7BCqNf1nla-vGRC8JK3dQasG*JR`zD7DT9%<5Vx
zIWT2Jod%tVQ2^IMgDWnfwttbN07f(*4DGFwDaBx~x(@-M{5Z~&F$|D883p7zPZr;Y
zj;A@AE42c(WDYrtJraI$j9$F$#@L8GxnfEK?Ty+fy9m6hmY>j&kctV}Ye=2ne36bR
z*Gu?*S=Cc0!K~`EvRRuP^S<R*IrpZroR7Qn7TUhPY3@NK<*R;-5q}SIE9>kuLwLxo
zisi0{J!gH3jJoCTY0qMdN0F<Bt68S7VM9-~2(_7$h60Fox;~!tpw~j!LBsJ0<qfaR
zbz0IpCb4G5xUy7c{fl{l3|<T$>r!dUj3`TddHbce{3DVDsu(;I>JJ?GD0b0OjP(&w
zg#zm<+*0$^%&U&Ct$!I;h8g3ULf+d_k2y2rc@A63on8)Py_h?D46+HuSl!w8%#^kS
z`-w@1oG*S`?r}xS^)kP#3tJVQgh7;(^-Nk|Xlc18nRM4nD5SXwJ-4cI`f?5Yq{X@&
zcnf{9odD+R9=6e^!Io;%bG_1Yoat=UEgze!{~Rlj-jT}-B7c!|;OtIb(t)#{7b!&E
zt8Xe_*Wi<In&CyqedI`*50pB&58{)=Xjj{gud=&#1qR80eJ~XLz@Bt~X7C-;dM^79
zJhXzLy4bk`G%AFGs#EnAgmA&!lEANF$WGPfbokPaIslNc73r+}%K?P&F#r(K`(gmO
zla~Y_Juh)1`hUv-gp7LufKXdF2Ov}*UjQH@(T@WNWjX=Keo^%>cqq0q)H-&6<IIEF
z!j7`Vlzdg0ipjUr*g-}eq&{0b?x#Lr&~KKXz6??y&@P_oygv0Iln1E~={-t)0-rmn
z52#R(vtN|@5b77CK7``Q)Q3>nnsFI^n&`hE^&ylmNPm3@#du<I-!nH-pFZ6Iv7Gx*
zW<UK&MHR};dinzb=TwZ~2Q*GRY=$m=DU3&>eXXELR=D|kA`L=E6Pj``3S+opf{VXC
zjA{88g)wdKtHYT4c*$W5<>mMfep2PNrL_6em{e7z<9>PA)Q)#W*hFmaWY|RX@D*Xx
zK=cz~6MtbkAMEL%p&Z<7!66^qIS@pJ-*{-*o6RbW;S342TR)E5dTG0^oSiy>T4_OO
z2ONNw6KIj@FUPEwjR9n}oi9eL`#4E})o~JkZ}H{$(GGh7d=M)*haN;3UjQBhxsPK9
zVL4TeH~jHLwiV1o8amA0u_^lf@%i+BSYWid3YYRJ7ZaC$%M}a?Ol59obZ8(oIXIJ1
zA10R^%M_VS?6-f8KRa7fvXCTC^U$uIkKMUz6J+*4F`2hhmk-Dz_~;Be0d`_O|C;yi
zwS~RqFQW2iMVRmSL#w2c5lP_lQQ?m@4N@LhGi2q?mbU}U6lwtka1h8^m+>hV6qh0F
zA1;@y%N0Gh@XZvu@|Pfp77Lec83YHHkik1Cf4y4Gj@vd6zWXWk0?Tm7;U5qz;C0po
zdTX$^qNgIsDTn6$4;@OREe=InrsN!C*%JK?IltdflEaTd4u2fpJ`SJG8V4u50jGwK
z?*q^cNuMC;qBVm95LUzR@pJh5vy{rx@kR$qsWlyrzI!^G*B1?*MnIqgUSAvn{n9{(
zf11wfOvfpG_pdMU{)jXF*L_ZiX8MOZj(2p(qa+=cMj~nx<<K~w*}&=bpN~HeGA&7p
zVB^<#;=kTM4*wna0#%R#ogM!Dd5AYqgV0_L+6Q5D7=C>lz90VLuoq#H{v%BMpl`o^
z4D8C+q~jn-$I@D33LS$a#p7UY5IWwPe{_tbK|?w<q$i{m8qzV6-i(UZiq(EPdssxq
zAOI_5Lwb~CJ)}dPUtg?@HZURoQc-#wGwJk#rbt!f9@coFNdXg@mggB-vVy?4^87I>
zq_{!~speRqN+pXGUPc=ytkDZ|a0VZj?dS?vE?a(CmhHR<%J9(`ZA5D~(KX*5f1g7z
zA|PLpgyr><?F9r^&EavmN_ydBwRTX;a={4&_1fRanQYrr&<WIT=6BQRW#8-Y9cu;H
zN)Gd#$2<asb-)pUq5+QS(DCS&aIs)1f%#mQB@0Bav0+ISvsXEIYY;1qLTPnpQJ2xE
z1l5IVbcYk86`@+qWFp9>N(c5Uf7n^!oLjp(Zv-8c^Y^*93R<{;91Wp=HNguiE|t#g
zOlQaC1en8GTdI|Q7S_({<UiOK$VO9I@f^8T1T2Uk9Uo1prLPF2kT22X`?74W1-P&%
z+w>=YCcEKT5(S!Srp8QqM>o--c#cI0kDF|w42W3mR;fjqAe>oNi+x<Qe^bIJwN@=2
z1hL8Aq!z8TFs{<ehJe&DtSTP>(+@~UTKkozBPK08E=QAlTs@JX!Ai^NSid3sCd(n!
zNo)0vmcv*}p;9zhhDc$QngNp#iz{CjQmit<kY=*}Z06;$2O1ftnL2@}8*@rs0^Os_
zW0!z7bE{a=NA2OPhlrz(e;G4%6#19S|G8WV3E?ec({autM*enO!kxxkZMiVQr#r_H
zt<>86rmERdWi0{<V+*(P(*jLa;mSrTw;d(m4uvn5lLENafCOs7y{~7nhE;FSX6{d=
z7G$=f?}^rQ>r|LfjE{@J%k1^!C!^P?tCx<yG5Om2>>ea<IbS`de?#AF7A>NY6yBP%
zn&z2^EB8LSC7il;-Emd48vWkUBTHem;fNb3EyScBD!?!Mg<xvynS_SZteJY#u@W2P
zYuV5Qrnd67y7?voU#T2JanUg%S+e$ix$EH3Q=VcH0B-VVHaaGfG38@&bKe<ivA@8z
zbIGAr%|)GKZ+hwqf1KPJ^Eqtj1*i6z1MU|%d8dU*ctYSP6-02A$~Oc~-Uhg>?(1kT
z&{ys^1`e4U-j{<1z3}qd2%oRqkn5-?a7y%QC+WUjt0Yn)#`)#Ck@-~OM9X0@HHZwQ
zl~^>L(a$T+i0}IAWopbBa-|oi>93GJM#SRE*M+20B1Yn>f9m?4nM|zyVa|bTj*#z;
zJZRoAoFe+ou9eS)X*V+$Wrg=wPJUV*1!OVZq&H?pV|Ju9X%C|m-eNVG11-<=4ylny
zNdxvIH6`K>q=o@Lfz*@|o=IxhT6Q2cH`QH{8n)6lslk9kf_fsUVG{0;8a8@oQnPX{
ziPY3{iCyDXe^SHeu$k0w!2L-Llkh}R!`HH%)ZA7#CpCPfw@D3BN~SLUNu-9M??7r8
zVkfC#D6Pcel#szZfz&YM9Y_sB97#=4@iU=WraF((m+Wgy>N0&nnj>g?hrSR&A#&iF
z{H48_`1r%yq_a|Di?pPI)SBOI)`B`v{_&i(v^-9Ee?ipny21AN+O!g&EJtdUiMU@+
z>5Prut7K(8ubWGb@P)b9M7432p=N&3^NcAK9wuzVvv36hZKxKTcFn;z0x6|<Hji_$
z1HR1+%Ep4ASDHat8fE>XX6Ss#tf>>AO1h&R@>8;w9o+Ksp!DGCnRdv}oYvVvQ`*rE
z7wmA^f0nvc)`|A$XBG81_8ujDj@Mhz=NPTzD#^wO&nV|5@*d^9L>$Zc&8KOQVl)>0
z_)esbzOK@{e6IdJ#<HK{`BtIN*?`SwxzKB}MtpbicA?i!Q>HWz7E)~$Ig#Dx=}pr|
zv;kqzZQDd?qay`^_m5MYW}ItkI;;vsq+@z6f2-2N#{NiW=*^d6ra0zvIi_CxG*yY7
zHxyIP%aGYqZr<~<(a9D)Z_07Gc=P2LF89>-1MtF@8_C(!{`yE8vpc_azg}7j+Vpnf
z^ag4P2<GNCUpcs?3+aS|YgV>N8DIJ`a@r?jsCVA!WutpMHRWxF)Oc%?Z=Q1b=kBM=
z9S_P+)sxFVH8Z>lRye~meB<SGc7Fdj{14K5qFI;mDHjx%A;<y~12`}_lTjZgf3;W5
zj@%#&zUL`;fnhP&2BgtQGqXEYd+Ta$RZmrSPpfM8{SP$+l8|J&f71gA0rT1ZYy&as
zQ5p56JDj_llSAcoAn;b59~8)hktY}#WKteLvjcU0Qa`rFSeAB>hQ>H2%}YPXn|2!q
zK_nnbgWbj>$d?0Ywtw4`v}5Uqf88d{XPms&Zwn&zE9^ecxLBo(G!{ujM@f&!foKP}
z`+5G>nc`A?-<`X;E0#dV2<q2U_uOd)51b_?U74x?&mG5@2SrJpy!rZMt-AfweIMk;
zp3#w0H@)Ux&x18aZNzkqj(crJp#$xLRbW|@W%cq~JtWu`t|+N6o9fkOf8th-X8?((
zVJ2$g2}Xj9%7WKuhj^0EiV_zR#z`DChs@@%%8C+63~N{;+hLV(j}jB~eH_voB@1xh
z$3mklA097{YBq8|`T^uMlbdUh44OuDkY@4K__?%^jHFm>O5P+}!uxFr!i|)}8*PbF
z86m9L;;gV~j-{Pe8HGrie~kN&jLtN?!stx;%8b6zmcr=L7H9Z%MgwX>92mWV^%yk>
z;tctc_2jgRVJ@AS6zh>kdvat`ge6b#(0c`U>t1=MDNsert3{fiN1=WHozPPK;^F;X
zdKr=ZZ>Gtbc3*uLV$jjj91&=px8OARIG4T@Wwx4SlbrIA>!k3Je|Yg9ubCEo!9h9+
zBsEp;WyEG=ipZ=5ZH}1it9xyV{1D6ig^C~@abedbZsa6j_Jb~8@O|a!1H6=+Zjl$f
zcqdDtJThW4a=)_Qw1_61`drQ<EY%`+1UC5=!9824j>$J3iG<zuw}zYtB?q#UjL~tB
z@RMEkwn%f4v{gD;f3g}0*?~0C((Vh9Kw-edArgmTZ_-^YJ$*b&Tl7ywqhx7|n4)3p
ztpG}Ol}w_#cMZ`=JKT6)l|6c$MeBmb;1~KXuxO+9!-kN8b4?N(@N00`09Fc<wbMfU
zBz>r@Td2Cg9t7><c+<PRE`2TfHTY{0GlBK|ew9E;DESlUEt@%wefc{m^^dXKaQ{Is
zqm~s5OJ@9*n{>W_yYb$4EL^VL+E@Ng{sOqDwVs#pDHjw1Fg2GUh!!ljtjiUlwgEYp
zvFkPy0Wp`6Jpm_|(5?sye-DH^Lk?d+V8C{qMSJTax1y&-x2K@k{r`s!A47?h>^O<h
zgCbcZhcn+izM%rsKNCz}cgL6AmuEwhSHYs!)60(uc>wq?0>5;2@`%bAoL*k1KOX_K
z<Lk&50T{!V=jXt8yMHhkcm_tkaR1;5`IkYybh+zizE1Ty>>tv2f0mQi`o}8KmKFMt
zM@*Diz?b8hM8>9lcpmcZNc+EDe&0bgN$%&D-Drx5P7#>t`|IxKPI-)Zygl%1S^RUy
zX6Bbv?>zYBw^wbZr+;^Uw0ZNB?DZg<JuG(1FxxOiMOuT~g4sByqK*P{0WsYfCN})B
zAWe*(Wgu+EA)!sIf1u?V<mHX!dDfP>#AeVr&rcY=<!eJ$LG+?T-tky>b>iPFKh(Qk
z?8TbUvv|@Ts%~u^r8l8!8|peLa?<Z?LD?aVy1H+geJCDQJYd2E%2*d1-(3Vf{bC1f
zk9?cmq6t+%H(Hn?GVkKIbxJOvr)<|U#+p<g6@)lS<>>#kf4+fKM7`2iFh5jep3Coy
zy>Hly-$Fyw)pgm%y-t~zV<X6#!B<q{CE!-Pn80>Nv$nfK>r%84SmizTtI-^Fv4?fq
z!kW-4ufqkbSCOo}LcU*!WHK3nut%5BIad3jWLMQfW>xI++vh$$W?2nkD0OXKZ-=}k
zqldgLZ!YAme@Hor2b%;ZQkn$Gn>HtB;ZgFgSrjF6tNl>ccGk_gCmLW~RpYMB2z8Zh
zGtC@j8M(tEv!Ljh8N#JHu5<paZRo6@BYM-@Xq9dhCFY64Bw>~&)kHBhbu!!qjQm?l
zRZ{e+CUbNtBGc{b@QbPD=SnB-NKxSwW)XwR1WxC`f9XKeufL}si933kLSQOUj<V?w
zKjm`Kz4=yoI|64HIA?8F%MtP(I%;Jj2`HCpYTdME`dXTCtT>L+^iz?TIoVWz-d6if
zokQw1TGnP=YfcK&aHQO%H$j#(NYtl=G&4)Phwh$t^!}c@_Y#etgR<$p6+pSw2<$HJ
z%ofXQf5QtfG`Y)+OFb2&s+4}CIh+?cK6e-uy^%4Sovk}*X_xc2y^U3(TdwP*OMhV&
z&$sn38aYz6ZXTypLugUQvsXZBnT)pFsNGFh2ir{hxa1I$s;%3YFiDguOodHlc8B%e
zna3XQqwevptFGeRcF8rlvd?*6^vODyp(m)if8J0gXB(SkUe>1ypLr-OdO~ItFI?f;
z-V1&ym=-O2uBGd~uQT9SwcKU`Uazd1vv7s$)xHJ|<tBWT600US$Lt&z{H$j-ZfIVQ
zRt-iupm8UOrrMLS7pWqgQ8vpCPa>g`mvnkbm)Im6#jkA~=HA1emOZ^xEMeX{a#;3}
zf4!C6>h(DR=GYKDl0V?|aZ-)!A7@5Zjf^G5>#Uj?^FqC?6&$@+(ds*R9r)lk&3`hV
zbZ-`&K6>IRkK^td`cv(DBnG3<$5A4hLE`5*57VKQ&jB|fb|d#ig|sbMmn&BQzJTM;
zf%L|3`aF~KnIUkRqIC*P>veK7DCaEOf6lo#S!2C&)~qD{Kar(1Dspk$9nPPMF3mMk
zn-xmKSC|#1>D1=u&Tf;7i{;C+xP=eoQ{WnI#~lZY=UnlMiDjH^kMG|%a;F)>%2NDz
z9!IMpnNcFzkNXbzOk#1A2<Ppu8{A284IQC5y+087=&sS;5t<g-SD{%2A%`xae_N96
z9B3u@B#htVCQ0&M`lW#5!iH|GOG*fmV?ru&@7F^@xtBbC9{orF_7X3Jiv(9c`}S0C
zMSSHq&4gzCRlnJf`>mJ$mf!l>?=EC_peIgCjP3ZA+rX{`znyFafkEPCUUY<e%>#F6
zD*%~d+PCob(bx}>y!rKH{cnK(1puV11Cp2VDHjw0F_%$`H7u7<lOI2~i-;EHXP45<
z1`?Or2_-6*t?(A3e}wVh=ifeme&LBELEobK{`h?P|4UDlR+>qUr{BN7_y;HjlNL@O
z8IgPX>%XTzUj9`eWXPOZ|BF7qO@TC;Suh^tufOrrf9T=QG*~)&`a2B$UY6mf|DFC#
z3p@gB84rLZ3BWA@<|kmzx&(Frp`>FR-F6TwG1I`bU4|++e{-q_8`RDUQai4-QjOY?
zx`C1eTQNt#liDFEl>&dAUp4oN!$}E|R0K)+lt|yC5=aV$+K+-fFOMMARVI!IutXlE
zF6pj+dd+tQmghXd56|Olq~H7&Rj=Ywqi%)&dl~hB|H?7##39Yt3Sy-(5Lk4u#~6C6
zTj8QKXof-;e;4J>!9fHWlWPGNG{K&Oi&0*0Fv?VsPr-YXEx-Gb-hpK)(+s*J--NtE
z)GZY>DBexug4QL-ygcDa4=+hy${bV!+@zU0Lcw9oE$U{q%C(sV3C??(S?jgysR^`-
z8h#5A-zpq*wv5sE5uI>|P8`vRSbkMls6<<^Y|L}we?6knTi-|<1TxfR!>Z|zn@r%J
zI}2UYCd$*o9r}ICn_9p=YmPdDh#HmWT(Y*&-H#C@jGT#(0tLGKvk4E+J$a5`2hs<T
z7DP;Rf^ms3)+N^5BctPT@fq#IQ!G+WGX=%D%8Y124V<eH<-CZ;-}gq7Sx{6l8BtUX
zS37K#fA7G1#4qgQGOMU7$`S=1urKOTR=;KC3K$dt(C;)pjG+6F?-x31<)abWtC--?
zful{vEo$Yo*?GunTfxlx><Th&St1j-nY4N~95EPWnXLz<4rX-Eh7+KP1c47+B52O9
z;A1R|QB$FyKuy^9+g&5sNzJgxl#y98inghse{Ca<`6opyFL|u3TuR^}wmcl>EpN{e
ziq{jVd&lfC8aw8+_M!}GerL+oLQdu%ZSPXa#x`z=@HM^bQDX}1g6mzT#IB@dVTDQQ
z8;VneZCvLAYLkbd;6;acfWE$U|0W}FC4m`jMiDqYbt*^T9n>|^BSj<%Pb4`;b}^0~
zf9W^)$S%f@-s^gC#fVbfs)!B=VEOp#&r^5jmu~|_S}?~=zb5c5ElHyd53$MuhI5$F
z@Q=;dfdJ3FZaE5Vr;r0-7zi~&j<?#9nH*wBoMp253MRK}K4S7bhF`+ucDYev+A+C9
zj!zlVmZTP~VgmIulUl_U)7m{hsZ|^pe{Cg4QkTsj)?tQDP;3NWlhvX_X2rb4{Fvlc
zJT<x9&#(uAu$Cz)kEqCI5bLbMgOw70P<D$-k?xdD8Ip+9r3(L(dW6tJ=75=UoL32>
zpa5*0WASt)m8r(3P(!DTPlB$K0<c|9U4rGQ)TKZT0b=;qU?7tsn<0CRbf_*he@W|N
zn$xEun39O=G`@jnIAoO}EJ{X%_X2$>9f`Q7bfkiMOQHL{j9E<0%G^ORQvn&GWU|_b
z*YaDKnY?{?Qn1z~ReVM8c{$vp)T2`fc+Kf3kX;Kadp!B&!7N}Bp`R7A;=sF{H8-hm
zNSflWd*wJkhUZP5UF@W!&U^^Tf2+(sJ54?$j_Jvz=4LC;SiJ6J8fgwYz=&q)$cicD
z?rHXvipKSS6vuok&z87J2r$qi$XL>aE=`EkU|rb=e7q07M7-|{>cL6+mVE>wq@kS}
z4@dR1M>1_q;T<0~D~uMrIjoeLFcdxHxv&yx_jdhYIYJ}u-~OZ{IH;0~e{V;RwpMLz
zzhch0dh|%e9Lf=XeJvmr?bLg<aoi>&eB_Q9aD#zXkpc{Er&ND(27H|oP8Cs5-ba2z
zo&;qXkps=0nga<F+FO|!<Rb@LEMXDWrKlPCISP<_V6)f`nZb2eR*<FFF4v`A+W<xZ
zcL3ja10;%>aHa=1lechWf2c$^xTGzcv!za?Vq|ukV1gn|?L{l(G$r7eGreG<yy2}1
z50ZNAuw2l{X!)MV0|XifQFc$D_XhL%JY|G*^@JWULracch{g^V3#H3;*g{)jh<PoT
zsp)Lr0gB?ZxXjr=)|?LWdy|s&SRLTEKU7xmEPl7p4)R+a<9Cb8e<OZ(n9uWD9pkrF
zw8P7u-^-_M(>zx)B!nHr$2^xgO1(BN>5=ANcW+Zb`K96`+UHz~qL6?x`EImF;qZnD
zmgFME2otsJ+$J!&9n%7Z%t@@ZVBMMIj3yYSmF%HD!TJu!Tw03AO^?%@%n_JD(T^Tv
z4hLkOH_8r^8JTMve{!8EOm6%2ty}wkY^SCq$T(a}5pB)BRMl(8bU-dC$8$zC1;mBp
zv!FS8s6m7rPj3%o4U)}>e)R)95>_%{Tn}+3UMti~NxD+{Jk*24hU3FjWV8YRBY4fh
zxGjf(oB8HXvnW~bpmyooCJ!Zn3D7-6kX5xuKlViJ>^1!)e<9uLt68?^Z|Pr1ERK$h
z7`xvsG^kp+KY1#yqZ5XnHcJXP428_y=sMWiQEn*4-idF&AxW;8QX@QlNDvA+-1kT>
z8<ij>bN1OP-$H$c%C``=s(d?ptjar_qGb-$14>5@uN_@ZA#Z6L^|q2W)cxm>JxZ98
z#JOxZMdTf7e>>D$A_ECq*GJ9V7;EG?4mxUd*_1m4I<Uj%$a75RTBGxYm$KQ%V~n=B
zEoi2PbWH>s8SNu{1oWCazl_S1?Hb-o`@ZWEu%^;_#bP;N(jKOS8q?}CNAWfpbfozs
zwg)N#lT}SuzC(2Vba#BJEDYAM5ifbm__>mD3DVlhf7iV+RTY1(7nIZ@l+AldO{$L6
znh7xy>+NXQfHyx1@SoW=0EDWc@27VSfR;=dbL2>PGl-R#sn|L7C0zqPLwCGH$v?)7
zVb+XAG~?VQZw7Dos8wJ;mmY8*9n)~=6^YFtR+<8VMfrm|1wd#<d1Tq)FrDdh^;U&D
z>HJx|e|1ffII3$^GasI|$C0IFSl8$M@n6%_t4iZZ!X)PqIe){k=M7x}yqwHltMlvY
zc$FHH*lw$8jn-91#1yu=sj3eB-Vt*%nq4<)>fBL~&ZD^+y3R9359}|ETIwO_+GJ;1
zz5S_;bma(7Ue3xkz3ZjRLDc8oB^)u+C-CL8fBd1`%G^g3i_$Wyn>kSVpVapf=-qin
z?cXbkE~q)82Hz2{myf8|?CmR#v4s`PSXoDA@tTT^)<t|N>+*e%7O*}bpA?d<LY-7p
z6z}nZNim&EJyJ9k$`lWUC%Hwr>g9=YN`UJ(H*Wh$%}hia8ba%|)mzC&yB}`c?_^hL
zf0gA))ykdfovLrkZXgZqb*`<d<d-6b8n5>XX;yWR9vZHj_t89dYdDUwk<q1kpG9Tv
zKvk&0QCs(M=}5yI4!N96zFHQ^-0GxYtDO2bYnCKZuT%T&fd|biB$c<V_%SUlIo(nm
za@};rhT=p$k8?)FkRAM`oqlNbktV+we_2%34Hl0D{n8SYUwgY%?thxm4Z-bePR%jg
z#I~APHez{cH%#L7TvBz*UBh<Rv<<Ek)U$-Z72{?&3}0v)O#GMQgPg%EV8Sda!-ndu
zrCWr9t~Tqcc5rLo0vj}7`bAf8)q72ss;$Pq+ug0#9Z%Wp%;vQK-m3SjY|_HUe<VqF
zkIKR!CCBA_xPKpS!V@+=KXl>E3ZWk9S(PJ^%;NCEyII~X#!%KW+<Or)gVG1XeU1YA
znxuLhhwxB}@gKQX+uT{G(By>-FyV0Bsnx3TP3~xD0k4mDcJy+^)~j~G7~AxvZI?L&
zDwVeDJ4*$v)M3>w0>{~vk9%9dDAn2R!>D4%*=nqjel5dN6|ZC1^m+UobMg^N$DF0Y
zTK`Rn#y^-Fg_p6%7!(3DGM90`2Naiqqd+H@v-}pPf8-A4^!4@g1J6t1%$x8t+~~hv
zULXE`pbJ#aGAm8@`;Uk4LP=oKi4MGA!usy#pWU~IUntq3{v4qGxG`Ff^e2!!s?-?I
zWO$YIr%(q9SEEsn8l_jbQjPlL*VCinurNaTv___oc>VM9dq<x9`l(-@N4q&k`3p|8
z#lut`fBx`Har9(B5#?auEv2Z@np$L8aV6@Ko~iSQQ>7w2^L8Q^u0O4Chdk0TG~PjZ
zs-3iB5{-MqPReewRk-^ydbiyKcg%7_<vC(EuuOoa6!+Rq8Rq>&t`@s#jgM8MJX<(@
zxO#qbPb$llwAI#VE$uS5V#!OE21eW$ce^nRf6P2(k*??)NR{Wc^<H$uOUC@*W4Pw+
zSG-_6L)!hk1u$j&gg2CHp1)K~$ccvdC1^neEu6F6gC%shSN*)S$!QI<)=uQSPISZ>
zE+@N@(245k$qk<pPBXkFdevzqAv&n*CLfYqFv%sQWccPCw&?dtF^2||hA9la#W^p1
ze_~m4Z*7^;l2T|eO5G@9O1L+sA+FshU>)-~gmUpIWXJ8O(ngPP#Wn|5Fu*)VCleUd
zz_;kuDrPKD3JpO4oMeEIg;3&ZR=<<9PF64rR3e3mmO(LX-W}lxB^ZKC!hUI$8rVeC
zUMxX^bqLZeVYvi@5u3Hh=p94mWD*9(f5Y3p_$HSOmskx{Qr&T{*9el0l2xBM=mn$d
z9LBdM&<s}z#d%l3$n%o-7;i{l34jt#w^C;Ynq)?eeuo3Ic46^qXrLITOw59Zjbu5q
zLLUfCgu;Q)L?~?xO>_l|*nBoTY3-Rc{{dtl<j|hVv$S^O7+!QVk@9%RDyW0-e_4rH
z5_E7EI_r)>ODo45w{H=&Bvt{+-7{#7DX#m@u>h9y9^rZ`fTiR|CfqT83B#Q8_q0{6
z)HM2;mv2FsE6f6L4=2djm{1G`ZrqZ9;ZA{}6j!!n4YA`{bxZDV9Z~)vP%l6j9H}(W
z8E;qq=4Qk`4b=d=prQ7CP=octfBN+%Pb%n<Qn`bCUNY;|l=8&Ow^(7hfHh2OJF$w|
z&7n^-oH>~6)`c)vlM>n6kc$wJ_jcIha`4*>!tjAu7-ZF|Q(iF+>u7tfpA2}1!F|4A
ztq>J~rVVR>CI%?|>O!IC=OR0Yzh_7qtgWRBB)@#+-5>b!GY*#Cb^pYne?KZ4-u>16
z78>F!gt4XxgfTcOeTy#d1oN0X64jo9mOfTRdns&`3j>JqaGNX|)N4xjR?8nknw3Z#
z<IJE-!gCuaH#9?p(mZ<x&(Sfw``E+%f$$?As9I^3e@!nEeO(+f*j^GFT@=X%oZ8@`
zbT4c;^-`!iAu9S5*Hb*7e@YM`nrM-Nz(9ffkWk>~m01FlH=U$SIT%Z&U}}n{W46#L
zjS=l!pIoAF`Cp-(nsN`mc}Wz)v78brcEV0plsUSGsv#>Uj<FFga$7#XjU!!Yl2=Zp
zVP9oi_U;%5%P8MBd*k{pLymP+-yF*OZlXMCG}Bl<6Ycfr<EHARe@!I}nO$<s-s41x
zwY-#*=nM#0q-j4*%*jPM1e?&OwwsFX+a&`vv@~l8*M*9;bPIOK7{TDMhP^hoa|l<^
zH%}qR=n4Uwld{CvZ~f#HS!rg3Z%fKj2U+d?ld{BeFhuGDL4G%hjgld>;r?y1vJ45t
z_WN6>il#K%ii;M!e-`3fs!wE&Q{`x|MAdjKHJ96cF8`y)^KgO_KU9eKb9xr$ic8^C
zP9cU>(N@Z{HCEu{dQMU5OG3SIed1KQM_~ufu^R2lOQcw@8X8wXO0zXroCJ3~$h%{=
za42m%DFxQv6}u(dl@o#(uFeBYUydq6Bgc%Ia?8ib!{g3Nf61QZDb`BmJ?}dbCD(_L
zUPINI@|(4^qJ3~IWYD+Kemq+KA1J?BNo&g244ga^x%ZOTn5+P6Zp2{k1J8vEt=zuC
z0*ZwzTD?>}>-zW^4_91&20ZxGWC7EHd=^8bTru~J?6DcY3M0`pbWsi1ZV}=_g1Cf3
zX_M`RxB?Byf4!qvN4f}6k3!tgr4xgm@68RJ)nIzNrh~JFo@)ZHg!|;1g#AZegd)g!
z@LX1}9F`BN%?U!+xf$dW&ZKKcd(&hzDjx%MgVxg#-Gq1U<1Xkb_ovCMmf|5FFz(|o
zKm@O*>@VZf$6`=v5q9=xr?Ssl1^h*qg@FRCL`fn4e~AkiJsB!69l#JK!Ce~nO-`fe
z_a*Hv6jDMnD<;b$N0Ynx-=h=@8&6q#&;F?3>i%U|aK6yoK9J0Jlh`I1IFG(7-=+=&
zk8x4K!T3J$(-N=NpjrFc)^ujjRn!q{?k2I#c>`HA#ydeX(30u?ELf;yE@2Z~N`2ID
z0$z&Ie`@rc(n1IkPCt<^fg=sS2IO*sF>xD$xXgd0I>_@xKO#_!ZBvNuR-3K^IdUD6
z<Tm{mHig=`NZ_BSD<73KQaX>T_g)Ud`MGZqFLEu?naP{WxD{N#ZMEa9SG+)l|25LX
zm;{1LGaLG5c1#r$2yP367Zpb9lYngf*tRL{e`!8G%*zKTi%=VPwh}<)+lZruuo3vu
z*=lsex5yiFwL}y?$GZ*SF5~@BP)t{Op+^ddSGK~neyP0JXh(?8!Be_9^e##hzgL;8
zo2$0!12{073fNJ4$G2V|u08-uB|i8H_x<oU;k<&s$>@Xld%}AY{wA{y;O~9>GybOR
z8|U$N%Fk~LzsEe@5q%TJZwtQZ9e3xytaeQom+>hV6qn)OJrn{qIG5oz1SXa%7zCHP
z1s5cL_!I?6wk&y*9AvvC3dG|B2oP*J{5Htp>)Yk~+vgkL;Dk5I>EZjYgQ5qLK9!`K
z)(lQ5VL=Vwe+)mrOQ|hg7rIbN0J<Fg_VjH|pFnvUp#oj%^vR*3zd+HY^S8dybuGXB
z=~H~3=%jbud!cCZ8~U7{K&Yc6U6w{7f{F8g&^YDdb^7J|Pj9k_r0839-yh%K{&?d;
zSz`rA9RBg+Ej~a{0=ynH2oiVr^IyZSZ+~kPDrG5r{6(K%mq<D*ympb~&%emwpY-r2
z8Vtb?|4u`Hv}xq<`|x*~;1OXjWFib?fN+nn9JB~N6tW`;13^d(&y(25nbpGj`&5;G
z6v1$^33gr)?4-5EOkhWF(*}sP5rK#o*dc++5X=lg2ML0P437jQqsu*=?;`;PZ}~C=
z7Aq6TWc;NgXfT7NlWPKyrm?-EJ<ym5!UlE7md}#d2wx(LLGOVt_md4l#cC^1*=pSw
z5RxU&HHah|Mtf)HUZn?47^65nd4BMJRuE1G{#L|G4nxM~+0i$CE}X-Jz)L})ZFkRQ
z90e!*=%1h~2sY)Nh}uzJY!u8%N6e=A6PkOjBJG6jkVmL51<%1I<I>kx_Sa$nM2c3y
zSsnnRKG9>1fQ;-G?<IU09IOQ4Wm_~z<jlD%a@`8oNUybzKYtt8oo`=9#(^S#HciSX
zlV?wZe;JJ^$N|`3n6W`OE_qEdPz(nRbWgDtXqEJy&q(XwB71>?8^VmgHh+PLEZ<*>
zGsGL$M3F1p1T8%35z5tla%fC{Upu!_`B{|SENSIhG+-Txs$ft;77a}9^ctv~cqP1t
zW*{hF#S!`m)~0f^V!$BCh@9GgD!;cH>J5#6kKOiT^VQ{|h0tdxW*c*gUv^U@nS}~&
zM$s&4P)ig@p^=S7`Lu4d(lG#V!>XD0Jw;R-;q=VF3YLY@fx^|yJlX8J#zr^7Y=;YL
zi-P8#$$3CVr-jjaE@KP~GjO%Hyb*NGWIf=Ljv_eex_8MEYHC<XdUJJu?Yi%hQ)(Mr
zvd*PT9;eG_9d(dKK5eUto+&SDu29j#I3Lj?izccZUC>20=X^(i(Sd4<Fz0v$1Hq?c
zkByGXI)Ver@d7RWFT%l;+CdzcV>p=d@(2fv^ZyABwsBl>Fdcmq2V_AhEyqDxlB=|R
zjTP3HX#=T5U=$x?A*Nz~32T>${vSn(%kBkU8zYRI$-d@)q;E&H9UGnWel*q46xXTZ
znQ>z7uLcX8T#umuBDP$Q0<>dAU02i^6>805+zJw$uvE7*sxOEN2W(Efpm&_7HJhUr
z4iz6g+FT#7dCe^tCo49$UeLA3$#p;8*xa{cxlN5@+a#)SjG<$HrI^uFTKc1VKTnqc
z4#{T}pZ$Q2ln0?);1pMK+$O!U(QRub_n;<}_^^tbO&|@iHCCPFjqLExtBr*7PIIcP
z@_i~mffh<zPEl7aFkaUJ(e&-d3tXv`!w4kfi;gGyJ+<ALtl}KUNU{5)!h}^1`-{#N
zIhZl*vR67QoO67Cbe{w6*r{p0&dj&pk3d>s%#6>N5u`mZxzGfdLLLJODq~*^c}DdG
zkY^O1ggkqD4&@y&(Nq0l9!SdL9UEO&@a!u(`$}0)_pGIPJA$)>rKn!pK*)sxXKe?(
zfhd4}c7C)X%(+Hc)1aeHS6rC`=)eMB63+!Ww>llcA+~aVDjZ?D`JI4$%fIENNbDdx
zNcu?pP40Y~ZSk^6Vw{G%Roq13FtU|KV#y>fX``-S^l<NKYJwzCQn=|wtanJn&$l3#
zcGp2_0TO01PtCBlH(#SDfBUjm!Y&)#ZZ?QnmiBYIEkX$>{=_{nKKi=U^?0TMj5-J)
z{k*UNj8vh2dFNl>07i-(_E7)#SqdBZ0*a}i{+j+TM%Gp6TQ~n#K3)6CW~_N0^_2JP
z>yaH|;i?Y-<XI9MU4qCOg#S`g8A1?fU4K#~nNyYx24u}u*BYy7-R=KMjosG#O@P43
zW>{<6_JGT@Iq2l=4&?{1-a1R0p_9*GN&Bw(zOMIwTbn!aQUM+&v)KuUSXS(7d-JaQ
zG)jIU3B|dy#igXAl%H?L4`fEI<6hT-jcuSjYHX#1l69R=pJEyjBZcqUpC%_NxhPg}
z{wOS0^<rdGU%Cw2jxEZ}+WnIRl`Zr_*=b2VKO0a_0J!dqz;)K;KXVh{d2HJ9eflX-
zj2|U`DCY}uWTFS)qvsjxZLC9r)Onp=eA1_ZsDxLROBj!GrCHoEPtho;)V{4$o)mW|
zm@ulft~kQ4M-eDp<@{EipNnR0oeQ};4qAhV@0`ol5pL!#hy^ccb(`S@JjqS)oyk4X
z)*9N7>R(6o@8oR-7k#&tj>=K#<5zS&Lo!Hzp><n#)6t0N&|WT`dGYEYu1f$6wqn~?
zYV<2>{Zaz-t>0;GqxzFKjf)hn^X%uYtZ)<h*v3p^!pXw>G%j2;iwiEiKkSVa+Nru7
z&kJSzYUAUYCNM6IQ1krR-99fS+-h1Xnf~D~;C$R2<w&92+*{9ih;@6>{JGQfWv=Ic
z^+>vmHRXkN*q;(=oK}lo#>Hd-E!X0W*RGswixgij(^7Qpwn&@b$Bn`Ko?ngvQ!4~@
z$A2WnCLH5VIHkj?4KpQX^vi2`Ru=9)uB0<#VO)*S&QgO;D`}MMwBskq$8n5miXhe^
zml0IILp4jT?rmVVWFQ{~9@B><`;A9`hKiiaC<Yt1LV$C*VlVAAmEQg(D|bUlk;bmC
zE49<rY75<Y*W_eiO<j}iKTaby&ZE}%o&t<5uec?^t6+K>%>6@vGVWl~avVSE?&jpq
zx016jZ)MayqSR;|^SQMI=E2rRJsilW=`WdLIyqkU<u<VWkY{oO596VX>*#iWEx>Nu
zl2J$r3>Z#v^*rK|AyUTdg$d5WQfjItr5vsu9e`36w{WIP_^ah|hFM9C(}=~;vMDX5
z^jugsCKEWRKCn-+dM%b2x)5i#g}C!1fZCiUsQ0nT=)?aQj_2+P=7jjZrm0*@O|-S4
zD%$U+l(h-x{p_%=6#?wwv%1QE#%+Xg72X7px~u<uM}cGs((zeS5vSHW3#B*9-rCU(
zuRRJzY3&?wITp(2-63~54$9>|(CaG)JUPW>NlMp*8++$*2CW3z@IV;edtsyZ9%m^1
z7fwy>S7bRR!lg0$DVA;zht0=+isksyOylJ^DI7ke%6a{=avUiiCYd>Z#C|N5gCzI0
z&zs9LEImDMZHH`gA5bPn&=>g+Xb{Gkjs62MAA?tj3|dH!FLWJ<i5K{d#sf-sOn?_e
zafr=7bgz^oo@$l)VG!8eaZ(C}vi6l7Cq<#q`tbUWlcKkbJoImBPm|clnU;?_3j2Fb
zii|NHxHlU!<=hfVZPcY|joP$Dv90aYjzqF;miDiU{Ft&Ex5r{u+Z^dZBWk}A9~ei)
z|Hi{Zuz8lmMo%HKaG&f88&Pp9&$TUuWX(&rS#{z4m+3|W+bB49M{o;T#{b}I?}h29
zwz8#^X5s&FRK5ur{{f}+a_N`xDHjx%G2sOh12Zr+m+@vLCzps27Yu)^SbPDC0k(T~
z4EPpjZ-!4vyqD}DzU22)e2Su;Eoltopy_UjEY|x)YUJ(vP2PTfxqo~4>0Qyy3a7#1
z?b{zW;0Kcbg5;Y<-z<PI3U1$i+<yBkB^ti&e36t&@n!kD<4^tiNrB@L;Q4~rPnLjx
zDc}p!C%*Ib$lu-Tr|^F~!ijV3=L}J&8TchWQNE0ld>I}|q;yz^#{sVn?)CR?zr4sy
zk`&&Aui>Nr_4V7!e_nb54S0d1ZvXi45+0BOp&Z_f6qY&i=f7@$y!@?)TGx_1Md<K{
zzy10B*4_Cvv)D^!u{1{O5sN`G-|1$I7b-kTEGA~4W|nGZ6ET0gH4{n9HjUxlh}fUr
z9n3-|FMxW;4w;8qXGjd0$|`!eA6BHV%xe}heMzec3iOmek<8H@CbO)Jk_2h!X@a+K
zb%HX!@C$W<I=b`g&x@Qxz#xfkk)S1!l+3h164Ag!<rB%}cVvU!#Za^;G-AJ^e1H99
zB=R?Xf1XuN3%-9Qtb%flFaU}`b-$~#yc4V<6HFd)bX5Rk<!1FY+6l#+DykK1>g&^(
zf--Cq&_e^7#Sn-Esq|!R`TV*?jmMBsc8c@_L~Vu9YC=iyBPv|SAzKp0PIx5W=6U3l
z0PFkF`hHA4o_(})d*$P`@T4w~58ICbZzqu8Fc+))=qNGAA!b*9(E&WJLWm&9_zzGa
zR8EkyJ5>nD2tTFlpbDYZ9JOet3ZeGGYJY_a87KJKOS3f`mF-nD7YE$YtmoHuUTC_4
z<l@0QlD{5S93_p=+HMf?6hzP=2<{Y%LM61-7l}nF6w>WgMHOO`au>*j6HDR9Ug#>6
zRve?vP6ewmo2(#zWv`0aV`xijmzdqHeIjO$Azcx(drGd0*<;9CF}oXmKM}WcB;F;X
z8f{O@2&EHNn#sCyeqFEvVWJyHoi1u&aiSkffd;+s)?GofB7tgZU2r~J?~@*(;F5Tv
zJ&#Pub|tx3Mi#%p7EjEpA|IU#Je$i&Jt-getU4)HS$-OSeC1y$A?;cfUU~7IAq}6i
z^|<CziwySlIENSuQ0nmlVOY<_<s5&X0qHph<Ur3LfBRkD{>cx2;lawg+kfNGA9Wdd
z`}ghdq2S*D%pp7g=A6K?Nv;5vx3IR)U<VLd`IrSf4Pqx|)(OX9aTP3!)axE@Y*;L`
z6#%#}SBx`%3PAIGTOm}-@VPV>nvq#QLz{z4si`*_AsLqblJAD6_cE9LNvX``U<Y*1
z*x1JM1~!iLu?pxgC$}V|#kx*zs?L)Tb{Y#qHHU1aH_9E|MNNuR@F8oX>P>P-ut<$u
z=`Z3v%hP0Sdz5))M3gu+J)lR5R;2;;Y=@A@@l)!5JrWxMmw|spgOF90OFjC42H_ZT
zv}8-qH3{*4IREJAy1Yr4@|JFu%`V}r_U!Burt-8~F?9*yXqPaSojt48n}kZDP*P10
znQs!ZbcIwyd+mOz;OL-jGzq)W%&MaW$If*_Nyfgc8-1kEF(ACO6YiZTbS&QiGo|F3
zVkb9X344yetri>I=%Y!0G5B3F$}PL|BB~ysg9N9We?1}k1jJO~^bqCgx}=AAL4z_k
z33sZHA;GI%r9ycJW}WjBPkdg7a+9#9`FOVuJ(n_`Hg$-jTTT>ohB&(O>$G!3muVFj
zB^Seco>}jNtRZF*w!IT~;9SbV(`U{1^_Qg;7axDfIRW4<3XC1JjZR!Q`b1#t*If^c
z-RKj6vFG0Tz}QnFPj~)$i1xBD+skh~%yy$shS_}*&tZ06;1g4@R)F?bnZg*_mMM%!
zpD~4f>lRZOk3M4x9rrFUg`Se4xVbJ+PdECu+KXh2zz=RN5+@i)@NvyW;=%|VUD;eD
zL^gjqdwp|}z{gJOi4M`zAa-KLpW1v(Z;|-ZkvrCr_<!s#9%km2j5t^42zN$%o&~Yf
zU>K~j_!o5;l~wF)YIswRS5Q%GLCB%0sIhGsKEfR;t;n2aMkSHC38_`zmW<<9+mdmF
zUT!R9qnHqZnz%Tmo(L?n=1E6`g5yI!tE7L%iGuhG8r9DD(tE!5Hl*hn|IQFHC*F#A
zy&`-bj5L%P%3$BrO>3_t`kDKWR5moYnu^DEQeQ?sS~;sp0aC|de)3*4pvYL`I2$TT
z#c^G-vx=Mrb;r!N_tJt+++A*B*U|`5(KO{^of}~LI&T6g;*42334&3%h#PUGNa24=
z;&;S1I@S@E2?Cg`uYuRO$2j2uah3$Nx2RgoaJda%pV}-RRpx|)zVTOMc~wq0;%fR8
z$F18Bvi%_TxNg~xLV;-eS*2i27#E$+nwe<A7C$}VNY(Xn__X0r)!fL3xYmO;3&wg;
z)@?1e4_gXtYbB<Utju<)pp3vlTIqj4cJu*XU*i$$gn5NF!efcbWDUAD*hjQy%_=$w
za$d87_aAcGYU7aLA)3KDn_mSBj2N%^<1}Y0{Ab98a+R+_U6y=?MK%9j)J4N_o<n6v
z-HGjpCH)OYqBNO)rI-{wMm%hyUZz7A9#-0Mh<(s<y2Ot!bu_keNS>vW&NhFSh+~)K
z$lu|v7Pj>CSLUD}ykHp<UqawdBUd@4mEvN3A`|&o_IjLAlccr|9(i9sB%9IFl4pG+
zU5)&6^=_MUfLqHsz|BReaYTreZ)Ou(*(nVh*#rn{`bisRv5JOwf;`SaY!i>Pw^;M&
zWo0KJt?QvRNyi#4O0(ce&*FdL8kYc#`wuwLI!<1RpjBg_OBy9Xa4}Rhy<`T#t}IO@
z%y)Ik##|XwpZjuG_g*oLTlINKXVDEASoU#g?w7_pmH5-dqcm4_8DGL=3T@dxw^(#k
zn#PAGxMsk;^`N@G=27QTnS*cXd3>QrRW8e+bfcUfHQXpP+lf>LZexFw8|cPgVf6=d
z@(&xl7FJF`w8`<1T-s@>=`xw)QxK62O>%zp6a-4cregcniDmDgVEq`xYT2JUKWMC=
zrU5#U#^GLL2N~r;uaSbuc7xTNQry#9Y}M<j>UfXziMtgSHuTF$(H_6Le%1py_MuVj
zN*}^lMgr6gYCCRpXQqE0huR2gQWcdQ42{c*m^8dc4V`CgYQA>R;<?!cqbsK+PQ4A&
zt=DzIkDSx(A-<h>&n5N!dx%RXf74qr-RQ8HcSgFTg&xUv)q$<cwuQArvSlTF;;hm0
zE}*<3Gs13($2#;MJ3@xn`}QN5NPO;H%d*b%2e3|vpA#oN^xA)m0z_ke8^0(SU5Z~e
z@VsBrLEKgYC5QB?;q?*axSu#!kd6v;*e`9GIp(K=el*Do99KED)nlR;&UmGle&#TF
zOwW21sfSl}+^~<c5kMliB{?pv04NNO3x(Vt({7q^5{~PgS{3P3H;q?+ElrRB4r;%@
zExTx>GRJ+!n2dkPmfi6UwQ`E4XH=WIx~JYs?jTjCaCn={kWe_;B?EB6vxE<AlYwKt
z`0IDcfX&xSJhRX7G>Dy;nUpnr$Tk@mFHkk73$pW6t-1N35tC2+)NzX$?8a`@aiTVu
z6?Qt(#_T0<E4~-Qt@v`iAW?siaLEb<OVYJXX69<QWm|u8hcVvZgi`TU(Th33MX|wD
z#OYy^K{ri<90k(pGH9CtO|zH~Xr%Zhu5YsIFj*VyLKGnF2eAt^L(-!?nQ}U#$+hNo
z4RRozSFpmn!F{~vf6`)72Rr_2mvHZqB6?8N`20Qn+l}xqVsNjEIn1FA&a-s^xewwz
zaEUe?=2Qf->uyn(y&D%Kf8q{yae#V0uoX=!-99pN$4kTfoxD0s(>kwQjMZ0%wriml
zT&)whb>`z}T;tkFHw~b{o0zuonHx)u7aFW)y_%>;mG8|7wb@W1O+F00^N`MAuB<cL
zq&WW}He{t3+>vUw%}QVReaq=91D)3|o}7~1g(MPn_Ws2lRrdV(5+8l3onHSiD<)+~
z_xV3L({F^A@hKM-mjOQo69F@qF^M`Sm%<zuu7AS#{mwtXeSgD~NP@ma_5Jd8^RJt^
zP)czoxIF#!^NqiOAQ)JA64Egz;PmU?Pruy!sgTH|IkWy3{rjn*X&EP)T5I|?p5~7~
z^V8qx=J)h4ID7hUc=WS>oHM7~>7S=R6MbH5!3xp>*IKDY3&Js@B@uTF{H>$~l3Nk#
zNPoEC>-k=DkEJ5rC=;)$^n@bB&<gDVd-e}XzuZNo1z2@did0EKo~%r&kdwl|M3^fK
z8fX<b#?gHR3<G$TB9++>f>&9hq%*u;ZX`VLnLZwm3@l-xoWE8ecUcKZA#{gOwGG??
zy+<UA#q=Rk0wQ_374JRng!pPG1`s1ML4Wi=24XY}!3->T7w~cg_=NU^J0#p;`SH~G
zOa6N&?pNb8z#V`ICq4*w&<u?&rcLrHl@R6K2qGs`3-)wPi3;I;cwH^m052(-1o_T*
zNz0@m+KMURGF*y|I(UgLi#)^Baqagqv8zQJKt`vyH1IKy31mty5L5pr#IBZXfPWY#
zA_70iDLF8LW5;p+e$IHcJli-a8#9ylqMP}BZ@PtF6Oa0L)+-sI0%DqyWs6y_yjVRv
zCcSyDc*z>Zq~V}lku+@MQ<S*v((Co|3~<u%y#5gTCqbE!C7p%Z(%byXR7X`g;`t?d
zM@|v=6BQGdw@GfO4Sb-{DZBFDmw%Xfm8{DL8nqMXs{t7Rt>OXwF+eN-U4pYu@<Vl?
zrYs;DTrtU~)(mqkF$JDjQ!H{VAJ%u42-izILJ}C-L4GEgvzcs<wSy!^oaK5GnB1hA
zOHe+eJFh{5&XqkmhEj6ftJswwc(EKVu^#ufPfkKMpNjIJU_B`tjW4ZjN`Jx?%zV#(
zKljX`tV)SxN#S499=yb|QL6zRGF>_izO#Urw3W`?g;Lj=Dxf}1M`!FKsK>-=S%;5_
z9nrDvEPmV}0TcH1Pv@$PZh-Y_Q`&0!){ai_U%k@1cYgR(%s`@GRD;h1^wsvNyV<C;
zPzb%w8XSqJA(JILY>5i-^?%%2WKPIAw;Axq1j}pEHD>Tkj@OqMs}r2{8K=%NpL_8M
zzh<T=OtumN+rn0W(gN;aA$0@QnxP?A%ox!u7I2Z-TDv%-FmzVgKF$jhHgrhHQa;G1
z=pvLG)ZiqGo~~|Lbb9JBH5bBP^ee)BiH;Z8$859yDIb&lUQuPuUVpNtS!zMk&`19Z
zA1Z&Dg^EPsV5mr|U^0bXxZqxDt+LdYaz;htu|+;?sM<ssrZa8NewkYFdpp-oHt2Mv
z@R3ll$r(%ubPKD{iVnPz7NTqU@jUe0#~ea9_ESSZE-TShZk2z%PL7;$s$A7Zd`3E2
zvwx(L(8(!=pM`c0k$*ifVH2JRW@GX}+qk*#mhUtaaJ75Hb&+=T>}bGwj{rm5tyZ@>
zYfr@=kIS_wzQt9tHn6JIDie@Jk|?oDZWv89QGab@tfh|6o2vOG6Jn5l>&jj0Gq`XU
zRK7;Jr~0^Z+eNyn2F+k<=;h-wbBH;hEO<#*dhQzURLJ{50)L7yC~#*Zt9^3wFUa(i
zf`&D9sEm^7Ss;uSvba)0W4!-V;a<C{YiKW}s%^s+y9oXF+-?ip>zYn*xTgl0HUY_8
zVfm#K0(!xLsVvTv9We49bo&%T1)3D-qAc^UR@Y-~DzvZ!i6M8*9iT-J^FCe>Y{HIt
zxIWa-_tZgg=YKF1Dt?$Ui#oK`IxX}|@T-s1fL8NlKd3Ga&`M+TuzNV(G~vG=Z&C!D
zB(zQDHR&Oygi={wVKDS{3Mh!_;NohP3WC;5ThU?-S4ko-m2K%M#YqKaH}mp!qBQO9
z+v(D(wSeP{Pkr65DlUrr$*T2w_M2lsiB#DG7M2pUoPRy~<y<o3Qn=`<*XML7HtFlr
z?62;)6O<qz=hIF}H=WAE&Kc2C)^hGk2LBKJ3I@|$`5~30j`!OTWK#;sjx5vD_;Rhp
zN}H}WHT;QV#hjXajugXyICqW5Y!7)>hDsydoNx6>tU`R<#Wzxs1!_()6bDeJcs=bF
z+S|1=P=5@TR?iEM)v9bB?afMryTVVJs5Di6RUuyL#`0W;zMNN(b$4*p-!dmTzPl>R
z6jyaOgxzySJ*UD`sFNejNsYCZq@?%??j$J+IX>k~mpE(3yjsIaGiSi6b~R9xR<STo
z#pUg9-~E1<7Rm0GNS4UkEpLBION{OAqa-sNXnzSZC1b?Bm3MVGEst1B0>z5vR%EKC
z%ZPI1wYp{Gn{D3x+YWe)ccU1`ZQk=EHScnpcPps&KgPTBBY+w2zqgEdeIk;WiolF<
z?DpT^{^#2n)PgzZ*(T5Ct|qhvwIu^-mi+N<(=rVws8zEhS*CB48{D=`-|jFZh-E^?
z-hasLmjOJMS+p&S+o}alDn%JKvl4r%ogN{5fo{#R&j@fw34v@G`LfN306H<r8Voss
z*k$}cj<3v2BijZ9z>_Td5u$@pg5E@^JW?Sfw;7>h%bVvBgitZ%T7b$XUHbzHD1Lm~
zZ1NLD)NR(ZOw#PfNE_y$ue+5^=uvR)kbnH70F9zm1h$HxThhXcsrE6VxKHQ%Dn=`&
z=X^$B+EoLMJIBbEkHCuLc|>PL<=tJ7JdbkFBe<eKrh}EN;_4p(amc{Kf#6j^c^Pss
zi^HoyHb<~})lZZK48jfJBWHUJjEPb^O(zH){dAJ-vlIfs$N!eKk)in1QO#HroPR_T
zgQ#PfPcXEgK&tl~r+~<eneFt=-My4@2ZGA_{`(OCH87+KrvY#fj_5cP#vVn`(kc|k
z5p+NoZ6CrugG8Q5^0OI(5;qw{;hkc}pj7)3NQ&u*stpEx<~-0o9jI3jZ))40C<}6S
zwu%oN*H=fjU6Q?ayG|lgV;1acAAf)ks|bEo6~XH&Laq!V((Z*9^RUqc&fS&5PpH~>
zsh=KG4Hd7fO?=-$nzfoJbSy&uHowAMt?hJwlk^MBb3MuHF0(t-p_IO+r%0ds4{mx5
zqSUQ5_jz93H$Q>o%RZN=s20yyF+WOkv^VmW5ib}j^-vE8=!(p(!}P1uwtq+IHo!_W
zSHa4&qS%0nyL=@mAbe%0TTs4if<i?+bN(c4K?(nPAPgUK&%gE=?<a}js`^h?pniX9
z`n&fUaaxi5W{qM8`l;X9i4GGVqq!Bv@oL+no!hOTT*FB-x7+EZ=hVU1)%0iEauuaq
zrlC8uTwq+zIf;f6wGw6z#D78R{9ew{H1nWX*WJz1!{&*5GnsUL=P^ujrm0183X1uq
z7gQv!=P&VoXS$f~Y{j@8=8cr~(WM5tfq5$`x#U8L_*ffanqW8LOd&opUw{SP0PA)E
zTLt0}V4)s`RI}>>wyoNBx*4!_P1O3tW@Gr%3H6&U)Hd8!kIHc~Z-1;*pFao+K(!15
zcc4U>QQ|unyRy_#;-YyATIuPmY-lX=m*yLGtY~+$7wUKUI^(VF(pxd>L@UC;5vU{+
z0CN%2q$a%1W|?wELoZ_!+IQ@m|2T4RI=&|_r1YJti{;B_vlmYQc_;eitNxcrQ=hQ%
zF<&V!q~+WW>zQ+UrhnJxM$zlS$E$nZjzE>?%QkiSCw>b3EYD2J+^Mf;RXx6|d03`b
z?ulxi|60+qW!JphA-z^FyI_v{CG1_MGa!V$@P-HsPFm@h&i{)c1zfe5jp*q`si=W6
zY2KP&bY8Psyp-dnx4@fhrJX`FbQeh3n*Pds<X}FtCm2`LkAIb@C#hWX#UGuACLNuB
z*{%sm8|3QEieZC50+n{NZ3aUFptwT|keQsZia||G$q+U>e0O&*v>r}YfC*2s+G#L(
z@Uudy*k=fcJIV;mu&3|0arN2*EqyDP6C@*N!tLqk1BdX+*M8Wk(4&f9pPZwUME3jn
z=?P$#?`zfyV1HEgG+d0j`Iga{&{O_epG;9E49ym!Rz8v#>IgJ`RBYT9H2H}Nno%EL
zh5xpA02)O{E*;oCb6soWYi5L=nd`8NucIjY;qfhb0|z`7jAj<)5#g<5++w4GSqnc*
z1x?>NS33BX8Ex0#S!>|aqSo9-jddU;hy!?rRy^gSU4KRQ_wX|?bHvY3I3|SIz!S+6
zjXGk|{@r&)8lW%KZfD~@2da9}j=sI2$gP`2qqsu$qr}M=x~KMa?Y*LL#@MiHZ<%&2
z1bbS}D4C+YECeIp)wFj^igu&jJ$p^tC_I8*7}MDXy-~{^nU?Y(dKQ_oco;o$NtB|C
z-dew2Q5;@rXWR|j^_ptFmb=tNFE3`L@0Yim{{c^u+!&YfDHj!&Lf01(0y8w1u`V_z
zf7M(|k0hxPzW1-F<BY6kfQaBRZne~1Gi_;aUXOg)-IMLzi}&EQCENer0SP3K&&tYK
z+lT6^%7TD+5PT5<EiXS^<mGn{uU{TMya8UUaHOoheEIRB_y@^<mE?zpE><aFKwZB4
zeEIH4N^SV{$}f}>fM1qBJAS9<M^KJie<;r{_55g2@i$QX(!;xc<JUEPcF&LD_edx9
zwNHs64L$Tr`~`e&CHZBzB_dE5hubM0UA;X2@a3BanOKs-oA5Eb_1`{!dHCBy&md_j
zjDpLbetrl)ASeM&Ule(PQeS@g%jL(1-}hw8V3PX?3V-<1FF#$nC*LL+d&x4Ef5s3s
zGFFlWj~8RSfbi2GV`K>u%amA7$nqxU5?O9MQ-0QP`{B*0S<2{z!U<(E0~KUV5RZa9
z3g+*D`pfW)XnY>O=eGP^vi9k9m<3DdGX9C|H&2-TJR_~W@YeM+ViBZQ!&m8rTCgGY
zz2r5vF5GHXC7Z7yCx8`{xs{1qf1782^Q;vCQ*=FLEoWz|j1pejzBNqdF-S421%f2`
z@NRFHs&v9i(~r0$O{IhV_R}lFCo_}j{nKa+O9DqL5NvH;v4<VZCw{+HW~7GSq6y6j
z<(dRwEOW|hQk7NJJgF6wX97kbn^#s4B%{S-%-7*FQl8C%0*z2+LTz!re{au^Mg})V
z!L5BhuD*6Wg`G>gfnri|NAS}f1g+4DCKPGj!n6`-eBYWSf0uGK2~xgMu1?BKM%wsE
zzpP-({3+8@l&`IMBMQe%C=|5fO`-wJ^^+hBZynY0k9PRF)PcMf2BD1fddQfvuj;6E
z#x%CJ!VvVWQQkgo@_Icbe|6aA@-pl>q2m-SvX!ZxqWm;38wWz!sa}pPa^)F8x!rp4
zBs`v>7cEgBO?m#jwl|z(irEdUD{yR0!fHH&<peU$Ger$3G@NOQ8Pe=CMNLX*I>{0>
zV_HkQ)nZUcb&4fgEl5v=IFF*V<9z_sS*GZ`Al$k5v<G4R8I}lAe+!VXQ_Bzn0i{n;
zGbjejTJ7W+GzhYE2G3!;h7%VbfELnCm1mDrM2^ioY*^xL6gkq8XF3Q7JM&)u0AZ_C
zXF&)E+oX0o9Rye>oIArofU^u;cJmyz$dlFzhD}oLb&1G$-Iy~t#<1P0Gj}RvpdNBK
z>x-qd4m-4-fVLi=e<lD<E<hOU!tWzco-sP_^;AQ7G&!}A5)?MYY?$D@8bjj=oKLW>
zrMexj#TM85uuVJW(vFg`P22j&CYq5*H?F)NiS(TC3s%ucC+_P3B*O8RoQmZRpq8;s
z$IKl-%{alY3AYRQ*&1Q3oi)YlUC3pXp$T%i+s+DdInB-rf4bRiX9dCBZD<AEoMvbl
z<(y_`07d{z>`c8)@{EUYNk311&F4lk?wFPN#E*yNX2#^mMh-D%fX>hQ5i$aiw%JWE
z?ILE(UM}q(IdRcaAzSs7;~o?#H>vnB%Nd`i-(%|DWPYF3(+bN6VyKQ~xtNXCjs(9Z
zJ>F7%#SC?oe^`|X&#nVIVAAj5nIEb<I%lT0%N>igj;u1^cPyE~xY4^xaJ0Y1kuz8w
z#m&rMQd>$7$J&QWX4;r&RNbhuCf_^he89b@Ft?rlQtCb{vu&NXmts(NC@2-9NE=4=
zB{iD!_Gs72Y-P4TF{SZ+w%~>g+C2Ea;Kq*HDwXSPf0@EKo{Pgd{yamHBLXx!ko@ik
zdHFN{_zgD>+0cK+razZq<mIoIKZJtb24Tk%3WT-L7S{;N3nO`!*Fbh8LEwc%e>;ht
zGh_1sMeAf)rJj1au^_pyq*TCnz{Ii6VBcRhtZ_HKR95Ax`i#k~HVUBJN<XDhuj$v<
z{FHHcfBs!o)u`9e3754pinp=TqPW8n-q0L}&7jR)HgzkBoy&;HvY$S(jQaJu<75G%
z5ylrs14&KnN*}?=r)(h4%A|w_Z{v?({SE^Gr=G@589zVA69ds0Os0-l+D)B#o<T4f
zAI;^%8~Yc}?oJ!5xzT7D2tmWmyMMhOy37kef0Gg1yca;k4}+)8!_%|Qd(gN+KF6aH
zL$sU)LXGg$5+cK4Ee<WFtM}PJMEdh9xDv_6_R6t>Hmic2;q?FqwR}>1GBGlanr)RO
zGF8c4Wt6z8U<FQ)R}H0i|6WzFLf{Yq`oY7g81lBab!!qUI>QYdU2e2yOWEDU?Y{JR
ze`zQrcN$+d_X?_tbdr-yZKvaJ*HbgaT6HVb7F4$;>X6vut>+d|W2m)SGtu))b@Je?
z7DhG6JSkg04UA{&A5ISx1(<Ok8KR10*$EkfS6SdLJ1ff0D)yPTP_2b^BYahPRpvuc
z>Ab?@;L`F_%FZkvx*B*;@Tx~c$VLBKe-A_fheg&*@EOOGJixDV9*V^-yw+dgJXA+H
z#5V`eGv(m1qquRNDJLwuUJF4sxUfi|mctl=puT<1NeUg^rPE9)zUo^4H_wo<D@JGg
z8Eu^*8<Lh?r!yj^tR~B-OyY?lZAURdkTaq?FbqPGzFu?($_VM{%;+v8$c%F5e@Ax!
zAQ0{W;<r-RIWOK4kiEoLiR~a%Bxr=)KF3S{PqEo?32jJTdG;{+E|a*G#Lh*)WVM4|
zDK^tC{7BQ-Y*2iJI}vyk3_3s+lks8s-6ZHwRigA-IY+w*z{5EvgSMH|yFac+Sc-%)
zu4iMcCI*WuNJ=Mc<r3N(SV?wXe-Pxr4vbMpfy{Y>lZ9fT?~+jJNTuYFM~wF2P0=jt
zhW7)*tW&5Kmu!1fMHoHz=55UgZ^1PyUfbYw+4pjs84_+yXu2S4ABv2w8L)H-r?+QD
z@k|GFjA{u6Nb^p=ukp;-gm-itwzYREQ?;|9fDAsL+KK#Tb?#y3A$!x@e}fy*R;|r_
zY*Ni#ug_gOc09Muoso@Kt)h#KFN_kj%6m`K<$Jcm^?@kV;_OXA3c^tXj>3dli}n^q
zG3<IfP5k!#qcUb|EURMGsu^oT+Ll=DdLoDQ8R~El*s}D%a_^+pzkmG4$M=uk`*D@f
zym9K7xM@0nFEy=CR6#sRf8F%j(3WySF$Qb8d|b~xWDl;iAT#yg*AzB_u2Vq4nS9Q_
zKn^;ub?&V6!ThO&s%mUsvJg3Rj?`AmgE*7F8|ZI2TbP3ylUcgD7+9saW<ute(-?ZC
z85a<gi+f`rzoax|J-c99l?FEUt8f~}g$2_pO)?W`!j7U}yWq-Ke|Iy&?gXafH>z#e
z?aFYG1b*mlnJ}qPAi#dYbFm@Y)LhzwTZL+Kllx?D8{zLY7i`Q$FPO{Fs>T4=<x_L6
zvd-!t`3mMF8Y9eEMImnRf(@tgWup%W#d~$zDWrWE-8I%+CW9lAiUN_XkOFj1E#nim
z_yX>$^s=gY%!*i5f52uGLmn{-f5dZHB~@lHJV~pqV8Momlv7MSO_hh#O_17Kl5$R%
z3gp=~bSc+>x+HZNA!KAx;ZwScRUJUfJ}d?$YzEp>Wd_>I4()2^JM9cl^rzuzyuAqP
zswZM6;Cp_y42C#zA=~!~o_v@=x%PCeytEzi$Z6I|s{>cme;lGHR7|@(Fdxt}agWFn
zSq%;wdSg;Fk6YPFYV%SZ1kAQsY_;Yu+s;5qzm6I%SI`<g;fl6X7HMK-0h!qG9;|=1
z#hq04%wQ?o0VLOKnp{2Exm-P&I&6q-ru$RrUFX|l&n-!%emkmJmT@S<mzvwHSj&0g
zB=iEBY|CW&e~OkFw^(e&Hwa=K$Ep}oG!tDfyT$WO+Vm5RRT-4U2h=j1q(X2!ND8LW
zMaETBmlk5JX3!?p8Z4eTt)Lue(vmGKdfMrc-E6Pq8EuXsV^x;maO^l(Q0#AzprOAZ
z_;8(zq?$|#6Ejdl#^fa86l-u9MJZ)jy!l(R806jxe`QplgnD7h-leA}pGMI{^$bsM
z-AY~I>A-x;s2jPrG(x}fq};Lc#@TJGJOafVj=RMG2#m0#djnXqT(m-0(D0=aMUF6D
zj+eI#3eec+r`ac1P`^rbOKB3|zs7C<ma+@SA#;TbUk+5bhFh53eATC!`^3g)S!hZ}
zH|r;Be}#!u5$i_L*yaqW!W-c`_;#^WaoLu%@<IA;#G$elvFCxxR7l3KvGlUqs!R=M
z*87>vR|6HtTnMB4)a?cI!_~l!?nX7RYyCe^4eVyLuNv5OS`s-&#%OKrijG_W4hOuQ
z>ZvVV07hOsK$EfMaRs-O?l@FljG*I$?!M~Xe^Fezn4O(P+xqB*`E<>Dr(Rt6Ul(|W
zUeXoseLUxD+&#|+FO>|}y!Wx3|M$cFEaz)b{kRJ!*RzK{hvjt5X(!Lwli$yC@-!Wx
zd#*SQm!^6~7H&5rALg|n<xX;8EwP2WNZEsRBOiBQEuj<q*D4!F+*eC*!+W)j?{r@K
zX*802*Zmqvu4@;+*>iRL-sQM<v77zY-Rx$!)r~6e{&y0m8DMs9`xG*Gq;vF<u<JgB
zq<<xGze3WT%#P3<J~dsd>e%RyL*0K!%qc<Rshb^V#P$r@hnAKd@1dhz{X0j_{{dsv
zDO#8DDHjx%F*86E0ysIBv5P4um)|!RI0r>hA1RWe<d+^e7gm3G*FGkWI=%tV;feH_
zisNOFQYopaAEXo5yVyVd^8V9~ry&Xa6xHYJ`|j6WKPaOOlUxm7KX&v$Nx`Ip0VKoW
z4!`|<__q68IAqS8Ir@fw-=<hv!yH&*`SUM)_!~a_35lh5!+!$lN9l$SKZZYJ2TP0{
z<AJe61^9D}`2c^+JD=E&EVT5DD}OnQm691?#+~OXIP<z^8_doLG&^pL){WVrxq+6*
zTQiT$quDti!|+cJhLh27G(d&sq0sQONS|m93XMw{q5`B+XoxyfpxJ60DKt=9rqa);
zm7eZ%t5D@RPyEC4__o6D{1DBqLZ~sdg8f#I7O>Zz883gXvWiPttegnQR(aSm5&9Xo
z!aW(V%oIC82NAUu^bJY(q^%=54(_tP9o><`$<`AtTUsr>o$u&<<%Ha4e&KENINHR>
z5BAAgOuT|cs%USCH`$L5w)5#|VDM<>U{pOe!08Gcnz8-iFMo`$Ov3?=X1M9$=#4vx
zt)GCVQ4xPsR-oB>a0tRVLD3a6?OZG9f6oev-kpJiUQh}c3ix9TC;?Lv+hk$JyPy$O
z&?J!$ahBFIwe|*yk|#}6PxU@q1_c?VlfSS{uZ`xyb6w<cPa$j(pTL_ZFpuFG;c#NF
zLSu}PMM2r7jx>%OLL0jh#Riitk>@M#(fu*M7nBuvOepZtKZY@zLZxaG$UOwQ#8uEZ
zmxMYOCVyXHn*QLo-@XrB%a`OPJf99H=%%z9w8!~jY=(UyG}QUTY@kw}3Dt8Bfyz?0
z;$xVCl$|g|^5#Eb7){HUw-GgqEyDCNi<eOW1svlu9-t%Frsd1qJTlfXk0f0$lxc;D
zV5+I2MwH~_G+KoiNXL}3eUCH&Q>eWhE6|iKoPR(i^3|n==_R8W$2r&wa%g8PdMJm^
zOFLke7f*#=Dv2Cs#o==GX2WUu0-uysX!Mhq4|UW!#kb%$e46?B9&(gRDuT)@_|Z#{
zd2WF`tFe`63G<|=_N6(<$unSvlA%FC<?H1@OHA_76gzI1+D!tjaPCOmQ=_UkD|lVM
zT7QPzsH0Z=@_vnAp4$`jc;_=Vc%p73`U;N8>vSl;Ok?9LQ}P5!3iw;>!Q+xb$(4G+
zNyCJ<GkQ6(Si_mUnZY*9I<!yhwIRO7Tvzp)t+gS(##me6ZeXkDhOx<1TYwiVwIRON
zP}gXr*y*%Jmb0hX1R<L?+J_qpKp~>!EPp#im@Z{n)9XoshEauj%kt;s%yyioa>P|7
zALR^F`lclx%?<PR4kaJWQ7U&;$>(Ag>m<X4JsQJ(N<PXPCd5Q0{r_0@*$mVI{thRj
zx~lARF^hGRf!G>#ld{hVYXEd{Ve0-mAl2ItNOys|M)skZ2|!mxD<Cu?Lq0;{KY#K*
zyHEf6{_Rg+|0)>1_HQD^`Hv<iP_~bB3rtE>GD;Tj6*Ja3Gf<S0Mh(g^%w3TnzgQZ+
z4m*XeQb?+Vl-3{A>M>D9;j5Pe0+tAK2USy+fjEw>slOj}W09(K5eJdQpgZD{-lbc7
zKfgt{bk+Bn6yN&Cu=_UbJ||SQ0e?#*jz`Ac;J+zU1%_W*&R59hCNK;amO^Dzyv2Dt
z(^`*rBxS}KjED*xE*}JbE4`I5=ay>Rxe0k?4!Nm_*0s;z(2W&a#*Bz*F9s<YK(Kz1
z(2WR>-@ksO_u62U`8)^=MM|6DL1`T_8RQkbrgoDB&_tf8a|U%tp3j%+U4K8l&|l3l
z!7e`!l}K@H598oIuLcC2y4HFHApNYbfWUq!_0B-dRD$yu5_PTw%T$}wPXM{n=2Ni9
zA%TE;E;Y7Xr%koAv9?!gdWRTNV&KT{ovrysUJD^9Q61_kU3eyY+LMakAykvYU8+!*
z;!RUKtU_f#K~<c6mj8T(Fn{*kmXAZ5&9%#dV(KACcTC^)ScT|n1XBh%s(#8KN8^eH
z`Ol#xOM&B|iX1EU5)?<L%u8)nK0j$UT_b%iNbRc;ySffrXkAKL_JvE{U5K<(y9=WQ
z<a%?Y)fJ{qHD}gpc=~+lblzBAc%!QG*194NTkF0O(5*VNrS5w+Uw@2C24sa+P|A0q
z_zt7FtK;Scc*l9JG>0)9lcbJVo`{hO!H0x%(a?r<p}=V6`W|O0Zg~~i%XiK~l9wn+
zuCXYDpy-&_aZSs_I!>Z;wI8dRvhp{&+QZPzF3Bmhu_n=R3l+;Le8+n1#>z`Zw(YMg
z4Q*M&_Z(j~bIc3<o`2(9SwNv96+QKaN(~60so5)5Tve?B(9FV#f`yzOOh^Pz)N_rQ
z8XitW)Mg6fPH5ish^{&~$FpYIa$8}hxi3gNMVI?3;FJoRzVA76NJ{JkEn6>1h<KMy
zTT}wK+gew^j({20qaK{WOT7nOCnjg>kS)Vy;3Xe3zk1_olYiF$VHpVB2{BhsrdViv
zN4#Y|=Bd*%5?9Q{G`reiY_FW=9imP-uhY2T;@FdEHpE*sg6kN~)%l@}#GYAw3yq}x
z#M^1aIEL2^HDV33cCAL7W_a})>0Q=-vz8>L@tD=En>SK;hI!<PmTnOBrQ#S)<T`7i
zvo22ADLD?O8-JPE-{DNKAbJeva<9N{&WFwP0G3$;ll#JMV)Y1###Lu#Qf@}O6>aZN
z;_*a)D@yR#KgI><mjl9k#&IZ{YVd%0mNR##!Gi}3XIItWFJ`e)GKcFYxK9n9=030X
z=U+;KFmd337N&6mib6NL%vq?HZywYFR+i)R5I5sn#DBdp-y$!@B**WiTq7CxGrF}^
z#v6-4>-4r+-~}L*Lw9@gohBcgC1}7bwYX)es{&r~c2qu#wq1^rmcw*(R8gqL>~rv1
zglqG#7T}WT4xPReTuLx8?#;@oo~WyQdaAyZ;`6kDW$Z@haZE@5O+1JeelS37hm?1i
z^!yle5P$mN0jzZ}n-N+-9x#6eb0_2_#}$UGYfDX~S>+AUVyr&th4ANPeN&w$O)Zn|
zysGpqp}Bd~3up>D_HXZmCVm1Y8rcEr>9Ce9O{e9kGW8kp@ssP7SZyBD0#-c6>YK;5
z8}rTb3a<kHyTmr(EtAro1~)KyH>Ns8UHPJM?oyo3oio3XA)gHx+)zANe^g*QE(>_O
z5bp0Xm4q`)+VN)>vf5ta<0jR4_b-j;{D1c8D$YPkCPAk-kY}b$F(CbrOs@3*o6%SR
z*~V1jJ}6t-m+nCq90`q)g-T>b+*g~I6G9hUf0n2D_fBLVuk*{YYE`mw@w$CH3pk0G
zjNE5F0cQqO{foyogDrEat}?nNf2n>YiCM~naJ&r_@i*LQl02vTef-UJIKaWvvz5!<
z?SL!*&0`vJAE3Es_^dcCU6o4nR(4YbE9XhO9mKi-JGJbka3EGLCbo}w2}#N1bw`db
zH)gS}!b@z0|DAqLY>|^OO{HUhawey*;Fl9w_9T+U|KwmQnkDLqXjY#FJ3TQlo|*a|
z;0G=(mvPM$6ah4sF}*<}moSkwG=J4vZEqVl68`RA!H?+|E{E?diU4VH=>@pM1#u2&
z>l%hEtrOL@R=2ylBv-V*{hr}Uwl=aHX&pNWU`xYYayXoqhr^+au|_CijL|~pZzCva
zCW0{DFcF1G9)#7xas!mGkvm&2f=f(95p5=th%p?1a3X1b;z*E6#1vRl%zu;;q{)Kv
zT2P+K_#ntyp3pf^DUbDnd`IDJ^xV%y2x=3iz|2|`3*^1fCinAJXczJ`C$y*h0))$t
zNa&zgaF1uv@uC9OojfntD^P(22#`^wh$%dVkztWT)_}tYufV$e3@fmCEYIjo9t$Zv
zGcc^7K=~Pl1;<o0tjqk?34f~`VU7s3s>sTSQCOX=kQ7u+OvM1ovh=WuwYgt{{$0ov
z3F{TB1s1Zw<gwrqJ1z;7<8>s=mU!h>Y2lRdOj+SbXNo|v9W7WngD#*FsFW!$oMq0I
zFxK%pENthz%@oG^;157W;UX~?jebcxe(ONlOnKpnwP7Ke*HD<ocz<t*1<(YFO=Kwu
zZ?mA<3h#VohoRoV;#OmX&qC=ad?aQJ$$U~wX^c%-D6B6i&zvoZK!HcVib1nVEnGch
zi&<DKIQCnpSOgy|=75KX0L^I|M2IF2O`x)|jT0edkugtHj^AQ=k!+?kC^ox=gGZxS
z6lW0`il;!ZL~zFx;C~#w%3KgPA!LWb*a#zPcvj4jW1;Do@=J)AR8}5*HfbyrR20cj
zrVwfzH1I_wl5_sxD95uBueyK+aeVN@55?;b^Rjp{n^mnSeyr+ASsy99RL8~Nix<UH
zG!U!D#jA4M9+5VZz~1R7;i!TfNgLq`CP<Tx4~oOZY0LcnF@K$%7f-IZkK1WAD-Mf)
zzk0#{kKVTJy!pN;-o1M#FUO4>%`W7d>c7PwfBkt`)nzfR=1u-Eay~ix?(yRT{Ou{L
zO(Y|{Bp8WUhR8}1JU(g(EV_SDpRu|fYq|LOmtWy;>}X=PxVSjJu>+|fBjCwvk^$!V
zbkF=9)_IEJX@7N5)ra%ZxD?sKqIfxK+p?aC&ho|I-nTy;wxhO$rt*X5)vU!F&*AID
zfWxyq2hxyJv1<U@Naqa!4LBGP95FW-r;}F8h5-btWg`HY6Ro-de%Uj-hG%@HWjBUx
z0DbqyF}LL#j&UH=vcXtpSFH1nvg(!@cy2s?T)eES@qc029u+U2JuhCD@7v=ww|O~w
zQ$jzpww$#Mv*@OMRW{Y49+ypr<oxKL<zzbgv3j3<%LCs+(#IHIV<dW!>84wm#`-nW
zc$miDWa?h8&VWv9=>W|4V8HD*>zTG^z%^s+t4pUhbm`O(T{_wxG54dKHGS;6YHv1H
z?P>Q_cYi?EbXiWPg`=G3bP(<QiU=zru86cGydMVRDM&&H!Q8;az$p=+67fq$Dm4Hc
z%b9#PJ)h2b*|MtN6x=Fa*Q0s)ziK)eU7Xa5i}K{RN0aH<S;<<M;#v8h#fTX;zkN3h
z;<Z(f8+jq19eUGINfmRRC_{|<FXPR$eY-f7<A3V1IIGJ^Jv}eaYaR_BgHy@hF9ymP
zjpPJjB`C5Y8T0@_+eKT|P+e2Zs@cVKR*veT;dnS{-Y(h}Dy~Z?xBhU_)}!Asbya^D
zEHyGk$w*DgVh2TX32Z|^B45J}N}tO~ayXwHcH$4*>5+wKYmoy<K7}}FnVY0YChZ3Z
zI)7iBHnN$HF0+`EYFeCD)p<d%9+^;pZHOu+Y`)QjFihxftH8lX$$UJ3cD3a!56mA>
zWCmTYciF&ty}AudN5&1eLr+2k5Kb)*xYYs8fBx4g2LXn>xTJhwwA*kVJ%H{{Z-e`w
zHsL-Ff%~AZ;ohqo5WHvon_Y>zMi;2aBY%|RGm^~_io+;i1*alsC5~su#S;Pchs96T
z>#BGr9yMilBk84uBdpmiJ;3JMOGf%>;(9VdoD5RDDKorI`94lMd)ef5p1JS79yT+>
z;D#7F(zT(39kf-O=p-Klv0cl2bT>G0A<IC{BGaBHuAMjPK0Jh-^4qmuyxl?fTz`6i
zABJ0cUI7U_Sl@@Y^%LB6Z@cB_vF=mb0VNkxg1G~rEQMV6v#59Rv*`O-b2JYSxI3K*
z_YJj}HdPP?sUUvBN$>sia3?*a&7(x3A$myq8#+LxO`uI~(5iQWsI&<@#ye>qATRk$
zeqz;zpKzUae~od)_d)&SWANkxW`BpV!@$!nmD$sTAEb%!P{^?jLHBh6+H?XM<OKAE
zogiN%xH~ujrA<xjv3*=+YJ54d%GCIJB3D=Vc4A1%wya8kDKfQqRh2MKOB_kRzN*AZ
z&{b7pZ$WRWO0?5?b26GuPTG1pK5u$LX4i{5Aa=jk60u{DTyOkzPmSY{^?%{s4>7mw
zhCOf37h`vUJnc5TU7I0sukE*!Bv`!}<2npl^%qYO_g%W#$d9bqB|oyuyOJw-v)2bv
zzVhtgZFL!FyUW1tzVl9i=cMCb3|s-8ljPmk2zZ^K*MPUDpf>^DqABZ>NqL6W<m7GH
zG}F<n*&^z`^&LJ>+aQb%(tmLcAUJq0-}DDbX={0hl5|(~oxzi1V(EdRIt3Y!W*Coe
zzMTWg8g*N7DXtfntns(sg|2tF#`>5GS)cC71q9!#M1I|QyxlnZ^X~+9j_hk)So5uA
zugfdG-qO<$P+Jvq8Px3Zvdr4BBH~@JjDI|ku*embIYrWw{ra$MC4cA&%k~!ZCM=tk
zO$!v8*VXAodHK0e3`cFBf8b=L0|dZ+GHFgbEztKiAPGGnSzat}2a>f8S2yj(OCq~N
zPYOml7rx+~e82L&UW?pzTmNw#{koXBcWiDZ9L^1aCFj31EZusau=Up9b1Bkqf}BhH
zDDwY?f!Q?*Wo~41beH!j7Y+e7m$6MM69O_gm*F-9CV$OZOOqqF4ZhE>=%aI3E<qA}
zR86T$((bs*Hz!lMxO_@tU$TezV*h-DB*-S&Ew$7dB{{VE#RkETAn*Z{oPM3;^!>}}
z{pItUHj@(`!Kvx}mkH>Fq+gKq(Bk9(gw-&;f17@Kl~P$cPIMS4wWh<-wWl*4K56jO
z0s<Xy_<!UK(47W4R6eUWI?jCU51-=uBqx3AURfgMFVr!;(V?`GbXaO>jK<gxwFBxM
z&WE4he|V8RBq@T8m+>^-eSLrV&&xPKYqgNhPyhJ#5?>e%LVGoV6+zqSw|`B)y!<#y
zZQ=7k@JVYCJNQ5nriP>kq2sFzAJV`KDQ>>dL4U9rf)i;FhG4<K&lF$NpbbHWK0lWP
zJAImE#Bcl|A{d)U#$Z)Tpffb*brc{7W88>>(t-@2pim^tSo_qSr|T7wI*5=Y{`9k)
z{z>nCptd^r>G#y~TkS<o|DOI9QQna23Fj!+3P)dE5iBPqf)6Zq7J;OaG~s>`H)^I(
zcz>m@izUMdILf_hNgxzz5*`Fy7fYfwl+B;41Ws`o@Y%-Y$L~(O9OGH@wd3m}AD&HT
z9$04}6|&u9+rmr<PcG<@*}PZ8jrL1ot;RhDJT8<Q{455Q91c%!T>J=ep%>qn+{7s@
zd-k-^5uk1yh%8Q{w(#6SXBUEau$ql0`hSh|thk-G(o=RW**PV-<sAEiDw|ytVt$)?
z#zwA9hV}W)s)-zdRo1lQRy1Bs_SBC1yFdPtKUs&01|bmiDpYt-$r)0|@MX;zZU+Li
zDA0u`1p;(H=Q$8896T#+=(GCs+pp8`<gee)dTS*>*-<jZ-Zmc2tt|})#?~pQoPRzd
zg3#0psTS}V6y&+~MG!Pfm?uP#=s4745#(+WH)^JQS`|-;AV^@@CW0V40^clJzK^a4
z$@N&7NL=XIN^%6$6u%NR3v9%SSE8HajFru5+=wsxN4|k5IH9fiXhg|XNJmObVNq2?
zV=VR@V}WDxv5CXT@!9h8a_^ewr+*vq+C8ofyrd;RW1obV48jK8;Z?W^@L4l%f!`Nw
zbi;9ue0{zVtKDPT!b%GR?jb*NuZSBxCy7<y|4ZD(I8A}smuc1R05&9oLS(_oSN0sy
zYsj4TB~s#<x<o?bV5B%sZBd-mqWB&>-bis$Pn#4c>r*LC)_R@hP)AX&kAF#N)}nq7
z+zThWiOkiJW^@#zQ^)}6m^n~dNlOkCXv%382gs4}O=vWE<Px3v`GCwd=4G4_;Fi4}
z8MQQTyJj)=x-cgT0&VpY|63BfpHCV}p4Cc9O>YA}%qK}xVSEieaZ;lhTA2Vp?`D&O
zTqW1pB(;jU(a`#FGh{}ZrhlcDH+B+ar9CO<MkOy0WJP4t85mnH2=X2MV~l!c*?&3K
z^{`&hvC!k(ua#m3DhO2bT6vhpYjhN%{YTPxT2lga_;?<#1LbB2kHp`*Mck;FM-hU0
z4v*KKat}&W|1Wv`{*iAW>QG3npTkSFi^)d?^w`kRLeo7{PCL-kmVcri3!T3as@+4{
zfXW&n-IG!tZDZasW96Gvr<jcA^SLQaVpE<xaogjKIPD(K7EU0f$A{d<y&`V&nA*}C
zZ$FH_qo2IM;3BD7(s!*WJ%TypGeb+93)xEqW>1mP9^)pAe{C5VokQbKWK+PjHFj^M
z5jxHb)Vf~W1j5Ys3V)Am>4k6jWM%ED)B|<VKi36|S4-N=O!YYHY`%_MKJviaOO<@?
z?oQm7oqMjDt|CKe(d#NYMW0b)>^(a9;7_PaowJapg!sAk+;Z%hf!Dl5cX>(4Nz`gi
zk~6loY)!56{Iy-Hb&jtn?n+y0QRc{48D{S1Hh%G}Nc&omq<=f}nztegPcIBVV_RAz
z9K+R8G=xA*tJmqfzh5myD;fsEIEqHgf_KuPT1sf2R=o2a&ATcUY`I3I+wUswQt!!u
z(^somSv-H489=wXa_?I<N-a&wt+eB;_f#^ls&BnB1uMfNE?HY(3wkT4I$n`Ar!#Dm
zvwvCGn5V2&+J7!vO_=#6cjcVBJk8cJZ@Bm}cE8Ni%QxM8_)TlBt=cV~^yIoV*Kgc<
z&w)K;Y9qU`Y985nSWzHX^~ko@-!+!l#(@h6T(^?;yE+ecZFXv9;Zi^Scvr2!QDR&7
zJ1PXaaqfA#n=Pc`zRNO5#+qAoA*0)8gT}GiVItcS!+%?H*O2}m!8zbUXUZHw&jq#P
z0n_aJTHUaXWq`k^Si3idYF2b*Tk+fu)0CJp56=!W&+_Y)Z6oYsJ8|H(2*!>%D3KyZ
zZurs4iPA(TaV}!XCb)IwpOQ8e$AnA>a{5$iEow3FeoAA#hqP8o3mIzChNs;Xnn4nL
znSrk4(tif}Tt*oRv()YhN&Wm5<SI&B*%^9K>?E-Nx6MVY%cy~LHg)lZ{e$iVJ!{Ne
zCzmtg`yh}Yz{8?Kvjc-((z8J5ds(j;+=hU)0_+ljuZ+M0x;nxL7Kl7JV=C6Ef*Yk;
zE`esnP-YdAzTJ~+`m*aGWrR|5(RTL6SN_6QwST^zJpnplw7FQD&whRO*P)39c7^SM
zZ#ioBcM{_ohUnY2{U218F`>RWCvQ+<&|;sY<R<%Lh&4hhGlmeaqK#dRAh*nOiZUBh
z9sh5EjqAG#_f5;m1oBQx$?DR?aE<+(dr$4U#BV;wB6pswTyyV3*<CenV?`i52PId0
z)qhweAHGm4d?Igfz`pRQd=Sfd*(k<qsRZRL<5zVv-~pGw;Q<s19Xftm!sMnsWQRRU
zSmz^m1sn?^NH<Ep^v@!8^F!-B496K^9gLEsiThBO_o4y_>-~lPXS9s)ftIHsFKIs%
zEFOL@`4U(hxkf!gyM6ZoS}95L5pL8<Nq;Y#Txhj}t0-_nQMxrqx-fN$rp%F-9{c@}
zZN$S)N1>$^h;9kpbx#3-0?q>PaodiqF#hYqxo22mFL}!f`_{XbtQUdE+92uDQuUdm
zT<Z66cN$dt2Zf6SYrCUSum>`VIep_wRN_aa1`GQ;VbVa;Vt0|l@l2HHPp29O2AA<E
z7Zd?Emr<EH6qgcfJtvoaRTrs$D+50N`tj3~Oj`o_gvIB#kEj1W@l9D{1xh{q^T$(s
zpb-RmH7FFc0ER$$zun=Fe-FPs{k=fQkfrePKl=Oq9!O_}S1yA5<ySfUiyr<&gE9Ew
zf8)>}Wf?jAXZQy#um{)+IRZvvK-dB-2PJ|J3G58QU=UJ=$3b*rW|gph8!AW<e+&mZ
zik$}%J87*kmDrKEDT4&th(N%T*l7|7L(DcL1)4-olR%jG>8EF!#7c)W4$^&}#*UxQ
zU#yG*CrNnq%XBx7G$ui?c~pdD@(7#Oq7i0<A4e$~WnR+V`1Cd31zn!Y1b=)!&5jt!
zw@G9d)UCv{@a-spclc(4umMg<f7GKOI+cjP>Y$D&5!coZ(`J<wF6b;{n7EWBL*Aah
zD2JKPWU!Qt-zJ#?d@ZsTpG1B~Nr{p<e!k9<H_bC)mNsLT@)W;k7^94YGAGNuE6=km
znP)Xhs;@BEl-p{ZmxSY6#<$tUEncIfCT(pfi53Ma&nda7t*%>Rr<L#Xe^47Kv~$hM
zIzgN1wV+ePVZ_TkkFF|eavBxpJ)4BdOskQq3C>7KIpWsH6KLVd%qCn+k*Dn5qJ4p{
zt&EVnoa{lcqjJ#4clPbB8^aKU2iLeUi9(^zevE-FiYRH0F!2xl_s8!8yYq9>p&;+u
zY0_b%2NM*jrz!N`wwp^wf4%AfbP%9h$ai%s$x=vmy&03}s^gyWpK+_zou9uK(kil%
zPFlV?yvK_L8fgYNp}_DN>JGlqqD=n9ygrVZiJq}R4Dxj!IjaRlNPJduK!xOr<cajl
zcn})xbxt?4<@uh}pQ8qPp}pfM?~x<Xu%1-h@xDf^szauro6J%+e*uOYo_;kjPFM<r
z7{m)i9Gn*+QlW;tLNf3uz=Z~naGqw_gpkuWP|Rit%v^;}&q@tsBpOQi4CBNguNx<?
z@LB@}DjITzz?OYdB<wcu8Qb<rzZ|enhIpYnFq0HIX6}(3j!>&R8$weI8rx0WNs6mR
zWi#{c;&)83SN=Y-f82Dj94RX_&hQl{TB&2!3z70q^3a@C+tv%C@JZERvSQ3BGO}9z
zrHpxva;Gt`C#_Kf45d<7s1b_`-yS!ek#8g>3>sy8cTJ}zxkl5hEgxSuo1Gexl1#52
zfSBD2+le_2(OLnvp|rejmB49XT|-IpdW<jGN%ir?J97W}e+s??BZ9`xJu9h&HqD-|
za0<!@r<xtDDUJ#vIN#9z?GWvY@^;Z)X(gP73uvF!pDim_Y()^xgwG&7#gY=Cm-OBU
zx~?L9z%lvlBz^E>?dCk`QBpP`8?o1L*m$6{!kAWw+<gVkMx7KYiv2}Ky@tk2VPn+U
z+9!;94e1J_f4;esLq@%ZxM$Ss(Pxc%A78RjKhKwJ)Vuavj5=HQLZi<1zhl%H?N1u@
zF4`BPewy~#sIz5{t3)yUm#KuZUeMJm9?lBq+-HbKHs@X9QB3wF;(;;22GuDZNJgE~
zstd#eAqZ5hpq+3$IlE&!dA#_rR~pjMSzGf!Y3NAFe?xmEytJQCOd(Mf-uq7|rf|qR
z#T2J6E~rfPTZYAEMP)L~Y{qs)MIEX5PPS_wvvU=d$)54u>{V3cTtmp`r#bcCt*DSj
zAYbrQ1pz54GogXGS5Pr1(AWl(`y~WT;u`R;R8Off?klI%bJj>f3j{Y;-(F7z-VEAK
zHHD3qf49SdcjvlMNP_31j|7$$M3Anb-id%e7UjV<J2?^Xr^5kG<@e2?o73jHb2z@U
z71^DN;%xBI*f$l-`^2L-gS+C<D2}3_4b)YNQ$yRySlQ?^iqp4lR-AhD8O34LyFhU^
zcXCK^7?pPvhtcLj#bG<!R-Ah9S;grRkD@p$f8udL#7&VM`RC7$xTO_x(SfoSao=mg
z=xSC(qlDFXWwRnW6hIl-<;{v{rz*sW*0P5|bYe!qw1n_E-HK>qc5NTaVEQ`}RaZwM
z{0SY2$Emo-9a17ZB{mO2=u{MAC{=+!sUa~IBLIuRiha@}dtY*Ib!tT1YK#MlLmVK7
ze<feGXHksc9q?Im^IUFWi+5CRJJgf#n78tDJX$Z<T4r!DKUucv&0|MlyyZs8@}z2`
z8}FpFYfQbScD@YvmhR-^{#%)&(wwX|GJU=;Qn!UVrCC@v{>6m2tRd$#{L#iepj<2H
zrg0Bq`>RyjjeDEsXmUw}wXl9}M3l3le<!LOjW`)(C08O&EocxMj13x-?TbGN7ztfi
zbf9Dpw36p5uC+9!0zb!ZOGJ`QE_Tv7LW0x`N}N#Lxf3TSAB3Ba?s0B9Eyi9iC2WPV
zQv_ZEi&QlCsJkd497{A3UF+poCUoR%g#mm&UX3RnLU0_Q**vSM2Ol}6n8Iepe^NM=
zlH_=p?F2VZdzc)sYE70rN=@9Q*?Bex4Dz-);Dhg#OLML3csVm?c6_yve2bdx{rF2I
z*HyACVpc=NZ&7>}xpaT}kju^-B24(<_9-8mRn|XI22M4@i|Lk%%WHHyVNR_ZkD!+J
zR4rqIMR~ey_G|2E4K2wbPbd||f0%LYRliy|(ah8oN6!7%1yj}0*nYrU=T%XQuxO56
zp#drZ%JR``JD~_M%zeP4>v=tHKx$|^CF;U0hQr(A>!(f$zP^lRe5JP`*_C?6v7aTJ
zt?#Gv_Q$A~8+04$t@9d?vOT}#bxc%G;a+%H!`|b3ssM6~4sbs6k*&YMfBDph!e$)r
z`K}wSPqq>(r-V`JL3-vsw1%GRvLKb9h;=Cov3h-tEYz@jWr1OILKfKi9kNi*ICo@S
z9}}OJhpFM@F`C9W(&(0?tvDq28#4oFNdLGS2TzYRa5p}tLGN|X{+Nq}_0%XGH!z6%
zxe?|(4{FjMtAVGmj$yHSe|R)i%-QH88A-|>I3zfck4&K^-^q#2sT5*z4t-l`DzhR`
zlEp?J<x6sY+q>bN7Z>m){a)olzNC=)LcS!E(aD$OIyhe*329#s^ATe%sa>aj<zLJ+
ze7vAGg5*5V^-2`R)>!55V)C|{xq~JAz2Y=pt5A*L*6iKN3WlUff6WEoWf|7DzaJmF
zS!{YH69%b-6;BQ8)t}1`)NczXarhvNR0&n{(hj-{q^gI`!GNr$qO|Ptb1+bpQ%QH`
zK;vN$otObRZv9Db?Eui#4{P7EH^rvBKb7q#4TLgowHCZ{E_HlYJJgDGAAj9>SHbsU
zxEWsx;}m`lZv*8Se~V0^X6>Q@&P=aP=wjPZbhnA%$fP;AnhF#VST%SMW)6@sw-h9k
zIEr_9)r_`_7kI$*L@Ush#Ana7N2<e{+Izc@K)}zT6*)*-y<8<XT`U6bduMOSE#H@`
zOt#r%+|v7^?9{Y6zGW)jdz;)<l+W?`ZTcB-d|Q?3G%JUx3e&hD_W6Hf^CEbc@hKM+
zm%%(U6#+Px;Wh*(muz4cssY)z7hx9@0e{kwb;A2`rF`6s%L3j?38mGYbV*rZAy&(z
z8I@yFt8AH09kLDP30<=34hT{h7b|ABk_^AgZbMmiAw@McI|_Cw+OLH9W;hmLCJFTV
zPB4QJwpuL3*sVj?1aN;Hx;LY+fUZO$VD5yj)B>TYLQKj}Lj1@^-Ze%bcBUehMStK0
z!8eF%=-@{5BzTI7vONH_9B|}2`xN<v&zd5dl`!87$90%($brUPV75^Shhf)sDP9G>
zx&S^SK4PyKcYS_!xa)Pe-;U1$?mmYkAML|Gnr@x1^TKF<A0Lj(ExfYlK3wnF3zK|2
z)mWGPb+UI*ygA3eZsoe(%+CT&HeQD#_t7Uy_v#hSW7w)wzVXPRi(uTSWf-SyF{A2z
zMM|?E8!FiI@0c$dn2ehq)38OLZw6!mXp`fR`v7f%2+~)>W!cHvG1$ohS4HyJi(w5i
zYZ#eb^O?CKlvkz`mxp5)MSpV7BI4`&CPU|zLS1&PE4(E4JoA!Gc8{Kc0LR<&<kxK7
z1jh>P%vs&UiKtyEn@p&#(K1|eYmmCb@Sq^R%wY~x=lDyOO)`y~+a;|$HObiABv%!W
zj8?)MH3dOw;Q`)SJ7SVn3+3!BqJ3SL5kMime~&W1nZ(MOQ3(#heSZQ7gQmjB>KJPP
zJ`;Q=-{$#u<h$l)&(C|lRtD#{^R|SU6Yws}{8kDp#aCN20DqBTHc};P(HUk^^J<*Q
zpz3_FS9Y4EfGj|2D+c6LM%98IncHJ7PAAjb6n>a~9KC<?{DuuxqMan<(+Puyh_*$M
zJsWwz)KEAK*jgU`_J6;>=OOcKE921QPPx9-fNQVut-g~1*XSOUuC<_<{47t};IR7V
z_iw|b<;Ul7-e_w<xhZ9UNzK3XMyzm(litZ296rMIYiWd0XYgSl=2X1R^lMz-;LOLH
zer>dH-o3|KyqUzxnUMp9JLQfVWdN7EN{VI04~_Dd-4kf5wSPMHaYo5A?Nct8B`H>8
zC|?`|B~EL1OK#_U0Z|3Q+Yfi9m|tMhqm{+UkIvNazFVni9oab+2Wr`|wh5!-{-yn}
z60e*2TEL48)}7D-BV_O$T8EiB)RECL4Klh3P4@h}a_`_)juvntBL(m-8}e2XE6pa}
z9ei9eh(Z?8w0|oK8qZ@6<${R5Zi446?zA{tGA_Kkl=oUW_-OZyC_LHI&Nby@>M+^6
zJ&r*PSF4snAZ5IO8Vl%>G9Gy|r6!VP618M@9E+_@B5T@lH6?|a>P0AVRw!|n>sZX9
ziZ67a#jq|oyt^n>cJs4u8h5V@ps=_JgS0Jiv7i$2Qh%f@@iZG1D36ke;?)x^A;f!T
z8;ppcrn4=8?-`cHZ!|&bY@T1Mn0BX1Wi$8i4`&NySXPofI+GP0kZTOer(AbXT@o}S
zt?(YY!JrSS3v#D88}P!bFpu+@2&*WR3}-S-%mz}ZszjEbRVh~19yFbZzkF-~En1NK
zENMaNHGeZk)pvdj8pg^U_p<%=F^@PD0k}Z%lBo!7{~nFYM&k^POB7DyF_AiNV0~Yp
zhN4#5VMM{48r=^iCz@Yv6b)tT{#xRiMLd0}os+CGgN|-WD0<RHviiRbD1gsHlEO_g
z{1z=gH3K;f%(6VGt^tUo&6`iZe*5~5FaH981AqRY5g5MXoB$ohYyL^!K!wL~-%CS8
z2jV4uUPMWCuM}iPNf{iE9b3K(8>3M2wGD0acWOr0641G{<CAmp)U|%4Z7WkDs<pX%
z^n2S_<~DvM@m8lozlaKtrJp#yjG~{3*XDQWilhEJY`zYg&lT6G@gJE&kB)K%RuE_f
zG=H?vmZm^Qu9BBefJBc=KHu1~SsRr=BWyG>76gKYPxV_P>fF+>Tbhtop-`I2ecFD3
z{RyWTkSv@S`9&bvj4GAmfkDl{j?Z6Sd0d}r7CKu{S{M}XEkr9+NcF%Hoy>&;+`aTJ
zwsgNs2juDw1)TZlX$aJ6+Jq=h;q%Z;F@N*A9hv>S9v~b7EAW{*mIfEp3zch;*D6JY
z1+P$xuEa^nsn%Lg=Vr%JzR5nIR(+s32<F#ETU~MAv?c$tHq-E?CZev}JSc(CVSdXX
zad2XuI1IU5`3mc7`n8$Hakk?Os{csiv~2w3A)l{EKEUVmCD&=5#}H+Gtr0Mu+JB&3
zIeVnTM7FKxm36Dfvv4xRyQg`o$NVN8F4c8x*rgLQR5Y@BSE01cf~1ZO^w(lBrD-qe
zY{KcI>4?qMU<^kJE$wu43I)oZt}Q#il?Btzvx;>6RTg^e=gO&3#Y;ZrkXzorm^B0~
z-_T8urwsjgqBWl;w)xxpoedA|Y=0<>P!CPzh65{*^CDvuI-FX|Xe&WYY+lZ-Jx%hr
zUfue1OIGA^<UT56Crl0UC3W(nDeB71$1Wdeb{3mmt<qXVzcX3JU!DIrI*#WCO14r>
zO*eLG9!Lh?T#7DB0Xl84mRhQmm=M6KQZ)Jsau_yw4Xz+}EkANrALVSRMq)W(j7_`y
zc0Af%4oK8&BdL3r>Rc@Tpo%KxrWD_&D)lUZY0nBwjK&3kuK4xQh0gl*;-j-$|7GZV
zaTvZG<M1{Ag&?M)c`?)beh;^GwN>_4yQIZb;;~!62AKZ?_O#Qvm+>hV6ahGwaqt!t
zm(foGCzs$ZLIi*0xDCGduc%`Wg4T$ls0Y&9Ku<q5$ZZ#W2$0Pw$P5Bx0%U*y`}0G|
zk}S!Oa@j@>o|-C46iI#jNYQS2`F4?)pYOhW+<ksWeX+t3SatdMdI7#6`CpLyqoIoh
z5C-A$@%{4GM=6!z>kD66N<_Xa|KIT^Jv|^e9s!;&czS=Z8u%LmUuyVN?|i+a|GTG$
za6ih4bM0f2NW%>E9Pgmdqa<I3M{11}*5PsB!spZ9KK^_slS!Z7fi7Nnql~WYQOXJH
z;9^J#?{&6Evn`re)2~s==XZk_BfLdN?QAQ+U5qw@l#i3|fB%$bnAsV~e)HWFm8Mvl
zpJIhHww-_SYhFcPwD1nF*(Ifw0LdzA$5Vz~y8m~cqUIc-r4mZEj=2At6|?hGoB-;k
z{3c1RF9b?RMeVHL*B1gSEcnSnkY>$8m6T+N>V9kGmkdU}7^d<iZgIUql6V#>VW<F{
z%rJv&gz=hzXtXdI8wvOS%08JzA%}|~>)R>6F}{CSJ&BH0_0-P#_Gt~U{8c#3SYj)I
z>@ek@tIkK~1sL7yc4e2*qmbBHw7$#eITRVk{X($nIxR_{)i=6M1N))V?!Vj6)Z(H+
zlbfI^)1{HY?*&XYF^tO$r{cE(X5rE=0}pCuID$Ni0wej%sI8RnOJLG4Lo-Jb<=Mk`
z{ZfBR3!}p?WXd-)$meO2m{iH775G!zGD@dVcwG$vCT5|vyO4p1>*(J<t?s_kUJwJO
znQ|3oico04cGmBI{PY0E@NC3S;O^sB3ZO9SDp)VH!HEj+=|Ne0|I@lA-Uz<a&1D5v
z<DN;^SZHVc$V$4|2nHE`<SE371lt6%5Q~2dcC9<z)~Ov82=wa4-9SY`ai(f_V|^iH
za}3P-T&ep%tg{($Omdi#)l4?Y7vJ*~H@lwZWD2~VW$x4|mHcIW6`o=_BM>9hapGz@
z(=91RL9u$uPq6Mh!;CpaP{Hhia>ny4;#pEi*D1G@Y?MH!ZUUWSM>Buj0y@q#5WIg@
zuW~Ipu^egUZCT10C>STV;?y$;{7pCw<&pJsI9*E`nf^&}=|-mT|8r*Uz|&7EGQC@6
zR6rD#*R3TLl@}J*AXEThyLhI3rF<mx{J%*0l<v!vNZy;fD&0+WkbGZH0&KkE5NlMf
zC~;nE#bhNJVq3K~L+To|O$mzf-H?BsgQoN{SP+w7J^DS|ds<Bjo~uXMS+TzBis9wl
zWY^_OblHQ_=CR7ncD<NnBKD+uCl*$i*8r6tC~Kn=eaiDDxl7ef%vDGF_MFDPq`D%z
zc$903grRC-gsaD@zDZCTtREqzpUWCihx%KtbB13w);V*97y%dO#UFO4b25L_w^doi
zK1lZ1QHwe}h9kqQ7MhC%XKmI4^BPS>Ms}>i27iv$RoW)6>U44*=G%GAeX=e?U#0!5
zb|9^dX`LYXZP>@`@w|_z1`PQ-9J0g5&)v7TkGua&jaORG<?r9`!q1uyA|V+u<y*$^
zkAGag-u=?cOlNq`3Oq$n_=kW0_v728`{mcvy72N6j?9Ld^ui@NM+%>pxmqxw3K?ey
zj!tMB2M%VXp<%_Lm!@6^iLot#$wAlEzN#}92Zpc-WLbq0h3(#^w+`nR>Lsd|N*a#R
z&d{CD-==D<oS=IvG15WaJ(qL9&>o6V&y+<uP|vfCbF4)=)3#P3*7Sdti9nvGKZdw7
zXX@WdL^F~}lxBv#Q2BedZ=k*zLMiK#yEsAKa`8A*Vxi~44BUiL1NrGlel#9mhdLnN
zi59Drj9etwEkzw#6rj|&;uDx5Up_r(O^>6%yue(>7|z&XhE09PftABArpkrGw>rNJ
z)#J$1yirL?;q>T)T2g;@Ns?CQfT0#O7<#BO1&*s~gZCl_s{B10dc~SKZGDwl+)xo|
z>q=!&uM#Du{zH<OIv7cR$1V?Jzoa`6x=-+!)i1&(J3KU_o%AhM=8`pX$q%L^>#(S;
z7R|0_0LGP48gB8#cU39CN^!D?FdHg&=vS7{S7xEm*iRZ%Q;UDWJ<>~TZN9O~P*%G!
z%&n^oK6CZe#V|(BnT@&bNavh}z!W*o{ZeSY9C9{_BsTe#iEONFncvLBN}One^rUrO
zFJUcWipGjGYCBXo*+)Lv3S%`K{ie#MWkM<^!%(3!RZaZ9R=p+M1I$Eg@@CbK8@(Q9
z+p2DE^eeUDER27(7iuh2wpX!DJ*skbc?-9t`F>0TMQ7LaAbrIm*?;AH+|;($YV;eK
zE!nhzuTn*SbJeRz3lFBRdtInx*2~^l(k>+3l*W<Id0vnGh53ZW5fv(f{7;Ov=9enY
za(UZHR&q85K<m&>?*`)x>xL8gVy$_DdYtWzbNz}iAEJNYiAUi?zp{SS^$4G@<M#Lc
z{OoXGiO$wfhjeb3tE$F$ABiV+Ouo$LqO?lrX0oY(7bq27^Gof4tcfHcRjGRvP6I8%
zmr72Y-#{;T6MJck`f|t^2`LNWTYO+~>mIU90p3n5Xd!l+L?^5?y+qLaC4vgU*b96;
zAkZ=k(?owF&2Ee0+@hHuy3xF!X$wlGuvLkDdpIcy{UU$nRv0Jvn%FtdAId60TuS$L
zz8ne-<uy~LuR~*7)|(Fzvhl#&NxdE-@%X6%44Ykxfphy#B}8;ubPyj$ldx*y2`zW5
z>xH*&A_?{*1hr78CYI2CiCPJ(VIuMU7?O$jM6G|fdOSr^kCFo@c6`i)vq9Vly#d2i
zV>r&+i{pgMMv+ArC-60c_n;V^(44kA^WWxI)g~79?clxkW)!vd(mIGbdn+8>&cvOt
z1A&{!dexMh1cvum5PE?J!`7p)eyhX16I|){xHu5B>?0u6RAnkvVXy7OmKd>qxiTXZ
z=v{v=@p{Oj9NC}_+1mj$)IgmAM~bSP!#as_J^0(vd*d)Oz^XNqjM9$Y8;Jx$(L@*=
zPfeXM2SU+s+G1SyXnQ-(7QMsy!f+j+4KJplCSTVhd?t+cAt}6w<3L?;e*F#(AK#})
zx_w=Rf-G%}9z-Ri(<em_YAML$#^}M&70`e744cF8RH_q02Lr8=*}8k7Y?{!(4hjfO
z!q^Qu&k$ocW%;d!SQ+1zoVR3>#X?KdmuG}@rC$<x@mhyt&$FG7Fg)s&XFIxXf2OBN
zcJiFZjE}#|>y2%5_u4ZbB{}$Te9H8knv-?=7EF4%Vr)g``*F;B(-eKK+QV-{rmlZp
z&$RO|YnJYrsd`N`kzE?L*L^g3CnnFTTcKm6a{c9qnU_DxlM={}y4qJ>M;-URTHvfe
zHF>uHq!vcH|I5n-;KHMu@j?Ce=LyRXR)^MHFZtSbs&yCV%vTCVgt|~f`8=fs#a`yP
zfIuQJKG3iF8{t#kUSj4Vx0AfoZDxO3^3wLBcjPk&<Y^`r#lIneL&XV%lZ5pX$v7*y
zZ&dfQcGS6yc}=i!{XNzDZI<P3Pmdirk1##}Bhdt5bR1!=<ZB3PM;}30E9p#xZIV1e
zSU2kc!cMoWAnYc~3c@<<EFerf#)GMOdM~nuXJ4JbC17P3wxjDOPC2jdIH`Ywb;5K0
z>!dH#f(1bBX!*YTY23zjGX%SJ7m^KHq0vq^YF{{}&1!`%oL@@jZAb5MP-?qwTsN~%
zXS&=02>u03FX=#Atw5}^<ODv;u;{#fp!JX$s`q-*7thIunRhz9UspRy-RRw2<=9c$
z2CLjHIu5JcpYA2W;l9%ZR)>H4PQ65VxJ}(lw5LG{Wq4iRZiG^9DV+nQ+*VrOfpIjm
zGcG`BLgC5?ZQaQf-U_`&;qB-n6yB{%>lEIZet^PTNlW!brss2Ao~e10?fJcq<37*t
zb8Oq|^qk_j%S-*a>he8L&wE4U@vPHHJdcbD$#>hhkk`^v#%}c9yk>umu-0zf_)N^R
zJ3nC~rcDBWG8?r*mslaQQ9Jqw8*N)Rvr%XI0XAwSol#KsmRteyfxmsoK;ZI_%=loR
z#&`HApT#z|=<;qf60y80x}#mt>b4Mc5dQYaLO=S*VK$%Sy7k}InZC6<y{EpbRPWY(
zKKka~Hb)l5Z%4-$8z&^)&87EZ@?Cx2pXYcKMvhO3T=OTRqgQ78+1=;c$KC$`S~Y@~
zvBww`0yH_7F)cR~mth$~CzslG7p8x#Z0zOp+lM=yO)_#fgZuN_-M@FWQI@SjP0gR*
z?&1SNqtHVGsKngn``^u%yWcwrMY2{t{*r%Rr$jodJh&+G(`ViMB@aKyWDLIfFHQZ{
zEu))%nm=TLIl*4(Bp77_!Y#qN0Tq16VtWycL1_x>MJ&|Jg0lX$RHIekT5NxEJ8#78
zw6)A5w-a{*Mv=`bi16%oBo4r)y_2{RiyN`H9?6}=B}p8H#4Sl2f^B=HXX(Amz17j#
zbe!>!XPJEGBBjhCabPYpY`+dekp(h~AM$8?`dCH{b<cH{?^?~`+a$mBQ|qn+WJfOA
z?b%jvXZDPPvcc@~rnMp#29SS<r2tz3sE*VHY0ZLFE_7L~xDH<+xqvSjDLK&9D2MW7
z;2O#Gx*3?a{Hb-M=mya#&HKFl!6q!^_FDZlnXx#;{R?%*GFAn&C1X`mmqeTF22*1-
zDI|MLw6$@D&)HCtrMiE<e{HIfkFUM0-YLQ>Z6v0~2|mvbQkcB2XX}5L)$58i$^^G?
zU65JC<_a^tZEfQe7`|E?r`HCfNW9?M=A3mlpbkh15SD6yUXgc4pibhu;;0MP<yaRc
zL4TTxs5NYW!K>h~jxAzkA#7s<>|~=19?u5YdIbW+j{Ln^#6rzjJlWcF=zwGx#geFs
zlStkTk!Nx#+L0Zm^y7a9Svi*7XUXm%r!nZ~B$g_7$y|L-o+-)OEu2A)A{#xNNU>t=
zM0ijfxDT($&QqA(8%bs3p=B^^R(A#lWQpx?5)6{JEZQSs*sSgx44}~N3h<Q@7N#_j
zYOFbepWR|V*9(*Pid!mt0`;Ll$`k)0)Th#BhWb>qcTk^Y>lT0NQ)9M|`qXRRrasp;
z@25Tst<Tiw0Ms+}S!j7neWpO)O?}A7q07OM)Q7B5;SB0S?3LR`eO8NDs2Mrs&Z0gN
za7vDoKbQJ!w{QkIw90FDPR$pCl@24)xU=SOS9b;mV2SOM7$8AV=1}UhUEMhtKx*=?
zPo1kpEKF)5_O^ffpTm97iCL#{AM}yJ(l5e&Ds5)CPc?f7_gS`X;XXBH`?ycN_HFKS
zZS#KavrvP~eGWi9bDxE%x7=q6^xfPCbx^^XBe@SUC^B$7_mP990-!yaakYqrnu%wC
z<#V_Xl2QD#xzBbBXOIJ@P}_5gKB889oBQy(qQ70;85n=0m?51+0a4C37WqgRHmf@a
z11e;{E5g@`SeVpAthr68&LKZwgJSDWB0u0FiN#-p{8ZY^ke_Py4)U{X-9mn9%=VF=
zg==T>b8Yi}@>6SYoBSMrdL};$*KNtq6zIFjPuzcUM{*w^NiD`4&wW7Bn*-dF8CQ!~
zs2O<WES-PDeZ)VNaF?IUeYRUTgPf!QuMcO9vdb1y#?YaUuE+d^V_f_(KGH2D>G8^s
zAlprtfks;i_mgNL$cqMhS~zY0y%EC-H`)TtW)tS1K`Gt4IMQkn3zMISwYSBtYF-ji
zUiiDSIMuvltV`u<(36=GfbWAn^(v81#HRd8eMo<hY0gXEo7BC>-Avs#tfi-Ishex6
zGUS!-zcj-8wks-iZi-hy>gDWZ=KI4?>$ZmqNEb?$Tj-MBgM?RF_VUa$V(`(_eO!I_
zRY%sUcD`1tc8%d*E!oR<r>MHJ8-g77#oo&9WG}S6t}$$I9`{@5BC~R?UL<kQD1gV?
z#RGq6V;e4p)L82V>~#Q6>UD&64DwiHR=_+2r3Kmc+mPDZeD7*ku;q8)$32<R<SPe?
z-`9QDY@BQgT}$0SdfRFo8;TQOf5-t)IQ&D!&0oqJscW-)zuSa(xo$LD-TsxV9wIgj
zu-qb;vm&sM@vSJP36v6^xZu^O@WCHCl*NDa{iwoWy+h@B(_vgP%}s{`n%s?u^7||;
z_r`8*X|$veu6nSo7H5aur#oXPg=W#6Ghz=Z!K^-|#oh<7wN%g}JhhTzidI^uU2i3h
zM$*m)J*hY8^r50&A$u@CRXg_s>lstcsifd@JT{CYuHYnvizK@goU%tTuEz*uTV;Re
zDv1?FJ)%~&+tjmrPhY=21!d^UB4ei4FGyAEFH6;{^vY-Ug?a^6K@NKpxQrvY{o>Fa
z1e1%gZk;XZGa9+&dD#=Bma*v@?)$>9$no;#kA*}H<4L4FtEV%IJw|7CLbtHi`Kc=`
zv-&#OCSz*>Wyed`Ka?7jzyH|YJIa5YQg?=p-swGei3X?7NZ1#_2nkPZ<^iDOJ@E0H
zc^HBsHFOauIR}jeK-e6LV?pMnuNAe|$DR>6o{V3>s>?pX%ssuq!g<4f?1+?5e({RF
zw|KbOZm_|=<Ar_Vm*k$;VO`4pVC3Ts(DMht;nLc6-?k!Wdd`bui6XuUOnDcVc<>(+
zOlF6+Qhchs)_1Ai{{tbD+EteU7Y!2vH87WPjTjS`{sssrm)L$6s{x&tgnt)Xf5O(X
z?amgGjTf3!&XQ?t3}I``O!Aix53Bki3Lh3i$OZ5+3g}Sjs`17NYxF|;B^uwSo|6K>
zE+mk%;0Ir7)HfMTkec}0MF6<nh3~uZnIzYyN2xu!I25PorBRgpIK5h_qBn;8FL`);
zZG@x64V^&kvR;Nf*!Zy@q<$oFe}9Jcdt)7n=n?6w;<Xj%d=Vd2j%K}&9+b>x4<ZL2
zrjImCvpcotIP!Z3t#l=4LsD9MEx;{w>4iKF&$`|tG#daYBMgMxX$lZAa9r*;I{V^W
zBEY)Q7DS+k(e@9M(aHDN&}0-&lkN_^NSc7)I`vW%vd}==I)U!^x??^hf8S(VqVG<A
zd627VFmY0MTBm`6q*_`Rcviuv$MITHl+g}*wT#n8`L=bM^PX`%Aa>^P4k(O0a@_z)
zsQ?)(HLZSXL-N(Y%u-BH!lFK6f-2Gnd@J$V1Z8;I{*RxaV66bFdiCN3t7%<3$E}E%
z^bO;^dnf1o?Gy%@iBsAze?qsG!j+Jmhul-T=&_hqRAgw^vM8W$sJ>UzH1u&yp-t1Q
z69!$*(z1lf^?mBj)9;nc&nTf(h$4UcT@HWJ!(XVe4u1G=YWlNuBZq$v|BPwihG0*V
zfdpd^65*C$IVcfm4I9NyB8-l&hWkZyYNqL%?71nHhBBB|Zc1aXe@%A~idIj91Xf9R
z6y|(sd)~zvGscI>+hW_|({KEi8&fBEXLp)xojFZGakse}Fj9b_kaRSRv}90PLDQL{
zM5(x-;J7`rJ5kxK{RS!!gCRZ%6<Yq$lobhdK}Anql)hyZO+BaAwiB7%7Hl9RjZo4)
z2^s1HgI*$&tUq(%e_{B)<@M%LCxM;l?6%@IIv7wWbk9Kt1FaCcEJ$DqI^AVqA3FQ3
z*g_{pSO3Uja<_=S^)rg4z_VxlY!~VdA_}SMDo(D_VsLs@#%9jI&i+Yud&L33ad3e8
zy}X>HC)3@x6iV|y%g;l;>*_a4yFPcC=?J{kUAWNo3oD@ff5W0rm<Rug&f%%BKv1W0
z+07B0vkv_J`Uq&Ng_q3;#Dr5>&j=ZH52pBy@ksq&$CPbL)F75{9-o@L9EL@|WsL5e
z;0u4F3{0bGcqt}1>)uA{(rgzO`Q1gDQwoNK4q6EkUl%Yj4Nv`@xtfgQfUd?^%db`6
zK9z0xuH&&vfB$9(g=)E++AA&+Ou4g?q_MOb)2R=f_`Dc|VlWUGV~VDfF%?&fyT5+F
z>dVR*5vUAkktwAmr;?p1XQZx^<;)oxE^=@&xu%-;$|Uhr_=m-iRl{RBSMd|CMeaKe
z=A}~ct%0uo=PSMBq~{nyO3QQy^cl*OQBYO!<qd4Bf4+4mR20+g#0BIj6m9r+N?iQJ
z4~*PDDxHJ$a>{V99;XfF@M=fr1;&gg)mxroNrEr!4s$1E6Y<yBxYwPuo#&&uve)KR
z2B4m2`SHBW?s7M$1q5srFNM*an;@Ek9iUoN*HSp@_vRkR%^Ykb1#}2BDmY`4W|MAx
z#@$8Wf9jNXG(BBot(m}WT0g>1F>BsPyvhT45tLc<{{#XVoKRpo^->B1+OR-zbG#y3
zG>;w3z-jM=3T8<MrCGlEhI2w|cf^>W6jqy7;<Ygu_Y~UhpFkP%mPw~tKJFpmLa(cV
zrZ>ur1fRT|IO493g|$LgyE=8!nrh1(V-n*Gf3}jZ8J}nMytO<pH~F%ErUZ}IwzN`x
zOm**)(adz_2*PCvL=Ira2%Omb!r<>_GZj^4*84Wz`ats!_pzB;QD$_TyV%EOYNcou
zu#Xg=cZ=xM3_y5r&tWs;?i5JV9|^8yl#9h}mDDVIzF56~;l5_hTnLO%_3m<>KDm1@
ze@VI@#f{OCaq{@1M<)nJ`Hq%IfuwAG$@h#OypKmmN~PyIQP{2RZ4?yEX&=IKP*8!^
z67Q~{@SYR(J_Pn_x`9B<cDA-Mb|c{99**BafaY)EO*+xot@Q>P)=`{OPeQ{{?6Tns
z8t0k<?nGg?wi_rIvf}s+6h?u{mg&UWe=QJpxW#V8HvmCP-}odzB9%4moivCvy99Oc
zOeyBM6Peu>Y$2nCbmo!t<8Bds8J9vuX|nx4@!tw4)=$Ke;+4KS(#Isv=~Wx3**#gE
zJ#%xSNgMsOV#Y5GO;cK_m3)Qo)=cOqt*2a!6pw|P;u)nitmP;%Gm2xIqS<9me`S@d
z?0u#Uw~DlsQR$E=(@Y0bNhrs?=4PB)DrqZgOGh)V=78l0Mkh^VfT~(blNzPwfaU^`
zWiA{~LOSD;-+h}C{A)~IVf7~VkJ-yolwM9wJwRB2(^OF`3MjB5X!|f>1zrj{9mqZ4
z;8G)T<KR-$`Z1rGieP!3(~!3&f9S&{u0^309$Ddnk-{Ed-$Fd53+j;t=57(4n#EJK
zMn9*%1=8K&bj&0k+m{n5lAJ`=n+D~;^Xd*HUEi3O;{8|T9mXgvvM;^$=zgm=&{RfX
zc+&9{P+mA+@$mbq(EU1Y*d7mxYV#!9f##>uGu!X1J@@OkVYk%6xF;PgV*zO?F5$Y}
z_m>v?b-j&2JZuX7IS>RR!o-p7tul9Oc-!!J6xD?1817MoiGDhwvfQohZ4jIkPjY|C
zz=HQP2o5SGcj|l_2<wD{htA~QE1@r|i6@wErw+YEWc&R;4c|JwmjM?I69F}sfnpgH
z0W+7en?fg-@R9@smk5X#o`1ml>*bI4&rdwE)Y7jFGNQ7iUlDdiG${W!VL+lwnMA)L
zj*vJ+o2lZFqDz87(__Jifk+|vh}D0O@qS6G<u6_QCCe*NA^Yz=KD>ZDC){-LYyMWn
z)hb$*$WJ1|L<+*{khrM%^6=susg5go_&|B4Ht-dBQ7U3Qmo%wUiGR`aANcUs43#}q
z@ot7yy%xf6uK}ai>Bp;uDce*W7-(p93$HCt5zDlND}l^7J;u`86QfsTQvUX-d3t+)
z`gd$-s2I}y&mT|z1xjGjiVie$Qg=W9?!G<!)?zK)FtEzCR;qGC;6yogS^+aC5|M>T
z_QWEA=YM{F@0wq}6n~B(B-3D9jO4PDZX}31l3N44O^HNgQimJf2baTa7Cs7i+wwN0
zx$w-r?Pf{{#f%t<De1CdN(jfC8Hr%z4@4BdOj$$3w-!4!DPPzLoh8RmFTyh%7M1-o
zJP$;aAw}tFM>d-rIXvm%X%u|0S3<w)n*00kjsl<iAdPlDe1Fca5$#4@GFP%#%S5c|
z1@v<n#2P+gFb!6Bje+&luW;0jl{L(8^B^}?PBLdPyV0!T=q^@8x7+otWMEuNh2Yb!
z?`Y?@wuQbS^azd&6|@*<LXY5<G@<v(EEt0iHK%B_>D{S??L&_q>|4bxI6T|<`KBlq
zL)cQV0J=kQpMSJI)W}FmBkSWNN+FZ^B@_uY*;ZK3(%%ynt%0F*9_QCTc=s2*{7k^o
z+3vpp`XjHyyMMahy?w7Bwq)UiSXm}WoePM0CrEcLq8(5u=@_~1eiR!iL&Jm<%V^0Z
zooewe`Y;f!l4#~k<){v1RuI#_V0xo;&~IxY6?Q|CSASZKdh63@6?cJ*<*iSb(+&_4
zYm+w0kUUSk^%<kTa{L?QR9Rfsz+q9Xs6c8arM$Na6jKzOJ@H!PqQv_rU@4jKblE6@
z!<4Q_fN~OP)Pp1-EYn(45>S-{)D8(8rgTLDK=B^*gCqccqm~O2*kc2SDP55Om$c)V
z2T6cyN`GJRPF`@B(q##t6ATH-LnMHX(tfM&BZ1?Tu1LTFQ_K7B4HB@(tT6Y91IH;{
zk$?$1=3@k4BxO1FK4IW6p)34r!<4%zQQD1STW;vl%H`dEE;nqG&<(vrl!{TJk}HYF
zmWt=;ve(4IF?@B(D;MQ=NX%ryTQoRqRfhXo!+*71t%9Daln*J|>r9J<sxq=DBq^aw
zNzIw7X=o{!A@SDGp$8jE6ECrI<})OveW!IIF<t7hfpt`g=ysh@3SCP7!q@CybNUy8
z_kK|0%1xvsUtM%gD_J)Ui=}c6dya>GWa{?%Bt5mI22JM`E>+*Nw1m^5oTPRPV9eU0
zK!2q1Dd$LuximGIEIo8MQlnYHB0(sSIp;>WYvY@B)o6^;(#yQFt{Gzu-8gpOBN^c(
zsW7`y>M;$?^|D1jYGvp4ds!__8c)?auhn~s^j%7aNo)5(qY6u>Hc0ue1C6QbGi8S=
z81zy{tnDM%GEAk~5OZ&>XnLr{5F#~|(|>kr2&tiWX)O;xYRZ$lnOy?`HQEA15qSGA
z7r92Ng{XYl$aMw%cw3D=J$w_egidNTxEZ#RqLUeQuBA_8c)9fy1EkbFmMwhUxh)%E
z*={jmGIIgh=0Qrkol5}>(2gW|i8f)N9(dI<Gzp^YGNUdVd>+F0$Hi+wao}WmyMN)7
zGa22|$4|>&`Sy++aW!x<-76o<x$5DV#KTw){tYj+-@{>APtL3-ut;}OgEMZ)(qi_Z
zOm~W2J$_xx26_#ZHtfo@r=mU1Dgq?WxJ&O@kW$v<{G?*i!bp?ga+jk@s$?@i(OX8a
zW*mKTQaE8{5oVtze%-7>NKNJMsDE3mc?fiivkFSce1Bf~e|T)$SHFDOG|v!LGwX)w
z`lSlJyJi#MnACEFkuQ!V8)*5KQ?D%ye!Mz9$sKa0DHJr=3m2|asTmn5!?YT4G?77z
zqC&laYD&UTj&u^Yqb(E@9pr9igyE(sm7k;5&Yf(>DCT4iwBmLDJS*8ugMVqoW@_kQ
zGq3lvnHu|1o2jt`n-Mcr*IjI;q56Q$G#DOaGn$GUGjtH#ahs_zt=UWsJ=Bp=yE^LV
zI&4BA&5GM&csD7Leg?TwDEyL@J`#*V#)pj(j?1M&v?i8j)#5ai7|NV70S(FiwJ6cf
zv8WX>?d&@iUMh2(q9iLE8h;<><E%EAR(P$Umve$TGmH82E*gzxPN;U(xl=lEu-`z>
zWvu+XIdBZwvG55*E3(V<<Q@zYFd}|9;p0=SA(r{U=<v>4QS)^3SRJU+^%X&rD4}9n
zxMNdwLFTBICq~Dn>Vg)K!rsy>-i=}#W#BaVjeE>g9h{?`Z#acV=6~w^NLytq+2wHF
zRp!5uGY6ae=}`YoI3?=G?`}xe6a3aOe0r-+#WA^avTM-U!kgm&sa4Xyt&V`HvJbs|
z#D83GEB0U&Lx9K3b--Gt-Eho8_8JBG@&PQTip!q$M)nVrxx&8jr<ufq*f)k5=_c71
z9P6L9iQzE0YhnO~T7TXWuXdx@7NR{`O5A?Ya0{fRqA8~bv8jWAWsAi&8!^QY3@6<2
zt!ST~idNjo<4|SOx|%Mf70O0Tgdfd1`gTO52&b^*mM1K?$|&9NjpLf-wf>rSL$c`l
zBj03-9F(o6OX6b7Ty2n5{GSH_pFF)NTg#~UHXl~LrqwXX7=MR@TTzm+VH7D!y5zA5
zm0I++gdcQ)lMVk2o9(3ea;hB$!k<D_v*(qoW4~r}fVXICpQ|R7nK&3MBW>m02FKb|
zzQ!D%Y4mBeQ#YTe1&&mAyE`6?Z!oSLjBlXZW1`_i>Y)>>p$2LXt!_SvI+t9FVJ4Hy
zFNl@pb+^c*7fFH~Net_qsgayS%ADQkt7RgWD(hT@$95`TE|ndr+1{aitnn-tj}`1M
zDjlmeR|>}p_WosKEvn_Bv4VX_$=Kd0X#Eel%o)0u0T&Gu0yQz0u}vxymviw7D3`C1
z7pi~xUVeN3bdW5PqPwc@AKwpuA6lWT0b#uDe*boe4PX>K>JCs?9lGy-c3%!Z7YdoO
z6h8i<zi*dFIxD<(k>uy!WcLR(exlAM`0k(7^;?-nc7Juh(gaI{y^x8pp#=zYgk`5i
z@PWw=B+vvQA#5kHk~6D?_48Df6aibZ$?Si;BD0g$B96?C%uS;q+9(1MXS0io9im~8
zsaTl|iwqITCV)%>sTJuNd~Yi08M*D;w<&dfOf;?Z&hfZ7=F}T|924*Ok?zC+kP)LE
z`X_UzyL`=@MH5FgZnZME;*sgoTR6qHI?8ABugFv4Lo+YLAGq8apEB*tX?hnW{Lp`I
zglf%3#$T*q@3Jje11jOPy0<w5;jAX&D98eeInd%LlYLG6O@4vepQS(9m0H-%?t&UX
zCXVl{1{5eA5eFf$XhBB9q3g^VVNED4O?Hrm?|suc&Fq$7K{Lh)=gduxY%7VCzMIIR
zgFU{x4H7FNs+Axh6xY_fiyQ%?d&+;V3)mIFxsR2=?kc((?j4<p&yhE|rSPKcGmrBl
z%13ru$$gZ!$<<^EX~yVyBy-|3r7}lyTcbCDxqa>XPx!eyvVUTiXW5xS<~j9%me*!(
z&3rm`sxtFZday*9yV6tlnjoYSH$}Tf=&e9ocA5?^F1g?H5IB&XzR{sx3B7-)M?KL-
z|Hf3or<S9qzbwZSAjH7v$O6)kxU4!V4)P-FS|y2jBPYy9aw+O?<YAGZv<Ge*MXJ)8
z7yX);VTixQu%oEh3)dS&nKf^YYt;I=RhBZ1#w>T~pUtSD6460<dn}`M;>lk&3<ts%
zG+e-h9l!)!*;^N^w_ejwrcZzSw*>QGljEY>Tk~`g*&<<Br$UQtl@-C*jBFdKYvK>N
zPJ|9AXogU>1)i8mpl;>Q_pe>^=9jla9ng(6EhQ}*5YN3eNkv|?>7Ct8x^!z(m+XnG
z{lwc*50*mPD=xJ>U()rK^8^t1S1#Q1Jb}uK`5?USIDrHu!020|@MeD!D>;jCaVfa(
zJx>sNB2c61^F)MCy%n&P$QoalbH+t)*PjQ>g58;n-$OhBtVihkH8HPY2u7fS$6!+k
zbO4rm@C_tC=}1ML$aQk>=Wju7uLTm<kCEG>@L(o#;|95R^S2<k3xs0aEqi@4iFMvB
zk~P--AjAz|g$_dw>vVrH)bm15sc28~>6Xd`1HZ}#l4#zw6cV_xGw62Ky-aXL!O9~9
z%HU{_m2rm|up=Rsl$o&ehk^n<Wew4z`0{lG*i{BW1IQTx_6%1w%q_-cRhpm$k!n>b
zj|-!sL$#A7v^O0st^xLS2alZq50n$|Ab1Db2uo_JEIh%h>#Bd)4FAxiP2S9O4gOhz
zebL1>d`(ZBoB8l20k(!Rp9Uv<c+$C<=qEwS(>?H`fp}N<#E*ncgpun00FoS;Q0k^M
z&SnxTIg2P9;V}TxP{abz_XUu13FR$-alhmX8nq}?u#W*w#(Dw10H>6yjE?ch8BN34
zP0!iQWB%U5cjTAPmKP*{xXttL)V5MzyX9Ka7i})^H~EsSBv!glA{)v77=TJ3L<Cp1
zf=L-T25zSz)u;nj**XF;7qWHyw!$;~W(LpnJyMp?*Kg$_jwxfSBv|TQr*qweyXG6%
zRi0G^QeSg6kf}kR>Fc=!+h-DU@A4Hbr_!r$auN0QqvWZY^_*LO)N}ik_?_a^JomYG
z_m@UyO&7S0JmZjvzm#P((hAzd1@II+-a`2n=EG95{8Fy8@4u9+x9hcAodwF%_)|q1
z<*H2F1-_%^YJNxqbH_9Tie8$D?$~zF%%2dDMp08UTAueKb?pe{X!{_nt_i_Hm2Npd
z468RYEnxMgcjHulHuK=<?5(k-dKIVlsJ`i)wmGtHj+_p@8MAA;qj&ej>>8zzkEkKh
zbnf-H)sSo^v63@`!at~nL=)?3Tto8Tf$X~lU(lxp+At3)KhW9>T~%Soy6&Jtpsqrh
z#|D|6`Lf)PUA>Yi2c=FcwYOW$1@$RIuE0D<eT4G#xO&5XRNC{bEytH(tz~W1eU24|
z59`l8`^8&Q8;x;rlYiMtVx>zZvd-ZT!tshiz73(q@qN&lq9B(GI>rSLI|?xLhB69&
z*IXz2)8^bzjnHFkxEDz&L-+oSSBKj8aX8@3NJ_=pGGETssm?UkpPsj)m6#t?zNL5#
zt?<X~rgvI@$|}PdQQh@Q8E${&{>Q5DQ6-NG$8mRgnof~dua0#?J!6c9{gYmm6i&Uj
zs?HcIjG3wq#<gpW3Jp+%mlG8Zyf4Gz0tF{#0YLD5s8HBn>(Ek~uXSj8Z>w|N4X~S8
z7XWtC8?W#vv(99k7~>zw{EXgI{Ai`)25wb$zQ|603aDzr{9vs5!;B%T$<xhe0_%m5
zcP#^n*0?Wx%}4A=AMH(LAe%|7<1AQV<%6m~LZFE8e@*%QYsz;o+Pp1z+BpD^DJl2l
z_{@+|@Qko?z&U3mzPYk|H`DXFh=qs#K{YIHlzuLC<xo}F=l`}j4fW+v9p@AuR?;zV
zot}0b3$&moMG)NH;@GxQSm()!DulaKytMa!0oKgSXO{sN4HE)2GM6!P9ut=|nitJ~
z0gepsr(eH*c;R_Ulz(FtLoP@Dk#~*g8qpQLt$n^D8owgC418#ul|o8|-0Co<l=<CA
zYtj{-D&$N<pipmxK7JE;>d97c&!m-<H~zl-7HUFELEqB0@TUVWE^`rP;KnC9bNim=
zmEMUQ>og~2@?7%#rgszC#u-Y`OzVw*6GR(emivVXhJgbEJY7*4wwj(vrl)K)z4wz2
z23bwT)C!YtAkV=R&LjRz?L|cTQoo_D`3iIp#M+1lW_WIV9!8h>(kBu4jOHk-S*THc
zL0U6|HiB^{M(GIT33-4Z4?7_QRYzpV|6j`@E2a_0)~ECxPtt{tw3wJTVLgR^+xO(Z
z&##xiW=tvTnKfqk?b}Oupd>JfJc8v+8W{fm&+x~~j}$hqSk5@U!*uwApZ@-HAm99y
z9hK*3HQX39tlCilM^jRRk)A0tT8;AAm?7z1XW1l~G%n-2tAm@!$}><mk)xE06^Miy
zzxV0?Wlf;4&&i$eORHmn;>1ONVT8lkyHUrw*m~BE3X&;*qQIJOeyZG{Qp|Y@DEMuk
zsU?%%&8W}Yhc&=Fw-kwg-X0gtM1eVCTy!1A!ff15j40wp$D@_6#XZIdkz+(w9J+<=
zvM>zHY?Pi3$_!d+X_AwO&OsL+og!oWU*~6+T|5e*nH7qT3UHH0Aw2Vc)-B1cm02?Q
z$&CAk&kfTPAXF<S+&`vZ9lzh9jACldKwVup=7lE2Pgg}2ZZl|46H+lDm*h5=Syvfr
z9C6nRX9a4n>vS=CVG-lZI*%K@6$LTb!CoI{Gy5ev!etf|)~KvIjl71*D4;fvKrN9X
ze%!WOCmGn8981nxOk!q#2G`wwgYu;p{RW{v!u+z~xFC43*ZPWa5R?Ph%U)*ykU($0
zhP*x(DJpFdp6!dWdm*ECosnb8&E!|S(8vqqLgcmyz8VhAqCdt!if*lB7F4YP<=1JT
zENX&4ozsRz?Xk3Wjv^vhFCYo7n9?&ImJ=4VlFvL--Ui*`RIsA3VX|wZf9?!%mrb>!
zf|$*E6RI-IV_88_H9<VUf;2FTBNhFZ8pCcLi8?QZh0Ij;9syi}GWuO--7}B@O$hwF
zP|3xYUhm+vWk$<B;^QtL7m6Bd*1P)avYA1YGYnV96KlfT#WLU>s{8F~9GMit(rR2`
z)D^xI;{{<-zl*~;{Copae+y<zg#6`KKKzake?a0;E&dZpzm3E2;lIPrq1wL#>`=N0
zj77G>z_tL)2f;ipkOVtHXspXoy4xXEa)#kd3s{ClNqbFkW4EGbKfnZ<xl+K~RG@}t
zruWfb<@mQPvWJ-5e<!p~OIN9Mne;g|bHh8~uF<0lpJRrY=wmy#e~HW@cfIPThL@)?
z29*ZnIeaU^{A~H8b1x?4Wx4q{J}s-U>FPPso0^o?SpR4ayY+U*9?<7ucsbf0LgNhA
zro>^Z!aFp!WhfbvX5o_L9OQ_RGBg)M!w*a*@tLgj#Lp{5wR=i;L<NQ!E{_(KMmKK}
z+=p;kB(*GtSG>pFe*%?-c*!K^e7Tj9+C8Itl9J3|CijT@+X}JL<0GsBy~pFz(6VBi
zNjij<Q;$XFzH~_@Ea|q#r*s2HUq}ysq;_wU*D1Bf)SaLD;CuVu4l5riX}}d0>s|eS
z>3&O0pF$yieD|9o=XxjZ7|2OO(T|I2H=~Tmrwk`%LCq%Ge+k~|eY9K<@~o%=yQ42*
z1;zkDnTL;V&qO$2TmpSV&UoqeE+v*A*65qv&>fA+BK{V2Rw5Z+3<fK?h}e9yln>{O
zn~sn*p)V`T;?&w8D6v4~x8)=?raq=FH)GNv_ik<wz&o28ggS03huZV9yRlCYJKd2{
zP-OhN+OGE$e_hGrwi`dj7R${_z66Nik#2}(X7}eqTg6hK_LP0BRZKu)4dv@>HRQKP
zFrB+Ul<6%UJc}_x%@D5Drn)7Sqp18XRJn7`{Z*BTD-nHZe0<4efUDa5eLr?U-&JC+
z;GBD~?@Ce3LA}?kIfA}Ja|A)P_5NB-%~g`3VA0A{e`t+{ELuaJ36A3u7E{keR_8dv
znpIPyb&5!|*&0~v;0urU={$zW$3{M;qa%BErgQhR1@AUKIRj5QS-&=VTqnU~OM|z*
z1VVNIa756^=135`K_j(j>G)3w8eo|Ky^=2>fGJEg2z3;^&DVw{5v_B~18U1W$k?O|
zhi25uf7wZCuo!W(*^#9BcrMY=NWa!LWLQATQ0mh3Icj4!d3b@@3r_P1Eo7mwjP-JF
zKenJK$3W^M7O27$8j%QMZ-Hj<kl~MCpJfAZC0jXU5$f7DmYibb*H>F~8`fs~wi5S>
z(dA%x>6TkUS`^4sM(j8&6qdOIiqxwgvZh;Nf0_i8t+)CxUzE{xDIZI7?w6KA7<A54
z+$%0LsPl8aW_BFZA|rGU#H1RAP(QNTidnaa_sp7%|AcUu4>qjhs~l{geD(VA`bO`n
z>ebhYI`5wn#Bhxze!Hz#@BF%|>o4{>oucn{q&vr@2=yb@QF6o<v1c7#Gq2=q{3lpP
zf8N7h$5%NQL8W+CUrWxn?*r<hVVrBWrX5~VqK61A&UJH@p(g^KQpb6mm;E_MnR<U@
z1aA@VMDUh)6x?NsSEH{dXQ=uM?=+t8)=f|1bGyYU!K{MWoI1_Sd>VpK&wjwbyV?LR
znVg!w-CsWBsCUcJT}n4%M*}NWdo2^)f6_2@yISH0%%{{gEnwI4r#P`qUeSumF#vij
zYWh`4T<4-KophCGl_M7S4RMSBu<g9v%q*OUDN|9&W&erBoP1Hl#pH_+r&CCempHi*
zPSCe@v2tNgP<Y;psyw$iq%bRI!6?`3`&y;srw9)%_HNrBEG{cCR*>RjcfL-Mf0j+A
zC5|TZah%Dt=zC43MTsUO8k6}r)?^6RJ4}WkJjrAXnt&Nw1Jds|nHK4e$+X116ZB(A
zmmPQ6Pma%K0!hw%gG?4urLZ@sxjHaaB8bbW5+SBhXTHn3CIF^XX%++52DT=>YWn80
z_uTdrILKYD6jEt566z)c-oBwNf5YyjHZvZ|`5?KNjZ<Ex`oK=Y+_BAGT97Lx6Y8^@
za-ka&3#K0qcPahIPsEB#AV|wDp(TFuvHMcUl2LmpXvvmaYTe<HpEqZJrM*E#c3J#Q
z5}}@qPJA@`TFq)*a;nwxDRFTP^>I7a;jW$CaN2pz`HWdk(T;Na@aCo;KaCK?=uwY7
zunY(Nj3=LFf^gbSxChT@d&Mj?wQp5R?QO&7fZ1aAd`CLM)6o7a=cn!*DyR3)ub2M;
zs|!n{mjM?I69P3dm$6MM6PGCv2q>3Ogfj$}0i_qI0eP3Jr5EmhZML{hB6abQPC!L6
zjq1iwv0&+7oP^|t)H)0gC{x&>0^+Vzc`29Qh}6X+I)RjN!Z~wQm^jzO#>fb>?C6rn
zI1j-a@|Be&D73&4^9_ZXUl=)se9cs3bBMqsOEx7`@RN?=CD%%AVhx|3ly+R&r4ZwD
zcuC<W)>O?ptkH9SDdSUBTDoQ@wtJa)M7dsS6Rqz?>(SeMf14j&m<u*d4~kE-UyIq1
z9rB<&$d1a2U~J8fGpk!x9SFiSuvfx+Qzz);kaY#A4A_=urJam1KQw6K+(j&?E=NM|
z?c>j1huNL)KDFWN$Q)L42cc!A`l8nM!U3dN?c@$F2^Zym{(K*F5W)EaA8PbA%f65n
zkXr}7>|?Uj5oA|T$snEWmGuJGR%DzOto0Hy=3`kCI*%!PnATgp8&kO~`0bYXdydXy
zX+cDHOP$B$kv#nEbsobjqpykI=UUk4SJcW@y^YLc<OWOOpOtx+kMIO?+F&8+x6yg5
zt*~EO=3PF2q7z6dEwHV78=1!_-MEAOLOSpAA)SCq24pF|Do&hhVq;{4S-o!dhB}Y1
z-&QVc(s?CDq&V?$qP5nvm@h)(xrh(d3yrEWq9AEDboF7`qF#|PmufX4C$#r(q}7Ox
zU`q|;{U~M0SHYZ*XCfpOD&i8X>1?T?ET+}!dCy0Gqy|EqAZ6#Z`>C5unawgJO@yv;
zd>qI1g2Om$p~&A-aKKTae7L>fAcdKi>Y7M$wuz0NMcGFBZ6pUFOc~QFj+hY=W%gx#
z%5KHGrbI%iRY;Q}3g1?VTt4u1KuusF{jKw>CV@-}<6amy=2ue^rt2=-B~xEYF3)x)
zvDTb_?2`W5gxBTcI)Rrbp)|jZ9x>h_Xz7O6^Gv$gf=9*86|6$3XG<GbunZfqx_C^d
zuu?)9drcTQ+r+lWX*@<_+!qoghGN;4L6sodhjLUyesZVEgCav_cVqb3)w}b9DJ1;I
zF{{M`q0-uvuvElFI}j?NE54Q?MUVNlP3xzBS&s6h2DT@y%e}Zs*MVk|Cii$;(lp<9
zSt4!mu(gSIy)rN5Agk4E-f(|%<a(4@A7VLue*ECB^d7oHzlr^%Yuc<t$=SwZ(k#Y>
zc)$z8B8<n(6^|-Ht7L8I<M5bIcB*UF{r=GAFUV0Nz2CDRAn>)8c9EyZG1P5`>K2H9
z((%$mOonyeUH00E7*Q$H;;HFk)*MAuoxBKKwvfQ$GF+Zo05o|J>x)eznAtjnM&=2$
z?8Zs1xw_IbgWGlrG?|i4?`L?&oZgVgJ@}5uvG#L`#CTwQj5KV$MQH=}%t`hxKIu=|
zNC}UeIvHP+-EP^VIDPbCJlQ)qk3*DyMd+N^tb+O)UtupGk4=58Xg?QBI#raBbBlYU
zx-_{DNVY$RpUn~#o}^rRz!T#!=3wSyp5SzhV<cHPr~0-z6lI<+VjP9A(|!@-1MXsr
zW39Udduf;7qUUz51(bfB<`gm2u7__cW_#DR`k#z^?0N65jR0zl0C|>E&(9ryt7T@x
zG4#j_XXqwtzph_DM{(gaAWYG$9PieCH@4J|O{WcwyDhHR%Xj<nrGxi0-(G4w4ZObF
zm)NMqMIAq6*@RE-d{<OXkZUIbT4R5%+Of;A5Lh}rDus=O<7#eAf5;ZoUp>mJjty>~
z^clii*+~MHUFznRx=*@Ih=PWHX9nY+R?;6-T-HOTR~ez@=;J2~DPw!V!CAAE?3Ckl
z>RM6ao?zqRnQ4|SX;g9CzmLJYsgWE2%Ch)UL0}>UT0bv+=wDx)u%>MBzAUl#d#j;_
zg4OYOXhc9`Z{W#-BTgvJ>g?VAzBDm|6~P(L(<V-6)@dJSi_m}f-_pc?xy4=rh-$;M
zmmlHZ{i<()1QAVJ{eQJ1K#3icKj=u+j(EA1MW&m%_w}l8m;wR$z3kgEZbs*=+ahj9
zdO=)(bJJedp*BV%Cd``r>;6GI_fD;MMge}!;Qs3VL0Ov%SAYF}zqY8rKtzYBo$}tZ
ziR{o08IMkWT<i#mN@}5hRYB{wbz(Y?O)ZEC5;)~%0P`~aK{02iz5Iv(aFUoIb=0*w
zG5PQ*%EvQ&%d9-pwZuz58_-|dc)94>IxqWPHuAD>avm@Hmdm{SY8x>-Lvvz2H)0<A
z0hELplU>q02L}fWp?PhF6-MTf^E%Tl>>Np07V!-!aOzCUL25iJ%daxB{pDuNSXJh!
zg>w$uzs$K>0`?~=`);$goBCU2V3la4g!KCxy?W1=@wPJ)0y8z2p$<0_mtd_I+X1_`
zJgyfY1_vs)yPy@Kq_<<V7q9_;QE}qBQRLlATM-!pOy%H7B8QGhD~2;P%d9raynAgc
zGAqXbQ^4oQ3>XOrIg)wnKJQ-Jip<ht^l6?XbLjlk&;)~Gs1B^kynAhHGGlUK?IF+o
zD2a_{6_S;9*I7jm#kf7ox9!ql7|7e?n)Vey-{u49K3kVoD6n-SQ$MbMzG-i6p3|NE
z?2W!Z+S>s`6VueKc-17heD5!f$G}m}_fn>&6q{Qzl%Q=P{gWtQx&BYeYiGHtFDx|+
zK>xSq;=Zp1l)p|MrahOkDZ|*=>O$fan@<$b=97NS{l#{qcxani3X8c*HngM}Q#I8j
z8oJTkYuip!jk`6UN)yt5ZL+OJVrGP|CW6-&Q5u3WEWFG;gCIm{c*0N8$;=^+wFC%R
z-k@vDObe8<Rpq&rl+sZfNn`5$$WKV~Mt(ZkywUI<C@r>C2#GnFq`KK^1<}5)Daac#
z$tqizZ3^)ko#A*-id?moT^qSF8nOmkDLK+|!GAIcZdhSV8}p-o(bFHl%xWZL^+qYW
zviQ|Fnd~8)b8@qWE?d40q499AT2ASsR5JIrm&A85YA+EA4`2+mj`o*?a+9gD>htQ~
zgz|Fzn+*Qn)W69hR_fn$$@}ZyQev9n)c-;KJ18R5TF$9}6KYSWffEiVRKY35^*T72
zp<j2K-zw|JYS*cMY(}XrRDE;hnQq&ub>G}FKn;x_ZW&lMg`w_@rmseub_6mUlSiY^
zf=o;88IWl?oB<g^aRX%98M{iz2{Nm!cR_|w-2fS~ZBKy=F|-pPvuFw<$PgL#Lq>Rf
zufCLU3^Ih;6Cgu4oB$b0aUC*bhQ95E%sT5mkfBu9Aw#!+?HQ1vhISk>%cejfLuEVw
z8S9yYvsx1fwP!%4<!}aM2*nMMX=muS-H=&jy$dpg>ITS=ZF>r2h@qVTnMG3=L59e9
z05T|%tNz!(Irs8}>iWI>qqUelo;@`blcgPMC{8WHl;TW7arnNO7^!9t-opQuk#eNu
z_XEwk*{rsINs^m|DgS%hU~tSV(U83KXPae9bwj(zj*i^?)d<2%%z5F|TtY_gG4I@X
zvXO5t&rrTQtg$$1qTBbw-%g;C%vfW%S9Bu{R@V4hb(rzMPKO`G5}PNnWa{};DX>R-
z%O@A-+w$jhF|o&OOJOLw?&uH8NF<t?Daoa7G77DKhdl4YOAKG#)?RJg%<z9M!nsB_
z^Pn>cjeYiPU(yg=cO`o&Ud1`ItlkfQC8{O3{Eet~?4BY-ZVx3)`Qh(FZIhxnlU9r!
zSj*pVI&U*_Jkm;G7h6q<6g#|ii9$}>jN;T<f_^!Ctq0egiEqbnu1x&bO(z4rrTSc?
zx4dS5gN6*_cJzTBQ!3;8@w|;lGK0<nWg^KAc$P3=QG<Kh!8b~h(M$x{fzT3XoF_<0
zTDJ*O64GXL2C2xG{xer}0yiIl@x=CFA<%yf|Muv__F+PKrG1!;j<>GwDb^${0z({n
zYd*at@fRD8>HGWFm;VBk(4)i(Wo~41baG{v2^S3umoVW46qipdD=C*ihzbOk*18v=
zf5+fl>iYYSUtTzpB;ZptpC3P7{`*3SvO+R#{pF9JFY$&}f+;UA)-j{q<*$ET{(Sj+
zg-{@K=HoZ``+W+ev&>5uLH_nTzx)$!{sMzl@R$Fkp+D<1{PLg6KVX6-z@G60Sb+e-
zGr;^JS?~eD4j_yQjO(x+#7fMpWSNQce^3Qy!BDVC?YscB<JKC}sU4`BGzwrF763S^
zod7iht>GF}O+WId!T$~1Fz))z@}zJ9N7?u*T%bw@hMEshqmZADziFNCCl!KE7|z9e
zR;7mOPJJpr_Y%IqPa~L<;%tT*X1D|3hH)M$L!<Qo<HT(x$-8k|pjmU~to$UJf3;@Z
zRFV~vtX7h|8?^<JWniAGPa;_<#)BBqyjIM+aa*8Sc;>bKB$}mULdi<AMw-n^ns?*2
zM6-dgY7aT?tsqv$DgbMrgPn_2G$<zV7=Q811ZUM$24Ot#PS_$I<*tIqpdP()^1{)7
zM{lhfZ|w#Tt?&4|Nb+1tSPVF4e<O$40Kg4_K=uH+%4-Y|*OYIV8&1RX^f|QDQT;Qz
zd^6@M#VdxHP>@E55mE_KV*<WOmq?ERC`5P>jwv|SnN#4}&YtqL8gNWAnKo!vdO{v0
zanx`p=6JerM&T?zjm8neqKVl`L!~G|a1HK}miLUOhMbN+;;5J@#tbh)e~~6<jQLr9
zLleGZaE_7ro=0era>fTzEl>tJ;TDx+qzr~LV|CRAg;HwhjyP3bb4)7=^f$mN<*|ht
z?cy)-*_ic4{i)nRISum2|3|#RvT$}=6u-=1@oCpcF=x&QGmQ*I0?i!piF?6@QkF*B
zd?k(>+l_jLL3zSq{+{l>WL3J~2`)b2A=1rEY<y7ObAoSNWnnR=G*TCqBZz^q0~;hY
zj*P%tUl3&INpw-1u4IYm{lOZQkKWB#Z{gDhE-!yY=Qw%i*Y73X%zGH+nH^H8s6bVD
zESlOKDU>tD%rQgli<eox7p{L#l=AqLwk&`eQtBU@`6x}y81*b4(Lz(DS>)|8M94$@
zRNOLS+{ye@A}}RK8kUzuX8Ea<WZaLraPLECv1h<sc~$M%HP!bJ6t~+kSfnx(`k8iI
zJO!CyhaFqX1Qdb<(k?p&K0qo=A(<hWF>hJ_%za4r^v5w<202SH!RroVwhX0WaGpX<
zwj2{WKev{@7fS|zDeuj-d#6t>+q`RMhOOa6p6#Y>foE}HQ20;cneog|bn?;2FN0h8
zbC<)<!m}G*VI|LY)3(4fD0PeppTslm(!#(L&&&qecjLH3yI{s0Nr$#VSeO1qsLF{i
z>Ww3y$Zk%im6jr^_noCbhSIT(i19;#P<l<2W*|5Mo3Z47|LRI`DLk8A6uJ4ws=hi9
z^Q6*qeRbe}?5ks+^wq&p-7_WJ=#Q(?wce^@C%V+&9UZ@=!1W-Z#<x9s>O6%hj`rL^
zcxgQlo0fF_P09$FIsjqpjIv5^m#Q4QVzg9Vzc%aCp(RG(N_LV``meAslU2H*)XBnC
zm{3}(q@R?3xoX`iz->54(DorpLU#k7a#)_Ddu5-QsqQ+lP<CDMni!nWR&OW(%+2Xu
zMbXN{HyiMXM^XegJXM~EOfX5C<sHdMvrHAjiHG@aOKke3RnS>0GrnCOZdg(>%LLEz
z5zVVDZMwgV@{U<42wUaHQ}a~8Fi0t=EL_vVsd6iSnT+XrZ@KB3tBj}lTcsQG>)PsS
zkO(%Ms%Cvh3#bw?a#ERWTnQnIx-Bakdrm!!5;09TBeYzbJW07j*&wQWEGk*Gz1}6Y
z9{xJjQ>*MZj~($dF~{{X1H$YHPTF{XDr#^!OndrZ+nGw$9#Ep~0aXmc6*}E(Bbe+j
zbp7@vX!b>H$FuNUeH|1UDiHMK<GK2}*yq!dhsbJD`Gf@J|DEk-7MB4R4HE%1mqA@D
z69Y3iFqe@`4kv#cyOQL#F~6@EXLsc~2#?0AYQ0r6bGt4plYI_ck|NtuVg*)Wf4&<)
z<H^TxNXh}1vn0>}`rVBN`Q_&ezx?sT?dykc-W6Odb6Qw=`TEmE;0um_1;>X*U#t+!
zC~^7v%jNqQ&ZWWYh8J+I6kZm8JG|@H&q_FKAv|8<^|OBkf&VFimrQs0j@O-jyVuX*
zeS{PH+82VTQxAEKZ<NoiI9>)@0w^8E(KHsk{`mFV51f|7nK$8QxQ+k4ef{w7590!r
z^GtAa`TH*)!V6G>DJL(|V<+w}zy9;`(}(Xyuw^igeV7h^@Yk<DUxp`tCLB9~IOfJ^
z-QiepL_B|7jPXo`R|&@&mkRe`RS+QNKXIV$2@X`CO2v$DBUH)vLsXIpX-0@b1xnL>
zq=M8L4Vs{LPg>~OGNaW9og=Lcm&6qssjZPK0Pg^FM_z*Wbl1p-`W|`&u8J*v?hjCY
z<=!?w2`#lt{lU=xI5@a&Jrjy)#}u+c#_fA*{S1Fj1+Lbp38Vv5lE}~~!~?`aC3l=c
zJ>*KEAlVUf0fVMc;aX_}w-c$cMeHUN1Is{Jn!Dm+Kb|w?I*6xGu7}vkmp)Ft9{N<i
zz!A3ZL<OEH37%-73+*y>JoU@8ygrs4Wf)kODIvsu+36siLV+IQnX=PEZe+)i76XzW
zOPYVpbee4yqf#?1eBtc_+Ngtc3PyW~;sh%+Ks>?n^bi}#AS)B-0&hhc0>9sqosKdf
z=$Q!ATBno%ri%hq*oohgQ^oJk6D=d813J*GGzN}-oE{sS=SKx#&JS=vEDVai6%5PX
z#h;l{-Kav&k7^7PM$V5qp$JEBiOWI~egc1440@1aguF<XgwCPy9=Z^&dq_)^(F#qu
z8(ZVDcabM!yoYu~*ex()CrBKGy|#ADqm+-UxzCbZd&vx_;%;+t7;O?`c;8L(0rEP@
zhsJ6$jl_k)I>-3OrpQZ?<qhwv^?iEEQpga?43iM^+#-c#Qs~K)B`HBNS;cwyQk{Q-
zB^o#kD;*RX80rxqkt12$`=VMCLi8M>vX~l+D;4S@ieuP4nQ;seEVHtPNs7=4<kZ}o
zfa87HxgqHrpe@e#JsXo4JiGivzBEn%%^;hUXNZdi&C2Q8_RFJGz@(BR#07bj$}w<u
zf{je2?Iv1kG|ha9*Uo*)#yDa(W6yu(z#@hsu5qlIkTKrJP<satY0s=NXYo)EZ6go)
z--?HZwVl92J;X;mG+^TZ4-Jrd9)iRrkt2w9z4j@*G(*|KOVJ%HLDGHt--Ez*2k(Jn
zj@z?%uZOmQ_e|2*AzgPe|9R#yU>etVg#U=6dWcK>2a;J36O0_-zX4L^KU04LiHJE0
z#5>zsI4dW(8qTUIl`FlIZCV8~B~H+E3d1^ZXK`B(?FnwvzZJI)>pOwldWes>ZNSI@
zZW|!|7Th*T;GM~7(lF=zS)A5G+rVjWr*oKKVEklV2RJPAY!7kKo|5$)FmZsx21rv5
zdz>8H#Th~~uUf`{k)>@j=)Hf2585mw3<UG2n0vtd)<-Us(884?*2Qx482KjQ^l_T{
zn|IG=w}xmm@W8mJ(LAMT;aa?lc0r39O)268rx-dq!vgmt83ss`9NhDnoe?76pYf+;
z!~k(-S8Yr-jQH7+`#}i8PBHyueq0}dU^ljI^H0UKy6;vv<6Pc$V>W*)oX41bYMx~J
zw&6@_5fsxSl<9pabEtEac)*OcwP$)F3n`!{Gjg-{HpozJZ@Lw%WZXA#H*HR+{enkI
zZ-BhSsu+KRFq77|eC-#Nab^2gl*MU)xL=hZCo$nCI9gF9k~?Pk6i+*Nk#cU^=#!)M
zp><Uu&yP?3>i}^fpIm>0xNE&9mz)KBSoQ&}6~cb)HxTiuX(BW&<yP`!#j-%29r;_Y
znDwxgJl-)5eoCHQtf{zRT5n|y72Gy4#nkgEL!p0$zRig$LrKN7op9U2#*$Ll%E4J!
z44NJ6Lv{58#IUKOAK|CAM$b!}ps7@FxP*VOkkUFhTXLeMQx1QVjOA!H{cf{~MouVm
zil3S`%tnYu{8Tc)9mGSGDcMN(b6hm(Y13?IE#G#S@=#d<4{%e>G)7sVCgWC9HT9kL
zo3YFY8ZEiCXEMS7k*|i9L+d>SW0|-;q`@#wh&!}C;HRC3!R)84VsMwA)<GOaskR0N
zT1;?r-@A!A0$P7^(I_4pSmAAVvT_>>7DsLvxB4E-@}@T@GDdk*9=%t0hEmJ&)<IlU
zYFXZf);nXAEOR}yBfOvGa)5ZijAZ58LtIODWCc9nXb1miN!mo4EG>QGvEt@P#sduV
z(-98E4k^|77`ac)(>R9LR<v2ethJl)7SkU4oVe^;pT>Wf#kW2ccPf<HsI9mwKK54J
z&FI3n^OCs|>8?w{JZ~+7-Im0q^fw60)PoWyzz9u2O;SH7)Zy^3v}PPZp%Mw_B=QK&
zmU-Mp&)9uyxN+_Xg}}RL?(=6f08J+bUgH@J41tBnM>H~o%2h57=kWIoNQ2=A8VUsY
z{g3?e4}5?5EjA98=0CCNFJ%~h`Pbzy;S9(Fz!+i&z}jKB>(>DD3o1f1k_l`Dp^-~N
zl>Rh`jWdI=bg>Q=$?Nq9cl8WQNY^NwGkWtdt(?~M8#zi`>8|K9p|YCpw)uT+uDsD!
z7uIn|LD}xLJ;E78(LF-VXF+UaU<jKQat{NKk`jN(tKG#h?X6LB$G}v;M^m5bYIO{>
z8*bCW_0J02%gh$t@~(3r_@iTKA@^>0t)z}eOmeX{j;T_i_Ny>2sXi?3+|qcoUyQBz
zOWP;!Zl~TwN#0(if2kE^)T^oDsW5X@Pu+5C_p1v=(n$4CMLhf#A+?7AhP`;26&-CN
z7><AXF0L^2i9#isds$<ggMO7|m!Ic1((>#3J(QlN5%NB6rnJtKetV6_@qlU=m4cV#
zz=e_06P-EcR!1v1;xu!St_B8;OmI3Ql)1o$&BJ2mREC`mkxFAmn2{5tO~($)+~Nq_
z=fjJ!xRxkbR%~+H*y8>wqhAotI8ygpqvwAh>8;>Nh^C;HA!X5P;UPcgOxWc1C{59e
zn0?bF;xVMb0DQ98{%k^;OSkiFo1(L8UO+BNqyG${Us0b?cvGvPKrWY1hVAIW+;n0n
z^_cS{7Ixo(ON>h=#3am3NFK8J^{mDy_|lsVVv$-c%n{#OQUug)z%)J0r#{?(_4j`Q
zVN?-K5o)>nJR|!go@u4J!j0$cGUJMwb*Nt0!@4yoaH=B5mina%WKee_lqE%8A=pMu
zWqNEj7Xriz@zD>ouBHq#D_hqd9t_8ct$nEfNvgffM-^R$RE}bG%d2n8h=fqNf>4?X
zse@=|(v$b5Qtz#4k%iS3ah7G|T*`lv>uOA@OKsoI%rZ0TwVSuB43p-W-xT{JTIl0p
zP8wxheI!e6<i(@YFAcPo9GqhydYXRx@25564;~#*YeqIuZxqp8Qy>#Hb41^u9cEcd
zdgki+_BjN|$Rd!L$do6L<xIZxMl3P`G^OTWs{25#Tg2U~tB|xucf!)Nvmk%DT0<-b
zv}i+Qx3Z6aT19L~NC+?^G7zyh7Yv$g5j-xqSyYFR9CWr7V0th~2bZe6!T8(tOTDl2
z5JtuZUu{|Ht@0CcYRgZqb!5VXv)eI!DsB|X_*kyda8nTMayUVEoK;J^Cj2TwE}`zZ
zLb3e!G-@*wzoJTo_b~~WRVjav8|Pn>{kDy2NF|Wus3m39D5~_fU?x0K4>+7!%&N1g
z3mw%o*WD;RrWKbn_*Y!K2z8Uy%BFFP`ZLWzBi%U7W&T^sWK}5=EvjD(uD>jf<f>9s
z82KZ2w$%CfyLE;JjX|Fp3ECj-T9sg&-mM@)(i78IPA=!PeMMaBK<IyxAe^w5vl!93
z%BG8-Y*<5wP{2T%4%){bxRivEaAFyKW99mkgbH8uK#VxUz;8lg+_kV`oOYCasajW~
z_`}8A-HOk<O8?xdh7z632)EHgQzckT(pbYUvT%#ak}Kw*U*pHWt)fK<$qYK%16q=T
zEK|u5X+_NmuAV@X#GQYd`1zycr{FY)@S-?^qI<vSvyukZ9VD?rp@pJxdV8&!ozT&E
z4fF&`YiK+LJ2jGE3W~eqYoz15D9z?z-Zh7eIxY23_b4Aq*F=Z;kAEfpuy$5mg+d_b
z`z}8ql0K~}Il&p=qpsI?>jR0P7u{qa@#!;$jUWH8KAzHXTd05QeEsxU^GBm<U|J6o
z{r#%$0i!WIl^0<Z6DAdfe!PA9Y(*W*UjU^^s9ud*7k8S3H5d)2eo4)B`OXTTGi_ll
z`zO%RaV7T`b9~m#&N)=#7>gZ%8R*7dG&)Wp;oT^uOR^kxo{P(ZnnSzJMT%l6Q-ngU
zlpca$yX<x9$$@|90=E%gYFl%9DT`N}a&a)HNAPDXWQ9`ZHCWf|K9)858nMxnCIJxV
z>}TgZ%u}0r=e0m}gKCr&Lr|U3FQitAN=Z)DlI2GiMl?l%W+9u&_3&Yt(ZwrIrswk;
z&n8FxnM>$Rn`LZIc2cySphH&2+`KelALV1AjgDhmZJU2|j~vb+O`TL|S30vUGsMlV
z%mkVZwR@de@1*LNIg^+v?iNmZMJcg)q-It})09DhIblZncX+xsp~DnsP10`RGb&m-
z23k&dta8yTM@qwFsn5@BRN|3)&ipmm+9$suhl&%d7`G<@BTS+GFJF)o;lmfrLI3;V
zuUm}}kAHs&Xn-j0DPtb?-ph`l+Wjq`T!21|(i}?Ln4h6<{H#Q;!xym3!>|5?uV$D-
z-`pXh41kW)qz6bTilZdxfi_=LT1yCZ2TsR?5FNr)RUs3^lBrq6zZ2-#ma4#9+`R+;
z9iTJt-vbP(q9_j6(`eBT&`86rO_+CI?VDg(-)n!wtMw|T!&f?>R{ibg^bXMChja*w
zePi{u`%79qpjyMQ_8FhP&#rG<r2@z22B$-g$495LkI#)xHHXJXr&7eHM$c!g4v(IX
zSUop772o!qfIokXnnZM1Mpc&&(n(cZw|pEmM-Ia#?1EP^+|lXzAIBuQ$O>g{WOH<K
zWnq^J7YzylHJ5>wCKLiUHI;EUCzpQL7p#Bq`rFsfFFbaMGjGDz@In9k_<H%z3+*7n
zF+cqN<0ahIN-*W*AeCW4!0^|<hCg2ZPC-lQIf2sS#sE~(6C6dV2V*=_;Zc;Hp4!SZ
z7R!JB`g0(6e)(x`ra%Ov7D&A6buKNjQ#g>MwEB3yUL~23h9U_Zf1$#)Ucc7NdHR1N
z{`q|MpqTP*wR*Q*y}@BKvPb<^Gh@Y~e!Jg&XPH*INBtg{*J`!??fy&dnB}HJ{Zd#a
z(Ew=uZnytZdgkq7)O1u<CIKq~KK&W%ukMufYsrv77Ck`%FkA+e#+Cp|LnEQ+&*?M#
z5BE?aP>V)#fKUqQMw_`b$Z#*=j8A{?j&{>8#_K-aA3r0)h;7zKJsCSV)LD~F6xxlQ
zAaO!*Paq~js1b!Ghz%oCBmRGnj1G2bQ(Mt<OvpB(H`pgS2!-5Fkhf?#(qdg$Et-Kh
zZj%=rJ)WI>txAHx=ip<<p~0kO@L!}6=&DyOIoUsHau^|1j2Mrf@2OdM<34{PF2(>Q
zrjZ-c7?}8rfF$GpNn?tU4FZ8RLEWPOjK!V`H3-I|0IEB`tZ0J-2EyktQ&FsXYX8^@
z{H`IjW@Ah}!*Qi(o9+Ef+7Gt|Z-7CVbBmUXQ`{iAW*P)VJ@PH2zn{roAz5;iBiQFv
zG;s^*=ZQXo2uw!1sMos)Sc-oyhxGom`%>I60J}iSBpKr1QN;_Pjmj3)U)xNUCW=(W
zQl_HRS}+a9AT$Hh%N8oIV^dIp<vwsi(4Z<DyX7ry{~2qD`dtAo9Od(TTjRZ-z>Abz
z1G<e!OfYRcDVzhbCu&V0T`lL6%#-ZA?#c%7Xv6$MkOiN1;2nqNp+tX2WpbV^jBAT^
zykgcbjOPUD)Ic-mm=t0~p}A<EY7v5+W)-d&R@#eJYP1J<<$j)5NDn#?OO4opSqNYU
zY9WXR+)|f5LoP(vo!Eu+=73&^l3l_t{gf&a%&tx?7>0E51jP`fdGe>CkfLn{j#-)F
zo#hwsk{PW!@e4tFfM0*qX?~$SIL|Kx@jSm!!1Me<5%>9pl*aMwJlg=zEcl*7VI0Hj
zU?_5FD;Y?+P``Gwq)4;tcg55qN-i6US<BIYA|>2;qjg~xt9u!3jdp0XiocyVTGE3K
zMq49xFj@lG!DtEMfzj5bPXq;2*nNTm+ME5LfGXJ)K|x2Qnu33VeVw8~0qx=m_8~~s
z#3QFL?q$NG6lYF!z#1pXn77R(-eL*SoXO6-WI3~PYmGsqmgIy?O5LPXY~oX1W-e(K
zdn%RhhH^TFM=|vSx!$plB3?y_GSfOIh#r((8JPL9PvcWLrj*-ycrV*d464ab1ix&m
zoRp?h*W|x+(OQ4I?k~VH1B#;V_=`#ZIZs-X+FjjQ$<ESt-%ogl=x=I7!FMyWb!jDJ
z^Qs?})%ynJ1KJH8FJyyu$|v={wM$Srrk#DXCyDM!Pu`+M-0W;ew<eCvx~&~^+^%qB
zS}{v2U2!!<7wcpg5hV+SdG!R|qV~FS^=u`6BkZDW8B2ffGl<w`-&Gb^@>{MeV9JVj
zRMA_#wi+mwJbIdAlx0r4-W*e-J}{ANjYp@I8`=wgI(n=DPlcnAN(kbKcr-<r1kvr2
zQ(bzVM2@?U8mKB!1!)D}T~S(5j5&wU7A2?X@v(Nq$gnD0%!s03?n}O#Q>nR3Gxtfu
zob%gBqndxKL^G@CjYM}|p?)h-*LiIbp>FSLIL$=@m{+~2r~<v7iRwO_C!z|k2M^VN
z9cZWqJYb=!a7{vIi6CXM>h^5#KuS<8v_tB1tw^)UM_Qq{WQtQAZP+iRf&~dlyR#nT
zsj#_9QdL#@0S?stI1uf{ThW|2zl0YrSOC;sGcJE*XHp}mm$Mq}#c5h2fE{>^BAzEU
zinz~gq%@(ndlzb>E!yWcQo1(Y72{Ujj9WX!&=5}>x1BdW7Ga8r<?1BTErm>FT1Z~M
zZ%A9oxQE^bUZY-b;G`Gl4ZH^IVc-O@gMkyo0|T#1*9P9OmoZJCXe;(7P^2^=7HZUM
zGn;?4J*nCV)bX4hf`i?yBh>zW)I|+|R%C6t>osrXb(8?+ov_(67jqN1)7oFf_mvH?
zYe!WCev0CA^(2hciW!WI$XOU{{wSn)7A8T=8tBIQ9DZLpU4vR|v<R5L{>F!Y;=`Y@
za5&oW-?8YATn``qJ^UjawmATHjE}&<g4KT>Fdxt)alSdh&JdtH<JvzD@nFliV(5eR
zVNpFiP~6=EMPVN+5+pawI$4Ce$;GU15|4H&-b+`MvuwAVwGZbyL_FJ(K2vgJ?rfzt
zS&Ys7-It^Cl7WJ5xT(FpadUf}U|`KP<F1@Wd(s=@Zs#!tH=l+$8oBlsA4tMt{2+g9
zZ%KsQyhL!tJfA%Q6l+wG3c)V2#>J|YmO@u1k(~6QxSG3lATA7Z4fu=1Wst#ic5%wi
zX|U|x#9rbZ))g$tE=w_-?i|l;nkktup7#8lTy=13K~cx&;hvwiq)7Xq<eMdQAbIWB
zRcZ<XJx6ucC#z|U%*SM-s}4oQIRbz1assb!<?;=z68#()9?G(@&j*rJcHCZZjVv|J
zLgbu1hp5N5QhwC}HxvhnW?8;PEjD|y3_%@-2W8pZ>jP=Z$;)?leVuW0dj&8j)kUrk
zwE({_U0>c-B4Q|hkDrOlT*YCVVRynCgX>*)!o%@nUBuc)`q3<x19|f)8n}PP+OQG<
zbN-}QoJrFW{dwdalo7-|B+A#5D+mXmBt$piqKi<`MZ}i`s7KvHvpDv}9Fp20)VM^<
z(&MzM$bMUzzugnBcEZ$QeSFWt!qd8Vj`W`1!|<3$+q-@sQa6Po*ND^^CY3G^tB#)F
zz25jbYO=5R!57=G-1Gz$)zyDtVmUV=*O(pnsL#e9inv)62O_pp<k8!C^v2EYHIsM0
z#Cb@gnG$AY-t?Asz2WG6Y|j^paQVSmReOI9<OCc)FPbLE@V*V0!_`G#luf0~jk(Ho
zn%T_4Dw=TIka<=#pzsol+tPsW(e(d=wBIg}13_yPloMxCOHV^QmQ8;Pxl`wwvS~DJ
z_rR-vWH?7CikH_+!P4}qo@&kUKKZP)G8srWhYT?UpMJF-h)ust!C1tPE1c6yh0;3a
z^xuf@ioF;%zkJ9<)w-uU<7<Q7UcnW)^JiZXBwTnIi(Mo}4fAtewD}!D!b>J)4g{9I
zD4N%;M~+vZf{H0rs^5Qi#4Lu`O(|XxCg=Npp)JKLz%cFo89RI$;=z__2-~SH@mM14
z5S=Hxvqneqj8l;OCLSN3+gN7pxqFOIo4a)+Ey)1nsd0Q3;z89#ShWw^kxvEp*upv$
z7gHFK<;JPjt4up%_m+wamK2#6U9r@A%cy2vBF?G4#LhFP-lr>)tsLnD)8Z<Mbe0*f
zw;8he?PfLb#=CIRyMF4z;~q|Zz{&IzMfX38Q|FgK%`OucGc`6dATS_rVrmLJJTFXT
zZfA68AUHLXP#-6kkm47ge}46C_qf1HiL|Gq{(iUky%>`6NFbQZ_t!-pU`ODhH6iL%
zZ{B{I=f&q9VULUu^O62;&p`$c(S{89^bO4q9emVam!kPoLSOsap!sFK>J7$#BVYk`
z`U28>z-TO_m@4c6LR<pmbU8%hlX(k1zK==>Nrtc`?ILN~q4y#ue`%-bX2mJCKvM9U
zcBCn$S%a)SKXkp+bI}0_tEIlwoI<Ha`FvHIrjlr=+-n-8Zg#%L+>MtM(q6-nK6R;d
z#QG$CdOfESAJJ0?f~8vvDw@<m!39uLelMtWw8`bwTKh{H=lP9tP3ASC7b663XVQxq
zP)_91ZI3opuFZtje=65xS|f7V1QFd#aybW>`Z7U&m|l(CCqLr|cZv|1XX0)xkh3S;
zk8{P><r1qNsg8#lb8b8~>NT0yh+es>MVU!2-Qy{^f1=l9USoPO#1yV7(w9Os`Vxip
zX_7A)8lVQZC_&w#?YDvK4tbpm^4=Ya>&h{2x)R-uWbenDf6fsK^;cq7rc2zqNAUZZ
zUFk1{ydsL-N0l>wRML|#B5AiBoSk)u!WM!X8btDtTyGEupe+kP-;%I`>nm5ISyhr$
zd1h-8izeMOvrg-O-d@b$%9riov7K%~)(?@@ho5@meY96PG>pb|=fH7uPP0q=wXy+6
zd`o{Vx_RX9e~_E|Yat8zNpa26y%?hL$@D9QBhO>7MfZge?Eh)7r!#+FQBDY=%jCw2
zrsCH)If!&9Yv~GiT^x3sR?|6+XcZEO4>M^MZP^LWYSqYl^7D=G_7ObBnRsh`j7d&b
z-8G*Sn?|iBa~jb~0(ed{Y2|gI3ukkr#S$Kmm(RCGH?1af8qtcq8nJmCHESVH;qY$>
z*EUry-|na<(;8FDK|s2uDqjrIW<)ZqpuL5m9`=6$m~2r;mjM?I69F}sQR^=gw+H4I
z&k6~J8P@y{SQYD1mjM?I69P3jm+{2`6PE|^7s~;@x6|<#m<$8i%<h*lqAU{xGc-0c
zmjN~cE&(u?5u_9+m)rgq5P!%Sz7&T5xum`x&;x=v1=<=$pg3DcjV!s+8cFWg_ZhBa
zOI|yc<d@5Rusypw91dq5&I{?l(Hmz1OEDqfZxln>E0K&(mB=QLvP3pP^i6U`94X<A
z6eVIY(X+mZCUJTOqL?fyFa?uSCD@STIxHLVE*meH%yZONN@dK%n17r&B&CWWVI7(s
zC{g(tLSlB9AY(+L`Za=rC{PJhsbeuk4n-*fm37+UjVIQxB^vLmcEy3xg>hlyLs1Qi
z7|$VM4y?_qwm}@mMp6T~sCH3%fY~S3IZ825-s@L<a3o5Bne=NlRkf)QsNnRgDjteF
z1Er$W^iqdOR0>K)u74Dy#H3OPzfxApLrU)7LA9dH;>ECIh>}2tsgCs?RKz;O2ZJee
z4xbD&Eac&DmY@TC2Bn@Bk9Ykd@c~GvLKu{qMFOat^&RYrCF_U)05}&gMTiEAs$UbR
zRHbACW$U{DN>mE`DO}ZKd=<r$02Elas>fm^^?Z<*6k7t1Nq-JH2V=IJs+GKnPOYi-
zid0PiTSrphh&Z6J>I}?8uJVA7sD1}UCaP+nvWZDSA`v69FPK9TsHj@uS1i>oz(;cZ
zJ2HvX$7-<T)pr7LrchUjy_15%R3f&48WvPCDJs+i`zWQ9F$av3iVHwg00E<U<to)?
zSYCEj8n#k)Re$%COkU2KL9uF*2vk<Na7pG~6WwzZpZ)k_^Xdqj(qU)6Y5w`Ie?w{2
z)=V!ZljG;l!C;z~v&n3Jcs@F5O;A45ydHI3JD(cgKmPSg_w!*l>RLnni-Xy;gXn{x
z8yOo>g&*kl!fxpwU?VENpZ1{Wro}F=4_F5g@P&ux&wrcO^V!K^+Z{EpUmY}W+b><O
zRP*-hc?&g8TWB}!+G)4Y*b=LVH|=6}F+XV+eI(LXzqTL7qn~D9j&#OI@JR8;7(d5I
zaIx#ByR?T~+JjuQS)H_Dlp$9Fm*`HlB|fqF)-1MR%U#WvXlTB&kNMJ*!|6V&PJ>k^
z*?!f{i+|~K)(cu<_SsQQWvY2iWgw0(XRq9*`Dr%)(9Wx1cijA~`K5V@2dwBAikx&u
zo}EP=@XS_IfzQr5q%kjSiTU_hb9nK-Q+a<MPd_y;E~TGz<Jq)1Z2s})7ybF+W7nN8
z{?at3<L=|ddwViFYyLBDKg`FU+D~&G4ZFk8_J6;VKSA*|D{ZT14z?f$trTU-Hu*gi
z?Jm06d_0;gn(1sh8Bg2My!qcnJO4VKp1%9oE*9g_bg`jk1zOx@pKfsV`_$)RkyQVf
z`!t6EVl(Z7*t~|=>}~++hWBnp1N9FT4S|Z)G=}_QHMujMi{{>#Ef#*<yf6@d*!(<u
zJAZ3lnI9Hyg~6XKDl9nm%46Bz9zd=$h<7%^12HCBFzFVYoUCj(|2pRiChWwzn6}{b
zL9CXXj{qnXdsJwd9Z?5(=aztqTAWhv`h%pksHEi=)`xrp2odfeZ3#jx;fr4hK#}Ug
zN<e*xH{|CHeBUEkmm!MFUMMc{3dQ9CP=D;Bx2T2<1qLu{U1>dvTl5+>XBnWF$Pu=Z
zHby{(OAfW<2AYfCCQ6kds0d*{<;TyS4-}Zh4P2@n&D|~)CmSx+&gsgfd6`N}=?QTU
z;pD`&V9T&CHli5}S!@LU7zJGLG<0%Ay^X+;5_>{N6J5Ua@%~D9Oez{DlTB!l0)Lbu
zRl+A5VjL(DqPE4kWN`_~rV#!8m0*C)p)){>Vm<0T^k&v+FhR7Wn=GN-6?#WBA;bja
z$5__kt_|?xSAVyKq+KCPegIi=+e7wh)ypt^`zHV13SYE;!=>|HOP8lD_jg_BGWBgN
zH1D<0@>D*4D0us!k3H`bkpo>uWPd#^>I1P4_iAo3Sv_A`^)`iUbl+od)9ZQJ+U%<5
zO|1=-J)g8^u=Qv@?!I1cZ1j-sn+5PX!{yn7k2V1&cXLwCw#c_+u1aMJ-Z~`SEjW?G
zNaS~`Rj!q}8+Gz+HSeywDac?qg|OFdx@x@iVD#SA42wD;*0wwP1g8Wm^?$SwrE8=c
z3W-R8VksFlehw&H{YDB8yhX*j$<-rEpol|3em7@u1iYljz{~DHxCX0-%CO7>ys+N~
zFPyLN!n9Y=?fr=I4vlJxdegyHTPI=EpcHlwyPrRQwx^@BMGd?Y_+-&F`aBuGe|I)%
z@aJQPzpGc>d^BC0ot#^JyMGz%mX^Lf*sU>K0<A#(C|hun*0HJlVVWozHc_%y6D4sC
zG^^f8+<qxLI9oJlC$Ua%i(G(F^<a|sL(hlV*v{ss4XJZ-49&|IzgoSU^ZO-hbY@yp
zpgFO6xgzMGoap_aXq0&z{S<JrO>e)+waV6H_5Bkz2-h&WWti0K!+*fTGoqbwU52<$
z`&8k}W3BJD-H5{t9BH40M*lmsl)J7r4Og4CT5W$_auu$!y9!tR&02pux+Kzy$oEQr
zoV?Y4FG$Yn^*?8CL!jN`Lp!yf$DhXMy2ctP^hxb9K6y7g@5X22-$pw0rf9ZfaYxKH
zpw$Y_+guN7bSe5j1ApEk?A>;%ZiD6fpAL}*e!$m@o`j?OajapmR+drX<4(@3u@X)!
z3H?eijHK_pMyg~{x7<28zHXGW>5l7!b%s@ZjvI&hw{Z!<9IkwWbJZVl$X4U<7My4_
zX;|9aFni!qPpa?cBz4zA-A-&|k;lBMSig{Xs2(m4u6rfRUL5git%r-N@~Xf2)~~c8
zR)-}ytcd<f`2P$YdPkQ57Y!2uHJ7o`Iun<*-z_PZS_c@Se}BFH`uXz<Pt)hjY5&vT
z#{WJ(U;h0vE(W<{8rAgMj~D+yfnZ?eBn)Su#q{Str{7=xI+87eae9ZL|3kn2{9_tg
z{+9UZIPsHfjo9&13+A|)v~~>qqsUK1yhGxRGQ-N&2P>47V%AL>Ed!cyCQOW1#7Z>j
z_uDXlHbhCle<aEPT?-BH)5VILOHqa?Fhw*7HFZwQA+S((BbK)<Fk#&IHc#_o2I4Hj
zpD?ekrCqr76x5887b@+?4?iP|Jr~}Z`JG<BY3>CSPZb0-zE!4H#6+X?_fOQRQVdip
z#h$-m)+%7ut|PYu{w|9O2=tA!b$L~DOQYQ6)-@DCe*!~;<0yjHH3@#)niTbLu(T#p
z^uyMq91~p5)+938UM|!}S|P-%BIspjQD(s*n@j9DN$k3+UNs6ybwM*_?2$G>lEEW`
zt|c-QZk4jcOa~L?uH}|7$Ri2Guvj0iv%Yx4AGB!DxS~(lPKnZTNpJWIibf_4E?Q}m
zF20o>f2kF5q#f;FT)r}89?3?8@ob~bN$0VwA<i$qE7QBt3XJosD+}~ZdK+ifBG1C5
zIi^_zsyfK*z2IM!vJ4RiMpG-o;L=)qnZF8jlGUCF_mI%VFJ(axmm8MK>*ADLX2}6+
zB&A?V&*m1Hy=ie06q7JxQ3fZkC#(h2dTd(Lf8u(>r~~WJNv0JP;=0+Jt}rLGFG$G$
z1>3V+p(Pj5swbE)nkX|8ACuUDh-Y0e2<Mi4fuZ-Z@KOwfNzkk^{TO+-$mp7}=MH&y
zmJvH=6m)C;l9PjJUFYm1HS1e6pRYk4#i9zJV@B(vR3PFQ#;aiL)P!-S$}-E%wV7^3
zf4rZ4<jiaVtsrg}rQ`G$=&qjOexTIM+Twl$RI=}E(^|y-Q3R*BRZU3#vV+y>F9mgi
z?xMLbMvY1XQz*u$k^E;0j&snMDb(?77|zn~1(Fm5ITHfWUw-A&Kj`7l)ES(e{u8?X
zD8umSU(?@xRJuZ#6gm(lg|_+@VLnObe+ZOFwvs?_!6w{IV&lvx@U_z1CQFDq9O;Ir
z`ihAcf;?52l%b$f;8nYl_i3};EJV!>8awyT+HFHM5u~z>vrzp<m*ia}6We*3ZOj>3
zP3zSPRh|#AEG9ElEJ5XdC&mTW-2kisOG*y*A(6Y6#6~S6vKrhTEgRK&0F6L$zeA=>
zk$(?qbg@HVUWIlcF7YWZlVb#%#jUi`^Sm!0=GNOX6L*Z5NXzk5`p;(ZFD*@E5?it#
z6~sSBTc<ucNj)!3tIY~hD2JW)Rf-gSsZ(_qytIn<B$;=n=|D?HPUUEDHQ6O%{mm@h
zO<NJS3nj=6cg0=x6V0j=mUu~%4q1)-Cx2{2T}!(WQTz@x=s;QW<pfxCFow=Bn?|%Q
zK4TG*sQUhCrA*r}(hymW*|c$9OQzG`->bDN7HB=EI0J(VW&qn9;_aa83QHX$wvN<-
zQ@_O58Z~(if85Vm=OZ3Fjz_k-cRB8%NIne3h9kxI-4tofOgJ;6s24dXxe7kc(0^O;
zyLlr?en=%pV8u)y?jkCY;*=S{0+tse9eGzXcpeZ_aNhRlF4es$R$3NUG)$A%GEm(h
zt~wcb-lPD|_d}HOCByjQ@)$-bhcZ%C%NT)W%9<Gu9>B<ue<QkbMBt>yBpIu2BX!r!
z2|{oZ28!$WK=LbVYDO-grgk8fd4D_1=zW>DN;0Fw*gG1#WK{X%SqQBdY2Y}@ok<#L
z2~U{KRUMHfacnLY3rUhKKwPajaLItrnDD?(09-RfGo!eFCjhS?VkSISc7j}e8b$3q
z&y-}v)0XZfe}F3<1IgZXRXQ(ujQ0fVB$Oa$6<%pD<ehc}>p<?E<MQ}m9e=>55Y!`)
z)7>OC&P<X8AxVEquntnw++|yFJmw0)`f=p`Hzc<O^ZNl>0iG~2P9F@s@g~JZ(_(9c
zTtl0t7$!pTE1mfe!A#{;2rorNol3WtnlhP}*JL_>Qi}LpQU==mwadZGM(*$C(hB#<
zXSZ;4j!gM#0eLgz;){75gnvwK;=aTtMJtkF?n<*h7~~Y>OVjwiG8bZ-)b^$0vS~`F
z#XD$Av0y7g6^oR}$oQUHx<NX?Z?yjRM$~q5_cm(Y(Xhro$5={7L42*4d>P9f#&SPE
zE5IAeoN!M9-byB$0u<D<4{ZrL<=MUo8BWTrRjliwhH${=z37wf>VKIV>wq;8o~Mzh
znT^2gX5b1i+OWenv5bqMds8iIH<)~O!!tt5e3+&3<ytzmKMIm9-wuB6UH1w*k@&&I
zllZAQZBK5gbklF?EmK4;^g((<R#6{Vr;-~1@r__N{KjFrs_Q^2Z-YI2>$K$3yqYtS
z1@vwuSLvQM!oC}y6@T1SbkvW<9VudQhKM@`wQ8G7Y><i-YLP)UGN#Y1lw)4W#w`Oj
zW3K81KsEjjj9H>uCA>A(MzD4>bp<Sl&h>M^0?WXNwq}y^Tt&o9@3#%;!Tf4a6xA#2
zYO_t+pr_Zoi%mIH#yM@20Jjmb-JD%POh$+NIfzNal#rc%Wq<j#jX!|RG4ZCnsAPW-
zGaQ-?ePcyDoui$D+}OcAL$-&EvM@D=f5!1P0=S#eD*%e<n?Gk?6q+fl2J+IL%T8|D
zxvE<B|8iv9_z_v=Rc1K$*>J&Y$yTNRG26pSzHempZkDXDnn!Q=Il97;siq<<()Avr
zWdB8qPk%PT=YM9_tLg8*Y^nX__d5i2_bP95&(|xx&-xTW6;3gPuL$*#2z5U|E5O_6
zTptWPG9)SW=L^vEm6c+*OHi|LF0<RZ`=E5;NbT+3v_hRf%Te)zfz;GIt#!cuspM+W
zhpdk--TD4p%_wL5LmfkJKZCU&j}@GCaHgN5-$pVm?0=VjKce4z@mYZ#qg#F~?jRXT
z44?UFboH!7R)@e4ll*EEbFm@o`;jB7qCZRScj*T6Za%B~Hgv<9|4(0migY)N<HBm-
zSI5n(I9F<=o81WQZVXniRndVz7F%hVB+HGQ)jIGg2G-Tqg)m#BQp`LjQJ9Pk7J3$2
z6(;9`jz)@<M`{hYdR1EP0r#z0#a?iCgRlZFMCblF2Ad}DkAHj_><0|?Zh%&Rmoe0M
z4)CPi#>zI8T54~m9t|GD#i0nQHYS_*moO0+9)DXq(`<#+&b0oj<n!-G6D-h$m)Q=!
ztx@D3-6699$+m#dm}<31aU15lv01^K$57(n!nZ#FMwoNWtX5kyDf68CGRLkJ%(^O2
zq-7gC$DPQ{Xf5Q;8YHoCS@aUEjY#h1_-!Oz3>KcEy#9<d$FF_ufqiW^JhxG|F;sX8
z>Pl8Ir*wO|aOFsEKF-2B-?G)4@p{|Fyxpt3%{>1N?sHgYfI0gWxj&*XJHfe)xHjR_
z(Fe|b-b-QINmrtZLLTwG+<5;lS+Un+mjM?I69P6cm(f@%6qgG#JSTtcTf1`XxDn0u
z74riZ#-s77QccxpB;BgbyjHfgk?nQj8+a4{z8gRTXpkI|mzpUj3Mh_9umSYxCxDix
zf1Tv%`_s$E(_h{VoV*BDd42l$<D~F~#J@`7LuXH3DdCJdef&86_AI4#c)j2SQW}Go
z$KL_(_WaGLfGyMoFLi(Z=0Q<=qw!Ms`<ni*?ef>Qyw#UX!Ng~xEb&S$y^<!;#C9qs
z+UMUtetD8K{OMSo2Z1AMW5uuR_52N##{b|Q|I_7yH@t8(7Dok_)&hIBzW!%%W`Hw=
zFr@%ETKSGojZ5t=LmNkQRHKug1m31mZvCY>8b{o6yj>r0eVTvy+xxo`J?z2SN|T=J
zGDAC|SpHMyh2F!;oKx$Bqcx1KEd9VcpJN_n;!v26Aj^YH+)8qgpI+*`NB&iAE#i={
zpJDE_kU>|>GwmB=EpoR$*3`;ZS!lAF-(?ODR6Mn_j0#B<wKCKfy-BUqYms3QT@vGT
z`#$;a?c?b`^ND{NBUDtUzyEkjFTf~aaNf#^Xx!<~f1dt$`s<V`SM#7y&82hJ4r;C>
z>OLU#D8Mz<=`i43ud$WhFR>#%kLlT2J$HU*ve;a{6f&7VB*CeTye{#FX0m0ioHW8n
zQ>Rfhu}FBqwMrY|W6wH|4f|FSB|Y2{p>%(xh6e$(Xq|sRJqkc%t#HAw>H(XCgo6ku
zyeoEJoT?61xCqm!Y6l`nU*}p$RMI$+++V=o3+w|gj53MlZ@<gaKk(%**x1By`fqOf
zqxB<C|33XKF>`}(Kz1g=Ae6Rigyl(#7$TAFBrIx;g!uC$c8)BmymRYh5&vkXbXNsN
z32hA$cLINGQx(*bJ6&1Z??>+cLy|P0qcg~P!~HNo8^CKVG%Ct{f%g_4;-Iw^fXYe0
z^SHyw?CVpc<&xK?$Sd9R+Z=z$`YU~zY)p0n6o*XxU}>pcj|I|Oa089%jbn{tc5?qP
zmo~VcGz7w(xer$85Rf<sRK3dmWP?-8h&GqFpQC?POPo+-nJJ<qo*};^`Z4WUj7M>4
zij=CCO))CtvS<D<(wDaQ#?rG19iff|1p^h|E2o>K10_H~`n?mi!`$6KO-f{Ne2}rk
zxWan>Wh_5qEDr;;4m^Y?jCl<15VdeBw5ZCEMkH;sWcfmgqA)D{rgTnMOVms1b)`s|
zH*bIDFDFN@O`RN9E^s`BBlV*ck0%ti8*VvJ^ZRf)%)bp-d=QwwKM0qggu@SOoX`CW
z<rkR|PS?b}^QE~fYFv=w&aY)L?hro@Pke)+&SU<dAH+~E#bWk1I&{vHnmA>pa3$ZX
zl0M-aP!M5iszq+YS9xEeRh{67-Zm}0G_HSZcYF==VJ0`}UUs5>7@ZB&Z3)xwi@G(!
zM~}p#@Su3QoC7h_7pN9E6bffLw=<U=TM-Di5Swuq1_=Nb?(pQbqNca|+KJX-u5O?O
zCANMHS`dV1OI@@EV_%qz`!Q^rpl9Y6Ls9jha9H0N*FG3M_N}GRQxi%ryCop*MC^Ys
zXEzWtDIT{ELd<v(^spQSRel4D2W~9ff?im*wtYp-l}B@snISyVB?~T^FvTtq)Bbh>
zc$m=}0O}H&KgKO;CxUCaG7lY+0Ozpm+EuN_e>E{?aYU7Qs0^>(jUqc$wVB#v^3@)`
z*ZWRZA7;r0t5u1HAEPVEiYRTfZi;`d*ODZsZ6SFbgfGsl7i*ukmwNkbh4u>S;aP5Q
zPnMYVF@h>VL~vgT>L(J^;{dG#Z(@qA?+m*YR(d3DL<}uI9YcVY(sU&;H7Ok&_n4A)
z-{EO33D#R<K$y1n^72!T+}b1dwx!Ik<6A9ZF@o%``sn#yMyOBoE{0>f78`#kA4OvW
zV_zcahbXrRMo1mMl=~CPeH@?-;9UvI?+d)q!s;YdTFAcYXsR|0>DYB4sP5`BWNmR4
zd$(p+Y6c~ho>Hr8|7>S~1@rQ){@-Gt^{lV=8(6=WS-<<BJq*DHw6;X=kI`wT1pfZg
z>Cfo&VSqM(hZ6Na26!ulvYLOdwY9Aoa=1(B#AJ^Qb}mL9pUMV{Oo`SX!y+8KiY;xP
zoncs9ifpH){4`V5+|JnYd$cn=qS10#)(Q+3Go&A$88}#@DBn26DUu7zBzDOLXNvnU
zKa9-==DOq<citC`!UAHhgV1JorcD@Jh+{7c7?6Z4Z>S8r1fxIo#m9e*j&eq8VO?e1
z$hqnKUVZ|}!yMm0QkAsfG1{xV(B6J|w$FIB!vJjnFH0uz7~rK7Mmc^2tz>-TN>^Xe
zZ$}>FT7cOuzkd(z_&nG7r(+JtAHqHzt-y%pOK(1*H;3`rfIT4jpgwa%@TW=aI~A!z
z5Me#{xpgXXdi-)nF@S%zRs`8JoFC?<1GVe`IUTrNJBSKP$FCcmvWTVJ^Ca{G_TKRW
z+YryCPqqYB-loK~P~LlBo?5e2`rI<qu<-sz!`n=*!wBl0EIX=P_eoQ6!wQ#c5Udst
z&g--{<;@IBPsi~AmP!@CvM{zT#8KPDqkL3;kiUakeo)%C9tVHAb?4r&-AK>-SU#Ds
zF*Y;e(Z`fW9B1<l1HBbA7v{cD0Flc@d|oAONADU}0!i&l-vMgJDg*zvkB$mkO5b(o
z=(>!v#%zEcVOrs_@%ovC$nT9Nxvsfga8dwfmkWIi)dQ>50lYMTdRw+am>vinXT5Qb
z(hN4wUzL&+8xnutQ(01y0Q$C~7V49B5YsVjR}(AaTLF<;<(Tmb?ez?mdi3%(IyLR5
z!<?&aR=Ef+Vwj<3T!gMwjc=Ao^T-cNGK*HJ$xP=%N2GXdZIh2eSxJ7(V65|J@@TWC
zzka_hX;qq2VC=fpl2kxO%a>*Yd1LxJZX^2v*&K$DLhyfDvztPxoWN*eGA{Xkx0!3L
zM2u={=6dI8!DSBQS(}-f?&5#}DU@4c0g8iszPl)xbaqGkdoe_}U9Kt$=enHqlZ*go
zg|g~4BY^p);-8rBw~<E^>4}ru<!@SIz}vS7S`QQbtX8kt3&l<%w7A62JSC5EvW=WZ
zd26;#W2Ao#sqAMQ<HKvqR?VN$NqXXpaj_frAyMrWjD@NVr-%e}ih}3n2|E+k^rD(e
zubxOzz;h{>&4dj_c@U<BYWa!rXC>n^Gi7wzXJdYCBvuM9+cLAMkPX3QTiNui9^lF=
zCR@$iP!9<g<`$=cbg1=qxre~&2a2E-JAf}tNSS|%b;04?MI3Cl1&eQf?bW^-(1j{D
zouPhT3GQk2H!nrJFV+xipYpRfrAoMeYP<o^(dyZFE6*=1+^R=td%fb2Q1SXzN;-5q
z2^N-XR{_=ol9ZQ5GoSast(zuKTH(DKzb=)<w2i;SYfxTz9d7VC>)4uP8!fzxHzgZ(
zNp^o}#U&eBI#=6X#-uaP=$+E8qPE<}+zgz!T7%`^>uS<WmEH6vQWa27y!zN1k2bVl
z0s2vQ5A9E9MY2O2Q|)VaVO`s6n#fXIl2BU>_Vt$INvBM@gjf1QeHpXM;D&_0@xS-h
zPOrFIHQu!vp6D#ps4dSU?>$+TtjR@3I=4Z5^hUpyw>e@lXJz5}W|khEh_da9=qq0+
z;ENF>LUXH8L-HoWWO<esCfe@C+qolt8Fo#$Uf28E$J74+&~HV=mjM?I69P6dm!UW+
z69F`rVRQs1mtP$isek<3-F)}(^#xDja~3T9N+12VR}UAzT=dFVEts%kdiQui53~|Y
zc{wQ${q^+g-=_B$Z~9dWFAfrB=s*1S>xZcud7I1`ILyhdHKsAA;E{*5fhl@4o72E_
z1}4%lje&{uZKL3mdoNa3Qx<}Yel_`WUNM?a%`2TC_O8svGk*nRBrZ`_a4=wFuz3bw
zj5|$k&EZU@Us2iQDE7<AtR2(JeMy<y73F=kYp}W2^Qz>&nx5UJaWju=opf7sE;wuN
zs*sWe=g3umnK1G76lY@r6}*}LJ;h-rq|pJ(Ss|RIWflIB$u6G!^TYc;z58A;ko`te
z;qOcu5&ZNxe}4^uY3Hn*KF;3-7tBio$pj1742==RoD4WInRFR>H(hAym{7*o!yjoF
z3wtePO&^b%i&Ha~Wvoa-i<Z&M^J6`jMO&Ce&59)SjaEst^cD1(ioQlwtM7R#sQx}(
zyq_-KB-<J*9Zm6kUN~6%TeGcTS~w+Y`{p69m~cX>Y=4*F;0+t2wZ3IgGV85Y(9ko-
z1%4~s7VD>xO7A>`wekviD1Y|j8+Pb)#Yk~g!%PrUfJ@rws3KW+w(#=%yT?TL0aUN6
z&v~6|Sb}L|r)zjIQ(DtnF_s#?nwrw|JUgGrd>R9t@p#%Yr?qp|EdrKrrb?t)x9df@
zH>Vqf1%D|{ut%2a*bWF&%+ppp4pVfRZhT<YmB>UXS>L7T6w|GtOyg3vaZ3cG89K(R
zGArjJpK0y$YQ)tmMmw7uNohW=$Ww%q@JTwB@$4w0<_LZ?`gYM=D#;9#>owsit|pBL
zD#Qrxmfeoee9|W*eWJWgbkCd=qL1aF=R|-7v44cTh~o7jYp&Dkd1bN>MR(%L@AIrK
z?`He>8{x<{cYWUJ_a*MuEN{v}M|-U$kPGBpu}}n1heW;ZDac!7^PZaa)uhZA^OUT8
zA?j<JZ=%>GXSu`CypT0bC$RHX=a3O(kzmpOS~~ODdrmuv&zG|v^NK`WX1tpBB!0im
z6o1%z9wF)Njd9PjH2V4C-_Zwb_VZ*hpW=-xvy<%lRfFiV9NoE0qrBKFamuTa|JdR`
zH}PRMd4`NGFEg_%iyqRV(MV-ZQebxKGMXGdebec=8>-;fH`f~G>;k$%cByn|NTS<z
zT{Uk<1Ey}SMRHE@at$;MaDEm9_}DvBaDP5W_~CeLfsgEe7w1dMw8lx(NbyE~3H)g+
zlhXDper9aEwp>(a+;VvwO*Ra)olC2<wVF$FUYt|u&r>~%Bm1MMF<}?+Muf}ZXxf1c
zB!Dp|B3@$v=9~{1@uCcYd8I0N&|=gCOvr>~pn<z(v_ZQIaY3tvK{b?R_#eZ%7Jmm9
z+vg?n1nLRr0gBGUx=1p?1#MdZyFU7VigmSRHiR#Nbq$aRqTh*FH_xsy-XZ3N6LD_E
zZtZL?ik*&gft-yop9OR^&OA2cM4+o3j+68S@N4Q2QQ)Ff9>c1f)3I`@OWwvd88Je=
z917DK3vkIJ_jPHXlCkLWSDd+)#DA5`0X9Rc8GdK!BEJkFYzo#nZ;J$}W<72f;1FfL
zD-zlYn9Zx2%w1J-6&RZ}T5NROfU4HcvXpXQ=5c7GYcT;riRa3fz>H85xSaFo>XWJr
z(7{A--tBWvOR-G^N#v_l^pSn3qB_$32zkh;WkMV>K{r!*)2r1o0#A_URe#SDP_mxA
zattenUpt1&ky^`<*SR@#t&L_@_&Ruatl+F#>)}%LL}DuE#?*i~ou1l96R1_%H~U*b
zmvOmK`0@bUC7bg)mqa#|Ql>60)0$RN1#Kl#R?~(<%yyw!^YU;t6oZV!eJ@7b)=;3Q
z&JAf3DxB}CU@$+mT_xd%;(to18Rd&w+P@5?Mc!pviC)^Jif^TDk7pS@MBS2PW>qqe
z1L!)@{MAqyZ-60eOY-~GkW0|9ang{;As0o>LTAs}VI?#Z{K(PfkeHl!cDRhYOW~A0
zLv6kI)K=AP>ggIg+-n70ZBVV?-{P2Ep=@%OH%SH5-7CaeKx^MR>3=03YOMv}kqy1p
zb3ej$Rdv!#wY?ghx<RwA548izW?s2t?ZytEJp!FLXa6bs1C*V!{{}9jP!n~QNMX1S
zM=AgcSN*~gj=Hln3`2mkK=jR8dtrz!4*U`OQx0}=&(sFk5O?+HZv-WU3Rbtm*IT*n
zq7v1#Lk-iAeFf*15`WYHmg7iZ9CN+JzJ<GvLyiR%65NrfV@aK2M)gESj9EGCPj8h^
zW1SEXad2Cbv3FBI^t};xq8mrq(pj!K=M7|?oghWNlAt#t7WD(&rg7JmR+?HHeqHJH
z{ZLfwnvUd~@y;u&khMF1c{XNaJ>#m|LGj(#xXvK%s)2EvB!6;-aW#@s0JBD#-=AGw
z3u|}$N-KqYc9vghEfF=(@++V&hRy6%--?<}Q}JBC08oNYEJ$D;9@cWqZNB-edZsl1
z=c}2?EM-chD(j~*Y$>q#R=9LIT+VV;ow3}9i=Qv64q7R^Hnmxm%U7+}M7LI~37$jz
ztjVs-<KnQ)I)BvmY3$qUnTx)~BB5%vn)RKxlDXy8kET~2mZo#tXUuSZO==lzg}>7p
zIzOAg33aEJ)Rri^a-qZF%}S{spGNWuBa5fGUAY@?1?JnGR>G<)Up!l7-=JB#(a{z;
zj<_Tb5xj}yvs^q=40ly;+W5$Om-~a_z%kmbZae~nbAP5mDxyHi23!_!Z7T7~m7;Hx
z)ND9jBJ82*@;|v38vsHfo+60|<eyN4A`wWlOpMakVxID*C6=gmkm>|yNN`>0V%JG5
z^TD(vCC`~v5}Xv0x*>U+x#$TA!3!mOfTK8-Fp_aG*XEZ$@04MZ30R;q#O3!_L+w4F
z4XmmpM}JP8+Z2}%SrtDdK^EiEwe`prv_ZxWqLFKiDCri~gjSQU#;AZgBY0F({4i8R
z2@~rUQX~X(N?}Di4~Kz5s9m(LuSrD;b16w~ZxO6Uo$P9b%iHlDAqyKAngF6TX^(!~
z$C60g4VOKOBbFP{lI)orB%<T;!%!F69c=5mDu3&sl1dWSF_ClOd>Hdmbdwy;v70gi
z^^b3v{kVf*9V*L?<i#xCxU3-Dq!3{m%BM35j2|w4d%QZOnMSpZSkRg}s_{#1$vGsq
zx&4wgOYU*2*RZ)O&;;-?5;`}SE00J-pMptm>S!*og45$@u22zSK`9Lrsi9dQH8lrU
z;(sg^uUyB59NhuWwZZsAYL4h6TQ1L@K7IP^(>SCfd}u7$uzDO9lGs$PTXQtU@kUsu
z-U<n-0MLBL$ZzEiZK7%^v0wgmhX>p(Kp7EOf4a6Zhealn4aJL{(mQx&I&atuF%-yf
zR;gX8^5)Db?P!YhD7ocHk%0%qZPg0$X@4iy0hYk{QU>ONI4PLGG8+5SaM=mMboHw?
zEVPmMP((WJTETV_9}bCuF7S57gZm?i&-sXe>J<t&+47gztPzNhwr_lHHI+G&LbXhG
zPYd;xy+R-cI7{zJ?;I8cJVvk6N`zix#VN;yNCIG-z#&jto^oDdzZ(FhA=x=Flz(6)
zz(QkzF}|P-YjIE{PK6Y!=OxV$bp59EN#`Zr*_&35o%5mqT1;s7d*{3$S_R$hcS^qe
zH2ItaAdzRu=L8U@KMA~K`D!jbcv-D6l%Z=tV&<6Ows^7{i)c)|#}f{oEW=3~g{kzu
zT_fU-)t!wm>NF&$K^{Oc9<{0{SAQUnI}!4r*ulOR$WvM?zH0={0K`{KfxF=qm@sso
z<(l<((5A4035r(QG22>kLkDZ=h4Eg8QP;n`BWJ6I%^$uK=z5rYSArZ!f88GIMY2bA
zrOA}YwLF~Yz(gg<CN`%16U-<BTuHbcGal-<YW*gXUFq@6yvNt)Ldsa9RDXZ1(!39v
z@U%#;Da2Mt>4@cPbh&oNwB!Cr5Z3t>#DDI4)?05C3%aFFf9Fwi+R*S=yH&lGHySte
zLpZ0q;f^c6>u)pUeX3++SIs0$VqUhdaV$H_52bo3xBb~KN}DRLWSb7!IIRFhYuQL%
znl>Bv3@QPGR@w8e^;;#5uYY%s+lN&lwgK1>12w(Mh)`&gCVw6uUJJmh!(=tn>6Mqu
z-~RD#=R(d$#9MQDv2(TNQCaDR;{BLMfL!#)eH8^=OYT%2*sGSyztR=WP9r70q7!;<
z<2z97QmhDTqS8^+>z;-@(-OShhcDbr$oJff)-$)zm5hzCn#a>|enZ!u5PM^^vED|<
zR%&ak{9?pUKkY&ndT+EQ9=~%y$i=psT6k<=aB8ViPq*@EIp*PsN6X~HRr6P=%iCZ7
z0clwvj+X%!4HE%2ml0k<6qodDDkzpK7&?EDco4+Lk=@5mcHi7QzTVuwXzZNuMmgwS
zKXi(Ikn~qcYP6<vN(rk~_xh>({%a}0((_0UN~twHjy`*Oo8v(%Pd(HFJ?eOHs8W9o
z^eFoL#NTtjINJ2T!b7B>!`F$jq=$PtB{kEeek#!Obo}A<+Z&n0uVu9k1xYl>O8kF}
zV}6dl;nVTJ5*eFYh^REcEx8}dg*LUyNe_}9U+Ddmdl}!4Lwe&d);R7zd^1O2j@-xS
z{yXIFX)IG7W#sYr$iHM7@RDXPQ(uU5lKjIsYQ->>qNHL%lGJ7yBNblD`xgsc5W#}Q
zss6P!!UQPR^!CeLKV;ctNLvEDt<@8to_nPc57ryMzNx;wd%gMBOq-WWEEpYs!}f}#
zl-3$k7`BomI(F75fdE;`tZH;=sa}%QNW+FdpFeh0%lnBLYD*HTnli{`qeT=x)SdNA
z3~G2}Z%DpmPZ!H;q1=oU`RJlpo>pDYmWRDG+Az!Y)Gt?jKPNlphTc183yma^)x$z(
z%V$c-`<N2Xbo7XR)@ErnQVbA(*3ndz?v#*e1&WsG3S>zj2qSS&j82|`z5G!M`iv$^
zv!b{lNO?82Re4{I2P<RbVLiS(U4MZhu=h1Tg@tsdQ8Xf;o>6>0n<C9wsNiN4?cxdo
zxiA`4_ok)4NW01lgXUV=EvfC$uDeWnih9Posr186r1Bn>W`fE0`n_X+aTXLit<g2r
z9kZ}$Z<DF_<78GR*aJ}XuwgbUISYp;QPn6wLrt~vgt505NWJPcIA2F5SYg3mSw|Js
zZ5`=%Js<UsTOdUiOzX<BOLD7D4)ZEnXkPa(K`w2LS!<O<{XB{}tZlAYLlp&WC;3Ot
zr(D`q%s5Z}4AVZoUk~|fbeZ0~5timPjSP??!E>1z#kq#{=wP(a!PgXb=uD3G(6>Sv
zGiEYV66J=p+iYyVF?O~$%ozi#g)<|E6e?qy>&K<<MJz(Ztmo-l8coj9>?#@7QWEDy
z08{s#g{P%NP|;gW-piS$474PdP%aoFe{4JdiCsSpY-IPh?$4abZV>k5ZbTTIP?QZc
z2+IycpwzM_JCZOu2s$R5PGYBI$_jwcCQHT_Oij0$t<#x;gQTKVK@}UB(rhvG_;l#x
z$ENQ(eI`3q94a4EyGd`Ce)C5*>Cq!Owhn~t&L^bzV<MVPtE0<Yg`#^4ibc*fe`>dI
zlfiDzR`8bLXaRVUMW0Avrw<cViRv65u7gFiaN|45DTCv)(7e!sj!utb#OKmTjQTF6
zux?UpK;PI@d54*Ata((qy&hZV`9U&DRDv55Yh+rgDZ%joKA##>wW1V7%|%JAxoWaL
z`B+miu^Q8=U;r5hRN~+MMQW9be;Cd!jzS}}gmTn9HVYrXeQ~x>DR0X28mlr*{&ZT`
zWYU$?M&32&N~d-JU5?7t8b@bGYSq3>)dG10MN~AF(&XOmZdIF0aAfVa<no8RgR*vm
zndTf^`jQzm6`?8e2j^xBq1rVlKDKG7Rl&JaJoI60Zc#~g_-4o4CKn`nf1`_qyL8Z?
z9m%uP3*^|LX?u#<&aCzQkSNdPqy!Sd)pD+T{W(c=+}kFOK4$OBi-$gtP*@Aa)m#iX
zt*BN@mlkKTh*gN0g%q?x!Zf?>%a_YVPz+<?)23QoDpstTDj-=mIRsY=O|qUpW$V_9
z4x!2O9L-*8p=`3z&vyqce{cW&k00AX!zkgUExG+Mb2E!WEC>xy4*AgLV#)m^mP58f
zgHwVs=yKFA$!mX&Z#R1HGUH%KTYkn^HD{=$;*(i1XmEL4b9w1PIplp)Y3%P1mF9&@
z2eRA0w*w1M3}PFca{21+0L~5at<=OdNkINwF6FQ7Rlq()sn!lmf6wsepYOJ+@D;x=
zZFWuy7b+VWn5+b6&O9KLYuK;EybA~wpj76H=$<91j3$}gFUjoMONV4Yu6B{QbS?Fw
z0s*?>E8C0|?VhUA(Z{<zxsi58fFI-cLylcT$z*K@P@48%+7c@Y#^<95$oy$M>9q4T
zGL{f#6#3AQA+RaXf3+W;DnPS2xHOauhn5nK;@>gV3b5j9mm#dUnVbz>Q3~gwT!$^_
ziUPPHQ>&KlT<8iYP<T0qt|;;tuopvD+5kt^oMujQ1v~?_B9|tiN}{Wvl@wN+s`ELZ
zMvs1#bmGC3?OZntrsW`c22f`$?eI`@H*M~z3UxX6gs@fIe{GX{23RiU4jklVt?pP4
zX>d=4;mYvjVquYzz*slS0nJS?n1_u0onbJ^@=EHfcG{yjRqt~QHC22oN<&C?Ls*_0
z4AD}y9(j(%WYeo;7~F;BSe79neV5CSF}pPx{5lSkjpuS1fTF$gt1>)s5pY=sI&Xw9
z)5V$vYf4?re_LD)s3^Cu22`Ej@M=KCY2#`@E!X*11GEa1BmO^J4bT)8y}s7ffQs5C
zJfx+)NO~^JcxSI9eMRlfp>OauR-&e`s_(I#J}(_}(rgiZd7+slR8)7&V!|h`m>YA}
zh;`WsP(T_xJ$$7gR7>`}9I}jwmm{j)7iN1v2RTOvf2WFS44=!v2z(H8B`xpjmze?O
z-oacOU9YH~gsz{PTU9)(<J;Q)SdSXf{T+sR4fH>oMk&WB=zlVmlEC$9n#TbbxZd3W
zIy;h6Uh{51Mez;p27o~^UYmM@y8(b&NWXPAAnjW1y8+W%1efCUitLM0mTThlAVYN7
ztK;+_f9b5RF2d<gCb3U4J`))8MmW6+l=<K?PVZ8Rp4;|55wDNFKVslTSuA(V3?9<k
zr9-{RV1H^4>3hBu&<7S}_Cw+4ui=(%Kp1z4(VR+QpPMjM2B-fwP`IMh$I0=GKTQTF
z(LWythV=I30+3s223N*HZi&`j1M*W0Nm-n6e@bvYV{2^9`)9j2&n=7(xYkX{eI><!
z<@JSPoEC3W#b&v<+XU;}Ua$`0aVZun_Ca+NZG!0n3dH`Pa{JSle{KN#M2PGEBa;n;
zj{pCFv^yEFh0j5}=sp>+jm}vsw06@mUlK_2+ucFi+k}+-z5&qDdQ^Pd;9NZ8VKHU{
zf14ve0i&h^-nUUD)hmZgQ%IAB=nZ6!Z@L!?_*ENY0tRmUbii+EfLeyfGzvAHb*f)D
z&P%a*F@u3Tji+gsA_B=*YPY=t#?vXQSmpZ-fKUq`U}*(bI9SS|N?+SXaw(yA6f=r3
zfE)0iB^NCO`~B{KcssFfg}28IEDwb)f7fpd$qU4_Fo=CP9=V}e`^Pbx?~M#0&J_69
zo___Kmk{$Wy1eo<)@afwTcCHU&#`0Amy~a(c6?~!jkK>&Hd)Oy(tyd>l}&@91;f7O
z*nF49HTlLkV5_0S9@k{6!3&G399tD9^PAGN{{khphD8cxZe(+Ga%Ev{3T19&mk$>W
z1($)QG8LB)_yj6{$&MU1486}+)X_dz7D=r@Fo1i;W5B1xnVaoXHUx+dF#;sW*N3W7
zEn=-*Rh{I})0^~2krbCVt*c)uUA?=zJm0;3B2#%4EO@M*f3ASP(EJxPf9P!G0hA-C
zp1)KdztI{UKQ8<*S`+cZ^XtIR_Wl(?;1%$OA8`N58_4T_3Hbs1`XRk<*VVgi^DVxq
z5h#9>ku^WkN)MV6rp>r?|M~g-ozCjNezIXncjt{_4LjXgPs_SnhPN!$vQ8#6&z8?&
zSU<zk%~LI9%`(nf<xFQ8^7r7<0W78?`t(XSf{bqjb--Y&sG>t>3SeWE^)Lcv1kY+&
zlXrVRx!Y-f<)bNfzxc#Dyli$koZe-WbFtj@*ULpgUMq4m5(KTF>BD)i*aC7Fnq~EZ
zR~E}nf4^MxPDEuRN+as0&MC>`E;%HdGr$xM0;$LbkT{nnh7z9BDZ;_D?=Yc1^1Fw{
z_(ALVOd~hLV-wqKgcRQC>797@_WADjI~j)7>N?1OfB15jzAyxe0vnSUJXT+Ss(!xv
zPNEL4e|>!ISewF(pjmTv7m?yi+bTH|(OMba*8TbVOC>(}@UGXJ94iPC8Fg{B`#WJ&
zh;lKF)Mz|IE@vEfy;^}3_>m;eWRq%fTpAjgCtcM#yX0(=%vC=>=1gO|=olFqk-gC6
zswP5zGZN<a56wx3IjfGxDQi4r3oxWsKo#)qYb;QFGPFRe31@uJp>s?+Pi!({(!a}+
zw$*0FaZ$1JB-*UL-F>o`>wOowMujlaT#v%)8dTIq84D6M-FG&XhUs%7c!iwb<+%^|
zL-W^sl~3|ViMGKXkB79kf#pUhw8yM%z7@uQ4Q@=FKY=AhF-XMq6Ht2#VJ1`WI+qU$
z35VyZ&yo8Io6l5x$n^59pE;XX&XPpkaJIKyJ-YVRpbpAwCuM?}iEB*7Ln>I9)8xyR
zOs17he_C@EpL|HIThvo^2g%g<LQZ2o(fmzUE;LTXv8HuPnOe<<>kVb2@-~hPHC-Bi
zxPwvLsEz_;bMx7hu;H2%2Kr&PbXdnJ$5U52R*H}|t7T&(fm;>=(@BWEuBn}Bb!@KG
zFg(rR)Otr^=T_5zlk!(ma9hg7uI{o=wL-wpFnFGBwKUf1h*k?p@s!lIS}j6fazZUG
z@aqXkF8h&Ug8cYdS3mNX?|E^`bAQ%<OTTntboE>Hed^^_fCJwN28I;0umr3tmd(vo
z1KUGbib`{dxEf-kW-0G*hrS5QffudBUF=suVM?H!rpj)EW2gz1^nA&|6I1Zy&tm@U
zA;0EO<yop<Qv~!Z+9e}UJE>+Am%pRD&NrxCF!q)zLay_{Zey)E!K$F?P<UK_3$gJ=
zFs$Q2amS62SFetjWr~249}i=F%lMRY;+wL;s&i?bGdEEK#v=C)xQ7~;nS(PmFa++#
z^qGO%Ch~*IErWr3hJ)bhW!bsn1Z}0F%Q6SSLp}zOGS@hWz^$?^t=3KXGlGIsh<+pp
zR=5Ff=`_g0?cfGLnczk^XK(|5$!QXJFEv2M)xmumYQSJrtp5$D0fD?s4T!}zqy~hf
z6>320wVxU=5R|pye?Scw4lEnb8bt-#7HZ&-_7t6R<|O4yuq|WsZdya=eGaQpf^H-=
z#GwzkB5;upXR-nUaU&~Wh8Y`A2r+Mz$Fo@hf%>no0z&38D<BoT6)PZr<m_Mt1i}fd
zfHc%HD<A{pcB}w79!uRVSpniMC1|_K3S364z@W=MzzPWY8(9Id>o_ZbF{r5hnXCZu
z<ZbADQh<QlW`jqjzN;ZNYQ|F^2D%L?KrF=if)t>^7*Bc7&%=KG+z$Vj{5lTS#604=
zdnl6KB&Z*3Hu>|*#Os!S8}Qq?f&DgH^93r5=1`Pe3$gKDFszRDU*ZPPaZEv+;s(-x
z66#_env9v=_gv>DjsQ(i4(WW3U<mx;j}?vpR3uk!`J~aVa0JLLB+3(P<_Me$#!8tz
zM-Zk5e$DAL*8e<D5m4{?c9~X)oam!C+)wQms<<KW+%=Y4cp@Qx){ScfQV=u?Pb?g<
z^;|w^EK*2vjzKnM%8T`uUl+*><a_J6U`PurlNYFYo(v$ZK1E*aXAB|_H%&$vmNFn7
zX@fb;7$CCqdUlrJ`qIyZG5z*LARU;80Qux>X@LlCx`}8{0=CX`ZR>OCf17lbY*G$_
zM4RI~2o<0k$<DKXMBNtVu|NIVYNlyR^Pi44AZ<$EWzMEtdqc{m6tu$F2(9)LHV|2Q
z|C_o0kg$QkG2`(h&O@MWBW%(vYIclQ>30gzk4|ue`Di9vfhZ>1s}P&%R{}mRawuYT
zx~koTe1&M_!s5f&JhNAii<9yO<PMv=;l@qP7<eS$JeNU#ptAz5MB5N>{+RYo^T1$E
zq@@Jn?jNfhJQv#&qdm(jCE5nBq#p>s-OYarc?T{mM5`hX))H~%#<nS_T;1H*WB#C^
zh}_9Ep+Mc>m%#*6uO9wm`kGt>%D5=y!_%>JojE^=IzvSADii$31_yk^jUjn|wvS5G
zZB*{A^4ZXT{PGm~aIq@^4^_`x5C!7QWo(12T-{y95%=8}wCjzR$y%w|Okwq?ETUXp
zr|cd4dq|{?U9oe&nuo$stPbi%C@cF3Ts-m6V*N<LE4X+{69Z4z(v|zUcuJK?k6bX*
z?^CMHb`;7t<l+IGGCh-Y11=u4<?FFZ#rtI{eq0hG1gW*FgD+H@`~OIruZqR(zYkJ_
z+L(8Ru$RwopYQ$wS>>shk)j<F0y8$3G2sOim!*~hD3=XB7@z`Ij+d`K7?XecloXwe
z2$wI4PX-Wioyxs)^DM1*<#~Dr3!-pL7N{SOcTSu81*8w7@BMFIh@D4-3SxckEN~>O
z%IXF+JXc;DYOqncNG~#wYf<d<380l|p5qg(^JSB<VJ#{-W0fH@L?uKZ3(Z<+%0i14
zF~!0JY^SZQ80R$nnOu)8#Wa6`z9k?68F!z)jqx8%va1n|vI>N?z{YHXJ(u6<^L@GR
z%kKn90~FUd1z_@6XzGCj*Tdm*1&=EnxdB&#kGuFecevyst@ihYyJzfL;3>AO6xZ%D
z`(LL6XATj5Zk&vDAjT>2q0+&dzutBr-p_)tRn|2iKVo~gtAt?xB3yq%eO)IcRuTIN
zA-cUY(Y$%r)3dX{hyWH2(E#fWE<Jok=9#2<I^M&dCq~>Wk5B93k4EX>=Lybx4a^8E
zN=q<XV<`6fzGI-Xhru3`9huBd>wz{3K8GqJ+kbDNp;+-yVT_+xDO6<y<%@lGs~gYh
zV?>5Hk1X8fx@Aq&2s0&$VR+Ueamf&Uq~r!K$2Egs#PHy$g02h#wyC1$P$fx5p@MXF
zx8y(=b_0JrHwQXBL_}zs?5!lat7uy^m6m5|hk4pXt|nR)hm4+1r7mI>`)si`X!XMW
zJBihc$9+-iCiC0X(^%!XhDdkMl@T;>BbaM<VUPeKildsb$+2AD%1F*d_2P1ZG7&dA
zYi($3#!c6Jy~I)b1!^%biX2%Xwu^bfMkhXY<!FD?UC$Q64fzSBQmstbLVG1V5fAsT
z;r-03{m`j=Nzf@XJ2S4BGDN~;E-n2~f|hEkH=2nw`U5fNu~igDgyyG{Ky~NSjEA|@
zzYG<rN&AWY%ihsziJ>4~jTH^HMSd@ig1h|Tdi_O0>I*Hs#s{v$2>%eCV0=ICRa)nU
zyIFsoom;<5AL-?R$ZP4N#<oKrr<duYruQ29XfST*qh9?Q`aoMm!BzS&a|bNN_v5lo
zEa#Kh2IIC)cyxp~N+{*1Cnu3pi4W(O?KSn8#*g>9@8D)(zg^aEu<o$Ic4w#2uC7Uw
zqy(*AyDR;oI%>$LHP*%K8&F^Ri40z#BZ_~@T;?=hOGh=<Z5<i8ULU=WHnf&JI-mn{
z?P`0vo-TQ1Q2IBR@({}7Vkt7{xrP$^GVNhktmA8Iug1C~qI^iWpP@lfVrjCA7kCqy
z)N3^eUywEiWj7cXjI0k4_39yQR08Wtm6C$%2JbX^a<yn9D<5S;65}@-X^f~@Ek%D3
z$@sLg=}&K%)g91%`xy0HLz`MOYyAZC`O?M~UY1(obd1erVkCK*Fija=Z+7W@hYnNO
z%opqF6#{dZ7-`sh{?;`W<2CwJrPkJ|5!4iqj(^v?f@~q{b^L@tz?LG7DXiBYrHhD{
zE#>iyFh@Dh=T%L_+&hq6Q=Y_(n(cozhdjw5;y8Z&Ax}z^1>U}7hVN<=J2{i|e%t*y
zQ=XL0xb&r(wj~e7a=CJk%Wr9>JHzPtx!m*nhB2SNuMK%Iyl;IDcvF+tqS$G?KuadP
z+BD|+=@)4WbDW!>?9^jyNQ9Imei^oqWA5bcry;UWDVa24Lla6vWbcgW`H_FN$uA3s
zPRQ_9hn9=HS1B_boZk;h`j#fw^Lx6{d|EyUWLHl~yq1)Vhf!?{sbOiq(2o5?|CMs%
z;FZUthOrr0NSPPpZu;eTNB6`hy1djH%ec(BbaL*Kjqlf%SrKYxMPivt>2S<Ume`UI
zXC`*%CKpjW@#wW>yOD*VpzeP-w;d7ckAX+yL;>5Eg5f;YOFLRKI#8~Th-cRA*lINn
zIr*0mo2SSo-afm5YPM<6lIyfv&cy$OcmW<caLa1r32ZO*&(uiP#>#s)IQDPc_ePL(
zD6U(+Y90ESv8{fzwYP)J>En1DOU5JEeTB7SH0z?WB{{Sd7RSN6RjUr;;B|9VK9RHk
z4<xp(-IoCu4HE-4H8_{CZ9Ef~aW^R_m#|0}Eq^w<=P4kE(P(<edb6G!z5Be=yYFuH
zA8+2hBJI2i7QESg{Imo9Li1nH{LtB*2T%@S_wmc_#|N#A<JX>Fq%|VHJpUi~-9FwU
z1YQA8{DQ}OPmtH6;TQ1#&+@sfOS|pxt$9iiFg|3+nqRWggGPaAGhF-l^T+RRbS7V`
z=zrv2d22tte7yP3P0NgRq=>@qZ(nZY3qnvB%nraR>-_H5f9yWp{G|mumsJ3km3GeB
zN>-p*fV6WiDwMAVSuxStGIzu>B9;-o4%92+ZR4L`Kku3+AHJIiN5>RG3)$#O4&!Pf
z6=J&>7i92*y$Cp~$DK}AKm~q*iPhdED}Q)JxDbyh-ZF7k=z1Q+mEPYEbXs}voFFa)
zZG%GAP7$=rI|P!+pPvU#({mWU6RvGf8qZk*Yy(;W?m+1*f$GUz0)-M1-u$Q-;U_Xw
z8t6|{@c(2BH7kmXij~(wTc!8)@!n}UUWuy<sAuycAFf4x4ebc(#R^|ex4y_ODM*AC
zA5h~QPJw7uBwRpz-{qJ2Q4oE*TC9wPb&);{J0ggcK1`N>J{9C<Cnm1z#3G-N&_X<{
z6Y&;F|NHdxRNWXIl-I7s3w@WNN*EmhR+qm@7(^A#OXQPKueSTs@q$+dYzv76s^N7t
zml#VJI)8O9;WIWWZ(~dMz-`6pTFPj=QP!9iLP~rkAwI)Z1grSfLR`toOo(c!PRYlM
zuu~AI0?m-=mt;y*Y)cCFJfxD75*{Jx@nkO0lX503xUgL=&<7R5!hVVlPQlALDn?jX
z;>+1y7A2=x@X>d&B%Z}CbSJjEk^$QwIS`E;AAiWDQ)056{Vp(gt-P^w4BL4~&b@Q~
zD3B!pChQB$$vmpmyH|1nzr*akU(<n8!T7eOGpM+9EhSxnFex%5bb<fxfJ8P1D-eJD
zS?~VN-+j+(C7t?jTKQ$fqIdt={Z&fF18@U-0D(&ioY?}bcbpM%4L5+DKv;}QQ#>8S
zkqwlwxQxbauxuF7ws1F>T}&7ne;)EnzAIj3VyH5VQB-I1bVyd3!!cc?#Zy5XX}1iP
z#e0t7wg)7s)rMVAHhPD>&JYDIF=#61bFpHdrt-as38J<`{U&n6rPx>tNPmg;Q<=jC
z6@+|6A_ZjYN=|t3^uri>mzWG=aJ0ZQhTg$EkK|_x^ecRYpdu$!4YuSqe^W2a%O@&b
zF`N0d_R%CCnO&%4GBkJYVM@5tvqDvJy#?;n7#ZQ?th}cgEI)BM4;(b^6z6ZB$4?W*
zZU_d&=?~DyZx`VpTR~qpWK0VGwY1FZ>!$6H@u{)Y$XL3NWTby6I-O*Y?1_r}=Tc|t
z%9@yEY99wl>aiX0?v?#-e`FIfY((>vjsubNtZDW;HizchKi}moT%xGxG1OTg7rwxe
zM%dN^-M9bFJG}2ALCzbsuaSIPkl5}PLdA)Cz2(z37GhVYZN23u7Q)@804%8BW4-I^
zH+T2o-0kmo_vmiZ2Y>s=Hg@lnC$qo~+iZ$E3Jsg4AfoOTPPcq)f3|!mTS8Z3k>P}p
zOwPrj_j8Mz>>ztKR&<T!Pn(iNFsZ(%CAn!zY_^Il)`b0}0k;vcF9d`d(eIlQKob@E
z2GOV**WH2=DX67Q>uruWSXvM39CghA#dKj|y2jq!8}Dy_*%pW6j3qX5|6?1Ub1aT}
z#~~z<W!`Mbr{wE4f3Co8!uSqvoA$?6ZmF^I?URhrb9G8x5`W($k+eL76G`O4L6P?>
zB=!-p>m(cMEuZg>Nii@|%D$E<#yyhHL9!10@=Xd0bpkM3pVxtVm{ngs2N`lDWF{yn
zaOdVA**OkhDy=xKohJgbpcKl0saLMgx-k+2ElBUf#7DFbe<Os9O=>Y;IwLtDNc`m4
ziS%61(nFTUzN8+H+xEjp7ivj!V7q7Ear?M!3GBuK8*8-_?zW>pasf!KS_i^yA$(L;
z$yJzzBI5p>ZjCt#T<fqus!058+gKyAe~a~&*{SFwvf8pMgL>C?6kx3@uWdTO0qcaK
zFFnfw*BD#rf7vk0EydWM>Lx)sO6-x0tvdHp9<_RupmUVAi~>BQ0pc9n;dBJNRmfRY
zwd=QUHqB24o7iH4Z8V(VBe;4?<IY!*#gL|bEnC~>ZZxT&!$R~=lh2|3cB0A$MtpRG
z7w|yli5#{Xaeg~QRd{zGF+}B{Ci#F_azUrq#2Y53e~i93?P90tw2!&GHRGq#&NgyX
zak^2-_Sr_Rs3Q~KNb|tiMly;*Ba*M%i5F)!5CoCsG@I(wvoA&Ve+tN7U@Ov(MGbvA
z*RvGg=<b*-1(orQc?}2r<ZB*MKu9mDX$dk0$JODbw>;2BhC^>~%cFt3eaoX+{D!wY
z8cYXne|fZYoqx+i4BiF%e{jnK(JOD<mF`J2(BAx(2O6%Y-L>whG|-N@<#F~6kCt+{
zZ0^mGlPt7QQ;%8t(P8C|!Z3Yp`My9OU{eIY?hLhodi)Iaf?FO9&b2E9!!3`Nz}IUJ
zW%Izh!E?$`mN@J#DZOMUn+L_MpjL*mjyU08f2i_sC;Gfmt9Hs-e?W;RL1aiQ)CELy
zK18YRy<I?b{c*5KRZnYZ&zGH6$+zj6cd{CYN8ZWGUSM`?{A%Phoa%)5hQndQSz5&D
zQp?)V*Jtrh!C-CW{u;AjV<H!`bj?|?;jkyC<m$s<aW?4UC1b>sK^#dLT1D_SvtYSS
zf97J9J)}7=<oBmM`z<qH`M<0Q`FU9ra_pAx{BYr9JRD<Cmxa%>R~aPR{F)D-3&NPc
zb2FPfFl;_sW`|V4Wuk$LjO0`hN1h7?YmB+tta-a-3sI6!_oh=}%P9>mFul=KIEtuR
zWb?r!0py-BrS7db<>WAFiVlRXZn|Ade;CG+vgi*9!!$HqNjQj7;D#3az_FckOv6}r
zkdQTidzdf}QcSYT?cV8A3O|18+WUuTeBMOzqKR4W-5u<`mtB2I=Pl^`#E#Ngm_FCl
zFx{RhL3)}CjF$)k^G_N2^SmbR47Fz$1kw||HvD$S&l!gH&i17dknTX3ED8tNe@kK`
zKM@+R3dW-@i-4s3*7qhk#+%%J7T6ZrasK003kxhKvY`Th`^)xJkAsDBP?L0^Q3f(P
z)Nu_>x4|#1@W53#;Lv0{;>rN{$OWvc(TrCHLcgcYWecZwK!GWo<`u&@eR5(c(DN!#
z-sG0u58D?WG{-=o<Za89u-kubf6w(blTW6R(ssX@3r-Cys_4_jMZVXKSP#%;xMEl%
zYvr~F=b=H&imlh1T6cX#7$ZM)!i6HnR(;$Nhi<^yeWNHQrd<*Ie2P&eF9w-PBWm^x
zWO-ALBUFjym^yJKlBb9*SEzcp3rsFS@VH(-ks=K3X`Xw@;WPk-KzYAsFImoZr+-{2
zSC;aM{t)XL5fa%@mL$JXwjjOnfY@Ry<6Y9}_mr;O78}`b69bk9)69ys;)2;<MK<8X
zdiR@IF-(%)BK7#SW@^pSOTpc1N6*xdSB<iel`r{@+4#9%L>Jf3C)VP0^?n)JmUp_^
zJ8rQb&PsRY)UgA*b8i;KCl@?YF@JAP9gb;xr8C#>e3hy^@_B?%i6rL$RfobH-4H81
z-x_eb9$@A{b^bf2v&RfI5_rG({hgR)5sb09dozEs(Htxt<r)dbx`C4|=%`*{ZVT4{
zXWbcGHc!mp_nF@~4dW2bo)^`Cw#P;F+WE|co~Fanv|?JyJzjI_mD^zx&?Pkv9H^-M
zY~dRWyULHd{J~$&<SD|8n&U@YE62ZiXYcYqKjr9`!Pztu12Z-@m+{2`69F-o@k<RS
zm)cYqqXF}`epMJ`1AkHnsf=>eSYvv_TwJZR^~q2kYXvLe1~WNKV&BRLWVyH3tc-pM
zM!V7O<e2srVQr>BrVJ8HVJFE!uJ&{g>_Mtn{3^AfjCvBfA$fkuF{Jbydb9~focgWO
zol=tVv2+(ob?Q0Ow7jWM4Nn5n^0+fT`{SuBPY?81$&*@|bblA$bTU1Q8JW^(JV#1f
z6+#M6rnCwxF%}C$=`o?ES#H&UBoiM)jjeDm1}yjS6@e64yg_(q5>7yM2Om&t?!!Ot
z*OU#en1zLO%2vLFuqU#QiK)z{D(tIvO3Twr#imtrY$`#l8}h>snM)X{Vctro+a3tN
zHkgC~IJJ8OQ-4x1@)N;AX%{#)*wgt*yT`@~&f`3GGhJ-&QqSy+{Z$r`Y!xH&7J(J(
z#uDas#Ef;%Vn+pi1?+t(h3DAH4CNK5OwTD9&@Co<()n1%h${PWRm(zIzz1~kNOhcV
zwHzu&b6pJ60)tpal!DKP-<2MAmM^UVC8*`=_xG!!0Dn3v<n~RM&-ZI9L0C9Q@7pea
zS|7zpCb($3i2Da&?Uvp=k_J!ndYd0i!U~Y{w1fS9RU}CgIyP;W|Ew|wgJI;l34L21
zMF19qTBo=N>&CI>B~lcNT}{V7bD>wQA$lz2BTX;9&Y?EUN!@n&cAYmPnX=tlaO<NO
z#CQ@5+JBxO#ey;2+s~RKl+FOVbym9W3By#Xps)kB20Y<<>3><3UU3OgI>x@a-xIY%
zK!Gg`nI*!7(DYcx!0<%O1d#kR``C1c03?QJDAT<t_on?46*E*@#>;^y(-EIi{UR1$
z<v`t4@ilO9F1>YyHd5QFY+HI)U&yq{-Z-N4X@8g{Bv5dWop#pI))LN2;d$UW_IoNy
z>MOa@wqI$In9Y}(#Sn^gk`SmnrURg<@Ig^nnIKuJ%7dD=IhYs`5?Cyby4V(gk`OK<
zkn~?;Qt`<gB#Wtcm^9%Z>%z?HgpqSd#YNA_*O;(yuX44md}3RlC@BTw-1W>(CaP24
zuz&t|p0?NOi?1sRJEG)-ucTLx=`(xtnZb{r8$vn>=<qlwqi|H#z8LU(wf<E_VdjJ5
z#E8PDfD!pcZbd**wqE(J;HqlHuNsEt2n@cbUL*V?xsKD!a5dr8BwmIplFhSK&}Nh&
zA;AF`UMjTcrXMD151D)Du_m2n>dDhslYfhu<C>>O)~&_p>0PpdI96v-z^V_*1gnTo
zidCOSKOUu0uFUkMdP{Rv=+b(~p;WQ${5;Uib@^zouFum#OTTE%yE0Va1yJ1hVa>E|
zaBviEHESbeC|bAFKAhKstwvG<fY33Zo=2kquGp>2Dts-#X#hzH)X}h5gS+|U0DoAL
zj6pA?IzDFyv<||N6#;xK4z$1?u&2kGJjfo>9V=Vz#QHc;yEE-D-`qby#`I$cjI@-<
zxbP|Mpm5g3qW)s9`JBXsa-eIaThEw<4_|AjH>%*#6<19KrXVZdBPGMPR@4%x_W<?X
z4F2|q4O9<s>L#<!H_`{FQwn!vwSR@|)=Bf#R-1yEk+7XIR&}8?Y?<^yU8s;)YkhFt
zC@Yv)T-NggAvLm{29dMHm9rrj0;JXAs9+vhtfzy5!Bm;#hde^4x?3w|1shyKefMqp
zSr=LdFX)e2_7IgEXGY4Mj)Dk@B(g(cJtOJ%IEwI*l$v3Rn;?~U^2E?rV1J}<N@!}>
z8?h(|(sduKr50qv8H9ZmGOU?a_$`$=$t>@BZG_&KfSX?Y0l7wD8_XP^KoJRdW)%$E
z8**(=xKe>^JZ$Pa)Bz{TtWazjXSWNgsE;Un;DkmD8@5OcUxv?ZowA_%i0Hj(6Re)w
zMn{7&)JY>XE$ZAQQ9ABShksRp*c<G8hchJ#``Owan_-#&cC50Cz2;pGNnmKeY+)>e
zF}A*R!2qRZT(kp~^(SN%rb`kKrY5I2=Q`mjCFWpg`Tb<G%1m-po>a-%uy>~<kbFq9
z3Oq-j>QjnsOD?(XWQ;Lcc%o=peZr8^puGtLH3>sbp@!#s5(ZjMe}9+D^9$8no<eVM
z^|aSqB$*U?!yZo2nae_jHEd%HL9HIB)Z1wuSV<LA+`3Cyut9qpl4-yLXWIdP+d!Uh
z!vg6Ah0S}i&Vnh|_GshF7xzq%Y$$BvN5Qe?OxtH0KVrp_a&52kMrj<l!nuARVK++S
zoO#|VC6~qz9~=Onk$*}|4h{gk81Ak=IFL6Jg(8OVRmJz(T~{5U^-Wi8<16;-lN5m2
zrXw^tqj6-@R)7S$C#Y{An0A461nuf3u&(hHdNIhR0gB#FjJ%nvBPeem=-~b~6kTIH
zvi=?P;=>+Aw>}%kwxGe4_;*xv*`>B>(cR&aw(Ws^^Aj%+5Px`e(&}>k^cHu}QS8q=
z7+=_eEu?M7^^+ajXS)&Pa^Yfo`)po)Ek#|->;_)*#@t>Tek|M;p5s2Oom*yIIE6Er
zHIuh1euWL8d0%<QK(`<Ch#`iZ1<1!Y3~_Ajnbp2;ebpY#wCl2$WMSD^<n%(k$GDE-
z&X%3yhIQ^H`hV_syO+g`6kJ+S?@s9wgX7WG0CZ4Kc`3*%#hmXO%el?;&pFjL@yz96
zwn}tpM_SsJTvkoT*T!Lh>a4~@(Ll?J<p1-~W#L3o%t&rL<yRm}OYMJm%CB$_Sg!oL
zoG2-tyruXKM@p`e3$akQXD@F#eN<veOLfoeT(QAEqhy<QtdP+0PuDy7j9NOiTy8iH
znGRU?Y%Uyz%7-WW@zu>IVama*j`D%yh9#i31eQNcr+KeV$4zVxa8g%i>25#r)Q{j{
zj@6m6%>|pC;?kL<X5h5J;&tQrL;s8P`M<?x5u=v@7Y!2uHkZLXGZdFyVHi6GjsbVi
zcz~ClVHh5N?pBMG#FxmAZQXt8bocS@@Ot<A7r@S|K*rnd^>b(V3(bFx=0QiDH%2)y
z-RsxxyGO09<LAH+Y7M}T=l6kM>FEJX;2GwbAM^C!(d7AH`7!+dD4)x`L{0s-_BnBY
z_?S6seq^RM8iXfhzW(Wl*Kh80-oD{iGkN{+diU>tyY>K%oC1yA-@e|-7YIf{u$=|P
zN9%t6sr!8Q=Yd1vtW}VA{P*J$OYc;$URZwjgYJIhFTdf*5JUH$H1+GSjPCy4{e>4e
z#W*OP7y}DCZZX!KRWZcOc4R>imBx5Ei!*oTta4$StI;Y_%Ql%<Fl=JlIikwM*t9GG
z^ClI4nRzxba}WdK2NoWepXdBUevX+Y=I78G$>cnhVGNSG_DClrMNzqEcIog^7H66*
zv&CqyG27NZl2$2nK~eM$mI8-k4-4K8frGV$94Lt0oxQn#&%uNE1F{Cu^=>cY$RU7C
zA3V~0BjZV{r^1!fq%aKOO2QK!25Qar7UXz;3Pno5#`m+xYc~4nW3k%#N*38#aki4q
zq3qaO?~Tq|1ip4|Yn}4YZblZM6luMD!V+__ULcRk2*%;lD6&EoGZj50esUc1KC8po
zF0wo<OJuq&13c(Da!!%&_x0}E`%J?7RnbGB`Xp8%2UX%|puOu!zT7_mz5D6@!Myu_
zXQpP8T3QgiBDD2_e<};oA_=m?vJF=xpSM>bP@02lMe;oh<bG2ad;>*;HPtVF<U=|4
z?$_-Fz2;~Rvjxre$-GzAQ%z;dLbym090$W@kDs;`7RA9?7_vtuO(g%=JPv5fx~&)a
z!=~(j{o3hfU%zgi9*mQ_rj?nc?X9tYBT;A!Xj`wq!IE5~oUGPgwzr5@0nN9Fg$g?k
zGQmLQ>p+yQc5+df2(`eSUnFr_>Eu)slJD7ni!9<Y6-N^bHPyDsfde)<F$Pd3+rqyR
zMRoUbLdQ}zX&P*2DJ1%!;U-Eq&Waui(<K}6Vaax3@Q_KcG27Y>MtQ`GRta8zA%uDr
zyDdj{DsrK<3~hR|D(`&%Zfn1&om>|yp0nuDXe<ZOMk<4Wxn&YQViUBd_O&P%wk^(Y
z{EGA!<$^XS(>nE+`~FgoE@eIZ^0E1}SP>u@6=wwj3*K?=l?#G~={d<Prw+vC0qjsV
zmYZ1a%5=Z_eAU0?L<Ds={z*xHNHy~5<G5|$0!GZc=4^FS3+35}iu1x|GR(HY0|zOx
za*5_H-&SBx_p_C_^jq9ka)p_ABx54TNyjl~O7i+XXG}71OdMx|W+RG2dP#0fvC!tv
z$Nr#c-1Ohcp=n7N$1DJjqgom0jM7;E)yPx;Wh0#H@UaKI$Z=v$>PoVI$M2EKZdUZh
zu@-0CR`>hz^x(8q*d%KFaJl*-*YzQ^>lv(ecNs@6;}B*X-(Ss<&&>CapK<hy60{l>
zO2DgUv*>qeS5g+Tx6*FS?F{YuUAsrtvW__2{j>%UFBysw%&wO_W%(4u$rx8h4c9YH
zA?t*GR+@d7oazF5utuSOi?dVdSv0(eD~BY@2~fke@q}@9;7ZHSWUxa=LB6J2>!{{>
zT1W7qU&nlq5t}0mX1U7VwIcab&Q*=1f;sA?n{m)uy`%mdO_95Nw+<SrIgBOu<(7n;
zmhjcya9p{(ED2CGQ`GJXU(|B(*1>WXGt;kGjynJxZ^F#R^l|lnT1CnEb554AfvpQV
zh1cynb+!TpWzBeV=V&Sv5+i!Cl*m^_k=e}4Z#|FfkEY$jTGHYJCo9y>o<*>wRPiiq
zT5@y??v~gz3Lr%B_Sm!}?UH%#z%?&sapumrY;nfE4LFU4IG>JF-6tSA1Jrf=9KMl$
zMOcBb<-z<ENLyZieTK1*{n*HP7uV3!-H;n(m^~D(_VETM&mcbEH)eXBp1|<0SN#c-
z(os1JyNvx(7H9f5vqg))5r_{SRM2C&M)^jteSEfly!NgD>jlt41HS=S->%ev_2-|f
zFPRDR4OFUti@^G3Gyi+A9+9sM*8zt*^tJ};*MM-B5S)sC<OJOAJYS+vlbnFtz2*83
zYm(D&yBGX7)>H*<|LgXG5mdXy(`b8)%Jb}sY8!0_c48AuP4a0QOEgY_7bV4ni}sk=
z?>6x~!L-pqP4nY6^b1_D2&Ltq(4J)(I0vu1Ym{^X)JIS(wMIiH5Oa>EZe|JtVs2JY
zX-<1XMI?!T`%CZrE<SA@jB`cZP@lFwL6$>otYO>gAh7OiST(y9O+|JNq@*hADv-X0
z+__MVd^-S_VPNNpA;jV00Oe+g`*J6u@HC#~+Zfgwu<f(`P7Zk`fXnHJiU!)W*=SDg
zb=VEz+JL-_c>@Ty;k*suHk02F!fjSM0pYg$x*o!RwVYi<ZT}B2T<dJYE^mYKHQe4D
z<r91&2E5hoYi?&CNBa=yX1yCA&@IRP2z1MlHWBDnf|myj%XJNht=oyl8p*!smZHh9
z-kGAYq`aLO*G_QWB7b8m=rozB7<vvr=q8~lt{Xo=s-YT=XBfZEj&_)q!+RZ?voQGy
z!)+md;fCu~kZ>HOnl7J157ONNCPu3~ns%3$z{Ffvk)BC!4@~R==T@QRxCRq*xkA>?
z*56%RUWXG)IM=h?toSF0{c{c2+Q@OA9UIw0xV0VET+g|nF4ues{PkMtmwkFQa~q;;
zO?AcPI7HWGecBPPIV<PdvgNqe=IUE6?d&UmPE!1ObKQ$L_U)LKp<C(Q<e*T`o?JXL
zqt$`)p!kLz5H>6}t}7ZKB|6ZU2um?r{01jc!)A_{yHs)~_I`GSE%ugEZ+k5WX2Yo`
z<0R&#v+Zx?lxTb2uDny8tM7X9^aiUU-PH{pF36SD@s<r8<2Z8yd;5ltaq?>~+=Fp{
zU(Djnosmg+)tfYQ3~~ADtDj(`rbTI9xZK_-jfyxK^X1=A^4M{lH)*geaVt%e=S&qQ
zoM7?~E;K&p_CT(e)(I`3GzQ!*`?!?FnJ&R>A^My3K(@OS#yGAjNA&QR0Qes3TIo7S
z&(K3lnbVNf^`tv$LrXQZ2fQs;K)Qv0o*vBO`VSAp93i->MUWC{<|F@6m3YX_7RrcF
z^*>V>=7fUD+^ipy%0HM(W=}uH*C<CCe@<P*1+ZgH`11Q@aiS&RPcqeV!^(Xo|Kvs?
zP8(O1s?hZ+8ikrhDR*mn5$mB_W7FI)%nJq%nALKe0$_u}XspE$XX=`nu@)eIcl_B!
z)j9ITv~1q<%UJ6X4ZSI!Q=CvbR;dSNilU|?m@^sl(_WSE&wZ*a#qs-bBulcq7G<Y=
zVW~T|UPn~>9M(@~2@~2-X<*B>#ZDxRY@~U@wb?)vfN81f&fn<TOKzR@;)WEb)vutl
z<L_bNx%PTA)HDa(<0O~ueTHX$VPhJY<|Vo;unJJg3Xo{Sf`6l!0YP~S%>ilA?MJXU
z0muVdab9~i@<PTgEo6%X3%ryhFqm4AQ!P1~m}yCLlX4|FDDnHbn8B1P&4m6bE;z2R
ze(Z6V*<9i<GSev);A+B=4#U2&H_fbnTr@dP>3^=Kb<6~D%Z66`H47>(+xo0;JvGar
zTA<l{oW>dZz86)xK46$Tc=7~GIrihWjJsK>P+mF5`Y(ASSoD_x7Y!2uHkZLEI24zC
zY#7V|{<lhP7;^&``yr(uc&uy1Sx}WfYoL&y+YA`;vn+U*)o>V{e_{ybef+yuiybXz
z7ReZmS}{G7TFMNkz~BAlP2Yu9jkzwdDJ|zsmp|t&G^cZ2v=!jm9-o(mP_Y9B+AcrQ
zh~6xYhN>Q#w6@o8e|qy4to`oy^y$9)`?6q<RLlB_U|1fY_cZ09?S(~<p6wF6K+C#(
zEM2HzrP@RJCXFjee{mR{nUagsI!pOO(M-+u0V)KOA*(*W6fq>8at|nMy98k=ljvB&
z8a|j=t0}ZJeryNx$Hlo)3MoG^r}%Br%8HgR$EEy4CEv0jDk9Lb8T~zXfq6m3rlxh5
z$ML8+Z>+8wBOZfWebslb$aIS8N6%%()D%;d4a+h;r4?s4f5wXnaZzkLEr*8+bkj3A
z?-qq<8B?&ExBt8p4RMn_jefWph#IH}1r}9b-prJ?x5*?Vo0`7DYse(>bwf$!D4C?C
zYB-Mb$RvimOeV3$cO;XTq#I-s*K0qSq-bZWmH$5=lW1~;cHtyChN0b^Orr9P9RDok
z8?Y6+`(wJBe;|y7vy^Yq?oNa;c_$EI9Pv02#u1l^FmA!!MA&(zDW{kxnkJuSt~5<1
z!dUC-$ZnV8YtPT34-!fY@f#6JO1b*D@}x6%hPsPV^38vDB#-hDFVlGodt`m8j2XfK
z`lt|E4Ds60q&I@D$y^@VjqhWO;`~q5o)1)IA=QW6R9TLZMsW*OozI@-&@Jj+Nh2m}
zC27QZ9UzUk{Ntn%M_eY2c<cS7(X!kfq!9;PCXHC@qbm@^b<LdZ*e&Xs0&>4&vk^?4
zdQaiff~NCvn7jcwDVM@?7@dDBy)a%?<@@pS6}M_Cy0m!_Viv6ssfD>!Rd1KSKj&w*
z|Gt)3-EXa1v!JP1aEJV3E5yo>&@poTX~gO>{fXo_N{u!DCGmUx`Suk)lad`O-f<u#
zu82bvsDbw?<RgaRC7Z=pVtY@-U>ou=?KTU>B^MjM+=;#V7yE`bz8rtSo<{OUH+2vu
zR@uw)&c2kG-f@^Y8@y)psPmk`tCF|W32{-5zo^lmYZknjo~a>qPd*p3%7OdE^fWz@
z=^LmQoe?he8F!0R$rz`optsCC_hSe_DA$~@CquuE2Owtn{ZJfjdg^QLJ86WyR+--}
zl~lZ3&3YRgu(I{W=i`5!Y{BwbCO8#4b3FU?jY!dRM!v85`o{W8QLMD1D&BIt-3Vi4
zv<Q^?Sg-aSf;Nnw-ooR2DChEQGL!KwvGF~L=QheU81|ZheXO<<eVk-Bu3t~ksa@)q
z4c44F<_U<|g$#eE91*!=%AD;0x4r+p?nMtOa2?>BAS@OvC$N9X0)wj+JGGWGm7A6-
zJxy+`1M&o$=i&x+uhdfYR|&yU!oq9UgO=HS!f<CYr>30kR*2?h&|CMende{6yPOT_
zeUf@BS)EE}Pb&eJEtQ>p@(N7tm&0DiOxo)@(ct^uR%yLZL)#&gt5e2He;>>JF_1R1
z%T^18nwRsmaj<_B;zzk<pai2p-GN(X$jjU^YkWs;nMu0AEpxs0bIZvrmGgc(ZW+)}
znQBgAI2qdAxn%%NGh@%vzD2t`y9`nZ?aum<C_}qDyUgUBz%Fyd<Loj=TxOTK1$VQ{
z=b5I^9VeP5(;h2LlMj$t>yv+G$<R;!nI)gr{Q(<L81R25lHx7kH{!&p;xs0lOtiG1
zCotmQ_$Qb7w7MPE2GYFzLb3%o@{>!T?HmFFy!MkzRKhsJr~1&z#%TK;%z1pfQ_H}$
zzxHcOFRY#T)G^}RQInH>zJ|OjWnK$f$(Pq%4v^)I<l{_v1Gh|*H%<3*<O>pakZKLu
zGCN*3ZKw|MxxE_yq^;n80p1sAxt9SK4HE)4Fqg4SDioKGcNi;w<hBvc_g65y4Gu=*
z4cB_AX1KSmQaQ17fvO}$_Q{DauoC}$H-HAv0LhuzhXc-o1p1Bc4-LxYUyEG+`f&OB
z@TV7T7B7MYua>XBFMtmueuBiC&MqE6I1S6!@5?VwQYwen1ur9|)_8gR9q`+pK5Ga#
z0?v5B>9aRL<F&$n3-I@v{?~DNwk=Q9IaAQ_i6~3F(nt?dCz_V%#?x<Ke|V52z7)}=
zpXqD<_vP!ue;#IKoB;t2mVbPINDqt#p@Uj<K>f|~$G7G8hriEM8!zYa9_R&xl*=D!
zz;NeBGG_oZr*zKR!JHs759gxL=@FR|H8p4=i$<_$BAb4Hqe(3)Y|s#9rTczRyiv7Y
zIICx>5*>sDGF5}r3_8fX;Ek8lXD5?eY$)`EK|<{#D>dBtvGYP&0;1n&Ak#Iu<rP#x
zB|A?<L8lnPE9|PRloeiWj49VVdSh}Mghn1TDw|ptGiUQk`hG5klIDs54>%YvjP{K~
zB<H;(Llf11OW`Qj?1tAGAFM_l!p!NwPOqZu@ea6|vV7hTI3Gk*b{?HIO}080P1Q>e
z-E2EkLRyhk(ztxz)=+<aI9d=}K``;`28^}B%8jF{YqMi6dyad(SWycf&AoEPAf$;K
z<D)G4(!%XE{59B=e75}YQg03ttuTM;JEN~nGE&%oxS5q#Kvirn(*k}2``*puNLbil
z=1i{8`T#YPR|;L_kl6+Ny;B^(2>hK${_>k#{)rENz`<$R{WlN&-q^_H-^)*_RNNqp
z-pzzjLx|T1%LUV5h(xxNusW6ic`b?iB(pIl+h3)cINO=*)k>EZCc0f}p-kX04Tb^+
z*Bh*VRMJ-}@0#wXtB%!dtfzbDQz~Gkl&Pl!qKQDHs}ic;D`Q~GycUn$SVPoRH@w+i
zx7^@FMELcN0C_cu`_@Dv>#vhNS`+i)%OPXK5-=%vr}wZXq8|aOWr~ChbCteOnobzg
z1VWV+GUX?(KFO6P-Sus3Jel}Z5_*2(SnY2Euq*qQxOx~Re@L4`Bi&sqFZEpJWxTA*
zs_49`1o`)u#!C^eMzc#xs{kMCklbuta0%x|sY-#^eZmWcbONlJEtN#eo*_<}VKCJo
zZnYPF*Fx;FRHSXA0#u%Jsot$U2IWw&@feli9EVXDrKH>SGlEj(Oq+GC8RqmEzhn}{
zb?}%J$)G0Ge~`0|R3T_ta0ck}7jphu)on90Ih;UL#s8VUWMJyt0zJ{cuu)BlyTci+
zjD=Y@C)F@mhtTvjTzXh$or>{m(4R??Hm^dw7VMUVz!1`zx*K$SZx||cdq7Y-Y1m+?
z-<K&l**qTxmGYQtF@a80E4F)&yuo03R~vb&&Cg_+e+sirwVxlCFHa|RncI_sr+Fz&
zzA_XqxuS34MF-Ryf!s##!CFVrW+ZObJdCqJUsPK)y9*dRZt=o_tv8YEm9PMZksRDT
zksPvRhB;`NdY^2XWq#?2bu1-~R~i}H+;A$MI$p1vd9M0V_B{!cl9!UCL#Lp-&v&y)
z*Da1Cf2~YU>>N?ZkGw^j;2e2e+V@7k(MD($XD=RQ(Tm#2qW9@8w^?T1;MRyjC}PKs
zO3L|ZMdm`g9>=0l!hKLyMUMcjRxIEdY;@q@dF~9nyyply$PL-+ERJh_87d6an$(K$
zmlL=uNkQO-Si|$)UhZ};PTPP_H#<0A3igd&f1FHvRGU6(FK)O~dvQ@->%|#`gLU)q
zQ5_<thx;AvP;i*3jnIrtbX)A3GE2<GzBRKr2I-*jKDlvw?oi%|;+tDH@7E1?6#~G2
zpl%aIty#?bhLb)-_TUlt#*dVJq^K?5+_a4SVe)$f%pWN|r<XIQ%Psv6lh;{+DDNXB
ze;z4n-&0(rw401>ZECejy+^;<3SF4gfvT@*+r+v-E4zbOt3$ohYtdauU*k9I;G_3%
z_#n07LC+lQ6)g?uNy~C)Q@t%iNOB*MLqz?;%)QoH;$%3_Owo{ET}4m5^3upq>fJtK
zOQTff18ojR8)@vNX~x?DH9E^dz-Ob|fBvaA3<_coZT@-qN4O|oUpNn|xGihuZU}Wk
zss_wxokRHz+oD^wi+WVFn63IDi>F<SN)bCC)*1WCzE8{&R;kxIv*AEre&;5KwXAJY
znjLL?wYI5iXoG}Ivw5HhAp5iN6-Tw*cE+D)-f;5NkZIDLyJzxV;E40aI-`z_f9uLQ
z?X4+(aFygs2d_5A<_32b8rRntj&hyvuAhSzO7=_wIR|BWsWsB<Zchbm`r4GLwi;Vc
zB`$}5j^oc9F?^+3)@$iVTJh$au|1ON3a$Ix%|AWc<fP$gvaw<N4Df7!UC)U&d%&Z<
z#<t<ccI<cWb-H!#-|JM#TW(>QfAGR|WKB^5_Wi6Xoh+F}87pEm+e}ZD$onNV$#egV
z^$-#|R%)32Nj{b-PVG-YhEk7H{``2Ii_g7=VT!WX*=U@jCpLehDPVV}7=s&CJ)2_n
zr>l%?=_(G4etx`Gzf}lA%eW}q97)SJ#z5(^bu<qol}I3&eISrAzA8r{fBKf+Tt38O
z1U~(WPQPJNDGZinKI!M{j?)oA(uNH@epcRp`pb20)(PienEUOKhptBJDsTd%8-;NP
z?zJKy%o(Qqig@d~3Mht#IL?xf_rNniHwV?A{s9=cW&_Yh7~I5hN-c;Fm>Vs;RWtLR
zgAaO(yH#^@OtFI{$`kd%f9}4p4#T>epUY9sHPuy!LZeFt)xTX|RuO|5Zm6EYHZ=In
z+I{`JXKc8%(-mOdkp#4qIX*3t?M?%I#Gk#KP)k`ZIF6t5a4<L}>=s=fcPxDaqRu39
zV96UeT}Rs#fcEcFGERRkV|#Wc$ye4KU3tT?TcyueexR>UkP%(Ee`Wc3@k&=1x^D05
zwkx4AJ<W#Udq{MLs?Lf$y4?<PP=p-h5po!?fqh~f-SgQ0tLv?GYQeFdT?rd^V}C_+
z_%5<b4O|cFCK+8p93b}tDRt~@?-CfoV{3PMULNa=QPeKl-0fp-vC{t;86CcxllGJb
zW~FmmuOF=x;bGIb9vySSCW$IjX6B}dth4f)Q_u8d7}4GvqmlVt2TO~W?tmB>e+A{@
zt)8vd<!#CQkr(Q4=f3wRbGOwt$kiNbr+1rfGwU4;)C}uG>SX9RfULaoeZflgK)v<O
zYs7w`y*$0M1<&PDv^p@x;c0rQlj;aR6>kaPGQ6f~d&ObzAHGS=YCx$GDfpqS7}$Y?
zEoqAQv&b!P+t^*1>r<27=}oCpf4A|p66oKCp1Nw*Do1+<b7~^g?crW~V~>B9@51G)
z2ce_<@?1H~=lZPZwMiFm@j*=IT!HVQ+e;&1V|wahkUD2>JVSe5de-Q)JM47m$N@>E
z*Z2JC^!LV^C}xIQ!46-*X`?KAVa4ppPyK@TIv{|sy|r%Z8X%ULiS3P^Nw=*9J2<DP
z)}9^3ZsZ}*qCPVBtUOQ@-z}`YHF6Z~P2%-HF2!Cncz>=zeAPRGcd0C&w-rC#yKah|
zMY&sorf)28{{v;t=^d8=7Y!2vH942yOE4u1GA~SJZfA68ATl?X!HXX!f6ZD^Z`(K$
ze$THUa4&lgOOZ1qhde9_bhp>N0#|I&^nkkyg1}AG)=gpuwz8Z1_xFutB~~1_vFdhP
zAoWP1$l-A2n{P;QOvO0EQJnD<A!KhT>YYr6C8)qA2CqUesbm$h$xMdCn=EbHXo^Ss
zSlD<M@&Tl1yk~7)2ojode{zth1eNeKIDNGwNI_H*<8^pQ+292oyA<Pd$ohse<a|;o
z81f<qB+(H0sJEarMRbxp8{+I#DuzOCUw{-zhxO<r#rA13lu0}G7}2@*DZ28kZ+LX&
zQ#pWyCZtW0&W|EP(yn6AajD?Hs831%YbQx)m)rgbT@^J!BJ@{8e|yG|!D;u27*d3(
z5<{Y_{c$vbrInHi!Rfm!kXoJ5RS2*qBr+kgjuG&#Bpnx4!gRG~E|7|j7MzKWbRJm1
z=(~2&L~45p$wX%DIS?e#doW^5>If;}p0MvG7)4Z<h<oiqCl~??q3t5PLYD(lG>KbV
z#bRJ)m6C(hYKp<pf4gcGlS!>%Q-YLp`!vJ1q?L*h*Ny?#5*OQh5mHGQ104vHt^)j1
zLc0JkyM%V;n1ZOMKw=|ks8TSJla7liNKpq)QcQO3+~6wNiS`GxXYV7%C~#|T-4sro
zW7}UANbP9q{AwIX=ta7AY;tM6370Uvj#iv0PFE0fFoo2Ke+9-Va2*{Zn-aYC2Ny39
zU7;P`5cOir-<0CpcM&qqmG-g57U?_%evB~F35Y|~IE^-l2pgv!MwlNYaR)#BRK5A%
zRb4%u&E`#Yc>T7~tA9>s=cf93zBsEFM-K7tj;nuEFREXTe0TA>K5dSCWUKRmeXY78
zQL#D@Y-UR-f5+8RgYFNjU*~V;)id+M5-&~XGb=X#^$Gs%XYd%U1`<)WMI9_fi|GHv
z;IfCo17-?`WVR8mN3Q6E4T8=3m_9K!KQQ*g!(jo9UkyjWS$K1#geePf$yo<Dh@33*
z@xeg(>gjFg?VwQOU$0;2-wz*}=4$!3s(L?dK3u=Gf2Z@y>fNF~TTIXE^M$_M)Qd?|
z+m9E}@?eCTnQZ}X_Id`Nt<dgzHrLI3F`Zm2tJC>xdRi@;$)Y*=eR6SKFSjVS>-urA
ztVkQZ)$~O^!FhrS37*VhV*x#jNJO?$ERm8CI^lLDpB*8;CG`Hf`uBhSt8rx&P6UWQ
zyS}(Mf8NpxQ2^ILdbPlfwU(y(WqvVV99~UMYs312sa{PQtj)|&clr05=GVgpOJ#g_
z<N18nV2tO$QSShkI?8i6s7E&KE|4}o1@10@JU${|aq;9y^=dIcJ*=Cf>eaL7)tmaJ
zIbO5wtI7Ks3eB2&)+{yOcAdYjm-FkzX}#>yf4AOxS)WZOKhJMkt7|fpLdST2fj5;u
zly11*7KLwZq3}E~*4|98Hqn6zcARx`e1F!;C>}lQ&PSIWt?8_A)Dj!jyyZ13A@@S*
z1Wi>Cj=#*=TZaV2@nkhy0v)ZoA`xIZZ6c))tXVpNf5BYBA*MY*W;+^^L$3t~Z++UX
ze?=-tClrto^*14}?;&Y0l7(i)UWt&B_28CUP2rH6nF}mH96?td$1g!}AF_4eAU%Tn
zh;-!A_Xpv<qCJG~tmu6VZ>CqvlgaGtq*+W)&xbG$I8f&Qc>o=@>HyRi_z}YDE1<^!
zGuO~`SM=~3n5jT90zG`WMGvi6I*9PDe-~0O_%8<yVq`{e>6pbjrKXOoq8CuhqJG#)
zaX<n6s0axqSS`+iAVdL#sK>094k1T`O4&U`&gQ2}dpTd$=WUo=&d=(L>FoWFr}O#w
zTl`hi*;##KKQxykV+FuuQ6Yr{5(qCr@o59}6>vaMh;Oa$u1SFq=t=7VZxayEe-{lp
z^>K&mO*HntpgAfCa#$VIs=UsBbm6vU^ZO0@$PJnSF!bkJdRC6l-;yERqfaR3d?B~G
zP}HF=|34SConJNcEgr3UOX?G@e@8Q&@83Zh3{h53f?llNK?O>VS5FZBj5N7w);>wA
zCqdmUjsYWz8(>5{28?J8j5=d<e|sBW3_YAfo1{a;FN*0<oDJ0G>JWG{M+-Q##7xNx
zgQSC=r#)17eS9!r2slVST9I}u%CajDkS0f92c@{^P>;I|!Z3&s3Xe%}gI23B$tW)A
z5e+>`3HmCDaNzyhDv5g$Gl+|UTMy%65KDJ%J&2`UeVjKFE{`cvbmQ&Qf3U`zV3m-t
zO5sPe3>gU|Kc*!~wz~QCwX4JdehpS|I{<F?hXG^_SN#t2c3s=Bxwc^xnF{$1zqlt;
zp(j&e4;%_+v1(_p+)1kpU`TE^yU9}C4MN5fIk=AVcS>*uqvx$UaKY3Ien^}vKvi*j
z08;^IW-*{f4v|5K5buJ4fBOq<?w08WH)oW-F)}fsXp{kx27U~Zh6f^(1tq6^%TnQ9
z;0!}*J&HcT90rMh9Z0F&!IuXJ#`q1Mv|=Lb@fFjU_)2saU$N&0k$Qo9A@oArN5pkm
zJ0!B~^=L?JXp0Juves|9d&GPZl7!{TOfDyh```}2w%hJSEMc)ce?rb<<^*r`@f#^Y
zQOgp4s|2%xPu!*iqv}tE1zMyMxWIyY*ls})$qIm2mY_!zsPQraL-+1sq_LKGL@~)q
zA@l9PF*r-e9R0xiFV3#!f6Uw3>@Qb8&gP3tJek$Y;U33|@o#4nYYlEWg4SgvJfbDo
zFBbaiZ-i{8d_#_~e~w|>ac@BRhBR><(1sv5sEfu^i6k3Pm*N<jm>#qHKWw#HT_i?5
z_WgWI*3E`N%RhB>2g~*eLO&>Hw^YizE}LvFn>=ROQuf0PmDe(q^chO}3?+SrlCZzT
zlpL54dM1hpR@L$0nl=-FFgt*uE7<|QGVQOX<p|=X5!ETdf0`_16O9Q>kYY;u^+qg=
z4B_{YmYgk`6MaCjs>j>D-R5g`3+aLC-g!L*D5XFX7n75HXaOcDBqpP{55-W?t%Yci
zo<O}HI0cB#AuD2RFjB;<xH4d*l0F0Mk{%?Zfzl&df;n5P!JkN5$6)zjTT8!5;NTvB
z@9mI3K%y^ze~Z=e<T_jwHwG@s-PxUf?a_fl&b?6fNME&HqV8-gI?z`=m61)y%no=$
z1wk6XY(U?54=90ELu}}ZXRF>9y%pGoe74M+M`WPckx~u?m17Yt+hl+7=~282T(J*O
zWzxxq-6OR4$7JY1tJV3u>pm;_Uh?i3A~lHK_23&He^uhr{iqTbMnsi#;oe16fBv2N
z^KWsk7kn>JFRb#j_=nEOKVKstndrwlCUUVJy#}l}3J=r@@Ub%Dvzm|Bmsj?z{=IrR
zX+HdR)x3bQ7VqFC_Tz<q!}9i`9(gF+9Ucedf$aG9@Hn8d3y8???(n!*v_CxFS<#IT
zfTovMe;4&heKXPD<!A*1GcZh#5{k`m#J(&HQb4Y0cT3hY`^WTrdZjLJ=Zp81UR8(n
z<@Bt6H=V(tCs&L4`^Dt)64m0^W5xggMgnzu9765q8+sDVU{FqYUrN86+)OX8FHcUd
zo3X2pVjuL!r)&v0bz5_bB558|b5C3Kyt817Kd~YLX+@2zKVhUTSD(Tl?vY^0C<Zuy
zrB4x_%U?}LBp2yE{n<_u*BjhsAJp_|qdt7FC41|_?LNW(0b=Tsm(fQ969Y3hF_&@H
z1rwKqjTps$_wx0(k6+$+VxKc_!fp7W|Gj^_{rN^GBdupR(dl2`-@*e0f`OA0Qe?`z
z>97AxzrX#RqAkQM;Fw5%{t5ui$Ul-ioj`eVW3+DMDL4vQO-4Hgh&ag;a8}BF{bIPp
z|1?Tcn>-Tar%}G@x`a{qT%*i1N>uysB;5zdTYgr5po5eVf?x6o|CaB<xPk(COL$De
z6`rDGlr6j<X60MV>sa|^i8)}?vb?8<voVQ_1qbuY3Z_efh*BZC@OXK;wNRERuz4-Q
zjO4vaB1=H)Su0g|c1h!U9uLojnCB7JDaz;`XBb5mhTrk`Hls(wc$#T2Oj(zi<~m;l
zGso9|m=1sNE@Ey*5c5xW&o@<7YP1$##N(5Krs)!eTT&kyf$~hK4s%@apvnQGToYz1
zNsiATEk9FXLitaeXEU;ppyZBOZYXu}EnGx;#%F;^J!5S~Hl!zoWm0HLB`T1v%AJ)b
zOES2;sBHOJknDArRHZe{T1zSNTYS67UZM|wL?3U2V!HSr=?{xhpaKGNLJ;wW*Sg}0
zIZt5^8`+7qM*H!W5GLkVffZAl&S<HXv&I#fMjc4Dq!TC2$WmNc?i?6;v9da*ILu4q
z@I|ty;;^IpSJ~M&?)RGhLu*}*OFuiMHTNuWHdd{Smv=h%eaRsxt%mFDh0axb$&7-3
zvflQzl2J^XT1J6CPt6GB&dopJPLghRBAw?JUC5=z(D?s-pW6V(l<Rm8svcE8NlPDJ
zJgGgi+CPv=fi{Y9jCCt~obXQ4N~WdUOFOwv0u3gZ`cvkv9p=qWt(d#X*^t;U<9#_>
zZ<;jkBl(?K33X2-3VE&;(-ss+cAUw7NyJc=INO^bXv>TSit(f_2xFKq9e)E|5D2A&
zq0s0=57V`ddBi|xF>snLWTA>e2MaIycYwCn%&T64mzSr9I#{_W;WB6I${wcFVFuJI
z<$gfoNib_ulgwO8g#=XRWb9fKQk^t9J~wk(pdtCoOUd}7Q%+_X4ZsXDF)a;$bFSoT
zAkIh07Sn?$*#eiU>)Sm4I-kL0?kA#TieW8fp$6I5oTd}#XMC|hn}FUGhvtOJ$v|4-
zc&~B+IxfSv#1zM7^UahpR;Uz|sUdtW*!v=TrTo9t+96`QMwedblsjHCzI-S6GUpt*
z!7MafG;eF;=YC#038roBq+~yTFNbR>Kd2Q5w4(VK4=Qq;7wHkJ>v#d;Qx9OwSsq;k
zyaqGWB3Kl8VeAr)Pts|I7o{)=kJ1CPUUmG)sua;JUFRlaleX>W2|zOLIx1paS3#kx
zL634wM%2QT4~cK72xAjRCRK7ZjXz>>8W?y-F^??H2<Ew^(9rRbq-#=tJJ*~xF+%Bq
ziBZ1f?D0zY=WXB$=0AwE$^pZIrU`j8oZl9`rqoW2K7LIaS-*^(+jOF&HqFRNC{kFS
za9@)=NQsB4T_2S?MX8OrzH)P7ORx?3z31o>dKmD2x^qV8kpO-OqmZG!P7w}VKx(NG
z=xFZu(?0fvlL5sf$i1_FNgjD+Hi|ck1<B}^mB91WnnJV3(A;F#!3gxuOge^XisS|C
zW#2_HcS-)9r5UoHEvKdO#Ny|&kzyPadsd;35EMG){#3{seTlVOZY$oGPE8v3HLZ;)
zt*Z@8H2z$r@cXhgkW}w@t0yK4gj11rMM&O4J3i#CSL!%;^ce1cVQu=@fe#faQ<4q^
z;2Vc7lLJqIsi|50*r{hlYBBC`>Wgailt0wWN=4}kNNX7OUfp@$s0{PIpVh!OYK_a>
z@$v~(0EFIy=+<+Pq5xKus1g)9_5@^29%L7MoAmTV`hbxTmbymwPn)f!TiHL|bVW;?
zkG1VHDzfY!mYfZLyT`Z!#<j4t8tb-i4(J7!40y*&s-XC3y8z)CNFx0~M+77=Z8ce`
zL8{BK=vHl+w2?GNzrUU4QVQU8JT;L+*ZCbpf@!-kK!DHe8TWxP2E(cy<2gGlNo3t0
z$PE~=uz7yETqD>=jhWONC(t=7C`~7|EQZ%-V}U8N=+~crU%?jBgB7G?TOmGT+DHry
z<`qo0n1tM5d~O?)H$J5)HP^tDSFzbLr83HU$vQWrHmhX7Tu!Tc9HunAk4ri&c#Ci(
z<ZJ~kG`xR~pyxuA5Fg1o6{1v9B<x2S*Rb!;)!Zo+qMVHqr+LejK(FNZx6}PPx}4&C
zQkZ&rbZ+o}vQ4TSw&_Y8D|T~`rdXaYR3O+~&)#c`-Q1?6Vm@E&LNIK^v`;A^xKG|W
zmbufh?t%ix&eeVR^G?3S^dK^|z(Grj+p&v1qSOgpZU>Hhsr1aqy0>@!1PsCSU?MH>
z_C!waYZEFV8|G?rZ=3S9Lx94CYA8oCSe%N>hh6V~^L?4ffwPpU^W72&F%p(L!jA4W
z&xZxcQOhm}w@3%m<Ca+=e)RlWD#-X^_YzDKrP}>U-t2R}k0&h!=5)TJRO8OKB+y%)
zO`&{rR=F(fX0@a@;?rSC4WfJZc8-%Azj{tg)56`5T!}4*Bxg_Y3#bp_{QFRHVk|j8
zqjwH}UL-m2f*EColDiH0P-tGptnl%&rlIr(b13=3V@w1iwYcXwPX}f~hGy#thhe*O
z?Yjnnzhd#b{^10m0(nh8oRAQgB7HZJF;sdkl5=>!f>KMSgbOHt{f$rm#)rQkaHs_S
z7eT)lGJN`D`e*ns;{jra5<<k7bzphCftOEznDb+_oUv968a%_8c^Jb^zQ8eQ#Wq@8
zP_HT8&Cg0g{~a|=VTz^w>kc**uW+qpR@raQJC&v!W86NXwf^}=K9y1qJx1<aJ@t|s
zQpLurcpEJ(>vmegk5(=eK+mbvQ4~9k2+{Iwv?C*;*1F?cFo{JK=d}R5EcZ0#g`U5E
z?OGmN0%2E=Eu|$r_yw1+z=SZ=5edUQnC%a^c6^*AEXHInmo12}m8E#r^ALvTt7xyQ
zr>p3*_`WX1&N^F<gf0{?DQij8h}KbCcfIYU1;E^H{vs*zk|fbnH8y4?gzE8pG03gv
zLNet-qQv==%P$lI_+iU0IWq$Ka4pY&mT-~@X(--d35QUiDfFw(sGey-rR1J)^vr3r
zboPQ`bc2%;^SKHNePSj`GONS`x+gZiw?#vHc?4PwE#XLRW-!MTFHOy`Xva|c`pnE>
z{idIZq6ZP|;|e1R_dh5tvxb3fw(@ICXuN5OH2bri(|kV=2ai3wCsyewhJ8v7Y0z`<
zmt;Eh_y2)WrvaA%7Y!2vH!_z|hc6VDuYxWomlT*7p?~Mok8j^T=&XLNtMjB-qfS@u
zK|H37d}kdkl(8*qO=o#3Pdq;%ZfkiI;?XV2zwJmWIvGrE-&DU|-X4CRs9=3mE@1QX
z=Lh+N5EKT}1fyJR{`jT&_2Ii(=+GAd*caM4Ye&8S%?_f*Sx_FZ%nx7KBCIdz)1%X3
zZ8mbg<bR2gC!zZ``7iaislUv%AI&Pzv2mo8=TY-$_Vx4#Mw4`WO(JyrbhfXF%11lw
z%YF^HF9mQbeIclTuJnZ`_K|}$h1h?k`M%-??}y{&!LV0hVO$%XiV%i<^CYpK`lN5`
z85^v4bLbd%#W8fnF}`!me7sy+xE#%juyX;9LVrG)fi<+e_~@1IAuozc`>b6mX<=M=
z?x}UsC5PyAROb~Ms7KXjS;%~UEtU{H^rg(ybaR%LHV3qHXnV*%Hiv2d1l;JC&pj?X
zm#hJfH>#H`a5Qif-n(Y42@0UgpEyz-eO8&cQ}^-5r>1)HaggiKC@`*;YxI0(<TdQj
z)_<)>uFu(##O*p~z;pYX+#obe+mC#kA(LxgHtTQL+QhlA(g<**|2BzHMIR;-!+wbv
z6a_z#__SXlaWut=*EpfgdChR1HxsA94r5zR!!{mi8W$8_N4n2D<-I>rA}3?MCK74h
z*deOvI^ARJ7?h@XVAP;?%Gx7MgS^*qs((6CqCsW-t;R;fITsHwHlkCA;mU&AB#}r3
z@>eFZNuouC)*UI)LEf^~+}F2u+-ee#w76Gx{XcQ<{yw{F5@YSidA!;+IU1NTlwGq(
z=;s>JaDl(ih>TWIG$QhcA9eFz{_|UIj4?ESrly~}VRZAq=6l(euL$N9nFJe!nt#>=
z>&B=UViwzruo%;#-Ynuq%>vJd?7CP+64BOjSD}Y&AIQRqH-8*qv5qo~Fxd?VYZNv8
zlt!`98Ef~fu2{>5w4m<s<5m$jhAxW*&eS_Xw_YPRaB?E6yn%m`=ORNCbpFfmT;eQ(
z(Qo3pw43J=7^W;E-sgi^A*iGo!hiaBdk2hKSu;Rir9E8c+k1v24mJ$HsEp@&ONZ9N
zXuWcTn+U7@!h{u#N!!)mOIWK>_Yu|^GJ1u+-a}YTWXJ{K3lUZmrF27w6V^)Z6~bDz
z{udC|N}~e^Yb}xPCakr@a5rIHpLIo8>&E*D>w%`pg!Mr8Wx{%(#7tOgiGP<h!g_$Q
zGhsb0Y6@XJ&NPLv9w;#r)>@)nBdqruyAaj`Cz1*4derP8tm~%PLs)N1BZaV5t&btB
zlF^*SBMGZv3<2I9PgspZrXb-SqO)1Vjhb<YXcNB-VKtW52)H@I+T*c~jK0?RevGiT
z5Z3*(x@Ik{?yEa&m2hL=Qh(&s0f*658=9bO^bY5My#da+p+A;=IR%$#M8a<&V>n2l
z{Shg_csBt0E+e{-$?K(UO>#6yi5kTADN%VMxYc+$X3jKVoqRxKCKYd>)M<o7G!nK<
zShFK2f<R;U3-gn#t<0b!CC)Sq0-bR*(k24$pJAlUyPw{-T^KPkCV%r7#%m`!6%6qm
znR>Tt=N-IX8^thxHIc@neEt3roO|~lx%u|i{rbAS6-C&~Gm03)95lwUBA?%{8DC(l
z#7t%U)?nFCo@S>&40t(HjHd8$T|aA;12I@oL1+E`zpfweUK**^Px4IWn$!j<y=Kmt
zDM$)uKdig-l^6;7VSjzbn$?g%rcaTonf|t_HZq5bWF}@^SyA#EU|cFFE5mz5eHGEf
z6WTmgYrE;`?uijY3<^$V6ras?$y&H#$S_z-a>4Vw9!w3K?|CV<dbF#vEo_NA2DPH)
zv&^SFk=IIelyB+ng{PAZxu<B;_rlA`w0jK5Bg}hHdiw<PC4a+QI%8>0Y?@J>7TgpZ
z-(A$E#YD6J4B@xt6we!CH8GPbbLnDprW%4dVS3x*oL7e)#MmGiwmQ?xQgY4DCF(A@
z{uhc_ZcVr*+A86;{kwFi^FOIS;MwH9mwrv;9@qy=S~<hor0yl<k>?Pjl(;P;u*!zw
zKLT5}flGO-sDEfL;gP~M&F|EPFqHK?AAaN+DE39aQO!x-VM?<p{?KhcNZ+3;UuApr
zSw!#+oO7&E+f!fWT0m^8Zt!6cm0=K#8!JhU6q?DK4<3|9ST7_oDiO4A<wIu@FnaPa
zLxZT-icMfFqmJ%%5sKL-kV`q^3g0jS7O^Bl8=$G=kbhTCi7m{TI~S5<&2$B&YR#0w
zQolM*@L_3zv~JW@VOLlrCusoUT-pI#q8-&VhRBjM#*!ARX-pAg-B|GQjzSm+J7k1d
zNlS4Ei<}OG)49pi_3QTWJ@Fcct>XvTzG!0l>Hute@kFycyR1!^uib!G2J8XWFh<UR
zXmR|7mwz2{q402PFu1aMs?F1?Z%S?M0Bhu$+{W?SO4`xJC{?f}*DO?;Af^L}`;C!n
zR!hUX`QaHf+K9&+Lr$8e%OqxH=3nSV&j5Wug1@lI(-F4tjP_@Yu^t%6Jj@B`KG+zf
znIpMnZRR6`<OIU_1FS)E@_=q}{Q0CsWLjo7wMKoUTK0cAjfw`Z;p+YS+*s0;eQw;i
z?C)h`E*wl}9I&uSN0Uz1itdm%eLV*RY8JJ9o&gp3SPfcDbCW;`-4H5)3z1c(9Hi{f
zpS(WpcGp<Yy0^w{6HG34%qMs0h;_8)?69bJ9f2;w&pdBVGj4=ETMBy{&P<aSNjN~U
zjIqXIC*WE47S}8)o&ni#h&#^LfaORfW+d);-tQT}4V#YLOV9UPFj^G}uUT-(I)*c}
z*KXdg8}JAi;Y&j)`epz5U`CAxxvZRM-{7)>GRH1kIf|+8OgsMq$iRtOmjM?I69P9f
zm$6MM6ahAuvEm&le@k;DsS&>GSJZLlAeMwo9u9>fq?WW1d$T>a^`30+UOWfy#ed%k
z1r$)PuIkoAsMOt6kR)Fw`GS(eAA=lzet7-%@Y9FJ!3sxU)$r~20r-LB{~-CMp}_(O
zqha{=efafBN@e(Z<qM_MnlH=W9e>jEqXx$<;Ds-EezXXwf4x?G0e`=Q-??4vP2*Gb
z5-Di^lqgHSLQ4x$2bz@VUY>vZ_RE7z>`UQI_!>U?U+>=@{`=5RMq4KU?eNd<58(l7
z5Zb9hOW?<D`19Yx?+<_N$<|moV63DuL=9Gm!ui1%FLZb^LZ!o`-lasIw)}X0G%^5!
zZ+yvd8Une0fBE<4A47NNZ$D2nV9vdweheiWKuT+81+rY%IC^V1x}4%jDG+G<l^iMM
zg}1J6jjTGv>O?`}ug7^nBRor~qY5iL5Qqf`KSl@$Ms|?(k7(3G49{NDeH!m09%fTZ
zBZ%ybO;}`;mx^$p`qqIzQX>!i$8e3jW)(XjLQR@Sf4CMtZ6Ge9UW7O)6$ZX8$N6A&
z)gESisBX0%py)}=R1+{@&X-=CA4(^T(S5Hn9~p+B@K2`En>#%EPaj?`+6Tqvhs~=Y
z$MZCe@&~Q64L_?^CR4$%xNXr3IqC)4q^BRpjZ4r2?|RB>#)b93OWU{BGd@Ng)h2lX
zb9{MAe<{GvMp{`nN}@o58DDBNFJ#j?m+A8@Nq~xsXGm)!kd0PG+N2(Vl1jfX$x0+y
zhG${cw1h_1Bx8B1pa&?L)ZD2hlQ~f&2z-%Dn2(=qRpbIk8#QvXA~_?#%&4vMTU>;G
z3v2v@{G71T^y2}ZUe^j#@X9^Gw`GL@D-=-Qe>xh^j}!hbd9eFUMx#(#T{|1?1UkDk
zn={@!`NB@GV7BTH5j7~-y3QuFY;Toug0Go2My*phLJZc!-nMoC1({~)sI*j)2|!bs
z2fx02EKe9rNAQZ(Bc}*sn%xhGJCtKgICgItUNfJL1Ycd_086xx!IHn6k9=Cl-c|x1
ze+pi20pA}HqgfP2xoxbj2?8d9PB^D~^}C#pX>yd*a<~$#jR8>8G+1$Y$=CVyB(-1_
zt`}5XKq`4aGigoqNtf&yR~8lHZlR}o#j1*N@A<T)o5c-QFFS^o;yoNgM{&n7w5><S
zP$L#4<EHFk^y(!+J&dH+!g{=xOYelle;Mv)GlrZabz?{Wb(1u8&~m*a%+gpZA>9mu
z8x|LNB@F62FRqU2{^I%(&M5^89dB#iDFq8%>&893ol>Q&#y#?KH37;_t>*=%<rUI{
zAh&>3yq6=4lyHRKD@g~j(>APAnd(mZ5?9c6!Z90x;~1-`b_+3`NeRUb31mlke<y+L
zT2JGTnB^$!2p>)hu)e3fqEop_*0t_*X@(woVa=4pXQP@|LX`;aX+dS0eUfTQDRZk_
z8aUxRIzDf4aHxbu=qav+^{FP;wJr{ZeQyey81m63)^|3_D2g~ykl0h!B`v?Ij8ZH~
zONHKV=g%rZYL5J}jb7z1#o-+lf0LZ{9m`RK9HGimC_$eKjMtT;#-evm<PPj*W71O&
zFdkOoRrXUYq#Nais4b_5J;k?e*W9)yc{_edFq)-|u7(f&{oo@j4C#IpiQ5*xYyEm3
zg~?FPXWS{w>TI$nXeq2)<+*prGgOANK7YG+I9;FWIFRr6vHdMo*pV}5fAW3`jPz8c
zER+98^^m;jWAzvq$A6Kf=1#rNLTbxlS35`ufYc6}W&V+LoGsOZ<vUw8K6}@8aC^I`
zTjn;ASW{yymC$mI%ytypHon7h%t|1U6%x+6m$c#8Q+d0czr5Hx%GaBVye~1MaH2v!
z?pbt{)^g}nPw!ga!)GM4f6<4eOl>%gReAZ4wzwq<(R^aE1IZRdO?aJx?KS~f+&I6~
z^$hiNb-%PP3GcaI<i)0R>3Fh}_d&av-cobm$U$4|DTe*ZU?GB;1hc$|Q%$hz**m9X
zF_F_FNqC)IUftQ9QRT2#dTOX98;7q|g1Y5Q8)@Cc24#I_S%Kc$e{nrh*3H%piEV(j
z_oO_=ZK%Ca;O<E{#rvo@^CABkl!4W-C;&~{q{ShgucJdK^|8_-`zwubcJn+@(T)wS
z(*609XcFY8ym>n}I_>vlVq{eZOLe%pQ-6l1Ho-3?w3jY<TRLC_{0Y_P;t3|jaW{t_
z-2o;ZBF-uuV9FFGf4E&7&hht!Ls5hr2vDH;>u+-S7eD-k8;2C^f3fNJGL9VnGyFZA
zklG>aSO9@A_cC^iupE@|tn?$<i3IT?2-V`lB(CHPg9-NAWLaZ8^>nvRfkD_()l6Iq
zGoA9-2uhR9%fWQTawGc!#%Zp1(BRAg>65_FoIg~DG2YW*f4+DOr#Afc-Ljv(2gq5*
zcklowke%x10HQ$H=}bujh+6~%4*+tS^BqLkRM2R4J^_(nn$;p8v<Qef0LW>^w*gU}
zlP6BgJ_QivJsV~?U_j_PG|qFqgN7qv06z&0$Dy2T_2XsHg1j8JAm=&XK_rx<=#y-!
zvNi_ZH6lLEfBp4lL{4+QgNPZu;^!b@495Vts3Ww9h&zDDY0mc$k(|xn6doNWaa}|h
z$dYttBEmlOjKU)~I6>YTz4V5GQY5UIJfa=*Ms&}Q8r@U10bW(sVvvs2wrJcS&pqlY
z!*nLHR7&>Vc~JbR?UdQS)JMb8IPo0@O1!cuRw|_vf06I;DNP9jrg{T&z)$Vd-uca#
zPK>-7HF(Yxl)<J3#i<uDur;q=UcPQkPiYcT_p?hIbF1TU&?%K_$9dfz2U)?bTNo%!
zeF9~B=uj5X-4rj|Lx&eUEL=5ok`k^;7jvD;RWdB^uM~|dnm{k%Zv7m6k#?=sGu?qk
zcWk5cf6ri^It^RX{iY{6(65+M)Jt2B4q-&9F*5kH2)@&dVBA|#sXYAUAB<#z-`d4a
z9t41Id*v=GLgttR+75r&8N<bwb(${f>(*RCsoQ|*C!;X;(gXVst6f`72P=SMM^-Um
zy-zFbb$g_c?x0&3>5;wmK7b+L4ToY50H}$CfA@3*LQRe>u4=o$*SfeDP&UMnkrx{M
zRg8o~P`q55E<22{U{F%q1-B6f6hPQ^*_WjXTZD1f(xq2B*oZx%EB0h%w2*RpyE@vA
z6c2!xrFa0mEXBGDeZ9uPf#zTeRddEDWfaLc>B;0QIMHlsF3*qS!NS_iO(rm(MXEBE
zf6<u9`gB?8>!vLNY`AHSb{zR5as4uDd&Fp~+Aud%D)VdaWPTG?(@BsuWkpm%DOMl$
z*OYdnzr1WaJ|q+#W|B=OpJwt0)~D&ny?%M7IO|Ckc@`h#M0_KQ@)Bn{p09P|2vwmY
zQ(*J{tnq0kZPPf!9fz4W5F`Gz{7aFwf6-pmZ8co_nUvRgk~WUo%2G>bEvHS2epNtq
zF0Y*E3-i7d?#o=<@?1{qS4Om)0=<kC?8%+G<ygvbEX6hsg@f8>_6@8NE^TXk*|u^q
zF$WkN_rThRtQz5<eI5tSxiwPpr;lvFWY+JAc-Z2!gAHeaI@h~m4mlTe$_?z(TiliW
z=x~S_>M?4cP#?%*s0RZAHI>!KY&(b@d~qTxoM{XTb}*cYDc^DCmfenKDL^eP-SCtf
z)Y77DPexjd=vYU^kJ)X;O*k5Bnogq|D|ux4<iz>^omP?%mjM?I69G4u;Uppxm);>I
zDVJuf7%_j{@AZ4Eg;mAc$d-p+4)XBlhvUn`4`(n3Cp;>r4=-O2ie5<iR+1iCJUFF<
z1$B7&cKGxtrMC1r(#J>%Kp#i<p04=x0hFf^>X|<3>4P(h$AhMiqWe?)UB)G7%y0E`
z!T{lsF-!Wykxog7IBA?BstTSyzx?zdnSCjOjkkaCGXMJc^6;OBc{0YL@X8;)e0zv5
z3@8C!AG9(;1$X%EUx%*`f0@x%Vv=A4#xJ`2?U%#!<gYBxAW5FmTEsz~N|KPx!P+1o
zzRL16K}sa~A<4&(tdQhsNKUBZvthPBoV{AbY=Tf`i8*2=nHiENiBBR4)tvKlqnTQy
zg86@wo}YRAz(pMY6qOlKq~Q@98@h;I5-E8~3|Gc9Nu8|9q8C82WX2LR5rZp4^E8Et
zf-GN2zQYreZ^ev282yY{%Ryjuv{fQVH;;w|A9)sF*=+62d=a)}NE0QcR{7G>2!m?D
zmu5*9B}l&h$kN7bNgF2&xGHJO5H71c{XTzndQ<sboHedV<HQoMj;5;MpsM8$PkPN1
zPQVJ|eAP^xPhjIh=EC$=w4pgGlwDxk<oEpa!OCc<5cT-+4*hxIq@C$0Gr64`Fe)A0
z$c*sES0HpkD_mf->n5g^Fg7d%c>9vmI*33R-cU4Pga-YVvH>rQch}0sXhA`09&LYQ
zF(h50RCEY#d(mkWs6t)7x4R8R8~{1@3DcH4Ge-EZAX}{6L0*~ZE0G<Z^khrm7%|1A
z6@#$EeQ!EJoo}b_ND#`d(#LK39)xn?E$9bIZM<KlZ#U^zuRTH;xo#YkT3~VQaL2l<
zw$UTHWE$gG+X8ut1cu{@YVic`Fj{{fV~v@JZG^fi>X9?o#UYMW(ehH-a>~!N=~9$<
zf8r>89Q?f<7Oj0<?2;e|Q6IZZFGEh#3P(kj=+>w13{g}w)+=Jz$B3I}aXI(gN<3=_
z80lxOuSzhhqaHCtDG$z)%)4;i&m+Zpd3(U|DCo1$hFn<xO<02$lyuKnr!aq3MuXG>
zBrJ4yzinaA!WwvE+nTWM*w(!n(L_PX*JeaBLWsU2R(kg%<_u-VeBN}IB?A~C*o<!u
zCmNN|xX{IZG}YD!n*0Ll_GqS^7O3?q{q1O4R@nq~&w33YTrgLN{}!wutw3~dXnzn3
zaiup6aKjU55a^cLeT66RF(iMQxAp`h*6-m7MhaDK?iC9ch0|t%d1v8rqCZ01y>N{c
zAuMKgY2{!L4#NWKj+JY@pwBA(ojjCAOIT?9u1FgMUL*e77H$rmTjsJ}@Popbe=J1t
zie*MchSNASB1-#;8!AKJEo?U<xLlbqUXjR2PF;`7Wa)Ne;bd&);zECzDJDKa-C?Xq
zWsfwg^mmN4nD_{l(#Q3x8sqY^t!a3nDV9*}x3Lx!j2B+EW*M~BR4tXx`CcAxFD|)E
zo~RT(QR>ask-ERY{n=EGl`_U~tF`urU40Wu@3yz5*y4-0V~?H3omy#E*Kp!Etyt{m
zP@e3S62;|Ab8^>}5hZ`sn4piXHJIe>S)<3h*GLW4>$2alQHYf`_yyGMYXo8$Zq{|b
zy+#0?unXblCDOUT8`z-52F@(mSSKGd2)&&CD_k$o)%dq`y(7jeT`w=_w{yJ{7gxI8
z3F;NDcZL2Hu6KmmbG_OM@5A?Wy%gpHd~eS?5XoAqj`7U%dhUOnuKGhdU1x=Ji|JhJ
zbJ=5$z~8~;svtnCRiW>ATrDJJc@xk*4ws}z7*?q7`CAZxb^>~hzm2<9DlE9zaJTH7
z$HA}f_tiBfjR>x8+?OrhR8y9&ubQm%GSzuI+l1J^(8H#4UNu*3T^B5l3?cjhwyxK=
z{uNzScU?bQK=prd>{Lu=3n13z<zzuY&?N2?HBa{|7|E(7goyFeXL<N1z5Iy=$Bpm*
zrJ>&{6?ypg;cxNi!Ue>hbU=tH)A6YG5X*xWfnq>L+fX2c)IYr$#g&|C6h7#Ev}7~^
zXT0g?#g>)D#vBmxAXH07e(x-tf7J^3$Y*6}tUB=-7vz8Umapa9i{WeFvBg@OEC-vc
zTDn`D7u2N4w*Y)6H7QD1bUShMdj{>IBiZTc#Cm!zvS(eytLLj@9U)&Wk9Sw<>27&m
z(i5dZ#=L5w--_Z&-$rPi!Fzn$>_n4Ep+AuO4bH;aAvSQd4rDc{cxcTWQ?>B{jDPHM
zTGevgWK4guj4-b4Nv0TD;?H<stKi!VH_0F_9lDF9ppm%XmyOgYOi}Vl*~hI&8AqAN
z@x)C%50r4&vOMrftCHu-3TNAs8bE#9L0-h!luK6L$`aKqTi?3b$h(6C(e@#kpTxfz
zC7U?uBRm!=*y*vs_w$)#&vQ#}LX`oh0j6XUuG4>ZoY`}om8me#U9RK1F8w)=P;k_p
zHMv^ZE;MaW80?S6^`gqtFDE9G>9K%}W%E_;ajcwBm+_I^d>2n>BA44q9H*{K<2blQ
zqweW1SdWROZX2^{qKzXq8(W`zWFFarcQlI{O~K$N<?_i=mpwC<Mnc;XDKK})uP$!i
zDb0W97{YIp(Uy5E`6@1<7$-wLcZU`t)#F-*dHKw$j-JxmIj>Zd-dpMS&#dl7TPoPd
zTY^O4lxfF^as=Br&Y~&ZckPTRoX1gA&na!G>Wo`~cG_H@Pqf4hOWzJFE^zTEs^)z=
zot5~5#9xm6F78t(StGX@Vt9R?;BYywLIr<Ga_umW((Ci0n7BQyA|lE%Wp}b*Ld9a&
zDxO1tVOqA^y;;-drKq7i#o^F)f^1pb-NGZI)&P&i+tptbOI#OAYG3Z0O}~uV@RYn<
zc+}*K_^By(*Epj*a&5u#8}{G@H&aZUnm$^~>4Z@iv0@A&y+3k7d8|ogiXNGkUWtGF
zg_`N2X~)MB-nKY6MA1=dIPl+{&`plSRR^-rx>Ko2?7DSPGL$pS&MH5#`N@#K_%gdX
z3c4tNEKg0}L*NJ)F(*f_Ds1S^m31+^P>FvIK(9EpXUrF5T-|<^ZkNp<)9m6k*l=ZJ
zXP%#10g+^%B$4LTQ#3r{l%H$XbEXKkB5Quq<7jV}inSOTNUCEN8{A_T?DM#3PD2l6
z%4#yi5LivK&ysWUsA7YmgvYkVD<|s~&M?tcryMZ{*yk>fI`+I&dFT1q@mzbq7BkgR
zx&=-ibC>P47$p@ueY(4T2p`}hm!=<MjbeKFaaEDWmnXIuAr-x(Kp``A99Q2ArzJor
zP_qOsT4tM<UA7n}e@7Lr&lTgsj(Sb5dbkW%5?((%%{IPjdw@c@mS<pJieww;X2A3Y
ztMpyF#?~z3wU!a{A7$)<KaRgyTB8kc#9rMuQz$I@8QVrNq!d|4E9CS_HUX%MFH65n
z=h><z5mx>8Q~?+7g?;vubcZ!>;u-U<D$}brn8uYqHULz9e~M?zm^U+*eth-82z`!F
z@`*g|lK()sDqiQP09LP+jd_Ob1y{!1K!>|DRv%)<udV6xZ3GZH_(A9IzP`Vvv7@Nk
z!sa{7B$4N?XHfDJRiaAu6vtY4&Y4#22<`H(f=!3<J7(VfcbEP6p9iYcM($g)!41uZ
z!j&01WxE_{e-ng3J8)D4>f9;>qA+rV_?eXYewPd-P1>u~iEogxMi}d6GHPF1PEuj(
zSHv}TIow9$arm>J-R_56*^=Co2zZ<7-hO^xb|(k%n6I|;<ZKuBj*C`F%VV}JzzeJ5
ze?{qCuJ+&kc)<cF4ucyYK3!CQkY<4R`=8!_P}aWtf5$#82UOdSke>@JT!$cwmSKSS
znF^lwAGCAte*JZy(GZ)0Bk=xm0I62UU04pW^`reHk9{(VqA<=>)UG5zk+%Uq0m%!j
zX#$e!772hR*3^v>h=2Xl<AnAtA)YjFQr##5W72uF9*9gJnL;v(T^?=xuQLq5Uo!0e
z0s-em9SFB95b&|pc+Ua>vRDUe3#4ub@%7`&!~X$zatW7Ff;JNZGdGvPzAY00GnXMs
zJSdm$xEK$A-Vf~T05d()3+yd{Z;1~{unrI)ht2;#EG1D@5>@XiAKISomPm=D$d44O
z`R$Jzzy15e>z5BdJe0mU=8bUj_T}46z!x0;6&yaa=H`T8R*Bn}zivN2aV{;qUg4!V
zR|;MZ{`T;{dH$?~hY{ihUgG()(*nmU2`>SE-{bp#GA_+F`CIvtD5&s+lsUZONGG_8
zG+~?ou0lV5{qmm=9G6emYMo}VqOw*UI5$Auq_w&D3DDM$=g)Cv@qhdj)~{h7`X&Eu
zY~u36!?{~v!Ah$e$(X>nRt>k0zi`7p<ylrv&a#pP;oDii{`1pkVeQA?J}okImU-_w
zh5~1Q%_z_-CXMfe0!lEVY~93dzs7ABwL~EplSO4z{+2(=x8|jcPhyoxkp6jT;}SOR
zU1(STwzy&G1k+r%Q+{3)uY{q(i6i2b-ZP=e)US)Ol|gk2EV8u=vUOz)ej8;B2SeLt
z)Sw~1QA5L6j+${e;vQ)a-&!263&(U&<3788gJdj|LJx-?*J<wckue+(lBteR*<sw(
z%c5O?NWnf^ULU`$m^5dO3%b0t8C{!oUeX{iiz3vc;B--eA4O%}%0`BnrffRB)8#K&
ztAID+*RA-?>!J{pa?A?b3gO2^At-H_4w4Fi5AJL@qxZ7)#<E%m9Nqc2-C~cfu^CT)
z=?pfRHd{7mCJ&eNu!Wlyu;{YCKmS}l%1Sdu=X_f}!kROs$dpx=EVX8mhj!MlD~tlu
zu4hzS@r)3RVa6&ti?1xt!YpZb$ssyUXZOm%8AgS-;xb)eR@%K{fQU3_3JwAl3oh!t
znE|0TEiHg3e4Znhmlohyz=>(t{A+Q4|B74Yyl<!cxX48b%`AM>Oj(Xla5$L;Ak|tM
zMyO@0Q%W)w$keZ&K1(S-{$rV`cKyLv*%}zD6k^xbj04?~gY?^`;0ySX1>@<~mdr?j
zxds(xPtyz|53@M23%R_lzyWuuUBU~VITQ8>Pjk^9^KHmQ$0@;13vU_L8*iC^1>c9!
zLv<))CYTNJKga%r3UCAj%H?=(nF712Ij=}lcmQ<(e48oD(iAW>n09|#aUcg%l)9a|
zf&;;G&;>r#$T%ae_+FOxh`a{LE#XB(QYj|;I*}<<lMZCYT_sK%H;jtkgcSfHz(Lq{
z)8AISkz>-6DXT$7X~6`UwBq%Dv}IcMe3sK5Wh8UKb}D&6aV8-ziv)sbFuNq;!2j))
zhZoE_Vr?tF5{7cKR~`^TsZ~GC0)Jz{e}Qg`TX=ak{;8N`Yox82SGhKt`rs<=*_kAh
zmE`~|q&F7Fi7`<cP<uHLZaL)?fB^t+83}s9-K?dYQYQ@1Oj$`eVW^3JuBM){(beea
z^U2h&%TOqSRe=2lumBH%#IG6H8M%a(%P)}4eOk-cA@ONvE&FL{qm7AO!h`J3Qh9Gf
zL()Ueq&*eh7Bs{qGf{@yx26PV+U~LoB@~m<rPfn6q9GY`x*5BeK||ocMX*%KYiKBX
zV?#`11r32a28^*sc1AOQF6=JZnFz(zkuw#Mq%&%l>;RHTYJ{r?B)VbGi6nksJiGwj
znTUFwJiKxaNkS~!<HIzd5f5MOkhkWL$Z35<-dZYVsX;9JFo@=m*^6m0EaO!j0w66h
zS=#kuX&1}7{zzHZa_~qEqb*fk4KYBsv$j@sJ$Pr`F<h(a0@T-koj6L#(04#Gi(;x|
zIEZa1`S=0&mlEE+djGaw>J2)%uVzcjCqo9GvLC}UY};TG(TfKdN@NfVV4CZ>0+=tQ
z)3lc`m_(6T`>v5-Dwd@}SuC0jk{eAhBO{Iw{4pj@iI=Q(GZaf7_|4065E+kswn`wU
zLH@SZV%nE)^rVM>w5vNX5GAFwO^x&wd3GpFcLm;p)+)qRAvPoH3h5G>%MwCeyo~h7
zl>&|z>Q$CBv#00E_og7jVO$t1xw)$QEbLR0u<AN75-O*!GbmUCp3z(m<?ryU%oDv!
zO8S<~HAvy*#xi_Ctx4au**7MFsSN)%y^aXr+<@8TTeg#b%th4+x=gkwq(YkSg$sYh
z+cSkPlXuSVX)WnCjlopqZye%{1xF5UFcwPw{^K_D%}>?MTg)oF>{j}C9&gqPhW9iP
zuG7@<!lS+<NptID+BI@6j$QUwA9yP0O<^web&?&6=gOAky1vL5t*g?X%CD#qEu-Io
z2$F9<qs)1K^?v4rDg}VW47rv-?Q)(Z*m~h|bE&wP)arE_jldjW6j>dd0VkhGWGLE5
zE1tc4%HK|0KEY=%4_jq?9Q{e*QCqhe{rPZ?lDA*60+d)2=Y~&`uYPGc6rYpMx1F>i
zEg)G`od(A>L6pdrjrN{AX?5P)Ycf0wSx&iRQ{L5o{x7XM-=XKzpZUdc3bbR(><8Lp
z_H=fziq8I|V1B3vc(hfEw(HT316mXW+-N@V-@nPb_b(s**WVy&rU2ah{?~{2xsFvP
za1-(F$^mr#`@gquAAY1%QOcZg9t}1A1Air}Z9e%48y5lxc)+bS4WbZ`v3Y=G3gCgV
zZ8Ly>AHymM&$XI=cmS%HqO1XM=$>+<T5iLz(2)!)Dq2nhs<aZGzi)4apG6fl+ePMc
z&xtny`0W880Vsf&HR!WMrxJ^F@eDc&vHX){+k8@c0;3pYA(XP@(>)kgu)rN1t-5vO
zT_ikZQ<ulyb;UZZ2puOOCf6moX5j6+bxHJp;BE@#qOYHkv~xgl5VWmAQvp`@(E>`H
zm%mOaWuA$(Ooy@U#_d<1%E=a`G^0MPnGbZZdf(3$3B@Z<H9)8oq(lN<HTPYFi-8f!
z`aEUG#DX;6gK{t|X!SiPqnP18<J6Uei<k*zb)A?AWpkaFucwS+R=L*k9mGs1o9o1X
zOem}C#C-jlQOq)8PxC#*OekL{=IbdJF%!zu3hV+$?AHuq-jD8{0bYTyscvGxRLv(p
zO;Cb3>&E7`m|LbiTUTgYBVvCHwV@stSSD;oZ8jVNrnMQLLm+i}KROUH!b!S#(M7&p
zL?z%>fU|_^jDe1gOaM8&>$ZMBtdiJ&596G_NF{{Yx=NHj1PLVffch=vN1AyLD47y`
zJ3-K;$|a@frWFgBleVD~;C_rPWncj)Tqqi)nN;#h*`OJ3l#Q0mX+cM8UJOY#t%ucd
zSkL-qQ@(OMz$beVCN2rKrFSU^Q6d*C4-=`5a+i5-QOoS|%c*X;oG>Nf;1gkgJBn{7
zC~EL=qN6x%N+raS(W4pZxAVHeV7soHIk^wsjjZyuvYi8th){b!5xdY?E!qW*bd>-7
zxNE!2dIs0FXg#~)Djs<oi&mlzt2??-I<pNTwKfo{>)|!sFTH2_wUJjSZild9%E>id
zg%-!fZ$j#ABou`?PSJV>ua4q>W425oJ*ZF#IwJ#<T5<vp@d3#xVd5$8^!Gk}qaYRJ
zoMv`KZ_SF1=`!}$+HTY0IG&AV9wDw&rHY<i_D26UEi|v-IJ%8$d_(wf05{|LHa7MC
zUVOe=s$1c6cfRij&o`b%vWjUd9Zp(1`dTaHk}0kzk27W(OZh@}6i=9csR&F-(MfH_
zOe3yTvqHOHAv7z}eF@#-ybByoT07%lx~#lQj}~kcNJLBBadH>RLj(S>zFW_Uu%#mo
zQFg_MhS7jAM1QI-%u#d(ob}~VzL(sTqZBHz%;0sM>Es4|yj?k#s1J!sOMULC+ZdIM
zqN5LF!9(h@jc-Xyhz$6DG2Ss0xzO>Rsj~U_5RdNme4?ZyHRIAWB^`++oVT`vV(eU6
zmP<=<LtNr*Uk=8l<WOcWsH9K{C6#NimHV`$o_$wf65@&Au6abo57(pz@1+P)Ax`8D
z`8JqPl&X5L)Xsy5k_or7>qk<Xdt=>S2AA8$;YngQw6Te&RcY^k!W!N%cy(Ll^Ef?t
z241Ptp{#y2;GQinK3psxlNJz>meBrM3m`-<umD1Frv;GNP4f}2+Eiw>J*DdB{0j%i
zv~6SCw(7REplxTFbFP~a@^3?=Uv#k8yJB*0lf>qTWDOFnK|~!-TY?7-_;9f$v_!A4
zgqGr7OK4}WhXJ&IiZEJ<!3LwzBC;j4=Rf;>Zw}n9VE#=-=Z@C^d|;-NoRG%}qU)g)
z?QBD!F^-O|Th9)(&S2s|&EZ5Q6q`Ae?>42ef0R$?03ZM&Q-jo-PbRR4uz|?|nsPgo
z>rG{WvTnC65Tb5W%X22%$gznL0S5UsjHc-ce}(~b312mT2TuOdL|0erwq|^KEW9|7
z*TelZgwk?94Q2b@b{LgTY<^%l#c8x0X3V^l(#h!T3uNif)T08D1L+*g;UX|y<xH;@
z7I^6@F{6ncu1BsPjbQ!O&y+2(v`Q(`%J%VOu@rX*hn>k{e8$emUYw8PZvQ$NBBcC|
zGQ_}gN`$3<rbVgnY>h*K<9uB}#9wL<YJ7ye>SQk7Vq=>s=H-N<Qq@x@ZtQJ7(DOnl
z&z$CxS<{`7b)fH~jx%~&qep*zs-QurtrRrK=o1AE%J*$CaVMjzf(8|iG_Wmmm2g^W
zGlj81IPK^&!XbQDgwu{*5RTAHD|?l2T55CQwFsww9eqYPgzt)Q+R;%s;(by$Trurj
z*SuVshkOOc?s!)BNBwd`Lb#o2z`~wyLUBecrRSUOmRCQ)@tU7?h%j>6jsgTv;p_7H
za9<=Q60+eg>43CyMYv@+mulnI+QIyokz*TGOKYx~4I2{(<0CCkxx-)^SHzF!xndL3
zHm=xz!*rf&hMt7!Oy>onN@90y-_!veBM98K1}x;9-#EfM!qv=N>L;Xjo+)Pjgy~%3
zpOM{ZuH}+)(^gAw!Np3BKM=;f-na&=C7j=Rbd=g{ly@wVad`pP+zheFp3Cj(d8QC(
zmD3xyiZ`?qO1{rU8%VH>8Ou-Q@v)aaV>wV$BWQ_tVfaMyf)JmqzOEbC?!IcikuHa}
zme{noKpmD|e&%0)s93A=E0qT>1@kpu&E^DpDG|OVH?;H5cw+1-tfvfiJNeWk`TzaP
zhyMeeDm6!!0T&Gu0XLU%x;qq?A&@C3f2~-_jw3e=z56Tb2U;yk)CM#f@X{V6w+Va;
zatbnYY6Lz1e+a44?kYQ84&^FK6t_o8q~!E-lGC@%?zDM5YMh+#2ArBsKPI3FNxvXz
z(3;5s2&-W_eNKPANvSM7J9<z`t?6;}?dfgaUo?24fGs`n{^Af4y;k%9eed~yf2ND8
z%{;5Nk%JCziL<1KDIG}7Jc$l={C@xI^!r98^`!`wKlwHO`#5dBHls1dXn{UVe}8Ux
z0yPNj)db24qwMtSpXtZu`^dKSl?U>bwAPr;SBQeRS{sDsnb}o%l+tWxg#{92`T6y8
z8b<#3R&_N_SffX7N)D#qz|mR(e^!kg!FC)AwWIZ5M{0SD)T7D93r%iHM22ue!p~4t
zAfav4?dE6kJ?BTvcU}Z#Mvib;Hs5Oj?#wcz7&mO0pDko?S6qTlpmxP2yO}t;L*DL=
zj9t%la7Q&2g>^7e%iSF?B&YS9-~kx2V+W4N+XcUuEeb{i)MpkQZ?;HDe-vooM)?Yy
zP;ll4xnG4}TY=6O`1j9&zvp@l{+{{~_<K&zfPdilKj9xNdI0{L<QMn{`Rnj2BMgKB
z|KU0C_gt^R-%~#Vf6wU|@DCib`~7A32a6tne?`8&4F4d19ezMzY>R$A3XuvS{Gk8#
z4m1+39dq>rFgk`0q074?e~t|CeujanPVYyChX)v@NV=q94HhXNxPf>DajAuss?-TK
zRX{S;46!1*AM&32TH<Fsd+K`<KLgowg0Pw*#j%6z2U2v9Z7-y!;u?Jvy^x)X>VN8Q
zOi;p-vV}4!DOnrCKQbxVjgXSB#~>F=e~{EW;bc4qEOzB0?JMXyf5lfDh|I+V;cQ=Y
z?rK08olX-jLRAW~eUh|KG(L?hmY&8f??dn0j~O+`WUbKW<#N}!W2ybgZ|bYYu8KM~
z5xnKD>T4q$?UOuTK4P;qFaFwCb}YAE=}nWbzkMp44*zoJmFT{oLH&}2y)@EdOJ@A!
zg$d5;rS??;cYHO0e~!sy3h1p)o;gsXiQJO1BZ~h_P2jX=Uu%^)6{Ig66Sv05=%uqw
z5Nqwpui9ABUkV#2Xk)Z3&a`Ps4XTuWl%S69;NF%kBK=n_T1jL|NK~T?=8Ylqqpgg-
zIFNALw0nBCl(OyxJ-*czDQct2nvbLr>&xq*%wY6#`gg&uElW6P%5tgPXo1{bSrr8M
zGxTO{q7Ms>>6>c?_Gd`Gg?ZaQOR4IOeQRCy;R=_k%NQJgS?ux=QZ~l2ndo81z@)U}
zTb%G#N3e{L0SeOQ6_w5R0X|b*Nwv&pFu##kB$QE`^vpUpP>rzU(2U`^FC2D-cMQEo
zQ#XgORxtrw69~T2cU!Ta!B2KTUlH_jYjCx;@JO~Y+tu2F<reMS|8ae8Ir5E#Uk2R)
z)eSY|FJYp8RkImrtw-JC$kL|3cY$c*&E7sCIn(Yt&N<TT{6s5FKJLnzUEMsQcA+Ls
z)hVJi(|gc6Og`DyW4X~{Epq(2m6x{NWX(!qCz;q;e#SrY)lKBe9JPfaG+V46*AhaD
zXj5^aYN3i8FwqP<Gk_+H=P)PdsiwNfZkbo<K14fz=w*a`Q3(X0ylbzMb7B_QUE{El
zf;CC9^a^Hj#y0aR$*~lUhovIt`qi=Z+zJdA$5skQr^(&-###wF+Nu2WRP7gN^anS@
zq^?{nVp$_s_iCVMCu++J$Eug=V(?S2!WnGoH636J^k4+YFnJ?fFr79SwJGa($ek+)
zL1-!$!>D)hx;DXl3@X6+^Ll~V6q^454I}oK@l`Mr12Z=?moY6OYq#3X7*$IFHJ30M
z85Ebn*a9Y(kn92ke`~i5-p{Y_a_(Xjk<_!D4luUIL9$DF7Xgy2+NQ`w1GL%n@0WTW
zN$24wf-XFdV~M2r(nm_R-2HPWcV9p4KR<nSBDJ%^X|TBa{L2pb4aq-2@<XF{7C;z+
z-RIwSKfX$dhF^PrDJhBgW%+N%zxCS-f#VVI&M$a-u?q5de?oo%|9uGmmvM2nPVeG-
zq(J^PQ<nS+BP~b@G<h5_S1@lsfByDKX7Q!)Cj1$GjsJaoe)`YTxEP~R0Pg<w+f#V1
z2n0FYA;0PM?)QJ}etG(CB-(1q0c%Pcqjje#@PZ8OjN#wmt)wY^dV5g_D*YM6o+fAk
z(z7&95Oc{Bf9NAERLQp&BZKl-XZaPLU>RfLn5c+FS<ji&APPmk3d+h?Y#Pm`RC`vs
zcCSEtHjDQB#DyopFzG-3t@Bu(qfCBg$|%L=(=4>IiBfuBh4JrYY;q8-%xiLrKy9Lc
zr#G;HUzXsV(j0v(6a#%7D6wvoE>cx~hfbc4JRZ|Ze>K$g1g=<T>29qQDIyg4O`)6r
zKk?s(XoIg#tWwVwK%-i1vmTPRNebhlvY-F`Da`0xQ%Ntt*f`8tAAIx2pXKiF{N}f8
zSn_W7-#GNQof6)KY5&^&mHjk&q_xYeGtI`(&-&A*v|xq7-Z!3q;;_xLRv4jUH|r;1
z;e(4ae~5(Rbka@v(>4<_LVG`OA(JO9`?}H~w4%}1B?l#$XzysjY}Mee@6}_9N7)=3
zF8^OW2db~heTkdsJzbQSXagyfQ$>QnH^zugUfxE94l$FSnmjT{r;@iv?~JKa7jHbK
zx$;on>7p;C<C<^>Su>}|m-7qc*^%pMNv74Pe}fW}MN=`_T`Pt1IRf=0X`MOd6_?|G
z-)bYwjyA(Ia<W@kf6ORL+SeY1k+yDMIkzPHzHNCbu;;jL{`0SY{`5jb&tEoO56BDa
zXk<Ml82vhluZybQ8zq=BDjJ}qpM8B=jHsGLEXNM=-iD@M1#Bh)BrS~g{gj`0SKbV3
ze^f@ek<qWLEk)0->DQNSwibmaT4lSkJf{TobM&Td(0L&pjJAo}5e41S`<qS<60{vQ
z@8ryRLxdXUYnmW(3=vc>)O%SwCnVu5xeh>((n3>Eo)L(YmACC=><NMtjtiOgG(B^<
zAy2L8l)Mt?aapTtTOA0i+-N+B+j*Ulf2FtASWhw1Vpus_v`MKvB&L|0O7hkPq*N8i
z5;aW{nc`{&gmg(2Fi><UVQRCf%9?~6Lja9MFIsNK5<t*qKPhA$Fs2643O{7iS(@>K
zGx>@)DBfS&S^ZYB$x{X3fY+{(w`awmwVU;Y*Z3S4YB_goGBqTWHP<eb0lJkie_N<)
z_Sm=szP+ID5TcYJNvMn;Vo5SOOPVq5pk-FvuWIC_9WakwbgZ2I6Guv^NyR-Y>|N%v
zy^DD7(ym|W^T&2VA>QuHTsb*~ObRLRAur-R3$GGrW>su*aU?Kgik^#nKQl`UCCxZ1
zC~8vWRAN+VO%tj7E_?*VPP=b4e^FW0p{H43NM4=HE6m_Xo;FROycU-58WU(~gan+K
zAfulO7gPfu%rfdRP8eSkWI%q&dsBrO@KW^TiVNe@Q>rZ&-ZlAM+rDedyEdsfj&O@3
zTz*$=W7RfJ;tF;9PIWWLDW9@lOkY<j4BIw!NY|aKD=$E}5rnDiIMCIMfA#%HJ3ESY
z4ZQ1-aIP%V=*oSL&8Z*T1)J<FkCIb=@EzpRk6aCrvf=GFofX3@+f9D{Hv1}Uz(gAP
z9k%llD$d(d^^AVl4hrO8AT=_IZWTLC#c2n1@Mz6>1~{uVtqzlB#B2Z;siYxr2F7to
zDolW6z;D)#`^&Z(Ss@W^fA?~pQ)AdYp_lE4ZJ^8>CIa<S{_^Pst$mhf5Hko2xwn_F
zFX(K)^=-mSPI6#+%bdpx#6`4>x@~4%28!ZdX!VRCa9QHq^H-b5me>n$bc5{<M<@Uc
z_6j?f_=}JCI_`YjRv$m(vC5}*X5{<LnIKkz_x*0lB|(f_L-n)Hf4QF~;W%2lDcSwt
zBGn8ccB*j~tHyDWsiBv6r5;r>M0h`i>rs1VogA3X_p{_>3v8y03-q)6RyKMS%Rt+;
zawnjf+n3vI4)%cnrCUlh_pjjSkj3?kdF{I{>}GDO_=3`JZhKD+Dj3VM^c1I_lQu14
znH4T8K4Gm;==z0!e{&`mYWDn%g}cRVJOX>$1OBogeb`pdYGX?gi*KiM8UGgh7ky5S
zPy{<)C^FfpPA-5lw%0mU)^<0);J7RY*W`K%Of2w3=f>V)YQKk8{6rg7f@1(3RMffb
zp$V!^pbpgxY}KAw2gM4$@7jk^j&!6E3n{&C?a+c&R81#Ae>X=R9sHxlVm0d8XU!~C
z*G;9FX_}I+^Ms%W@^7|))nJ#TgLoF^Mj-AARHSa4O0XP?R%HU<kOBFcUxi){RGft$
z4J^*<Z1iM{{h~!(`(G(K;E-RF%<1TgNiv?3g1A~S{M&emyWhiU*5W(ge?7_SoB)QA
zgRHdJ+@8|Gf0RmUKXq1i^wCmbb~##WlZOd=U@t+V8P}vTQ=+<o{m}HL(u`y4(b+?x
z^`7bvzG}QFIaJ#eFL?xe|LvoD^s!p{RQ=VKN8*+bz-m{z#@jGXWy{0TA?5i~T4#lK
zemtcmR<&c`xOq&=C{YFn3P1R-=uO>|@AFZvm&-@Ff0DDl+L2|vFxAfy?B?WMj2TrO
z4cR%!>LIaWifT^Ttv49d*f}bx8cSuEO{dG8H^@0Z#NLP)!YMazXJnkfzCesbN}*b!
zQw4>UA5o9plQtR|+=XMxfif;f4j2$#+Ho|^<j8r@<U0iBYPpzNRQ13KU^$|>fL4sZ
zM!AZze~ZawfApsHx~8=i=-Se=S`nu+yn5D_fk8Z_TT;df6J-S91Owdl=Hs%J9Ly-K
zm`^Fj;p5<NEkW*Y_ne*@?_7(lS)ZZhFpQQrc&rIE=uCN_m-$+r4`#+U@4x9{#x1Mo
zJf|3FUMPhJ-f`RmR$t>{;A^(y;=!v&<WkA_e@mr6Ei9@>te)!~q`B?wGG8v{m9y)s
zhZ|SV6@-@RQYo|-R{KY+K4dpDk2tr2a!_c*$J^d<&7dw7!jrK65vC`Jf+Ss-ZbG9<
z>qXPo+gHXi6QCVbu+cqfgT3Ty;?mL7-ab+;NmK*WRJW%uZ=yMY_0(4Pj_Rruw$zUs
ze^Q1MR+^FDpt-9k0tVzZ5<K2Pkat_qGm0xtRkF{Zz#Ae8z3|pOQZ*$*Jx(_zi(RH0
zh4GYq7#$$~6!r1;rb!iGh9;L_?^WPk!UK7?k+jfdz6@k1a54~ifOrL{jFe~i{}QvT
zieI9-=W>TDRp>HMJl;W3agKu-MK@5qe?!;Vh(JOqb*m?oDr2B>yx$dyN(ah^%S%B=
z0_<#sa^JY>6$UA|aG>^ayNSIPUeQQ#Wt*&Ma^QcsVUx*7bKrYN-YeVIK=FajTKtTm
zg>t&}2IU`@7)sFfT;xifWS#J+XQTJU$67$3jt|M-Wjmj%;liM(0KcbjLp@u!f6tzh
zKGl2p^9|1kKbJl#>)0vl@EB!TBdoReXe$HgQCaggLG}gYETOS})Bx4Z)#^xTrA?4?
z$S}I?nv|;>wSHhYmUYMiCotVU*-~gmbE9naj_qYi&5korxdjtu1T|1v#aqe!9>6zO
zl^62ORprQc#e*FCwwW$ctJL+Ge`k|g99>-uj4b5siN82RPZ72gCK@k{bh9PSa=;x+
z99TSHi9N*`SiGV{Eo#tC80Fg%wL3vG{dFw09A#r-AE=%$Rq@oEXSk_E#&XZ3d^AHB
zS*~vgfDUP3lx?L$OyP502;*pFr%RqPe=@EwxNpEhWFQ`ejJA+>`3wade}QhcURqu+
z`-5{dv)lv9sf?O1+#UAaS;%y5e+@hvYKL$WE8M+&AmLn}MsK!smh$<iTg7^X6Yx?%
zJtztbf*h#+Uqm4qCD;c?+1Ac!atCu5U33v6@DIow;n0NK8JVx~FjD;3&$NJt+oL@k
zIm($Q$>VaghtrjrgWQ_Ff4E5IWHVdLo#`tS&SYmwU+Ld6U?o3y`Q>volqB5r;axo5
ze2z9!+B`=a2G1@v-H(k+Q)v|II&56AK~LqDv$s-5^psXGYUys=&$sj`F5_Ez5hLHj
z&5eKX$OboE<G&o<ANzYEF?ad(=6~)$4HPeE{@2UkZ727S$ncJ}e?l-yyq4je7W^7V
z&v4dpj<Z(78(0EQwu`fkQ~lWNzV0ACSS6eTw0DuFkI-HnB`G}&L}%N}6>gaCp9~a{
zrq3v}qOQP)VRoIV8x=x-0j{ProIp#xo|HcA1g~eceVw;u{~4AlSCcV%C#0IGJAQ9P
z`HqW2ik}b_cv|)Vf5QnG{DVE9C;A8v=qX<70sZW952*%GNgAuxUNUv`fLpDd(c`$R
z4QuE2+c?9g4q7bxP5!nLo$@2*>sZknFna?88!vkTt6G`|J3>eE5q{7Syw(l6$>%q+
z>=AzQ5S_0TPu`HcppX5O-JMotw~Df<>B|CcxP4Hr0Sp>-f19@e`oT%ED36|QTHbJX
zuGA5OG48;~!CTZQZ@7&a?{Lf2J0{h4Dcr4v#Q-mN*H7v1s(UVoqpQUqQM~TS_w(C-
z1DD_d?d{#E_FVd~Ue^4|hxO9C<;z-Tr$<q;s5srqTAplmnRCB%l>Nf)a>-M}?ERy{
zCw3v~_tl&_OE`3mi6|5dFkhx#X%eMMZyf!&D*@p*sM>PT?YFZO>~VY9U2gmW@T!|{
zX<2UUtbYYmq7uG3!I(epa_`C|=;8Or=coSxW=LP1mjM?I69G4uK~)zM0y8(4!EyyB
zm+vPTq5%q*k|-J4fA=*D9)4SFm#QFkG#iAkabqvp?r_5;;gdQfC(ojIP&ELpO?r=N
za2+g?<Kig_mx4D-#>s;gYcolXt8$X1ph!fPlK^JKW`k3n0N<@A8Od+qe~VJ0>Th>#
zEeJ9{^@Zd2%%zuBM5|}AT%<4?H1}R4TltLtQhU#m>p%kSe`l3?gz_l$_}1Y_%7_{+
z*J+Ch8LLSjA-v<fK5M%Ut3{erST|{iz0p&fuxB^5YVQK!sNYJ~XO1u>9HF8XPtm=$
z=;)qZ3Gg3Xi6u*b5V_|vf@#E+vNGecjMjyU?n4*5K=R>n6dFe|acM?{j@DTgIPYE3
zBE~?xFKrFBf4ZW5Dm)<<SZKw$q=<KUr0RY4G)<H%@?z2IebzZ;dRd)kuG5FH@7){<
zjv`HZ5qp@Z-uI9xBtKhwqCB+A?ZK`t(jqvabZTLP8+qTTjx?b@tjcuOp<3wrBy%qZ
z2t|&}G`<NQBwpzfd#L`gOfcwFn%_QDuM$M?vD+M@e`NC2Hu8d|-k{ls-||`@s_6#=
zV0yr|IUhZI`y}=0k8Bzj-~9Zo5lJT_B-js5q-RvCjFFhQX!Md>6T<xcwd<=Bf0|vh
z$#uk52cfm$BQ~<ba>3Eb5pz*{68kc0V=6aD70c+)I8QQ%axxD29dKz;%)AjQ7#dIE
z6S{HPf7Glw`dDdNeqSjNKyD^qyc&FcW4{)oN^QY?TE(MEBNe=k6jm~{&4Wsrr6m7>
z|BA>APku({m*jHT1kTslle6$`8!w8v>P+p(*@3I80U1n6**WI8=BnVMQK@GVi0@{`
z|3kg6*?(*@c-^Xq@;b5~#4%<C(n-YByZ_Afe;PDq&8fe#jJ5x?B#zrmZTeG~Fg)q>
z6{0@@c8GPS<v`Tu5Jr)-V+fU9)IAlmj3)++3h`9036^p(@?QONuL**3YmWD-AtJI)
z@m>uE;rTl*jzDWAtbf70H6TbKK0kscW(wFft%1!B2{Nz<1*Kz{m<UaJX!AfBFsTsv
zf90YVm;J&#Vps0CG>;hNZ9m2sgoQx+g<~XfW@G%iEP5ofj02yOv)V!VCeSaNRg-Ln
zQ7_NRL}4L5<utf9`}$Ph){B&8a%9#1*|?k!xI>-^*nz*a^T|9v5pM-pfftIe4q7-p
zvm0ynovaLm7Afdn-J3Tv@{AD{Mplz&e=~B239!Nk#d}<{M&y4+=oYOhj8auKNyYD8
zdr!<E=TLeJszKB?i-CYmq42PdO@FGMTF=u5<De1iA&2jR@Xm0IOd_7b=)F*0&oC~l
zK5?8uR~kQ0A2iRrZ+X}}Z_{To&r+xqX0zF$`!b3k{S4EJTTrkHQf`TgJG79DfA=;O
zx$$Puv~Htx#-v-%j|Z*RgS3cBKqSPVnVrHhzyb{zhJVT!h9|?yIF{@rAHx}~#d)`c
z8RhbV=fT+-Z3L`!U&LvXqd)T~r%jIj?2qKMfq*SW^C&Gq+f90m(~pnA<`7FRaC$la
z0D9`ZS<(2JH;7)yytJ*};HGP4f1m`+%`4?>&koutuSAl8^|9(ZG1z9%#_jOgHOOX|
z*6cl5=n=ZC%<YthEAqUL4iwC?9@eY{**XTjBZxLuXrpIzI1ofzErM0EvE5eU&Jv{F
zCG39l`?`sdiGGYlF$p)5n1f3mBfW5LrY#5Jo@U1%(t~El840e#vS(&Le**LYOPk()
zxL4bcp^sz9B(yOj#uqH;8O9aCX|`sB-eco1LB%i~SW>pQ@@2H4)Bwl0vgB4e=YARR
zE6p&j!XIN5Et)y>4r}F-2rj*nus#a-%*M82hJzO&g!`0zer6PczHDaA2tLH4X5g3-
z95lp63mNSzXJw^u$vmhCe<m-frJm8?jv@9I76?;C6>FDNskA$C;LeoI2n_RD_(sNh
zgf_ErpbmT7@Z@(aL~%(w_Ow!tYetBTYpd9Y)z>vjyPV{Sy&kn%u%7C*RQ92oO`f$|
zq6S$2gFt-0G9T(1Pqj+d@Z{%t8q7jCrqxtsF;mT(3ZY}odxdkD9vQK%j(?ZjdPyCb
z+x|qt*x!Lbhg=6kcqcP4LSz1HcV^;slcA{nvDr|rV?LX_jO$`BgGv{UiJXC$-oHxX
z24r^yu2e{m2;9)!3*>FYcSqa?;KrW=oX}Djsb0m35!+o}l$^#9UX1;{fEf(o9d3-!
zuV9Bv!c4o_GCSI?G&L3aDSvf8Rr9x#&?DCSDq?PZ5@)|T8~bIE%5&xn4HUz;QmUc^
z1Dp3$!cn-j=r#%Bj{E^e#B59}>E!4Hd>Hhs%w#Fqd4I>s+IAI_x2vd`ELB-~Xj@8K
z#E%#Z)x$f9%<^M!OG?|%=4w>dCs5^u)jglr4TvIziRQ(+!LUA|8-MPQZm@n|pc@?H
zwr=owuj@wJ24`y$93r+IYdj-Dva>O+B;=(AyJ{FTD>BwmzCvGcrN+kgjJ{w2fnoSx
z>q|}(Y5Qt@8L>X3FMD^+P|BP2dBn|YCjwyN80BUXeT3F>XXc3Xl=ZE*^>nds+GLlh
zA9rlPsF|M!sepcNgnu@*MEHE)=m)tRowsu2e4w|31fT@bV+0_Fx=es>8Ef}fD!H<%
zLm6sH{_y#k<@x0ljq@{>OfI2Y_5BD#1p;Qh8`dDWyVZW}OlatLWp7sh7{&*T6z!qD
z4EtaFe)rBsa6azd8OD{R38tqRbYAnWB%+MZDeCP*6kSS7+JC!As96urf)dB2-OZJ+
zyOW9x^Y-peV<D<-+fk2}(<xIqVm-p>CcF&eN)ZT4Mhtp~$Wl3p_j}tUCIn61%nSbe
z>DVJY5)L;qjtJd$R5Jh{uQ_5BjG^7$`#J*WBbe<xMw~J-noiW5@wuZAZg)wEQEDXv
zj+-_|;1V#>iGKu%du05+zyQlYjAq+X2mxYd-$t45z=myx{m9-yQdWjhZq|24=-pu_
z7CLy}D`6*Q)?U;%1@L3vYbFhu3gK2zdW)%htQ;|Q#D2(LBU2g1?V>%0z8u~>i>XVF
z__le*V}w4G(4hEa?EgqYgJKwl(QWMy;?rmOjNQ9~9)II5?fhZT{{<u777&bBd<)0@
z{0Ofz^SN9%H&1dp?bswq%aDgCZoDez0vUAYe0UhDQ+9V|I4M3I-+X?{dNEcE>5S}8
ztcGw8Jj19>e+d0<dWN;WzmCFk^Xy=BALW0j_gSD{1ElC|TP)4PdU}KTx-yvOi=!s<
z&&=p@V1IRtD#Lka@j$m+*ru^Rz1YMr#*zZ<Pjk?rOFF2%Uuo%#st%c0Zdhxn>CC~s
z^QeTyLHw%X|KEAk<@3Ge)`}#o))2WY+!3J_D2JT3E}roqVm?t1sPQDYi`{wk)f3wf
zT}q*n79rTxODXDQ1Yx>jf@0`aZ=H_+pXtCpwSOb0(e|dhZk*+chSo#u4*r*O-N#ST
zTQIv2MzI+-M;DgIl*wIe(C12uF$oFQlwp0v$^U3UlJ#u7MDsIJhBkXyqF>-weJHx1
z#F13fg${mpWyHiQi}F(RZ6+@!!o0!N&gMN(Yfaj$bkItTt(Cdx&04)xK?EIFXg!nU
zynlU6`K7CZj4SsYn9hEg{<4q^;mHT3_KWL+bC2jkX*;zcOyHXk1HX2pUDa{L`x`3M
zh4?=7tQ*eQMqedQyO?jz5jJ1I0})D<y4h2<UpljNB;4Z$=B~0BpWT+`Juq@yyL)W9
zODiyW+D(39r$=@+1@7*6p_@$Ffk8juX@4wsy^_ct^H2r5wl&;L;Jg99q#jB~&W$sv
z=|Xa{+(}jLYB7T#uipDIds--bMk-hJ?y0QDn|834)*1W4``d?c@7-#E%=>g})?#?4
zd^rlLjDzKy9j@P4&etfD)qB=Lf5i<@vRn)D#ai{=&voQ_qh*{Fs=pBle!tIUtbe~9
zr?j3RQAVxRBu|lieH%+Bll4i95!#(hOXDs%0z9oh+&Lu29hBW@P-@ruB{|LpsIP-N
zy!OHd_1^J5jQ#ZHOVNJfE1y%*ik|v*=CV85E!~Age9n6|WXs<6H!sb3(z~-Mx!<JU
z;oVT^!cdRGII0fA=M?pqJFSMhmVcJFk&RBW1*!Dpp*nTb=}nE12Ey;|bkBW_8C#ep
z1vsaH2FqcB9jtw$2w6!A4HjmLmT}dN^=&HM+7ma}s`Xo^B7_LUtt-lA?CsSsr=btG
zNsG7~GMVl|B{_M(7Ti7WV%t@jUO?pd16-`~u)>IebwJ^V_8$4CeITDX$$$TJ@zuG~
zp~AcK6!;X#OMO!VHh1|j@_o__Y}0!wPuRj;{hlt`()A9xb`2L=G7Not{$Y<qq^T#g
z+4!DIzyL>_M+z}5sBlexeYa%6up9e8Jz2sVRwZhnGc83fi!p|`!c2AUs)%7Y@rzfj
z$xf2gpy0k&kBQY?pwNPZi58@EEnKhkwyluHwLgB}zQ6qk?Cw(wmjM?I6PMC<K@^uQ
zhafDMxi%S|0pYh;HyH#2e+n;0c}~GA&yYwMew^TWFQ%O@4c}>>#30vfoKnJ=j6&(l
zZFD|MKm|3SOEuXnFSSvBv5khPjHU^rS48o>ZIqpEGc_a0aH)|(1f>Ba$#E+M)(xYi
zIO(MtxLbQB0cPi~^-{*_LHOaUl^SBTIE@$@SY~>qm4Z=*Gfqf3e`%#$H(a(}1V~h2
z-gO37l^8g^jb%@J&ms2T)lQ2~ItEM0+G{u5O`Rn9l~p@I7)%6G)46Ev$RUm}-x4I(
z+FEm{bj!-2wA<RyYK0!(+uBM&U5Xq7ZW<2?c?Y+nliNIQ?l}=S9VxlK)9x%<kPeA3
z(42+~FmkDllIpm}e*ih&BP-8sHigx(?Eh;F2{UOk*m#^G$Zi4hD|LTtM=V$b%yXmn
zEdl1M<+Ia}5J9;GT*JNMAPs%H$%DAQkz@_~!C78EJ36lyn5RE4`+0@!TGVaPH7EGY
zvG`^g$#vSa^KV((S)9+n{2QKo+68CroqBb)#mY9=yb<cTf3o4pI$LQ`S!fqOKyAvM
zSGw4gHl6O6F%QbxEA70<5beB78oIlymOeymGcy>u`+1lmlhmIa+1+;>*=NPJAbbyz
zJLgdwvbj;SHrlAu*6SPVxRnLvQWaQS^Z4JBBEYE7hQXO<aLPNyix?Ocl=DtF<++6O
zUxuZ1%60~(e|%7pkLH5p1WX0ubBLED$7m{HMD_-gFxs<Mz*HWucqiO&hNc)rL10rt
zg4y;Gn$m$QJ4z|n(9{s0Rr9xNG?gfb^|Q=D!!eo)dlM_i2MBPeoQJvD!ikZ<&9=z(
zRqo^M1cPY>DPBFvIS5OG0#0;6_2&;q79pZaxiVUWe;>qH2Sc7ffQzb)kABraC+ApB
zQ=M2FW-X9%zl4B2LkvfN<euJ9n3sm=Mpf+bDcrJSz=+77ox)vEltSH;e7TmHGQU=t
z>G~njah|17fTgPr*Ie$RWpIa_Z?x%fP0A!kK$%otak!4kGE^LOvBwE)JQr)u7*0vA
z$FfFyf7z>3&i5p9S=+3(h@4`Eb;D%NP}c0;a3ydB6hiV2egc4t=Z>CuE5LsD++nz?
z+9fpv9MxisqJsc0VHdotK&-wz($~AQ2)&CBw7~Q_Zc4ufo!%?Gf;^*pgferP&t}E`
z9**8R=<FNaw%<+%iu(FMab8U`wprx^w{lB<f5``Nh}b_x9**QTpMQ%}SZ8<1_xvDo
zedIH|=SCgvr!uN=`tY$}Dngd4@o1qq^__z;ZcE(CCqp(Pxp#sw?mWar$na?w<F(nA
zE6)W4$q5|uMltk}Bqwl;F(z_-H<Vr%aF_5{Koz5K#0eZ@D8_oRDU_ZQT^eqsd6Nb&
zf2CZ5V?%^iH8HH=m@hrZq$eU6zXy&5mTi;gQBiZ_X`6&Jgw!k-%Ivs<V`9R|RK{=;
z*Js)%Kq|HD^a91Gi#=)CRzZ|LIC~nzIsrEyLDECR?JmwoA>-faDTI&*^1@SJj*%9V
zBt$1Xr%xd$rFiUlZnu#WA2Ng+>h$pIe+d}mOof3`9}9C@1??osw*nd73m%4aUu~4Q
z3m$ks1JXP!iZ8wmUZ6t+WKHcPWN{qgz|a}}_^L#rg7&^t=w2qk?#YPn#!z+i!hsw5
z=_0)iiVElVzYZ}|4&%0ZweQ&XWy&y56>2S6wCqt=<#8{M9IMVvdg$+u1E=Kcf7y?<
z-#0yHPI{S?ls+vunstme1JB(8yeN>T&<@QWxI)-`TVtl_At^%l>Rgt)aA>ni6thQc
z4&YsV6$m0+)^&!VMe61|LG5nZS@R6BtdCu~2XVDab;^$oD%eCfZXwg?hIiu{w+g@7
zTSwIoG`#w#WwOh6l`_=aN(4s=IAN{;V#wS%DeUd4a4_`7VYX*wrgu+3%>kMmXZtN=
z-7uKst4Z&y1{D#^?@-q7Pj3(Z1ByN7!IuFS4HE)4IhXOp0Tcl@mmz8oCzn}38J_{U
zxA{OBrUHMpG0-MlSqP#ik?^uYiOZi5dXJ<3tiiZ64#vzKaxe?&6-bj&u7|P`UMQuB
zq`K_oeRGruWS3c%cbfvwM)QlER>(pc<k=I~<Ek?1y)tjYc$vInnwO#gK7{&lWJQbd
z({&3KXUPTQQKFOxL074Z_MqyKNs^Z^okb`S;>3Sug=j+3-UG6AN_+dWm%=%>TPjO9
z%o7(a@=TGIsnkGFDpg5sXB?1H-Qa*UeLNNC2l>efq-{n~^61xj8)B)4;v<0^mS?{a
zzN1xjPt4y0eMOQI1p|)#c_b(~ZeGf~EYl9}=51SrBNryFY)mzx-z>0M(nneoN^dRG
z(`0|g5U)03l!3D88AGfoot~ZX?f4~GMeA_pF-_)unIhnvQu}i9^$Ctz!${WI^jzaJ
zLsY6VKl=qyIV(*ABYTDT4aX-&?Z|yXgzNlFu{&|hJZj=~U1QT&0>BssPZhr%DROwK
zUp<62Pnn-QM3<^089ReLH&%;sfKVMj3NL>U;z2FxN~YX|izVPdqp|LjU2zFr%zEu1
zst~S3wC1D0v7nr}s<O0-*@x5ftp0Y;+XR(tQxppIHE3B2&3j|E?*1t!n%WW?8#oR3
z+~f!bW%(=mpHW58P#b0994``D;?tSE+7qvJGJGW7;=btV1|DQXeyv<ot*q-bVj6!U
z7EG4VF>r$vbrz%v&507a-y4XD49l*Aq9A|0MkK8y+SCMRr&O%#;;nQhucFsJN!SxI
zjfrO$Y|jFQrO|9(RCr5=4*toXGQ9oNeS80W`rpVX#)y;l^4pIm|3C^x1c-;G#6U{+
z=YKB0KmDVJiVOahLH<UyR;tP07|4GX7)<F4r^R7rg~<vOxXD+~BGM!2rTz&5+>9XM
zWm*%adABL2^D@8j0>Zk<VuGPRRhkkDSOVKJabbieY?={_O=f{&gwfjb5L*Gs99`)M
zlLIu5+kIS(CP`Co*9EK=?|1VV;%FkZ>v3VS!TdgmiWojnmu318L`R&l14Mr{<t`VJ
zaKQ)e-Y<h(a$+QVByXf94)pGUya7mD?PZNpMA6}BB+Y`vbNels_cN*@;j!e6s^swM
zZM8WzRTm|QaJGkdQeBW6th6g#GZA`Fp_B{ti4NN{LZdB#*8q_ohiJe(bH0INovp!t
zm|^w--dIK8^+-yrBcrgLVYYv>1ibW4)(D7i1Qq>kN$+Nj@M^=xBY_BlzRmaMf+#aE
zQv?rpPIbZFGApp^?n{qX)YlLVk!c6@ocgZK@T)zp;yo0H?iSFjY8?m8s9I8<7UCHR
zwQz2N4G>TGkgS#+#1$u!RIZ(ze8}BmpB3m~>?w-WL7Yd9f;l4iz|4ObOhxgkLKnij
z1FG%)m(rt9of0hJ17zj~I#&`hLy0iMsnM<vn+*U1HQ-%|q^*M;Qr~5k))Y+t01WE;
zG{F%Z*x=ZZ;24ldlM|c_p>oxR?JRtrfQ01Ig24xnzx+xs|AvP@!(<<D{1vADDBIA>
ze=q;?8Po<~@E`#&BU*p6TY%|>6K7olTR|w{f)lzM#7@jW7!2jN!Gb6_)x*7+CHg>A
z0B`{z?pSST^AxY*UbW&oC{+r;SpB3*K-#ZRcm;hQ=BZ|06uHWeMz9K}53^*0)jk@O
z;)__VMeqcf*C}`%L+s);V?A#lM-F4Y!4YYR0cZ1Ij`+mJ0MdV)P{XH&Yb8pk^2eeV
zEt0q;tDNGw*=23=^)US#d=+5Nxc(x(N=O))xQnl92Vd1pqYv`+F#Q{R<r@5i`69jw
z2*hhI!t-P<b7Q5aw}(v%KjkmtvGtAcG#B6PyvAF%`BDK#4nMaV%cU%xYqsgi3C~wt
zUdpx4rFJQWMpu7H7Zfu%)WKQmR$MpTF_!ASO&Oh~Mq8>L`;a71yzG>i!za5XF@jiQ
zPdWa3LF|lm0IW8LVx2yy+HpagJ4iLD7|Qm!!>Q<=_FKGfb}jeC^19q{#ofYs_fO+X
z+NDgN))v{*wdCG30OAgNZo?^-Q(4@Ui0hGJ6=Qi#H>H1qSr|CyeS@QH;N11*{VK>U
zAhtLCs*cY(-JzmuHz88lAlyV8?nf3FH@NB{${TmO;;j4n?6Y<dH<oB=Z(QJ(vs+I$
z$`*akO_pND+DS17+6UQJhs(xE0VbhT(^JsgZBvag?d}@6rtjF;)O=$sHFo)EQ&-w+
z-1mprjrM=q*vv((+M8Tbl1ML^?G6W8N2?rVV9_Am=yo{UNMqK3as8@B8W)zZEoO8~
zzq@~ys&36*YwfetYis6yLHCSv1LFUv?pX)zRzae_;)v>q%dsLq<Nnk;6wVa&m?0eG
zw}*J9X%-uMJ;arwUTWHPvn4%u;0y%IKC%bYK|FtU=z>=3zux-^_g~Ky(m7Rx^G6rb
zJE+HaFx>i0sT)cn-jAzC>Vb8buJxC+B$N$w9jLVC@_oPW`A*qwN{*E8ck(8c=KI+P
z&cq2~oZPu@uFH3`56VNX$@hD!m72WrzViX{kgNNTlZV`N+sQ?+rgmMhsfVxppKMT?
z5!iqB`-h{mNJ0n*`NbgU>An(1Wi1vF-iy-VNO1GNO~JGGUu_yNCqelRHw^|;b-G#L
z-=%eZ5C_?kg3#_zvw%SaXzdp?3%qSN^i;FpZV)>$lbRU$Ma=@HK+;9iEJ#=GaCp`i
zUTev}JinvVf1IEVMhOU^>=%{#nMdAyQK^6bIOZE1;gVPgD!-(3$~Yt|tX{d8#xrC2
zxkfcNT2B&kX-?&HoKl?a78?(fyTL}tajkxMWgkK`$oUu~9uQn3NA^w0SPfbFLVnMs
zm4fo73)#ue!(?x<6P_4z%GBNqVyAauKBX}Kr}C_UY=ZLD>TL5%Sibt>2-{a@>yJhp
zUi%*pOlybDZ)8_1vdrG`YxaBxAJ+xjTlrq5)ipMm5v4=bZTO+Xw2LvTw$*>zkm_<{
zCHT&NH^JJ#F295B{0*xX2A7e(DH8)TI53xS@D>x7rUNu7e_TtC8@Um_&#&k|Ff@x~
z@c|45kz|j7y*cr1v4_Bkor4`@H;4WEDZU@w>h9^0kc$`4_J|^j<X6RdvCDk=Hu34>
z!};^WyBDb@!z_Xk)90@f&<~vc51bk`P6mK!3Df5v)2~mQ3r(LheH7<X(#O!>mY(?h
zD8W(>IMD~5e;<v4)L#nv0R4UO-*dk(8so3x6evjdlqhrh_?`y1^faj-oIUO5-#-8H
zz|-<Se0~_0tA%7v!Sv<F!|xMy{5XZ)51dJ7ar$!#%?u>{H*sdPGDLz&)K*V_P{()w
z`S$fMU;Yk^hW*!f<&$Cp!P)6Y{EEmdH`Yzx(|5}Ne`($cn4?vzNmDC2>W~MPhA-2D
z5{70++q8c7t&S<HglGm!VA2}uSSDzi@)|tF(wgV{-Zth{)hWWB$};>nPbH1@Uuk?E
z<-ht*`#pE1QU97AzD^IH5~`@RHq^yhA(5WiSf!Rw1*X7|P&LTrCTW;6PU(z_GaT_s
zYrEgle<&ESQW03fGQ&Z6zNFd8o^wmH@7zSRGKJifpJV$K9EQD8OA%6J=xCf~+Bq+x
zkkE|?*T28~NaMcIsE^Gv(s=}ua)o+&)P_)2o>Vx>Xcv!i>YX~Llk88fq7#WSoT)s=
z(%I_A(`28fW-W13k9Aj{b|XQ$1ySulBCl!ze*uY1IV=!d(PnuI%FweYGupm#&yPwf
z`i341QbuTViz1byNFAZ0Cn-6E0M*q=jAC5iFbO|x4AKm&90s8&tI~1qG=E;1PHTn8
zM5X;N=zlSO2XWo%C9&AIUCLNY6kVc$&%&!}-^O34mzGay40hv8S_LMmE<ODkPG!9&
zf8l1AS5*4RI<sFzSk+6Kf%{%EuT?cDm`pY9etFhzwK}0i1}g0GlIG6h-_eQtiAIJ_
zHz_ibLa+6#MCmozLhD~Y&QD4b$}DHsdD5D*dN@dT8#_|*MpX%qio8676i-E5*X4_>
z9BJc$*rrqpW}O^Ig~h*Q3mzD6D`O;>e=Im6_fa2Wzp&C%8EKlsWoE6mj2=WMwYI`v
z`UW9#uV&r57Yj=WkXz!_`-njBT6AQo=-1J*@*1wv8*XW4xO#!mErJP$o9)Jsd9?b*
zyHOw+H+n%(uX$<1`M6?As#%Dp<lj@@E%|znjkCAIlhs@^&TZNIDWe<}fDGOzf1Z*|
z#cv4{FZa<1q*JquvOjBrn{g)6VUbEwN#PWUtO67Fd6Dw9b1BHpL8oUkoT^e$=hZa|
zWtoH78iN$*_ymYvd8a0MdcMy_V0E43?lU=?#9J4sR5x9Ly0JoGVv+2RNk>DrDx&S(
zu*)Ttl0XfBwsxh=#@9je-lSA>f0%dB*Bp{+g!lfNSJR+Z!J?DEP|RlImx>t3vnV1s
zAQn|?m_0gjz1X88uY|Nit7>3b<(fgc2JugRK-g=t_KnKsS&28Ie9pgape+1;otUj~
z*9tncd58A8K@=945oVmVpdC?ANfD-<4)_=Im)8lrI~tWQMX9gh{$?bVe*gv)Y7Kh|
zD}=Z*o`QD2&xjRlV@0kY#XN2aE20sN6KNxFjp*@l_(qEV_qjMCg-u*|o$Y^!ajKXo
zG_zz7{`lwg_2JV<war!p0Jzo^>9v}d^o=Exfrf%JSr2S-8w1=Tv1C=NMWB{7_A@+P
zQ`%8>BBqUQiwG$n%H;nAf79KyH2$Mdf*CSs!!`zkT6=5e2%(|3+_T9`uPZWqnpR|7
zfp3EnS8H&MSOxeBT$d&BDI4tL-bFzYN;1@DCP-G(@h*}yhPAZMFrTHr7lhMi2zc;3
ze*KM4|DYd#nL_Gl`ZIL>QReYklD~OT-5_j9SUq9#R6ws0=96GCe;ee<RuU*3Bggjj
zBzCS0XbhTlvLqf3N4l%W5y(B2L|ijbqgt;eZ}SPVbUSkYACl<=RPJ<gUif~PK^u(c
zK-$CJnDGXwkrvI7(Mg<pm&If|?-oU0Gh}xmxJlWB^Zb_|>D5RVHk&^25POP+lt@FP
z^WuyDx{dx}l-AK#e-0Rz?#AdVkGdt|ib-;DvbjD-r;2oLLSJA~%r?6@+2tvi37;cK
zsS8n^9hIUcv^8PWpllk~$yF!byHUwF+RCBIi)5)wTBM~Lq=O^?l>d7l*N&sR!8OYn
zxA)PN%1S1+{i!S8peqmaXM^q9F%WPkz8gVaVD-6@&os}8ml;zT4H7+DHwv}3`EnS*
z4aWGZF)i=JmpD@yB!AV}K5R@gHF8jvl=S9K-W<kogEz`hpr>xk8<fmALyM+B#X%eA
z)u1+@ud~(JK&d+q;SA_L$Z7hjoroUZ`39mgdeirjc3M12u@pJ?{?>?gN!aNa^kq#n
z;H;95CljhXS#{Hf&LgcN?=P>TC6?B8seXyF^;)S(hFqh2+JDLS!~EIcyNJ&8eNK|F
z3{ANe&hdHhxzi2J0zx(-)1Mr#BFhz*y~eQ_x~h`^)p(<A>LGJYY|pimTZd8I;1)#Z
z`##(P!!%%fvk;IvH!^N|uX^B~oPQcFLX&vs!-Zw14>P@HET+gnBrctIOMu(SvBSu2
zaEwRK`#v1wnty53-TBJ$Ya4$JYx=idODC?rr0*i$r^ZY1m`>BNY0l(E5gwVcasG-h
zxhfFjd^;I<80d8dN*7(|yBJ34kZEVSEAUH`_1Edm5#9UeId9Q;9a-j820UB0<at9r
z&8bfypY`EOe(r<%I3ycT+u#)6<)}!jm}!#tJ8my#%73fr?=QD&|Dc3+%IV=vZeVAk
z(|R8<H3VW!-TzD<97rD=XV3=YG5XdwW;`<FbZWF<R&>|9MQN9a=w%+6c6PhZ7`)6Q
z(e>VR5cN|^4hF}qG&R4LjK=%^R|=ID9mx7qU445IRbt?91*Mxir7boeX2%Abm3O7@
zl#P%G%ztY5dA{Frz7I2KgYhzY<+o+L6byxzDjLYk0r$QgzRUtL7p-AP$$!X!RXIUj
zw^{sEyV=G`zs#xS%bDiN<`Ri7%qutJzB=!z2`k;uZD)${I1e^>E295?Ti#O8O@WN`
zFRiQcT1AektG+Avonm&Q1Lj;|#->{fbM(y%%zq|zJIBvRtzoa3^R2K~zM8$Nx#v5L
zIh&pAJ<Ns;_CoaT@1wK9GH2CKo&6S_eV9QTjOQ`fxDVs~Bto~z)N(bNT&ksbak!&a
zd(Xz<t*k>NTns7h0+De3M^g(PhF6RhbIq7<qs^F#5Tl+s&vV1oXBE85TiS*#atts7
zMSrzGvGaoYhwctx;AFGL*3?}VDemI=ao%k3+=gi4#;+xizfES&znTU0{no8aN@Bse
zu`xI%4wK`*c%j+d(nVUP5fHfq{#M?sy<SVS%(ZpxWaLr6H~45`xNx7dYmgCBf7-P-
z*tO#f+F-nn;lh0wuN8yt_0Sm5_j$~EUw>2HH{xsFW&ha;-{Ea;z^9^1ejoTyG9iD)
z@NYUdhZ(fRcuiyQEpO<)n#8`Bl08|oPWIYM$>ZD4JF27vNTzEZIV93eQz=ynZ=N3|
zs$443wc=PVfw)|*l71<r5wTxe8iD#XB`I<IwL*$WtGROty>f|YS!Y+86vx=-Sbvvy
zPIo$uuB*FPFKEp@Rkz=gG_`4myzxe{zQ>wyT6n`X(Ws@oaiu&JvuHIIvxNT9M_nuz
z65N-^l$ce#j~Io2kwKOU%v`%!YZlvrtcBHJzIaqMKi$`iHWiOo3x+^3rQNtjdG6jr
zEj5knWk3p<6n3Pw{$AZu1?qZ{RDUh~87pHKs$CVG8#UIo?wo70JjmHOZpA7?8^QPt
zPjZ1T`JG@`trm@XBIVz&A&z7PDUupj5$3)u$|6{QF$mqiSkM<uwXv3+7j-A;I?d{Y
zv56tO!Yow-kVEB7fQq@XCI!eYNd4yGCeXx2qeoHAn=j?U`|FK-jv0g=DSuz7Hv*|?
zJ23VRCgOaHk{(M(GP=YqdUQ~=aK{RIz}J;=xP3tn2*otWx3~+hCb4s6-hiXJj|U50
zlvTZqpmXG0c-J)Rh(+s5pAV!nTz!`%_X#%=^whjLa~m$;@f~g;#|6{w)~SAQ%xKa9
z9Ep`R*I!cSHJi@6=6z5cUN-VN6pmE1H21l%cF4Hslj1EH@n-n7KpbB1Iu6z{qwHHw
z`D-cc(`!T(mAgeqFh2hW<7zb+mtkxf6qiv=1``4@GnY}+1}A^*T)T4XHW2Oo75ssV
z0K0h3-0_U0YiA~15_g$Yc{RC3(&YEEcmpIYP?VT7E@(<5xCG8wEY1Ru`Si!cr*EIm
zZ%^O6qMQt~N*I{lexC&0;Ph8;x~SD;gkT!Q^!9c7=`-h`={VCNIY*?!&}U0$_52Bi
zr4r&qhj{*Eq)30|5$F)~`QpFla^Yze@4_iqK=(OW=5+XyMsV~tshl|b<N4>e@1A(7
zK4(t*kN!0O_VV`h*VDW)S|j7ePQQG8@;4+3h88BFnL|DO{oC~W)Aw_$EmT$zR9tJN
zI;aFE{3Ka-4E?PLl_Y2sK_jU_@-<eRNqVIG`TLJ)cIAJEZ)=@_>P(!>tP->>1f?_r
z6QJj%If14KO+^4of}c*QK6DW9lIY==RJXNe3p$JDyb7yw$P~IctWq%zF2kBFRPt<i
zou$t!Orc^f5cSi~eEOB{eMe=HF8vcqel2g|)1T9ip0<HqOT>A*z>I@yhWP~aYM0Db
z6VyV+rMrLK#KxL|U>5PZS@JZe=60{TreL6uj4N`F(u+fX@~&W$E7;^3{!BNNdH$ri
z@4x6$DebS)0nwS%3-weNpZ3?hQ~CRXt_eDOzY<@PW7TGN?o$BW-SPpVzF!*oXhaoL
z+>mj}goc8s^j(vV%ZB5W7|7o-MK`+<uifpsj+cLOjya(Y!iysVj{gT<``b0bi^dHw
z_r#K$O>B#v-mJN9_Py}A4AbwAu`4fDz;uj_2Mn-ES$f809}02D4{Trh7!?M(9Je5S
zPJ><;kZl?md9vZ3vClr-<#F-7G4laqC08Ut#2g_&MoMenSg9zuD<>a6SSkP#CHgK0
zztw-l#+uQ%CZ#!Muv8S<4OYdaH-svi!jennZ4<hDsB;@b)$Vo;@S?L+_AVxGHL=lo
z$t)=Mh)}g5n~{cekS(JS5<T*qbr0nmsC`^Hg?K>UjQ^%nlz7ciB~Y21X!$0?nNTo)
zQy*&@0lrKU-fIH@Cob*6@+pNWZ~1(+Za{xjQcATzL8m+|IXdP5Xqo`%&neC-b+w-K
zB(osrZ{e?mXvhic`wGVD8AABA>CIVXdUSN*#l+SsD9@$Ho<ns&Q8k*jyn$awMha3V
zG(yd>-C#fJWUrb@@6tC2`uZwD(l{o?;%)rWq7?b;v58)(8^i)qrxvBaR7?^V-cNt&
zFMutrGuiunPVP~XWTjXY$@fiazs$B{eh{C?gJUV>OPq6#KR{iOOp%l~&gv!gmKt>E
z^A3G|y&oB%5g6xhE$BxiWPFb_xh<Xd#VexL(`Qpn|ES#PV)2q<OtDt6xA!6`Y1}}%
zG-p_(<4aULes0CLGh!t+c&b8@&U$|htCHQ_I0qw1>Yy<T0YrD5x?3U=f=r9rfpwz9
zBwaTQZEMmnhwCU2yNDfqCgsXCuH8D>11Kr}z`XkJO!ffMP;AY-G;V3!rH>35HgHVS
zxlz{FEOB`arp2fxK0iV^3$Ky+QaMu)HCpeJ&PWjmjC;IvMu&_$agX4-*~EXwnpt21
z;TY+RRxw#8MfjtR;6jfx_y3sOmICbOXTZA=OqfT{FH+~@l#!#6Olq-Ilx2i=O(Lbz
zS(g!;=TH7gA9+F~NvgI)AwM3Ua!F`ZLeOTqv(YK`qFKl2??!2WzUB<Xqvs-}Vqh&9
zN2WNmlm2O9KcNpQX$qTFO;&$61vA2nlez)ZXfKap%8<&u$!mwQX_S(yx?R@!V_O-j
zT%>_H^LuIS2JgU>Hg9EZH@XAXq@%E=4vLiun3DJrEbj@Hy9pW)PYMg1K8ScJn6g!D
zrMbp3o!KnNc%d#JtbuOJMPVfuZF|>)a;q^hy8#>!LwN$(K8P43Jq>?qEf0+dCnPk$
z=jLk!9aB&<<w`?XEi2NDVx=U|pnHMc04_HIz8jbU;6hWn<_-c}P})J-eI;A#2mLvM
zWui!@gd=r^o|Cf3gb7uih8EMNZF5_Bp!6N<7F)1ZuS*>qWlD%<zVM#i(?;HR^E2R`
z2POYl+7)GR&X8?OnVf$q{VcD|V44$=s{o}b_P*ozK3X<3*!KuqfGax(kc}e3ch-;w
z@7hS#UQn-d1um-ZL&ySWn31-f$PPfA8xJ?Vu$z&G%cCJtG}FE9?vzC|%=VgRRT8I=
zNIGwJfQ{=h`;i?mW}{+1hB2$@-1V;Pz3qMvRp%Q6*m<Obx~qR90Q-`3hK$o+I$m0G
z+(f}WuWZCe@ug<rcK*JN{O!hhz@LfQ{g6@7&uH1Y<*xLGJk6|Q%9MLr)`vNBZX`!`
zd6@xM{YtsWojH{W=5sWB&F76!?}lUmwT^1}n69E_*kpIZeEA}HHRb)D##%lEp{<U(
zyOjg<RFt&GIBI`b9r(y#@5x~I6Eq+mqqII`GZ#pp2TKOqYk-nCzIM4$NL#y+yH8r{
z=us9Au9qf37}h&aS-F)<&E2|3sJ{zbt}4iStJc0<1hM&QSpH~t&yINB&BuUeh+6uX
z4vvC3r#?FPJso^EK?CAN)a8#Q9%&E<?^S|vEQz_7cj|w2k&&r5l@aEJUvg$udeEBw
zF2%4G%c}vWb!&NrAZ@712KA~z943DCy=G~x6x+D7bUz0JZh27fkENFU5GZuhbKI2N
zJ7?83+!grb3NEL0l<ln8rJ}5tzB046AOr8F_T22)!=2RbrQkZV!bNR=NVpcxF(K?n
zxV|G??<ap~K)j8a#xcZO%Y<_6JZf2*OKNH@Tb!+tO0yB7-Ax%lWMVpT3`E8-;G*06
z*6wUson|h-cro*d2$+W%HxDzdTm|0ES=xpcoG%QN_@=H|<S*Uw!HmhKj<%_zS+%&0
z^WEGGIM*?gc<{Qo3sVt7YX)^Y%4i}banK?*=fHm@(Z9N$)~ylse5yEhEzuuMoAv8>
z#nx)QHZr*z@d1-6<_gF3UgdW>e9Sc5HL~p{Xh6J-xxz8TOUV%A_MiNeEiWbH&b4KK
zY{YJN8wc27RO82BhsaQWWbb!%n%x8qhzBCh-t~i=n@w!n5-UJ6YyYbs<Tzxu&LH=@
z?#(3nKKWMf*!f~oDRt_eNnxYtxNa}LtQw&E?*i#ee(^6$@Bl8CVQd){mtoc}6PJ}@
z87hBGlk>I>zVENlA6S+E!LOMe&v?D>W+u0`d&}gMHo5dYB$xdCg5MM^%d%{zmmajT
zEeZr5fDb^BE2rNkIsNeR{`K;=PeqdzPJ_ki>(>eRf#km+`A4HC3m}Yw>FfLS)0>oN
z_<HAyq*RJ8%by+J_2E^4;}&q_3l6WAAh&;4$QSVEQ}|xm#kX~OipMBH`L`@t@)cTI
zkSfUJcEG*D>G1Q{k1sN-FNHVZW4MjqzJ0y?{xS~67$u~2(=Xp&!UIwul*0*(@XAbo
z{CoQK<?o}^9x4Y6l{7}{8B`z{KVs=#sPHI3MGTE*Xov?ypGk2+{7U`v$8Xc{%Rhg9
zXgVY7%r+UMGPFGewY5S4XvWs^ebj?52AtR|GK0A|T;Mx&_eD6hCjr)saFP}VeSxz`
zXhw6fJI9|FEToJu5Jmp<vz-3L4}avw%Dd^$*z|ktMo#~k{u%6bMX+Ow1;I)Xx;4Ra
zLgAgyVw(u9ypZH?7qL+@L_v04DtmvJ*Q3}~mjr=Yv1~1@w$*uf9(f<`hgT!R{0uWK
zIED^R!cU3ksAi|p-*_ha<S-gr8whXnJ@^lNWCv=ao4HBFZSMfD;KpOr0FMUF3u}1R
z0SU{J3UCx=9$)J43Ya)a&VJs8#BP<ZKp~~j8XpaX7Ft>su7p#*LLmn@oHT!LJD}LD
z@;VgSTcM4*XP)0IVjHyuvC7aMYCGr9%Ykp9W3Bj+ep1xaMhYY>?xB5)G%@P9+aq6&
zWrO64k<ajOU~C;CpFTYz%8}DNdVFM*(;VBiRF98x8c8V7cX|J<A~tFUiW3v_oG_=^
zmkm)L<^oNT%NY;fb^fj)1hRh`{(Np_atOo}1|J#z_K*Auq7qq%eGchs$?om+4AIE$
z7p}nmE+pu=?J?hdPTXlDR=dZvjum+j#4+qCSP`eemig*0#%lkVu3+Vypt~sKRteik
zGe|`Ze|`KUB93(Ur1(i1INH_pDzZ&T#6RZD6ch8pR=5;rbIg-jhK+x*H;`Oi^bHHa
z`$Hp7RleI18C1e@mb)u75;(E$)cYbM@nl7bce(YgA~tF!i4zy~oXAKFt8V3+5W7~n
z$z*<d&m3j*-DG~!8@Xxsn66<JhDCQ-@m3KV{g%aY>hK75+EA5K#_>Y3E>A-hj@~m#
zn8PbuFi?rgiHj~2!mEF^w@fJ|HHy#sH~1SuPee6=Dy>>Cv7O1uRVgohmM1mc!rb36
zXj$JYI#Wu>@;j&Ut@RY7M7O(yiem5J1Zx0A9@k%D_p{u3c&l|@GRCMNWMv=^!xsqq
z;b^SGFnQRrH$wuJS-JF3AaFA<F|)d#Jt;Au<gp<I4ZSCsI_ZBi015Tmv6&;@br%Lw
zZSu^;w27<Wg3B>$qkYyhXkJYl3?KS7GK~=ITP`H2K8|IJM9w;QZij4%vyNv0?NT&b
zH8m>E!2P-8VzN4<Gw9QBa;ViM6uUfUj2!(KRtKm;$^dHX2-9rr%zk~O%39`KKlTvk
zaW+EKGdoZ-Xq|s|X~e~SA&-xm`^TBzaB;;oW|;WKJmJ0ge$OSQ_FE>GU3NlFy^O5&
z%b8xLt6KK@+4L;zK@h5*9bMUbh)AD6z1Qhbre@a0l-9Q6yY%ikt1dBW_M=mMtMFj=
zMj@AmOl^&6@@8!tPbYmc-eHLwt!O&-oy$-IafJIvTh4#R@3k*`H3C#*lMi}1F?V_!
z9Bm;+0sEr#%iAH@!`F07c3JcG(i3_o9MsZJt!cy1W3~tazO~s_XPECPzO?Eo8UY3$
zOqT2=c8pYh&EC`X8a$8YUK8eCZqa5$`NcKc2ist=u0EJji7s}l&$4HSS)iO9W<efX
z_T=b4-?D#iQSq;w=%IEYWf2rIAH^AEEZv>`+LiB=^PJbgXcZjof6*6HiW(O{pNl0w
zA{2(R%b|zScZ)@<rA@J6XVo>nCXc+rR(Fr&s^`aL+Fs{(h9BBKn8dd;Z5NW5hqmPQ
zF@H#a99sez-GS3I<uN}g6dM&b<w>1lf#kdzHj{sXG}R3fJ9@6u&UX%&32iZRDx}#H
zWzueOWJRBn1NHjWZ)f`C<anl2?Po|<JBhBV9Lhmob51^4Ghe50n$|K&0=XCl(US18
zI$#Y5onz-v(UC_c^d*E^Fk1|bJ&CnoLVn&${Z^1vEeCC(vL~f6nB+imHd<{3fN6`H
z1DJo9p<#{4wu?v?@__=5Y=b!VQN7E7);IupB@2k7Qjg<+!&!E}odXnlp;3*>mBjXv
z2dpQlh8MqQMtES3O5laDX6&Js({DV}ft+UKJ=Iwail6c@?yb(DB4Me)U^;f)l)x-i
z<qCYs^os~4zfGGI+kNfktv0kqSZl{&Uf6$XLkB`C7+del>=H1_<TFC%X5AJ0kAaP<
zD+9?}=B?EhHQO%YnRkRrr?Y+Sc1@b3HE$SQA=PdlR;x866w+A{n;9f5Nq5z1!9y&m
zJ-k>8E_49*l=*HJu~9RgFka$winZXR05?Br@sKOJ(&NPaZ$xgZJnuKq6~h}X{KJ2%
z^fnwVV5N;0SgSQ1C*}wc!gegdL$|#cpJIfpd<t*Iq`5>1+mTUT*QS=Zu!`+uoTt>=
zz~2qh74S9Zzm6UZK52o@v2Z8Q&JK>`<B=u)>P*X}I%Hc4oH4^tZqAV!TeXMsvZ`*-
zea=v}UBd3$;R9;N6I@q%?iOw_4Jm*2J@s;<ca%-$tybHO?iH&M2?p;;TuD|aQ2*!3
zd$@A9fvy-{sbIJ#89to%(ljsnTv%Uz)YF<<D$;2N#TDB}KcCv{+FpOI%q?fq%tpiP
z2Jnhu5Ctc1PcjT@mZX<z^<~2<B4MypZW=JDvQ(%K2ZOb?(d$ibl4u4q&$E9!z&D<y
z*$>PWz!@7L_awl91gXWf={)F<2*y?gacq-eZj*MRw^iq2XtDj`bd*&BWtdpE$igMP
zPAxdro!GfAT#|d-X!qUrxng&jGWo~aUGr+wTYgUovTakWoB~p<M!A5HOUTS;&V1|W
zlEu!@rI`m%mlCN9r<Jbswb6fCyHQ=YmW$c<Q*7mNcDSGvh~GO0xAVMiFx|o-oz8^N
z9a)tN6(Tpv>)zFE3qP9ZeY5%fHgoNTb=_DtWb5!8YdH{F``N@Y-CphI{P4tB=Ot_h
zh9j!qa*jcVvT+!GIyZ2~fP1=J*jedM2af%1Gv02Pul%P<%G#e253My6X0}C=fvH<y
zr#8b%T4;t1{eCB^V6EpTF`-Q2+$M@I9Qax88~+PtFC1ruy5as`wo#b;mtkxf6ag}q
z;f*~M0Wg>0Atfh&TfLIxHV)3`Df9wM6D0UowKY}Uy*pdw%=Tn1M|Wo@x4}8_>l^&1
zD7B<kOTJ_Rjd~;qB0mt|2aw{!9|Ir0y<Fd4et45O7-kh1F}(ji0Nvp97dTziYA^sy
zOBmk255IoqTxdG3bRg$a(qZVgrL%hdB*BsdT<Cz;PX-}>$xA^8(C?f7Kg)%uRlF-M
zL4$OklV(nbmo&ho*GY0fPlJ5@?fu6Wo|+GIhUWbB{pEizeFv1V3@i@+`u^f?padqZ
z7(fBD)UQAPGyMMYQ&Evsb7uW7`ulC9(=tvvwbpbtUgxjB@!{We^G6bv&JKSC(eGZI
zGpF3}-{BX3($8DOEfC^dYo$hrE6I#Aq}(y-Z$-ovq~i;Ll91;fi!ZOAG>3!^fAjx|
z8F`O$mG=@E#iolR<dk%bqkE#4E9fc3==Z_s(Kr=Kl5$I$bViSWIOuqVL)HC%RBkl<
zZUe0t9+3g*KiKe!DNb&KhM_ds@JK^mQE);EPD*8eS5ehXq`&Uv$`zHQb-7Zcku)i~
zhqx>amTND5@uL1E<x+E3QN%<4_?Pir!O{Puh>1Erk8l1Z^;T`YNj>CIAYS$E?jPs3
zjdtE`j}<#h$-u!$4lXrAV?P}H83*5OpcTUl!GshC8(s>ggjY_E4z3tpW`_ufpl2P4
zDd^~bn*3{&0xTX>#AG~1VzP><BzBqCmgZ5OSV>l>XBEmj?+uT4&C^NDYqg!2-Dun0
zHdwK(_dg>Kwk?q}L9rW<sXAxd*1DpF4=c(Am&-iLJ$$z*%IdDRAZAv_$|I(|Cb4Tg
zcA0j!306$YNeSRd@tR9UR{V(9&&2E923j_Mys|OXIK}YF38qn&JVqzlg7K_uyyo+g
z<VrzAMB1ZV^bs`MAll!R6^KmCC(eP$QgThXX@;m}b+?$tc`BJ<N*vZilroqettPp<
zHd9C++(vv=@WLjOTrrF6)61gqwxp4NZ8i3mY}#lWjb>KkHk<FZ&5F%+OeK!BIXb3)
z$Q=k|W<*Kv1I16a_80`@ZoSP%h^Y=Smso$zRucMQ{Md=ErRmR3bYV(Il4Lj9RHN%M
z*RGA0+>Q8(B~?ro&hcL5n9(1lhG%BB{RUbwJjP_<9K)k!mX}gP&F7Pg=q8lAoS?%f
z<o4upd~C#ScOO@<lYxz&1D!O?;z#y>{*=$`wa=Q_fhn_+I11)$Ry)_S2UV=KES2zm
z&3;*JF6;A7abMTpu;PJ^a?)Bs_A4BCjb0pAE|kn=MLN4Q)v8x8pA&V%H1+EIl*t9j
zvihN>T2sw0m-OP*a79-g#FhQo<tysa>$pRplVeVLEvzOWMc-@RilE8VX&*&@TL|^n
zlID3lrKH49W3Sm$i`1Kplev#dp)wg?^(q1_DRH(S5hN2nZ;8m}C>+Sp9ZQf#_}mfT
z{5B{U=C=g$y@)6c2_|tmj;@Pnp;L^Ti5-Y)#h5^syo1o`Tle4Wsuya<kgLh6mu61t
zyy(?w-4eygMHQLUD9T&PqF@+*YFJ^%T+cvyto-@&kD<Hr^S76_$Ol6AngO~6zbs9I
zmZUoAu-tWR>fdwS+q7sXoujkv;iCyx3-{=nVX_DcCb`}d#FpbWQt$<^&&2JARcvgT
zUwTz1kuVPyVj*EZ198ruGue{7DK@C&_I`!rD-ki5MyseYK40;)%%?wp#qCz6;{82b
zK@JQ%x{MA!TcI&5$oFhzV5NLnE1G<Cc^eQr8@vJo*G!-|XTcCTCFH`5sAvmDT$y_q
zh}}(IgF=R3|BRPEs$yecBeh)N5$tSfnwtsIlz;P>|C}(#Nv&uiu63Fz@p))|6Vzs!
z-aD;#=eIZ;cG6*eS2odqj5o`~A`+tK=RRSHIv984+$Twg6UR?|lESZ%Po*pmtJv5w
ziihU#)v`48Nda^d+NlqC!_#ffBGOJEQ>H{Wu24i~ee9IDzvnB6Qed^mFH*(1K+=QI
zEK;eZ>P(sxY%Rxzbz+t9x`uVXZCLH@(=t{RGbo1XbFdP`jiIi8>-<w#?eEhXR)S)b
zc_xNDtYTZ#^lG)X|4`XO&}^z)aRJFo;$72YQOgP{Z90@vq?9@oB_wZ)67fHqSC|&l
zX0UWoOxiJ?cjG)4+97Aw%f6KTUQ4OAP@3Lg`ZuK_!D6vQTQ>9H0ksnFtPy(65nws0
zjHF7&4A6MVRg7|fzpbZ8n5!~HxkoEn@qM{A75`fp0-+6<>L~I4<CUs1P|+|gS1VQX
zL}}3qppnpr{tbI;FoMp;PKpSg2@<D!mY)it7#GkkujdlhHXliRau&W&r|w;km)%yP
zcgn07CpEW6iL!SwI{NP2A<Dw|y<wu4-WN1d0ZFG_>-yn;wo7z@MQm%kRFX-W&^l?k
zrJc7z^e4jtkjBh7s4ZyYxHJqVG0YqP`3!^H%n32c+kk>A^<Y<qK@5XwYmQGbz!-+k
zJwp#3R<W^Vnx?UnHf@1+G#P&^(>=3Hx7$7|W>-YR?a?(cLorBGo0TPZDmAet7Us=I
zuG>K3&${k^h|RzJlYv#d)vfp?28C@+WgAkvS;cb*J<`!}ghcJu!QYM23i^mli=*>;
zR7_Y)XT#iRH6vAY^0cd(?+zdZYm9_2K}K-#OPZh)r~$%DBw;G;imm&5WV<a7TKX61
zS}GjaYM?d)2&^NU_txqSgCnPZTaC6G+!dpxmB@2{LZ#I7Abf<%XF}y}1FaZd2xj<6
z`&)r{n{#!4YsxpuZlI<T*Y0smYo$3g3B;-u*X@UV%}%?!JN)v-+Mp3@hz-|j;O65U
zm>UhV8^9IA5QYX%+GFGkRa}-`wRd+z>adf*GvNx%ovbl~*l=*a*C2r@rk{!Br@~cN
z!*>0DSgRcfb%SOMog0m_8^jgka87yaNnf`Dr~JgIHf?RTHp?`nm<=e693nV@9_KcV
z?8X}7{+=%*Y9pIIWwblV0Q6iLV9d9AtkE`jiyDYJl$)<xMw@$0TMy`UmC^q8m@YBb
zu3IH0XmizRw4wQ)-pe+t@3zje(M_b%r+ATn6%17U)-~_Fk+sw|15f&_YV0)QPfhQc
zYx-f+$o$xFQQ2>PO+#YkUE55x8`G5!paSnc#Zm?sINKI014`FsaGNoo?kqR6q|$xt
zpBM4?OhL}}`kH4|$_YfW&~>x*-Nw(BcEUO-{dDKQ0mcnZ;R<DLWOH<KWnpa!Wo~4b
zZ)_O`0Wz1dO)3<ZJz_E^mv43%qXEE|6L%S+1VK!MN4L3m89)JlMG1B`poR>lT%;$n
zi40<j-g8S~l&2(K<V1dPT&Tx|+vC0^GJn?{ncXPcVHw<(t+FwLKUTKVk|LgzlT0HN
z(b`(Pvc}H?V>)2Y6U?e2t6`4Wd4bs^vU9w8Zk2e=Ty90YaJP6fnOYdvY!mOe7*@n<
zVt)S+m#r+-+ZKO+`0O1%dt5*(g4eOyc#Pl;XGWH?U!@W+*W1c2SB$yZcd0VmY8%5#
zS%FB!lHxHCDaVX1`An18r2^Y$lNqLz#+uZqFnu#Z_I~El`<XgrmAoXBDclsXKwn`7
zjygKmHZav3`A5yMmt-}^9O~>8I6IEpWIimL6`5nKCXSUbIXb2&&MCCaa<jFWl4MXP
z2dYENrNn<KSaj9OvNK*euHiL#uC>l8NvcXW7U#85mmYc<`+pl|rHc-{s^L_MbH4Zd
zK@rYPTeX=cwA<+J^@kG2P5y0sg)7JjU{W02#}dAXBP}MUp!gI3sz6o0*h*h$Y8CP3
z@ir)ir@R6Mmn3iVnC%$kbjP;mQ^OJWJ?@5Lc*?h-0Oy$0c8{gs&0?E#^KAKTwvUN=
zpcLbP4+?vriCsB?Qy+h-2vNu0ANf)&YZ+e+e3}*`I^wW>b~z%(NlO`n!y|(nkW+S`
zjt_D`j@F>rrTX`>*f=xtd&WMf?*@Wi8eIpuaPp)+8kD}rBS0nixwgm)T^4n0@Ec$G
z3Zmpbt#ps^_`o$2!pwr-WyAGWork;RotW*LUgPV!jhA$S(&>N4z$GaU7Rvq|a1AeO
z1ubhSvzL23$=xirIZw}4-)6fthz7}$*ep}dXjnR4DwSkhl>AxZ>ocG~9$K3E=d!Vp
z-UK&uo({f+Lvx=c5Uw3)l>khO;>`R3Ry<c>R4NZDg-nkSkoOCrboR6<B&kZ{V#YaV
z{^x0%B5W7%*Jppq*mL9AMu!=ngy4Rml<y^;pNUVwAG$UEc-mrwmgJrlMbK`1T9Gg>
zSFy@-xWp}qG`@&W>h`s|__rz*+-L=yt7bWtxnp%WF^p&PR!QMLVVyWmCySEAV<|tI
zo66rr;Z^UCi8Qxxbn;b%V?|+J(6{%)+6qUPLsCMDm*0OBu4SgJp3JmL3Z}KV(wcY%
zsMiBfuP1#=TMJ=6yxL$=)2tKa!g|u1*?QtpY&3p!e{4Iv%_F(#q%IG0bT5+>LLk3n
z&P7hxc5J7^SA~+NkZvjsTAX{?kZW<knp5-u8Na~SmG_l4O;7yll)QzA%4y5ON#?Ni
zQxJoq=gWUG=Tyn^&GrW2ZPI~ZUqk-V$=TRYV^LO`JxSJ-+5@H8<n{LnT7yeIvLChJ
zRf@WgUlPD&-(;!OFruGtN$##!S&PMjp@dYbS)C;{Q*wWyB`m#k)4z2v_g7ZJF`!7-
zb6N%`cOW}z_!zF}I!>Yu7<X|*&NS6fJxAQZ(J{9aei=3c1R%YXf0qFl4HE)4F_$rk
zMkJS@za0w#F*lcyJpm?v&01}58#faEo?pQ)0eT0R%NdeG@=ydwlXDxO#RYK==$&B{
zvaL(hI8rLF+qA#_p5aQ4Hj3@l`lStEXUH9LhQnuGNZKb?j59vDU;^WR3WmLQkQmRE
z@W#iiL@^-~2=*rCq&>;RqMw*dsyxgl6?%kE*<|mvCmM2Ghb2RQ_@`6`6^c>;RM9@q
zhI~;Kycm3}vlT;;l!8hmMd$Mvkwry#%$!TmJ~C9$^eH&QM5=c1hFKSpd;;Ywl??}{
zBd}JEq9c4UEKX+*;7duRMKYXofQ%49l{z!jbxCKA_(k0aOiDq&MobD4lwdKT8ZKfn
z!Ry+hs0r0#5x<0g>j(%ENstlM5Ke!I>xAQ%PG8q6;49Va$(!KnEW##`uhP)xoqiQ;
zV30aXG(lL`D`EnshYTnla@MbjOeokU&KO#xPJ(T8P6s40ktz;ghseHC1yo&lb|!k#
zKIjv@t~q4}RRkh1UkqN&mjkF=M?@1N*AbXGR%uv4lqxNMLxu!h5HyR@sfQ7%8m(YN
zDfpV5GXjG{Mp25ZZO90#{S%NPfQnMZNN{Rg?3jd5$Fh+K)j*-IlL3@Ts<48EC5Jj+
zFj5lMk%5myF6#)Onp8$cGqD;s16u%|QZ9i~z-D*|Kv605NgWHE0}6FSG^t`C1C6kz
zgA5oE_Nsk<1r8ITbB2#Vwo>XlPJx(#6tpT;41R)A49}uUd6NSW4=R})MKy(MWpUxr
z*Izf!4*{zP?|#+%_@Dp63zN+Xd0(spBIg%pXGc$-fYUV37H5m)!H3Do)Oa0WnwOKd
zoi68wyXHSXw%;DKlXhyT>)2n+Tgcc~cXZfrm9h_iSP%!_HKIB>m#%@0qX@&)MnWJc
zwE6-pQ-QzK25f~dD>l&vS9wpKG%uHnlY?n{*t~qc-@KlFY&+eX*S~+5Lfbb}EMneH
z=k2)$rf%7<rss=`<;nEC3uD^#!}N4E`DXF)P$i1)L*XNgUt%O2GOxs~?MGq5eggYp
z3!&0~ecNusT)QrPWy;>t^YEoNPv`SRH<4eN;;;sK&_|a-E0bxySu9Vd%Q~MsYQAe;
zG|$k$LXIH+q&;MptwRi;n5;rO<Y*&Yl|-y~IeOF_TpYJL;rFxoyXNWT^e63XF>el<
zf4_R6KVQ9V+YjgeXqwZ-%r2I18giD%r^ZEpCxJ@#e}kSIyH+lD4dJf8VV^Us-GVt;
z@(=75J@CWh+4;2kg7oFw`U{e>TVeWE*pMx3a2Vqj9B*)oZLXD-5qvz*^nea@oyj*s
z%;#Zg)BaL-M#h-j`z@1u8sm;@@#>D<TDr1mlwS%1>sWgTHXv(yA1$&B#b8~b-Q9A3
zI=8>g-pxMfimHEVt9iYgP3CXTra#T*(_bzoXS4SAPj0%4x|GRkX7FrdfGZH92Z|3A
zD!PihQWr2hgQk;ZPfx|M=NrPw2E%2R%>a7551jX)Jv6lRpglgX2HHk*8H47qkzBfv
z*OljE&`iiGf-Dn4wxui6tyZWZW}9Pwyj_-YE;S|PJ=zOGe$P2y_R>fgCjxF03U<I5
zp*T>q8VpdvS!CGV^_(m}&VH?$PL9u;_mlQ*<GjNuK>GL%?xBj9t1G*Hq}|MPSK)bQ
zZ00R&1{s6RNT0A7;y6fhw2tcfZ9uI}8Bqa&mK|H1|L!VeeZTni@iYJUDV_g+@zea`
z{qc19*~X_E0@nt^W*y`_{b?B*+y3Lfj(>&~G5*DSuBY{I6(GO?7(BZU7jgS65d9cj
z$ObMVj|^2PGWC2|ie@xU*$SKi$9hDtTLmhu*X1$q6P3@3;TVS0Y8zaU9jbFZ!=oI)
z`F;Q=YsB?KIEigW%D9;`6m^+@c<(s8rUdN8ug!X}cTBxSJuYPnnvw=UGbQdrGi{_(
z;Z-^nzb3+(h-*T|)r|;ePVa~k*SayVw+`VlX7rcaC)#sz2t#|0(jLM=+{@z<EhC|3
z-8@XI^<sA2DHKkZ(g-(uGi%>o9NUw{`{sYk>FILzZu)MiqX9i&&Hj9UHd4>*Z6Uc-
zG*GtnzI0bb+lzLAV*Bj8Ia$nSC(Wr=+kZ@y>k8$1Gk;lA>Kav3Pr(-TA#A+Qg5a2}
z-dbs8UQ#-0o*GPl(0sdiy=b1Buh{E-8wwH9YrPap)~&py5?%BvCV4ACZ-xs3gn<n?
z&`ug?#F2bMYTRI0r$+IAV)gp+W={42rDyol?$~w>L@8|rq7;V*qEEI+G9=mbxK_4x
z!LpNUI4@I(hj+U&gcETkNE*NyO0<#vb(|>H1&J#lf%!6o(`&GdG=P(f*dj87GdUcA
z2XW$XEO7uMiJFs#a01l6m*pS&OxwR6?%mLczV@uMI2<s<^+j5L&_>?jP~5xk>0Q3<
z>Q-gjtx6n`-*dOjF25JIO5%mSki>V4sR0$JKG)!As!veZ1FHpdL|pe<G4*^}6OZ+5
z>Gt8A5yxH8N2r1vqOW0O^-vgHu~EuZ(*JI`lwK$=aziLDUkW#3Ip0fw?jiP$ZuPU<
zx<=I3sWC{38%SDzSBhm_DVA`1ND-qr6eLTo;h!P`CVWVO9);6AgUVB}9OI>Q9}%+v
zs;i_Ip&(4jhn!I=a4L6DsCYP)zD2Hd2q>N$_F;?;fUgGU2fDtw3c5R^wQQp`jVLFw
zyQWIJkBVy7g|40lHy27g#x1)Cw(6Fd?jWxFoNIy3C~Qi9{nG#~)Zq^<+Dx)*xtv$8
zr`OkYeXjLuv$OiQ4u9Pj)4cs+ng_j@ZlU$MXUD$+TDPD7@}TFxEwrw0>J+WKN+P*?
zFTQQ=z`4ok`N`RWlNO`&<fzpw>Rz0P2h{}QrAIgkYwed1u#^6Yj3uoF^}$K5k?tM$
zyk*eyX#Ua;^Qhn)BA3PZ{{ehnhZ2`zY#9`n;UppymtBe(GJkE(7{uVn#1S#U2(rnZ
zh?OOY-fFE~?Km1k>qwuk#Em*wD1u#@m6k@a*i}Fl@LnPh9Q^e4#7|#t9-ePLT@;)Q
zvnY%_J^y+V^n%l0!RevKlM#YxB~H)ZPCtI;Txz-==mO4_qRY_#mhO0WtAwQz;+`&X
zcWXeT@=DSr=zssS|DDT)+8E!;`yfHN&q*?;%a=5QD=(AEiL<}o{q+394NuMgbkXwU
znA1{2+jjaG1AEe1GHzvP`iGxC`kJ3E0@rH*GfqybeCAS_yI4!F*OGz>rKcbI8<&{A
zWz1?VMTp|`TOUNKvQ36ojv+!JgCIeIT7jWoNK^n(1AqRh2M;to1xLRKV#GV)e-n}%
zi7>d(1*z6Kl_8S&g=(T|<=F&+6>Dvi+=Bq~wkf#V>XGD@p<UEMms*9fRsMg#)LNT+
zI_;rvIk*K1we$jkOzNugNN1T7lZJix_s2~&@az~pts+0(DrrCbX_Hqak)d4h8aaT^
zvG1NOG=Izfg&s*i$YNjh0-1o`#Xs=~6g*&pOp(#b+2$d%a7!mZbEeSF`$dnr+13lt
zj#;5L5P{M=dM0w8Me0Flb17kC352Bc36D^NaKv+H%VLyC!_c|5)p>LTBu&Bv>bL!~
z4kiL6bM$H^U3qi&M5Ya5FE#fq_P>DpU(%0r?ticROQ_Dsa?Efu|4u;AiKG<)FRxR^
zn_|p8<Vgoi(2mOs*cvlKeFNWIdMAedop&Cwrz-0vEcBk_>UK0hvV`$#seE{Alg(3&
zt@G`^&!t02S?cCGN%O-<um!2sx|o@D(ZN;mb)6E(G^k?dHm&g8Eg7MC?@WFwA7kC%
zB!88E>fI*Zr`lO+Derq8B<N@xQu);DSVrmOs&#(SYDuh7SgmEul=q-xfw>l^^gW#<
zBrdWgiP})wn)D=CExU$`Yph&FDfXoF3#0C>$aGT;nBiPotEy!TbsO79{)0-Hgn32<
z9!l#fa;uua71P!hO_<229HQB_CS)J2v43-Eb4EZWiLLhYbbl<Psq3l963^S6M~yI<
z-Rwa^co+JWq7gs(JwEZx>mri;h2q)b_jr1rv?oa>&+!21d*{@<WLe0ngwT=Pw~rqE
z?aS=f)AP+g7j&Tp2K4&pZ#Vu0Q1GTGXd~ll9iD#s`{~!4KhLGM(RHAYYmGRekAEUm
zOQ50^%+%;Q!x$jgMIAkOsBmQCl?P*VCe7fE5CAoz{Q2$cY4+qVUkVCRNZ+SfGc%T{
zpS7@IlH%)GX|<;lGf$l1F`OAA3p^c395&!Mukl}c@Rhz=@2i?10hp%HWUdL$(71rC
z3Ry%JO&tl}v-JOpcF`~v+<KWm{(r<zztGDcsIYSO^si9#Tlo$@{p0jkuWKE_mbA?a
zMn?I}hFpH~@rX-Os|4R98l$h4u(Mqfd;q;Ime_YVm%BNd1e2&pxMdPb$m6q_>s)u@
zs^Y$n$fW<3or#tg+J_115KkF4*@7z*Z;&2RqrK&+A$1eXL<{KMz6haZntx01cU=PN
z5?cG3B{(@ZL1Ys$q>#J`Xrqgp$Y{fj>SFug4};VJpL{p@)+>V#$Rq^>$Td^d+<OD;
z9WWW(gucWSPTL$9)qp)=<}(}$7^G>^1dm|~B0_jzvrj81n<m&QK}8H%ljpOQp~^)Q
zjD{qol^e7}#RN?3Wo;PUZGYCl`3WLC##Vwuri^;Gm2a?>;{<hxrvOSweGu`M0$NdL
z+*BG8i!6t5c)Yt6GR(mCCWNrqKR@1l#WbFvo!XGEnhGe^n@=SA5o&f4Gz?^iAc{+c
zHU|-86a)WiIsKFq=75Hl0q(~ZoXlp4jv>yIB_NxK+vjcx(DL6jdw(oXT{Yi6ze`_g
z6rwaKQNEL$VGujypks17rXC;%6*JD9K+BL0W^yjM=VD{VtdCd-NA{sM+`A5H5M$?K
zP(#5`nKsR>s0E$X*=_GYO10)B&T38bzyVzx?tKSU6#~;^>;!Oz#;(mBOk>tL;Gt9o
zT~=GZ0=0;62T*y6jDJi!w=kDg48`SjvgBI3&Na3x!)el!7v<jXq<xg04(l>T#K(*!
z>6jLEbt13vs5)1?0+3xBH({O(suN!)Yys|Mr6SbNfp|MNlSz~PZ|x*&7}XA2!bigL
zAhv{MTDQ~w0jP&Y!_71=nUSZKS3{zxSx<HMYsN6kYwBX1aDNUYJ?vHhx0A48U^|5I
z7{MMx7&pvtF&SQ2eQgu(;`;JA<jg<K-ojIglObdrLCn({$&DEt`D7C~7iDr)A?DxP
zN#HQh+XN~XBiuvWqH>xUkZpvP(eVl#;)Qb^YB|5r%|aPgZ|bW+20Z(=l(mRFnp2q|
zKAXc!zVCy29DkAy)K=!@i|daSWsw<M7d1t?9^)*K-^s(z6XmmWpFgkTcLmit&2o1i
z!Qq~E5HvA%Jw{m#Fe~wW7Um7JFyjPuh}SW`zB2L1kkio7irE?%Oap<E8C{ln{ltAw
zCtfbgJbt@04acxe=01mS<x%stObk6v-%1XvVPt((TYulK*OeKBuu`X+mAM|z!+dmj
zh8R#EGr;|>A8y_U_*(|}FhL#SRfx^6%e#_cN1p=09F*kTM|fCwcbODi-^XRCu(a`%
zgY7ba)?)CrVtGO99RKf0!xulQx|!d4+i;%M@8oq9{|>D(M)23=RRSZ}sLedAd1qds
z=GJ&wn}3!FJNc&I6byA!?0#0P+EL+}Vh=AD%OBNO;n!|<VIRN4M05xiF~UE_U`x*U
z`?A=Z?Da4|9o~6NH;&<*Q{&BKCstWsPT|yA$~ar-hrL5%xGUQbNf+~pLm=`y<Aspz
z4#3GStkub8%uD)aJ{uwDW(Jp&!!=&5`=2Kh-G6DyxJfbZH|*Oa0_IPXHG~<HO&d*P
zSl8RQi}P`AI-J{>N?dt!hhqrDxiyT^{k^j(D;%8HICyD6c(TA5W##C`tvRBFtd+U5
zhDqJX(yO|$kI7-gJ4~9GEF3d>jb_5ZyJvfYXB#J|L%fdJ!ZE~aWK!vN6K%}rr?Z*u
zVSk6TZ;LNh1N7{2`xmaq``Ez_V?=%ob_h(G_w@ZugJzhZ4)Hx@+TP#%s_Of8b4#H6
zcEa=0Ztmr_@t4mjEI&+kouRM?9a4}q7)*0U^N>Rd(yW{zdt5yc^6EnjpqZmi&-(Gw
zDF*aIe_nX(U`<-n7(ajEQ)o&P_O)CozkeNZ!Pma?7cZW{pxVqh{C%9k(7=?0xt{yC
zt}6QLj$jDwm{8*SM=%8WT?gtdHtN+PcD9VXnzL|>%M@tIWy*J#`39FcPEdz<0LHIA
zJ&+Dz2=abPsQ>*_7~?2)&{xW%cg!gap)^y?f%!k&DU5M+JFH2`#8FAPkVul=*MI-t
zWGsgX>JU#HiKCKoA%H=Xl>Zk_VVLiA3S%6|4nY)|ilZ`r!4-p=`TyUZ!Wai}n;h$y
zRL5*Pb222m-|Yy-XxFv@Gd^rSXwSG64AK35k6#S;eH&681JXn6gW*iSAH5hqdNIt+
zBD<g6{{x`&zuF3AZe(+Ga%Ev{3YUza848zi?=%#bw>K9jmp_#mpnnvNR`%_;Z@;|A
zGJQqgs{4Nb_VS;Xc>&--7zl^Ie}9P$V3aUkA5=hLtv&ql&%^I8f2|~#v=l!6qR%e_
zO)n(Tq=EG9BF$falZSs$<1f_N1b_H%>iRu)mm**|{QK}XqAz=ndqo_V)*=oZM<ZNt
zL_7!+n=;3N_yHFwk$;#1r+lwIzkRYY@>j(>;;kYKRAk#FwrFZW;yl<k9z2iJ{Wz1a
z)GEt8FdfD3clkTc0poAN0z)1Jdiz}&wT`6&QfPvZgwXw5-9QUb(C;Uqg`Ox`MvDYB
zbLx~ilQ`uvM-VfoCiKkm>QE$5P?xc?<8Y+=x#u*~>&9Z1n147vp>*0U0bwU%cXM_F
zF#w^m&OQM#Z~}}Q^o!-!H~CH6gHGOcMkAqmDfjXBJ6);oTT&OBA1y&{z})r9a}kDK
zm_ieX`F%S9yqnP*02(7yP!9%>T#-gY#H|45D}a~AohZOj#qiuzP3h!A(qmH?Tr`no
z9+=@+0I`|Tihs4a^hxB|9$xr+C#&yf$p)*n7TV}1>54{yDpc#H%0-q(mMB1;2jTNM
z0+=jVHzxZ1HYQo3y~29;N;fzd<;s~S39ABu)jcEw`;aRco5zv+b4c#Q{ce0VaF;|x
z!J}~}jgh1|(g+goYNUoZfVBmimY8CF&V%=ky={Hc9DfK=LZ5uj&EBMqHg-R9_*Nfv
zt;vbS;h)D=Enzi`Y)x(UY(Hi0ke^0vkHh78C(d`{aT{m!IiP-ueq+$Wg88Z7A5if7
z0onlG<=FhOz#A=4N2#L3*h2;0utFq`*b-bHb37JRQ@(S0RW+!_;4!v1P2EiEhNT;3
z%0oA-ZhusE1Yx~riG^-wf9<`;(#_`ILpQ$@+q*H?z}Ds{{;}9<5VTsFv=mO)6g;uZ
zFEH*?@0mMp9lX<miHmM#rJJko>@l`07@MVRZD8z5w^N?O^xl0K-;Kov#+U>B=NN7b
z!r7mZ*@q1G-2iO>4>|344)EZGmo@*#lIwBejemYUb>y_4%U*kR<L;$wu*jsm;yD~L
zPI#yZ_7L5bL_3C*V=XK3PB&h*qJ*B!jGfI)AGc~bR4{#qidkfuY<~T0S9^-`y{9+>
zxiZH!OP%Mq5Au7V*+5*!jN;KX6$k<;d_`@A_hD~Mr9g1(qK)$!1Vh!!b`>qZpoLw;
z<bT$fzg0n7DQ$Mq*CZPIfV`XI8%V00GdxFmDFF>X^N|m^wY%}zz+L8);W@ZV5W&@r
zx|Y?CCC#>V!<yF&?pk!2u4ov;{zBdT>%7fB2stc&3jZL)*kAutoDV3@{QzwM@11aF
z!=-8wJd<6S+K-ahni=mdJAt=TAKr166Mx}(%X!CUJRk_LZ8JW+8GPjp!naSx8V=V^
zMbqOZx`kR`eABA84ZQCpvu`ynQZu$}+qbc09F!oeZb&+Q!|>h|NY}Ca)W;_{lsfw@
zP3~VQWFK9$cEJ%NE%DvV9N`@?rk`l{E44LTRf~uLu0<=HBJloPzZcgPT;(F3aerkE
z>!^a0La7C+Y5po@7&8xVa2AHLaNU+#NDyd&vSz(IxgF~w&MiiNh}-V}%x&#a0IB7)
zhImxP>U1Eb@^h4UGHR@0T9;>W=8)%cEC_hZDNEUMN8Z_WwRKGgJD%Y=_DnkBIsK_)
zd$c=hD|+PJ>o9BO3cu=%7JJP6UVo?+pmpgnQgkO%%C^8T*>tui?{>sHHwqcIm)r}F
zZAT=x>=CI#qgRtWwr5d(V!Qp@_EeQQ%eLTd1`(K0=e}DolPF~YM1SxCgR^GUh)Y8r
zsS2Puc48NS0AI7|de=Cd-w@>L8=Mm4kmfgxW}%z~(bH)jIi4vYe%w+fUVn@6AZ!%~
z-EE)Gf}zoNf}s(60aDyv<@h~&wmcFia7*8t9<9EE<(>4P?u8Y=3uQgr+3BForS*?*
zikh#jFKlIxn8E7~5~LEz@P@X=dsos4{549@*`K671NT|9yOmDGkSN-Dg<+uITf{n_
zbM+ewv2c_?1sskI=&HDhZ-2_1xE63;W+q03vZ+~ZR(r}8E8orPlb^C>Yoo4NX0@td
z>wZu>p!L~#eNRv@11hQT^rq;oKu0V!uZ87#Lh#4=z=$WVpKtO308_;Bd_JIx2b5s(
zrZXF9KdiFSTS%m(%mjMRg`N08^eW(9H>O+NW;O*?5c9|lB`&?(Y=7X2PfM+FKR;Xi
zdcv$2lI9iqefHT`xIwMUEJF)f%wX`!o0ye(bT2T#-V%H*dawJZDM0|=am%;!QYpZh
zX4xryZE2j(YiLG%m5%F7h3)FaeGuo?#R9x=z8R{GBq<`Lon*7ahU*lcDmHCrGHq%Z
zM&?L?$3wT{kWTbQ#ebHOm-<|sMWV=jV#LdP*m2q>L1X{Q*WMX9H9L(u>n?5Jr{^*1
zVN8?zajJJGSLj{~P_@&u^UwX)oPXvP)LC`z+*NnJm;YQY?jyAzK<k(ftVA~|h<5b7
zR=a7*u_#i`s;3oVlZJYT6Io7u&kGKZ&Yh&|KJx3MU*>5>zklgJ*j>??4o-lpMSU|D
z25QudzT3a)Bkv2x!FXK6@W+m?^Wjy$>Rg*F=4ui(7$2)8XuYLx8>Bw`5sNLWEnmM>
z{ng|iy|VKP^<{bDx}Q^oFBL>D)BgzAy$2zEn5h<N`OthRP7X>h=IfdhPn;=MpY^Qp
zel@S@9G<Q=!GG7hWyA_@7gIc~JDBCV2X0+0V@MRvsd01;?@Hjd#%oYcIPI_SdLhX!
zZd2KXB&X7cWc}h}(OjZpj@aH>mAgfeOzzPIt6o>f=-TE>x_v61@Y3gsJf;R<$9_JU
zik#E`Oc!d;)v&#eJhpY^WMCM_1-#WFp4xAH=6QBrrhlKWyp?UYb)PwESjXNvN_B^g
zjVUpi^Z)EsZWgpq39H;pTZd5xkvBeWmzET1x04HM*JK$2<;?yeR=JatjPZ4w!37~s
zYHRY)W2e|x)eYA7w`xzS#m^hJ*(L^?s^j$aWx8AaH<k)s_~pN`3_V5QKVwZ)Yaw3B
zXlmmJ{eMb!cX;x()X;-m-{+C(9a>XkJFt1pIhB`QKLu!M;%cU*KWB>R4{W?!wmt35
zIV($!35LVOa#rJg^9FKG24{-tZ+)&@#fbtwZvNyVl*UKSf20`aKO_mzj7|N6#+p!V
z73-8mVQf7tq}jCgRR=6*B>s#LPl4@seT{olxqoz{W0^*BuG?oqH6DsR*j8)gqI$9s
zuCk6>=6q)!dBw_?*3rtB(QB5x_1|g9H5nA3m~CRjotE5?JwmsjmP=x=EEGwh(Z{_#
z0db4apveldc>49r_2>(=p!qIVuohnzuWpt&jmmQ+QX_SAM$K=!hAkH@_3N%OPA@4_
z1b<TSz8!(A`ATIl9mwrZqMiCUw^jH{rJ%LkuJR8O`2l|>E!iPh{JWSKZ0v1+cJoX^
zwwp?60aEpUE?tA2`W)Iv62Rd}mx!?6nkZ|z8l)=#5p`?o{6msS_wQwzHn8s?%blh<
zJh|aWh@w!zEoOqv94|F~mqXD8(x1vE$~^J3S^9_D9>=nu{n}Z|l@{yZ40Fy$y%N$a
zz%rY<_z<2OZO9hl6L@aWQS`ma^79Q}J%5uz{tsw61VfjBpcxebIhPS)Cli-An=mJr
z(V!Wg0e`n}p&3E~e@py7(FBUhIaf6yM?fa`6n8PzA@*&CeOoiq)LlDDto0`Bl$904
zD_Tj8trJ?rKAMUy8xG$-HNP%qEzGd;p4zMhoQl|IEe&dKqt0_1U<VvKc0W>wvz9M2
zo|Mg-m>T4VG$ejFr`%0yCTKP)uoE<Wx~h7GF3ZFFkGzDlf0i$!83MA$kL?Vg&*WK$
z=#|BZ%wV@8M$T|7P7ry;MqlL3u(S7^pEe_idBcyZIWpjt0eawyi~(icl=as;5Q+1N
zDeH8Z9}W_u4{dL&*K>nG+2Ggg4W(FPvNg$f*}-2W74LJiqlm?kNw&lMk;OV|8H*K2
z@ij3VD%B8ue}$G9@&VtDdP&zMPqe0zVjZs=CqBs%AH#85C51GyZ*q^Xq%cLitKps?
zX*I+$$!v&<Ox!2@wb7^melVWo2nY9CGJm;4IjyCr5{KxfoS&k}A4_$vj;JSc)@nDO
zLSqe@qhL)=*h3CO%#h_UYQB~hBDNDcvr(x;=pVgLf22rzmKpev4d-q>IpmQGn_Ko)
z@n%--yy7Hkr)7JdHpa=aX~nWP%Qv@f&Rq=N<1ki3Ec1~G*4~DFU{;l(Z0|v^<`urg
z>8R;K#E;s27%G#c=y5y#cC5T;+S%xIJI3~W)bPGGqiQB4vQ>$Xy<8IJu!jtZBxMuV
zNOA&Ce_Htzz9zR}%}U{HBf(=m<e{dAm(!5s9nTK`YjmMCY;_Z1Q+kIw)Y9*15S2Hr
zlBh?8W~lVKt6M)#Uj<>3{`Gy+O5#oRu&f>-;!5?fo>MS$Q9GcDl`>`#V_X^vqjH?7
z{oSZKo_Fi*V)khvvN86hzfvYjZ@K)Sdspqie{iZ1u*JA14|Sv_cmS6r*C~j<R6r@O
z=<2j|94cq>MD)V?x=^ZR0i5)e*AG+;#EU-Vr=kt@*Hng?vWtvcP<9zfJ}NxaZIacx
z+0p?Lu?vU|d3Wp>?-K{(t^wRyI6%)vUtV*qEgUQ;3ZO4F;i~Hs1t}%yOIw~({%z?i
ze{k4f!L_qD>Fdy&k6{lJS@k1wAT*XUdDqfwP<akYHswIb36$O8WpZ-=Ccr%r&1#5+
zlBM;So!(?2^7fC)KqabX1z!9DZ@YLi%Qiu=ba!pYT8M=z$gqI)h!FFZ!<tTPf>p4S
zuVOc>(4SX@Uq%O%DKS^B)2#MmIY!%3e~Hr^Kt|sRcg2+%&7nU*+q<~3N4PVCj(tyf
zq&qXjd!pg-{)~-V1-L6iTMe;LGIo<m<96|9#yXBS)`W($4sN@=GdwVUaPF=hS_`pI
zDY3kj`Oh6(s{E(F&+X*mIAn}^C?ic`@Lj{PX7($U`WlWky&n^ffo08uN4l?4f0@9?
zyRV#>S-QtKRzob5j4ZFm?c=_Hj|1mNF}uWKUcP&&Qm60f?^fw=7*X1s{F^l0QT_oM
z<h#!&Fr4H<frz8FqrlWDDnyif9e0%Pp+Lq_x1+#TxG<8c_VW#tlQFr^0C|+|#}UMf
zT;rit;wqIN5n^)?3nOEjbM@QzJ$6oK*aq(V6Uu6cg$B&95X1lA>iG^1cXIUrsD@<r
zRxrE%6u38-J(um;8(WEG&pR&)W!pPz`oHQVUqhE+Y#9^*HJ2fp7ZaCYRt+eZx2PGM
z0q~bpsTrexaajQdw2p+{bKxwx8(c6!*#K9`=O~DsDk)$&;2D){b$^FbuxJ&7trCk)
zfW;}WkH*b1V#Mr?$k0HPS&`jioQ>wpAxqSlx<1acD^J<S5B;ny>!Cn)ac|i*z+!B_
z<7xQI%+i-=(!<d-a)H5>-UL6>3-&|V6>-1R75GtqMtyNy)ML>_b0dDAHCey(@O^>y
zc%K|%JY`&EDCv2{U1{e!I^+m@phqrypUVCzH#pIb@S&cTTCRq&*A$GA+H7kR*#j%K
zeLzT<Fs&?hKGhoV3YpuTJM4T>55=czfz&{RkHtF#n@mMspPFB@A;Ng0j16sw0ILGo
z8X_8hsJ&HMCu2YzD6P9c;)2o0=M5Jq25q=A0hH(7ppf<|eG?BWdz`Z3SR1JNaakp;
zhN2Sr7be9NWqvL!WdYrPq+^aoK3}jSVB}GQ9C(&g1g8*8yPEZJv}lyWDJyk+=5do(
z?A-)+ua|`eM_HcfBhfG#8PZ^4M4Ry4<P1iCn-Lj@4rj%LB%rmBKh7Ua@)PGK2RkcI
zu?~4A0yXoveNa<{7M<1lY0f&!_OD(nc0tC_^3XugJ2660WtMymGrWKO`7S#L=Q}Y`
zM}A~e7$gId?94DKB+Of&Y!Pdp!6N3SB*z^g`L9r1@wnMdV!P-X(y-#e`7OfA$erPT
zmajLkNa}$&-lEsBNNy{VKEg9XFIh>$Gh8Q(Bx|GQEZ@9)kIrq;Csx8OyEnLCSV0K$
zTv*Rl3<Twuaur{*krFq#INIc3lk<(M+$f{)IXB3<jK;^5N^#80$Q1uD8U6feIXXI{
zB!o3&#v4%$<~1-Ur)2XDDH0u@c=w8bcAb)ju|+O1MnZA6=P^d}P!5?Gv>MRSk_a$N
zXg(j%NsC+!wh+gOqMM|0epf1V5nADAKa$cO9&ne7`+pf?NRjs`^Eg2j4Dfs|)nFI3
z+e~hwG8@c#s<9)D;2_ad_xaF~6yiWYsOd$6Z6V5nwF<-$oT5#>V-ath`&g2Hma;x3
zG;Hr<=2C*ogdyHa>WQOdbtp;9I+g*w4YNEcaG|9HV4hhL?wBTYw=QZml<umEFrgy8
zX;p*`%KHtfBFu7Sc2`wJ4ujZ9nPiX--=-?U5{k9<*98Q5o?FM9>}Bkmhj)NAM6~KJ
zb3HE^XqZDh#Tr%F>U|}h)4bk)5}{lh=&F*=VGuj*Nx&NVyGlB&QL(v1-chWakG!T9
zEG`lD?UE$9Onci>naVAgE6V^y`8oT_T%gGbU-Ij#`k0I&tK_~U%84Y*gl%Ul5adW1
z%V#POJJJG@q{B1(1EOF<$wm}x>F)LI77>t8?k`NYjDW;ev9-64fF##{9(4X1Cv_Oa
zPRe2)0pV>zV7y{l?S|G~BS-W^(eiV#G<$$PWvo%FM5|g@y?BUh?T4Mj*h6dutn;kd
zU^xawbd@d~gs@Zl1uCqE!5hXa3|8r=Hq&c8vR!FG<k{iPvWzUWATF~t9wMiBxNs>+
zd&7{D8Wjs&vy={-5gC4eZAMi5{pvw8!Zpo^yrOMJlzj`8U$lBhGoqoLhogI&5d%&R
zni0t6d@}+)t23>L(Y_lrB1Z7eCIM)S+>rhMu?>M16nh@FA$F+wd<%k-<u?$wTM)^~
z7xR<R!^LWGl1|$eFW80dLuyqti_g}{QJ~l`LA}uUDD8@7^{K{xM^M2{Sj}+0+d!PE
z!0kz{=wHZX)-g=f;rTqo+-g}T@v>O1QkriiXv<2AVeQ0SS~oWkv_oYOZz<@~D7vAb
zOS7njxX^(Ml5K}u$%W)DMw)gSAZe?NaW{}l?COY1-FfENoiT30p)0zML{`$dsjrP~
zM|V}*k)jtjlanofv>i!Sr8ssQX~rA2<DFI}E0x6=E8FZl&NrO;m@IC(_QPnks|6BE
zJiF13^`;g;?ELU($9m@sP%9Gy)?mkaV-L`ZJxWkQFIj{cHuk1V@&QKu(gn)pY9pvd
zbo~3>dx5ag00nj`z#%3X$z=Q?CW*7Rjp#O*BuYbOok{(Fk^z%&wW$o^_%)B2G-xU>
z5btD@K194HWkaNll&`JP^>#M|oogNQl+6nQV~?E2l=13WP{)<_fUslsro$0oBaqVW
z2)|IS0M0i2%91>sv4dPf8T}iT?NH*G(`HZEZaYiQowQ3M>AwEG&$!*alA=G|qUr|h
zmDlc;RX4DITh)wnMYq(rWW1)(au~!;%EZHo*}qNE4UAXpU{BwPhNX75)q6@BH&;;m
z=l7n@Gki;ss37oFDs~jaPNS0RRS4#9D(1zqOn}hq++!7gUnW{D!o<Hu_O6AZA2xP*
zWqE=-x33l&UK6+2&peF&B&C+f!f4bKYk4Zl=+2iSiX?z*bZ-=QLjH$|$`HEPT&?CU
z$B32q(7xqhe*tC!EWDRtY#9`nQn&#Vm-HkeDVNY?EDe`uu^Ev8+_(9$8D{~1>n)8`
zrjtdZ)7m?w_ae1Brc0#ut|}Xoj4VI5cj=znjZa})I98v|)y5<?i?daiay?UvO7CS9
znX6?{r>aW+eQKKK9*;?^$!J88GF4YA3AK6RuM<i+&D?b(>>Y&SoFT~0Li?Xk{{^A8
zPkfJ1f;nN|V|?E%Vx7O?#hUAXVlVj{tz)yo;ZvG3t;=SLFHc||qkTuVou8tM5|A@m
z2dxA?c3P2AGi@Uk{6rI;yctOyc%i%V9myV~l-^6AyD257SQ6TZxltNxc%xJvYbi7N
zHq|)AGB1m)xyVbW!yM7=wfVK|cM=*Fuys6<-a3^X&qnH+9g3tXNt7dh)=E?0!Z_g4
z`Rqg`%RKly$5<W)YZ5p#XnqO<Jo^Imlv_JZ$aR<`raCr#g{KLk2X$({9#($Sfn_XK
z`S|GQE^Ui7I9iu#qn9uJssnj{FJJnA1nT!}G&hS_*)!vq(q<q3QCbYV%DX~wlEm2p
z_pB+-QmN-dNxn{ImA?mn?=#S7o$}Z|!VYrUGR^Pj(U<<-S#IgCGE7Ui(@fJ=?u5B8
z(<aTJNqdD}IdhwP+(ANWrtc1f3r@F`jyW?$wh~$MHW>(C!X+{-g8&M|H<FCd*%}}-
zY#mx#N7VsKD6&LgC$xOuVG1yrycJaH3?}kS%BKXJmg#jCl?m~G=p4H~#fYPiI^)_+
zWHSmKbWt*6a3^%pe`FS9%~aHBFCC*<l<4RZ^d90VJX)Dgsbq2YuFR$$n41UP0TTq7
zUt%XP!7$(@K;`)<(m@9j4#LGl5j6hYN*p%#w}S(ZYWMD%>CSR<e}#i!An&4?gi{Qx
zIM+<5Yj+}UF0$HxN~U#j-3~6t5XWSrT-@%tj;?Ppemnfwj!Oq`>y@?MXPpQu8Dg>d
z{H7aFIZgfnS#F)}TWFZ?-<X$|jLR;ZE2_Bke7Vf~JEr{|{9eBj=FM<)VAjzT*av1K
z80vBofeE~I@Jb8%=fS%fh7P<c+7A1`tC5LHcjD5iS7Nn)o9z`YPePgydPAt%xk<-U
zI1H-5UXI+2@e`qrSa1Gqhob|t3`WF0C@XTNr-;S{60gD@FJRvqg1bP!6`~IEB3c~x
zR^#`UTl$NdfyABEI65X5w73|P9`+1*zU)|4j>SsvbNg^QFhMj(_E}M2nKUpvv2OyY
z(@A~N$#!Oc!{}+9k)`rLD?>L9X<9xmJ=-Z;XSRIS0=*fK4$wSWBKrW%4O39gWl>e3
z&t&e*gIWRXiiYPt?Xu?zb|s3NN83Y@7NAh?30>YSVqLV>i{;ntwmY7Ypf)~AT}LYB
zX9^3-C!J>a1?b9oMOJWrnH@{1nHF7Ogr2(?^KTS?hUIYR%j}Hgl+RM)iAg$&34A^q
z9@&sgan}qF^9ROg8-_t)ChgU;9l-^ot+Nrh^+W|dg<@1|7UkE$Vt?<63Q9%xM=_!o
zrZ9F|QnT{!$G4#=`R#S;w?GDd^mEAoO%h|St!X&W)(*Sy@g9dHK(@ESGx+Fpno2jF
zksv95WVb^iB+}hje9t~}vxs#)1Ao$lVjsH(X%nO8LKGGCXFeyBHmm3<LLBvpaC37w
z8$s&1w|qk!Vs)@?`{+BUT1|Y~DQk*M69%ei=aNCUVDQh9H*%VSP1mtX$dL0y3s*HA
ztLz6=Wka}1M2h{Ce2|=k5uT&SQ<InBlK47*RisW2$@v|+Nyd;<qJ>CR^PpsvSt6_J
z*yMNAxg555Lvi3Md9!(bJG?Q9Ilfa?85qaFi8J1)h9@n5RF(G9LYQYN%}7qPQB}fC
z&7JzLtDGRw<Ox+GPgP#eSbcSD5_B>RT}4mR<O!Lnm$IUpsmy#+QPfrg*Nv$BJ1v-h
zmEof8^V{PM?fhWgqhD?ou}-6Uu?EafvSWb;qh(PzK_yIyH)A<le+W*?OMWxevpM&q
zn84A^S(92bOXD%=%F0SrE$qyMeMvQ*pM-jxnbzlK_=;%JC4+stnd%Knqjs`V0q7;_
zI`8ND2vcUx!uZbffnrupFC(YD6A1qW0s<_3FqdI$85Eb|L@N`Qun#dQm#VoL3V-MY
zhD#Cvf2z8=%2rQL)!vfKZEH_Sb}yMj_LA2RNRbpwsV!NyxAq`QmPil;K>U27^zi$j
zhi@<E?=L@n01jRSGTsi~e;W)x(EMvO-*hy1W0V6keE%{0@~X9Ue4P0}tpWJ(d>!~q
zZ(qO!ZedP*n71z;O=%C755w1E`hQ>B<=!+swWmx0;%lL-`A99j(U53zJ2NKQZ@+&3
z`9&A@4WErTfBpXQ-<N&@j+_FG!#{t#qz4E_L9l}b#Zw#p`1kPJ%ikJ>N?EIruK4%c
z8cFX|us)Ie<ySrYiy!{XjUk5Nf4S+8HjEzrGyH=G*diR1&V+%59j_7AgMU>q#6or^
zK@gS3csq%moH?spxK7n*6{#niO)MBTG3^}Dg^96gSpud_Dl+k6Vr(heKr$jeA|JjO
z0Gpi&#lUdKeR@k4nC+&LiG}p3bQ#NkSICUzaOF7(a_RqhB6%v!Dv4^ISjb>`76s{Z
zxrTBb%h^vPLm5eOb}s*;cz@=^*_WZ@5P8^%B$U5yS~Nnx7Yn6X&QTR_SPn(yqPfea
zZzZvF88cao_Q*2s{lyMP>a6F{nw29fj@ZJ{Cx__M9ablW=?Q*RH%Y8qvgywBl(-d+
z)4kN+ycg|8m#507Z{5FDw@8s?8$<6FjZ+a^b&DO<JxiK}O*lAB3V%Gd$zDv2*Ed@#
z^2GGG#{4bboyaQBJUr056MInW!{fa>tByr}&qBDB!cM+;9T+graqh%k1x;+u9gAb%
z<H*NyE!)A;ZB|$4M3-}E8ikXuUKxbfM)5pMBz@^lU32Wn*C60UyC|+j2!bx$)CqwI
zGeJxqgu_3JBhR8te1D%)Kd##?n{mpZy~~(yC9!k)5@!<#f09=x_B0Wjd8c$3&&5Nh
z1U4tZ#Cjuov-vvta+=&ACV#F|EB#vfVeMyIjHpyv#3)Xb#yJ|`Jl55CQ7PvZF`H0X
zz>67L7ZkGWyBX5??TgcihSl{7zLwQz%fWW5Kho1|s`5!$#(${G*z3{-41BX6I#m@I
zu~Zd#a=gA$=uE=R0t=If@}!H15+i`FOuF={V%QXx*`t^hPBH~*N}0qh6$!mryg@z}
z<1v$f7UsrYHO3ZK*^VY^e&S-9^HsGP@QR&6Pbn!TQOL>ncvU%u*DX4~pl~n|!2pUP
zJ}D7l3s%9m@_&`0SU*W%)r}%|Yd?SdK6H2f{%z(PtTNX1l-W2hOViq#t50UJYn>9H
zr0SLc-7r2uc4mw5o6#(IaO`NE?R&u2+sEeo1I=>o?qb~AC<YZREhuJxahz&2T*@qE
zbXu2BoFYcAqmP^AV>K!y^FaAnW0TE4Ry@{{@^-N}CV%I=Dk2N0EG$f@xbZvkcQjlm
zz+&|yDQDxc{2j?F@A2{ZI}$1i?w*|IW)eF&LtuIN=Vb7R?1~LFE1@Y$)Oa%`F*5en
zzz?ysa#MtwVj>@$E?6pw3#SS;7Kd!-XqJk@7Ev~G+~c#~kScq6=i;NKN{X~+;_P(D
zU4UM$xPS4ngOE@lQ4&&;*iV#r(2P=@yZQ%9Qxa&Etdmd_H79{umD-3z4J0u$Cy*+M
zP}QMM;(05{-dc7D0`RJpnG-cpUo%Bfrd1)hxi&h9Tn1V`vu#I*^}S(7)+leDWJVS_
z3G)|bgu;x%N<Hj^(yF9UNYqzt%w`0_3Qqzv^?xhHsuxA=NhmbVRr^_#$|q57lhC!7
z7^xuZg$c&lj7BkmQLLMu)LFFg!u%x0P$rHC&FHAQH>}764%&C&s5g_?8E}~_**^<M
zh2VI!hNGrHGL>!(N2Oc@Rs6IRSUC_?7}Aw8FICg0+K+1O9$H^Nmxoi4C#rB)?zvK8
zPJca4vVU$<OU3DVss8g=Sh;BJv9;5!ge=^**<qpEN?=(6wz;epwr^i*F?tVj44ge$
zGK-XrGeZM+7i{1-h1PS0Tg99vAo7EJ)Et7uBtt2#iU3hm#x0@uz>JyllQZRuI~NH(
z-(mOKaiys<4m)KswX(0AX3mm<(0W!)$$yD2=6s~^U<&M{dP)~AEW7)4J_2hr(1MTg
z5`(B3;sO+r{%RV!<Z6$GGIw?44GneF@6gaCrCT-B@$rd<&?*cqx6BJ31iGcm`MqcZ
zuW00x5qQy)^-q9RI?5ZcO4s@Uu*zcmO<1MtaR;o@k9|L^(o=j4tkP4x4y*J!I)4tU
zJWf6ZR(Y&=Dy-5|-G)`V9{0m4mbhk*k3>~$j0&&6NIHh9*yxnK{=Q}_uf3VXPR_E^
z#pj?ZHgM9MGTgr^s&ahx8&dVCIE8#pIc&XGj-#wBhkfzl*cF&gym%c2R^dAwqZXd0
z`SoxVNC}B^UtCJ8VhzK<`6c#A?0-%QqSTGuRw4=9IMKe=x-=K<CxN2K$&Z4=C@S#S
z&8+r|jmx!6Wz&uh>wCkF90zRYo@7TH+a2MB9i4<7c_A*}i4HnYBX;;#m;nnEP>B;T
zix==22)pv4t;ER2d7A+UtELY8Aprt{u)9$hK<$e&Iv2ZUC{Q;L-6Ue?R)3DLMCWd?
zqr>{%vLhzq!(H*`RuVfyE|WFKHX^O!Dij8eIng^<!&R*1T*S111QJ&e`_0<!Gz|4s
zb_qHaKWSlW-kWgsb#fhFlM1F4Ew&&dmKy5e%P}rr)LWYIn%3t$2g3>1$#{x`vY_kh
z_quVQp+c^w?OzR?;WCs;{C{W~k7|o?4%vk-Q_5dm_I{L22&JDddpV#YO0{231kwz?
zz5%_qvg)%}`#?r{9Tt+jniO2$hY=#@r3>98yvw@0S^IsOS>CV>jqh3tI;jt;h!xDT
zqFSc6IbR@GamL-I7QSL^&EDFWp=`i7&lp*mU2kb70-xtS0b7=C_J3P?*`#8VjX|X|
z<AKbh6FgF}!YScv2w&Yhl>6ij_r4q-Gn|q_5xzIrMB<O<nVTdZfe2oa{)~7Q4D_m{
zxs<%dUXs^+#!8Ah&f9g|^&LU!C1kSgUU#3YEP>yOm30*HG>#NH^Q0{*#`Cg;AouiK
zqA&-rlnEW_DBDkP41Yw$k*RNdnv0KyeHW{O6`+VJ?BOEwqZ3o?I5L|TcbLteKB8U|
zNQLq`PBtQuoYH5au61@V>Id))X_wrB=XERJ$7=x!JBq1vOGkpZ)_uawW&+wO(2J2i
zf!shlMYPn;2DmttrgzI3Fe;<%6Kn$lC<FcpwgJI&vSuD=8-JK!D-%Aojk~wgyHu+!
zcH?;V_Szl8jST0=4xfM<B`1e;7jASjiJhET4j1uBxRJ3z`OwmIi)}k@mx>j%=(A|D
zQEeTc!-i}ODgt_vYz*6+ZRn4!NEyyaJ=VqdF{g6rW$`4fsU%q6q&-RbGWpz(rX8R6
zhCDTJCibMAUw?*OesB#1fq0n)*EgI^N-oQzZu^Fvk~uts>yn99IrF417So#z-QL2j
ztg&iO-IGZ=wtnRrrVj@4_p!R%^X)s?UD7)|%j?1mMmdVl5nezQ(a8^r+n$EAtgH9r
z7ERD@-cvzKVK=u}1U^nLfV1FX7J61fI={D%ns+JTuspP`E5U2^q=oH)gW8-#t(M$P
zv}P_=JF&Jjbf1_G&S&eB>HMYNU0+P!Mq%$a;)={e_C)pP{{y4*#^IM?Y#9^*H<!Vg
z0~42vKQSqnti%}<f4Y074fqykZnjTJ>`Q!zFUjZUQKB|Vt+fX}s27TSwUB&OEJ+sM
z3t4=4IKMm`uG$n%cqXTcmv05(hQudHG+Hhkk+7PImmkIF_fjg0>x|1tsWmQ#-yYw5
zeAML8gHE{U_~;C!{#xN8{JzBhb-!{nS9jH^M9|@VLYBB<4^Kx@M@;l1jHdJP>&wRn
zmrlhQAOY5wZ^aoIe=gPUQuehP!HoyF{8#m2_{W@@JKRCOBz-AiT^n6E-SO$JtMKTl
z@{$NPIx2pB$o@UQJp40spJ7497JvMBh&PNTp}i`|ilA-r^RMFD!*4linh7baLmTOW
zgp|e4*rC_*DJ?cnSZvZ-=H6l>i4|xH%Os3a%>?Q>caARye-mKG08L`qVqobq<I?n9
z%gL|YWiY(?C{5_KeCkN>T96C51TsB+Rv`v#Gw%n8DqfpWl{CUI4Ny<p5lO|F&jH~^
zB}XSiDHcu*CWA2<`gyhEl{P|Ybw|f5?S=7ffcUf>(I^4(8zAzye+;(Q(4$7K_YJdP
zX%@`QDpczQf36v8Rv|hooOAbH>OP1egOD9}deFj!Awyioh_e~{9V6TCQ3_?p!pZ@K
zf>^BzHHFp)t(m?Q=ka@lmB|b6RfPQfRTh8Z=0|ka!59CQU4JxUWbt?Ld*tXDU;viD
z2BaINfMtQ*?n8oYAy^0K6xKs5)J$qI1B!qVI7gj~e=sBzYd||e;Mryw<C1MGC<Ao*
zA44j;<D-=^*z1sb*(xkoDFm+PDt)O(l;3r%vT58zI-t90$#eSO3;o9^zK#A#WDR$!
z@~TQ%DIu}BnmnNOb#Ku&v$)aS)04;n6Av1^^E5_|a{#wOVMq0%NsI#>=A5y5yM8lv
zd%}X(f3IY>XQ2R}Kw!Vzz-(S&_Ga~F%yu4e%<N&4)ey@FG{U;TTL^T>)VEXW5y41S
z+@^GTL8{CKc{OnAeBqiX>C~YhHGW1Y6X9+77df<$&827gSKHg*6hH%A<vl_zk~W+!
zeC}R_$WF8^)MAGTYw7->mSZ07^sddZ7GjxZ2AL3(d4HW?YrxW8cCz$3%WIDyb9A*`
zzd5_5U}g8r$ZCjXi!j3KY1r4HD}$4W3iL{RWmpNeP2yFiP%&@OiLX?Z0!40G&t+wb
zwB<M{rJ&$)<!;bfeVWdCS9Ep{O2Zt-s>vN0%}{`W`u>cDARyrRUHfM(#6ry^1Kz$8
zqd|z)rhlG#Jz1xl%hVS#;AVT~{G<dF*)@Y}Ar>xLNuUgW4+dON;MNxDp;zNY<C=oX
z(^^}2HQh39c_K^`&-LjrEez9{@I9qj4UhZq3oEl4!GWynR%VykJ)Rgc5-n_}40WPH
zn@-_TW=tch7FCW}&-zsjGZVh2qNPX}8wQ9wD}P$4)|sQ;2u`VvnY;9|-aV-znIk=o
z%S~0stoPRXt~ElEUTSlyO=i8f)J{`PGPn78Ty8vtd%l;=sTeuE9^2l&<lg-@YiR}Z
zz~-IxVvUSLD_S^%8p#bizPW1080~PyL<O{Av2>|Qc_|zYgq4RXtT*a#ptlk0aJ6b|
z6n}^eXFlB@^<Ns|e1W|HQ0@=0Lc?(?L<j0>NG)YtH%%Jo)R*#ZS~0i=>uHr@3i)Hm
z6jQYn9+jdQF~szxd_fePHV4yXxIc($Pp=Mc%EqT@{uf3!WdrQtJ4d$+bzyYlZgks9
zDddDA&NaA2qbYo4y|>V)!eYTp=9WM2et$fdTXN&MPI~l~_1@}5j!FMOG*8CCJ2)}L
zz+CRm=Xfc?<*1h(Pu!{~m?Q4YdQyDLQOCEM#tB5r;Z~cDALE4E#1Cu-(BcxpRUSNu
z6ip%Mc|%JgL5%hkw`Nz$ySI-@sW*BySk*CMxr$9p=&HCdCUm7NjtSjC;c%!2O@AN4
zC^%tMyK78c;b{(Hss3Z0`j6qv<@iYTTQ|)+IOW5%x=Lh+=kq;BF+$SGTK~dm70c?L
zdN0Y`^iJcM@j#`^Q*|G3K2F8gv&tzt_hOYzPM7T?hS|E<t(D$*8~kt0qLog_v6usf
zI6=H<GlHvxGP*BiIxQoOA#$B!*kw;XWQwU}`vM9&YKs%+x*}S?fp(lkQML@88Hzm3
zjH)clMs<oJaZ#MR@5v39OLM3HKHWJjoX~6KYT=YsvTKFEb$135UcFWgJBFsv9`|(n
zRKKqA+}t};dm#UYVgCbLMp;XjVQd)`m(fE!6qgXB6e*Xp%NZ_zkR?+T1bzg851?fC
zvy<KH;rw>^?M3O%3P)hEd;8G=-;n$Zk{b<m7C;yU-P?Qj^+`%Je4P2vQYyuV<!i@h
zI(|~%xCfm0fa52tA@^6v2k`YJ{O|k4rzzdVQ<R|mQ<g0G2t6%G6=ZThxP;^Q?d{7!
zX7z#3FgicK9sWIk%m+Z$3-Z`~e?Np9T7ggw`L+}qRrl-f?#JQxBBB>9g$q~w`??_0
zSR)+TAoJ@t+5N>gzi?;eUH4z?`d;26yMMYr`2}l&osdzmVzm7&!Lmc)ozG$i5v073
z+TSmtQZs|XxZ6@e3Qu#f(Z(FGjY(sOCN{>lg``+E3D3fRvyFw>*K7rJK7KMXSP`+s
zsb`*lZq+ihBDc!t{;J%kct6_6XfzYMC^)qaXay$%VG(u$ibzQ53sAhch!>v@!~V=i
zxDIW$0LROduo94F%~uf7QdomqBSM0tW#L*#>H!IZ2i;o*d@fhPe_%dzqgAC=XG_oU
z+(>F2$jx9LtbmMOIH`65St%hsOi-q4l&|rG37%l)vrvWdW%QSQ%^8t@JjI^Z&L5(f
zs@QlHSQIPp!e<{8>64g9Q#qB;wzOPQOMFI6Ia%rvIGH|=rA*|z+~bnkf-DZP$t=Bz
zQ^cp&*}vt~M{*b>HBWtz?|SMppPAH}b;d?Qx+9h&U19A}Sv68#=5r>dy!fO%r}j8l
zo5A5h%aqq$y#WkI#9=;vZlk0z_i_0peHy_RbeX^W2v1k<UxuP8%T&O&W6V{SX?O(I
zwq=<*tU03XvrHXQVq+h1Gk1%q)XX?x<W6a(wh#@!cM65Ls;l)KtfnWh1Hp~Ok5KTa
zq@nGjSwV@cFxqb>x9GdE+@h~G!uXvgm7TV??t@9S*#+3_7AoU^d$YnTI4BZUZC)Gj
zF{9K$cNl5|l<l~$5xk7+0y<lK1O?I^&17<#HCN`e?ZaAu3$?HocY+JUOA1=HU-hHF
z1%vSnai$axmd)!bDUQX@zEFBe#jQ}{WKvTHZZilg;DQnk)K1`n6(C_!(Ccl{B+T?F
zF`#UZK4Tv+F4TN~%#?D=qNg4)q%|~~N4<ik#K>6oK~pk`(4{aSbJ6^YZ0Ss9s1l22
zoo{2Id<wH$N2jgj7JVH*qs%_xNjWVL^>uhcGtM5;Eb5Uz8P>*a-doC+=d=<&_W9DJ
z;y@w0EB4Wj%L?8mMeJSHin2<eDI70&-iXR%jjZ`_Wnk)m#L=f+uDCa4@=g{7C!X4M
z@8mH%c?b@_66R(&R$!*Y2lj#4pio2j#0*{?yxM`8#z;ueMHusjVSrY-QQVB1h?S5w
z4}Aqv9XZ23h!SUV+7`pbrEQ&T$bRA;{NVcI9Pk~sQ#Yk@ZdBy*l_<83b_GP0xWVS(
zVPm;PUrf<|cc1WpUTC$DyqvOs4<_*ZF>4>O*2>hTeK>0{DFLJ&sVUqoqOOdhke|8D
zb2hA9qVvKH`KY=s_I*g*O1W`iMZ=^ZnaWy6Of{_1dK&wOL=MJ9$@0&odTOYQ$M=--
zWNf9Sv}deiDSY;UHII>cn%~nI(_Xq{dc+gGP*010l;qLUB^t{tVW=!#1D=%D{8;(<
z^|PC`{P8;Itxy1(xnzLZ2P=JV!^44j@5tfKTa+V#xw6)F4xc7r(6?11QE}?z-1gN-
zR8l-ti@Z}nrB)#>>z%&iK=yP@DvOnb3|0>1XzfOe3C0{=`6(UfV+1|tiqD4&H}%21
zWP_`Jz{?OtOiB-`^XPVTR#4}&bvu14f`CF6hBetIuB1fm>fIhTcwS?;)7Kj)L9pRX
zjP1_-&UXFQ#;t`@`jN$dw}?7#7sMJxTaboZv2WqUB*k7~56Cb#vEv=P+c`$wRJ#g~
zOFMaKLv54IG>~Nvd$Mn-gk?K!ieA7Elf>;gqgU8*R`8e!-z%b0GwFpQ+9!em`%*Jx
zE-m?GyMAlqW^!fBLo;%(h)N?2g<NAdaVoR(|EYH_KbK)_8599Hm!UW+6qjkz8Ovf8
zP~mg^`Q`f?nZ=jFoA5O}#=qXb-2D4yT#PYFNb4?t`E(OrkOH9`Ucd;i%;mSgT|VCY
zFwWYl$^olN8l!cmDv+!{;@Q1W;Z;(V_EHirhMxI5tBCj`S#=Z^!wZ*BX_u7K8F7D<
zeHrt|XBSum><nF({@F4ftR0+iL|<f;*G_TL4|!Tc`g0s{bogi{?JiyrqBL6|$KfeB
zNUND5goWqz4L@W+r4YnyG2&OmIHuW(+NSRmkctAPX`RA6j4I#bNRz2}1`W~XN4TrC
z6JV)`6mXUv3-5xVL4p7+@rjw|Q-Ob-Fv^Be59ybMF0LZbL4N8y!e;qx<=`6W!{~?X
zvHU{0U=eMc;ivHuFTVwYZoc{OT{SThJ2RMZg+arSi`|wf7=+%%b+tL5&T82o*~x}k
z0lr*bN5TWH<SjaX$eJl0LPdWzTPWJiRgIcV7uR^jkR?M(NFxHxU@OZ05|Mvq<uA(d
zy1iPjyVKU5w~^GsTIi`kJz`o9%e19y0xK-K9bLy@n|U@#!9K9-*>uXY)2P<7(=YAV
zhsJMnB92yiDx;~`0y3vUsTb}w9ukFa>Cs<}x^H}y=Xvni54!azbw!hH8?8;bD6oy*
z4iEuRDCu_f=KQYS7{=}!-KBq#Cec?P4s-Z`b<3~imo!`bD{C$4(r3?2lvHM?BfLyA
z75}f8c}2Y6#myqQS|}qAUA&e5mUYxagC~!LBfd-ei4~^0g4W4#GZ8*tQ`pW#Pt$*v
z=+#(@x{@L?ASV~C@84(FtS}QDW#t$Uz)RJ-Q)VgbV#?sagj6Je@>hQxTU_$72^Uw#
zXT?T`)mb~xCIp^>C1xYE=^{}0?BdUr+d~1vPKJk;Uk&LPlJ(c@fLSJ!<)U8K<g<KF
zVWi68+CqTV<t;T8dIv78Gnr1R#+eqW3b<#<9v3@mRmWkJt&E}it<36IIx)qDn-(YU
zVkuNCp66FCcr*FNVs>IkG}%7E<4pu4Wv^hOLx=cw-LfjOm@I6~Z_nB25*q^s_9=Hn
zb=yMO<-jOnbZE7NRdvJX%B>)?+%TdYLrZr`FcT##AzJA&Z~fTF5}xn6s%LjCaZA^r
zon)F&G*Yf-=(pt78PW#?@NX7Zmtkxf6ahJx0Y3#30X3JQXbmTq3f~zr1Yzvio|i!1
z86|(X=<LNSC7f25PyfCA^I1xA_<Y32NU1eG9^V80Z?8{U1>B)t@KM(%Z&bct6Fv&x
zU-|F#zC>-8U+E=1LC5F(WQh;o>6O&{Ogu&5qB7TiefsM=nV0|mP1#En(UCLd+R2en
zF<hK+0>7<mH(0j;H}mK8l=p8=Uz~_u2it#AhhRW0g(97(A3lHh?NQ5z4?jwxN2cEH
z4jv2e7iXi8VY&I&$Nc<oxVoSPLFmppA>$(1hkrbFmjsgN-Nk7m0_%xM{*y_3I3{tz
zM_ZG8uU0m6xUv`p;W4G$9zJ(hrX>rDKw^c`?aKTS7cGM@O4nSz>n;oyBQ)>7B!Pb_
z>~ERu*oI+Bf{eO2k|N@kYzG$BWMQ+$vMu1l-}#zn+w^<~tlHAGE2|KNAzM9rAS;Sm
zNEnn;AT_n9oIO(r9I<A?`_5Fy=C5Uhlytjo-ll~c>W|m1dWeL!w&iK1dVC_HO;abl
zjTSoimNz@OhhtmQ=WT;<YR|eWBn^LmHDR}4Df=enVH>&@(!=k4TlcVW6^`|H!UU$S
zcT3A51XS*jPhc5>4_P6NU#PBqMiH?}HM{xR-%Ye4ux%ae=0rPejbwJ}QT%zFau#Vt
zj0@p!1Ix#fajB$8hW|v~Km0O?7BC1lqneibR-`<lQoY1qy3u9Lkk)%4W!!&EU2*Kd
z2RTi?aYQjWqNMy2_=;X*3`d~XCHj~j)?fIFb=7^e{bjGDvNM~QIoazHwPcQKZ}^zZ
zGx4Lt6{MI)OHfELab1zgYf$-`>qo+9V0*rj!!Qzh;gtc26;@4!LM=7&S*W>;L!mN4
zCq|WyrbD4m4%qNqLZKRmLfC(Yk30LA424c3^wXaAAQrN;cPk#oLL)rYmOPAwkt6l7
z0*{#l5|MUG8^%Hi5n2xtDb37<LL=bBHJ^Scgf;|H?eT7o)nj0ePPv_W%2;?DHUND@
zKz44-g2P2}y5y#%^EXcWF*`q;-0JN<hPAOn6>^C9$W%ACUmYP)yk>vbf4i0dcMEe4
z*~se6g<ScRet-Z3l@Nv{$Sh=8{jDD$3_>{2t0hMAStI}c`oFGD2<SoaN=rW9_tT66
z;0UTF`Kiyv<5sM?A}Jm7SO(Cx)xM_s`r+|OIR{sJ((WPs;UB&Yz%vB1<E;jv6Qzlv
znrd}BZ_WU>v5j2*-YtJPG|&q0IY?kHGFJk+zZuQNLnF#R^}X9Dp+ckM-v@CxpfamP
ze_tHd>P~HI=dsF0Phb18@Q?*0TammUgl+cBV^N$i2|20UXFou1-s~)%Om}rDsM=Ul
zed{tZz+So)_RoDTN<uw!Y(jn9W+G#Q%^oNt9^-R>4a;}Na43Hio|k&jzFs-?KQCc4
zp)9w2R>Wjp_zqPdH`?K^<q8L7HyxDMmC%|jw&x`7p}Mm2GD}sP`Q7m1dL=(Ohx}7R
zBPS&CVkxmvtpL}imgM^dy{mzTGd5H;<%hn3T4V_wYZ9OywS?F8n_s?-99JSDHO$jv
ze(hQnJEMuJwW@!c^&IGb3V(MsV>HqV7E-OZkG%~?P!9^k8=8XN1{xu2TJa8`x3SQp
zZ{;zGyy1AlO5L&q^6q191GvZylZNYLthY$6`haOyI#xPCawnYu<hDS8(KfvLklP|B
z2OUl{cg%(1(}Cvtn42@WlFZ28>ij<Fwi=+e6<r^6TOEG|I2~{2L_a`<!R)p(0lK%C
zk0JyaeY3``(hjR_np{KkFR_S_TTDLe5uWv-b*jda%eio;H4;iJs%Xc?$c<v@ZA&#U
zMv5|(Bo)=Df;Ykt2O!jeeoA|(=x6(Fb~M*lgui&?_dg~|O+dh^BTPDnr>>^@(l3j3
z5SEvDB$j`Z<YkQ5FKtw@Ta;9R?$MY=JjR`OC`>ykV)O?J>$f1lN5__GLyKqd7qy#h
z40qC}`Gktj3hAB{6#^Z%XUiuH6s!>F6dN&cr_xhmRD6yE_*`-6m!--om{CnJv<YRS
zfT7n^{gk4dSy0;1Zcey3E@V}^wn0jP>Vyf*$O?ZlQz=Dp_>;|c`Y8n^mxzoSj<$1R
zL?H;G)`s}}(gwz8+TU6(9RE{CH<+Rt&|a-b+UHME34ngh6)(AuyWP)uhOvLb_-{9?
zZ>uM-S!nAOn|yM$W~F|2R$EA%8teNuh!hcmawj%^T>LviRIxRQ8USV~q}{o1+G;Ap
z`rCgBIhowCEdULO^sa41D^#p0sNdZKgtT7mg3@~A=yG>&GmD~3>jHg19~bZ1FT-}N
z@7gcy3Q_o2izl^#4MNLZs2bG<jynoF-A`t7!*1?&JO|m_z|%t;hWo*uai%pVj&N2D
zNMV|t%~S&twUVo;J`bt^U|?_VNkp(!*wBAYY<)Mi1fbU}ZQf1S9L-0E8|vtrt>V+W
z7(oI=;Ny(hDr6Oa@|vW7+^0|(SyV>V!iILs*<bpKXx?m<L8^(mWjR3fihCQ*w^S2Y
zASzc=eI8%|oen>a<$k;)=TP;j-Hryh4T0ZF+e&dcEA>TpD_3PNInVj^N`1U>U*%I4
zo+r;uU)SEuwL8m$y+lhXnfZ{L-k}~4T5oYm1ejF;eRj7Xm<GxR{?Es^08EBxx=+z4
z+oi2rq30!LK=e){6iUyYWd!cEQcd+`Fu(+Cic6O$=NagK7p<#FXioj*!p3bZYk+NH
zO*PlX($ZmVu0D<|o6QOGyk4JDQ^0M%xD_51z(p;vX@OKHkg1&CNF!+5y)i%076+TO
zt|nB|CMVB+%lE9UI9D8I-uRTGY%j~-w1PQz#M|d%*OfByC|e=2lFZAkX=44>EGhP2
zLs561wD3KDP%+>)A&<DI?;R!gus1Rnf?25bbo%p$Z;ww#J)BLHYpGIdVAjgGsgZI{
zLAaG6pt62yq&)Iq6}X@(i8@Sqk!0ZRYm&i8Irr^VtmQY#bBZU|S!gj$@y~Rsd1e{z
zR&n`m*xaYkENo(=+^Pg*Nn2q<_=6+mXR|$0KzZeVqbCk203?htHPx44E>OlVb6Iij
zkOCF9W1FnkHUc?mB&a4D=pN~SbU0bssUoedH8R3DoDPDnNz`!eD7Jf<!#I{>fIZ&Q
ziK78Wgrw7gWRQBf52y6bct&S6sIgkv(QsB&w}Kd8oWLi%jRZzme02tbdz$H20aj<I
zsYYghswk!UifR<2EMnA!0C?0S5?|irsCW#ll#qI~tw}~mJP3=Lx#2Q}#Ced!AMWVH
zNR1Rax{6|e!vU!n>9jf>*eIWcwUQgyXe#;|F^q6{6uQ~}7=|w<3N-GgfGch!2|I(8
zl5}$^Hz0VZnq&mRO=_<d%Njv&Bb>Wu9|+Qa_w529TA>K9NCxm*g9i|Mv8XZlHZi^G
zR?w&!1p=7~wZ0-ZNS!?F<26Wp%7-)9MDhDn4VV7$FT<ITxD+JgFs6A%o@%PW6h8-X
zDCBTEXIw&>6{)y<?a%PT?}4K(irXu}HHW{tOTuq4%DJ8#;shCLg|-9Ni^`G3h{DW&
zn@H?}JPlQmOuvf6W2r3QhInB79L8;|speirQic!M{lPKK69rXNYc^Ly%hk0>zEy?t
zV8FFJY(6;<h<tKvs#R<=;3v&wZiMN2;ajzH(x!t)k*g{CA&y81V|7h3=mnBT4r!S@
z;qO5`PcU7I)M{el7=P78d$e}etOQhlHu@e}Gdv<P_sE(%h|a`6AO|*LsdolB|IS$!
zhwRVrXpiur8!>zt;!&sBfVA*-?w#cjgr~+F20@(lqkT;>4B|#|Y0|huKZx_mZ1Bgs
zIUAa{LWSMZ1~WW{xNJ<vn8DG^2#+DR3reF`u~Ib`k9Tu2a-l<-*={ZEwX~vtt=09G
zXZW+LZNJA%VYB^(?PaL--uY?mT~O9Kg@iQ}gkomDZ4eagx$Qi14H+@NFYvl+3vR79
zyS{zukIph-eXq-OlObJ9H?1qWH4HPqZcw#c394#$w?5VBKy1pTCCQIc+rcTNO@q%B
z)gTB5DFP5@Bl-C=X10u5Gjq{@!g5+{HcG$BJGUNFI1kJF6^D(|i4?GMzoqScyS5)U
zlYST3*l+=mh@J<Y51<0W<gob*#apY?wo`0os;z65wr<(d?X+{3yW7_u%_e<w+NNG8
z-Qq85UD?*QzRmz8AYL6+)otHg@u17Tz0_TGy<fB9LwWV)#5ICCxo6);J%26mSJBz7
zUJE*K?ZvHmG*Q1??J^dz?Z{<Er~GjxNA2=I)KVMZw)?^v&QA^UFO4dqdfp6j{byU&
z|Nrpm-Twjn+ozS6VQd)`mts&h69O|bm$92dCx68o*^V5y@qNFdA0-4!v3NJISY-AR
z;5RvXa||R<A_cG%#7YAHeyZ3!tB>h9))6q;88%t0`yyKoUj{k6zj^q0^X^5P!HZzQ
ztKsA40q}vu|3TuQvx5f^PQ&o=+wjw!l*-}f0Y64ct?}dWd%(Z;{#HZ48F0W4+~0Zw
zX@9;}_yPR>NWb&ExY|xn)geOA<6VL*@k28`NKG&i3g95v`=3Ala3d4?QjCti(jWfo
z`Qy#6H=Gz}K!Ahc)3+OXU^ECF)Sv^>Z-(!GA3ooF;84q65|;=||KYdqUj{bvV`Aqh
zk)6^xYddy=jC(jY3QdnDc3NbeMdA#y#(yG<%%LOj-AjN)Xxbs-z@e$931VcQo*?Z&
z`zX#f2#w50yF^@Z2aDTS6rRN&*6+jHNK_`vA(9#xlQ`fLRCzu8jTc7yYvSKS)GjHy
zOE@nn5Go{BN8DWicv8n-2b}nRh_exZ9)lfz8{)74WzC4-9L{wFR^vZn#m&2aeSi7<
z$4`F+fy@3)Q&FddvM`3>+w^lBh4J30;p_ApB#59aVg_PFgbj{p;nj#%k(btyPs5E-
zUVyft9sWwg2sjv}+V#<#e4Lthsbd8b0$N7%o?qL+j9T3!QdTgbUj!wsrJuMyucDuk
z)aG}dimU!H+<YExJ|x~+r#wvwa(@CmI{df9TM!1khL*p1C};s5l+Jt^J?hlTXiIMy
zRKf*kG*TFZmw?~OZk79Kq@g<xA*>7`59Pmne1#8;Rvam+YK0wP7*Qo%AVr15-HF5X
z`=@Vd-4|T-zWF=BxkXM0W8HA;&?*gTH^Pek(uT$Glcwj{*~4W&Ig;&0Yk$i-E!i0Z
zf?IwV8k8p5ZYyyg_A^9`sFuh1k+GWd0o{*~dMtcc3BzeRs(=U(ipN1OX&#a<iy_rg
zH1R2-`0Ckc2KiR=QfB3t6Y^Sj*NpfUV&t>ER}{)SudEOuGRtT?+Um(O&qjZ&lq;=-
zMdCOmGWq^Q5v2psGy5v-_<t-Udd5Ug#JBU>c_)qD$H|fJL`ezy64D~__>Y-#lb~nF
z#2@oA?J8gLEEZccf5JU@;>`n(CtNO3S`)r0y3Ay46lN#hFL5X)NsnpuvZi1>BA)Zr
zBv+dvp0N{#RTm<?uKNXrFJ|IH)*z-8k<w&=Ojo_b@TLm}mHJ!d?SI8GCrRSpr-_aU
zBCReno+p{a--o;cWiMmedAPU!E@NZvs(wcua9Ph2$3o6Z%It)@dG4~dj0dyJG%5?f
z(oPu~>5nV=a~~gOmglt4XE4)t1?Xb}8ktn)BmyR-E~81|(=U^r`%MzUvT&WjxLVXJ
zt3t&3C1G^v;p#~oFMrJ_Vg^Y$;b|#p+Q{WIBA5eXVNh~;iS9?@V~6>Xd)-`KIbjT1
zno&ZH`WktqbwVk}AztVkZ=Hn3#+_7iZ?eLm^;p`3t;JZH<Kh%b{~7<^l5G{N9BwiH
z0qc2rwnwBDk(68CHpz9A<qGFhjbO$gl-^g77}ApbUsRrMq<@KdxlUuy(Hkwrtz2EO
z**P-D55*+JOFo>Ai*s1}su?aemF$Mf>}9-SDm(I(5;YK>H?-(jYr!~8>1p9PdghS4
zqtQ7hiTZ7dHQ^KAIJESuvApPJLhrpTb98By$@@enmOZ^|=Bvr!PV=X%s;oXV(IMMk
zw$;$@;qMyTe1E>zwxiPcW~J#fG0H!T^LMQ_7VaFfqwXV1^R(jOYOshL@*P#7y5%~W
zV>W|eQ^tcksvI?68VV}ky6}-@ZbA6kP9iy$tn*Oye9A%s!z2ZgG0ig#_6_~hEIU@c
zDLJrh^uy(h(6sq<$on4Ci8QO^-yummeao|XZs7}lzJJjfNK$td-|Xk-d&Y1(ws3mP
zr`g>|lkpqh;hsE^z$u#QUv0@E(GQJMP5xx@N#j?LshD?i!P;4YlFc?}2~%be?wWw<
zW&tyAVx<u%Qoc)eq^I@)wYHeV%=&<UyEx2kevd{VRsX?^T^_ZrZu^>3qX099yy>X$
zqmH<shkvv-O6b016x`pslOvMCR8oTG2rJ2&yeB~PSYu-T!63%iM(tz0s9Y*ofgiq#
zeD9PqW%8Fj&)M{AjQ4iDCjFXJK(5fgTq|2IsCCB{JQvbkwd1Ix{&X@&EXR$zBAJxy
zgPjR1iWpW&HK4?Zg)tZNBdc@JMPR;S0L$nyNq;|_pZj%y7}eEw4NdCjTAqXgx5hP7
z%*qS9m3N&F=rYH~uOH55L=Wi*C)l<}SaBwCdc@L#lGC)y<JrV9NZZ8mZBwOBp1K6&
zQxbdXbbAhUt;k4a0GmzBJV(iWz=F@tpgpY=5_D{0X7nG{1r%Hi5to}GgG7EpK40X8
zz<+8wr<p;{Ug+U`$R0<)amZ_2%`)g`|6jZZwSaRfWMys&HzRFakV<H&XH=F}MeS3y
z*u>oNG3O}D9Ld__t&sJ2^b+mqA+2c7Of&jbtFd%H&x)*j$dzXcdt}}<g+iLE8*V~v
zS|{$OZeyA|TfMFkEiPSS_K3779c&JTFMp6$59x?A?k7hgT_tOBjGalSCYg388P&-d
zvO!0QxG@~TEa$OU9FcDHlc`Nptn-4y^eiRRZKy}zwk_b32JEd@DD||OLuAZ;yxKJY
zZInR!DT1!8{f#>%g~k1`=3-T;ModdX;8ms4!UN=@k{?$hZgIQN^+HRbin!Ba1b_L{
z&vN)1KKx;b*8_&%;?!?t8#)936IHE800-m{0>%xj(p!M#fcY{S$ON{6uzD0|{ilQ2
zhnd3Uz&f)H7TGdd4tKN7OQxX__(=W8>(s_sN67&!LW-w2r44_ip_F`PK0As2HjhH}
z@3Pt(xvG5D)5YM*-cC=m;LhjpF@MQvVZFUXr%nd3&x{DH_U;@r;`X}Z%ABHv9L!Q*
zJN4`iO`p$qT}QKdLJU^L!xy%L&4+$ta>Oa}*xHv^#d*Xe%qkA?c=xmIh&LV6)FF2H
z)RT9IYZo^(iNQk|brZyr?76;ET2qxo6;?7Mx@v{$+)$+r2-6GV&41pe6@PYl1u_TN
zl0Qpj(<)eAMfcd~nhoP#;#^8$-o1FgxY`5Sjpln=32$$eQg8mSy*EmZ!aUbDlz=C0
zo6WtczSc0#_kO#@SA%X8Muzc!-)?RM4znyb+X(3McJo35-F-4M=5i0&sN$hh$L?sI
zrESfjP`w<}9dkf;OmJ40Ie#YW^V_C=YGMB#O=e~2uvWc7lbJ!@_CP5|mhIJq>3Ngr
z`L5#{R$Br}VW{x6%zZk$Okj{^tpMk7DbVG)EP`(2$5|=C0qSGNP@BiC4BBxP>gUr7
z426!;&Gupi#%U0(@XE{}K3xEj?n#CjM7vxBX(N=LsR^B#SM};{s(<8cP1OrhmemHQ
zU^|8>lR2VXC-l!Hrfb&E=-=*kqjMKp&6K?4h$kp_HarLdz_!}mddnlBdfv4%xFc|=
zI~;{e@{wuwQ$y7Wa?(MV;Q9XWa@hgYRDyG$W>Zx89E`K#0#RWE0_(0VsvhbdheV<}
z-zoL`!Ezg?xQBjrihn1kcU+Y4W=88;$0hY=IA^M|KAq>5vqtR0!UcgJKHJ2rLKnAB
zR6s4$Gmf;k2ae<nRDJo;a0s)TXlJ-Y#vfR5g_sS&+$zG}<Zybsqn$&#-!;}ww^6_<
z97ca-8)Z;mCuyLARL8ekGf(hBp;KeC&o$3v?0j#tnG0~H)PFSdO!J()JV#<BA+Z+C
z%<$RQvd(1t9Qp~?ne3ZExWGEIy>tF7TW3y~xIJ9cIxlY?1KknPuh={Wt%WjIZ64Dx
zl=rXDJO)eG!Y*napA2FrW(eM?S85)Ep$jCiv+K6IIjnXyhhe=rY>qXD_cn&l=iV`?
zxOlPo5`8%p#D7i`L9j+&uQ9ybHY>YHCCsEhZ~l_IW?`Vq5IQiGQ<wmHT!Gw9CBYk&
zmaM!nhXr4J9kcAw9M8q8C2cP@cb}h{TcIZ0wbwK2F?;d1F3;-VHrD{kyUewtbgWw*
zDAhHy7u%9C#22vQd*QRbpSJw^>vXaDp{6-jmwEZptbaE9U&x|F$0JXB6ffwuD>~}L
z%BBanZTvGHR7w{vbth_<4j3lNb8|htHIzr$g=2Cgk7?%9;^pqVa+X|9tK!w__==Ny
z%<~stg<EMfdkS5yWz-RiG2V+&`5`D_VeAcNrN4|J9a9G`-S%GlmeuUL@9xo#ZA}|o
ze}*Vu)_<>YpHK{}`kfGjMhY7w-hAIY27%FXw5L;$Tjl+mUp5a16oyIdG+(PWH&oz+
zh91RVHU&-H0n8Z(FXJKR4nZqLIkIN5XazmDm;PFRcZa=P@-J8F+w{|S*YML(7>?$*
z0IXdSaC0uUOYF03x_M*kf38=gBnoA2WOH<KWtZ9Z8Vmt2mtoc}6PL534JeoJ4GaW-
z<Twt#*Hicfq824lKR^&{RXQ19ZatIRU=D+x?cPIYu`_$v*I&veZCU;;S0#XBmrPUQ
zM~b8<>E?T*o1adXx6`Lr>zbe<LNLwSw+85irvIR+(PI+;l(*2ly*Ix-YHd7S7rLC*
z*3uQ|Ino`U9xOznfHPh2^bi~{y)|@y0X;wG-$K`Zjon*wZaLWW$ecA@nKFR3nI~}p
zo9|D*zx{mDJbbN^&yV?5|9W{l{dKB^aVaXFu=)Ca$}gM+Wus{z(%m$F|Fija`ciZ4
z`_lA@!u*e({{G%nEx$0FNfVsfd&Ci(pb2Bwc%PKbuR1tg(i#aqBzPPNi3F#ABbd<S
zwvn)(UL(wMc1eMoa>kBx=H_1M%A8I2oVmL*_r&c)-nHl7v7^(ZXiJxVeR}XZs}rB)
zmw6;D(dhMaJIa1EhL9#p@)(sR0C_Z7zJiPGBvUS9C;wVVM=b>DM9MK=qemF^)EJaU
zI~j#%nLl?j_VO*r;DT~C^j}YZ5V~Fpv^1Vrb6%3dgGxHo)OJBrEQMA+P{&{!=nZw$
zQjSL?wO+Xp`|(g0NV;`26%pBnTxV0QEyacLdIf&MEst&1y2Ku=&Q9XxYBY_GDrjH#
z2d;7wS8xQAXMN4^^Q^cAf--<LH4=D))lKR!s1VY<)G_9^sgBkq+@g+uaesGEN5ybM
z9b@X&vzF@EcH@gaT+{<jp&z;M>_Bm+yLKw3n#UamC$6PpJc%QF6tFmWF{x1}H)t3F
zwScb~&TANgqp1TGF^gB@tF59S)f2_zzIx25^3_*y6n{SE2zq_`j;SSP*_Tec`al#)
zl+wkCWJ{<gRLR+z+UWy-4uZU1n#ixu#~p;YoQmUyhZs?}(WmzgSM+U?N+Is;+eYQ$
zME0Cdi6$A*%8)S`<q7^Gg=sx^>il#YZC*Kv^!vTLbXlZxEss08NsQUdTyHRv=7vh$
z28AzjPotJKcO*TL7YU|MeYF>f@>-%rA&B&Vy|l>aoH<c`9Y@%IHc51=DfyAuj+j*N
z{!X#I;<lpwqm#Ajp<RVRzTXhlyJ=W+eqBt2WvW$Qlp4Gt+d>yr)S=Xf^P+E_ae;et
z5>j*9YLKe9ZqLg+CrgRorC(<G!rdO9m&SROD4nWCHj^SO%{P<r$5eB$o=07vmSwbt
zC1QE=vngq!EL;kI{cU@LiT*6zU7VTyf^!<QM)OJhpy65exfs37^)#7zhb+r@72`OM
zzpmFhPpUZXuqr*sUU+)X5koVNa4ZBOrP|s}M)fj3i@j6iH!5nY)AjuZqtHz7T?)Mv
zhqQ&7de(c~A?g98OSbN$bs#(i_SSW{hEXwGDni^aaxP7OktWH@!e+dltBwKPp}5ai
z-w+hQr21y83II}l6AnXvvF$F>^9;+7tCKOa{Oxz${7El=CgPm8|F0!|@4L~>kLK50
z-rHat2_a(aDPO}izPh2sDVbBpb>M(DDRMbibJ#g9vI>`Aoh>OeVa>NKPA-@l0C`zU
zL#8|`r2CbB5XB2+AJbCD!Snaoju)zpVyO78ve%2W(i&9W>tV{d%P^vTjt}sTthCqL
zVT89dN2^xK{BJR-Ygz1^M8wu=dt?&TvvtRYf-_`nzFz^^^ApC$E-+gNj|+w(iFlOl
z15xkL(wd#nIj;7yAwcJIN#XTyEJY=kQWpNNhUEBvscD5dwelkYkeeZ6=Y1RRtpiBe
z!=66<=li#ReEox>OY{A|`3v%CG>?FAd*`24C9q^nKlr!LqEkYX3jx$~Lx~i4Bm()c
zZ}{4r9EB9H7>m{)xi!&B(oy&uK5n%WGNE)(G|3n;$Dx(``%yObqw2e82i*@N|H@qW
z+x(M%zL&4^&wQ1?g_OSfS9AK-oWAsUCQE;95=9I6890Fwo?xMbRGKJ|)6x!{z>xwO
z$wF%>=s-#f)4Zs5Cf5(Z<rI#4>xGq;LSShkUcy0|`k__r3jkEIc93w`S6QW&7XP}9
z9>B16&<@XE-}$?7<P|;-KylTfZysz)G`(Ga=G9J5jI+~|$%T5Do)#a6^=UgPC6_*P
z!D8$K%1Kr+r6GK2iXnEgJ`BQs*$ohdEW|wk)#+dXOdyNF0hk59B{DFzB=b1(SzaJ|
zGVnMoJCMU5Q(eTP4ODt~I~0<|qxB_;C-A*z6#5(z`AgTQYT_Mjzs>4-6Sm)Y?U!kP
zQuOG*a%ouNcB2+gi)QjQ$bzLCU+KAFoZ-$H!~p%AK@7_64C3GI|F!MjA*$*8xmW1I
zWSf@~T(9H$KJPFV4iFDTgjmg?x-6Q)gG$qFPjhhF(;5ZFkGbAbf_CI^EspsCa$UzR
zY^HXF@X<F%%FbN=9{25*`xQhjsmS$z>ylwZy2bSQIHqI1OPjEQ*qC}fN#Q(sB_U~7
z-PNXeQS<3&FFX~~P;I(qmCWnybvci$yDN^J4rl9v!x}uPsVZ~f2VSl({D@&_N{fx1
z7Jlqm3O#1Q-!bICZ_cB_h%9kZ$)(IUnQy%eDwZtH&z6~`8sLkw2>s$ljGzsFv!*No
z=S9ux6fr5UG29I$70PLq^|s=4ldIu^NX2o7vGh_!%EdI^^ZbSN6bYc_yg2pS9hi#v
zJ_l=_V8)e=&P2?dtr1)(c!C?BYKcXMhV4j}&w707CEM4I!r_oNT<v(eZTo7NRZ>vz
z22|8`j}Qy+cvKMMB0Rp(U6fjX{t6@uz0jiBm?0*A?yRCcxppAV%VZgIUBFBtSVg_1
z_;$N*UCO1iZ&#kPGjxZU1kBT_&#&iSJ5nqIltlJ^KW~qmFE0lmx5`7TxGnkufxM!=
zQSE%Ht>U_*c22}9YS})NZD4octOIXa?HoslDp7E*sFzIZhNDFFc$W5m^*etvF^loR
zpTl))%A>4t6JrKKfvoXR-5MgTP}BX()<9k~-Bx70n#InM`R*9WhZL^CCsmk<<4X@l
z=weYB?sjh_(53NC9P+LuzR>heMO^1h9;glIo)24yL2lKQj|;y6Dg}TP-MvB0yX<7(
z7?lZTy7OPG7y~2ii#A_>w3Y93VW7)<+9Tp&Cv=BJ+<=afJ8kfTpmRZ4uoLJUn~y;9
z%$F@76cdYssljD{?y?iM!{TkgW^ESxgJ8?)`A}l(VH4{)OPzwB&kW^-X)cAu$$Q+1
z*I~If@G_J)=x}em^1gc@=181yI7nDwCOBRAvV6w^w{wB5m}72#k2`TYEZ+uh<k`W+
z55f(ca@H^6_FUXuVOJ$?Sm5cH_Pi6b!?JB)mXB=3^g)<u%C16~#muq|30|Ry4wl;@
zKR*}tm;~#cxE+>n9k<vf4<F*ul2Hh8)}e*Mft|T6RX|F-z+#f7E)MJrpGRd|M=a#b
z06xahVy^wCSv$Ky6!Q&>yD#pMp$%0=vJ<bvlC1;gyL{k7Y%FLsJ)Vih)Wag1NX0aC
zh3mQ$F7N!rCuMgbc38B@imAH)A9sjTz?aeX8Wfjd83YuUEKwyWm&XJep?`Q$*tcIl
zetO}=zQA`;yuW|E{P(35z$nKo$myToU&0Rx1Oq20e3K5QKmI-a_VU-6LZ-}_3!m`s
zw~C}=JdiX7-)$iI%ddR;7ykGYE(Y(W|HY->!*b5N_S1i+zaf2ID%=TFI5$S?O5wsY
zYa|l)48l)I;ZbZ<o^U~Y`hR{Sm4x>zy~6K`t^vIqtusk&g;15MG<e39e?qPueQC|`
z^Pi+I?U0h`^aZ3BD0E^A9eH}ck;`U1hU=HN;JZRUsJ7GM?Ws*NA-Bm&=3S<`#@R6n
zdwfwuX~Fy^)f{BJq1%Sp1NdWvV@|0Vf6!doA~N`Z3kr>?A-$BN?|+CX+T;})O|Gu=
zi|wwu`HCs1gjZav!Y_zV%FjERbVZ|!X`j|={U&Vq?Lyy0TDMSFw~8l0H0(B)@h(n$
zDcakX{KZv4-rLM=C)vy~af$Czz7myt-jjBVlFeLIlCMy<#oa3DFv^eL`TgyS60($5
z?{XZXt+PKj%%Ovl{eQEKqK;LPAW%Sh;qD*rhzOo}BaN!O2?N5}ir5c-=TiSA-867P
zrI|r5ZdkzNwz(BCt{q0arM><^8(B9Qyb7*?GRY8}CT%@aF8TTof^N}nl{Z8`DXUFh
zlkZ2nq9^LozMAa&weTJlvT)R&B;UPJ9FR9ClUG%A>2q9ZDu2J(RAyF-kxXOaC63|p
zrXwH0tDGqP<r1z$PxXxl^F}&b-Cz1?x_?i`xGLfZ#)inAuW5%ME1#)n8gZ9=RjEyv
zkRBM;IhE`?E+M^T0+(td7<aCAiFFMFSFBfX7HM&dnA=u8V+{$CZfuQo!2+SiGeW<n
z8nB4qkrou!On+fgI-mu`Ju^pipoEPT%&roSXdvGiJ=3JSONtTM9WS%UoQD=~A%yDU
zL33uL%bdp5xU1qV=F$~Oo@$r;kZ^EbL3kg%-u3b<bON>y_o?JtG&WNU_<PypYD1@V
z*Ji|#huPQ@+{y18W!86!-oVj~vW<j3n~Z%Cy-nLn`+s}Bh>cLF^0e|y06gZ6*m@}0
z+PKJw6iq=RvZxc1eyaRm=p|TbM#`cdpu1E>LzRA!vh%thf@3DIU-@B@E*9V`DabM9
zMSCak!sgf1-(}yBx0>r1dw!{IW~&Wyh*wz7cD%-J)0ar3$!nuuSebD<c(;;eO_XuK
z&c8gEYk!n#b4|K`6#?e;dzpx6v}jdSY%5te^Rkn_XjPNHMslQY#4*gAWz<kr)*JXo
z_AF_b#p<T;kJ$6yQlkcQG$d^URo7v6SybD5zTvQ;H?G~StVyYG!-~no?grn!HQ!bz
z2vP#>2kitwdc2AIJVI#5&Ycv;dzcP?@Y5f^Pk+sopQ#J-e7(uVCQPeIV{R9GiB3;G
zE{d+8uER0|!}dAAnCdBO^7mVg41CFvU#H(Xug>@Ix7iu(;!Zq(!Zih_;c>++e?un&
zj>7RKSxU47+FM~EB@~oET2KKpuL`Qof1<{#D?fjGX_e5>mAn}<(Q0k6z=DnoQy15>
ze18KO9QNdHZfGOGR2n)ig@)LDH<U+GU^V5rZC3+dw-iZ%3C!Zg0wyv0adoBSCrfHE
zB;`EGl0YqUQFKM%pMh$?uVKP;)B<mMjpytSz{BYA{jfGA$E3|oyPr@McZVoqd*^$C
zUsuJISiwr6+to4l_;cnk91+rfmN{x_yMN}0sPK$A>is^!91X!8bJVLN@RsJ-NBJSd
zXY(Ky0e2O|r$Xd0I9JU1j>splm%*i_++~_aY%pX$dsV&56IK=$uL#poTsO0@*gIR%
zt|kdvC1d#sY8kq?^N3wjY7qMn#f<7TMZvd&xL%zz7f^e}Pvjymm~owCBei?YJbz}=
zX0_rcU;?qvck=Vgd6>w+(@@;OiE4GsK{G}i*jr-HYTcuI<0|Dk*+W&8ElC<}9T}kq
zd|Ah&0-pvJPz({w;}D;BENZ$)<1?{R7^ap($$21b8*5|!B)Xb*s;+DSQr;;Ht(I1k
zn=DZ{ri@y!S$YJkEUi8@YN~lZo`2v0+hZ-@lM;MGZNvI!W3|~~Ik|fxwWla-Dc^_H
zaE`R*j?#iplw9C|X{}0&R0l}Zy2ahpGFa1gGf6d6_v13eWGRt&pL;D+pzZm%l*%xV
zHC4meN|7*%H>o|d+P4&zVZm`Pvyy4q*+7VOKd^>bYrFj)9!`)Qj`yGdhJOLL&Z_UH
zqW}WBt4@}tm|9l*H~?t(zZIUpkE5OE2km>kf}7bxeKAOOa|X$St9(#Z4mW&<S*K#s
zga=wog_6g1ZA;9C;sf!>7uky(nU5AY(Xi)g2d&X2s^jW?-S3TJh`7ya;QoBPHTw=~
z7Kir14Bo6>#og!Y1qcIpYk%8aBKFc4`q8l&&45~pqbbf4ywm6p0FhDnKeN`!J7M;8
zT&&kRQBT)mFnBA5fw%10JA=vW=|S%@e=CRal>PELu^?;+mH0i^Q*$)U_tQ~HiOKn}
zA;i|G2g%ZhQ+~A!LOZT#s9EroCw@@}lA`u_<ji@!=0JSyct)S|*?)!5x(V;fD_j1A
z_%5t(cL85x-hr7z?vPr(-)lRcnrTW<OoD7_KhYseC7D&dmQ|8J_?PhOSe0v&`pz@h
zY+O8~*p{lRN$S;Yp2#)5HpgoZq^xzf!=Ofj5GkUwN%Hzr`hbAX&~J|_5ks}i<xQg&
zyPSKUDO?B!q*KgG)qjH^Tymiz+f#O5l`zFwTCsNJOpxGj=S*;=r5rzjGarR4pwylN
z61@(Pd!8e|yyA8E#McbQboxx|^-}ERGCBsE*o0zffqA2!C`Q8LQDC5<^wfZjN{NPI
ze4KQ>aqXlsfH-ZcFdm$!ea4{>weiiaqeCG|LCxyVISot?9Dm6{Rea(!Fe_9`hP2j>
zdgE?e52aAnj@Y#xo|{ZBrM#hM+>e8APDqUyqXv~)>}EAQBIH1QYI&?`O?{@Ydr<_9
zRaWXNsjpPofC-4rTK3Mb$=_nzf8sbd$7n@#1U3+!jW*5VJ*SCt&FhiqogcTGlhXQg
z0tX*T5Saa>SAXq82_BeHJI^F+K2)E~BAk!^9<hwEFW2`I_CRigYp4w|hz!!Me$prC
zh`*saatb<_X;$YkY95Cxce==bkPqvy3eoNAW1g4zJ#*$t2MR5(gI1n!toNDnh_v%C
zln|CWD_1P4d_wF_Ba$mgU?Gim;-4r%b!7z!gt1r$EPrDtO(VVhB%Pk0Yb^&*e%OFk
z4PV#5?l+*c3)h2TOll8WuJ5Z4k_c_kZsy5`m13I)e)sd5R^jR1<#_;|wG(|d^JUld
zJkZA+GY##5KCop}MLvV)<U#Dx()aQxU;M90F+j={dDNrcj_QW?eLX^iWZHJ!=4$q6
z-Z5kR9|W}lLPM9)_8JrdF*uiR4;tYC+_zB>8omnzSOef!m(lhb6qmuI7ZaB{92(97
zq;R(i9U7Po1iWO-dY94m8WflCVki}t&ng<lf8t2{`KPz<?qp$KieTf{_|gCR{C4-B
zyIvUMtzc`1zkR)n4>XftuZF<F>M;EFW%&8-&pp+-sysne(pqC0Rgr}L^k8ie93N#>
z8J(FG8pFmvzx^_FcfS92;*&tZa!*OgPV5UPDo75?b`;5V6qUhLBlDj`vAHv+$j~7a
zf6-BtFqf4I-!xo8!_=^?)Nl<0(;HjvsrWmNjLs2%=`haod7PQg*X6gbqtXb@Pqa->
z=8`OIV!U7&vf6~Nw@FLsJ#5laq}SVR(o&pkl5F6MPEFCZ*|EcG(Q+EMVoz5ra2!_%
z`Kc}zOqMp$Ui<dZPH1+ceL!2QneAUMf45iFMN9b>duc6v=n0a&-du)ZCf`$>nagMO
zPbR3QipEQ-Er!r;oHK@2GUBl)uo^rWY~K`6oOXUP#r2`mQhJY2X(<vUBeO3JX<^1l
z;BdV#4WUk!1)t5}6gj#YDu2otS-2;UWAw#rcSvM=Axu!nOkC+DS2#?VcZDK5e~Js!
z*np{0X6H>5n>(QR;TYs<7A$est-v!BACaU*y3M?5Nd)P7iso~X{svyH6=av%2sv;P
z9u}S{EVAV$zrlzAY8_PoVzRnA7U`)VrOp~rx`gXXxo@iHTx?}!f^a7E91wL@ma+nq
zEEJnNbBb0AlwJ!>9ah6YSub$)f1~A;C0fsqzn~GS7F{d6J{*l7yfDEb7*`5ZW0de-
zkDNtlkOdBTSxv-@#oE_8#ZR?K)PDiJQ$?|1EwrBi?vSBZ1)Vr}R#7^V&yVDu)>y%V
zr-_|B4MCkppG>{uj8y@$f>nsMw0X&;Qy!nFR0llFet&94r9dreK#T;Ke*<1HSAz;x
zD5)DY+<$)9(;Q1Jh*ht-8LLs4*CiuOrB|PndSx0n%`{J?22rIZ(UTfSEVc`^ED$Ph
zVl>&Oyva!`98x_q=OT{()K)%eRA~5V=h;d9$<PsTI{dGGrj#%|A>T|=&dyR@sp1KZ
zPl}H6W~-g)Yekr9wC9o1e|a-YL6S6sPc;@FW~5+qY?~3~azrF9Bc^^T?xpOidjXxT
za)b6H<Rm@U*o0!%W;u8mux~!={?~_HKXyToZfAUX*j>c4aBgD})F)O1$KJG4yl@sX
z4A#n}!M%EzTf2^MKsh%b0FrJ!{*Y)wtDyUj4|_UgX@vofZ|Rg)f5l}o!b#K5vLR-*
zY7M_SQ$=wji%_vtnZ>`t<97|h3i+$!^8U+c5?<3UvxhRMRq)%xE_2!-O)Z(DU9B?`
zja}FIhW0EABD=4>bv-!5J<?vY*gn8k$k2v_3nRetB79?dQZ+sKrxTg7MEv=wDI-9e
znhHj}H2B05E{sTee=M^`EYgQMC$)@X^U1~5(a2fYLa~Cmd-hQpH{nbgMY~j^%96Sw
zN2gKB1ny@W^}Oe;3YKLBZzT_JrOsP7P%2eF%|iFT?>d7*wePSGVUJ=;PWi(YOkQNI
z(JF2(fnS#!HJ*Xol0_i53TR~r$99`9?rS%O#EP>}(Yt}Jf8tO^I7uaR1<HT#271g1
zK9~WUZD2%E-Cy=saYk6|$=SJt^l8dJ_7NTi(ox&<DIaOyr~#JI(L!O&@PHs?_L)QB
z<sdpEV_;9)GPeOG5?XV*UFg62h>8GzK!CqmEqcvGtH7M<qD6?Yd4VXY`Re9@Fdu>l
z2UTpg+*xd3&u%@pe1DusA4(%2RV(bTCxvxzxQOq8{8_71X>D9ZZAqJlQ1Yg87^&jU
z8TtK$xCS8!zq1G;rMp$uvn69Yjgq|ycy4(wLz;kh8$)IRZ+43<W{4MDgJLm8+=(=e
zg^1f(C&~_k(jn+}$~O;>!RV>DTVD)l;Q-^!%#RO`%DMY}0Ds^>LUCBEN!H%~DSo*8
zG5-1{fBdV1(VT=>JJ{MUVPvCJhP;jzv-TCMP+(c_+FA19oE*ta!d#pqnX%wyIdZ<x
zTn}3kIq1Q*vl{DNe8NOprp`xq!bDCA1oq8GOcBjdy^f`d&gW;je7mAmt^~ny9o4NK
zXB3T1KblLkrGJ)uv51T9u2ihqbdGm1U$ACSi<P=I);Jn_GD6zz`C2P$$8JY1U^m*X
z3h5()<g9E;Gh8{GsQs2>dms{}ZY?sd0L?jw30R`B<<{K!e%ZlJWrdm!FnGP8F%`}m
zR*QSw1+`UPr^X!RyZrfb+X2@saK2lQ^z~-|rLEdi{C~9Gf`k_kkf0l#wk>ilV{P=B
zm)V9!t)8%4+=eY}(B6XApMxle!p|+FB3{=)+BVX{=Ev>xrFWB^ff5n&_hp|(=N))&
zt;I3mZINW1(AK|KB(>B|M3TQQihMj7T&#6d6qUH_D6U7*7<29D*wp(2qbMIe^wzs2
ziWKs^T7T$ZBZ_Lwb@sl&4-Lv88RmXCwyJ4*`C)8&AI7%Is*Q{j7F+0!>Ir|e(~AJ<
zD3Y23$cwb8uu-w0U!RVf(XLpd=2iWSbY@Z4KYcW{d8k*ltc{r7OPO_u&N{;@=cFBx
zci^rZkM@M8gZ2v?dXxJKQOzQ^phkI9p(FTd-+v*S!R0lmg@WM=CXDZunVn<ei})=p
zwA8mM+HwrsjV^fIz5byTqx@C3yiriC(^8UK+AOMXNB<G0$g#g6ZNKtH`K?IBZA>?~
zz66$V?4{z(<*YF30_-_^R`~iHpt0jz@(YBJYD2(ssNT)U$a%R$H3#*|#icf?IU{hb
zt$%+FY6liBoa=-t_|$^C-T7H!Q!jb3?#p^z7_podTAE&fxvHw-h&>AyHqeZ*IhKB@
zsIiDn?k!iL0wG}?uXCLzg3RT+VhyU)Q{MvPvWM<@kyq5Bd4Jte+OXVMXSSowO!i)B
zH6ut8j`V`l3+k*wiGoDLn)j-+irzb_Cx7a!_sM6qV(KNL+v?zgfgRPYI@nNrovIgq
zjDz$dcR86;Gc`wdm#_H^pLZ<p;EDh+*96yWgE76ycG7K5HQNZ=c+c&&@$@?%d%3)$
zxbH$ddtGO?b?*y!-)MLtgNY$AmD(qoVrh4`-Nsr_kB+3~#P^j`cJ-kiFqh*<zkjyq
zGtD<!YyDEfC(S-r8?s1u66v<<#{$zGjqTV<U~(RxR`dA2(&8l+Mw>|T$DiczcRc(K
zgQI=_mxg{FmyyFihaY0U_XJ@i*+dvrjN2nD2NYY#Fq!N`!f;4=`eG92aTY=%u}c*z
z2h)??ZD~_F0VWg%c51egFU629|2py!|MkO^%a$vY--q(sm+^jjEqEqR;}=y^SBrE)
zQx;eoZZYAd6wY(n_{JVx<Fbp`=sy1sn4Oc!m(lhb6qi8;FcShYHkV;m4JTE#TT5^4
zwh`X@S7_<GFvA%RUw~u4=lD8kc5S>UkY?2;LD62I34-?TcZMP*Qtu;QFLZS*k;9p9
z-W0Fye(iMk-QDHw?weObomatv$L{Ty4)}%Um*6!T8-L${@Am1zK;RMZ%ojX8c!E6M
zAYZ`0ukv>pmuB1kEj}j-CO&4$nlBmYK^vjTRDd7C<>{xl@9%VGU#sZkr`)E0eSW+9
z?QSZJb4F<&y1#$9lNV$_nSdQQ6^-kD|GE3+?$1-IHB<o@D(#%L4OE~RKVs@p8F|%1
zWuw-_!hg^!|7M6N^vo;bX~RFi|JqHS{N=lmZ-WZZO(`?HH4Niu$Kt?rv}Sc#hTjUR
z*UZC`h`@KK?u!sw49dhAg3#YHgoVPWsF0Qm{Chzw%(o3BG=Ka_cmLp*-}7J-L-)Tl
z^vkf0?*85VAUfJ199T<2*f15hMOb&JVu+dSNPohbs5HgfNgU)1QIYZ6WZA5uozmU3
zLr`c96SvCaXf3RNRzKTcZ;Nub`R`;G{^`MKaaI%aC?`+(V%M}h)P82SjM<X0w`zG0
z{0#Wc!!^8Q%NoUQ>}LS<wB$R8vhdWq_SZzzDyuycC*_?R5Vhiid|i9_(>>(plrNW`
znScD;SMTI{K8V+8xpwf<T3Lu6gqJod@V($g8L#JjT{AA%3bX5IPs2Eqp9zlfAYP~C
z+Q!R<sI2=KxNJlfC^QZW85WZvxnxL+Y5Dxl<>$VISKL^f{wQpx#oIy62gR}FgJ27w
z0tem#o3E^_SPtcTf$35`r^>p{!ggA|9e>=MSB~(5aPvIT%z|05wo85%^L5S7N&qXr
zZHCPr4`%1Z+JVbD<%0PjxNKC>!z`H9*c6YfO|85uf$my)?NPi=%e8|S^%cU$;6=<l
z;0!NY;ZqW0ZF9w@LKV&8jrMg~t{uEgpZ`AwFT;!2xj)(pBb%?1FkRIXQI!Eo5`PT`
z@j5Nn9$uil_wOmZZzgf9*ATLHyPR6<HKt8XQmqYLP&RsJmYQdZRv`_&b51U)tTufK
zP1jtEr$&)GUXtFJd@j#jKZ2g6R8KnH>(|LUt|i7buhmS49fvZ9j#@kP>@?8Ky#M@Z
zb1`lL5oo#?=%<I&+iyyUepNyqmVZVh8qcX^&f^n>Yd>b7DEF6Ulu*5tC+;yfIH`?!
zg7mkRH#YM~aeZduuFSGeh>ruMOd+*NI5zt7M%t((4`p6m!;K!BCuyvb1eYRptofk6
zG0VJMd!!b|wFps9uOMVYk+?`yupXMf<xPs2RXM6lt&H8CRYi==%w4Op{8cba7t6aH
zGnR;BY8<*;<0=Cwpz9e_=?VEtNu?9^gUa)$v~S|GAzlwq&E0Y0;CVFfq{e=ZtGe%N
z?#gSGxwJ#`U_jRfy0%_1KLB&V%B5xq44wa%%{dwuNiYXD)lo0wAd%@E3+jMTb{tCJ
zD0r6Op~GX@o7m<#kb>Y_p8xe}%VLRJAaHS6%#~H53|(1;vy9`y(S#=JTa8M4Ud5uq
zzC)KSIvOt@gCt8ep&i(PfQ59;jBV6`3rypd*H{J}EGa{>hT;B~dpa6o7`$~8CFdt4
zLnh`1Dp~b;%#EraDT@O8znAqo8Y%?cq3;iuBRd*ZDVjV*sy1w6DmQLRtDt<Kn8@<&
zrw8=@9=6jPrxkB!dR+?++HyxX>_C^HI~o)+Us9?_x)5Occt5RtQXm<=%nVuKTIPgY
z-=de#ro$R>Y;FAGY^EE&$$aw)i%Q2RUl*6>I~pqs3)FyDQ|~PX7bBaO9XuLC0ri(*
zJQ^#1H<pAbuuai~k41AA(}-keu=;K8*}lnPJL``>wajE!0EyH%*Vi&T63TR(&GMTe
zlzV-)R@=_fc?!BMN9Cp>XdGMk9e?6}f_rhpy<Y@-Bv~7jR7cH1H4JM4aWM}qdu>j*
z`IHB6)!=A;%0rBa%)5_ypqDS$W4H|c8J0eOg`7U(F`<fj`G^O*pEGtk+T>lul6EAS
z$o54_8ZMosxtTA|E{L<0){dS8qkucsDPK#eb=bA*lW0DYGDF$PcW<nd#85$Fl~`Zv
z^xI3_YNR^!;yjDVwgyUP>6<6D^o0v6Uz5k(RI^+`j|qH5Yms}%8MQ@TU(ai*r4D9)
za#rdFiovrMXZ*tJs$H@K3ZkFRD94V>=bNQMK=I5e3v^m2<amoPhv{WmATCvjQ{z-1
zss8ls!kE4yzX4F<xI33WwQATldhaoBZkFtLQtZ9v-BMdI?|jziy^e|U*~QhIkNXZ~
z-!)4?b#uNf0hb1jYo`fSeQ86J<X#ScNn%OP_~isJxbSQiXHw^fLmnocNEQA0z6mA$
zvZ58@%4j+mU#p#jFsd3$2b{N}MCCnrw;Owj1?|c!nFF;H8Z)8lt*F#I=|LSg@q+U0
zKFoMI9~;qd6%l6AvCtY<B(&1?PABxC@pgfnblqkSE?>C3ym#qC>caYi|J|^EzMK=H
zgC>YEq;I+JH+`uNjh>0EpHJ&iy1xK3w}j<zEakSrx$9j{)qK*b=|D?LTtQO0k3cge
zrphNDTRNdk<D3GR#|W6ErJJa7JraW1@ws8NRE}mQ*-8+FXt|dhJ`JWb<auKRtMH%j
zNU3p7DJKvJa`a(X?tCT!FJ~Qp;VWmN3Duok>dH3n%4zRg0%l+LOxEA3RB~-InbE|T
z6R0{l5{XS^f=hcu8PT|2omi8OcT6a5BWmjUDm}%vW|{G8AsK!3fqbj+;UaRYRhd!e
zwl9t+y7!UDRpRX9{?e*X@c1?UV?dCiXxek5Z2iEw{}v#JiCB4G|Gx!L0GAxm+4r1I
zx|zf=&LSsw?LOvz1w?LtV$~hyqQO@u@?3R5w)_s&A=&!W=Wr73*h#a~lI;Kn=^Mm5
zG<hq9W40TmxcDlk%%=DM1Bevdm%&Q~69O|imtlE86ahAuvHt@nmn1+MGzlzOA`f}K
z^Cmf$VL%!e1|K;73XYeTen1)*f1W+==H-hL9xcQHmw5T&w4nA%;u3g%jsKT+skg~r
z<snf};h89NT(PASTt%9oBJfhYzWn_0!vm-3a~5oTjJN*3_m7AFJoL<1s~C5F_~qL}
z{GgR!%F97m7L*-+|Hts_!=HPqHB?>zDsHVY4OD^yKMm@EsrV^FWdi3Kf6V9;o&k~e
zdGbu7+xX}A--hnZU%sE_tt1oD_LKo{4a3yhf*sMf<`9<w{3fDG04}m9a2M5m5kez9
z(+d!ad?ySGiAhpEjpy;aAQnvALj6GVr=R)ocYOQ-jaBf&|5DR$<2Zcy*YIN$(gtA<
zF-5|PX|LA^^FgxU1CgCbe;5@Q*Wr2+7tTydW~5ms3%v@ar@N|%f=Q!5+%PZHs4h?d
z$R7%rl`-B8{ht)Wy?n7eTB{1w%Eq_2V4DU_%{%NekSfcp*73W*JApfzbMZ#1HL=|&
z&<N;m&NmPR^TyiGMl>*IC5Y=};ssF?jWGJ7h3iPT9O&86dqsC0fBRpE)oz|Pu;QFq
z?(T#YSIh@9!z$w8>Hj0)py(dR$G_4u=~e0zF2rg#PwQA2ADGqtv#~PFV6rC26nY9)
zoNy3?LD3ms=~cl+(f@`pJkmRtZCLOZk;3t1@a<=F13u@Og>$?Qd|t5NTmzpb_#8oU
zAc$(%EIn&_9|-0He|@tlT8l#RLJi!{>IQ_?F{kchti}lj^_m&0u8j0pp=1c9*UE5<
zbBpc#g;4EfX#*)^nRDV!P+889)Xku>^jv&f@nyb(%~vs@d^=oWoV)qjz)R;u;XZij
zz{a{EDTwI2)oX=H=+LT_%h6@CNxBO`+s)htXe#Fd_W@0Le<t(<Tk&STf=+iqQ&mYu
z9T3cQ$;N)Zw(t@u%fBHBzna8lbw?y?*VC(Wbw^*|B+(i_cxHmLYN?Xe!7$~<STB)v
z=YpA2qOEdLtzRiUPqib8U>#d&3hTz^7WxpA_axb^5?(vhx1)=X6}vvB0QUuX{&OBI
zO~B?_K-*|$f95~~xv6q6uVbGwmH1`qV{`3L%HC8b7-H#dsvFY(GBH5((i>aoG&D}*
zD4AJUNkHK$t*)6xdBSW6HM5UzPiycfE*T2D79C_wSt@H1p-C=E1kGZLU<*?tiwV#p
zwT*p_4b+V|a?fhy0-zuJ%)~g(r`*kpusuCJm3-O+f8tZ;1ZSo<)-W!o^^9snW!&>L
zN_@s$%7U92TUdG%T4gBn(6**mFixdsQJdqs*x@(ZnLfU+N?l6Fv<y9SamC7LS=$F`
zWXDLyZc%`wDw3FTb<bQ!k95u`%_yKlZ0XOyO=BPPcp3gmNve8&Y3wb<IyY_<UQ!cU
z$jW1le_yi#&3iAZl!*&)qecNlQG;0AjB5drp%H-)>L@EHD?O4SJXZ@HC!(W>zAP>+
zw)C_%ro*JQ)L-ZCc^qh<9I3h;vnfk1VVW6#%_^spKxL%}Wyz)GKXii0v!CWroOf&C
zG=5o!Y!-FMs!%*FSd})KiRerxaXpH!Piuy#e=WM;Q{Cm;swHY?(rQ(!6XY{jhw_>^
z(<sx1(QB01*7rk2M~xuxT)CrBF$$FWy%EKtPHVwXkS@B(B;FV@M6r}Z7Zs_s;idIx
zPhw_`V_?#;_glMGm#C47CC+l|L*2K;)fklMyP9HuHAe3Pr?zCWc@`-jpOzHIVRY%{
ze|jwq*^Q;RbgWsP#oCpW;oNlE<xN_U{Ca|HmLiAK@#Ua`Vg?yr(Z9%cr1l1(S^GR$
zwK~Q($zD%++royTr+#S!B^47ACV`~cQjPuiH?DaxBsC!QaLPB8fKrwT1D)lT8J^R{
z$A3Sq4um9?bC@fKdTX+Xzix~oU_-cje`MB@y8M&6IKSLQGvnGG--%VHpmZ#C8i)aa
zH0x6{$jeadDi|&IzMa~caZn!%G;Id@@A@b<E_}GFMuCe12-l5*^h8Ep9I{gq=Eqa*
zMq(3z-YP^l61s^j%Nw541=lLppGp_II1#Vl0g>&ZC^NK>4N-8uE^?32g=l+;fA#zd
z(sK&z*t}0vykZIOGtV_Tft;(X=lCLNU`4u7r$sx5mq0}wX1(0gS_D)PS`ch{6|kgT
zn;ih)a>{W=mKGHR$pk}~&Kj(1{Wz<IFmmHeGfs`dIesL(if(&-_w+?NM*_WiGgS(U
zdzkM88fL7!c{f!_cqli9)tFzYe=$z%-bwAKO1`|4sv&)4`qqjH<+>z}8Kknb(QU7v
zm-8&h*+}PhcwxoDh1tB;;wKdewDUVdIo^e>&2+3GWmTq}>hgl}X4j59?kz+3FgZ2E
zpp?09-qhUFEMNh+k2WLKa>Y1B_%l+2C)G1*i*r4dEzhzuuHLVg9Sq(ge{^)?3Edr1
zxQ3!(L8}?Xw?%$s^lS(NUF%dD@|wr8ZJU!p!jv+m^xlu@@3k06_(xnb3G&?-NEg?-
zXIu%ji~+I{uEb2{iFa8Rs#I4ne-7^oUcv;m!TZ!2+R5ciT7oRCtbH7Bvn21CpO&+3
zwWOBXh9%wT1Du2IHxu8ie{_~7k0LKF;kE?V=dNn>C*8S`BO6y1Ma|Mu#h&D%r!uod
zw;5YVAyF=OR5xT|rzM-kb3j+vC1Q?7xbManr?VO_Q7}j)`yOeHbU7MHWI<exwV2#-
zVoC(2kZSdm&cu0%R0jEL*II=g`9(f^)luSaXY5PEu4a=wRcj4?e==HGR>H}BC)bsf
z=6WHPkkO8OCf$%+Gi_!}X;(RtLs70B3WYossdYaWwLf$yB<tSIg^%kDhbEqw>N{jB
z*%1phN=h{c7nfWfzQnru_<JlA)!N*M&ZwXaK5ebtuKIBw1x3kh+(ReF9S&3*)CcOb
z9cW8+BQU(vNw@bMf3`-}(~j4rhf{0M9JjO9?{zSmZJ!*f{V`cBG26^nJeeDpnl-t<
zh%<7Q_@~IAd4t)1Jq5m0mKEh{r&q1k7KYB%b)%ZJ6w*=Ocm()*eODq<nnSj1>MBMC
zITMFGO$)QK+Ch_*qQ!S=1AaKb6mWG${9J%TNEBR8iv#P0e_P7q&zD*=>T$tQo0|I8
zO0kwp2bGsbP3+x%eppy|eKlJ(z|5Tx`;HHb0L`#>I6f@AW76GpeE4b-7tRcU1L}XD
z<HN!QX1SUj-Lcg<FGpum%ArkL9p5=rhdj$+&I^yw?dNL?FVqCm-XzIuNnEJBE^ZRI
z&p3m!wt}}Xe=yQ7mk+kz;kOVmALjNBKZR%sZ~PLXZTz`Egov8~at5E4OzNFKgAg1V
z=G-0pSj_sj4yJBs|FtA8n;DV{sqW*&f@G#T*!qY14a8oy*THlip+?;2q_r2!RA*mR
zb;#OOXZw%UVQW*l|8f-ZP@QW(UWV0ft~P!G;SgxlCw;(jC~X#hDpG#((EseAJRhqA
zuZG$kE<aVho2Qwd>hAvsWzSK5m(lhb6ah4s!QeU*0yQ?5kxULIm-b2;pnu*I@BVoH
z=|&{+33n>)kIy&XZdwJ5P^0s9`1JKAd{9gyJ2_ZA61m~8e+{2+ewlm7J&VzWC;Wb>
zx^$xmT^fUX?DDri#PCmi{0WWOyWzjF>DSO)j9&ZUKf^!JzbFBB1i*zcT33LR9<{LO
z+>abSO@NaQaN;^92t2Msf)F`;!vK&=&#Hhbw>eCg3T>AsOd2A8oMAjBc#i4YF<mD8
zSH<OC2H%U-+<;H1k<kC!z^Ck}*sfZiE3AqeSdZzwNzb9}@j|R#tkVWo%p-YvBdpkr
z%DP2U+b0}MdUol(FJ|m9(c~GszYwbz>$HKDl%r(%Mp#MQ$)91BadM5_aKc0vSNYCd
zF%Ms?*9KmQjMDOd8{tK2)F&^-39izl@>Dw%)A3MDNmoFdFMA<mFIH^>GTfz8{eI=>
zUG8>4v}?QAk}PKlkkIK~w>fhVO~-BS5Jy+L!H-HxVV06AjTo=gDaP-!m)+gtCLX2?
zNMSN)O84*|#;=}`6{(Ovg@~k{(iK{OM4RcoHgTM{ayt!wXK#qJ7$KKaht%7gr)j*^
zF5Qu<X+5==r6C(y?WZ><I<-%-P`c~X!Y0}6%)uqev%c98;sd)d3fee>yYWH}e}zF@
z{POFEd03^Q(UEC2lo>0#EK?W+^*-uV=b5Qj3mb%CHN7I+{cd@k8a>G~c{Aqwgv`_Q
zKVolZ?DM#PxfxND=%S808PXPIa*IIIjFn3E5}neSy|9Dcp3n8J&|0%LCG==5)zriX
znQYfBVy(!IR=O3e&n#K`(KzZ`GB{0m9g_ZHnl(v^j7-K(%yTKz`xBZ|i+Kdnyo<Cp
z@eM6#u8Iz6iraKo^H>EO{Q6ibo%#yx9`jq7r9@wU+E2Qfs9WwS)K$?Z*usN^`}@M?
z&*Y5Uj^Az13hq=RfHdPGvo6Qdexo#0dIa`6Yw9obP`DGFr|&eN(bXKvK@^7~XW$8T
zaa=SsjdaSw%{QVZLyMjqx_E1N(S-^*lAQy!^gAUUPL48~Sa#IspkLyom4ke9k{K%H
z*n!V~a1?X6u|{nh8=Ke}iO}SYG^%1};DMAY+1-A<>k4R1%tLDdH^S1TM(;C*$*h4R
zz3(a@k&TmqI11^Wn#S?=F==4!A_=Mv;cJTD!RxNe2M)>ORYkS`;qIP{x&0XL_V!bs
zua%>-=Zxv@URrzmW9F|HiM<%%;A-N}eUZI?9+CgmB>(G*$bwPZG$Ow1a$$ieL28+c
zF@qVsWmjdpj4aoRAefogJZ^V=E|O57as&T{GB6E?=vvYI4PZ-oEBo9ujTU)$MftI>
zlQtM1d6iB^f{VRsd@P)9@D8;~T`mqJvf=WnYisNfgoSAhRjR&h9x1wP;1*p*3g#w%
zMKh{jy6S`wL#{#Khr5ehDluw_*=g7ACpsiV_84&1GVm$vxYTOV-OYV*oi)VZ^^_c9
z${?m}Xo_@oURHT{*W=}K7M}ClLz%%h`Qbu!p=Wt!&N*{*RdmU{sWQUMq1l|W)CISH
zhO}CbXWc-unhCNG3%jQb`k1AR+>X<Kh)ohz*&dZLqzWfJ7jZP^<MThxIGyrI<(ldu
zXI$>O(KuOJ7u{3Ebg_-3%)c9A;3=f{aAl0D#9kC)oJLt*Q}jKU^hg^N)g<pWLZ2Tc
zlC6s5&p$V!l7jDV*!+wr&2G!ZN-?Syg$)<WATbTIZ+#cb0M4B26f=QsSF4eK!8lb@
zQ=g&sVcDrB`LYq8z(v{Rq3XL@mTKfvb~qPi)l^?_z)Qu8DoV|jB-YxvdRB79V0Q%#
zA<l6tXi<4|%^tNa+SLrdbs0*M$O~wZ;p}kx?$MqKwN<C0x&@+nIJpHXT#$;cs|ZK)
zV##Va+t&_|9&5L3Fe1jnKiIf`-<Cse8c%&#C{A61oR*yc+0h_)H$7$-$%kFn>tKck
z)Jy6z=U?H;YLr^GJi~HQD6@_4^HeUB8EY(8EDM>Z#{{aVe(rlz%xtUl86%3NMRyVy
zB*<baOvtKvluW8#bz10|mpIB~00FV92<g!yPb>R|RGb|f>gt6oTEh8%LRJ~l+@#6B
z`>{cqQOPqlQuiT4W#Cn<`0ZBxhk-$MTakpa&#{%Wkjv~)xPaEzv_DS1p8DCY3d_mn
zoGXgXcT>&zuKJe>(OkkC*PTmn@T8~RnN$&-H9<K8jHf25j}YdV@EAN)LYO&?>AJ<?
zW>}?F(S)6<1X&$$6z>{;MN`G)cleQwX{F*mR7168joLvx9jPt*#vYd$p{5I~BNd~4
zpNsLR>G&!x4uYK_9g~}lv;zf~VRt2ZjKawXx)MFOiqWHpzCy21<%bQZiqXeDm$1XA
z{4$r7><N>~w<aWN^GB41KaJK#&R^=hC2CWAs;Dj+7l~@to0i>wrcr}Pa!a*o)D{@K
z$kKRqnNf>TdtEd7xM9@Dzb!3uS#ioHRQBhXQ+M4Zct#O1hbs}PS$ENEv&ROLSwVZP
za6zQ;uBM{x+2;AmVU4p*n=6ME>EDRr_?lr2vyEk=(WlMXX4qddtjrG8gTHbrF&Nku
zx@Ib^)kb<s^{z61qfj2jRBy?Q*77?amy;yR(v93J<>Unl3K@3#E8@%aLt;fC-76_e
z{q!21D%gBYwPlK0Aey;W%3erkxECMbM}d1D(%o{MiBu8FR_O>={&;RurX#wqv4tJD
zG>uLgOVdZ!fsv3kmFQ;C>=_2<E0VMq6%RX_X+t;dL#<|idNOQ$*1cSsR+c2!rIjV+
z6=`L^_z%^wJsop6U!9DF>>;mOrBamV&AnV?_+p(cpFGf~ecQC6*IgL!60_M=5|yrW
zMUSwM?Wxt{!mUBFZeEyvn9jOJy;Lfyj!C0@7DAP4igBgyF+K0EI@Lijht{uhs>3zQ
z1ZI27F0Vd+=5q_@Ym~K)yLk`zr02ZZCqgn5Hk@($(EM9G?TZl;*0hIxVdaiEANFlJ
zozDje5q7r251!&SclGAibhO8d<%3%SL!t&kB_rqM=e46PiRbDmFqEb2sOe0;pE``E
zqti*~3xwX37?rcMvz^ZS0bFcj+snjOsMcAxJ<g{>ZYb|1Oub5#tR;<mT+-6>X_Qyd
z4Kb2QzNT(<kIaR&+j)+@itIv-tob!%uX~IE$;6ZP3v(?`i{_45_kS((--(xDY#9`n
zNs1W^mkTxm8kaAB9t#08m(koTD1X~-+{h7q_gD1e=E2RrRNn%_fG^Qzfdnxu2S_3a
z#%M%q8B0TfLnU&8?6>c!o{Pw#v?52AtvuK&lEv!kQ&p#{n`5IPp@far2yHzJB#g;|
z7RDtOk|3iY5D8}j-*LjZtm8$10SH0F$jYRJR>nSq1dGbUNDM-2W%;3nM1L&V2q`+g
z2gzph4lhu)PaYCfa8C*C0y{{s;<FV8DQF(T<7w19tRqM^OHSw%3`<EEt=K#GYG4|Y
z$Lwsr=W)-bELmYz<NAmxTb3}S@g`u5fEjqtp86n6Q28m|h<P5t36pG=3?$E>1gL1k
z-Um2jbe8bIWXK|vZPtr;K!3qtjED(RV*et5@Sf+J!7*$;W5Y%w-byMz#1JV@DB`~q
zc^XL}OgR{^L{?|XLUMTs1aURp6a;w&JSi<OK2Hb_C{UK1pqSxHNjMc~HKfyd;>kmz
z#P_Ul7SVx(WFS({AQ?y@^1UFO*ZEyo2@H7Buo4*Dq+lg5Hc3M-5r6vvhrD7Bk~WZd
z_DNgejpGTzlqXFXLp$Leqm&SLpTm)if<z1z9)^sK_QFTQLl`T9@>y~sXqP3Vt!BKF
zF-Ty`?jc|SGsh5CiQqVh2_+D`W><hx@GjqjJ;o3c?4>*!${;GkV4#em@@yID8x5C4
zA|+5CS<(UplE*;3P=EFXxsDv;HyoZu&K8brj2R3h8N$RKp!fh6>!D(jg;9eD!I0%!
zkSMDmB#B&j_Tq&o-qh!Hb9gzMm%^gR-;3hitZmC?A<XLX57+IFhwZE_g<jn_s2A;v
z7sUY((JC@Cye|%bwl-u}7mf=Tz~+m@)u|$1ynI=_YwG!7*?%4t@7^91AIj_Y_~lFZ
zQGEDvSz_d~5)Liea?viC09N+im&^L9nU~8I;qa|Lm#5Y2b$xxLaOr_eNah&tH+WOn
z5X@>gw*2-e6Bobrx4-`bF-0)7SX`Z-uZI3=5FnzO(=1<TW*8mUPx!*DJU>9vXwJ)B
z*rtL*Y6wmo4}VVaYO$zShG-4eQ6_DkU?%Og5Lh_gx{+TK#p}8`Et@=oIxhZL{8YTb
z1!i*$kLK->AtfWT9m$BqGf~N0uZko^d;F|8ygF&w-k++)=i=4Y`n;{`MR8dC_5Dx$
z?}wk;_Hz0AqByJCpRZ2jyuK(tHRWkjeJ(#Y{5Il*z<<bpp8pOr=)g7B&uAqR!NyV2
zQ3kVzqwQ5&H`VNXS<LH2H7{n(N!2#9=F7*+rkqzxXy4&vZ-m=D)~k!`v2|K1RE8#!
z7Gyyx$pBi=4l@bvxOgRC{;>G5{!kZh#ShwA1(FJ&4)3FK$%P0vI>nK;#eub0@}k$T
zkfBz97Jt?N?Ip~O(6lznF&1@*IU0{posU9Llm<0OVMGl^r~5VN%uNlFzop_PCW%2(
zXsiKbqV(uzELn+wAq+^LVTynt4blVs3i2apkF~LvA{4&?e?EW<T^z*0`F-Nx)lD26
zttrsChX+Ilk{-Q^(?OylxtWv!y_z$qI5Uih(tjZ+G=o^&9h!w`RM<x30ZuRrb3*~j
z;Qlmd<b%l+LVn-`x;<ymQ2{cCr0}GCFDqQQhd3Umczf$CI!U8eYt-HA?9_DfZjesi
zj!&l>ggia%y75E)@?o!u#xZM1Piw{a#s-P@`TvIT0}T*c_fr!Q3-;Tl#!llos)#0o
zGk>Jv;P(ROdoat?*~VciM6=JL6XVUI6V-ariTWXZw{uQn!@cdhop%z=)zj;{H;!)l
z?#Yhs_TA_8S+#7d`NyU_Ys%%I@8<N1zfpU;3LB#GN=bB6BQ(h?>9EA#uc@_@Xs*63
z`SmNm#|ho#lMlR<+NdFu_=I8gB(sITg?~Ewn_g_HjNXe!k{(u8+J{Gy8nZ0*v^J4R
z(%MJai1b0~qI@^7V`Fv$`A*C_88A5*LsC!s^wxZmxTU$Y>8;aO=&g;_(qW--(M!$k
ztuoT{iDPo;u?KDKvdkPJ4Yo+P^wjrr7g3UY1kQias0*i@RM;et!)6N*Od?ru?0=`3
zJgH`jOSvf9;sgX%AaJ%=TwRn+HJ_b7uP@u`qWVwSET4a>o97#fdwxT4MW5hOxQF0U
zxFzLDTJIhs<vab>XlCjqcCTi7`QIs%iHV)mVEyL9>FF%P$K2njTNwr9WE|oUhcN}7
z^eztfWL;NblXWf`nJj(BP2*SE8GmfjvkRl10OJi$nj4<v?;M#Y`7_5_ppD=j(jZ!J
zRRWK14`w~M>j|FAZv5Py%TDxjw>>PcFF~poK>p)txvb6>dwZx&l#V}IX^oDRKOP!M
zSi-GQ*wKdRP@dM6tedk!D_&JX@#ghkQi#fF9TgGjuW)v}St)f0((zZM^?ypKW5Aec
z+)~Mvtiy=Ls;{+fqpg=VjL{_cW+;5m=J@X3_U^WATyypSo-xB^%b0!W+T61X>MC@1
zLBaW_5|myX`Vr~|g@3Bx*9Y@E5$Xn|e|&h;K2MHN8b&DW6`?VwYJIKss_MO@-WK$6
z!qn?W?v7ax|31pyX!as=cYicHL7y5KJFFeMSMAvQYR3+12S8yvlD=c@+zowyw?8fp
zov2QZ!>~f^UKL_TRfvtZ27`a%@kEYZMrHeM%<kl_UcEhFsac!ohKbxhEv@Ucw6tky
zY17iKet@v|Q74h)A2)4}Ivsy$3s3N<bM5H%sB@~LyS?m1wfI<`pI%p&%l(}r)C2#H
z$(bP0Y6@pbhD7-5r6)5T#ebJ4a@o!npD?g2Pp^vq{_?+*`paoC|JU_NU4QPMRHW6%
z?%yENj>5KfW0L$WX@sVBfZC7HMD=0b8^vD%zG-Pom(lhb6qkFy9SoPeO)3bN-(4Ea
ze`V73{I9paKS>h*r&k+JLD=XWtjX@Y5h&p#JIZR1iXA=&O82yP)(U5C+Q(<F@zbCt
z7G<|?q$XdJRzCeD(GChOsFhXed)9a0dKvDFoSi;J5uJ9X;OvzM+7+BX{rKDzHh_>W
za$yG{Q3Q+Dv>^#F35ma%SGuk=!j1Rsf8-!}$pR;3pSAjWS9tWc2w`1z?^mDjT0krf
z{psh=U$pn1TvKZu@MZ+AwY<*Vku)1zZs}8!K0fP37Wg*0YS8g!Co|Zir7<`!ka;2g
z|443}lukqs!4^+=5K0<+BnoZE$@%H+bCZdW0+?CPZ6?8?#WX%*gfhhB&o+}Ne}!VR
zXi{a}c@Tq;_^0BLzoGNA{-m4CjKV3ance=~RLLkWbo6U>TMj|$COo3G(IBjcEm|8;
z!UV2#lZnA63}f<Je#EXng1Dx2`N%&J9NVIyZ6u#5vbl_mLyS}l@KE1Iq(vSHY}3bg
z_S?(b(|?~hVB(TNRm17GA5ZBIf5}!5^hrw)8U_37Kd0ZH{=v!4qft;OrF1Ax0dWd}
z#p0rYG;ds%GUM-n#R=5h2y;#jv8M`i&+}QMG>VQ!aatqeTeP59BMc6ke*E>viQW0-
z`;8K4p%QkNl+?GIJxNt5aHUBh6Af)gs>PjOQgn-Q)fQ-C2bOb5F*8TEe+kwuEM8Wu
z>l`lqg$3*jSzw2(9&Mx+7SVC^iq$un72Z?+A$jpBq0%}psXDptSIl6P3{zON{eCSe
zy_a80E)XeT=ee$jaJ<Q{CB-BCT2emBuO-!*U(;w4C%?LWzm}BV%daJu2W>?xtQ?sQ
zoQpeieaJF4M6}~A5sZ`of2!Lo8qi_LQ8&#bMp-u!NqOnQP%VuxR&h=T`$Q|9Vh>}o
z141#lVN&2Z#RHa_Hd^3bYGL}FOAYhxlwUBa=({0%c2f62rAB%fr{CFoDeuIU#NF4A
zlrW@$_8r^@X-&ex-UI7}Q+J%V9S;Hb%T9-P_8#DNRe?X)BqptBf3=WZUb6>13Z~$s
zCKN61%u&=B+*qEh516v*NxZz1bhEv(9n#IYc((u<L_|9<Dg&Z8LYFX7OiD{CsZ8Q-
zSaBLR7j7Z1jOkjHO@kRmat&W^HG?c$l2en6=1li(-cNU<L&;dFbul9dA5GPS>AH?v
z`uVm#19w7!g<0>%f6{ssA7=&b_ciVpGeFoG<2)lC45?|cd9|QO*0<-G_^)|OlWad<
z;>W0#(+2AVC^H#4rwlt1656&|V+ACpP0IH+e-=U7+{$@(s#u$oL4&|y7F@JkU*8FV
zE%4FsvYM(|1SjLeqiGYoY?nSKXeFniXo{9rIB!&-;%sF8f84+&S`iH_xX5mbIpyaR
zf#gnw-wVm@)?q?h;k8*%*lxoK7mw2}XN8-!+r-5oK)TT6l7s2F4Yef&QzR4}vl7zE
zRk|vfCEe4fU6l*~F$osBp;Vh?NdRu@zDO(ON>;kA%YvI;Opz*9+AL2_gBtQ&@~e+k
zd5jFgxy5Rve+u|24oM}fiXVYPC8r68JbfSVXg>}yE(hQcqiAQ|IOnU4LyW=>96C<B
zj6<8X>%t+%B@&nO><VohDk&U?Lx))ulue0Bb)H+gE^SmRpWcIN$7z=GAi1<v#wDOk
z!d7u<+D5K=aXU_W0l1;7$>T_tYjsvlTSI2;I!>C1f9b6I0GKQpO|$1r`kBQH>Oo1y
zBpdER41Cp*d27iCbRe3JB4UKvEXgJ&HVSWJ%cddcT<huvEAV0j_Jfbq0_4ICJ65Q4
z!dhRmrl;?|p2&a}G8jgn3w@A4gqq3ohQm~ocS!3%#dVW!FJA=_LUq1d`d$y=Ce@xL
zm!^die?AJF7asu>C8tTLD=Fj`8FUv?FfRM^shZs9nL}(dpW+ny^Qi;1qkL++c0KtN
z=Q3YjK3~mzicvT$pQ`B(q<^f&<eUL93Oh2Onk<3juyqikecig1diQlTc`U`CSe^NK
zU&n3Jz#*I|&cH|DP|0c1D$=aK9+R;!E(hRHe?{(`cG&tF$~eR*?7*Spw97cOS-UPA
zYWO}^u}m9>N(#r}&|wzM`?{L0o>ni^cRx8NDk*d!+qe@uP&*3(>{2^3!)=ojw+dEL
zj<xE^%^F69RO;5wd%H_@b*7d-wM4)cR;J2*Fdu=HC8uIlW~|&_i|1Spz{-Z)z8Og;
zf2?E_c3|ak+GVWVtX&sYGA?$8mASHe2P;bo$6@7R7M-xNV$u2Pjcu$fDeS?@iY(y{
zs!;~GYV+<4up*16wST{|&+W2;YV>Z@7>6s*&Ly8#-CD2hy@L-%&9}gCezxIjKFu&R
z*2meK_*o-EzPEbrf$ltZ{9t&U?q)3sf9G|Ip3ah(*Rg$n==$-Q()<uWdR=Z_o|U84
z{@l>e+v}_$3N3;6Ix8D-Lg;p{^CNH3hkoeFD|>Q#Dvm~DkL>iL*WiruV8hp&b&YQO
zF@;Y4%&d)}-I?7}^<(m`^el0RW=}K|f$M520p5uiZgt{jt42o}p+uB^!KKoWe?T3%
zZ1j<qru2YusCDm9A7|A@`e5l00AqBIN@rE-Qsc(5?<E|?^aq8b_C_w9Tc3oOZAzUx
zP(b#yK|2I4#VA!L+Smxi22Tg9)D2hVWxFVO7MIi&RObBHLd8aRgzR*LPWgpai{jme
z)+a<4gE5xdDPiJucZ$w6e%V1We+VyEEIxnT*G%^LlP?(o^XZ%+u0sd1TQsb>GcC6z
z=jp5y^ynItkg%L-x%m#Ugl2&y8HzhAJ^EnnaA-4Aa#L#WC}PSn!=Y%h+MGE-RlP-%
zMvVQ?^Tt6-Csb`LnpsDFaFb^`Mp{r9;`=47wXq8mEx~DPlEKd;H?tYxf2sgD*$AOV
z-I}p4W7=V_9ld=cW)<{yOeB@+C^3`-1DIy3AmeVj>k1_ycx_{2F}Mqo$4=(Qbj>)*
zxb_{pO4XqdII;77^ehQh(!>mm3srhfwa}aNhg_H6mY0Kv<fNlLoQ{u_#FT8NMI~d`
z$2fy+$yn75j_Ho~+&jZYf6v{ZLtbVO<GQOLT&nf502MeEQvimu7T0;Yf7_~d*DfBD
ziAuN)nXcUjKbF3#OqAS|vm0G<1^6L!fmHI`kE8iD|3iy0HxD%yqlfVz$?#*~T-=#+
zPB5Ss0CXnDo!ik65L@j{Ubeg5(elnH*P?O`dgbWFF5leS{oew4e^()M!cr!+X=kbB
z=#D)?E=hB~6*QxAGaG#?;~rs&T*+V(RT?Vj$ln2$$l35?mY}w(6-$hJ1)cL-GcK%g
z<Gme9=nNm@vq<wE-^1lTDL&)XtPXZ5DwOLzPHHF@F73YOq~7=%?w+pO+bkdR=>*$L
z%M5`!PP>Am4kv-uf0(=J4F|7!4l04d1PEZ~L8Wxe4ReQegzP@0&Zt()GvjG<Hu{KY
zLG?&RxfU&Gv(mK17%9q%!R+7hY{sSZ|6i1S&{8=vu+be!OitLjWg^uS)~i=JGw1v}
z@OVtyH?pVvoU=b@+2*ByKn1hT&baL#SkU6<RPI$~cgUS5F*)OS-F&X|VVS4JxX{`q
zNcJE73g%e<TPF;Jy?|=0Fg`W`b@eS9`+RwO`afVV-Q1Va_8JwJlL$W(mv3nr!GB~P
z`1tqJ?>k-8ug|YfyIyT%R95rw&-3nOryPXze*7ZSb@R(kgr~{+l$T$Adqy4}{@eXI
zuA9Eqv-b6{{p_>^Df_`I4$hi2ND*00fC>aG0Fr=1k20K?O%Wg#Xi1<ZKh6j&jB!nz
zjq4K}uelJlRZdeqF}fm3;giTkdw(eB`gr2DJmZuF>gDOvE4zYn&h)!Dx)(W~-XL(?
zN7^J5(%S}wr0o&~nRtwE6GqVvo29W>N~4uaNo30m)wNk7Xwf2aF&z_*OoudyF3yFr
zfs4wNO&CG61r85ua2SJ%&@Y@Kly|fVq#;s7(C<WDc3B*P3KlIjyVxjhkAM60$LGAm
z`N68-jOS*D1!Xu42wEA;JMi<SX4Wg?7dDNsx@|-}r?^}}gd$D1Q6VjO+8aS3NopJx
zsO-V`PcNIM5|RSHU~^1(lw#DJjvSxDvQ0UJ${4sYmMBtSfmCAN<4AKpA|6U;*b?fY
zaFeeGL!99niqmXidpx7OiGK?ADFy;5Cz{pZ&T+HlbpWRVhJL3p+sP!RPR4%M8<_?a
zV{f7CHo<^`vSikd6cfQ!Fe|*%Ooca<J1(wBKo)lhYf?EAg2vJkAK$uPQrcb?Xj`Wc
zT`ppJAS3bW#VlaO)(yY@9e%`W)k>XL{Dh9Smbvwa7VBvc`xTfeQGc%&9j!>VjUT^$
z4&BUOQ=tT7MMLN%mmuxSDhSDa=og!{tbdMeLTNJW?P__Wz%5RyUg_2HPLWx_6WbRy
z#p9O@S9(5FPRPWi=q5ddR;=?!(yf`9>sFm1$rv1u`9Z3t)tK^3W>@}|$nVQ=aTaPw
z<G7|)#&H+X_^xQ5Du3X)W%RV7^0(SGhFvQ)l%&IRJyJAWa@{ZMqLf;Fbw;sDkM+Eq
zi}PqIDe6>ZH5kldE}xq^3oMpH;m*_-G_B-4zF}DdcIlhS&onNDA9k);v><5)G3Tew
zYxGG;+pZ*Bnq={1T*VFIn|4M<*Rjs*YL9ek$ZqP;YFC&GaewlYn+iqZODPzYFEJvo
z^k+E&q!H8-KKUb|+|0~S76}T1>$TEcw5ph5(Waint2UzXiu_XA?a2&DdM3%Kouqg@
z$-SeDu&#q8rNzizhf7;ht2fXqlpf`ZH)@YN#v|l+S3Pj$6e+{Tgn6=`#U9*|>Loj(
z?d4ug^;#~M+<&j0+eD`+vh{f@w(?V7ss(VE-2@&ce#ZQi=D1X4_g2X_qhX{uHQMy{
zU5P^@Z5<0QdhF;fo*D?!zQf{dT0a}*n&1|m7;TomA_XWYUCwQ0*0Q6HMSnEHymoWC
zLXe=OALxW_WSQ)D1Mx41ueY-|fU<ItX^-n5yG<ag5PzG;R^Y>WeyAC*S)3XQeS4Y$
zjGbTS1z3dQO){nXcG!CZ%(^uw7cbXi`3^z}l!W*}gldJoN~m%_zmrs%1MQLTAXVi_
zC-54n<jTm<Elz#M9koxg0|I}bDF||V4{tQ3&~Ir9>E2iATHxNrH!r}&Kkz<;Z>_Mm
z@h$)RaV8h%Dn;63-^IC1qfX#8&Uxju?G}%Yx~$!LyT5d2f9?JcW}lM0m(lhb6qn0g
z8Vr})1sh+tOm7;RhnKNUDiZ-Umywn`Czs$ZLIsy5_ZpdDMW3f%-hX`JQGCw4p|AAO
z|9X3W`tzxujMkDV=ZAlNf1(FU0+UV*APm#O41fJL{QmU!o@{2u@g1h=4}SaW+tA(l
z8u9cT@#My6-QWovK~{q?o=JK%;;B668WGUCnU^5=8f1U3sBTOVBI!>6s=0$eWiAGo
z=XSz{H=yLv7T*Wck6xbK&GMjXG4K<40A}S8Trs6#86$CL>6BsKo4wL$U_wv8(vK3(
zOJ=>=E1Y)>t$8AynO8z(nUXe*k}tO_nl$LUpr^E?Xi72kdp+B?$QDtJX0IdL=FZnC
zA>^*C%l?1Ikf~-P%04aHaI{@`O{+vtX_WG-(hAKq_}a-)(6H*bRkjpi!z!E65-5zl
zjcTR%?J{@_q|Qxd^}1k*(Qez(!4fa0XK>t9kSIc19WXz^Fih`+C7It{Q4v;P=oNcW
zalnveKRUR8OHx1mTEa$9Lo)(JVWl{V@T-duQSE;~F>QLSI|wlt!@!7+=%V5&W-PRW
zKdyH%(6-ZhT@L`w;E3<dYYT!>sE+Y@p0NO?%oP0FnWAb~q1FMvji6^9U0p{|9VaLq
zCZTqpO5_N6Pn8hHF}!xFgf-0CUR-6OBN8D&|A2)^j>)&S`_v?vD3=V)I1S#mcqkVC
ztOS3kkBOW!gI0{CB}|z?JGu#FHwR$t1v6;X9|*vDK9NZ&igpyQR9^6;X2zmSAP2}r
zLv!cr3?+=GR^$|4$WeVp8OGHj=~9|M77=uB)k$_#!+P_WTG)~Zmh~6cLOVQFBTtH{
zAzeJj`#Q;Tz_1kKh(sq_mog$^KGj+{XvBZ8zOzzzCDZLVNW$<E5lCfI!t_!qnT?_>
z+2Rbg^5swrkU}zT^#qaYBvrkMDkZ;if&$1{jgbnC$WK<K<N~1;L5&SNHY}26rec#{
z?o%|!7J$~CS!4Q=SJ}1ZMqat~c8x~7tt?LFe9|uPptg&Rjyq^R#7hKAi%pqeIhTLX
zju5$x5bcv}L|+n}dPFF)@OLV}av=_hc6ftmyd+s%Dg#JEUpr1DfzL(Tmn-N2gRnvK
zYHYz;^iZasbK)F}sxMazUUP4I%JN<F_NRZM^?Y9C=@k=R^gPM8P$SFuRi2VNa+FIf
zF;AR(3DYQ?QTx_UpI7++YM9d#MnHd8d~;rpB>HSEWj{Z%mAHz*KJp5l>-8+2yXZ&q
z$2zWW(Zs1f)KMXNt;DC+$I>zzpYcfjTD6HTg%ZTP=GJDR!qUV$(Kib*dm|&>=oM~9
z$<QlY=#3>u!GVD_b>IwUp(@9&q6u9~k7JTEd4^d-Pi;G327?zdFjk`iz+!)No7pYi
zkQT(*;W{z_?p|quMrWX9Pj!2iA&=fMQ+PXEUw1NfwL_ok_oLr(0_%&LEX|NHJmvK4
zcz8-#ZO8wHo^DDXItrniPhI{4=>s}4DW(X{r>-g4=ryZaT~V!UAL>IArJ^%QHs-2Q
z;_@+lw;q!_-&pg$065ik$)taGJ;fcmo{E8&re}JKozEvF^N}liQYD#)iPxt;K0ky1
z^zU!0VF=4gEV^a$CP^wzYmpKJ)kd!6_v>%}S!aa~9RuBsoY1n-URzE-t@BA>2#x%T
zaikpVJdLunG+1J1(_hJ?WEXMI)wT&g9s{M*m=4%J22x??^?D545#N6lNB?L`886~Q
zk_GVi8Lfk>6&>$dd69$kz#QZ<xxgBCn7UL!#pXfUeagkN9?Pqi`>Eb&TQv~CPpCd;
zu?8zcI@@nfI>&ecJCP`<h70^0exF%C?Lz5;G=Khu55MBWA8{}_kN<|D-!mUR{4xB4
zR&#?eN)rfcbk%x|Fdu&?fAW!RA)%!Y{g9hU?37F?rj%JHi|XNZPj^*gsrN0BxW%+s
zRV}krLe`q)Q#&g8U1-LIJ@K@)Sf<h_MF#2cY&KAmGBhFYMK{ZPXS!KlLHh*mL92=`
z*@0PDRa?pmJ?<{=Te-s$Zy<p}Yvb>Y1hpivxB-lyx@-{YQqX^=5{nA@SRbVPsVenj
zbG0aZN_IElIy|inxHx8J@F29f4EaVv%T{e@UNsvhMQP6H=k+T=SM*R+r)@4{9!G6U
zaoq{-Q4rR#70zSL^&oJCN8c}fT|{J6j6EzByAM^@JM94Q-07}zrFdu7RaZIE2AIcZ
zy@956%y{u2G#!7iqN${wb1+aw#q~}Y4vV`1gP{;B?;e@%O*hM{X@5*VCNg2FfDgqq
zGKs2<o0_)D*5{-eAG4VACVC;WUG?@np25qB?v{QW#$^L<H3sX)m<jdDxUBc+Ql~A|
z0_wWuy`_QHeG5Mxk_<;1O*{yx$HoNRPeu0FBpGW$X?K6Zd>D=mnB|xWJO*a0xv3?Y
zS|I?NKxDtRG+nmgRXE-MAH0WQ*nn4zsl#L7#rhmv&S&Lzm#OYBs^Tl?38L!SjE?#V
zw3|%5bs4{tbHEoM9?j%$@tNES^I<qPV1_Y?cnr*77(%_F#k;^43*h&O;3D>Z{C)`g
z!x(K~&qK;__c(rkU$RMBI-bV`JSL2TS4?t0Q(&IUNe%V3Tv%0w#m@5c_<S~Sa$`pF
z5R0N?<+450&}69DsOk=(YjPM<?U9s&l65^At%oe<5<lhWz39bYVRC=ndK`}poGs#H
z<wqK9H<Q@cQv_?6v+77sv2R`mmC$aWCZ#oUu3e7n#jxUkTJK0?v)F3Bg8+;(vK*&9
zNgYbZxRq>6=dN$WM;%Nx{W-VZr_K0Df4Ux9ZJy8dyH#~|-KGN<Gh6LB|F5^mMSWyL
zw4KN~Y3*Wt5Vsvf;$X{8U`=VH9rCd-Ym=#pe7bvS4@HF@eX}GRTPO;Lf-wCmmO5yi
zA8qt_)iV5lbDg1*47~3d9^FlG0*edrpv@HTnXHcX%i9kM(_SkYVM}P~9}42xzIL!q
zJE(fpkKmD~Y0b@&#RK^*vxQCIbUv^h-0zn!>=!vy5u2UOk@MaW#Fo$oi0$AbOBeqc
zq$6})J44$|^$x2PkGi*n7guRq(spor*nVSEt^Y&^llvdd>v33@(e@e?0XLVynFA96
zF_(cdI46I#SWRyuDG<H;SMUcys`y5VEX9r;?c8QJx6zz-cjmIuDs$PtzYiL4n{LPM
z&Ws{VHx!TJb@3=d;=3U6yxF~M?vL6CFN0;T#LG9q=pg(uq0w34nWfV#US7qQ2O{NA
zb|{8OYZQ;?fZA^FGzVDlBMNWtyy0T66$;~dpWc6~T^?<RUiFv}bbKhtgpw>hlTMgG
zVYIW~etr40q2fNH=23lk*}QG~1vqV_4o>{}x=9^IGhUT2Ryq~MkH5vY&F6+tkx7R1
zgx}{mNbh7&K0$u@O5!hceuA-%LHw6ZU)wY!{t<s*f@{D*QU=x_JH7`@Ldh6nfjvQ3
z9VLI6xE$g}&zzDjypPHxqwQgHhy_N7k#p9bg&0ChSq*F}Bfv|DF@gfoLIz=FjzewB
zyG4if+dD@oxO6O5F14Z{lq^i`0bvw}(W%Xtp`T+4AtIO)KMkQYTJ>4@G(OiZ{0g6z
zrC0p33ANIa2e_3)u@Rb<tJAJ5J86HSB9nh@%C3#2FO>IMy4`n`vej$5!kHoIo%%(b
z86(NgQl>{VRr>shwq2pDmRRg!>DzBP>*9^<tzP97_OzGY>knzmQivPn6tLEtHFDbP
z_6-M`BO^_0YNO)TE7WE`s7um{tU5_M&_>5}VEKolE=6&Y%491uShdlQ^>|BqV|IVZ
zy0y5XMhQ6-i+#c6YBeCWIj1Q4q-#kDOY37lQ`W`sjyE*&dSBl2+M6e7ZFZ|t@Mbhq
zy{bx_bhr89=2&PNek{Fu#6#lEkCZsN93-foySLec9T+mO>JN4g@j<>)4)zZ97Ciu}
zHNpm#A?O+=`i%Cph~l)tu=exgyXb#fet$mst(9cudSv8E=Q&Nac32Mg*0%G5@R4Yj
z5!y%veh~ZUmQk6DBaic=Yy?+L2=$??H=GdCN2OijWH|@UNp{m&j`JLNBc10tpNu<&
z9emOGOy4i|M?oAtxqb$wS$`CJso3_`Hf5}b<#Q{GW716KHuG~USozRfr_X<%Bx*>A
z#+&dXtav4@Qn#{#2t|JXvsejH+9@e^e(I>Dk7kY(LyM=@q@*T2nVV~w^+6+nfQs~^
z8ts%OvgwD){AVqg|Na<hPKvbBI2R}3W@62B&-}S0Xd3y1js2<+8Xx!1M9DM>lX8NM
zeY@o=T=d`ZAotx@(Tis|AwqwII8G3s)+0h{wVy`qID-u9(St`a#8>=n))5zWGmXdH
z^8FBTFT1i6Z;79Zhixg@PUSNlEBn>#7)ZEtlU+-P^@xldmcxRt*+wRj)-zf#Xc8ah
zk6i2{TuR&iS(93nWZ@i1(P|DD*CC|w9pHtK<0J&eOczoT<&s|3#4mrJP8YHdjmwJh
z4WXpi#@1AFIW<=2n?#2^q2~55W(ZQmjj=<_|C&&vRQr?+&vKTwX_)xc^HQ@>@IxU?
z4wt#)w$@0RCwCtsZN8VkZKQ_+rr!3D^XOsd+s38U<FD4}y#MVL+3VB4qwCk_$)vUZ
zAIrNI6$)i;WOH<KWnq`v_8JTUIk(XT8_}(oatS&Um$kDS&I07lw}rGDMhXN}!7K}x
z(e@e?m%)Vv6qonD8^r=4Jh#5T8$SmGAoT9%m(lhb6qm8lIuw^F#~U{SA!nCj#~UMm
z&6|o6KZO!scE_jPr?Wxfq$hA%JpB;BHxz$@;!R7!0Z3~=JUxqVpOw;<*O3=0WehLJ
z-#uSx|6stg1RQw5{=p#>d82s&f1l#-Dwm$7d{-Y5f(f4svf>pb9Vinq83iz6@Be-J
zYo`kRdVPG_Eop1g(kc?)pLZ`i=^&(k`|S&F*N0zr44yXYm2ZCf`mCKe{lELW$<5!Y
zu7|4Zq0^RJN?v$L!CABAQmskO4m~MFzL;#-WQ@+iz`N`iVh8bGlE5=Bb^@`>aVqyA
zIpnP%YGtKUSarr=q^A!x7zx`)?8|t@Ng`O{slyvj1(puWrzoX@i`uDJI29&;uJMZY
zLcf!&KSs87XXU{xr?QRcvSkrBrX^~O9yshN;rolswMB_n+pDdWcJ)v)?A%PYB?OVC
z$p$3TDj};ykijy89-1S^iDm}uKTdyMVq-KHu*Mag*|QfF8ZUj&qr&f(6(UI+!j?h|
z`(!<ppb%AmtiwvhL(y*(vOMH}bXUwX)?&PTS^VN&@UrR#)=SMUNL8-{8j!|o0oqE^
zR{Z);{D|c}qV_G)0~@NWB^vUYgW(AUc$CVAme;GyE2NPhYahRUip9=9b7ui#nISBZ
zwHn%AlSbqiErrcS0jOi_tEDU3>uGr-!L2mas4QuDC(*3j6Z?J2_}a#QRJopNPVmN5
zxM^Y%r1D4NHth7=YA_sgG_6v($5^FP#pmS8zf$J=vRvGSl7yt2jL%6pKZV%t74mq<
zY1=3CEK$?9n$>D<L~Xsk=aGtG%3ZR^MU-PfDQ%;ot8>$<`(i2+<WR8&A9N*`^`<_o
zi{(&KXDZ(pqxU1Bv6|C=2iCEv>6zA4(!=)5q6K;0#qOKtREtSzrmoBooUv$XQZ#t{
zqMb3J>)KB2Y7c?bpyzp3XDDCwJgd{9k}uS^`dpA(lS<W|;S9SI%183@u+w>&5hMh+
zWTmLEsy;2Jil&J+ZPeWN(XY|2?`0aT4BS!|NbxwAS~`P5*yg~0(^s*POAfg?tfGN2
z1I=sENWZ799wFa*H3F|_(Lw4!eXs9o4?fW8xh8l9EpT7$^gR9ztytKxh1NFT+NgCo
zT|lbBXV}@QRq)an+xVC=gt5zIEOm8m$CfqUAbC11(Jo|tt#`J>JYMh3z-Jxrm-+w~
zz3Uo8k^#*M@wA<Pv9$(A>C{)$ZSmlfjB)N>@fgwGj)%E19;3c{#AAW@Mm$F8G(|qp
z$w0@m;r)!!;JC5TdyV0PJle2j45}emrkwKJO(_?wMDTaYxDo6f4dAx+<%!?h!kxH}
zVDI35o8KKu>Y2A!MJO*tZaz)cQ%iYoUs*s-TAqnL<U5L0{jooG(&_p_QJ&xY?5Osz
zoz%IedP#WfRM*pcoNVd?XAdnM!st?1jn}IniTHRsMhVQeZ4{*A#2MVs4hwyi>Mq1}
zfUBv?qOCX6jsJrE1_gQjl$X)=8WaLJHkZ)_8)~;w%o`!QlhQ*Im;MF_D3`8>929>e
z*;Zz*OXBY4s!J|S-XhoJ?`Hu5APG_;MN=tgO4M=zJPYhTyMT4~uTFP=dAWSO{B$Mm
zobm=7c3;1DAU|mN3tApp(>VZT3A(Qz-7jxiqm}C-7i&!<my^$4?&kDL;6(z?a>41<
zF=Tl{xj;U@$L~chy>0qm@tibJc*}p9wOmosfu^Waw17N^%jwszpI>zDzE;7;kMY)i
z`}}(O&r3TP<BVeMy5D}h#2=VIkw^KNmDRrc<DcE{FMn;-*2b)pXFwe&t-C*>z~sr_
zGM*j;Pi?I+6Ffl+$lO^Q6vdx1o-$OH(HTL(2u3D;7EFyG!!lxi&(Gat<W+yGampHM
zwdx+HDupt9F|RdVkqjzJNfW7EmiAa|$X9uAU4E0DVshn8`iwp6dlZMRm5F}}ER!b!
z%TL#!7cpUhkW8z+A9G`Aj_S&v)XL;ZYUKv<hGf*fZ3Xz@7G|h78b6k8WI+X)bd7XG
zUkrQ`ak7e?uZY`%dVS-e9&;2>z}#xR)#HNLRGRhU?G~h_n)G>>Q;8h34$Dod@oF_>
zIuV<J&-etF6^a}r9{uqJY`v1JQ3!CuZ)7hsZQW#}VUHwcoYUu#LF2QRSBe}k7cS*#
znpUoktL?K1s4P=;BM~qf^-<@0m!66oAAfmoTTdoyI<TC_fZSnpUrdz9lDtK;M9GV<
z52hqaMaAGAeQM9k=Y`~py)fIT@|RzA_Ye8)XOWH8{cn=|QJc}-zq`N3L~KQ`bSCLF
zf-BxrtUDo2K4i6n3PT~qu7_1@?3vIUhjd@9%#L7My>}DI$X!U3OCtcI=$t1xjejRr
z_M^s>g#i<}l%AJ}ISgeoCn`w0N3(*GgkNX*Fuk?E2kWi<6&CKbd6HGZ$_k?!R(070
zxa=0TT5flbSCG)4Jlms@2x2xA<wjx*k0V@~$=M~Q+3m`#uB63Rg7V4;Jelx?dac5>
zdsz43;vm1=_kjE1;=q!h)(Q|60)Jpeu?vRjGenvKp;?|_1-oG);;8cX-bdP4v%~8)
zboZmMg02fHpnnp&lKmr0$X=a{Hgn^1VlibxhKzm0SdaM!Uy2;IEFcUvD_P?zn!886
zf~IiN@T{pS6c$Y_gvI%}1Vf}$+H8bjZ+BNvFnx?YIy4=vxAvC>c7x9eO@Bh)C7})T
z(1dvfB>$3q3y(a>uqy6z-l8ut+q>>w=ly_6bl0ZGZd_LIraoFfXPHRn$yb6GeQrYK
zeaHp)y5_yf!qk0Z=r?TAHjB-hGA6q3{y7)-j@i9KboCoy-VMhJ%-B<b=fI3!F%HS9
z8#3c{c-e#f|G~Q(h81|Br+*I5fft-&TL~?#*8vLTp%@_>M1B!e{gl$hs*yklA^?;(
zAM+K}$8tt`zL6>Khhha{-Sda%Kpg9)npbA<CgAY~@GS|fxVwzEXo1jQyKvu)&wbn}
z^i<>M0DiRI+FviRHu&cRaPq8-w+p%DIfvjlKR$2gRHKw)qh0rpXMY7J*K?7ltS54$
zEV}Ww(3G3jH8m~{`U!opqx$vx=~evGG)^RkRThxXckZvr0@|3f?gwNAXzLUi_)c-{
zVHMlTiBT=wS9`0T*w!u6oG`xg$^;i<)%+NY#BAll{bEF27Hel}2vGVtIBY(VNjc*w
zos1b4T0AS`ma@c>&3~WEX_|3nnV-xm(_>kK_1jXVJG=k%Jxae<>GQY^uFSvwZcd1A
zqY#7BBIHtU-ObEE7S7v{9rTiUr)I3F*PWPH8nvxjF`?SqL{_%s$`&n;ZNjNKKQYOl
zKK^{qij1JVgVu^e26VqYSrs4)b<*H(_ZwtVJexJOLh#@jD1Y>%9Qavj->N|N{t>7R
z$D88TaR|flCinP33-Q7vi?G$Cmp6#=ahXPXfy3-QGSg_ju`O3Nsv7BG@AA7lGAPW{
zaci}>Z-HUoevI%J1s;yykP)sS7|?fO#=|N$_DoiiK0cA`=@qh7G$fKBt|)|bsHNSL
zbfZi@l7-JcZ+}da6$hC_s9h>{Y_13HAK?megi1dvM?^Ad$rt1(b0Zohb>)8Pk&UPV
z_J>1t^9tv?$)C^L^N{_+T!Bww1{|ON#e)b=l(YD*Auw#iBPP@CZ9(*$8EMF`Y^+Y~
zAL0sh;FWm!Z+VkPRcv&i;MQvITN&uRC_ZmEo@!*nPk##GVtnO>6W?rihww(zN(gVI
zcTk6T<jO;<iLT4-GAproBYXsN$VXHg&9|6?WNs|$rjbrr<RQ!fv{CHrk<<ZLW<EUO
zEJw77!|<KO{;-OTJtLvb?m5f>C~i;qF>?St9Oq$QN({s1c4ib|TFfg&oFf^<?t!o1
ziKzUpEq|~L5fLzMM1)C1v^mU~Tx_|FqT*##Id-GZmke3?_mi*@vfX1_!A$0hKYn{J
zV<vZ)k_gU_T{0|Y#J>tdxD18g1g9SrgCBux_s~`#le=F#1hnE%9#yf?hl*NzU+qKn
ztCfG_k_9m^Yz!xdZ_`sD$<ti(*~06%f-*jrkXNL*m)cZ?TG{c<b)@%kg=awv4ZbXT
zzolAKExG#pHL=?3W*;?6gr>iGQt#qiE2F;B)mNuZl<t?Q&N|YHkQ*J*tkpFdVU+jH
zoBtRncUd?81JL^|OP8Uv9TWjDm$4iw69O|am%-B@Czslg9F+o`4wrF}9CLqW@CK=k
zS0t*|-F&~)Yf7EeH&Lvbttr9<90(Pl?N{G}KWi~|(^JyV+kC}tfCmKGZsW{OF}b!%
zPJ}t15Z{4$)p<?zlBHC=)~o3y^EcwaMtqLXGJLj}d(d-U1su7Yw`KBi6>rJJ(Sk8s
zvdBW(HtW4b9z;-VL(k!Y6(N5H6`b!^M^B=c+j)~$?A^edw|O%viI;<oh}!ymc(59a
zU$cZz1D`|18ztjfR01g%_=`{>jfyl4yv-(W(NG!VwoqYxJ8v*CbPv3tj%^uYBpL*6
zu<OXj#4laguV#e(<2mT9q(NjA(O_;Iis`Q*xs-aykc9qTDFN>D@>ze#DYKI=8Q{%<
z{i6JimBj%Cpu1BCNf7emiNW*z6g?@R(UqwfPJH^isf#r$Fy;e!=hZCxhmfX^H@gFC
zudMyZ6|b?NMiEd27z1cbET{YNi9s9(^!3XpC>muw+;E<4r-W%6M)XfDAPq@hyqqPe
zw2~wz+2r9d@!I<n|C)dDh42M2%0yk)t4$|IimAtq`ILNMC|88vOHE5`?@!EYsXkFK
zw&dNRvh77ktdOu@oyrbVVw^68G?s)rBXQ0Mu*t;|pKsSccqNhkL;bJ!y_PDwbB6gW
z`8byk{Is3yYED{HY1wY8FRD-JxB6}ioRtvcW4HCbQw~CU-+q6jCM<7$-`S{{eCktP
zKK}M5nc?~0*piiX-v^N&Dzk?Lx74c{KY3+>lNh(r*2dc1A~xu`7{1GgntCT#n0y2-
z#COt^vR@_YVk~|(jkcp0PjT~)T5XVa6mjlf>Y)iwG+|vFfZKn>H|~m(#(x}&1St)V
z5_oY)1HA-^J{EuHpH2Oe9w>SM4n%w@UHu~7iDc4ui(km_Qp9BI9%kptVPqzS8So%-
zx;emn*ASz5ykSUi6feMfz%JiS)FX;>R2yfXTC(WuDIEB&6x-IGP@Dz^f#Ow&WbJ*v
zHc9NBiU?u0wHRJRwEZHi*m9-~zG?+g%yg|{n_akaP8omVGCMo}a%QyXV)>`G%FZ2O
z8r8~YI?Hg<b-UG;At-dP3atyMv%cxN=Q(pr2S+k%lSWnZqFID!DS1UFS$>s1;U~U<
zDNOsdzxYf3QZchD&3v1onP$S+%zd)aP2Lh%v%D)Upq;E=m{&!g7u29&d;{MmFE975
zEF|;pEyRER|4CQ7O&4{igA^S_43mKI`-OqRwgJeKz%DPP(vkRj|F?viayYwibk~Hs
zLZO2rbLvej&nEeg^F_l#GQV!NYqtD~sO4Fz{Q(?W6duKj7>jk-4Sy~6Q1_(xoT5q&
zTK2ktC7qHD*b7r@rd#h;T~O>pf3*r$1+t&MTJ;%5IcvTiuLKtCOZ7XXyDMv-MGf>H
zJU6hjm!Y&B6ap|Zm(c|qX}5Kj9Btf};WRlEmygf~Czl;_9i@L<klU}%pEo?+KF~Xi
z-XEVge{ZH8(8#2~?zit7|2axv(uz)lV%kCX&mY~l%`ZC(y=Bg<Kcc^%j+r!?Sz)}*
zFJF20JAL_?PL|Gg|Ate)@2<nUzq()P0y)E$@nBezv)z(m-U;TcOJ)ZXN;<~Too=G=
z&a`0KE?Wi8oSJ`{4G^<H5aU`awFfbZf#Qa2E9S^}f*Aj}z0q8Fh!C(80Y_2zl|Pp0
zMUYVh2SwmR`W%gK;o;`LLHY(irf+mQi=d}m{0s*BJi+h5Lza&H85nK*7XCO`dx|an
zdH#MdU02;lIF3BG2f<xTYMqyfi2`N>GzyARjC0oqMLmCQZ$NOojUL%~$a~^$8ijDi
zL-2H&f2uo@aS(E5v}oiYM8?&ge=OG%R0)-kRf0=TkuD(}Cg+H4fH$OnlNm7+3iTLG
zdEkbWz-S8SGoW|P3j<T_z)WC-W0uQCfsvAN2S<=go{Fw`=otW0MtkQ3MiJ4GD#a-A
zMVYgzJC%QN7FcEx8#zk|22vb>@rZ^mMZY>Z9TVjyD)x#1U3&EtY?qQH1@$o`%_@md
z9fM44y`r$HQEW-Nfst#BtHHfu)GFmm2-PTuS_L<<-$8tH!KHUTnDxiQZ3UBobJOS^
z`lX|!#rtFqRwUEvSTAEpInCU%8a)F;oHGG<5gC6_-;s=iOq4Rqk{Fl+JGEQxC}kZJ
zs#Fh%(tMi1XVL?H-V~bI0hZxZRih1JM$jRDT0e`oI~u9TV!9!Zpq<a9U`qJFS|*(C
zeSRf3oA}_v6B4wUGe+4dU<_r{_~w85`F(1F$D3$svftv}`|(UTpIh}gu$)^usNg2t
zqqKivQre3Px#~V;T&1EeZR0DYnKW)$$Gy5?zfZtzFD}m?_7kVxMUyk!*-8S`Lm&e!
zj4Hkjv`{fTHbZDD2CBQ1ag~UIrmYx|$WX(a&(_fe((O;O`ZSIOf;fboQ6q>V0OrbM
zM=soSo;EK6ge0GfSrGuE3#A(hF2KX7srP^QLbq3&p6+TnYs!g{8w%ybxW%b*qCSh3
zvy(>^-G<135QB1=$r?+G>i#YJ4PBImZT!Z8DT~W&R#ProL0+s*MgE+~ooC34L(+*R
zXL+X#J@rw}pvQWl-Kc^<Az)glE2%=Ag_o99DB1U1r&A{wk%?&(9}VSpa>t?FF!Fzo
z+|`><n24)i=CXy*tA<DhdM3_W+Ca}ZrnSCE@v6Qf83$n;r5Ejz6tc$D>s8d28G2v8
zcr}B$p!Z&ZFfAGlW`rp!!wlLht4=L=Z#4!Q)6cMOX;bNPw74EVzZ(|V=RAodSlz56
ztM5BHa`x79%vr3>$4DsC=6fyPy)S>d$l#>HYm6G4DF4qOrDd{{QK7dP4WwdLUVdG;
zzEc@zK@?U^D;X$|q9|Uju6UFWKvJ*dEFn}617|Voagi5?x>YH5kfFz@YLQ~g=xXHN
zLMj=zIF(E4VZ4|g(naythD*4hhIr|sXnmhDzA|yRX-Xy1kpZU~7e)IF^80^M``kCZ
z+KTzz3$BMDI~r)wvr~OgCsnCFtEQv{Mwh=b_Pjb(U#RWI-Yg@_y%r~u<r#^(OaMA)
z!6sZuFlF4Bmn^$hcQ4~LjN+&*Z7RnAY;pDV(qc@aWuV+E6l2o~xJ>|iJcZ47)wa6n
z@i8{vjZ#xWFQueWL<%Ws+~R*!N~%wS<>Zo4Q?}tGX_(-d1=?aMCu*czb@wuE6H%GA
zahuRg%<ly_RC78#0i>5xE<6jQqhg!pMP8cic89!lrLag)<dIPw!3;7W>hCtlpm$+q
zaY8j7@2k6-aUqY=t&IzLd{V76xqhS-*FJNGp)WU{7c)yRd7NAd!Bl@8^kpznuY)db
z0w&=klUlv)2Q1tV7?-5RHGXeS_5fdL|6hH`Gf*;dj?e~5RxsrH(zUMoK4pBRNcy#%
zvIzqW(Y$T{TY3B6+zovkcJB6*w3*r$Ihdifuk(q4@y(ab>-JilIIpYDljZEgID=@z
zROl$m?j<Wt_5I8EO-6rx-NtXyGG*TO_m%JOJ7AnbbW}$%gATF#k&6cC5Qb@h>)yZ@
z9<X=0pKypGPZtEZc=cwtN>g9&n-=@h*j)&z>d<E(1#tq=22wOUX(cbRF|WQO%MJn;
zMP6Ga$^FmNV55zBhYO{<$?~2GI%}b{<f+mN?BJ?fD+PzlNcDf}*GkLiF-D+JjgDPY
zrbhQ#oTx_6$hPGywoxHBp*gUU8KW=0<E*|@8E2`W6Pq#~7{M%<>)LPz8kkk~ZH@D4
z);PxzDP*5<i&NRB8u-ikgi%?yAs<DTb?(w-yZSz5d?lk~Y~w4sd6!vvX}gx0ErDIr
zT4uP}(wcXD;4*)J5K$Vp0fey3S$N4xQhon2euJpO+V~CVB0{{ClFnGTj3yVEq!2;X
zk<K89M_JnjLEL|d|F3FOUq%$ee0(=I-w%90Kd*g`vp#mqm~c_GH-UuH3?{#{*-E`s
zcVpQ*CQ7lU6DO+}p#D)G;XI{)F;mMi$iSM{^EhTPDHI@8x7N7Ic~3!({9iU>8#g#{
z!d9Jd2R{TpdH)Ytl(p!Wp|l+o0Wp_xzXubSdwU(r0UftTd>zRL1kTvv1(%_;9TWjF
zmvKEl6qh}O9msz`y5s!&^XD5Ho4;(U(t==%Qd?cy1h!S+b0u11z)tQAui!J}c-6I2
zrn2siPGz}@Js4{XEGMzW!pv+!xH{EzPm!(+kB7S}^AbY%zWnlJU<-dH8PR6dB^m*{
zcOc`rjZK!Jo9<k=^QX{6SI#xgHQP7&E<SX7&^Z^x&=r4>zQfMyL$D%L1k71?EihxJ
zZ0t5HO0jT_K4D4DPqEd$WyeFn9)d;Rb3Vdr;y{{YA%m0>CXHbOB?1q_B_F<_!>q4<
z`Lqi^b_C1w68=nP*t_=MCfA7LbBNs8b=zI`rV7yc^Jr%Kv9s7H#kEq5P=q55vG;2i
zoB9-=kKcbF`L>PSiXI$G-KTFReS_SRDSJ5>P8eG@X65B7G|7AS)8z4vDxAmpBYC>X
zp$rIbpeY|iF=x)Zw+LkbuK|84dK@SM2r}ae#fm_QxD-Yk8Pq+0HEKc-F8B!4?OA$j
zE&#PXO)z2CKO!%mCMxh(Q?MG?3N2qgjbE%i`!;`us2Q)wGfQ6V`{a4<?EM3Q@b<?c
z+ad#_Ph+iQ^EQ9Nx<~mlUD*@~-u~1w77?E8(l|nq48_w<pYN7v(No?AsHMbE?{K77
z$R_ghSF`-H=?$C_Zp`R?u)x`-e!6UzJa0cUMcR6ZBx;qT68G_G@$K>X=0EGkP);xs
z{`G(L#$6X2uxV@nMJ!jtw?BqIZ~nP78mWpBrSTf@0SKjmt7lifM5W?6geoQFj6(qF
zT#7;ACq=_asqm^%+SC)06K>7YBE8%4q2R}gLL@Ff0(bsuz){d)`r|t{yYgjvE~AEM
zs+P|M5oi$jthE<eW|C;O!iYX>_Mf2)V4Qy~OQT<2+#`W1M!@R1kL^8?fQfkp5#NMp
ztfVlDqc<&^;quGlmb+IbQHkn6WppghLZp;VV|dAVY8>;_ZAzm#sivg4mpXjfJ<=SF
z!7GI#REw7At%yvCT9(VvjkZu>d%ceLcKhrz{BC649}F(wCuE=7tcLlPt#A?25M_T7
zF3G^;&c;8@;Q0pBV2J_cO#b#e9sZ3sKVxCXH~;lTzmAWg!{5WN4)}s$gU~y}iYR7|
z4AX%TYiu+-nULHPOjMVfsB9UPL@9f07JbFarQJj4l0;IC#tl}?DGkXuP2TnrU79xp
z+AgOFA6d!ntEtnaAUKbDyH1_vsLy}AcO1`V;ExAXSb_4l&v*)}iM3{9JR@eZdR)Ua
zCaa|*i-<Y<^cvgqIIX>C<`I%dGV9K#W_go-y7)A)zB_m>AYajinuogJK{)Z#vX=`P
zyvgmDa8RECqv9ZUW<Dds($JVmOZYT*qyUftVmL2QF%S!8Y4-x4G^Tws!^wY8wTJ_^
z*x)^`Hkx+4Gx8ftGd7dWR^r;i?sCemGlzK4NrbbvvU)J;{}FPWt4Zd@RT5zErj|Og
z<qp*9l31;!S^A3xF>CzK|M(EAi0vxM@y4viPao#0zZ6>SA#fa867K}D+ktzeAfVU=
z_sBFkyicAwTQZZAe3emyMBRT;$3xAXDu9LdEyzw_2~HU08wB*8(|!uUbJCVCPy3No
zZ-dDl_UtABnq8paS}Z8{KCXp)%aEGZuGLo#H20~16%CT6uN*)F%*Z{=UPl^mJD+5w
zI1cYO##bVjOMzSmz}~yEjR#*oPfOty^h~1;Kquf#?dkx?9uynoA=rOd-*t!|bo+Zu
zPwN#$21BKy$ZK4xDB7dwFgm&QI=Tyyacc=z`k2W2YK!LX6;Lxl8}@<C7$P?3Hrp!W
zn(E@!iO7`+%=rR!v8J40a+h@~ou_$}3*MuH(C*_sS`#7P$Qsm(+@<Fl6i+=@*$ddr
zMR?>HNwYj|7A+2&2*!W3pTU((ND*l(ZFDh}p4aRl1<zAq*3!rGlp?~yRo1TBJ5m6M
zhZKEi7j=-)x?QYtiJzQxSiIWtIWP9!fhw&^20N;h3qhpjZ)|ny0KHdZpqze2q3ccP
z`S!Z-@^;!+vxg+-M1j~vvbo^_$=$Y>sIyv`=dMtv%x5)z`cQxMkPM~DhvYRbl>aZ_
zrDMK>Akq6I1jdNJb>#)c_D&sx#jKZBT^BOA+q9f*T)FzX;wB?dN_1f>Z&M<85w~v3
zypZf(m1HL&y2Vw8BwNQ<8N8KLlGnJ9O4_~nI5<pDe_fb_8B4S_*Ip8B?^6LQ71B*t
zCSlqT1Ge*$XbOKI_hX6_`Z~$25Z*~~-A}>ML5#jQ(kFG&6zRKiiYd2y3f>rbzByT6
zsqJz!>$q}O<5I3{57c7<kU<LeAxo)<rR;kyySDeQfEy3$s4sB}bSA;8*I%!##@rrf
zYs5QLW6L?Xi5~WN37?M(Zf!H<Ej}Of)RNL`NhuqVN>YEyYg|f7?LlxHTr7ynKBUA1
z!L)xIRMldRwTQXq?iFC8L7Vo04Y<2m0bA5_<Iir6L(cz%RRztqjEk~38@Gpi@5RC@
zNm2HU<`F#h46s3g_wfwM2vA)eP<<1Adshn(njqe~KnNBH2vZ$g|4;8+Q{+skuawWL
znZ=vjCYOIoXKEhwf-~tLjQco~p#+Tmu_3T>LtuC-m9O$Ub38ftD{TO5AM%1LsW3<A
z<4VdB#;?8D)!wB7R3a$9u5&h!hA_>3?Eib^{eOE6ddU8(0vq@7ZOv9eXu^TN&tDDX
zjjx;7&1zgYv3o(J97iAzlZZZK#TDUFe9zUU_WpkraAQGY_rZ-xv?PD*@~hwFw<4WN
z_h=r*f_p%iKJ;-95JYhGw)gS*{}E(M|H~-|JykKl)$ee|cQx&6e#-^Fn7%7rs(I)O
zE=9v6qK`{aPK4H1*`L?mkz;_glX=irOIj2zU^?y2TfZzFr^_=L^rDxgH9sxA!Y;11
z_oX!kh?Mr-zb>ug#xMYtYBWQpPK@RiF4Us!fp!?bDyZWw{H7^USKn*a+@aOKURiwp
z57`H$mqE=g69Y6bGM90`2NRbtBRMFS`i&i?f6x1`&p+SEvHL)ukbQo7zWe)be*hMR
zRyh3j{VrZd4MKZ0STMo{82<TV_;&Zp+``nd6h5BO-_MICy^yg<V+5VYHott8!|(Lv
z&s10kKm41Dey2l$2;hgmhF_`ux!oLi4+Oll)Y<}g>Y0&lur>%CzhdCc)BRiU($cB?
ze|Z1aN)<6A<P9Nf2p<#A^>`4bhVU|E9i;2%VvT4IwBxr^EypF1P{+sdI$eX03F*00
zS_z}#$UP0yPMu12QfyhEnmUa4n_3?G5p9ZLJjT1kaw?4dhWoj5ZY(l&@)H-sC%zUt
znmS0=fIg{;q+)Zk@tnJrnw)!Px%o8Cf2Ro*ZvMq6CIs!+?9`>ez4T*qvE$>NaRNFY
z;N#8!DuX@MpJ0FFCUN}gVb?n=lLg*AAq++Y=ZxQX69>4k$d5n08LQyoDwe7&Yy9vg
z52)Pj7r|t75Hz?`Jbyi+5Cu^CtL=gm!GyhbYa$7qK<y4hQaV)<qY2Dpk<pqDe=kBO
zmUx^)ql{Jam<m{FEr6)24_4X=bp9Atd>Xw!5xZRvu0-*%ta}*&>ImNE6UXgJ$jhh6
zB<JIwR`N-m9YeM35zUq>c!W|yY1PRi6n#QikDkPX5uC?qYRV^*G!?0la7Xh*_^TtS
zd3*)jNZTl<JK+W>j9m|;Y$oq9e}hkLUnEL+BHJ=c7g>axTqTl92WmES!GR<-0@%rc
zq!whr$%8JW`ZoGZEED}DeUBNXBTtx_A>)k?E4A!4BU7ZqvBA@<@Y3)wMGcOmGTpt_
zCz@@y?i0qdZtvm~CMaQ1FIc6QerX>_0U*nAtn2E@cv`l4JV3v>%S90qf89>i#s?hr
zxg=VSYMOHkn!2##!BFdmmU7gbaLOEA%I&i;USuh^0!1}7<jbqWYpXNe<)9~2uiSD2
z)UDhWH8!r?PKdhoXg0iH=tFlV2xsj9y6Y=A+Xqzu3t3C|0SmqG(p@iM)I~Z>x`A2c
zEYe}(hq=j@6jA9W&E_n)e~C^@hdy4So$%-{oeXInQUNJoJ=9lAC@Z`Vg`CK#vSQ7B
z*;OwiEl$ssaA%KC@CVfhVIB0=ymFaB=8WyAggWZ1+G&3`(#lystFU#_Zx4_4u#>E=
z`!JM5!UC?DaB3glI&SD#&vd0L3@9Y1>de!npCO^wf^_pMe1>%Nf4R_>heFJb#pgmt
zRY{3oa(L?qW>I4!!<^vJ>yctv%ywZIbV|rz)+3}^4ygt?)IPidY{*ijFTq$4plt<w
z=gYXvQpoKj(lSM{SRh0Tb;!n?g#co`S(Y?pE)H_4kpi5%F;2a-_$)qxsFY|qnw1oq
z*Vsy-?J>K~1v=@*e?AOKeOKVC8MLolYaVF<Kwu%&mrRux%7kh%y{2#jC-nbYe{N>$
z&(0o|B(*EQm7>}sVLiemE6+Y0C9Tj1)sfX#@HLNg9YGyf%636e2a`5SZbI-9TZHy`
zVO~?+;ufL${Xy-<EI5g0!QRJ7JPK60CCvJ&g7#4t&~&WzfBK;5v_NZGLT>A3BDMMH
z&h_Fc1^8*3iT3&Ce%hYCbu&@B{R@6&SX1}$D@z(q+M7rUc70Ab-Hkowz;i-1%_3?f
zAfI*ik#DcZ>ytQRH{n@ki>EYOwBRYoYlc3aVn7&lS4@tz53c~5VnNt-J*QC$<Nd{w
zV<kS_bKNObf0csR?iJmXFk4XaZRxEeNiwa1uqs^bA!9r9?BJ<f8s<Z-TO2NGY+N3m
z5LoL03%vg5!*k$;bvKg1Y9DC<LCNC04}#K>=fJI{u5NtSdyy38#cVw<R61F+GpswA
z3USTSDT|{5(UVcPk;Q2rX#qjc>bwtvp5}z0o3c2ke~OdLx%WL-k1;r>TpxDRsJY}`
z>NW!4C_Q8Hr`-n%zU5fY_wg-<!fU*}<WKvk3uq$Cq^@fSWwr3`7L!1SWVJGihU;GO
zex8SncJ}mk#V<f>^mvX@ogQD**r>-(aME?>*GV$>VG^YN7=NPAZ`PHjkW^aHlaM*&
z=q+iIf5Mj=LyWCttyX{wc2g|W`xs7B#SmHymuoc~r+Jre;+Yu=6NYVkRYtuTo&xN^
z(zOrlN>b;WQE#6u>LTXYZ>uH+nE13dDw=UD*hjK3?qeTGo%e7)18t(VzRz~rGV(`$
zwgdCe1<uN(&9+_l83-(RyZ8(QFF>gieWvRQf4F`81za7=ovykXyc1ycIq<z?^8Q%n
zei1k)td|dodT>@KiLZV`>Bb*QA8UP+)~j9><RI6r3Klgst_n^FvGvfktlaxhqBTOH
zIUgnVO*^z>QNS3J+Uv`X28HtWeC&9ITm+yNTKQL)i`c%JEV4_6MJ2oBHMX)#vxlxb
zf4*WF-G^MrYqSmLe@D{^S%t<mQaers{DCEDAN<K74sOuB@{_{mSD)k*v0T<Pvqbi!
zaC|bDy>TQqH$6wG%v6>=soCNMPm(OY`*;#nL<5(8t3msi*0J&|db&>cAqWhr^Ggkm
z%(toB$K8+M%<aG2p!inG!usT6wt7>mHZNxKUKp)!?7d8tE1InDb&?~muu*Ze2gz#e
zIaa}4Xv#TZ9A5Ly2F;^;@rLZZ&;J0@5{)pIp|l+om$AJ%69Y3aH<xjXD<^*(Ns{BZ
z@m{ac2O36VM?{ArRH{-(%+2GzP0VTR@x^oST>O1!f&hqx6xS-Z4=SlhFp<c80l_c-
zyztAnm)Gwve|boCvCL_(;`06X3*Zfoe}Usgqc0YK83~u~e_Vd~%DFK3dBu<7TuS^{
z{O#~tU%yCjI0A0?f$JBmAdY{R0zZJi@AQ8f7oOI`U2zK#r2iTrbNtXq3tSRRgaSAg
z?Dg00KfUmXKHxVL-|ydF{{7M}z-Z5mhs$q&ywD9Lfk`JWf-@=9<<Eayet-FEhL9k0
z=I9sx{Z@f=j1x#>@Y@o|Uw-A6zvIoHaIp05@}DsD4;s#y*Z%Ua%in(xJ}(LG013{G
z(YhiyXr{SE;GRjkH4t1su3x}$Q~%^|*Dr<(A|sMUBLlRIA0($nQfXw2!t2V?NMeOD
z@h-ALn_&cvR7h}jkJs@UjIeoR&z)F`zkP(6o{n=7%ao6w2#LlCA8~{6nQ%X0@lIo|
z0lCL;nUD!cc;Yta_!NJ~kr^!g?$CN0rVGm^&}NX5f`^r4Xz3Z^dVq~f;vpHg?m1k+
zyh~uGP}>M%3A3mRMlfgnGq6cNcmufAkQJuG2+tgsS(0zWCQ^S(48KLD4JsYAqK73>
zRS-r4YmfW~@gjMDtED)IR+Q`~vTuebg0?)Opl+7^YeHlc|7U-)A4xM!X2pS7AzIR)
z?WLzWW}qE+HVu-ut0p5%{7}+<9oMEIUC}&U8CH^4PUzlgRDf_FMiTAAT^FBp$)co@
zI@uqd=5>h}M4Q*7#Cd*w6~Q#Pfw}`{uaIyHI6X$PRhGpNlwdT2oyypnI8$7gx^=kJ
zIPl1tBu$Szwt|0KudP^7l89S|XbTK7ZAiL;t27xR&nu2pm}IN=o_WaLBCPL`%wWQK
z;Md@G^%H&3CTph-&sdrnb)t5<5d$F^pk>q+0|b!*gjcSoM@zvh%5OcIoW4?#$610@
zTt`7W0Z`J;MFTIiBU8l`Sv;xb<xjGG-PNDJ&D$4egSUTe@mlHmXbj0_xF0P7;T_hX
zgS;dWPF^VzW=^Yxa|}N%LIW{QTek3)0W7Xs7Ls)Dt)(=DzTXQii9S6s@IIPjQ3N3Y
zgsq8|mGUvNd4XceICRk+!B#~(Pkj##K?%1lWFaF3Ciy)q3-w^acHo6at;I<zC`L6V
z^8r1SMCX41inMRQ6`zcLth|ap;~kMgYO*9aBfmd&9s(4z-n26ZAD!6s`hj{Ul*M^n
zxAM~>&T9^NE7!0}q9le%*2xyRlXXy%SaG#Tg564(G0w_(TnRhLMLuGRSn9^wJmZOu
zL`j!98WWU0^7J3bIjL_~Msx7{qBcnMrpB6fH2{ClJ)gkQ&aCx()#&w}PneJMeA?R%
zuhTr=X18Z_N2L#NdzyU3!kRxp<>-Gk2G)UmYyBO%C5-o5Qk~%NFr&cyuOrX<5E{lb
z3SAGU8qa%DK>arD)IcLcpmb`A_X-U=6neq6d8UPG=l?v{rUhoyrd$6H;@VIHDDtb3
zJKcY^878GzB*BmE+PK7w*?f#^6OWp^RXO1?mPbM&7H3Fh2%S>yBH7GqG=U_9zzVfZ
z3ilQx$@?eyTV>IM3e%E!!QUz;G0<O|Y1X4x`&+3$%->>|r=Y{@)H>5O+q8{lfos+h
z_;jN{#=z($0#Gkt@P<WcCsuc5Sv*;}G9-Vyw*A^xYH}!vXglFQq;|ESq;WHLJv{q1
z7A0jYz!h;oFn!k$@ByFnsS0xj??<1nia;wC+lB&8usr@e;We5GXs187Kd@JJg@cf$
zx3oDsbi^^ROadt6D%tj1Q>65hq4Cl>vl+zrg43;aX0RIl;}qOly#D6}w+5k)DY$<v
ztvYqEVo_Y!(ZfH_AO_`GpF~LS7(UjmdR-Ec#G^KeXVFS&j8P{<_lF3evtB98!oYrz
zxPTH&VSv=ka!6jPNO+K!?ZsB1$=J4d&GeiU^_;}WKq%Thp%DlZ4G}qt);Ol%)zE^<
zcan`wUNys{=FIS4j-gWuaXUBPlc;~%+I6e_|173+kK%3H#cT?fGSQQEs%WN$j#L^^
z_Piq7+}O1$AdSIZD`3)l{BD%63>jm$5K^f0yw{9Et<V!RP)9sxB6K$}Yd$Y}Uamm5
zE>*XdOotrO<<*&@yRC+~Q|81lG9mfJuW!wNlbd6JD&5+2n`4AVkv3^}OamKs&0ZE%
z8?7$VT2!ShMHAiB0c*uC<DZ~_)s>eZogGqtKY<$6m4X@wfq}OzR$Py9XTVStnI6qF
z9mxvmP%V3C+l5Cc6JqxU5;fGq<`3Fy<fm~kig8?Ii^B{zv}O|dJHw6S0qV@{2?y?R
zrDz=jse$ph(&|bnj|mf00BQC5*5!;Ed4{UgLR+qK>CnjBu1y}8Awyix2dd7OHbK;X
z%q!VJoZi}y<+mQ48?FcVAth5_&f-T6Z3REfG3*6Kr5Nd)gnbQd1@^YUUTDNs^Wjg7
z*@`Pu*keeeywCHQT+l3PIB~0_FE+;Tabr>}jvm_pkei!xO@6Vp$v<ku4xXZ{YjqY&
zD`0DBu-`sZ)16@+YJdyYp&A_7x7j6sfuN}Bw3P_JkOC95bt@655ynrfB#K-#6dtGH
zd{D7LsLCrLHS&Ba4$Q8*nqWl_qfS+(gG^yfNL3o4+i%_km5SYty$sVjp)E6*IW|yN
z?RwZOMNuT%6+jmm?~u&~D9CA9ih|cJcTd}Bk98N;F*7Et`>p%fHWRDbX6Lkj%nsXe
zVQ52@JX?5cs2|bS+#SU?xHY-fA+x1{Ml=(gWwuB`)D&gghV5p`NRkB*9qijAkEtJr
z*mZP#-XPwL6NLu6!|9-Dk_JNkqQ!mMHc{3F;5)i}nr+hPskl0Bo7R@+$5$grrsz%)
zStefI(v{+6inobt9Tdn=GR*ORmtF&536D|o_$ZB;YVfEmt(vxAiS|k}tHurTetWbD
zo=KFN2GU%~Nm%BrXh-jfa-|5HBpOIZ!d!(mQ0fjMe;U-cE2+7;c~zE&oVCSa<|PBG
z(DOmZMZRNR$Gqs)-q^-%HNT;X*Fv0GNlY<XYVHa46z0vaa^4igoytUiAr%u+tYl)!
z0**5=yZ&bB@B3q1O>bRxOa#3jVk{M|WRDe03vo7oykyp^wfqUaV1ELCBo|C_d7_f#
zkY*J%C7Ei`sm7jAip;oNX(=T&C8X=7O??E;Q+$x3`RdA(NR%p&PW7rYz7Dr0<jm1h
z1Lb}kKeXo3_Fneh_0!jX5Pueb>aIw8q9>HJ3&q1q1(8&5Qk~fe@fqJ=Za^bi?6N*9
zW2>PqXKXnl+AjsU{7}z5AqBHZQS?$ZfU$+X-2w_^mFS2((~uLwGjClt|EA4cBe9RB
z3qmp{?5C#-@A38HG2||G*n|Z2ZCIPi8}}oN=iZs#y`5=n3<jEi!Z{@p6f$T0*$P)f
zU2LOnFLO6D9H|Z`MQw1ND|9<dTTMH<mXo~QW_I66Y<kYi()Q>k#w!h7;7}x(BaH^C
zY#<(RmMJ=^XV_pSC8mVjb&`C*PUZlkrRB6EOvRjL{@vTqN?FMpH%FB}W=1(M4=1(z
zR-7HNqOzY(RCq0a3nBWPMjm)xZg5|R^^9ec;Yg4M;^y^jPizN!)69E9H>_RBTsv%t
z8Jf|_{_n_-+j$S<sY+nz2_0=_nlUj?Zs(ugt^yXLA3yHWTG)mZs&>z}s&q3L&$z6%
zCxs^;g<lZOjPV+5Zx8^LdyQykoh#VI64Oo`7dP!pZJk?xb#N6O2nAAid-G&?MS0`F
zRS~4r7_+!N=l?uE@`!4c${soKXe)%3cnJuvJDz+}7qy4b5(_#*p@<ZQ7zEsKd`~zC
zO~>o>|7$$RjK1%4pLIu9?O)ddy}*TX)DcE*1+h|e5v&yQla8)(nh7=T04`7S$p_eN
zJCmq7_o{v@2`HV+Q@%<Du7qgfD`(>gKD<7UDMc(1v)h=(gqM6?yV6GbM(O%5CDI$!
zm!Y&B6qi9&7Zd?8m(foGCzs-)9Wo2#NaBe6=9|&Sp(U3eqa7B1=m4j`z-gyeodIB4
zK=<^m`+UbaYWh6TM{+I%eGGlKv{k!X0hZ2yd-}lc)<{U_3#1Rw_r3qWKQFXa@hIL0
z3&Pzc%bY&`Oaokan`8y(T(G+@Pe0%A)P2sJ_Fw&G`QOK<n}2SWgV9<rZfy7aw;O*z
z3Sh!w2bwvdyYGL0c3*FPSz4V!Wr3jLS}Qe!3OL~>$+~00ABs>Z8ik@xOB%MK;VXhl
z(qJVGQ0nx*Kjvs_(=j!+k^cC7Jf`%{qsa+*I?5HM!@k)g?bt_Lj`kdA3zO|pWB`!~
zu9wItg2rm>q3KPG4zS^RmcAcBTPWrN*q^`f?hiWlGo34cXxG2t%x}YGc=vbrn}<GQ
z*An>NE;8eA%`oqfF71-p(S#C?ap|_3DC`*mvqG$!CAK)Vw7W_##S}MWTrt2Q68VSS
zt>)+>$&<+@!imTZPa@Sb#P~Bke^jEF{E&_up&^WRd5X@xkLY+E-y#tBF>d{Je8ftZ
zK8ft!61voXFIZd&1DQiie2k}gb1KD#KMiOf&{p$A>=E@DeFuM51D|oI4{akGITR!V
z3a&F88eqkQWr8^54>`{}D^7KDD)iUd=G0x!t}>_6mXBur_k(W1++@0X$cOTfw3Hvq
zcv9`pb)G0|sg2I$6JQSP((VSnaKVs^BYd$F23(qdHy+oaCzLi7(r_Izl(O&1<v6t)
zIydBsb2$}E4n6Bp(dpC$k3x#G@g_UmgYBZniS6kSPM{1X%;Xj>)BgLEF{6HY|0s!)
z%o(HX($NOw4Bj7qdY78W<4rU*nYZ}i-CQJPV&V+$uE&Hi%&p)5>vN~Fz|f9b>v4Ar
z$mQLCX$dtUzdvI?kUI}h^*}NJDb2WTn=^=|OrW0OWVv1hEvgtESeh1K5-NTM&<D8*
zp>ZG-_G1{}DTRA7z=KI2<NWdDO{_YsW5O9CyEd#t=Z<DdMRNjeURb7#NJ29?u<T9j
z>`I4{&dVmo37Un8aoS>QVyuqO^{|&w$G6FUi_$Vloj%Nqf%_i%eIG^Bhz>F#%_>#$
zf0_-H<;dzmWx&FydD<pPVVG1$a^!PLEYwUoc!d&cnfz^H)yFM)JNB|zk(K1_X!1zv
zE=@iYj}#`KX^XANr=Ixh0R>Shw#hzzEAU7pwH0X9W6EH~qsVN76_-pIaSE#>(sU|+
zq3ZSE3dK|%h1fCzPAje$w<{rkZJn@M2=d)1tq0Y6Zk-V{tRSbGFhV-PG|2VnX?q*5
zA72JH6J=N%+$=J!)OsYn=IYvJIel=O&Zd!Qay63_GN{_q83ySnY}+u%0Mkxixw6)T
zGh=OPf2oRWoPa}It_vn{oc0eTfbC^}usoNleV*Zx3VOc{l{9BYn5z_~s>hLmM?@9Z
zbj~HEW5(DPg{h+~9pYkp0qXi%C@aiT5$D*%Q3}f}xVMVaMJi5HG|MJ^vlbgCeP=}8
zdPHNKYqZID=$HH9>LsW8L1n<=QBJl^fBrWo9A8$=NphmK6mQFVZI<<>@D}EOl4*;r
zxul-PtKlJB)XGhAiExrRE3aKZsvlDZD;-5n(^Mj?WTa{`g`_5{ef>;wD6O8q+Lsku
zXNUT@<agPnK4AcmKySZ<chZ~ix`^<aqFI(KXDv1+%QFIXJprgF2HPY{jSRSX&U#_}
z_%gVOD397EP6Z64Ev_40TaF3I8CdtWe+2kqmH?+7DNIV!7F&~2eGse%7e-atCMls~
zm|4{-)^b>*<*G-QK@FlbZG#$Eh6W2@yT-a;p9F@V;J)N_!SY4HdJkrp!J`0g!whbi
zfbxK9e7~<A>N<qzqIPQnAv(>3lEuOGe_AT+BWJwxoxN1pXO`iop0!Jb^-j-_f5}E&
z+=fh6k)>z+vDoW+jsx!p85hj*v5YX>dvt6TdY`me6`iWxo}rV8(}Xs3GLlK5uU+t}
zA5#V^9p!Etth8X#n6`y3ydT-RWTh}ed<@f;{&ZDa?1~R|KEAMD1C(m2dWUC-B*QeK
z4UZJfufaZN-K2VW8PtTIMl||qe>qBEO1Zb6S(QJ;c#tG^{WI~?iP)tcg01&{hE*7+
zAZ=JBIMeM)aA8?M&g=Q%3Fyb{)6Q`D$4~2#%y<am45y8Uz%ebMyrd)3?axftf`%+A
z16=)o+bX1afUAMX0L-K6Zv!x0L|be+KRqK3)F^^NhAQQHHDv3q02dW=e-l@LGt6LB
zso(|8`TRNlmM1b!_g6i<hsAk)<4joZ#SA-aoFlYhht<qzRm_g2b?*9+W)Rd-<~3b<
z1EUyWv<=A%P4F_>Z)R&;gbFdN7pI;z;XAhE2U;>KU;pF6TQbm!^-YPVcolC7a@i{0
zti{GvyfflyJx-Mu&TaCXe-g}ETa)M7PMd4ToIz4Vf!hX4iOiv^$#h$F9&Q#D&$ogX
zpDEe&LZDnO*bO>&iw>#weTGvQHGLaS2}&1ceF@?P&kX9Z$ziUTcJiF}!fiHD_9i@U
zR<4^p^(wsOEB6Xs+31kTvApSj{URU2GRmz4aW{E)TZmS_chH~de|>_(?b-Bgg7kS9
z%UOO3pR5gg`#ty}S_*s-g9Ww2UQGEM5*fJ*?}7~PG_;eRO;{Rxr^Zo4j-j$1Ehata
zLW%CXhyP&jub=v8I)F<MZ2-j*t|KHr@0quuj40=CuRP^OiJWmF*t}n5iGAkG{otY#
z6`B6%;G#YUKMZ?<N{+i2<UK<zovu$mXYkxV@hx%_(jb(ui?IyG%*8-MXTn@*9kl7X
z)wWgE;bn}40~gLQMWT2y@Iv7cY7Q`ejceM{&Eh{nMPn_Op|l+om%%(U6aq3hm*F8L
zCzlRb0t=VSs~wpEzn5mL9ie}K8#()yF|~DqCcc+%Yy(Xr8?@d>ll6ma%c37<t3$pQ
z(GhS+WmI%gLCqv9Uis_`YI>yF8x2MuT4Y&h<1i;!HcCkLO2J0aKI?@qA&zX2ix2ZH
z=zvSx_C4|KjgbWXp!#viuF1!C3Y8WTI)|e!@U9MM)rcaKCfi_xC=7p2%pCh0m^PDD
z0?wd6mtc{f2Ey<UbA$FyfL?XOtW5@{-)IPoy_mMQ?ZFZ0eip)P;sG1KHtPCb?GUZX
z>E}u8)F@L{KWA8s?&5RimkgM|j`^;dMq+;QlG|UC6O|qD1SSSeZ66q2fjLQGW0uDe
zXAP{a5#Xfo=o)?MWC?%O1T->sHGd|(IO$wsyaatIPg8aJN)%TdI7E)BGR<_n(G>%-
zn|0DG>a-FFSnWHQ;#H*T+tX%6^C#_MD)N@bm2_I3(yb9*+P3$#M1Ub7jcMPak*?8#
z!TxDVe`#QvCYN$2k11*lMUu*vXO0?5>1Fm)<HNx2L6Fyzk>!7sEFwh>5*u}Eg4_mW
zLSdz!^j1)oM#;!k&vX6P7LQMt5V*aYqVi!=YKr(lSlvIRR*^fsU_f&w2JKkkbGD{8
zK|nZtZGL@ufBMfPkjX3IdFAo<Z%^R?Y7m<3uK1~^?$>{IzdZe|l{#G4M2B#6adJ2e
zId+I6W??=9hu(i-U~<FPzCGGLLl3i=kuY*|HUY}6{}aJ9y7NQ7reQ>eF*R!DnYLX3
zDX>BTwXJWXYQH+j{I*>m0*;c2I-638{i=B_tjDdYadO2(IMMCVNhr&>j_0QgJo3DN
zlKrAjW!5fBJ(<s+vlLta+I23nR-&|qm(k&2Wu)Pp+h>2aL#<qbcFK(hTsg>*qlRxi
zul7>nIK$H~cQAq`p_OSF0T`Ov+|3AtAdKvn0+Zn+HfT)}?g8yaEn|=Ir`leXoUd^3
z>eg5|s^cL50iR)To^=U2Ck*)sNpna76qghUMKaqvcIWtd#>>(RU>Pj(r=Ml_4}SPP
zH|AAQ_uqfm^jqylcK_~v3_f2G%po5I6LT0Y36>qRuJc)JFM_m>tg&0fHqFqJkeZgo
zGE6<Sau;zONLaIQ$5EpzFY1&d#F|<-FxS??nQ;Nv{QD(e-#Q9Y_X8Utq?qjZ*w$>u
z>s2)OcyqUiZ2(5jeI42(0B$2?LxC2FmzCC4N_c;nUqENp)=eJlGf}9)Dj*jb6L-g_
zcF6iYlOJR8Fm_3J5{sj@LhC~$j`xzGcD#~y_K{?W5`~Ayb0JC)R;jz9XRnBDnrSbE
zHToPb)E*go6Dka}XEWtdz8`+SW{P}WqQY%7czjqZ=5kh8d)K74Tf{c#2C-!6gJ@6_
z=Z1f>=JPt!U?MU*ton_U+Z`!vPZ*UP<6Y3N&=PLH9#`rZO>&|uY)XSpph>MY2SPt<
z=zbYf6pu?^%$4AFBWht;%mxlNI4|ck%(4onEX(sCr^bh9mWMBlc8}*-<*h)vCpz~^
z*r=B?0+%P2BP>ho0?WcZ6G{b_{2b~bWxs#*R_V1Aic{r;GQjY2zlYm`5PG@o)`;l~
z*|iu=X>U0ct<evP%PdKi-8*P{>USgQOKKLWLG|@*@M*3va@#y`-0&+7wQPH>PI9Pq
z!hyfXp<DNp7M$wGP$+%vbLyp_YnxN~EPJuxdZHGLMdse|Wqu34*51}|a@e0MI3a&6
z46IG|z)7cT<?g_X3A73`cyR*+JQ_EnYMWMqR2Xj9m$|)%dKgoes+=7xq3*&|6gk=5
zEt~!gj<o3u)^!GUjmFoy(H@l_Z4T$UW|_dTY_(;^ba~?n8Ir<V8*+uQqb;LoXx=~n
z<-7a}&c6~ql^MVBr|*WvuJxEWj@^Idm_Q{0lF#8zWMg2zB7M7j0wr<zi;x0-O1spx
zV7g|Y!vk3{kmALQwkMN@=B2K~k)fP4P1J^h`?ilv`t7h9D<K^(QJ#wRlT6{-$HC3*
zlbaD9MTQ_xTR>{*tBu`pJ=*E3ukMcPsN5`7Dqb5a`i)WL+-BM%^MO{Ir0jo!LOIb}
zWyXOuH!|b!jlIlxJV2KNU#3j{9A*R^zkZl9WBsD%TEb}b;RyAsN-E!Xx*aTecu>p6
z0++I$b6HX<0fX(so+bZ}Sr&K@fbMsgWwrG0GOWJrVeU(*`D{zdwlH@RIYZtX>1XJX
zZS*sGV<-LGf&Ar&vME_Sg?@iHYy>zz6G^A$XvfF2Vk?vK%yVpIyiod<sdAQ9zqDJ9
zt@z=;Ntsw10Jl@F6t)Wy|IvxUVks!7v9%&}+;J;_pp;La0|;_LS+yKI=T76t$G2iP
zOnKNjc0(&PtVGi9xxThLpPtOpt2B~CE;7kR1ReJD3W9V>Y|lZEHbQ?BesFE=6wLJ1
zsq3R^vS|el>0isrmkMRwUuyvS3t{V6I_&cmEUA?5pMxbuEQvm(Hg$X;D+ZD&jXQPK
z1%LvzIiof;i_;nz`)g1ipAD_1X1R_tyK<CTptXM#ogPwj8d`JXs&Dwl-c{f6$XkqP
z7uOo65T0k$f&1`^)6p;yE!)YYRPt2fb0CQPZ>!p5r7(YN=$*5nS8cnEKKkZ6x#KW)
zC*Son=l=nse*|5Zp|l+pmvN0469G1tF^M`Sm+`b6q<^Gq`|+2zAMR*0pW!d7em}q6
z{r7I!0j?EcV*2;jJ9iueBf{_rs36A3>Ca!M-|qgETSzTaV%#47{W5FP5Q<GIrSMm~
zHh=zwPXED^KVV^D&GbKC^s6gQiIsNxWBLi(r?tZkLx)qPq?|dNNg}Dn#w`)<RB*UF
zJ>D}zRe##$6mS>N84@=@mjZ?W(1Bpw{^Px(ES12s>@~Qo>d?$hQdmO4UeTu=UjYz?
z?Qgcr8L`YP+dbejX|P4L=CiQtLB%5s9hg{vr^I6t_ods{(XSAE>yPVDR;b`(_dnOM
zo{q=Wqj$Qib}Cg}Solh)<Yzo*LwVPftLra?$A1|9(o+g$ZMZi?#EYwVB^`flXgM27
ztE)P2*koBSHdQrqKlmneWi*t|_|6$`oI_QPgWyDR7i>*H=oE)D!=r(aJk7&ZT*`an
zY~?x=+#^jfSXF&ZLEG^nR@w4ow?SLh1Y<W0Qq+6z(+M#Dh}Bw}C8{-pnK!@t=l8J+
zTz{^ju*$r~KfbTIa-RVD6Lp&agdohaL1ZZu0d2O)(&vE-!;8ZKf0(DXi?<Jy4zY!p
zJ<Y_~-M<~b;lRta7)+cTf0^ATPWqbsCQL<@ZePSL=1roLY8YxI*w@R1!hlHi!T5yH
z(yZGLX89(wWhw#*3TJwNeLh0x!OA8^Z-1rpJWMSay_J(nKpH8js&JH%`ZHiVvnV!#
zeHp_@C@e8qwlk_($?+Z~*WJBsw@DL(=@7R`%ZUQf;I=5{RGeF=?S%8Q3>5_1GAs6#
zna`Gk;%tbdXwnnA$moyY4l;C*-(!$rZ~|(eIibdm_WfPmaiI#zZHx;wC7R3T<bSFL
z{1F3vYWrO?rwn|1qj@v4_$DutOCy-NgT4bM8C3BYFiA_ax-{2*C!IJHKAG{a6NAMp
z{*+amZ<dxtggo}BP|aH(rCAr8y2HK$CkT^-F>nGy4At+w+|}Qw9ba)L|Jc}Bn-rQD
zFk_dy_)|gc89HUiliMz9Er}Lx#ee^+GA^zTU7C4Hu81-A@1?ApEM?W+Y@0XEY8;$6
zUXg^`$<4wfWegGGuNB{NC9A)CJ8q*v!H;np1;h~h5Gz<~%H*cR7Rc-t$|Q<e8SY)U
zzPWWGcrHA@f0;&Pbt7OKS?D=Gu^ECaus|sLmMb~^-P>`S3Yu%ENC6`Vm4DE_qI0cY
z$#5=Mmhs-QujBPgfO<@FpM>(`>X=|`yEM3@V?c94tz6YH-Ns&}V;W^cZG4+-nAJEa
z8?H!;?R1ktu8(0U#RwDXgDj8rcXh{w!i%G^ge56qTKBAyz4vV~KzDOyK3GyW+!vdZ
z@tld<#fNXQ(jU7tQn{;<Mt|xWM`@%RtJ^8SgU}trBHR!zZ8P@A%AW3??s$*}MKYE}
zm?q+wA8+({(#isWFZa5<;}sJ9HhO3H9j`3fTJ#Q8L;zc*Mpj)Y8suF7=g(0&kAFCs
z*GhSn8yYFEu5py|`ZH@YV}%WBZV2ZIEB&LO9h^5-MD_Q2$D1Zd#eXs0G(doNG2hYK
zjr{Pc%(uHyAEe5GRhfN)3UrY<su5G&vERXz4x)JsrnDfS<cD6{yD5rpwzel^dOmq}
z-#g#CR+ZK1E~5@j`2N=mBpr7m`%$rjHW_637_=!=5LA7`^}hb@?YK=iO*&Ne3dx8B
z{cfv${>_p2ZV9h)uYW!^<}T&1VWa0Rr2tB#F(2$MWur!|jc?Pavl<6AYIhQBrxp*Y
zc?=_QOBA;sWNoj%t2-`aL5GiVAu~kjzn^^T5)rE^q44UpLZLhMJD8$jHZlfNR3jO+
z-*W!TBWd?3kmHk@(B$gEOCx%^qr8nCZr!|N&;twH;!q~}?SIkZy$AF6_M8Gu53O#X
zrdpBrVQ%z)zn^yeZ-NpZ<9~zd<Klxhe-Wp5J7>q~+7ik3?tT($QnMFKJKDsaW$6uW
zOT_lHD81HA7Y#4p>uz}Aor5AAa(?E!QZGJ4_d=S7im_;ZANF1-cZR<P==gApTs|%*
zC-8)$05RkoMSm?CPr&&=NHP8Cym5B)$1e%pXVxRIh{hTB()#5yKWk4rsI4hjdAIq{
zGoqI+HG6-5Efj9uN1ulx3~1S<@{y+}tKF~`a`e%ETfl|IOWaev_Tg&k6ujqj#551Q
zuC!f>YF-61S(aZE&qv(2X>S(n3{?SyoAuWtWqC=xZ+`_S>%2?aZYPKwr>G##jBM37
z>E;U;?gJ3=sbry_w>;w+#?u)S;NWLGCY^z#ZDbvCUlL2O4G8H~2!-qowU;o6SA}d@
z1%r4bz@R$6GsxTuFz!({gLJtEn672P_<lnZC^B+6C!%~+<OnDjo3DhV7Euej5g~O=
zRYF%qSbzBMi`3a)jyM?kAciBfOqIdp@j0Fi<9&ioKI$9;5i~DF86+i;`rtt=Wrl99
zuM&6GM{*ccw8G@vT=oYd)I!|?=X(_Yb-9<iFAjJ~zobD8)B^f)%P&=bif3;IT(C$Y
z{4BaR0q$YcGK-W}UQ%tMjIy?WFMO#iB+5v|=YM#L<qQ+g5F`IG$Iwa+Fb$}`%C~x6
z;&rWfsD^r0d=c>JN<bAqZ%?}I^c*4#fwJn6)3@sJlUl3%q3Y$RDhGwFi7;H(kDR5o
z1~f){k1BvV7QrVe7WwL_O>beP<E5YF=6b1DI@gV;@;PerK%@NEi)cJA3QFMxqszm-
z`+r)JlC&0^X0>;5kUL)yP8e`^G0NwImU0y3RN|TxX=FWC4_7Lc$^#pShw|-#{H+21
zf2{i&83~psiFH{V>S*+88ivE!G%Wi6U{#{#5)s$2_Sw~DUq^s~1J913PB(HA-|2_y
zZBU3Y&tLt?*GctEZEf+XQ&^rko+CO==YMPvBbH`~naN^Jc4-OaGBPA3s*k%8W9fP-
zQ}s4-S<MdlNWfSbL&KR_1~FYdl%27eaF!xyX1rM=_$l$LsnD^oY0kQF34HmSRPu7<
z*H0&DCLDk`cee#Ej0c<%#<Neum(Oqmw_#?WPhC={4nQSjd@`UpkUBFL>FErlFn`2{
zjq(FsbuymQTQGimFte8M>bdX^Kjp}!WxNvIm9eH4(eVoBXbRc^PYB~?`BI@^MdVt_
zgMPSp=EpA??;IF0Of5RdPxDWtEKPEOgIC|J`6uRLquLT`?4r?#ldG&XKrG`+Yjq9n
z`Qqy$W#$Q|%J7{@M9pNyJWvp^a%5S434^`1;;h#*SFHgJs3cmmS>-tBozA3JuEaFa
z++6na62gWNq4-T8oQLl?pxK&mR(!oF7jnIN>VFNWaR$d0J(gR{Qu?)I7Ir1GEMtc1
zn`i?I<Ji1#`u+TN_kU}~ni7}M_8JrcHJ34oMkE3_IhV`eIx&CMT3v5jxe<NOui)eM
zg_j(XGs7YXkWHFxfEEqn6xh}<3dd2i+BlLS%Ru(u_Y9?LOTO0Hl4Wi45NCKVKZe6I
zGiP{}YA!CgR7-J`n;<TwCa6oSCS;eRJ>lR=wI@uDwF$`yMe9?Ypn)gFVajI^#Kn+`
zJyJ;0US=0#v}b>4MyfWc6c>{=iCoOl`k@&MSCjCTrH#NdX>TREs2Z?_h@+6KpQs}u
zn?=b{!lqJ45VTDPDK|AWLuz_TbyR{)jE#_py@fuc+@uJpW-Ostq9(Ry7#7<0D~83k
z|B4qx?JYrS=CUq1iOV5CpF&N-q=ikr3iEOxdxl|B;<JAo%(z-wKdhnBq-=kx{ZW`d
z)3O<18%V`v>=zO&k`vU0glAcsgcl{+1QBDo1grtuWXU!*VtRqS@C^NG>qo_v)+I!S
zlx!Btu0$LsqzI|D5$Z}xHUbbSnc7%^#Dq0+b)~d^CAlJP3z({Cv3(>0daCK1lH*7=
zGp1r{I|qMKIBDz0jtiLvFd}nlv#4WH!5TPiu1!;HMr<KS(MeP`iJYWpvt%GM7h5f7
zCs4tp;&9NWUuGv-Y>g~VYI}>O#;CRrh1A~SK$XNa06;V~+XEn)+sVKUwB*(=Ib|^k
zFa<VkT{wVhv$(L+DyAP$SA&|Z;0hJEw|?*#k~V*W;XBQ0Xx7r+!cpK#Z5E8s#anei
zqgf4}QEPh(!$ttf76CjX&WxTtf9|?h)AQ-<=;HKj<PseHz3bkc&gY}q#L=Spmv8f*
zj^?NHk&BB5Z>E#^^XJ_goJ|aH7(DFW09TQ6S~L(DBD{CefDxh{NNWJPvBv&Qu{Uwq
zU^sthTsES;!9gtEz&p)V77aCz=H+ayA&W+}m=M~S7caVZv+3E<Xnx$id;O;SF#0w>
zdGP}K?LPc=F~SZ%kFfE{d^DL~T5wv-_kMIay_%hkE*CM)9{qjvX?*(Q^xLs54@?3J
zbb|3Sj6|=6zKfY(AGaWBqkjI^zhN5oicNo6_?ujvpPy{#CFlictN4OYa?^8fC-cz5
zShMo#$*djiBdpAEXyv+#l{sA74=#0wfmb^821ca!jQI1^23|APy0ve>3d{ZckB`5O
z&TO)`U&*peNt<kn9<b!A-!<(Jz`PTH`9T2YA%J5^9`Wxf?7ITFTlQTG6YFSRwfuj|
z0d0#}$pga*J;boW>xPxI8M99lgx5WF4R6LrxIO}JI&@{J4>wEqax$4NCMrua9XA4J
zTWkDnDM&B0ZJb=|)5g2*$LZ|TXx5erC*5DWx7{l=V1yH>b~Zl-VfCPU@Oxj33Tp5H
zD{`rx;>ok_=<4I#ru*A?@}+xuJ^6pxd_0|WN8LZ)zqNlq{4$?kT>iQ1K9A?WTz&Ls
z)34osXQNNE@t4t;nT^It;b{D?=YPUZR?PNLH8uGPUe)Uf4(Elo*K>X~pU%do=a=2t
zbTU5cM(5|_i_6h}PA@KI({H=07m_zl4As15Becb5oCxxj2any(jxgis;_83nrN11X
ze)ZGY=kC*V+<lx*zjQJ97?X6-<Gh1{S%X7Uy-Vq}74P0bTr`$AP;WFD<P(Mlg^uKz
zup@@SLq6$VI!u4m{WSeB?OwYdE=NsuFWy$aabgV*iH;iXNbmUpy$aM^Cjpy!!Z4n}
z<qn(8*67+sBC$rw)>s9CDOrCNav$GYrFd)wS<y$653PfUrLs#PSG|XPtaLauyW5;8
zHyziuW~{?}Sa-^YH9f?Y)_NV5$gNpy#g?~osTh*@5xKMuky&>_X1x!Yb%@NAvrnlW
zl8kJ&^LAj&A;>MuwR?}PSxCEIS+bCJ!(w}C7c4{0v2EfZ_A$dMWRZV8B6T8VLTwcG
zp3)%i5a&pRITb~!N=DDG^Uv-&fL(x{<76_@xVj|O+vQx0*^|gX3FZg@77`M~Em0}D
zk_6ke)V-d2tJgAVF%*&bhhgou=i2Q_pq)Jzu*n1b#T|lSyQ4dBagfo-w9K&v&`h$6
z3&uJ$f$U@gy$4nQ-*$hXG_b>xDQy{wsN962$d8XeSy4k2aV*iNlJS25b3|LC4>=CJ
zKSq%h7%T==K0((M%9{HVeXm%hAy#Q8tkU<5)e;7F$a=pWFp$=<+Fw~2R#RN2%|yM>
zC(@LAq4Waxg7kv+LR|_u^g=&#?q|;Z%(<UA_cP~y<}8o*ov42dpF!Ek>MJxP(0I0Q
z9RdnUW#~JC@>LQiNl5wv1IZrJcLjw&rs<VH3in9q8<h~FGfc{056S-2vIiwHd;5G*
z3VZunmYBSKrvq2P^C5)*OjBVnh)`oj*a`LVIwT2*TzaiT$>as4%my8v#14iC4zL3i
z@VeL#p>{QrvX_5)I6xE>4ZRMzRG$nW8+71Ai7Pa%38$j)^?i`af^=d^L~y!nAA%7F
z;g^NTYbu9AL_zVw*GiO(G8+}ZCMBK-5>}5E2ebNw<d{ISgMFv0!PhcW0y$L$r$H&A
zt4b8!gC)S_72N(>3E)z!y+H{<XMzzja#w-&2r7H91fzeTu&>2zfJ~08&_MXih*Y)c
zVK719J8ECLmr5i?#bsWyL5Uo35+XquHAB@BBkw0nSmMaJdP~xM;xJdmN{3Axy%OLt
zKG{d+4N5#o@tF~<i<E06ijcRWVUrS1f(b-PERIE!U07#F1=$Ns0jBnm?%=?JIxY*!
z;0kku0=|E|kC2&g9&A4Ws6Hn9d4M8%#dh1T#9KuxW1+eJv0s$7;+#C>-YmOtPKh_1
zlY3^!Ju~F=c-EJKp+X!2iGZ;pKp?YaR}H{)AgnCPis}SmTw5SDhHT$k4#jjVdT8i1
zNL9W4=468gQK_hEK+%X88sN1`Ygpy)yJC{zMWcU|ozPoX(AyU;+!rt07cbJ|S^7#x
z_Hh|0RypK$6@U|J7lEKmjuj-K*6mJpcW5IBD^AHiZW@D3*~pt?<ZIM(10^cBl>NS`
z0N|A=gZ1u#E+G{fb?$Y@9DGsUs6;H_F<OZtq4J=hEc<XVoJS#b$h#irYsUe6urcct
z4|HN01cF#9c22+o@P~}+t<XjoF|sh!0z@AyfhbuOpX}#sd$5G1L{h!hf<b2s2>m84
zeh(BDZiK(rJE}UoN33y|NK;qt5w#B|+J_VEV~_T+NBh{L{guq(JLbv%0AKk9S(gzx
z9uxsEmvQhG69F-op}I9EmxRe3p?`qK!sYz?%a3<5i$BV0tP(6yku}JsSXSlE4`?($
z(M(vgv|8uS%v+mo1V2%}ek&BNVO;w(Qn8tWP9qdcuAajvQ$}H#qXMNcN9H+@ILA>s
z&5hiuB!O=n`PA{P6yxk5?K8joGSie*>N7%Fk4~d1|F{zMnAZhqen%@mSAUha%G~07
zD)i0}`LU!R&uAa>Y?Udw#8cVIFKZ!COFGT(Q6qI_nMSAAb`2(4!EE;p3hA)yexv}2
zJArL?(<Lhss2z`&Lh;Lyrqf?+SM#U8d}J?iu;?X93Jo*fc*&CLq_Mb0JFO<j741KG
zlTBVytfEMlZG6o?5sU!kM}N-2^ArZ3A(M4|`)Vr>@&bTnt6Q;!sN+l3He@f6dH13|
ziuXtBb=g*F@*`~*J)9J<bh9n?WnyJ2>yv2y!70)&)d6ibsB1P-6O|U2zJ1p2<b?Lt
z&{8Ngq=iJP;iW7}WmCUqlg%bi%4KJa{K9s8$e*7&L*5I)hPT?_Jb$d^pp+H56O@X6
z&JU2CLwz_?bZuAY#1U=^iJ5W^i8Ck4zpEyAM2*Iv)d!JGsnbOY9!Vc(sDkizyw4Wv
zO)!J-_Hp?3;pOhnSpd++1SOvS_;MG1Py~Y5Sjh2Hb^7byr%!i3k5sdrNg;WP(BTh%
zik{Iu`6VYyFL|e=F@IWjI|VZG(4H+s>9SK*my;%X*N#DDPK^u;4kI;upD4!UNLfYu
zV(-fAIB7aled;<n_KH)k%FV%$ePpDF&y5D$6xuPTiQ)|*NTgKw?bc>Vz~njKVr@Oe
z7{E;gD+?xDX>P|AO-eXbX`(+X`PfJ(b9{MjV8WC%k3<m75r0JK3B;*9YbB|L(bA5y
zl-=fqAb@`wW`WGHH|st+IgF`)ic=~Az8X_=vaHcfh7sy<0#6fgV&czPLW3vId%&Qh
zOPL{^v1F};Z+>F{dcU<UR7#M<aa}UViUk@aIg4y5-IcYZ%mZ}6#s_RWR$uEkRX=Se
z*?qv3cuzniLw~hJFjBQHCw-Ru5fXa-^{JVAnM>S4S!Id_)RaNCY)%trcAQx%s7;&G
z94W1y?M=?-GcS-zC_lv|m95A$KxN8$PaS%=WX<>rs8+qlau%|F<VG+dk!?S6oUd>;
z|5iWBtJv_PoX;2hC_I~dNiX>lNhhpQos@4r-EX-c0e=aL{Wi4j$HCDEZ^`L?7OTS=
zo<$t=v(x_Pep5C?0iJqBpSQW76=1`193$S$<){1Bil!kTyu@A*>z1jqbWxCI$~)K$
zHxw!gCO2%qtPrlG(CA{9IY_I<n;~L9t_O`Bd{qsEyVemodADFOj{t@1zNQ>&%h6&f
zt|Q^6EPr$<nsvL3{H>d1ckDUOnb~vYXq%$bG!afV_+*}63*F74y2z)l317z&Tidp1
zPcoRMuy30W=~vZp!^XRdOuii-N6=Ji*4UjYAQM|BoF~jyVXjNvRIyo!hfNKYM#<cM
z5&SbbhB>4S<2*)2!MY<tTl}RdDOt(F8Z)zxP=5?#%5LNX#$7A}HIXpRkF(`<H(ph{
zv|3=sZ#(e{03$5xrhM28BY<2`bC6Pu5-rh@U<F7*gX_6`-r?dnG_V#dhekz9!D<?U
z3|rNaU<ILEih?v#c7heC5ZI^_z6(~xmWE(un~ZAP&@_awz6RCBUotLkbMXq|WBJhI
zEq_~x$_b;)@)oWjM9raE`%w#1^h$ALt$waO3)5<2G2V2Iu=NbSt><2|sT!MgORh!c
zCTVPGT9IX|F}<dDle<zVth2W$TPY(zs!`c%X-I?Mq){Q?V(wfBK5edTI;RzM+1+9T
zJPHKWNw60^ycZlcd5PiukmX4ZQEs-(27fH{>0Z@OU)j!0)Z6O9Zd4ajZ*E>)6lvY8
zW$Q*CTDH=K)lDPE?%WyiacMhe-l=M5VAXb=7m9H{t1<@gCJOTux9B9CwrtofCp6{_
z{`<CrI3u*y(NWO3BXK;)MT}FsDK(BS8*2clfOw0s2BfSQmoFKswTnzZg^?YPcYk>X
zYh!5!Y9Xx|&0B3ZJv5_kh2x~P0KtjUTbytz`+$E;kd5LA)*57^2!Er|#ZU7nl1)2O
zsX;f6jo(ybeCeqze6e+x%K%f~7fLjylQfSeL!Tv%>xAedw`<G0OBIux(^{XVAdm{@
z+3ZP*t->>nSV~)okEswgIc4Lvs(;KHYLExz^-^!V63nq^(FVW@2~(F)cQIx<W2haJ
zSvRhWiU~u{BiL>Ia@*j0Z5K?St5kB;F~!=QBf#8ZfBOuJpcIW3oQ^3pJE6ktunb^!
zR&54Gv>=pICo&wn>;Sr!Ljz+p5gM*!J^ORU0V!m4oqqIOQSc}&45Q9JzJJ{P`nx>+
zi+}ntlpNlj{uc*+VII<lN&h|l<L>9wxnDtui6V=N9LxI{zT7F0Jd}09k2#u|BVnI7
zOKUD9$l;`rf9`N04&I&xV^FRS#r+@{NY_uPaesNALvuy5+J=qhqD>lk-r6xj!&%06
zGa2t`x!nSyaNtPeti2WxX@8*@@(~-Bfk7E-jGV9tJ)`Y}Mhq;a)yrH~l)$-<xV#vZ
zuw66-|FF_W8Nt-llMW)2Q`ZB@6%X@@;WG2rY`mb4*dTh&h8*iiF-|YoW7`J96en%m
z>Tn!zzy^b0fqFM(2YV1lVF1&qV{fY%Nd$i)BaE?!c8Ux0fTKDl+kfDLxE_`k#%-$L
zV<nyhR48?9cJfvpsyKaUVzWH}a6~!OGzbwTRQ?^8-z3^?CGcp9jG9?;r<DpRsI*hu
z_I~o-n^Cr%-J!4??5DiDe?V*B>=+MM>+EIS#2@Y-z?e6G3vV{P)o(wfx1a7GH1V?|
zr~Ztgh4ItrTnky$p?}jio+3!XGRD;O*t9hYN}~D;)@B?(Q3q0c#GIx>Z)sjRdUDMX
z#PSgahBg~7c@!LqRH2m$^1R&?;n>7*f1q*G5r*D73d}28Il>6AP!lezylZIpDsK_n
z0hM=~cpYo%vLV!M2=zBDHc!TY(GK9!W*8Y3hq8PEeas$paDN)Ja8YXWT#Jh_U{E;D
zwztc{PHT-HDdnVh^R*9TP{Eb;c<lo?)RNi0_7Q>M%7K5rLV)3j(M*kMNE@0KnpMX9
z!<~fkCpN;nCrtw!IMxAYoPjR(4ey&M=Bo(<#k>i#yOK%?2A0F%+|0U(t)>XMx7jZ1
z9rzuZ_r-mMfPb*(*LjeH1t|MkC`|lZWas$vf~m-W@zhr3P}r2(VUMtjuqrj;%obsJ
zLg88SiR_=vnFbdt%HK}nm}I~x&ZFHX%eL<INO#jYAmVuNnv0=&LrGrJHQnSO@R+|{
zvu#accKclv@OnviG{6bmlmD8%xtYW<a|~qdHrX|E9DnWpz)fz@Hb%q}$)3*X(9zA0
zCc5VT)0C;lgT@0`B6x8)A{x2xTwL-LqvSBF@`i*YQVV+<%V2W(fzr+;jF3~D#BtX5
zzJ&4Gn~+8cLN!KChm7jJ%RR_=gRyDY`n+Ih8hHH;w*C8eUU2FfDK8rYf5w#?VwW^7
z$D|3{uz%jz0s_V%Ki%KFydu4)yPE?glv~MoWIHYjlHkktu2Y8?S;Y`uwhg~^xea2;
z$K@*u2$gl?Z`ao=pr^E>fPwAh)o5g(B;PNyHR0Uaeha?F#O}bz_h~>uUvZbUY>gzL
ztnG}XEgAK3U*e!<LiX+{zWSc-acv=Bh&KacfPX<*9Hp{;P$z=jlMQ7d*a%3*rS1Md
z#v#R_W<)C0NIiJjD?9~_6a1QK0~bveEyHXc_5C+3bZ4YfJG(ZGX8)(zs>-2n(p49O
zzQsjvkM#`pv4JZG?{;S*HXH>c67-sa2)y7rza_I8_Oeg6L+Qg;r?aooBy$kpZ6{K5
z_Cs~ZJ+kw4B))-UoE<Hyf$Yp_k8IVd<m6%tp=FV;9>1m{$8fDhK~HWu6QMG~b4ctg
zA?&`~_=ke=oiX750Y)3dikA^N9u${>Vi^;c_thQF0n4|>)*TT71GAhox7yepsR;wr
zDj%1@6$}#rG?!p-4HcKK;vFh~%W~|t5#8%6xQxpsT=WYMmujjcU5%@}OQN0QRT6vS
zyRbL@{WL)E0g~cNGg)vrLjv86!$$W(gJYZD8r%H%@bdof<x1Rm69haq@4q&nH>`ew
z)r(FV4`3WY^Zu>*`N>*zdc5et))MLA^?T5>TpkI6wt#0n;PU7hN_#>cJwU&|rr%w=
z+*{ha_?#tBd@7Qy9;u}VOG&0H2VIwW;~g7+>iXsCaIp)Z3GC}GKa?&XXaO+}xbDJ&
z93zN_V7*6hMZ_}5I&_1v)~ds0ycETuaVB~SQuktmNZLyuO(=??8=frLiZ&P*v9|SJ
zit_3R)qKMYM{q<0!-z+4L}UZ8?g0=3gYy^}BCv;57)eJLJRXFR^pX1*Y8Y|VR03~*
z(1qI8@7EC_)Qd43L4-&xoBA&*PO2Lmct;Uv6%p34v2zCzQN#^pFBu`itEMtqooaXp
z5hgMc5BUg_S|$@GvFtT|1jYan(FfsGOaz^=dD<MoMAGBqSYtw~#tN}vn0iU>GK)|S
zjSYAxLXErZj8Nk)fC54qZgr|^kFbh=8ALGzu9C2g8N_HX#SMwZQ_(9M0ZJ8dt!Phx
zVpVViMIE!zly(F~op^?QoS^7p^e7>Ea4uNpjR571PTYeqlBQm(E?2?mR$)v53ewa%
z1S9Fp5P!%UiNJ^)@<!5WK{)Y7t20+GaL_R6g#vGcwP>vM2XWMVWFk?GW7r6PYcWsV
z%GU3{dj!WDz%Y|xi${QAb_OQ8$a7#99eW?t5ZDN9q)}MtATB9R`E+ZHi(iiiUEE^L
z*%}p#qQve*$&5UA>`s)?aY(=XQ@N+4PLwUOvOo>n?fGkJrM&KI%j*=IMD5b)$H}j^
z_lG~`;WIIz7@L28dq_7}osx8a_^5ZIG=KiL`Ss!NQ>kO>8$btXos&9zK|)g<Rk@5}
z)D3cL5LSt+eQB?!w(Q&MeANrDmn!^s8uXkdUZ=s8p4vmgNK?`b>3Nnmp=e1;RjcS1
zd7Wy%B>p`yU>OV0GmOrQSSvA5ta*wxgkbCdb|<0}t>eu4f_|UN6t8@LDoOnL7u)<t
zZ~dfANy*Y*x#727i*0^y{+XWThFG;Fi`B#ntAcH#F(Jevb`T+?6JH(su!x<OIWksw
zU937ql&RcRe+7d&`^70q;(n=Szg`}l#WX#~eid7%5J`8Fi$K*bFW6jhmSHHBOY|U}
z9?b?yg4w5TCAYWt#&Uaqdl?LnyVFzEP|HeNm!GOH5x|$Iu=mNGws->xO+=ts?#4)H
ze$s3!x^7_CN6S<^$evsxV)ky|c{LZG&!|9qhm|i@4iCH&uG7QXfJ;Y)x>C3gTuOVy
zx*jgyP5e>ng28=;MAJbi%MR9IH#j1WYC*4U^(Iwu@VXP-(<p3zpsQz;cqepqdJ6Ze
z$Ta5O8C{DR7Y<Y$G8XI;#<s&Q<NMOJ!zl@f!!zqS(iWPhN4<_FM{np(SX!k)A@m_m
zmQhHA%E~?z&RV;HL1;7VtrOF&<@WY^$@J+iiHUtM&d>7_7WdD=r34lpg^^)Otn=MY
zUS_qcvEHWAFpo%o?v`_$hGhe8*QV;bOcHiJS1!7MLj{%BVI9ENt<X&l>e!93t>5RW
zyDK$%NwK@V&RuNB65Aog`JFJIhGPR}X?ehXU>0Y*^h;<g;O&E#1GxVmyr*H<fR|g|
za36SCjE_B=P2=X|#bpi6ZWIM#6)R%*5XuA5JBn(k`P@r?+I!MN8-q{#Mddq}I)g3O
zcJjXSP;5X<EqS;P#Ka~HWs46yco*>a0q|1>_#y6c68F>iY~YS9(YSg3zOmfiUK@?W
zc&GdgXd)~maC$B&I6b|$j~7&<^~BC)=lSuh<D>!-6}iuJvY;J%ayqZ2^iyhF7Yr49
zal<52j`9e9!PaRa&67NTUfWoe=Tm5pYX0m@4X5$gf?13U@R_>WVG;WZiAgMnQ>Dad
z#d60x5#-SDWM%!*)w^@`JCm~u;yBWBIls2flU*gn=GSqRoMR^=tsW&zR+ZhFy|1(P
zvr;aZdURT=n0_5=$uivenmeW4TbS2@or&Z=U2`XY^%EaqTytMfQ>7|COHxa<__iW`
zlWI5yF(J}y=YcXuKUEAT>kNateL0*g8N~3ZX?s+}KFcueDY!f3H`JJrwkV4$<@gCD
zwl46){5zCu&*goi?z)s?=K}NmxHj+#(R}Vf9+j}qb(2)P9@TFceo>cbI<5Qg``m@!
zTs3Haw=c76%q5VH&4!tTzSe+tNXeJgjZl>!%u`AB+6Qw#vPy$;yBSvp;Xa6S|Lv`i
zD{Qw=Qx&9ZGn8xFfIdV*b_2aH3239TQd+eMhveyeIzaRWfX<`?t4XvOKK}fznOynD
zu9ShAkSNZkd8l3$+@q@(oc51BP6XSx^8(X<+e<o#4=k0>Su5VS6z*zkzg7bK#?BdV
z)1i&`lvCEk>xgQEVUHP$@3n>PHZ$i%Lbt5jZ#7a-!se%tz@pWt%W%)IWOZzFdzJrP
z9tBzbPH)F8)6A~N)G}VSOatG{1=$`~s)Rj<=US;<RGNrFN^Di*v%@8jG_|IhWm&U-
zu89a?Hr~XjZo#U+N5c_s$*R&TwmO{Kvnq(r1oWRt#)n1hv@CH`7w*HVKs{cQ-!4`q
zg-L~SU&;zMly)7<dfvyX&W~#gFVuJ+KlLP!irDA8N$mQ7Z%C_vqj7e4-H8U`BHa?J
zT7OnFhuaOAa`kz<8qPbPqskHco`B4MSI$c>SskD@eGZw8PJwcM_#Z>OczmjN!E<UW
z)M|;y?YFTl>0fexifMQFR;zEYnhL|67T;jin}@l*K_ZW4n&e-=GcCBB<T_e2X-#j7
z73ssa1TvazrqVy)9}v7@nptjNie8%(NcylvZ*^s?H>S79!&DN7WxcwBjXc9hpwF}0
zdcRzwU3%i_FC;85vm^G^q?(TI(X{Ma`41s3ZjAU-V@Yr5`#GksgO~RI7y7`)!Y`+u
z8Ae=gaRD<q>x}&$CZWHxmk~K06qk{1KobKqG%=UK*a9bi8_RASx$&;A=-g~1qWLJk
z0<93uSTeGQAWnSS>>+UC%>wHnyE)|dQ^i-4Y<5rgNLmB!nPHQ~s$#ua;?oZkpFZ52
zKHa=~QZQNOw6Jpe^kowGg5&>!<Dk)#6@nQhrcYm|@9sF42A>l?z`0WRSp4qrQ{Uez
z;c$j{z(?GF-&zo9zLNL|{C=k2d0tqp<6HR<ASi#AAai_ZrWITfOp_kC|NhgrH$0&)
z@KaQukDqRSzUc`tLNYLJ`uz2VUVsuzIXM~2G}!5ve@<U+J{AZWGG~sy;r|~-Agy6e
zS_1jq_k8*XzWf#^EAOWNg{fc5HhlVN`YUd539w@xJOEb6Y`+DVPm+1(6WAGqR-SS2
z$3g7G%t#iQxDJ;#?Hwe4Q~cEFx5gLR-rpKd!litgWoQ&16eW#Q^<xwvfHaO{fWoIw
z@f`_en2U3?s&*m4Mwo&gT!+)bBi=``t2oq%TH*b*OuE2(;LI9%;YbDZ!u4?X9D-(=
z+X;sm=RO~TLN*CSnvBmmvLk%Dq~UZ(khJogo`>{aRgcSuB{+S5lXn$dm5zm%In|p0
z*``OQPnUM})ttQZLmE4!?>_Ix9~V384o-$)DykmRcS!G5+MgbMPv}PE@#o{~5`e5D
z`71}JB-4`Q0vOjtZC-=6(UFxS&U8H_oLxnqPJmC<{Br=lLmcokQvB`-P`jA7R=Zwk
zt$>;XZ+`bSt$^`=3WF<}1;2S)`;yDbeZ3ZLYqxOBA=0m_-AXa;#ZbHV9_d3RQ(4AN
zhqRktp&+i3TwP&WE626kt<oGwd!N!=v!>mzLI9elp>|8j@cY&3y@$93=e8`pt*m-&
zJMy@BS=Fw{->KX6ilfSXy%JtfZjGU}c~(Go6vWOL7Qvc-ZLn)GY#%K5ptVF}10g*A
zdBRyArZDv;K-UO94C7hhzln@0;OXH@ZJe2Y!HMtw^TU@veEz#&xa_|)6~lIn6b0h;
zHU2b)Nf5#!X7RUU+IfvQFprr7W;C3ESHhHqW#IF41JdH|+Et@J(WqzOK+0-y;DE)6
zWrJ!c<3JF9%V^&3r+P4BHE$7UDrnFzf|AzKPh6i@(NB8n-+3yo`qy;xWxDy8xT=jY
zmQLMCrSa2PlIsCi1p{F*v8y;+h8%v8R%Nbuiy`#dXhm-sq%hckLJA$TXercGKo^?x
zNJV!ZLRc9>9?GBn_zE97T5+TZDLO=SgxUsMfg&n@5_cnq=kK4trgd$Onc>_DlyX1?
zy44P4*@lcOcw#h8qkK9|&$F{9w#hQ8<OEvPLSwZC0Jr-%Rp>;zZfC+i?q&w&BzzY$
zdX2A`KmkU@gq5N~ra>F#3lAivXfCE}Ar(8PwC#L<D>d9zNg5?|^w#7z0Q61YRqru=
zwe<Ucfs(ocLElUAO_cV<bdTPs><LANqRl)<`}~%ml^t{PDB@c%Wvz^uYFl|b`g}j9
zqT#p9I|j^n+R@G_FJVD76bRK1ncf60q$C(@k6XcDhD>u#4HtUq@J-+fA@PK>jLjUe
zHN64*!XwZPOr8_($YFfbgfscb0rN{$vK4TDLY*K%Kex%f^7<|(^ySw8bLCA(Wt<C*
zJQWDE+AmfmX-accCn>rvLC`F?oB6I&*#Yuiwm>g?=^=)3pGr!S<kOiOO=m#0vI;^z
zd0OKOzk{6T^+6I$!-s~pe1`SA^tcF}BmhL_cA#p+<yZ=lXv4{kkvnoqD{?V|(Z76u
zSV}@N6<gXr)|j^Tm!P7v7U)nrs5(Kd%R)iUvP?u9G4nE2B?r=+=VmRm3Cg+A)f%^v
z)t>F+$ZF3PI9lz0Pwli6S>0BlRrWO)n0RZ{sCEk;!K1UxoK76hG8;-|qT_VdEpQEg
z5@%dxXP0Mlsg?{DofBeT`@XCfk=p}*+1}VQuY#zlVKn0}Wu>V>FP;*tOub1shs=3_
zvs8dV;1cI;N--d07EHfNo-)TPojuR=6y1!LXHdEXQ?Rb6j-H@U3k5=+ZOe#^mOj{&
zxc;f)k1M`2zs;Tj)_rrDS^2|k5d+pxSxQ^^G({FhJ=|-T^-J>3&qR8(nJr|0i2$Nq
zj9fgoo+e9~_Ds&Tj>h`Nb(gGs)OVezjCP^egEkt;{ww+6C>usj7zrDRv3e~V++*O@
z;`in;Wwxlm7!q0Kn(~dr%vMW}ODwbA#3g@vyD5PZm!L@ZTzIG??zD|3dTOhV;i;*-
zG)B|f;U)&{qEPK7-@M(#%PQu7yA^-TEA()M!vu_Xl>7~)X`2`?7zRQUV9sZ?ZmBV5
zyu>jjb-Se6$hy%g#er!p+f^cO<c1V^U`7#ch6STfOa6R;9ON|ME}DnBQdsiiCSuZK
zYHLKh<QuVhItpEwaBxIwP|uiFt`(afH>Dd*1r9EBJlGYfYM+v0pQZMHn%SpByHqqj
z>2bO4L7XV4!H*B!ui;H2?#-Np{-)8QOTbJu-86Pk_mpsnWC-A#F9Gy2ZAeH1Wo-Gs
z5R^3|bwH*JA}D&ul8*x|0G!oNR)g$?pXl~B-Tj*5XgL)U+-@ha68|~NKrPs#wsNr)
z`|!?xWJaTmd1VTtO%?Zl$=NFvkCFTvQ9a7#)^_1kESVz&cbrecGliPfLs~gLi9(9X
z)Xi3pY2Q#Z5>!~>Y2Rj~5gUmSw^{6~QuGB~v|YUGYWxmj0d19=Pv!FW@i50Er*hO{
zudb`$5JVwKk$ZsehonlhcT(tnHj&4VYp)E|HusXmV8|<#+)IsratxduQIksioCF6w
zNGKtty_*r-``wf0BdaGh^#&|}pBH)Hn%PA6#JE^$$EnJFhEaIxOffMtm36hpMY-hX
zqSenKQ31w-e8KK22iilY!bkuu!ID$Y$IoJ{A>EEM*dt`C_@n(+=*eGBj1d|$uvaM+
z=Ew`~r7Z8rt|&W~)$|=Ge{sbNrI4eaJ<LVtibbsDrkpI~mdEiVsc^2}S@e=@d{z-O
zKhY1?BOCf6NRmJ;XuD`}_Sv&!oIB`bkA4O2Q4(Q2T;$$mYg4C8biLVg>wLZ5bIg3T
zUEnyDL?tvaxu2c013Ia|Ti4n9194K;HHVaHU&l%NY9?b1!eq&4f7L_194{?&{29(U
z#Zes&N4&EX3{bIp-n^`LUYrC~<;qE`bkZhfjA$z76CZG`l7UeMx2qRbdd`$IeTJy1
zWaF;uMOA>BDb36GqAK7aJAX#&!eJ0QF%#(OUacK6_M|FIa-1<Q+mm|SP{;O}J!+vJ
zLtl@4sO|_|TR#rlfB5*w)b5&ZqQ-5p_oZmkb;ed{*k;NXTf75JKK9b4OlJ-s9Fu6X
z>AbkDs|u^lyH`;dC^^#mdp(8qlY-9mW+ya*X#Q&QFE%^LaG}&JIoR!Vy#j6PC7L8-
z#7j4c#|}oSFM4SDlY?D;jGbsux(v&x&ZBJ)aCvX*_ieMEe_Vw#_qJ`pS<Erfy_CfG
z(g9t~$?X%O!mj1=bxufPB<0%%a=vy*RZBL^tbHES>%&`zwNDW0j?eZG2TkR$|6s2_
zfJ+C9Q2@rE&v4d<LF|i92&@+H5`Bg{{#psGI|Jp0H_Y)SAm8`BZ{2ir_Z?<7_mS~_
z0OlM2(3ukFf9VGP@Xk`GlT7<1*D|lI?idqprQ)oYJxFP0rVG*?Z-jPU{F?7JTy>Xw
zo-4I&6}yH{*j#%PBJo5uEaF^YLScRLCS*^Bwd1x#D#S#prZ|21=*J0{3ews^+)<#;
zLCh(S%G7EF)rI=6mB2gA@ad^oxm1zPvn3NDhqLjve?##_e%WNo2Qqi63aJH)g&b$@
zbP)x%pM5byTf2%bLdyka*3ovn;hUG86ods7>LtJ5yznM3Y0wTfOO&(<jF7@({>M=_
zwy*X%XVSO320wPcd6~<BNo12i_pG>UZ<$vv=c$(HWsB=?w^<T0dMS^vL`q=7cYN3M
z#{ym$e`~=SZI#}>4;Ct_`yrOl#WTKXI3hPuRZ=nu{t!o$0^Hgk;fO+!ruLy$NCX(K
zxIyhl0n%cdBX|aQ4*noVK%xesC2d#&o6QxV;%#?>Yu)o;$EzH}Q$3<=02o1ZXMBN=
z07y<RjJ>jLvQZ;L&}$rIqozAK6qAOHY7<pWe|=_{Xtk2{cGTD;&?;N5ju@zB$vy+s
zECJvCgSUJ7`~<Z;4BGs1Fy(SX(%f9x1Y0HiI?bLM9=C&^8J@eFJ!TuybfIA0llP;g
z(b)AZhu4Qh0%IUA`-q4xPrQD1Z-}6)L_PP0TxfpCrOpHG3Ig?3_@*ghi%YptBA3!j
ze@ml|4gBo@ia>S0?xoCBA@GbUtI^CZ|KK0Ug3IfQtO9+glj7xztb(*&Ju7cJ3}PQ<
zp2BYZI)CtwDIuDl5=p<$v9Ve-`Nx*jl#4g>n&zSlS@+enY_rIsp%muDz-DY;7;O1d
zL~1g$uXCVQBIdAipth>z&VARH|0MHTd>5+SeFgp>@!#Q>alA4U1T--+IG2$59Xx-_
zZrd;rMfd&+{y^9n&hP~U0c<x8&|OjQimv+FG>dlA-|vh_(UP4+ieeQALs_J+$;090
zNML>%Fb|vEar5p#Y@$sbqcg`J2BiVyKZ4v0-b6$j2+i@)e7*&6LCQ`F0}x4xa-F31
zcQ=Gddf-+H-rYpT+Mk>hl<U2I?)!gr-}a@|Z94_Zw^bRWXwQg1su^nWcVCVlHc*F;
zQsbb$KW=_+&L`kqvY6O>ecWgR6Ix2nphufzUVpxuADfR&qL&7n^h*9ej5L!4(F`F-
z9aZ!57ck$X@q=`xoXp>@>!WrDn|(IF%qP)@A@PJl976DZL>#k?!HIabq)mT?xH4iP
z>}CF5YCcFE?`{GpU)H;>w0JBn?2r}Od+p@(iVL>EoJOcTWec3&Fs%t(-dnj|371cz
z6dJBraS1m01Kt-f%nGM|R$$m8?7+rTj}B#M>>9-`jQfz^mmt+(b%&XB7gLy^%WP5f
z*P+YNGQ~z02EiB#Gv4aG7Lb3rvRB5?v+MZo`hKpb?rJ#a)0zqIc{CT#dV(gA#f&S_
zq+mIwo~8!qQye&I+e{9N$2-BnIqS$>O}CBr%DQd5a!$EO*IqXf^(5kKeX|&7mf-3Z
z=nSTs3lFU1v@J%<oSw2G)q`Ex1kd|D=qV`bu%8(oi?K}L1`$DCX`6olSpm{ahbonQ
zwRAO|!<2nfz_EY3TvC-T!3Z@S@#^2@KKK6as=uRp{Y=b@alDM#r;-b-g4t(h15-=w
zD!h4kIbry3c$Z_Cz#B{MunyiNnX@FGeH82N*C>8pO+E}<AdkEv_lr;$L|TMu8s~I7
zO|zv)XA~KvdQC-Mn#zB@nYrI`6cdO;Ng&oi9KlBCCn9(jc=rtW^DOXL?tUTn#e62X
z`;uy0`~1GPZW}Lq>oKo$etS+fLA*S`X<LtAUA_C~AF2k=6k+DNE}qW>C+bp>b+!}9
zBQv-3lrB=!wx#0WG)|-JC7d>y<R~|oV0=vIKI!()t|T(Jul0ZSZ2!zX8kX{y!i?6(
zi8R&OoO;=(M)pdHsgc$98Tta_W!e562b>9CzT^<Y*I$zA{L|$RbBdOVQ;lAv1e+=J
z?@`Vq=fs{&)U)9%IUlyua@w>@MsaD`H4dR_lq|g4)rUda{M0CCy-s3T-H73wf+M;6
z>bFT%GJrf1<t0xm`f?Dn7vLV>62z~Tm^m^I7P4E%?U^N8Y%bQBNpCpALpxRT$*05#
z!^w=gw!VGl)7J87P2rM<e4sZi<ZC5n#xBcufUn|4XZ1hH?}(T2%`+1LG?$<bHx-w`
z_5&!l<^3I#2n_Z&0KL9H{SQa+9uSuiIUW?3!M-gM0W+7NXbmWr)dn7*fBd+YU%vk7
zoh%dm_KU@*Ac94!`&pt=!b5nn#tJkl&GMVSypN*aesTWEir|ztTqL9gS)+wBA&X!Y
zann(>7Qn~&SsGbnE@q}3jcvHpuj^rq64r*(K=A$}I`_@@MR;y6v|dQVg-M^JZmCJ%
z23}UPRo0PG_{e{+>$5X8e<}weXne9(3X;wuFH6lg5n87k6n`--=Sd@$&+EVU^q@f~
z8JZr*rwsY2CWYu9=qLUjpSkF_F18?#G`Q9h1fh&(eSh=1UNc(Zkyb!n=fT`sv;u*_
z@=@xdX-7^)XkyZEk)LVvtq0OHCj>6_xAj0|2nFLfa=jIlCF?o$e~{W7xrjDgZLswf
zguyuL4W|MKHz@k2^&mkDgML(ig5@|A1BJ1})P7kHWa5S-5%M2p%`GReMp!+FkQ11-
z4xANWxzNU(9t3FMBHt}HWlbhlqA+l5N=gd)EpJ9+UN9yA?U!2+lo8Sc>!~-u#rPD!
z5&SrQ3XCcF=n??(e{A9t6Tf#QXpjN*m-^I%{KS~tL;5!U{)Nfv5_IAZ=3rA>r;%d6
zuJ!p^W|QX7o}On^dNWXZc4Z#AhWYZ$<|J%QltOa^yfnVgEslQMNKGT=`Sh2rnu@xB
z=VL_et45kG)nU=J`qyKgF(parL4&S+xbuiQ`$8V4>he=>f2$I#M;{~eq|Md`e$3WM
zmh(w7Q%MRspj+xBNTc&C@_Lee^<l}h^4C6n6k9F**AkQfWMgeIEl@_&s7S;TVDWb{
z?I9vX&CTni1J2NVJfGJ((}G9gy&cuoTvQe}z{5$hdp4CtvxJHpnX~67ILp^X`Jr!F
z5Kbp&d~wI)e@Wf0ubuho#?Gxqr)`@$uBOeJZ`GAjwIqnCxMOY5sRr_1?MQ@RxAxq0
zb315o6x49EgZWg}>&q%?SoncCKdo6<Nmf*ZSCxAdH`3TadM3RCj))YC!;{fSKn-pI
z$0|a>UpsvZ4WK8isrke<Xh0EEGss0gY!DQQz%a_%f56J1pl(RS3SE6-O+O+Qkiqa#
zwgf~GFc0n_1rScNR-eXM=T@Y!WQgb(GExZYcLv<Wy}G{JA}&x+LM4HV>`_p(z2}SA
z!Js-fNrXd(hvwFcG~4U>#Efum80D{Z6gBlQ+E~Q0C?pWDD@*y$EqJEAqWLwhuSQV7
zq9k5Mf5{Y)zm%vOTYRqauEGf3#<jj{YL#8e<&iu)P1k-*?y|-6IJ1-oj{aX+eyPx<
zLa|{4S0UhOyc8Y%0Gk>d@XtroZ7}N?|7Pw$C$!-<+2szU1X9q|1kgQw-pL(W3RK7~
z59+Pt%L*|s0Fx%h5gO83pv)!m(2BO4?>lSTf9S$pDTEr2!|=^Zt*>5|f15z&G`)&N
zEJGi<f13VM#TWen`|~&%!(@n>(1$(YnEa$i9&bgXvKtSI|Ji;IXvg}fSm|PnM~DIK
zkhSMg8bDwM**P9XoajXaZnZYI9FK^-3-DWPn2}#jA#c?VIuwV40Vg7}&3XTweA<vZ
zfBJ(yG+J8ps>W2N@1-(`#TFk}|64yd<mYOfQVwv9V^-$@djio@4x8JpIqbTy8EFXe
zg=`r((ppk%;%jNK0@{2{Kl_vjYACHHt%>1N#}HjFPdss5Eu+j$$a7f1u8O586LXnO
z4U3vaEocf?HFK|(6d@acKlxzGL8i;Oe>ILB+^3QBY~mh(b@W@igBKip%p5Uh85?Ma
zJLJ1$7aWL#5~5^kSmRwyTvSRs+Gm4zwez&WV7|dF2J$*V*Kwe=l_=DRfG1Zzm2k37
z*YRT-fg&-&Fj#*XTL_9eV&c8zF1DZ&L@$T2V#$a|YmfarUqgzscp$)vxX{TGf4^gv
zLMuD8mH%qlt}u5rq4;b0-u3~l%Ym(6R%ICK^L?yuxp*1Cf)+>^f2yrHq{cw~c$eT3
zk<cviuk~Ig?w{g4*KpPquuz@%Z&@4HSr;CQPEUyUWml^u7R;8q;iujSjN!dsD(=hZ
z5is<l?D=V+>=x`JX3#R8jLfZYe^q(60>XJ}n;jmhlhmW_`gB{YbrFk;^>p*{(Ba35
z@4Aw4X+NauV)4MMUuE|#<3WuiN($-DDSED503}Rx7WAT)8P`5exr)5YvT<q@K^3Ej
z?>HUfN6h;<M+#K~MH{}M)+vkQ5MqUe4a04|<DwJBLrk_S9loTLN``N-f4nhLMGtL5
zW1l)&YhG*&r4j1qC3bkrb4q;BjH3YgqSEKvL5W)^?161xN?qnh9m}=Ionn)^wHG!<
z%FZJ;u%2L^rKW>NRCw7cvu;v0@9PWLvzFzu4Edk~zN&Q;F-R;$KH}#F^YQBF=Ra(K
zvKN8ER1A;cZWRYH-bv7oe}Zrj*54`)Qh*T%mXET(onV!a0l3uWc7i2Edvh8HmL)GF
zEJuWLlxpR;i|KijZVEi2x|oEFen@q}3Pkmw>SDoM20vHbMMX{$bqL7kwoY6MWZHUX
zZGS|>qbz+{XLJ<F2jLyTQM514LmpXUUe1rnm~v!H=VSw3&c7;9f67FRl#g3ib9G#i
zuZT*9747I(q$-vx9X5sbKxpC7y`<t^au@zEw~vdZdH<+=yo{xiNBqv)bcx^*zoXub
zve&!(63gi!n*8Bb$ypJhw8(0R>691NWuJp;#f+PwfjdKk&qX%&afTjxnuu7jyFcMJ
zx`$`|3|i~A1DxmMf426`L7N)%DDBq8IEC@Z3z(iZ)Gpp1g0pH5CGoP+HKHoh+ak6*
zO5&iW8P~Jx3NtpGcD__&k-4!%-&JIFcs`y^&!H`!Q3Fv~@)PT)VoD>QJ1O5?WEjt)
zlbq?sPYwa4Rn-h{**ju>+E)z5YX@0_>x)Aa2$W}f+fhD?f0Qmlu5ig)C7O6&FrLV=
zQTCD20i{mGrGDLBPjbY($ALD5G_MHasnK~(B^q5z5JmOXiC<^Vl-*-+93%g~YN`%C
z9Xb!|!!^t{QVB2ZxW2T-&1s7t?yM5AnQk*h>AT!Fcz%G1a31rCk*mP3Qyf<)wCN7W
zcx;{$MmGx=e{1j!L5c{ZBUmFK$#%g7{UuB2U{c`*OgTOKaed7u0!SZ$q!Y%4I}ih9
zzL&rzwz=60l*J=@>_i^8p+U3{G?(7p(9l-s2)24nY;I^2k*OxG0#z)Jdp@z<1qF&9
z-g1!-`;Zf3TQ9QRUsdIiheWLZyWVLqNHfu_?;T#We}u7#e=?C9H#=$R2wX~T@3Tnj
zh_w+yH~QAxj86EUiEQpP5MYieiivFQG{i%MD4_-MP=2T3nlVTs7wO;$PIkXuOuVo)
z2$>E3r^W`D(4XJ;K^7@n22ayho^UwcoJ^$7F&&RmJL()st>IET>P#L<m>-oB;Qd|d
zY@{9Me|Cd9t3eGr=4?~Ng-vbGoP#@0ZDh_DaGSgaPr^G6tcU^90+3>@3-%TVR@5|Z
zX=M|Ty0fr&WPbLdoz8W9sRS>Sez6OP1G;gX?Ir(69PuIJMi7biA>%sAX6PrzT`q{=
zppV(9m+~939mrLO&9|e@PS{az-{*IC)~Qf~e@cOnNq+bKBUno`nm%0qXn%c|KYolK
zH;Fca7GBx)-LC$oNJ?(6eYM+dFLpB^Xy<%#cHgS60MBai{N|44x462byuQJE$uWT}
zr^{O=wW1%&<sNhiJ->FebEJUr0LRM-Moou(z9-nZ-GhgC7!PL1>HBGJz$_x|Mne%v
ze-Z{zcd#tRcbhbgp;P7c#%%>uAhbGZuijQDOT9WF^X-Jp-#-2I@PKzrRP9g9e_hl<
zB3FajHxVtwJE2G-Mudt5)@KV^B2hUtYB9Txo1GJTI6G8VlPOPz+QLd3<<uN|%M4b0
z%wgNUBIK#<9S9Q79xa7cphz9;tyAZCfA|R0dq2oF8v<g&#J$kkq(F$kf#EY-Cjy|{
zZs9bu%_~arK6qd=t4FEH4cLVYC|p0ipEBjlDvY63)dTAkBpv7Ssm!Os-Ip$M%hN$^
z{f1Q^l%L%~Q(?b3Yj5W|!g|}PbgFRQ`BVrLU(MY4u6Av_w5kVdbvAh&^^t;^e><Ku
z9rQIP#S5do9n%G{Lm%C#^hO_swpOI(GkshgkV3p01O$N`;+^C{*(rfcCh`Fo2{M#m
z;tjy`+K^OtGE9pjP9^?3xrb3u$yhV>N9KSU<%vJ8#)$80TVw|j*xf00UFCH%TO@a~
z%_-1z+r#SKoawKJ4mq6Zoh!?ae}^1apX2`$UJ(372W;Lu#z>?>5>O>5PV@}!y+iK^
zkqo2k&#8%u#N%TowR?-6*aV)&*&I__on9Ioq;_1J7`pdosZDW-!i)_kKVe5i*|Il}
zh-9QHIM>ldeH;wg?KFj$ZH&d=aQ#oY(;Kwe4Nv*zc%w7rf!1o9b8^dBf6h7j*xZ2e
z<QdY<9sT0Q838#;=aij^p?SCy3pK&U06qZ+8E(Znifx2`U^K`0rcGM>r|#U#&e;qQ
z4+S`)C8@~EL3z2K_zm=L;`Dqke{+h1t+$$_J&dOJaM!4(c{QUReT4ZKvb*))PAYS|
zJq!Fu3VP1}6UniLBTdgve+i4D2ix6qRSb)Fvy44`eo^L`%~tpGUfl;QJ?-F8o^^QW
zP9Q{IGv?MEVTd>VxyZ%2j+W!w2U@xZ$XaZ@H(Yva9Q~OTty4&TwT~@)eE@<M(mfEM
z$cpaPE2@V`3>50l0Z6I}M=mn6$VF~*@tB26eAtwODGEhi-;3<af5ASDq6OG{$(_in
zf?;Ae%E6>FA0AOatOt8i?>NxrIG7-CFzV#UrAn2ZFYeCbhLbp1zVZBhuox$c^dwFe
znX+5UU&9u;7bnm2s&TUGBV3O$#K~FKh^%RT`g^;K=Hh?z5^mkpF!s0J)JT#?y`%4s
z<%c}NEU%pp$d5h;QrnhZ=3Jl~HKBR7rsJ+iLG!Y|C(~qKd2OfkQIFB{Ulqk4dJB`B
ze5HZcLkrq;z$>@EE(gpEOdwg|{D0Q)=X6=Tr2ZFP_4D)BcmD%^&B%wB5jh?dm*Ltb
z6PHjH9?1bew;C57mj!=tuGPVA;yrH5xV?G>&J;7cNo$kUNu^^RMYu2Z713%;XRaf+
z=6<yIOIcUyc_Hhf5m;&DSWNWU2D3!7*l^DepK7rNpe}S)om0C`w4TPwSEOg_FPr^q
z-cHiKQUg^3#rm;)bYZuw`_kNIIc|Juv;6nacguP0U6a^us6l^gM5CQ(w*4CFQ#C9`
zRM#W|uGHb8H=pab8DBNN@fO!55QTbMJq0?sMrp=*i&q&(s-G#o<XEfCx`&C+`|5WJ
zl0^@RA*tU%+UY>2N9%XWjTXa;`xFD+4c<3YF9Y{YM>Ux~0rptQ#riGYNsv2GiG1R2
zW+JXodj{O`eaC;3%<I}Z6TW|LYR;v`5szsCb8wz{>RP%}c~-vlj5qfWRE{W`+z6<a
zbnz5SsG;u6oK9g^wp`@940|xJqNjeoBdAO;EWqi7Qv5tCFUZ40aJ2WdpjL5}Yj~0I
zv^J4WibJT0)RH-%MkSobHKevIfG5)xEe>e|o-}yefwO<5wJl!}<p_Mv?WKWm&Zm}F
zZ#|H^IrG6o!f1F+_&eYU2Bz`yR-VvPTlECR?S3yHoQuVfC%J}wR?ra`_l{@|Yz21k
zPIG1V$x%w%N%l041Si~>P+X5As=4u!TII2;AW{mZ!K5iCkNf?vcr~H85MdM+{~YbT
zqeNdlnaqFtJU?&o&aO9D=VGZ^DZ_*vPp4}1mvZh{T)*=Ll$63dvzdt$9)j8=Gr?NC
z)g?2bFb(h>r}OZ6CfVVJ!Fd@;{`wmq{sBM!GDJ*t_$zjOuhSq1_;*ZwRtQ6gii8Cm
zQQRfMe2^?aPLjxuB=7<i<d@fz*v6R*@Qd@8$-;kf2Se#zOz#Bj1PX}j);>m0?Q?*H
zw!?0jeHZLAe-f^Isy)qY&oPFM!gf|u)VN>X6+Jp|aS=X9k6tm^z>`hh-$&W1*MvN5
z)5vblS2P0KbFjBo2!<#JJWD3W3ldW^0ws7Ssu`(`s@Ti&iY5%abMZl{uwWxTOjKbG
zs$zdP&nvpnAd8kCqzeO-nvE+9-HOlWOM9#8*v<BeI^f94z4{<^;N9C1-k9~0ZnGv=
zj<;!Mw*)Jik&s(R{Xv?MkYWjW(P@rrX14@unvqzoJCaOXO=4UAgqd?tc2E996SWQK
z9iD+q7%w=6RJ0rG41v}WWPFl9^v_cyEIog;XrE8HY^q(@<~fL#LR)9y9f}{(z*2Fs
zb7d?(htH0!x_?R`RchKfwj+gkoE>Wi1*zzs#~5YHC~(Aw{VejEooiaNg42M9a|>CQ
zd~R8+bZG_YhoGZIZtLo_wU_b)OTk;kY(;03Y7R-$6=a-P7NkU0kKy66PFf7Oc`|<^
zKD#oD^EOpF+u;5G7kIDLOr@CB9;yV_%yNGj@68a9;Ry_iC)2LwYj*2-aHt$A1!Nx5
z$*w@M>`Yq#qXNtj%H6V-5~!S0^Sp}V(e`P2IL=p;1f&c~s9WOO86qK+moRRNcTybf
z{Xsx?ftSc^3t+e8K3<KWynPVhT~>bt>jPk62VuT^(U4Hv0C=Bw0p3+t{d@-iZz&{t
zE+x2tMhVBHFtqpdS&MymcSfk_(j3IQO<)ePDk|Xu-VHI2quG!JU2tp2>Qw0OJ2au_
zs=K$b2t^1^#J<ALOMD;{Rlm`c4W^W~^*sDe3m$~(y1gH`U@Z*-ACoB;EVO^DcHeu)
z(?6xMKqv{kG4QgCta}3|@Cmm&p4^c^F&Fe?{`q+omrQXrS?>MMfhfGME5$~=EA)uT
z^v#aAD@F_0N0hxqK_Oq?CKL}B6m~3Y2J|T>2MP)$Y!Zr}F9Q3dcYN@7A;@vXw5yNl
znU#H+>kUBv^|lZsGs1UaO!t4c-141_OX*Kfq`lVDJt{qSiA<6UAa1;7m!y5D+4r2l
zS1hLkCgnt6z8Z4+0rBr3l)s&DPKKQRPD18rJwIPe&hwIZIHWbVd<;E~ZO_T`gC9LX
z{?xkr^*d^QzM?0(lpp7-8APPjV3Q3ajQ8A3#sYd5co6L$894Wu;EsQ-!AIm~lW2<D
zcngY%Hevbt4I+j7p6)`9uasom>$w#^SA}e~%Ssee9AkT;iWD@jk_f^Co_n|H)!}_H
zDw59D-vA0%aR7Fxg4n23tov&_qtrIT<L4#}9dP=5A_u9*$II(YVxsa{M1Z_Mzn%UE
zgRT(*3T19&b98cLVQrTiIUWiDHJ7nXDifCxL_aB)Ya<?*0pYg<Bp&Mpe_5f!9oVD@
zo4kH#Jcoo$=sqEyLkR->UGdyu68mPR$o?z!N%5Q_Y%=3nMZzZN@rmoh7EYa?=f?U8
zWfa~O9UUgIPZ=XwG88`vsT3iVwyG&5xK8kC<M<Fg?LFkQwvG@ralirZH4~H28?|^y
z$k1{r9vV<lnL5cdwyZGCe^p2HDj!{9g=Wu>i-Rl~9@Ug`fv3B=8-x^@<KXqa9Q5Jj
zC1@>3wb>vt7rt%fX+X(XR-k5lls>JZ8owiTiWLty<E$)qqAxPzWts>sr&Z(05i~1$
zKt{C)MCnu=Q6D&tbpxA+i#adLsl47lKeVKG&`^G!$?$G_1zU%se{2-cr~M8vnCv2_
z__BI1$XY)zJJ3i9FtVYp5n8NOh*ds1ACt;4i&)O{7mk^fRDx{DAhBjU<;-cjFAirj
zoEL$(#ztH^gw2PVto3#kRIV_atAUzu<}#FxEHfu2U_g%t7`M!f*Apku$<{G~iwDrX
zR`TJSHJy-w7S;i$f4p#hRV2n*vxE4CX`T%I=);evKBt*BCAh^jHvC;Bh~r{=xi2r;
z26E+wK^uH|_%50bxUq?*jZ<)~M!R_;e>f$W{j_rlN%H)p<pk&LuLAOyfv^3JT^WH|
zA(vjRr)x&nN;nOC^z-UnY6+$UH_B(&=z}?&*Gl7s8c#0tf8fEK-ilL#TU&A3I&G!T
z+Zlr@O-BW}u1kzXyYcT-Arg%wjA{e}YJQI7#gliyTOZ_MqZHZ+1pK7ZvLpN(dhK6P
zT|A>}hx~C>3(1!?y7pmNxzw_MDutzZo>RDE+hvcFQG~boeP^uz+pOSmxx?OPQcdoV
zTM1=o|C`SWe;W$|PDhDojdeJ_bnuv}P6y4KJ)5sn;rY!<{dcCcTIszUGDenZ*R19N
z`kaLUntXt<lG2Qrl@{=Nz5u6e7>PnVw^4wz6_f_DjAP}#w!^{rNvN9AYjV>E$}2OP
zF@jj-cVH~;E~au+_SLGnDbdB)t!Np_Gv&Iy-@@h(e~a*4{E}&hdmJ50_+>c<mmHl8
zk!<v-qjP9<AB5qnk+0`n)?9PPRx{5_r0^9bK8~01zlUN*V3n6gb}>Q_sF-@-9sEIs
z!Z^bjem+LP_%A`ek>JaMJ(cPIwwSi-(;-jJ>Fpal0x*QRU41(~FyWl{*hqR=*J8Pd
z=BhCbT3#z;IbZ>^i6(c{m@u(`1yEBW@^F<kH?h+JWX_c(OE{0-4Ni01cu1jne>TY`
zw-4P)VKVD{t7lrt<Ur+b{wEC1vcq;@E9LA+FHY6g^84$@)Bgdb61D-C5jh?em+@W=
z6PKk$H7J)5E*_bGIX(XR_Va-!LIr+`=JWaO@W-JiRU3z-s^RPVfqnobm~wIuB7Xk)
z`|$1XrJ%^DIdk+4|Gv)PG}bUDE#ds-S3dj$fBYGNm3PB`1L%8MhYvr7zvBW+h#lhr
zu|i_~HN<?7%sZdZRuo!!#=&n#v63<)nQ_;#3eLRl(FSXOassW18>98enotK)E5xmt
zN95U>&=NFS0-yzHv<!gsPUC1T8tq6(u8z@&!q6#$KRy^P=@*@Tp;3Xq+As<5Cq+Xl
zKFgEv7k-SU8VFI*1RG(E%&w?mue=u2parv1e6SusGwB>6*Nk(Y^&p^#3}u0@<ihx<
z@k1iy&*>e1omM!mY04>q`{p@y;3|NdV}sqK?>2rPf$NpB-z}sC;R9#Z$U8X!0OLB#
zUOg9^d`h<Hqiu@M^h=JSVAGSKPI+GNm@^45kAO**ZS#mImai%5K1b05&!fChA<h#>
zJqxM0N#C3mW%iscY<gBIez$@a6fZq<T;0j*lw#a}mt0CZX}!)S(LE(m%z2%zUYFfS
zCvbDoNhe<Sl%`Fcbe-^b3u!_4!Z8PMC$E!|2_^qxw`-%RTkV3jogwzRqFJfU-BMrB
zCY4Z5-^rCwtHP?{O6GC`J1@J^If0)S!AGtH$&Da-;4W9%Eu`zh*WNIHy>7k1l{jZo
zz+dWrO7kYZQp)?Ku^?p^%h|h_xON_;UFrm3s{z^-^_i;Xy&gPZSF6C+34XVL76fl&
z{dyPk*48oUO)>9ggM$C6`L>&tV$LC2pVDYB>&r$SHXy3O!Xs9ADVR)zf|u+Lqs`Ge
z9qg9Xk`5%}@|F$kRt)R5?Ai+4vWq=i_P#fNu%bUgypYl49j$+~A$~PRLdpvY#4-Lt
zeG;3E#O1CD#k*ioB98>M>!fnX7|ZO&m@51wV@eQfj0x?YM2jD<-P>%>1-xfQLEm$c
z_^a2s<<enY<B6zT4TBbjIh$oW%OQP!PQeMe<>xpa_)LyV{PL&3-(&%f@kx}5oXRYJ
zkXL0Hr38qk{Q2|y(2e}<wXUR9FuiUi$-Dkmf;NJAZYOm4YDuIy)cR&RoWg?1_n3v4
zr^8tWY9gN?-E)!Vyk1xJ>OgU3a9U?KnEG&Yv9DDJ7CBURXw`x7%)?Esy3Ht7Ql=gA
zcglibB$L+F&1x*IvqOc9nBFPS#X^mLa+XS^a46wyawZC7&lDb$5zCrC!{!w~4G#t5
zs+_i)@dbGTeyWe|xe3iQ`ZQ=XK2dvarOaB%@^0D|cveWWm3QJ95R<8G#*HlRCT)pj
zVBD#j9C|B?l`%{y+z!G=)GpUj`E+4rJp(WAl*bkBzpT8T$aA5Tq6Ngw-{v%bml9~V
z;-z$JalSAU`r1M%M>CnmT`J2|ifXA$Yhz+gld+U|42Ap!b}!T=3inO6o|91Jc&ECg
zTWO&#>FK(kA|?8&<)&ZpLcJn|B%_>&?bb3yp+2;Rk1SJ^MpC{0_%cNSb!F`>QEfAd
zm6R#!Fy84%Q33UJ_5Wv`vY+#R1&LB%cK$v|TPf<Tn!Sa}e%cmz#?7EtcVbymN7L<Q
zc`s>894p*4h?_imD~gpt4AEMUAJ${hZg5O{Yt&qiMOxgAa@F2JY5OXWZVrSV%Jbuc
zg3#^APM<m@u_F-lxzy8<8imL0r~!CxcHL4tVvGUFM_~*KnQSuJW?Kn=D4YrAtM?y#
z?Cwa=*SVSGYA={Z#=5n%8Hc;}jzwG}XYg=aVm5Rp43=h?bc+JY4T`2sY6?!<tST#h
zoyw`wCLK{jxI>4^jN7XTCzNmbQb?*p=-v^Eu$)8`IM-OnIj;bVw@+ff4(+Y!*Cl=E
zM=26!0-G%v?YOqf)&&@Uh;%uK9Ch-nItq2pi708R;Ln}F+&Q#xoGVHrWz)hm;?K!0
z&xcc^!(I|)6_R<JT|1TaGzaY5oa_C@s4X3CeaUloxyM&pLdVo)k8b;V$f*dHLS50&
zb*0;6q<Ok5q|_`Oer%<k43m9io-<6ifJ8@($9U^4+-Rd)Z|6XN->jCEX1pFE(4`{n
zX0(%g!r_url`D<NG%s?4pA_48*-bQ<0#{iNE<**)L+p=YDV#7CPZZUvvd*)wX<Kky
zjkQTW@i60*?D(mXW5m(+z=M{{9G5YO)-EHR<;1Z3o-On=;w4#DgQunBB}=3~f>F~0
z7GH0PCHzk#yt>|h7YnqM$W-<v6DyF42@sP^lJN>T>HP<Z>D<|ujosInQ{Ws%i>!^O
zXrFDN8=(r<Wes_pysgMYm71AMl=0{q?*on;#q0c2`!?pi(3Q@QqfM8YP0Y<V`sNPd
z64$3%(ub-b;^;lSu4zUvtxb$@cXl+Zb-JA3itmp+Rc6h9*ksW)qLon^@#WN}JL%bw
zAHw`eOpNDEkm+zU`7w>I$dAeHu`2F`AT%%|!q)kZ0gIH^Ayg`dP$g<C6T>iW-SEGG
zE^~nuR6=TllEZ@P@af;*zy0*}H^Fe)e+U%!-&kY`tTErytHK&iYKs+Sd^Zfq@<;>o
zTp1X^F@Yw3Y0Jbi@O3x<?o)8%=$g?VH0l{Rkg}N^Dp^Z_d4Xmq(?A&OXj<<_J6NbH
zO9Y||8T3Y2(p-AQ`FR$-($l_YP@MJ8;qYxZd`VK(#u!U=sKY#jpT^P+pBbqN2Er=Q
zN;VHk!?*<0V3LZrXayHWTYf7jLsf!84ylzR@LO4bt?-;j8ou)o;>sBEP$8`wzaWQ>
zR~)I`FWha2K>_Y!!oE!;=|&0H?_b~3ym7NA<9QGm;6WSw;yl`JXr1I8&D;(nfoJG8
zFFchi$Nd(m^{x&9F85_<$ccPCRK~un76t~b)aqIgM!(@`gkTI4@w&&gqw~x|g04sB
zN*_yqs>{J}Hu>51h}{>6GGp!{c@3`hkbSwl$w&6COF*$axpEw@M!|UR#*rEm9C}fy
z`KraJbpNf0#*VY)XiT)3Xhoejx7P%$a;bsd(bsip=j^vy%AEaHTa5PmuVD<9a?C@M
zWLDhO5Q_e1j$&^m>D#t&@x8rWzqBz|O6#(J&AybxQ&9xDoZ6S#*Khrrj^)M+9tN0_
zlLolKgl*-`=R!gNgL*zV-ms)2gvn)&9Y-CO8FZ^|wTnj`WSO#uR&$q1&XWz=-eZn#
zwd=<mJ!uQ~V!|#uABBXnKv3%BiL}v3E4Hj8jxK9!&q~$%C_3$6gJ~9nn-?p1+zCy8
zu#pJYsXe-3dmVbau#@=4BnTK0ybYmcQs@cB&-2=oSg#o0Lqn@JnNAM*cIGslhfK63
zCWgKVDY`!=+vpfn^9&lt@R{g1yDaj~$;c?>CEII9aNSQYp3emgX$=M(JK?VOD_VLk
zg&m1S)6}4Ea1Ae+uBQe!dd3*pF2|sMOVT@?qGLr|jE_=kW@S$<g~uBN=_ud(ksxyp
z`#jUO_Lqg3j<v@K_eB}iR8n!YbSz{I8(TyT63cSY6L;G0p3&z!d>5jR_RJ#G7+9aT
zbCJBfTlIN<)#SJ%U=_|#=Wjk(G?qe~s*aQoAU3J;+WCL~iPg{l0OHy`%nD_fgc2wb
z0x>q1ffPLympVBfy#bfE_&FXN2n03ZnB12UIUW?3vBEhMm+O}pDwk<O9w&b+$q(5s
zzp8*)RJ+QiDU#yz*sYsC8r^(<xqiL;^l7MZ$}>1@zJ6~&E@=4+T25MS9DuS0ny(+t
zFK=3-mCsc^q_r{faq`{EkuR?XyikB#1-QIAg11D6pYnMZ8jN)E{g&?&N@1amm3xB?
zksfUR+%CU<{rsY{pdd#W96x`)UjFw|i)sxi<8AZXkC%9X49HJxh?T|A{P~~e_m_Wk
z913TxeEcSVzt6FBR(W)h<uAYL=HGJhXCXGhH-ELHKe}af^WWxQvcMW+uXM}Uh*`ru
z#=1cje8_Bj7Hopj6!x>&D49iN{e7;WRp6R!>r5WRnY6XcBWDsPLN<TGn^h3z*_l9`
z0mK1F9Ed*Ck@Ihs3+yhhR!6rmAye6SRzBiNETsG)ZWPkp@f=aQ4ss6UDC8QxC080~
zJ#wGUo<|j5u_uFa2#+=>E7ne!v(h?r1~oDE;&G$Ohh<yQnY3~ipQJO9(%h*`J5@d`
z+KS4?Dd)_ofZoeuV-9~4w(RIg4x1*+VB3}CQ2&}P%pMi{Ivt~23|!K2#c_)lh@MOh
zhoC&z;s$L(Fe&)ymeUC3w33rWn^s(Q6B2gapFjU-sx$xm-m7Cl1@XITMbuS=%G6Ho
z7phu5n-f$_R3({VJ|DkxmY*{nk5i(STFKXZ4_;zUTvAu3(m8)InD=sYoj|w3Y>Cr;
zL$PVSs#<BuP{pj3T+iD`lu&ir)+<W0GAADbMyH&ZEiXIgwwt;X1|!i9kL*%dssMR>
ze3!zE3Im-<YP(r%l#E0l!_R3`SSpM~^Z(hV9M^osqDGiF`<$B1MtW6CNBflHvaRS0
zL6Wt5lFo>dZuoz&&WB}N)7dIQe=4T;ve=l$ge_utM5EGfkMtg1ad6hm4YuThV(wb!
zIF!a<j)5NQPaEizLYce_s1%3Wwj#M4`rHR{dp-;)y95!hUUS%|nzc4^lT=2qUHXY*
zGq>vrZwzmJUF_tXB&15R$U#PO8Q61$rBHYk`C7?)?wx;*m)DrsX7x<>&+A!uUGu%R
ztWMP3*|s|ugtz&YM-t*Ik$uZ&LT_~`!bO*1m(a`fo_{48$Vu4`B}?hKqnWg4%l4VJ
zY|A?6&$t5iGL1U3vfmVx^d?=4q4R#*R9XzXw|*5WEBmB*yIMz;_AdTvy)}%bV;!&9
ztV3I*hCY96lco{8kqY>hK$=$fTi(Hb4co2ZFVM<&&Hh+?&<?#NTf(~HOHictBdOc#
zw$OCfIorSeHQJGMM{b(iJ#~W}fW92l$7nTJ<~H&LKNCgizIx|vUel&evSQqM^L|NE
z`eOZohO_=416yoEvOf%?eYBg!Pm}$JSpxi-ECGMT=53rM?tZYN(W>u-Zs?bwmndCr
z>2&YrNGd#t`)rdUs^~M~Sq)lwKW1$$Rn}lv1j5B(F&&o>%+8yXcpoZ|LTC_}#hi=V
z0_KWi-mTGWza8{Gx__HlN{US;@B`E4(W&b$mk0RSX~mG05^Cyc#cJiGc%2>MO|Zij
zsr7%>wtkdmd;bh-tOy7-=cO4`8hvR}Dq=GcBhB9t!;V;X#J%xjzi8emG83g2#!m>K
ztoIYy=EmcA{MMI8U1n74L<leqD4k^+of*njHpH`X)MFeb2NC5`amyandEtUtgA&u$
z+*%A{&bFUreu_oJ9I$M$)p$Lik#zewvI>7>CLF)bLj_ul8eC0H!#%=`-{}3Y1qKz|
zL_36uTkqmZGMx}iG&J)XN*V|f-Sv)^0JT9D&~X*y>}kU@vhGixLC>WE<Tx!&J@3}d
z)<b!g5(-+c<TH`)L?k#;&bvoN!kF7?Bp7XvM?obyET%uyq==!uBj&#|ma|5A^do=E
zmCE&PjE@Y)iCODqK(XSuo=1%C8<2DsOWTEgUf30bph#T0mgN0-{FSJ#I4)!u;}yvc
zRn&aluStpIYTb|dbxU-Ps7(u4&BM%wQ5uJCtS*=~-TJ0;asN(Zt7wcum^^auK{k($
zVXG#>Fo7PAts0O{McXqt%x)GNCDVVb47ulEt40>FX4Gqyfo#|Vqz8LaEpzOOf97xC
z%D0Y^XDRy}x`MPF*Z001F~s)~!gK71!AU0WDxDdOQsXS!*);(kn=Eo%@)e7CBQ5ZA
z;8yg~C`13CX5vO-1n;18qm3}^VUbrX;ZTu#(pv%Oqzhs{F$JGA#bKFOY+-*vF+AyQ
z0kksW^>r)jJk(z2mk3PxY_!H<saLGQS}{K>NTj3k1b2gERIA%O^4@{L+bnZfgB8mN
ze_nf(c-S&B4|$JwW9GbN4r{Px8LO=3Grh!a7TemV$d>N6D7&8N+<$U{S_X#x!AS==
z_LIN>8ifJwR**6`#O>OLBt3r^Is)8hq0mPFoVO7OOWn7SaXqM}CmiZFgHX;rDu5B_
zo+QPd_u%ep=ifcZdk{$kX{0St-J3g5&cJCT5xkb_&!g9ER$r?*@P>;tU6R?V)fV4y
zu?j5K)K7vJ%iAs0YWLr6sYQD(x(GH%U-Bq)vEtTFm0RDY8X=C9AK!m;uO8hFp^#`u
zp}|_N6DVYv`fH907(qt#D(ZO@^4vbMxLSA=Hdb*vxQ_?-EQi5Zgfm+PPN4);&2eGN
zGOSclPqyrK3^?XEDo%FmByKGIw2r<}DY$?eXI!@4;}gC|QclbWUF>3aP|?eSZJNiC
zCkcJb^_UOF|8+QwFD8F9u$JRK8dy=!=d%5>ls&6tSwIM#jNNV`OJ@t-zkV>{E40xT
zMx!!%vV#ZoHqpampsP5w?`(Z87I_ojXPzuiCkv%?&EACvLg*F8V-CIos~_0vQnD)P
zPQdCp=T!_B;)r9V3^jF3w+rYtAgh3@;PzI_o(8;adVo{))N)cl@ySDYcR<C&kB7T`
zP=?)^33V@rZE`s;ES`kX`uIT3E<@cCF$K!Lt$^NkLdk=kU<GzKtmE4MuqmSFSkB2<
zg6=us_EE7Xaa0}u20QkGESC{E9u$|+Lp&1$Gchrjkv#z?e;c`OWVtcdSM(1I&Eid9
zAV95SfXzHsw%BOxb>a!UiGSZkvdG8lUEMt!NZopmEY^Ke{POP$zkGYTem?!}L#m5q
zPJ<Pf&p$4JFF5`Njt`B#SO8`uTt0tXe*Mb1F!*!DAH}(p_+#<A!%zMEMS{Z>fNuew
zzgPud(>n2ie?Kp{Muc_v{gw9NN(8~B!M)!2upSUT;5(4dzkUAl#0gQrPpCfMKcD{l
z)T3&IVGh*g_a9H;g_6Le6BjL+xcvFA%O6jFFZ(lG&RqD0zrQsA9peF@G5Bc%z+Zpk
zmw)2RUvRPX?()C5^ha3Enb-dEpUXcGKCc<;02#}Tf6=;OENG^=Mc|&v@G3D@BQq2-
zNFf6>ern{54vcz7W*`$4fBs^4WKsMkBD3<h&D}ch4lDV4*)u>FEp$Ds)D<4iR(s;=
zD6op5jgYkTj4OW!Pj_UcHN)>eNmkk+B$H(o=>4nYwd~%L*C8pHgHpPgn>$iciUGp^
zIZ{%Ve<^8;Ma`t)EEImogS@yI=N(>1$-u);;)OI!SvT-P-^q)cao*zvXM&4uxPp0?
z!Ol=S31Sa33MvSeZ-dQ&hIfGbU~?eaYHyeqMTJ9@RKZoC_;iSplH*h5$!X|kFhRJ%
z9bpVs7-tnKySB6=tU(b=;lI`}f>Y9l*5gz4f09m#`#yrQl#?PPkXQ5_%Ec2!*d4z^
z+7V*CGU2IwdDb?*EoWW{BTrA`*%1>+&M%(l=%n~r&Q@i8<0K0O!F~F?T4zp`Q}ngG
z^Mr*adV@pIM88YC;2+Td$gBqWDB+mnzL(G3jT5A~q?NV|EcB4@e8g6MB}^f6B2q#)
ze-eMT@YEa_NbvwXwG?b5IEgN998}jS=50w78-YMn3nVcs4+<_miweXmMahJxw}u2}
zGaOe7KK(VGlZk``A*_g^SBvV!XC^A|loo@oRr+Fvf`ot>kuWlwpj*_qH@UPbo+8R$
zgjZB~GKgherlDHvlA1<gk>SPtwd(S)mzz!=Hh<r!N;ig`o5t<YXu$PSzr|<-Fs<wW
zsdGS>vrV-QlP>OLv9-ZX1-e|rEjRGTt@5X`oh@#F!+2_YWKbMigoD@2W0dSk@JW3H
z9i7y5hBOJiQlOY*XVX%ounQrZMv>LecK($1W?x!43RP6rH7e_Bhr;PxM_9Ra5d?UV
z^)M<qK3eA%rxs6rnLtR1n9wey6ikK8L~42YGvp}4)9>Hb)!GTB#UPWYgs#HSEPt1P
zPaZfd+cV+ljiisYW)L@ieCz(2c0Lql{f3*q^AU>ythK56(Sa@LFULJtG+g4hm*h_#
zQvns1I8Yu!0r{7EP#!ve_c?kbdY?<^n~{86?82rn(UZ)>&y+mjG{c{MM{Murmhfw8
z?WXoG;k55(R-H^1i!~@Z+lR8k9DUrmU>RmGes&P~wN=VwbINFD!8dU&9Q}6Q#<FI#
z5LXxt7Wac$KflUjvi{T_aYzGawvs$x)SXfW_w;MU?4ri!bqutBj2)8}&>Gd5NA0O&
zPJ0>)hpB<>$ysx^E;JKWxA=kXR$%HQK=;_py{*8)LYYw#<X9KU9kH@rx|lIJo(p%$
z-x~~W5_1D_SI+I~Tk_LWY!@N_=?+~{f+C}K^&_<a(^fyhXGIM(U<ojAY%AtYj5<40
zDVS5D5cPV$o2FcUP@%TEstR<a0&6vEmX02p^hvC*Jw~7|-xU(wB%Gb~5lDx5$M(cy
zt<SsN+LjnC<z1cavk(sAh3TF&4<A1S1(gUK!>jU-$=P$(7|bqp0hU^;^mI*Lwbz=X
zVc82kjn&*_<jI%RBDTrM)MNhjdTdy7JBRlqhk9u5Q>n9mAPyQ4f(0x8qG5@Y$Pd2^
zOLBq#f~9vHNKhMRvuzL<q6b*~WvAJIYL4YYw`=K?1`PPuws#t{7fU0Igdg?s(qBa#
z-O?b)Z6;|-3yu2F)gGW`nOsIdDQM<ZLNA^7-0Uiah$*ma*A~m|E>ShDx-uoUS5p61
z=f+sUwjW-96`D&S=Z3WfXJvYLt8KX1$ZO%wa`JZfd)ASa!Z27u<k8N!0Uo)&@H66A
z#yE(PtMn89;`1@jBievMgrMA$@)Np%!eNqN<Zx0JiQ~sx@{)TdxEf%c?+Z%BV}GKD
zbZjOV?j`l=>G(>EorJpI^$eMpMc3Nmlr!(u$zh;>WaLRva#yCN9foB5uQXaqZ>z`~
zjG=TS$&eP6E<42P=ud)TSTD^-fBaCz+N|8^8Do&avVA!9ST3=};ZS7(9@<0has|S)
zHhx3yx6p>GxL;iS7CpiV3?}2L+Y=YzuyByB-xUXQ5wCK0pmS8!?SVL#iCux%>uy{U
z<oSPpCHGYuGm3FL;Ib1$r7=v%QI587xEmGH(;PI@U)9}*qEWKC#{hh)k!`1+EGWIv
zl}}llte|)D1Li;tui&9TZ4U}Dl@olhoJ9`pFuYi-nA;ym4U(<Lb;ArI2h%hSse+9*
z+xDg%vjT=ws^%LOe(1gS(<bDbc0Hgn1Tjm0{*zo$L3ZCM2uv%vmp$y57tImPMa5fA
z`45Utc)_$X!$~)eZbDM;^*Fk55Nf{7$<1aRmMU>!Ju|R0jN}sWa>5d|A4=<!GtA#(
zZ#@2r=|=NpY#?=qaJu3<&1I_`4FnHE`fxiqrR~L6Wr65Qm&q1zQ<(EhN5se@oA6D4
zOvqjck<gje(*}zSZDANFdR%Kn7{`ou&6q4X<3n9*YTfiu-tWgx8q_-&AioYB^&{P_
zy^&sb3R&DeEyo_>%Fv3s{@v<4sgT!YauGBvd<KtDF04Dcz^<w(he(yDG}9)V2q${~
zz;V`af-=l5<H*69C}SSY;{1FtP5jz_1J|0n)IM()KW=i-&TUsXBQ>VkW1IAb(I`2(
z7T_A`StYyUA27h632g^?ZGw^I)vk_wT%W%RfuxE_l)ss2Ex<jdb;mzq*x#&8V^h6;
zYiOBJCuLdZ6iAHV%e{mlBt72p+7tIouU#E0{3X3PHKgDe`MvESQnq)C+ZLFA^;m8O
zh;N%?n4aUe8?=xrCU5=u@ALB&*6Rv}^KZ8ep>?d~JBVvzDD9cmemK9KLl63i399Ey
zWGLCs&6eB(+_U6%b+p8vXCg(R+EamVD^$m#!@={en-)EeV&#?vRx}d&)hBt_D+PF9
zUg^z(Ky37)>)!8`++%T?r}(#jxZ1-M<z1pt=f~vdnLiGTENANcmnzTw>uTvXR}bM1
zv!iUag9slh<8;Ho3>3s0I6CB~DOU~UR?KZFpADQ|zqzyIzVbDtTq4y`>Hj}xmqo&*
z6<+=8!GlXq3WeP%A$rRr*Q@B#Eot6#2v|3Z=*cW=r`9S#s`C^#v6Yp7ax<_@*@2n|
z=raieBL_F7{D>x{s}@Ut*W<+Hk!&!^F@T*>|6V-{(lXFcrz-lcq_YFqg{o6G0x6vJ
z5arREy>sVI8C3^$we-u1Cb_xcib={|dJ16uf*)lk-Hv<m9W1!<r-!(f?@*xZ&%gcN
zod})BfuOZ?&$~Dze;oOLGIW1H?2v!P{3PgLP94O1E^kTjOo~BG?nv@wux>d3oagv5
zG@zuLTEzLQ)iN}i_=MrBDWNow`mvqRn>z5WJH-&@c3w0dB!?tldix8bPXxv6quLWV
zwW^{gnFqij)b*~n#U^rE*VyDI8=)xQUF@+t79xbu%nCK^zSg)RR&qJcwS4t%;M!@U
z+-v%V|HBaSw)=4X=lkc={{aI6sBxDOIUW=NIG4fTIuimkG?$S*0VbCaE*=Dz99SNe
zfB5>_$ImZ35ejsN=KlWi^1qjnRHZGEDyM&afAKF+0+Uuu0P+6kf2QAG{!vh5)SOxW
zi~fFFz^RR9Rv3@-*WdW`Z+iJN0ZV76zXIs@vJRjAJN=Uu*g|X>4~QiZ>(&tSNib(!
zLR(QN=@>_MKZ>1{X~DEz#|oS|HKGmLe`JBQiEFJ?r%faSp(No}%n|aeO{58mq>ZFO
ziZlZ;J^OLe7Dajlg3D7>qA>K9!C${<F8mvn-uO|Vw<b(NdL{jkjMw%gdc#jqRRbY1
zs$j#cPVI^qj`C|m3<{VL@L)NhV!~QNt{CSo%Ym#dvy6B_kKTmolKtgb$lud5e;F-U
zuKbj@1n!#aZ2*@6T%Q{3I{h~BcL%O^^8PTBHkgl`87-b<1c;2Q%zOD>bn=pP(M8>a
zul`N;qG-~Uq2BVmU@>RUz&wIFX|~BDP&8k8ulpWZk35g^LIs>>kZKW9a-Duz17-G-
zF5GqP6#QWUZ3tdCX1RQl)k(>?e=D)n%cQkBlUR32tXQ%-Rjn?YkxJnDrjgF9?vkd>
zjdUOL4>M_l`Cyqve3I1(!9a>1Y<8_TwYyEwv@_tSDe9flJk0eCY5Ek(swbHeNfnGN
zrlfBNu=BPlofG(Z6TCAeOlE}90uPzeVJ59JUpdX(dfxg8Q{tQnz#nQ#f6FSqlgr1s
zvB6~<)7ht}xU!D8UGfAlRRPV8{7hE$-VYwI%U$667=M^S8;mzGe|?I2D`S~(x~O+k
zpwPdnzHPBm)H!(Tw=^2m`nHmX0z@`gc*Kg=f=Ndxyk>LgilcopILxap83@M3XA0PR
zG3+bZl>sW*;Yi6o_6BzJe@FaY&=zG8?$reU)hLM<UJxLS@vHBXn5-l=cTK#%3kn5!
zB#2!lkz>YKW;e=I;cqEZf>>3i;H{%Vk?wo;((T*uwAgn+=a@6IzKfu@PUW`CfOU^G
zl7JOX#CRrnX&EeT)8D@)4@Fz%=QtiVo2=JwnqR^OXDe{c1eRjhf9F~#Rv8-EgoLL2
z`SXuy82RN}T}jEHf5S?G4^6HFt${giW^}o#PF{Q@`n&0{lCmZ@Vg^{I!y1Ni#-G65
zbK&N)V0Sg_f~LTzHx&M;z&O^fLnE0{@c4EesS#6Je#YC~jbbNd$}!EKR5YO$WW=IY
ztTDgN1{K0$diEY9e`aewaLJSphdj<CWAfhY-TTO-#G(e#P{Gn&^TrQt1h(Tl-{1hF
zKwQ5lTBh;Qg$pRA$f!|~am!k`lw@n?o)6Quv1e(RrZD!LJtIvtSi6mP?)flnTYE;P
zo%xha--}|W8zZ#n>?2Z`wP-$HSXx@e!jnR|bR<a7tD!vSOUYkgtbYOT4s<Dm4lCXY
zxa66Upho)Id@1`gk;ZMx&SdgyDO+p3VhEH4_BG`5Yv9+m+q0g5+{|U9)l4f>ws3h`
zE;FpOk;{y9Jx}zMD>B3NxZ;f*M}Y8ZJY(BuQyWPE)PQ|tZX*>kQ10=$ja1$}!)N^9
z-6(caMuCIFlRm>pB!5Nz|IBQT?|Xxvg77Q(oG34)U=rbIW^<gjjXeY19X@H#A{yM$
zJs+lRYtM=SSD&)zdr|CkVt^D6^bcyz2rHN}qG`@Zm7|Y({UVI-cLsCEFcMlT*qktk
z5HV`M3OFrCZXe`*^cEI}){HO#qpFSsEy%No4kf+~7+rayb$>zw$wr418a2A_GA62K
z>!{pB;^SzauH4#5;Rx145eWw*u9>nrBsXa1P5m?{ILGpo_7dD<hL^c9*3KDtXVSdf
z`Ot+*RRVP~)0rM9p}g(R^Z;X)kq#McY7wP}8cK3d8*z5lh%>{di`v`^-1f4^)a<+T
z!6jj{jl8<nVSm@=OU=7{7AD|&i*$r>y-Zz8OIXEp%|=(AR%Dy@h%<b0q_V+RsaTVi
z^nFC~bOXQDc~weCQUXFa&T`qR6Mt~f_QIph#_DeSDOu)BgOp&lX%b^vA9?0w_@#7{
zT8M0ebsvGhan7PgN$q7+4Mb@yXFp=gYPkU&4(de<FMs&24F!!qOfx(EsQXBKj88Ut
zQ(wzuI}dWmzRb^)?o|1Sw&yR}%8_U#%Ll{~mZ))S`h}EXpfF3)*x?NjH50~;vy1jb
z&a~PS`Jk)4R_RpU^Rf9~<g2n}UjvsH^~yI^>fy|tb_PN+fo3*))-hlAj_8>5{OGYX
zv1646O@9R!>KP5YO6yg+1ZXPe<%}*@;y#S=Xs%V+KbX^U1WukK`xjNe_myLes`ZEB
zYs3?dyUNb9Tn6p-=MUcJ-l)ZfDY8~?on!No|8Bi-ofZAV+y~lIt;WlKF2M_D*ac1|
zzxUHs*~uH3pcHQ@lKR8Myr#?f+=V||L0HoH;(zWN70+g2+tdtJ&y4N~>R=r+2D*>D
z=Y*vTpC#CeA(-@E@~{gjYjpNPnJZVU9O#9#J~LLn^+}=45#K`NWARPf06}0D^teND
zZ9Ymb(BuVyUu#QfY?K(P3n<wiu;#YPXXv$55>T-_(rfC0-@uusu8#a$3D-GE%ArJe
z34dBID_l=Xh$`;neBT|nZYL)y%<5*xa2?&*BCin)Gs52B^-PofS9F{ya)>S>S1<Ye
zf~?tJ1Cg^e?9M4<Qz)aWrhw(McRxpzcc!VBikBU{rs?SF_9EL%dgdO)e)o`}bAJ8S
z?<lXD{Mr|Ey35%_m+Od>>*(CKP4!pFd4Ft<`{vYSK5>5w4d$MEO*H?Ei4|4RoQznt
z|Cyu6bTQ9YO_M7a(R(qo^-6NF7wpK;aU|lFR->Tp;6wa#llsO(HMhM-4sNCJ63ua*
zF8BulV_{42EJX?R@e>uA3L%YS^O*Q%j9C8)L#*{JapnK;)G6;qeo2iJ$Jf5pfPWxh
zQ@@1Vh*D|Mn_V)g!`Gw%+mH-J4sDaYNi%m0hBBQuE>G3Z30rhi>Zqj}F^aYT!PxCB
zE`}HLebj!O4ekENZKp>lk~R7N?h^hd*5fF*hKUE#-J+ASu}kw%ej9(v1s+Iuk5KCB
zGXk<y+KT*~HR;mqYSF}pJ9&`50R(h_J6D$xIUW?3!6`Ttm&%+7D3|eJ9y16jQNZrE
z@5R!WEMguQe}kmIAnBns2L~W5!r|@j!!J)#DofXiE-fV@U5-9`x|`=m1Wzk~eg$}b
zbQ+$9b?QRbk=CfVjy|9BK3Yj6BwE@VY>4ZD=mY%*=K0sRA0K1}`rXC&gBK<^i*qQg
z(}HMLl*C07XHC@fD_z9h&@)2_452kZ&s^g0p8VDre_?f%KjPMeTl*q!<dJB);nL?!
z{}@*|T48MdRTV^Kufd4xSw+dyuMzhpQm5~->`xSJ<)?>rRTg&}lC<J7>?fWB<ztZ~
z;T$8m4A;Qdm8W5hJW;}ow%WgN9VD*nBQ}}QUq(`zj^cTEnzZDHUF1vAP7}MJ7nbhD
zF2`q+f4Gx0%BN2*(}1iiSAGpo`P->`Qh2)hBukvXVWwmA+Vj|69hInomn?}J@1M#y
z!K23rw64ETY*Xd5c^A}OW{jwLMSa37)OMM3jtQ#{K?JL;9>)ZFCtT2-?mzwYId3tu
zTLOx}x2U|uxw$P`qM)na0-vt|)6xs{jLY|*e;>ixPydNGH{9yC?}ytDpRc6=jTHKf
z>eG<iH0Y-x@zpeQS~m1mrD|^l_<5Z!;0hqTR~%P$A6s3cxz90Ua?RD`<tnZkXvzvV
z<iAh<FIFJyh~UthZU<<xHHo5(P+)Hn1%M0(9h0YotfW3&lXXph^-~FOIbQd^T;2$f
zf4-CJzt`moQmBBd<RZ7Xz{E<9qj0v0Obt|j>GSTi{;bUDtg2j%cdGp%*iKbHUzHn`
z63+Xn;Ik=Q2H}WKN#UiSnNIYW%wAeF)xs9WV(RnrV-mBKtG@imQ8_%1zQ1%vn3@GL
zCRx(z1Ya}F_3{Z>2H!lldG!oh-^$UbfACimUW<SQeVahWF|A2DRe2Kbnx__AnoO#_
z<5cAFon4fp^n`bBC+5(ul)-Vd^8z^v8$Z_2NSfjs$83{HJJhtEkoeLB^sUSULD!BY
z8oo7nvCT|@oUH{fLnBMk6?@ru`@+7xzCHYJ8XQ`n(ud#veu&RC0u?xQFqAJPe`)xi
z-w%I0{KTn_yLt~~4oT&KrD#bu(g)*!LeDmHWC6?|%n)eLk5)!i;(S4CCMLv$iHHd2
znu(8Da81kVpMU;5uqR)?G)kxhxf!PfFHMbEL@8{b?K09(W}w<V`5Mm*LC)$x(^@o~
ziwZepRdYYe{B#;&Cy@os7A@!?e>~OmQk30xE%cN=N((&~ReM9mDUmdCAf7mNPsaF*
zk5PN#O{kja&t328vrqx5$Xeazsj5#MvtBBV2p%iF`L2Am%?t{p)`wBw+h&&Pv2>gZ
zhQ2%wY3MdBQ%A2Ub91e{&PBbHR?g~F2P5dDKRSzT^?63=gM6NGxiRv$f5d?HnqZiB
zVqg<xOB&NPd+9KmY>e9^l}gx6zCf7(^n`MmrI?oaS6SB5@rK$<MiqKYb+x1Y)MB&M
ziT5c>L*-nSGOSkPAqoP5o8Z~k(-a*@ot(&#CSU3%YbbDZ*f5+zJQ`W4wpA9TsL6N!
z2HEQl%Hgr;<8pL)mQbDNfAfL?I@<bit}X3RnLnhsgRw>H0gkFOk_eFhB+0rbV=r@F
z1TqQZ7&~KD6LO_U@iOGetj}>Hnl*f7o|gB_O!o3G*d~tsRBzC?G_(3fv;xsj<XF%z
z<ukJBx72UXk)(J8CH0W>v*=2)FiJ#|JDM?@^5m8<Vr<=wA*5nsf6O<ZHpUJJP2ta|
zE#xt!jBr$8*lIjhTdvW*LM;@k8SP(<S+;t?hbSi~_bpwhV;BRX%s$H8V~EhjdGO2o
zlJ_hY1W%(U^(`guIi}$8(w6bX@G&Vv4UbBqFHe>Q*ES7mIEMwv>4=F2zoq0o$5Gmd
zJl0(ADS$C5vG#s{f3?^%W~E`E%p7#rZ>{LFwH7{bB1OAHQ7=xv2WY2-bua;azP%aD
z2Ln#QvzeIBzEcXU-=1J|l4Bf~5^N52wt6XP@JPt!@P>pm8v>zp5tsHRERCL<LAUm1
z3==zpOqA8Hmgh}jTB3(jGeyp&s53!0GtuLyN1f4+!S%Wcf0*7oaosGqM(~2-b7I}&
zCc9ufZ^1;<;OF}{Ffo>PvJ1xCgR9MgRQOH64>t?OETjC5W<jb2ek#rAjm;|F0ff~P
zMT4E^O$YukYm^-e1ZhJ%yXR5AP_`%xv>vUzhbWV0Gs)^(1`24DFvfqtr~;7=&iVr;
z6^!t)=_U0`fBcYoTI`4rg0LROKFP+2t41CLUQd;si`wYW2}>;tTaB+9_zqPq9-35b
z0^u05RSZkIkQsCR9EVen9r<ZH)d8{!PPk^togs7TZf3N%9<sN>OY>2X8FhOf)ePAg
z;kErJ$c%c_vc4O#rJHvj1({KgS`K$ZHb`fl_S(j0f5LEIpcKfAx}DU#Es%8_l!RHx
zJ7}3v-$%=+goF;U-Vbq4d>>C%AgI1|h%qBJ_akQ1?W9B90-5T!41KCM`$3+ee!Lel
z1S-9Vecs$H_&D`l3JG{Gz_nB#HGJL=GNT?LpYMk3T>VF=0=Z~e-w&Bl-^VeCxN84@
zl9EN(f6bRU^<7dN7$fK`+d+02vcDfPqwelgM21W^#Lz+BFD-NGyVO34XMs%(r;PRv
z>*bg9jIi6qi<J^u!33?jA1$N2O926h?@U^X5Bn*XQQjx5$#{&#u0-FP(lT5|c^Yi|
zMCyYux*cuhZet4Ph4FU9EMf5wp3kj4K{a#ef5wIXCK(3B{rb3e>jahhOs9bXV%Fm$
z&wg<PJeuzAh%HYr88&lufH{Nh{&7{AXENgJUT5UNB%`qdxcvH?6TaCgF^*p3M7n6i
z)97u$jN@&ZLB^@#P41@Sf1ME2esP>YW|v#bTZ0oY3FlO7anz0BJev6OMC%Ef`;$zb
z7x-C94ebT&;wW+MpI7<)_3hzbK^y*jmk~K06ahJx!QeU*m-A{KTmb@?P-`9?O#DFd
zUyyv$n$7_TtDt-S+`W605-opD{LxY>#UIDtiLd5xuYkA&@Kb=pz0>d*%i$A$j$8&q
zIsSf5^|&OPkjiphZ+$2aY#&^>9+$dn9wk6O^rrjrNB808r>W5~69Q%;tu>}%0+K;#
zHco;a&P+7>tGS=1zec}><|^gSmyg}#&d+ZKmsN1pPOYM>2$a&7A8Z~ye;>Z(!pc+k
zPha$NX-9T{b-#pxYZxXTx?mUtDYax+b|^ggWOguNlowL_mz!AWS<sQw%VxQZH&eTd
z;sk+4v2iWHVv$iT&U-7vG}E5%uuO~zcknSiKLruxJDR7WNs{WH^0YM&q*m(K^b})h
z{kl9+!FGDWSN)Xa$tr|be@TV|qDpK9{$@(c^o{qzgZb9<y>Lw3N+nRLOkW){ah}eJ
z4bNZ+Dt#ba*YViycY_BJ*RlJk(RdW7b>?A?Nk_ffGMk{5a5pHV*AYc+tyRkQ9yG`}
zCp;VZwq#HX4sYx<6CzU<@blpwNZJeu<E=CmUAeOk$Y#eIKrDAifBAi;4r@XvED;Ul
zOFk>PZ1%sw#aJQt(uwgxSy!;|8V)FR_{utA$>c?(+zvD6v=Ycx#~VDf;gkk;0-}jC
zC@Mfq6G+G>-AW*vJ#P?EF)HsvL^+|$O2Vo08ilNlI-4DD0Kph2_W^=dXkRuhJosxQ
zqE`~x?0JI-#89~pe-SXksFD=|t<s2JNo2F<4I(mz$bE=NEsQIf8uztyol^IqUq~w=
zb|sO`o-Y&OyuFW+yAaV{3En*vBFA|m+cdJ>^9B(Y!{a_g2u1%W5joXF^jEDT+dXd(
zu`xdGL&Slw!_N+Iu8HI=jj1(4D~W9Oyg|hD@o|@|mG>*cf2mP=R`j59@69}v<4P!7
zoo_JFF+}bIMf25B1x43T)aCm6f;4Tr_YEp4hRJ=XC=vv|qGARrK_=&UDl4~r+r4j4
z!5Av{p%Q+SwNO#9HI=Lty=?+*_P#*{Vz}Ig3RvMKR#e<TCH(8>mzA~gc(eBnDl&%5
zeW*wyyjB$zf2ygtM1|@ZWL8qy?ENwo<zv{~g^Kc8c+wRWUsKWZx!cNYcDwftDimYq
zK2(S^7Bm%=WBH9Oo>JAI-1;?TyXOreHU`dph&UA<RzyyhL`)s1SLSZpJ#P>(F=*~X
z#G(LMCvU^=uw{jvXk=wH-tKvWh>ihsA0h?>=+TdBe`(Q3M5v0y>V>Q8cF$Wxv>=CH
zbJF)>6YCB)gIT+5_S)g*^c1xpJVPA`<DFG=XZ0%Z7?Or?wqB0@PQw9Qxpz*-Ejk>~
z9qyH;>S!>Q4o86f@!hCa{~Y>3Z@oW&<b{=G(NXRI)`vq5<OZp3@Br@dN~4rZkIyNU
zmyL5ue?|VvIc0j%pHm*NKZUwDt1KP{FKn(8PEk{BnOZv?RPPDJqZ*Kd7t#CC0lg`9
zo<_b>YQ<dj>==inFU9sH511Z~i)ipJ^*2+I<EMpQ&-LOi4Q&p@dmroG_uNfBsT=GB
zKV$xRAp_?)u_{@98QV1GI`<(eVBGJg_%Sxge~lL@FT6#oE4lHGQ(CIpx&8AlR&j2&
zim}4TNtGY&?)m>)xLa(@A<r{g*L=9U2WxM?4;L3N<-dn`d3(1sV|YCkfNF{=3b&x4
z0epDNhBr5l@okzifDpwI-^MB8fHIUxloe!`rkt&Y!NvNqbxKpM44oZGz?Sx+1af&d
ze>&7Xr}D%4k?Z&@hO{Q>$wV9-1yA3GL1xC$0(tnA$$|VNO;!5pI4b0v1*_V(|K_1r
z9dE&$qB>6U9$s?}g?%WCJCWN@S!gWuS#6TwM6_z<S58eE>Ld=xwyq_YWoeag+SOGa
z7B$Dwn>SQA;kUGSEjr+N5*nO$V4e+1e=Bh6Gc9%#0feJTE1wo&A~+7+lPasGIw~c!
z^V2Rhr}~d~_m09pTARi?gw6;JW{Y#4M-TAD&c$_Tm>pNNk>?hAp_SS?w??&Z|7*;x
zu<)@KkVi{Vy@p?#TAW|uq!o<pJ~ZYRYZWFn+w;qNT~~QnZYBJpjZjyC<u36qe@#nO
zSNYh8N5m=Q700f$LPi)@4{l@Lf#<COOj>D#SkKw2nN(RPZP0Nksk^j7+MxN?YziXT
z|17J95`o2Q`t{7FmzPRUuS^!+KLI})H!{kd&{4{r(l;~8{Zi;N=kU}w17fx@&E=Bi
zY;}_0=cPP)<5S5Hs*|;2w(YAlf1^CV$jKH%fs!cDV-0;7q2f^BC*$-Oa~UL?Q`^34
z%S|H*YhB&+a#UAJ3uWu^{?zbRD`D8$bii%OZfKNg4FWi&r^kX-4*ePlVD^Q$QU|Mn
z%Q@Anez=y%XzK+ksL6a<?51|SFXsL6q?_~F8#F}KzN{$D*E%wCcCP@uPGtltzw*9N
zd4MPkFNR|)t;x#)2RZOe8U~nm&xKbT%Z4@&c6kAQ^R6)Ngs$xhqwU3AVX<+pQu5i|
zUA6B<O!p>fcD}xTzWE<H{Ju+<LH-^TmvK--6PH7C9$SB->1I7x?_zcH;mg2>@9$nd
z-Mu@Z91OEc7#Kc%9t7Rs^jC1YsMTPEU>e2n>Fe<G6X&4m^FklVIU;=w{cq`89Uf6w
zS|RAJ5Qj%2#Zz1dC;B|o8u07r|6_uqm4t$$CRnGPUoQy1F#dD+<<o~do)~?15^8YF
zY0$Fa(n@~@+zeVOW{k;PO2RCOntTmd-kmfKnmL13)zAdV1lmC>%K!uL{fCDHVdv-p
z2PTao2*p4~soekdVVwvB)-+1S0V2r15{R+$AeyLjAm52h24*=EmMPGUB<}yW&KV$+
zLJ$ZE(PGQlyj3XRWDT;HH-E``rw?hIAA{USEm?o26+v-F`^rE{M*wAl!K}+rGeP3k
zAehUV`f;D8NG^^z-lZvIB*sBG!wlJ50QqfGQ)Is_$X-p2(zyR)Ls5$4=Lo{kP2DoK
zKm=@9S`+#2IuSB9t$QNb9Q3UF0a=bz*iiX9`TRO-O43Ew0R83RQJ~whS3|}}5Dkax
z3`&2}PfN+J@+JjY4XOsB_^O$QaLk$(;=6}OFy?;U42TRv@~#^2J&8f<`#<?F6Tix@
z_wnn)KL2byt#MCP+;b}|D(i2;IdM^0WX{TS;-b>@t>rm!u>slXZOTZdgxlo_K=BX=
z0x9~ceU9KGHJ2?F@&vWc8@Z2y7U<V?)}()9hU<>E4HqC&&>fX6FCY(frUm(J-3=%*
z3)hKxrypphT_?I7j_|`SN2sy(o}g<_m{$<z%u-la(lV~>F5@8^RooE>eP<ipvL2xs
zIBdyO)-x}10L^TmR6+4oLR5V*?raOPZVO12j>e9+bz49!nC(G!*#a8FdcYY|C>DQ_
zOyS3eM}_&vzHog$Z#iCXF?U(OuCexZYWCYHA``g`*=eUp^*x+trZpAF;+FF^N_8TK
z7UXXmHi|Ko)F4|n>L~iU7G%ptsetc6_SvZAF$C9;JB;*)heu=2G}5z~D@MA(+=WJ3
zVeO67?l)4($ij$S9SW!5<48+nd3t|e#RpWuY28tQD7()a@zjtDc948Mfg{B7zXo}~
zPKDB)q)QEm*r>7vld_}o#Vd}Q^qmE51H04uYLN4+uNB9wtgph_Tc4ZsoCV4^bueEF
zKTkW=bm}2Z)KY@-rxoE$!5$t_qHkXP>4Nen4E`#8bZlhhk-lBMGF<T|98yF+#We~F
z7M|{l^qqgdHnbY*0gG>&ZCbA)2TC=(x$;XLK;)~Z_=)^9xTXKUTzzin+t{poD*QxO
z6@N+)o^LbHXTIW>T6Z2eG+n=sQ@Q&b^(jq;&8K8VN8wlCLPk!;H9=c4@Ex$p=Z5k`
zulY`+nMkddPP|(1*>oHe@0Z4R9!h`8)4H_fO2A1xA|5YtiiG~0(tmy5{d)U!_n*mD
zff{Q=A@m<#@B9r(B^X*51l532;_&-FhtGFE^=Q*cC>8JoLq<7-|9<~6bXPv6V_hpK
zm~u*mO-7}-M-f(1zea{Fbvf2OHpi`;Nslf62}GJJL}bM17*hJ~G4cjuSxtXQ)>3@$
zRtIe*C--KaHR8jyG?gZ0lK2_?Q3t4v=qV;EobYEXbJSROt3#+bj^QGQkUo86=?Y}Z
zmZSr(=X0H<+F|0C>nTgV#psV~{ZYBV{OKr=Fp_bpM4`D-Q&5I!HQ||)M``XoA9L{M
z(XTZZO_&=)TwK49=4SPlCC`7U{Daa*&G_)ynxKLWt>`f?*|(I;x_&WE*odYLDd_m=
zNSKD`^VxIMzggQ;TrzK`n4=KW^9RwyOR#L}kf@pkQ%6r<yZTT+K$vJ&;rFq1aDUJ5
zUd>4*1tM};+N!`olqlC+`8ZX5Xw1ez6{bQgTh=k9B$b(2(4S+Z!w!GhBDbe(#6IKU
z+r`>z$q9M6mnQ{x2wyvd+c~7}+JrN`bdR&r1McSmqL<$JPUpKgN_#gG3kbu2de2p{
zjEx}mTV&GRmc1|V^LK)hQJdH!6IaD@pjydI<%9FvQNoAp1)b17lk#az^t1ZPM?ZUY
ze6ExL<(N?uF&(p$cLRUi`FaElFa^DMu89KLb9J*Y&|=#W2aeapftKDa0zrpy!Q``-
zchU<sU+)jp*s)%tC1t7Dhek`=9rdiH6f$(S$GEL2ta|Tw6`raNlP4TUzt8LTDor>J
zulBm$1<yKAUu1PyckK5qbl7nh+wp(V>vM>2=|`^5w${b4kmG-KOXZpjkb`=M@rumo
z&$eNz--Xl&jyYo|a%;S6Vtp@JZ<l3JOE!sOW6$Gi#ppB3;qiRh&N6DTC(;p%r89be
z5ip^DUHgZ=?U}gCLV1k-Jl9TCy43w;)m%^(py3RS+{>5mVZqLc=9<7nosBi})`6+T
z%3s>HA$6fh&f|Z^QNlv#nuYYkX;r$*po7<|!y2{0Orz?t-#Dz5WYto+Ce3H5cNphg
zTxh`@H~s1=&Cg1lBUPKWZt?0Mb?x~~=BcH!)WB-Zgr4$rdk-y%ZWmv`ZI}x7d_7;#
z`lQs?V&Y4^N=iZ<-8OOLla{)LxuOKfg?~%4I*Hv^`Rjl1{7;YN1|>>PWIg8FmMA&F
zb@p+&FCzy=*d`b+FHz7B`PD?m3nfZPz@6otpZw9EUc%IOPRq$ZXWmX{S0;bo#&Ou`
z!~V$wR1+V@^L;ox0i*Z&aH$14A3o;nHXof+76*q^<c&`_5tMkFn@bdbTz}vjd`2<j
z?(?a9@$P?A73Aw#J2sDbV#EOl)rg7QF%AFX(CDx>@D)yy9-!xhKW{*htqajZ`EggI
zuJkx8%OhH_dS?EUxUD8jX2)b{0+q2>W5+li94AN_hYVEZQ&nZ@aiAFFT(Vlmn@_}x
zRLhvQ@N&%i*Y$Yv?SSg3#uk3bA}2Q<`uf3NqIQ28D?|BQ*478cjXeEObV=@uGeGc<
zl|$3iqB#f|#aPcFn^|ru9s}~@$l^Gmoz85_teMw$ifVYg8wD7j7ARrx%weBDhymmJ
zVISJyy{ogn9do?!48Os}<M&2C>7|HMcgSBK)hBz3#|eirU!Iobm-OBpv%;q5Fc!`&
z)5(_xeI7c0_pH%ril9^#_5)`B;Jj1W=rV?}vao@irT=H_SW_-V;c@=_3m<-^n;&Sg
zZ)g4pOTQL6eE9e9XK&RTh^c@Ih~r7+3SvG0bJiuaBMOBs)L>_$*vXlHQ^Cr0v}7Gl
z^>|l3*B{4ISMFarGEmh9-ITPBmp+@s8eWuji^IczaZ8PD6k%Pws|{&M8J<bM72d+{
zjqw(K5xsW4k#_ymSjh$LR#k1x+B-FOF~=K{0H8A2+*%SYJmey%L&fU39#uVnm1{;*
zI@EmrR7S;zM+rhUhM!xkO>tektqpOJj)k<}NL)fPiMn|VDi>q*(}FVvXmy4>HLxtZ
z(GX34dJ~h!H#mL13U6xWYvC=;oszy-A{&zSV=3#tk)%0>w~4A}G>T!Sl6rg`MdZ?6
z<wsb`#Hgw-s~OTHfLD);ZAtUuU9U^iI{%jIH<6|D#n;%Wc^`ipGgJ`cOFEI8DYzD5
zqR94N6|gW<ymGZqu9xnAU8=^XadGpkcVo0SnOkib3j8)X&%d5IO?aT|Cq>8SVaJ1m
zCUPxS)EQr>n^%(v*UekVn|5)rOeO_%_x*or`Pj0Tp|l+pmw`nFB>^{=U_%!&e_3sB
z8@Unw?q9)=*M8uV9CAn=4gvB~Uk~V^Zk!fv4Z~36t)tqOTxo@-|Nfrgu4V1570Z%*
zO#&F1)si#Ad3k1r>rHaXC~p!eOHK%{90wtSqMU?a6{xI<igOeqsl*;c_R1O`6`2Sk
z$+m2jwJsicYn%eX3RD8~RJdnlf4#M`&L|tam$#spqflU%T%iIeHbSwo8LU7-Ms9aN
zMotVFVC9?>8J3hITPU1>f~>O!R4%_UPUW*uNx4Xc3ZSBJcHok|$aOfh2_$bbDE5%y
zNRh>nL3u^3G-p6@`3<KJU;qj`*cOUG6-@~{V$p??qV`0vC8Z=9kQgG!e>ueTuAH2}
z-Ww5#m)=_tFsqMcJr=aG2r32^Jv~^)EPOq9q*N%p$X4FUL30R@fa3)lMEBrSa4hHc
z0aPqpGJKpxMldRP5oCh}Rpfd^Kr1(32Z!<&{9>^gju&DwkipoIjqpPRu`z{T0_6&o
z6$=BrCB-g^^$<nedJU#we+uO{gqhD`GscdP3dM@qN>oLwILe_d6z3#?X@gM38xdTx
zEpMZW#tUbdB+8-jn+H`q9)pT-v`|?^Ph~HIiWV{u@i7z);HDUj@I&|_SmYc?m6#+C
zEL2OTSV%I$!6byRNiGXxWJ4nHrbS95U&LloDX@bGa8IGEgYOa-e}f}Z6SxQkcG(sR
z{Bj9MEKXT=ID~(elY%0Z9T&=j0(($FWukEO!w=Pm|J~Kq>*;jXRHw_2jXeEnGQC#S
zo7w!bo}U?{zB#Y{Sp8JJJ+tk@`+D4*L9B*<oF^UN6N*+tvJ0dk^1OPj!2GoOarR+W
zy;I*IlMoXo_#|M#f0=OD-@U@W{d6#_6CwavttI5-|A^Rd-fHL5KnvhNj?B>6I7u?&
z51@q4q%-V&uS7^eN5u6eB`jklGLgK4pEJ3T4xofj+9Q(Scvt|L>J5Yg<*e1jJL>Sb
zSU|C|omdAki2BT$k<Km|EeJ1eaXFtK^$t<JUNIjxli3vNfBg0RPxANOr>41Ed|y>p
zljhU%qaM$0tAEYw%lYKGzMjkOrk;<QTL1e7S|077W`c_VA)TH`JRRK*dN#{uHlK`c
z7S(t*os6shEJxF3d3$j=S&ZkC+sSk!8+KO{xs*_vBGd!gqf#Kl01Qc!^Ut?tJ=Oo6
zTu<)m%gIR3f96+}JXL?`bY3*D=&v+?Q2X5(%CD6D_toG3@lVwz8Xyk{KV9D3oNrha
zu?lvjT+IEjQdMtfH?#Ta-Dq4Zk`1al88vY0RJr!?51*SKPemtX+aHc+(*`n*Ng4>i
zs7N`6IZ$G$eL&?y40HG3<sr5Y03<8L77wppRVVY=fB3X+&Z?7l$JK}Wb927to0HL1
z4eh2)J#7}!IJC^)*NfS5KCTySFCjnuEQWkD`&=Bz7GR&<Irh)7Q;KwJh~n!-+tC#9
zM`xv{ku#N+rV!H6fH>a|{$S^TW`#r69tFVJsHI>7U$yamko73WsMdrONGQPx^_liZ
z9!E)Ae;Zdy5Fi1q$0i-dpGJ3ZySBRlPB5rCD1^zQ>JS<duUUG^>~Zpg<;cmO|Mqh)
z7o<_Wk8*Su!!%f5aGwSYOb)2&Pp?E-)Y61Ly%L2Bxc>5~wU$8Dap}X`aITKg7251f
z_Yma4b`OX^W4bre8W`%*k;_Aus8M=!+p%ujf8cJJ6>c`|lC{<iveugIv-ZK})@}iM
z{Z*Y0v(tgX*8zIJ_Sw2;`p)RNY2CG(-L&TPXS*p5-4u3m6K~GPgW0`jg1&KjW?N5!
zp0C!N@LWr|VL&^xOBr%uRfcSPmOYD(b6SQybh{bgW;tnwveIxGZwGk4s0;M!sKW~I
zf5m}rt2}d$<QXb}G#y__p7r6dhk(84=>aABbk{-17l3k~=t;xq;a&2qJM+ut*_z#-
zdr%64gA(nMCiKv)$?Z$qq>;TR+#`vz^>~C?S4Y1Zk9=~PB7J@1QK#qqc=T*NAI786
z%@wHT)9s@ASWh8*adBBg@4MMzvfqdle=*lug;7^0sEVSbqg;Y9Fo_S<{=Ioj<$ga%
zejE7et?HUW1pU*if<-@T@)j!z_#639aM**k&Di*bhHxXaZH7Zh4b!y&>k#@}Ebw-S
z1$hfBz~ImaJ#U_9_SY7yX5Jt(Z+r=O+}wHeZMtsytqi=^oa~1s#`E^pen8$;f5|y3
zx6OMuD2@mnNh%m27o<e7OvwTg!Ng4+%I`IK>a{37Zj<A^Rrfg#i79TeGB~T{%J~JN
zrkBcx>PiC-oF4oxDHg(S+gI#$Y999DY<3BCUhBm}>UxWCkfxlpe4BBBQL+UUjArlT
zGo8+yz7+Fz|0^J}<*o;_0AU2qe<kzKIw~7UTiSdD6np8$TTKK!oCq8an}hqm-HALq
zoQBBByh{>s{DzV!4wERFT}~ahijz2u6Zudhx3b?#oWx<A#GNW^G7qp5eFr=70HxL^
ztAi_Fezl{p`sx(e0T99<^P|CZW<6PhV#ti9gb#254fIJtKD(248Lduif3IDM#NGZ`
zN|ea?ma-}hFVs@lDMs=`j3n2^NTPef?=$~GOD)~`>ib}39b@?V2H<-|S04zUtLUQ>
z@OarYlj+sPqM41aFUGT*<?Zz1s-B9@`^>`kU%zdL+6U1V#h!2qj>x`vsA>3ZPyM_3
z?Bh*++n<ILZv)A_RbA;OX`^laW)L|KYgpc?hGqN2gqiIYQ^m{73Y$Oh^vBKhIr30&
zmWObb!%ha?DkAeRA_)^&_^q4C%^zO+Uv75(@Z!uMa-G>Ga=QzKmC2lWB68Ok1cxaE
zCSW#t%azV!a#+p7zW_8FSwojW{vH$pF*TR5Z9EeJG?y_p0w#a8TT5>wxedPiSM<?y
zutub)SJrEwyJx(}o)W}Bf^7B>WV68p8DNr2^7BJIORBC?Rk!U!yWAy_l0H5}Qfhhl
z$3Y%`xI4YveSB7Qu)=AuczF5s0Nfz?FGzlA^uYoMqu}uJ_3+c9lxX;K;txrw6n`xL
z@A#>o9uzq40o;EI@bq8_9#cPj=FgG)VCcvHpEDizWEN5xruD{${=oc!+rT~j{PN?S
z%o7EELi7Cma`*4M`Bb%*0wf;(_Vq3_kOH9`9ss@Ypbr1}`{CES-&YoewG=M=$A5q5
zcse1Or!j_~Ht_uEXL<OA8-L`^%DcmVW7n^tyA)pg!@qwHe_;Ny#nu61D~-{*!xp@-
z2AR1RDl{c*1D?DFNsJFaBr$gM{PbXCKt?k%+NQVRsa<%-$Qi%@+^5hn{D(e?B#Ou+
zxKU>@@O#26I?SSgAdUoggGC3iR$fT*uYh9%X02It?ww%PjyajC!Wq1M@R9@5!5iVi
z45P^~GPHlnWvxDEO^2P3-n{e;NQne6|M!8EtRQ8p<_0{1=6>Q&fVO}cG}(uD6=q+6
z&qCl9R#ewP9323|43s}=`ULj!bN;S(LVPn68xSicz{5L1tc)P*u7^1E(wH@vQ6#$r
zvCa^mv(J4k#5Y5+0WqRLh3^D0YN5ebh;@ecd2)XP3-BIut77MXK+CX(P4L)guS=}C
zDSfA%u-?4<4On?sN2}h9Z=LTO^R4rh%mnDI)qp3f6*E_Y#aGloPOko#Eh#9*=c)jv
zx38!sF7@IaW1n1aUe6dO?NUVXF6N>=3PY-zOU5C(^4`hYCWvtbO?gVYZ1G<5{Pe(f
z(eHn*(P&^Yn&_L(X-nw*lG+FK?SO0mZBtC~K0rH;xgE7Y7dW-b{wXNjsjB3;SHsSG
zVOCR9G49=!YEbq&QN4Nf8>pHTR=f|YRtk$X+Qx#Do}%uA->XrT`yKr4E8aj-r|98*
zkksK>1L_bFM@L}0ee%LhK0Rmx-$S+6jBgZP&K68cklTn1QFvF_E=}!97F#oy_=+Af
zfA`db&bh_ZH&n_5<jdar6ceoI+?Il`XlNH>Qc|e=_F^Y0W4e}eaHFSo{K&=hMxNFe
zd(ad=XHFNRyW%EH+ec%anJatV#I^p{&f4X9LTxSB!^i>EQB7H2?k#h?nUc_t&)OZF
z(B2xgB!BW&DDrKQXK`k9)y$~F%uTVBUq0tqjkS^uo(+16HyV9Is~=-brsC5@p8S@I
zJRoCo&htGbiJ>*U?iEfeE6%MuSDbrieMsMmKICs~Bk<T{aG1D9&VlYzyS}-;W&T;i
zi}uDf>Jk==SJcid;dC*Umk5j=C4bWiQ}^|opZ4mwTxyrxEvksgbbC^m--UV^BfPH(
zcBF85K6X+gI#y#=&5Wzd$nGmETsF9iu|jDQCkN}3;ua;$&e1)kaf--i@!sTmBL==2
z|FLs0&zUc#q$T&s@k(qxcP{SIy;WYtr58n_=b>_-0U^<KtG)kcRt{^INgA9WGb8!_
z154B1!%(fzDQ0&8?b80#`PFU*m&S}9AqV#wte{X>UYG8S9vFY#1C~e#V<g&<d^aA2
zNI0pMM_HCPXymA=U&=0Va{?%&QZoo+QVCV0wd8f~!UOEAMNP3jbz8GEsbC||V<85c
z_Wo8OHq?8CC^`6*I>W0rq!{|To@Psmj%TK4!)!>Ag#gemW~UfA(v*4zVk1VoC`NaK
z7{P3cQaMg!lv{rh<y>@q?|8nFonoVC3sLdNWkT?OpgZ1GEmlO#5P}Socf1v+D#87i
z4H;X{*#z}88!~oAXa)VuvW(@2J@K0|S{Q3uyfn;FDjVQFUu&oMYA9C%VL;tzqAm4K
z6IH~UP_883qugbp9nC%y?MU{SXt$VRqJG0hIZ$(EhGKuV(NHLQLwB1CP+NA&Ptv&8
zPP^W3Y^M$Hy>^Pgaj(11PFtFtcG}MDDmz8*h3b>PSH{8BRG+i(Yoay=*-N1T+X-#i
zpEm-fVT!Pw=sj@`%U%FO*^!K;f)a%$wLHp7Tr1gYx^5-=qFT<ctW`6aG^L1P{7sx`
zadNgFKjeSl+44-fmEjmaO~TG0-@s2>AuV2opCO^xc&-P~Ars;ffa~=s*pm&|e70IW
z87<X!E|aKQ2$h-sJXI2=+EPtrAy))!SFP%5OtHPKBv5CFOSLi>Dh`9lGVRs%k3*O>
zI-w=ZU|z|7gG@B&i}MU<wggm<7hyqOU}-HvoWFlf|A9JTyB(E2RJ@XFL{NZAqm~UT
z^)2VS1@iQEUiwRv$99yq;vL0V`%Sfl+LtPiyrL$@E>!2~fH9U7=akJ|im0x(o5`9F
zPqZ68Jf_hT9<E!-#Tx6Fi%Qu@YEe5lv(q(7EGL8!s(3h9#S_0#yPKzWm8WD@EH}}O
zAgO<RmLRdywl;RYD%sGqocnKMAE!Z(vmM3E$F0>_*vYyMIrhL0V9fn*!o#LV{q|jY
z{BG<xc`Nmtn)l*}dLV0+>D)M`1Z#6;D$93m)**QHVHap-{nz1Eqc~DVPCd~0cv_Pq
zsBgn{D9w=?$3y)(s<^Z0s{>zD1*!)iRXKm;v<t`jS$gYwj}!gHg3>Mo!NVeJ*e6kS
zNPQW*<8Xz61Lc`59hWVjzB$*O;ld~xMI)=r9^1w^s8bx$0(2bj$07`MB@FBDClPLy
zQuHLMEA6$1HnUo&soUSev6Edpc5+>!(5I-fR;raOp3bH--_~Ep_s4mHTRe5${;7ZC
zt!5S1V{_CXrz5nk)cEn4H`c`fl(WnbvjQ8Isdot?YJqp5UJ3Gedv_r&M9*nYy60E@
zhjEJoBFBY6&d38}4kG?E4kekyg{f+x)XZbLwP|^tPD}CqhOcC=6V9w#95xa+3<)}0
z5rP(E_b4n&OPByx4|dF$Pt%RCoaKMkaX@l7%n2ON(AstbgsesN_--sE@QVE;TqC8B
zY9ZkjQUZ3uKv#*WBiSjZ7LASVA;?Lxrp2V$<zc>j8xNyXmfQ`qp$<B%!t}$`QS{md
z9WEAjqx?mp9?#8knxu43^~-o}99bjI5ExEPIm;BLAH)skr984|peOk<3~hguXAPJQ
zvMaP9AeN8oDR;y+zuje@GgIsA^IYw7RX(T$yY2&U23#&bZ|M$Of+Zb~Jv{Bx9c2Ur
z<7r1}bY~@&WLq%o5<4onp$WVkoDH<0368VAV5BX{w+&4&LeY$5rz!w@H{VmfI?Uxf
zl!5BG&Q*cBqU5GJaDYQzU8R2x$_ii3O?Ik7J4&h0SXwIs>0sh5$)ES>LQAq!7g~~?
zy3mqbrwc9R+v!5Xbx9ZKV(TtpGE0Utlj)|Wn~F*GvWso<gr-oFPU3UyYKGMhO-=q)
zli&C@RWs_hP#e(o#)PZaZA$xdjX7hcPbFH`YE&iB?Tw5)n|7|Io9BOyTM}u_SdV<l
z81*1QQ(N>(eAD=ek8O_i^+l~D$IR6Hq^Bzg==Nex9?Q+vKfmTv>6!w>qDDv^tN!h+
zlI3eS!=#B+ytdqO+b}NBAVAGXb{Q9&P1-wJ%2yj#D9!5-Y4w1tlnBGQq@v9WXb!Hq
zfaajpqV$a}+SYFP<|}{g6J3;g`;&n4WCt-nbGTF~bQeD_(@nJ?wa#lRdCZPl&Wc~P
zrIyPaw01hJu&s%nrysuM{(J4!oB5?k2TRv%T)F*QOWKVu%J+|!RKOa6wlkbshQVf`
z;i?z;$430e@@9DFEzy+Qc&R_-f{<NqZ)^GaO%X!5I~U)ESRqyZ$};>6P}RO>kyp9P
zm);&H?o9#P4rYM4P+%{ADB6y@ev00V+$u0m)uD_|nXT_jWQ#~{yU!(&heIBmY7=f@
z-x^$%t<I$0sLs#NFL(b3Gtz8xmqGp>6qmuI7ZjI|Jpm|}ACw-Q0e!c<lpaz8e|EcQ
zznh*W^xRVy0o)bWi5dTeC#R9^9)gXb6`WWwPaaw(2nB`4HDMS_W8{8!EA27UeF8>>
zUw*}paSgu7L^t9MCNSK&I_EIK6=6abOvuAbbR*tig0YiWV3IIn3KrC17P`=GFo3ZM
z^Y2M4fWV|=3l{WY7P=8{Fo7wWe;R%g6O<EPP6~6Fh;G2^L<nOEtf&2cB7|i`02EBv
zqfGQ8-e5xegnbSZxOozi7ff6ZGtrHBg9+vH+c`{V+?rTXFmXN1L^t9MCZx}A=P-er
zddX$M#O*K>-H11s5I)D9!-O!ng>M#AZoM4X;1e5iJlfgGMmOpWHn`7qf9J5lHBq1&
zHpF2zx>0Yi!Nz=d4jD`lB~3$yJWNI}=nXEwr@V8>07;a!4H;^Y484<#ZqyrWP?z>j
zT5t*~h!zbQdXWsflZ<ZA>tx_&d(3>Nkijh{+CamGIm$*q>J2uu&wl5yVHh!tHEh^L
zHf}rF=tjN4hVmKk95yr%f1`NA#$}NWv46qQk9>m>>9gQDj3}%~Wy8pIl@Ym<k#6K0
zj0m3!&tXJZVnw?IyR9-}b~4h9e1j40^Wiy+a6_zVe%o4El!^!9NH_8=Mi|j><hS-m
zE$lm8a8~8|0%PfPp+3ggAvD7wB3UlHyvk9HV{)~QSK=gfjSUK%f1u;>`k`>|EmtSC
zYN;jdkWN~sNw-v}gKW_F+_Tg7OZd_Ltxrg!+zMYb2SZE}kZN5>Ly6{i55miM@>SgV
zeV5pVX@lISHR^c4F<#Z-*LGv|5iC=#kCWVDQPkzn7$c==uwxmPNzkt2v0PM)$#9+y
zZ~kJsyg!|NtZ9g1f9-P)YWyKhb|yq`{mjNO6J8P%4^BnWR@h9SrfLoo88fmtUw{6X
z+)skgLTXFSo_VD<f*3B8%_d>6vpL#Ri`jGwBETi9BVIPC-8>d=Kb2YTeJd&I8XerW
zTO(S(Pk3%|G+1_kC#FzT;)&mQ*1s@5L0FiTn0NBTC10Nkf9`oDG3a8zrDP^IW6JTP
zdj{M+%!N@a{=ZA^5#)ij<)=ljaSp|x(4zthlBNovk75gIq*1z5Rp<$pk4uB7B2wxH
zz;cjd$*RW5avSbsd6)%joC^{rjY}ad_F`O82he!|BBPAp`XG7UcBFh<eu{T+$y~Km
zi4c!S38Ngfe{LZYF2=riDZ)cS&dTbjU!t>If(4w->E}qFl|r1DB=HaP7bIbo^@hL4
zWucYDQGw!?DExVmGcsc^dK+6rPM^W|MM*)ha%F^uv^s^7QDXK4rB=|v)}U|u!+Ovs
zR~E*-!Y%5!&sX(HiQR2bEUEf}igaeEQu@&$2)udxe{0D-jl%<BF=#a#yzyF4>?wk~
z@y0uz;0M9&lqHP!x~>9ZF?du)*b|)8N7c%3>!W){VA+qoD!**x$R7vd#3&8UCubL)
zlreP|RT+CD!O{?>*|TGBgiu<i?~5XOMg<WP>KaGI*bIokbXZwkn##DN*SOKIIeESq
zn<@8me;D73OJOP1uqiL|BPulEz;pC6DuJ)m%Ym;u*TVO5r&#;NQ^SReM{--=`@nx=
zViLt+<yM|y8PI9DdZK>4vbcL$d3=)wloO%#Gg%6j4dCv=p0N8X4Uf|_{sf^jN&Mo!
z((=vTfe_<^MhRwZqsIk?WmZ+_`&CS;#SIGPH#s#=P4}Pia+4eX#LJ8FYGAtdLVl-`
zn_UQ&G56+0g`5_PZph(%?uX#$;??{6$IJf#-?v%~mqGp>6qiB9HWLFeF*BC|ngu6+
zSj~>(HVnS|Df9x-h!UwEU=Sejj2G=~i`<Hyite5U#q9eZO39LK$@V0fOb?D7nIfr=
zpGaCt{1i%j>&}<%X)q|9^aM_e%MStkLh&ys9<(GJfV2k0<xTwgtdzEVo%uqgjN!}i
zv*$Y<4+cC-0KWw|9vs5+D5nR$PAsE;qZ~i?c^*r$3S}(M4K_r1VEw@F;B@@!@_VPU
zqe=!FALFh4_j>8xyVh+a<TVBH_gfcVpaE&T7Fv@8vf|f2;z#$rRl7vR1EZp>C8|(C
zRvU!1L7Mnx1ci7Z$csc?7_U!}9i5lRv!<V4KSeY0$G6#U$0J}{&01A3%+g+e22HKB
zI^S>C8wqZ;A~4%_y_3j%Y3&ZjgH`%Dqrh|^;5&z$IFxvfp7Ars8gU#;<57-AS>mv(
z{%-HJ^^&iZKd%jW?$GG0bk42tz)d`~2dQ{-wD$XiNu1FVhrDE}PsHiJjySdEYs9Ig
zA0bYy=^AkwyEAdzOb*2Ri2yo(?%~!M*p3=FsbK|zIei#8j4N=fEzdq9jbz5*6VXFy
z9kkk;=qc$8Plf!GG*%;9aj7*%dYKP4prUQ4G)S~-ZQBqTE7AECxxWaaTKiW(RBKES
zML2Z!JCPGU%4$2H886YhmHq}a&6SF4uH}+M5a!&G5mFHfia#nNRE_X|dY*=$Hj7lz
z3@B-$H->R(R6BjG(p5TQ8ppou+Xzie>R;h<33BanOp9=vyCuQY3UbiMkNLGQe#$id
zuF}+Eqm1P2H5nT+ckIGIpFxMuCHJzb+R|Qf4<rMcm0vak;1)h*CZoQ^?^GaQm8Exh
z(4a&?OUquBENf<q$)itypkl2At6R-G*}fFWMj8tZ5-+9_INJajloP}i-g=yJTH}4Y
zr^_t1=`=039+qr4Y;B{D`lg%W({I<V;tvJs?A&(bY)oN6bvti|TP?!twl(z9E4PJo
z<Qttb8ikMdHy>J|M1t1RswUjsKr{-$L|X9`cB&(Bn6aI;7X-t9X&L=8T|8PEb9!hr
zuV8o}FXxof`W~|HLm|#hNxC#%jkZ0!w3Xh6Dvd4DRjFg2!K>E%NGnnBsWqlz6Tf8p
zkwax8Mcm@pZsUMb8@<xD*=-T9Y4bm{4aA!(IB07qy1QA`YQ}(4fnKE{mkSi#Uhc_B
zOJ=39O3#lY0!n&+LV?^cD#9y>?D|@?Ft5Sby;71=<F!8Ie60rG^ntx~=uz&km(^Jc
zpysbz8_!bSJL$Mrjqj(2bi^`Fe(!9=HXgX-JHLGT?M(;q@^3tE$#UOU)IStshlRCt
z5JO&gNx@mObvy+S&v_=Ynp`(IAsr0*W%aSk(FhflUY{0!^><7K*AB%3H}>=AzVTW8
z)L6x{F!wx{@nkrij!iQCTvo?SS|my#X8x9u=wz7wZASSx+AEPBCJ#=P^-|DbIl{9w
zWM0pmanUf!C$=yANdUU}<RaZ)mbQ}A=v%SXhGLXcd4;ajLC%pcb*`DY`5*IjEW-+A
zZe(+Ga%Ewci@FyH1T-=-H@6X@9s)a;q7F9{mxH+(Cznp$9-e<-1C`fbzWwwfGow;?
z6aEdC@!QvLFaLQN+1B0)h5Gc5?=Rs2DG<ux$)Pa*^vAzWzrOrsB-&%*fH9HAXg$FM
zBm+e!W4ut|QNn~plA2{B=9?G`x+UHP@oV+ZAHSW3JAeJLv8@dA2P0*a6BO!bBH{cv
zSk2hr8<zw#UDJOOM99tK?0HrK=f)X%nHSmfT6j`3{+Kt-Olq6>39!ZMm&C6QH?EQ5
z3~=oPYe)s$<TnB(%-^p`;&>m@vmm`*LgktfFqTCHbpjzX0+TyE0)-Q@f|RWb6y)(t
zl@kAv3;v&U;bvt4z2HS0sjc#Rd;Mr+Fb^Hp@9XjPo6dg+G#J?<6!RJn6u&IW%kl{D
zb{yTW7oGEZUNR0UqiXa0tRZo2O;|&|Cm%)vtr6DFI6;^h+e1{H7jxwRwG4(=!ut8*
z7q7t%N(Swdg8f%g%Ki>d`|0N8l7CVXg_5&H3H~;j#Dt$UFQ-6<n=}62U^MLA$hwKq
zU^bF${A7PR$KcluM=N#&Zlt(vII1`sD0eYC67ABvk;Ww{)>~vnEj0LE)+ie`)r~5x
zv!hsEn<lRnMyrwPw!GzTbh4C9%m(KX=FHMXmC`aPSCnpY=Z>Z!df^%Wfv+UZ!)COm
ziAX3oj2LHmgWEzf!t+KlQv5Pd+(xES)DWz4T=swKvgvCcC1*-Pa-x(Qo-ZxoA@5gB
zdr+tIVL%lsq3uk5xx4_NKwrNPD1}10nf&_lxFp1@;b(@i;K%D5!FA}F?9os%+C-1@
zVWm)ZCecb{#3O~lnc2h@zcxo0SNv@&O@P9K+sPM!6(Rm#z-BEX0}{FgWE-Jf)bRVk
z*lYr+1lvP@AM(XFZ`os)nX+FkVPNs*)$awvxsJ2HMi)6ogG<=}@D;l2VzQwK1hHBV
z$p-S3RWxTCg+jA37pHUl|3d!4dWC2M$)A6br+@OppLlTC`~Dk;ey`KW)4xxD3+c=T
zVaKM=gbh1d<?j)eC)OR#FA~{CLMtyM`Nv7@<V+iXb`JNcvT}GmlD(U;1@EkwHm4%S
zcJ3)VTh;58FV42?qEN@JCQ(hIiE31=jtaGb6=?wo??$(>`_6PLzlsDHeA4174pjBZ
z+IH?e`i8yhc9`J}Ab7t-4$<Ek2rn!%tC6E(Uu!j?)>5_-ot+)WnI%FjQE5$>Uq2E8
zP1L@Bgj$<$9TsZ?F5bqF=t*#KG^ADCVpJb%b?HKl4(fJOv+-pGE7&zf#I9I&P02_q
zjO|4CFbW&!N;V*;pM<XD2*IkpI#rd?<mBfZ6TPC4K4Tv+mNgpAHGwn^$56nxH7h~<
zCYpz(zK<sJFBF>Rph=#?A)xByf@yzk+fqb-@{*&*g&=Fi6~bCsj&f+E_d$AG^bJUD
z+P2?&n|p7%m0#mH4u4V%Ll%T-MbNg-p_%6RQ6c)Y#!Vd7Y7|}Q#aQ|7-XTMkp1RO^
zOFs_dvVpgbJO6Vm&PZWf`78MEQsHeTvb7T^1JkVleL63R=O-p@2dMKIiA#A*+$u1C
zZxs->jh!eTMPmbGN>RmgP$s^1rNNBdK6J@}nQpX@kc%jqjAC$quG@UWGuVmfVbM1r
zRVng#4y4Kn*OGX(U1=>a`{w!MBJdN|Q#XwW!{Hpm3_7_0Q0)^_{jd4^-mT*>6dMp@
z3O}9$F@`K19*3CpH0UP#XFz-yiVcW=A;l!mff$VNt^QMTnRsMGtMH!lQ?-=Ks0^!`
ze<rZ1ov<Dje*;#DB7pMlLC-tWE&M8<;*zHXJ<5~tjuw*pbGD=y39l!$T~4<%^*=7p
z22L&o6i+c1<w$TKQO#wRBULBghxu+2@|otnII#N&PMV<@kPXP@K8@g?uFZgd!qHkZ
zpq=r`aX>bJHYuif4$xM`Ba{y40%t7amm}><RmtmA4d>;A<)Xn}H*{Ru4HR_>Bc6kz
zK>@OzQ0uN*uiIfrtHd5gQSL8d9hZ0mNlMYebCA>^fNB68BH)(%->da$0)I^54vVt|
z6A5SWEhiHXlh}6{6UdtTWY-RVW5!dT9UXUyw?6Idq%4`sJVC?Jn;n@vdiy!r50;}Y
zO*yK+ru$g#SDMkU$Cz_`zswY8GpWCiV(#@8?%F%H>%|BmGRUy~tQ{=o+%OZ--MJ~1
z+%P9EiEvjz%<Rn@_6jwdpSTHY*{FiqBnp$34%@eKEYE67eY1qHw;rm0FedHI^={z7
zPY*`3dZ72h5cFrq8oRpcb<%y7n=+Ss3SOZ~XX@GTY(59V%67tj$@859?S@&M>D=Z8
zb-|YBq~}C+BxF9$_1UO4G9jAB7epx`m1n;-{NgIuTzP@A3XQi^B&*-LZM#?WQd6cn
z$JXI9VnpuKO}mo|wrkXXl3}AJfs*=GS}JIEU5u$4j_0NsO0<qrjc>Qy3v$Bkr0cm@
z#GGtNs9eW|UOo?LoXsREl`yKSPxb-1>eG0TU*GN=TTKZsBYvYDW|^+j+#q%y>Q?%o
zYq7?D#+^-EW!11YLVddNF71UkXsBP0P^R(dC+qRye?Qz?LgJl&S7{uVu+@Gmv$wVg
z^mR$Csu3-K<cd+ZLfoQbK*cQ91D+jiuxn5Nq2QflLh-}jIhM(2usIpwt#Z={h|#cM
z_$_2ql7ss2`&|G-`=CZQvL4wsot>K|Le-My4bELSda2%IvEgp?Y&%*2*FXwwqn-wk
z?u(0|3Euiq+%-pkFEuM<^$Y%0>7vQHdRv$GjN*rfv?p}2HP2gB3KmqVc=fi^<xqDn
zhBShMqJhDC9gZ<mHMMbS6`#<Y!Ku?g_?q{_FH3zv+dn-3tpxy#-9Vu#tLe?ZJpiYK
zlI^z#t|?XD62Ii+@}~cA^SCjw6=Ilqt@+>b`)rlN=+<3-yV&i;3nLUgv!4u1ScVc+
zt5EC<I3-Ov@Our;U(pOy*MF}eg;9DOJvE3V(YjV2Zu%c)?Im7m(;xYi@6jX|uH^i@
zNsgU39BYyT<jfK9{hQ<fIgdnoODg&>iJhDQyK?iSVk?9;&~Bel-8qEYR~$Ib-<I6`
z#M(El$fG2Gb}Au+9|hPysW@P5-xaIeJZ&JPVf!W2pSy!8>=N4SSILz3#(g7J<kV0}
zu**q6o;hvk>wFAhKpuBDdQ5=@Y3LHnOwF{kE>kLZt}`B}bh+uAiZqT>I#Qn1<GE{j
zqsJzq*e4*=()fO4{k9SsW<GokvX-JUqdD6*>z(3%_XBHIypW7`M7APiWjV@p<Fx1Q
z2X-NJ)3Oh0P0l4W^htL5)&-0$=)E84vdPoH^ky=EK!VeVli`cTUBNL*fxLMPlHRF!
zX)M2JP2Di_GRpj9GLOCH&{$7OBpL<{PIr0@PDsy69~tTTMyJ8~jX{I+8#^>OukE#P
zB*$PTK2gec$(CNj{ZMPKT$c1Fy>Gg%q)H>kJ5%kiE;nC!kNH1^*Eg5JO9c}HG%_@o
z@nr-PmwZeaD3?&^9t3~mw(&k+!Q<><U^LKp#p9`xv@4f=^Jc4(isMrfZ@DUah%c`E
zeHu^VK}w{g&4ELS0=gUhLO;Oa)8~m#-#;BcJ$-YQI2mRY7%_eNbprZ<(_i59P^-xR
zFfC#F^kw?_m2;u#I?{!lOG%fZ&zA1$?L~s65kS8JyuBEN*EoMpoas8y7~#j!=Tn+T
zBMAhTn&vv~{CFVvK)=E1?UzqKKJi3oPn9;v2u7v~?Hd@nD}soGk&tjjpz-W~qc7*T
z7vZ!YNwAd-Gw6x%bNmzGze)c)mgf-XszZK)I?fF<=)zMPRYnAySdH+bRiceX@{=Fw
z*QpM1exUH2u`7R_=DUVM4@C!ld~?=fa?EL=v6hwu&I~=O`SSdCF5z98B`3|Ckyf>{
zG()7Hj9`lHVb*VH%CcHWAsAQK3xF0W0;wV|DK5q<0)l#t0>@)viGbJm8C{%}A++KO
zY?>KyNbCX6-pwLQGzB(##eiD$m{rD<r#wA5CK2;=f1iJ#M^OwasfjaI6pobjdc!ci
za7hs$Q+x<8)2J{)r%S{Y1VzDFrO;$!%W4#{m8T7ETdO@rbxY%1f~#g6sAjYAH1KnS
z>sYC6fnLiq&@B_4dNh>8i;+7f>46!j%%)<7b$b4|^;KuOYORF#$&f%T8ImArg3^;F
z(*R?OgYkb(jc73VcFCZGBtrt4Eo^a`&p%`vw2KIZVNw~IqL{|0KF@!>Xtsi&bLS?_
zk)i0!mS|oij57zeo$_7r4vRiTpk&Bx!;0%ZMW7=yp!EPUZLuY81(6bj2wOz|y^KW}
zCahh?vh2AO5QC-bCCiU_%c6_)4peZ{wS^0iOi6z|DGQ9mRV-Y`SzhnR&u!5Q<s^~2
zM=i)inRdI>LTSg)h&^i2OxdUwH8-CBxFRi1`^%(^V@AqtDj`XNrbTj1B`h;!K+9N`
zja?(Nq7^J@xWLP_8O5YWRL}a$Cf(~aXwW@fbZ-^8cN2~nqezQ|*u)qTG9jiZWGljM
zVvK*(Ow&&d7GjLb+GLhBgM4Q|RL$ZS5fQ26k2*y5BKsv-gwySe-!9T69nsaz`0m|{
zFy{HPsv`)dRmbda-@Sm=&wuha6L00)xAFGFHf5-wE-8cAt_%p|(%dd(kZ!>}Usi@1
zi^v?uWlhkG_R4n3-?nK&J!P9F)HAkeLOp+DpC;5ZAEyZw)Fn-@!<JNL3eAL6t25m&
zGl#3_9Qov;eLg4ip_ITTlsd)GG(DAFNL7c5JcaE#hutxVn@9MdwS#V3$SKUzeR#YN
zKfG2!eW2atuo?=aJp27Zpz0dR+`N*KH%5quE|I||=c7@kPcd?ku&kU&VFpqHQA>ZO
zsO7Q(Ay|fF=$$ZUL8v<GnwNiQ7LucOQ{_8lF@w+=Mh4cHbj?wIq?z2zg~9?_L!L;N
zXezt9m2w0rJg0!Ko0WyUlvUU||Fa;ESY=Z-XOVgSm{0h;GCi;I$_{RqS9Tf8xz1zI
zRLa&Xj_z}zE|b}yunCI_yv&gJoLzrMd;<<_C|~7Z)Q~k!)iCN}OyIA|yfkY$87#Xf
zic0M)YcW(0O3r+0GUzND({M1AdD6F9ey$(a@qTblbvC=jF{PX{>{g$raQnXd_Tkgh
ze+C8!LpwM{^#Azs<bOa3Op=437>h|U{r<1%*QcL)sHtGa8K)^s`9JjO_s@S*_vB-$
z{y9$O#%X_7t+{csL7b36I<3mi6pA&;bQ%>D#k*w!lKGF2DL(`pAddo}OI%`p-0jd*
z&I%Ko(XTQ82}^IDe4OivK7ul0pa$O0LhCA!LL%+BqMsoeI)V$1v|Ja%oWM}xLJ(WX
zl7DFE*tI&!Z#c*etMf@_CXIiBumH7r^6~psT@k)|C<cy5*Sf0YhPLR4aFofS8+0Ub
z;tpCR7fkYjBf=%k6SEY3>;z^Ly;Azp-=IZ4n@iEx9z6g`{VnZ&NO8BB<FlK!00@pG
z!drNJxxS8+5xaJ1@npGHnLH#v?q8ar6x6h&^a?(Ywak$;8(R8cHnV@qn=zIqMc(zw
zj(gti7(huTo$Vp7u}uSW-1MX6Ku2_1B8AqzDH56>dcWi1mKm;li~!3{i_#Z243^oy
znw3JgT4y*QGgYx$rx_-BZC#g%=TOTpUM(>tTZuNbQu0_(RH=EgWF=fOE3gPiUMp*_
z2#|7dJwRKfmx4OY!*+k~ds!cD7hj7H@OEFHMBD-F-TaR22n-XT1_m!T=7a#sL|i`_
z`3+k#<wnL1cww-<T~JX3)$jQumsf*6@EW*THuXC;OQnd}IC8agwKi>#+A_KZ`ffKP
z=|l)QD>EvQKg{`IjOFJrtfRO0^BvR;=A`F*>Ji~xl6ptSnwNh}Qtw8GZq{iIW^%qP
zKcwe3=(tskuB~)gpWP2+9rmM5aaAvt<w#d@&6F|&-qXU;ue+7ih|DXX?2%Mm;Y|UU
z4_j~`6;Jn*=i>mSN616MvqccA^!x1x2Ep919Z18OG=tLIu^&hi+Xua;Sl$mLMQ?VL
zYw=J!R$>iBVxxZzpI4(X@)r!$U|o0m5~TeQMo+n7v*`CjF=(Vb+~=jOk27@>>|C!5
zEqiEMALt(9bq_78(-HBh82tAB)HnO-hMKj^p!n&Q_IvzsN|bwUtdsy!!rP60uwxWX
zGUKqvvez^ELJkbHKEgACV^Y8f`LUicrPdvmdpx60hox&BAl~5_ksiB2d)7Q--m5xY
zJFO;L-j_R^&mCMz+c;iLv<3`{Vs2Z`M1;sLS#$6$DH#l@orpRZKCAJ>S4scENvF3`
z<Tf`X0x_lNh+A>`Ic?~da@9R1l7~J|&%}8t=Uru;)e>vwy+4<)@E$aOe*zSa9dsff
z>5EFxn<pQK;v6|I#^;Kd&O5VERh=%Ktc+&^%PyU|HtE~qVb$2li6e4fC5<n=q}Uxv
zPrE)+%HygoN`lq(k(Ma(U83~7>bRvRjkY#xIF}=`bu=Q91=zd!O@;~pBg1o5yHvT<
zd^|NKio3UzRcDb%@n_e6)LeE`bJ<PJ(%a-aa)#bQl)BOjH5c|o{+^e3D(7HtX}5G(
z*Ye)%(z{i!&=}lx(vs4oy1{YLzRV?Tr7>sn?dTE%-%%+_S$_+7XQ3$fJ9_eF0~*ga
z>`u^fh_;DgCH(3c7L|&hno#H=8!T50oSfekXjS8efG<N94(ro@Hyg}x+j>?^R{8Fd
z9I!%laQZkL!+41IN!?&!B@;$p<>XRB8t4wyeYxQ+Ig?iR4R1l25INw~@Xiosd^Vu7
z^mzfKB0_lRq5u2~pZ-Zd{z!v;Z|9FN^h-g*r+-g>_8s*rfNA^nfYFdIgG+$<Bq$bj
z3G58QAHMT-xZNOsHd5w0W?EbZOCoiuhZ{}=tY{zjSc$Zcf=;e(Z`<e1GgNmEZNxIq
zC=tQs0|$8{?%slka{=qfrL?Dxb5>NG%*@S3EsLCPX6Q9RkzyOXONI7=*l0BbR$T_W
zW;MNeZg|H`Z~~=<vxw=y(4jeu^M5?J(@-$HWYZmb%E3r~P*9M^?$w?47}(0IIOA!F
zfdAIxL&R-n(GJIlG;>m4K0b8O5yZ^^jAJnOo*r|6r>QjF(pKfP)tjESa1<7XZr{H4
zYV3ygki|;m8spq<^aHAHIj!y?@2d)?#P=(ZmlEIs@<WS)xd1ze`F8Od^6+L<oIBd*
zDUIkLuC>p9QyS6ptv*3?b5!Mg3u|x#N;75cZ9x?vu`7anY)}R0_nmt-#JvzU(nUFr
z2KYK56-Y(N{ah_l-c!YDQLZ=Ulce)un&KsQW$e)puFY9KvgZ%p)rqG+8W*a9;SH6N
zNwx#xubN*8(b|%YsS8Z;je!&f*2lF~b%a}2yW^FAAntY%UQrDJl$Jd_+IAu0BwxTB
z)*Y@G)j~X5^8$intcm-Ns&ax?*>N-3SJ(l*#38>DJ3x}@Ldy@j4uu<X<_7NJ#?4e;
z=7#g{BH~wK2A7P_?~es$B-gRR484sRHxqq@7qn-fzQ?!kg|KZiBdF>!)CZ<I8kwwu
z=EZ+c65wA`fY++9-6g&KOAp=4D#HsFm-85yXXZrvwEn8cco-`G;*UE6=6@Jqz5--#
z=JUTJD3$gC5A-s=<`tnmIP*r_MXu>iw(~zS`Tql#LH-^TmvFxa6qh`N9VnOk^d6Xh
zln4Cu+so%W8VA~sN@+n*MjjPP6c(jWioYcGB~1Q>5+(L5a1_!ucW?~i++4wv#1RT7
zPldAtJU_H8PRxzfOO-*gz%fgD`oOGBBe<j<(z8P3Z+KlAcf~!OW!oUf_3WN>@}<mo
zS)=1)?6Vq+)0BE?ywT3-Ft4-Mz(hTN4e+y*q-S~^TMXjUjfI?EUIU#+3F<zKBUs2=
zN98dj3#$5R?oG^S;5=mH8Hy2VSn>>^ifz1BpuJY756)^{Gk?Wn<Ndcm_c<Kis<B<v
zS?BgD$cw*f3;8R;2CV&45m~Q2i&{0kdMiZla8@(X#%L9%?Z;;WQ_;#aB|`Ln%KHad
zHK^jwLD+QYqyG@edtvYxB9d6;g3_Hnb-$io?*1MF!Z{I&4Zr`mb2kJB!ci(HV!0ZA
z{yBWR`=ys!)00=2DkWt@Pe5_ts4=%hbCWe51v?&WDX<aTa|p=fwC?NxR}`8+tAgwt
zO+j%mM7j_B_VfGDUHMXoi!nfdE7*R#>@rDnj7g%|NIQCQ#}o|l`H#=i42DRn6`vw=
z96RC4)U;Vbw6egZqDo=z6Hx_w1zM^J&Qsr#N2UIy0b+G_DW3fEA%{Vt?htIM3jPpG
zDV%8}u7qO_H-Fbe;tbCu%~aElMo)?Qdm1idsL6XKuoowbLJcO2AWa;9Y-Zw!8ebZ1
zTiv&y)mX_#fr+{nIS)1APsnPCNk#3_R6@gfFCB1Tgkdaqbiky+37swT!IpsYoMGt_
zXGm#@Vudpv3OfcwnwY#M%7H7?21Y&T$RB1xI+ID3jMAGcUp~!qAVZU4G)ZBILT*Vc
zp&WNPKjRarPh_>qn?qZFulc2Th4%=V?n;@WB;Gks`pjg?mPW}n(K(NliIkH78qbk#
zsR+b84%3PP*G)GX1XDyym@rig$ut3^Kn&(`*+Dc!mG4Bub-7$@H=WDvcC0jhz=^>o
z0_F6nlPKJy>BCEGu1vXVBat#X4Wbal*uHV6dI05$At;;;sP3hI!i*m&ML_i?D6@Cs
zrktSN<|q^u=KQvpTACr+785hVQY;C}@hs{*FF0k%cE%}qiO_2>(uP<xR@dX4n;1<5
zrU3o;UN0-VI`7hqNY-1d51i5(=|Ss${5SZKFpOA`Gk%nafoGywn&TVK5wYZrepDBG
zvf=euXchW$#mXFinXP%w71T953n$BHjDD#ymjr$4US{V#L5bqBmvCehSRl%nzDMjN
z9=nO1QA0FUQ@SivZlx!sQ6%sIa!<{bFCz-aI3w1Ul@neGx86D04m1zq{D<$LxzbPt
znhjLngJWf$-43pFyRxcLIeNNFtFcc8%1o-WTFkgf$c_Ae^_)H$Ut+^j&-sKymV8S*
z-JBo@tQcwOz~t46O@Wn!FhhWqT#d&vB9;L;H`|07ToszPku7MSB8<q^ty<QG4t;LQ
z?xstr#mSi}!YE(0-np94t_b6p26er-IQCPksHRj=Cl%8Z9;P`0^Wjd~Z04O2_4Z(?
zV2^#K9GX>sD_;6)vSr2+tkQY~>F!jQS*$IE?Iq4EF>}fY$5y>bE;qR*w~{a+Ch`Ci
z!$D8E=0r}JGR4QXz^GRNI@dLU+Wk>wT8SLovS{p$P3IJ^8XnvT!i=nQ!?ft>sbN*W
z=x8JDPAP#JE4am@T|Dxq%?ail)29@&Q}aqDjxo@G3!xQdZ1?}<(Va9*qx>{C7YA5p
z;kf#zZixZS31{jRVjyKlZuE3&jkd{tVcw&6#0agXu9P^OUp+hK3NceTi3+LNXbE9H
zL-y%LCxpgCvSM-7vbC&><{OlV>2j~cWke3M{vYS1PJ##e5}d_p<BKR>X1zovEd5R;
z)FtnKpj<U^#Nx3V#moXlj{GrnIV;uC&X19iwwEkW$KDmOFEvpx)*v0zp^uq7$MU!L
zpG&*r7M)B2_GxDM97>X%VkfUPCXUxVrw1-y49MlBAN~~E(*rm@Q)8?8l^oAqsPmB!
z`d<WSVj1DK4egEP0%w*8+m;K}-i-Hz=a}4or=7Da%F4J(*!AKNXLgiq?kX`M&k$T!
zb$7d<yIAa`en(lV*UHDGto2M+7M;xEYzZU1w>lRmQp?fP=W2~HuC%3DaZ^cGXTjSs
zM2Q6Ccr<d7euN``YDcYK!SCY^5C?<SXDfa`y$J<)hUz@I%Z0*-9q)31!Aiu+w9A!$
zM3fGV@6X`(u}n<D@a#9zzy3ytKk(*fY|O10{_#zJ<bLSz*YK;`Mq3e#qw#_nmYZTp
zFdZ1NNcKT&5h2}j0}{y9A~udp1JNLt#ey?+FL&{;1AwV2N8!lgwCsGmMKX!)LArE3
zx_>cP<9Zp2-8NgnLScMMbr2Rx;NbaxyTL+k#A3J2R<ID3XsZvxLUKY&!T08|P#dw>
zZL<|DxWP2e9E62n1WL*L=CRNlvDj_16)b#4Y7fE!Cs!!d(3{7?Y{X)>%~r5*|L|i6
zLBS|trDl5bP}q%7>^0d65>#Re3kN|#En%gEeDhE|Y=mOB&6cqM8<NgLumEd+2zx(J
zoVP)--()LLn2>WG1BKCqSFcUHbu2C$vDj_16)d#NImJO(XhpbwABCPO`!TCn?6%no
z7AoYN$6%o(5y0OK78{H1{We>{LWZ347%ZgQWD)NMi;YG1ew(deAwtr53>JbDVdcBQ
zVq?+0-)1XVu#j~gg9T?qa{X?9u-I61@3+|s7T~kaW1wI_q&Dvcij6h*Zj-GbK||7c
z3>5gFf?Qqjr~K!-=I^8a&7f!;zWd+C|KHy}Hv+iZ(rX9`!Uel!E9Pnu+jgIvSXj>d
z=XRg^ZLUpkofD;UmdGDY(!aS@-VB@|g7|jZa|vkwDsg@*u)OTsytJG?-{qTsX*qbK
zBMsr--(%<Y<?Hz8$~mQ&L_4ugC;G7R3?l-GuYD*eq&K`}9a@Ebqw=y|LOPz&wf_a4
zXWEw$G6)j`G%_`p!4@YImj^yPD3=~ZD++)52gG5KEH1!6fYdWG;+qrQ44;zNmkfgO
zCBL8I)?~9cEuKTG*<7lMk5$D-ig!Oc-hJAgF1x3*RGncK!HDkiy#s3C^b4FSHFgGo
zX$jrst^4}SxzO~S=uw<YNspnjr8ge-5-jxqY6Up#je_UcPn_vFQXk>_QMpg!s3(7k
z;8N3Ar=9N)Bp*l~sKd9*=N(Uma^|!z`d9n)dfC12TC=s5j2qki`L^>7N&=HsbfB4&
zy8HQe_kH)Jm0E*hfk5F}BaT1;PDm+|?wIsV5fr3ZD4Hmm_$ZnH(nJyHp$W;?I2oo$
z%<%K`N7vN+{;AAEIR<K?32*|<Fam!K(<s{!iYMK3Qw4fM^`3<itzw#E3nglX)@CTP
zgtA~ZsIyd_SsiH^k0QT*<K17>_?bFOXS@GG*SA3p@BZoj@N2apm;?sFidisgf_W#H
zvo47(B2YTUmAhTUm6i!fu4C85l6yI9<*xc8FyYr+81fc9@)Gh%a=+k<>YRUkGJOZX
zG>5(BKIEjXJQ$<L`d%V`V`}ST8wJ7Q80nHmouWac_SYEpQ8USKhY;;l<Y#Y)cy2qy
zR?V}SNEt-+DZoy74Cq|KMrAO=<km9n)Y_*yKcSZtB4zzXfiQT+-AlOP8X%Z;?pM2p
zV@3-St{GU9T|?4BK%X(C<gkAS!IcL;7V)_xm`yzHwtE8)4$P?!f`cOy!j9ljaob5?
z$O486v2-OE_Zq#91v(l-Q~D4v&?zRgn}Bg#1mjB2xZmmxG^}CH$w$)r%_6Sb(Y#o*
zE_P{0Yq!oTNEy=Rt141<K7h$d&V(gelpr{x4BkrM$Bj{f(U#Z@LeGEOqT*B93PSqN
zOi-zzbWVAh9`#z<<2Dq(VqUMu&(o?n)kv+f{_y4H9}gJ(^kt6pY_mobmXYocVvR`5
z0^OK3W?=mE#5R-U9w1Zt_;=hu;R+tg`xP#j-vmqxivW6u-1fd>23%+!3IVJ*_I*cM
zLZwP?YJB?bP4e#L-^zdGzf9QcKI_^SWdTy!Z!Mu_w44hBq6ti!DiCTvOMHPQua4v?
zYUFcj;&ZV^r_o-LO+ES~71o0!+hhb21|JB{M#6Q$*$k?JBXP4TFi%6c<WS=Y8t*Cn
zQ0swk!zdg3cy8wjSgLIS;{m8F;W!Ft$d*JezBfz*?X{B=2_1i#EXaq$l-M(6Iah$(
z^oV;7c8wOnR}Xb<?<NHZSJCuL^7YNO(To^4(`o~lwT&nNQwaCi=F?a@6c;b1wI#;}
z+NagP&wR(_tady2n#izjl`LUzRLCvk&fc0T<6?1+;2?K^PAr;S^eU5t0H%nrH72<f
zn8`yk!gwm13h#fb^zU;sL;u5*aibbC?&<9KT#N`xGALRAmCMJjk=09qCahyV&SqD-
z!ggf-!=_8`V6wHmFMhOPwN#|WO$6j~WqrKHDgvstNRyc48zLPjpj_B2_oS^6$$cx>
z!#Vam4vZk;RGbr!cLWNZX1HlJ`U?dC>(MeE@rVsXsg-}6%z7akfdi(kUDE$hfEp$M
zTgeR$H%S!8jF`?$E(E<}&P--j?oHNrs14e;#b&HzwA~IzL|u_89rt#{UA5CI;H^on
zT*ky!e=ta;b0yE_`=mjp3||5=7dei%5)XlrH3TsP2P+bPDPL%7`SaW*a;IcpbFfLZ
z)s~iTx@v#GV=|t_VrF^bof<8s^EOF3!bG-JTN%08pkEhU2Gq=GOhLH1R11BN!$E44
zVutH}2B=3FCM!rwuP|RI4zzBaA>}h7j;Lj|teA4BG7V#o^wqb@l~bZ9r$SZJg9NJ*
z{t`pxaeb!7wH0BEO1ozt>xWyzfk=n{CQP^mnVAhm$uNc3PLEdq1HnHZmqFJx69hCe
zH8PhGq8@6up#&d6Lzgi#Koplfc>ySw9!MXX0lk;HNFTC)47edc(HUBywLM8^e2|RV
z;Qy?x_r=X)pzx)mSS#VR6C4!vWjYXElh96`Ec}Xp72f9SKzL;Z<E<~71K|~as<Sn}
z;@@dr9}5p8BrVA8f$+=^?OB?aUPw;4BY)~$rwH%1h39w$sq4#bUwGow#xk5N{J@{G
z>!6G29UB0Dv<ZMb2*8?Xz#jl`Hd^~s03(0O?gKHoMF7S3{jL?rdI5gX!sP%&C@j*P
z3}WC<DMJ@4P>utrNpS_hSfTWyh3f$Tt_+6YWB>zyimL9cfW?jWtw6+|ix%bs5X5bc
zKidiee~Rv0K~Mw2tU*}T09SEBEwoxRfi-V!1JPc8aO-t4kU<8D9?{hUs9AmmKy!@`
zbQ=H%m6$knTX(i41{o-NL+4I&Nvr#o0B%w^U$#IFsxomJclvBg3^K%Oc6VG9Al%!d
z=`C_lp((b2bTW`Z29t|i1LTr{*j*s#Nid-f$U(J6PVN5Waxscva=FNv*5op=(QOnA
zP79)c-WUf(8#%LkbFwi;8BAX;TCcL(Ah?ew-y7nfaHDbDo}G+g5W(b%kulf?gFc=u
zS-;_H_!$4!p(!IWcdvroHoyrFFytEU#=!`$&GhHLr|*nlpxO@L4j7j-hS9bII7kMI
zOpbL{J6U2_6lP7Hn<WnrXwvM7a~Qf3E`dycq5kvdAJcH>pFb@S5QXyt51!sz{VZp>
zr@&DldCg3DRFFaj7K#$n?hZ5k^5*3xQqs}(J>g}-?4;RMmoaPvHA?sP`fQ{sipx`Y
zRG>{t3-<^fR-BSfSnk^H(wqVn%syC>#<c*8JBKR^v*j<qpU`}6`XiTRL8wN5<T=ZK
z0u+BN6c(>{k)7l38!j$o46=iWEo{nsev7b+1hqF@TW-uCtUY$paorI*9f|&N5_d}G
zNO*8Jk`&vAM!JVV4IB)biF={geY%8O`=VB`&cYS)hh@8knf0Xnr>yZPiTfNdkhS_F
z{x>@M9Xn6Wna<FXi7-d#YR|VUxMC`Q@i`ZVvvW#5I-R_PCuKt#b2g?uVY*Y+gyC%_
znyBUNT6m_)3el$Im9l=N9{6874$b>fqD#s*n32Tif!5s0lu}_%m9}}KmB!RJFl)e8
zIT{pQy7kA>hFxm6D5WfAu5<8pD<j^n%gS(+7&qoY?h~wrPuFD~R!rKrg&9kKT~XU6
z$j<X{ktWOTs7H=_W!iA##0}|iIy4J(t>tBfhPxG62fXMBy7BDdlV1}EiJ72W{IYod
z9s@}N6m)&edF8nDCIc5Gw^dXWI-IsmJY!m9ml%i|aCLeASgRlfaAP`f+v^p=nw#Jw
zvIUFiZ@u=}2^=Rm<uP_gtut(Y{srS>+Rwe$z-3SS7KO1eo(}S)b`3VWqsR50x~}_#
zQVMt+^(&$Lk~CxDbxR~+4w9~Ei|sDS)c8j6x;qD5u(C>!S}75I4hp)f4OTAnetY(A
z^xFaWwD2L>IK7g6Ufk)1+!~pWUsG^7i@GKRsR;#lGqWk4ZPl`n4w-9z?<8A5Ee(3m
zd>X6z0IA6@yC_{g5HKl64;(}3S8yCO{TrW=!C8TNg?g{5co4?9J^r5xQ!5nENO{o1
zLlhMWtJcW$PUL&x?F!kwljDtn?N+O)bX#|I>5b;L(~1F*jJ(xpVNOl+(ktdll~1Rg
zR9@+H{pL7-ez?=(PjK#kFWP`8;JKN(?5se}ZY4@=_FQ+ntz7yB88?Uf&kpuLoZO8)
zbFc@;sokGH*Mnu-OZAya@?jEpN*1O9)PK!f4~FM@-7kZizqQRDxa@r^fsezwr7l)6
z2jOQ7`6!7yU6;vPqQ7Y;09+=Ft>cs~Ioak(*7|m~jqORe>)N}2C99{iGM}64ol3LZ
ziIeb}dc6~e%C}}}j;(6PA&;SDic5*Jr{8QOTp465jZ4Q5+uBp{b#j;8K%5C}Bf+J`
z87ox1gJTMpWfM_P<jqHH{?-&%nz^GWtSroph+TfI6W{i)&lrEl)^f;qigvHBTIQM?
zq#XA*!|=~<%bPoYk0@WxHS1qD1y7{4ysKF}Q`z~A6d8^l9I43bvYSdc>(1y>29afd
zgxBOv`Q*2@R~c#4){xDiT?kY%&cr_xaFUNDdGkqAnl*8bb1=#rbXSK0$liSIc;`J;
zwwzv2<W1UEX-9>EV5Mg{=ifJao%2AnnQtcO>MM;+O>P2zI6Bd)O^;ef!Rs@;?E9<a
zHrLeNvHBQnj7f{<Kcl?cs<L=~*k;Ck!Mx;(+S<(NB<QOunog0+hh7>ED%&9OKZ=la
zqHI$!=cKufk}BPe<~+$!ai|vmqK%r#uZ<NiC1bbz`ps<Cgjx178xkJv((ta#<h5|1
z+eOLYx)e4F|Gddv!~X%yFyWUmqbC#rGM7PB7ZjI8+yf|oty@Wt9Jdv|&#&m?2!Uvl
z?{PyW4Rp7r1^8xVkc;C}90wAjLww2K&*T2cs^-kZeJE8=6_@Yr3rq6wuY){%emK88
z{BQ+4IN^<Q`tb7GLD36I|0+ojtvNWQgavhY`F{BMNlI<$I@5(x0?_5?x2L-~egx%d
zgrc`f9X~pMR8M)FzS4D~F*=T;-<L9vM$!@zEX@rz#PN#OS3cg4zr6hTAPdl2_ve>~
z|2$O4I)Fgq57ucR(H#Eq{o&_d<l$F(`Qt$f2>$TjH1vBMu0$~A@bAN4AAVZ#L5D!}
z8(P$GJ{SUyM4NH>Lc(m!QMvezsMlcdDLvAWEJLJ!<27F<lQu$kN`L~|MG22~1zkxL
z=pFse*mvkR1Uabt&<_%Xci~_y2<?2+gG;0{)2;$tj%bz^;;T^T)SX3T$Zfg_G6ww~
zx>qvuT4--+RVjkTido?ffXXv;0>IsgXEe}Qy|e#Ea3NGRPliGlBv>2p-~|~)n`{=~
z!3gkw`p)1t`#12eTnq5p38J@6whQo@?)u)G-|Rnd9@gNM5z=(Z-2y!AhwfgSmqAE!
zx%2*JR2^J{Cwo;A+GKwlydl>{814*y+TV<-b6J8nK?#M;$*8aZ?_HAgUBS=$o1!}C
zRptCE1h^nn=+vJU1jK+M?VSmj_BYuxzk>jO>x72Rf%Ad@iU`)II}<SNZ;}cR1lTnJ
z#?ZZU;IbgVP-+wIErDr&lhk-1;IvPGBg!di8w3Hma{$}OeIg(TO117x#Du{lO=w1J
z)&u~>^tdh~U=eu@xoYR{Erkh#Nt*B$JE%k`N=V2zav~NH*pSQiaBnG07*gbVlPy$#
zA}EJPJ4#{^iA_rN%$<ptFxV2z7!l_s5g3f1MFt`k5gEyahdYOb8G;?df-U*&f(8%5
zm{Aamz>MU@gT1pLW(;;48vL3F%GFI61+fUwXg#H1ccx*2V8`g-Hrdgwy77lAA@p%A
z3wu>BQu6D13I?iFc;{Hoqzz+HXIbrk2d%K{d)%n3DdAPLHfmyrt&!p^%`m%-1}z*d
zFpf^Ow8=Fmp|Q2ZBtM6rG9s+#Vx<;KGG9kxD09^`O6S@YFjleYiEXRLI#<_sv`v1m
z%Dw*miSLuvJ<m_lE-sgbcx@}Hf=L!CwzP#dcDJa`T?ZqhWZ+bG#wss6P~_f!fiBIw
z=5eYZ#Z#TXMVS6NmA#pUXdThT_!;{iX`j<zXHC-53%aVLv7kyEiRjgUzEYTaFG-p@
zbFp<O3bzuSZElifnltdBJ=yRfuV1xYmbF!WMzrnfZqan`UU;eBsOjTqlI0O|E2T)#
zG2J^6GSD*>fJGw7D&3Bb;*QpTr&*caB4Iu48s#S2$f`{MbvIRX@zO9f85AisWJy=@
z%nkA=#VXsX-{`yMq0hQb-mrA!vMk#Am8!mCm0Kvj;ad4qQ6RRl)%wQHR$aSl_J!k5
z+SIUD4ODVhzrk7Iu{rUitG@dZv`jvJqwifyZTA4i@5qzs@-U}r_45&b%v0gVnR$~&
za~@xGJ4Fo2$F!!JZ6{)oQ=wwI*G(l1k%G`oMm=^%mCQe>6pq3e6gs&luuK7bfes_I
z?y!^7gYu>Zf;v+c0@jUQ&{U7Q9`Y)~tD1LqdgdaEd2~172KUz-$BC)1XFMp(#^%CI
zp_OqYISWBAs4k=+Xe?EK3z8}~@)5t23)z43r8^dlN_b^w3iXwl_70{fzZ6WbyY5O;
zk|uqaS0%&z!+1T4p!5cB<@t%Zn$ly<{97AJNf&Z3u-v#NSz*x^uXb}WPjm;^(minN
zmbRfZ<OI|>TftekAiKwFs)B^0dUT)H$uw7*P4@uqre@=tG&_%f;+mbG@UbVRx6FB)
zOC8<ZV`-hYnT9#BY7cpa;q$d)f%j7VM|>Nq&#1wp`U%+&Cs1>yt=)kRQm9~Wy94zm
zcc6LnDeySaJ*!D0duCLgbpNfNdK^!V@XB%#Us=kL6*Ktb=4gec!fC@1VyJzKKTm+(
z<<IjBI4WDxUHx=_)30xAh0@0m)C@Nxs%z^{S#at24H~nzzPG;N!8SCFO6NU_7Xz#u
zYX0600vb|+s%*{+^3o0KwQ%%P)_s&P+m2=SgMI(>%*GkGlO&gQTuG6NmDa=%xrDAI
z*>AfizW6bv<7rCS(Pl0tt99J(_7cUWVl}FdP3eG+lXc5~yxLN-+EmZq3l!i_3-C*a
zC<#33maX*_t9aW=TSnu%8O?J*Hs4p4=Zu$JyxJGNwW)7AsA;ZI6K&X4vMc`(+HSJD
zg~T~GSIw1IN7?g5#duoH)Uqu8(!*p`Z$3BIO>Io^aH-<fyR}%|&v!#X*Sz$GTj+&B
z98O_-=x6eOP0DL;Cj}Kz3sEJp7){@^8YXe%`PdSMmeNJF{#3jy`J3oc>Gpa0Yv>VG
za+G(JOJHmg#TOX6ww@~BK=~RRe*fp;w}+qRRcj|lNEw}X{D*!ePd+^PQa*18lER#{
z)|es8DM_&p4_cG@^%I|4*=K9%q?~BG8dw6Z=G7*DFPmBPI_y_5t9Zl^29qg4u?FGd
zXJ5sQGJV`Itr$Pnh}m(hDD?CFa7z>b_k=wx2xe?b<TCVb=v4p#PTdjyv?2gAMc3Pj
zyXya115+8KjhiNdWYOw8HF4Y=|5~;gjwo4<TTAs8HpY_sEor5Kn8CN^j>hg{Tl<fk
zln$_e6Ti#T6CX3TxOC7@EE5mi!PoKe7w$F7uD$WW)<3aW52@vs=_F!emtI>OKRt<`
z$*6(O$Il<~6zBG*1mV0Nr~I8@m04Er&LZWpB>Q33-#&aa=JC%TJ}T=TpO?E*UPmqq
zH@m72D=|_?5XR0@971WA#R@(k)wqOgg1_8<+ZX{7=gn(HGO0{gq8iD&0+s$-`75v_
zoSTB@>&kPB;7omZKeEO`c^OnQX9z1+AW2f2i51UFR>YIK_$}1`&x*a|BDQN?KW`jc
z+ATpjv7#-#$*d_UM{n8lCOsC9oY+$v^&xwh)P$?71#B(3nsS_5!yZ@)dh2LlEzG5V
zL3uAQj9hs9nrz8Fe*3U8deTZk93H3YrTK!QDdU8}c)zYUg!tvcINh$zeP_mB)(BAx
zCnuu5y<(vhOesvwinKXt8I|d{1WsZhD;i<E>K!4S0^7AP_cz4&ouLVv4b@U27(YA<
z{yi#ynx;g5#;2_99zhFbU}o@mREIo&rt0pfHvQIqRCVFUS*{)j^!J|ix6Zfy2l2r5
zqkir(+40e3Y{S<>$W*vqmKLh4Or;{<=keFdV>nofDB#Cb%Nl_$Cvm!@LyuqA)~p;Q
z!)7F>iPGWj6ps{)ah&p{PVu|QNJD*8GfKU|73u$JDSZ{lLphlFe4|thlmPaBJZtu|
zV_Eo)9)E5xuoBu21@TTUce5`IQ<NBcTf2vqwMD|p|C)zOc5G$3=Vxg5k4%d`%Ii8`
zv0nJ#7*}7>Ja9CUYT2NL(RNb2TdodBa+dG$ii@U^v0i>VsQAUFbpxnx6n0*VH?gOv
z5&1@b?b{SNC>QvIw}n#F@l=I>&wIs*0@;KgHd<$qh*lVdqj}sgAxG6bJAbJiRHJ5n
zHu_x+VI$w$5Y8Jw`Tqqs8^T51P2k3v=5D?ONEn(E&b^&{f?LBw9x;apS}h;fwASuG
znX19V=%=|S*I2i+=8bPkYB`$MozB6h`LlWFy#nivUV7iMaK%q4<7HQF4=kBk|BQc=
zWBMYNb30FKPkJ7kMz#jz(3Z3tFFLjr-EO0jHGrGPHpkwGv~T|0qjWHOYq!0rxwTrI
zbiKpFw)4jNPM@b9{FNz<VtI=iK5?m8{n!ki)ha(-sTGT2bB5DTa<~8G=zo+x+|ZXn
z{vH$oH<z(ZDioJpMK&jwVq71df6;u(z5Mp}(}OI6j#62Ng661FY4@URpf3h!Dj&rl
zKJ+9d5+kGu@fw-P_)C2A8sD}zB!w|Apq>kjiQ_X}oJSHZN*jMYb2=nNjLI{1^zk|J
zv}Lr9vLQA)Gt(kQTKZh-I1}c#aa8=yB_e5}X}-H!n1TovEb@iZ;Dyn;e>2hP@5e$y
z??rIhn~p@I0B}SC!l2G1?;n-Mtg=BPz^a}NJd&OC>5oF`+$yD&BFKSKK5pt{jqrZZ
z>C-kBCxsrEsBMiLX%wI*`7>+ex5e{8c<8<I^#1AbNm+Z^HY2%}(9Vu#+~S7H(a?e0
zFNI`1m=TPSc8r?r){cine{?%Hx_1hpOL1o`qhn~;=Gx@E(fKj7H@?v2fZ19pqddJF
zOz@QgZ_^3}KzQ6X`3-*BkjeFepV~BG%RCl@^iXY>Y?Sg>D2b#6w2Ri2H6?`h#M00F
zCXR_IgVG42ice)`Ro=LqU9F2_bXo<rQs*U?GGgYolV0NJ<QcFoe{IaQm~x>h6NQ>#
zdqJ|`x;PD=KAFiEGufh;Y7F#?N=```6;?BHox3??qfO&{Gn&k9+3y3@Iom9lmvFH>
z(H$1cZ=_JrLI^{d{hB{Xlk;GQq)iQxx?dz2&eaZW6t|%-L4|z8Vy`;0m5|9fvKt$}
z>BfM<Io&qUYU9sEe==L7ywvq2{uuXF2rrmy>x-0pWM0QZVU+S7+rXKdS>&s}ArV@`
zLNGt&U}{g2+4I7lKl2&Pf_w8dRV<uFwDC2dZ%{7+I*pUB3>o+v@-8L85+k<ECujdH
zH1U_BTm3a7lKegfvyreasb$uMn(n)qOT^8Zr!m)rSk@xMe*_c=bAn_uC;rHAlifrz
z3bio7bu*scAD_Uv({udeTc*F^Y!*fEb`(hxd-b^!9-I|ck?<x6>DtKIeX5{-O;89%
z*vQ95=w-8@|K5aLk;IVy8+3xUtg2ffvFjvW5KQh#U>n#;EnHj^PEw)5x-MD;p{2V=
z7fmIC$AQbof2J;4k@_RajxIp3QQ@e|lU1ir@6R9G01XK&sL(C%>Gy4R=mjl4k+`%_
z114T)V+dw}>e=i-Zvw+EIo=&?tP@TTZ2ol!y%++5(TFweGE|XLe5&9UXs~Qx^~rPG
zlb^~4pEvV1k1n&@8Kuk{Ip@Y`7+x8qX>X2gy8hrMe^Rsyn_C*!S{04zCXP9+j~T^-
z(6mVCouCD~Goxs!1WK4NSm8_cy%&W<0s$waG6t<Wl2Noop%nBadl@AN^r|P>%P5|>
zsJnEb635s<u$_zoq`tjR7oeoCZ>X{!c~m+CN{&by3We0MJfegQaV3Fv7&sQ-lzWs6
zJ)508e`2)oQm?Z4;qBpn4+|Ay&1&5ApKlMp{wAm2>BpZYDw%_y{z^l?O#~4GO(m!Q
zPG25=&ds75#WqrXP8o;dA@v>(RIxhNQIz<2e6%FCja|`za2R+?sv!(JW1kLbIEFU1
zQzU8>WX&ZqopCoy8@T#5^!Ts5wIyXSL{~9>e@>;2Vat+mj_cN?D4qakDR}avH<_O-
zi1f;e`^(I=)DsIUQG}{s+8Trx2bH^-WsTUgXwQ)EnE2gln&*{Hs~iUeTxbs<;QxO!
zxqmIHZS&|h&nx3NZydJ*1!7EviAsIKBkD*l0B;3&oU|hdCr3=N9>GA#SyKM)*nGbo
zf5BA3naJk2WLkkB*xpO_@%Uuiyi=|B=78J{ZO?0`DW@qDdexI0YszVYm|-N{NoR!L
zbQnkoof>lxnQ22cik*?!YGKv|RA4=#>{u0z*j%iZ;0UbQl5t*s|GMpVgbiY!i_6s(
zX<Zc{RSbd4K@}!)V?7I+M7oy#cgqAZe*)*+&^HseBh2wGP#$B4&JPD~Zjpbmi&Y-;
ziqkJ)v0H*!7#eZr=AvBY(}J=eY_@<sEspEgX1f%|K=6plxnXVQ-C~)8lXkqjq?221
zR_xW8sopX&@6e|oxBZ_gK^sCx^30Q>zm*gTsc1zA4)9RHO%$ke;SC0lKuXG?e?sY=
z0(O`XIwoa0l5I!{fr5rKA<HQf4kb}4>cLT%NwS<Zhtj)NTF8;j_qmk<+qqC%szh8*
zxs!Pb@&<&SWGBBz#EYsY+0VGf^c43bT!`&|2@9NenXyryaNF5m&J2B|w?dkumlE?0
zXZ5+Fp$#Zl$KEK3q2vbJuw0_wf5i>ti?n4@)Kg_}$g`r`QvQZ`K%#2Of8TbjxJ{t$
zxfI8eyIZ=Z!l<L%-_ipND(yIy4nmD2uv>i@ZF-WsJWElob+FdkrkGvS2Xl9sjJb<5
z`6)u9sLdE@*vWTSm-I8s;-A;`gFP&VO!){1kk6c?_{^Y3)+p^^O26Xlf2Ya6UD??s
z@0*vW&MtV@(Ndb=o};BgapY)8wzhjcrnd}s%nQg4VTf+6NKzqCAkDoWuCUc*o*4+$
zw+kqe-<~;T10xm0G|H4Xa&uqLnkM<fe5k?Zt;=P`gNykrP&>RSSG7<cXBgn(ASbU=
zqj57`D>&6?2Y8*2wUV1&f0=}8KD|uqo98WsU@NtGd7Vda2)ZUg^ezj}%ySl2T7sem
z*cKj;x*P5GObLKZ2t|c;fY4FyK`1c7(QE^=8<o3qQh?MqTflm?1*{j<vvzAWdSKmy
z8Z!Fs>RCI!!|KHjn`b<iB|)|v{B-p$ZEB@ZD|6#ln!tI~d3ML-f6`eR!)M|NVwR=r
z@5~%Gf-hxZkBci#x`XSN&d`3S7Eu@R%7NV;cvr>?K%YJD%MN&_EfRc3!crCFYa~~8
z$)McCD!*X;*4d$^4;Oz`%w}cm&Dx3COqsEFr?NQ+;|#@f<}nt}uzE9$p}Hr?x3$(K
z-v$^#NDb?Tz5!!se<*t0lYFH~c6zK{1S&^5s_mOX+Brcx{<xyE<#^N-%N^*TmG!<O
z2Na3}F%JE$K5X`1I5t`Vp=G|DQMw-DY9#VrLz?}vh0N`N;?<~hbxwgyrPYz@l}H@H
z*bvxlr|sgE3jKJ($|-h=R}<3NAgB1Ud9o8rC>!^xI~1|Be@7|knCw*HP6;_s>?q=_
z&|VFS*mz9*s7@cd<#-5=GMl6tRv)d$ZHR6b3od``_9WHgNHu)N<|_qWuAFv`hrkL)
z#e7G+lRq@^L)DXf5HkQ`E9G{LM+H&Txg*}WOKHi))ga<dac84e`Yx?#p(B(TthkpI
zJu!lw=$`T}e=>EgQMF%E^yezbi<T=ppD>1l{CvlXBP%TV>B2#My_LmbZ98^X9PYEr
zepl9U#V)g{?rlXXS(cRxg@yM$$PU?!yUnEyN*JZNvG<}Ppdcv134YnRdJ95JyUOB5
z6E%8(8XTajT{z@JUO8r1VpM^0`C=bCR;)D|114>zKsEPR1)|NkslHm@1T6{;xGYJ>
zYZv#OcDd!<u)0T{IR@kXGW0!eGq$@8uhRb?C`C;De*pFG4{s0u0HQ?+1D8Sm9uxsM
zmvQhG69F`r@y|UcmwRU)Fn<TNx+L=L8!0|{`u!wNUp|cAK79IY>?w%oRM4kyznv7l
zkn~qcdg$FLC?&j6r*Gd+zr0APJzXPRR!U>&3iN-ZyL<g;RHPA#-YWI_F<A97kJF#&
z8fc8p<LLi$nMWfDgfyP!CZF<nMetS3*Xys}e*Pc}p`Sjxc#6WM;D5~)l|EQOFdLPG
zfe=YY;4NXdasEBs6<vhg+H%ihb$&i!ye1C-gg|@0@0{?q$s_Y&#{${C>db?USr*}?
z;8;D)Sly7ZwkJawn`!Y?lGD>ze_3C~sP-fl?h3clT>)opuA{jwk9GySo_{-s_~!Zb
zmwubaOd866KDTkMK$v)MJkWm=n>yOZ8RkU!*u6Oy$~%}ny?OS8W$1YBXlz&}j@y`2
zsQEl!4NA7Hx6ZDbcC``8rIXh|7+dXVlx};p(ISv}^rL6mLbU~EVQJBqJ9m01luje6
z28YFh+E`T}5ABI8vqOd5-h@rLROkikq<ypezt~M@g$8XH=&Uy6k*j}D@Kl<bx1E(4
zb)hpHv&+fmE^3Q>(rR;5TixiwHG1kqbKqK-=mef8s{%<K<)&%DcE2jLH?)hUAyuG;
zMMIUk>6xEm?Hso#LSBelE-K-aZV?hApgk1%oF}O(zm9x4dfS)9#+_U2e)9clPt&MX
zOkAG&MTRw6Ptu*ChZ27==kYaOB=-W=t7Z*OvXfw!^PH+%l<2wWD>9AFGitP%v8*`G
znQzj#Um|hyM%&HxB^%jAChGn5qcs{JSAWW})+)2Y=~`AOxJvwUjbb^j`D~IYcx97F
ziP>5ED{VU^ly7;ia@5E2Sz3jTi2e^2KFMSD{u;v;pJCo`DRO_$b}mKo*}?Pe3;Xr{
z?ZbapjF8Jy0h3Pu{Qe=outo_J^+_owoDZiz{(buG!{0gB5<;YqG)K7nhyMHH_Y-^a
zwbW2alIKWI#X;veN|HO-lhahX*ycGCml0y|x(Zq9o~Sk<+a#1-Lmrl>yUIdY?)tiQ
z*y=m>RlW3%;>Lg3!1C~Hyk4qH&ZkRqh)wCU0&Nyv{WHn5d-8P&-9{y0wBl@P*$L?;
zh?KIU;Q*E0%p8U?hm0#JlmB!clR;9*q_`<uY}LRN1?MbsE8C=55fcZ<xNd`;5hg`G
z+H7j&c2YtVNjnZ*DW(CRs%y~T31au;>z92eiIi5tid%mpA(?n6CW>jBu#}R`l;<M+
zhy!WsMIv8kSSRZ~Y|s0qAieN9HeSZt<!(HMDa9c@VENUkdVq=;RVH8e1NKTJ88~FQ
zQ)vg9PgMjP&4peSF?-OQ5lgsHyJEMA)@EdPRqW@x#m*o-E_Md+iDGAvuZW$Eeva6q
z6=A6q!u@|z?*Q$^yCTqTta&d{z%J#dEZNG8^ZTh_x+4OT#F8}(q1q-+ODnA7oHPpF
zG6hYJw@{(A6G2-pYjwBzXq`j~k+bFQ2WX<O6iIu4@_k!QCH%6%Sx)DNw55X<PV0l(
zqBW5;D{Y-0)|Qc47-^oTEkn`Lxo}xqOVvbWr{aGyXlz_BY4=ifEyoyJsV%0DWL}e<
z-EhORPoORb1{+p@S4H&^!S7P<g791dtbDvlu@uFpRZ0(VJBS<aX$8(>-xQgrIGvZ0
z_U&;&#y}ON`mur}&l1YH=SkA$&@0n1Nm>~x4Vn!}wwQCj{90ZNMg*t2TwAso+P2Wa
z<z9cbI%hb<oL80%o#PhRL3o^xvV65(i;xU%1<durG1@B>OZVUwS>K+CbNI<h>mfZ-
zoK;Qr5UE(=mD!4$oIYVQY(&{FnoxHZM5V^*kz82-veD1o3z+tp+G?Ne1+>(%wTGhU
zuHe48Nu!+;6uS8y4T(X!RQG5qghkVFLKlDO|20UHC}moeVgB-~JpGGa{!D|jr2aPz
z{oe4%(|=C?$eTb{07tU!44C9i&UgW^JZX_gfiu_zgfmG{zPug8G0JpKCe~jDtB7ew
z>jnlGzX<L*vLbC5N)NPzh`KA<-g7ZhzTV}{p;VSP9<ETn?4@$KGA5MY=gGeE&&7X}
z+QbSpmbKGRCfo8p-_LMQ06d^KP7rN@=m1iJnIn*$;QcUD0MJZ<G?dw<+#Q2b-$j5p
ztY{1M^J%%KLMuWuF#B8}nq^%1WU@rKD0DcPKzVNY_AvphLNJW+s+T$@mgf<^&Youi
zfb>d=V#-KN2qQol76mg%=K|>rq6>du0%N((Oi+HMAjjxJ>$6O(9%vun{hGcJOcE~-
z|08_^@G@*F$WvzbwkRlQLZ@gk`2u_^kYofM1<;9OIvQ7cvOs+Vja!hz8D!u<>b^W%
zL$pluV}Vq8PhcS<Xku90nMrs6346dYG6)qTaJ}OKPKzj)9M6ryf~Xzy`A~llV7%yZ
zd=rlX=G_L<2s!jIg2jOdN=l$!15q#mO%aSZtz2LLhw?y|g}tZ*D#th#SLz;sQwMUh
zKtqJcoWM=Z#W-c>42DVyfb0xNN3fqW2qOZ^p)BE-BKaNc7^5Gifdm-1X^fjOVT724
zOh=ZM(3=K$->W|{t_}0Q08)Q&;2)KWuq$3er338Z2XNg1bcYV*d!{>UaVyIRAcP3a
zRz*61;(HsbY%@VP$kl-bYBUA(tZ}hyrlbL&VPO&&#{;V?hLu-Epe?{COj8LAC@sPB
zijruVlVFG5fmkiri~;eDVDlD2Pg+=Gf*67H5tMNTFJd{mK^~^$WLSTmN|2{4h#2dL
z2!U#ZaI+4mmU&bJ5soZ~1`w-Tc-j#!70U^B;yobpETM#6`)H*V(N5t4^<5(s{lGA9
zTibqw9AKEWV#3A@V?b$ZXd|O-WF;C}u46?_?@0rw8VkCS<)jg0>iZIDTRA$>N=9|z
z-$Le!EZ;;hp#U)f-C}<&2g^)9fx`?eC&rEVkao2`?nJf$q&|Rs?Wd@^t}tySgcU{r
z$Mowpz$zxT61w2Xe!C_8NXtH_yN^^@hCtYFLc!`Y0Nd3UZD%xwWnCP}oet~B`uu&f
z;CfUXYjID3tq49alR2{*;|$L*U{nVlH$@%1kf~pgImpBbVg!Fj;8@_qh1ol3xH9nG
z{(G)M9~aD;*0tRyczR~b)ySfhVVLa1{g(K>6Ux{o{)cU|E4Ck2<c44#62scgyH8u6
zK3jEiB8A{0-|?|7b-M?D{o7Avirt=~X^QWs{N<-ZSYQc!6bnu{hy|8`N3r0PgIHi4
za}*0sIfw<y`J;bWaLPd}P>&wPf>RD+fwJr<7MyYn3)B|}k>HGjI3SB2MS@cfVu1v5
z6bnu{hy~*0Q7kxR#<I10-F*hA`{IvmU|-%DcTq$a9@(?M8N{*A6d9~1N#609f`1P=
z(9M&Nk+@~=UZ+>%@GQKP{LJx`H~L<XHcryw?}hgulumz<X-(KVSO`iv0qpA6rV6E<
zwm-wEaa+$*7YuI|&Nh9Uoe;9o=E`A?xSY4m9ZGgObKG_Cs6Fb-Oz4B?x@>mnvW|7-
zxN;u7d>CA4s$b#Y9B}0zEpwY}9P~B$2Cp6DYO-xz4V!1YwKBr%<-{t|ujK{lEgda+
zq`H2lxxjzxD4d_%*6C2r%j+t~83sRLj{O^2^_<NiZ2H2r@)ASdVZvH2<Hi1nd`j4>
zxSAyw8(BV@c9QUysR>#r8#r9!OP`p8o;K`e@E57I$_k$phdNpPEUN)6z}Bvd*&bBw
z{A#0Bcfh-jYqL~ug%xUioqNVc=+QfaC+yg(RKkCU7QK$!&Q4#{CF#>oVa*DsgTsXI
ziGMmOR%>-4DSxXkzZ@L?gX0;?@wPopX%~c}5}HwPZy!XW<nJT@#Gc3sZ2D8WAXK14
z>cS7`g5Z!#KTqp24CN$4fIb|(l{J?BnL`I(m8a?=Mf7-<g>F9hQO<HO$a^P0$`N2T
z+U$Sum3we=FK+JU)bF<+tt@9-88+4mu#Ao(NWY>@myW`~`c~hTQ!4E^i4R1CGAGr{
zu6wd^kS8CMQf4`~ZLa0d9l3g;{0iE1+(mBJJH=KeU%x}LMOHn&2M)t&D#p_q;PzoS
zt%A_X@zJFiy8Wc`e6aY5%or=Q(MP!{=a7Flxas#<H56x!&-?pW^6Dyt72bpC`s%4!
z<?_RoNj~5zmW(`ho^rj~yVP3H1HA4mMt1VA=@J@1=U{xh1-g7tiP4j{y>@ydw70ax
zH*CUm!0nEM`qd12?D~G|k516V0lJ+N?MdWR+pFRMBQb<rE3}ehRo9NCQ0*BMhkAd&
zRNkL*LY#04Oa9F8Klx;H3s*gshXP;;X$<0yIa@7)T}gQz0J*EN+V*D4&pTx{9i+0h
zNN`=;*>F+ABePfejAl{Jkh|C{%8mO0=hI|JUO(IUdZ>TC^Y!?Dl=Jmip5c59;uYuX
zM{hXaY9Yy%g}1q$5h2-?#XM;KrRgw;b8aqn9dY|QKdqlaDKmwP=k-%?h<gjT$9pJL
z4tLgG>7m$Jf3rI+@Ba^)!_gI&LH-^Tm%)Vv6aqFlmw|K$CYKX+A3A?;aAos*Bb&E}
z=l6#nUy&LsoCd3!_um@e8<PJ8$rp`oEPyZw&HMM}=O-zZ;m<RFNJ>QhSpM7bx4t|g
zI359f7vS<}37*<<>XkoN9;3oI{`;1m<B?24qTy$~@nJkLePDXvFTcG1^dJ+Xz~3<V
z{`&s#uZNmdqm4jG&98soAHoen5a?8+Q1Bv}KmOMI_VDM9LZ&Q*3xDyyZ&M_#5l-1a
z^5<V<^H;w46Awo3nt#Wk-@9pK^UvllJi!uSCuAgyELQ&!Vc96*olj(Y5*odb<hPSp
z$(d2YxR0rV6kgY4BbXd8n4~dU7hqx#DUD293(v$8n25okd9#0s0YMBBVo(r+MDw?1
zAZSd#86;Yz`(ga+<<ZCh7s(K!YkO5g`qqh}$)m`(^TZLRr5{EFni|3=#iR5!T@81i
z^HrekzD)Gv{hlLvYoR)jY5~t~zxE1tVZXE&tUHI0w3Wn4<zceq)gI-kwRDA{GRg=7
zld*;N0>+1JQ<r~7rGsS(kc#l30FsM}&@k<0w<97@RJ_Yq;?ur|N`xpS00cU=d(LDq
zCE(fwLkdq{9>YxWUIMmc04pn#3PhS_NVDa;1!ww|Ku0cLX-5y>Z$@Jyuig0Js$Ic(
zRGJIer*?#T>J=)|;&tm=_r&dEzO?D3rQKNB+EGDqXZU}-)iN3?Tq~!AmF6oxjt$B=
z0t=8J(|NTxl6s!gQ$M~M)p)j8%1{A0My+Mk4yk<k&kyMlHgm#TbfpDEE65stg!Ra@
zDE;#Hf23!uot%cOP^O&aPd~6w>=#H6&Zq=i8mn!+N_lIgilDQ(bKw^ysb}-r?R&W9
zS&g3B{YQVmYR3~LGuR*9>qn~zc~nH&Eot2AaPDq@6FMUumyrEjkLu@LFzc;^qC*-<
z2(3!nQQ{zJ1Ep1DPQt1JVw0(#lRJSHQq@F5Fho(Q@S2w}@K!#1t!`N~6znuR-L_?@
zo9=0yOoQu$Fap0+Ld(ulPH#Qnv+9jZ3me;Uw`_m(GOf{Qod7j7G|AUzYGiu~bj>rj
z5*YsK;ZPaxb_Ut=oMfmvqcGqq$`87+y0}k?n}Nx1@ATZbNp8k5OGH8&DR>jzCV#zE
zza|4)<R}#NIIv}$fwwubt;lvRTJkfz-%b4MkKdc>%D3cQy_}BR*t_%8+Iqq6KSO3&
z^YVZEvmqgQ0aO!SQosqP*GiPaYVtpEk;Rp7Z=Fz;WbM>YO$;yLQ*bUhCd$H6x+Pj%
z`Svy|HWkf$H3ueD&k{>uT_Y`pPLQ*;w7#a*A7GT&RZuOBkex-bV`r;ZacM|Mz&mhh
z9D71n5mvpCS;d_LA+cU6Teid8IW72ekUM|xVk}+~32{*P$S&gvB$FHJR7tU!U{9o3
zN*>ivm*(Aa%B()pw|>s-8GIj}b(vCY6*p@kagY==q~ux2D(z3BwPt%o^=KM|nk`Iw
z<&w|ad2(Lp(IlJzmR4Z{L65@vF~yhZMP>w4+X!9cb`+=>;uA9xI+~o=u0kpXxM6=y
zBwFWl_XCX)0RF@cVY?>KuH6tGDU5Qn8<O+>Wd}R!U9;wD8&ZaDh8!o}Ce81f70<d|
zjWGeQW|TB30pL+aV?DaF0WQTkn5@SwC3cJUVr;{Rg4xrY^6&<Cg0Q_HA`vx}N7Uw#
za!DzRQ4fsZj({Qi-jcV<R2_4|$ryiDGi>xA+H=-7k|BO&3CLaM`~a}icA^n7+_^ul
zS+hg+r<!pN*Fm!_)T0-1y$G%`t_=t;89a(juQ!F~N`6-BVeYjVp5V3cg!K$_o-{|l
z<>)T^3iWt8zS@wPpe?AoJ|gL6*sR|3ay!Q!XZ|G$u&&6E{kpU!BdenDI>>+eul>q5
zN#VBLbmh3I<VL~%kS2%tB}JIiGxnTp`^YPyz<-FgH9q6BaX_3hs9tIF2ClQm9gz@p
zNEn6>XM3om)YVKgCZO9C_7_O1@Dne`Mdv{n=W6PM;>;>vCY`X-)Kp|rlK~K3PTBD6
z<7=*OtLqs`hZfg}z+AoJ0IQcldmmAM=!q!3$IGU?q^9sc^0L`@l-$7UbOe@y{Vlw&
zuH;<cX{MMn9jH-gni_LX$!Le*NGMLMD=y|j*Ho&kd0w@#rzr!3h{~R+45<b{?8G&>
zJ5WZCLJ#V(x<ALz8C}?<cQwT`U~^z=QjY?i5(XCqIMpRtb-oA{2PLS^JYR%=<T+~4
z=btY^wldP#Bd0<(lUT_aYQcZs=YSEiDMoLW+ezyW5UB$Ph~U%%MEfPTv?EG@Is}!C
z6jmw+BmdKm4~1CPQD1v|ZO_A`x_E9nyWDDz&0*V@xVy7KgA+ERhWT^f3XYw%u9feO
zon{Dr)H|rx(48r&O7UHYsHPHslvO7k5IFrGPgottNkY30Dt@osa*IBDFScGZ<x#8t
z7}9Vn>aKh~oRp|uYkr331ag-m3i#Sm+uNa693xBAwQFE&6dcoMH0VQGsMI|=^cCeJ
zsM0%{a$to5U2}1Ezg}r!)o87;K1GA0Ew_FR>*@iu@J{H3r<z_xS@|A+&6#E3Yp}LC
z-uAF7s(tTSmA&UPx?algoOVN4x<+hus9H|4^I$_RnN=Jg-%N0!jOr`G8yi2l15gq<
z<6wi7@qp1G?$S}1l19K)C^+7)pqJFIn00FrGMzf9h}=#q;njpE0KdN@ma2T^948!R
zs0e<1mZ4G@D0ljJg$=xyj(s01e`+O+w7RQEZje;4Y_<8>-@FPX7LqkVz!3L^*N!GP
zfH(kwk+(XdyqY67O#%V#SGNdeoB+?PTPVeh;2qwJ3gMn5cO)6zOkyQxl8v^%PuW88
zgVNe#A@<oN4fuq2qIL^)X|GT?dFY*}%_LST4wGfkeNs^)oQlzX0~^)wf24X*e3&gw
z*hQ=qC6U1kRMokEIXDM0I^XXW&VI+-{AFKm&h}jxJIc*N65{p8x%r2x+gk7Nl8#^f
z0?Cd9IH%#|L?`FJIa?|xIgtM5#HJF@85xrmgta)1X1}S(P!f&F?~Zn$^1@qJb2!t~
zO1;C4Fk0;(+J@k$3{xBjf6U$Em(&66M{JJQ!meYznwi5v$UBK1s=@8a+B@+_LE2+o
zn&yv4F)IBee`PZ@n*xsOz5WziP|kEOV^Geb@>^8(yI$${kz!faze<8%(@Qfnh+q;I
z8Ty#tck}xdHoX{9uXjdc+q`<~W>3Fiy60p|z4<VcH@4|rHM&!tRN(>k-I)ceyVR%t
zhYP^*k+ye8EI8EQrw;vx3!6!-<Bah{c%Qt&E8YqFf7J~&OR2-)Sn*=|uvc^o8)Dc=
z97-OxQdp-OOjXe-sYCt!U!EBnw3k8t9uxsNmvQhG6_#2GCYRZNAD(~k^!WSBFLyFC
z`r%~!IWDb3K{P9uG=rfTkSH0N5ivh=d*F7O5t&>1GWFtbi628eqKk81(u_4mSe!M(
za{z0>J>;2Onu+Dw<RbPqPx;$p(PHEylqcp8PkE%w?=Vp0=chWJ^k{gLNj;dJ$=Wcs
z(ml3MMF)wm_<hKnoYH?QQh93Ui?(KMEwrPZaICnjokKI)7x@!E^7MGG%&awfLfvqe
zo{V`^P;`M=LN)7wC5?P~9sv(U?n<qU$(l))uhfg|SU6B^YsbmJ7i2!%f|jpzWbR`-
z`8?<Yb8ZvOs1N8U&3@sS4tylbcvHSP%wu|;&N5dHN}s1nd((eIX8v|~U9`Zu2Ph*v
z)HR8Xy^<d0RZ%Fj<fCW0Y&&%M<>Lx%e3osz<-tYsb>u1e+h%{}P&e`wxj(MTUAQft
zW0!onhgsg($8>Ur*|Q@}S;s7SYu3yhDWc!|v=*x#Vc#01_kh(xdRPM*p!30^Ikn&p
z9_IBJ<3!_$4V8ami6phPS<|fby0)vGGrcISe8?@wG+SIZQKOH^OI+y{{n%z2A^Ulq
zwQ43uw=zE3dez4Ap;>9RSu4=)GfjMQEx#r1_E9>@uLm83W@cafNp5M^7CQ3KNmJzX
zk6?Lhu_<)~x#|ff7!g8^*3lNHG)uC1U4^H#pW)nE;;Da2T%KE9RRe7dHhtu4_OBwZ
zA8pdjQe}=aoQ8eIwCkj`8y0?#b$vX)EENM*1Y@T!KkojR2#$Q3A||<$LI-cAKd0C%
zKp6~^6c)f)dJmfZrk~XD!+*bh{o9v+fS_Uj<F0Z-p$r5+{Ya123fiwIP=3$PUKk$?
z@gM?WBf@_mQee<v>%iLZWx7M<1Zd-%*6-XJ1bS4e8LUBQ@2F#)plK>RFvZ$h=KJ0@
z7Ov(gqD++yeqt(lEI-otGKwGh)jpT5H0s~e-Ph^vbJnV{%5fJj6X9U#Z%wNp5FCNj
zMzggkkTKBtqLtvtC$?G}d26E*)?3shp%+d9eX4(?tIVaPk-IbztwNzRmG`lILl1+j
z)KZ<)2oqp3%!XnYg{<AA;ra8IA9>t28uh7pPjv1<X@TVB_l9va2}HDFZm@y6bNAA_
zMcp;7<~WmPDmcDfVFMAG^7GWNlw`Y~h5Ni)A(XNhgF6&b*jiBpS({c!(Zfb2P((0F
zpsIhFfUbxzW59q5gqb6llr<`HKU4RFK32xhLrhj7w1_zrR9IqM>|4J{vAe|8+G+}i
zCS#EnUB~^pMb%Y~hC)SV<4ky}TwTaTQZIRE@QZCQ0JhXiUX*yjkMmyNXraa5m4lMq
z8Pn|9{Wvw_WO;0<Rc{{bjuc!eB9&dU-S2<2*+8G18>4eOaw3g#3nR4HqeT^Q)=?{3
z+jZ_bt>ip)&`hRi6aI_7cq%k(i__B_$vQKB^%+6w*$=jJo;+8JW6~UEnf>AOn&frW
z9h%)=@9@D@>!uJ;(ibeSp-b93NE>0s^P|SXXoR)vfZnaTM+<*y+TlA;B-<rc#SDM!
zDjCL{3$`_^t;O0oZl+lKpZK5mU8cxypN6MDDiJx|D~SYz0*~jm1R2x>K?&u;^z&cS
z*SpUnxkUud!0Wz2UNHoFpb#vrr7gJ)!9ehUc*cMNu`WftSOB7zTD3sbpg`Uy%D<oA
zrlIB6nPYHX1Z74Jl<aI#dwVBPFZq8p**%Ltxy)Y%sWJGgjdb>qo;hS#2|3xP0M<Em
z(UT`f@HOQ-o9=QL#3nFb)aN5?W!xLgI3)}@OdCIsb4`u~-kjLiI^A`~q80|j{mfg&
zDyALyT*<|Xv0DBwEAFv|G<BWgEUnbOveHNAtM#HStyLHZZ>a@Q_%QT67hr#H;6eKN
zYMB&L0HGk$^9dhUWg_G)*wVdBhqM5)6$m);EG9z8JbBmhD;KJRJR7G>OcLu1Pji8>
zHC^UzkGPwWv6!@<5%B66M!OFLPwP%%G^fa8@498>%2e4Dql+y3xKaHTb_d0K%RV-h
zqU_5qgc?hll<9fG{-}enrYnDwv{2oYauvtlLuTA*;t=w`$avS%>eq5s?8aGmJ*Y4i
zFG{RmgJxE$1}dkgb;S?95IKt~H7&2D<&1wT6NEEi!Mhudan^lRS;_%P8RO`~YaFI=
zn*F-hY~dvXX4Pr5R+O^hqW5v#X$W)Kplfz$Umu-&N*BWOFzxdkhi8AJz^TG5qFdQ0
z`#BpZzsC!Ji;%g}(olIfu9Y&vDO_-Utn2@x-)bMx)M%KEy{@L8*SmmnRyenqX+Zq7
zw+A8h!eL6hDw}KEuXC!sxt>njk>(~=&<=%Fe$5ILz4ll^6Q<SymK3hmvStrl=y-kS
zK+1MdE_$<XVqTcSx)gtCplYRSX_t5j7T_pGVf7@v``~OHz>;nqj8ya9GzJXGE@s}I
zLPe)pnSizH!H$TL3G7FR<Ae!9zY-zFBCj7A+k%K1h0<yvMh1!cc`Zy;S9n91irL3v
z-YbKv@<9;(9EiMIY}F8iR<K}te{3zA>_9==SAjy!E{JYn>eGMbHmZ#yM7R*({qZX=
zRxUJ;tjbpdcchKZ`h_e*Dlqrv!x6i1mD4J^(Vi4Lr%W3dNIOjtz~8>?ke2`X?D<W|
zrU+musnE3q5D(k)Q6rjqG`b7ruwN0F9w`0=#ScY}rNE3c%jeBCPy^(^>t9{~xrqrW
z#Sb2CX~Kcx4eEa|UenqSBjqimAAr`iAA8@X6jSVxO!UT5Owhsw92wo(XoDBDy)DGL
zeKkfS;V@MbIGnTVU2agQ&|U`;+u#-*DhIQIuovnL&9R){S}{re*I=)rp7k<&e**mc
z(v<K^>0MbvS<P6+S7btfQpFz+c1aU)Hs2izREpk`MKpht{Oxx+{gb}@Wr|lQraxoX
zA9WhOE%0|fCb&SDTqF`!v`3k9gylrBi_m8xJCiV!yt9{GO=2f!4#+v|d9vg{!Hjgz
zo_27RQWJ3_2vpZ~#202weh0ohOD~--(;o9@e!1plaz8U)wrvFJ9HXl0)iPbsmG$H%
z?oC#_mc)Nf$0f4X`qwz_a)s<UMynGm6hST)A)kp6G&_5gw&S!+5jpVqX6*6m5ChJM
z<Vd#)F~FOKe}@nQfs@xag_x^J?BtBV0(--|gcv@kBqyvPM(5C?bI3R}A*VdMA=Gd?
z=2v-6h#e05BF^hYyreoBgZ@oMy_UpIKPIvcZxVlTcKo^n&{3Gd?l%n0t?PR60Gs1!
z8!tJu7ZtF%aqPOg$#%Yrk;Zj5H=h=_n~x28#pd>Te+^>at!?ZsR17!V=PMb_?eXUR
zO)fvC`?JTz6%Op~s>0%~_r|^Y)n8qV_j_!Azo&rYGuy{%h_xuR!ZF0s4$fB%&u$=l
zkad6bt@Gd-lp(h@UA?NkL%sY3!9~E0A;O-4a=95-Z~v(Eyhk6}LYJ;(FO`#vjoJkO
z3e<Lew(B#zB12DgE5pl~=FnRY62Ia=p&Z!^ls66(N?v+a7fH)E%-BI{fzG}5I4M<~
z2O{ry29|2=Bh&jUp5<=PK%xGS;WXrve!hPvR6jq>L4+kWoy+&P%DNucXY?sYA9&p;
zDQc;Nru=WT#c{dllF@<^2K+7E-C6-w4TQJ6J>L@QJ1?9)7b&^$uMhJ9K&(;BVWxcZ
zF?}<CtMaKV2cqRz_0W9NjD_dwx2NWZKSSl>6_xQ`ST$Cg-*6PbH$lv(O$_~|<l#5w
z%jT&-J}vBt-7w~Oz5^L9RD||SG!iDCn&YOKe|1`Kp~6+Q$Bx|EmbALF5AOsV{tx6K
z6$qC>{vH&Q(L)mgF*uh|?<*&l=!+i>2udRLV_*<ql9&C9A1i<HefQxExDP5i6YT!|
z%ihofO`k^7pm+OVjPhXi@89-+KWlA0T}Qf5YXG_e{f_kQ4o_er%`o)Tn8Q;*^PK0|
zGhHt<$MQV-eUf!FlSpXrw07_V&o@NhM5c21=l!pp7D2TdJpbgc{_gF4_kGuk?a?UX
z&HnSZ9X~)Y3Zj4AL!_PJ{>Q)jFT0OD)yhlL66N@Ve*O5m??yg};s#BM)80GRD9-4_
z!}~!2Kgx<j+LSZ&9h!}OI7f4l3I>G`3Kd8CLQ<zARXS23%6GbF4y0bqYNbn`4^LjR
zQsGqYEYlch9SWVul4Xe$^8U(4+IBNS+b+@(H)V#<OGkfZpYCm(jS8LVXS1Aml6{Em
zELXZeiJVdPsPi6C=CbVYcDvlcs}SaLzb?xijnXia+p*jvk7li&M21C2l5cX*-aWxa
z<y~O+{JvAc4C#6Ok-Ro9{=W0<eC)&WO%Ff(`Hkdy{f~WE=6#%r|5TVgO}Hh;b@5Wf
zgSIMQ-7bGeZ!*g(T_=iE<S0ikPDd{7yf~p4Fd2`>Lz&IyOD1OX9g8WG>aO9BjYwfM
z;tw94(8BZR63!j0Ndd-Bl7ko7AahId_|dUg_UdRcNkS{}Sw=xFp_^nyS#_*Dw2ZzT
zz?Ja!RkoNpW|^Q+87iMnr+Lmco#tm5VKoPLI2eCFBo>vBYgjKT3ZC0-CPnyBS$xl!
zftYB&OxIzUaw|s7E6WzEsV8boX!Fvjv8;{a3<(X_ogr97nW9Fb7n*L4f{YpkDUxJ*
zRAPfNqVP1|AI$Rzj_Gm}>%V2tlAvGnT<fgz=zC$aQ6}*Hk`)YtEebwP{F77!Aw`;~
zHQ#?#anBNsdDO|qO&uxpO4^AdpgPygihLKuF_Njw#24e}cjMYL8!MGU|D}<6vML!p
zulDRRhK*6+9mU~JqN8(lhne|30XjC;w__gbEn2eC&80)2%E9%OgP~*q_HyNr!1Z!I
zw?ItnSQRutu$Jp?&U3~E0(|JXS-qI&jQ0$*V>^>goGKaSEb`cwE{`7;e_89vSYvPv
zqpPCtePLV0g@ZCUqx@Km$@t>0V@&#8ScC~VtZJ-+TQVC#kTp&XIy8b$R_2^^zbm!W
zuenk?``V7NWWA{y*`+-s1@iD!B~{Pc7Gsa`&SDCxcn2Kn_D?DKZ}UP;?Uw5)Jj$4X
zMvU*Nb2+<a&<M(ep<j{OWj4#L-0d<9<QfP)#ns*+cN*R7)P|PUk}oIsJT49eMr4#0
zHH~>`Ju!bnOzmfK6Dysm8M|8UTBW=FY1cqhaiHCU8b-6au`Vv4(!rHhgvi3d3D0Ka
z(*<V=Q4KV?WvYGY1fiEGkRLREjRPtIV^4V%K23tnZuX@dRB`Ycyw<KiL+G82zjI_2
z>;+s4UN@-VhJN<rq5F=b0N`}5XC_}c3IH~v^E1ON9R)nVnaK@|feYzl%TBMQn2w9a
zWtSA(>Lwp=j|8c&I!Xp_-dJCn^c2q*OOsx?=LuT8j2&rl(l-7lX1*bRXslDg;zl_w
zpt3NN^HKG)nA1^Ob@;pUP|NjNB1&)-^7%|jMD^Fk^QSRo@#%}Cr!eV_<)~zm+WZQ)
zg+%bk1O^lygZtS8W+OR|pB@*d9ZpmBS)8eaZ~zk0$VZaB8Tqu3o?0#p$#!VZ3$o43
zvhIo<?9?jDf!OIeNtQx?%uQu2BNy~sb5SSzHvL+UN|Q4B)yI1PWD>~Rbq{Rexd3iC
zT>+)3h_l(tjiWI^*-)D`xTCKCuOOP9(pr@x7}T+S8(#@bDuCC=z+OS1mM0GNXmnGQ
zN$b*4`DIeq#L-E8+I*;CAu5bvW|)=NE?F0MwU*`8O2=ufGL%<;E8XmZSEaslaXijh
zji>WE(mccp5Kws=m#na{sfQS-2{3aCw^yy~fioR17ZS*?-SliW`zGpTAkY-XCN}NQ
z<N)oaVlRhyOR&QODnmoYIa7_&j4bt09Yt6_r`%W~EV^e2yq24!%x;rp5||wqa+gxG
z!>{RJtd1&Z-|}C7dzEP{A62)aO7p_NJBKfEk)MJuXShUcnBIuE)>8uWv8D0r@OFrb
zB<v*Kj<&#u)5f;z`N-*xrU5J{7dq{k^J-EBPQlP=>Nhku8If8k_1!I9!K{`ZuHai#
z-b3db=S5_TOx9efg90c-`-me>>9AXZE2&UWq@CNG_~`MM*peR*Cz1a-V7ke+lh<y1
z&Aog~luxhfn~+)lw_U4z>5mzcKWO|9!6tS!mqGp>6aqOmmm!E2YPXM*A0Skh!4@YI
z0x&k0;awUgf3;W5ZX-7gzRy$W1+0h?^#=q7GEQ7{Z`=B|*i+lxOM@bN>Fbx0D2+xQ
zJ9dgaj3rAP@*_ol;wbT5DDlfQpQeYSLE&Yv;I%k?6F?h^enHWqv%&*NXF!}@#g|8=
zw4-OH2bD609#8W?quuWe1d;&S1=#OA!ef!sM|z$}T}CH4njiQ*k|Y(%IC?j_nB;-<
z1L=dq{_E+}q!_4@(WPk`#(yuT>1`TeTW=*QFaCI)(gqrkCTQA5=~NIu{u1A&&!gIA
zOwl{irXQO9_%4Q(-#BtnQRI|!)~=Vu=pQV9SDGJ=0XDH#E$Ag0->|v`E7GYUGH%a)
z&<UbtKYWo@bGCRE1>+?8(C{5u+cQ0fqMZ~eo2;;=B%_Yo+O(AxV%3(jN2iP`-qMka
z<tN8)wbI8A73H-ZHNyHt7HO0rj>39yCKX3=BppBsVG9?QkmYq><nmg2|F}<ik5w#x
zZM~#u(sQ^R2S90Ee_ZJ(@{xGPz$uoGkAs>qs0B+s*8D%?E?6|<Xb?B>4Y+C^DMR(8
zD=>A~l(JDvho-?#rcDQ>LVV`T!V-_$awofR$#5Vt3)maZj<F$h9f-(ln&%vh9OtkF
zsKstGpKFG05f0D@-MGzYbo4&Bk<qk&kwnpz(5i-oQ9?EsXuhEmi91wGD!+VH;!oQA
zM8YNp@oyG=ZDy4CTl}6F^Bu*3I3_8!5;R;>tb~@4N~)+`R9F+G!g#-m8<825;PiF1
zlr7PYdNbo+Vo{LV8d6Ry+6X^iW0ps%HlQmI3}0~vo~Ormk7c<$a>^|JtP-PtMe4&Y
zy%#Ew;LZO!#CuiTn74+!@%WLv9Sh`!4hSQSvE5*)bHSNO!j-NE%&hCZTUY%flSz^d
zwL7xRtC{1L)h!0hgwrHew7$>U;dzTw-tUc&8WW|?m;{`R#$9iXt`<SzSftKNhlSh?
z^QK<fiSdp~Tx#ZgsqCCeG-?ii^`Cy+;@`Wkd$J}W`VrlQ>t0ivh!#fa+b<#V;N^ad
zmbwdd7QuSVBSssgr;;AyJ(v43HxJh?ojOvAj0pnP5s+NAZxdZFG4HoY&YSY`+vHM%
znzD#T+I;x+)oAhjPgi!I&*1N>v0Z2FhH3=uq=QDX#&zAaS-R6)XSL0LdE(UJq$;(E
zk(1E&H8xf@Cor-Fu1~BDS51aR$9X?ecb-^two+HqcTs91)3=m<srbT0?#Q~2UFW^D
zZvA_Qkzp2}JYy-o*Ixh3DxT?_&-Ag6wZ&bxPJKk$x5T_e=cDXJZ0xo9N#sS}hpmuS
zapbOe-SFN0&0$Txq&h1iT5W%CIFOM(Uwb_L>G#d)fb%wzD$VNgMvmyX;y?MNLAaNp
zv>g)zGcYojA&3@gw~vz_AXJx9O$HK|zd#_te*(PJ`L`dxJn<r^fKNd_Uw%CO`80}Z
zjb)1K?cYD2;sdP&Q(kUTG1$cR*MGO)pZ-xPAZ5;c{08seS0tTfUb;x~*WdW|A9(l+
zOjf~ff2XNGD;vK3Z~GT8xJB4Ao(L;YYq&<3Z;}Nc3fV@&sKB@mkCV7lGAo(&*QpB5
ze}WmwCTH>joQYd&Oy^ACMA9grZCC)}#hHYPt^zMzUaEY0@;iP9rxD;V;!-~M^4*u$
zo}Gp4t8<X5vA1sRhQ>@_;Jxp%<)b9-bYCE=gM5qoj&X8Fs<0A*#Z~LZFgGHm1qZ(j
zKYxh<ZXgL52y)}<lQ;ra)sPM}Sku&0f1+WY)_Hz<V(XJQ3ZA1KjwG5uNkg(R0V(Bi
z{#w^P7k-yAgUoERnlyFYYvK6~UVAq$?vm5FNm}IN%i6kIZ92N4?J!a6_Fv`DsL2EA
zcDlOEN7+Ir8!xk5lTo{ruww*$eKmdqmywd&m4!`k;7_{7?{kS?*d2WEQ#=#Sf6}6X
ziYQ3|ubFZB+{=b}VS3%Ij5Ewyg-jLMzBI{!R918{w`wkyXZYe`I@I|{Vb3mA=D0$|
z6$&QL$UDHbi#4)7yYwp0bF{AC458{fiR9q4q0KUvgI%b*wT%2rK;~ca8*$Ga<#cE}
z+pTgPSCrBf&U+5VBfTJ}bk?gJe`u(i7RJpsd1ZR;&pAar7tXJgWAdKKTPz&++@ZH%
zS{OpKXtE@rDWEzJ5!tdW`Jjv?`1*j>CAlQaoTWJ7Q`hlLbS@<6wM2PoUXp{jB+ev9
zopsgvRGKt;>3zC}Kkt$%+A$2Nbf?_?WKo)2FrlCqM*Y*|;u++S7|e?}e^-{t(Bhmo
z|HNb#lT4M}6fU}Oq*ChXkjNvB#Dyc#Xj<$inr#K~pPFZSZg=G{0iVImIc()62fpKq
zZ^(9}cF{n+b0?8%XRr4U<`3#0qYt$F0rvSepmv0RmIY7@oOVpK*h~;ltXC+>+`Qtp
zPoUP8esNVc7jcoSvYFemfAuTmBfgdGTz#uV__2SD{-iY^SuQ82r6h$~Z`M=+L&{)O
z50q*yj^3%%shtJs+<hrtm!3@CoMl9<h&pfYe(6ft_LB8{SEpFvpc>!g`4vQ>baKoQ
zjo<U3-1Mwy^EEkP@+bl=-ZC!$N1-E;SRQNZJHx5R($mr@6lYe~e_*+tx{8?_C>jl@
z!^(guF?I$toEk8#y6HA)*3x{5+I91h)TZlES`W4=Iqnp(F{VU1CQ2<*q|TwUFdfdY
zEp<|vqD@UojB0?FMHWkNe36bE?6zpiqhys|rw3ROjV1Wn0fbLgByoVH${IGXy^?Ee
z2zw=ubY^U5<hh(Xe<faV!wrqp^U4nB%8zAQlEdT74qMsvX_OqNmUxMOL05wJs8t6v
z>$y8cs~P;9pAU00EmVv7hEgtfwbob_xH9hz?XA_qQI44pFmV_&7Qm$wh9PljL!{8;
zSw~9^*{qE;^iW5!^2om)sXX$E9n0Zv7bL0VQPkfoX|223e=km|J=5>jRNZJqi5PyK
z)P@H-K)Yig7TLNzs1;Ig6q=8%?5@0LL53Qqp2h|El-CPO<k6Z`@~>o7p`OKvB~cGK
zs&uN&yR)wHSmIE>Jh=~f%F~cc^UPP$9@&Y7&=+N^8_SAGLbPAb6Tr1Ji`tr6jpg~1
zE)CLEEtyL6e`%79Vgk46iM-Jy(p-#4pqDkbx0aT75pC)&Lt`6eC;5kSYU{*r%xfv%
z?D%v#E=`Nc9Eag_&p}wSFxRzqt{jN}ml4dIJ4#$DU-W~jsYYk2={<=eKKN0iN)Pv}
z9Yb4(mHJ?5v9NSK2Q&359j6*<8N8f@^t?iS)(T!9e_Pxd#zl>FLo0EFS^2bSIRj6@
zy)_ZZsBPLXejQP?;=WRqDixRr;R7n}dum`M;k4tVj|Z9!1=T36PknM$`l$GLpx{Hg
z6FM{#_AIzhBPO0)URxqh?O-QLiTl3jV&^^3qu9`OOZ1!9EG6zzPLijCD12;r7@CAd
z(<`J=e?+iox^!<dnNFM7(Vt8gEpW{-;*Mo(4T-CcVYmnCYNVd@pZ6;}JvM7B11(b1
z_4Z-3l=SvQU~n60I=!X|VFHuzXy6hpsC>EN-1Bv(j@O(bkH?Tozwx@$<m=LOE$Y(M
z=QRyUEdU_;k!wwn*qoyfUd(A!rbEyK0^6{$e{C9+&-$o<simU^(!8N}A)bKd=K-`O
zTFVwutbIi)-NRW?D;Fup=6sgQwJOl~afj<9nA*%0sI?suA6WU34B+#v4cRw`UC{*A
zZTDKBlu0)%KpK6rHPcwI>530Nz7M~a*Wl7PCXJk~!D9!~;5GO_^`0wS2vVbgjAq)m
ze<s76!m`jBmesev{@8|*fA;+}A?-+6_T4wmcwB1Y8!r^Q5lT^WW|f&J`VyF7-PbcJ
z-HZ;a)JxeBWGc=?c}x7e5y~GdV2v6nMXz{&!iaR1@=N6tx6wlUarx2KRx#7+6ke|3
z2mWhUx`hU0^hdBdYI#2kg|OUoC3t`Pe@Q`H`u<YqR@go^98whS<K@l@I}kXg4_ou0
zwcYE5R2Rj5S<1dWq6O58!kf>EXl|I6;xi(;rz{Wp=@_m3Y;_1O9+$_#h?wL#>hDAR
zSSQBW^{6=jJRHXP{<v(#3L-$%+IIioJ|M<~@ywa9yg6^cRg<0_1xyWymKsl1e-N5(
zXfq(D{ge-Z>{4rbSFP>QU3K08!i4<Gbdu5=<FTux9{_*K#2cSOeXhJA{w+ePYaVkI
z_{CPfQWHnoL8VRMQD+A0$q|+~oLf73xs1?P_iwmlUQE&qU9LlzA8**eZ->><Cz6j~
zw^O}6L0Zj}oc{RdC9~n(=}{=lf0WUmkRAuh6X3BfpUk2Ho^E6Tct(n9`j671km~De
z;7jdT%LJsWQ9J17m+ns*Enyw#x&S624M4Ek5KsDjYs$|_e4r@#sN;Z}mZs>Kj$G4#
zS4_AtaoB69-L>;zSkTkX)k_6#la2%&6v}pW?w&y-hEu^s59=%lW}(^~e>e;}t8kNp
zD+V{t{=j2f%D{q~<PqY-Y$%mxQpsr#C_AYz>;LM?Z@JL*4O>eeSWGBAGlCl?Q**_%
zbDx2TyaO5Z$U(d#1)=lw6z@0#KO!AfSLwMAC)7;?>2gvHR1np@Lq20*nw}|7De@jG
z9?EY$gV$tL{V<10(MhNPf8|7R{XZY+EZ`*fiGHiNWyZK0ep_e9BV0_<KZNApaYU9Q
zD1Zt@>)l;Qo~{?D_qivB#9=0aJCXeL{KeM0nJdS_D5Dpk*6<lQyvW=$IIx%GfDpMI
zCimiJ^BiyQLrjN(GzgL_pZkvWoDHGyY$vrM<x+Gpo%gaf;78h^AMAz-b7X-M@15LN
zu7BcZ4i}>L70@0(0mlCy&48Wum!Y&B69O|am$6MM6PGT!Kq-G3$&MSl@vg7v9~hd&
zoxnhdB+nSg&B5G)90DVD4t9|Da>(zeihEagYn=&@44{!zB8#=L)|S(cNlssO$G6?5
zS8XOQf(5Upx9=0+4-)@_M4+>i2M|uf^!97|^Rtx7;p>PmBc;~(^7uL6yWKx&2)F|9
zSAhMKH}IU-saJn|9dM1Jb@=&I_TfrIA+^K3(M4Jhh#w-p%>FNLf85DJs1(uBM|$Jm
zUfy=U?Ks&EqEJRp-+t}r2ctpgpe84UM?Qc5+w^_+m1AvZN!%hV{libc|Cku$PepK1
zqTr-+*3Jayjc^_&=c3T`Qx=>-Az8q8KV$6EYk;#@Y!H7an~KGvc+6|YmN1h1jv1pc
zL|^@v$euxv?1~)~As~8=9So@sUwTGk+Xadw#Wu-PIsUTLHbrK=a8^&wcww}^AhK&t
zYM6~g>qT?WP8c69V!kZGO@*PsX@@*0D=fsNNQ-SB%4fPIhgC5*Y0NPt)sx!s-Kt7I
z!S_#2(%pZ#!w?`^5zUfnH&9NW?YcoI5XMEm+q@7(iawcQuCJM9qMcwC={Bd%2l8;U
z344cE*)J&ne6vOy#$W}$HZ^#x-f2}98l(uI_-?wN#Hm8RCtq*oQfnn-FbkUOU{&^R
zj<tyb<bq}y);T1*goU!JIm+yfh}op_g_^6GHmHBh2HhOUnlru>o@|(TSqdxWSnyL=
zQJ!dwg$BEKn#3o&Nv4;sE7$yM#uMgvh~nZ3mbAqvqBNYg64JIfQpk~7?pnY0jou>J
zLSHCXj`kcI_opKBkSKqG=k-fSLEBh8NYRFTJ{s;52K$1`#5v}6`mvnes54;?j;K8D
zH9UV0e1uNXz)RuOn!KVq=}PC!nVdLPosNX(SojKMPp7F+X=9?V?(qfx%U;_SAd<co
z7UwZYHSJV;Lrf+mJhr+xIfFvsnxMApqWr3|R0j+jjy|tQ_}jNmL&B-Dr0G{#*Dj!S
zPyeOru3~O`WBFyc*NKC2OH!&A$!gRo+g5*e=!1p}UecHGFzBn(UvDCXjxLkLdD3=;
z;vJevsaP|UR6l8E3x{$ttvW2Emu|$m#+?<$*fpJNV;47UGsKnEi4LZ#f(P%5zvjMV
zJg)oEYUkCC6OW<af$G;#l3*w@liI8#H2Sc&zL8;`hbRs0s5mf}T|mCrr@F^-$D4l{
zvUQECc#oYe^D<XRIjf8JoRc*wV6-+|In8xhZqzwq+Kt~6S7bWJ7Lc|awFQSUSuf0@
zT;f$O7A9r2Fd=f2T3$!0s{GPS3-oFSTUN91km|K-jJIZ0w^m`89Eu<pRhS3iv7|g(
zg{2B>E7itS%cANvzWe5j&F8`2r+9yHdpeBi7AVItD_Bp|OL4*JI5j_d^ZAuCEG8oK
z%)#HC;sIDp@!|rbT<541F>JL7-QU-&El1|UA=Ut;22(cEGlQHh{WKPnfuE@3V9b7&
zbzQEUgNS%2!)RX~r!{B+ej!VB>Um0PU0he;W3DHR1&+<hJF590ymSj4e|~@5Y7ld@
zy+2?d#%Q7!F@|@vY!gtpI<hSLA&j0Je6NMib)pIsy&v~|8tu@>NWSm42^W5u(rQG2
z_HL(tPXxwIxY#+5@7mz?BMNt){`uqkAHMw+1a9~j?LgOwwh1V=U-_%UGn7Gt`kB9b
ziI%D21Xy@zhhYoLMo$J;_cedOw`pgT7ocqzpg##3F(etKhRw-eds<lVcSxgv5H@H%
z?&k=sMD-IAM~w>lMp(*T`ilGOF8WHZ@q1m0yZ%1yzE8WaO`1{TA6;s?5i7z3VSs|q
zAkYTF>!YblRHPBTBq6Lr2SyuE2^XBvDL3(GiF#_N8?zNMFkC^1t73nsp!&|?8v<aw
zW|0&>D{Mq(gM^`nK%0aj)-2lo`P;9u?+5PsGQ4LzPoOY!L6`7kv=<f(;YzXEhSgfQ
zEU&Ape?OEu%jBdZrGPyZ%qdeodF_fZq)5WoX&^E4`qY{IdbNN6gtg`h0rAk;1%t5*
z3wYfGv(dn)JOJtvLvDW)%uvjw!rqjUA&0<~tXZ|9qGoc(w(o42S4!x-zh--(L7=xm
za`|jqu~zZd0x{*jr`b9?<eH(aXN6j$cvdJ?x>}22CWl&O^d?86>oC%Gsl**P@IwY%
z+E{jj<8+1{v*vkzU$rWg&^?AU9A5DPGDBF?pg?24Aguw?N^gH!rJsk2;A-4@r3shn
z!QfjuvyHD(k@OX3l(h1-pO{*Ljb3Qw?~MAH1q~Kg&TF)Ao5~n2+El|-tI>=t3?+1T
zJDW|tqS<=8>wUFDv#=zAt%@q&)!6-hIa^yoFc{_{KP9#1RBFxD=1{2h8>=*>K*lsh
zQEf41r1Vy*ocMpN{xT+#$fLqI#(CrohEr~#p1Wa8s*A}%u)-}*nzg>T4#AIAHs`%S
z)UpVyb*}GPu!}-wy<uH;_O@?5S2jp*;p7^-`Vl{Go8|t)OcpBvZ3rD}0j&q89MJwT
z&6kzT^Jz#u&xPiis9GgLXue8S8MwikPv!P38c|_o1kit0@tbwNm9I4xL{arl&K_KR
zE{-!RKlHv|SL)ToW(8d)wkny2)C~hG;VSQ0P<UlKOGPK5@mvTuDC3-nwZ+8I0LPx3
zvuTbg%ZI*~4&}XVA1(2AOW<uEKo+%oxb36y8U*h@pzXsjv$oGe6YkO~;<TRzLoE57
zz4ieC<N|+|G~Z@0_YK{Yx0qQsudS-&+wg=ugL+<RP2Utb%EdBa%o>bd%T0|Qv#j>5
zHLhMObTvLCTKkxwVUR-{KQ*ZCbhoHpu6-L}xh8=bO^ruiVtNdV4!l});B|{$Bi4Dh
zT~{fd76YOBtDSM-hNx#(T;P_p*1FkPzX)%e4K9Bjr@!72f>C%of#Xb8qw5G+!}*AU
zdxKoiHFN!gZinEI`0Qh>A+F~!TI#lwWv*oqISjP1Y|9p!<mW-`y1bL_ca)m*a{WEe
zbC1#TvISS`-RC*kS%Dhq(x`(g<<9s3+6je`fX4^QbDV!TKqvzO&FBLH1jGCR0fN*1
zcz}Ok80n$C)IvDSuX2?gJvXQ9N62oD`NB+vt)7|e=dnn*ou2Y!9j9kwB2s$(Y&DbO
zJ&#;5jIb_``lQ~~@~9(B#Q3tRVZ<i1{D}04VZNYYwtCUxD&xp|rf5aOoSyR<=BsZ_
zq;8nkq=s?eCvI5)0FFR$zvk>ptA*Y+9eQb@-SJd^s8_<O#Yt>N!ymxot6yr6P8zJ^
z{y026a=1&Y8f#?h?FFq$;i>x%<zK=$gRDGU$1sw^rE~-2`Fpm0a-!UhCs)WB?@tAL
z2zo)ub)H_61?z<-$Pc`u-!$N926Uy>8hSKTEoiD0XH5p%vwzoX&0vD)Vq9u%oeqOV
zr(H^a_H_4b9)c5Clf2^ulVdzTn`f*43$uB4;ycXd8OC><%`?mo&gL1m2WInJ$B)nE
z0gG@Q;ycdf!6DarbMAcyLVC){!+Ty8y!;kgG7I6p*}RLG62r%g=8YB+bc8yZM+c-F
zrEYBu9t1e*=hlr4Fq_Rc=xjFMH}{!)JzxNTRT}<(hV#k^iKUET6Zci<|LAZY6Fur$
zyvJ}JBwB!RR}Sav>USN^4<xM3GCH$24(IJd!}&lV#a%g@FN>>Kpv1bNT~t~{YIX3n
zY<$(3facF-x0pRVs+l0MU>q8b2`<-mvBulkc>~3xXc&u^)86xA2J|XMbEK_FnVVrI
zW_1!H`OwYEdHq$N&Nt6(V0~ej)>XYZOS62I)*?xH{|^hBd6bu-v>g)yGcuQPjTjS`
zX)YxwmpWG<m;soV(^nwof1!8f$jxWZG>^Ee>zhzToQyAs&tlONIeq`E$GK23bGxFt
zK?)S8Y$Pf^z=}j_+hb_5_K8I5Dk2eUgnS#AXmEwdPhw2`ybMA$D?*#M#h@x?hBG0n
z%F3bywiCH43YhOjFR}#{=~_eregJj&H5*<QY_z@c<2Ul_^Xuimf3w63&ZM)afB*Rs
zUuY$m0wIv_Hki|2|2h5s@(&8N_d@BQ+*)IXg921<^vMDoBdSd4k9?p=^n|=dq%|z7
zfByR8L`Hsfo<vC?ou!baSoS-bgBLW@?7Inm96#@vR@2GF3Vs2+wD)xOCW8;)Qn(;2
zGe!}6P;gmh?p}NWe`Eqd36-KcE1nWM+2i;vIloI&x9kYf$KT3~qH;+l;FGsE=tT%5
zSKo6qDDTKkj@;t~k7!`(uAzE_Yz)e3Q4CLAb#0~&N6@h4QN*?4(PEZ<q*0^aDViF@
z{?2%&P;`JeV3>9+#x<0aq|qC~Mjum5cB6jv%{5j^`z-lJe`7sJc9o*Kn?~3-{B9aA
zQR*j1VEzoRv8gKZQLI7@oWRVcXtZ+yo95LB<{bs!F@DPp3Tqk6qpJmPWK54LNF@TO
z1xD)P2`l<>G{j%>+{83NMMD&KDTnM?QAMlzm#iS#iF5a6-;CkjtD#u$=QIY*Sxw#>
zoHWCshoy|ce-XKC;@4QFZAgxWP`#>*+T<BHVP>>h<hXCDJq4K+_Y|aN%4`QoRdkh4
zJKh(f3rNDWb97ZQOS(6PO=aR|l7{;z0C^YKZG99k8=aJ-t9IzN8G3TblvNbs;mAha
zeRAm>3t^&d3|oy&k{GfXLEV1^^m?sA4OLd38|9OffAc;<tTV1zcJi-Y?(zMtKWUwZ
znvZjGnj##!Gc)amMm}L$-JjK?MwIsw<K57lTymbyEbC!`bm<KCZP`Fe2pFnwuSP?m
z|5A#QK<yGy>HbKnl0(#A80%yRb;Wu@UGUh+vhu9z>0OU)^<Ot6R?YKzjN2*Iu96@N
zl<(Ete@;olBJ}%cSR(;KLv?ODG}AuJ!^kI#-dif$ET_Rn=oB{h*0__=63iaWHI%bV
zKs-uv4H2acLi%Tu4ILnX{fS&-1WZy5&UjpD;vbFxhml_~m=1gqb4*VN=KE&2Md>Rf
zah?S^0e<iNP6R5cg5^|0A5S%;G|WlL1Mv|be?l%B3WWf{1auhLAeU5{NpM?)AJ?iP
z_mD<v3x0z#6|<`rNjhdBW71v#gcg?A&QbfUXZ-zv98odh0)-CJker}UE4iX!K~Ff<
zyzmJzET?sxiAIwo66mB2I+%Oq53^tb@dxwnDjVaDD|xqspaFu%>BL)k_cy4y2~6mg
zJVDiv`aYT#Gp9WhUf-|OZ?Bd|W1m!eDIZUn;q#}&S_Z}Md21$RV8Kn0N#*pg#6Y$f
zP2>oa6i3?Rz_;#PAZP^yjZcs)m!Y&B69O|bmm!)L6PKqRK`581Vj!J=<@@QcAOHNI
zkkNM^A7;ja^=ryu`Sk7K_XPmoFBSShtu`mW{ClaxtG4jBs5J%`0mW*4NMZQ{0Dt-C
zm(PFt^qXE`+P?~@THI>9&T09Ej~4V|d|bZbb67!y3_PqExOR(6t5yyGXd(}MS{|H<
zK;0)X`n8Pa)g@<)m~8cbYEvwLL_iD?4+Uc%MS5T5AalheD$~qHc`8_NZh3_HG^;$8
zk9;OjnDzJN;q&tF0k!Hw2oXS1GS<U)h{?G<t?Je32<9U+OG75CU>@3$R&rF{AB;+Q
zJLn2a=`2W?)~Gc6VrncijR?7E2yhVw8e-O_@dG_fvcd=(sb4*RETeU61igTXYSeDi
zaR2@38_xRzv)+sEhUZZmyE-41#}Glw6;*XwpLU%Vp%EB;w)mO@pYyi*6w%o8t#LjO
z`>+UJnoy5jst-470-}ibw`hX6+R=cE@PjpQ5VlBy224>aKUojo9=yxfNEOSI;t^9~
zk4yBXYD>d+W(b~t;UJ=U(x_>0>dZJ9SA%qs-;_DeM5RPZ^A<)6+__<Mj^4P~aJ$eT
zD&5p*RXiIWTC`jN*d~s0u!vci;+6+*e(id;;4;Z=6h}ltBj`P{>XEUCqOzC<g1IJD
z7X@1ZNoqX}!JSwDUt6_~6AM}&W`c3853IFYUbEJJEF&0yrjtjJPsqa(V2Hu)rNy8U
z<Qc;*mYR-)gy1G+DU1AdxjY)g89!p-t9BL7kcQGmPHjG!o4|gmW@A<P{ZbWm7#omt
zpCEpMoW%Kws8usUytKY`t<TLMx3X1(b<&h;I=9RnI+!CXIU|RcOAa5Q<T!F&lPcvi
zX|*|~J2icOwYSyg&=D(r-*sMj3e9Crb~ngsQl3@?scfY~*fyJ<X5Ao6KnWPGa>>|>
zZu@wZJ|Y+A9t<kog6SPe<lhk9=9S1j1&V?D3il#8A`2s1CfS<iNe|T)Uyiio4L_ZF
z#AuOXrq%FcTAHlDXUIXE#903|^Z_2NH*41-PYs@bfgpx;)8=U1>Kk6S26x4CtG=y$
zGo%T^l%8}bIEZ76gfzd+H6+0c(>!&0)r#$;<%Zp%A0@r#N7g~v>&&b{!Q~|{wBD`O
z*m-uza)tYf!PH`cc-~(bhZ^S9FgrHJImLB3{V=0OII^GTi*madA-!!Z$0C1~PU@T$
z`Wone8QW#p!d%pq_h!v0Z*-;iE!ycd>M>^wMFz@F<4)w)bMv@nHye*`B}eF<Uq1%d
z!>HXT$8Q|{+C-{4y6Z2EBKb>$yTxCen&Xq<k=W(s%3+ShjS@wfHg|ZcR%M98A<8xA
zB+5C+5k}^ghGBe~a|y0c)pm_3Sr&|8Di~#dIauv|-!foN6Jd`>u};kLp%;mst#%&^
zlsd_<6&hR}XM!wt8y7grHWNCvvQExh^1akxwS6{y)ez_8`O|3Pk*O&590dx5pCafH
zmdkk%V}H$sJmaNB62vHcqK%C9j*@p;Y0%W5%^E8Tt`lPy{)jMc+*+Wd2NBS-g^U(|
zztP6t%DbH5JG1><tbd?7s}+@#8wrP-7-uH`v^MUnw6$Ryn-R9I!Lz_}8g0uujvb7j
zRtl+EVXL9<IGPNSr{d{|VJE&dE^C!)<3S`7^ff=ZK|+ilN1E8-UsxLAku`ni_i9zQ
z&)+p`NO8^DvNcexb)I%&=z(p^Nj5`&@z9BgAY#r=BhKl(8W(?NBhJItOhrhv*ogCK
z)jk-@&8h&SzGPyW$OA(ojz@<^T%XLXG$tU1h=>2XMw|t6i5d3(Lw9X|^nbJw=fWCt
z{vnMx7yMe(a$6^kpuf8nC&=w}H@aYV_Tv1jdU4Kz$tiPVGtN09%NujX;t02Y5@qu`
zG#5giN&-;Zo8G3Vfy@#dr-7(M!3Uw!I_l|#v0z4=VP_*sSSbB#rkzFU!kc1{1t<d~
zo^*H$+P6v&x!`0<Q*eJ(PcJ+0A2aKpOAv)0N;{a!YqFappg!b&;|^@pBsbG3r$9ay
zJXWlVA6F&tiu4|=T-PDNkv*Y*^^rWH_@*Ss>!~@Rw=^e9>pA7g(ta*E8lH>WdhYZV
zT$w4(cBkXU?r!OF3YR-<bNcITPLZuIg)aoR*{)D?49!AW@J(&Nu>~sh^5dI?Hadul
z{ag}waAS)>iA|5@x1zB!l*o<>SAwkTnyA^)WpUi}@YuR>y?InN8`_G0sD#ke9!>-H
z!_6{oW<+b)?KMMKCZ>9z{{}bf-yO8W@cdRqY`;3ODRx@l*xqujZ9D40Oxt>e-)|+K
z*xuT#cqq*lPNloWeDk<MCHKZ9j;e7Z-O}*JcYW8r!`66#uI;bqX=~JG46Rb8mgS~j
zL{ZuVO|e08$r}!xMoWu-MLYT}=7k(euqa>erzeOWzi-s16C{XGONH{HRjX#v5%NdS
z%O8JRK0kbz(PB7)nxGU0zE%gSC7)sBSG-G*75J<Q0V0v~Sq-7fyV{tvk$u*04^3nY
z#5Ug+n+?FZp0Q2V-E?%F?W;VO8eWgNMu{z|=cdk8zJcEWMf!w)vD|_vSj!EGdLtr?
z)=zxs1Zlutu>K~oE27cAD)BHPssq;Jh}rDtc<tN@o1Y2T?gS6@d??gi>}3tbOPVQ8
ziLyhNGTCnUyBKiV@~M}7g@fO??Bp)m(eBv@w?B_Zvqk#7kcw2w4s9u$;Zkp-28I!-
z0p!HwuzK6iWG#k&7bnhR-B7OVFJN9_5oc_N{6dflN@I|J4fhEulo$r?sVDVz43c$a
z^05TomX^Yd@wx<7^&rzm%_!G~XDBX$j<<DC_>|x->n2%7N|W=IlAdCg@u+V(aw&j?
z(m;qa1#87=i@z;!#k6K1Hf!Y_*~ItrnCz-!-e`eOf^Oq~4}E{aqRy}857u|pd80#<
z`&k%uQ0t|&+xRuLak5$aQW@B?6;}H;8SEOMT9*mnrNtuVrENd&w#63OCjv)#fDCtp
zwwDjk?qY|GVx%wy{@0lugm$}8sU_Y;IOoGyd^S6(zP8+DSM2mmO<f2--qfznPtYya
zAQOeHZf_5N^JqLH!W+(1@2M@X#*dNubkcowfym1zEHAGvB8@5y(XhqbuWYdvt}o(o
zJkhHQ8!c#eH-(PTuhDD_k+vS+g%M-k%p-j4exNBA(qQjz%BlHb8OB!<PQ9RWMjWqD
zZ;MK^XIU7Og_wPi$Xd0=#8tY#Qb;Oywe;Ii(0NUN#{LX1_YoqEohclZ&QvIj&|^{&
z((NW4Q6j#=UC59pPI@Qs`W~xABk;S_FTnC0L_C?~#kyM8jmC2KLa-l=`~Z4o_7}rg
z*uSK>k>F~@jJuND+esc%*=Uwed=OtP^M|Y(Su_Muuq57E@Ns6mJK)b5QKuk}Q4TTh
zH7~AzbrQ87Je~ZIk92hsowwGf<KB-v=w&$S3(%!Jywf{$j?x{MUPK?rj5mb&U7)A9
z+e4yDFJW`(Mc!SmM&-qYkmu_+mHUnA-9zN*vdnb5T3nWqoQcaaJGZJYT!^`LX}7e}
zO^pAgOS^vVB3`_-d&Nc-#I>Zmje^GNAxXJ^>6FvdVm3M7DFn`#+xto#+>$|fdABlW
z*?^uJ!gT8hp}a7sB(YT^R%%{yCBD75ky<4uN1b?bk|HP6ASD}}m5!z)Zay9x_-FhX
z7rbKQ<Fn2S7AEzMT$cC5_!!ZNqpKw6xxCv1&jz;QJq|j)c0Zr3=ycrMA7W-darzGg
zO}XWomqGp>6ql@r9t@W}iAEWhV6OoR12H%`mmvf`D1YB?+eQ+8&tI{R*M}~5clKvd
z1ZdLBU4T0@h;u+&!!TrtwUx+}K~ZVizy7{i%C=})HbqA+`e1uV&d$z$^UdrmX={`h
zN?4-;e(~QZqUFCyXyYMhCA5!xM+v^b9Xuyi*Iqc2g&>@FEJWc|cmW{^ZxikqC4AB>
zXc355hkrx}tZYbVrFc@~Af+sMp|#0R4J6Oj4O}73zK|B$Di8aFr`EEgCPIq&sfDMG
z#}cdro239Lu=|)>N5lRT?4`)FSuL>eEEy~=ug(%AXD|oMV;wvPOc^!`$wO-GMFB7J
z2uRMdrw$PL42^?jAM(>6Owc@sgS}{Ztb>(A5r1<yJWYWuI})TckM&BB&MWe`XEH7>
z3L?euc_YZSBa$Ge7-YnUJh5KDh(J7tw~#!?0+FFaybh0HTCtQ}i>wzA90Kfwa9GNb
zW<+cTgYma68wQ%5$D4qA!DR``Atp=_;6)@3bAa_2*ij(tl;?XOl+!Lt4pL;7BXcK@
zh<}<{M6Ey05{E4*;asLc1ht%FsDOhD8F0kJ#q3{#y%th%%*B-NF$bBt1jc%;v!o$q
zKBfSP2}!`g6R?KlK~|?;SVq`ehJ}y^-ZQ}n6Zru62vqlh5g`q8k&!0^)<>7k;b}~a
zp>`rDlb<FLG%tvl^xA^Cgb)X6pGHWnJ%1~L&Gx|h;F!5&aL?zloH?EY!ybf?orga_
zp2q?a+{%&@K*hcI{`=y`W8|wwV!tnb{^eKTtE4AJElCSfRqMs#<mJnUG(Abq%>kM=
zN(Fij&?J%?;0(~zQECwLR#OzOmy2b6bTyflf_Q)^-c6dOtSe#K%YS}p{&mz$ntxL0
z_QsoK)nJY{To)8L$g{k`{0M(+F8~F~n%Z0-jTQO!BIJv(d4gh-%(WUDFnXgF*jn0L
zAiovt=L@h}b1lgi;H7pdPx|s@@vdG@kILq_c=zL*;zRkR*?{rkkE;@kIxn%-swt~x
z#nrO4_r6>$*Y&hqwFQ)K{kNRWCx5S&Uyj)^4<d$SPB6a4NY>Krbk{IzdjqppUPk0{
zF#2vVI$8l*oUls96d%yM4jb}WCN)G;Cnq1C9R&lLsMJ15ii0$rmAo$<fHa+8g6r6#
z(mmAZvZ2GEZK8E4xXrpQw{GdQFKLYJk;a%Yq%m{@Gt%*%h;1oc^To#?U4J}C9Ckzh
zwbG4LAGtURxilo*I$AA!+C*uUJ~)lHK<WSs#D4zi>|;6Q@oz6pZsNi&ETI7eUq1+G
zzk<=I3P$!3GE%>Ip@WgZ(O-t<-Ob~+4(!WJ+h-<?VkQk@=9boVf%Pcw+L!^FF8MqA
z_fE~ArGf_oG@TzhyM>?VNPiCy^IgGwsHk^0g6HiXZ0eE5hB2hE@sZryQn=X&G)eAE
zpLIii0$8qJsNASR<p#A{?w)F;tn5(y*#X-w{lup!XP!{{z3wBLh9g@1P*uy;5N!}X
z&PQR$CpVL3B~iRu*0ZwC_S8x7@8WIo8W&jh3Fe<R$AtDU;rxjch<{Z1RPUtI85Xpn
z;p9bev_5Ot;D6@Tr{aej^V4R&tcs)JfA8P&&p$4j=4$m_QIwVZeg0{F#Ve7^`n=$)
z;#EDVrWdD^YIb^6&!^=#aQZgpUhPh!gKtdx=7aAS4<E)vFKk86XD3B|E3py$sCZ&S
zl+o~O3miX%iApL&mw(7eA8G$+p3j?$^_iS5FN@#ma#qhjm7i)J4g3MP{J6M{<X&z@
zE`t8rPm7anFeChI*3Gh>PZq0UvN(s-TwJb-W_5MCnEd(2>2z7GntC!v;GW>AlljdQ
z_wW3s2Sv$YJ2fdt?c&!;kPr3bbGcZ|tMk*<WKpgzClxYfd4IMjFYn|^98yJnf8Vjr
zNK$ElCZRh-S2I9Uv!;#*Kh*4QDtC7dyPJ|uPqEd9>{|(f?^S|ehg5>7x;tZVkF*h6
z3O8*8d8AAnplLlyxw6mO`0b8M_ZZMQ=kr_S>#kwHP(<p5BI+@rh-OH7b+mS!?{{gs
zMAG`hzX|B%Nq;AUEPpUC8`Q1ki%|&>`&(fQqfn2}-2gmL9k$8(ggznlDVau5G7X|+
zx{s1hrbK^Pb>17XKAqAiI;F>sW$B(xPji(#$SGrYBJO^{l)nKbKmL?Ks&{VV?>PIT
z$zOyX=HJ1Ax{HnBwV0=1lls_%QP@1fv~7jk7XY3+$A1Bu&L$Zx4bUW~VblIyUMlc+
z>$?Had$dkBhbYqN<`6~g7oh!*OvcFk)fhW?F!^gVu;j19L6Ua+clfjA6z7fg0_T!R
zb)I|tvMN670!{sTvM4Usi)Q|9b5WO*85;hg)8LbK1Im3@>$+{Xci8gm6>KjKnBE3S
zX3Pdj#%}z5ULVlryGdu&ehl;5*ZM8<+eW7S*gfhkkO9$9nJgFt`1m-kkxibz9cuIS
zGhjByxqN=MZpzuW)8*yWy6NV@UcRH9MyDgJRZ{c+-FPnnvPJC-I_>ZK$MxlvoRy!8
z-<I`ybNK#05M&*-m!Y&B69F@qVR=9lm(bQQD3@AxATR@zYg(6$b|5o<&k-NRxs>==
z{M+GMKR-!uSOf4Y!1I$;@Hy2JXMA3;jtKSe?^AAvHIW3D2HSe$Lw!K{;PJ7~|9t=Z
zfoDcPowd8frLl@3P2y5wZwh<S*o!(Mbrm?nuDM2D)j5B`8o;R*zOjM|-;rmPziX_Z
zULr5okTO~`qoDHeEW(I?IzGTeKAxY#m&?55){FdL#WPV;W?nl|aiN9{rHbYzw#GX8
zg-VE~Ho29`wG2^+?QrIl&nP%pLP2=RgcKtR5p$9GkK?_Fr3BBUkt3DFWf_l4_(fa{
zA!}Q#U7k%<(bXcOUg+de(HJxGb=1nII*6{3pVR!r`4ag-;e{`MNluwJK~%%*ZTR*2
ze)#h+3X8CYaeROKbqF7n1SXxBq_NBhnEv|r^!@O4B-?6joCR(Ws7@AvBdBUJQZm%l
zg5Q@aZ2=3Eh{UzRX63)XeoRBjZ*P@xDHtHdfwF*3PdVeEsSf|~OtCBZE$Fi6i@Yn2
zmPU^~1ipiG4%%3MEyD^D#xO8q#&1R<+8r4w&9Fm@2i$!Gc0J{fz;59Elr>P2BnLUa
z4L^&cfH}{{L^as+Xur)OhT3@s9zFoWf%1Uih{sSnS~-!m*tNEStl<58Imy8K8DD&u
zTjOaug-#T^l1p>zaRqj6>>;ea(E^^Lg}P-<1$WHy8HV~PHIMpn9tDVCqec*Dak)+8
zgwV{2neg!QcJ_p4-p(Ktz8Z8|Di)oO6_@3BAW?t*Ef%1@CU#aZEyUOxN3eMCB43xq
zPH83sq5Z|Kr+iK9`r12AGp$%d5A%7p*c9XhWtHtpmIQeip<LfH5Z^nj=_Izp#8s#C
zHk+C)-$3P*;oZQVbWAxrryO1E9)#<YxNc)1DBn1z>h2CIL5-uE=>z%k4^hm^z{n#$
za#?>DG<x}3PFgC48{aE?rH8zcp7NUMCC{R-EoXcr>$2ke-GvniP2VFd+YF%uv$zW!
z`OAB+;=S?w{!({-{M1`PLR?6uYMdbME65KNb)ONfVq>+#==BmRE4`rRD(lvIAy=cU
z2cbvZc%*Kv$7e|X(39aJPgDRyobI0p654;B)wPK*ilH!%wIyVxc^UwLFGaFb>&Cq<
zSNwh<mr60G?2LD{$h%bNopCYN=7iZD%K=R<2d2Sy%Yo0+O4F$_B2kc5lut9SX4F<i
zEzWujU>rAbj}ybR>6jOp*{naxG`i(i2Y{{6?X?5J>-<YzH}h4Nn^Da;ww__7Rds)&
zkjzUH_{l?x<ynyxDe|%9CUCOWuW`(%>LFh-JLvG=YbaEBJM|IHkh5s$)Y!v{e$zrQ
z)b95I&FB{Q35`^4@G8=JKQdxt{O-26Me^B{GPs#6t8tt-iX<JI$+{X5=!NA0L!11K
zfgdEl<|zvlJA7s%xT>Xj4a=s8Ml^q4O`c1r_?&I<_IWE3EpZQTYT&Wx-ra8TxN%Sd
ztv<I`C|?VVDlV9F{sXo_1LZ5*prQ6h1g7$cCF-`97rfH@h^F}bKy<~uL}$x@8!D6F
zDRLZKA0*lmDa%6GcPM?YFn8>p$er#9?hr2{^E`0+NJvjHrD1;#DOT7Mt`UE|`yyp|
z7uplsBe0kE69bnU<9gWu_e8@&VEFy^-=Y<_BZnKdn?(g3M@&o95ndjrC@gbUj1(7^
zsJp0UOE=>)cp>{3A&W8s#V-WA9f6+WUI>hQ?uP*G;<#2jghdJhJ;l8c82Q`}0p%F?
zHRy8MjzCXwF9b$D_d`Hh-0pwXP>S1*Ku>Wm1V%phLqHf@rHf|>$L$F86!$`4<Z}lE
zGy{(Q3BComy_`Hkp{KeR3QN51CkyBg30?eZ(A!byDei^B$mebdghZU`XZX$I@96C`
z`hmy&#_j}@HEjsY_R;qg_Zt7m=L!N#**~xEvt%Oh=*2=zCi-*2if?~Ytc?VucY6BY
z6pFDCRyHv_h<$?(tR6o7@#Fj7zWoarHu!I7!KWaY;NA2q{nQ$pc_F8t`8PVpV6~gj
zpXJi1$#C5N;$E*bECt`DgA&2&YuA+i45c33b|pm9IA%cEnCFl?Xo`r3K(Ulke?Qwo
z;;MEL$5hw|zXFxqHv9?1_S7o;3{U%;mo|MMNCCZ<e|;b#0e82eeIRN9fBF@_jVCHi
zhJEHGe$83wXDTK8Qt6V8XWFJcC@l3v%2CgCiuGu}U!oeZ$fGTXbFF=K$)05%m$R$a
zR2W%B&uBKU;^x^)Iv!4)3$nD9obnOwv7ApO6~8EGI-Cs)KQfd<%QJCA|J&9$RJrV6
z%awovCtVbXw(zA6sa~ycf8EuJn3v`1<%#|Tr=_HF8cn^r;4WU{$ji^gs~qX_16^>q
zbe-dtiw_HlC(T8WI-2<u&b}<EO>?gG==BNXVp9KbdjdBT7R6)h<p}^>r{qVR5e}5E
zoDq(-Z#R*TW(y2GZb_ZY83p4QE-L;^>vu_`%t>V9tdjfivXZlSJ;w=kK6VVkSp!>-
z-do8>jy<u3Yb$%4)_cQ*fIV)DWC7|4t|km_zv$vL6GmK3Ggz_!;~Ndk{{h&TZ#0*o
zv>g+dz~na+ml%u&DVLUjAeI7G7?&=AAgzCThUf&^rjIi6qJ>BPa*ldyRSg-%bX~2g
z6>lBHtMKrMb*|PcXK`>gD5iQHA<DV{-^-fik<4nei;Rp`MOs<T3@{TTizg{r)zV&=
zaF+HmM=xGZZFWo{nF%D$&yS6CAptIJX~^7HOzY$HTc7eoDqv4c33fImLiq}K1?+!e
zit^hEqVFvmy%|DA2j#Wvd*g<K(oBrZCMB|*xZLCTtYvOcKO$_0xUo@r8&~*}r*TFV
zE#I##G%k4@ogyFR6rs@8w65P-+F^Q@X1_%n8RemnFGvh~0^&U6Gf8QdvPc5(ybAT|
zAuuU%cg8NlJ&!L=Cqmof=fDFg>lT05jbN2EScG!Cg5N6xwljDeKb9qj(rT8oS|Qe`
zxYMPD&pD#xfRbHhO1V9>K+ouob2L{%+(x4u=cY354eDyTgKQtq@)cD``wmr9wt2&r
z{pP;hYcwgcYl}DOAOR<o&;AcBX>O##<mhL(K(--9qr7PwU0vXhZ*U?=Y|4LH&fTzl
zOI}%I6-VJW$g9rs4SCh|UTmtN->%d156CK1;NuF{_MTNFQ{Vgd$v%TYQSkkUm-Z|X
zQoY|$f}N)OO!5y4lu49Sjq(YB^0>qx77fnf${o}N&I1<HCI+h|h1XA5^eP?ZRv7uh
z&1ECrL0woaTTM6*(76P#aVmd=cxCLztg<doUhk}42ol4)UGnI%cO_F0GFFLGlt3)`
zjWgq<1m-RGq}An!7?=6qPMZ3zL|D?o(xQF)JXSf!b(T$RIv~f5mvi`iiMvRd&Z^Uo
zCmaKc+-l~Zh=<A^;FA(AC<)J@CCh;5Yl#Hx(=1aYrJ)M9TpW3lWhQ@^wpFr#f1YJ;
zl#Z=H5T0k*0qgRrEiyfyWZAOf_w7V>c%Eg992hMLAD?I01P=8UAg3o-whDC28^|(F
z@Ph5F>e-jG4MR{idIq^sP(<&QC9^os`gv(*d_hWu#GEoulVZqETz!&~RII;nc5)Os
zT@rm!j9G0=$tdHyQt*F7R<J<re0PvSXOOqAmrIeAqi?k=5w5w)iKRkHvKQ1zf#1ts
zl$}xfWza$76PEpB-eU%?yo1?=28x>w%`sV8&CZVLSvto}<2SWbah3|)VS%U<Z5fff
z4|5(TjNtNlbwt?DcF2{WuAZsCGAdfnz6#Qti_1stfNh7y(YSy8-SlC?WjuYY>?HUq
zZ6`9<mQAnSMJE%TqIQBt-PXyT#uu9KTv-+A5z48*w%i{y{6^R})|A6Usty+3&Mh!5
zAe_z!F9GtO0C7aSuL4LQBqDezTvU2ugOR97KJJAcW=7!9NKS56FG_EayUQi@EcrGf
z?OTmTv9B<2QbT|1g3H26n<;N@KA8(|mMjKhzi5f3RuG$cdEm4y#;sKWR!ct%&rjp@
ztSVwy;dE0t<-coE(mHP9v)XX}UcoE1qRjp^%SMMyv9DMjh!2@zKGjV=z6ygw?Ba|w
z)vT^1BQ|;@JfU&8^v{^>%(ZT^2zq&T%)d8kVTs@o6Bd8GMz;y=#k@Gr#}(PlSgUp)
ziR0;w!>r7EuD+XZwkPq$GOA}@zLeotD>G5m*bq#T8+--igHopHs`XnvUtG?IEG{iB
z@?jDXM#K+-VAJj2n}iL`w40$ID>c77L*EKFRl;m%<@^T&%ft>vIWJu+y|jI{{*QHP
z?3BH;?0htJH!rKucE47;tu3KdHiSjb@;ahZR)>DRtNQWH`QPPTjGmPw^rQ9vFv|Da
z`|f|ww=yP|p|l+n0y8$3@m>uTm%jl(CYM%+AT@uk_v<UR<muZ<p1wTD@_v1L_~F&~
zlhVRi=j`e2>q*fSNq?1e!dr8)5`>m?>9^BAo}~mUC0^-;QW{M!NB?_zH|I|pJ>8*h
z=bzvHS{Pda9?{Jh8{-5zY>X>Idf_X*ZuGK6kjtM9&`Z(Fb`e1!fSjt1AxcnHf+7xz
zs1<*nKhc|{ue2^qFH0*li8fKBWj0DCTH}eN80BCp+8JS+ZQ2#5C59wIMG=v$@#Q@#
zVT#olJrM75&qX!lbXb*DRcYoW`i!ng@Y#SQH{N|o%1c2Pv_DCqjrEq^ZIE#KFD1qN
z%1>V&s)2w?cx4v`LJ3&UhEb*Hc6>=@B?W&<{xzm92<gLusZxJjTaO|rSZeEf)RS9Y
z^XKGv*BoDi=9@chq05;9>baz-uQZE5VaXM*X1St0SsnD9)q|_n5)Gvhj&hq5D}9Ah
zJbvnDR8BbXb3dnQ+Dd7v*WmnWPUE|BI^V@}I)B6Oy|9eMVo(gxt=^56^#aEzO_zV(
zDj`!CPOoogr)qt)E3T$SVETNnx-?E$G;6%6hh*umCHb3Gu`HqtG~J@+?VSdKl-0_9
zL<hmvzNqg#^DfG~$BH%cPn6eMD1$_33En()&Lty@y|0{O<5uO^+T9!1SsU&pH*dpL
zhuR5w)_B&2dhe?|WhwWXC9ea&8ts4d30|^AF~(Y-l`*|w?m`A2tT8QQm$S(VMrOLE
z{Vj(k^^O!GTqbmN51*{c%pJIyO2G=N!%{D4HmOJp0e8kEKXD}?`?1FnYo^s`X_f;C
zG;E7?C?u$(%6cgjwfv>Fy+${w2_I<G2W667A{bm!nVN@b4BiD+c7LTBl0AQO&9;`>
zdg0L@)sp-f;4b*<xsNPC>*#C4uGy2WAn%K{vdnY;rgsS}(MTD<$ET`X&GVuQl@d^L
zUSujx`M~XI8D}|4Z1rr{JKmtLT`g=Kw1_yP!iR9yrp>H7xNx9Dlk%_a&dUHbDsuDd
zaHMwMHXngSaO`-RMd9<gL6v`oB%|#w1#JZzQA2$DviS7*?cqQ7&H*W$h0`y;J;V#B
zmCzn&Z6J6;DD7XTuMdA)a;<BKkTQa}_=o;Wh;I1ir?R31NhvOE%462wC>1;vBF#qO
z`y#OhdW+D4#_qXkONy%{x5WU+F4|XnS7#FlnD&NsE2XJ>nv}M%t2ciE1*^@HBRJN6
zF(xf5G1&(FhuMRknxg$kdq$2>N((OefBk{FYXY{xeWdOqN3B_bx{vicINq8*LidrQ
zwmpYxn_RVLR^(}|@WKRV^$h5OCtpF2^B&O1ApHA*Mw4rUF#5e>0o=Y=x+^LtubE#g
zL3yj+*11qPVV+U>T9SXJ**Qm@YA>_%k4B~Y8k**SVLOP7cUe9+Fi+e<D_MhFO7vAz
zznDA4lIka0Il8JrXjHZ|Vs@evaXg<*#S^}4CG9k$%%@A<%b4d2_U9W|k!{(Y^}^7q
zo6d8fP1d-Mi3(Z)Z&@tvjxpHzQ{As9cMl<qO3~t&B~G@Va^HV+`>Aww705VYKrx>%
zAuqUbISqSS&Y!o8zQLBrdbh-kw{pm^axxe33<O9OmE#I-kl$pt3P{)DrcC3tO5Drr
z>^Ix-Pu$t-#$TlM-E^7*XoOW^+!**VPQpePcF^O?cXuakiQ?`E>F%tcqrBVlTc&H5
zWm$yXaKN-i+wOni+jFnDb~H3>x;dVGbLkn-`a(NR;M?=K!K`9uDo`5ZeVj<<!Rw-?
z-FWKT&wU1*Jm6sWTzl<smavgK#YdLMyq*v-{C#1k2afqg4Y9R*Sezm2A6^Na5k|8R
zkKV@3S0T{49*2I1gwe*xu~J(<r4Ar4?)pKX&FTF3_oshCp-)x>r@a{|v_R^9%6TEn
zCiVDhC9!i7D<I%V0-I8_RKn7wf#ffh1ZJja>B$B)IO6e_Z7B-UnxO&;7a)^-d*X@M
zZXujOvuR=OG$L9BVPT=jKPiyg>H`WZ^+@$?TYU&Z6N*!w0FoM2b8ad`qwvIJ5b~#O
zy%5v+!4ZGAVi3xN<W{AM&wk(b8O>A=aMWiOgkSpX_oq+b9{>FG9u%hxh~8aNcfGbL
zT6*buO9Mp?O?lcG?KnqiT1;a77$u}PW7Jl(r4VEiuI=3^)ryAXBODdn&MIY{Pz3Ui
zDCl${xl^#ak+8R1HedvH?2~{H8)3^4(H{b+35<Wtydrp(ahyQexl2gptw|dW;MLmk
zak{r3V3oW4QdX6b`6~uw3x{s)R*+CjYCvO@e%*``QsNOpCkiW6C{g;~zc<5#1X|>M
zm^PjV)+I2_&-8?22$1bq7cE0`ft?QK={uY4X=Ma+G@ro;rS(9)S8~)R$@Wx7fMYva
zNoap6QXRfmLqpimxKYC$dDaLMJNZN)IUJsMj`EIHkaDvMQhu+34#Cp($1?5Smr3#r
z$FF0J4M6U2BAylM^~Wy?fI_K<gNyl2$imQ{Wle%|<mC8e`oL9zPJ=WU;S3bjuI1?d
z?q;2l%5Zagy2v+m)39uI@pd*7zH~Ii8QXt8K;*RCT%cJ_6ogK!TLq+RPo1<d;1*Wo
z<eP6W4gX)i(?YN+{gUMU7J{|ZKyhMW{E>*morg|JIIq_&tUrb*&s+K4T{3Sdk&UOH
z)A2PtU#P=s_m?5<IFN57E5i&4yMsR9Pedym_4J*LS($`HtO*^3Tr2MY=xDVV>V$ug
zCg3L0L_#=D*||+~wOm|Bs0wK2n0M_B=#hrNQN`$PX^)gA1yQKD9&9J?id7-jcJy`p
z!OdvYR6nB(MQ3!}k5A)?@=7Nt&Ck*D=)yaj-*9vKwb_0Pru283msS468@hwpl*{+c
z$}e5npK`2}=KES)wgdGFX4a$JUZj5{2piNuv71C%(V|BDIML3&Mkovd92BFe%qiuj
zVpcZEsF**+i4qEO`?|BPQd;X#vF%+@6I_!tuEx|^?t-wt3u+HSIyE(KMRYuUSqOe7
z$czuN)d5e&-)@<yA6Iq-3a|-!rW{}C9Nd+Po|X#yfl{d_{3C8F6>m{!z=<yR_zxhD
zVAI{h)QUpg2cXbpAJfWGJJ*yMg8E<kiCrAJA-)Ep&u<U^2kMsV@t2{r9TNgGHkW}E
zJrtMGs4*v(`H&!+e~zB>>DQN^o_HiWy)&gkGC((IqZq-9lwLSCtMY6qm?2i%*+kz-
zBE6C5z%%EHc*4Mwp|1+*Ax_lu_(q~sX+JA*?f(mFX$1Xm<JZSk*Z7^j!G}QAyJu|&
z$D9^g)lwJIFh|nX78r6DsC)DGAL0@@Xy%Nx*ep?&DP@Que_&k2CI3!K^vxv*O!%RC
ziCD&hn$$qGOa7LYh?~4<K^mkRNx%8Jj71x!!O{}JP|21TV{T$GwEN1oSpKq1n588E
z?UFy1u{g<uX?5()Ul*}RE}3!IF8OVdeaS6zo)wY;M()2XBLPL+cSwGWOKuys1jiJ&
zy#WNJzhuQ1f8KaWy%&)y&B#DEap*OVtJ^Hj(2F)WqK@8RR5}@`H?5KSY>+1M!Yh+C
zdDKO7&74MqR>geDiFvwuhp35b<T^V<cS7YUu%p2jN=64FR2Z%Cs7PIO3+n7=W#<Fw
zKkrtB$zyyRPkeb-<oO>*GworuPA<#l@x-q!l87VMe+H*IfnN=fC>I~ne|}t^PiPaR
zksAI?qv-rxtml$sAbc*s>gB<gK1V@8A~=X@^ttC6biIWU{JBBapq9F9LV!zn&X)jh
zN!65=jMS{(;uuNhR~aVkL^0+fN)+q4;@p;4fk`o9EQu6Plu#;$n4_|fo+y_fWyNny
z!NiE3e<PPEs+#ubj?BOrh8B8??#N7#bJW&N_&A@0h)S7t<2vfWNG<q%CQd<}-a^U|
zW(*ZHvoVHxK7-+ayoe<ABoj;$;OT47zw3x0$VEoe$x;hun53*~>3V(3rj{Mf)Ini#
zP`a6?M~;&dN-k`MM|6&HZ7TT%`>MFuz;#P!f3d{mLRH=cwV=JDTOHxYY==|q@M)NT
zed_-E`117M30u-qSaV|d=eH+6pd>J9MbH`{?eP1*hA&V5=)tD+ka1c=mN<l8zkeON
zFP~%H>o{p9*BYA$1;9yN)qqA*o|vPJt9+-P7+0NWzI!HXxCK}_3e`Pe*~q(=^|Fsh
ze^Ta`RxbnzkW39VG7H(R0@Q0El@frV{0SmzzI?9oWyLA+!yY1hD7%{luU(S0z!7Tm
z<@2X`sVCvIScHp#P87~VMPSy6+)zFxCbaiHHOUMW?YwAY=5b#3%dCa?S~6>~93`_B
z>YU7!WY)>rOoE=J*z2`hRMpYw>o&{Nf6y1uJ+`CenK*B;JSa~=IlbW%+UMj+)v2S9
znRL)aoET%GD)~&1?*X#)#Sm3C9p<BD>wx0D?Xf&6TLDbz3GKaXlSil7&S0K(jk3BQ
zE@rt(y9QJgLn)<FCTps)U}ccNu`cQ$_Fi6*Z4%p@2t<2?lh-5~ZhDA<-)^BKe{-C}
zKiz_nU}Y&=aqb8Y8wu_p-XVAv|181e5<tDScZ;^sBAa1_&`@S%c2PFVB^5AW;X67V
z!4*>oJ?1%{LHHE!VKQgEu0qX(5J!53=yQ}pCUl99rTkvy4H6U~nN-<xt;@c*>tXag
z9kw}}(wc!5{cQR1wuEU5rPT*#f4$pOE*eftNvI{kzZX@nT@=-56Z#mcdh{xiAn%i+
zG;_z$f=D{79-+ljc#?1gE##7MN{YQpg5;VZPWF>u=dGO676|kIpo1+Xp@*?T=t+gz
z)w`6@Q6^+3bZk(7q(YC=q7`L2aze`;{Go_8N8q;@x_3*mSfPa-9oN|`e^fdib}{*K
zeyc;0v3#YK<$b$X0?3r@TO|~Cv~a0Yk^q!c-Ue)ze&2vdfnaRD{PHUwextFUXs&d2
z_#@2xma*{RKf}+y6LSZyrP$`-DrV#&U_Ow)rIcg@TR@<6;n-<2h?SI4ab}EM21^_0
zu!kEuNpsN~s49{K+z*z{e~Is4^3uf5hZHvTU{7x~_nk-S%E3ev+dqzHrNJZdlJUt@
zZ^#PrNAIb|dag^*BbGc;GGQ7Kab!bgS;O0tn_X)t;VgKDA22nQ^ED*r1Uf&t7g-dc
zv8SwJyLI7?j!}_n0gu)(?}rTmr)-Ru4EEBLgHfiG$)%mmUagj7e^rkq*_IVIV?(`q
zWpy;A<u2qM1<90t+d;ps)P~)=c1Ig1m|N7ZqzwY8IFzfkK`%=G&(Vt4(*$*c-nH7W
zd)Mx01Nbv;^H5~m3Swn0Kw!1Ddy;`Z=~!{uNLVVwUFip?e7=9K&dbXw{@pZv-|>@>
z$R-(;v>|_6JbPNSf6X!f6IxOkr^<<4&Dh$?W6lyz;f}$9eHwBkMofr4xH&AOVuF&o
z4$Gs>;aFI7P_wCTkb=LBB&PK7#oD;Xd-*P%HR)S=%B-YSqGXc%SvPxMYc0!mqVh^G
zZgg|gs8aa=ZLKA5IW$sC8|bmmwX@<eNH=?*d2ZJ*!fMonf7TUk5P^;voY0-Me0Nz1
zzE{q1as*GwF~GjMQgBChWREW_1y$h`SRRO;TOq8Y3mg(R@H%BBI3JnagZ^cjc=d_$
z^ZN8?Kfm|nC<w?yjf2e~)~U1y3r>Eg^a#k<Z8;Sble=8yCX(OY@rH7oQ>g`teQA+3
z<hD%j5i}3?f2qI6og9#cA-Gq#l+!_$i_#m7ZRUdMX4gNSFmwbQongh+afjv649j2U
z#U5x?t+h$}_Oxs4@xHQXV*mG8(fFn;eVoYVxH|_RG{e$WJmAhj2k}51Pbb_RhC4^x
z-R#@6)IGyT^(~%htfVrQ7yUkT3FCoh$-cXzOOJ2oLs_!(7jH|Zza~s<x%f7^(-z5G
z${qEtoqXT*?odBuPW6aFAINSxd20bYaPZbqf-BG5g1haASW5K!H%0aT0g8Iwmto5p
z69F`r0SN^Zm+qM$#{%A7x3`)gYY7C50ti2sp|l+n0y8(4u}vxy0X3JgMFuB-y<16h
z>oyF&=U3<tJeC`{%w#+hJIR~rt!-|dp1NK74(+9XzaRxt++<6((t~10q5$y2LJ)-c
z^ykE<uTSTXr|;gCo-A`9teignm;^rH_*ZbeX<)KKFr&ou@ooC~nR99II^(4|R|+qS
z-yMF!>5CE$M~Dkv;`GI8@m!35!%M#7bs-x1WX&_ZkN1?sOW<XYj+c@LN;ppVK=Jg;
z$M;X1gckTIiqE%?r@x<Cv5j`jDmnf7?I}FaN-*W*gdftW>7U=HKc0Tb6cT05T=<56
zzg8q2<AJ0x_-O;lpMT-gZ}{+g9IU*X{*6Pwh2fle@Y7$@Pe`AaFgt;NFmq!7Dwq*o
z2?~jOroy8cX7$3kMpytL(f0Jk@Bk(tAfV<a`kgKWSclQ51o0Y1$^e}TN(i6n4&Fej
zGmynJEAT*h#<hQ7rZ*4-VEFx$5Cn&mOhyn53(%;*5=QR{MQ!MbnV&AGaZBc_nrlXK
z8+Lck@fLQH33;8YWZorzsw2tvQMe@+IT{G&uTo9E#~Zq>_eC9pn;ZjgjT!|bBLH=s
zW05PADd>}vdxSWmLp;)!Tq!w^CZoXr^*N1#<Z!mU8x1GidBrDgF`0ND$N^m*og6ld
zg4rdbJdWfmOp)GbOc6{pvRg43yupV$;}H2GO{1UF3`nkcNBfz7O9~m0V;Nvx@<?|Q
zHqbbq6n`ek;Up!=nsA(HtAyLaX^7WYGUyOHN%4FGwaBwZEmIFO^jVUZg~B|GMf34p
zNFybV6go?!m!MnFg*5M(68LJ?m#n?=4qxm|caqNff}y^k4_(13IxBFRIq#DdbYV?U
zz`~dq+!sH-HvbKOS5B!M6G*mQIhAHsLs~i;vN=?6e1}2!gWvx7b87DVNN(==IF$2f
z*I*<r)7ieG{kp<Zl1I-tmbt;|BB4DKx}zL*Y7)w`ezn|UJv6Si#9+SsJipH0%T>E&
z!YQTMfY&Nm67_}g0WbioB`D^=V+l0a4iZ>qa3yUhs%)TtM2#<VksE-T5uy3&FtPL}
z&bPYr<JWAD5)DqOZhk^E$zU4on0B^9gD*j5(a1g07|k#T*eZ>246mUy0iE-s(s;$J
z*IT9WE+zoY=-^fn8`5nI+_DcO*DQIr<jQJhWQPoY9ZIg8VnTIVy#q1jEG8vlBVss`
z8ErcKdD0($P?Axl0W0Vo;OQwWYD}z5&TzBe*=^pPs0j2F?aC0C7MkeYb~m;{8C)Vn
zOIh+Ewit$S+3Liw%c>az<T*pvO#l}v?W%Pz$832errVew?4BI%g@lnU*R{z7^R}WJ
zZHv4UKcz21Wy*ET{xBdclLA_9YhgeNrj5TxwxA7v@ka8YKakl1$yePev7;qQW4;PY
z4S}1yDeVJF3W~nWoN~@K#5{3vz#t^dYis*{-GhodnTB$sd(IG4@@7OLQmpP=b8UmW
zCvH0qe#z96Zek^5qpOGT<vO|58&cyjpx>>@VBfe5cAAtoc3HH5P>xRTY{=26oT0PD
zyroiqMwrWWdoT*qnI((7D&ZU>XfLqaPOQT{Fti+xv=dyuG;Ru84v*PMiXRQtBkTku
z&JF%|JAr%bL<?q>*=;9qzn#<^Kf_KM4hPvuP4x&nQJh(8kFk@!Xu8)DglA5;J(d79
z&)pW0^i^{{(jq)CuT{t4L5q;UtnB1I!XmhT!z|B_bA|Qd_a|9|@|O83RlZVwKhPd3
zzMo(XHT`33p`v+!B`9Z5RynYQjivok-=VaBOV<{$ajM*r)Le!&pZ#mVyhCWX9l8eK
zrrG+A!=r0}47I+7XrtzKtELPvE^}gW;9gjPj>jTpiLl=ytR~1w<qef$D=GBacJlFm
zIV2NkxL>+T-4?@Ions0mHJ5yz<#uz*rocH9-gcB<AE2n22C*uuzMnFsE(gj=+vt{}
zsm}>B<vRkOMhE6*5cnJ|K;@QwS^A2Dgn9g)11WlA7(karb%jr=BtrY9-kiG6ck_QP
z9kLMBOEo9kQ(F!MRtq2*UnQB)IqA%Qvb}6C{@+ooNt^P9`${1x`&c9Eu?H5r3r-Fg
zUaeGX1t;fO2-WrAbU<<&;<EG{7n)7sxyU)TPfQ26p`Vx@I(2?4tcjviX<QI5_vbwr
zV1#ok88DsWrW43jm%Ms5jx*x7PsWM1B7MlZ*bO{~hvw{@?shW4FnpDn`M!XEHOTm`
zx;>^2Qr6kL%%`inTzM{GHw4#8!c;_DWxqtaXxujxmrL25;!X*VfO_z<`vO~a;8X2p
zwo@4#8KpOh<C8GDt?`qtr1wlNQBN)H1aWe~!_M`Kvr43MObKz{*+8Gh5>acJ0$ByH
z?V+gb-MY`*DD`f{SPQz8@HC@;gP`hdKWp0CwY>Gp34>?ex{fzL3J3s<nR+V{;F$$~
z6cTi#+wvZtNk}UhNV$%Ld=Lpb(%puH;*u$@+R^ZGkTfcIt=cLKE|%QskuYtj85LX(
zF(ehrARFaPsX;66rt&}taD7-wLCzRl89rlo92`X}c<`iHV))b6R8bgz7A?iuor~r*
zC|k;MgyzgD6kO1sIG^gy;IOsf%qbN}1-B@PDU$~=I6^M^I#-;OI!sz_?^OyjK9t_&
ztSK1Oe8fuT?*!vYbm({I^hKh(9pR9LI59xD3<GTrqAeu`V;oh;aKLmlF9lkoZm~^`
z+`^;fTi3RuNyRL#^E(@Vxv9fVV1muXJw2<}dvql-@QC*VO}e{DY?DlQ!#uk5r*vM2
zy{cRawU2n}Qt_ehFH~6xpJC*gik9IS&^Sxq4P8~g{MeSa^>W{j+AVD;^aOasPIi;n
z<{E*lHRgzR?rV*w1I8MYH6>a%7pMSbt~Z@twa|Gb6JT)VbPu(E42@Xd>Q@i^rOn%s
zaOecZ{w-FmFHIx%+>Ty01;;y*TB_s0mCEBvimqxcCz>L)A-^vTHB`gC)n@JmrYTDs
zs3;PiAeK_4pj5t7N-w4CcFgjYYWlr>QNVN5NgYR)9qFRzNEf4ybS2A<^jVU!a&W1+
z#;x}yzgIpiePeVbZPWF{wylYciB6n}jfrh*f(cKIiFsn%wr$(S#CG!Ke%|kI^}4E8
zcdzcMYj^Fv{WPNzfPW_0wNt|LpA&=`y*i`*#!FS^>>lDRNg~7Na9o&?aokF0T+_4D
zVDR>;Cm)%1b<P6qyq|U?oJ#K6z(oRvp7-&4TWAad_0@+j9?_X_)UHdWGM)~AX!P3S
z?KIDa`mn?Q+_2hdh?gB}{<yQ>_<bL<ag!2MRr|;V8%;Wv0a9N7Yb($hYbWf$_)#-{
z(@|XWTrg8#o-}ieaMB>cPW_iDCuz_e5V|Mh8NJXwc*s6W&b96;I8@_^M5oOiZ#eO7
zo9nv$(AQG+65d4Mc5Z(%H{9*^&RnHq|GERy^hd>d8KeVKL!eR$d&V6_Y>GXdPbW5>
zSZXXYW&4M55C=!y-Ck~rzFOAATk&k7PbmhH86N19z(|0?Nl%OW_nW>r7N9g`Xe8VR
zC0&3O@PZ<cUJL4ZWyJvJU<WCMa$$g8?x<-&7nP)0K%Q$q<}RPlXDRc%rpe1Mh1E}v
z)Jq|}{7HAkRprOE=puceg$>}-P!^zqCHSJM^CWtHy;ySoz3TOGt}8XUx*8Q6p@4^~
z-*wfwx+=-Le&+xISMDv|^zqEyLgdTQA}ftOB55OdKiB)&2-abWaefONXH%qTeK7wa
zukktw_`C6);x#rkWI{WBG2f5Wny#`hyP+#{vJ*v@2tR^O9D1hMK$o8Fm=9*==7@}g
zey5;VhXW>P2`UJh_ejz2%Z6lcTl=Pj)cPsakfxi$dloV;GK-@ITOi&@_e~Z)?lphJ
zj0RVXz)*okizQBNhD_#weHolW#jB(%aK<ADcpUw&K;FGcG1^jNmm)ug^a1-|jQbT2
z&w927QE->~f{s@CY<HI$CO9*Q)cO^THHYM|wHbv!dIFse;f&>qCu&dy0`NTuENVc$
zm|0-p)z}T)gQ*+2-RM<(PZt6pWV&Op33oqGG-q7BHUbcpp2PgJ)g9$n^YMRS;L@w{
zV{4Whe&P%_K1pLu9xj%s1B0uG_w~se?%Bh2RIbuslnN6H$J#Pf)iBW*pkz~CV1uCT
z)^0O4dbFSRstDSZ)^XuN<1FySKf@c*72rN(iL{ORv;^&%j>XG1WJEoWb9NlHd=2`+
zf1gpdj7AHqvOR@#78*YMhvhO5__fND&~@kc83U_pCq+1X4l*C-)gE8Q>51{RHfv8p
z+`-4w%2>1dm1rJI%MbQWC4JN*69K~UE-Z`-U47U$*Q+Uqfg!lz^(XO2;dBrbUow!M
zYQ1eRc}L>)si>p264@tjl5}#yzv#bwJDV=S-Iq=uE0TYSiX-gj@qtOVb$@I|o6LGT
zWFqmr>~P*~gv3;bI+g4+<Y#gLNj}zJ$<|Sy&a<<){u&XziH{iSZpVFSJOx5={6-kI
z*;dJ}<EhaV<$`u23iKg?nj0<qffX1z+84Js`e6VMk6}^DM;w-u9!7wxRYh1t`5Teo
z%ZP*Lp8+;Jghi<!F)%E<Ej>tb)Oey-oX9A4YLPfRhsCducRv>QWlR&cxV=hhx>$_C
z!c>G0yAKP)iSb4faqoc7Bm`pqyTv?(8cgLY%HhsmxE5mf@4-dk?xC!u+g9itR_GD(
zeu7CM)>QkcU#T*g56Pgbj3)fVHT`?&0*x)3jf5+uiGvI@fSMVWq{k(XZWls%Y{Dq;
z8$Ap<tYjdn=>8Zc^PIZG>&RKi47KuOjkmxTog_YVfE&-J#d0;9TeebaHqxjvH@2Xm
zzo(Zidg{T(Viai$2o4DcG4!@W_FWu{e~nX=Ic8kJ^Uw5C`$WOkPmEDyYWj=380|!U
zwxWNVx6@9pK;V`gQ^nHRq3z2Cf#X|aJ(+EDj*Ub@S3(AF09J`^7wE_sRt`92k~g{P
zW%*;-Xxx{vj;$v}UVoq|H5Si=Ym)Y)Rjw=_jJLX2B>Q6O)3u8oE@&Q|tKgYf>;wh)
zavLqL^PJRFE;ezuFj*k%B(9&AGS?VzjJ0x2TN}v({7Ds32?_PWG7fh)C&5w{XM1)E
zO_suj<}timZ&)gb6PFyu@J>Y!T4Q*y`ErOjXsqZh>6BVYJAlS|X&@H%h3F>2_g9%?
zzpITcdlbm9T*aPWOO)JqDz@!yc{|yus`N^8q0Jh!rA|m`a-R2AzCKItgXMs&sPuYm
zB~l~-s_s)YphvI=N-#71rLY*wD?kIkAL5NN0_is(*u63xcHK5F6KIGS{Dx%%YsyE(
z`8h`wmq--c2BAiyBO}0Tw4X?kpp_k=Mrp#q%H!QdnLkLM3BV5RFb}MJgzBali!5M1
ze|O1Rm7zMp5?Q%~!Wy+O6Mm*8m_Ksqz2ZIv?rF5#4N}Hz-C)bp%+34ka!(T1gYbn&
zmM3mC$T`+w>la7+gEpQfXYY7}yF>G;c;6zI9TQaOsJ@eg>4rs3I}o^1;UDz&USPYX
z(jQq)mGdKotxY@~?Bpl;(D=wZGIdNl(EFs1uLW%&y~*yEOsdwoO2?!0DoB39T*c!9
zqk0r1Gj^8N5-~(aX;jq@X)q(^X8XBOErx!D1ivARWZB=JXtY`hC=lFsP!61C@JFjF
zs5qH!g{#sihpbf8-xs_WyGx9622Iy(=}4xr{;sSn@7cj~=kHkYQy~(uo0WAL-tVT!
zM_iFA2`C&D>(cwT#656M<{>DhRd*!?tcvjtQI(zwtw^x4Ad2;;cZ0iV-s7k(uByxw
z4>Z|XA!l>3)+xDFRwZQ@>`;jl*!wX0=B%g(D49wA_1Z1;FL;)3c;K$xME<BLjDesf
zdWYpiv1q;E`_yGlL{*c?OTkv$t@XfVQqtWW(jXt=#0N|6p4PIYl(yl%8rkR|Fyx_%
zeh~2|@x_MM_%Xtk<}VdBszaAi|C}!cg#ugXedDd+Wa_`SHnq*UQ%jAujw$5SM@o&v
zBW+pVV-~mD^^ZT7zrzO>TNpa{TlpF`!4<dg76!p_A?7DP+}C|SN$3zZQiU}mySBw5
zhg98{YLsUt)DuRNeaG#KLN%o81pX-hauO)4lcOi*)ssBB)0ySbDfnG`df(UsA=FcU
z@NcmvFKH2%??J&_^niwAc;L{QN|LN*rsQ@vpgBPPwZpHNe2Em`3^I{Ewqr8Y6`Y1^
z<Ah%AF7GXnb5gp1cE!&#@PNrTKk~w&3}qv=W9O(chYfRT;hegZ<s|ND0@e*2^iDfa
zM8C}CA<knUpH6ry9+#;nz3z%%$`IM5XxxhE+77EVv%d!|Ow1VFp6}_ZnPo2gz1Oxr
ztXPxBxI`?R%Cfe>T24*^Lz1L)Tq38XipooH?A&PP!;tO?hoJ`r8q7WMxx??S#{lx3
z{-$F#$MWV=Vvd9)aO)@)@R6e+9#b0LdD4HnWe%ZhkX%;kja3BM-Sx1|z$08t!}uk@
z&|}))3)jP=syX1Be=JBujsc0UxxSye*ZhBQjQc-u3=<UZk4p^-Z6;*}PrU%OHIwo}
zW--1GfS$j^e}K^*FZZVkD12xWDRUorpX@%bZ}&TbC=>{@j)f7MAJ^wSpV>kz)lLwn
zxc+M&Z$}TWeOnUM6@lfXaE=vd05~ad1(CT`)+<3VxNDXCBumk5NE?60&i4~CH~Y@F
zr|N*fB(R*RTB;P<IIuXI7?Xex_7Q~G&3K&!+77M^FHn99sVtCMGfGk3zGzA^Sowm?
z(-I}An6LRwp?kLD*~!l*JR%cqY;v*6xEOULa{+n#be=mO!ff5H!L7%ryD1O45-)cp
zW%Y>G4)yqewQ}%i?++yl2Y@UNYleLYDHUv)$c{*N0ubOvd#Tqm(LQ_N*otUw@X(6-
zJGDJ;lt^xcrvpd;?eWuVXv=Pwa+{_PGZ~<(NHb8r26k<fz-k4HC5Urfiff(V4M2#x
ziE99yc{1yIkgCh(;F2*|@}o}08w@Wd)Yx*P2rsQISi`r|=EKzLig5-zol2=v{QN*y
z1O3us85%*LWW%gv;Xrwq%Mf8L@9|wCtXu8&vt+uQeFypt4>GdrL@z&)*TePYcN5u8
zMS8BKiiwIRUbhut36K;qALSD7p^lwejUBJd3`Orvq*!$WP}pZJ%*N%-d(KiFaP9&%
z)upmp;Aa|ExzUk-Pj^TGw>u<KQa&!Gkw_I?3hn2SvJJJ`1FVFssfJ-Bl8*3L+c0=X
z>isVbYk?yL&=KG`OqAZ5d~t3e15{gtGfWI>@P+jR>Oh$A#3UnL7k-GtU;SZ;OFumy
zEe18kQ=Y^pzJU<AzY-ST8LNHuU+N5tFC67oNa`xVyiBf!48PGOiV<#6v6ETsB>LP~
zP$=>4n<x$w-Lnlm3)_na%kfX4o$Sr_mldPJfzzyc)?lY8euc@yEdVTh6SUJw`zRqz
zGk3;t(jP$riCrAu*8Yl9iP(J4x1_CF<hN`p3-S*^m(-8z_)b*Z{_BD6WC}-0FZRX@
ztzmv#%MHmzos>ufXs-%3mhgkZy9}Lj&8=${1L)^sbxOG2&+{S3eL1X6**gBG>Cg5n
zKaiI>|FAEGO^sf2bA8dBl4kU|SH+bm4?ygK)xhRUQ7;C=m|&cLy<w$djq$18TBx@Z
zChDEuIDd$=tcKSUyGrA>A)MC4a_w%m^mi6f!lglwFYfR-a++d4v0FvCU3!Qvi^dA0
zxo=Whp7v3gB4xj1liifvl|R4Y`zP_9|CcO*Z}eS-@m>TE-A~BlzVV&+@q64VR<7wL
zWFY%tj7pGk6wBive0j`<u^ubltH0qjdDY|-#DDXra=ceA(p28^hwtmEu$TS%(a-Qe
zPqr;bjpU;ph@@W%nBgO_`CQVJNa(KLr|V*5@%5*Uo8|6j(McK<M;R+D$ye79fcJ3%
z9yT&rlmJ&ViM0dtCSxXQvwy!0-N^4AYK0vqgc57XG<0(z=X|#ZPl~SMkpt0;)J59;
z^tNP`E5ZM!e3YM0vQdk&b|(xn$tM4Edt466lGJ1s(M#g%F_B<fRipofYHXU0H@&a#
z#Vsve!!*Rb;audl=qObH(TJtu0~9GZXvUB43;Ne~GA_N8ghF9z%Qm*$G^U`?nVeNn
zEOQo3ac=S`q=b^l!;7o^hgz{G+!c-Q6!H~67|Z!2Iq9O&Mk$+6*?I7LxnRKX48C%I
zqZ<j?p}fJjw-;#Nncs(jo4`d)+4l`tx!ZEEZ25%V0gMNqISXtaXotQCRA5<&ycT7K
z8MNt=n6c-4Rxs-MFJGJ1sR3_pcGT#?UO~UxiwTf7Auh}B^7tf!+8EiD;HwPJdot)a
z=wlqbhba{X8Y;u+^YlmE(9%a$4v0+C9~eU&d3SDMg^o*Wl<CxE063h?dx6duAxx2F
zql>htAL7a92}Z72V$=FFZ@3xy$x3Xvdk%);A}rg^`c$IDn`RKaGFhPaPEviKu{oX1
z%bD?(OzV`3bQW31V^`ZDbE8|xVPSeit#mniM{c-hkgFU{YV5DB5$P!Z<{4I4hAu?c
z1Kd2AO@}mQ&iV!pX9eHrgV`YoMWzt`%{#-Rn=w1Y&pTG@jSvT+c&pCp-G(UN@b{pf
z2kJMCI||r&^5A+ROCVOg&d(%Zby0t)srnAJ5o^gq{lUb&H7;J43}2|>uWZ`Ba54L#
zM4Q3PfXjtMI+9<e!;K4~9e2OKIjO6fRQ#5<Z@7aM@Q{zmq*ohFsi;@oXXzNArd7_r
zUJ`D<>va%_wt{|_rhp!i;BJ9uO2BD`n%gfP+M@oozQvTt46#wUc(e%)q>VTKQv60M
zRPC>&Gh%9Vm|UuWzi;Iir6)|sx3<WSyDzr6hp`M;w5h#uMcf#Xz9<e1YP#q6BX`zT
zT!N(NZ%p`#t$GA#8<4R*y}x_kS5Iyz;N6rhjoKW;61e=Qn?lC%JNvRLDUHrjf2q{0
z+A;qqGw01&yiGHB(jqwss76)-zVDI{gJnaTe@(cl1#@M$3Ew5pJI2g5ozQj9-g)=X
zsK$Ne!pIZq0Ltd5oW{@Z9<pZt)+Mc<p=_?9uB%uKiC~J9h9=r$t7Q@N>JiSGY!?`8
z%eFHC)?(RPWMPQiG$&U&WBWA4v9sqOItQRT9xSOrg5Lq{%1)S4pN>*S`7PJ~aZ;AQ
z>D&Mpj|9+hq+(ER9$K?i<F2u!<R>%{3g?%C0(wX(Al8#0KqWn-B%~*k_TEp5;a5tu
zs1@^LK|6D*CxcsTmF6#z%0@YUri7s59#UeHxfS#pTROG|>0R0DmpRR`38ng^1liW`
zw{Gqx#y%3CSF($2HRr}Y{h8R>wQEql=-LS~bxKYFc8c84Gs#DiTeo=0xbOK5hQq%=
zs1h*EQF)S&M6Tm!=!wASb(etl2*1>yAgf+dQY`59Dg~UAyopgOg)BL^x=@3!dkh82
z>LukSDlR@e94{$xl-6S$*rZ9-*w!5;&p-u_$ni?gSOeYll9J<C|2mxGyD)LsBkOmF
zN&8fmdK2unVY&SOeGU((1{D(Xi^tkCM@j{nHbQ^|`Sp>efq8(4`biCeFgVvI8RSS3
z#5o^m9k*y6QO2scf>s11JQ8dAxg>_oYhD}{NcfTxaw9JM7UxwVb?2g>5a{-+OkvgD
zoDwnUhyzk`AhAdty53Fip9PxsK{&-$cId!HyZm+%F9ZQf6D)ld#GpnYoM2xGpdoB-
z=Xa>cuJ2I8I$sKZ@Zw*zGy(L#_z~ZOxVn{}%UOGdG4=l`82+UHW>cnV75D?;9yN(l
zKnE61$X=-(v2LwfBmU0vnbz)F6)x5Y2d61Q?M?$$<V?Naw=?Z@m*<Wf8KIb=8e+&?
z5MKf7%omb-1G`Pc`(|W|xa~+}3i+%a%Vx`>y~LnMINNUgzazktXMts+{S0z4OZfH*
zgQ4bZ+yBK+oITIvhuo=!Y2fm%?FaxjdMuE(HT3h;S(A@F7r2+h^f?Y*2R9;hkr{RV
zvf7)8{~T{l7MC+nZllU^Miu;kl?FKI>XxfU>ctAnJKoiRVSv0fhpeCAuj;+>=@_eU
z-5dc!39f!%R9E@QNS4k(l%xe0)Q3(_RCDNEr2BuXQeF*@n2baEmsg@tA&>d46u*qN
zohyshcAJd+r76eGulu~$WuE-dQP)%P^h=mSf3uz9-`WoZYA@kH`M^9CHSg|0n%H?2
z43bWc5po=E{e^3)WHo!5Vf~PzpWg4On#b#JNdeZ)V}F4%cDm-6Pv=J@9_VL6W?lu0
zLX%ToYYFxf{@NgX_NQ^OmmCP;grQDlJJKSH&H93^RJbN4F8DX(=_xv40TiEteWAVP
ze0v0aL$HmU-Q*tg=o#j-*4)wUGF2kU;Ify&J9^pdl7B*SRuI{NM1#EsNP3rFRXoXl
z2M8Z*5ww$F5@lHMvW+gmVP;_XO|D^xAC2Qx6ZSj9K12CDDXf>cO}L>UxZICn{;w%C
z&0=AVwD@f0knAK8o~{4z6(b@i1;lqB;@Up>qwmG@<TOM}Z;|fJMk3{67hyZMcEtXV
zy(#ICi5`&cXse9}JE?Fr#4XgpM;m;QK=UvDlOnBDHyP;D_CJ87T=F6)TYDv?gw*1H
z_(}oY&Od!$%kkc?(`bI3#!?Ug$!t)46rDx6I=Ar&>z}cdlfv1!&&SMk`4_&Tob-*?
zZ+~@HJ=ws_X9KP8eAcdvp!m!fd;kg}Vr`)H(j1OohO@EuE|QJOg*7J{NQB9(!X>r{
zbITXytU&iiylB~;%vfS#Y+9H?Uz~ADO7v;5X&kbH<SB|^%`|tuKy1u4Z3Y}*0iWuL
zV3Z}_-qt+(Dpq4p%ZDTIx_rsb?<TDmjhz*8g{+1wK*L?~lpn^Ks-!T(g;}L{%&lw{
zN<S5fI_mJE>NRFAidc;Y+%n(c4c^c!0z{Ig=Up6aZE$6N3z~N)NLS%w%x5*c&X7Z(
zbgwih>c9%zW1c&W8FDQV18AIu!#{dEBWu1r9UVPZHs8EPbT7hHmDCH3MSSd8_&`xG
zoGrL6Vg-xFE5dq`+gyh8*rYA${W+Dkr40<SO}}?KEUT)K3(iUa3Lb_R9yZ_N7iOS2
zld%j5Q!_=u!S6sFDqoH1!{ZSjeY*#*qi48!4+oXCvx#5v>d_TH+EEUqncTk-_sm*7
z4fmRKp;qjIxY&u0HQG~0S_YGfsUC3fH=)ba)UN1D6FlJ5^<pA|`kH6&n-B4UP?bT!
z;?0N3%sbMgXY=|%7qQDzL$CP9mcyi{!Z2-%KNOv*feQ$2%sb&)%O&6~xx!CYUcV>0
zz_&z1DZGA{fUXY4`(GThc7$;5YkrW8)BduuZl&^7Tx8~GcafVW<Dw6=6Hbj|pD@P-
z%ElhSs$W8l^ki+eoJmi#&}h2rX)Pq^J|&lx(QciGRlQmPjg|oT-d#w#l9g1E^kHU_
zp@ZgeVz?MZxAA{1eP{$C<_X}nzF}z+9bQGuV^-9nF|S4|jJG@aRi9Yh-PXU@;||?w
zQpQjVRqFU?D3}}HXgBLM+%=26+3hPgDb^+$5KuqSK-()99|c?4v-_fbv&)+B9cS0)
zP3}@!mA4!NavdaGw!U@zsEh#pVEm2&$K#S;zzCV^-LJr265~`n>jjW@{KZ_s%}jMg
zEUrVR3eUK=h)lG|ArE#Em}o}!+;@*2!Zn*z@~2}vS%8D%M}=fie@TPA%NQvd^ILy;
zQ9tv=>79xam;QJ0+ykwq#vkqIgDpXkT`v{x&RNuzKotZFzl<L%0Hb9cK4b!jZ7*5}
zgdg&6eRfxcFDso{ndiGFJu@E+#{cZihM7hyh+9-0p<+~ASmHtpZ5%W(V9@H^6l-=L
zM_naTq_}@BpKDk$-p%7;r^~L@b-0nC@MEn#ipnOSd%RHlpEkjQ-0f6Jd8pltM>Vtz
z6RvEUbZ?DT4?hOtxho@6Ml068W52E}OIy_*rn-QaTd{^4xg;R0SM%9c?FT?`?d&Zy
z*yZ4rmFeYZ+JT#BG|wV|jY8>tqak?{_l2KOg({xr1FP@c>bl9=`-I5oDEIqc2?pV=
zlOlm88foD`w_~JqK&SPuA)L!Ed&_ZD$S1D7>j*b%bE>AsPE#l8G@Vq@-zTRy3B?QO
z=miT*84wi>Kf$&`qDh6M-L?Q5-h6~x$Rv8PyjA{C_N`6!L{|_KK;c_Ca{x*<p~&My
z-s@X&gX4PCx}^0_yG$TV=V}p5U^oAEKT;+!U(cSHpxt32?;7n`HuWX(?W-3G64&;j
zhCGKjwbVLF=G^_s*jiD*q_{Pf?n?^OO&6%CuM36;k|15=sI^KE`2-BKI8I7o5Do3i
zKw#_6a>t8X5!#mZ`S!LM1^{}cpWWxD!_da|v=e^J1BxxqWCbA#VK&%hY!!$5tSu~Q
z9mkY&vKT2{J4~A}=6q!v;X+>Tc<Wa!hUrDNoHz>lD`sHdeWtT2L7Ee!v`X(Y?#%;4
z#1|n;#W-PzsYQY=$dW5{ZZr{@w2H-(!3_Q4eRc3t%3ewNFsk`g7`*27BH|4E(e4^D
z@wKju)LnC<r^}jS*Vf5zSzX5$%sx+`xd~D#%Aq0zCzV1e{5MD3dMU@K7j=gzdRI`A
z(`ZHm<3#Hw{Zza<wjPLhlJo}{KS*_wR21keH*CD?^a2+>XX_JAXTL1z9qi;dh9`Le
z0l16=^LS<&lb8v%JE)E6fpDEj3E*2hUG_WRR*0;jU!X>YV&x#wy<B*vfTx<@2_V=W
zJ27ZR<~`?j%&4Vl^3M<-kdZe&3;bwisYcvHr_>HOf^s<B{_+Z?k&UV{mOx90S_WbR
zaz0un0iOteaBlWa{|x<0<ci<wgLln@U<w{whI=pKz1(Hq)?LTC{NtzMGupie?>b2j
zJF)my9FbO1WK*?IZ2k{R>moQB{%Q372)*}okJvF0%eQc(&fielSk59MXIcTv#{^XY
ziTe*y=+x^jiDNFX=7vaJ=)3A*6BRIbS^+N#9wYiSIW(27&NY2n%;~4&d<_LWxnrV0
z&2~Mw!D_VMA_L}&<;)F?2?djRs|mR}Ku(xX7Z0fgo%(nf7puestnwS24m8gk?l2U=
zE;vRNgp*;*Cfs*YhM+_GZpaEF2D}>O@R*0c+Quw6Hr$)zqkp*QnY-)TlEZ+kbdi+w
zYJV{z)M9PN$0-Tqx!7-8DA_ld%#5z2VHg75;gi9z1_s9yI5qs11y~dqQ}EPMG&eBU
zJ*BrLbI9B!ees_xo8J?@=wakVTG(<m#?1|yA#>RBn<8sXWfN}QQWN4;7gph$$AuLg
zK4nu#5<Ky`C!2O_5mIeFN>)G&z=ako>koBFqI9l?a6r({v0qb&<^cyjwYUvE*))Hi
zlLVOTI*e1RI(s%<U4E0GC8jwYK}!~SMeIc+=m6HLmxs`1*5Cnb0<UMsQ#>(saBhzE
z_iMFyc{8S$_}gO=bjip}MRQbMEktd~$`i+EZMBd0X`e@aXQ@)WSGmCL%K8~!sn}CB
z47?{3Qz-k1u{mdAR{_!Ll=rHpYp$#bxRVB&K^cmt6h`x{+x_!AiO-@X?e&&<g4q*d
zDXB0|b9+s+ib{~rgO*rphE&!7kDu!TBh}A?qWCYx$**U3au_5giOsYE<*Mfu*bgc3
zB9MJ#koxGF`g%9&9ss=PVp%q@M;gYwX%R}`1qt6-Do|jrBhNiAAMl(Ytz*xc3thw{
zKQn({Bv(6ZCd}y2a2(v@=kMOiLKQ&iUR0>#$=T0`gKr+p>ti9JR8|w>VxZ1DDJb;J
z-b1HEmk!%qi?;-458;8q>7aMeD9RNu$9ZVRB(gVGpncD@<ObFjB`ix6tj=nu=xk$A
z-?K}NT}}JG!A_Mh@)g`+UnQVVU8x#R<Ca_n5p?-Z@ffIgQ$R<mo0pjLm~vR=(NC{}
zC|V#{gA+25p>IGGb$Y8bOkrw*vn_QjBVlpVs@k)+M*_Kzo>)|=m^d}cYy3e{djAeA
zxZvY;NW+@wf`D*8w}1X<H#Y1QYxYOo%3(o^<ATSSY11{Fy$sk+bdg$d*VhE+q}h`T
zsQoFqN+a|0@?9TcLoIVZXq#7CVljO4RZu}oCe+XCvu65ro>t^VJcHb0OQ{G}YAt-$
z`yPRr7kVO2ZQVi5cIGx}&Oz{2ek1Z2Q_CZyRb0|eaR~?;#xLt+_sd}9G&Z3c5pF3M
z_Z;wxR>UKKPE2yWSb6{WqEh)AuC{YmwlrEQzACNAu6QE3hx+z(vBk?%j$<NuK9^))
zP`^iFE!X>`XRq=Hxy1|(!cK-XNcj)Mu(Th)(T@L>Yx0i=X9^>cHzz7QAx(lR8g*{`
zwiWe@B`<j7j~mw~5+@2CE>Mm!IVOnZH!&t?*q(<b-T(**I-Mg;1X8!IUl%Reqr@d9
zkuN{7ZVrA5v=a-Eh+~usgK@v#xQ3o;`6&`p=ZEyj0zY`Y`MkL7R$qo7o+Y%t{8LF{
z_BrIzJaa<B-Z9SSO^V1o5m(G?7IiI*z7+f4IFRVKfE~pZxvj42DPF4{hB=KU$Px2q
z)1u1b92@t9F1ts(K*71YuBa);pv!}=HzXpThn4F}-g=&0yX|n?sR%a`O<4;*K_K|}
zR+t<1#Qi;n9F!tRBDqpu{49^s^>ip#a`H~X8YTO<5tNttx4KAtl?cXHk+yv4M|V_4
zBB8sN$Y~aV8@}sF`i7O4tD=uuao(4didLlE!4xej30<No5Wbi`c2YfM=7g*=zXL6q
zx?)V^BzbDw1qa#4pZZoTGe6P1g>t&cHM&)cAwxUdc})9axYnk~rC}J;{sl7Kx@OuU
z=sOEyP0S&*x@3v<HXZMmJ6kck&85f+#D?C?VA%F~ki{6)?MV2=g>Igx%So6h0%boE
ziPJZm2?PooAnTtU7M%dw$pthj$~#&w=4&Z<t{_gZzOM${*#8z`uB}yNhT}qCU7MRq
za<aTj$HuzYWrPXmty-w`Sc^>mUQ(-*!<|tjvEvYtZ?OqH+U?;Uj^VkSuj9~IfmYtI
zB66o#7EB|-dJo83?KV#OL#ZEsSu;0{ZvuwpQUy5*C_Ple!`&|9S&pD5?ITZ|LSJAA
zKXft#vQUj|)zN?EL~S8_kgkyU&@+ErH^?QGsI4+NP-?$C>9roNk1F}H5|$x}F-}cb
zqZC@ZkSMbMF}WDNtrK?JL9J3am;YA~qt_wA&BF0-nWIjz^+$1Xi1iaKY;_`?rFy<%
zSE2I-knV{m_0WrpQWhwgWMc||5|%{D%d*>-45BGlq5$y{)!d6VVldx82t+G!A0j#!
zm7D9273e5g${p1i31*%>4pC>c%2k|J7F?||S-Fm^l&DVrXiaZgsk?yX{LVU*t~fd7
z%0+0|onKos6&p+_1s|bPvhZ-A1;d$08}`EwSYole-p4>Rq6K-y4+?U6T0SYD58aM!
z2<^Z4&T~WwA!qEbL&H(<`>(nW@djM*n$HM@ul32?oMhGW?~=_y1-<0$a;?){mIv_a
z`V-@pS-|Lm8bc``ej5IeYYNH|rexV4`;Ph<rJADX*Pl#0AbvG#k6e3#QCV$#>BFCP
zKrG!XYU{=kz8rgk-<9&K61wY+#LM-0<qMOps+!9d5}K|S^6%q|$+JzWBVIH}o*j}_
zuQ2CoD9C}NF&70@0LGeG^Hu$SoFr?tRu`qjL*LR<T>Rg}E|b1?MSkC33t?`E8OduW
z#p016f;ycMOQM{eW-*V`qG%6)eO({`Xch)_sky2{y<hgLmq6Iv`0lA<b+?4PB%FI7
z-?1zSiXQR)^>yDetRIWXVJ7p{=KB1&DYJHAit<MvTa{Z;^?uyrgu5>JC_7-y$It9C
zFwSs(^{F7b$9GTipBy00<JD`309yZko5;Gz+s-3Nh5hGCRIB2@2=Yb4&gpnoAgIAA
zZ`z`RbKLBOz`Q82HIkA#D3<%>5&KIi|C`nRU$IMUSAJOf2|f3p6(t^e!adukfU7YL
z8fHoKEw*33m_K2D=3Sz7M46cfvbh{q{;20<En|mzsz@<PAqB0hB+1ui(EbiN=h0I?
z6whuPiy5+Z>n1WblBcQ!?LOTZ0b@q#Yk6#r3VvG;)tUN6>$q@C2nW>4JPukBp0!)9
z>DsHgj4(=-X>LL92T0@95k+_bTo2U|%K-ZU8GYPUrv?vEz)e@Rl?5W}-@gY+Z@zP3
zof<9$M-`QJ+qCYzpx>*GnG4in7m@mlm90^PtHDt-o|gwbJ`W3cd6;1H05ww~B+oMk
z5K!_1^GBCydM>?Pu!<`kiwK2X<c;m}NPOtQCrYFAif|M&q6FhCqtlX<c3h{~-`|@{
zw6dYe3S~+H;alzm-tO(Rc_=Hc<ReFOHdUi$L$W7#|E{Swgo`p+95=M(5P0Adjq7bB
z;K=t8OVU&n<Cl|>`-JS>0jbEz>yWKb$cuj$9ZH`646IelUi2i*Aq>ra|0P*IcCEZD
zJl;0sX~r)B-#R@3essMtap(wLAl5X2^h)aOmqTTRncjt<l3nqolSO_NJ6huLH0(85
zaib|kt~x}+wE?>bYwoFZ*=VW=f&>E~FE<9$@iprW7TH$NG#hx30D&z;>)dV@g#O_L
z$VKYnW`yG}mj-t8nRoIKX|67SJrxznl{=r9Y@}X#9jFlx(wKF88|J;T2=tZ{dPGvh
zz#jD#0bKpYKF2TOI&J7_PDh+!$w{O^^O`(X*Gtg5*<eD^nSm+(+%Jy;mDc>8!&ZNK
z{ck4y>CA)mudz9_KzGR#H&W3u4P(QQAr2QNz1JV0p29JX0xzt$Pd~)-5)+waKSa4|
z#D1Y8(tXG6o%5zGlC?#KI9$<ZIk?kHIwMR9ve$NpHg1*UppCp%;Z#G0i{UGAu!X&9
zg{l_>I7ZBWo|q6qTHleW{f$`d1wZF_e#~k(U<_<tGfR}-V5!G32H&Mch)0j<^Ko>M
zJmK`mk(B*!?+L;FRkq1Ou`Qbd*GZkj|D%G~@?n4wEyzZ})|-|r$W#y@<M{MV{;#wv
z$w6a)emRp3gJm@lyO1@3167QuW7-m<v$om}C#$NC{H*&Z0<kqmsKjI|cp{7h*Eky?
z(4WiSe3##5Q%*w;%Z2z;oz>(398>NMzqJRco?ZNoR`;<^<2#7j_+0-D?AOI7JH)9^
zL@-qGw~E<If_nd4(1`-Y_>Ejb&;;B6@K~}8!m{LDhpfO9p%dvD0im9Vm!)sY@jT>N
z9XT#y^jS0<QsN1vDP%Q?986y@Y<Tl|eb<nR(A%loQbM%4ZTIO01VkG0^tQ3xzYq49
zIa#u|MD9Z@+fts1+6T1uSBYK}ls*DE9or`?Dljs`EWcljSTE{O?1<~F;LzYkdWNZ&
z@vIG$xfh(bFQP|p0~=ziF{*z2L$4zI+x9!h7^zAwyMBdaJ}&wSCNePNO*pc(pBe0E
zq!7{op&*j~xq_0U$DYEF+Q)jH(kB)|*BqfM&q)&AwajQ3I?K0L!=m8)A4dJRQwemt
zdpNp$M5JTF0$OvSgB7qOUkw4u)?Eh4hug6DvJ0J59dTr_OyKXco!>@{OiAPTRl-yd
zXLsZ&zaKW#k?gXh)KvIGpBgzEt8H{;lzubjs~#U)2@eqBPD{^$16Prg&#XPNn~Aty
z{WqZTZB|>cSw2$@i`e^399Qe&itp$Od_xAuol?H34C0PK5WKnB8%55yKmtr_&T%CQ
z1tnjo^*}K*lfbFH1mFhuC#v&?r+WKvsFr!)U8xrFuksRIIc<GQi*H976hsJ3b-T}J
zF!WBpIBnDt#>DzG@Vs=iGw?s4r6(>Wl#byOz<PfLw?-FDc{F?%I?W0A*)=(Ft7Xd3
z0NbP?uI3uat&S~SkVN|&FpIjOmTNT)@yuR08V&hp?Ex6N_T$drHTbN_qu5MN!9Ga%
zftJ#YabO8^VaQ^;=E6eg&xlb9ACy_AFE=uUlY%PYZ9;b5fP!YyB58u#XYAKzNr=K_
z5HmqMr`}3?>SY>qf=5SN|KN%1^l<-U*T%apnq2GIuxk4Deq0KZFeTb#FJk7wPkhdU
z?s%L*iWykH7M~CA5SFL)5{bDGcfuz6YYzdk1`J-cnnDz57hFv4v>JV*TZ#U)7O$a}
zxf=7WOv$rWm4c~8wW%|)K>Wc}#mZFQU<6-p{63RmAykIp#|_D$gq0-fZYD(q8c2L~
zI4g9dP@4C59U11n#nj4ts7#&ZeidEPIMWq;0}Jq9d$(Yu_QMQJRZefRE&@F5xGMbe
z-PCD#S(zC<Jv@BAg+GOrk1_D1tB5o>&6=1ZKbrgQ-=v~SN`<r}Eb*O*Bv0(2_^|~f
z2WHZx&NeMfbmUdyK;Bu<zsYDWJiBRJ{4L)}D1#EjX&Q^*CL{IJ#Jpj42SBQ@A>Cy`
zYb9_rflF_61G9vOp1i7wRykw|^*oTA%j|kdF{0;hKFP{wey^zeVDb#dW0{b@c~I6P
z$s~8g$vO6EV@KY!Rec5p-+gAIttZT0ipnx@4Up`k*=$&~v4Bc1F^G`p26LVb=_Imp
zixPcn$88^yIo0s{NBIq|Rs4|erg1$$+Y-3pGkzYyXhD7!a5#N{E0>Lhe+QTe)GP^`
z_%CXc!Ffn37pSWnc=;2<wSbC~y)iqi8cKULfTM>s`r%HVV@J)(h}{?=)nN7;$_o(l
zn*`HpJBHj-V(RkvvOkqb;m(_IYK&HP!`FKHPw(1{nz2*|z2bn6j=e-uhs&P;r0CQ6
z>cq-1>1wimzT-<jWXM<`wbmc4qM!*c#`hYoK~P9>kMuQApWtfr)7jS4I$?_Q2nbfk
zk~8qXPbwHI$Z-)01H`PwNdr>zCd&ZIv~K#mKAnh}7a(m-9DiAP9#r9!G(r^l*GydM
z>(3&DiyQg|D+=E}ex@=r3qGDBRH~7^4|EBt`3j#v9usbPgTLK#ZM^j`1C0zLa-NH@
z-c$P<{$i$ma($Y5@l)~OQfUGtRAa~Y!N~$7f=+%(u-$Zrh1E2+VPMFHMUVhD^)lq^
z8=ZGT1vZqBE~UUVdd2Pflz+a*2PVfji*r8XIUg^&Or3ki(ewK;okz+A%RNpKQnb=x
zpo$M{2zD<AQ-)%GzJHfdH2D6Qq2ceUwQD!X$zWK)7V$|aqfy21a@mh=S44SXX48ir
zI{NeWyG*vqV#rs}=YK)q7`m2Uu#1i&E9NrK0<wsWWfJIT@WN@eY05nKpH@|;P+9(K
zxex1AR;XQd;mU!*lF-WXL2##Nw;Q1mPE29Hh5<l4S7Rsm3ee~wgy+f093@tIZ5xvs
zP3ojek2^8@!TVDJkc4l7%dK(1z#&ZkS6gYRIz*|ktsclI-5v$3*BD!1p}|RppGrs;
zP*2NJwyz4#&TwR!63!5_a~C*WAK!PMWOJ}wdY{L<w3HTA$r`|bL{`dgSaF#pAhdKU
z$7wZ5Nloe;VT4J0B>cm)v5cKssHY8Y5E^Bxf;+PyKDkJ8nxCkPM+Eb2$OpOpkhQ-v
zh<kh*4c4FhoPZoy=FbJQZZ}a^u6Cx_uNhn5!x+3A$rxF}Vr@j=$>WY<C)$6q&Gj=B
z$OWi%6z$i_W<<kjQjA#-9<CK$!M5ALp3Sx%{Dz$V_P2b`89DrilYkz*VtUWb##3E7
z7#2_RUDl${xVn~K?775!?szn~fi`heJ|P)4Eu29H)+{D)?eFZg>Z%_ZnQa9U6d(zk
zPsO_C=X#)?ERR;!#r}8COg%ki42q*!`WFQsY~mb{*|*7DQ0;r4j|Uj{HNN{4d`Ixo
z5AVjuzaM*fo2OaF9d>+xhncaQ5Er6$^B}#65vY{-6RdZA;``03pO<oXdMNHyk;1~b
z?bHl&))X+n&*RAg^#`EmyeG55ogR?;q0kG*@G7%e#QxjCx#gyHb8qs4{$ZxG5Jqbd
zJ&IM@;VDK3cZBg3=RhV#6vqpjusY45M=y*U+dwW5Ah-#EX>K;C`4h@ygDe%0H@|^h
zx@RO<rwUM9ytAxpsZ^Z6bM3C{(7dw`$~((jy|@9IWkwm&;{CLxZ$0(#7^QDDJ~{F=
zk5}W{C*!P~o}jh4GWTpe-kP-X3A}KbYHVM;DDa|>>90*byRz$JMK?BPt=CbjaSC#h
zx28xRe~N&=CWF};deV{hiEs=sTUS`0;^h`o1*}MeS@5hyM#&d?i-(+YIov%}=S5o-
z8BYT3%5yb5SEUUtQk1=2IkJ>?9+1-CoeJyPj3+CFd943I*`&BcC|UBZtpp+0U-j(T
ztlFFNdTO{l7-#hX1)7nv)AIk$l8q95gS5FFTVV7(kKR3fg-FiV)n_GocHZ!T_F>*~
zz7_B{efl`M6ePIb2MHVRK~(+u{^J{C6uSxVB<2A?{UYzWrqk5Kd7PL@w1`AndK_>z
zk-SQlaJxQl(D$2+LQP>Xw7nCe*Q8$KvM*E#g4!tdoKfqfZ1`&Vj*ovP{>H<!Dnz?;
zs>wO2Y+xsFU=I<WAhAY`90A(XvX>JX&Jx{Y-;rdlk1kV7azlsC>W!sJT&XolZ`BK!
zXmVJ|Q2ecL^6lugr7SinfoXkVcIdI{1&k<_c90&q65s+=DOMhiayrT!*NOnW{e-9U
znV7qkliMzO?~^6^YF8(c%^XaxI;h}$lF?&~L$#XKxb<&@ncc5STzdV^i~=p`Auj-#
zkZqluS8f_c#i`maOpo3SOt01X?5Gh)a+a>ZNXP{z$SO~c5Kl<nRxb6!NM=Z$L^q$l
ziU!<0lWw9{mB#_MQRBAp-v+XdpTaQjHw~|LaGk%(1CggWGZwMe6#lfAEc&fTmRb>w
zs4GS@syc}kM=UmCYwLa;#!w@v;hDr_OmTm)DUEn2DfuPpKu(4f`8db0!j@-1YlYUy
zfvz@PiEKf(nc7R(e%D;a^s(WPA^|s;!mGG1MuBr*w#^mBoBJ#u^{~%yn%z;HT3(j0
zR%EjvFBZCt$edZoGohrX0&HPAUfz^0ZT{9!(eOIk$#4UyRr)bk)^Gpcqsao8$7`x=
zOEfXa7I6gMj|xzSw5{7FoB6bWJ<g}m$L_S0!`9(3K3Y(7rdqj8Ul4Tr_SlS_DvK%k
zVg)uG?l;BlJt8bbT&C0=I4D`9cBQWQ-lIfjOC(8Zvxh`xkr$0;BM}p#?Z6u`QriS(
z5sy#d=GW1r^v8=iG}#wfucp7wS$~;mIv+}ph0(hCc^8N77nPqdd&>ua)W6igz4L0(
zXHituz)!w-vv3*2qEiR`h`jK>nM@_yi{mmQd<LGPPT=nXU*t%$6uaR11uAsX;Uwl{
z)-oz%-p4T{qKXPVm3gSA6^f4_cg2ONB_18b%ozZMw_Dk{82lNEGjRQpf8`~ar~Pcq
zi857Z;O;Q~V)!-d%rGh#RlZ3nkx$G)6sJcjlx{%2;N8uKDul{zMERYW3s`feFE=;m
zL9(x2O^=Ji9p=G9Ps#S8CG2b~LoieB9Ag%4R7(Fz4cuxQyEhZKceH-nszt_1;deXz
zJTR+YdSb4%BR~FM;Rk2s0Xe>MVuG%M$XbB_{MYlYqAs=6$fVOX%LM}Kc=h7*t9S4F
z8f7qLcRhOxExiv7v_C07J7@bqRhK^pPe04LPFqjpAGF5IPR*J`^20?{nV{@Jq4ff#
z$#JMD87}+vdGzS4f3`izF-E9Ba09XyDc5uVoEG*7j!#Rwaor0{BUn1aDEvSowi!*d
zA!NZ25KoVspq)!r3`h<RR<{4!6|#ks%|?TI^KrgNUSDbgDsa?5ZJGEzPUOzVs_9M6
zUwFvE{#B)wm9-mtst1iyTgp5KE-O4XOs|g+%Gty*G}^WC0jqXlFW<mPVSTGFK8~Fq
zH^#ntn%e~<993sj*~e>7m^-04{b}d9Hl}Yh_BpRltM7NE4cE>YCn6OU1@zO8l35q$
z@j;Rl+afsa-te%%s(44|6nb5p!S1GQ{b%o1b0l3M5Oumq#H1t#vOM1P`BC(xJ_mcn
zubXG<kC;#I=ReO|;pzTRc{aTlpep8PFfB#S)XA6ZcE5|Mt=2954+=vVqts0$rq3LE
z;KxiDK(HD!o{HA$G9yh2jUx;ja7kXXU53eOc`j)cWflSCkdaUpit}fG-n41VszOtX
zWu@(gs=eihk1Hoq63+E6EZTo!7o?x}&5DmoKOIW^eD9W0Gkh+6Rlmt&D`!g&(F%rr
zOCJ}_8v71yhGgk~HfF47(MKArqEOAIoNSOUvS13|!HsF=w<IWrQU}Nn@zcr%Y*YUD
zGX-Ly%qj(nM4l>{yZJm-R;;4_$|dvvD~CabeA_HjAf6`zgRZ2%WNO*biwZN4MJ0`N
z-V~vNJUg%Rcd`__l(w3+7vzE|bYRA*9<j#@-cqe^I~hkQ0rp{2;jSnJBizrermD%~
zzkH6zv1CyXFA~>ZPwKGN$6+2n%p*Ztd!}1U9*zQ}8SR*Q8<vZsNHLaTFa40!K4`!b
z*HxS7|56!P?UCn@;l4s|A#OzTWB5sz$SA%BMY3Xe|ElO}Hh_LcM!?Bin%$8PTRlo5
zqf^TjbL2AIxzrJ@7YFqbbFNA`U?a+ntDrZ0GRYXanBjtM*hds$pZm8Y53f|{)aAu!
zKP(N1tyo3hRM<xiKR}7~)|^}Fi8zY}<qx^mBV_4nKE|agjr(gpppFaUReKAaAn5|@
z2}(;vd8tW34<)V5)O`tw59!ngbgr#B$n@u%PLnqQ)?oE^vxAvlO4=<=P%YweSt4q_
zIIT~8@_n;agi-c+DTDWJb(+@lWvYc>W4<-eZ{A6+9sDq$C@C`_k1n8m=+(bqgs#~(
z4S@`|%1D-7TlO|P+oJoIe3PwXl<OMyWk$|>Z~O5(=Rcaa85(!%=c6+yowz)hkoGM}
z*(<uh>TIAb{@*aE7N|Z=)ec<OI)euFNk$EgpOgU99@EdpWK03y<b;BGS@O=Rzr*;z
zG93peXL)Lr<j{%J<v}94S{|%9GdT%h9gp!Ou5;^LdaOUj<#{QOdFNy_4@#}SN!9_t
z%z&WM80ae=FDZ=r&2w2bB-9?cxh5NlCg0ig!yj;-^;e2c3H*<sDF+>?<Y*f0wj&DH
zQR$8Wty8{BYfO-&DG7_T7AdW7MCD%~R_7GPy$EXqV|+L~xDqz+i!Bib9a3S&69Wdl
zPb{1GxkGA+suYJFtoR8BLXOP0{6{FHXqyt$1TbnIOUv{M{hN*xx<&BLcdyzCxX~AH
z>$OCi#@vHjXPHSOFKJGrx+qpjCg}uEpyr4&cKTR0`T^YAOlmb4XgHkikm45*Gp&rq
zl`Kf2Hyk%X4nf5t2&ZM(UoWbWW#;;WQ9E@9RKWoz6_lV~4j6t<fd`2}S{V;4GHTvj
zp_FUvsj|?}HmNZbWzpAK&5MNTLXm}%FUHzJ+me&u7OjnI{3KDC|L6UMuUOdpki-<u
z-HaWt`tuoNF4(z_<dGN$@;V1POAH=$%rGeuyGaBmC9s=?y{WpH<f=GJ(V;ZwwJyPS
z2d7Y%iD*R#97kHQ(a70u5G?#uh+r}ssx&B`%80t|D@o8%Y|Zdm*svX$!^qQ0qY=`z
zC{8qst#Uw!`WHtu`!)E|=0HnDw-*<|rzN(7?)3KVqLgo7A2GL~QRM<`kajydDL@zn
z0EVplXH9ak4aZASNbq6v2|}Qa>eGlz`&s3d`(Q|5;Pw|g32tdgM`sqQPG#sRhZ|GB
zlvB);ei?wD1b2Vm2<k4(%o*{@zZ&LH>G5Dt^uO}paM;ETo0@=ZP}>~;0Wq=`k6|9A
z51FV&JaBL;=T1Zu{Qw127!eg^%G62Fm%Ps24Jhb$7_YbT$8RkC7{U@Pa+SIEZ%V_0
z<eX0;H4|t}m%p=X#VgTLh>d&*m%IlHnS^Os1REgU85CBZi&V#qPvEj_fTOHzYC2QG
zZ)%05V{}tBg=QT(SGWv6QHbBR&>z>5E9CNPs7q(9*LMtj#B>8v9a(Z|@k8fnSqFf>
z1dbe8WhFXFzOLm?&{4BDjFff?29MOBhTnNZ#faEm%8I|8m~XsGqjVMZxLdyf@m^fA
zbK5>$1rM1j3A$q=xFHC%rkkP40eR@p1;nm_4wwr{qje;^@{{dMr;Y?koI3CfP#5(@
zRSi0RM$$52=*mE3!y_`W1`aJX8&)0{NDpYg*DGJo8>7nP-$h$NmkW9wb0&P6xZ^gc
z=z{$0TD!5am-<h7yR1iWxu+!wtE(ous%Np6B3x`i@j|?}ztgXLGRFtl3g#*nn$Uw8
zzH1b#!huul(WLA=r_fS&n!ipBxC#;2krLF<X|1>Xs$>VM7ZGZhIQAZ!_TttSuX866
z4IWtwC@a!J;uO@74WO&oeZHS_y}?&@iCB+5KhuA{m}RO*mzf_12D`X>f!Q+4j%pCr
zyoQb~BRCd;zH7+%3v5#s06OzXmT^^?CQGvihUFVR-Hr`mZHSTh(t^Oy6|(Hw#lTAC
zZx}&s!DK*Lk5HP*(zkTV5M%QZ2XkW1B`#(xp3cq9RfL6W-Sm#|#4du2uEea_vDriQ
zht7GenaxRy=+RU$V-C^-Y>Os>Jg%k`0XmLV8eDBpHPK|4K6iOlr~()dJS#SO)Nqu*
z5ahqsztqXQ1bvs?NM-u>5{I>N{*XuO3}{>pvo!!uS^2aT<(Izg6++oEjrDcx`QP;B
z5?>O4Fwv>0&C*3lm^M}bn@wb}^UCi_3S-!oYw9c~4lC-<RVgVyucL2=I^58oYhRBr
z!FS}d8j^Dqtn_G7U6U@Qy=5FtqL!q+R`41_1@~acjKSZwQAGaeF~QWE*$>ToK*7cA
zlHvl_ii7@*^RL{r95w3}(j^^U9bZ!8_H)lBcs^WG{|P`lo_cEm`})@n6j)798nACI
z>#U`U!9QUBmM_=8_0#}nQRcum$yh`gTi}_Q9&@(!bNd&zGBVTAYAvOP?c$%;zt!o>
zCnZ?jCJAZj%0%?vW-ED$sXyndE=&eEkl*}?#+2&~-VhsIWP2-tyz_9#ckX%hl!CO1
z9|&Xpo}ixp2bn--zcn%w0yH$2p$<0`m-QqfzkdpODRcVs{r7_;p{0l}{hMC%*X#S?
z`(akW$slxq<*#ptGyz5l6EtcjJSz6%@8!$k(@d2cC@3_Lbk5qrKpLr|&76zEq^V*c
zdzKPVA&bfYDhH?-pduFEHob97-m*d~N;U|C?C(+XG?pl#NdXGQ!~v42UbiWO7hrs&
zjF&SeBC>z-O6J+Zo%%*Ro6$^1EAVI#%)TFlN2Y1I{gW}S?EQXl$tnr)=#7)lzkM@y
zdHE+H9<BSda)YPJ>!~x#NNB?n1t2=RtyH1WOO0#_mPCLLrUrZ^L{BK6oYYpQmDTKs
z`jE0mp49|?=tS0*8D-A5I+N}8QQp!3=)w(_Q=WfG09it%=Bgxz4fIa5ZCTM|QoaH)
z@O@CFQ&mb9c**sB;}FQ9Dv+P))N#TDJa%JUaQJpn>f*|qGb~6nC=ryKHB8wxbMf0U
zrK5nC9+q?H`zU0HGgZvwKP_(n5tUxa>-$M(%h35|+xuqwdTyavv?hY$9Xi1qxVj?k
zna+O-$X{{@ul&UX)H|BB-h;LFybUDAhgainWiFcFFl^cylowuy9b2zFSvz4t|398Q
ztmOp<fnh$CYqMpwy&_I^r|sNF|IagRzkVV9l6<tuq2tNe<Cx-G{Abx^syYpw3eF}-
zNwB@%k+5~&npg}v8TYJ-#(uz>TKgm8Zcl$I?~-L#O@vLNOxIS=RK-4pM)l@w=g8ML
zPgeXZuj{rcr>Ypgjx(f7J~O@=V@4l4ep%nNzJtyr?FKrh8T9xzQ>o?if?}CeyeWw@
zo{~Ur2VzcwoHU7+4Pt8u6+vVY0*Vovz}4Wj`jy=k@;E6%F1h6nDmtvypjCjXJ2-#b
zm^lNPxV1S|Ue)A1sUsJ;ML--CD>GZzMz~Qu$nzM5*yx4M1>6*ejz@f8r#!@VU>V}t
z=XWAT`G9EEIzUs+(=gXmWC`~(bIMy-EKT&}2l*T^5~Fn9Rbr)b=aJ~|L}Ib2o+6Pn
z*oZ_nJDiny*j3D?tE^pnup<AeEy;h4OESc>syeZ(IN7CPfE%dj_C-=<ErIF6(1a*L
z`|nH~HMbK4hQ<s-$_Z@y7nY6Q^|rTp`%eC^7JM0FAajeuN?z0lK2s#(xpU-^qvghU
z4KcNalXi&tN|B#>mf&2PvMt>_uNK30?3X#R>XiU46x?X%<%j*V@10p?C_8_pk1zm&
z2*#{10QqQJp}2ca8f~CXNYiqMtNT3DkEqPq*!oAZabC5tM_0O)X0d;d*ZQ4vVBVK1
zl*B7H%jPaJ8Ks0)VYbZ&c!psjjQ8`dBDp)V7W&F?#g{5-)-CH@y|C88JSJ{+pMAJq
zX=m{PY88p?vHOpW6M`aVb!30_h;_V%lyqTH>f|C(PU71hXdGsbFyQ|X+q`Q4-|f((
zs9l(@nCTgv!Ka$0b?4AosG+{St4-s=P_Sj)HVcmUXx33Sp4RaqXFxEwfncRu(>4U1
zu2=90#?tlDYV`BleqZE3DC~<SyX#A@9U6A{Zo4FZzVCIV!_$uIHC=zI?DSNih0!a{
zd$_-cSSpAJEvpRuMXFk(V&|Dhn+@xVlFum1UKnQfb0#_s#iFiprOTN%PrOaFd0Ba8
z7QEEaJ)iRAQ?zRE@wI;whr9k=cIDDKdX`^`X6ENStM|x=lYTxB1LW4g+DS^HB)aPb
zNrR@YdR5u6kdk8q+Y*0tG09ddj#I)IJxe#H96ZYW%f=2eNhe}!D97Yfi@UG_kJ;Qm
z9mq{AkAZ9bB$ocbr%dZ4ZLKdX@4y2?qMR8_&&Ag7+$g2^HM4!(M%gw(@>ue$3qZ-d
zCafBsp5)ki8leO^rwLy6)*Ij*SV^}<HtgLVfP@F?mA)#EycU0Cs;CdL*4eafGk27J
zdVPfxdNkFPSjc1&W+SabPc6iH)ilhs>Uq(Z>Q=O8CeJHhVKuY&U@Exuf}BPjE3B)t
zUVD4Iu|G6D`4tswkL`}n6`jfFwocr8p`@PO_n`_rnnx<|;%rYY-+${n8Og-0gVY-b
zSva>F!Z>t%@$r9-IsX<r2Vs`tk?W7R>rK3^jD~p>eckaWqB=F)OwIFPv>x##bJSwc
zPe$50d1r-!_7h_3>m3WVR3`-BHF>bdwa|TZ$o9B%x;e}1_FR~E=6Q@f2Xg1Fp~^D;
zIzbAvE|yk300^|Qc2-4;X?>~mSv`_^sbJXMFYRMKR7OK;uhqIS<gE0n2dfq~Ot#lj
zVdw8NOPBxsR|@b!K#0?!Bl})9o_pzQ<cMfJYux>j#bQb8t)7_3aWVVt?_bQ-(VUl|
zV<HrnasLQwx0@{@8#b45@D>x7dcia(m+@mFGk>4J*9SoH!ygBJ`0{f8_VV+)(g(-9
z5l$Yy{eBSe1&4nHhlkc2oDj?^arpNA@avItY2kH-m*!k4cscm&;oZD`Qo_Rsae<e3
z{p7SbcH`hB-{EyZ8v51QBhH6sO2SLP%L1JsC6AO~oB&=zzy9{^mluvgQwgop4CoXI
zHGgn4z+b7z-vC_$ytR){ffk2%i_haT^D5&wZR40SONaOk(&9M#gtOxFw@Tl|r|~bD
zcvlLh(T=ksCGG00LL0=<r=D}9l_jW86!JZx942!U>Ri96V>xnjd8~e~l7%Aut8isx
z8iA6;85bNQqK&F1I*%&f=*I1YbK=}{;eV8@iqsPIOZT@QC^wnP+`+~;H}i}uVDh|`
zjC)x;Kgp9cEw)Hcb&(6L?9nUPCZdg;yq(szcAd3!TIW%&pWlr?cxHmLYH1HyrJ3`g
z$f*qS(ddH<@+s12kA6?`TC{~OYbDt*^WEONR#>Kjs(h1Oke@9&C2fn=MdM5Hdw;b}
z*(a0S-E&(<7f!x<0eziT#Cx}*FWssVv)-@59;xnN_s<y-Sg;Wk;@cPU*W0(3|4s|W
zD8)qZhkt&5iC<_Xn1c1=1akm!fByUM`^(=cRYV=mI80$C{sW)>{Nq5Ld__1799+V!
zHD+*$;NT~HFhMhE>rybN5LxJj>3`4fUMx~m;8^&IP?MiCMX(?ME}Cdx>|L3?Vl4tB
zQF(_;5$W7@Vrveu#1Mip98?%WVAQ4p*weIdk}304HAae(fP-8LBKCRuSkZ6pXXrCh
zjA6M_Q;pg)WH+2?z^*6CK2ntL-JbGoZAlDf1x#|7PJ#Prx~aglGtbm*q<^Sv1Ia|q
zW?f|YyR5lOx89Z@Z7*yA9i~MX)`K7cy#$FAvRXjjOE9hqU)n=>mo0qNb=a(0_-nJ4
zD#dTcRH+-=Cujy`^(ecIBRNS^@?7oP*chf#=qzE=PExrs;Y;Z!FJZXD|7@9l7Vf~2
zE(Xxcq&23nZGH*CDD@Q~7=P!M0FTj$CWjDyKiR-b%5^RtuADB#engU`Bo1;aSo*$3
zMBcP9DK)d0p0}qtjn^LhDn>6P&Y@2{h62DWu0`aLY!fQSRUc#OXxmPRzQ9?hR%%d=
zb!XVh@zfHs%&SU`@j`dWbW4*4i-=~G8a;7zdS(YpG;N0zEj=kKY=5v{P2qW`ubpkS
zlRz(e-gqWNi(o~g9xX+)XXXqzB$ufHTJ|V;dNm1?{5IA|PG|Dd&J4)}04JH1cNJK|
zsXnjs^+|8<^duF^9CfQjTOGeuj5LbA+bQX3!3pG^HA(Sh(nMOzqz;oN1BEPSD=ar{
zqqt=uOunc^a+c6z34eOVo%vqzRmx+rf$VKo3YM(*4syrlgYls+c&b(7{;YdaVVRhz
zmX*BXt#y9@%jkFS${sp5Y{f`7tyKwOw?)>>X)LBPO?m9_HCcdn(lv_}N%@hkr+ywk
zcLjHPlJXzld>!9;L^B)E6SW0k(OUVNKq-wD<E2=>G>jffMSp>7Hc?9+MYJ5F*`Rgx
z5Z^)CDbKL-5Fq!FR`;t_OT{eLgExNq^0%!84k;#6T2col!>*bg^0>~O!epa%4lbP_
zS7#(Qlrd^0i4c5waY5B@Y)ysBlGG)g=fnkRGxBqmYEg?cLO7Hsclkp9*PucE{@iJ>
zGN$t2!?#;5lYh=L3)yO~<4>PAT_$~C)@y2(RhMODednb7<g}ykB`YkJcf1Oo1!>n>
zG13HrjHPNU2pRX4%n@yzqsynE`4{({Y%L`m5!zm%StpLbJ%)dcnH!Wro+}S!gfiA!
zc!wlV<|=9No3E`|D=j!qtH};Lw}KtoB(;iTEBI8~%706(Ur0Wf;1e|(5z$`PM?_^5
zlfhBCb41*Zg0M)#>~q*+k6KZsiYQF-=`I%Z+dVYj5ew>At!h8Xc<HUyBX-5sx}toE
zAD6OTBu10et~?j$X{)<<C3J1&7_BwexFZWV3YB3q1L5+H@{nt|*3ao9k{Ggw)~f8k
z=`$xG=YMrlo@#PPaHds;9P(g!>^mlNrHUh#n)+gD>bJ2wm8uep4f(7A=2v}Xs)X8*
zGx;Q~rnzN&v;Bsd^)SI=Mkp}j%hklWqb)j&3auV9E33g72CJp629L7}zl9ZyNPa0*
zQN$qm)`i*JfL8igWOWnrC)_fxS|3|KsY)+~Mt^n<v<s{Jv>iK)V}L2O4>M(pRkoVV
zElcaKI^>w@krweLj`dIl>||<gH*mL8%<K-7^r<_F8;&GU!**C@`)IsCdD$BML0idS
zS0ebWeHx@gI*n^SRDn6wdTAlst07iBB=Oym%xv!-kwaStvdgQr?J9^?!~5EIT+NiY
zNPnjaMAn^pSaQV-j3%?XzJus{;De9P!xMp8^KKv2W95dhel<-U_wzcJwL0$JV<kXe
z2I|veB}h_&vfIZ>>BwlM{fiA{up+afH8unasbf#N{FBm$;YpV^2oacqteH@fd`Jy$
zq9{RgSO<h6CPvdqy^>5ol0iqmZ8oR3n}5R+QX$h0k%m&A+dZJ;ZaU4aPi=RHfEOU?
z=-ReSYvqD~UZA3em^@=X<SUdb*9tbeYb9~FwJrG2fa9(KVGgFlF#ra2k=i1Q^4K}y
zUrUp<jEy5pePC7&CTjTJKDE|PFcAW!T0v2eHgAF9BzeBE(UK&?Z`7%-D+g(~vwsg9
zwydsP3!B_K%Aq#cYd~s#%&U~PVu!$HO8a@~?A(#F^A-}!X$bE9n$J&8G|lzP=lYjU
z>Px#vBaS<12u;q$1YiXF<xnBJ8$F-Ll&DMB=pL_jN<1!OYx&*)AT-24ZKePq3^Ug4
zT`Z8MoG==M-QmgCtD=O|o@rC>!hdkErRB1_ZBkNy%f;30Gd$|yeHqc0Ecu(5((2Nt
zF!^&VB#~3+`#e7FX4N0dP)f#AO0pz$vO@#c4mfGVDOORA>Fx$h<OmdGuYEZZGq=Z8
z9hlG)7ZO+NXh`F&QIkyf9OqsyqvW1BZYftxQ>lzn$hbPm;C*1vu2XNfi+^VzYcydJ
zZ~4lfv>K|s$%Op<cD@dRjFc1K<$jZNmO1C1!PgrGuZi#SGLF5=c@@WYGOs)N4x`6o
z8e`mS^jMOvG~7&to+lVUNq|acZsD!sEkFGX-WmWks9Si)9nR&B2gKZ@D)<ALl4;jc
z*6=6e-TnA8a@oTlBGU)q&wt2u2Y<%FAAvtaydC@*M?Zo;gzpaiJkGb^58=CwKjYx%
z;SW_l9=vQ|5LB^JJi!PmR<CNd0e0lppa8++arOmN7Sunj^S2@wD4Wh`P;^aRIl)y-
zSi74L%Sq(IU7NpFQpq*5t=Fvdk4Y*=E_<L(WcnabAGz*;`Z)L_K!2Txw*%_q=tn@E
z@ZACR$N3geCw#X-eH{EeP$xp%-(*yAzlqtLemg~d*qk**eV!&5aIRCN<hze6QbN}A
z6w#V_?XK0N1iH<oww^tUpa9(59?(9RJrgcFKucx%7@#Fww}F-n{s5q*;%x&h8T|mz
zQoh?j`vkrSw3P2A(0`J_PXpRXi0$+hv**W+fM9kbdxWAVut!k2kw0e7+s=3_dlp_Y
z8~j4{td_D|v3>=v%Z`Vg`ktHQEw430bGPUskK~&vWAC+I;7e{|S2@mg1o@>0ziejA
zBL2c|?t-k}|GNvv2PS(kK1%dKC_XUV0r6q%M_~9UoZp7<VSn%=2tIN>Z3FnjTnm1W
zTrZpOJ&b)GdJ}WU3$~ki?Ze8Kw3cuVyN%b!IT!FJ&fZNNH*LR>H(Jh|7gOH&WV{)<
z>;Vpu>4Okw<hlbn<KU0L93tKh=!~NuK^?+(2X-FkTX2W)-3Fd<@bi#|2oX;|&fC!Q
zumLLe5Z-3BSbwPC^EmH<K7_a3J@wi4zy|)nd@p6c*M9{5j9m8chsgBg94r_X^u&d~
zkxs+@&#N1PX{L>!6!SIZyZaYXI3|Ue<ovw65v5^H&i=e6o-AGU5b!R#_r<#;0Coy^
zlB*;J_OAEkCP^XCOyCnY51$uqr-0Hfll=8Je)t!B`F~6NkpYw{|4T!^S2q0cpTj@m
zRh1pW9&jNN*32lgMVKETwD^Ez8;SUvcBmq5Cvl}@@lq;4_a<4`t-(;bn^$cF)PS(a
z0Z6^=(#P_9<ndm9|064ejGKaf4A<*(R3N`iF4fpu{vS2Q`t7y;@a~L5x|>fa1%!7y
z;M?+uT7O=WpBm(+q86uyh*fSxV#(S6q}TNSH&Ay6tNC}>Cw1lA?c(jF6QGC}{@E)B
z@}2mLHb6<<Zt}k6#9xC|h2$!Oc>JhqNf0r##OM+I_Uv=F9^2;e0>Vl7G3N0af`}c=
zb6Pb|$o-yV9+ORUi+S$WW7|9qGKL>x9H#&U{V-sh%c^lsdHp2gU}NkS<J_smwrMOx
zAt4?K7PnHkE<6HNBji2NaSFra{r}-<{sNbwV<HrnvAsGI12Q!>lTjZge|?b6YQr!P
zgztKaeE}=`BWYzYf+lH7dkgqhdMdOhQ}X_oj#Jy1QhFHgG8*m5AJZo>J%!^K4i`)0
zv1W0l@q+{hjIS`vlF3D4F`;py*MV6Jbc7PK8RY0IYHpoLMM6%Hw9b<ZyBwq!=(Nm8
z&C9Udhi3{2R4}Jk<#`fve}Y|-+uQgOczN_%AII?fvEm$!joznl3vjC0J+W(Xv<|q1
z)R$QAcIs>F%bGYmPgdys6|P(Ox-IM0ZQX9&KP1&_EUF|~qxlB1Z?32bKc60$Ga%;N
z^zPLerf)DIV3c#~lBVGY)O;_?nyrhQfid~)qA!19+u{2L4^L->m!V@K6ah1rAcz($
zhsbgwhsbgxx5#oN%08D<KRY9n%)1K$F*%n(LLMi7)mm+D8#xaCo?oGldmmOKO4Qq7
z5un>Py#iZoko0i93xb7{%(iYFPhd}OyZ`<k$}h2ZlGvW4$@N1JYD7^a^^l?@v$I(l
z#m*V!lKdx?O4};3RtTpkFbErzH9@{4<yg8HpScXeXP@PZQ9jci2vWhyC@_XlMnx4(
z1mQ7%13Mw2%3MYTm6LRFR7B_pl@vKD0<t3JG7y^hWFXx$bRwDAdr(0X_CZlJqLw$H
zL^l^mQ7YyH@Z}_W_yjYG9njc<QF@)SqPtjcU37cSd1aYI6WDIqi@un~a`r++Wuq$;
zWF%WCxF-eSi=><>Rz#;@M4=#jHdpitK98_}B1TMD?2g5wA|+-}p-`+~iS#DO3&mpz
z@}esg<i+CG1bHcm8zIW&SSYy0kj#-Vr7263;RYxzl%1K3-qImbGWa|#n!ywbETMQl
zqZwfWL#pBgM>s30rR*?A#%x)<!G@J5ECi^ayyZeUP@)E#q&&+oCcuPqp-6e}isT4?
z3a-O9NQpB(NQvSrCP+zT4lKLRER;1WnDWg6Z6+k?jVcg{JnMi!@rF1y$U=bv7-t$d
zR>2odflC0?Vh#%`3aZ#96_TKZ5gpO^GQucyiWGn(I+A%2B#|X-;7=6;lynXQLNOfc
z6oae;q^Ov@m;rCbEUsta%SkBUE&=O*La|DY3gtmbY%oM6m&>;jH)TXt$tSELODu(0
zD60}Q-TUdM>eUGpi-_u@>chYPqsRoE0F?+Yjju<e)BXKhxgxmWDN8!@R-6W$ej838
zP+OpGGbcIdibQk6sj8RFsF@sJ4bE%jWPqv;2W?wV#>&o`ufMi$j@v<7E1G?OIB3Q#
z<~YD2lL7Q(mIK5+QAu`aAf(8$n>7#_<N;1J_V=s9NppT&w<p!%tApx&{k5HI{Ql3Y
z8Z&>aq0qRk$L&;-+DzV2J#DTh=k;`!=j7A7`eHcvx%qlx&<q3wWP6J76O5FsHk<p!
zc-$00%(LnDPD(ZfeY%{qFjdEY*JrJ?ei@FxsOslta#2r;aC2Jyqk3DtJfT_Rs6KB`
zC`C<3H5RT%Y}=T$pv{uk-qC6GLP7d*^`?2>RIk(z*>gmqvLQx<n$O1a-}d)!mdd5&
zQXxcdLDCAt5+4auBZq9}q~x^ueiq%^&Z2wMXVHDUS#&OoKHCH5L|PDk`4&9f4gnaY
zXbu77J#CQ%ZFV}zR_dL0M)oGB5s{~RYmlm5EXAL<!)6R{{PXCo{QdB$ZLg+(ud0v3
z_S5y5K5s6o&)1h%KVH=Tt=?QunzNz)JOW(yR(!Utj?rntroGV78>0>4|50^(dHBog
z4|k}zYgm<BiakGM4SY*~O`#Dph_9?s8tfETy_G<5qCLR1Py!n;67k(ipxPkWuWja-
zupbAT*Mi&AHhaYy*DP+aFKLsv;*5lK-ip%(9Z_U`gJf`5{I8{nU6VoRCWFu`8Nl{s
zGVqJg_x(cOve8l>Ai50d37ZgA0179o1M`ET>i0=~F&TcTzf5F*G!_9~?;ZXYi9e&r
zTPTX+PDk5A(e}D+Cd0vKS`DtQM#J-g^iQj+=8t;v+w{+ATVGDsv=sCO+OVkwx@A-2
zEx0_f#yMgOP9$W_@v-2(w%u)seUdCj^G1-v(X9SJ!^ug>%o{k~SfY6Yhr?_aJtDx)
zdyoquZLWbFm`wD4y}~MW!zyeaRx3W=R0uy>dg<f4HFvEo%gTc7>buxYNzwH2UA!ki
zzIj&O+3L`>ntNI8x7iNYxE4pJq|+9BWKrxqfiu^#!2>*-x_C;x@RXhcPmdKvsf(x7
z3r}e)Jf&N3;;@T&ezsERzFf6!vg^MCWa|1q_wj!=-%%ES%f4fEcD=0Tr`q=($#k9r
zXwCGrh3RbUUrf&%>bez%MXbR9O~F5HPF^pgZbId+5p_p_1qXll_3SeWY0UZdvfK;j
zc$e$J0f)c5cLh4&liS<fS3vH8p9p-9l<nE%uIaIt=^Jj!RvDJZh0NNx@1-7K>cWJ2
z!GxY8N9}BXZ5J8T3mLQt8Pr7vYA-EMY~T87^qw$UOEL0{L-qkF#&XA-*7otHb<dq*
zssl!Y?YszLmPYIYG1m<!w?jZJ?~vHtTqw^~nLTFBi99EqSrD`!>_hIEER#;`m-kH8
z%cVyC|GQ^eD7v_3+F8*%?wOjac6d1${WiOA`hC=Y{P8%~%`TuS$zgU9D9?f7ZwH{^
z1j(I?ULV<WbRqbZ4NhA=x`*F3qb}3(ZcG%klLuMK*sRdDUS|_KN=)nGG#xe3!)GUX
zAb<l5Yq>tN(Q8+f!Ge0*nf(0I;0jpLlx~@No9BqAIm^Sb_VUi`qK?SE(P}b1;5-_P
zM?VgK#?zsm1AEw1hXRpz%|$(u3(0EQ4$ek(wRSNnkJI<>6pp(EYzlyMe$IRIwVk3o
z%ZCT-6!QsJlOj`}q{u5t(JfLSn|=u|L@lrTE(dK<YdiAV$j>m0*D0mlA#yorOVZHI
z<YUz@xgpRIkanpPlO#FruhU8ve)r=2;jjNpdVRFivuY=Gjn)(}sxAiYU>&(^x47$A
z=bGQ0ankadzM0cY=!fSkiPny7yYBSfbV1$cjLW>?jJwL<tkd3;{ESOGyFBCm56e8b
zmyt~@69P0emqFJT6PN9|B*uTi#ebiF`})HZPe}I|Y9krJ$ryTwl4ePJ{wR#3C-Uu#
z%6}7{>RO(f_(akwk}y#Dn=nz3M=3{7i3UaVd3yW^|63~h+!9P(V8)Rmj4bP`8Aj8X
z1npMdV8Y-v!Bz#f$Bj57KTmiE$OY$siRaX(n-(X>oEBQvGDV>ox2AvL*z?~$q$zmP
z%o((*nSw}6_EQYg3geV7Y06l0Pzq**=;ac4<MB6(T&O8>4Zpr8SEO>xs21PnA6K~|
zHE~1OBV3UL9fj+;5^!`qRjtOTm#ii6Q*(m2=~tcEgcc-eQC3kE|6dba?b7<vm0!)e
z-UemnIAk0XB$Mz^(Q<#h#Y=9I#Yaptj4-Z1PIQL#ycH>%Y%g0<#Dq_8lqrxZjZsO`
z0i$b{yq|Ah+Vz4is67-l_8;*m_P=KP{3UZF&jZXn@z`Jsv*vP<!i+-^+e>XZ#FW5Q
z*HIYxH`WBj8Hh>WV6qGPZ-;!z9|v!oXJ)_YDaR;{jAOFd^i+SwzW^^D&gG2!rRZ#i
zIq-uBuJaZV-X3p}pbor6!YSj<=UnYZKqZyQ&K7+L?UJTt_J(7X;23!h@*d2{XsFMh
z*FA`EOqiKOD^9IqC^b`^K5Th8C7G0P2uCd$SUuA^lAWmS76;xsdBNb}u347+vR~8D
zvVcy-mufFYFiU^B2qup=C5plF{2FaHKCf*zJ_+%IESTkS$*)ecY9AatCQoxNX$?$z
znYFLr(%3?Lvd3!U$}_8_C1Qj&<e}gU)i9LOkcbTn6lW`%O(g%y0-ib7umR<fwJFO>
zdAftRgv~W?86s~P6{*=p%YtVV)AXy%lyS2OxR^QCQkQ=fUD}pl6l#q55#E~y(^2@R
zmt0rHRhuB;^f$cG2n&^>cH;}Tdc%{~Xbw`3Re@WEuIPd?^|o7#qf>srPH?8-?c7)b
zXVgNt<{Gow;B;kJqWt2OmR+mgpBu<2*oV)_VS*YtOd%b!l=vv^7=SHq#na=d$g_<Z
z&eobqbj^S4XZ~T$Lkk0pylgY$uOC(yG2YwSb_&3Uk4mZMUy1zn^&B##Iutju0S*j3
zRR)<zNnwe5jU7!9j+R!MR?!qHtAyh!SxB14Ws4V41FP#G?Z6IXi<($p2P{*rW8Yss
zd=x@Ft5w$dtVs2;+F+d|)=^e;fOW)a=FlEu9qoU+0`-t2DEctoyUDUB)0e%%`yDvu
z^z@N__t;Q;8&V>r1k>6y7;NaHC8j#l&fG8ns~OUxW(v28mNRDC0+3uCdAMfwh7>_D
z<B8ME*nL6-%KIcq#7l;>dXHWTLB0c8ytWO1l6&N;#cg9D6wG=5F_0Ckb50WHP#lq!
zSIvJ;xMF<`rI@yGNIV*u)Nn{XK2tX4kbI0_Xw;Bxw6E+lvTAN`WNN>Wp=JW^GO~K+
zP9v+QfDpT_3`onoI@j|z<VajHKvN`zqqxt^9JyCd7v(A{i4ix<%rFU}#cSKl5EyZ!
zL2b**$W}}*t6kQ0fMaAV#uKxjW6Bq#>>+=SQFf<=8LAzk+Nb!{++Oi@QT*8~L0mUD
zoMmRSatJFa!f|S<UEj_zXbqwBc@0~jbqRV8fxm9d5|n}VlR}}SM9Y2bALHsK356z7
zg(-t51*lsdAwI__nZkOGSjWL_CRE<MR?sfxY0m2SdJ0=oRb4PN6&13?)SPt3SL=WM
z*O%_s+t;Um&m@ID#G&Zz@86#M3)=k{T5$r}ZWVdSKmU39{prUZY<-A9&fYMtLzq+v
z1UN}ao+O3VYO03L*%OhLG^_mc=O3r;$xqeU18Ff@=l}~oKB$nAS=WulLAL7ET3o9v
z5R0N7Le2qgoVHsqW&AVboMKL5k6nLReUwSX47oSD%xld_W|j75gNu1nK4vKeREY0F
zp=ZRCYoedjfHS}oFrMgdp49YbOellXyf_VkTjksZ=}|Z=XV4b$C0o<nTev^+@)V1@
zo*~j&G!@?x4*al$1xgC@MUNBQ7Ijca+5*UIx*0W&5ubz$_lqk3Pv&;>gm8cPbaQ-U
zbAqBHC1$v#>O_s(9*UPNTUbTgu$pl|9$sOB=-6CovP0|HNiN>F*=T=)d51WWYs$pU
z0ddA~2iC+{t6V5<H7fU*VIhT|QZ|OQUDz2KmqLb=f_D)4rFAV7sae`;C4FnFmi>4z
zzu`=vgh1%p`R2`aX$Gf&st$k9@7N<K8lW{hNbzJS=F$(SYvEjn=U87s19RN;qg(So
zPNA{P2{ofoj&Y#W1B~lO-(fFcDRQR}tcL{mb2D$59VP01cf!ep+>GP=?E#3E89vu1
zdp|@l6pO2!Iz$;C*^qo^9TL7<K*CNPD%;q@IwUAzQe3{Ld?-62>z;p8;$;+0l!g=;
z@=Ru~C)eW?<t=fQqTbJ;qu`X=ckpu^q~hl=b@A8-d|hm_=y6^0b((>mQTC{>OWv!e
zQZ4wp=%KpN_xQTxnR-Z%__`SJuTyf&Gf5^X;qB$~fGzo&(Ojiuu8`diSUBd4-Kk1~
z!V+oCJF8N5J^NHig!X?V_YSJ$tA_M>PcbFu)l=e<*)O=f7-{!t+~M(@KPYk=1kF4i
zS+b>B@8I!TNKIs?(&quMmtwXa-vzH{Xg<tvdf4kFKh{yY(2<#aRXh4Fua|sO2MHdI
z$9hWC!|_-zpRxAkSnGY|lM4a+<Wt<)!}7^-*gEZZImMom^cjEWkHw{RcwFasUCLek
z;f-wu!rQ23nHJh=kM_lVz^TS$t9DLb$5ke^Ifew)l@6AvElKIB0vnp2T-z&?V8|Mx
zTC2|uni<n9oLaF;!hV}t;zH+!*11k>%i6kMziz-R>(|xngmd}pHfYH!yeYjZxyfmw
zal*asC26=?I#Peitj!LSIMPjRwV^=LKvv@;1*9a1(%!ndnTsn8Kg`sj)5vn6FJ{P}
z_-4ePQlGER3c6mZ-Hl>v2O7fphVh>0jZ+Z6D+_dURu%GxU7u#Db2Ab<=wOa-_@(ZS
zF8zBl?#KMrbkXVNXvH*bL~*@{2Q>JW)~ohlX5(n2X^nqMuIA3;5ue&A5dHCp4``fv
z_;duhPr;{TT2+Xr#_C+RZiSPc&o%V!Fr&p}={zL|2kq#&JtyVZjdxwW-}WO3%4(ll
zeNJ$$`-|hzMzaC+Vi6Qg89UhzPRADc&?JxsRSat|R&K*Tf#uDUpJs<r-69@U+~GkJ
zj|Lu8bCG|Ia;8W5?a13jh(noN^SrJYpNI?kq-t{Wla*5n^h_p;>ghM{r~fUMt3p|_
zP$auR2XiTn4+VVqE%j{lbtS$5`iAOz?0zV2L<2MTgP}<yxt|PSqT7b$Ed4zIsci6F
z1oD^P_~{??@&_91E871FL%)?}_~~D#pS^$E08D?mU;wN_RK5b3pM2YsOJFAu3SCGi
z4};i=8Cjf>b{#CKrBglJRUZyakg5wwi&UN4|Iap@|B~r*Ngtm&zfPfk#mq)4i<s`_
z=a!rx6kHzBprasm8V!L}`W~C<jdRDvPoLgSiPF?)ZC=`T@2PpE#6itX3iUd?Edg~4
zN!5SC4fU@Q;+v`gUCrXI0&m_zbD-h8yV@{nf>~DC#TymACc?F=+XjRwchaUfIA5qh
zc_gjK*A6fzZTU`~d`lM&;>}l8wf)Qb&Rw1>(|HS{S7^OfDQ&es3wB(@O>6k3$>7P>
z1ojl~x?B;lTyOawZh1n|T4uIpzJL+o%KU#<bYoF(OD1v4d{8gD8blVm_9*W$^s2_?
z`M>MerFiVw?a}YLv=VAKDQMjhR<6oMLFP(=1-!`VdUM@tQz=HiL4&0VF}qS(u4E0>
z+xTX2*>-s>bh_zLE?m__Dg#c*+#0*V_~S_1U{KhthgUklN(Qc-@hcZ<wdoO@4AFl;
zGi$xc7o20-nw~k{TP$=c`dY7WQzKcv%473YnEJ+JxC82B<y(byoo`JuUrV=R9wYO+
zVwX9@eJf+sl<y`uU0EwVUGUBc%0`JOjdE>CO7VhUefcbot^B6w{idzGeQ7#Vl|}bH
z)4IW)HCo#)UD=9nMpw4%y!ev5PVImFmhW%-rZcQ=iFdc5(|*hI4m?S@s#r`VBzuNn
z-k0;V2iyMs<KVK^9d(+e8@BgWJ}3b)6yA23hrgQTbE&n$$DYQf9mNq4nmMlPUZ)Z%
zK<6@Lot-cAJH?ai4(fI@VU7qwnfB728{BhpmX1Mt>R_dTF!pJ5GZyh$x@muXU2Gin
z&Av|B=KqYa(t;YA*w&jyXDe_SV(z-oHS=HJzu1*d$vSM7_@2(Ky3uNsWL@@`t=hEj
zoXvchq4B0^%J6dN_S)QTJ>2%_jR7|Ja0sC9{O}um3vuCK9Sy~@W~LPjy5s!+w&VNO
z;+iGiy%IOrsPD_<ze^pirQII9uW`OdB`($0IvFk0;mDD7BiCwh{Wood{{s_d<(I*v
z1rr!FG&VLMFd%PYY6?6&FHB`_XLM*FGc`7skxULImk7utoqs@H!k&Kq_TwARLJRy9
z-RGBYZ~u9tVjFFk;&AxKZ*SoRD8ZDI2dNk;bol*WhhN_QQYjS5oVoBH{{1>5X$^DI
z29iJh%n$#>mp@`<<=x@GvFdM?4L|(*@Hb>|i?Cxn5>{wpzaq>Jl6mJd*+xPu&p7zc
zlekkdBbjk)s(*qruPNCGEGH0H+!(C~U?HHSR!CbjkHj;u0suhc^8-EzC_v|H1P35E
zv=3#ccf$ju;x~j9Ae^)-ZzR6qmrTQ@wy}NqUzZxO4RX@w{!O47+Qm1a6oClXmIo;R
zP}&NtA<h|=;lWzMg(tzYv*ZBD>Z3=EdBaEAGvmcm=zsYvi90(Ck_9gx=`gfs-Vvyb
zGz^5Pf$Bm$?Tg3JuOA^;*<(h?C1EIHOP^%Ki7Pb%(n6`&lm>)AL1_7u89fo7B_wLK
zMO}C!)TO>DbtCRG6`lJ)kU3Jk=JHTxD)pj6kkr;vp0RBD($qVu_NKH7w6g^aN+ok2
zv@w(ibboKGjpzfXYVTDirH>OWQz?}NONs6(3<ltvt>YUW-oUZ5t1ES5$;d$xPsKM-
zq!G>kP3J<nQWdELO5`w=L$>lDpUd<wv{vKa`>*<xXWl0JU?WF}Q^ng&4s&dDz*Y@i
zUlO$lC91wJxXA5mdC`#%RA=7S-Y_j|tfp29!+*WjkcacCzY)iV;cD~t#+Tlebd3nt
zS@+T(O|2dH`Tm<O7#sPx);sm}B!bR@@5#ZB-B{8_3sJBzASHZ1*H)%Vkor=M9HoyI
z$#_hcQmv5s@N0P!W&Ie(yJ`*7MUkYT=tj14!8?|Hsq|>9N~I;D@Dz!%eUsE7z2YZn
zr+?$>J{}w}ZENojEu()vnvGMxLm5Ec^mX)e;PZzMBM%<a0%>4~@zyZ~V&MDT-@i{y
ztUDOyt#V_UfmMZ;*hDdI6Lh~5-!Hj1$BZOg{`CF3Fy^~Ih7UV@nDow8L5!LSA_evx
zO9c6IZZaRzq(s@VyAd;oZU#PP&HiNVaevR~x!Xj;7Z}AM$X3WUh`^Bi*c{AKTd3~i
zjedmTwNV4xJ9-FDsjcdUCo_Pebm{c14;`R3Vdk{57#P7kz9JUm=Nqknaj!OTZq~cE
zgC-4TAtup@;bY#FEM|qGh1Tku%y~9ci8`>z@Xb)`rIL<PO9FIn_q~>(_H_)Y0e=Az
ze^TXHU6RBc``}hNOcS1c+U0ExKc?3$1S^w+Rc(_~bZvTj5IsbP>cM_yYXxF0xDtPg
z>~UB4WuKPz=Ug(-sdYV_TvXi_$S|B4Q&Aavqvgjg>FlvmJZ$eCoMO#5IN5GsvSr8^
zIAnUM-<P%(fr%&dt@j6<+8Q!VXn#p<jVsrjq`Qz*z?aUK=WKObhVE$P$ag?MdVlSt
zD6wkrX3wW0!d9z7u6G%9zM0PD<V6m$l9VncO-_zU(-a5bnB{gFA&Q?=^KsOok4?_?
zy7~L;659Dv)6{k|Po#P<)GG-X?|eG3eXxO1KH32LxG#d9Tc(o?)8=i!$A6&J4Jw}B
z1HLHk3VboRm+sq?>j%w5qw>xj+7F!~we0e&-bpnmcRc3KSB6<kV@eRXp`G;lIL>wH
z{36%LaF8jTka=#ym&2vxib*A@Og9Eu$u&;-1#J~HPqNfu?wnRzK&U%ISZF`aBeQll
z=4zQQGzn6NZQZlJxfV&yv4456(o`)A-&+BY9y(NHI$fYjB`Kt%E_IeVYFj#GCzYtw
z{5lbA9&c?(aGN3zlThnx*coP-!VUwd+YG;M67cv3U7M&9Upq8kyB<~7EcA83=l)=m
zaH`$GG3{*_Tes^-;4PEBt!%7jUEq#g@U*oT-&(5zoX@ysTltYNo`3kPDM-#x6H4Jm
z9~4qDVQHfjuj`_MY2hrIvYU!4PZ!Hl2~oY&I}fe00b7`~eMY2<4AHonET@$z_a6US
z!M=4{nvG(6vII3!yt(|AN3S@=R(Dv&Y_y*1M1#yQP3F_C3I^w$AU1=9*yge^AlN?t
zP2XZLw8PR<&M0TKhkpQOY5rV8pVeeWm^oa>8D;f!ULXjaJyTU(=|cLgJY(>ooLBi^
zoaKlxl&kYeyJKmuJAc$LTLST(6kTW@wjMfc=b8jB9YwM%)=?gpN~g%sz`U!A;(E+9
z^A^n{T;M<tmbQ+XYhtlW-9Y(77_^pI#4DA~OtyaNoNdTr>3`z=N*k#Z^Jc@Xu8b7F
zU)wS=tb?$$!0@3I5ma$s)8>c}_dUbD=DFV^CDAWw!u#d>7@1`2^_Uc;4JL2mZVP^V
zCEuodpb9fD4VLbK3S}8MD4U>!>^&QJ98H4SE6H%Gv_=O$a*7Sk0Lj%qzyEq5C4c?e
z6bh?5I0GUaC4VRT!UbJCZUoWVgADlHck=RK8S*o-Ng;#f2(z4PP|>8o+(PI_L<<vj
z^=0hj68((G1SwGAmXBXPZ(x*1NJA+Jm>U=~d4<@Zd_O+V7XTBI4O=^B3!Y(vSFqKV
z8Lc*aRo>K9Yg|3gAM-FT^*;Z+u^c8uehwrK4D>5pP=E0IDN{(PnUz~tT&oBhjAGCL
ztzLoAk<uPm3PuUv$1zISx-~|hXIn5z`1TlqfRP=nW`sR2hr`iA!BUQJOhWJAG^|~j
z7sydV=Z+kZ`^;r7hYH)(b|c3=V3iMt7vKeeC6so0bO9{o`!HZB+cjWcz%~Px^6h~V
zCQa8RV1HjAM+WSk9Q&Q71l8wd$U${a1akwc5SQ7g7z?aA$Fwr9!0Jdzm^D@j-^Z~^
z*!neApJ!XJO8DNF0K9;ILHi#60+5>t!1J;ctbWQAB>|x97Z}AkpSJcD7#%6yil>C{
z;}|7u&uff6&$eKc@TCa=?8j86n6Um8vUtIabAP+CK>lCt2emwWKQBv#-i|D~Bqeh>
zz%cJOAX)*}n0|x)8Q_kTb}UtaOZYwxT*CGSaGz&efJ^xHz*RZ8xixT~m!$xAA<KC`
zd(ENoX)$D&CmFvhM(96wfms=s%cvX+%xaA;@UOt^NNLAV6|;oz<CrCEFKf&`&$eKe
z@PEBCA$XpDNeK4%hjA_kH@8j*o|dJc{!^x?QJJt`U=%(5l60V_b1S9}JRe8r!1B7r
z=CdpdG6$Y?Ql;0Es%K>=Q>v*9kg*BN6@X`Ds0qYW22=XTa`r#Uuz;pgG2p%coFk<h
zi&d2G;{YXWZ$R{Uwgsew?~tqRKopp<FMm<JBn=B0F1N$y({hyXxg!Rw*LBO>pNKKR
zYTkuW5|abh4hZplZRyLrwh+uo^?%+I5n9DtK7`xOrbq?PSX=Av&`G}$0_~#Lo%>B9
zc{1PQ%DX6^@~IAY);9Y5F!v!2ftqh^?UQU0WmaJ~&^!h&rx`x?%HllRx;-9Ocz<3L
zB|KA7+$oBZ3TIp!!J`a1$)ogRa)da><-KwQaB8!Z<MNms;dqTx_sS9Cm9i^2!v7K0
zu~W}Bfy+I-ozG!gIP8OmO-E@w&WUc^_Cuk(*7;W6;`W_XKJGC~tC(!JIk#Xxy$dbn
z6nv+5DdMTl5erys;^Dc`HCPN1rhk@-78-RHhT-HZ>I`mq6;X;X61A&{6T@MkQqsN1
zaDX#YFen-NQq)ym*_|<pSyebS!{gzT*(>3=E9TDUOG@0URg&3nR8i|H`-VlEh_y|q
zVOK^8=hU#N1jn=?+t70`W5VMtyvipS(A1XuJ8&ga`u47zk>hmeUD|YAeSf}vyL~;3
z%vEK~0S(t=qCM9cKauilbtdd>Q#zbXlhVvLAF(rXIo}W5yc)OYHU^Lg6HRs==K4#)
zxOuV9NACWn`e`d`<J@8uQf$_$W6Ef}CbNXO?mUcj?zOrT_NFQCKLz;9rvT46rnqE#
za;mfN7-KL&e5uYO_s-w(6n}Y)syj>V@4_0rS^Z8Y$ny#~hHR&fUGa8dnwleAKD17g
zU5&XtFa-=vUTF`?{&LeGZ9WyWh54%LGzQpxBgCKB@MNiaZiJ9<z-vK4+3RM6%iSHJ
zEz{lE5rn}xdxyo#LTQ8Z5cguwZ`fC*ulu2I$F*r4X!B}Zp4k{c<A01<c(PhCH$as6
z#a<`u?PvfEuwBZSdcO3AiCwlce)>rdZwjdyNI6W|$~kdNmeUE*a=|B_8|CFpY`UgU
zO3G>Qa;-a_aTWRWKezo40hnC7+;G*6A33Sc?eao;zU;w$x$)`px%PZUKiuyd&iC`p
zS$MwtgAjc4GYC!djepirQ?WQ#%=;Sxn;+QV!$p@pDK|gXQ0~f#iRZctPo`e?*mz$k
z2FB8J+rugQ_G2U~52q_K=kbShwm*K-UA38<@Xn{N@4h43|HR7uG|Ppg`JG!!y*|BU
z6n^TVt^`3lVGZQeeft3v-60=byD7hOeTaI>Yo=9=p8b`9%w{4pPFDbiJ4?ewXE(bl
zQ*S(Vb-J5j;M;96F}W@FxnuG;laib&1bRFrALrma`~pEd>%KfCA8yqPM+in>_E(JR
z^>CZ7BX90iCX4k~XEx9&zwps6zV_~}zvMIvz&OtQ4`Y*%m+>Dv69F`rQ0p%gmt2}B
zDVIamB%1-)x8K$z4G07)z_ovup<^Nxm(fE!6PFj?B**~>m*C$d!GHQlZDNg93F%dy
z9Wf#V$PZoz3yc`w%G1kW7mQ<i%C?~R4k~-X(fe$T{S!<!9LHX&)%gBiDEWlCuQUm1
zJzHqqdOw(33tk5_A|;#r=w9jYHmnaHG|Sp~p6_|Bo)eIXcGkFg1D?t9lyMiDtwZLr
z@H}WH;{=PCZ^;)CQGW^3f?pgP(wqC>(Rf+52O|ZjVolk;SuU5og*`0ZmFVDdG|haN
z;Yl5Z;*&9kW!y5KX|<XcTX^B<P5<2HrO`^D(Z#c-yoAaY?nN$>wNu7D8XtRE*5V^M
zd=SAZi?uf^4FX}z@n3%V>skOH&5IyID`YmZ3bbo>n<7qx5`Xkn4W>LOF|uy^Bu1Qa
z9^;>%n!AthFaMb$F0}{&x_|t5;TK3^X$mi3MbNhU^IzTXFMn&P*4aHqN@=YzqwEfn
z<j2m^ud;rKl!-~CVdc-Ce{{{0Z&enCAhgvjB_-!`wj!wr^~t)U2)%24MBzhAP=3rW
zZ?w?CFD;+bIe&c2=_=9sdN;oF!URWB&lF6?0Xe)WK)m;+Pq~hYAhHPbRg%P<?(VCp
zsn9k`I28)tIXG}YbZSU9tF*Kil+RmQ^0Gb)r7!^PEKc$&CHcfbN&A(PbJ%rIC$++D
zac4K8QU_!=D&XqE!KWa4E%bFpq=dFOpFC5ySS^)=k$>x3lN4)o^Ii7yoNfx|m4q|J
zP|s;x#SY0cuUrbCoUTT6DtlQ*s!H%^$3ZifU?t?vhj~NVb`Fb{naXWIDVGia^Rtx}
zrD4#QH{)Noty0PHye)fFN&q?HXik%KZmHbNdQ7&Oam=OvkPs0^X*U;0X0WC7Hn#<R
zbl+h|O@9i<QDUFbaazITA|Mv^vxR_5)*suEq^n$5kJD~bV5l^&wZYr8fJ`$g4>zeG
zj-yK~m@6!otwkDvRlL0SJ{@PfEt?UNW73$yBbMXZNmYWkRV89-1w}R7%3ZFwZUbPY
zBIalB<BByapf<6-i&{1S0`2`+*?`sxqSlrToPU0UIbAlO*Mh300^WQ!+m*((B?h6T
zyi`W?NxvDaXYk|Q0N!6Hi9r{$+QnV~VZCoxOeQ^HFBlO}PxveAC(DqPr6l)j>+TG=
zQ$kf;rOggf6)v}`uS;&n($86yysZ{E>qQ7^whWy10Chq9NFhSyHU!<$+bl$MEMJ;g
zO@HXO(?TQ{7;PEfc3O4Em)^`270H_~1K!I4wC>1qR4LKIR9dsg{asUK>#28X{GDo1
zr(xaSrLc~kf$o4pM(+osC*K~62|zwM)c<-*-UY`Ff}Fb{Ca+_sNKFKCYou195Iwlh
z&eTD1$5>3*_HLv|o%$nI4!sg2M*_otGJi0mC*Rmvf}BMqbsEh|Y1^R|{gE)Vh&@Gl
zuoHOzcdt+=I(h<jLh*fGT@J}qV1g&RgFKI+cLFI2Y-6~D59yRbjB6>V*<OPCh<>BG
zY-A<3<}fO6%hu!-NunnksFx}zsH%$dr3TxbhMXshH|>f`Su*1{9vvl2)?R{F+J7iw
z9(xa{H264wih4`*u3d4V8&_yWpitEEhtK0LctMr>gunO3U$$*_v<+`268GtSR%#7~
z4Ryl9f^AU*CtEBr&Zl&wpJ!X+6}N0_JbKBtVvq1iwl!XH%eET+@Ok`Y+iLjx1lt-@
ze8jdI9`3V6C8a$f{m?h96@@^*P=Ec-79mP{N(XkbS4kb0#!`i35(6YYuIK6hnZkx0
zLB+!6*WYCKPkQ+?4d&eX-#GL~Sx0vN?*7hY&IV!9QzT5e3t^40>?mzepV5;YNEjV@
zRqrM7AZMVd+0g4`Nr%C-bXUtX5XxvGPL5HIBjRDLTsstc`(h=R(K<9cN`ES~#~+P1
z`FgjhKLF-K0lrkqu`z<>xhuI?g?RYO%Y%z%Ow6dv%t|G#^O+ruBl__wmGuDq$cT$l
z52L@KRmnT2_}N-D6ggZB6}%Nvwq>gg*wh8LMK16=8Qngqm_lj7w{)lJ%;>TCbzs^=
z`7j#mC@X5zoi)!vSy5k4#(%mCWpff`a|Y$(Xl$VD>5a5cLfNB`0e7KnPoiv3p?nyP
z4U}yV9^I2rcA6S|vkPT+5@mM^<-=%fplrPG-aiRt%bU%17s~!5%Kj9}htb$T869;O
z;YlbPs%^mSLOGm7Ih;cIFd7>u>sZ`92V|5~K>A%EUrvI2IR)}jFn=~MR(+}U9FSEI
zDui7iUr&O3Jq7Y%G&WF%zSMdS%HXLX-39XPB*?c@ARh)}3uG0~P)@YKyGcCO^(*RD
z{W{sDuD{)0c;Fjna6!1C7OFm3$frfkrykA0|H3CJ!`y{)rNuAw>LdGjhBrN}n*AhW
z^wzcKUXPwZ@mBZIbAKq{RP6A`=W3sH4h6;uW8jkxq8e(82dejapG1EgK45OvJ9Cng
z4xw9$^Cvm^q@)@BNhdiRB}mr|@jVAQ%Eax0R=LLya+11czV_IwI?Z1|Dwyyl534Tj
zc}_Kl{inJ4BgrF#`MU|6Xt~)#-$jwnIn6}1pci=EvHQ(un|}~yjb>r@n@=Ov#D~uB
zYg;4*PhuK91xmPwcC~@*Sq1a?yV~*VCh5+k^xY&L<cv}}pnuM;HUJ9L_W#@3i#6nN
z_&3zbiNimsu?MLIy7k6B9Nf;Hou_tI=EtR9N7YjssGoyrP{M$J2vTzpQga5<qoS_^
z>O7UDPiopdP%OF$4<Tv~B5KbddRX>#NUakZo@AtsO8<udbq4`;X8=7c`J_l_e*Xto
zCjghhra}_~G&DGuag7)g0XdTq87G$x=_I9p*y;G|`%h0aFwO9jWuLF_PyczU*;Y*A
zoYdhT-=EwA0gQm*2PJHMV{|7^&~Ci3ZF6JWwr$(VhQDlV+qP}nwzIMA<mUe0_kO!y
z=A7xC=|0_4^-P`Ws;3|mo^D3$Lw0h$k|{;T6EUCRzI|yVu!Zm>i*A&;<9j}&e%!nv
z{D#K0M>jiVr|v76$Pu7Fj$Y>Z63)qBqDd7hDwxzbc}Z5}7O+hcgoeaK=s;CMZxBO3
z0VJqKrydf#JAwV&#BD@Ayvf^Hx8nynqSVFNdzRf}hPIbOLm7sjRBa%1hw&3lsg&f+
zdw@a`1XUTG%B7gfC7lZUj4!zA-|*-Mf`IZ_5x*;;$3~>3h5u~_>i_M9Se_d?TMV;w
zFQzinp|HjCj;6v%f%_7maQFfkv%x4&i7M4|%WL=e8J$VG)o;UvvyMF8ww6uGD5~Kk
z9nZrt!p3E#kywZOBKk!XJ7R*GQ;k9(%rSVDPH#?*jNvYJcgaWzVMXW{(cJ+?38Zq2
zq^S{oo{+pZand!Cmprd#Yl)E#sm)0oaZ#ay=KF{H8&47b-(cmf#dScMR70-ExCIAa
z=KOhQQ32!BS}@t|k$JzK|9<|Ci_l<<{ZftDAQ>8k%sJP-QMiTz<9>N5xldoOKQhEG
zF-qQ#J#19>ukQL~#ic?4cvvK4rBdiFqw@X<0adH~SS`_|D(_!+q!7Hq;yd|E4s3!t
zAfcl?qQWk@mPSK(nX7;mc4_T~eU;tTO8gyXT?d&khkyGymWlh*EK)e!(=2;3%LNRK
zJVN=SSV-ePq;D*WM;1A(`lHjf;N%%^=nlA;`VRhHF+(rA&1BPB_Bm-Te5QRB2Hccd
z`G?3fZCS)h{nmnWwp|w5AP^~-zi-+PsfI3Y#fm_2Ll^I&F+l+0@1X91hzy9J8Umqx
z0x*2N8VitHI>qWB!MrzEQPJ^RD0~s`UU^rc1afffYxTn5A*D=_EUQX&q8iH^aLEEF
zg_FFDXq$3!*KEaQth;umhM@6Ijksp&&|sj{iJ<x%CMp_y*`QEz_`<@|hSC>q+&W3E
zCzPYSduI$`a?=1xJFkD^=y)8AnC~hsUd>dnL4TJca6qI;WgR(LxPD<MXQKOt#F9ba
zFnUJx!95picfavy%SS@bEo$!|tnWoBj|{TkP0XDbJ`h?!SN}EJyI-St)Uluyj;9B2
zV_a<hi*<P27No=ElbL@_<1HQ^C-0H{+A}@O2=QQ@=?4wy3HX&MVWf>n^fQkI?8%yQ
zo4vzU>*EwTylQs|<Ur3GoATz=@fTg*;Dd<LTt%<4XrO{Sc(=u|V!WXjB4X=Q{W1Fu
z?$SMW@HjKqCwxZDDz`|u{*Df!c;#~IOF0>d&Sc{0HUi+2(FN&}4!5JUenQ{(0-Fr9
z-9520f&frtYMH*~hc|pwjD&*DZvG_rDqu*4R7_7mX>=l7h92ONEUJHD_~+wjW03Ou
zVq+#f6$BisdmwmXs*<^;0q+(~(x`QMD8MPUAJV_NR65=%g!Hf^{-4>PI%yfz-jSe(
ze4b-Y(c4|NRS4<Q|8|6J{Mo%Bl%$qJJqI8cZWBp-mVvcl-Jl6V69n^%#Ui_}w-VrR
z?QaoE-j|WKVg1bEkMM|3A}JVF&)`<sDH#8xEa}9&PZR%-FVi=Yw2cgVa8>aMLNc2#
ziIknBn^^Mb87d}@lv%`@)6KhEF#cGcY-}pDMk4;rr{9?L)G10P3UAFm_TQnhB;aQp
z>v$8ek<|ouJo32|D>sw9f^skc=0<#~@!$nTIrD#4jhO?^$so$z)IhDFuE69vTw6=d
zI1uJ35!@r@-Cu-Fuxsm7=6|F@I))Y=r14)9&0*8DawfQ)zIcv<eE8^(*H#b)e>?5>
zPd^kZT3nFG6ER}cr{?c^=ijN_05F)M|7NVRkvm#ST))owbPaJvxmKhT3De{VmX@Yw
z&!-9`X5XpY*N!H7rM2y6!X^cprs+hj`)x!_f)sr?jpqiZ%~pz$EULGTtC#7R(&Q+H
z8T<4}up9AwD&4M-4XBMBGa#ro8hfbmH5(K2<s20$eUqlN1Nxd$%~?-t0N334mdXf3
z=F!Yx7m5@#-qmNC*gE6ELv$Rj1v>NIG<Xpnj~|hwk^2e_%yPDAFwbe14Zj$GC>b2w
zp}BVfU5RhIIWp%0Q*W%#a!K6xSu_<=L#p5v70`pM9kaWuuaw8PKr*&i;d;EHY(W<b
z8j?@7vmF|?hsR)P5GQOc03mc{LuUCaQrfCHWz{{?njd%j;QTzaj(=l3Eo?z%xD-9b
z#2*P4su%_6%Lg?P(>tcib!K_(<I*sRf#0@Tm5N?z>ZaRlOW{TMCq=nD;+gX0$W5(p
z-bp=>ib$mGo?@-Bg=^~S!&T(t823jbtJP-%;|OImpyZ*9chjz>0JNpY_WJX_i$75+
zrE}$fO2aA7u&c@+hmq0Y&DIkZOXqbDiFDcn9)<#QdGi!TRU>+{+Qk`GSk3duQ9}>U
zwyl(ubk!5i--|A-l<sPa$xY9@td#P`uky*40@W4^kp-*CX&V}v*LGXi99TUa8{#;H
zsxO_<r!$Xw*LiQv0PFmH+`-G!Vl3mZDurCXb`1l0G8ILo<GNL$abY0yr7LuUj;SDY
z95;V6Ob9<p+X&87alYsB3X7N-214%`x>eXgITYM06hZgc7vKyOW^R~<;wk{F_mYEH
zs{Hy@y+AS+hT-`9X{x5gn#;zZS1cF0Cd$`|5mdT1W<coyfMYI-`(oOpu9BO8sZdj$
ztfN`ib)ICy2_EtZ|83=Fw4XYc)Ib!8WBhesTbi_XuUS3<qD=*0TVYT+cMGjsMZp%4
zo)wI1N&9;We0oiC2j`V;w!Zg9sU-%X*4Y}&HWFPd6!G`q_HN!E3dCEOIeJvqEdq_Y
z4w>iv5z`bqfX=Sa_T>mR`dzgkBx&4Enhkx6axh@JWY=QPyY!4R8P2BG(U;G28P23Z
zi#m!OPMQ1|W?1ylrnfrR;>mn?g|D?UnY6iu42$m?<}D5L^+{ZhmahxRf2^76fw^dV
zN8Ll+qBmzH&qpNtVOtrM^sisF>kB#r+!aC*bUsZb0M6a6f`7UY*7EL{xf{Vz+Wm`C
zpP4{L3}PDws~>qp%E(!>odpF%qOQ>2V*SG_BE!`;<kx4*o?Z2iGY7%l56Mxy{^#@Y
zelTMSWTe5cR;@;>Q~%tMGT4cRCRQW)<iLy%$1kQg8}o*eNJ~#W5Nu^6v1j>|I>a%r
zX>4??0PfEHmgg!lNssm*F*!o38ND_0*^Il&KkVi(G`^C^QzCWMYP`Yybfz(^EvOJ<
zES1i&o#MxYSm{{-|9Hi8cib>s7aldQqX<yvf6W}nH@fY;T<DlH?XqbFy*@J8D;<F*
zs@qO^Y%P<Li<LB4(W8HB7cs3(oU&H|7l=%M1K6xDY(upRqPz9$;Ge>ud#JCF;D-?j
z8fWVlgtVm#kAxw#wl@UKt>!G3@9=$^nKR6HFg-#9C)&39w|q~RH||30up)`9riU@O
zAx85t@!;DVo^co2L%(z7f+t|Hw3ph9UmWbzY=bXX8EuZ*+uWKa0wd;D($r*QE;kr9
z)n!49ua)Y!ofkyZsPZc>oK%FV=Uen&986YJjY`;+vbH7i>$1PY^itTB+@HkO^3R-6
zJwo{O6D`$u5IsV=iI@oAgGad|yK}mh=xL+*%V)lUYjJ4d{wE!bBc*Sb3neA<fpiK0
zTXOz*-15WcwrGjO5AN;z$Lsa}uy}*-hYK2zqbIjxO_q9B-=M>gXZyTTvt>`uucIwN
z>wF~7KLqH}W;w-*k;OhwRl8dJX}P~jd6~FAn4*5Uy1uW<k!!i{jYQZY*or~gU1vDk
zh$ywUt73%lff;~x&uaQxwopU&h7M2!<mb2`$)#s?)3mxQ*6h_>@1Gb)3=i7xN1iVH
zTy9DY(X`bF_O~3__E$2+p}sDsySTP8w9w6A*9uq4rlfaD@E>TsCipYemsG!{m`G&J
zqhge;Hcr2hy6mI5*yGJS^=p!mKQW%UDgZufrU_s2wu6E?VlM5AEafcY0N?#kN9?+^
zD^4HJ*i+_;MLDO{tJ?OrvML}gsXw2G?<`jyHj8a;Z*K52B+DrN8J?!Q+`OMVf(VZf
zfQzb)g~oar?jDU(tNHY7ho!}3p!CPwAO=Hn+U#N8?rc>D)leHruCL8jJ(lb_-A9fq
z2d&4<-)hosY9@`K#D$ucfMINrAtJ|Ie@{%6_3+XYBHIq`!om`_1vck}nFG@6k{#_{
z%ws4I?}*Fl(l{v@A!cKYP8Hr-?utrJWGl4hw)o-qr(?wCX*me>-$HtA6<7)8tr1or
z*YTY8vQiaYt~^v}glS9EWgThCVx&dM$%tqxgX>k#n#?g{C}CqS0L(lOp94l#<C3Ih
zc*7DiX{(qlEu6NCNRH`DonZ%A*j<l~L}rwzdG)9n&xy15#S+Yh312w(is#=C)gQPa
znd`9)K`Q21fP2M=>Dt}Q+jEX}sIJh1?vJ+$Ns(hw8&pjx?^`q|Fn=^+NGU|YUU>@t
zTrecCO}YPW%>WqTp}-&_kPfh+n3RH`g3c14ni7bDe#>Eml!9Qu#l-%(gW#koeA=Xk
zk<gLn3Q2++GE<fqsSR!&hS#Z4pno`+%_qT4@PK)e-0Lr+(N;CZLqW8-yEL^cc<(o9
zZER6OT3L`Om9~$whZx2gYj~+5_F!<)p7mic!xM$!Fad?8+aWgLOET(CiVLNvxFfkg
zyt>vQWmij1_YN^YEh{p%RH;FMIc5&=KAEFz%9UnkX&O;3I#IKI?Q=iz*erOm;;q3`
z>ihsdSW&Cxs(NB{CNy8Y^fiuLtg?_Pj@`=UGsaC#(k11(8A%=S8ZnmR!rB>0`Ie5-
z%*It+PC(IlE*8dJuys0Z0h(}&t{4Kt!`Iikq#W*)<T}l86kldydRmi0PUcP07?ixM
zoDS9e))R{dI8FLG++1%TS+E7AU?gatKFNQ(`T_qxyP6<92n0+FhKWoJAvG%y%V9|f
z14Rt!0E`4i20ic{PU*w^rXJ0_7fm+_n3H3|w}Wv;B~8F{a0ar2@7*xtyP~F?gaTI^
zMuv9AzhWIDS9^TQ-m3S9GW}nzTL0-tulj#V<-BHk{-7=mAhWDUlNXA{lvb;xDD2jH
zYgt<h&7>Zx|5Flnt)2b^Z5!j2!``_#&unQC#GU7}H@#ppU<Mcz<t13@XLLpJD|XZ<
zRZwIo2C@|HfX<_L%{o?P(@ARjO~aB44Sa?Wi#KRvSPw(u^ht8jly=>Z{|d!=h>QC`
zoiwydydFi<W-6t>K(kVxZkTi}5$+ai-&8Sh?RzlWffe=qK34tn6$}f?JyuPL(I(s0
zyI4^k8%0(cE&;!(i7t=Y6IWy=w974<!<kr<Y}FL4bGj-1wMG9UK15wu#+??OveDGf
zIe<>$te$NJ`gOo+WCu(Z3Iz)(75bl)>SsXuBa0w95d0^!ZS6uLp|8HgKpem#kcqi}
zbA1Z_{w5726gUz=OeFLNl1CC$$Y=OgOQ6`0=><p<Sf&u@_!eA%ldAje^hcArS)0*}
zZhP(AfR#1%#5GlmF%)dcdRRBnD;ep&H;=5#pPgFcv(r>f7TbFjezIuSn4rylv~^t)
zfi^IiFvFMi>j+q>3l_D6Oh)ROs!x*9L8rBjn&7w9A%;iYq(%i*pP0@U^K4&*g@pi#
zDg$^;Dq&Ie4MRXrdX;f~_X=D!8RCOl<puE6v<&ct=8@3LGcdGX3nIxXSrU2Uo^Lp=
z7?d9qFB(wX=S-2N)+}?6iLxDf6kn~_@rZ;UFry~<FjSgK={y<Tsd3Y*DhfLP8t&k{
zuv1FEVS)-c?#5*<O+wh&^D1TM*Gwuuhyt28j>W5)XqI*-;76M{PF|4@SQ7~$f_4{w
zcF7+SL=O5#W<v=ZRLPC}fRI3JV39z{EC@#nSz?2JP~^q$XXAs~{K8{>n14dtOayUt
zBS^B$^iILugE@OCODnLTVMEr+D%zg>L^@PQu`)UY=&<h1=*LFYtITb)R42Is!k)gJ
zrS2>k?L|~GyimHP)e%af!iPjMZcdY?6JHn~gL#*)4(Bhwxp}o}4L;yjRL`qfn(@~0
z(fDVEC#S>re+SNH&OuMFhAeG>-2gWOmpkxudXbH_aoZ2v<$?j*lg(DJqfl~$Uhe^;
zs}ESbBUSyI^BBtpf@N3$o)1C?nC%E~{sg6)wkp4|2;KD#?BY?PH2}5NM;ASHrRb<U
z`^F^XgN}5yasrp82NiHyeq=VQ*{O-#$Vh08OJ-ctS~bTrHPI|p6j3hm<KF!6iI9*Q
zF5NRJF?TfeNfcJjDx{&MA2!_o`<>8ms4*Neltha;g-GI>NLUQoOGDug_`hK6E&V^(
ztonntU`QrzaG{ul0)apU;)4QE*pN;fa3(}QSiJsUFW3|A|3#5|O;(SRsla4*87pTS
z%ZOTx*e0x39=%~s{2R42wd<w{Y?y+y2R5%CVVUlj4@j%~#4-K)-cd`>LaIuQZ)8fT
z``j?eHp2lVR~D1o$8-8n<M1tndjIBCA?HC^rA{874n~#v<`(o|-p00w0c(~r!^QA7
zqY!Z@j<T562F<$2;fJPGf{kktGT<7u3Ca#h7i&;L?$)B{t#GYA&3Tun+aoZmtr(%Y
zEN5X7D28m1y^4eGic38y!}r2|5nG@2GD{r)`iTms&Qd{Ck)tRK9MVXu#azMp(-2uC
zPybs3uTYc6lPsk!H$)nCO2HrV`|pAnnBWhJauEKf#*=PI|NXdXQAJ>XlpN@MjzPf>
z0+>4!;x9-L(f2{zt6ML98Vh$A*jPVnRP3pSA&l2&`-f=7D)o*X>t_;8x7G8}j1FWK
z-#74@+#t~ZSZ16lb%<anz-%dX8A22(bl;>IfWVip^A&me_&7Vp#%V}@WyV4@Mj>Ev
z_uwnoc@+Pczh4wOhvt1~mafiG^5-Z?BN_G{m32I4YDXz?!B??jge(R7QVQZ*&}^sE
zi>@-+SvLh~p|9T{Ll4fs2V|^K+mW;Lga+Kv+H9c*XSn3<FQk2?Nr>+7x=pjg4M?;P
z05dUDTO%tI1Sn<&a6&<|pU3ag-i9oYm#P?zwd5{<ubTJELw(xCyjl-J!-Qqx=s-A9
z$=uW5)Onrs+E_uuh<9iGquAJb-yg4DD?G0rgt*~1WMOuM)SUEE>&RHM(Q+9|no|DG
zQ6O3l`CjNUhIm9k?Vv<hHFF6>dO=FUfJo)M{UY|T?cZ0WLGax{lr@$QfdpFNdg>NU
zos|b<{0VN8a)E@fD%=7;c>@L>!4LhpR*0S!C{4(@HZV{wjSMbnf1*<F5Fh$mA!h9F
zZJG)FT`o`bL{7hBMyOo$z=BYqr+DD?U>x%BZ2ujcds4t+eW4>?9GemH?@?@m0XE^{
zTB!+TIOzD{1s?jx%6MKs5&NLPTNQ&7h4vCTHG>ls{WCAcu#|?w5ig*a3K7AjoR7XV
z?*5suP$UlWhI{hM$2Ih~m1S<?myD960d(?8o18HvQvCV)rdealvleiFy@EDRLq&M{
zpT}1<fLsLD7c@s#sY{d5t+w0b0n+EZZ1A*;(dUu0Lx?fEgvN}M$6^QjOaSmF$fcOR
zO_bQ^WuT?yg#Pz9qLP+MWY==&MMj>@Ve$QhxWOnQlA81<T+SJ*f0`1()ZW1pjstq3
zUcG+n-T2>yqWP1f&uIU}o~iv~EKx8D|H@_X$YExEt6S?CeKY`8GN2Z73D6KO%NogB
zEx$d{Rg1Az0fmAeF{UN^!+~%8Y_9hvRq1;5MQIy2Z*}mc;_UVZAcgCe^Q0?QR}@(D
z`cRkN2obZvQgg6~hF5iP;Xvv=sfzNas0kWnco2=6$H!wt<$3glZd&c9*d#p-^XO5o
zz4>EPf(|pVkE+hnUL(KWA3z0KQuFe@nEgRkxhq!n7Aa(26|-UWNFs%Tm3Sw$`HE?l
zZz0pQWtE0F=<8SczASa_5V5vC>O60KQ}fN3DWOL$7-`Zh_Byh{0~|0yiXY-=(^|Js
zUTUloB$ewgy7<_$qWVc2vI92y9urr}{u&zC=g}6|P8;ocOf&8NUckS7#O?K^oeB$Y
zc-%W&dOu&o8}12fG}O~{YY({WO}Ci~ko*>GzN()CHY%BeQ4$Xn!>_xFi!W(<Kz?5x
z>YY0FTURwn$G7`;bVnOVYOSQ-SR!(iJJ3XQz7heZ4FZU(gyZM{A)ApIYglCmLh>Mq
z0nBstpua+*GS+x<B>-CtqeiW~MTZ*K8nA3cL!@rtp;HN1;j)c)MbF4CjB`*iyDYW}
z(w2ahL;BKmpy{R<sCtPPWJk<y>=;&v+(&m?c~pHjA-EXM+K9S^reFaFf3gu%crroj
zTkcQmW}9icHNsAa?I7NKob^YL*vBd8jr`ZGQmWS0lv0jn1`w4e*^}JNHyC);idv~I
z{q7Cwdl@ak#WjWB@BD+d{5X8T{Ql)#IviqiyGGklO<8>E<Bfi2n6++Y;O1-*bpH?`
z#v68D*s2#ZO%<N16}?KwiEOM0*GU`qSzqh=hiO#atm>68Qp@<rF8`WYS@|sbdFvI}
zOlkU9m-jKO9>9IP6rM{&pj|X^fB2^V<0Ea#DMr(Ky(ph~vN8Wzs#3XWl~1(!!tDjr
z<6;BVa)L?jxW_{;>SMbzzZ>0+QlR&Wu0)14AFF|0Ld=%8=3dLx^cPioVO2pSK&3@O
zr|pqbl&7V~{;Mfy_zNS$t{%D=0>*@2K}ER%q17<~FeLWglY>WNGmI<66EY=1?<#wp
zgu|j$S-<i!k9R9y>&}j&wrcHkaf|zeO;cELTf03=KkysDN0iJKvWuwt%~8T4gQN!L
zyB;Nu;KR;1w<`yY=JztyD92VzCQP$YMLh-}8b?tKC^!x5v_G0Kd}PGQj3G3#<?77N
zQ|*Nm5CC~o*s2onbBg^fV;H++qH%IZf??%_Mk4+Stru;ps)W7^-`QJ2+2TDER;PMQ
zbhtJZ{caN-BFA|12R@?+dHmrsoTGHPt_5$!5V_%(5Qps85lYSAR&7+mJC)u6X}H0y
z#L%T7;zoER(vSSPm5p*#nbBdCnxnI6NfhM(RI-X^QzNqp)Jh_rao$?AnDwHgp54M(
zF@LkN+?5$e&__qvkB?PrYayf+OBt6tla&0K0Do~|Q9-RLL{8B=Eg#yMhO~{m8Lnd3
zCkm}YLbQc(%*!AgPC;MqgE5C2|L3;W|7|#_pSeTfu9&RsQsR7W@s@x{z>765Vj^k*
zz>t)D%d4$?<|K!O&;5O^&nv#33}I%kI`zYKfNW_g?;v60ap%m#2kRt9=4!&|QIhP*
zu4O2h$7<lsk><mjzZd}K_D*i4q>5+I9&ta9fde*)dObiPi)a**n!YM4r3mzd$3k@j
zf}0JPE$odIb_&_tF8EPLaySoy?@S4RHj;kGufy-u=!z*W_yrrg)G@)aBer1Kj<&XD
zX@MJru$cfvHfov5=5%JGVmN{)Ke+z&Td1L?;zRokkklbo#2_Mqt;)LPTqNF##>+}f
zZ)UND7|oX`D*0VKLVYocvYsT0URk`es<{Fskgf-^iTTjJis1OYqu@wPP*GmMZc9_3
zR!-ukx|x>7z}+TI)coJ3rn{}EnBSNiVSET}EX8)CE$^tSE#5%=8ZX<}CUlFOWc?j1
zQw%%>a6#dnb5~ppO{U|K=GT{Ej*MVTRU1wMU+C~5@43!SETv>>WlU`x8e~=YEWG&Z
z_aT&1RzNU2@Bh@U@LEO3cGao?%|$4vL&Ed$3uc>=vhY&ces?$STyTTJ`d<t%IHAuy
zAC_Y0hx9pq{xC`jd+Z6yO)=CnB|szz$=p~Cj2l~zYeoz?WXaX{W)CdV-klwBJua4n
z^r0@6RQ6&e2Hx217q;j-p)l+bGrLS*57$^_v|xE?l^5H>hy@~YgN9Uqh9xNpe#1o_
z8Ve+sGE1=?l=_pl7sBl81cj})gp><K{2dKD3A<Vi)Td3G%}74cRZZ`a*c7lxoal_X
zG3%6=q)T|B7|wn*)NnBwMGoN|o}ppv=KcO1ovB^qAu6aH=Wz)ogISvXd(1&X0Xa6e
zy2!)`Iohp>arAu(6vGHWb(!EltxQ=a9ag7}Dr;j>H5FT)7ZF(;P(8)Q+uilZ-l~mT
z6clGa`c1wbIFb^ua?Z#}lMy$-cmq_zAF2GJ?1L#s&|W!n9O4V-;!xP8Z%%YSX-jv}
zHzdD0&2Xlj<QtDqcE+dpOSi4p3;#;$R@ukyr<A+m9^h+(TtVoRd}HFDlj)Q?S!$G&
zTMq`xl$^s~(0~U;uFxZ;&HXUHw7Lf%5{B3pPC<irRQf-T`2H`kAD`Yj85cndcgju{
zk1`0Fl~JntXxiRy9{UR&9TV>zGkU19$(*#6ncLqiP#?Xw>JB_Cp7lZu1%nS?2M?EF
z&Xx|k4oJsRd-X3D4lD98QXG+p>N@+UbMYLk2M%KxI)D|Kfb2@n!`DB>q;!8L{GwCh
ze@T%mE6eZneBVWz7e8OkTwi~=s;{|cThfyolN}V6^!RkXKY8(e`VsVGawu=heuoM3
z2k-fNRFwc^^mjfTOHr;<lm3=ox#V-M%>ISA)>si`-6OfeHri9KCgKzHc9QQ8K2FJv
zr!XB)bq2h^xx*$OGAB|tCJIV_jji*r#LmrP3pON%_<Md<=ZUr#GybKicU&td^zS1M
zbKb+m7?#T5&naN^G!-5b5v2F$WFj%8p-WRi&Pixy|6R<>+}aa2#L_S<H%v)hRe|0?
zP84v{e^7tJB}>lz(rblrm4hF3!w37Q8N^gpTLdu6R6HPh*9rs~omjDOECfEsg!)yU
z=^-f!>d%HriB33?SZ{9(Jcc&#rA_}WkzVn_cyWWse1{i~VfTgalO|lica_ZbkX~xg
z_mbM8l5ax;M>*1Yn*`1~>o+gM`Pz)FLHk{r$*74%%x6W4P<PEU8yrcuY6*&p$DV(p
z(gH|qW+kQ_SRIeYvJer1d>!Vi)A~ZvIaN-6*?bAavgjY|TC)VfnWe%(TU5%krG=bo
z&4x4nrR5iBva?FBHeqbe;c?~;4!G*UIHko9cMsZ1#@B(YKD)Ei_q=q(u)hGUIOJD-
ztA!Df>CbvNKHHIo6JQ}PJ%ZT#h3$}Rc@JoQ(x?9}!?x($IT}gPeYZYOyJ-NP20aD6
z*gLW<(x?3{o6l%^s1|!qdKe~-n;!3hE&LV8pw(IRP;t1U@P_<qq3>Ch=F)Ii1A59j
z4)i_T{PPa=X;Qx2+hytTdG}5Q4xiVYom1>Qc9SZwWg(wxL*hKvuO1T-_p~@&6M(Fl
z1r{_Gv|%W8e)FSbF%nTb?P$*jcAXLCt*G7@%c2j+J#$uTHVw_7Yd~$wIM-B|M|u(T
z1ASU^12}Po4F9I>t7t~C=MCKRsd|c-0$u3*LSaDYA306rxd}SCsILcQks1FPo=InF
zv~e|#6iw$n6cZ>&KGgiu#c&=~2mrNO2Ip_eE!1(<{PMs1-c##Ev+ZhJ$2x9`N1>jY
zROcZ9e2&y=E_m)M$qiPz<G#1^>r~Zr6>>6kPjZ%z{midAeL6o`Z(YEL5ckzau^Lf&
zW&~9^0ZB<*P4WXMTAo4LS0PO0Qpw#}I9=yr74(_~VRr)GecdKQ^8xyOih%M?8gbvY
zg786@>6-g!8Z}wtZv>~NuP>b(bbg7a+b#m!TVM*zoCBREhq#Ofr=3C1w!!-Oy3vCR
zTqxzv^k*L|<#qCh%xCMHEonS$H<A366VZD@@ScCj;)kdhPm4{{S4MU^mu+h?<W_R0
z%FCL%#@Q$t%5Pzm_e#di767+`G4#*^HQX#vamx~#Dz>)pFK4tf;xqV!{l!WZaXay?
zWj3C3h2zY*lXxpV4b81T!?s(mOyV{D;*Yz;;tzzTPAJ#d9U8Be6uLIx5LS1GRBz@s
zH@7??^UoGF9(GAxKG=t>YjXDsZ1BTP<=-XtQNiKY8Oqx}{na1bo`B_IUy!RkPiFaL
z*D|Y4%O1sMk+Fb$w=M59>Eb>Z%Mo%qdROlQgb+mW`B+P)t7vj_TqQC~re~W0y&5vP
zq1L4ri6Ju8@kX~k=5~`9&8vq(DXJ8`oR?+i;l-p9WbNZ>OPV=kq%eBq4`aU?0q?yh
zyW6Ue&Y_3A!#$9FIJz0huvKK_U=Pz`eL$u!$fz!&?Z%lL5WvwGC{+7{j6(qHZ>zal
z#=iPSD1)nBN+SXp1E6bCI}k4s8`7G5GpgGcgmpr!K@HO5pT7J)p?P#_Vhc4{syO{A
z{FWF)8Cs}zaYk|)T~?HLlGTc-dMef>te)ykhDTQsXsuU8fcF4b4VmLC5y`k?&k#{c
zHemNgCeA$gB9j+!#|HvW=^ZYWOg7_7_yyWhU2Bc>KuhnG6L3R}*Rn_%G$B$2e(sTJ
zvAZS>YHq!!#A&IlLMH+fX*JOr{0w`Cj@x$EQHc6rdT|P|2F(R29@rg!+k`H|>3{6~
zR-g1a?gTDAs!TyQ#frrWvUpai?#<m^cDpa3E^^2lBf4ajl-8^eq?C+BpP&!)xujX)
ztsCi5z4oxF255Uc1%@&=Y4mwoIsD_Og3&PwMQ*AzIeJI7#(3~%7z4sA=qmX)$9^(i
zX7#WT0@e1-<4^KLAw3u5LULEJ@(hie2&h!`DNZCUjDtOqEbUh|4&4mi=!Ky<@SEpR
zKJTNhO-vyN2?c@%AzV|qG6YfG%F|C#)_GWKc5@A*bpT!Q$CqJi=t5-2HqSG__fMkI
zezJKw)b6;E1$RQL0SJ;da7e38&Nd*#;g;bJVjsJ7sLH!^nL#(EKyVF<_G*l%($Iwb
z6h!$hv_3F|U3yL)iw1_x7`v?|!%0FD`9_61BrRQ%iN@4W->#~}K;CcOuIqX?0ii$j
zMlf=X6%Z`BD3Kw>Xq7PIkO+iJLW^u-=eGlRVNfXq(|{V0-w5AChebs2D9X8eu4BO8
z9vo77CNV@+M--zA#9}JuIFaa?N_E|+2z(~CvDf0<cJasjy;^hkI#96UGQcNgc|p~M
zc$(s<LO6grw!+aq4>9qgPPHLnc9C4o9as@54DdnIS!af1?)U0#Q7iq@4eyfB48=ER
z4;(k}Pu|fVmg-Yrva@W;<M!zLP1_udQv<@EVrN)|-~eB=IJtXdg(_Ex5g{o;VSNq2
z_LQLXOzCsR)q;!*OXdS77uI>xZC!Jz*x{3JsTl^r6d~xyeY+osb4DpsWGwSETYBz%
z0^E;tl$;pwo>}aK-V1q|!Lda{X#YbX7d?S0SP|0cKwpYvRf2=!03I32o*Ty6j5>CD
zV@=e^==nI^(#?45)UR0~EOi(~bDV4guEc{rhgXyVt(%%Ou9}r{16IjBtZ#c}gZ+|Y
zoN@nM=bRi?g+b!cvkFg{Jl<>9gr8^|1F(a;6Y^1fpRB$tVdWi!eM7Ey)XsWui&=Tv
ze2XUfl&6O3a!gm^PJAci0=?@-N!{~0*;rt#ByTd&;%dz^WkcJ0{>C3bwxKAZH3OPk
zU3~2N9>6p*kas-p5bYL@%h=z8tNbfbvnzMG>Bt2&rr+Dd4Kbo1%r)ZDGKgso1_((e
z`UG<C@aEz_h{Q_`BIBrXWsni*Wd_C|Xo*JtAtU8p>%po4eyAxrTONHmhTwzIrQFs&
zQV3Vp0@isz8G?gF)hcZKce}up339;*6M-x3&<_H?wr*&njms&sgkf@@*R{b@DItC`
z(D;i~TySJ6oSjEt@Y%F~3$*iL1K@o9sE)Tot`{z<JL8=)<Um)wRiOckEuObg|1!l9
z5U3@Pffiv)Z^-ADw+w&jI63UN$~lGF-6<VcvZ!qBwYzD;uf9+8Pi3!&z?HJy(`cDQ
zg?gH@=k^atd+<E;V27xkIC>)QD8pwZ&Z8X1xnS2q-P4|0WUK`GRnugg34qS#`M^^d
zEM>pk{7P&W6kE|P&wnNc90;DqwiCulw_RiF=&+Kr?a=UpA;)d{IP4L1mIZ!0uaJ+h
zD#N(h(mug<`%uz&X?x7@o=#~$oM06g`!F|4LADgvxTm`gxF6>Wg%~tx@C_}Wv*r2w
zV{aq`oj#cK0^~vbW3O5MUY|Scmz;g<8Mc|-;x@8^(5qX2l#*IdK<<qb5AN!P`Znbn
zh;OaS`3HID7o7Mx^flfd(^T-vYBu(G2`&mHUIr)NQrp4{i-$$_!HAYo4bF+9{!)t>
z=eyuRAG+X*Kc((387eSC%1;G0)E|R=PZyg!1sa{K5D+y30DRu#q$T0AHb#RFQ(@Ql
z{laD}gcBReIkh*A2!?RJQ})KrX{7|6dhztdU}Z}@%xhzOjB;z_7+~!g^XRnx!kalU
zo`0^usE`u`tN7iLtU3+JHmhwdcuR5ed6UwjDUwqODd)8i_J>TuE0iN|vvD!Rey2MC
z3jgbV29T3*@j(X_U|5yhKW2G=&v&>-)R7;T{JSw<uK>Y*$FMvkr}kfq*Ti4hAd)2*
z{%bmM8k)i;#J~23OWtpd{1IwGV}{8+S2O)e3XolMde3luYPAiYziWY#!Y0cT71#E&
zP}@Gy(g-u5x)x*Xle}KICN6&!4;5=pu`=Y;X8<x9!tG#>w==R(-DI8E#|KQ^)4;g3
zp1MZ&n8Gv_He1!sEZs3iYaz#<hzaT;tIo}klq<ywR{dgluH}vz6{KP+xEws4<VJVd
zC(Tzj<9J{W2={`ROtH(JOKt^Y+a}zAr|B+@I2+HErTQTUobZ`Tf!Tohd2Pr!(1)cB
z;sL1Z8uqtv))foaK&OxT-8p&Mz(C#_tqvR=EJ!*WJ{4F~D9d85-il%P=BRY3c(+;}
zc&3lI(3ZIDW8}E*6|Z2Bl$qL?M3N!PkVnWl$rQF`p$)}4`!lEw86MQ>#hr1^L1V8Z
zQgUff{N1}#hz)wUylJmmBkRscW+V@YBY^jl2|pT~d`LeE=?@r#8HDd`SLEj3nR;Ad
z8Sa#8eN(S)dmM*j;j9J~Y*A>b%)smI3ho7u;Oh0x^=nO&DtA22Q!mY9t0`71cQjLN
z8{QX&ELYBEK&DZe=FU}jYr;(5Wesx7%PrXUEGeMDh=C7_Re0KsIDPnjU=o;<Nq|UP
zj7u?M6xrBFs)YFWwiU>c&5DL=nsg-TaB<gACKoWA0D4~^=8wmZx1%0u3cqt;0=e&(
zOWBXVeK?Kk0zOEMyq{n9rzz|^Q3cJY7|kwmj;Z)@l0wdur=*ZVs!a2mSX+-C(9$$8
zw23Q~e8zrn@<N?gk^wG^pTMWXRDge61*}>Yy*D?_%~+ZoUV6iDs$XocEl}0hs)|I}
zYqK2IYZZdxWF(HY2ODO>BNh^$JXk3FtPoN13D$D**UQ)}7-$1R=Z{)~9wawdJK2q4
zp{Uj|do>y!JT<S@`0t^)SW$u3*1GvHQW{nb!2}oJcn^u-^l(r1lO`m$Du4r-J10VL
z++Q06g)*b5sSna`iN8^&QO3<o<PD4tH6%4z4Ng^5qtKQj7oh}gQ$}>fHYG!bPM`uf
z&7JJbPQ)AFvNy;l9Jm+wepv9BqPcPYW?nDD180&4d(zw)rff4Bh1v9=An+5S(FE1-
zGL3}?<+tARC5e(`ZI157Jpe+2dUExAer52mX}WHsX}d2^;WP4f4J@0A(`Uu#Vkh6%
z|76x#bs#Kt(6QdHI8nq5pq!8QyfHr8I|O7xCa{f7Nb;8@Te^4~{!31Q2sq@K=P95m
z)2NYfHuIN9bH-2u67o8viPxDP=5PLt=!+MfnXG?#t(5)tt&7k<W&mhsF7vuA!@(sv
zcvMfU+MZH{?rb;cilKryshEcrIFQmB%;=f$<2KFrQ?5j8qDE`|>IFw;2T9l5!Hllk
zFd+FpX>kYXOZR?~*{34a!nbajCdY|_UIhc|PI!@|cDF!_8C;RvvPtTNkFSNEi(;*T
zFb?mSoqGN%PKFF91p#1H5J{6QwTOm4)3y6m_^v%ZO?P}E{ewMpDG{~ahTH0ky|r2F
z%D)J&CezEaFK4m_H&b&gsS!daVT~%Vq+2uCZW^>-5;<<O{Z=zY$O$X6#Se?ekr@3K
zsLD!kGGv7ek;bET;@R0^R^H~8!B}~SXWA45cu{PIaJU40e*^%D8LR<S?}-Zdp#k;+
z^s2xt1Rx*OO^LWHX1N4Syy5~zRmBj+u>+NI?4veAB?3p6&<`sL!coCC)8fWz0%RmI
zT^fVdMwl-ZQjJWbfmsZZ9JOWj9LOuYs60@5EtZAq>|?NzFI^%Cb$QB)0QAc9S1(^F
zZmX1tkog3x(SRZoA{=t|$>O#ZWnX7+p^PYIQ`yEl4s{-|@xw(4ysRBN5Hz3ke#4Zz
z4(#fxqm|%bnFGhHl%r{JTl)0F=thC%J`^+ZFubkdJxV)_Qne)8?0B6a&Z;^eLHDIc
zYP}{+|MzMYLz9_`kCkc8f^H96r%lfb%COmJyQP!oT!34EN?Z`dnI4!%zFPo`Tj*HY
zV;Gy3ffgv%W4{|5P7N@J`EvAg^}jV*imXb^@?R`zMrIH1ti?omO_wPzr@Mt?P|V?_
zr@2=tKC{k=(<)qJ0iAoXSpRBQu7Qu2`MmLZHXl^KJ^6inIywSX?#~Pt^`9ryX4OCz
z@YtRhx&TsrjESUD66f|ZHPnPfgurEflj^I-ioQVsURx2khwA8B5TSqa!ctyc4^~y6
zuwV-l2`wQ))+ELtJXA|SP1K<i2%gO2zz7Z&4V1}v@o}e-p)vP;lL-)(`%hP^at;^x
zBxf0SUWcOb<G@x|#SguUqFdquXD4uR18{0N<pEuzwx$tG46#}&v@7Anv<ctDa_s@m
zwwD7%5xdb!1{O-WILf>JZtN_d%@^6&d92ebF}{1e3{`1&lRDUg%uTCKMb-;qTcNL`
z`uQX5B)9~?G_^Ve#;r-~?r(ppQ6$sySO4PC^Vv`y!)5BxCvNn_=L&rPIMY<%l3bOC
zdjU)$`Uc~Fum-`}7ZQ<HiB;T{eo+>Aw=d3RIPn+8@yYDUN)H$8G?7D`Ir&8?0p&RL
zmc!SAzSuyJd|2+>K@blX9BLkR8*~yqHrjh&$U9GUCC5BtQf``h_;s$hq#W{&;=X&)
z#(>c!X|;R|JH`5Yxb$pPFT}c+AvCNRR{>&8j|Y^m-utUh+1J1~s}d=k<sX`_2J>sI
zI^b&wxO-_n7N<<Nmc4~%DlR=!!t*Xc4wvj)Y#<p<OXJz9W}JVmvpUmysa2n!K4jUl
zyKk={B-dD1mB1}fK5kuBW3%I3Hjb3HY0oan=Qvu`u0L+qztA>??T;nj@|jK!j{^>k
zhh5f~Z1B}nyW3}8AFqTxaOmOg{*V(W1pb}zIZflK17*GnZ*_|LiO>9`xLp?##1`M?
z3eTCo3`uzuiaHI{UNe)Q+6?-TQJo0;V0iBf`zWm#*S<FUklZ0S9_QD$(7@zRA9vXB
zDOD^^_iXK&FgZ%4JN`18(i*~+GX|_BeoHQ&WV<YfK67l9Cb7W5cbZJm-E-m`YxPXq
zELWLFGC0*8&9ucK(w7N58j|jB&PC^$vD;C|xMPXovAK!=MS{2&-^kLdSlWV`;sxsH
zBEhYY(29J);x?k(Zo-9VHx;(AL3$5@#=!w0A6);MV=G*-auv>QLHSl}g#{D<1X~|S
zwovE-;1AuBT@w^<)RyzS=8ppF1C(Igl0K+-Qx;rDTBkB`I7$4D=)@8RhQxv%py1P|
zKR?|3E5z`?4?cYtbm1es%IIG6RqOmC$=PpnjDoJlo^L^vV=Fj0^o7qUtl9@m_gZY8
z&GK7Hp)Oj}_uiIuU$^$4)&P#ZG3VNcm~w&zv<t3{^eqHZP-0Dv7Cp2ZiHKLf6tpQ=
z<WUYPzyi%=!ctb*HX)D7^A;SpoZ>6f6IArC9X4e<GBJP&H|1J}+UmE~)tj=nI<`US
z5zWFU*f-pM0a?<ocGz*fcE-iqa3x=`9DL`P^}(1-=rzO$o8Z15ZxSxo-4yudp^)hG
z{E{dC+~M51py$RQ8nIV_`34os!}b2pa86N_g+KveW=!deAfQP3MNF0pQoP+#ModNl
z48}(uH`1~|LS_XGb~^MC{y)zXje?-0yohm80WNjg;{UZGJ-*YN{ep-fK@k@b7e}>j
zr!eA`Jg}`Kvs{_O)DrKh0fY1-1V$$9u&|L3V?75-Fz8PC6+3p+vSa`8bZPqXb+-`W
zpJ9j|6xVXyvjz6qds?&mP&Ax@<Znr4w*%Pp^*GeA@`QT;nNG@2$$L(Yet8mwk|r4p
z0O;*sMgZSxMHZxd<gSp6`>_}3AG|(f#XlfF%#)5tj&+gZ6`i)|_k4f!9~6S)i8^|J
zf8AZD95xC{L$F--eE+%SpVcm|?36Ub)N1{_o!PLb2aFNNTTE91!~s21CR)Ri#mni@
z!xJBe>b~#4ek#<!sZDQxp#*>33~)N10PMcoKcoKLE6u3IX16VeGV|gT(&It7%e>}@
z@sb?lO1N`$`pj)wBTJzCP=n61<$tSre%?LE%~}X*9TA-3EFNG<w2LP73{kF$q}3Bm
zc*1sdD&7ZanXQW*V*YxDnsQ-@9zFR?V5P%l){1wKj(3^=Gh49|f%gbPa2pDf1@PGN
zmMsO_3Qt@1SfufRaS{Q3_~RRExzGl4F&L^JJ2Ud!uM!VR(M&oaJNCgVxgB<43C1-F
z!XS88Q%A;N@^RGA+^8&Ea$Q(JRFvwlZ|QXJo%5T0MqP?L>4&)<np`o?!Guz$c?sMX
z)zte$kQEzj{NKlRw;vkcw4db%DFFN>=PC}L8gK{>p7DI=BQh@}MpRYPB|CQK6uuMi
z`*)`Uhb4W9+<ARKoO+2Oh9Rk+p<78`5bxn5c{cSGK<M93^l`Cph*@SJWGkhO^U$cY
zd3M~AQv1HGPAlfQj`!9Gt`qw6sLhy~>U?E(zp(^sh*)Xk9_lrQHxK;4HNZHU1qUL*
zgp%9{g9wkp@|nX1b7v;FkEIfFZ)tOp<@Z)N{;}p()8WgTWOwGUYX-f=$mTP0cW#1A
zz_|C{F#4ile57i_4Eiu5a`G`+(+nt-O~2}0=`o=OLjUTbv|$Ddgn%JYm?z$F!&O}L
zBep?91}tUAIo(dq82b=*Gr;6-&0bVd5Irp#ioL_WEYqlX<DPg9y8;J18vg;8Ju;h{
z_YT7=7Tf9#Go0l!M4-kyR>xFEe5G9$dyJ6e8L7S8!Ny(`{>rbFn_&kt6TMOKl?v&B
zz%h=D?Dkzabh`1OnQ{;4-!-|*9-yMTSj!&Do;7gWC;mY(#R(ar+<=wPaRujfAAk0e
z+shjWR#4|TH?Vwuxz~m=j{@?T*t;OLzz~11xtYpBK?UkfIl3}5RhK=H<jgu0?_$DA
zT?EtSZVQ?cD&9-@>D6J;XQ!i|0K4hlR;R9DW`ndtj{=h2+EB&0^>pBpzO+=WK;^6R
z%)cO}P19K1@}uShn*gvJnjgewfPjM#+Q74Y=~cqc{x@l9!=0&UNah1)nrMh_Qyf`A
zWn7R{Q#1fGzPg_%p2Ngpicw`?bgn9c?^Z*#Q2$3;83HF!Q_>P&)=V|gBeQT)(}FuP
z&R6j_U4G=^<a*~Tx`Bnluqa#~-}qqDtrQ5$1V2~ZKgJX;901#*Bmx`kcC&Ew`MnEo
ziLaIkSCnlCvyl#_{s7kBN!f@i!z?B@eWrxqj7zZAO&Y9assfl;g7ZYGEF`uA^(4Vd
zo$j3#D&?cGBT`d2;f*|4f|bD)g<w!;a`p5f;|%6EOml-Q=I8!Ig3qPd*<RDqXQyD>
z=~hp)t$mp`d%*58r}0_lF@w3BamN&v!-3~Zd^#wHfzIeR2HqEsXvn?y$<1Ws`LXpM
zGqi(Lc2L!7Ub?eS(HNlAO%exs>g0iXZj!8ky3*E4QR7OjH)#I$-(ES@^5=yG*1Xl+
zHFHgIksW0y#qHmfD~o;o>YfCfJWbCB3fGI9r`#8o@PIV5EL(`NOj#psWp2uR_J&2q
zAQ^jlGi3}keaPyK(oA^u{DC`VA45}G=7RDGHChLk6lzy1T9GS5lr2il-x2^c#d@m5
zTh)uaFe0vpnZoUKP4~scF%(S70^XRcUwLUF!{kI*Q5vWa&@34r$1ywh5%yniv=zcq
zq&*t48h|^|du?3VbBP_I+RiR3{J~cu?AV^Anp-zl$B41A(HFjElprxm96QF)UB$>W
zmSLZ-kL(Vsd;!W2-R73haD`}Oq***s+&BsLL9jVvqB!KzWUIe-3cGfk>HPgC;rc?%
z>`4XA#}bi{f@1`A;x?h}4B^F;`#B8vYwSalL4ce9qEt@Wl49cGU&5|LQ@-^OtUnCF
z7466JbY04J%e1Yv2s>MxU+cA@gF)y>1bNPO(MGNJept!v5<9VUzHbhlrhDPba8lt?
z>_?hvd|K|X_u6WFJ!!HAE1PJsyb1SJ4tC5cnh-!sCo$y0XosVA)Fxv1EqG_u+gjlM
zj)01aZLOh5L~(8F0ts2R8F}!W%Rd#dDr7_E`<I4D+~!9asE?wbPS9bVoI!63?cs@|
zdy7s9DeZ<eKSJrQx5NucrS$Mt!u5}f+Ngv#PQ!8+Yiu*x91OKiu2n~wFg@1?Z>%8D
zBda}8v9JV8iA|6t)`)L;xX$xzU|5`w9$@z#d2s3EW01nf8&e*Nr_C{9y?=r2C?~CS
zq`J+Ul_H+%T^-usF_>mrrcn9#!6@1vrPFrmQ<NlgJO2L0B62EUG7%r|*l6YN6zT1O
zVhNZ}Qc4LLO!8Kvlx-m@KR?GDBsywF@=fC_8U7e4Q&HB~p;Xvn^{6_+^)a%#1URu&
zPco8D$!lQvnLrFqb4n57Bc4Re8PYY`^YdV6r57Z69UHMQ4>um|d$>;R_ch45O1Mkp
zQWTd9=xcIQ=2kALt~|HAy>?D)KoV@6m?hJX5%rD@N)t1+eR_|y4NB*m==hSe;3w5O
z>+yLk?WZ?NE3K+Bs7gV(9S<uF14i$RzicL5mZudvS9xTlrwzS3r*jTsUhrd>f_0Mc
z(ZY?lT6fOJ&7=AoB!}KuKg&ya$Bv@SC_M!3Pnmx|i<oMug}we8N!1Y<meX>_ABPu3
zty+FrV0Jbc9mAp)jW4$Q=OHafw_YgT-2Y_O(KN9*Fx16iswf!#*|9|B25<}aJQVY7
zoN<e4aD4YKE^S@H*bKR|WZ|NO85z8v*g-Po<nvwkZ!1c8=UQwXgYHVS!<mPPMTOGN
zcG>|eJIk@y0-da`M_5^SyWW)&zx}{=9ZbUI#atUyB~UNmwNu2A6S5@2X496HB52U$
z2tBv_v_GKEG`xFt{TE780_c=2W%a5Om>+Qns-99)$CS!AtTt@GlZW@>3E&I9^nm4B
z)tdUZkW(463Vn_#*R)Z6bZKbg#nvGhl03B?^Xn-&ji7Pk9@C|kb#(DQXD?<ITgzOw
zX`gP#5N9L#NLT&d=ejms32wx{V?ze-sm~ib8h+AK=S>H+flY?yhuk=jo1w!N(1Bsr
za+x1O$LyAN+voDi1CQs;SgBq87lU<Sm~GE><Ml$1ra7m6KkG)%w%wfQgb?Oc|4;{r
zmXNvsQm*ptJi4H71sA6K>d#G6h0N-&pH;d$ki7!stow8SHrgpJz@TYK@s+QXgQ~^?
zkNxNoCbS<ap_Z!_1fYGB49c30YmlTb)eHkL<KK9dx=v(+4ip}W`m999Bfq_<rim^J
zr_QvhORa~}6<2XvCeqcm`2#eLJ}J+&J(}OVrzEY5MRadU2&kUlY{tf@rDpEqwdlr8
z=FvLxd3IL!q~ve1+#Ktggb<Y4|DFv_dfN(j&TKFk;Z3H<aTHS!lO9FS477j<c<PUH
z>+eaFr@_de^(cdgKhlEv+Ma8{98oGv{-TN7&|L4-!Vhcy1`ms?t^d!CMq~Q_5Z*GS
z08_(JBy*9$1AtqoQ~nPCYe1C0swCN-s<|ckZ&NuXnYm;RnM+<j-~tjPyW5tn9NMzF
z*&qO7`2bS#^!rJketx?Cc>3wxz{!hXmDi^qznv65koZ?gyy@)8D<zy!ryqZuetnTr
zJG`!Vft1GJ<?;W3pZ5H0RKOAHhL<`&dr&WI9A5ezuUn-7XYXEr>ic-lXuK3&4(UX#
zWuoLNHGls5k6)f-Wm<~p(!c3r{`U6c=|4}i2o97`PM!Yo$5VO$MhO%2$pn-dPJjOE
z^xM;4W~y94L7{@AbJh+D(nuY3=3EpeJt_)9IVGUbaFNoYK!5^R6w;!S06!rVVFVzV
z_UF&vPt%>h{@mq%@OVeLc_u9fwe-<JL~ZBM*5aiLADJ(IHy`-6P|?qnak01f&=Le#
zHIJ6ps-MJ<yPpTG1i1KF9<&u+!xj%VRzBYA+T6b`+A6UsdXE)rg_WDE7-U7>%B9}U
z&rT-K$;NUkhSZGceube^iJ<fc#&aJ<q@<UHRd!Quqm8hCe$i5v_dUO{IFt>-pcfLS
zP!0A{77K8(v@f4<x!;+b!(gz^vgb@7oLBSc2RM_8Rzxr}<y~j0t~ZX(hHR@L`?LlW
zWWJ;uZ9r(FKg$yaPzakm!Lnxe^Ws2A^=bXMAn@AM&jTDi8sQ^+mX{cWgt)~E_w@gd
zw_7iq`oOq<?;@DoMJ>YYIXh-5;}=D3IeQ9An0+010mk={RphN?y!SaI{bDz03DOE3
zZA$vT{#~B_i4T84f+mL3fAi2k(r_gr#-x9r{)PzP01Rnwgi1>)5i=lkAidfe0uZ#Q
z5X0xv|Fg2%d@it!l7T}wQzVn9m;=d_HMkA||1Y9{t?IE+WB5^zQcsFb7Ku7xb!h6j
z6#<j{DrOw_tWiPfQLGNBYp~!$g44%Cc2$$4@}Tcg&5>AEivTEA0D+01**<JAKxz$A
z*bpA2)|3###{AEKKso(uQWUBv87&L(R=zFSBQyA2lgVqL(XU5;-j)ett25@oGA;8j
zGF^**Ibd`MR1Bosmvn8*<Rm76t|xaOlXaNa{exv%=HHabb(w6v9rtAdjHI&b$sfoB
zfM8cYSf*wEO_`GIsT=^a?LmEHo3|`-8HaGN>na?`WzebJqt|Mge^ahF%0=r1&0`k1
zG$3bPU-3XLU6A=><yz+7m8;qy4<UYACZz>`AXFuqmjjtng0zobt7ZPrtd(JFwc}d|
zsv)#=rLOyO0cN%+AH7z~{2#3qFqyX33h4=Jx~^^qGGSVY;2tZ}I{#<l1^Jb49<zt_
z=H3oCC6IoD9Pw=IOD)cFOUfs39V<?=1xCQNpe);(R#2fz9$DVaZ(7;GM;?=&DWcnd
zzhxSszLSa$!@$dpy`S-3d*9WOP<e9<Fy4r6zFP{HdPl#q9!irT6&P8Qh9XogGSrB?
zNppahR<@Ngq?cR4vqrjH=$>^hPeR40P~_XxNqz(9X{M2-Y#OIKRIXBKMj(-Za~bLQ
zyLrAw@iI_zM1?;utB6nD8fU~Wm&w(CIo_+r%dD&jgDamWQ<Z&v`<l0Rs|96>gG{qd
ztZtOK<<ttGwiRYQbg4Hi)$&yuR%flh$O??Y%BIC2TlK+pL^aCbIo$_1XXVa4HlxTM
zn!LrAE>&~4!NaQEw3VD>EM-0)oz^K|l_%7>QB~9VP43@Kg0Dq`=8@DGYSN5<ItqtY
z!bebC$c=$Yohn7^nrHv?Zo>(~ZuHJ<weT?j3kMDB7+cpSQWRwC?3krOnAfJ<If|xv
zM5E}KHtayX@qMuMU;>u#-8QM(hPN_jj*HhyX=f7yupCnhCG-WXGSixw%ULNXy_{#A
zldOhwDWVh&OyBf{8_^{AZ5X+K`F`#Tq<xRKZeIE7?paw%BMLXM*3g|&4)c+o8Glwl
zQCe5VntudNIJ02R%C53{*2%)!Rhp1M02X}?KV1Snxk0q~gg1%+QF?%7(caH3RZ^f%
zE?OiY1EJpB?9j;!>&P@JNgm|nQ2lv!bU!+Fo~yiL>ZnUM*YvJ3*Y;$87|T|Ba6Gx&
z`D=#jjnL8XK1kXlD;GI9TRoNqNkwj388hziWnB%ttisi&t{5mqb`l|dmn~Li+;FWJ
z!nckcJ@vKjzE*T0r+e334B`e1@}b9DD;Ss?Va}nWYbz|m&yhfq0CKO**n$g_>4Q{{
zHT~35QcBI-htEMlkopsUP_T@mbXdNnrA-Y(GytOV>C{`ysi=Z{tI}Y;Rp-%l|F)0$
zntzyxU|r9vG`mNg^egVs<zZSftR!IW(RFy>9$S#fAaDlB_PvTDkm*mZ1Kkm_tH|lz
zZP#_+KG#7H0N-2FZ%lJ8!(i<S06O#|6NHL!rk0b4Pr5bntaRXi=`G*RNy*8o(2rQg
zuhRq#lQ1mO5M5#ah;6N9-lm(VKESuNcETWbn$_!AU1FH&SkgmYLcF71(}4zCAw&xV
zmZhhnx2*)0K3})px6?I8sjtx9$q?GcsnMQOAidMjyuQ!P!6O8aixKG>yb67azjGbP
z5oqC$A<#{CR0c7Bp6_++1d4hwRf#&wBB^%nZspTkbD5}}r?{`!PM)Ijy@Pp*{e_NZ
zg-TZMjTN;^lq0uuWMtRU5<@q<yn7Zy0;M~`pE-`Z5`%K?xYhGHvgyH%e71#f9Cmy~
zX9F*e;lW$xyv$T{U7^p4ugztbN?51oy(aLqB}auUE1$-H(y$eres2u|FG&qA{;u5#
zxU4xZt-y*R`sG_8>7xvw1D1I@az`-D0bH#|Vd$8+wjKV{dz72Rt-|;hXW}D;(ml>;
zQ}@FVH7e|;nh=F@Zl;{#TET4D0C>qIH+l}=j_*bZ9paYqT%47c>pJbK#dZa98zKbe
zSUW8a?U9OqwyISf1n)Xx2fx%gTktMi+16al=M5Ti%UkQ6qH-UEc2;c6&9Ksf+^Oqo
z(XlDjX0ouDV7*z*V7qCF!5f#*b=c5hy|H=M*weQR`(OL~nTN2^%&9=eI8_sSe_w}z
znq;kvTh$loKECw8I^;466|Xdnu6U^uT}5~uCat!APMt?Gy9!RfRk9n4V9zW{mF0M*
zue0alo5_$a?Wkw&p=AVO8;GWp3)I`a)%r@Z=e*Naq^pL~lFmWOD+^={D{T}zWI>(v
zO!ouKps`}}pq3+MbWG^R3b5n|Ci0k&8KbU6ejLVpMKA1mZh5*}d9)Gs48UR#skb2;
zS3S~yEIBQ|uS3s=dw7D(Xs*=Kky*bgYVKs1_~lMUX1aAUzNN5gA6Z#B;l^4{0V~JK
z)4DIx%COtEeW3V03`e)VQJQIO@S(ZGzbH?0pMjzOT6Z@(06DIY6y)9Ri4*kT@7G88
z3`_d2$hd@I8(JdSK3M04vu=^%jqj)B16D78LpMLL_6oTIgM@iB?f<M`kd+VFhI<M;
zZb<>Iel%w0R(|l7LRbrn&FI+@YVtql>boVQz@G4Z(6XcOIa>kmgpBH5zJtz<gXygm
zXy+&jFY|^*W1O6GY9mAeXlicDy;MZw_8mFYj(Iv~lPFDPO$K8;Uz8n-3-up{eGl4y
zJ#xoQ6Jo?`TsyzJquTa@sMX!}C~q@>vKO|C+IPf4FDHj=ITR{X{xB)O?d2E`gL1Qk
z`kg_;4W806!0Wz0`zR!0?}_Rox;E)(gVGDXzUP*#8Fsv|Tj<7a-LQ|v0Dc&DF#B3T
zwU_+Mk;PsNRz4DO*CO|zt^K%Sh4*TId|R2T9*<hjJsIVN$2h&M>&22S2&e2LEV=DP
z$y%T=RM!f0W;4C;XsGux%&6yc5seSCUEU94EicFDEJb=dH%J<VUd#4oI+Hw-q<#Hc
zab;_Ub2--0EcItldjC9zGETU#bju9P8(s!#ek(;Bbs_Vqgfk_xGEfeIj9zVj$0wlG
zv2EX=y+xa?lv;*e3xZT`p*HO3#c`fChV^B-Dyz{Hab-)knyT>p`v%E&Puo#xzWvO1
z?^<npVP+;5-mk3Vc5U2}sTYpf&J0l@(;NDFuPhjMJ>IzFtZ}iJfKDekn{)Q2ER#^$
zm+6>6J1Am#ccF|+I+rjUC#ZmbXX8jg?$D{J)FA@B$gc}krW?+s*ey@H^`354yc}{A
zjdaWVgIN81Pwu2&3yU3H5|<5UlJ^yw{4@*D&p=pXr>JN-^<!&%$kg%=6dON|MKZSA
zUVpYhqi!R1T#B%soflJ38l!xR2J0ysLay6wq18a@@@nAmyk|JBA>T4!UCD=vBMkS#
ze^OI<2+QdicX0OQaMQny8rozVRC<Uo@}3Y;WSTcHhY+!y)kjO%jpK+$gNf`wqHjC5
zwx>NgLcJ~)+b!tnYfVx+$EILY$+36794q_QpV9me^))bwm$88*6qnKPFcSkZH8YpN
z(;z07BS$5h0Xw(9M<tE{e<`Y6Vm1T;p_H>0F}A*v^GB2p0*Ld3^!1Y(OJclmj*epT
z+u^D5S!zui=}u1@^xj5)NjEjs%Vi>2^~$j@>69`E>0Q&~+}Z`{o-~I{j3k|UAA6j1
ze<R7K^rku%8APgz+6yg2P3`&yjb60FVf2)d=r^h#Jz>|@Vo$#(e>~&0t}}u_yeWH%
zI^%uAud^3G^Hs(2^QcN^Un`U)VHD`!IqF@e$#~|3xUI(5q;R0PX~O|YU&n%alx!i<
zFccm3krPTYV@%(VPee`%#iSZqEMYYDE&GG5(Mxt}r0TgKVQN$R5nGp~xyo|IYsCW_
zmW1)c3ipM0Xl|I|e|Eq_Q=Vi<|J5IZw80KhiuT3wqpw|HJn#&_^<VP0{jUTjN!ASQ
zRlGon(p$?kn04nA!8SQzWyO&PZ2ex(MD$G~q4A7#$FFh|BWIao$}>e%g?QGc&aUa?
zK14X6&UQjB;)*Y;zU64bS|vB8t2$fMr*bzJIDDngkbp<ff0Bd?VF2EdYJ8-c6Une^
z>oq}Va*q1#l8aSLYtwgX!hJ>124`^1dO!<dBn=0urCi#`Y!E4EgP|-nycH2r741fO
zvNx)NaK!9fjX`{0x8qY&f+yO@8E|@o)6?TY@@)EH1b90l@%~$t)TFOvZQM9gaTb8j
z?bsS-d8u8ae@|AZr((lSeW>w1n0f0Mb8;bMGa%`z<IXxeD5b)wwD3l6*pwmk+DtVu
z>u^$(xT~ZH1*LPt)P>}LkKLEl6iO?`OTxRKTwR%>Bh9Np@C7nZVs7yKIP>2FycQSc
z7|fRSC*siDsp?W8U1G{cH+YX^)XqT}pYaqr32pA|e^fm}nK(<(Hqr+4tB@211%T`X
z?Pmf6$phMZ$IuLV?Mr%bmF6-S7l=+sNH6*6@v1UZ+(_3-Ne1i8WdzrEqj=dgE>)Uc
zD_qT|a+JclaZu@)qXwS8lM|e_jJhP?2_P#`N1&gmdUL0?YVMG9&7i6#R}wirqt--9
z_Daf2e{RY3v}?r*JLyjtNCHg?7Q=7O?0U8zsOK8{E{$MHU@Y{l+~JO4y3RN^==;h3
zC@>QdYS8~Shb+X`N{JqMyxHKu0P|QusF0h3tPM_^b)v0|pc#zoCer5Ez8_k{+g$W^
zaf8!<2=R2%g%AT{{&XV$ajQ{o`G|%vjJ>e)fBncpunW~7tJj((N6KIU&p@_q?McW^
zx{&itGc1sNGdw0&SMyc)LTn`W8>1vc;E;o2=Qfi~q$Is(!s}<U9fd84MTL~6PK@%x
zMZA7;7}J&+E&Eot!Wfxf|MpJ2lHYGPa<7?JqfSqXSIHN9n0YpsY=AUR^$<6QYR3+6
zf62;IyQPgye)f`CuLek-7G$Nh43AF^vRx|=1leBSuL-if{`UphPRfHpRvN<;xGxH_
zU2D$?1(Pn^A7p!B_Xk<2HB;t`Vr<*klW-k6a9jL@LWd=q0=-w_YnWE4o;ix*x>`Zr
zh_+!ngLJ);^Jt4F&EYemt&~hFF-}8Vf5%2yiV#e@b@G|hL`9x6JKsDL0a`NYT`yVi
z{kYt|EeAbPtlr#uaf{ienzCyVVD6e^kDYpAow&_wzVZBC8PBP;hYWbwl_b<UEnTfv
ziE5qo+^%-UL}dm_HYj70FF`S==1x68SrhbMG>bAolIEt=8i*o~=`#I2A3(#ve+<$#
z^*v|z)W`dWuG1Vg%}Ze2qA*r=Upz78%iP^{|2Y|-3`;}W_R<x9N^T&*GbvzzvrJxI
zHzS6HH+O14U~c}x`8b5tBJZ9f#jfs~{?LL&BT<ul@OCCk6y6E!EbWm2$W+<4a7S`!
zQF$;!=RY~Q<ccW`gZ?)rmtdIyf89&DH@T!iGnQHR*|{Y+$-oY+J(vUKDst8bLG!JO
z=3DhH?e!AQcf+X-bHX>O)cY%{<+cKEUY8PE(8sloM-ousnYXT2Zi4>0fFaPtp#S1h
zSDIxuv13WcbqQx6tB2rM8o8~R|4wU{6>HKq(Ovx>z&hnYog#ZsQ{mZte{q-ELjXAB
z8MO-5Ar45zJcuFlMoObRTT`E#JCj672+6!&CqGL~A+t@K)3R$-chD+hE&xgGTyIoM
zuvCi#Si&7y?vz!PdU_{ph^i)g)>`JcuHp~XjjuKl)0ScA+3E3mBwC19JBwDA7NmL!
z^%|3y_hW7z^ZbBZiDSAVfAws-=$;8bDqlLO(Ls-!^qeP^x~~+!Jw?d@sFk!;Y|#`{
zkHG}k2<q{v1#4)XTXUgcy{4r>vcqwKTc!XnO)76neA>=oFrGoRb9f7l*#-%~GgRmK
z4h+ytm|;)1WELRrzsLgC=3H#e;q^qKL`ugZ7)Ac_D_{PN4}T6@f8{ZQ{WHG$qcp>p
z|1SRw8$2!ub_h@qjMcT*dxH6b+<@+z#I7Q=^l?%DW)W9vCc^&^nSHT1qF(oM;|7%t
zBZE*fqLB%VQ0r8{eH%$~)Cmq-b~r&*amGTTM1}obFu64EHV&n4VT~yD7Te>>yoD@Q
zNE_2EgLOBm7e))hM9x3=kR9DB;!4*+v1(uJmh1FBdc}XfInAW4{-vnjerDHgt$CZL
zsZz1?uvP4|ZlSKWmaSjEt@hAwH=;Az!Cx8*IG4c{3=;x0HJ8yn7!;Q*Q6<I!UAOa5
zCGH6WOXTU7LH#=u0yH(3;o2q?mn=~wD3>){C82-Ehu>d+IY?$&is<MoJ;q<3Uk`sC
z#w9qTgmP;7`gWifMk}EMOgiAI&Ggqlr*DTZBh@;W#391cAN=;$_i1SPm4z533Q;;|
zZ6`!P4p6SnMWN|c7NS9cSzN5~!@|x_wxDc-(8w}Vww9@*SZ&C^t?Fj^^b*u6uht7^
z^@4w*V~Rv6mhYM34t_T!qhyl4GhP_&&q`)(a}+*R^IYA08?~@Hj++l!pk|lNFBIna
zLYgk%_Z59o1{VZG^0(jR^bdUb3wG8qO#e+?-|94S`Z4{DwCfDvfCPy!zVYT1VL5?_
zAu`#S1TIlX6E7#Rk24#ih{~QOi&IBC(oKKXv7vG3T7Wc8Y5`I&G_L3{x()-Int10V
zIU0#h$vie1#61l<5aXy9kbXs`Nj+SigL+s<cEYsmBe&Aqr!rQV@(syW4$V^@ZScE1
zvR%xg>~ykh()H3jw@Xe$@VDsC#U%Eb43YJx$!?j<XqyYt641btF#nve*UwYxdPslN
zEt02ZMVvY3JWXnpisY>hpZ@**+fQHrt^^MF4~;<ofEuZgtGE2q;T&L4ZGYt7UJ8tp
zaZ(~;sKUSrLV8l`GOjj!oel<lw9+;-tv{$W3KI-KGdL~XQ%9yhw^R;@U~Mha{b(Bt
zQ}YaIMU@NsMO1h!{lxKQ6#b-6`@4U1#ZiAuhi}v2OO<BS_z$1XOSdLWC5-Zj8ybPu
zCe(gyd??b0UXl>jTbqT}Mi4GIqj70Lc(hABRnm2_rKMSLX+m0sLTM_`v3)}T!&<XQ
zIEo4z(VUR6Y~YH;nngQ*|N6$``WRN|+$oGu#=7awI`Wc4D!Eb{R?!;w=FWfYN%iA{
zw&e9pK?3a8=$S#)XkF;9HfSOPU#5n?ywJOOrC)YS2mQXaW`}e}n{T8-F*w93mJJYF
zmY}_eF;)=!M8NYwc?O_bKPXFRRFPnw@$>R<2R`0sx1QWPxqA#Fh;{F?08_~DwPg{U
z>Kpq2sz4i)!%wEhA<mV#N)Ug_<BVC**7H`Tr|;$|!3gIAo{xIb3J8y**W=|<FAkoS
zvlJMiA1z4=!F9Ztm%?WbxA@3w=Ab&LSQ+)qQnZy@4zWZpNjvwm^9&T8&F67-&3YxT
z?^q%3V%x!CP%{-+oRAQ$HRmrR<+ewK8mfT{yt+Ih0SRb8MV4(+ua19`${M5dn=ojN
ztMuZ$H;JySW0*S0DXpJD$~s@?HIFp|^nEr6m({RN_gIvIwjytKR=|e28L^j9Wvh;P
zibk}p!kHQ63nZc`AZd&4mx4-^kn@SvH%%U7ycK|0eAwz;7M|HHY*yKZbzNeI!xN}S
z&P|`XM5T^06d}VniY<Svh~u%bW0#=jt>RGL3yUacm>1YWtHmw!B?(FvxIAp}=-aqY
z@rkUpW>f^q{gX9mp0$iKBk=C%WP9dc@XVZ-@n%)t2v~$f2S2R4SQ0EOsi*sJr+5Y6
zW<Q*JDqOYF^vu1|_dcaw3z~C$T8o-<1k6$MU+M3u%EpYT<%EA8P=p9$POA{QYj?V!
zF>}K3`L4mwCv&x&uNzOPOWai*FciznpQOHB%dfq;IOJiV4sye|qEKd=mtA>d&KR?w
zs6OISz5x&@x;Sz&sK?yfViL6AJ1Kmb+Y_c%{3p%bwX}0<1g9ECY7`exoLOxjv%||O
z+dK{RTjwz#3?qNwJ5YVOTs9$!rNXEuW+Oe<ZCN@r4!>ZbB-dA$f(*q7fcGo~8H2nU
z`~IaMV>CMXn^O6UN$liIL!$8^r68j%W?*vy<611(n77M{oT|r;E6gX%Tq7DQAqs5*
z@HP1SD%nX)>c!QcgiPAn5EI*5-ML;sX9Q1ay^J@h#HD{Ec3J_F1@#`qos!}P43pmJ
zwF3zqf&5C*kR~tCrl9eP7FyJu-lkx1WCWCv-nkPH?M(_dWyX^G%{aU8U^;yN$XsPq
zQ{Km})l?c=+WXR>#n9#)8F4KrR18+p)XdY4T$Fppinqp99v8XFW<eEA>v_#<pFguv
zHlTql&TM~xO2vqPKlIRsp^ml)(jatLa=UcA>xhMc%Rsr`IR-6udzVtF+c!AcTn$_X
zPSo?fV^x^gjpYY6Q|($#9P&V_*oAQx@J*iKcS#N~&PLqik>AhNX*cpqIS0%cPMPC!
z@N!-}bwCWs!ww^KF)X{Nyj{YMrqe#XkfR)SYCnH?)Y=KH7ds;@YZ-7Iri4Dbk##vi
zwe*q7amb+aQKd$ZZ(Up-RbG)U7|ef0f?L^w)hQH>y>CPQ%43UiBh?M$uP}r-uQo-n
zYUCGZ1Vt=pbxx^vPZK_Ztd6wcjs2{P8YmR-kJipk)UtTGmK9c15Y9p+<chiKA(-&F
znwx*m(`s_=yHmHmTvFaNARxEsbR~hV9MqG)raZwdr(H&S7Fg+K*pP!kwdc@LoeQaj
zJE$eWcAYE1e$bw4_8!Bhi@M$WBv7sq(lV111ODtt6w%?SMdFr+8!qsWBowUio-D_<
zyqA_!VZ%n@?Q$`GpPRoxdYj><c=J)58wr0a0)D9%y<d7an0E;1Bvj96^79pDo>9a=
ziS`<rO0@NT-I|KGWOnw*o3`}p7ez<~gt`VMaBd0LYjLTNtSrr3P)tm^`8<a*tIqD-
zFe4R6Huf(di@s*4R7Xr+HQisjJd8LxZuKM8tuR9I*1_0&vD5`Vq?o*15XWV};1qw9
zThv)@xZG1HJB>)%w}mbfcOvCn5F9Anf*BYF3>3E{&6lCZ`Q0qhT4r>Y^AhONlNqU#
z)-5pI1r{mMoT6W>u+*>hK>6Zm4i@kc_D1u1EVhW`H5t~kE6TTTU{;ewY%km2Mf9$v
z`T@S5lgz$EUq7Elab#CbL8&|G>JWc|<qwo&x|T-01>{zu_YDTb0ypR#biCkl-=1u3
z1f|0)S7c`@D&ayHDP03$i$Bzu(=XGKh3VZE!ivFklieD^I4&+0kRYR(&){sBcXX>5
zZ_~|-f!53QQU#d`;4N2fzP@%NBB)x>&$1%Vc>|0vR*fTW@baL%0A)vtn+Ja~xe7AQ
zzolw7NM1%Xn;w_jS@ML*-b3>JEPEv1k615PWJGbxu5k5cm~CrUSnK=F3vnH{5Sl{p
zuB7<@JVS9?nUA=}tvn*T@SfMW2a31iRE`w`)y;B85D~X<Tvc;_sieT^bmn8VZXeIA
zbwlS>tvkNk?RVv@0QkV$?{t5zrr}+0zbm8>><8R_R}h4AH{E`}n8Z%bFsgX@LAT#A
zSw%I~*Bu7@Ki+=7UYh52$)HxdoAl>W5<5)>$y(*#?SS<3d1XLQW9I*`^^$MFHBNN<
zwk-NBbPGKJ`FLg!2wYk}ra4Dsb)HWylX`WhrC~EA<dgp4L~6cu^*;lzd-a!byfPC3
zG?zfv7Zd_9H<zKR1SfyhTg!6fHVnMaSLg@K5CFkPm20Z(S$nJYmc+NMJtf&(;zRb5
z-w!B}l7^ZYz3ufuBbk(%U=swIqO7aGD_wm)JiZ)$c_vdi<ry4SFW)PW3N62&rO<Na
z0F))DUS6xO_gbTs<53QyHIW=nuDzW3ct_yn4mimH$2(`>zIT709QZ89DPfqSv-jzF
zDI=5v<gkJh5OqYd3GMOt?d8isC!@6rHeSY4`Pb9S;on2q1#3xZ=c_+n53#}!DDqf=
zRe@~v<FD%b;nxziPml*8NL$N%Mi6LW%?Nr>6ssD7Sd2D82uo*-bk0b}m{XaM!9HEf
zeJu4lbxzqJjcb3wR>;+lXvta0Z=c1Ipf6fttpbasB^X;5`b^!@8HHE-<Vk@GB~ojn
zBQXZjPi^B_iP+d7Ib~?Tj*d)-8*pUhAu+)XONLW<k3G%B(3?m{{;4fEE9qH)M#fhL
zvt&g;5tgHJP>u%-0i7YXcUz;RZotAR9*NUSZU=IP-uZu`TeFjyiBP(=b&8G4Zk-Xj
zzWTa3FV`1l3b6|HMEUwnSAR;~7rB?D=)d~T*R~a1{ZsuOr_Bhi7blCjs9eBXz`8<d
zIwaT*f+;9%!hDE{mLVulbQ@M?df*cGW@rGJvNG(*3J@+&NI9Njyi3>I)YXnJ6GJ~A
z@2t)dG?ss|)LI=&@jgwSQ}e8O%%$M+RSrRY>>PIBQ*ynIm@@}!DPPRZ2<I_!I<qzB
zGBINi1XBmYUckwr)|kXc;z5GTokKKTR7m9M6p74Jdg#<eqEXTockVW69D)kKJCUYI
zp{6s^9vgx{BMlh>ib=kUdt<CDw|;~b@8uc04=aBhP@yqrtWa;N>c%@_KC83*4L6g=
z!bLa3(CDXHC)uQrIB6x{0N45nlTLz0Gfj2J<phlo@0?^;zZ3CZ;-^0o4K=YzEY!s3
zSeePj^!L?)IKq8177h8sp&;x@+DNr-h9wTgWV#pAGDbIx5)5k+=xm5-l7XF3{x<A2
z%`kt|Cf9}ig@#Ggr{kSeJ(E;2`~<4Luw;d*FOXKK`V#5wRej0IHmbfr1(UKIUsm@H
z72{xpwo&&+rmZZfw^a8mS>8Iky}XwpR>u2P(P=KkG|P++GDKk?d2d7?cyJ$8-Y@57
zgmdQ=yM2^_5oO@nzni*W+_w?RWLjH*oj8AU(rgGb#68aw_sgvxVTG--Ogph6AwPU<
zZNHqK5#9}pO4(_BZj3BQJbXZHznqpax=}gJYqIBTh-rc$!m_qoX!{8}4MqnMmu)tU
z87;(31()-o&sz+pb&7iAt^ym9!Jq<IKa4yivKz2?NR)QA<ZLC-T4WMv?fgSfFRXt#
zC_tx6=9?BHp@5gYe7-_mZp-zA!p!{?lHM@wwVy)XimEa{!OE+!WW~y>Kw7c#Dv^4+
zl0z?NoqB9h3YVPDdff_)S%SjAlC>)^>y;cO>evcQ<e;R@0V|uVz$nJ@5K1?XSwK|t
zq7@i1DPy)@fDuQZUbVk68)Bkm0*`-Tr~MU*`N_ZG`ir+*e=VnGgmUzXXeaqb)NDFV
zT!1Bx2YSrpSanKi(05UNsvcY3&k_E>Nj}|ar$vOQvi5wZCH1@J&vtV67Jscwwkbol
zzTJB0ba9bdxs@kI$6_&OV`E_?9NJZZZ7#&bP#*Ugy?Y~V921>Hy&c4Hfjoav9T!h+
zB##T!&GfOk#9<TKbJ)uEH<PMoce<*By-yR5J9`?P?RU}Hl8zNZ*hyz2du8OI_t)8l
zB`b7xfwV$rmq>5F!%?zwNncOzSKi^c(63Ho59}^=cKP%sIvYGH>$jBIn7+H&TxTPK
z%spC@j%Gtlv@9mq?Z2sMbT%PMuEXw|nvW*5mvb}1xt1`u*C#osNm21h&N~V1<^8<P
zm#u6i7=LtMS8rP1zoscBj>~fX#>U0U*{z9kvmvGhcJM4`|3;!4xcTHXhrMug?EpjX
zUE6coL%jCN)h3sNRw~*?dU6totz>`5chxA_z1mOf<A?X;j&*I^$sjsI%v&fSUpIFy
zm7jpJLfA=s0*OFz?bdlsQJ!h3EBf(Tx3&-+&VMR&gWgGxovaWRiH=~E*SNjr2&62S
zHIs2J!Zf?0YGd9<aL6<d-ChPgwHteR>V?Kv#I;kLrOOi_N7`mRoYjR%d)yCspSw86
z`cLuyPEuKY&MggmwR`n4DQYj~?e=iEpsZc>qGT?@v}qxz#%}VOq&)u*7wD1b3T19&
zbC<S(B@mZ!@D>!8g0Lkgmkn+upnun=pTGR=L3g6Rd-ds(l+Q7^9mOGn!hMP2vMV5k
z9;x;bcZWiy59L=MapiYC<1J|f?^S>)jr`<%*$t`pRRXd6(&LFOGJr%Q5c-I2rvgd@
zuI)L%qzEEO6W6vqU-}HbGF7_i)NxuLrTuV!Uu5t+X2wDVolvytVmqa1!+*IkFZ46;
zP@n1h)3>IN>DL2arc1Wd;0GU-l9$R))uqBIE?u;8KBR#EL)P~48)|qz^5YK=qzjj<
zvdJ8Dapc4J0x?sV|Gp43$c>F$=%f~*R_076l#8yiB>+f-K*%a1P8LE;n|*+eiz@0o
zV2PxL{79PM@@{yk+JET;On==GNplWi$BZXe#!A`mx#@5s?4Xq^A*nun?oaCI*cHY%
zLYHsSF8dQU8kP3@CRhY&$5Uc<0db<VVJNNmg=zy0Db}ZC5eI;u<h*MrIXvYE30{G*
ztH>?cd+1j8Oj4_obyGsaGa!crLJijgKABtmeG=L6-hIviRynGy!+%PBhgJ1{SpmAp
zu=P+nDT_{wR$ZF`7_>zEr#td=?s_gXmFL4B0#RQ)Wn`X`3<k&(g=r)ZhF^8vU)TkA
z`feZVQ&6DYG?XnyV>jJMy6BVuH}r~R;OmBF%=B%0&+8ddy8ry6p38z!$%K_d`^xNX
zLlbjgJG|JKl}>rxwSVK~57C=IWwfJ}UXuB9m&|L|H<0|I^!OxEiV}%gdl2za>Iynu
z(AjVB;u7+m-H~r0tuwTN_9{jrR~z?YDYouy)1Xb-bRz2ml0#Au4?vy*aU_iX-BJ)3
zLM-IuH{A7`rZ9qnTr*Mta!Mm{G0g0@fkOr+xnpy8JxvfUPk*S{2xIHes8Gt{BoHS&
zw^K~<U#EV{yvuJ*Rdh`5eAk(VJz9~AAL6UhRDR2_&Mk=|P(|wNv__fhB^K4H1U^3v
z@ee=PW01<BK)w9?Vfpp>%fo*Tj6(>;rY`^Z^`Q)aGYa!EtQJ(nV&S)cU4D7^`?_nl
zC9SkB&ZqprZ-3Pji6=ky!X#@nTpcjI5@Et%!1r+R7Q>{qB>FRVJ&K0vM2@?;fP|ZM
zw<UpTB6^$bH)%EjWii}FrByTA_cdj{i*7)7?fkl<rL2`(Z-JVh5S&>pQ~w9>rUTCO
zAIY2C(RRbK&pp}yINBk5g(mWWv@bh4#|$-osN4aAw0}<VBWagAx*)5fg0252KhY_J
zA@&3Lkvp#ZaC@O*eD4<TY(Ri>%7s~Wc<zE{Xc?)H8O6?9sWBK#$f8Wd(zJ4N{WgUq
zZ1gA3RA4qrO7Ug0UCPQfx0f@cfnk&H(b%RMMEvx*{yI)@Gi(Br>FRf=IO?l<meYQf
zR!|LFYkw*%{T9vkPXLdsuuGG6`#3{?X-Q#CxK`esSz1iBv=d@G2okz$t$mKj>pdY1
z!B%ZD0N-tSHlSA(v&%Lx<5b!VAPc2=V2K=1h1-+1HcBa4|HZ{@#ttyXw|j8456BS=
zf>@hoej>ox_6f&|znXcGo_AC+(@8HfJ8M0(aetcYlqNyL(#4o_eZoB7h}9rNCQ@O*
zK6cb-TN|0J)rSh5xj3PC5gkyo>{$@QERWh}HU#o<vCOT_MCib}GN18UiQ0j}WrMD7
z<`jNHfLe{SQJ5Ylzz5@GDWd~+NIXHb@^0CuNIqH1C{M+)=H?q<+Ey-xqK-)PXDySM
zI)95A&3S$1N?>r?`mA<JR$w`87GPE?>eD9oF~>=G?_^!%U~?V~yRlZ#VPr$OacsD>
zdlO9q*Zyo#_45W#yn(3A#u0RWc#~^GmnL@u&1m~7UfF50wkO*vYRV*5vYl`~+T5Ad
zkb~-48Ba>MLpAML^ZK>TeXaAiFbGDO=zn)^)m2vn6wy9+A5cf^BP;1F&E*u7ovLnv
zUAb#>6RWkUoQq?0MPH+ZY_ZKb><Ni!pv@skovq+nJh<L@)<F1NbzQ<HwLf_?t;$K@
zqMdP+GI^RTWc0k0?MdtsXNJgV1Cdo7^bJ|D!ci}2NfAS|kE8WZzgXO^NBS|c_J0m~
zS|$>penJ7Huc?$BB^m7l*RH6zreuA52g7Jq7`C}11a$~5namky;?61>4(O|(V^BpP
zrf$r7V+51Gz+tt7C*MIH4{$k?2J0qv!`Uj^MP<V*OIMJIc2(<O%8|3Xye)2KSQEdR
z*0%kb^ZkYSpT{thm2<OkaU)G}dVd7v+W+TiiW@fqGEEsH!`<dbR*a1RkvG}r2<v1=
zZLQwPgKmP%Aw=aHd2Qd3T91=9va9ywD*Dc}=Vz8oJkL-w^Z7mR_m(Gp_Fd%(=|*w_
z&mw8sY9If*7!OUb#BXPLkF=G~h1%X_lsCiMn?Em7NzpVW-htmn^3Eij`+vb5KQ?6r
zFmA7GVx@D*;d2s}F%qaSo9De9L34SHROjlF7WgqvHBpw>Bo9l;)#{bbc9)e8KV{Z@
zgX$j2+nh?*<VbK%iA-EueEiY?=tPd$wl^hUee5Ld=%Yfq=~TKymagA00(~Ep`Wd=q
z!pknz4+w}+K^n~ylANd&D}QDF#9!GQ1^;1#)&BGrR=X$ABjKbZy{)Y$zpx_hWt=Q9
z!7Kk!N!NMCcX_!nZ>?i<I3Qu1vT5b?#Q1prLC^KIh@!95Xqc<()9d<FB0n6cKZ#d}
zgTM+a@=hn+4UE#t&ztLU7Z@}}tw802PrOj<l*U8s2!r-8-%A-@jDPJrawv8)LvhI`
zalcRN-qmOyXHX;w=S2&R9(-}<W{hhVKd&rl*Zb2B`i@5OuUY)NvZAe>Huf??`*N?9
zhkk>5(6o2T$ILVEy7pzm$!IyWYtUgiX?UCM39h`7Gu!XaPd<$@*d5AJw|zr{it59)
zt9HAr-?MU6k8>y&Y=6Ii?>R0r)2+))4ez<k+>>^^%tX`7y((rR5cN0o>+qJEQ$pn)
zcXX1on67ECe44t!5QyVlFJLb(b@(>{f=I<S74jAc)~S#U4<|Lj3Vs{zdlx>~OIA79
z0P;S;DGg5S`RVs>ZZ_^Asp#TLMy&#C>rWu_j3^x0;xL4JvVZQn>#i|m<nE-l5E!D5
z72TkdgGErX__JFgbrDov<n!sDU8KQXLb>ipUcUCBm9a$&R+-#P{L`OqY5{jpm{Oc7
zVOYTlV<UlKuQ@g+=Wkg<6xJ~2vh?tlXgng=fJE@8|K1dhH&$tz-y<4iM}>`@EW!LN
zWSer|A4JGD{C_2&5ugGqzh)QOh(dXjW-i@hFJ8{tb<Rw>zjQX)sl;A1<EA)XYh7Ng
z%u6R$FA0Kd9_(KSc`I=2UWGb&;iGc*eaDgUI~-YSR8Hw)d{7=7Z-kUYFVag^dq@>r
zSQM4l3cT+w@2>-fa6$V`#tBSW2g*c1=ZD3d&k}}VPk$u(NzD8Ok)OVzaQiA8+O6`q
z3gjaD%`Ky{0ok!eF8TfTA<>|}$#Th`X$fqXB%^d*PWgP(bZv@gNe-8k_Xd)b*$H6?
zUoX%2$=~E2lX;Za!pd!z;&f2P!Je+YE}cR38~;vSItxm#2++6Lu~}hIl1u(_`xIwX
zvOz8}Uw_f=Zl6@%BuI7-n-eKOc|+h-2q)(tpuM;6QV=2<Tzn63T*%65SZVX-=i9ae
zp<Xl^%&jBjdx*2>HZl^02tqI7@A3p5okkHgKjo{o@P=LE!<^)*Hxz{C6M|fkZfhrJ
z469x)xy=n|RA})xe2KlOo$`ulb}Rp;Ejs7|!+&?`uSiX6IT0*IJ7eXNI|K{L*(6uJ
zq4WZ;a!~0^8*`V^BYHoG@1-rF`(K4{J9#WV6AB>jr;TvJ54jU3e94_S;Y;qri4>W!
z_t~(<j5yeP?8JIN(m*aj-T2$VDHL}k7v1)KHfl^>tz6Q)`FL}*|Ns2u;eXn9C({aL
zZkM!yB@qEPmr-;NJAbWOOK;ma5Wf3Y_z#5TkQ_dMz(DN8#qO;|Zi_uFwmk)l_WvJt
zNDd`FBs-RUP-IEu@Z-E^7`gwsm;0}~<NNN*S)08R9>J;o`;R@)2T6ZH(o2hb2OzA5
z{rkuMuV*QhrRzwSky2~A9R2om$HS8bPb1(&7aX3Pf#+%*U4QCK*NJJEgR{>(pWbOj
z7tm#iPEbllO7snMybgcA|GAS4`pX&p-U|%Q>LtY189{UlVoj?VLV=PlqGkw%@wAHZ
z3Ex2@-7&<sbjLi}yp&PyndfW1Xao{kBrcd!rklPyevaQDA-=|2izuv_c}ufRI6N7Z
z5Jh^&jKZkMD}Pf~u=#7k8s|hren7yp;!~zN(eM>&i0{gp=5%<9QbRkvv4vhmA9y{*
zcN!zlGy`>myGN16;`Ao*@}Qt##3GxBZd(7yIA=trrknAAJ~GDi>}sw}XeF=FqEHRb
z1JlRMhs0geVe60-{@IROAd?OZGB;XSZ;Q+uS(=K!E;@gUmW0wJOF4iI3361+1R(Ho
zcz`%0vW>r(_^W}$>OU#Upt20$oG&~TS`9oq&z_l5*19axtRxy$ml}K}H-Ce0vqQbj
zsIx*CrP+0w;)HB*zul8<rJ6YF%2!--jd`xQ;1=f*x|jPwr~761NED*R%42r<5V9|R
zQl5i`%wZ%sei3Z+{qgp-dwP4{{Tf(|#tUNL{-2Ls{9uUV+A9iWiSfVx-T&Bq>#5ce
z(F1vHX~{h|UK=F2a<jKG2!DXdf|X;ChK5ale*fHeZ@zz>aY7K9f>%#THgRqmt!P&i
zK!$a1Qyfx?n{j$Az=eM0B9<Pfcf#m_Kf1<NJmyQrm=w<B@BMz{@n*aVK`0#H(gNT)
zm4G3gm?k?VPz`J*pLH_~50rqwfCE2;LrW#J)DOU+qr!q(#G$X_(0@uuKZsMFTCc`N
zfx!#FIq8DUM$u8~jN&GWY&adoqz`&^-_wprogw(yPjBB*OX-T4ccZbyZ)CGD*=tTw
z333dH@1hCC7;7aw=X5%H(d0?rmZj#bP&)8h>U_r@8%ht6T1yekc8a9QpzJ{Lby*JD
zQnY}6G;8$0L5o67!hh0wI|Ca+*L0*A98;0xT{*~a@SI@C$W6bUdw7mcSnRh2YI$4M
z3b6gQlKu|byr#)$gjP@&eRY-_tw?i*ldQ>Y3y7$NCfn>*TMLMV2kq!NXJZ_S10v4Z
zShVJ7)wEmQzri|BzC9*sL)xr*EO*$cGD=X5Kkx$ABWw=O-G78E;#d@vY@k$WW=rUe
z8iZH3=`>9R{*x#+Z)%D`3l|0j%Q^GK0KU4<lp5c^S8b%gpdYwb9ld~5OM6v|eS>u#
z#LQ8<fzoLyp)W}JPAPq#&?}7sN!KW$|FshiLi#@aNO^vq{A)*6m0lXY)5kA=d=Qq`
z|6<n9<Gu}$pMT0?Pt|Iv9Jt(WQqtG$CiR+ATXHrr9B`W{X^WcMFD&kB4I=QNZV>6J
zB~(>=ef26!^JT5lpX$F-4d?Db)uQ3=j;mh96<YMcShoP0a<Fk|w(i}`0L)aH4$Xqb
z1r|`*<&~#_xmC$K2lQLZMW=;yzONQR8eyfsyIcfIAAi-{@LAS`+wz&5p>&kjat33B
z5B*?0z~B^;gOhs6zx9S789-R?`xV!MWIz#&9`GsF5LdMk;{A1QNUW41fR?1;>IB-9
zdRX!EsWKNR_2!arDxXCplf$Iqjycf0`978qV37RjoYd*M!?@I4ToGW-%{pqUc&C&J
zkK1L}A%BPaQUx+4gV0l5OB2YUI%$Xf-XRA{eu#^Yl0!@Ff*iK?Te=lGK5dd`NAbMs
zE+$#EGccafyipcjO2iml1;#%`3eB7E@th{eCC8`~2xEs`vZrrPdX`;lNoHFLRKCXQ
z`1YL9SP{sp-@qdb!r{0BS9p}G^m{Z`ktQ?>+kYIytR5O%tV3F&*Zs$%<XtqXS^gGr
zgA)p{CBV{d{N1VM|Ay`%rG2vJ{UtT9UeMJ|h}r!;k_Sb)4UOo1rdm!P4IhS-NLHOV
zu8O{;`ia=_`at=!KW{PIq5#EXG!$wLxV;?q!g$wXf@RNNO5JYsz4~U2kTA#w?ufJ8
z!Z+?ImPYUPc8;P#WK>SM+(A<z9>0V|r8{o!Vqa#{JUXO;^P;FqUv#^XN)RfRY*j*0
z&O13lHJAT@B`SZVP&hS+M7}MrTQY}kT|57eKsu7t_gQwa!w%Dmx$tpOkLh&*!*(xR
zx>DUL)xQvY#uHEG)lhjvl+dxYJ;M8ZkSOi2mO=^=m9_Vl9Hh`>=!NH3s?O=~WxY=9
z7jVR|2(&%TNqy*rL)ucH7G<2;ZJ2#nyZ9kBhWk48Qf~>Z>aqO~QzW24m$88*6_+50
z7HYRJf+dzD0yQ(20X708m+Yn`3jsNoAp|}se{FNyxDo!IUxAs<bgmx?;{BzQODC6{
zb53)O8@rRV9*>5m8H=$jYAEWQ{QEu&Nwyi<XPK6B&4)M(2rrApgJ*XE8e;+z%oyvK
z)Fe2~ly+2Om@=M(z^DKrG7}ZG5zK0;agte2jh$jnrpA^z8=s)Yvf$Eyo<*C4z+%uO
ze;6i(vXH!DQU>bhEt4wJuU?QQG`dGS<I}H_Nh?zi&!kfp6ZpuaPjvx3Jl26atinf{
zEMVlQBS~0?P)U-I9E}$oQ%aKJAz-;8NfuJ-fdQ~`nggp<)}=WjQ%;b65$(L99uajx
zlR==k3MT#PnHG|Ci2;(8BpFDwHj4qYe?k$GW7-Ju1d@QH1l38Vok)^p+QT@I6w^Vb
zIt(f_w6|20iX?$QlB{ec8Y{A{l^9<8B+=wjVh%EP+A*t9=S=E>I#1eLGDJ-CVhf=e
z?KhTS6<Nc|$Sj721SPG5l8{j6J@rsP0PIY25CmlURWqkN^@Wnoq){+{qwUtPf09c&
zT4f*scTftTafB?3w$4YI44SzhlY}18J4}Jqdznn210=s%l~)8kYb5j9(!w!^Hzr8}
z5+T$A%HBp&#VQ8fsjop_pUeV)0yiFW7~od;1o_B<veXW{7!o+GaV%(^>L3v=Aq5t!
zOA`9pw7obF!6&VNw-6Ndg=!%<f7%Y5rYPvNSsbY-ZR#Od)QW6?;}HpEhL9I&`P$GK
zO3QVQMYkMSLZTySu@Mm?lcZQAQ$2mkvKRHFZuVz+Q8JAq{*GmPdE1uFl&NL$=g;k{
z{Wfn)CYKdE^|XEZG~3a1K5=YN`RrNt;qCFKvY;P!&l54|u=-`Msf+!xe?83hUhZTE
z<>wYH^3(Fg(X(gJAUpUnD>3J3300<TIc?{}51u+_@5*_-Xo_;aqztP1wLGcvpX$$t
z0!0Tr`>2l4zd=v55=raq`E**Rft+4y@#HX3738DKNmC=1{Zuz6Ws}AeN7*mgZuSBN
zP{|SIFWN(8B9DNm41p6$f3?zF0h%%fF7(lpY=3dw(s*yG=~?#ta`d9D>S?y0{rlZ6
z{rh3uwzK))vaFo)i|VYJ(L#9LoMu$Y{<X+Yn!K%w(Z|I!b^X(;CZCN*^~X^=E=P-L
z^;_Ol*oVsof5EbrHhg)QsF_y&;XnVywm2RD0U(}D7n8}+jjeD{f4RdZ+F-Z<5;th+
zJjdbPq@|S{e6c}8?KqvwtCm}WWVYjo+m%=qK)ph&3g9#uWmp!-RYb7YvH+eZ_Dlsh
z0;!B%7QptB$aOgy4WQ(00RKH?JJ&<DGn*jW#TAlQ#&q`5nBbQiw)$EFDLepS8GHf{
zMm>hLJI3mmzGKtQe?EW`CEWQCGkV=KqnDeT@roe4y&?!N1`&jjnv=ERKe;CE3zAdp
z0H;I5u`Pz!ofg$eIW5Z3dEPeF=kD?M>W<_87(0FMT?_qPb3?%I{T<-H*5DHS;5mow
z{-)r+2JsGu+}CoUkM|&lST6h#IQfb`5rRph+^Rs~yYhpKf6|W<f*oWOcRQRO#~#}<
zL6m(H#33e#;*Z9~y_eqiq|pp<0Wn^=fT#z#02vS?A|X+TdmOS-q<&mukCjtg#BF&1
zEdJBtltRF^$*1#KJukaXAl?-d#2aD)L|_4!d4!$rkl*hF2-hPpmvU~|i?hB89j*lx
z_r#4A=4;&8e^u6SKip7qzQ&DR>b1<y2PGMZ9RlEs#F<OvaT@WlMHb``TS%d{g`l0L
zJ}#%#Vm_*-)3O=On)+l>v{xv)7wDtl*eklBKw5weagIl%n%bgXZcF9;V!UWZMSVV-
zl;_xp4&v#m<Qvh}?de1pzt<t#Ni;gh?w*DOW_Lrle>|KXt%!Y?`vIO|?X|z5b9}(6
zWiL*$a!8y+H)Y?KKeVdn59#jg?xkBTy1b_46|YIV;x*~kf?o?^EyR_e)+DGdYvfu`
zYhjp!g)`+o1g3*ZTw}pG3UQN;Y7p0))K#JwB*apRm>ulBf49$-0NpC(jrD#YCt8I8
zEjUndf8?k-IPQT-jW|FnwM9K8U7HB+)i6ED-{|CL6i-*uUUQ*fU#+;&=7EZ8pGh8l
zkWR|zjSHF=V`&GPG#x*>L>Gv$A|tmW!Nv;z+c;qFFD@=PQt0uVmnS?g_~I<9XE;S=
zbNVHFQy2N<?X1n>@nSwL+l#t6<Ky;xGSJQ%e>kQ0ftB)Bc}~~qkF5MKWwkaO`vpf1
zFgSmC$S9h1Hf#~S7*~LpU3efQ=$aAyu65+}f~1__e#o8dV=1a+<QjYm#}8hCC1Zdk
z>FrGLAnpAX#n#X2i?T`Dp47!W`}O(3-rM&-zx(TXK3(J!K0En1P*?0Wy@xrZG8D}L
zf3X7TwDb{?=m4aiR_%Ck%n`X|ADi-|sm{u?hI+$A!YA@iLyr{LB*Go76*V$&uLdaE
zF50@O^2t2QXR}FF<n;SIyC{!GsOacN5rMiRxw_t9om?5|aia7FEfod&E<SpIv84+s
ztNYQ09u{p}W#9Cgw~P|>z%)EPFh``|f50(Pa@r)PB5lEg@cVi(sq**barVo>!QNot
zYN@zDY_Qgl@+J_(_B7mmo4wwD`_CS7wcX%psM~6FH-NB(p8g*Jl${H~+YowRP#V<(
zr4gHh^2!!Qen&Zv3H?>OAxF`wtD|Vl>L}U(oj_v0bx?2Yu-6f9xB}u3`Ew)bmBhdW
z(tpbzf(+rddM>kSK3|mc%*UaD1;Sz24j!NWlN45H3fd$W4|IXsSJgXi0nU}KlwsKT
z09`4t{8BI4#c`PwzR25R{P*8}T)Y{7IBiF#|9D;ex&1L(qvu_AF@6(vG5rS_gVy44
zZXO?nUHTs-*o{6)=&_4+-x0gG{*8*0Lx0%C`73s@dd)7EG_$@kaKj#9bwGq2=vyJs
z3Vr6}k#HfNayS=mNJ{XD4lH6pzB<HJeR*`}ZGJo{Z%nc}v)rt;To#sEI|EX5%xTL7
zeG7&BAR;tyhTbZVvgZt=?`N;-gF1W3ewdd@TW&aAPYe$6Eh1>|&1HyhBdA&U|7O1s
zf34^`#J{(qw}kk}>POhb(-ME{Vw}(0s^~>cPHn<#omNX;bHQ^P{3b1x=K^Wr-7SrA
zobH&v=_^h5o0rW#r*ZT>s0}w<hIc=-q0cR)9YRQ6_6aEtp<yp>q2d1l)k9A|m$88*
z6ql&QB@CC{Lp%qUEwe5uf4!Tza^yA;Ky!Tse_&yt8;zq%rK*`(X{(&sR_1bKd!2X#
zJMr(kagiYB00ui$Kv5z&J-|cXpyc$|B&RP=$M>fn-cV0gI8qj-_pg)U8<Kx2$rlYx
zRw-dnP4C~PU!SD}!^e>iEhQo!mVbAArk78s96zB>e5jXCR;%age>pzjjgM27p<k?d
z&Y$ON2z)3$49oEVGDzZkia)3H%Wv;LKFK_2#b;HWU*DhpdD;$JqZJCw^!vA`a6_XK
z=wJd3!Z|<v`{(ra>8Dvl5iNxazwys67i1c1gaaF7{`#Ao{@|NG^26xe^k01R+x$Ip
z`g{6?zn~@92^j?=S2Omv1j`A8=Yc1&iwHt5r1tlV=p2~=Vcc!0N(xU~u~D;}V$G7q
z5Un(eH3|q>HVMzdvu4T9FQ1HrpcAYg#QLCFU&!;B^@}Ln4H~89=Yk&UpihVNm;A9M
zA%88d)<V*iwPm~3I-M!3ZST4b54ZsWyl@g94Z_F48bl%e>_3sRKASZ6sN=Tiw$7R#
zsi<^EgVHOtFYd63Yj|1>T*?Wj^-j0|ghH5c*<${D$YkyTOi`cscL1geLb|MAh25fv
zPzA6MfMg0*@YYFpm_mbY2sVy(qRZB)EPu{)Wu=TJG1G@|ZB|I1*$2k5u#0?;<vA$S
z&s=uo-A8kH*0*VrH^Scz+227^O2NiCJGo%muS-~pB#((rARYN+eGZ(ixI&ecldw{8
znHT3R$mv5mKJ^Atm%{d=!`!3g&f~Qtr1blQIkHZGH|KE<vN=U(l~++$i@liG%71j+
zCASvj$zw98>#;+c5E}S=>eP>6E)Cu$M*dwaj*JzEbIKU}cZ9Mskya!?qt>ASe5zq$
ztzvSGLb+ZEoJ**H94bPcVmm31(r7@YlvV6QnT$Z|=aMh8>!YikG9O2G7()ZEPFcu4
zcr^*VWH`a}gARYrk-=0hz^@#IUlxKuTo1m(P~AE?bfNm@gCADGBOLSjm$0)XHvztv
z=(8ns5$gmd8jH&9=tDLjm#&8o@~xL?v?VEjLhkfL!xL_Bq?A4E1Cp$;dahugprNIb
zS+~1~L1HJw;n_AQ>XabtLs1*C6eyu$Seh%hda0z9UsIC({nYUpwMa=p3cqLf=WY>w
z8(u*y-cGTuZFp^W);e|?HJ>%uRqHo58Pm=pUG0j6C+u@xXP58LVoJ;BIlC->VP$TA
zX1*FSGMN7=-|+f^^jyYbi&>hg!O0m$0LiB>o4+pZPoi}k;MI12QW@dVl<mol)E@ib
zr6fI})57`L%bcf)IrJT?cG0D;>e^~W9y<V@ZMCAXN{#KbV%YN1=J7kNsGT5k-(zcc
zi|8B~T0wrNHYKtlJ2#t_qMWh#oY|CrlECKxtn=@ud=IPAA!a{)9&>z7Emm5zLRMoD
z@08G~wxRE!)Cc)Xlu8ij>;AIzolC23PB7?lr#bu3wU@N{zO8+2@t^ElAyupjkC0;Q
zWBemnVE5s6@{hpY*ZaA=I-g`}OA;=-UtuRK<zjh4nnlh({{3s(T>1ISGACnygyO7a
z^Q|B^ueIc-Evu$g+o$pM{_hPEfVPs<`g`w^=ujw5ytWc)T%uRPIJcE3KVM|S>mZc3
zXY1D!-}=V{=mb~8XJdN3JSNUXJu6`Em_CQK<eW9=nykhr!$2tfka>KoD?Gj*-n_h|
z)!<!pYMT+nz1!F-SdIIdZJ?rmaB(9A=c;ymd>DF#_Dm1cUVh)$(8PV8wFG#J48J#j
zF<Kz#Gf%@QRH%4|@)yt~oW{rJFX8<)(`vrm4t;lv=o}f(#|*XEI`S9b!0O!`>SBI^
z`H)rTxupJ8$(M>j^y^k?X`#_;+MtS73QKieGNI5D>V1b{oIrytNMRp;-)JN@UG{OH
zg<KkqG0QuZmpJoT=YE&Y{V$)hK+4A{XN~cCC#&H(HCRb6taUrFQW!mCVO3sFEe-ZV
zzUE3BD>FM)><;-e$No0g_?&LDQXDp%@;kATAk@+@oU4jMwHcxFVJc?Fs=w+p!})$6
zukm>`c)8%*)jqsJ9;B9kzTm7sn0dL<W6lig3a)Un%f?POtKm5{SXoYr9^d`TtQ=>7
z&Y!WenN?lAQpL+v9FO_==<7OJ4bQ2;D!d1+aVJ*B3&1l}zH(Vusn8*-5LTg}Gyo3$
z?tOep4OEo!fPGMr6JTM*%2aAam3t~YYS(Tm7jOI9%Hwltu+k}ivhTx6TY(C%vN~1G
zq{69gFcs5t6}!%+()hd@yf8)geRzcehV+V;t$4Xgr}QSpI&IPkyo1&7oEoekMfZJJ
z0kI@HV|A)n9V>hE)w>O@vd@xw9XeSJ&#A#mrO5v1_Zl88cOI`y(RH7-l_<1b(n5bF
z%{f&e8!H>om3_oQE%?s09iG!|RxSqDT}XKlq})e?8lBK>OyMPIyHn!qlpyjS&1rZ>
zt$zfJQFRwkRtodcn1<(b)^6LJ{|7g{#gvz^fh81|z*H#|0yH?4K`thj7r7-be~K^Q
z;NEkepzQGXV24k;{qyeQk$CVX2zVTxzYm}<tp0-4hfW3$U>w2l{A2iXXDvEC_Ij|j
zM0$9=4tkdR4+KF+z(EhV|KJ(!Cgb$LqaKHX!S~+XmHGONP!G_<DNaPR2}$39&I7ss
z`uurki_oBFn4F)UcmM8su^ou&f8>U5KX&N_6Br6OI5r@0`1#lHefL{KsL0lYbfv$a
zW{|<y1nHcfeS-Y*)ee8^%g;KPVi^9*Lw}^<)<j+4pWzS1w^Nz}D9zS6kr~Y@r&zDx
z(NKCd(hM=FUDQ`vAe8Gvc|N^QG;uj%dFMvTCJN1H5{kxh`~{uPa3x~&fBKEMlAuWL
z-MPA2r!!o!I;G;j4Oi?9lYe)vuGZ-cS40LPyb)LA4EylzTwSfx8Lkiw621{vsAq`p
z&ehdAo#6_iL8g~+(Zo<;S46uM;+$l)kR^D23APqD+X46G?G@Etf|%%92;lJ^yq;N2
zBtDyH69KJJ)AtDW;mj;Ff8hH`lA3_Lu&B?lwhe+=fYM^{A>FChJQlCfVvrcr)2Hrk
zwK*VkCQ|I11Jqn;J=N}@gKgH*5!OKY`I!dHp8S3$e$>#ygE-Q9kBug&$E?+P&0;4i
zRXJr>1l2pZBC`~}G%8tHWMA&x$f1<?X`F=dS$>b>9IMTq>@(M?f1iU#?a|`s5nLf~
zgpnupn0PW6b1_pKgI<SQd%XYPY(h%iUP4O;76H>}d$^v0N-kMV<gG!_UQ^e^M8>{I
zek`t)6qYIx$3g*|bURT|G4j)W)vh5=Xw7q_MooGCSnge0jYeXb8pebL!S-}wz;Sk#
zBJmsB-9s^=uky9oe>5`9IWhk_vzm=GhR?X|VwPrEk#K95ocmJs8^gMI`^j3(o=mht
zmCIY#%ZGMV8w76<q#sRi`MO#>JuY~$kL&9a=bG1#-j5#J(qoCj%#NKfd?+Rg<GHbS
zIbW3Q2@6R8SLG)46+&3@b=}%2z0qRnb;^j)*sydwSHk}=f6~^d;u4Ctqe>BVrzx+b
ziglE8v`$r9zYi-8&bXH7yHa_nz@U@?x(;+B6ts0Sj-(q*HD{K*pW>NjQsoSGv1CE}
zV;{T4x@2}5oc23CH@xpMOv*YBCDU*<_MGHWPT7lbp?&{o^{(idbW1rbv7c)omZ+>P
z%F{ZkW6c1#f0hxiVThydwzx;#GM;$jgkF;2>a09gF1UL=L7K!bPM<-G&_dgBo3Wlj
zW5_LfHzSAEaos*b#%n-UZjMD6sr|#{Hd4gYw0FpzjV`Lm4ouXoScmvhJ@zUUxW$YO
z4j4_a?UPhakhkEWc;6iTmw0dn+(O>O1Fbq;>l&Wcf7Z-0&(^F@bj^>)XxG|$v}{<Q
zmw{4|@fT~hwRu3mv>E9ZVY_K?L@?~)f|?D3>pm!OYn$b8TZ7NhZ(;B`Vm3IOs&|$7
z)ZlkbGfonjEBCgx>D!zN7?j20$Mq7eyW$W|{qja3x6wqbWuMBP`67~2ucZ+^N7}3a
z)!_5de|`O)X8)Y|-F4NeW}W$JTLs*+aUS~7)i7DMS+%*K)U30wSy`j1eagB;$5MK!
z4tYysa)vbEyU|j5`&N^K9b<jHPobbS4%VB&&4?$jSw`9{dP>eWuCmd%Q2N6l;u{pk
z4*3R!u|rHgl>Vwpg{*G=+crO4x56NsWdrY;e*tb%HYgYl3rV-jM#hxIDl*d_*_o(y
zqSt?pxLIVz{MJw2pD1bHNc@QmqmFBhO~UBpEZ*$NOC$s^nz$DZ(7uMePVA^XP2`2#
z$6~hZ*ptoFB@RU(X~6~dVY5gOk`XT}(yRtm`e%ve-R30fSHA9Cn}fTR7D09iqGqe)
zD<^Zr`XPU@EgSc6-RO;~N$KAs&u6M85#{eOFYSL<zu<P~r{~@OuEjism$88*6ahGw
zA^s~Ammj(>Czm(BC7c3aj+fQHC1`(6Fl;4^6${N?m4p%N#sIaRT1>067LOQ`M=tEr
zhay*GsOlu2B{+2yq;1}dVXfM@<JD985_Zw4vmPv~dHvC6%KB~AE9+OEzvjroJB9ai
z@jdQ}a&x*%lv(mzq7Bw!R5I&JnE0?eP@Bl+Qb_*fr;qs+Xcr7B;u+NtPuPDM5a751
z3of#*n0n+&)?*${83mA2%TQKj+5;gTeM)(VyGkx;np2kHcNwq(X!Y7L2XR;2frk;B
z<b5kUX14Q^k}uJw95WG-$DmQ&9kna`ls;!Hdf{4aG1p1&G#kC|OWaSo3xASMGa?cB
zmn&3{F<f9@u+bUS%2%CjPOE<==?R$m`D~-%MuRomrpX$%nA*c=HFEr}_~qJIieILF
z$?O|rcNI@pEEa~|H5He>&eu=*TJWKQTJdG%X7D>nRw>L~!j8;{zwbu=F+*n2l1>;?
z8!y&G@6p*Af`&$bt^$<yCMcO4SV}d+6N~e{>T+W`0-SlZp+wevRndP{Jrf#_{;xHa
zP;ltgp(R|E#n!t_WmCPRz97)WCZ-zNn32kb1W8txc8w|0SiYc22B7S<twxm_-MQK>
z+FlX)EE$l8wv|-aPzPI2O;<_Sl~mq8X^3@rBptr2<XFaX>j;0@l3Z@cVX!EFv^L9x
zqk_C&q&He8ti|?()=PiaSW3~EnSxf$D5ViuZEN|a!-v7a3vbX+clhu;ek63h+&oFp
zw`)fhlq3+;G=elB?8Shnjmcg#FgbFkdc2U;*Hm9_q^Th<Y7NJnD4|PBUByf}FPM#2
zVy3)}$GZt;@gSCSH+vA9ivSzM8YC(AuqD13DV(YY*>HMKI&FVB>h+9o2P)WwAzxd0
zGgRcc9XI0`njau>j~r{SsmRqh*L|gesqOr^>=Ti7piZSH+|T8}UF2Y;T61rJcdINy
zKKMCh2L$GWvx!Cbw|fgWzufK-@#eRmGNJfhVuZKK=}v)RLSax~nU3m$hStuSwaC{(
z-<y5BUG>yzzAk^6UgYDt6^FvZ#L6cQ%i5c7vn1$QNp48KM<sW_a6h$lT!TbOBlzsc
z(kQ@EfkesG_!JUPmPvfA)U2aQfw0s3Ju%r<fl;IcvMJOtcTotU1mB83@5XJ&cTO%@
z1i=j>+*BVw$>9(F_+gMjc{ltUw>}S=Z^=RI@8M^@^u~Yt>_V7DCxGyJhA%WP<%N^j
zhxvY$m`d~Yny4q{n{`->63)Vel}}vl>zT8BWe#)by3AQFoao<t=9apC=->95Do|dF
zDRdkY6d^Z@o)3R+n4tI7svVNdCh4Zt_L4t3fMXZN`^emVzaFQ?Q@*=6<;O9G>A+p?
z%iV#{ws(KLdPFdFv|Nt*b2C)jc&r?cJ*DZ##(dWwfi{~j3PL{s5yRkK(|kby1O-Z{
zvD0idUt<I+=*@04U&H;7%e@oXXug&&nJ}^P!sd%YvmdSG%eyanJ{20qgeMzn-BjaV
z(akqsN}M#1@H}$+;}IVz`C(zEz4m?6Sg1%aJj#D;ALn-UqvA`Yn@Vl<ZKVZpp@0Tq
zq+4#>inVfs2eLwO82?M-E;<=ksV=$3K`cxf#eD=c1FLc$`E?yo?YVL59-1H4d%e-Z
z5%hJikv5wk?5uoQx5A9ZE3FiAaJn`57g>qJmaQNPS@`>AgG^yM>fXweM7{1CF%uA`
zQd?idjg-Qv)YjP2ECmw}KW^ldxVM>Wa3INz!}mtqs$V3X0G#-hj9jEJ{255<t5V40
z4FIl6VQZyH;nF!Sr?8dU$e(9JsCtntkTj><^&)S$o$8yfZx{Y+<v$dbT0@tyfh80H
zIG0h2H7u9V0s}y|6~-lAH<!XGI1`l&D=3$){Uw@z$e)MrZ(kl{5?YF2`73|a|9W|Q
z_~)Tsg3(GCZ2tK9!1pzRK(87=C$!hiufLif55Lu@?Oc+c5r+TJw_oo~b?0kRVvwXn
zX{|AZ5<!w?bz^N1$Uj|_sDqS-lxAqbnv`K^p_-In6yGZr`{~ugRb(9m=m9dT771Om
z+KrTdl1zoq7et|1=KLovrMN>&Y0;e))<KO9&wN$Jup#S2nG0%h=j-Qj?RXU6LOn~A
z-*=jcS~0V<qNEp2+8WWkBn_;WRe@s9sMXD)zFtIRm9w}I1|B;qE;%NJNsUVDmh$!R
zXr*Fx;9~PB6NEG2&gN6r3%d3+UqIo4zK8jLf)^BaJKGNx5e!~4A7qiI>7xbHbUV~&
zD4{agd^bR?h1c%RZh!^}eecjf<^~+dF3;H&90>;4K&#i{PM^!zNR->G@d7D?vwWE3
zvX_fp))<y^wZ6fE2-4Lv0j@`MbO-DJ@wGyn?`7b;2xQO_af_FMr#wnE#6U*Uv0ROR
zN?s!%40Wjs#DR@9%i>xpg5{YHEma3lBULNuvM@zhs0Xv#>zv+?DZQU^GH#uCAUsX2
zoo`uBrr>djPhXcWhC?&>8hLCn*?r5=IFRGA60QQdXi>+@cdFbL=c$~o%NYT#%9m4c
zB_5|2?A1^*eLi<FJ*IGFgn`g1JhK3QBdPT7b7@ZgjTdjODP;pUfflBqCX<Rd7C#C>
z-?hoYjM&c#ur=7Iw6VYdj{WmC*(D09RQ3|SM4Oa8+8JNgyIwh8#V?BxlY3Tx&m+F<
zgaT8K<_<{9<%>~41PY1T6tIngb9Gv+2(Pb5TM>rq-{4p#C}A}g3ESo-DJ5=ymz+$e
zpoc*MuUnc5b=tx3g|TOsbE6fg0->p<s^-5)z1T*h6*Ow0aG<*_ZjGMWWH&$@@$2L>
z)#!^~55qRm!wQf0<TW^$=ZeSIy+!#1w*I-Gtd>fk#DPYx&x^bT00WC)vHopwLb|YC
z7X+*l;XaqWSoF3qPQt0&tPp2^#E49Uf9iFvYjoTCZz%z6XR#V_Lqr(hmZ%0)t84Py
zu9t{gYVzEg*M=MkT4#lGt{xo;)?Z8?OIO?!vIvIxYNkx6P#8DR@>SYcPs+JNu4ak*
zbMv<?kP7Fx&sd>sPZ~%8iB1oZcTF0(*WNEgka|K4!=M|nb)Zpt;2E5MG<vn^fc8i?
zPIa3^6MaBr*{RBXy`@^IIF@Z&eaGS?^%N%WFwETHHBMNA1H?&cPu)wZ!Q&o`PdX^p
ze_-a`Qw3?W+h{Mm?7HFH3E$ksZ(ZV@kh;&S`}sN$lTe9SZRitYIDovLpAowGe4NhH
z_bW#PL{5gBi6Fmym(8Dl^y3#A%+2$^ap=!>8rl5a{LTjk6TqG{#K2T6P;A`*mJKxy
zJ|x%<!eEGdmWv@~N+xMN)HK|LC6fkI<HiGz5vK*tjpVeFs_o8c%VUXLHLyGK{z&T~
zzSlC^y#GCMSdvDWW4asnTJl|@u>8C1HTS2E0$RWCR8{y~YlPK*+PZB89e96hi+e0Z
zm=_R?3Smt=)PXnSj8R)dlXbQ6@nWNkUqt&Pe;XnoElv6gA(u|0^J*@rp`JJTkG(w~
zb6W`YlLRP<{&H_TI<G|`{a{|Mm=Rp|VxLb^+iq8;)Rz6X9g)4ccPIIItj3|e%ibPt
zJcY5?QL!^qfG3rIHth4!;ldMF&AX@<$|{artq5pupSa#6mrH`tRMK1Msv%qLIH_ci
z<p*bJKew7-Y`-+^uGXWk9Kp8dV6IEbIu|S^+oxl&UT1}?4wLe3Oex!@mcmemlzKup
z%bFA|$VfF}6-#5yO7&FB<EyI*^jx-tJ*x^W1BHRwzpBuG(#F&OEooyZ#5~E6YBTWW
zK2-&w2<<x7_322_CXYF#?|ko=Jc(J%WI{;aF<oI^?q9#ADP1W_19poQ7emZ*T<*w1
zaC@ZxTe;)P#FSr&5$}{Z(C8SCxz4a1Bezv$ufmLp#jZYet7aQH)yi#NHM-2>E3?RU
zPLz~gl-rYk6Hx(Kcz;Ty0*dg}reInMF;g;P+92^xln8uI4!cpJpGk?U*EFRo6ama8
zD=vkY>9>-d-h5C;q<C`+x)Xan-I+iTUSaj*WGAG^x3sP2i>!lPU!b147kQW!+wN>a
zCTE0_u11fJHltheR6TiLKfP<Wbj%yQ9rH8U#x?4H{7OI`owv{ONn_q-Y&Xfz_v)vV
zhRjTNNQ;V+E?~IXPvMq+s+wS;pQ=aypZcj%wmtf(+qzlPPgRIrwDqp*uTd$AxS>C6
zFC!puZKL-rBNUKl8Ml8Oq4<4wyJ<&mF~m&CB=g{X$_T|tGjt4X(|i2q<U8k7F8K6$
zOeMvCDU}Z<jma;0p5~PitCw_JQxKeR3bxpAF~mH#Wmu!s$2V3?uEyUffq)4y82i1;
zSdLuG-tm}f2M(f&m5}%3AW$}*QSZ({KvJRKbZE2~Vy0xEgjRPtGy*VW4Exa}w0=z!
zx{^WYZD%`+A!a%*!7BYht&FoYSNDb%$9GqMC-=%sbMCRm@pfNfl)xH#vehAyGSp6#
z^635neL9PrbM6#~O^7*5u!)9-ldK{qswB|G`S{q3N?o>Msu}g|8L7Q<wF$bm$vyR2
z9azvv^+uHKRr@k#db<MT;Soaag07rLxuP$tJFaXMqgvto0r>w7np2w_a-*ev@0iyN
z4*!o!`Xmnc4@NL+mmvm36PI94CJvW8>L&;RGM9m)Kqr@#1}2k#jCCCLU!Onis3=V;
zy7ZX3`rp&@?sZph!8xPIhvx5>T^bO>$^_l8Q_;BQ=fBOj-RBy0Y@om}5IJY<#6V22
zngBg2lZILY2{vaVdTiWU^m@7pC0O47{QTZjZ~pu^=r5u|tOaPASfih8jWRLJW<woQ
z)G<(p;5!SuTHW@4>Ajp<HLYn?PGgeR9NRf`+TS@!-b`NPr?L@B8r$?kWh0X}qm5{R
zsbSZtZicw>>-DD5)<Te6u+1h!Ckl?L5wq)f#=(A-cglNTucBkI(GsT^J&*cj&azcv
zea%^I6b@v^*$!dYZ^+^To);jERWYA}FJGzo2V<WwH*w~FSDyLOOQGh!=C5=nF5m{7
ziiE3`kNO-iH5#I0fgK>Mi3(KP4soSsnpH67JS>iJv^DP8bI(d!5Jg+Xx(|$uAjmQz
z3=R@MqQ&wY%JZQFHb0u5-nrBYZH7%v$kChY&HK7wFAge*+~mov5La>+uukk-$Xg$h
zD{`I<h%PRF_EF`uXP-iQNbxYcXca*t_Thk52=sVil~(NQMK-0vEI0+`6r|gr-6h%_
zz7It$m->Dr%?o9QEB~pRIhn<?HGLu%Us8Z8DZ{q^#g5RA9S+$0VJYZcPS3)kH0xP<
zwh<v(-<+SV1KrSfR-M{%b52I2V(n_wwWKVB;>p&3vyB^)Z?>#!Q|q8t_S0tFf&q<U
zEsc3ux8RkSnSC;&#$Hdwz+>^gv^=6iMjYr1M}pDTGSYmX1h<wMrLdTx3M*Kw*RQZJ
zG43rklBiHu)u^jQl~)e2{)`T3M_W(IP=xLIVN)Hpb-Ox5WoNr>^U-oa6GM*U)e1WH
z;%EMUF$j4mfjZlq%4TdEZ)CgDT`5~3&x$*5u$0SBMx4AEBRVjXe2tjNY?nRnp%uH?
ziWM6$SjC}d{^6bMP}ny`zsU|xv770e>|hp2*BAazOER<06Peu`$h<#AT8hh>1eGn<
zIvCiC{N|mlMSmoelqo79LaPqfsHp_Ww>Hgxio;PcAwl&HKfbTSr|9qD@F`+;IG5l=
zgU%`Ww+?S<kpAh51;x5eZ7P*#!Yh-OJT1{HO<Y}}X0w7*T6E8vcgUJ0qgTjlS<^UF
zCtT0YH~!NoD+FrUwwx)&Tcv|pu1Qu&C%B>-Jya1-D_<M(t)g?}neP$BxvMiwyM>j1
zyeuh7JNp~reucc2n@dcuW_OY54Tq#c0>jfMieg7-QS6oN%dLGX;%||vaVk8ytEKiI
zB5sq~9?4aqdReGhKjUqMQL};8Su7fUcN5->PK|vmwUh5|INNzta(5-^hL1z+gNn&)
zQ^fARKC7(TI=q^_7WZ)95%HJe9xj$Y62jSA%+dv^>v8j{?_Ep`0_G1wuRG;A=JEa^
zU&q?TuRE7M?WN0E=`nc!$BT*0haaP>_UD<{uKyp<<$jTuv4JHNmstiT43`_C9$>fP
z3npJSmvN046PJ7DG$@yRLMEF6Dx9|&LnaRcf0W-o(q%4Tntom2k>qV<>AgPetZ=UG
z)sFGo8$nlvm=j=Z_bzxr)0n2$t<Ov#Mh4}Fw|g~}rb&QSw>29hbYGNGY7I0F9F$j-
zXm_I%_DT;R4$DpFSPx{TNAe1FGPqKLRU?az6WV$8&549I!fCG9ovCxCs5$0R;*452
ze^ksEI~+u7fwQ%+gOZRE_5V*o%J#P;Aq7G!33)vx31RwI_mhyK28_1fFnE-0FVO5b
zFqEa?^|(Y>(V*{jK~p->((AM~J1!H1WvVhDAb5`WfdA!{!}GQzBPTv(yL*P?L|F{b
zv`v(qwk*A21Rm*)De$e`Zl0qRr8l``f2OtL6lj#tl}cC|vkov#2pOVgi1*p6ifE?3
z?dzgrWHa-<-i~bMHa6{n$RTq2eW&eA2O5NN=0TYNw2;1Xz$V*nA15I4gsH-7wjTe;
zSOHE7a9q|S#xZbScn_6)kB?&z7ln}buph@}DP_gPlxP_}Evw#HpzVqO-jZr-e?#Hk
z4jtMeO@^1emonSqv$_j}sx1}><Uc|B0^zHli8wzDwJ11RwJ`36xU$waA3kd5zLFnq
zwr4C^3d@>b-)@YE#2Q^z!4EfgE~%&m;ChkQyf4+*ZED#L=Q~?{9_MF!S$^12oa$p3
zpCC;a{IEMY+tJeVowB}#w-dyjf5t%;_dM+tZr-~BBhc1+S2Pa?dwc8Zfo&;DYG~}4
zoIYelrOh~>=d*X5TG3t2?#-@{mkCdC$+|)u>6FS{$q3UB;QQu0{cmW8k`J&ULjLqw
zcE8fc@2GIhgZ}A@ei@sQ-Jji$QCD|>X;R+<8=**dH-KeF;pRhvogfT^e|}wHHN<0*
zLEOyL`Zg??HyDn4(;s5S00i!cqr;Rk&~=V0MtLhR9;KkmrqvOfuKfNby@y$uG@#5n
zRd`2M*Y<wh(-(D(_+y79S3^AJyb)Fs>1}Xc9wiT)!U39}1~V=dLB*QrCbk;qy_a+f
zz*i>=GW(WX>C$&sF>Y~oe?h^yFLe}4Y83>!nkBWRF?~Nx0+BGxXtFp1GHkoHPk^K?
z*-9(v{h*Wa;7UlaZv7l0&4QSqOVjBz(_y74$y^u#Hc#<c3%lkt!xV%gF3bukSs?bQ
z47Z^tPLm61j<hr@^f9DWrP-wggB@wIB-b{Mx3o5mPZN|9<&ZQ!e@$k%d@IuUc(4O`
z59f=QRhVi@5&g7EIy<CfHk|XI9uR?W)Wu^clG0RjO9s?yNj6PF-^O%PJuk5crw|H_
zw{gAhP`;{L+PFQH=V?r3%2Zoaxdz6?a8oIHVMYfE1nOyvKb^cAdbZsiO4NJqP?iQe
zX|>m~9o85oxwg4#e>zyfT3b4jM0q$ur5`*q*KD1oJU`+cy_>dOF(2>CldcWN!$6rx
z7IzFGg@a+~wg2f6i#LsB8ip8qtY^E@#bRvZh@xzdMR`GkqmK))J9#S3fvklG9TO>2
zh#4X`ej&qhNJ&kzQTt!Jb6Uz~t*P6)|E%;w?u=u0!w*Ahe<CL%U7I{cx~pn=8XVCd
zpD3O(k2hGpDToci_;Dh4K-guYFytL!r>qizRLl_ju<K8q5Bnit7DFpD8An(!j-fz-
zqq2jvGhBi$jRwg@Z%ME$Hd5v3SnimeL-KLlZpspQ&k^Em$`WxrL9wu9<*+Z!YH>Xz
z?$Ka{n@-R!7cf*mx0#)eZ4LU(bKdMu)69PX5yLohmystY6qjKc1QVB9OC~)^iMy%V
zTczCEo+`DcsM`O3=>RhrC*JKIh;5jU$D210qy8wPK2PWC^mK7vMV%0$RoCANXu;4g
z7@7>IB7hDK)b*ymAC0kr#+jGlOC}kAFEkd8;pZ5R>`!yf(f}HP;$&-0i9~BaesG$9
zUcXL;LCf>`I^BbHIcRU3`gNOLCLJNy_5MQB;`}lZh&t){NsCWkZr-R*f2VKVZhezn
z2T67qvQ;86siZN-;P!}gA?cK}5Lq_si8v%OG18`eArC|O;uTlEH(P(TSk`HOz^oh^
zjfAM>I2+t}cWY;LqySrMD#Z-G)C7^nQtBB65>>}j^)S<us)Ysosaq7lXI=@Sg9;yK
zTU1c}vBjucGve!WTg9#)U!pDG&O#8!sPDjJwY=r4M&sy-v9B0P-rT=vC*C*vg07?K
z_I0)7jc=i`yJ(u_TEc?F6N?ppJ2FzSvFsgcuk02{O6sUcF}=5>8|%`Q-ZR?5blIO5
ze9XE97;BiyjH^C#WBI(t+USS2`}n4j%KOrIv{3nyC^$o4w!knNJFv&f+Cx?2uzhE_
zF?et(m;IiGOpm!8??7r71oyfNyUWF<2OjF!^DIqmuSMBn?{-$a+`Wo_+-g5?Q|kXv
zcSt(e-pZI(eF<*A_Kjn#TkT7Ru<g0ZTo7wIt?iq7wV9CgWA1-?-CEaMqH=7xv+!O8
z4u!MIj<97|+n4UMdp3;z^;~FQ_k7@}Y_^dM&$Gj$V1fG+tj)aV1Zz2@Cs|i3f-TBd
zSin^)go|gT2iMbZt5uFz=A<9j?)GW8d|KU}F2Sm-bFjgW7mm+STMV1}efVJjKQ^!Z
zYYL5}Q1QO%rGL@lPX98WOTP#o^pk>)e%1ZRQPcXU<cmapJcaxH<sKA1{NMNoZ)4oz
zmystY6qh0FA8WVGP9|$8mm<gl5|>JMCOm(JB8tSE4Y9(+{Z}k#5glHh<w##b@C1U(
zKx@x^7?)Ck6^lnW^apU>P@Xacx(8S=vR3FXxE+a9n|QVvQ7M^d2ertJoVK)kzpk*O
zRI}K_gj5+!9F_bB(~5GYerH=g)b`zRQG>u*IOJc{e1=>>CuGqL<jADT^A0)p;SB?J
z$2FJH%Nr8{G&Yx!Za@^bF?c4*2nsulExi8#P0gQGmystY6qn)OJrtKMfhNFz)YC)i
z3;OBrAOCug#Gn+>rLXCOKfQcB{C?o<I;#Yz<<qxAYBw5$4r)>OGTP;je=MI5A2`uA
zmBb<CKP#6%zAUWeSJGFMs4wZ9wUfR;q7n@<5QIj;q%ZScN{ccZl-Zo{54EFUlJVVZ
zfK7@<qUJ59XmJG$QnaWwgIcwJ=}*%$&R`{dzNdbv;aJ*JU*k)A;z+(eJvf>4p4K#t
zQqSc(u21Gsm?VAuFDqqgL8VlFl?9<z?7}Q;?`%=0D|P8>ggFswTdRmh!A-(?D~wJi
z@%w`Pepzy79Is=r%YT>LEI{Kwq~@LRNI_`pf=dD)-u?T_=f8aV4FnE<`%mhMc7=jM
z49mCjXpPWl*lPJo&rUe+v|B(#)TCKZ$Ycbn4QvgcmV-g-1Z~66`ZcviVS-U=7@V!y
zr;c@kp{eqaD7Mx*-`BCRa@D8EGBq~RQ=&p+(<6?rqtavg8=q@e9QDuT@Oe3WAgx;G
zoX0L9D70j>XCI8&(kcjl10G>!&}?nePM8?2uUd&7&0wc>l(#i1flG-NU{nwukw!0V
zwkC6J8RV`_NUKt)P4$zv@92TaiY?7X(F-DuvqC{kCJLn8qT&Ai(>EIT1xNidd=@$n
zptLa7Ee}qimQALW(aMHRbf@mMcXhw&b_Ip&EL+aS2;g)dmVuprBG*F~?8DW{Kp;gN
zxI!5mQ@RptK$mx_DPbqXQFj}CTE7<JJb`LKuHDf(3Wc$Sjh4^nM;Wvg1ALVA7!IvV
z0plgjOdLGdR*SG~jwO7QzW$L=GPQ=%5A;!Wa8E#Q9dVOG%!(U@UZkVu+~`%Ps9|Cb
zB4;I4!3%|!!DwfH2_R88%bAu|IE)W`ZCs&X97%p6YR}iijwXepUWPu;OJYC^z)->0
zOp<m;BBh=lf{u?l88Rucsc%hN-C@N_3{wa`34arg25of-v8&c0%o?=q7}uJygJxcy
zw59oruA3!rpJ_fo9>bCh;|6`ILXAcA4Nw@|ZwnN<gBMVL_}4ObqqzOh19wF-jD!*6
z-d3J!O%O8lM1nbtHYqpFVz~M=x(~h2qJn-N(v7A>QVA~eSqHEDVEAzjSXC@5l69C<
zWj}aRvB#P~ZzXD}M@rVHM`JuwQ}9SaWU5Wb(l=vCdq)BufsY=AY`twyH8TQ*+OtvD
zq+&-l)BJIN&1-A9!cY5Ht8Vs`T}+8V1n=3>Y=GUIo$r8xQo@2}#L61^8RtkEQ?cu{
zmjk4$dr4@#!mJD!Q@2+PS82d%OgXXQnI|K3Sh)k1k(d}&cR1DnM>EDDy+;+Q;K-0o
z#3vt`Leo=r9aZYZONK;aDWvCCS6f?{^{vH~YAsWL)zw{VkFu<drt6Sc1B$I6OiSA-
zmzFlq;1txdQ=Jt}Z1m?clhe^#HEOOU4`Jk?>oYtyeL<cLK|u{*MH+vm6?rWQ0gbaO
zu8tj6>KfY}qmUR0uw<;Mf$LoD!d2!F-tPu8ura%VQ<~@=rQPnWmGEfNb8o50>IkB1
z;0xP-kk_2sOt5%O8T-0Dw#w-@R)Tn9D?#Jz1hV@pF;Y3M0DUbl*|j|}SRQ~buan=0
z{Of#Wce9|atJOk!xOQEq-WS((hUdFMGS-FI-*axh6DBLfWDk5Y9+kMMxx1N~`fjGy
z;8{^d6t{R7S~hAJdK2qb+#Tbt#gulX_e+a^D$R9)7QDb*m{Z@0v_@g&<mQt8xX$)W
zY-^oYE%SpqGSY5krXg!C42Ek7{z6?DQ-gY|KigXlnIVq07_QC=!}rJvL8X1MmEPW#
z7HS>y8Zl}$lWrldX|hU{Wt}uOL3g!T-;)?*np=L+YpqG`u<aaHYx!JzQK$iyy@$zv
zugo;|)K{8-X=R02GkY*2+|8a#sYNtH#ZsQ8w5+sHT2^#KFkPh!jY=F+mS}m}R48qr
z0=*f~i0GNQx>8Y37{wB=LWS_E%%6CyeAgYBbSz@Po=t*FYMax)7{)}nNOdizqpXEO
z3w<-^x+b=TzMiqXYdX<va}rK>4oVV#u&QfTmLWlS8#nfY9<*#lco^pN%ra%TqLR(l
zD*X+KdZC-&o(QwTSb0kxvt*4kS%y<<fnF(_nXYP2-T&`~V+N~R?YVxxQ=(x2xr=<V
zY}_g-O4kjfKe5rJQ`QH3jzqbpC0U5oR$x=|H6f)K(*PZPxrU88Zp4VRBFGMZZ)0G6
zC4)FV>DK03D#U=DIoD=B>y0qR@Hr=*H<LR`>?(XqGV0fHi>FqKXA?C)Xt-BA^ET6v
z8f4B$6HqtN&+p!8r-bR7cQsmiBE)x=%&G*&ULdQNrSG7hUuu>@#f99-Aj{U@>$o4>
zcf;)Is_i~)jF8TZyRjJ&tQir1E%xZ2Iq-xnE+Md@qvCvAuItz-=W=g$CWuZT5zc2u
zaXaJ|&+|<D)K&!??Kzc(ZH43QEVF0jfonUB)I685b}x&!JYJKQSrjZZw_@;QOSWYo
zz$Ix8W$SLfHA5y6hk|j8I+r-#-DGNDu?aa?)`FZe%2Fr!?fPds<BD>BzjDxEgpTpr
zO*#$ZCY=Vm&J3x~b}i;65OI%Po0A1mW>I9lQh$|R(~(>0$nNTBlCJqj$;p_H5S@7o
z#GOYI8Y7RifjQJ)bXN4{fQeq|17AixN3z|3?f{KI@zJ!$vOmfS<24_(56viaz9b~e
zS>+Y@d~SNHvC7i!My8v8y-Sq?YxSs8KgX~`>ayC5LJZWa(ehN&A3ZAlT*%bDPOAzt
zj^QJv2g8m?tB(4JW;=Ea@2kY7<AR*Fx-OhM?VyDUXl>ebGsAQvK8`zslN*nrJZ!+$
z*#&c2m>Bq!mzLRi0l){{(KHiK3`gucJ90*&x^84$nJ6O0*$v)*?Tq{T=Wl1v{Zlre
zSGt+1ICVBVsMp41`lcRxZd0x1!jkG>MKCM9)D7TUul*_OgaXU&{ecq^;o(Qz`%@`{
zjc~_%e~hBt*^s*7h8{R405P6XR{M2_^T2!St&^(beQppcz$zf}8TYObDl5ENY5tlk
zgz!=g_y&`|RT`~-bl$c-ukCJU`CB)?+LBBDZQDyJp$u!=+jhlEc8A$Sy7kDZ<-qoG
z=E~)zFssrfJ7cF!ar-tm2k(TJfxEd5*#&=yn@c02RQGiAnc_X&e5QQK&F9{GZa$gA
zWtCpiCj=#8Fn81^qxiF5muBo<xGv3l|2$Yn*unU_C)VYE`edcAw_;t;0<`DU&R#9Q
zgmYj#8eq7-eD>?3)aRijI6XoB`9!ZO_KvWp{C@Y`bA2Oe?^K2z+Dl@y=9Y%v2^?#a
zzVkg*KD(y&|GNm$uaVsRBa?L3-tZ{oL8Lq2KOzbNran5_d!pt{@s3f5QN9p`Sns_k
zlnw#XUEsHW;4Y}NCTyZT+y$TO3eMI~gS%kUnqE-ypTb?rV!6q^5qFKjvd?pBcTOr|
zx|$ho@Q(DulqLNJDE-WUCY@S-z3f5rq)PV<;<ncF7R|OB1Q8V{b~zuh$02yx^gE{d
zUzTNR8GV!#7C_kH$nbdXH+Z#`eIii)4}msfnF?i>lP4z<0X3Jg5CjtfH87XK?Kvlx
zr;H|)e>0@j$A5kN{Yf%{QUn`s<Inu`{_*thrx{&il+ZpLe*E<*zR(JU@^}EW!k|6;
z_S50#r+>^~>t>SH<Ud9pe*5Jxjr_nI3z9e{tu<!g7)WBHK3E%simx)q^qZ81sMbWa
zUg#qJAWT$RM>~KzK-?>BJdDuB-_9O3chxjkf0=n#B&KVUl_9<xpex=JpN+{FL`>4S
z8!)L*y%SokXptm&0+66;f~J`+dCpLTj+`og5*@?H4{ZlT5ZZ=$2QZuXAR6Itg^xMi
zzpv5)6|VLrDo{>RxdW=UB9Jo8Q1v!71_Z7|CYNz?r+Zl`rjc&@s(Qu?MQ9WyNQTK5
zf979!J3d<pY|8uU5#PlX_`+fj^(nrK^P;NxgnPvY$r{O@SRK{K6?~#2h|uFoeabk+
z)$tj-9iv($OafzL^B9pyjw3!FBLV5vnL(QDlhTsV2#CutWD88sOlW#vh<}fBGRQMi
z$yh?j?uk}OkPU3S-=`*l+#@J@!}Q+fe@ijA1iE5ZP^*NG*cF);R**c3lz<6!+Y(Y%
zXy>azS!R{u3p2bzgm!hOIJ)$@kdSWsvUVl3Lt_DquJ7Af!@-E)objVGC2hAbXwaYk
z`j@zrmgwyzC<u{p$)C!S>$`MZxOJC)N#3P8jyo*)qFXN(4O*CH(92+eMTdPAe@bNX
zshDJ^?NxO52x~BO9OiDSNWr4vH9%2353};<qW3bzk4>7=NHMdjY}GnKPeZnsffJ_f
z*zm|Eyxwn^6FRzN7r*t|d$2;U3`;6Oz^d!Rp6?9#l_9Y=$XrO+Pp&!T4Q?eD-`*zQ
z_S$$+?-<+a7mFwIHU1Z=J;-1?f4XlqrGClpXG+FXChK*T$Y2G<`4@Q@w;lAV-f@sD
zTG2Fm&jH5qrtc+&6=`P~CV3R?NO)ec-HUd$9jJ$Z-fQGCXu^dc&|$vxkZ$KowxLa%
zp$#a@Ynk(jad(V|jp++rn6HvqCks_MnpP_bc@CyI$6AJzX>S>=IHlx1e<QlQ#WLnc
zHRU)Dfq3u-^c(X9wHr!v8qxG5rwfy++-ZJ23vV>PVaaHI4>3CvgtIk-srM{_a%p+3
z;Hy*<>Ry=OtlBD4iGPGm%QG*U9u=wD7vnmys@iN|`OS6+9o_;%UApiB9H{(|H)r-=
zFR+ayw*)Zbo}p2ZEd5nye-dYemoBa3V3IS@@-`K290v0U-60l=G>5n)kCy;S6$#CD
zFP#G0&M9?DpY5W@h2VgFNe&HOnw?Gz>dEs>6djn3z%sj5(zb(V3%kOmm@-k@qN-R7
z3~Ku&RnW~7GXps*T486ph|KA@Sw?@yqSrnM+p;^tA6%|#RS|(Me_mwEK~WZCY+OOr
zi!vPr%9`1`Gk=t|H^e68{FIC3b&2DgXECJ08_VXE=jnQPwf3q|GB0y2kH&k96*gU7
z8I>#V)o61)E7MXckgu>af7PmUF}c=lLnRaiNIgT_bRmuJQC9CqrC(ai<+pD&k7J!I
zZ;IKtWHy&=rkf2pf3IgQcBY%==3P=(F5Aesc3UT%6HZolv=so0ER#Ng?{XxS^|#AW
z+WS%(Cy1q#ZoCMo`*#7S6hdl|in{Dn^JJO3hov`_G%MrcTYjAic-!=~;n(v0im_|b
zO?=DqhNn^Y$2^?p7h3`U39rNyjCd7KY3A%<tlSzENxwcXf0yLM;f-zc&Tgzi5a|8)
z?8a*41dzM9ryE;#khL2-jP#8+?!L~uEG;YT>f_PdTu7TJ8`EHCh{(E#ulZhCmdHMn
zb~yw_pfbNVho)T~z@ZVM>gx2O&#Kf!Z&N>HY1V=zrGIUh=%uypO3T|0J+@l-)WZ$^
zFoCF`1li{ue*)P%9GD=Q3xV*ChBw5ZXmN%-wa}Z$w6G55*=p-3;l$T+e`}u#8ge1i
zCecey?3zy%HqaVGde7;C7uvfO4j1;<D4briW!Ecb7G3lSx8tiMd17-bzA7px;HU6i
zI+2oFYkv}c!^p<L21N~_Oz1O!Y<|ebk&F9cqC1)1f2S~TAZqUj*>}`TVrhAQ?IO9M
zr?TIr8Tvgf=6(fdOY+vOv<%$=!LqI=a(|^@4C?Fq>p<3kSMu+|o#{=+BFQ(NGyB}F
zOGj9e$M}#1?bop+n>DEV6Yg3}pZcus7EK>|Xs)XpN2+{lM<vRlG>|D&jk-i_{C%sZ
z>|dr9f1NN^UvHIEDoPD2X9}8+%q<<g8AZnsit-Ur`dCxujV*G=tUFZe3baJ#Ydhmi
z>HDb$dp-1J%?&;+(zEu#_q4RgJP@VbY73z^ROi}ul3c#3NN42FF{)h=A9uNCBI=RC
zO`R#Un$ni$^d?t}xt#3^tUb8eQ}|fHmFD^?e}lv&JFhB+RZp$i1=nVYzYf(~%1HFR
zxC9DAYtEF65V^RHce*%%r8|QKewKA^D?G35OFl^ie=oLic5H)8t1f8;2mOoC0*HY3
zq7YnaZceLI>qtGU=L0%r2z$hC*m3l{bM)1CNyS<_(+yg`*GAHca$riDW(2uJ&l;2b
ze^l4TY2l;J;Fbbg-wgqYXO727TWuC3OCyaOz6Izy(Opa9s@;_qiR;=&t!JDYH8osx
zxTXV<Pc|Ar8Ok*5s?i{c=hFWnjRwS!KAC!4)iyzVVAVE3-l^KA+1pjys8NQ@p6nHY
z${}YSZ;Ux&Z>rwIY9sVPM7fSMTq@<Le<X5YHT$O0sL0W&{K;SXlTaH*`m&8Q#Do=p
zrRhFqD2JwnImjOXl!MvV>byVm!5cvkKx#W-fbvt#qD!`=V?`f*=D17!jOsjwksZ5=
z9vH%kX-gHs$^gl`7_jpPdY{O;A>=yy`_->N`J@Tqf2@K{E=Yg30v4<i*28~!e*qga
zWQX{FQow=+kv;?XS^;}0K<pH4*DEHFqmk*Z(Dm(buVOV7xoPggN3hfJ)x#2HaDhCx
zopG%-oV%9%uIj5j&Y)K5@O0j7JecCcxZHEEx#M_vwvSIYQ-@<a@~TH-7Y@NTtePR6
zT5ZQ8{OMdirmLeDv-YoLwblryLRY6}@nflA{*av3Q~34jfhp|>@PQfa2zV!<9VcHp
zg?K~?$^YaqZV1;;BPQ%_Y_7xmmX1xIQI=oQw*LT-M0iJ+ktZh<moYOy6ah1r!HXX!
zf6ZFUjvTuU-S=124~#5QZ(z{KOxt51I~%hLvbZqf$-)kDZx;FeP!gq5Nmc!tig5?1
z=`OX3I-)3&hblR}O>+8tcz!*6cu{(C!W(cny?&p7evtGZBsE$yIRIf5Os_wuzdcHc
zmab>Iw3JHG<><SoPjh@w;Hd}DFF?Nle;gm225O-jsgU|{Uwol^dwkHKt&UH)_|)+s
z_62kN`|Dp1k^xE)Z2THO`nylBhu;r9T&tu&iPN{Ahxmh5Ae6^R2jM|YfBa+me)!T$
ztz$`gB>yvV`r~ctTE4Qvf+U4WYmHeb3?yk$Pu2#Z;!jy&`lFPFCar1G`a~D0e+OYP
z$qz3cRwZglsyDqvLrb7Zq9L_u%o5}GbQvDS<WW-}`joU})Fe&p&utNre91_JQ#425
ze`)HA7MglNg&0I|zDE_sN9ko$qZ$Np)WA9*wE8LbU+ADp_eRh)NG~|cTBL9D5|x&(
zpEuVBgAy)`>k}B$OVR9AYoox#f5rjlsl|ELlrGwlc0wz+ryX??pZ*+nv_!SH9d**n
z`j`l){7X0&Ee+1P9?A+$c(1dYbr*BV*^(TgK$kJp=?DrRi$bD@b#Z9Riv`8!wm=M)
zenw`WS{ExWXgKEIPF30!%WiBL$vovA>6d5zzQAGFC<Pi%o>vM@msds2e?#lnYPf@1
z&^lqmpq8W!#TZW6Ak*hZMr*RE3|I#vKSuRzOnAAo*4IsWFO&^y0HZ|BfZ~Eh$?%a?
zb7N4`7D!jB4{Cr`7_`46su>rTBJAjmI*HkI*^EP3NY<B*Sj)dOg+)udTOGB4?WLn!
z^%CH`2smQ-6$X}_De@k|f2aIi*_)@V;t~P&lK@zqatT~rZ*|3#?{M-eTCN=w{7xr-
zi*_`E(C9{k7!3A3$%R&0Kp5<ZU257X*X0%30OhZKP#7%*eA+FyFvbYyVb9&%NxV0=
zSoZ?(cwdS^tAYx9aJ!`#L5#Y5M}}dv79Q~m>)FKy5JixBPa9ZDf4rNg{C{j<32tW_
z=!EaVR=gk(!z$g&R*0n_3^?~=E2K^lyxr59I*ETSkFo@}vjB=fb-y)r!teJeG_H{7
zuI4Q7w*zgxfIS1tQsBrvZ_f~|6=1OwyqO?sVI6dx)fxldw6!jJuk%9Vr$BiMO1s7j
zHD&s0jxlqIjV_{Q^dY%6mw(dG#KL&XoXU6NXxcwE$*Qj{z}706w&X3+^f#w$(+xGc
zL}j(qVy=sz$&?cz(%Gsvw#Q;3aVXl952#cJl%Fg5^`M?L6X6rX0)}c!@6<EIyOf0M
z{7h2DEU~n>`c%QAGtT5&^qd&$DxaieD*enG-W)Nvvn4RP<QSkumVe~c-gV^FWhl?V
zVn#SD%#5nzBr}iNqS@Lg;!MMmvClG%vxN|=wP)8~6D0<s#7kjG2NzI;Rm0&WPa9vc
zC-JQGPv=ldY2{UxssaiV<`UlXoVpYTc|$FuN=GNaxnBDEe2O8}%FWh}%rzLxzbd|v
z4aKcT6+)uIfFwDk?0<^~jYO|!Ya*&zj{&1kIJ>N%2YGi|7I+Fm6c#&a%W7DvXqgtP
zUh|4RZ+V`LZ#HZ3mC*Uh-D+B1i(tZ_lc(8@)e^%gqgNY2?^1D!=mhvRb<G~BV;SW>
z@7AQuduvvx$^P|FvH~NFb-$!!r3_LEy=MWvlel9t(n-FMjeqpLZ?c`}HyJ0nX;vzl
zjC0X)j-GJwzC)tAU|9?;C!JU68m$(T+zh(j0W;-rbPd`o+KLspQ%NMQ!^%fWoy1t$
zi&u;!I~Q)0B1VF@lwnCfF4GvOBy`d8^$NRRj*5!YTiw{2X21I~(2Bm;+DSWer1W4V
zaDw#?5){cV%zuXTygTPX%3RRkFJ6>xR8&o3q<CRgGtu?M@$sB&n9JqG>Bp3}J3-V|
z1d0IN*r4w&`l6vQBZt`l#_N8m>_B>{_I`=Z3P(v|ul0{-&osN<@7x~ftsqCX;Fidj
zowwlryM`RSa(h;cI*9{u<wA^nUa{JD=r6IStAcTSX@4eh-jH8}X%R#+GltJpr<x-u
zbDq_R#WAOD%(Gd+S?|q2n4qr%TSGM;cC7%**A8do2n*|1YstGYiDh3PCX_H}=;fKV
zM&?-!xIuX~kOnDK^gf+r99TIct*!I%`6hi(lmcqs%f6FRp-WcX2}`Q9vl^LO8T;6j
zMMA@<Y=6V3@`SM2cB{72$}0t!++vuom3nHj$tt;WyWH0|Tzxq@utH9Cr2O_nukxiV
zn8TbgMZ&_k*7!0?a?HV9ILdC{)hI;E&&F2K<_ZA6$&t2HE63IzBP-`36tu4(lzNG*
z&t1cIHzXz$rzRC}RqC*9)@uc|=0K#eS0iXIw0}>pIs876KlnT)r$Bx}1qyrronkXY
zj%{tqmXrnv(Mumb{QK?uU%vfDxzqIapZEk(PBLi-e)`F`ifFf$Vfw-MHtta!^Q|D2
z)e{Yl(o0_-tPS6$gO1f4MKphI{SjLO0oI7k!-?p+*s%=IG!+a{;@ryP{b(C2syajh
zU4LQ3yC{X9jkolCc`Dw<tGzE>>8XEChwszjOVwvk^p{dlY1h;VF~BKkRa%<R#&}F$
zSRKKU!ECg)6P>#X>uuB$df_Phhzr7_)ybBYX2PY3WEBaesY2`8Hx$s#HLE1sB_)v|
z)=`8faz|DPtk$gB`TMt@{M;Ko^||@Xa(^BGDMc`rzIY|_dpxOrVf<?N;NIN1#JQ)V
zs_xkE*6rO=9=Al&c8e`#5`ZsLgD+3?VV>!itF;3ljL}zUXO8-}7gHMzS`w*TRpCYi
zF4N`SMTf8q4po1?*5uh~t(8qWZ4cSG#B|oyjah?feRa?931kX7?U?~*OI;vf_kUZt
zoLi=HW^~1_yac#X!raO$6vW(F_vr7;E0hi~sq2JoSGC=CoH%F9vn{My^K6;fn*TK|
zyK&TcXu?j}m1mCPDB=*y@vBF1t{%lHHLnsEl(vuG*2{>O^5Brq7rFyB&N0p(u$@hd
ziO$6O+fw`@DSJ9o7dV~IHwn9ut}PeN=ae}MTIL>|3i*xI;Uq+|RTbx0HBpYtuFY_k
zIn!MJF3xMV#{U8H59)E3ktZh<m$AY*6PNv;Cc^>ux51z$Vh99UtMQbVktZh<0XUZ-
znimt7=g>DOmoSqm1b^E&^6szD513(*#TO75Yj@HC_Lf2426I>ho9rQ3?93eI_qU2<
zQ4}S2+Z|<gHh|j}MHTD)QbkW+COv&PT%HbZ9xYBmMF+u5PoF2C8=C%trbX|j0HC~u
z>FL|_^INTrr_Y5xsI|593G_SCw>upzMA`v#2j~vK=@<}bg?}!zHMFnpHIH=dPe%mw
z$WH`*>U4zs`}E7x-w(PX`2OiIBIJ!z=<W3B+u_%VHh!4$?t@k~MK}F1<z)qy{!Lm1
zkAXla>wNUnZ?y5vf4+SF+oykkqQ`#ct$IRb9HcmXE0>P0ql-9wt=FDdfRH9oi7>~>
zMdfUu4f0@H_<uAV&;&Z&#cuU0uO@{Njp-h?pll4hQ4Z*q1P6xLR^_~}{i0wsACYNl
zXz&$7sn7ByJugr3C4c&B*-B6SYdU<M4j(I5o%cS_mKcq7^zDfb+~O)Igg|Q3aaNYB
zSLjGHgexg%=dP^w^tMH#NR()a!l(jij-A?Rjb>Tt=zo?af+Y}SNxsMRD|{GerIn_n
zPPs(SIRzoGibCaX;qd(X)3^HE7kcVL_g&x|!C2ylpN>8dmytnbV|8)USnKY(x!Nk*
zCO&BGY_=&mW6QRd&(mQarw*4w)N!Wj<IS7^BG#H4I3ZYVaKMtmC)ZeT#a{^t80fpP
zfLxCw*MH;4ZMi@$$FP&hSG0=ch+Lv?a!SteI`1&N<6KzBJMd9f5$iMg`L=o~cM7C=
zZ*gR#L1ZjMDO*;w&a$;^to}5UZqTS0MpCwsK6%StQ`h^ptfOc2!CrX3Z)BUY!--z-
zH7xwCpg^u&fnK2ygzgN^W&~!?sq!ocAK-04!GFOTT|5__XP!RHd$36fy`Wf6?#?PQ
zS)5$pi~K^BOOr?57b_$?{|2bWF{n{hx-RTBTzq*t3nFmX3yTw(o)XNQle^Z{(1#@q
zvR;~)PrjzmMw7qOX2e9FpMMIB(kt{?u?r%YDj_Xi7jabra8`RJ6X0qz9{6;{Wz6PZ
zTz|&Dr{%U9(~n(H7dJ<-%orkNBPVgmsYFh`Qa5KD^hv~e;u7=R3t>bm#$uE8S2F=!
zbfn&Y>{cK0lh5(Gp%YMn3(e=m!aR(3dw@5O5w@M1BR-A?Ip0anj6{{SHA!`^7f-Q}
z@O8B!e*kgBKtJwAq9K(5SdkcKoBKXePk+Uw+(fM<)AP=3P%Bo1)nSF*=(m*4C0L#L
z17%7yC>O)liR|R7Thkr2Slo6t7}zu;*fyi3H&MwmYW($KxFQep>nh_%P4SRj1231f
z*HC^z4uag{Z~vS=A3lyiE9vBYa+1M+oGwCHto&4KprnaI+Iz|_d$S|0kB-uP2Y<?;
z5f0(IIRIpD$xa;xo@XhqAtEJMlsOL&T}@NO7o*q}GM@4$qrX_`k_PZzfj1*$qc`I_
zkjy>RMZ<2p!h;qrrQ%~s&AI%x{%-t?p`_ZU@kH3pKMXm%U{;*BV+JD{_HIt%G+o2{
zir>rPe^K^Sc6ft(%6<uPA1_|y<bU+3Wk|gU?^h?Pd&qr_xI*e8t=*QKkBUCylloyd
ziD2?7VBAq61H|u>z{--LC*N^TLjGj(7c0F%3%QX@A*LHdunTj&rP>4bWbP8;ei1zf
zy3E{7zVEAZNYR<(vm;vuh})^>Ya+k^`Gp8D?A}NBOS1I?#9+9daC?ejfPc877?zNY
zmd}>Wy2^-|E3(o<j#fcPBjjvNrmH}unm!LFy~6iUBCLa7q=v5YJPy#e@~@XFCmT#o
zdWA<ULn`1<J3j59tJ=A;*M1}AEUO!GlRFQ!uajY7^*+*DXzpU(kFwv-U7V{$3!;iT
zuIQ+xw3E`>7k_Q-oTf+v27jNgF|9@wAdM&FCta}sF+n~}=e|e~KkIuQEqWq_x-_?x
za>;Mz&R5kS;alx^!P~CCQ=}w4p#rh*lk}anr98s$^m9DMWJv4%ipOcjS$Lge<MaG<
zCMu-TZh+RcMRu#%MaMlE#s8#e2jW~JD;_X_cdskMSlMmDRUt+i>VL@w{)|G*+NkVZ
zh1gyGwZ??;lq!i_c$lWoH$|C1x&nTLxb_dbMMw)?;jZP%0I?7HV(C~5{G0?_?vI!u
zs$^Ct!Mj;k$K)jjg?2^lRleR)_h^2n<X5;{DO16)o)4S1-F%k_-GL5zmDVS$l8Ac>
zJ4dRtVjcHT8E5XDf`5NK^gxwV2qSd8Ek;A74Sjxq7_Iioxp1QxEgKI7gHhE}_sS<0
zT(XM0E-|U<UsS&YV6R~(D!9*mBMscm1GnO()%}YsNr$#gm^&xmQK(wYJ$(1FqYw@;
z&q_bBe??UB&AW_*=00s(zP+luTH2D6965of3hI5`h26{ACVzJ<p$(<!4#SQvc!2ol
zbc`wsU-I`m8tXvj3&uL^HkZCyVBGG<!y2XIN`Qb@UXgl5aedrfjA+-Y*LYCgq&*dC
z3GuJh@G_3)N*ecl*Ymp8_u5K1H%5mz(TMTo(IMk7@o)=p-9cdUDA8@BLug1Th#@jM
zgxqC`(a%FVq<`zt&<1(1Ei6Wd+UjC-XnuH}2$BQ3CBfk@HacWNuA6iLoIL%zb9Bgf
zr40fO^FW)c|EHrv)+%GnE~7)7oq~2Z&JM}$t{0PieQFi@4Fg2okz3;cGjVrJ5CH`T
z{H_Ti@*h+#gc~P_WVd8c68PFX&ajxqsiPjl7*Tys+<)Gs#zmv9t<+JHTv{&6%h$eP
zJyP=5#J7NQnVV4^Brh$m`R@3hyX`HFIPhh(Wge?(V=m*^RDI5^-Wgwp(V=tU&dc*A
zohUcwO`5ViKX1}Sf*a;d0Qp^W+MG8z&+{hdQ(=PZB4(Z~3Cr1&b;N8_DOH(FSpj=~
zUzr(intwzAa$+vbIZ@3N@&K+~i5z;dt+0k|Tt&U(VpO0_*Oc{IIJIP6=B^&v4pTb4
zd{xKP9_IqI>&7jI5JGQUt;0F=yZ0EzAU-DUZ7un^=F^;jXNpY0y_rC~ZQLYL0JY;`
zKVZg$+xa=14Qk?+K7jRlLnUsSDRDl(<sVKMhkxuuU3t!C9xpWMWZl}Q3?rPg5{p-E
zJ7BF)ZU3_>)%(1@O@0kpQD&L9Y^y8p7gpYHF>Tj1?&T#!hi&g4^)>9uXj8s(>54Gu
zOv5aE>nH=#z0h~nw?xn%k(pMzp<=ULz__$9QD!(>z?Ane6=|u~&fC7^x<{&>s}>hg
z+b`IL`zd4Rfyq*2Y~5M3!7QbO1NFsEmj9f<ST97s-M~VXzz<-Crezn+{tMK=?3aNO
zC=>%VH#L_5HUc9IH#RR!Wo~D5Xdp5(m!Y~fCzsyeEee0Zk4`^uIWwFgSp)(cyG{)>
zsA0K4T16-n=jym|bcTFJ-GAR_xVN&-tCQu^`H}*D8SZk)nc+P1#>-kHq>#1bkQx75
z3pM+=1XgjP1<9STAeBH=4w`%pCTi;h1>@p`g`(*mh>B%=-HP=}bwWd9f6^!r)lNwu
zE}cq&B>8_Xtpv(!PPITvoib30jVy7XLfF8-02WF9OhJh_g_Ubjr@(!Z;#3Pv&YMvJ
zW&N!M=G32YU@`xea^S3;Qc%_Ii4s)6)gL+Kj2eTo+9`ZcATQoxxysHSVcV9QQ#D9}
zqCi1^>|z*85W!nf1yKujaNV-A1kFe`XGx%Db5(yBv@Rw2YM27mcM5M??yF&PD*YK)
zgp5vMDE+N~#avxcLZh6Z@K5P){bXwC&#+NzE~DTusY?MnQ0`Jl4Juus(t_sX^D==_
z*MKOf#@GWx?Q9rM2^Q5?F;HE#xdfO&r)t0vx|GTRitb+ssOCb0VCh>lP+iHEg5~U6
z_*;L?{R0Kn6|WYoxJ{XsVAZV@Z-TYrZ;b=A2Mt55g&@2X0^x>kP<$B=#h;1#4tP;=
zXUVc#v5O6JDb{&1SavC{OJA<_12P!-Xk`FVg<%F)R^@(XFqg{eAJhUYNW&H|{R-vz
zr+|g^6D<yq>`X!8T8l3SJhenF7)+?q_*Q>_pDq|+*K*&(F{KV$1I$_>y7%?h!=sbz
zwwl7T;rqY;BTy=_LJqJI6|9`Ud;R+K=;(%-%oJV0A7+N(;qvwJ;`q(ki$!pM@hl8a
z&sM9&#d+Xy_^0=)Z;w}Jt3{yk!Q<um3VS?;F;D`_zRP2%mk3tc2apm5C*#09fWm*k
zFgQ9IzW?iw|6aWC@1OkY3RaumJiS=HI9{wyhNq7n56>6xSD113a`A92()0hmS%9^d
z3vhhCTAZ)my65U256>2Fm+vlKEZ&aJ%pZNX`03TzH_P`Yt_4R)X?%+BFYqO1Ng8*(
zR93m7EQ7Ml&FD)g>IU7+G;ovXRc3#pSK3AN%9o-$WPayTm$}rslhpQ!1E0?tF(j2E
z9Pk&%5S$!&SRDs~sF&6Q*eW$zPeV+@psuI+fN4DqVJwlZ2Xzda<Fp<Tr*HZjyAGZ`
zxTib7v+?Y-TDr|?wY+&+jSgU=SIlz5U_aMJ+uSKN+ezx*gHzhDh~`}C?iPRH>CZ>H
z@!an7b>sQsR*U#}!uCLU*zoresITDoGfm7=ovYDyQe(M=8{C}ro4LV{j4}5T;STP}
zXl{(OlQGh#jFINXn483;w$seJsCq-YSM>h5Cdf12Zu}X(eRyzwz8n{tlzGw%E@!sa
zTWg|DL}&T*lAr$laQyDam5+b^^6LDTFnqJT_-S#`_f4n6pNA*I!xI_@&tNtuBrUqX
z5rgqUY}8h<B1B5zn2Meb4+54y9==^ZUk;DL{kQn&)$%+_%wHWHT~Wp^n<K@KUhHrK
z6A)SwM@C?i)FNuo+2$3sSZZ8zDO?D3DC^yspe&J55fe1fyr>1}y=H$RXaL*Hn4koZ
zXHj4!=TcRLZ4aub78B4naUqf-iodN)FcJW2QZs{BP8gL-CLA<OxA9^dv!;lUZ!rs^
zM|a_07Jz$8l1*kwvRCvwKy%7}9Ga|aAo7G!o-oQ2Mn|c}CdHbdiQqMnFJH+-<mnaP
z9pbRo*pN6umgwyY<s5&#cVQ{V%(NBal<?>7mMsk?gL71d8G8NB0@G<h1vqPR7?9^e
zu_e7^LPevubhhyDWiakwX7jFMip+3(V4@`Na-^8#P+a7av726Up;V*xX3rKbZgMrd
zTLn=RxO%sS1kW!8?$0}V9IYJ4C?vx&Gr~=)@VVdxJh3<1QiXrhr9+(uL?-W&uDQ^R
zbMY?N7B1HC`~%ci_h6GfI5Z@$C3?HiZM0?Q#$qx1(z`c}38}PXiW#nn0Z*<HL@6fL
z?Yrl+2hWrd`3oYA+afBDN%6ukO}jzy#Iirdd+t*`{on@panF5nvfJ(M(OK*hnYPS3
zHZbAn0r1l^1u1_Tkax)D!bU#uxq)737U*-^CD7}Y96V3)Jx}pH*Sp7Mks8hKCZb;x
zh-J9riqgh&;sZHAlIT|v2ug54;6)Ic9fl>vb`29$6LT&I1%yzbs;#trgvA<B<q(qp
z;@42{1^u22m59!H;O6%msc?e}&Ixr(+kk_n+a$g^zX^X(7hcOPW(w9tP&_Oq5U2WG
z;W)@XD0fDS&C#Mep~bGG#X6<MI>AOK*ytW$!{#|#MLJs|Hb&$c9jW1TP*5JqXc4LW
z-m^9!#jlYnA|e78yt!UexxV=M><z2|pX{Sl6zs`kiH?#%GIAzo1-Q8mv?wK;P-#kL
z73rIneMEn5W@sPAibF`=FGslwgCv7D%d}noO|V+}MC`IT`f|p3r+kZ<-rGQS^?wt)
z%DjJ6cIqFME3wO_Qqm?^+5}6RU}^h+Wlgbq{(^Xh9JzN07+IB<e_ZE?DAZVh<I)h-
znx-OYA6A)1NN9}+q`j8|h7eo<&s=9rC`_fXlJ|d#GyDS8Jw#RZ2<|_-Q22)8I7MQQ
zp&`-Zd>9@kMIcm0ERM8ZB9A{eL8^OQQwuXoN+yM@Qff8rCA~))596A+;M}7BjASbp
za8<}}5|hGFWvSBAhQBDc>S~pFzlwTB^V_AdAKyxD++=QCQSN>NQs*}y^)@#kw!VP8
zbRU1?)#BQHO#Q1e5q0WT)v1wHr$$zt`d4*wI^#|B9X$&YLda%tC>3ggokpO<e<2ey
z@lEY3-XD;(peXFWuR^Mfa!nC!>6#+ZS`o2C1x^^=8wu(@B0gno`hf|7Y+lxEWdez@
zaaAA*r>#ZzQY;7{1S`H|0;x{Ixol&Ck|KXQuT<dSVc&|fn)cYQfQN^h0$7j?ssHW?
z_&(5a2y(yXZVa}8Is}?X2zfxh<OppXCe(R*i2WymYbLUe>F_ot;9-ygi!>7$2Wl?9
zu8Vj5zVL8?CR<eaa{mdKpFRgz8LUD=LoHkK>Aejph6k66cT9qvzdd{6fA9aiTD>%R
z`^RBeoX7vX`sLLd_fT<p@pACd@ZHnnfBta)U(1VM&sINt1wQY6@b0y2$r=9#%npN>
z!KOkJ12i`=ml5$V6PMq-CcA$_oC|*c$Lp^<T^Q6V*!VSm^uJzScYp7CcC7&g7KT5+
z@8SzXpvdDuUL-WbkH3bmyWe`M<6c^p<Ug#3AK!*<<Vyi8XaP)H%iI74I_+U?P!wNf
zz|6DOEJ7O*+8pH(cn}&PKYjMFO3`S6o_mTcaI(N@1U8c>XKKnsKI?y<#C|g(?2nYK
z6_l{2jJn}#ob8OV4*J=&;6nz)G_*(wFNLSmke<ZTIAzG)6VH|+Kgq+H8dq5`e^2to
zCzRbj=l4-gUX#D-!jT9F<xi5k8JXz<F{t2t&s8M9wu8Wu#|3`2@yn{t&{^f2TTnd~
z{?-n_DOX^{6`+Vw^<RH5GKb6^k1inc6?K$&q%1>3ctqF5neGwjfEClaHX@zp@4*&p
zk2+4;uz6k4l^|hkUY{(YP>1JrD@IblVPKC5?<~ck!>`!R5(7fP6T8((8;OT#KOs@t
zM3IdGXL|}Wd5dflrbSmG9C=X|vGrCyZP1u)YW2<hoXcLG`z?Q?4^LY{G3pUk<m`o$
zI;&`hnG&LyFXQa6r|7?{x_qOq){k!~!;;c+$gfmJvpOpy8Uwwp2osbs{oe1a2<Mf^
z-0ztpAzgDA&O_HwbT(IlJ6}Axo#G4=*3adrCtG?Ob0>q)6Ma|6QCnrA@i$x2%$>xI
zk`tr+JK8hfXwQH0iJGs4yjh(whcUpPwC8s94BVTo*9w0I`W;dZsB<2OCwTOvf&`n6
zf_^%&O41aG;oYh!3=pK9(`k8+^P^>757%n5S)H}X#kW447b;o`9}+6sEs0Bd|MYH7
z%7|$QQ&)oAuFfFaoD0NbZOjRvy$ag&R1)`)@|3ronwWp5<i*|UBEj~mpq!^Q&Qk_c
zoD=9|=x<R%jY><7ciH}^B^3Fk%<t-p$qFM2PRZt?P(q-k<WS<?_7*Gab5_GFg#eB_
zI$0f(kKaEa+ST+wzNrA@d5Hltj@x2fzGO(p!f;jBhRJ#-25)+n(!Lne*8JcxMy0n^
z2&`nRrzC&HN9E9$V^CFx?4=5pGEg2*TuPF~=$GvShr|+2`z?`+7HrAAVHME!L~jTs
z4#+X|JH$2u1uE9k!k%*OY^%q)(u&olk=B-Gial0pqk_4iLz84|XBEd&ol$(tJ-{_!
z6&YMuYi*nyGqTuKp%`YhRZ}Nz&ELc0+LE&rG#-Dcm{94dATPB-vN@d;wo+oA?%a?Q
z|Ka<_jn&D`e?;R(>U?Da-m%fpQM_a0rlWkKansHIpvDaauh^M8;@MM7p7U}HS%OgT
zJ+%(Q7{sc)$eCg@@)iOXB&~3Z9ggKVb3r1M$4^=MDI@}`V5q139@>-S#?I1%J7}+^
zxYmDmi*nR>7^9ih@VL<idx>wcNeq@A?E<SnlWUxg%Z-zT4ABO<$&<gne0yt@mNK=Y
z2(Z4r^-7)vvuDSLi}@pnk;$IPA3;!ru*P4+Csq{^t5iU`++nSSWQ_7Bx|Hk(sd!Ne
z{I&2zWK_VZLpNSW$aB~~#Jsu`Hp!MuX-9vWn&ahfE55qU2?fvytj?oAPcLJV@n-U0
znFLZgOu&@Wl|dNWs4I3qYk}HYsEjExHwm*%UIlAEZ|r)^8n9KGJZqHbj6%<Cwr0oN
zb4r-*l_o{h5V1#(8(mHf`$<b{P5iTR9`c}kaQ%*p0&#=i5F}bJwY`=!xAyp!4H<s}
z%IVNky*BUiRw9b&^DYFVKm`AY#7o?QlooKuP}5PoW8&3OK9P9!d|YLyF)Utwt3+OQ
zc0C8T-YR)k@+_!Uf6bAR`X_e=<|A`FtL|M;j8h_OYFO`GVon8ULzAOHUcrIHQnqb_
z_&8mTaQ*!}bD_tv>0MJ4nzWeQaM6EUbf;OKnC^_J&S7y&brt9Al%3k56lRw+<=iy9
z*tBcm80hx>jYM~2=l6kWzvrn&AJGaqkmS9S&CjlwmYCwAO0T?QX*}2DId&)cP&IJ7
zOvK81w_Pf2C8aHQ<a1_2*k(Gh1VDQ`muQs3=0Y!NyIj{zp|`!@SlDy9wJm=VRZYwi
zg-NQ|YNv7=o}{8lq9kGbd>;Q9<YoyaxH+k}W}QMV2?>V<=TNxTYh1?`Ua_dNu)xPz
zTH`uo%RO_K_uix6f&ekr-!Tey6l=PXo|jJF<#xU)!0ua6f(<GV-BZ4n;%Z#u%Oky$
z5wJb=%}DR1+F0_^Mf&x^mOg)^ur2=ya-KXj!gsT8sceH+B1=zkt8&z+NYjGbcdTrU
zB<UK@<&X0X#tMHIRkoLV#B^6RYbJesrUXVb3H(3Ilz_yIu>03CB|s~sn*XUXB><M1
zBmBoRCD2mG{>+I|h%FKL-{eWFZG`Ug<#qQDfGPe?3T19&b98cLVQrVIv>gcoH!_#8
z5CjtfG&q-$Jpm>T-5Se|9J%e@Ur|5MvLu(_#3mVk5-F)vRsGm0`>+G_bc-cXB*nK%
zPTwavy&Rsu9Nt~POkM=5yq>;%n-o5f_+Lo`Iy-r#gab8wd7b|DB&BxvdB%^C5&%CQ
ze+T@wrz5C<GZa2i_(ZAG(Hn&j97xbOkLKzNj@{{KlyN4l@M+cQnC7Z@`umr^9wZU?
z@ylU<B*;k+pxpHJ_3+z-6JMr0`yd6xV5k3087!2xCQec~rM$!M2>3VsjuYSg=li$6
zeEo+Kxa_}as+^3_R>d&AmZ5jTfR57-8V2D5;+vF+NVu5-q78_hm+|=PbTHZr1vWI`
zj|7au1fz9Z2}(flIISY@K(u{ASAo3Oj~>W>1how^H<gAoN?2%Z8sd6hm4@llk9jJt
z`qy;$HXS~ZJgw6{O$o9fjt>9U<f(*F9u=mODT5rIkqgqXOrn?8Za{l}D^R2KoB`wz
zrG`2KQ{Aod%#e<61|crSz##q};Tv)octuE6Pb+Lh7ClPJCgoHl-6G-k`|B&M`;M!B
zUb^ps=cqIYW8HLg9;F>r!j<jRhE+SI={%e46qJ+}ZO)=?nUHOYcUrTp0SRvTY3fKS
zyd4{JpY9fjLY~p^ZVQCTS`maoi%;hDdevbu4<-t&Om$c`2>?InxD2``i@&YQLODW+
zPYRw<Oy<%c_$!j|>`6HsBZ{)tr?qK+R&sPEjWn;ZRp1DT@`-mzSn&H?FywvkxBbxX
zmk;TgAbD+~He*xt)X+NBa287P`ouxxk^WG)%x)LzoJc&_;%F;HmbvdS=9y_PUm-n3
zj$~oylok2lwik5n(ywxsu^N);e9w%9RCPF0@Mw>b&vb^>2RflF1y84asZZX26R(#U
zv_Gv3$#Q7;>`7-SC0vAoxi0cQY0b}l33g7VU<Q@80ZY@Vx-!WLVj9^qJ<Sxby!jMX
zs5LZ$b<B|@D|fw&O<mWt;ED#+9jXQt$lV$++b!>sPxajeUCTAg)mDJ4mb*Mtl<r5A
zb<*!7VM$mw{DE6w%%^0Ny!Cm1&PT>?%3fJ}6+F#+MNUdk2SZ3PHU~nmIiFa8ety~H
zRU<_P;)2&3=VEjjCp|HGJmFc&v?UMRO(^FVI~SeYi$6fOuOp1R+^ah~@M$nVG=8Z~
z?5(4@W<U999zhp3&J-oujC<^;%xv;q=sU%=2W6C=DYcm<56Cw{c?-*bC6xbrnopx-
z``BgMoP=PisWql+E5(BKATSNB_Et*){G2iF;3*~-X`4Z3AwHw9On4epfn0AD0f){c
zg7bCLXL76-xOvTq_BUrt+(n3S+{rFDB-MV5zgw!VWfV@k#qOl)m?aOPj9qG0-@r3!
zm^mj%eC!+;#{<EE_K1ytw!BhSio5Vmkd;u7-L&;=u57B$Gp^h<jwnR)j4t322HPcF
zWbAY@3r#a<PgtPT^Y3KdOXEzE@`Slc3@t!gmV`<pcbA)Ku3(SgB!e(6F1h(N+8nX0
zMc;HymY#w#CgwGbtFps11Alqetw1NYkav|X>zSd{Dnnc1RUw>z8kW(85@<6^2Zz-q
z-R#n@vg7+Em87Wz{V-1#bC@dFO2V!(Pw$B*PMn)Xt!gBe$P14E8B;G&C0pW?=~Rof
zU0wGg{kn%j*ay<v=#pZjua{wjVePjCV}JoJxEztvgVGGqlZsksAE)2{Iej~P8qt#K
zp<?BlGySu&7|0BN@|k*KQKE)Q=aL$B-X5c7ux7Fnlekbd)KJ02I-j`JYS8&?aZ8M-
z>kjOQaiyFovNDNI{+G$0K|Z(g8Pi`(_w<Na%NbpDwl~6gHO_`Lzl(YykxTWFXu2)Z
zTk<pkgvHeKetDwQJfc^rVIyhR8RYYAnGd^axaDs*Bo{<~)C+7$N?TFO$!z7%c~oTR
zB1;TW6>fUS6&a?2!s~?$Yk}kM(nAgC`Ga3@L7rP++-UzYeKSa}cYLN^3nxR>=;z`5
zcF-D|q@;GV!U1EJAA4N=U<+h%{-^9ds|utl6mEp~nkdI!acVpL{IF91>4dd@WK=DT
z>kq3Vq{C5vdfY*8r(gFR5a0n>tSSerZ&6Q(!b4Z7?5~~*<5lZJ?xCJ@j5Wb2!Ts?7
zk^26A#JU+Ea6Q(6BMom(4SQ25CenM$ed7$oRIRxkk7U63=k*u^qd&o83_I-PF$Sz}
z@E8z`h)_eLg55m^IKVPK?x~)0jB}5{soZ|%Wt32Va?gM~VEhxC*PPLxVP5kc_A;+I
z*0-1!VDfHYGa#>(bGaQEkPz=+t^cQ^#M-b1D{w;v?x@P;9?yI3LuRFdu5TdwYdfm<
zOUJTZ>+ssXb}o>$Ajx%enE}*Y(nEho-qq*S9z%JkxwbZ^qb_Z3HW1lW?J7OB<0Y3N
z6RI<R%)s5snx3CdhP3v99%8q`)t#tkQARUPfZRYaW1d01p@N@r&MUgsxfz@Fx;E?6
zMQWpfPx2b~C#W!AF~PjMIX~5M6=hVynck+!oUsMQcEu|*AMvR>8%aEb)vk|rrLW)h
z5wcW~M%iCi%Ih6b?e$Z4N~+V5{1iHh(yX3;<}A%TBe~p}24!e+ypHN$zS_x`EthsF
zs)M3;jk4Ta_AI=f^`cPGU5yfCqhFZ&TB`CkYRi(Ru-2z#`uGaBa;_?UXVRg9vDcvj
zxvCj6`gEtC8{F3ZW!%;Y6NL2DJ3L~SE4{zFy4f{?EojMJzV(+Ly(~BN@;jbdr!wq+
zdWDHY0a)S;wEn6txV7nZsKxI63j&69)v)hix^hI(i$3&@*SlVT_Y3M=v$uS0dpmyK
zb!8otuv#y68@uR*W7A7!>6z-&vFmSO#ojthfp`Jd$;G<PkMtCvr%tY0$#qjuL3GLf
z4Ua>8`QfIk*eBm~l~MMZmqPEB70Psf>aAB8Phg+AJf`&W16kM=KTR_?U#6%Yq{P*{
z%OCCgD}|}DS7YBFQ=jy{-aT!9OZ7}EZWwZ3+h2JvT)W+_Z-=zh=dRiY91^ZZZ1cCa
z`tqWM!TDh_Rn7?)V9%=qrf`mVt21N_R@md@t(${^nTN;<Ae3M1G}KCMwigC}7UF@M
zo~f&1Zr?F14#Spi{2T8SOhkXT!1l}T6jZb#rk7y{ZXUc-km)5|x8UNuuEwZF-r2pH
zo1lf#Zp6B&!lm6X((tpJo9luA5b15`{{5^g3M<VO#n{+EYY~iJu<5(tLgQJNh1`4I
zt?wM6GLb9g&e06AX2Np{q~iu3jOHT3cEQ%&)e5*?B8Qn5;AqdWh*7uKss0ap<mR-O
zp|l+n0XLUH1~3x>GccF&OARQO)y^iEe|S9n{nOv>Boin_u<>X7nxCFO-TihqlWT$#
zTAAhZ_q+H4H3;q1;+3#r`Td{Em%9%$(z=wS8N$Rr^y~Mp%LMtz{0frzC9O4P;1@_@
zA}-bjq2sH}FGKq^#4|k6H_-REzjrSlHnC|-bi<5IJ!OoEqw&wn;oeGMxuBtse=Sf|
zpu}yX)2b_4jTf4jP-vy8hUA1K5A<<nrd43%K>3pxGeC$jE4oL_N)EfOqY;G<^I*Lq
zg8;{Vi?@^IwE(wTiBxoxEbj#B6?>uz8zRm)7N_Lzr%b1bfD_YZ9T5R~fvQUbq^GS2
z1LLZ$vMjC?2s38Pai#b>ao{Dcf4IZaAc(Cqw#YxT#n-Ap=9r$C`cRagOes^}Lq5LW
zH-lhAK)u=xtG{tqslSq5IB8eAIBlyjDIGdLm>lGN8ubblRsxirvF@rc@kW~!4W{up
zNszdUs<{DEWbnvik;k6$J9Fw~c8CHR#J0l-WrTrHiLk7gE-!AbQQ(V;e=_-$TUe_=
zMV~v7xnqS2oKRrq(KlG(iamhBG;enSk;F6H*jYeoL5_E&1THHj+V-53!7>##Q;}9X
zc20y$jl+*&&JK_*2RBk6ZHWZ>k=KeeJ;ycA#o20}#xFsjQe=x8xq-1Dywg>XHS(9=
z6w`@_s5C{#BMY7n%095Pe~O)Eaww$FYhX}9>ln{}TZpjtOQMbl*1?<Qe@g<B!5WMn
z$^tkWx1s4@WZB)j|9t)O*U$eTue*HxZyZWV0kIK+U%ux@L(krLT)y$MC5uMVI1qvI
z6kI3_80~04nOFgzmpfFByt?rn^jm}y+dMI@o19V+7YCLFI#8h@e`2iAvfj5ISg2}C
zq&-zK;#0)JbK@h;FSFuf{PfRdD9!rUa`$Dq`@mc^R=GID%ZxZ!`nSbZ5D1Qv#E!EC
zq@4&Zpe<Ypj&yJb+V8glDRc9dA_}S)698ib<<;t*3#8k30THf(p@7Qw4Bz3yv{!;u
z9VO2cXr3Y4b5TWsf4Q4E+<t%l&hx&~toPk_#&Zu!3nbTcPpU$eop7aGlk9IRH|ODH
z^b}+<)WBC(Di+-v*o1_Z{IGN+<-Ogv&3(9<A#`Lg#$3UW+2?nRAV=oCsf^e;q?A3-
z!el%pL`463WN%I$P@(UlA==L8M~?_)1Qw{}QJlNiI9hQfe^=(Z1PyxcUPlFnAP4mn
zo(>8gW_&5Wgz_|I`3Ep2M@f3^;_GnGrUWokQ2---tPtOkZu0LK#+8i$!IbU$I$koV
zrwVy&+cRlf;eofJ2DUyk<8+1ID1y8{ObLcQ9ViETsN@Ziwg{ZngC(9*y(G&xdRJYv
z7ttk;ljY<2e@T?#X$q~=tjG@ys^feSz6RZKDL<u<3ryr=g&uo|T8qBmlAW4i&KipX
z@d({iD=AYwC2#^`%L-VMl4Ex$ZF?zl^A)pgs$F%=?(EWz3SJaHV^a^^F4~Hm1f`TR
z^egZ4Vot{*jwy@Wa&*n5Y&p8-w6df7_c9LSeB{0*e<W8_PtYMHD$PS9ZNuYwb0y#N
zK8EuYbG6YW=bNA(>%nPL!zE>8)4F=+WMA{Von2TIPPrLdG3vY%ta}=a9wIclsd@xD
z6<5?|j!N4OZfo2)CXmSi;%a>Hqm7>A=5#wQzX6XmB_c6Z+u(7ka-DJpyj8|fe8fu*
z(_bY6e;u?)xzV^<<u$S^$~(5jB-fEfdp#StH<!*qxQ>;)(piTy`s`v0Rtdy<HW^)k
zduZ}CuDF;IYunT4p!{p-2fyid(tZN81L=x>q~l{NgjV0plS6m@j8QgtHkB0EgPvn<
zs2-CeU=^Ne$Hf$N^GK{%*2Y}MRP`zb!jWxTf05}VPx&)U8`~|vv1O~r{99Yr^RwsZ
zRJ!QS<hyE4t_GB!Xlab><<6VE;1N_%ov&inl@Q#s^&E?65rK5(S<C)*RwM;G#{;A2
zj2E5v4iAtOgNtY@{nUyg7&iMVcEv@c`z(rsQV+#`Tvb7SE7@BXMY4Sm?`^f({L`4y
zf9=^0ps=diKBcfub<`x8NWvY|PL*4SmG?@J=5-5X)(e~?Mb?Y5GZ^MoycI*u{k!zl
zReILd{WYa;7I(A}GN_rC<9>&|UKL*)2xY?xYv|(fPP%T>aX`HsvK}?f+*=DN6O%D=
zj%!^=z)=~YVgX>Bm108B{js7-FFEBte>(mwnV@&VO7q4tF~le5SCX#TP4Br@mL`>m
zRhrF?(c9&jDhRN8UhTDcrXdLjoUxU&I6N`kH}(QmteMF$_X5ELK`UL=4U~QHb%0#g
z4b+}yG+&M^jp|=XM-*Jw5#&b0gz+tUcGi+C)r@(s;X!$*buf><K5MwWYd~QUf5}5l
zEAF)o$XhtI;{0`O1H$^u^j#MJMcErq=@q`E>?e%oc3E?KaKf}FYeU|R9KmYcJ@0n0
zW75ov(6&E`<^i(NAl2vuIJ0%$B4MWlOxcdIzQFN#b)IUTN7S)%kx-Xo>{HI;BfR!y
zU8~8oWpdYUN7v#qx2KVZdY~g%e=O&5$F{j;@_gi5wYEZmJ;y3tgmxOwF2=TQqzYYU
z@+B!8JdCoR*;UbRm1&+C<qN;V$|eZ;8c5tz{!E5}6ct5q&iK*PfUE_E{r>qczvOI!
z)O>S_BER6qDS!IqR^{9^)uOC1X)$BT1WGfmob>wo7&lkDv^SQq^(nUse<OSAb(>Qj
zlfIwo^_w3yFG@D(oUudkAkIoHW`QS46;p+cN@ahXDyfCmt)>0gdgOSlcNaEJ#F=iK
z9fGe7(KoY+YYpO1Yb6(LVm62$U}F%P@-!8TmcCE!_xDh^n%hXZV0VUR@->_KVZBhE
zk^%)6^y2b)d}C*?_xQKCfBU2j*B1`nMhjf5W~lQ_i(Pv*`t#S1iiz#8#n>OGmojxD
z^|kcw71I*k?E=ytKsS0}5U=L*PA#Rmp+AD=*ar6YJk6DhFdNXm3ywVx{<S^}T%3Ka
z%#*7r%`V8*0UpuSnKPv0ylF+;jV99Gt(fyYA0-FZq<yBmRjXqle@g96r8(r+U;}fV
zyYxkpq#KsUp#rW;m3*pEasB*!5tz;vO6!^GpH$-1l&}HLi=_fzE-S9e`xhHOx%hM%
zTLzUivb%b99H2vsWGseWTL|@+3!&)>2YG|w$i_C51)LTETz|d#bqAP>tgkB?BTOgg
z^XdU$iFll6{glJke_8Ss`6;E;Oh^$tD#wCH<#)P{`OVxp<MABZR&k^2N98qo73H~U
zQ=pH{E4I#Cy%|zzH0@SqX&x2ZX9_oaQ=2pldobnBisldB_v?0f>MFWDm9kL_+ub`r
zEljJ;mE%kF)Y^pEhevLx$!z)_T$&MU1$~m4p-~+iUr0`Ke|uB(>?~{;B#|$zFdb*O
zz$@_WoqFG=RPOWqvsn+jnLJXpNEq)qCjB`bAYN`HF5cHhNf{IiOo@Ca=LTpaeMj&-
zVs86pXfH}+k9V4yJ!aKDE$q&%ALzu?{Nko6Xb`rg-sd~Py@Q6<>FxWly`w0(Wqw=!
zmr3gGWY=!{f1^u*t~4o+y+)I!u)~_vI?KS~IQ9MLp);h*Wgx%qG3@qo42pm=E2}Tw
z37(V=Tc@W<Tmsr%$)d0pr%8ArOgj+XyLl20U5wooeGufO=gIc`iopu){H$BDyFF*R
zP+Tt|WfRLPoO&aZe$`9$&|OI~edaXiZg(9&uAM5vf7=~R08PU1!`;i4jvTk{Eu<_{
zt+F;#B24Ra%yn1_!h!hZ4ISdg>()E}gGDRhRudHVY0g*{Dc0co?wQ|8kv48%Y-I1F
zwYI#rc3GkCXeYV8^meqq9TI)XP(dj%#;+2=3lXlxu4>yu*59s>{T~DCA~XtRZe(+G
za%Ev{m#efL2>~~kvAsGIm-N~u!U2A_x7#M#3IsUKt{0b~v>g)yH#V0cnimw8zUC&t
zf2Q&8uYcX~3}6q9v5MiI1sa?sxxgodaB18_^L&M8D7|h3i+7EXE0sqeSn3e_a8%ek
z%u&9~scYkKjhCA6LRi%U&5FI)stvuRecOH=X3)O2w3-HjDUGVD#IZ2j0rE;N!J>T{
zr49=UKwjNusKcl!Q8%qZ_=dTV*H_P@f7FRmf&TXg2WwoRxL{ylaSrm-{uKlMSWif0
zi~<!nLSrIaffr1~39wyodee{e#GgxB;+F-V7{Ry`1mAoe2-{B?&kgREl6i0Yby_^M
z2pn(o3F`-Har^UW5&97CV-y!)nAfg{<hB&wplp5;Fr%J2Nj+(R3A3c0%CcdPe_|}H
zV_NF(iSfn!f*UcZ{DSXboja9cfTQy$=@K?kXGH!fI^(!B;!!XK(61M%Ve=)n^jFe1
zi5eAGRrE@M=@{NZ3TA~^Kv1+vpejj6fMQJYVP~>a<Ocf%&^*%Oo+xAIlp$GL!!GlJ
zdMKj~ng1rzp~)h^Q?fVAn_zOnf9<7vdw$*hyf{LIEbgQnK7HMV4kdv}Ja;r@47q>!
z<)6dn-G_eJY}7d8I4AwneE8*U=vqEz%j`MQoExL{?9_qdL8!q9Ksix`=1+)~s}?0-
zD_~|#GaoHiL_RV>;^r$ZB`DKM1vu$n1b^1@v9a66puG_N%E)i4RvxX2f3eVymTQvy
znEa4{sdz}w=u=z@pTM1z@%@eftMCb-k=qxP557&X<&ih1M7MZ<s`eV5N!Biq&P%b8
zLNaaj32Ik9i%*sPsO96!ChtiZ^xKm6*0shdzLQ!qsZGCH*0l(b9O}Oa=d7jWGm;6s
z2MRR)VYXpeI-)Ns=`7?1e}@-JT@Hcvo*6I&0-sk_nKDy4@px~2zo}RznLWJB8_7ov
zTvhDUL!B3NA?dJdpmvTK^fHK}^LZA&R2?iP?n?feyncUcnel#G;tO@Uo&-faVU;Xt
zh)9A&hBabah&qCnFxNwg+RUmp6pDl~*(4RCbB;<V2xT`Wg$E0ye+tWH#HOS)B7(9a
z7m*G3rxK1`jjTpx=Stuwv+ET_+%K?z8d!B0Y$)jqY~d+GsL@$>M&8g6_ABb<q@OeH
zkeG1O)qKAa(?T%MXXY=4&r@P>fNOYfjK>x9%k$ddq~*`CXw{cq15gr)rj}i@4$PK9
zG*%;C9hvp>xViiaf5jt0K0k{pQQ%XGA;emS$0ZHFUOWJpS9%^977e`+>f&jCM}g+(
zhdQ$X@r<Aa+HbE@(RE;4luW|hwry$sWDh9oF%twwJ>QNiw^(v4Nu<lU^8IA=7;wlz
zS&!<>uQ_B(B#JBQOIUV{Xz!A>>lwp125^Ee&N8!N?g=gpf6}T9v|P5nJIIy`i*cmb
z6)3_M^V|nz70PV#_F|N57SYFg5P2$Z8zRz1KZzqb00FHe1)U%O$n3e&-^ML2VN!bt
zx74y7$J407etBfd-IgAeji*4zRD?JOFDHq07Ofv&w)1BeBNhogJz7SO26dA@xu5a}
zPbr5`$49Aef2W+``^|lk^@BvKz=SkC;M&zdBBwZRMJlBv1Md1A&kI=3!^z`eD7Cgj
zlNh`OE}A=rI&ncgy}^{%;iZjdUW$IVjr5;Fo$c4Mf44#&gl0<mZX*gTXXOM@?Ly(#
z(=<te%UPrm=*&9wIyzV=?PvL((9cP|)bddM+GN_Pf9F^-GmlWM0>W0~Te-<9DDN|e
zqfuLpZA__2-b59RTbEaF(~@be6{{IaVz|aWx^qhHYieK)QIq<;gdM7s0fPmf3D>v-
zQ)UOUTT=@{L@!kE_&WQdXdwb~wjwP)N}xI$sXo=PR;tRa=Ki#0!oy;9n~%=YaWol;
zL&@$oe~wJf6Z6ti86o!2fslVrnV81%-XF9Z9Mj&07;tG@QDlSD(h4v%P9!C)m0QG)
zqcHn6emq`;qV0@e(4_MwDh|4RJ;D@uQ!D9gDig-tTJBGC8<_y!xB#XaKOTw9DPSmb
zYuVFL{O{@>5h_!YNFlwIN0P*)w}<9fpWD{ff8DAxi;Ptmpza=zKaD+RgO?0RvsN6y
zyc3OYH3aF*L&Oi1NGksh<u0&=9U<w4h^^f>e2|MhCI4d<I8H^uF1SH+2G(yU>FLbo
zCwx2OqGe3g^>D08LA4_5`IGZZ2<kCs7dyVUbOXl5<j!|2-AG}XHn%L@^ebMGvAQB=
zf0JOC;re&TT(wIFU;Ma_zU82g^!dYDg6n+=va!THItBN+zF&Pa<UZtY$u%>|93$2H
zvK1rj&rMjI8`sAr=Lw5Nh<|@j=m#@zZ73tOEDy-FFvIuHfB7NVJq!g_r(l$z?R?6g
zez=r5H$^5htMG~$Xa7JMJmDm>!vB|LfAB(tZU3LZyE1_%!UqL4-}OI`z(e(mcX6u>
z-fcAgCF(v^A1)|~+=2}x`Z``7Ct8y_JX)wNGIrt<gKOXwj``-3kuwve$3NiuJ>Uk5
zhDV=GXYyj7I^Tb)A1Hl$n9+gd_0xiNnAc*WQ|awLbaMRq`Y|}oxOX@7#=8~Ye-|5V
zdFoLuT~S^RUs-`#`9x;*?BT+*7vH-uf-zqo@i~X4RfT}>=K;sP3~?~dm)_~&w;?n`
z2z`w)b_ic`qR0he?(Tnj`~2glf1nFHy!{$xV7!RYGT_;MO|M$xG%w`vCBHj{x&<9t
zU>-@R27^JX031LQOT(vOr!X1-e~Hm|ZT%8jJyTAF?<&jW7^PYqST<;y2!^m?X{G(X
zw2cH+-6G;tzzA<)l{`1R;(VGFUc=SC(@>oCuVMFj*nKGU^b-H@ZY(!z<OHU`hT2Re
zDK*KUo;^DMnH=6CH*B=Fh?YiyzP(XMI5bd}13s1A3bNGFAWTiT7Og-{e{}ELHwe(-
znn;o}C`^Y;r(xCvri4t@MA`cM>1&?$hLc{Jds^lW`GP5J@Im5=HZpijJ|J_fRt3^<
z8cmVsL;hBmQgyH>F44|wBGxU6Abc1a5EH6*1=kN}lLzBQt<@Rysp{bJRCz-~9@0Oq
z&~U4=lF`KwRf;n{5IQTte+4Mh+QDMg+n2>qaiAh_{Z8L9d8Z^ZT(08Q<mM5HGt$;V
z`+Zk4c8wzjZOwMPQ(SJ8TJkz2p<I%?xku${gEF0)+9a!D<-DvDlDzq?G2m@@TCxdL
zDK}lYhxVqldHdz4U8aMH6Eok^Nz!kay07IchFxM<RTs6UjaxYLe|Y*`(;+AG%kNxj
zl)C)&LaOQKu@gZS$58UUe%Vj11eN>1Eq)&`PoOu+^YV>9xeKwJL0_x-CY=v$(WLvB
z(FA@8Gs~~DsH8E0e2n~1sz|;$Y10=JR^62~tyz`Mn%1l;FsaRd$Nxp&<@<jCQ7AZ!
z3T19&b98cLVQmU!m)^@81_3mea4`WDmxl2sDu2n29J}p4Us1UwKt|TS0E0$m+8zVB
z+3@Wlhro!PgB|3(9P<02BvPVORlSUR93%$XJ*AQ;lHxA5T)r-Hd4Kox>F&+5u}cuq
zsi2onUls)~B>b-=40^W&rGz(X`SgAH<v~jA;qwF^E2S~;3GjD>Z+EyiD#8o}Zxp;y
z>VI$_tb!4EkVeBioU5PV*&pt$vfkzuFs@*I0Hr`$zn-hi;V+;5d?yjXH_t9ENm#9=
z-@?uZD*y+=uJ4V1TYw6?Kh(|bQA+3F8<4O-hzo$U%}B$bbbzAGvtmKv9AmuGqF`0G
zHUQIk+ynNskB570G`z7E;YGCx&qQIAn&D+eTU&qev`-vvWu`M3e;k+i{kGEP+(`<w
zaw5Rcv?Hw21z)7ZWiCC}U7T;!6g={id8inUPlbbo7kz8QJ->f$kUYN0lpG00Ifc}o
z0#YAoRXwRHEcLVom~^_sDHKQ^DbbZug_PRB*H%38lYUz{P=wbs3J2;LcBf9*MlGCP
z@jHJD=_q%V-&W(P^V7djCao0Op|UFTq#4vuRnqE=xd{j5(?eyYQGt#!Av(v(n5xl@
z%wrV<9MyF`3Y_u}glZQA`P`uUQNYS>RoUA91bJpX?zdu9d$U%(VncuxG3rGM3!iA5
zS$+envL5~Te)s6W#DS`Wk+Dp`2V+u?ZUcYM(V}f1p@~5_spp;4TcjoV3?P5KpPt@*
zy8HERmIUX#aPa;4`(1uvjS>c^4oU<t=D+=I`EvJRhKd+UAz?}V&&kTOVdNt^jwFE)
z>AiD<EGl5Dvx`>&97mIO`?gz@B3>W>T1~7ehp>hSEb+$85ce1@Y{rxRE9eg+fPR02
z?7dK4&ycko@M^Qc+4!JVlwBnZFz{4eXkMcq>)+rGuf-jN1O#P9MyXBIwR3QCt3x^O
zNGiJWm_LXjXiWdFpg4?h_L3+_4~%PGT7#y0xn*R%q<G<k2E7CWm%!zz*f2=U3ZdmF
zuH)DRv^a|+;Z(Tf5HE+bGKVD5;vRpQ(|hWg`y*IKD}qbyv=U<Mx(W@r__8ZvJVjD3
z+CLc4c(!pW#GC0&#X)1xrH?r}#K_MV*Ym@8c`c%CHqu}qQG%83kqA2sJc?|O*zG7`
z!)(hsg0-D(tqlPp)NE>hU0~k(rA&tPCPlaWd&$EpI~(tPfNx`!N}1BTH~)Y9`sGib
z|EdHm_*Y&5mO5Mi6qoPy5%e~BC)4r`pL4KitXY&ufN<6=p&@`yZDAYuyxdtGgfigo
z`slYj3K4a*IU;F`i+>0(v2DPH2!{-@jnaPK_#k1`w<t6<H1bo1f@||5tfy7^F@N|o
zO@&o|U+%svcOM!*lYxKm?E`-VRU7P7!m8};7)xmgVs05gxR4tufWT}Iax&YX7C!pS
zBOOJ6EF&KW?$t=85k@Wz0W5+*L-ak5?~p%HS{YJ>5)6HUbsjXOxPOIAt&Ht|fBuf^
zzQU^a?7PCbSK0{c{BrL!AQ1|z;`SjU+X_K(I?X2Cd^{NU(Ct>y#xZ|Rqn)=jYyt`>
zd{`KYmDc-C?1zgPl+#)nFJRE@N;_k0OajbWW6Vx$PBci6*l120Eex=O<zctQOf@)+
zHcHT5o-!LtAl}|n=2je*p?qh6K<xKpCPW6ZzzjWGLu|N_KVoj-Ss>r?yfvowT1-!6
zQ(#m@lxpVZ3qta}6SIFgSkkK#%h{7l_MsL(7@e2^g@?{h%rc4gM;2N%j@DciRbZ4A
zk}QgJQ4%PDD$pM*PkV~xcn(>Dn3^S19*nc-oFY%}Y8p5sQ|<kk`URxIB`Th9e#QEQ
z0o~0KFzrY?P^Qll(J3t6FfmM-w@geo7uCf4V;QGmC~)6nfeU|XYu05i+2UfuiXBb{
zR?aAli&g2e0@b%vD#54M%Ys$1B0|VA;rj|%aLH42SnY(B2c&c|jMHf3W-Y(<&Hv1!
z)hIP?$6Cd3PpA~GRwWR^GeS9z#=5S}S&~g%@z)S32F-#`5xaaoT{wai&(FL#rC9FA
z6w5&zr`2jlr0IWT5tAg^28h{p!<BgUw(?Comc4|EbH?ij4T4IL)dFJE?vSxkZHe?K
zpfQf5S7%IH0=RL8d!;+k-U(2(2}rG+QkgAGwNA9fX7VVBu`Uj4&oi(H;!Dd(R#pyA
z<s=ApV{$*GoMa&cOzM{9q#2@Pdi9Rk-B7iAvy}7lnuvc@MkE(EY9i~bbSKZQ8Th*P
zK%>ggUzP`(80uA5*=8_toL5ubql6rO<xIDa{haSJ;3z{foPH}+3mbcY)9N!=Ev~c3
z_w_Z^6S2ZRUhvp*{WD$g&^q)h2y^l))9B<^D%=SI`V)k{W$oC-1kPIdl|xb(Zu($`
zxL~5N`<s7OSI^;~?FK{PnIGG3zQ1W#A|(089{e+!9=o5ZKBvL5TO4Wi7CMvE^~D{>
zaF=8uTS(Kx-QC1sHhXksGu=~k1G-ZgqQG{<A5H1m-RUBwqynktw=LjT<oCnN{XU>q
zEY!r8wPmQ6NZA&i@j=E#k!+kHlCIV2VdDT#)hd6dj%ymmbg^;VqFu(on`Dcm!M6LO
z%O&yUZcF(bw%gflsicJt>8JQDV~x<-+|X}L5N~)5GC@9b4Kh>qU%dvIkU4V=GR?m3
z8U&!fLp0Az$XW5wAd=GEQanb8HJ4{c;aM$$ReR2cI6qFSPAQj(ub?&8b^*M~^-zhs
z4kLdb&+}v2RU32#yB8@tLJtO1ub}O6tzV90v3_^GSi>n=pXU9Po%nMNL??n?ox<vx
z>Kcf-9v)W^-L~!-D>URkvlPgi>y*wl7@vN&oQ=C?IlIioZOdB(Xl6yrE6N+5<+$G@
zZ>zJMI4!T=*Thn4m?5c`OQqE!Shb?Wy&QjHbusf%yxVaxsodF4vtN2FJ_`7pA*Rg^
zwbAGX#}a($d%5AUL{KDi(_@KAthHS|uL75AfEua6dzuRwZlo8^t=L39lV&puR;|l4
zvQZw;5R&ySpxR-HA6>uuO~1Y(r^AdMvuAt?cr9QyE+in3hE$>iPEPY)fELw1jRAj@
zW;;^wWD~2c#G19_>;<v`N+eqvoJl;JMAQ}z7uii#yMsEkO#B^odORb=2tuXWa|)}Z
zlATBS?`_HsR0z0~IL%rzoM$P#CDaPY)7a7ZwooOu6YCk;=4n|gT(7NMl27|yu{d(M
zpnMufb}lW|9HN5#b>L4zE^u)N4T68~)gVp3|KlHO;%e>o5-nubFuvq>wB+<Ij0>0U
z!k5%tni07|nRi14Qf{pwyl5eM7AmHV?1BzYCE}fgnKG-$>vfl&)}Zxzn7earjDbbN
zYk;!b7eztInkH)FtyQm9fqd#csGSCCFlL-^tRmU)pvK&JzhO>dbV(Kz+I4@cGD`GH
zvp}|j8vEyl>Kv=Z`d*Rqo3%Gy%4ZziXe1|Iw7<v@Cj7*ncT+tEPwhq;%jmvTm-v#`
z(<KE_KuB@x4%PSi>XAI-wg+r-XIXMQCWU7#U>@)CP7bZrU?1CgY*y+wboYF};o!9g
z@6ILiI7q^d7hsqlSGNKe@dbZt=36?GQ|D-U3`w`E0r4pHJ^1MQp>&PvjY=^sNp()`
z8$Gliml$3Mty5YTz|<tE^hT$EZM_-I-_G^`zH`7}E~ySnE+;j1m_+4D_2VqOQ2mA_
zqx#)M?K~WA9GaQJcz-rTp}zHAzNo`8K<b~;>vULVLbL@cvBMJfQbB*G)`SQ-l?<MB
z0=9vn!Lp4G4VHazYmp@QeWYRV|I=S-6RfsjYMa~{{om}cbby$;O^2ldvW>DAc39Hv
zn|4?-6t>!*I<Z%ESh|Z^Q1?45qZZ293mYsIB;(TCexuCuq~Qkj-bj1FO>><^kwRay
zchhX7*|=%Ob@baQx!Hf(Iga*KbeF<o<0HFm+`zEiz>&K-j<F=p>i(%V3C4{!&W^bn
zZky{Zq1<Wyl<lX0=`b`VGK`02Qa1CrA-m%#Co7apb7y4veV`2+Ju`FgRA<p~XnW$C
z*k?kTGiS*c?|&^<!R6xHOh1#Wuv?F;Y9r6QH?iS>=@`zLdP;v=*Cz`#wdb9t9ZKiN
zsVngQcI!5Xng(zd?-C9zt*RHkYWlUM4G$}ddB?Ht-AKa^0#S+Z6cY=gBh^(l1p;jZ
zi$rR7*VWyRyrAz=m;s-;BQ(s^v=+T*cG~?i{p8th7j`Y+qSb^F?>96}C&wSSwvL1i
zE%v)wd`ZJ>x6Kec{j!d?<kx=LIRX3+s|JvlVHpe)12i`{mtjpU6t_|SCQAnj`;dVQ
z`u+uEV7`~pObQbNG&eby!QeU*m(K+!!hioc{{H&U15eD3oN0rM=x5HPIfC}fmGG<>
zf5kaRB7`qTspuYU`G-j0Qy%ebBTYtPOac9x=K3+u@MucVM;!%pXmk}(2~m+y@D$gh
zPB?_ePlfhlM^`6Qo7S1~&r{@Hxpux^tyIMgJF$^h_!I@koc1#FKc9!Ym)FBzbAL&K
zU_!g;+s}jl10^tN#e|Mokkj9PPTvn-M%1Jf&NwZJUy4tEzfHr)*JNdmla+C;mFlbv
zIGG+!S^+~LvM%vcBar7P1_7Nld^lNHL_|*Zsz$`DJWS%kQxX+=KJ!;#zZ?15*qzo4
zv>1^EU$#=RPHkYEEoXet04bW&E`Q-A@$|k(-<79uG!yrfXOF0!u9qL#6X1-uL4HKd
z&{DEH@EZ<@-AIe54AV-^5M??vSdU6@11d?tE^Evw=5U7QwWF4_Mh9ARNXMQ(CwUM%
z^(%1SjkMfnU4dl^)hO8_kELVxp4V1Z&=D?P+wa`#UzxcBTy$vOx#;3@Ab&g6d>3S3
zTW1wqQJE8+XSsW<E-KPX`ZxufZxDKyT->*=JXYt9Ch5esyhgWJ=D6arGiI0_qLY{M
zOl}$(qAjQOv5^uKRNcxf+1V^)gqc|Ap0&pu9>UD$91z+N5!P0TCg~~NX-22;V{7O^
zmqOEX(fs+7(S*YsmU2i8a(}xq#0VqE$mJMfgmKKNyM!2pz^s%zy2Jr-#}G3hzY}7H
z*>?&t3P^RJu8`*?uKit^_f&6n^~N$MW@wkWdKAgAsOktexPQfo7p1HFU!pL`al7`=
z(x1}ZS$hN3joKT~uF&2oIk(3W0Ixl%0yT5wrp6t?LPB!|s&dRIH-DmnT$F>LU`7AU
z<Z!z!qS1`onJKP_Yfp1j7#!c7&9SGt)f{`Y%gr&wSXpED7IW-*fYUzaIH2Nra~$Nb
z!5mRC!R2U=Yx3ID9tALA?;YF*s$1=`M>}JWS6dqhpP~%}K;3R=goN775KpVQkC4cO
z6yu8GvQjEusw1k3@_&ZBOjuH(gbVn)3e`zw+|1ZOFYnk{Mp?m3ts9OP{Ys&4k=>3*
zO<GHGv3EXd1J#X3Z9uy&j|}pDDi74Ry8dxTd>7Y0pn^H;`bRn3Zky5=hH5D(UJ}=y
z=BO00v}Yx9Kn#&f^~Y@0U^WO4kttIjAzf3e$hCoN=%=Qf{C_-1^ajdOG7Y7ECUuga
ziqg#0316j2hauLEZl|jla57aPSWv;cHk0Vm@t`X)H|v|I`#y=&7v#>7x;0R!%k9)T
zXQuW-E_CH{PSj>jgeAve>;*Ynre3vx+o_X7w%~4*L*+zYP=kGWomIK4FLH%G%3l-B
z`#goov|c)j*nj_<{4lMdxjF<ai9pQK&JG{`^Y;B8-~OQRHNE}Uuiz6UI0~Gde#Tp0
z8?vB;oqps2M|rlSOdFV^V}=xtXuUQxC9ka`-=+f!Lm^VxcKC-Mb_}iWQ8epAea`r)
zg-Sb?S@Ec4I1~NR4rbmem9V$Miof$HX|2Dd^?8-Q4S#!W?(<Yy^{?sheL8$8a&_c;
zWXJ}%UJ@&1nP7Y#S4EqL07EEOSt8Dmm8^`m8!dz6e5^)#{1KKJvSpu(Bx+UeX{6b3
z9wMxaP*_U8<M@gIhP_6Sq&^8#F1AaERd1e<<QipLufP3F3*Ttj=jJy8J%ErDmi6=?
zD7Xo7_J4#GYOl&Jwj<K?JiC<DQ$IFkTcN;Qi|4U0A1N;#wgHq__%byt#?3x7yZv&p
zh)9_NE)dX|09C;_!O#@3taNGe4uH~z`g}Wf8&Kkf-3F90bsI1U_PrW^L=G|OId@(P
zy^5k+D&y*f$QHW%z%xI&^ODf!5g^T9?l$;j8Gp6NW&$yCC(fH*QNQEBmF;#Ja)9-f
zbJ}HxsSB1?LYimFO6+?`S*u*pgIbM%kg$2v{R%G!93P~Ok~}`Z$3_=jk1WsfIbzq(
zvgwru&^gj!$|7agSMofj9Cc=i{tTf}BhM)kui394uP6l0b}C9OW|ba2RtRI}^n6FI
zWPd`dbd8s=8&~_rMzEHWO_n}lkWtVoW(}0lWEbhKTZNbO$>KU*dy1j&Xp3WOY(3(%
zltV9XfwKBobys?KCY_R}{wyRub2gtWXY6bys{|(*ch%R(gl(JGNM`-Zg`Vsv%Q!09
z{^-K0EW_mCR(DkTm~}^8XIHl!(54?6{eMZMPQFr9YjoZw{JTgPMuj^TuMoS!;|n}1
zr>NOm)8}6E&Gy<3@Yz#?8knK;Hs~<DDX?~0kg_yXFs{0~gPRP<+0Yr!4P1KQ5FC8Q
ztKIqjZ3u)bkH~^pq{n<P-VN0n5MJLomB^gJ5otpOgfcjWKDq7@XAdb?ysnaWOMgyK
z!R9WQMtbC%p<PmYy1vFl&o{!j8Rr|jU4k+y6eH?ddqGg%ONKS>H<w6+iZz~E`sqfx
z+7>cSL1;wM>GALJ^pqb2rI+GM(#`XB%~Rm4t=_GQ@4$8}@6)pd^Wy37hY)>}zA4EL
zkAyw&nIb<uN+|`yg6?&z8?@ew?SBqRmxj^F+e_;`+OIwC1Ltr?Q|T!8EBwB;sA|~j
z9dt#%8pU3^>e1fU)eV8K=&Gl6_EaTZ^((d>t6{HMSKPUxtL6zR+2+pLBL}0c+)H~s
z+OIun1Lr#h)*f}1x?F>pJn?`iZby_pADjU$zw$2-m;{`uWP`a5NOy2897#hGzmswH
zXulx(_XM`sPPXv@QEX2n2E;8dAf)7SKs@&XLbTt0!B!_aXmz_2U3{T-y{1~KmrE7Z
zQXGx-)PINX5%m9;d4AHDflVhAm%*wq6ahJxQT`?;mw&534wonoCxrvxci@+=4<|l<
z7Rheu&JihRd)*~n*_TA2tV<aR!E{<#ZUG6qpS;p}@3*w}0E?WARW3To$>6Ou2+dwt
zhl|QOFR3$;LM>^SPR3;YogPg{R?ctUla5-ieX75b=CC$XE#Qk4>fA~(c}>!Dmfw%;
z_B<JB>T0T0a$g(WEbn>=s)fsZ^IGJ8VNuwR`!P4|Ui{iZqJhsR3QxT@FKLkns{k`^
z(9bt!k%}JWtg>FIg)DdW>50BgwN+@nX)XQQ!dM@9{COHJ^wQ6J>FM{jfwwIG%;zmC
z7hLZv!kX5nJmH@zmu=C|^JHGuSjm&~jrUJiy}!y}@5>S6SZTbC0Yw^S>vflZe6i;l
zoA<YjnQT;hs;>I6zREF@t;H7m;iCQxDbH$Zx;UQzO?7{oEIg|Rnh+O5Z<v$Sc`C-E
z`ukjyXz6W9G*grzRu_9uh`m;q;LMhMfFEDQueXnfe`jka9Ti0A{`&EdURVT$>^z`3
zN$Sx3{JZ=9@Vg{i?W&dLIpu?Ybn(4O&$qf=M$O@>9mnM)RRX&wu`6m+fT0i9s)M|`
zx>^Gw`R3F<r)V{Ayd)_$XKAj%$)M=UCOO+j7Jt$+4jbnbI3p<=z3Aw`+p6;*N8?Nm
z?%;6N=RP&uqtaXYqyWm873yTAd8Q87LE@8{PI|uGR{50a%FT4pTTsz|NMcDvw4^%D
z=}xEMBOZZotBsoSk#$k911prXh9D;rBQ4NEa}&YHJ6ezgX*s>3RR)dJ-g;*@)+mOd
z2kON-6*+Yrrx#*Ebf;Pv4UnFNg-&r-Rl~V|_{fYWJ>xvkV$Ab4>qe}#jxmc+T@v-M
zAG8QaZc1-yv%Q|r6W}j@P(y1@?Q=z~k(U%`c5#Xdb-%@xfGLL#{a&6{pB}+zTzXCU
z^==_-P-yK;2)Zr=T@m7dY-t<=B6r~KBPec6PGEUswln!e6!xPD6~-V(>n%GSsHpvn
z?{aXpeNk#rZ;vc?mUMu+<f%(uumnzS9W<?<M1*M&w&1sd&0czcl;#FdS+HAJMugPR
zU&d`v-|HIBzHC|}YzPhBx7h`A4zgV3)}Sud=Y-k-p~L9tkbB)2qe?0Fph@Js?86*Q
zvQ3F?`Wa}_lH<dLKQNvJ#RJi#pnL;OiqW5hCM}~1?>`w$PMPjRlT%`MpviQsvqP;V
z%}6YJxUe4hOhtfy_lveRyia?IVnR!Tc7p`oF(EHCs<%V0K#H4O(hhJKqMG^H9&8v<
z`8XqZp(GK5Vw@$tT(8zMYNv?&{gy`H{l|dhyxziSP?47gv9MWzXgHapT*>nOk7$$*
zQrSCzwKb{o2ZPzT=^;w<2BnAXj(iWe<Fty@S4;mNz-_^Q@OsP=j5z>FTMa5iTtd<Z
zP-k_ih;y9oehC0a7ZuY!aKjc9cLH!h?Gg~K09>wSI}k?V)I|O+8*c%?oP5SUgfoci
z?|`@s#L`)oFbDgMA!sgs;shkOnzk*`TzhZaeN^d~D>NEmRALo38e8Eka8maY%<}0}
z8T_2msEJX38GmGHR8Tw+%nHgkz^oYkNnqA^uh4xmn4L1+31+9n?k$aOgO0WMDCFM(
z9Wf}&2bd{Akx;i3Dh0vU-Gxd)@B(%$MMz<hta+2YYrV!@P}~mGct!tWJ$t}9dp#bh
z@p_y%&L_%o;y3~8apFFgsK<#VpU5D!r818thvbic%zx4|M&l`UfsrKJt#m6ZU5Dp0
zc~(ooxr`Vr%%x~~D<{%$S(3C%fYeMX>TNk}aWV`+N@HMGtY8G?>@4<ic{!2P%*@+Q
zJ%+Wzn0|KZX*gzl;&D@KQ~o>B9L#tzGL-)fQLSO5KGSL&;|uAPr=(npbtibLUtwrY
z_j}KOb7h(#UY{2ymDW;aEpMh()_upYbcSWzn=XH>9quJ>DzwR7>dj5E{MV}2S=euq
z$Xeb6af3uOk;qnw4wMX0k+B;jb7RR8+L))O@rdQLvGeF!1IZXgjNIEAMm6>#tM#KJ
z!o<eupfXsy+~$pM^~}>zp)=Oo?s7w2clyhJp&n|(Z^bJ{yk4aRq<yUJSGm80!2ZF>
zG+evtOV6<k*RKBD;_$sf0z(BC;!>Ssjgama30W))&NGCAHI}^)TgO=!$Yo4&!@5$A
zJ1o2~Ww)@FSdn<k%J&HCRtmfSzzPOcL^`klASmuEjtOcvieqB*j@xO;*IvDyzQp%S
ze*D|Y7Lw9^FdPzEv^cZo_TnIb;j_70nP2x84gDnW@2(kMzo-tQ8sBhXjnh9@=9<ZQ
zbU{9<0*(OXx$+Oo5VAhsJ|6x9or0K`@gF-A12i}=m$4iw6ahJx@nSY7m;DwelYjdA
z_K)+g8=i#a%xVAVZ~b4#^XBhOFFG1zz|HXIw~c=RN-$+5y#}Va9lrlHeA)cglP%0R
zeG`9>58uBI-N@%;J;%v<Tx-OU^#muA!l1Qd%D;-%gAdM;v<1={$O_=CAVyk#I$1GG
z5jd>__7urX93@dAn?ax)sgdGC{eKDKk48xRLRo7jwCpKMK9Z}<He@{``q^qC2Mv&)
zaMDs7lT5cwdefd@7%20AJ;`c@$O|<H&nR#mjeNet=3}msltmZ0Ay&?SNfLX(a+Wzc
zq2{fGbJja;-!%8}<26)A&*$N_mC`Y1>_h|J+2{d^$}+H~rzqfgiXa&!ihl{kj|CJp
za}>@!MVwt1iGs~6OEA9-PWHIjwPkeOFngJ8G&5#m(r1?2l1r;qO=*$QHKC~IHL;G(
zL@D=qASr`ts6T=AXr#?cIp&pSKt7y}m}1^%9AANGTBu4sevqEgq-Z?EzkNLUC`BFT
zDnD1{=-q|@({Y+FO1Mv->VNb%rXcsve*EYQ+ST9>H4jrHyEc-2RjZEWYp2&FNOMY(
z9%FsXng|!`Vf||EEy+q(o`Jihg15xJ@V3RFZKY6Du*pKbt30w^R1vfqv_yu&mCA68
zc%ZR@Hy24OB!?Ik{t=r@7w4CmizO1kP+t)U?-mHx5r`{-Ao7Yt`hRYT^ePgelT3pe
zNrVsD2v@ZcQZCaLu1KWsmT1mKbExkY7)BVmcE|xf#TXBMO%6Sqp)lw?CyofUmoe1Q
z<iu3w+9nC`E=YBmb*R*GfVUiTCL4c+DY#2Y+|5w%)=5BUqQ7*8IRZ?b{t4J!vYO&Z
zC`<QWc}egvPS!r_=6}@lD9oqK%dCMO=38=T_Y#ngQZmP3MUdZ7ToL4Vly89iZuW~n
zKB>!`)h`D59oK~*zoT{s$ajlDey_oDkl#^U0p#}-mxKJC;^7*|?<XG%W@;9PCkV(-
zPXO=$fO|F<(6Up^xXEn(xCbaMnc`~d0gJpq3H0GI(MHW`qknwemXj4;2>uB_GcL_f
z0si4x@P7;5%jjdTf%jk>dBqss&mu8svf4^06=+{A#E7-K$=a>M8VwUT=6$%y`*1h#
zL_0VJQgS^9Jb^-yUe-jMd#s_W@kTe^`rWR&aYdTxoyV$R-)~j7fFG}+yDXZbEQYQc
zW|t9VM5+*R@qZ!3sLdlD8&d>Ik+CmiS^W%smwYjv0OTYy6uvEo3s&(sIdL?N?7T>t
zBQhaqC{mgZgQt-xTFY-{aClE8y~UB8WXx7NX;h*-WwV_90>=?;-QsN??T2X(5*qZu
zI&znH4TWp0^pmyrdoopW`)%u1S(`nQNbAZ+JC~_#u7B-FgtN?0IQA0FV5|^`<Q;HK
zQq`cm<lB<A6b4*R{n;IgAQ-Ku2N4nojvDonJ&Hog+ET2zN9ic8xJT(I-`JycvoG&a
zp2i@QF3Aif`AR6QxGv|!I<GER;;G`3Bv4d*c<Iha*W}Ecm2#FmRdj%dvb#_z*U2u2
zUO*KKk$*yxTGDz#-96rs&xvbBOKLfPDw@JxIif1o>E1*sWInRA5v?g1X@oJ8jxqIe
zDLF7Hb$h@XETdCJDETl-5LPnhd7kbcwVRYQ(oJt=qLmVn7RaQPsu~3i{r#gH>@L;N
z?{lOs7+&dGgV0SRA6185YLvxa@3Mt_uQn)q1ApDrS`+HvSi|KV3~4CCgL`oYqvN`;
zgV9mDql2+5%aY?WRdFQl*IyOi)i7$s8rV_1SOcH1hBOQX{D(D13eB+jcYyy#&5;~w
zH=^&YIZeHR#k_?hL{g7Ef6#g(tfI_ff8u&WV4j)g0P;t#H&1A&1vAyjI~7nD<?d_<
zB!9uv;Zjn#+1O5dq&fXpbNvemTPVjYteLPoiu2pWr#CJg^)8*&0!!ZuJ&|Q!dqt1%
zrf?R<_0(eP_wv7-a6%T&Ol5ewU<qlNt;e;w`54yAu5k3~#xadsfCoAcpxjKXht5y<
z7sb>MPTh<$Ke<PwoadHKMe6Wk?(qQ6HGi$`f5p+Z{{@qbiJ3mnrjupfu2!$o9}DwB
z7(nd|!LDYavHY!jcRHuj>qd6S-Q#(ALI<hq+rhDL3?1zvd)pg}wB~V2zu1;jdaDPj
zZG&HTzNMR#EZf6zXd3BW{MdZT3l<cgd^_JHPxs?_^AB(@tx*bPZe(+Ga%Ev{3YVNG
zCkmH=MFu689##(u0yi<2fpiEce`{|Yw-NpBU%`(_fpR$<&X7|C2Arfy12nA>r$Adr
z2o(7`Q6uRJ(pByL`<~(6)x%fU)?=+CA8c`V`8XtJ&YU4HR|_d{<4|%a+BCEhN=+7W
zPAE_goE2ZJhJv>!v>;8qV4*5`<rIvn7bs9_^$AKjP_F$Q2P&oaSpv!Ue@+5b>fNJ7
zD+B-)1z^F)7Ra*q3jjHZH^>|^Df?FiFft#ahRl+^6A8IEq)Paz<h=k-uHr4Ih1^>2
z1#?O0nGOGv0Fwe3CsVIxfr%SR%$Ra57=mkHZjLjfRC4d61s3h^&_c~;#=e5NxAorD
zL9zIHp?gDtxWB?URC@zFe_}!yq5v#cZ|YMieenx=7x8yk8SF`ahp&zLJ50*@ya8TP
zpLdZUxo<(yAncviQiGJn7(*zC`d9UODh|G27uCEMMZFPx%|VQPG%UKN-XCBniSLG3
z(A=?L_F6g?Y`=;S#;ACweRDB;ZOK<28>Z%yu37?i&Qak%?5mGke+ea@YE?z4`50IT
z%*xy1u}YtZ3QD#47$~{LpoA8!8x?k^nqp62jV-1Av7*G@)U7%9ci8d9$>Ft(QU?W_
zZ&+O~mC!I>4AEL>zKI1BXtg75=(zQ!mQZp`Fwxe``^D-vV+RZ)GHTv09K>?%B?-pv
zG|gS46nTPD2_}6(e~nS9y@Yq}9{8GrH6Pzs&{_w8nVC_4r|_?YjYSb&k_ZM`;GrmC
z{qYih1q07dKKpF=@+Ib0;oIL1|NPg#@xI1}rBu@BO^(kG4|ktFeNRnNi;b_=48s@8
z!{zCVlfC^#U~dqH=X<Nw;`BJ=@$t`ZS6{tY?X4Dp#uq!wf8!O#*nwGr8cH8!2S!Dd
zZG6B^Q!YF{V2KGf?GI0%4$n`Q`!5!&m&5ZfcZTm4Z&z!?-~I1ofsub&0MPMjalATn
z2Tr|<Zx?6F^V9vs*+?b+>YK%@gT2p}Z(n*1#pJ8yU9>+%OXsn%;j`o8<=9akhkx?2
zvz||*v);O3e`CQ}Z(X3Nkf#Dqg)$Yy3%eKK`(R=Ce0ln6aoTaE-Qh38*TWZhz=G^z
z9{a18)F?I(mH1znN^UtC3{sj!E4xpI7w12&yuZI59RD&ryXbv?b+9}hUJU>E_G|z5
z$<M3R$=RQV;ni~gEFLY-7Qe*(<<W46jXGNm`+Fw`f5ZNdumAXZd3v<Bia)Q84nM{G
zuPpX$@zxp%;Q|C1O9m}7MF!(ZO)*KE74YH7!|@`##Nz8&QUqF9e1f5h0Ikv8(f;TB
z=p?>c{CD_fZ}s!vPgXk}-*C^hREi_9XP`k4D_C1lgK&yD+1=4Cj^mqyUk*;(8{+cx
zr@^0wf4?36@ck#_LVo}0J&{gUnsTIyP+bIn5lje3OTIhOH*em=pUzL0KOXd5hY_3~
z!B_Wsb&E2bE?yt)9plfNA0BQ4Ea1CNud&DZL{Xj6Ek&j~+AP4Iz(|*I@9aD-UY!rG
zS0}^StCPJ6=j!e1CYaa8xEV8z4{OW?b_<M>e`5jG+&Fr`FA7^x-5s6<bpK-bYWdxA
z_%eI~J;e<bO}QjUftD*!JP<@7i446lfp90v9+bC$AjGqzO%g$@6@?Jthl60sF@tHi
z+ZuELc9G#NoKvI?2vB$ckrfIg1Bhh4gn%&UviC9!D$K%_3B406uh$xbB@5O<=wk%S
ze=8yJonUz-jNUa^X4(jrWcLY{nJ<E+Yx8xqTSrY-=Ig@ri7r018zSOe-5|papo|q1
zSwvmkN0*t(IdW>7+n>d^UQm~cK~Gnk!vagp&<PWWl04?M?I7Tcid+J~H6?6wrH6xn
zQAA7bgBCqe%CO4}6)%F2MrG3%MUer4f9}8w3UNtriie}{QHv8lV|35YAXA#l0feq3
z_DCp@S~_tfPpo}ovT#(O%0Pu}e<TzPq0MmD2?f`2Ew-Xap8X-UBrcv;2wy?rtdr9P
z>-5xVgO;vTm}p51wu!$VuE~j`Co;bmwa;aPz&PKBz_?x#Sf0o!x2Z6xeV)`lf78Qx
z3wc{iJz0jtLk0?qi=HBYks|10!!}xxlnlvtbt^>@kYxq~g<(o-7a-UO>X)~IAd*^0
zlO?Dl6~xgwh>M|Q`T-14G|PUX0_|viJOXrlgPVXjO{-Nc$=Cz}a|S8XG^!`+YZQXf
z)f4<jMS-0Kd3yv&UGWo$%g+Wde*_5ZPFCSw&U_m&bFJ2zHIQXy?8%2~QUl?u{%W#k
z*eUpeK)QY?{!s73C$f2ZQ0{elAeW~HevmhkMwyNX%5+3frXzwf9TAjCE0swrl{yF2
zAYYPxt{VM3EoY9x{irFDZ3=_4MNez~018Mm_*I475rQBH4neK^6=Pj<e?_L4Z}6)u
z2*8G11yRe8og2-PsKARcdWIK3@(1u3>^u9BGL)L!;N9Bb)j1`Cdm(L+StJ^RXR8LR
zdK7J49Qna$P7tyA2D(L_LDWjiRv=ngc@|ZYqQQ425ZDwf0c{1L0@^S#ByD~iinMKp
zg@8tdA;7)IB$leUB0>}_f1v(H3Vv=9$P~agPVn$!xPxC+P(wP|ZZJ%%4QikPftq4b
z+8l-nr}Qv0p-?i6Uhap&DZ1DPG-BCIU8!))^!rI&wQVr9>V25HUSVpPl%Gt>PbTH3
z6N&3Y;`QRq#O6A&xt1BS+Ui0KYA_>r^J_O)MfX*1$nol9*=7Mne^b6hVV>6NRuopT
zpoTOGyYFSqGo0>-CzSneas=V$$d=!;6s*{HL69qn{j3YBdY<IE-%63EZrN`}qxqd(
z)(GjhfdUxJAOyfy*b3~8v>Eopz))@lYY-QwZCgOB19?R483OqzApPn~pbS77B9NxA
z4k{4L^<$4No*!}Ge^Mhe^$ozP1{Pn7%~U0O1aF!+H9~?R2;g&A0W^$Xz!{{BuK>~~
z>;pMxH+ifS0n&vnU!Xvz;D)jl#rsym<AK)O<OsTXJID|aVD^s?G`5`(d?evtZ6lo2
zdY^F8=rWw>q~+^HK_4C%7%b?+D1#N<9a@XUENJn=zRk^Ve}u?r&1}?8g=cXC<Jhua
zJs@RfBoy#X>AzD@fT2eMfviRo6pkxNNN8}O<{Y4?iDWAoLGE8aQmgF)a_LHiiwktv
z>92K2HlNk6@B6P>eUtlGy<M^THp!tj$)Prb9AZiRZVMC^6T$G1s#I=)5eQ-VHwckb
z2pY&3Y5>eDe>2c-rX2mZ8Ge}Cgkkg^h^!yxmbQ7-g>S|DawK61q*k8gbRBt;VgV{+
z-WHq<G?maG-)}ggde1VrWjDPI@n5TTll{M2OGr8^I2#Dp+te%?kzd|o<m^WnX|5ln
zTywZ<t_W3+ynnRw9YUlxT6<r=SV`+I+JA{Q-lOgGe^5Lo5%}VaB<nW)+=16$5LwXM
zdQi{3Y<&^8HC4K+W80=<+Z`PHasy1abK`G1L2POBZqx2_x7o&D?~L4L6Q$V<ltvjY
z?15Vtlm>HC<om2aHN&;#ap(l0BNpUR>JqZ{-+M;?0iA+xwY2R~lp<Z@l9tjfxa4}R
ziA($nC!+pdE}dH1aITha<6JG-HRozc-!D9_R+tv6`0rGT{?^>^|M}%5mqC9m6PM?Y
zCk~g5Vi^dRRP{P3mr5Qd4S&Tq0YO;Xu><5bfp0+$i$P}gFtgan9`^62NLG<-Hs!V~
z?ZiM!-4t0b7QcG5<?>~b%g=|0m&1o=Z5A(r1+SKu&kNuW5<fxWptFky5KhDL@@@I;
zD5Y|EJ>X@e)EY03_W{4{=}tqy8DQvw)15bfBRZa(KGGVC-;d>9rGNWpTJKJGMk<Jj
z0@4Kh4Z`X7mtPK&gq0$?beq1$r>B?0_roZ<l~FirmQUXf=?9}h=%5xPFSLW@&p(#W
zhmRxK%1q*#@?quj=a*#|`IW2}C0b89XYI;*Akm~|aV`p-ek#^8KS^m(TZ38~w1P=i
zfI&;<knb8}9zn*Hi+_>0H+I7%7A_X>9{0=o!p*8Qw&M2TIlxqYA4F6e$`+}aKonJ>
zXl&6djjy=HGA7Yma7vl$Y2I^@LYSl!e`H-(Bb~0wM&WH7C4!&QwbH4sa6#ZTuFxA<
zaNv>TaUL0iUZC23HPYMOdUP(mI2(jU*WvS_-(r=yyb>8jKYvo3+YbkxPj^lNX_RS0
z9=+VM7M;tCe8CQWc>zUdgoKT4<e7c)N?4`F*)yH>)8*R5An-(Nw42MKr4m}|sVtA=
z^SB>IUMz=J2ufR%JiXh<T05b2*eGV^iXl|kxaH4|zYjWRp^guN@xo|dO=cTZWH7QD
z@Oc;1q1MKS3Mq9AbAF(V0=17!c{TFuE6~T#t?*btd_VN@@k|HU+;_5)HF5xOX_nfT
zd>|(te_9~Z3qspji}T*DDc&GxfO7(z8YwF|JtN=^!aa<$rvc+3*%U4R)SsWL>}dlg
zTESeDD(r`=J(~|)bNGsdTlG;xp8k9`baA>cP}>h(Be34urYUJ~?uI@(P`q72G-B~}
zdTR_eCfri*HASnNe<Z|8I_~PVkI5T7loKJufBqyEj_f~}q;$yFJc&m!zBY`!G(v2o
z0V^<svM`Q}y4mjP+$r>_*J_w*d>fW1j8f9gYB1w5=!MWiJXctFVE{>KAC*^mFBE-0
zw*APF1*dA0O0?Hcc%GSr-YX4_MZB~%;%k2bm+Cjv`A(Nju_px-rVS+m#+BfRY1Xj?
zf2uZ8o%HFX&9S*r9;1*}T&!|QPcc2^`G$BnTH{;j&D=~5CCtr|RQKgD%lwNYI+R*4
zW!rW+{~8zon3|%4mK;@UUbnIW2XF=kPK{jEW`A>79zJ|4O{-?6@;?K?q^BDsG#z9m
z{qOs|N(l)0^u`^G621%HZ_y)n4928Je}f-Aqs(Kw%hf2%MvIN>lPRD<Pub?D!UZ*F
zAL22t4KD3BC<h{9j{i6I8;rU%iP!`Ff#MC5vVrnWQZ~%~K}ng8mnIOmNXiDTS0-fx
zwKpVX4_5_k9fAn5r3w!>FFiDRP|FH_)TXSMUb`nNVi#=*bml6$3?s8Gu_<$De@)y_
z8nO9O<r_hdcqzjfRt>Jx6sHfp^fW**jn-_8L7ARR(ULm_`8kD{N0x*`Q4QOULqZA5
zM8Bk2-BX^k;x)4_R2@@8y3=A!VYQ8y*giFUFmckxi0E6LT;+n}I)84xjFvDJ4^SMY
z)hyuWz{RqM_A3YPL94K`?oPfqf9IVoZd*EcrdQ#JW|w$nKW;jOlEkK@2)gFwHP+Up
z^~l!N?0--~4DYz!Pf%Vzm5GXeqC_0ehE96O{><j|oURm>jIMie-~70E|9qETIa6xt
zv0?wz=AHL6BizQrl@`*r6oGxpuRKkctEQNCg$k22_N;EQp+WXSmCy3ke_BYuaL|G-
zqf?)x!;l`w<`v1)JG|~n1JY8r{$54AGsVDuJV+kL{gTeFR!EpSikZ4)4l9({mE&rQ
zs)zyBw~p`@fdaMDd)`~*nx25%FQwhL@^x>SBY%E%Qhjymb7p0=d1N8mWz=}?Wo~Cc
z9oPJ8$0+oz0+Rb`tNSQaf7iD@(~*LwRSfd4>%%$L;78o!=co~PPA>af?6I%)yocE2
zS7&GXI;fq5guQ9rHAq0&Qh9M$D~+c$lpFb`j?U|?6Q!+iEgv&n^wvm=9u1xMmg3bP
zC#AX@XXKMn%Gti)g^qd{vJN|x`=ziR{C&50-jv?{*s?m~r##-=e<a|(9~boX?zy#u
zY}@~h!69H4ze4v!TN!SZl)KX2AbHZp%Nnh+)qCr4ex|ISw0?-59@fSobIUTQH)dl*
zbJu1C;jr$bf7qC|cS7S_DZAF(%G4-MulZ(+hn9u24R0fVrDsbeKx-7?|6^AXG;f;j
zxOBn=m_;8tNj0vNf3wzpA-y5D$L>jfxo#uAw@C!X2@7M7$V3q^YyA-|A_xdQ?v^d0
zf#MBYL<8lW7SS;K2epX6Y2nQ+-!Tmoxq)(t{m`GlY-CgN;;zW}mVc!9&NfK2BAN}g
z|HB(3uuAw)bI|m5h^w}n1aeTxTz(Z-_TIr1^ob2$ISZRde?Nch{(r#N$T@if$B5iD
zz0b73?a*uB8lh$gHDZieMac_k*8PSjp>`joea|mRYdC(aV|%UqGVa>;B7k4-oO>;>
z6AvjPKr|jV&l#ar{{O)53-0*OOWt<$1RaB2{<oxIT!V96J{Q^mCf4Y8hY$bx^7$X1
z{)wk#`SRbie*&RgC<`$x-{?_`2s*;@m7X2uluDt;h$uCplnDrN7ANq+I`C;Z7)<6t
z+t7`EO`}nmV3g_>=K$i<L}t(pu?&e~8zp{U`@t$zWs!xhVx*@;h1RA=T+gf0W4iih
zo{FpfwH!V#hmX~sQR80*8|h}|1Yy8y=uOH)$_YGVe_+@g(c`|K8SNEqQ!y!U2t7sx
z;SmQC)DgF;ERS>w^AOTv3OrOr8^+HVV9>RggfLd5E!9Q@X_Hq1nYEa<-GBN<>#&;1
zsLJ1k&K)Q%jCIRh#4w;@B9#*Z8>Uc3vuQF%&-;_M$9A&n>FqPF!k4kKKVfT{D8R?1
zldmXxf7dkq_-5+>2wVQAcgH$o%xPBf77OK2t#St7X*Nx+@3_Y-(mYb7*5Q{wgJ0nt
z_*JQJCGyz6_~ae~;b=qef}vEOaaCY-i%bQxeb+TQhp$YP^Zp#qS|qtzyKQwhfAw2s
zsNI88EPd8h9G`e$+whqeHjgS^`0xB#(*pYb1q-GVcd?g&O(zrrF*%nJVkZ-qyfP=i
z0uWBOelsW41_a8$lyjGXO(ztW(N_W%mw`GbDt{z54!-xVs6SwhsJ8&45xd82U~e<<
zZLx=7keM827MtXdzrUm;s#H>y`{9m}7-)AlOQQIY$d6RZ>Dwfy=iTGW?!!@=$%|mY
ztLf$I1o%MWe~?IYcJctiX_#JqO#i-@QaOA*;>$>>HNHH44)|{OcNzkY08?JDzw-u=
zqJQJw>H9Lr;`<?gR`mH;=DYo!kqRPlK%Ril;9>utmw)YK5w@4ic_VOrmP~_`R^bb0
zo50efZz!)Z^qHuhs8A@9#^Sr<@rkPQ%b;xb&~mb>jthCLam0T?Jr<tK8+hf!1tm0b
zkgm#0{E`+>p3)kJA~A6zw#PcZ>RdHjNPl?^{|!lFv5n2+(>&W~i+tf@_Lqo`Tt;ut
z-M6Qg-EV^<tqmebGky87qX$NV&_Us2Bb*G=@4rl6cb|H;Wm8BYaZdhc<@EcvsiS-@
zyF5zV<<dE87rPuJZhbR3tAw|T_mcUD8nG{nwh9)z@xw8|Dn}#nyx5*&zvCo&#ed>U
zKNXu9l9&G_H5Qc5&HndD0oShpX8p9_DRCYK4N(MnC~I3i&HvKc3rf2-+slCPQ8PDE
zi!`%jE4AD$`YFeQw4z<ec#iW<dJdup-rmx4OOCX&(+>AVw;OsgcwzLQ!<Hu_M&b|#
zjb(}md2LTFgYqw}y`VH}D?DAsLVwR~@Z{s*waP(75&a-f-BoY*t&Ltp8*b>^j-!3N
z`UKeEM2xZejPjLyJp(wIvtb$KI9lg(VxP*hSE4Ls&omdu;q)4D>v_6F4=X^%$$BR+
z_Od}1u75joXN3%~%DoQ}q?+tjT#NXaLiz-veWcxu`Qx_<$30JJFwWO8*njEIDUopm
z{?UnW2IC<<v@ST(WB1{I-@g9z<!2B$?GGA?Nk}LQF-$-5SFll_)Gkcl3&CLkbW!RF
z1paphDH;Q<!XZ3&rhJ)pMtPLLhL-%D$jCStrP}!^0s1tQS7~Tf6yjRRyy*9q$g<6=
zB;Beg^o_Wbx%3t1^DO$BbboEW^H7}i>$Lki?LHOjTIZa{At5L{XL#CtFlNoVfXT}v
zplmG6B*3g<qf(YF6Fu&aEpn@IB`OS&&Y-<i5FR5b`KPQmJD^Y+&6$L}m_rq+y!Z4C
zOLW>IrS4XXTF@JAJpkIRPz;=nJb(W3qfGpU)1I66%;*l3#@OwqJAX@lHHlXOIT+d2
zpnMt5qZu&!0{wJ4b-c>jIqMkUa-XI~oUGVgquHnP&4Io}zn!*5pMGX5VAMq4*lLB1
zf`*lljyzWjnz+~Yp2IN0U2jU#PZU|ev$q_-2p0P}(9_5S$V2SFg7poweyBWJNS-O2
z2<+<k1>C4q2IuM9P=Ac%F|Q_*I@gs^u#k|(KJB%+=dtONX{F*ag3{hX{@AQ{tTS|z
z>ew=BvF<{r6+8K+z{|YzzFqp)L_C?8F*xV&TE%__l7H=9L9v`Q1y&p~U9m+yjn++3
zmFJQTd3>^^uUs3M3uR@;q{2Ay)qW3_#IU*YDm7hQR)K!Z_<yG0nqgC;qC#U}?3C8)
zTPkx}KOP#xW7Kgl-%6$2nBh<r)n%M3hoE0|U4A3A2*a8^;$GEe4UIF#Qj$I;M@HT(
z%=c9*Sd2>E+qF%E8z5svg*k*0DJ8_9843;5&(E7$*R2|q&~rdm`sATqV%?2E=LXIf
zSLQ6B!o==%t$)lV>M7A{2>7YK_$v!@<uKKtBM|K%D7r-nU@`6^r;g)zLa!*}AeGh~
zXC*hvI5eVzezabR6jWMD)DoWFUQ)3=HSQRBNl8UeE?1903#C;Tl+S0_e7Wj{8-2E%
z5qLm*Zm9rYv)+-zK`LU891`9TUM}4kT1G_TgTzX%wSPfuapmk+ro7@;4ahl~!LRFK
z@6;L58Ve~m`0JWR5$`vA52lvMWYf7JbVjvz3e7S0YZ71bf4DeUx2*yjea~4jj&LhT
zDkttO9JkeSUT>;6tel6{$)6RWW`43#3X<U`b?tc$xH2QR!CUl|@JK7k3hr+J_A2hL
z5I)2H|9?))VWmpHo6B@ARsvbt>9u<YZIytzzrvl5*1+j|EUKHp0M&$Zz;7O3Fie><
z9`;kWBU&SVk$RSLZ_ND(Nwbs8d7bszw~sS8&4tMdr!kfBldy`)Zh>W#(}<LGZ;4a)
z1r!Lf^|u4<i4dT55K)DG<!f&tRJ1~&+TAULqJQ{m-{lfQgTnlF%XVkS@oi4&GwmMk
z7s>YG4zy9i8|a0+qyz1OpoUq`;re?GEh(18v)Z8-r7PZKD6y`WFmw@`uh{(RW!+Hw
z=TLBHFD#c=fPpLRw!U?38@jq`^FRZlEaN&+ZF{@=cy?8(EO}k^%pRKR`B(#)UFhCw
z;D6^5x#rT41Krif9gm2|6x3fjI;?`-yBv3A@S(ot%~i+o?|IxiA!&sqlqrd8O^}V_
zA1GoEYS-g`)d2-9eA7rd>~Z^Y&YiP&+cc}U<w@-Nn}IYAH#hK8m0D_i*h6Z(wnrL@
z0ooQ?+4!}UcII9Dd*uB{*wFL_dyl;DqJNMf^ea<a%_z;?FV#TpF=nyd`K-0A&SfyJ
z!PYMVSN7{1yjH^S!`$#RPuB-3z#7u30hgfp9oALv5vU+idW9PXs*d9=dN2;_CZ~GN
zTRoWaqlj+wU)_UYdwQxc^?UVT*yhVsukFEj0^I|*(>)lf`pw~Dzba*gSA)%l4u8r8
z9JLV&u}H_U`_aW^9D^{y4;(*U&#~mnp5sOyH1tyhH|#6f+MjbQqrsbWx0KLy1Af9c
z=VJs-Cj<uHy!P{=4Q!x|5y1>Tc|&PH_wG)cfBX+VzyypP_N?Dztaa(GSW$c7!Cn0;
zkQ+o?y4IPSNw>&gOOAg}sND&9#eXqU-U;vh!09azJ7CV80X8b@(#^Tr-zk6s%38pz
zKsA6lf1{>X8C}3$t|c>stMu-kInUsG=EC=kAGT5yRT&`qR?-&qHB!2(`p=s{I~ak7
zXkoSYK-uYf$}HC<jIBc`t&5~Sn>oxSTsD{Pig@!dH_&+owX=&UYx5)bHcLBmHpsm^
znVTiA-Y?T-8>WG!^;UpN3O*k*Qcp4*Y9JbZ5swi-+9uoay;6JU5>B^}jZQznT(V2n
zIV^vlO&zd4@XyWq{{vb)nwRmmGZO<eI5U?animt7DHjbWf7M&bjwCk@z1LUd50qMz
zNG(7ify!D8_-3PTgE<XspL`hq|6wSFq|g#;R(1n($fY7BisJf6spRFai@f~ubo>7F
z?Nj5$3P)ho<@=u(;2$KvgXD*XE*3x-4VUkKU;gzXr84}w@e8HYnqQW`JN~5Wvj)c_
z0MiSu&lZ87e`tR-`X$XF{(PIhtML6Z%{SL)lnQ*Hz%+q>gWL7r-+z6QiCHPU317oU
z|JTR&r~f|ntP^R0&R_oX_fz-*H3;q01(Xq_%;ldyE`L7#){`xZNuD#kk-Yr#*QI;%
zdy<})EInxqQ6oK&EK<A}<An}CP13`6DT#NBylZ4Be_)UTG`sxv>0pr}O4b0FKxe-L
z^%Pl5k`z;7nPH?nQ4<*W%x{tX<_WW3C>tZdsGhQt4PRxnMPV%TqxHZCKVUnNyi52<
zJnf5gSDt7EYWqlewiIb$ANu2dnYPs7eMP1`AI!iD%Ma!GwVf7VeNU~^C-2_!RL~a)
z3WyY&CpAUSF@M0hfg&rN{3LavSp!H(BW`iGN_pY!KrL9ylJw>0_1VZ6M?S(Dr{$rn
z5!TvWd7Pl=NbqaWE`EWc7o&N1=5mUl*TVXdA}y=LJ0Wo}8P;Vb{;1ZVYLk@1iXuqQ
z<gD%!TxyQcn#`_htey-rzD0`7lbWJ|82fwUqGVieiGT0PY~`{{v?no0L}v4&uylTl
z95qjB=gFZ<8-Z;0MHBpzvzdoaN`D8bk}YQM(<l?3P7V_IrjPX7s{2jpHqPa7JbkpN
zGy5O`k|m-@5cnowE=V;vYk&bLzrG@%U?VxQAT2ET^{hi)34{89S*N4;;H=Y8-^n^X
zFLl;AZ-3_GLaxC7doyPigjMtt&78;!Ywox6B^dkdlibA%R7#p9bWh;ewdSByusgrp
z^*LxRs3xl<i9bgDR|fnT0DqeUru-1HM*7uE+laqXGV|;><`|9mOBi9xlbRD-a~0-B
z!R!?_=ouF1BJNO>urY4V1TDC3=qAe7ok_%A=70KG6zUygUXs15r0*052+FTMdqRE^
zw;4qATAsK$DMybfv?o6*wA>YJ@^I4g_^68Hj~M9;yg}k5uk_aQ$^5AZ60)D;w0dnW
zXlUR`rC1v32dxXW9e)ozg!<B%U<^DD@nxP7VoF*o7@M!9fwy{pK?<4kIJRCxW-5$1
zpMT+{musRFbU&DuLt$O1nJIg4miK2`4XcPBP1`owEY30QV$;e()K$7)IYFECS&U=#
z$nQR;U!$nN2oEROP>TJEhFO#86tT$4sBTpvsSGU|FUBIc5?xesHbt2}-#ZgcMjm4h
z(MzUNng46y<TcHSoRpO+<4Q3q(guCY4u4y@NbPDGO^#?TxyNfV3YR6Yyh4Y*Gk<v;
zmWVbOs8zzL5{wW-^Tsx+pG<y1{b^*BfXPjt#{IK=Qr5_OS1*;RLos!joq$RcsfqaD
zMwd$fYqTYDp^WV|xRSs&jWSeZ%Jq~6kx$H35qRES9lO&w-SR+P>H=6+7coMBIe%5%
zdV|{ba>a#*vpto)yro!>IIKOX+!muLDc;Q)ti<8BG9ty{L6uqEhg>Da{S`FJK?KNO
zP+4Ii2a(%KaR*un%NuS-580@9ia8gI8EKYFqnZ5FbWVb^b94@21{;+e9-UJG@AJ#)
zj3|evzyGrC7)EpDp@$ux>*1_fi+`3}d05+12iVq?bo^<L>Z%$27+skhYKkRN<JTQ5
zt-_I73RgK<WWXE$p7F|-EVN^Pllq`6+R;)k@@b=#5_Z{<)S8sy?s78mjci6Qp5xuL
zg4_<(j*FFHEBgj{q*X|_(yLiD*-+py`03UUuo6`~uJBiljjHXoNgM&K?|+>;ae1~B
ze32Ll>5Jxq-sS4-_0t^kMIaQ$w{zn;&4Kg@7I@)d@*U|@hNgn;2(4t#k{#@#UG2oD
zY<(>t$r+;o5L|a`?OrP>%pkhhJG||pSgnyCc5?yF>6<{5v$nJPuIEaHa|GzSRR`s|
z>APIj4r|hF2j(lT4Z2NMYJWSCCR>DChwiE;Q2GEX%I1`%v^%#44O*aqeo`uF*9>?}
zRY&5*S=-V4sr{YC(@d_T#bVgojv?L%BGuD`_k9cD3{P^H*_z0GrN(q{G3V>@?c=H`
zBh9y_xPkRdw{Ys3f9Z?EH445eG2O0AQIe2@ns&sA(hCQ6P{a-p0e=({j)JmwbUfDI
z210#UI_DEV9+Y@5B$r&>zF5+XEYx&1mA`!jsqA1A&tP4;o{K$ff`_Gl3?`q(E6OXO
zJl+?B+R0}Si!)}JeII0TvfsprS(L3YiyF*M9l|WC!7LNa8ndXwn1wTuu|3z44geOe
z8kOp2MkAwOWcsygwSPh3X{m?dCi9N%Gc9Y0lpQYYx32Q_+&9L$lcE+{Zkv(PGXC;r
zNgm-OWx7L<j<&buL!tYk2l-GfmA(VB%o8y*Bn3Ss+ztZSEx1h5tI2STclD`=yMjs4
zD~<mW)IN<#ltE$5odrxgS?aedC!HMg+Qy2SO|fa`e=hH|1%EUt`ngd?ZWGO&2Gpy~
zLN@@E1-)*Zc%9V>dIA1ktu&Kk=b?hWi}l?jTE#yazwN>zrK~_p2Vqe=`fx5|**XjT
z?1L9?Ru$)w>#mV%Be^NAWuP?mWY@c7?TJ<bR@vz73Lx8tU6vyx4stu6#W~Rk9gMuC
zG+O2nS3emJ2Y;L44C2+%UUWNpE8ca=kQMTCb6F8fi?<VH(@KY{a=ZJ>B41+VCYaF-
z=2&T+D3-6sOeH0L%qEO7_y1;;zFKH)-wpuIy7c8EmyXK8BbRPgceF6Ks!MjC^;KY$
zTkn+=sCk<!gE+_b78ODSzARE}?_1QinO2ADC?tpT9DgcuRdM!0Hv<CAL88CUEio2L
zKjS_(kP|9s?O`{;I*Jdx&)repxzF8={z><_K^vh7A1Tk4Vss9*1)uJ|U)`<rfcw=w
z#joA3?x^h_n*4|hzB2Rz?XIj&T<`@7${%*uHtC1Uyeiak!=(Ff^MWFwJ$0S9M_xJw
zsTsgL@?T#aI*O~I=Dakxt|f5w*+XiPbxpo>({p9vyY5a-3^-sjD>bt6)#*VxPR{`#
z%P+#pw{l*Xni@-&z!HvVS`c5nVmdKxSuG{r?jg<h%<Ax828-#g`{Z=`Khv-BT$h1O
zClvuXm$B<M6PNDLFejJMO(&TF_qS_KCjtR~hx9beG}B4fW@gJ9K#Heu^SLaf3-g#E
zPsPI*hm0u`VnHd8a(+6F>B2lt4x{Lv2(+nAgR;(zJIp25P*0BO?i3EFESZi@X(T3F
zV_@^l7B^4fEd7cUVlR!rQ{(=w=x|WZ+MehvCANvJ5AQ)t)*Y`$jlA-@7M{3u+);CX
zfAr8Q?|UYEBav6lOEU3OL58BD<44keF*8TTS_Kwc>5{W^-Rf#5&x2j*QoI(kJ?ZjX
zJZ7h6lkVvc>xT%6Y-o2xy{R8!P{H|g`rRWBfmJYfOhXE!G+@U-Y9y`$>1r^@8>P7x
zC8Wxliq?@!7*m_cnVoN77UWRrP#GS77B-Q}j4g<jAQ%x$OyF4AhMAgIT6rqrt82@8
zo}P=Wp-Gjiw6>tY)WSG#sH2bPx=igD==Qx?D{$^V&WXtch2a0koS4un<nYUI;wvl!
zr)s4%jRkD0ZI_xzCWE8i$@A4-H)>sHHrnKb*83V)_nK}3r{BK)YI-BFGdXB~q+cKh
ztu7xW2Op0t<4F<TvkM!Gva%-`ZY1{h*i!l>k2Sj7pUyQqq8_hpT4%f>@6PFY2U=$s
z6jSYrsWNi$x$hrcLR&5N{ISqTyy0}ffLGT2a;F1Zr-+Xy_Ct_LfiZ^GhpXtkVqa~Y
zj!xwY`v)0x;T|<JX3RkgQ!?a#pO!hCa%H`OMxe%-MgAYNfsvJ#yKgeWDs_Z)<_6B(
z>(aS0vQ=(&GD>5t7iS&E^W?;S4>P+C6O%CDl5KVEC9Tt{B!MymC^J>U+M+7kPmF$%
z>&RYukJKHxyczR-HT$*Z&<YdutblBb$)sngI;W!B^Q<D`rq0XE-S&ikk==x=wY%z5
zH@l4d$`I`?*_HG*`RZU-Jm(LL8+O%rS<35RDKW{f1z1^noOy37AcxHYa@Lw4zc10Z
zlE?n~(RU)isEmg8ok%dsQn>U=<Z|ChuSDK%J!n{U?L3XZtCOZV4-Bcqq!v)>(hbLe
zh@{N)7L7`sO3K;0ba>Q%C`_y8QFANt>MULYouAu}M<L^*H-HU6DBpA_bZ06BHi#as
zBhanHH;c$RYFGDfK$3-@IR7prxB1W8Bmg8W!u0k66AigXTThWyB#cKB@xY;aibg_l
zj@IP%kz}2xDzn3YYy-~|-Osg6)GO2-kC}7cood9UA76eq$akiH7Nmv3LuhyDNG+rg
z$qp+!dQ%C@%IYmU?notk%#GEai_FH&$a;@O0!%96jr42jgq602tN*Z|puFtyZ?Fl_
zMv?1UeH)3_@`sYU^WzH+nNJkA-B6f?6!LU9o&H%S{$@1YyPFeGI)XbFcP(M~2(p`z
z*NhU_{{ZO@a|{ZXf=wqA0XCN*{woxhaI-)sf4y2suiG{dzWZ0Gw-)GQI5S)U!@#ps
z7wE0QyA?eZO-?y9|NqdTxJaZV+j97FkR@}Ooim@tmcy4p4!<AHZ-)<8QiByvgT>+P
z^8h>{`4c1$8a-G5VF-q|uftzQDbetC=1WOQ#Fyn~$9H{tA#l6`RJ!2wVioWR`IDhz
zf7+w@eu>W*p08=YIlU-}-~$Dw3CsuR^!M8z2bqMG!kh3IZtb_<-VXmATG0(jff9#*
zz8=DaA`s+oP+mBo;oHB%=fj^Z)xt~DFnxUcGBhKflJOj{@uV?YSH=U$7FB~WUQn2d
z#v7G_6~!cPiyvA>X4MEY`*8L0CMRWue<GVXsgF`>HmhRuDmJ$Y=7x&vrtC_x!}Wrq
zkCZe!Um@Rpq#8JkocNOGX}($?=Gm$85+!)RxC{Nr2xVQdq}{OOnltZ(ID179s7#yG
zq><(j)t96R^HZ-m%r#4(M%}`QJ+4`#tR3x<zbRFobes#TN9uM#J;+i<gAGSoe__2T
zYavmwSrX<9=9p|BBscR}{hDGT#KyHy(I<psraeW^jOu8ENsIi==3P2*WlET|=4T(C
zA;w;7dF_yo<dj@&ij`}kXe%ow`v#xnm<|?<tG8{MZ8hqiw43=eMkH#-sX+?Gm!lk&
zi_uTz+$^FAQ43P0)k&@Dl+MUCf64kfMcX*OjEnl@uE#U+Sy4RINFcA>bR^J1;VvCV
z9VlKMbY!0k4Qo2)NK`ew4q7WjVeNuvk-TKO$23r!$)s<o!N(-;&LXF{K65o~iyUar
zT<~6%yI*+5DtEi~Xy#>&=~#POie1i7u1}{i&24NgCF=W2vL^PBay`w`e^AT?!aQq=
z`M%LCtESjrtu-3qoUgD$u8d~P*1hqMmP&XFJ&oN^?2_XaF}Sw1V+<-sLTc5~EG@+*
z2KSp{zAwfgb%{Y`QAn?UU<|U30&GtV-cVdtoE@0EgMR3^Zu*}3H54ziWVWWu`@LT%
zu!0_rm1o{ExYi4xQq#!fe>_sKX~fGJ?WTh?<{i}=3%hs5!kSpuR%LBwsq-D+FDfDV
zg|*?@6aJ!StmW_N)fw`V_rjB2(d&I7FZcz|y-=55+R2)h%Ck_Hq+?9$TBvIzqw^nw
zbV)m*u_wxCC_aBg9)ooGEU2J%{ur%Mjm#1x^6QFcTQ-8m4qv!ef9g&NfB`%w_GVYd
zVniHUMp92#)D)jT)sG>Y?{GyeDCoyrQSa*M3Q)p0-P08f#XWduzg;=e39qQ-T6&Bd
zcI9kjrD=B+yFP;v&{oF<{mJ5SrsJ#&i2$jWh=30ug0w32{$hVIkA{AsUGxiA`E4MI
zntJXmrqR^P%pJy<f6pcjVjnR}68ngDyN~!nE}$2(Coq1TE~GZvhqV~f$Q@;9GG>Nm
z6Q#%%k3eT^ofB&_63szYgpix{*71pRrVW5a;NNs|3I0V`86&hQA-dMOhu=;wMnd8l
z_2uU%InM;WQRb)70(#c?za;I}!y6@zg}ad>HHCNM0H*{hf4}H>St-cC55WPbNcf%!
zuO=AE!`SC90lhz`v8lmeLzq3dmde0#3bW8GriBzS2<2rAmtNHa8EQ(x9mZsnStidf
z7!VkZT&s~!4;Z}WigY)Ft{ct?j|*z!QC+j%H6OAJp!ZaN1fh&yNDd3P?xIS3n3!5a
zb*ttPBXDOIfBHWE-i@JsZ70xm?%tIF=LP7V-RYL%3OacPg>e2-U|Q;+ZywoZb@oJF
zJ+oFkDb2`=M?y@4FFmLD<9@m@-uVSC#WCg5*`SSCnZW#Q_VSqL&+98$=e$R!(__-v
zoXTllafv;Y^sTfbOF475aZTcX2R%*FECEAAC6QbBe|yBZ0;b+d&H#GS%1i|~$Kk94
zQ#^uP@3@kJv3y3Swm;`8Oi>(ibod(vz*Ar$XQQ2fAY`015D;W57JW_LxTIa@YG#HK
z6PFF}mMr#=7q<Xi6H9Y_++52woIS;70*|X6(ng^5Ex_8cJhb^n!N9K|)@oO0Fg~pY
zD*yldf9o$1X(_e7g%h*3^_HLVmfKurIy}l{z9n+$nz^|)1w_{|b2k(_XYQK%z0BRL
zz6HQ{V}<l?+VEK5gzOAia8xyN!$$}lV6d=gof97NO-3LMrH=gxkc^yM&mX~BiaY)r
z^&MEr^8}hj+vkzLK(%*%ulgTZo)HuZWo~41mxfIz4*@ur5%Dh*0y#33aW^QJcUmWv
ze_;ZZ&94tXt#}gl`eD@yXQX1z`|j~&^`*nY+pb@|;!Js`yMMbL%!I;U$C))`L4ql5
zoaz3?!jFGFKmGXln_xiuCoJV1F{y=j-AjDb8n)rwcAwL;WB8RI1M`rYIs^J#VnIe6
z0FT{@qy=~F3h3tm^^6=zSs*LHl(Seke=sOOS%yGyfM$N5OJJ62WYOz1h!LIwm4q7}
zfuFI$W4OxaycAe}bgQRsb(gHFjWHHWoRdo9Zmc7kvZ`PtEJCQTYz8UE;B;_GR`Hf=
zJp(QE9e|R-m877dV-^vkx`ErEbA}XpXAo#*3K^8|Equlfjjn(UL8_Vdz%vYKe-J`c
zB<n^CkKZ3(65cbgZi{=QvlfVQ(xzJ*3oBC~)hDiYlZl&Fr^P96L!J`yPOCnrGLNa;
zs`0_>t}A4TO07q&?vAzuDv8!~ge9_66BbY!U({fNnc9_AAj^om!3L-^8g+$StSfLR
zVi$}<pgRS!2;K?k{Vg~-G0BOLe|TQDoX96@Z>tJ6D=%_BWPD2{P!LgO9sCFR33Zb!
zYNaq(p=IFJ^O%!Ra)A<QTiOz}n@pro^3Xc*$h-mLsP_OzZdnf0ZGN*^8y+e=^u#wA
z=<;yoClSzgNxo^_l_ci4YUS&9kA6StBX`?-EomLB+smyt(F#1uglDYKf0vE4JE68R
ztuTY-eAKhocb9b5QLo6!UPegI$wg}fdcjgbb6^BIS4p~EiQ97+<&GKG#Ab@*?CYB(
zj~#uj`#rUngOV;b>JWEYqWrCwEnr*?zuk;J%yFZKJ^DD$57fi2vV6+^ND$T2mLf;w
zV8`eQ_;PL^Z;UZhY{tS7e~ZnZUA-)I1M8w4$ORI+pp@K9bvD`)^Ky*1p>M$jJK`1`
zEOlYX6fk-R)yW+UquyCd!9R#%72`<KoSbt!#gROR0EvzXsTkd54>});7Xd{!V;u|v
zlepWYv5_#oH!|!&MLDwA+pVoM{_x*LLeY~b&M>W9jUq4B(%KJCe+d&5ComD0y@xAI
zez7E<;+i4kDvTeNnFutLt$1QlBZa_L!q+Iz7t3W5T-rEEiM1C^p<gtnY}s>q4yC|W
z5it}JnvFbrM9~~+6w#^@j!7|+sb4K`TLim#+S6r_Z!xc=F=$6ciyh@Ciz?`3Ib|wj
zxlxj%73iiab+XW8e{prnzsT(>Z6kTJN}H?rl$sH5LVXOk+vW)c^UI!-$!jLjf1}og
zAL0Zc0cK91k@>xQTHUo&QvpV&)wf&WL-W}5RFL~L8F~)ma%0eEmqiN!?@*4lA{5ub
zwx6PYdE*JXOVPAHy|4EP5;{$c5H-R5jw2b0f|U8A7ruhre~*2mKo|(*c}*DYNh#H^
z*iwv{YO$K?d$ysgf$Of5M2F3J(3V1C<!g)?=B+M_wxNMeTr0@^5eBchhMw?1iccHP
zp~~)2JEm*QPqNwcQqKtMg|R(<^g33M(%0FN=yx>@eO1$ePbb8o?U{^aURUox;^r4O
zQ@S1!e%JToe|u&1$ilg+Ht*#oz)hCZQ_M7y6q8}#t+kHQ;Vx%oJ7(6APBUxwZ2IOh
zC_T!cZ_6NNf9LgC${=AB)?A;k4z?5z8a^f4_FaiX=)pA&du8qldtKZqk_d_E_3%PM
z79}C}ja4epkVJbLrw(;@TtB-v%zVYfQK3;)t)_W+e;hmH_a~7@ifZ3-dTSD?@D?Ku
zSl_6ZP%bQXCrqM-;_I9+kx=j7<CS(8XHThOP4GNrLXj=c1IXv5w|=+wu2P`{F|Pc{
z$6-@8{lj~eB>6bp_Smj^S28(1mE&<U^b}Dxj0myV><`dOV^A0_tu?Jh7D~tL07>B-
zhxRC4e{YmdnuQ1n@3Zo|wO;%71;C=T&N8iK%VT0InUj*ZVPr&0;jM*DYQ^YV)-~zT
z6MJ>Sq;Dvma7t<@9~hIG)n}TMWH^+t`pk1u!}XnWQbX;vbJ9}V@CvU>E!e&=jnhZp
z@WFT$WWi-pqo3q^E?ej;=iDGILQTp&!@YeIe?#xYtS~1X5~V~M$+7E`)eHCfJ=)jn
zL4_tJAw?_A342fUoCM1>9Nt=@1*T2I_ugY!!NQ^IS!y&7744M0XG8InHM*vJs7BYT
z&s3v@lgyekm*|@5J1cZe?8pNB_G%;$?2e1sYIFjZgqFvhvJGh{zIm*guC&VscY=fW
zf6T$%c09N%POy8IT@|OQr#!qv)9Z_~yCt@-7+7p^ekZ=8^E-!(^rg-RPwzZt0#RYe
ze4s{?7#>+@YK)(k568I19wHl0%$$zd@PA<XHHTx>#8?=3+QsKSDANuNf^5|q-+L{d
zli6?t?i=Lq`Isyh#qoa${s$;sIpYdtmxE0w5|_a#I1`sniw-H5OJXMu1L5VHmt0~e
zJAacE-NREy_<`e-ID|%bmWUZi-NS45`7P(d;JU%3IF}Na#dn9h-mE1#v>=r(+N`Z2
zNTlBw`8M^@xZn2QMfiS~`kT#KaX~(4fKGxP(wi?2zpZ$3mNRd{*Kli}K0K_xty<TO
z<V<kUJ-)8O4@we~PISsMN3#3=d-t@uZ-2EGTO3{b56!#p&s`(=l)&>G;BjNL9)U+3
zh^o#Q&t&+i2T#4{Tq9V8pcOz+0R$Qa`R2~iq(;Rt47$|_1i2Q_1jtm>+-qYh_?bVV
z`y&b5FPe>EWJIf3@R47owN=bmYFo=KKOn#~;fRtxB%k&}x))DnnUeMhJb{{pdVh>J
z1w+y5Nb>0b%*R#*g4MparuAT1>Ewyqft+JrY-tN9=_&N?HfzH}`skbZ)MIFTMhvR@
zXlls}hC*m&#TJGk88zXu2hBqv9aGM3MRA+_)3O^rqJJYvJ@4|EO2d?OWk%o3C4KR3
zp0ofQv)tCCrEU_o;roY7R&nfncz?3tNoDz-@vVAmGkS_hCQI-7yDroi=BhSBOZ6_H
zCbR=FAv19~nUUL)4s&%Ya-j^<nA1(<y@$AqS^vf(KaH*Ml3Ax(3-mt4a3ncvBbf3n
zHw*me^+mW@TaSELlABwN2VtKW;O49x(#A1sOvxeJYmV`_1@cKt$-E!dh6;+OVib0i
z6(Mh^mz`uM9DmeC94#i_O2p^ql;noNVrz})w472_;QkZ$JDLRJu(e=LjWr%W4ioux
z{ys!`H1|t#N2~F2+@Zi)Bnx+_wDN^mdXyL74~49fMsM#qcCM7O>u+9ld*<l*rD2^a
z0xb$N=@CzvXi>!8E@{Xf*`g`%=xy<<ns>47z(Bc$%}RTXh}$lnkrY-_-y*YzPj>PM
za$MRioUzZ5Iz^|0=68u@^`=LlL>c1|^B#N6um@;uOsqA#yi}xc9-~&nsXGYiBop!{
zvYiqFmx5&{C4Wim&;MgaUT6*6w@__Hei$HL5FAP7PLvw^UF};7GrP~7r%G#>wRQ`=
z(^oa=JVG#wo+_or-cSy4ih9Kw)(QJ#cc=s}1kUlTH_Pr!McA2wT^lQ>a<#C;jN2;3
zL24ykgsi0$n5;{Zb)#`XvTii*A?v2~g~(b`oT?Wl>wm`TnPlB)JCLkTCne#rIkCG*
z>HJX<)-tc%g0sBQc-FydD!oA$Em@FP46X*U_|~D}$1b*p0_mOZ{_Ua|(*}P~XN^J@
z;74t6_yugMH-9}p{rdP7ld60EJ2VC3fMdzK?zMl!+UVe+?OxKeWB66U2j;mnp~}Sw
ziDyg4B7b=7R!Ue{2WJn`mmu{_Ic%M(&XMOV8WsbUDEkmJ7HPKkQVX+JgNYbZaU(nh
zEvavKME~3?Jcg@&&P~zlk8btUt?m=5+L)M>SdcLKAwwO8VgglSiUN68$Yzmr4ATsj
z5-Q$eKJ1Xz@fM-LVqp}{8p1KlNd<yP)=JD-Qh&vrMbMQsWKq7i@*Oxdu|hH|zf61d
z!-fT;i7846-5%lL`{Qfs_l#a2t9!3!O+qrIO}7@vG>dr+U2z>lvP~`fsd;Xl**+ga
zWz4Z`Wh%QG1u%EtRqph0t%tzw4+evXX{`?cqb1@B1RhlR0>X}+SdM`#EUFO*!<l=L
zU4K?UiCk+bE;5-=RZnlbXt2~T6Zyt{@$F%|FT`o(mi17Z$rUhVstSZiwGSC95OqzX
z__>i;7HcQcRaFvfRUq+ikBTGrIH+<vvU;1n_2YU;mh4!9rBY3FQLh{784;JdB9o#l
z-{$e_6R4SKX<k-mo9L~|_2^u^9_mlim4BuI>$<|^hT2PA+onGHN?N1TOM=gGSZqu6
zs-6$M^}BIxtE#f4aij5;V2U(`^qeA%fdG2a__G@~OAGtDTG=zzVL=lj9MrbYTUaZY
zQ)Od2{9^&G@KyoSA|!v_k?=lv;(K1jso^C>8U{5g?bUkf$z7HE;vk89%qtr97k@41
z%3~#LB^f7C$n6}zjwq%URFGh?6ye4%XR6mYJJNhyIMw?maM?*FRB-UY`fmhw2%|!&
z!`^p{%GH0@sf(WMA`FCP+(lfkv6a1gN<FQe?5QYhBgZ}YyfG}8V3%>1XBZRW>hruY
z6N+qp+%tFC0fy0xTK`_Ikyls0;D6KDtAFNA;^3p&{$H+VOUsU8A)uL+)X-&~yF$Z&
zuV1`Rm)Iv>x&PpKK2w^RF5NHkI?-W+i4$LgOPkXN#(B^yMJ;e%6RpMi+;(aa2Xrij
z`iTX1;#iHWUGQ|<Xk4&{(`epP!)aPysD|Uhvh6Qi!WkPqvw|~Lc1QuI22<~}s<*(|
z(fAg;+p>k+|1VVF1SuK?r%^(P`S>>j9w%|tMo_j^(}>4f8MR*X;}b-Z2aQl`-OO8;
zf^pZ**pLoql;bQW<WVmO+B;bCe<?_`{FkA$9TNdLmth$M6PFokCpLe=3}-ler3(Y=
z-F1N8y4YKgQ_<wo99p12e}0EANu*?3S}Q)tmS}Q@{5X$qD9i49C%eyghqt?rM@5|#
zPJ_kn?OO-@gXB+;+-P)X0fbS|z5VFEJV=R#uLEBsrBZxZzIXhs_d5lSdjRDN_B%_!
zEz0kWddOome?O&r4EKM>Jl^bgBoTa&z%s#vYyb7_w>z09+s@2b60E<>Oz{X}Ni0+n
zYk>T%`9|CH8$ISe{cN@8778t?cetl;f6P5i>6v=%F|3GOXtm*vwQ9^-Owtr((V#tQ
zwq-=@fxZ4PvqO;>Tgp0YT>ZzR!OjbB(9n2)lvILv>&7VD|M!0(57Dzjcq^T+fxE{K
zJGA!xTOMic>_{{MovBC4^U4$mG`1e{os}Wpe@S0O=ZOVs6zF{|;I9um@b|y+r?LES
zcFsl%w5}fiOVSMIH9IM|S3NT7H20s9>QB{Ep3lE$0U_yxG}Ho;hRpjQff8(Ct);VP
zg;5iK&_;mNV^4n_>^0l3t!$SjHn*Yg^u1h`U%cUQ><QWSqn``Z&S577@#U0O-9PO4
z)7IC|jF^Kt#L#$wa)L!08_uk49C1$Vi$W!At7Gt6#_u_=$eSQ4e0^@7Uf%Bho@}A^
zB$P7UA3yHG55zvCc<tnc14iS|?%Un(t<*eMDI^a`A6kDV<f@j}9PYj3B}rqnuEISq
zxSFj~aaeJ_EQ+4f2F5MTTa3_v6t`k3kt8P$-AW8w5CSF@=#LpH3iB)<ar#xu>nH&j
z2A*BFk&JRaU9j^Q3vF*TnFlwpG0I^6FL_RvA-#wv$6Ot(XEC0Pnq~34-w+f<tyRnG
z0?cczs4jmx>pXH!c$`QJDCs%$j{BXFA>8Q07(bmM8}KVFXYfM=LB>z;i;la<{!o!X
zg{_3;p&+buwQZu~tm7t_B|^lPnnfBt=LVXrR<8FAC5_P9PV}3z`W1Mg5GMMc!j<R1
zfp_?bbEsOz{#k5avNf{R7SBI4c0L!zJ|4B>>`8x!(#+9Ot(|a=DqIV;6xX*4$t!Pk
z+|+FJIxsLA=dED0(HRrZJC})j^@=S`_e)ZJ2CKOtYBj?cQ)}dZkBI}K)(`0^_I*lx
z6E%%^Vt!JCtjITgHeXw7>}9IK)uXA$_4d*x*wp!X)C&ik`d*Rm7EK`>^k|y#oC7y5
zzZ8Ey9e8m1H3Z;L2`H6T6}C;!>{*@*CgzbPQ{VpB;|!*xmzmYE85fNFL|SG?alVHh
z!3&i^$?<K_WNrZ3cbe&*)g|2=q@M4sP*^s#y7zL<bczRoWvb7f)XJUWd)Df`6~0}H
zq?f(Y8MbJ$k3FENQ0hya9IdCN&SrFvqk(^BPTkCKWxhK}^SHz^aw5HVCo^`*v&KM<
zqn4tD#!&a3Gls431E#RV<x{J1L}dw2gbSPtq>FGZ!IS|Mnx?Tdc%WPqrQ~ch?|%ug
z%a^AVGvtE0CHK$KD@D54qATe%${s~hAK~QB-8jXSA=n1NjkhTj&|qAnBpmBdWO;uf
zcn|PK4uz+;?(%j~P9!M>4uEHfVJy2?XU~wXhm@QZ$Q<maV%nQNao}=8?f97O=a{l|
z4w$3rJ+c8e1ldf2Pdaue85K<QKn~T#*J{f=IctYp2u!5qbBNVdE)rL{8rmusJPNe_
zCS@+67T#`H<<<f>EOKkX7i!$P^NoK>T$03`_|_$EqxH%Xw~@A_#J#FCQeFw^PnSm5
z?}}mu2v%=s;*G@Rkul>rMr?!iafCx{Z)|mq)+?>9k+#6<p2`!RX7j6#D>R7dM4BkG
zPN0umxseDof9_$}?HK+Rh$|dRIf35wM*6h9V}apx1zS2+EpfGDU3W^Oqz-?_Ef-Fz
zwm=S@;-0i2a*f3I*DWI?*Qe@oR@ux3=Bqs99Bpxe#)hF04KPhJTjE?xys?Ec?H2}z
zSfWo@==^!&c_NNLymvDPdgtji604zh^xESU4!)9S+ele)zA`%a3g_0koNK#sragX5
zU5eS<k`qigC@lSxWcM%}@tuEqN!=F?xg;mkzAi7tcX@ff?xkV%THQ`xr-xOCRlN5_
zn`NHNB}ZF<pb;9)O)3!ZARM_3owSj-VGYtqzEFfTy>C>7fbn7=)HkAowbCouU@dF`
z8RRvpm!Rey8H-g}*UtY6+Q+nuoNWX)tZ$V-LSYG7fzMY6G{UnZHv%V5<mg0;8=I0_
zlW*t6#BxtKS9wF&OzG(zF9=a6saiqsZeYtB$3>Uva3>{y=GBiX=aSryhfmFN%{sII
z$@@yYztFSr;@g3Zle}Jf{o1mZu4klg5Q>9VwdK)6BXRAa?3$G4Z2VhYqY+>}baSdR
z&b7T`?0&w^t|bYuW?Q>PpRvK=?_l85A1*htW3L2)yI4%F7wnKDa{mr4{zY_Y4bXlw
zBm5r%%d7d9p|l+n0y#04alZ!>m-9wGD3?%gHUyX9aVM95e*Ej}pAV9leR@62%sJ4)
z7`J@>e)xUCjxS5@eUL)OV3+?asTl`hXXiZLwE>9Bzp%@@e}4V)r_X-_fdl@XM##ko
zWg&*;dwF$EXykGEMz29QE0HRQi0q=$q_Ec5I*i{xF9)N%0Bu8S{FWM{Fu^F*Mrsf`
zc<fe*cTF^ZeIi#Id9H7LBQw+{$kJ3E(p#cJW78{+=TYf3UHv<E#ZkX4hcC<FW7TBT
z_=k7rrQ1>@2m>DdrBh~Y3izPBW@SV#t=((w1g(uqxZsRN(Sq>kWO}KjtFYYC3CvAM
ziz#pu-{<xn1x&IQl@yP(!bVsuvJp-4M<BHp)wbt<&);d>R~+@y-3y&NQ0UUuEqAD3
z;WDYbN^RI&rPP}`(<0UPM{V(LaFJPDWESxGk?!|-QSrQ6i)*Bb418QV{)(V?P127S
zYX?vP=sf4O!(44>B|Z39)v^=3SOnv7jl`-K3@TRPB~cFn!;?$l#!{=nP-J8Jg|vnt
z$$Oc9^ypI$eoI=+RQy08jnZ@}qTuV8)WdfcX=7t)&9euZ-IeeO-lr}pUYb~$F$2*R
z<4m|P%>*}@Z&db}Z{p3&m$5s{%je@=2#Q@0vdg`D4sg;~+6tqa6H6MMX2y)874heb
zMzSQnnaeOQuCxLEt^@UR$CQY+#Vp;Ex)e=+moHn^jIlJ1r+Q-63dy%{7n@G=RE1=k
zF)1{4ODUFrwz4T(uS_DPlceVDHG!Ey1np-vlhc5IpYJLet>^|#=(C9T#9rd*nWD#J
zI$D)X)0Xw63sq84q7F>`zNiRbH^on+cfwE$DJ2a3W9hp|?0_Kvz@LT)5|c+sYaT;?
zMoM$qVnVnOD1df2&P&Zu0d&MUt#&o3g2C|+Wy>tBWxxxqc|Z)^_?GtTEPWfZ@_z&y
z5bY5B!kD0K1`olQE71c$K=2Xlgb8+<|DgB*fe->ge-i*<qZTo^3lRkO=IHAw=;%w~
zLO0#!LT!>=0T@7d|2Dv2d=Mbjg@6HnW0mmeDOC$cC@B!Lzm}W~bIkN9zR51W<%n1M
zbje_h+;vQ!x<l7f<#b8CQqmZrrsNqC2p~$j^~rq2juNH*#`y>r>O6+}fgEWA(Yzn{
zp7KtIEV<VPERb3DnuW6#OYFy`f^h+@nSw~rR;R$Q(fa9S7eb^Zj=Rs<e9S|CzEtC@
zCEPLFq7M3Ow3|uM)0H&yY}WUFV(w=@LA`F}PHQn3#hN=A$Qj2JOD?8diuALkBeyge
zc>_N9V`(`rH}x#GJV8m44>4+bX3LKK2P%x9QojE{HjgeTIo)2SHw-Pmua$%w{+K~P
zYDf0b0I=wQ#=$)D_vOps<4m-FFL7kAk<KM!^^4vKgWGUSp#nlhMD~GrG%3XJzwDDk
z%cmA@KpPQ}lZld&gFn^Y28D8J?tQ(tqV6AdZ1Ony8cC|GP^x+C1I=_g8eOdokLv@I
z;WXQxHUb8Dt5&?NV#w8eD5(Wv`kC^CIa_Zp6avkuch>O3zD5C4Jz^Vw=DKEwPP22o
zR5y_h(ICzYGY31;;a*)>Sn*q>!!_)ry4WYGKDHO#xirg3nLUFw;ivrwbX&m3E;gpV
zj~sdY&&erCW8SQI{c#IloL>~?6&>b@&u^7FsI!T{Dwx{ZD&28S_0FWqql!Il#0`Q{
z(rwj;jrO>EXk6CnFF#{{!a_?WD|hj!YA|IM6Bu=v^V#a;9=9@uGCNy~*7_%SheGeb
z)i=##wq?A;>R>JII#(1KRj8o%>Eo}4bXcCw=#cxzY0od~JdM&uc1g7<1M25!R;N@S
z^h&djG}E-UvUL6;4WoNuKJaHW>>7s2&relbREFzytuNQL?P8aIV#Hjq7iqfkG0|UH
zp!Fd-zP3D6G(Ti3UKf}VU96kDz8FYolDP~8XBHIacRjDPb7;C4)e$=iUEYJPur8M-
z6r!mVQqJHy-2CMoVESh%RSqqm2>V63iTyxk?1=LTt7|?qA1GQsH(M&0ylV$d8Y!H%
zGsO+-U7hJMAOqrm#fkDBp7KUG4|8w5g%UHaLHj)sPv=#7R-X>F0PzZarfSfV3UFy(
zPqPj!pDxh(IIGeZWs$Y!f{1EGZ5Lx_%mnWTe298VWkmj$r5Rd!yRyzq6wb}tk1PEA
ztWlIlD5G!b<P*h{!TDFH5S;Mc7S6YA=Va_#=u;sy49#MHH_+!yaVOYbsmB{g>XnEl
z+)~ns;{V!na0`WSqXn!Uw!01QxoW$^c}_(H9xXg=S+_5u(aBkQf)&7^0?-69O3(Ss
zuTUD&d}-lxv954v?wd8bjTrOLmx2m^t`?{2;ZZF@y6%t-$Z~^(vftjxf$mn(bBBjL
zaO%Ku4f~;gO41~amGcaJYk4wymaFe=wk!nILTW6OZow`ExFEcqOB;KDk;-POFY|4b
z;@qUY(h8fc;}PCet{V^L`odx>;+;4q%xzZ}_DD4`rTDNyfN9T)AmCRZK!{dEFuxeF
zOtZcbu}svi3YyR<tw=MqYy?dRsoq)BRrvsnE>-S-=nyoQB1fY7k|L=-^mXpx*I3^s
zXXXkOKb3G*-#BdBws3Pmd}K~ccfMZ_<9V;wc%Xe)d&_->xpo6j{*^!2*rYJn@fzJD
z<O-NyQgt;(KPJ#B;r!_VbUgaL&R4@XP$+q9JyX~x-b4G9gWn`o7_<u=)yJkMK8f0%
zdw^?yZhaa?KI@1kbQQOucF7-Bq*Rvf^7-xBc3Mc?;EHDdgR%o-i|h0C`qQg2S5uy-
zXohP;Cpl!lW^+l6uC!}2y4vEtPjtgt8LV_-O{q<f4d++w2JCpV&aJ|98Mmqwou|=v
zJT*nzAu6oA`h;w<MT6{xYF$PD5h*k#1PA+nHSWvm+vNML&c;&e9>$%5aF^tR<Z<?4
z<yS)_G<>pbd4?fMoaN)~8igoSSBYOopg`vr{oL!wwYdTc>}-(pl(%e0{`{$P_0>(E
zGlSN)Gs^_k8qaMFs1uc7@|VW<o3$=OKFwpTZ$^S#L+W#oKP?ZVtjUkae`?l<XI8F%
zZGgVTKC-Y|<t9I@oeewY_)9Ax#opBKEs=oGQuIQ#+C-kD3*|=MV2A09(5j(HL*M_#
zV*~o3|Bk@8>AZRTI`}3hB-RUUX4c+vKoYF*xK|hnwlly{rkly3q!TXqY1$tMdArAb
z)(L}dGEKl=ldLz_)9N_2L1YYOY&*>y&8yM1uQSGtF>jU`PqnqwHhJu-?I`lUuq_w}
zm!Y&B69GAw;Uppy0X3Ik3qUBBQG6$re~RcRzyI^=A9u1aD@CyJXZ)J4o?myr?Pk`k
zmI5R!-+u1m4b)0#4+{q2m0o`T>+*f~WyV@*N!pVC8M*xawoH((MR`Gz@}#xK49ZiI
zG>VJ0LFjlZD-WNfG$bsN&`3(as01|#`RT>0O^hh%7|e_@Vl#?9q)LpKA2Mb`f9f+X
z!{R+jM;5;f5Uju>7xE=j7Z~N@e@WZ{LgF$fpqv1;;zT_Sa@5tJ1lq9ToA{BBh4y(>
zpVc!uf!bZ6XVRu4^kfGr5=vAD$kz+bVx$V>4kl!3={yQ$G=&qtsUhXO2(ZF-;>7N(
zRI?6^UfRDT%>e1nuj))&fzG!Re>vrB1xju<JDCA5oY&FGep_ha=OwKs1Uh)L{A)>I
zp}=5>I#$Zrh*r~IWOVoGA8+6P@a-?8^yTfJaVdGK0HcCme&$P35NL1h@*`h+63II{
zra)l>7Y_mh9a7y^0pFG#f}<lfzJvaV&>+wwK({$X2NxG+^bJtyM|>5^f7E{TK<23t
zNb*!B;#I^__Qp%v&%5Gf{Pfqnly?1X*?nJjUy4SJ1s9iinGpv|e_I+Q2ULzCc&Akc
zS)!x?JnJMlvfLxkUfv1>VZB975|R~A*pJRrs8yL6(#g#r!o?UE#NRXgh#n?c2~rKz
z2op%eK(-g6d8(q_tl{?if7{Qp?~QhS?!GggI|W*xvCEwaP6U<mNs9f(Zzj&NoL6&n
zc+8QStc`p+`5IH&@~#O9k^HiBEM?j5ns8q(RtV*DLnn7$AuSXD9b>fQWNw-R%nJH-
zQ)v_|wf0qMDWo1(nzNL$bq(ZQ;7h5@t-|Qh`o$pGXxPG-LTbd8e`}8Wl0qTkXGjX-
zg<`rR3(S?kZcA`~$sfak_;Otx(*;n0VkkgQoL`peFVYPJ@?Q$hvu+r(?c}#f<E*EL
z{wnJ-Ug>Qex`{(}frpZw%RM|W4(1?f{h@G^<nB)@@h1B-O!Om#1E11xj%E)rVHV3-
z*_@@=f;ADo<K*usfA!z@bdH`50N;1KE6SX4#Jx~T7pf(wzGf)%5NR`&#2ZNu5SVMl
zc-0Y2l{=HB3Lm4USSHzUr!!kgu5KiG<*L!H_Lne}b<-}}S<GN#+h=V+KxjS{nncDo
zJ!wqbCR{i3Yejk0P_&+TfMWMg&HIuh&^!ox4}$J&O^ZC|e|#R5@GIi6NyFf3eu|>&
zVho8m6P<|1DpaFC7NXlBRD7(|GZk`2pUt`KtLAZ%&q@_4RSr&R=fib;q{9ho9=46E
zq&hxW+dE)vkKM{GxoOT0Q?Ibdn6HZ4C5L)sXbO)wKuS37FscIWLPiSO9ONrk$*GF-
zW<K1EqhdB}e+R`a3q3pbDr)SvwrZPJ8En-ytxn0c&EJ=CS?NFSx(0-+GOE1+q}>{4
zAf-MnL$&`ptSH|kPc-j5wJ>;p&PZG*I;$$JdhCqOB&9}moSU|wbA3}bX9FFQ8R{tu
zgu<QcB3MxD0C|rj!S1{Ke(0QxqqXGKxkxR?n;~|JfAYBD>TM;oEHT8iE`($*p?9)y
z-Z6QfZt<D-73Und^sI?w!jD;ZBbJdmx=yyHJ<E>NoX5E80+n7|7egZ|$u4^6s)j?w
z_plBkYyp<+A8Ok$NE2EvEWxBYrejOBvqPS?%Ic^qKG2aLG8?)}&f3~j>bB;Xy3@Xm
zg6wF*fAm2}t0z`~^PZ^z@RVnPqE0e|2}wDqQ`7+{Q<@cZ_A%9JkHpcT`O&o;dM{Ax
zZ4U3W^J6|d#j~d1X;-MUEeC<^)NjDI_Eu1vvSvFL6~;!k^~2uy``$3<SGL?k;pwHf
z%xWaQ)w{`SvdOWh4hD!_mr`~QxXj9QIl4vKe@bp!zmAH;+7rAKUwRn3nBqGdV=?Wi
zC`5C1ESAP#QaFo_PzX(WvlouV5GZ{TTRav+6xMm#UWZV0J|1WRZ>)e}Fjk?%V61I!
z6E7yvftZK?vC)`Kvs|jpwMm0rk9g&13{V&eR4I(}u1)Cw>1a%AK|$^&qcK!kM4ewY
ze;VV}eR4_3;?l8LM{cWhY=T`e79)AJzG5u4R)P*P6>qGBBc)zh&Z}7ok9(!L8Zxag
zw{tr%^wL;bq#RRat=eiA-ztMt;&a*VuB#HprlM~+x_0P2b3$rGg>-sW23%fI&&NR>
zg^<)`PZ&ApZB?S@Tb1r)H(QlzJEvQfe=aGomw3Y9sK(jo1vItl0o8Whv(FQf$BR1)
zttoV2eE`>IX7I_LxFZM*<?bhCzp3_5ZR8{Gt=cWkIe0x!bTYP6T?Fvnz9-0-*aI3*
zgeBX2t5sGzaciEIIlG|8C@UYj0mxxPvXkXlv~J~Q-N)KTKbMB{DK?#iNuE4Ke<VY+
z5?7dPv4PG{BK9=aQP3CekG)iVR4EWMoKs(=bro+N_PEWeNmI$(m4ah)a_s2+5WdnR
zDYvR@7_lF(^Bp;r-_t-W?(JqMD<kQr8?~HF>sPII4smA<r6?Pcv4y=h_g(9pCGV8U
zlb>4p`u61f+SeO^B3&Sxm|kyie@eun;8(ofke(yBTfW{fVO&1|zxs6rWvx&aX3QXu
zJdBSuu9{cttq1Yx@jD&g^UYpk07tpY{OZ>jCaNENjbXxe*=r2?`TEyT%8MlvJ%y8g
zuEA)8gViZaFScBVF<OydVKm#X%1~?v4J!@4;}soh%N$h>H`EQagMmWGe+_kGlvta~
zG1VBVwB*`B<cHpX0ZU#He4{O82zdsaO;dM|{@4{#N_>vv;nSHxn>MtIqnwqhuHn}R
zMLj1N3S~NMl@anKe`6f;LVg}P-)>dLhnPV6_A4IDX&%f89<~zHxoVbLaH_2ASR4Q6
zlYQScwXV!{_0#z9B}cryf3o=nE3LX`pxw-tA9Q6dY!~ruzuvEBykVTR^NcE$CGRXN
zltQ5?1x{s!k7LwdQdKB#DORW(RuvP*cV--gK`31DH$~MpR?u<OH?NlOsbJ#J%dO~c
zhf1z@W-ZEhhe_oOb**qT>}>h;$7`XeKLdD6D;P0;)Cxvy$4-N-e_&kw5i6MJKApCL
zNsAw_0;(yDHGQ&Fa(eXNH)pOU(J)S4a&z=wOQMS6nU(fT0{bu(Ae%#Xl*+C0phtne
zzRjV--nny+tlRT`XfKtean6zS*1zRZpXaGz@L2Bf2xm1O+)9gGdh62_tD_C${taVa
z#cT>?Ze(+Ga%Ev{m#efL2?9AZmoal56PIv?C&B^#w=;()o(Kfm>#10mflVhA0Wp^`
zy+I=lIX5*gOl59obZ8(lHJ8z-F(-f3THB5s#|?e=SJY#YhniyXRs;+IzC=!d1TZ8$
zBoXwW$b&?PuxqT=fb;h`?4I7;anD%0>y`p(0Zr9(UDz!0@Q^h!h)OFmh)dv;tWY1>
zZ-sn%FbX$AqZ1zY{6>rL)Ucwa`KT3(D1uMt{so?RaTh${R)TUFmO_oDb+UhDoGmRx
zF)e^?IX4*H3yJ1bh(S8?DK;agsl{k1V}~ASO>-GDf^)94(i^(?PSKKl5027l;3>B4
zohF$KElp7{qI2Y{D3vae69g?)!4}d>w@Cou(o+#=VU^IFAqu^WX%#End%mQ$!kofG
zcAgg&FD#sIYZ#^$*^0vEiY<Q><9w1`@!hZ#&#*{DzJ;a6ITto8)hC_he>}anK_2bX
zICRx=oMU-!xyP7K3VTe8YR+|xi|K_NCr6ZX-HUcj37{n9!PD8Q<Y7QjD!KLWQB+&5
z5g28W<3oyiICj92nwmiiGs1J-;l*e=`7w%h$+;tn@$@vfn5LXKGQ)pn`Bk<C7Kw(b
z4t+>kj^XXF1*{9Z_IwH(Ych?a;*`rp>#(~vvc<&(M~GGw*R(;5b(^OeT6sKyv-C6}
ziH@nk*diLnMNCt*4C<*Ycm=F97l!4)mRh#(R6A$O%9z)~=|e}D0gD=wr3AD)@+>8w
zPsnfB0t&@Dr6@QzESY~#VJhh5wvZ>J<#OpYnuZ*3o3PXcY%VOOue5}7$_~Q`AJZ0?
z3UR*p^2^<e|Ngk&egFDj``tH}7uWlX>-T909Zvpi_iX?E@~5je`}ebh=U0E-|M2$p
z*Ox!P%x<d5qrbxZE6l_wXc42HJjs>6eJL|qJlp;4@Bc_Au4#YN4pX`K>D{|mXGVFQ
z3^cUjo$ib*5b)jQ)$<=;zu6bg2?hIJU+=FjN}s!b{Q3I3=hv^V_l4#!p5m3T$I}T-
zU@W=IQ=AGZPJimaiYcva>fq#v`r#FJ-(Ow6dA`4Xx%>Xxr@I&XpRW%RpVnBcHRku%
z*zT)~i_2-DP6vMtzMPOVUt>Z}G+y28K8@Zz|LLFC+55}ei~p3}*Oym6?60O>+N<54
zc7NV|^OEMyv;CXvm)%=c7$|eAVGdlPDlk))R^9#8?yG|3pYOiAd~vz^w*0R5!K57?
z9I*3%7u)ZjJh`KS#|XhKD#7p<Ay`Jm+GmA;&5y=RIF^5OB4CR8jTDpVj2qK|Q=GD`
zv9_(2wlzM6Z5^4@k)izt+hSVT*1XU|ZHwjBwz?c_t1m{@ZE5(@7{A@@sP`%v_F6@=
zkQoj<`%zhRb>Ikt65^ekBV;=0!3I`G9Mu5-JVtGbSl<PG#M1C;vbMVSnDqAY;t!B`
z{Px|u_1b>{#golz+J4mNv*V4n-km`<dh3c9P0$5!BsLHta15})P7p$pMzq>SssciM
zA`u4HhDh#F9j3|#m(NgF)#O1Eq+4^lPhdG3H9J3>R+sVF1Qs4E=VyD2^GmyRe%1{D
zi$HY0;t9;!l5}*1CBeC%VDjXHg7bZtXHRG4A3*Xu1D3;o1OcrMdi=Q@rmR_lJc=dg
z9hM+;fNSYXgO|nv^0I)uY{4XkR8W(yT$6kwBm-5E`)~yfU!8lpE|N9GahyGmMyea|
z2!Ka}HWwL?giNq@P6&xWSaSuF5~L0I+yK}GoA*140I#lqtg}K85Z}S5ki#+p0)`CQ
zKozRhGD${%plwL91Qg346){02ww40U!!SGP3`&J`t8H;0NcHDk7by_ut=V}gdNT*J
z!-1qCY#>|NK#GA-fS|HSq1YYzI4?!(799}=o*CAVhj%#wB^mb#GqUEGJw`<$_qk9^
z^A#|eJ_eX9Zh=Y10dzDD4Qoqd!7N)a%XkA|0<S}VXj3o@Wn?;p(+7Fh<eEcIcmmUK
zHP;41VD!3g;EyEO^+pJ3gtY6|ga}f?$beH0uu*u@29D4v^Ql`Y#Nh~bRtlUE{A3^#
zrxw5_CE0-RX|KRMaLrKxG=-BX<hg)`)0*;;jRt-|_;X{1$g+VD+`;+biHi`@K!CG<
zMhG~6kYou_th<T;v9~!5inx`5y|SN`A&?~ljetuo7CxNy2$amuGo?=eGvFH}bFzqz
zWXB(aIisyG$H!wZXY^alv4!#Y!gzd1Yuz`<W|6Zo(pM9$G!R(Ss(RKWEBy9BC;51{
z66BMBIS`pjElbS?U#9|psbE@b0_Mnb8G3MkabR+H-arP{niWTmfDDFkC$(#qRt97^
zfpWqP0pH+M$Pcnd8~7C^0qbCW9iKR7B#c#}7w}A^B<pam29!lkDF9@o4$AGwuRQJ(
z?2ksx!5##n;^V^u0WMYbW4P;`*#8UPMTDg+IL{r;&RkD=>Uk4#C|n-9xC?YSS05jL
z>eaC>9Y=eNSI3=V)mWT)EY3WZ8r4`x#)u6F1l(HlnhNYKSxcHVFv(!mW+w=}n8TbL
z0ju=@2^ZPIw*=`!=tN#A?}K1G;P(j7HS`3B^==J$74mV|JV965wsS33s;TUrLQq~r
zQMEYXT0EyTy(wT0CmUm%x}A6UGyKbc6J?Utr=Nr8sjkabKx4-K-HAa&!DVcV>S5Nc
z@^TTIXWcb5lGiK9X)6Nzno2dmn@XlOJwio)QLzsCK%FDjw;_a#J8l2}Rp+=TiaZc2
zqOEFMdFS?UtO&YO+loW7kvD&b1}%-gG`2`K3}g<17(6AfKN<GXiy`-KcEU@4)wA|R
zgIJO4;GfVI4BM;PD2i~EN5u$$)Bqi%6&(5#Ro7MPba$>bMmi@l-!Wa_4PN^w0Mh|3
z2*M3S!7|Kjk#qIA_6U6-p$AGqA9;~&(2uMcw!&hx%Ydw^9hhdiP4S`Ytq=*=T^=U{
za3Cv6AEqP6qYhP3Bz<|k{SnZA;+|yUW7H^OeZ?C0x?){Kyg`e8n=j-o{s9sW6eI8O
z4~^v!-8|Z$cq-{0kU(J-bI}l)x@D&Br;CQ_h^T=xovuLyfpvqdm?BG*NEs9+3$o`_
z=U-Z6N^t_c@X!@emCeaxfXZ(n1ZcbGeJs3p0Fnt=RQ~V+Kx%rTMiEngP;U&nLl%i4
zTU<yRP|I&*U_mmK_mgLY(0~#mlIId<p<s(bNoJ8O2qoZFnO$c>mp?DTq4e=-^Z?!l
z&{`ybO@M2jl>kwU5I_v3(%A`$&|~QO(5-x&x|PpsgpX4)J)a!i!KR185%UU;4j1(g
zgroU7Z-%4g>OJ~c8Vj9&neQ?;xr|rmnhaK9^OQfv0>M3T5&dxtO-6F}MFu1dVAcJs
z48Lk>kJp2>sVu4zRYXgcR5Hs^jl{!s355}!LCiUSxYsk+>Dye%H-Ii@O*VfXU{FX*
zKT0^k7T)~(^^dUpO2?kY{XQh&Xw-ou5OS*{cRM#S$&`FHcdQ`~ZS9xi{(kx7$^QVP
zm#FBMp|l+n0Xdg(7Y-BwGM7=*1}B$=nkSimy&wMl^^X^sg_Xjaa2x*Sr_ZmK|Gmtj
zla|6LSpM_-OL&2x1US7Y@`70{fBbj(_2r*4*~*MyX+MAbwoEO*WbHY{+LOi*4cb$Z
zRf>x-UO;%2w5NE+N=k`5Lu4&bxaZjuH<HP1Iz1z62v3JsBZCgLoqD>YL;A?0On#<+
zbcN9r+CPK-^E0Ksk#e{{B|5<-;UBr5JOwEafn~f^?J}|mL+_bs5b7he>bKmY^5lZ3
z^-U2t^bhyy$7I-jU020wwk><AAt!(>R@`z|p+)xKn(|a6iw5}--0gme&<X^5gG4k)
zgju2oNLHK>j6mzEWLWG}B#0KnY|lY|KZw1iypPV2TR^r^#kGPvUh8tKl)F4akx<ev
zMw1}r%PfN=@L9B*^os1kN=w*!ZQ|>@g-|HStYk3fWiV%Ckjx<ckW4ZREiA)NCoj*%
zAo5xmgd!BOQ<2N2$fnV1;QDP;Lk_4wD~hM=((l>TgK>2-cx-po;UX7~zDxUmiE|oS
zdcC@^s8(}OqrA1Z=e^3gxwrgCx|YYx|N13FE>On5HjZoGbX^vn@@c<JaeEHhF^;FW
ztvBPgdUo7qQo?Hb8^vwnbif+(H;mhaPB>|w8Mn{5*n{FU0qY&zu#|$cLXC?t|4U7G
zGt)*m=U2?!nwdM7Y}UX?ZgGNtyfv_uEF*_2t7@14RvAIt2`^jl2as=h(I+3;OVKeT
znW}0O;_42ZI3?F{9ha6slao99mb?gb^di%ORq>=66Jj{;=|s6;lX~=*(yH?|6TA(e
z_Enw721-8X0m0{vwU^182INa;{lqv_dSS1jxM8g{1RB}=dZnS8>=Gq^JZBH^BrH4I
z#4P|J1T$Es@T_yeW|<-En`))hBi*~fcD1$Bt?Mlu<9^|*R(4CCzw#=w0!-WF`(5Ig
zIH!3WUNvBjMx@JA^d_A=BER>=k_Q#EK{C979If)GelDi@W90naB`!%Dd3?W51#L4h
z2p;GUiKg!R@@$iqi^s};><c+F4Kb8~>AY&<EUDN5x-rLD53{!O?DzFoB=X6^xfLtt
zvOh!HnXpSI(DdW#`q8}<v8B^-<aO-QG_&msNSv6tOMD>}0Yb_ZTQO_t)cIuW(eFOz
zEDR}H4LM5opO3EnJ*0Pet!dbbKZ)yJWJ$g;Sw4-861Au7&1H6fB?0d9q-zTnKuRb(
z*DD++vwp8yu@4hswc<QdV9g2rK%6kpiwjLRm*t%JYBXSsP)3h*wcN%67KO2D><cRy
zlB-77`ONa%kGN1Oq1`ozi}huhyn;mqMv?2#^2-I<9=U4B%llgx{*B(8@OnipVyG?c
zRb*4pEFQ16t7hnbZuLrUdd9-nQ1rrEH&ax2J4HaDL9ZzOxPzh-oG`8^(xzQBYGF~W
z?9x*HAJ^-hh&_5w@2xijCGsopzQNuG1!wn<)TMEPuS<1lwIIE+Wk67O)Cj;!!3k&8
z;;;cFkfC=|?Yx1gIZ2I3{;`GMNOG+GZbRFUww6t^Y~&GtEFRlTS{<T{+Ikxx<X$;~
zfphU_*C^B}xcf^H%E!bOZ1}IGh**ikK?|*1hiO@aTk0!!I288omWaI=r8y%JT`Unj
zMj{3Qh<1>Ou9S$*NVE;I&%<3tFQUV8?+0=j6KP}KTHYD6USr={78<iZ-S?|SI@`J8
zs(9D%X{PsoE$S+?X{Y`jv;Ih{WZ*H+&(anuQw|&Xg|Z`=h63Vjj|x7*#%wBolcPWv
zXo5jvek^T|MUR6to8zj+dhF97kT$-XMbWMtv37}B6_?cB1vAC_&rKWx9BOq7v8&2M
zu54`Mq?WtS*r*|rS~p28s$s+tz3d|(?5cI$n1(TbGFnq_lYV9dpXTJ7$0e<<u!_4v
zHp<nOENPF9Vo#Yy#uB$_IA_$C{5V^CL8F8xJ=-j&+21X4<P_gnKaZNQ-GuD3j&_j!
zt*A_EBRrZJmRTh%sHfmE?HuPz_JPQ3qWC~;Hc{R|XH)Md;WLhALVEvXd^T~t6Q51g
zo`TPR<{8|N&nAiwz-KeX`|;UK(VoR;bMMcZ?RM6YdE=GWuQTn~-(bz>8>Hii5zg9M
zZtwx;1#P&%eQw}+`J-P=1~w-HQVWo<TE6$ewg7_F{|9QrTT$B{E=vcAPr_vyk@I^s
zV~#b-zzJnM0Pko3PCN9#+wStV7b6>T=a7wm@+1s4?$CWt;k#nB?9$6}jD&7J%FBn=
zH65nu=&km<=#WuaD>OE2IvsMLs)r7yFU!|AU7DI(c9YGV99Qb40AHTcU4}lVmNrvH
zs0e=JARv!Ej{W<zyYYxR^6eP3$&o-|XrY~yyOTY`f!yNS{vV}xTT;-b?k<O}bwoRV
zeCznfJ`s9t!YyTZ_h@mpLv%>pq&J+Z%hiKA{2F6Tr<uxbopGjozH|2uwM-{MDf1sN
z;J06K6vyiE*alv!xTrgiN`2o~Qp@pECu-+hdYJ##$@8(DtOvK9G&P<<b+wI#<u<UN
z<7H8ddmKY{{&;F-Y_x*od)){bW6{%pFqXksI{oV)mf{o_JUtL%O%xvpu_nqp5NqoF
zB#5OX2(*uctP@30*Xs{GXL>8y?#N#i@tWaGVxkzo1%tc4|MFT;_Zi6+(fbW$S-TIk
zQcn)d%_i!}eCOX;_XEezT9&$HVfCaYDB4yP4_0RAdIO3pDE<<T4urD)ZjSjS+t^Du
zavJJ!)T})ANFxLcPy^anKOI|EydAlxf8H4W1XYwOdY7TJ9TNdLm%-#W6aqCfmvM_L
zCzpexCz*fe(?7re`X-53DWXeX(+B@`{eJt;8z&ut7v8wxUw^!%2Vj&iK@Tcwp?w_w
z{P*zN+uu0YhRm8_{PXAU1H1Eg;$Be5J?Wga6YeRAEQP_jC`@`(xTkQ(N=ga1W4M)1
zOTR5Xu>SP!WZ@aDa_KH!8ZTUGuZ8C}y<s{i*TR3-0tK<e!bf?l5br<Lcb+2Ze5o%c
zhv%#r`Cx^Tad0*WgA67<$aB$=k-G7huD>q(v4R(1JjbdG$2{KA{T)t$MDXR0m4$-I
z9i-sLEy{=bpslbfR*K-ONHR+iB>>4P6vOH~f>&lx2~oy6xs$Qp(L%tr#&WFM6gel1
zMf-m?V{90l6UymD46*DdX>Hvb*s8G;j1`)eAiydPEy*p>nQQ?3Xg#A0ElJi~6RrYV
zKoU#+lXo$>)9pfA#3yIc(US3H3&beAjU1wa*C9F(sFgE_U(1|yLEyz#8`X0SzP58J
zSb^7(oC;CQ4Py_Zq7gpAT1LenB*Z6-x^91FltCAS_IMzpjKWZ&)q$Kc+6e3IN=_M#
zQPnIlN?j|XlBFhFAAM=^aNV!#=vABBl+DXb?mY0OCKz-^e&$Vc44V*t&oi26#a!qZ
z>{8%q6mE$)C>)2}Q6Ov!b18@@MPc96q=o`4H0R~IP_IiEPHxV+%C;Q1(z_ReRbhYH
zk+JO?kc+*M!*t|^tHUP@2`fsv20=?06k+*HSB05kVR|<uY+abV(51SN8Jf-1-AGDQ
zE=1RimI`kgl6EEl6mF#v2)tPzh1*1^A=&7T>zM+mibRQ(C0?nbAbBr2tetw{rF_<s
z_NRNAB43Vv>HCFpx5rU2r+$|7Ax?kN6xZ{9dkUy`q<~FRKWT%BPv@owqZWn{Lfc?y
zy7sy>i*Z>@+#Q%^uGeQt<mp`n=@VhN1%=r~@xb+KpEZ<Mr=KAAlN|8$6o8YwHUnKp
zcX@aD`)m*Ub6A^l8|Qv;4$H9qgdBDW!|ORLgSIk<wMD82L?R18pz_Z`k)eO{b)ZjE
z9sPz-6zTmMhWue+sD_|Vx?p1n3N-}j%^^troDgIUrcDsWbd4$i3{&Mr&zr)-c7FG4
z<zZ2Whn&p!oX*{dwRAm;1TAh>QrnYCaE=T+qe03;J({a_B7`xse#+40D3Q!<T8_|n
z$`SpYc_~3ZDJAR-C!|bxs)&Cr0A^`tcDJO<ciKJK0>Buq*VLf&vl`R}jXJ1?geP<n
zeXB(F{&HPCX{QJZ6%f-oTtA{>eSK+xqNQTwvl=a=STk1FOg73Kp>j=DnX?`f*(6`1
zQdX4ZY39DL)NzYXslT!*kuT52pHt(u){dHzwIWoLpFs1=urn-rD?fjUUHs^@k|-ff
ztQxhoDy$78(Xyhj#)F%R{#jA@dG?fx!jBUuFaaO>4GGk<<cL9>CmNgU7d0tjkXikr
zt@eQFmEx<-Qo?gCim$GH8DLjJ3JPh3cqBik5KF?norl&469b1fs{y%K_p2?Xs{Nfu
z9B@iRO!7DeGU(#C*X@7Y2NCsr=T%BOXiUW#A35YyIGUegSJA3v5MH{wv~H*sDxEBx
z0a9)`vsYedHDi&y?PihIN`RZOXgrXwV+%SqZ8V3HqHjv%wnpq?s4tv9$S5Vo%R3tR
z>)Tl4`+L}M;2}DdFtyj44$;pGrreKlEU_kW>wJRs!l86Il>L7N+!|eGkY@$lHP-yh
zr!g!$D(2Ofafr(ma}`qiaVBe{_Vcew#%o~|yeeOc+8oN)e)=k(803THn~L}_<HDZa
zk(4<L<Pg6emU`IgBjj5pgYmr^AJ{72n2L{K2Iab&zQ!j8`B?c<bY@8Qh?gQbhj>T4
z6qVWPBg9(;V`YCo6_Fi(T$uuq>P;a$>c6Su`fnPXh)FVY`d9YfYJZ$TFeUTiiJmdB
zpW~Qz`==dk-fgVs>lb?W4C_2?UtM9Y-FOc1wF<M5PFU+XwC5FOs8zvQ<?ZT>eGPp}
z3z<5x&v@lc;Y|LMAx`e}5R+}dOy-HeQSM97Gj+CyUhjWA&}2XjbI#1L`7zDeu!nov
z4(NAM>4syu9k-9<tF&L-$JcTD&{0I=e&#+*D#I*p<9ggSK;zMDh9x&#?4rq`s}<;s
zW^`SS3k5m|@U2Zy%U<ALKLgYp0+p{Hk~jcz%|4SOaz_HDV~fVb4qetd$PbTU{Me<v
zL3W!1OGSSS;-plx{nV_T7GA^u`?0yTeg3xt0~*bi^<Y;_oxW?3F$b}%3=1QMKWSJ1
z@%Am-7sgeqI2+}9@3_D#?FgOxLR5Vj3)9-#bTH#wO?RGAj_TG>n_8`X;O7kkXy5uq
zwQFsOf{AAM2tuH{=gxG<w3>>h2FInn;ihfss-%CmwZ!_CMBi}f`<?BMu3QMbQ+nxW
zcYH;~oQ(@gvyVf}`wCsZGeQ2mM_p+L<-Ph&7W4NUHITc>sH{uHLH@YeFa#8|N7Y@>
z0zCiY@u@{L9Zw)X#wH?gTETN#flGLKJ}a$^wJ~(58b&le*3?u#mtAG%vtp-4m~LYP
z_N0H+$0&u?GON^=r!aZv8P$1ZlVbigolt9Qo;Y7ivtdS?`!GJc0f<|pllIayLDrUF
zdoQ=!`h6u1<-O^y$=B94>1A3fpMW^q@VFkw25wDd%<%$;TD-~RP}yV;Q7APYDkn|u
zDs2n~(XyTC=oYwK?yiwk`o|c^{;oD^Z7+Yra#W*sDM4kjqq~ObWgNLRQk`U1*5B3R
zwoBqx-63?ov<;NBwOzoU^URSo5Tw<kD|fMVqYei`U(^fZb8Wh&h!6H@e2Ydwe(?Fw
z&D1DJCZlrTPJs29_1oj<uCuC41+mn7_pw*r8bmX(&0tIylohSW?=^EOzZR#dPVavs
zGvd+&Riks}mI_odxfc8~ZJ7=|!A|o4nHP1kX<Sc<10E8Mod~Bh+(lt)iw8T;g*}~<
z=$#&1XvZYlY|Po!8IMU!W`XXrmNo<zrddR9eJoC2I&Uz`a%zlkGgkI~7E*P#AH-`Z
z`6Sl&*-<SA^#_bshb<fkgU1;*Y=K{%H>5-RFPODZ(%`xJ$XN>p@xWON2KmXX1zY`#
zW-Wpe)*d-(F$Jh!n6gN595`h={?VqEH>LphbPZ`rg8}^70k9|x188^87kZS*WK#8O
z0qml+vQKmd{{_R~BA1t;v>g)yIW?EDO)3+Yk^U?wmtc4~1ed?9CzgM8JpA>`pC4o)
zR*GQb$M`m1y?%N4{b42@gA+#M^7ZFK`~r*;#_PpvVZ-vr-<NL>pJ$*Im828-56b0_
z@5{9ETajImBs*y>;vhRENuaP;^uiF}B0KmfB@(YdJOjx9q72l;qjiKqnMFlh3GuK(
zh(Ek}wTTsIufZ9sCM|zq$k-4YX23#;A1J8MoEJf_ph?m_9h9i)q>HCB6!~^|v@&WN
zX@_~u8m9kJ7A*=iOZ$^J8&-bnaI^ft2{4XvC##YL?=jO6@gpswEo(YoCk`u3+D$us
zrZ3W(PT6veMMm*+Tz`$$dro%QqLUOtltQHP5K*R&s=chN0(yTgNFz}%aaBK+71e}_
z^&2KQ?vJ0<$H`=_5UeK8AV0Up3UI!f$VGuN?NJ3;)tVJTMy>F6idfYe1*%Xf0+t42
zPI7@FwSfwNN)HwT@l8DgPmXscrN&*h$W<5RU#W09ewRKJ3MIp06$q?8#vY`S2i*@`
z8bMsIpyLMgtzdt6DMO$gywupvn3L%q9<@t`pA4Q?l4%F_)Dju%H!;pYX9NvG@=KFd
zK(U1(#ds7FHzhu<=+#;a?ZZs3hH`3~ORa-Y;OA30YE-(4fx&ArG@ac%I!>l9n{M52
znu%dAujk6o<dV_cwNJ6EWDx>E8JHmgNg_dB(=^8V4nBYCcDn2>Zh<(^@lxqOCtm;*
z7WI|#1=QsD`n-Iji#I(dN_}y4j@BheHd|qykh9ct<t@^)vSLI5zIh3D_E%2EipGRk
zp_ppP37$^)N=V3|<e>Far#TE;nge($A*hD|jpihQybI3TwsArQGjnlU-aD%=L(#;z
z6{f~z!pDDhq?mrTqa+>H#!LJbSy!%J*>2xk(P@S<glm@Bj3Y~qzd^S4=VI}B@A-<T
z`BIktk$FM3#(|y|sTk@(*_5=lZ4Hb-U2T13)BEe2b2gzIX^564Y**@A1&V*t%}|uu
z<}9Z105umpAJ7*aVNTYTr4?5Ump3m}ZdIpwKihvnj7TVxE~XgEX*o&lna`Sa>?*5l
zYu%NtCrp8Bo5XvoXmFz?HTt(~a-yNL$^6Y64a)ut#YdwxN4?=n4OhhmJ3X@wkgv>K
zUb-<C%{-j9H?{#q^@gdXT=izve5K-O@%W_uQGN*0uAsgi405zNvUg{$eJK~YFuS&I
z1XO=iCC_am1mUR*`xSj71nTuP{NlcG!g^=lI6>RdHy)?Hu@(6HEJzB7QszHN2E$5M
z8*68Ub8f}+To2p8$|IJtRQr`A)ooo;u_mNZD-7Gwx7FI)Ix<oV%DgM7-rif2;0O)Y
zc)qK3QvZLg6WExaTJLmx*WL-xQ}ypw(BFTncLJ?M@TOYFloO1J6;;c2ymF~#EW4mW
z*wk75R_A($`lqsTf{q=z$$;!@Z`;-1kSjUH+2c!;rjz?C)`N0mc2i(-Ve$yxmr-Rc
zIrX~HV$?ZtmMbQ_YfQPdYQ*mdNX+KYux^%2Kn^LpVz5ky66QA^s@C$kG#1qyk`R9?
z4>|SuKKXb1@ep(?-9EgbUp!T7&YCULq?|_uUo}nRyHWwWLY0!@^K0WhNeTQeDJ7FR
zoBWv;3qo}z?2-9p?h_@ho2zw`t0uu{=%YMiIaZ{GCCL`q#~q2s;~?8Tns4~JQAL!U
zZ2DySd=Gij<4GrH&KnfQ?4HKq!$^O1wy7}`^i4%_;7lL8XS!O2gS_WrIH<NpW!6fi
z^bX3#D(m-!;^n6$1x<>hCU{)_v!rFJU+5o2RcFBw2w~8A8)Nswhkt$l_NT9ZqkOo0
z|99L<r5fdA75wtETpF?{jc)l-uB{-p0?|MfC7BfEKnn6e8~9*d__{oRrj~!&pzl_H
z#MMA;FKFEzP6eWi8yP{j#56>RZMES0(Jxl2DvCsOB_m!%DCKOtr1N|#UdC5{&0Fcz
zKbD7Y%fsg)RkSu$hu&){{9>u-0k)*7NJlPKF-@|(q;0H36tGl+BTG23+Vk7C3Q{Fd
zO{2YVl(1u^TCi1~d8Ol<mk56rL*S)K+O&Q`4im4m(wYv6i&S-ia3mKpQB|ayCEPxL
z{aMa^r&FK0_YCJzX(IsbLLpI$Z0)K3r_j3<?#k}6nIkWIsVVs*<KPmb_pR|fA^W^^
zv}C3p8&{uqOF}7uxJMFmJJ3l$QDfk$1Wp8e<{)qGCGgUk^%D2Hktcs=f*PAjBj1LA
ztH=+->XhA9kuXPA*T!01tPSm4${eNMh=U3$a34X#cEq0s9t9#7dyTd(2h?E$$K1Za
zauvoe%yn)v*Hst;#3_#`#8!@+Gv{hKTjtzMGn?~2mvLLkH;-My*>T%j`xt~<VIK0h
zp%%{B-h8nOwkKI+@Ya7uzfD4u4#Jp|>o}{gP{-`;hVz$NfXO4cI@-$n^;X2O+RSv1
zvT$LGo@Mbo*^Q2>Mb_PA7gN^T>`dk0aVuG^ZxfAoLa!b0kj_QW^MHr4xv<jzk_nHt
zmMHO8On5F$ZzhQAS<BTk5=9V|)idO2M$+a!Bt56*-DYd%^Adk27o@(|LepiDUX&+*
z2>71j902a<e6N<SH$u5EA$rFFlnRQ<AU`7*!oV((Xc_A>JBx&scgKiK>F8V|TvKCA
z$y2%+?<GQwr<m;~fb#8rLaCl|pk`6tHp<h~%SvTd;=J}rldCPxs344<xtN~asGBMG
zqMQ97_OS6KU&en?20CGpSTaV0pxTU{=kwQl2S<|;BH6+%lketp@g6g*px!p=)*6M=
zD`=M|-RAZBc{Ll&h)cmFi)$cyL9Z3WkGDf~abToZ5VsACC-AfMuD5ILQPAr>Wkbau
z!h4D)7~ul!6;1GhoP6a@@?C<Ni78X9)YQk_6LsCL8x?;L$MJeYt^6F<{J=(EgUaS?
zS4fvm3Wc~17L&f5$zdV=Nf3vX0iqYyuG~%H7Q{pm_7HjHj20l~KDBZ`xi2Myaw0Y^
z8=<EK-R}0#6`aQ$3UGn*_Q+3smUd;S6)#tCvWDhi<y`f`DTVToH1#o)?g5)!)f}gh
z3Cf(9?x25@tE>q_?tAx@S3q6vipE=#1Q%)}q$zIA)z`RL!)-B`bKySJX$H6wnr?U1
zR^>PmE8TnC_ta7erB2ktr4HLHL5B=3*KKVa0^D*B^wNXDlqC2*xNpOl3f<rTP;jgZ
zDcGqH!>Db#OV92V+@<oW6-;@W|CJvJu6JB{|NMU~<^)DP-`O3jZ*e?-*FD|Br=t|r
zhKq?$Ji{lR<r6RRiD&u5oqXPMcUi&8x7&Dec!F^r`N>wpcEe8F4R7k$s0?|IJaTkT
z{h-#5W7&R`3F9yaPmC|y?Y;dsp0;oL4dc3FcXxUixw+l9F8dpLU1Hk8wlN&Mb$I2Z
zFmiwOqZUgkos3}5?CNZu+cCRRTE_b-wR82WTkE@0v#~_DnxRc2;vp9>ei<lXzU!om
zX&E7(nsiZ1?8ChI^q`9?_Wf;YVU*ec2;tu6v0_0EGqAls<GEgxr5+m5-O-DwfFgjI
z0p8Jz#cqMZ{{-+vD^`lLaWe<!Yp~{OYf^uy+UQcQFS;g9%-Nc2vIDkGjT{Gf1Usy3
zSf_W?Be#;@){OQx-HgEcE39MvNz-$CX1e_-hod$Btf$8Po98`+RA&Wt55?c-NF%WS
zxg(9}l)c-LCV=-l(ge_*b)<3g?T$35^nDy@;^e#|O?GhH%H~YV^iFyGzdeKYB?^~A
zlPD1bG&wexaXmg1m#WbyD1TbbZsfQTzV}n;1<YZQ#XrC>*6wi+klQTg7UU2NGTy^D
z$ZigK{Zx@fiK3+baksq)5CgJW6j>}5i&Y;#N)IoC9=`0JzU@98EDk|M2f+;Aehfe_
zH2nunh29MTKzR$px7XpX_gWiI*Arc+wY78w`W@-+9_}nessSh$Jb&B;1S+xV!P|Rj
z<LJIGzYYIBNPGWqhuT152|+IV@V9S&*=bR>*USf0)PL2CH9^u0NxJh@o9QXglLNX8
zU7@~rR!V%ixIU6zNP1x$Ydwmj+{aj2dmQU=OBY4Tq12!*OIxV5%;!>zx+WWrZM|p$
zA{lEt?t$*E?bq##?tl4>FV;qb_hojnoJV<^YJ2MLFj`_SwPM>p_c`h5M0R5b_h5KN
zSsp&hSP#<9RoQiX`3+kFw>6*56UaOS>47c47W`cNKqb$vz_%~m)AP68?+XXP1S%L9
zzQ6AH1zJ!xnt@zPSr5Y>{}_JkzV>3p*|gHsru=hy_~T{hN`HQp)sr->9__tzv(*Ed
zmJ|-&fkI=dkK)2IWG$F^`KLpKRf(u+9lBm3Pb$wNR+}BfR>*qRe_8!m$*;|N@?L>A
zy=0@;S0Al9AE2*o7kW?ud7`FK!n5df+N3A#X%f}MTiSCj(LG&mKTHa6q1%@ITu|QG
zMZYWdv(K075`V3Nkb3EL1@*ebpwQyd2Gqa4xNp*4K?*1AKvoab6zl*z@mb=mWF|4H
zp#4Ikm8ga#s@0iq;`Z8T4SJQ*LK=JNn)NwWdY$vv<?~>oQ*<rn0Xth<B9SEFiHW^P
zV6rNtr9`)GycCtjg=3M`8$;!k*QOqx5F=^PNJ8Os7Jn`_C<W}2^iMd~1rj+8gys0x
zL5JnAXX3(n^4N`}{_K!mTu4%RF4L1!UZ5cwr#wZkQ)3RcET8-@OPrOo5-rjaT@1pN
zHLusF9(mm?BbMIFfE@Qz|K;vyC2e%7Q5+)qvYT8^P*RQLv6`%+p>ZV&mO()Y*{N9j
zXL+|75q}Eu%x$S2HD4~BOUB_8XtE;#4HrWjH|=vlzo@U4?vl?g*GqRR(yz~~jvPF|
z4HIi85Cv=7X<j+n<_Pj6{#UWf`5B^Suz6>1O%Sx+g6W05Pk!*UccJmjySgSl_^9aW
z1zwRJf>Dm}T_<oOI|Q#n2>qOYSaKkjbE&uHHGjE*A)Q_fej_zFtE}_A+q`RL&`2fe
z)POHY3}94(-hFan*l#BW$xmIQ{moIobB3VGEq76!#9N_n*-3TMx6n5M5Cp%_rA6PD
z=T?UVl&l^54*#RMl@f)D0pBaPLLz^6XXVy6NT+>EqKHtX?{BI*ra=3cxRAc4`x>?J
z>3`tjka+V>TdNhu51pq}lBT3o`8>J%hdcHtb{XF|)V?ksu#e9Y;VNau1QKQf=!7FV
zp9gS?lw=%>b~J$rlj}Bglu%WbmMSz<CcoIlQG&>7-8>?cg&?}RiTLACk+$i)aNbB;
zQp~gzVnet!^yFyGeI@nU@l5kT({n$~#eb7onTukN9NpWVYL*&%<93fSBRQZhW(18_
z7;QfznHpxQI2@f%M61B1zID-KnztD#X~Zhk5=}P6!%4f%YqM?(-!dkTsMW<i6itNk
zoCnai6O`6Qiu|D`C<9QQ{uqAJrbL>=8)nu{;tl!9PVxzUva9_;{G<&Oo81NGJb!1i
zbJFCqPT#osfj~LS^ukVWeoE@Rd;?@aAtjl4ch_uyWb@VuykY}n2qJ{o3*6WMQ2;a6
z^wzuF*$omHM7W_{g!R0gT^y)|pKO`Mxl9>#%R;x?qbXHf#VJ|O<Ajz|N+~NQj%p6`
z^R&7?49dEYeQ8dMvSnk;!@A`&D}UVt{stCgynG=>Zn%S0pz$`;eIw*eds(HeSwr6X
zFrwHyFxsxYz72%7=#&n2Qwg`Oggc_dIu<?-G%X`}osOR|*VExVFXIU-11-33PBqIJ
zt}%0<StVv4I106ZPXLxW$%)nGNRn3}#=cc|F23pG0w%@Jr}N_i$}(HipMQp9d%lj?
z+GPozD>!RTg6sMDV*|3%Bq=A<`PuP0gooxl9i}tuy4offolno%h6K`j2jF;2D~lo8
z?*qBbmm&KOxsy%N4gVQ(u>#{Taq0#Lo}J9nAA9=l)4yJR{PO+pl*@;ge{)sBc@#f^
zHgK=y(V-%x2=M)sXHT*X^?waCVb;@@J}Q&=8;sg;Ecia`&;-H?E;gk<xs*0|k<VwR
z#u;6*D~ug9MO6ouIF{D_ezt{GtEr1Trcn)kVky#=AE|w9#gBZo&$TMG`fb?#7<OO9
zs?M9hRiZ9}lc&Ems}h!g&=}}-DzcP$LPB4yk^=e4Mrk|VvCt?VJ%7>;$M9h!AmcLW
zZ5(qgX~tcPNLH0li|V~E->^eBRw|j!l9i*BN1+lz-bH{|w^(@k{ry$iy-=$!&ArH>
zFlrTT`-eLdX-KsBAz69%;Nogbk?OT}jgI5jH6zNbJ=J*IuCPuBb@_E@R4QY;n+Esw
zY=ZzON08vOL3+H>7=OZ9a=+0I8xgq<!BfBB&yC3BT5+Y7p1!!@_-Vj4bbvfz{dJ7O
zFksqKwHoW%O{?GB4sVT>0-TC8O`X<8@JU5XZhv&_Ch|7|W-)fj2}N`BSu$rss+2eL
zeU0R%1+kKMbaw`){4(0lPnoOz7VNmKE4?&YF3Pw^Bv9>8Xn%sn7FBOka%nF)V@o@Y
zsh19VPEk~UE6zDf<(n!{#JV3d+_7~ydi>floR(BshW{Mq)g)csHTkjzkRwzhS%>f+
zY*LM3Ml~Ks;WwrjA!!xPB^V9(abDxe)#;BAjAn+B%<Ox*X)T;NPHA5WMl&NA%?ac;
z*|oG|DxiMfSbx?`|3&3gGTN$Q;7TgC`X1^E31<+UlU>&c=OxLR_mU#J5}oE6qI2|%
zKrmqXcHs=aJnC)E5D@ac)DI&-=Lq^;NX|qXIhX!KjuQy)g1sTdITv_ChI20X1i?A4
z{6YLCFwX_%g9*;A*DD#$PTIRuoLzr6;5d7U*OQ#R#9-qrXK%r`)0{W7jqsc|v`ry8
z&oqrNo&5sQ?PpD?E;bTd7D8KcWck&vdphZ7CZAJr&1z5JE-a???PM%tO2U)>Cu6VX
zL4gpmN%-fIv7jkwTmRl<Y=0*>@n7DQj}w=nv>g)yIX9QF5CjvK^A#y5m+)^n1eZ0#
zCzF3X{rdd#fhS?#oRys%LjbPVB-I8P$&)3ma-?^AI?_k}5QRqC1lqvh&Drr)!rC$G
z^hg+J+se^uBC2Q$ILFWEKc78S8rO;Py2t>IIW9)ZL?BYqk+OnB<+6&j0u~q~`Mf;N
ztHXGjv!_EZn-vsxj%g2NH1k%%uZh4jP}YCK#2M1dN+qX%NEOS`k6t-etjcICGh8lY
zyo%wV`gH^VBr~|6?(qxM&he$_J3r^0!ds69SpoMZTyXLdsS|jE4S2aEl_?s{GpqFs
z*hRf%8oOaAQ8Tz7!D5BiP?SzFr|rn0?9+2FdG{GxC5k=G5rI_%Z9y&l;^hi*nnr)#
zImtXnu4I44I}GtQVMvG2gGO)fhi~tm4}TxV13|RuSmE^P+rj^UQZPwqlTTq%i|L==
zr_YBEBiRHJt)w;akK$2T-N?t}Hpyog#<f<evv)!IR5)om;YO-rk;&%rF8$6y4B83W
zUs6O)4p@y81L9Ea6VC}N^m67uLHvJigv77@O*5fkqzt^1tDa<J+K6$sT*yHKWHp?2
zDc&WSu0whiPp?{G?!c2!Gelm<Nq9zqYd7-o2F!b|FF13BqLoY-H&XOSsWk-8EpjS>
zRgudh7mx#C7>LClf~=m)V>)j)TA?yblbcdblRj**!l$r$G0Q57oUY6wKJtH^=Xbx2
z(wRk}M~Z`u-bXRo@j!7%HtwTnnFF<;`1a)#leLT#)rM^uhUD&s*~srl+eZRi=nurE
zR{-wg`$9k7vV6qVTx2WPDWasaLlNa(=|Bm#gn3IaMQ%#D%Yuc&fKj+`C?F<4VT{)_
z<PIUFQ2z8>HCQ8~D++%C#cqG3rASGRGK+nC0i}09utqsZrdrB*<bunPQZp&-f|IF7
z_YV2DT!}<v7R_=$t9q2^nBcIOJ+C_+I@}J6-+9|A8&vP{Ug2AFd{A04N0~)7E}P89
z+~`v!lg^G*c=nP?PW4ZauN!I4ZHkhLC{mQ_qRUf@fxHsEWR4CldhCB_V6lMm-~BuN
zPZZ&ynW54Q6=&30B5Ptaq1R^xiZVOF_sp^eiZQ3nb`w3=X3Tzt-hd7b*4*CJJUO_g
zc{5M$5q$w?PRGlZ>_F;-po(Y)DRj*LlO|2AZC=Dz5x#ChXBj}rBi8LK@w7%{Yg<I=
z7jUL%3EC|<clNu&S-pRrvmy5?s&mSuu`#d`9!J%#)hgE-14N9WE{<hrd+bu)1NmEx
zBc~mUQ=hIG2_uKG<c3&suN0fXU~L$SzlpY(M=V8ro+nESBL4ziLuFb1U*BoNqc5q$
z=wq5!q%P70_afv|&P!|@6Va(r@jI$qQkW#u^`-CM6M_4F3J8CbvMJK2>AxuqGXV;n
zFNlON+MAW6e_s8EH-CNk{L`o3s6v{){OhMu{Y)kyC{BG#Ux9J!L-E)29Vz=X(pOcE
z90(Z$nK^QrkQbJLPtyUxP<}urR!6`3QOD3CK(jbh?Z!{c44R=VgQr+W6Th$RV3De@
zh?~e__-~#{TI+wm()zs0fAz2Sd!9<G{+JG*r^APYs?uP>so{~Z)AVl*RY5h4Ay?2K
zn}<MC1V@1x6=%pUHb&e1)~QTeO*uS*&&&mSy(DgB&v~TTcOD{InL-{aq7CC4I1GEG
zk(%%@<$Rh+t{=RKiiB=NxPJchEv@@PtKK*7k<L+2N{oL>Pe&A%SxR__R%pslWxDF6
z;`BT_`$Lf$Ri^mF%qe@N_Jj>Yh~<Z=5h?ENSU30KW<Z2siZpi}kTE%~KuAZrr4Jd1
z4QHCB=h@j^zdNDJ?sAR}DF8z*Gj#ZB_YbQh^&N!ZFnP^<gA=WtQl4h$*4@PpMERx@
zFH!5m5)6NB=3U(6+&Iy>?Lr2(jf3kKt-G#eo_`m8Ubv#@2GTV}SEWTo|2g&3l0!Z=
z9yy!ng-Z;nH05J%^Af|=OAL2X_;5qyK{LXfN^zw&;Km&F0=!t%q?)wUq;*gVs%zPZ
zl*nXl9YJZ(7jx-UR7SU=?`}DaCg;j=A>U3{6l8xVoNDTU`o@(3?NX<>ANH-+k61kz
z0TbQkK3HUKG0VEn`uC`dhenq*=V(fuCZo!FkqsitwTp|FR;uJuZpxZkisRM6V0#%e
zEXDV_tt{*jfFUo@pj=9r=0rR$t<XMn7iEo8i`Q(e6y+SL?y{uS;$w3T`_gXBL2jJf
z=yrdLQ(NqmBl)VPygW(ONoHc!8F%_r`(HH|E-mTpQxwvI1+N<_feTIWTy#F=S}C<6
zb5SPSRs^&~`h3wR0hOPMbU#ST&)q2~d3&&YOr`N$cU1hW=<@k=dS7~dC7fg2>GAr?
zNv0`j{}H!WV10hQXJ2KYxZ~~BK>5n;)i8hi2i;zQrjkql=-aD->zx5@yxkb6-EwGO
z)e{4B4ET7{v}gAas2&2hr;&FAOUN{_!xRD{wB>#pUxbdXb9?tK36LY!ZV<+M%8*3L
z_1hB4PC&x2sAz8othr)T!|w@pPjP|vN}$cC7j)+}2Yq96%|9KaK($ZZl&2sd-*JC6
zse(xsFt^X5^vxu5F_GvKobR1XbX=E>sWw@%qm+~=)db9rspvh;v&Az@y00R0_`RAg
z_KmZJE|nnfT2ue0;iS68c1U+<Y)QMzMd1-qHKRGfZr6J0)2y~iZ4-;s%8c}MYEV|n
z(Zcg@ftRdZdW=tgg>=~4S)0Zt8V!H`X{5uy76q*i;>=u>l+q>D-$|BKE+nZ{w};0j
zZL-~}$EG^(XjLio4OWu)%o$_-kea5dbRSmrE*&$qA#{WF95LueKAx26PK3(OMa|pI
z-u?t+jo#lz-Z+!@l=z}6@vCpRSZZybABs8F)LW#ww~HkMC@PJ|#myD%rVD?Tn)#Mx
zzi^{Jv|pP*QfL$-*S*>V+%hQ_)cYblHw-Y@jmSRW(zA(cJ_2*wrOcT74xBHzl2t4U
z_nYjjE1>VZB-za>xqrI$)m85%v>BSx?SRGzyUky6EUR}>uDyC)bWXZn#A|M!OdCkg
z70^a)#Q^R1>AoQcrW_2)lDbI&0BT+=lTR;C<*N8ddYN<EX+t9!IyW$VX}zwINmx8>
zN}05E>W7r;*0-YfEM$6uI~MCb;VZRzKl!FoeHG;4opks=_=Mf`m!Y&B69GAwQHVDa
zm%U>uDVI*oC!Bxg^zY|?9wZAZMX>Q}{AlkUpAX*;t>{(-0jP)1-wyEsY9+LX!Ajvs
z*B`$RUk;yIs+E_dCHbF`!;hDt8Tpls7bF=^T5HVAcuJB*aj-TB9Uo=m;V)7eQWi;R
zBqLxnf@+ksjyNbisK_fZo{bRW4^Lh#az&b}P0JN%n@oSm+>9AHqhU$>OoB?yc@dza
zCTa6&p)^1%U0j{1$fwiM%DCNz=<CN!Q(&}L`Ioerjr=<D15g5OX!$8@?6ZZU5e~Xt
zP`}(jF^J%NN6{`?5Ht$V(0PGm+Z11tb*E<|X{VR{3lIdIt&TET-pu`J%~!w}V$|72
zTi#Kdci4ZZyvrKccX-F+cbLxc*)!xhUaX(d!}u@hHyfGQwVoH_g{Fg8?Mjy%eJpY2
z`^eR};&VDo<aR6<O*N1FY}NO3NJ?vI9xd<vV~ZX}C`mqyPM~&0k9;C>SS-7L3w3P;
zk{7nr%|6uW)EE^y>UM$pQs}&tP>i2Pr*P8duIztUOd{@X?tf`N4eufEZ;UoM-NZY5
zNKvee@SHTC{WtIbDbsd^OPYBexujVSg-$w6w%%d6OEdyUz^8mAS?@E&;y+u{6?!e4
zQEj>kT38J4lCF%k6jt6jTg{o?n5^c+uE<pnH|451yB$(hL-6)Y)ew9=Q8oK7Z;IuH
zGK_!o>Te&*4WjS>H6A|f8q1B=bog~geV<rv07Q`3QD2JXmm|o<@F6KIdHde+W6m`O
zkXiPy^9+JD-Y*9dA2WBS%QcbYc2rl+Y-7g_sE%=AL*Xw*b#xeogo=iu2hLAXACb!m
zC{LjC-sHjgi3comwB-1}s32*m^F2&6J!^l998ZufL?T~lty2<q#b8myVz8UQ5}H|f
zfCOm>gjmavB!Q2r&`HJdBuO!XFSA#{PJWsKZIYOstCuMRVGOh~MH<=o?5Jfc*o=J8
z)-)xRs7Mbj)e63*($QUlZ9Z$0b#rqN_?iX~=y*^#Y&Ez}QZ0jUwp)#l8NenAXNG@)
z8O&zj=&BYmcmzI1U_604O51dKU5-*WItIXA9%Cs^_CB%=L9J>VJcZ(JUt(8>tbLa|
zKgs5Gr(C~p95Qnd?)Q&@bPA^R>3!MD;_I~_Fed=EboFocXuMBab>{~aCBRxLm6D()
zhx0^6KLz-s1y4tHw+uFJ0sERNM6iEdZ_0zP?}PO*HphbAkgP8y;)Y_ME;qw$MR-(M
zXeonSsR58YR<*PBQrhOErXHtB`zG+CgA>}B9qrjrtN>$!{V)iHoopaCgRn}`IAJL)
zwG`i2wFM&rcGNC|t+$kHT;nvnZmHR{7wBC_@rI&};af{_rD{Vf0)>Xw^X7kWYM$FR
zN#m#*bnb%p)Y;w$y4u-)G4#~l3RHjph9X&1ps5fVVb>y5TND_4M}40nlwb@!2_5xo
z8;X=a+w2LRNGcPv=l6joejL(lB0#Xk4F4O_uuxzKSyD7p%GpR((_a+&@ZrywFaP-b
z8-=&w<-c(%rC<VIij44$Z%Kb0YXc5n`JO6iWxx*+Q0NE)QF!HPLRnY`J`V>37qOY2
zjDC%yfg%P1Octla@8ZP5U@}x>h!n?B=J$0vSf#2gvd~40co(Vg+IUOr%c^)Aujzf6
zN~``e9KH;PPi$3V!6keuGvQ$AZ^^2Pj-JzMYS}UbNNE+wIoK+}(P@8L8J+4oP8HT$
zN*xHEAictvs+e2Vxr|KoE<;4CP$)xHv}Sz84vnrfGUiBO0tIm;Ty#*yw$1h(zkmM5
zt6pf$$H`}=a#Wz{$lHMcDMi-o><QNyzgV5|^fJ4|--32s&BBG#()ntL5X(=)M5L_Q
zv8eXxYGY71mRet7jy8X-jAkH6Ck*;(hLysZr<d8q4Cj;5P+l|m>1&2s6I4spmcMH<
zMX9HiTYT|wW!q`?LV+(}S<chSy}YXMr7AiXt|N{5{B`uTLBfVFa1}PLrz%V>Rbfn5
z%~9@xcTIM3$-72h$^P}nFrUUb;5bPF<y?58;9{}bcT#|=Ws-lbq8OhWN=XBmn?z-w
zDPJGa;vKh+xPGnHKeM%?Rnbo|mSo$f-BbGhku7?j_zqs)tyG7;)`_S(X+q=USufoo
z0JG8^a&32LqDx*?IxXW}@o^qm=^_Exrd7G2!!dmeVxUz^ubB4%dh@c7*O|ACHCh3&
zcP;-2rUCUdAXtB<h126O8otUg=|&b=va!tA4r0>*7mTr7)p~;_-NvPjFf)5xaN}`n
z*X*q1VtXZSn62o#^Yt7-{Dkq?soA3gmg-vy6>~At&ywQLlw4a<T-Eqe9t1#3^sRj!
z6&p|lmzqA0qYO&e;yoHYAhjTW*s(FuP~5T6(@?(9=xKjuzf+?}c}0bVd*`=Y4c8lA
zlQ-0U`g`(*-wrRzTZ*^8DQ_v7S6`L47ToUhrP;HiZMY${qiu{5y}PWLhJRAl9*`Y=
z?q-p&o9-xYcrnVotcJpRs@SFbq#T`b>qomu8A7rjA1ioja5o=kVC;tXJZ?suUS3hw
zcNcQ3PHKNBu-iLTlkVywY<XMnYp4E~GC{<~z`~C(<Wuc~QL!lX(&K#^SqgnT4Qn!(
zWf{yXGAzFme!Y-U(hCf*v5>jwFguqf@s<DhLr&ly&tdW}<T_=%+ft{{&;;&g8q+!j
zYu|gRPGO!m)+uPB=8cm23LCR<Z>&$i2|~MbuJKyfVy-|gR43VRu{y#0dy19ngk|LF
zB;U2w$<=yHHfY?gQk^tgwQjei+Jm4Bf$Mr2e`1wEnBBOB-!&0<EsmzOdiukYs|pjU
z?1Z8$I2WW1u4CKf|GL}&te2s*9TNgMIhS#b7!;RW)+fe)bNc)1Uk^MlmNRd{*YH8_
zF0Y4Q4|M5<J0_j(zI;D~AG8uoIf*-B7B}?UKfAAo&s1tZ7sny#5Ag1{w~n;@oWb)P
z!Q;jN48Ri{5!IbBo~iIt0#E<MIpAhBZdxM<I)ESsn|ye2Vp5{!cnm-#B7z(wN<z4F
zl$@k79oAWY{>0@ET5$QjWMh~xl1i3*u&%zh)y!Byd#lkpumRPC<1WP|uW1?5MLe}-
zTG<VFB5D@J3)P7*QJi(q^79JJhqFrDEci4VK%H4bUWuDQ&M_}X(gI2_h2G2QXm~Im
zz#i(d`CO$GOdC(<6Pn<&MItaON2efux<jI8-nx;0M3_WSD$S&lBLuhWywsARZ`u<Z
zZJ0+FvLiMi*jQcWpaDP=8XE|2O%JlQRnsdiS4>YD%{Uqhm5fRc>s*(#hFNQ=1ci|V
z&RO|+QhczDNtxtUqM0@A(Hz296@EJf3khAr@Zxd!Id2<OSO3Iq4_aC;SUij}Oru8j
zaC7c|eS|$*qkJiw&$gPqKnQ5j`k!(;<R!;YxNf^}3BmH3XGsAx%J;30m;-97L|jC?
zMtZviNnnDb-$8Ey!uOFXda*<e_)Gdk^*jp6)zr^WwfCMy2+hFEt^uak09V#9#TFi3
zz;%v+w?<8ID6qgJhS(B7T$(E$3eTFH4+DXJKWd}HYJyT<PSQMPWkJOfJhfj}BxGXR
zJ}w2RASv!`i2(u_{Om!LB}kbpDb8?`(GXo=fnjd6%*+R2QZ-Az81{VWD|0{(>a0{R
zVR;{=XP85lyV%F|za=~s+=6FXtIJY~`9EbhBf%lBL<f$SQ$;sNt(9twuvXxh21~Jj
zR&mXBz4!Ud=qr=kEIyMs@M%TR$e2BMbaBN;Y^CMeDhxd0zevjgi3C^xpFm*0xg#wn
z5_hEKMDhYHC%r$2mV=PYY)W(Xpyg!NJ83zQ_U^Qt%(nwArxNd{<y7KzT22wXotE#2
zO{V2LVv}h3N@y}Irv|bgEvJ&V({dtVpOACve=Wq&kh2R-$N0gFH}f?88X=BU-KVQ?
zv9QL_ZzuALqA{A)tPQ!0cEY?L(bd-DY1X`jFS1=+(=xNg08YcV@@P_8hVcDshaEq4
zu{R1(-a+?A7n_-opndS_LRb^TD*PAra(MXV?dy+U{()}3d;2|fMGk-_CA{mtr&~bX
ze|83kf6MonlPT>&Xz-{}c>Inj<*-8;*f)IX4q93aBS^k#>$lMAnRZ&rW^iO4Hgqf#
zG)-j<L1N!p=KIz*mQ~d$;-30tgu5Umj}5mtzKjaD;cD+oR~+@P?(nrce9owXkv8&4
zo&<Y?|E5sIAR_{uYLG2W$}tSbI8#Q&e_Pc2nbvl{eIxn{L<Yf<j#*T%a4vbae$S<)
z*>`C|Sw%u=Dxb7{1_!cNZ0VI0{jJAwm~Ua~i7F~Gx(VU>{mb_}?hQwMYCe-3G!lh`
z+;m5cF-e%nRwJ{-Om8Yz%H2!v;!H(i5MX=$QD&O5Fd_t}{M<E5N_#uj&3#@Ce~1ta
z@Z^>Op%JVBf@6x}Z4bmu(pdH&F>t~#<@Gswh|^J;lCJ6J43W#Vljoe2PA5u%=sK&Y
ze0aW|$~sIgxTd3GP%vdRA!U`urf6Q;LKWv+?rJ;CgqJ`MrFA*6jY;B5E1_?7{S0k7
zeQpv=IY8<<u`RO5ezNp3gKcrge>C|T&Kd)(N0Lo(o=$9LGsjQ^vVuy%)Z-|fZZx3o
zp_URV%6-CJ6YqRdg_ifdUFs2cXtKc(&CGf{W=yj>rxX){4qQ?>%u+(cqtoSTi5yl-
zVy1+l`H>bD``Ux8<;1u_llV(i6F4Pm(I{nAp;8|iS6f=}^Y4bN<Eg4xf9|nRixETe
zIpy7|ADQxQ)l((!_U~@kEwhkgLqm#RzEXix#-k#dZ525AoD~W(w&#j%I)_kcbGfJz
z(Kf?g$>sIEOrxmhVnjkcT^jYagAw~w@D%1O!u_uo@al1#(#&&5l73w}5^?QFp5?@m
zq!I5|1Vqk*yhqdHTGYb)e<a?dqYY2!n=rGgz~>dizGUKC>w7Yk3Bl6(o<nlyKcu{;
zke51VcP#G_i943}h~$Ox9_jr-<vqpG)f)f7<vlX%o#j0uZMnQB?<?<-`F1GpQHl4L
z_o&3{<vohv?d824vB~AV9kEH}y_L}9@*Xvi{mOe(^7ir`HKV(te@SsKvAkExcQ;k7
zl4p@f;<(uVVf&h0#!BE*O&Lo6Qq-KU5~kwjr2&Uf$k0`WjrjB;Tl1WL<Foi|WtBr?
zh{f0~9WY#U8wX~Kotw7UKO&6CeaIGPUO9Mp2@e9$Pg!L4gP?=~#E1a|fBwr4i3V#u
zJw#$xI~?+-AGRv*e{8_YC0L0;n){k6L)F<5nC0Y45Yd!ESyuD4<Dia9sPjwsCX0_K
ziSyH%@paVQWoT(IRmQA$3kIXT2!8usMBPh7bZHSV60+uXkL&4{L@PBl*;FOXKDU`p
zTb-#1y0Cqo(bPfG%Gajg%^c^F5s`F%uZ%@1h*cuG<B<KyfApg7T6e5w^R-ifK}{xW
z*@c~H!ZwawYWR4n%-j??RwYZ?F$0F~`POVHBmz>=WJ!C5S?EHxv~SiVIpEZ4$E=A;
zEXM@-r5l@UvDjt2WPTG))-qRH`YOi^Hbp9PmCC0BF}`4d#4e>GZOaN4MmC~+Euf@=
zMKsq{tju{ff5!TjxAbP1f$*AX%oawLxIIJ7r${7{0$=T|ne(*seTHAG{xqlu-(~oH
zl_-jQs_-_?qnDURyT_HxV=#m;k6jo@GGXBqCZF|D{$(o*8NA=JDR!?O33DU2$NTg+
zJI?8(o;C+7(F2+(q8vhDFpbD|Z3<T9(Lqm?{qW6&e;UkAWuv#NHPz`}&2Lw0qSJ}a
zN662c$aM?bS`}}HP;!0COS9pri{$RzO7{Dd)`dX7C*=QGZGBt=pXQjeue&0L($bC4
zTRy#OXQF&?e<P_;LUt&(vNR+DN{;>~nR3zXend^%qjlBN{0`)7C~(J$_E7LbL3`Nw
zgX-B9SrPc)Vs<;_ot5lX)Vsf>+D^7Z4SOK&{sQ(u-1X}90JYmo*L&fS>(+bWk&4!9
z;K&v0feG-Pk|oFR8mjs_hrZ;qLJb>`V+;=QJmtwLsjjaf{09fjKQNcUe<%|HFqdH&
z1QeIy>pv)$x!@<6e|b*-`uN9_%*;yRP52r<=6}6?JpK1+X5A<$P~zd=zdVH>NP$oe
z59Eadb@=r^ho7GQIa4ir)jhR3v_HT8e3(Xl$RIjk5T!9%4<LeMWa(gx7b^UeKm<Ox
z_u=#OY-E5az(5R(2L5C?h~d#X!9?M=mprc1I7NWdRfO>{f5!hM8a91K8azk8@w~jg
z(2q329FzC(=?Pb%nWk1@&MB}Km-94@X^ZevWGPR@a1ftvOum-`Dc=rJm<i06l|c^T
zfp{Jh4=cl6SKyGZkOk6cCludI22ROL)i%nJ-Q7n_M^hQOFF9g);M*Y9-@g+awD8ty
zH_8sMK*0dPe_#LhU0BLXXnhF^#%5gdm+wX;x-HO`rCAG*{$R)oQnp-*)=g%^+2w0O
zO}!;UAm7tx_$tq*XF&eCb)9bO66*3Te8XeD{45YbKW?vYt4@0{Szw@yAnSHvus93I
z&z}X_St6phn9vI%S8Qt)&Fj8_rCBJTs-3~`yk@pcfAI(HU-p<)DkzQ1zxc*4IX=oU
z8!cN_L|w;9&E_+jyrxsYsQE8trY_vLI&xKE2Yw-;BTzf`7*Y^FP8H!;j5eiNV1>>K
zOE_b-#c*G;cFJTE75F#Nk}><EO<BJ<c7FmL&CC>wwAaCEsRg^X*}?KAJb2x|3dt+d
zA1xEAf5S-et6pC7tJ+%L%9nG{w=7ui4M>CunG3d!B!@YArWm~ec4GIQk)pt7`4ib6
zMtX~3i~xf(WgKl#ooz`N3-fHX;3KKG<Ws_1=IJ)1H}TY3SO+`sWYjE^=XeUjMse3+
z<cAHIkGrC}sjBP9IpJ|ZEgHc!^r|0ECUW<^e<Fe)mh)*ppb4Kn6qyn(ZefZS_EGd4
zCAbAey^0_z5~y&2;C`7$LN<EvNNmUm&nD!7*k}+-ak&Hy&(d<KAhK6Iowcp0UKzQf
zdT2;UR+gD^=Y?*VF0@8iYZq$t5D$~Qi%F!4#aRRod*Y{TX^4bF{4*okn-OWH*!QYG
ze{x3DL$EO;8kRghBhrY%OZr1+M4E5Mlm5MDL@ZxUXm5isEi)prD8SX!SiPhiL^Fem
zVh$1%$mD#3Zf7MqcZxHqC?q5rRp)6kmA2B3vxoH^BE>8gnuw>QOV_hL(n?t0{ER;-
zUG`%0O+a1XocIi*dN_SHGbWi2QodZCe=QWq=AbQOkSRc>+i3O{&=%U|*NCXEbIWO6
z<+HnwJg3VWGmRa%XBrvzZoNJs94}#CxiOhX$TKL(YpkWl88ydt6HQ-l05OjxWZn8H
zvt(9YT$aQ{ckn#r-U&F{t+lvyUTfCgAszB+P_G%mD4i)EpJXKG-#yyC^I+%pe?9C6
zu+AkaHpQxsf+Dq-`HTslT$MPd%0rNpb^ZqoA5|Lw=QtgoS*9Wyx|KR2z<4<yy^;4L
zX%0ipyi)^U<+w$y&h=vSP$uaFu8$GvLg8jGd>0Bdy9`{~qZqih{)VFEb(a-wRj6c$
zEf#gIz_|3sra=*e*G^td@jYc`f4W@O23YEFTvZ^F4e|^`33y#TNZNUe_W-#U+(aAu
zoi>4+xTa3i8$KgqrUM=_E|fxSb1Z8EHOxhb5a}v4tY5P}()TsTad*mgW)(S*t5>gl
z$j@q6vx=|Z>m|~Pikjc*C779GFIDMb9+Ka_P=O@RL5!VG67j^;jSGoWf23&hgL}>s
z=BMo8wlH1`FqGpEn{&g7(dfpKdwgjH=32D%v(wHaR|L!!0gG@h1gx!J6R>(&JFcJ#
zljC&6Y`uXy;pF^GSX9%dgK}0oT|)}EE=uSG$C5kGu|<-!y)GJ{1aa&!ZhXO7D?@_Q
z<-TP-Y@aKD!4x3iBt2=Qe`Fs+-YdrdWKBUcJxrkl0u)KYyr=Kl9J<b^#e9C=vH_%0
z<EXPbADy|R-elBy#nW~zk#^4DksDvhq4cLteB-cR;T2a;GyK}Sv%byKg{3!ig(X}l
zgYzx^|CuYz=8DMAj6zW%&J?{M#2~LLra+M}TD8(>d21qn@}=?Df8e;;CZRcYtWqb7
zwoj%urEdyQIIQWL>Z&}^dOCUWmb1Ha@rH{-^<`a_!emokJKZUPP)_f<9bB$BsS-<;
z88Aked8Fvcbrf=3>&@&jVsKWePr8%1T9cw$HMHgYvK43xXTASPBg;s=oO}*P9CRL~
z(X8=yMwWv{YDf;2f9O5N>8rL-avEY<(v}e?JEnfk8(a-JQ{28L8|I=lMmE`AwZ`Y{
zZClgzhP`bq<!y4%pQy`XXtkAm?ri3eQ*3M+@5)@<J(<gxRW-SLR8mSR<T^Cpo4Sel
zrc9K@m}9#I^;#5aiT&<wG*~N?nw!yH2}57gjpjf@@U=X!e;Ykfd|)?vqI{zpJ<a~2
zZZs;#+<$d9dfN5QZuCTLTQ~aAID!%6)K}v;?s_kdBee}Us#NBu^Cvp{scSx;Kz_d0
z9HhY6nc`jj@YG?SDSo@V!#-2I)DVx|^I5bW&=1#MSnU^-_w>V2XvCR<+7L9z94hur
z?oTwgQEKLZfBNP&$R?!A;AUu`b|v2$6u<3nt}aAje@hQM32j<SPqp?N-QO?-$F8|0
z^Od?@q5he-6Cu2&N;s4O)<P^-14ZikOOO*_*(JfB3_#Zs^40L|V%f!laB^{?7BpQ=
z{LlsCgY#;7OMSaY;hci~oNm?d__HqthEMHbj>W<&e+*uKNh}5!-Y+o^jKyY(ld!~s
z-NqPsk5!f=NW;AQ!WZ54*OemY|BU!d@N@1tyMc+=2JbwI?jA&xM4_a*fk-PHvOECJ
zv$r{e{k&-<lwN>a3p;9^#}Jc)U!ZuO69NdYytHftvw5`-GOwI%+15%_zd?5twcriL
zeYRNMTE2E3ixh723)0}(Rw|q}?P}%^do7%upq<awCBfUw$z5kv(s3#dGZVGG5M5WP
z;TETM9_kl*=<Y7+O1EFd?^y2WrZ)Uot=V8(J+$)c{6BKBoBAKAs1nkb!G9<d0x&X{
zQFIR!0XUaI&nhUFZt5qKe_i(Y*W2$;JS~<pr~O-h8$Z3iJ^lMME?p^P(z@yI?@#^<
ziGrbpNq}Yw^z`GO>Fd)Uqu8R1(=+i!@#)96X&CvE&~u#7<60}#p(i*&DkrTSL;qDo
zPrh)jXt$DfEeQqbQ4rlOKV7U?lqfmbL5&g>Mkz|AJEuEoP$u2|e>l+H#qYi5u!4jw
zlYT$>C2X@Bp>5_B0GCWEF|I)L>#wsdGhC0e`6;fMpOb!tQ+)~gyOBnJ!Bkm>WUQ>8
zb|})W1Fp6~L3kvN2hAf7(<~kFQ4sM_uVe+6uz`f5=bwu_r0HC|3|!o@Zp8$5la_`_
zG-chvN9e-GC8K82f7lL@CCT?GeV5NdGNt#Vb0<*ghQ06zyFJs6A~^`YZYMiR1_;CY
zgj(i0S_(8~M2FVEFWWUM!C(YN=tYy3XS8JW-SANHvB&MM955|$N5c;ie?c4@R&Xbd
z4YPNMW6ILWYUTefj*&=aK>dU`rj%nw-y{9<f&J;C>|~j8e@5d{9z%dIF3YMRF*hK5
zPCQF^h0l;s*FpEaEJ<;)rF2<3;!(=y`-~>(2LvcHkKpY~oX(?Zg~2-(IgyxP{1cev
z=<>-_rUY5K-CdrQeJZRpd9zfF#wt%$erAN^=Oi<{FERr!+fVY;cq=GNnHYH{T^2Jx
zW}{z99GM}Hf5tHIoEee7hx!numhCD57}%Ng_I&dBtwf)%`g~Tpg}vqxo&52Vu#W3!
zzqQfE;mY1@TI7EMf|<LNA~Wqu9P&X8A@0h1-%f}MBZ7|)+FC^Eusnd&nfJtL0AEoA
zjb8IPkBq#IpUeEjz4WPgRpUo8JsX}e&-F}^m!jjGe<MH>9YlUE+9qj$X#c7`+aJKK
zI@`*#H$AbeOto0_`fP~-4exZ7)<%ppTes1!-)rKzx9^NK<bMW>XA~b0X7qtj6<+lL
zk?k42&Cw=4Bht!DCuPxB$_!i+btBi3?sNBiie{M$X^u=BGdFg)!pc=z-KCT*3OA#*
zSi_Nie;;w3aa7>ROFA8qKR4$_*iYaTBy>fxh`w%pI;kx{i5V>i9VVX*@^td#FxsAH
zZJeuo>t`MhIfXBord~Ryi`MLku8W!H<gKnf6EnL0{HhY<IqB3hDY9K3qzpJjOKj)1
z@g}b{N`qu=aB^)_s95GxH}AW}B>S_&Haa+2fAp=%1(L4Y>{{f4^!HNNR(6|?-Abli
zBws4~HrcYp5A0F+Q$U3rJ63UR5ZB}4vpbtSBL^5Ux@)F*NLo(IAt2KwFQ!@gxi3K4
z)I4o6r#+)<Uz#MzmUOS#uqhjkyM-IO&85oLh8S|kB$bK3upuTvF{rJI9SX@oVh%io
zf9~5WaG^C58b-;$7YjsTB-2KWvvIbLSTu5<R=w_4m`jcfw=`r3G(*yf1;XnIO1qIS
zD<JQb0<H=hbKER*FbWHG^%8WKT>Yy3K}4#=<)-e&9w-i4AB1s`wgN|N@ly%1ieDbN
zfgb?r)NY1f2z0BQIJemb3AiGQ<EQPNe;=cnF=o_ut*b|~IEopv=l%Id*_47ke<c_s
zhC^jLUP6&;Iy*`m0^u58sG(Ib$IVP4CcBmc@Y>;bvqv{b-QWY8+q0vpm*s`|hL&<O
zbDSP(9HO*L)izmZL%vi4@7V4~$_%vJ^Ryl$hI}cfvo<I2wY~Z%QbI9pYjbEwe?g(6
zTqXrSH5uNyCbXJ|oXq9YC6B=8nhBOxbt+YEYrNmGtyI0K<xT?+bKA}>X^aW>&?oj+
z2L7SGD;Px&R1<v8m10WIWtTR`Pq{j?c_bv#a%Gjhre6!nFlF#v{aRs``n55EAFAF|
z6xiD53MR0*w_tLm-XgNxFKIihe|>jE+$%4ZVqjC?oLjw3y{3oa$M*fN{7b^tB1B7u
zAZIlQ9kP@Md_)Q2a}HO3cUhI~-OrK6KsRVnO~)bGaaGt%rN|(y?>=;4;@ossFtBj1
zECo0)GUASh!60$R(ru7@r*s=;Un||NV!I)GNE!2>@!D^XR8kt4A&*7>fAAaDOOqY%
zo1qHrC{N$f^u55@@BX^!!ZyEkUCrqB`=wi;W~k%`*MfuA2W!DW+S;1>-D#E}o}z2g
z+uajU!I4R0KH4h1)<><<OZ(6YeZ+5jg+565$rbvb?aF@)YNbx))r=&uP=7b69WW(o
zp|byfsSJ_qg9+WLgmjVxQTD#{|0pSq?kkM8$b@m7xchWQUeljA3i-Lyd&}^41%ltK
z!P}|bKX*1L)t1rKTW++{7P1q?0QyZ_=|v^K=ny|ICE$4-!<|y`{{W$0*EN^He<%|J
zFfuup+VLmGe>RLiE?*9yV8ISihSF`{o-c>rhb?p*vW)|opKk}5z~F4i-nh)YH}Aj9
z%i-(RYmY_2%#Zh1v$1?r^0H&{AV#7Uc@CISG%*e~$W$edn7D-FF@l#gp^vDIh$)3n
zvQjKU9y~ASc=wtT;DGf(P+_ht4D;bUz+W+9vKUgse~1#D3xZr2<U*ko%B`Y)m^A`R
zg|;5aSL~FkN3MG9ITB`h;Wkyfti@FAf&fl6@*a6Qv=sMDbLEhR>O>qC$62$Q7FpaE
z)fH+8mdoFw)Puz>?5+diV-Yj%NYKV3#mY+2foej$5@`>M`K)&T5woLf$K|Z#qU72h
zg@To3e?hJf*<wyxfG*!8x9=in{Z&>wDwHtW411o$jmQ7wB$6XLggw(-Is8W_kpS(G
z+#Rvp{+!(V$YNa8DP%n@QX7tePyBIBP4M^53`JQjUkr@8%C-F{`*MF%rX<$vq3te~
z>B>VaYkV1JolUo}B6ieHeb0GZDXe22X8%R6e;_l^?vJ?0J!ythNBpEXQnMB>S$Ru~
zeG)Tuq{|uXNJ!T;7F$(1e_o|4w@?V(+o{w*RB=^jiY$VDn-`UmN<X(BC_zD9-jc2n
zMHjYCin^{_RH5YeQ8m<0-{&+Jm~*85`&Qx%{cWj?pxP=kH9}t(DjwfFw7WjuP;$Z3
zHq|xfeM=KHR;z^aFlXMg>pIo_t9Tvn%$;{(a@UOK+sL9KJL~OI85;zHTVGYJ=O5BP
zfF(<pQNc750yQv~;om(Jm;3Z5Dt}3H<hl*M`&a1bIhc<m04|B+D%)dERc@L1mb^n=
zDwE1#Qk9(M=LZxQaci}@?aZOOEr|d@;KM4*<)4dOzTC+4`}B76%PYA>fgoe8S>C=a
zil0dSR?-S*XfaX?ko@TT@|SxlwUH8^_=8ds@W=Asj(_Rl4$$!m#ZMGH{D1oP$Hd$S
z?GWvZIgu4;v1U%1KZ-vlGAD|g;d8nl?wQ0MaA8KIte$a4?}at0aR1DdEGkPa0a#q8
z?h{`Vu4K4MyM^7uvC!sqQLb`jlN#JAi4>Zg(~A_y9|r`>e_6%j-LGG6ij*FOHocHR
zxK)6(0q9N7x>7;ZIsf{ko_|%~d<^-Nj;SM+e9nSv42KVQqu>b%StyVoq#vYGDs$G9
z6l0Q{E-R~66#79?&vBi6<tN^rG{?p-o265CHk(g7y~fXKp2rO)_*iHYkXC8Kll(V5
zs31o)v8;IJCnWQe6?A&l*Z?xezngzBtzPAG$!>hD^L0MgYD4B7cYlG6YCRF9Q-zf>
zQmM}g@QM~MywL_5z<VW>^$VgCz)O-WtHuJyOl@E>!dr0EP&+TkI;K{_MdI5ZpAv<c
zQZRH^lvETtQCON&c!5D%6~))Q2QDbQQ_wiEmMBOnrU2m(BFVoqiJqxJ^Fa_=HQav7
za=DpYnoq)JU=kyoj(>Q+m6^m&ckxbGXWCu<%BI?t4LFP%qgtvTyBmSzrG&#q>eF8E
zBSpf%fQ1kdVq^~99RHTjEuuXLB)y=vK~V+KCQ<|+;5e{r051jkmP0YX9DLdn1B1oY
zhBJYSB>y^QfO?nX2B@hH8lc-K8^BDR5fr1@cjD<!ciP(9KY#D)g5xsd*Ufb?KNWK7
zcG9++?W>O?f{;DQAEJC}e>1Q`1t+d4fA{H5854hOzWRO)qAkz{h|&nr6m)^Z3DB0=
zU6ll%LLst-;Dgl!r=-18E6X8*gKNF#8BlX9=G1SQB%IzJv5In<gD|$coCURVj1+V&
z)jp_|*FvElHh<a&b%@`JLETU>s6qKNpw?_`=04ZO93NV--x=Fmc`Y5#)-@aI+Rid1
zQ{B~X*pwxa1`A4jDmql+WE1~f@1^y*cqnw3pOv&81LPfg3Yhc{K=-Af$W$(s7P*8@
zGUtTfov}hLW{jUYQ7XuCK~xn&yFShF*SOBGf7{LeEPu1`Ay05b^(Iz6w-38GCH7lx
zFy=U&BaXOlAnIYNaep5j`iah0f7=g6fE;v94>mUMH1Ke>2fkE$rA{*`Ps*xRO^{Js
zMJl%;!N7f7xDuukI$JI?;9~PmBU~26?zr8!Pi3&n^pp$r=Wqw_I{|Ik=Xtm?nId4l
z;#MOKyMHrrLVC)+;?lk6tcW8Dd*`{IN;{g$@W4-ZEkruizUS}bOmW8y4IIvSjSdxN
zfTj{0CJuQGa9_^#N3+Y1Pe)o`uv}K*hUVP0y>3dy#~eh>*5OoFrQcfb95Gy?E#dHC
zw~FI3#B&i)OWGtnnp2^03VSM+XTLK*&aS#u%YW_fW3BSrez8ePXx)kV>wfwoSPQ_;
z1|c<&3_yj)hZV3<yL8`m(I+fGY4Va2in<?TO-v4S7Tnn}6PI*dS#FMss<&O|4rT5`
z!&*E(%IPWQX?z-u4R<VV>%DVAOnrUWiNlfVL&DI32RWWoBF=M%60##lcR;5bU1N{P
zsedXGVn;PU5pp`#x=e$b=}EBzBt1Sh9?GSKNu@<8DQaz-f{l!##}jsbOtyNiP`2UF
z=dfR$-VL#&*JpfEN)~$^MxN_P(@l^RQ65#r$3q213G5{VzP3(~Ypl5XG8m`Yji562
zJEnsql?WN62gD%#Ao&zloJowa^^kEjg@4`BLRT;LgF%Hx$WXzxoK9D%)v$mZM^@y3
zKOHcGM4_>Cf$fljZNsvrx#cRZ&2#2hfV;y+Iy5rY+=ihl>el$M#gc7K7<P90_|DZ7
zbM-P;8i+WP9VVe`!7><Wn30aN&g;k0zq~p%&1R14W2pztz(sn`a5|-b0It40_J6yO
z+#36BM*6Tu)ZpkwK^-Ms(Va;VNL91nr@0-*33k?aiC<}I9Ouz`?`1%}Ly3UfWITo%
zj({^K)s`(nn`<4#R?IIY5|^y@xJ}*Hp-5bp&tCFW#CGehPLQ`;-_AS)L9qEZ583g{
zd4T#qcr90jrhP}%)GAozu;|aHB!5%x(!3P4I`ha<TYR)fawrC?hEspAM{<`x(e2zL
z%Nq)==}ll9e!g^9FK;*h89;APfX3zT-*3Wwgpl<#XB)?*GwbFb%eR|9^<48?Ovd4o
zzDPzrHg~?JH=JJb0i`k24(Svhkh?1_wP0zA+ofQfArcO&Rj0DmFe68l!hZnu94CQ1
zstSXg`D+!wxl_L@G)5?HZ92|c&U3Y8lLdgZt7ln1%v>d(QoSUZ&a3pCK9!R~Ie&pZ
zS!<TrD_B(6Xx+8B^Yy&WyIq4b^o8mQ+wyYlgyx*^df*ncL^JJiN;XoZq8nZ=wyV}^
zef!bvr?At{B^*guJy;>T=YLfkxh=MC;Ao~A)WFpo=m7=t<8ek#XsdrV&e~nC47%I=
z-U@`_Ea&kS_FP}#t#?YvlU^CG^iSpO$wk&SEQxWwMCx^jHWzVqtY1(ASDa(Wk>Jmu
zm3$gbkK5O}Z*!-Pn;9!;1t_RXtI%6S&9QG}9j;y+HH$LqTD}a=DStoL^XH~|?rd2(
z`1J>P0sA|6`q^I4?)p()&~iQK1=M)Km2sF$sxY$aBshi7FcTb<fb#Nf$qP$<IQYx;
zj+WiGI|yCR(Hdd7@$Y5)yzjd0>TAv7L0CTk<5cSLPH=Da0d@fy-aQRx*6LkB5ES>g
zTAi^?S*<~k9W?o#Ab+Uho31(t(&a&Lqd5`oRJHP>!=2ZHl>5oiuHE#5f?dn>V5~EH
z86U67_9!H4q7&wXZ`<$^MR$uc2v+^ey1Jue+s)AV&W@q8%Xfy}{xk1LBL)F2s+Y9r
z=X&X~3-^r63&Ts|!b#~t3+Z|lyrU-oFap5!46knq0K@-z^;)nw5tac_IJ7;+E(a?R
z2q4qDqkLab4xP2~2ajT<0E?%Na(PJRvY5y>5!rl5L_SfVZrrD_J1BBn*{XCpfjKur
zQikCT3wWXQHsQDF)4pkVdAs=^<_>n8m%)E169F)n;o2q?mzDk}!+&+y|Mv3pMwUTK
z5p2ATKm6(WW%F&rRj@`2rMu5>o7j$^1bE$PdKpaj{g3X;<`ZYyL`mwB{~6hRf9+Vy
zFU5vIk`1M`#*A#JBpDVvYlDFJD%%jBq%>qRJkmwFW3+){I$B2(6m%-eN`gl%B>2P8
zt68Z?v|zYYO|yk$eSbqUVaVDh-jj8W$s>p(N$qYllE&3LfiTfU(pn$}i7OH}EPBb~
zew$}8%8au}Es3wdXuJ9^X*O#4rI)F}3-pf5RF2u|3XW(+peg1YX~bII&ea6rOqg&r
zw{x^!&^594)9l=6XN7YUI~kfAc)s51-UzzRXwy>%F|)+w^?zc67ZStDQv-1e(&pz}
z{j4&eOx#Zsvp*JX40Qd!BzM#@klVJt@d9YURlB6k5VwYoqj+j}rrzaA%Py3xm^YEv
zSl_32u5Pf@*Z20Gcaf#_!_oN83lp4$IahC$C&%IISffi$&cfs^o}5H+S>`D}M+cIg
zlT%4!I2TaJ0)IlIZf>bKL-I}(Q1)9Lk`<cly|TL#@_lq-wZ~<yZPX*`^%$RE*E=y2
zdU|}QZ=396qmK?RwDCqe9O9VT`{E3v2Cnj-AFUgYaw<KI35z}@V8qd<5j87RsYah3
zRqa6|&!dXrRTe5Na$W0hi##{tF`}DVM~|!n<8j0l!hctU@#O6MXpiujtTHJ$5zKJR
zzII^W{%y2Fh9pBfl$L7Ahy3Q09SXOR9Wt~*jk^q4Qh%SR;;t+b9)|Lg9AD%VCw?uA
zF%pI4D8?z_2%X&n#TaV{gQzP|%sGU@I8JfI)j>pAE0oooGYcS%+F_z%)G&u<4lxYS
z<kfh3-hZTL?TJ9B0q%?nnbe31-*+!3Eayeg6K?0xp!orI)Jk};_qO9$>PqOEalH&(
z--N7DXwY2VDU4TeIHNX)>*W^w&Gmd<fV~Z^lZoXNF2LR>B&`I_ap4)EV06yaoj}#J
z4#}^tVDXI4S2TFR>6a-2QVWn1V=h3{CjOEk8-H4uLDO=KsLP#qvQgL-@G?^HRjT3T
zn94|-G~xzKQ6KA_!Qzw(O?3+uKSwP}pUa)dIi!t73#WUeoqAVHgKe~h?-<DskW-JN
z5EUFMI_j}aDELQ}(T_NBT*UEy<(wT`(zwTMws3}W)zgTpSkGUH{0p9Y80dUx(k_{!
z9)C|xuzkga5%KD<2i&zYOtrZ$^c7{wtXD<lw<1H@U<9CnV?zZLR6O1V93oKklx{_E
z$T+SD4jJb&;E++@2^^yILiuoKP{X7y1c!{<W#BO8H}LWLFsBXqx{2ME1Z*8daGrBq
z0xSyX;vAnAaLmArb36g#km3QV$=Z^O;(sv&M_Tow%R$-?(i7#m9^mt2Lnu_$Wmo6^
zT;9Et@BmZO7*fXQGBM#bY8cwyp#yJ)c6yvU&DD%aa_t80&InL2;cjNQFLxebKtQ;6
zy@5$x=sXy=3!I0$!D%RgHcUpYmN|~H%^6n(@1Cu=z6&-aZF3x)8Ql8oOA}=ee1Gq?
zzQ$5lI&m|u7p$}RHfpIeWb701x7J(A36_n71sYy!;&X6A!KIvYIVZKaYQ;4%i%-pb
zmYz?XnseofQQtZ*0fY;9MjL)CUOq4N$h_XXfV9FGx8l6iLK)|HJPD^E*4rma-QctV
zw2&UxG!Iw-;4hqLAa`WKM3W^BTYm<co-jJ;GzU}NPM_TvLlY@os}0h&n5`lXX;jOB
zL}%IBhq)2BAU8tiP<{&4n&KWsFgk^P>xl4Um#B#W1lpAUcBz>>KK&6pXOT`<^wL;w
zqoFn*{(k-P)92qQ9(1q&L{d7+3T;&I-CKSnPwAy2==!xh6Oc7l#~2bMSbyv+wVFT#
zwPS1e+-;DSpaSCutzTnnh^J(&2c6Ra+7V$BFlee{h#XsMmG^bnSVz^1NIJD)#HYxm
z^o@_yzxIlc@fto?QtI_*xB1d-J{7APYg0V+UQ?)y!PO(qS(OlviWQ@QWor{SI`9Ov
z)hfYJFg?*a)VDQiK~srjEPw5VqxBi*rATW!*Or0awTXCD4z;QN^7a#UV7gLETci|Y
z1NF1Q#%^yLRjivW+~0qGEB#)n*T=zc=Cf5A1RC3IwWBQ&Qqm>mM~t7%#TD{OU9$)^
zM+T)Lnx)L`F2;bwed-3o$$D*j#XenZ45i5J;04CuE^uH3pxr=GuYVfhL_)Iafb^1(
zPDOC0KLH~4MbwyD#9Y)Z_%5XeLruEnR37G(TJ5W9ODm^8@4@Vb+*)EU3=7Jo{b3}x
zl&=RI4<k;`gHIG5+IeQ1Dl%@u%?%36u8kZUx64p$?e{p;|9sLiBUDUZ;-w_2+3PgU
za(*s}OlG5&7S10cjDO1ZbvDb~*RXX_`qW*WuuTb9-xuKrBa-9=$l=9+9$g>QfIDgJ
zPZyK5sR78vd#=rmr;AfNAR<#WwOLM!yB)LEp`R7@0Ry(rY}U{nAbI6E-Q&kU`nakM
z2nC230hly325xh%uXi$YuCI4v+4cR^jmZi;*$(hYb=VY^wSSg&u6UvX3a|UaSpBFq
zSC(RmE9J2E26yv$tl?)$xg>Pu+P|d4L(L-esXTbxE2!V!7(d)_)k&VJw~T8-x7Ajr
z6{GoT&&+nNZ^DJf%%eLFc=(!X;yTuEZ_@W(W$gVIJh4(^?pCR>j-a98g&I0&T3X-t
zsMF9Q7T|EjI)4r0xMH1#aXwS0Vbpi3(<p+v(MR1k6~?44Otms@cd65G4VKqw7{?Xr
zG@Rq|It}Oeew~J|tz~r@rJTUv{`Z9%6_Z)ue^8+&5U6nBPIVebe&4S7LFJgC#_L@#
z2BDCT?3Cv<PFOPInD~ZFtFkf2;TWR@J6O?Sna(qup)PqhfH2}4)bMnapvA)&+1u3!
zN?1b7jNyrb1})uyd^TGCc@|T!UxF<E0ZPTym+`hU69Y9cHkUDT9ut=y2q?w@p10))
zD47WZGd^>dalA4U1T;A`GnWB20wb4y_b3YjF*uijbO<PaL2q2S4ZizV=-6H~A}LZ7
zurRQ@ska4MAc%9=*6_j24%(oBCvYY}`tSQuGcWe+v*U?9cHu)}N}?o+B0rLPQ+uUU
zQtg$K2>yzvrqR~KL_xG9We^UI9z;*#qo=0IF-a+@#*j*~+K;$UPv#swB{i-=XmX34
zw58|;;i<cS2T>Chi7w+rWEK?qSV7?*v14&eCJ{Qq7=dcBaYLV$*cnYS3dtb|Squ4M
zhL_+oP1I$qYl%x6d$q*88TG_6MjI!t){iQ-XDto8VoYnSYaNuMtK;z8#=#w{IgkCJ
zFK=TFgTgj4L3OlQ7rJAMv$h^u6r92(i4N6x1*KzuWd+rP&r1@kA!o<Uw6QV<#D`X)
zHtFMfLNjH^*TGCJ=5{P9RIo?KfYi~41*DJluu+v*rDK8GYSJ=J3Q4sP1{#b3eHd+6
z9#k;nyrAMg4ZebD)XNblO)5NQfVpa{M+H>q3JO)_u?S`<u0;!Dut}`MnCcRv)IjN=
zJZYSN8KayGHh?X#w~>t*H5tc_!@?NGi9Syz9;2`!t8MVLWExu*oXTuSP_asrm@5Rz
zV-M)TdU5OlKWNcJpQJ|PkWyj1P#P$>MUBhCMYL*zt;17F4C0!aLTA;3@K69+&3LS#
zpls~WQtKQVP);5qqOXD)r@$RLd5prpTqst5ft$FDQFxe3+%;BrTyC%#sBu{64X)LX
ze)wVW>I^R2;n8mvzyAFnI0p_u>GJc359i0nw=hs?+2I|y^Bk9W&6yTIuRg4<PCs6}
zTPBGKY4Q4Ey<T2jCf+pv^kx0a>H1>5Otk4ZSzWHN$BBePL(<sgqz@nUO#=rcuCi%=
z004-12y=YAczv~cce-4kEndGmS-f3-S#KqN``?dC?EJ4K1iD-=FV~*}dTcE5X8CFL
z`Rd*B(*}*A>u<~7-e3H*`f^s#g#Coh&oTcBGsBE&v)!Z%sQp?O23^=~8r$L8XbcvZ
z88&%8C-5dKyzU8fyQ9t>ML$Bf$}`=6YI%rmmF_Xb8w``O0Mj|p;Ke9wZixCIL)5(t
zQRn6*-EwI!$8}9+8Z&x`#*FVVuNw^4IA_^9@4?B!In@19)Y@pS*1o@57e8KJt~LwF
zR?TN4mx^PIRBB7;MA$^Mi5Mq3|0>~FadG<jpX+G-_5J04(&DGp)o;tIAxSxZU;KIT
zm&MO#v}wFqzFVJBYq=pCK^B>{MTW+if$XL|Go3GfOj!MN@yqJ%YVj)l0jX4Es4S&<
z1wWFkKOP_7(9%O>kgA!H-%K)Kj~sd69WoF_GAhyx`GOi|QG2`)#F9g!RbY<DNT^BX
z$l7XyF!o$hnC3_hk$2~YJc|o|RD$b}N?DAb=RU|_1IT2FpnqgT9PkKLpe)Gl&tw!8
z;a-tM6aLnb8p9}}O(pj}*Mvb@++QmWQ$nr}xB4nXAWkd$!2J%_2A2#daz7XBAS?3E
zCqS<o7<D(4eJ^_XJm>K85cG1pMz1zuRhzJ?O<2`@BKqS&3&mTeD2)|=DIs(vy=RLm
z+6peCJ&%Dh*Bs>}1jwNURGw%H>e(77HWT6#CcR%>W+KRb04J@Ebpo)FzX7fsaD@^r
zqNwrQJrMFyMFG^RJIZ!$-4CDw9M$8UVL&7*ZbwaHskdD8+X|=$P=FNZ83m(eMP<(1
zGNB-Lm`1tf{KYDJC2j71qC!lx0}p;*>g@B>nGN7<doU7$YSg<gCAI1EjIq;0z@hs!
zIQWDBKIJ&Q90Cf$_To9L0t&_hNbd1oVUnGF9=759J)X-*y%&UTE9C&eiY!-hglW&J
z<EbvkQ-dg+w9c&&cB{3CSqLExt2_WA3?Kq31~A1HR-{vg+A{ioZLSGJC@+kdDb@@r
zI->iIM|9pNVD@<;uZ}c<_TdzO7P0NY3Gm8I+yBjj7AJ6D+@x(AH5%>veM`%Br)%FD
zFLn3&mKNIat?_WU^{w%A7zZ3Lhg(7>qE3W9-rLx54b_qY;v+AL3uNfba82B9Ns`52
zVg|cGE-!#+X?g&E<VU;#mZ!>uK%?=3oX49fB0?|oz~dAK?m(m38n+6}IDpHCw?)C^
zB}<8iN(2K!I^aEf*SRW+7n7n0!nJyFtKx^KS^zZhe!2<7)C~|I$!ipKw}2lK<3bg2
zk}41rG4TN4hXY}^I>{ga3x#psB?DIn`ogi1sZ~>D*BiKhGoHr36^VBkzP>Ub>H|*q
z;6&`t0LUkQ?z#&u4)qYGg=q#P)E@#Qtk*z7ZHpFkfVYLHiI~QRj~wpF1>g}M20Wu$
zdoB+du`3?(k(EXyY4BHyuRw<Acnf3|09!367zi1CM`SNM=E-~1k-y{GhDAYr(CZ$Y
z0U?SDcp2$`B972!mbf+cy)5yDF8prSfAUx9?>)7&*{MYz;(t}H{ZFPqE>j?{UwfvA
zF7*J<ejPAq;RUFguqh%~F5@F00zAu|wHx<LkM~I+bC&p|T6tuuh(zPjQIxZKo$dba
z-xnWo(V6<;@E~5;;wvMKay?`zD4hF~?#)P-5OfoNs_z`1Lq%Ga-iE_VlUviL+6Z;E
zSETERn7tjMYFF$Am_hn9G8n+K#AkYFL~kh+z9~hxt=K-0xWjeF)7lP~l4grn9+Yyd
zeIL%Sr=4$EckPT^H22?sOuUto)kaBo7$n*$DRy9qB737PGD9*hrED7(SG56+leo94
z+f*Wdel$^an!{LN8%<7HJRNTl>I8R;rMH`oPk!6n;b2%;=S59@P*I~dikdoAF7*(V
zi*P2R_}$6ooeY%>MR5$1@!OJ-CviOy8xNP06!CBwzhAgZ3h<zoC{5yvFmVGs?|5Fl
zFITSf4A?>w-uB?ENa4-);bb^q*wIVLST<h~*LLi6>qg&cueP<<{{d*0VVChdI1>Uj
zFqc797Zd?Dm$5E3D3{S2D3^cee<0YP&j0@W=aXcjr3f~DjUV%W@1IZqdzx9W1_j{h
zKfgc4ajifok0+#;!JPj1@9Ed4f6YKEDx*&A&mX^?rjegnh8{?U(pqB%8G;cY!^zqp
zRD6|XXh?qQJxfZ<^S8{ce2?ES@@>kuI@6SyH0mAHnSPSYKDi6pj2C~3m}c6*E^u#A
z=?+sFF6|$W{`6t|$qN&lRcm@^Q3?Q8=v3isG<|B)4^28{NLLK$n29uyKAG?Qj;K`H
zMJkDSt$Ajd^%~zstxYiEdeTE<yIGoN0*~3m%bwMcu!O^AH3>qrkW9(>Ta;zOmvEIg
zk%~m4&D6(F`B5y(yzze;Uq<SD;0Ft|G~X9Ub%Z$3ufqs5z9CA;M83~h=u7;&k+KAZ
zWujjySC%PcNu}AZm9tRxr_%nyJOY0fX*YgXc1(AYi5^B{)B+4=WuAKs38z1UW&KQW
z2%b?9Lxut0vtC~7Pm%`-es(Beg3uE8Jc^_>`Hzv=M2C^$QG$P5_b8H-PM>IN1?Zw)
z`g`8&#MoPA7GW^DO&6slxpy-GOp-$0`2?6Hr6ZN)Pa=L8>HUkf0xV8|Q4Unq$xaLF
zV4AEJRL}tJr6d;NohjNE=dLWt10v!T!X%M70uSm05mCZAjQo5+;3HNfW=&u{H4SsL
z$E>d#g7YHa0$P8Rl3A19NAcq-`RuEA6%mBCVa5k~lFu~|4SCANE)?6m31Sez`2|F?
zDnV3Qpu&X&uh)3gC@A20q-?Ymf!xOfWn(}r)n!XG#*hgu4Mav0Wbs@RaIm>)CorDz
z1-FzTl%NOL?3V{MJ@4_@zhojYtYl@UdGYLAi0b+P$tZs-l&ry<>cS%V(ZVA@sn%ZS
znCNaI$V!y93nI*ABFqIM9^g2o5*qziJI)bnbezY6w{-{VozQB*%Mk~9gY#s^G;sin
zbMYp(GeVt`riDwH_<b8^ce;@GX&PCvYaK8h4TUSV#`IwRBx_{pq0WISClFl_ngcUV
zBfpm!T3vs$(+BsDVYB>8vZ8W;5ivIW%6B+@4kOdUDTi?_-N^8_(hcT^69VQxtS2%t
zATL2dq&`{Kh^2(XE{DU+5eMkHNw}pBai8nn#Uz+A80|qA=%c@cMi(OCxe!(XcDV(Q
zbh!mHX;?T1^9Y8skw;`I(zZa^wD|uh(hjn3Sh0W7jYX8MWgQ8>?&t|kVIn0>X;Su3
za+Q)*sFW2h1zqSW$5oS*WIUE-ILIJzse5VU4PM_kN6xzKT*dXmJ0x3<WvbFt2mL<m
z*4LWYM>Q>#t3pv}(T<LwMEltwt4uh{YZ~!5x^bWG)qSAc&x97uzjoRsyB&A!XpHpi
z^L~F$MVI1_=A16NeL7N{@#FbS-r7QM)qnl_-*VYP>-7~Vh`3?JU&@NhI64J9ilkq1
z5KAZCqgeVS8Ot@N+jdEhPdn`?8mUpNvdvi~7xz9XpdCvN_nfoJQOwkm<oE0l%d?2{
z){AYI%U7DB1Lw_HzFK#!(QMF%?tnlaW6*z6MAGH7r2p}`6>oaCg#OZrnc-C@8P{n(
z3w3j5)wFze$4o`%j?owATMn;`TRnFt><iQIR-nv>9mDMoyF9135xsD`e2%{~r{Cb*
zP;Sf0pR{waf7)rG!O!~#d{}anxwZS+lTOcT+2L(s3`!`9`qnVQE=EK87G1#-XVHJD
z?CLL3U=W3IutouV>{CG5oin(>^=ua^fbiAm$9m>iZFQrP<WvET_ZS0sdb}#R-(s{k
zdPBL9^%G*%f`U@&JYXH>t&+MPVy%})lmL>?NB|#13-OTp8K_Bxs{HrvfVLK(#wzn@
zBJ_nF2_RQ-lQdt?t>#^vN->S@+>d`jJ7H#_=uqGcCYwik7iC-Wn%n|Zle_?-7(C6X
zJD#fOukDQ24Vo=s`9Ov2;*;XMHYsKZFZs<jKHS8o-+CNcUp8+#=DfyvXYpfHxe|Al
z=(^PQO9l|%19$(D;>pfQx)kHI0{@`~i?U?(7KfB&t+|u+eQXO{W*H25f&hOrhFe*l
zL&0o~iRBgvUSt`q5d`WO%cWoBR{KSD&D)>TBf7^u-QKT3(B#A1qF+OSl=k{d`ZY=u
zTn|^3$_fTqOJ&1Ie@bv^Y{F9ab_^bNTSDBqnV?;701xZXC<yfOk?pPt;sg6#6XczS
z*EIW!I$ju@fL2ZZqQdQj_0E5e*97f{cf2Ni59oN!5by7J%@8klyk-gB-tl^%Y`8rA
zK-uJu*Fn*6$7?na*@=gAfsKv8W<%&lE|?!$FX)+}uLnhoh-lBXDQPK}a*J)fy;Fyt
zT$*<mI(560U*EOs4EBM2zz6jL(HViQYx3vx0@0HFwnobamM`uFMpu8J41D#GGGcvD
zFK~kTLwkWU2XZrHsLpm;acfw11+BpgvZNU^+#awhNsy#k@bb3GdG&15zTDy2(hV-`
zYp7ec%vVuc6F?`^(wmmoBDH&L9BC7}^74j_D+{=7<14iH7c{Xpnbh|#N99Y2+<v&f
zmJD=?7~KM5*A-k;SG#|W>83Pu0xIo#cwXLEE`(rLVJ#_Vp0NsTwN{X@b7%J2g56d%
zf!GxQVV1&D;?o(wQ3?o3IJ1yxO8MoGymOI<CCZXX{-DaPrtPjurnH|vI2|7(*q6(?
z-RcSfsc@ID>aHKKwGRh9_3o}&uA`R@y4FKkW35mb(=X%e5juaCJci4x<*M#_Tvo5`
zb{Bg0FA&!{)K~yz%)OKya{|0my6Y*V%dcv<T)eo*+|f{I>aG(n3OzsU@^96&!oTmY
z1pDDy@alQC%H2gN-(D^XAc9&{iyR22;7f`{u>vXOSKq6eus$f3nxOvBSZcyGb-lK3
zLd|^t_;{*BKMQ|)?}(>zgf#QA5l=-u16=IQ+Z|yGYc54-F;cwgZYLhJ?LN0lx!}4!
ztzFXqGG-b2YFoYc^O~=zm97%@&aB+{2~#s43emZPu|szOS?H{}>?#@O(hWOGNooq(
zuCCrIZpb~VsZ)rdJ74TdJMBHsN$~(SDk0<_XIYuAr)+-+JN{y){c`Lf(oX$@ESj!`
zpvs-OU3&A=d7BNPDLY>9iUe%b5%%t`bUJtN>Po-j*!`GZ)JKu@OUC;#M=XEuCvWtK
z6ue%pl<;lP<JnccnbqCOx;xvV{1#z1*;>%5mb983TJ-}GDyxM;y==1g?X&W8rxsV1
zEGfqS$qy=wv1mAS*8;o|*4l}L$F)YQ(>vN44OVzR)3f`#v8fQqpA-!KA7TOGEtkQ6
zC=&rNm(fE!6PG4g87h}gCMcVK)>AzF_V)7~O+r)RwEyVe@VD2uyT9)66r_?cKR5jT
z_0Inw1tY@pfpbNe(!;kuhClCq!DI_F3S;7nqQkc@18Vu4wCE^kk!r0}qeVu6wH&l|
zMEIYg7Uc`23T7tH@WAY)*8)Sc%pAg}k#TMT#)XTduto}V@f?(n92Dq(S3WizKAkO_
zlv!)SG{@BVxuRsF%u2--4D)zQ?JBm*v}y|!cvXAS5Gl+G?aj_DYpSIP;!*_`DovWh
zIms@)kohIZ-n7(fkhRuQ=CqNk^fr=c!*FlW$pH=EZ4_q7UXRl!SiWLzl{IPu<e3;)
zvBsw5^8%f>TscfuxzUb)GG}i;K@vNoIZGVxsCg}^Deaw4_c59X=pkNLuPf(_NbNAc
zE+(y^sEFa+DzKLw6diHKb`;g52Dy-g3*KpPz0GsU3556~bt5%#5P%QVjbfxXmpRoa
z=tX-6p|$H>49Rn@$F(FBoCQ-30d-O6;zko=Oy@>V(b&BUc?Sr86wT;PK3s>5+Hc$@
zZpPk8au|={<>WmC-dZgyF^n%^x=l+R{c>bslE9W)v`ndULEFg*mq@E#p<mievXmu2
z=(ORu`1sUAdDj<-=4&-W8rx$qiT${j#>Ld43Jn*-kWW@pK!JWhydFu@NHdNO^1S;T
zPB}wLep^<8L(&6(shp;QXv)jZvT};~L7w;N4DPNV*cfHoodtFQ%^I#h{P@uig&}f!
z2q&PocF2d1Yb|hg*I|&|Q8y^ZiI$@5^lHpr^6naUhM2k1bA0z7!5{7!hmT9gsc~F1
z5Blmq@$BfySbU~*ef&6ThgT*YSjge6T&06<E{txU3mfNu5;S^E2-OjpNR4-;KuS4c
zw7zi)gt+<?NLKOKC+j7W$T8hkW;zwJK6&isvLqGSkOsvR1J8iYrnosdz$3Pn5i4YM
zmOvoGoZ0pQOHfV(72RB(Cuc6GCJG`OUYFE+!?voVC6<n5!t44#hMgFaw0oz8_|CCy
ztZViFxj;t0#5VzZf1b4e=J=LZrCOhmcgAm)%iJzrstk(9n2~b0!($|<*NIz)$cE`r
z5waorz5%jf1**IpBV)OPCjI^)vII0F?T!eUVqQCwnE>exjB&gy`^yig-Cz;eeEo4T
zUteTrZK~|YiD^@sO3J14RUoh7o-B=giXdsI0!#!&GsjXrf21VddxUQb))k}<)xb5u
zGs!Gyp7^vFk>m<Ve1#dlXEUwMHq<(&5+{>o`%*^b5=oDxU|EK%rBEz`Qbm_)vq{Zc
z0o8^)RP+8f8P!w;dXHiUc~_L8x!S5C(2O!e`&0TZBDIaRT5XIAYqe&`hdksmmrC(V
zxzt0BuLz|9e>P5Stq2p4(IirS&u8DeN=)A>$;9r=N@@n1xBh2jRul8vmu)qRc;Set
zI272cBa(Bfr;f@br-V~G!l3ye3?2=l0_z8~&6qYym&DGWf=Pdx7B&`_zsZGL6pW9K
zcKK{iK%pc|SzVEWVMg8wSJrZ$JHTT_wwbPV^tqlBf0xilV@?~>y4m>{2a0EyOR+Rh
zU7dLFP!75b>kxR_6!t}uMHT%)FaFSuExDL!&eRgUJf9G)Q=FgN(+cgfw$aKLuWdYA
zmCT=?ipP@*mSHchxNs0sIma5Jk5@)P6tJCM;|x2wUAEscziRgb`Eq|%p^!Ffp!9!k
zJwrW}e{lsm9#;cuNBC7g^lKzdn+9Zyw;KGnB#3Ie#g7&)2$u#^UY1PyGjoI2zKaPE
zIvz?~m@Fe8{pGL?<JsvsD=d6nw)=c~pX;JgpQUe3hbsBJB?NnXnbx+p1WMONCsV#O
z@(5YZDSXb_3KpSrkK!}TuJxIe9j(1Rz&x77e>J{6z!dBRl;7m`fPga5XnWxKH==mp
z`8T3`<@_7<zR~%&@Vk<_F|g4x-Ff=m61xs)sC|duExQBGz7fIwN8gB`zIgJD)}J14
zV3ZRlZVjCwt_NXfi0XA)^E`dtEuYuhTV5bE6T00#geB4%OmWv0g?O%kDZcE!^nfYO
ze;9GR(11;>pa;CTdonh<ei^JH@N@#5=Yh<0<btJo-i%CQN^G-wGc<{3qb_2T*zEU%
zliV=~RdQTjUm1@M6mly%UK#M&VRb#B%Hv9<avRW+TL?f*{SR8M0WDohj`=9%2=>L>
zHZ$5bu+xlJXb#bK%qLhbv2<1H%cU}Ze+PiFHoS$<VKZbIKv^q%DVKW4@c<=<(t^72
zx~m=mvUnoM;(mjIeS2bP=CQ5yFd&x}5)s=;q^?M$W+XEEB@%`(<2s4-6^ZnWM8}_=
za*&j3c3{zlDe6r*J(G?w*~vGv9Wo_|aJJ)3H{WI8RKf<&-_4e$k=x5aDDzj{f2%T7
zk5I=5J>+N?;S-yntO#D%hbopn=N8_rd4X9>0yacVLDf;&l-6rNWt8qY!9?|DAp6~E
zD-5Dx)u3=gD2Nk8L)v!>5QIrX1QQ4P6QXz^2!beI0YOmj8-XAUN^#+Cje;PqJ7Ew+
z?cH$@#P0wg1XJ9PgkXx7p%82Zf5&4Xj0z%+xwW0OTo1xR5Y?MtA$TVHu@FQN?fhG?
z5KM7D7J?~W#6s}06da_ux)ITQ!b=dKX{iNBf7HM<RdvYhZ{5%Qfg~X(y+o_!fUVUy
zIVDF9j=_!WtSo{-Qp3B&@h&cD9~3vEMM4>(dmESsi=g-<!t{}016bsDe^72aN;|+J
z*jcO<SA3-OwgQ%|^QZ&BBFBkhw!X*~up(34p|E2ps}sCfke7QZ2KBTN+wf~1Mx{Ta
zD1}bkHepK<V^{|-LB+0=`yXAp3AzNH_uqq-^5vfJ&c;WVa#B53e>Hn*ZX-2|Uy9DA
z09*}RRVKX<QL0BQgRRO+5-#L0_Q!zLWX*=&p(Okp(?jEz@jN&a12r%>m%$4?6qmQ$
zD<^;LTRU&$HW2RpE0it)ay&j}_y7z8wrmGTmxH??Rgk;tiroJ%GJHvVNW0Rmyv5}T
ztydH|ocTDM8GhFC;mg2>-|n8??moUM7!0#0j2zxR4}u<W`V^cRH4a7yrj;1pz7Buh
zb1pSqPjmt2O3`KLyQRPQ@IeVnJp?@wNSA+j_+UV!F-p=U=+bcv^;hmb%8_Gw=G!&N
zv6}mFg6@U&(jNYL`~8k5r8#rjfAxRbyO+1SfA3loPy++7!{1-;`~y&eDJusdkfAoi
zw||DucYm~Er^!$a^T)R@L(}ppp=br6$hAhSP?V$s0jsrR%0Egdl4U?v6HFY)LL`6d
zQKvTP?;k!S3<w|o>hI)z!u6Rh-5_#a$xb7*Op&SkHNsLO^}@p-#5fL2yq|mcJMh;l
zeYLSyXydgFm)IgWIW7zu3<G5*TOjHWChP3A0^gec&e{Q#*g6-yE!)QDQNccs1^GU{
zM4yhhM4YBRkot+(rZO%rj`K48xS)TLji`_|9*J&+lxARTwyM+yp5l%|74Ky+V^}?m
zGX`yRY$LFe)(;+%bBbq?S<!6xvKmq34o!R<2=@>%2qH1AUFh$E$gGhcCS7!(aGx}v
z-7_tVOO7UIErpArMDrkW4Q)x$K)6HE_IB9_jgFT};sCbl9IxoGUeX$Hp*?>~R#AyO
zT0B=<?c*zm0hu!fZ56C2)N?H=$NTx$U;I!S0?R`rBZccBKjk5(XDI;RI!ieu&ax+L
ztOjXH1cJLk3!sFMJ-~zPqg#x3b%dkXm80Qvjuw7{<o@wFwOYfc_(IW0NDd34+wchU
zpgrO*P6^q5t{4gH0MI)D&?|o$jy(YS0sz*<*8p&`E`A`708}l(Ask<Jh?rN<Qte}5
z0^~Vq_!<L`-)d4cTWiI|fKwHPoy34cl)%6%mOp~NRZF>G;Fy3;zot%_dZVIDZ`<A}
z4svF^M~L(h=B--Fd9`4Un@&=f|CjZ!iV5R73{o9%LEfiG$&AEz+_8Tr>dtA!PuI_=
zptM${{1DQvuXTmFp98wDvSJlIrCl}GW4@k=Pcc2om04=_E%+cq;`BXSksxg~Uzyy;
zIcxb8m6OHZDqCrCGpo8P8y$JCJXd9(->jj^KC@pbPl&=F4%n}?MC_`!{d-qyCbVoN
zOTN5k40(?cZSTb@lDvPJla(UJlk4Lj6*Ww9vja~;&BU-F{6t~as^!yp%$r%IN}ddJ
z6&{M8lryNpT1L(?Cp&2#C0wq-Q+$X~VfVU9DHv$i&L^P3XOBdr=+>=5Q_Ss?=%~1|
zokYEeprptpmF&ELAMa<{GHKnd{j8qV`x$IH)vaVVe+KPrIq!dOIcveV-{<lQHJ0$p
zJw#as*0d6@@nVq7&`F?`c!d{(W{%QFD{;k(Q83vpc0m9<?}uFj<K$?kVV`?bFKs&~
zzhJqIWGLm4P#VpQF<YUe!i}qLDS`LlGi_YH+eSra5!$hxMz5>*Dt~}iOEH9xR~(5_
zWJ1G~npQXkCZB(PJ^49PDg_|`HEZF+M<}Zj=@AOLkjf@vAdnjnp$kKGy2m;C;JAf%
znwg#Js3)K;ri}bSJ>7(Q0)-5O`tGQwd1S*GoB0G2E4ZXhVO-W?9<}r|>Ntg-$FwTJ
z(&-fErciHZ2~#E6oO(7=WaS3gNM2P#wozoIV{DzJ9CCjoMdlbC@CJ~;0O!-}OZgv)
z{QnAnmQF;Ej9;-lokKjG=bo#la#%T)!xd3ISs@eyhbOGS%~s%rR^atka4H9sK4FF4
zY=yqi3VpQ|$_a+XR94{5wZM&Qf$z2!%F>xa*b`Rh%~t3OtvFyUv|}1wWrf~ph2Cg|
z-e|=ooST1As$30#l0WnO7f!6$#J4pJoKR0P<;E%1PJL~Ww|J9R^wevtxVfBtrx~$h
z0w-oq7uCn(f`pg|LZt36P@HlHo!}Y_(rJASImIob%Dx#CN{6woZT$jXN)ZZ&OgQK$
zLaFIoqi;eHDxX@>9Y_&25)Y&Z8_8!V!lw6)C_;Zaaxnq&#?(zS>y;E?BkkQO!e+h$
zD8g3a^%P+%aXd*8wq9^NMJTmlQo^nMte*8Cim;J-6N<34<a&y*k$3<_*h;*fB5Wm&
zrzpa<^UGAT6pO<XN5l7rb-zh^PhR8NoMOiHzOvr~6de;3SG@;Z<pr6iYpqes9Gx#P
zWw(FA$1;ch-`v=1bb>kbc~-bitML%(P@-m1<NuC3lz`-K`UiJ-6YkK*cLnBq<__Uf
z?r>%Y<sv&6PqgpA4whSBBX3s?*~Si*OTTrNa>()Qpp^a<5Zm$xJ!8<m19k{gu%GXc
z1PjQT?{H`Jb!YT-EBd<Y^`)~t)KuAcR^5N`jJo5-*Xfpt81gem7)tNDJrUzy{;{$+
z>ExdVq}L&7gWea(=5fwpRgkiYyMfHqY<cq<^-0EKQc)Qy6{p`fsK?B*B=!bf%NK=~
zpank2966?-Tld(cXN+%3yRJ-_<}cprT0<#tNLVvPzIL7T8SA(Dq0+YEr(#P_WaNJ*
zOB_`rB|XIs4JuvCeT^OJ@@=;Ht+ZLTEy*Yz5B8Y&IsZEo1TRxQW=3tA73!p!*Ia7G
zXXgAoy2U!r9HL08&MQ{0$L17UC!A#lMW@k1xfM1mc?Y~Ct>!X$&BrBjI#{_@`la-p
zrsN?z_6}%^`xi%VK;8+iDE^rP$-81k;(_E{Bl!$@*YsW|?=F5<)=7plvn!{cy|e)*
zm1`?NTyY+<f?+{sx*~4oh?J|Y*SS)|aXKEV6z60~?8OMTRN*ZU2>X>b<R;UioMafM
zLNRArwIP4De`(wO2VWd3te3%mC=&uOIG4fTIun<U)GR2MUpOcZ0+gtiZ#XD1e?jBo
zgg4;S^7dr`dLZc&Bn?`#H~?WaEN|bI-*!?eOV<-!D5chPIr{GDZuXxvc$xw91PooU
z|Kt$(9j)jBx@>-j=IgM_bks9Hi{qB)cq`+3K)-=#gWdoB_Va^erlklrevP;J*X!HE
zUk|egsFer^%OBq!;sewmv{#Ele<!5V%lAK*FAu-WWb0x^FYV*|*JT>{m9^-Bv?#4L
zX3!!?QY|jl2BG7ltVL3sG^7zE36Uhk!~Rp2*-TI1B|cHCDeYNHD$__~E^AQDS~}18
zic8~#Ch<F4Bdvh(PBuqc#gLuwF>_#u1><;J^8F?AZA_KO@+{>0AXTz^f0S3TRJuu&
zB#if$>SH%cNH$09svRtA`#|@w;vAT3xy7|n<(GMvF5=9hgQAbBS)VOVF*8aiiT(Ux
zWR`FkAqhVpf;{PY$_N9YNQ}gj@mb=?W8^~3t6E3oK`#&&N~;bdY1JeCjx<-Bc~zxZ
zzfpeERz|~|_<@5;&3O^je~OxXTE|JGB3l&E4EdaCUgITz!qhM@Pj@$p2})ShKiDYK
zs?jLO0T5li1Zpx6ws1FY+c-Xk(n^+h7$-h`&LJPyrzhD^7!_DPl7z!ZXDEeM1WHuX
z+F&+0RR)AwiC@fm$+9L0XToY(5>Ls=To+JvN*JY9RB>AvXN7Za5~cR*z1?`K@E)sP
z2J;l7wcH1nuR16qe_^fO%yq13S{JSABP-uf;f$uH2Yuy$t6IvumNix~Fl#Uc{c{aZ
zol`*KtLEpZrE2%b_Fk-B<Qq40t*r3AW_W$FRupf9Eq}XWZ7?FVT&#0jg&|+D@rdQ5
zs3krJ*|KB$uCpm?i}O?(q<zvT_cJ+}mY#r@C|y95ik_>Ye{GvD&Eyj4`2K4mnap}%
zBk`&ba?5O)qd`ff0l7TiU!_usA~^V$x4RfnNKpUdb{D-C(%^rl-MvCWDxNa5jnyQB
zsRb-~xzE;2+}bG^6Hk@%Wcw;{2|`~-HaeW9uZ<3;Hl}wkkCYb$ICsbLI8oe^Rli=`
zJb|?WteUA$f5oG<{`5ZBO<p?9qP(1sK<S<ur-`zX<=Sq$Kt*8SBs*7k4XA@qP-}Kc
z%E^rW6!uXKYWxxeDkbd}hNA<u>aDsqgeKxE`%kfiamG#gzto54W_!c2q|v{-iljFp
z^XDKY?ToOdim%-sY}n<Lqn%B`yH_bJIdLS-*c^~Nf1?Xg?cA;lfY-e?yXLxS?1=?X
zIp~>2{1xi^C1I7WsU!>N65658F7|rNDv2Sl1MFJm&g+!9SR}xr+@ANjTg;<T3$Qz#
z#*&eoIcp@c`fe~3>!6x%kh<xl@^WO0$1>ZATkOh3DyhjaR*qn`a8eI(@W_4u?i!iW
zTVuake{jmdWAcdYi$@)Wt<kLe`U`Wp;9t22g6ljtam`~T<z0$n0&OnL8pf1bDDBU|
znCA;kf-z6m8jT>N_jjx{6UC!jvCHesMEOWw2ZzE^r#j8nTkf4eB{E;BeYe+c4S~wZ
z4~BupMQ?=WMq(e1%xUSVQo;v9bIMzx!jFdLf2LLM1kFuU&s*nOI8Gy_AT_Ud435*D
zx)C?ez6OtD7;s+qMm&yDzM1L~JkG{Vq+aRzdLLKR4ryJtG#P_a|11WljHY({M`x;O
z)q6$7NcH2RVqEWBRE)E4iHd36J4MAr^-NSuy3SDno&ROum`1ju0+qnFNyq>RUn<`l
zQsAX<*uEDe^QHCO#tlzH{WW!a1vVRi0owx{2~07q&&2^vJq>~M07D0`xDJ4_Cn1L_
zV+4*mDMku|g)Y&(qQ;Z<Y>HY<4DAf@xd&EoMmC|B96l%^0g9JCJ}4F@MOhhpJUS<N
z$X#qP_}RYgTxv_m-YdfaFN~eZKJg`hfUns3q8^ueJ}4wK>3r1<tr^k6^``k0dRUS{
zIck}(&}*cK1<u;mM3&^#!vipF#zXUp=;&Re-hmP<rRGsMm(o5cIDbl4U)}UM9JIDw
zSRT9Mz(db-ivhJ8*H3c0UK1VPH*4J*;jlKq+nFHk!}}8CPRmnT*2#}dIoxDv!<on}
z{uaiQgJfa5Dot*PolP9QMFN&{yBjhAH%nkpIN$!}^g8Bpl<xwSPouT~URb62qwWc$
zlMKkoR#r8vfJ{D^9)D+~*vODs52`^&-ESPIc9A!Md?Q?U9H3jldld?He67~F01fY{
zH@Fb%w?l)aP4k46<|S>98l0b+U80R3nOgpqp;PANl~1a%X}{NEY1B|}{aJzO>FM1t
zwQwB|=i4%#`EK3uA4>B{#W8r<j~sjyKXtartv>5Vb!VZa<t*<&>F6;$bIg^qeZ3b>
zW8aKy;Lq4xIGs88?$DE!Hvcb!@2_tUe*??i)0&sTe<%|IFgcfTjTjS`TtFzpfB5))
z_~Y%@J6V{PBG~vde(`_Dx4XaZ_!6x2LU}X%`Q<L&FdBsRidH5ZE$r)G!>7C7IO<4C
z(v<vxa`^grU?U%k5Q8KkN^21tA%Y~yn!(y2bi9>?XkMg5T9|pIixzL95TIG)hgT1a
zh(^+OAV(y5OC%8+5}8TJ8BLhTe`ozj>^CFC{)lX?0IN8%k_}(u)6NL%fIqD}K4^f{
zMbawaSa`ZD(u?vmP8jVr$djbz2p;qRkx-)AjC@=n^OmZj#bU_R5GTq83n%`B<h%%~
zLyJ<vCF#8$?(=cLZy-m70wL_FmkiNv*FzHon2tB5himq8L_{Yn>JCvyfBv$>k<1Uq
z->&SX$r4E8ZgC%#G{PWo)HD0IRT4^jVZ7rM0GIhON|3$fPvg^ex(GsbTM4fVo3H`Y
z8X=+E1Tf2sv1GEX?}YnY7DKx^eI&Zg$b9pws)G?|XELNyrLuhTFYja>V@xf(hD(l<
z4wBHWah!+;6;E=~36r&rf2;&KP8UZ_b)k(`x>g(GT*NvjX;~QOnKGx&eV%kNo@=y=
z6i(`WGXo>u_|HwNNjN%RBbSdOhmrrMy3RN^-8Y%DX*ACqWf2&DK8K5|o>7RqRxDJv
zFGg$Oo#y+Zk7d^KZW=m+md}BgqJ|p-5fhK7_dE~*Wd$nxE|CZ#f31hU_p|s6amPr6
zAzz3@*wgP6iJ-O=McQ|cL>Sc@BN2wSE)vnl?S4gpFg9>{oweP9h{CCE7wY-E3NeV_
ze1~Y4-IO*$Y1L_g++WYgeR=SEo?C`^eWb>ac8JtC;y#fYL%po$`WUgVjh`-wWeZ>s
z4g*KL<=~WPj3eG~e{f2~#t}~(9H+?TsZI%H6~7wIOi^&Tc^kAOw~DUg==G#4(Ll+p
z>J%l?uaV~{ne7oMN(w-5ihN0lK~8Fy0=x3RhC4+KLwTyIy~?VO#}Zbq2;?*tjF9K)
z(<#cxx1y+pq9GHM2y?eOhOSS!u9Jd4m+LrV$m$Cp#VtkNf8!Vlqwsuyhs$hN$jW9^
ziOPP@MuyzLNvuYt|EBCgaiFm(m7Wg?avpoDP-1DAHYDAzn{dp}M7Sj2!X<=U(6TZ?
zo|3f(5e)1q?Hm9k(b#<H8Zn!-Jm+g>8BYi<jaOuynW(dwary8{>qK(44t?wK$SYWw
zsojr1|CB@Yf9PwLCMZGd&L{j(C!D7+6ZTpf^C_9i7L#z&Cj$zQVIVaZfii+Ich3ee
zmQ;-_vOkxnZZS%QqNOlI>5d|vpoq6qq^Ln!(^0f1DB4xIIBi!DMk#5Ra&aRmH&9dl
zHkPQPjvwx$8Gcx6ODXL%eHxom<#UN-Wkq`m75!9%e=2FO3aTXV$<$S&(ioTclzyl_
zpw5{m*yX;RQvZ~;tCWWE=S^Nsp0lMOF!4GSG)DA!iQ}=%wR0<VP^#Wf)RSR)oUQjM
z8yXe-X1>x&>e|AbFJsOZ2Pl!CJc4bGo)erZ8)xeUh{A`!k$BDH>{dL_#hmLlbAjnv
zA4y?Af9po7l2K0V^>oH`yM|G%no3Re2F8t*mB-n<ZjPU+V*CgCNsaktt?EpC=E}{?
zp~Y1JEb~u!fSD5i$D%uN`Jfo93aV>spJ~oT9^spiM@uoOB9FgTT7!a8x2aVVNr`|b
zgHEw2U0zos*^N%lp;Ia4KzLH6Rw&&3#B2s$fAXbDb{NBl*rc0+uBGKPAr6lfS=@U{
z?RBt}VtK~nY@7BHh9O7ngW<v$cb~r1Ey9`Nq!ocQ58|r!8WhIklvL$w*_x~tF*z(j
zkw*u9T`3Z1!-COWDN>VIy?K{Xq*huu>vk+@FvJ~8kqr4lDUv<Csua1tNJ0iD?gwY-
zf4UlAsZ}tr@6wFalI$G5b0xJUdSjspd~E6Rm0T9)oaDUxac$+PrN2|-j)9)<+c61T
z-?pQ)l<YlBRmv$CE-#6^Z0h9QuBLH`!RZ~%7DHTeUX6y_AWihNm!sQ^BpPJgUs~D<
z8SI|YY7y7Ces;dH^qkI?B7LN_H6!yAe;{cjI8Hf|Nay-FQ=U^<<m5!(o-ziMoutNH
z-eC{G2ylD$4mjd;DzDZ|sfALF=Uye&E6u<)4Sldd$nlh>Qd}65+ML}~zuWZn`wsZ%
z1**R;yMb=rpO15r&v>o{y)o-*``=wZ-7P01+xq2{nUgnFY}uxG!Z9W7cjR1pf2?Dp
z3_8%`L1QTjq{7AHlr>g)T^s0<_tL6+_b+H+O?a0u7mO1c_VgDFamO&1Azuh{+0)-9
z%;hZI6y`Fp?-J&MLZQQ+?}BnfZjCqSLFvCT#9KPLRkIl4jUC;h8!hNJZ0KeeeMv0a
zJGu;UhmI~syuG8#5%FwCmn);Ze>J5cQ|&@0`*U5~()Z<@JkL2<>Blzpn=Rxx;ewW|
z4?MMtubp*kKNonO2r^B41wwZEwnJANC=@ia7iI8Y($xkEIE{1v_qE@puN}xZOfOXY
zm)qKNy<6FIZl<hjZO`PrlE{3<{*7%9WRk<B^sei5Z%ymAccaO|+Er}f4+geyMYhSm
zDS7z^c45Q@m%)E16ag@oVb(4am&Zsb%YQw0|9txUK^CEV37t0r$FhV($s#w4qfKC`
z(i_Tc41K3oPpwc|B=yBl$Nf{Q&X0M-&tpr;s@gB)zQz&#MeAds$+UqvCn_kRiGy@i
zUZR&~0p(ZXaV!!OH6r_1$JaU6nJuKeg#Nmui*c^-vCHwP4S$0VdEG>Gv~=|Gv44Ae
zcsl$tEU491ct3slexP=vLFk|+P)=CoreA-az8*gHREup$A#qIpXXW(k^VGF`E=C_E
z8ohMR+J(`B#GPO!XUXU-o0R#06XGR3=E*8p80EXi0IL*@Oq4o`z2V9GrE2v<QJo>V
z`CppOqUCe59efZ`4V=KNZ=Rq9_<uOe6h)E8!rSU;{+D<!T3X($H=3<?OpT^TGqj{L
z)z2`yip0?_WISgPPp%*Y;cd9)3N9&<fv1h)^W1)m6^&LzqkGO9D~fysL_f@2CP$HZ
zds;du{}S&-OT$}%qMg+dquJ4<x&^C)6mDEKD8Cxi>zCC=CxU@1E^AM*E`M2PlKVG<
z4ORqnT)b=)6XnaXA;&!(OE31XD*IUbkYnjN{#V;VBj#~h7EkVH35r=%xfQx!1OXP~
zX;iw#Q^`>?#}JC-DKh#TxoZR|QHeO#h)B^JSn4~tJuNtsWI^ryJb_V+Omc@Lgg?@|
zDXU;|U&0}&e@rCXP^e-yvVZhmBu~HdGCbry2_~pnN{AW4IGIRpY1?VsXy&TgPS%xG
z<MN|h_;Z%b`5}l=jB-P9sr`NiOe6%4W&lvPBJMMEX+HUec5?&nPgoe<U;p+44aFg5
zeTWh$@N&psc*x6K8ik+Br5%#F>~Rd6E(oju87FHs`oK8`JJPwjO@I0;-|T()7QC##
znpe1}!r9;iO^w$t#%eM)$<}Z}vOb=#K{bSaJ@C^A^>9ld@cU)LejlgQ8OQ4w?DW4W
zHDjQ{KMGyWc%&e-b-~em5AXi-{Po8#KjAu0&;O;ah#rNq5X1C6zd}GTse=#GH*V0v
z1rLZXMMPsZ*kXl^5`O_GkDXh-Ob3Je7PJj*^EYaa!UUsKJ3b*-pSm)awv9z0s<oM!
zerp?9wwX(sTNQ=gh)NktuQ;AZ(QBG(^Uhsy)W4>~*Xi)7XxBRDJa)mzh0&4VT)@q-
zqFoRMJSK*XhPerFzo6aV_@bHU(N}CWw==Gd3iag>`xu2#0)I_iPnF-i0HvjwIX59K
zrci|{pMCp=Cc4>TOSM*7U&7l;C>6+Bp(r>jdH(+8dl~qQ!#*~jna~X=EsS;3jg{zN
zOrn*t6LdpsDS7G6y*XrXJklk=WG-v_tRsNzK241{d15!s%s!p34iFaC?X)&}&^T*B
z8!eQ#RSO%1iGR`^0}X=ASPK*JPTN}sOO&oUj(>`QR!xpqz!3G!3AOLgeU0}N|KGO>
z_MB4ln2Kt~{)_yew$|j@tripf*(rD1J4gA2q}2BzYbbM1{7fqy^B|DAm`9w!wV5@f
z8(_LXMC05x^yQmEO{TpBNM)T}5J6UeRCyIW8ZftM#eWqP4O&x@f@5x1tXWUJbyFP4
zV~w5rCuHL*)kfyhvSK-@(3R+Fe+M?fu3SMWwVBFRfxgdqMS%o^aVyM5G$!%gr1iRK
zWlHPYM5CLGIu7bvsdzPNIM#{k(ysu>f$O^b20cp{c=oCC>TK51XoefALW_!bADQlp
zTChN+B!9bRL}&mxD^-@VG)%rJ>S=~fL+cFWwq25?Tyl0v`s~fSgvlF@&s*eTdM>-G
zY(-e@wZMp6%WM+oEOB}W!7wWRrd2lOgfX}%8e^|E*GT43B}%!T;<#J)*I*E(d8>NL
zXppN_wTMc@9HZ;Kntb0=VFZ`&;q}E4)+eSb=zrLE=#$`GF8u&8N?aE$pU=?wk}Da#
zjdOljBZuX($A;z%V#8q$q$Rsvx+t!}USE$ori|l__1JN}(0R<&K+GwX>8YO9XUj9U
zmMm}Y&QUHG);bB`ao9b(sQM_J9OeZZ1xd`i0)pvY3-i1#4^4aZfiYYdJ;*ZXf(=zx
zOMhClykNf?gIqLzu;2BNdCI2}6$Nq7-36xuysvQWjIM#};>y9P%z*QV20SG@54!_?
zTP(KG_mmZ5Q8)NFK0Moz6C^OUURHMk<iI_Zc5AlN<~1`Ku2})|p2y_;s4X`(HBDIh
zx0M{SN!C_#$eIr36!PEGvRkS6+{`5(PJes@X&Q0K%2~9QMrb)FzP3ryIMjp^erW0@
zaFS!fo<u75E_gV<t!(~NC+8|y5kbw;-WhXyLgKtwX5VRR?H%WYCnvol#T7`zzl2q!
ze+_)B5LAMLUL)t8C8u(volBEzj`)OY&=h15Gj|ePdP1!XlJllNqNb95v~O}kt$(PN
zCgT+k!g`8VSsX?j`7mPr{At&*qy@CRUe;5`lFkT;box=3pK~j#(7nN{EB3ph8`~bZ
zBm>8w<93xLtk#_Z#{;*1eI#&or>&{D_MoyZ-R6b|MDxb=vfB1)&i&b?3}wj-)IjZh
zQ?sdcsT70f*|&4>b77`N-;ho6Qh!GPAqenpxU1qs)NgZDg%T%Y)fF8<Px1Hc2s&Al
zBVgIKE@o^viKvB-gFdU1xMZt>rF#%HnbpTTG-D3VVZCC@SFV!S_2OCThc?dIOB#Hr
zyFR4qf4jH&1ih`_7W<#;xce@2y#Y|r68DdSUU6>;6awZdg8)?GHA-_OR(~~+eRO54
zSl)q)vissTUUkBL)tzm)=J#{q-8@|Gx-ckeH9(l4go|(ExU>`6nrk|)CB<u0{~Tss
zPW6<xs(;=H=b`U?R`u_6{iwT{)r)jl*^S-klGxi;-Pj%wGZ^izX%8L6?rPT_JFs3j
zbX^_Ap>}irDpt$tD891lSAT3{N%Zt`(ghE5++X6enC(k*QZ0KZ#&oFLQ5=Bto-$+x
z|2@SIub>zNJ_E%KT;VZQP{TqS`P%X-<CRW|$4&R7rNgAf50V^PNf<pb;J-@Y6EYe)
zVI}8-KkcQ^ewh19*Nrz&pOHvG4)(^V?-jQ{L26;V|HIw>jA*V9{yDt-`6ib?164`m
zpLai@HM=t}Wws;EyZPB@`%E6~i0j|`L@##U)l0ME>g72lrv4N2gZ~42pV=;#!G9<e
zmw{p#6_?SdF(;P^Qz#A?Ty}YrMv*3Y6C`e+xqae<fD;G!{|CxzmoHN&9)HA>VOD_=
zr}y7aKo2;50;fT(P6mK!38(izPyc)2Txhy3bRp+b(q-u0(ziN4ORzKp=m{vg;QVY5
z5{wjd0bM%6(0u7$A|7!~&-}P1JXUiL2lN{VH@NdJ?>|2AL^Nkk`&<9j|9X3W`s=CR
zf<|DV;q>dDPyPW)VA6_{w12c~JpKLK>G!9fdaPwFB~R1G-+!FCk)IP2Ef5pA)=D)@
z1jT^6lh%$&|0pt%xWg4~8i^@L3{mv0JOkb$>yV$H6AM23JK<q2>CPIS$3S(A;k0c0
zP&^Xbj1aM(ZjR4W$^$nv8ivwjO7SzTOd`M=U7lhuE@)O=+&gko9Dg-ntoEz?JqcN<
zh}o|Y#!cdln)u|${T$CbB#QXO-=95`JUTRS*b~Zckw9j=kRW{ob~Gczj!djjz{EhS
zUu;QMyybmCWEG=pz;sO{%_-TI32G2bJC|`ua|Z3gzR%BS)JxoTAj}i_=128Q6r7Uz
z7h<_f+XzCvou54q1Ap(5ID#0$HVY;{+x75UBJ|Hl7S9R-l*tqajF51chFa|@Nvsme
zl9h6XbXl3LZy5c#sNldrNsO}t{~n`(oej)OP&&h$v^@phHCpGLrMR|q1TkP_*2Frc
z&Y}qEHBnhX(DS4SVdRoVxgRw(Ytw4@T9{tuhy5%uvS=bF6MsT%{8(pv!H0eJYM7|F
z*8WTs-=$pI=`DGfOu;-SX%z;>3>4_NLXnx5&n5WH8GG9f6$p-Tqx(=nk%&o(ARDn_
z+x>=EvF?GDvCNo9df*P@D7C~Z^mOKrKz=h)xl7QRBC6;ylXT~NA{b=aKtEf~Xpi^m
zoOTIsi6(8LuYU&5#XwUtQoC@nF_K45lAKg67<po$H|)StJMYG0Jm`?cm@^x2l76e6
z{*Omwr8IrJN9se1ao&pW*NGN;w!}dv2p;sXyUI5sKTd<AWYyXpYp0iow95M(X9kRN
zZycFdFVxJrVfWkm<-IFj5A#J>$3|5e*m3}&lgzjQBY)rrFj_`A#)kSF!6><4lFI?(
zeBvg#W=ai$CRrx-rzHF+>!qN`G7@oH^p=vWRJj4dYt*C%nDi_rOw0Mn5kj3}Zc|0}
zr_{)iRv0keky9g8YKVif<(&FJsW$jl^m{IfK^LE_F6L0_JKA!#Lf0nDfquWQw^;1%
z$u6ZzIDcG(lYFs|DO7t^qO?d!WDcQ@qe1Q|m-D5zlJU2wb0eE5$(dKu9b!hJI#yE~
zglCkQP#^O021YU>$^(@620{3pa>E?BeWM`EcL8NdWeH5Ul;Xl%4U@#;rYb2@2`=^+
zbSxv6xGf}elmkAJ+cEKgoWQ1@NJzwSFU|3)LVv>o<HC5xEF%#ay7R7|ZlqOY5ymLn
zrXmy;DgCV#@7wRci}|`Y%sCLjxOb;D$Tt~eeU*ygdsnGJ;wnen!dCl6fn*f&ceQWu
zK>G$i;J^}P6yvzYub{C+YR3Q{S=MwIADqT^*f-MHes?oz><7XWYR7=m+OI88p|VT}
z_kUlY`d0!KqG7`P7pI`nyPXlF0ofO)Dt_->q=I;SltQVQ1o%IUQYaLYN__DsHQ!t(
zSvZD5a6Q-3loqM(gQ<eer&n~no}ay+O-ft*Ow~n|XkW&&iQM>3Pjiowr+%`i!k*rT
zBbytVGE!4wYN|Z!YA>~HT?=N4RDx45>wkch2MOK5@z)I)MG?|=Kt$oQ(6$bhd<x<I
z;2}MAIcVjP>`T>a8L?9x9D-ZHt<sT$2*)MflbO_`%MPZr1S_vh^9aK9Vcd0n^oGu+
zLw~D0Z>1|eP4?hXia#|#?&*^9jQBFSP9N|i#rdFSS}S$;6AwZJAX-J9tLhY@P=9sp
zxj135FInF}UZ9*ml-*ZcY=TtQImMaL<s=kSauN3Un)qiUXX9~ly}FF@Yx^AiJdFH^
zqyEq<-R<kv<=Ke?3HSE-OW<gDZVH{Y%RkgaRb|cnF4xXsqy|QDq+{ZOb6T@X=>)k%
z*JB!4YdQm)NUlvYZ9=v*sLm|`%zvwdB(EZ-Hw-SlQmy+Jp&44!W7+ab!ZMA++EL`D
zQJf%XJ2|Lgbk{N?8Qq-1?l+fY<O_8M6}X^W=u%>7MqCyeW+2Je(uUjFyEFcp3ymD|
zsP0SxtHVn(Xick;k2u%$r!!L9VSo2E<ri&)OIpSBs{pJ@NLb48e5$caLVxY-dnBZ@
zJ1N?x8n@V`6_v{ob1tc@eyzf>EdgkC=h`23@7;JkIlTQdQaOd-_>RH>uH-=AW<!LX
z@nKPC`KtT%L}Rzn16sG|fCqJb-cZWc<XKb8d)cc~zAbv@hP**YPpUghY(#HXRpd~1
z(wAtIcF#FpRLoczPQY|234dO1dY4IpLZ)2@Jk_KdFEJRPrAO%F(Nl0i(ZlqZ77~*Z
z9n8EAek3u;k)uf@$X-^H2#S45>YjQ}cF;bxu470B48z_5j)~vDSW%90j*7nl<^nTw
z#n4nNY}tMVVA1HQfEgFX^%6@6!5qi|>e~D69dUF%vF*ygQt~axoPT1NkoMp%B-MMU
zp@+&P%#ucDG58h~l`03F$x%%L?OP&uYElDSJCgZYW7WMle}F7{oZ(%l4Q1hEkbHn_
zmF2z~B%Tq^R6}oPsd9k1mf6?k%FE8bIh3DdiUQ0EhNZW@=6Q--F+PZCgPq77v))5>
zf>_U9z#-|A-I>0$vwuU+cRhP*^O?Q5jQxVSj8)+jrQ3;bNX6V^dR6bi8P2hs27z)J
zH_*A2)Zjsh?9DI_Cpusx8TXx$t_41t==9L{CpsObuPo8&sE>NE9{8kJx*qUgWCDMr
z0On4~H^5w8&*XZtmfKlxVx}MwTg>yKXv125`@1~Z=w;lUVt?gX^Sz`dOwFjyO0)Z@
zR49LS0`HVc102fMDiNBsibr%3zjPGe4vqpsyKHB0Y-aEFT1a8rKo7n1O}9nCQP^-W
zVr@7V^OqbvW6N&2!4l@VBCnL4C8akDL3%Y5JHr%dRnNCO+PdWNq*wA5(?VN!t8IB-
z!anGZe+^T+y=A)7%wF?*oltlc^KEsye5*}k&mSq~n<AC@-Jc}8;xkvn7FAo;sGb%7
z;-4+htB-@8f`$IQQOzF&?`z+;NOpGbTcECOJvBt_IsS|0GKn;t|J~-T!92Q&pTFPU
zpZ)>gEu;3A!G9<e0x>z4F)S_<m&jZwz<=S>?O#v7?`##?n&{GLI*k7wpLYN5MiqdA
zu`vDpZI|Ymz)-+R2#vKeeg9|rvioDC+Id-hru>MVzJHyDna>r(Xca{3oX7%1u!<~C
z&P7A%QUQ@4l(8ztdmXAgClFwv<i~x0Rf=rY95PZ=ysaq2tdMc0T&NjNa<-o={(oXd
z#qX4zGvIKfjCLVczuU8M9>%-rNDiG)cd@D^JXV>`LwXiZ_J+w{fTyD7l|0Y{iBaad
znECty%$HnIm6ceyiOif?swCkAa=}DAqLxsSHT3qkAIfr&jAO-6XqtbrHj9}<yEe?o
z8TO%j%-w*`237$!;<j}1B|!}xjDHnhB`8plYUiIW_xV5ud>j}JF8>nO=>(%j%MeC_
zRHRR~O_EAF_tk57)<FEK#~=m6!4E{AahOYCK#xWc2L>SP(s*aQAEjyTbP3lWM#so?
zzhZz3h(?X|JX#cZRTZ*m$kGUGUU7I+J&lXKGQ8D(dCtx&`~vw%NaLkD=YMgYiht#w
zOmH!CNYFt}a)S<1JkGA$kW%!<(M?|yTr3&IY+kAjLU`zB($3UZ%31KXU**lik%Y7s
zdBb=#P&}D*FUL=#f$=yeU2eEwl{zI>qLvEi1m}+46%t4^-pw8vE9K_?_jBNF2<8xW
z-_sM7fF^jD0uxHz?w{u8_<t_@#;c{)(k|GZMCsXd+`JLi{Ot<fq~#^|gvwN^S&n6-
zsyL<WhD*8*b`JLWoQf^8%oUYXDGS|ZNPQ`eOKzxeN@c+{BPO-<IySgn8C1z-(&_of
zOfuVbU#(QRLM*kms}e0xYN7D*^mP@c=u8OlbUE3*ZMZVWncSI#+J8-8p^@h*nCM<F
zC5VZ1*b;SX+#S{<&6w4C_cd#z<3qD%SEz#8U16bAZjZ2JpsW<l3S|Y;@u9b3uU62q
ziM}^|RE?h<Cmu<@$cUrVf_?eF2#f4&6@!LHM>Z6L1H4*8j;uXlPgpeQAg*gCm2V4P
zZdi+1rY?E5eb>0H!GF+aRtmNmuQ#a4SxY+`x9mJeie)*$fn=pf9%j~Tprs4YoN}`7
z)`8_XhoZEkd&M703_J|?SnbhLkjNpjTq|&<jy2T#eB*FPxv$}3-Oa5+(IA8^tuj!2
zX|Hs}%E4ez;=3yK*g8WDTjIH-IL5fte4WSmRV6h=$@^5#SAX<dphunzyDekHKyfYV
z(c{%SR0H*)jGXsb^m%Yc+EuHajqCREh896N#gXHQ9W=zKIX0Y<?+sQiIc@W;dsi!$
z)MF%U83_l98#ZaA7IPSXV3S4=($>?@)uWO1#smD29u1>2Uhw}OZNnZd?c4de{duQ7
z8hSQ?_!n-{(0`E$UVh;w4FydE|Gt{EmlhC*FK#wZn@HBg7&lCt1I3L?wt@PYl5N!P
z(vs~8PeX;D-&JJ<F&IGH(8L49jZHjIKV#xiySJ~(2aaglc=ny!2|3@OI)~sD$DUSO
z^3FC7+k2N2gS=(WH&VQQUkHQ^hM#X=Xzc@%k9!GIG=HoC#;tu`1E4*bg!x5f`*=o&
z#+g)r-W(K)#SyY`*NAFGpe4^h^W|@!b4o1?)}=;b<06b4#Lgu^!MRd-{?nqE-g(X4
zOme%Jz9RC8<b4%5wC{>~p&U!DF0+kwR-kRRp!R>50)dV|Z2=z5&Yyf7i|B*#mPQe;
zdyjt#2VNvH8eaw`+$z<5t>s*)B`PkVy^ih(EPZ1O?NLLKu)A9O?nmbkAlCNGjR-8=
z=@v^}+7a;WU$~iT)Y8G-BD5N|dr5w#UTNJj`-8soxLfvkoWR=g@oD!TeevJsm%)E1
z6qnIg0u+}7V<;<skKDKszW1-tA849I7GJ<%knSFjf!w^9TaZIw#LmH9>~0SE`zbz(
zq$sH$?)D6zX^A51@v&GBigNlk$?4PX__q7-s&R6{8*plR`#J%=kn|rUHCi({0AV#u
zZ$GAgJV~i6T}Qf5O0DT~bnofcJnuDl>H+iy3|;WNcL-&Fj8=33T{e%Q{yIEmI_gy3
z#db?{yp?er&@&KiI6nXL?e9BTgq9-McpHDlr;l&DKX;=FMh0OGPG5fP;tSLuv{w_6
z7uv$~*T1K)yU!!p%1qLj{LjeguW!@P@>@YMNJ3FsYs?HqkOVDG)&`;DtBfMPNNGqj
ze5H%D$A|)d%(VRQ>S2{4N}3NdQY65oA&ikQ!$dh#6By;>e@XmV3yEJTTPwh-k+PCA
zxoU5R!a5jx>xCTDK=vX@OZZr1x(w+>Jkbf%?gl&wHAmz@PY@X;uCtckuE2cGRgtnL
zWNyYZ&S;7xegio#f?80ED48|%UZ3}QJ(ypBqcXyOKzMH)=UgL>XtU;q{D6t~u3(B;
zlGGmC>^i1_yp;DN<ujOe2$}*~Ly%j}AgG9wQ%eL*;c{TP2lOtl$4&H%BR8fe3n+|b
zhG5nKu^@tULuYfm-7AonM6e^**M(!aiKy2CTo{Ssx+<)<)WS+tk+$=-@e`VrbvUQ1
z;6BxV*b?Vd;U0=c1YGDppmmQe^SU%a2}_(9VV~C>y6Ym7(Lx75Qjw=JGc<+mUy^Uu
z(raA2lkpKt>{m0HJI%PtGoJ6fliU__k7g5SYx~Q}D5!5`8F^Ym^e+EuOc2i62gdz@
zqR}n(u#epIdUwnOdf{b@e4$Xq?5&e7907}eZpmAZi2ej;O-UU3nqSz0pvh<^DN?+s
zo*?p7cknd=WSESy!u#geZ?dKPxzlgZosU0j^ykv!YF6jdd~`Fjim5mrTM8qT@su)?
zYog0*R%6lH!q<|MLJ|o_PIm<VO`SXn@OI=Bmqel%Ihbdaj77z=#Xhx`<n&xNGHL^V
z_r!9=p79BhOP$(J*MyJX#HAN_glTA&_cv(V@k%r>8+&Oz6p5?!Rm!yRK_3g3v`}0s
z%OZv!UX7oEK&7NzwFg798=wUnF6!tyJ@2EL$H^-IB(az4{-V(-FR)ocagMQs(3<pc
z8SN~0LQ8k3xCOGV&Xp=}PK<JWo9E?!R-3L#A82y?Y7o4i%;W*9c({!Z_Xxc)Mk9@H
zY^kL3WM>t12Y~8KBu{^)#OA$Gd1%U{z`!6r0J}~Vo;9){h6SmDL+A8OMO4ofh>0&-
zatcBvP16B*?Koc)UkxFz{cO$w>}RpfDPl3Ub-!a2d4YhVv9&DkqLqTBNjGbMv%S4^
zNjMc(NNTbCRE-llI1k-CpGrmP>v*g=SZ%3ymmti&MhLSc3?b7z2cae;AgB?Et_aw|
zP(vV#3%?9QUG8mLTMff(+NhJKIu(bUY4mO*^SDEzJV*5@u17rU7Z(;LsSLUioU*F2
zua6%2iQ_`^{v|TiG=3IO*Z^98967T4iuW~AF{E95M7!c&izGb{>=M&T2eQeRIP~#t
zfrbjSV6~6ib$?Fe!#+*Pk&(lu>cUL_n_@FW?`&<_dC&+{fRppyefZ<s*S~%F7YG{g
z-#9`}NDdvSD*2HgsX7akQsVTzJfqNL#D0q9vZnMTSfNNcT?f{NFVhZx6)gl!0iv~j
zkF69)J%!S4aN1*Zu_H6+nplQNv9>b5?|man)mS8*su=MpQYmBOBaP=#@iAWgGk2v?
z|Cx4Qr`_j@XVCPQwsR;aSb-A+g457L%1sWbV$s1c9Kn&nY_#@@)<!BBeJmQImv)qH
z#Lc9_tt!hcox<Eiw3q^aH}UV-z5&3XYlfuMwZc#)<Q3H+CR!yFSTnTk{>zUt?i-Ez
z)cs~Udr(@SvD01?E#q~hmG+J)6<bZ1b9e5|srlhqo2Rzx$XQ2TSjE3&&K&>yMK%y6
z5k60yh*{TtQ}*-K009W+^c4V&X-NkR(J%lNn2o^9I#4OBwc5ykAJ2w%RAdEiq#zF$
zMq%0t8%8C67t~tLeOmF8sb{SZ>bT7`G50vl`1C$?#mn=3iU%wFX5p{BUzF5CLXvyt
z^fJdA;OCM>qU~!(>?Oz=<fujddR+=mVh(4px|H;+B3dfPg$1D#ltk9ZTWFdRfzh^Y
z8cwxiBBpW{dgx|<_{gm>Ju7!trWDFuir8wF^(yBF^g0SxL~50?dPY?(>Zn_#_d}Yh
zhB|D9*RfUD_1<htv8XQUOy^Rm$a>nIVQpfOJMICPlIx`B`1>*>;j^r!)<gM8b}NuG
zud{UKRriW#QD_CYg|u}g@#@D}j)k7f0c`UR`tZ^fg}o<#z{So(=z9#BV=GU)(lpc9
zstD6t-jZCc2IVYRb&_9`F7uqmU)xq~{>-vR>54aQ#dNl%{HAIy+xBQ4SJxB!wPSf^
z&7&qC^E#+AxgH~wK>>iHjJ5o5$)hpOo?EKFWMAuat(H|qym{Su9j>mZ!&9~3%~(<<
zK1+Gbq4hI=t>0|7MK#p&x>!Z6YH%?&*3WW_`7T)jui;)=*z^+Jhu9QH!L8}l_VHAG
z@-@mmYVR|CEc6~0n%Fk;b&o4bsOB5{D_c67_UGdUvKl#cbN(ousH&mZxlqZptw!Tn
z^fLXbTZDPf{u-6Eg_K{8DxK!RiI+sT$$2jSZJFkOTK?N|*_{9WKFzz8Qgz?es#lz)
zQ67YsYn$wQrEq1hhV164oiRwJ?76?WC1J^Q#)KvF#7gbjke_=Ox0|WVLU-mSXEV*M
zLvqpws6vT#DcUa5mn^ky`z}QuGa{4vFb~^O)eaHg?H!enw9w~K3x?`<L4-#u80P$%
z6=7t5JbU98ltR%6))K$EDi&601F@(^`FQUK<o%=`_|%AJi(^>N*e`zTaa35kfp+uf
z9QQfbO!pt_{uD=L@T*^GRt{D1$SQqp!G6bVw&9v-T`+6@w$v?m{+NnGg|im3c*C*e
zeH8H^oxMvtph}m`eX~er``}%ZG~JSHcqwLoaO0UfYlNjzc|39#sP1;*U5?x-s_m41
z<dM69;scM|4U{h&xf^<K9l2Aba!`kSukFq9DFu|phLPhCUdJO*qkP?mt3}OK?DfIN
z+M*<6-;bR8IYka7rld7>C*pNpQC&QW@c>;VWi;=6fN#e0&SQ`>TGt+e{ON-%(E}lW
z!gwI=fk%E~L^4Xh^Nb8!uRm3>J~uudO<m}6d3Ji9xmEq#;XGx4a*R=cJC7YIOBq-M
zb)dR_?oe4T=vshqhw+&pjK$@LK8A1gY|-{Xp9dJLcEX~4Z)2TvRr;z8=ygNQ2T2Et
zmkm{S8&UH1z^$|w=-tToj#Hirm2l)!m_~~0$9v<3VW6yO=zXtqR90B;nl{7Hvxi$B
z)iLgoQF<VFvBdZ#XC;FczT2h<bB2Q<KLSQaRbbE$gmF&LT*%)fUIF83qCfmnA@P5e
z-<9K+!G9<emoknG6ag@ovHt@nm%eQ%mjU^gR&FSXe+z$=rDgrdq;5u-)QK+Rl_6hu
zVG~v>n8P5o+l;))MYPI7Jd%q8#DQyxW5yMZS@;~Mz$n|qm0%Sk;fD@ql#QsIp$8bn
z?C=bptppOHjCC{e+ihkJSdm24Y}r9pH&>J-?ZU=MNgBjJ3Va+PUR2=s*XQCg7-cbo
zQywkCe=>=2^&?ZZ8R>+D6qE^m#0z$JmuOJ2ha2C{xW8RdY+#2QOoy9k!wmO$Jz*E<
z-H~o-fS3)3I+<(v7Jbm>4!3jwoy3vAgd^!3Abvl`VbQ4&;$5C&FM{0)@Fr^I(N_56
zU%|yJI*!CT)Yy{Q_4=HQ%u<@k?*xfBLHvG-f8L<e%v)KPp_GwongESJa=+)gIy;6?
z=d&;b-U`P=StW|oi_F>$wzH_&slo3Akk_nG2XWQ4_FS9^LC#x5O_Ea~YPcw4MyTo?
z1-h^M7g3`PoMo_!;U`cP_cXdX)=5MZN%0Ow5oL}`yw8-NaG)BSt@KU~_L4Ff!)6<3
z#I*9F%73%LJvDv^FEZ`IS02l`VB&n1Ws!$zdJ`+}MWfbyVNy>Yl;0#6E!FW14y_0}
zk3bu#s#X4-Wh=f@RhqY+m3)P2Wb^r7!@Py6FDOJ1l?@ehT4mZ^(m%Hg^^S<Sp{lLx
zX>qbEtRmQ8@M4xLm9T{gbn48eam{r;T!c5{&VS}45x7W6R$-4P`~1yF$BA7vaO?%a
zY)rs)2F%o<*Oh;<|Ci~=^HtS)%Gfi@yOp1^vXhLgtY%eRn=jtD=_3^Fk@c>33-j(^
z!l4J3W!;)kXF5AA1~|^i5N9uhyBt2&$xsB&u0}+?@d^{{hGkuhDeQ$_u<s?J^x|%C
z-hWr7nXtx;0^wd(i$9+f(i*0)h2<#%fiE5}t-Zv0AfpKV2n%@MsAc7a;BAQgHH<4I
z<-Lf{BwtG7DumnkxZ1>drzJl4%6WSoyXQR{H(}Seu?MSMoK`joiW4sK6*;)h!IW4e
zidzFfbxyMLm^%)zFKcotVhci89Yq*jcz+q;Uxg>9HFwFFtD_3l@g`00L&h!nMXt7u
z8TWH%@_C87sf<CIpx%n>VvpDMiyXU2;|O^tg}u197?rVpZ*N!w&RetWVsn1ZB@4%!
z^SvOqtHsJrD3evL3-j_j^}Y<yOX-z0*BQFxz!vMlNPXzJv@_Qw1K2{%uCAg{(SO6_
zU26(Z<oC8_^~FN`n2c;}=3b7-s)-|YrdZ~6ow3N0rjd@du)nS^&MVMnT-{m$pAVPl
zQR(E1wiv!gxJ-RY@Drw+`8Mm{k3#Q*ol;^)3kYMU6nQq<FKLt-jf(7%9f<+rV}d%g
zl27>$#nlZyR>gB27?h3P(Owl#Hh;D0*PuGJf7t|0=$xSZi`%eSgHjDu<*gQ`YVPbi
zGrrzkAR>ipZ*0?DFqGR`rq+K_n4QxM4gRQ%Y$IKJ?<x_6cv)|1Ob|x6uA9&mfQHOY
zPFG>vhqt}I!@T7Ajl&lyv~0~l8)PX0GiI~Khj9n@<**HEU#7J<v_{grJAc>j-S%Z)
zntkhcD23Ht6e0wLA@prPR=T5nFaT$x&Wu<nhzWOD9`10TY~{gSei(dI9^8oSz?cHr
z?M@i*eaIb-p$Cmq*2f)so@uSE>~Rs(ArDT(N!u(X$&V7ieiVT&6VO-V<hIqkBq~n7
z3>}2y#F>0CQW*C)IIp`>0Dn$4*yHb+(@MoAZF4ZB$9;<6x2gLx*?7~5NQ18Su$+F#
zk(5-}+I%*%?eH98IMv<gsEBY)kt&VCaku_A*il@Z?p;p6C8NO+Q=;B>6i+3B#i8{r
zBW~M2Z23*(`1n%a-1LafQp%L|L(j2N*b1KfR_$QQzjKkHK)8@b`hTu%JM18Ac0c7%
zX_j7g)7`qR7GUe{psvcWc6y!i?&S7nm^C!Y`1Rnt7qpPeYg&8Ftw-6eHq=wAp7U!9
zMC4bCDX51Z;R<y!yF;DK^sR(*XP@af|J?NfA58QkDKPVmh=>X~GJHj%kKQQn@W4cW
zfH)s1K71UUj2;}0kbk9HH62>>Th7DyHtij4o=RsRQHpQnH@U};t2rzf7e|Z$`^VL2
zmB)$mbmNFNcPf&q{dCxIY^w2tF-OXN3$b+|#NoGQ{Cj^S^Cd^mNP5$&`<km51J-;M
z<F-SbH99bq3EJ&z81<^cx9IW~SJw+)`SZwy4#TM!PEgIhg@4HPkCEPQFZmH7d(Sm-
zJN^m>#GthOV)fc0J>a6x7T}%te55G-#66!aTKI&X$yJ~M;H7&*PDPHX+m}CiZ3qI(
zE5f$N{vEFlf#aIh(}6yGfcSrMbts**p|wZSIY8Vwy?MNoA-p)f!b=I)pHU^^G%$YE
zUB&cG{>V$60~y4}s=><0PjN7vTH!a}*c!h%gZ~#8f;(uJ!G9<e0XCQ6jXe~XajY~a
zm!OU<1eZT{D3*VoFTMZo^S?JT3;phi+~fr+CC!?|#-d<aMdBt;zSsLL-%5VP@=IdD
zkQpHhfy|6fEC7ayAl5sR`Vo(#a3A_(ezv7oY-K2jnURblUjx^W5^)wu)Bpe;lPnZ~
zkb0Lm2D(@{pNcf02L}1v9&G5zV~Hn)Pf5Pm6M9;KdK`ayC}RUfQkB;Nkv*7QMkVzq
zGP+9h8akCvLMO{A>-f`bE_jJ<z;-q>n-mSf_xL3~3U)I&ajiAZ*d@#=x&=d5r9b5H
z^OdRCVWuBsE-M(N@G~m&^$i-b&<K=5Q8hO_<n_QL5qv0S79|RlBbq@2%vlX6y$b(^
zLQf-Ntw(<yOV~5UNuSyDoFH$FUe#w%!eTW&y-KmzOEkv!ymkh69dcQ7u*|R7)^XNC
z8RbKsQ_uuC(1^C_iw&K!Ub~xYlROIDRzkaCiOCu9DmkYnDUdIxNl$;Wds?=B4i3w~
zDkp`4v>!PYdyjK2YoLl>X}S;*eV<pc^rei38C-wn`dn3X?s_3X)@wy{#wXb96LODk
zb#lv%tE0#)S0~eu$7{ZY`<yC`75|cDQBmgn6#PpXQ89#!Ve${%sPq)6WA+l5q%3}@
zXf~Qk{~oUuy~xkdtfIqC3GM6t@)I=)tPzB;;8;9lT9ble*(FcaoAve$ciDI9J!nuz
zb*6t4)ssoVLIzm*@8j_F;d%4N!Z9P83NO|4<;NzxKn+6kdH``GBGvTguhZAfr;%(1
zO$y0l(hteBoU@kCITZ7fv6aRU&9DVtal(n1mvwb(c;7azIQ0-o!LV3SJ(dJ{lQEhG
zf-^ghgfmhYq%1#C_*n}JUkMu{z^H++k~4p)n#!Os7KYAxAw}{8@)UQ!B++#^uVSgg
zvg$j)WXLR&2R%Vx6uHh?KHmWNoU76~LcmVngH2M6Kh}YB!m9<ffD)~t_q5-pfSP^+
zjzGdAy)}+=t^vo}{O&rQ&I;6vxm?Gy=PB>E7KI^Nf{`!Ks+0<i@a$HG-gLa5FA;yQ
z?qa_1v@q4Xs2V5uT2RIH*;pk=sRibzJR9EM+E3&=Yw3MX(L{;DySnqVwJEC=Vof7e
zLO9<sdg;ztS>NJ-!RxIAo#vO1bln5R$Ha^!=Gw#VA&NZ+3Fe|=5521@JOF+J0_c(G
z&j~hFd#UoW#~0w?0m^G0ue^#4+v0yC;hR%@ow9bC`f}%iV1;fg3MS`^^nm7CbjRW1
z7%Xb71)3Nv{xNaKk5lZ;1hjWF{XK<d2!XsYaqrU{Uiu*F=H2h#zW(dWFT4{@-~I|C
zIMxgq0q>?C>5=%X=Dar3_xx;x)z<Rn>7_R82%Rtp{JxB<4PT}WD$5%!A0>ZU>-W&g
zQLMw@kL%zV;M&kJ)9+e}1c6{}E$qJcjYX)&A<n2u5uO5-JT^S?_%bRyhO2)rU3t|1
zrp?!B^Qofg8FwAS32D~A1i|a9p$D`yX(P~kvIPxqS=G+8_7km*N*HHQN8Pr(hlF4z
zbE}grE!~7m6VoabN>lkC+s}UhFw8YUlC+Yyd-g3zLa|-4PC|w?LEG-X{K(_J@u-j8
zf1<Mm#cEAv+G>tKE&4z!o?FPR=QmRK(z`_XhrOm<i+bZpuAIXUIV+YpubJ3DltuV7
zbt0y@Zrfx(-3$=>0?GPw9iZk9Za=E8fQXtsFjNrFC5iHUIcN+2|4e_9=&}GYBkFKG
z$ETKC&+)1Ec#d~xbH`41xt%rcG}~HQ24EeULh+vbjZrawV`?V#-uz907>yUw<A|}o
zMS5K1S69zCMUAr3mCf9@OqZ@aa!bpUryR5GR4$_tv*s;7e92y$wy)W1v%O@mzfJRQ
zd1KvnJ8SVg7xK|s^67uV;Enlc>hsamvU2e~=);F}g0A|K^3WI0@72hzWt$bBL8%45
z>XdVq_8y%Kc9=&DwT^*~?5YH(oO{P+XJMtoDOWi)=m<w`L%k`E>*=|EEcF$c(|yv#
zd(-evy=pc$Qz|k=3|zJms{Udz3eR!MS>T4{3<KBKch;%Wp{sxW%=%Sdw_fv6mV|+p
zi^6Jg*D7m^ET_p993-FDeA3*?n5Z?jxA|1vs_&`Kqo&M1=wn8p)ysMrQl7WeORJyi
z`Zpb$Rj~>>_7T}&ds8;plwBm@T&j8v8(8Ak=bHPKKwC9B(rt2dmNVv`7dk*zbms3<
zh*Vv(^2_pJX*GXkIa`ThEa63U&_R(^KD~uBtJa}a?z-^H<b`#eyEmoR42sCRI=fT1
zDcP(@bxc$Y?#~LO(y2ASyY4!>nMW!~TCDB9DH@exW}@vnWKv)`qWHLS3xXFhQlBki
zSpCTUcvF~#eAu8B66D(2jO!YUNaT{nSXEZ?xHj)4YC(UqV0Sj{yiie&sjZ@O0U!eH
zIt^$oPvw(?=9KfB&TvypoU6{8$-FORDsC|*>qdJjI*L$>i+Phut+c8+`HPkKwtPrX
zv?$mQ_dZ*uULXK%j`i+}Gw<XdoD>3fPPL;=vl$g_mun8UM+{TxZFP+~yt20KMA;44
zL7|d&!|Hz#%iejYEtjhr#nTRTyy756Zu1?>E$WJOe}eJbkAAPTXO=z3uUk|BX0`VH
z$l5M}v;Fi+M8t9Bi>i)39Ao{W=&0CLDDx&o$8h=sdSvo+p!mSD<3RaJ-ErvsMukUk
zUa0nU+8Y(j2ClCxJPy?Uk-{UGkg@e|jpLl_YjJ;^Q@gL=SoaV$etP(-OgNAcdZmhX
ziQ&Ub4xmvWs<&R;IoAhOl?SSCR#hJOUa2Y%y<bsP9{E0<b;PA&Jg)W~I(LVB8YnFN
zVm&^*xZ<sq7GxKCbEe|06>sT!!TuGN8Zv86zp!S=$LDf)Bj9gQC<hsG&E|p8#z@h(
zGtz(A4-D5l>I)1}hDwkbC5WHBE^CW*lmdaqkE5!dtc7~lKQm|1VDXi?>+>}?dbzV@
z3f{_?^uTJUjGvX$>2T;B#@2A#_KEy<AaLCWg{OIFKM(x71H+{$DCW;?gBT$z+!;;s
zX2rnh;<S`^!pH@|VdH8k^Zx^H+WUJ7Wp0;6%`OiDH8Gd5!Z{O{7vCf)mhu4vmlc30
zn16&w@$mD@k2}dsKfdgGCau!KS{Ock+x<CE$LAsS-btZ@H^V<eY^Fif*;+?;jR0Zz
zi@Ln~<ICqCKK%lM2K*gI$N_~i5d82hKN>5fcg_u8`PmE97HD=6h+iBm(bIuCl!3M3
z)38J31Zd-%)~~TO2=u5_GgyPr-ciRoL4VU!c!(5hYnktB+gP|7r${nYHsVvH!eiqj
zjW46(W4zku(v?R2HtaqRyAKu5py`k9)=9erCkO;b*3&>*nzR=X9CJ^+1}BYKX>A3q
zjY?Q=QIlw1I5HAF)Y+=ArKJ(LG!d;rp){3y-@XDsCu@cz+ocgE&^Sx1hG><*uzzM~
z$KRj6@whKE>bbdRI(s0)Mq`IP0W3%zL@N2L@rypoy}2{HtiC^JLw6)sGvt_tTn&$W
zzfZ1J=3R2JDNSPF!_e@T1>KL5ez+JN3M|g)3(zTCEuf+exylI3N+>HLpd%j%HI#*e
zQ+JHT!$HT8gA~fpQ3y)G$%GL#Jbxbcz$jvTe#*T~3|J^Dz6X`*R2qb`q2OmC)f5iL
zKJ;5NrAXDz=D%m|C=lX986W#4ijlJAx5&?~<tUAR;F+NES;osO5eXxWAMGOXflPO_
zlS0!KnzcobGDW%*bYwfus1d7*26GcfB)om17SaMhJY$IcwQD44@B_*A+ka`A?SeH;
zG_u4rQt?Rnbl9s<p6bjrqw=<qGmVQ{Nm!=Y-aUC(w6kD@q8Q!JzN9b3ch5d4news)
zD@5VKT&`o*cH|iJgUWU+s~s096IyJ)(GIhO?dXQHBWv%e_QU#%IZhUG%wy7BG3^n(
ziRa9{ra_-A7u|#tbrNZeA%BF{`|FV2=k!9pkG~=h=HVpz=(PxO0bfU-r7PGKukkf1
z6nkDvWIDrL3A1C!%whKTVcLz;p?#ANmF2<{1!whAif4iqO4sE=u}c^CS0#9)?2cds
z0uC*f(4p8cj_hd1(svpvd~w7QYV2%|tb}D|l~I(nb-c>^Gh>jJC4bco%JI>kjieJX
zF7wlhsH6L6DY^wh>K2+~vJS}JRo||uKw@KeBc&8q)4K$2T1T>_QR3o#%m8C1=e5x>
zNn2?B_`K?CieyS4PGw3S?f6oKkNo?znz>KK{s4E2@VBnA2Dny~qmgPm=F-9eVG(+?
z9IQ?$KKCrkZ?ib`Du2Kpin4%bbyVf(-XHSF)BaGXBuK~EPKHK$7KW1XUR{%!P8zY%
zmh#@25%)`NF?DMM)g6<d2(|a>QVFrq^`i}^K&<^$H=`&S3yyy71$Ryh5RQDswrCyR
zrONb}aYpMp{h;+t;*b|arhY$7i}F|+1Bu21Tas*&J6`b2+<#i8Y4{pi(aI_OcjAIM
z2@+N3Ld%(gib*bs>@F}#?p4IJCn$N!Nnp$vqGM%Pi$U|qHXr5%CAr*X;E=J$GVPV<
z7(n4cU5w76=^M%mnsudQ0GxFvQ<Du=oaN|jD4aW$ph1gjj@Xpu#Sl%m=kAH}g+FIZ
zXs?9RCNA~yMStE0P*h`4^)J=%*RR9p-G`nmM^&n^QlS<98JU-cNy|sxA_|f)l-5$!
zJ#k`MJq)eMG$%H0_?|Q7r!n_a(1r7)%Q;1)B9nyaC{EIhM47R|$Y=SN*q^j~9BW3f
zR)AGKWhEzk)!q(;b<p?LGd`$+v@1zUcx0Z=LwXiZ8Gk6deFdCx|AnYS+_8{6>7?c3
z1&~iBRr*{Gzk-`5kLZa-h|CgnPbC#M+k=SS35)8+D0Z_(7%y9umf<97u^Q;8p2xGD
zQ@-9=6jDq<=;!5-+H<Mm<`2b{t~VD|ubS6Tog-)R!qd!AUq#h=LD!5bE=%K-FiOoZ
zKe99?Zhv={e`(%HOB-^EB|0mt*Yhl;XbWl~N(hf7Y5wJ%tZ%WyU;{6q3dzwbvR&xz
zDQ;FKq+|o`=KEdI9o}m2#YCpeuNizXk!kO0A{(ehE8)>?sg{nSofO?ynKdP)-un&A
zT7DPP)**f<fDJM1DAt`_lf!W`^|Jf(e%ymY?thbVLc3tE*M}DEgOD(WG|8XeVgaT*
zM3)l4bNd*YGF&yTW?U_C0^3%zP_|QYQP8&5>NWRi>ae=05(;=NrAsQg1ZN$!%+bj@
zf8@zlDr@D4Z`1b_yLdafY^L)Y-+U;rym6o3m~lHf7X#O){fg?BAj2%}SC&GPy-C_1
z3x7=Ka7zc9Qrs}@?<k*1`@7yZO8d3+biZ}p-_82Q#J{8VpJe{ZQ$DK0jdAQ|-6-?#
zsNO8|@8+E?Ca=lqL8uu#Z%*lv);&F=x{|MPcGva3ra+-oKAytp)$(|~B;E+RW|H9b
z6oVEn;Kqn+Nehc{7b32sD^gz4<8*zdqko(c;p${lVQJHkW15t`M)8>?l$)oZ@%xjs
zYQq(>r})3Bt~%zyo~*9A-Wyg|ifSRiEfbfX;@hh$Mb#B}v!TJeiNVW!NN;BJ%T<E$
zLlCHxv`bY2xFBq3Vpll;$iIn~IA5Ph&Au1sLm~d`@qzyHZTiq4$72cTCu{4u&wt-w
ztWZt1XU9sW9L+K2)F<QR%H*OC#h_jL`9Zq_*mUwo#q$wz@e@0LXgby&P1(a|Jt{D<
z6d`bM2^>DXkVlLsWJ(LxU&|*N`0ZH8F<-L0nI$U|?x!Eo&P%c;ML9=O%FLd7gY4ZL
zDvPq}Sr>a+64{>g$TKx3x_G4JP=BlWD!T%%If-)|PvY=NnZoJPw<S=f{kC+e*`;$e
zT7RFD0AQU^VaD6|(UupXSL}<7`pq_&uH&Joe!-j1ZC#MYgp$xR8Nc~z?`J*+6B{WT
z5>Z#6dW%dgSae;iq&ZVoEfOxX-J-AWbaDf=5t?FOudg@jr<Elq4VUZ3=zl;=nFak^
zwGTptfL7RVgJ7WaQ{QA41f&+|w%n@iC~mk5(osIM3)1zz(JqJwsxxj~yiX^6;~vPA
z*kxNF`tkKybv*iHLpR-93P<utr+N*P4a%zs#{F{N6I?0MH3~|jd;QDKz!p8ZyHlX}
z7w&96tO{jrWOH<KWnpa!WtYu=C<g;DHZzw2HUc9IH#sj%Wo~D5Xdp5(m(j;BCx6{q
z+m0N!5q;NJ@G;2)hjk%Y1sDQsD{=xPfFS81iLei{JP3pcyT)n+D1V<*?CISd^^Ucd
z-h~yz_SST>$?D>%Qx|qEjZ;arR83;>w<hWoLQ7IKgq~Cygoev)jIq?pxMHc}fm%8z
zi`%8uWJXUQv_xeL>Ln40heVJt6n}SnBBq$VLy{ij4CB<|Zr4P$3<=sa87nySg05<o
z$V$Nyc#2G87@p!Xlr@Fqn$&TLxW^@e)M9;WiA6$9Ov1{9VYsAq+^mUfiJS3gla3X<
zC+;e7*C7qny(IP$DtU#4#*1M=p&sK(5*}~u1W92nDw0I0gw&H%8g~s+)qhyRlQbxX
zyLi@!L&CGJHKdxf^f60IN;Jm72vp-5;RUc<NO)uAan~T}kUVMa{sea=X{S&R!}_=h
zeYD${1P^$PNqE2QxHc?4B>ay{jEi+-Km`BEbQ9)PGOQrDiTA-Z!7{Cl=^*u?G2@1*
zhlGi}$D2Xf7An@*q{gw0#DBn*^f3#kR?8R%L1GDZNEQ>rGHsd|b2j666aa1JaZP&a
zCG;UEi(z9rPo2kIEeRVsW+7m=Nl2V}^M*CB)!nAMn8l~NmH;;jNg)Lq2!Ju_p**HH
zwotT6GLg^)b}((@8oZ%;SPy13sDL%#PG-YYa2RvzAtf(k9|3SLV}Eb4wg)6?Y_-#{
z5B$e@paJH0u`x+cuCccSXY{J=$(!NKCB2-yc`vkC=pmdu`R=>T>;L|`-F$y~dbZoV
z`t<WIe*NwJ>3`DZhqLovw&!mOMwXM!Up9Z;{P>2ZA1}A>c5eX2Z15<Ro<-ns%vl0t
zS~LDT*?gZc{j1GSXMeBHHqX)@8lWk)b?$Hrm8!-6_!R%#O9_^eExck@qSq{XT!|h~
zXNH?%lmd0e_g6y2vR@hkE|4+#Q60ErzZw6+T1Bm6AMdXNd^A@$o3dmBs;QS;gAhcr
z`a%Z~KxGBMox?qwmP;L;1$I0DviZk9|COeFZtxvSr=LE2IDZ+V(&opr4`=7Eetr9H
zn`FRh^WyDpw>>{iJpKIh=iN`Qc5ipvMAL=mXQw;N{~WfV;OJQ4Ifxn5Akz=fDh0KS
zAMqAJXVVXj9|N{epKe~9pS^pv-M!hoc=mkrdi!~Ia>aZv-u`<F?M`>w)7{4iGt;}g
z+<rX!bpCGpaes<Dap~{dU*5m{;q3D;YXi%`94C1G98Y2t4ZWxB_&CM4v3xjJn|-Ra
z6<kR@M@SU&jIbk`gK)$x4~4$iyN`I}{n=?I#zQN(P{ZOt4w+;)CgPfAxJe?HjMrN%
zU)>ABX?5VYz>~D`0d$&OKnGNHh?*}4AQ%E$U~-_eVSiH|8ed>hiWu^6VC%L3TW@y(
zo0Utj^_j_gT>)&t5Qv51HFrcf>evGiM*(w+Ew&p$wM5>)hHz%qLHR3yju0#uAQQu?
z0lEULYh*l<#pE|JuLp4J{VPnh95rE1wPZ&2_5heWK-K{OkHPE!nzzNA$pUjmH^<zi
zm5jgQx_?&&%w|Lyo{?iexa%Ib+<~?4W5&KTW4)6Z+f5*OFlrA;>Y!kG+%S6&i}@0G
zek0&{J0v^rmt<$YCp#~*umbBL>&ymyk#}Z<u8IAkJZnhS$}-3@n^0wQ^6u^xrNIh`
zA`z6Z3ngk$5U_kq2^1<;P^)5p49Q(BQrF<#eShLE%E+A=P#~<1LccQd)qXoI0?Gb0
z+VCZ9C|i!&?we;{D*B}pL~(w5Cy2(tg%h~V*x!3YRDpYe*2n>{)u@_)LzI{(K(Znj
zYeQy0^jIl?cy7_;fD<%S^UP4Dgmb_tq#sKG{)8OBR>*q62T`}f&2zzDU>dFnEgVkK
zmwyR}IAb{XXs{IIzp%&sm5@@hFk)z}+2hr%4^N~9aw5Bx67aKVGJQSp9FCg2Q4n~n
z_WuKKBwlzUmAiN&>4i5^vDx?h{=O#>1gfT19Diw)c@C2`f<H8R9ON;aOrqi}aQQ0H
z<jn#q5E8@yy&VMvsP^T{(HL-bz(oweqJI@>(C?zFa6i=f5_SD9sAIZB9qn5T(SE%@
z<D<&!CAyAzG_-|sLy+;FoB1O#K3mJt44qUEt(n_#6}+N4)9UaMb4BSZhk&!2a2+js
z4S1{UMvu#N_rR9W0$ckI#x1cWa)~XLJ+@e9VMZIbS?IH1>vSC)yRE=s^y$es$A5{3
z0u!C>R@;Af01}U+Q@LiO_XzryqsgyQv-0sy8xXSw=OV!|i{l~#(aT6ZTAT}!WxXcl
zz&4|_mlCyy7Q!EHYvdf1<9KD(0-l~DK|iKNF8~gyAn(-P8W&ol5Q?m;#A#YIKI1Gb
zdXKP(Apu<)q8aKELuCjnKml!Q7=HrL&nRmWw+>xU1~pV9an#o$JTH$gnsi5g+FK`w
z!Kxq7k(95*zbgYS;$QO|EtPMDiL@=4NcApEq<+anGSiLBbR%=NC3EegbM2!v3p@+5
zLPmz1KhCuk>9=*z0g|KMJ?RKd{jnS~LNcroRhT21i6cd;aaJqPrah7t+<z&<5~vR&
zMhGKEjk96s=JHq+h1?05K_8UV>Nx4oadM4xB=a=~B1pfmT|(Z#Qf`PGvI1R`X(qW)
zg0+Jl>Tx9$OzQDbLQw@n6luD~;mlMlTxbEPGKlk-7FE@#l3No|L{AhAD<Cr}@@UTQ
zm5{)Q{>FYh3}4S}EBsaPTz?eF)dD3ATIj6;R}iw+n<S72qxQuz7ACWAhPGQ-&~~-E
z(003064yC_oWC-l`6~mOzcQdzzA`}GSmSdGC!}tO)>LxisGgC=)aBtiK#v!()|pW^
zA^)$o9`q{Gm8#f@UUP|$G0d~LU(-AW^hz^waFQYCBVOmoOs6hww13IK0?;KI*m3!B
z?0{M@PhTxfI@D;{1v_1@!$rpqe7HvQ+;W%rJ`2W=Zl!4U$RQh)&?O7G6i31a2JJ|+
zy5ZRa|G)By!}S^4!8Z{5b80y14MuS=QoF%O#H?H&=D_FRsQXwu<hE4Om)*5~r|#Nx
zCO%O~n-jk_Cw^^(Hc1=9XU9io9WX(mDd9fk@eNw;z9f#y;B+oM(tN=~R_h8r+X4xm
zg2m$Sp21>BR;C^VUz1r@WO`D>kdA^jq_<jz7XJez#Z`5e!G9<e0XUZ-nimrSF*TPV
zumvW6?OIEZ8#xZX`&ZN-up)}o3owjLJDmV~OVGE)9tOdr_mC`hW)AcBmwMHMRc@E<
zih)^d1J$llNu>0Nq^M7&yVp*4pLS0#yLTUvI<JBSZ@QOn9q<dy|3P!1vz-S}4xxMb
z(f#9~wQ+ns@kLrA^5ywH@UK1Y5dzl$egl?&FF5W!L2ZMEFW}3SHe4U$p~x|(`Yx3_
zmJ^&#?SP*lCTYiizWjZsYp1n}F5RXd`Skp<`)wzOU>zx<(0%=}OD_mPVK5zlRo40L
z_g}kjyU$W=uS|2B^3Uq-_t#F8e5oi#%_wT;tev0;nxUo6xu`I`Dk#zitu3QQANk^c
z(NjVJ#3;FIS^;*DbY;bp5?1`~V}M1i#C6e1t%m1|7=O!iVHt0m?ipiiOAk^{a=D-M
zq?a1JBAksyvo)}QER|Rq)p6;M$9<W{u}&_{vK1x#+P^%yNy(Q%uOXw@kfdH=6+>;O
zB2S^HZMm2De7$bVHlk-^23po_qw+R?&SqO*uh#{|*9`0DWxd`h@BLO|4p#BC7Vi+%
zf`Q*K<QSBu8F;{^WVCRN^-?i8AefaCh#V5wn`W?|>S!!>!@sO|QZmxpJH8DH*&G`#
zy(={5fDh@;6oBhpnYa>48Od78lP}-f(fi>0nqqWoV%Wt1hE9<WR|K$&!D%XgF$JR(
zI#NKQ7NATCPAR<P09Ccw4>ioxx((G_7SnaiapzQC<=@a^TZvdACe_Kf*JGwby5KPa
zQGHWzL~RsmoT<iT`_@ib>t*fUzBjNFAYN3rG?S^o!GCV{NR5i!WCM@<xgKW&CtiUz
zgi=Vn&%>m9Wsm!L9#+>wF@$q}9_Ed*#>^npcZG5unDS;=LNMjHGgn(qBRMaXSt)Tk
z<!#PNf9ts7r!LoK0T?-!{Mn^qZp4*cPW+GnV(RMd-G5%c{q5^7oCbEUf20<i)GA{k
zhVDmsw8SZHMAv<<&zvA2X9%Dor`XB3lq2BxtzBR6wcC;LoMqZDD*c{+N~5Ac8P?V~
zXVN}Ztnx=Ctp`b9U)t1t9~ZVx4I1)@`d*}`B&D`ZkKDesN{{IpKewvf>R;XNTeth%
zqZyI^axSgi5}2R}Jnwf#fVQF(G-KrOS<#4|edkQ+c%XfuVQX}RJhY%Z$BdN8S8e@Z
zTghm^t%zkc39YF8%JMUR0En>`NQzZ*o*TJcTF8<~LPfP8JKTT$QJcMTqfevXB4!T;
z6>+@YTSTQX%gXyBw)CZ4X06_8SKq?-N3@3$l5%*0VRu6Bv%fXXOCBbKau+^#qb`=A
z?)$NRUJVaN0^Z54z-NdCh9@-{bTZ6<E!NCRzbs)$kwZl6A<WT#{XB%4Hy>mlo==G-
zXqing#A&d{w>oKg=Wnb#)Gj~eMWvPu9Rhr!#a^_wCU^I&qydH_<ma<cq!#wDcE?=!
z4$^oi97mQB>RNNX4e*4rae9dtj(YFU^`7duPt#OMtdSNBuPh)nr0K32ikD4|tXgE7
zKHXO~*+4=ffl|kR8WE5?Qh`dHOaQ{PLko^O*!_~<4o+Ac_ut*LT3uiFqXo7_puiHm
z!zG@Jr&-akx??<AI@Eb=vtEX}NAg$L1>1ER?j@d$$S7@kXwkEM#%$0H(+SCPDqfk(
zPkoJ%xfe3-vf~uTE47>R)V9>ijf+6IWbB}i#<e=t6my<`TAirc!%DDHqF8l?79Os4
zqH~cFIpPCbfq;6uNLiv}Jiqg-ws0w066ddM|DvDGZ?2_34F(0#&gjo*lwmE8pg#$s
z3b5r~Mo63!kXLUpz&QoZNXdeG>T7%A6cS`@Tp~iB(zb)=?|HLbHgbRr(LI<>7FrXX
z+`K;fa@9<Kw)P~PK`y9hX3{Rx$oA>3?HdNi$-f`S(am@4=&aHZA3ZBmts9a2Lh98d
zf8|AP+3HsUU-aPIR^Ng$GZ5P3Ud8km?p_Z3`DW_5aTanQ?oG~OHj>-UOe!NOb^Es%
z$pzItkKf!t_jWoddBZyzLn*!EG)qm-S$bpTg@hD;Ytyl!IbF}1@^(ASx&+IaG9oX-
zVFpj1i$mR)CYN|G^6#2BB`4I^+a$6xVuY2hz%`Yl8BA&ExhHH|V({0nrI{ZMW5#rd
z18jJs@_|7$)VJ0n=$^{VS}f0Ilp$s~ZB4UUMjHwhcm0%<Gn59;mTLhUO(J-<%Xoyu
z2cDOIG^oxqy5oR<o29oPF&*T&W4d~bA!^@pe9}UT(CZetT^T-v`jeoTH?Si9HO6jf
zMMC2J=P4#eUgp+E2>1U8QE>Ad+-b`bRUz?zGeU?#UWgE)_8k#I>a}Ykgj9bT!*UA<
zL@JP<cV@X06cuQ*1qjwF2Q#(dZ4A6t=xGUmgpI>w`buZhwN?n+AV~7Wk%!em(w+&D
z;V%^=ml&nY!$w8Mq^xqzgBp#BeQxmYk*`rfCIy$Y(;e+Y`#6`FPwLrA3Wuw9!<R@}
z2%sqRhJo92lwe5tAQjQ_P}gvYz^05*3M?${ac_b<FTTB{{%#$Tt#%wv268K%-q>D$
zqCYHi3i%)f@aN!*;1sVS65mr*WTr%R4j=1IxX`lTh!=o%ji9gh&;6TIJ-50Kay!7-
zS|!iTXKTSt`r{{h76{b7FIA48o(MIUa8X$^q%lvmh3kp@7EOGAu+VxLN(8K>i$++=
zZ>7E+DKsRR?})rlehl#q#A_dHxikrXi5n8%Lh=RTThx9K@eL+LvqyhNEwtWAd<$u7
ziSKQ3<W-iue>9F$ts7C*Lh8f5M;Cf8e2*?_-^5UtZ_%aR`#z9vEVYzMZ?e=9^V|8w
z$O0i;`!(i6zQE-D%a&iVPQ<;La9nM~r={hwAt8Qwq5Lng$ZtmqWo~41baG{vxw|L~
z0Wg<g)-F4jNSr7y11L*HmwKEi8-M-4=`V0v)VLS`rX?&dzm|VKaxOGod%6_oQqpDU
zyQNQjdXQjg2hcBobivbuQIMXIf-ay-$7g82bdM2^IHceFx+XYQbN?LBJvh4b)4yK+
zvEzwo&YbpN{l|Fs>1FrVZe&4o%Y;?Sw_iK|gOb3c6~q}c#Ms~eS-$VSj37`8E|g3C
z^Y@QsSoxZSXn};twMOiO2%IFV7OfqV{->8TohT+Lx=7T@iyS(voAfJA%Et+!@pLDM
zHuB>?IT;4A#ZmNo`0?o>s#NLtE0=?vC`SRUm&=_fihsLF2Jv^DXyn~0tga><2R$9j
zYJ%BG{W)U{P$2J3vW&DO`6XF~HTamE@ye`1SeWHz%qjpBIJ7yJR?f$~kXBBIJyggH
zs4K!A1vf(&KrY~M1;pLT*Ylxqz$)a$_R&&Opjw|=CSI-#sK=y3<>i<E1n#<(CZGx>
z7?md@UVp$<DuvR_kqU?1xW67Rl6~-rUm)HDA9QsER81w0wi8sf&W#Zag_vX>(-JhM
z7;R)&SxHMM6*INKi4|eGMEoQ_OPUkOKQ+(3=%cZ>-k@b2V@dO?D|bVTh^y5S%E}>T
zlXRCeB-Z#C^3I{yQi_-%uBa5>Qyf;6mXNF3yMM5!!?hLP<FCHbZMf+nbeg(G2Z90;
zrE(OK8K?V_rEplhu*qVXMw9(Yti<|U;^gd_Xt(v(oMQ&AY$si4pdPB6&~}>Ih$C3+
zB4I4rhBmK5)gbOWo~+kGtQeq)7q>_o1hY*^EJHNN7S4;tWi1$k{iV1ZEdF>&+3can
zPk*Ve`WQPU`r1T7&r!a(hcAmRp4)n&^_gBv+rBmU-~803_rhG=-7Iqs^l3}<ZK}k^
z(qz0!C4^m6n5_B)Z@fkWO!Xqb&0Hv|ubr+Pb&gJ^>Z#}=+q<h);Vr%DzW8B{hO(`e
zPgTm7WLa-<ee2TMR~HdOwR}h%barqVJ%6`LmfCC*XxnBFk;ul#(BYmV*m{<6)e4ZG
zwwm<};3<&BW>;-bb-8an!weO1rMdco*lf~^Mk?JmOC2pEn<!FWjR+)zPhBoDr}@dX
zEcVqvvY(PY+!+tXv5%X3xV8f5d}u|FAr227;bXDie@kRYRg_XxcH2^q1#C@hSAYBt
z-Myl`CBW9`^CU4?_fzexavvSLDcu}*96NuBwXZ-+eGgTnbXl1XGN#qqS<JMdVNy-B
zZCG%BjI+V5rOKAu5n||>SGc+xCxUCHtr_vYBK~c0CQviwZDqiaBC#xeZi#n;ubk@2
zPo9pzb*Ce*@2n+~t4KF?*2*zUZGY-}G}h8lN9^Ps&l3lTcSz7zH=ba~U?xhklExNM
z0Hbs|N#bqJZ>4Wv4Zib1#(?!s2N?s@cRR=!@SZuy7<S)qkTK%D&q2lr_4Gl;DD&5!
zuS=hbK+pDG3*uZPKy8Ni>6<mBFdQge4_6fa#^5KA_CN@Je=W4}^AdIw0DtK$F8^D?
zG6R9id7B3?+Jlw!&$GDu@aK>3zkT~RIl=PdzkVx4P|9K8?D8w#8WNVYS}#B2y(M8#
z&I0DR)M~MUQDmo|^J8V<+p<$s^ns*|SXY1gRmYT7q>1`)h}y<)jNt1+!SCUf)r{?D
zyO?RJ2m(Ka2!H3Hq-XstJ%68{^0)qK@AFoA>i1>$ec63U5=AXcNbx)ZcAEaJNt6O3
z_`zW#)x4w$lN7>cnK(nyV`H`Dw=4>(9yHa`N?592K>1*nuoa&3O2c<vB3KzhUdo^2
z`i2|^UTLKkX$&2)=*aehc|wwIlyLq2?N@q^O5WhB`7^?K03oS@(SOT>6trDg4=dzk
zw5z3D+MajkXr0T@Fd9iijEHH?y9OYH^2^eY6nT57%ze37AOH+eT_BJe3NEiRXQ2sA
z9%f__+D1j4whL#OZ00yORMaT_;u^$uen(6U-Kap9`zxF4)Oxm(JMgi~z{CA&a)qRG
z94WtNP3{!+FK&_z)_)GT@SnQQB&X+m=ZY-z-HHs&*2zUGm-3K$U!K6K=ZGs;=`EIe
z){%ASR?Z|EJ(hYu^O3m&-cJ*=6<SU4T=l)mMK<stoOrxw>RUPxe_k&fB<>>&=>tY8
zDyhojQWnVF8DBF^)UiYpPB$mm5w_ftn_D}}iK|&|kmc544}ZG0Z^?bz?<J_k8e_Mg
z=)?}QZtTZQz5k?VqZs9~qUZMQ1l!E`Rh>xO?+k&mtPf8E2^)t|N>q262MhJMK8K-m
z0YLeK=?S|(_EXXL4rwH<%0S9A4n1HG#}vwJ1?+aLA4RFkw@U~03I?-|lwtb<0<`-&
zdN;5)FBM!(Pk%FPST(Jb_EwH|(}BhnFBRki)pR$#E)Z69bcE(L1l6sS0~b4t*V-@=
zLb`2U7>S2Zd8GB6=`bUV0vQj@psc)Pm*tx6V>o<VseGEcVXgA1RZOgW{#g2LsXKXS
z+LK(fQDrU!*K3E3{uDvG>JR1B!1SGX_vZc>-wQ0WJAceZVYj1*XZPG$Cpbr+rDUY2
zKfeXy`dN^GY|;X91nlYRT_Ek6Xw1LxegB45$nPN5&Q){t2GJT^eshJd)4~sh=K79M
z#-LhU7anVD$&~6FCdrNlneyk;Poa;>YJ4hW%RqIQ)}^_qpQ@B_bF8{(axdh`M$ML;
zrbN}j7=KhF&W+jiy|R#^-qx(J+eB}%VN_z(j4mq8v$<PCzFE)es^IRHWCc<aDo0Es
zf4zd<gl7ScX`@F<QXHHlV>lJ&b+DqEP?8HC<;F5exk~~-+_e?}kdJPuTz1R&?RK{m
zT2g5ZBZp_)QfLbM)Tj(k1s93@bZ*(WnsTE{Du1`;l7aWrE?IEBojdx|Ykwv?@p)HU
zREpX^J4y2m{7+Gl3t%TqZ|`p?H8Uti%v=73)Ce2bp4BujghE%}tR;T18$Ok)pWI&6
z8V>vBwt3Ms*&Ux=9&M^b-o&qcQ@NGetPZJDiaM^2+Dxg_5~s-vH>^_#2HtmjX<_?{
zz<+-7i|tg4OqdB;t_-jKPZMO*V){Q5R!}|nE+aS~gOjN4ID#7>&O@22M<Adz11(0#
z!}z1@rshz41@*h-+Hhze-?V{GcN?(*P-`mnJCE1~tmof`+mM`PPE63wkJiFlfUh40
z0jJ){m~lvU$N#EhecR>USt&izJB>YipgUJjJ?kXx`Z**}4jq|-8>f(pagt~cG_<Vp
z<REg8!s|_s((FY;$Y%5ib@VvVxe=;|Pf7&;2dXMZs+Y05C=>xPmvPnw6PL&OGbopk
zQ9cBJ8#xlb>sRm(h~caF01RPl#SXBy418PcVGwLQhh#95J?!7#DzeCCvs<zry}=l0
zOU*~I>SH}gDe)$h__TX^-F<kmP6Qc22P0m;3XL5K|7(Sf9z@VudaK3jck#~$rHset
z2_L7FwfF@59`O?nd#fXk(Ac8^AAQ&dr+JQlHTY<Je3^seZF(ql%#r)3-6Ngo70uK5
zt`mJ8{`LBgonoeyOrE~d$N2XAy8CN4iU3AQudVp<eMcS6YH6bp8j^I%h#&tJUw5BJ
zvTZSibILzZ;>Vj9T7G2|lR^}g_W&~#wL)l}@DQa%!i=K(s1zV-?gbx|-Vq9VCgq2J
zm#9}MIuypjND&ZM2!#>=e@DzAl{Q)_uZWMr{c|*)`Vn_^EGu6i>br7UFI|XR$~mV<
za*_;9zI)_;VYwX%SpQ|YXDujq<;s{O6LK?fm37{d&eP{izBbc&nQ*^`nOkD7orFzQ
zY({Vs4FfBte_6^|ONVrmN?GaAO5aC+dJ7C8qPL4~Ll8%bfhjU#_v!FKZ>vIpE;0Bm
zb!ebCn(uvZ6!AB(YS7Xwc-63O$N-4u0;?@|!D8HUTRPE^;{Q<zueWe8UdEUv*iTsZ
zT6lghdvQt0fLqJ7Z90VTP=s(Sa25g%-qG6XV5omhXcs8M$+J6ctFRGddB%}{+o{dh
zoCxU~&Fg|=n#a;^a(kog+MKk?XC9?LGTK*WLI%*8YBy#=#(>o5>3>Eh<b0G`Z>ZaI
zg18MCDN^x_-u2!<doQ&&17V|9y@3u+`k;rwcELq48igY2=gl02!<<4Va|M_PSwQlb
z3~O2_uR3e_^#Z`>s~S7mOSS@kUfBV25>EgZWiktDgc7Ww_j1^mssM0=eZC=c)6)4g
zuBY9A&ovYQeam7Os%_pxv1FGcMZX@1tdqu?g%+s0ys@)@RPD4jdBtfTvIpb$;^*qV
z*j0ap&olR()DNl#c`lB_=}Qc+kpq31!hJS~PH1^o(VCJ!I(0vFZKc(Ju%HgxP*+CU
zF7vL`{R;K_y@-vdhc^CXFEXr3UgWa4y6#P4KxzwF-uEU$=7_B(Bi{RHU_d8yYs(pQ
zIj0=K&0<^>-;8$>lc14;^_`h9<0P78bE{&~chirFQwX}lJy@Txd~L~EmjyDQXu1p{
zsqEbB<`DNLv)bcJ52`$W_b_^uzGA0B5<f8lw!E7z4XbFln$$6ZTAxERXSJ6AN^Dj@
zx3o$Zu=q9`CQ2j4zjY;QF}QlSWg$LL{5h3qjg2?tD_mv{Z0GZjGp9_LhT=2Mx}_!#
zUWO1B>%;l=W0K3=yJh(lC0guad4G0!OWLE&LVMJ<x~VKCep{Y@l8hApR`t0_Mg|P;
zn`VS2G0SFh;^nM=vkcHgEu)I>Sk6t910v(zKe4!hG9=j@h4?FuLa^W;%YBV01SY9i
z?f)fk+$WI-<9wSUi2sV*j7b~*A+_Kg7P5*m&*+KW-G@KlzW(y%H%#fo+ka@N5|~tj
zPEmX>uUIOZXfg4B`o`~(E{o6sX_+v%IuS@xzyWn)Yxp8|&IDW?V%PeOS~10l&X{g;
zKBx#dur1IvRUU|9Ypv^j>l-Uq(-K*x#s<9+6`o74IKR%KS9<#QIuvL9E_PqV?lYqb
z-shA!MuP@U`Ef*tw1O&@r8-#MBiovybbxxsprnB2bE2_-leRS)i5uxG$`@ro{io!d
z2{(yrODA=0LRytVZK|JR`w0{VNwKAAC<eEr1t1$CE2w5{3%=9$FW-6I8_xRFeHJ=<
zZ7jm!#h&VWbo)drpF%-gJs0HRb#zUt11Fco(CR8{rZo`2<vxp!og&wM672KEV4%BI
z)?NU{81*K95a={YN83O+aX8AuHO1IuClC(W)154s!``@5lcXcpP__$k35QbuaW<0E
zDF{=&<{ohb=+*n9*0U1A=<L*hdMIf-pMCQ978YMd3a8E&3bi`7B>9d06{opOgUb+}
z?-~~;DovBmmMcQ_JPb_+eFG&v<Rxs45R2BSmR@XsB$;u*reP!Rxg!NXBrfYJ1emZm
zwNN3t^LHiob5VAKv<o$j9-Y-x0NZi~UiTVQl>(6tFz#{ii-fAS!ehmRDXNm$tl<Tx
z&fqZ7#BzAx5qP~m>uGw~)^FH*wz<;!$a1AlcX@o1FmyYd_4KgMn)x!_RSY9NYdY}A
z<5Vnv=kHv^GcqRIc%|{g{g16b%s4&{g?25<V^6Qi18v^rmRG#;pcm}y=0_`hj{=ka
z)%%p?%|Pv8Y51*KPi@J3#*RngS@q*bzt-L0EbJ>=Nv}&Q(WUd_Ytdzj6eYU+A?Cw4
z{o7CZUk!<v_85b9yK+^gE!|}JS-M;kTbF--gtI5(c$lApoH?i4TNlS&4)ak}V|a((
zyUo+uVBX!LwyDwR`C#GUPG1AKtRGis_l2kLHufe>x?F7Q0!bHsL}_WT+8)-Y=wY)z
zHC$xUp%;2UTs3Q$o$eN!jU9WdZ~VD7M@>~(wvg&AkGz|zj+{l_lyQLR?lQ|iM*1>;
z?{VNNGN+B3=3i30HUg*F4Q$Oz+*D;|Ydaov{^?l`NI6xpaB%Knx@KTk#MiGZf}aJ!
zh59Mg6=#FjX#S>pkmfzkpuxV1(dfL*RkzC=!}x}|QSI-XYdorvPA~4XoA<ojW*J@s
zUA*sC?RcT}IpS2kWZZ6${W>mbbUKZHlJ9oJc4rhdN`!V6aViU|>Jg(%c5+TN9Mph#
zJn!PLMg^q@BE%gZo}FCXVXYb6xO(UgWLT>+M^BTXU&$ec4H>*$=dk^Hp=Dm@csABb
ztw&OH7GI6ID%b{mTjiuOD{L3#wY2i%0mA%e7?BO-sVjWkm}I!v-6%Y*tMnNh$1pRw
zdBPGp_ghMxnu7HXcachTDaHQ);(X5Zm$ADj6qmuI7ZjJM-7zSaaG?SOmoTs>oqrr%
z_V~A#zuxi0^wZ1T$fO3tP}u47_q$&w>iA&_z3(^!XVvt-DKs+y^l#$KXf#BEfznn_
z|DlfW{{8jKUq1g`FdFt>-<40uq!P|f-_xyRv|c5rZ|NSHHkO79=7?*YP%~vEb;tut
z!{_M^rJ>oCZCbziR>#nyl+9oe41cWg9rFZDQ|7@_tgUIjZ*61Ys!S0Vk=gKfo=O_)
zZ)toU<!}Af-si3~>Yvlym+9_fLRD$44Rx_rf}+2~HR2Mgg7(OeC^g9DCW%aO$z4Xp
z8LreyYrEdsmC;rr&<ZUxB!e)O*~*%8OSA6WM6@!6+*DW_+E;KG)=Di!NPiVmPD}ty
z(l(HYiiB=Nxc>h9J&pTHqdqi$Bb|F8fuYjVy*7k0dQw4wm0eWe)H`)f`s;NJN#7I%
zoyv18=~h3UCi^%wyu?M_*HwMojD(=Y!WJYlR|^OrDZDBOteCL8g4j7Da5@T*^h1$J
zQr4R-Lw87q3CS2KkqwzrM}O)2p^Z<_<i2Fx;%Mq2LXSel_fwfRT_#iW^h2xgt*PfJ
z-uoGpAEit0Q^$$#{}AR0hL9wXl#P(`nrT6S*S3Y$PZc`pfN=u@7IGU?t@X_W0W9Ov
z4Fr`=c<Ds&R41SVVa<&98?_>8RcxZD{cGyGrL*9^ISjITSfXGwEPtJEn#gMF3L=<R
ztvnh>;C@l3g2JWKQaQ;8>LP10V_CNHeG;=%5hm(`u55e|v~=a2cwe8EU+=5<6=8!F
zn(Z#9jfv~4bA&eCDnl6I1szh)yrs{X`y!}%K~VE^m$-ROBW1#%tc?WYF^RR)UN-!n
z55pY^%dfLjAC+QSfPdd;{~#WIlB9_aQ%UFgKc+8tA4jrD`bd_7sqz0Q9`&bd`INjP
z$7#R0)^x_S`%O|~6;8^~v0!Z8Z+wh95mtMOxXE7Z-#uHgND)a6>HHihW>XX7Ay7_O
zq31LI7ufGw8r2tCGoj^3S@K1_7RLJ>V(bkwKB$3qn$s%9Q-9*=GNc#rL`VL_4miEL
zF?r9<pC4(`uI1AP$h%1;EmgUsrK(Qs&3a(OuHa^wlOr(?ky?W8`FJ0lT=fG$B#*~?
zEN_frKdXmvo8ei8(~=Q~*;+2++0F@1exBV5Bgq?gWG%)O@d2|P6v<L4K3k2z8z?$v
zl$%j}yq=;)#(!zQM~chGzton|HIp^o?Zp+?TT-9-wm&X2;m8NU83^iO_q4$2%ur^w
z_~%tS3BoauVg|teB?D2gGu-*p)jJ8GDNeK(q^<NO{Z(!NpFm*0KL2T$U-`&6{}<-(
zTG|6A+aaOJwZbY%bn&XQ!NE-T$r2>F0oQwx(buNyX>eK3I5U5s5)Bu}!LrqeK~Q!>
zX(Pz5Q(Q44ReSV>!yq{~!Ks6IFPj;&!;G#vOumiaNL(L-!N0(GCYc%a2Si@LoQ+6w
z2PZxOlVhE*3+b5ij&JDij^fZq0n#P^QY<_6VOw1F?=qJmcSC<Q4<uL6Mu*I3{i<lk
zV_KwuM5UZl@JxSkE6?$&#%v9^(4HwnWFn6q%<;QtB!j`6F%*+@2R9wET6BW<$3Oom
z1{{dXLnPzgV6{X3kcXU|r3HNBEZvYe%O1n9>N-$pC%Bu?0<+u{8#)}}KBI^{?*m~<
zzht!Y@s~9`w+VHF*-G3P$t>DcZqzb2H1j#W^zD3U!EAqxRA+qY6~6RYzRbYaQm}L>
z>h)<jZEIE`Ih<aVA;hHm&?ZZg?F$NgjB%sLQKTArqRXmYlQG#2h2M2lI4H_&T_oDJ
zbkxh!eKE4`QZ)KdQb$plg0#=jB1^`MGc&Zhj6;a6&t)3GJEsCiDyp?_CSyY0$bGbe
zh0iuQC31gKtH;kQ;wNWeL_zj)w6ul^Z34v@v(?nyIE#fhOT_#*+E*S>!lAMHrw#**
zk{8z5qdt>tVaU98I#-i~CgIs+JMuwPq_athnYLHO-sZGf#&x-W&9R)?)Md^o<vk5l
zOqXps<=t$PltG?L7H4^HZG)BWu`+PDK5qkD(o=sk%iC~dn%g(Y+klH&^m=)lkz!|^
z_vdUBIpcH;3{+QKc%glv#>{v~k;@gXJ6Q4tl=>kj^6d!@@S!TaX4Y+__)5FiQ=K8P
zCx@4Ic%b-r1I3sX94YFR>=um|dWI9fIY<QM3lpwl^g9U>0VNaiXN>qJF(S|rK`QVw
zMihUo+-(fr+7a6##1`(Yle9x#KSK1rv*hwO7n1>ZOg7oS-%d;>jcF~l%rc$z@zj*D
za~x%1PocW>E*`PPLzR7{Q#=v-Wj}d7M`PzyTRP4@aHOh|zZ%Dzq526)uiA>Km17(e
z#)jvZPnAsH7Z!4@QmHbJO;N%_9TVnBCsThiSF6ztS~}arxEEWa8!ee6u^UG7Q;%uK
zcYPgUVfC189VVo)jjI>qAE_1Nw#s)Cr-q$>I)Z9rdVPIVAS^K9W>J9=%)fQ>7D0hT
z3^k3{b1O%R{l;J8QVtYnHL1&WywL$1Dc)AMlfuV#BhkxcyK75zsSI`rmOW*#SlE9W
zDBe=4Gm<Ge1NxRyozct~Goz@Mx5U!chj;9HVZ60e=fim5Gm%`W)qz(sF+<^+T3xEV
zjjKOXs{?XU>|DXGSaT|tjcdN4Sci^033)-@N^gpmK+W~*_F|m`qRfAPahQ}Og+FU`
zZxe?}3J}2kjKe6<Y2<EWh}4eQ7KeYeaBrQY9rF5d7@^M)WMjR5L<LS*$IxJNuI-89
zu8EesdnL6SD_BwYkN4ze92!Tkc)hz{$6bpQ1UPD(^Re$;wAEEwl04@|>Rw-@;R_YJ
zxZYeni0^I|6l$~)s{-6VvD<23{zWJ*>};L;YGBg(cXrK4CBOBW_}o#^Yu11Jkk7b6
z=hVqS6Yr)LQ{$ghsVpyvSIgXCRH)^5Yhz5E`?*77w^Ni#htfiouj%{Hj>h&}I(603
zr6^CJV$jqLaEK+jK2$7yOUP#MB!@WExVSm=2!|){3p&)pw$&ujJk_~+C)(<ro!5kO
zyxYXCt1@RO{y{xk<w8wEJf?p&@$)rU@|nQ6k>aQ?2?;%LpASr1dTgj2Qyfm)FwVHQ
zc5E$kC`T$h@al4WZ%c5un%Cgo16P+2#qEfqe`{84M6~3wOe-xCi)86~x(CV6R<qla
zD!0oz;r(G%E5GeG04p|JXtJ`ifdeRVP?_N8$?CvCa-l=WEaJPt7r=j$972v{n7Np<
zYyjx+lvXbpc59L`YLjYcYC<lU7Q6992lCZu6K6>56;=VE1UXO%8!mkM%X^r?+z0Mw
zN0{xu>S@KMJl$%#tzI&|c6?DHQC&B_xyN8&6a7XCR*dVSph`g@C3b?kW_ZQA;mt1x
z%o3sTH2|9q`}IU<ez|{O?aQU{J1gCN788%|^Jd&KZRYE}-sk`+EIK(-^cq*!Gk=-M
zE*Yl!q0O49Pp@8_H}g(2nq}ljnkMj0(h#(%yxOu`yHF!D+^$Mbn}koFB#Er`=Rc@&
z7`EwZ)^~q%Wi*qufDEm~h1XQVT26gI!v_*z@G!`FY}rzrXbXQ0aTLB<Fd(I6oSR``
zHweL{_&_?tNDzhz%SU7e_P!t&5jnMY$~EPqQ)$(Wn|o@5^|E{UwY1p(_AeokLM=tB
z92UDOB=8zSvXskHfD}vXfsqRqqI|Z9<bnN210jelg<_bfepqNOzW7(%`7S)KUw{4|
z-6H+A3T19&b99%tyC@Hr((o`60x~(5(Z??*mvEs11eXQ8D4~DhIsW7I?+2cl=FFS$
zHGK4M=hwr}hh7BDJu@1HFW(R02T+13CkG_Ylo7)(KMh|GpL?=pGLB=?f0_@!ybWE;
z4;e+z5k+o{)*}=JM`&R%#xoUuN+`ld&NZS2FZiJJ5Ks^!DL=e8F-s9Ru7~a^BH&yj
z#(<bXq@1WR2z-Cif06xB3$kA*8^eT=J!Q#9e3jl7n6aYottNb61L_M$DaASSv<~Si
zo?sa$y8}-|%_4cB1`#HTvyNJR*ns(XR*AxDpJvm#&I~$p;uer|%*zS2fRdg=@8$TA
z_JjTvaFm*9;qQ&(c-DX;m?XJIJD|gNn=ysNNo=?Cc)EXp<3yHqJli?dIn1+r&qy%L
zqrvDU;vHr?C~Bm{C(;PGf}&?u`w7L5@26;haa`}7qMltJX&l36vOaMgWYoI-xI4}8
zxh3viG3#MPrhOD0<8Vvd6_-r$Deh(#cf}3U+D`U3A7Q^==aodaI$?r;oM%EArqpEd
z_4EE|0hoV-i8$hA(M2h38T$8$joUwU@Qh-zlKlMsl?>ifx|LmCdpqW4AU}2UYc<sW
z;{2nQ*4LB$7r?Zs%8n1u=JJ%xXE2*&lw(^qn9(p7Qz~uO)1}ZZxU4wJ<ab@ox(@B<
z6dtad{tz-|BkqsU8=e=UjTHG5NrSdxXluA)sPcbpoD!J`BcD@zs#SWI&7@JNFmd+t
zkcC44GvkSw72b&?F&c0o0j6vuk~=sFQFVNW?7K760rJ>CK}H<K)$z&I$M?*6yS{N{
zd?FVi&RoWT96t?@Bp1-uUNQsfrf8>#Dnet@qcXWjdK1s-sUg4;aIu{ztHeYanRZG1
z6!m{WG$=@W$~Z>es+sNApMIg;5;ez6+{1?#5QApkTJ6RG90ps#xHW41=fD0X48<WZ
zKSVMf{6Rb9Pc-D>EbVe`oMkj5&9cW(EryV!U=EL^4vOPnwOF)1;dolzpCnt<ZYEiG
zNzkw`J_M6--LEjY#5o>2t<>+}T(_J@Tm64=;{301208?arg#Ghq_ZVpw#_yk;Rw+3
z<$VmCa!b4U>6k?JZ@=x|ZTK~y^W^pq!+#&cukaJx@6PatXQydy6>qE2Y=>+#%?;pN
zXK9Dr&@>Cp1owlM2BXy)Z+EBcWPp>QfZ{e0TXKlx8lC0EsF>pes<gXA&Ja1dPR@T`
z5jbZd@zbkC8TDlut8rYCR!9z)2ro#xLfBqPq!zVCt#M3yYt&pq7_4Dl7tb3uOx!3G
z-EWHikBUF$6F~A`YiVfE$dayhg$#f@KHw3POhao?AYqVxjeaBGs7uR&=ZWdB)+`1y
z`BX;b0~(XsIT?<1_&7ycGij~cWg>r>#BG8-dcocWXAr%!P4jRfWyY=PXN@+~uR+uM
zHf_p?Q~MTkqS>G$U(EuQK<Ct2-O?ilTkTt3;=PcQv`C)8RC#`7Rhc`%YXuC{1ox~?
zODa;79}Gt9>P9-UxYaPE4tct6wxS!tKCb4wAIM9KoHgECt5L$o#7JMOL34krjlD~)
z1|;Cdyk)JXqj<wwO-Fg9R@3#qQ>_L#o&?3cYc<`hSJrAeYQIUXMmdzs-@8`R&3dC+
zO-J>vwVGa{>uWU~#T(RWdWzTAYI=&7Yc>7)+@e-<L)>$%=7zZETFvcnr&`T6+^JU6
z_dZ`$r%_DX3GSf)v{W$3#hQN^OLKkI$cW_NJ8l-rM&WLToviSC%0{_%(j$AJY~%$q
zV5TVSo*O}9LQiFWf87Yrz|0u6qWPsa9{9bO?&j<3M#`X_s@d1<ib6mUE(O&b**M5<
z(@+$+QWmQQv?bGXyCPCe)m@iMR2aTnC;rGK-X)i)06(St9Tt)1Q~H0g*x+*Cmn_J6
zEF5vslrUy~O*~#MN!3(QDLXU?`q;L=9Mx`+joHZ2kZn27Rx!5D(hj+0&eJm$>@j*y
zXX*+4+!mT>r9;9l=k;Xzs&AJceZNJ|5=cak>yZR{Bmr!efRdE(uI#7Ykfc1nPBI5l
zV^FAU$Q=AUV8>5G?2Uf}l*j1ve?w>%AxmUTJOn8ScRCQ9!-xO8ef`^)f1+{?Z~qM=
z_yA05;oa~(zbbU37#IxS^1Hz>S>nM^AZW_qA=lw~Tw@0sSQ@?z2S9@+P|uszZ=uyQ
zaDYl#2gh`W4IOEMripk66l-gq?_1kgxGGabrVAV4El|m0!z+J|r%~ZGJncJm#Zmtn
z4qu1E=YpqK_%EI;xtW7gFc4OW7E)@GhM{{AKyY}A256zR6|{C`#)T>~NHij-aN%5=
zXDh-|OCvBfAuXanO@+0teF1<@)&xn?OFSF;V5^wJZG~DXGOP*O`u)rIJnjugeQG`v
z9o%J#fi}Z~bdG;%kbzcAW1*#+OXKD4)SEW+(^19EN&%FZBrR#?xlPs)MHW5}4U0+E
zhbrvn%>bcI*XeTh3P62y+W=$mgc&x*_lq~$Nrv2;g4fe1ze&-Hqn6^qeJ(eoe!Pho
zPNnZCLOZ6O&*j$**Q#O`J@szN@6=Mh)!cb{e`<M)zi@w&qvT_V^Km|RE|;8Rg@8q{
z|GB3=N|B?580pye=jFqi8_}<XQdvct04`_Z<6joEn5UqwNZdvLF@kABL70h6=Wk7F
zDbN^CcoXBy+T7NtN;tPQs+y;*@$X^WEqk$t#*NV?zfg<|MI<n#o*?}Z_#>f~QEH1=
zV8@jeZ0UcCs`H1n(9=Iwy-4xwrqZ3nGb^`^A)?vS<rXy5#6hQFeF-l2gwdE4uF)|x
zI-lOQ*SC`XgVvKnF7ZIIb=;9JWd^$LLORIls8Yt|T36*LU=q2Kboy=k(t2fiGBS{{
zEb|ze!)?aLPFJ;^g{+@#QBS*Yd}zz7Equ5uYe#>zHe2>_USHHEE_zK&XOSIi6aK;o
zMEolv!tN<c)>I&)1xs-%9sEfGSCLj+wfs;$Gx5}%;piM!Y<0f=lr3zcGO5i`KizqN
zhPP8o0WxA~P0MJJrL^t5@tpe9WJ65hBil<jlU<$1T{YN{2G!Ldy^Q)+j?(jM?;uvC
zy0m{#p&4ycs#l~+;fS;J>#1ngEsE@T6v0RlCY<f71}$1Rub39Dryf@oOdm~v81Eix
z@!jq)EgoZA&1d#%qEfs%h|Tsqj&qIAFJ3}S>+eelhumU_)jJF}aB#_Y0G>ue^{J<N
zZX$$J()%_-2>T{l1uv5~jJ-=WKjF7j{ZD_$wNcj`4ilZXl1W}JO-_86sn<AjtU^}k
zxs4Y~lr^$I6n_<KoOd%kj^n*HE+ij<{qo3SorJs1y$Y#AD#T7?U%4Bg${I!f%wi0t
zxQ|uBj()Z}YcY`kz%!|9|6+CXZI&1f_4%^gdg=a3ilW-+zp<7ZuQ-WtA*deMwk3bi
zk9hbo<ycq!D@LQ$DpbwNF<1W8)jWI9z(k26lT3FhE-Q2Fxh<4uko7br(w-?&ls@HJ
zd^;*wxZHCQ>udqI*$@{<+nxC?+9VZnXQE#x{<JbmGUNL}6fTZLTUe(409qHOy}90~
z!ih#3Qs(Cdr001v7XA>()vY`|26ul7Ha!$lXfzkPcgC$6gG<Z!@qJf_3@{}pf9<VY
zEL`2=0+E+7D!ASKp)tW5ULP70Ub#Iq?tG`qLlzOZ^Np8w$}6u8wWPM)8B*u{_ba4e
zfF`D$?3$|(7<ySbM*f=nCoxy+#^e?*pD18b$?=-=PNin|ZTE?0;s3z(aJ!eWyC@Ww
z;o2q?m$}C%Q36EUmkh`#a(_HtuP^`m{Ld55o#xD&a2x*7Umu@OzdunEw6{zOJ^lZ;
zr|?2aVA6@n31<BC$IsI*PyeMtt16CPq(7QZfBZU;CqHHkJx2_=F<N&R0!LugWVB-v
z1!oM^JLeiPQ&+sub`>xHVxzIB0ptWhS5(|RLB-#%4py})Y%4X@Dim<Ev`^Z!o;53-
zd*ZqtRB6^RFBX>&$tWOy_vFV}pAf(ljtkKTis18zM9r*Ni~_u$km#AWZXr>xA}FL{
zLW;#KK(IJs1O=n}Q|<0NlZGkm*B8&TUTc_V%A}d0(W`a1w|UN~)RXu;JTvQa&ST7R
zj!M2U=Nbfc25m;LJh_2jog)N6owdA{es}%dL`GX?w44k|)!>VNH2D#vJnRu*3<D#m
zG{FzgUkt2FfuRyl_!wA)^9B|Y-%m))afe7;_c3sif%UX?ibf*(!sEDm&AgHWamRVf
zz}tbi3$B>bBXJkpG0V+P+y#iRVs68JWC6&oI1cs_uE)V%Li#w^rDMw3y|636Ft?(u
z<B~FC|9GeoMllY5iz(0F*Lz_F(_(>we#@cmJ@n}DK6uHjUyMIo7jvL|A-Q2%FR;sV
zZxNJ1F>OZ0<WAU#klYCz5s`a=0|Ig{@J>AL8jlIbqad%*cogI{7>|Oy#^ODYmr%Tg
z_=!k74Dk|(haq0#@G!(n7~X|=iNamuaY48r$vZK)Yn-@$*ZC0Kqp%|a_egx4&dV#u
z;F2MrRkBMQLfM`9xJ2gVsFaR<t<nE`wGnMBQ_|M8Byf_q%oI$ihs?k1bABHg_zRfl
zXhD7x`H(1I;d=qu(#V=>yZAM9Pp)mdR(CH`sd{(K{LI#tfm8YVzVNqDGq}7&Ex^<7
zs6TVMpf%Kgd6bZ`mQ$r=mNr{atz6i(s6v#MuUXe+9Z7<hOT9|p6Kj3#jgi@sia@rq
z#EeI4O9~e{i!q<nmMkzU;Y*53S}Ep)c!>1hcS!1hRLczMBZQbrIg)q*^W0KNXELvx
z$`OgDQm>TDQHk%TlQ}9;-z?dY)+e*CRFWBpd(P*7#^GKPx^cLdjBX$9C0*Ob{dhTa
z9PTBn8;E;O>jvVU^Sax(?<rm`B9N3Ax8=ZMWgYci@?Tl3%y}o#c(hnqVv9ijz}1=9
zoJk*%w2`^ih|;${TRliDhzPxrx>k1}=9A(<0SR#VQF;VOy`*~}^{vn5_Zg&K=pK-W
z`2!Mv{Nk4wyS+0<6a2E`_EPv|7y-eAhjC%=^lytsXg7JLgupPdLL-!NO!%fQSvFAf
zJ*KoFWzD;#AzZ~PXS^}%&0$KtV$Q3&4E}ylD_{83B7Vi|Q&sMk!7Q47k0U8zZ~?13
zDpKY2)Mag#ma9Z|<hl(1l39oAcumh%e9r%WA=_=UjoJQ~?_cJ7UC^@&zOx6dnTx=Z
zYpkF|8Ol6XX>tnfkTKu{#>+|cg*ZVMPv@Pu1=dm<O!zFtx=?`1)<i)r-S4$JerFRC
zeC~#$_s#W*fJiWj0grU<DB%*$qKjKfxB#VVVZMZMDUR!|mJcwldt0bP;)bgoVKqd5
z@<x0w%NuDe*YcOQ;oH{I;i@9_7jIh=ttXF15NSJ(N4y>0d?>3iwcvbVmPi~>nWYlX
zmu9KN+qGFbw+<vr8rX7x1#BlvxqeGkUz+3}8C`Y}6rHAT`{Y*gZw<3{0fVUVdRO2P
z&a<WGjrMDU&B1=PU-w$~GrWzBtWe@(UD&eMzYq9!_QYNv@x+Cf5u^{y^5QSV&X0Ay
z*(dtOM@`nAS?w1l>)YOS-sHS>LiJwHg;J*(UV8^H`yoMnupeUN<TJk*kgrgQLle!H
za(Zx1G;cUuSw1mqng2wXx+iUy2+k-ae~py#yfJBx6ZB26HC&68@(}|y1C9;_m;*X!
zV^$=PK3(D1ZgAPS$)meI@tg-`N+!!zkZl*5nZ&7_Hx*hMn5($<6t!-S<?XA0itzJB
zOwt#}KA!ot35GJhXJFvykxe=FEf3*lq8Cml=OOmPVR_B(Tq7&DpCL*vu!+)ee~5&~
zb;PK@2}Gn_tY_x-3tNwD+ak4YqnGe6Qa{)ItZ!18TKI@}rJm|bGzf2OTQ=Ysn?mgs
zEkYDowQ!NXzgf7Ld(v$9#_zfUhFU9?gb|Ti_8X(a$_=4ZswBlRj#-JC;&5GkYxJZ?
zhHqh^wQK`dtd<3`7NuHZO5@S?e_5A#R@?WZ^_V@2S*I`>tGU3e`U>KJaRKCoAmy5-
z=(X3sWL&^Ij4Fm|oww)ht#Mg7m@sD&>*c<A6mwg>dT7Ge(G;Ss*fu>lT7v>i(zRP`
z`U6+YaQ7uOC?O;!uOo9hBJsF4uO(}7L7NGZcs{ehghfR>UN$L*l!&3Qf1sd;xw#C3
z1A=PUUeI<8ki-GfUQ($iGT*TOqwVId>UqY2=?ODh44j=*VmBKPpCTbrH>;y*cZv4l
za9sV{dWbX(sU0<Rq=)3WOo2b?A#sagG3`zz#(5%JkG9F#RYs6Rv2&ud;~Poj_Hlwa
z1;7;w*Jm(}S*hAW{^_(l8}@o;CO0}>M3#<5mie*OjU@7cp4{^bajCs*mqpMhB!BB;
zF`%bFHx+*P*FV#qE#_&fp9qEq^`HEsd~#b$W7l3S?I%ggBd)r+Vq{JWn2$Oz^jt~U
zQCGao1hp^l9O2X;HHHwk(60~aKf-f;7rP{Sg%pM71_tEed7gcdZ+h9RYue-tK0Pf#
zs|j_|3(Bfu+Zyo7;ofU~v&FZ(Xj1e1UgMrkc`w>c`PvK8uZ@bm0?6m5bYa!0{iQ`q
z6bY{3C#k0{9g!~8w3_FX?S7jPz0{s&|E@=nP1n!Ync=i-r$H*Qt2LhK(5s|dBucK^
zmpjoYA|sAc>fUZ;yLSDP6(_7cxXwe^oJ4;h)uV;n{2z(spBy+P_?NM}C={1MbTJc`
zsL?3Ne>&{{`TF;r<V8ynY`l#><J0r&?zi2@f`uTI#_99-U3{St1bQ`DDID?q`>)fN
z-N%t?olDY?{Kv@Y_pj5?@`*tVl0cN!8Z!eCBq6IOYlFb}DubxsOKAu+{X!QpuOkRB
zWAfdLhm#bwBypILqJd12aF?2()&#Rr+#c@<e@bKWOmW84?(#c%$m3#vO>NrmBl>Z1
z@VtMpQayGAeH$DnG9eR9#F-Usm^IIf;R#cJn~D2IV*Oe(LPAIE%bz5YSqq7zpoJa<
zpc~Ora!{xu&2-m<Tv|STI4fZcv^}(4$b)%EQueDPtECs#;z&sznQMbn`;*o<Yw5VI
ze_Cg}KvHL>1pXA)7ghUBy+$jv^IPimoZ?@4eOa&PX#EJi9wf}LL>MAZYjqox2;fI*
z_vp4V!a%6HtuMKVy1L7*&IA+)b>~HQqz{Qlc5R7c&Dxk+-Qye&qe|b@5A#*`gHO^f
zJp${8VHA|{<d#U*ba^aDELuNRoEMxZe{$B+*^!g7!gSrBnQ!2yJc!_NoE6?PUst+o
zZ7i7rJ-Ta9Rv2Y}ioZ5a2_3?Qc5S41vqaX~wPKqrE2l4=ueEQPxGMYo=h!#4J_BkS
z+BaIG0JD(GEq)uEh#O*{bcby-76koNw@sI1sjzemKWB)e%%hMnaIYc^#mqeCe;9m?
zu+T6Kq7`+mY~l*}HS8DxGE6j$FurCGhSV3IbDp4OS-Y&7Ti=`Vo?5*tcS}=$!-_Bn
ztJK96c?Vn0sA6!Z7c-hOGn)HnB(-$XTt>Z@e3JST>(vJHODmkV7dc)(|NWt=_F8zg
zWvDt*{L7@TG3f(pKft7qa=9^Ne^nr~awD}n%(`|+@l6-9{-4`*ZDUaX!_4|E5xhp(
z8@*Uq(xJrZ=2%j_8qWiRl6F#V1w5$sh76I#M;6r6652M>Jr3I^31Y7z3Ta!v)ODmD
zbq!CPWgK~DA8LNEelr$YZkB;uuov>Bj`4L{)O@DCJ7H0ZYH|;i5Jq!Hf4q+-2uDfb
z|E@fzWe|SLXn#&H36FQT%(cuqz&K%PdtvY*v?}X%Wsh2wh}=Y}=Y6R{d5U{47G<kp
zyWIfCJWwKx%#3;NTNoBdyc5CIAT6@hRt?*&8fYF(W@VU<J>{AetLS)wTTo#ADFRCl
zSFuC7La1q67GyGESPBWJf8quf^!>0PT7l?K!9p8;4J@bw`3W^rP~rO6A*uu{G(s=J
zhdRJXfo>sFd?P|?PIA6Uc~_$&E&(?P1<+a{yA*E=--k%yMJrzUd9<@ZaS4<bLHc2$
ztF%}#-tYT>W?hYmxU?FF$*aT50dmP6Ty4?pYB_-3KfvRBsl}z~f8qjju%b9{=&5Xl
zBTA&r;s#a}>*4}KeqI;ot-_MFLDb6u%G0tbfRtLOp0&;b(j4p58s~gmg?633)ZRrz
z{CjJRp(rpU_Qn_^ASqS&DKQ2DNgefuLB>e20(-T6tuA*?@r-c=GG1WS`T1`(1lgbI
ze3H_W!sn}5QdTEhe?2U6u(YkW7j!LHlJ{A8sf2XRBwIE!rBJ}0Xl)7?)4k@w_N9Ju
z-ighECE0DHLq3mB#}h86NqrS2O&*&0iEYHwqF<}l<eV}^BeOK{$FHEBx|zHHd2BM<
zuzaL^EH1bWx9>>tp6!c*V(yV5!0GMFn|+)=v_nvBZ`OHbe<|qe>gn^Uz~)lXPN-H|
z-a;w^#SNq~r}&ms=Dk~i?skd9K)w~bx7~&<y?h$C;i{0<_$dgjl(eTpT1u+O)66So
zCNCk~Q-l@^Gk0@E(T2XV;u4vAD<TJHQVLbvscFtGt+lJXC>^UziuE^ggIHQ~4q2ag
zJw-EO^(nt8e|4q&bg!JF;ijl|sGIQ`YzwWqMRTm?`j#}uu^+Xzp2ZxM!U><F*TSdk
zdDo0eo9SL{;Mt3ON*(QKo)s1;TTHQJmy)!v2He83a*t!fBGneV)FeK4`@U2SElUEr
z#O07bv(rlF#5&dCI2vD6^XB!4p@P?a;qVsX-Pa2jf4J(tb0{g6n|(*&yOLF_jdoc?
z)~T!0uKqZ{klh>PgJa_7T11ubax=N9_QuG%N*rptrZ3^2wC_4lEU5gr8ukX1vn)PZ
z)MT~OM-abFMD4?rND>m{Q_S?|l$r%7L-`B2E&ykvd8MQ-VY7SppRZs3_W2hOH0+Pq
zm9iefe+$SXeb0|RdYbsP$TxnDhoXU2Kv_OoJp~|-qUe1YR~tT0JFUnxBf;NVzr|Lv
z=AP_%H#ntQjye|lT}xR(Bv@O^bl>{MvZyvg;;617K1C`#Ha^n$GAcgCtA8$CY1F@_
z-Ir<ikufz^xuj4s6AqUCox&7^28Yn&TACaPe+7n^Wl(}6g*nsO^IIF0us+s1wDQ7{
zx5KH<R(UQh9p9yiXcY>jsr-)ZXK)yJrIr?|h{%KUL?C@s3yf|?xZQvL&f~t)s1Mz5
zrt<&_g*KLMv=R<oq?IyhiV9BFpK|xoyKHFSZR%>ZtISk3VZXfwA~fa4sasO!?V&OE
ze{nS+Bsv8v@EA=)!@-d(a*^WFhuMV#MYLORgirhAt&`tzc7C#xqm?sdJ!2?xbqjhj
z`%QJc9P2nGj9!e+5lUG}EyKC~d%1d|Wb2#Id9T#cdDBZg#HiT?b&7{O;($whqoSzV
zFQm{H<4t$43sBQw*UtOq2*ne^KB@2Ce;#Kjr4`T5DU;XAZ*wGc2`x>z28M;qPQ*Lq
zv}hieUc$vlEiBO-Xaf{p@VE!81s*>Ssfub#N+Ln`pe8_$Z+vpO0A_X+>gETdH5fT|
zUjUo&@Rh&D=tG&s2b|@rk3{7uuj=YTC9_URxl--FDo_lgWGuOnXJ7Z}<Hf_CKudK>
zQfN)a$J^O=ibvMzSuD8n9M2Lfb|83fXwKgau^jmLa&D^5z;oUgG_!lk6;!jHJ;A@z
z-;wHn0SsYE(3i2hC={1LRTmTiIhXNQ9w&e898HfKyY0TeVtQ+Vt|IDNz+%yjvm2nd
z1i2MG6hV_7lA`bR(7#_wBWfhh$J(A*zsn2YwXKmzisGk8M-JZyIeff(|K;x8i#CH7
z!Gc%Amu~~$1Brh@;-IsG2M|uf@a5<5w+AVe!`FL!87Z~Km&eZm-|geQhJZ5wpMZbG
z7d+m317(d?_yT;naShJb@nM9cPUTq|cL*mqnbrY*hlsD!<KMsh^-dO|Ifc#}fom&;
z@ljiRw}{bZsyu!(;W58kq}?D+Bv4wU-@K$}_~nIWTEuGJ=M`l~CXUeZ+?N9R2|pv<
z7U?!mG&8MDPpqnIU73&btSd9@NCSVZ&ttBxFqv)z4@ARAKaQp+p{$DqMxmeJz=x4n
zBD%z`^!AZ`d;W6w%Z!QEMj>4ozW%&R4~zz(gTlu~I2ndt|2cfS`^3SP1Cc`Fn(>d7
z!>`{5HuAYhT9hbh(m888Ndt-EX9j1rkl^SX&4;qvmvP5ww+cE*c=r-u5u#9$D6^JB
z%x9j?(T_YcFNIeo(Mf+Y`Q6CpM&_Io;1q|f<g$dKWI7Lgww}>}11ND4cL~phrq@~e
zDnIcIWegYa6S?LQJm>)uqKvg0`FsYKJKZQEf8ujh3WRZy;{(I^a|uK%yqcW?>E~4t
zqliAtAi6~kDs6<)Y9<BpcsY|%K!97zS&ucz*?G2hS>A#dM$h)LA#YbpSdRhhetBRt
ze^`-+AWZOl_O<$AoQUWI_-@QHZVN)qj?Bo*(V~~~cFx3?nF+oKs#n4)b?#=o6W-4l
z1Sw09BA4LZC?tPrzcO3gotEat%FM7Hq|6NJ8z?iwaDXy%nzxmi&EBcZd^elQ%mqcZ
z{=^okn?#UCHbjt=n?ufreOBB@ip(IScq`=}>U244f))6hu?Ch|6QY>wkfIRXjW10J
zh8d+;zQ33(tuSiW^8-ti;Y%-4r%~B#oUKaHz#ik}tk8eP9$>tdQj6EJQ>&G5p+-0Q
z^axr}@;4<<j=0?>Svi$+l&s~{ubfho`js=Xr<&rqa+<!*$+iUa`(mxKU74QRc6AcP
z4712U63di^rthjr2Qvmcrk<JBg7w<;&NvVjPm85@PNFMQH_7g7x<_#C){dpedX!`7
zp`MeKT;YFWibLpef!uZo{p_opffC-p4DX6FAlJcWUxC+J6s8$(nM<Adiv4&U248N@
z=asCAJPecO1H4~AnbtGDU*i3u7Cz26xWoHt>CkG-q-w?c>7?l}<4wBRtNKRf@u@L`
zbemO-?=|i)-O4~b^9<4d3hn1wyweX#w54+LM9+UK9x!Qe^?+m}I)7<tw(4iw#o2CU
zb~_%Fq4C`g2+lpxM|fxtbtFli4$oMO4=DCG_A5QogWO6F^a8K4!|D_-!<cx1(JAht
z2e2pNE`Fc~xE631H^7b~jQWw~x(UjvY@en8P`Q2%-NWv%1jeP{9M&zt{JxU3eqF=g
zPMd$v<JPFns|-6Ml*hP~N2S_Z;c`&=%YgGf4tX-J*Aa8B{|$LqfU+hg9R=_%LDl##
z9mU<d|9t=Um#_aor5(QiH%-OJ7IRaGVfZ<IMaQ5Ft{Z-o??KoQEs`Lj)QB5>fr%Pi
zU04Uc4tECAEYLPIqd(GU#8}H<6fiB$@w9)(i4{RJL^LFbb(HY^Xa|X^2}RyhB_n-H
zP|Di$71z_M^fkTO?=%%x{cE`UHr#zGT(wU5G$qIpaCFHRE^rlu0gvL`aJCMi@zAJy
z#Fgk#%GO4k{no*hyg8$Bhp4}(9mxO}&~?u=((Idt5Een8A$sTID|}#k#gQtbXoi0x
zhGU5gl|&Q@cSjCSpTGVr>%QZvkInlC=N^<6#=7C&V6K|7ErgZxB^wqq+cKSIlTD*q
zW3EMJbt6nm3dVpS%1=W>(zv(#X>*^>W(d?o@)d?~TRkxXWnDltl@Ti`KJ5W9f~*$|
zseqjQlPsp`rIt)F$97bNtvP)%i(`MmmPS+w5wergb>2c2KNe+KQ8=-_c*7aZ$|3`l
z=QKC>(`&m`n4m`&`AWKfFn$}ak3y-yS1WiOZ^RWdf^z`V`VW(%jl$skt+EDvLvzhi
zL!dZ$k$g<_*|k+|OHt#gW%PJ>j_A-O_Q?nTc%L0nEY*DS(h4m_GM3q*Y4U&6v(YKC
zn=!(m*>X6<#jQp%#anJ(=CP1zndMdn5zxPFAN3J?A{Ejx3RvILQwa`26HJtV;fn1+
zbRq&Ur04fRlwJ(wcrS)B4{c{eF#*l`t{-EZ?Izi&nMtK0&9YOtOKS+{shyLh(3Cu<
zQbTch#^ag_x20K@`bQf1U#5S=^fEwr*EF%G)zsu-cJ{^C(mBxV*malC@rT(Zsc*6m
z&pTqQZ989B*UUxMAe2j6)NH<9PI=Ho+frt<(2qmT!k4VjEOCF9UMzUh#%C>g(#b21
zp7h^CKdl73_fy16Apiz?N=dhLxvh;u_0rw-zAab$=`(IKl@G`%(xrd22@ee~+P)#v
z@W~geYr?veZ8?rgq_XJpV6FQ~0i1DK3e-gsGujBHHIuv%&gO2`9Jdn|9Vp+8KHqYk
zm9Lk)tvF`N^iFNwH=X4y8(ljK14cmboO-fg6IW_5WJ|Z0>`PHe!^abAG<=J11_KX?
z()%`i>+BiH8|ZWkF_nKj=XYO8uE$Sg*{|ko>O)EaG{c&D8;wfRR~=;NeN)3fr5=v#
zF{ENvo%tJg4r10OwA8cCL5r46-ln#sq7rhdK_1XK$RHlrImjTt(mBXx->7p?`E={z
zZ+xf9u<q;}WYFHcbCBUXpmUHz+}}CKA==H(K`sTycMd8Ogo1w~)wTm@FTZ^T%mTr1
z%GQ+Bdf_lW>nC3!Ly(ex=@}r<CDseuCZbz(MKcowEx^Y$>`$u{pekWZ%*jr2a-@tX
zC(sJs+7|U#4=V65sBhXgz;^(vVoW6*wt_{aJ1RP*!p`8Ym2^sloz1>Q6?tr`;$$ux
zNv+W_hxw82D;<9+%qLTE<B}qBwGtN<lRNGHElb6@u+*PisdXP;Du#&3JlGrC5{C64
zTf(58VM|U|+Y*Lrq1~F>61F0V<Ds^MQ?zYM*zDWc5>Do}E#XJH&z3+?`R3|(Tk<xx
z#3jbx`qxzq>ru9(hx(>*3B&bDKO>u6ZP<>Em(sDw&+j~fmY33W!XWIiYGkJvZ(#EW
zC>;L{tQr|AM(3JYa~RCbPJFLf!wk(fokqW#wVf^W0H_dZG0;xe!}dd$H03BF3sYAN
z;w3v+FKS$Wmto~77=J+yrw96`{)W*H`+g}{!rLC`E4}e-vvpGkgu#rC?I^gPe7hi+
z5xE@$GiIG;Mh+Uy8D*M{EclFIL^P^x57oB3)cB4zflftDEhi+^eaW~Mm-J+@@})Et
zt4S$!ooT1SQ!eT>rS=hPf~8~MtWd4rb9uIU5kBsL%1-I9Mt|JES_&q*Tj!Mjiw)n=
zTkL3Yl#QJ(&&@)iDvZhtbx)dE<pPaL`EH?Q0p$gql-DALdI>Ui1vGwiVj5L0b#r9U
zA$w}7bI{gX=?7xdJYwyLXWDB0be`EG+Q=$DchK~96?i&=j&eJPv3y+9fwxDMg~(2w
zR?U-{VDwb<M}KF`;&C}Gq%vELK5bOc$Vc;QyRzw~EIMgUCSNt}DD-PS*UIEtqs*jV
zO{o}`%f<bHqyUhyqTN%NC<2~1q~A}W_whFLejH7egI2!(_~$>3<!57-SD;n2zFYAJ
zTCs_vIg~STv@1p&+i@j-$C&B=*pGvO3T19&b98cLVVAnQC<~XtJTnuMjUXw1)mck#
z95)QU@2{9+a<C$jq8_MWAnVuxdTZfZ+f&g52#`aA6zI<{rO`Yj&Ccx0+x9TtU5!MF
zA1RXhth>*h?mq7JFS~coWICrjgTwCSQwQ=w%U{qkXt{F$$`W)hU%OxKwMHw?ULK=0
zkvvYmd->#tI|45=K;8h$0}pq9&Oly6C=bYE`!zD3!hJ6XAM?97Zbgo_I<5ox4e#Zl
zhu>a)-08e&t%8kT<467L>1Frlt~S9kD!}gdue&&I2o!njP+kV^zWvdC+Wk}u&8k3Y
zSu+0l_PMJ@e$E(r5Dc}o%ms#^1z6Kr8x+M?J%$kj7Tb*2Zsfap>VGqT&&C8(_)yq=
z%w`(R>=lVggXX8`m+3Kj4cR=p6(JT;*-3?LoRa}VKp1f^LqD3#_SsA~nRu#8*^^jo
z&he?9fyN2OYGm+f(5pVxS|NqOZ1X{ki-tY-21f%CDs-(A4M4r$lHy|I=Z}XY$;e#d
zS}Rf?EsQ0^V<ZjGyE2l02C3dDLyaw3&bJ6iQqABnu}eQc-1R4u_t|YGVmACq78N72
zh(ru3IA7~ZkyuCJ2~rIzNivO8bG}3dg)PvsO}t>6)+kA1%{FP#Eoi-xr*TVyALAL*
zeL}^@)7{{MYNHomc_V?83(f49tH+3g93Sudh)V}NMqTK~V?^qI)1!q<P@u8dCok@k
zm?Xd22C&EYWZg{N^e5>qMiz1>3smUmsgw6Pg_@K)^I-rz_$7`{UK(ffFdLiM<G4pN
z_wESgzV|d!{~N~xo#BnLLsd5592X^U;eJAFqb9KLf-;EOF4Y7!@=4Qi99n|>egAML
zUv;j9X>+qImI_0E&Ogb((O4-dCY6-=Y`-*mI1|Qq$-=F=#>gs!DFx)5BH6)9sga14
z$I}o^&FGZPTMfX<5!D(FVL=TJeNQz*VD7Plc1b(~HFh0^TI|Ze-A_R|m3PkJm)7LU
zXE;|vKSO&N3-qi!hI&`IVjZ$2W+vHnV~ZJ3Th%G2LlLxpI`^wO1prB%*zPpqw}Mah
zDo^dsyJrb6DTh|-LZLyE0<fJCOpJC(!<#Q(-sUy9yv8fbP_Fs=?W!6uiVPo66Kk|o
zYVusJ3Cep}LMlivotEIb7HP&BV}6xmv~CVLX>_n$_5Fvp#~fJ0RvC9N7M2x)coXC-
z&KAG<zG@hMNV92zFII?MJmBXR7fGWcxvyBH*R3+*21&Z5Rq<A3knVFSaAT8g2aCrM
zE}HP%aF6|*o4@JLcls4_-}l@<DKot9qxjrro#Ydxr>dHD5LX*-Zqltwmn-?=gl2uK
z5ibR{-SWvD4_K=}LbMcI^_=5mXXNSt)nr=g+4p^a9|TW%N8502VhM7cCcN>@r_vg4
zKI@g|0B~LfY@{7=<8c@Dl1769M^$w@CCInDLQc9><{NvZ(t4p+Drv{uLt~45#$8l0
zxDB<wOKy!)MI1I>s2H4c?#s2r<r@%6B0!1fItuhfQA_b6I<ZHoh1ZiGQdg6IXO<!R
zg~F_VlXX)k&rkW?O20^)*UG$jGifk9#SW?MTKvpTNn<hfdBrvEV(C(-1@}feFG#<_
zIY%S03%5S!rPeF)PL<T{^G=y5m1;J-WDh8hFzV6HVq^u~%=*;eCNqxe3kHU#l5qZ&
zZYtfE9AF@^9t#e`wFz`M3WwQ?S7a(qMkuj=6myWkUWv6phv`;%h_-l~XbaA}bu$Ep
zw%W|Eeg8pX16fNy4zEaTR2mMWb&CHVhW{m<$a^4Sx_(rf`j>eR5h6F~Q15Dk_aG!?
zjjdPd=XG7^C5yX8mtWUKq^J#-m?C;ij%8bzm6hJ8*1o~ESXAD$d%hmqf`Jtalv*->
z0*52X7Acs4%QYKhi?+)7a6ru=vIUiRy2F^hUBR|AQ*_S~97D)JXzZQwG0VIKk(5qX
zt-=iohb4=;a>+WSLege<8|uAjSXX3N;-*~m*HKIncxh__o=FwWS4VE@PVVlb;sczK
z0XYMqT)Isf#eZCza=Tlsk+iyiuD0-h|4182M<v}<!y0L$toAoyjezC;<M4{KQ6=$;
zyip~2gE*>Y--tN^lADrz-uCWNX}yr6sHD9*a}<*f1%5^52sA0<+^xM*YQ5AerL-fw
z;xF<_K*ivzM(UbQae~r~Ju+8$p*!ZHUY*68E4u=PSBSfuzbnM8lXr#B=gi#<Vu`Ch
zrG)+iwqXh;m$ADj6ahGwVb(4am*(v#Du2svWGNEe`z!JXiY5qxhm^8Z&azvY-Hd8C
zn$>8!SG}12|3wE7aDhv{oNmvQdU2gjF2su<I0tNf`1PO<KR>;EfBNYS?ZK-+;LYLt
zF9+Zknty`kMn?w^pd7;C`)`MT9<?@(pBH|t)`<Lg{yp#~J%2$6+ynRx5I^wz#eZ9n
zF=+S!eq0&D{V^Vk9CMO)sok-h;B*=X{2jg~iO>J~{<kMBMr##a`Zayb-(J5z{r72h
z!Req(u!n#D_LN>&1ckw{XUel<zyIg(%hTUysZAH*Vf*p>uZOAS9}>g>97OFL(GWz?
z99esCE-Fl~N)U5+wAPjbXWz;v$$xj^7v!J6ln`K)UM{ur!(k-uO0V+Hmhbjdez&wq
zM3S-GkB4jcHihxJYnvaWK*xilppXM3eh?=DzLqJs`f=&jD~o<7ofgtH4=q1P7wJO_
zOaJLD>Wgz?ND;)U&3<Ezbv`7T1}3e}vv|M*Bt=>4(2}*D$pp)Nk!Gz1Wq)ps%zT|^
zlKDKfjOZ<CQbUT^&KLbs@+nERe#<lc*7EAL=z-VNVvfc4?%T^2^A=RyydGLE#M(Hk
zG)HHSb@poT=v>^4cglM|OGvq$W9DTIR4;ugF>TFATywjuof;9M=USaww@`woPw5f&
z?@44yEwGB(DygyFqMu@2zBQTMDq6mw@`h)mya~qEKB&70yoi@rzMGak)#FkPa%M@C
z0Pc-uj@mk)>u9DrUZc6L;VlW5LGUOUe-^7#6sqN|K5}f)rfq6veDS&U5YO{^bhS0F
zso47XXUMxAovOn6N_MSl`zsNb4%doanKG8)ZHlt3zg={w^e`o*{c%Wflai!4=5cVW
zj%`jAN$eES>{K2K&7fRk$}n)pF2|tuZAL(sUCof{R0amF?1$#m8$`71XJvUNf5$^O
z){Jc<rS6qVI>>|Si9KH4d!>gH*=;)N`mnFhNXI@d+e9^Qq4-4KwyDT#-{iE|r2$3-
zVA|eSSLvJVN_nNOM;cH@`yQaJb`xqSq5`#L^h^5Ov?Ix?DCr+c&-&JWZEvJ5ajZ%?
zH53?pX=V8dSWD`zb=ymKOUDrJe|j2b&&Mp8c9-=cG(M=HcQK@H<yX{nFX+UG8D37H
zGhrR#<$eWqUWYdsbzVj|HU`EV;SPjz4ZTbf_ZUNBWs{xtM>c<G>9<Vh6gV?WHhPVB
zyk*1%b8lQ>9ofbi({w22A;f4C7BQa;7)PZEFVaQae<p50MKcS#4igIiEV+A94KF9_
z;-qdz<L(hiENGsA9ih2LU{ukES)f}+&!ANX%{>45A&h7pyO#s<C^mn(W~VsbK_;zH
z-oh+l2QMFIfGmZ*2?GUO(mYtuyafSt69KGVe<TT<qLAu|i2^}QhPWVj4q_p7sf6gx
ztxUwQcdgMg?A^-6PFnjidW9TbCQ|57McdBUOhwNUlAQ1<Ex)Gkl{;+Qq_qf6&!Mt1
z=Cq(vAtUA3ah1YGYI%RQ8zQ!~9LrMA3r;wU)lLv+4)UAq?=n#tVHdeQ8!0WcZ4Wj8
zz6^e3%NplM%X8pMO$se(<Amh1uZ<*zIO6&$>B{D_`I2&G5xG#T$Y)Ph3JPxHOSEXu
zDR!B_lsd+lN?kI`lAEU<$0=`=K=$G?A_wfmp@)&=jXmi&uDE}G*UlC-EBvv(Z5n7x
z!ECV{B-scZ^_7tm^%bu=>RbAqvxQDgYnN;?hN^^>RkXJX$}-t}s>vb}ZjQ%ji`#5j
zPCf5>8n+$&mNOC5$F3T>#i)KM<lc3T5TU2b;!CC-$y$eUn&RTI4#>(l&~V&NCM!43
zYpI!&*C;~ghC6?|M&&#N+gU^zw}gtl)V!X{x3ALoRBL3%@td2YT{S600!|etinQ)D
zvA1$d5*goDl+FHY6$M+$wZ*qi4HK01@vhV;Z*NTv=$G$&`q6d~dDBa)#iuVeIhx(+
z^{B}WGgh*xpS3e*Mab}kIdt12V>!oac~Shjp$<#WdRTuspY+T=UO7j4?-`5nTA`nh
zSPZ|6^eHDShT*W&^vRHZ()vM=ev<mBkbaKi{g8f=u!8jOImmE9@iR+r_|>k@3j)t1
z?rhTcpN^Rj)jrJ{)TX6+)JCDk4^W@VMF6wYUqSu)xC9L<^1C!GK>s0Dl`aFFb$=yR
z)iP~hRd#>7^y*kuo0bn`RZOE4bw?yajrYSIExXWqpDtHn>+^hYz{Mi&biG`@K;yOP
z(afW4-L>LiC63^{PEmhHB?#=2r3ORhm^({%$5Ll+94Iy7EI}6MS-sAva6zxTu9s_)
zzh7@~TwTxWwO93`Cgsx79VG}pxpv`e02G*b_QHSH{$E}A(oGk>z%|`$m)PyXmqNwv
z6s|c$;noXZ3iE|84Ng%wd*Pee7f|h#OVaMjusrQTmSTS)TU5;#vQ)uh0<ak>7uV08
ze9LUm?ihqZPeLf%bp0xpx-gTFBJI=%ribqxy^z>Uq(fTwo{cswfjM}uusi2oSe@dx
zLGORTvUYKsV;#C!!w{h*C~pK&GC%KuaTu{6K!u7v!|@&t)LO3c*^yz)EO}sOR0Gh~
zPtmdDv+eDg8Rz4RGu#P?3srYR`M^E<4|Ufbl?OL_Klntq-L;2Q%VC!E5qIqXk%<xh
z4~Zb4V@p2!j|nHOA4CKvsh>&&=Q!R^1Sby(W1>Dx1Sjd+MaD%UI7gR3_9zz_q&`jr
z!H|mZ!}36SKM%wu55y%8{K(ri&X;laC@w-=xr56XB4r&+0<KawUQC?DXSCkH+<a8H
z*Lv!YryGWaa6Jx*k{N<?7G3ACXslvLa90?R3p|`xdOZxxvAF2(qL<<JC>2pMo(6Pg
zguc{!-ud<;_c8(y(!xa?BrVYHl=v{?Xn!3??{9-I&K6mPXr>_Ma`UZYX>#>c!<mu7
zODj0%X3}cM!e`mjUn0pT!@WNMx6h-Qm%)E16_;_kH6#o-H7`tMZfA68ATl(UQT`?;
zf8AP5Zyd)Fz3W%>G0CB~yFR-CLx63CPJjduEFF>%I?(bW5h9QRwE~oXpI5y*<gRB|
zq{xvBTNKEe+3xA;>Uys}X0=*RC24g|l_dm8)hQ{JM0MQIL~0@6PHzyJCmCaVO)4?F
zLBL%Il$^>F2%MT|+>z9M%+geA9iw}ye<U0aD3BZvC#=eK3^PbpLIJ2GJw7bBXN6({
z?paeT&@rQp866MqT0^o#rHu!BBC4SgG$xiFQconIDHZI*s)ke(xwMe5&DLV$n9ySn
zgcYe<+=3yv#NG)jG8Ksl9TIcgD~Z`Wq?%Z;JxG?=@JpE>$hgIj!em5{8abB0e`4IX
z55c(Z@oDWzN)PRD20|1MrV5j5C_tJ-hMr_e;+Qmshz>=GlGsp>G0x4fhG1?RM@o2d
zub~TLS_zvGhvCK%KvU(g^4JxY3ki$sI1R?)+Cx*u18jU6XKx<+2zT)DWWkI!tgQI7
zhgZZ|TdCuiC0QNTfO8m)ec;k$f0)W3XecisjiXn5Xxgv@_GfPKDTY`ZiYgw`dEB$)
zWt@R}xL_z?H93d*pr>zr3MO<JQw@^FrU*vIb0wBgNMi}Z<9r-Ayxuzur%>JdSOLFU
zDX|IIt<`a8-V;tVJ_V8*^&tr`5SAtYps8UA2_ZgIO3ipYLm>@=1!kJXe?5Qze?Wl0
zpmh%uw*qN61BUcUp%RePDYk=6_W?;<zzpTdH{YyY{_ov-_3h>5)n@hL!%v&I`rDh!
z|EAS<SJywUug?qMUS6&Kvij@l`}4ZJ_;LMua}G#nhLJ0AF0r<h%y5y)lsoyW)wc<c
zf3f=E>gCnyS^5J|iXXKOP_^dxTYvqJPw_9bcy_K~my&*5{o|kiO0`&SVN>|=<%hR#
zU!54`7=<G(J;VC<7zrfja<6BC=W+w?Zbzoo_g8PPu3x;nc)d;>6VmGW#b&d<zD%^e
z{PV}n4=*+sn{}$&8)sLS8?12#m#YQL7OR}Wr)s6Xy}&6AZW|XtZ(3ShJb!(<dVYQN
z`o(&4zIy)bZ1r;eakJBSoY?b=f3Km?<z{`kc^@=5X5nPsUwycKy?(!C#<=zO_0MlE
zzPtK3PSQXdaL-qme~p;{b>EH}9`Jt4(PQ~1=YytW8-u2I0#AZWf`;(wQLa>pG9avR
z>cPa~d5~AN?9%6+r0&LrTYoX^TiH#^jVC?apH*|<cBh2anp-G!LW%d7^XBUEPg!bz
z{NgN<s|9KI9A+2i>^MM?o(uDAbr2=ulNSN5xd3uaAFo6p*;~Z(mV>aXLa7LwEh8Yz
zbdV16L4*cVGT@@DvVb9~#)fZnz(a8id{PO7n8v*$h$3Dh&~cRmQhz}_nV=oKJ;(mu
zfC>9y{9x0&M<Oz3nBhyX#2y#?f+hC&<1<+z5|Ll9L?kVr&k~Vg-LOQTIH6CR&~*}Q
z5_}TcB=oHi$<<D%3owaVj_eHyQPyzb2J~Xr0BJ_D^L2a}xnZ_j9gu(IuBUYX-r?<H
z2ztn$D%@BX0rTKg>VMD$alb0%h7^{e8w-CWLx9x^PGeGnI%kfw_Jk5&$P!kvVpSw=
zS%D-Hm;yN&GNjCNVsFUJQF2sf<kO__XMy12jwk|rch8YdYRLvXQv@5IyM!Yvt#JnM
zA-FW4v4}^2&XJ%aWal!-3rZl*2X1q(M-n!P!+?s2GwiLimwzQ9UF%sw3v&>9L^x0=
z($oqIF}EWGUkeiahSCQ~aMVZ(m;yCV20(6<Kn{_ktUIMd#Xum0VJd}xP&nv*n2Fl+
z?%efo)GmTZ&2C>Lf}Bb%9~D8W&LT+Vdqj{Zt`8zeZ8s4lO5<$^s8cMc3&aBCMkos`
ziv1krqZY|h0)KSy|9-4Mad!|!Lv3DB%%MD##dsK*b4D1A-op5csF~}jq%wP^W$=#T
z5l3BeoEz9IN3~fIhg$KIizD%^hu&S~(@E6+fSUqJ!^-_PlN!-HH6nV5W7D2vSDO5J
z@^G3woEGqK>V(qUlSyPf99^ZX*uEj-kU9@5um5xL4u9s(i$oJ{TkpbJCd0qEN8e4t
z$!1G4cj@B91u+Hn8<Q=NUl5{cjjm0$U79f(i445Y!G$uyskz2~mTK0Qs2X7gjV|U^
z1!Woq!3h;^i^ie{2%~$&51Pjh1IVMpcK1l&Cx@4R>5j%}bP6s&%DW4R3x5by*=A5>
zc?eYHwtv?_G$E8HgmRsP3B}9{V3rhB0jFp;M4ym43fs6g+@C{M1^K55uv5HufZN*=
zPKz8YDwLfL;kAzH?Su|{Ub_Hj1>r~|^D)Y{Z9;>P8q#xwC*@DA?hDYCbkz2d1fv+>
zZSt$p<miZcreBG-P^A_`(&~uJU<-yxqQ~JKI)AiJS)d)T2J0Phd2hf?bECujh3_$v
zq<~%h9wTvl(VY3ZvHldBJSq#2SsOLpEj)l)wA@mQ>>4ETHe)9&&@|-$FE~mFp@)*7
zx@E`tGU}-#B_*o4QP~?nuU51m6fRKBQO$ryqG+CKO+d@&nvzA0IS+vFn1&;a7Z?>N
zsDBH!I<|!^3=KHhw#Y`10Gs40N2)p1?n>?jrRC^%_#v*u?-PyXNaxU6cBGBJR?@p%
zjI}ZYg0_bM!RQtUI{ClqB-kYQ0-+T&3JWP1PDX!+GRW0R$y`6v3m(-&@9|ofYcBCZ
z82%IQ2^nMxpPy@C-3`TX>ouAzqICf!xPQ=yFOw3{bE!(Fm58=^u!1uVl&tz9t^D4R
zO^jFZ*%cWMO5Q>`q8^K!tUsZ2J%FS3JHp)FfSce|Q3_T1&6J<kXLvJz2;O`zr?m-Y
zW(%}aq-AqUNJ0=`qMhQX_@b@dK=pQn>Y<AZ7xjiTf(>QiV(k=^n((dB+F^}WyMN_+
z#IYTLhI(ZL9Mkx&fE^OOwfhkL!Kkf!v7zFw{dYk0?lZ@8eURhzTgUS$<8fKwolcbu
z_ldMAgC>!>do~_lmC5}`j$yU-Zp}-h5>4B_h~Mp$K<<Krr_nrjiX6jpZAa2AS1`M_
zAXB7e(NDAk*K@mr)b<8UUZ+SeB7f1jdt`Km;uZR@)mC1eMOo0dO)gvnRJX0K)tRH#
zL#*3fc$8ibqRck$*tG|B?Aq<i%=rGy)^pnu{W;WmQpUI1GHOSZxS-4up4^TWdSFTs
zZ%9dy5vX_Tx<s)VIcbVFqpd`l-BB*!41>Y?kuO((s~zFmQ6e-&#$X38)EcSXt5enH
z{U_~_oy$E&5_JImCU*ZHX_5%tmmvWu6_>G1DiZ=VHkXl14knjz0V$gS_O}fJDGUMw
zpiY{%9Rw+g1Am~d4~{8m=^=i)0b*<d^bqxOyiKW2FTpF4wbV+uqij-Abw}A~V57OL
zi6(V3`VDehDk3B*-1~}~bMND&i)5OMp)e*U@=QS_n-Yvtra@F&=+z83nz~TTl?JeE
z?y&G<zM_NKiQsbSak!^*I}IS2b1)F0PZOa}5OKTT1b>eis*w-&n;q6Wm2e%@OZ?~6
z&B>kvxUINIUZ;ba&h^*Iw4UF}*J`$Pm&<P8NYARh>~__TstLdiy9zAYyV5u9s^+4d
z6Q!a$FU1RLCd@yFi))HWjFp&t{yyQn&r_I;>!o*k`fCcqxSQ}FLbFC;QiTt-afU>)
z`|#WQw|{?q{SWG3djFHAVg`%x0KA)i#IF{!WW#aU_xRlj5*5dN$2_v6CPzJMffMq=
zGVpcUDPa*i=KOW^JB?!H=dku`7Kgmq(!`9Q4iy>*Vj0c&zPE#!s){1;RLG!j1SPGd
zued(1qObI7zvrpA>VMPj+qC<Vq^gYxz6uK(bbmwSxSUc;QVqRG39V%F5Y_Zj2$!Ye
zEjp@|(U#vbC>W|z(UF0}Y6cYf&4jJ+oJShI^AN(y5b{v|?8jH+(D8~R#YQn%@aVcx
zGCFh?AxSq%xc>e1M_Tuet3EfM5f1J*8Lo~z0j6f&YJwHYB1}P+>hd%_&n|U&a{e-1
zRey?DmNr)h0MUJ!8ge3EPnEGR7Yl>B8d_Z-jNVUIAz01O(u)jMdJ~xiL}*oYukFD=
zzG>y?+MVti2f2CB=28?Clva6WS)c9%N0<o9oM?H3gupaS337rTX_ZawkGOuV{SoJ-
z3#eNg4YgfQeQi5PXG2RQQF$i&OdX_NnSXTQ$jL!FrG+w+i<uu2+^e~QWQJd5bTn)%
zJJ;+_dBZf;6+MT%-W~6bg=W_Wll6M^`>tML(({Yz#o+AhBt^x3S{$vXOr2P*$>wJ5
zhO@XdJ7=_q7pb1l-;%ZHOR3#CW9RZj^}_TsH*aBfJ8^b-enT?1s##?*@`;Wax_`Ut
z10iDH%3o(BJZ+oVXt=5B>rV@4e8m(nEqAgxo?KR|6vLsK*rMLF-e<eN+is(p5l7}+
zTv@8D9?D{xB%JL<mCs~FvD|!$Tt~^t<)3uy>!sXadeq^SyuX=W#G3b85?vD%0*+*;
zQMji0@kxTvD>8qIdIGABW%lf1Z&dLBwO{U<Uqq;;f4%6efiGHGim$Du%{_W^fL+7^
zZtG7in#O%+!|5QP5saH1?$28P3)jFG`~=L)?uA^ZW)(}RPEA{4s`G1jFV#x={6E(f
zNI{pue<%|HGna8tLlc*r3n{>V@yERW^8VwK%tVhA8mk5GEffYQnkm%o^;sFicQnH%
z2zU6(_=vBXulkyw@eaB~9AB;u7NucGXs>3t&acl#s=x=sS2Fwxlw*OmB3n<w>lrjn
znZ`iRTlzcxhQJBEOixG^-pKSk5xuHFLE=8{k@-p3ONnD`GGw#KH3ABM?QntXE$;ZJ
zh4OaBozACVw{fBsKxT$>TcW5@ETFtZ7K<QUAs{nPS$sh-Dt(x^qBOefjlWD@1I-@F
zYE;y#T;DbsWKew5tg8~p^4AX2+8IICkB|~Sh3lkLRb4RiVJ6j&S07geY%}vRN=T<F
z^rAX4t5TFE;8`_N$P#9M7Z43scoS?lTz(jSeS3fUYlf1@2yNZz*FT@a1C1b%Dex$a
z_NTvpJN^Fj(+D*?2Pq_PNna#SfB$hB?tIT)&r24mG=`|NObCagPX@ivW>TcD*}}5N
z2*<9V^Gf-+=Mc4I|4Adne6&*NKv(&S<aT$yH~vn00eVIod!0pp??_e{!WrIgx8pS+
zytE@E&>h~JAX>fyW(JX#IbroS7G|sO;KZSTU@)6_Z-VGKIy$!_o8Irw&P3-E6ntFM
zSH#=hY4H|4RgRU~O`J&SedMS_q6P0y93AOfio4OFg*fX|yk+Z_@O6;KEp(9p$r1)D
zh;4O}$1^CYmxO(P2XM)L&o`%6VUH@g?+f-E0lA=j>IO%^=5850;0Xevh_$=(ecxFO
zSQQ?NKxCKI&RL;x#)~{BFP`FFPB)k@d9BKGzS+c-@*IfWj00G8R7xwsdE&SOFNKmA
z2{EOP%BFO(NFvZ2_>f`s<!&<~i^77QoMhlVT|+u6uj#6PYl1h$=ti|?Tc~X5!+Ok)
zzkS#Yhy$|?`co2!jGcA}ra8g1L$Ge{JEf1G%E07O#Qt(6cVaK*t-+-Ngp?AVcN8`Q
zo=etT*(A)=B=OrN4N2Ct4xUPypR%Bm8FJ3#bxE;Z&YepBL{09La=yG8k;VH`<WlY)
zv)&G?n=sdZV&xin<Q0@se6TOsonyA&2)n{4nP8%!WP)k;TbudiA|tKSb^k1QB;M%U
zVu;0-Zt>Q<J7S{d!~zXP+O-gwai@n}<2jd|!9OtLJ9NZ#OPMVfbevFD1Jb^nEkjkI
zIE_WvxuKN#W}R>GxpHSbjZlcGmr`am8MjKZP8iaE+1FpDQUEbqA7Nl)V}tgg3dJ`m
z&cq?u?zAGx_pR9(<vr2eXkU?SDh<nup{&fZX(f>E<g+ZN6&e7VVv%3COtMp1yJ<;w
z1wz#5YxX6PTD6EZ>@|nwUQ6DYE;x?7HK&Lw>O)7vE{Z6b!GQ`IG-u&l>G$A+Z%jd=
zx@W3?-g4D@S$2{Gpo2Aa$UCiZOH@!th2Et?DI<)sOLSDh7r9MJGpX<7q<Z(0jem;^
z#ksck1LpOZJBxbT{R7}wYVQ*-k{43Z8fE+Oc(cftPFO!`(4uEb$~szWx*Nrvd_#(G
z&C!$+0G2HyrWs9pjqNR(yDCN*ojNt`fVbj*7;)w*bqPIzOe|}*I$ujr@Nh^iIF}OL
zc(oPYbl3&wTOU+1q!}f&m_k&zgM(@<fJzWUvkIrUVr&EQrKLLZROw56OxSmOp0fid
zPKxDJ@*FQbR{ak{BtL_EgYA|st1MkswP|7M77J6x3dN5G>PqsYrRkG{<mF7#fb^7q
zwv$rOkqkZ*<AOD!D}A(Wzhc;JgBb}^*S@8lEIzf;<M9>)qUoRjmNMs8co~zbZC_+<
za{il=8q?A?)ho@xoa$*=y4uwyS5h@}!s)KXS#2<3Gl_=i)J8i@D~1#?O)GjxZlzyN
z=km0qwt4M=B(-!?Zf$}hfIyBaLlZQARDWdbr*uIz7`BYtnjZy&tj&+^PS?*&&5jYv
zWl8|2QtMc+HTM+<h_@R|KCzFtkI-v@Q)A%)edGZme!aVoJV4s%Bae7%i{y6v`IH8T
zYdzE4{T){4lJ1`F??Jq|X*QnV5PfcMkV!2x!9gli$o`~x0QU+w%E3&3R(kG#N`cVE
z&$*ACs}D*bH~bDtK%3b9d4L3nJ*U)1CBR9&CI9&Iw?<fNk5b`4NPL9$E9V6rX5J$g
z9eU~bKQOF7Hb9&jmbVQn&`*xLb~R{9<S~~JAyIqK5Cg3e*mNFOI>MkNu4l6!Cc>~K
zwqxSuoOvXIHi&pYSVLAO@Si+?6&bKTXeu&5-Jgowole~0BzoFlCR1O|jT6Nr)Ar8B
zNmj?wW5C^n_E55d;5m1NCE4QFW>RxpB3vO0L^soc`CREHs38^NLS~dRE@})?@Ly)e
z)n`qqFbsRmks;NlOM1xgiEAoh_DlIqxAd<Lm9|GfLH%2=3K@^>D#2NQ#b_5R##nEv
zsdGitLTCe4YSuT-Qn#x&Tk&bda&zyJp~BRoD9t@>yDoM!?3wD>)amZ4;oM`|!n)g1
z9BcFZYifS9HT6`Hzcqb=sqQYr1=sYBvu4~|VRHZxM$KL`3atjVw$CWA6)`UL*>4vN
zSRWKe2dMw4I6BbwFT~M*0mFmi=z?tV{Gm8H>cC^-=qTsyI689mS#eb9VE_2v2&9S+
z#7q6Z4Wv?XG2bcXb}V(f^HTI`0qiuXRL3TjVP2{T22kznry3@7@u!>pE=2pIbEmdn
zygAq8${g(|%8hbzb%@D1nGLqx3-4NasVB4Z2DTH@={0)#Vaq0eJ;n&F?Eol@BI?nR
zJ9(j%l4eP{z=BX_Y8cf2EXY@uasc+y==bpEK!3!22O*EZo>9Bo-q1Pnj!|f`hNxKl
zD_f<8I>JjMh5G(QNYc_%{Q*8S=*Nuf6opWvXFT{E)syc?QT%K##U^R{-vX3(0+b)2
zj`WsycFNmOS!&sTC<BkpyXN-rTlZ+S0I(zW9hx~E6q4Q1NOOLlCDYcph<pZ(zSJSE
zuW#OL{&#=|;X*FgV+@uTb)kD-gK~DXV~5&$&ih7?0SuQw%?objMS{e{SpCeHU7FJY
zax!TB_)1lMy-PLueR**yRTJm3GSBexbbM&zReTBs5G#{^<ix`j4{?^Xlvd4+6cwk1
zYS_K3;KR&0r!?z9hi}3Ra;~yABcCG<S9-QH9(tXLUmIg96ixS{It&t6p)<>mLyt0d
zO0V6!COf!8b{*UsB-iij<kp{IHu|G7G8~@qULkdxSEK~jJEYVDzCld=#-v9JGMejB
zxpFQ24>phjV&|8^e<%}|!FCQ6m-7`V#((kq{LAOhFR}>DCA1C&jU6O>m(oOuWR&py
zsuWVIvANag@{l}+q%<zRi@kL0MPrbpF_*K7(jkpNSKp#7<7B3Rx7;I*ir<jwA!gc%
zQlx*aQVbNJNaRXSEWM$XD9Ii28^ol8pKc!JtVOC9QYo&BdnJyt$Y|A&zYRT-xPQvq
z`8C@jw-g!XJq7`VR+`I@HN7f5Dii?S$f-0%l8Z~fRFY|39;=1*q0>py^Pt~ouS?WM
z_A#N1_v$1C(&M$!r@zSS`qm%cmuKUhP}uQR`d<4_ynSnZTC0R4*{Zf>p=OiTRkEEr
z=aosb#^lL*oOM~Xt@kC*Fva&Gt$$nX6=hCu-Kss6R6<32PG!>{#RJX6vwUtrxoS;?
zN*HB1%_Mh>8Q_Yqp&3_UomE}OWv-#sP1M;izSB$9L0!*@s;Mh!plAxEvZl~gseI)D
z>nngXrqOlQ-0*Pn0qKyBg3PP)6pZV0TwG^u#^qTQKvT+!xRhzLz#(s~s(;9_ssfVR
zfWD}z@L}@3O7&FzqdJo*Rvin2P%*?zrV@7>(o~5GActRAe%&84RPZ7MM&G=Ru^ca(
zaTpLefhwF=Riv~SKsO7)Yrh#{wKW34GV-HQlHZX^;k`t&U}G4GU*FkZAD=IOb}nK_
zzx4Khf4szY)F3pul~=-s(|=#To_>4z52sn8loXOi<PRfr%$&6RC}1N<nwPZJm`N}L
z3af_HrXZ@nuI;juC+8?@PXUXL$qqe1l!RIgBbdBdRc}p!EvpHml=UMCKWX{VfDUT~
zSj7k{Ige0i2e1xUXWdDWx4F3C!bg#4Tb-M_l+@j+9m*t?IU-Mi8-F6B#C6j0;|_&S
zxyDUzBZST?j+UY7s3kBkr@aQI?6#;P;~@DrmmE5??xLzuC|aV7D(KnyD(wZqhHk#s
zG8+OM<458<X=(klSl{@NAk}mu*|S`?{L5G;fP%o+`WNGR5L69R396y)=Ca~vN4<-2
z05k6g$_S@Ad5(B>@qa(g_W_S407m&-M-k^|Vt_z1>pa<qA9HVGX?m_jmZZ{gkDtK;
zJexr|vndz!TtQ3BS`%DxIZNdT8N-OSV;exucK7655xrDR3}%_zW4AH^VoLk;xXo1H
zb)8_*0>0Ed_^NQO!oLAoQ~p^gRE!(xVp;UQ%@7jZTj}Xa&wq@8D}CI~Aq<5?iVkd6
zqixk_Tb(rzegoTqFs7rpfR+s*^3ZtLEQ}p8CEaEq?VIXNcVxlfK{x}9b6hiPvjR_B
zCFKB2-sF*3b?!=z(qt^2dvgZlW2WSqT)E7+gv1<VTT1AfYDcZC3_d5TH$j_)^@)M%
zwvhTpn^I*W&3~#}t_C!j*R`c-uEhd1_j_B<ya?o4+v!i=xP{46ZcS!shV5-JJBuJj
zTTXew-_RQYZrs?Ww!^ZbT-H~LVWy;N(8R23fm^;EIn=z`q^m})OPBlN#Ja_zNe^S5
z4aLr*H5>mKqQZ5IHKRnV;<A`YN!ki2_)qn+wHh>#u79aF(`NG6Yl*?8MyNp|-&8j2
zjDN1m=>kO)F6WNQ+qw_YHclX=qaK<#Z0DlvV0$~MdX&hVd`acZHqQKLzKCdq1)eV|
zdKt`@B#da4@MyjmgfgyA0-=oRcCxdty1O4jx>#{jT&^YP>a1|i@!m(EEBQJ-9p49_
zYtX`lPJhEo&^35LtEICCtA#&=Uy<Ok_xAtzwGAd<g~qX_Qn5!qd#M#xSN5#oJS9iD
z8of(3Oc{<aa<yMhcf>m}f&yx!_`oHco=>b|PaTA^9fbj#=hG*4Mm0pSiplbvYgFgw
z7pbL`U`>P4vi>k51{;6B6+X<OS#tvqXt_@L=zmt1t|EzvKI^#c;?E*7Bm88f=Q`V}
ziBt4Qap-B9o<(aCg{Fv7Wtw1FH=LAYq}J{YF_JGJK<zE<F5_uQI6GJIwVh-U7ISP+
z?pu3GQ**Hif`8!5vfzdtXO>NNH{ihz!`k-XVC`JCW)?@$3B<y+*BqWeP(rz{DQIg;
zet#mrx(Lg-K3Ifh)Sg)!i)h_)29Ip?6sQ$+A~VRcuYQ*+a-8CAvFKlGxokvCKlI0b
z6`>VjMn`yjM!7P_3BEr-(9nXw4W*C01Lbw0TN!e{YOJR{)!$*7edv=`nTC63Q?M#u
zBgc`UHk$S=-JWfN99LN%yDKKtVQ#OW=YONTTU`CQzLLgLzX+i^Z`d;r%2XxU7ZpWY
zoVtI;u_zy>PO%)7>8P6->W?3FXc?%dd*Znyqp0_LKVV2F&5Y90Nu+VYk}5ghj{^sw
zAOcZflpY93;AA?~<`m89nJVvnU?yG*ARNFc+2b<_K!C${6c_UTU_`*qCPO6LAAdVy
z=4#Nlu_J_<Spv%s2efy|F`~^6A>`Y4Pl^S~J7IiieB~9C#JQdM0C5*#b7Q@MwtIRY
zPzHq?PgS1~2pC0!hXVnlv=Inc?*{?_qXY|JvfP_sy$^uK8UQ%ounJ<i<sMjMs0qN}
zb2sRJa`9;;TrdmvIYd!`@Y*hD{(l+k3_=hA{1>N9jO&xix{T^}S@!@KP>Ogs-|>21
z%}`ixb3L4$aK0Z4&_5>%obTK<y~oO{=}^>!w!F#!XwfxES_`bE<=ueE(%SU&)mo}-
zIBXm)jB|H$2rY*x-Pt!LzZvHBZ<mDK-6dh;Pl1rUl6I~n3`o0`9s$kpuYU#QI~Lf8
zRa~sKKCDLZhVyUSS;tyEZ?B>$BYcCq@jTqXY^ra=3HGXdJL+uHGxew4Syw<9lyYe{
ztMoq<DmzMYg!Xi|)bn)R%0Bv8T6x!9cFLSnHpjS+x6RuV|KYT@tu^mDSvEmP0uYQF
zybqvhQBFr&iRwmf-1(@|O@D6@54F;b+X)0V(01N7(erveoPsWxUglZ}c*{CVz(Y1S
z0HF2bs&91N)owqYD;RmaRvYUl3VErPplv*N`HD&{NG;GkGskBXpLqG0QQo+H%zM9$
zL3{AnYmPs2SCdUm4Q!}NrKOqyR@<@y>jrqjB<a0)OSc2ym{N-S&O$$Mh1$mpW8sTY
z=#=S$wfiZtd+tvkt5TO+u2Xi86sb$uoe_MXNL^fz5nQP1mdmru{*pP^Y&2ru6zu*6
z0r-^Pm%)E169O|am%*6>6PF!w0VtO_ASsxC^!U@?@4r9EBD562#-H(v|9g9X`tOOW
zfHDY+F#PlLDZT)sgz<VH^1@g({Q7tJ{`3bY+n7mOlRr=nzkUpC<j;a)kc6VN7O_K7
zNrDyzYlATHRYnnBr9?svF7%N05K&M}OY4Y(GJ}e|660=!7=OKZwa67{u0hUKlf8g{
z@JF%*65bep6UK;Hf+$HEcbz1Ss&@j+OczPsKm-z1AZk$bl9$t$Y)7OS<}K3{5OJ#?
zNvm$;&yk<jK?LVHKLv}@X#+wVJwsaF&KI;G$K{mp5}#$>h~x@0WbW15f{<iot~W08
zwqDS4gx<~y=iH1h`Kp%8H>>x{=9}GrNb=3)h|oTWc|cS*OhM6y`<jy*X$9g;T=`NI
zM>C8cNyKhsqIe5Z^ukc|cjn6pjo_@#oZW!_>FFG-bK<COzc%!`CbR6O6=MZZh<tKd
z>A4=2R(h%*pH_N)dufFs{Xkk_dWQIHr<H!iqiKcdy^~fR*Sn?_#`j~=3KKVf6c@Z?
zx4bsB#PCuxqd$^bG?`L`?bK3S;mXvqquU|rMH)fOZb&b?siUVj5AeGY#-N1-(~=T2
z#ER6ErKn=9Seq5j4ZS-g@Kh;#XcB}%8H#-(VpzTlf+<bDw1j5LW}(Z>OQwM1cJe%~
zgL4d`$EBt!*y*-?n&MT6qm4U%S#083BxcFuF8&yVvu)D_!azS4UaEXoS#0hT>PFL0
zR1>5z_aiZS(3}#i5z@CEvO8)UEG$AM8Oaq(C2P?%qLQAdfs&(Ss*FXCODu}*Vn>uL
z?h;AecFBo!HN<TmBh0dLPy(@pe{7tRC{3&^<wUY1)zoe^i(IvF$9Io^TN=FZ(lt+;
zM_`Qu_9n}%OSfoz=@fOQk;YsGP#++rp{5c1wBKzeI8XfTd1}EjUakIGP1rOl$Ts_T
zTBY7M>Z^-ZX4EKY7{6azpZ&0g+6wR6US#a(DHvvwt0J{rlA0W?W~QsKNFG-(sCYDF
zeYUt#k87w<gtiqJrUJQta#gNIbxo_5#a|Ms37$+pt?aYdZDgTZ9pS?B)*(jQKDQ5J
zz;(%zr=XSuNlNW+3~yRdT{Y^JR@KekReD&{ZyQNj(zkz5^mikAg{~jB!5l!1=7?lO
zsFV;h42-1ij9L|rfoyf;8hR8{LeeX?vbt=uc6Cgh-8IV%)elO4B4~Ovoq2&#w0fRw
z^_Mn^C!MJJJfrS6+3Ols6W_YUj!kB0340Aqq=J;HG>%D<_S{y_{Lj+2YKMVpEKIPi
zn-!|cY+8rvTS09RBOTj3P%U4m>`Eq>9g~;<H@bbHS!>$2lSve|ol!d@wDeqR7^_a%
z)#^AOPlmEPj!R8{7$peCrE{WM3cTcT{jK7lOGTN>n#l?^b+C%u6pI`L4H-8k$CgHG
zne;psBGrY91&WOltx|Yp?%QrjncGt5#l7LYfbF7#Qp@adbZ1Om+p<lzu~%ROpqfz}
zy>UwqWvQyBes8vV*$<L_xF%=oJlAB^T8lWuxWyBA(54lCm-U3ci;hWT)JTz+9p6Wx
z@G9MU-!CoywZ0z@?ahqKOYzjK36FqDPdxg1ob5)x-rQuUi*#Yeb<zK-9-WVkI$q&p
zfFLy3ncYfPck4YzJP`$DXZIXb(E|0QpK;FthVmjDcwo>ITs<$i{y0imE0ooYFguBA
zlN|`_6q{^+mX0OQ(F@3S^<eJt(-%-=KW#qwaF=n7hr77h9IpW9MbI;9A9$#%1lYj&
zu|wV4D57wBW`)^_%k2lxoFZ->JaaFQhx?iFfh5HPgO*X86o)rQE4s7`Fp38r9M^on
zQP?|O-S0|QQL4C8i~!iHIUZnXv;y^KUl%d12VEC`F{=BSCC#`qyOdaDJXf!b$gbYa
z(7xx&$cjMQG)GXbZdf{=rc!OjR@{AMlrMp}-YXf)X_}?ydQWanZs6@v&~uCK!3&C_
zneHw7JamI&00r96`P<?xPG9}@w>Vv;IBAgc(XKM9&GVsIKpV=b>a(*z&-JJ*&{O^J
zOJl}FZ|BmOBYpp+F-LDJ1LR9%w&KA{W3Ko1rSSoJmrG;L_jO(Gd;!eWJ}CH=7B=YZ
z>)WlszvKFLu}fp$)CT?!rUe18m%)E169O|bmjP8R6}K`cDQpG``OZWR{(sK|s|c6D
ze<%|HGndg<0u`4ki3BN^`YkD!e@xHQzute@NM=}y=;%-S8Q)&tH@|O2_Q8X&(obK1
zZ0LhEN|>PWvlUK;>5qR;-!@-Hv@<b@bMk*6r$0WXVdPhqV3a69=^Q{OK_yYFc5*HX
zL!Yt)?TeIv60mRh;OZ^PORHg(&Lah7CPlOo;cf&8e|igQm8wNp10z+9f3k;|sR2~V
z7MPRJaWG@^K}5a4CQ+L>Q37htdYYZF$hW5lCrRf(`hF!k880efs{f@GbtAu4np#H@
zeORnO4l+6eqY1(WKVnq&a*S3BH0}k)mn$$j5z+icjBZsg8kAV71@$Q<DptL+(2v{i
zQRB5<SmW<)yggMiUNSK<e=0C+!1V6&L!6K)f?7A?K-d;*Ic~%w%}WiMl{GBt3VVhq
zJj4;>HTLYRkf@*$;}!Pog1~2iagRNFEkNsqxyhmq{dk$^qY*wXRsXWeg^^kqxsXCK
zvyhD?8BK|8FT`Q%YQWxAU9F59a5aNQ4{<@&MmB2bc-`kZ2E1@;f9bE+v*(;pYGG{2
z=hSJe;U*Hoa)IZo(*%QI#NW_NG2aj-suRr=A%(MUVNSc-$6XBqQ0k}0-J>yez}Uq$
z<{QjZ|4V##BaPf<K9v>Vs@U+6HI!QRn)Hhwz?aZ!B14gH`C{q#yyo1Jte+k|gkS_;
zs{0(aK`4pFEDh#=e|q+C_Grpo(W9A1`C1~+{2j`7z5d<Rh;RjW5H*^anWYN*`RMsv
zfxHn%0S!Oc#yw7<vzAn2mcGc*vBZeRQI7umgn*x?1dRwx3^4sQrD1e@{D<f~dLN>|
z#K#3k>~B8(@8jD)zW#>U#q{wffnuVMJFbdh`oXUl*bN%lfApQ-G5z-erz2&oJM4tn
z7X*y}W#T;Wb=p|%kvfEC^gE44%qlJJT6J<NFFXMj0?kn2fiTulruV%aoVl7w<V6%V
z=#8-OTzbX%Wfr~C)4mr_ob{h+^KIIEVX1<1DLV^6W00U!Em*rIRVA$QSO_+fEkhb(
z0Om8XNF{pof7Lys?RqOH8W$P7)j@dlLdq1G+Pvm6(yY4-A+Ca<49%<!<9p;VtQALQ
zBgLqR`-3q;qZ-ge6-zfuxc>e12haP!S)ZHVjORgVBP@o>2e9Z<5C~VQ9`Vn$`kTQE
zG<zP^+iB}H$=3>%HSZb_Ah|D7!%kk;LtWUHlZAn?e;<qa6NEADpN$YSTBnP%5cV`7
z${HBD2x0f6Oo~3VYXs^%P|tw&l2qTS(u=^2O>5rlQnl*~Y)SkHBB%N@7Osjj?!0GZ
zY-z93WhZ0EU4BnH9L;hw^seKC;J3BTMl(c@qjZFP&y+a47#(t1>CGq%eHvG*=?vDz
zQpXWve=4b(y}6cGsjJgsTC=yR*)@Bswv+Aczf%v!d%{DrDHJaWh`~Fv)}_KAiSnBF
zbLuF8sq0TxF=rX56B~;4#0Kh(i$Ti;VUc3)K^^OqSxl6znRB5&b17D~R<yF!Xc%o+
zyJC-TJWtP~z&4xTTu-B{H~sAk6g)Fkw#ClEe=~$+dJVn=#r-^+<fhBqXI@bwL#oz{
zPSRBj<y%f?Kd>#`*;g`5v1nOlcHZr(92;r{@2*JGtxL0$TZ#f~>2_NvB?@oD8iC{O
zpmVUeLC=meE^ZD>RncsSPzghCY9)Mhfecwz?y(rulX^U6Q*)dP&6k;gKd$Q>rOxcO
zf1>;ZB|T%em7E{5jC*ujt<hNVn;RBR1IA^`=<G69QAzFeh<WG;)gCP6l4^uDYptv8
z%?nXWzy99c>w)T(-Rl9{Y29nNySOMe5ffYGYIQlt=X$Le??sH^&RQ9$Ua6G<+X-6P
zU8fa?OEWdg+1*|%lxk~r!`L%o4Aq*>e~C9>sFWVDUD1J7j*3e^VpOoMagKKb^d&3d
zSU6ykuX`H*MkpT_bkA=*Dr)?Ng#^_VJ$#9-7nc&WSM+Mw3XM(DUg;jKbgp)D?`ZC8
zEqGYlu8(EpI)d>1x$&_iXzl^Jlk#ni@xz$9M_9~kycu&rc*;qbpT9nH#M-s+e<`!}
z_Fyhe5V}pxGQ3r0YfHUVa8knNANvYGAIjG9tdhOvHz-`%Wd+HRV<8=bxt7yrDV=>i
z)q}D&Kn?GOtnJC#e2OGTsvRK2(!baiN!dq~-K=#^n)BLi)?LckpX>Qc+ev0^%rYCK
zR4}dZYVC|=EvwfKFmLU$`cO>?fA~<UwhNN#{Yt6dv?1nHT@EGsA~7#Z_j{LK+6)c!
z(8M^rEpwBe1!F@tV9O@ii8%9b8lK_05<!_De1QLFiJ+|!F6tX@(*wpEmIwppBPGHB
zeWwyZYbCt5cTPqJs#hkX1Gb-DB53+m@Dob}6@m!y-ZwV}s#j`d!1mKie*{G(NfY^R
zUm_@QLf*7Q7%^T~A`IB}#T!olN9Y3wxAFp>FoleCcGFY2LDD^9q$`Wm+&1f%Q<SEC
zXqZ&*rCcEHa>s2U?iHsz|5n<#JUF~Oz^zx>1M9os-VNX9x$*h}2z<^}Ukn0Ga<RTE
z!UWcA@(vXihLQsnmXc1mf7<9QhZgkcnu1$>>fJd?2~R4#FeqzdyAw%HcQM?zH<!xk
zlh56CZPZYIW%yEm@oY?O)ymmjslgbEYp{%Stic%O%o>bmaXrD9q2qMfO+WWEZ*SeE
zoiJ#ny>M@+)oNDj+O<lhJ54<URJoaG+X`ORf}pFD{klG(vyb_df8oIC9M9`c=boIY
zp?`5l-3jgM&8Gdwcn=2RV^d8v2Vn%6syY|fj+4^7#Hi_y^kz!s2q{DM^I|=^YO2Zd
z73y(of+|K9I&8k<f?dq+va?v<(L49V31VC4vVu!HrsQs^6_Ft|bNw=0;^JdMNXLZc
z@+0qkPGW->8l9w9NBD);gO7zdXNA$Wt?8D>Ws76BP!HR0uwI2*vkiJ+<&Hst3d#_}
z$Uw2!ZDU71cNMC;^Sfiq4$YMWx41~qTqyV(W`rl#m%)E169O|dm%&si6qj)~DZPKj
zb+h~Z@z;$k3QG}e{2FiV)7|6d$EH=EgBLcK?&)<CU$7>jz3Ry0uDYLpbkCcImTNCd
zVoLrnvitecHI%Q#fkDE7(pqCC4kQV$a%XK2I=;#d<j+zXIDqeQ!FYzd6itl$bnj`A
zBTIzBv>f44Y2ZSJ%Q$h)+yqBG>qmc5e?o!!J!fkLS=Dk@a-vs1+Oe>X+R=KZ2M7qe
zgq3txbUN(PgZ^YESi5)l6JB#*Pr9QhC`p}Az8>N8sZ<4HO~}%W86>kQn)p4Q^CGAj
zx9BCa<h}dd*2-8yCi$`W(CBXXNY+d!<G5CzX)D<IK4aRVS|KXL!q`g(Wg>s{)R;_|
z9TAe-(?TM(uu}CQxm`UZ20hMtUa|4C9Q}|=?`bwh3mtsRZC|soB!YCa(J`<gU&`kK
zF**lcen6*SL}1-YA=yObc#g;$L1CtMS{<?0kzTY`k&25@GwFmhGfoPp%wSMdVf|EN
zFT8YXrdq?1OBu!su7z2-G*y31!y)|X+2$)skK;@<oO4;D5A-ed{Y$0NUa)tqf@I!~
zt0AFgR^?`*PVetV1MRRMlo|NeZ%F0(?F{iJl#gCB{8NVizvxf&J5<yq_>(!K*q^qL
z7|9|ydMozwZjg0EJ}iM@fa@8wrXX0k2(5o~82F`2qrm{(-E{wUk&J&y7;MwZu9+xA
zqc)%ZetG`o=^NJh?&Y621IQdMBCHCp`O#P*y>qVnUY_B1Z3)t`8fY8WPoZ_dfI6`!
zJarqW1sq}gi26NJp|EGAMw8Qiz<ECi7?G+Dkz-G-^1hFWRjWZnvZ+@iK1D8NZhXZ2
zIx9ZLYy4b?V%DGC=DB~{JoNqy8h;5cNxS4u5=;(TwvkIK3F(VPsb~*fgG15mp^gph
zi3;nI#X`x-3y0K+V68~&jjfcC!IcDFl|v=fdrRNrfX3GBl7cIQ37BW0YKU1$#jV-3
z!}q7xGVcYmzKrgf&z6)H3@Er|Y>OB^2CuZyGk$TWT!z=tbvu7_z0>AufTOVqrG^ym
z*Qw;H<fS5Om_mhzZlq$ibvqdQaI}N4p?6Ln;h?sUjU0o-SG`;I93WH+SPL5M63vde
zfXCd|cUzJ^_2k&zbBm6o)URwAY0K-rqnCcEkZ`fsV$pbAd*+Nf0cHU-*XpHiM<H^f
z-9uT#%;!n?s)Bzxx;u(z7HRq6+|Z3(UJ3$$9X4W(@Q7lQS!QB+BFTwEq9c+FTbH_F
zMp_kzk!E?K*(s|24|SpLV^B9tNp$4z2L9JIJjwq(*^d_6A-CTXnt}%Qr%+0_Rlz`m
zMwU}pNiPiZR6b3S%u>zP>R77Sq55*g{?$!~mBn>CT3vr@?}0MGS-o^hr;%4d>7mGq
z?PUj>w#B35)f`XiQcC;Mom5YCBdY8(`OrC`c3K$KLt9<5!mpSOMo8TXifLeK!s?O|
z6T`A1M~0j{paEjoNU61@OQ8=nG;y)-NAynIk-h32S>C3)D?D<I_is4YRz6B)CVj&t
zY(4LYVC{d^IMH`XUF}QuUZd+#Z@U>Jl;AGATssHU=2~wQoV(A5+OsgTNn9El`)m{!
zHFG=3iq+bB?<ucR9HH9mlo#6~ps|Y_^P=QLAX;$`)Now!n7844;Fz}=eW7Ds4S*Q}
zU-(3^k$U2hV#Dq1$GoU%bVzf?V_p@IL3HWE_eOtejLLB<{c?1CJba(oayU9w=J?p=
z{U-<!vZDdD((gWg(r5rJ<jkJgw*DO)E!rvK({fzy6Ur&{dDV_SF_)D$K*)Zy<h*gL
zafTx;WwlUi9-cQ0kA^Mv*sAg93Knww4OFdZV#9I#`@jb>Km&?dy*Hni7V|?mnF2JG
zX8eC^c8^KqGA20^|AH*AwLCxmZ3GRlYWP7UnJ_cuIQ;7ZikVv}Z{S~7wt#gLOcM-z
zQ<1Ab3IYzebZF;kh8>c!Vr%Y4&#`UB(*66lm+&g{c8a~<M&iC8ad$Kl|2Vt_Y>P?T
zpxdB@<!}*jON^`tYj(qi<BH(6;d}txHlsi<1a3(=NWb*Y<&D%6!EM9s?BJHj3O9xq
zt1tP1g$v=rL756YF(y-HXa64tuf^~q_3S?w7W+|iN4xtEj)A+EAqGSf0yQ(2F`fYv
zmtG4%E0@3iCkmIpJ}HZTJD>i1|Mj4Yz*+?xZ{wwXdVN3qb7;k9Z<VpO`|HOc{y_$m
z@z@2UEQRjp-`%&vZ!OizOUo<yhjsV!d)M^*UKAL#C{SC=JW(KM(JFP;24&)3S%LIb
zYZe92n>3j`iMkM)Nm?fyz;qCKCB~B;Vf^LI!y;Fbv4&f&D5@ua;mTqLmV4&%Y;%i}
zu$Y0Tl2oeRDKfLU=txcHH_2s+abEiE^kj83ea};R5?w`6cl}SA*`(+9fhQWT$h(#&
z@S=oBsKnda(fo2hUj~IU1?tzvME4L3%5j)Ygq9~_Nog^3%fG~dV$%&jiSMLm#J4&i
zdu7D42L+08Fe|5j+{x_9T_e?+7`Pb##wv3DjsUpn+zmxrTrg4%rJZwsx<)6I{K77~
z?8qsLo2I7?8b($OVbE~Qe$veTy3~@mH14U(hT?Wzo;T@|14?Q(RhQ<1E|)U)mpArZ
zP=u)M(pn~2PX(YM0{fLJM}=IpbQ>y4rJU7LuB1HWtO{9wrpO=HX4Lqcguk`aG6to6
zojUeLeLWdY+#+8S?+DBbW!ky7{15Ipmgt0dye3sPn=Hx4GP~ZO@+MS{bS(3q^d>n%
zp_bn%WBIv^Kc|@&>5@~0O3zN^;G*ZGcRT1h&G31UqEcIBBf48F^{w;x2}_vC>uD1M
zU#lFt$Y2M5{Ul0L-D2jQbvsC!`ty1=^967Uu%@^)GK?s^6^)kp54tLR(^hPama_~a
ziIcXd_byig8QIIWNtQNg-t@sd>m7C`GcT7dh=jarXI5tr8xArNu8%KGP0Cx2GXtz7
zRksoVl<1V?;S~pG%~Eud5At@6PBXE5Z}nAiPEK}z80adc8hN|v8jvJ2Qp@SVUEsy}
zILWoLv1z!pF_E?!SUt=3-c`MnOHN5oT;xd$kgL6}r&0a7W}{K$`<~0H{E%tgoi*!B
zqSeopP9K4`M5k{fv(9C;Z0u@2&NAvOPx08h!H{Ha<@5{;fy4z1!1xR_MC6r)JvT57
z#W@~-xf=n1wF)e@lu<8ZBS(VHpdD?Na8cT3iv0?h*kMCRXq650$>_4-8uLn8ov3(o
z;Y?+v>NDF+b6y3UQM-*QH*VKS{M{MbV#FP{=_VNE!GE$<8m=3y(okDrm2De;$wYFk
zAd1iXTg6IFBiJ$bv=seA6$fA=0NuCLntZc=#zTe-1}bPbqqwaKgxaU&qoKMcUhl8M
zkX0efcoA$U!=$=SL$E0)JyO?c)v=*cWWAEMr;N@j=Vp$ezb+ftvy#@1e#Cr|JY1la
ze2qqBw3+ce-_6&=P?tscx=n`BK^cTS_@Ytlct>|ND0?=uqbnF=l`(#;YKW94w}UHx
zn3&zzuH`G4D{WP`o5X2l+{^_*KdwPp8U2DymJ7MY8%fRc(A%&=Qc%GUJ6*}UjuuL|
zqXKUXxr|bBCLTt`)q3gerJ`_tAG7u@6^btw<gQ@q7_NwQ=Vz|hNlMjieJx}a8pj|W
zH&g>|%7Du<oSypH&w?D~tqJzU&{scy-pUG$wYNvaxQK|;n2Lz=8`lsqu11~MBVv1J
z9Wj3FBj#x}qnEUwwVw2jbsVaGE6%MIC_9<ObFD?@k+7uaRG989U~A0JEboHyZd&>t
z&#_h@+_)k*AoG`T-C|agy;xm<Yt_d*y5~Nwg!Z+rq2fS=rWx&F{k94Lu&l&?4F6??
z0NKet&F#2LZz%3qA!sPys1P)xKdC~1hL!p30>PB&MkG2V_Tl9LS&+o(;RDM9DEnQq
z=Cf@w<+{}<Q)(Yx9spUUS<?rW2fzl!-tL%qT8i7s0}ZuH$)-U6X7nR`OOk76@i?Xd
z*^(33mf{9b(|cn>aRUbLkGkxCC}Us&0A?oR1q`l95yKAlL%_H|uVjlL;T&dc@5fz;
zf>NBnVUqTg&klUc384MlwXZ?#=L*Q2lGsY!l%qg9xPn8)h->+JSyBw_VG}PD=CSlR
z1jpc&z=avb7K}cyzOYwz^bzy_53T{cT*LU}=K8|`tDv(`QsjnD;wvD3CM9TjKg6tW
zjv#dDX&^M!kFybmQT^z=439I!Bj7R69s$byK2JZ<U;6BB5JfMy)Pj=={8dF8Yh>vN
zEyeZcI!*zO<tQ$M2A#*L31%Ig|6n&(YQ)&-{<)||kB@)Vk>`nds%hLD6E9(KFAhBV
zP<H3X__(_y)YAhiPo#B!1JO&*7<~7K6wV7pgb<F?GQ+Y8%3*-<=;44=s6~#R%jl2|
z<%2@c7mi$oups|}sfKw@4hi)CRMX@qy6U^4MkSko#vrqkPw|*`uYe*BxltvjoO&>m
ziC^_`*0KYbIl{=CGj%Nak+X{JP#PT<P;f+Qh-P03b|bBg3O)81N};Eokj&1#bTSf}
zO-H%Dz90St4atu5m%)E169O|fm+`Vc6qh}fJSdm_R4ECUu0|=2e*%pGwf*(s=Z(k;
z3+_$04PWEa?Zf8VW>g=o;nrIA__7H<paQO(WKt+Dm1Ez3XHT2^QS3CCpfTw~3-<k)
z4J}_Xf?g1U!WgYP1VIq2$c*t^g`X0Fcq4=+1n`b7T0Ih861r7>x^u875e3;oj}i$|
zp$S7I%rHt$(l`b^f6GTyziT1&C&|WeFmjYEMW@%)+afm>#@=eC2Q`ol1+5ZpvrdO0
zJ%}e-j>?{ZC!uCZJSYZXqBv{U^5qE3`&lI|R{J!Y)^%poSrfMhImf-6Ned|HCG_sL
z7e<6Vq$7?QL#<f-h^y&ZTDum^l;LRI<T2L+J}X!SkZbeaUPkFAJ=Z3kup>;eJI^E(
zf{c)pNp`E7L^}%I4p$63qr?EN&zDRX1xMAv5c>2WlnpV=vM&g^u!Ux;l#NcZ(<j*U
zWSyi0j>eC(fEcWGibTz=oar6YCd?AvG)g(Qnn)>q1q4-`@f?@Ie<%|IGdGukk2Mnl
zHZYfwOb#c1T1#)^I1Ik$SNIQ9l|)hxAPCT;?E-t-L2ipZEoSC4hn@d_SW1>;n{ghl
z-5wOjvO-cHA0p}K?q{dFKX%9S?)e?C^D2<>wmbjm3_sBPX*5qd>bx<^f$7e_y1(~Y
zTgT(b1GNU=;rV;uD;=J|1YTkIiQ!eo9G*OyeR+O=EbFP<yMEXcUo8)l?wLmL)+Z`V
z*BleI!$0Rgce)6yRdnfVy2!h?^X|W$6ra#4g6`j6yEKnr6a?ED@CpNVzrS}sc7I8#
zjhE)%^dF?V-#<Gs^Q{0FH3O)fBboqcG^2{0b5TKh)B}jGS`!0+?>u<($Z%OR+2r$k
zFtZeYQL`#YQe;FmF@(sFVW!+uQ<&teA6fj#42$0=JEx4ZlCsqkx!P)v%6TJK!-X6?
z!79|eOY>G_x)14HJdt$=egU40nrHF`I+GY>u9KN>4`4p!YI(C1du~JnXQU!YxPn|z
z(OytXDA62x?}sO+lM7MCv0%vdJbdKcOlHP^daam&Q|Lppn4`dF1FaZT<i1)d6Q)Tg
z3Auz7LD5UPcl}6$B!tP#>n0?f$er9*O=a27*b%%79ME!@;$SIZl#neU=v3rzHYI52
z;{=J=6fWR-o5j*PRGJkzUHlPTJcn;<Z*yHlFtBlf1p4BXSwb+X3;Q?ofz!sIG{*IR
zuz<D-G;C)spq(P}x6P$3PCjogl(Hf`X;ADsjV%nGq_xL-$7&xu>|*XhhwFnHv9UN#
z6Sh0LB}LzdsT`J4_)6E;5>x6thc1nMBx<qdwApb=6qR>w%RLH;VPxXVZG$kVp#8{M
z?rVN7n|pqz&;>=&JDAIJNyqG)qdYf%$NXSY>Z-9EAbB^Ot5VXO%Y5%;&=?sP-ds(S
z%8^u41Z633O3EI|BS6LFVs<zWFF9G_pe7k%+X9bR#%XRVLcLeE<2le-lw2|pHYNWl
zZ6`X|b}0JZlz6UxX4<NBkY;nPMFWoE##tHbN*qq~{0lc2D%AtJt)`c9?q`mFl|N#8
z4kd3pWjX6?q8r=Y%{(ugL6z-PcNwN!us9VCF5l$M8fm`?1*?|B9IV}6%VTnp)f>^m
zvdD5F<XD-aF0!1LY7bvh7op?Ba8VwAFYpM88`edF@}0U!tp1|92sz_3_KP#{Dbtlz
z(UjPy7exrf$~B)@6rllS5x!b~CsVF#l`^F?ol|&b!Pc$gbZpzUZ9C~W9ou&D1s&VA
zZQJVDwrwXTd++n#t-7qWp7m7CF~@wzlb@RWd1HTB6&^Dww@oz~zaLs_O|>fca#Xmg
zkc(1Ndn<ZiE~ERHlboDpxo%s;^6ObyVx!@sM7cC`rXJsTGirDIr3KfzGMY>`P5G)a
zzD!ap8fw<vmO!{%Suw-!dTKDg2-+k9td&ySg&M4P!>Iz|-m#|%Kr#`@<GJt#qk0@F
z!PKG{5hovP2>L=hA?(ge7#3?R^Q2CyU3BGc=2-7fofNK3(lOVpQW0ox?&6JIKgPjl
zt+^r-ejdkX6I%JmEm1w)qM4zE+-SlYi`r7Rilz~u?!YNmP3X}y*}uE4b$OIdMD7@U
z9im1Mec}Xm9I^EbumiOU(>9pg7MV0^(;uz{u!eZHswYTu_M*Sw#|0`pBA{NCS?J?M
zrbF=IX-86_MFH8sMQhWzxXxLp?Q*dvtf5mqlnRJOT(oLZKs*0N+;GHwZ|Ns_z_*YB
zE)h_Y7^CZee~eD!YQ+zH5XWY_eCL>c9lOKVIgBEAZ1sx)Xu_gKIN)f=PTn^p%g|Ib
za~m%zR&A_S7Gc{O?h024m8_yY+cb|Zu9;!rzAswz9B8SW*-`>6WivJ8^|mPL)A*Gr
z^}tH(bVm#NS{USEs%~5G_|cb478vx}>#pmjNauj%C{cDyJ)%dh`M*i0yhvX!Nepqq
zB;S$w(F585r6vAo1~r3kR_}Wtx_K$G%#!3ILE6Y#Ba87&%ECODPtvEocSQYY$@+>o
z6IhoQNNiI=P5<UFpPX#gI`f;&?!&E*#aP(LB@t+7{-ZzzLlc+j7qBK-^;KzI$fOZi
z-=j0jifbmy-o*(#t#08<+!l5M5>4?6n3KuRFoxwam_zfXv+v<JP~yvL{+u2qu-VJn
zR?5_t^1n0JXGD&Ak96fZP7=C&Jr6qbJh2yKK$LR*ps$Su=;hu_Rhn#Q-b4=HT(>?E
z4i<xl|9iyzjO*o{bE2j!{HCG>?2E&8zpI}hIJ=G#9Rq^_vv0Ku+Wk3Dd-(1A+!%88
zEs%Jh$Xaha;y8CiIFW4nLLNeZ!sku!aC5$UdU$hcY;Hf!!ilTf4M)h(=pB$8k_-u%
zTj{@hIC3)*xMY6l86VNeE>2Ie-L>(0ivg;u=GO_V34ZeuLE=t%oipwUpzpEizM*tW
zgxRyp+r?{_6gvsH0QUX3KWqu42eCr?j|BW#bDT0D5Tff1BSe(8TSE}?5&%oN?@+77
zHT!(LJ<u|6&9^u#Gm-q!=lSyLbCT^%2LY=ETUuq3jfL2#k@VxLV!l;nC-j+39|CfY
zmr0~x!swSBQ=mXFqlyQhtJ^EY1rbAuq0iT+Dq}n=j;gbAOt7$fMDd&2hEFp3l0KR$
zh;{|hUnC)spiSv>AM3EMh*@D(jJ|qlS;;eLg%W9B(9s1|VhT5)+5|bw8}p#8PJiIf
zGg{oYy$D1)<SN8VLlkt!OFkr4jWGT2dR?1VlVKmB?n#Cp!vX+!Ra5e6`tFpdYo7Vd
z4b?~ebZ_G=e{{BdZFvd^M0}=4@`J8TWSr%PKu2}w)M(--GR}gA=4<Hi(bZVjXfAE+
zh6cxyU>YElvR|^=-OZcphOo(!SzVAkwqoMBq(SOX2$6P8qtEE|M5e9w#rOu){-(sj
zW6S)?_F{n&6aEd@Qkqt9mV%ecq~xLOprsH|e&CcwE(kk&GjdgcE+n<%gS_En5lRx0
z&2{c=;1+?2!wq%#CO*hQIaX7DRhlD5OGK12Q;3%X!8)^~b{x63NU9)8JNyHKej}gv
z==9~2hel(JeY_c^Z+x98{Pc)!CRzaU#`2yEvBR#8T!WM1p-x2r@TD%!qg4Nt+?&dl
zwc-T_Z;XX|Ttx+P9hbH^d!-LJ4ZlBmu+&5nii6{OuCBsXZ*Xj#u^!<XHBKC<GeV>j
zv@5A!Wb%>YnpBqVNauFsvs<T3uM75q8Co8lgl=4!ki^z(Sv%bjRO0>Nmkw-&2`dt9
zFsX6}xN%WACCV%T$S4H15od@KGV`5Q+rP@`*b8U$YeKu}l18DCI(BgLrrN=96ZdCq
z*5VXNDa!C~SD@cJ{wBCU_inIcAE9cWcJ)D9qcQrl)(ew9L-}dlZ^0$1ly*cU+5zK}
zOP=HECBY+yZ<6B~SiU$REa^*KwlE8pT^HynuS$0G@fw^1rrF<#=_KNtx@c!5j=0Wc
zJUdw8r^clPHk2e4EiPa_8*?&cbcLj-tvN5bR7T{BB;LJ~_#qUzxZsx_b|c;}Ek<Z*
zX_b3N`$KeT7^^_i;L(-u|I7^>R2(T>=0H&wR&XV%r5NflPP9X6N)tZDZ>;=E@(xBS
zu6g@IT}(C)xX&8+C#BbQ#4yI&O8J>Y5~P~Ljjq^nqjvr+Q*2^)_@>$w?-)+eTHX7&
z$6|S&K8QtO0_}^UXA*%jJ3u<6%^R8Of;NJ6j^>`e<_lLsW}S9l2TwVySFBY(&TqjL
zjkm=^mZ)x}Xd~J!JF7ownFsllH<Ra}S0-OFOV3&ZfP<sp;3x5|Ik!wc(WG0gWvg6?
z-R}imW_GU<Z;R^0f2M1?Z6cPp-qh+XPry^~*LlYtF>#e+;*MU6u3MYGb@|u0@dM?X
z)c#kDP)sy$DlQXE6tq=c#Ps8?W!O>2cj>4Wd?Npq3NPc^I2O~GiE@J8VJrGTC|3#5
z;`aayKx@CDkKM`zGBmngQFldk6MTzShCJPmtSXPlq?*dMc3CumopIpYj?pdY;(#7B
z>v~dmu7?U(Z0K5=u*ab`L|<m0ZDH4^v@g}}0!6^8om*@}2my<97P0*R^>04rt<Qi7
zUSaDAAM@P_g-@dBGzxG-bt83`pb{<=i|+{llt;1PwdA&&Y5lgQOzh$yps!SlF7tC9
zX~Rk<96{{a-GCJWLT&RJwvxC7uyLLL?O^0SS6$`c(Z1=nbYHq$bI{6d6k2<COU?FA
zxpA={$vvaI`gum#2Jvm(16h(xG1=1hgrNPj%jHPR_IIY-pP8ox-b4MR9$RospF0E~
zU_#L^z#4S<)HvfE)a^f~f^{|1w+~c7xk)Ew4LW!5R`GLo^ez9bLNtZe47DXdKl2@m
zErOK4H+C0;5VuyPF<KuFvP?v@uM*w~i3m+Dmw2U4bOh$+EafC@=NE7Udj%B0z4(m-
z%ZjlF0+!CO<xuSL``^cIcpK<YUl%lNdgkQN6@A(l!&4@AiUAhve+AsQXf6~`CPt1F
z0~kRHfQC%WIy>Tjk1E=M-G4X0w6K+3u6|ouFF!{WE7#%7$G3>-ORSvbVN{o`KQ3k1
zoOw1uNRDwyO+psGoomTt3CSF)tnn{fZl5$kd=0LEq>StmAbs9`=QsAxSL>>KW>@A>
zEm16?>Viy<_rHi@AO}#PsJ%~Ah`h_k9n^qnm#YbTFD@ReT&G#2K$vo|hkkiBn{V#V
z=DI}#tN3LZzNm%)i~DO+=Nu$h8Djuo{RA|gN}K~&OyuwUGoY9uV_TxBfob^E`{m08
z`|3l91yhLGyvyv0`ImKX@z}euB0&<YR#-Y2C@oTgsFb6eMd)03xTC%{tee~5O$orL
z-cGwoSvH*)Tm&;+)ss%`lv2b|UYQKiPV><6#JXt{siFa`J^)%*L&{LBd~Cx)cM=hD
zpDd;z*~jCZ;zM&X;2EM*+j6P%JJXIR{U8}>IG!h-dal}nS3UkPdvnn#%m|6jQ-p{;
ziaIPr3r(AZ0<I3v@nAD+LYh^;s|qmXsx&pHRFD_MnzVLm&ju}CRUq!vpfKf6lp{{#
zw}$rcQEUf4#_7DT{RC`oZth}@IcGvAb7`yV6R~JI>wa#lVN=m+)eVnbW+)p{U8|^;
zt~;+-$NOwz?A?xB(pzEZRAGaK@H2d!Mp}fG3EHkZLxg(Uk8jr^w?%q74Fgz5`%S}d
z&+I?{vIJZH%=4jPDL>`D%3Zw<5QRy>&e(bJ&&6iFyZWLv*X_z_AgrDEW%We`cIgr3
zu)^<=p_EddC=jS2G!b%cBRNRd3D4C%bX||tVEr*0>{j6xVv0H}gfV%A?T~ivlI2~8
zc86AGH9r#HlKB&m6JNA3SPIx$2G5yF+VRn~5uU0ywt+*%`7^Ot1}98Ysyd-;FjK~^
zd(@&+CVOib`)|#3OlO~c!ZC@|7&c^zr4r88LfuJowrdpK(u@$hVUb;}%3)Ju`7Vkd
zJ8=yTJN+{gqjJ<X1Hij~tEL=&5s4Bap-laYxb6#{%BQE!SlP3yMe)aK&B#~V0*n3X
zx(1X8JJa97My+zXVE+6VIUXsh$KF+edhE_IgjEUJ(**-{`^y$-71EUmc@X5o@iL{*
zytj6Z^m>Lef*@Xd@#WSgHFG9aFh+;EYk02oN{_y)8I2e$5<*2s)M*4G1Tw)G29I@h
zz_4uV2~T7gQbYXS4-U}2ghUG3`wi+aKnlkf1W9Z<^8&mVti&91n{tkOlHiS2{O#9D
zBryPwFTF9t705epWKBxryg;+?gHgKE37r`dNEA<}zEo^KcKMoXG=9~P+R41Ha+*HZ
z^S>|~N2C8Tcv-DT30H3N%=Fora?%?ss<^AW=lI<@LI!nlG0Z8WrE#HZc~YY+o${Uf
zr+d9S@_C;}&MoCW{5F%$O-LVY@_h<AWpRDC`(yorQF&VU$rtEuSJL+XZoIX))TqGB
zDYM8R6o5&sNymNmA5Deo0c011P*9l#Yx-b1?X)UwgPYH#o$oU5{VOB@`4N&T+1e@#
z-*^o575D`Rgb}6mt1Q>EBhV4>*81+6_&d)H`Ns}4@(le5cS0fTOeKltmkH_fSn#$F
z><0oQa5@CE_Za?``9L4yH7#Y9+ui5>a=+gK0>H;5J8A`^N5+%5>gA(KJ$al9aQsOs
zc+|xZ$`kaeMiph8LuOuXl3A#@G*WtR6KXw{M5qE@MfCl*;<0c}=|&!u@85WiJvnF0
zm-skY32J~}a2pv+{x6lUeT}H>ClR8+)%ve4A`W>pc*V)m9$SN<?)n4|IxFpctBYJ5
zA7HjvbFzv(3axHsJWvg{y{Y~z1BRBCMAK>(%bs@whMl%}SME4s?=`I?VT;5y+m?~X
zu4IYW<AQhl?34-8a$3c%;!pz=OI3YbP08nGWA|8p<VEoUOS5j!hsFj5AXcOLrBxgb
zk+)M6jAu_ur34=RD6Q-LsKW8?{qR|32AE@M)f#0!<ybFXFKuFo&Tq>2Tsc!!mZwNy
z;?6ZUq2Bh{E5$?UGgLN|iniLWu~e4V&B{S|_6)34%M&+D$w+-KnTFkqnp#Wz>upb@
zs9;`C#0hq`I`!@<+e)dOH5RJI{yM4VQvJ5jj*UBPOh!vFz%RRhu)e~w;sv4L2{1jj
z`7La7#3Ji?44%`3|NYkm8Pc~IlUqeT1g0PY@$!ty#7N6a%h1d|Q<0J}|Hk9t(<FUU
zoBF4)F6>GatP+Wi_x57h))lz=%Bnl-;)E#wdUHt8tCcurD=fgis(X8Nd;Gj|i&$f$
zfp?aACz`Ph9UJwsFt${5Bm#?r0<@KSiMQUl%%~M8V694G6_v}v<b{0?FFT2D8uUeo
zC2Atm{&rYucsF0jiHt;M3f@t4OB5p4kj=wf1((Y*eNtwIm}JUT+q0Ur$eL5-S8YE~
z&v^a4l;2-a!w6$pC@;RnVb7nmX57?1ffRT^9vRaR`J*Q4$gwp3i>IwC0SvJVCQSPR
z#|T?$cCy8&lwtf1!DsTQf@l)cIi8D&N6qX}AGLys%}<J`FC}AY8#OnZn16Ky$D921
zSA9R<3CIe=L_GJ2DIyrmH8;BXE+>L$4HuRmYGZ7abJZERbdkcGi{RI9NF^D059Ram
zgjq{2qnakMHz(x*)<0^7mw>%F7gUAnB#nl*apVab@fSd%2C-xRE{qJV1@$jXH`+P~
zhY@6}Sas>jG@k-w86h{*KijRWm}Tn88i7w_WXYYTD}P&#9H?DFkgX6ol2_sjt10Wm
zucX7+8W<SrmNw31D%lR4!<z1XR-$hM(dNs@UT=P2D$bGh?t;V_0C>sfOxP<=jR2uc
z(y#adWRiCb++<8BFT7;yyEz+p`Pc&66njl|Gjx_Giz5|7AA6U27&(#U&pboR?cS}^
z_N7T^m23(+!_H2FdDi`~o?5+$jdv~Ai)Glo=+AT^-P7fN7dVerzqTAY9~C@E9>;wr
zbA6<Gl4rs;+Ap)90h@W+86ep{_{`3~%vuzRkXTsv4lhxx$``>{!C7r~JU&y=MQWqF
zi~DHk)}`P=#>P%f$n}JsVd>9wSEnOM<r%DoDQqmuH?13VW(&nPAwjW#;xIht&v}-c
zg2#+z<j!bpy5oX-ZVtKM7ru>zAoE=w1#s81t^R+aNlI%6CEO1*{y(VI6k1{$7+|K9
zxMUtmfYu*7oi;e1=bD-QD-jp5IOH;7#@J2Pb?|mU#1g@dU;<pT+S{dD%&At%Zyy8~
zvtbrIGd_HU$TuSVEgI=UN_^jwV&e&-pm=Sl!9^$-&3Y6n<8u+zzbjCk$zQ=Di(le_
zMaa@g`qnQ^`pNK&gn1^lPWggmsVcakg(@K9076-(+|2h1FzEypY)NE@G{z=H^&<bh
zZrp{+CVanso*}_m?9pMNlHcCcA(^8o1<1@r@+OMaq-Gbi>W<RFN9FulLQV8)BEgcO
zgsJ&zdRW?H;e2RL1;(<B{~EQtVe7TdtXLI6l2TQ6jQOE6_Iy%QL6&Z3Bi(eEN%hIl
z0Y(5bf0{=uuG(;`e%Ntu0a|~$XE%+Z_F;{1?Yl>No%q4y@V(M-8KaVaZvJ#IDtblW
zaox=FO5O~)q@hnrw1m{%%o?IqHIW4zPmj=$7JqE}1n=gL1HhGyM(yLc-l>NcZ|>P!
zoH@+<2u_0;QW^eKIwf8af&(@H&>d_|0D>f*02hK%`7pu1IqE9vKorVUuN%+;|2B>-
zOLg91K@WZVnu4hJ#*JRN(=}yEle~X3mH(u2fvD<0eAH>F<G#-)`p}E^^V6%tB@$5E
z=dNY}t<UScTOu#uH8W1h6y4*PChsXn36|kNo2$4C?jQ!Hzv)mGPkv?^ILTZT0B$e~
zs__<>%6YNp>j%X3NsS~5zL`!vK@!voaXa0Sf(c46=FbZ%<{PB+bKh++w=$<p6V4}M
zM@D+Q(>|S9d9vg{q8M})7|a-SVQUOzrB^_7;cN2PO?GdAcm1U6Z-GZ7z#s<)c6V=%
zJUv}m)9?`Nbi44k<huXK!z1L40B)`X)KMhAY(RqO%@QV7%|;wAWjeFGv$u3|<lZWj
zwipr+j!Oh&dvGau)i6c1`TIVs{()iGpXx9$FqG^Ivxf_CUe%4a%o;6bj>6=h1L_w_
zrU{~Lx28W%?T%Bvp$8D=&k+*N={d4s0eHUnzP!b^yO#&nuCRZ_f|D970J3K<3zTCH
zdPY@_%$=|rU*<0sZ?@c>(A!PY%6W{nHb88`AwB^2R>Q=s#9VA`gc7<R<nWzhe@rNC
z`ARTy4%4Z(C)AeYhx;ut-O?4l)2$rw>GBZCnx7a8NV}Ah*n(FlGMCfuK`a_3k_h9>
zKw@)YG#e8UaMH%VJHacK01%yuAHE2lF5YwDG5uo<;0xLqbR3o5j-For%9t>99n{w(
zLU`StSw;3HUVEyG$O$geig6P!E%WZ1cz*tFX;f#e-|4H0j@nx3y7rQL*_=uAEEKaD
zoT3SpZ^#>T-8Z6i<eqHFTDzr|PDLxiGuz0vKo`d-Q(Y`3+QE<waI_bMN7S5VmX@l7
z_d5^+h7VS!0~B}Gm==Z$8JUllksXyk7A1<kNpMKHo=~7g2Q`U^dPMcF7u##dFFg~D
zH?<+V0<(Q>Qk-rv53lX^Ch%Ln*3YBc^E~b8#<s_1n96!ugTt!cUlmN%L0i9g@%3(u
z2Y?H-B972pk$Al<0Qf;JD7uNEta&;TKI&*Dwp8z>M6py#YS8hV<$52&SYX6?!CJ(2
z=1S5cNp?wmJ|yrdq+W{of<J*9iPuO(uw7*MN-9jm+<gBKubQ|vL2uE5O^`!Kwx-F@
z4)H!tedTR`X*XR!QPb4vPrTeJZrqUD-pRM_dPZdE9NCY-0Dum`{qFFA{>I}EabQ=#
z66WC;GoW|oOrhi0y-co-%+qM_%<=0d$UMMfnv^`-2xFxt%}5uj<v_4TJisRY?lDX?
zGc)>_K@md`9u2^^kD$~71sgbxBcaTv^NGd670+xgLw<$V0I}Xu-ku)&<+R$YPx^vt
z;l*C`<{hT`0f4>?(3{o%xC5!OhA~vqmtF()O(41j_diA#4Yl49z`i4tRd{=RO4`TY
z5y6TDcSNH37^WIv3fHsEOA|?ZWknnw8(Fx1K#pgC*P|=G;P8T~UAxE~qttEy8BJqt
z4$yJ~XekCkHPSWty?Y7tl!;a0OI4CBrDI!JoxiHz0r+0sf()Q`e<KRtp=X-kBlRDy
z5;j<)?~`2z{F}3l(>z+%_MLun$32>ThOrdtaUn<L@<loKoy~@ZD*d3Y)l*_;lL#0N
zhJ*2rtK{p1pzV~S*W6Fv<|Op+*jW!?$s<wS3P|{jx1^E3whX#h<#gL@?Rczs#LLv4
zuX0)r0ywE_A$ka1>|P}+l9v-55Q38KjMN}*3UuBIW4*G%3rzbQv?0U+;@(G{^lTIo
z-eYC+Uy4q7pOGB($UW9JLKaHQ`O&2y(`q(UyT6o<oM;!l=xx=plDSF<8AGl6q(<d`
zg-k;vw?oNn!X?viNnn^SfMdH5m}kBNpW$%=0T}Rv((@OSLsSVmo}I>ce%1I;Qs*TN
z)-b_u#ljs#BS8I5ocwDU7kp6UL4cQ~=km6E)rK6Ad}Sp~8z>xc9+jAE6Q5)I<{_ki
z^WY_5CLX7#wff^A;GX;N=J(WC?Tx;rU#UxyA4gv~{chW<BOox){QhUx)B!Wy(~+b`
z0@$AAu)*lQMu!1~)M_o+Rw<`DL>E@nk>=A_AG~pFbAMJUy?<^Ic<;I9pE6a?`R0Y~
z$(rWh>Wnd)R~WeF$f^E%Fu1$kP03h2eUBKMu!{6a|2H+wKeEk`d(l@FGCBP<CHM_+
z!{PG{(tRddKP28-Q~5J^eJ)&gI=pnOry+X1`t|~61w|r7leD{s!ymJKlNHOi6NZxx
zhwAr*w0aF>E-UsWw9X0mB(A&6gAK4IS|Gl#U?f@6t~~pu3z_|EqieWi_^ay-gP(H!
zFH;H4tsWGO?e%|auoNL@BGeQ?TdF~THDv$S^z^nXH-|>^lt+a|g<TfcRUaXMh+T?^
zZS0<50)mbsBG=CN&170<PkA$Mi)eWtVwG{>uz1}_UBCyqu%qqhX6>LunI1UB_q>pw
zQl)#M>4Z;!fL?oH%^IzNEhC5fIDJEz(`8C=mJCoB!?3qJa(tE3FJs&#-2Dkyj``DW
z07r>Z3rr|B#Ub!*+kNSqQop*dW#L==F4F$`+J!`>=Y%abX;JvoqAo--pQ2mm6b{0E
z7hF{dpFWv?X#SO;*pm(Vk6LRIN=muzkwaqhh;zm1=?6o3@!dF?ljz+gN&OQ7k~m&^
ztP-m-?Gf7~Hw#a{(>eE%d=m#)cit{?`?P567eshT=sMEGmjxYDv~p3tbMc=pQJdYE
zSf|N`p*GA*?i6*};<?t#ln(@N^8Lq8Iv)vGYe`Gg8E$&U5;d8D!*W9W^5CaN<vaep
z!b#nKmY{@W36t;WO-+rOS<eoy#I+O?-e8T>96i2ZiG%h}iPJ=&o!1V48^UGA&fG_N
zqeQ(GBa0oeM$7q7<R15o5E*Hn70ydHik@k@9x-?yPXDnb@r2Y-nY(&CcQw-aCng*b
zri13Ex=7SC$lF9jo9w)D+6;&!7Ld22`dhLT#?`u!%9Q)ME|zv7?&S1-o+gZvB(`81
z{rQWLFE)cQ08dYr{UZ;c65obzgK1(7ss7pWSDG%o56=6{lKvd~*?du4(QiQCkeM@z
zaoG}oxVktNqyUiuNU!f<mP&dAUT^IP-VV6%5)JR`Y+VfB!#ft8((PI;>1hRi81!Fr
zDc4tfBr>d0z>rcvkkZ+r5yxmVcS=)zRukm}OM);fvI!sM^dCC_OnehPv+Q~}DGe%3
ze5GQyNPYg3IH{F~gJI{oyDPd{jxqU94x}+bP68sTgydR)6Vh_X?<iICTE(V&mMo}o
z2)$0b!$?8p<mO^?gC>v~sG+9Lr7M*}v#kd4t3ZX`An*!nT=IAi7&M@M_X6RY2C!X;
zfLgBJymRj`2WB)tWW%GQ{0Y^;5o}{9IV{7a$5CcpnB_$;Relt!^EcURzgUes0$nW9
zv_G8|V^9bz#sIw9*0JSvADKOD-x*<VSbJq)7yJX`-gy9cB`6H5d>%tD2<voIC-znV
zt?U|tP^l!ka2>JcuR|VVB1=0_^kB_d`|CAVd?d9TQkXnIvvW*kIRO*r6^!;Z8*~NQ
zkx^c&J^xe$(WkrKsfE-Ec)A%)#JF~B({d~^u!8W=xo>5NX!y9wZm(5rDu+$s=VNO?
zwh;7Z5UAaScc_}`DUbz&_((WtgJLDi?lV<aNK0C!Z~tdznkY?I4+M@<U^w|dLnF*u
zCOoVs5-S-%?Wb!{4$yv$$f!X_#l}bYn7n`?9^{%PNEV`d>cZEuND1TbR>$2fR0Q^&
zT*VBa!Y+*RUB#z8!A4|c`GsTGo^i%UHyyW@maV79RHY~PI)x{Gee2zUFGP293i+q`
zWZbuEFfIHbxV7(iGk*{gmq`}AJdHTBpgX*5v?dO~p5_i&I8g^v2vmi_U8SP-aKn+&
zqitHPx~xJBW&E9nD6qT&r8t@oC2Gx$!B$y<X72g1<{BQe_Vwt`+j7^Vv>kr$zrx$X
zXj(q<8u^%O?D7z#0j*L`y;OrxhFWq9@APcF91KCdud9^}u?lJ1oV6>eb(R=_89<J@
zs}l!kL=L+y?>x~66s8VCP-gxz?Y&zZPla!fSm?Uz0f$Jll`T3+Wlg-}Hqs1X<5g5*
zHF3|@((z{>e_L&`@hq`@Ot5z6ak~R(nwhe(2TSk^85yeTP&pkFmA5IP2i!XJHYoWg
z(RB>F9VG2@%F!%qXSvZh{c&-*T2=avxJZNjBNeKkTlN97t$tJgKYfff#lRj81%!z;
zWp{;#GG)p62jI-x;^;!93k*PLk^Vi+7DP^M1zQpw#w-Jz6vNp6j3wY$W_+wd)&%-E
z2$L?a!(>z5Zn5*k#&>#6s#OZ}ccDk7F0GvN?39@}@!VC=Y1|LDlqj7kvODukXh(R0
zWtvWZx2^qscErFYlxvyt;vYgpMjLmj0e>}Y^m4fvNz|J$ClziE|14m(kqGTB!IeFF
zljX$g&y-|Eq?S6vJ9D&0xBdL{_8(?9Mn1QyyR+gQ`6^>N2j4qo8sHR(k^y?dFh=S|
z*8x&;#I%LzKRIc9HRv)RnxSNAXQFnMA1Cpr!(#pe8BWnXfeNOREW8*7lEK_@ZXU-*
zz{f@v5FvYU^=fq+KjK82RywGYkeEtd6Pa0f7ysh?6YtvSsoqGI2dWwz{W$FWAqv_H
zCDL)5tb<l~MK_=d^S~Lqv(ZcL%LIw~rz(eDg|ZMK4tej^eKo<Ofi$!`)35+^ebx4r
zqmuFgx@Z~Hd0V~7GS^7IvUWpRagI#<Te~Y2#BS^8rjJK3j{%5|_zWnnUK+qzBqw)y
z;IPX#OHIx@v()p<e)XYSj>r#LfhALB<A}o52WX=%MgM4Q30bR-F7t|{@?M#LP0Da9
zWO?Ebn({!jikQ>KC13`CU_z5R9qippx~a;W;JHxqd5#d$MwTU>Lb#|kb97O=>kJml
z@tc7#wWF1NU;_Zbhm!h><g_y~;+7^SE=uVj)o`6DhmDVkL$u~X592~zYN6N&nHM{5
z&`QpQ)*Fr-bch_N`5+$Rq2XIRP)d!Zl)Y0R(i%RBq0o)n#G<qIs*bH3F-8?Qzm)u2
zGZ4{S&Hk32Y)&3qiuEZAdDu{hZ7c;$+OysCUE)&op8+5oh!HV{!zP`Y_~rlpnK^fZ
z=Sn8sYZ`T;b`(PCB8cSOZ>xpZK56`GE4tlS*;%y}iDwq-D1R|wV8Vq@1Fa$XzVQ6}
zKx6$Lz^U+{1k816)l+oi5m9GPqg+9dca}2Nf_ie%VsBaH3^Hl|FEmkpn2OmJ<J6dt
zn%qL<loWu-ktqMO8Sbzx9w~~@OL)U=S$_7nTzl-mPxMW&>_p%Oj)<SlEQtJU#unMt
z%sJuYQ~wdfjB9$Iz8%UPEVR+!4mot-2WZlyc?*DmpF1f!eA~utxTX8PCq!b3Ib_Rm
zQ~J7aRL!Cf8c4I`#;fA&L7I1CLCP<uctm27!I+<JWFxK;7oC*Era|5!r;^bQ?S+wn
zSPX=rEd#qoA0L22Pm3N`-f)bILTX8HE>oa6AODG~KXZA8?n&*G`a`~@`SJ<bfBk9p
z-=7_XmHEF8Ltyq4;0-FulmkyHB$d!hES%9v(Cq+R5E57TeL@zlgXjBl40<{0M7n5l
z5X!Kjv$YN@0}5SHiYqscm<|NDwO<T@WHp2P4TQWWH*$D>Q_u>eS>(4b)V^(J+=up?
zJ1LT0RCs_?hqU0Q5IzBwmKOpTI7sYSZ68%iKZHL7zCIrw%V_=r8>@u*tkN#FQd-AR
zV3*XFgfJ?gBh}{`gdZBeOm0R6fk1&T=u&aidiCRqJX3B6JN<7Ugt33wO>pC2E!8-I
z^TY-poboFn$i}FXsV>1@RmMwoR0bR^(%(u2$BqMnb44lOs&lK*xAoJx9$0W_W{4?F
z3<{H#OqCN=N+`2$A(|S9<ale15{@R&F*Dn&p-d_tveKYV(vU*Zjyuv25Du!t2~fnZ
zCAe2(JhCRd<Iyi<8VaSLZ!Y3aG>E?1?5_CvuDY+ZAhqY_XehFE-9CEF<FUI*4hz9w
z)~^A39?vuJbmZs8A#AS`uB(qIqsR4SMvxP(Q<^-tyhPlzv=#Y{b=wQpC=={$G+M$&
zVo^29w3Afn@FRgXRPvn(lo;Ar$ql|_w-kt>+bXW|@hWsCi*g5Nb#=E{CFt7I2S9!t
z6pe)nV#O{A)oUo#F1kH^*|q7A9&-VyN_zlqe{cT2Bb93D6v?C37SLnVN4Ehe<_4Wu
zrODE|q>K2+D|oNv+pXI5^;w7uZ(nC69(J7syk&(5F>MICv61Axb~-M36dUQ8mMgTP
zK-Rt%?=gDmvdH++e{`--X1h_a%-&>MD5LH<YTE{oZzmF+c^T5qI59Se6r-@fu3Lbs
z4@*kh6j&?@ioqaM$31c{2|>|=R8AJuwnM*AA>jq|4uoJh++wacn>2gua0(tu)2R!8
zsA%d81`kp$>!U=Lr(i>?vN)XR%c!El80!2gqDUy^xI^r{3ZT6%$(5gItlB+-xL)KW
z{0-<pB##keHh$W>s{uWRX#u{^U=o0G=f`9zrKqB)GVET{26Y(7HuY`MhEo7s->ya!
z<zf_Lsd@m2@ShdX?7=C@7!pjeQ=rd&#f2;2#dmWDY5*^Z8U%@Al0OyvO#k%+=8DW>
z!>IIcrkCil#3t{EM;Em1JGSncQjN!u6~2f~k-Bo1@e=NiiN6H{W@F`~a2|jMd6Ee=
z#Bja`MdNbSe2FQ35WI7qTxpEV*%`)M4f280YWs7S>e#p~G%&3sVerR%wi^;xb7$~E
z1mz5A#;PLaK_cm)_SOD@?~>*EEAWadiJr8n?8M~ZpJ-St`Y3rar*hj_F#0DptS-Z4
zZ{0h~wKwX^mT?P<5H``?J0U=%=y>}j<RQL1`!8;bhyndg!pm1)=D$&M9V<9{k~Eqw
zlq*e`M*|nd#)MvqZGETf`0NSi>onz)YL?m6GMXOiA|D5@!YoG=lc|>;%WMg*9(6<M
zjka}qW}dB7nuEquWhZO{X|@|0YxyqQym#LHl-X=2{HqK%xpx<DHc<fMR%~eQa|o#N
zO4dnN8DP#<Pb#1g{c+%zlGOYp(K-T?@OGCkBVWT`C1IMlIUCuW{AuoQQ6!1o7x`l|
zMmEOz1p{q(J>R2+1`k`*x1s43PDppk1-5oY?)y5%-DnC3r)>w<If0Zb3y!vSM}(nX
zfm)}?PKZ(2Vqy&jnri^X4SV=GMcuSO))kEOKyy~awGNo)Yq({WV?wwl0%k~|*$7-j
zdYcidWw;|kxS@ziA?$7%r&x&he_WyDnR-mR#_;+1$Ly8^_Z4|BUy;oRm%NZ%HA!lH
zQY8;B`ZCwClra5GI-V(_1Aw<w244h|M@fHdYXYmym7ccJeINke^a7c(!xwr_{MJ^@
zRBl;=_M3~eL&tc~XXXf51rARaODES=Nfg3tV^`qzT&PO;_rRn-HgzfEDlU!q_g2P0
zCgYA)#Wj`I>4xH4s&%QRy2Sif%`@=+){!}`?roZI5yKe!2{Nynf&=;WI?CxRx*#}|
z#spE%*@5I)gg4+%Znf*9K^qLhkSm7^2xEK%mF_dD!nK#WY2~#Irg!4PY^C0b;;rj0
z@h6B!7R2E?9u?dW+<Q>)opjHPzFw&wp$w(sX}^o@z~evPyu2roH*jSo>h_2C<KT0M
zPhpW;!%yJ>w<10B+UX07P>}e@Hpw2utcAC^%h#U^g%SlbnY&?Z>enAjw(DX$*c4ri
zY&7&f=}QHVjPBH*FoME)0I*{WnDzf%L-XJ{P(fHZlZ5EWQxsk4z*6S@e>iLXPVLR#
zp5G(l@x3{&?6OLjMTOc4OCBV#u1U=k!rN?)Ny_ATioXJ7>7euFkT+7_CWr6-wLai7
z{O-*MIbHmKl}f|&MPRL&m1KdNVy$t4J}Oe4Ji}#aZ;&>I8#@UT2KvlA@RJRS%rb$<
z;gpFtTjmN7b=kKtMCz+7$Ik{xzj@PHw%pQ6d(m_mv_OmN@-Wv{5*I7IGj}IYwL6-W
zOSP4t9jaj%C$0{zZVLT@5pbu=Cd-$ZO5*2l=xBMK%)#kqC#T+pmHlD!%z+jNo#c`;
z3SkzM^n9`Stb6YokNFhEzR16t9X>3N42t^}rA8ueig24<{^JwOpH2mgS%eYkS9p=i
zozGFwC8(m+wnEiP&q_wxPEeVeNxME658$nilX9K+(0*Bi4cR4Tq3ocRR36Ayv*YY9
zdF(Vi3d7^}RuU#S^~A}J$ayYgGw;X(g861G@y7zQ37N)q%t(5&_3ydVi@Y4A#)=)U
z>KaEJg+!ZjlPisvD_a2T{G4g~sAeZE4a=%jh{Ix22(S&)T&O=_h&pICeujKZu?GX8
z*=j9C3z9@N7=~DmXBM2yTB5DhY0Tvdvl1H>7(Ey*znIKD2fg|F<7eh_e1?1e0x)zm
zodB!O=>SC-lCc&S^*SZ@4#wW1yn}$=^H64+X7zsRCVS#EIZpulwLZ8m%+`~mh&Y|M
z4riP9pG9#Qvc?6D`a*eNh_g_3pJut)debrFhr#W$vhRrxGm?E%b<hoM*tV(e#7H)L
zX*SUuMQMfb7(Vb4ADX1*utRB8M`oq6mcqf)30{xr6*WImH&>JAEIQ$`)Tl~kLrKLA
z`ut0fX!I;_j7k6l>a1<2%@N<RQy^^QVtPzVWhUp4yKj+kK2Srfjb4#)m%aIK_V>i{
z<<dQ|<PSUckZANf511IQxmPrYyc?B{9zFe0%7AA#SXh+0%s_in{mLx1o11#hcFLS-
zN_Xd1r~a^<5?&mrlTfd|?_%D_Ch#vJu4`CwaBuTJ_ELb4$scM{_vhW?wuXl+!MB{s
z?MGk-#USHnu67hmb!ATXeAhr=Wh7-_sQVE+3Vs3*-(m~KdhfDD0vMRh0OK29)Lxjt
zP8z_SE2rH%T>}eNL^W8nE(W>hdS5@GylzyOgK6$hzVLOiws%i=lI=X()^gd;RR8?Z
zdgH6GS`%REaGVY-$buxv0XWcis9ZxYaG6o78J>z_=Cbd1g=oxHx0&-oivn6};>k}S
z@9wOg3YyeL`jm_DOb0wsm2@6F+w95<J`dv`he8T=0|%@+%;zC7|4MDzjUK?K(6!Hw
zc23B&@G)>A`f&ScA14_QABM%DNE2K|DvVrB!#QBKfm}T6rz@iwK`#!p>XQcH5Uz{x
zr;dRF2UJi)HfJQN&n>L)7dDFY!{je7zmoEwERU$0jcL%<;*)L*NSWwLD`oA-=rB1W
z=%IV!l1_wbZX_`;Meal0dvWcwyp8LVnN0rYg&WFI;n~iXzpwc7@H-uYa@D;$n#n2M
zGFkw2GgFV;%^*6?@!i)j(iba<_EOGX4Y3B;2r6d9%mJgq4hzRW5ub6X3@-0<1!-d_
zWv0(qM3bXO#6$YJ0#)GBZevY?r}BCrhtC%LQS1#N1{^%VBkUUCqwq#Oh~8KiikHS2
z6to<Li_F!aiD;~{A?hTLL>U;6<mY$;Ixv8DC#GfEt$rbpB=PD=cjD&_G>T#+Eglqz
zmX`IO6{&33tckkuuL(;>TjqUiM{UJFiK2Y2`&ygZtl32UV$^!pG8;7UOfNRG-I{2W
z9@9R+6?k#314J)W6G&T?k4q=WhP9^5i-eaei{{b^C#dX9U1{k=;-`7n_F~ei$N<Ht
z-o=@b<VsAR>v*&jx`}2;{}~V!gyRgG86Xb<()eAeF@U2UN}s@8vanVx@7+Q+$Ap<W
zTi3UtSyWKGR~e1U5kH*frw#6;^1-29fkYLiytDVpzBjOyz@i`s+ygq<AGFJp_+n#H
zw1!XWkQs5wwe+?Vn`IPtg2{KWN`S$7pw&vq_`C{xq`$&)>zyo7jZ=zESxD8_uIW8`
zdea4rbJ;9%lqoLRG#vS&dSW4N?}*AA%`A3j={k1|e~ze`%O&0tK7FQ>IfX+1g+Y7^
zuZ@k>Je%E<>Q)^)lo9HS1D!6VT(bEplg&q-xSrkaYr8m`{9=BTI(HTp{}1PJ0y}Xg
zw@r*&(YscmT(z9=W#Yc~Ii^WdTArVJ=q>WVkm2|das8stBV@slJ5{LQ9?W$#tN-v*
zir3kMGwhjyF5AAST?U}MpU3-9umGpX{sUtDloEfH1jJ8vs{oW~0?$#eHaHA*YAzI*
zEd{y+#$Gma1;(g!w{s3q0SM0&Z5SKf0%b+R{qR?pW=_>()3AX~>rj1k%_}A`$O=K8
zc@t1y><1rUH@b7b2QCnq@|d^OT_|o9X;aoN{Mj@6SgMa27@|lmIq{c@r(OHKdT#R0
zu`Q{oi3-{={d)Mx6D7=^j;5IUTDK_4F&9~k`gPGzBA?l!Ov9n70h8x_srn5ykx<o>
zyQw?;&L*<@WF@`$=16Hdb3CeRFlg7GG*rWn5IwpEs0`na<2FyQZ%Drqpbz>8{A*Wv
z<!c@p5C1Oh+AUWPl_IW<@HAT`vGzaYFpXkI+tcDM%j=C0TKBRN?{5)+$H0L=U?BZK
zi#tW4BH;rMcod>w0pFa}cp~<AA${C0za37X&t2L}1GDF+UI*b%*nm+^|MCk;&KzL(
z_WLMmLMDo1T1V%qds-c5k_~b+m|MGnrt|jN6c5Yd@V#W5Cd;79>x-PtvlHm~rIL(l
zbDbruTNXVooJ97v;HK*uh-xPLS&6*m2`6+(EPSy3BV5B`2gDa$I1a#aVD)Z5bGu@b
ztY{R`Oj9Xmw=BeG?26I|oYnYLI^AMaa@-US@22E};Y(iwcl!+=*i6FpVf=jP9hd@d
zFmHo~Yp`x0{qW_hDHDW!PqLT((og4`pxX}MmwqZ;gl_xrkUoqn@D@nDn?kGIEzIrA
z)2Qe)r+hatI^Cm6JD^U+Kk)nX_hah#ObF%u3aS4*s%gGO-hhV>RUgi_hWQ<iKLl)9
zH3t9JZQ13bLrv*BCZGZ=YE0U$Gyc!*wkR5wNSpgBwt38Nll_4Gg0+(GvbO8nV4;R?
z)n=~{ra2HdBmVV1j9^kS3?pxFWOBgaepg?SlM9iPE_xcoUHBFa6^0H9lCTQc;e)Ds
zi>8P!1`@wCk%<>~8{tD{od*=O1}Gai8N~V;7eprVh;49Rjj9TcJ92$1i0gCong6`D
za)5{VS@Q>eK62TzUlKJAPjhP13McS6|JC~1bh;_wnv;dL4Y2J8!;Uqb5ndPaMA&;q
zxb1x1sHhr7?Ma|79K6PumFiBZh@@fw3>f%{Rg7A}J*u%?qt&Y#%7$Hs-8IOOxC?hG
zEO;t6!OQ2T4dn;0CvAzQum-m-7<iefu55FPP+8`3qCCpLVkXN(*2Z~b*;qApsyDa5
zEf_QllOWk`^QM;!tu%?1VZPeB2+mzBOYAq2R+-K2)Z3U9IoT$QTvU$#i!~kuxQ8|4
zj19W}C^Llyq7tMYEzVI?4w$MT*A~o~#sPK6oAy@ZjcVHwP*qDwvb84SdT~-@5!wp6
z6Qp-@aXFX%T6QT`6fsyJ?=38?e|#+dflA>wq*|;#?HmfP{gJ|-tPQjSElV8BnACqr
z%<mDQG9krjXDiaG@K>IHh#n&VoE$J`dB=7MQ;_t%>h4=`*Nqk86#s^n<PaSQHouRV
zor}D56jfP^c1Xb<BbA6esvh_kP`7xBI6k1^)DPr>?mjfDQ+I|FqSN)Zpp~<7(9~#7
zU1EJ#C74rrp!e%*vDEwAr82v3#ODlUwj{5k94DNR_s=eel%noTFW#9S;262SA_ha#
zS(;mprY8qu1&VEb4N$lcyL&L~qIW&Nz@=)*T@fY|RL_(EtspHliW`pWTNTMGFC!Hz
zYGa@(#xj`T$g4H+Pfj*++bLt4lN7g{pq)Y50fI&Oy)4n$r#`{u{o-`MJkOoL!z80S
z-;@1Zb1AVpyrtO@^8@#qE?n;_ZqDg3UzA_1{Le6=CvHb6Ot4&*NTn6wNHb3X1iGCo
zifn0F&Lls3{74FD4Amb%hZ}?UC|@{nr$fx5rcVfd#k_k4{`s7|w5HU7DZje>Up49(
z-W8snpn#rrVWRIgb>BJ64+=U|t(}EN4(mKj_bDm2dy-!ZAIC`_Rc#wuQ4)w|3uaR>
zEzyMBRaP;C^qisdb3_lW{x;?=j|Mt5O}>1(T!}Ivh<p$!-!&50_H#NbmK3Ku3on;m
z=PT`!9owG{_mCQu?YsY11J!kLqku58q}2IhP^8q}K!UZ%#Zoaq!C7ma86cU&e?wkj
zG5>F5J0*_;12v^4jfxwPYmmEX135}3v>lKckj6DhbpIWqsI4lcy0z&uny1r@9#8z*
zI_mktbh6trd1Z9(&B)4mh=mJJunidtMVFInGIO^Bl*q>D{ek=CIg4Gz%i|$wWpv<h
zw>go~q5#oZ36utm1-!z~6;MBt%Zs>dm*q3|eoGf~_<5(1hZKv&q>R*lgI;Fy{q^Eg
z@L#}6&#?Np)puQre>#;2Z^sfXMR6dnqi}e--%;W;BtiwMR0LgX$*3Vh!!#<Y*y@=C
zavK&>9EGJ2Fx^4mvv{%t8;(i9wfgzMV{!%+BVe7PqJSVnQEiozI_8CSLV3#!SG1x&
zO=;DPVEXr8nNeRlZW`2O(;#<GZXO*sWwqP^{jnb%w{elaz-ceXw`+5HMB}H~>-|ET
zbHBqKK}#rfOX9zX*omPI$FwtFYl*%cSe|_TZlTx&-3h9)Mp@xtZ1vrey5~0@-!`{w
z2iW;>L3ncag%AvkSRBSG+|<!{FU=j^`+x5IhxMUFMr4`@SJ1MXr-2g@6tQ{2cFV1v
z!kD|q)R=aH&E77#nrYv?5n0)I`b7R*a1qd$0ks!CaVZZ&3XQta#M(eJ5<l8ye4G`n
z1yl4;(JBygsIrk(c;j-w8G4sip`xb^z?Onf)IUcriX->0|Bjyv^;X<1pEKW!?f!wv
zOQM82{l0Gdi)~6j0e6vzcJc80X$Tn{8)MYvH;z2(m3|-5>#hm&Tk1*58}Iol*3D_w
zbRB2YJ+<tRGMFXoG%HN~ScgHr@src#U?VeltXJa?BdANiRmhHielhQ+l)b_Y5DxFG
zzQo(?P3<jN&*W<;PGub_>J38?JrJ`SSg5#9VOre}8QD*dGQaxI1(YBPTETQr!bH-D
zEMzB<NH4-mrpeem4tH);y1v}XW)rVRFHr?94#Y`Y-sX9@bfAN{7wc&6lgYISYz+*1
zfX#tRAS&<HecZaS9{gRIHKFGaAi~!-0UBRwEaRM~kWbm*-$NRL4K^f*a1r1c9z}<N
zwT~`|<-x)sM&Y>odjp+N51WkTZ4Pz^*4IvM7BFe6?J59Lk2!J7Xj(uiVX0TkSx&Rs
zGMcRO*e;BLRkRqGJb=D?Xh_e3*z{Ojv4oe0T61VjKTvq~U{k305^X>RP#Guu`}ZN?
z_aB)@^4_8TL8<9A;3*k(^lmV_P$91GQSFdu!}=Yo+4wT|j$L8G=NCK|cRh<hF&8no
zIVke-$x-kos}YmccD>@bUTR6{>(Z#Wqr3TnL6f~8ol^B{5yf<l+obHv>R$qZi>Mz*
zyA;i+Q1;%&hnIQCeKaw)fK4vyZ_?BAO3WvfG<o*w{<W6{w0lN>tV$2n{4cczOFVGw
z<405=!zDSPQJ?Woe8~`OXLrH{?(i)3nqUo!c<yo_3Hr@DxIZm#z^AC4y}`#AYec%<
ztGDrTDJ1Qh&&?f=@o)m)@pR%zh&$HDk6;1!Vsg<E=qpL!ZZeiqfO0+F_1{^siPKW!
z2<57T?55Ghm|frxWSS#qo~+*{n;_<DzLqf^u0-VU3}Kp#BrzPSc@eW^NCHb*&XIfD
zVUZ2*km;&CzR>-35S2|DV6UyoF$aZ@`^20kzdWhhak|@N4ah^ZEtKV5R8BPFoFZ6M
z@|M*UMvU396Qj0P0m^CHYyXT(u-!W`mHrv7NFztG`8nTWfm^Y#uwo|SM?35y?-?Yk
z^U-Edh&yvvipv<ew2Pg@D56!#jOJ6txCHYR8_ub%;q)Lr4GJLHj|QoHv~W9Ae%rJ@
z0$Dyv-Pfi{tOv<F`ZshI!G0l3iyHp=|87|m1Dl_xlqmou1Tbq#-7Y682qSCCtPT@p
z%0&(pYzkmb2om7x=C{Foq8+Oen>LTj{0)$FNRk{&wkd;5AcXYDD>bAuUi2zF_=!v$
zU_fBFM#$M>g;(PQsR({%0KI`5%OPx=ozVyWJImM!!0`pdD`R!*z6et_9ZlTXjchx+
zFYL5J-r((^`wbL`jZT{nVf_7qw91~lKTBPkyZ6RQ-UXnQ5e6NCY(sK?+<u)0{)3j;
zm3zVJego88BLD;*&JtLJaacZHFNrn;;e>hE-05K7cHi{|QbM*p%Uo#3VhR!B2cBz0
zmhZTDW@KD*f8r+{2qmAW;fLMx2C5DOd8MSR8M|+XAPnunSjBxk|3m5^3kmQHGfqG)
zq+KMl0DuH$ACi=BH`czX)-!=bo{`b4o~U|hqYn31a)hLD{sTkFiar&n9~yj%Yt!Ta
zSEEeTgsUyhP$rytNsFO8f9gD7pdwSS*n`nUh!GE#ieNaEzp#h^=dh|NLe@H?;j0+P
z^w-3FIKSKn34&t1Un|p=H=!TV+ec-fixN#P1RN;|&}-*667oc(dv|OiSN(NvH5^;X
zSnhCTp;LH{iqiG_y&Nr<?ltKx&p(~t(s?&9bT)dyCFF@Ikk2nS;hdG6DJ*X;uO>!-
z?~kDwpJ~Q=_-(<KtXbuAWirPdYDHWbwVW9Kv&k{Wa}iGj?u2YfU%7K|d8(iWy@rPs
zfFm^>bqnn(L(IJHvAEI*Q-&zcrE}xG+jG)bwPp1xb~~Bx;C)yFikAP{lIkt;veK-O
z+V$=Nu}gW+)I+M5=a*=t&i(HXH{9cW^bYsOJeLaBS;$o}0+SSCp&hqSj7vm@wP|`m
zA=421S@_+xS=zCi==rB7`!u|M6W4|aK&RnKxm#$;Dv|E-g;xKZ6>Z=sV`Eh^752n$
zS}7urj6zLAEF_J!n`!<cFS6B<ibOUXA~Rodke(DM!6OMw>QTkUtX;c(kGVq0{|Bi+
zR=+IWnSxTDVQ8_}o7oK*dYa?Qpddg(|M<3yTs7ldrF)S$sYnGyV+oka)%2}8;+Tye
z9aVo^2-OWbx0tDOMu!$xk*fDF>u{_sK06#nEP-WgsM<Vdhaa;ZFH?!JC0h`K)&vW&
z30uo=rh2s&K$hil4bQduR@%tmieWqTv=Pt=<OTR6cq3_K-0N@Pjklor!X^7rjtT#@
zi|9IbY8gF)=_RThy%I&`^BkTa&A2)W$60@O-m2$739zsu<L)G0XVdL0+-V-g_oJO*
zy5Ita#Cvp$Y<GzT-_J-$0<&~AWh6e}XH(15s$QGRBPS)HXiUbH^Jg7A`C8aQCNrRO
zedM2K3gO*4F>3{72DDx%k6EWigPv&|9)Zis`CGWmYOH$O;JFk%G|VS8RW0rO@_K){
zef>JX(`uKy4N#2qzzs-NXgd;OoQ7QU63UG^(BWtH=s7;LT_t2L1eN0fEvZCjRickj
zNs~aT^sFw<b6%oLUZQ<eQp`ZALUzSRbje3MDFj*&j>*@p_{d!Hk=aKjsThz>a_fqh
z%q1_Gs}Va-6faOIX_gSXHv9nLxZ;1xu73^_4^y&!*@enjFKqo&Munj*{-;t64q8SW
z8&zDR4O)B7vLkK4T)z}K_7~SzzML+8p0$3im47}0*A(a6Y=o=!xRV>!Z9l+MZ+<m>
zjX5c8goT-cS@5IQHiLhngNhs4Zc>2b_Fm~#VY<uUU3YA$NaLazBt>JRltX`l`7c#c
zf%CS}0;gKkzEi!-=*}qpc>>w8e<I1jFoy$VNdUj<3OOvJO)8L-Wm_tNi2yg-AoGM<
zo(?2Ygr?5han_%r*Dg(00xqw-R{E#e*XVRE>l0==o94<2Fs`SBvlzf)49?KN2>dFQ
zs{R7INcf_(V+I~>1f{nUE5(1fh`jEJl~YWye+aG897!`-FWbpMCy^s@Gw=_`3^>#6
z-0>S@9mCnH@BKl?Z&C{IwjT|8>S-)L3VMJ_J)4$BA~Fhrnf^NGfr~aQZ=EpI@;q5O
ziJ|)G*O<3%w%Ra>1-%lHn<QGE_)5+Eh#Mqw1+eqTobH$CI97TV890CAbw`4)nb#}U
zz|-BwrPr;5a$MT=-f&!c*_ZuMXs;!1I^eu=?0mr4k4@=NF&-?|<1Cq>#xy4FuI(LU
zQH7T$%!vFbA3oY~`>26E$8k+3a@}kR+yoAD4SK!zcu$)wL1q-8($gk-tzhMhV)JEW
zc{<{xV?h(uqfIqVl{kOzt2dQ++rc3t0LP$0aiO%5n$4bzKCef!?O6wnTfmr6yH;bK
z)%j?+@`vMg%`FIv<_^sp{u1(eJZ8rJm^rP0&08T{Mul~5x~5Zf>x&`XM%H9);gpNU
z?Juf&%kil=myc<lf6QFo>6ltEE^(?k_8jL?J@8L^6Bk%;4$glL2VhD&o=lBCbd5fA
zl|E-_uw%$-jb37);Qe&pGX$kVNi#_f&sXtTX3$*F+C_3zg3Kgfnj}pH^J~z**Ex~H
zyWl;;CarLGY!YgfbUy^&1396r+7aJ(66xhGgi$blC+$p?hwv(qcYW_W&;=L}UfOHN
z%`XCWy~89cx%GdJoC;`#hL$z|R}cgK<cI+RUL5!7{tbws%o;_!255UFUVk-Ct-Tue
zvsP#T=F-|+RiQVJ72kQduKF$<OK&>TWnxO?cTDQ7z>PV*71mMuKBc#vH)Ql~)XNFI
zbKw2C+_L}vPvn~u`62@9z8PWt1Cq1(unJ{vWOH<KWnq`fe<%t9G&h%lk2MpQ52z`>
z0^m-!(5NXS1b>rj*->si?HY#&=U?T@4;siK*E?j|Zn?4j=vZ9U0q4N^OU<#tW2Ux<
zjESB$1hRVsdQ{&Hf%|R~=KhkaYd3W~xsq)<ZtJI{Ifq4gtMRa1bw5{L$`RhyeRrFe
z5Vy-3tKvAT3|&y{=D}{7e3y3M<1TzMmK61KgB6Oc!GG%h58ID9^&(psTaA%R;`;od
z2@Qux;2Lhzh9BC^SgYuQ$gM*K(4gS9|KuLhvZU&V(f+E*_I69q*H#^u6^n0HYEmV2
z)E-bJ0Y-(9=2g;^65z&^u>Kz%e$Ib%06`gJcMXOc1YXS74@e2g>5X^0rpJvyc9%fQ
zDPikReMjY|63uvqpm=N$a4ktJAfaDDLU#ocwsotZZygSIBbFAOi;aZJzNCWOR3l<%
zh2+=(SLHT?M%h3ep|0DhcLu`$0nl*0YL~%(C=&uSIF})s7ZaCZt|`BNIEC-Sf4=_t
zCNslQcoTkxukqjK*W3TzM)r-CLP<6K_pi5b12qWk)P!C*(9_?)Pk+4qYouD&lKe>e
zAUXa0=QND`mKEqFD^MCk)F}`oYZWJBywKrRQXrnCBo-Jie6fHS6ak%(KQ0bdDWYUk
z&`6Pv>1lz1uksPu?M9e?-I)>!=SM=&voTNhf*)82ViWTezcUM@r}-nY(2cY#6!nl3
zn)P0IU_bCad>`J84Bo>|R@Wg)N+b3_`b^#zlu0@()TQy@Sp?}B@X4FxB~T2$rP&`T
z%m|eEgH(E&hvNn3!RH>4iUdg3ELwru7OCVh?Tf@ikP6Oe56IIx;T@F3#UxcGQJnNL
zL)r+mu49PqVTkTxXl?)exR9HiAa9Lci!9`gaIX3jM+Q(KfCOU)vs8ncxY)N?uqjan
z1fWj{sh7cjC=&uSIhQf&Cli-9n=mMs{kAEYf05GTpKre&bP-sqVB^pD)&4uZ9sW7A
z;#&v`4RwEgJH#7?K#|AJYUO0V-~aBu9DZx5_Pn$#$seq{?_axS<oBY$phbb&TIPuY
zL5o&RXKhdvZ)F9V7p++~Y+mJ(4VtJ3m~8U%tB11`jh3U}mSR*di+Wj9$u3ijl;eLR
zf1FrU%QkaRCdytv(v~J8vZbDzCj~CF+yI&bu}+o(brAKEC{9Ku6rD0O(+G+gnTvav
zaIF*(g9^^K6tx&Aog#(JyGCPG5Ho0H)XlCpP-GTkne7_P4HScNJU^M2{S>WHT28E`
zxS)0Gm0UBezdnAl8k3E0*SlvW*9E=zf3pKA7v{dT<Gh1c+T2r@q*0N+R+pr$Vs^8g
z=&?SdJx6cIBT%A<jw5x)Ox^I2<eQ9)y|Bh5_KL*PtENj_5@$?-9pKu?lIK#KOL>lX
zYZNlxiqm4O=dy)3mi}|T()2if%DE(pkjT6i7N?vi7&6J27D7%BgEsJakg?vie-jM?
zkN>SRI>!Z9xL94~;HR?5sH<bo+xC9p2MTHP&t=S>vA8WKDz;<A;!KRMuAXR&fH4^l
zdW!#yNDHV<cCPr7i-pbSR_nS&CbG&%gkF}3H^{{IkqIRtLzu|K8)V{(GM$4Hmw*=g
zNd%x|L(x7G?ScvBtI-G>Gh2Nce|f4amS#uRcIqR^$j9qUuHSWB7a7j)01+kAQ!7K{
z!_U*`!N&*_-9rrq6*RIR!bOy#dc#(@$C{Od(iZMHP);!HN5DO{RylJ=v^OOv3fv6j
zOi*B2$rdszuv1ptrkVUc5K|~B2(Dpv1Hu%4RfeIJ{sw?q;NX<tE%>q^f34NS*hJ8w
zV#HR5H=s)=gxu^<ZUC2Fj)~es0efEI1qU_?zXe)WKz5b=Q*>i!)k&xAT!Eucud8vG
zERhpucAJymX3iwx*qbXkOMWKuEoaUUG^YvN2d&9Mj>%Kfupe1Ft;m^mQy(T9WwG7Z
z2jFU!de3?I5pXrt3@(F)f3wVyz<sgxB~BS%<1)T>AB3;j8096@F5*Ts8*t-Gaq!#m
zEE9>N(TZHWQLg*oY)MNhXfu(^ZqS!~FMR>1V0c9?-YD0-Y5?P=vE$f-6HL*`3tqN;
zrJ4REyFV6(593iDDu_C7ToL$3unz^1+i_S)-eXs*4wR~cC0{l2e@c?58ti>GwS!;A
zS}SS1R@ME?bPg|l+N<kd`tw+;K?q|<s1+!ZDO(RiP>ELxzfZN`=x4>y^Srusa#Cc2
zYdOJbT|AG1!TGFwjvmVK!C02z_)?Uv5}N+k6@}YCKkfB6Dg=)`9bXK`6wUY|jjO>k
z?QCYMR)P-p?W1Hgf4E+%f>$VY$hnnV`1(~X0tID_pHaXbzc2D;sTZ7WSDNv@?PdN>
zRs*zDN@`i;?Sh>7Ok*Qwfx%hU3vErx;EX+#MeIATIJlS!&cv38^sqO*m~s26)Q)1q
z)4B>OHE=2*UqP)@_(Q$#YHy&Rgmhss0(W~4fY>T2W7{5pe^kL<3a^jo0Z?EyJTkjO
z4`51=pLh2Fri3o^02;2xU3Cx${sxN;v4v_xK%oRz+f3e31A+6&Jg?(go^&<J8IsPP
zGJw2M#6MFRVD^>)S2fZ3gFwEdOGRM@^5<~Syf=K7^sEd~3sYAlxdoW6Cl3-RZ{6@%
zt|BNbC%i3re~>l`-OzydP4-=Cb+gjcE;a?OWKh#$+`eJDoYEz@yOwA{GqqJc*=^sl
z#tOKTN7m6X=+24Hk32v(_G|a2Ih4G{jQ`=A4Y9Fj?Ur8mz(M!D+{wr!W#!!0atRZ_
zcQJ#OT-H)Km4m3M^s-!w+K7!N2Qv;is$L~yc`u@+e{j*Vv)l8~ubS;ztlR0~i3|-P
zvrjs&rBo=CQeV5*PU)E4K}+>C<GTySj`vu~o~_|>QJd8|jThMM#y$IY2-SrbiBF=R
z@Or;s4Dp=>H^wPkv$84XVN2?_qnhyg_#7)1JM9M77Q~fYD$Jk7NLOpv6dR|g<`SeS
z4T~PufAn1X_DG4UnOly>FfKdxb9PhCyCSxi{Dre;lU_1w_>g!Vqk<gA9qW>YVlsrM
zq9iIY<+!_KXsD+yQVZ}zbwv$=_VcsB@yHc%yWuhiuGeWqYKLH`rKWd0AuP`@8%j%W
zLIFlwY1tl&@FVZchfyAH3=cd)v2|XUx&J{3e;7H&^QKX`Cpn)+X~2&#{qWjMC`SUN
z>ow_eU3fTP`ulo_5HX#mNaoIj<jcNx(b3Q-@OV#0n^J7Fx!J)AJZPvk58elN1QjIT
zG_xD%vCBe$6e9f(YUEr3XOc}7yy-QxAEQPFjS8*VP$O6PbMIHVg5*%m8%pH4AofaB
z3f6v0ivJf!V1+Z6G4&}F0Wg=bO)3+Yb8Rvxmr#T=1efN%DU^TwJNVN-zkYj@S>Rtj
z$qi1Bw?;2XYb*-pQKV&Vpn_1y?2uXDCtu8tOeFhAc|#-&R^@g4XP}UT5{jHtgA~Y@
z)1-&LS<LzBA3<;ji<}e+@UG_+Oh#-nV)KyA@?X;TC_g1h0P^Q>?~mr*s%vMyWqlPe
zcWZko0yR>ncLjgJjo8ZMBklLF9{JmHSCrsfIp5P;Sr8mAGYbbEI_Xe!4Ely}p>wu@
z$=@km@!rd-KQ(IM+>BeX=%}ZPv68(@jy=vk)MT9DcjKi>9$x%~y<GVjE2Nog$Khyb
zF&vXC@fy}9E~_NH7v9Z6vN}aJhO<evE(q?q0+9sHe6N3f5-l>YnMtttjYVama7~*x
zGL-pVn-T)KEKF9<$W`wUN+Ch_62LY4!JA-<@b+u>?e_Kd-?x6#l2t-EKm6;jx9|gM
zjx^qn9B`KN*x&yi{(Sqrr<#q;a`0c$3(4f`$;ux&XL>mV53|IR%?!NZ2?wLK(5AZR
znk*8naY=s`Q_$_p=$bh}wv5r#5xlp_1qLb0OB8;x(gY|NBfzMRu#$_R2b@%mh5qJE
zDe7s-6j;IOsAC4OD49|q(D>dys|k@~^_-Fk+)nRC#?*Gs0gNe#ll)K1uB8%M>TOol
zAy}YyfxbP@m6JC|PH(a6gzFf>@zWUDWo1Ip3$1@<{=O`$p|l{|`%8X4uZC7aI$T#{
zyx9oNDYz5y`Wk2S0;Jl=nRH0|eaQJ7JfC^5hl_=gmr_ggFutT3(yp{yvm_CvSk5sU
z20&EuY$+T0d)k5}nAbA*){GtQ4cjDS`2f?{s+oFRxt0}`SC;gbm-eoETvSto!6!XO
zZV`VET(Z1*TZeyt7hPVST(e#Dj30@kigObKXhc<a#yM`KEC=U?Vnu4TQrG%PwssD<
zXS{1}_4&V|7peDGq;nbLbB}(K_mk5H=|>0YM~CT0&I?lbP(SmH{xjd`KeF6UF>E?U
zKRQUid;VKOZI8M*V)PHXY8Ysj{1|iL*wugK7B<2d8@}0(^jpa=@hbV$Mu3z5)Ac=j
zx(6jS19I%;-qSrO&8SN0o@Cnq70iakX}R?Zj>(3&WaURg6%#@=L+L?GYn;)bu4yat
zZ<uAc)D@o%4r39+!W*hIKeE9uS&nl{F|K3{qSGeM;uf!U+Z0Q;a!oaWOv~l6@?w93
zf)crvbA5vX%5b7FKcz!qQ^xo0Ey^<hva~Hdm&@3vj+<y(BV@VAfaxhJ+jP7^X5Wki
zWV1U8s(x&|N0vkHgg>;o>nLv1>G_V&@y@rt=XR&D0%rwkd_}g{T`hH8uR~#EX}q;9
z6!rnqnOdeE%C);pqGcC?tI4?bm|1@YPreqnkjV__TqjWhE*Ha>W#U$ev&Dei#WUS3
z(?o(E{GQ%{%gSBk8-K=SIBnJE4(nwP5?is#Y3G;M%j=J?6MSsG^<_1<$u;bJL=@0W
zsIYaumT>?IRE?>ydVdXu?-z@EQE;`z2okJ{MUTXyW5nVPR*lHHZr6!e=B|IW%y(If
z8!|NlUa$|uqDNxUW}LeI@IqT)Trd{5+bIH_nD8%ooR!Na1~7%T0VZ{9b#B&vY<kKO
zMjGVYIOquwtiY_2vhZ^PjLs+JKWZ^*?j<^@I7N3As7+=L>6roAr-9`hrQc4P7VVSm
zRuTNvM(Nk7on8lXio8(P^k9F^DTehw1<L8bQ633MI*P$k`WC1?4?4&S!@IDf*iA%V
z8C-W1f$mJM6Kw469``R^UjR}G8Xc&UC*g5sr8@&ij&K3q&iIFYDqn_Jnt?SZly(X*
z)&*Eg!22P!VvvdYzyO{SJRp9j1P=(`DZx9QF3~%^`H<i}PN5vTn-qU4q1$n%U6HGT
zN*3B@03m!Bzs3U!VKDN{)AZ>R=o5TA^AI`@*;hx>r}C8ixv6x6Iq=o_cbaM!ZgGHe
zDU+*lPFusvHBZMWZ^qJ4@0p8xp&0;i(G&z~eDIYr4o5;oUgO6*`DiCjEs_`#M0z5L
zS(U_WlH@K1^n#&rNZo&PEXH`5v-$|->tUMQ#4(#7k3V5}uW-;zKwpN2%}A10Tz`aP
zaU+B~eVNf+a#*=dFj7g?ZbYq5h#!+_;Myq!P*zxNdMaBa?nl+yuFLB|7&nu5oJ!g$
z++01z0-VXn&U9awKq+DgW~yR~r6<mVjn4CsMmpTE=oBUzYnXpjG~6O!Y>tlP?C~lR
zHXOO!?foC8n!HoE5B3KliH_noop~s;xsDt36If&Uub>C06WGbhLY?pB*PFs<`KW~#
z-nyQmd>vdVG{SQj^fJB<Mig-(mou#acTn`g(%duG*D3Ng35T6}ig&DToZxF_^*tC+
zITCywX?L%M^@o4joe`qMBkf+!=4`V&H&zS@>@1^9fY=K$uer*(nGo`}ddOuVtu;by
z+|JJk844sm-k8Rh46P`qqjqJb^TW^x&^!c}Pjx;@))-EKshP8OJa7Qxg&MmV%Fi1r
z)(Fi-dEQW<aVV#2!25usBtV=E>}1ZHQ7M@H#L)^3x}AU47UK5smLC`aX-`L1>M6zV
zuGCX*M^x%g{sSuYl;RPUx|8vMO5NG<fJ)s_JfKo{6dzRT&gzF$>f`KQD)n)8FO~W*
zyVpv+$?mmMcNyiVO5M5hL8YFjwVzX|9|Xt&mAd0#$AdQ;k8mJZ5&GcXqC2WIAg4~V
zEYp@z+`Uyf+`SV1=#F~sZF#b0f6`<P;`y=ZEFUKmt2C{uXYpdOt>^r9^Y=>7cv514
zJ+82)-RvE3Lhvh3GS;-WyV+abTr14+3DYl*O1k&k*W3R9p|&=wm%)E169Y6bF_!^0
z0wWDKFfUAHZfA68ATlsDl|e2hf7MyhZeuqNea~0uvHPMck`hHNiUHEjPTR#UFqq5&
z?FM=9q(+<FG#(@lF#Y>Il;gzibzsMFip^jW%lA?uMe-a{N|BaB3PqX)5(%18AXR+k
z!1&SB0%vuB64)%1oHS(4K4^HHe88cU<^wFy_y~g>jEg_9(24~m39VsSf2x6!-~kf`
z)@PSM#8@Je2IAs6sKFX_B?LKX;v)`JxMFoI1{-}Y!Rp!=%V81MsEnDW)!*S!a~m_B
zV=rzFTVN{WZ?Qk>`k|VITmuXCniU$j1}?P)g-h6tR>2c3eGSai<{AVe6QAjCo476^
zjJh*~gCz4AqCptuYbFeGe}Mwb63q3NEr{>8WPe+_!Elh2>?$(eYIQRUX4I<w3=5mM
z!?J{&lCP3;fLr?8g79dA4umcD*kmk)`R-2aJuBR!E>;n2g@#<kon9ED?viXFTMr4M
zky{UmVkfDG#J!L#LqmZ=Cb!e0A?7pGz01Ce^8GKESd!261&ZfWe=e9-dO|6-Tzf`U
z$XQP*q+qd<pu>$#<Zi5l{DEl-S16h5sUCWE2dL+ib}?i@PW`R#v-OeBVt1_vwN!tG
zohtGYDzKC;Vq1~%i`W|GE@Im}se5d(KfLU3B~&6e)N-g?eJ(;>;cp+cP)iS5OQC9V
z9WbN1)1e##a}VHGf355ufW~Uww+&NUaVH@*wKYc;_F_r83mK!X3|KO9&lkW4i7!WA
ze?5Hs{quA<zPg&thm+e+bN}{_i>v>J;hWj@&(rJE1f-|4;rro-;oDPMeEBdvpPvF@
zvExNKx^^;)3Qtkxh)A6c#{tWq4ByQ@&W5+)D=o&P6)7d0e_EC#{FgWQ=SI9e1y?8*
z|1kXc)4$;+_=m#P?d9d!MgdAS8Um;A<aQ~<G<ptfmt=g_w7O;(zMWmpu1`Ks&Zj|q
zLKxmp=JV<GRp7<fzkQj1cQT*Mr-2p^-p#J&(D4os07<p3<sFa&?#hcVNFL%Wi!X>b
z0UQ41&70x<fA#GAWI8_`-oJe}e4Kun-&x$_^nUWo6b4<*r&seEr@`8n_%OYh-Cmzh
zZx+nxkN!FR`C{_T>`RZIf=9r3XPAGDnNG|Lt9smT7PQ>eADwn)_HA@#z8iG2&km9<
z8EN7fyG)ndET%?JN|*hO;2a$ll3V2DyiSq8>QY)ve`F&gfL`#3f+(0;HA`9!X9OWF
zoo)!mC@7B%X*zZUK|2VfazU|DFin!Q3LLMk4|%M}W8HwBb*V+R7Xfk+lI@ZxrqNJB
zwo5WoG_c=c5VQuG6`cG5K`TIeDuQT?AleH-w1Oa_JK)d~N+btB&|Ti3Z9<+PUz*0|
z-_>H{e+`Ml$Sg?}iHjacw4r7_KYIbnNQjGAiC)N1KchE-==CDX%%6ngY7vbei$&}W
z4OIrgd(#O{;js<XI(Sii1#q_3{(vCbQhHVe?n_U~5f0u#5_^uK_G%!(I)Vh}eIUWM
z0*QQw1YSad4+sf#Y*DafbmFm}KpY}E=5tQTfBU22jU}?d3e^J^m4;7rvN$Y&EYUM3
zkY*j@5}hBCwmF?QG@@Fw90C#OT+rhZm^%3xgDWZ;P8f+dLSjod=Et8MiprYQV=dVx
zDV$&sp!^b;SnIkD)o4i>VM1gdOh|pege=R1@FDsua3Rh@I5>8;VL?KepXLhM^EOI=
ze;Z7?m|evJ<~%@ENXc`(R{<vwt`w^{28yTQquRZgtaGvCEnK`JusP>mCX*hc+NWkx
zPUB3n-7{&$4{ZxVU%bjZb?ark^=$iqsK^6|^0J=f<!L1^OHD4zfi4Gi1JE{SH})K8
zncX;}fUt2^srdA}7v!B^a3KE(XP2Wjf3yt8_sjPrN811E!w>%R)vxpU=bOI{!!H-}
zUvEFf^Vx61-)}E3Cdb#;lizQK<1ZICi(&lFB~WqnTvQs!-(WBDqxoGBiUNa;1+Xg{
zih`howOkoZqvcYous)v}1U0#>*Z^NPB++9wc#DSGENq5@kAcG40ctcij^67te`<8|
zt2LBKKYa?ixWeN`fzTOfacD~B+Em33P@_9c(WVwC_d7qY(cVD|qQWB^48?*C%>w=l
z8cUq~4&Vr+F2Z|n15}`6DvuiB#fdc|ExdCoj7EBmu5;T$*SYP|B5aW2?4_jaAW7(P
zn!TP|o6<%+Sn57JsO5nN%aZJhe;%%{CbY0PRNc3<a7;8CXNgo4elZ|7Bf$aMvZ<%o
zPPD>^IX1A=i=IB##P@oiO|)+(^U3gW^67HAPO+X@#dc|V*QD<5QNVBo2fyu-Qq>R1
zJ4pJGR`M=u5*j_@*K^}i$6bK#)dkd*E<l$RUiYl<pb)?C2j$1)oGc~we?KpP7E$?M
zF7ft<DCKC4gP`9BJ9mqd`qd^<%*vChiD#B>BXS)RuNXy<vV$Z8z*Xt>B<c7Lq~?9P
z^t#hIx^6f}Q$ptP4w6}ORI(i;LDSJs8`~ubeUks7?FHiXScslgQq4y3GTk9wx$!!w
zv=vgZ<<Y{HM+;jXEvy`(f5-~ZBcW6jSM&s4WC=yK_|0m2!6K{NZ)}%({mhOI?_Tep
zz;(M5a)z;59tb&mzncF)2^rx7yjw#B@w@00{ZHRW{HRZEtF3Nf;}wBb#_-#b?ULf>
z*0O^jWkGL|Ukz=yF%7gQ(7=N)THS@V+t_0WX@n!U3wqsV&O1o@7#}H>+G`0yNu%0X
zQr?4$3#Enn{{cyICzpW|C=&xUH8Gd5O)3+YQP3&LfBn<nzx;J43&Wf{>rfEQ%AGzu
zDIJF!n$GyhlA~`(3?NNtRAn-L!#8TNv2X2Xbm<pcZJPMQn^&vqaY1OKI0HazqmMIX
zD)Dm^)sV{YklWIClV54JLz!uPAppG~x_qW2q1ogSj2Ff%m1H|f&E5Fa5CmG_vymC(
zQN~Zye*r++H8Zn_!dTr#H6x1ePY?7aAG75p!S}Rem=-cB{Gjyyr6}HX3*cHWTgIi<
zviyhqJ&&qA<FDpW6y%c`R2o>R_f)ZRWb3?}N^0T)2&bW$4b6yw=o(Xah;-9T#`gTn
z!r394Z{xl7YG$5VVqf|lR5h4TshTR0J3U9Ef98tT<f8SNQ(KXt>N~l}Q+BxLCUe!F
zD>~0(aOQvJdu}OQyDB%m@1x@v&dZ9)HyUa2vFqiIab+`Cy3uz{TPC?c<E*uFy<z35
z>7vP&Y-3>yN($!&$4<78L)W3RTA$^scAoT<^~l#CxAwX*{V%28X7G?8Kzh!d!tG!s
zf2|05Arq}YisFJol7~qOq?gx5Pk&HcU$%b!c$_cVj{*Rn0Zw{&)<+B|ptmBzQ(gKJ
zUF+6!H0~)atY0u>_MHp8wI%6HH<@~#C+$gjcWP!o>{YQ(9^YcILY>3w148_ys|_4_
z%NV~6Ua5Q^Gi+_Gqh=dS@emrb%rY1Lf81<u@aNg?W`ADu1{1vS(C|izSY<;>T$wH^
zsTQ%}xP&QMRiA9JaNy{~Rv&_J=s0I-i+VX3@qx@EKv;}^DIO7Qw6FN~k$rppvip6h
z9<V|NoWB0t#UD^Bp`-Cd+k@$ke@x$YpE%WGz*0yWlK&Z*J#T3FTmWH^B#5+@f1pFJ
z2C{E&a5Bz{w25ay@TCj5hpbRg%yW`rmlRPFv^1RJd9Ga4hmrUCFNrj?eC{Bl*Md+R
zpG}Rt5=V}PDm|&Fmqc-B87Mk|dZCf+@Uj|Pnz0au3eIzik_=kEMhX{84M;!()q+RK
zq`rhAg?KV7u1CC{qEWO3EC{~be?ZWXJuenU*>$BG@^7;+jEEYngk<8EtKESxv`gJl
z?L;YVFLzYCCxh1No_hC2&}%`_uj}0hu@X;YRejndF1L6c<CI#cUcRm}p%vwEpw^<A
ztaB3}ng6A^hn7Af)EH?=5jcY;iUuL%OeV{vWd&hnPe0_`%e>}fpvDaqe{w2J1eObP
z313DiNhWTP7RWcgXfszJG5bdQ=;n`p{upkb9ktN9^Tv}iwvYSUC)XgfoA=){rfeUd
z+3ah(>#2+n!(d<~eeWTjT3~*j+t>lu<UC=!NMQwrz~=_jW4?@uHO!ln&WLIg)qtjG
z#0g&uYILAYRssh#fq*rJe?JA(pwq$xbw^OcD6$;zBH&`2uw>MXBHK$^Z)=6Ju;@Lk
z`PJZwQDxwX10&XZX=4%ET)wv|h&I4kIkLGcCqONh<A&Id7HEOtGH1#dF75+$<_qli
zr16r;IfeH$-jU)jM&xX^J0Nn-|LusJQM>~p=RB^vb4A;%cb{H&e`x7?pt{Y?H9(cG
z>g_<4Q@kCha*CTkmAl!!fGRhwZJ@g1?6(9nh`EpXzXPhMyl}z26Hvv7OSFF{q^b#2
zc?yf~g;cd2LqVwpf(KS#%PptQCctZUHc==YIh!DFU#VzkDl4-esX~@HX4KX9x9#^U
zP7){@)YE{0a`FTif4EgM<$RV@IKtVN(jnxb(SYN+PFf5tbI6j)P5h6BVg@2#(K9QT
z=nQRCz(>z30w+$5Iw{(zQ9BD*SRMLQg{d1{lYPjH@NVY_N*nS&#pk`~9O39;Z$3lV
zK1V&}j$mSLwN%XNxdIwq)NexAIe44i%<JJ+9>G&iUqpA;e`j?-dkheCVp*L6ONhh%
zDQTUixfI?L88eD^Nb4BI8<1B`{TRg?kQeRf0+n4*e7S)l5_(q)iuQ(Wv#fR8KE2^V
zZbA9>d<%jWVBH<9ic$Q<`4*e)4*3@6|Mq-~QM^OG#d+M$w^;AH=Udy{T$69{RlPmm
z;uLSsw>ZVke|(F(*}d{DZd%*<){3*=kZ%zr5?XczczVfti)qG44N)9sC(2jWs{Geo
zG&`QR$+yf+dROzy$~t2O9vPXnv#|kxXY5`24>tgA%L;)I1)(Ew1R*dCm+dySR(Xk=
z!F--~bZfYRBf`eItFoTYk?A?QsEFuSiilTOA_fZGe{&hLyBSqH%*~lU4*H9CpNb}j
zdI7v&y#TZq==3|)3xKS_D14`S0VtcUdH$+;0fBoC?UnQ*<O7@EC+hfdO1*c{87R#E
zcS_Adfgx`t_@I=tk*ub_DAex5zrTO`%h%s%pP9b@Cw8U$fYLV={PZ(_bzZ0tkAHsT
z@1Dd*fAkT91gR}f6iMkjb*KYd!`Eqtv?7$@`L6XxY$cobNZao`IAtX+cB~W3O^t_0
zv9;Fue)Nr%t9gnfQ)453i&V<k_?5=jQSobh_1|k(8uia<_ifsJDq1xbT*`1|CLApN
zt!Y&ec6YIEI%~EzStWdsswyQo8Se?NHTKfBe?}12TO?Hh<)#!*<EoUon#8rGle#t$
ztxBOb)py>0L<>w(YMIljbk1XgklsZDRYkX1zw_s>Kg+o9H0opbp6QT_Xo1Gk2j@ia
zA<{}Ir}3-xgwnnC?y6@gu7-A|T4!4GFGhr>{4{keW!WBP;Xa+M5Tz(c;28?xK&Vp$
zf5`NqD&j~PvC3d90jRdY=>1sGLr9lrYm4r#T&uY{(e?6Uxeg#cUSP`BmC9(YyN+G2
z%05RLw!#<RD86<DyM1kCMaQkunpQ|5-*72eJ>ck_Q4NH!7dqO?nND0%)Yiu^%PBIb
zbCe&7hb{mg9k+Zm@VHpDjiO)CHoGd$fAs6O%a#Q5tO*O<@~iof-712jAwS!GPNCT+
zqR<_%uNP%=p%u3;R$vUn$Z6<my#*#~GnSJRhYSc=dGbQF!^KST+OJK%pxwHEbfsc|
z$5wiM&>@zuVNtV)lH*vY-{dPO%IFm5OYTpMBHio2N2QDDxnzFqfrYXBkHeV$f68x?
z!7TLNepy_JA6NaP-y_f_NjQwJ$M&`OxO1B5UrbvyS(|04puZdoM6Q*7hB?W~++r{H
zV7cw0wreYj%SR8Y4^SLJHG5b+)KDvB`$M%K!ylyZ^0AjgKF;Z2w8GTa&oMgj>Oy~Y
zJCE7+xmd^1L35se!4p@MpbO&`e={YYlVR|=#yh`8pIXg+%{QC$`+-6`yebi{dZz8t
z?rIfn{NO9@bYwMz91Ir>4=oRsM{?w&h8-vep%BdtX<vg++-je+nV+UvHNyfdDnF(>
zH!H?7KA&*$E&nrvnj+XG9p^`y6L;YVS}D&*Rby;Ra%8kN0xq5sDerKLe^F=dukmCl
zo+z==rylLGnlLFW7HfFqX!9Ym_P9nLS6KCV4Ghn)a}5m7z&3;7e@$bzlCnN@IqMmZ
zy@r@q%_zP-cUYV`%!T+A6pb!^lw3SMIfHh1awfL872sL=@r?}^R*s8jZVvs)a;1bd
zq~`w1^WxgC_l6^olAwnTQjV+Q!dR-Mh)MrQWY*b=hWe)%xfD5s)>wdMT}x*@?dcYf
zvaQBg>D&yc-U-0K#5;$9QpT05hvAOGbaw;eGL<V$9zkjRNQ4*JW`Fx1**SohLDw}C
z0yQ<4k#0Z~0Wg=b{{ttNDBLNX0r|Ja+$s410^NnT!rm#51q73zHKdm@^(hn<GB+SF
zAa7!73Oqb7Ol59obZ8(qGBcA=A10Tj<S7(?+GN0P8*&?V8rGhASl|CJ%90&hcGINE
zftIYFPvl2Z6q;9q=3BQtboYC9CO{Ml)*N08=>+PRPzSx3Km?x29Nx^Y2Sn>twki&h
zStV#cYAc&Nrl>1OXGlkgHg~~M-NROe^7>8i4`si7Dr058idT$e#by-6rKc>L--qvi
z9ae=AlF!}Tw12xp_pfW!mlS}lm_Kh_o^VXS(VA$%C-eT-ymUWW(D5kL9pxiv-d{~K
z^0V@nQ29geMP~dFDyhzRpMdjJ;m<uIirVY;Dr%q0JkZR>`+cNUM2G4LSs*IAqU?$?
z=|Vx3p{?V%E&sUR+<By`2Uc#uWs8P?PjBdKL=7EuuoxkrZ*{OZvtV3J0W2*w&YD%N
zdyT_H!!PJ0qV8BQjVa5zUOs}#M|wH+kEGfT>`nO+Hpfq|+iYZ_+ZY8Bfi=9&_SxX4
z{_Mx2E`2LCc6?ryv&E$nRs9P85tckoPTCH8$qhY)E%^QX(LdPc;OB|)X;!d*2N&Wk
zvto{DCk*%YxN7@^L-7~K8bt;QsXdLTVaNyFN^B^T(j(4BW{3eCFBq{rkCJs8)}z}d
zNQ%@T=1AgQMz}n~oQI^*+v<po)nCZjhDUb=VuBz`&pdw}(K|r(Sc@p9@05o1{g|WA
zm-8M4RBp+8VhHJ`yr&3y|J;y&cSZ}0OWwmO?~EQq!eaGq<o$~Js4dQXczNz90FW$v
zh`Uv5oYq(GdF_6@rutaTHcDgs)Z13CnEnk%bnbmw)^c9UWsTlJeQBv>Ge7HOdn~2>
zztobI`+0m<o_@nol*)7ONIuDw^!&u^iyPMh5zoO`n^lEW#&Wk_4rO#gc03}CB{x<Y
z+VH-_Df;NhE+z*uz6kQooL|m!5}Yba7F|jD+?71zuRk0a)XEd5hYwGWr(Wd4Xw&N3
z{6D_Rew3Fn^(hnqGnc_!I}?|8`3)$S$LA?7f5EX$(Y5i_#)=+(4tn_dq|^KH{pria
z_yI^6>zp0le+(c`wEP9_l(%NE8Wm}I^lSL*MQgNL^HE-`wIO*q`RwK09G;22>;QQJ
z;qdqSABnhC=$YLbabuif$C)?@af3qeBJVyCKTT@}<Yg0C;z0-)aJWX5Lr^wAq3V*z
zmvrbU5)N;fijONHuIOY2n52!Lmz3x!D1Xz#GrX+Xx98r_r}3B(vTe-7Ym`ZB0YZU+
zB1j@lSM20xYcK3y*>mI+X8A`QMw=uF#l}3_N2i_1#-dZiMTZ0Ll%oRO@vMYDfy>_@
zej9oF1K3D8Lt9V!X8qtia?mQ+=sWT4YxC{x{psIM1hk-x=i#4UPw@d0DDpTUuz!N!
zhTq?ZA5VX_WV4qlEx(!m7(M*{IW#NZk_QJZ-k_~Drt${R;vzh-H43nJ1D_LGFA0yL
zEyAQ-&DbPfGvmA>j0+dm%9820B$RVGzly_`i-%R2B_3Eq0~@iA#Y}DEytU#eQ2sN@
zS*<MP^rXOrmNM!}uChCTvfj75sehA076dVvq6@xGpUK`WACgys*Onu|s(i?(9787`
zE`stYMX30G&OurhPPMOOi!&+IR4ktTHB!R#{j~6Rn4U%L&6-tC$pY4isz3u2BvmWR
z6UvehD&&$AkWw_`MGWO-^w~H;qXKu-HszJ5X!RJ`j*p*RBRO>Vx{Im-g@2|c3bYOg
zW5heTQ+LNPps@GmvyP#rcEd3UeR_+p`;Q|#N6j;b@XnR23KDml-BCAvf0#iNlr!O0
z-j<b|XvaI+G`g<7pTC#$q+a*SJ}3rKE^tSc*R={>Niy%wd<PaKl@Zzsi@5+fuat3N
z6OZz%G4VDOKgJ{FV(<7ME`O4L)_h@4ct}WVD8}mBY@p36d9}ZIB;Yx#oO7S4jW;7!
zT8d^n;$=)W+MFJW88mh|9nv=ZR{9|&txLOmjE8y`(n(qul1&SwkLmW9%kO0iP!;d%
zXlKO+TB<ihUp-kerv8@yBj`_fY2-*<VkdQqVa@JGaO#7S3b<AOZGT4~oI+cx$Z-QE
z6R2869)xJD=K(w38xRE(%j<UcXF$|TwWECnQM>HXQwM1hw^ObbuS2w^x(A{Uz!Q~0
z>@?e1zon4YP#ps_T?uOmVU0a@oR@QA;m<y<is;9u6<GrBN}V-Jc-^Rjr>t0Sek63H
zWZ5XXBXr~#SgUa*U4Li%)T%rdJaV>dxSy|Mu-ZI@j>p4R4a0rDMM%mxvueyu*CMit
zs5{d2eweIPJl`Rj4?62Hs9x`dI}Q=oG~aR1+*sY*U4MK}vw1s|?rn%!ELr;g+}pEj
z&PD{D|BRffm2X$>bkaQ(5h)A2yi1u1>Kr3L7k`iye96ynDt|5Wr|E$6j?^YD!Fh+w
zos5WT6n}9a?bjNOl7*(b(#QE41zJ1}K-pjyg;8e#2|d{Y#|`Z#nY@Ftb3NBQB%w4;
zj^;XIjOXR#&V}RaRmtnq?WD(Edh;pL$3oCKzsLtYw@={Wyo9<?hwAQhW+d}|jy7I1
zL&wqF$Z_r3d4C+*1jUGYbs&UjrQE<O=!SSxIUyo?@ol4=G%qsA$;LpZ=Nl%IgB%*~
zF{S`{dCI2gR_743IT(u73u#zha$b`$UwmGZt{81&Bqd2agpTt#$|(Z~S1!!+IAS?6
zi6fjeq(}pjan?4i<Y`<AO#D`lOA@uks3omzYn6AT;eXc}1-F?H(`f0OC^qAz^ve&C
zF<*P%J8rQ?2`HLj3kYD1bm{-cumyY^9N0UCEt_dGCFq9YKFqcy#Q|Pcs*3ybCOF9s
zICt13AJEE7(wIJPY)w#h<Z@t%#yV`s<YtIY&bEVJ?69T(qUisnbXeZhBm>W7L+QZE
zDCO|X2!CAXQr<6<K;Ji$@H7W16)dbfYpQS7ZIvaQZ7RPanmEMiZ>;Vxg8i~z2g!AT
zsu^5$1;%O0=!<bf(lBaLa^Et9rIl4|c3sg!GE2A*b5$LU9Yb?Q`6%ykl3;D7xQ@^~
z-#g^Ov&3Ca4s3QijVC3XdY?f%|E~il*1V8#wpjl4099(jV{@5VxyZO9-x!>BpN2!F
zLkUi_Y_#<JZyVPEPZny#qaX(Tgz|V^5%p*y;-_`a9)ZatAbQ6Q_18xw1%Z`p5pNY0
zV1XF;)ly%w<bTZAH++{d^(hpWVR=9km%o8GDVN>uDVcxtrsH4FzaMN-)|%+j&-B&)
z_4IuB=g_K-B*q2k{{DJMH<-Xsz>b*>snOp)b)OG^w67h633cPgw=Z2Y^0A<&@At+!
zCo)41q9I3}bNZcb74+0<QR`SS;>+>D*<`OMKH~GJlo_IUZ~&j+_>d-IdOT*Zd+=Qb
z5G4Vd#~FXiqF@0x0qYwOZ`-vnLV_Vhrx=mm&MK8Z=#`{a^;M$8G&UF(tEt6oy`#E^
znNg~v1xfx=Ry`CS(pQ4ZBJ5MSC;N?IZ<zf3#-v$-Wpz|n_FKl##F*xq8AJPxA!pc!
z!5Aq<Tt;VFG32q_C3P_bp1at&SQqyzhhhZc4c~u%5&pR0Oa=5G?@Lw*lRIb0sNia$
z9rU2b;w#CrrvZ33_n_X+kZ0+B-;1|IG>&O-h9a%UYrZWjzIKYDJ!K}Ut6<N^2gT3L
z0geUS#{#0h22T0!^x)wC`&mbV2A;rEbE?up#m5tPrvcvc9Mahizyi5jahm<4qGF84
zk>-D^bSavgt5&;p-Om{!(ReqqITSbXGmcoe>hVDs;&QjOG1{uM{2?|gG#gQc-d&<y
zkC}N_32FG|fz%Q+NF}v%h6-UObu$XH)!aqeH_CKpgeCAPjP`RvaW^|z&U4u*w!(Jo
z@hS?!(=7JY)vckU-(FdD!wHf4(E~LbnXG>tdS|@%tr=+7pqGO72z?#70zC#XdM$wR
z#pp2VeFJi`Xvk9gCMA<>_5CC(=u8N4g=7~5H{D^Akao(;i;wv?x!uGsmW}4B8#*mh
zxxIUqBO43Y#`)@`Jp%sP-d)7`#lHOyN(C}jMuMJ3Gh22PtWn#oytgGi!bz=>(prC;
zQ(&zpcH7|5*f<b3*$J?&<QtDOVvNPYWQiSelVrtdmaVKhXK&SRogEmUfxvB_V~lRz
z>sw+u*(teaZPr~hqpT_6B0(rAm))N(F-3zAmb(TE)H$s#W=vF^q%K#fdzQ*<?vZ|m
z%ZXv-r9I1m!JrKMlpM9j{1LxfE>C|B@MkFX7`&lhyIN)I3^BZoYL%k&J&QL{tx{ma
zNIz$_N}dcmd6U&@vCyIBg6<BYx=FNqq8Vy}=&>UKBRuqDayi%F#08^Ju#r-9#s(wa
zuZ>m=62=5L(b7}F#x7IKL9_3;d8%d@jl;OmO#VO3s$W4c&c7gN{1A7t-gti)*q{42
ze+8b$$(EDc&v-R>a@Lb-D<A16X%eAl6NukRlL#H@vdb^gBtk(G!T%CXB4jX8!uM(t
zt(~~MyQsXV2?C}VwLFMYK?X+x9FLj$#d%0_%BJe8A9YOPl<tfLMID(|Yh`4kl`y&M
zi>>Nah`l+tPR_umW0MiX&@_L5Py-F7W{+Nh>dGueKMu{7r6kmdu1YkL<i(Zvq}O%5
zX|ZK1D#?MCbSoMNrW3(b5{x<Ml*%SDOMaTq4`sT7uvjm+x?ZrhUQ%LSk`JYGc<)&v
zQ0UW?<}5KqohY}C5^wiJ`=i9`ZBY=7BY|GAE!w>Cb=;=5M~_3kJn4Tu>}Uf5Yv4kj
zRzmANp+r9Ux}zfn@3$Kx=+?*&?Pe=&o!xGbh1Dd=&3b94SLO;<jkpzL$NC%K9>_L(
zo$!7E*^8KjMV&G>`f&{{A-CHyp$tUgN)W2ybLjbOQ%B#xtNZ3MjFw%LeR_Br`_i(Z
zmDWEtB=Yo4(@@#V@`HaZG9xxQVm(h7WXLhoC@g5K^`e-Lq5SlJHFBSFJg-Q}_H|MH
zJs7xqQO8b=9o)lBW{hjT<%_87iIy96Z?t(g9mlc3*uKnOu4~|;!QfiJE@d~kkxcrr
zbLmzM|Hz`>wY%l7YDXl7ox4xFRE4tmR=d(8yWUs3u1(PDNOMN0`?QIuvs-0+&S)x3
zjIW{~Yd-JLFBPo2WlR^Ysmxk)g~jsD9IG@Sd!6dod1ydx=_f=6Ql2_{G0fv`{nkX{
zU)M#00+%uMDHWGOT`Us;H<xj9E+?0rsXYsS9}vw~@c{@zr145XZZ>=ia#$F#bFhQ#
z=8(Ui;-iXeHhZQejk1Y>(adnjV!c@PRk1CX?~7cXZyvwge0VZ;@gjiodinBgQTT(z
zze?hugT*T)oKee{pUYqFq|^>ykNC1u8iOy7p98+b{jE^}XDIwdMSQ9ITW=MP7+hn2
zG|qdvw=sPWb=|3q8<pnZ`WSthukd$;v=!X{^~;}cWM%f3FE=A|P8k6*EMI@#{I=l4
z=Oxd+k;23P%fFXAER+Tt7b%=l-r;*R_;2|gCw}<n_iumt`l}MS>|bfBT&&QbVpx8b
zug(c=aozHxeh&gd1mr<P<QjD+g^fmkIbO&EUzZ!Jy-)^1H~J%uMqz{1x?60NFu~(Q
z-k=*I9uh?!rTu>NgTz(aBFn^Vq;H8zU7Nn*dRmpfrdR)+rsAr<FE`(on@?4%;IvOu
zf-Hoi!+$faDq)pJ7wR<2LlKQ(eITtwF9G&OJ9&9f3m2R<C}9vDea|dRnq{Vckxp(J
zLRv(DhUk49-_gS$D~_;}f`F(NNNnijqN-}QXt;g;`m?V4j;o%#_d@4ZX(KGS<<_CQ
zpjD7mKDZF32e;0ylj+>zSpz(07EdPVe6DDp+4)-2zzshw9X~~|+a}qk!`i@T0$>kl
zgSeV#L4%M^vl@2d7l~j4x{79hJy?u0IVu6%>-26h+5ko)2ze~WFi>fhOrKo&)!J@W
zid9ypUv<)G*5-$&pr(98z{3k8A35%sE)L2PQ!z#sYICVH9(45bks8P%N9Ixyhgr(R
z`(yf*M_lFm3EyLtAL!TOA63reFP>zTkBxpA&D2cOWL{S(THIHjNQ;(#wQe8k&WQsk
z6w~)Sc4Df!Qcm2Ts(g7SYrdBvbTf6~Fl8Sb?Mp0P&cxt7#-qMH>v_i;w1$J;071B@
zN4ol{W=Ka~P{&XrtNRR!@3CZD?azCPc6f9*JIy?$a^HChE<WQaW&X@l%DQCjI)^On
zwwA7YoU!zTi;bmer{tY~@K7Z<)g2$1Mx|Qz&74(k3U%fjNPf8Nt5biev5B<3D;<7F
zzN~!RwMa%7D$i}-T+?*kpMS{~)cxG~d8D04m-3WP$R4^HgxLxIEtHq=|Bt1gCR4f+
zGQxRqrV@Za+~DyYPpOIrVHp4}i`oz@!*ivThX#OLf_*3nFXVxL6)fY?6)bCu8CdZK
z-4OBcKLpDZ!&}VUUcMBMVX%lkTDJyS^U?oPJZ7EnPF)3$K{P`7qkxQN69Ra+Hx!8g
zY78Zm_HY!Tbu7%zXcP8@5Dnf0(Ey=gbm1UG1Bc#&$Eb&B(j0+kQlEopg$4%q0&Fh8
zG%Jwih8JJTRZ7Z#DH&(DhDOqqh;2#GX<S21-r;<5#?WX?7<WscDJPq?1bvXtBPlUK
zXs*M_+}GqDcr7J$+pk}N!ZKnir4ejNOX;ZbLkS;70MMXchF>Zd`;?Re*=){%^-U)G
zG)=DLRirKAswB~_z+r&nDj0y8I79J;mU>rwbP05#iI$ju9nZa}iGCe=Ag%MAPT0p;
zl5k&md!+rUUpHAT)4d?=5t$A^=?g%iJMOzc0rr5xEj4p)0i%^w=2FzC(*ZM#4cwSA
zp=zL<Jc-#Mgp_7;i^4$e_e@4VKkvtcgn!Jq5g)jR5)O94Pp`6FQM?sC`ppi$gsp1q
z2MT~}bUQVFT|?+D9VoP1`Q20($?l=dW5(kvMCd#BDA|X0>Rd6?d8*EH+T<&uSC4zH
z<pyWxkr^sZ?;o2MdxYP~>bupZ3pXFD(W6T$PjAn|H+0+IHvK~zupG+{m&OZR(+dKX
zfVlPdzb)TxK8-}HNihxPT$uiVEWWiG`CPlJQKIF4rE_=yddn+xKLpPp6NYpg(mr-5
z|5UW0)!<9nS+^7|0zLyH#be%D>nB;giNe=YuHlPjAt6folf_+)d~Ps+WW(|s@hVrI
ziM;)#bdR%a<SDLnKgONS4yPvD|DKy0V_CIs8qcklp&BA_tq~zlVvB`@syNfMmNvfd
z=%k~6byxZ~)39ofgw+}&GAjWCrAJ4cv4YUr14D$j?75mXk64$2E>>v5fPep>J8U%1
zU~dn=A+^!MX17%U-b10DP-xA`+(f#D8ng)_RZpYBWo-nViZ>b{wAcbi;&JcrXxAUK
zLpm}BghIXtPHe?GnX+Up_{L8Ffk1x0Zd+uVRRj*xe}u734|{l?lgA86T^H-TG?5sO
z7zbm?AD2lCy0W+WikW1f`0^w8E?LERFMMbP+Swwkc27+^H2JPEBZl%?AE2C7UU-IX
zolL8=OrLbl7Kbv>`9N!|5>AEwx?rdpZjHBnb5J%_=ctr_JzqiBbL9=UA-EPfS1W7f
z(5>T!e@9jL+%&8=*S!~N<ABt|NIlnVMdEe|ahA)dF_CPzyjZ=Ts`H!**VKZ~t@%(K
zrbv^WC6hU|&V9KLKP#X`Pu9whj43vn>4O@M6wID`M6T8W#kjLupYqFC_OC|+kId_b
zRq;&@i8#c$omK^>zFoIAsRmRd75Q*sl@VUXe{ofoNN1-d8L7*yE!9Wb9>GJ}_UT@m
zi|M{Pq8=Qvqr@6!4ygK~Bx9NXsRi}af)o1d#!X4GtCUw}9MS<dHU<q^M_xfIse<fk
zwZ;u4=dyO)(&={^8Z}DJeNJJlm!ZFX;|9m6<+f}kt~bj)CButV)~Y~+A%TBU6NhW5
zf79v)W#n2%MC=|oEMlNAbXg4Pa>>ZMPJxO?#w#hK?poHlCdBp{ldBsUWp{HmsmG3|
zz+K5Bm{F@J0+xQm4(aIVQi9B`>arFt9X~EqcA~z=>|r;uQH!I<aX$axN^l$Ase@O!
z-09#|OV-0wESRrDjXH9bxu#u@ex<VVe;3tXRu<Z}vl1N;x`=1JAIzN=JtSbj9d!`b
zUV1^3HU?qT#uOVUQ{CazqUCF0xQ5hEhd&U8YlNrjn*E%C;Gr}6F2Gx-r_iK2bD|FP
zT+TE)(gO?DnjZ0H5C7ODP&zM^?n^A$k5iXWww6Jlet6IR5s1ju<^?X$zo%TRe`rMd
zmZL#gF#%usA&NM=$ZuW!lE`%QV^qjgIV!C9r#gm{FYm>?XZO5Xjxp8=g;NF%U)JZI
zUmwalR;80PZdxv*{M5j<um4yfsZYt*ZV#khn!kDKg;m|m&>^)MMo`fS84o-sf=+JK
zK}R$NW=tH>Y*L6>r4Yja<(#ure@?FNEIJs0iF>A%BhE6h$)Uy;SfRqk7TH^zt)*f{
zvu5)oozdr>q-)7Ow7Uw>4Mk_9T`W-XYpyB`%=g)~DC6xN4$obxk|ze#f^o-n-anPZ
z-o(!_t#5hGdXoz8<V`7&VAU{4w$?=rGw&OYFo%i!sQ8+b*P<JQwosn8f9=6dG{V{V
zUjaADVHK~{717N=@%8qYK_2ery#e(o$Pc=+>16tE3H84xgnA>C3RqvqAM5zTrSSnV
zUAE6f<LH2n0M|6mn&Ruj@iF5dZqKqd6|6=&<7lgew`yciI%>hv9ix6IpN*q7E%#vR
zQ>bM=<y^kw4y`vs57q!v4nGnj%WgBYP<N!KXMskQ1}Y*Pg4d1R>h4`CB#iPw#_NWP
z1BKQ2j=AGME!)=?mofDz6qm8WITMv@D#8KXmo5P+J%44U>Xh7^waU>fH^xm-WaMwl
zX%B+0N^L<;EGIRYrnQvLt`KTF5Ag|S@tIe7fwVV!hoT26>ushbf5*DgkByIzPhdWh
zrWD!`aMFtS3C5b~8|!Wj{YA7HU`u`&y>bj1<hRzuU>n=oFlbcJW`(Ws>u(EQSDwOA
zC+U4&4}Wh9kI~7Y^6`Fnd-t^ax?33nSmh&3&tG=(1sPC=rGflyw9~ggrjNT1Bi6!|
zR+_iuUsh*XE+D@Y7mk{l*N(Ydm?E&aEltiC#g1E<r?*9fkC{1PA7Kz@QqCG9X$}T<
z#JJ@qnoZX7qxVH@toHd=Vr2ngR#tQ)G6~J0Q-5HQ(7+}Lq7b%rcqp*@^8G3Je!=Iq
z*iX5(4OT@qs_NQycA4aAA_-s2iHbS-t2qczRDPu4h3)~~@_D$zhGP&)ZsM}jKnVjW
zlu&UHsH-!-FLf5HQ6c!zSv)AuIq3gz7VC_{5U%Jf1I9zDQY3WNUy1w$q<0VptS&!d
zhJPhp^J=HS;kY_E=>Am>YAq^*ap9o47=O)1p%{zn^Uj>5oECFJKdcP7!jJ%xcDeFy
zORl3#yM4#-SY%mTFU9JZU0y$Cs?sOa=T44vS`J51$o(AT>7kO;th$o+(AK*yHFmb%
z?C=6XN)}+3w8VDa0v(=|UnMg0RtYojgnyVUg&CS4A~XGRGONA~TO!Og4OI~}q4`+{
zODd#ZDfHN4CDp6Kz^QNiln|5xk1xF2(^}`{uTlypY|wGC5EWTmShv`Zc|B$w6h(6q
z^LCiKLJBF-mi)qhUz$X?7~sl%#MUsmzSa_Ft9qn&eVw(eNcn7Vw>bjj3il52wtuhB
zmht2@Nrwz7^5F+oXe2ou=!!{az<97i(>ah5)+jeO6t74OWP<WSSZCm-JaA#f*5FsD
z+=$ByyRgt2zxonqwx|?!?6S-m!~Uh#W*jO{KhW5Wk1Be5MPnN<9x^tOZpv6MH!S0#
z(%=?`Wjn(<CxWt8Mx9902_co8Uw<cF_<l;O8G(txPXA5`tN<-VM&|&$6SRhZgz(*)
zzdwEa>G^k1yzL)Z%9`bn3^7b!$}0y)kAC`GUP*ZbO_Nk|#<9yd<&b>aR^Oke9kFqM
z;rQNwp9RR-DG=hkGntZ?Wfge`D#r`D8CcHx+yfgy?SM>vr9s{VOW7;0ynny$l2>{9
z_qvpK{W<MEPP-39p4OpH`ATQR(eclgJb}v&X#hHzRuDR|Oa7zCBzpGP1JIt|d8u;2
zkzo!wX0b(?sh!q%u8@xJ3SwLpLj^U`2KWIv47@U=nWyDw1rz62bViI5igdGt=f9u7
zlzpFg*Zb}><GBGeT*|oV#(zi!X9ZOX6gF%|%d)yG?j~3%TE>*m#n0+FTb{MV!3Tbr
zI(jl=H#27+&Xxv%vX;(}#@GjSQsDZ_6Era#NE(%evbvjq*63DaW|GVO^;&%Ht2mqv
z2fw(-4@!<6v9)r`FUE=IrNC38LSJVDZhPeEFFq-mU1<?&L$DrXCVx6cIn{<F)Knj8
z^7s7!^vFPVH8@>0+gqkTTiKTF&vu+`|F3DetfZ!!E;m)5<-m-n-ByKz&U4+jY+7QS
zncdw;<5fD$Dw=a=;@#x?_#7TN{a{#hvzMN=Sk?8E_Wb!fqUDOJnY+3Eq?)S%rOkK{
zE!fSK^$|8sdv8=0X@5LrIlw;dh5eY@!u1CAoG)F7D?#gW*|X19XKdHBl`>OkG@CNX
z;WZhCzhKC3x!wsYP1J!x8Za~|_?ey*N6OU+*+nQn9v9ZS?3&no&9l9mMrXJv)?C_Z
zK;SayN46TMIl=pD-aL#L&-S*i-^JMqWX~(w5r|w?!yM{jzJCl&m!mHGfZPt?V}=WL
z>C&|LN4og*c|<4A(pS=$@<Mm1^VBxn1iC`o4Av=_Ct*VVY})0Dz*C>n^(^WXfYP_8
z?iG_8c~e7GTV3M)fiMJ)%{i`UgiL1=dAon!x_dsFP0(@iB-Q>Bd%6M59M3a^qg1k?
zXtScS8o@!s*?$+~*=MRv8P%j)_>c`IbCZlEq1l(Osb0<>WsumC<-%nk44TqRAHx^|
z<^kt>o>W0)<<^ld^~$6^3esZzfcM}Pcy2TXgX9teqF+P-?0o0AQo%^+1tS)He`1&6
z{xFWa5=UtIOZr&2LaLQD(m5p$QlHrq&QtYuGDM_tn16*U2{#JbKpp$_=8y()Q-d4$
z%~p21fbV?GvzA)Bl(%(fw!n6~>U#=W{>>*J`qzm0#M01KzCmx>n?Z%tyrT}EYt7Kw
zK;eI1vw9kNQD1j#Wd7RG*dmXxymg|k)0(fn^uE;uD@uWs(QU+OI4~Hj?H-E>9o}}r
z1A&;0X@7l&euw|>0B_j&%t+|`A1-AWJtX!s%Vo(1nw@jNE^dN3$lRB+_C*R0jn~uJ
z>1c$Pt8@zE$+fBN8uo*<N6uu3E6J4q94b^U^KQ+MRj<W~s!rRk6D~O5E^c11ENmq?
zj;&{=UDW+=RCS1uH8{)t%seS~NKUV$NO5VQCp;ucF({eC4XRj`E#2L3i83#(v^{al
zAAew_9YMKx%?qRf;Ys)WF7<WL${TycmrmT34xg;L{{ZNnJb#xl^(hnqIG178E)xPa
zF_+;*I475B2`Zlf|Cb30DmZ_137d2La@Gn5jb8{GBJsxGm)-URkGVDA6U$DOmuW7f
zy?F%c-Atdlh`v0F*6R9XA2jhD%6yw~$nQ`#@>uZ?c?ZNlNk(WJf?lKmyn?Z2`2}UQ
z9{rJ|7t|KtbciU@nQaqsqZ(`r0JTWO2s5Ncs=qJrT)7G{9di$PzIuOO$cv6PrN0l;
z)6?_r=iN&7%87`?eEPPd7hsey$Ow&Zqg{Ua`|^4BafVu`l0xE={9|Qi<1q4LvDPS&
zc<GSJLHm&SV3BwOLLn2FjqeK^kC`|kU#p3oNiJ)MAkpsa3gRFb)Qx>+CRx2a6t<zn
z+fO287(r52cxxlHalC&^@*${s>apCDP%K{i;vFs;AL!g>`#Ed2sGXfM8R+~0?rRUd
zMK#Oz8=Putmh3aur{Ze}9?J-9ky-T<G?HvMWSdOT8$`CtLwWmnJNxunn7HDEJ{Ei)
z8#nX@j>hKqNbf~PY>WsH=H2A+X7m^na86uFmT)%tx*<I%OQC<G-k_&*NE!CcC<LvA
zwQ41?SoC9mJrx$sM!^j6wQdqHP;2nBn^Y2nli^qKlh%bB`^g#NhM&CNX7W0_f1vOz
z4{vg@*80K4n&M*3<R;Ja_bhXb73i04$=nIzMs^^85ageP_b}3P1Ucc5NfT5x;#6ll
zC8G56Y!b|CoyLC^xN5BgxOgG1T3g{YY~gATZ`U9V!eDyS2WgopOpZksE~qXT>B5DQ
zq&lfeYCg;ixu77|lvcU&Y)e>#jK6&<Zmbwm8uz8#@Z?rYB62Lcovvg&PSWQ3zF6OZ
zBQ~Q{BPXM_uk&4(8N257s=-{rNXi7!wOVBa^mWS1>|=i%Gii*Mc_Lgg6WVARA(qM4
zW69!{T0j>+TBKs0`6hx93#w8IKDH<cJAg1z>MK2^`k`zxC+$T6&Ilc+zey&_3MJzL
z5e*Ha5G%LrTx57fCQ8^HRyFK(m_=c*)J3KUtIb)*y%jDo{?5_Ypg(iLQ1tzI!HFN1
zG#de!80>%YpM{1M<r&U7kN*Y`6GC<O?q6R%|M}D3l)wf5rWF{xF*Q*!EZ_1YGBH}?
zVfk8~(PK!BL7jZ4aj--uG>XsEg>~T5vIDdaWiZugqhDz>qGT|u=@zGy!P7+Epc`Tw
z2x1-O{l4~tj8)qr(Nxi(CxTMe(j%_tRrE+#|IB|=an(PT-REWZvGOxwa!XQnUb-!I
zqW&pwRL@c#qRxh+EWEHIdNhp_qdlQ@P@~8lAmelp9u?Lsb+<|^k8}j{5W->zJjDO=
z_z4C~v}Tr+XpHs0T7t&J-;!3U!q&{%cK_*HS@#83edzu(oI8azBv`lH0RS3IAXq6C
zv|)d<#4FQzHWvbq560fNk}SErv8>{qw~TB8iYR<sIutXlca7MO7c)pH;k~(lL32{j
z7-M|CVaA-OeVGPKJ+yEAw35c5`;o@H+zSJZTHt30^6|_nKpk4`o2uNXIo<?i_PP_x
z+}}zwYlU6c^|?5&OVxXxaTxZQa{<%h75INKc5Ur9^a9#@n^WOF#-*L&Ff(`8?$rI5
z9#J>dfxFM0)7lsGBLz7Iqrcw9qGT2D_yVsu3PU%)??!BLwDd+-8YD4-=^9fRX6;eB
zRuI_2SIRC>_#1gz9;M8p+_{!sTFx0tvd9v&71}9TlGMw7MT4}T`#Bwi!`Lky<XC^#
zyN;^`_8Qv7*oU!8zz|hgKk{qea@BfoPh*jamqJg~(VC@ITn{HDY@){;rP-=1xbC~0
zmj2&&QXaa|mGibf$<6`~#S&hGQ_G=Ku3m!^%CE4kmc-T%zV`)C`|Z%=ugv57R_qO)
zAkXjR8v7aZcLraq&2qsgD~w~u0k(fdTfa@1wkBCJ7f5pA-D8jW3IDe#(_qmLH>*8l
zj6wS1jagfA9?x$o<pmF>6lAq0_G|m{%#;%)Po=t2FUAOMK_~??j`b^xPaci=9C)Pq
zOZK(iux#`DW2b8~*E7u7QYOx)j$W{nt>%5!_#|6}PIvOW-eY^+xR&i=lWTvKg0r!)
zc2QUIE?EKFaF4Az=_R@@b?$=uExNY8HhqV0NZN{kuVzYIlRWsY_okZ_rTEifbEXrj
z167N9-swAZaj~iq#;-`?P7<56YtNuNBTOOXvX(p6gOp#+Bpt@!#3n&IP2H!B{I|<A
zxANbX%jW#|pXv9dMsn*SOI&{r0W(CkY9Jd=RovSf$PM38Gv9SZ^99{y1F~<m-k<?G
zLA*}0Z}yeWU_FTm$#?xq_xAR)krd@`$@~G}btcFi{5SDkM<+n4&EC}Q+xOaI9Tv<D
z88|_FjT5`3y&uU|!@TRM_I|V?f|+Nxmig<Y=8Ui!{@*TtYr9-qEi8X};op11Tmvee
zbpP`W^W2u!28|p1sxAFnHOq4wUn6!d#EV?z_JI&AA)^4TNs#LUA<83sqX(moOgR=1
zxZi#E!EdUG<}MP{x{=I`1qe@AW!7)S7o9qzPnw}c2(vzCHg_$~VO8__9;>~nSn>IV
z*w^oyC))aYbUMSvbz*<r`Pk-kRaBn4I(xGP`^*WuaH<DUt3Ow1cYT%HL0;3tk@4xl
zz<qxPXpIf?E3N6z15~JQIDQ7gV7!^K;-H0#zsjKsWt=efnnM+BhjiCA=8@jQOENL<
z!lqW4`1&2nxd4UbBz;!tDxuXunG*?*m#Kw4LE!o+scqk)*=T=dL6<u6doGz%QX(H}
z9qwD0sCmj2M+<@ylD_1O$Rc38wvQJ?BYcE;MLNT4x--2p2R}Xa8!*Fip40_1tuCML
z>Xq6lJC=^czEI`K<E!$Ms2WCksw92ka`P_4cXP%Qbg7N2=?qs?ebT{adh7`<1%V2)
z!hj`oULj8!8Y_RO#Ok_zAz>Zh)ax3B4OlW`Xqh11PGKL^rkj7A!ZKq=WM5ENIujg}
zo>vTt8ZrrVIIE7)|Mi83A@QhrGfAho;x>BdF6cw`EYE%S{Y9TTjn3}iG;8Ndr`TV_
zY`UEd`cCFIZt`mDbA@92k|ow8n(N6>*O$&?fern5D<N*``s4J~!Vf9BWxUr(`@qB%
zUiKQw2aj1QS#;F4ox7@y?DDB7p$+55n^UgIfY6w|jsT0kygPzguo(6pQ{W+}0hVFD
zU-ZX6{ZLlG_6mzD%&_7QKU^D@&v`*K9=GiRFUa%fe@v+U3tl@zAD1!pDHNB%DL51X
zGM6Cw3MiNC6DpN|2!B8Q_v7DBJR!@O)Bdl&jZg0%Pk%g(s4L`5gINCe`;(uL0+_J4
zfO1T7T>kuJ`Ss~PBiNQR7Kk%kYo$8Q7|FDOMQg`|pNgE3gtb!X!g%NBNC7i~Hh}c=
zr{Ap``Sw#mo>nq&W`xdF<e5DColei1drtW+{9e+;{pHnv&!C@qr6-bv(q#k^UZ#xb
zwE9YSmmri`sVg1+3bTI9q!kx0V~;>5rA&uVhM2!1<B}M{=`E)G5B&=a=|+elKVBW5
zvfl$%V$sSnK{9LE4}pLdolPNTDAF(~OopOFBKDFPmWhQ=2@@O{kG112#Ik2VE?*M=
z{f2J?hc8)wCPmgRN;IHj^Ng1i^89Z@7HActYlTMDmR&69!Wj;pEmJ*&Anju%E;N(I
z7Ksyd>F$%Z;Y@-aA#~E_m9#)=W`T@%Pp%)EliScKJwf>=(AkZ&rcRE?C^uq7fW1hk
zf;7k!;%bG9PB=Ja9UUtw?{MR|rzbdTOitH3+-N6%E2oBvDR6(0Js`Kta6RtQ+QU>N
z#RgiBWg8vEh!VUNjXvLP9Rifsm=#jhI<~a_e7uT-WNOc+-LugOFc*xd`HJlf5Ry%N
zcbfr*YDDIi$!N(an^>(l&SaEBrlt6cO=j4oF_~TxkDH2j_8=cH7tSo;3ZY^yqh@&N
z$2|^zHy34ne3eJqi=>=UQ0_=tZ&2+*R$gNHCqV8-zTL1Z&qpN}<)eTuHZN|<ElW3y
z=r()@wKFTn`6#v(JIdQMaadYrdCiXwO;@}{fXc~!OeaJ6VAA+Gjlb@Vf1OT2HG|yA
z42}sy0pm>V!<44+%hxGJ&8&5Ee}0@730T8_1C&k2Zxw<iekr?9Wrs-Ka#F}jN`gU`
zLbjHv{!Y%6r2G>7{mPX5Jt@2>RaC5nOZdCi`bkK;Q4f;IdP4SUZm9(m#j~QR7DV;y
zONy=NpG1<Jynr}E-t>gn`YfIP>%2aOrt)-72)<^Bv%)Z;pVws4buCFJYbP2rVv*i|
zh<Jleg=))IUugR1BAgjZ-k8P2RhJ4;21<gTlm)LzshseT5sG=~a2YAhLauwnD2B>f
z^mz(hR%Gj2jn+u85<6w`!4iA1s~t_`a!B9iI_1qA36%l(YvSIuM2VFWQl=$LiLG9%
zAZu0DD6j0OGT*Wupx)wlm}%u6v^c(hUm-*jnez&{qU>Wtj@VQ(HE+-Yo9@&l&WHP3
zlilU_OneTVEg9<8kOZbg5Y0`K6+ofgi50hoOjz335R<_*VAR`8dA1I_=0&p8$hVwT
zXC38PLE3c<M`bNng7sd?%ubHc%SpQIz><ydhOWl4`Gy_|a>})=3M=8Jj47{wWgy+S
z@&MO-OOi+h0;kY1;%ghqlV?rbcouQAx24nu4Ox=SPg$bbilV<wF%;%}LtxUDyfEvG
z3@z1GXQa$ttlDUPFYw1y9q8odY9;1%Bf?p0P48FJ{yQ3!aC%6C3eK|R+8L*b&o?hy
zxOtyHS(j^0l8`jGVIP&3S(*udQsdv!$lwoE=A?-+j$ph{OfG>x&cxlt+Qz3Ez#Zjo
zC1cs<Lea$53rD>Zz0L)#yjK6Qp?kssQ_j_5`6Sdz?`fB~>p96R`J)U!#O(NH_V~JL
zW8kqVcw(R@>_RzZjI0l~`+Y&B6^`wG)Rs91x7+=xAk0-^EfMp`w{#?buu+euT2A=A
zY>BeIO3Hc>Bt|qN+}CwUb?5+b5>+4Vpn%qt2jd8NqkWQ_9K+jUW?AQWG~_A8*Z5!P
z$2&aU?`b#g2rzH&prSf`#reHh5e{d3vhd#aB5sO*SN1YY9x3}CV$Hg0x2?n-TXU-2
zZrcI064#+t%XiR96q(q6+k(=A=|;ZY!H<4d6t&BcYxUq&ijUjSXwnp1LEGe7t%c8@
z*X03oSlQb{P=#bDaD)nCZXv=_9${AW9<2S8J>j8_uLHPW41E{9M@umU@IW#Lh;6P`
zXoPvbz29*^m9Rf!$0IGHM84t7*RqY1T~%2!mQL^%FL5_Pr~IjZz>FT$Eg{(=+b+|r
z<NGyPOR^iGj+Au|apli#(s9w`>nUebc^)DU%JI|n(>`gyX5io$Ux&Akv?<g~fjyGX
zJ;acoc8AxW&Y1?tJ2}&c$IWBJFZ6R0A3-sta?b9w$hIL~Ym5x~Q!1w1zYUhxeO!V!
zKQ58y@#7MiKM=8h`IKT_x21de&?MKA1t;BMu)-;5h0K1h?LD_OTV3tDA?Y1)Ulfrd
z<HgUl;`t1WHK(Be45zESkct>X8pcU?iY=_T3bT%6_t}oRRId_Ze33@(Ak_DA2O;5_
z^Qb!VK=&X~x6?hCe0X&GjH~M&Y<CVC3ZUc0L<842<*vbhmfo*>TS;2r1k<us=#!2@
zhCb8I!QRw%HzCoN>}TvKj1XHc&N*-ciwX5G`d=bhF=vdjy+@}^UPGh9Km6;TvE?S^
zd;u~-p<VEgyx`^zE#L3lqgxVpIr0=2EHZ7ec;9;NS9}4c$QGK_#mi{l0nFXNq~-EG
zj`b^IYdf5Oyk75{4?3dK*&Ocsu9$J0V0t){)KAs}rEa^o=E!dNaG*&EV(B-GOoAC0
zptPelC&Cw|>B4w#Lf4lY!_nCfLHJX6YtD?8-C@3Ixb~h*S`D0h`})q(G0RJWCf?K0
zX1XLlJV6dl(3%4iw*lhcyBGq{%%Qtk4Bd6)G@dsq1$7ADK|fSK?G;batj)%k7~-xI
z_S?tSw$ptX)2|L}y1#!s{U7=0UTBvw^(hpS(L)oL!yGEYf0DBM?e6D|$jS;HEIrb#
ze|vwo`LyX(*G8jVf!#lC=z~_km6uFP!@*eg`49H6x$VW)$pnr`KSr?6kF2BolD!uM
zdQVtu%*=ZraB9u04P4Qu<UM^OghA`{5ihh}qpu{)tbBL$uqsgtjDzVV8i5!3P@_L}
zl$@k-O?sA}f2jV9g6dbv)^f12mn_9hugdK-w+?!4wa^0z`;;qWsk~0l80Sm%lM=YL
zp`Q_8RX>WZlMByNo*7iqxTFvHE#7aEzJ>_+4)3#m=fi(C_S$i++{oC^B^sM`C!1t`
zKjayR4(S`k!K}sSFwR!=x`A+kt!R-$@t_#cL~+&`f901K_+vh+r1*w38^yECtWELB
ziH!C<$c40%UOVo#R#2TX``71kr#URv+eK29=2FRCDya1pqCD5$E$khf{X%go<jCxw
zOEhOzw9wCG{!m+u$34>Pfwp!M%|5U2oZVkg<vYn|s(i0Uoj+UXvyb;pH8F}!KT)R{
zrPZnUf6oc5*;8ewy&o3&J0w!tb1z2L<W8b-(|%Gj5^_x{nJP0a*X&5o-$@&yrOe&I
z)iY)L6^>STGSa@?6zd`GkT!8T*h!oPZqlV*Y|yTp_15g6bK_)4si2sB_$1@0aeS=7
z@SsM9dM0qO`}$mu)Z{Rt)9z!+cSs(nYv%ase;s75Oy|D-Ff?Z#xbN$rxqy*XzLVH{
zIaVI>=JiEht~+*LN5^`@NiM?3(|#xMZC4j<0Z(uC+3G^~o76Ju()!6<Wn^wIa|MbQ
z@<~UPC@OndK{{8BqR6Qr$$3Fi&o!Sle9t$}T566xZQX1s3BqnGJ+;B6D2lVoAV`<@
zfBDcJ6TDEP0sCS^r5>7uH&v#Zj=#kj+BRvZGk20vit;fp3BUN-6ty^8mTPR~($i5O
zemyiHxn7;gFLsV%e;F6bY&&qd$z9oRvdwWJhs^03E0Rar<KF?)UM7#MrAu5IN7BZu
z$ES>|8kdABtp%6Hdr3tbFR#kVSx~3uf97=VfWkspSCcNfn!HwRv-O++0nTNq#ma+?
zKsQs2uOv8!JgRd{Yi8LL{RH;Z+)VbLvPnwG!@?vg2oV=Lr1ReE3r1YY^O}+MS+`ta
z!vssiyJ|tM#KMyTdm;7&4tHE(z9;si0-YUT;BM4O9Ed%2lFxJ{y=50Ijd8e?e*s+B
zy)vg@DTB?nSOn!ou+%;^tb?0YWeEltc|PtWz9D!@Y0Hf@-x!|iX8l5Vs+0Cr;i+Ep
zG4NC;^$a}K+dDY>@nmMFPIyH2&n12tJT=ZvA-Nm6$Y%oIc0(cqaOdFv0e46n@IErs
zoy6}9cy!vFK|DI{fh$UC$AcQ#f8=)(2ZFtw<TJ;rUXQN>_DW^3C$`@J?3G|RS}`(G
zbBR9`I0Zp*&?DP{PU73%3&2Z`*N?Clto4N0WcDp)uxOa24)YU3z7H%i<9HRkVgE8B
z0|5@!#*?ek0D+vfdH46@!%z3W<2jZ+{zD_MLB&=Ef@e?Z6}uI0gH-Hee}4B|VoU4s
z>>JQf`tu$K3CK_eR>D2oXzB2rY<xrgNYsD_J1u21IJU(O8S4ZMspvowE49q`u_YF*
zrYWLLRSkM0DS0ft;`lO(Ug>Jz3oDNLlWiW@=C=AXDEt@O0b$qNiQ5%81+7a9NnwF?
zLEGYv;4ryPP+LPQQF808e@4ei&mCqCH6wYpYHXo21{V^_DiR8*+<W>72XwY3m!!1B
z1~=e1%Uw)L$X-pJou2QX@~}r7^r5*YF<Z<Ft}(B-XdSl@Wks|DlVc6U%It-^tncfc
zGFJ^43Nu)e7F?NmZHaY4QH5L9sF+OMR$Fgh>>dDauo|8BPrpewe|{0zAVGD@32d)e
zAaUV0ot#NENJD7xOtX^_3-=j&O<)dk`3$eb`qiMiX?#Wh4yn?`|MDeWgLG%U4y(@8
z7h=_!`vg{<Eh#$sgihu$1xt4`DXW-v<v4?9uV}M^XJ@T0#Iv)%GoEedcy>zJF1jfd
z*6_G9P&XYrGQfLCTIOd!x2SfgloPN`G=1V~QmAfz#ax)p1v$*ELvXabYSFtg;(15U
zpy(v81Ez~!<xeVE4_&nKSJlBYO|rBgmyq6Cov2H4txQa7KGli8;&0vW^Zh@+0at&Q
z!G9<d0yZ(1u@D3k0XCO$b1o;BBP1%9e}oNBkN<r8`;KQ~Kb^E&JX1n&Go#d4#Sm_x
zw9yeEJ>Uo5C&HD<4<gbgJqn4@I7i`BsRAPkXKI@ton|Qf)Gm#s)rmwVv~yzNOyT3C
zmdjsA^m_Em>sf&tEm-6*v4L`^l&C`zn3m~XWST2Afl8m|C;mDCR!r70<Hd+ce?=%D
z4<bt<@%JD^kcKXF1bjz2szq6g&#C-!q4Wxz%w1YlVVHumj_Al$S)~%Z%KI@>VUzkv
zud@0D8Y2hpHu}y{68R6ip&vz1;YfwakWNKnc2D+Z)W1k~+^>~JsGv#J1S6HlU~UoK
zQ-))y9u#8?+7LMcLeDYF0}|#Se<>xRq{=S1LQ{ovY%n4}uI)KI&$UcQ2HGzM9f;S(
zEDv0Fv#+xp2UE+Df>|8uF(t{0Os~%p7DNVoWg*SgY}jb#Nqd2=MtlHIJ*K3oF7Z^o
z!m;NsYDD-=-o~3147mFU-a1Up3-jdRB9B2651Fsy;OW#%pKs$CJs~*xe-S))n35M!
zi6830;`jeE&q$HL&kx|a!<1;IF+*hEB$}_%_NvgYIbw2p8u|}4^vmLMfhZd_p!^=4
zYfAwjJMHMk$x2ga`AobnBjOaYqsYR!aE}C`)2a#~o_d3h0mrC42|q?R4EeX15aiGk
zMXQPV@E$@8GgFwFq!!gdf2thR6CUbPOQ1w)6}*I6Q)LpvRG?1mNyw8;fot&0=~u=T
zc0v@|l&X-68Pt71G6RJ+^6-?ya#}^hc7jS%hE+hU>kfIOQN#1P0P)bM>R=biKMx5&
zk&v$eXQc(kG4HD@EYgeVSV<U;M0;@u6}S7QcZ$0`;Ybn5s<sQNf371soA!rSh{X*(
zWjTaoij(vV3LHnAb~5G){UZ9qdVWa122)l3(FPp}-*HE;Pps!|NNSv;MiCb;6g@9*
zCLWdHx#f{6&xxWigIAiHwgStz7!mR#l}^!|B!#3^M7~c{n)-?OssnM9Z5Mpg;#F#p
ztZJ%t8VOa^_arZ}f42RU0#*-~H7VOeH%FB;gxIGl`g9sRY=w&ZUfyZ~S|JYBu1uXf
z^+nCAU|KunS$YOFmzG5fvieFTCe=tDt#ivr)tMjwB{y=;s9~nAA=O9}xZA|;3e+u^
zHw39hR~L}PzEv2Pg+6&U^iljhy<$@emAc{%->+N?tn7G`e`{+nCV3+7983zn8<|N_
z24OK5(zMo&jf!J(Lc;WwQmqwLGp$&MPxeP9dV!Xjtzod`I)pH2k!mC?>+;LnN}Z*}
zn#Y*s>(AYv3woy4CAuecoYG@L^xq{8Gm!X)zS0_H5d@RkI1^my?$aONzy0m&Z@_Tb
zf5KG0s8DU9f8+X@UnSBhu&(-%-wi{Vc%*@OWD|*Xn8u?osSE4C*JT$uk?5}5=#Map
z*_^|Cjx3hIq_bgSK|qEI4FO^s&G>%M!Aey{5s9v3gtq`CuMMxbzN`wb;i2zkDz5tX
zvir8|K4+<FV~h<E$O)<O(^!XlH<2nZj0qB`WXn+4e{i8d6{&cODz-67`E64fZY4T#
zB^<Mu`pTtZt2~zx!gm=$SOr2Es-z9$8*&(U#gVF{nrYldshCB%po&74Zjx~O{`F^G
z_Z?TglFtO^A=E5Nn+3z2VW_|eDz+}MD%hxS!t6Y`7<X}ZGRK94qXctmTm5=O_IV++
zB&Hr3e^;L`mIMIPxMk-hF*Z}01f-Kpizb5YA}>rYvx^M$qf+}8qhrWO`CoJ5*XCUn
zr$I)HSmC@T^nhK3o*XR3$_wl&dk8f`{&pg{kE=Ow$sERN>4_BAeA>*CV3^}><gE*|
zIgd_N3v@M4<&1>Bbp}2w?<uG0&0QPL!Cer}e{+O+#~O&@&CH>xb~1<GmwsA~R}Um$
zU2ys!b?E8KX2XwQ{62|TIbZspwP40k&$B9|Z}YBFo#(RfOSU)U6d}veQjo$juOt16
zIM+7J6G1=IwKQKMG+wx!c|}(IO^EN5@(t&9`b;55S#RJeN}RIbQZ)DFe3}^wy_!LX
zfApMB({oJk7`LuJlOSDgnDoM2Dz1{Em&QnM>SEd$X}xPy$)Ne+<8^pLQ}}a5CP@5D
z-aq`1vJbI<7fe{c{PnNp+wSuSwzk%I^!4GN=E<*iBVY2toabm7+?bFD64L;5y;`A5
z!u-axPU17s#$%F>fT5>XFvM>SaT%e`e_UZYgGtSZMU(xAdpuOcG2dR(jD@DbXd$^n
zWU3aN>(isO8!8Y~{zW>w5lSa|B|wP}kj31Mg9prM+&EjV=t%k+4mtyup(z)>OLH}S
zLORSs?TB%3w{<pZzg4iJZ)VM;PZ6vEEugcfEOFs2(@J+e_G0aVl+`C%J7sH~e@W$-
zB@pvo2)SX=q;7=nP@%*brd>U46f*lZMOsdZTIj|OYUb4n+9ti|?+l0R;pNNC`#0bN
zHCWaa5c3|75TEaWXqZPAGeXpJ9n*km+&CE}zo|l2Gb7xV3UzLu9$o0ZrLT%w-MSH2
zen>G=v&CfDwtjZx@_HvzL^h!2f4A?L*3rI4h%%p9%ynNF$M6|fZEc%T%ox9E`7Zku
z`7ztpQ*&<H`q_7G+kpC<kTK(gmH6&)d|Hw+H26JrQD13J9#@!+!vOTW`-?qxRYt@e
z4zb9mBVh1{D~;_YD}Fszi>0`=Bik=?tWGn0hRL_`>k)`oBflH^>m9?KfB1fWB}eb9
zM}Fb7%51Izin*K=aM{*LEc>(_g?Nt{gt(`FxgGRSFEj-GD-EHi;G8A&v+ps50WW{w
z7W&!u8pDA5j5XZNb2Wzn?=P~4i*~RY#Gr9^SOiv$){8D9-aJC?j*r*YfQY;uA)dE}
zdb{xOD=ng@;+#eFv)@ZZe?@3I&W|fEty|7lz?*e<u80Nq%-LF-)vkmc^iYTMmhvxR
z>PEi2Q!VETAJBe0=+#ZoLtoRRz^?e!_*eLxWnLTFxQ!i!l|EMdsXKH#Ud@2}1#GBa
zEWP}II+Jl=UDRo1<u;FE4lw?%$?wkAnX8)Z21<=#Tz1zq(Dz+Pe<|>fq!cR*a<y)S
z5CTGIq;2NKHz8Mwfq*mE8@pSWXWg@N9pmsxm6&#nCFnC9S_x)#4Z&R9a5vr2gq=e?
zb7d>hTDs0@&8+L(eknJ1<M)?2FSfh2&=>9v8m$TI!L=s$5LQPNW>0#uK>D?&mq)%g
z0CshFVlM2)ueh3ZT9G}PDTnbkRlPz0?T%79kJ-MbL{FQ#ixwOARW7|fMb2@^?WufL
zbHVMY3vbU{o?vg@p0XJ3gq?^qd#NphlpbQ9IxJM3L^qBP;)&Gr^jho7+wOlWPc(m*
zG4&}F0x>a{0X7070yj37y)G&<e_fAUw+(&wuh3)qq7gqtQb7<PiR&~)3mCQwv^D&&
zaR!N7cQ;s%f%M<^P_wf;)0wT;>uV=9Y!C0fx)LQG9+J}Somy$7>(Ig@e|sV0(tDF4
z>?wHBG4Dj_+*3e9^I}HDQtnNQb%H>>c=!VdQDR8m;WB6nQmCn=<b}usfA3=~rg=*t
zb@bnv3JH{fgnPZUd^#Xe_Rr|VR4@QHNS^Pq74ng4a1@G|5_%p}d!fRuXD@;oQlD=O
zNH(8JVQH9#;g+C<dERlx2+|5Gr6v-E`|Lx}o6SC?Lz?4@6^@p@Dc%@pV(t(;6%jAe
zI92algnBl^f5MiNP^__Ve@<Jqh|YD_UZjug7>fl>vKPf0VcCmDkujG?M^Tzn>{^tq
z6$^qeSH1V5A{|3OSrwnn@Kh<gb4b|)5}w-Z7}Hbd9M+_0SkBha@pdp+3YJVo4NZ6|
znhPkyFowl;QPCd;-}dKt&0rM^yy`k{r4*^2qxYCl%F8id2mqo@f1|(_Mm*aAWZV-^
z2^Sf&7d+j<Fj_(=--ulE`4pqs;8Zyb5}q0W0Yy@~;Noqng0{{rhrsk)+Q1NK=fXK}
zETJow%cd{^HCpk+5~L`e54r(2z6IV$*u#-<5ABBL@(hQRQ#L#ZB6<KxN-(9yFmP~c
zY>i83-ZSVE6J`c~e+r9S!PZMex(L<=9<_byWwe+YL>a(zjsZ6lr3cTaC%5us>&jC|
zSBCK7!53ewp8fmh_3Eqh^NY>u>D%u&dG)7P=l?9LuP-itSYJL*GH5SWe^`CD`sO)p
zFP^MlZl1#+mD5wY)P~V8Y;ACUGp?rn#p<hq?w_u{y?A!9e|lIx6PE$U!BhaI5guKC
z_a*)uDsg&TxOSndamB1Q=j(BS{wfsL<32c;$Lg!=?q6<RU7SOqzdZRa|9$q;X7lr#
z&sVD-Uu}MR`+a?R@vqe%-@blz_V8?Twt9B<{p<Cc`j6Mp@xl8m$&IT5qF{j9VJMbf
zeWI|{HvHo$fBEn^hL~nv{q=8uFP^Zba{l)9>lY^*Q1cpaW(%)9{0=Q8dQF;qcS}~S
z>6kYy%j%np*B6&he?EJ;E)#=g_4sVFSzn$P-d_Ixm(90NH)or5q3w-F7v~#{@d#W@
zEhdy#j}RG%Krr(H&Q7edy?|d*1}0p5`Q_^I<;BaVf9uWj)#HbcR?pVIY+mfy^6}Y^
zYbbQSS)Xs-WK^-)#gp}$i?^3A*Kf9AC~y6F{llxXuP=U?22a=W!7uRsCEiTGn!0bX
zG2glx9?qBHZtHWjfX^|tI>5>B^GX>5h$|&el2?~F@;L-aw~Q(w%0A2_PWasWI=;ZH
zGQ(5Ff26o;^_DRzAx=j+1fmUx_xuY&LSSvzIwWvtA2<f!*eS@;3CNmM>GCn)b%3Y?
ztbG8yn2+Ga?LP2g-GkToa8wC=A0xt`3UZ1J>=hvwYg-X3j|>VpF`FW+Gz2?@d5;5w
zGTlx;i&<#T4hQCXZ=?9JQTpADvNvmc2aJDne^ZxZJj%WBD0_J11@O3CKjhzys=~o3
zv?}r=CaTi<36^8TSC$x%2C{J^*$0RedPhLqYj6_~^$x8V9GHpoGG;ztnvY}iacpzH
zkIj{CY%aUBQf!H|l4{mh<VZ8Hz=(84S;qA$c4T^E*hvL-L5D~ot_uV>WVQn(Od2_>
zf3(bFP=dlXz1oQghawJz96P&7QyESCFvLmRCBz-Hx`G$CYM&s)9l&c3h3^ZmE=Ta{
zbRT$izk%0Q$!$q|CF}sFwiA3O#7<a(h!_D8sV+i7HT;sJ<QZAz_NvOqK+w!|1}gX=
z4OPCFwk6=W)k5J4xzj=j*Xmj92VCL=e;hHcbZXiZ`JM<3Tfiw;e5RZ`EkrUq;9Ipw
zmqg|V6KSCELy#B?EQNy{a$m}7Zsw$W1WIFpQ=#xa(gCHq$%y7HmjT-L1r|cI*W5MY
zhTpL@L(A)|fBW%5Fp_it$Pu|fW<kUI0@etw_*ZBEk-1fb<eMUQwonvzIaY#Hf6c4`
z3N`@ca9KbJxP`)%F));gBB1U8h|nNF<rtTagfZBF9;n_O1^Z!`?);80z1LvJkP4T~
zeV045RFn&i@6eKE)G_$~rf&7)M9%M($hqCHZf{paYCHP19sSyler-p;wxeH*1&UmF
zw|T$W;9-X0<kpQPQdBAK&ZxWDf3m)mC0JJjWv!Nu<kNEx+_i$(OCMZE^G~J4RVoax
zt+{zh!bpX|!aksld@3zERWSMFSxSd@uM%_ny85TH$W<IfvR{#AK?@KsYGKBj#5kb^
z5<(<Za1XRJM+Kpake=7v?cHf1(26;3)#4+?L9V%xPpF90K>OsV+$9fBf2z<iEX>4*
zwQzG9px-_Yh}Y>Z`Ki-gcMVUpM0`LFZFgze@q5tnx6Qs+o$s@6@EzDU*X@9PV{g~c
z#ucIZj#+(|*!u2K>=-&|0n&wi@|Yt@*WqN+M$PkrTS+(TE{_(Yq4a6+c(w(c3;CVo
z0Sa(6sb92WoIDXils`pxe}bHbqP3PMS*^(usjeW^4IpEug|=;ebE_7}>5%m!&lYL}
zB!HCf!?n;Xg!4>prv*qj&xdYN0@YD~rw1}#QSAfu$r5CT^^aaq(AGR#;>c_Z=!k~Q
z@5FvH=Q$KIT_l-UHT*1r3c>oLu)o#<HV51Bty*{^r27hEpkM)je;e5W-G#g(&)jS9
z4yHTNlJY#N>#xRNt4E3_dbJ<<2;PGWzYnvG<GgaT`!L(s+aQbu>T*T?kslK%fPiWo
z`61Rpfes$0%HclXp%PdUOsV1Ou<RIjihkobQgu{-J@i-GomtS*$J>I#?_uez#I7y9
zJ#AiQD_E&j$}KtTf71;Wm47?lfC+-XgMmi?BJ8;_cxdYQ6^|x9WM|EY{@X(Nq0K!k
zM6ze0tNJZJIk*yJC+JS-JK?&q?}S*+z^0>SvP09g19`|ugCAM=&3c}Y!_zA*ag>R|
z#`8QC-Oa5vs`=>>MIEPh*1U6`E(`(LI4Wa`zMK-E#!3JYfAW}`-)z7uQSaEP>WYVo
zGPno1muM4RPzzL=oV!_3BR{1<p$W_$f6W0Enp)T9%7Ln4&(l6!2hHlX-zbq_IF*FL
z=SGK<=T>WemxBUpK`nqWh2LF)0%k{{cM24iO91KtZs8L=#FyA*95eG(L&@E@n~aoo
zztwhv_@EflE-nii$U*}t8!eFiQRZ2QhM8vt;$N2gl3TwC(BNBrvsXSi|1fI*E5iN<
z>6aFlAzwWc7d19AHXtw{Z(?c+JUlN<Wo~D5Xdp8&m(eXLCtB59O>Z-|5x)CZsJ8~_
z%W(Kjz%U>wc7fhn_*V2(G&$wa{QpCT<d7V4DS0K!>6?QrX_p)h=i`}&<jV5!%RwIg
z`gD4K`tePhgBQVqSBLkX4}do${soB_mozylAb+q0;2ntgf|qaJ0AA79M=Nai_F`iC
z9;n~BT-T7=VE-6>Y7clHP&S-i{`LO1CrQdm5ncM2-sW$w?@#}InpM}%3kc!x?_ZzN
z4WmKmpbi?82+ALR`_JL$r@zl)+h7v=ls{G;e*5JxUHJ!DFG{qYbk5qrdLYrH=HOfu
zI)B|N)-yjzX>oSu4PQ8WgSJvI%<{*#085ERq95!m(W3n<+RFMO(Mo#9kxWQm@O3KJ
zQ(NQ9^`wD3myvf&^}~4@FO2q&oaZRf#v&z2%#s7?ZjzGqPu6|7g1R@!4n&lGmJFj_
zy|okI)Z9AT^2BUs^`ZaC;}2K(_|;b034h~58y{T*vU?@HTl#JxdG_U-lR1wfwQ|MZ
z+4tDgzM^#_*+vyRrTaMx@g18>_-V?eTlB_=bNb+v6&7OCJLNf~O#OspD?rAB10$?!
zf+eN@O8DnP)kY<p(GLvONn&zIcAjP~^0$Yp$1C7w=}NA{)dvyPMwc8CsE0{<IDZyY
zj$op~M(p41nc5HYc!B3Sm1~HCE_;0#s^Cz<t4E^eBr#=jisLlhnmzKZ=G$=hl*YW>
ziP4IP?wH2C;6x>jFcvmiE&~ej;I-(eUvJh+86~hd&m)0Z>t&pXSRQA8<nnSi4q6+9
zca9@#PW^@}+!u)}q`)<?V{Zk8M}LBYM(SX~BO`8<n0;s|*?<!8t$dcmcZ5$y1NxKa
zF<coy!U*SK9wjAbVput9LyPAW$F|Ih*O3MQxutrhJBnIGdY#Ml{JyD8xpl1lOH@+!
z$XRN$sBd>X(}<!yS1~temEwN#eX&fAaLa8~uUzpt$`@ca9+G~Se@)6&tbbi|@6!cT
zIp6GCqk}MJ6O#o2noVk6T(lfK-g2N&BU~<6<wOYaM$3U#0+ieIKCKY<F3VM>H;s@`
zMcNDHlZx%6QyKfmw*-|UdT+ys8i3P-q1T`P@&k1->n=)Q+KgTP^uw(T>Y{Bcjv=#(
zjRpLEP@r}|0Xz;0=-XH}xPKmZCJOc8MQ2(jJfhf`XoLjyz+y8=%;{vW(i{5_dNYmt
z0lk^E{5pCwjRf{8y_x;-IeKG)#!NT+=*@8T>*$R^U<B<R8F8b;FQPZoDBlgev3~%1
zGu!SqdNcPvT|kxl>CGHL-w(Y(u>!omlitj|Pb<Xz6up^7*g<b*oqzVx8{TUty&1YZ
zp57Q8g@W*1@f-Wc<2StOd<DM&FCMwiHp+9Ccul)=C#MN%!*|mD(ym+OKfLWVK9h7K
zCVV%qkWOd=pXU`1@URY8CDbE*JW2c_J~555<`deMqpox`waWbB%kQ-MYg_>0#m!9j
z)i9m5{MFV@r$ye~UVp!|i2|hB_`2a4?839J;7nM)BG#4rtd>w9ouXc|+Wl4E;F`w8
zI7`1$^>M8wl-$nd{-Z1HwJ^?JuDk!2oY`3=Z;q-xRdRK9{LzuRMy!JXD=TViYXvc+
z>5<>5y4uQGTt`}4ZRVStBRl0ebs6nG9jaD@%?GTu9z(YFNq>TCeQwaU`u<YSdZ1ge
zB<nlQKvONH8fxjYv24y~>J7WcbuBKy<r%!F?Ds&Ht(6-n5^K~PlLWJmU#2+OcwPme
z*a1r8o}f==#W~Ak6sW^<c}yZ}<!LV3@Y$KdQMK0doO8Q8Rw#4!^PFMm%WJNzUTSx%
z&U6sJ@j%fErGLXSJIpq$hPx(f_3*+t+7Q82#%Av4LFy(R^P%K7ogYR??Rl)(SR-sB
zHB^bW#tX=`_rVF(pOIieg`AE+&xUS-a`hvFQ!9GzBTx8L+-YPD_7)&4MZ0tkiv3Q4
z3Hns0@2Xt4D|lN=au1DoEu{k)5a;DWpl65rb1g65fPW6P8*IIm1M5V2jS7?dXheHy
z_Iu%!<V<i%_BCr~OCykk&8c1w-}B7vlo)9w;E-!dxbYdm#Tk`51bs!$w&yj-B?&h|
z?#^*P0l5auGb>Up!s)pg9@~8u)sudu#YceaGeoNMFvAgSRz6oz3AU59s*44CIl|<J
z<~TaLOMmE``_3Ry`hSSZrPW8MT(rB0$|cQ64_ilS(l@iZitnYxrCM{#tQOGMt-6K2
z#^N^a^HwP6YyNWyX=Ub+rYpKzwD5tKhECbQwPDR>`c`q}*xg(6jpI%-wX$&-n!7N`
zGZn*ztCoyXX>~K3?$nN2U2za;YdJw4=D~7~u78}^jNC8|pTT%%QLOtIE7#JUMurAW
z!=RlTJOCZfHjh57>6&s{DBGU~;WLirxm%_AJ5~abbyMvMG(QdGt0Hjrb<80f`;w*f
zMcqoD8px7+a_DPXpSN+mX4x!NFKRmVyNHja@>VD+GJ2(;yF01{^^{b}XWy2)+_y!q
z>VH4>6lg<t(<yPce@oS9K7h2!D$1j8O@~s?mQjJrYAyTm<p2zECzzL(m*9`?HuxhA
zHxNk6gR=CNEpj@@D^74JR$SdK&fsQcjhy^x3u-f><w&bGb8XJIpqewhkTtuNyeXPs
z{8ct^X8zsiSQlRfV@KPJv&~j7vvoC>+kcgsTbf06yE`tis$4y0>oQVw6xwvgzbBJT
z<=#q#JJ^^~af3_bCpFkOf7_i`rFGM4s0wE;e^^ohAGIyKYLuJ?Eqs&h`+-c(0H8yv
z^`?rBAf%78t>y89UWn{E%)Po6l6;gkHGSbp^O>@wK0Z4$SJhS|tsa#ASL;FRw|}gS
z^Z;)_kiV4e1q!3Q&{6fRB}AY+AP%jSUL;R<7BMHaU0D<FQ*N_^#*&cjqD?XaxgzhH
zZd{?15)xIORIlMj0C>%r%4lIt+(>wR(YDU=TncSnf|Nk3n$D%tE*`<}b3$8|{Zpg{
zL8Xmld<9+iP&ZfmdFD0@wG?=5qxn}nwLO1rVXsYTR1`jJ6v|b`=(Y*fE9S!8bt6jI
z)fP=@yQ1u>F+Rhcw~lkL!>>d5(nI+xyi#v6lG-1~^<rtq*%-BhDLgDYE^cdUvd_8p
z;tP?w`fgS7IBhVNk<a5^L7R)~IFP+!7286#fu*aK4Ncp;8M7kIY)AdJ_I|6nH(P(%
zF1z>TR)=Sv{{9+zb=}z7pV#qWkK(-cztqBn&cXP+N*Z&|m2^LiG1q1ZVK;vn7Pd|D
zzL?;n*|7EQu<;Hz#6W>YT&6@}l#*_B<d6<Nt%u>aJCTkx5|=~G>BoJANPs@R^-1Iq
za91(74P?`kcdh;$`KmDSDVP~^j^BUnZ%Nvn%57TPW)<BxAlMx@w=;1nQ0I}=7B*)Y
z-!6V^8t_dB#--Uv1D{lEC0u~~(p(tFZF3lH16(>^J6nFVHub#C3TKx8bp>SkAy^#?
zwS_Uh39(lBnU3>)=)0fk$jOCa9~9;3F~l-51X$G)*Yd&fc3rNeHCNWU=0|_)*$#x<
zW~0i*-Mh0?K}H#)hAUE3k;C~_J#_ItFM&~1n5|DGRry}`8-UGcSIroC_sqqOF~XIA
zU%2PYO~TS(84@ZWAL8^b58$XiL1240gyC*mF-_mxGkg;;ZPw-A6T5mvRs|WyZhon2
z&kY{5u{q6E6a8|1Jv(mB&QX78=hiYGzvpUsfx)MT?p)YzLI%g=*L5m?yr^171<$iL
z%qIxt{xv7l4oJtFN6}qe{0KmzKF!fE-0i4NGg6}uOk}IniW_DZw@ttM_SVg#w-s@-
z)Ns<kcg})e+VZ0QtMz%{rzO<i)a%#KF|)mV5~9_##~R!U|BJTsehz=*4i`=_A}#?`
z7>+3^sisXC#|oY2X9L#MtsM%_^Wu!)Q9%W`lm~1ws-1Sj^8Xw-7nY+wh#~DppOxs}
z6;~>xhj#pIw<?GSl;y$APlFf)a;v($5Mp#N7cr=tz3l9088pr6$G#y>ePZ8?X{Xj}
zKnhCO{;*2Jqf)mPVlsb%_(3sEShOiP<qF5{-X`ayYx_>F?VAMJBPmx_%(2`F)9Lo{
zaX5x-I(j{sW}NOG=-LNc>E8vY*$>^y_Q=*Yj`r1x1vuYqJf|^Lw>ewtS2vGcwmbD~
zqI^5DcAoom?ZE;d?F?#+p(b_1fWpNcRxgEP&(qW(uxy)gSY8cp{}BSczrH{H4=c_1
zm+>Dv69P3hm!S?f6qgR+Fenw(TFs8!HVnS+Q}_i|5k>t1!2mn6m#;-CDSuzCA1|No
zs2eA|0jHXe?+wrkNq<4oLu(obAS^=j@w55kvy{rxb)`#7iAa~D|2^H!<rTqG3!pa;
z=z`0u(?Cy<zClIp?iU^6Z=b)rwa*b^8~Q#37uy592a*l9%by>=y~wPr6v4)S<Bk9N
z^6~P|3s+qQDYUcAUq4^s3xACu(5nW05C%i@>+k0K%kNz55E&1IjI`F63K@_Dmu{>L
z0^_R|8T~G$A<gxjF0xGPR(ql8v+mEY9}Ro*?Q5?%DgnxJ$qE<_GJa~}ADJaqPe|gY
zdytd5Ye}PNcn?u;V_UrqjfQ@!dYnbVS^-v#^vy`n5R#81Nh9pE6Mq`rGq9^AKLb0a
zd&G<^HDs<KZ#MM9xOMnBvI53BzvU4ztwe2NFVi?GBMgLR;8;s$IF2x>TPHlC<`m^R
z>9bvr;#*7)Ijx7&!aCrsaij-5xIRR9i;jwVb80jSFpZT`K?LVVI;!kDuJ@&Me{Xe;
zI?9>R2Q7^4pwX9CD}U8E0O=`Af-x;rq98Gy)R0^2K1H-}^==kbH@DJHGG++uJsTp6
ze&((JPG7gf4k{*VrYO8mdqy_P1SPD-4)u^tDyC?rtljRhItb9Co|s;nBpN5I!I4CK
zY^kRi+w1ul=mx5Kv)x{U0)yxM&5eTv5v1d-6OYJzAaMbipMPb3pzNfRlIKw5nIS7l
zBN*@3;XJk`zB)W?Zl&{mJ0?nMp$%~TW-!T)%#G<HH^UO$Q)PH?K%Hd1zjqXD5-Jmf
zvt5>mWGoFjIl2iW2q2p;ce+G&*t*YImuOQwYrSFmGe}HN`zK6)O!C6?$6BX6>9Z55
zLEbFM1=@FUDt|IlQ`C&LU_`)Y7+VP*Q6}r?OTY#?%2Yl1Hi1Vi5VNbz=G&P$-HV{O
zuvyIdw6<eLQ6RJpBOpe6W0O$SLZM<pXOg!<FSK{u*cnHlEJI7B&fTt)t+YoWjedf^
zFv%x443~&)E>C<K`plfXNB8-i@$62jk9>NBz_RPaK7Zr-wSA&?I+L~{gb9?6c~eZ1
zCUB`>rrpVEL9RAJVA`LwQ$mJ*C7V##?j4Hw7<eviF8LQ<s$1{CyYrz+l6SD5cV#BW
zEisYXFY+%+YXaOe*0xOY8F9xXo)mYr)Rl2(&WgLbcetMucTB2&9Cr+XnYhCs6!<Fc
z%>KBmrGG8OUETUf+*OjilgHzZS)$j)9XRm}c2vn-Ac9p}v#|1zdNT#YU{Ve9d@Gsc
z1v|MOC1;d6SRuc@CBVa+qQgMnoL_1rE-59f0K*}%wr3PzndB!F5xMSB->|M|n3vsL
z!dt!OD2v!GedoH(aA=kjb6SVS(*2I&m@x#%Xn&OEieov=a^x2r2icOn6tbgb^459I
z7QGRzKpf$#<ZST~W7?96e{ldiw2*c%)?tTn9SxJ%mF<;S#Z&iEQfX@l(M#g<)W+v_
zu8y?4o=2-_EN$hjx4f=Zi<7(BTj68f&cI(5Z(K<`$iTOid1z%eF0dF7Mt8G(DSY-U
zsehHY3a;WLC!`j-8{e~vTyW4U_asVkRY6!4c9el9Q#JU=MTa#gp9SQ}upNSK4e!!K
zP>V7YWi-sv2?ua^vEqbxic6a%O_UP|FV=b&Y0|=TBMz-q%5=aCSz4>KV5(p#?W)5J
zv;~t%TL6=_-VbIi=nw(TU~)sI3F!3B@qf~Wx}r2?U7w}fng3N`+!~LGDkUi2J%Jn_
zK80|3D$8c$MWq~@+WXIJ`R4i-ZOyl|N-YM}-z?ST@_E;L7uq>Gb2~NEdEJ6BZK?2n
z^iXR0h3(a)nohS?(gRI9EdZ~`(t5Pp=+A>)C*SLsOZ@}e{sUyHOjkU`#C?G-Hh*=O
z$6pa#>`do0T8ibjat33dZW~jICIl7Toi2|wb)_&EJ^j|r3uE`ShTPUNrFxFjMu=Kz
z@iecylv=<%wPSPGbf}S5!l0T0aDfq*l(qG!Njd<5{G>_T-zO?}_@qhPIetzhmq}h&
z^>IC_5Xjs{<pEv6wuZZOTJ?j!K7V31aY{!y-MNv;Go;cRLD#7DO8wB}92WWYm$<&2
zq(zEo(V)P1vUH$X>^X%<Ciw~LW3GGX=DT%k<d&c_De34(sMfdk9O|75pIqy9&RU}}
z*(M;kk%SE)_Q+$>cP@okq8M0CIV`Su4#84}AqLZ+&be_g^VuFQr8E7wo_`^pN${f{
zwhUjp@6RRZJcRdc!!<uHX$bo=dGF2YL~kyD(_NXcr$$xcfl1g|qZ-0?PHoDAkT|k%
zt~QnD0@nH%*O?+Zmu~5Myo{xzN74$9D?MMvinZ%Vq}Ue+VAk=(8`_mHlfzI794UNj
zD^saj^{}E<&w(YREe;CuI0QKAdhM6aOe)6)@$xTi&zhHE%NY{_H8z*w+9nhNH#e6d
zY7ZxW8_AC2y7As$p&u}Z`vNo?vEA(fa+^Wlf*c+OnaN=$$V(3S{Zx@GlA;#Nwt5;+
z%c98ISSv~K;m5#-&rhfKr%xB92gke-P7d$i2LUfQ{1+S=tr?sU%qlUw{~G@G%(=91
zo#4`(D+QN>-yZJf<)DOz9s=Ho0GD_<I4z)mMZp*)q2J|22YmK<+&Q%?fqh_n2oC!T
zpP{IizrX+WiDzOzU5p<B(^7IfN2zt10bHWA(ovut;3r1{;AQ}czQ7#_(Ljh%=~;Uq
zL?a^x$Tl+f@=Z@@GXmd7@~#Qc8p$HXi7b_&8cw)Ek1OCCxSNQh?co!h@shcRD?QAA
zi$wA1Ol$0IUgD%6y^(gj9}9<9<q$T>Ax*Z4nAG`2DEXXU7h!7Xq%{s8u(9Qo$c?|>
z%9~RjeJmt!8V%QZJ7EQ)ZpBk*(plI#jf-sKZ!-}Aic^*+oyA6zbdHKJnRv~aG{tfv
zg?2OzQM;j!_j2G&!{wDx!+${Er>$Rq(E_}fRW&7;aTA0llW4nHmPz1gg0iSwXKD*N
z?LCvh_LyMyX9XZ7^FbHa5nhabt~etNGgBr-oTflJjxBOl>|>qIy&S1kY1yE^`~tm|
zn#Bu{ghM!|tTMc{k~ns+i3F$&y7!Wc_HHuO*%TsnnYnYBGO~KRX*QLWjBC+<W80G-
zYnigH$DlKvJMp1~2Q2mj`GA<iY><<A36+PWncL{qAQkC8l3v4^g)Xad-$ttq%vd=?
zE7?nS%i8Tnec&J;J(2bmJ1$z}vyD>Fs4n`de<RN&I14Hixsje;D9IsBg`lMj1(oO}
zN{|y2Vo>UuXzgj%Ez}Avc{g)^EWRkccOxv&YM-O;6VH`xJ$C6!LY6SG5`{cwbv|gU
zCOYV}xK|TM$LxsD6zdD|b|NN2BJ^#f^u)4nW3VX6nQF2J#1gWjJ`<*S8hoEDjd`W9
zV|-VVs~iH~-X~EIsAN{C;oGmLKL^mj&qGv0uoM-%8U8n5vp}=37F;ENG;1LOTFuj^
zfB*RYmv4Uv1_S<!Blw^(?gl^nN{`lq2Uk1{Khv{As6wkjumGth%oik|!9WKZSQ@?!
zPg*)A6b78q`V(6r$@bv#n!%7@DDSW%A*fA-2838z3E$7Qk*KODB2R@3d_pLBEIz_`
z8ikK|wa?TQM*VYm`aV2=eJSz`3jW}39c14#;uK5^2bE1DQfdPCDGUtG(g+UbdcLt`
zvou1m@D^nxJ##!wN5)lHrIv<WY64gUftu*vw=b~Yahe2@w9+sWfI*MqL7fy?s7bK(
z_qShp+z%M_xw$7e2O$;H#tsMLAwp=2uwpq0McSNa&E2UtRYS*r?D5D?i%Eh^n8|my
zCe{H3621%#iHX-kW%kQv0YTO4+!g^5Su=vs64bs5W+CJd50ENhYE8+qZwf*1lTdd*
z7caM4-C)3!M2|YgD9W*UGaq^-=H8dNNfp3%Oqh=t?0Wdp4st%zkBrL+s##v(X{kb9
zRzny_8zx}_EoM1?QShd#+NBk_#D(dgr?&D)qZWdOql!ESxG|2kayIv*%;rqEDS&25
zD4>}<S~bv``T?tU&C&j3DY-MX%^F0`OD76zG={2n&Y~>TL>5Xcu5i%XQg!)~P0|~k
zf^vS!HCJr4^rkE~h!Ms`&Sz<wk$9Z{vH(Df6-KOTrUIvbuY>C4vyfOFEJsQt$6qoo
zTdgNs&OwDZ0e);&FfOg=BAO$qOID)9{Axxs62I6_r2SF>DO2{9WX2@}+|So($OL4M
zD3XdZ^IB8MLUPv9=;Q9I^rX;?OEtk-Qp=Ld+nVyG-=j+7G)j+kS3T#tBBi~)@k@&k
z+NtMt0c^5=N|-#=w)MYO9&<6eM#H%nUE^(v(f^A7A&IjH5^tZor#}l2J{&YUZXa+w
zwxx_yLLF?tpM$piWBC5`r3aUZg<u{dI{q^}Y1XLaYa88jR)1*H`WW2<B;Sl%3sI!i
zI;e@f(U3br4q|G6d#z3&+AW5tu?Q>r_^VPyEe%zFQMZ-pu`I1P7Nw1rGNl=0DbQ5L
z&}60=L#oOv^BW|0O)}cU%cj)+7|HZl&UG2DSxDa`9jJh`usy`NEd{l~xkxhNW2Ho&
zU)mTN3AGd{lDD8}5hnw;R`h-oVhs~geDs8LSaz6@F+`cmetAs4mEBK5S~1Dh4H8<U
zoun~;h8QIw^Qw`rYjsQ`k@PRpK5A*@oQ=4H40%RNSCyfLGFkM!)r1acIOXenGfh%R
z^t>Gs%(FUil>yQ)C-odzBc;EA1O>^NcM}>G@I%aUdMcK=EC}zfP11Z?QNiu>DsGuE
zTj^D(phDh7?>SxKZ9g-S0%Ee+(@e$?OTt8dn_5(Wy1HEwOCVp1T7H@{uaxR12m6S2
zk-m)ST+}Yh0@}m4_k2Vchl6B*J5jA52Lq9Bq*{r<l$`J=UAxMfDR;GK#@eG$ecnyS
zR0wf{X+It7Aoe;&`4knRX{CKduXf}WN<VFGipz05S>SRZ#pO}U*A3zrA(g6@LdC*=
z#yjTo;RP~D*#o<7in!w{?l8j~HagnHB>deIImSS{BPzWk=oltlFt{vD|032=OB=>%
z?jfA(Q0x6hFgH{&S2av)W}N!nDmKRYsE&=Htti=@=}U*8(}3IbR4{GB1ma~ePm@YB
zsboJQs)M;BT%`bX9sf;GO<>MUxw7+r99hjkhLdS~z*Lli!d%bEW=~VWV{yU$Zc`a!
zm1@fh*1WlR#e@qz#BJ#jEhwbcScksdoh)?}Ea?g9>bUp45rTBI<Nj?yM?HgU7Pq<2
zm3R}Ar=Ir(tB>1TJF%{~WD077b=0@zkE!bDA)1B!F=d+z!qogR<L8{Y<_xHRN%bYE
zv7VlhAhix-^C^oEZ#g3m>5BVur+Op{&yI=LPIkA@ZJ6L8PO)>oF@kJZd6qMkprCEL
zc^of9O)o-Qi;ppix-nr^aX;=-r2!M?@;AHG4lBCUayU}d&CQsJG|S}O;#763knpJq
zFMDPoIcexHEV<2`3U|Hlb(=YV2kvq6RhK=7+uZH4d)yBuD^4;a<pgS&=g3*G(h2_$
z_$#Rm3tFw&y&PF#mR_<>@2Yi%{q0fW=GC6-HyuQCWnC%;gO9JPB}7EQ_gq8hAU<|*
zv6sOM?cBv3{I^moFCW~y%OJ)qJYr%VTrnP98bj>L*FH1qdOu)dA_yjbbP28ewdAY|
zClYetd!6pFV4mCTyRIzINxt`%P@GNi+-?b_Z(BC4ywxR@4Qs(j{B-Ra&GCVJBTojN
zI#mmEv3`c=Hgvd%2T6c|%-t17tya%q%zP7Z&{F7;GKd|B!lO%~uFh{hZdK-RURXq7
z73%t7sE2q^mMNp#9hrK6bRnC_F&#O`!VQ_jsFfv}HAfS2%fg3H59QdIX2^w9sH<^i
z&IL@Aq5If__7Q|2isj7FrJ^fyT+JOcI_1w$&{oakR>l?6&P~ls+V;SH_8W~RRsPp0
zCU}wVMWc#6Z2J0+pC6vT0Fg-)KdxrKiMeJX%8Id#$XwG!=Wdf_ok;Y7m(;GwhGe!O
z{8rID%nQUsV?w)m(tMY3g$M29v7E-vMGh>o{nm2sJIjaZc97t-+lp6Ko`{p_>8S$<
z+>g0<!^PD6y{4L8F(Z7d&-`**|4H<kJqy&`-`=184|JU3nU}$TC=&uUHJ5SL1r(Qo
z^eiZs4OuD@e`Ti}U~UQOHkd=B(8(N9jb0A@_lMfoXFr$S3E;Ayo<xx!DUzZTx_n>g
z@_BcB*?l+*STr$`X}-LCTNu8e_?J;^RC3Xb5hd92^0xfvBc)v7=ZGJmR0#ZN{B7`C
z9`=G6?7{F2v-n|$y#|IY0>=oB{muul?%Bj~r_e5#f9r$et<|nSGm#1${`>Op9gWOV
zVwL;pzWS%<m)#$`o^?Rhk<sP#ZRcKqU_=<caAk<KZ29xo<=gH{Pc}~`#ZvzK`F-hH
zzDD#6L-eRpQa0!@ON0`OQuys&1@xS>Ao24FKPV+2G7^yhL_%8EcGxTGFmSyj_LkU7
zB0iuGe|yB-1N_N#;*Q)iJ_!eP{O&$(3i@20=^4NAi=DQQ0op0yIxYhU(hz~@CJ<y4
z9=QQn1>h;Z6^A`=294m(ay%;be&#?E5IWnp4w$>C;hdO}oAl;5^G^_j^K|l?!u@S3
zxI;pU%NiENC6UY~328B(r)XnNvgZuEYAw&pe`!K5sqV*{0g=TP;$$AKM*kUJT7yWB
zj)|fKdOGaUU8ojch}dFhhU3{&{Epaq{2jIrkcJ9^lh)!Srlp7`?yx&%bWbg!%>v*N
zaW%0SxSLmLIZ+_8rKj0t<m8mNx|>&VKd%O(j)9CTi)Nx57cp4sVK0H%ak2H_QJgLO
ze~~mW)U+T&AI_G}9eY?*Qd*)_wqxfQWb$dJ02ODmS!JB4%$izi7=U8XmBTDHE!c+g
zY7ml|D8#p?atfn2@Ux17QiLf!Qg({yqy|SW{qeIjEnhG2uQ5>b0fj-E*JsESjUY_h
zf#RozV(INd!=P&!<CpV8N#}l9@VWE!e?E)W1{}uFpoc+QV2GhPAUQXZmS=Xy=kS{z
zqb9XmNi6>H2sAA<0@f`o)6e1<Y<#ilmwJb}r|%J#^^B?Ig6(htgbO}!po?X;vs{#G
z&2~6&g4yRN=Ty@i6sCk~=)IyLf%zqve_BKBEXa9^bRnAfJc#DNxFQ(0#OjgQfB8bb
z5*XgTsJL4m()M#OHs?qI!3Z&MbPVaNJcjs9OYRk;xNBt3neXW$yqR}_j#k=%8&0RZ
z3$TP?wC{R5#naQexo|~{vH!HWw6jo|k;=JiLUg;;bTrQ+&9!_uxWm!9dUts6fjlaZ
z<s|e`>UD7nQ{PUIHT7yBiVtFUf9ky}Zl<`-B(XUYgR;lFHE&wqf5tS00&0>?)9um4
zNAT~-H96+P%q(EtD$jXw$$yLO^_$mXU3KeS+fRwB)h>WRpURv|URn=7fv2Ds(40^!
zZ`6ytUoVQG|4{GLi+ZeHB>FO~H|s@Bd7`P>^f=uJOE@z(>PFqK8*T{3f6}YEQ9naB
zOcSwLGZ6RJbM2fia5gKwJ--Z<JORR&>|Wn?e=OMX^Wu9W0bz|?{$6}DS~mW3u~Gs0
zJbWpYQBL69hu^<{``7C)jNq`pTvxha7Al!F%UgUDsC+K<@*_Sgf@0&Gj99b*Gz){s
z0Cq?NbHnSh1Fms)X{y#Af3DRMFu-{=*cTzS>zF2}ni3C=Vrh-@{iqu=S9uE0bY{an
zIVy3id&Kc+lzVia`Z;yQQU6<Z-<I8%h^kadY3yPQ$IyiVOdEh1s*Ir72v#GTnlPeL
z4h#}1R%3`?X|3fgT`|dV#S6DcG+LZk@@!e=)KbZvnvhnakeV`Ue_i_u4xOynQt~`Z
zAzgBa8xG-2ltpv{!sYktTO9WtM}4lo1D(Bd1R&M27r61&!WXW6jaiy$g+6vqy_3I6
z{z=4aG|VzvT^#|O?#oiK6Xe<#!M<z;gE1l{YyhLr!730Wk^-4QSUILhGQ{3B2*+wW
z!_h$88gaI6L~?^ee`{BRk!X26W!mlfoqMBhZ?>^c98O8#Ip%y2cLXYC%F5Sb=bKbq
zd*~cGhNLcWDOYn(rGhP(&by8SPt|PC&8ZY5Tpcg=q`An%Lvo(p;4h})j?DAjS2+s}
z?rZZ=RtY8h_tnNz+5Ny%Gp`9eHwEB4L+-g6k>f}f`gA+cf0^YM8=F~vvEpF)zb$RI
z7@hW2cuMOPy)x%i%}&&WtAMkroK<Bo;nZvI;$}Q-3s#&AqXQ01tO`OrvXHR=4ao@3
zrHm_2X+%~}c)FP2<f8wzgF^1g%cfd^Xe{<)&<cj-Sg&oxac^YO^)c$$x7;L@fLf?9
z9&&rV-=@{Ne+V;Q5uW;<UJjFv9&UCiahEng$E`t2fF8Oz$Fab;8JD;Nmtox|vaw0(
zu7W(*EigqCZd5x411R2ho}%JTNxh9q!7Di^BsmDL6o6<`b6@}?5{&bXRn&`?DJdrm
z_w%dV2P5&^=^{hpPW)j4ZR<M8;Ek`!B;HtcoZ^hqe;;Kh^?B??w|X`ISd^L3({)*{
zaptmegrS#?%PUZVJnNdthf!BHG%W}9L1uX43OHE<iO(t`dM$VSv3`*Uj?cxA-QmrI
zzhv%u8~3qN<773#4V2^a8l}Qr6~1KEzby1M8_i~*FniM3y3!$8pnTbLVs^`QPhb<g
zAr1z3e=jnrl|Mr1<|zC7%M4nE+ZHu&c*?PL00Z~*xEq<IVS5qI>}++XK~0P=mj6=j
zQRO%C5VDDQ>L`4O^y&<hhb0`d0X|urVV0TmqQkOPxpkB#-$PjO%IBuHs)K5jEE-*@
z2G1}^^j#aTD~76atR9cN<S3tWobXn~MOOx&e|pIZ`=`i*X*AR(*n=GigZo8&Iik4)
zgs!=o@FpiJw3o)WY(_NnuAM=M=>tXoK54SH!sMdDxEM+&bDSbd*Ep6jm##IcsvylO
zU8R5#=2k^F?v54OhD#M8&v}_QRddofU{$%=f|qI~yK3uGYru`Z6}&E}huwkhF$#O5
ze`_`xdL?PX(^I!%puW2>%mq<P!J`Zuq+^F1w$J5$gvTe3yCZ-gOa5L?kIwc0B80+i
zhUqD!eb24;<&+{)i2k}91I)#6|9|DW-1RFJert63Rt2X(5SANmRjeL>#FLhe;z|a2
z(;b3%_o1V_a;C)GnUZFGZ!x<Ma12)~f1RZKDf|}d_TyO22q@dnxON=NCDD99-d$Am
z#$VsUa^IhdsrxkGxjqxS>qZSToCs?lsQZp$?Ku6`NLAaW?e{evBCA5J8Ftg{Azy=K
zu1=Xw9h;V~8}Mjn<>Kl-V{XymR!&E46}9Rh|J}CSbIykKb`#Zj1Er(d9ebAEf9U@(
zmFeY8o&(v7x<^ogUh0-hH(}zflU}+KxVKn)x_;5~_u$ouC`cuKw#Bxy-fOWfwG9?~
zdOVJ<qEd>tKMTipR)^!p)cMlAxQ^;|Jl&k-au0u}Pe`k#n|-Bd?kc==;8mV(pEq7y
z>L{`EeZB}c=vwQ6YP!PgtZ#1Bf7}Y9?Wl#v+N%Xj4<4D2wFD=dmG|ajE#W~W?|0KF
zh3h70$_?8fh*z~*LVnr2V;Nw=26EtvPW;CWL~%>PS+AsGj2CZ<bz+q@etP#Bf2?pZ
zP<F!A|9{5L1&nL76Y#MMO>=TnLU?VRHS`{PqLA<GwI>RjBI>Z-tGs%;3$g6JapU@b
z41NC;m%)E169P6imr=()6_>zcDky*3Zrr-@eZN9KV1~mP-T-4!+sOv#TMqJ8^q~kE
z_aP~IP9OUFo#9OsrPX@lg%cbDYe^J2ocoQ^)AOXKkB7(4hj%BW$*W+&o9Xk{3HXKP
z|Dbu$*~tSahcJEqKK<=pYvcI$$Uo8=k$;}=17Gd&4k7Ri;5QKY2gf^4z$1UiYtZn#
z(~+V)hqCS|k0Zz&ULT{E`49>ed;I(7zaDf!R;%dbRxabyhtG%K4<qU}dWFVLU%nsY
z1tBO5W(r2RIQ{X@>FeRs2()cwahm`A@jMM9zZKpEVBTrxtX=R9G&7VY=b}P+Rd~lV
z6Vpu056k!VL*}Dp`Yn&^)A4`KY2koj{?Um?nL9*zcH${tot9aKXW54gPu|jF9c8YS
zI|>E1>}AW|GRFJE`+>KY37)BAsvJ(v28GPFLY-|GB%I2=vL8K^eHk-LNlBP4ZJ?oy
zqAnuz*CRz<5&DXhu$rOPecm_LHDvXNHuhBRf^RHNYABdJiCQ-CY-4|q8x&|pFaNVT
zQ?M9e3f`Tf-c=%Fm4(<Sde4}KtR6|AlyK>HJ*$?a*_7=F(c-Kgd`S=!bWy%oL`Yd}
z(vDS?PP6)n`pSG3F;vaNtiJ+Urs&5EKIwGTj8m`mbw#f~(8N>iAk8>sf)xoL>%YX_
z#Ym?hh(_tKVyDrAz^;FQ$n3NcM~Dk`SbDIw(t71!T&SN4QP*gF(GehJ-5dKEI_GC(
zk6ZW`jfyw8&cEy$DDB0QW<2}x(7J_#kr+l%VxPaS91?75G+gU+eUnC9dpc)mr#vn)
zD>IWM+fyi3h1C$-n?VcGekNm)VA#d%1vbn{(<JM&!x7U?G(vyD)X!4BIx-utZ17`b
z=Ft%CrO<rYrKZrl?^P|43dY>fs}2x<zE{nX;Z&#ahBKkWt;NVr05Sm-=Vl{eOYia`
z7MU0n6FovoxIbR+$wh=<R(R=^$gYBFP+7kcV@m*V+moYHh+#$9sY@!Zrq3d0jqcX;
zIj7jOuIR&k`h0&-A>fL>r&on5i^y&OZym3)Mpe9C<>iS>2NzptMt=k+rHeJq)t5xN
zbM*`7@6Y;xHs94`ra)Ub={_)sJ5C9bywMdp(I%ug28_I_dX)`qg<ZvTR}5halxHU}
zFk~aFWhmQhMdO&QW`vT%P!1~+#BB*?eM0f0A%@T@PR@UZ2t$!w2fj&!A#XY1#2ZGK
z0b-IXc6w7~{Jv-a?SzMFysxiMh}{)9cS|~#5Pxzc%@qjTpaRiGp~jKC*Hj=S5)KeA
z`ScR$hE=cd=R`&s?SEx|t{rC}rte}kH9F01j2;RPj1BB9nBV*hSoyFT{q5sf#|e@d
zm@w}ACdz+2Slc#hjOB#Dukc&cdIf06fx`e?6F5M{P3o@+8(x_a#^s<<F|9R?0ka6$
zC8&lRjLX}HN>vKJx6%R<TcfMAz}9+=y4uUVNTd{;a%*ibeAtr~MdDf;R(29|cQ<u=
zU9NVGl}>IR7kpC}*pzNAq+DrFrqv>mchtczf|-Ba#SuXIjX189JsXY%H!K!%J^sm}
zE|6e=vlzERBxhW{mFN;iTDDd5x;a~XnCC6CvYcbKQs2#LoA0;t&6j1ADJ!L-{4~0i
zBcy&kWc}x~Y9%zo83^ZZoK^iku{wU7l1{RzU<`Kp-z38zcHx}!d_|I)gU#mf?my38
z|MCXqA6!yS&zIw9Dk%XDmkntuJ_H@UPKTFIX(}E8&X;a!Dk1@Amyu~I93#xDK6cj(
z=MLCI6Ia`J=tUwRuu^t!!=@7|)9dW&gdUD)@0ZYNDi%v_Cic^d83dp#XK0rhG-in%
zV>pv>78_&E`1MKy1&$<VDran74f8K~s;Ao}U2NA<R18SmF6m0)A%XhAq|39*h$}8N
zw~w*wZ||2eYAO~1&zDGQDlUIv&Is$<KI32Z$-A}=6e|6qgTF_AUT5aLq&p_3-(+aJ
zchuNIC55e<?(~!T_+Hxa2_<qP8u>UPyhM>UbX+g`Q)yMW+CE!a$!box5}~Pd0N^7I
z#iH9RS}Abx%%$!C36`COEuT7juPvWC`^W54|2-|ImBi<+%Y0t&(FA|;wJWU=kI4|{
zd~QF{<>2atYfC6e9v_OduoMfm=i?UfrX7Q|Hgr9j3ti!MF1xi{xWa1prI{>nY5L}7
zL#Sm;GraE5uFGL-*;O<^ju$sz`5s1F8|Qf2tf@OP+sKq@aR5i16@_gF;F}fgmrul%
z0IgVdUA0S&%L=~|`;UJ%K&8G%yE<dz^iNtgZ_xoYf!Mg<hMmCy;-BY?Z6NFwXKVwc
zSDvwrc-`iVZGZ}P^L&0hV;d=Z>c0-T<QxYFc`ZLk6OY6-Kej&ivBgwoyQw2Pnm}N<
z{#;SH;kYkpLaN_5e60BO!~I?3S#Ht%$-XZ!j;m0_U+nu5W1N2?H`h;3M?iBwC-5e?
zMUpKeej|C$qx~smUfE}rWo2oGCet|cu}Er^&a;cZbfuIKft$E$qjs@%dvDJEs+%o^
z$V|({7H{T5ZSjVBBKi`V8h3`gpkS%H8nwZGNwzUV?Ihr5jwB3hd$WByyeNUHQpopR
zKuuL+C-1QSsPkFcQ~|6}p6pLNEQJUv#Bjq@VSxCH9F}q=sA9NfN;E)x%akZy9Y&3f
z%G)?14ePXGZkQNljaiYoVF4MZ@^Q-iQR$vQuMdU56li09;!)>#WsCklw4$A>m%)E1
z69G1t!8|h*xAbf(b_xk!2lf8~x4elnm%)E169YClHkXNaDn@_A%l88igh*q_Kz25~
z3$h4|*jZc$xi^daeu`ujStPq>MzU<!z<4xlvRJI=VUaDjubbT7-#mW0`R&=*Er{q;
z(A%dkn}P=t{#Ozjz1xCP!W*@H`nLV;PD<_J^9Ua+r7`dc@OOl7cYkYCgdPf>s0a9{
z``cg@v=|s;H1t|~zE?8khSUaGnTJUDeS+WchbDpszj>R=6XuBmv0^h>{J8x4UU_=<
zl&8MG)lqq!XNkyMxc~d7zuric!MtD}tbkc)!3IV-pk{$T2lLwpplCAJ*47qxqL-+6
zDjt9Lw^@KLuU%%DP+t<iZga~#G9eFbUZ7Rv;T~~*iM%{l4&^iSv>sPu1Acba+Wzu5
zhpKYPT<2Al@i>mH@Bv%P+N#`zb}{Nr3JafToZ0>aO{}dS--B^RwP&0rd0~T}b*0VK
z{Ba97aslgrt2j?G3b+VVcd>ywi*dTsB*qY;wk}4Ox<Qn2nLS5}DeNtmRe35RNe|=#
z*uHXmjS94OLd!7EDo?c5vQD3W={rt6&#ufP8baoz%15K*N((tfLOQzgo1MaF!(5!n
z_QclV+Pr9VW%a2$KR%U6>C~6Nc`6xyuIs|_QH7RVDU*HrX4N%V;4DQ(zPZa{kKi6k
z2W|;X24oSZ?u>=kk|D^0Oj}dq+}E*%Mho<nPa2(l2<f2h*@qhX%Ib#Ek%v{wTgl%@
zl0{~op6Yj<5_u=&vt20|Eu3zasT;4MhTmJ#XqF^+Bdr=k|HNaQTex)U({OBmZ{A=s
zZJG`E%-}a~Q-h2>a`}uO$GCo~Gcpa1Ne($|wCrAMk?=$2d!f?eB-cUrv_^sZQSCz-
z$4CQDLY`H<+fVV@2<6h|r4!Dg${wW~I=@M%1fvMhG)#0+#+K5S?pi8sn8vCFqj&i$
z<7X09PAeW9d<_;=3EKM}Fy1OLh3KG!jb>t}KpS0sN2v<l&dJG4GBSgF?g=qvv1!k{
zUL~@EEfyY|$K-F>#KZmU)Rzc*Dj$FI7@yDtT{6N&dtH?75f_1{A7e1~r<hAwu0~Xh
zLuLI4m^-3r@$<+h9iS{N{ZDYpai;6MGW9++-dyK>63pGkdB$tpugA<WFK`<P2R4FZ
z_icpb&hvH<mn_9H?^*PEm@AzPLTk%&W$}lkr+AMogxnClVQZ*41b?j=0H1#h+q0yP
zwr6QTLTN%ma`I<}Wpwf^(Ae1LZS!5q(pO|>ozz7eombKFMeN(Uyc@T5YaB;Ba+X*A
z6ge#)hMI;rM-tf8z5y)n({3bsrF5>rF3oxg11>S`I<0f$#?a^^L+>5ubd1C714o<&
z96Dgz)!&V)8Y_j4P;;e@`~H8)ZM)EIv01YnVXod3kJ(4yq+Q=fG;N>ml`WN&1NT-z
zc)z%}X&d4W=OS%WEmvk-#do;9J`RyJ@YN+bJqD(ny3cPaE~Ntog4Y$vu5cEJR}<*7
zJ);cXaaem@4WsGd>C>VLYZLgI-9CT2`D24T^?obK3i7B)(QW_Ra<hNRNz;2D;M-WG
zQqHM2zy0Uym%n`ehY~R0zj=h*AbUBL;`XinbPk3kZMNU>cNAI$7dIsmfH4cu2{wk1
z*xSI?@OitjItXPTZ}-;Ub1MMVXtid8jS?mX=-4J;O~gZ{*jj16-+3c()u$+HYHZ|R
znF@~0KVdwL%0Kgm|E7PgFzR31&6n-wLz8DR@DILyfPB9rP5=Zd7{x_OO(uvGlJz8x
z6hP?*t$d-aQ41fv1;Obkf>iKVo2@ODTG)cA31|@oYNGGC{XhWITB}H~b|hSaab5(U
zjZ&dht7QA%pTFU-uQ2F6`>teeGg`FHZ?_guIt0Z`s}!-pIQD-aA9knSwBdNTH|~yA
zWYv(yYHX`B6U5^_HW39DK5UG|y42gY)(=;UM=9YzkC){$gJ>ofaGwaCOmaCW?$Z)N
zhQwVi5a3NiO0m_6Jg*$tyB?fM?mkhqtk(jmpPjG<R@E-$>v0a>Fy*Z+#|6v&a8%u4
z(OANM!KQi9jO%|o`e0kAJh<61IP`L>r5Im@L9NdSyZfScz)mTQI99febE&YlaG^TK
z3RJD&Xs*a%Rg{XCt19LB3}A)5L*6GVLYt!v)up_DCLjL%!q&Nb-VMJOW*Ac8=FZj;
zej>PCw$*!as)vNVIMstr4OIW$#%_oxZW*MYhzt2tmwjICdTa>0W(basY@#0u!Uq6&
zmcd;;wj1D&eF^-*8rY>1Oh7zBr4CEyIc0d+oHSyta>^0x;l3iG``4!rZ1<VnPA;gw
z(oU@Y+(&bRlLP*L?HYKUNA}$dLh#z>xtIQaDkXmc_u2C?V6|FDm!@Vl6NKqX?rvjD
z^g^2+&Yt&c06B=dRQ71IR)eQtQ2sJ>yltrE@VqGaX+PGpC!LVwUT&aIb&|3NUFjd4
z#Uwe(hWI{vWK6S9XN=358^4hyHqSKV8?WQ+Si?Dx6RS1MGhW>Gg$DTo8@+1*xn9R=
zrbB<dIbSHS%r5J%p_&^Vt;Df2MAV{^{9F-#@0{g!y5kuTRi7lQCDa*b$bo#4rV{6F
z<kwD-t5klD*5z3(MRnN2_|<KeNhhSkE=lvkxHWNaY5n$&PM28Z>`$ov`Gmkf?yPUu
z61)~hCtfSQRDIPxT~S}vK}{8?3$^oGTcRhOPIU$C7Q!%FS!X2bR6OBXs`lTje1$WW
zFCFB3W{P<TpG-LwTd$W$fGQt<>^}L}H%k_FX3D8Wx31q_D7DN$7k8ZQB^_Lu<LiXD
zZV5h9zQ-r7KB@J#7BlwFL7aj8qSG^12c<S+tq81=-hk)#6qK$jfvw=UwYZ@Sx4>h!
zbCqhZ?#XjD#j+Mb9_H~4Zu1i=Tk{v41B>LdxbxTh=@$UzKh{Hi3pyx&P&c>NQj*@Z
zk0JqN{_~&PmzxhW)pk-P%T_)A=VZ0Wq2(i9yGs&4tMneQHwUc>z`C;=Z2m-fgvPmZ
zAwjF@h>T_^fAxu}g}DI0ap8?yW;ykA-%I?{QCHbquY3n7tnxF(TzJlayWCZKLS<YQ
z&_(IMRzCDASyxEskZ+QI#7J?dJI8jo3Tw<v(tm~bp@qxXLzwY;rmW@uF(^O|AJp92
zEPT|Pe!FtLo;R(6&}v!r!4a+2N?4y3-ZWZDV26yS8hYqBPUa*1x6`{0ai}z0;}xa6
zt#a*>ykcnic-c0lS@SK!E3WUY8iweh7iw9XZcob)^t>!#@A2h-`bCXRbE)UW{$51Y
z04z3op_=#hX5VtG7Qw1z&E!qMBI+LmQ+2~z&RB$YO02@vM;AQz6|5o71M3TWC&C!H
zl289tY{6T&vy*}PoIu)bX;R$_a>GRzuuddvezpsYvueZkSnv2v!C|kl>ju#X71B)c
zvKtuL^LlO7xJ&(i9({pHC$`i30s2xVKm5Uoe4Zt-y`*38{B;Q(!_`Q+=;(xZdat-S
zbIeJ-r={@by2s(9e%pZ@s<)q&1V9+zv{DPeYr}PdgG+KR$=WdpYwS$@C~K!`#8dCL
zx#%}sWV<M&o-X#h=)_a-S<5=KXB4|iN+=trV*g7Z;zZ$pjT8qf7Al3Ax|WB^(@JOP
zJ`L!X5nqvYUD=dP6c15VKDUhv!@F;_^fy|v^2>WLy1y4w+C09DW9zqx_w?>feykJT
zB|%c9>#33{uh^6++eJ^Z35t3{%f}VAstD4LJ%f6(Mr&Ss!|p{k|5%}>HIx_Vy#UM-
z?C^#?Ke*3-SbLO=gw-shGF4fyHL#s(+^P9Rgu!aL(fDF1`vd@N6>`04!_y)!>tahh
z{}vD~W_rB=$4*5BY`wCmfQInEPP(|5!py0|*E8DHz7uQfhGmyj_9~I)yCp&I81A8g
zV9EAu$*!&C;1VX9?G<JESub4rRDQvhc6!@e+Fyo$$uRV?e9?w!Ykp+9i906y9qEJQ
zZG$k%^sBIYJiQ(cQIaxVnDO#<k+JWW<E~9whzBN$Z<VxgA3jk(m$WErj3d4$YvCSk
zqIi_GtQRR~&57eT$yzmIMV7Tr^wKBZf_!K%j=nlmL6SDf{i>Pj7syL*fDx4ir}6vl
z_jY@KQ_`q3!iMoBPz=ibJZq^5bljLP%lVngTRFIi!p5-hc*enNVPhKbJ^qbFj)~%t
ze|yM>J}wOZ7*uNL1@EUmJfEf;^=ki4S~wMd9rvcHsKrYH41zlDt_86%er7Dtk%PYD
zi0L9$d3*Pc`NmM1-;A<X9uy!E*bMff!s9F``&aDSYgvhtJ>0Y_;bR=SzO7}{o%?}t
z?43xFa;ER6PdEPuGgmqzm%)E169YCmHkUzX9}|}^g(@n4ZX~(w?ysmHutx6(U>Irl
zxEIK67IO=7SPU|g!%UFf9P<02BvPVO)$X#poy}%}9(S`SN_>c<MApjX+ai~r?w&s1
zeK;Ar1QDGIdinfyQSd;*ze)noyCo<kyiv>NAIo1Jq|_chPw=r)8UvpI-y{5WkM~AJ
zz)<i+9pIyX9`A!y05LGeXuvx?Dw#4MwLw<qArgKc;WvEIM9|=04pVu;JW-&f*i05b
zF8_X1o}NADsUPokR9@#<A}%gI{`K>p?<C4VisbXx{Ad34{Q2(JyICSGdSSd@zWlh$
z53EtbM7<;<yiLn*e_OuZeVVDZxg-n`F8>4He*3n6Opu>ZCrN@%r1#DZI&ndGXBO|1
zF!@o{i3QQwG%)qy6xAk$Aca+arr<!M7Bu8QF&0E*PnDXmd0q5RS?&OV<u=O6iDXx6
z(`pj-eu`RJVV#;OD))AZN@smc1m*MF32G&*Ppc(n5yooQYE(>DgcUvjb1n)4Ya&&G
z-MXxQ2=xX9z($R6rl{q_oewZnnME}V93X=N0Gz^#BJ4~Hd(A~OJ=l9TR0Hbkhfn?v
zaOcpdjtCE%645~mr}aFWe9GqeYH8@`1bk*{>a-z|S5Qq+z*bvPb(<U+sfCed#nG`~
zuPng^VND<`5OQ^z$Jbjz%#`d*QK@Z9B)b`Zv5WR1YFeW90_sCnEf=lFY60@}c<-gU
zZsporA;A*&tz7#k;4@?ASdi__eT2`k#)8W>Svkx0n>jAF93C?}g;c12<&iVWja?G6
zmmT~??{2)XAvTu~>kKwbW_x+2M%lt9H?SGB6HcXzlwlG5pF(+n{QN?EQ4&%KFg;~|
zg=-dH?B&=!L=7?cAx&iXGvcIbdu4H`v7)K!8BzR2UkAoUM7`#jd{fdzS~)jxhi>8a
ziJ;21=Mw3c=iT~A?W0mi4_zq)c}e{!I+rvg5~eBXBksc!a%NXS^{g^b?Gwek2O2NB
z8+h9n$L`-9?(p~B^;QeW6z6rG+x6Ceiv-?u!hY+n*MmLQ9Y2|5W`5lHw>4<yJ+_iY
zNJ3XBsEi`z1oYPh(0*D9E(|vbLf*fZ0v1X;2vuMLN(G;Z8u%}3;_kygzkU7FmtP>>
zFW>%^Q6Y;J8azW>e$=1N=I9yi@*RJBup|qvLx}`QHCSl`PY?is1~$N#<<9DV0QrQA
z4EjAold#c3i0^|-?i?~AO~633hD@<RX}<3qNUQ2o6rS2L@~=z<$L60fo<`-L`NMw`
zDvbL3a`$z)`-D<;UWbelWhI<E{I?}l$P870gA<c1flLr7B<o2kDS&C7fpUHu5G+4t
z+qXK30B$PhZa7=xnIMdB0s$?5qCg<}JHyY&Vd51aJ&FmJV4M>wglwV;rCTN3zkm6G
z<G#VDpV;3@=U!<8>Azp@9T<YFw947Y#m)07q7yc`1>EXo2Wm{b>P%bvYa#+uep=X)
zDsT6lxlg+V0>@~ex<eqdzhi>X;5jRemk#d4gVCVFmQG^evJno%!8{ayXLsN7o8mOv
zlYL1%VB3dF{*SOq0JSuWB`+J#Qi&xmp{I73F|qY*cvC#V(##p^+F0iL%N&$IG>r0;
zg*oS&T#hwmK)gYs#1XBFJblJclEDK0cDS;hgE6|W8m%xxD+ft69vMi*Xb>wMa|(1_
zQdaNjSLKrKY_b$el3_=G4w{rAVK%N&BNMdYU9t+qYHNxw*Z@8qW-L6Wp7dqA!nK&5
z#*o!zYe2rywln>f;2i5tAp4mTk1hYucjA_R=zG2Thkq;_O)mG^I{`<&7X!G;K?Y(=
zKAb~jgwVJzbVJmK7#Mtu-5O$RG%ocpsUZLdIFa_~-62Ex?%*MRezWl{1ov4ujF$30
z6v9<Yh4RgZd~NL3a5{4-#4ttVZSWs7dTWSqCit6#ZyO=otJkGF!Z%@)OFKLo_s|_v
zag`L(4|~HoTf)|MW!&zM=b%ijt_tVc8KIQg8O|-Z45Qi=qoG)LLf;nyimAk{&>`*H
zX^iy<-5Ki<yWTW^<1cq)G0Zh+^S|g-xet!#eQ-aLBUb9rwssbi5sOEjS8!D86W-**
zeJ&c5kpVy5z<{RTo@FJ?`0%Rh&y{|`8jR;x`XJI8vLKXe3VFT?xP(TY`lPxz8$1TG
ztR@=*rc0KByYJ`wc6W>S9+?ds&yuxDgC`{IX7E%^qmH3}<OuPZz>p?aI<8h#%7nq^
z2+gcoy290BL-Pu4&i355q)+)tepP0hYl<7kaxht^JWC>Rrc;-w4V>qkQcW^cMUI;)
zx8SFxTi^Y^Y`8s%t10wkuq|0?<{a^^>eNysk34I3^EHkS{b|{fXVScjTbZR~RSS-;
zOvsXaPq=n}SK?5B+9e@lYLz%Cg_gC*WxRU16jxmCnR_$sg|!g86~yN>Tw)H_U$4KI
zcaWbz-Zv7S>Jbgu!|*uanyovyT#h}|8?aI}eSZB7oHPR5;56B}3pg2s2CQji8uc7l
z>V2#Zl9Kdx&9%P<)pU_>zE~OB^kQYZv$VJ)%I-XWBU1~B>x$at+s(IBph!&PT-R2h
z;N&5#UQx|;ZL_T|$TFQ%E1t=GeWLqm?NApS^1_X^HW|25&32jZ+-B1@iKxwLd2crr
zIGBi-lPG2?8cpNl+0eOdE>R@#FB8Rmm*K@Up9re4%6^MF5%C$FNU9eKUM_U(Dj4AQ
zm?=Ad7Es=N<6W=Td30UX+vZ{$iVCa{1w%EJ0WuW2ATW5nm|#Wf7bDV>3UpxeA8Iau
zbs~)3s49wd#~Pue-8K-K&l{1FyG0ehdi{yu6-E4J!xP1KD&j**W0aj|dutINuVE)@
z*Sb&-yV<H0Rd=Q>0VE1oGe=WOM}`1dyzFUz{53^<=Y`gOrg*N1k7hF>|2{Q*qjPa&
z=fW|DOvtw>;Xl&8{_mCW$u(Qeix9Uc;S*(}gkmN98#?`o@&T#nYaD-3!RlDG*YRJW
z2-=U3E-C+(MmgGkBY_<P=sk3}`KWx(qyHPW)3A_h+Um$f4+PJ6Kk_Rwm6O#C;#&xR
z!lo6)YXV&K00}FK?-by2FPG8tY;O&46kJAX*9N#~M06_|p9^p?!oD{vj-~)t4Iwwc
zX{&D$-+}>3W~O*9zM<Jn*uPJJ!xPKCl~WEgVc)Jvt=#c8H~a1p4jrqGaJr4=9U`0#
zUKnjJP`-3`lPhe-yW96X6;byy-8FW9ga`9!sb>?V8V4rq+q6AVBzq)_*=~EfO-}I2
z_NN!>I;?`SRo7`Hnln|{kOT<yU5c<ifQztTnplGk5W0fy?NcG=hn^J(8(^rvHt0}(
z?SotOl<0RzTfzTN=@tBLGPyH$EU|7w|Coi>Tv%7Psk>?o4#VxOyAt+I>#l5n30wQK
zGyBRmDtA@gRRt$}ay#p;Dkb52kw&+|)L1srX)b#qWL5UULQHk2Hsi%Jc%>()QsLcG
zX{e5R%6f=WrP(U6dg|F4*J%gnfr^BUmZy-|aI`W{ihfzg*EG${%XzGp>!_zHZPif`
zx;Go@l1!}AtFAp_>5%x8SO9iFiN88Me>E*^73K2F6Ft1P3L8c_7b%y>@0ICU<<UH_
zTB~DS+oaAiX2o=TRoh+Ev$D$gLExCTkSUe+ve@o9w+y(_FdMbr$t-i>96}*p_2)V>
zjveky8?PjO(<|4+VOF0`a%d%J{K+ou%Js}6wbfC>L*<X@V^h7j`ANTX8DG7!e}&*e
zqZsowXB)-bxfJD|+P&}d9gVWwl|!V~8Y1Y%8=J~l;d={Be;*S*Af_y~?BOHJigWXZ
z9XEW@Vdn1A3!+Nt-1yZZ$3#~rv!Y9ykGwsz;5B!~HfwbY7pO41ZFYa8ygg&-=AaDU
zJ)L%1tz~fL1+wDJvbsMTb;QQfe<#+mqR!K;mD~25H-DabA*o%FvHE)DtF3DYNM=JW
zy*^rHoQ+}bDX~Xz@w&c4gK~TBZLvjir#y1pUt|pu66WFErJ{lZHVu@vja3h6SoPRc
zycb$VDn@h5iY>5H)B`T{n$~BL^$Pc*F<VRNj_o5=Z=|U-ed|PKW+zuXe|By*?jy1*
zbtU5IkE6Ro`V>B%F&*pFrWckNK6htLO8Xj1FrrqMd|=(gs?A{Oo1vGs&88v52&XSQ
zT-0o&HzvB-;%#r7?yX+7wX(Sf_O{o7L%ML=h$A(WtyEGV*EVc-x^u&J7NeD|;@NQz
z>^Ab$)^QlSxX!J2_CII&fAVaD5nlt{V%+uD)SEMV&Q6iV1(YvR+ERb;aEc##XNqE(
z9c!OvXILO@s;%}nN`dOwO-K1?8mwWKTeDqn4V24>d&u&4G&kXTC-1Oovpdl{0+PQ&
zu;#u#$CnFRs^l9kHud4!(3#VvUt0wa0d;+r-dd2C-hZg`Txg0bf9aphL06}$%@#K6
zc-bKI2Evwscqbq=X@MfJ`#Ib92zxoePJ6KZe;_tqJ_s1a)ci8Uv>%2px?;_TzK2==
z+Pv>n+}acUH3c=ixUkAG<D0tu=-e^B*0I689kQ9>Owc#cWc;z;unJ(x<ZB+|eDtLs
zNNRiyc<Twf(zHPfB4by|b-mrNwixo$DMq}lJJ0-@V{m#RH)Y){`~DA7M$Jc;!G9<d
z0XLVysxT7+GcYihAp|}rmoV`c3YR;RDzX7Qx9*cFpaFkRic)h80<SPxIvEvkifRBz
zWs5lE8O2aGuHIfPRMlXx%Lyp2R;#WSCN9XI)l2)A`RgFu=uuH^bgKsyBX@G1x#}00
zf(3`+qy2+TVVG56pvbC^W}g!ZatU$U9D-h@k0Zs~=8#+>1^5D0k9hTh{c?H@+P&p|
z8M#Io-Z+1tkF!Kfx#X}fDkp9G0()^=1LeF&iElL~jAOEhIGdatvyYQy>+=wcLnxEO
zZ`|I>m<3~n-GmUdM%yCyxr`<c{WFo9RI{hN6}cmn7$J<<dxX+>L}dtDQYb<qL9o!U
zEfys>D}2ZA5e23tjM`NIx$RZ11k0%Dvedl%3yOcrH@zx+V~X!F;F?wh3?9D41q)_k
z_)OWe{6r{qVTBK;Gzp^>JHVQCC$FUWwS2*(y6|$X1SwtG+D{Qzz~_l!u1YUIv8N?f
zA1H1le^mo|Y>Q=M-aAVC{9gb4lbkE#))@D4j!DNTxULQ^wRWk0_jn+_5U!pz^xSsY
zn}L4<UAT#16``gMhMcA!%<guO+jQv|l1%WcROa<vdg%xhDP@_S*g68snq8IUO<Fm~
zpPHhXGeM=P_XYJEN-u$wkap)4!?|&GRl>bu1v*e{y%Ur+>S1+)m*zOFZ?WSIjZdxK
z$dS)2<3m#&u%EL&bwlUGd9=XS)mUc5v&MhUhtF4iA`P)YN|pLVzb%0Hamm;~K*wO0
z|1Jp(g8<p(n5d1EF^T!%-M_zm{mYkMAzCco{*y+C1&S*xV_1HaO9yi(1G@6PT!Yt^
zUPTD1QD|ots9A(hx<unD;LGx06qvTQfkD3~Xk1M&Q0cV6xwqvLBGPY=%0VJnp)`Np
zcMc?`+6;L`m5X$hsNmRi3FB#0x=bH_O{g&HpUcD7<>6D6X4LSP(+p6oi7=t|l^%jH
z6KDlR7;StYX+$rQ5av*}&<a$L>p7!gX<_x?dQy^ur0Ze{!WK**phXl2MBgKPLjc2C
zi%2k2SZ#zr4U9>fbW);Ni)h>LFF${9+&38Yk$o3B2dVUGtXmF7InV%~Xyulr4bxsb
zq7ycm`st|cV{4OSqie0(lH-(_XC^if1r|OnjK#9nLtE^p-2#$wwcx>T3uyFdO)%QW
z6->GcX5$co)*$q1y`THQ;{j6ON8X1|cLAy*dALC7FX86IS8+)8a%bwsQ^bF*_^h{%
z5Z0MTNcN93uep@If9<FT7R7DFig?Ggu0HF6vAL*^U!&EfPwmrji;hWM(n*Omtsvm(
zG35}M!c+16DH<p<IpNrnYm*nAT(tFP$z{3~YL7}%7KV|PH8sYa)~Oa$sw=-)92!@S
z_31&IaEF|%G}z)w`-Ni(FP(p$n22VVa{x~`#?)W@%*S-6FQ$WTuD-}pXSvrCphEky
ztu=+S=hsPea^4u&eDzM1UU;7SOw&~yW;2XV*T2TIE?$g1&G#AYpK`VC?8=X8Ec#OH
z$yN+%V_oVg7MHe1+1O*jmi5f7KX1FRY(_DBprsypLvG61M9YSJZsLEEne&^u<mO(B
zOa9lw(Zq6qjL^8}b0$bdlr^ab?4<Or#{7MmlZESGgQ3jHI!LgMY^f7O4XZsM(7+1l
zQl|>-Ql~b!W<PZT1|lB*M7a~xHct735=0>!UE0X|QYh^q6^gKfU0yVY{-35$#zHFW
zUnPYC7YX@G+?hfV_HciUQBK15rc7+gnSHj!-I+4km!(V+c0>4`DU(*^@RU*}=3q?;
z%sU=uFkPsh7lsK-4(_->=uw6ZDxri{#5FXdyd&(JE}5f(J%uc!b1AMKIqYOW9MU`#
z1QLc64Ce9LQi4$1Ni@IjLr?nTr(g18Cr$Dkt4ifU{I-?M115j&J|#vcOz=#<f|@k6
zKstu@t%pT!>@o|NZgULoDc<d7>{1cS!}jhZuW+tg`-XIm-F<m*Zf3R_ob6U{K2T%h
zpau9+tGL^O;G}XV=lUrt<@Kth=?;WX^+5QL^zX4ec$LM+I)Ii5%((rTa2CU@hrdz-
zShr59dL}bf-q(N4DN3H_KQZ+btkI26A1UvqxgM&BBf3RJSaxUHc~4)@5KW%f)zcx5
zRaK#RQ<|KE(TosE{X<!)_V;<+SRfN;g?Ffa+!*;%XVvIVoK@#?X!mu}*l-;}(0E&w
zgt<RfCq6r|d?oc+H4bn@r$tyX+*#vLU6L_bt&7g8v`v3{dVK$6g=%zMrUeV+@`chy
z`loiM9L&w6M5|-6I5#U}+;XSgxa?*w^}M$Jv3JdrXaA;q4hf|3?Xe50Q)?*6JXg0}
z&vx*vAZvjP8+wj+ZtWMV(s(%OXRlT|lh#&<oA|F9c%ItH2n$Uo<V0^<iHJEekd;s!
zDJabAJA8kQ(S{vUH_22w&v;-QyTfh@CYLZ$59R6Ebg)X6fi*GK!U<D#vSpnv`)o|n
zI?2nA@)9>j87M`37)KFjEdil5&<PpWjcO&ftgmXTs;7cm^FtZ5TCj)vxHifF_SK+s
zdnC7P*eLWGCUwiFk~teElPlG@l^QWvAjK`U3~GOBo;6cj89kKTtpW4w<9J1_GABlp
zQ$NF1mvdE>mvI+G8s916bru&n)zpGxIO010%#zKx8XqgC$2`U5nHg2D)Rcah=1R~Y
z4H<&;_f#r!Z&2IfkRM;xcB00ri95*YZU+mMebq#sqIE&5uNVS3bkAxjmfOos$NIz$
za!-FeO{>oNRI5d4)Z?aimCmm|wkKu*y*MCh)n*6TwF0&0^l4V@cnHZ?U4vJ9YVc~e
z%VF1<Fm1ybE=;PH<K$F8R{EFusLBa>1Jr)R8!Qc}JM;(Gs;t)b#Al-hcpo(Dd~64s
zgt`z~AY+2hEn0MZ`j#G(W!Rwx#@*{v>#2VlSi1&ertM-IH%kb(>v)x2uDV9Hr6jg$
z4Ck-V#)I#iSb6c-dvT<&!;5Pd$GvvrQ3_HUp4;lQZ&LE$x+ZA@(%V~M53<C-zLkDA
z*^&Qf(%$6_*a`}*6D#XnEmN<w!RR%xmlvKpFp*(zF(Jo7vn90GUUA!DDYptSd3|iC
z_b8a^&%DZwLk2oloWE?e*|+%<ylCX#;6)>6R;|qVo{@=bkopgK%}5Fj3F=$FW;D+7
z^4E+8if{CqQApjg*L=&UC-{ml8TEv3ykpcOzws+Z{bV=ZFd8_!_XVS=p$y+(hyMdx
z`L=48!G9<d0yi+1an=PBmkysQEPouk@jhSCKOmZ0aRCG&(pY04HygSIIRrxN9IPPk
z<&fV`6<I~HIo+C(Y(L}>1FfE}CTrhVZM}S3^z!NO@O1caa(;;_fr)l``nni=q4BTL
z2n?`9V^nZvd3ss?a;voscs=0dwRR4#h`$rw;qJzngewfbG52_xyPN0@LVq0YaTeE|
z?u^a^$ZeC?v_<3d5ufpg6@-rOBg_y%A{2!cPiXl9dcHG3-aQiOyBnKKutX&Zb9aAz
z`tw1v4z$XFe$s7zdVD(kc9=Edb5bG0^89k37v34=l3h%~yIp?&`||bhdB)mkY1~r(
zfL?z8woH&OY?G|fCOQOwm48i)Mnm}}gsdFBsy6W-wFVT<qY^x-(UVflBxHz4!MMc`
ztwMMOLBbzS$!t>fxYof;)uMqQS84#AvIpiQtvG1dVp7?jVbkbSxKRW8iI23qV$r9&
zo1n>y;GyDEo^&ODiX<z@mrl!CqkPCSEkn+tC;hwutMkgDrDv?<T7QoGo?LBHb_+&r
zh<;+hkW=;qTvSaoLW+sD9m7Ia5zYjqm;z)6;)<O;L)&N-*fYWOk@*9+^go5a6=YCR
zb5uymVFWNHG2t7c;bJJbpnOb!bSSK;mQb+l`XAYO1L?hTQNm;=9FkKe<{9JVa6pnZ
z^d1CiZ~5s5>}5fL$$!7nUfL;S^-cB?^t#z=1sUZK`%F}Fr9!c=vWq3yRk6|1)*5rM
z5%Fy+=8^zUd;vcb{E^q}a&lL4oT;WSTY$GjHZ<Sayp1u4DxftXTNS+PrmMRpL;KU4
zv!}S`gn7!GZFR;zA*FfsHA?9`vNNPTEmS3mB8Jkiua-n?`G2OC(D&25*dn|`HBou&
zn<nhLssWY#87~yUr7b$Xy|C$|q7E}gW46nVvDAiasUlSklN%Q|jHLOsBBRIjXcl~>
zqJ@>?znYxhE1HSQ8kp(riN+OvT1_1)aaGLAy$KBAMxI4oX_0`vZk-fH4=i4{9f~WV
zA35ywn2`xeQGZ{^SiYdC`A4e0qmc9AO!&7|eXl*5;7o<<nlgteN$&M_#8EF2!qh#S
zt@24F9h*D3!BB=Zd=f*jrlPA=y8J1cT0y>SB%?JI@>+CBk!vRaKwLVCWij#a9)HoX
zR@h&2uT#g)^&qyrQ7Emjo#!oid(UrhmH8-84Vxk3ZGTqTg)`msY`RHn=qoPQ<J~TR
z>oYh!O8MTkrKW*(<-gp)?AB}UH^2s!QvUh%x^xb#v63q82RR1ss2*W%uff%9c@!TI
zsA(QpJT-M6kt8hE`H8(6U}nhCyrEaS)HhEU$*Y<ycJR13=nsbYN!ND9?=3#ws_4zk
z=U(ygL4Rc`tn1a+6d!xR_b}%b$2zVO@h@mLMMqoHic6y*A66N}4Ot75%$UcBD3(#0
z+EAABHc8%tI<KMVDXi)qMbK6jf3Ss9_OLUaP+ZIiXFP4^c*e9X9UkqK#RK7lsajDa
zql&;X=Jk26*7x?dm_eQ_M^N!lx5BVHMB&_vPJiGbPSs-4a?^V;yzL$+xkPxsR_{MU
z=B?(QJMJ@o0?gH5$D;#dE}nyTYj3+w!1ZeCWA7dOy-Wz%Z+I|SIF=B35a=cCbL{W=
zUE|&1`LLgydk(4tp?CNpbxNd-3G+$|Qk_LZrrOJSXr4k(VW_dy+Onqkj<0s-wVYQ;
z!+*Ut3+`~+YZgD<51mmQe#04^)b@=1Q6ATa+rdtub29d2jN^5z8(kL-r9sCjTk&4;
zr@)ozxEf~4D%8K((0}nIv+dW|gXG|6ziwI%;;$T;4L|mBo99W32Xn~S|0kMURkRON
zySMAGnBF3SR-4cKwT$>FQ<W(_Pv--V`+tvIzw9{L=e6a1DK7-3fjuoNV|p%oGTfIh
z2X;QD?1X~|9r5Vv;k{C?cK>Rz3yM_pez}MDiw#(fta-24tKGj`Z24^S$e(Noc2B<x
z(~nPQ5n^LjfJeCJm&0!hO7&?eqG4ZxZVJnP76RiYTpY@*)_H>|5qr7Ahkt$h`hTbA
zzZr!a{!Kgd;_=*Ua#~*MBLHs8)-B)pIVo!*q{XO=BGF7z8H<W;8ynzxId~iKa0zLE
zekW*FK6x6Jx6S1{P^43*5Ewwh17U1Xq4#|R3Rj&(k*TpkPlSc{(j)GdUGzxT_$*6t
z*FToS*X8iJ=`%b0R|eoZY^l>IZ-1gQqe=yYN4Olb52B77F?a1iM?o9V;t`!Dh1hGT
zh$kExzcaNatUv~V1%$W+gMj2d!*>)g$y!yywO2vLeL=-QPH9-HRkQu~^9yhLhMPVO
z_lo9*q8GcG<;GzNu(3FJ`M!tFu@mgHTvkiC-ru=a(1FWwZ+2?QsgxxljDImUAw?EG
zF9VBJ)SD*j=d0CYi~@iw^fMQ*t{0nec}Pt!`$@2fpq<jT>qQzwT?@2zdAG5sWtLaS
zr<0>Vw%d)i-VM5sLz{Toq`A>{l)((OOF40_T(?Op?#2egbK7_v8$Vi^xFKi$b8m0A
zS{Gh54%gB!vViO4z*XgQeSbS%w0#B3sOzTLE8FYY(&87xNRy5f+$J5l`K)by?1vrC
z7NX-4xA(4V6l92bl;*29UJcveQ^oG5p{-2~>*Y3cEYBO!I5_@RH1^P{;rP#GT&9<s
zKp*PN>jjf46L=aX2V*d}CEZ_HHt{(@@~vzJ6|n5)R$8#k0rpeAZGUWlRoQINRoU#D
zdyNzV14wxIA1a$#L9mgQBwZ0_Y)bxLJ<qErC==sPDWCNJ8|YKyb>&l5Pv1jo+Pbz}
z4kS0rWfSg?t(WwH9^#et(xN|sv0DQ=S8L1_pEI1bS5$_&>h-B)!!^`j<^zPQO)PTg
zuWw~kQJ*q%?ze{Oy?;(uI;Hx}NAHT~ILfgaEY$bV>Bu~lRv0~w<oFqe@tYCe<<s?8
z-?;+b_|o6G_2`6GZP!QkgK0SF`WUE>y?Dges<5BBfL_(+VbB_y#v1vml}%dllwPx`
zNw+h_t!{$r^tIL=YCo^gXK#PJhWB8+zJ|9CG<^Rr6p&Sf3YWrvC=vlTmw{p#6PHe?
zDk=e5mwBivEPv9P#~dH$q%5buCON%5$^3ozc=~qIZUQB=vDQo<KPKQA$)6yta7Ir?
zqEM2jeonvbq(mbn9rz(Bwc^L}yW?-Ye^%sp1@H_U_P>Apl8GCEj>xVN*V+oQ)Dzdp
zk4(hrl!+e-aSi;qw1V`=TAMtp@)Kz3AY>Lt(Zq~3SbtoVMoIzbiZse*RgZx(^RRlC
z<zf}-C2$X;W*6VUhS@_WIpw`j_R}r{c9`v5)w<CGxyShKGtE=dGU!gW%9Uq>zJY68
zsXGi!<wr@Twzo<E8AF@t^cvV13br!?TPWlABpE9qm8xsco=Tbn*gtC@G(^8waR=s8
zjffEy`hWVR?WWhXHlYwunxhsan<E-*U_5hjPJmVA%%hy9X-zN8JJGD?mRE%HgJ5;!
zvlNtdN;ZkGYpdm{(&=U*zIMqWN-70v=WvpO_;E_B=^u96<>>oc)e6zERyb;<C|eA5
zXBoYy^3JDTbm+9R#RIaPWc!L9SWqgvj>V3W27fus6zSMWg=YU6F)z~C?sSjM)&W{D
z@q!TLYM>(Rg#;{AQy2=n!~6a@*vDpwNhWmG$x?v1>|~z_G^DHm2eYs@m_<7TTU;O=
z$GyP<>_t5auN7yw)-b=FSj`nA&=^75tRivdbW|5sqcm=2WaltfsF(weoZx%b#hXAN
ze1E+Sf4zP@{k@169Eb$a^vBPqFir}D3dFr|z%cwd{doFyBwCWLkh~>*Xqh0JjlAcC
z>?P|VjnR5$JHV)Fnh3N&UsERKkQ<T7Cy6j16VG5tk)$w6j}#;Jm>oY-&PY*^vwTGH
zXCqx00zO>`VQ5Yn<!V>}OTt(fSF42_Jb%D;AX!Uz%`#nv^b(%RE1|5p0#8QGGI>64
z2#g}v*~t54%m=QhF8bs;a!z<$PzxyOHS{<M83{2`^!VuvS;Mc4Y~Y6of{b6_cU*-3
zb6B7vfeIH`TonVFO~_-`i6c+@h&s(idOIyP;7}0Q>H)e-&)Mp9<WlgHui*`MU4JUf
zCE<K+TQOMBT7bBeH5Az>ZSfDtIOR-%XqSaq%0}xk7)T4xX?UQk58F9RGgi)xyGoIq
zrKqxn5tlO9qjCpFabf)uj|;h3l~eVf${VN@bD@z7N1kWYcGs8otqV#8UGg*}Q&u$;
zDoU;v;srAzP=}V=%9^3E2z3H>g@0wu&)Cxng>vE?tYOm%l+yxWRCCJ-^=qiqb7_IC
zTmc>f{x)*cfVhGYmX^SB;bJxlmkd&ti(CSWviear=z5A<)4O!SDj4%S@+@22SLAnv
z?3KnoFuxBJBSgo1tS-apXOZjXILaAZ%GyBVOFNj<sAnDcfRJ+I#1FdOzJIfUEevp9
z`>8{+zhZAKt$jWOR4ctNB(vm#NLPgm{X<tY+geWdoqz4Rq6POL)o#(Pb=0W|C;{Hg
z+Je;<V(c(e4_##%yV28QT8L4PAt04eXnp#ueOeDy#HgdC5%sYBe98{<+wq7Knt}z@
z?v`XzCgX+XsTG;dmj)MQ(tm&gaA9A!Ml~>8VK0IjNFHTx{fel@`CVxJ$f!0_%m*JI
z{G@rFtCMe!fXGQr`x*g3my4;6v+UE=EOfc-$lXOqh=tdMwPq&u2(QtDs{!e``Z2=L
zCko=b8F6wltu5aur{|pRZ~RblfoIesO>$0g9^I~QqO}!<K^v&T`hVRT7!Y3CakV@=
z(lufVzO&Z?kHSfMu!r=Q*7dgPVcrlr4tI6Jxr?J^w4Avzo!nQ<0Q<bcc4@r0=pHVp
z&-ayV%G^>6^yY|d_n2(V4i*x9G9`ocLOQkBJ;QI4KU)^}xOPFo=Q~koR@i#4QjXc(
z>=OL!4X6e|NDGv<yMOPH^$aZ}Oa8F-6GQidM4E%;o$~EP;MlX~V)V+Mbw%oxU44tU
z{CTFzZe|7Mg-~~ELB;iy!o@9cXl(hozM`H2AVJDJFnJ9WH@n2G*mp-};M7I7R&?h(
z#Ci+YC%0p3iSoUhZ$Biak-PfQjSG*~fNpWPriweyF{a>j%RG{v|DZDBY^Sw<pl!_w
zZkg6*lDplsIGdXP?dm?eT8(CxetZMNR-q>4xt(n#tegFpNFl<J_Fu@*{^FOxe<%|H
zH<vK#ClmrXGL=CtCYL3zDxrUQ_4mWyAAdYaX0qqVIutaoa6}lu26`|mqKb%+7BuFd
z;sQuWk&qe}pDRQhf%ad-z`yVGk^G<L8(Ua0fuZR&zUOC*&*!|7JXgpxudJCyA_d1Z
z64`q>dsx~y<%}Q+_H1Tc5ev^=-gXseyItsK<LRIcOCxdj#@v}hq78ptBMTa}@%fGU
zkUSGI7w{=>kO(&JQT+Mb{X2a;{Wsq^P(s--{PX=OzCaB^do^GX9`x|z-{I@ipFLJq
z0!cC%_{+%Q$G4#y`NGx^B#A{@YfNJeAW2p@SOWr;s$%(+8BE*9+o_<@KHjD|MwDb%
zrpH*+1|6&3nMH+>&hmei$ZtlzO!VAZ0ao>xm0Y$8+Gy*bht@MXXn>SY5|?mdnzlu{
zDNhv$68Ds6i_si!mmld9;Ec9TezcmSsib!h8BQ3Rkts&wgf)7GQExef<sAz~1e{@u
zyJMDija-VAFn0mB$V8zPfeh-hOe(Z7395&->@*^kdUSf`nEiix`p>R3IYZVr$^KlB
z2*<BPZ!<ELc+oHL!l0@86(h$fM+uZgF7gzg)bzb6@6%{<=iTlvIhmKU@k5AaB<)fx
zHO`W!s!nx`lySwwGyTX7F-4G~GI2Z^i)a6f+YzQf#gQ7FAS(<-sY{WeGGR~zZ|kNS
zroi&Llend@M$>=953%?<<>`<jM&|3F-0M1T8PitujB!?F{#ca~Ri`?N6_aC{Tv@Nc
zl&F^C7n@O8FG=bxn-uL*8<Wc#X%6Tk=K^V-<6o7thH)hfEp6BAAWL<&gB4S#T+DFe
zqHbNi4^cm6`hc1;g_<C4QBNjTOv=`O$T>opQ}PBi>QjG8Cv`MO^`7OH=b9qdiaYUz
zdDe7|_>}ao%j_a$#P>5o826uHAC%XWKa*!I=R}oum-A4bcOmO?uhnV`<(AiuI)7}|
zwih!-^e7xT%WfwVFDkEEtw1em-X*_bv?c7S$1cwoX{y>qIi7UYfu}V!@5TKzfuWSq
zVvjxAH70)@t6|7prqXNSw4YIFN6!)|OwVdbOD^Z)+7NcPPNm{>DHm2WDw7_(u@#-O
zU(2J37CmaQd9UyjABG?oEhNJFRxSlE*E$LAZQ>^j@9!0>ZY7d5w^ER_eubXVn%QgG
z(YF{kv9>&a*}iY~Yd~qBy3Vat7j;)zL)%U{LtlT@XiZj0smre@YcN;>Han)_9mXBY
znhx_uS<^xPqOt~Qb>rO)^1P;;*pzv@>b>QK^CGAjHhFBVgi9K<T3F@R_qlrd*-WD%
zhkcz^q%5l#nvqMwFACBi&i?2+-7TCq2=1<0R@g=m4(<)IH`Q#?3rB@Wx8kghy#XUQ
zp}>Fi7*ib+YZ$#;4N?^tobP9?$w$kSF5DOJ?~+fu;vVv8F~(|gK|VKOreL77>M_gH
z{TNYEc!3$Mc7@h$yX>p~4)a14U1wMnwBn4r6%C=l?1ElTwr&qCl*WMzJ>~~A0pc;L
z_dP~%cT2UL`g9mCNmZgi1O1F%swH@kHF$qLo+jOY=Ar|oHIro6w5eD+r%FC@;ryjb
zCR^CmvzW&Nwss_4<91AdI*ijm%;nf^$q$GiU5|A&`Jwq>XIx<^3fgkyQck;is`!9@
zkhO>48}BGb%(ze-XFbtjJZh>3I8mM_es_TvU9^zkB3q>eQFZ7RK8yev25n-CBwT-?
z=MS$GOK0Dj{GyTUZWanCrP6zvMTha{r9a@EKt1CDVC7IfJM0cR$+j`w$5fww)p#Jj
zkLs&1%CS_$5Mr>|P1RKhLXF4hD+%9Sn7tQjJWU3DDgAgzmlc%<+WH-Jxx@GqQ`Mr&
z-HIE!+(GZG%Q4ME8Gb)q#+Vju=H`EQAl6!7fa@~p$5a_W7`JEFw8MDg0p?w*+^)E#
z$}RNXsw_zgV17GQ1|O&*+IF6IV0K>EFm@{43g&ydpr9>H@QYngyHgi5L2@^;roG&W
zizm{$?S{dq^2uxCc#dtZ?~Eqljdy1H(|HZ5Kb;P3KPU0(Cr3(d>}zVc)Q^9%SN5YA
z%a(LN-zFXAtr=AXPx|*(M%0B9db>^e)ifmoD;>Ymi}jVJ2?Z%;(|`3kmOazkN@M4@
z(%;_1n(2Ym0{nP{wa`tWjDDdbl5T&QuKz@^rso0i5gzjFNtY$Ba7`zNY3b;Wy;&pi
zHD3peSWDM7bF_=*CaSmX+T4Gtxf!pW-090W%v3hD^K@nk*ZW2kv6O0R?tCS%T#y@4
z@pXt$(!~l<wgz*o;p>04O)+b%am@w8>9*#P+uLE&Yml2_oUWXj)SUiuM4#O!)k1~w
zN>Ioh?$UMVg(n^MayMWa1NzX(zo(C<{{a~wpa%+NZe(+Ga%Ev{3T2nR)GY=AH8+=$
zZa@<QF*i1sAp|}sm;JUXpnoiVK;8y=`1|#boh_5JCi#4szx1!?*WJH6J!Q7Zcy_}-
zKX>_oBQO+kfM5-I7=C>lzVH6jum@pQ-kCnw4!?d3>dx1q^<+is)_dkk>tMwcZty-C
z%8w?kyC-W|X6Hn6?opcP7@?Zw=a&eJh|Y@rSR;y>EQ=nU*w2l~6Mvd<d7SgdIQ;64
z4F7=ay#bFJ8ErMLL+{{>51@NfGY+}2pL{IO6u(T@<Zm7yA{ghBmI$yoK14PF8_j25
zV;LleX*^kP<M-1vUd}h<H$0c|T{r(#+q+;Kg`T#r5!siq=Q&~d{a7UA`!GGrcHkxZ
zJocNbXq_S5FjtvT5Px`}0kWV`YIW!98}w02#cba+Nz3+}B|DyVa>k>hN!XyB)!NI$
zCvS7i;_T1Q>;1Ww2Ttkd7R2K73h`+JVlpYj29cN7965u8jThMdYs!-G7Vqt6wbblq
z&QG2(t}#EYv(e{Vi7VsUTlTk@S8+3!>;7i<D}1eZMlEZLuYb+9R-=DjkLb`yg3*kT
z72@nB_F%{APKD@AlI>i3%NX~UJ($IFZEKUq+coZ;k<6t*dRmqn&@eXITpL-Ucr?Kk
zYae^Fk=AI*-Yz$h3CVTz39@L&Hd+L~9x+D{z&F~3K(YsfPIe)bgp{s0_@OhQvC(xf
z3rYCm)W}8KB!71|p)yHg)hsRu>k&PS`Fcn9$tiwl!(Ln+QYfdr7;gwE6iN1&a5khc
z4abjBc6H}8K`)yhCp6yC>4qUoTQ&!rv{kNvk14K5tCE$etP@{q-}3X-p(mgA#N#{r
z(b_FStmH0edehb~JFIMFJxeKt$X7~)F`s$=RGM8h(SP<V8x-Q4B4@-AIj$%f!6ZRc
z$Y)BjxxiGtM_ul>tq799>f$k?O0!3+j75+=Er#aRK&kYaQ#YK_Hx|FgRCI%-l#u6;
zvd>#9-P^SKmT|KIn+8U2VocSzVIe-PK$h=x2oXSMfR`wbYYdVL$pr7K(C24vXH>B(
z4)s_&*?+0}2{OYosl1r|Mph>$SR6>RIw)ZlDc*(EaU2XVbX=keu_LRakgu>hs`njO
z9VPJ!tE0zve^$qlWX(i9ZC@jPAXdkLq<X-2uC|rbjjWDBdv{hx$?L%CXvFQTjz+x9
z>S!<M&+1T;7@oqN{Y**i#Of&23;wp&!<2+g9)C9ZgC6r4u7aX*G3efF*Gu7OF3;x3
z&w#<kWNWAjmL!H7C2pawd^FOUiL9q8FnAL}c&k|ojnm02h04*hITIpMG}fgKb|LEx
zb2I0yTt%UA^lX-wFd<21*=X{xp^5_*ZeR{{)tyth<PA}~u#Po0mryXK$n(~72tT%@
zHGd@(og~of>w^+%qv(iY?IowI0IXyM8vdulw!qOa!o9bJT57Ltp+<d=ZK0OX(B~_*
z1vTa?+d|!Y>$ad~b9q}(j<R7}P{!XDer9%GPbQxEFQ9~drS{vy55{EL^>d6qhf*o~
z@0WW<9~TV!QwCf|mnsgOM3>k<x=4EMoqy|C$SK4R!Y7iI9UXnodDe&>`9y_$g-=wy
zJMf8^&;%d;cYI?0<1gCy4tyd8(Tq#)LMY|})5qfxo$UYnhFF|EYm81afjH`srgre;
zSiP6;r1=-m*zs`ZBBh{5!2Ez?jEsj7sAyTJh>70J|B%=R^oy%wTIsXZZ>2wj6o2K>
zH)q<MY{HK2!hRy4>H7af4cL+UtbwmkpLORuFrT%+E5xT7-Tiq_iC6|0JDOb~ejwfx
za!F6_&V`4Px{>!(Xm8;?PxW{4Pw{j5n>}0WZ!D4C<Qp^RsCBUzY)jfW|8~c8)$S+%
z@4W`3I3e)&_KR9-r-D?WzU%2TC4Z~K=`%h0?WfQ57`LB3)82CZ^qHR3`qO6`>GJ6_
z)tmRLPo8zO%T&W^vut2h3s?Cz)y#b4cGZw%(z3RjWxSEuoQ{-f^=vqVRVd4T79{7S
zbZAt#<q(!jQk38gD~mW74`&zF?G+*r8v(7(V1+lI!Kzr_s$y*rMr3=+kxHDP2aVpc
zI!owFXSm>=XSe{!bfsKszz@Q>Bp9P=zggCROBH1Y?jy<re>^$lN}=x;oMa{eQn_B!
z+~T@DQ@E;cDUAOE<+%bhm%)E169P9fmw}Hp6qf_ODk+y8rU(ju2gEdI$l(Jp7VxgE
z1JcEC7o-YuS6z|&|3!u`Jt)a5t&}aa(n{nE=Og*eLz2z+MmAp`&L0m?t<sGZj=*B`
z@vQ-VAo&+0-!#-%0AUm~A3vHupQS{@$C(c;rBZxYzIJ@3)1v~%Er6c@w*WXjS`E36
zLOx{bgKh3}E`1b#@ZnMmq*|gf`KihW#-}#ORyMRkZ-ZuT0~AJZ`s?HO2bqnmEizUM
z?pG|L7b!K5S6}!r;W|hR^1>Z5WfYpF6Wc~EKk<C#|GLf3*aottH8C!PX&l0vFqClm
zTK#+bc=&fRKuu0Kt((7pJcI{Yflv+`luDp-&Ch?DZx4Te)MB%-Qb_KTen>Vyzc<yL
z_dM%f@~lf^h(@yxl9z&R41v&y1ta~M?QSzpY&XIP<)_xctVAt&{ZK8Dr-|`lNj9OU
z+l-YC<19aM_@g`TgZVT@fWcZa%K6#nSvMBy)@m{izHv@#JXne^=~|i1<3nUk#@8AF
zX2*v}!r}yf`LNbWrS%k7oy_Ogyz0UY{ts_?eAlP{>fEE}mioxKA4?>Y#~y6L{Qpw#
zq%;pSH$Ru-Y~?LS&Tg=kEwU^gR0Ba#)H=HJz5<W2R#ZKkg!wF*H5E)2R<siyC(;&C
zYfq0x2F|2&#+~w0U2r{g9g!npc>UiO+e+(BY%6Jh>#^<jYS8s^Jhny&OK@*&$65on
z!D9wA;+{3x?@_k<(s{*MjGYx)O>DV!{ch)HW?GyW1+s!|i$Kqk{6r$%F3~u_#{{C+
zc^6`hu-4YC=a=MWxRf-4Xd+#{-E5r#3TthGm$_XcDnXW3;-wS{>GTp_JC~O-5Tm(z
z)5uAGh_Xp)2xD-vB4B=tDW);!)C7;>KuI<IPt-fQGYqy0@em5@aiEtsB`%|uw=D0f
z+>Uoc)uURM`6>22Y=zbV^r`r4^}XqHto11R980~`McquFmHM~+^f{Jz1btRxyh5Mj
z)<@Anq0htc>gcn=>j3&(g;z(P^LXubH-$cbt4(tdeO8#c&*`(mV+VbL7anM^wpY<7
zuace&x6-FOoIbU;!h3gMhFD3YD_O79cgYSlGlD$S%#pxh1t|)4*{a3-Y&%Iy_E6qd
ziYwCT8#`g8**MB+rn7yNnWw8!zRoRN%5&IHXu0z~rSfV5<47x^+xM_GBmn!+E7UuG
zA@c5Yx2pWi=j&Szy)g2*^50WSuJNIO#Pc{)&DG5CY7oA4wc5lie3H+(_?3%G-PW`H
zswX%-DjiF+DwSu^mOD6zXrhedj;@uNwFE(?kCO9VOC{s3sA|&PNDV6Mp&T`tu524x
z(Fx(X-##6YKqAT1vMk#!>MA8*H!S>rb_HTLl-p_3b9Z|LPhg&xQ0$9=TT8^~NqCMX
zsuHnR!f1b&3UTm_UYi5Sv66UTg}9P@r9xb_UaAluT+va)57FH@C8@OTtd>>MuBRfg
zvi^2+Wxdi@z4zHBF;(AdiLW~(a-?V7M538@y-EvII6?4+s+Q&iUBi*sXv498nK=@h
zlsc}2hA6%~s5gilZoN;Vf~Bp8MhkYa6Uos}tfH<GXVly1s{b5%&Z*B#^m2khuCXIt
zQ9OULuVDi#^L`kwU^{S*uHXGw-fq6+`_vaFLd3AfOnOW2(8_zA(!RLbsKcity!8!H
zWpwQn&5zv+Ix;=#UK54abUp`vS9VA%>Qf2}FORc9+>V8pZPo#Y>|{AiQHldS82W)a
zScs$DY+<rYW#H~&VL7Tz`_wX1a3o<Zhm9kr#16#35P@HR`;y`~t>=57U}#1?e)+O>
z=`Ok}V5yP`AGLr%rM6Cx&c&sq&{noNe3o$|Jp%IOv$WWz>$M_GFQSfr@F&^N)&6)y
zCX(R*#o3+O1P9f&^rhKVwoLkGTOJj2!nHxrPCys?9|fpHcEfrTpk^naW)GldB4kaD
z_<1UZX5=rh=syp<#OOPhlpJ+Ql^G;aT@skcEA^RA?xGYu`r>9Ok#l7PQ6CJwS2wPz
zUqFs02JJ#cqnFo3s-}E@0hG^}6W=8N3IT^+g>1JbcftRV^L26uXau3TOLm7^2~9_S
z15!&Yu|=C}T`TasTJhT+8*$7nwD0S&A*YkQ<fvJWzwa72;JixGw_d$F-|Czvk@R-r
z3K<b8bE9M2t)0;%Yj5;Jd$oi2!hl<)osQS;oG|7_kFi_1wZck&*kt;F+SyT9jjObq
z<F#{MXq_4tcUK25yR~aA9O_Nl542sW=qc7{r~e^D@S|l{a=n2M)w$>!eZo{sU|n5F
zIh0XjHk$G*NPlCn!6NapK6zE8##}-Gd+GwTJX&Q2Ug9$FOMVcQT>7}BPW4q1IW#QQ
zs7D32rTNfgI5dC}dOlWtR50~B`OGoNf7~KLjhDfHC=&uUF_&?=H6#KyIhVZ%4l$R%
zkS_&)xe<Q%U%|)gLnv}M<P3|0fn0J)gQkFyq(EB354IyDzB-a2$=6)}`hG)sZ7C}`
zw!ATN0b+5rT+VQQeDe)grHTtqLUc)8E}=RN_@wM878|HbL^cR6m3AvSu51Hxg~SHQ
z$?}#st+Xd0yNVU?1-n{oN@5oyRU3(ml<a1IiWsp8NyWt^?QU|Bqdmcxm)Ta$!Nszi
zM=lD5J&dZ0noSE#iB(N^j?lU3%H)W$J;BLQs<sNVqfG4y?5*S`qdF3;*ggije6(Ox
zOaxY^T5S=`O)=Qr3^Of7WP)AT#6nI=8lA-C5=Yxbwo$W4!6>&y3Z}_&5tvlCiKwuD
zx+J?B9EV^!6__`+eV6PQirR=UswPZ|V@|dUn2VS8fr0r-hB}sP8Wh~E6d^+}C8(_;
z(ZK+ASD*$*dseV$#(H8l&-h;8-du1`ngg>bEH-K`w{<kM5!R{EHeveLV&a7z0>}g)
zh7c9DF&GB!wY!RN$-#tba-}TqiYv;0Cahx6ETLGW#wIbtG1>NjtxIeks|dAN%o2!O
zff?}!UvZ?RiozBY0x_^duo1Ro>R_%Da}lP&;gYl6!~<F(K|*m-p({oS8DL(N7@fv8
zFS1iA?JgpmmJ6#>wiL18K`6E-@N5%?P*l5uIvCg<1SVC1`q@m4ENSzrt0qf-*aq03
zAaZIjPh*#^wm2}4R1%YfM61o1NCJ$^)aIs|iS~Z}dG!3Sc`mp&qxb*%w~Il%B7V%I
zlgsn-qy7C`W@U;#A(n4uy3w=g`E>U7;`nszl0D!?2gi%Wcs6mgy!`dc;^o`L@nY=a
z@{1SK$pSK7SiT3yH7PF;@Gbd&+l98QPs<BfG9o(dVt;>hFq@vf9WM??2hU%O-i^O3
zt~7r4$Hf?uKaQc$WHFvB<_0aLE%9bNpI**R$MYrH*jN7;e>gk-W%}jNHdzoEQt1fK
z&+sIIT&2x=I+;wD3tfrdJ8U3l+h`znHK1<LR6e@ar#(M<dwH_3tJi0LlmEEUFVoqF
z@vO-WN29-uUX7j|#^uGE@#*3)QT4Dhz=?eZAaM#_)F6s|s^w_()M5FzqnFcn)6sMH
zQ)EdsQ3@QjLQ#N&{AGWC#VWgKL8Lw+(BL4iKw@B<&q#n6)guXeT2umf5&khpLyW5F
z;Y!3x-cq`Xw|FP1o=|~*Vklk`^(_%lASiPJ7%8bH78<}3l6)0QeG>{87te6G!b{2<
zbWlhyfFu+mAh7~|u63}iLMnDy1+2Ba{VMHNCUj|McF6Kw(rzWWzDhg5Q{<}zdVx5^
zh4fPOz6qIfFCkOfB_WgCBxH&Y&pMC?#I-}UMyN$A>OR2Rl*J2wZE!Q#i3Y^H2C#t~
zl1Nx{bK(^*7I?9I<?F#bI22a0_FThd7uV#l;+hT_Q@t8+ho0P_CwFs}n^@AYt3e*X
zv%t@6tiuu)g&Qd8se)&)c*?o&TDUTR@6-T_hT<K;z6L-M1XMPu6nubPQ%MLgFdLvf
zz;aW7Az5NzOir4AQmlm^u0<-@C%{gp1(>K$g>KQJpa`hZ3jwY%lo$Zk87#tx+qD+(
z1bFgxEf`Ev1kfJU%1jL@44?&9Z>;@V3qhO!+O}!|io~Kp!NxKLC`DxeEf9@~ZnOX^
z_ch;Q3(&ZfQ3XMNsTOM^28d2zh6j<fJoOA9AmmD|#%dFPa-^>Xwhm*Y42q0lTq|-9
zNK`n)nlF4J=&Vlz5ds;8qf+hZKxncu@Hqu|Lp4^(YY5EpjY7LJVKWQ<vAJ#Tm(sjb
zDa|XTG<Tq!X$a(nBdvYSgy<^PWH&ko!~vJ=bI3U!I-5>>_K}A`T=4)5bn3_?AT3An
z){z<#<FhA!U96lln)D=^k%IWU7Ek$G6}>-e8B__xz4Bw0{s&RC!o9dEv`bu-vJzL?
z#Z|`xO2-4r5bQ0Xu7PWyzWZ415GNzTa2=DO>b-pZfU<_L&NoW1W`L+{RpR@UcqEu_
zXGWa8z_uiaLb4>x9zJ^{Q=p=Q#98DAGD{?%5Sod96LiR16F;K%t>s)v+cEsd&LWXs
z7D?VEi=^IUQIRfZx^yaCI!TvK(jmBPLh#M5!=7DSA!i^rX!whj{zlF!*n6VvmFv;o
zS}2U3UN7HV2&ew@%`5x=)2GGaV*c0B=;PVq)8&ajoqitucKP|@$%papqh|-Ne|_(N
zJBPD>_r4CbJE)kW&&HuU6~R(K0ESiZ<?DmL@2(+|j#aNU%mH|0hr5GncsZM%ob9Y*
z4#qt5jgH`a!aA&u?=NQK`Fv*`k(VC8<iPO&V%bFlkmh-Kyz+x|oS&XuOfP_J&t4sD
zQSiRqRZvgxda?Ef86?{Ht3-TTZeu&^peN;jE<|H(*48uJ4YN`~pHtHvX4+X_s6!n)
zt&!jr0KPA>YX3}E?{cPFRM(MJSCCb@qX+4Z9<&2g?EqD~a|#{el(%Njz;@uZ=U^~f
zaSa!u5vzT28sRmTm=IoQh84|t9i;8Ex{PKs+rcVm69Ec>y?CiUYzSGjz|aDHStKcc
zDDEsgNZ%K|ApA-o<c+<RQ=eiq4+7>>pRs^7p#&5HSwV<u9Ojfa0I}s-3G+|Vyj6+F
zpm>Vz27j~$SW5>qMHYL{f`mb@Qtbtz0z=}4qpDy%a67I-=RrAcJQ%q{y#iT`Xk8Kl
z!Crk`*MeeaW%q~K>FjKBJei*ScRD+N|KLx~ryu7R(?x?_g36<yqs#et_T=<(F@N&=
zcs9SBKRG}C-ycs-r<3_&c6^3Z_n#J@S5Vn7O22h>uKlxf=HTNa5XUumdq7)X`|wtG
zg15Tz%BrT_xS)k01Ei?sf-we*3Vd6o{4h~b^Ihs?-SAokEm1%fc+}S_yZ01-_WO%o
zXw<uh#<eZ%&G)T&v^t?P7S-gfIr-BVt9LwRZ|!$6)}y`2`7U;d)U(6Yi~alIiT%Ez
z!xQ_3L!&|xv>Viib0VTE$kuE%*Cn)`fxwMWi(}%g3|f4Q12BLBMphgr_St^IW37i6
zF~g^BcukaXhn0k2w!|7>f&mDB0QjHO<>?I(x0n86hd1nX!#R9KbjAWEn~WgVlxek-
zupg(~p64qQwl@)*W?6#^%JH3BgTW5VC2Btl$}VUS<qpMWeb&`aXR3JOJ500#yaWdN
zCpqa7>H`dgT}(^y5f1ydVeGwkUHZG``|SGv@r#Xd3T19&b98cLVQmVR%zr2d0XLVS
z4mT8+;p;yrmuJr^segVR|Mv2ioy<Z?5iEb@kN)lHW%t{z7ugyVfYa;yj@z{Yp*&7V
zKL#`X{*UR~?n_U!vXV3;{WEg<{cY-6zGj^VNjjI-8ne(jNHT<;tPMi(Pf6$cS1Aq2
zrb%!5Ko4mcVIkv(XAi3swIu7Ao}z(7k(4DhX;YJi4d!<KPJdQ4CUIhpvE8NL<UWn#
z{+imfe`G`O<m7Xjl(3=5j3#Me(em~4Q5B%_LVMRwSOsR3C+hNK%|$9etAzwLsNgA6
z)mhZ6SZ%VRqYGxP)Q=@6`5vr{ZYPr);vq|{S+w#mO1x+xiOYNxnZ$%)KCm?C+*=E;
z2Yqk|^2~VV(MgDhq?aNE8;nCG>d7vDH(8e)R*{BG20RpdCV`9H>rkH}x%f#W;g-sa
z?VnhIi6nKAMMFkS^S`K}MN6Zh8V1G-Me$WMq)&ynSJ5i}0e{r3ZwGH(_P)hiJHBVU
zwbSs8-in6S+v@+*UDsD8#!o?LL}}MnCfd`w!Om5tZlmzV20<*E+(ipNDnQ!zXhv^C
zGp7+lnXsj@sMHT$N(E9$$LZ3gQ<A>N&llliluF1ja9$$Yf{Y>nsF-*K@A`Qu)ZpAh
zi>b{0wTFS?QGddqJ$0!kT(<Nxw2~l4@3=&}WZ1(<5Hp6o7Ve;S0`#yaM`l9he+t~u
zii$D#I~I_R;&M|hrm4+|n*b|V>?xyMPu#R%unzj(YM4hVVv<+5rHI{LkTsBHH)xY|
z=7>0`37AdVRcu?nUZC-oCkKK#ndOgYEfwVjt#MPd?th5pdSb4F0Ei&84NGprPS_TD
zh9G69{SuT{(F-Ct-_ctKD5!MYY*<?UHkhC#B{>Z<XI`J3UQjpneb`aHZeMuN-xfKH
z02wAlAc|9a$pz6{5(?1rcIt#J0_uUfS)qP=WGhR>Q`$|jUC7HC+l!V~rX0&Xm%|M3
z=Ix!!@_(@-ucoDRM-iW!EmS{y1m==ocO1)3q`q5jRGKoHzh|xKDCSL}O|?Y_s=&kC
zj@1_9ge7O}dta;54H*fL>wm`>hGKI45f!B`gcX&ZMdlYBR~bW2GN@VRkiE8yI1~nP
z(8~IX4xJUwxxo$PcD~*Sdd~3O68`9eSc#Ar{eSDi1TQF44^!B#L`Aj6m_aQdudbeo
z;~nsh>QR*O$)_&R&!&0u;kP;?_<R_I)N3%&m7ZIUOMBkt<Z<cXxl5pNRVmG>SrOuF
zcJL_FLx}Uqz+!ufyZk^f?q*!I6Os;=o=$I<-(x3aJVavOp*5odsvoXhWR%_bLb&eO
zZ-0_7Np^C(=iU;Qr^9_8o(YH=g2T@EWmP9<w9snckAM0lZLv^6T^&Io#F?fU@rPfw
zy6#a-P*9=tRLI~6=6nEJT^~yw+aV|hs+!GSqO6o3z)PE(elRjcXM|GbC$Bm?u2*ir
z%-g(9Xf?=keg#@F-}!2C6#C@VUa!Z)<$v%-qi`Aqo!qpgH}6l}u#)6^LTWzbp260U
zh?pV7Em2_<Ej`?(s7P~(Hub3a8{SQW7wls{y$qXkV1r;gzrOE&o2cXGDfT7;DiHYl
z?-ZM<Y#PlN+X9fDiLBxcL(<dk!@u9Y{rUCp6j0OKf4HlhG-V42etJ)@hKxgU;eYf!
zzgu2jkOP4TGM;+-h|S8d{Q5fWG}Rwa#@Dv*+!loPngDSb7B>`KJfX{ko2J5>snyoP
z>3iE)W-~L0veYT?8&k<+`IW|(QT)nJ`(C=zsDDnoZ`1Be)~2!O;x@2MgoCBOHEjZW
zn1WWLrAbi%(2|zEStr4f{~T#;<$tY>D6F?yk%V42iW5$XnOl{)v@~*;CZbg+l&138
zw;$0%Co8qgD^S9XG|o`AbnKdtwVO0Ne}8??<KAf0=jJofp{S=AHFkPH#?~QT42}E6
zRI^tPEmHT=yW|td@#gBzV&SwFSsf8#`DJRRlw^CDh5K@`LI{jFr!G)PUw?RJMev;w
zvUpX*k#nNl9fG3hSUV?rpb%-!3YJ*#%v~2Ce=t4mb9Y@_5YA&i0E|F$zp5!;NBIRw
zWow09&c5X^=kB@q=&=i*eAqz&5doA7H?AnS97p9heNt4VLtHxbvfgc3lK}g;GjdE+
zzIbN=*F;_Mn&MO4X*}OX5AhTVPm6zFy*B%J@ez}u;vtLcw%nXPDIZF{fi-eW_uA4e
zspbSN-F0j4PR02=A31x>(N=C_MGBu5CzywLvur$vEL;XX=lg#9mO{EKmTrgDZ_&n}
z_H`ZC+z5mA#^5nq&U5N)*}{~SwAm!6wX`{>z?3%s75`V*lRy6l8=l^o3T2nWe<%?G
zH#C=V)&&!{@YpJw2n%`e^QWi(17lAss+YlkC=&rUmqFJT6qml?D$xOPmyqHrkAI)C
zyNBwXFuI{dlDH6n1l15UQ|OX(uvW&S9plu`5l@CMOyptxNEA*+J`K2)u_DmHaEc`3
z*$Rk=bbJjpV0J(Z!eMA2K5d6+jgWNS93sx1kFq4?);F>y(^;(MxF;zi?YNA49YiE8
zY^2h_JT*RuR&ns5iclnMHBmaMBY&tyjGz&veudHkUAt+*FCQ5zP<O*e+Q(#Mw2$5_
zpck6Nx@Rsua!{f8jpct)%%V~{lv!1URaRqs&NlqIibtDVV1&F*&nt1HvER2MgG!FX
znrI4K-FI9Wp+~6ZA=`J<`M(-rOmtwzn090H7(m_iITCkm;}vAxXcUS{KuWuf406Rz
zv@QQlIR%sFM(>pQnm|_Dop$1fX)0vfp2iCmX<j*!C6;O{jPr-G^kaONH|s~>$`V~b
zr*SB<6Ea~*4$6E8(F03Kw`b%kE&+c^k0oUw@FlB7<X$l5nY|{aUsFw7(89c?h8=o<
zoV3x7o&7mGlFLTJ1TDNA9DmhTf0&j6`4Xu)BBfW6an%Nr#-cJDOAbya#tYA?4N2yW
z;gTWDM{oxFVZ8sdQXdv+sPLmcVAjM*0ByIM8A47%jg6V13`(oJm>J`Mye)qoXl5A1
z)|9L^7+|dct2pF^ybGL3o`ubx%~2kr5uxL&M(jFEukzCbVZq(GT*I(N1;sE#)fICR
z-<QUv4xIBMs0P|~hmb?4^^TyJc7|_HA%*OfQ_U&tDWs@Ck-TppE)-I<5sEaIOMyII
z&M5giGB}vRn;7>E`U)aQH-&%M6x^;${UDcw0W5>J%?a@!WCHVs&}W1(&E6>a&I;!k
zhn51MQcjzm&^Y{AftOsY3ru?C^L+hEDxstEAJ0wPEmkEuza2c#n$ls<_OhT?no?%1
z{TvTQd{|#aBdHS3z?|F<l<}MwGS6!IxX6)7N;#GN{razi5nky7EdYOmnDih*QlmOg
z4X@OG8xc?hqdC?cRzgW53_=5Or=38ch?xVuZYZ|iRs=7k30K*&Ld~QVh>k<t;bdzm
zoI`xt0TH8IHV_w#0R)PUn}ar!0^_)BF;>Z3{@-mlw5MFfnq$lF`f3J~FFwbjJX09m
zQd-kak;}%hnV;j{$Pj<-#u$SHW3OU-Lla7dkNA1#P@Q4jX`brMDua4M$UZjj4F6W!
zRqh-)g*$FqNh(n(4slO{Py#OwaYuqs!ZQwWAweiNnOydI+okScyC|iRHvBlY3%me;
zd)hAP1fbe%ySCgAsMvEdgsRFMSL7qAoYow}`V+mm+Gjm*Z1sOKaHzMXVnu)H(O*SG
z?5?lmJr0)fh)W4gds7W3QaNRz`j>VPZoj4QR`}4IX5VFT01W|z_jZnk4IMm7bZA5I
zkDsHXo}Q^U?L*bIa;lzkEu8oM&Mlm8ovVgh?R*t|7}w}%g>rtDYsXc(ky2{%Kpab|
zAj7r_GO<rMWt)ErG6wQf_bDMOL%8g{<0yu8gN_a{lon)_IK&%rJUO!az4zXk7}xGH
z&Q;jmUB)2p>BlPlVw_{oyW=vx$IB%#j(>Yhq_h_Z<D-`kZ&?o-w;2qr2t%8}JnT__
z#nH)-0xK}6dn@r2>p^X%4C>nzYf81BQ?VoAF)BF*262B+!c*E`Im8_aPp*h@fD8Ff
zwPsAbruoj*Ty$g0X09snoxWUX*(9f`o6R26v8iNiIMCl+$0me$pKAW^*gURdGe#T^
z@KBcKYu#G<jL&z`r8$<a8~Z0GOK$7Z95eh@EaN4|cWF{)KtY*ZQM9OyjhDJD<$>Ln
z@fOzwx1xW@(e4me>~Mb!1C+jd7w#OxFsug^bQ#ne3c9n2^!C`MBn_N8xjSiK5O)=G
zb9tCUeA=CcaEN#@4dHTk`6BA3U}txc7gX%!`C)3j08RwETyec>J`bt;#-a-WUcC#v
zu!}9V;MmzV9PZooeNOSzj<OU#_?t_fK=}gs5)LnH(%qR7W<NwY_}JI{`g&hK(r#l^
zA2eawChu2whPs<OLtDu%-_Nh7{{XqE<W`r#e<%|HH<v+G7ZaCa>MGBF)Ai}Ex8D!C
zZcD4^(%*EEcdu`Ue-Cn#HhN{y4uAhVqyaLZOu!+aa&h?e&+z^5ha}orX<ku&tR8;-
z7(~yvs^zFz%i1|>XIci$s;9xZs7x9vTBfgBTizS7wutvltRaNi9>1IdoTW&b?PVoJ
z%XTBCXj!X-<v#ys%GOSQ`eW%!oU;VnG~3&9Gp|JgMQLyCCvSPy!&`3BAb5qakOpYB
zmp&Uo5o=s(gGJD%?Ys>sG`QrQe3P3wF9SIh*~DU%K34wg=3ccmc}2$WY3n(~i#FV1
zb3S!Wr*mhadQMxNX87B@!GgktxR;#4T2|;%U3y{+YliVBv!3;TOjvj28>h&J&f8RH
zc;@D7K#4V|`06a}8fy!amew6b@Y%sdfRuH6ddQ4m4hF|!+U6MJ{bsX*C*?c{CM)L*
zLNcq|sy^OlU2T-p_`h2BTo(9pvSEmd5Vbp(M6EN50B5Q>R@pqKH{Nv3lNRfVJB=pE
zCoZ4al=)JQ&F3+H8P1fG+|Oq-WTj$VUy*6`J3q+=RK2Q1*fl(036u>T4_qZP6q%oL
zejnw2LVR42J+J7pXF6C<L&gcFQ!l&_3a7eq>^aXfUtt>8I?ogzk0bT!l4t|t0zLrT
zf@yBK=>67D<U)@IQ2B6!9t#?pXm5W0&s)uSr@S$eTjG>|c9<ed=1^lhK=lE3=b#uB
zIdj$r(wU_xBpi)a2<7xLA)Lo;HtID_M$suC`cNlxl~#AOtf0{E+}xe=+=biNFpsA6
z+}7v0FPY+Au7r$Qt)uPQy3LgwyAT-3jW$`IUF>V!|3-*TU^wWLgy<gFzu`kLI;AwH
z_r&OeB5SCBoXqA{+c7UVHjYC{8TI)bTobpLZ5DDAn2?e)yzYjB$#QuWw%}manrHIt
zgpe3zuCtzRSLl4s70rl$b0)2v^Wi^@$918(pdv1)WiWM4dn20uAM|Ln@?16T=$C@x
znP9QWCqYH+CDn^Q$?L(%+1rEKO;8pws>tPm;NF6NmMl?fg(+sw;B-5~JY22HZc{vW
z+VhT`TOlO%M>Xgkud*$=ptvpc`dV~$XIVuM)dXJ1`r>6IzJjCQ#|0)h#i)J?W5(Q{
zk8yVRo{h0MLv~<0=4`R9mB4LQDub~+bjqQ?PxJZh8O(5TLR_4DUIsIcxcD_YMhZc3
z#|$QaD6eENG5doum?2bl(^I=WGMM0cX9g40J|%-0OgjsE`lX<_PX-fIS2CEG{a#SK
zA>tYx*>mC!A@}`S4}E$j+YgqjHIAtbCyx?aB}8LEULs~*q37lHm6ae`!&d(6N^n}^
zOdgp`W>C*2x{kd}^9#kElVG9`J7%3EPX`Nsdpu@{91s`!OQ3;bX}8pX40r9)DtpRp
zxu&&lVRxz_q2}XTWRF%EP7zDlo82p*D%E^P2EA^dgpT|7NyOov+;;!IY428}q8KH&
zTagT)Qx1YeG5`!zIh%b1ltJcck#<ZOg5r))Mo?aXGGg`zK^aWxM7s}$GJ@-!P)1OH
z`xGdXUTe6pXE81*_6WY+A33%<4oiaKC4a2mD`wl<-wz2&`c?ST28zKbVDKX-UU-w(
zUNy<tJ6hdY<vr|a^*Kd5%YQcmSV46Kz)BhJu)NR@+-~`L)&)hdclc8EjyXl)oqaZ=
zh)8Csn|=L>3}5#w)Fs7z3L-&u&7XvS3=LJgGX>FxM|PT1PF%>MFFdl-L3zE9Wl=r}
zym^aU<zoja*6W3j9is9!E>t9g{q{N$45|=mniMZ6z9vOh9RI@$HKnIE$~$TEUiyps
z#aLOl>`$-2cd1WfiZFJ2efs{h!K>?2?-Y(zxEWvaJrJ1335d_H!dh=D3G3Z|6+*Xl
z<+0?7;BB_9qsQJe0as&hFGCZR=gKpb@^&I2W)NKX$9dtJzxC7Ui#&X;`;Ai?^f;E(
zqXzO`+r!mXHh&!{C!j{gPt#wJvfB$9uX5g}_2dB3;Et4;R0-=|)N0A)o8i4`N$`+!
z;#{=;9cBH{1ZSRD#!_rqKQj3pYfX4uuT0H#XKJ-lC#M2Nu^;gcrd_!}{SOT-o|jRA
zHWLChH<zKnGZX<dmyyRoD3_G+DyDyY`TgJLFFTo+C55N(HQd@?@8{iryLOT2yZ}kV
zKR<S1KqClrYEYhEl79I4@9=H+XDfD8Ci$K8VdU`h`_Mf3nsN4$ah62JR5*iV@by6C
z1%{!7vwoA(@ai<<rFrF=F;dX1vS^6kCdK6mD;~9I$zC)IEi9SglZHVzjJbc|AI4T%
zKAjv~wA)Idfm(^dOQSTLq&)!AU4*n18tq3T)M(iNp;}^y?_jIp_|}IM>j|lh9^^5X
zv}vE_YYNZyFC@cmh!QCLQu94?+zxGJgaN+-?W;HxWNV2Fn{_a*d_<e8Cv#sJ6zE+m
z5o&v;*;bIj#fn}t#*DQ*+=zeN6?S<eyv7IZVm$)9UEnf{hZ249{x-!#uvhXkc`EVH
zfu<rS_<Gc8tAtUiovm_Q7)q80TSDX_rMQVJs`gM?jkYPP$LCPuYaf;rf4Us%W&&p1
z7<m(oB54Lpfr0%#eMA_lCv)eOzy~c1jVK)QDI>Enr))s}op$T!1)qOsQDfSBMxyxB
zq_H3^SPgGY-dba};iJnBBvEiO813e0k?-A7jaPy=qGwV;_DgllDV&~5jTtUFv0pmp
zU8~GodW=eaF?;`#2IGY#s_dt1|3(_L6HeO82nEj5uK}6*Xjf?{r;qQKu1GlNE9s{k
z^RNBV`R?(u@Jv9lUx|Oz!dtkiWxWe}tkDhyhj10t<zRsBwQ5f4(wl|!uw@SY(g_4$
zwOynSSHEP#N2sl~i3vVB^-D%C6ja*-Lg~_<{IufZ)>K;~(hr%ga&q-(04u89p<o-!
zB?;6}W6~`ZnKXrBtFrf2IXBYH+l4Ed%7ckobLpr;kr@vJRgQo0@o7HL<F4`(^C*Nj
zk(Mw!6uL3x>m(~u0ZN1nRVYmhy`M?H$zMjP$3*PYJ349vJg6BTvg7D%_X{>29fO31
z9Pynzvw`G+OCw)L!oENdfvJRT0gU_wCw@~v2}zL)<q15Rao22iJhLEG5&G0r4VT@h
zna^mqaAW8yXcK>H9=nIw1TZS#LB)SmVLp0cJ`!<$z~q@X%g7{IPWtkJzlcCC{1lBl
zk#)xghdE>*ODUDXj249Kbu6ur8^z}}-+@KpEVjD0cHY_si{qTfes23KY@}pkpV41I
zrJ2dP5AGhcTKc;Ojl>V^ub@=I;)Vr(EihU9%e8(jtT}(p#>vbyJk#cSa!;qhGWtvc
zYk}7q4b};-+I4?GpF!~fufRNJl~N!y`4*F>cNv@)iy*sZvPWTkJNdV3F|hsbo9#a3
z_zr4Se!7qTy)D{kZTb+?Y%RE$YfYRtC~2w`9!Q&4!N^Ik)wHZ~t}cx0B(MuxtK_r3
zC#1;peM*02Hiv_Yr}~9Mqc0TrSoCO~9SXJ9JE72+X!Blqol$GpARJQmZMCflhtPXp
zK-wD)NWxosRyc5>B#r-I=F)2EXD*GzJ_@>7o&`QzU@BrapgE1ih?**;+l(@=U>!sP
zNVSr$k)FIun%TIpV>-JwJh#AtPzqN?Cl{Ae01kgJMrZZpfn?QaRd~5z5B<{AvA;>!
z8fgI~a|ykX2bn|HUkpH!i?gwrw`j3#;1zTNH9gxawN_P_R4#a<XG+OT^BsQKJ2`hZ
zQtx|blT5Rk**0e0p|-7q7F;SXwEzDsP?n2r`<SNf15p~G?X%;sQo2<bRziQ<D6Fi!
z7la$Nb1ss&&En8&VE`Ui_CO1`y}zG#{{sNLVo#UBe<%|II53ytG&vNPZ`=eYmzwt~
zt$*kB<+u0$JV~C`PiNx~UYOvlzBDK6P>|$xP9zf&bt07C#2Jy_@i$3hh_kVY%a+Eh
zF~VZin6^wIQuk4IcxHM`7YqOR-81i~57f_TOwx-Xts9;G0eH?mjm^&_CG$Ybv{c8>
zRl25^C-#k$)pI9<<L|CY9WpP(LM74EEPst@<5Ih+<ABGOUgz8vd!y@S|C;><L-ZsF
zFB+eFT8i~ySs#g;WC_lhzC6dCaeS7e${;&Fy|uq*|LSZLB|4Utq${=%zP)HuG{kL6
zCa+k$0JX5*7B7hO(ma6mnQa<>g|c#@MU2l{NLI|PN#^p)u+_6BUu=|4Nwy!bHh+s2
zGLKiDNv2k`7W&q~J95jG#fo2Ks-2G~TOf}*v5)q?nWeHk*EpABrGwon^Xk$n)4>Xg
zNwa>{`R3Y0sYRf_W~=b(pqbp_iJ$~Xm`6j5M)Edat%Wkn$h`L0Ltj~&zSueH2fk<h
zq-dyntE92td5mQm^j6<<3>3k}Fn<_dKTp5j-k<)rII6)Q6dZp4`zf}gRziC?cn~%m
z{`uqZ=hJ^@njB@NkTfEH7<u^Tufx>x3n!Q$Nj1_^TpYrslB6#(2`N{Zsu`YlU2z!o
zJVp0prIjE`%2Z||m|U`37<gtiVU)6bB;kh^6220)RwxS-VYYmcs&=N_Vt<{QI_pIW
zYM`YkV*q($iMG|brArfn01cl|CaKI3d8H33GD=*BmS46NK5+%jp`!}|DQuV$q^~H>
zivSC?s3mhrdlcZTRGbL*)}wPmE8Nj3CyM6yc`uCX6c^bz@HKmWf}(N48of}&(LSB_
zvWHA9J;kCwS~1$RK$!p{KYx(l6t38Qa0RAVSz!))Q%XVuIH|pK)I@yOmKz~9^Nk=T
z_ucT;La`-_+EOVfPR+98Y!aAxSeFDRb*NxfzeTF;K{?s~QmEd7tVRX$jb^V?N!J|+
z<itM$iwgN#it52rPU;<Cqkd`0c1Ux@u!TWGK5itOmkt<DiHe+>9)GXNkCU@6tu25p
z(}%dE*4Y<0TLJe~LB|MiS;v~7hsA|U@)T^qyTIgo(>Q7)TnKX>cS<<gy?sR@M?z-a
z?HDshf;%Sik?=+$A3J|hB1d8#%vUG!iR*)je4_S45;+=8JN-Qq`Al&L6nA)}5<$9|
zD#%aUwrQVuD7#R+FMp`jNU59vwaoq_HvP0tzy#q;n0Y+_bijH+&q5I|&((o20LxrY
zJQK<(?IZCWS}vq;L=e*14*W4}Q0Mv$cSz%uUWD{*hZ1tNqSjAZ9<4j?Ru3uFB0Ser
zKG%bWo%{d=pwG{Ro8PPt^>vOq9!nj<J=a<o)Cx`J_h`MM)PI3&2P$2SXjf|w-U^3q
z*0YHeZr2}T)dHlNN6Pc<x`U)WWYEi~5TDz12l70-O4=;;(aML09G^xy$zMWS@3|s%
zTOx{dV8cwQ0OZfY9StDR`5y>(wAaF5*fF(C6n6wX6XgxWGxh!=$fNCW(C|f&XX5%G
z<e8}b5Xhr7rGHP-z>a=7QQQaeOjLJ(JTt|8AkRc~+oqR;BagfT4*4Kfvp#}`f)^y<
zLQ&pLQA-fC&0Q$s{S>1!E0`&EnRYgp1`1kkG1nam*Oc?5Uo_rT`|p|&qVVW8bkuqW
zq2x0Ehj4WaPAPRx%URt&r<h*U>PhDEoM@S9hZpAb>3^9}xb=SZH_c@+eVKot;EYR{
zh~f7EY_Gx7ZUt{6BTksdg8kKq{bfEiCn-ufG5b0HYU7~!Vcj?I3~KKf?i}_lO)qVx
z5A8D`fN-|$E%T|}HwdyK%lJz@cV$~wcLo)~4`=M5om)aM!%1hMHB|twR`tj-n)Z*g
zl*5Ty)_+Ldoxn}(3Whw;xS3+KbCOcEleqCDewfcIOMhyoo~)y;yqzA?N`v@T^jc1{
z7hRP9C)kpt@WnFf!>6>p<x07mEnD_fuj`JHrP|!>-Cy8T)(8!1M*K<{VF(#~1>^^c
zU;utcXf;vX5%N!zHz5Di`->nypch|`_(!G>0)PII*bl+`09NP#dphMraUZ-tQN8Mr
z8Ou+-TcEy`V#kX(e7Z>(UZB>OD3!1zk5hZOR$*G8O93}>JFyh!m*oTtrYe>yx%)1Z
z!@z%%{{ff|K?@gFidV{@yiJ&TH?W~*yS%nNR)2jgwOvq6Z-F#E_QiI>n>-j=0(n4{
zv3~$-L-^p#5v>TT=VYK*`7eGZQJ{P1gukqmk<tSIy(WW^;uU~i*WTFs4t*D$DDpVd
zZozB-^um7{HS3$Bl=2{qaXXafl~ck5|F44ZN*iHiJJwuJ?XkOeapKiD-HYH-@+J%(
zXm*DR3<bBY>niCf!MF3;atXC$IrrhKxqlQ<UAvhJzQfta-EI{q2m--P%=G4k-z?Ca
zX}9c2YehRdF{@$L_NPAS!UkkxD1C%kg|!;L-T^1Z_JXh6+J-4Wa!H&9rz<PR{z}Ph
z!x@5g<k&rtS@hw$Nat&}rSq2jKg?6JIl-4}JO(NTzq&V6P9R~>k<jJCtl`?rg)6Jv
z$g3z{>AZ09<wJ*R{dMA`rFyx#zpguMIQjUiY|Tn!;Wur^{tv+Xx0hiHCKCcRIF}*F
z0uz-BE7yOp|NZTkoh)MU{q=44ZN~yBrG&C>uwDx*VfgrU_uKDs_=8^lGDzWq8~&T7
zejSul!ni=sUxt6}K3r0hQeG%DgLO(s)KyI{QW{@`(JUH<JZeY`Bt;q$5lJkg;uwi3
zgNJ4{p7MJ{R_+RgsDx#qtrxU#i_lUi<GrF;NSA-tMi1Xft@F`eiywKos32p7Vf7?-
zNjIp9fFu?cfsKvN6MHf(ZD|9ornD}t-yp(_UqoSWnPV&laF@B9Qjj5D_w%L6Oks(w
z^lUD(G+i9Eb-JG;B?%#xHT4`>Dz?GmuD*<^^83(ew2Bd*_c0dIbS&|JB__&Jqb^H3
zWlVoX@R0Y5E4{FOQQA{s*A_2am9P~rOS@6R2`_MZh=3+o;S?@Pc)Kw=s<h}Kym}PQ
zE{8*b$E4|F1DDb3k8A#!tuZs;Wf7;$D7ard$|@f4X7SECWOcSRk){m5>3oVYv;<&N
zUak3B1REdYt^YM{#1)AF6hXSUQG9_KgeHGYDta5t@cqx>Q_5Zwl650VOY$EhvmGWQ
zoXdiwTqdnGW>OOYnabc`2u5K{_Bcz|qw&keB|FY3Vuq-mVl>(z8=E0rVC0eiB-SS*
zjr)Lc!jeaIRAJfHU?J(V=ixBM1!p=zM8`jghsj9G1M5nG06%)rOL4{H%=eZ}wpM?i
zqBlZ94LZ%?nk#^FRS78aA&dQLfD#p^V1q4G&hTJBNaA)v1@u9U)d*>hi@#%{$AuhT
z=FeLyQo~F-u5&o}OC~X|xr&Y%CI-Ki7x`95Hj%ogo~Wuq#TCmROWxx9Gs|N}O;?#R
za@u!Ad(n$!@#aD^Yk=2N&Nxf0f*pU?3)sq>cY&AUm*??gf>qUNGztYhf@q&S_D3KH
zT17F74@8ij;$sk`qq>41JrDCpx&=Xyq7276I-M_N6Jawi+dt9yZd0_3o>a}eg-=7#
zM6pViKyPcZK3s)vbSMh~m6CP|-E{E689$Nzm~jexJTOWb1@(OF$pR+zq*8wd^z&;8
zTwlNl*Mn;v_?SIsMNqC*JQ>uVv{78kx^f-or(LRfS{Nv6>bM<2C?B>)fJt!b>Y5XS
zYc_r!jk{o|)O-VDp>fS)Tllq{mjfQJmt%ldIWE%C*J+u#HG9RU9fsi)Hq7%&W5LX0
zTz{<VuS<Jqd`j`1-D=CFxK@8`vv4A)+J&Y}HW$6aP(`RIxpuoG`}oW`<ODVICrfGB
z)n-%Dy2`dUO}NVwCyC`g3-6S`PdEB@Z3;G%sM*_Su{LrJs*9XN`+RR^-P#pEDJk`$
zD>@%3#F&%54TKNxhXfLY8-xLdO6WKY@vLAiRogU-K&hLMY4_7VpFe;7{PEvZiwvLt
z6IW15qvHhxPc?V?X$&n3N>kmPf79^-BCbT-j;69wfC8Gp3+uqgVTV*&gQnxkIQk`y
z5^i}yu4Zvcq>d)m4Voe5AyTZPyx*60khvPSNHSG6;;%@htc^cuJ+F#C<JJD=skG{^
zVfSg+eW>~jn*Pz<I)Z=T1$9#L0jHr=DGzB!-Wt%dI)WoZpBZfht%HhA@0Q{b>AdG;
zjLf=Q6_!UDfq95%F$Estdp~|g0iCQ_C8YuoCeS)MwcF^FP_&vgJ3fE>T9$pLMemz?
zCi6soA<)=?5|5@LDLHs5kjOa}3-dgkXEQ=Q@3m=4imoKhSdxEtZkoU6h!P7QhK9u~
z>Qj~V!`12m5XRsY`squiMlS>aF8HdKGZ~jffItO7J*4T47>0mX<|Rx8Hs?Lgtq7*e
zR;FMx=jE8$^DYq?3l*EQT5TES9Kvr^q2LNym*?~WA<a->uPfhXf&O#TRC_HNz>&|o
zL#ZQT{y-hxjk<rkqhXA6Y<9-|Fw0ExCVNHkA<mo6O~rMnS#hx(o@B2DD{$f7m;8Y+
zEQEgN?triqgQL4I#o!oFISBtArqfEcd}`7qqlqU9&g!K)RawH<vG-UDQ!a-~C+w=?
z_^QHKnV6u*&q2sV>!c%q_mv_meD<_2o#3}R|JkoSCZvBZZ!f>TN~>!<A1z~pbw=Hk
za8Hiig#4}Df%vUV&H9VrXj2P6q&=W0`#a?~v<Jo=($lyxkF*EY(lY3FS#XrA8W7eB
zu&SrlRp!|c>&jWv3<RXRdSK(EqnI*X^NE8$56DAqsL12?W{TGWT<CY=&mS{n1*+Zr
zjIYD?eS?1yA=ZGM&U(+eo+1U6fZ%$HVD4z16uXj+e9)Jzz>S?~i+DQbmQRJ0^c^J5
zQdfL!&$IemnRfuJ@8wwMM6l%ZDEq;Nz+d(4*hX{ak-ptg42>V3Sk~AAyTz*=(wPbR
zJ<fSG$G~)U=H_<Ko>BvOhi4;|Co^dP7o|}$p4WfuOq%Ulq;qr3AibgP2vKl!^q!f1
z7NQ#{DwA@0FYN2GYGP|$t3^ZKpdLDtg}Hw2F|?sdQ{sw#CiRR7k1(`FAwfUV(p0C=
zwQA`bz&L)_0z1M6gW)>x!X3b?{%^pmf{?*K5_qQ++eT*oZm}KmN}1JBJ_ol9&ePd?
z#z%j)sf>-^*M}8p@oc7GQ}vo!nlU_T^*R~(bzSAtO*fu6M!k6vR8K5{*OKbS9NPcS
zjHL>4yU`hbrtF+cN(Es$-cL24&hd4;`O};?gg|4<WA(E^{>*Tx!I4y*7)}30WkH8r
zp}kzG_2U(v;THbm=>GuX$EFwxWo~41baI!-e<%zBI5U^=UJVlgF_%F?9w(Oa0R)#H
z2`dhND!&5smcX~7hr+1cLtHfJp?|+Kd@ae{^;)uQAOY5rxa7?6<8Yo4?1>5Xd3SZc
z`*f!?bKDznlHETtz!w7lg1|v*m;>Ne0lR-@zg-F;Ek0NHXd#rs$KiL6pJsodz~c<S
zH*k#)>@S=K98rLeOrP=&pKJQQzo2)tHqwfJ>mAJ&X|~Li<z1e6ohck}rWQ63FMJP%
z`0anc|8*zwwgeBBzS76|>-K*4eK&5>c*~Ww?BRJwFSG)#yktRfs{{M-Cwts|8L3uY
z0+*yeMz9}GY#8~PVGaUe7S<Zm!3+Yi*UZ|$6}?I@>l+~qZcZay8aJ*HA_>DLtwRof
zpcoLZ$haFp#-HvyEOOO2SDBHkMC=VhZ}3wicViM~<MO!aB&gJ!=RwY>2~-m#5LarK
zDAIle9;g*`q?H%V2=Ps%&oyCJ{=}ViBVQ-pQhBbu8+pUBq*6<cIysXxN+d6{6j_<L
zQEd7XrRYW)De5U2&lRdirc3G?_XyK}NyhI!<7$$suj8Fi%-Ceq;AR22@d9is<)%g|
z%eC_#t&u&~sF7hpt8msPOX6SF$e!9OG!kd1+vBxR0@pT-{6G)*97EA?CugSsH*cUA
zctA=c#g&_)G?z-wv_O!q;i`hs8T3q`l@;=%^{nSwEML^Co2(f&-Nbln!@912dBwqn
zk>a+lk)taXVX{E8n|*b<R$9;T86_LEvd(sr8!2b!A=acbgpbarRqm&Mlf(+OO{uxW
z%Q}it05?G`N_<_OlWn;P61~4FRM*cXg*+2Za5PRSh?(SmDVWKlTVb0qmRqQpP?)B#
zq7LUUElIMEA(uWR=}2FrAA;<EgKP=Pl1Iu|O5-s&!n4E2tAgijbh?FeDZSuI&ZImt
z{#e6+Y^R!IkasNUe2DsBOQ-X<XeT8z5+e(%rdtz3+Qk(Lql)>89`lHzNTF<R69$a)
z`ItjqeVr_kzN&9#XSsn{Iq~aqqUHRf3`GvNzQ2Nq@Z3&!avP6!zSI<dwo}k9XBHmc
z7;%|z%5~O5lX1GMZx`)zeY=EL<VREsGRcl-9OAl&RsGdrn?@kd5qUJxl<6dOs3Y^0
zzU95Qick{;aZp#Xi3Z(7dE+dT)X{H@h{n`O)j56-HpVpjsC24V{|L6tq_QaVoANdj
zRir9#QXD$!@rZT6E2e9IAUy*stNjE<Jg}s+B{;4l@o@bMYHmmBzCFNxmF{@Sd{p(K
z4aP8J31M*i8JCTVp`GQ<*Blw_=G7QhESL&tOtPs;G}}+bC1VpWm7ZOD$mj1nypMJP
z3&<XcWw(yxz1=I*Xyg{sXwn0?inQBCx?d5^P~6X=8A`Gh4D#N8lDXbC%yG9e>w87J
zF*qk7R(E)NKdW0Pk1@RrRJZuf5Q5Jv;($I{1#j4YjD~?6vb8ZnRvI9XzutZN=jrj6
zhd(f0W>5dp3c)bqHxN8~&Oa3brvk8V`M1Sr5`8Pi*g+_(nf4s4!3kwy9e7|nEgi<=
z7!ypR-)J;ox~?&Q%$yd-VvD1Rg+Vh^WFU&uXy*5=9jsDS7E$Ob2K^!`c`g0K^<@?P
zq^JE|rsAr<*zS?-zEnJe!hbRC5OxVpOb5X!Xdx{_(sI;7&;^d*@Di?!wi}(Ma_cED
z)fjF#j7uo#$=vE@%Sf}~GK91Wg)&q{8^%`v81|YVNqT92I2|eGj$RnjO31J#Xs7QF
z&w1SwuKL`3COQ`&6~~*)F0|C#VVH-sVinE!#dtDLFSCnTJ!IpOqwaXYN)6_a9a`dC
z%tw342BIj!7uJZF_IfdG_RFgQ0^r8H2B6WZHNXfUHLt*|qyuFSd3FKj*w7?-fQ?dB
z6I>Qg4WplbcQZfVCdaz5m<s{v(&0=nEFI1SbUK{BSwAh?(Tm2AzM>|sFjdAttD<?m
zM{iRAsxEv2p>x}&aegI=OY<iO)P!j|)MJwJa&pd*x?jm1PRtN$#5&!PIYl~<wY*ND
zo0e!e27s38ATjHuNd7iUMebB&>dWwBl*FB!!C$j~2AGhgyt?Ff$2J2{P`vN6W&jxU
zD?)xqHvq$Wu3b2)CRR9zS`+I=+6qUmfel<1PQTB#S1M}Oq9>OlwIL@tSGABd?6_*-
zRwZztxK*vG6{Xi#wMvC>;3UOz;2Hqku%F3wbv)l-ctCePVg{$zknw6ja3jT`K6gsB
zO$pO~RofO%p;XsmPvp?pLf>=;Y}?CuDG|vP%|jk_DU{DW7G2*jCIj!%agnGeWZ%xG
z+XKb99X?*C@@*d|uj#VW7$U<sTP@Yaug>0CkOn_>_~6dM!1awi)q&b^J=IlPc0!*O
zy`M8-lxMA%U^N$r*(m+gM_7~W^m@Hif%|ZO?$&D(G!@Y1D6Vv5w{~|<n*lBsM$Tw6
zdWx^v7riS!;3)r;Jl()?#g-J`Gs?FZTb|ItP5st^`bv+W^iog3N}R6Y7@*Rv`l(cW
zxp{?YyvGd9QMY8CUx9U0shVnbrBy)%?^?Np@@q4>(SGLWOR|>pQeEqQItL56{gCT_
z-dh<VXf)ST9bEP4(pg$&sGza3({nmg%M2W#G;*ZgBa)KRm}UmsHk5)}NyFN(OUId|
zULE!3(A26UjTo{uQl9}g=`moTnpVHFuKgvOD#Z(zDb8EOdS5;7>)=)aZJ8crqRxx8
zveBQClF>cX^WH^pm%Zdv!&yJGI6E|d%`iJu=7Vc{3&G73|C!z^%fpmG4lO2ar&3N|
zQ*gOI<?{czDQ%qy?Oj{hNJ@5kHjj~rl6p+$t%%KmkI3Z#G}p?$b1pw{d1Ed=P<y9b
z-cxZdoH2Ci3BF?{-!s}s<Oiuv&*KNY-k!$y1kaGgJ25r<f5H@g(FzQoXk`8c2Qg(5
zY?r}*C=&rVmtlE86PL;nD=U9HkK8s8?)O*d518g~_ymjt?)H2E(#5zgNEO7cT*Uvs
z$nYgmLus|rDF-AjwA!Vp;k@~AX65j0ki##Rr`OBJ7u16j-hflX>(>G3fuv86G-%D>
z0E9&tUcV2&f09yJx}NCLQX<mj=)0%8dH8_fX$H^}xY7j=ADjjnL7;z2<-2;J>l(j5
ze2CBDyNOnM-^JN7&sK%9dMPung~B7v)Y68y7Ci^z2=L*L*IzHPh$Tg^@oT*CU!Pwu
ze_yzenuH`ahCjbw;scE!(5pdNFG7If$6v$O%Ws_QxR|7O@`sVbk8cAT`BL;8B<WdN
zYs^f~AW8Xpu*M4{!bN}2`dLat(rMC{CdoDFNWo0Xk1rk;DQZa;Gabc1iY3b#x+Lk;
z%BWRCwB<F^6*_WUKawD`5fWrv!)ak1@HLd2aX8NQD8L0iJ9D#t?O;UE$f<#3K5#!p
zCAH9|lLqvnG)j8mr0t~HuSEqa`Vu!392W$fP|%qk%yEj!2&aEK_mrQPRfAe6REvf_
zpdM{0{Op;AHijZnZ|G;59fULC3`1KjC`p}ZX-Z9WZCcvkFnuJ7vynEeY9i83XmnHB
zh^ayqE9ytOC@mQ%cT>`ntqCQj8B=FaYBny<Fe;8!TIR8g#-~qplmf0kmxX?Mp5NAP
z(*2IyyP0^&!wr9gGZXKeVw9!wuM_3{DbHE~R&mN}>9!{4`E2a?$TpR6vo7&`f<8@9
z!cuxZPoIop@+1;Q$9wcJT~^S=AN_v)mJN!nw5H!ckjVQ^QeAZG3U5%oev-b%&B&{2
z<f-u#pScI9qMDLGm&~`{l@NkJ1)TYny$%%$<2n;${E!g>b<qIHp(?_F@<bjo4SGij
zmw^;36biP9{VCY^=wNybm!A|X7AP~ZiTkZR69O}eOCf}x4`*XWeZ_^q(feRKUhdGb
zmIokZMu2R!l}wl26e}x#Hb51;AQf@uSIz(^l|YGn_Ot;;yqh;Wb;4$+weWh&47(^A
z6lgkAt>h{+PFT3@TW{Ohfc9QIJNgq-t6|=<d}2uJ5YKe##E@nbPu{H<^^M(%^TPCb
zU}vgj8%B9;XDTI`n|vI0Lrr*}Qvd(M4=Hs6?m<oDZH~wT4iFcAN~_LBzTDxf^Hmke
zXhM_L%gWjmPW+0uc@b1c?MAk*hl-~|v7i(PZ3ADA9e$3~$2i63H&6^BINwn;H!E)3
z3*_nNdab-rCcL+w%~wr+rix!Z7td&`Okv&E${C+g-1fCtH>nw+Ir#F(emd@zrIyDy
zTPr_1C5%#?J>(jHcdIb>ILC-qAod)uU?{Ssp<RE8<7>T+=W;sn;3`xZWz3DKNlUO>
zOHsa->su|SamP1)@FcvmxYS%j7dvle+=n*y7-M7W%P-VV%V?>VQUufyJHw-O6!>T;
z@O@e#{etnA&ss=pw(%AzJ&v<(*M6hEw%UzICsEzRyRL12!9*|8k+~gMV>a=aXC$pX
zfc=#2YK#Ljn*j~2^4qxaZ3ShXcH>*uq8I3N45QSlWWtw1`LWf{v>)tkOvVihg5!-z
zAb?IF{aqT9i18ur%KLQ%Ip+P@NE^i^&$CsMW$ety?n2T9>oRJp@SIx9c-oGLbeI8c
zPCNlHiYIoL1Q#oLf76Ct;lG~%Q*)fHu9lQM@-<dK+vgQ7OeMnG#fA=^E*rJ3iMiB}
z$})I6L>9PV0CBO5SuLLd;>=)2w2xkBoV87-(B1-?_oRvR-?KM!shrSmDiKBa57VXP
zUGV=l*o}>37smhAVYSW-a8_9BR$*<5Ra-RSo1;R0-%Q73f2|v6<qIHBxu)Y$0g(J1
zkK4&+M-CYi_eyN9=?_s6JPDk)NpBhKmxDq*)Lo|%RgU2b{zp9)8IFQ=uqXh02iC0!
zxPjFZ4qr+7#Ua7Qq;+Gf*^eut8tD|lD6<WVZ$6xl1I;E%zo0vb)^JqpNf6lJ>z>wZ
z%8e%X?=+jd1`%{|=GV5Dffy?ze@d!|!Zm#vQ&OJ0neITL(A~S?cFh7QA{2ISl>{@Z
zCKPWq*CsvXq{BA5CV}Z$8rSuk>Lb6cL~L)1oyKBQ4yF`;&A1%_%O8Yut)z~m$Y~Pk
zSTS?8Dx$kO3EqM2%X%~5($s0(NT-ebwg#2f&F{8zvktMXX=M(#wy~Ew3zO9k7^k`J
zGmg5vmo6DAFagS!Ul}VO1atKfBA3B`C=&uWG?&5PIuimgIF~V6HYb<U87r!P*tqBA
zpC5mFl6)*Ff{kC}$N1~*<LN(7;~}k40@U=cKc3<^Y7pA1Nzuz-ra%8Z{r2?tk!amY
z(vtkg$m!4Dr(xs=OF2lAvb5HiPRbxj^f*}?gpRMWl<~WihR%j0Eu?cJNk~C=#$Vn&
zY*Iu?`Z6nufy7FBGxW$gQxZmhPs^XgdN)F>ulO`xXc7QZjnv#t*;j6Fqs%HLlvahV
zb{9wrHtO<C0q_9~!NQn}O20|%=*wHh0!!f0Wz@JrL6u)+wa-E|E-}x-T(jqca3+lG
zVc=C)4i&6`)tS<EwlITOne=fyvqbD!9dS*^Ko`eAzs5rYv+U}sII7Eks&berqPh#O
zbyWz!b6hW$Ja-XlT0Q5tk}ObZ&GZsq-D-SoR~_T5o_b~uUl^0hcrraHkRGoKE?1*J
ze!Z$RG(cwIx8$FBr$;rCfW^+LoRN7~*UodPN-%?2?f{((Njckrf)$kA-&*a(u|~+`
zZ-q_sQb#0@usR}VnPz%_`@%|^S;ep2I_AOf!ILoPF>~G7ZdnfRXe;OVFcoUh<UiCX
zEqy>`F=Rxy6jgRmH`*D2fa7lPozHVJm>7w|x>Ym@T38HU5>1SE!diG>G#M!>77a$e
z%Adq;H_}EB3<7Y)o01(Lb+(~dgpqOIhuQFDi4S=w1xHydhe^DDC+BzesLC5b&&d3c
zq%5<8SnDkMTq_j^K6s12pqM@ugFnrh)=Egfr-f4#2_Yt|G*ygeIIKn#9^Fc>;jXO2
zI4hh~XWWFo8Kp8D5?Aj<^6<Uq^MuLn+N(j~{3El!kz&pM&fBfgl=1nIVwA=e9fJiC
zq#IYnEFVpd`eV+25r6k796gqURS8L~xr&n$28mV#a<n5y+rCE`LwZ|j<$UmoGCyyj
zV4_;r^4_*5pe$d(roM~gvZ0_+nY?#5as!{$TdXyiJnwu6lhx)2dA?;?5wW{$AGPqx
zHmN!L&1AtZY~j#YdP;(dRj0C|%wQxY%_dD3S}B2=OL|p*M#0qgo9S1^`p@teDq5I7
z;V($%j>v<aAa0aV(2o2#;8XftRq>t;-jTEVn%QS5Zh8rd0H{}Ixe<no#@_kuOXH<$
zWm}%l<B2{4<!ME>ySp8#evO|zg-K`iR#AWs1cW6Tzck*F@mNrtJZDU~(U0Xr8d0xn
zEz|_x-Y#2z3pYjy7Zz(bTncFO+~_1QUDi*j6E$v?<9ThdqFi9aQQOtOWu(xp@mMw`
z2U{=COe0h*Rr_9wlp?Jvfx)twCvFiiJ-h6fFkNigxARNA%WH0Gt2<NCR2sRisk)cS
z{%n-fu|B~<W65*Ms)U1;-Z2(tp({<Bj419MdwR5g2Y2HVZx~BIQ{<X=;n6+cs?nA%
zgRPXB{sdFO#=mm*8}o>SB45`PJ>p>mp}Pw~jCZ4#M!-rMbr;*{wvJpy8g_`mQjTBc
zzlm%n47`Q~Wt<VL^sc<{6y-}#ujM`!_D1FoY@$2YQC)gkr3v823T<A?ET5*GO>)ca
zgBO;6g52<S(CGM4O$J4Is-v=hIDd?5J3XvJO$A9ciCIyvIE7sl<OWUaZ)k+=yvd`<
znp<f?*Z5(7QTKtYnZ<7$Lk<8(4vJ&9B8OBWxO!GyEViX?27B@~mfiFghohJFD2!XO
zlyF)6VC+6xd~WvH;Y?X`h%@eyI$JusdEHKbWH>_oW_wzB%H0r$Gy9a8-EL8<plC%H
zr4_xOdR}r__UBMNvn>)_PFY%q$!<;cqUCMDFScjyysRGfwQKult+l;oqO)xa5EgYq
z?;<Pn2caLs9m;emHD5s2?zZ3bquME9m9DZZF=ysTk(rq}3LNs5vU<vE-gf0D=Jw!!
zb=SQ}yw@gf_0{0QH@K)x;HF8vg^LY}I|yj*2%1RHdc|9wju7GbCJCOmQK8{XDE(|-
z7|1)uI%>RSG+s!pzN9g%O<ZxVa4b8>S~zwiU5hstio-6~u6p)bS=I`$YNR%l7FiSb
zm(&CSs)ZlfG9M{!UBI4b!*Nq>pn9!;eeP&I*m3q}+J7!yX29w+@acvzI4^=)$p_3p
z@qFnXQ?3-}{s*lf+-n8>gf@_(;H?Zk5VJi+P9*NcZMNcp*E=p83L*!J1Aa!d4YR8y
zoJW~*TBzC{DaqDf9BGDPm2;0oT1PQS(c%4QW4|}>dl%XmKRv|<cq@gGSHG5jKtahq
zmel_xmcY}@M@HI#;<|nOd9T$|z8Y=WBlWZORvVk!#_;yA%muumvID>PGGN&CWYc7{
zu7TSAB|zOv4VPUfE(Nt0=-s&YwOe{r!Xd0U?XKzTuO)I8j(ZO`*CtlUq>J{uk@i}U
zavNh-+{)?)G+)6a$pqH@P&|!)oK)Vq1<SsGWqYM{cdy?|)?&@izpP(j<ng`?;#CQ{
zueQB@*JOJeO{}I#e$)1PY6ooZzJDPF1mV0NGqFc%1%mJnzFI5&uEirJEqrg96J0mq
z%GIc*Iwp)1_caqEPO+cGGp+1k<NES5dX$78sMXf>T@GB3j-PKIPyYo3Do10Km%)E1
z69P9fmoZiiBn&q;FHB`_XLM*FGBua*W+f??|0gE~f9}76AJYQma>(IurU(Q$ah)cp
z0mE{Ewt_w=@*!5L=mhE1xc`0MaBn4@?_~WtsqhEO++FVSaX9nl4SBWY7E)-ng`%Oh
z%s~_WFB~#wZ{Pv<Cx#y4ckrD2PN5)`ypcmw_J#x_ebI!1jZaE2D`+qWqFTMx0%hqB
zHBd?Ye@hmKF#w*XK+^9q1@b>=<v`jT3`@u<2?nTW;G0bh4aNAagiP5%u(!mKR&+1P
zUzCLDjGtAsT8D&JCHWf_Pg|cpKxb|A0<9cM8lRR>2xdTwL&?n{RfSbj!LV+g(g&+(
zHUEg<p|D%A;ROd>-6BOtDEP3IiXo%~PSW4tf53TsS_7Bj@Gt{raf`JSxSGJm6j+L5
zYTj!7VGi8Lhrp!Rf#KB_SX06fSWC7JPw~yvSJW}4>O;V8HD`aL!S;Po6qBmO7gOC_
zY(5RYtd#s7x(N5DeQ-=e6F#q0{2q*?zQ}#1YJN#MNb(6OwIEb{k>L@NyMc^@l)f5P
zf8k92frgpNXg+*EU_;EqG>G-}GYc(oM+&!vhj<J3N*|nR!20{(<W5~XwC*OjR6K0C
zLxC~f)Mydzr-o_3n*qn4mWmjFcT`1e;Gud;gTlYOWkHLtsbW^j9xb>I+G9cy99pH{
z!y97W4|CC`f-!(wmHf-_C)I9nd`hjGe+tfL<n~vrlTn`r09ls)6%JO+$AAl49iJLn
z%}0X^n-IPD_S@mr9)h(ZsNW6$_~*X@C5ehJv`pdT;`n%fXJ=!thF)+!F44_%)iVs=
z&yVM4yQc?-vrznnFuXljEM{jXA&-}T`Ly_9cX6<o1sZR>o}Vl*#%tJx61>Mpe|e21
zpo%r@7jR`nYZ))#v_!CLzwo%{EXxb_3umZzJ3GVMv-#ofY_T`Ief4^HKl`*;&gK1o
zPiMgN%M4IY7PFJZxpT@mxp%Yk`Ni4c?0jS+zx8JJ;ppJI`KP^vOTnTx%lml$3~$1{
zb%glx<YYd6XgU0gy-tEY!A^q9e+EsBZjAj)JpKJ)_u}V;U;XXq<lkZVZhrP*cGkyD
z`@>&{ABXSvXuNngJ6!BhE|CzfLKZ8QxfHAl0wY_@MfQi60bk!8ewe?X53j;EfJ!Nw
zNfFS6!9Vp6J3H5G@(>IxDH2^_kdkBm0tiqxTr8^;xUB!N`gjm-2m>{ae_)V{62%IN
zU;zlLYaqbp6|61-)=334(8}XMV4&zcH-RW@(b?%H5SV@f7niIEYW0w}c8S8{t?*^M
zfxtHe<h2e{K!UVmg-d?@0m<tD$qx!hz7~*0r|_F1lc&h!DKdE*k*PHzFu6*sE)1fl
zW)_Lbp?>BFhaV44u^cg{e+^E!c`PhNpiAe^_`EBEC}1&a8qM9Vq?ohpzgU4jyuAGS
z;o@k10vLaP_oM&&=Hp^<dj98O_~mHv@#5!rIR9<<_2ReFKYf_}XZY#j&D;3v@o;p0
zela^APG-Lkhaczj^BEAocy#b`^psx+na6wr1@Qu*lqIrmpfMy>e}bBUkje=uIF-DG
z;H|J8xjK|sAP_*4QPrhkjrKecCObg)_MGYqnIKU~7z^6&&mYWp$?CjLB7%^=y;rGV
zd@%j(J;^W=+m=dLVVf0(S|O!tbah|qscnsVYP|#Xv}&u=Bk78Ic%mMjsHaZUQzz=F
zd1{oY!H*{+6fkjAfB#^kjM@dO;j&R8RbF!|HaHSNSVD)BKsi?4o^P{JgkN9!b_F#;
zZLj+_B_arSI5u(#il9=rArpwB7o8aynTR?H($ONwHycoR#()eLR5s8UD0jEOG)1Ez
zB~BnvK!Wz@Mi5U!ca|{dC5FNbw!0a_7pdmq(h&kF`>0D4e>H*l+n1K(Nt|Nonu4^6
zVw(U;#CIs5AlH)*VQ@nkpHaV|&_bbD(hBKTxka>H>u@RB3K$=x^Q@xnCY`oIu=iz7
zu{GwD_7LW5e8rqH(W6ZCC=)#jPmMY?<ne@Ah+>0`tO!-}ot;)u%Ydy|J>xwSPZ-$R
zYZ@!D64EWoe~RS$CWsPZeI5{nH2N-;%Oi#bN(C74kOdgo1R@hkDif-THAJCFwgnU}
zqvZ+&EK}Nz7$Dn7O3DcZeDVLck-(TSSiOYci@?yTH{0O}76aTD4RI+H#k{@*d*lk5
zi6G`R2wA|9rkogAUIeB8Dp85vMIEm6TC$bG0qrgwf3wTJJ9NybmeQSig4j^h@c*s6
zy)Pdst?^;Ahw!16D?V(xf#Pnj6ANX?K-5_tO<*^82I?y$f0U&;=SLNaKV2Lj9lSYM
zd>nR9XNN}z$FDC=`fKN?8n0?ZUsX*vC&(vX%zLh%CjNZ=e6RZc>!(v8;_Ii&q~+V=
zTM+l^e=<H)RL}i*A!Vco4?d9OMrL{%5rsTaC_u}0VGWtDO#+DezPcJN;)e>iQ`2g%
zsc>y@?RJNr$Wu}DNj9!m>eDTiq<&!Xxg!!i#1ZM*ZUBEkfWMecGRh_yWs{6DnHp_s
zO!ceT<i2TYOn0bGjkE>N&Mju&QLYpXuyb=&hD8==kbksZxSD#HX=ap5oC_+6ZEWgJ
zAd2<!AXn7VU<C(N(}qo%)=&|MBrRMmRS&T(qVSBGV2YX%U(FB0B2xT%4WAPU>JtQC
zu|!y)x<wFRn&he&c|@VpQOrh|XG4@5a~0q3w^|vsYu%<PkiC-c5;C%yWuU4E=(3I?
zp>)n!xPNG?6-%SU+e1;!+187ALWN*{b{X3SF~$u-7=l<)!WRPfW!oYMO_9K5lPN0L
zx3zUsl$In_go&XjgDTE#o0dTRoRI+_MNzK!p|?e)bxTw#sP~ndmk0)3!&}(SK<bG|
zH-(UXd{+r2z4X&}<R4nkKj*WqP5v0O=r(d+-+zBEsYdF*FMvu1j~3r-k|!QJS9O2$
z-zZ?Fn6Yo6xS|G40nmChxXn46pB*6-d{<IeQZMB%Z(m&RaCrz!2LHQD&#ZnpXit%K
zZ%VX}k?%Sl(ynCP@g(=YS$8~L?pas!GNbyB1eOik!if{&t@v+ZP@$XeY39;}Q}t}w
z_<w$$@Tko+aAF{q&d%gpdqCx98pifaD7>B||KZRQMe$wdA~&M|C4<KpnL?1Zfk9;@
zD#la$+XU%Kucge?)Hip(l6qB(tvmcZ_V2FQLT}-bvR>T#GNCB_YK+Be$3)5xIVRHa
z{AXkbUhZ)!m#MLZBOtCRrk4A!H2TlApi$kD(2c*+_+msnVmC2EL_@)2!3EYUO(H`p
z2=aD*rcE^8v1_CO#c$01BRr(-46VSv2hVc<jLC8rZs7G@e^v+tzDq|a3E%fy_WU0i
z&|D0c!G9<d12{D`mp&{jJxy-xxbg0<=pQhL`vMpZxZCysx#hvPAg3TRr#WQ)|Bxb8
zMUkboTY3W|2YHdG+BcR+%FFK;dHMC__Wkn9m$8>1qEkU%zJI$Ycp%}I$t)`zf4?LA
zcGp*<B8*V*MBU+1*VkYb+%XC+J^a>RaNYCY*Vp_kzgu9H_d^~nhtV1#>n|L6M+omQ
z(n_B)E_kkhBIU0C{QldEM7DnUa&e6(AFKe%iKmRw0VE5wIlyWGlS}X$=yGoOzx<t_
z#f&}kY18nQ1DdRF-qgcA=L=0+fAaOt32#UFrmJyp2qnwJW|_E$DNMU41h61o<NO;w
zy>%SEBXodl@m-7XT-b*NWf0Yz3_$QAIWB{!O@_KEJ)?|J)^h2QoSY{LBXO2__#9;0
zaA>z_^g$#&QBwxz2vH*@{NJQORP;ZpbB!fyyohFE5QRkFvC=3>nVTN4e~&CDddppl
zLlrf~K{&noLJP5&PoZ#PqSgo>eWmD*6nVm*&9YpX<1jqIaezx@r5GH6GRh-!fN&uz
z#I#%$$JRHb)el!Sa8;pi7&jR9np(B@0?;ttis=QrX?_rwgo>6+rZ`iSbB0*Bnl(fp
zb{e7uXHgSvY+9lmLj#2he|KBAM43ba3|#Ma)H}08KWm9THkRlYEwPgugNh+9Lw?71
z+|`>L`sKW>8YRM*u3D}K8IM$uJ-V8@G9J)8CZdAzhVNz@ywSMo0cz0$)GiPB7}AzV
zK4-4{>sR*Q=l9EhCQO2r3Jv+|Uw^#h2NrlXQD00<!kKjW^WT@>e_sC1!D4QdVD=aM
z;pFAd-!JUW4=f9k1ly1v!cmvDlmvqT%Lf}#YP5}yMy4oht-7Q<Vkm!&i0KX>u2{g<
zeU61a7V;>Pt9@r_zj)4+X7F|8HHW5==2CDN)Wo)Mwj|?%2j!#(B(q32N`XU!$R}^r
zi>0ucIQmg&?(Vd5e|1Kf6gfnt=<4X=@?&qw`{Veb@bW=%NLtS0jgaI+n&5qc?-+zN
z6Wu7wEc5}0$%$mm1mZj=T3TV<gl`90G@lQ1qm|A+z0S-yI0&t#MkJq7pw><VJvDOm
zv>>&HWD}+)>7IvZqOdVc%;xiXh}MfVCvmt;zMhBdVJ4y`e`X{<=5-mgaC+*9x93eM
zIsw;)^vBVVV5Z8KkE2adKpflYJw|%rTsX;ljB+{D?sOlck}z7Gz_ZatK)HLO>lh8?
zfZ3@#W3-Ubo)E+hOjL}CXEFkb&Ly)M1Dv^xKMKU|P8*&n8T)MCW9GcRda2Q6?E6@S
z4wWU<dU=%Ef1cVRtWqc5sdlBiU{Ca%s$HyYOO+W$(LYd_ZPp%8u^Y(LKesC5ShgDg
zo0U$8Fv;eSu@OMpdOYh>j_&DIA7kZRrQj1W4rCOl-JKQL*WBJ?InW{4l2%O`Jx2FA
z%%pQyC(oz`8-Ainq%glJwn;ee-!S3v*dnzsa;mcRe*(3J^!_byu+~Lyjhn|Ro+`9{
zI_ZM&5K<gOO)G6R_4lVzPA#B&XbUMv1$ZB(DIB+aqTEKe>LIam`>8^jaT1PKLgYq0
z<fTdoOW=O+vqg|r!h*Z+Q%@~{538q^z6{cVV%k_ppi@Kh2TETpf}N;EJ#RwVDKWLl
zGc3qQe*xEC3xZlBgm9J#Q60U{gfPG!w;&3<yI8Ju^eHUYpvzcx>vi;Do~i0(EI&WH
ztM%+EBahXyRzw^2swFpN+*!fQcvX6Y`z+Dq%%sv8YRRn~9Zw7tCbxe~*lA!jW9MLQ
z-B}sChqi>9GB(3~AY=EC&dAuDlk+lm4|yYFf3v{{S{Iaucd4l6+QTZU?InhHq4P(r
zqki<WJUa&yvnkIDX~rcHTsnhinsG2l>Ugr=r#)(XG3&XvUHjEJ?BdLs!UoKs?f9*0
zm?YrL3?~il_{`ZRNc_4kJgD7=@NlDJaVCwO02`xlKC(~h|KHsY3qjkvpWOUlNY3kh
ze{_JG57C3V^b^x~DTPXbQG9y$BZ?qBLvrAw^mg-O2pm1a^Lii2h>+~I(0Iq9f8nXt
z$55R!%v+t0p{(v-I-&7l^D2}yu6MZ|Dx(;*jm8HXjVn2xDWi7c#qW%4^fTpKQo_eL
zk=}vXGC48l<eU-?J0D!)AFDMK8~to8e}i?@J{?gX;<%2eBIA(M(T)aZ<iHn9EZjIm
zdAQu-5Krj3VCQ3}2~l`G#e{0?U6uSKz#cXsYJG5?5A-1L<KUJJftg`-LPb({=f`wD
zY^@c+j6HErsbL7|Lns+jW)3PK5gJqf%1f9UYoS8&qb8aLT7F-_>@dIfg|AjXe|ZUS
zsg6Q@6)VtT?bP^N)f!VLEaA#3tD3=@wPM4#=~^*dd$n^F7}A3)2n)3P@~RjTumO_}
z=B!NFL)&OLu+c{{CBwUsDf`h!G9{DV37L{{A`dq>IK*&X-evmR$h-aMqj`6i#%tc4
z)%e60v=&j>vwWe4w&e@`=p(+ce-AJDLO=S5FEHty;0v4+vRs^0n1`psTYVI(6wJkc
zdLIS&i|McEqvSQU$;6rU;h2TCdnx*Vv6o_?V3hjAMXWu*vOZon`TpzaDA5C6T1tuw
zPy4}ULdBp@nX~@qX)A(_aY9-1P-Dsm0I$qNRiEhY{MaGz9&5JTVr{0ee^I-*_Jqeo
z2Cek%=cG4QXAa@|#y8w}zpR1(d@4>;H=5D3)H}wh59i>k_03B6xaNqS_pLbUIPWjS
zY2h8mUt>ewMpvapm2T)(xi_?p(MLBs5t_=?O!vZ+*4x+N#dJp7vg?JsNA<OGr8kZn
z)6A;HVf0Mh1mxoZ+xZD`e-EaV+K0c|S9`3*S7)~OWf=^+OxAt1hql94kFA?w_ztY9
zr*C>3;>Y3T<s9Ndtibg%+zQ&o@{jwChWgIP*i~V2)PtSbn>XlXevKgZZddIiZ>Yt0
z)6f;4$nX(9Aaco6cD)(P)?e2t|6%Ap2{Q%mW`G}eb8UmLLI1^Wf6lN{N-k>q=v7^J
z_{P=A2ZOln>gCJJy>5}D#IZNv9{603oj*qL(zFv7HgP@46T%jKGvSmhbPcgHH%H0+
zSXeM#IAz+Y(qlRx*PF_O{AMOa&g}`@+M}MQla1q>`&F4*Z^Y}XRdsffBho$JF?Gi%
zUNy?m)3$#%+*#Z8f6H*xj^1!MK;6;0u-g}ArvYv;M)dkVSnXZueXzHv<{BufRS(VE
zkfNdH!uucN&QRoo&Gb+rA9atbR&S#@Wc6C(-<s}as4{kA;3!XNk7Qhxh*c1$%!2~4
zPynVBmjzOo@I%|Q#Gz)wI!=mPhTOe1BQt3S*SlR)c6RHpe~J+ji0!ui>g%gXrrV@D
z+#$u$f8EC8M060=(nOvpCy&&tTa+KzfLtKCP$8&#?t{lxc-h!*rgYAKOK<opGB({i
zrgzS|ZFxpAxb1u0m}FT5z%;OBO)5hCi~KFhTXou?hI?hhbsBA9S##A5sAOs+L~R)y
zNq5rK+2u{;e=2xkO>7r^w3g6rW$#YnXc|05ht@ZIqic-cFb$*;*|KzOblV*mFg&4<
zT4|^_T6yT%jrlt6JtCq)Y-nQeHx6siaYy@$@nrBKx=H(b=b0VWYp9O6+J3@G2|#6;
zaWPEtM$vuVoV*n_nzk24dDHkYrsw}`HHm5Wuk8;EfA0<7)!G#w6-Il?$GY(HjL9U$
zshpS2Fc&G`=P(G;7?p-ECAhw1@gNsSB~a{fY!VL;a$Yrky3XOKTY{a1;zGf)l!9{U
zBy3kI6O>k;YzWHqq?Ba?gqd-iXnHFk!VE~loPyiOx?8N#;iMPJ#W{#=T{0{%tHt70
zIjCG-e~lTtb3w-L`w}fmdm$BnxmLSv!;2zgCdsao<ROPZEzO{ov!`4f&Ql6qnP$hs
zI<ldZO2qNFX?0r3thPkzTJ`8u4;mTGb3Ba+s~DY{BUJFe&_r95!$41LGQkMHQa5CZ
zwyl<`gVegAs>OFaz4xEekrKxt7if3}6~2kFe;T>pKK+Ve#8P)QAm;tQ&BsMLbOjf$
zm1YUSmrQ2UHYYoLrEr55rYQ7*Pk`a1;LWUoD+7*Lt;xj3PF+J!RObD&saVGor#9N2
z&4EsAB7k@2H;2-xRWqFJJZ&uH&mel9v9V|+=5D%YPFuf|Ic=9t(9w%Wav}7ko|mV0
zf60jk98+P`C*8?ISM@f3_|OQ7n$aX}1QQ1kjzWKXm`~h1Xv#__yw*(e;~6frxOBPO
zw-J?dv14pt2=WrpP7Jk5L@_tZa8M;HX{K3{Gv~iii=f8t^?h{D`Z69NSF-dFqQ-eD
zI;@^F;nqz$SP`XV$mRRSOI{>!`u?TiSS+Vnpf@Kq6+N(TF?d!KmGWw|W>-&$swMLf
ziu_V?AiTgUjom7h)oj?P)Kec|db-t~lelQIX|<U<j8Kt7gms6Yk&BkVPTVS3c!wH)
ze}2FG7j@;OB$vT|C=&uWHkV=6E))Scm$8c}Cztd&E3AKo3D54IuYce0ye!VV34ex<
z@!Ror^V?=zq%n$d>!xo%H{k^+!IYDeFw846{r>Cpee-3c+KSK#6d^Z8>rRA%qg(;0
zd#1vxBtr1GLKLBXPLpPU-OI1v-=?AE>*qROshAONq$pm(Gl;-=9L@s`sT0+%1*v{I
zdA>MrC7FMaX5^x9-hcuJ6u8FC(YQ%kcU)H`p)E67jTHIeb_p26gpng9$-8-!-WHg#
zV(hKM3UDQw(@x;(j<_;ZG2l1PNpi<5pUDfv;+&+AL4(H5B#O8{nx~zLY`c~-+a{%8
zfP1{4ZZ~X|RE~kOA2B~bb6Q$C&P@!;ds+Ca^Vff#Sv`}2fSWlKiUt@LYHoDP+iM)d
zXST{yG9HZ{4ED)@;JK<FKQ!jB<YzAVX`=x7nZhs7cS1eO%+u8w-|QW^#*KrI@{VW^
z(mv8Tb_9`4ewICiGPoEq40?}a0$|P$bk^LGR!sSErVsW?!dV7tCf^k?c#QIHoY5nU
zo^yXDyj|GkTYQqRObY!0vy+-xIg`<YJra4;UN7%(NFW(zRwKa`e?(rlU*bo*BW-cZ
zY)u;>@rs+Cc}4yuZG_N6<wx(N+@{{4J!Bm4ObKG#9z)|&CQ_)(++N!+aV;$-af`Bm
zOv&aP#}<k{TcxO(Pn<#I(9PbI8N%!xtv!Ey9+Hj6@2*|Td0y{#;7FZW+^4~TOF(V4
z>bwWf#zYoX&5K&TW-q^zOJ*;xwPwQAaQKo|8fAY-GLoaoUAF7$d@;LKdgPL%O9mY@
z{km*5=a>UEndck<ifVn&S>arijFe1qe}iOJ=S@=RDUz-M>GtW}S@%HsLe@R>R<nP3
zB!3fpjW|9`p5Q#jJ$aL+8H?mG#S=x-x|Xk(Wj=6~!HfB9`?_)lO}=+Ua*la9qZYmL
zlJpQ78y>)M$`x)$E*UZm{O*ZOIiijfU+$pjnMZ6sQe2jx#57o9uTcwl6VDKc)?bl#
z*V5`e=@jk@pr<Un@!@qJAZKk9QMG@ieOboL_#E@QRs9Ghkvtar;qbW(Z&?R=GPFc6
zmT3XSGvo|2lx%?CPhA_P8<b7d^<$`mX4;z>b@LeNs5eI08Fl*@>H(TqKcnszs6W61
za>1+-4~!@y#XCS+hCTzuM^~_nV4mBN%RLpWjHU*P?*WP)(KM)$(n>6q92kFSG4_6R
z1uH|8ai3QzSQ&qey&r(bv<N%9pl-LI#aw~;fazI-dGuUN55|iay5?4VPfYfTSwE+^
z0vNqxDDv4NZ$XT=M_4pX=6H?QK15KNbDzMk@ODkaG!$nqp|nIx?{lk7crvx3Wa&M{
z@Vqw4w`g4t^wA2H)H(Qzv($fqYs^w*M!_UkSoMu#mO4<oB1<J+GRKh9yFJywdo3Hx
zo@$)wsnxD`L@_EdkFz*mYm`<p#6+{%+#W-u!yQ+%b%mT%uJF!2VAQ)D$)nB4tbeCI
zO7<3hpOtGHHHYGHQxzf2CmW~CSe@0+j3C1ipx=pZYePygQ@9T>WMzM_yZx&Tx#voT
z9O1)DzkN}tejRc;#RR6N1I5e69N;`{Se5!r;_o!DeC9ID>Y{<A_&l!VPNhXii|a3`
z*WwCWAT<Z(1cEwAbpU4Zja_}?dPuu`?p^XE?IwrV$r`m(s+G>P>x-pY*BBYEMn^$c
zK`FK#gxbF}=2;+c)&GAX7>5~ODj!P|4vghW>Fr6y@-XBX0u@nH(YTbELyKsWWhyTd
z2)Biju4Uj@Y(8dyuVZmp@hJPkA!|Tc7g@r~(t6YW<&h>As}olshg;rL!7`P`jml1a
z=`;`O`Z^bxZcB`AGRa3!IWY675}>ksq384d*<3P@Z^#o+0#<)YT%Dh3cO&P<-TEof
zikdLnd94xVEu9p_(s~#fZk8aT4tn8tN^jd8ZPGw!%K4{Mc8@y|>C7~$9tkXJN>`@8
z#zI7EA<rdoO%lGsO7c9Te)}@NTlJ&RI2S?##q3d!ud@vi4l@QD&JKQ?(5HN!qGLf{
zgD_rCe@?NP38{Y}fO8{+HNgWZ{0aVb^XWft-~aaQ7mRGv+aIATp8!Fg@NW8<zH0pL
zG$tiK@^^zV4uqcqh*t(CjowRH>_7uc!?$Sz(qiP)u4(-dT0Q28fZ0|Z9FIM0=tvBj
zCYnQ_SXznSkG7F!RawL}m1%@;fl3}5zT$Wq6~2b2{Z4;fan!%2&G%{ZB}-KsX+sys
z6Jc-gXGyApfv}hmHj<?#g((+ieukvtEpGZsYrEdkD4B646f9)8<{}_uDVbYYGqp79
zrY59C6sU<l$MzLD3~R-fT1~>RhT}A*Fi}NCmTr=8{rlU`JnjugeQrJzou_bj0JNE&
zLj4#`C(wV2r&<^~m%>!;PQ58iQT`g8R(5ZR69wbezy>17^2^jrDXs0PF7C_K0udq>
zVlNA1%%B=UaLmc1E`nHz*R%#9(K>EQAtc=EAY%w9r@@h4AN*n4xJ1JJ?pof^ogC({
z19JH!a7{o85h!4G-wTNEFAM@^<S5m;%F6+$O)h^z=qn?e02~9H8_v9|WUdbLWkyAQ
zv^ULebfKP<Cac&`)0~5t63F_q{^7<zN?mq}((aO)tp?XH)7YDmi{{<|{_P`-?4ibz
z{Z>m^vfpZp$$tMm_1$uqeX4Qxid(tDhC-=c&5uEi*@7Yb<G2c;L}~QBv82vr#z9MO
zy3&6=It~ys-eV57FX~WoAlaB&z;AJTPpbuQ40j`!vh`_YZ8kH@5JN9{ZJrkC-Q|Wr
zwt3oI^4H{PAg|5Sj@Vn9ZKWU0EpctHxPW@V-rS=tq^a67)!s^sl4_1GIvFpc#aYe5
zc;=)VXjeTb*Kqo4L0M*|acpYsLQkJ}Q!#&7@0YI~1-TZQw!LK3%;DliBHKE_LzPH#
zX`rPgXV<4b7?TudxSAHxo~v{dv-zKa5_EWLId0paXVa|sxRYkv1R1bYzNd?0b^Iv%
zdf5kic-jM)<EP#x_s7vD-S(>T_yT4!z%VYJ$4nyR^_b3w${&IeL?c(CCC$7L6)b-_
zeC@3+fWjacGpb@G`rbh$YV43w8Cg*g5pS9@`~M-FBnsXPg;ahsKyr*rVPF7Ai>?2!
z0LjF|wMPIXf%#1M=aB0F67{Bn|31YdD(;uWyk5?`0x&gGSOiREu{Q!HxC<~@jp>`a
z5-&M?r%LFObiNtfj+5;T()7*sIjVoyHtyevZFAgk%mQ}VQO9G1`82#>R>)x*={TlT
zt``qOyQ5Ny5GMQO#`C6L(q{l7Yj5FGWjoZ()3uM(Z=R#H`ALd-;R!j4%yqs5a?-)i
z;Vh&A?$No8au16YA<!tj1K8G+Cn!5nk}ATPPPWOHSt<AW_U3P*ruw992@yeM9cpsk
zdSy9gH2azzw5Kxl;4A~0{wlJ)Kbu2^3IFWTPw>&HKf>N{Y|}Hj48W@8t>P~j@mD!A
z$)EoPSe<bUm%)E169PClmm$b46qjuADkqm7MJuI$Nh5T-zd!%jK;9PLgLPVvrezON
zXH9$1e)lF_#4#k(kW52Z#?hz;8r3$BLNf#FSCVbaZqk|FQhv=dOhMJhh*;*EB2{M@
zN%AqT@_A*N>ys{7hE;K|q=AkY9@SyeBz@O2RvcAjRgu(amGmxJr8Stp&%54d1Z<Fw
z(Hd2MyM0EPc2tqb23YK3XHB&2_B-h_*u>Wn%)5`fHPr&`>f?+Jr?}3I){W6HM1hq8
zyev_W?2s^0<Y`xZ9M@mW45fq*isQk4VU-nx&R0dtWa&EF%{CVBC98P!u+chi+7%`X
zO(@;mX!(mPLtcT!XBjO=ebzEXPui)+b2}vw$pZagA4&N7l<-PqVyL-$$+yu)EB^xn
zZ1B97K^QX=12s4}m%-pV6Sw?ID-Q_?KbP|V1t<-sw3oquC=&rWmoRf46qh_yD_wuN
zyD75XFOn76ed}cR@$mS1`0%1l=Y&Uas(by~0lkp)7bFc@>>Plw8oJjX-9PT7RF<Ac
zdW@7>)8pv7r%ycIY49`ykREyiaJ+K{XhZ`&s{d3k^qlzn@s8i|+6XJIcRXA6vsHqu
zUee4{g7A=M%E#g@&wOB5$M{a!W2ApZ(<*s7-l-r>VCrZd5DyTJ|9t)RAd?_IyrA!b
zFiJ@~Ck5+_AYQW+MjrV)Q9wxUdq#}ohe$Mx-kAIiX%4FX(~uC%DSh`mi@xKq#*Q<Q
zzdL^0&&sRNMe;8x&!Q5Mqn>Q-dx*yO%nOCyt4<0-kJqTXKS&(y=w}v?hgpA23P~_E
zBBe)dU%}3(jRm&67s2@&TZE{xRhUu<Fem|5D*@IkAzFDa%3zsz{`gq^dVW3peyBBL
zkXRwN`tswzFN_AEN$B9|Q+0p*yZd_hR6|YDkwVgv{twDtWs{N5$+3eZAxdizCqe|$
zDw@u!ATSJ$ZJvlBKZ=DJRZxGaCE7p>(MSprSVJUfMG;12M94{0hUv_IBKeaMB7cBv
ztpKZP$VyJ=8fKG|S_k!P-O$m84oO_XbE4_8NH5BhG;5u{LY^cwN8mwsz=$H&$;jsw
zGM{2)$HS0~dBkimQQ{XQN7kqsXe=e1lO6?KD>>{U&Ct(R>s2Wb#)f|yADBu$*FZ$!
zRHFrW+5j>1!KQ+Uvl3L=2t}S-TY=o&&d?L5{OVoptirl6zz@&^iePjN{j%Nh2-_a&
zH2X$7L?o#<GOg@zoE>OZXfzOYd_8Wb1S-|were*{RZs~UJ!?7g6e%0_2WBYskAEWC
z$w(QfvvE=gc44TK=GA}1<P2g<3I)mE)P4Uh)>S%<csIV&m0)&oC#og93-9etQ>=Tt
z(-it;cVeBG8%)?JK_z~}%?9&Dpz^Z@xFs)S53e9@$O}0NR1g>Pf~AcEG%AV4|B}r*
zyG1lOdA%BfCZ}sn8RjVpUeI2snSH(bwX_1!HR^}O3Qa?x@)mzu+}nbV0jgdmdowv&
zd5l5nck-QV^&!p5UZ#3a!{^=`&+ioaz3`eng2uEiB-p$Kqy1^#th|gM<EzK?TldRo
zC_atNULz?KAW7$%_Yz};J_*M|ZiU5Ad_GLWVwk>xQpHhRHI(84rFe!?yn@p03B|B~
zdGpQ*rNX*%LaBeCuc*%ooVO;I6uE28HYAr6iz|p*vWxYCo(AHE?2@X&8sb8BaaK6z
z8k>`oJ(^BoEorMP#MnBAjFXLA4pulBIb-`;tBAj*BjljE>Ks9vK)$b954t`xA??Co
zN7K=K5CLIF)6q@}?aT$!8D|?|RbyzA#CZj7?}QFB@>zdO++<M#5YOVjK5-kO0~G#0
zb6?Jf#_>s*EhDge>dzK#-dM4+_siJE(#})5^R<P1s%zdcqhH9{=ct`KCVhTw4qdee
zVU*@W*WWvu__2$#34ji?Xa1{;!)%gNA5&fhaF)Sp`o}^zeE9FTufKfx7bWHH+n+pD
zc82yV5PX04qyLG93S3V??0focDV1AoDUk=V5#DPHE(g@Ga2WW~9gHG}(CCZN?>riW
z@g|-X<i+v4*zv@Spcu+DFvMXr<NIC?W~xRMk*9$S{>4zzTK-Av^D6$yPx(7frB%Om
zhp*k?Q<5rL<>HNk>;VT$|K_AhIs&Jmlx!Z-6n%db5T+!R;K=tDMoWH&K_#rWT+k>l
z9N8upSP9#}a~>)9&O-z%L&!t<vmRfNL&YnN3`UCa(Atm}QKBIeg(Tfx!sGWZKhnBy
zwCZE=*~7U5rG-J;-5Eu;VHCrP_W|hVZ*cnQd3H_)Ic&xp9{nw2n)9#<2%-Gc6(se%
z-Hm_DeOfIL0NP}l6#}Vq)+h)a6zSg$f>^j+$~;IzE<XrDpN^no#NveEk9S7eRN=;0
z4?N=Tr%4A-JUUz?;ZfyN%eW@i%o);}OQ{0ZinN7xpy?sKVT_OI&T#%FPlFOveb($3
z=aEZ-qaD3)su7%gIggJV4nwo5I6YXzr>TE<#2IvC@!>wJCEfs@CBRZJ3JrdF89$e9
z#x#=RU1d<kjnk<2!D=4HawdLe?VXpE(V7gnq4jd9bCwUTDUi}(&4(Nzfjm!_4<rfK
z&}9>M=WnS|((Xkyt4k|S_-hx<4xMX_3%)R343`?|M+fg4u8SiypY~0<C@%9(x43`y
zU5<oB%$bXXV^r%S;eWd6v{bR&6(!4xW({o#`FIfLs#@A*#>ZQ6*94%xpJM)I++7Sp
z{pwt;C-}MTAj5pvt^Gh5AFq6;jGKAJ*R&WYN3n%oO#$Wt{@Po9tqrN$H@Ezj?A^Z0
zEx$JO>2Swe{u<(}tgk;Ky8e6!92I{P4LwUYUFpWTroOzpn^VK3#vm1f0AptxgY4KC
zWRs6n1;B<Tq2gvOLyq2<{VYOR)g;6wuL0k<Nyu8LF-;CB*Yk6+ODGU^rbae23582l
znGSBPDtcJ{2==}t|ESdcA`2P5N?jRG)lr5`m4wCnzP&D<D)@{oLM-xIq~(8+--aBv
z?LgUS^u1yJX{=p2*e9rNsPgwvxIv@}&Ce-(M@6ILI?&cwVj&1wGf~!$$02o@oY~#9
zvdGS0QRqo$*>I>c=Ce{MaFYp)<d{6t6!@+zW^T*XIT+JOPoEqN?52F?_?Q=eV$)iG
z<Gg1&t03;!v{pgB(6m;~ey4w?HA-v3EBDTqRTb8aO=}gjH*Z?2@a@pFRzuw0v{pkr
z-?Ua+!TwEa5J*=t?C7^O#7SDY!Dsml)74O$8>7p%KUyc@0NX&^R`b-BQ$c^f#<|pH
zR^dP2W>%~0w!S$nmC#Z*tUoVwoR9&6s%J+rn{i*iK@_~Bz|hdM;BtT37~zeZm&K;b
zH1JLYoG|E93|(bs(+ebcI&U#^S52gXe7UnI-hAORvbtmI1!rDa93<@!Wu_3LZ!sh_
zo-z*AnA#qNQoBWkbwd<V(g=glK-?(`rB=8K`kVC|=7zIN^0@*cAS=IyxaD3ywYb#~
zZ_nvkX@Rb>#ianFs8oM%OsMJzDGuW%-5{rZA`$QYocRgY`YQw7PIvM^u^eD~$i+J3
zF6pNTxtnR%S^(P3wA<CRzhT%1pfc9K^CqUkx>HCUQot1YicL+KnkxLacHpI6w3^c`
z|2u)C%a`U;h+DQ$xe%=&ZrDPlVzmOeu!Bmi*R=^w+q>U5+)64bRN09BKN@cNu2F3c
zSnekTTjzx?x~^{}-EWSyvnO-a=kx2~Z|p>L7?;6+C=&uXF_&?kJrtK&UMnhpj@-KK
zeZRs#usWhd>H;hlo!RUL=v#uk6@54en)D$ldQKnu`=zdyWqZaRPqqtqJ(eYkH<1)6
z%i;SVhmQ~EFAu-HX&kI@1Xc}Sz74=HB>xA=jfMsbAdH6L%g^C&Pf{ww*O@PrQfs~}
z|8{(*=SK~WdjRppZvdVjEdsZHXy8kwyL#j6CH#JV4DZ6&2rG`ap|?!ERfeqIa?h6x
z;Z%CcbMrRzJOx<K@trcy5gW~;<jeD;@)CW(j{5=A0H^1_fBEZ!%#7%%HWmfbR;Uj1
z65~XCN2UN@@^6NQj6E^-3e;0*N%IrqMmoMxshxkDbj8sTS{N0uN0Z@yK!rmQR(_&9
zI?iWnpNK`GLnv}i4N@RqPLm$~WXk5PpFgt19V|&)p%YT+o_3{2$px$<W*;&xfg+nR
z@e$*et82QYB?^napGm&mjTSAvkd`jd5^>KmrA!A=l0%U%*=_V#DM+~mdLEV5Peryx
zc2M*oaV+ZZs0&4}3Qmic|6eNx3v(VrcjF*hC8WR1mk?kp7k_>1o?gE^{4ra?q^;1>
z4PSpggcqnmXr>;NVOg2s&wmWx9zOM8Gn!IJ9+Uo&Ow60Ke9pGhOQuE|Lo~6S!2Bf~
z3~2#Y7iz+)2`jl+kf0T6${#6=<rzQ^5ax;#Pg+Ka&1XgKtEr#$!1}F_J^KLXddR{4
zGOz@-XzE+idw(M#+w!uOQfsw^qkT-$#lYF}FVa0}`8;By4C@H<qvvkV$vqIA;H!rS
zbUQ?Agtd0YvDe!nIud-%aKuH4Dd)vp0of5IW?$FttE$b*NrhMMtJ>3951R8)y-9e~
zJ@gIg{Tg8JS721^{$NIpyj?jcK<WW@y0H%+u_DaG41XUEK=i_rpFyN${Vi*}Q#0Ax
zktm#?Y-;X{l4!z+-Qv-TeUY7W^KB-gIzk(vwVPr4(kwalw&TAz|D>gvzb-Mw9-G}#
z<rH6XRKrMde$L8W)Y4C_dt4UWm0q_*vMI*DzLY}Dp2ztNJNSK$yK}%3!Do>1)j2=w
zG>n_#HGfm8J**-Dl9h_Xy&!-lY>3u~wgxFlVZcU($!r8!6phUfhaq8GlUtk}-|55H
zlXpT}e;9kWf<5iP{<vo#c715zojpT0>%lO!_Y58M>z-jbh`-%q<N&3oQXURaI*3PH
zhGSW2m_ht-IOORe(xs56S3fxsi)rG~^HvWMtbh0+OHL+uQ}uta<QvV$oy;&1vIz8H
zL*^6L8KX719WwBvERWEN93?nBOzY{F(dt%cUE1(o%K{Q`WMiua>za?cTfPAlvH9pH
ziks3LOGGAYOE3Y)!@x$MhZdJmNgomxn0ZmXloaNeQmHL-Jj$D3I(&ITtnC|8hI!1c
zlz-v4Q1>Q85JH$SONclnjDMGO2*-(R;oXkN;F{O(AeIzql%Hz6Z#IXueJ0Jk)Xrh)
z?Dx)Kg@PG2Yg=;#?4bON8?s4&?wDJnP+HCBt})*t4rqjD5qG2Rm>)onSkT{7)DwWK
z65J#44bj?ud8D2A6x%gT$eZkrSMpG-fj4CN%@{1{tO%#%{bK#Y=NQChMJacmFV|dQ
z+{@iJ38+aJ(D&4jDa0i&IzI0fD_G<$3G}mES(jL2D@9xjakSB*CR>z$)>H~CVJ44T
zaq{%EygMXQ-!?G9_9Iv#gz!e?g%FFNy3VUWpP^Ed3Pu%L<XM7i61<Y2i8!4Ll$fSb
zEtVnjFiqr9mtR>d_9EvrS^*xXoRgQA`C}^{e@;0#EQ~sjf=Izzs_#{{FJ#k7JupG9
z<8y5<3v&!Y10geYWk+MM<{IY;Hw$Jl?JQI9+g6-&7)otei;h(eM^za2rbTTle3u1i
z+XbC!mM!E=?J2)??c9`k6e0W38Ry#HjCKxstkHX_^V;}Lik9VJ<&r!jjLKotrZ`Kl
ze@wS2bVoi#Xtjo4ushW^Q%T>a{;k*@>z-4STHv-tceVCy()o~XrU)(TPubAKtykh<
zQxDHP$o(U~QlFxaNfw1`w%ss2Yk7t1l|m{et)X}-pJwyGI<TmBO0n@8L-s;N2`qyK
zpRz6%X}|Rw%W(rbGxn3MCeauRMeIwaf5?oR&4!CNZ6z<QF-gL!Y8&-zOI#)LGS-k2
zCoXTA?QGM$u-(3{o2925c};fNmK9DwalJlcJF!&uX*`v)idT_I_5q<s%;cSWY1k$P
zMZ2Cb=+^$~X$<8B%%#GhN)mW&eIkMYn2oHsopMs)##0t57#gelq=IL+itm$Ee+XG(
zY1AFJ5*@?^*SdTAU<`{-^^jM}iDk3V_0~%jhjZFa!0X+1?utVqKz!%=O*d<BRF@6m
z3g<YgQ!{8A>NUTBkG$~IxZvK4kJDKXIQ;(bPX~DePxavw5<NACPe^nC?{Gq*hj{#i
zL=W-u35j0K_Mdl9RtZvS&g@kKe^m-|{oKd$8my~A8T<ZKGQD9}aHk5H0v``0Zr=Fj
zS`N0Q{VbXJ;e^B3NhfrZWZjA#UBHf3u)C!S1=dW?qMB~!)uxluE*Gaso0uGtZG^r(
z;w-)G+Fou?Jf|$gG)PC$O1W!|FUnitwdtYYYf4^4ykF9uJ12fpbZl=ge<FbnQ_hM=
zir;#Amv|4@@a{WQd(CI|5Id!b^@EMRbC21TD3G9^edgn?#{^>?N(GMh0FQ?N#n*I@
zH-rEy<=Y!h<6UFZ4v3n!EN-SqD@B@ewy&rCc0b2~sZfBst*p7h2P~~aUe~kFT~KI+
zVNw4vaKDd3?9Ah}XE*=fe<3ytP{huG^$*zK1gM(-0?iM<{rmg3zkL0j6>|9gpU{;p
z1BYAS-S9I#5{8I@x&9+R8(w+{o|V{3ZEzsR?6TaU46F@bhX-W)0-9XY`XjV@fetxz
z83#vuWJAX?LDN)l2oP&)neRv2Sg;zWh%{9+!c%~f$A(89Uq*$;e{i+Wr7Ms6=kV}t
zc=(jLN=Dh3(vu0WH~hE6RS?Ju2rXwzQ!vWbqbXeRmeppZwWYT<Dh}}tYNpT$%jy)2
zOP;OLTv{5sOB2H?5K2?I_w6fu=xF7Z@hpcR&*Q@Kvq426b2o8#{QdQ39`~I`eQfRt
z&LglTK{EV9dUiJue_%y_LvHy%O6p#E7sFFreypbC)LEKoX@PY>c*;*hgHodHaU|~3
zW`+O=BHX}`-lC0+uq0eM5OV25z}(yMJ3ZQ84Bs}<##~+M8KWgr#}yZA>-+KroHivF
zr<?Cn%C{sRnn<txk}g5Ll0Ai3Uol0Hb?!com~g0H;c*|(e`ht%IV+7#UboOuX;{dB
z=J`%d=Jb3y_XkD*2MV14Hl+R(3By6=bl{W&ArqM~D!iQHh~sW+M%}h(qF1D>b<LTq
zh4R!(O~S7p7$f#X;RDqPxaO4pszMvKNzf0>0rJHeZOAI=ZlHqeGdkX3d)p0^Z*4K|
z+D2wud7w|uI8RQrU|T9!*uvbc_!K@F#-2_4GJ5OvV-(<e;4lsN7Yy)d>Xz*9XiJm*
z{a5^7!>D}!A0IqP375ftC=&uXF_*DTDifD*h9@bPmuV}(fAi_rr=Jh92x8>gIuyhk
zbFFEXp;_qjETN3|YLWu!=e5z(pM+-#{ql+A>tR*C%FqhqLMG56Bn~7o89y&r$lO$x
z0Ra|9Y!Q<ok)eA#vaJ`W!h-H39J-RyP~FUj0xi%ku@hFKSnUQGCtb<d%src)i7Qt_
zjAM+B+!sl>e?$krjZfzevZT4ei-DgID_j@Mrd(wsg+$#cMB^k^$np<13LsuS4No7R
z4u34H&?pop`04BOA-+HjLX)OIJAodjKYyRT9X^j}i-@I=G$;QvGAp`+d@1f9Bxyog
zYfPtSAX_jvS+Xots#?@TkqyZR?=#m52HhshSz<&<e}Tk|7`H4%V5F1(CGtDSm&R!=
zh#=jF8DQPTttDSk<LDSnqCP2D<n_r9=6`AZ4$`jAvLX)A>biamgvBC~ynJIJ5Uu2N
zv@!+AP)1p{^<vo{NxGrPM$J|X1OvL#a0w3`m)Fs41W58W704j4(v4AeF^In&Gk9Up
zbP`#^e|*_gFWIRHR#@fV&`t-8mm}P4L;B<OHXJC;dOxlTX5EH^5drmrYa&5DUcI)G
zUN~tNYui<A)maKstR7~)!hei6!V<=p*MAdhv069-Z)q((#y5;=1Lg}+Z5aIqQOy`7
zTu^_xsAdqTn$hnN)kvViJG^M8YiapXxs3sZf2-<?^==Xy%*L7SJr^GPOOY7D>S$TY
zBbybL3oA5T(c|Z!hps~J3Kd*FC5r0X{q)M>AT~i2aQS9ynb+W<5)V+#*8kzr_=!?W
za53GkNYOf>WX(JJj%&R_Ec&*p#GYM-q&g3G)_Iw#3%W%wybSNH+%h+bG^^8%>I@w3
zf9l&&LU~<VJoK?dG^X(GHiai#ZjK!d&7B!F6*gzxw&sUyHu0Y0#I=>R^3=uvEGlc#
z5K>;<Q(2pmbLIrZkbRXkDvH>lMv<zYS;o8akG4cPjM!AkXsMajsVXQ@IkwT!o|4vd
za{su_f;x(IN>>kBYVZ1d#+3G$m0Ztfe~7||aj;%7qO=RJ&eSah*s$VOx-A9RNMBvQ
zC{mW|!|bRf6)yxXQ7C6O%gtgMHRVEnmlFJ%R5;;VN34RLpxSB&SxNXhXhGgr6{P2K
zCIz5UAf;^axM=NX1qMM%9I<WETMabI>NV7RCA0}G^}O%2GmI}wt6Yc;n0Ih;f4-p<
zM}btnja<luqg;g*C>??sv2K(^bAca3aDKt}D|F+iWmwgWD)#iymyF4G!nTdEW?H<L
z^`hk*W;0aryoc*G#!_l&fc2oJG^H0m&}+_(Os`lMtq7EtI$Q|1+lT?7%)rLFLnvd?
zenBtx^{}QbErfGlbRj^d7WN%Ne;NfN^r{PKHdj1}qPunTDj92-z1jJ<fr~_03Igs?
zJ4j2TL!y7NinKLjxAhOsl~Xp`*@+j)>MAN&QpyQqv8O_{)#_ar;x_hJ^PF~%wLn((
zSj(HT$6B7wEsj{Fm#48Jt@C5eSni5##VQxWoOqWVYO+MrgrBCL-Aeo{e<!fQ&K<_?
z-J$bMJ2c$g538objO_pp&WjLzx}UY>snQ6G_W$guiuzyc-Ho0)%zh_N9ah=xss4(6
z0-!52*Yz%ja*x-GE=ET(xrbo{^Kw3|=IMR~BV|-g>LHB3{3@q^(954`unvCuZyNes
zSCP|?>F=>gx`5bI(?W=ie-}W4eF3qYln7MwXS6d4qk|wvz8=L+%FvN-*_Y9hbp|uy
zO-<fLTLv+kYNDB3+31}H$z~*%!Mzyk%t&VtRYW)QazRGqUy2H?_mL41k4+<dt&IGY
zGP;?UOES__WcycH__Zi@IyIq{5N>hmF?x)s0jUDd5$KPJMtz#nf6zn1ZU}t>N4F*R
zDEb3Xi%YWA^cREe;lsbbfBW0l-zXhS-~SVbQWztv0HW<VKiaq<8bA?c`8k@BqeK;J
ze@&%|r7Y)w2JplR_&Obs>KV|+H_#su8f&e1`ZrHb5^!-KFVKLPhKR93dA%Pkkg1xN
zNIX?C;#0&@=Eg^wf6uexWBj$xJd|erb2@yR4xfuuja4q4*~^SLSo&L&DwXNrNK1`m
zD=6-RDW_Q~!BIuD1#S4PKt-w0#+?uq50L<4YA>iO&kSk!W)R_G3=HDy2;Y#yz$-!K
zuOfsA<Xu$PBnyQi+$`Yv{_As@)~8%Xa|ER+h+Db|R2*Z%f0g2z@#|l9<l#J;OBvqS
z3#p*ZC0OrW6tA|B(2}30hNR5evGVqLw>T)YNv(ItWAyn(4%90KvQC!6#<n-J;7og5
zqL&+;N3C)1W^x|6cSAglPMgJ8HkdPnPdf4w?{3*{FNoLTCZ@+=+fDYS`RTaQyqR73
zOSor_m)oihe@)%Fh-7?8cDCW&6%Nl8Yj;Ldp{uQ0Ioru=wr{X+5?Gg3ivV>m2+h1-
zgE$<bXusHqW@w_$)~9|qu|VtbOdVrFk=8v^oyIj<Nc?ST!&Pn6qYFN~G_BUVPkwYg
z2yyco4efkzp?z&APpUSw+!oq3n_ql}r2VE%_*UAXe}3vTAM#}G&h1BEYwKmZSl4aw
z>e>K;K0`eqs-IwOd_1*i%$J`(y%6Q*fU*|l=8%=c+`p!NSSd!1O*O*tImB;#q+L6X
zqpVMhyY66Hms*neikpKmTt&w2=MDEH-g7+}<dDV2VRxMb1}cSAZMAhP&GeO`*<3XI
zG@qMRf9ukNHyhbnX2!TDw&syCe^J?<&Z_owHq&t5KC#Vq7Mb6)-pL;V@F{{F=tjR5
zZxc7>S{a4rbHFT*7rAaem1O*FBkzWIo2aj}ZJuRU+h2TD>(fp`J)b@q-=EfYb~QR#
zPIukesca6wOd}BB1O6JG)GUUi4BH?7=YUgL47gi$;l`E!2d{wO{g=UiC=&rWm+)dJ
z6_>%a2q%BNTg#3#rw-oFQ*?HTbbbPut2An9re`|R*(I7?lvR{umBq>XFOC62fvM_y
zdQo$sz!z)`pK;0EUpu+`>E-?F<;NqKofp9>uXkU6?i9X|_)|%2bav;J5)Rbv>+jvK
zuTpA<&pSRwN&tL3{vGhQ-G2iWu!pkv;2Wj(-@JcO*a8Y4U4H8$K8N)8{#$yN#%5Z1
zyidJl>8**fK2pyEQFyC8<$L!w^?XaT_W7N5`z#IMD0$d_(@~n3q{Ds+bx?=>Z(sj?
zku(uM9&OkKVWW4jHVe)hfx6aN0Hk2jgOZzL`E8R1Ch@c=y+t)lNUii1wJ`^Jhgw)S
z^j3d`jxAE1vx&1-IJ4L!@p&(LXQ@KrZ$0vmkXj@0Nis_NO7gs?zNy8e7DrMa7m1tH
zNxru>Hua>|+{0x4M90)IG)Fej2$IbB*3(kUk)%WW$zyt(=B3T;vGor{NBd6t*}c}U
zq;KWG=eX!5J2@EtUaLG>cJnaI&Z@=B*o%KGUe<(BG5h6Sfv!Uvo5~^&Q1sSkr<k0;
zfN;T8C#dS7$i;<%T6H09gtn27PDbl^Z8{(XVGJLfZB{HJJKst3a44%YSxm{jcIod(
zR(I5|`emU&c2zVrS+KshtNhk!5x<3=BYPd{+Mph;mpj|dmmX-N92rD(DLABGKe2!R
zKEGc6H`&8PoLS>{zyJM`UKmgUg2pNH!UVJX=a1c=FF$j#6n><TIHvrta`(?)JJ#|G
z#p)>0EYe{B7|fz1+6wKWm5{#01@l33SM^8BXvN}Qc9WT+kr?$Yr<m6US*5I8nk$p!
ztUt2&Lko)EC_5*V)10!FgIujQ#$A7hy;yIU<S4s=L@m`P$#i~7&&{WeQ78*n%qPy8
zNAe1GDly7jhn8Q?&pG3&f&ARmHAPX#wivZOb3sHsQA@LA*Jcmbf|GTv!spd?)o3B(
zBIl<Jo1beaTH*C%1@&<U#VDc=6Ggi^gBox_gPy!VKHkr~M^5(Hel~Kw*Ux`OY8Uv~
z;qmi`=OR3WJ}~}qithS&XV81hB->WJvtBs#HP-uv^{AW@QF=~^tw$b}S#Nnme5Z)b
z`P$Ro!{%8&)-M2rdG_k+d8V~h(NA6xw#d<+UGiBmZm+Y5I4p#b?)g=gcFKB!FRl7E
zKh{<W&D4@_w})aIg~xN~Nbi4RJrKJf@EQ9e;J8`^?}YcB@%+49hhPOhBR%ETbqG;#
zPb?8|?B4ly@jTYjue9pjTH~R7K!afE(d9SpnbjWxp}DZ&sKE;fp^=}^UeGYjp}v97
z982iGg!zqZqq4u%{~C-@%bKeaR1{TY>t(Wb8`r5u8y-$OVMdZ$a-Dzk9O)7FF1o69
zEWZFRgb9eItip2=;)3b}8@TdXsTbVRb~bbgi^m;r!x$6MqnsM#`2IgX6x5gYskI?m
zpu*#j?|!(}Z#N@m^<#<gT*I>GC7qVXNVZVdfYiCniE%~%hMmg8@0H0w(ssDpjY*K4
z1Snu~6S`iN_9>9FLSuimK-UUJ_XXU(cZrA!Uk@EdUq|@bp$Vhrs@j^0krgVPMEW{v
zq44m{sa%@q__z^7gGCfEgJ>rrxTj!fjHJSwXJbCb^?~OFM)}qg19zV!MOvYGPUQ7l
zF>Re(x{U)s1mWXkn}SrC4At7Q*<xrpaasgrQ{&_TOYZa%<N#8D!7|K#bhvts0DZ$o
zn0!Otd|-iuMWp5w@4)C_0mH*IXLn$9NC-$x1lQpMAjrvKgSpWSlkKGdKbh^0(;*<V
z31Kqd#`zEeV7%~?i+4xF+m~#3D=dF{v|&qbW3R%4gE5n?x9&k;Km?`u9)xQrU_^h$
z9)#?W?z0gq<9ef!8r9&1fk~J1`w!>&2-M1jxF5H2il>n)>uuk=RgEiX`H^`KEh~Tz
zJZ>@4@@-I=3c^&0Tn(JHO}zHBE63^k(=3QIaqIKXcH5Eb1CLIO^3BPX!J&Vzcj4+o
zOqx6L7L*5J6gM*$XbDif_XRuC=__?Kk^mk1EA=zi&zD?{YqVX21w<~_+_c()gn**r
zp(VKFZ4u8yW!0^{=~th&0;1v^F1j}N`UG~`K3vXP#wfr=F2e$oPh&b2ghCw8MwiR_
zCM8|Iw)K*{{f%@<JwLi@e6N4HbWa&?OrsPRT~iWerf*$;RXiu5XoOoXXxC?UC~qh&
ztW=<4dozt!b$m^XIgV%-%`%6K&GU>995SIp@pQ>=gPXl?@>O51qpBUQDsSj7xL5tJ
z&S5_3Ep-L0r4~P^l_HdO8f@=t$%iZ>vsWCcph27NN_Metqf*mmV_|;<<v9AR$JNCh
z%b9iJ+Eq_f9WM6l+lv5{)xsbi;`s}NmKHa*XB-|i0w{3$ftO&6;sc8S809lX0Ic_u
ziU4#<NttI~&N8ldUd}RV|9TMs<M)6f08a7#A^=X&oh$<2Uhw!L0Ifh+{p`~-<JzB|
zhwF}hr#8zTO1bFy1Z{ssG{C8!@Vo0hjjYHxe4Mu8iaa;mBN{EdQJfk(g-tS8+ymrf
z=NocW+8H6$G`s3}S8}kT0#4PCR&eTGb4#h~!L61Wx5{t^+<t68T54T+@A~&@`D;<H
zv;@DrgNH+86>!--doLN8K6r_b#Q%jA&KR6t;IlNhuJMvM8tH#2eRtl75!|ReqXo{i
zIozB(U|i1xDgylXb1$)hF~2WL0fq;Vd}_+~2Ae^u-t4SvbVuoha#6T=W$8t`N&|W{
zVdSa&qR-D$OIla4{Qz`8i@(<u@pspBXR;ApslsM9qA#-06s7Wq88BT4u4qw;DAdLB
z=oJLoaaqBg6=xSNdc2~4<0Ti$2H{~rM3j%q@(h@CNJ9gat@kAeFqF8s!786a&d<{9
zETuHiVo}|sKKo55pC$)hT$;3vWr^B2d!SOebs1aEqn9Be-h6=@NuI;ZJN;kisICa8
z^RkAmuP!_3ZE<1Q$uMLrJ2{bgjKXLw-CB-NDcmgOQn6*b0D^0OD@{OMu5MW%skv-v
z3f?}uIwN1d<k4}TzcJcQC1t0%B#nuitMbd0FI#~+0$E|Gzlh!EH2IS!bHO^LUR@+v
z#u@Tp;zzNE-iI!8vVu%&<xJ__`yx@3%h_kpz>J@cjI`ivwr<esid-hqwsr1coRw1*
zHwGQlNp7_E9UCowt_y0_PI*SE@&jiWlWJX)9dtFz7(i&{*OE8w=ou(!nZLWzHJzIa
zqw-16cZpM;8ZG5hwScMpm#3{JmJ6k0SM*fB!z%sBv8dl!^U{|)7G7IHR2@5TZ0_Er
z54(iPy6nBnye(g{)y=tl<!4{nw{3Ofl#R|1J@yV>#|ZL&u6YD;7uHH%K1&pu=Px?(
zOt@sxt`)T8TrXIIOZN9<<~guW16AnaMc(`0>MC5#E*~CRv7jTK(?ejBbjPLQ)_Ml(
zgvB%d$f)6@`!z|mF;?0Wn%sc?wzR>$Rzw?SZHBs%Y~e@gXN=8(?Sy`hD>1RsCD=Xk
zVn?6)`}6At<^R_1S;d#Ze<%|HIhSF1Kogg{eJk#NRLCRT{rU4BpCl?`p0y8Fz}(ub
zKS=2!j?L_zt;~Ho=+A<D`<6esR70tS*|<<Y7TSP1q%--MYLs8~P(M;>txqzFbYGuq
zFHnAH{kxsc-?<lyV{xAK?5L(q*oUg}NOtK@<yLJp2R>3ulJJ|_*7`;Aq3YkMHY$?a
zMo{K|ta@){CI|dG;%r+@8><D~3+fGKsd1KNc77f^C@gEg)G9W4VVZksBcS$irO)gM
zD-ZBm$f(H!idAT=s4?yJduA`SDI02CW-F$XJ+-oydO2;uR_ic~jC3A&Nf|-xw$0zF
zSkLA45_ZZvRAhEi90(Z27o*f=Ra6_vvDeOjM8oG3Z>8BZsST;QODowkh4!HbJ|1$6
z@^@~XP;MA{^33dQc{OFts!z3FCAS^INuhKL32oM?yJmL78U#yEEu_(m@wb338mZ)*
zaj_N}7o{-c;_tT4p@pUVx5K~VAW3bHrz|~><Iq*py=5*rHIqx0Iga7VWezTyt@8SR
zd1V+Em55R;Uw%COwm?MxxRkK|B!x-QE&p6{F*Jh8-Us+LR;jf74prX${oB_+eEC%g
zXz-8RLM~Qlr&3&g)JNx_S<+_tj?YnO6<l1DNYZ$_`0N!5Dp12#@MU?jItXQ4WToG8
zX%aR<ykd=w5+(+y*g9ZEG=?m(l~R9y-+3X8s&`T5)P|9tvJ`BaAE7<9%8&WOpQ$Re
z`s?!ab$R;K^qCC&gKr<C-%_U%Rt10#Or=z0f=D4*PwGek9N<VPpJ*%8!bfjGaXN}1
z75vqDYm=oCHeo6PStNmq=zA_dqJU|xRVCFgop1@-c@cbeN`+djn(cpo`GL)UzCojp
z?7ND22EP#2`Q_Qe9Gx!?UH}_7#};72>Qo!!l-ci$Yq1ztV^1_S($$^`4DmHK356ki
zTG$ZlRG-^iKV7UIfRzsHcwRp<MC*ES$q3_o)5{ScOH+_e7}fQHaRz~sHyIpC$_mQ7
zXcoD07P9d|Af*e?EU>TT9w>Exa)JRRBh}t9cD7z@tM4ZOh%prV&<r}7{<msiOs8sn
z%15<bkzXb3={G2SMGe>~x6*t=&2X5HJwb^Iq+ylBn8QsD8__tEBf>`MFWr(BT*2B0
zS*GVIoQ-8U-wl=__*JsknCsD3$g?QdrwRTJ**&8Y^?utD(Ca46h@cIB>*jti$DLS*
z%;mS%VP8~rA^vR{t6|1`W+{^*3asxGyxF3F&RO9glM)#^OIScN)sIVwGW!qAE|HKO
z*0w9WlXY;8mOA5wis=|ZGbUjF1+hd#PNj4}PKNM~BQ$H6&$Jp1Ls?vpv{qcPzmXLi
zLv7D|vCT{3_lK?dB5t&Qw5qZ{VXwB6OC0HyYXjw65VcVL6t%GQ4BD=<K8B~)NV$!R
zaA>yz`BU7);7&s}2FF0FmwDB-sA%*#_IVjrv5&=D#l{thsG${--SVc8PI^gO=f`rh
zwpzrn9{&X^34-M9D+yZ@YPXfc4L&sb&;E7vAME|S|D0O0ARI7%qMm2p^_}xk)v-;C
zEA;%>7$jN(xgb9E&4x{%QCjHGB87cr5nxZu2sN-DSkS{5R%5e=R>FL}WG&mX)>>Mr
zr5R;^8#fEEmPa=2r_CpSial8ixx<9h>?zX}XH&-dN&(U%?0+GLg6HU>Z5+xA4;(Bt
zo&lT(h5!|)VJjGaI8-YgI8<-k2@`d|if9b~K^#f~q}HJzBUuF}V5tA6Qfn&ZgnAH{
zf&e3eG8b|ws!b?^)?CRumvSf8<tBG(le>~U*}KS-bwS9aE+kLZCjmk0DhQ}8OI$Ci
zftF5y&Ll6R4d3=E@`j)cB@FTqq>q>OZ9Na8NSg`_n*hLnspJN@O;{gh4axnPR#teu
z(rcgREvEw5a!M_Z@85?(H<yC7J|doqi*=YiwulsYS(_j3k>_4Na8lKhE~WOKIL!<|
zcShmHVFTA&-cz?;Ba$<Mh`7y&urcY2Xct>7M`@QmPrF#Iz2M|6$73n|QHm4g4Cat*
z0$BqRoi{pv9mL<9w3F18j&J_LD7kV_X8fw}8A|V)KPV!4nyoga(7tb*^j8Dn_mBss
zUYhOe%s826K)*m(NwqG4&?c(8A`#5zHA7)FZhN5IVG}&6GujFlXzLF9LHGh!G_z(K
z5he;Fr+IIa@Y!w->;|J<Nl$D`1g$b?Jsv$&0Y)N!pf?^L$`~msnPZ0`{UD-f^o0OD
zKbq)qq-=A!DMu*UYP;qTrbVzr*W3pDPpc!KlYUxbRPoj3AvgP~I!Q%I^9E1}U_gGZ
z-?D~ev$jM)h7-#?>(`z3?^%|xfuFy|q5Z6XPF~uUB-Xuxo|+3bWseabqg4jCKke8k
zDP{S8sgX;;4W~vxQHDdK{Jc6d%7v>VqtbFXF)G4`6Qk1o|ImREL>#TWomkFh>3{Em
zQG{sz2$o|!&_4z=PO}`U{m>I5HiRQ5MmX7jE9)^PTPt12dJ0OnZNp0M>EI@3w{~|&
zwWm1ZOvNt()NF_wHxN6kiX+DAs5Zxu6fj4BJ&(hvn%jK6ZzI_qofq0D%D)b`Qv%(G
z8KZBlr4zCU-D|5<Z%XA`R_a6j=(L4uIHJKVyi`$*4*X8cE)Q28PVW`U1o3U9`8?Z#
z{EXXK`@U}y+GdM}!-y}&;h4%F*>+ESgRV+5>*KAJXP_VP%OJxQip(boCv0xp5ijw7
z140P*BY5zcWL6`5G=c~FNEzl{&NC^t0PkT^EO}qRq>MDe%f5b?wRBQ#gOx0T2ewJg
zY?|RBo%o9PYi3F9=|)x>xPqlV^u~=jEWP#^G^tDOi6y-r6~8A|f8VoR61GcVeH-0)
zh)Sg#r~1y}?4Z+jl`H9K?Xd9>OL6^wE@Q9n4qbLRl1^c*-7UTW9Y9=(jkz57#Vvkh
zc<va|7!i}1<(=8}Q6{q;J-6z+A7$j89W}|}7+~_Aqu0qW#`N8vBdq9#i)f*H&#BA>
zJqs0A`Z=CwPVbh1wslxZqi7?N&jBZYeVm>k0{ymj0t-<_mKc<Os9;!UFc~3#_Wb^r
z<?GX@St`FUU8XF(2UN+8t7K`feFA`KI62NK)V<(fZH6lJUkT|$2@mBEGrfwl#F~tU
zK%3J*20*ghxTsQx5|H}tEy>MsLqt~E&%K;674>{n8$NS_FiXYkf0Y&jl&JOi<fMUX
zBu0@yDYIVP?Gj@U)~swVKWyKB{zyR2fG|w8n}@*AAkWMCAKs#2CBN{Ev`l7Jx{;oM
zotZ$9-@Lf)<$PTjOXk}_+uK6T^MQ#5Yai+Pqr{)(`C~7}F?jiQnYUX&t#+k>d%S&=
z^ayXCbol^ppJsc;+sE4Xc>C1v8E@}T1Z|2>>vr}1q)*!*LcjaeefmCs>yA<%Vck*U
zBdj}0JZIgp_C3~}`n}G&cqB&ms4lYkcB(%|Dzq{}COaI7wg}2LLEDXYe#5;1y#ZM9
zcOm-Ck5aea9O3Yc-Ol7M{Y>y~cyHk3CIyFg+wU%{Mp(D&*~=}KZ?T8><IzXEnK|`A
zFA?>7|8|fuQ>3f5>^EM231Y{K{Ydi^ptUJ)Sgoyo8=!SM0ex6WyA{xSD~z@?o>eB~
zE#1?A787Mzj}lktQ1T0c-Xkd#!(`rdP>ffd-;uvGrOuM+l2n$c?*kbaI;S3#qR)|l
z2OK2t6_2gt^cI;0%Ye=EMPGJJ)!mRXfHRt?W=T`Z=vuSjSgUS-*{#810$-Izci-p&
z1VxOV;aHZtB9dz*tY3}dY$W&NOMQi`I^0HMFCO+t@6_<*czkIX77q%so;khQnSYmd
z$`0pO$kF9NZUFdPn~n3*%T_|Q=_um{PjV(7ZB2si!5+x*MuE!SA<zruXLYF~$OEy|
zq?fZUJ+4#KZ9YSP3;ku!{HpOQW(;|~qG=kV)MHCFSmA?Ojp60KF;JKqrCN0o#m%+J
z$sVu*0M(g(ldhR4T)oy-l5>D`omaT|?V8I|jtjy&vy#X*<BA*csCb8FD0ONF%hEkk
zGbC9WVMF*CNqLey;p=-7b2LJQH21Q>{#V{UwlLv_b%QH^Ve0!dC?M`KwUP(ElhPSM
zMYgyX^Lo=bwNp9&&cl3r=mv>)Nd&ht#0?uHI(QL4!X(MdP1KaMa4J3kG>w8zW2ULo
zBaq49OK=ETeiDE(`9+CY4;?@0vq-58@FtWr9<gxjRFIo#C5Leg{ZRI&ps1J+2h)8h
zYLam~iW&trJpe_G0v~{)CV{6>)TEc2P*i@+=fMz)i#&s$hTb0op18;}$Z4p0*rM@d
z$>|#i{l5Ui$zGRHOCA#eHJ6bCG!_9kmw|pdDwms!E4zOIQ3gE!`R$J<o)p2E)Bdaf
zGyZygJN<SV)uIs?XqdkJI{61Efk`VSX-Shf{r=bK`{~O_wamq72vhz)`u6*eX=wSH
z4C6Q%hHI@<XBa^okgc?KO!`OBFi3(aQV*Izu>Hwql$B+YSWc9cO4Lbh%EsRcx>-J5
zEiCe?4AXx~&M4|ciAN)<Xrw^#0Kco60Uork(J+)|S2Ih-`E*rwvP?On<<fG{+%kY7
ziA=gday`5x$xNElyEJ+yL@eK2lHBu?`}%RB=yY#ED(jekk{B>&(B83?@N;0W*IB^h
z9$%{Pr5_Vdx;)p?FBR@}=G#C%i^Q`Hdd%aurAdFiROZ1yea<st;U+zGo}Y$>OV4!~
zp%Z1|OUX2|nepVk^HRoO?U%>4v!L$yxsaq#v{1C@<trm_l#@tO5RwWyJr>OdsY6Tj
zZ=oy&3eJ!|R?^}KK0a&xT2|G}6BGyXZx}p5ZM$N>Tz6c|7-lpU%De^uFNzmCC_F^k
zpb&qZqkotg@~=kt{M@DS*Z2u4%p7~wKb4*~MsoF>=167_BYhNqNeaC&a+8BFg?5bG
z0E#JXwnKLa%PXNS1s|mpLxo+aOHLT1NLC7A@3{$k=;K@dFStb4(uQdzDVW-ECc!%!
zm);tf<7UR!^X>C$O({=|ln{ondILp6EfjxD2a2GUCCM5_WLT4l_o$*xT8Ns5g?PD7
zw?x~zaf&t7W<&J7BQy7_x2FdD9qL7<4a_LwvNXmrC)7-uJCtZ812fa&dmP33j6O5{
z;3zjybc#8YGihHJbt5?SQm|<0+w~*BFahcVEiHtiOt7b)+S@T_><3s{NMu5a543-@
zAOwxp9F(Ml1R<CriJF8JlT=EH*V}zqYDx<*QeKM$DQxte<&2H!GD*=hAvkARA4})S
z_m(EOL)N_VI|g!o9YxKHyXST-&0X(b9X&U=;>rYt^vF$217d?Oc+H7oM2MJip2wf=
zX6_{yOmaCeB>2snvb9Z)^xgtu)W3g8`!ZXR=FZ+s`rY2h=!$EiN17b|i@rH<u;ZK~
zzdQm06bUZe;gz)SGhEO3Sc$M8Ja8O3SGV*`L>Ckv?}EZm6cS>b{R4eaSYVtsAfuXZ
zcS7MTBN=C!%BhF|IcFdzC_Hv8U+1X+HSJx<ud~#$*;=Z)h4b6QxRY23=#GC;P#>vn
zICjRJ)I+}B?&c1dwsxFvR;9Wf;EWVs?oZ7WDZrT-MYRc;hTh2UNC<H;4TgmUm%be_
zjlzV&r7sgM3zg8x=oasASEZR1*t%;jlLjaAZu*iOO^?-)%?;l`dEa$CT*=$;CWDT#
z^Z9Nk7pR%@mh3rOOYI7n5Bz_9XXHoH`#YG-34CO7*~=x@OewR-<t2ZX>J$enF-Z7E
z*6O=#>M4l1`VvZBqBsRTu}!%w8oTt@LxB*R)U7Mk{7t;X`nuO}sLad!WC3KSna6@(
z>fiAm`t3V-=I2P+bh%h22|#E&DzZ9=_rDO9(kcL+oy)g-10n?Tr&fPn(soBfMWgq9
z8(ONPwN>X;K#_^U#=x;ZP9faaJ3)ulvdeJCq^Wzcuq<LJ&EvIPcW>*Ov3!?}q{N3+
zDnUyogdTbNfKsL4k}2s{q%H2fh7*Azk@3ehKww(cNEGzEI6?|$<|AZR;Ec;6Etp|x
zq+)3*f2wj^78c8Mu10^S3i(#?bafeYW!O#+g}42a0U`2pv#yJ*JuL_jG<)=Y?e!xB
zX3oS*3gXuc(TWyy%s;_rRW2Tvr;8!vfNpi&t#l_1ReO5f4nhBz*HOodYgf`R>ee$B
zFh|Z<7Rv5fn>k?53+mzgq_K9;vL3x@-viJWJgNuieGlkIEUbSYXkwIVPOJtUSE<|6
zpuzudh<cch8O*)wuN&uc%IjIGM!$Qy6a;fsIkg0{ugiYKT$gR~?pK!CJl5G>J6q>)
zTdU?k5?|E~MSCerVJ#fu@93r<ks}SQGMTLN)N&e5PkW!z!-lqtEXylwptP!|wIi-N
zNAX*gp5&6i%zuCFp;OL8B_Fz7&Vb^vDDm}eh9ZQ(0BFDcI{h}0mOoEH)r1P9vugU+
z6q;#gL;nZ}7>$NVFsZcF9w(eW{o}{?KYaTO?R%yl|MXq?goGWy+38pONsyTfO<UBT
z>9?g!j>b%W{~|R`n!tv!eh-@lmWFTB35B8A35(R$pT2+9(FP5LXa<J^K0_VL1Wi-s
z!BZ@)dA^@*W9F((5o9WC_+Oq%8tZ@3_&m!0^q=;3?n<NnHl4mtr!PsXO6#z1B%mc#
z*hE^yC9MKOFuo8|HJh7gr%IdkBt$6qlQXTfw(^!nL9p3t28FMD0s1{`nytv3TN=4@
z6Vb{Pa#Mf)JGQUrVUU$t>cJaR&X*RLgK&@wNxM<Q_4~J9Y1|JQ^||>Q={$gtjJ9sm
zgD}7h3QsGX@hZCrXH)mwyA8ZPY{&%2<19HewX16&K+}Dh8g`;w4^^-)hm`?D(Z~bJ
z7$at*1gZVSm!gD~t!b74Ep2_}V)o%`9l337)_{NXB0!9^;bp8tY~Pe|h6ywy70l(;
zjCZ#0+Xz8>bvfgOKkbZ?j5`=dg{U61QgDXLy3%dth?;BTch0?Bk6AVL9kFQamvK_;
z(~Z*a+)(?o*qK^(Od=ft|G7#0rWV0pUlPyFT@c%Va+R*v@RFR(=1hnl0N5MNhKOYB
z8GnBv#TJvKja%JR#V12G1<kC&Og-^2*Yt}Csw}gml@{WP;|8q>nM-yFYGS|D%UA5?
zo4{=5+EOnqZ5iIXwxeu6aCgJP6sQ6y8|A}tL9g8eh{o0{92w)wX(PEECahv$XJXVH
z)XMISHyAA`HLy5G_Dcc3hSsHkUxRlH_<w(&+HQP$FqTl6-bP~{f7yf_Bm%gDI7}Ol
z^eE#e9A*gC8Q}_4F@kZU8<<cUIv7@3I~Z067a*1C*OtP;{}2lE+H?usYJc)~8~j3o
zQ4-Avw3b`{Kj9a_X#?tG;1~BdmE=+Qm3n97@s#%2nC~Wc*9nh8vdY5J3a!H41j&Dt
zQtyIf{sOfEc@UC;C7X(>m|F`@MjS0N8u?ham7%pp>`X_c@EQ7cMOl7pDYDRPGexAE
z#>TB}Y3;a?ImgMGoy$f%(!In&NyyHjrOT3$0v}K3&k&&2rHpBlMx7Og%g)!7FWVch
zq4z=&bO{}RrqHKtd;f9EULGxIh3S6{iXfp=qlB(p)Xw%ar8Ps-@GHX6#u86SURN%Z
z$5FTXp2H&rHm4q&vY@GWF&0M3?7H4EuI1f`BoStp<a3E9`%&Siv#3)mr(WqvF6$5S
zH@aE0W?JD5*4w^@?g5Ej0RHgUA;a#yfDz`Ex>5OdXj-#g`<6W_YHL?PWxIbBsRZ*2
zctv~Ho%wp~NBb}{YFB2lJS$IGlfDe64;RKud2U{OUMff;aR6e*$v$LkLjMY6Ovz;l
zGCt%w-UATkcFMM?b>sCG0D5TEf9Um=(!Yt~v1PN=dFZtm@V8T}D+ON{K|x%VO>9%=
z;_$%7^?ZI18V|lUBD8wfYa@T%T)O<ajBh%m68>;_k>h3>m0o&@JZ;Ht63?RH+h2GI
z2WVGMLAFJTDkqk2Yh~?)7h~cQpN&jq?#TOi_!9euHP;4-c(|~1;B74vvR=KoHmS9}
z7IIB0a!c<diCMT${Z>w+L<#vmp~hcJr%~yNsfH~^Tt021^rRo-E0KRe`8A!{ntXkv
z6wcKmLd${fco*&S1^M`3l1RrV=(N`zmSU?v%o@c8CJ3i?&OTC)1E(9EmX}O#G!v$*
z`CBnZQsEK2%XwG7;j^L|-#WVF(3Rx)+R<fva`fo3(y70$^YLi2W0`~jLnQ8HC{&=u
z^}1Z*rYCZy;dbVH?IB7#MQR-pWtlY=JNW{12zc$%MC4$ls_n$BT674FOx9~yxuDq8
zb=rw+V{B;8FPCp>l(J+BNy-jEuYY}Q*S}{VpaBE_2EznqvzNhtC=&uXHJ2fp7ZjJV
z{{ttN5H21Bmt2@DyMMq2Dj)v-_SX~7g81~R-6bg)s~GZHq#zNCMp_hN*B*b71Fh40
z{2ND54(U{Z289s1^f38r=qXa0Hj$P!O&G12QKKgOgVgsKOO>A9N*QdU45jy4pyStb
z*Z(nNN6aesRT-bQg{I4+WZoUj@h&_tEezq^2^p|nl$zv{DSu0Q21$uvFOf={>C(wo
zvMvhMs`3pP4uzp~?yDzKQH5uh#}Q>5Nyas;l-e<S3EO*+XOSeVv)q8Az57w^iB#JX
z)wd8+u+6(mQwN^YRAAnO*${p{led?*)9)v0kJ<wBS`XiTp27zufk`LOVK9sS;*Wm}
z-%npC*4FPtjem243hJyKI4V;OMoNaWxacj|P?+!f9OrW)SYem?-yiP-Y5Dqjq*)3E
z!H#;Yf_Gk^LMx6wPg7`pzJ6S#nDY~6pd0o&gq|T%ijlapi7<sgN!xz<92BTQZ3Dqf
z1-eI6J;Ei%iWqw6tf2|}xP}3D6^f_nQznUe-OqlHd4JKK5^FuMoRE<;lC?2mf^%fq
zkB%(&ddxGrIMzJtqi*L$*ukuI%nMB^D1L)7>m&o~C-eqd!tx+8ttvUyx<I*Iv<cyv
zw{Ak4nx#!h#e@_TIxKO9i{fU=Tb(EH5?Z4B&QwRYEt*Bo*9GYs>SfR^Oz4@+A|ATx
z@E^|@yMLCiA%O9G&Yeo|ObVE2!{%kh#dV+$h2|Zz3??rplCHi3XCZ!&15_HjmakK<
zq#aITKL#{%K4ghY?(2k3>Iw+93-+INxyPMjIAWZeP!5MYpWuoq4HPTLc^l@AS#Bo$
z9?%2M?=UMGJ(N&AN-xO&IZMz?Vu)Ft7z_=#rGJ>$brO7H#0~AN+~hmzTGq&7F7{Xp
zv$g71w5KN2oDWg~G*cL2lAbHRB6Ey(Q@EVY$wgWn&PC<uh$&Q$>Cbb*6O$=fgVMZK
z#Mv~4^))1|VLhQd2EI-7s7!^7Ef}UkjX=jR7n&Yw681MCuND=b8Aps^v!f}?OcHV)
z!++|?M;Q-d5}M?AIla?h7--wA%-J?OYyUISJmoF3W;xB%uFQyQDAfL%bkBGy>C}HM
z-P<tA#hD~IVA7_D3^t{EAemK`O0UQ%MMs6gNKxS7CCN5%=t|Oe1+U2mj(m<<@?{>G
zZq!DeG`Z_~mCWw3Nj<yKXH(DA|Eb#A#DBJ&W{&5qA&sAWLFS^Suf+#Lo`(@aB|Zb>
z-keb?i{Z#5XT;Pjruj#cMrCBXCU!Z_xb8pg%Xm@{Pli~E__buxI|`kMxqfJa?#T#>
z4Gtt*R$*8G6O8j=b4dUwH#>xwlLM-#0|~LSOv#DrN#1x}kf)hfa>DJ)LO7O!w|_ZJ
zUoRg4mI=@|)U*(qS+OP`nob?pbV;>tr)eP_Q_ikxdVjK1-G~-#*t%gsTT7lwlG0=D
zHe%G{e!tMu3(u3}9FqWF1c4Jdr3G!vC8-o&(T^f>YF0KL>SCGS<^0mPMOxK%`VsX@
zsyBov<xm<Yv}acP4RWhJ3??$h(SII+ns#Bio)TP;@wtW)kAaR?vy?p5RN}D-i**V)
zGL2P+eSjcvea46i(-<bKtd@^><jd~eF@kMrb;-M}bZow3iO#dE$8=;<Qd?%UoN#hr
zbt!YU$H*5p)5_c<&^zOZvQ`pq`0=%Jlcq)3I`VeH&C|RsO3oyxTiIfdZGUlJIol!`
z`+zO>m~LT<7dhEwi#^6QTO`o?BAsfBgqs`K;wDX3TO`~RP7Xv7#eT!qYxvY}*pF4m
zN%XK~)Jd$u9cCJ1X{p31c3WJMu}Ehy@&1Sk3Bi21LPCJAtdK~nnTB6kAt4m3R7fbj
zA6p@zwCRf#5(2$eAvv)7EPst++T5rJ&?DPY1K78y<T^k<99i}D_%4F*`9Mt}xi^AP
zU(8{9;2n#e01@_Vh)NGuM{)v7IFT3gS7HYTtW!&drv3r75rX;8sEzEf^pf}xYL@~$
zE0`97LMz@;ypxig_)tm~Oxin!o@thnwF=S8+hpHT7|a;G3;UM84}S>wd5F-6KzdBy
z{yT(b?1aE3nv_(4-V+%*efszN_rHAm4GnB~|4-<_2XwLMTix(8y=sNv7ToY7zZ=ZW
zwGKw<Ia>0d9n;<;K<QW;z6~dZ;RZ<Un${no)idQ3>SF91H(PB0ECV!6g@%B!wwCdJ
zw2g(TafrxMB_q5AEPuIgc*XvuS9lFi`(8k?*FT5T_u=%V%F|2y7w0KAbK(T1ARISp
zL|U3K08!d1DB|!IC&WT)%V=#B=;KWYC`G8af(i=Bqg7^0OT%z!LR<wyX)3?8eL(<(
z*Cdjh3!qZNmJBY|(JCQ}HHmio``gdl?;U%6ZhjM<I}j4(VSk1@i6~Ja0ax5;UnrPm
z@B&=|wZ~jYQ6;V&T~*%KDkINLGJ+II_%bvkCSLDGX1{C}5PETr(agGlsQ(xRBdljm
zR=F$`?=o~auV|fa16W#UQbyfpSKSp%>zAfX6k4WbB4u<_CNdYwL?cEmE}yf++)|+d
z_VjNlya|hwlYax{yz(dWQ+UYd9TR1#OnbQ`C?JN`1pk5E$Hfw&a}4-oq_A-5@$gT|
zt%?qE@jAUlsf$+fsQ}nWgBQK(iKcZ^b{k9I-4}ED7(%hZWU#e<oU6oQtO!Fm(_}AK
zqj8PUiXwwPU!9r5<7}Cm!sB#IQ}|y)50+QkyJop9&VMMZV%EsHin&mNVRTR@-Nn4G
zR_oQ)x_zvOtM%i~9b{uFxn=Q`!lqLJ0_E%cctYXn0i-C$0^$(_pN~&uq<2ymn~mx>
z`Iv>~<P`Lxm?C5<i1YO-NWomNMcXb^^rdfpKc0cE2R<N)bp3pDB4*?H<_ukyBH!bs
z=UO_AYk${nq(5cho=2SN^mMB4_o9l-nNj%JRTq``_M8CUerSKLqAvj5quN-iY7g{V
zYNg=%L+>W#6*0fy&DNEsXMkEMR(+hiD+Z!FFn{gYmCaWbHlFqgr>A&RI{6isIlx$E
zngB{ZsD8oduOZki6A%_Hwq8k?yZT@ku0GiHtA7tIMoBI`nCU6791hN2c5sKB!WEYu
z_8(rs<Q5YtdUnN$P)R*{as@ED<pOSe-GN|Tx$Z!q_g{CQ%pAY&KrpUccL1qS(&7WI
zJM@@uaNU6*T)FN*pck$?gjfu?Y}vQ!xCn}sbl8uHAam7YTvINH*z5MC3WE7^se%As
zSy-w74h)?Bg{2BY!Ahxu()+Qc3QF7M%NzuHwNz0yRQrjse~Z|2Cd91@7(KENy~IH1
zzRewm9_h*z2Ey!q6=2dSnj`zS+obOi|6kl+puZTZ{0~ChY4(@Fe<%}|J1i>=mpq9^
z8JA$M0Sf{&Fqbhl0w{k`ZyUK0e$THU(3figbIIXw$SHyVxl41sLkk3P3bZx+pmUVO
zS4B!C`I7th{f4Wx<;b#SM|PTrINT+d91dr``DUnGIlJI;&Fo?lYltIijp8V)HL6P?
zS|ho{2@Oe(i*-r0*Y;1HM7B4<mCBFMh)%10CE3N;j}{jx7HfYMwA$oCShvkyf{UE2
zx8Pl-HcrtJlds}iskV5<yHXWP6v$lL8YMdl!PYMjKlZagek%5Dsg5|1wG?}5+pNKn
zh>gPpR6|>X94Xl>SSW>Du|~z_quQp_C8Fwzq7R?isO(Y-rUFGLGrm+TlX7meF<Q0t
z;$0z=)4r|QBvF6#YR!-EMRXjaV&Z}ZI%|8QHiuTnl%mb6j+4o&ksNa|O=xzU$+W2A
zSfsrxb~IvN;E<HmR)Ao^dNtszqPAHuX~K4@y-LzPfUInPG~!)u+ksqJY&#k-L`rL!
z)<WyGnG)2b(LO8D)fk+FXlub7BxD<gWu)y#Q*uI0I;ej`nlUMn6UG6eh2BCo4u>E$
z+rB9~(Mnh%;Sge*844+}^(I0KO08I|P->GDv`R0SE1a+a#a2q@)+=b?Jn#;alzszj
zCfk<Cru;UsxJpeI99-?S!Z;dc>xJBvYZIk7oTtsq>R{m3i#@{(ZDuT4`@>+&*B=HK
z2VL4Gu>*hT&3bX@P=l!oE~d@6l&jNbTnb0j-nif~7-20;pt-GKekQ?o30G3d@zLYQ
zuKQs&navMBj84Wb!M|U+?!{=i9M7kY`p3U~TK;&r94*H#_Aj2#rpw2VyXTlb2G}lM
zcF%EWF++&<07r?$seb?{#0cBN!;>f7i}~#2aJ+v!>Rvp1-n|-sS{^@n0y(-@zke8G
z<+o!9G+mCT%Y|V@Zd1G*FJ@=+lkuWQ7JK#g@$1vk_p?t&0gnPhFG|PwevU6)#=-T=
z+04(58fMv7|NPg#VFk`SY<f1C9A7yghT;v1U{+pXF*lElBwk?G7bEwk)-~mQrq(ru
z-&TLE!W(K8+Fh;P)9G~9PqbS8(NV*7+eX9nRika}_*{(knd`ov&0mk_ZG~{${k8k4
z`vDKw+A$<LSsq1JA7I1)aWC*f3dutu2IU?x;rLN^c=pTECj8s!^grFx^XX5Pr?Y8y
z*!|<>PxkM-cgy96#h<(G?dkH}*)RTN_P&4n`D}7JdUpEe4YWP~IGVq<w*PhVN9<uu
zVY^Bi#IFL67uc0aoN!oC#FNnNCH?sEqyN=Jd~IUQ=5Id}v->ST6`s1Y#dzLLPT!xx
zfH2j<zgxb)>n1dKgqV^%0hoxhxA0iWSH6c0%*T`QXfZypecZGQ^EOFq-e%gCtGs{r
z(x~1iN!26H>@YH7%ekcc@ZN0>`$eEw;->fN4&GZ!x(tdXYzM`ASHW#}UCYL<Yu<g=
z7noJ`0<#LQpsM7dAw#1KjXE?2fU5(*)i^Y0XrwD3?~)T;?U1p5=pB;Cg9I=q584O#
zi#~&Ca858&Iqsf1%zoJYID0kgp1FVTPTq|^;8=WAc+ThUf7_+G8WyC4rNNE}9wNr+
zAxe&QV`+FKOB4d{4xpF;gd=>p5P%_|>6#P@q0)9KAR$F5SUgbx;R=cNr-{g(!6Ye?
zPe$SiGAM7zKv7KrB{BX+Dj?B&P(;S!VEBU&;CMb+hOk8lWMzR%1Xr&ta2S8ULfM-p
zNZAKVj|NR3Dofdvfs$G>BB~{Wc7g=#PZ5I*=Ze@OLj*G~pb!GCE=9oh^B(ljyCo7k
zj{^o9$==dWO2y|*qg;$zi1ZZvUe?COkr747AQo60;kh6wh=qO459XLTfaW57C!m69
zUwulJLxDM@+sR<zr^+s&{nCG^ftd3SqX<OFQosGgv`akvJ_4MEuo2+2-XXwQUEw^y
z)gi#?5TP^;4c<qn>g`eh*;>smi??!*Wc47!4+N8pD1`tH^uxhBLYfc&mL+(;0wme6
z2?&AFDGGAr%=__AK~_cZN3r@-1xO&%rplhm1EC}XB?EG<K?vBP5Ac7}twLM}ljc;a
zHzdt#1FrGSD@LUVa$dfbd|Bg$YihWYYsNL#)PZv9Ksj|qIjtVIYVgp=U+%2tCjpTQ
znX0!kRSMY`wj)CVsQN%b#A0_ah!~{6|G5?Mu9o6nAOkGP2fYviTk}@M+^UNE@Yex6
z%})hZYIP40a^Se|*G+$-AK|E4N+Ap&5rLWP<vgH@P{6jy$7F?Q^;iOIJKF!7Qb1P~
zdgGEoFBx=<_n?b}U@d%Jg#Z%y2=BEjM5PTGQc^&Bp=7J)QWD8Ja9IWXW@d4DRF6v>
zz^4L#E2RL9vH?ty?*vR2NwJmhSD;h&$PnOZ(e7hbMlKa*$MJuGj2!8qM|OBF1;yx9
zbh{LVfbcm$7a+aJsz~D{3yb;s(6~O}+_(TPK#v_pB87w9e=}CN3b*T4NaZe8n9_w6
z(yF#%mp-e;5EW^NinNEQ2q5!po=;x;VvAkeFtXfym0|3bTJBJFg<m8~t1FTy#QUfa
zgSizuK+SJod3}F9J6ZIl*U4!5wv}GT)9$y}y&p}_Mw9N%=)c_`y!Qu-50lg7;$Su%
zAB^U2&)(xG4o;`b*#QRreqeRmu^rEO=K{S4Z_cJ|;e(UWWU@GjSCyJ?OGmI%n1iz_
zKMp=+<k4^?#6IGO;v&cx;N*h{Ov&4o+!h(`hlw!Ax!r$o_9A4fxLLT`qK0dDh+zrU
zT>{^wQ3GwaDn{91q!pj6CVKP8m|)WM^;mxQ9F82}rmF|GhjJHyOPc_f?gVhmYXCQd
zSssus4Y-!_(2y?&IJW8`P8a|u_6rD0Rbk{17A!MuD_s4_lhnNx&A=r<CM_?HzP5XD
zH50k*;=6y{9U>2B%089;>JVt{82nzP{vTZ(BFOlR6nKy{L(%{`cV|hz;?-eF!Gkmo
zSBGHdu(0^`FAwd~;ChypHc=}BiDSw;j5NxSDt|LojJ#RJ;5$_@@*0Zq09AYrXM4yR
zatPq-?MGV39>jiZguJl*n#DkDGc-#Hlyf2lRwI9Lh+!{D0}v9d0D)9A0ZYgmltl3<
z)rS!w6|8B$UCYQzD`d%b`C#_|!1Th`-8$$cV!%6P#C%0Pk0J<3-eF{16%g*h9q!||
zMc`2z?vUjU*XZ`6tLsr_ZNT-vExKxCw)-Z?aPvr5AA;-o$o~Oc@~49eWo~41baG{3
zZ3>rt92W-!H90mom$s}c{(k_n@=t&N`qu}Eg!tvn^=nk{!3w~YsF2p^0Fni496+-$
zi;M6Lm^fE{=fBgl9@cBqT5K$%9AII6HBy&0SJt$}5$~MvcH~HWGVT*XNmMjxF8A&g
z%3Ui2uy3ii{e(|%&G`JQ^h=*%7M0c&H)lZndo_o@jzO5H8UDQnW`6{mAg@x-C@VB@
zkWr85lvf;Rm3dVO=4+vC#xT<ru~}QE@x2JKpYSQcb2Jlj!H!f&GDRFA4!PSZPSymN
zfm5VF@eMA-C1_FRrppX1&P6hwluv&@BUL2wQx!Oc_ojuKRydJSYlQFpu%b8Q@H4`5
zPd-6Wvc<Phcbq?s!+&+*o=;GK)2M4Qyc>>9@^+(snR>qTD;Q^3Z)JWP2tw|sa3yA(
zF+)qb;-%;b5q(k>>FYE5_4@VUj|l;+j=~1tef#;49$2G<>2;^20V(XeKmXBvfB3?|
z)>=8smh`~dMh7SfVqiP3wUF>l^o>OuqP$39q5JUPpFcX*@_*~+alslmrFsrgDf%>e
zau{Vi;G_5`(6sNK-&)Z}4oJ)SX`KNTFj=fTfVP9J<IDtY+sykeh~5k3^#tMqCSXu}
zm=Mqb6V2=UvFVLaA#PwZ_71|DsSwFy42(`hJE5dihbal0Rj3#y*tioAK_yE+k(<}!
z5LFObxhcib`F}CTA)92VCUPWSW_hWt7T{$~cf?^?(qK%Sv>VhPdq)vmoXXPs=LNF~
zpPl<h2QNYh7wsS2I{}}a`$zXt%;?|)C`SXHg1vI<#9n|k*vUL2wJ_3Lpv8^!!ntrE
zCMH=rr8kv;OhaAzPa?#o<=5ofN4aWCA;OW-)wXoQLw_Bkh6O8aqX4_cAZJgNEOqj5
zk~ih0`gF{0hh9YO`0BUpc7$lI@A)X?_~;XW69cZMPm0)#UflCB<3zqf3vJ0bSqbr`
zo6Ic_coxWo77}wU4mk}BlW<z;5AHw<c|qw9=DzKH)OjG8*vB6-_ow+Yn0W^2jN@R1
zbyG?6Cx6VmPQEz@cwB_ER>J!D0U`upg|L0tDMH$M5vQ1)^%-`F(BLzi7a{T)_TfxE
z!!<aQ&%khYt3Jbdl*z&~l#jczgY8~L<bAf=C8M~LEe0Y`7u-c`F$VmFey=E+WSdNX
z0bfjtV{N!lRhSg)I=@rE$YwhbG!?={Hk(0PntvN@HlO>r&1R4;u-Ow#F0<JT@`lak
zy>BP%cCp0!ctqCrJdY@t=JPHDY8WW9-ba<Lb|UDBd@LCn9YqZN6*yy1&LOGO=6Wah
zFwE!DLy_8vD2Vh1#1mTe1AiQz)CP=(+$8$hGb@j^oCj$D(@*rHI4hio>_B5l^#M^_
z3xBz~Z@?R<+u2BtG((HQW+aEZza6y@$vNwUv3k&_w{upZg|-pNvED$d!&YJ6ym&~=
zdLOYyUWq8fgnB2eQGgcZr<h!0kuZ|u6n|N_2yO49nwhM^E*@cAb)H3od0~*$=?IM@
za^m!v%n*Yp&l?X0@q#E4Be|U-<gI|&u747tpoP<VgYrC@yFi98lH)j^xE;WQ-7Oi?
zj*OZ^h)v6{GwwY&BBnRP8htW6dQj;_m5k;nDHWgy^%?$3*PWHLU@LGL|8f<yG+t?$
zZv5374jllcjErnk!!T6g9c#X49lujP2reVyjcvY>9V};cPhI*=Z;jt~V~p#;+<%1(
z?Y?YJiNFwzT5Yeglr6Na_^IuE#!|9*H!Nk_`;4Vz)Vsh^GEC(8c#S~}*V$dhzYV+F
z_C9NOcX6ET?kvM6TLC&-@2=VkEwrtz(DpvF752?bTcPcJW-Bo2U9c56CM38RD46T6
z#z+FBg6{r<gF=X;gtK8IDBMbkv46f7UdJY}OIIf7?XcuB$V<5?C6R@7Hj`zPsFBBM
zghV>63Y^}_xRrz^O4uvU^5N0k1aF706Wax-^|P$$YO}HR6%XCmo>Qt`53P=lU)bsv
z97A-D+s?%&(v&PkZpU~jOV8T?pAJ|6P&$R})F9PO%bZkO@K4p4{%7VBFn<$;N?O{2
zR2418psjFzIEXlP8OY{4lNW^t6&&Kr>HHuLap~Z*Tne(@O}5hk?iwe8Gc|W9y*cwF
zarFiH;#msr0;kb#Fv#1S&WWR##yCp(?%jAh9j~z!FS#BVud&{l4`-#8F2(pt*3(|*
zR&uYTVlFrH_3I?n(|~{l-+!l6&p$fo__@oy0RT7@uKRD7nqd%h_F1fqRVpTg9)9`v
zkMDo^_8SOU_v1flgzP}kIR%c$&+=1ABYXsob-z=C7t-{-?~=F}Z}U2$)zF2;jxFE1
z2Me}b8Q0g%zo+I1TG9ghs_{8XHRP@|OWj6GNl=Gon)LU&k=8cOl7Dd<mXdxYC^$C#
zgz+>g{mi&lzo{#X`e*m>y?gjF&@~$Pm-j5VDl=qN;5`MSYSK~@ME<Zhf+LE=02^eX
zwMKMk)G5t5X*ZqRaAlLmz}ZH+V@riFH32MwKuz?`+ZPDHa4jT3msYr_x@d(1CkCWb
zAz2G)$G^Y*#Bo1h)PLvdyTEx&DZ<wI?$N{u!W$E;oTI!@GlAWyH+e$Om`iWTaU9i_
zeMWDn%sjW*7*JTkm#$i3!S!)O_RDSpDe!TDliLL1dZZ$Z0#8w!fiMdN9E1aCLK#QG
z+@F*gPAXECzK-a?W?J1-hQ1o|`tdVl=clJ~lIO25U|X)nkbm_Yb2SGGRTw@cwg--c
zUawTh${%{OQA3t~Ud<p4_t9Wx7wHY#UaIW<jG=q3dfx`DUWO_Itv-!rFjp@a^Qk&D
z3d7uPG}ifPPo{H~8W2Q-;Zg6TM%I(wtWzU%?T*G@R)Lor>m3S4xfbVM$xj*IoWrpI
z9|h4{kyI8r@PExo+0&yABv($uFCUvq3VL3+R`(_;#1=I|&Zl=1?iAQ#ZO0O%E2>Cv
zCaikO^Se?pmg(%vIi;1k5^ajh{C=q^b;mlAag1qqE=lkl^{{ehRpOi%T2FYlC2`&e
z2~BJiIh19^v?q!bjy##=?yNjHEmYq(bx0(&x}{!`;D1z?XK5XGkS^8AMqPW<RIJ^r
zm1PLSZmC0LSvZ__iH?zaC3%ak;$~gtRY7akF+8bd0-#K*sS0eWrV5MIR5iqEO$`3J
z7RR~P@45*x;bibb>n47W>t3&K8@O#H1m)%f>)TQ=8o;cwndFVMev(qhY^C;o(C754
z@V(WeRDU;y{s;BcH4jK<_?k02Dgad!J-GO6-SZTH&nw=EQ1FO-R4k{{UanU*MhR4!
z@I&57mwA+;Mt8>Q7%A&=&#VYyJ*Dh|7gEu0v>~%v0$gPs#!Qo%u3=sBtQOi%AZ+Nq
zbcE8YFqwCKKdi2!_o4?^d<J6erwrL4_jTV$Yk$!j&7oC3%BnvGT({s5&#k4ebb+e0
zeOi-UUsJC|jD|x^^5k47RmVZsKARmYERz&v&m{L~{L*Z%m#nDqy7q!RjjpZ1ShI}Y
zTNz{PF$11zq(fPcscZnQrCFW+RM|p@npL+x!oU~5YzHwMAYMGT_nlawiW&+lSxEw^
zihsMbq6WuW)$RK6x-Pc%i_3Vbl*Ovq<TqJ@AsNjoxP`a8Vm7ZBC5+vv&c$tGgftho
zjWJk>+y2$H-SDX8u{vv^2($+X0i5ZXN_ibZKFlQj+gJO-RrvSHJik^Rb9hI^KK7hY
zytKyl2yo!8U83a9cL1amdL8vVW}{b3*ni!0i=ECr_F$*IU8u9hCo`4u_+%FM8Y=Bt
z{57FQ({I(6s^MhsVBEI^qsJXgmZq-v?uSc>MXy^r@VX~-AM38ufp+<r1-v(`fKA80
zlL_*xD{3N)?r2e8Rr&>zeZ{O*sTgxb``QL!szRyzs1AnFwm^`B8mj_jR1o^53V*w@
zv$`(Q!HQm5KD&J080AHTUC%}p6&Q#PRY8AdMQB|<RF)Jvs8O8Vpft%nN9E_2%+a9M
z?LKwAtg31CbQICP1fk#gs)iwy)n@J`A!V)Go1%W5!mKRRrV3xd@$Q!;Xz-V<?)l3S
zz+^;V?b&xF<eYJN<+~CX#Ip0*b$^zRbn^{rqg*M~Eo+zW2BK*2=h-atdS%}gIo3ch
zh)0V3@g)Zgl3stw0qb2y4&(I`DvRUw2%OJqEdz#kKA%WWd80Gp_6}Cj<@E=w4=esA
zkY{o@wu!&Eeyr`|TjcotG3A&}Hr+x-YN-h1SiV&UT+gH1PvW<E@3-ZU<$t3JsoR6K
z*6}*}DjDTVGU>-?$CO+o!-dMR1-SD)GVP$_uaRltY`jIL?R>FTqL-{N`mUGdv(6W6
zdQ6p=a``<v33KePUmaF+alqGVB-E%(E>9>Vti@sg*XyhqiXX0yU|&f;M!xlVr{C$I
zuAYE-SWxhMo{e{~DNxQT#(&8uU)@X2osIPE%}IR@F}lP&{K)SSGy{xjI$-Tgw=}8V
z7Di{L*_}S9z`=kSJ<;Dl!2skrt-qB8J>x$h?U6xVPkUsYH_{%DWL%~66I+!8pUbCX
z&^|n$j?H&LJ{^a6KA(<5>{s*YSnor%TiF%oj8&(Sr3ahBkTI(j!hc*G=CgC77EbjS
z`bsUpoxV~#=(w-c!rAbZ+Rh*3D>3@6`%0|yb-vP+%Lja=>sN<z%>jRiuf*i?ysyLn
zuKP+1#r?h#->-eTRfzw`Z2EQq8`<=2XUs`I`ceosbH7@N)OKz&<<;x(i=}hW)lup-
zocz6~(zKiuw3AOX|2>HvgGT{O3h>e~pHTPv_B7UqGBO=^Wgr&tACNg-gFTuye!{4c
zEv=WaVbVxFG%MeVefT%{-2Vd{kflkN!G9<dml4e;6qo$FD_a85&X-xdD>Z-51wTeg
zt?}dWd%(Z;^r#`=46yjYCjh5MZvaO$;766e>H~hR>HFz1Jxgmdt-Ri+*|N-5L|J{H
znOCB4t~2F@Jxw#uiPkAUQ|^?x(YQ)pPmd}}6BBnh4}AKZ{{H!|50V7&^7&ytxYJgI
z=(aE49)8<!;>Rt|evm@PV7Gt&ZFyLL#{V`cyfYq22yI<(+wVB>r~mx&^)Fxk0Ror(
zH%*nB5z0ag+c)~DQSQq6?N|Ek1Rgvh4<btK%w~cBX>dYa*ap6A4@P<1-3B)LYZ{Hh
z1fvvNtU>7D)5N*~8>&1cihY#!`;`wiuG$uv>BdI-m8j6#^b^<DRq20c`taX%Dz5t1
z_V9Ik_(W2*&N+`$LQvY`U+04{E2)Ao;L$vpWb2S}0?h$Tl}hx|+JjMk-ab{JDviby
z2H_>(x3*h5a~)ylu0u$xQm8|XwQ2l74%1n2q#3Cd7S*VY@F*8FQ6T9S3HR?`zR|j0
zaMeflx6pY6rG>F>dvuo!zAM}XAB<G^N0;rsE69KSeeN#LErX~UlIXd6LY_tRlaven
zl;(}Qx|OwrLL()6WnIxsk@kvrx3Et-(XlK7?Pj|F4QC`PB0Bs1Zm?V0DUx5oF>LS7
zxzw|hn9nq?Hm=H>d>O4=@o=FThco^_v+BS!7AG0RP7SiY4W;jmwk|R)0jr}53l$Ws
znniykSziYZu1rrgxLj*E3xo8_%o0Gu>|d}$%XVT;t$G1*a4ls=@kJg$D;DN6()Z-4
zL?t%qFHYYuNq4k?X9~y|kx~T4KOd)G7zuydt6hNzGKahLpk(oC!^Aa2yZ!#p?d!v*
znJTXe17N6-&S4(RLj_>m#~^BT6jDbXD)N7^e07b}qmw}OVy+WI8c|PG6=aC6j#lxO
zU%A%PR5j)*vp&gSfQkWC)JLmdb(YG@o6g%@tH21>ipdjI!g~j4A2?HFPZgv|(SCHf
zDjh5sWPm?PI<=10`u6jyyUvd*Tt)?cc1veZ*$rv|+{!S?gx*RuOvH>08Yp3Ul=HVR
z!7IoFf3dw{cj!^UVk!<Zm3+Z7R!+@;N)2cEZCr4B=Q`rMnXHOJX@)rqM0?|n2_1x2
zW~79fiDH`bg|b|CEHr@dppWumDmDI~yFq6JBK_`aP&)ze#@!%VZhoCA{bOYTtzqP6
zx<8EawfjRAqVT@A>z>(d4LO*C35O1DBn=juf7;Fzo1p-pKwrP7R9>xMpG>X3?_?7H
z^`CsKVT9i<U|_+VdEOD8yFzw%16D+%=M^1h?&d+s^h%=J@c_&H6{w`edHcIjty%M3
zta^vL`gwOH*Q_Q~2zzzjucky{Z)ryoidy(+_S-ID0S!I9@>~BY4G=!Ut)*JL$7l`l
zcKfV%#DB?3+btT5<)Lx+-GI++;j$J?yL~=r@K&|o0RnF#fOR6WI^Z1;V5E4^u(-?9
z9U>y8!xQBL(*-QQoQHX~UM2C_l~D38#M3*-QyUTc%X>G-Grd*4r5D)KYqGTKF!J*a
zG9PzU`mK!YKypDuwWF5t`Je<Md*+bz9X^TW4u5ZVU*6x!rNV00|Jhm@cl}YVjMQ$>
z%Jqk|0ud8J(=qt{u?JjA+x}#wjGHE%)UtZrJ{Ycv)>#Dzm;M)of$-bv(d-?!GTL4|
zs-u@5j`kB%FRupsaq^CnYXQr~%{v%?@j`=s$KW$j+$Rq0;`*)t<Z%StO#Ltb6^EU9
zd4FcTXGQ(+%Xg%lMi>h-b+BvIQi3|o{t;Nf@<dxRT$)H$Z|%CR&EDhaR_(p}msEGL
z8)?I9S1FU2riA`JzEh(!ZlJ8tvHuSO^`b}rIAuVMX!j=NANQB2goXF4)<%N4zE#J)
zxw|~3Bw^zQM-CIT03T;=ZY&)axAJ%SlPyugo0-$2`ov&Pi6z{mh@+zRG9l*u-j>?4
z>@7Aknbfj1F#Ub`{P2HQFUS^`!G9<dm$57Y6qh!eFe#UCc|8Pw<R%cl_pk6D7(x{k
z4{6lWboWf8+_IWmlv9-5(;l+_|BwQMG1wmWgIOg9x7`?56;&Vg0Oas(ki)Ns)BE9<
zH>C$F9D&8*{p$eyLh>(2ZZtGl0AUmi@85?%o~1;?*NHDJrBZxZes+AP<C6l%J%IS)
zHvq>ctASe-@P*@l9pCsmhv(x{co)8nwBq+R^p<09OqB60^*kpEFS(~Y&EAHdFM-xE
zzC&}2(kT8)o{vxHrS?HO?guv9kAJ@Zc92OCQg{;{!>#@I`hNK5(5eN=3i5vV>-!<R
z&<ccdIA}zn!3;nC9=;xaZ^=%~B!4shkQ{z|8=96c$ueGlvSp+(M3rSgvbpqNj29}r
z8ZDzQQWB3&vxYT~TC-LWs!{&(=3tegmTY5cDN>+<BQ;{pYu1})4s;r+3LWIlKXDjU
z3lC%BV6`w7S`H{H`KI1Z39x=44~?Uumo2#xCXyy1lGM$zqDH~MZMg%i%8J?u&*p8d
zu^}5pJVs1^j&hM2%1@N5YMJXW+nt<H%rVPF4U+f4wTmoegto3HvTBOtM=F70q#QKF
zo<+Tmn%c6imbY6|I$@=0C1{X$6GgBBVKJuY_aOMbhVmU2MiK;mmvRS7OC^lNo|djD
zerW0Q=UBRypogVvieS1~nmh`l)h^}7ElpMmi#?-%MN84%9u>3~pck@jMFkGIhSk{g
z&RIpsMnV>_KwO9zj4Z-2f<!MY$Nw2|q3;78naB%k%v_?#4T^r_bCxE}=Y469Pez99
zCPuvT_(q>XD?X>b`TKL?6`m<QyZpWOGwyce6r*XHjc|DY$sG3#TSYtJq|G{zmn<b*
z#>^~#NAHZ2M|det!nE7qbxC~8<9yHv=lmWYws_WW2cAH}s|}W?VlwK$N^0RPT*IsD
z5VhAr0&eh40f5v9nK^Vqn?7}SjoPA+zNc$!C{88K`F0mcijzIpQrt*Jvx{s;lpRoZ
zA?M<BV%uf3ZE%gjCCaz#IMnOyVHiPZ;};TtU)IxSPIW9UY*0IlkPgfobO(!*<O0L?
zw0KRihsA4(dlqj-)P$ChX2pO<<7mg3T+l#%3jtdR+ymiKE6-_FNt|^~dQy3=Qi&eg
zU(&Nl_fxvutdx&!{k;Mwf2t=!=JbLF^Q@G2l+)SjbDA4vLMg#zT$-r9k$2$JI^nT@
z8IOG3DCDisN_QPQa@5iU&DhO{W6z2F`q=M4urxwx_>Uu4Io1^R#Bfc~VTi~6ZV%VU
zx0UuA5S|murgx56Axgw#+1xji5<(m5&7#WG+k-G}A^8JHQdV)aq-qY!Qk29IrvG8S
zkewp4uf2FKZ-@(P^vPdiN%_H#SPN=@dMMo|0n!i3AfvD>La7cVaiUBhNZVb?9p(pm
zVbJ%SA2r1eSwKT^CktptbY~V27iH>_4N@9efVh$@B!`e&@S60YgBhh27NDmYD~i*o
z_E#s@baC}c*X_;&y%Cm5A$1y^$24Vf2=eKk-M}e@BR>m4FS9<u35_N%hO7vGpA(Vu
z?4tDXIT7}$e~yK=Fw4g8IG53C;n<J&13^pCJQiot3~i`hnX@%t#SMhXG7n0|UkdPc
zA=zF=g<(@`A=v#`HcB{4|9>p6?fC6&vLc41>ry?=77r<GUP5%GyUU8X=Y$AI>o<1h
z%g@DulgVGiMbWspSg;h94-w#hc~*tz<x)GzR4+`9#z%4v1J3wzDg$8A`emAf3%#J@
z>;he@m*7(Ka_@CxY-aqr6_i%OJ7~G(S%ymG#a75;hgExwThU9K#MUbAjy>06?`V^V
z*Hh}Xs(tu8?Q2h*<=T(5zn?{1rFo*TU9Ogzk6$ll2~^>ZdBm;$7(~y183p-pz=(l-
zC?R7zXg;-=8HQB#24hu|=*+(ChHgnhhFDOK=eCWEtMrRpJ6Y>nPAQ9Q^$Xq$>8C_W
z^&dH)X2+-(ocfO(xIzPGT24Aq$ulk4IL~C8F`z$?q9A2T<7kF_y*{T~Y{l<4ym;w5
z`=^T(dgkk<_1uZaE#7~BmD3d!LfkakK#>Dcb~HZ;M4(i|XzP&59!|7pBG?d{O72-m
zEN(Y%MMSre6U^+=i*t@pn?kbwtS~;&r6WRuZr<+;ZTMWOHeH4FbwjVwD|?2=GGV4b
zl|xWTCu(D4H(lw^%rw(2a8bx1bqm3%3Li7RtG8zt>W^vU+El`S`@qMuiQOSI@@~y5
z!$>U&GWw&UYVO)LChD}&t;$+bUirE8xSP>Jp>9vRc_obYKjp9+Xe$)-T+3`IE{_D<
zUgH@fz+g*xYh5!PiEDbRwWA5AF&l-~Zi^;>B-BlNo~$*?<XKV(j3U>n<;xX1uepYd
za*CNW+m|c{7}hm^H|{_*)HV-xl)bD^(4$m5EA7wLW6kx!x~Qyflkn_^8tVIcZFz>`
zd$t5v?e}CW+`dlXyjUt6-X1#Pr6dX^7efZUsvsqeAhHVu^@S7%10w&E5mhto45*s;
z9eAP&Pj{?w87KH^1$n*yWT54OiV;u~TpTLD1>Ge0YV9B(A3MM3o#0%sl~Ql6>Scg8
zdYa>VHO4o0P9gpUz2djFm%)cD6qkWw855W3O9d&H)6grye~204@o&$6ImjXiDWapF
z^ff*`J|BKPjAF6g3M2jU`TK!h7!5)Pg|CfpGAzISWBGFUIHIkKNrcJ&S-Jf7bs0uJ
z70pD6nvu>~+i3<Qs>>|SMWN|c){J>CrA6)xO4uOR2BoT?<MQ28fGI{JQO9<~Xi-iU
zF<DfJLChAle_~>myK(v1qDCzKMieCHUZz$r>wk&OZUouf@MMF~D6B#O>6=$!zS8fI
z`Li}#WJ^h++UEB>pT^L_$WRg}GGQNa$Yn;-)Tb3?8b6ii!n~8mJ13!-U4<xuwvOJa
z`+}Ruu{^S$z@9oX=Op&Ruq_OnGtsn9l16+1dn#bDfBaU3v}Ens>Jd!Jr=+DSd4&dz
zTi+QkjP_OXa)l;Q(2gWg(6efY&V*&66GJYK0kHNG(Jv{GrtXzdUOLNxS3f15igK<~
z>BeT?JsDW6h~C@K?>WG21D%xq_@`e86hX`!L<yw12l>M<{SJBE5xnXJlob}@Exo{i
zkyX~pfBO{=@RpqqMp$=y^IPrZX68q-sJ?{7F?E1Hu3j6Ja7N!yy(7lG>a{<lUK=ot
z2<HD-zjmVES#*Cbp|Zw`NrX$xVxFSp{VJ9$A|7o+OPQxuenDp5l38#}sId0cNJ=vk
z(>rBF@5t{eO?j0nhMBTAT~~F*T<w5|@w&>Je<_&H6_!<AGQ5N*v@k{aqs&$*+9HjI
z{3b<0*+>>T33aloqYsR=U8~AS7PAdyeU|D&A7fed+Th+_Gq#X(RWh-J#U7PnvvVs&
z-8MuT3~)Z2C<^Ajv7u8<7VwQO92BcbtC?1JA?~%JjBAe-C3^N-5#uiK*#=4eLEW!$
ze>~~Ilzg<pv{-qiW|w5?$_*Dj<t4l>TSMe^Gc|SgUMtG8dn=k+fuh8qCfO;e=90#~
zuj)&Jp!A8x0p_T%+)z!@#s*M?q+KfjHjkV%(%aCkpPx*#(QFoXW7K7qX+86K-Kup9
z#C5AS5a-^+*$E`|o|)RDzREpOU*8zUe>OUsLi;8mu8`;ALaIzU^NdZNeO`atDx=o%
zmXTSEY8=~EX32ohtc9wZ#%W#TVy1b6Q5<(jAH~@<kfaQ#`+n=$M&>rOn=RXef$H$e
zMk5ztv>hZ+)<Eawit|eIiz`fSAuz0&!*V2d6K7YXmg&F7(lTzg=Z1g1+M79pe?HNS
zb$(n^1<t4_TrnT)biE`wsCZ2B*vi81lSQW|)=tM!wV-UXYcVsogCK=g)dN`0tJn+3
zqHEkm%F8BAGLOehQJYN`@4z0SK!ld6W)0tlCz_pjY64lp)5=n|p{Gw^7@mDsUb|>B
zTbggeu#3}OxLS0jICQ+6J6hY6e<6*i)7B!hldmuh_th#-6+b6z7x3FA&uwK%zN*Y9
zo+)|7FiLq_zOx;;AhfUkpr$RQS&%#q_{N@yquj7d8VdY0MgaDRDPrUqVY^$)o0Z>O
zCny`xM@eDu6ExQfo6%3TYT0DXN^9Yb8u{Z^BDEl?^|vOAg?iRd-lu(ie>Bg}RYZzh
z7*{Hh_uUg#_fZ(k|J*FAH#&$IYE`v1LGeR785MV5T0BG=iOhti#K_y`Qtnr0v%%M#
zGZFXrys-!EoRCpbfAZG_dH=8^0iaHF40ieNLc>4`aL#%B*65NV>F4n7-(SD{`SagF
zAi#eJLN1tPScqZyo*%6he?j^Xmv7}c2(V7+1w@qEVKI-RHsS;ZtOK8ygHc|9MjKd1
zztJd4H=ukQ;M7$0II%9!3^5Iav5xY3-`YW@YF#4nRLP(x!ct)A5#bq%9_iCQ^HhZT
z=W_V696nZgMvedQ+j%sUi8w(R@ETf?@{sl-LIB0$h!}J1h0&JLe>#YTeW23bs35!q
zLkcOYRc3jlVVH*y7h~Wd{vO9K2w>nfizM5n6*eNA6D~%wN+@E@qFsM~{$Aj|BGiZG
zJL9<nrG>F>p#U#DqC{M&S&sX2R&|%@Je&Q;LyqGd`8oINinqKn`;B&!4M>rMk4r;h
z=Jjr6_Tz2=0ZKUPe;op1vbKWJ-V2L+oCR~C9Z{z15QQ~xv4jA_RZSSuTlP$OMjLuL
zgq+8kmS%IUm+H{i)KIPsG)Xy+O~v`JrV%MAeBhvEx+2On<-yt{D=vI42@P2SgCyj`
z%TNlEb_XcrF+<J6nZ})|tRJCN7X@Bk*WgYj2j62zeSsute|<iJ=IqOcPKBJ`v2grM
zeOWoIs!dEsU`<X^TXYHJ=%Q4m2RM7v^a71m)cJau_dbE!bU<1Q;>tU-U?yn*$l<J`
z*nHU`Ts&=)**0NV8>kd=^a%a(-U!<gzb*M7%<Yv^vy#FX$7y_9p3c*7g3Luua%_bv
z=y{qrv&L=Af1KIsn6fkb*V0cbdB<Ioc<eY`#<T{@*lG%#om*&}MC~1n^@fGEnPcJS
zWT4<t?chgJ^_qn4%;9z=p&0GexK>H|Z<0<M5CP-(WoBy4_I_g4ah2>Vj3CDBRJ(<m
zU3)8za8uy|_9gY~YDMnj3af1oBUiPEbA9PsOI=q>e}7+%o6B6=bva#U%zDLBi_uz6
zU)TV8XrINQtDR<DjF1<$0Pj}qgjV(qQ428cVTm+ixMXZeG5|y45H2(hRc;koeu8<C
zdb2%54l|mOm(E?OJ-zZgZ|*&gaQ`L;LPSNgQPmAy2o{RyqF|#+*L0b$VE?keIo9^v
zK9^%Ge`aJh&h&C)pF=GI8STw1VCtZbH3nYiqv)j78aEPm;RFG*1z|y5PXFb+?Z9uE
z!<N*eqi(&7ZJ&%f`*|zZK9g)WDBpPDjy=>%F0sz-QYJZcpZ6V|F<s*^Q^B{nYttlG
zwJ4Dj2TPE3w%m3{yUQp~YhPX0Zgo0Um~nk>e{=X5*JF=vEu$*8!$Bp@9Ft<U1nGpe
zwUb;g&rDO=+NpN^!E8x!$Fq*HJIEtL;4CLz+^L*eT_`uOs_k5#=~YMWln6?VoU5Vi
z%v}5XLFOCFA4}PekFBg`C5HAOAZ#e=&Y$X`W*KQLxs}>Q`o@0px$kkJJBZ%tS@>5z
ze+tm`VvwrtuC^x)|Jrj<d{qPV#pHtNbC{j2H(PY|Ft@7(loeP34h!8b^4F;UmeRpU
zx`Z|o$LV3Qj#g(-bq2e2mSL{CxX>29nUc%iV7$=CW$U%qb3tQ{reJeDm#SIkf0ygI
zN(YR;;f6Q228<U@iqVpZPW~^E-;K1lf3vVy2d75NO3oV~-U#PmoISgUN*+3Smd8TV
zkKcq-;KbTB+96Kr!6SB{uC*+Sb|as5h`d{rs>+woi!aD6h^R(nDTvKPSMRc>t7mV&
za;Kyfm=jlXIo<qnA_R?V-|}+eh;dtCUT5BPXyZq$gLz+t(c=iX5hJ{Ral;I=f92_V
zPKpVQnEZTQoCQ8^*BxwX|N3a70^&xJ@^QMpl63yGky2Zazd;5HgRfU+rwmj&6$`hN
zfl(M(H`+AtW@}Ogi}kp*COhW{>)VM5oPWqkU=ks$H>)ADi0WTEJo4W$^y=d|DvO}}
z%C}^H%1vlQjpe<+rL!3@j_H_Me=|}O-@ax0RW8NWKz}8$<??5zmV4sRi0ntqO-JEk
zXgvbo58hV@V7l>`FWb|7d-S_CR<a3RZ?Nf-hQ87DZjc)Nj=9`W3HR-jpPiXDr&yF8
z^B>UU0=P+~-;^3VgUOp2Gh_NIplE-W#F+Y1_LhIJ(j&ZKTHGUkB`fYH4)02er$QdS
zQHTEnqRgF_p&CCE0yZ$0!K4=zm$%$2=6}`u?JrM1Ze$jO6w#&I^uwPXo;F`LTr5sY
zL6W-Xw@rFsv=TZ{XQBv7cVB<&UN)aN)=Eoa%J9$1?(1vEMt;hQi4uyD&RIKCjFQln
z>70u~r&ohw%m*ngm@^Qt0n-LTCB>L@9u%-Xm1y;ZXCol|?lGuEss^yya;gMv*?%}t
zvv#CxY+gy2t)OD_K}2e>C52+5v{1kCY4(6cLdi}hHQT$SPN?r;z<Y!!56aPaVYEL%
zjt6DLFot2>R%aPe{!LV3=EI}~mw!DUvyq<`WH+Zm@?>3Cj3HraB$0Gmwrs;gdUnH=
zq?9yf)uOJ+kbocH*pJcm$+Dqf1%IxF*2D73q82_jd{>@SUghcQ(8?zBXe@#6$5Sw3
zr%)qBWfwWqq=ogZ90q4NaAS2h{_APaMyB>rc8Yew_%M}Izf<h7_5pi(z?zePz(UlA
ziKcN!gQoJ`f)?@{NHO2_p+*hirtr1w(DNu<Id|mN84-2w$iH*|{Mhx-K!2bkEa~sA
z9~O$NiD?fh<y}In@t4H5dH3h*%MZ`LVpDame<e^v13^~B(7okH5;BAsyU+RAB&F!I
z5u?=3bjAs7Bmm09I`G_W40&u?8z!Tl(`XbX7^2DKJ_h*&SO`pp${G^JaWvEWJRPj7
zsw4&`$}-YZ!jk8vN6as?(tl&Rrq2ZwvwrV3FWu%-ma28+6G)H);^^?VBvmDh^2kCa
zk}X5z1fsq&MJmz50(Xo~`5i%p3(i1T6oiL$CcR{8mFF@t;kyh0SHVz*>eATw4mmJh
zF;b1x3LCMVwIJ$ERAuQ75+3iL-}1ay%=$R_9q`;JI0R!|cVlA^3V(YUxcVH!hQ;JB
zgBR$$<5M)~ce1aMy--@(VGIdKe(EMm8hE=Infr9KK#-GKXzs8;xMQ3M!o!&bUj?xv
z3obll=pu;wt<t@hO40}=A?@Q?@qE4{#Knu*9OaUK?ENI=EW=cZ7Ch<cu8brc`*W;B
zu~PGyqTQ0p*@tl%Cx4*reW5a4m!(Rnp^dv?Rn@<5iEj7i069Zy#EEJFo~aS%JbGjd
z%L>r}anM+fLCquUk`<2nUb4c`q6RDcqnm@}XmK+M7iIqhVDL^a6;D<p{!)#Y+GSpo
zZJ0*d!w6-~eqg%0&sI<&P0ZG&7}Qh``-?kxBY|S8t}kn}WPj~&S0N<#-}`7?@`Pq|
ztQ!_K5Hcan3E3)$oDQooMJS1Pe;+1hPfNnts@q0B^IUyd0)$+K5YhR<>R#}5nL~j&
z?B{;pn7UUEF1tb=)aq!%r<S5d>E6g#DT?IQfx&B(q9(c&;Z`i67{<0xIK6CgPAEql
zGb@|MBFms$Qhyv;FBL57m!t;N4RW>XmiCP5g{3{i_FYPQV@cSc*0f88u?<|O*<{L+
zJ;&OV@~1QhFrpmCh*E2>mCvE&-PuhEj>#|64O%E4Te`&xPA~i=Og%P|WQG4}7*jRi
zj^v0GCC|DdLx`K6S`$9MgFt=36B^T4?jVpxv4XhBIe$98dti(bj%YN=X<e`8wf~M<
zzp3$8a1n_AUYa#s#2n*#sZO7RsM~qcaM-V=m{oe`32{n7`;NUD?+0jFXjWE*FQpHM
z(o2nfKB*o{u6sYLSL%~5<W1JW{TP69z!@?zHO<Cy#&%Zbq>~XKLbA-sN#ViJH8LkC
zp!b{WnSUR{xMJqSFdxXA81$7gC#{jrOMm6eiBY{Ub7I)OOXh^VC1kjE2Fj>T4<L>x
zix*`Y5p+V}n2+<Cwf+1!2}~yka%&;;MxwipB%Bcv5@XQk&~-uJYBYI40^yMgAb>NB
zyAlZR;6;7I_r7?t(4Hq_g(mxr+w)p_BrRim9)I>3{Ior9ZL~CP&kqB<JwJ@{_I#c^
zZO;z`X6^Z54gbgXJQ>0?-kx6={qMHtjaj8TPY40t*(1C21^S}y{DkC=?z}N4H|M?A
z7d7X}!fC{CWOKeipVOST^Oon%HYTK`Tx>`k%&qsfVF?V7Y_Rt0wtLq@(2Li8aXsm4
znt!|V){|_(H(2XIfg2Ao`g~?g!l{1Bu(S0~Qczfw;DCRB+h03)pLS&?7f*gaZJUZ&
zvq;<2=(>|OB5CTWu-S)?941!0<H#ZJll?~ylMr~skpm&#hFWEl$q7z9ap=3(!<`!|
z(W2TuT{Y9Qp`*?bR75Ny^X<<O7$H;HJAbXXZdo7B5tg|>c$R?lHDcSa#m&zW%v2L9
zoOzCLoX*ZaO9+IY=BqX(k+H~dm5Lbo8}f~~Vnxg_u2>N>%m*rB27RT9Si$#&sIFYe
zF{&3<#0=YasfdwMV3V)eDC8K|IZI$zS2#;x7!RB!Fz5@;5;&&)X9*46i_a1`#(&Gt
z5{~02pCzowP(DlG26y>c0ymBQX9=96^Uo3#kpQ)S*;xW&FY>)LkHI;{)zY2b#i-jh
zaoFQqPd2;U0b)$-@E;TqTT*d8hnDAm1k5q**li(-R(JAl&r!(fU+r=cJYM9OwN4*%
zHfvU9`CDL=PFjbC!^^$}hSZb}b0X)o@DcCfk9PBhRkFS}<$WFretf5o_{f+4BGLT^
zfPpj6m%)cD6qnIg0u+~6<ty(2cDG38E8GYKZ?f`)m%)cD6qn)OJrtMDDjF!4c<(E`
z0k@Ym@GEG4e&gH344AcJ61PN{C1TT#uzeC0e$1ZV4fnzcc*PTVH3L2&&n;8ZCcIw!
z%I_ML1yua4uTk6)FaPr2#78wx*Yna`KLdFFro8fcC8iN)%QLyAHZ=OnUAvQGda_2&
znUul%Bs2RovFt)ZGLd0^Os8kaN~s)e#4}q%Vlu6N5^+qMG*e%S(A71sSNZuw%94~s
zkqEuAxI7C8AE&2*D*x15lgFy20-K;+sx?GwPyrK|x8-rPn@|B%=8q$VPMBrD6*vU)
zO8d@SlDoMSW+Fu4RV6@Iq{z289u&uOLfXVRTthgoXURW$yo=n5PZPHyjW?PvX-%O%
z@G<&-v1MLH538K|h#ooY?LxDrqCfoZ<s&~xvqS|eaV7mOTuHogdE`>Ll6y0%5><IL
zntS?{<rkzsNZVSb?UbY)xEyv~OJ?3Ss^>PXT9<sDMGjGYouwie@EJ|}Xe7`IeZCQW
zBmjdpZnh!;$0hMY^XLSdkrW9gz7`3zM*k>(3kyF<jqk;<1vGwZF}5g<!Cns#r{pz>
zMat8cJ_6azNbS3|#8}~ah{3;xtD!5eG6tio=vVREar{;>W1)v6_&i=G8K5$Dv$ZKk
z7^04g5yYEriYD@lxXlmKD5PRS4!c#*#hN1!9aGK@T-UR^7D_P8)E2p&v@v`uN}5K0
ze*C(qp+W-_v+5M=QSVYpVI0G2fbs(JtYOyL5pwVqN14A!zmI2wm4bR)x|3vf8X<np
zOLuY%Sw3v?UII!8G9gXqvm=E1HRl$y$TmVbV2+v-#UUe!zBqVjjH)pD^&D4XM3{}P
z6t|#>uy8Pr&PFB<n1_@+jTq!J0^#a^c<NcH-56_x6H!k?h2p7<BPl)-l<hcIts5x_
zyX5ugMhuKOSkTo$aEJ&~?P2f8)3waaYdAW?aaFeS#0x4=JXKx;XcvjrTa|;AJuc)_
z^pqq{JD<s^=*do9@Yxb>=YyVlTE7Zhb!lYZXzPM>*RvnxeS>u~+0Oz`NY9~vl{@Ks
z$6cD+WqBcM6d`Mz&QgCA8Ie7@GnkL)1QHI<&*U&3&65PjJc-*rPx9V<n&77-O3GMt
zf$oKgQU_6};|YqI7^s1^r36J=X3)~Q**^J@U&nD%n$<z;LQ_ela6OoTi#%FQERY|e
z!uhgZ?WJM@W}NG1BL#=kF3uT$Om@J#kR)-%l!lk@UI<{yOna#inpvS<v3p87d%R{I
zFC~;Oz^kw@ng%TbJC8RekS}LCrIh<rPFTUT5U<$01I8unWxCT0uUFuujBzK^DPw#d
zujGblJ(`%C#Vfs=1~DgXf8|VcI~HsP0)?CvuZ#l?zz{M;I;08*CJ(!R<_7RmRG=ca
zoh%0`Vs_tT^)KM1zzUxqVJ5Q7B6yG|$R%4d@?!_XTdYEE9%#-p4wDZtKnpd{9QW?|
zMEm-jvv2V|Qwyem&_Vos0^-CbOi6o)ZXqyG80oZtzh2L{PZ;$pjkbf>mrdqzt8F27
zW{cRTw0Bg#sQ>lI8aI-EskazpIv}_EIFM4V7K%>NdV;=`o{jYBTPfmV*}U)g2P_jX
zNXzj5i8qsWLJPHryxH~vRIhnGmSYEEsR-zysHYFalK%FP#{;oN+Q_4J5U4O3c7S=9
z<|Wd9krqFG&f7tJrRCAtaEIX;&Blt@a??ZP>6G2d7!$)sptTu)seLvGUoiK8bUwK$
z3<w|s19c5FjjWxPQt;}r>~lFOPz{@1KZ^7QM=xp*GlkOjfh!A1emo2AXFq?+oRuB#
zS~aK4gLEE0Wlq(3hlQ)vjIp16oiV~Ec0Oq~so9<HbSK$FItW9k&XO8-i1nr}Z*pXH
zBzL94;<^AZXM2TzCCg}}uG?hv#!LO}470(RYu#O@QeaFWHm0?z98+h!H|mX;j@Nbu
z!nt0z^TFdju09e~T3R*m*3z-R8}j-kHIkU`$Z3$$D~(%tJt+Bal94?Y(E;mpC<o|@
zq-10FN;s%a4G;I~SlU&Uk01@stqvh8grt<JIwiNOiumY%Aw#!`jn`7$Y<_*<kU?PS
zF))wMA8w>0iJsN9xjjGwmOt9+y)+2R+(>Z!CRsi7CRzH!WURt^5v;#$&z9WLb9?BG
zyYwDWxPgZIjBhvqHCb>6bw0dFUxg^Mx}zs-K4tN!cvh<oo8MEF^i_y(c$fBtwQdJ7
z#6i8&6(o*-9q$wrW@4_q-uj6>ib7)@0@F|LQ54Fu(i#~8uxd>kR5~6>s2^-Se1JHK
z8p5HWPA^c`9d*T-g>Lav=@tfKg=pwD9o>47ZqvUo$WQG5Z%26Nu6a`qcDE$9cQih6
zd-n<tHoRiJF9fFQGi@MXB{STVcU!4LoNyyv{Oa3(DaBWn<y@#`FE*meswFm`mz3A^
zDRVEat2ysZw(oYgpn?W02Gv~a2fbA1<yCuX0q`&V!y9x+oeHXSuEq|<#lBBy`X<9K
zK~6vP!^`#l3yDr%xt?r(*;pmQ3;AwU?54^6Hmm98uqWFJDR(239);Ag0>en@l*s)%
znKQG0!jIh?vkDM983qNY=`DRirQ3BeTLs_FiR=T+^QPiKqvg!6K<?UcMNn#8^+6SC
z)s}LRRE939mRCViUAIfYve=;<)`l|jT9v}K@3YOdc7oS*Tva_wSvK0NbCfFvS~R|4
zgQM+=q(8YcIbu#Gs~QVi<!sQbJnCo9^E{t_V2r%q>^^R+r8*RdrQX=bwn5kEn5ek-
z78Cr?z3m8(94b`gUl;0Q6UBFxxxp??flPfwa~)0AFj3mM8_%gkYM$4*F2zw;GF2t0
zJuncJ8q~vRc1^aP7v{(&6HoJ#094;dSHYY8n*%g|a}fPs229D!v#_aAjMi2rVTwtA
z)cb~JZcA5kZuv<!cB>5_^P+0uikW`;GehBW{db7?7lEK^x(Z6FJ~XIli{rE6H;PH2
z>Z6+6v1s&KH=i+eH(&fU?@b6&VqG=2!6cfZDH2VuNEW*veME$Q6T&r|Tn(vOB7mL=
zwZ6<qqi#X0<(k__jhbgo<=*0`m%BYOAotJ;$jwYEAhEcDH`fXnnkDT#`;Y11s#ttm
zsgeU*yQciPCj6AO(*5iE?Qc-|eG!+zhb$DAG2sOh0x>X?5g8|!4*Dw%f1zHK4O|$Q
zapC|e7t<}aO{z5c|01L$QuJdIBwJ`~OC&`e9+4Utz6~&ZetSIMK5QWld7P5Vet3Qz
z94!F-I-p5RLw0V&;D+b-;oo-vk2D@MB0vZ<GQCgqOv^2}i6mU2K`YKJw;UbK1V_Wy
zr{8F-{Jz|HjdhEyqMunBf2y=^mi^Ww*Jj}^lJGFAEO}>JOIh={6mJ4)!Mfc13~^@f
zrZ&Rz-}B2GG$1}~llv74nBz#X3JP%sNaSN8uMs@Yto@DV9e=T3j$(5k>ofX>OdH2<
z@{@gfl@Sl~=)-*S1)DE~O(M)<6d!0HMUISvNW@8JpvuLqja8&}e-g1-eX#CfZmqkQ
z9XHj=O)}L?utp*&V2R8#k){$$<eg60p&@XcjR;{((-8npeIn%)%BH`9F#0&AIUFsJ
zmM^KB)QWfVjqPdRmEb#ESrV^+$B!*Yw1LK`A!AQdytfq8@K5AJKf>hITWpr09I{J|
z7)KXp(Fh&ho1Zi(f8KakW|c}{B%EC>(Hm4sQs*E9vqVX{m(0eUHAZ48m~5i08lI2X
zPM<7Bi~7{CR*AIM+*wS!cbc0?GG*+mO`W;3|4l1f+`8Ao6EFUX4o3->v3HyW#oY0}
zK2!~z^|fauu1gLCKe{<;nJAtEpO?nWQfPzBvd>Op5yUrHe=Xi&|Kc~NxoHReAu+<c
z3lA@xZh5Gv-5VYf<wN)1Wx+LGC@RXRvcK-Q1Zd|pV#k!fx&1`PtHGAOYT0;P^VPVs
zi4=6slr}+S2f{SK7!poMPn^LIKd5<Jul(~_^AR$R=<t{iMqgXnR-37mVJ?u))fz%M
zu#HM9nVUzLe@|9l<NJ4<lN>9vU8@<vG>>vNuQ}+Uv@G!sj{K^usuu@ooVB@|meAH~
zdvCl|tx}1s@3Ata+3by4n~|Kwx{tQXf{X5TI3vv(@di)4S(L1;Fup3v;Ku-{ZA*?6
zqBe%&V(X4dM%<M|wBSe^V8WRGy6$A{!9p!R+Tz56e^Y7ELaLTbYdn`c8;yWz_WM)n
zRn2>EzO@9{lAuzy0EI(4N7HJ(y%pi%9=AufTR)Xcv_vXp$6(TLI=ILFiV<n%vf*rF
zFIn}50eN>IYUT<F6`Oo<WNOj!X`*WbjFH9^Qm6&%G;4~L%hTxm$+IAc^8R}Nw#VW3
zxStMOf6!atRe|#uN6h)?Nx#{ei$Sa<rIdWn?cUEzi0?hO!R;<RhWmS;UYCnKgE+&@
zrnxFsKmK>ah{TRW62}}IZsVn`h5GGH5z})rcZT*zizN0-WNFXX{->-w>NUXKFSC+-
z8s~80N2_=uEzeKSx1aqv?Sl$sZe(+Ga%Ev{mwX%-2n04TG&z?sPAL<YHU2BX3hyA7
zHhlmh20FKw!G|mq0X3Jw7AF+9DgP_X1_{fJAG!Vo5Y4uivD7UT0x&k0!4@YImnQ}+
zz<<Xa|9=1DAhWSjcoTkxulnEH`{BPsZ@SWgjPJ->p%r!id^>~+1QN#j&MCo_-~a8t
z9R93@_NqvC!^iipT~+cqW9NXelg1EDuv115xy~3bbf`+$5sw{t%*11sj3MzOUyct(
z29!XCU+Sr#$B)h>&JK0D!bGyl^g%KiI)5o)Lm$sxUbGUeKy8ba_{x?s`gl&(B3q-N
z(rPS=2i-wX6tzxDqSgg$$Ox@{tySsl3qdR6L@MDFFeCOCqGU<IkL<*vB3%V<MN><U
zXv-uOSsLc?$#X$pzXjPzroaGR`G}*NlzblPBPS548I7#)S|QOG;n~9~iGbTBT7M8q
zHuDUxJ0y}oqZ)~3eom|x7Hs2Za*ThyHa5%!hadL6cJ5zit<u5!p_#e7CRgj$Uhp^@
zM9#Z>L6Z^^)g66-!!NISXe6q6j5vNtYDN|2smJJgplsreHUh01ye7oHVNHf_%K!69
zFKg82RwX8=6c^|%#f}m)O5}>aU4M!jp`E%ZuB6-)`}r)}9E19xHa0yEB11aE=lox%
zZd??fW`V7MN=xB@CteXdsh@<gYN1S%1tf>h$zn(_>T-e!4$#ys^wlep(}ocPa_+u3
zV7l{w>E{kuYQe%Tj@aCJ#O978UZ77-YRAQzj5C29cphTHW3Fc43oQ>VWPi>~<8)ce
zF0bK$nVZ;_dCwxVB?Lig2LlOpQRAZRMH7z!MjSdbtN_2xSntw4&)`{t0BbAlr*mN1
zV4sryC^P7wql$T09b+A2#<!Mpqa)!g)SO!>;Q_CZ?)v)%mU#mSN3R6nJ*K%%kKl~p
z6<jH}&NWN&z|$al4g1oT=YLh#dgF?PVdiQUhF*AW8f#zX=DA-%3wy&d*+jxS-9wAX
zcjBlMY}wNy9vO!KXarZ6d|!5sbTx;UmMbQ%$c}trV$WoLddZj=NA6XuF{zy#+C?p6
z>5&EuW65M8F`2%X&!vSnyc8W1{bojPnkC+EI2AmCAE!%>X-Y4|Q-4G)Ja3BYx!bq@
zTJjVM5zu0-jjG6_+Q`%Rnr3$2Z>9AZaKfc29co+|N-0Guy{Vo+t_S^G$mO(Ga>L~2
zr!03uR<n~BU0MhCv9&FW3n6J)<KBIbq^*o-7+<lwEX#<_U{P_uHTK2Dd3=DUVSx^|
z&}pzX^ewy??fk@YoPU;{#+!?@93sYeIl*G2Rkeu1-%o?Me8w`w$joQT#AV}_b1B9>
z%bccjcDyhJZ_EBGodBa|9k`E4GugP*=8|ZyExac=VseuaWaF6U<|6IH!R*wr`tMV(
zyNPbw;%~oE8sh6IkEkDAKZ@w6c^Q@*V0&~Q&z#vWWG3D!3V)^gjZ-+Gq~{?bj}Y0!
za=?<{y<fv}n9D0>w^z;<uOHb(T66i-PBS0o&C-ffyK5(MmBg8Vx|XScgUEj?$<>KS
z-<s!4bFQdb>p3nJcRlP4Z`r2(r=S{n4!!x{Q_xCl$in1D3^8jiai-eg49KA{su{y|
zCQ3g5M_wX!IDg&wC>*C+_u@E}b_0$t_s3C&wF4iG<5cTj9H-K5z|r0v$8h@WQGYOw
zmDZg&R?=?3@ip9NY~EQ``u5yCAP7{pk+|dTA;dTbjYPV*>#e;&-rvuN_a}<(n2&3T
zxIX_H6pr`$m87NoYjrr4t6OjIBd{TZLoEsH!ZAu>34c{<xq)wF22tz1vbrjq<$e7^
z8v*j8Z&xd=cRch`%yk?27b4OKN6OF*IF^zNIQA>J!m-`s-L$rfk1G;@eTFZm85_85
zB}u^rew+QR(jDd?=Y^R6;+p~MW4B@lp~fDB?U7dTo>%%>VijrE<(F^_+#n#|7GihK
z9Ti?1Zhz@Z&bS{o#VapjGrT2$0vaIm(QD(Hz>S?B!tbWb@O(C13EOcO-Fxp;;tp;A
z-U;V)BXI{ekm07oHWC+RMOvYKgHiQf!fTm=P_x1FvJY1lofo4-@s-M`a>#VbuSek6
ztkE(4_CVXGj(CXCE1sT@U5&5CbCgz<kYe4IYk$(5M!_qg?Btp>21?t$KD7`~%Yh%X
z7r&*115uIja;beHk2hYO!niIuR`~bhURa1PE5sKQR=ltfoGAj8{XN_tqin&g9T98;
z$|R_v7Gx=AOAP~Czwxelyb>qYQL#>4w$5#uvPs=|dG+3p&@m0NmOZoLeAeJ(w7>rL
zDNh|(BF=ZQ+y$mxe);rs?$`laX0q>&9l*yE(K#NP&t&5j0hLRH(c2HIXLZ-^*I6R(
zbJWfE>uWSG+<<x#SSwz6wO(C!zr7#+2NG4no|nOgEESitAPp0j(G4uT0rt0!4lG><
z1QcayKbOIOC=&rWmoU9SB$q29D+~cNmm!=eCx6viU2j{t5q<ZsAh0jn0^#KhhhN3Q
zKz6g)W|6dkodRu*Jjn80N2x3s`Uuj0-!qi2Y*kh)$#Rf9B%aGVB!|P{nKQf!C3@pb
zC?OhQ{Ef*_$|@AYQHA7;cToj0Uh1o8e9=y+n85lfdlP-_9!xCh2(Q>=_NwDe5m7-5
z5r5UDV2CBC6b<>5RKmQ$6&;HVMIn<ZLqM<044bIxqbY@m@e9VtMW-xi)Bn&j8+JkG
z$}pa?jte{4GpR%dlj{8F#v$u?#T!nfeM>T&L)8;Vxg14DSg<2Dkg#HcUUUxLU@9u%
ztw#^M72`ux$+5w5oJwAOq<x(jaXzRNjDKK7D#6<r8gsa0M7(HQHWC(Zi;-Bg4epZ6
z87@GFT&I8^q+oPN4wC4=>S%J>6}o{*-!e9cy~h-~dLOGE)+|Mrp}U4gmEfh2Y8$*6
z>(V^>rl37^Q*)ieV~)Jg(_La+WE}w80SPZi^wDDg%Gfmw5}9ig`fDsjC$LPMu76B*
zVY9x)W6)ks@R$~FvC=w#!`un(#}o)pysUcJARJT*&LpmF*oef~@r>b;=c<+<CG{C%
zB1x`NF$v313B!fz3xs}3o{lmWky3=|I#sTl7cfQ^H7Vd_&TyMU45h-gr%H9ZwHG;E
zEFLr0pouE!N*p#WOA;(1;#kDGQ-3%_IWOMA>YUVK_y{pw+X|$pE=SlFjU2S$B<Z5i
zUko-tdt_&dCw+^B7R2MxPd_y;&TyU+j`d0N&wu^fATn&m{!>Y|VC|#%aQOcC7@DT}
zWjq{DPOp3Y)+i2~nC5M7Hftv%!;8mXKhJ(Uo%Lp|p+&>X@o0uQUc%4huz$}w%S$+i
z2u`mLdVpE{;sNIm=Q``d@p1EZGVY(Yv$N*yi<iy2_Va9ENAvE#*DYqgY+=x7){bUV
z#gf{qIccZk`J~@Y7s#VcZ`zAN@8|L7Gl$0rXJbF!qyGdwHA?+ry3QBzt}h~dfenY3
zZ~d^YweVQX!74^PtjUEe`+x7px7n@vHsvjRTXMC%c|IDA7Xx*^KC9rb%c$UgBlu2`
zPFU*1PN?2|zcjTD-JH(PXZrMdF#3;aejZOQ+DSFZy>I^7ylQ?qqs7BX+n=3riI(-4
ziddXJ0JFCONRL6R5bS?$F#Kur+xXqMd0~E>qSIhJ`m>cB{&akN$A3}#%uehr5r8~p
z3x7}%t0<kb6{fYh`yYDOhyv^HnfU&>u+4_Lm5R3+2}d|sEZBQz697<ru+G!_qZQMe
z=gZOSdSS&sPG0H1A3w}y*VDf=&DC&zJ{Z}{!R*8Q+z!Ug^~E2}$NAOuv*}>cezJIk
zokx%KkWa80xxh!Zl7BR*70bXlpV;C^uK6f=NvV*CWWXMUaDZL%*d;ysJ~$hY%Y9~x
z+y<K*wrhQ#xZns9NPsI0ec^Nf8*;QBnWwWshyckOO#u6e$cO<68NknpnICV11VC0<
znh?a*qqE5dK^{f|1~qY*-#bvPr40&(sVu=T(S5jjM`4Mp9Di-{TPO-8%ij;Gmbiwh
zMfO3}l5bHp=Nq8R<whv36Q~n9z%Bj&V4BGW5I>MN>yQ;ZKvhwifiW?b^@Bn%EjXN6
z{7;;MiwB5OfM`_8eq#jKsufV4O-2AkBvRy{M7k(R6Zd#4*ojpR$OK3S%VywLLJGq{
zu&6`63T9|_QGYH($Z-PxNWiTKBFdmLu7s~<1A*<6W;hf+B-tucCe3kPz7}w?&9wqt
z7!km^l;>_V0)-25F-4zo)Sbv3BNT5+cz_lj3~cT|$_m)*hA&ZT(N%n@Mc~f(QtR0p
ze5n<!T3!p<g+S7^a3}aqSiTCK5IZ4tLhgie05I}6aepWlB?kQEO0)%$Cn;E;_Kg0A
zX_k1z7N*gJT?nzpfFqQV9)~bE-koP%jQdmjaXuXMhP~0H9ZxQscGP^L=BhWE_lC_?
zJL_Ff#+Q@c)zx5h+5GV9=j%y3ono}<v)<_9S^KFsocDC}`}o+yD<TH<0{emA9JngD
zmGw>+D1SCc`sNn2D?p_i;x9-Kz-OPqFO};BkehW<13W7zt#d3WNJ30G(gBhMm=wzV
zr46OnfWI4Uc#@3e36+r2yM+-x6)-K?XvCA`Zq`s!TB}tU>OAk0EjYU#wTtJx?uo#e
zk!Fc0>4h7}{{gJ<N^Fs<J1Yc)SBcUlD>wir^?%w2=RE~v{CI%Gi9wL0%hHITMaLIh
z;W<e_)sqm=g!AdbAp$efqWW@R5}?yQ7BC1HK^`SQ*;E}{Mn&KbK^{P&JjhcmJoIoD
z_yVK~zDa<dox<V8S-qG2QpvHRu&g97c?^CFO)ko>^4H`6%}Y^E#qs7r^dYSkD!bAJ
zmVY(6z`PG#;CxFLxT6b9kI(L1Xat~^UK(Ss!K?M70-EF3Iij5n%C7^C=rTOP31!D4
zsv*Vx;oHXl(^>>xP=HEl3rz+)d78_{%@O*DIP|n0hh_UXymf_l4;=hE3QMmhB%{FJ
zkF(O6vwYagS#g^<T+!-oXF#!2*+P?wCVv!p3g2#W=OgZCB1*tD-^yNqD0k{7-<3qt
zA*0-fj1Mz(rRJ#bBQJQP-hF|{MZgh-xI00=xzg>AuNpA)AvpJR(7bu|x7Tl956<n!
z;Xzn;vZS9JGcq|+3rBAkUyqXox%d*XDEpOV2xO&C-GrJbil^m-k8x7H^uB06HGgk<
zvk!m2p1nfYO#VO|!&l8<I-R%EhEOlu`b|%T_|1^?6o+BwkT2^YpZ0+h$=krmiq;MD
z0!ZjL)_pV|<G~fsxZ}=UJs8582ZP@SG(x!rjlhp5PvSsSDb`Xx*B=j9{j^N_%`4qg
zZ+hK&Le-H4{1!EDNh`!9$OGB+j(@`K7gE_mE9mv9D_dxSRsjRU7Ml7YOn>vRg(d)%
z;!}s!p}rzS7uN$W?iFy!-3DCjl2hz%`6BO?rSu*tBvAGkt+M98R(}kkG%N4(0rEZs
z{(b)Dt)lMvaD3jxwx1)VbMelFe9lb0e!t~XE-w1+JfkM?f%@c{>7?IW^?&-4@f3Z3
zc<o)`XRkexKR~7v(Pb7KGSuz*BO*YcPhupZU1ZODmF>N)XIyzj=Wf!kcf#M0yD6%>
zDY2|3$SB+7jd%}e=og>N+h~F!=}&1M)+7{8RQxuYU|%Hk<kWVhhFHOtO;-ZD!bw_(
zle7h#K&!06NxlOosY8+UI2;sVpa`nzWn430q=fJ-!3f0w7-8oYas5B7&4icnA3GBR
zHZVAsfsZv5mn0x8zklm-{QLcngDk>I5p4V#PyE;G`{BPs6J2RxZ0Im}VT0-Z`F4mm
z5R9-cbl^c~)qVfB`*Qe`Q*FE?J(53+?7n~P*p<%(ydViYX{|9M@RSwC_|DoObi9?p
z!?Toz6vH<<NO>ff6pWPo^yXofqOn5jz$p@n(h#&rkYb{osecKKa@LO|{^$yc-zXCt
z(a?;tk|Vj=>L7)7$XDx$9CU%)M3R>9T4cHo=~X<@3Dj-_o{~&X<Uw~386~cxE1y?j
zKIW=OSralhV|vbLiX?skITTh+bciUKIrQF+AFYhr$NT^sl@SKQqj4N_-HYRx+6o+B
z?vJC?!b<gM9Dm1L_u@FFwgN}eQonAFGfF!FPIG>M44(}Y5d<nXp?Kawkx~SK&ncQ&
z1QiAmw4Hc?yt|(niqY2N9rFmMh}-jqMd8tLN=tdeY04LtufD;Lz&T+7I3;lNd`os?
ztK|)OvSdEa_ww%IILrI`g$cr0`{4M)xW@RC%DA^2Eq`LZGP=Z~_}k_qN))4tf%toV
zM^TA3p?@JF4RACCx*o@R<pPd%2Xh>Y<oH~qUb(lbkE0h}x>1c5t)Zs)45Dg8sYGz&
ziq@x_qb!pFVRw{e6j@B%MzNqirYK+MeZqKwXeOw>Y{$kb>IcF-v9VSJqpx7oRyz<%
z`ib6IlYfa9ktIF)HC`02#g<FIMt|Loi$b~*I7b%Aw|Suw0X(O;C)ZcMW)#o+{hCo+
z%JsQ2V5?KEPkUHB<*1aR)N|gy*lbd(g3Y!e3ya_<wt@L9t7Q<>^-g@}Jd|U1)NFZx
zvXJ+E7vXbkN4G9oQU?iCH8Cm`qh?XsUj0XsbANPYn4xP{j9EmF(uXaH(X3BV5wN(a
zQp!GG3De~+g_7b5X$nGAK-Vi2-pV%$UDaV`*5~4}cdD3Lt4uYqr|-Tm`PBxU)?;x@
z3J>8h91iaB;D%~6<>WZRp%%tbM3*Rn1Ma02be0=$F`pgEJGVpAm}w+0WO=Kp&STAo
zyMMU(H&R$2MI5;vtI?drk(R&y_8~q>kKp_xjvj&WBfosOFa+y&iI>{bWy&1h#+okB
z*2=qO?oMCjO$UE<dPe)xx;}>3f~_yO+ev3x;>F@95y|dB`?<O?*~H`uHaWJ<+bdZ#
zG){{$R}OI|fPH6N>fvbIIIN78zKQF?h<}ti0;e?N`urA0nc&J)EOp|Mt~si_pN{Ds
zl&WLvP*ZzPsdX~cR59V=m7yj-ujmERkaHn`mvF0t5fEU9S*OinZt)o96{88;r=j|N
zke*kDj+Un9KwNZ&Y9DGPeCVAVv$>PJt~v&_+htmAu15E^@;a7vrAuyhX~QTTgMX9e
z(k_gXmvoK=wGU=OrM69x%Bkv{wA2HcWjUXVrfz73rpv^l7xmKAEI615v_xcHE$x-i
zXgQU#V;5P|#)PQ@Y~+oALQ9x6Om)}Zjm%X*ka~?UrB@?U=`a_iu@u?dR40`j^CDko
zsk*bXeEW`ikuOD7<^7@sDC%6=f`6!}&+4;4+4f0anVoPr7u_0x<B&o+J9(`{XG7uh
z5$}~WPzUYx9p{0J;vL^BG0IooE3wrNdatB{r<>-%b2P?v=Nz9=Tl-$AV7R@@Qr@K<
zUOnc-@VSFkDEff+kG9H~>mAA&Mb)Veic`LBp*~9AG_kJpF2;9h!ck^fY=8AFGe~Cy
z+Nr*;x6f+4P7M*)WXC)yD1rWph<pgcp|O3vmgfMacYnEGV4{{u^)2!p0zISHYT=?s
zn`z8B=8j;YsiXomQ-0Kzdpa*^O?zH7o}uJ_m@Ld!X6>?jtfUlB6Wyn#;ZR}#oj_v0
z6cAc~?d$Nzx9{MsprCZe_8lxqF@Ar2-vXl^q^(kBrJ}cZ&y`}aaX=73>DBG~TCSJd
z_c7NW+`fY}LIwL1weJ8xHNM_q)}{8HQC?AzO!(dW{)=lR7c0;DHi|t_E3q^&^ktN=
zXD~nwk#b@9Kf3But=D(Ci&rMk2--NEu7sz&zRfXQ-OJe@W5-;dYwK6c7L<SSPrLVT
zOwS&HskmUx#jaZhN{W^dskkIb%3j1%(ZMzObEF;o%4-@DOLL8dRz5h2436uwZ8(Bt
z_gM8YB5XF2C2p2^=?lv2d`+i`J2u_wfPtfNNkao#0C2Epm|W7(#37R58kWe&-QS|n
zs{h}|`mxtl*G&H=sH<yX@xm--B=x@U0n6h!|1_2)KiL~)`=E}v<t^Dpje~7F`(KDY
zov;v3Ddc}J+0OHq!G|mqmoZ=|6ahGw!EyyBmk|Cg1eew)ER%ov^5;L_emlu5?59_9
zvyv(;$oS4!6pY<V$_T`gBFn;)Nn&a7l`of2g*O2>+`d-7UfxdsoN9P-QUE{x>*pyv
zKp<hP@0=3+@Ymnn_tWn+Xck-w$s^JS$?n&Wu4?&~SHeryPZ~os)z1iG)eR}3lqve*
z7uLyTovukw1(kor9Lyq;5nB5iF<Ob&32m@aK+cxJAZPiA;!j#w{2tjs^qLA;$%$Nj
zZ!d+mSohYA9NfSwDOpQ+$uiwG=}miTo?W^f_LRKh3=g`4Fj1U!((-MC&8M>}onzoN
zduZ@h8oV~>PuUJcvsP3bsoJX=IarM5EctqVGBU&^U3P!3uR<@pHjNgtS>s)(CInGq
zo4AYUd7Ui?C7U&r*Q^ogZup3EPg=$vTf70h(CGVExCASyLAPwPBvaVk8rhEUN%B_U
zu-hIG;}s(Lgc1-Vvdf-wxyyRwjKHuhk5_KNO7mOf-rX31)pGT23?{ng>UJEr+6AK(
zb<he6_`ZK2-4uGwGL$FD)q>;A&ZHKW@H%&>^)9k-#1~tF78>mMi07?F1!<z>ShPDo
zDGHLs76`h97X3W!ANo?-u3|z!FdbLM6qI14c)t!-i{nj?KI&CMA$~8?QQvdh7Mj<;
zv53}2D5q<Sw5S<n-mcn!@RDkjV#g>;SU<KiUb24{gC#=`HU5?Q^eyv;^pZvxHf%DD
zMoTF38lfdY)-@=f^IAkH`*kg=Q=VLlzxFz`-XCUf{8EC@rr%V^l(h`0rX6gxrazz)
zh`Pbr99ye);Sr1uHF`*WQ3XwP<GJT2cy7Hn>vP+C(`eI=kh$lzq~A!%Thgx-M}v>$
z){%ciqmkDXow?73n9JkbImvQK8ic(-mA$~^`LClmH_<}LuInj1IxRe?N2T;kIE1wS
zT}F?b_Z)fFPw)=R<|nu}a`{Q$56a}(R05nj(vnlecVx<7P&l+TQj;a$W)wZ5<;|D3
z@*42~2BDCpZ?t|_8nc^Ggb=rI(@r}Oy%K-eIK>K?V514|TJ_(b4x{o+X!Yu8Q{{~M
zCCP|#$OOmP;Dd<=1`^3R%Yh*@MSKwP(4J>Q^KryO6Cd?_Bt3?BXtIKwe<1N#NPI#v
z8b`gBBQ>|#>ERvS6ozZbUFViN&f~_3!kxrL^Bug}LR_?y6>&)|OT@)y;!?(iZ?S*h
zCOehhv8V5UZAqydY}_z730$Qzh{7wVZFJxIuaktZDjbT{jJIe{UWZJ3vi>95lUw%D
z9tvT}QCe%ikM{I_1|XDPp*?Vu_NWEYpP!7VO>VhOZ4{F+sdXdWrpQ~THJrmKp0q+P
zyGtP)S?}VLI^X-kfO;O{Ar}COjqHDzHWZ}63h8^y5(AZrJBl7rnN%834MFa0*~iQn
zoAT_%6VuF5E}tc(F-p-0GlAhqP7+^7ntX27KkH_Oys`Et<$bhoLu|*?E=p+f?x@Id
zV#1jYUrv|9v1uLZ&#f;L&&^$`xx)!lJSOf5#cOFwt#0ked<;wPYV?gXPX&LTF?z>z
zFwFxZ0W(K7--(m$k6?OyGzkg*!1Ss@Jdo{G$T#?2)%$}OpVCf%gb#jqQ_XtdyPFDm
zLmqccxfoveGcF1xF>18A=lu<5zhQ5gA_Bf2F~8rb5x?AlNCHh`X)p`#m%OlG8$Xlp
z;I}Tm+ia|Tm&FTZt+k;$5WRmu-&kCJnWxO)IA$8HLhRi^=9IDmaG>m=P$&oQk8vc2
zEF)w~DH^n&W~}Kf=j&yz1Y2j_>7NX@^6ik2H2$|kN6lw6A3}<mBDiBq1GGSR-yxkO
z_Ey|29j!=b50ef&3f^mWOJ`Q3^A9VjaIzybx&zPr3URTd63>SH=YM}b(W_a27S4YZ
zMzvbXyLJFDAoNiGSoEDkN;$xxs0klb$^oS&f1EnJl=HtT#e`E%<CaPcDpyrvQp>GM
z%y?Y*<u|@U&g1&}MtX&|R=dgasNB={EgR}LNv@6co3R!v6_SjZD<s3O!h3Ve&KJX;
z*^jf;!tE@O5c@dQi(h}0N9G8yHG(s&an`U@InjRMg)u+D`lYbf8(#{0Cmfnej3Wvs
z$n;m*3b?MDBsrJrCRvM8M~qr7)lG8Wg}R9$Ia4>ub{kjN29#0VWU%jC-6ZQUs+$Zb
z)4E9?MYFXNCO`XXCkBNcB`6<RI+;#=aOGqQ4i$OLcRt4c1(pL%%ng@;H7pbXFqgqp
zDH8!Umr>LPCzmcSEDnE?s2^YuAd^lC>}`v=E%vl%d+K5P{)Z)HO15P?j+1N;PGVW6
z$dCVw?tXN-`*L{uIDEQT*98?3g6TfKcR&v`eS)S%k6i#z-a_~Bv-|5+Yvbv9qswV+
zEnR_rNBYL&i-kx#0O_Kq0LPc$fZw4#*7WaM-@jvNmpRin(D#3d*XjMFy=DqguG8`F
zk3SAtnEiCYxX7uo3VrH)aEfSGPKg2%r;d1`!!#1N?m}1WsjB3eFn@h%?%qER{~lWM
z=)6Mk-9JAM{J;S!A5sVOjXre0zIERZf3{?Wb*(hL(|@4ue*Ng0m9L_Uq)9^UJ>pbC
zI+}#{2+A2#By@kDNvbGPoqI_QG?+zLq=;Twn_7w)C(@J)j1yB9tQ_-MJ~I206=J_q
zj*#u9p=|Vouev*FWrJ&X+Zi9UK)Tc<CA<qy+mLSJX%g*p&%|>|5zqVa^G-z<T1pLm
znrx3=1=|4&oGDIL#uS597X6H(74N%<Z?K=>;4h3j3jTkHi`p~B5%)8xSyQ;A!OQW*
zYmNx^oe|!RjR~NPj+@x*p_o*F8QS3Prx>kD0cI3GKR_`mV`lmTzpRsFl$)Ej-VHXH
zgi0~Z!F~vE3swa*Ghy(4ipD78XMpR^i||m+QemcV_I6$D5p6eo<fA7mW6vsa1foKR
zgJ$T`gWZ2;4=Nr7>{ZndlI-%mud&Cl58`|8oB1Awh?IZF!B_33-)c&!3)Z(oMglAX
zDU9j?j5WLB4JWggb0t8=C-RE%(48rj0!ak+VeT+_<xSkp9bVxsJde9bwE~q8d%5#h
zxJxVCJ)RZOvx~wLvtmOLQ%{8{X39}=2G@h{`!j#0)DX=nfGZV;%a+ngsUL~}h(?7}
z>z7sASuPF<1RH6YvUK#E$&)s6?$VR574g+tm<rX2KX{2HVqF|>N*!Evn5|;94{(Gs
zXUe;(2j(E3w#s)oB^YJR&~1qAaXrYfjxoLa<dr@rp>plPrwg|u5Tq2%#_Dq!WY`ta
z9P@uOfcshrXOWx*+E#)RSpv7wNwnWL6oR?4Sn*JaP|g8)YJ5<;Ghd5d&53W9hxNGA
zKzdKXWP_rjw>>vIBn(<>3n!OUvVO+xc}d6BX;zG?C|YCYWYjzt8>(qMBXE2%P$lOi
zAIi#je=ZUFay;7~Lfxf@Ae||mCk-#{GOvG9d|WLl>dxDgF1w2y`KkDddlR~$(#!=;
zy<2WI*8+s46ly5;HHfZKS=NiwQAN^iLxnBRSo>NK8FSprQE~QTJgWSX(6GpTE;vsR
zDwb?8M#EN1SF|O?%vP_2c{*NE)R@`V{Vm;=CEH|`fjGu}DJsSxD$Xbb(qODbvK@b6
zOEto>F2_%zq_I#|9m?BIH@f2IVBbnHJ$<XUbMXB8bj*y;>hs<ttaFV^PT5&=Heu*S
z7ZSFLVZpL^8SF3uXW+(+-`eD@r`<Sz=A-BOG~(FGyZw2KN~J2w+Ho7jtT^2RcRl4(
zoQ)Js8xknzod$A!QW%AAsey~Wgts0w6E>P^G!vE%x`@Rw`0qkGIvH8!<NXLmD2*CR
zh?mhbEEj*x7;WrJx))gDD-^13Mfq~No7`1xV&h-(JPCT20ZBZKXrGo%1}VRW8V878
zpUW0Uwd>KuTIHNsj3zz3<p@=tUd^iF5c!U*SO(j5-`Jh<6X%D#T617QVL+ptJ>fO(
z2^EzOSI`P|^;F}!NLD4tR3si28w?fMt+Bf00k?ln|JPU@QdG&!$LgBeG#!7-Se-{^
zF+Fn(+)#YtSiPaVF;;JO|E94z1y7ZN-+|{XUCbut(^H?+!WjPhi_Ej3ex-*Qp@;(9
zfMQv>f}(Nc0TE^>&J*KJ7AeDjYCV)5)o*Y2Ti}$^gY)1rj%ldoKM?J*Z5N5CU(=Wd
zeC~gjXiDi>5>2UQi8!C7rGedDmQpR>Ml}qwrbW@FefXq+`l<<yXU7BZ3gR>WwrnV_
z4iw)W8tNJ9u8gr$o7^?n+nWoKmC$BFpT`m#iSSv2#bH*oS`dJV)C!s>&KAi4mDMl~
z-5l<+D$iml9Ar&XYL$OYa9VO<xM|kNf7Bjn+!{6B&q&`}ZuBPLNCk1q%g*UzGZ7@q
z`VYXI&6lB^2onM}F_$2k7ZjJNH7q>?u&Z5{=`}1We{Ua8UvElZtZ)PtFCV{NfL}=d
z3z8cRT`Yhw3N9bNUH<tjB^rL-_|Z}-#gFChjz8(;ivq_zfcW9J054yxhA@WvC`@CO
zNn;fE!Yh9Qf4+ux9`B`*VFs6<x0io?{QXJh#Y*8#_!>UO|K2~I{`XXuj!Ka67eiKn
zgv-BwM|%n{G>|aXUl4@{efj-AmtUX$G0s|Ojh{S{`+&NDkn-|-7|^x+l7;Djg(;08
z>V%0#5V?ynN+@fqFj;__1wqWLW?|?Xm)SNfA`W(T*1Q5<74aBSsgJxSmkBp4ElHL>
z0}VVc*)6V+j4(`TpcrSR-L;}|tTXdkeUf+O9Y1ZYN;4%huhpd)%Pjh}NIUL-&=nV2
z=sEWb(vw}Q%<C|@f0@bUy}Xu}e>W^Pe>Kz1GL36XquE{BESqTaw%?27OxwLkEhjl_
z#uL`Dy0*(SOQh<OeXzq`<A}T!3hBslM-LLL7Xddk4FQTQ=>;L%PL)N`H<Nfty(3k%
z=vVB~*vdjNT6^%P?=LF)54UfVj%|`~<doiCz8D$i170=ooG3#}SPSTVn+q{UIGS<B
z<bhrg)0)a^vww=*V&&Fk$*KGlF3XPDc6^j;Z7Et8<#4P0%p&bLgD$hx{YdKr^V~@I
zT9<4%EG&QC*lp~wXE@3&qcBJ3;+Lak+^VkM4VJV;VLgsu$)Vtv`&cR9pZte1HtJdm
zHu7>08<Ez+Ay!-YmeD}5W#$uM@Z1|{i=(qrew=akd0A3Ath4-y%kNr#X^@qJ2#SZ9
z*)Oc)Fv;7j$KGni_XcbxnM-)jYdRL`p**7x(szHz^9;y6lzC4z#ANb|xfja8^S9bM
zNlpuc6AwX3YmrOdieIK5AV<mi?#v%8$DZmB$`lZtFh-Bm<lEhrfhuHGleB===_=05
ztQvdM19Ua}DHi1@FPN<Iu35<mhOQGgdYu~+oaD4;C10xa8PzU-BK@wV^)p%bG<y(E
zpFe*mi1PEE5_gW^G2HtljMOBrnv#+YM0iUXeC%aY=3Z090>^xl4!8uGLS`S|b#NoK
zsRC^?e2I(m84BUjAJSrhzw*rS34VYztF`%Ck6kqfHA@`SJe!zrJ@4nT6qoiGR|`v#
z9i3sOic<z3Ht5gSn`Si^-rA5}^nS{+4Yz;1{`GG^B}k$5`Vh-Qpda#=pLU8Ziw--f
z95*X6tm*96yu_@_Yj1Y4xC%UU=i|vVSrQf550Z!;d$VMv!3gj5gL&cRdh_hcJ3+o?
zc$nis(5+Aqi#l>I#SLJ~Fwy&w;_J<n+)*_d_>xDx#CI80-i-s50S-X!PgFEh419kL
z=<iJPmtv6$Dmu=ZpvA+0aF`9gg=ZHOj$FNFAJP<1VxTu0fU){%z3RbZ!uOt2L2Lln
zYg39j&!>_kP^*bgD9<Xd3LJ0qEbH$nBq{|$FVzU0v^`Z&mYT|A)siA$GR3CyCeV^|
zX9~H#wOOTcX<f_4EO~caN2!Ozwe5fU9VL};?9F<{`&~aQMmYAx{R9vD2Z_-t2mcUU
zt}^@n;f$?IC!Bh~)|ccU+1_YTDNvs_sv6-rOsn@a3T3PCouPf(dJE5etCrok2TXUM
z@;#=6obnrc+LT`VVqGdug|}uY4BcgRsjzg9>A43@4<9r=hLK9UK~VRYp3#585)Ko@
zQjKm(Y%))37ryD+2H(4-H%`KUgX^%~>+Tx0GQ#O$OvcljgfxdeHS(pu?Gw_$#{Enw
ztJ)AuCT%5_WmBf>Hs0oqH=ncd0eKOak>(vckCGiC!ZbsvY{RnZ{)8X}|8~>2Y9xw;
zV<5pkQfMqW&YN7%xPkNJa7KUD_pWBE3C5EjDk);Fq$Z=1`*QP|*MuXLI7@NGrUI91
zl0))%WQq|SoV@YOFjF|;aJrMOJ$~y-bB~kGf_Iu(IpBTQTUd^aG|f=XaPkUiiC2Z=
zo};Dj%-<yC4f{6iNGwb%Ik!h4tqh1|9R0eS6Q(o6%2PgL$p5WdM4x}(XKiMHsqIqZ
z)Sq_72&9BWGg63I9;RwR%=B_CbJDr|zv?A+6q%7>keP^P1N@@2eQq}0wJkCB>M0_|
z_kD_5`<doQvK%axLnr&G2eyjC1sOF7(=;b7g{yp5Bnq;rtF$GqsT{DOhJZ|MBr0#8
zcMc6UMfZU%%1gC<(Mx~ZdsvQu^#QI4H7~}@rcn<GKg~pn5pt*_Sc+^au^DdtJg|PS
zPLQ2ujW9u(=cN!A@F?uNWaL~0JdATal2vS{3L!lR)w>0}2Z#```OtdJt$~VM4}y<d
z#Bzir58$4WoPJI6JtdET))kIEMycogeg+Dot^|M(JV6gXb|HTSF%%;hM+T|Sa)w#X
z`yh=3<E(M+47v(HyR{H<$~!8Ui+A7ja`dzT*MfeVdKh>F*ZSTITmwIb_t?_HHFFNG
z0lJY&$oU#jcFnmTRGmG5z6ylbG|n1!C|@vo?E-r)j2v8>;+@hs$)jK2?y8&>r1ivW
zfgVYsO>E`V`<#D<DhxYo#+gGw0QdDzu}&zpBB=eO6j*8fFv?jtxCLee8~MGo2nENp
zi8VN8%>XN~zn5OP<6zyy+W{@h#^U{~#$CwUr0GL`IVncI@3ZSl*@Y^Ylydc)w#_{@
zjJLldRnNz>BA%^eW=Fu|AiHJQaK$EIP21%u*ksMajV*sQ-3?4FbzO2OY+ReOoZukS
zoxyS+Esf3aGfAno`WEuiI@QlD@>m85qdX37co7KXJ)cIM(>cQ$<J~m+pqgYA=TkI}
ziqMGUeb7kxaGOYvb3Fo$`r8*8%Gqq#5Bc80pZ;BTkmZh@K7=k+b3;u9@J{HmU+-bi
zsiv9peT09_^>`A&?+?p4rveBe!E_VJeGwoAj1JSNdd<_e*B@(5v>M??&9oLMuHA7+
zF`^Qq&4PxGP>yoKS~!bxy537ZD<|WZ=8>Wlha&AqDi!5~@KVa|w9;k{&<+Oyuh)1u
zx(d!A*`e847p$xc&aR8O4zmukN29$Te<@_pLBoG!hjAHQismDO=X_+i*)x(n6P6V*
z(v!5x`FIScjvb{_g1MhqKY&|auIt<>*0r<@D5kwT0lPv;LfT2E<h58+SAb5|M6GSr
z%i87aA@!Lq$21<Hx$VM0p}k#PZu!xYmQJRRYy@B)t>Y$#x8l%LGp4=sY`4dJBBns^
zIoW@}iMxaNF2hY88b_@;_k`}6B$b*Yq<rw14nGfY2nlpBt@8l;SY!j!^Qd!n?L4dI
zESrPcIH#%+<M*SE!VpOQq_bhD>%D{_D@T2aEpxJcj5B;V`&J{R??YWnx(=|GvSZed
z19TNGYXGeY;UJDF^g;#z{l_!4-=}z}ql$l+v+=sv$oa}~Mc(+0pRH^pTHXoY7uwq2
zwrqN%`);mm{kG%Uf<HU|BFMGB;p$pT*-I|hD4O3Db;tcH{&(X?6zTrL>iMqK!T@X^
z?{@5E;iSD@J28(=vHtug|BU5zG1zqHp#2e6L6dRLiSbhVvrEl(Hvm|}shS>`I=Cm0
zwQZRwB#d+E4!o`Ek!`$G`Ylivz-wP=7_O?7&-ag~{{uz`I_8&wH7pbYGB}s9Z9Ef~
zS3xYn0Uoy*LM(L)1WyYcAD4kOEEJc~@GumY^2;D7m*Pz<n180HJYVqi$NR4jGEY_t
zZ^C1^_5a@9AO3!*Q`aEK_z4MFE<gSG{UN+SMZ#D=X-z_rnST5=eSP??CtI0G9+Q4Z
zPCvd)UCWn@o|lZCG=``_Pa8qxCS$yS@G7B)uTm0Ahey6xdSslm>ZCN5IVhM^;1wA+
zEzJ1S(W#SM$$tm{-*eTxPUI6eqM0*No{4K_P^ozoMh`MbUVR=YiB}YD7@erd<I|&&
zVZDiaNams565R`+{18X90>R!J$Cm3}99wD^;CQ$_j;#Bn{Jn8(x$ecWrFH?1EF}~S
zF@RNmqBKp*mzkdiE0l$vpTZ8GD=4B8sO>=Ubq7WEX@3lUPtlx2P@@+f%%BDG_If5k
z7^&~AXD!#DXK0dU5brY-o^94SOf!k?p=|{*-UY$yp7I%;TOddov>CzDat6VwBOQX2
zxjdJ7H}hRZ#sd3pGm+QAAQWo!a<w%mr(!(NQPN@mmYemr*mOs=v#~)Mo{Mo<r8#Vu
zrlq9hiht*k0xS-EpRecZop25YzQ<iC>JX#hfQzE@C|W73uKRkl4wv*Ab+!tfpR)x8
z4TQI#f!BLejojjUu6v}`P8fqd#XbHRwD8Ie6nFUNyodA@*Zgy`{jPWP&)sg8{iwWt
z9V`hny5}L^PS41J@Kz1FvSIz%!qJV(ze@!`l7G+$AF9B1cU-m>A}_2lbxB+{2z!|u
z<`7!3Gw-O}c+(|THpduWzUh9np>`Ar=76nioU2=5Q4xY^{)uumElcNE<lwt=bGYa;
zRZLDQ%2L9{EL(I|hi=t3aOM=z*Ifv*M~uu~^CV&nQEf5$$n9)F3a3~tBt|RS;E6+~
z&3|4X&xbt~Kj~zl`H3yNw7vFYbn?LYX)v4XB|+ZLrP7-up$(GI&y$2bpCsHQiP<2D
z`FWC9Eu^PK?3ZXet8U4~C=S)l97)CaXq5L*xz94vxjN50Cnv#iGsiB@gzUhxM@@4(
z^=8f&#@bdX5OD+nw$)ah*O`*7ce!YG!GEjOGUTXTY1d>RE>%RT^K_Y0c4bL>BA3<q
zX6oQj*hMWaJ9s7>YjH|#;J8x#bL<Fx&++*fs&vYJPCR1{73k8Fj^0o|#dj)iGS}AY
zAf%^=W4(@pGW?)i*WWM5CS&<-W&g|j0V|ZyT5Vj;wZOZh8g|fH3VEJW(G6)-M}Ije
zM7qu<xBYw$r<Si8z>30@HRL<eTrzVCUb33R{o2iuL)g26I{;~Nf*Uv6va6@0oZnJ-
zi?!czjz4NtK;>;jfBuJ=CT4m$@sHdCA5*Eiq`LFUCY4@U-7>2@=MWI&ma;9lGW$3W
zl}FmjFV6^TOAaN<zagTkBM<ql%ztm}9Np0({$gYm>ST?NsQl+tZ(r6yZ~={s3_`C)
zbX&aPQ>^W$(NR5yh36cU=6Xe>{}O@C9wR_(I*9c-h|_{RXF7+hL@(u>A^yYC*EKws
z6jJ4+j#^-V+0R!jY}>2S)g5hG&f&7^QZ>R)NA@dc3`nynqc^hwCB&knGJhertdU;p
zV>t_n8A#`2Lp=oaU5-RFd%=+8-Y!e&D4wvLt>vMPvOZrP>UzVetmyVfBbtKY-MggU
zaot(c@2FjN#I&QN-|@RaNx!GKzog$&bZ1NYy%yZQq_4FX<kh|PtmV2_&su8#$`M?M
zx{QT;<JfZDi(^aevLm<~<9`@Z6npoh^^WV#j6jEU5o}NO0z6F)M{wp@!nmAhFA5eh
z!`M=?HJtq6FuT{2vXokE*E@OYu3Wvn9t&ah6%DKJYwUkX<N(Og5Dasz810xJX(3UM
zlm{KSWF=PTZ%{Znps-^f1LFmIgn{DreGEdjs|M{^*~h?Ij!y@xmVbv`m1Jd6*g2Rk
zy4V&?=i-#2YCSKf3XF@MX6R{FdfK6<UFqqDo^H08D}wM3Jtym^v-ge|JFb@>&Y$<3
zE!C}_^LpQCjPR^<J^z(sw0!8+_1<!fmJj55H1{3DS5MzkU9(lmnV}n2bdQ~K6z%nY
z+)U5ay47Oz&COYd?tl0G)~4fnD|2qC-rt-%{vVig+nZjun{!KP@An|ZoI9#Gd{a9F
zZ=L)*Oj>e4vHarbe`3?GC(f_%qi>ZjES_5<^z2VzE}@=190D&Kc4Mn4`18j+L}K7u
zj&-n<NTGfr_o$U#6j75&mdbf~TT81a#)N;+m4_Fju++6{zklc4dZ=`;mY~-V0d^)q
znOm_gOnKL!0_4Ed*^X#%L>i#XD}51CB+4>ImuQw!HTWGCNY7|3iraY%HE6jR=fn{j
zjP-yqoGDjR{jg!kSbG)a!~^tvDz2l7m0EpM^LJO*{BxbWOF8oW?fu~&o*c{P3T19&
zb98cLVQmU!ZkLWVEC&HJm%$b%6PLj~1}K-SRV<u;laz<QUVlI7VysrtrJw00|GT`N
z{yhz*8>`60&OuZLZTI)LQ+j|v%6V?q%JW#i|I>Xr{UN0ex-`Gi{~_Id|JsR`&&7FB
zv-7lbM2+)|Q$(S2E-ITI_0Gd*t%(PRFMRRfk=<lK4AObl0JejqD=ThVSn<b~0E=45
zDQlyD)M|K|khKy|8}V<V^1th^WC`)~P-{t*YP4nLUP{$Tn(p63-%Cv6(k~AmoKDj&
zb=phP)$350{8<0wQ8q1~C;FXJk?d?#TKH{)L{AExkrHQc+afV44>KRZ^L~lJs_0=R
z(JjwW2+9~Y!<%mAW4rIIjtb)p$NG8E3SR$zjL?}C8E2tKOGeF8!NPO0u-H`Bl^#&v
zv}DvTuQki4H*r4pSWr!kt!rz?daEqpe8nEOFMtYWhTpvEU^6~e?T@G^xI1Fn#-IY;
z5z@$kL`S+2(#ZM8;w+fybGpi;a8wxoWZg~6ME4MoDJaWe8Fcfd$KZtpruohf$k)Yx
z!?~&NmB8#PwD`L8xiG;kT}A{w?mqX~2D`0R0XS5lSI9xQE~&DFQl8}{RBY6j^vg`2
zQ+;lhpSy7h9<k0kg+2^krJ0@;I~RJcSiI2lPPQ9T1aX4KTWxnY+wN|%eTj#xva>>Q
zD6~lt3vNxhpT{JfLvUhQ;A>DmSCsdERD_-?hs(jq4gX=kI`TWFwhm$4maDFaN(TIh
z{H@H5z$s_KPh@Vyti$65;`SCP69Stsu{ke;_sRhXp)*SKHkBL%2B&y22zdvIqctjm
zbmiN;vEhmP^KRv2!twV_yCxcia)?KWWMo0S<b}ZfNtdTh#USCgl5?%5TrdfLmklh^
z>|~K?O#w+DtdK}cisP8*{G&N!VSi%a8!2Seywc4qbR4&?XJS7T$e7A}jrT)xirHZn
zNRQA*!;xziqt@7?L1mp>{0GS+5h)+wmm`ma$Q6kBpOQziRt0B&N%BaBQ=G>CDS2e8
zgSLU2xkl#NkkfzEri^Uq$rvAhRtYDDSy-;zl6rt+5YaLRwblJaeucpqw{wonP?M%r
zldA^g9Bl`HnZo5*c;3MT*r|1enR6EKT<*78A72LMo%Ci#?Q&)@2^h?(^D}^bd>J`$
zgP-kmCUkOEyUv&8{RY;U`SS|hEwM7?6Ql#PDkFNOysh9Hb|JB-dmKf7Zy{tRDCbb9
zO_=GgW#)_A*?4K|d1-QQEcC2rY8EDiUJsZ&X5SQcdMs6PDR%B%-e;(xqznD~lN!XB
zPgf_U%J)qXIP?9qXiOQf^eb}u0MXCSq|}6=Uy}1wEv<!WmZcl}V`+7z|5%Dl^}7YR
z{Tvty!KB&dE7%(EXRTL%B4XO0G8_u!%<L<g*5)4_DDQ&QLMuTej6r0EAo3jT2Nqzb
zp0;LfF2^Tz48FD^HE%ca>?g2KmF2W=<@H$5`#7@tku?=`SgO4fZoIT*72mxU#<$yw
zZ=2&q#TzV-Gke`nyyW}wPVl$I&X>9Zs%^vjj|Dq8Zm|@7i?fS=q-^w}Iw)%h@p_`f
z%x|QK0^Br7Ra_rZC(*AKg<+VcYq`;A%=?5(nqCU{lKU^~AU0m1qeT!8;csDK)TgB-
z+v_WDDWR|lbL_H7=&1W#B8a*PfXW`)Fq!rKI1QYmtzj=nTUl+8eJt+|qZ~$)!CICi
zO!nuX%ok_uYsp7{M+V3cI%=6x!}7#4yVwixq#7W}UcfUwb?`3ReTHdm1b@f<J|S@-
zxNhF`O9x4eeIZ%w`b~ML<@6+a+a<E})R)F^nQm2Fb?KY_bf{vC?|kqrw5C){aY6x$
z<{EEAr80UB^X9XV_BPd|AK*u{N{?F(@8!o<>rp?p(r)m7<N5x6G@xMIWOy$>wpx$+
zv6Xg%AD`2P!S<5J`dmy3eMfe+OFbkd9{4<h8H5WklSnJ|SX9xv84C2>aRxDcazuB`
zBT^!6&l?=a0(vQFEpMc)aB{Nu4jk2rT}T4IthZ!0(Od7xlO^*>z1LTl^s~M%F4z~=
zgm;cVLhBuWc418VDkB;x`ARH0e2iCOhxFjQI=N|8RdhS{kc$mFOY>mn9-8L&?Xi@(
zY?Rs%_Q$<+2*N`W5=GqGv%w`K>dVTRZl4>Z|8{)VSLgM9z9~83x^HGkAW{go03o%d
z{rtV(F&Zv@brcGZe8yPL*IR){(sZ5NR^apDye|ZQt|jR*eNAU^_&m6Oa_qbPGfiQS
zzWPn20Lb%#1^c7h5UG*DLC}02GVZ6m8P!}=ppBGr;HVR<RTxbZ-}&;ma9ns!?bBJh
zcBGjPu=ME1@qkPmd-<uv<_%goQVW&Q{iLI@*d(YMj>hEu@_PCYZ+?-43T19&b98cL
zVQmVRj5RC?0X3I#)&&y+Fg2H9Cod<LeqAgre@J!}YhOrt`0F4KUms5I4`0rT4puk~
z77y>=55N<WKSA=K(FY44jDo}akHc?IQljDK#1BcS6hD@KJO0+kM+J^ofaZs10gjKB
zAiTqC6sC8TN$)6Lg)jUK{QVrp`F$^q3=kL|_yFnn`};2sGH>?F*~_h+8cTv<Pdnw^
ze`&@_jEP?PVW`B|HDf2ntbG`Z7`~hxY(myr7@B#&&~%_#CLRD>iEV}B4g?z4VS7FY
z&VYSN_=qxGeZ;_9jmJ30+>X2w-UN1qudm&|xA%v?A0}j|gf{+QzzN0F|MBA?Ob{?`
z>ko*+108<;dHDYDYY#QcWBBGbc@3xoe+VfLKf?>n$hWL;k=fEPoelYDgw{+xk}&*z
zy%@_zCWyuhkk~_%tiB8^LA4p;IkS)W)#)RW(2RVmpnHaj!+zm4Jd685shAgbpzt~E
zK|}u5iLyYY8urNp=i{T1QTS$lv;0o8$g|{=nnt)hD_@pWWHPQ5rObMwofG6$f7!Dt
zm@O-(2?`kHaf-6w5%+YF)XK_%j%mh572h&aRuG}c^gCRDxn+E)jc~RH609w<Zzi)L
z>i8JMuuPa;5W%zx!`Kx>D(%^V#{D($HwjeMvPlnY3SL@U^_0Bw{RJ4_NnD&pbTjWV
zya{{sZCkw1?3}cPK_z8|1e<)&f654}v42~hw<AF@1ItcIxdvyT?zHXMm1Ug$%X=sZ
zg(X}<8H4Fd3OwO|38aL{GGjo7@_SqFB?<7??hPe-R%b5STSyd6x9!aae>v<e8qc<h
zpi{yawYNX?X+Lp~CXHiGOz@of8j_Wle*Z^!x`+hjIFPDAi11coda;)1e@Ai{6UfSm
zA>i=ilvXJa4Wr+_t}=jFtO{XQK)V)Rx%$?T?gb8VUeK_CoTdzBrJXQRPZ9AYI}EL7
zc}I4YIH>hBo1AC)SH~r7`nc4tshw3?tLnsxXP$F3rkpVpbGLJZjtSX0J^CfZUuEbi
zt$-ZU!*`mAJ9w^~9c(*(e`3L-?m2X2bW_G~#(_PQC}vamLR5a?Yf?y^Gmc2>G#eFz
zsyv0V5}g}fW*VdPOv8Uab1h<+-f8#KO3tk3JBu1cf#SwxKn4X<9QUPWX;3vIMIFfP
zR|~B1%DE2p9ZhcCPa5H^>S-R>oS8fOog$thW?QJAVszq#o7o)<e-Mhwat^x{E71uH
zb3GQIoGN2dlgezy&H@djux@6nJ#u>glx;teG$jD}yZnD$5^SXkOZo`z5+&KTgv|(L
zf&HTC$=XRIf}}CTwf3#(StkFOIO0nBYNmfaQ&hYaCMyl1EfveP<s*R+TqkzbBL_X6
z+b$kML6|>uuC$twe^M;Qnt!!eB+lYk4<?89+B+@QJLuxoO8LnJfKtAPc#lGn>-x||
z=pdt9E?Pa8GnV?*i-q7R6m>D1NPF66=`KGxP#_12%lUap)xXR07Hcf|_f2c$Tq?*r
zXv^h(Q%0Yv;+XTQ`6E)(jMQ>1JkwfWRd}8{{@g(-O0iDpf6tami}fRo81IC0s)x8Q
z8CALTxv}w4d2#*7err`;x3ou;YjkccI#CXRpAwyjK%J;gd{V)kGU>37lJDYENe^<M
zpIa20xhYqH2m|q<0nALz!U8${qU{`&i)76OP6zG3i*1V2DUxGrl=sdB^JhdiWw?x0
z=9BHd!+Hn1fA1h(X!qGz<n<93b!ckzXuIQtd9f2jx?f$~!Zm0)@OV9t3$hc@6%ef!
zWN?5;%am9j6Yiefx4TNEIg_)l;^5?hSMB?Qj+7Hh4Qk7g)+K&&3^X}G$`(S|UFg1G
zZ=B~|Z*z7uP*+Fo4pvHoi{US}Bht*h+7X@mw4dTWe~}c|D5iIxhAS5q{b#sOPRiI-
zyE_&K9mJI`%q_)12YIJB=vH6Y1HQ!rU~FLe%Q2M`u+-^@Dr^@r9*P~*_B>><@sT3}
z&TdBCJ-<f7u@T{u<7|iZ4so`FctPc}d4TwV5ZH4RJ`e&ypd4h>1@r45rZCw<{6LuI
zS{>2sf9!Eu?YaIK2)cpUtT4j6L6PL!&4h%Ep8Uw{<$ObdibrWZ{@e2n)NHo~XP!uD
zsVHleX)NR+O!YN+2=Bm2<#<?W!%>RO*GtU1T$<T9Y?&e+4AG&qS7X>-Iy~9F(gM!1
zgonEjv32U&^yY$7u?ed2LVLXAI?FDmm257YfBFz~nEvIfETEA6>i_8~3#f41a;^Br
zYb+!j*&f;j-hC5Sb5*1Y(d`8i^;V6zK*92;r8Gk_Js-Mk0+Kx>O+^4wb@<0QH#*hj
z-00$g5mYBV=jt%Pm0ses@^&La1wyf}@AmisV+j#XS=&Qw?lP<$b8|91@@u7Mz1nz@
ze}>D&5z2FYGh@3Egqj^X3=~XSwT+)Zcp5mGC6*-roZYCb@)et0Y|umWva8gRoNM<o
zX@M;|qjK#yYfpsze7{q89rCh4vT9dP=j8bUKe1;`Tr-esZcmIU=2R9oC1kml5Lvnz
zS=ir)xILn})h58iPH_v6%Dn)v%N<WsWTNd6;(2l1+k~3VnCzN%iz{~%EV3`J&K>&(
z?am#qZMJEFoNBZv$nRVeWo0tZzvLEL#gn!-C%=omiVZof_d(=*HeB^WDtn7ivOwY-
zEc9$lr*_HiJyfYc7oIu%{{z|54wunP3KIi1F*cXM7AF&zbWJTNm(XP_n}5({_kX_r
zwv)_k<qWhiHVoFGAX<g9NEL}Q^!o=0Kz9tCX^2M?pB(u&orlC7p2{kulZ@vlwqKn%
z_;B)S2_ukR1oRyz9(1J2JD1S~kirJr%Yc!OP&mwd#lI6z@tLi%b<!$*i2RzkCO)mw
zm7RvA3KNRuH$9gL5BJDVBYzg|+*uKu{X^934@DPs5E7Tbb$AtEG)5R@wgWf=8`mT;
zpn?>+n3<*pL9EtOZ&ZT13`nJpC=`Y}5h0%xfVSH$iKWcTF*<NE0F$N8?wnL?_MnPG
zh0Jr2DVj;%S1~yiYH@#B8!o$%Ho`i!B-b@JLMJTrJ2R%X%2RZ0vwz{<D%V6G#$VJ2
zSOIDbM8({vnr6+9wnluQXR9*Dydyn{D-M@6S<a>K10HK=zK*ogif}UhPAqmP_YYR4
zkS*gCe?`kzc$ptdVbDTPHHH+njOt6t>5B#?9>#Yhui+<AFWJ5dQVGMSG0?=7DZH>I
ziw}<$Nr)6nWM7Z|i+_iD_i}(Llq9d>;L3Xq`|a~`q9VKY4hQ442+lGt2WgtsXLW@}
zpm*Kw#4pjHGxHVGo;*%}7af|X3F_?AS)b-5?^7q<V<za(Ju5p7&*;6#NBWl<*UH!I
zX0QM6$!I8|2fbFtX7?mwznh%bVweFZoHsRwvAw6b>E0oh8Gjk*A*_gH-Ohzrrf!a7
zC@2^86#tK7>;vejroy#;-tMlpZ#w%|YP|kspGO;nm1}*z-QC2ssQDsSY^1Q3?JN<F
zI@DoOb-Gl+BoNFfE<H&`%JAF@sAA(N=F5(YPZiEZpw$lm+Gs%~^ID*r-Lv2dhdaQQ
z6L&NT=mg8*Cw~C@T7*RiLK|Hpd~@C&QnMB#m^OV(OLtPVDsXXZL0@8h-8Zw=ME!F0
zF9j;-`6Fa5R{omu)m*A+^O`w=BBPx|UgVgVd2W<mG=7n1t*i)hp5bz?(?I#fEwbDx
zAb0aox#{v#fr{xw3yV!mT2azLZXzGhCDFq^3z(Ha`+sp!wdWz?<cD!8V@fjdQ{b9p
z(sAB$ykJ^Oturbobnvx3RbykBBz_)ERy~%6F12|lUinm4SpCSmTvxUzHfTTMmk`R7
z{8+-&T{R*5d8?~1RTmL>h(~HW=Kq?9+1okeVpBl?!Nvks-$G`w8>?_0_uRqiEriWR
z#v>is`G0k-8^$R?sX}6iw2!0tvY`b=<CeNphak|_x-*UvV@yqFO$|I7i=;j?W`+3m
zvHSJ%y8C-(O%8?idid+xF5X8K2Mu(n0^!LMzkeRS?0)a5a@vtX(w_Wb<naCL(B1jO
z^V1+nKxwTpt$<b-l3hD1#=07xCwgH0fLcm-bAN)y3avv&Fo6V<vBimj!pLR$NX$2P
zK26FyuhOKqB@A-877S8oO4;pVJ>w$h-jo7gn4{}(Ud7U&2paAHQ<A=kywZb;2c=V+
zJD+X<ygfD67fY>BDxxQ*Is=YO5oR)^5_k!{ro@m(#*os+_>rV(?o72>FvB=LsAV@N
zwtst!{Hj)fnp<PHH1UpNZ58gfZ7RZ*gq~t9%X&)*MQ$)`DbqWbkZ?{|rTSAKj9IA^
z80o!+y9o`DF%!;sueNYI3P5phhUysy2P#rf&Q`C$9>7ywjx%lc9(o5w4ODi`EKZq@
z;tDTO6b-`onjn<2ytHd2B~HiqXp4_tYk$5cCH}EgK$D?FdA>L{7&X43&w?kvaA8BA
zZ7DYG?i#;!nrdX*^SQO(a^~w!d8^;XH>Nn_eU0A&B`P!LPp<bM_Q@Vs^9G-EL`l!2
zT^=O?_3V~l>sD6_x!-fac>&P9Gh(;A?!53*I&XA230pdELvc1jS>?bvC3d`S?tj2m
zP#fMI@@D^y*|MX#%6oGr?YWb0&VN0qhKEl2&3Q2TCsm9Kua<vtm!f_V!M<~6(Vdz)
z3z#<blfl$DNITbAbkuI>EXrBHu6}>-2-MA}nIQb0WfuI67l>xY=xJI;<zFx@A(}fi
zF0F5gfZBhw<NJEm4s`0Y#*a;&zJI6|c8ib(#+BU^>}D8mZ@aN2HyGa8cDqxH?WQjH
zxg7$J|Kf7Se!cKs_Y|+&Z<^L0$2lcOjUMqs>7n0bG*qn!GZrjo57}P5hBB$!GTV0)
z=d!HM_8sLjiK(9@b-Ui8B(%b*I$BrmOk&(XE0o+_*mADdQCu>8{jz$v*FjM~_og@3
zQ#~u4fuLNkAf<)Zzf9;2it4c~yXcnS#x~Frzia{>BQpz_hrAWm;M2(c-jKRDtna>G
zUU&Zhkwz4k(M$>x0yZ(1vEeuqm%JD-Dwl9=ES3S>w+C)4Z3cgmVhSSWdcOoDgR*0>
zv?+mRb%`ad1s2||+S^YiE0laiOCH)&_7GX!Ag8V94Hy*KNr!mGQ^@CTWDg@GBlQzb
zH`=FOc4K#KAy3rjE75QvM}}=V*}2bl&$FLfg$DW}#SKQbG=STvK{v(ZAW6AdQ~%Y_
zS<Io?{4o2jrKv7~)SRnLC){n@9u#<cCvDB<htme<57K61+01v+)@**f-1G)*_5Jq#
z^be=qmzS}oB@+TRF_%$@Hxrk>*a9h+b9F2%0^u{4p>-@ce{-jg<Q$PchW@tnRJRut
zmU;+9A9^dq?ZrspX{ZlEq(vQRkb1#`o`RmQzMbeD*W8al;|b@Vw|{;9{TENn=4as$
z6|+WOw2_R)OdQ_1YN-DBuwqp{shG5;qvk*|DVkf71d4c+kNCIre}B-2c%wNuedUT|
zk|wtKcl_h_f1<gUSfKw4{9n`#JbzR7^>(i>{s??;t_L7f+Fi8bOe>lB4SWjEhu6Qk
zMov9v%H}up)OmHJ&NQVZNuV%2%V_>>7^(DCzVF@lE7Xr=mY^%0M$z>M(q-;eM$mbd
z^j21*`jHB<h(_q3#mFta0l4P*;wMn%(dfH)YtS%Ye+LYM)6l7jzY96sQ*NP$_Q`u3
znFffv?2SHNZeB)cXnfy-EM=$2;>fHm@?c)ch@T_aR8}*v#a7_sCJ-`+Yys#9L0VW-
zEo@L%2Gb+Kg4)!hz(_BE&AkX@@!ew^Y^n%VLs~)k_L?a2?DuWx&lkJMrj}*Sz?83t
zz^ODaf9uYov!(tfcrFGkZqfXW29GBG;SGO$mJMy=jCuAj;@um`*w;U+4Ir!5$6>Rm
z&@m;7Sp;`Om64|k9(ZX1am}k_YBT0ak3tbNW@^(?##=XzkH-9Yq&`~n=5g*NPCW`Y
zNk1m8GXrHUTsV{Tv8eF!Ut0OG^&bmzz&a+Kx7BwnP62=VK;!AzjGDR8GlCgZJxAZ;
zKRuxtBtAGF_=s3HtvsWTIz<VE_J*3ek!8hOmQu}mlxP_Z&=2$;p&hp8BU<e2RK<w1
zce}C;TwI`X@^9L8<xgFG)~+|ymAiV(uD8@4l62ZXVb`HGR@S$`0lS`hCA+?S>qe=9
zoYazH@c4gCcFws!MT>!sX7U9PSIlWe-}DO(d-_zZJZ6Ks7TL65{wA-(JGDLsY1`Tr
z6PmdDq6xK%36ujue!zrELB?8an9ypCj+)SRyx+luwj6zv2|ccvVnSO!zSD%ZBX94l
zxfzwnPz@{i*Vs|e3`)H9v>k1zL(V90%B#quv4MYcBwNb}t^KIcZ}w45aUF9qh|3&J
zxFtXhSn56E(6gj>YzkO8?ggwN$|yZ$8OhVgh@c^w34QC-L6+aAPZE93ipIWWSKMC2
zbt^zm$mKfq5GfGPRyV%sWSqN;icmc0J5Vk;)+utWVcdHF-3U1a5~VA$%08NjTCI{U
zK>~k1<+@-`NFS{f`&n6*;=&{h`mAZrmIl}Bqx~-D&OCs3tH&9k($9t^M9GAd&+^)5
zcr(xPZEWA+9f4f5QCo`*jCqCK<V~9*l0$lHd&lRVTl=BACD>1`rsV0|uA#HcYH`x8
zwbb#Z%inw*^A9^}Z@(Z|a*+Nneb7@38ghThvr8vIl_)n=W=FT_$=EyjiBo)PU>XK-
zzh=gf4o7m{`cc<%>q)Ew8OP!}2ldHCkt>Ec=XR4CHh`Dtvuhmt?l|DVZP!mISUSxN
zm*1@60#mD*<9c!#cto>${VpWlT$FcPsYthazrvFo0i*)&P^CGu8+t6Yi<+LW#NmJM
zaSM!2QIcDl?@Q?->DdXrdaGdTn}wOHW{Rt(pHI@twcCotZB2)#ZPLJx`i>%}h{FgU
z!yTNi4C*|QiU3R@o;9cqb;MRRzl8oEJme^Lqk2QD6pojd6fpLoGTPvpv~%WGthJud
z>yIicdhMG+llM-TsW|?kvKy*CP78led#T(NKFb04qQi<qKEmVNx^t)cYjUhE5C;7$
zGOTE8&g)u`OR*;fr;Fp$>pK*DR1RQj3HsH`ePMQv)7!lu(&OBSYZdn1s}#rUlEUbE
z*AnYPiT2oQSRMaHtnFXr`dett51u4K#C_Ombblu$_qR&X?6{250z>15nTCH=Dq&{6
zDc8I&2`LWCA(tNLl+;jVflJC~Y7NMVaZdNtyM-s=CiaHB?2Wmr#2w>C^|R#r+-8KW
zFv2irag=gucfle9IGgBfsh0#n4n9mOA@PBQTQ1qMoeE`GbEg-4ftVg<ib*D%)N1HS
zt%kYQ^#aL$+U6}TxNod+?a6;kZ+mDpEEj7U$|r|}b&ZAgs+u8jHop%N40ykq?Vc)r
zO|z$t)S683qlq1GXYtceA1T&4a&C8{ehn+fi^wu<@X)Y$p(z=feTEYkqy<yLF+=fG
zCtsJShI~pl38pWX$F-7e)L4)|OMLaK>b_Pfqc<yM_N|)&CK1k}v!Q?9+`CG(B?D_`
zWU+$bmim^R2l<gpoT@s!G~|J84~810?ime$cVcCEv%>S-sCb{#Yup*%qjBAo=p76N
zKXdz!j)&@yPPXCtf5%zF(e2+W3GnxgR@;8&66Du`bXun>=zLLeyW>rk>~UQ^8aL!2
zcSU*)n6JaH+-)p%oArP3qW5v8e41h4s4cZD+n2W?j)FRjFNbS#mp3=MSgL)~Chx;%
z5QcGA936xe^p@~h9=qv~RZx94e9i@Iwz^~5C<x7U%}Q76m9h0&rJc->26<GDgXt@i
zn}XBa^WM0oqU!&sOmbUYeI!#l5A20QX3i!`8{)P`E8_>&Ytnz`Zf7vSd!8qOoya>L
z7H6U>GS0iHl%ys+i*AivUz4`yND5?{!lagsSScGewP-fBUzoX<gpxM1?sZ$;TK36}
zodV^`MP1LfpDk@3R=Z9e4(mO41pF;FY4R)#8|pYOc6RLNao95P{{v#DjQjde0|F!t
zBC#Fu_k-fN;9q}%qR%IYvk`Yoy-`)w`<un#o8pWE0|EUEKZ3I=>wm&QLT<?oQ9Y*z
z%V&L7S1a99@1kzzB!A7&H5L6Eb#iqr{4!%zdb5W*X_yjc<Mft#TH<vId73$N{VcyD
z^2i3#KIu6<Sm5V#LTLN7_<fMKU0(EM^dT?-XXEgey3v19M%(ZUp(O8cE1AB>Y)p*9
z4<5fQ6F&-0`y43I!wLp9ZTy(3ir$Yz!(|Bt9oC@EE@4Bx8&^nJO*riukam_T@n_f4
zxUw;=@D}?8DB<dl2RCaXmMgPV6Zs9vK14=v?PUTx@*>R~ADTuIgLd}oC68rG=rE>n
z`qBS_q~Cwu)jq{HC`GThv1s?uUA<AE=5+JdSM6_!(v`ySd!l*fAgpIDHhHve?UI3L
zjs=<ah+cPi57i#_HCOwX*hCsR^k{9Tta3ZIaVjbdWvLxed%d=HdB4GfCFc;I<-sYY
z^}xfBJuy$lrR9Guz#&hPpYNZ){s&ARNfHWWZe*9mhb#~RI5C%@+$=Ab0hcW&myNA0
z2!F0$u>h?=QolgIqy@aUJ1;xCMPR^*>oi3I7`6)}6@)^fiNvWa3KA8#|Gv*GMTrb4
zixlZv4iI_nZf|#XcAjTmjwKm|6iTuZGP4U7aw#s<K&iUGfz-!T4P5<EOF<g>9SWMe
zz`;aau+Xd>K&T<7Y(XnV=cLUawvcf%Xv+*qe7%1vkgC7hAhDQB&_UFnwhYO-M9?es
z)uF1fVU31a%%ECMc-s7FP7P{zotkezvP-x}CjK-965mx$5~v8dq>yT#r35nS_Y6{z
z14x*q&SuV#T)&KMQNxx?rX1L}g?lM*b`#``du06{1=j52@Q2InsRTB24KmExGGUF}
zf+T-&{Ro!rN6?_S_yasGkV^Ka7KF*?V>}h|r??bRpN}Mn`mtovAeCHa43VZdkIa5%
zFf7gshnBd%TOd`RnKdY7-y0t2^%@L`T`R_GT6_(zL5mwg81EM8;{;Mwtbr|PC7(|)
zOCz685+rq9{3vipE<s5qpIPugXHPL$TfcvYg$nn3n7(vdDIV19=i!dj$Y*h%Y~uUS
z94vK<D+Oae{ftRXx=qy}&1{;2X}2j1W(u3Co5@P@eZb*x<i3w00Y~KO0tr^FOL*Fn
zzgt?Ush=fWs`}!%N04UYEL2XuE-tHfI|E}?&0Q3iwe)?!oVE0wW1(vO+<=8zo9}<u
zG}Ja*qlA|F%-8}p-e<0%aX)=*x=C^E;f_tKTO6+33hBYuUk`8odp;hXot;gl!|SV)
zsbBr@?(9Ed_-1nPc6@P|fc<nd{C)U-`1UZ*E?$jC(?hsaG$;$BM8R^JT6FbbhL}n_
z8lDAA|9bdt@@6tT4`0Czv3jcoE{1>eG;H$EPw}5C@%#`RBAEQu@YB!#3fPOmN6xNJ
zPmk6P0Qe$I1JdE>h~Ho&x2UQMBg63R<aBcJ`uun_4(u<4;pOpkI=(myG`sxUhv|2(
zr^nNA$g>+SCTCNu@dEywotAu+7dRXMi)R<kWB|<U0xpytU$}Vsba;6&8NGiVPY;Ke
z&tD90#vi8F8ut@;dHl;53Y|^IXVXjPywaz5HNKo&U5v(;GaB?;KaStNJN{<!p`Q!@
z3ZJ7Rynlf=or?Q%vlG3X5piGs;IMPCZ=-WDgrnv2Uw`D-PsrCt4qk`~s+CglFE&cC
zQ4L}m;eSU5OCB^lyP177eK&tOgDd>=)%X7YSHDiD=a+vShF{)If4w@1qsjZ>x2yN(
zPu`AyAO3Om<IDKl>2Pv7IT^}$1UIo0;gqzV5ar|1XiTlWeLG4gl~h5Vqnw7zi_!4@
zcyuwj#Jhj`0dOi$@K+05>|nvnTW%Eo92*E@zSG=tMAUgiJLRPI1b=_cZg=VsK)Aex
zKmi2GI!G;zngK;rV@1^5V5H>e{&)XK97puvy1_`b*y<;N=@K%RFe{!xP`SPkzp$MF
z+;j!NjUNPHy9IE|*8py`2EbbqxI49~=P3rl_RxcW5ndVoLRD7?oEf21jNoil)QqXz
z5ctkY6pEf@&2_LGnd*O5D!50rkHnC@Insht-0f9cx=T*{M|bgLa(j32<o4EG>fBx0
zOz`A(7vk5fae)pMw-Y@k;z!C@L5E1L*Af4=8BS|aGU8jp?mcnU8Z&#U4QDl5&2W^Z
z7DJw@rJ1;o<~0b}BK(j=uR?CZ4|}?!k15cWq6oC5rKRxW#nylQ;b(c&EC48MU5qvu
zDYeG3(VQr#B7tM>ABh7;g8IdfMy@NQk;_9!gYpe&c$GTiik#;H%>|wdG8c6BI3%UT
zUQ&}pMP{nR(Yz#P99v!kGcHLmfNDh|T2V+e+X-f#$XHv*XQf6I?~e&)A`F;Gp*_%=
zZxi8$z~$Z#A((%?k+(lI3a^HJO;|R!#)5(wfwytpJ?yijh#K%l3CSoTOSw}CZ-BH?
z4JAyHR=FAPu7u<gQHtIu;o;0%RCg(nJFj~`z?(~o%DV0&0}_ei9Vw30yI<D2QxYr!
zg;tbk-Vp&T5NfJ7GoZv`3gnHakCxGk18%64YV<bA1|xrADh6|X>N5Iny1Xk{-<CUR
zl{@J{+{p{vNppgvIYH7M2$E}y?EO#oQz6vbrGurwPp!$CJOs{!N(v-Xw52Ry)SU<t
z6ZQISPLN(9n3cywkZjorvUAg}1c|w#XI)4zU^<{f!I!W~tGLbE#UsZN*(S*lt)h*p
z^~7_7k;Z?$8TYwxxh?v5rM=}n;)kH`w%qY<^iEwZWZP73y<CuOTX|I3HUh9`+liL5
zZPt4@pLH#hdmnIvkwCv!A^+E&%)>h`{)Od_z9kRoy!aQ28{d+7RxI<ZSZFtIs~QX6
z*3!~YRPtkLEoP)!@V1us!Kva+C9}YW`c9GX6@-73f>2T^lB`Mcb3*=+1dNJkZ-rQO
zrx@0-{z4-cWu`r<&P6WVv0Zo>+aj9;4_W=<A+WS9wNmdKsTIM4zZih)D!}zY0M`Y8
z>m1-Z2e|H`$12F`895+4Fc$yNnxc56QIXuQ_&jMTC~|58&#P)evayrPQ-%k$8=prJ
zhq8Z`9u>nR_jw?E9}J`L=H@M&nk6U-O!D#|wJ7WO#hb;fwNSEzY7`WsOG%sF^Eq;H
zz#K>goZeQxova;+6mw0*wPs}Yo|J4dXRiL`CVw$7E~_vuTfq3%7Ue#QeB8OPrByrB
z4eXGOm+5Tw*?UH3W!|$Zwub_gK_vbI1Id4={xdRT1PuRKMvh?KH3bohA=HbeTv$|U
zUE3$wqW)|goB#Z1o+}|E6|>LHWM`-6{~muhzvx`nnm58zUH%U1zXbU>jVuh91DPxA
zN_CTw#iCUE<PGb)e&VpLtIvLbu;l7BJ;c@Pf~%WeySkORU~{3)g|<5iP~S&Q6ySf=
z5v4xr#oWD8rNA4@3JfmC-3|ZZJ)7OzguqiwV12O~p-Eu!ZiIgZF<@e~Pl(I|!!jul
z2&(1M_Rulq?7%O9ibez`m_VZc#>|zocDDs2?W1R0o1;FEw5V?XGW$A4iu+vF(2xSJ
zP*o5=6TtwM_Mip0&_7gXjupyyU4t5OCi^!F`iKQ;p{4Yq=Fc6o?!fN<0Ico2@0Wo!
zEEEAbm%&si6qmSwEW>|#{^RY}C!GhYRj~1Eyw$%xzdil^Gz=Y3u??L?tw2Nf=l7@h
zfD9|^LPtS)gWZq6y01^a)u@9kEwA(+th*oIx~k<%R$kDeJZ&xWOnG1xv+t}83ge?*
zdGx9^%jD=HmrR~SO<-n|{CM%^Dk58j7-~e(qGnMoiApIU_h^5L;yBAk8vd+BhCd*C
zlMG!UgPx6R=pD2|M|E#(#vu*jCR%1`KIbvrR_RTBl2e3kg+8^IT;RptCJIVYXDwgu
z(D_^{$YeS6(wIld#CarsL35-mG;@er;%nNwJb$n{E+0Ovwke=f#$hvZ)7W(ts;r!V
zwnjC0b-uu)Orw81ea@3%?8c8IcGfa=&|-YFiku%@|MVU^Y~>6Vn@76)_c((CE9dDz
z&ZrO%<ctdW24_^gAH*5#P{wK>%o&x`ot#mj-NPB>7kuiU#lHN~_e~>!UV3VeZ6;!^
zXPN+tngQ-08Bvr9zER);$)IKhgpGo>vkY5MLA#&JGUR_zi9h_CECZlGWF5#dbHpQA
zW{!M|W#+vf#WLcy5RT0BD#Q(0fU%@Jxf-cioVVF-K$K;MY#YRF>;|#4pa$(WyCvKx
z%jTz$`HT8Di@Y^F=Dh;2H+AK(*G&3Ih_jX%Gg^5@&Fna{GcOQYP!t+}n#}#@aXCs(
zA6e;^4px64IwySB%w~zq<9Y{d=DwZG$xD@jb`_Z~7Q}_SS=PY7E>YvrM^D8Xu?%*f
zk`Ef2BXRLd#<CuI-!-p!W-G%7#Hi<+@Bxw_8^aHYA0)N4qa(A_3h}`BQ6b-mA64%M
z#ShY6eE-o|Y9)1NmRh0Rm!;z8C}%xilTie&nZbXa>Q+{fsH;LWw7Xue5ns0>_BC}4
zXcr1_uZXyDExlgZ)~X;3Tg$yHE+%k6In$UYI+QL8S%@qz5#a2;)0TG@O)ffW+n@W$
z$L=oPDWya0wL{eu89H%$UJbk<qGC|s!VNDSdXtm-xKUaf$1sL|T_r`HACwnR+)cs*
zml}V?rPk7qxvmz9Wl_MZzO+3xz!3Mh$4#slj`c`VNrm}$cB-ss47pc!kEGnF<j?up
zB6;?^-sxTGC$G0uuwot(rn~`3nq+P2$*x#Wsc{FM^nP^5Do|+9`51I5P(llCuYm85
zut-vKJA812t)%XZuoc=J5%#=0!qzj`A7Oth!~-I1jkr6))`0Iv*xI%9jF@fa2s4Hg
z%AqQk46DP6LFE{3m}cl5H$-fsc5oAR%Y}42)#DsqVjGmOQwr``$SH4a2#lniLJD`z
zAQm{jUBWRnPv?s3otwbRf@)pST$hiayldPHU&l}zEsQW)wUh{ZB9urqhQ?)*Bb0y3
z_R*F7&ZD0t9~<nn)zWJtns|oE1J#$P%<-~%aGIAaG*!*R4s}@&WvHiV7S=o2yk_1K
zcEo(_WVwP-W^&SbQ=w(-{mjb9pxT9s^PIW>=Hwu{LX2s}6}F7LxP+156UG#bhMcaD
zP03gDeR2js<=cI!4*d609SD+5k4JxrCAJ-u>ZBdO$?@X(&{PMU5S|GQ`p?OAfLRF7
zJUG{>r0&dhDzrOto!8yDPCbMDxlV<6K(13G?#^{;!1r^V+O_ocj395y7vkUv#lIY~
zVP8t?TP7>3Mh|X1Yk6lem2$kEaVYV%%=f<ShP=66yY=?`0EomQ_`k1U0a1U^<PQ&!
zsqy6);+5=|fgAMGG3dr9p8UC>pr5axU*C-vUPdp?$kyAm2c{v*zV>9K;l%?ABQ-nY
zZ1<?_|8Q^g#%0q2So{m15$3V9;<PVHO%fJ|T9DBw7`GQ7$ElRqW1ms2lD2Si<3)<~
zm9&?OEoU`RUc$7g9D_t<@SK01m6BlX8I43S?Xfwc0Tm>rfB%H0zTXj=(}~SObk1}k
zx)fIm8?4wwXrVGAzE~mNac)~=y^Ppv;uYwWw^wq<I^%la!l?6R{sZxDIRmt$5XJ)~
z>RlyGI+ZEKv64En8n~1h!y#b#$Lt+uC_NKf55L+YvU2!HJ>TbkCE8Nv=&Ud0NgtD^
z=>0+F&|+sXy1W%8h0keTDCHhhk2>S_DwkZsLo}~k9t+ntt?zIh265_OYeTHdIG3zU
z6#w#EgYyDO_;%l*+=Uqa0oNNC!IyzGEE57UFqd(S7!v_Bm(i#(CzlWnGX$4vkt~^i
zUdQ`?KL7n7iP%D!Hp2Nd`CtXWV#*Mz1<Gs$NG8DwppXvmT6_X9?RoJ8Rvb{{0<I$E
z)-6tV@RWG~SkASlaG$lfdapeGeY{78GiW&@b!rAB5xKQ|Kp|N@IrdCwK`_qD-y4Dw
zBb#^N-<jkyVe*9}&sfSzVJ#8`Stu-jm`aImPwk^Ce>)LR3p(79sI~LCe0G6vK9{d~
zb-i8-88u(8wRzX4b5uihHB!XjSEn3dJQiGcUg8}o#V=gZpX+S_bu*?7!$?*{?S`1n
zq;M$zrwn_@nYBXMG*ny_NbdQLd{!-;bxN0~w36u!q-~&p`I#d5=Wif0GRm@luf4te
zB;(LkPcvS=P5wL<+zVRcUac%N<g=E{XI?$hzQR?ZK5HkLonp-?QvGyxad<y0W$lGl
z#XX>ta*SOrB$1trNFMzRethY^Jv|?OUtJ=|On7Cc@2>}ZV3iX-q$wHU;cEZ**YxA?
zwTDWMObQ7{<~Jv&KVGJ8<r{f_<|KhZ>AiDH2EA~=yO)4VRt0>l#>y9=)xv|4@vI>_
zFTfsqhy(vQi^?LM{FcdItbA*HbyPWubda?id-f!S3AW#DR&>AuT%`mq)f3TlS)`Zb
z>3~)r?G^F_4F~`pv_K)EjCHZ{?Xt`}tlI1{sj>8;h|%;AQBrOs!E-BrIY+{(lH-*(
zB&UrQNv$QfcDlQ?Fc@K+pTJLu01lS5@L}mhd6!Efcwcs5F7br;uGdl0x(;Heqt`*C
zn^;0r_T3OQScHOHr-%6X0f<4Eq*oB#P`5N@fp#TzE3cNe2EVD`;(KYU4@taeV#kdc
zb7O)4J?eL3iZ&+)*=Bry4Tw4G3)k}+oUmSB(5$jM8j*rp$@qkxZfpY<^h}lz=N_2{
zOlP#K0sQuyu>_YCeB=f#1W<4c9mK60v`Hkhl0fC&a@y(17jY6{IW@nUXYkA-#dMKp
z5KE2#`7J|StSlfVVN+aj0R$Sn%dssQt%de0Ny*cP{dtK=aJ$ie#-f4C#Q35g_em&-
z-W`-epF=_LTc&TZ()v-3%<vgwT12#mR;$g0h-o+?r8!D)w7@zQ{x^D4p1Ao*eVXw+
zRzAvok5W8H8T*9MIwy?H3FGe*Z2U6i-2lL(=%)Wpd08m!Ab-T%Qb9QbU=933F&#er
z_VVLz-+zI`VtV<14{n95Q)malW_qQsSdqk(rl0iP3TuMFWF$d_Z6^!&Yb9*pgJt3S
zbg<y#EhtyB?I*4p5h7$l&0)-Sqt&J&XqK3U46&>dzMt)4p{k(BL>Dsf8$zM8_!Z9c
zDg27B_B(HdQ~#b0Kc>T1lB)AM;1<LZ^JT%mkyHUU_%+pkN|u*E38l8m&P>1t54Sp8
zTyV=9foX41QV^+9fKs8LDh%^V!!$1eD2BjGg|2RWLkJzCuu{!lxHK6fwAQGeiX=Nr
zxBmV8m5#l@p)bwn4CYQ5FnoY?XJZ73JA{?%n2kb>w)1Yz_7xSRp?GP=G_tM^2%!8r
zH6&%$?kaD8U-yawF2y;wLma((Zp7fhZQ(_{zS(GF4rn)rm8W}~*!iBQm!juNKiBP{
zR{6F5$CP{2eX2cyRqIPFT3?rY0^Gl;)%&Tb?WxB?Pfek&b}UreMW_L~QkdO)a{`?<
zbqJa+#M}>w!=42EHSZhO`;FX?@60LT$R*{nTp1~UTfNwUvQ^EQmF?I34n8PKgT4nc
zw}_tICkT>r{2M*_)~Xc6kh>;`WR==zhH=P&2n^n(<cEB%ks_t9c~FVzIWySo2BWWX
zwI$cI5j6>uBXFSlFTUNjBe*6(eIsiXOtm*w{@7(!>Pl=IrUrGY_OZiuJQvl|>E0P`
z;3_hI1a5qiL9lvU+b}ZV_@#_}!w6I<xC{NF?z7(fW-r5d4VlDs7loF$EHf-n4#XL!
zgz`6S3waBvVdx>N$5lz&mpt7GeQeUx*l^{?Z6o(I_glRz4UiSLRthM$3kFQauP>04
z!#v3L&;}f$xpR)L9W{~KmA4Xh56&Z*9t(1R=(txOcQ#)Frkp(__w??TU(fk@`);WL
z({mwdY;=7~hubAR4QgCWQY&O|^CK(a_S-wk<(>3cNagg3dd=~QoDxq~*hy{Ci@Hv2
z(v`03B($xKI|>jb>ZWQ<`Lm7Rl%n-|c8yf1snk(%V;kXyQ-|dpHd1bJ2`T#&k8-|$
zUINYvm3n|%{UPw_KBW;Oy({A=prX*xH5E8lxyL2}qttJ%aM?l1bH~&uFm1+J3VAxV
zQApeSI@bwrrv+dEaJUEH=-P@Ce>>WH$bNKBMa>;oLtZW0X(s2#c{N)_Dtop)1Mlx=
zUGuqiE?Xc~+^)%oZD?UFDaK4#HIw9jd?wZUV9xW)iCA~>3^2cVoE>_Mi~PU$Hm9zt
zm`}8;h1QN&#BFJ)a;jsFGS*O~6}eKR05)7p#5J$YsK)jwTd5jYx{Il1{9Jj>8Nz8;
z1znQD3}=soXB%)WromKI69nc&(&uBTg_8QbIbqh2eP~k$XslDq+UHzL0`|9m;7cRe
z9r!71mKbi(tyb$`YhZR7VvFi&aMXh1?2#+33#t4iSEIERQkx#z&Ka>bR+v9)LabB}
z0g{j|;fKOQc&G#ZiV<;+ur&bOU_{(Oyy1wrgM48`-0l90M#L)OXeQiHo;}3fmD=V|
zms`~z$ew8w)6uq>t<IyEZuc#Jv*!SiRoF`pjnR5~c6sybheD_s%I&1h7%9Svk^GxD
z`hDCSK$|Y8>PY8i7$|_@U`k4(-Tdpbk-nU6X9FI`I_Z`+&_R4>w}LtZ2(`|N-uUIi
z>L93ns=svnfB~uKG2`q2Zs-Jiv~Kq;?T0F$ee_^NwKA>f`E%??doL1yBosSo3s4t8
z?Ch^fHC=C4%bKC#xF!o^11qx52GokXFPvFdAaO`R|Gx@ncF;>IR+H!t6wPyC+9P0j
zF0!NY6P=6oGkfMG)Y4=60J#wgUA?*%otocdQeBXLSiEnLjpZ{`nLSi0J6B!|Gf`z$
z<L+^Kcl|KnQnNoUoal}WUVr}yN$mdsQ5{2bmw`1b6PFREC=>!QH<xjSCntaHS;>y$
zHW0o0EA#`yVUfiRFbs^HaRa%f(YGL{ATy^qWd8q<M{%)4Nw(zH^eG2fZkn~Qs`zSY
z*?sS1_i{MD9X?+*cTRXiPIYhJI>HYUpCqwpP3MS&)zrQH=>9xPsVtr|9+pyTJPzMI
z-sbd4lgAD+c<?LH>62rMG}wQmRif24(P->a7rY7YFL538L0TC(fFA`s=hI(rzaL~)
z_VYE!dzxCu0=$c+3?10z2htgm4c-P4P5h2^WAeUBUU$93&sR@(B^e_u-)!<5D5DKz
zvI)^UrO$f=)PkIAq<^_a%aMzEgp<+7q)X(-Sx(PC%Z68an~GqgS>u1}rTOdY+u`3s
zJ+jim*w9(>Lc=P5|2V`Sn4p6Tol+tYcR&B>z8(H(xn{q$?>^{y151muj0~)-FvfRQ
zv9Nf*%a!m30u{^s-k#Q<Vm$IS+khdpe*FC2H7l<#)eI`$0=BJU^jmyJQCcY!yoJ*>
z6fxGh_{V3pFX>2nq-uZE6ARbVCo6lc>!G)6ACzFfL3wB*f)5krqkYYc&GX-CtzYV)
zOuj17OYqLtgR#^?2D9O0933gSrO-odv!?uIyFS_p(i5}Ekt$L+4NA%%J;Tw;kRcPU
z-Q<8Ns^<^LLHIDyL0;zWeMQ!V9FQ-+@m?@^)-X8rc$2PVK_7qSk^%NW+8~7`Y9%*!
zU>PZ$P=2C!uZOxTSCJFex|VW!Giv}(f4QrfQ-bXDT%1F$S~J?M$A_r1t(LvCE8`0^
zz;`48?cr!`thadEAXSRB*<W5?mVHs8z=76QV-{Bw5VN-%yfR@Mjv=}P@_m%#cSOQ^
zEqy+*d6Wzi*l&MJ?FJc08QzlGw<&rmu=i{gz)mp0e5~jPRRN5!4S_$X3gBRba{I?e
za>1&4;kjbLu%OVWf3H-q3YV}Ycu);Q??!cXJj!MjwG6-?|5(Qe51^Kjg))ihXyvuu
z^NAmUP}pW0A!<hR3K;VNp|ugnIt`<Dr8G7{khE)4+T(v<+nD@{gzVfN?rq{zP%Bni
zJ^$P8SbJTnPLcN9qi$X)XjL>Eq-3AYH!!?X$D8b4{-sH-^GK-?tzg<(Fa_jY=+bD=
zl$}tnNXy!DUw*fg+mbI!kFr3uTrwC=A_S!uxmyWPtbIx8*Q9G0?g4GwrH(@yHPAYP
zQCao3q+fry#xf(ziQWu#@4)j}!jAqW4eL;Q{x$vuZZon&OF6dKTm!`UOgY-~Df2mc
zv{s3-?#?xErOYi<f6K1vF_#3GhFw!)W&Xt;zUpRg`|6FK@leJcg(^-(XvO2T$))^c
zbwz2#DhNflhC%1S$aPaO{Sf<ngi+N>qY@8Dx%Yo`MCYV@E!T9lPl>X1IsnSOW87#f
zf-6{lUT9{B4H{S1?3Vg+BCU=$_vM;d-kdam*h_-clH^O!_2H#*xYPiwye8mk!l16W
zWCxf`??+X70f8L6eOMQ8xE2U^$8qri2wVjAJ+F711=xv6XX=SX7)2mxXIc&`!ziW?
zYkhxP4b)4mi$$NSP?bZ4iu>18sE*KL(}M~%)inKw$@SA{<uzBW^VLqNS~Y{h+|X~5
zlW}NTRrZjvLhI0S+@V4xPkp!3aBc<_!)9E^J3HPzTcDm}U4~%znss_?(uej$6-nA|
zuqP^1mnJC`1*MvGxbnKf=3}W5k?#Y*g5`fux2VY(=KGA%Tx@?ks$hqxze8NM9CvBZ
zQgi0EG-$oyo}7s`XISdcu*(4ro89vKMB|kvt1TzDFo#9M!+I*WM96me=+%{_3CFL^
z4cfp_OMxt~@<6$h+wnl`g!2>mOwS+4D1pG7gob&Zqs1jP6Qk30W;VQmtJ7%2CboYh
zp0~VnNRX3&k04X6ZsGvCe56_7aAlZ43aFr@R3?o5SU?Oxg>QYGb}rxGm;9~&KVh6j
z>0X1fD(g2czvmV2ZolcqYV-Mg{~4&_AY|YjhoB0a$@9BGbX30$E5q9q^l@=mm6D80
z<$uW(H1&~moYQqi(fd=-v7#R|1tourtAwonF;h@<(!$GM5Fp716T9y{25t0TF$ZlZ
zK4cL3Sj#;op^cPxj6xeu@0x{@cEa*67=`ZFwJNQI_qtJaL(>|AQ2{8%4A%Td(4vI5
zwk6#7Z$IyaMur(PSr|ML1>vMy8a#5c;e){=D}5Z&=$XM|UR@YGW;uC!5{G|}X=ON_
z$WnHkPDnILOsx$zI69W2dTPbSdMk#I**ZH7A(5<_A!O!17(!-iE({@CZ|y#WjLF{_
zH9Le{tS%;=T|0zqq^+DD`5-)noo0~M2t%}T1UcIL{{(V}3FO$KPYfW(45;b&@*5T0
ze>c!ZI|^lPWOH<KWnpa!Wp0;_H7o`KGBTIJ3q2H<TcRw(0pqtHqb!^U1f$Xax|e}9
zEE57UGnesR4HK8}&pjxYUpOcQe>o1l=U2oJq)L=TEr4Ml(^=U--=>kb!JHaXed@#1
z{~w0h6^zAcJF2F84tCfI1x1mMt3pnHOmh0=>Gk93%eyv{6CS~->Eqi3bVJfNNE);_
zIRIfbOdo$v|2#>lEPY<-W2DrYK8}8Sy5jj+gQppQ^r5=|=VxaiK0|Y~f6C8lo1f7%
zOI_#+bbX2A^n8$3MhZj^XbklEUmw3d$-=A@!N#BQ!vB5!c>2#%V_j*1HcUn<5e!Vf
z|M?Ve7(iiNn3QosIz4^=_w?=QADnDkOwu#?3+43vkBN=^R-_jsNl#jfIFp{T0?|*_
z2BG7vEIsokCDQK9JAG*Pe}?3wV7ALI?;chuq7_;PPLaeYkz{K~rY2F&)P#wA*003=
zY=qcvlszO(Gs;TN_-eC*6xta+TQB&a0m6x-UBcJG({+(vl_%-WXm^7=CFz{VgPtHR
zlvbUM{Pqc%&sSBnSqynK#J;k@!inFIoDo(n_7J6nYtnl^KU*2?e+QEsiwsdu{VVNe
zHZtn9%1m2foNtJEW%4;dD`3drjs*FfU?@_0y&c<xQjm%fgtz$NFbQ?j3JXlekv%au
zCoBM4M1?GZq}PZP(k&T}1otUXDDuBLaLyYHJ}&}AZjVcJ!?u+&^e~@%huD?!!dk=i
z$Sv~*g3*-&M;ag_e|U#+NG&V@q(LnjohoK&Ua~nNlxsD1TTL@U6YC6{l^kb8r6W{Y
z_)<pW-Pbw_rrno?e!AaXLDlwwri$e3tv3hS>YSpF&hZh0b(+G8<$NA(u_ffORUS3j
zm@1hC@AufOqWFYM{^b1zR4kIq6(2B3CrDe1-M+267DLGwO^J=oZV0*Ea!EPDDV*c3
z+U0fYRewwCHC}k>j<ntx#Uw>%9<<;b4H>V8Td)<N_#zJ*a151<=3VBS<0$7&J{<qJ
zP-CqHE0xk6=0um-tt>17YnKMDEGvKe*AtRjPtBd&{dE;eBZ6z45&a3l>X}g2B}yoP
z*><Am99caO{Kjh`(fs6cnys43X;sSSw<567b%dRgn%tq+oZ|J8I<_mSg1V&V##$p_
zZelzD6yubT0hi_MkI^ZTGTf0l0;9N;(%o*2QQin=y!FJMdPFli2nDS*0&agJoUV5W
zNvzt_Fx4yGrrd=J$-M5`oo*S+W^3(Ol1p*6OWGD`S9f6-mQc{J*}l~-2sj`b=;6N^
z19XlIIY>mtHn&(AvXr)mo)Z;#MuUv~<S72fm`n^U`D9G4H@%fnR8s(B^5s1?r@^HJ
z^Njkmm*S+xcgEw5?P^U~qa%N3#@p2~^D|ev)$E!&(;G%zpszK#YkNE{Iv!k)bL$TI
z_Pp8z3iy`szEWAV2ha;+kUgOHyRD_Oo9YPP4epj~Mh0Pmzs~0)6;REB*xs-!sY>{Y
zo7WnJRE801UUdy`n}<dP$|_i4sQ{`*PDUB6CnK!dQo;+=W25}9qYi&B9Wu$<nhmb|
zqVsc7Ubpi)JEd23m0D25`DJ7U49$X%_wXwEV`C2V@Cq~KElV!xWs255XB|<3Sb&l^
zuyYS4&8J+sBH+Xh&@Ha~CT_AT9~s(k+dzH+9AoGwxorrfQ4BxHZ7YewXnU--Qx4bs
z_GcdH-g5Br$&~EvUY#3E!bR=Px0VvIK`-332PMox9uIjfO8-Rg{YaNcuq-+$XDITm
z)K9N#U>Viiw70c??X;k+aPH!Z21bTcez72V_e+i!%DL0{ua~Q^EFb{_m&~v%HGetM
zuCcoz*0<KtmQ-S!$S;KQzxIKVfuZ|^S7m3sOGVjyA9=xys+^JJ_*WrVlc2y7nYhvd
zHt8J~9Mg<3RZ=>fMhKT3mh>k9R=1?=io5ebPD%3>A3<u8XKVyrDb}o_IBEFGQ4<Pr
zNGXlmcY2=O3M}aSHDGlGNQP2&?|&ESE=W?Rp4)O56$lLZa2erHF>)U&tMH47m>MN{
z4Mo~etakjthd^{CmZI>^umwx5vDgkcVTX5}hQc_7i-NZHD->2=;)vZj&q#{sSRF%3
z9(5(GsG<CJ1>@GWa8!z)n$^zv3f&2pkxLCsy}fTsrd{ernrUjjY@2JQHGlowt_sE8
zDkz-ntUiNWl1>f5IKnf}eW`Oxij(HY=mpt07BpvT+g@arm2f8A%Q66gTt@)G#fC7x
zb=Sjt)eF&usy{ewB?E(gKyhh8J!yk6ifs5TMbM2Ac+T4qiKa6Ym+XRaO{DTJt913^
z?YFWs%x7NppPip|w3gCKn}0DPBadvlot$cD?^jZdriI)9xG&W(<Wd3MLI#ILh13j!
z(ic+YI4XV=BU2dVkPKQl!(Ncd@6xWH(30QUeP^-uj?Tc3&<-TtkGHO*`pOFv_$E65
z)oJJ!b;DbAm(9?ypDiU$n{(@?xDN@EJfLPw?FnUDt@~j;P=OO1-<STfEH?`E>&Mf7
z0cWCwotJ?%EEAWoRWB8nwd*)2m%v{z4S&sI6*r(^!0m1ikXr_Q3vvoFbDBfu{|_l{
zT1ZK@WyOOWY}pi9ReZHmlHK=Cc3+;}KA&FBNSzf<gH_$<w+{G*<ZqBXXmn=*ghA*&
ze{_GmN~sKg-uOdOBJ#)bZ^u`Cct&tM12lj5F2Ld05`;B82US{&Capo9rA~YWzJDIW
zIIs88$UuSV!7)%d{Q3FYlgz@toV{F=)L0TsyGTkb1KIpx?wYS!rQhWFH*?X<olLq%
zr*uD#sXU(l>p15CnQR=gQMi}0gH@rl7KR4+-V<X+F2V<Jub@e&M%EcY=k=_}3L;kt
zdOQbxfDI!1gT~i&F<948Q^K2|tbcI%QvLh*eER39M@OO1#&-ssKnC!yA5Y;10n1>0
zr<CwO-Os<fZ%@D1RI@eI*J#R77t-)vCc02YXw6$xBn)47H;|}<7I)V1+q`bH&g@SU
z9Z5kfasT!6dsmHo{nA6ItQKTcy(xGPpNZe}D;x@Mp<bb`k}r8Ixde?oAAg>WjJx)E
z<~actX2{c=HS}J*5O1<cXy*iZ?4_eaw8EsQb(bR$%9xr$E8(g?fUyF-ZKy=!C_WMO
z{1$2DsJCt~{j7CA&9mZYprzB^8TjI14O&+6wFnu4saA!+;h9j4g{8!V5Q@MmGAZ*+
z60m?3=l6H%N;3)ff_adECV$=)-ujr<L@wu^L!fj_ALgr~lrE3acvmG3SL-i;J3^xZ
zSDRikHQI@45F9Yoc?ZXIC(Phw1I`f5b&TvO8}UFR#+ex&vx?SE7^(Z%bzM+-R^FG;
z6pUAf^xEGy0~#q7mKjB|{7>eQY|?xOOc*~6wCLFL8oNf~cOjMrYkxtBvuh)AE;~mW
zsca|n(34doG|EnUi!M91<F)#<8d`|8AeFAQ5QM)mypRUYc&?$uA%aS}MS2~4Goi{<
zleHH63Y;<T%8fCkj8Bxy<=~U{N%NLPhSR}~hP4`p4QFv_E5e~H)Ndhe1ZxMDTgj3{
zz^sR4&sh1IAa8k13V$V07}sp?aXEj;kzo6rxU@y&D$ILVN!!?ZN)I#s7;WQJ(o>Xr
zl1|wBSM3`(OY^NImBB=1Dw}?;&lgRdoUqNFaw2;<a|N<i1}I((eK8+ejCtE%x12(a
z_h(%%L+~*@^0A!M7xLe7>R!u#qg37(3OyJ87tO~`G-f6myMHErCCM(wFO)_oR5cD8
z;uj(b?R7)xeOm*7Rtc0;Hvq9>SLQ+oAY>?u<Co6W@?_*|?h0btxY!jC7ka>=ln9Y4
zmWcN$VWXgGL>KxDZ3LpPDNg(~Qogy^2M4j6tSGZP&K3ILr9eydY;5F6!AkNj;Umj*
z8Pbb*8n!j9Zhyd2vd)=2U<ZLw<T@GodIjcFt_las?rSy5b;1xgTI9GjQq3L$N_q~x
zp$+;Ea3tcL(W7ykat%1%(!ka&CnsS*TT{J^=a{Fw-Cq<}S46tz3s~WG14Xt7ud$(c
zZoKfeCf>&^S5wui4SY45Q}oUFcqOPYG1$HlHWMx&8h`VtPVjD{=(+!PI|GhN31#3)
zz%jTYG-lMFZ?HWKUy<)*WE{YYQzLR{o~gF`D(k(kuypy=m(5O58N-is%9>mz&JRUG
zjm!>V2tbgwd#1;V;w~k5D#zivW4J95rTE@dcvy&goOu=3FV*bcaPC(NI?pDyXVp?u
z987bov45j=VszYkpRLPe;UjyO$padj-*W1R!~qe^g$5zF)Ws5qyL&YV*PRiz4a3BT
zCe6WcZs|6hTW&Ax!70Js(G>O+5;H!eut)I1N!ZgxrWAK9>?_I_3j1pIg9>})*?}BB
zxFD^#-dWgJ)P6}}AD-O!(UH`HaGY|z7sn~JD}Qi2?T@1-!Q38=<CN=84N^t*<_2k$
zDYvu(7KD-wqVi*2=G?SYv!jNme$&ep9~H>4=5<!~p!qVgqIphPJi|1aKMiHdg-DJl
zg>6JfGfM2Gw9GFpL`C+`jHZuE<H>st5O*{Micm;*RiTs>od2<cMgWiCo&&!vtK{fJ
z@qaDv7Sn$<Dp<<SshkvMw8Fa!_aS$H+_x0)cI;b}0s(f<>|ay-C4Eb_2>jU4!K9v~
z+D6r)S0@D3<PD8eIT@(gE_B>d3wDP+_yP*6)wO&}WT|v4cWv6hd!aP$Xp=R?`<pfh
zX5#6-VAPIH&no6tg{Mv9pfx|^{}#yR`+pnOIyxRkZJY^3h?gp>XYS5WmMk`~pjD`1
zN-i!EmK~d>qPz4qU9blQelglzt{pGE8W*n$Z+bQE8m>zZkva1-w<*^Lz2_F+(!mG4
z=4Op>xKVMX^xV+M8Ud=}z49_zX(Xl#!keJv2G8n+Gz(qs=smQ}4+*Z8xjX}?WLVAs
zHkE(=kCy-+_5i~J?&7VtWIV%Grs4Yipz(r%s}|sRAH(ZCH%r3`)*Qkru<FA)i>dvJ
zUy<NsWJ1shM-{!)CC%=R*wK;d-L1ZxSYO$f{{>Yz8f2G&H7pYWGM7PJEEJcH@H{A&
zGrKIC0w7(t(7P=01Ank`sD=cBa7NF2jI)?ghxhg|<21-k;7kiZ%iI_vyq@uXwjD~x
zoSqr|rvOhm#z;pPi;+idM!9JZIXZScxeBMb?>GWLuA>cfd;%=$Pz5P-@iBck9EfP$
z)wKCwq*md%{;FO22hJ0$<zk~(HTqoY;z$H%PHqkMoheAJTz{xnchFinR;docHavc$
zbJTXF&GXqP6<1x}!ssA;$x-{{#!5ejr`8%~ha^FY?>lt}`_y_Ut8^2WbKI>d$C3hU
zD~_t)YqGddF^2Eh*6-y|EZZ`f(#7;tr}}<{@B*^7Sl5w(butpj9hBVjv=MzhaSRl!
ziOL$mRGccRiGN#FQLTH{mJ=jfj`lpp8j}&0pq8Z(*9+v_e3jkZ!Pw)CV9;#bmd(5i
z$JWrqg8uQZf2J0NXnBj}Eim2kL)o&8r4jyKEbW$r<%ox`W{r^Cn0RHl3H8*Z>=db;
zg%Y#78t#nbr`Q<5niI2M5AiRX{yIkQ<iGA|^>uX1Ab$>p<J*BV(uZOin<K98CVe1q
zD))b<k7@cS$~i$CTZWfo{i0$(3N<-3{Nyaq7{cE8+-}*K1u9p2af}<z&j9E>7nCr|
z0Le&Rn)mhTjQH$afLtXLp4kcY5I;K?gmIQ=^nN_3j1HpDiv3axB-Jyv7W<_a<Y$7#
z5`j&+K7XU(A+}d%XP7_M@su{IgL+cS6S{pe^xfqw)S4Yk|3!1wN~mu=Ex&>-IScQ+
zt=!XftcjXPXfmgYdg6?ciSMFKYOQW(@|tYfmmph#@5Rz?`Gh2?*c6;?&OvijZyF|9
z8z$<a((+rR^T*x5bELEfrri#|Hv^@G<5=9iH-96~Qc~Nm&2~&;L3ksIa8BZ0bY(Ea
z=<&=~WDL=f?ICXrYk^fkv;B^v(^~l~dd8vVj!%xo&&wDRV<d466ukGcD$D()8Tc?`
zmtf!b-uTGDy+=W==jrURFUgkbe2)cls}kjY+gGW8V}z!M)TfnV?S$=c>N}?5N_6TV
zoqsGJP~0FO^jJPH7gv2|3h}*_(}zaXKpMeGWcP?#HyAz}QA<yj5w$v;Q%BU^G2mmz
zfwRrQZ!P2DeaF#Cj9Z^dAz`HLp{xw{I44lm;yF}5<hSRm*kl!_is{DlqEw1bIG@-$
zBJ$kU$@tQ-B>Mckc6TSfR45i~FPu=Paeo;x8F0I+Q>D4*QP6?ffB3jbCjrE>-%b|?
z1cgxMeRYlmbPcQcR7dbO2M?X$$oom$Y30}*dM$ulv}0=oTHl)^$UC+iHg|jb4uGEP
z+}-B?dl+uojqM4cb{Lgofp+)2#ir3hO*Zxv7X};yQrX<j^(_WCT>@A{QKnwC?|-LL
z({rZt4AVj_p|9*e6!zi{c^!MEAjxnKj-<>AJ(;_R<+(s|?ZoZ^C;Mru`<ouH*r~oQ
znil6e)ePO$%&?M2pm{xA|1Bz=CXKAo^GKYQX*MB>B@Jb&;(d2LFwDUj*Asb8IIi9P
zS>d=IO#11-=$&vJ<~TeTAX6T@!wR!<=dS1f3-%Mmmw`1b69O_em%&si6PFrlCn%TD
zWh?}jm&7cVe?V(skAJ@Xc91OWr&r@wB~@A&8z$>e5bY`{Q86Sbk}Skb8j==Y>GGMX
z2sRRqUthYvKEECQIrQWaiMKb?Uq24<27(dRg-HPl2yXiM_w?=Xdry`{mqOAr`NPQR
z=l7{w`O3Zsk~Eao8ne{U3PZ|U0|J#Q8sZCS<yos!f7Vn%XEP706pa;Hhn`}x6L|_<
zv{R(aw!$c9`AFg~R!ICx*+arjM_I{*Ty=MlLObkt>p>1$AYV$-5<as``zGDBrwX(a
zudt^ipA&h|6GTRd>tf~W1vX!DRivy5j8ijnMzh50U^x_44LU?CnKkRZ9v`fX(P*Bp
z_NmfNe}L12<S_X4DykSn(6*-vdR1R#Vw?@?7JyQKvg?&$x3nu~z@V(6y`x*m4Tcmn
z?xKKF&Iq(>knhfbrKD`U`)I%}xkkXE%@9i#ioYF$7qcWQmaCSMnwYzu>KWj^8?H+Z
zVAkcqn0f|z-V-JT&>i2`+rp&av!jaZ+E^=se;IVYGY*ZX;zt4Yu{i9wMjWR0;q^8=
z#s&v%PyN><QbtH`KX^Yna$4*Nc2Iv0+<J;Pf|%1|Pjv&tE(iy1aLJuPY_)I(?$}NB
z6p^XLOI-@md~b%|F=DMnAen#>SRRUm<DDda*{Sy03Kpu+P?>RDk3w80@YRukl%F&X
ze<m9sR{SYz)>y2ZadT|Y{pjYRL$Z~>$i1kQ@Xq#><cYPWbUS|}85b*Ub6lWjlyIT8
zyFfCLDXWK*5BP7+T<mcjcvta;Vr@R><JBeJibt<RgJ?yh!AocJEA-J)c%ZPR8tS!f
z(hte2a*oWuypv_E3;K8Qi{B2LT3xt6f90ZY=K_?i!3<m$)=ZXsVqcltc}h?9+ph4O
zHH!yqNE$LKps{Uxud};ei>;edD0;f7!yQv-Px1d0x|>2fll_7e+B+r8j>-C7m_}Xf
z_BvuWWKnchNIxf?|9BEbYXm6xuq2AHb+qb3Qq9^>Y;T^2;6w=Z2otA`Imu?Le^;+b
zj?fat{`jCVP9po*OY!R(D#&qqYWF3{H>aXSHc7E#YiWqbkVL^Te{Hzcz%qe!an-h~
zr^Jm{lSY2lVkuYMwRk3B^Td-)pV41v6zq1Bq(aBc5s;+wHIi9&JRKj>A;(p1%u%I`
z$=S{1f(P!gTCO~c3z=DhA)1PWe=&=kTAQ^x;cU0ns6B>+O5%pj@z@4KeP$IiEb|&}
zSoeiza-{VL(B2D}yGor6l)_alOGIQ8PmLOI7nEgJbx0{KPn;!{@m|GJK$q7g)Q8pi
zG(A9ADM|>glfP8#3LC|GgzL%}=33i((e#u2_2?pl3-s8_R-6=}06M#*e^_%bZ1S!W
z?8Wt^m}i>@RUnl7r@4eDKxA)}ko8x4p%y<A1ZvZ%ru8u+ZUv)$3!}*dRZ-1Yoz0<Y
zBRXJ~3MQC&-tE~O&vPr=!m`VW!vxxTQc3qzO9!;>4p#ysl=nrAad`zfx)X(=im&DE
z`sA#Cp)TtRXRjhf4LUc&f4Fos0HrDMQhRczHrCrTRF;aX>)M?&VOM=%yim;>ml9Y7
z0c-tK>drRb&}llJYS^4Q7??d-!yMR0rL#R7;zxK~ZB{o`ErNk=&rga^2l3@jEatI!
zxJ}rcYcA`ow{ni6qd2%;*r!)paH7J|=*M$uroqdV?-Uh<PVAEVe|L(CM8Ta`irsg5
zeYkP9;(l3qqN~;Yfs*~Odknkjy&@%7foOUq;AQB&`%!rJgnF8@NzJ>btk&v#$J3f`
zcPd$RFk0d4+{2n9;z|#dGF2=<k~<Fi<+pnty4&W_ymxkIy}#?wvh%|6d6BB=1ToYy
znu;g-wp!)d#9Ez-e~x|DXK6=_6zTS!<cet@m#tLhJMYhZCO<QeY-(L#Zpc*J+=22l
z$ph_QO;a#<USm3CZN6FmrBz0O72btAR+;#wKG@B{%-)>j!jeJD)ST+8O2(y}HzG9{
zfj#37O8=k@Rc`%3=^qdXPd9k*ez`nOu7c_%I+x4IG@;*re{Cj@^PTv=M^TZQ)vu7+
zQ-|Q-0;zBMcZ8tm4Ae7OA6Z>AcB4L+_}ZRg8g`vC;0_$x0a!L{>H?x|ZJh(9Ec@IL
z_x0sz^~DX<6(q&3GM81?a!zvf0O^N=Z#^|EPAF|<<@@_Nr*N{f9QJXnlZy!mwAF6w
zbhJ@P=<phKKG%!=Zjd!kb<W6(YiFg63R^YFM{~zV?W{|kz_r~1Zk`FtDxdW@rS8oL
zy%RXUN89_Czb<D|PW~6qutSHJfi)}>m(lGg6_<|C2Pl^?%q*D!a<|INESLub-+?5X
zmw`1b6PI6jFBX@N)GStiYbh3AZLdn<!;gUvucyoV>Dx`>V3<|H!0`Tk5OjglU%~05
zR)Z0OX(@*HpTl2|oP(y%g+9o+l=Lz5-O^E=AEdB!hfwsPt3sR~3=01Y-6J7BD|LKE
z(p_++qoCuHKTpp)uDO>$^2C!rp8tOT<HVD)-)@etX{rq}lASewMOpz#52fM<X*IOz
zJH+40ROc}L_UGxkBF&Tz8hpE1vAVKSOyh{rGd<}s;SnPtKC8{y!`LV^Z!{2&aoa2O
zbo1Z}<5hEKB_E@DULNY{rpa&tcILE4-XC7OUoY>c|4s!3kW6VeXknQop#J%J@)uB$
z4#o`t%n2NR{X2YrKmFNDO{VRS!7l^mT+?0}%t14y?4SXersJX3CWBof);ylZ^a$P(
zCP-S8Kfitq-I=%70t<j;T4S%Mc=n&hvC9+EFkxi>gbxqQ_OjLXbab&+Pdag(vtJ3K
zg>X+sRMs+da<aAwzL&fx0C(|Ma?l4qvI)1OM+32|?+##pr@MpU!mi3yzaLj57sH<8
z+Kq9uMy+|z58j3g0iB;PNrxmap!-yGnCvingbEQ;&Z;wnwQR&7+bP~*(r%QjG~wy^
zJg;GjzKLB7r(-sY&!PdQTZT4y{%9f|K?qyawL)15uT(HoHpHm<TFwIy%o*J)#8ZmD
z#&icO;RttsLotS6h{od1OO7k0WWx3O1b>ABMgpO-X14D&gBE<vZOn{xle83CQfaH{
zs2zux4Q8*ry)KuKY+nUTLy38n3|u@kYQLu(DoT__H07{Z!tX0Zc1Vu8O@aZvWSJZr
zDFbXNhl*1nfV<`J<Jv_3Fq_L^tI!vfLo|Y_*H@K)Lo|j-X^*w`?<j|zq8rL#_r#Z$
zLqrjD$Z75=mxbVlbw(8>iDdujN<&o=IxYEauAHb+pi<!z>v>sf%FDgNKBEP@-bw~s
z4H!)si|by>LQO(C0xn?#s>UY%3UvlY0uWX6S99U_#XCKTY5yi&;cclvT}$opq@H61
z&qzXl#+6o)C8uJbyWRQ%7qo&B4T}|uDyFbKT!nL((S4grnD|KAVhygTz8AP^#!|t#
z#5|g3a}`XIQl*SZwlPQ0x7=r|9AeX2GMjO%@o46~hZ;+u<ec%?UV|Q_DH;_6H)mJf
zW{ijs6<7bHUCWs!%Py=U-tSjQs}+um6{*U9JkVrR@1T_B?nXlOX(`IV$*;V2Lalg6
z-gU+)ps63*wX;5fCB;zHzIX~gYhURcrR_p!_$<+@Gw4XeumbZsHWQ&0NZ|fPDn>ml
zlg=DS#U!=kXj>{~tG$^$g?SSNlT<soyHCMBtHnYY;V#UKS}neacL9Ozoa%X(CN`{p
z4yRp`YetwKW7$BrnovU85@`)iF@YnwgR0|GB^dWIGNB#m4!!glM&@IO9_L;Ls3^5L
z^HySDiWIRoEkwTxthQQRbj{lU{A3b!FPY!wF|-(}aAqsGA1@?Vm%X4B0xOiJ4CwkN
z;8s<5JNC(tYe%j-xgPCTwLsJ9Dh@M$D^OTXb{57GdAz_Y+Epm}Hp42CIuWj3XaN`5
zo#rA99ej_%Zy&H_@(M$oY@BYd8WX6O#Qupe1J`lJ4ONsU9@-Yhm@_~F1D`OHzh3=*
zpfBSpmf^V5xc84oc)OywE$=c%Yx@MG>O09@>33gWJR%LzN-j=_M8oa(Xi`XjJGWsg
zpYm$z*@4@=+m{u&Df7h3CSNDCm@)l=+1GXO@K5xw08M?8>WsM3524cc`yrtzxd=It
z?do24B)D(@pFmETSNALCK{1YP%Vi|GnWK@(tp)E;gO{RYbcxB?6=n#=k##2P-%oc-
zu8|_7XnRgQOzSZV?IeXLk`kSNGFvD{-B;qIYs9uipZf9=JZO|}kb87+)uw1u2JUQ{
zTCAuHszx<Aj^N9{e|)(x*Mxwe8ls@)@@3wg+dM1e1uB3{(_!=yh0_egSn?)=WOH|r
zT>6}*>lugCsTnM<KSGPzzm_=6w<Quuk^R0v?&<{YIW1VrESP?k>}+9wrMf8%sOa1a
zwgY;ons<m*(`%Q@z}|j=Vp}WSw+n(J(^Lu88-`d@!XK3iSaaqrH4zKcI1vItltS}y
zGW1-%LrYyb^vo@<5WbBI=QD3p7TT{;y#^X*5u0iNk3ew0mar7E&<p%9t-X)oT>uqT
zB&&A-xVW_pRx9y<`M0nlbmpC8f8g^x%yA|S_IJ<Dx5GO!^8c6P@C3(<cDo!>8xl{f
zKoo~;&b)2Fyj|62pgaYt2UOz}XaPA4FVZ?`9wjx0-q=m}|7dyHPfXKWif%oD9X>lG
zqG0HAWhe3Z6A~$@Suv5Q=HI)ZV~*HXkKhO1`6$Z})sfhwHsC;Pv}R7pf34W02BD7-
zetZn_P)5Hx2HRCX$O6VWprpl3FZF&be#MBy0mBK*M@~qH>KI}}-3kZINTgJZ;FyEf
zC$jR=^FB{l(qH<42g1Y=ubT&SQX{imO%PZcnWSP?^ry%i|HQF=eWot3DR2mO>k8wS
zFFCWv|4QVyctrpIG31|*H8Yo04Coxp#TRzyLS4NjiewdK((&!_>_SsLRI7)-B@Box
zU>+6S$FS3V%e<TD@&6T5ed?EiH7pbXIhTP&1|<PFmmSR~Gk<Mw8%GZQo?kIQzw{39
z>~P58%p8gWX>;kdaDgDs0e2S|2U$^D_Z-WR<s$d*@8QU+7fZg#k`pUMDsy+a91e$k
zNY1QMjV?H)njHlNQL3Y8L=BF$JrXCl3nIA`qm!D&C5?86oCa|kLv%T2Bh(e<XLwXx
znejkisA|swM1RL>lsJ;q_!B4ymYms9;&zWfW#gmhC{r>DQc5Iy3QbF@&p;IyZzWe3
zV=TrOT%1WJaWP>MppuIk*(e32O<r~}LzD(cj#=7hkeAz(>}i`uBm$M~4a}SanFcTq
z6WJIMHlj9e6vy1gjR};QsxbJZ%?OIKv>7odZ8O3H&VP|GMl>p!rVz!on5Hn3*!022
zPVk>mSP0n3C>ShJlc%86CXI1LLwh5DYHtX^u$nQ3WGrGf$y`8{_LN<sW>YXHN)<m4
zpKJke&sJOP4N*|Vbe7-}w{=d@B~g2t@HDBd2Sb3GO`{Y{f2INM)od2B2~x9P!UJye
zO%TeMZ-0sjqHGGd2NqG2V(z2OUlc!?6UHzT=VG%=5L(Gr9}Aj;w0qTMQX35>$<lOD
zaap<XDX45(L43}$cMB+pk_rdOB?OG21ggm^pk~?^xN~S$%j60*{c?0gn)}1B1yeFg
zT*;<xgwvGTESc5Sz-B907b+{pfY2JNNrqcgVSnR;A*v>O7XwwTO@riWW*ehGN`@%O
zuyZZcv<jdsNFyAsR*MA=h@(IuKKtgI?#+L%M%}ByucPk!@#S=MIlZ=kWJLVy?)B(;
zd@~u2u4i#<w|*YIJ0E;E{_r+5Oh8Q6c#8KYcoWxo_Tux)%kfN5w(D=defGA&2Iv`V
zfPXlhkKVXl_uY8%ZZv6v!)f=o?#J$XTp-X-p~!Ih7Qo<>K|mHS(V!suig2J3@K;Ws
zbtgAxQxkZ6e)*s7`CR;PdOp7FPP)Ip{?Y#b`nT!y>iRET_v`uex0^FR9KY{=zj=Rk
z^ltP=_tVYKul(;9-Py(Xtkco3U`}VkA%CQDh8P_Thoe~QyLZEIR^XeQXh@;Eo(#M9
zgW+U+jd#B~25jijF_wt_gsJ1-%1l?(+&pcXU9@QfD7=BKS3_9|o&XD(JjZg{J$Dd&
z()}=gGwxovuYoF3B_q7yoe6ju6&lPGc!4EoNdCI}$3OoC34mjAmp2y|ryCh4WPe`}
zH)KO^!zbC6VTsehr7p&kldHjS<eFnRc*=A-nq0a#yZq~i=?^E<!F1$kcH<?S5M#WA
zKa&NMjq(x>LkUpcE)Z~Jpws-Sdo>vkPe#+X-K!TbyEmf`)6*q$-*M|0*4)}~?BQ;0
zDQxc6we)U%>))2d_VI6F^({ni{eN2|`VtPwzX=)r2*3Oo{%!u1fQiThp0VP{jvhG*
zK<!aithqR`Vtr6#umo28<dEh*zKgkw`YbfJc4ZImdTL|ezWLQL^5N03a$Xo3;b*nf
zBK#ELryGKKm1`EHz)demsatRc*uSJt5zb2y+`U9%C5m+|&>tGxv>w~EQ-5qzoyWEn
zlD8pUdZG_<+Aqja4q+h{L7sB*H9dKd5673+)5+i*Hl4X&HTSFOG2*<3x2V2a#gD)X
zC{YaVXg!r-9ZF3Bg$;fQrBcwn!aQ1fDtT^jl$ahx?o%85^5Uu)T7-{|MIY196p{x$
zH#Cv>OdAjtRe35X5^K#U(|<FRUoIM~#rVHP!!kpU9$Mhes3N!EL@AS@<;OXpo7a%G
z;Y7wntF;9u+9cFdbsNqY@V9=>@@54bABu-@9S`G9co-M8jy1FUfu5DEX#si)+eST<
z801ioj(NGLF<*#V<=VK(3kxa>-iDLKQ{YeGTlb~cz3OO=Yre%{7k}U4yyRPbu8n=J
zovFQh9Ua(XvF2qop><L^0B8I6=)={dd0fdr{mOL{>OUa0<>FBbFH&0(pEw^%SfNxx
zGfcAIBo%czE#`>wDbtfraq!D8i^UWx)g`7?_+a`-_FQ7<@qjCNlClkFRxjI#jGBB}
z>CW%vOu=UqN(bD>%YQol$WHhpOZdZOzg?+BXj|iAS1M%~oDZ#_uMtjK^^TJYZIk1K
zwypCvac-{jHgFObPg%C%L^oT5d@(gyTCd5{PBmFts>!(TNTtu4(%(t@m1~+3^2(xR
zQroHFCvTbl7);IwXBVT}y{H`1k^R=kXq^%=Y0Y-#;cZVoW`Fhxj+bB&+OXeBC@Pq=
zsw8wLA%$4-hA#DiT13D<!v}gTSjF3B#=TmET;QKY3mg+QHk8?`Ri=uhq6;XC0e|)2
zpv?MKtbEkE3TEkJ{c76Kg`#7$!@VsyV+Q7wPrfg<0zfNOylqy>tSiHX9h1hq3mWM<
z8tG1Gq)R(yseiBHQ;$Zl+3lSkmUI&=Jv2tBwq0XJ0<u#;wEN!9WbL$sh(A-L3h+Wu
zIP_X13Eoah@6+OX)I#OYzT~+7=zY9$#@tcLkgBhY>%tgaA3FN>0Kuqjh-3~BjN)7=
z9UMS#E&vFLv}b=s$>+R;htY7E52e3CnmEGGog)I>YJb~8*8z?N*+z96&RD%2D1V%@
zWsvpmV*@9Z;x)+^vlUX-TamI&D+2gg1<_I#Ae!kGDRb-z?g{CM9#*sVL=U~Whu*y3
z21;UHJ)_`y2xk<l^<N~b0uZ8mZlXvKAMR`yC4KT8hN}tSY!>E{5$5SJj$EUtw@*ai
zAf6B-QGc^nvCo(adbG5jJT(CBLu6nBXLQZVx_|oOV!QP1AjUok5<vrv9wwsNl~mW1
zD3n!eskI2eI17@{Bb3_F&P43fD4<TXFS@+6!y!0Dkk6w72ln<|7;*6El9A5QW&x*Q
z@wS7oelixc6b&2-VMS7T3?)QRu}GXNk)Tx3dw+HKB!=j}1fXC$Q1&H2_1Wqg3&pyt
zAs+B0fMA%VF9G0>fb2v05&$*k(XoU(%})8)_F`GzHc2sTI!N2ZiOONaL0Z|oWx1h&
z+QgZW;Y;~qHf^e|W%w#PWOxp|$ndjI;7nAX>T93sXW!nP{tsg3MpO!AZe(+Ga%Ev{
z3YUyEEC>QKF_+OIJ`|TE<}AT~G-lH7_qSi3Gz&XQ23Da@gAYy-ttMHf>PRvdcOM`+
z8gWT$B)*6{6X-YU@d?T2ke{Z-8j=Zp{)Kr(%)jcrtU1&R_icWg;qr_yu6u5>(04P=
z?IkSF>l4JxefPn%aX+quyqBVjkB4a1tYyKe5IdDPUbJpvYmEv{F3S6VB#l9pBg<gR
zsIf1uN?S+q)2Q7hr<-Z2?a1-INw3aN``#I!SCBT)R(>$<n>D&;`lkNY2U?x-`6!ml
zGsD7aqxPL=zGh+X1Jj0a9u`Tmin_DqLFouhC(_Eco+n}Ddj98Dt|PsbCkyiop>w96
z+7n*Gr?#BJ#ri@SMrzN0{5%|&Y2sG>q|BdN!LS}DY3eClrC6S`en`OS9Oqe)P7d3#
z+G)PZ`K6U{c5cdS)?2pzq)lV$F7ut9lqjK+&w(P}zKCD1Z%==9CPwS%KK$|HDUUmY
z%7-)<x}y)n&p(H6Prpj49Er5jw5EQb9)5lw#K>0;tx1#LY3~t#7oiorLL_gsNjX0`
z-e-27!(?tSix9}}-%~`dtWA<)HbpU8U_&=mxh!Y<DB>?hzRrSeG$joP%AnUnbJEJ@
zkg}X@J2_~84v(f?npc+Tyh_jN(*%VsZP2GCn-h6s2a_2Uu8Wbc7wCM+73?|3J)$|M
zJal`=TEaE$kvn*QZAy%%>vgkT!3Jf5mHdno`q@Cy8RZ~!6hGfUk$jSrFDc@x2Eiqj
ztnW0q{mcVK<s9BqH<V6%O83-__bQ>iR5!sW(7lJ~nSLlFvbO=UC(K7q_#agWDuvEY
zxzxiwpI_Zl(#$LTMX}8SDDWM1y=S>e3DZZBdoePPWi3vBXu8i-rgch1b?j1bm~xMc
zlPg2306D*wI)rmo&P3-oOy6H=-kH2^amvdxN7qv|)mE1JX(Pp4O699Nd>a&LwTcD?
zOgQfiYACAJzdqEvqng`*wrh{EJib|oNRczZQ6X?$2s|Og?WqXK;X{7(uBk|H&8g_J
z`f5FBXiLn0gQR*%N}8v1WYtZ2Y$K;a3Wt8F$0G4XVWS0cL80`lzV@Ur(@ow*D@kq$
zEr`@oQLWgxy<X`RQ(bh6XE?ggyTL0wqfYUf`hp0Kzg?h3A^dt3O)g<04O{flmxQB%
zx3>0V@_-(#XhpqV1OsfcMLzYuE5J7Q4UW3zL<dWMG!9;?K#WiJ4W%bI;=+7@T%^Vm
zr=EaY-4Zko>~OH5ZF*U@w~qD@aBx%)=oCjT&>BF#YFod#SMW(0vXh{0KCdL3>#GeD
zg7U^lE|A#F+_*N-p=#-}kddJ-=OtyXkUAau9GjECS~ls+(Mh(F(7n~gxj}G&hJeY#
z!@t*mtBniQh^pg<SF|WV6tGjYIduA&Rz&5!+qzfa?40-9B3r|diqmmr<@L^7*HMgT
zhI$n0s85)BGFt|qoLq4-@QQQVB8rU)0J^$)EkGRwgIny=i|Ax==SCPav@pA?08s|S
zbSmK1N$ApGQjkJNn?b2&q#Bc!yFOFU9NSHQ?tQJdlH&+)6{!oH^}@!tQ`#VRixIMg
zhn!toF78H8Br6D3IZA7C$9@T+A{Xt)3V(Y<!ZWs{l&T;^$^DXyGTxs@EiGU`&6v}X
zz;R3;<eSQ4e*#XU*A8jWbXpvt(0cZ8A=~+mlxMmIb%#}Wm=0&jbd9HF-T4{Bxu0Ht
zEzAMZ3XE2e^b|l^UZAY<4xlL0cCS754G<3P)yMH{*r3$?0z~dW?$|1j{R-@RH*c_x
z5|owuy0UpW>hm$G2tjgbr?qfmNW22+zMX;zr03U3<4Yi@uK==SEsAd~CV`3>1SgBh
znBk&#!JsYW7eseQ3|dgUBL*!fpTVGi#q4*&pe;~!YW+Jm{(|d`7_^}Fa12`Ty8{L-
zDc+7jONys4XsN;LG3Yz$hB4?n>V`4s3w6U7wDb^te;?GSf_d-vK!WR?z&AnlUBNf0
z$L-*opzvVu4Jdwve+l-M--$%sL*sZ_kpD9F25#ZlkHFqca4PBaAnXkQ6-9=B2V-ww
z9EkuA!rlb+2Vrl5%N?;dLHP*wR-&V}w_|UD&`Ch1hIz4K`@6#Jr9e}Skk#DHtQE17
zJA0WOOucMu`QKHSr2?l+N4&kX%)!p`3&|UBHIt~Tb8Z8!CRUu^>{ZxX?p#dtGwh9n
z_~Nf)j8WA5T{UyH{|k;Mz|!V_%VNd7z*5pkeQUk_ymO&?ek(Vfc`ipcZvO^coU}!0
zN-}JoP+!a6fXy1W$X;i$*#H+{vy@qZq>!{9iOojp^(Dg*nho~R&@4qlnl1QW6dumP
zw8bH|Ahj7MMkj^YjMP@8&5_zVas^ULQN~)+org0>ZJ<6e#?Gh>sV&lfj7V+yH7UIP
zb^B-n>iodk?0f$RHk<!FO8ozK*sR#sgRoh_WgVMsT6rIQ*3<rbcs9o<^2_(YXQ@m<
z6E{H_=No@n(+Vc-4T+s1duPcbomBLSn%^b8t4rkDk)S~Jz#XCOIk`LH+H<mJK<)X|
zixBPAqbcrm^8W^!BzriQ@*g`A0yZ<35eGLE0yH(1K`thjaCtohmvAjN1edPvEGPmj
zFPGTvEFgcM2s-Gb@rcGTHt*k>Kc4}}>2=V{0x;|4^?T5}?4KBezJa2bK1JL=d5dX|
zzQfebW!KGNeT$BIN4>wKah)H*L6V^I@FiXQzute}K{gglbm?ol)qlOc@BZEOrV|@+
zv2hR#1>5}nZI>Qc6ytnUkBnD`zW>vF+5J&V?R9^FPU$`bn(tqmYUFbUFM@&x&WVh`
zBWFZH<0KeX!5KXJ3Lxs7J?W*+TZM_Ra?*L#fZSlx)hZs1sNzqjfU{b~8Rl4PC0!<o
zoi>p{vz2T~7+KIui`F=r_yD^2IuS${)cQ2LlLDvxlY_L}oI>NRlWA<it9-QHj7B~W
z`iOtMLC>{5v=u%JBw93<{6ymG0}?fa6#QDE%pyoOn#gW~;AuHFD#2P=6B|dy#keLm
z&Y38DD>j;i1k(&rctGP5OZ;3<!>c*S&R8FQbPkTSrW~}c%No&=C+WM-`Di;!+HpUR
z7O*P!wbbjmNT)$|`!v^pRK8mTUh=pUTb6&vW+Y=LFyjQ$SeA_>LwY{)!9r4WTCnW|
z(nDu@GCP6Ly6J~Sm9Yhv7K~Yp!Z8PJDr9gx4@AkaHw0oC`M`b_GGJ*Eh{xM)bk+or
z<p|_G+I4#{odsZ;mwZgg*88a0JAAZ0jz-QFy-zL@BN&TC{8q|;*8{+*yIa^lp$32P
z^UP!Evc6<|A217yVK4y|V6?jL_qNO`n%W^_JJKr-=otdp{B0$mKWb(MuLE$<VHOu$
zn<PNl`?)7WC7L#HxeF2>4RF6iQf$lqq^eYP0E-(f8+BXM?dgt2t8Jn#l+qkyq9xVi
zuti8N;Ky9T(n8R!<uxrv>uj&Zst<o&8quxYi|@72*^*0^EQ^aHx@b_>sgjPbGh6Nb
z6%xHS4zbokF)uS_baKdcDBV<fD0TnV)sk}yGr^Fp@o}hdaXDUABbjSKYr~cI=t>La
zo_G>G`EQPSuPo2cLgFfnVkY|Jjvkq`36TlVc$(ld%m~IgyVO_8U6K~HzOsL{m7??w
zxsPzL$NO5^K4!FrgrPjmEcLK#x{oTIt7aqy+2?&}b6&EECF1kaC3i2yE|LjD=%$ws
z=h)nl`nqXbUkP44^Eqt@l@cyVzbx4c;~NW;w;TqE!5^20Y^fk)7;rvXyF3}adS6=R
zkPW%!pU&m9A$zZ=6hU<em5qN2&_wEwO8T~HZr{6j>{7N=Wm9&5ARVsw6qkpg%NH(N
zR5^NSdy{Nudb8sN{nI?!B}P9gd%_&?n7G8ta7rM)q|KZsH}nNc=|kNWpleC-mE)J|
z4D5!^!CGs;`;|yo$+b%A4F@8JJF+GvLT!IKD}xWqd^LlQ=XOks+M|C#3PEv}F{3G|
zb6?1?cB@fA{&_*e!5Ly&IcO@W8wO3K2`>eV)!CD``Ek0d(oDn&jtl5+Ic2LQEDhf+
z*c1yW8Nn^t6gh9S+Hb_BSStmeH9Fg{DYf5NiW{-1O8Q1@s?xF|o2n#V=tvpTY0u+i
zYAk0XE;UxNIh7i#dw_pQRp4AFQp+6uaXgA*I*G#tJgS;8;!&KxyYXSy4Lpj1)jC5P
z5UEP?Mog;G@*tI}q^-oIY9u})Q`Nh_4x1uf5Q)4Bn<5U{7SpfJrwF3f9{P#Raf&H#
z<I2r(CH02p_;_b?Txm-6{tBb2X7FKCMukD^6!|q76$-Yc=;vq!4T?7r?8-q?N!^f9
zRhq6as_N|(8C50u4n|c=_&+czBvL}c3m8?#>Ech8qJ~|(%5ZM@Nz6sl%WtECAsBX*
zs@Fq&Q2BhR(Ah>Q%2@f!$(1dQ;$G4}I4^UR>3`?*UF*kxoP?*BL4Pe20W+7;@GujX
zfAlQIe{q|7KK}FV?+2NgmBO3wXZRW4UfvFW94hOe1sOjX2|`OVef@a|A5f7n)=$<7
zEB*B6zo&19&m+|`*YF#Ga*BbH#t_Yb0wajrWQ-Qt+5!}^fW(3z3yH7%u+YdLkmp~9
zhbVu;)Hn9w{xS(&sUyo`X)*)?m8;Yt^$t|xe-kKEf=eI|t^fV`eHvQ6d}`E<7fQlN
zU40DC06qgMg$y)+64hA?Q~mJj<tjpqxKV1H1$;t?SOt*Pf^0%0%A#hoZ($EXInj)i
z<?A|zs04Q|5T5eXfep9;Iw{EOf#?cy3#?}Y^aXW*8JF9lY34>&E3FjSipbn9+nlOW
zf8|*h(6RxLS&VvF=U90`2$ka(=Je?foM({d)Gu-FRP=Sjs1fU8=+X&ty7oF{F4z3X
zrgzOQitJ2r*^44RX4AIWJm?txDw_m*SQ}Tj_}Zu$q0dw10mwNBsPN-7>4esz^-DFW
zC1=(NMQY>>vR+a))VN024EMYq55~NXe+iHg;1)9nt&fyGCRxjjc`X-+5(;$9zJgPE
z&eqN`2DA<|Wy3gpjdR%g)wI^=!e3`!A>+=ibSkZpLs#EUX%5oOlIxr*&+B@s5!*&B
ze~-Z;a}ej%o!hU{L`3KaDr-($2itkLr_d9^Dt=57F&U2Db-X(&6C`_XoJB58fA5gf
zdx+}V2OW)`X)k7KE)R3Nj+@Dv^V)D@Dy!>^AV|-m?Wq_@G#fLADjQ5#XkXq2I26%i
z)vFJ$Z0hGM7m$7A%s#5$T3PMl%P#o^O~P56csSdF&(xnKt%bwoXe8EsJB&V;ENt?@
zzTBZ7DRF3fFbLTPAw+L<K8)vHe-Q(yV`l4ITTe~Ts)b<EBZO}G66fm}fHwkh0l{_R
zbG9m^H4E|8*{L-+fmYuEe$JAGatrvQVXOQ=1_?W(H9g$X?YRnW4khs`I}3QMoRl#|
zWlgpoOzPZd!~lIN)GKAo4$F35JGYS*dPoA2qM~y`GX$$CS;wy&oEpr+e_`ugth|#x
zx6C}k(W}SpWFVU$av0KH-)MsK(QC=(rs)~vku*|>rWYM=(gH@GaLs72nPOZy(sRSP
zN0PwaT38PwkJlx$`;r3B$)sA~8N!g$awWhgd1DLCTCgCt*nyQ~>olMh-;{oa7KG6j
zn)o;b-PiTNiv+#<OZ0o|e_ZEjo<5avCI?rO?JrAK!^=gjWWR#iZP}YrU8?caycXsZ
ztOWNs#-7z;;4o>_az-3RZ|E1Y4m#Y$FLmu1K{}o1%>z@{BlS!E06DSr_Evu|8@;y8
zx#$n(6oNlMs0pdw^t;&u?)1{dU+VnD?CG7PNxu3_0<`I`w-Qhse{3mdcg=e4<<C%p
z0AogsL0`1YJw6Q4#yZqMy+pM>eP<t$if9*fkuwb<bE_`SMxWbyUrG-LN>}UT%Td?u
z?kRCm;(q*Y+m`pQ(lGODG)x=fj82V+-Alt1a{~+NJ2Y(E)}TZISanqAoGE2eHQRg5
zu?SkV>mH>C!&^G5e?}gz#OUR1A(ZA{Zz&Xl|CqSrrz!SkrnPr8{cj4*DsX6wvHYu1
zDIWwleE9wQx4(V;ml8bSzhQ)&m^o6s^ZS`z4NC$@H2p~LmQ!~Nep13Se7sx`wi5ik
zjH?Y_rvuN!Aqb%B_9L`%dgUZ_ePi+3v2j)Ut)+4h36|C}f8CG1u`tzUh!RyU!dsw{
z$A(uPUq*%3@bvGcE06lmboe$MK36oo=6|71Y}xs(fXS1xBT+q|waN2nw3&Mh&e?A?
z_G0EXU}+691PX-r)`rE%bgQJLrIWTaF{whKG?m}6eM9;|(1b@)NfN~SIw!1(`lvFb
z3D)-a*PnUVe|H}AsryZ29+dV%F{KCXAQa-6R6L{DC|cy+xpOh@r=zAW8HrA?Im6^0
z(!6WNHK)l8e4aY~qMQ#+(9e6Z;T0mn9oUS)vBQMJBtZjZtJabQARj6Q#oM?&g|OoT
zSUgwuieFP1FYO>w2`qdmyOB&`s}z~O4m<AD8V=Lqf5Dln{dVSSnXep8&1N($e3pGn
zC(kMSe%<YNOyO?a-Q;mYxhL~Hr;YPj7&TS*>vflvxtm3<vW%Q%A$#*a_APKtJ4ck1
zk1{WoO^xby++ZN*682)bw2ZB2V8%7lSnM26t%8%sK|g0KCYM5R;Ze8!VeYk83z8fc
zwV>yxf0kph?Sk{HwySAu<kuXc$hdKh^u`*&b=5Ko$2J_dM-4FRdXk+(x2H@MVY{bt
z)a{SbzPtNsw(Cz<=C7wx-`c2VJ@xtAVt>?C&gwzJdd+Lioj2dP;?J9NPX7G2Y2GU*
zTn}BwbuKKG{7f1l+qqkD!4b6CDazVO+vR%ze_(Aq(i-joD8uQ5xo{6a#C%YNh%fWm
z<E-V&uDb!?v_QBuqwOmJj`QMyntaYML0j)YMY5$Esj{a^Z&1$NAs;8d=kStqbk4VG
zO|HbM;^nh0o91_tAGvW6l@r?Dw1|5a2v8Wj!{v>!+<#P$H{Z?xmeG%_-T7`tppJ4e
zf6G1YW^nF@NF(K@mG8WO5v;&lwF95N?uU!DzVQaeHLPla=-+<rQiek#q6Hw@=HiN8
zDJQHiowg}l50vO-?mSp8-seD+V-;yPDYVYFANs)Bm(pb|g6nn{pspFOmy*rlP&ojJ
z$==@72qj(-u2b(tS-#OQ7Y_`<j4(j0e`udSBSh9_RO9HZo|{u5Pw8MW=V_K(Z=Xx@
zoV@X0&iQ~~_Sg5dGpx+8RDe`*g7aLk9tb1dIJ%wyrA2{QUu@OGQ{j{PDi-wWMQdlA
zX)$+HDwBGyWz|Zq*4mWi-~(>4W2p+(RoAYJmI{bB-foPR+A*5=m!v$3*gY$Kf2WjZ
zptd{ZN#MOBJmw3Aqcn4Kxflv_kR7F1Pssx3r<p7=Dc!PAe|^yOMp(A0Q64!&p$HAh
z1X83pYdIH%Bnb!=viS_<&KSA<Q3%4Jj9Q>}ss-o55y?_)?ppd=fkh|TWqPWm9%%~x
zeS&s{87$~)CA_D5>HM79yVCroHdF7mh;V%s7TjHH5rjhdg@~2;U!n2?!@oe~=QF%V
z<p*5elVo`k?#4aZa0#9x&Hn+x5x(`8fi)}>m%!vV6qnV$H!YXF{w$dRr?*%CECT_5
zu3>jFFwUln7IQ=jVxKOu+=PVBGV8+6M;omI`PpcJBfxMkyH_X{m)NNqPTR+$l}=GR
z)sW}jBKDY4n6I&HLQ7~#lWox*a$-xGW;5r|V+Zbi1S;;n$?{c}d91kjv4(ti&{@JC
z+^3pJ$_$&7lrQ2;<}jM71XuBzgktr7*JFFTtbNt>P=i=kofT>8dTe~vAI5iWKil30
z-DWwg6>Evf`znI$uR&8_7bEL}%PB}>iHnm@UcQ|7pt=e+dPUfKe?lPJRkLeKOpMc2
z+X`Ex4AG6Y73+=i@DX>Fsbx3XRHjOH+*784hb?6Z$>c#llfQwKx3~;iKLAmGp#w1>
zl}jm4=X~8_(vPf7<<vOR>V5lI$5Ne<%j=(A<C^qtDJW}(jnE#9`7|=C1bImST>l|t
zt>$v_w_bTiGubN^2HlX<tQ$dKMw>^-+fOot>Z^H}t&hw}w>k#|X;5fD8rw2*nQ+d#
zEz8r8ank+}F!@TC_B%_(hh}$wFXTBQ%kRvvy3$PsW2>8p&^nEUA+Aw4<q2bQpxOgA
zhnpar7UZGFX>p0uo7h(Z7v0DLDhPfCjjqkdQ>h<lwC+g{>$tSu)^a$~Y`tqnn1x45
z;&EKQ08EO8v+>2^<#2tuq5*@7=PY%`0zNkI!(|~Hdoj{R&CVzgL)I*RWi1R^J|D&L
zsF3K3ndH%;M5!*-mm=$<e<X*)bkH;{t44)X-)-25wsUS>tvFK}S{;-dm4;2v)1<Jv
zyw|lEJk79lvsv0PcB5JCm4uLtBCdeMypWdsU{zW{&kc;9NpA&ZTL;W(w@1&6acK=h
zDeEhS4+X2X|86b&3|A<Bsd*G-pNE_uE|z_@0?L1EEgU9yUTSH0x;AR1p$RVcge}xR
z*V^znK#l8)BCw7ZFiMGe`@o^&Fx%@M2#{$bAsO|_q6Z>Z_}<93X=X>MX!TR>VT(=6
zkS^kx^4QQTtiY4&S<m8uI!KC2g*(3UvI6s|){Q|kAZL`r*&GsoO7;|beZ5PGxO?8-
z0+3h0K`Hcs5%_G7Xh0eCGl`$CkjNp);7f`2G__-5ZrCJ(+kL*Z7&~I)kUOWiCpM1T
zQ%h^H;jD&w9wB;!&lSC1rQwA1<sx=fZflK{T-Vn2FbKn+!ZnQVjB{Db6P6S>H(;i9
zj2XKcQ+N~wM~!8FtTkrisPe{K3zS@ot|;5>^KvWH*R*jG>1PW5trAlMoUWPoZ9tN^
zsNh=FaLtO^_bZ-rQTe$*AIqoFTzsLXQ;%(ae6<mjTDWl52z%NHd%=j)oBsan7DcNx
zw|_Tew}=x3M7$BZMIM!L;Um&p!n7OlTSCi@1ecKfW(?PVGHx<6ixHWCc&@fQTtj9Q
zl4nAD?Z9~vap9)TTF#4{U%`1fb{Oz8*&mei;?4(A8r%O0;lhnMFRm0L_Y@MW-TWs1
z{G9vx6BAe>!8rU#&sK5`306F16?5XkIufk+Sxz@(;!Y$Oq&<OK93fQ?KbUP`y13Od
zypu>-3zQRoH;OD>j@rm}p|;9m+^8G#35c1+zZT<_#uwIW7siV;uQIq_j`0%n+L`eZ
z(%+l$Lf)@Ey(8h38pWVtN5V@;+>!7Sk{1Xs@%9@LUR->sFtiVEM0g3UR}x-A+RX_s
zp>GGmOG><+@RAZQ6J8R*pC`O_#HJEnJ7QA_ua($;RKiO}h#tqDt>HGrUb}Ot@#Ea?
zVYcd{T5hqyD`(JzfVb$HY6iN3Iz!7l1Df;H*(H)H(`T_<>5OT<PvyOx2q~nPTZ?8N
z9qvuYC!uacJ|RYSn1M^wiv$$g^gd)0u-}Ezk3%;go4^qV>5aH1@Lc1e|A<_ZFzrTM
zlhCq%BiAG(zZuuGjGN?|BnFS8h1Ap2k?pMNTb`!^?klvzo1CYTvOn~BDq-wfpQjS)
zuXU5_&Ql3v-uXP0bfbOFQwgbO&Qk$LF$}*KZ9Y&X#;-h4CAI$Ck*bZl^O-6!=Kn)g
z7bqssjF>mS9K|H&wKK&er0-2Jwd$pQa?$q$AKcCRm+>Dv69YCgHka|TKNOc>3qU8A
z8>R>hm#_vcjDI~qAOC*+<3VO-rSK*^hFkyF>Gk2ihswIrf{Y)?TcH&V|NM9eFAzu=
z>j$R<zx?^{@a^Hxo@m`l@{06Ba`^du=tjO~?7U>`q%lMj?358iZZO6R9bP5u@Qakh
zvf-I8mK_--1rr%RpB*exBqOx;Jw*ml5+jR@C`QVOnt#B+XZeZjPez#iO4&g?=Z><H
z6Ta$fFNL<)&(<?Ocz{n(vXpSjJY9$MDxPQsYP$nZ$tTX_K@Si%in~rmzHY#Lx~pPg
z$*0|j+RjLs6L$eQ6h_Tt2q@7KdgtS#k->U&u(Z?eb5+_3u)5z5u)}8uMFfG$&M1Dl
zfg)QZgMZ&sq(uZ3df~O1A-G@X6*+-O_r%5or{eF4jWNQL-inO{q2$~`<k<Eon_IKJ
zGO}WO%6eg8cIrXvX`Am-UJ0DB7J^7n_FzK603t!#%!wemmH2F%snK%^BBZm~nPae$
zK&p5g`O=}zFt(p4{bXe94U4DIMxb>Kbt%1%=znLcROf<0w%P0rLR^eq;3<z&W<5s4
zZDP|Y#l$!d+Wa`j<UM@4LhKEYkd81!vZoYkgG-67MHk1Erm7%Fxa`Euth%%mu%kdc
zkBJwvtH2@0T+1e<ND%mOU|2fAw}}|$Br!ik#Bz03{P~P9knq;BqI6Np8k|7z>u;aJ
zQh#28%S$XTf$5T8KJC@sixEsq;Xn-rkQ1VsG*DC75~x#<Nr{_0r+av0Q+fo&k6YJK
zTbDqWr|_hT%gFi=%;Wm{w(9K$T_BKJFu`3I%wz%ic3Ggw5)r+|1kH%7<iZUYC$-~Y
zO~#p!>v)zq0krg{>CQa=FG;kSxNRy{SAPuF%rc(+dAfFjaF(ZoiQ#@VQIARZ1?A7K
zoDhXm_2i>NcG&@(bJC@*rF7|V!o2FdvnNvVad!s7GWwo1__B~OFg`h0b0w4V*YU_&
z<W|VL&UOjDk&4?jE&4=BHb(TEKXsIwr@aPd(4sfwZ5p(&q&DDIS12VEJJ@bTtbbfh
zs!VGf$ziBi+M>Q)td@%^OD7D~yn(A?#GIYP5a^QPggTkA=8D;dw<+(uXo=)G@1Bv~
z0P$EfGNUMr<K!jfPq{`9Lvh*2H!Wz$B^)%nS*uLXY^-j4kvFh{&zZJv{)6?|ITcQw
zFQ<!ZpN+0~Ew0x%%EAn+AW?SN(SL2bX-X38U@h5P;`WM0u+L;8ogsl+j-=e5Dq=SC
z;OfZa_&KB;mQ-MxqGcUL&w0uy6b24d)kP}1L><+cTQkj6<L+&i$@}GeHr279pVdg7
zT-yRH^JJZ{HOS~9?n@M+u(s~FGBa&WWX(19(AGJ@f4|f@4cvRFq0%U{)_<K?J39xL
zG%HkO+V;e2HWH2ZsMKL{tT;v`BB}DG=ZxS}LaI2@y+K@i64L2Dh;v<PHGD|74ao=~
zykobmqj<+|TSxgyx2>D~LEScFB~~BaZR>WuvD?;Bd-rZzx8EJQZ9T=?yKOzi^=?~_
z;M==xcf=-l+wO=>?zU~jCVzL^dJB;gO&aBxHK*L)i9+_T>ODG9TXLe%XW7#=8A-p=
z_<2hoW_GswTL1`3=uvBGvE>nqbA7l}&EL+11fk7E9%z{dTH)cN8Ufmn0B?3Q0&;-4
z_-QdD>&nqMH7>1?cu~t}@RX}`=$yA|X>?V$mik_lsc@k9R9saFm48EAP~Kp^0Jm~=
zEP45iu?xAK3ayZiucvd(ft5d!KCI=zHO1bY!6~OXX8JyZ;|kgXUdiCP`-ZlRavWde
zI~;y$%b47;F_~0ZhHd(&U2%J2rz{CHsuydeH4)$rrE0fgtt-Bm_}$X3P{idr!q#>L
zDqqwY?p7-Shh8u<R)1vvpMjVB-Q<dX=uS8EL@$y~bX}1&bz8|5K4D?cV>u?uwQo=;
zBzqXeW7`gI(n-nHJ!+X}hwe!FYh3GRz)d>}b7<vV-QSj(=@UuUo<7&NCY6NIC!I6R
zu81BTv2R_8i1;+?F1-81Pwj=u`zc0IQqF{}GrK&6Z*yYJvVZNEfsdK9<k^nXHpELo
zJsV*$SQ?ktH51F7lW~4i9rXQ!wBoyCSnn64!CT=`-vRw=!5FOP*k^AJ#+P=#VA_()
zQ)|23KPcn^vcft0K_M5AlM8d>Q7*;RNtV-v*5^0W(n0(gyKA8|*UH_B)lj;ps<|b^
zLZ~{t-}gme+JEnLE1tb|U`s=9$*bpCwF~dkSBJ%3^4!yp8mln!Xm;`z>(dpr0$BG3
z88?rdPfeZvVjj0@+~Ly+urYMRCjNYmb;2UW+Th(wd>_Z#wzf4{7bvY_F=@xHco%t|
za0cU6_JXF^T-WQ_>9q!Mj$LWaC|8+FIMtx}Zez{!H5U^4___T3;+$oVruT%o{?6|B
zf9gxA*O!4cEE57VIG1sa7!#LB<N_&|jxI9`e^`by<nRG34A|Z{KyMB9R`gUfIpxs&
z|3imQNt8rO&U%X;WLXam=lRVHDS7yQkcTf%=eMU%SEUbDI0B1@w{HjF2a<n5@=Zer
z3m}Yw!`qL;A1_j(;p5DQmQpD`EdO_WrsJ~$$8P}f!A}8>&sIYi!|y0eW0gr`6u*TR
ze?9}BPvJg~_tMBff$70HP&xkj_S=)p%u3-+_!~~`)#taTf1WDqs00~*Fh&V!-Qlkv
zPvL<E62|%i7~z#U{QUdy?dkWH>R2%jSTWKVqKRVA2qMSSy-?v%QjBK7G>fgT`Jj9q
z?tkWk6+nJ?hge9>hdk$ZD7~ZYG3YG3P)0VL<@NY%WW37*hzAgBo+xQYXy&W_{QQ1s
z?tJ}H6@!{Jz*04=!)rJiU9<kAkbwqLqB^<5R6kw4Tv8B{7U0`SRA8Nrt=Z@lbODnI
z*O%@OEg64N+E|@E3QA`i1xj{Kg<p~Kl1;GWLaTvQyi<{otf!(MI%-`^Y}4Fq;uun_
zt7m2DCA&oM7#HrCmw}(CS6TckJhkDeN!l(R7sn44C^W(wzFbAugW(F)Sr;9RM-bIn
z8eI{M960K;QgcU$BL*i%EK-|vO!*5^$2%}Q4Re1aUToyBQt!rKeT(nXrCyyzI=91I
z`c!3|>wFf+l^0)i6Fg&Pb3RL>D`vP6f9C+8c0|2;qt23C^Gwkv;u|U*g--Q@!hd@6
zl69GUX)H~8zRM;N*n@02V%Pm!GT5jIJXVMs<69=S5-j8eaM(A*0yn1WYc7#au~Y0D
zJzswgu}*e8mas$VmPe(xe4Yc+lWCMC#2Il05#?9y5qY{4%TBCl>XpsXs1btMMG3KQ
z{Ru+&f`dmN@G)Cj3KmW|1~^5=nV1SBrlhd=<m;E4-a*cBk}ws|GZ$1)*hC(u86u{e
z)k1{~3ZwPJ;m$cWMmacI&}DhC#7g#ZEFOQ6|I7S8;Z4lI*S_-?X0#TlU9Fm4lW1(M
zSTzCsDuaoc8IL0dmjeow72fNH!PB)y6Z#{mm%cw&zt~xjk#vH+t}VzAg$La)$lCLU
z0*ycm@(*x`q<cFIi#6N?1MVozKxnir(+j29Fv7<=wi$srTTr$ejSwqbx2_lp@(F*+
z>9$H*@vuG8F**qJW`}gjXyGL7l}^og{lzMciEz0)RB0JMLym-kuBS{N8PqVy&WBT^
zl1^rKd+10~00&4rYquguO%!EURgt5irsYgCMbQ;pKo{|#nK{np7xLqKNsI*Cqd_|_
zHBu0D(e9RXipvLLSYDWABpr^b+?0PC>+_Y|dBZM4zP?4NUqN(Jj^^0XG`HLwkJhxK
zZAe<UWcyK-Reh|>4M`B*%{9F*OA&Hrm@#y3+R=nuYdmTlfORjlrH@K8W*9yL!O|w#
z0<A4zMv}9Pu#sdhBdksM5eSW1fci*;=7P7K^kV?p@~vj(3eW_P0%#5%n5lpH-2vJN
z$3e6(%SgI0K-W8q(w9410Ge<ErEPK72Wn%5QZobYA-sZ?z=X*!xkeHYw&Qn=Dqyoa
zeJ6f{wgS0&Yp6W1pHyCp0fgV7om9rhq|ulu3ZEAi%Ti%HyG=e&sWXf?mh47<xNF80
zQ#j&=ubuIO8a~C<yW{HaV;X-xg&;uRVX~B316Y(ThaJ5QI3obt$vdqLu&WxJvD;mY
zzLj5S5Hl=|ha_lZ%=Rg3V+KRmziGhk$Q7O6A)u08X#XGqQwdXLjKxDKOHM=S6$!a2
z`k<5h3k_sLJHU+#_FWVRITc%iLwZa87#Pl#(Ry%mBw5!|%0)q!1<!wWp0kibplpxG
za4fh_w~17?rJ9yM?R5FG#~d<NX|_$v-Rof6psbBb3RI+9R=1scO!S~jF3MfSy_qM3
z%_!^IlCvaS^y39v^geG=<bS@uq&&P;HdzwAIlu2$;~}G}Q>|P(GtkWn)`cZ5*UnNG
z=1bw8%Zx#jWsyE<rfz>n$h`c0tqz|W)l0R2nJnKH&F1ZpOzkw-u582PC!-oL({pJ{
zVe95q=00!AjrgJoKYAuvZVmH>&!PvCWnHRq-1gw}frS?;AJfD3%4>HOfdImze^CF9
z;9{+MXVswZsTd#@fz8a_l7S6lbvxqT9q_n&9&+b3W!{BSVgG*uwo{ynV24oXQRtce
z13z0+dLzUAsU4VkXSVOfcXkVt@1#BR1)v=71Det9n;8O^Y+V>A>xCsP*|JZ3d-buK
zV3l;jYTe3BzMDuD2c9>!9X0GPukFS*Ul70?O!LB2TC!Qe({67Kj9?t!Eo=SV<$xr}
zS+W`RK(qv7gjRp_NVEjy7%S*OXbGB|7lQu)Et{D)pk*`SfoOSG<A{!d56uq2&>LaZ
zOt1)<zue|A46|%IxKY6+fEKyARBjdl9NrzkIsO%x{~3T!=C2&N;o}vOeHiD$NOGgy
zNI&i3<m-+OPH7_ycLEK?TM)nL12DLwhQ7J+#aP<%dIM?m4G1vUj#}yTxI_tURB@}(
zGJa#Ua}l`Tb%!rfJvL<bgXm;4#;5OXTLie>7jqv*j#_4Jd|O5NTY=QJO~aw@tmqkF
z{QJvJb1c3oyH+Zq-ISRJCjRgl>#!8ICGR(9&h`HSqRB*3mw`1b69F@m(L)oLX@4y%
zmpf}c375VWEsg=wmrWNfVSgEug;PeBZGV1zo4S!NpJ%x|B$ISI!^e27RMjYgR^&`i
z5fxtfNq3%kB{3aG8RTw?aas6f%JcYBHl)7D+;EEajZgY=dbAqliNno;q4x^W-XItT
zs~io234*jfH%_9M&kGTKd=oid3lSwKJa1v)s)tmV5h)1GL&Ot|d4DRqx~xp*Q}PXl
z=q2tczsbB+znK99^lY|((`H5!*>d4xrYY-2^R?MY3+K9H8x3VX&$K4~z9gQaHi`ob
za+CJT8Pm&M?0!vKB;FUS>!}iFds;DB8yTv-u7u$XOk99@rK_Z$uq%&2RMy*pGqWp?
zL8pYbo=T!#Mv|DFynjH6IIgXg4H~l-Y`*#O`6_=6XQDxrz*F^Ibx_f5o8@NY2E7;C
z*^BoW*N3pCRlmR!F5W{4Z<jf2q_Ado_1t!K;$S>yl#h8Q*)U}kfVXmzf|A}hABr*|
zbi91H*os69C#I*eGt4q3yq2Att#8arEReARnJzC03<H?`ynoc+59BGPAZz&h89C=K
z)=ACDzbg6V_Q;P$MncczUgW2!ra`RTl%{(=Q<nJoh|ijoIPtP3+E%Hc33ZMof~&Ap
zKS*4o6>96;bBb8yX%!Lk$RN#<lI`lL8qv>LBdy;eeXi3CMG9Q)oaRH@6^m@u2XwiA
ztom+r&vmax3x9=vmXQljezk|KS+k9t`;i1aTd9i;B`SK1*pkZ=Q^wpdx)WoKH7aP|
zTXyCsJI4VntmE%Cj#yliEcCnSbFQNrX6_|j#jOcaK$_+=RO~|ckqVrbC`A1(WrdvI
z`Gi=}#5LA7+~$m35Fh8(2or0VMZ_KCawIWft^$`nlYcA08okY;CbR7=7To8jfqK_x
z+l=Nfd4&5F>Y7lWKb?r57S&p=O#5oB=V`YxLr?<ZULCp|zs@<i;wurJww1M83`t4I
zrr&roQe}PArQIS@^Ew%&xRekv+cPtrd_>z#Tk&<~2h>=IVZXvO46v-HAiFG_m-(@7
zmP9af$bVgw>Pukc5ndLuTu)oely#(hKL&_+Op?P@C+Z#`u8Y(IC#C;;V240~q23(F
z5b~>PZDWkm=<Zl+3zQYWSZhOv*$kf}bu@*b93HEq_jb7!%_WLUIlA8EohoFqs00u1
zR3Vd<11%jiP3_*)F3;^<9Osk;=y%=g;&`h7dw+txiWkk?gu33qsojv=0a?_<H`MQr
z#ZY*IFz6*mxIuihZ>Yyf&IIh}ULBY#-m>l}jDfhOjqQ$wKS9b^%@Vm1N<7-RG9NtW
z{;`d~C~IyAZ8y6wdRESaU(|TTfD)Veo!zzD^>ugc2=3E!y^G6AK$k>xT%g*X?In7P
z%YP!XBr*7ZxGZ=2uf<PIXU6^y-5ANKfcI<%B8Yv#?>W|1S(v&;Tn_8CJT&DP=?|4M
z5J*|XTT7Ya+~i)>lA&vD!5}B7o}qWDdca6B72|hdlSp|<ZuhQwKx!;YceZv6w@ue?
z?hSfnB~f*Z?)C<q0#Q{*f#uuAAXYuJY=89!FRL>n3Q#HF1PsY0)#*m4|4jF@TlGNq
zv!fupJ}D)>KKxvIh`I|$c#5G)$hA)0sJ!ZG<Vy=V0|-%07pO-$s#i5{>H^p{q+?<-
z!(ugS9ziwn{N;9$Wn1I2WQ>Js-}K24#I+gB;q0fw(Ci3LbBnBfbM{?Ef|ds=MSp8`
z4J(4arMr>ohu(x9oubTr{XoCp+LC%9=E+WTo4Mx72S{g?T*C_38k|@6mV(ZYdtXP#
zD<2L~a{aJcYx`%-rnQ^ycCC_IifhZPn(%8|w`Qx}*580tEq4w#HfA0T(xP=YG;{5I
zTv~O<3i9o4vpCmsZrPv@b91$6<~7o24(o5lS=C{99i3#vmfY*=7IX@}18i&Cj`eS`
zWBaF@!RO|G6gb#K>k0@dQ2qnwFHOLgfi)}>0y8j|F}*<~m;C273j#7Wm+>zOD1TZ}
zZyU!Ee)q3fATQ~I_I74xW_OFgfD>0~3I#AM7ien;0Y%Y?5?NFwDscMi`^`~DN~K&=
zB5g5nsNr#LcXoDWzHj!<xTTPSaihQ@E^v^NUBE{&7fL9(xlltXC8$vemHI0!RHI-y
zrO;A#fkG2~27yDX#dRc5VrXZ|fqzKZB@4t_T^~{=aj67S@n2I7WJO(SfogI+N`X`M
zYXnQlDyAR}?DIfkNd2#zv5tOZu7TBjWSGk8%gKyUD!ak>$VuEP){q&?f@C54ppXhA
zEtrEDQuFaJeAfQnTthbQ?`V*U&yiEeHTkuSk((r!;L23ozMO+_>h($x|9^lakV?WF
zN<q|JofX%N{0_l2Yi<i%BCYqI<N&q)R|!h<Up2sKzBaL-sy>SN9#ywlYS2c8Iee)}
z-71ALRlmc(a=#+e`Je`ag2`MDLn-?;g}Fst&mCLm_qI?{?<-58uw=}ENGQ4YB#guj
zK8j_;jQAafFlxQ8tcIGB8-I!5t4Z8aLnNiz{2C*;a&MJD@>t9j*RU0s1Iwzyen+8D
zm3<DlxRySypj>Npt1wYZ9-|q7)aFrFEHtj}AsE@_ds%8|rv1tr0=71vBOzj`cWU+E
z&FUj!ScrWXVGK&r#~VT%B1TqsF(oCpt&)r8#vyhnnfpwVk5H-`)qh-}_KAY^5-#^w
zJwcd!3eOty-G!*0HuKvsF;(sdgOQwkYXmt55x!)`Fg^hLRd&%GF33#6M%%M*zUkil
zaWUy$oS#o;q5F1vc`~_t>lGj!b${)C=)S`jjCRz$o{VR2aZ^<IRFxt;2rouB5MOId
zOh?bU!|UUjk9Tl-{(oEdVln!7b~-)p4!gg<{=xtM`j^@4;_A;`_w(uOm+Rv=p1$vX
zy?%eOcQX0C``h)8ui~#~-SOG<xU<O^^`smN3X0Za*gP7KCnz%~C*yQnSry4Lrqo?s
zj=Sl_?DYNVe<qhdou1Dom+wa72?qM=z*5@7NBPt9=g;u(k$-DO6hvg*ux9LeEbYnH
zd~@el2gsAX11<HwMcy&-{uR^wOTS_+HG1GM=P1Yr<W8)8ZOt)RJL+Bp41d^tKYcUp
zUWTtx(NPG!d?1ONk!pe!n;L=1az=Q!98p`imgrR&H9u0N6&na;M+ySHRzXE7)mm8?
z@$xNDM=|*8?tdTu{FkR+tY)ZE6wa^D&W=8!RiK4@s~Q^;aShEvu>c=Ft*N<2E9;Dj
zf*U<^Kdp&+8b#L9RQRym)(qWu)3fR2;RRAAWWOPFuSPQ@(Rtwc*S~z2eSbI`%_f27
z7xt&;Gt9A{rLSE5Ec>WAq`>F?ix)irfq!}aynA&y9e*EAW^cP!FZa7QlMl0_TM_bV
z^z#HZAZ^cQSB}v0@m^1^rq`F_$<-Xv{nC$<lhe_+(+_>*I5d{VN4WnIcLKdoeqF1N
zxT`s_=%;`7wx<?udCx3vggOZAMxZnZ{W6apT_F*t3STcuBqijIJ$1RbI_ZlA@^$Y3
ztZpmCf`21`F1Hp1y8*@$D~N-n76G@bd5Q!8#Xvg0E&xR*VaC>gvRf<`U=0!2JroNJ
z9>E+QRxBQ~+bjk=T!I55!&2Ur&8`{9khZhg5}WfL;pjXCoNLYl2$%D!p9Nh3h=GAo
zfrNr!iEa@Y2+mK--S*8$O&<&m-i8b%5*+d_Dr~JX-YS2Fcgo~LD<iK~Mt)Rf<hrbk
zG?YIc${%l7{z{1roPAOLDqt+{N=BF<Bg}L|LMSboGwFd5szs3eCtu@ab~Vp?an`%p
z8qqL-3NB%cxWrOyXVzH2?5i!;=;_RP4W@TjYtD<1tWg0UpVn-Govyn~pt7x$i7byP
z6Zuw|;Gus^<e^Mt8U!AMd{5zcV1*}C4|EaU`$un&8Kk11<yYXjFt1ejl2evcpkC3x
zTS-*50W*vSPyN&-76Z&QU>%@BU_@?Z)d(}En2AyM1l&rCZEgfwdE{RZA#`A1Rie=k
zmBjn+v`TDpZqlSM7;Y8R;LKaFAseC{pf8NcQ4@azY2s}TNF*rkm$88i3ocG8f$>hT
zkk|x@E(HJR!yMr`o%IPJzuVX9?4>;>v6WX6Thn6_TV<KpGNiE#X*@T3eh7FY=pYPj
zlniZ@3~iLO%~(UT$W#|-gJg;YO~ezU&HK^#ayrKwYNhNQpoiH8n?^;B6%m8EVnd@U
znWBH6dn&SA<qMy)C^D+gEYx1lawUysk1a6|U{Ni{R&257l2VjXc7!SB#Cr#swsEQk
zgo+Kj3xtgRh(JIWg`7rW15#aav^573YN+*s|4P|r?*zaEqS|7&(E|ZlKl^4AFtm7$
z9Oz`Lk@K#ciXb1c*_yVP+JuxAf7o-A+Sq^4{ut;BcgkQ9C@PEH269F!^}(pA^a*CY
zbKSD7(qMZO4K8VkmO22S4gjbF0P4VyZD7bYK&uVVYI!q|G{dGYJL|%TG}@IEGD8Yg
z-bxBp0D~@osi&LtQC#}0;3x%jQx^FnRnY~@=f_4c?^Kt32vHcdwp}?_^v7yKx`lte
zCeujXW1+7he|F#ev>bmeMd02&<r279RsEy3l&V$PLX$C^=6d91N2h-Fz=DS7o}+B^
znXC+lX*NZ7L^*ACij2QSmQ;wjMV2S0LgUfd+3{%n+YJMzy#un%PL+y+Y%JGV34eBM
znHjO2%iV(|`#(+N|0*n>wV^?kbrgTIq1>&Z$f?EJ>KdBVP%8P09#eJuXeG%vcm75E
zOL+`(*Svz<l^%oK_0|Da8Nj#<U|fbiw+wx58T#BZz<e2Cz6>y52AHpDGc2zq`b#bg
zFz+9B?^Y(L?j6`R+Fg<s$<K{|qT(-Yv7y{GplOM0tIda2D_l}bRZ(Uof$4v20nru6
zf-N45$Vr@#f7A%lW`fo&iJ=n%vbVg2vqg7>ze*&2yzXz>G2+0Fv(P4x<S*q4n(xPr
z5jg>bDSy<6HC^$r*58EhwEjNv6Nmrrqs!CL@!8}C2)F?Wa@)q!Qv`fMd19LfK39)L
zKXL)1UuBa7np={9-=Gw^vs-`D&G-d)O|46RvnZl)*ZghOk$1yqb51Dz3FqRVp-`;+
z<dt>w?%hw54;PpHN<r=R4v>T!aw&<o4pb0PRi3;V{P*?fWVucf>?o<+B~z8xc}jOj
z>*kFptXNxHM^nL+@*0X@7Ljx<O{=Jxvdc#vt*sEJRvtl|{O!~QacX}9acTo`Y6Ed<
z1955tacZ^^DKZl5#_}Nmr)ausF|Ezppl#41u^byVLxFzHA9{id_{oyLF;?uJ-ZzO<
z$!e5E3Yy^UJYc(VKHw%Yfsq1}{AGD*z>IrNTfjsz`d=8T*c=d&&8UEgZ}yT2h(GAv
z*%yl8_PqnCb;I{Z2xLqD)=6GQpeOKvUBakrX!rA`Z^|%w+dI4?1DtucGMYW3`F2J-
z1ho2zmj9N*;x*tzF{iwSX5ap0UPBWLim9xnNhPW*X8$kc>y4LT3nmi+HZ+$Z$O03U
zkp(H2Fflp^fBP%?2h6Zo#SIVy*cv-PZW;I%<P>D)G>6RpA5vVKWG|8|dEkT85|>(D
zZCqMT-zPc!{_y<v@Y^fWWQEgU)%5ml0`8Ff4U!v;o-BYc2-Dk->92#7%JAozKO`k0
ze=PrY{M5%s1ji#l^M`u@j*pfg%;7Pp(p)rY4)Q2<f8r<b^C`6Rd@qd*6qp_`4=~5S
z-~M=zSy(B&2|vSQ{P*(q@ZUofU7^s%PewYS98LfHcnBQ?S{UnDBVk#gpZ`wZ9{wDu
zw!k=GV5Bixcfcqkw04s*USQ}-zz{2@SxL<b>F}GC)T|^i6m-h%+(tZxxLu#p@6aB`
zaFZlfe_TEv17_iGAV)Se(<|fgC4B<+k;KR*|HsALNK(E=IEx;N4UAq*Mmv6*qQ@vt
zVTvuYtPeeHQza`1ZFL*pP_jvTjfsPP;r6T^x=i`hp!rf%PRML%t$4b2N&D!TCiTTJ
zuURR*6xeD(h6P3=$hsmh`(nbba8qe^E9^Ntf6Nj^!S{sXrcJ?=78-mNOy*!Zg%Dc`
z(Fx#_01?sB-ymy}U~9~a8G$&IDsOY46?bJ!RvdlSWHL6(b81*^IYkY*b;oF;E~J}i
zP9If$G%{KQGJ$~0bMM!TO-w$k>0vWz=FG#O3r(uUEz*|hIa~ht$haJei-mu1XykFV
ze_zh<gq{@MWjqypgD%-|gO=pz5UGF-G&r-koT(x$jMiu*h&-vZM#`tPF)X)=at3}+
z0r@f_&W8HW#j#)xX^tv$s@@R|zmUOI{ugD{mRgbpawbdG9i$6(&DninL9$kO?iwtp
ze4e##cXCQt>Bd<gXDLDM^Yt9%sFt3yf6G@(6qY@6nAM@Sj#wM))oYCt+Rc^VsaRvS
z9$RKl3$FdK#153spa)!mXq;1Qi4L+SC131h;$AiQO&Z@9gVI`jFU^{>RPFC_TBWX%
zkIrLyUUVkQhDb#3s{h5v-9r%#3UF>j<cu$HLKYtOVO$E|gP>MHfDVLpz;%<qf5Xa;
z3&$Z9iJDkVoxPFFGMs7bs)K6Nu0=~4s&!iBbyJ}a%^Ha5E9;w-!>iUq)-=p>rG;La
z)bUiuH5$(?El#S#X=~QLv(!<uqZCRN->&5}Q)+myEvq}5I=(?G_T=R3%A&7BX{nt&
z%uAOfX{4rm&Y6~<lfON*>7(QEe>8hXEQuAxh*ApHRX-r&G&3iv)*BPQEe`{?S}seE
z%+f%{BX}|?Dzct>irUpwx)sWboUr6ZiO(cNrM2)D##tAp4`op(T<|=TK2#`t#*DMV
zIxHrzglmVi%ZrOx-k;m`Tphpbb1l^_<D~wQanipf=L?qxiAVKJ+sa&ff6a_;yKKmu
zbn9DdK5iC19(`wJ0eODciUfkdj=NUyN^schZ(F61JSTl=IsN=T4J}{u*4E3tt*vHJ
z*k7vO<zgCR^pucN(Cy7PH_M!{5&cMUxx39+QINBIMe(~97Qa$<5YHzD%1U-})zMxG
zv^0*!m0fno0de>7l4Uv%f9Y8~B?tg_1)gCU#pD4e2#g}vuI1|mnD<;2=4?mvCcb3C
zoPF1k<9whNIs}x~&M4;1m5w-O3^h*kSJb9!Y3-UZ<M0wNJIwO{&+SGbUMQ(%&HQVO
zGH?w>xiFEf8AtK@AJC3e&l{)ZnOaqY@)cpIj1tB#^r8{DYbYAwf7vGnilE<5F`uN2
z*vmz_xQeP{YRW9+q;*Si`lf}`-4Y+#3gvLQ<trwJFSA*tv{e4F%K6tI>Ro~gwLqiq
zSUE<DcR6NF=b_tCl8j<8#diFPaC9weKHCc$$j9wOi8Q_L8^x1`*{8b|AOaLb_MT=L
zD1L6U^uk87n9I#Ff6TriW@ShM?vuqVhQp(K`o=);Q@chl>y@4{Os>^6o_G87372ER
znDMk#x7D^edbzghB3WUb9RvVal@_%UjOZ{p%0=1M`8EQ}MMqis@^&P<lG1s-)3)q5
z7dj)P9VHAG5`j@PAm!7da5ZEva<H%$atn`5*9CXWm8hLCe;V%@4+F)|9T2^+*G9xJ
zd%kLEUCn+}SQJn1rPz!MBa~mr{Lz6CcA~hD@z%&tY)@EF?Ly=FdKJ=)@pRXEwSH{D
zV58$^wuv(lO{ubsYd!NuYx4>YxoR6*7gO3aXLh=tnD*F4i*3FM5quonyH=t)hjK_J
z^R4p!RZ8J%e>|u9di^`zas%8dNxD0=`-g)-YydY1^Ym)Up_uYVjfeevkoO*lDI_78
z>2M&Xl<+`j&clZ11+R5*?dC<exu2EusdC1-<njDELAsN^e%Fmb`Gv6Q!hAhJhf_sm
z#~JEE^}74>(+xM{N^(t{7t^SZL>C<p_lLiFxZctEe-JCy%6f}`{ZVloZRZqBRF8dq
zbANdAk+d7Hul#t=+FtCG%L{ubD}w#IWC_B17_lDwZSyW0j0jD`>>9oIPu0MQ)q8)(
zR1JnHOaHDQZakgT^5Q41%1+1ff<bk_=ZmsAkmfp#cgyArVUP1$8?C$>!fstZDfWO-
z*k6?Ge?fXdK))#4Lm>YFev-_v3jNlq{rVyhUYj62=&)A<@)EzFaQ)ka-9hrYYrSXH
zez6{CIM-+xCfu@buqcqdYmi`L+h8p_>gYO`7E-n554RLl2OJ-8T|Ff_^A?YfYlWWl
z^-G2QB*XF}Cz+l}zVzL_-bNIJqv9L!w4LI-JkUJ5CsN~o<+X*VGBztXJFYbB{2t%#
z+daw&t<=c+x)dE^1~`B9L_V-a8{PmYx}rpnPfqs#0|Dk@pqGI)EE57WGM90`2NRc^
zFfGOb&$nYSEoTS>dTPwTmw`1b69P0emvN046PNZmEx~_q<mcg^U;n<7nOP}<<v;l`
zzWw%f_vfy%uCy>VOvHqcB*M2}JN^Q}2<yV6ToA!bfBk#<zWY2<ode^6fRWZ3(*dKb
zFvd^T2BG<@1Pp$Z(h%eLMjsLkF)=TYmepT>{g{T4FP|FUNRaE1u!0VSI@wsEbr@$8
ztkMwHNEm-(WaoA^JVecpe8_b^f)v`}I9o6Hpn+r3HN#{~{@yTLH}j|v1S&V~3hP)Y
zB+ZjHM;>W^LyQ`iRL0oF=f}h|^C;;Zev4EW20-dU?cTuvdcv0spr37QW}byo`TdmH
z;&Djdu`<k>l1}TwDkBVpcM!HGJP0evV`(Mm350*l1iO(hD}W3XS;K)@Od3t>X3LA9
zY}hpls}_Q9!uED}v67>mIo@u~+6izgSS-BmHXorlr|n1;^tz826tDnq%NhVF!d8<z
zu1+#R35)vqCNmIzg2{{&$z;$bB}ad;$!DVZU=9+|kJh1-nl^JI)yt;CdvfYbBkE3L
zvb29|SjzOEF*zg9s*x&~bp$BII#?pGZcT;~j`dg#C8BV<acBze3Px)1aoxSMDfN@@
zfV-MdN69<58)~p9Ji3J->rht^4MA@0J9t}9_!-_dP<)cN^;EC)HuEfR>vxB@=Xl#d
z70-LyfWU^gp%-4d&fCoW-qusQ;BEcnN4$TnqvV}@+}j30@VvLt)BATQhX%qIYQJH&
zxlqtbyt(c(5=C(E4ubZCF9|xZQNo4H0+)i<r=oWrPB6^a&7ma$ae)0zSfJNJ8hkz$
z1K~x8mAumMuM%Rr5hD2EEy&eoHRCt3&g_P6m(6yUdM1!+LE*J;Gh%h^W`X;ZD(ru`
zX4XwFYN5qoZXDxb!5vMZ1EJ(XPg_l4j<F;77feUzXBF6|jK#unvnX%pTN@<^#04jE
zv+33xw4@nVJb;~cjw<?VvD17+p!6Dcynj5@A?(y!I^<FL(1#6$$KxEI=ege3oTe7t
zPUr}eU^;_6<!)A8Y{5((4Xu$cy)S>-7A9hRWIgITY%S&7fzJ(nDer|+OR4NEW}mJM
zT?;3Q4zw1w3X~S8Kx*BZZXP;647dv7VpvazNQP=e*UFHh9(P?-Dj}K55Cy)%pA{aJ
zYFiYl2Oy{Hv>O3lk(x|Rr_t4YrjS!;QbRQ~rlopXS_%*3%;1E6w535Uj5B{dt*sJ^
zg9vpC-Q0UBEK6z(x})=0H<R;6C&NZBQ?Qd{BM6U{s0F>A+9mw@*^eODQF^RK%@_`n
z;JNehI!Y$2Jq=SV@Rr-NK&0Y+tKWR7Ubnmx$Or2b%jdppO2{gq)-F!fax~byWzdL1
z&AQjMz{u2(t(qvUV)W?}FiU^D!jhyT7n|~w-IWo(oww$?E@o5BIq*@2QiL9PTEBNu
zgfaU-$J6T@Uyp!`A?{%GK7$P{FU)r>S^L%t6HZeW7WXAe1i<t6CFApo`x6zYhVkwl
zR1UfQ?}xhr0W!hYtz36UL6FMHK`i*V8F3U9Pu$ZCD^xz4UL1pz0`z};>GQZGnxc)Q
zEO@;nrFeXXB&Rjgtx;s-uCmy#@xm!p^S3!5?pvRd;?p7CQW`kCpq9}u`7gCFK^}$N
zR8y~1-k8^8jBj~_3oyICWT<tw>!Xw6`c2wVs#OXbu_TZOOBg9!l4!$*w)oUk$yvMH
zWsAx>uWUn>df|~us<3~~qVKLkAC}l%p<&fUi#rjj_K8+F+w$ghQE1cZtrY-zq?&{O
zVsEN#y>6`<mV^?UAm8Lp9F*Jqh~u&4_gF>OZ)0Wb>l<d?3;%6i$P~<cZq(DMjPz-@
zaCfbDWm!t3QGJhtuXW$S0`e}z1?)RY%?i%ieh9+A2yV+;XBB^lKnkRA*{ueBR4p7b
zyIG->u}7T(dO>w7C+t5bijtqE7%C}x>fp`vzln!w5fwQ$9?Dbk2!W&j?!zBHzW?pp
zzd+D}|MCjT_tg79@YApKYKdcol)HbX_gM2O+EfJUj7T7=mmTN!vaSw%n|8F1Am>rD
zX+L>X3+GA8v&Da0WO!7%(r+V`gGjK9mhFDd2MbeehbU3yg5Q`*UdykvzO3R`{><OY
zR9f|)Y4?5FeXeK*O+WFTgQD{an6xSN-Y^HW4)K*Ny3Bp7dBeuwT+BSCNLs#5L!go@
zILFG!bgQIgWF~DHB2tAy87j|ld_(#{(1b^BpQz`VICg(S{D&LJkS18$-`{@aWj|=q
zr`a=+dEt8nAWAQadL2(jOSED!$wsph1M~DUyJY^?gEsqi=O1H0iv=k@rkzV|F?X+l
zD6#N)nz0x~eQC0Oz7`%*k}-G%JY#{?U;)14A+sy6Y@~)f9jL^#&~DmFGDVZqPYjf<
zjMPJz*BpP$UvnBnyDtFN)h^ZPTr8zICLb=vdGk{5R4JJ9X7=rdBCaX_og(}HQke(v
zwl5ybQB|S5mBjK+v-GpZeMd>vqw%evkJ<=LrOJTr4TTvRM`6m36sl}E*NQ1=uA+<?
znI_>(m;GT>wgU&TM4NN!Q>H0Jh~#oehLuE2oy~vMg;qjRIXqK3Bzk?SKB!#Gm9;DS
z$?FD;hEbm^Ec3jOvwo9o`D|(@-Dt|r19of2q5#+Dd}`|h2i~44H*nMs?wjXy5}8%$
zp^jGklFwGOv8=_JUWfatg{|ZWGcUk9O)PE&Rzl}2Os#~@76()4{5}4@=cKAp&Lz>y
ztR^{XhY*_;IiEPH_U}$lILQKu)U}0k#~QbG51Fzj%K?4sDb=k_g$JrNe81V9$4|qV
z5ShvU0kS~ym$B3>6aq9gmm!)L6PF1?Ex~^@2e|itfBxf*CuTWw+F$yw{@3@<xBtBr
z)&(-9-9;<IfWzfKf4un{5){+MU4&o`<mJ!*UVeM~XHPZH)qMv=Ss+Dmt(9sNB{WmY
zUbIGr#$-iF5>%1Uij<^CQ0cFE9FwE~NkYm84SdYTqu@-^6y?vKzhAnMpFb9von(Jn
z3|zoB|7p~i-w+%bpy}W6#_&8SNZ-VqyhnL(!asa}|Dw4!vvlc68dauxTU6R0s!GTE
z;hv1|gmFKnPfL#Im>{$6OU@ZN1!vsT5Gqn_Mu_UyhvPFVqcqIeL0gzLb<grOGp_jz
zYR9-5cvjzIKm<x&gCC1_B-w>$zj%LM4(Y?0%faeoJC$Wg_Bkn&loVb#$fWsci%RL3
zbvW?IJ8K4U6_0`kip@w(QAq}Jhz_-?>uJPDi_D2ZLViyY&JxLLdd)dX7$yLVKdE;q
zuPJs+vWa>RiIwc|MQHSQrN5FxmdT?bMxFnqu9bAQ+@I&0WZ8jNfuvL_0Udv@t5oI^
z3#FNm0~MbEg3e8mNhW;?Wv%euK_80>qm75rpFh?)H;y?P>gL?caVkmUs(V#r_-mgB
z1K-$1#$0kVK&5olU$%><WNM2>os<#P*v`fb8h9*k1z!=3{Z*QwWZ3%I&L3QwGwVU7
zSs(RNm8HDYgC|pYCEE=3Z!>@66K(LdgM3`Twb%SIwrIb+k$pV=s$~Lkt!#TMBcs2j
zJB3$@G@ZB+d1jrWOJwPrJW<Ub^dITTa&=2jXPIC#(okDwH1@hY8ENPQLuY#`{I>I!
zBphRgx%Hln#C41i5B|{ltR-F-&E%2{?Vufzyh@e1NA%@URzw65x`uz~=)E%he)l5I
zfKE>HM-w~2E96bG%t_rlYNYNlv$<AjKxY&k7<wcoMH$V>DY{Y~y(bsJ9J1AXuz)cy
z4MB+dEz|revj<fssu|M71W}%d4!`Nl1dKrrnm7iGw_VjJg79XbtWtVh^lB1a+elZ+
zei^A#&7|x`YITws-`jr<i9(~4ju0fXM5?SJcu~Cj9`hW!u9k199z0r=rX%%2fWBJf
zkt{<+zMD9@PBh6askd>^EMIZM_-n|X(Sd0tcdat=BVr$Px;Ii7IU!n}*OhvZsooZZ
z7!3L`Sfp^SGlg|Lo!?z?JCmJMZ(5m%F<Sj}%Ff~`Te-=0u2O#n8R@i8xK&aAD6yn!
z&#$96Je%^1PyAANxPZN4;(L3FnPj6$5iMDA;AaPJq$oTAH<I)vJDbX(^{PXc=>Gbk
zO0}hXrI@4=m=^%`lH{QE&@e@LHdC0w95-7m2wv$rK2zt$ct2><rnfQ>Qgu}+^zXX>
zM@T`^nx1FhAqs!pAQihwp;KhGGmR09{I;jmR}!Bt1ZM64DtO|c4(#MtNfKG4pJ{WV
zw>AF@QK&m@Mas6Svjt4LG0DFKI?<BVYTIk8fZ2>IGN<b`c@O_6LEsa{nJWDNpVvL1
zm~wIfo^vhc#4K#d6qU{cr!wg|JW&=3_Ak;cA$@vA_WXb2bNUlVOMsEu5&ue>34R^&
zFkr<;lix&@O;w5#hm}4v*CD^#ZY*QURCtqSQwC33uj9&8*D60>7P-YrA-k9sd8%*q
zdxdh!#%X3;q$`r$E6vXrtO_tRtZI<rD+E_7agSUdjr9BSOhDY+_Ru8Isu!tgWIC^!
zRprP{UO9ibo^Dt*#9zC*t+{OTWOFB3>?qZzI5#!t1#?6?`*zMxE~}Hl>KvH1a%jRV
z2)(vRMENtDEI0+~HPNec3wKN_IdHwBdoGXV8p0k*=$?9RK}CBVMl!xVYSgk*<aL;B
z+^kLyp&F?bvta9GP|&n+z16}<hU}T=prKn^UDJP)HUaD@Z^_>gGP*?mM8i~b2152S
zoNbs|%}@?h3N{vJaYxc!){05X7?BZAy1(5F+>%>zT8L;I<o?){Q&RoZMh$XSjp3fh
zgJ^$~SL2vc1FwN{X#sEZ?JSd8^b|tnRVMrF!hK`6rY6q&08)9(Htm}%E3UF4sTZ<h
zTpxd(^<a+7dF`}3EV=QizKxKUIX6=6J#wwvGG^uzEy-^6rWf=bR&Y>j^k|36Nd}ai
zdwD*Rb{2j&{fL(F6zhY%+jje^AA6iVTB`NqbNralPUr}xB5-;WUKpTtfay2k!Ej0y
zu(JyvH}vwZ-m=A)M}DTA`T1UF-kFyOj9h=OsODQrkB=q6iDP@5pmgZCbQE8)e{Ufi
zx9BK$d!EZ~KWWd%{%Kq}U%>BSpFZ-muhLz85-34Dp-xUFC+dI~C|kA7$j<{RZzdIZ
zs8Ef1i>D-7V)G*y5edYj4eF!X=A~d;{sdW?k=m$9VMvt%4RrPJjF<131dV1Cuz!EE
zO}bgVQBG9mwF4;~)qU&qxh~!jm8MSwqR&@?4;rP(#{&o_sQRM=270QX=9;jU3Bu*Q
zCVs&K38Toao3Nmqq(}Hx=`CH6%;~D*^(9^_sYX;srhYv!cNF*WN)zP=w*E7fdi(#3
zA85k5Zs2nz%@<$n^z=(d%U)g%P|$zsQMp%Iv8A|zr=BUV9p#<!+VgSRKd2;sPyF-e
zV;49xe_`$L*~Tn85HGM_PjMfMK6BKun`JGAQdQ}me%6+QEYI6j7y{n{Uz(HWD4NWP
zmWEMjT4(sff?OmXXLmOGQ3ch@HZpQX%9_*z1WR7Hdd^tT!oXA@y~<amhrEBY>76Kn
zGbGRA7s;PJI7Upp%`NIHeJRS^s5|1fM_E(~_pb1|wYmBH8VWQ8v=e{RL1xet#oBFK
z;&lg?DGD?|KkLT9Ww4ro=wCQ{Zn=YxIVFjq{c{dHs?ZWTE=&f7y)*~UEYznJf9V`5
zASb9H>kZjSfQ*gq+)_Eb{ceA|28>FFPjabg8kS(k_RwbA%1}t4=%qP&0%T(S)gwqg
z0dh0U<hm^iK5u84^DQ~aB{LlQ8Bez0q@blJ4t%d1MFJ2^n6<+RUr@U79P4I^)<oA~
zPxqO0?7TL_j8b#pOZJBvC}OHC-82xhpXvH|)F7oTG+^04SVpHFAnkvg!b)C3Wv|uI
zzerc6=eqN!`ED(1X_Cf!M`JVcl(x&$ubSsbUpG=uy@nH@0NpIYYCT2o9}XGSI%+g{
zYZ|x9;U7Dm*v(tuA~)g%=Ks>ruicoFf`OHTti`W3ed^HWoCD?*3^SAlzd7s2TW$5m
zKIH(aOWGZ%Oe1zgI(rmbhL$kZvhTRFW7tC(f=d4ftR@+kk)j<F0XCOFRTmTkF*rAu
zVTLUymz_;55(9x<$Ct8AEhqtgm)lJ(Bmw4@PAw#VJ@Xm(d<x_Iyq87>3QP}p4LCjh
z_4fON%*;yRO}GrF@!!ka!@m!eb)^LvKcOMvz)gSucnCKTNEquU1t`GV>E}Pww}(GQ
zvUM@Z&!iuc)6ehIF!D8P&r8;xG=^xVJ!J%un~d>7hg(T|_(e)$>F|{gmL6G83T9G%
zes!>aN|B7v+K&`jjgnZm$U4PHIZ+cB_$)t>{n-e!Unx6?mL4c8IpeF&_EKnz<7~a)
zg9q4#lBI-~%+qz1UezaBf!glSr(~NmdC(Jth2pBSk*_!Ce7>q;Vacb}h}z0XnG?66
zITS`MWC&WKHSN7VeKIn{5B&l?DiQ?#qxCp{Up43vO_H2A4^X(aS*MsdiQ{%vpGs<8
z+>TVQ>ocq=_s5AsFT6J6dO+*kQOGjmWIFF()+k~8f)czgLUb2ZO9G8rP{nm&tQQuR
zs(#MWke=F4Tz59oC!6AcoP^0#|1-fhDlqV+UQxMXrKKxpWxb1o=<kmS@~mjo2j=R3
zkz(hK;Y|7ejUM}aSoHJ4!m1;y(m*j#i;+uNnp2Ku)1(gfE=ayROtcan7P7HFtva-}
z0<2zW;$S_a6HeMA^*pCYx>cZ*DKF^s{$M56mP1$bw$D^CSv1O0>R^^F=JDqtM(<de
z&(76^6VPMSASG=^E}~D>S{c?=$L9uriNaer(vowEYZlWlcVoGM`bHFhGXm5?xM0#$
zS)fx?D*D67iph{+j%2FESzH49@lffUl6-6pUUD*cPBViu$3vxaD+gF-wo=g-j_CfX
zxI%ow=aHO@aDWl_$mX<=gfn-v8LXJZ;|#!<#Pg{BdseneNZ;W0OK7m#f$~y+tJDI}
zN$q%8lW`^hI4@NW#I-vO(k<J=(qm~O(ALyIrqekU4ZO3B3!NzAuAz+s6VcekE-rRx
z8*x|^8@j>fuKQ~A{`j<37=Urr;L4X{ba)`N%vk)|b!~{*XvJnwq5y7GU5O6SV)mEa
zj0~>}6nv`qI9J@%K=dOQ(y(%WPAFx@RkZI*K*<_6rC|+PSX$_n_NQD8jzw7T7V!Co
zNp)~NpoL9uYK2k;Efx!Sr55(J2-w`-$-$+q6X+`tSF7&-^htrq#TC!}2;Rsudlo-I
zj}0`GOo67A&=PE%smqF|1}8A}PFopCS<cO{kmv?~-_nr=#jr#zDEjDsWhz#kKv5Ri
zmipO98Cx=FmO&{!uDD5MgKy8Ei58j&8*<wDQnysHGWnK4l9dYI(IU^)($AR>-Fb&>
zS%{j=OrGWD)~(R+OSN*u7^6oTjt=Zaua0<bYDla?U*;){sI45Hoe@uERum@Bsxzu^
zmDVl?FD`+;{+qdk%Ib-Ks+C*o#A|l2DBUawE5dzlDh{b#X`M|3xgdmdyA#9cT#R+m
z1yMO+Bo1aM&D)aAd6`@=cW$O=_nwMm@%F_4O*>yX#ceq}aS?JnH0K#$-M<JK?KC81
zJJlq#k~gyZ7|>uJbK5bypB*oos?Kwu&)9i?7bE%EvMNS{l+oUQ=(=mNYF^BmY#f;`
zwqECMKCIOEHvPQj@agW28**j#M!o1XxDvcumSHUEr0UUiq}}wIB`vk1)K^9i&Z$tD
z`Ea^4my6L$c-C;P=N@R_;kJ#8vCZ9$vKN{FA8Xk!MAa(p<NqpDt+)gk(jSbfhgGkm
z>fwR=QT6550b^o++mEuDPs_RQb^)~{p-}IV@QMsQ?tB;KREGOqeHC-A<u7|x*EKWU
zh%@slJ8-5U_InzoxWfZB@tZ=SI6n+4tl`eAk!6nGi94MYoR~(D_Q6gBei{~v*+};6
zoSdMmW9qtnlf*!YW0|xUMF4{WIT;OfHwx4xIlz<few9*x)P_rFFLT&!JE5dm$!NuV
zqil>)x}Iu}5fkO&0MI$W8U=dXexHD<LUdK>VxBkP<U|{>(kw~iR7b9aV&*O*;{42#
z*Eu3y^%f06`L@k8=qHE4soR5o9TUh*bEh6sGys<ElG>O))ac2p+(MxWh*~$ecoYOt
zs027%$hseYf^2!V@*q6HVuiG4UbMKwe>c<5!agM>vmdmrgc+61t}TO6D18gVi%Z1?
z_zc)@KhwE!({*?&1?)O(;w;1Id4Za=709Ohh)+s@jR@^!%mJ?=28suMQZi7!@=3`s
z`-46y0po?zAN(c6e9?D)P%<ZW>-QvgzYPR(LRmFce$sH@+cw}POm~|P2e|k^@$NVA
zGu<5^DKKj3Al?Cz&ME$nZ{i0|uDppK`94<9p1XTEGH1>Se$ab(&Z<QIe|QaVIm5m4
z9sFuV9L`*E{{tX2mzP1!E)xPaG?&0sDHH-WF_+<88YY)KRV|l)<`l2vU+=#kbXHcY
z!12#`YoFfU5C0yj>VV2}=-31Wgx%ji4)FyUtJq1iR!(gF`A_%l@JB1P&ZT8Y`mwtE
z`MqmezGmkIEzZ--)=r!UtgQ7NTdz<AXXnuut*sbG7rDfE5;uXF82NnhW>KPLg&10i
z;ze!6wInX3EE(T_l0-#O*YeYwFliAJs&sGqU^mjumx*peZ&n6osz}60_XEtxSGu3N
zX&FylTt!Y1x|(M`<w$=)x{Eqbtow|5OXk{h>S=AhAl*|>h%T=^ON)s?IF%=B@+enF
zgxuoOTXIt;JYpNNBIk>3u?DP5;U`Wy#^<PcZS!;YKQc3aSY=J9Grz^U9IhO%l?tn^
z;tKTAjFvsior8vyP>XNA)-;}@N-j<7gN(5?cX;G;TRiZw>KB=ccvk&}RY1O8{e5Mx
z)M0k4Wmb-AVUNnM&$j%mPfI1by*!;$Y)mt0OyuhoI_)1}mP%3=N1v0kaREp4rIecu
zDKYDan51@pFqd-K^?%3m;mszB0xI}$+r*e-CFj8=p_7(A`<7uDbx;)A;9<y`=dw&7
zau+HiS|v+iBhYRKV1af8Zf5>vHNq%x>Ac$#;u}H4Q2?d8*5_=+QiZIthEjYe*x+h?
z!(x3LpMd-HichCW%MD#6!E(>O7Y_*1BS#)ls!$MryvTh`Z_|vueQQVs;@AMLwdBE4
zSo4=qC|UK|N&12I2>k*T1?nrVT85W3voL@%es0#9lx4Len0lD$9Vs7aDoY$Bx7aWw
zV+tJP8GA>bYbeiI4;%$r4+(AE*yfr>ktk-`(I%nsilLgA&FfWgla=1qmPwu-pQ_8x
zxu#x!OV3`>+_Uz;VI(-bH;<FN5?IQH_T$soln6;GjDi*4n8_RR22V7qg0bu7^O&k<
zm6WaUZ3SgZW3XJ`le0CKfzA#d!ZuaY+&u2L%~#72sz|aJe$7e%dz|9qq$&Ae%;ISB
zUO9v@XY1$nai8N}2I~v*6rh$h$a{zgFPBGuiwXH>t=Cu*gOrtNbtsIq;!8m@9wOB&
zT#s<KnJ1iVm_qnWu0~{3(EgLS8j%;w`_*Kvne{eXYX)4WWVD@P0cd5!L-#8NAOOlV
z)2!^7OjG-Qs9@lsPsz`^zd2V_il&T-$A?G<84xyhi*uw?iZME+dbX=prF9@<ajE`)
z`|P{QFvP2ls8O4)69%wWoZ{L+bHZ?SW+$W(Z6jnO2Z)RvBS=kzWHkMZJtGEE+Pe)n
zl=lIvXIsdU*UsGn1}a1bGJ`BLLRz5~P779TDKk?L`aYbh<%6ME^YK)z-<y9(o4&Fp
z3a#Xm3F6dNq1OwSM2B#+I9hzlE~!s{@>~El=o?`lC3P&A7WM&DFm&U9uDRc_3`pV2
zI<|AP<Fb)_9kv5<JX$4g6}t+^3BsEZdu{tZgudWGG2)}}J9l~MyHH}~&3zzT=8ORh
zbu88P8y;FAh`^-Wz4UNDYG|ci15Rg$g>;35R2^d7z~l-xNVZwY&&gc+vYRb`5DzD1
zDVWHwn-hKrcyHPiSP|JzvkA444SnOYoox^mWR6d?!Fol|*kE~7$NR^HASlEpuXE)+
zUj)6cArA?i;mM~1Te7G<PvFN^Dx<vuVJqzhB`z(%zvNpI015WPj^C0RiF@pQ@^_?0
zGU(-fPlybX+TM#BpHp0&BqPdy(duoN-t?yi1xOqGDOA;oc!TM~OFDWUw|u?B=hImM
zZytOsdyW*t%!cJ1?Bpc>w4J4(6OxQG&CM*Xo1_ZgjC=b=YWi?27cTQ37LE~SUfek|
zjF-GLdw#WTB|f~y<rtqvTurci_(>bAam&I-LBsLm+kotuHCexKkh6n6H{*w0+S8d+
zf%|OQ%*%pncQ8Iu9M}`ITZ#X3DjAaX5y$5cW8!=LT<{aWFT41!37d{Ao4c@p{{ZE4
z72%g5tt}Ln!KyG50x~g|F()Y}m)Bb@EPv1~H$@g}f5oEt^kd@F^Xc+>y1z=C46{lY
zm|nk6f?ja?7MupPnv4)kOEJCvoc=g-4w^m}`XJ|0(#O!>mY(Y2P6|sigrX0<72@H}
zpzv#Gj)b^Y>bOSIESTvj==tc!X}#l`dkREP&>mub`1AF*6Hm<EUmd^2sW!-nc7Ns+
zX%{3)c#Th$e-kIl|CaF+-l;f_^i=u1@H7Gk_g5=!?yMBkSatL*QXt|%M?!p-n+LDP
zj-hp<fpCo5k)g+{SFSLgHTR(8xe-t&R_3&o2!;!4XHI*~`^WR}>*@9M&#7nuBvaZ=
zT39AYQGfkB`4=dN2jeCH<^)c^{(qjnpMD>yCex13;Fm$?T$3!F&OtM!?4$virstv7
zCWBp~u3gdr(yry}*N<r!d3ion-BX~T9vKtg{HJr|vI000B;ByWhXZDoGP`rM6@>^#
zY{oE(k;V356~v}_d|tC-38sy(PsUW%GIX-=w5r}5^5NGdkC^kC%JbZn5P!>WboD`%
z{4&Ui$v^#gneAtyD{HZg?`SA__d;I7=~={$EJsn)13=o)6ICUc0(sXm6irr`GKo!_
zSJ@QCVQv-NrkyTH&?Nj8s|uKg;x0a;2GbPf2a&Id4kRZ4vM6|`rH_?LMrU4(zC3Sd
zJ4MNbTw>gRCAwfGvZ`wiS$};al~Hgg3a4Q=emzVc0L7HF;|{@*F>3DvSg|XY(+p94
zp<I#+hH{~~gXJJ5XqPXQLs1HXr5w6RLC4z%!7(WpUY=s{{z`fJR@#)5%AM-+NcJb5
zc)XU3@FAF$<Ul}xI>2neMbg%7oMqetSNKv1&$^M9guS8C+QeQZfPc!25-_RU34*1J
z3t~|A2AC^N%60?6peblE{oP<JtO%7^PZ;<r4hzSbG&oF_3od+h;9-@#;VE&nug*PP
z5>hjv;T_84WN#ub<Rtt?Vb^ZtWnbaLt|0e-b3<_$Vq~D^wWQXxcYV0iJd{_L_m@}F
zFk$4lm$)v>`Yx*In17>OIa0+~wF|jkxM|7_ZlT4aJwsathzEw*Nb$o+mu!4#icHGo
z^()BRjkK|u<x_@%Ex9t6d&(~S?GSBB#iTYzYIC4?%z0CsSCo5z_=4bC9VvZJ@PBqC
z4s=>62XJ(K5xj7xb!;T2s2n6LP}*vrR2_%3|Js{drE<n<O@GP)-IgvzdKFJR18;jx
z4vCb)gzlVfPSk93p7*-GagGc&T8G)aW{0E$J%`<1$A_7U=kGPXwhjuG4hlV6|53*S
z6jd2kj-{AV%#br4-SGfN=q|s&^g-jXyy*{|NMaf_Q5DO$DTz+C(Pf(?=M(sxWBvmr
zn`Q#W%~skcK7agj@&lFXpLXO{I!qPIu!~$_eJ2m*cF(Asr$~(yozFrGb!x|isO58R
zad_K2HFh)aSD}$UMU6Y$XhG#Vc~}vm*>`N`_w$f7Jn$ekDo7h5Xl4JgdW4KsNk^&D
zT0{w*8f$(eOrer9v~M1Bk1B^uDMo|I-Y=+UnKDklp?|MOkMvh#m=<E#q_k479**{4
zJDa5=r1>XWgaCrMde)Wd0r6aGAC-D&vFBBLQ*9qC%zexwJ(iStOPI=UY=ZaH9K^@t
zdP9oL<IE`IUaao*ek=im_x7G@&=!=|X>NT&>^+Ke*I4aZt+(|P3bL-nnhTJSx?R_+
zoNx9Vfqx_UR7HJE9Zf8q(Nye(zjH^@lo?q<l)Vgx^(HBgcx)W_lpCFd-QFXzC9$pk
zv}(XNH;MgGS7L4VX3>M=WCzM-EJ_T`<WWWW{hFC0&ABtis{3@t$Q7um;e%#{m!ph>
zh%|qx&(8%Ddf=66LA7=IC{3{d<ogHJNd4Hg<9|lS7Y-=Pn~r+A5NLu28;z#&DvX(?
z>`E0=2)(&&)t7Tp1~}V=tU0ua4^k7L*o9o5Ol*XOhNulx-`F%21{r-y(3T65&N6#&
z<PWBR(3W&1f~-5&hE{1Hi*f?<SI=z>rulh%HEQFEF2c%HE$ZghClKobIxgjD4sY(3
z=6|xKOQX54mep;`MOb80Rp;|*2YJosWll~Jk@j4>(Pd{-L#M7Qc{{Fa%8qcMSE-4z
zf;V_<imXdxxQav}Bp{V&J(e$w>{E%YZaWWs^(NTHLtb+OSFR3CPV221+qIFCjJ+#W
zX+w2cjf~f$b34DD-<rL3z875EZ>xrviho%OA0ZrbCfCpxuGVLxVn`b7XLH$>$Kp6#
zHk70(D2LrXn{K#b0vAMw>sC-|xNap7$_QzPl0~&MVjz_HF!ZrjoXZCuXxmVaxuT<3
zdEzmwMB@hJkkROK??Q3K+wR7@JtF=$`-h#(NL|0-&fM=PpyxvH;c%PWic=v5vVZG%
z-QZP%%-m)9P4gcE;q9VmE&QiC3shq>3FQ#7dVyc|oLdfZh~{9-=_qxF?udHqB}wck
z!L76qIoV1t8RbE4B@g(bVi`eE#e}1+K3FWH5`?>Ar1tJ=IaqOwp6UVlfa2e7_@2`w
zmhDQlRSeJ#q{ow~F6dTJ!L&#0oPU+|xj`wDSd8;CdJDZM!dJ1RJICqKjb<S)u!m))
zMc*g`wy3r!S%hAk3iY|-Z%Gn2xD6fJysiyqxFK0SADW8T;rXou>Y6z%RfVEvk|jWG
zf{yRZSUVEf;ezANzt;Jn8E95S?{t2xsT0k1?<?W9wk+)RCfM#Kr&77AWGLSJ4^*oM
zp+)V|*DTx6IUZYTJlHmxTs<tss9I<zKc8Mt{{k=go`RPltt}OofsZv50XLUn3qUBB
zYh*2$fA}7l!=G=zJ?SEBy*3FAso-tryj4uINO;sVd8*B}q-^kFc9zAmJ~kxzJ&zDe
zN~{^BJ1hkY%Gbkl(u6*xnS7--JN<M0sP)wbHjS_i2+AkX9PU!(*XACF`SwLpqlt`p
zME5kUFLTe0T9Vj}X=mQhOPMGU^-|mNzvq?Le{0ch)yfo<j&7@!C$pEVysm-E_QhoH
ze8-OEtk3VYa9!18AE~iobC}R+Wyl4`R~I#Q-&wQeJ6}HYt_iTXYtSfiv2(^KA41ii
zssy38f0jN>y<4%l5=*|wf3I&(|31xFXsZI>{_ERQx?zEo^RWY`qH*2#zq_wbze}p)
ze{G)?(9RJ}6>y3ubdI+@cvBUybu&;gUv(ZF95y>|Rq#}lM7h+5ysl>+I~MC(=G?xr
z)h8og2Jb@FF-p<kQ~C&7vVFADi2_njb25@wFzO`(AX?@+<B~j>PQ{0&E4OA!w4f|{
zNr_&2YHS^?+3TIo7Kyz-SHos+Njvqlf8R<Gc@J(dQ+<F{QZdX`)+KE37r_9#KjBP<
z<q*TWlvl>h_%8<HaS-M?FJ-I*7%ah}2D<ErCE#&*s4*q&M_tug+99i`ST-MdVv)lb
z6_;by;Ta*1+QJPf1WX3wPcUS9L(WDo#TfF+#~@{R%oc4X;($DL@@0b{F`Q~Ee}!qV
zc&eb5sjaDm-Z9AdnM#*8#hOe;M_$btwY1sLm|-#2q+!QX-pvd=?1$vZ_xfix^xQ?O
ze_m9Or*pWWnbENz->0z$G!MLjt7X<W=G^bA#9Q0PB|)!QF1w|$UUuc#4nH)x=N7!?
zzRoaQ&X>9bst;#I!b?$rccS(!f6GaEFx5;>7pyoBc(E7ZO(5kV%n+U;hg-Z1@dfMh
z7nNu|8TnGufMpb{X#g2U!=_+zteiTYi`%lI2m*R5gZyjYTSrDH0pyf1Q8GCV)@vKy
zZ$exMP5}yfsai=iQ<af^o8lQ840P2xfQqcyVV+gVt0|@wuWyvy&<ruCe-Mo1gnAL0
zk*h!xalc$!Ht>}Y*iytMx;H5i???kD=Jx6W?=gy14uhn;RuWYcPG@hy16@#6ahjtQ
zl&>t?WMnK`(VD@KWu%5?RJhS#uM7a}d=&dJQKaokA{hgLeN=E|2J;=iT^BT{B1g0(
zU9jMNoXI`TH6YF&J})|nfAvv0=N^#EmZ%~KuKETSi(D>C_9YkU;h>bPK6&^k5#6&o
zfvL~9EUwP=c)VH)l>$a^nKThLmO>PiwInIYr7J<uI!+pNYtiVGZ@oS&Xj|*JG>;Dq
z+JfRjYh%C5L2kYY$}2&-`JB+~9c%VG>T7lZ4}PXvETL<J5UiJme{IO0Qhpm}i#ycU
zTM(;lL`spXe)WBz@yK|gs#B#d*#M?CvBmVOcvlWTrV3+LJf!W}8w5bh6%<U&S(AJ>
zAZ&|41(Z3}uLdeu(u#;0sdUvvg88iJa)Z=h7!R;T$x`n%Z?r-dm$?>+WqbG*{+jWK
z>=-0uQ?gh`-07vjf70B}jcn=*fZXKvcS~j>s2FLC9}?YD6S_zTL(w6mTgi=<n~ocs
zABM)xK6IRY?XK7oQcp!#+v=d_DlWB7ZY}JqJMFEKP;(THFsqV`SC*5r|D#GW#wY}O
z=Q2rhxw}ph1mCexG5`vre<0=xiVwtGLHP>iO3i+1%$2G=e+F{})d6!KP?+&jE0$jJ
zSZgF(4hhD{*_#TpW1nc=kprYQx-sDgCIJ{CJ7xUEk^uBv93cF#Bmf;Fg~OjP2|x~Z
zq|Li00rv-ZG*(fd2L^aSvA$z(!Sv#bp3?p}t_RiqeHgTySOk!Q-Oz_=6hlymZbtD2
z*&>%coU|>;e|EjEQs2MHr%w;Ve+1;Z|B?7FC_dgj%HFx7Z&aHeLev%PLdUb>OtH(2
zqmi@R-sUly6k0Pg>Hap4K~(5KQo8u2^*v5sfS<(84W}=PZB|b~0kJd@x1@P1spzWA
zm4iVA!-<(`&d|Sbyou{dDaYxnM<cktoLe1-@j#}&f5?RquhVd%sGXnLS+PZ%kEn=(
z4O46QTD>N&8xFF7=3EfWZ7bLBvz&dD*U<L0?!!J<_!|zffO6JJ!7iN7sJ(dc#66*@
z^x>rpE1`)?>2xk*fxk2t)`(urbk`vZM~aCwc66}z;}GWi-F)cJ!$Ipm*mg#)mRT=#
zMCRewM9$K!9AbFHdvQ!}@S_(hE<4)bW!KJPgp{EeLA39$v~J7fC`CODFz&@{K51y1
zaCPo1^_iSXoS!kQSB?_@0dudDE|-BdEE5AXFgKThE;cR$GBh}qK`thjPY*%`e@6~}
z_pcbBFTDrP4mraa@=ydwn@g_;7YO1UaCd<pY(;3@b1lP`i}c_3<4SAUm0Zh~;wTBC
zj8?N;l0$y{IJ32+5>hCQB@_-dTMJr3EyRIS^MQs;8H0p@PbtZQH1nTQP^vyCfs&Aq
z1S7RQ;hF<cPJXRo)cj6rftY&ne}s|vy{v(9FD~QPqT;_=AQ8XDDNr+)!UTobpTSU#
zr0ijo19N`|m%uI_BMB^}zqbai>Z(HxR`PinQnaX#AVqu08B#2}ei>4%xqewfPE_}#
z;@+I3m)tNizZY)EoP8cd<y?9xhEeqqzGmqqmlCq}S{2AsaSckv$ogxLe`neDm<{qw
z-L^uZBzB$gNO;rjsJO<pzt;k^b;*J?E7diC+l%z_gnb?T4#}Ztb&I5gVpicRJY1Mv
z3WQ1)w*%@*HV82IPuvmswPMtKUezF#T&qS{@!k%ss<hq?JY3BD8Lk?!Kciwnb9E`r
zmF%^u_`u^ZQcSMR<tgU1f9Ah>SZMFHmP1SC^RNJodpn>;n_1avXw?|cz->+0Bd2L-
z)_Vxyqp8<L5t7V)2clR@zitQ+Em)7D&`PCWX*gTOS6ULTHAFH-H7Md4&rl?R_4cX;
zOXB*e1tT}XMOp5n7_Kv4WQ;giZSGUc0ZZA>E5WR|Mdn$NW<Mhte-OrQ3EV>|`vggh
zp&Nu@?sKWSy5vLYuX={n)T`;jYM+T_KJ<K*S>mxz@a%y%+(n313|rp&=9})-dt|pF
z+23~m`S1U5y+y^+G3SUG;r#mS?C8adznYmWwsd=D=)RkrO)d{Fjz{B={f5xJIi60(
zm*;_JU;puO`u*Ybe|S0$G`sM6az4c~UT3#ROaGMD>Tpo<><j!$lKSVrc+tJNoQw{~
z)A!w*SFgKw<B!w1lJ9=M7~^3-jiJN&bbLO&a^RSWcsstDTwjjHS2LvXOFxc3oF0EW
z`S?EJE5a3_!$-LP5_g8mSaF~A^89=<6EuJP-uoW-T=O3Ie`W)6A#aav)aY*u-M5p=
z597=J3F)Z&XZJ(*9lpS{BWN_5z9;4=aCIj9kL)7hXzciEF;zR->kh9^rY`W{^!)$b
z%Ny~d>FMOWJM8}b_6Ps>_0QAk#ns=t?x)k~&(|k$H2Jmr<@(o){SV{cx_@2&_$K~x
z)}5S9PP#fCe`%$50t8H=C(Lv_8jacP!-r8iQ5Iw-jVN_jm!obpIlr1-9-qRM5b)Ij
zz;piqi1#<R;g-)IX)GzC2k9{wOI9LaZ`X{KmiF;cHZ!*7$X@Q|wpJtzwOuncmyC5S
zZDA@B-V6RB;VTjzh>LQ27Z&c$n!y58h(N=KC!$hEf3MaL&!i^N%gP!vYl;<lxx<*m
zwuDLC*1)8a>s?IBZGlNNKoO^5K*NBCK^_KW7|1ZF8v&T)78z7(h;vYf*z#x3>5EIm
zcpu_$<UT$&gf1XIaJt->2|fm#mfbNE#Csnfy2{KuI<B^vW?y=nb4$@VB3lK7fqPbt
z&f&p*f1}D#_cB2GVfX#y-K2XJz9vMYB><N|Xul9r0Zld)1J{QZn<}9-iWB$>B|vp0
zHqb6^H74OD9h_={#}a|k5rNp%vbhRww}O>&PEkN7geih!a!E1QWj}!3*J^5}Ag0`)
za@$-7SQ9lXx*UxS%$JdF&aKhw#6rtDpc??+n`uA{{(mc5&;hYuWmc}e7O><<<|MEn
zPTF!nu*ZO9^;X3=;oBBGK?=7!wM8f-j|+})H`>G!EiFd`NN=r-Kj@<bblh|ilwLr1
zL2ALRSCltiX1x_v%Pg-mvr$Aut}#;>PDJat#!N~v={hsBdA6)GQ7NK(_yiU6R+0Qu
z`Sd`dS$`pXZ22t9l)lhY%4fMn>GM!N%Z5z8_RM|3<Uw4}DSv*(ck%m-!k6Y0UN)rg
zY!S=1pzz4XN_l6jtosMIYoNz{S&Pk<6>L^edYIRlSty#UBa;X+YguPzf=nyVNoO}o
zPSK_0l=_t9th7kZG9;%A+*Sr|D+9Ncf!oT!ZGS~pWb$K+Uejp4_XT_wMS6o2HQzzZ
z#YxdKu^`BcS(|$;R%h3(^nDX58^UP0>Sb^Q@|`*lDMTfB(e`+d!1R`A;6Z}QizIwr
z9wac_yn&qaAn-ltqR&f&iUH~oZ37V!zh302L}-2Vb?PayXtsdqJssW^^tBB5&20-k
z2!8`n$#+~^&!gD9CE*?%k(tY$2K8?W=tvcjfz$&A+@X8Y9-w=-CM>o?8{)mJGZWn)
zwJ!&&>+(KCebPQeyM@(tz~MUJa2@s->SlWk-u>ma2z#(T2c(_bOsmeYR_`=Cw%MtG
zSOioC<N`quV5x_{?7J(qJTCs9wlCeH(SO3_+1nJ_IhPlr{e!}^Ma1#Wx>C7;9L^en
z<Bx}&st@qa-EOvYx!KYuHCtL27|{mO)&{U>16Z^HELt4~Z5RxvE^VNLx>>I#iZv~5
zv1w^wHs!OQy8L#0d3t<uHlD+YnaB1v;RI_DQ(3?X*lMiy3~+*qg9k0zMl3RzQ-9tq
zhl4W1;gUA%kh8(WTL*gJib$N#ZRz>h#mq)UjViVcz$DR_tOZ7JsRO0A1!!c!x1Bbv
z>X2$O(v;99k)^biO<MN;tk6OWFh<`Mx=#yqCq@}eJ~hr+6Ui1D13&jF<wl1rD3<wt
z9SFP;O+iD8hUu!BwxWc<@_eHN!ha$<bG}c9stIsR_}jp+3$!U4rFa7~w*_HP5EbLy
zO7P~vuaxfY4=F#;8t94*A59Taw9d>b(ci!MLdw1*S}OZyPbmBT3e8Q~*P)W@P|0<u
z<Th;V*s!r<14podBiO(ZY~Tnsa0F|cZ}kj7CdGW)e84dgi=r!tA6aXb$$x65WNN~h
zwufVb5_1^4PYEoFa%-aGZI<GH-5QL|RTxn0UO#ZEEdfKRLJ_NK)wYIWa~<X&cK=ea
z6);sUm7A1O(q{WBo~%r__wWIpI|gDVx_>Y`&Dy->u}DaGm>jbo>?yiUpo_Oaw(4Ln
zp-5P5*8ByijL0v%g@ZkWdVk6wrHNwy0w(*(o*#KqOtn7G)?Ptl?JtSWbwJAz{SDXq
zboex$U_VSj{UHTsbP(`8XMf$#W-o<3(|Y(pR_R~Vg%7Mh<|RMs;Tl)<3ZF!Y0ys?m
zUR)A?MJ)J6c=Z=1{k;orj|vJ@W`DKw;X|Oiv;OMLcje3;EO?EX8Ab5o`&nzu%-Uz@
zm%<*X{+T@pm8Y-=vITpvIcbp8hQZC1IgOAFU&E{!n6qfwY6dXsX9xcIonIZ00`r;X
znEwZzk4j9JA+0SG0Wz0yzXueT{Yon+e_F|o+%^!s&sXRNOq0dp1{ecOYupBMi!rw#
zry$PBhxq>wDejA;mU@{uIrJux#oAw0srmHR#HTM$r?;n%XNi+xRtW>s+qX&34NmWZ
z(?zW&BLve@OmE+(-#&2;nqDV*A?H%k%g}dApX&L86qZ&9MK8K5#PbJ(!tbFqe-h$e
zspB3=t6-*2L7y-FI_-B{b5DWjiP=+m{{8KjC!UxsoPlIYyGa{lM5}N{XDUwxfqp4E
zPs#N1-|1W=x@gJ@<L~%iGX27nUlY${psa<7Go+W5N=`qBi{;g?Ur2*itZL9%;!lr!
z5<`I_G>M~gC*_ts0FW(-p-$q7e<)%Qid3s?x@XdTy2j65SZ|`Rx`^)AA*Ro$f<}qx
zR#cSlFGa~T=gt@%C(q!WVlW0op2I2~_eZJ&FOM<~BGUS$K2@WO&yh>7E#mZuk!U$l
zQwyqCsAOr11t6X3EmWj_Nhj-Y>ca6gFiw)T)Jdy>lFeZluG&WLJY6&CFPv=kb9|}U
z?=*R53Y~CwZmbn(r_gznEkJ6fg<Z&XpDjQ;X1E!tQd;u6Kc1Ccm;QGxEPr2(+PI^O
zFz*G=B5$xMECaDFpyO7a*6^A(f@8!ZUAmeS5u7BsZE$E)*i_Z|eA+>7b5f{la*Bww
z=X9!UPBt}k>b{b<<35!HjX}^-6J-alJO(F@vM!C0<SQZJ`FLsezO09%P>c&%+_sUV
zgpclnucqu0QrXNg1odFXc7Jc=B&}Sg;Et-T>9p>%t@Lxd9`hOP**a-|^*r&qj?q#v
zYj9FdFjE5H#MSz0R1B>dR&)81EflQzFkCj2q!Gr`?oBsbF@X!BDf{NY1Eos#k7^|l
z$_QzPl0~&EVjz_Dsk{kv)w*-}zyoa?>M>Vz6e~|WhLvdCfE+R!eOK;XDDK#*J%SQT
zO6bmr_}`o#PBJ6)Afd}aop?!^GK5reYXLnMg7*QM-O4y-Yk=<YDnVxMviv#gt9)1K
znkoXPh5uA%fog0f0Y4QAkz&H@tdVr)+?RiOEg*lj<N@zgEF(}7)i~Pf#$p+jV1Qzz
zc6GHJtT;wb^?(Ry@$Q!IY0sYRYPD4m&<&;spNUsguIbFDMhmHKi)lGT%S+`Z@EF%;
z_!fFmg|C813yE=0tN&TZ6YPPRY2i1@fGw;oY8Ii#3OQa$09>-f1#Uf`WnUZ3a6z)%
zAew&)+2QlGMCz(Nm12(9T(X3yP1vD&R9L&9u)_z(i<Kt$;zN%_E2?*DKiBl>RhmlS
zP3%U_D1wh-cegqf%w0XR3l*gbS!;(D6--~bYy;@nywsquZAiJ=GR4r^*AJyEF!HE=
zy48Iv5M6ZD5h5TrPMKHRVWttK(M`qN6d8Y2@}@z|L5LN)mTg|;(QY;TE7mxYXped!
zcg(57Uf8W()AF_id!X3G(oqeULTz43O!2~{C2*8VTTP$Zafr~ldgoG2n|cwnnwg>0
z$)gojTFX|0oElXB)qjk?KEFNvy=Zq(S%zq*KfXWt8%POA8%ODlIjN@~e@@??ejR_Q
zw&TVi0M}ZnUI5J$84fiv1K8Hju_0-j5&DGn8i${1DEEjSdSHtCsCnd5eG3}zu6(VY
z1%CdCdSpzzZuUL^oC%U{*x^f=Fa^DV2MNrHg>p6`2c%(wGR=7YF5sn(Yp#(GDF#Zu
zp5rK?Y439_&a+r+hTof$AW+ygquhVZP^9iah*(oRtQ}N_8D~cxuJgQOEh5E;pd>Y%
zv{GkZsv|%;kN6|It4?1$ev>5w<f|jG&^bz3^}NZ8CC5}^++SVYJL$cy7b7><0}I!?
zC?5)-MqsIONV)p8pK*YI14xiuP&N7F0S%C)nDJ4&<aAGU2x9{fui???J>*=y%2EBg
zwDrMxsyBozxa^-8sa$)svogDuMD{U>dOb?gOH5LZGAUdJCC<3dxZzEuMBH8Znt;pk
zOTZ-^p|vaF_*zTJJiP!h0B=4Ze5>c?D-8e$rI{q^!y&7eb$l&0f7el$8&V})&)1Ro
z-ARGI?&-T#BYo*ONy0WIaM!@sC<U1YJ5t^%;<F0-)kpAvCro9eORA^RTi~s`((;z2
zM(=dx*^MV(9#z<t*`4pQ)exIk6lba28&0JImJxqkmR$0!$pQ)P-GcA3TOku}vH862
zB-CjsN{RQ^sX43ke*`yugK`fwc;oT;>F>FY;KO62Q&VGJV;#%1UYxROo&9{gmF7JD
zAeix#uI;sl(i?>CN{y68)8!-x6pc;Y=IgD`k4xGdsF`$+OxiugtvdQDmlNS{Dg)(>
zBwc)NknDyob7Nf7Piq>NRX)b^TkP@ME3HUL{SY1{)xWnqe}B;2m#2{I5CT>Dt73mr
zeiH*!l3HH>`s12%SXV46(VC~>bLJ^$-BNC!>$En?GQ!@_Qs2b2cFbZOv>kbl^qM&q
z^8WLc$2`371S4x_`6xA8*rY#4saPt9KTetZIYYx-^Ph$l9P|Um#GR<+`M}*q6|p{0
zIyeCUAqfG^f5rgYbC!p*bA2|LT80mq-{vuOYJsyqUE5G-WRzPM{e(0O$+Z7QKOrdk
z1=@FRD8^In?kEO=@7PlK=PB^Wj-n^{Ky2y>??BTa)(yz?fGQufW{#3b&&<}m(Fg{7
zG{~VE{}gMMJ3vxUwgm$}5hVp>Rar;w{}YrHlAK=Qe@{b60gBpX^&dn@A&{Ze?~ao9
zXFvzkRNDt;z@A|FPomixCoX3Zp)&3)7DtZfn;LpR!d+sy=j(=kzRs&DTYw%3-oTmD
zbPv}!+wab$uw>>17xTlQ(Qu|8=oT%x$2&!emwO(MrJlLva|^39FztGgE>9d9_qcsN
zAh}{d89hm-$?Jy$5`cjg9hr;I-NsL<b^igP{T7$Oqy-ZKHZ_;ARWB5mYJV-h0i~BI
zfGt&j&56sVyp=j>#b}#1p0cb$IC%r$Cpd2S-&@hcbsQhg(NEwa?{GA9D7J!ZD<i!5
zRCqv&fn+-sY&2hdy>!1lzux_O*Sp08qa5?_*Y~^lfx#&2!hom(@bKgB;oIHsJ=Gc|
z-8+SNGOD)Ln0XLbWsD!JMx}qkWAZ$f^~bz_*OZ7=EF9WDDMDu}5}8eiz{+AT1U$uO
z7I^glYEmE}J3kO}h^r@R^q#U;Vy<$?56+|&4NhwwYO%f%8`VuVh&HV@NDW{<6#lwc
zDx?f5F_YpWNq5Y@$6YO=Qsh9bBWdqrJsnq=fgZW)*j9a^HTnRtN@4Re#$=P^p{yo<
zF1xvh!K$E5uPxAS{*Yz`;@9pJmkPr^>%v94>bodUl(=Qi$Jab#P%)V+YTotvi$;MY
zTbI8eh9tY6zZc}IArJ-N7bk2d#RVDLo!CoR;_MLStlx*GEeB~)fS;ANrqd<q$z!u%
z`)R36AT`ta-U4r>G6aPI`cwadR2J5M%Cr4I_`=2t+?%CVb=~<Qgu>(%iXGoVC_UAU
zkd;~`sIa5@NhpLRZ>3m%5(;4*1aO6Q)0fI+N2OAgGdbEgTwleUQx>4-x>Lm>Ae7+m
z4&X#2Q*2)oJQs$%Mom;<@O-?tI!*>vfppSmMUA^!8FWUm#S4^)sU^pCq=6=XQ=P4R
zZ7H&F6nav@Elh?@%Z}6A@^6KN<cSJCEI8?_IgAtqg@i?r0k686fRr-}d%mmbkOUmA
zLq%6B)-m@Bsq6w-yIzu4y3!%5DU8CHC+Tb3pXzg{`}m?@-3NmvNR_5eoi`G5%Qh%b
zf(UzJJa)Szf9)$gc4Br-!+H0As}_;Y35V6>x<?9!QD1hs@7mONcXelhw8)*bJ1M-S
zHWN!QLQkl<YUz47OrS||9O${4x!E-*7B!m@C0L;LiCKS0qAsG<8mR^&MxL5;5w4?4
z3LuK0^RxqqnoJdFC2ApZfDL(A`k!x^NX6i#xo`Gl*2)p}r;{`#$YQjAUek{SaEPUQ
zipeI}^tV;x^7C^8c&F^+I?pwW?1XEfIe~`BcQt+O)atdvB+oXQZAaDlc@q8)O`7ps
zm1<E|W`J(tBZ2pX4LebZ$2tE*tN6e6CW$ofG1hE(e5@|1mM>QK0HH-?q&IaktS4vj
z`V=fK``2d8N?_Rbob>sB!*oHEGUNS17du97idZbg567{nj!z5FyP->rW}u$a4FNlW
zRD9!ZIeViQMB&_xW^W9X`rj@}y&jDU%-5cK>U(I^iW36$C*pN-YsBYCnim<PkuQg=
z$JSfIJJFBuluFE(!k@|5J--3#;7_SjIIXJlQ?7KVb))2-m<=<3t={hiN#YbAOs{nb
zjBPaJl|}4N-7p)G1?O7-f#WlRPb1?-dZb!~rB^+T9kL;kf(95@LG2~;x13a9?B(<K
zwNf47up!unTTx8Hom`(Z!Gpj!cAxck1aKf@6MVe$TOEq!wc=o|qrmz4d6}F7N4>y%
zSC}l8^U$Bzy~137$a#awUqV_*uX|gO_YD3-0(oK!?UcX4;BRaKS3Re94E~mt*Jfd%
zjltgrC9&7<epl-Qu^td!Jo#ha`pU}MjE~dyw|269-_5y8DJ^av@;(}ejmGO-@6+q9
zi>LZ5Nk7$RPJ2^5x|ONkT`<*4F)-+QKmE*HZ=ICaC8f82{Bzx4zczXS#)7uE2d}Zx
z4g3`rZ4CA9vxa)<#R_4(6Du^Qu@vLL*!{+DjMNObpr*@zx;W*gPksXFe2H4BBkwc@
z^B;_pMJn_=<ql*f=8W&>B^&IT9&IMyvEAH7c(O8TTfP5R8E=&H)azZZ7}(g(Pi)+_
zpUVC^Smww?Z)Ylic8%_)6&IXOQ%W`4L9a%HAmHh(lbdwZ&h{H2B?IkvU;9gLAZ8NW
z@cv=O=Y}(a^iM**W?=8$0!;q{wM3GOmm#e!6qi9=EEAXTc`zxLhlVYie}MPEoc{j&
z<Ao<<U$36eS!%3eC_7uKP>V)M3T;tW{GbIIwX1V?O{4Y7MGdc42eXp3W=8cfc<&^m
zfEWXSmD(mE?tls7h_<(Dhz!`E*PpaJc6xNsiSjk-RTwKRx@#Eq7&GQg2wV92AiutT
zzWn!6lnTkT@dpDAEfR-+e}2A%8w$_{>kmRO59;vi-^2HpKdIC-FZ4TtDh8Ar^rdo9
z7^XG4C^Q4%)=d>g`#c#`&J?OB|NQ!KAXmP9^gesVjA*QY_wcD8mq!RL86dRCh>!$I
zcQ%EI&^q~$e#<c*F01VoMlj?38hb55iM#Fgj)AG9;{=QoqO|kOe?UnZED%Z^yBO#+
z-&&ty$q2yJ0p=aNZL-3JD`CvYoabi|Dax1UnvpfKhVr7E<oiHVkvDD0t&!zdi0*sm
zagOl7%li3sg$bWxLKqt3l?6xPOgST{lu)>}jD<U3<uZ(A^u;F)uFKZo7Z0&DBxbW7
zcwpW(9_&#@Gybe)e@2@Un;h?0N|nmX)hM>&omf^)^~<K`ljkVUiWD&n<}qcE8%n7#
z)JbxuXoJpLdD47M*-;99(bW|CB1`wy#5-^0j!#$0!W?zMGAl})8Y%iM`EyAbaBe(p
z<5qsfOb_yG@v)7MNfssV(~|aj#f4*TzO`q*R5~~5*&^&Hf5AY>VbhZr*l5+I&xQ~g
zrJ^S?p)AbBW#B7MBT^0Oo>oL=RgW>}AV(`zjWlX7A4%)Td?-bcmqwFnn66&3hB%53
ztajOkWmb<`tWvXLAEALDJ<M%0my$DZWept1?*tO*>M@4sg|pscdzp8+s;J|5A2G<V
zDo1N1CLB%Le>JQ&CGSXe8Ad}nkmqE7tUI=<E{#nw=Spo#e!bt<qGBwVvyQf?UShZG
z=UdbTEvb&Zz8x;5d2)0q!KW{Vhk(julU~75p%+%;!e^N0ee0V5Q5IG|faYbGAThl6
zwz@>?NbAV4#l+z}$Pl`dT3z{87+4d-+#VAx(v#+=e^1iBi5#C5AsjPCQHlC%iogb2
zWuP=oM1=2w8ljB#0$u~V%-+<Z6RC!eW58+VRavY`L7Yy})$@*AGB8)7Y3F&z05NG5
z)v{&YF{ynMpRXO)o{Le6bTV7>&VlGnb!E<*R0)(-%opW30o5`f6d@*TMu<Jj$qYYR
z!6mVXe=6>&eQ=%TlF~A;l8(5OOE#&XCF52;BNb{*k4lAF%l1?t@p*D8ggzM5)?ARl
zAlw?;K3hMZ89BAoyzxcQ3ksKYNgviaH&Jb46)ETFw51KG^ZKOgrji)E^BSBJyVkM=
zfC>qfG*n`s`i6_+E?aTJdS<~ch)o&cT-^Gge}3LwLhss=Loo<#XrRqUZI;F}#MHeu
zcc{yehkYWiaCV)S3)4txg@m|C<=f?Y5Z!w=5v9i_U-sSdfK0B~$Z0n*Mk#1TVD2VE
z_9<jYg!M)|iVR6B8G5p?dzlXA*orT)RhIhM1c+#T5CI}ub|64R;*JD}NZyG65vdad
ze@JRz(m#U$5ls*3`iPdzT_26lbGtsNFq&JNJ}ErI0=cz>>_d9OJ8WgBO>?9tbyH--
zPNWBy1*U~~2K;MHkAi=#Wjp*ye4Y${ZgGi~8}To^VTP-nNe_`aPkPA1zlHQrgC9eB
zh?X5l50Ur~(sMgDl=M(PSC!5Chz>%@f4aujDWand!zb6+`51D8xRAy!Ocxb;Edfow
zJeu4H1B|=UWy|vZDbkMFgGnkJ5mHpb9&gW)a4CRBvKpf=<L|ib7(Kd7y}|C^%a6Y;
zpQd%UMz(M7Y&&Y*+q;T>&b`aL<89x)L{I?J5}JcsIJaJJCW1mbW{vsH2epV5e+v$3
zkr7Yc9TFCk$Ovl1M9~yZGH;==d2MmuC+2x)i^d!#cD59AT+^pll+Ffm0lD1%Zp77A
zEJ0j7;t5N0rrbpH6w1<(Kx*ff%&201g5!43k+9zYUH$at^%`w(TspjreT*})59Ok>
zeDuz@kLeXUm1pn*JEDpOi+uR8e;grPbIKVE<_W)9k>hfx65qCi20sVGEk6AW$}za>
z@bfAo#`_V$eLQ|%8S$hu(Ez}tfG3@a20+^B?LHIRC1=-vobpjJC3+HwEztT`$K94B
zF?6mrrdP*ggjm=XN!@P?Z)&uiV1rXBvj>sa9tLZtq}*wNo@DcLcPuNWe^B_lh)ea}
zzz!mEN&8L7)*BfsoM%X2Yk!wUTM)p>*0Z@pl}KtiqNCEyFB5I|=%LTsR?Z(O)`DTp
zsx27G9l`(P(P*a`AjT?Lsg5*%I)PkM^jx=`a80LI-}Bkv@aHar5X9&91y_`GZeBlS
zmtDoXOrHY$dIMHAzJ#kCJAGi$mqueZNfpMEo0rT(iy!(V#7H{Z<4qQ=aDlUE0Z}uY
zz;lm2^)`<f#hDY52qPU#{>J6zf5~Ocmto5p69YCiH<z&x1QeG#e;y~7ppGq@0ne8_
zk1d#g$;4AO`7PAhd#!iXd%vXJ?6}c;mHL4bH{!*S@P0h;`Nb4(1b-%d%CuC{RXpRG
zb*MVac@xwZZr_IgeSLfScM%H24nS<sKmK_N4^WXX)=vP!E1bUnIsNtY>qs@*XS~Oh
z5S1#8A(}gZ5kzic34~L0zsXqZoI2P~qQGH)UkuonEHuig{fTZmTY2qLfOo<<J#wPH
zgljcv06T>CFzyJc2a7?96JL{Y^ny>O{`LIH$dJ_Hn^`K#Ps!db33w3b4hRxrly^%R
zvsEHGTY0UqwlEw^BE0hp;jc+>vcpNY+23*zE&<GtS)3&AKA}<!PREo+4dJSpIjDty
zwd2e3x)xS9l534EL}4l^u5*Z|mg!UGJIBgLs#TtcRvTmv2R39bsn?Xm^r|<eqPP}^
zV0zy^SZs_yYX(M2aK!3ox&X@e7+B8*8$?zvrAE8cW~Sw*{*+ntJG*m;CMtCook3+A
zp=kl0-mr_5HRA<Z6`p73!#q_g=T9+zQTUQt${QgGVm4fO67Di3!G-f_;3#kPww?Ac
z|I$wI0mDI~f5DLB^>%^0=|`N24Vdx7O3BWhs2ak>IPRcbRCG~xF!MDjyikhfuO_GG
zY|N;_9rJb0GvxMic5&+zJBj*b7mjF4LM@cMQ$n_shm!q5!AL3NhK2Vi6xP6hQH0!T
z&LjNFfFP?z9%v&`VK!n#lgqla=}^iG;FLa`^noZCy%7%l15q&h5<|60u|Sb{L8b~h
zD17TLqAOC%bhV$5i}ZdADI>AHK<^^ucX4%Arwzt7%nD8r4t)J2JRJZu82yv*G=gIE
z{sG26mqhj;*bUkWW%bDL9_&Vc)*MSe?6?<pb9KzHJ1C%8{B8*}n-GjTAi=xvc0!@U
zA+cs&lOd2QNSLRI4Oq?CE2OxA;=6FI=7yE)QZ-2ivNqR1I5qDh)7L%*Kf4Kqd9~V%
z^f%mZyPXD3OLn9ZecsF~ajD!bCg$+3Rqi4Cbo;1t!Z~DRC|~|%x*oTGg2w2O`NXI%
zSx;MFi@Ecz0nE5hWeg+J-YARR)j3K|lKg1Y)-qT&fdzXOx`kJ401IubKy1Lmwec4&
za8kiy!MjW+U(#cUiFJ%N+IK8TvH^GK;dP9yN^`!ezT+}4mySP`rRWMs^R>v^B@Pn%
zCITuCD(|v&<vH6?%~t1syD|(Sov#vcc5HF_beOnuyAc385m0=pv>zV^lr(xl@{vGE
z2g=^}9v&sW90EC7a$T4EAZMhwk)P0UDRcwHp9D{U!_VW-f+t{6fLc*`#ufhl@g1^8
z^+<RJMueK8x$X@+c435XWA?a)c_{Ic-YpXou4W+^62gGf_ALH?9-L%CcjgZW_#Tu2
zR<Y|X>=VAIM5oIzBd6XYFoPimmu_6}{k7;O78F<p?+?uLK2|qclaR2G1PY{HLa%pC
zJu^HTPaCJ6k#Ukqgm}ViWltW1!bw_<#jb~6B-0Ikoe`1&>Ok#Xa1yi=k{X=!XA7sA
zO50qtl(pUP>>*TtE0)XMEk@J#Mg%NyMCSrmhe2B-AEhslve^pwGd&dOhEv$Xo4y#Q
z$)G*8eI5MkKt}?ynW8CRrtYgeW;pQNc69X20Cs1xHhou68K^zaK$Dt5fQ3y#3I))A
zF9`Ahj(P<22W*<&s&*SVk8tW*s7K0}piVyT7<HP@kRnxo)1RGI_s4Y%7r^5v|1$f9
z7~hP-t%%VS=2b9K4<5jUW6@~L{BoCT&k&v)uNAHj02*YO$&Q<g<CUIgZu(i@^IT1b
z36o}(Z}ydE52qZqQa1P6?Q@*xE=3(RObOSnjBklgnoA$88D%vvB%?oL+r*|!f#-oe
zPbl_K+H*U9PCW0DhCY<Wp0;!g==ynh-#2yPu$mn01C!$+u5%i|j@y*hbpbXort^DB
z$_^hT-T3M$qs<kABTy|e)T|fWv{;qPpW9TReh#t+CJea<K58%z-03^Nl&5e$PZu|b
z??{3)jnp;@3U^EbtpDGBvNN0yT`SY|14ram&I%-dGg6c<kK;M~G63CpLWKb^Z^iQL
z<&e!ZbHd&5$MFsfzk#dC$a~N4JAU?Lc~Pp5<llj?T%v~)elSL~Gy-8^$s8k}U%1EK
z0oKY$q0!9dw3jv^j6k<!dGllQ1Vh5IK8JCF3r*~eX+mmXN*B{wnzHhmCk*RdO_?w#
zZMpA%tiLhDFYVi4N#vjobv9*CFi?ENPmY|5Bcy@y(Brk>WusXf-(LN;5#}h@*kvc4
zZmaaBKG}j!={xkvowa51yhqjEbhh$(K<D#QLEkLN>zYGhU?Ic$7p(rv5ItUxUUQ{1
z4|L0c;BG3<GGDsBoRe1jlLoW>O7E>%0*+x3-+mZ+V_a}-V#pcfKZJYskC!2>EffJb
zmoY6j6q6AdCztQ~GX;Mr5WeSE_z#4kfZ`#wT1tD;EA4GZxvloJ+L_ZFcK-ii3n(zg
zcE93e4~`QI_5A7qL=Ha&IsEo?dVBhE)@E?RBRDm@eII~6Ncstq1}zQ_Kv)gK+t1<8
zgOtkBbE3ycsWm;0e)sgo<Ff`&GXUwKuK>qqXCSVjIa=kl+U9>Xnr5ky-azk{I8N(>
zv@%j4dN9WK@vpbvpCk(_MX>SL`0D?9eS7-%sfw<&KpO_DrBG50fB$@n9}J+dE(|6J
z4|@3Z&+z@}kDhAfC22|ifpYluW9UY{vGIZ=<4J1~XU0=jAo{`DAawkcjc2|}iKLu4
z(?bdxGLwRtl3#z$9#$!$6<UV{#gHi>xkR4$ea;kxiL#a-Nrl-6sZgnVD?s(cvlr>n
zGZE{e;Dz_PUnNIof8Byt&1al2+U;q+j$$+&A=0S1DNW?!8j}{CP!H3Oq|<DqHEGd7
zTVb3p9FeFZvK|-l{V3{<3&L63qdv)_?5DsU1GYJ12fu%H6v==RRrXQx)gzFHl;1ka
zN^W3X3hhikTQB6un2|Aiz3vtV=(KQP?y#pMpA&gXa}XINuCtMEH`sj6Rgtn7GPj7}
z!r358;x{a3gjEY2Qe>@J?|x4z<%PAT=g?)XyC`ZW!083W^CmtN6_ucEzXDJ9YlB`G
zgE#QLi>iMEDuoL#fLYhZp)gi0e6};H5rm`kvtyR%D1Ll)M<a!A>Iuh3)FE;1GyD!#
z11|!EJ+0PJ+_k0Xh~L;!^s8Qv&sHiHE~bo>fEoke72JwwFATW|uHq9}6X<N@f~%AP
zVU+7RTqylpE~?33PHD#$%LMu;_bk;t#g7j{Lr{Ms`tNA5XsY%!tkHTKOJ$&KL*~8q
zWh$P_ZByyO;x40U`z+(uFR@$gnmo0+5i>+NSicsgrWDa)Ys*}GwnnnGx>~*=37Ul3
zRIdmlQJ||PR}7oj55-+!4Dt5~n1or%*a*13ZH=Wn1k(xla*o?ea8cB7M15dS0E<#Y
zQ+a0$81eVE-r9Ph0s7Wk#e`$7bl%-Nb=ez%#;~V;l;HURfXKG5_RcY?U%Tne@>K5p
zW7*hR5h$!XW#DUk`j$47+@_FQ+IHkpM&^d5<D$YP9NgV(Z8=N>wNbd_zTQe|f{U`y
z&t8|0nk^WA3*BBAJ;3h*FuSp78rLGOX2y*psuFgZs-)7Zo3at~Np+cHBeh`=NF7T@
zZNsXYGLb!uUl55UNgbb!G6#klr<&weHW0cxgB?xGXao<O0!KC|haisa{Pe}u`8>Bq
zdvOSsHJ6J)G8fajAiZli`KH9#2e`a2tw{5f*JtN{(Jo5IbHUs)R5`HW2iZXl-_&kg
zc6fmg<x^u9fwrCG7EQ}G7$`9X>^X3G`EsTten~Frso|-+8QtXX(ncU$=FgNU&j~B1
z0Kua%ho0Ioo!%x-#$suF!d3h&l*W5IRr0LtJGP&zYi&Vj9Z9<M-!tMTKQql4E*hUt
z^cnqsF&>%`(#Re%^3!}dr6hE%mIdF;5045CQ}N4R^LW~V_;y{=wrWzhCL>2&6R|Ot
zdDcn-jHl(88Wi19MOm%YST;1a$Ao}|$DJ3}OqkBmz%5Xfdt@sl&tLU$o;6$3#-8yU
zt<ZR|uF2ee$!)IbPrBx-+A7bsjo#4YvD(mov9H-@RY%!y@>SmAX;}D=Wg{Jm{I<!S
zRhL`xSz{^K=2h@m(Oc}m<2PmrWu<UXPv7HxO0uMz6}_vaR3mLy47JacsbtAHCyDFZ
zbVKcwcirrCClocNly|7rCVPWphY?$5FqTxXjfILY)7+Or(KI;X`IdRe`~~FgwJGy|
zpQPEOccF^14lT!$E$Ay0q>uBA!>c;XvOId3eWr|6tdMGtWnDBsYU&_Z8^(XkiOdrV
zFWX09M))j`6b~yS3_`odC|2uoNtO3j{MURlRiLVQI%$M;>=mTO>Ik<2wVj-BTDmK3
z+@Xn>)8t>SfUEvp_id)N$VYjUD+0oQR88Q});i0+tralj`i1ptwP>zBn&+BwoKWpX
zl|;3bWSrIXamw+vr*Td8%JMFDj}0!}OtbqfR1TM3kJ?l#MGe#Ku;SZb<GapwvSh6}
zDNa|B3nA{3r;F0kgSZ~Sx66NXPNrd@Og$Bj_v2V<m~1@fA3J8CEweScKV=AinnuH?
z-0y)2!fLhSp<zdH$NRmG@|F9&ZuTeL?}2jv!~I^z^_};79kmPhd*!&|RQ`&&N;Ui2
zX|%*+-SN8n5170cO4)_B@3>2&%;Knx>L_m9rIA%d@C$42jUb74-lgq=pgGs~B4|$S
z18#^r4xPsyxgqW;UcVvkR^a%5@D1^G-jCc6FKo1VS*PMZnt)sQt@!61$F(J9JABn@
zd{b?yk++a<zwDx^8>N17A?f3O(1wb<ULtF~!KIc+IKTA4zS+g=_Acv_lFOW_;$2S4
zC11N8`dH3UFJ3Co30^NyTCSF)2mYnQDuRBmk$l~VsvL*&$vAdSz2y!FiJHg%00lU7
zmthMg69P6hmoZ=|6qnGRExUgp-yrP&`S^3Ei?CV+8*k&#KfQhI{@XRt(I~cIAn%kj
zcKG+_E`A_jWnCB`ph6jceH*^-{_4p#W?I(dAJ)UKA4B)%y;v`3v7WY;d0{=YirEj=
z24&)>Y(09_nngQ0$R!$**a=~w<>SG_QHpGpF`=gzsj=9Q#HK7ZCb@rDY)h4kaNsd>
zP!wgaKS`y<8&RopLv0io*K<R?khRT@K0<b-ZOB^knB|Hxrf2a1vXh+bio8+wP-N!G
zLKN>0`{+t(x-8XW9@Ek?J>AbpMD>%g<<f`!gVm9OOw5LIlaX|Ewzp!W^ZR~w)5$3k
z<K1x6mlWAd28x>>HBWyQIjURHi_I|oB)t}IW}DyiFjkTCm8&o<9Gfl2c!zSQ-|8K|
zbV2(GzQ~FpG`_ajkFzb4W@+!69Lf1V&(~<>0S5~td7~M*DJNcE?ts3|L?TOwH7+kl
zxNyhi>n_P&BGl6jU9zKiSsu!6+FbGmdT~(B+FR<iq<E)LWAcB>Yj^ihV<T0b{M|#1
zg}?V<6$ReG4P%X^;s<kt$t<AdGGAw;Nd*V%pq{%;GUjusVf<yD6i8{nb-JRNySCc+
za;lY@r108!d2QHwE29mPsj@W3zPz_^TDOn22jhi<rJ9W;U2t|T@?C(Ykl$=b$$)IC
z-)b_})jv_+l(T<Q1(BDUHL)a;7>KdW1-OFKMnml+<On)W&xhFff{Q%Ep3OX!*8b+D
z*bGurom0$}t*{a1C>H2iVMG;x&UC`vn7ZDL99pu9@tDSox2WXQaCM&w&P12&Il^na
zj`pe`bt`Uow*qTmLJiPTK_Xt+>`C)nQ|nxq)qrEB+;o49x=SQAD72a?n2P9W>XCEM
z>=Vk0l_9%La*;8)UqN6;9J7AO*}FbHmz70N8tWILM*XDjZGP(N^Ang6Q#vxb$R2BC
zaom;0=oQ%sT^0=!XWnC~a;cN3@+28T(*Wy^HA!m!;U|eo3dUT=$~<4Vj1wl|6LNy2
zKwQ*GYL<UZa#efVsXi)X?=kwUT;9~gzIuBqjY>D7rG$0-aTZaN_jPG>!`B29Q;e7;
zMFB`RlUeXJNy_gyG~|a}Mo1oVk?bg?4`b~9Hrdgz%y%o<o4l9FNkQP6|DAF&35T9R
zs~lH~>%8Rhm03$DJRT%Vo=93K+s%O&Av~6(p>ux&T@$4Q<fp0^YOjJ0ef4snoPoQX
z<Dd;H_#48ECBZeCxV{pWy0#>|abN>MIypz(WOEGDj5j|XL5l8%R~S>;O(*2R3=kP5
zu8TMC7Z`lW70ub&cHD5JEUt9eM1+SKe!Kb%POzEuH!I2YYFn?OD!C~b+f(h();!O^
z%6Nabb_f}PZh1)BQ@peR9lr$PQSuUaJDD`&h84cN7xn@t%>%*>u@|sW0I;xYJ=}2<
z%8fNh(?z%o3@HbE7I&TFvW2-Y9ZWB|+zE4q*l6*cAQxzpis3~&lf9ivrOroOpDs8Z
z=Tp$DRGn_cN$3T$CjT%y*YH#7#->O5iy(i#`VtDWP6oGVwU1hIYW-I*V^i8W{IE|8
z`zNqd|KCWBa`VAC_(>i)@2jXqa;2Tmd+Hh~K`;TQ2{KmMo*UGI8GpBE%(tT05Gn7u
zGnsxn8I*=mf%e^^WFI)sG5*EmY|ruaG{<=K68fF3yHyJ|+uqJgxIEiuJGWmChU*s&
z&CcHDaMjGboVK$~mjWci{14Kg{^XZ&-z^lEvGy+&m%*eh!+)34-*10B>8$L>3;R_~
z(I~cIu#S|-E}DvRnq^d$(MY6mkYAcz$9s@Q*u<bnW~)H+BpJ%nBI%N)#s5B!AXt!#
zmXJSzlV29Z*rYp%&tp%v<EQEpujNS_Vh=m^hUb}Fm;bxci~USS)ZnFzoFtk&64&Qb
zJ1=^LS+C337Jqw_Taz30O0r;T;bdpIDmwZ+ew28g#)Fydb8E(DDfgIKl6{cJpp16u
zWy|v;*SwEUU^p3ip6raX+s!7C0t#g|=W~=~>#BK_uP>P+9#%PGQ5nyz)N!^mnxdhO
z40t_#usY3Xs54=8V#K-r8_^xF(n|UypD5$J%U0x|h<{vX-wu<6Z+o_3&??BSMxy>y
z>`CZkoBCGVY)x0^^4Yb@c%uy1&Z=`+F*7F7$Q|)DnE@2yRn#u^c^#RWTl6#s8Jp=z
z(;ya>>xj6pV;4Qv;x)u&-GpAgTIH)j1%W1J_0R1i2;>xs_iE%o@qJgkDv&~plmGWy
zD`oW6Du3AMuH(;_=IZm?(|?_D#74FmGyL=8Dc&GpWnB=Ry>QtKKmQ%RJ^k5=%^QMN
zTAmrdSPwtH4^7M0yu$@87HVsmmlj&ZVmt@g;mA(R`ZW5T;@P962+dBEX1tdu`-Cx}
zl~|ZI`Nm7eilUz7E2+O|5%nj@9@1{yNJhQTYk%r3>()4Gdz(&=k%@5?!so2hbx5z`
zX@hbY?!Z&aoQrrc1H^&SsEd}bH(<UT70qnN=Q?t;c6;PVS?p{`>vaviwRI>snW26q
zYZfi@c`cZ+SI*TX-oSIfq!=q}9FHXVQX-%4C$%NzWy5MDZIRl^M!>ckcYULsk(o18
z6@RzVvV$9HBkb#Rn{C{;V1+0gmZnz9`EmWwD{HBjJO#CtJ=bSsQ-1<Q*r5P)_Xt&@
zICvSXmH6oai9rR^@dDf`i;!hA?#TPd%0*T_n7OAYySt+#d5Vvg(NB$#YWwlaUT7Q~
zAGz!siD4F%yIEu@v6-ff#rYwGDHx@}E`NOSF1o)3S#q*71%Dv2t;Ac9o#lS|b<G)`
z;Q1JTm_5OYlK1#Sj$s{lQV#Hk)~X=<=_DWP51oVs(4p{%KlE`qK_~G6f9SnJw4KBa
ze;80v$L1zCd)KP9F?_4}aKqI1t>UWD1k8G3AWPABX_kP&+Zas-jal!tJ9Wx8^M4@1
zh2KkgC*}$Tmuf__2ohB{D{CfrVc5cxPW2Ak)~Vz_)@-*#w7E93Tc0SS33aWl@p1N7
zF(-(1Jb(i<*r-crQbALr!OJC9{8Hrw!->uF{@6KNy*s}&ItW3r%`cgi3vea9Y<^QJ
zx#7Yy>5ZhV)*mpX$#pM^0VMVa{eO&cl9+6ymhXtbWkE>r(nw7Or`7`rnAsg8=M-(;
zvbSYQG6k~Ano%P4QVBTz`rEcjZxwWc+PX=BN+!iww>}x~hiWPHS3$KL#T31lb?d8e
z<vKtkjX$^R|86u8#>zsA2CQ!~vg$(H7jrhgSO8QPO6#HG$w8?;PjL}Xvw!~x>KH^b
z)8S-t-LGhLN1$U66<X+#zZ>Z&fuUnFTsq!OPZ(B$Zin|{Gj14_IIJ6ITy|F_sQ#)s
zmnQk7*tRR0W=|}X^V)X_-kPJocKj2gK1);K)bNR@9OAeV0-Vy}S1jC*n;Oynwj7J~
z5>L0X@84eIB1Z{j8cB<zrhlfO&9<fz+#K_a+AqrGqdmKO*_C4iR91Jkt~lh)LjWL2
z0im;T<@hV>Z<ZLcE60;7JLZ<&@#sOqzHZbt3F~^3MoI0kl`sV}$A_unbR<K81H)7_
zaV1R>V{B5GPq<7^r$Q^|5wBxBjRH;_4)2Di?;Uw--LRr|;NKLL0)NLG&9zY}k@CWp
zqodLoH4#2AgIfjg`@;sAZ4<E7XeEeRh7L`Ce5w9xI+#0;Q)G6_^h;as*MH2GZ&zQm
zQNecxBwEKeBJMAm8!*ao58=i>43cAN+W~lGA6^3ycWH7edrRJFI}SQoZS@%O-5h+q
zKXA`MaHGM_q=6*n-hZ3+iCxz19yN`xyOg8k!`L{H>t@ShHcX+LjYsI11Kn1K4Lasv
zwng?D9j~Cfq~=J)H3Dr=(~@Fs9;Ie}&*FZ1jwg<d7JDu+A5Aq+pq;-*)Bt&|!QL;S
zVku+ZVMTVO%Lg7|qGp9;Iu&4lB{U#jOHghCIWZ7#;mspy34fqZ=LW5Na_d2S3VuV)
zrP66dy;I(nNmr7*l=f_JdSZGH=44}J6id13zNvX~vNEKAeoBNM3J}o;#htM$bluDy
zv>#`BDIxQs0~I2#YK^KucphKQL-x<zRWyV$pywsXHOnf!$3nvN>^!->bU9xvTzHR>
z>+<oY6yLknoqx;#4JaXaYuL2V+TglL4mN7K)+!*^O68YqLV|5>a^Yb%o9g?IiCwlj
z_W65(WAB+$C=&ZnbS@e4=~rwW-z=QHZ<i%p*-~cHJ9{g!PTDP=Hr}f`G}x?>PvQ>Z
zn!EKm_-(Kq#T<zFFWE+|mN{29=X0HrgzW3h3^T&p1}u|V3GRrs;jHw5D;D0mv%b?(
zBbM3@<{Mwn^I8Hbf9u<MP3|bU>HplBP7#+Ott}Ln(Ht%&ms_nZ3zxB(DJPfP{uc^=
zL7-o*A54+M;gC}t0^}~u^$H(gBq`9=Fb=Y;T%0<VE6YKyzrN3KwUP~4v0}-w&w!=T
za+Vy<`@>zyp}62Ahw5S?Bh-;(5HUE`MieI`MpS&Li&~;9A=(vog%b#>t`-H6qSKn6
zVJNy7vR#VgVpN+uCR@bjB#d$!C5+~OHcA*R-6&&`+^Fm*R9iO%NBFaA1(XD=!O>At
z6OC~#hc+sps@;q2P=$n`l1qWxEbJ0h<4f5kj&^Snmohiui%X>zqpIT&1#2XA9NWFv
zD5thbX2*<eVLI?}wlz3|g1)dxjM`MVCZKRf&O(lbOl1tUikJi$*Gkh>#w=BT+bAe)
z`@p2Rvq-i{iy0}4gXY?`0?JsR9egw$<OqS1NeX6Y#H1=1&9&(YVu-d`Fj~n*%>bQ^
z1OuIo1Y&eHG9fz~2~1pq-68nUrYrbJ2~CqCu28gnNO2`~qZN+PMhm$j(dH`&6gJpQ
zTRK@FWGQWz1=mU^Gv=$X3|AX}t=V=Y>WT*22PUl{*?y}ynAh&D0aP|Wtn6y4CIMWa
zGF&AQ>`__TwdkrjCcGHe1WqFoaBu1EEw0v}wnlXtxUHd1V=-GoTus4dfuWRBqsVD)
zmVp7ZhQ?PwnUQNYs_8&6pEivv=2J1hQh0j^upUGxAS0Yg1nwFGJlU>)S^$wY8OdS^
zp#{=zD`pxA##_zwFd{-MP~$~_#%T5wL@-FL&%XbD@b>qM@!-Yz`D8YDb9FqkU;C%$
z|8avKCexGg^e`X`grmXV20sseJdE9s*W=ObFlO-;vcQl&T5KhspnllFhe$_*7Y?((
z8T>SPI~lxm--)o61f?f`EfOl!mj3)4e_P1__xA-Ro68`nc!mG1lOfhX#3fb4cZRdU
zCVGnyx33GGx!CWW$eY{7l6^vK$4p+4>SB!^ia&aGV;_STi|I$R)5$p``p4^^?f<*?
zv)RSvUj~EUPG|40j{Ruzaq!{l<Hg>|_|xF;SHB$i4`+kpv&r#)K*pmgu^e*>A=hJ~
zI2?_}q<V5P3dfZ)@>U$hFu0tK`aA!$4{h)5Qz2~f*_Ns%Rv#_5ljn$1P}xTWRWVJk
zDmoP8JSi|W2Ze@+X`$$she-GtQ@<Yk^I!jVRx^+Q&#%tTj&5bZWWK_x5ldSh+<=p-
zCj{q`(+z%{oK2>GZ!U(Tu}gM?8ypO0v+?xYQTOw&pJzY4nGI)S7rP6uCg(G(@hY`&
z;<m~wWQs_fx*rJrXa%zUKst(%)P6jFJ~)_8MsLQm!@<GJSA)0X&$FXz`ac-{Hikgw
zv+?=t(n?di3wpksTun#g%dVE$rC-J;r^6p6pPO9-8NMig-1=ZR#hpmPh&SVE_Pp$h
zY+L@>VXL#YjaFvoM4A)S4^clviy`ibrq-kFQiK*CEFqTY8G1sQ_JRr^2$R68Zc{mj
zfH>k(s6>=w%G>7%=H#tQnahC6>XQ|RH8OnBb?+ZNDf_O>t49yus07~%rzoW0ZzM%*
zt67dlEW<8;6cEHnTc|_6q;Dhy6Z-%s$fdLlf+f7D359wR%RQZ9^PG+P>m|q{XS1Sq
zgz{u^c8bW^DdipH?7l>!-SKD$9GWqok-U*NyxtanvA`j#YNu>Yx$MB+MZF>!_d=kt
z^42%66XJ_S+7dq1YY874?Apu+PxjaH`;E;1waK1;X(icnxKFa@vP||Q^R%7%jGp?8
zp8AZQa!)LnOHb6E&|9+p=CLJc2dsachZa(Y=M<k#Xr+qKD&@A&+I*A1Q$AU<H{T^N
zm#?Icz02XdcfXE5Urf8kFK0A<S-0CMS5zsu#3EJJ;)x%7qyN1co-B77MfUcAf_ITF
zC3*vYo-nRwz)qsz2~ZAh1B0(tCg4Aai<k)(Ww<-l9pp=*)UDb}(ZI}lAxJ|_DF6wh
zN+aVd$yW6d^~ci^(xj-q<VTO;(d7JcHXWWKhjwO1?*_Yv9JX_Dv@~c*<6^rF<_JCZ
zcF)f)+BUP$9zJA;-i`j3QG<Htu7FC)_x0_6uP6=hs<`aB(217#=uOw;_yc4-{WR{B
zUH10nT5+&ugX0}Q^%_GDKShu>Z@|yBoQ#HyH(+F`-p;i);LN$TS$A_%3f@k6H{e9y
zNZ%Are{EBh(@Im7<9(W{T$fE1*R7hNs1X2%mWM+?!%`o;d|tjEUX15SRFPcjO^45a
zbVql+$~E_8;3`G)w*e=hxE+Dt%^59O32nkDK$S5+gpb}h<yi*_zSdz&Tya>@4ICDz
z`KH5`>kg~)Ha>@*=&wkBMf$TJ>5+o;NI`m}AU#r$9w|tV6r}FpSSVK??EJb|fB=?e
z_L9DEB*X9nrXX5EDpgYn+a23NQ*hOPvMfd@jjZ*l^*99`PK`S`2S6AB;{kj6fM7O-
z>m12I7b^;flVwi$7p@5qJhATJ@W~G6?a36%j=<GmPalC>(PMZ~z2zK0Ictqk87Y9!
zy6`Sez`7<V$LKZkq6^9teGz&1Hi$L*Ld&9%19^1(kKPDf%uYX^{%<_}72Y?0p1woQ
zocEszC8EUbn(V~zzp&hNGXD!tcGHxDf#<r5rzY?(wMTO>J0J_-g?Kxqr;vOstC7bw
zbI=N}k3mTRymsOA^L%Shs~C$%^!$t(8d0&SuAcj5<{ER<Dk(8ykau$SjBcncIv0Xc
z^kl$mjSw+$D48lWm!ODZsbCj>=kRO~0~SJ{zbK}ebwUI(JRP~RsssiW-+?`_DXhAb
zBB6(2d$nPW6kl=*Yllj>1EXAHSRxPdk#!SJ!gtw(lPh_9G`|5S;EX+{zMC^8Af&tz
zCn2Pg{9zvY)m6w+<9#~7D$5SAhOHnA0w2hAdES=|_dWVy$fx1-ba;G!)`OOilb!si
z+ypLlBj8e3p##Ikl+q@gWc^b5!+;FUD=wz#9xi6@!Yo~^hXS>KXi$5Qp*_gZ9%LwO
z=Z{%@;pOQZ+bI%1F^427_LfDWUA5{~tF0zTU6kq+Q6lu)R_eJ3eW6|>!h_c7UW6P_
zD0!U-_WBj7GlVP*j95#5+QDlB_C6E0_X9f-iu$l#3!C+AU-M#7Y4~LMVi9|e`H5xX
zXb2SmE!@n+%6If5*BBP5m6NZlr$@JO67V8z!WkvQX0iz<)SMrtcX)`zF7x@WC3btG
z;U0-S@*=U@dxb4&+gpWmBJ~9KMBXl+<Gaf8(W%)oi9Ug+Ez1XgiX3@wAMUv2b0rC_
z!(BXAk^%tu=mC5&1=PL%WJz}ca#Qf3IeqbkZ0`WO5p5)LWs<sDG6Y%)luNCgs=eS*
zc5%K0^ugIF)?|}`eQoc=0%PTTOTvb_qM*SDg#ficO25boNZmor0HCW-#9W31n*(R9
zl>xyYL#U*7z7-Vh|F!pR7+xN>_kZthX>W4Pb7X`?7RToQ2M)qq<Ck&YEfkj#2R9S~
zH#L`$Jpm?{KPW8-mk?z^1ef2pEsz29mvFc(_kW~Qb+C#-^0$vR%4$)z3HqW%M<K=W
zeW>=q!blzuo3;UMY3(40zHb^{QXh9Yz@(D(Vmaf4k`54bQ6W0X|I4<p-m64dSx+8u
zab(rt6nc<ioSWcrh-i!opwSI(i;_U01mHcsLOAdmCMbJ0H1PT_)14ddNMq$L20cB+
z?tk9pC_ej$9dn4J_*kIL{(?~BT3GncbBS$koD;WP<01RQ<Q-6xLeBfo$DvgLEm$>X
z2hl)@z7dRTi!6+undXFa2S7-E6<|txMF;&&j!{J&dYW8Nyw&7S%jkEHebFd{Ao<#}
zFH+@30*cx8ViNu}_P10OV_e!l>*N<;fPbphWP3tjP4Gqp)Qr~=0V}rN1HfuBT$S*D
z4GuJ_6wDF_Su^fZ$1;tQQf;JR>wNTF`y|RaW!<eOb2Zd<X=AJ7SDxWU&BUq^-;|}t
zyD4f2$>n4>MWsxnvh89wl{=*!bXJ8TdSw#a2hS|vL>Y<i;0Cl#fwet5CXW;Zf`6oJ
zUUHov3f(liJ!Oqj=q^Vt1%hB1267b}3$Qc5Gx)EI0=4^)+ITJPH;z=}Lce~euRGu8
z__h3$+S~E`U=-+vZ(iHkYWwHv+I)SPWw4KRjOWxXixxw<Zdt;JeTQAS*7XgX02PF}
zcv=y?0ZQca6ww6L-cX`u-B2JAtAE@KrLqMesJXGtzk|;aq-)Ah^Li8G>u>AVM$Ixk
zUL^}Sq_86wycgjkW`$Ar5MH{dqe5Td0W;zY%69_i7TU+KY$ymB*R)n!QGRzPu_h23
zf--L2Nt{AF?;M6Nv!ha`s(Xp!HVcQHT>i{YI4Gs@7veNtGhcoqjjDL?mVXVP>vLO!
znXmdtrPV83yq{{a4)!uK#6%t~h201fd+M^9TAvgpg^Hd=sV*rnQ$cD+6*3ZLk)~f+
zSu6-vg+8y9Jm!I7lvs`8V0zat=AK)3BL|NW*dyo@?O0AVqZdjVit-d*Vk6n)kF1Mj
zg&-V#TdP0?Qt~LNx>b(Y2Y+I=pI^Ys&vp3dMmXtg5j#dLZsgHDi&9|>ao&f`0KEt~
zV~sM@EQ>b}V9q#D4x6<Fif~F1yPO2xKQWCAw+HZDd)lNJB<qE^Gg&}>GKg@1W#5qI
zaM>ENg1;BEEk)Dlz2sU#+g9~3IJ<@UtGom)C|X)FXhy43Mz|E<fqyI3sfOocA#RSx
z)+>kI#s2uw2E-VuV#synaZUoto9RQ&b4UUe4xT@ZE9hyfdo{8w25YA-S9zZGmtpHR
zqTWua7OATONf2s+;0(<7)^|Igfk1sI6;%DUEd)h3`NQ#AyEasU;@&}5Jr>(tf8QUu
zVRfSU*luFWrm1Ti5P#mH*frDjB6GkA<dH&%ol2Bq9@)B4&@~}wIs^%}9OLA#%-d>Y
z$s2dkx>0c6QiE9|dFAC!K&o)UQUmDQPffVuwm!6gxXFw6Tmx-j$t8{|25MD%jYlB&
zD%ki~=Ob1ize|fFkPaTarr7bwT2OxH$Xd+aJF;GScyHZWRDY@KQI4e8{V3<CSC2cj
zg0VZjdhU%Vug6C0J2rBGVt}3xbS}7M7XWBbrt86ur1(ER%M+}=bCxG&-(C}M30?%8
z_lytaEz&$taVZ$=gmC1)TWM^l-*;SLlF6y#V^Gf9jnMPxN)Wsi<IBK)LuBJTP7l}2
zbTWJWBZHuQhfl%D=ODmG8QhYcT+h5FDfTgWk5kL?&mF<vUMwrS%VP5XD3oDPAqL4{
zuTZ9~(#bW{?+j(&lug$q?2kHqjZ3e+=Im8mU*Asu0nqTpCYN#FEffJXmvKEl6qgLU
zEGK`x8p&=OIrgrvs2?!H<#7WHp)DmcFt-H04dyTiCh0@c=<Yep@6UVW<5H|DxfCk_
zESHNS-@emw`Z~$!`EYtUyt^2fya-l#J-vLH6n-J`x01N%?Btaa&Zz0-`}EI8DYe7T
z2|pmEG5GQLcfhATJ{T2nhqCy=Z<RVecu;?I2KN}9&l;D{7~G}L_*D3OqIEnUrIQ2!
z;i)ryj{kc3$3Zf&?=DeZqttm22)j@U2o(^LMQniBEqx5XBXlGnKi~5{oA&|aJ?Uff
z9>Av!M`9)&4~ejMm!Pf^WUX+}ME9P6AgeNqPN}yNSqYlcIS{$7UkzG#tp*~WE~I}4
z<%!&QwC0iF5vi^tOXoGzang{8j+8`?=i%+c%i*s>m5esRx;WWjg_SUU{(hh@Kp_nt
zm#y$oO@IGq`f~U<LS-u$pUH%i9;8EFW_r+0SQ{p%K{$Ng-7u2;=Q78$rzDd~1R%>E
zjX@dVVk|@u5<(1o3N9>3z+?5--(P>HVdc|vgX};A@FQgPo<1|rk@!6xmjpxG40X2h
zse0E^h~5Ze-H62$tEkp6`09pABC0}%g)~0oC;$a&=xo|$mzhfo7cT5d6j{zUOHVr|
zqBRTY&H1Riy%Jh3=r%h{$2{>G?cvC|j4#=hJ2SG#HvS-+>*gfUeW9B44Ca3iSdtVr
zngQs&4ar5Jq#ls`upLQZ5Xu~w{lKihfy!c_Gr**?<max_8?=~c++3fDqo-Y*XbVEc
z0pD9~yIAX7pIfGF1vq^x$2%;)S$eka)g-lU<s7+fRKkavLFIWZ-BbZ@6fz~}P@+?=
z8lF7@Jhn`bC_*=+5(bTslVE>wT1=)~-q^^3xa2eKo8*L!KRSyl@e7St@D<K1EY#H*
zzC=NRc46F^PDiO(N&en2^n_8FQ^H;#tmLhSW#YP8(HMs|67?v<m9Vp=1rq>`RE-eW
zCKtLgUbxVVk)#-}*IWhzf->XD8z{uORfJ#{5J*QTORZFar>;}tF~)!R1to6?A)6(f
zaRTfD6Z+DFHm0|xIS;J=ZiTD>*q+aG8z)OlyH#>zK9MNZ*KJ050V^8$nPoXuUu(5h
z7K)R!v&Ge1f*V!@Ipl3oZ9AkTDip(&ymqbuY)hdI>tL-HB4a1PN_zAYa6xz#sE+<?
z!btc$rErK5(nK`n|4n~sStv!hQVK6)phpNAD1#i{{rUCFZ=e6I1fKdY?ZjFf6Qsgz
z-}5WFYX|}RgKzvE1e%?9lM)eamNKxwZV&3Y^8R@`0Je@`v~HWe(V{^Dq-{G>^&f!v
zF2K8`f<M74s|DG&ez8#0FeH8|5%fk-_$<BR`ErV0>C?ZLt$2Ux_v!FuI(%e-TI+(x
zEx`iS6(NGsYXU{oSU;;?sAb6nO#8m@fEyg@b9HgXEU#(g*#QbdX%VBx-l?Lh{FarD
z=dy&L3WTy$p5yw43<gGVrNvRKCDAfP%!a90uvxh6@6X@)*jGIC+&wdx2W7BcL`V;o
zG;AEfO3fZBb~S$i<n3j5H?K$`Msk|WjdM;q8Dl9RdjA@LA?lA)M_OX>q4D@}uUJqy
z9&^L%Vi~KnCJu}xn2S{v$3`~e?U=O%Kpl#hF(X*_*g{mv7dd#+x4x@XQZ~5MReRJm
zcSz0`=Uh8wsKXm9AjC2#Lb^_CaC81?l~wMGD%Cnw1^j<DPFG~b(}HtdIhVJ}Jz+Ta
zCn)c&bDGR4wt8-z+X^eT)W9_+L#y~B56x(xV(RKEau_S8qHng2qVK1p4Ue;WvWMF{
zhwR#KUw3mSIyN;J6flWU<3ZuiU=flUlEvz$=gs;+YwU(FNevW_^#Qz^ZihB&qgKLB
z0_|dfjro5nRaucTl@3F{v=SaVS6m4X&DV2y_+y&y;vD|aCGl$xT~fZK1m4;hHixF*
z+MKX*OACJol_{r%^=hP{y5`RqsO{4(HN4S|gvrbOtkpZA5%N!CePal;i+vNK@WzkC
zsnV!yWXxhwM&p;JW=z6+YPD<1nwO@ETW7rn&{=<+A7{O&F)A;v&wBI3CjGKJ^O&uC
z;`#5!EQhBEpitMQ&3`#lB&N7O*J_w^6%6UGhPbTs!!zrIF>!(A+{Blx!q&p$)!|lf
z0_({EAoVL1m}$Uv$mlfznGl50H^^s`a+bv$;5}yDL8U8|IbbbUCt77VIN!WFLq1)R
z>kWVNHtg51SRN2=&D#J=Ywbtsb_5RUAk~NaY27*G<F0BSAl{|b&pBbM$==!O->gzR
zDq>i$w4qZYv2n+yY1nOCS2>3=L&=nP;{a5cncA;J*KDPCgrYobi~(xUhB#0BL*Cp9
zuP9DKov^`v)}<-A)Qo@an`@;(iS{ZA?PGtRACuaPdfiG+zkaQ&qY=hch;ZuaVAi}1
zJ@>orX0-<qy}5cfyBE_&!G;xs1vOy$MHj3Tkv!&*A?a$39V^`Zf)!0aIm>PpZ8`7l
zd4WUyK*7q@RXY(x2xdHS>+K`PVGc5BC$~u@e*Q=f2}qAuJ2-jE^_He_z(QgAT<3qg
zb40I`zhxRmMr@6inB>hE+B`WRhi)MtSdYkP>e?;D`KB8rO|e76&KsnwQOMGm&5saE
zH^BEw*P{#lPs4&H$TubQuQcOrQeIiVP2z8x<JJ|Zf6bhD4v+7XE?s_Z*SY2Vxqq9@
zROXKJfZa^jnP;av=4r&~m0K!6n=^k4sct*yW<wP%2Fj&$p}QL?d7bZd%3Vp5JjsRM
z%H0YF7JJTB3dq27tfnd6QEms_3~Dd6&f@$Ii(?yB$XV#J>3ZF;X?5U*mUe`)af_$C
z7t!n<Rorh(VWW^}2Lt}Wg7-Am%2UkIBcwad6;z-BK8zTWE>3M0m2BjR%V>WyPfvSi
zBqm}>5LbBV?}_vD?`~5Qr!c971$V0+$;ConF{+iGdBbBdJ9xQ}W38X=U+L06%^<mV
ztaO!uWVc&o)TOh7zFo3&iGkl$0V}Erqra&FzLkOvOO#j<ldfhI{G!SjJsOPvMTKzy
zbUAptLb+a!KTsd%YIjb^mGUB(GM<TgIB>N(<`h!bW(kq+*F!2x)N2*k_(To<A1mYn
zewT6IEffJZm%&si69F`rajY~af7MvYj@&j7z0X(Z2TYU2DlWjF0dH*}w+!YM<P^j?
z`4IpAA&+XZTPjJd(X@ITAJm$mq*`9>WxIbm+kJib`26tx=(&p~iHp=df9o6`So}JR
z!ysLB&IIqe=kMJgkJeJa^${0a>piZBzZ33qc;{Wh8%}WHsdI;SkzM%)fA4rYee1*N
z8;`f>6YdW859J)+XB%vxka78P9sYd&?ZH-I&ytZh5^@(J8_;T!K>}vn8PW(iNV&_Q
zmq}bt>HRUeSzU#UBTcX;A}2zeDuD&CB19tA6c`F}zRo&XlL52lov-st$>5ssgJBrZ
zO99ixa>C6le8f47ARj73e`kZS)F<T>=HPfT9#esn)^#9gpG9uKk21m8CivN7)1%Pp
zwM(zj=lmp8eBAKe3r*oB?Md?9erZUd=k}?|b8=5Zk`i^+Fs|gqcR%2;)6=J~iwH8a
z$$5g{R7c!tR>NVSETPOffQ#2$+$lIF;U&RPPUr9n6^nH7ttw$Cf70xRg_CS-k#fwR
zX0V_c&-GH}%rC-($Qt3Gxxw@H(uYWl_@p`PY+H&ET<)qzBILM<9uh7ox0ioky&#lf
zFL%_-OEOFo<ccr#esmn@m+qn?Ga}ASbabs~Nt%t57N<TLkxQHu-mMM*DDeXwLPafy
zB2u}1RsVf?e)xBie+opzIE3V1-yg~Yc49)zXbnt6XYk|i?%TufnrhYA8jDX(KVnbT
z8J#@UE+kthX+uCA87Uch<$?n=C=ks!pKGE%)+7yO$tkdVb5AJ(Wbd=4*h57_vsb9N
z$k~2+@kb|5gT5rWzGG69$<B8%*%}{NzwIY--~i^d(9(UWe=@B@x`wA@HlE`KJS|KP
z^3HdzFzUIEPM+2=S6s=LMyI-}TsZed`FL&`rpU1gG|{1;By;GYi3qmzq4IIFP02@t
z=58Xno7i;|Rn8_OSu3iiu6eQVr9dX+35t7rMkJ)eykn$kisvSEYcj1i?xUgVi`0Jk
z>Gd6*45PE@e;1tLYdz}4p!4|*ucDeGcqODzf*@O}tmmFw>vURrg=jGPu%ez7m)f-^
z1$l48#i0#f8Dd(3yyp-vf!Atx3&!3g!>)Zn{>qI@;zJMBf}^@jNo8e3EG~Vs1Pfh{
zMc-^$N$}x#o_{DG#6J$4sPX+&F?Q!VySXYdqbll}e@JAq<CU7I?tT4Rh@tAWEDqHo
z-fqwuL+7k}aF5H0cS}m6WCF^V64ZSz*L*swsa1VJRgR`o9|3%^oWEKyTK|7iD-SZo
zoFNv+>t<mXHPe0zhB3WY>=(ua>Uv>uL|&<_Z3G@I4bo_7F!d`<dl0CDZ%z=MPiQDz
zrPK!Le=z!14&+`3u}^6AgT|QicurlxnUPNz9r&2Y!hq~^S;G9Z`FPA;k6{}V?CN;w
ztzvl2n)}zoVoK&6%b^!^jA5|@=_&g<Dy@%mYvs^;bo9}0Du-JoxLyv&61-74q@0nX
z;u{r2Z^b0bx0|#T#O>oA!XYKECAn7d;2<7pe?@&kt?_kg^B7)Q?fNp{9&fC66zy=X
zxV_vdg0HA|io%9{y5{||eY)m%WuHFvupRg5f-2;1z>hE@vUQrmwSD>)eSU#`dP3@$
zxb(&}yB0WQ%CceNH4E(mAt-{4KdoQT{AaM(ai6ZZ-C>`uIP7fOwokvK9(A9t_0Wyn
zf4+LPefs_MYWsA>;o3fZXC2!<T}kl5eL8tY8+X@zy2HpT_Gj+X$-(*Jp8Iq~amRhS
zqP((CS8u=3KD}@c65e>9uDIT~Pgm6Lx=&ZU+;NwEx~BTFeY)0rXU_-;QJr>-G)?i=
z(dV{(x|3*P`d{tS9cE5Xd+yV7(bX*?CqA=ft)*dYR}e9G6f1976<lMsG~GBusAwZ*
zuiOf#Ngd_wKk=l%m$9ZL69YCjIF}LeFB6v}&Mm`#rJVkG`{SUqvRVZjKgL`A?bF-g
zzr&z9pkf<3%T5_>x_`eP;tyn4Sr<A06%2JhzI0y?f7VjVTGc%P<&6NPt!1781y(Wp
z&RSIHTmgy%BoY`{3??z2L;s!5@>u{Q-~Xm%oM53@#*piu)44ZXkt{L%`SGo*p8WNB
zv^kJ}ve0N@KE`VXW0?UOl|fg}h|Wm5nS2-JNE}7cagClL{nP2m>geV)GF!#@S9ZHC
z0q+%n-C&7U;$*FKlmXmJojnn$AI?E9c@2V9P9ndy{<XKf=sm1-UH`TCK!eILG%M-d
zD-04Pg29>wAzJh^#`rW`kJhi*GfUyWN(v=^v%o1ZSY%1^6CJbnUGIJ7Q=$VlX`PHt
zE0M(((@0EGxMWI}u%eR_d;w|+DQ1I^3hFE`Wu%W!QncC}%jcY%z0P_3n7w5@(I5-Q
ztcaQ0XZ51oj7J&?35z2CwS4O{=LF`sm_1JjVHjUB?__M8^>Ptev{nVi2HEv2T|*du
zgA>MV0Hz74z4n6AXU?Df6nuukSeg)Ebg|6U2|he#m<DI%5ziR_J9FR^w4Fg$!dNN-
z2z3{WrW9?j2ObF?nrkVhU%Ot51w<51Dk)<#g+_afJI+@YwEE-}xgH_`$mLPInTVlz
zI?~TVf(4}m*6X+BXht5ZHhZyEq*T&>QY@tKyqn&jK~aE4@5S)#ODyHoW4}n^Yc_!(
z1jg4ZWj|^q;?Rr6$ky9bVl?#jEdFxI7bO;4P$;$)o2GhmZ%5;gRK`}jG-g!fVM>0<
zA5bEBL@BN2CKMLOJzvh3vI?@&VreO1AA6l+TM&CeV{TY^CEfxnsuzAf+mbYY5Kf$E
zqjW5`X55J*cealt@!+Kq`8n+)Ym#R51`jT^lKY#l+!R*Y+`IXTdkV7kDlv}Io*>@q
z_+Lg3z$<I1mK5uW)o`3+cT>wlR*E3{jV;f>=~tScpa@(e`#pIgm`j3Gl46a7q9wMC
z1`SL(TPbTlYnu8eI?vI$&gZ6osWd)mbd-Zr8L9U7$jl`AnSU-fGpv=79;CLfp1sn*
zNGs%9A96vVwH)W@r$(&Tx6Q9Loe8b@q*-)Hz_QxJtc-8eeqdn|pGyAn9SaiyN&CSK
zt>qo$l^%AGqJ?`y*i?9G$twxNPFciC@=vNPXuTqF(9fwX=twzp!3BMP<wyqd{(}cA
z;U@iMwaSHq2V+Cx>S_kq5!hLUqiG)YCr7Ds4iiS-QQ>mM*sd?2Q?kjdXQbeZr&{mO
zfr2EoZFDfp>$B9h@LsrJIWOK<1QET2l-XAVThgz0v##IGmhq5x?kgpOHd)@xr|Hu{
z2ItkMh0^8%0Hc;{tAciau~-R8sIf+Cl3=Taieu~4&V0R$=Oh%1eFcxm*q=_w$H}Rl
zduR<2MU*vuM-AD{pAd|4ZD($5%BqX}xq&YscqxK-jW0Ru=SsE;G1Mzu<VsSWTs`CB
zA#L8-56@<2GPm<53E1B3up#qTHQ^Te%qLs53M=&Ki~`-b^%_incjZm6M)|#)hkNrT
zU$pp_@}+oK61PGrC)g@iN?EqtVEbEfB@n8xtKZ(wl|qhMy2=M}B}j6Gt5k{ar^?v^
zTbOb-A--N5;_^Jv(2uzgcMwBf&}5VAMsmG{D7RWrcyeyQ9cXf{z!jF<3Eyt5PzOc9
zZ_}#0JFB$X;IO-Y(sw-dmWcnw^nKkIW#m<S295Fy>fxo89ZN$QO=Bxv-oqr_uDpB_
zV)IZVmx*BgoW?7s8@wM#z{bknF5n{x*reTnc@O~`BE`;M-P})8VE^X6T4lrLzM8Rx
zhau}b9KQK6I45sV&GCJ7+_UR;OTa=p8LBP0fsh?5vJEy?tBf3Ezwh=Vxu23v?2PS3
zy~=+dRHL_f@7~0$y&`FDeiLFIWsk@4vo6|J|0^)O7=92*OUvZ3MsdUGX*`RP{=1&}
zK%x#3wBGFD>FO@r{6FZ1;0Tv--z^jYIhQf&Cli<2)GfaOO1GNTErSFDyU$gZff6Va
z12#7?m%$b%69O_gmw|K$CV$0POON9=48G@A_z#RKQluWhBEZb-W`N#0m|M|P(cRM?
zy8r)BN|s-=<2V^7J@^qtBE^ps$<Ma=(b(qg@bYo^?c}HlCNczUKE5{~e^|MLl|j!<
z05IM`^YOF!^NY3U<$95etaVPVAn#G0e0*{cWd^WZ@-M*gDG=l}GJnTmyVm)3jgwjU
zDo>E-a~_xV$$Fb5B)LR3_~Tz6zaMPftToB!*Zk1GzI`12J@i{gXV|C4b1-0``TOS~
z|3SdUhtwdN1l0Wcr}=*PLrd*+sb>Kyf&gW`XPy9s-Z00;duLn-9Z)33u^3BO-^EZC
z!%4uR=l010f)y}1wSR^h5|RDw*N;YxyuS@LhG0B^)(T(qHG!>P5xoJ_3$WV?3|q?E
zrR*p11JP(q-YLJI+PT?>#hi4^Znm%0iX}kLhFRw;3<kk<+mdi1`ck(fsBziau>=@H
z(Np~~m&jP`oC+*LxD`aP3OKW+AX`Gx5()xoU3JV>^GtVI*MAqIk!wjG=TiY8|B|;1
zGGh%TKiE|K@;dmCLbs~S!0u!g-^Kpd*`Pe8tO;6}XO6PTVC?<m>*CmNl-d9(8^jtB
z$T4D$VY#AX*T(j2^!;tctI62plzLn_Sfl71ABS^ZK<kamzMCkpr`;?-Nu-XwQow!)
z^7T})&5C@w`G1~P_-Gu>tcXKY?r4p}WJ1z02i^IXq`Q()I0@b~B)Y8jr6x3igx!$H
z*e_+S%LRrc)kGh3fhiRM{e=o+Y|iTC#x>bbM&4`HNp`L3u=pc5?32dVn&M7F%0^IN
zpQ&Rk2f!sQv0#;5*}3p^eXWX6S_vf??Q)$;Rtgm9*?(@fkG6JNu@b)3Jza+MBA(<;
zl<r_Vo>qJ=;w5oIR#dc3M&37IKGljY7mc;zdj>f&9%nX`4r2+u)A7mMQdI7MBa(?J
z>=~Cz;x+x%%cwh%{5IT*ZIh94+;+BrGs*S&htjmlDR1n79wjt(&qF;GxIIlT;J1Ty
zK@P{h=6@-PkBbeGz65Bk%{)V2`cauGv1M~_5Pi-aNepRqFa5LDBXwmdGaRWyBX5_s
zv1{+=F<U)$mIS(#!H-M$mO4s9Bbs6DV{0h&XZPG`2vM@;P#bDY<gA}*Ccuqa&$Qf9
zV*)y*_4(Ad+nA2*MvV>CcDwAMw&1l7xw>m`1b>Evp*)6`N(*FsOt-nNJ0drf{6aRB
z@-=o^YnNKF4?P{7@a-mGU_*Y34f$kt84$U40OOIBoQP3sOMsc(T;mLfj%ziKT;kVG
z*heO{yYXu!@CJUZBwWF-weFWq{CYv(UHLV2ZyS=;8so+!&xHRQ<<|N)WQ}XNx1_)b
zGk;6>q^&4(0JH|(2yMxTW!|_2j0@%T44(x_3_=X=XQ#`>1meujJKGgA;bpCLBT>(Z
z)^oZiQ3ruVaMwJYq9N(s9>LR*oe`6E?0J>M9eKKve1WH{*$?9B=s>EFe=tv1T0hCt
zm9z~!-Hq|%3fX{PH?r1mxL~9J&+vQg>wo^G_(Mv<hP{^HHZ_yicVR+YX$BSWBCwde
z36i2Tu&<w@c)50f4?`~9o$f=5Mqu4BIcSNubp7sAs6h+{(<WZL1Av|-7vY1mfztZP
zY@noV{c3l+gP4Sd>Lz+uuQ)lo$?qA~O5#mq_H(GAkO&txe+zunH=j5Oo(-RT^a)e?
zrvCr|?0H+4ao;T!0XLUX&o3_lH<vMj2q=GAVQ-ta5&rI9L116H59UiA?|9@;1W22u
z>)--Gk^$QSW3Uyib>~=yEEm~+{hp&lONpgKiQXVCipG3j@g<MPyXT%eGUpgwa5+bE
z9PMv%CE;(#E|Q=@$x*gnD7%!^x)jI3eu~v)F7{J!T3VmzYUCp{#KkDqn+UCHQ!sy8
z38SD@UF1aeYjIJ6y(7HKZ2eMjQI)L~9X_>R(V3#y6bbTBu{jbu;%IXuaU@L=#yc(c
zj`6Ov6>vtYNzUr1hUz5R7qyLt&#2bqxCprvQu`vJrHVN+-cdA#<mytg73Ty+Ft>Tt
zC20#!#U+hRX7nrVS1cygVk;6;ZnS^(3MNYJS7FCo0_G5LtZf~VO$-|ks^eNr7AX+7
z%_|U`#74=%Nov1lc4F#BOVE-gX-2EIU)6~wJ;I;~EwxcH#w*w-5gVoC_5~hmFzKyJ
zXuDv;M>gFn9;wZt5G_gTN4E8*am4^if_2!IRcT*{)8(FPcFJuRRGsGbEfjyIrD?5@
z)7sVso739NtgtH~K-^j~THIzUXce2Isw)&QuLd*Jrg7`X%}TKGSZNy%Uo>%Zsjie}
zt`G=|H2qaGQ?)NDxteUe3=^o#(dJ-R#g=7CtlIQatE<%<L<2^tSSnVZTnr&*dl{{X
zt%9Z*O+46JvL<Haf|1Z{o>hP0S%D}WWiyxvi-W*W7n+Ps#AnvGEkn%<!&1dT(QFG*
z$QwgVO$+!eMzyyQZWG&D6h0jz+fp@BbHse~?YF~|SFjv~yFVZP{*QlR4D|&wrYasI
zT-{z?o<4f?fsVwAj(MM+DS0lNJ>BsA>~c1LaeX$Kx?~@?;n~?@F`a*3Ia)sc`R(F|
z7mKsS)WzkCr?aaC=6IS+O|T1_<tda5SE|bgSWzTS%LnW)u%y_-qesKD`E2rHx_C7_
zd-8O6IeoiWN&ND^*Hg^=^ArMIEv8qC8v~!^;LoQwv)lP(db7kWd-c=w^~KqDv$wB;
z!DKXprBi%A$CnOnx?F$z<EyLLa-h7N{^(W1Nn1w4$z_8=Z?qMj-pSLxcf)tH`RnPt
zO%+auzYISP-{S#mJcVeJ#j8l<%`!P-Bd}QV6#&Z0o<cl58eZI<FKoDzi>v<(kM9Pb
zEG}kO!;9f>&wsT4zxidcxW4(*F#LS6_~rK8PiAk1UvJ-BAHRQ|{x<ye_NQn5*URDj
za&|uC>7?YS=bX&A&xzvMWHJR@zJ5Ik=Rj9U5+^YXH}lCbyIx$px%l67{;vz@b^dQ`
zGzR*^iRN&80yqC7w6d`UZOQLQynX#~`Owx+iF^WRB*|V3!l4HS1g4yQET_X`htXdQ
zKg?dvh9~Zun4y27KmxR3z@=0mAfz}@h!hDnZqcCl3>RnEIxr$}@o<nWLU7>%KnNlT
zO%+O{Ekf+S`K9=Xpxk%AvWNLT)XgtJk53A3Y`!4|r8~p`Mk~DG3joCi9J~h<8xZoT
zK(QKP6DT--4p4Ai0|j-UpbiwN14W9xLA{ZBgL^}^fXRQSdKP6L5Gq;K+ajq{c$h^w
zc$lR7Rt>HpU_FS>1cH@}x*hIXa9EhQ5hL{$-$=<5f;9{9Zkun#93sU6N`Uwc#sTj-
zTnbXbtan1B8oZh+b_gLQC2+}<GzGx00(1`Y?-b$SaEtEXiQwUy9;%1Kt&UGX%}0Ru
zD8&abJn(-R7$clii?0=b3;W7jkhNfsKn!azoCjH@Jc13Dh)LH%5M%qJ>=XhC1q>j7
zal-}R7Ih3EC<k8==(-TFZ43B2L{PQh0y_o&XZTPmdXNR$J`|&=D+`QSe5(4jEpRfs
zx>?N6V174C%L`OtdLiX^qYKjpAI*r>`MnTxzt?|~wIb*7mJ06CQ=?Zf#E<ucWq5e-
z9z6kcKFB^jVd-ETb)TLIaw2>By@c0%KfLD83a@!1yr$d(rS?V-xTyzR?g3YNoRuDD
zrN>#3-sl5LA5i*$(g)1p@Sx0`8vdEJMGMXq5ea{hL1X{pmT(x<Y7b0cyHqUgl@S1O
zh01@>h_|8&Ax5_>I(Y=V`$BNap+@BJ;06fEa8Diq5D^5$^6!-phzycGAjB8Z1VQg@
z)};_h*mH`zg~*^sKpy<3plyg%Jc1??Xb3!QB?B|`5<egV5TGDuR|$a2geN*IHV8Jt
zN^(6m2oQ$DSC0+9oy{-K&M&9S$N=|0KFNQjjL1o>NRq4ynL?un3|vP%zNc_Sx)~uE
zGn~LmbBD4)t(bS6?!lgYN&K+%4T6x4?cX|r#*KHwdzZ<5tqzv!L$<oE!o=$q1{SOu
zm`cp5QZtxK9S`-0!MuR#9rj47&v6^}T6F<D4Y^PNNDG{8JlMOFVugD<rT8Mc=(vB^
zrRc!7TNhNB3usS9E(M4RIvVbE89MNNU_X$QAS*)($0Y-M4=lR?#ywdc&`B{N9gGhQ
z_zKhU2|j+85NFz5Zh5~$XO0rd!=(4<X{^NZ<B#{WGF_&<dP2!aWcKI@Lh9`R>Ek^s
zgEXZtL4|j*--AlhCETOJT0Vyg7g>K(VeK?mdtFI8-__1{wewx=d{;Z)T@HOZ3a|?H
zgVmDSgj7oG*a?Z!7I<f%m@2x&LPqRE8llpJxQyVE@1(GVo9r}cmlR{?*a9S{vW62u
zZm%Uju(+<#;<{4U7S})Fyn?NuP}V0FVg-fzwHJFmzqnc~&nO7Esf`*R9a4X}tB6S(
zg*&9_F+Q}iNy1V$OD3PQQeyJDv7r`~>FaOkyVlD^k572}!P6(i{cM%-jbfPMefjEf
z|LXF(EfT$s@sAv~$buw8H_nuhV<<nAB_gkd#fJB)d|LOZ^1ah*#cZ;GNa{X4C4z<Q
z(~}d7ME2=O$qRjP5^@0VsO5hi@2L4Rc!yokxZ@qAGl<eFW~EomO0SsJ&P{6PCbe^u
z+PO*X+@#8gOvFypu^bBI?Nkjb!5-bOr3e+GP#IymLenMaC`3fOo$1{Pf!wN?P`iZ4
z$Y{`sk^I;lpi#&WQY)+XcNtOyPmjA55zT$U3X$D0$Pk|5VPpvI=qrD(G1ihh<_DHr
z#gvT<Vc*9nawJ*aTtI@a;0#=X4<Wjg*yph2yfu)WEJZF{D@ZC#k;f}?Bm>ZB2W`c(
ztPEhwNhf$5lImJoH51gCM>uUop1z!OB?Cb-wH<kFmEp@Mf>ObxYem2gTK>61h?;^;
z4l$H-)uL&P>M}MX=CywamKOmWcZgsY%>wikV+@rlY={pmZ;&S`8|95+b(RO_f&dyj
z@#F9=7pVG?0!P(-5Bs+D+<w@%r6-fXpUOTxWAKIc>51+2yw8F_+9Lm6PWtRb{XLq2
zT=R2iMqHot+0DsD5A5FLs-eA6dSk^m?7rlx5qpEiBNZ8LY*&Ar3A>Z*Vk_JjS<|+P
zZM0jYNeog!KQxblB`~j)*Ft3WNp=e%c7sJ#Kt=#&(046k++o1V<xYl_0#ZiVtq3Iq
zz5-Asq+5tYBH=JnK@nNZIvq!-uq5Y!AsSIWwmURqo>h_EShAb&(L#ycK?q)|Y$3E6
zfe*P$gxJ8DtU6ML=~Yb1cS58Xy<Mi=C4{XV8~`X*GNP$~Wy=_sH0@esTuV`m%EsMN
zXf2?zfaMxl5L&n2XZwVVnmy%pdC$bi53Ra~pX};ZUPaUY15qRF1(zY(Efkm0R{|54
z=qx-bm+R^+ngZYsmxt>uAb-?LemZ$rmB_3t4ZgpMr0A>eeFwNw%fbpfP#&n!@_dRb
z0WRtb2R8d&Cr4ePhh5f*{g|wBS|b%~T&8$?Z+^Xf?Ec$TrDzn}&{^=xi1+{e*u@6~
zVuK4Eq6%Wt&wsmbyFXj0IS%bTB~e;PZN-;~i)a<I?<|tCa=$7xvVXdgV3Hrn$@rUm
ze4ZU85nccM{N6PqU*88`y;Xp%a(Io`B(v&-AaCN#+?bzKVi50~O|^Y4OwE*m)8UKN
zG4oTza-@>8oZkwYok4F@aInD}45(=C%*Z6p36;#E92e-vOf^-0zil?xFe-6*X6|Xx
zTtNoOm6?MPXo%0)nty+v3xGV*`gpD|DwHggWUdbOUQsZ`1_3vseGb0M1<8j5cMR)L
zBI_VI*a$3+L_y>)poEx&)JjPC7uHq#=rzw~UYP)A@^#5Zq>vD-^^y+=`yq`=?Qli7
zq#MgGqS8aoAZdPbo%QwY<HZ0OE64}4Ra$_br4=wfDCF4+RDX1qKpQVvXxMrycN;YJ
z1$6ew*Z1Y9mK1d|*H&}OvJ`l-HL)sOZ*37w()s1yv<R>&7`xsgu4oBNP6guM(-Iox
zSG0s?#_cVkI9F$6d`r&yPqzu9Ss|GX9vU#pF>dnkf)C@~@i`n8^65O>5r?M&Z^7ZI
zge^GS=w4mKb$`4?Q0V+j;sfyZf|VO^7K0GZ`6_J0SvE=|v{D|3vx4TzlYb=8k|h-c
z%O3-@B-;!$=Jx<?qx=e>ZDw5J*V}nCdg*Gk!7?;XN%hYxgAuns4eTRXnJeZp**?I2
z&bL2>nPga!A>0o$8-X`qW+Pz(X0|i!x;|lFa?LD$1b;4GjQ-;oD;cSj*{W_uQtK4j
zR>&iP6r_yI7RGw%z{dnBNSss1D3Nl4mf&I5Kuze5!+2e=^+O@!P<aej9SYrKNrHq6
z(L*2!fNzZXMQ2AwqeLCse#NX@fV-a^8QF$-viFN}>~YEQs!0lpokZ4X<^6DgNEt4+
zvv6RmjDHEO#I^P(9Ke%nXRNVstoHRsupB$L$<X%JHy?fK*?Kw=Ii2?!f(keg9L$OG
zQZ#P&zW6egyym?$3c-_h#{ImXdR{V1d&$mP$<lR&R$yF%=du4i(apQJ^OA4t)EJ9b
z^PM`i_cMu`T6OFL*h;vnRU6rinAyx}^(+FJ)_=-DIHxzZXTg^Jxmu4;?9jN^>_f7O
zc;?wco-f*&GIYVqLnRSeQnA|&<Rj@b{Z+}rb9s|r>)he(<Uj2PJd5<B_nX{+gcEx6
zT<J$oW!zb9)$clj{hTVhw<X{{f5=JFVvifsmWZW%tg-L|GN)5xUm}&(B2+EYTFkXz
z1%LNwLNdt@glX{Mp&SJ%e*bWH<R~btM_{7nOJ)m+teM=g!djcYijzQ)P0jjus#_H=
zz(rsws;S;z#6G}+#Pt__xB_9x7TxXP%Ny#KPj2eKfFgnCabrB@{Ylb2>|M$)KPuQ#
z@TP!}>%YF%v;LCcVfeT*obIGu1^3>k;(tzFu7fTIg+c}!yZL1uWQ90r@a?R)-_QHB
z5!8)aq}z|zU<_})2h)nEd#!O>IX1gJKdes9)LBd*?rw*Tc7Z-WjzxmV3?qg7C}i(3
z(k8il>?#SgW}|C>Wl_i*v^LZ|;PL7Tzm$OpF_N~`^gs|6XyLF0gWJ2n#{Z!`s+Yl}
z1rq``H<xj`I~A9f`adX_@bE2}0r8iL@hvicCwuVa=*{lP%*xV)Q5jhbB#ex?)tfQG
zxNvBD(ZB?yeeclIF@kGcbS;dGhy7t?;S%gB*ob_*y>-7{-yi;YXea<GwqdZwD-@vq
z`uPxFkP#nT7ywi-H2nH|`2O&FZ?!CKe@}6lFxOV*vcMdyV)latQdaJty4uqDxlEpa
zWVrU{*N>rl^6jn7A0!GeK<|<H8m~nJYb7vgrK1htY<2PETZ{75yg}d${TR58)hS!H
z@dJ$OvBHmEvo(Q9>zDD->J)O4ne?2IVNPSn8(6g83YO->32rG7!`>?j*z)G7tWPK-
zni{r5^;cw=m{;kE%icw9i3r5skyGM-+NO8a>yX>W7>F~-R$>$o=g5gIbA+_Fm$Lx4
zpb%U?8+7}r>V>VC`d<22=gd4pT!S<VS0b6m{H`HWF_`AHWS{ypS)u%<VH`0zR;;tF
zlKiWca$4?fYPLY;=L-}~vCY8Nz?Y!*9%9PK<&iN5{p#trx6>qRL9sKP3n|ZkA<yLo
zW<M9HK$0y8-QbNOWkmEc2W0DQN@fjthMRBpW?6w|P!^`NEO>GaWxL_}epzH8OlPLD
zc$_6qwE8>)5b`JqT;H{AU|v)bSxYg1{gQZ-C%Gh=Gv$zE0et7ooys9OBkX&xa!5{!
z5c<zs<yZ=1n~ZbpT1?UL`U)w3wT5))o!qF9eU<8ZMAaENo69<-l%Q$dQ1`y3YWDF8
zMP!t56Q@6o#D{5I#{Fa<Nj<l+!g>?YlfF1q#;>fTrHXYRMKDqPkJgo|yw60^e$}c$
zrRJrc$)al(?X2$R_q?!(8#St^)tFmG*EZGHDI6({0=2T+T@;bL2qap6p*RceRJmX)
z?01pHAeQ+OVAF(6$qkobFa?tIEDrwK`W#DPdL5AubI_tZsa)7vQ7QWt4mCR$OVWIa
z8>uvs%@&N_EY<`;f${Tjo0s@3i^gCMC(Av_rVnHvF-aL?!y!iy7}Kj0GsSXry|kDq
z6ELtD_j%%gxUkgPuPl&%oqE!-wX3=@wcpoj)+;Ysy_?|Y(Pf?CSizZ5(8)0Lppd9{
zYJ#T5#0&&#AnZ*yT4XS*=~mZmp@}-BmpqNj;mFgx2$4A)XCbM9h(WaVIy0N}O_Y`e
zZu6}Krk(54wM;&Z!D53D<{N^gg$){?`%?T!KIx{DD-qg~`);X!q8An}5hb6mKxGRr
z+gbOC3KW?Y`K1CaI3(PtScc`cE#Ye~Qk#e(jly}gdF`ktf-`940J~A|Br8EO!sESU
zC2Fe#<c}>~ArmS)@Nl(mG=b7)sPke^H)_#J==vV%WWi5<S4|QZ2AY5QBs1xWqFdrA
zM?AHRRGd#lFtjs&l8>uOut5cyR3(mW$KNDCm=i<_$wwrdpM~UKFWp%ORbnhMf@p5h
z*QZtCB?9vXpX>R^OLrciM@Wh{%vQUImZS%dg}|s*>=g7kJ6SlR?m#LsagRxhoU+Hy
zYu>L$D@?N%afQxQ7ZTbq&h=xq@<P}XenN3O`dfRP5K1b4x*?y6$SIqYvClT!%1B|g
zWqv0_9<WovPa+Se_{peGMdl7gLy}08G+$9Pt|vaxFvem}e8#GPcAxkzAUZHki3GcW
z=i2ESCH;3j*GV3D>>Sypx_s=s?6AjlxbX09Y-dDJfa9y#j(N^@>@HzDwp(n+?ql2p
zY!rd*XOOsmsd492Ju1#C4&COiT}Hi%&oGE&Gp6@*+C18xd(xAWe3tR*`K)1^_-w;w
z>ea>Ae3ddjZFFDa;55g{rc)vOPbkhvFrfcTU2H&fv0IC)sut_HLMTPfPV-zHfXxqB
zwmutPET!XJ`^%*Cha4>jx4o|$j-EP_k9_AlNALB2jZ?%o+)I2rE8C0zAUsBM*h2l(
zhW+~EUM6Ki=l)(Lvmja6ci63Dd*{5DJ1To^B?a!hr?-5##kiV=yFoC_zdE)GcF7l9
zWKgn-ok4yr)4#z}whBTI2d?VoWnMZb8RP#0FQv$N*?n(#dEJ-rc>3$9v!jnUJ4mtz
z)q!socTlog8PhLx%+uYk?)_be>|YBtMzWV7+AS0THkToq7ZaE7_btGG!f-$Q_4NCV
z%*<BK6bfzpVk`-!RXAe~i1<#-jph|}xa&jYNi#2KRd`F`8-{O>d@%>wgz@F6W<JS>
zUx8MogL!=TZq{QlkEQ!Crmo=An81Y%BV?54)1!ke#Fh5KS(@dN0d1xv9}gdm40r|%
z<uTLR)}Hbpz?hkqzAHR`My^yu#GU1el`0g>gp@zKJSxq`W9L~3?g~#;g?dOZGs<jK
z1AKu|HW4xz;A4gqdf4?btEfRwd7eEh9HA8!R*<tXo|BG93%n>?&tw!V?0{os+Z0PZ
zFG#dbmjIyq$ku8f7$u}rRma5+DN;WtJ`m4@nLwBrjXhi{W3Li_o;vG6>pbq5pWE~<
z7%RU@Tf-h8@5ex@*>AIE^-OEFCi<dTQnp|?x#P1o;?9ev)}Ci!;yjrUPw1CXqi=}5
zEb*t^CXPiDL;}dyO%muAO~~GhB)=xS39c>NzD&P9Ki&MhQnZ#@7{34g<0iZy;Gfp>
zYa_HW%g=w7Z#RE`%w)6kltS{H^r7YQ^ZPPA`I@7Fmn@+)M(a^RUW0`<3WR1Q6;sI@
zx8M~dRWNw%Y<p{p+6Y8HQ*6dlEGJ_{fzR?0*&m*K?c#%jWGy3QCAXuAm%N#29*rwL
zl5tCx5<X|1E<<_=Pwl-hV0PdsS?5e1aDlK;Ty=Qz^)lvv)2b@izt486+sdRzdtD(%
z!l;!DqYK#D|N4HnP8B<9u(;wJR(jn<mAvq1W~zXj@+!@{dj(!|WJ))Gcx^MWRv7Q!
z+f2q)gPA-p7Hq>Okr%f!)idVuVj~$>0KLr*fgrWw=OxBj%xGHo@pZ<@!7fmnsnWJI
z#t5%hvR~VO4z(86!4=*AOmPalA^%8e_MTiM<sE9yskTy-&ftz2CPuM+M3{yreWWR{
zaNKvQ<|V+Z1Rqc+mpp@9Ww2!OK94G*s3ea5oT{W%?NOe0GIe#q^d<-!E_j%XaEF8_
zg#)!{M2@LA)FhPE;#;N@GbugrBugfmveJ+RmwZcqb;)C@NHRsr`K}9@7ms^v+S9py
ztQjsOR6<S}P>Nd3>v8=USOA+~C<=ZJ3`z+vsfPj22F_IIMiu+3!*ZA3xsFyIKSC5?
zIE~VEAp#Jd^#21Qd}!~3%-CiAD_OVF=1Unr*};7wj{Ma9EW47_<ob?cr7=0Qh)*vS
z&4q7&tZU9yOO-dxFUi&LeVp*>DF?pp4j&01_cgSyU4rto>u(UKjSfni9l9dv*QFvp
zQ)=aUq0-gIZT-=g*ENpiSLyI<&j>!b@KOWJnCJEg;CU^?{+86aCw5wgm8NTNN4yiq
z>M~ke)qiM4IA1+86>dw<sqDGLiw~;=&Pic^uD#>VbG1HhT29eWX~z{eCntP4EM8#s
zbf~|q&}?@TXNU*Xw_E!oSL|DRB*{=vz*5t@vAb5DxGeL1XX%|zru{RgPzc0(+01+w
zOWbThrS(c5a7M83l_NXlUe1^cY+2W78B6BUdLOmNVOuApq(jlyo7%r%pY0?%)){Jl
z!;*4I;HAobed|pt5>9&=WrNG?z>4e@C7V_Gm~+gN4QG-{NZludk>FitIV#zCUFdS7
zuMkIv!syl$7#fc$X6#<^gco*Nct5Xr_gH&U${#e|p4;!8-gb71kT|>cG<S+Z_5yUF
zQ8e9SQ|LqvpVME|D%6&{HI{R6|KbjR&$uu=tWw>%$umw+cWob5PLD|wyK<J}mX@ly
zpO4ZA45Bp8KE-Grar7}O)a#qI-4<U`{npeYy&!vvA5=P=%$>Py=LyYU*&LfY-&pgn
z-fKlZT<_a!m)7>a-DKas(V3l5Ro)40TkIL)ob8(2$zkn|btnCd_Cj*(nATT+8+$@+
z?1yGmlgm8O7gy}J$;_+O*2(MTQA%^3@8%V1X<_o--sdwf%H@sVDEx+Up#(v=Lat?<
zZz303C6Mqf<+4^Fu8`~d992H|CNh#yf|JXcgJ+(gutO1sE+3$@ByMkL1Rok@tI%jc
zScG>uLt%Xi-u+i>qD>U9c!n~6QND17GL3$xGZckJaDMmB&GBi~SDv9v)ZX+AW#;@k
zXDAcZ^be5JXDPY4J<EM%%ixL*oWJHyn<(zhb?;x*XAVaw3NJu>%6DDA^aB%-x?=4I
z1cIR+;Kkg^HyZ_gDVKcv9U4BQT4yvi4N4pCc|{^{on+0$8u7)9)xSl5Z1Q#@B|=s`
z`>27pcd8~Wah3^l#0)0_>`itRlRJt^&jH!KfhD9~xK2a1{p<@40Mk+8$n%+cv^dRi
z!r6LsBN~*`pm@gfxkQH2O^T!U%fenxC|2*YWkIsSx><O%`&VWk%bEWS%88VN*fy<R
zTsU=Tu}pV$iQAv`7<Pt#T)mA%0k%0XBplgkaIq|%GwZ6ak{cA=W$O$23sJ8|g&kI>
zoG_d$Pu!ldx|{%v6L80N!cC_dEb&kdAdM{lhZz5Cn2PV5yi~)@K99YN7T#Lk_meBI
zqkzBu_94%p?HO!~28J2Ge0WB5obK!f{{!IEv!n`TZe(+Ga%EwcD%vdz0XLViO)3)s
zGnX+ZDJPc?t1JbTY%Z98qOD#&|MKzklVoA5WPrlhaIsDcqE#g$Dw-%7TCGVUL;N*H
z{#wUh5P#ixj>>a1$!ScvYcTzTxW<sY`X&D7Su)A;)6J_jYXL}sUXL=-;!MdgzZ+9o
zw>fAhw3>(%#UhdGic(}!<CEyhJ!citxX+3jSJp&Pl9G-o%B&oJiB)J7buG`JtBiM(
zV~TMSWjAn5=9XEq<Te4ud$>rU>EqGh<xg_tJo@#^rdhQFfogJf45F%?ZLCw=&;QM9
z4zx5ES$~r((gxO+JG8e#y2%~DW_n|v+RUsmQ*_uhixKaV(;~Ymek}+|rf6a%nI5xJ
z>TNyx(FpG-($EQiN7uaiD$V+AZG`r2n4L1V=1DZjD`lTdIS@)#u>6{}jhW-?tR@h2
zrPGI9(tKGF-2vra3|^G@+=LNU9=mzwQ3PW~edLvM3^JAIqv#JSla%ES`$WD*>*fzm
z{OR|59+$U#wQ@IWF;Lb@JIQpb+ImirYr{u3NK~#bb;Xr`gSCz%B=I$#8mlI)e6{CB
zbS}u)$6|>Ops1omW7*THNXaX_YUcXdLkif+FVugHJb9#0iZEeEaI#HWD)v)j3$Tww
z<iaP-QgK9=k!u;vXx_{>86KjSz{UB-_7)yA(nV{%s%L?gB;iY=2gB>awTIWbzJ*rV
z$44`36Ts7de!GhoGSJmOiWlANb-eJh^)Ewl%4ajk@5<~$xZy&J8(tRN$dnG^Mu>Wi
zYW<_Ip`{mr3ae43TUGW5Y$RiKu#wF)!bUY-g$;I06S~(tL&;UBFa&o2>bBLTSafuj
zxd4V^E~zvuV8Q=dDbh#_7VaQeWWEoFW93&sqZwI$0S&&}fClR@fX1MiF5KL@zAJFR
z6z;$QQ@jrcW6@z83agevz%UiAEm51*<YKPcKG&j|<*RGj)V>wn`;t7_&A@Y0R#%(k
zLOgTm?5A43$pd9`y0Mg8T*rpPNYC|Q0*xe@UTF*p#me#iwprFZur)R`!!9goZ3Rd_
zdL_4iPq7KwM-YolePeJY(bjI9Ol)W3Ol%ty+qP}vjcq%b*tTs<Y`?K4c5?Hb^W9sw
zs;hQ&^`BkU)ob@!kAx*zYOWfXv;|wg)3LLDi+Kn*hPO$=J0)QrPW9-<a7bSkmWPP8
z?z^h<6bo4#i@|a|E#I9IGq_1Ko)05Lm(_PhA~kb(8R9q%U25Wh{)X{uSVD#DB1Iut
zU^uyG)6dkBub%v!@vxRMt}G%=2~eka@l&AqPT=yg{x$J-yXY}biN7j`rv7hNj@QyK
zIg>J(c`CTRZj!X^B9@@XN!#8>8F+}nQ`Y5DLN^*{R@_xFXNt86Q5w2=^w6^Q1oueq
zZW`2BT(>oUUqe<MT7I%qut;*Rbd=KBl~sD-!1r`pdv%a#tLf-RV5*>au8LIdW#WK$
zj0@1_DH7hRUpBt+uegfmZ$8XpEi%gWNtaVybmG7mHH-)+Rh8`glJ~YLKr@Yiwmcbm
zC0#4y*`s{4WuAQQo|;!7fnJwgPwc(ucUosEf?<C|yty@fe)_z*A2m@crUjL1dp~|E
z^$?*Ut%3_ykke2|6ZrCec{tT4)x44`iU%}N0B|zm5;_uUeUfp=EGetmN>={DicDCq
z4;z5?w^A6k)du&3X<--X!|gENHlJq0pN1Kl+^!G(TRv45?$Wff4+#wT>A--B#B*TZ
zbP@9g)nQFg2*J&1kSjYLMdcNV7xqW#gwo_is==MKI-y%CY#!V+{(RadpD$W#$M9R|
z=LC#2TWE=!ggyy&;&j@ym=>b>(=U5leMxxp+F~GcCb~0w#|QBzhHK%8Z#eKN)cb6m
zaC<$I=B>Z#<@kYDw;0WgYsMD_s*3a$mZ4%mkmL3N)cals^!lewwRjUnQ?|=$)A|iO
z_`d-yJ|_jIy0L`p{4+IKW}1&nbZNjgJi2|B1c(?M8&|(hT-In{9Wig;E3&G2a0{cN
zu0SC;R<Lr?E1M|1UTW=7E1u<EvN9OBZW?KsQ5tyXb$1adc1hi9mUE3c?DN0V=WMs!
zbpw2uD-(hzG0L=8`Z%z>W8b*OvU7K(4+(QcKQ5)Gt}TOni>nI<r7t=j&<a32_Z-^8
z1yyE_d0bdUuF4O@A&#irF45?Tt+wDWGZw$yoAOXuwnaI_vL~-de)kh`mtU2z4tue;
zVU$+l^(SJHHzuSH?oM~Z;7wW{XYrvazgCQpXIGj9n5e}gV|hBKOMgDtRfNacC*Gu>
z*oyaPd$S;yCG8GgJmizg_?iP=fE=Wh3jC0SB@0ttElca*6D*;12D1?waLX|5**e*K
zm$^pneyVcRc3r<xz{U)>H5b(93-K&%B`zs(vuLPQy;EVpR<k=yHu#)=W<3oZ@}?5P
zhaNpd?CGSsEAj{Jmx;#0+EArBYyWhNK`-$H9dk<rj%Ib|C9x?LWi12vO-)8;)3=Eo
zle#yxe%3f{Jxr+44lQ617FHfl%r0vz?4kTifM2Pf9hNL_9fXxJEKcYH*{G{PdSTxx
znQ_;)3=;F)WEafcWw%p?DnUtV`MHvoUHUVM{S!?K6oeeadgl|6vT(>1f}dyHmy8z{
z7`xKM6QWdljZO^_a5e?#FeuW{o9w#9bIW?@XFbO|hmdO!imm1jm|YHC^L7Q>oOnZ?
z9M|;Szf60A`Ril1BPA(%$%f^&qi9BX{bQkoxsZmkLNmoSGVX2?-n9Df>lCwb{&Ua-
zJT-jCqx%FrUwgas`^*~4VxOhL9rz=0QFR}YoSTF9sn@z+@J$<lfZ_^Hr6Mf<);kKJ
z8p4ZZZTGD*=3WB()#^a!D@7cc^p`uK#BA9w!K8RY;HRL_A4dLWp2B<ukb#Muz<mIt
zcb2fB(78Pwx*BXe)Oo3};m6Cb3=lp9<wY;NIN>0Omap}k9QCuwJB^*eR$Cd{v%62Z
zroGqi`hbOTPm!9&+|G~s&B5hNtH-BCOUwR2FDS}0xV2X?NUj<=`GzHKNfsBz@`*>&
zNu@NlZb26L`W#V&z2$rE^3=1AuQtmglB3=qIC}N38T>Q<1&bAld&3jt5Z)y$jTn}m
z7hu@<`Sx#A3d=OlT~U92(Nl@G(`o|G!ibMnnq)i2!#vAtIn~H=!;klg*G>~}a$Pgc
z-ESMRt*x+mV$OdWOE07DFCRNdOjCL*@>x0^%ksW!T0BJFeT22Oo>xxdR9EZulO%T$
zQ;X7whG?3B+*arF%7)>?&Y6b`>qN&Y7XW60k^9R>zZeOkDkzSiAX0EE4$xU!DNwv<
zB)(>jYL?%8&=~D6R$7ISBcc@hMp1G&oW)v-pY2h3FdJu|gT2;U3)L3`Mv?v(s;7v(
zph68kL5+J{a6_6P<|57|HydXm<ogBndyL2L7fcmK42kW_3pZ?6=Vgzi1)qMT16K3H
zFA&o4)j3NG*plwL5#1)b32f58K|*UMcDIEM{^_|5E-=Cb&Vx<ZC|iTeLas9f&J@q)
z13&LI1O){P1u3d2?oJMUb)&&>)r>5STDulIBg-;7*ImwZgdA6FS9T1tZk@biwNC|I
zZS%vq=+BN75kj^BpN53@Kwf7efPoQItHEkh22;}9$D1Y@!j7~54~j`nrb%uz$rl(j
z?Kx5H1hL^i79D5kThXz)#;~rM()xB*!^mf}+tvFz^7mf9S(DcWxsWS=ue9Qf@)<U*
zY^sYYH%-ei!BeAKVR0`?ulI%TDrh1Jhlu;S>NQSNQuv=Q6?s7lCo*gj0=!eC)3yto
zI5~5LbE}H@*NADOqLT}L!!R>D?JA(xgpBYD0tvNwuPsdU2M&Es%JvfD;w`mLwQduA
zDr^BtU34b?t`q5$<Yw84hLN55TU!C(U_g8+2{ArzUoLW$k)&)}3a|Tq-8-ti>Jt}&
z`AUzoe3vHo-SNHcMvrB5z`U3*m;-l1FX-J(O?tL4EpPMSjLE1^4v~1qrLdOp_{^WQ
zBv8uOGNxGc)oJgY_%x;ck00=yN&#KTF0o<Tl!G&}GkTd_Hdpd(t8sOhN<p8)DciDl
zl31b`)4#8Y9~$9UsUkKZ+A}sB!c+)TpaIt}Bn|F4__ar&RfqqQ0Ih^E+4Fytmb60{
zyExe~C_mJ`z)-_%9E4~WDi0qxQ&xaWz*MwH6)cZ<vXmLp47c{K+en=8g|r<8W9{}E
zo-?K^6l#`ajAxxe@|=dhE#}0`{Hw`f!X$K~V&bAR+pY}9EOh*dKX|9Ec-!9jS2SGG
z-<+s!=rZzdN?WUY0IgP>4oS$q?pBuJ>1_*(^M{iTOpRTC#*AnQ*;e(7#z>l8Bt<dn
zr3Xo(;x~)NoCwgq`D`!rK&sFQwo-ml8!Zba!3h#x4GR<{JKCmj)m}I{#(onnh8cA2
zf`)8fNpD@67S2Pi3t6>hx|U}p8^<^-ZN`RU_&w!HI^i&VK<#EAf7t~oK&~B<(W$^r
zpm8%W`_RHFojwMVowg+!Qop=8`WHZ<91?*_@O@V)YW1OCDJuSvsu&YOWi>Tfc>nhM
zmdwH<iqDiwOp0~5KBtJ`uEv+&bAJW)7n0H$otsYfx}*!C2cENmgQHQcxZdbbjrm~Z
zSKFW8ta21UW10aP=cS0d=}sq^Q{8>GuY`>Sf6N>{!vMrMwBBPGjwH8dx7B5NgSkh)
zpF;DGRVB0MxOgajVTpHOWYXUf#c4??->x$<GQDyh$;##`-T=4>2PHOT&rI{Pu$2;j
ze1Wrka`TX(!)7YR_uc4Kpc_DCnmU%@0zs5!&RGfSQ$X9N89gn4jTQ}*GtE&!1TD=U
ziGBohr&S!8{t*HK*KWGuKTp9png1(>^TTJQOiRG{?hJ>h`*Xkiz78bSjq_?1?KRJM
zsZ+PKCtz9Y)l6)$8;Xvt8M9EslLrlJ^E8Fmv*^9q%tX&3p?{!XOGx;(m*n8)0Y_QZ
zl;?zK@zyw!fqId#tRs3;{0ejA97Us9&*zhG3jldpDh*M3w;2eG`nTgzWnA$=isw~O
zZIimU7l8fK<*7VXQ->zMbF@jX<*0VX1#&hCY!0}R!DhqpOWeu}w%&P9yl6AP&+mF@
zY^BxEnx48)$ciOz!5o#!Hsomnhg7}DV{rTURC4%u{L9Vk7m`luOup#h#OX><KC-I6
z9$V>;{ql;3hm+H6v@jV$Y%Ul*ZGTJ1??gDBV1Ph>B3*1|jGwPIlH)OkCF!*0YITnI
zb9Hwhs3>K~`O08b7<w;mNMJAxj!2n-!U$IHtWlFsw4Rm^-;5FPoz(LkhfPFOkg{jS
z*!9=$qb4B&VM&hm&L|SFjN6JKbx#qTc1W&eE19qO+F{GIx4FoM$nwk7MqosCti>dG
z86Xs{A(WxYlg@-cQD+rDIc^F5Hkt?*n@7+5YI05Y*^uJPhwY)jjfj4Drujew#8k4&
zx`)YWur5+Z#r>lsU5p+Hevz;v8V!-V21GRK61ju*5mTbX03f>;$juUsC%KP9%ZD}o
z5(_<YeeAj%gKCy#F2wUZV<bcP6m#1g1@xDJc~N;L1=&iWwUhjTRFVbNAN*EN=m7W5
zZ|V7{BL7?_!g?ll0^E}msiY#{I;E**C!1#!w2I`!dRi89^1QDH%P}{^ouUjtA!rnb
z{MZ<s2hz$b@x$zz2pX(L@iskS7!kR+j+I6~!Z`rLR7=7Q6Ct6j5jswh=SJUF%degI
zEEtr=W$ID+ynS|~Fd#Cmnef{hcNmkIY_`jk*B9xMcONHCfT71VwiJsgGd^0goheZc
z^eSsE7#%T(05$U4z_qELm#ofoq;G<+aVyIay(kCIb{)cOd2PRQFqe4~uSy_QXY+2K
z3@)s<+<TCMiHC4xUokZ9nk0uq0Ia9x2-b#_Hm|OZS-qjI&N-~yNpoG&?P61IotjcF
zeJWA&SzzMo)@9)nGt000OZ)ZZE2BTPpPoIClHE-0uY#~>(99GQNdFv}-nyBg#V4NA
zsYa*VC337ba>X!&(S@(*47F9!uAmLnWfsk<&qP)enJeA?od|;9@a;~Y0!;Pq+i{S3
zS+X<Z50aUlz}t#*!VaQ4Fz7gh5DEVc)evXqn>|X?-)PmPoj22$5|pO$T#}YuW;M8{
z+(;C$$+8NQMY%V1&AF5^pU`xteyZVTUeJmaE30ueE=+xpLH^QWEK()t^}7_V@#P8t
zCYTl>vgy|>ebjACdtaFs0f>#8{$|ssmu(p_!c=915L}0tBdszA>$EgeC_g?w!&2gr
ztj_*8P@uX*(+0SMa~ox8qny^A^EGOoQAYU6D($bb1i(y;WQ@R)umD&lwJdzHxm(s^
zd+ykbMmO0(0JyMK{54cE`G@iB=pug(U?o<cK$#G>%aP3@edT|%0h0aYRB0C#Xh>7$
zZzk8~%fUweO(G@^=Z|rhe+7hw91Ck*fsHPw0khvlkF&*|nJY>Li~Nq)<P=<S+GUrV
zx7cnb#!g^j^x)NY;p$u+=P@)X|Bx|4SlWYzOczvC(-1U#)~d-gqv*Q7C`ILix|FRP
z$@4wcn`_Z{LfDvr0W6GRST^dx;5+-L-<>Rol_RkM>cl}~(Le><mO!E1LMKji`pdH4
z<0u=BAdB@=tQNzd#z!A?SAA}6&AuM+%XbLN6PG>^Cc}#&xz;sJ>YMdTm}gYTG`h>E
z!%Zmo(l}|%OH1IEH6zqv!qJU`{B3yQv9-q*sWS##!CI8O0A&89-YDYBE?jb;)$m?i
zEvM274DAmzKl6UWHO3&}J$sv2@R4`9qqZYXkT;nfn^<{#HXV~k))Ap3*rPuq&n3&v
zC2eM}t-|NfuETfd>UTo+$uaWw;hV<4TJJ|*K4)-)>|xZjoE#X^3zI6n;<au&o27k2
zbuVYo&&vvv0m<o$mYgMTNWl|B1oW10*sZ$R)gEOpUo<hogIH_W*WN#7zx@+H&@p!Y
z7N3kJ;;wQi3msTlY|i#EQmapwDdeWK^XL}^9w@gcz4@-7N?f~TqX;`)`rzEWuYcKV
z)CT43w`aB3>Zc)d$rV`xNJDT!ML3fONW9$w`sq`y2wNF=ys@XC#U6Yi+~VeK!LlfQ
z<9i{>8O=YC;B)ur{|hrRr!hr{k;8K^$e7xhyI2r0v$Ccokn>Rj6z%cE5&fQOHS!s{
z!FhS`5zWU$P>5{lyD{mJ$+&i}0`u!kn_Xw?eH!qKzCQV5n^JYD>W!$Hv!Ew8brHj|
ziC-Msw5e0*-HMeiH9mUX)@#}elLZ|v5}(mpVx^%H+9t<6QEQ60iE}bpO6O6?*TGeJ
zj4UtK6_+W9T~#&!5~26h@y6IvOGDR(vB|pl2f1p_u(vk+V7FCGTvgIcW>X2)X_PN8
zRJu~gk=_s$n(@kM%+21L#feeM?b2Ghm9zxPF4fEz`xlAifmX&L+?!JABB+$<Ins0l
z(k@gA_$4nI*keYv5I_uU%JI?E7kuo!^Qb>P9T%9YV~rPpM<Ye&q7!EgXi+@<3ubkI
zR@o$~FGO@GSdD&v;1b0eM#D)e)_Jf2E19^0M<zyxgw>szQ;194I8|(oB~^YVuM8Qg
zuHKRyY;$Rw`fB+^&SJ|R2HNoIQJ<-e8|9Gl!s0)o^6PecoK}K+(WxbIN<XPA5vF2x
zLS>uC>AWLAjnX||U{D{8jyapc@NwUOkpNzvKk7HBhON%YqXOkg>JC2$eW&3&ja>3l
zFMu#<)~tDf%&xY-)$?<a8$+LX7<fhxLLD>@%d<OCQN9(GucJ(UPWu{3^S(CrA+-k&
z)#_Z~HZl*~v&=mYLm3v4sIODoP{uYc0po$GM~(nw&xz1=Z213%A1Wz84J3-x&R0k_
z%3vXaFEComhbCSQ$IZz8&1pb7HC_V3D#FNo>HDDi1;h~MCY{fF`g(l==;$bL3Wvlu
zPA)90#Qm&3tk^|uD-Q`b=*>Z^A>idr&DQ;z2D&i&I{I<qx?g~Ie6H=;u;=EwTKc^_
zxi<pr^`{Wc4`Pova^+^s(hthbgUjwTa|aSAO>^~$+zc|O8FneiG~9J~y%q(s<ir?8
z8r44<f62e49k@4!j4%Lr2#t^OMiowM*!i<`J=82&9%o0fk|s|re3;}J{LaH}ZpQak
zlH6+xO2scQyY|^e<eBE7TTWlU!!?uIiLwCly-yW~;v9vCqmablickfJztQerr3NYL
z&wF!a;zk$PJ$<&m3cqT;ps+LoR}|b>y+bzfhe(?=1Pu&8l^Dkosr$)#x6k!;eopJ`
z_K&Xt-bU5;#T%2xHNOhJj_<DDuP61*8t<ZdA3kP(HR{vH4OrZkd_CMwzkJ?2+kOE)
z4v}UDYD+)nPJv|%2MHwmam)?jCyG)ooVYPbYd1HPKZyr|L&VVEqE5>QWoPKk76#F(
zm?3-|<24^|i~;?yx*M@^te6%ky!}WcW{vSq9rngMKjq{R<K#x+l%)g8+3YN-F@m${
zWgp(guQnDN7w*~<QntoAauWQHcP#-<jLfD8GEcS#d-4JwoPO^7wxLq7Z_jhCq1y`1
z-Ak;!>Dsz5YNW-A@De~uWg`iH+<`9}jjU0Eq280;oUZJ9`3{DNKXeA+vi>F`$)qAN
zoeUgKjid{8>*&Ue-U^inQ|qI%%}Z|CdmbSy{`+k4t}%un1qEY4BUuh=uowUkFP>k^
z0`YdtJe5!z!6?O6bqy%L4i)UiHgOvC250dY-|7uAT_I(h+u_osGy^S5AQg^Sk-@u;
zK=m4{`v3BXl_uj-zsv<-`D-P3EJ(SK7$u2Dmr-stVEnVg$R<e_pmWqgfB7;z6mBCd
zSZI-X<uo9^)Ww<50*gg{vONL3pf^i3CZ2r?QEP+Ykya_jG6cv9&mhsPI4ymE<4d(R
zqDPSG&?rmPWqDn9zVuEG9uyak(=@1@Qd9st#w|~vR-pfw5*&M3Xbw_2U<3PQInfk&
z?MRXm#+|mql|gL2otTZrfH9nCCaL%FHx*U;s_kfcOHD&GbC8q~bm0J0Xe0jyW<gDy
z{3;*;5jJx<h%j^fH}44T)B2RzNLf@3`NC*3@TOMCdL6dIN%&cZC&=um59u0S3O_)9
zXxvS4iicnOGEws10iVi`K!XEHjie5(DRcTc3FBz4q~S60TFv{oMG|71aCcs4x+#&|
z?aC0;{8Z&djU;z&NfUrKc^y{&TLm0Jo6*zV8`c1xtwhsy8{HIFVf01-Vl#wUrxFPk
zBH=oMky3JoD#>uLU;wjr2ndFKrw)(y61tP8M$8ZLzI>;iA3??2PHohc?R*Ecfu|h1
zCXKt7?;Bj>OyTSqIsPe`r`ul9AIkRMo`y6?SZfStx*%n#7$AU(7den3X%K3W+WOna
z^h8p~M??<l4EVc7<}>u`K;>{x*;%kXd=wS(KLrWsjf}P!8fDGtX@8B-lMj0i&hFnw
zqi)uQ$1i}F;}@;0;#eC3WxduqD(%HXKC9D?)ihV-PFPQ@S{>#D6D0rtwI!rEBpKLi
z4nS#3ESND8(S(SBP)Jxw7DX?A&FIlEo6}$&?%)%wZo)zq+-|kF3`tZ^=ulz6QrUeF
zv~KLu#b*MEu)&ew{D9PWu9GWLbRgN1^klc;Mni>S&q6rF?YBv`)Ku@}O6Rtu*K~et
z(y2S|+_-~?y8}eg9ghjo`rguojZ^8K10Ky(#E>pc<jTg-Ca0Uq4^tC?xPx?Ix31YV
zVYl*rjNJ)|0?csMGQG0{pK7lMjdi^Yy%PiMO>(wnm^L0PmV;T6cybM9dA68la8&c&
zSzVaLWcZuErxe_gnFr1DLX;oUSzmM-{~1kcW2VaY8~_5zq6G&-N3$*!zh;ge0gorL
zHAtrW&Dnqkne2Tj3W{!N3c1OHOn}H@Llr={iO7Wrv$z2u&_rZGG%xs15Eu|2fK8^n
zHGE!=3IRhBIaR(4+Wr9&{*JICQTagApe6I;g(oFihcLtaucg(sU_Gz86cE4apLYtQ
zE4uMW<Af-xkukIs5cgeVh=!aN5CR|WlZUDlB}(?G@?qNJSltwrn+5o(RFiB#AMo(8
zAv9)=inc0|;a|pPpSDu;ItsqR&r@-U7wRo%6P2Kk4<i1E6zmC2bhLFDoPSGBA#@3{
z`|lCW+(wADQaWarOnD*Li13wo(|!IF0~%Eh2QrwA(OCL(k2<+db(WzR(4Kin3#nbJ
z<>p_crdYi=lf$qwUrb%iH3V@i{VHA!p&2gmSpFJiJgk%q;vf|XSo<U*FU$84)e?6L
zV(}o3h=f|hNrE8D<hdFh{SIsO|9kjm+Ao*vjnvB#R(NW1(-5MxHd<Rmx5H4VQdImW
zc{lJ7)R(SrpAW7+YZz?}KTN)E1;?&DYuQnaGtXw=R6pVQml@p57r5W1OIm;vBO2JZ
z{Z73q1A0=90c9HcFZwiq;70GZ-_9y^6`SJxj>~x-f3NN5R&+rj3#}vaBNQz>;A^!A
z6pVfUz~6%8@4vf^f4+C(z1BRHo#y#qlnQFWo>1tKhKNDoEHiIV$!F%8DHN)S^SUWU
zkddl}O315e^{l_T8)2!>O`k+h3ypGE)fTY3BZ3R--5%iivToM^;}Uw};hF9KipPpu
zRDxPsHJwxg^d=@q98;_I>*7J_aPYiPp4;iU={eU80v|r8-_{{aw~cQ9(Du|K-?l`@
zoS87@EmV~VDur3#N`8*k@c!zRjz6yN(r1#-%tNQ^G)ceL=guk5<_*j+>tR=zlsmhY
zTnLfu!)>a!JSqYMaOWsK8Bx^@O1pvpkrnv-?_p!N{;|9a?q9Aj{E`I=4CT9LR$%jV
zQ*@8;p4tCbM5jmqq6^R(9>(3;hD`Z(SeWbl9VJ>?NNVk+YQt4PJX=_+3empSwCY6f
z-DGVuN;0zH4(*tk)a|05fKW-yk}K8Ena=5knfRk$n|v)5fUJciO2j6361>(e<*X8|
z?IwRd8avXHSfSYnU#Tw;TA}K?B;b>sg669}JdU^^|9G{A_iz^RmJNFBZMkc@@6U8c
z^?O+(>!A07+xyP-{vu>1{gy)#5UK5TicUH@Cju`PjCtZq-$|uF<>`x>h2y+}-9BO@
zLCTp05Gg1G<R{eXIDb1eiZR3;4*7v>N*}#@{iP}i(CIx}83$Om*B*^Lyo(Eq4SZ#N
zapTm?0)AzD*vOID2V+rgGW|l0h<H!nN<fAWPO=%F@473{5yy;)^N_94&UOgI`z=ZH
zd=(5$&R7}}lp;>fcs$36HJep*k3p6csFN}?BdF>L!0z{z2Rkp~O6+?zS*(_rHX+#y
zv3bB@Nc<Yv9_K>9mnK!bNj-tN36<)qkP`2Ld1iH9exz8Kd~VWk3~`mCAS4mNEyzJs
zzw`<0D_9u%9hLKRPMKZ!{in?SWB!ZW$`krK-Q)4bO-s(2d8Sa+1ZHs|R^zGUUc;io
z!<N1pAkS6P8Qrciee7u1<L;cmz7$(_O!9X+r^<QmTLI`BxLdrKlnY0k4R;4$6cuKw
zapx~WB<-jYDPyH)iiG$=$uCIAc6np!qrmG_jXg=&_h@YBxqA&AVu2r306uLNl1jUN
zo0Q#0MIco>+H~i&>22PH`SxLzQP7Z5x~h->z!j-BxmAu{xI!iv83gl=U5a4-XdoG+
z(hdnZU1YVc=qU3=c=lWJE=1CdV!Ir_Lh~JqAi2WyQ=+O_dUhduw}YjwCYO9K3_MeJ
zR~IFJg`_@ySM?sbkEzgh9nB@d@Y-%ix}fUwcFm>|{_~pI{hWp4Fx61~fHLaFx0hE1
z2<b?%TY-K+%-0pphi*pnpKu}?P0(+F3%-1lm6Km-ff=6x`<vG>Pa=p6zRJ+n1ny8+
z#(|nbf%=2-t%sF)^1X8&8huUj*h0j;Y%50-Jt?JA6gryg>65~i&P$x?iH$0_e7Bo4
z$%8%N9pm@>4x9fcc!|QS?EeyF(jtZ^;Qps71*B4-r3us0_W{6izWm+}$Qm1zJkWYV
zwZG2KH%@mJ%DAb8Mvz7*iL3EGUp!%vSmPO&kqm;>&7L1@pZ#7N)7HTb`~wn3kY)Xa
zNW<}^pT4n)AJ|4BtTfYZlGFYYP!T@Nf)=^le($eLE~K>#s_w$XIakGl&TSM9(JayE
zKve-!f86_UFF*(nO$m+z0sd}djcIi6`M;0UZG@R0?$g)f=XQ7sL4PgWSN+$;^ITYj
z>*A}u#I)%(^T=DPp`0UNiXlAG4NXM;t&jTu{$ad6M0r-%k@+BOb)ue>cBzT@e3=nz
z5^1*pL*q4q!ZS>DBKbWLf8)eJGvh784=Nes9CSY*rvT{F^LzSIY1YM3{=#r#k6ynG
z@pP4%OPzalz)fNi#<`QUl=&GlXT=g@$}DBh`Ctpsay9@bU9WCQSzd+(|2x}<AG%l)
zDkm7WV^w0EG(aecLqgg_KrqEy`(P}YwSVF)g@Y}^R`sc(ODH^KC>X*}Dnb6OPSFN6
zuOj-v;tMdTp<6*dD$iw=oG0ltqY9OR?nrz0Gm2S;>hhJlGQCV_V&q)#BX3Y+YmQ}V
z>`rzM51DM-Ey}32MZ81cma9YFA_$bVIjfcBYH)uE*`hAWlT2yFbT5z!KH$)vRKGWx
z3Hp6^6Jvd^@p9PkOXczL0Uno7AcaGcODNOr-w_}!D@{8F7Zfcd^PV*6?|=-dDMq-z
z4Ww;CRX1kYyM)lJq_7NM%Dml*)%gQb%(+_U=E!T<r_5RX1Whs)54enR;i%g(_B7K`
zdzlLxl%j`O=rr!a&UG1ACZ2LpP7QYr=1ZR_3th`P%Q#&`!$<Tq=G_VR5CVB)zwmjg
z78!t*wVH?5=vHEA&WyezMxLmUhDEtpN-dW{Zq|wP+MJ!{57sj1TJ+j`R!OdLi?zB9
zqoh~XnBeh{g?vN^i8i^L$P|1EP~qsib7^U-ZyO|fBYA*7u%Ovck)Lmq%P!0$OM3a#
zsWg`DnM*wfrqS=HWPG@&HLRnP7#&&52>^DtYo1@VoQY;SRP4#E6wctQWl;{L4_LUA
zM0!(i)f{}}TkzBQcCB?dY}<GR1pg4Ttnunt6HsZF&x>yhwTAcMo0#F!xp3N|@4+eq
zZL}|8d4HIiKbHF7*iN%AUyfkwt^`U}<$ua)@%O|xAOt>TjxH%)x&DsnPHdcO&;}&g
z*R1_)@j-y=_QdP;2pzpjIh}dzwyH`tJ6e2FIf$boz{o!Th@GcVG#fZoEWNX@I30_1
zF>`x2j;?d@kW|+p>E_|LQfji{CMtSMUS|dGV=qt_Ch2NiTWyV58nE1u-rgLkh@bz6
z-l`t7T@wzwK-Fpa2fkdK)FgKn=n4qP_2@`;EKL3$Y3qsSW{Au1!=9-9P#XoGbGXK_
zCV$LEVMJ?!N`cgK1A5v%4i<9tt1?(jK5eSdTmj#T>oM!AA^N9*5)q1Ly5`PuQmI7w
zb`78EUlFSm&%UEX)0X?puBoM)ErJOpCDk~xT8)Pz)=uzqSXL{!glbdDXaJT`-$h@%
zk1*ITvn3@6VV;ftv_gB=V|h7^lVgSApt)#iq|$1>sZ=do&W`rkbo%-`PBCXy@F6(~
zzC7$^8ecM6PIG1H^WXqk@>UGsc5I(37=JyEh$RG#n*j%u+(`BD_CTgHv+YaCQe)Rj
zi*Zj6ZfSd`7Srz4h!tlM20&Y-P3<dU1`oEj$#BmjNS9Ofnd8CGKhqh(Z($9Va{OcK
z7sON(TGk$5^#&dn=yTUitUb!2qpvt4Bk&G0a=!Z0_dd_RL7nwMcOyq6`7Vst{U<O)
zfI})L>muX&E9;_e!GT<nXlhq$b>x5!1yYI5`mhea(_iK*B-Xp}HvyD;PN6l9RNeXg
z%Mi6e45QhsK;U;}2MSbnq)YriREr4|Ev=K8eh4(K)t`mF1`48S!S4EhQ7x`C#~5t1
zG{YJy%Kva4)4rqxpwnWx>7&66V$<+>=;;6%h&nr7aFv_MMq06MFR1Mal7YG)Zt(_w
zrVwdUj+&AVXWfwf9-5nqd&D=PLn~xIBLu4D>gb%;vsS)*sz2Wj4-T)>glq$3ELSY+
zrV@1kpB^_1WbyDLuELDa4R0^%PnU<SnO0i8uN?ji*csOI#f%MqWvDhR|6qe`_7(wH
z_wbKjZszg%=*bfjeH((vW(HSguXP`QALj<JLfoBNtk3%bka}S^!l)r;q2k?BG$u*A
zX=Uh$fo4>vLI9s7-m%N&mE|movdcWfBF34=5t#R>fw4)msQjg9Jm(-;PEvTs5k7=J
zBuo3_Z*{YH3i(Mgnp>-m4S)Nc(<1=~_?TGTdV%c?^zRfufm=nq(Qa_d>Z7~Y=r3I8
zuko~AkA?4O%HV2;u8;PEGQ9)zaa+3tB`k?#0a@Izj@XJAz|)U2L>78YJr)N0pw7K%
zs<;@R{ZGx3rr0^dMXocBGnE3zB{Vcmij=AqtG6qShn9(Jy+S;CX2mEA>wf@a_yr(x
z<3Z;7ubQPgIOk1T8<YgsPCEahMkx*-^`}>_WR&0I2C_`UF=G0MFG${~Xgi8K{`>sI
z+GI$i$;|`aXs;S7*dIc7tk1!a-4P7#8h20G1DWR$S0qNp(!1EUm|$XV2~>jm&J+Wb
zVvZfD)xT$csv*lU0^MSj>wW{i%HIz%y~+|4X{*{Pg1a8jdsld?(4)0^JRI$pyTCn|
zT?O)8oO1(pnEq_Ah6qF3COz<yLlyojRsz!Gc9m|L^2;1<iUt^mdK?Z*{4feFQCAT(
zJqfXqunax1ie#DbvfhP)BhIL%gyYW|Q*bf0FyZrJ*W`Np9oY9zI`I_HAz8bo%q@(z
zj__EtIDqWghsax)ZX=V|rQhl~tQy}iL7rZb;0$wx*2bQiBoieT#{4Ts`eD)2+bc2@
zl<P{%$zu82uJ<%uL{}LyTi!Fp<S#Zu#UTs&AF;Sb4w<o2?i-tAbeU^NNudj<Xq`t6
zl4?TR3v=yy;Muf?_E;uhlt|pj&Z?qy)!9lsMTTRYNkFr{Fdp4rVr@N^>7!@(Sozsp
zsUlODlHplt;!z}%u@Az28Pj0f3qrVDw2`9J1dz3BJ2FU<DT#W{@#wY_@)Pjaf?!fI
zYCE)U@LM`NNLa`4_ZQvS+FJpzW2J|(+-?<1yFhFmnw;U}I!YAaX|JNPqR714JMV2C
z3_@4UJ}V~i9rzoZtfG#5dx~ES1hLZl`9mgBEO!<t#uZimyd5D<m+m%A?Vr6OW__em
z_%_4&+ZxBR^We0wpiQ2Rec8}_3!%nLp*{%NtliSgQ{4YI@yQrWa{(MEb6ADdeJV)#
zt<#Po8e3>==d=?5F0F#jDA}lly`u@3h!@(R>D22a;+l@k>j#+;^=m=IER<}VdUHT;
z2l7s8?gjm6$Vc)95dESwXPwR<evpQ8u#HbSQZvHmk&rkpd8l}Wggvj<!6(g%F&qXy
z(rPH8xDZ3ms?UA*`w29>wotMPLmLdnowb6g>cn}A)$b=@52n~~Bs28RH`c+SmVC*Z
zfEc8kDJs{Yhf<PLSUN5NV!oeQhs@8}emU}}T<mPtW2W4)k;tEggVa^IVpS&;qls2x
zQfbLhD2&yn8QxpOTZ7U>>i|&CrxW2O;@y^6-^?e|Zfu7nzv8$))`qGRI8L8Lp}EAH
z$zjgABn%BO3|b@00Hq@6v1{|*3>J9TEW|S|9m9xT-$^B%Qyk!~qC&9SW6?DSO$|Wi
zaL@U3+Fjxfr)y5xF;jouR(y`n<C%6qgC8oqY$P8U%a!QnSG*v+U{oJ+bP`WiSJLn|
zgscMyv2||-V%f3IA!;CjBg^L8OULSw_@K*b{|_B-;K`X&qLDjF<J}w*)w3abd?dw{
zhh+e=43kqiDVgz$&ql__d!U25yogrh#Z*rtRJ$*Nev~oGXl3nu97#mm_klU!VZ6jI
z-VfH2rHK)oAI>=6I1SiPg^N^}_UFQUC8@O4I55AopRno&{5o??1&U~Dt3WU+8fehm
zTN4N%BQJsj)Kum-Rl`}GoKgCw4ciwKQW6~ADXcWUfUCRZW?&~?xrck&=fV==yz)w>
z_L1j3=KC-X?<IwP*nP=kuU~HhM|K(?4vPbqsg#kEs=4Gj!b3Nde&}D$a^OEEju}q%
zW8$vlJz+$^LsCd9IIB~9zY>QhZm}}&($O0r0Q!i}%QX*{+Y5K$q|SZ@jlp|!udcB%
zwWjOEY1r^BTbQdXrSnD%l5)LKP2fx{q~u+O6-%I+!I%G#A2J?El-!-SN*&$3OI}g2
z$0@G>_oL_y9R&SGL7-h((XqQxRZr3T#5Ypw*7r*J`%Al4UOanrPm^^zFHp1V5o)qt
zt&*GG*Xr>FJ%>>Kckw1bzL0}!O+E*Gza#~vzHW86%hO?nvErL)g)#S#t7W?Cdt(m6
z$0L?U)egNB@l~DqmmZSXB~?>1y*yyy*!~2Fsa_9zIrnOXL8)3@VxlH3_`p!C^#vk`
zH}T`YhcXvqT7L#OS{mR{h!W7PIUbA01>gNfa|W%2LS+RG4{q!J1mc{PZy&^kkeKb1
z{{Cf18Brm6sI__dtU#ZY*mf6@==xN+utgYl_qI2g^=GEipx^lEvm5;9-}fJx5i@Q-
z1-Zbt-e#|^3ztuxx$Z=?@CM_KuLE`-?@fsQ!PmY2qwG5!M=N6E9W?-GTI*;UdtmB4
zBfR^`x#$r#FGPr-BktYLzU4vFO;};*zGmuO)~-Aa0>boM|3m64#IU;f{P2_jhfw1p
z<cXyAKm8UK^$>nqwLdmgorcEg3v`6us%B&P=al9&y6yA2OsxOq!b*`88TQ=84PupN
zAXLL5MuUZWxy5{_06^SEM%SXAh+0qX<#YJ0b`RP8PM6SKT;n#CvEkC9#OKn><P0AI
zZKAxt&k3@$O-Lp)>tHZ`)mhf$E#@VnNTP(UE&Wx>(##A$xKB1VGQi4$r92jCDU^&7
z;MqfS-{IF=Fw7U3xN<IzI1)V2Usl7+R3=zms-PcLlj4s!2dGX?IYGtB`=d!Kp_8bi
zFtvrDIgpPpJSnR`^zTw7&osv|(&ywZa~al^vOHhbprM|g)30-ESQDs=_p+|d3SFwT
zm$?G+(9SJKuB~KnRCsYC+SZuMxoSrBc4zOof;+i0NU5owRI4(;`-Lf69qH(4#YdKf
z&Mo3FV(33h4amaS(4>{lg3?mi)i%xBL2V0?w;D}i8~w8?X<~{~Kn<g}1^a{<cnj2S
z6IV%4HX*pmdP1~lwZOOze3DN_4U_#%FnBr*63R|VvVjZAaGJ3w-8;aAfdik3VR>G{
zn7$m;OjFLa<Q&YkQc$-VDdA7Wc(JZw>O3jrf-l&s0{EcfARf+PJ9i4rVZoG0V*Zg<
z-SJc?dCr;9i95l47fL72I(fe3uY*%hU;3-ec*(t|^-$BpG|P(s8g8X;=(!BW=X^XQ
zS}PLzpm6_%F0IGzI_|D2J@Vu_{4V<=@~wXOcFTg_Es&i_gR|>L;}fhNWr_s`n-d{K
zFNH(xIpAk-Pv|9(0(h|3Ntj%sMrTwxhnw7BZN&36DCiYb^7EM!-OVBsw$ZK90&eAp
zX|bzlDE2J!!a7G_uTRe%kJd;vLcc~ROB~@UvR%N}Wn|UAJX;+|C?MwMuFM@CC^i~C
za~G?QA1MeP7(eU~4_KWfNC3DvdTmcQ^yUfKF(6FKVy;4LM9&SBS*-SNy6W%GWQt<%
z&Z;n^aByT?)1MFvV_az6p2%{gEhwk4iZyMz@9&Z)BQDEP)rr=BZ4<N1%=nt%#0EL>
zFZloR@S&`ifYT@Ehz>VydH7yQde>-Cmh|niY1Rx3oBD5D#I^eQrIh`$h6onbK0C#8
zr~;trM{&3u3`X*9s7Y2yuhGsJ{oy1283skN(|50jr#%~XDyo04rMbty4|@n@iRs9{
z8kk{Fivu|s7fUw#f!L~G35LglqD2$#8)S>AR!U^tQ;5+dLK%yCCMekvu4l(N5*;He
zjtf<oF!?tiSgS3fSQWg(pX*peDW@}8V*pToF`BYLiQb?!&!PP*Q@dSKrgg<<hRVRk
zA{k%wI2m6qv5>-k*;Qh&`6HZ_+Trgj!E`J{?|N%0D?LYr<gYtw_Y;Ogv7YFCN2$YF
zjOZ2v=;yjHRu|f=1U-KLH)22}QZXgkW^}_-Dxyv0v9JpZ+zg(&W*67rO}p`>;V3}J
zIZnG5CG_sWy^K=iTxhIL$;55Vx6`uTL7#5n+~9-c(g$-yvOi;V6-~zStOI!SS7rYb
zUmbyWA<(>h{d2-39WSf`74`?MKE)^ezBEG*+ks4?j^+bniIsCTw8E`L3)h&vv$ncj
zpdzA~-)D?2mxjtnCy(1<xjpGI`<r}ol#1yG;fUPVAaQRc9yGz?NXnx*h6bgbEVWf8
z^}l?m?8cz(+w{kcJB>CG)(zd`MUgM)jw_!JrX`!|!&D02j@=pp_qU_E52WN5iHH9-
zWn5{74hU#zyAt$Wp#NHPB<U|e0m3U7`1&(?b}3>q-9!#6QK`^Zq3_cU%8b_NsFg}L
zTtX&PqXqE0g^T#RjzkuGB*ZIV^Dz*7IhE56P=DvH7??LHx3Y*{{ylTQdwF_4+Z(O}
z1v$k&pPs`01q#9$YtCtDAT7iJO6-vU(F%=Lqvo{1g;T{$T2GJ0!OM2@02m#==Qk@J
zbkXh&)H^fHd?L*{jI{!*2~^T;Re=I;7nWMtB^wFO25<upMC0zM7V78KI^xB%rKG6y
zY*cHj>K2~b2JvmUzI9bfZytf6qcSl;n)hIs{!UYn0B+DrGrh`E2G`k;7^579F$Chu
z;AYSbemE?;X)Qw#F&oMiz{5e?$$;ZcPN0HJSvuO8CE{+H^mB1V<4(LY*bzi>)Q4+p
zm&P-C+zL^bduy`~Q$)X8)Kd3@F3aNii;{_0ETT>>9pxr{J_><3GSAhS6mE)NnoHh9
zBZY6O%3IU+{<>_T1o?9I^OJUB)FB0x-R#t+Qj@7hNpBM<&a7oJ;Gsdcr$&Ve@=qq5
z2d!3fd(&YF`wgsYRK9sQ%0hdRmQ{#-E99_dW@oG-2~SHtND|~jXFi#MTQS`%NGudA
z&uJQ%s#(RXb5jS~5(&vci*W_yOkcDM&r*pXo`6ei_|}xQTrZZe10v}J3%N6gNINh>
zCI__;M&!kfc<Lx0z%SS(9*Y3^qtedbhsxiZ-iR)yQoZqnDMZ6#ob@uwMLicY`0*3U
z(`mg4{^ueDsG?NS<<vO_^L2eeX6KNWq{4CZc_OKyL;v6~aOfjFVeTbP7Hr-sQ$Wdj
zZl=kLj+teONi~QIalbuCDH+YJ=3GFUicd{NeR$h-n6qXKumt0%&m=XIA+vquQ5Io*
z9FBQUolwSgn)p*4%XK@DJsWUJ{*Cdc1!EK}Wpg|9|EP90>aN4*5_>1Cb7&gCU6P2$
z2$tl2!E;XQMNruq1}!?GAKqk{ctXC`AiUOQQvyR58y_preTz?wAIx)pOr!ci=M?$m
zt}+jDqnuI%piGjt9OlC?Rgidg^&-)`y2<rABaYqrGs7Jr+^ZFKT4i|`4m{v~Pyd<V
zZKGPa&)y=O#LNC9UHU@HR?d@2pZd_|-q0|(v)wI>me^8fqh@qE!#=SchrQt0zt2n>
zL-K63c76KZ?~WqON8obtF5GH%o}$+d)7*k@kSAphz%j$}HPcaN0GX8Zuin~Gz5==2
z6g5*$KGY$mB6G`IuF*jtt5RGnJ`aYicj;TMQF2p-&fdM&A)R?PlZp@|uw)N0Ae%Ta
zcD}IRsBA<DwKjE<Rk!V|;a;foN=3${iVbI*fj=TdNXDf-T5p8&yqbZ(@W9#DzvD1R
zVbwPTs6L(4@|N)hNvvEvTCLa4??ppvP0{@Nb@`HzQRwGPGM|@zG4FA+2px?&KW&~e
z`oe&gGks3t-I@F`aytDzF8qx6vmp9b8X*mO)s2M$LfICKBQM@Y?Vibh=mNpXl+<p%
z#Vh!}nRpTUJNcUzuO#jJkFhavIbfk9wT16H(JKC+?>_iAlQZjl)u}-Zd&nrktLIFf
z$8ppzh&I`#>oVOT&05W-CeQjJVqTJJ1$}S}3pduiVN?xx`YjPyl(MWCb-yV<L7Q?v
z;HRi!)@4Uj9q7FOA!m^}yRFZ><{95rfcS#LF^d@D<T5yX_YBOx6Q~%5c=<29kQT;9
zivY&;%~$&$EBZeQ^xXgtyRXkz<I<@5xVN$uT30=Elc!WI3v`PF4xKuM{59c5!OECJ
zP28Ys%$@4;5e3iGs~68KOk-Qm96eVk{;MEOwkfSc;O;=H9C25guEmsl%q4u*YNx@4
zPuF-Na}WEYXWqV%4>M0oYTmMMNcwSj`H9ls`OCuBhtKA>Z(j(H@b5GeJo7-*%PfH3
z<vbiyUH@pPged9sH6V+yQ^2~MoaW41?i3&CvMyDTcZ|wf<Uk@B&gbLz=|1DR_YkgX
zka!OFPQ=pNWB(~{r5EYjzGT4V;OKRp>;v%5*nt94Z7$2p^N3EjPzu-Vw-WP>IQ1nk
zZcZ3jXO^2D2@R;tnl>fF_i5?ZF@WCZ&?I8`_2x4{E37vzh6i>(MzqsHBEO~8nCsGN
z3&msodW)<%3Wk>wpEtIxS0*Sf+P{?eYEe{nD1i~doT^hb-Tsznc3<b)0sES8?#h^m
z6{fH3%c9BJ2R-C0xVp++r9y+n=DNoEc64XFcI_+E%?~K4&)!%ZZP13c*%0PX`|a6*
zuvHXISvC`7SZXl>ww-=0#j-XtF$U>{64)wZ3d}hX6;|Qij?yF38vZo9XTE*mt=19#
ziK=R%j@K*p6EmBxB{w0H>og4|)?rLHjtic~-ZH2<S|N!AO<SH(S65i4jxBD@-JP-j
zP=?e1G7%6X!xB_{z|9LD@L^@tg2c73e{-UI>M3%8$rO2-xtfD4IAwM*v#I4{u(j4V
zg<>J34tgXChpC_*6{$rHcNN?b4Sd|wE1R6zBL@f+u@#vucB8A-A8BN@eCt1LMCHvk
zL6yfzHOj&fS_;0%oF1aHmn<6rpf%P-Lg}xQQvl<RmOi=MlTt+JFoSa^0F8&PDaWk(
zqN&GBvc~<Jho2h*e_9X6ZzJ~A#tdr`CO2l-9gZ|x-a3bQKMK^OyRqAIqMJm>a|zy-
zT{#JvXr;TW@8#le?#+{kmoHh33=eojC60E&>bH`Y_pwf(^|1N@%fLhhbr6R}QZ?oj
zz&%d1AxXgZf?wf`D&-Fs+CF|Z$t7(6lz-avAuTy*GnX;;J88ShFrHGv>nI}3Bv*9k
zn4%b$kMV?@)*oLXcE$`8R;3;Ul0OM@Vv`CJ11IBV6SSN~xc4WM;aJF$wGaJ#BVCZo
zhLXOg>SHR(;EPh~DsGxtcNGaUOZvPFz^7S>Tw2c@i67i|DRJV;e6F)t>#aL)?jU7?
zsS%Ne;9Y}PH_sPiEnb8LW!A<%wmVmR&1apZnYFrOhAyaPa!-%Rq2<?O`l6WgNfrg_
zE<>EJx4vb)=|<l9>Z8yqAAuZ7WloKBB})hS7c0`LrK;!vPF-G85vhaYmB*naAXwVz
z9j7#At4uLg&Z;WyST1jYxe?ZEMA|gRr7Y*2V<}?uw>|E|Z>>*j+7r|3(fZkmjV#Zd
zC$@{)t6Ac68wEcJ?Oa`=nkP2(poIH~LdS968I3h1f^aeUlQ{W(@@P3;uj^FY^B0Uw
z95C|jehloQgcpljm+CyXqAHGNMTEN(^;qaQ?@KCZhQ}NVYA|orOuII5#`0?3Y)f6k
zxOhYcv9{dV_A*9IUKee=853TEdjsIS*e<OoT>Hr!LKF^7N8L`Xm&5(@7x*NewZs3N
z$S2g3k%Mt^rJ*GAGpEH?(?O^CeK0};np#<S8XL7Cz_Q!qlqvlU+Xc0SV8}A%fD;9R
zilH0&JLa*ximaOFWYj2cLi?MJcE_g2hZ%fF1M@=^Sj%Fo0?x`q0@cKS*T~qD;jR=c
zIn|JbWl>!VSPLLXBn#)V`z}=;1r2cS9#EGf=Gv9}Bge7&1W+e?gssNL4x{`51A%Mk
z@*0;{mLlXY)QLhF=j6{VO4~^T@KU-gWn)aQXl*7ipefFSQ#l>Yd*~mzf$~Y`KdQO1
zDr+_!PVW~M{imCiEK#unaVlh$&Yqo;8rq<^*vnmJ%{U9hR&)2VRCw}Cpj3TQvT-9W
zUQIzwYxG0YA~s1hsEWXFpcone?P`!uN~cY=G2yR#ycEwPR94)DjXY`<9jwwZN{+g_
z__Dy)4EBnxmcHT0oZx6wyz0eqB89ES5TT{3pT9?d%s%JFj<RqXF!Oc1^i*#Xc9`O{
zQfpMFO#@&$q$Vn56u|Yj0;wx+bse6bK<ShHo!497&x(1C3zu?O*r^G?DpDbkHbu}A
zUezYIIJ~aHAxF~OlrFzXyOurFKAVAEP1i1F5q<tDmp#1eBw3uD9X2^&Vyk<eh?*M*
zBRPCBJl~)~B^dT`Ax{t{Il5ZGh7ByM;}<r%FyxNW;S5U3jg70j8YkGROEPN;Uk!Dz
z`uLLRVo8S%hD0Ux7BwtjLRTljd&#F%H%N;cAzjeWT8Q|HbT}5}6fLPON5VMrj7lkC
zBG`e(d&h61hc~pAhWTCg?FsO{a1D(loPrLk($&@8#>vI`4Eu`}DH+<484~yiA{R*x
zgL>#UlPRpL*<jh5rz2A$u3lQ`n3UaXsh>Avc5rGscsD5k$o&8~u=3=@;AQT3u(f9k
z4OIf^spo%qhES3XKe}T{z#2*3!7U;eS(Hxp3%z-9;}FD{k6xaq)vmW5%r=G$*nj!5
zg#z_Q6i%Gjd$M)C%THA-cA{&4aV*R|d+BApFUudeAD8aP@)N$9J~Gz$1Rj@qpt3Pc
zV|Y4xa%C{69PI<1VmtTG%v^?Dg@Czrxs-f!r@l^=hOI{4F3#+CmEXGruwA+*op#<d
zEs*6l0W1MFy#ggK!&x^LZZa9dSdy#3!5-$zyIjJUV{+?Q2$%v^fb{<Zxj;t0yq&Z!
zn{Py-if1s_loLyo^2dXNJ5K6U*en%$&`>Xx8B&wiDM&??R}K|_p|90Lp`7fkSkb<?
zU4Aq>pIkzrzrFd<|GxQkHoLn1Q``P>KKu3NEQ}@}+YdJ%ulC=KKed0o`RR4|aM7M!
zOwQUo9-)NWnItr*a)$Ul9gW6Z>$`WOc!rLssQn{~?e%ojp1yzo^Z4`C6i@%*FiYCU
zzY0V5zO?j_&K7)sdS^vd4H-324B4c#=I&~6jG%U)JEmL_U$T|E(Pn~&Tkmev)R3g^
zmb=Z}Z5!P^+)|Dlve(9ir2sR5*q+<c(d6=aHa$H@1?h}cmHk7_ww<vi*wtH83sOpG
z*DU1(OF5OLrQQXE?sS119LO1(-xs}Ng=^8%44JdusWx$cW(ib>a7s#nWDqAoI`a5_
zPAWCjK^Bpr;#U4G+TTjYyVDCF+VQlmbv(7lbiACG9Z%|_n`!I`>Iv?N)DwMs)V@7x
z-p~Vwr%onB$FfWS@pgsynUm+!>Gb^c>|(qq5?oOvQhJgGl!Hc}82|}=RzlMGAfY)0
z%o}SBh!TQ-mk~%9o2m~LRyTE^2Wr73hv=ilLW=}2*$o8$ON(eC6+OYS#F~IUgtvno
z$%3S1A9RF_b)S3Nucj!M-IkM;{X>ky+x3NnBI$6xH!~GM(Uj*lb8<C1|9JkN@$_ey
zJD$Erd7F2BVsw7my7Nn@%WlX-L7u-IF1I!n+doWyH9tW%r$~V@Ay!Wt^#h3R69Z!a
zkX}zC#t^Q6>;mAPN=+rvR%v7!VB`{baKLo%+@8L?xSChZLa3TWR{@8X+s13O32WS8
zxJ3{w@aGUliXrg;#*`IhSBG#aN}~+n1a=6q4w3Fi!_Ss>7-`r#(y&J%jbclrk$a?J
zJ<_m$9%)#QG^|G&)*}t;k%n!DG?YsaUKS=zs8gJGr!YxE!eGFEg2Dt@u+<xm1}19~
z#(X~&OBFUpS+7heS%QKjTG9WbpA$X>%mA#gD9APo*x*DdC;}!S{F6Rfi3habRD-4E
z;YN}SFO*#=kQoTWo?^4506Z}|tpy-E_!p>u4t<z0&%oL>H>+7r=sz)20kXHOaVs}7
z+MSr)8TBnxE4wxT7fSp1_aw(i=(KPzN?(Ya8_qQ(Kin>%SlMu9r>K)98}y}2gqE<C
zzX8&j@^WNCkkif`tQ0>*f;U#Xzm*}!4YuAg(MLmFKSbEXX@2}NfKf?<Pq+g(5#UsR
z>j2K=IhTh>PsCj@yq^<E>NR@^r%7O?-}3<fp}3-^HC$0?0In!Cui}b&2UnE-Jihb*
zqsAvVnAZY4XBd(#_`4-WSydqn$p~E4^|~q)6XlvkRde)1V|#)g>Id}jMz2N6*kk&N
z7K+J3*(_3o7Es8xt2XS1n`JqwvmbANmbz>8h5*0OPnl)4;#9t8-nheXPqEp=iDIM5
z|1Yo1yo`L2e*R+h;pykEhHula_UTs_>F2M87X)nw*#Tqy75HtdiwZ!wbG?oLG!c79
z&FxN31g+fP%ig{nvW#(dSKz8tKLw539*Ic|0<d})lh>4X>%iBMdECxUQcPTbdAXNB
z-NBwcau-hmp7}e~qN~NyfBJDhJX_^jAf$pwT|1)L$Z6aiHQ&z()QmA=h;$J~Ss23U
z_)Loo;l!}&2lMxHV#dO(Jw!T6c%a_~pFPpJkN;}v%LYOEm<GY|wn6x>bLQ=UWXpD-
z7`y0BJD`A}>e-Le&ptrQ^H1Y{#e3HM!?a~10K{Tjw!Pw|!*<I|s^qVeX|;e*{yN&a
z%do5+9Mamm+>M;M1YqpnGk>;1TU^$XiSuKUDJ_Q`f0sQ^Cgz^#(~G4ibh|tQ4pCmd
zS5V-P$g)GPaI(L`G1?Mt0g5m!%@jZpBiXS$v-%V-A}I#%NVlkqEJ)l%F;h+V5A7*`
z<>j3@0#7qzqcOdappMx3ue=D;a)e?vW=(>p)5gjD0pO1*`0oa9!5^29p8w!^)!ja`
z^gpWmFT!R<(wBiXEE57bG?#IW7!#MmpD8GpFkmj1fBE$1$8QIogyqa>f9o&()BDHa
z-$M}{kSXmZtqc=fOn?12_!m+rrj45bm=kjP`S<kw@Ow`+^XjevC@TmkuC-D#pg=RF
z?4&g^G$sS3NMS_)NxDi>RFYDXfRa~g*YTbckdhST*V7lxJyhzPpTIxSIDUyBEQ)z^
z@-hP8e;8U@CQtfF(#I%`A_)}1FG(6@qLe{u;d_~5YE3dKlUsu(GHMdFDQ{GI<Mn%Q
zsVE3qBn$_60VR`akoN_+-9iYtVyJb`(bK591cXe+OuZ#EY>QFQwLv4uTCfUZE*Z%(
zM&>*C8o7|6&~hrH62eL)3}nJK(XWVy`BH_he?7(4bF@8r18~e^5`ME;)(Pc>Iky47
z!J>p=5;5W>@C;GPK)qT>ko^>?za<=bMlSNGNP8(omeQo&G}=3p&|5dS#kBy+TA;-s
z>1R>l^wTemq4nFZw=IsmfOclCPjp3D%5cpyprfmb$fcRM)P8xkyeo|8pF~ux^<X!R
zf02X$biFX-L8A$icUfEHoHOpK3Q2?@5>HO(S|;>vFDeyn01|pHDv)HgTW*wb8WaB%
zkFzvu`IcO(<7=)}8Ox+Du2rT*lPAGz3Zg^$hO<_hAtBvS3^;j5ZL+gTSV=Eu{tM#I
zT1fmt*^01Vc9en7<SM-#r_Dn4y>XBue>yu(T8j51(>A1=coGUqx;yaXH0MNK;3Pbw
zz;)L0Z3pIau7F#P?gny_X&7|yD5)j%=3B&zaindliH%-imvf5yf{0chl~kzu<d!vk
zoMqPHz}+P0Zim^pr3a9A7_{K`uO<Kr2>3uL#+CHLMcV-KNy6YO_)GF~q+-usf35RZ
zw3?+F_~#Igwxs$O_|IBeZ;~yHoUrz8e^vjN*fJN)DEUfTo>SZ#5T{V??iurWiqt*V
zEsM!^HXro}fAnd^fa8|MWIOBlKW;GuD>JkS?+Br5&@5(1wV`yCze@E&tCN~MkFhVb
zx{l%}d(fWyqaL)Q)1)>Ju8!mx6g#?`*6El-JxKF}by7<1MA@~~YPZ&7E>r{r0tDqz
zmw`1b69GAwLDv@*mpfb^CzpUmLIjtBXfBumm$w^fF3$pg^3D1|)zOvuM=2uc01j&X
zBh)K&PH552-%*VVU^^GWrqMR<lt$tqZut&`ng?uWmihTOV$v_6&IM(fVOM^97bPwC
zS&;^swA`RjXKOB$1~&8VbRw@7sg9oa(iAo*s{$BS?&%-)8&^23opUs<VsMdP@%%?&
zK9b6`le&+8-VFN_0}MO2s4>YM3`0p&)jj1E;|W5o%ZAbKrR;H(iRb0cHaIA?vVo&G
zl4UBIACD@3Df0*Zr4{bmr|>ha!WXn7ou}|>Ww(xJWn|<_;AHmtlSL^nm^G4ec7%Az
zdY^+_dG}>|1}B)dmhN66JwxMD4xIzSMQ36PgdQ+|Ik}}fKYy&AvucoZOv?dAM%qi6
z%q%2FF#@mFVl@w;!#-4aJLPd*D^#m==Uu=7(k+%tj9EU<?!4k_y=Fi+KHW@FOX-hV
zeQY_%PLK~4i7`3Cpy_FmSSiF~YNtp{K4U?61@Yv72ngy8=;tN41{5BL1_8@@Nh#XO
zFOW5Vp3V<_0MH8{DF0SVkmf=>ZsrXj>~QA)i8muf`@9+FhrR|Bm33WgHJf~S4yT|F
zqhHf~38r|^;Q1kftv!?h5op3CBM4JROs)3DI|;j_$dOhaSLh>0zRy!+iBC7ll)I-M
zE#yyMV{4gZ)6B3=J`Z5TEHTtoYF)+z=Uk0{!{{7m7no^A*U_C5S}gC@gG{rXPF>m>
zrv_H*IrMP4OS@a6KwHOkRE4vgwmo&0x$@|0Tr0oW?u>n*r<Zhz`EooLhaC@%`8-#2
z$B6e5-XTb)!@}uT(HQ9PYC;gQ$*+xI{<n@oIz7m{KOyn_WWlvFI_mVGp`(U_YroTf
z#p-%L$7TCLY--DjriQ3VikPHn+tN#aRQjXRK=&SyU*YD&yG`xEFsmfcebutxT+@i)
zBhf*TMLLDJEDCV|qPt1Y0(fn%XmmqQ3Qy-$@s?X9stB`EeQC)>Q=48KQPiRt!T^n-
zOHssN;5f?=XPZ}83L8*VYdEcJyaV@tOTnx%lJTx__jNF<lwr`-26DI`cz$UqR&`L@
z;5F9r!QKO2cXwKGg%-g9Lc?*VzX`2MVVD3x(W-OI82fUx8rnO;iu*-WTUB)>L`2%^
zgG8oDQQgD~Fq$yH9I~IRyn)e5Gk`qtC`uF8<r5miq^|1RY8oA4&2e{+k1;lX7n~5j
zKi&i@MMgr^O8s4eU?44>{VFt#t9o3OtiDoXmsTLsm?-}nc!vIgSv#|=To3o=XnTVm
zsV^m-C-65lNpeSz3h7c`9+^~9YbHHZl`LY+MHTR!!n;;1E5&^3v*PnzdWS3)AD!*k
zr(fkEgl^?osVPPtXeFfQm6MQvu%n!0f6;O2)=AP;1n_JZ`#!pqP|RvcXYVEF;l6fb
z-MyE5m?x{oBxygSG>`Yef_z~*_KM#U2JOwG@Bk65(l1bl8Y9LPcN)fY|JSQSsxRC}
zm6V(b4Cs_9lj|wax(49Vo0wz1-Q6SJKSTt6iHfl1a}+Yr))EL@$0ALC42kB6VDFhg
zph{McjUKTs9w#CuBu`Pw3)qp8K`X9)sTd<1di=Jd+6-RnDc({)a3zTpHoU9TTIV(U
zYRUPKHEe`ZU0>y0NRDgFs&fjkal;;Squ@eE`qq7InM>`Q4D!dzsnTbhMmI7R=|WJk
zlq3CHr}u!cVx&6~53yZ;0NpGH<(3k@>Tn?q(`X1}zv6M!YYqjNWq4G0vP)heDQm8<
zj8zKKXDaos+8xVv8?mm&S~3fQsPSgvho}wk805(p<L}&p&p<G@t`}aWTCJ9#+^nCd
z3_5}!h#0?=cGpb9tYV@)o$1*khdgKEMJbM?VuF;0RPZtd15L|+4rzJEQXYuiOW*HQ
z-I3{^46Z6aIq_lngf*y`WjzAK@?9^Xb6ffO&K-pRR8T&2t>yIyzG(WTR2h@jM^27s
z$hHu9>$);g={lQD&6P?sh3JLQx)CzhwM#P%Vp@q*vy+@F1~8rb9b_Krs*^M}JG&TQ
z{k;E^IoqnM^?_f1KzB*fQyK1t5R|ok!&_0EwVU3mT1YO{?RX?WRA{~a#oVZxE*(aV
z!SC(polp{Sp`}mmq@$~J7|Xm=6CFc8AaeV9RIp=P9bDPBt#LLj_bMVuKfb8~w#J}u
zH$;D192SBp(Q*;j`V4)&h*DVkWeQxEH`Oj?pF^3uB<(SO$Ma@t@HyH7Sl+5{h*Tz(
zeD5+;*ZWs}PPDnup+h~~>w?Pq3noda=<6@t{=Umd-J)P=nlkkZTKmY({~;wbe5|4m
zYvN0*g~RF?Uu9~}a@(vh0}_~Sjf=dl=yI%$*y~huSq%)@VIE=$P1VO+PoR1p8NL2X
zK)H8J2uet6|LRZ3<MtEhG;geHq<@+CQfXsJF6}=tHcA^iKpKBDt0$cpifG~ex<zd?
zsZG~|@4=zgrt#Pl8eZY5FHG_55?%M|Vs^X4jd^lf-+PUP_UwMKqs$~OIlA`J+ZtP>
zv{!NceES~**xR<3fi)}>0XdgZ>n{_RZ*wjyf6I>JIu6~}SNI27EsCNZKp{Y-s?)&T
z-A3&OcQu%vRWGLhe=*e4mi&mFa?PDZ;>4CHiqAvJa`^Wkhwo3<_ouHn)Pog{z^dW>
z=K%ab@;6AnX=tzj!XOOqzlMK2OQ{ThuKdwbBJ#)bZ^u`<d_izL0*F8S6yWm3Y6x?9
ze+;TL7fqUjJW9Rt75MrR?(=*vjSLigrh=Eh-v4-#nc2#jLP5q4#%jT|DrZbh`^y&v
zVNK$H#A}iZcT~7T7Zsjp=9YAL!oOeBeHHK5xnUmjrEh^w<l*bh!J?GB7K(Jw3%n(%
z5KD?MtdVvZu3D#i`kE;(%U@{_q^PGXe{aks;&bNRmfgjFrtUyqzSNyG`%Aw7ugrxu
zX<n2FOsSC#b1OB>T${MYmON0CU}ig}HM5~@Ud-o@%i^eu=jVF-T@?}LN6}rP1_R&U
z#;Kb5HZoe848Ot-nB`=1ufY<NolW%`$wQ(Mm!vfAzSa`m(LQ(~z9(NdwP`>Kf9Ym6
z%RUk<T(B{s8>B!o&hXWEsfOQd_sdUyd|$W7;vHm!b68f2x>L#O<9<xEh^c>ly?MD6
z2{;M}LRBfMETy6>nXhjX<*JFKZN7iq_NcNNpa3-0yWbM;QLmVE^<fTHJWp!jrS3R*
zxrGP?Ht^yv6g-fM{h7JRR?T`7f5aOu-@AX`-k<(63u<JH@E(VM{(1@zG>|aX4~W89
z9De_6`1$l_Pc`eWzsBGmg)EIBYK_DQA~%rZ@TL66Rhpb23swrv%ha!E#?74{lVI*$
zG}Dfg5J2H=G5jkw5cH~_5gfG!gu?4<QV#mj$xDoh@#L1qVva+cK0*}9f4+QV8_GFm
z#vIv(H^Q<3b~DDfb(jZ~UT8m4YO%y_ePrvlG!x@JQK@BLni-Zwq3r*IT{VpCF{@^R
zGNVgVcJq=B3b$mZ7R3L*9XFrg^)%!5w$803q0NljEgz6+Cq=eFp?v!hV@D*x0C2x-
zQFCy^etRD=bxv;VKs^U5f0k}IixNDj*%gCd4#BHnx<q&MVL}lAU9>KU)Rx?Qi71+&
zWSi(T*M#2^!fxH<l15!;<8?xGd5)=QFpa9HFmE3x_{b(KInrY+xRtXR;0p)+_U@%w
z>jZm!H!1CHeed@}NgcUh@YzCUYEeo!IqD^^S}KOP=1axCk`cdie<~Oye;l&kMq{1b
z-9l-EVblG>8PwqYO}4;o4ajg(lGLxFOQGb<bGCC;|LqWO8d(dDG|!A?97_}*OB6i<
z^gM1|v^;|9oI6E^7uL-7rE<iJAC=*I&LX<-hdor=v{4OCfQLI0oEYg@QmoG@hXp&I
zVXOAy^F_nvvV%8OfBDX}RphG_fY<CE_EXCd?UBdY6Uz~8oIq`kVvJyIsdUbJSFK!X
z&ZLM1DI0Z)le;Px&F1OpjKXD^*TmP1-4$L}LIWuQw&V4GRN`pR?7pTiaoEK?=+jFa
ztsthE)6FEcA{IrTQR`mRTaif8>;#O&9##3DLs|&0soT+;e{IcA$AMHQr3EY(T%D&&
z88cm?+T8h3U3R7mG@8RF^{XIG;1+eO-QUBS=Fx$03TJzQ@>xcSUUP7pNut8zN8lbh
z%QYTO;dztzVnWJFFU_``|4JinPbNuCy%jt=H=#<qG4W~ZuV%AExwOkn(;`%*-Kw6~
zBIu#;8Cr*mf5T@dv=t*?6FqWPm<c|#qS3ht!b&{N_N-?u)rAJDT52NavbmJWi|y2{
z-D~TGv%2Se*2dfsQ?LtND($LfjmI*E;p|2Xf##xJ0v-&Z1=+Ey@G=68WZb2%-1vjf
zoE5}w>&V)V?o4#EDq;95<(!vUU9g=g5e3evJm7Bie;>3aaYj9%ycYo2b2)|OF{_<Z
zSj38&GkCY!>e+;8>&@EYca2Nw#x^!N&8Hhxn$ebDr|VT`oc8)(wq4Kz%Y3Hk<7s@$
zB9a@@u0sQ2rFw@8VqSDPa>A%KZsyvnFQ!Y0?K%@lMZa5aI_!(^Wq!s%o#;<7b>iwz
zLcibMf7I{uyU)}a4gl$XwY9LPL1;6En#@y@qJ19CveZ8XWS!AcFp8}NdLBMWE=p-3
zSsvYL*)~tnv+_a$^(*L?&)JV>u8iwfgVG=1N@5G>u{02<@8e9?zY7&Py@G+h1r;%h
zr>-oRcvX1{SrT#*R=t_xv@~u;cduVhM%RP%e}*^AwjRA6@ESwnblsT1BTpp>r2a!g
zMP>h}C&U*I8NqR(_&hm$b#;fii%}qT4EeTl+=t~ng6y(P%sMBVog#;VtA<XE^5xDX
zih;eGH4%445jrzo)eB=22jpTXVgf>qEvv5S)%JE{&`v5FQ^qN2ARAxXx31%+Z)9Zi
zf8ObEv3X3YY5iSC7l_7q8=e1XivpbH9HaXdg;SyA-R@ANdN_VjE?k>CKXTV%xl%4&
z3q|R4umrugiD~CkIR#zCTvmv`#)ZqB>ERetNA$B*UPtMbZbNP?j^T5q+v1D>)e{Ua
z3I(E}@D0DQG$`)N=&pcJ3kT+afL2*Af5K=f?uE(^oJn>oDY<zdd6=&~4Xg@7*Be-E
zRIwOZzMPbiozmV_@DZVC23D@8ywatqB5X(RDT^yKvB*q7{))Uccc#a(n*H8Ezgi$)
z-Vds=t4kO1ct2}|B1-6)K$!mA<@av|C8q<8i(66STUj03BMN4(V|>%kzRh2^e>&cR
z)K*w$cNzeaitV}H0LrUuw>Cjes;omiF&acvyL50r?5@Q!Em256lWJ2qx?M4hotw*l
zwV<EIiA$sdC7d5Id+05FJ1sH;%g#GEF*0`)X}N)ODCmS1v7w&wN+b@W(CU8l(i;_f
z`~Hp0A89QES%xIgr0#lpv)!G{f6WspuLq1ka8i#R1WbnOAN4olB|tV6`fR+^ukvBM
z)Diqdyab?y#xwC!NAXOrY$@OK%69acSDpY>6mpry3AS5tB~yhLxV5q~{*Fe6kL#ZC
z1c5tI^)12U9i|*78%?F-)Z2$}TFci1Fs*-`vm^KbXZbD91<rP$E(ix0LNhL2PcsZF
z#hDRdO7484_&Cy3<b>AL@VVx^CZw)*OWPQ8Vjhnp!h+HpNgjuTsFY9!8Zw&)+5MGs
z{C@#dK#G5tfi)}>0y#LBfsZv4m-2iry?@0@`{(;_C!L+u%6Lt_(Ta@qR_-=vY<|kq
z<tJ~iGCL28vqMtGhQ>Np;!l~K5$}w6D(18i4_N+ZKk?$}8uS)kmc0TTEJkZGKLgQH
zwi?@X_l#ZF&et@`s=viL2IWT4_}cj@fk@E+_`rxO@%KiBP>=X>jd21O#}2Kb=YNb9
zbXb{a`;vWt#DRpJ3O2$QZ?DyV-`-FEp6alefQlV7fBiVc2Mk787aBwrfSaFxH{Vac
z*H#Nls(Ye%0##dU%#;LH8RKO%uN;3MWYTbsq{)+HG28znT25Mojsv)m!dZ<BU*k2Q
zTi+uHxgFO%Vj8l^#C6$M9)eL5B!8w~A~kpZ$9wFfFH3>yDZfII5-p38fK+@EDPVri
z*twESZ4w6qg+7TY=agPqI2H(6o{S@M$@iE3>D<!dEH@JjtyNYuqnb;O?HU5;KoQOm
z=xKkvU>Rm5#5qo{3`9cp25I0#y6Yj1lUxV`x^eJ4=$?E@Z$-?T^7W_(D}RyFtb*ps
zgHMt(wxqP@Pm(jXL6LntqFLR}_p6&|l$UInI<g^dDD<F!Rfkv>u&;FZ(nW^CXjYf(
zxyjk?yYeDwNN8gP@{qhSk%C|!mbW=!NAf0)F3Oo(3}O2Pp6f8~$eUJPpsnVTgS(nE
z)(8Sr_sVH7!%1k3wiiLnxPPP>m=>`Jd0MW@+*0tuws;sT6^O0v@AwXqb0KEKv{y=J
z+hbkoCD;4I{j=S~X#=dIK|we@tcf=B(GLS<DkKTd^hD!5#(ZjGxD0i9Hf!BlCrnT}
zP(Az&A}~ss>H4w44Z+nzJx_f-Y$k~E4nQ?P7r!i7`w_ZnCeA4%{D1m&Gx1U#psxln
zcH}KSyzuSKk{rsyR}&+X_vtE`vLS)wM3{q{r1p7mGr69Xx37lpTq9BT88Kh0H4?}v
z_Irz@(s30^KPE3ula{x<aHdjdb>XBaG!oT9wOpN~ndLPGp#pnlDHcmap|-fP1^dX(
zp|Xs6bJtM4O)EPYEq}9wZ`r4PNO$oh7GL=eJf#kHVjif0I8YjO((-l#=F?G;hLl;Z
z7FLjxIvZzmNEOi%da_rs8q%^c#?NHP2K-9P27V}<2S+pf($425{vk^;MM;}lxMdm8
z_t$$Eck45#z%18#A+F{;1NJhi(nL;&pKiM}ZW*UTp_-^j(0{|eh$>ngKxPdpcz>ig
zOf5cKsMA9TVvq!rt!y8?ZaW6YBZ*Sw1HgeuRu+$0S=&-#_u$^I)@WD?=i0v>+hysA
zM-f4SRQ2j(dq4zO1SZgt5tv&1zqw3SA@L60YqiAjhc+@(rH?#h1}D5SVF~U+{s=GI
z;p}2|LYU9dWPj^(%gGHGqugk7w{vpyVK3HBZ$kPAG4<k;PHzCb<Q09==?x%Dsdh|c
zyJSl5cXF1@n6eMQU(Uog)}iuo;b14GMv){mxsDsRY*s%J*cu27Tfx1pD9XAzHQZHh
zSdLS&_<RJMt_u34*obFRGSmUU1;=-=AmOyuw#MR-V_q`vm4&_nquD7|{0p83Y9=0D
zy7M8(o5)mF50X%NhMuwcY^Fch?>`aTnh<Fu(_&fN`_BWVn0HI3ZYlNp4-YqQo1H{<
z_$q*9L43(n*2DQ`a@S6IlJBBP{{iv6myvoY6aqFmlhH#Im#0<;D3{iPE|z~#u$ZTR
zz5aHPg;>5<3Y8Jw>BTxN2xp=2BuA(^TN|Ohn`S3kH1VT-f_nGjVN)6wh4FTV))*nB
zG^Sh_?*-G2T!Ru?>lG&A*+fhv4^1*Mk*;&W2vN4~v5Z)jv?2Lx9Y^vNDbbGKCb~jL
zUxYm}w~0=KG>@^zH4XOIc}{<#BINb8>tVDh*j%AGWA14%PZ>Mm!pABjUQ?D?%meYG
zk36=mN!GjZ!h_k-`>rP04hHDF-l=kZQG?bBA7F*S7TzjHRSIF^UiK2d!un8wKwQy`
zs#_I`3|-cClH<9IyNmpwH^SkHpTvhQD^xpp%05mOsZkC7vt?1qI=6rA?6|Tu?5yBi
zEGyn$3Ma)Q*jnIu(zo%O+A>B&<}`|6Y+>&BVy`n6s|A3Ypl=X7$?%j#8RzQaAdw~c
z^7LROtW1yIoNr<bgHYDp%h&+@#*C#{rkh8o_iPc_$CIYz6{6VnMD}G+!bo4aI1e$C
z<8rdp>XWv`PPQnb<D!2TN)%3;d9@oN2O>#*ji3hgDI~CS$X@t80~~Iu`#>QlJ3Yfk
z1%>*w!hfy8$>}q5JNAggl?@B!38AqT3^C!S#HKnrO-mLjVd+GfElH}Zm(+8*YyiBb
zBmaa@4d6MiNy|}@QVoGAp@d|+gw!OK4cMHkE-ujCrQx+iLy>>_OTmaG^%m~*@v@;2
zHyUx~>J~$zBEzGdF(jIzW}l1fS=;MyIwxDBgYZh-nQTYsESFQ3VRn3=5tQ*MB}r4?
zkE`rh0n@0ouy8MlJ@5vFB)T`C5GdSwe=msx^cIcq7Bor<Qr6x};{Z)GGBr)Z&;{j~
zIHJGJa}xdjCc1yM_af+fEp&vAYZoe-O!w&?<*r|DquZ%ee%(p;0KK+}lRN*UJ!mQm
zb-34NN9YZk<*fE|G-@T#-D|TW^qNMG`e*4wFQn1;(l<a$^IYvN#w4ia*8A&oxs|uk
z4@mL*ZW|q-Q_YuofPn|RkB1-C2Fiz2wy}2gP;DJs;TnHqRmtCa*uLr2doym0(+oEj
zl|y}#*9u+vdoI7B%fQ#wSBi2vwn&9s6}YaFv3>KPTI&xbWi&|&H(DbP2>965FQI9q
z*O2LJ3g1^)6+>wkB)p8xU9N_Rx;1gN_hGu;2foREm@tb#X?9wX%XEaYo6SY!Bzj#6
z*Tm>c9*=*!W#wHdh7+YFwFxxd<3a^9U0xZr{6Q|Y9{sfMWXchh!M?40%G|7cU>|T-
z#SHCYJ7J|Twsywy2CVCVRBM=D!aBF2UNZYvrgU~BO*=(zhu+@M_2T_9Z*#cM(S-pi
z$b*z8F(_jRmn0{Un1!_Y=7OLqU1$XQ*5P4+ln#GkVtOvRg1pon#t?+{`qzlQ%6*qF
zrosd9<+rzgQKCN`FW~V4aXWi3MOB4$6LGOYQ;Yz^lTnYC+P<AJkcUUj)9u6b^#1km
z`(f5J&?a1wUOs<6#1|Tjur4gl2`9ty$A6YDhmSMXsm(6?GlC?4X{|A%U0Go$TcH7=
zjd_22D=5xa8cVqs(5ee!)sgZwMT=msXvxz&xwL=Dm4=Z|-M-fb0pwK^c6eUxl%x9*
zxRfb+Wvs)<r|L#)0g^&2z`P1xJ`C8Xy;hhp9@}U;V=gVOl?*Wm(;Y^fi{h{yA=JE%
zt%5PnwZ^$wzQGIc)x7c>%NHp?sDSe-$aa5$8=-wio~Vg~Fk>q6yoqy`Ie5DJUu9Va
zg`@y5t)fEBYu9ic+r(bG8B46vll)Yy!uPF8M(Ykk1yEbRcX%WvzMy8T0DNutuV{Zv
zX>x{|DK3Ewf)%ZK!HxLMF!`ezAf6aX=Z2niY1MLFSWk+^cCG8mW4k&hD{pfXqK6x4
zrzt}$?ayVn*R6QHord!qzgydKv!zUN`ck+1h%Uhb1himZzL&ubFBF#{$O03W7mo!g
zmk^9DoPR$5{r#^GvM|ff%Aq#GJ6x<oK{N}8uhI;oKD>F@ln;$S@21fbiBBYXj42-#
zpNZnGrkoZY!%PW2K0l(wNE4qQl?(A3NoFE-k{U@9hBSevG7E68)cP97#qY7zKjpVL
zCVe6eoqHNvsbk@DTVf(3>{?=7_=wUbULtQVhkp`l8HCa+i4(tZbmf^PAq}dCge2z+
z^W679U!~vXr^+}=%z8^2N_Kk&`;=!?JR?z(Z`8|M*5e!NoM|UsUdoJ#r-7f?SN2?@
zEAthLRBp*CAd~i1NV7^vuUK2v_94^N;GQ~(L&aW}M=~Bp#N<)&4Fj*hUZ<r+HvBZr
z)PFg{WFcAUz7$HQg1c9ynDtcYzckB?Jgh7a$usmTbhVz*3P8AZJrAQZpEj@Stm(G$
z7#pe2$MTyyM81+%goylcmbBKDE^EDLh`oo(cuUlgt;RaaR`9*$U8irhxbJ(BrH|IR
zmPR-~>oN19kJmLg;q-c>Ew)tIdnIXHMSqVO_%ox;6Y5+ijb+@px;m>qptQJSd}t}u
zt?1~+UZN2&C<19>7}u;luQbrF1yfB<_Dqa{>_>{UZ3XSE;6sm2+@VnucThAP?UH@%
zA)>t$#t&M}QsBzcH}R+yY2Hqnj~Rn`Eivpkx@y9rcasI$%ISTkaWw`c{Ed<F7=KfM
zLYZ(Ys)o^%_OnHx`Mv!-!VBY8)Hcy}q;Uv=d=#wcT!{;1g|vF4a#v(fT1acl%zR#f
z4dnwh;#$%&%J@zFC@qW)i=olMR4=DZL5nLnnklF{)>EbCD7u5M)Le_4IV!pf^1l|X
zjh}r;MfJerbJjm3oE#zCAo+HAD}Tbr(aAQ9B~w;e%A8BC#Mu#i%%5v)sz88(dB(U3
ztW;gjhcao@7Y#J(O-r_L*J&7~eQJDGuYq0s?Q+gCN6Bl=iCb9aY6o3^DG?3j%4RQJ
zw&CV^dakfU{6cZ621+Xwt^3?RVyyL#zA<ya(YZ)sTmD;f61SPH8p4gEwSUDvt*nkN
zFHKqGIbTGzk)k!WK`mtxC-x*Ge)78O>HzG${Dj<{b;-B~@)4kK#r-h)EZ_IQ{WxR9
zy>`&oI#BeU>w*sV>Hzn7)T*+~v?f$kkGB)ii9@R8uhq@5P08iG9_j}r2<kV_I@IUe
z{31xgp1(K0F`k~|IXGcpt$!)k@)FJj(XFcQ6&B-;jPZlyaJ>2q`A~}C)5v?*L?oR8
zGtW|@@Ej-6)6?{u57~{?5Xt3vF$0G+awDFKXDN_AUK_pqK}&eu`rE;xrUKkhDmk)n
zE|#6cLF${c9bTq*nTt{u1ZCbMJM!&r@+wQIvsTjtcl5ona*)~+D}UNpZ|P}+RGHDu
zUw-?%#)CD&g6nHdt~1l@w#Jw?Yq0Tae9Uj3-XDHnO`&6=Nviq!^CA9#Kr_0qctuPt
zfBbX#_V8sU+V77h<w|R#3j4gzl2ZMGs0BItyP_%i<BX=_l8d8hE`cO-W<gC!C+Z6}
zF9|VnY5ye?4K2NCY=00yB2CnDP@7Ca8k8h|LT08qv?O!}xv4<s$nY@FqK;GKjo`ha
ztm2k6+L>}$St}h<;FZ3nCRaGbt90ruh4*=~d#f~{u*%P~s9`7SIu(wTs&l$Ra8@%#
zVNm8*Np4i)bDBj54k#B%4Hn5EyP1ml#Ct_j5-Rq9q^a%(Yk%uvd@H+Vc7KHTLdoX)
z%>IbtZ8R+!g*C>?6IVrbr<VPukH5f*`(nQ>XLqi`)i|OmJO_B1>Iu`GQ+%S3?$ciR
zYkxM-?Sz@yBXoO#L2|vp`4Icl6aUKcRB3Tsta;akG=Q?N#d*Kp^So&<?v(dKlIGno
zzprnJld`+_dVe_=b^KeI=hL$`=c}f)>fB1<`kG|AY$J3<Ns1$H42uTO3Zb?gx-z{Y
zXsfUZ)YI|y2r7Zj%@iUZhg-nYu?LR*&nZ1`x?ZPrg(x&7>l3H<)jD$@drl65dknfg
zt6v>V+P<B*-bubSxpfHDMX6JNh*S?O+;Gs-ZQSz$%750Y2S0vui)!BP7&81bC8p9=
zNaAzicf@R*31mgDnemK8x*py8xIxw4)Aw$RCs}JVx7xJrvBgjs0Ul=Ab}XiX7Ru_4
zdg*Re5mI>!^F+6+>$1`ZP>?^JXC2}|28(NE;N`~UA)h1OP=)L&gVpYg(E5s^q$DQ0
z*C-SQfq&}$HRn)p?n=c>XbLE*1(c6jq=?ygmm}_I{5j5Y&9cYb4it2RfSZ_md8?b|
zlxUcA{IOdVbWpgmGz7w*vpAQAN{Ik?TpP~M3u;ATmI{(zy76%2tgg=86_PqUOnl!t
zb&rRb5T-fqQ3b~2TG@&44Hb+EHl2$9QZ2+odw)f>>Nv@j^$SH|nrjw0769XVtYr-6
zg6m_!YZRny5L7qM3K~y6=W{9TL3lL_Tw>>K1qxU#yc@F{S=QO{;rr%cUG3}I^#^pv
zmp9HFK{Ms`1)bR!bcaJ*7X-$OBwL4IM~4C3<(|_j-@81l#4;s#KkNMp-%nkHCmT-`
z3V&qCW6vh`ntX+h?)vwIaPfIb;2^x{;LY;CB{oA$daR|8Ld^)|TL`@mAO7?G+h4x^
zgJR6`{l9SpdHZ-Y75wruzgl9DJf2y8<aZ|pu}iQLf$}cmUr|vOR~(P44PTcBnn$2k
zs_XV6wo-WU(sX@Oi1A3sXZo$Ba*zmeD}PV-qi<xU+6+me%0+yORLa=+O5=G{e2t&}
zox9Sge=QH+mWMA@n?cjR=*%2Bmknjoq~J943aw4?E&{s3J$0~g&v&!27qhe}lGc(}
zM4*5ioQuXNbgQJ?(n-rrM2abJ6F=wn9qA`Qvp&j>lQ8iP0p-p%;t`5Uvs&BtuYW(w
zu<tbJbN9?-9zj##MwA`_qcuXLl~ReF%L}DD_vS$HQfgdm7Z+B?NqJ`Wrv2)gh!P85
zmX5_F>SL4j%cbfeBbf;2v+9|nLsLsY;m}vD>>c(N0Vn|4b_7+AU)}COTRB`EJhD?y
z7bm=>_k6(;_=06Sx^;D~Ro+*rFn@1y;Dn>na1302EAiZfZi~&<U23MgQ_Kx9p`+f9
zt($jt;Iit9C8|3IzKevXZuh2+CA|!lF56C0nkDscNI68*TpkY5HF6;3RxFchii>>c
zIbXfdDFYfyA#P3mPvbR_NGonFBE$_ut*6VKY|ormx$dylc9(iAbGXg-bbo>B&m6O9
zr$sQh(k^Q?v|XwZB~=F7Um2J#^ybULTj?D|F=3=x&AWe%gmtdV@uX=xczgkM?|{4@
zsJM#hHhS7M2Dfp)W=j5WxUqRQdEC0~JnS&Ho7bx(a@@bzcgW58ZTREnT(dv^_cHb>
zDbr)inU3O0>20YDvF)!t`VYcjTLW8vp+S=TAFgRJzL&ubFBF#%UP2U?&Xz900%A_L
zk(Vw72n3q>LHL(}H7pYXIWw0rRt+VWzj`wZ0X3KLW+f<pQEwc#5q{UNAkdfe!CZ1U
z<P3{IfD_kg3I#PR2WTq@#p!%al*&5cbPAmQ`h3H?z0>)mXr+^F`N7ug?kqVo9Def+
z_e!Pef-5xzm(>17r|p_tBr+knSl9%16pG7C!4Wsz?5GA8Yc4LOrVK%NFKblCvEU9B
z$5|jSx!gv7W?ZQ~O4Ug)frC?AKwuYxM3b_MG1#<RAaRo-BrSWz=8LIWS70WS8N(Hl
zWJnn%Nw7IH-mr-w5sY6vO^#^sR2`+%qzb7$E#_Mk%uymFJc#Yx9L%Z=hchQgf_p{G
zX2B?;ZDa&Uv$=?PrBK^=*}+h=uOKum=0_2iN>H1BLoHnGjf(J#Sqc>nEi5pILyL<b
zh(XN)5yX|SBo!e<gl!H4j29lk3{K6yN{*Q;=8%8|2u-g#7+@n~9oP(q6kU$dMpmp3
z95yMroEA^T<r3RHg@n+Rgayx~`KHxnNwSC7WJoO<_EgEns~A~}*`n-nZ5C7v8nBEK
z0v<$v%Y<4@YVoTGPKq|K4bQoa0&|O|J-mT1DeQxU2MYTjS-@hO*M=%KawPnROGw2D
zVIt_^HOiK52&|-PDhQZlWG5sHC0T6_sAz4YASEr%kk~1f)ufCxZTFCvno7He?ZP%x
zAl$RukbsS%g>AMdP9-!wb*jncfVpZ*hBtzLtFb*rz^Y{;Byu$`o)(97YEK1HGJlas
zScza#by#Eb0~@zm!a+i|BTQxQ>#w^PN7#CTo&CD|=fD1q`^9Uqctm%8b#`{Vzkj2a
zKnt<g>}_2JXiD@FhiK+pd_j&4&}8z+`vICv;*$&+lVkE>kVjP2hcaYL(&R;lj2VM}
zdduXyH7g^W*A-p&?euIqJG>Z;CoY-3>t2oK^U3Vo(c<#2@8{ng&PVf!i;EjCr{{A_
z@iG~jV)NM~FAb+NGQC}3*GJ+u-2Q&|YBn7oPUc75s~0c3H<S1CrHyZXyO?0IpC+*3
zd_Fm!UmDD|x4PGp%jwl@Jh@!pvfcWBVe<BL^v(4BQLuty+py9x-k;%3q!j!&t>@?G
z)5Sotn11i5RRXiTRRRddtJ#}&cX)L&x2uEG^Z&W-o9XQBWY%U0$KBt$@4Ii0;^N}<
zWIR72LQWb+MD~h20;G6CVJgI)lN@)?9fm*bzMH<8b}!sl1waAZqO7Pocp1BY{m=XR
z*Fw;YxOrX#)y3t!D3>01h^!>9*qSkuuh_9n)dMz;_cr|Kp09=<&rhf4FzCnE-`l^h
zexA=SF8|VXKb_8hzB=*a>AUWit9KXA-cEk){(kktEC0(`cXBp8>B?lRCCUjWAe^4S
z_R)Aep<3U*9fuR8oKueD7`n@U*|>W*8qcPec=t~Sto-aC2Z5V=>vUvVrK17RHd0(u
zSnsRgd4OK3*`ua>jKsc{%iT+q^cWkX*-ZKSW_gyaEJy8sK$aI;XF2y-o<lFhUZ7q`
zH*A36xtD>6jFsziFQ<gR^n`LRmh|j^0<UxLA%22Cc|(xi&s5+y>*w}=b9ynEjpozY
z!c}UnO1jC_`(m212h#x)ho9gaiPcA7S`_hE-VY%S%o%$;fx9JUkkg!~+<<BKG$NJg
zGYC{~1jp<_BX2W;f?~^un6iL6g6N3sZUa^Ew&@a}+^B2ua1vz9qZ)5nkW-LUO2z+R
z@*rIlR~8sh<uZgIZD=ijDT<YrZQ^m0W(XcV%K%MV(gcD$K+`sLil34tY#{4>!Bn;>
zgFJ{bh_1mj^Aa)3O31xXdO><Yd!g<GuO(+MXvTWjDTO?Ve16!OpT0Z&|77-`Qv`kX
zUt~c8pd|I#0UF0g0HEe^u>ym9MiWZdFSZ65<VK1<aP!}TR8{<cMZ>uwRkec9Kh?kk
zO#uCNl<^qYkB%fpX*Eu;?Hde(S0unLdhJgH{-DR+TG`aLKs(j&Xo%4T!-%}S{@CJ)
zf)J`S8st={?ruXg&xHYi>I;CHDEmMSIZ+KhjSWQT57EP~Y^c$*o%h^gL`3@Mk`#+V
zg-eQs!$uL3yJ)w6Y}4t&sDCO$G|9}W12kEKPhrUHj5ViBADzr^6r&yLkCv_avz+gq
z^`~AV^U`y8=~=z@ceH2q+OvAyjnx+c=%sblk=j!2iGp82%_+Ol`;7dyv$RzklY_?w
z`rIBb-!yz@GJL<;&wH0vAPQ)7s*FudjzDTP+SxDrlz5we)l#7ZfG?=38Ut9)WAN%&
zF)6Q&5JkgIw;B<l5Od7c&ipwE<)DvXg~m2(Z$$#nPy7c)u*MpJOIie|lu7eWEF+@S
zwoeMnyrYl}+&_2T{W_YRj!w=deFq&PlCRJQ=Qo;%bTb=T%Nl`ZI&2@!-lWMHSu#Wu
zt(Uj21OqgGk?u@GG|@+a2o2Jd=#}}4sZh0URj4ZWuTbr?%Bn)uw5&e0?~rQWAuV3p
z?U>c}roeFIsaH(FlH0+#(Nk>3n-Bxrno6$d9oWRMJ76%_S1fBIFnBu{xXlP9;t1}7
z<d9N0CZ(MKBN3w=()Br%rM=WAbl#TK>T%SE735@p%)=5pP-Fj`;jeuWj%a-A!x~@g
zIe+>N-kqI}{&6v1`MZf`Rqvq1*A!L<BQg5I0~D#EdrkU<97kIl{Q*9-504MPhgf=i
zXdgw*jrL7+DM&A9FVv+FAK}AH-@aL=A;CVNlh-8Z<%7?eKMi7FI80hrMkH&c09&^j
zfqDmjS^zFDjBYy@CEo2Mqp;86+DQg#6;|^JL54_rcA%j?2DeSs3uq-Fx{T6Iz>R`z
zhD9FT`}ViEXdmz^gX3fb-EexRU}MoR&$^KyTXo_m6l}FlLi5h}%syCht?)VdQu0T`
z=k_y;JK5$3T1)H&3fOOL257Q)`+dZhg3^?Kw@{kW-6$;~rA?GhyhiERV{+^<Irf-L
zp%-E=&?9_4M-L=qKOaI{YQN{I)!TUqXRM0q11Z%*OGUus9M`1+ZK~AtX^dEvilq^&
zQgMqBk6tRQ4JKND<>X-KrBCUX1eENwhju&ep=>AH>(45L5=lv(ke^ru&}c&W*v->G
zNT_?f?iV5z^eIoNUmy+{o^lMEcu)L?e0)P;#m5z07^NYa1%zC9fF^N9ZqN`-^kS?-
zX1BkxJ}&qj6u%oAm%$A$6qmuiEfkYoCn%Rtgfj$}hp8@=0sglgsxJHie+amdzfD0a
zUsj9YbhG7;EF}?2(K>pC=a|imE%@DFb{!gMkf6btJzZ_xn;1Tfdfy8l9a;`11=_{%
zA9zyWoNyGwUY!&er9=p?$4NHBs_J73kyO)o(Ik0PH)%l5nOtNbIke&~_3@|*WI(SS
zr@RwYAx11?`HYwWR(D?$f1f0{&Db6Jr2jv27D=ybHn5NukfIz8s>j~dJ}&J@`5tM<
z8CbmOA`F!O(aqAQ_n&(jN>I3!lBfce6nF>a)!)aH4`l?2vJg(LCmqo&O7YPN`PUis
zX_w>vPO0uY0l`<#FlWh-lW4Po8+qBriv$64ph^t6y3hkyp<ophf2X>K6-&XGq8Xc^
z9>6L2G7G>dZ5Re98-&HO=0{h7U{+|v$&LaES^|vjgRK567F5M@vixrs8P;JTgBF%T
zI$QPCrHxdSAvwMZnl7omvej<OuvzYuLW-{@*$RnqBeMu;h2hDbJSqmih1PVKcbHjA
zhPKKxO7Rx#a(I+Tf2bYdo5CrtHV&9vQA$ftWt3iQa!^4i1is4XlkZXw7%P+toRax0
zl0U&?@h+T4U1HniKK@E{;+Ekd;gNb!>t%eul)A!~vc{yMEmn#ql{+lz^~qi<%P#v_
zaPY!WfxuSZi2a-CDI(?enp25YyRxZtJe5+caI4Q%h8n8Ue~Pv}y%C(dk+q;TeluO<
zRBIT_lq#QTL@j_zkj(W?F1DVqKwbaTRrj_O(wWZZhcb2DOO?;Hen;i1K4qot0!s=g
zWoY<3Qb?n<mD5XIr#?-*K>M%j+1!_gx0$C=(=~G_EeINAdKrgR`zAzrVvC(r37cij
zMA`N8Y6{e{e?dJLOGT4g`$Tz1J0BOCUFJNS)xER=EpQ64=S#2Nvj<_4E%4ejXj?ZZ
z_kc*7_&y`L)i*F>v^HjD?Z$mm==nLH*5`Fh0r@;^dwWD|#3fWrB<vz@809DZ{wjQ?
zwfp@O>I`@cWuUPr)?gG49{_4*cN8b5#h}hO7Znj;f90*-7rmLGiaLU?jxwF|DFw{z
z{x_7m>GW^|MCDEG&2hS`=9f-@J?K&hUOu|3R9w`K{{BV-PZ^!Ie7)3;!EDRelEFJ&
znp)S*ZZLCR1iz42UaB<Ba!EeVol<@R9o9JLn_gDJY?-f1;K<{U=tAl4B)n7?=ekv(
z@pl7he`SSBowF`M?_%6B7qf;R6^xj@M+HM%wzQ?bUn%9?$&VD(Xk?lO2q??DmJ!_3
zV+AU~J!L9{c}mY}o9QggdZnk0(B84p352w9c42HN6SBx>9xU<7nFp0K^s8>tlvDH@
zX#>U@Nm9=_NV!gK{vU6z_cPo$a;5cuPYyRWf9iETZt!cFPljQ$J8SyySwkx-eZ5&&
z<5-qU;!`!XKts)yl3cTPO@C05-k&<|`|QVUo;x+PDk?@=H~uDChtY@7I?j9+ts6!1
zc9UDC!)L}t?{153_38nX>u`4MN2~_N{C2f5)=pe|gxQf|J%}kd@13z5c9|Y1`K3dh
ze^2$?Dw_NPO?AHvT%5)x{}z&vvQO{sMba?3B5Bx(gH~85dtQ6%M~+Y1GcX?H`g+NJ
zL#uVLK;^)%95~h|%M*k8=G3K(OWs45Uao1;cL~+~F%*Lp&dqVGoU$iAWLG`;#}0~6
zq5#_4?4ZbFY}4$Y6XhOGZOY9Zy87g2e>pqa=<bw`8O9S-J2&adSzb$*NKS-9ohU}}
z6(>L*(zzp}ywz(CGIdti=(Q7`m}JLhGt_R`ZI#~CCq+8c_JBSmS)9m&nxM5>hbKSp
z(D}IP!P9l#?q<|B&p95rA%k96vI$!O%Bk_<F4h6c<bSP)1?B4|PU^iMHp84Bf4=u~
zd+kG1T@a3PCr(wbOXHOYZXy4x(XD82b$P}Myv0cIHHX*vW)JQiUf+Q96Q?;bV30>=
zf<4gMSD=J5$SEb>kAqZ26M|XrC088Mk@ML=CFa76Z9}goOR*dv`z$28bVoL{pXNXL
zUQYj{4-XZb5QOSV&b0&Xc6iZWe~^op2vZ7yb9*vf56uS@wUA^D-u#rdV@=(CR{)y4
z#^T86fK>71o8kaUa#PTGr_8i>&{&wgXL(YL%HP|)o31m9FTWT?j|UHX`)1VPJj#3e
zT3ojQ+KI5AM%{6}26X5zC;xx%*MOo3`b%z{Fv_n!@hs<nY?W8s-<3|Ne}J6w#svaE
zM^RP%g8NB~@*OvJ*o<fHCzTk#kQOKybw}{8xo^U#-f`cA&3NLz38S=s--Hc*;Jyh+
zBU~{5<NGFT)n9bqgst+6@0;+|UVq<&t?=A^6Gm*~z6sgOtKqz(%N5SuHug2wm^bat
z#RvXPA@9FKja{e;Wo~41m&c?94FWeXmoc6J6ap|gm*HI+CYM&TE|&pxm+rGJDSzyt
zez`=m$hUXOM+w~742}+;GdOw@gLt7I^yo15YtHw3koT57`3$k_&i(#O)}cRAz(Nn&
z1*U6R9}t5=B3TkfvcnTYb4P#zd62|hCV3CZn`BW4DuiY1X4rK;zUHU%btO|+87s_X
zM#Bhywl`E-@(LV`JyqTc#@Q*>4S&8dU7C3?vz=#NsT8xAM+#V@BfnXOjiq={Izp8h
zGViS%Ds5R=*`C9M$KJZ9Jn~AFil0CU_3d?Di>LIOcX81oD;J2$V;t-~LpU)8&6>It
z%8J*h<g?@C$1mc4?;kJ!HbQa`RWhOb^PiXe0uA{=NE8PMtB&1|zq;=)zkf-v?3!9>
zI@AAgy8H2WCsw|3tVx<U)ZRNc=g=!h%zJ<ma{!S*ZojIYzV`B}PLw$XGm8oA#VthV
zm9<GiEOG+2fC8Cig_+L!r;tBe`8J6@(G-XRGU(-4m9(+}<!;;1K?|fyO<bCHrs=v!
zugX)Co-pi?rzV{fcw;-05f!Ynm2ZDH$b60!a*Ru(XI0E4L{99rksR3_G(2P}xh3iC
z*Cdd^dyf)>vDqW(7lmYYK*pRyOau9{YA4{LLUi(kJMHJndzGBM0rxJbA*n#YK|;l4
zZlYBwH2jl}UcH9HB%b3e_wAvG!YkRMZ>;Tp9&Pp-<60is9{eJ&Wbulm)9`=#SWk8&
ze+}t@kx@W=Oo7uEHk|a^0%ya0&r?x3md+@uy0RfENL~f1ousHTVQ`aZtWFA3BW#yG
zO`8c;S?4?F$!JIKu0jf27r0W3-9jp8OCdQ(fQ?(RHqxljVm6UAksc!X!K7SJ#tf}O
zEpadim!#-4{wbJdD_be1x&wcDWoT=4dQs51_r5_X1=hI+p3;gWmIAPp0=FCV^m+(S
zZ=+G;sjgD6<9|(yx4u}Uw_kU2IdEkiu}aFMC#Ahf@b9UPLANvz1KH*4x?K!XYt<SV
zfsM3GHW__R@ks;Ixb$Y!$ZBy)QlOV<W{EJGHTovevo)<Yo*106`lf#lY-ab>nf!K}
zSFw=orqO1B18XkOEX(w6%qxv`atbBlw~Md+2Bf32hpldH=oog__(ZZwBoI;KQ*oLM
zUMca#vy}7OUWyPD6*l*4a4lc#FN#RpHi4wbd~N)8I6<OGa(^CF3{fpk%2~f$(o+Gz
zac2nieoWl=6g|1fi0OX@KC<W55DtzD`RU3=Jg<3Ba(^D$^Sve3YR2V~OJe#&5tm~r
zU?U;iEZ)TKAtx?5!m3}hkCC0J1Ve7J?udAhiT>D+4k&&Drg3nQH4`IbuILwS%^c0@
zRz#*&!n)AVp?L^*?xUJrw8(xyx8#Z7t9|qIN(A$jsg*+tvyy+7>LaC?^b<dF<H!f!
zs%)ks7jXKp8g=Vi9to|UW>2j{DPDC)sAD<w^I;T2T8v*anqaK4<baJ~RNDG8%bZmx
zX&CGA+Z+v%^_Sn|CpW&06~=6~W9%I<J<QH*q()OBYRqgU=-c(gdDW&2MUp5H4}ikl
zo(5gS;@8u$oS%Ql+mDUZFlY=lCChqnbrd+UAra~8FtJ*9!NKF{(`Kb&V;;a^qtW$@
zvL-k*<b_mc3P@Z`G#2FPX@_dm)-h+C4n_-BmSN36kFYHd$^|27a4RtjVKRP=i&<*j
zm?#5Kd>5F80_*XWvdz8A1M6h+Hsx@+$}$uApeQA4QxAXaC)H;{Kd8prYiO&1x-q!9
zhTct8pK+E5TTv=R5{aPeCzaTF{sxrleol!E0jN|<k5~TD9L5UVzE|1YJ!BPxKDc59
z8<YuFVksAl@UUu?Qn*UiH(0hux>7||4PoN>c`}weAwBd9pYM6<fb_cLWi>AO+u8Me
zs(J>hDvW<0vySB)IWy`p0}c7Sjv;-0A1Bq;Qob%wH?Jl7xe>#kT<%p3mQy%)w#f;U
zSAmY5Ee~&p3&WX14rkZnvE&7({PJZZ*C2zG{MdDb!}OYX5+p9uE6-Z<k5y0{YL|`b
zXQ=~zULFO?d%^C>q`PCep|-`hH=@2ofz42UL+F1tDbB8X{<oo95c;IhEwDT(bW1GT
zp<AlxXNT_Vl9!=d^0zm1-$85iN{8@dE3pYFP&5*;{8>%;_95PMbAtJMCf!(dxiU!k
z5}<DrS(me~Yx-cvR|LcUPb51^bjlleMzY&ZAwf2)$PLwbo*u>)c=AovEWzE8+^&3y
zw3dImg`QX2<zT3xvr@K@Hai9z$cjN8dSdo6n8RLtX7(apETGI@B%Yi#Kb^h6DVy9=
zv)6nqfy7my2xm|sMe^!1_vg2IZ7Co=IC5o^iWFxAmIvoI69O5oX|Sw(l|(U{tXyoq
zlHL}Tp{>FY9{G7y71tc{M(V3~>*B<nsZW1$NEM(nQACkl43VQpE8slZ#@LeQNbX3E
zRhDx=E24>pB0(DGgZGMN4(Cat5yajvnmHi%zMpG$Rl)XU-N1Wq6&wt6_WcA<<r7=E
zo8?lNKQx(d25yv_R@jf1NcF<2R8;bkqE8}$^mx;qn|xBf`ws*LZsZDOZe(+Ga%Gpv
z4KECrB8fs20yQ+3;Wh*&mk|Cg1eephE}MVz>0hru93%-_CDTTLi<9$KATO58n3MAO
z1f;ac8IUvhBvop)mgTco`tJC9Y6X<f#@YvFJ{G05D4`9BhWP=Ubedak#o0n2uGp@r
z-Oi)OG6GaBHS#$6rdTE#+m7>C%9;D7J|>Nrwe{{A)S5kw0bvb{TGNrJ*1f=3DU^TH
zBNd8NWm9s0T+kv0x3HGm9~Zpv$}GlJ#Uli&+k{C90}gph;iA;jFX$)hUw-_sDPQ#_
za6(Dok$|ld$QrI{nEg?T=+d9*F}}UL9)4apECeHr_tU3uhxEZJ5YFSsQh1c<mw!*6
z4<ARa?Ly5Hkj?@06L10`OwJf#L#%%jFym6dzcslPxmo@e$yxQ5+n*HaLtgUy>%bPL
z<<X<G7BEK$^FoX@*~((~`d_wD*Yabtve5>{#F3+VPQO9-s81<m;-R5%?OL+Mqg+*(
zvm#hOj*_F19!9w>`=(jE7fz|gC~`ff1@t~WcII%9Y*C#N|CT9)$#r93xQ2h^;lx2w
z$na$4>*GdYb@iNAj51H%q~$kKXbz%^p6qEUB`oAjvs|B{L{npE=mW_^GqM;5%*g1i
zy^MW%5O*1|d9-a$B6aqh;N|g&vri5)ziyk7{BoFtq=f$TxCk+bu%djmI=q%fC=CBY
z?*u+_tQylLy&Fd!hJUBHHF|#%MN9}$IIzrB@@&=u@Fq;+*bZIv{X2t^tT8<U5#s8X
z&?M&u+l)?n&r*j+0L&~wSV)i~+@<9R)qcCs0{LU(Quq7N!b(7p;76)I5n2F3F981l
zw4k7}3CcmD=UR6&*Kc@E{&t?Nl!MV_SJynUjVc8`UWm_S8aW`Cq{M$eDog<ztvu^i
z)G7)Y26@icx+D<Sf_YC$Rs-ZAXRRiG5XV{coES$Khp9U%yAR{W=N_yx=-e23QE;K1
ze_K;0ESHd4*7kRTHLWeS#8JREPNUi+wT5w&6|lCYNNts{+K(IvLmiz|Z-bMx#{$dg
zJM>X0upWqBE#~$pPWpc%dCQ^|mDIQLmQ`_x)TJf8&2T1X9A~&dTt*nCYBPo(G6g@+
zwYQ~=bHx@dZ|0?(?~m5P-NkiBVQG9{N2BFspGoJ7>Z#RFrwvzmu;y&5PupaIG`qYS
z=dVD#J1G|W3L_0X4QPift+YVo%a&dZ7>({{tT5!}8@kVh;`M*A&5Z?BV}k|}?b?9x
zP*~A+!?hx%IzuzyCpWLY2C-#XceJvyE-AfW3K>mdu$0kKr3~`(giH0ql-DaRVG}W@
z{Whg$B*KjAFfL%TRVt?Nb9nc^FQ5PM=_e(S;MYWf2?#?d!s_W;dBwUz2Jij!mEMD}
zCZT>MB4Q((_8Wh&Fu@K?SQ|b~2OJ2yV{D&Wzou3re2~!e&1o-KZRP=*CYB*{<W?T<
zYum_D%|qllRWZ_A;zG3P73rBOy{2FL&Rvn}Khxp!bof~H84dnR*~d$_rcN9bt1Z>4
zl$%TtDteY4JGf-XwZ?9m)TVK^&YHXkBdRY>b*MU1ZfSpJ%T36NIdBs{$M%ic4-+l=
zph8tZ%Ho)cGNw_a(xTS+?@!-|><bcoXr6`5QJE+da!T8{0y8$U%J2p)x|V;aH+AOp
zb3U0Ggkg~oN?0r7a)N7@SI-j#_&7E4<#`_GiGI9SH)tX6%^kWK6MUl+usE!(O4&%t
zJOPJ7>ezqo8RT&DRDzribx(N{-KJjQeM?i{IO#I2pOXmnYIdt>EGiOf@K;>6v~<vh
zPb$yTDFChlf^4a%EgXR*4VC&l^M;uxm5M}FN~I%*aB8IllY%ZLzlM^7G|g+LkS0z;
z;vl*qaga_NmvE<AgmbZ7b$nf>W41AQ1mpTfch-M%GW6XSf_WN3&<TE~1Yy7wSh{H4
zD@=VvNbD9eyxEp<)@cvUH_ABg5H!OmNtM9OzREs0Y!$7o|9P>k1b}%pmjLkJsqaNO
z`q(7ty*S1xT?0n?8pzU-7+eSmUYkW@mkZU{ZqKK_@osN6OS>DgkLV3n>c&7T2^%mW
zaB_bt)#8<kkSfYq-<Nq5?Gtu#HJmILilh1IIT$tjTEkA&y%_Xy{5ezf$8`5#|FxsI
z%^gi;Km_#7RQpNIX#65!y|=}hGWemlH1Xv9`&)Z+OFOq1c^wVcWb5aev-)hehu2gZ
zTVY0ftT+@bq<=|OclH2zSCFVWb6DO7cdviWy59Nl`MzV;|G(Y>Vhf|k8ORIkjVu0p
z^#%$nKe$(KUGKcle|rV|4SdZD02rl*E9(rdcSTRF9i~sefL{G9L#g;mh29<OH;JpJ
z#5sO0VsYTzPFr{E_s;;B`h|cp=Lpwnwb#5Fr6Z}zeCC|6xLPpOxPL*;sBv>)7a4!B
z#H#?okSxHm7`0aS_pPiMqrWGY8QE&L=Nu~6$h$dZN2l~#obnWE)_10?S<p4!)5h~U
zVwH^jEDrfquHlRZjys!6EtZ|^w=yHO1Df;dERrwxPMNmXxEoHSy0yM?N>?neA{Ik-
zzSnT>R->EuTWVq8x=6l$x^!WkHN}5WcLulTAG~6jciK$PJ6KQMg&&G|c^tWrr(dUQ
zTlCjf;^w?eHuitI9er#6r#6~2`UrfHXIX}9tj1J8NadPKL^&E&>=b3=7hU8Jv5iBG
z=eixw&AYR*JsHfrV!m8<Enc5n&oli@qoEv3m6k{s(lYOJa~jw6F#2TkRkVL@bbTIj
znVYd!)9ZO0uiMYex=B(y(!Fg;u9-{P=})V*^iLj^uf&=8DW9r+HqSb-5^~ry*ozRr
zHdPQh*pbS}-s=-G$WhCe;=bEhfMVU6uR=ZrIim9vB){ra3GcXi`^EZZJln2#j)KtO
z5h}$;jIp?Q2g59ESo$3!5!-)P1*Y*#z5w^UQPb7T-bR6LpIp|6yng{t=ToIYH#aPZ
zMWg^f5}82vE}Q!v-*Mbs2XPVIBrME!A%k$~>gW(mmme>n?EV}Gv7iaG+&Ca`SY#T$
z@R9lOQsCvF=<oF1)9*YSLcaaMzOqp$9p3O@KXKOcerUbS!Iw02zr<vU6Ww)iucQ@0
z>)n(4M@HL#(O9~qZIFJnzvx^Z{Sch`v(M$}@{l)JCp>5<LJD2q@3=vQj*3-XFTmb)
zJi2n#6YQQc53^}^v+`_SN<A&Fe148us@$~NJBuElXvE(Ez3*g~5nLY=0yi+1p};d0
z0W_EKGX*GrOK;>h48G@A_z$dBq^K7V3^3!(EYMpQxfMMX-JbT){r`tj5+&R6BQw)%
z4vrIB5-C2ukD{Z8e+E5#eY(6peZCqRyb2aP4zJ$_;1`;Ig64<L4jw=`1H<c&;g4sn
z(eZWRi?lX|FVDXRzT49$1A#|?<%{0}oIZJiG>6B37%X#*D{~Bw!jbR5_j7vA^P_e;
ztq^6E%jwV8-=4H6+cZO?tc!#5q<F2WnK38ue7E+5nC>jm@=C-~h$RwhX;}&x%WsJ_
zw2=OC&t;6QW30UOvWBgr+KQn*&GOQ?^n|TJ*Cr=v{`neU%eH7qS?l{9Xtpc4A}6gv
zQLeFnJkR&EIuxn!rCiCb`g;20bh38VkYpbyzL1e@;iHkZ8#&4@wj+cvC|K_z+p<|U
zTQ%(~v?o;Tl_V#>%<Wxv+D4j5KkMgSpt!Mq(sdOh4wVjAb-P8j$x+1XL0<n-^r%r)
zyj!lkQ!M`AXwR{#czF{Bt&DZS@!ds@!%vQX;O#fxzV1dz+XIm@6xtpnLTrsy*LGhs
z<?}V_qnvgSSeus9%?jrd2U&ujWx?AK-B#9GL9h)izLeQ)z>F3EcS_sArT_|E=%k&m
z&3hP6);!BvAX(*iJQiItTe^Mi{(5<R`gavl))_^XhQEG1r56NNIUff^6@iAIe-Gb(
zpMLLG6~FZN48Pf++Bs{N&UXwEfu%*|&7H0)P{Kn?vIwxzRW?j4qW#FBxp?wz2DOot
z!<7~VBsRerwNb!e(9K8?U!kjktiYOC>Di!X+l;$mQ9WcxVQfLj)I?NMx7x^_GFcU%
zq7_h)5DCb<Y~PxA)0}geW5-G>8>>`*Kxf!(pxX$fji_W}1oo0l-^gm5OQlg(@|Do2
zlGC}4v!b04Sb?;~_)IoqNy#KF-4ivC+qOpXQiIA%A@r=RFGOBRsd-n(Mc}FGw2sL+
zA$z%MbXf70ZxHtuSX$0Qj*jN~DP7D?s?_IIXb#cyG5&B`kawLhJ^oS}u$ps!QM-4f
z0rEiw6BgJqBS*sG#Z13gd{>`k?_A&ZN|odoSqR^xg-4}ty9pCLc;89?Aet~fs=zVQ
z(S&o#+jnOP!l+{SpPo#Zuq>j8x5C;IM8&ph1ztBRTaIMaa6iX*uN*)x<$z-#I0e{G
z*v?~K<YjwPIw<Yudn#Bd`^X@FasOVJ^?=iPx%kmm>gE>{a$ckZ)A#^mGqQN<HGpmD
zLhsAL*!dSXZPSXhd$dCWMd)wS4vAAHz(d;IuUSMzD@*82#n?TZZSnn_X238i;B98%
zQ?}B~10Uuz<46T<Z`6)6fNPu$?UHR0A16EI?tuVP{yj)n0GIDmVIik~+pQcn(JO#f
zWYUY2Wc5sYerUfniM!PS>a_z(kg2Yl)oU(*Z4!i8oJxoPQSP!)l?Q)oxhv;es{NWJ
z8P}&<g_^wS%HmX}%9?R{TWIFE3qpahk9z%YpqE|34?#tSmwv@@zVmr$TXr;oznuA(
zFtVq=Aym$8>c0a>dwLsx4@$#IZ;7;(zYQNXgK4=nN<T!e`Yrqf%msIC$s_#p>`gr&
zXg68y(1YL^x2V?$gn8|kV{aKy(YStgK7?RsUAeZibB&8~{K-d$qN$zs+ZA(C?9>hp
z^c|df%$jDp#VJ7EVBU?qF{QW3b^9CE_itL?9avuudB+~$$qcuDZqoy{=nVF_ix)X(
z9Npptzwp9_2R5S>IkvII^pSUP$2*d@8{QEFORO^Yj@{24Z4}0~%Y2A!KFeAh9%mS_
zj9smAz&pFx&8GuQ8Q-l$^EhogH?s!B+Ag_u(9n6=58dxTGwUTD>g=)IGTO~x&ukXf
z_Hfhj$<C_s+<A31%g(xJEj`^`gm%(s!N~#z(R=*%&&zt}7aqtlSmyHhjt6qIQTd3g
zW1OFCSO3n(H#b@24VSyiE+~I{xxZ8`C*e-`(Rou&eBF%0_U!yC(I_>aomkD->1&mJ
zDRsNkCzzMDcbvnxghO<9Yzd=Np5w2ZVSM_`g|jYg+gNBHz$SGcFTcSv3`~8w+sf}X
z#4XC|Rn)W+dzpL{0B7W2arUv0rk*Ud(UroB4o#y&Hy;<m@;Rxqgv5XIZqdv3GdlFM
zc8D7+mzsVG*f9H;H%{8mOD(%ck#(NLgB`buiVY753AK-hx#Qm6nq}I|<~4HLUblA1
z8}P*i`8U5v*}L{ljSV{^Aeei+_WEa9U`_U>MG*3ov5mN=PomVXTgDylW6>91d6l?)
zojTB_#(S8{OuMX+T`^%mp$!2OdhivDH3d~xPmgHap|T!&9k#tJU)$C`UefzcnFL(y
zF{X5qeqm_>>BEGkcPvd9WUX;4dOI(J<(3+{{$EK@KsVJPM>6)RYfyn#pUZf#>gXx~
zt~CAw5>6w0m%$A$6ahJxK|&@Kmu$^0!vXV`FwQPue<>?hsaTZR!{9&*t3suM3js}M
z%TD_?nnYOKR{#0=ePCDKf0-ydD*{@9QC45mr!#Qf;ZYNCe21CIf*0PFwXjN9SCzm`
zU~U&2*~?%Vu?4xMz{8e8CK4J(qTLlF`sFRk#nW?AKs1a=@=`dD=C@GPe(RX7x<hb6
zX-03&e@q6M%NkqRu*5aFi5lG|UrRPrx7#qN9L0>&-bRcPUEgvpDZaA~z6piw{#RL2
zYbg}WWM8wTiUO>zwv4yi%3{VdoY3B;0w;`#oQuYk2j$2uWeRreTaHJEz~p~jHkS+n
zP+KyWE&P2G<L6|?`%LM*a7uA5d)kp4yXT-wf3PkD>^1VQx}r2f89(DD8HX%5$WevN
zp}TTHADv(p69zw*IG#<)wK)}p>q5t95ixT!YCLobazVt?(zdQDu)7G1tK|Y=uIutB
z9ISo01ca$-@Gs@cDJw`<LRI-$Wk5Z+L`zA6AZ1RvgiNCEX{X%|ree668X95-%U+EV
ze*tT{Ylla`_f&;aWcg0bbe^vVLxur&Q6wWH!XR*wuEUBokquZxsTon5%{&_MSd7>*
z@L5_i^Lf*p7X8mV#Z9wZ)U3&B4Hhs6(F?M=W%o&CD6B*Hs<crH(40|Iy{t)y=p>@&
z9-tJ9wh|#J1M6BRnnrT1!(;B7DmM&wf5u|NeN{l<0KO?7QlM}p+48S+)s&AOtJ!G<
zWh=Bb+mlbhms2s-_i2@G1YHSEtmhpf9Je4<+IG{GjM9jaGhNA-CJnZuU~F(qG^N1-
zT{Tg~F4-or(r>IM+w<m`f?b{bg32k=V72Rbz(K2*iV|yf1Vl6EG`B*kK{C#0f0CuN
zV*ax88z(%>s44PP1ei;Bp#kgfian;<uFH*^=wm>H<QW)<I#2|jBrNwg7K9C+tw#M1
z1}NY&Fva#9&L!Vy`O^49xl8E`5tdd^#D;cH8&Ts`JgYH-ok~=^mCB2fxi6=3t`a%o
zy@4A>_y}CfjY{v?Ni|kB_?IsFf9D7fjbE)qm$$-8%{fjsu@tGUk)uF-v5^`hsr(9!
zwg~^Gz<-Va^mR?A6^jnOEd?UQFO=Lsn$|e3t60WWlGV(kzt|A5cJBcG%Czi^kO(Xt
zfuPNHYDSAj&o)l3SW{HK5+~J^T6X*rCN?$rAzm3K{p#Es(;))YU%su&e~QUTCfnF5
zwh&DTEt_9YNjC?VomsdoJ6wf%lAwl2rcAQQK3z`ELHsu$$5@*IgW2GU>Z=A)8%)Yx
zu5=Xk3NYigb(-LXwPEEnSACLBx7&S9wWHDwGuvq?p@gn5YrwErouIDs1t<_&)&RdZ
zFYYo8c}2-bH0^e&jjeM8e>>Cca*!Hrg+qdDS5AAQ#{zj^=lT7w>d|#dqMBH$MOEwN
zR7W?kPgbv2n$%~DUEkW3B?d0Ybv2Y*gH3Wp@-!5N({xM~+rm<JxjoF)5v>f74|8)5
ztKG&lcWlF{7oQq)y7FEoyeqSY(kJKe#HNYd;%h-gLTuDjUpcEOe^VevagS+lnTs&W
zY*N8zMr`%sr1z9219hHsjY+rcY0@NK`Wx&i@ywCDGJ~oUmhQ^?w#^yWgu)f3VB*|_
z%ChCq9ra9yWF=UhlNUgMQ!`-!Gfe+vZMrM%xsuJ&WUWcpr=<zhW2XWjyxGyo8Npbr
zbh_h7MidI`%bwuHe+L=k7_#4xI&M4xI*%iP3q7-`mG5g>d$W`p;lN6+wtCxj{7C~o
zrrEp`-kcO#ggoOIAt!HF8v>#qFhVPNM>~f=YuMsDTta(nHRQ}`)y<RNi6E9mZsuda
z2;LF6SV?-Xnwp}xSqg?l%+F0Ax#`AGh+OvAG>xaG>ZJQ?e{Jh||1_dCmI`*yJknF#
ze!#k#iti5TUV@dZqjh?~Y-bjT)1!%dWS~H0GT3Dq(tLL&@Q6aX2!lvYzxI9lW0c7g
zA;x{jrzMH9c53Dl^I64~bD?7i?dO&{?^oqWnGf!htDK%|s&ewH7Bt_3QIguo_YL}@
zOGz(G)SU8mf5nrkNow_b53J#EqPl3~>C??!gTiAq(#^UCo50F^O4pzr?r25_k0(M#
zamTKKQQqhp*y>N}8ng*QH9-0~@x-`3*flU}*L4l%^>)xO9{Cu~kAcQ*Fr<Xlen#=@
z0~8^TPdLTpp(Mc|_X7D)SEaQVT21dvKpAI}U(KEXf7RR;`y5)IhrO12tNzhc>oFDF
z3pTt#4<kiD`L%M`I5Y)y@|y8=JS!ikE<xz&Sxu!53)CNva}~Px>2}DQoZ|7}PLxr+
zq7`L3jPIiSe7htiqg*cuz00yB$+>GJW?iv*pertM%jMj;EiO$c*X@qJxb%UH(mjna
zQiy(bfAVXcan!=jj`I!|&1G7cIlWwViu96+jdQ;Dr&Q&;mq+dKf8Ot678HAqhC6!y
zL9g5K-{^FE-H&(nw_{<^^Ia<f#*(31+q4$Pf3vwwAc(C)$9vD#wlvTE`QOvnO5vF~
zckF7FPHz_KPp-ll*B!fBM)~fp_BeI3_tfh?OA4<%mu79qL&(C;Lbej!DM@v6CJ9`I
zj|EA#dwO&n6NdB>nhP$YJ(gA@$SmD?zwl7IwEeOFPZabv03pr%MOyy-`tkG+MJ-gA
zm%$A$6ahJxu}vxz0W_C^emW?BNpEB~4!-+W)E{U?3$+1;0avvhAh$I7Hki{O$tj1)
z|33_+M{U$wYU>8(P`2wmiKHm<<04s}{yNFiuP^8CFTZ>mIC&AQ^7{1s&y&Iz5`QX*
z51l=ErGzu;^!@MCzh0%(4nJr7fRx7I$K&4tf7`b&Mg<(9EPn8<Qg2^4Jg77W#~8iL
zHEx+>aFo8_Z-u|#({r33rIU$5#1D>B=I!6#e|wQ6tQ66uKhrP%?c@8)e_yKTmsHm-
zAb+RYD<&vZjC9W0K{471Ys1O8C`@{_D8}M~02Kka7{G;DT!8h<v-D1$^I|F$KpDL~
zIU9sQ)`bcVw5Teenee%&KmbmwKmPga#GZWnwNZ8;0(eeYeNCT1p>>8gxDv-_m`N6V
z@VUMU@B)mlt8lkVjTD06gjR%>0(VvlnSV%VyAU0oAkkkwqg>rOiw0pL=aOvFUKZtp
zrA2=ZTsMs&sO%+-;+C3>6W~J`l@(j~CNois+2m(BeZO6eMV^t<v-KFIFdjIU<j;8x
zzH^0a{#WTzYblhSWZ%-Iw!*4dbs6tAxslM1GP#k9Mu@Yt<xr4<ZTpewl}#8aZ-1#v
z0b(Kz3fkPKF{_0PFcN#Z#+(#38ph-)3IpME;L?Ar>q}ykkh+=|5PRQY;w35ZK1F&j
zoKl?2GqqTNVC0|)-c}&)8u>R}7r7{eKh8}u4q0$3mnCEl!;_u$=#++loAxCxdy`^Q
zhR$#u)EF(I3Y;4E9g6feYUFL|sDBIX_gdnxnNF3Q1o6Ph>aCbI0le;I4sTyH#8-N9
z&MW+)ZAz=MJ-Ry1nPfq-#A{{Fn%pJP_Js}{I}&MaT7$M;cuWP^8p!^)R<1E00W?-N
ztF9+`R+7%j@-*$|$D6UF7s@8Kq%zV96Z9n2b-{H@Y6DROL?KQUvaXeXU4Ql~YqICy
zv%IvL%xhn#fEEaTl71r(+>#b_6cS^>nzU}v?Px*UiO;6I!3u?*!{v4R@&(i;Og4~8
z1dytx4pnbJD=&LE*yxI%l5YV5yoD;0$$O>sE6t3A*?!T1S#}Xh;ga{$cHA%hzSdC>
z_)kKe7?_-WFy8bb&;qRCD}PjKT@Q#`-nR6%-}j&n*Kp`@$_L>cXLGOw70}RwpJoAx
z;rCT1F$D3+S7}aZj7*cT*}p=;+M<{IEs*6H(uXZ0^lOVX=q2=|%*p=a$)8L`u0mON
zl5XHw%H!;$AXDl}I79&h=v{tY#Z{7mU~U(|$TyXw^4;pXUPXEtbbpey;CgCx<s9c1
zDuYexij}1I&>?9rwa$Nsed{b<fN5Una*;?`f-bU9qrvZ+Z9*(NQB%d!F+JFk0iqL8
z)*gYWx*H<fVeVk82#D?&Coumjh0$5s^=ck$cC>MHMp(2!LcL!9kK-D~FexXFhn|T!
zET-dp)@9julCwy3A%FY^!stue#kK}{Ufw9mdQ^)Qo3dk`teu;rIL;ovxFvGQoaf@V
zWW@>&Jk$$-iJQC}%4sa_YK&~=H3(>2;yLXOM?lhd^)d3m(DI$MX0zZ%jhR_5p=A&y
zXC-RyWM!Adrr=!X^?bTG4Si1A<x`W@jx`%uJXVgr<mr_E+JBZ$E8L}px$crU&xP)v
zs8?oszWZKN2m<%Aw5PfjCQj+mu(V^>iUyO{$Y&|F4j2VqFu4d$T?;S6=n5~Ofazd)
z@({dOfiZ$}JcJj6jy=yJcp*8-336Z{iO@`RGV*P>IR+IU8Q-f?m8gW5lV-W2l0HjF
zjf0k<u8hDchktn0q$h_iffPp@;!)S<n$zZ)@B0M-rY%1D_4N`Yzn3P_a&7>fQ`Ur=
zzMDwtPI<526m{3_$CAt9o4LzHMwv2f>Bee}l;D~WiAstM<%1PHp?O8DP&)br*$a2|
zNi>GMEJ?M$x!z#i6uE@*v*t+y=Yod#)+?Sbc|D=q@_+Q2`m|6ij!@=RPGnGdjB6b)
zYT|C1fr_{TFzvwmLEY+v!Mc#mnsUY!iH8dLvgo6BI((l|rVY*+`DZh=qDRtusWH#%
zR;8tgR!!2i8gjRCJO*9_IO*cGz&ngCz`L6fTP=*`e2#jWHNUAIK<=6WGwfSCf&$W5
z<#<K02Y*U98K&7kC?}I%_%LzHo(Xp7v*8mRRx~%jXU=u%$YPzUd3McvR5t^rkcbrA
zYX!kV&oz#HrPO2_kV5;{mQU5WYLP4FLvcGFT5|;x6O6&IC85R0{X7XS3#?=T;>ee)
zun3yVP;IU=bd3dYYEE$}Z#49crt+o%X^h)ebAPVMfLeZ-*PQ5GmaK3iJ&iY{r$;Ie
z9R~L5cAM%R^7c#XbP#9TQ|`OZaEx)rd!dclz_?B1DrPs#KI%3myz`1rg0lNzC@H#j
z2C0l?RxPVfZb)M+e>Z0QSuxOBA?@i(yS31S<5suAOjx^VWd{WyhURGb7IgLrw)n?j
zHh<$c*SARvd{slNZLwzIEkyVE=3Hjhph=lwrqf|YUt6uob#*hwSR<rvCyGh-bK9qy
zB80W$o#Wdk>BP<Jz;Bx<i;0oc``1nA9IPCMwUe>(<QrW{**NT@gDPWkm$aD0*H+w=
z7^AN|Dy38Fk>E2fr_izFNSK`y@)PcH@PF#Hl=IR=p~Kv3)On~&8SuBhL75Ww8_6qk
zQuTOqc=GKEg^yfQ6}47}c7C$1xaFjqQA=8~>z8^;`E%}$H?29fA9*#ylYUo8XF7Ue
zV>^I3uO0tBv@6;tuX0Z`aZZtH)*|~&7q6mF=;@5&?!~L+8UUM-%FOFa(wpVf8h`M~
zwDP5}ecA<ta>UNpHXmmk9@YcKFXK2RD!6Z2xeWra++>tC;72=$antb8cUT72uR_b&
zdezMrFd9WnjNEK%mQlRJhaF(@)Sv2+wUtiJ6#J>*S*-P-&sQvWEV^5vR^H3jf=;$y
z5UXI*GZ6(aS&Z?Sh%!(dI$`&q3V&VBWODAI3dLA9qoIo4L>2W2ss!O2OmaRORlGor
zOl!LXRV)a_H|s7?ap_!TqBPxox&tnlD(wRlV|HprQLY=14k4ErY;-l=J3z&tN%9L1
zWN*i~qW>>(MPk&_`kA;gQarwJzM1XaoiU6OY~W8t0;Yl3t{&baL&*Y|41cgL19|s2
zcu91NIn{3!qOp7^VeGm0JyQkaJs0PR*CWwBL^^fiHZr$tkE{)2jnHw@bFB&<_Y8&b
zg5vz-%8GGKTb}>-q+AKHizNNoQ`b&gwK3_1iJDVBuCh`ETN-ntNml<jms;%Ly6Meb
z3e4_|hI{@c=1Dz@bnD(I^M6EKWE7v+rC^i~+7xW`PwG=BZP5zt&-SK~>w|p?M(w&j
zh55KWhEVp%_^}vrTU1g4w4W$`eSjjmWKa{u^+UlD6}c72hx#~5VZ1k8RY@1ii(Rc8
zDig9zr}y2_{jPm<Iec7>_$M9Sunr8jK_N2d%hj!anXP8|??jLfQ-6<ISQRGXc7^!k
zZtl@XpNN?Qx%Y=#UOl-h`d$6-T$VP+zT7SYfyk9Ilz&y3WcY!lF*)fV;gjoP>TC_G
zx}zwjwo*^#o{~5i5&gK&TTM3fA{aOC4m-r~BVriMdZV~CnAbvN9CuSJ_8Zfx{{dT|
zXow1BZe(+Ga%Ev{3YVM(FA10OE-Vw5d$l(ym!sh>ntx>@mG_^&{`n${uu=pYKgKuy
z?epv9KQB#mrG>F!GN^>n(=UI##Bl^8tP2xpC(z^c<Jal8mw$1pjTbGP|NQuUVk18l
z@H`NB(pqB%@RSwC_{rKJbo?oUhac~+Rx0`=Pw^YQ2s#7hoghJi1Z|uZfkOhq)ENYb
zae0POaeu7RG9P6${(7pT1YZ4J7W(P;yc+4VJTrJD37`KMnYTQQ(3Zblg52FN61Lj4
z{eI8#<E8A3%tFB$S5p|xT<ZvNZCs2MT89OfeUy4`oB9eUtm5)}rL&f@wlXDwQLy?a
z$vcd6@~Y@MXt})bJ-!$UWUgi7%4Dj;$PV^ONq+$}#JEHUMN+*fh$QR686k4EWEJ&$
zLd)EQmU2NjaGaz0%&tuYrMy%dHmAFB)6xiq`Uc$E3edtC#aBdWvLwk5&i6~ArEu;Z
zL39uV&#EBj2FXWu#OCzAiI^a~UrcX>T#@r~sG(_c65|)<tW7bbXk3?@3<#rRNaU2P
zOdqS8Q#wt-llW;X(D^xt63@@OF#-=MpD}7I$IP>>0JqxXF17|8Lskp=>BQ&hWNy}O
zhX2g2w0dM??{p;mFW3&ZgqM*8FBF$y83YuS3@a#?{^~B80d%*S>n_a$e-~-iXQL75
z-7s6K4ZSL-AJmC*c&b@D*RDLOPQ9y^`Z(q4bVOEO4^~X{_jOunnkIKPRaur{!I3^S
z6~|kuOKL3ya}|--UsGL`*Fj@5pQ^5?8M~&x?6!qZ-)PdePS$aCX=OXmKvz)~3sa`B
zvS#GeH>-Yhhei7;s>rZaf5+iUH91$&+4-HD0tIDdv2%V4LP6*_{^>XQC9Sb)eT}C<
z->>=k%PtwGg{F+qErH*^?8YTe$8f|^nE?B)2$Xt4xYrfC*^9D#KV%&nGlR1WTW-Z6
z^5~|ER@1-DJVci%E;rBUGG}wFyf)x^tA~~Y=g#n2b?!q;D9VO9f7}}s``NMk&w-tW
zPxYt)uE~q#SdnLB7gNDDE>d(H=f0I4u?;C{zHN?sS3W2XmUWGj`9>SI(4V>4+;0V2
z<p`B!SjAF4R<4Env}z2li67lJFe*oo(2}$e%TC)Io6FJJlv(yPw=-f;O*BoRWeabv
z!LM5lij-jl&d2Qje`we@W(vQ3fZGXxq4>cZ){Era;XOhscNz7)F{7Uyky;BStnHvy
z7M2hnE^u%)Ql!zGHsgFXw5m83pv~4Tv4rD5@{IDd>J)w3^k^CoS#z%j*xz)Ks3bFN
zQ%_t|8aA$0>xEx8zdVXMXp!kU$sX5Gd8_gLtg&jz<^qqKe=d6Fwk=YtCQD39p1O`*
zLrC$pzg|kjl$cu#-_L(V<#-n2^H_77-*0Mmw%|q~hSnce&)H>G#V8hBgjTjXuXua^
zyp+c?=E6~iO6#)?Vh){0Ou}bG^-AMyma-N@iQW+78jIAzZeCZTMp~<y%=gw8;mMoB
zX}$TGyu!6!e-WFMvTCc*Y-n_E)-A52ptED!mYv&(X7sy0;%!-Yj+?e?(G{%wrtR&=
zV_Tx`$Jl;dc5JGy>H)6G4NZ<asC;U}+;#2Aqp}xC1{mHb_Vv=?f$-i8Q}9^P+qN%#
z_2sWO<M;X;N-BYNu5y|tc*L3RjujvH9)ZsPIJUuJf8R>kPx&4JJVCkXBK-LG2*Zv~
zc-IAVu;9Mwu(OiuX+KDzsYn=RueQMQ3uXBJv{(7u(o&42Pkog?P~}(ocbtNknnPjL
zLWbkl_wmJieQn?NN1m-+P4XK2Z0+_`e^k5uZcqIlCw*ZvP#C_2d1y!N1US9e@wVxG
z<$IiAe}%a|E)Ag&(cD4%p<NvaCH+!Ibm?`8To4Z2!fwSr6vi`#mC5*OM|M!HJo%Yg
z=xT0i+!c2TjUxC3N}ug{z)ffS*`C)^Rr<r9CUMiA4+O)GAN@2bH})Swo>hd5g-r7M
z2I?|S``NL#{aniiCCvP7(-R7Se#5Q_rP7<WH{~0+4}bsC6Wa<(`O}wyv!}fWjAbvJ
z#`kMaHIpOnd7_cp?NaZl7yRHelKbAxQqB8+YiszUmyrc86ag}qu@D3k0X3JwKp-cV
zaCtoie>e`l*RSv&Xhjd|1vDC4UFiz2w+#BW*ux;0WDm(;9*6n+OG=_7%XXD3T^V2+
zaJwu^qDb-ak+Ph=Omg~gcz!*+I~$n12v&JLy?&k)zL5A|No;g>@=6J3)b#po`p=`3
z+Trz#7f5LgULJo3yxXS-qXPC&7B76O)YF3pe??=kkI{Loae0ivUiyT0h4&-1<M=3@
zq!}V!j;^QwzW)0l%VMR7j(*Z@e0q63{C=oQ*G5<uCx}6~Xs1u#4)g+)70$;=1tpB0
z{`h73eE2xdT4`A`wSRwnnTD31itvI$;YsJLoe57nVQrY4i^9;WEIgE#1Z3KxXq+U`
zf7oaGrSX@1;a~PNuzh_g3%WK4gA6fU6sE;Z0u(yD$BTYbo3HD@EwTFh=^815(vt%(
zz<8!xxeHz~cgnn-ig5ybs8miXRN(DYv=vsx6<*{$sVG2QU16movLZBzv*dqPW<s+T
zB=qha<t5d$Q40r#DU(_x6**@W1*EFge~jv;vRfyNiQ|kkn?>t}q*{2#%i5xR=S;z(
zCy{g&mA?mHUJ9@6J^7l?Xieecl=&mmAs(<P^XVIpm9N`P*xD$(bAP+*un<I${%?03
z)?mDWA8phYxeOuRY;WH|38PIk3qmusNy0qlyR~m8DM3W979J~aUZI=J&aSA+e?}M#
z(gbWa)&+-m7o{$%S$p$SHK!$@;Hd0o3T9n4o8&HRRxSvGfhLpRgeU-7gxF?WaWcvI
z9E%i4^4w;M^owL@Oj76**;`u0CO=GFqX(A_AQgv021@`*skg4^sajdTsCCq)aWM{H
zSk&=M=t_B7UVO5q_s*`6Kb37ze+ju|uXQ8dGGGMA1>SN_czXqI4fh?;%0wipe#EAe
zxaD&aK_!ZuS!BWlorp8hh08C31BzGtFImg1rMF!N==S_LvwBb0%nIrVZ^-7x5&GCM
z6<Ua*4K^#G$-ZPDj<2OWuXF2_u(3Kz^Rg|vM2Xtn?Ggx1pz|7ttq6cze+***4998G
zS<5AK5YaAl7^!IAgoF`o4=T`<4}lwJB5`WhY{(Y7O4|T)w{KWiBZ<Ck19p?@Wqm6g
z@(TS<)V+*F;dx*E?K|MTa7qnC?!5zLo9B$MDpl{ZZx7J*nmA6DG+LF3*xWhl9jlhV
zg$WJl(!waIx}VToz%>pWe`00Sp7}&h-n@4j)lkbYOr+&nzF-TKZn&CTW)mXrM87)7
zc2lmaKFY`_F#8@RAzcSK*KhXb5R+iJXN612e?NCFEmYiD9oSm5s`B5cZg9v_vnZlH
zyRl}yHVjcRc24xfRgYOHAdeECQoPMQ{9`A&usSA(kMf4at7LJxe<i9FaKz&?@-FkI
zHQFalvSDq^AhY&>I&#I0Cz_1#eeK!L*vD2FaYN-(C+@p7&bfx2UXtDwB!XyT+&B#g
za1iSv*7$wG?fx*OJx4V#G1%#^iJEZ`LNw<*!BkL~%%kcb-u?FF^S?g*tOO4DlSUv^
zLAzI|uix?`CVrR^e|SHAEzd!KArGlU#Eb|gRBOx(u>%jR4WFh162k6?#9HfDYBj<K
zX}i7&AO+B-5$KwD4QWPhWqM!xMqX7Tk$9@hpeLH8jHO2$&!gy(uKt<3;;8>khtJdD
zW0hw#_(v|&OSdLYB#H<{^&;ga6NHMMSC1VWS>RgZjLp)9e-1NHc>$Cb5ycm$N~&_o
zEgiSqgj4Yh+{E{>eS`af(JYTrDL^QIi4A67fizMTk!G>B-=DsfVPA02hwh%Ic~B+_
zg)=?qh~NSu&6S`qZt0p6E8V#_hoj?@v5#G*rxyO2DYKjG4Q!Ycm+*1wmY7(5XuN*h
zD<0e=@68VJe~iJT6AM*jFuP%~Z1{HO0258!!Ph=?VLpUe<75A1iia$}c1Hr9i9Kaa
zW_>62){9pGxrS0&W%lHXDW8;i))i8ZV*{^pErN7>U5Fq=DOKnr`DC3V7Smy$9(rkX
z_-sq^U_3fyGfG|_uZ%*(L1s7~)k)`A<fXXPbkVPuf8ad+x)Of@)LeQq1j5=<bc?C$
zE`>VI%O@>Puh#s|H4+y6;c^DLl@K&RIEqV#%Q}nJ3q=&I9QN_!YHLh2dG}J5yOEq;
zK}P6ng_E$5$9H3m`Pyo0mjh*4^K3nhicQW-$ax28nh>yY!RmeMN7n=nJ_y7p<0{lV
z6<YJqf6W`Nc<APCvxok5n)fO=>Y)QtmqHE0F2w83ghl2Po5du(>&536`Q0^CjtPzM
z()KmzMCE>Vq(EYpvmhkPLFFme?M&y!s|wDl9i^We)HOxkR7f%x6vWtFxO$Hgp-~C7
z;HNV#HYyHlz5(vEBv)P975F|g`!;`bhnrhpe|37?ZWU$HRZhZU#xJ~bt$H_;<oryn
zdUrB$xsDs7Fw&2!NOVq@i>)MWr^b;K0JJfK^zwSQ{RQ)xdO+dEwREhE?+#n;5l$jD
zss#uCTsr^{!uW+s6_j^L4IpS?lwFKGHiJp0d{+yg%W^8KIXe2#Y7JSzvbyAo#&$`s
zf1V_6dd`3wAMHqk5FTjVAv~kZ!mLwE?Dz@%2KLELL;gl&pnj!jv}snjhh8sc%eV@9
z?BJT5I-Io`?J<*#0oM27;))EN8ph=g^2v5!2Wn<qa|^C3W6_#@@Y{yeFRwfTM*PL@
z?kvV%!rg=;My`O*?JwI4p^vE6WU^6~e=b_C>q+Z(>W4LSprlvyW7X;q&jzP^99!#&
zYP&YKGqCjax)wr7qh3m@Y5PR7QM}Waah8H0qK+fZN~;K*)+0R#>$bZGWzn&5Y{#Z!
zeyj~JgxEYQtbwir1#OL`0;+hvl3`!;BnVq0*l}(1bDFoDwxnQ0AFm|^jrrp>fBW1%
z>nP>qO;-R^C4zB*V_ILtB1=jSW&S3H5zS;_?^L?d`_w+LX)dP69+a@<zB42;r^VZZ
zTc{tqqK9c~Ns24i%}Ygrnb0iI_9nc6xu{Suj&0vabRV?5zpTnmqnWtHDxk<(`|8l{
zmpTM1R9vinTc`G=-3Z`ySj<WRf80b<0(dew9U4#NsixO#oU#=vN{_PJ>>@^Fo-ivU
z4PvJq>USbMN3TPA!|`-0BDgV*tbv;Oh?hS_$h+7dc9e0fy`{WWc?wrndunpnfS3yn
zmtM%=`od5KQhvpbZ#&}H84Lo&>E&r-MavuqQq2YGc9*ZVIEhHn3T7!DY20=mvn#65
zd8Ae?sGwhpkA3ZE{X}VAH}aQwSV>cFIa8A7*(q2}wd!0~A{3QtydC&_BNQ_zJg>${
zVTgLYVRWByRmoM{*5i<HH32r51gc=xi2qn12OgsUz(9B?e02q|{{v#$aDJDO1uql=
zGC7xVRV);j0+%f&f6ZDsuiQ2e?%%Iax&(-6&TxhYU@Y*CeFoCScwLYxh+Vme|9_Do
zMQTOr*wso-T<paa=bqs=Bz5<t)7{61r?-dquZB9OJcGmT?Q;k6hn9EHGHAJT0LmKZ
z-oAE!e9#)Lyq@GmS{oxTC*Qq%^68xcFHZo=Oa28oy>kR{e~vt3u%ByeKgY;Z_$;3w
zpO0}|<_B$cq!8pK&w)Gr`S#m`&ceQbWj`w^8pSqr){zq0gQOxAi=-s72s6zhExpRi
z_f%D|k#PL@*!()bJ^b^~l80awc-Q^)^&$R2z{<MNk+ljLyKjGYpC5j2$!5{D(lV!i
zW8HoG(lsNWf3h!v77ewv%(aGAu_$jDloOs=LwXUdeAeo@*A$_$*@Z*;c41sbgz><I
ztuiJ|xP+byi>FAu!mj7W7{?k9d6<<ykg_<DKT(X7#s7raOY%vsACG!z@t}TATn1{+
zD8sh1q?N~k*kp2cGeu3yTQWd3GB!Z)%6rogMqTMxe;!834->F?*sQCdC1ivxm<|Sa
zR3l?)ikcWC5ibcSEI8z_=8NNr>+o*HjcH61YEiaXv%j9+Ssjy?cS;Xl1&x*Tn5BKm
z{Ki2QRy1H0T6)*zj~+?3dP<{PV*(v0)V}2lJTIeTMHbtq`Dr^}Z&YwF(f8b>_Xdgq
z6-^bPe|bfu*g07kTJ}2vLGmS#83#ggJoa^fa}cDErW_1$+|A2}!c2-xL-l$%M4oYV
z+7k*j&ugJjKfNau8oqjWC^Q7=Zi3=W6?~Xbe4UifO;o`s$9SOPf>0zp?Ts!phzjk*
z@H>)#C9Q&|gOi}4dSepst|Vx9f?Z95=II?tf6!VGc12hzbFDUJJ|1$ExArdaa2;wb
z0=yC*-YShe*oiYhav6g0&5#SVkwA14ic2VIhwK?PviMs59AIOCjIQwxo9^<KI8`H`
zR$yd?!E8x^#Nf>^vc!GN40;^N=0_OH#;5Q);4Qf5F^|T7pUPaRq4`^sqWF%UE)tb!
zf7#F(Tx$!MP{Ee2V9~(El@-CcS`zJC){9LDTPx$JVX%O(3mk8xxw&Z&3D4u`-e+T}
zwdn9jBi9sDds?E7ddey9pD%hQ9Ba1M&fKar9G4oKgtJ@ec;x$xrjt}v7UZ!NO*EDD
z>6DUaHms84%2HMSEiI2P1UH+(#3BoHf5i-&D0-0I&$JZN4hSqSBhn2^S#D<?9IzzW
zW0(#M*oNZm0V_?|d8v$~Fo-TU=k0M#pa@+n7`)8KHL<dyVQZM&5Ks!vEW`CRYGQ65
zb8J{=qXDAQaT|;D;T4~(;40^ibZ$-Y2XbyL*fr-ihRHy8uX9;4cC^A4=koV)f386#
z#qUVx))b@D3he3Ii{+-FROA8V-eEnATTj&%S-*so8ZYfv0ZkVmrH60Bzm45*dD4!b
z)aMaQ+X_l2T7k5CerO8gzIO7`WMgXpj4^%ie!fO44=AfgP!8+zHRvtbQNG!!O)vUS
zEh!Ru2g7W$Urz6pvs{<i897&de?RqOl5^zyeZDNRI;Xt03;F$BtYfY6<QA5T>m2u2
z8AGH39@#ZC6c?)r-`zGe9Io^Ytvvrv<6zJz4M#Q(4aH-id0MJ3=zPg_qdg-ViXo8m
z?$>x^#Wd)ogn?@nL03BQ$&T0JE=S(KX((OO;SM-r3@eQjP782E+$7L`f8weO9NEra
zoG;o=Y^G<9{5DXvL1~`InGRo98kpSnv{rwf+1=Q_uo^&-$;?NEyw3eW^7TE_{*bve
zk9<DC(gEFTeZELWvE$BYP&&{Ax8!^>IiKQh95Ed>m2$@#1&?FVptazq0T#^2UPu-}
zV_HYu*Tv{b-XKZDE9p0le<}y}=lbdSP{c_j<1}2|7t5Xrxv1bp%5<DBpE701)G|$h
z;X_URI4X@an7T|cr7F_x!e<4_!Qz4i)A-Y+J&GLbnPVH5TM7k}+X35L+h8npyaZ5e
z&6;xb{*|C(6`Yg2YSt!asfzHQfBhv+l_@ko#mN+?r~Fi=ta2GUfA>Z%)s&vgAsfiC
zh+CHyAaHhyp<x$>#ul^XZ@Ea%&pR5|Av;-q{egI+p}4tw<Xzn(Qe61L-QA;I?;a2B
z2~pz9(0|?%B9by5fB2RV!6-s=M@x8o%ZC~jS-QLBo4@5H?lxYA1+96T5V6KA&)h$k
zQnh%g4XkA5=;X8je~7a3*~xvGTp$)N^Z|hYmOyF0GUz5}j5&S#8#uj1V&o3zW3HT?
zo%#yCAa~({sKV&lU@k*3edD#kxI*8}dEvrhU_~wC$JcnT5L^@?oY;I=hcE#5-3_#o
zfYKbaR|83TbM0ne^O<e|t{;B}R1E%p|HX1Wi^2rfGlwHD)QMf<E`LW&7+x0prg9Em
zPr$_^+j%^nV6t5WoK7-uAfMpicI6mTW{yeCsJ5q3mOe~zFck}vcj0?zdD}wnoU#BF
z8Q$IkLu~tOZM%5GXUUg&)20$$=bOOKcN}8&7aA(=uzEGKb@E&)u-z_+ae^MrO_Jn8
zj7pMxS##33=HE+_#t&#7_N8tqB~q#W1@;m*S(lLoFBAbZmoal56PLLKFT?@iw|)jM
zKL-Soz6Z#c!3{4J0X3K5OE4~%)ASxGm%xxO34gIbU#<_H9dc$k!{T5dcWJIy_<$fu
zfwqPb$jT<sog=xj9OUk=@5fpFtdd%`oU`sh5_iR!A-Uv7a%Zv+-1sJY4oxKdM&Gax
zM#Ls~1rekPp%{^yn2o1iCF9{nV>Ti*T2jNDOGBRX5ttN|*aRsw#Fanc-#~Sl1S*SB
z(tnVU@nI&)ss0PTQle1`aj32b!B|W(uyE9XH4>K0T<-}ee37OIO6!Q+aDhJ2V@WQB
z?H)I*Y%@!SeqyR;Xi8$T;4(1Td_ql#(bf=df@B+u#Y6FREJkRR1qwz|`!{1D4s|{P
zRY%|}2C@+u<_)Q?A)u;!1tU_3SfjwQ5`S&1Bv8dnnn5YoS)@rM^<HjLtP;S^DFwR+
zwNlP@uY@KQ50(TwrxMMkFk&i7M|g0cG?>hIP)TiiD%6OlI*Wixia%geA#YD1E2wA^
zsBeVDo}h+CLUn>tKvkPUeGy+w4~&p#S}JDA{lB?^LY0b*sQCeTQ`Ov!%+;v6kbjwk
zl6`qFhiU_b3X$swTqdETOoV^MtU+doYAdoyl`5Y$g`&+u9uySBsMHiTFCq9rF+E~t
zX-e>QtWjnw_-IoS*&1+=B4!tYGZe{nEbghVGp9y-=}8(Tvps=X;Z8;s#6+m`Av#hW
zi^WyU2^r5^B?Z$%L8-~i7*T@FhkwVTYEP9QpK8?uE``TY)dUZW!{*VYlJHHiB5Zsx
zV^EE`Xs%2MUzjEH(YN2WPtOqF5n=wK{pY{_jr-XZSV7q-1Q_33UY?(vfU#-6pIlC+
zFRuo}t_jB9w9f~#SvMUw++Y6n)9i<rv%#!usK4=SGM-_HXTjDh{UXmyI)9F#zp$Vu
z*8T#XL<v@`7qB7xOXe5RF0x&moV3rUli|y5cGf<B`mBA`eVX+Wwy%D@>LAm*4x)}{
z-FSB0dy-A@qPw2lOoyG_3jKONb#F$4?<SwlJTAptNBJD%rx<Bmy@*f7<4HeJ?3X_}
ztFUJZR@j>pRqXSn7<E|t@_*)HW>=@9@h?sL-DLWvo7Uz0dHc8a$M*X(>Mve&!`WHn
z;s~}9jxHf!;fYQVPZFYI%IEEq2GhT6f0(?Qv`?FFLeL0^1+e525$A#B&nG8qwx~jU
zeqETlq4~OSMrXc=ICX?yN_hBiJOnt_kLO1ldbCfL(+_8($ru9t<A24E_W9=hY<6}1
zm$rR3n!UfdaKp)m_UD@qSI2L<kL}-YetPbHzHBcpCl{@BLyMt{7(8(Qf|&-xVaKX(
z-VFUkVZc|SA^G-tI&41-hSSM4#{TIPAkWA4ouTN_{WYabuyk2!N;x}ubVc8GgX_`7
z=yEjs_4V+5H~giWR)4KYj!zS<G`5h9el27LA&a<XT8r0hu0F5<;x2I7g|m>eEc&%E
zU?c3;0O78Yp5j)dC%#Xlr?QH4)FOR=$3@7;MaajpAaqV7zaZ!{Xt29`Mpb9U&cZzt
zo-ijqRR7HJ_*Bd1BJz#^NGUt3DAhYgF~KRgqIRt?!#4f_@_$kc_(&`PwFxryXYlW?
zGKB1`P7Jv?g)mGBV2p~d;C|*UAyCO^3aD&Qd%=@XuKEWMLI{okA;ddD`Gvd=i=f!d
z-lb5sgaCfw`6U<*^+UvHNV``mKz)n+ze)x00*74l*d92!nvFh;e(R>MM+p1rTi{Uj
zz?3{7GL`+Z0)LY_;g!<JGY&BI*h<ewBF;iM#@C)&QWR%x$TIgOyU&NCWlYerhVojz
zOJSLxti#w&b!!ZZ1QrppRtE3DiAoDy;|`o5z*>GMPRcGL>pfNpeb@9UWh;G(+$VkJ
zwL)1I8Bu6K@PY^nBJPzeB^gxZRn;m6RpoH}IiY;Zl7IG#TEM!CC^VKOLG(ydhofZ9
zP_m?lC<2gp;oAX<TBpkykJrKpj}`fNJO2vd=_k%?{eCXFW`VmKI!LK0mIQ`IGi@~(
z)z%<7^woE_^kDjF^wCYG@7lq|b<5tf^GK`vb}q3%oHg)9tH_4~-lFF095=yR@-EHF
z2G_g=VSjSscjCk^s;?b5DWuxv?!bwvOE<9oup6MV{i-A1>s3d|YY4t*fbx=9bV7O2
z3FWdMg2&)Et|}6VXn*3PJN18V25(+>pRW2QhYPwKF1ja-`(Vjwt+PXGqk6D)H3QaD
z;63Z=0!U3O&|<aR6&}5v9bI;>-%h3<2DAR{l7G_iDf##iIc(m)Su;txY~QT(wx_+-
zzQH0T1fKw((G!05e3M7OO+x3qOcR`)Xnt&qe;iClgNw^<{xT0Zi;yCc&1_NpEv0_h
zepe;I8e`dolhFPE>~H6caMQfq6nu^|iwhig;N(nzwp~`|gysdndOIhw3NU3CD`^DI
z>VNN<&+eMSbUTHWd!;Z9s}xrDOHWEJtdf($lthmmN#o0_zTQg6O3-jKg7mhR!-rkR
zsBAfb#@(G@%?|0yNOI67uszl-6Wk{bY!7uA2fAMzBtlXPX;r%rS`l<)2NDYCIuHuF
z*?QP+@*g-9$Zx%ru|4T~_yO_K4{UF|)_?7`HNIEfPN?J7%67^w9u71X_Q$7X-<`sU
zgv~J6Ldlr3%l^?{yi6{BhLod^eVI;-!nOR5<LCchMXfWe_67@-ExT}*;BrViaK=<z
z^1KUYv^Oo~YsBV8IBo<gdRS7LzdaR{=*oWYsO2s6Cb}1T^XgTqoi^#CmKNblX@5b;
zg2)SEKd6raFJJR{)Qz4Ftb1V6dKBOxHXd!0uSBXljuGVtOLB&eEWw4iz27E5LP)3s
z(Hq0o=se>7n|`*#SnPz_T!@&}30(3^LVyQh9iwUxAAJg;JjC{=Iy(-=-gb$^4!=2}
zx(kW|V*3aR72EVgfNRzwp7;R_!GDTG4d<qU0t4lL%m5pLm!kI0S$-cOs6@2T3Jefr
zHLjb3T%iMo@nO4%N7NTZ?MkWS9D`(9?M55o@h&GeHn|oE8sQhb3ui<ZT=)-D-OP8b
z)^V~w<aG;9Qs7OT467=a`ZHtqhCV{*YhHVGVq1gx(YKR^lkxRzIv9=T?@p=+9)_f!
zC~u~HF1MupHHPI21ON^f-i0$1XXj_Xf`fheB;bbYJ<zd0nxA)CcfI}y-(6jA?O33l
z-RskDSY+nKK|$&d3ZDNL%4F7Dmyrc86ah7tfsZvimk`D-1ee_xFDrlJ#I{6=JR%S2
z$nIw+yZ7D8=kEEev9rPvSk-<0=zwoX{shSv4Rsbk7!BR$ukNqCl*;h&!UsyJH6ND0
zJ3iClNrU4jfcW6M0EZ`w5XSH`TJ>YK>Bne(N}c!&d_IQjJl;zq!wjAe6V8XfKmXXt
zr0nzA%Qa1nMZt4tO^JUcAZtcE6SBt0I+5<LA!|cEedqhc8d11LQR4I2!Kx%8VQ{d?
zD~}8|kiAU+?vy_75q?5|wT<-SIam&C(8C`{?|WSchPm;S=7;E&Xj9=$Fl#uyH^1IK
zcmM6ml9d)@d}qK3%~tv6*Dl;ZU>&UQloB4W`~7eCWA|q}Yw~|vdyb%s29}1`GBU6-
zg2;77p)h>D%a!bnm-gNv;4$Bhc6m*=Z9f(I;XcV^D`f4;$M2tAbLHc`m`f1~@U8yv
z8jcZ|rC_5<A;T1k`bkGuJ__I`R#HasvlA(L8B)U*nzy@hPAFyC8Rhx#WMnki#PV00
zhNA;7&TOzv84-Wb0Mw2Mp$DngDXx=xz4E&C3`*o_4)(<=3n4IJpV&I+b+I!Y5cdg@
zHf=XL#NbDeH)E6zIVc8S1fB79gvN9?0ue^P!n~#)odJu-{CryVIZIsd(NZwus4R?f
z?dle1SqMtr2ul+i6)h7cE4xfrx3GK1+p}1wY|sOkec6AcORLnL;m*ObUF9a3iy>jj
z8fdKWII(F8y3c0@@fyz9BV-E9gW0kpWOBy&qC?7Lq@sKCY%>laNfD@NYbe%Qg>B1+
z)1>?DJh}SuzMLP)a|Xs%cG!?~N~@67#>tR2eeR4at*(dFk2iA+-#PYN*u9FZAk?ig
zaVPq`_^p4~OQx!@XE$I^8=(XC8gsn$1Zjw_fim^K)V`-axQ;Vzop6rYnZq1>X`j;m
zOSDy^dxX$At)q@5$MU2}PuhAsuUWxN)Qfh2em|4MS=lt}+;yszVT>H5tiA?HA?q)~
zDcZiaJG>RU)7S$gYb+xSgp$zen6ufn6bsp@USfYsvCs<#`U1u3X+!*=l!r%GJ`(Mk
zClr}>QAp8L7&Inc8zFP*4XR)hp#Wm&Yv0az50TSv*RwGu4k#yD0M614rJOJZ+oyR`
zT`gBzX6kw(m*(Xn2vqhFav3XLn~7XAFSqBat<z;4iyj`jgD<bAj*KQI8XN7&av9Aq
zvFCrOqJf;phL0$FbfuDA_6FB}(hszi7wUZu8nMbfSHaY!WmBT7=R}EIqOOUuhDvdZ
ziRrDcmSdo`sY14O3R!Z((#Q}e!!apKuCR<|MgWQflo4D54x8%<R7?boe{P!FkpKsk
z$IIlVnS4Gja{SVcFEfOki8Yo~N%Y*EHFAHG!!&O)mR|Ga7^Yk@yqRGr>x~QKnL(Pl
zAeqB2WiDDvi!R;Nut15zZthoX7P`Pdy%bjIvCwa7Q)njONZ-_@&{p$8Z}eO0PRN>T
z#cJ;Apf4Fc+P2!P)YU4q^w6{;)G?O^>D<Klv7UF<)S6Qy8FgJvT{Fv6J3@L^ruu)n
zA6Hl1zWk~gY~v#`qq^Ov+WM42cj`E#@+;352};oo(~QY-bSf1qMlJek)z&)p^2#@p
zEBaj3S5h8TqJ<4&6}I+|*wocNl;avm;-;DlfweKNoxSYl$46rgo+}{#z-X*I?+Wmf
zUF|F1nu@ZTVb_qk*mWsVED5djrjUPpt>mS!Od&Lm1%FdmR+_08<KcVavO<i-$ITwR
ziKK#K8l+m`4{rT}$5`+;QnWA4z)kI52BG#7T?1Fti5qfo&#m8M4!YTtxi}b}zatSS
z>9|(HwLaBcLp3DN%(erwkI&{1j6=;D!)FP_RB`Rd!&R}_Vw+TZq1{AJz0MPXf@J}0
zJ4z1C+SPW-I634*7oV4q9WN9ndbxRR_f-=qCg(bBHK{tM7@fYGYMlEDzkX@r2`pEk
zdSK!K6Ly!%9WO!}`~1}vL1{;f*LuDFUI}zpsL$F6?<Na~s1BDV9xp|I5v!Q=L)$Gk
ztcSX-|I<k{vwt8~3-mUxOiM4&PrPvtSeZ#H0}~x!^}Tv)*-K!(VBIlrd!Wt%X!ujX
zcHnYN7G5mgiQBjmx&gaA=^iVw7EDu1W;{TdA)ifbT(xMw1f|NcuPt1$hCT$3pT7s9
zl9(^)RsmSuA&mhq0CbxZfg>>07Q}1;>NfJg+~j{cSoTGikp(Xl0XLVi5CjvGUdJex
zO&>3tf0}2W9)ElP`9T+EXU;4r=hNT=Dxx)U#*~pbae+ArKk+vjdFIiHH%F9_c*8S&
zkJS_657YDuaUS>O*}1Id9o(Y_=2MsFFpeX0aAG?Aed2=mV@~qRl^Jf_KXVwfl%LPh
zY}vuas4zCt)>I3HZA*eqDkhW^@=AMgZ!(&)J~Sj(wpzTnKY~g+N-Bg8Ue#FDd1bVl
ztvcE8#JYa`WL>sT{fjk;g_l#!Uw&prRE)EW+|ydWntLwmX+9RWr64bk1_a|6tpb;k
z1uqnrz*H#}mtZF^!G9d~f4%;`vqjj)Q?hH4`ani#S4mkYkVwYDjf6W<_q~0DI|+A^
zH`h7&?OTqIrwFT@n2o1FCto@$qK;N40B__zZxNwlAOywz`II#WA<L2Wj8BE8$Uwf(
z7GCVSF|kUkQ<HpFYkqxIf4#i!{@wLkc4*kAhS8b;&EG$Fd4HThL<p&oof?kZ{Q9T)
zzWYN{Eq2rI*6CUWTQ7SVDcE_#92@V*czM6clwys}fqLpBdkJDbK;u)Y)ah(RuPzw4
z4C*VYBO%v+{QA+Tk#AoIj>#Gfq&dQu{EVF2%I~t=W*Uw=s-ux_eE>|gmZFIfH5aUI
zpycnvY2F6TiGQ2%L`(cK)}<HAC*f(Gkqy>JcrL1h@~+15U36qIs=DbU1E*P{oWO+G
z$5$IyIe#2)H$Xsx>@`Igt74cPj0c$T&&tCyXRzh{W!NP?m$5vj<j1Lt?%`?1OBa+7
zt%=TErP@@ywT@924_U<&yJXv{JS}tcNm3i0r}>&PYJXg`)M<zh4(pUlv`IXhWP2CM
z8(x}8!6f3|<M50h_v!H_^LsLpf~LH6{8c9#bHS1PireKqEau3PQ}mvoKC`avEE>Rx
zCbT6={$=zgMc>al&VfdsM-~c*iX3+8$5|t(`t$yQY=$qQ{4JLdt=}H2=he|Rd8BoK
zj%X}smVcMwV*lW6TdBxi9pt~<QxTCmaq?W5w~EgJ3~Js}CvKJgK$t6?9K4l7%VoX(
z5SSNN^b;O;z@t5A;A?Z!GWK<1cWSzd!D(5Iy+<vnl3g^!Zeqw(q+#W?ou^S09Q1tw
z*ybfFN<P#gX>5>vvI87U=ml|g>@yb@HD`C}Xn*y)D%rB;6fshaE=RIF5gEG(pNtMs
zKJ{epBqxHF;dWCi%wKR_skf{{&8=jIE%xJFlE`Gufu7%IKawNKN6oF+`SBzvZY5(8
zOBS(1d7pV7ifp!c&}oq6)>2yPGUFtHt6ExsIEXy!rE~{J?FSd!7>#@@<@EwkmGmyj
z$A93E-l_C2HSOk&I#7UBIkevf#&zTzblDrkG~sd_)Yk=Zn{}=!Ng;XzP7}=SV^0hR
zVpr(sqJnH}{pWfGEs9>1F{a3qV)*vjCPLecD$35zm$s#9P_t)?^23U5ub0nZkaiU{
z$EEs&z>o_Q`K~Cvt^5ryo7rAG%rI|g`hPLO`+M1Quj#;g+(k+LK-ndfk9PB;k#plv
zlEmnhY?g~0t`NSxY8{Fp7CSn{pww4)6BU4WPD`B+o3bi$i2`BPyU?<6gnGNT>kXwT
z4(#UiW{WKk#jeLQ`|+D}iMdB(*jAUAn--uMT|z#>!g)Q)WE+Cs{#)7u=^!U4(SM(4
zwaTLsfyZ$vnai-YorbPSDCJ#HrPz(&?3hiU$w60ixf}clY(^KP4uE&YX3gcz*sKV?
zBR1z)j*MF-=rP3&u~|{RfX!<52Vk?}cmbQWd>@a^PWt<VTY|HqcyB8`_L&Zy??q<G
zI1#j97ofRc?Y&82BE9-{7p?%VD1Ti6wk|nC6I=Blv4IyzZ!FBW^nx+P|B)Ar1-qlr
z(6y7Mc!y^sFLINnbg6W($xxaYZ?$&FmrE9-3?qA~5%Q6&b)9sxzpNdoRDgHmWM2=u
z>yf&jIb>MBcZ4LhG~uH2usT-DHauQ+&zNC^Dcp3=xW;Xv5)&>?TCOp-4}ZE>1R~>N
z(DjycCP{I+^}QlUvB~~MQ<d0GcU(So*GY=&%<Y3JrqqIMu728d^rliBfD8^df{ylG
zksjKPv_BKA{|G==-67T4<%aUCjL+9--C1iqJ4K$~=xR7Q7C<5`KBx85d``xT4fLui
z7@c<=T@LO=wIx>kXfza|Uw`(<Pv^(Ta|h&I4;1?&`4N?|VtOYo`;o*U3CTN8qq~E|
z22vR9=hOGk-4bc(W|mO<4+n9)#Fiye8q6*cE`Ge?w+l>?<Z>k!d|ulM$yDu+pMHE^
zvnD~x4XZNd!nwcT8+m{2vHx~Q{^D^qp<?eAa?)8~Uv(mG^S|*#ML&GY1Agm`cUpP%
z7ViQyU6h_u+poNz^i1NntQb_ly8&b`?F2!DNuU^Ac`*^9VZ_V#62o1&f=Gej{{Vey
ztCw-SG7|$gF*%np>L(MIeJ(FOL`#VRlv!YAI5H$f3atMRT;0=i73UzmBg(;WcKfLN
zcvH6d-q_~T;q7vGK9g&Li3|ao%eMw(!^$sMS@hfl0OJXo%a{HxFD!pg@&<Xo=5@KB
zthbp$l1I+L>G;>>^TAeOtw}!rnP2+9kC($ghhB6f!#*|4WL!+m-#-p{!vPx~l5Bg!
z=$oHkn{S6dG})e+l{@VZ+vex@My-6Q)=O5bXT4{hSTFP7*m$1|<*n9w?t`@~)^cZg
zM0+Q8LYQdze2y?nk-dK*N}8gW(X!aqiA`Oh+)^_x%h~-D@h2-Heor|<@wB3hb|P1Q
zI$1-(>C<#2hb%~(SkV$bR++X<x@k`rjH7UeJ+1g$$OAQy85ORRl`l8ge99F?SxzlC
z=7F>0DoI?hoHHJ;bjViXIqRK|Pu}MI;a^}!WW$j5j7vpPsIY(Ju)}JX{67pf<5ShV
z=vMV&YKEVp*JNcJvz;E`B-MMS)hStx!aS-_c)#H|q4AD(G@PQ?$}R73r}1{sEm#w9
z&tj!1J|-xblL@|$ifqY3J<deLeC?N`6lIy%N1qEuvRHY~?F=R4x{Y?dNYW#q8+$n?
ze3q${kn3oz-dcZb%yP>hrp34kRpXiud982_BYW4gJkFcJN{c#G30!U*)JHDXc*(6G
zLj!>KN_EWDZ&t8*%z_vVJ0;iaIYSEwS;BN=dZ|{2*6M}Zsd@tnY-!`}PO#yeHvnFP
z4MdZ`ZD0-cF1TVLv1*7rMnj4AcYbT~*yf@0gI&r|uI+y*-3>C(i9JLoB{+P@&uqIy
zc1HS%x_~2*I*p20ZYU>Q@ntLOq?29+`l!&%xn>a^cV34mrJx5vy&H~CQr~Lrac?DX
z>&nzM{kbdW^gGwNS$DFBO0H#Sed_7EYLqYnC$;(&4(qv%kqWGvDt4>YQN<lytF@ND
z6v#kOylsE4T0ux@#k4m`xvIVpGUT-g8+ov#j*H%;1d3RZ3fHAbQPM|Cu~?<&xwDiW
zE4X4xB!87f%+{iC1fO1dVXi{kN)Jy1cRpfQ8#=gJB#K>ib*}$Q#d!N+I2Yq}XC}sn
zJ6%S#E5(t+&23|%pn)U>MC$`phc1jdEb_YEViJED8&OlMT|?WDViF&Kt`B3I4QRFM
zAzhuLaF7ZrPoxg@T&=F54$;|)&WisIm|&|gB00f1L38fA>i)=z0!U{GJ0dH}tvumk
za~&1D;?38(&o`ka!eB!3ccG>&jtfmHvP8b-_#ixl@>yf!gWwRRT;u%x9;@FQiy!kn
z9D08*{1Ig8z0-oO!5<3J3rXL=PIMSlC?e;%+UiVVjERa`?*cyPogwJIXV4w{%J0Is
zx-m>DxAH>Yp3hxn&tHN4joTV}<<_1YcZLogOuNj=$Yv>bB`>9(g4UWh41Nbz5Eq*$
zpAU+|LSI3tP(ulMcw3pW$Cd0Nl`>o1BX@sQsgY9#_Pm=02vD+V%3v~=Sg*C48gMK_
zzF((-C*whfKv%)<>>OL#5VEndG-XzU^7$;4cdbNVGTshh65GK}Ntg^EX-E|I|E<a7
z9`jr0`zTF4?K<CA6nC8OE6N-5eRcXl^L>y{Cj0RDzT*1M`M#p|Tju*9Z3J+~uv9w)
zZz`C(40D0WfQJW;_<_LSt|NX$@osp!@0cHKGPLWMUsHTsbYyV+Kb!Pt0C(4*U%#c(
z{R^3(WI>ma1uqi<GBKCI7AF&z1T-(de+TIF_xm3QorQh81U)CIb*zMTC#exmERq?a
z#KN5~^0(a0Ij@_%Zt{9u6KU8bu8$WFv)qhPmalgC)>06YB)b6YAi=a}P_qh}D+$h*
zXhCpMmT)q3Ea=%R!*k{X$vermD%fb=czdn>_40oB_t2}2q>K%X1+SQ>`RCUme?DLk
z8C+-(RRC^&e{X&q{;a8H|JC;te96e#iY@~qqgBTE#v&`H!~HH5uSn3Ekd<a()BpVb
z*;FfUuUEZ;SKd=C34Dv!AUv<`af$-gJFt@x3R$09=Sbjd=3<Ro9io2h+hmD%4yTw;
z`f_@(8jn*3*sLG{DQ9qnf?A3Hf3?z)2AY=YVCAjP(UC(*0t7A8Tw?YoC+Vzkp+leb
z(;-6al=H5psGq0ex~Fvm>vC`#@0A7HOV@{6msu(Bp5khUb`u`xo{Y0PKZ&d2%I-bx
z*B8PdA}S*vTjXphxZ^da3#nY@5_#L6KIC_MNn}S$;VGj-Mv53=E$xIIf98a6%pu6x
z3ci1XOZti9`5x;ej#;Blo@5cuGD5qS5p3i`rzIY+_F*Y+mMihe(SqW7jJY9~Cqr#?
zk5=DaZ`TDAlnzw$#|enS33w_}d>H*VY5;0w*w);{Z+q#3Xp|veb4bz88H%^pv_-vw
zRm|oS^&$;Yz($taSq~9Hf4F?A9!3f=j`eUkjxI&1^q=A@X301h>6r8K@gntm4r>hx
zeqYO@&nU!gH851@U_aKYFt8HQt5zlT7Gk*hUZpK+>Br;>I#_wj{mXQ@>-rbbLZR*p
z3g;#<DV6~qVO6;UV--=TDJD~4>Gtg%j)&CHR+Ldsb~n<SD{@@Fe@!Df+Kx+$mhh5g
zS~lsTJ+)Wp{RVqVi{gYlPy>-s+q9G`Z!2s*<chRZs&811mBo<`sV|+fUTX3n73a-9
zMRE!pjU<PWUpG<ZphASsA=UP<H#)BMyDUnZII`l-k-=Vk<#tT8is06h##7OX3Xora
zl6HfYG8a2-oKXz0e_-KV>=p}^v@JUtZb<R&Guu;9Hn(R$B~2Pm%*1Mr@M>4@$sO!X
zpp4tIBGeSWI8vn%wcr-)d#lKbFw$a5MKRXfbexnvqUAgFWUl?rHJ?07bst1uueE6p
z7ZpA)jg?}fW%1@%DcLeNEt&Gt=!jTqCv6%XO>(($m~~Rme}OA5Uo;*fO-dtv(Rhe-
zP#D5G*R<hYo-Wfn)wyfaXs9jdl@zn1A(Fns)jM$IiT8R&9JB{^J2*;+pk5>SJu@G4
zK{*&}n_H!krupAgu4nD}yly5$!pVi2<6Sc$bR@Piw`W4v`Nz7+#eDL5qr7;g;;PRJ
zyPPOx>s_W#fB6bN`pHw4v9NaXOmS}Pfna~AMq|WFxB;p+!x0KlFpb~{Y^w6JaI^w`
zkZ^7~Jq14>nLyj1KrC8un}Hc{PHHmLJNB7C-wQofY%i_EvNrcYkMz%WWV26&9wUX1
zYmRq84+JZITm?NV&ZD>c1pyC+n%g4eh$F&C#{W6we}MDlsA`Aq5DW@nn>h^V+l@;R
zRle|7KKUDe9ZsjcDH_ukpYGM3Y$)c~es$vs#FCC-YvZ|IfnM054jUA>F&375_JU|b
z1-|N&Co96y7*UFs3$S}j3(_L8bNk2B365Lb5zE!^`%0YlI@1JH2K=7$N3pMGR}r2q
zZxj>1f7(FPGqfocyd0R#pllCq@?jN*gm<oKo!Ubt)CGJW=<9|D`Rr1e;w68Q#o|+E
zUk8gh%pH}P;!&_>oZ1h6P1_{Ey6?akzk~3|sW10Sn&*~7!Wgb2madq!|6gUDv=X#j
zYcEQts}9!;2~K2Q)lbxSL9_n=x(Pi}3T19&mz4!C4+1hWm$6MM6PFs4J}Q4ZZzDGl
z?)@wF512=ChA+SnV9VA4>0-DGQU$qGSLFVGk>PXs*p+tWb&+B%iAxTL-#kf<`Sj1k
zr|%Ch9}nLyfRkZX2_vVEACsULoc;<<jap4c2&O?yA3vu*o;a79o)>zMa{zh_eYf<f
zPLCig^$?04dMm`~(V*~Ss1JY0I2Lpq1Jp~N=~K|>vu~&Ij%)5IIC_kya{BY*w+Eh>
zEu2YUO1nuLWJD`-<|!j_qKxOnJo#_b^5R=FZ;B`*@rD=rZksp6AG-2u7)eWwYdGf?
zPsjHQ53L4iHK-!pFy2P)nN-%wi8G|ff`aK6t!Hd~|4y>7Vo4U%mq~wGX}{valSeyY
zd5Ln7Dy`D5%Omp^<q%e#5b%fvPVwZ$<QGBU)sU|#2Pnf7*F8fdazA<yl+sLaRoJe<
zwbIXA7%5!jH!cblQxCr!rSjX*8e5{k)k%hSn`lCMPr#rbtqa{itNzt;<Y5djJtyH}
zqVk*s9ex)+I$*1LvRZ$w%;RYm4lj{*_-->A!&=3RSLQ5_3T^a`U%W8|B|T?8VR`-v
zJv0fqFsIjuyxJ^<YQ<hk=ff1dXbK_|=(46{Ji{ZmpcVU#SQP~y;o79|5(P78dIp1!
zb66`iM)gIM8U7>?Gy(8)V$J$Kfh(SThOS3?)eW?CRHxXNEH!_>*VXc$6C5+m1%pcC
zRtzf9-)T@Y2W94DSef4}z%qX_h`D!C$VJBik_m?y?M=oa_Zf>H0kxsT@tD7@;haXf
zBs(!9idn*<=B8F<@#f#CP8$+MJv*DSUej=6o0i(7BZbBp)guQN8~QKCMA<m6*zYUQ
zUzEBe=K;mW%A9{R4}E0U6ZBK?(_;zWr4s><LUoZwx&V$%ik(fBCsVuRUfH{aWu*YB
zp39r;Bl-FgRqmt9tS>#CY$ox2Hva=t271k$)jgQUdDQ46p6>RjNT~yAIY_$zQ#6Tu
zka((CGL5LPK+4m52zs%oS)B4HJNpx8Jy^^dv+x5a&uM?&9Q;<Y!%emkv&K!fQRvod
zcO^W}J~{cFtZr!qIax7RI9~DxbB;Vl2q)GPJQ~74l|2H$3Gu0kUw+hT@JC>XJ(1b@
zCDvy_MPe)47ULLgGzMxp-;367X93Jx2Ads!gaKOk11!Z|*Km(e7NEXel)aYkC?gpP
zTQ*mA)iZxZzu>gDD*y4l`}Owm@b4fOkjQ|tmA`&I_!s09nKo{cK>4^()33j$9}mCx
zWRu-;#%WCaqj+>cP0M@QJ~~b-<XS7$u23^Yi&yBF<xu?3YYH-#Iggg2a~MHQMvAnH
z0CE%^#q5c*XQq%Bu|m&h{u9`5THdP-r4?bT+fjd(e0lrnHOpY%8wNh8fn>?gA>c=z
zZbNz#Pf6-z)eby4$(+awn1r7w%-XcPZ@|2rRl=M$l&vo&*IDW7fE+R{hc$SV)Dn7>
zyl5^`w1&6ca-}UZ69)SEx?_c16loiP(hL+|KS7aflECjNszn4P>6`;SXaPT7&$uFg
zV2^*a+m>Pum022l?9~AIcHL9#>~)LY4<4YS2=k%J#=yNOwXBCN^{>-=ih)L$e!i9C
zG-)YQG-8lA-#@RTXf~*hxU9NTpq>&3U;?~VFbzQ62L*tlbN~mQmd%UQlo5C%6w$m6
zHZiowxH{0nT7t#E)ZGTw5<I}ZI)cZVS4)5V%dM-O?uK!7vcB84IsnB@)9NNp38qgh
zKV?;%nE`<{n~E0LBk(mQr6sAKJkq9GiVNGOd$Y}s@@QQyuU)ESU3JSfG*X7z&yWtM
z;>K<h3X&TJPWF_7>f9f{rL;`zkw3nWyYA1;I;w(yIydVnV!Un<ux#?m(sp)`Z{>e?
z?sv?^og9mt)64+wTqkhH&<!?sv_&InyTBtYx}~_Zdk-gL%f}$QCFZYBQTMa1cx@>u
zd=lGTZ+Khg)3?`wv&5AQFfWn0X}N;CG=fP!;_g}mZut#snUuq-XI?+eXNhm!4Ck5p
z4IV*Qnq%bg&=b;fxjJ0Yt)(aDN{oLBruk2Rr)jB!N;Vfs8LX;eaym25CX<e?^Vy+4
z_=-*m7S){|e{bxnyX15NoA2oTQikpC4!$Xquz~^%34`3w?6g<i&pzsudr(e_K8%6A
z;n<FFS6~$dp#!~DOK?80+^L-HolsCndBgWLS3;Ze>0O@GtNLM40yI$1Lvep^Dc%Qk
zk8)LJaR<=nyS)M^$rQr>&MN?=bZWk@IoK=sU5u9MrdL=#9Vc;-NAZ^o+Uk&;leWZp
zm)>kBevr=Jh>0dK$(Ic(-}7`z9L!?1c1vo@=3sEI^9TY~(W+`?z;o1RSNK|8Vdh8q
z^kz<il9`4&CWz>Nj>+iR20ni~Cf7<lBr{TUOg>-AJ&$xJ1y9#5{L8)1?wcB+9qo>X
zn;K1t`K^1J?lauUtoXL%4y15Q7;(H_oJvhcRW!t3+B-v}71N@mzl#;}$ZH&p%%bTj
zi`T&y_2*pO2v`ntOREshP`(8Nl%l)P-@l5E@`hD(BdUtU{YMm<5gdO#3*zd1SU<f?
zcc@t%c=~*;uJJ3&fdRU8(1|@cmj%^_T6gTwgeI5mpr^V3_@~Zf$iM}~fo>ML!27KX
zB?+e`)I*g_T^L2_%zP+%UUE-2Dm99ag2ruZs<+)-7qK2>it}zOb+5%6@5-f6J;y6P
zW;GyC^aR?v_f*<af?<EnzQjmBHDYaR=yX#fc5NNBXJL*)QOCi8!e1-YwHkT~p+Ge9
zpG!Q$I<C}<>@w%N%4ESGuC3SC1X14s@ii(dl||%;v^71ImD&{{1(WmM5bAOhUw0X0
zccab^=Ndwtf0YukT-PbvQm>h6sE&;DoKKG-g>kv1mPDGRDq(;7tbS446DWEN_UISo
z&@iIdq4!)>E7`G74rg?#qMKl;b@?@~YP_nW?E1jQ*HeaCTeMvrRH9#3-(Ib^?4@;h
zsk7jYNefGNVH<qmmI^yzi8H7AnKqUtIl<<*0d+B!-CX16cUqQG15IJ99Adoi<%=SL
zilH@aXAZQ<fKpMZZj0OMil$THTQZ+~(3X`gmo4L^Eg`)1Zz9XJExGo~K2=4~b)jAs
zS5U~Ax9R{;t~=Tz&D-A@>K@S@){TOcfD*2jK$i#RPukb~2g#U?mr<WH69P9fmq7+F
z6#+4qkv#z>mk|Cg2$vv4FOq-h{O8+mCz*uJ_adQ2;hpKMGlFO)39oYcRM~A%81Jgx
zQoo!*EBb&WAbM89@qd#4ql?5vS60)C@sY)+n}<2m5hbA-MsiTtsNN9KaF9PIn)An@
zNPJc6+6#^A!XOG~-K;o#hS!J0p~r=n4OK%5B4o3%q39wmW-uY~GGTvJVr`Vrv-a!B
zuM37=gwP6RdcjDLRo2c5MNU7*n<-q^c?T(svGvoCB}P9RXi5vTp|d6kr$Pxd_O)x`
z%ic&rm|O8ZDS~AW=i8U+zt3-{e@-=dbWRY}?ynyw{=xtX>xioqj^z3IclYh|drdV5
z9zEElKPbDO-@9t%YeIjLAW1=KE#g!`E0A=z$_W5dj^wousHoISuc(4a?51OCDWVlx
zhniwE6Z=>8yjwDrVL9`kDE?&S>j))0MAWM&D>;uTbl{p1Xu4ZBa?k=Dr6eukbCPM>
zq?`7XD1Z)Xzda=#Igtn50W*qRCo5lf*nG-Wk+K+)o8rLP;M9Lz8_OACRigu23FoYL
zJ3m>;xde~5BYG{A^9TB6MX^sA2@zF7$aK$K;%7gyqe@o5ULvw1t8X%}z9h5zJx1Hk
zJ#qKF!J;VvQxd7wh1c05C&SRL#yIkNKVL(TrgcNf&GR)*I7<8t-|S5H^S6eQA2t*}
zSHX4%MW=-ZvsZs3f^xL}NR1fCDCgFp0FP@#0gm1_zA;~%>EMNL)qfc&GoP3!P_?IJ
zP4=*iW`R;%6m0gJ0pAwWFq&yjvAHx%acLV&OZc)_L8Ahpq~1gW?87*sXGVqB6wYa{
zNz%l%kgk`43=yt;BDr4a6yc<bg<OX71b&PdT;2n`SKxmrTr2a6zzYR1VJjcuRN%|a
z=E_oqGwNKVLj{#d>{N=~T|60!f*hkRo|F=STxGp@s_wB+T1X!$AsI(Sh%(1JOz$X*
z_U|DiKY}S)M6dUyYQF~rlw;3t6oS{qBtZ#6IB@k7Y9v$cEuvgz)N`b+GW*%6X4p}m
z+y&M>{O*6(N_#7$Yqh!z);E46ANR<sGH+?hyJ8LFHxP8O`_GO~Dc)jfN_26>($BG@
zU85aS|3A1K=&PBH^R#nEFQ^!%6xMR=IML-s+EcmfYl^A0+|+8_1J|2NLJbAbaK4A~
zLgPsVNpQ_+MM`R;;YG@jZpprbA8j<g+;pp&N<x3z#;1S^GBq^U8O~g9Z{_Ux1WfNS
zX;lwro@O7T{iJLCH1D&#>Yk@l38utxqr7R-UZHYH-gS$;WLk}Ox%3zDR}oWvS~HCv
za_A;XHX2&_vAhuh&cpaA%8kWC&QCGI6Vi>nqueiz{sRX)j1oqx8z{pwJOi@Nt{R_f
zt8;%>;+~fV6{V?YJb=1B=gLgcT&4|8jKdJ3t%<*s$HX77mr};9xbwUcuZ*CoFXGcD
zuz8*B4!1Z3$lu{ctwM1?LR6fRH5ReXf-tU;?-!>3s_niN^<7CTOC~VX?I9=fN|`F8
z(lOgAlty9H?-)DuaSoB(Q3owNlpzFb_uzlQ^8kPWu%`H6*{O^XRBhB{Cuk>(Hy<*z
z)3Jnu1BW9O#RG?S73GbgUA6mzhIZNnp}hX!p<TuGokP2d+TNjEvLWhj_|v!Q<CXEb
zW}9g*^r#=YR`xmIRKSu?xsN`TDOWy~J5a4@ounq_uBN(uIu+vI-Q=Mc#$e5NVbXu2
ztP(afy5sfX=CaoJ@hMzxK5I?vye3{#(G5pv5)|Qi!}iud*j-(!<R1+6&STLs2oqXk
zz9VRfzcIe;a!O)-ElJhqPo)m?u1>i}hC3e}>H~Ayl;ANH+fc2&r((N%xRKmusu8$V
zubFNxPIJIPQAun;9eGj}#lbBfb##A)g#06KT61)Fm!p5w$XkIS{0Rq+yuo{cJAME=
zdgL9u<~?I?s__MQ^}iT<kAz-mC)sCpCG!VOzI_mqoT+|mzzS+qkDhx2^^6e0FKB$3
zRr(=vcwdMNKuUu0VyHb29F7D3=7+Xourfu?ZO-(82S?+xs)D1s9`+4cm#Ke9LRwWq
zs|>d<Yw^4q7kAZC{JNU4wG=aEWqi!pNMWkr$15xONhu0NY;@IWfRzJq^m^7-e&YAL
zcw|&tFL$2rkd?Z%pc{+uOL*wp_lmaUOjY-PAg7~((W;uV8u@VlV&FbG13?tQX+=%m
zdd)fB`DzA~JrV^J+O`GF9-Igx6#W;FZ(979kp(Xl0yH(3fi5;Kmwzb*DVI+VLIr=v
z5q|ftSfDTIgZ6f3cV=gbz<?81X$l1}Y!_%N2!%wmiBee<C@OIJ>-)`-Cy^0lQ6gO_
z0V2=w=5}XiXTJI7_F67Aq|kEVkV|nvLXlo*p_)LbIW+183QejDC1~<riUzF;fjOA^
zE!@w^(hDd_$UX?97BV%T1-iMKxnzGKOFto}khS%Lv_LtNOA4v7O9n}O6iI=IaA{T{
zwTw9=2P)Y|5rV}21dOL*n36NbtHozyJW&<v$S{K>`*;m{jd}|(O4Hr~jAG<^TtZ1)
zd=3s}VF6rIi`yqH6qbHAg}cn3Z4`>Rud*1V=JP6rnzQR+_)7iAjUlz(0ttV$Cc_-K
ztX$m|IHET7XE8vX+oCCO;$EtObMI&6z|^lp1rqnOwZJ9iP>7&c`th)6F3Ok#E@6ff
z8zl-X;`@h*xD_ASFo%h+!*Y<Ee3eE4Th#wng4EXkR)a|L6&VL<y?3#pVy(Ww$QV9h
z2bmH~OK>}W!|(}#RMpx%Wbl84%$L=$%K--9vUd(#<WIox>K><ryW%&<r)=EM0cD-v
zZ3si^SA;ku-+fAkh%_GsgK8t!gPXKr<rtHKYU=Oe39S{sAp}&_Ey#G*NHGTnRl_#o
zw-PL6w}24h9?H#w&<Fv7yD3;n{W@~6J~jx;H<o@yki?^%Fo$7emk@uLDGB)`NQQ)q
zASw5v=a+9CMBigVdgN641r}GRo`qDBi?tUsL}|Wm&);O8JcJB}i0omHCWzWzeEoIz
z{@06f_v-w7GV9)5oy`2|;Pm`Iq5Ebs{V<*$Cg4Ci>i*Jw-+g<S=NE6squC*{EOJ9c
zsYOQwtU4+TVzY=GIqH921q}bL`)=}n(!CB}S!<1qeOIc>LgkMy@!zTu>*vMVLLTNN
z!o0}CV$tIm*%{r5rE(N4M=w^^(Y?AJel$CsoWr2Mz5U+*|LUjN?BeoIUH9YZ?5C@f
zIGTLye!lv6vG-y8OZV5SAKt{D&$^Sd$w?>Ukpdh}D)yB11R;NUJQ|GwI3GTY(uqe{
z;XKNzyPS@OC;xCDRrd}=35$I3+f_c))m&orU9}VmF-g^;A(oJ0(OS9XYC_C0N#T@I
zGXV^ztz4b4wGL^6t8<npVo&Yr%b$)f#+P$v=e>ibH?Dqt2uLxk>uvY<fBe(aA!?s|
zIh<deogLkw)u4X`TqlVI1#kn+Ofku_X6U}1oK2?hE{;dzQ2c?=y*ZxE#?$ja^UFVf
zntk_fc03!0JpW;Taz4Wx`^9ryYd_1rM>fjT`~onHf~fNg_6?xkn3pfRH`B@J-FSA`
zy?MRgy&r#?9WCwio8uqHFz9?XKA&BB`I3H!x8uvn)pUO}zMR)U|LKSEhtuP4CZGCX
z5C;sSBRoIFlMc_l?en;~oY(Vy`WJ_NMfYX&1$`mtAlO1+ANK9hmRX=NfjO{8KWZy+
z?W2>~B0%ViXF;tk8Vbpf1t`t2=4X@yiX<@I0p1jm=rs~a5UuoNB@(v0RCt_j<4p$9
z$Y2ACQpA6TqHMWshwWK05#3-raZL2&*gm<Koqjz1?|AyJQ>4`N-^lUa@to@({@6O#
zKLBM>NS$|IqpYEz6@KbPe>@&dC-VRU3$Z=?<+_TrpjZPkw!98xmlOu*6lG*_*+jEt
zL?dmWiF^kfZ!&wO6q{|L2`Uq%)eSVsa;#|+O;CS))bX~7rl7!pvVMUAvr<!5s?8P!
zrsZ1&M(i;Krj?rl!+EG8GzeuO2oJ(ggk&f}qT81ouxgYX^fpewK&Sz4F{5t~DF`&V
z1%%od8O&PIFC&dWk=JqDht@r%2-+?cbD|dj5JyEXB%C80am$9P$Z8a5G<Du#B<oAU
zmNtJ-W03oLZA6h41^?b<1i}X`jG_b!x+Y^q7r$NcMJe*0?7H~Y91GRwS`k!G5zsa6
zpcqr3Sd@CM*%0^y*H<sb@C*v_|0f20aut8eQe#7|)#7mgaC8ckZ*M=EpQ?y}#7qr-
zK>NGZ_N&IcwEM6^p6}*^0fC5dUM-R>-P3=Le3{l^3Cve8|DnjP<rcDQehjj!-XMD$
z;JXd*-3It>Ll4o09zu7~Lo8leR6#G{opetuLv1LxCTm&<w;woawP{U7$aa9Ggt@cb
z*x<YY34NCl=-Qa2XsV#sRSJ@8yG8~yE;fG;yT~Rgp!c;s+jLT-Q%;$8@+Jj7w-$e_
zM9ElrB2aA49n^QyTdrqPFJIVvF=UjcmUgiJ&L~T{wt-X3?M@T46>CKsIA~K=YCAx4
z0-bSQwy>a}z4KS6cbSlV`vkC04Q1G`mfAaqHl=&`|B}Kz{C}P1dj7vglYAkj8)#+&
ze<6J`Ufu2YM$jBV_knuZ@^FXzPI`X~NSodQX_H5Qw4ViBgR~9+Y-tb%LT&>gw}Fsb
z9Ryy3)|+D#1EVD0U=#zRz`UOtqkcJ_o*ti^Enc|fVg=Atv+NL>ib`~U5}MgY;8-T0
zTDRY)XS6!iw7})8r4UHPb{wm;6#3>QZ$UXq`sDAoMw9c)+4LACVd3)xW}Sa?-l4~-
z*60U}DD<)T*&*N&skx@9Ng13r7Y6S}MXpVF2Vk5NX1y_j&AIK}MpS9UwfWI088mC%
zMMV%4LJ;i13V#KjMDMnuktEbZJO>H`PPgsLRW#JJl<s1&J-citKV`_IWtG@l+vJ>K
zMQdJ!7)_NQ>D{kcQLaki4K#lXV%p!oZ=i`*9_#+Syjr5nzpZV43j2uKHZQt;ZSynb
zzFhf{@<QN2kU?mJpo3sLI?eFIceL~BA0iMslS+XByDiB&5+=#e1SRwVD8*<->MnYb
zidplG6$rHI*LPT<Xve%G0u!c!{Lj1t1Vk!g(rX)vpR1I+Z2$&?5Z8ZJsVE4#YTAhr
zjFp#bBdQ;D<j)!5Uw=TGQ%?S|3GjHk>Z>5%O9)uhqCGbVY>D>{8f|&d;Rkn)d)NJ^
zq=<IlxiJl5**k!3Tb}A5fY_E!b+jVy`FZUtfM*YXWTmkAM>wBt>y{Nh-`1_w^tLW-
zqB$p~w23AQYT726S^R%R<|c}0&%w5SH)WD;9r63OLyswwc2g$vqVx0bnHIvL^YgEn
z7DCy<c?&=o)`H<i#E^3t_LC0H=z!MoaYgTws#dBXqXr1gJ2-9`7^$ByBy<CnVykzb
zxA;d@xm6HiM*q#sx|?zY7Z|n)#1HLuZT`k42O)RRqm6g~Mq7W0=yPr`+RW?XQ^VHz
z*~J{NEbSdgDZA(b2cQOb7hf@=jo@<s$cfgRuRyO#7F36n)HtCTOv`V>{ENW*`+-%h
z#p^2t&L!DIQ()UlKk%#%52(~zptOLS2ZB<$IXv($9_PqY@^-J9Dtg_t^r{K}a%#^d
z)N=t1$x^oDTp2C-y!5psW#qHcbK_h9Fc8dYt3;%fdx*%A!gX_9fS7EXXl8{gu3P5+
z0W<6+myx0!69P9fmoec56PH%2Kq;4Sc|8P|@l`LCe-F<A^Ze)gZ%;A{J4%L1knw}D
zS}?65S)i<060%H5+e7%pBd0K0Ml{bM3?raUIz8)iUI%FfnQ?uX>7xu^FAkPaLO|h^
zYh@wcfoA$dnSS6EeqW+AiQ+QuNJ=PX)=&&o2yX8z?!x4qQ09CtySus<g&;sv$GM`c
zF;jIFe=^JaP-KKR*dY_-!s~JvE68>WA*<<_#F$hWM9H7?s!Feks<jd*b<0s+7+lha
z%=46K(lZ95PSBe}mY^8ZDU7-@ddX{KEzhX3q|$3pjHYf-NQa+*1xj{z-*Jqx2FHLX
zxh~EgDF*6lH16`IV0NyMNpEC!4pnHn)}+%&e?~eQ-?8w{29)O{N$d=h@!bYzhlGEJ
zx7O?mehe*;rLaZss+L$6t;SI`xqVzi_53LLe04!KjhezGQHnOGeb`MwnkI+V$Fgpt
z``5EaJ1#q@WQLkEz9h~;9HmA{j~UIN^~*>(@H?i@jKi_w5)s5~36%Y0<X>CTgjDjp
zf23(y5W{xocD~~tr^?C}^{!BW*Q1}AEVV3lX}jS5`~?nefFJEjatI8kSNQtYJiWd@
z{nH7c9STb{{PpuGywE_xSU-TWLOVVD`g{2P^m|J-pN~>VUX%Wi9De;6nvu7B;CsoQ
zFO6Y7vshyU5g%pJbNC%?U%yN`Dn}W@f9zn+(W|G(M}Shkr5N>x$)Zz~%vk9lXZaJw
zpN+g#DS(5RoHUeCE+-6L3T1UWTXk~q0NaM-UBYXY={BUd@FZu2F>nW-l6B7HK@AWX
zMXs}vx7(OETv2UGN_yn9Fxc%OphQdPaU3u*#_Wh=#*ox7{fWDojnsNAm{}vNf1J*{
zQ-jYAS|Kg0qtDjLoNLg^9f@139h8ch+m`BW?bJNeXo#fh^=>~z&Jsel&`1EZEU)2|
zkQLs{F510^uO*>5xomg-+Gy+!icSe{q1(yw6$G6Vh^F63#Pi<WuoVdJdK8znt|ST6
zYKga&b=0El5|e+Fkt_besT0rWe|A>os<)y(-wK88j4~z<;SPAdA0AH6Np%|rScXSi
zp4Ib*%FSRQDLLl=>5?1D`;AhwJ4#O*ok8>cn)6hh19PUq)~;hGGqG*kwr$(C^TxJq
z+qP{?Y+Dm2XZHTSbE-~NKm7~3p6a#wUe{GM8$5>(n;o5EJalzqH}br1>g4X{bWE9L
z=V!E3HM(BCm?X?wn~LgIR=k=1zTr}bztoV}_<oUyf?O=_l2{1~4#?w;1MU6rI(L)t
z!ir>3BY&XnyviwsRooY&$AsO~yUjIz>h&CwR(Mv#;}V4=_UK<8qhQ=6IWtWYK^b2c
zX8uT!$S{^3N=@Vq(29WK^kcc2W2K4nH~7;D<z<;#!l1<=>FsZjqL9zsd&31Y3$<>H
zu3)f~Wi&hHE9p)D3CMj33*AFm!i!1V@FXnDJei@8q*h$~sw+Ht$VqT5!udtbG#i|}
z>b#Sq@Q(sU+EFz^B|DzRtS#P40a!0Ouf<+3RI{GP5<al_8(yLqsH_U1y)L~^lar1y
zM@=Zni;qkO{<L)6h*Lk(DVBR^T0LF>DAm{<KlNwx6+K<Z1Ozt{sAwnb#oz3!2^PM^
zj)bAA&*fIwlz(mbvf>tTZ_%0Vy6P*a`K#9S-ysE!9pnfyJ0WbKwx55eVUd;eYUiBz
zfe?yms)sy~=2W+OM>euT!MFcW0q#;Nn@QkR_c7L0Rcl%BZ^boFiyoP^aAECD*bOCv
zEBhSC#ah3$0z_1|=d=kCw<fo-FbV-B%-38!x%Na9dbBXnrleH19N62cSFM&G39WF&
z(W%rZ3aYl)shw83;^~%a8D9l9w;akeRp63SX<Ltle=Br+mni5ix;OM$l~J!8eYRGT
z&|huDaR0G*MCNOmk?;P~NS#fN=hXC7s}1T;fMdPx0}!I%QrT!p?$m$|SyDOArMlBh
zqI>FpYLH|lm}lsKOal0JLq}x(;*v|kJ{J}eE@RXL;40TMt^`U~|D=+4-l#LFT&YZv
zk{@33$qrK4s9(EFpn@G5veM6n_<0~)RoTk);&kA3&aRWG(zf{PFvesE-jwJG(O0ig
zU+6L!1CEaPebO?DG7YH`YVfT7S+F-!#cXV2WQS0Q>db>enYB_|Q;zkDx*h)j?f!nN
zx9s3aZw)ZScCWQ24UvJJDQiyrW5Z#;#;pgA9m(CAh&9QsB{_2?Z$(OoxT`%w&Q6O}
zi*G~9y4>2D_;*B+5_pdfdyI1}CzBt3r7Mfy9iZ<s4qa1m+_mu_314XTSC|(6HwE{$
zZ}GA58%-d>Kxw$AJ~CxXh*o48T`tce-bSd0<*ybF-W_Fu-b>Io+LgvmCACUJHMP@&
z9arzvAonQmTn6Vz?v>2FdW|H&pSKgCS{E4c)2z9jhGp1Q=n!wbJKB*rbngL;0=R2~
zUY@`FRONv%1;5;A%7vhi{8mO8^G-%`WwkALRJdKm+Rp(MY%6kYOnQ<pywKZszV%`o
zDs-F!-{`43gyR2=&ZP;uFrud6|3guyf$Gtx0Zcz#zdw%SkmaYHPMUwt`!)db^kqMC
zA+8k|R_qomG7gNNkB>+8{sN^j*u^Gt{dbqU`yc0XZfi1CZwQf%fhZ#_V$~?5tisJf
ztC5t(jfL@G;7V|b1OkJ^x&lH&Ya9q%+r*^#QO4OQ(#i)qNhi8RNEp969_}y4=A(H4
z0O|*6rW~inV+o&U^^(QLK$XLdx1Hf!W0~xKA~UgB)iFwVojLc+sESXbpB-9AM1$E*
zTqlYfjP_Zy+$Es?u`izATE%F%zl?Hw#{&Cfd)xXl|74Zj)+rX5>WKMP$hj-efE0i)
z>Dt5QmSloH9Nf>Ei(rw3P-T0`(G6(10J3M`T2j~i!>#!~1FXZTVHydIiX==PuKe-T
zzhjz<c#vvRc8FXwL#VVWv3<W&=gi1UM4a?`l9g;Z<5EJ>p32(|@?E3}H8}E6d8QnQ
zjySo-!*gblf^{vU2+3|hC)lM3txy-jyLU;%a@X<4UC0IIvlg#*pV@e*HlSbA0Xm}T
z8I+sNgQXqc-^O4)Dpsju^?7tZY#{Xa#(|s<TD%$FvX{IyVMG2<5EPLYqvLfEh=x0X
z_S+sO3yMDR=^g3ie<>U(zn5^9l)P%XdB<lDdI#6h?nyn&wuqFeDnQ8PIE|n0X#`@;
zU#3z(zG>mH(TN=Xvip5VZ={kd0oEeQHZ%VC3!AmuA!8G#7sy<@84b1na@q$~zN!+-
zRc<_dtucXAWM+oou<!S&?NRN{Z)}+qdKclQJl??lB@q;iN*HA8mg>XiqP!?ch`q7i
zBsn_*sS7fsd|;_tQuVGPVBy65G%<11%o~EU4-H<qGX4mWv!)`f;gX~r51<Gm_PXAh
zF6Vr(ZA>!c%cxNro+9xi3}f;Gv#d<0&~2JMV?nX<uxxhrm!)Xtu8JM@=s9F06><p?
zZD0yuLs$1R&XWlyl8x>|pJupamasFm)F^1(0G)Bnh-Ps=RWli<lOb*@j0`3P?ng=7
zEJvGe(I^jen!aZrQ<KOJ2K><J?re1qXG`)S1`Pz3XE0KhQ(y!M4k}VeCI6zdH)Ug5
zz3=M35^@lk2)4}dZNJcglyDcz3*W8;cT#O;&!vP-)*JstwT$ZNm3#}7j7X4k6IXS;
zzyxFy(yqE`J%i&pZx&+TY4Y7KhtsQ9bHyGU&WqYg)~j$8FJU<r4v5l4+CJiJEovkr
zPx)I&pU$i8T`ab{3DhHKb*t>WS6Iyo@qRuYqY&DQABd|J+00<%KpISli^qT)xwGu8
z03~C2CcSgzTz?_<=cAOTX}wgH->}3+_5)=K#x+1Q>Leq+P+6})94rON@XCUR#46x1
zYhO^(Am6nt<aw#(4<Kq@pu)OrP33CmXoLXSVk3q7zHb;^Lb%I4k?u(_nW#7h-aM}<
z_%+c*w(Jn0IJ1zL6(q4o4GXCy$7o{LEq^pNCQp;gkYSt9E8TgRa%RppDO@axH~`yG
zr952n<%)twohv=G87V1lZa-Y(tQ9Y9NRm7P+uNs)z?dM|0)UeBz-H+W{le#|ZNR|D
z9JQ3yk;yd8X@T*XuOsV}zZYd&8#*419Cm7jgp#z?>kWF|{s||6uhtyBJ2^M6-JpB-
z{#2dn3g9Z`08haER5qL{J_qxdVn(kN9k;>#1v{rx-kuuVq!vIeUw1U=9Fx56;;)h@
z>Ci_=U_HtX1ZdQ|JNMt31^}B7_hS)Y{opMwoF@ffWR83>v6Z`}_7SsC^|BKbW&WJ^
zYaD*lf~yQDi5{UoE8i=vh)D||B}d0l-q{BiXWqjrRHFNErK-L0gUPU>9$PV&QW+sv
zmLD|XaQE}{6v85VQ`JD!PT8$uNI`Me;~koDzR<IE11h=*eB7#M7C~JZuK#`7_`YKq
z)+uI%@jH(H;2}@LpFy<~IzRDVmZUt(;(UwA*gNZYb!M7T$Tc{xv)~RS<e@d+{I8nq
z9D_{(u~CIf7r3O%MaLaG!Sl&V(KVm48j2hr529XzvwtpXvhG)++v!MgncQBd&RAc_
zEYSG?o)ZJI48>0nBpy}=5oX*LjA*NI&0}^70yKFX7n0~Sk{|IL02*S=TgSDEokIyx
zOz!R6TPc1wwZ4v@1h=X#2i};W9p~|Ofxg-9p%2kPiig_HlKTCx!k4pGD5;8+U{G$p
zKx+t)oeAmc7LRIEA0H@BlQ43|1a2Nwd01Uk+{FgcU8Vb{$x^u`3r{c3uU}cV+yAcw
z++*QH{YUy9SRx=xb1|h)1Dj_a%xyz3qc4L58}5d#|KBt?$Nyvx!Hg*=(g;>@!P4OE
z=raNLPhYPuYkhboh=2Eyj#G<gu>+blm7=oYq|6c$1#eRMI(dlAQ3wD4sz;lHzYfs#
z6C+yt(-yyUY$X91*T8oOv((&(H4MCmWw4lVJQ2U~YxqnVX&TvU4ND#{?8KP<k-V!V
z_35{lPS=KR!zPt24Rvv7HkOGV?fVd(f`9?Z17ic6``344{GqpD^t=m=+F$O0T1+)A
z<&z?U$H6Uz4wFz}yY@Q*I399dYLgHv0+P5E`$pBtrln>+dX{t}To7q1fbUmE7h-*V
zvB}wR10Q)zz6EoT&CWsVf|^V9->>kuItzi#g}TekoQ`{q)debpS(SST$_bM2xhw!<
zvGHQp*Q}9_JI{oK7y_U2SNSU6x8I_h<pRdlxn0}CiZC%;t~ls2Gz<CdI!N%Q%qtV)
zYZtsHdr^YH`QCJsXx8P3cqyB%xU2cK$tVWqYATI~@nCICRq-4$Vr)-}R8Xg8YSLZS
zO-T#|k5$x-SUgv+%HkD}2Atgz8enN84)o)I;NW@Xf}?2>*GXv1iEN7m_}Tot6&z?w
zdb%tm>vIv3ux-gtI~1ED{Gs4rQD$%mz_#4fCb+Z#yXk02j%gkSMDw<5)fla+r!e)e
zBr#<xv4uG_`C-9cAREL)4BYN}SRlFzy-RT99EJy?-8AzC^69tNm|O+J+$z&E(>H#=
zeK6e~YJOS|NguPd@gz$AiA?cG3NbX6FhlK`+D<xXjaZ)n!!9)Q#VOKpbKc0k3rXB$
zW}unei}?w!e7(9w%c~Msy>=mMo>-Vd8xb2K`<xl1#RH~wxmvbbM}dU;DJiD#y4^O8
zyPJ(_C01dnj`#-mmR&(Jg*Fy^%#tOb!B5XAB+5f*{IGO&*bg53oW**~yW*8nQj4VG
zuUYMi`}|eUDzyvokJhKiee~5mT5Y8GA#sl9mxYztu~hh0+VA{$N{?r@b%0W=oX~9a
zJ-w?H&tyo@^NA%BRb>8Y*8$c~`n9v&r}6Viw0owgB~qBgUt1eRcdNTa5_L~N3a0NQ
z94%isg=|*dnhD4SLRF!C0=OJ$1Q4Hc5^nS%Yr?Gb-ydxW$qBf-iSgIkOQ*;|dE_^a
zS{0kwL+m*zkM+?=5JUKB8Yk_-qWe~mofj1~AAURJQmSXUBI)FVEW{m5&_Z;3=U3>B
z3m?d=m{|6n4t92opnUV?H5L!RM`aUVn(%=0d)0<iDmmq2PQ?mP3$Usz-scfVn?puF
zU5&enA&91*{xmxEubJ3ZpDD3Zr(iLEouGAuQ#`}1voKgaAhz1nQH$x{qZhD>9(czx
zjrv#wTN?CBWKg)W`t-1o7l@nBjXp!C*54kdrwMf>p{+`z6l)G(DGX8obub^R>@UvC
zioFvf;bWtm`6Cr>x*FJ;8&mZJ;6UVjuRs+;vO*P{qCQsjxZ2IL%m5cSj-_pbeAR_0
zMy|*g3OWdhmXlZ~b<&GVlkKaK1H(?4wT&I!tMhyAVDAlQse4<~JwN=#nxbWSX`uxb
zG`n>iIFLb%%_zy<2IXcz?HxgKc7_$|5&Jk&3Ip?CJ&~Bppz-IK*=5dOk}VbIZD4jZ
z;nY5I8!|9{IM$bcY)g^hD?=(q9pE{Whg{hqVZFq&Uo?x>=-?;NIUerS=GM22jk9-+
zKA7#Hs7?t3F3oJxD-_1@TzmCmlsZ;45-F-0pQhBLM<l?dz(MqY8=oGHMf0v_Nxy!v
z54<~j^P;$xJxn)O$rN^wOtLlI&f=PQ)%=)hPgZR<DG%_Rcz}{PY9SiIbKa7oYWg*m
z!^@bcTQqV~9h-iNnuqfP{=CmM#cmg~239LB`b>Vg!5{QeqIVorv(?wnR5S-p$tSsy
zx0(I2wjH<6^O|}9D3_NK=iKvdhxY_|M-P7C#_O(DnKmT~g<Ka;${BWkupiGw?nO^Q
zup$xzhl|b*TuJtw`!Wsr^Q~nldb}KE)gp#w%|_RGDI54;m8_XGU%UCMUNEy+m#nUE
zJL+gfR?3beh}f`sGW}?!ZF7*!I_<4v)=M<4e@Q5h7+V2gNca7IkIYbHjX_D8@3C>F
zYf6G!+PKP>Cg`H5@4`gi0lmsHFl~^S&I@LZI+KckXy7lkeT{|GmIU)szRN%Kex*O2
z-pt!vzjH}pWbJ~WylF`x%qR@%Q_MCCe5bzW5JEr-5g2z-KQ6O>19m4R^iA^vv*gbH
zUp2;(90oVQD5dLtvz`2+N|<xSR~&KHb|F)jOY}peZu@=sW`+yCSB^%fTbsKbM@s>L
zJ}!%TYM!0z&?O&-B=AyX3lF`=8i;$$L;*~e-<uD3H$kBqZ;4a)-OYTH%ko~hi{q=W
zz}goR$xjR&?!WQf*W36|aG(;1vksHuu8LoRn31V~urR>S>vx)cUqHX=PeF#qTfYmC
z!$#ew1#NDrfx)Y0ZjeB?7khp_lpC`_EtptlXKY>{JR~SVw3=1+QxXI#48b)Av#vXc
zsu%UWS@;{d7tiFhUuzJ}!Z_i%a!5XTq0;IiaqX$<lSJO%jIx7D07BU9(7rTd`N(EZ
zS!f)9HCZ&oLVy+%Me{HQSgYtE9wGhEV-=X3qY{J$UM`u|7{=8N0v=du_zV#wMRg@j
z94p5YN|F&A@8H(YX9Oarkb}ucqqsIHd=ENZ0WK?s`}^x!UGpy}dJoHP#xx?!$(Y1B
zRC&?x7MNwp<+T)Znat_>wRd2p<x%=Zs6i|s`@VUsia4~o4=%<+a^$f5Fs;XIVCG)$
zmO(l%KKA6!MOk?Ti#oP=8e#brhu*@MprJ(-ca$N?m+tE#OFgF@QAcbuE!N~i2wM3%
zcRE_==!X>O*h{Z)D#pZbCBN4qv4oCy@N*eG$>f;%{fW!zq)f8td#hH0K9I(Ws~5ty
zUrlSVbun_rnxw#cf#fmn7!a5Lo={t)el<_7W7-AF=FydhXUieGgKsg>N?vrn!fu}$
zGV;QGvjAO&j!7z3UeS-g0v+ek-+YrkqZa~itk~U!O6L)Jk3z>0dY^1Qc31aguV3|n
zPwH-h`@h#A&NRYIbYx(LG{;bKinJV$e`P*{xDLLb@2lrnQ@tuvMS4R)lN|FrRU)G4
z^*NCwju=bvy!b4<ALAaq9WpOj%uC~;%;Cdi`Hj6pdYoTJkZZR_b8N~q$HJsRBOJt{
zbw)c@N?rkNvfRri&s?!53FS-#ddxE<Q}(F3^7z8G99-iG*E!9hQtROTq1lBrJq7^1
z@;JY)O{wn;Zzg`3U)uE;)cJQgB3y%nJ%9_V?`#H1ag=F5HpUg{o-ArCi3&wt6b+3g
zLb$E#<ot2y=7Yze!;mTc=H64B-@-866_`nwX#4BS`^D9LS83+mY4l#|oq>nbhUtNn
z{-nQvplvn1A}|&6%r3=fBkZDX69u5JSiwh0C2D+z0qh*fkfKyjf-xh4H-lIuJ#CrT
z<RfxVPOL9Bh20<;QC2|n=EMY*O<A_NZbb$Ou_uK<J90|1WSNxPOrBU&ps_KVs~ovW
zlAJ8_?kPl=luoLQ`^$Sul+9#%`)ee2xQgaALrRE-OKIU?zA^{5$xD9~mjr++`F=2+
zxg5E6&^C}^dIWAxb)EibXZD^Q9j!mkHzlW{#vJ##pm6Y|)LTxC4VOGt&$Oz&WxK~@
zS^qjEU7m~_;|S54p05O#jYQxkSY(m4Le+oP@8SWOtu66{daAZUBSabC!kbLy`8*P@
zieh1p+6=>*;wm_gN+4!pc@4l8tcz<^p<u#^?wn*1eUL$wEp2GjqZ7rRAFS&X#Wqfd
zJpkKac?qYX_M|o!!U%=FcMSP$>Ti)o6&|`V>jDH!Mlc`hFH)t|mD<>jOEuY<YWR}v
zQ7oxKPE0QtsoXrPO;;%*78CxsM`B@ClBap-T-hp3R#SSbd__BWau1NDAgD0~UNN!m
zF-3`&W0){^s$w1#*F9y-Ome4q)JaNpL6$<1#gs7HKjJ|)EVozLQG3F~xw;&(lzQtK
zw44-`J+g1lSvBO{QbJw5Aen5rvZ+m1V3uyq9mf>Y4g=Yzk<r~=N?i{3na-){S0o}&
zP6_y=G2U5tlGTGJ)ddVIsr$uQlXk<Y;gt&B*n89x8*bko!NAd!hYur9Aez#AY!onQ
ziV|SZLTx+cg*jHfHP*WACH$SEw<-@dk=n$J!`JQ1Wpm8vvH62#LOvu(pgExxb{3z)
z+Jd;xY|AtuYnd#Q;{atO5e4*=8o^{40X*C19{$ut7s{>X4+Ge7`Ehf^z>rDeaOP1s
zKC3TVYfg2~`}TT&|CfkxP<(m%H6tSFk3+<8&B<HHbhl%@;5#*!D`~o~FaUDlaMd87
z=d8O+Jr}nvS9Guvo@=<sj-~S3Kg-aJMyNa0ihvJ`>XW#KNh{qK`xjaS1cG876&NpK
z$|oJ~!Pf&C?;?QJk52i16@UB2x~*#tg0;K{w?#w&WQo>-Fbxi7b-csKW|^s&qNk$}
zluQFV6_R}Ngb_N2mQ8GLO2~m?pQ@_T9uxCiCsd$PfkOCrC${cSz9I_oP-zM#6-$B<
z4tVK!sO%_iFF)VuTJGKA_F3-uRR~e@uWdc>BM+6bd{lt-Y>Jy~uAW((R+8iPy`e3k
zDmjfL4y-`*nz}Gt^=(&zh2qFp{^8d@nxu2KvUz$*$VdWb^>C--3J$epxy2rf8|Kg~
z=b9jN3xc<$O>LO9S})~!l3$64X{~tIwJ|PhSa2v`pMGL0g-RAVZ_#flc?_p*ng!aA
zk24l+cXfd1MR3zw#rE0#q`AbcYcajNs`tTs+%7@2?kY>DaxF}=_EPI23}}0Ck6VG-
zTi6zUS)s(r0<zn9KEtDXZ0y19WzfC);rC1Mjfdmi0z&`FdIyHwh9a|jms^9#3{<=o
zi5+#YGtkEtnnyp|T4bv6!pxrlqi8jDYNa%#;d%fL`oDcgIr6nBYj%>O7a#adI;yk3
zH&Mp=UicR$7UqttJ>Ud5EIj5(te*B4N4DwWX6jHYB+iW9(=;f;(Caj*d9FRhGD>LJ
z3)gNumUMqq>tXQ0oM3Jo$1ae6k&rU_)8U*YP8;ugsx*H({mNI%4aa)vJsNjdDnU+^
z8o<bkpmiW1q&XmDf$r%V5)FO?{+B=2BsqeA=DIbcKvio`eo+aIaL<;Lj<u74qQ$w{
zzxNbQHooe_4Q9Tl<<`5e<OA;8SLG#l*R{N9us(yj4F19lGGb$Ph~oJS)%FReahcZ#
zvh{g%U{gD7gpRyNtIEhlOX=juOARWgcfjTF1F$yVi4|(^mJ`fP+5lZ<Ey^GE^_H)3
zz=`b1vJNA{{O)Qtn37*F$AOvX2$AnZs+N~oNxJ49t*j*FNx9J&E=kd2;C0zHRVBDF
zDWQ`!Q_5p)1tdL~GhzAmg*L-WqSicq!0{j?Sq{LQ#=fH)a|%1~HvD*ytF$%v0YEtb
zpWq|)3KlO<4giXk$ISJ=7we1@cPxOX0<@HDmZ1Q2cE63lUUDTZwZY{CUhl$?-=&A9
z^FJRL=mmBaxFC~I-JF;EY)#JEV9}*c7i}=<lp}wJ95SfxA{$F8MzjL!!U;t48Jn&n
zs6;R)2yUW(#DBKO<qZy>@?ti)Aur;7fYM!_(j0Nn|GjN9rxA{WAcHcpF{JJ5Vvzwx
zG^Op2*x-8S{@KgIi4qI(1Q`0w+{gv&;hm_xs5-qjgkOH}rFdyJAB;yMvhaB1Od^W;
z_(acd#LK$UmL50H@1;teCz&~}c6!|j@S8saQw67TKowH+y+78!?u%KqJadm3wU8$c
zkL>B~`S^c(WB}8|`G9lqbAUR=OcRn$TTh&@k~G~B*~h=6anM}`wQ?yzJ&fac@csM~
zS4g58;@1h<r{x9E69ROO?eY6aDe>y?*gBZ0ts$~|PgVw5_QCEHBW4jNGZdyKg-MQq
zu?u>2h{yK|J@%)ck}Vs;&1?**sSKx$3;V(PVA!k(di64YdV6xdIaNu5DL3ae?;^pf
z`}}ebAVG<OV2P4o3_A4z?z{4RZ7NZbo9>Ns!CkG5Y0x(osDP$h8Z6Ue-|ki79W9xh
zC`>Bi?8~`2d|l>;cclV>b!Q9Fpt}2v^rw7!ViYs#LgREj;hy@mYFe107~RObY<2@5
zxB6E$CVBn2Uut7t(`}Uc^wM6na&1U>k&Wx#rXbHqN%ZkuM7n_;ZJkfM&iqJ0Beu^^
z%t;|=Q%6QMUw}%J7RN2sPD`^d`&oD79@((88VYI*R&IK=#!&PKGtlWfsjhxozhRi5
ztQs9yW$MrMDuHDsN<O}&%B6=|vvkx1)p7G-JcHQEvy?0YxXp&d8YdI%2ljYSX{vB}
z8FbQ=%}O~6qz<Z>bweXywtNE>v?^((2>KvJb<#hTQh*u-hfxf)qx<3Hp%I&>XH(Fs
z{AP;EV}Ljhwoczf9`%G!U`p7!9H|9P3%W=K=)BW7S2w%dZX5b3ia6}Y?cW2%1_e6~
zpZd=B7J5V}%!5?xhV`+k*D<wtGD+$H^6yWc@eC7sOJSc2UCF#|5z2{q*+odF$yaSn
z{quC8CBUz=X)a7p+7yAxA2n)V2o*1eS|$IIm=vhEdH)_IEs1m@82Eu;l+@)^s9QSN
ze&I^gvS3B}RWK!RYNyf+8D>hSghJD(yBY}>>{)FMsg?Zx0|bVaGwpFu*WZ|=wMhwU
zoBr%N2{%&9&U-Qs387k4$Eow!k;GVsW8jUzeE>fbP<|9|NO9&*r2wb2`vU01a$D*B
z-1fR%#c~LgADawddi4szQhR35zf6^cq#g+T5zi(jS8zQQ9r+5sT;W;GT>TlFl+WJ!
z$LAG%0C)6elK8^&;VKr8*T;uspO6x%&XWNWa+G+^K2zeyjO`!Btc{8cNp(hDI9~Hw
z75~r~x~i_;{t(R^HvYt{3?;Apn@r&G^d6wMEnGrig!DthtDEWmfun=&=TmBL+iR@T
zkyphP$p~1yNNxqen|8{G5Ty6UD+jg<N3I+nQw=A%Xea?BT3vPEEhDzaJY=G9TDuM(
z#XCq`N_qe_dr276l5IKF8da>1su86z7J!mXO9G2z$MFVgmThrK<QXwMA&ZkvSOMYD
ztU(7q8Gt*<w-Pc2a!7|rZdXl$=t<m$9f|uxv#&6YzeG4$T)8=ts$J$}WoOQ&7TYy}
zuA|z7|EYR%+HrTnRI*@RF0v2di~O;~34yr0LxJKXv0O*1BBsh3HmrFk`OGWB2#~zM
zENfp?-4^7c8T$yz6XUMmDF^@Us$P3nou0K}q-m2tAb2tVWETh;8{zimT`~}8kSx~Q
zHo%1OgD8te9p(vH1Lu3Rfk<o5u30>ZDsIKG!r|d_0#dQ++7HT%&QbBfYZoHyCj6{0
zw(DL10ypW|#h!JYXdtS-QJybL3>Y9c<@gP1NO{6fJ|uWGeGaOU*|hVU#u-{KbNJBg
zC>N<)Wg#YB8#<myrpb0+*Bvs?&HRA0ZmGH`+iQ)btCm1r*af9E*l9aTWo?>8wQoA6
z&!^%m5VW(9<T2X5t#oSSmh|It+bf@tI<6H4yv8ig{aHRHO>2_%Ap*_YAHWDot+o*-
zlU#oYVMX1F>1b>fjFF`4kluoe`BzGVefXwbo@rHD&T{QKDxNG~XfITv`<{L0NmZYs
z9gSy9H3o0ZAxbqs*wnX%=+Sm|85xV-U-dt$+`97TxN3CD<k(gsL4#8|W`3wdrJ=Y$
zYHOOW5Py!A(W0~Fn3fXaNFR-7dGdb>J!a-K!8s0;v;l1}a=^9rMC?&JV$YiT8Du6|
z5VwLvK6RH9&@NAT)_&6dy(L2K4?jnHgptB>V#dlg=tsU<YO2B2JyH3|QXe_r*Tci#
z`-ikFl_XO+SlFd8`=owsoF7yidMsW|!t?~A_gCNd59(8cJC7k@M><kGGypfx<rF$4
z4j&;4o`-t)6ac+S!~-%fA$})?tcUoXvi(4F@6EbB_SW0ZBY*F{-+cJ_Aug`3_ip-X
z8dIV)CNEkQ&JuMN#PoO&k)y#wlo0)5v}2ST3yBtDgEO5&*@L6v_z6<!w(k2K)@n3D
zb;e?@J<a8-JyN#HDhE)syg|<)`46>?b&juay9KHX4rs|d5>Q$sY*WHEzq8>9BbmC$
zq4kX9YcUMG6vLC?f%Hs1_c)!-x?|2h)4#>1|2db1NDYWz;s*er*^}p|HTa>2x>=U+
ze=_~N->^$MT~;W7*u^f@fQFjgk`fXrPt!oa!qVX#{|ID1wt4AZO5Yi|&cvNRSuB7n
z`&9mj0AN|v{c3Vy_tF_7nW%b0j91nKwaBWA?SsQ33L|`0Gx~n(_D24B%qF5AjfgY4
z-#m`mzvFUSMh1}zaZ7#h_aC6Wcbzr9BdwfCIjhgDTi3QSu;A>Bq=D%$8$%_TU+ods
znkYVtF5GQ;`gLbX>*afz)&yKw8+U((8ro8j40xp_ou{z9J87fwod@}srU5<WjHrcJ
zVR$?BSlMBM&Meiim5$6&cP0ILQDoaa|DxW7UI=X=L5;BPz@q%)gJQnQNw~hwFK!9X
zkb@}`tP<LD7ypcxYZhxtudd~*8>EC;S|1y~@<yXuj~QxEH%6NgC{DROQOs_-aik1A
z1*k^F73EHwMx=*H%24ndhMgc9dTudTrIaE~>zKu_F-M9rYAETZP~{>0GiKvN!@x}`
z8gXU_WXnVs4?~n0tEHsFXe}q%361mOuosPL0;X)XI;X_Ke!AD-H(xl9_{Il}OqKI%
z)e>#Z4Em%VNtkAYc{ocDMt}Kz5Q>jf3xHMHsQyQt+mQww^l2)Uw$@!wDV&H_m1Hhn
zos1Gpe~hFF{a3hgPoqxu5me8zqj(FfVp$fw7g%EW%xujG=|<vYU7s2cs;zt8rxX{&
zPVS)u<p57vEiX)=ztP?pE1JodycNwjRo)*Bi+LGu*TRPJaH~JbRJR$97nl-&I)GU+
zD((sJsTt5Wtzvk&D8w~yL{z{`dz`Gt)X3Uq_Ikg5vknT5DO$(x#g_F!@+>w>#=Yu7
z*(}(e=xw`=vo`_cu#ynTZ;~Yli}g?s2d}7;WQ`V8aN2)kUOE)8H2e$iP(hsd&Ugjg
z3!4%19u~s8f3B{%_!8bfI)kC+iUG`H1V*d-`H6LxUp_C?0iOF5=JIw3z|1{^&0=aD
zO`Kq$?Txq1Z@t?d_nU_^#A6yx+mm~3fu9=8l&nPRqLkQv?p^YpM^)c?8-D}L*$&e-
zJLQd=3auuTO3B&|@)iK9Eu-1^2ybrL;aitGBllWlc{e+Jvy<agFP&{8UVy;*c!dwT
zo9m2+?a4WZ+k<4<WACjl4w`+(y&VX|M6?__`F3`8D7kIk8(f$`x4g7Q?O8r*)3z#}
z%tmdiU1rlZYW{y)GQQ+`@K{c=EKichW`ACZ_|}{+y9MI9cpd7|^!yRBFXyjaX!W_j
z$2|4qRa=)+8;m7hvVOhvfFSniNjYu#d(+wKlHglM)8QHa7${T6y~#<Y%d1l^)An43
z;ehHqY`k-oYHU{_pPz2MO0*}ujI}zDA~Oj3gXC%*OB6;tJ+`3b7(uhl*^ae`&%ZLe
zo@74cr}@6vv1Ro^BPbFPN3jj2-ny2P_H7%g<;6QJcS|MmJ#-o^fFVW|sO0((h0piS
zEe49v1=p^lFQR8_wF}mm%?E$_9>WO=*01kO(*2gp>Ki1}4ZL{&WJfN6_8#X-gi!`d
zF00R0Ls;cwmO9DJJ1ugc1rXA~>q=jFJ55@aCI`TCZr!BIm8Lq-62WRs7gA|Sm?lE|
z(3^G!5(_X4OM4|RAT!=IWw%)PU#7JO4&=R`ZnScAZRs%m!r_tuA*If=nEz5hIoGY4
z8HdBnT5!wXbuO=TKGvCjD2SB?Z8>{CZ9ilb0=Vm(l84j+V7SSWQo4COQ6jhqQ&gF7
zd&4nsKv^z!-H~l7xv#r(*|E+ju0iP&OOi8JfR@c;y|3jCh<Hf33!RL0v(2&JQAU!+
zGF_r~mT}M6SbmAl0YUv6&{4YM@KgUZKRa+0B-X$ctoet$Y%~x7gx#JcSe#8On4e~8
zy@KS4<|`1P3%4?**{>k}<SWiLy3{^Wg1gBW$P&ysyT0WoW4In7Kq%hcMnA}>!l3{S
zNPo5KDJ>HKxF#sDSd}s?MJLp*n0M2Aq@6l+{+64gyA;$#P1TOzu1{dzXGwYKV7`~0
z0K4C8Xj|0Gef!6v|2eE9rD4v&a4TrnObJ=15MafLnJG3HEZDoeL27LQr@^iJm^67Z
z1!f3Kg986*)qP}u=~&y=z9vU=7l{ArJsl=TCxDv-oODI%6cpC{RV}IAN)@mMZpv$n
zv+=#6fi{1+L3fVeO)FeR03qMUeRLiB@+AJ?+(PivXJS4AI~4lD7ohFnv`V!lSg~+G
z7NG@3`TPxZ%;Huf(6i4COkv0)jAc%^tVsQ3qA+PlPG3HaA~PVkzAC6H0}=FV)=d!A
zOd7Zoz&!a?FJ%#c0w1Ef_XGkFz&w>N38;8m1J(e?<MKjLJ7$Hho}8mw63~4HtB%}d
z2TgfVcJ>KWr<yy58Roj;wWK4^iz}eLJP5-S;fribjHTR}#IgnTT+t?Z&qF<JZi{J+
zHVfTkoxJFLRHW8Yaku|A377UaAlQ}<Z-+A1{6$13bWL$E*k<}=&oj3v%Ew(fh(t~M
zX^}7bLj6qxG{p)8`|Sb*%N&u<C$PVAFWVAEn3f^~{lOp7u9a{Lt|MUcH2yTnY)q+7
z?tu1{`Sf*o+*cBG#reNYZRRw`PFB)?VE_SON=Cqnmae^a2cqv>9l?CswXg>#juVG6
z4TuT*j2kTd?_tV$HU=1NMfV)Dn*P+F=DuEY3f+m=E0NZeVNVI3G_1s-H=ZGH?GrSm
zaud#E9IX?a<~{rUcxR_U5OYT&aJrm)g%wX}c&U?Mup0O(D6wfggJN({d?@(RD-ncb
zCvbo&?Qg?w(*%KzcQ;n>gDc|{5Y+n$ROZTOeNF}KY1#l8B)4)_lSV!=Li*Fzcy#A`
z{nYQOqj+q;8dB`?6LTaSIm!==*Jfz~@MP#wju4yb-mpNWyaf)1guGtLr-ZN9MIB~8
zzX1(NU$~9Rs}Cb$y8|Uw7@My(HDSV=OK1SaJ+D^KK}B2#bmb%EtA=lFiGtLcV3UHC
zo2*bi(Tb@e%t&IMdcoJQFI{Y=0c7Ken=HI{c|hPl0cKmvEd;eOXP$NX6o-s4jjHse
zw*Y5~E(k`Q)TNKz%aM6dlf*@@0Z#DpH85ew1Zl`dNk6c_Tk0Pi>M{b3AZda8&Q(CZ
zjpE=WoO1XZ9GOjqf!%b!;W$30qkBALem#%%52}$$8zlMXnvQjLy}uQ{G6&g9inM||
zgo1-CdLx;Pvf1e+3Y-?0+I*t#FQ>_zQQD#wduYOa5<bwV%JxU*=xW(6A2gL2%JhXr
zBYoHty9EqElM&nDJ67Zu>R0*t)pq~}WK*wrsCn6NEtm%vsgwNkc+oOz|2$sFLGSUb
zT!mW_qY!aKN9?+~g+>6i40z*u&}|=nZnuaEbKvLO-G64}#)<~{ounF!CMVvq*dgQa
ze1nrz3d5!|nc?_ncc$fDcJbs&3*@1Onugwdyu6V9B?sqjPx-Isew{7}L;(WInvyvG
zdTf*gKfck)@An^7l=2g7g#Rx70_KxKbsHWk5+7<>S~mf30XTWQA6z^befjmJJ`eir
z%l^ErQG#)0$jhIY=USpg@#p2v%*vB5HHkjl7wDJP#r^`M_W&~PDDFDV-hdwE8IgBJ
zVQa_8$(u%mTK{|xOyz!kVgPtQ0ys3TcRGgdhOVAgo|@^i5kmXm_<!*q0FQC|6C}M!
zPjadSt5u+6nfGTF4aAmEC41i){Nd-%{P6M~Bmcp~t5eIHjx#48`o3#B<^Slr^~;*u
z&Hb{(@Z-(>aq~kVZ`uu8JMHW9eYJYGd%b+4pF6l@A6m#-Um#eDfC01+7dNRqQ<!sQ
z%alS*sMnF5OC<1jECw$+xN%{UH6=B@o$oWEgoA(AhXMJvuT_5`hV|v2W9On)(2<#h
z-&YSKCCuM$8m2rB@z%KdYWxazV*Cvjuy-EI2bb1}NjJW=u&1r<aDr4~(?q5irDxZ+
zCo}I_mB*`b`#Ud@?Euhpfp(wa*e!hzom73$6nn2}N%A0ZCy|^`PE1I=P#b<BZRgdp
zxpYv_#cL7{(uSViQYba>RY-^nn<i0%2NwsLx{L3Z*^>@0Kj;pEBDxqKtVjY^JB3O2
zYS_N50F``85#0%!H_}<Y&FaO7CWJ)qDGX(R1U^|Ad_aMjUJAI|4k2hgdmWUbnPJq7
znTI4022)u7T~JXFPp0CXD$*a9UHf|<f>KRt*ys{vlfrCLpZj_`R!?~H(0wV_r93sv
zd->H46U~COu_g|N2XW<QT~SdS|K>RA?fLHxJbI-buc6YI#s~?Q3uy_|*<DDtH;sLX
zsG|ymArP@47zw~~QG4`FcCzxLNr$2e{=4>N>Ws|cmL$#EsztTu0(V60ysN*SDiKJ=
z3))w23Yu4CGc;*CW|A%2qQ8(Gkar?d2cq6=%hBM?^kH`AvN^m&yIZQquDfP!IW<Ym
zh15#2p)rZQ-9u3XDej;_KY1lr$jNrL-&xqL%h+i;8V_g|fHEHa9q2Ckc7<6)MOAmC
zI$C~rL3ZqC3MVSavLTqDvF{&96wr&48J+z3P9nNALnH{LL}W~Rr!fUBe=T>x2J*C+
z(pm!yv9YH1*NKeDnUq57*~!XIyM^1fX}|pLSjZJyZdsg4znQfo8O%QOMTPBQ$Egqd
zgjf2qdkrv3Q)ZmHglobHxfLVT_m6G3O6&-Bb00<fif4a7^tc&&)aE8DZ8YIV9;d&j
z@C6j+=bD$-lY{@`_t|%T-Zy_GFaHD2eo5Gmc@}KFDcA^H9z9XVp(v-+Cv?+hHMvl}
zC2@<#s>Kf+Psb2#NUBw2JD6L=fmlg<O3T45PzlgxaPuk7BQB;Fy?1RtDJ8V+L|i0m
zgR<<S)DLNnMTh6oWBKC&X?Q5#`fk}BRNom?k1q6@Vy9t%3g|%u*4jfVcWKo;uVgYd
zHtW-qUk<+59)M@UWV59v{4)S9<C*ijxCF{gj<uywC7%3d^;Q_v(RV_lp@yupu3{|8
zJ_&$JhF6RBB?`+!{ho#JL=fx7wxSocvZy5l(wLl8-i%AISpsP^v*TUB9Deg19T@!C
zN(m0ES|queJv=`^KE>VsB$O4Wt1eV8#9E<CKU~W@uOGkM0*j>;t#wlc$tG+r=Y#pJ
zxe4*Dd!QA_#UW7Pi9HMQTK1-G>7DU9qy^Z&1-4hDw%Gpru050c{gtwlNI|S9E+keM
z6MU1e2K29|y`l|-%9`I7Erl>i3uC_ntNs*q7LGw{rFw*H$vuvi0hg7U0w)Wc`q0Ak
z8>1X#j7U|&u<vM1cbQu2s`GcCwP2gr8tj6M;ti~e|4B(I$+6wUAp7*?CO6vcNFTtf
z$IIQr%g5u%%&WCKCpD5r(S0e7-dcnV2n>K5AAmWtTdypmuCswO^r$VD@#zfevgkAR
zBtsDcamQ7OGf$3|@{yU3UI!8E27**=Zr;v=B_BE_ee;6sG*U$%NZYD|)kyvw2}Go0
zBlk%i6URo5j>)%^tux_i>g<F{jt=N^{1eIi5i`V=O(B!eZaYNz-R4V8`-`66s$FB0
zW4koku}5JvwGHEYWGWFu{DC9Ul4LY9ZS4PF?i!D3Ux=m#X18(io-np(yuvM0qRd*`
znTr1!fRy)8N<Es|q<D_?CSJRI{{ERf`&Wh3nniGZ=XwC^Znd9{vqx(jYr>W~R=xFG
zpXXdv_45@~w?)WQG7Bdh8OHY=@LuQs^pv9InTv~EuHPoq;QpmBR{L}}K=qa(8NgUY
z?0WqyvF1&uPtBh2V*M%KoTMqiC~NJ@IZuTM<X=;~m58svroAG?C0FhVUZUt4A~mc(
zMk0#W!;1hqy=N4_%Ax`p0o+x0`_o@@)xOL(p_Bg7{>pOIJ9zu(tiB3}li)g%4$b4f
zCQc}aIFDus)1C+J<T(>~9E>yi{7z!#;EnkV^B)#wrzC0<|LGBXGOdHsQQGyJ=mr1B
zm3uxM(47kC(sh||f6NLdY$77!fZtdnUeF<owhI^)`X^d{cl~$2P3OxvqgtZs8h?I4
zEN*4}_p%85uP>(iPYiyONk0T&3F-cR|DO6c;D9zOz*Mp=`Vab8<sh~^kBDk~hHx&2
z2{3=mp>x+TTab%%QyK=h>LZ#7;TW~grRH29J~`ytzJ9o{T!K;*&3b`i(SHn|>3YXq
zRt?7F8Gm~=8%ZSVCxVYXfN(?3qH3sBF(F`}UR-5?lr>@J=^AP14n+VAU5?W~w|%lE
z{v{cfzlVfud*BkqK0dUUEH%9pD#|!CLaEF~J#l6|ZypN2YEB1#*y;`4CWBq*Zx8v1
zwNXYFqAu;wER5bnf;CQ8TAJ(!_1Mj@A@xvJRbEj%LDhe#)*CRbsiOKNZ+ka6k|bnr
z974T*Z@7q{*%pAFb(RSj6n<*hYk*58#pD!z*Q_`?wtv|7eEMj6PGXClor=wdPi(z6
zd_KxCEuWE~%o$$GI*h7%nD95ms_YeMDcF<JL~p&Y*gY=1<!3)>9rQ92av;k|Urrjt
zqUyq){-j~%5_IsV29qP8R5{J4VJyXkV2&#CCfk?}w2EeS7q$RQUy{-NRaGd*h0}S4
zBMf<K?&a5?b7lK#+v%B4y<51u#_e)aI@#%b<eV`~tf8>KQ9=(|iRem(#be*IsF$7$
zEzxiuXAd~2Xm&F3B9)?~#&F;@^EtvucKu4!vBFs}1^WVA7@UY(B#7v+n|+Jej(=L(
zS$Ddwc{BAHx^x1Hw#vrx)Bgrp%m=?~g97VS#)MTxZoU^KZFdSR(r*g`t(yB~n_&MI
zSk^{RT8|Hcc)*Ml9EwUU!sS$k&R3LIedIhU)+sBx6w^s|BSau@&PPU@;7WIDJZI|C
zoxXCOWsCY&r4SfwvwlK=>~j543YPuG7ZjNL<M}u5Q2hxIG$xdOBya};iMIspES6kM
zs#vn9wxr6PiGpBsKal`LpMbeG`r?4GPW=pWO8uKW#Uqj%X0SbzQOlUhOs`3jLrPDV
zIQrCF557|cLsZ!JVOL};Lb8$bq?)$jr;MWU7hf*|>OoQTSU>B%pZ|T!a;apCnZf6A
zX%l-u7nMFh%dVurNKDS#^nI#r|K{|itdC4y?D!JJ@ohIx{u>yklQaw1KyL5&c9*V?
z|I55Mv)Z&f6i;v-eod=lhsf1IemPLnc#b{4Fv*s;`nb-;?cZ^hcj&P@^0uY*zn`!`
zqA(j6F#}TwW*Us9jy;`9rlACm?`12T$5oP$9S8v=Zug?_oy|MV3hTtGbS2!9ddW<t
zB`Za2N;;)9gTR7sG-66tVX`k_k^wc?lz5{e2bvm4Cnt}POneX(TQq5dnQO1p&i2DV
zbpu1r9rM1+FFQBl1?-V+DQ81R4uRgrbuF3sB)TYZ+=lY&mrM$bDFZ6vSqD>TjJLc+
zN0|YQ{HPGq9^5yaV46L}uPHAWDc1jdCY#DDC~4wvj6t1CM*=D=6lq88MWij{ziN#9
zts3$4Hcf16@Ubj))bcwsf}7N#CoPe<F?h!+Rd+>%7gS?Ze2OqY$&K|vlRfQjpPXrX
z1Isdb$b+t%2U+lzQenEmqe2d(TenstjRgVq61YjVN?9x-fwsR7eP=c#ojT4YHub9F
z15%U7%s7%tvOMB=!-+Oa$z8q5a}z9fm-H=NI(*fnB*=nuNpxm!{(%x$M-VjXTS^AS
zxyHcKbP>ECYXChN=t4U?PuRkUIW&IxfVy1{*&q%jORJid=s{Lc&WW^V7Xy2s5ir0;
zLg0&L&6cwdCW;LKF`hc2yhR}t@6&TJbA>3w$<meU{Z`sLBB2#$&#t2OvxqR~i0nhQ
z_M^z3_unS8WULwGtb&M|Z#B@QC+=KyLWrd1WKvvRY@8MTHX1z792?0Ao!SJ0;Wqi#
z%WGDHA+uF`9e-U9bYTY<A90ji*y#ZS6)tNoK~gbwA9ykc4Mp@}d~zK8OTpc3qscOG
zz)>GlI}D?t?!=Y|X^y&^Anvk>5s&ugmId(wTWO2#Gh{zX@VCexaCm#%hVnKn5+bGl
zxGp&rt>{mnF?=2?(r)vA2Dq1b<5eo&i=E_O_UP<tM+lzXcu-YX*Aoixmr(&gV;R6N
zL7gCbXN07E$}QK2W;WzG3fCVn0*B*gYAy7XE?Atk9^D&x;Jtkt=&2qHkj1YGEw?FD
z%OoAvAnH<1`;YL>)eq{y@>|0nVN?L2F}M)T>d!U3M)y(;uk|LoMda@dmA+$EU(bU1
zZnV3@&ua@<0I48HL#z%kWh_8r+b=gM$@5~p$_`R{!HXAlb5-@UF9VDY%2~F1wwnhF
z18Ilr@92W2(I1&<7kht@^qI}>n`F0C9AhNAk%qQOE-7VV<*3#!<tpi)Sulsf6#dL*
z*1vu|5g_Sk$Stha#X}BKalLEZf>s1or)1*g%sZ%-29x0lVStHA6jA^Tp}5)WHBT9l
zqpu}r<OfIShVKF@mh@6=Ubq1CFtXkPIgtXBA=w&mx$Isqtu4XmKkdym{7ryd0lc%n
zd%&zWrv@W`?6e~e$EzTPvV@aJWIaEGDU9U3&UjoEi|Wyh9neW1eUn;jw#|S9)9-Tw
zod#%P_YIfl&8gp|M1z3f(-Ke*Aj8Rd=}q)IV<=<{3&xuR{@--Bnkw`|c%33c(X7NH
zzY3&bI^YDhL+H%mO@cYSpNW*<@M+xhTxv4}&q_dB^Mu7~^WA!~=H&DLdLVxVB2DsS
zX>pa%>vQE@pEB7>ld*SoR__`dL(U#rlzD=uX}|;_#0@4s5<~zBjdmv-KSJWcVE4-%
z&xPYbDHaOx*tn4${R`Xe!r&*zwB8HgT7Rvv3ajLirxYIy13drOJ2avUgW5i1C{axP
z7NgwA75?D6LVgOS5C-I%=ldATMtBjIW6eg_a!iz31hv3#cgI{j$aW$jyg_;%%O(AX
zitX&;WNK&&W15x&P7evf%=RB&0ZJNgF?|n!nkr>U-@cE3ov501;!^(0>gWCQ@6+58
zL8tH#pX;;*Md{5gzPwkEf6OmQJMW=ccm6)N-P~`B354nDt8@N8Zv1`jGm~jXJtiYw
ztKKIW&&iNuvmYi3_Rp(D@~e#&V(1+r-bmb>>6UNzNQNmy-a+YFkw#)uM%qYON~K!>
zW#(e1)i|1O8_{8W3i0t=I|^>@Bx?^Y&7gpdGDpJy>svw3sLgt+<fX(IbO`z%--=s)
zG}GGpWEv=2p0jHqO<6AX<9J591QI)q_-I!Mf)%M=)6VhJfc6Z&U(b3OLQsG|Eqy!9
zrx&uS6j_13!pu4ydhNw^+iM9u5;$mXpjujjEX{w@Fatcy$Z4!q^n(DG_mAheAeqUN
zR*P~;j;Npp3JtTxJEBcZQ<g*l^R5L&bQACS`*)-cNKe;;l?Ybh)zm?#)nt`3*rIUp
z1j(L6{J)?(?LB6gh6wlM!-{e^zS6gXdhzyNnjW)7k7?zwIVyKXI{IRpe7iU_BhfjO
z#TT+7+*l}RW7<gdI4*!1!Lts36Yz}#V(yEjbzsRUrdS_6Hcbm_{A5pAxWa9+5AK<c
zzmDUVMq2C^NcdTsT3i)GZ#<kqyw-z07(HB5C`G|8l`lfE`paq`dOr<gB)bABl$rqk
zp0Vf%W+Qo0UK9E8Dx9xB>fUv;64y$8`KRp3mt49^JW8KuI=z64<c76{55u1NpoI><
zwCMLts>ZfzXX6;kMr&*tE$NmW!OWv^?Rt<n%_v5zf|}k}k|3-YI!d<vTYABVbL1m9
zPI~c-VjxwgoY|!0i7-pt`W|!Q1I5YJaNu6N+fJR@-ayH9jm8#<xOJ@E(QN9=@+0nl
zf7Xp<PZ89Sb8Eo(9VVO#v(a43YKMngGm!QhD&BE6wky4T?WBbao4EFcg+95Cgg3AP
zuJHyfh)1Zm9pa2yyS<pZ!)`D+Fz5EB#p22YZ?B@M7I5ytFjEzHM|OR3;y|fwVJxbH
zh~!9vVNM!l4>kuSlX?H4m93EjT5|+1Q_RGjn>5hxAY4GXm!aF`|6%GIm@|vEZR4b4
z+qP}nwrzEse6j70ZQHhuj-7ODCokuocdKsI{t0u>HEIq;$UKQ&ZY;p)b;xKeQkt$&
zrOIN{+v^mAzd6%V9a<;uh{w^}%5oKd`H(ydrhFhSE~1v7P<uu-VG?l2_^Gfj(vEON
zIvjPn4glLuKbyO9rlJwj@pn|sL#=og$UT9r^vx%woe|A6jpk;oiAzJV(lU@Ra!5qI
z6a#?lvyq%qQWvK5qzloclagja3Z|GW_)eoLVo{RhWxRcs@?K_^4ihxGQ1w!u%o6&(
z^e3F|LgOsKJOBDL>lCn1<Sn1F{;?rHH4x6EvH)kO-XaVkKY(o=xd(p0I61ycb1=|A
zsO{s)h$aX}$lWw8Y`i;@O;#@02FZ2M-v(3~Lh%t;|E8<=Po!T^U!G8ILiOXOC`(hK
zKu0_$^yf*`pr&om)jE^EdsX)re?>Vv=(a9`4(8~`%YXgb?d!?+;eOg3ofGWD`D5Eg
z<R`qTd<;?^0EO@EK@flkJSlf_%dUu-y5_Kkmc^7KiP{tX5*o1v10e)^Pox`boCF9k
z{mIC2K)Xml^7%MN_fNZ_Jtz{|n{z`EX2W5Bwe>D?Z${e1)W>fYG=c)NZ58~f#6`f)
zOH*;->Z|;f=`z~3M(+r_ekO~8g?2@d$&+1iaN-)T+t+Ou?&hFF0}4`gWhX_|b_{GL
zId$<ZNGrPUwRw)$kLW4k%>At6WD1b~(vh>(h@{j&+qdS%8b&FcDt0x^q7#f{?%PJk
z{p2f_chsG?AuEY2jHuRD05P&lnae_)1vawFW)#0Bc$+_11>2uzh#SlkzPjyBgyF<=
zLr_Q)3N28wUc9fq2>e5n3}2JoPOL9lVNXEEqe28_+Hgf0TrnWotSP~F1p|N)mTK+C
zScGMuooZ4(p$@L$gF6rLs-0WCt_w6LR%y{qv&yE)`AW9hpb7cSx;_RGwMv}tizU%t
z-_!G__@%iWU;7B=WS#q8wQj`IzfbDmk05>Nfw<N+K^A2xg-ot*rma3CmY{zc<^Q&^
z8J+poSKyY-I7`ymTe0dqynz6(U-><*lNkWQ&d_sb1n*SrNMFgKE8&Y{ZO>oM3dNS&
z@@*5Ez$#enf1Fw|9+k@JD1y)+#Ay)qN>7tGHLdt1g1nF+X)eJydjWyTDhlV%&rZ(!
z9r9=0Tz@8AYUO$-NU~&O@58)){Y1>lXI*3|PT1av{n?AD%ev{-Lns71@px)Up{{fe
zOrD@9Pfxa#ciE07t7#J9(3HyCr<hb?V=-t}uFKk9*#O>YckdM9|9F;tQ<c02><7d=
z1<-=U!XSY2G$3LQd&6G`Ol|a1k}_(4Fny@{eFx}C3kyyYY!$*bZ-p!E@m$>EKRM%i
z$NJa|C9T1b61)WD4C$KzQD%jGqS!>wiZFQ+G!{cwvRFEp%cyC!PLNEcYL+I|(KIq)
zHZYk7F-%0q!)Lj6>&=?8Lk7B)Hfx`PX06C>B+a#aJn$^6>wG&{-3|TjaQ_xGW}$2v
zfb^(#z`Ff{By^iU{r~D2t~4cA1k^Oo28Ip*M)fbR=jRgw1>OK(F0P&sRUFeMH}Q&7
z8A*dY1Sc`cTTzVa3MjtFhp8f_jNo*?cPFo!zFH*2t}I$xKP{?~`iaw+?FZ&PC2^qV
zf^|<OH~-WW@;)Ye3BmUy40=q2>C0rFn41r;vXe~mVeopE((c30OW&8r%losPQetob
z(VqqTg}no>x0BT*fWIJSbY`$2yyM%a@7wLeNt?Cur3Zo=mgx-JWpPsboGcZORet)?
zUH#vZGz6s;UU(BPa0MxHNy>@b9f4A#!-MW{F{sI=5hTj4*GiU)yHHlxXxyEfs{kzn
ztS_d2@-vPLeNIw<;Ry9jJ}M=y8Nm&}daGPrx2c$rqkbJbk8d4G=3%O%E3w<<1>prN
zj@q`G=fCHJiCVGYE8<uYao2nnizaek0V!rlH-jF|4ZRkKlrC_OKANs#u=4W?BD2V3
zQBhsT8&`nNNZN;e2@(k=cN*J1sAc>dl?>Ze;XM2fpU9Kb7O}P~;<nKaw7W0>$VGe>
z@fLMEj%jHeP6O5A5p|2f!<gxatuRHI*^x!CUsDW80$HSRrfS#30`+R3PBN*8tT$0w
zrRhMHhu)Y{)93ECCr6+m5{+5MrAnHbo*aALAPFYL5!SH9m=T+22gm$ihJGr^y2}WU
zVIrUjmL+fIsIOD0ij@BfMCJ^j<fF7NLX3Z-<{8HumcJA?Z~M@IxgB`0m2yq#f_MzP
ztKK7n<?M0r*9UUPZ8jsUF2N<DMvHQ|$;ORXh`DeGoadHazSgBTRMBbK{;;s^pKN|2
ztf*!B<=5Kq{#)_=%c`t|^eC9t;R~7HBgRCeWqQ3iIg##CqiSFRQEVgNb^Khe>&bKK
zJ|N4+Wv&wf>lTCR+kJ~`@?z7G#Z?LK6!@RWwI<<A1;eg7?-ij>KF<0Yyq5$ji^>Dm
z{AlE^xD1yiI0LtNc2@a2zIJ3;dKntJT4s1H99s++bHb#sdmUMVQI2##{#-#Uf*}uf
z1HIdMd8`<!{@$PxRtg`W+IHo9p7x~844RjsC{wAo&|7ia3hG?gUw85vYK+xvNeC1t
z-%TzpJQ*nlPeVJun-W$<yu~Piv|(J7^vEylAoIL`_<hb0!_5WK(8?^`=srMp4RP!#
z=1FcTo$;=D&!r@qRa=gv`ts(&`-@6H@4ijku%0+o#cBvXWS}14Gpn`%^9wyGCYQNl
z_vg@c3vs8bs5W#bKGYV0`P-n{Vk_htJq`w2tBFcsIuEr~4oi!V-$;xlHPmizgS6~I
z?_K;nZ2!I`RS5XJo-QEB)bp|M@D~F`fhSJABKII}s<p%+J>*Q}p`6Z4rBX6_i};G~
ziXDcO>oiz<)*v~cflhppICJj2u^t|=^pqhp!H$I;UJXZm16q5}SBZ<*ft4|8$<rW(
zT36@S3s8XNh%yi@<38y|vEbASm)Yt^wgkVjXI<8Sizi`<nQ9^$3c%)#X?z%5H+e+J
z%2?T6_ZB#ti2>e#8J-3qxMb=ndS;i?yZ#<c8zP@LJbeZb^-#rIY;L$ka<pFik`6-e
zCbxneAT9r>aG}~nQf+<L0)BLwvAfNzVsOQO^ak^fH=Hhg=(?5L$PD~eTaROE=&&bj
zgvp4oo5MCnhCD)^;d+`C)Ap=&=Yg+$3~Gur%TRe-S`Q;aC6TbDI#-lH3CL{#+*5#(
zAR(N_Oi>OXQ$q=fzTwioq>rvHt{Fm|;Zk7Yy5R_j;_A<A<w#|m9opH8{$)}Fm1bdm
z{Kbc35Ld?Tk1BRZTOZd>Ni|3o5x%A>H?nzH#oDltn2q+MvsRDIFK0{UK9Lm@h&RvJ
z*N)=ato3Y_c@2Uo-rOt>wtqE&{u8sz>M`p$+@=qJa?6At`pxS*Rez1YzIYXRcm&O{
z^8`^YG__yA@o4zO{Wxs}y1-74th?{`GDjc{kLruVsXb=^&#Q&PIX-24w|*iW;OL`+
z@MkMuGVVBa8x-+Yu|O4W!EGyk*&4MRJH^?@J71t~e4!a$)k?RRXU~$FGJy;k>+#NL
zi})UpO%SBF5VU6guf}+Ssp(R7X^@X~%ZS*GXXD@6&e$RwGQns<iOwF5zd`>(0B4@+
zi=aAJ)r(*%f0a*YF+yW+crmEqY8x-ZtjynH7AypVb=H~5rjvLpBMc4sm_skYM#-y5
zY4JCcpOUk!0;cml@@fy2<@(7m+RyQMsH$>BUi)jQ6F4Mq^=31bM1V_R?bRynAFn}!
ziM_?gtozL~Jf%5sI*dE#F;v&Hhf2hDyUxiXS4&AU<Jcpv+QcckYC*_2jH0Pqd7&o(
z<M+0h|K7n_)0`AV{^K)X`EP`Rcn3oo@Lub72g5Ni;P^Rjh}4*r&2o*CfN#sU;IF2r
z-us=I?W)WRDdte9mUNrCIK?<74^TYOFgao4%-ksWaLcLY#f0!)9hFV7?5(lH0HtQg
zImsw*U-lHcBbgfWgVW4~>mF)ApY7uOpVINDm1i+S>#t`J;iQ7VWG}(+GUPQHkGf8a
zi`IZ80I64X%S0#-UUSI4sL|wU<JVzdR!kFDa@Ad1JcvH}XIYKl0EnPwtxf=i6iNC5
z)vAMmu?dP@%^xQg>WVPle4TVRoGA(=B<6hM0n(6j+O4SEt)+2RG}d!Z0%1*L7p-tP
zA(zoEiSeU;jSf4CG$aE>8-0N{8+JgX8@ApJfJG&s!fY&52?3w4)+H{Jb3dKA;aE>?
zN^B0VT5p~~_U-bB=p-;KWEdRWwz>fuEh8lJVkjeD1zk+R=)m+R`ZrgW%^uQJ&z=!K
zcO?k=JoN0*GIGI|mVs3|T$HJV2Ln>K(u?iT!(@5)jt-H+?t0?MY}P|^ghZH6Vui&8
zP^H-a3g0#!4wOyXPuu4qE*HP4<O69%6D<(al#xe6av5qP>$K`$r`e9V1=M=du6EOP
z(ap}5C{}+MnIgxDz5UfXJU51|KN+GFIce_UtT;NS^CGRtBgaWar>;fCze)?itzs~g
zQp{X2io4!yGoID>0-#n}{`M^TOE%g9NbcmHe7n2__*=p@b5zt_hg(YKZZf)Ko0UXQ
zn?Jaz1|gw&bYZMqIBy2BgK4PD-UijdDYw-_3P5b+fr9U0=sv~1gu0o{KO|o4XXeQc
z_zj2c<B+z_^)!ugC<ECKaX>uX$Sr<{*sF9rAwv+wJJhE+RT5=vpn&pe;iLZroRrK9
z125XH3HM7D0ITy1_n@XFtYy)u66iytrnB`jzkA`-YT8uSi4+WlGvXaqAJnZK)WtB1
z$wB5*ROorTLSK8idKKd|G*+E~Q>$;r<@V9o_A-B~5L$q~>IzP#_kni>s+0!;)0aWJ
zZYGUls^;vsKWiKzh}Ad`v@rw$C}tUamN4qc_vDj?U*{?=Pu;<cHl8JP9S{0nh%H5w
z8MA@;aGRFS7#~z!mbO*t#aT6G?d=6~oyR(O=en>kVm3`dTRh{vmpQ;WbT?>~fUv3E
zvs0*RhM@4=qdf}gOiM(?J#FX~eXeO61JnZ=@Jkw7zJWCY<)C-&^P+#5;&6~?&T^c$
z;AV!>K+bR!NY8<_W_3qWU+=ZWQJ>%nK&dIKzL*L?sX`Pgx9u1dr$(3d65bzP15Y%0
z=bM2?vUooElM$%QT9}Z?EtATaQV8x3JTMSmC%6Zg0Vo_kVT=!~()O;IP=Gno;_U@d
zL6|twZW*8{(r)`0QUL_HgSoZ6d}jkca_mHAcG;DBWbG7|J%26n)NFBiMkm&=p`?JZ
z?Dxc-8yQ`-S2onu@n}bqAoIO!g?k;HCr5p=%@6XW=GUr_0>5z|8rJQDBDBea@wo>T
zQY?$!PiplW#~J9cGTs5gN_uQQaXyP_>}Y3tT#7$!W8j{#5CAQNc|j#GY;e1j-Xrn`
zxe!K75dw=47#&>mVz_@Nm6{4&wp0;PGsi+wbY>&M%M|v^&G*Rlzo9Mj$ggnS;p8o=
znA<QRCjQXpf*T;@VxGycs}+VdU47~@o^jT+-pi(JubwMft-=&4bZU<nCQrG7#7&`@
z#Z_HW=?w%3?*aneT-2qD7AS{s5ob4D(;VJ%5K$Ea2x6_OpBi591g^cIvB^Zajdfy$
zg{(Ajht*w{qL^)H+Ajx*p)k2P6ajuVFQ&B{7aml!glAtv1sK)Au%3Ha(U0olIIQOF
zGd=4`)&J0iur4+DBbFJQ2~S^A^UMxLIUl}1@7^A+4FUWpk_lk@y?(sI7_j16802--
z@!5V~e7=CU?$W=3w#b4yHdER_PIw0)UW(+KPUXaFL$3weD~op&<*sb^)xb&!zqUJ*
z_G}`xEQ~n9jc}iNh&@iyl}C_A3l^wonTxpSolcWi7jP3$j7_m-${ksx%2(>E0!ukc
zLEHm;VF05pO-IfRHbopED67GPF`70(cR{LYO`hmtUZx8E0-*oYS>Y!F&#>C?m!|0}
z-wk7{HVu_rfE$&3kRf91f{h2|>kILwwoMS%X!;T+M5<ioStaGckG~$G@6p7G1xvc@
zA=0==|B+v4$eW^7S*?ZiP6!w{Vu6HMdi6B5*8!ry`!@4r^ddngoc;L;P(lv${>3IQ
zI?^b}U*-5q{X}L8U0p{bDbMYW+=FR>AaF~TBXC4a`*IuJK?ijvP?#$U{SaTja}S@u
zWxjQ+7qh>C%jMknc}mzJ6zlVGlrv{b^PZu-p42_{o#9${-6A2_o{z|-o77Y*rz5M~
z1OUb>N!1uysp<o(T|D^|N5W5bo2It+7!2TnoUzwfUi{O*h=cu##o)!s7qPi9ekSDO
z!P1Hiv!qCB@gSBQNf)Cu%#$A9a=WA2Jk6YW5XI>*s0}_HTu5RM6WUi;TT-GVV1Z!8
z_M`U1+&8Zy!AZ?vb^yDci1@j50?gMeX8<yQo4un3am&ya8#WuxHtiU>^W3D!Wx{&#
zR--u=q)CrrL~dMfDFuiTonw?UNH0groZ6-od8ToHqqSyC@|CJwRx4ko9%0yGzf9pS
z)zWvn5O&lStl@?P);%S^LSVEfJoTYk`=9YE^7U*dNN#dQ;zA}Q2!Yf-eg=I%H2{2O
zkDjb^RQya1I$=c@(lyY0fJYy!!&_me%~e}Q*LCZRAlMpHb6z4K<5W@0WXF7bfm@46
z<ScaOhqjbKZs9ZWCJZ&EO{m$p%_j_9p;ns6>?GHt)8Zx!0{$B-6>I5>;k8`vPPUlt
zO)az7AJgyIn6yu2rjuQT6ESHX11PD6MqcNk#(6}P&SP>d)@QJo9ia}+InAcCfy&F`
zl#+;vqgj4CI!~W?NG(%-vKg|G={@4+CH~b=79ErmmjBTyu%IvNQpDS0J{QQ(cl9n1
z<UBxmfCaNfVH4}I8X-eWFWJ4tbRR1P1kx<{Q+))~l|*U?j%PPYZ;)Sd4hWGk_9S}=
zis>GcSp(eX>pW}Ggf;O$1Q8NQ3RfQD5Cpvn3fY5fS+s$lpG>u^;4-uZbz^GZ9px0V
z7!u&=wb*3a`<8TMyt4CL5OPFvivS=lB^eF;@4ZAJhEKgB9i#~bW!Oaa{(%LyDG-jS
zp4k3t)a*~;=v0rbw9UjU0`MuDuWp?*tF1PVhuTHb?4aos8`E}X7^oC6+`O~k<Yw9^
zG+xkGyIGQ5GzG{ky4R#nS9P~HGvv0j(#M@&58>F$)#Tm-&NZ8E^igWK7rnF&;U2Uu
z5Q>^|HyBi-Y}3v)7k}Tkx#?U|^FirZ_>#1E)n@g7jrm@$Ri`kH(V#Km^`pxlSd<Bn
z$G<eXz-1-ynx2V>fjpjIba*^keL!?XBWr(yLUiV2{|}Z-+vB1`O`9EMpaOLNU$?3{
zsuZ8Up}%Y*)tpY$W@Jv}RnF7g=XaR0!bSsMT#P$BK1GA562VR~eh-}2%cbKRrw<ni
zoV+K5MQB_t=olnjZ=TuQ-7ZkF7$$!Y0)WrPu8Nn!Q_@D%Gwb8>XnK?ajDf}fD4`Z;
zltKb=bqtr)6F~R33&JebGlgU9oh)|pEw4HjHISFTCk_Dk@S6+Kst$|+jA(rCypQhn
zikNKW8RI~>RY`X4<z9!q1o;1CBFKwRSPOi-xW9P46jE%kFE)^W*Wvs5;B+Jiz<3G4
zc>UEr#_}4WHW$#uK^;-R;`3U}I5Krjm5^5mObS@10`z9_RVkZRB8n&|87|STB^pXc
zIG2z^5{%>yec6hxIVAIuO1US;<lHWsLsriz3E*J(jQO8Nq^b;f@11cMY21>QLT%xB
zSSPf{L*f$JC%a|Ge0YK$-|NNTriyIGcy9h_bKK{MN9ksXVi8UDfa|`$9sg%3K*@6R
zY2S|R3V_CadI$dfvMH1xnBW8x>5!X8$xgNVjCx5O7#av|i2=1;dmq|Oyw6u(FTRfW
zG&G`(hUqXo62R<bMdTG2_cCN`7aBz-D?z`Ods`N$Ge+g*;z##Z*?85EsTtBe-kQ%e
zu?GIw?Hgt^jEH~9bKL2-;yzWb_l&26{nl(q37BC0Dq5TpFJa0zh9J#v0>@eHzxT{{
zL~JDTsZ;FGD^wga$J9bbZb4xynUcbGKQAd&Vt40T#=(8}-j+z#6`yjk;7xWfPOHGa
zOE2K(cfoz`LpGOEkPC?-6fa*&fKl$JMB5EfU3hh1F@v3G10P`Glcd@`->Si0r{r>1
z03aEvp;IEWCyqb9yCl?%W8!dd)37A?;11F-&y{x03L<mPGdONJ9-#*9E6Wr{3m8q%
zTID=0AyRd$mSS;vuAY3*?-g2}7(+PB>{hZ+!|`MT+pu1c!|~)2p~qWzh}hN8D@P*>
zn9-TiCOWw3H4!=MWVtlea6-)4%+{~40t9LWLmpRU9_4V*0O9(fV$I!hOgfpwcg+li
z?>Z~4&5Q+>q;hC&cLqdtQJku{BjzcTlVX=66+>w%Vf(xd<F<Ed<(4F|uF)xEn4~8G
zTBw=H92J^w#?os_eq5v0D7u>@xLBpQ7^}IK7=P3?TpT{7)c+~m>gs(`jmrad0F?dQ
zu2rq=hnkFcNPp#l_R<9+Ck%Xo8%Jbp!QZq<I&478OeCFa3<>9cSW=o~P=fTNqnhE!
zDly1*?P9xTRwxz+I%Biku|`ZS=ZU?_uoQa5@b2dTE0CLLxqx_;yEse{QXi$G1-GRB
zeL=&GWTuw~*Vp1pE&H`)qU_{}15j}@t%=OF45Q-EqK+Za@Q6;f&4iqWS@G<-P-dc{
z*x6UHIpR}*2@mdZ@KNHoag!FCQLgbsc~r)aW80?_XK3e#mxjD}=0G{eWUF%V#d$j{
z)=b!JD_o1L#`8#hQ8;}62^cXwUipc;;<%;%Rg0?o0prOr#=Jx4k$Yz&PH(4x{7hqH
z@q^%lu>Husw<$5E*uPtEv5JD-msF@)ZjWf~-vdj9sXF@ir%Dvka;6y20SHzJkuZuz
zqY!7&+p(APSX(sH^Xr?C&EiPi&;>hj)kU3p0_)qSWm|LGoS&{bO((hxMP3FZEUn}`
z6?!60wgQQP;2z#IU_)`<+k6kG3=jpxJ&UE0>PzYYIjngN+SS34e26UiL;Ihx;lEKi
zlJ2Oad9ntv%{FsJ(@{Pi0nQw-(7@(K?unT6rr%`~>yrMkKU*owWDLm03vMqXx@T(f
zpDz2Bs+<UIPgQquMvql%H7h1*%EOjDLPecWgDl<ksOlQxPhrzF%Cb6YrvL&KQ^Uc&
zFq;1R!d7II+RLz|%(dC@r=Bj#^}g8j&n0q{O|5nE;iol%p;meqpe>*SDP|P#R?3Z-
zHyjb&O1<`QYKc6t5&9B*nVR>T><%jWw~&)yW=c%3>MRUB_qx40t1|>huO^=OSCZ2C
zBrhUa=Vh6#=C~UgrSzVY5Scr?S5=uS9qWpp%v+^azYI?MLO{e5q38HCUhY}sO3=8j
ztQSQ&g&sFsdjU;8pr(D?s{qxsES8j8_ct(&hKgQ+QrH-qZbc}WkGeP8zT*&gY5Qsu
zn)(_}N)3DPcAVs_+#mlN&qka>+YZ`_P@n^FxdIDMIsX-T2?H5#>bO)9GNtT|(<ZAF
z%Y~{lcb^K5Vw%o3@nO`wossiaXQH+(Q9&{v_fg$S%YC~MfMviXy@Sj>nV*c;5Z<0g
zOgR4%?6jVEaY=}l)5d1RzN+|r$N@L=W_A8t)^gdz9CKl$fZlNIp>6Eo*IGSN^9Vk7
zJz0xV`8-~K=*f!Dh1yD@@M4YpY0cGbDl!^E!s?4>WTylJjIniSYrUU}PE>Re@f=Fg
zUn)1XvyU=YxVo;!Y6q#yFG!VL=DGhBDIz(tl7lcZrTJ0-|JTu?%K`8ET#GoDb|vB}
z7FkpzhL}f?X^%mVNUCS(4q8}8(OBIXJBk?>)za&Sz}Ca!_9ME%;&39k*t7H?ODuqo
zE7Eb6CkjbS0d;cAhd71+{%*hu)(s?I9cR-t$GGYI1tt}MoFIc5UexNTP<$mc5x|P(
zkR*7lU^WjBEcS3P4$!_MDh4Ajr+?jT{{9OhbFHOfh(@4JR+nO51VKOkq-lto_Em*l
zlp3gc(ASJp@zRX`OZgU)!`VHVuv`Eb<43^V*spQLUm?@}odH5Sc{50PnAjz75O4FO
zkN9VE%6CveN-BtbWLL^dCq$J7S9Df*-Ma<0+=Bry9C&zWqRLA{iU~@D#Pr;%7{6c_
z(9m=RS)~CsNA@eDbGjI!ncu8fNVyd{U&W@|0$ST^h(@F$n2Fk}{4*j_zoJ~m-s9ga
zb`k#gzTGrkp8&}nMFxb)zfU5VHt%50Ow{Y*BSUFV1vn!1<!$nKZ1O&W#>WedG(+tK
zQD9+!(=WSfo#h)_m1m4NG;4e~cX8hR{c__gNMl@SqBg_B#3utL)3b>ySutZ_VTNWe
z#9!I$=njJj(~H{a!g7pI%D+;yTiZh>rWqlM6^B;?D2&AN^!62_GDET8nmu6|J%O#3
zH(2w=eMMk1ObOuNiKV<4*qWrf)6lh1rrP@eR#iF8zcdLWjG!)V)LXVAhZC7$1?4-1
zM;t8KFpL-$ro;?DEgy?1eQAR;68m$Adv27C{(8T$!Wz)ZqfcsBgf)7OX+AIv-t0Rz
zXD}#$*7kl~o-AwBs)tp2=KUVLA3pncwz%i{WFOCJiJN2c1qI{q362%WrvEwFXfYTC
zI5K{+?(yR5dbxYOx%@a;?r*ff^m=(pYvh7>_k+ig*SKF?Zf9uzjlpno5*2l1>(`^7
zx&1KxF@3vt5|wMf!@lly2Y>T{tPnaNsk{1i@!(p2b?W@my^7boc^9jlCX5$yOCeP#
zSlCw>FCio@B+)N!bfGnpV|JnCJqw8gNb8Iktc)s50c;5He;Q!&8_@1drEN94GJg2}
z0JVL#>fWxbS@qiTrPEjQM3dWB8@k9xaP)+5$UzI>tzPUNZJ$oe9@yMOwS@ihEJ>~e
z%ngGazE0$q=Y~M8ib46sGR-?%bUaI1E7W$QV00!FTK7IXG5f>?&^1S_Un%GSJT8}s
za8X4&rdv52Y6|kc49*K6Ahmls_Dz1;2F^JLy;&~pN9>V$ra5)xB5FwQ)lE$P4Oamj
zJO+8AIROK|OF-^dopyHi_ixM|0sT)$Csd25*1<RohuNI2Ki{k$2m6^Ax1&jWr>0J<
zYd-2R6G#=p3tnJDyjj!7n}0I_9F4B3SUHr_lyE1L>vfzuc)fgIzk5EeFlDS>4QMS~
zAIAC7DFuFN>f~>{97g8~t%pD`A;F9~fCY7l|55x)-DlYe=R;3pr7DBz)2Gu6Lx>L&
zWH9wv$DX&Y(saFM4peL`Ry@`orbeFEUZ(;Ds+*s`Vg^>kG$H?5`{nowIHY)&9HV&u
ztMRwy9_|i>h7#V9kK5RpW}YaSO~RUk7;a2z1kG{K--2%94QO%;N(pk*BudgeT93q<
zz@#P97ZYU;Ez}+dMNyu3)%~xal!_a+igsbBCeyt9uhHa!0P1M0QB}miCrhGIRYG|>
zgm{0TtMtC=3UO6NikNvR03c~iyBo{HI@}YYdnUaGdLf-4Z<`SYqBsUl6BM53x=v_r
z%rG~lYhTr8WlOHl)^KVb<iexi&|O<FtBRCX197fkn%WHq!7y;(k~d=)vr4p=1&Q*p
zEE&!lI$57$90#SxXX`m-X8G&!`rCg#s8flvo0E&HO$}I<?X5->!1Ow^LT*W3hg^5v
z$yUz?=AK?eg$jQ0W-lkqx6drq2%-*Lt1~T-6@xG<6cw@J^U|~Qk!WO4e4M%ymZ7tu
z?1HD&5c9Di{$=L_<In@CGB`ljLAR7a)!_Fa{JFSy_i}S;>bKqBfbQL}vrv6;fsZDv
z_~&nG^1yWX^xmi&;PK(Gif0^JdH^rft%NjhXW6|kOYnIdA;9>yo59+d?j!$6DDC$z
z9VAp-R8vrrB}JPi*Q9_{Q7#lhoIwm0O{3MhxkG5yJPLvEwqFseP4oe=oNww6_Hwrs
zW+<q9J;kY18LA|KZxVT{Y<(RNW;}TsR2*}ar|)?hbUF{z020d-q^1cEOT`h9AmYw`
z&G9uGgLOxB)34DT#~Jj}lgcBavYpo>@kMM)1M4oF<OuS3o2@n)%K5cMUt9NC04Hx|
zTuvfSMd#}u6riXYTao~xI71xImZ9hxp^Y`6&Z}c`HBgo0n9MR|B*~BPP1KOg1vi&P
zXHldz4=Xx!0Cn8Yfc*t)a>#a7WN?Fo=`E<E;#z}oDOX|bmnxD~Ox887RDr+{?JSp_
zd(g`;V3wP!!ZnK&P$hM?R$*-xb9_e_Kfl(2#o3|^ZSGcJvhAP?ZN5@JMO_MNRm=~S
z?t9aW<7!&@ned0V3~MXsw9b-ddF;WJj%Hr5Mnf7v#qYop<Hd}r5=&-7?r*EHEYdH#
zuQ$Bytyh|~(CF9m=jhB2y8}IWoNMR5_E<4760BSJn-K)LK0efqob5LOJv_Z0?@x}e
zGPB3gruGda#YQnb-aCrewr!QSg?fEGj}VR93A>UpMLJ&cYw;QUf8PmbX8r}&=BMs-
ze$%4>+A8X(MoO~#<mXiXQIxsc1Of%o9skO>D79r4OJZ(dK0}p*&^7;zC-gAbJjiSt
zy$<pAF+<s)kK6rt43!BG50D^=3ycno4U7@S3gL#fd(LzaN2<&N*MItGh9ZQ8F2<0Z
zN}Zdg$%AN_R94i!u~M(%-wcQ|0}X8L<LZY3_&|=!`7DMHX!}D`Q!*C$9TxKF0q2Lc
znTdy5Zz1!wkN}Ybx76|O4#zbZ{5~V!j(p*u0eEz!P(HcOi_=T{Ls7+@U#uDAcn5wf
z6Oits%K&WPa^5!_!4qb7X@kcm=^!hiPzqG?(vtjDjuY{Zodp8IV${okx*oUG(~V{T
z%M8AinIwsJt(~8Vg(gQ&&b>y(e19v|t-w*qZ$+lQk8|*RMg)f2D{FbHM0<`(x>K^~
zketP<NTptZ4O9D3Wn0BfDKsv(*kwB|duPRXBmfqU-t$)yn@^VQx&s$1dOHDcIMMFc
zg0(g=?qJQmmrO#9(g31;dpT1mezQoS5(A?YDz~>v2dHh->7JbmKq`#7R&St3%vSY|
zN*6<?PH*YjdUo_Rzcl{3-v9lsF0OXJnfD&k>PyH_KskQePwNrE{*%7>b8M$f^V6e+
zN*h>bhyp|CR~>Ct{L3H)2^QFNcJ{vv9TSV-PlC%HLoc9ob%+1!aR{XWoCQ_Zs@0MN
ztA8S$Dv<gYixnT~fd*78sVwRi+3aoS!w!QVM)i3E$|wEzqzd70ZQx!CVR%N<V3ZX(
zZ~mIp`-Z;9Qn9XSkrWF=iie%BOrdud^1WhxydF__YdlWQJ_T>N4lV(dnUA|S+J2uv
zir3Hy1Q!5{0JNJGnesjME$3@%j(A{XPJv))7|K~pC!y2C;TAs2?^x|VMXe|pi$s2P
z%K8s6FQ(eB7kO?wBel}TV@uO4d9vv<Cit}$uU0FfNc&dnKynri$HX7X^jVUv99%2j
z0F5C=zGEmzCZG>x%4NM$+d~OPvNq3B;m9!omK30iYEH>wUU=x5fJ>SkZ=X#+Z!MKd
zz88QMX9}qm6?y!2?ErpUV{crV*;bxeJA2bmlDoG4PcSU)wdOc}3b=B%{_)sRCl#N3
z%B<ntK@GoX>$U{0x*-;eYwOTAKy=nZiP7HdvfJTSl>2?(A1BoyLL(=4BG#|)f_dwY
z&jx@iC(c7MP9o8!Ya^B$cvwYmA#;AgEGWRn3d+NazNa{@x`FPCXh36K8yU^vXu5qh
zqK@WOuApDWFtJ*&BzO`RqvP#T-J#nZO?oSzN1PE_1@r~73|^WvEO=3_8Z@EhH{oqs
zjCWvw*Iq6Uabx@N5B6*9%0Q#pRNaM0UjuMb%AOqJHUv(js*v@mX=+g&6<HdYIajbi
zNXd19(8R|Ol^5xY-dAW7T1-srZ#Pb~4`d!5>*vlNJ`mCPy{JmK>&ekJB9FY|7f^%}
ztuD%G;fEwTo>0}2Tg&?39G&U?X08AI4*_Ni1Wg2234VVhodB3LH(9l^j{m!5JqVzN
z27Ap6s!-dXVqBoXsy(rI-`!TPgA7)VCz8%edG5TgL1Mtx=E|yvOu>@7b>Tx^eUxhD
zn!TdliT1B}G4*jbKS^HDTyeODb10MsE!N9M6F^<5;*nm3UZZKZL|V$X_<8VtbG|XU
z_)X3(Dde2s_4+wW@hu!^{FhUD3m1@to7Vf)_;LGijbw{VW05s8BoHZ)a;ffD1Tx{2
z$p7O<;WzuyK6O&cDA+`iZsX`kOTi15Dm;eXRO@GH_^}ShFUZE%5;$;i@bhhyV>PjB
z{*C4k0>K3~o&ff#KJHcNMh9=0E!2ask`fN!!EB1~G8ad1QZEbce2_dA00unU7n?PP
zD8eDcC)8?`#i|UY9oW|wtbMqKw6-3E6`dqoM00>t?S~<!7i~HZIjmmJ3)hT{>8%O?
zx7rsU>7_OzPYISM#aEb4uTm3F8>Q$3CHR1VO2@-?PhhrXV1z2V%f~v0<MTU3a)Mu#
z1#4_b?fyyKeoO${?u~5*G6(3!cy{W=INA(g4y|5~BIdL<w>esPpwvSlr!uieE&aM<
z9LR{M$9Y_>KaABX+~8M>xv)cai^{=>be%?8l2w~u7hUHnK?o35Du-EmJMlhC@lDY_
zqjfR-t6i+>W?w*T!Ru3CIC#H_1pM69$(q4+0{?IYI+OvreZ5R21PYJ_g$T#dXIXNR
zR(QTX&U*HC*>OpE6Yq*o%!r^l$GWOMvx}DL<{Q0uSqSzAO%^Gy`R(TK42$T31N1ia
ziINZS??QPC+sdunw?`SI=n6|t-j7I;<b4l<+rjVVuOWSRLKFQWAFV;DB?J5=vLRy5
znVDbNNqkf24%438<N$cvv{UD42Bdu9&UegEwdAm}iz@fK#ri5P+0$kmNUuIYd7Het
z(DP&nQ`SUd%|RDsG3cv9_jUF58>~<&cGk7W!%$qP-o3BOP_gPka-~(_yj-WGIbkUL
z1s}%*mR>Zgym&Sz4Z$R9F<QwC;rzoG&r~DaF0A%Pf9VCQgbR>1-Jruc7~Wisky^1z
z?ne*qw2>ZE!vtvtom3EIk%G&dun<Gry`nu6w@^*?sVzbw<e_*T`u<lmu8XV29Yt^d
zY;8NH0C~JZEuiPIFeJvUfiaeC$<5bw-y{7zgxuKVkBS8U6}3g(W%>CMGMx7cs4OHf
zb8rn3ib88n?JdBoxR>C#!OVK`4U3P%#n(@2c0{AZTgP{)?to@77Hjo9S^rUVTwnGn
zJf1r6Fyanp*_csI=YER1TMwg$eag`8)lA^%=Xo%%*i`-YZuY$(P126#*g2y@qCK!5
z5V#4A(^tOPXOoZl_OvNE_qlUP@LCq8Y<@1z)+VQ`wE)Ow3#Tv*FmfUEM4zjN?$P#)
zMi{Qz9u}+pDYcrr>nwE+pVKf&-Gdz>jfkONDgkxQFkb24<(ZKuLDzu*N>W#5W$S<s
zy|XNp-(oa(yIGJeI3K!5gCUF4xT7Yb-P7$fS$uGCxpV(boRR;|?sXZo`(qGQM5S2K
zs1~OQ%>ziR#bI_p84k#C48Q2{3ZHAq6fy7|x<z$$)i38{<S#OjU=m&;k$|&N!y*B<
z^3?M5rzry;``qB!CocBwGhs!6jgK9UoY%tmaTLbKYGp-q02Jxk@`yWII&@)Oy15J#
z^m&FmttDHqmD|WTSg*5({85P@2`HErn5l*p{s4yZG-#^vc_XkKUhhcFfnNHgV5vtX
zFCbe?Ai^Q#jUmQ(F4K*k71D=bIj&@l#Xsybh7gw=M#r}g=weZQAr;Tw1OP<Lwue40
zgXcFTh1Y4syL$G5nZp#nVNf~dsz2HS@II?4-M)A1^m54-Ip^!fo4GT&eD#wcr;Hj}
zIstjgn6AQcieX%8>(!(|5uOH?zQm<-`V#(wy<YhF`lyGJyZ;1bunF-QZ!ih3V0lO0
z2{>-r;A%N<Fke67Fg^Uq7xu=Wc!Xt<WJ@}pPzJ`3+<yK@qp&SNZ2)BaV!{N!>n_pv
z2q;2cpt#!GdGD1Uf9`7HuZo0FSA9)kVfn9Q+ifD1n^Qm~HvOwJ{m!7bA+-ag2-n<X
z(=NR1zD^y_S24+HP3oXidH2dxLXefxI<yi5=AAz5{s`qE$~%3#0b9YI6#E+hX88vW
zC4e)a0JEksZGxhtnV$dP8&3SMpBTB)#yV|q9EXEqKd<+-&&G2x31wT6!ENe>me=<a
z`X0u<iH<oi<(1&$)z;(ptGEp7S((Yyhry25$HH>+?Z(?c^aCrKKN2G%mFd1F>omv%
z>mJ8yHt54sQlXR%L$XjLWVs6X=2Q)*!ucg*#_}YnTC&!R0t75!08WInZ1l8<VYW%^
zgN4pEaSIfK{Hux@(-fyK2)*Vukugv<M6BPs5-ismYbH}@WTafA_EVr1gTv$`zr3vb
zDWg#E(5v0JdA0CFGB<1mP~g=3xaGuIu0a_{OZ(<U?>rD)&2n+dGV(yyGpOgPdd9^2
zhdg{48QBIEM%`Fl03$8#{0+wWngb2hsOyAXO}IKSI{3wr#+g*zMcyqOo#?!-T#W^~
z?Z2V8tH=IC;AV-i;~%x|VP@LIGC4D!8lKEgdG<*XGno@twvaWMn43o{w9O*F@ahVj
z&#by4XI><!2*|v0c4U=A+S|!%IVt+N%f9_yU9IAwz*eM)18fu~gG&D^-DW5m3txY(
zM;`Y6SIyY$R{JiK%F<&sNt^Cs<I1BzrOYEeH>{>*6lw9kwWM-l`zI}ViQ@092Az?p
zen~`iD^oUvlU>8~W+qX;{5gQCz*&1Odo@$OY9Etly!@4r0R;NC+4MEQxLjF{jD^eF
z9b#fVOC_Bl5}?ZoMEwA2I$kbpToyzN<nF8oB<&oHr#$?4gbEDTHYR|ZKErwmQV7OV
z(!Rf~Llo;&)#T1LB%}&F1wB+=gO{Z>mj3#IIqqOO?OQXsZqvd^u&+duUMvHhbv=mp
z$VM%Igsy*(UK<LLETnNAKAxx0^Cf!-WyD|{4d#&a0k9bNoPmi_Qp2^?G&(iWGHhXb
zr&W-b%a-zuXi1Kph~8lsan(0iP<1j*E0OudjSi+|S)B5@7`%$MW^?wC$R26q>p48b
zQr}>z)R2crfqOTexrUeh(>ep)Q2g=eHW@;1Hq`Us6~{zu=0R;{#q3RW>8!}DI8A^N
zi>h{F3LvXc95tU=T>_uAC;KThm-I63m3QmRFwr=|NXDtZ$Cw~Us9%bY9EAIAEC-29
zfGPA3EetnomSVgy?JfWwWW7Buic%%hZ0?ujV|Q+MZAo6zaTK|o%g=d~zsQ07Y;&;0
zu(yB{^!y6^9qkCM&Y-%4i*aWpg9z!iMazL+4Il^ZfWMikp)0VSsj*aZHa&@=j^%1|
zjvdL+8od&^Z~Zj&WGS#bltf6;Klh0>F6sX{#F429N1E=lwqR`%>E6HSw&-Ad+MVdy
zksr%8JSqyoU9QHcde1l_kaX&y>VloXlL42hl58wt-p$1apR3+Eq@oSpQy~;WDCU1{
z15jM$WMC_=lAV4L;<FBJiX~Vs3~Ub6pZdht{imW5H-Yn1e#YZSb~DVcMoaNmgXasw
zm)bBMF85r;LU+Nvpmfh<qq+Nh@7L4GXXX3;7h0g~O6Xcc!0%g`tZFV=);gfFB0Nks
zXdg-OT~K2th*E~c^}JE}NTo&r=Y&cBN#iuVo}Zde0hF=Wp>jNFRo17#s=%AR8^DT9
z-1<Stl%q>0-nai@WPj=wwP4jBv>+tPHeCjFHp88O>)OS>#0JH%z8E!!=A&3LNNc*9
zi?*XBsfwQBy_(2W)M3C6eWvO0zdNs=it|Tx$-<T*ME{>~=keGBcEs*?ng`H9a+^I6
zR@et-_;?)%Ul?a-W|s_+<8K72nnX&Omd@oFp;S4k<}Pjhh?T)OA%tGMoE>-silTn1
z*^&O6CtpTqp~=$u21`%J5y~iYk9V$**D^LuuAC#nYNR8_yRFGTaRzf9?Spl=&dccs
z7!u_CB+mfCLyP%i3&pEI3@#ilf#Kht{Far4!HV#_Uyo1kLKr5^(CmIOrC+P7t%JQ`
z=G_0`%gt2}2XCHdMY;q-fB15GongZ;gKyX!1i*uFCNtgaA;QXFErCeuU3DHtQ(kGc
zSx!F2?>d4iIOv1oX#8&L=|J@CL&R$-M-Z0#JHX!Iv{r;In(-+abaV!|4zetj)XV~^
z7|*!`GoXd|m}7jqNYXt~X!gR|{XcX$fP?-b9(){yPCuh|xMVV3e<&%7*p*|qoa~r*
zd)Kg_py+qY-6izYiMCMIQG7E~Tt-9`nIAZpGCDfHXz~lLSLp#prOD@w%UcRlau<C#
z02srega!u_1WRleeLMV-R&DI6f)|Y33_bn=&)jghz-+YPj8wd|7!4`;`j<b*mkR&r
zT*+<`D}4_W{DH)sT?;c^<-aJRGx2JDJ@l;2TfGuLTu%5r+P&Dm-^hhcV5xM3?e6|<
zLAmlYkls(r`um^A!CaIUI-!bZwDW1vfID=U7Bwe=Zdwu@2NjJu#fRKZX18v#dyO|R
z<^ET6&rEMrj9Duyoru5xXoZA0O>D7k+LQfym-M;M<ty1W;Pn0TowTVuk=Ci?X-2wK
z&L^T#jb{u?AJ0%Nunr2o98I%4u$g4u^r2J9r=_@D?6d5{gXMW27UBG%L+N2T09#ye
z%rr+ziK=aK;7`{uTiD}UycPc9dv>cjCi(|Xsjft^wWjYByH-uLsm=`>HPOrrCxJ@~
zeq?;sMbT;{W_;=?ex^a%*HjGpDtaka+`#X;8piU&4#)GmE(u9475Z|aVJr#`Y6y8q
zak>_#@ZccK+T87|7`Qx}USg~Qfbz|~3J$q?p19at*YYJk5jguKL2wt2UtDqZlI|OB
zt$3BJ0#0BmO#Ey^Y3#vL%1#MNeKOAav7x*5DSo*Om|YDeO7y9mvdV_S2udE)zF0ur
zu3+WZT)gTa+Ew;!0GP<^GrKk-40<*2(>z|G+WhAJ&j?sDn8(AEvjb6b00X}?_SLYx
zCW=lP)IibHV>%+-yy}bwQubWg!U3;0%DkT0GL!k6)}Z<Tv_8{sx2*{1VWT6?%?M~E
zVM4KXzzxfxn?OslCT3W7zmBa`V3x-YpOZ!l7$kHV6C3^RmSq>qm_-xQ?4{Q?ww8U8
z6ytd}PY?daO>fIp`!Aj+Kv7h7c2>it$($QK{}oYPSJd6W;OxERf?)JNG(Tds;zY5{
zgo=G2Szz-m==*tTEOM7f2bEPeR#OmaybCex4zUAi@0%*&Q?~60!+r9AwCC7&#OV6F
zR1#+gcrByYn=86q`FjYJCA;0L2)UBAP3CF9ucE)O1g3~2g;23N0Io5Kg3)@IV`J`P
zcn9`1(*SRiWV3(tm}DI9>oonL*EF`okA+!yJK9l4<NBUh0R>Dt#}h5f!Si|I1zJ~;
zF^w46L$!=%Bb?D~4Uf951H{5KZqjnnx$UWsPbR$Z3ovyWfD-*(+3M(5H@wb(rr!<O
z4Sef#_DV1|Q)oq58bwYlcO!0I^Oc5ZNPrxTvJgOBm>pBf#W~spY(#MGX^IPUeV83L
z=%T629ttlh))SN$f{Z7BR<Ia5b`i>(an|pY46_bR4{JW-mRH>w;lyv??It{w|Jk)5
zY)olPF8C;EH*3_CfJ<#22T~5$t52F^ge>`FNk0E!kd|5aczQ^8p$bCxWwiEhgJ7>2
z$&AsO?o3vyk<Q9L`7g&36V7k^<7a7MrLAGJwb&FixB+U58?C=yLJxZhJ?H04frIB6
zyFb(Y1f<Gg4HSIxleO>7UONvIxB&~hFTV(TUg-N?lf<B|e<nq{*u7cm+R{&&@1El3
z#6<Jtxto<)@#b~lp1`4(e{8A>bvw9rukicv?&;=P#7xghXhcDrJ0Djb@Z|x;fhdW2
zu0?F5m6r>U+rjrcFJJkotxNoh&+_xj?MSKB^3-p*D`FCjxl8_WZZeO+;MtQM?O{+w
zq=eM^)ZdH&td-}WUSF5{sl}LuZFW>l>7<R9i7Qb_jzr}QcN;$<E;wkW{E9CKu{#Q@
zOb0c5|H|akJt>?AAm|%%B+!y|ndZc*d6g+Nfi!mu=O)ViRHQ2)vPX8Z(-<(!d3IJB
z8;zH4g&$`pFjLrzxKKu8n?h+9uw8cN2w-6igyqzTx0Ela`!9d$-*B%zQdgXYteWRx
z2W2yKQ@1ipV+vT_uH@uPBEIN|Ys)Yxl4REy7Fv3yRN>oVT#GaqHCa7;(Gr|leW9e)
zzA)ee^Q6tZ{D`#QzqWpoNU?f?ZCd}jI+TGcB*P)T4%k9T+H~AuyU2}~db0KWaRFOD
zcd2?D2<#TjKr|=Iq?%x}Y;Q=~Nj8BkoXQKJK3ZP)nPg}p%~wR=>F(9HoEu2QW(}@v
zF@G3?#v-lea-F{9C~Ol=W*R80pbVJ_FSYg6Sy<&7P&Wp&O~mCoa|#U0_R|oZnrzRe
z@Gt#CrFgtbH+qpTDmTtLJh>sR-IDW;Ty&h1qgWM+wzraUiE$D~SYo_SD<3(`v7tl@
zHm-b#`}1Db+#ziNUb@UVsZEbeG363F&PI|=cc$d><+ya;i88B`SU8`h0?F6%q<tF{
zy%_l@*pdh+VP4a8zL6B(+ftN_x^5~iLd_P>gz1fWHf)jf;wDZ!S%{@89;2D$^A%#T
zAaENiw%N_NEh4RQQnfU0C~N7^Tp~TJ$eyHGhU9?e(iw0f{^#aVDiQQNupY28q_x*L
zA3iMK%V{H8zGyoO>jWKhL8awPn{1DGka5l;z%d5cP0KS8s6i6RxPSt)priHQv<MK2
zKN@K-6ddk;4w7*Rx|}v#I7DwrAFQDQ$WB0p>PU6R1h_3m{DK>oK#iG!8Maeb=&yEM
zA)DRj+=vL=KseJ|_C*=dF{6O=aGI$GtJ~-@j~S&P0iylY0|t@Y$u(beS2f0kMhIje
zpT-Wb(A+v>OD5VQgHLn-S_EracqoG?*O5dcmkTPg;A|eCXEN@`hCCJaW=C>wz$>gT
z+#1W>tnn|`GU!B><xs(Q4<B&eS-4bfY#R4;75igKpU7?K^zUrcU#<Brec$@^UFLVn
zNwLg`&mUbUND9*oG@L0k@+<4q#~Mw><85g`Bfk>sO+z1ETuSC^H@%=Xot?Hu7;9e#
z!^*T(!<N9p#B%-vWNTfgQXuUEZ<R>_Z$0DXP{BD-%SElf`)<R+W%BR4*A$%zJW~9~
zfhwXg+_eM0t}X>_OP1N)53NNjqgKmXOB0rJo)EvbO8S~<7}&zl0u+sv)1hp_ce}rU
zYQxx<N@9s#O5ij3h`cwN>y|jAE}(xY&Q`wS#%sM#cT*B{^EM5t=WMzr(a)AC+T_Qr
z`Dv{D4(i?Yo(^IxW@d#nlkTf0{6PjpP{H}2hQT~Q@NENO1ACBM-N)q#&b}@8xeG$1
zb)ujM4R?-TV5M<;vS-0eHLHT@_6F2|quUE^TnHzP(cVD0Lz!0y@<+-+?o<a^<|+J%
zj>=_Eoalf$Q*Tf63j<nhQmO^IJAriEN-l8?zEBf{arOs`>+Q?m=`8U}D(Sz$Y`FhU
z$D)J3B<vq465xtxU=~H9sHPmt8bxBqzCc+};#5`HE$D!mJk9Y(N~R~QX1MnOt^$|Q
z(%)xl+SX4zf4S?N_O8Tqs(jCk-O_&$l<6&$ahl5JwSJ20opXHXSmIKp_CH=CAhP-s
zhG^gt>hkex$Pdr^a_~c<AXK}Jo?t{z4ic`+O2FypNiFJ>PcyqNukj-E^T6BlT0&e0
zyL9Y`U4{+)pnK^s&}gBca>+8lqFfv%>Yyyy7I??M(U<PcYfj^rxiakpWb~7R=PfK@
zNqk1qTs_6)90R$Mia%IYi`~OFfBzo<r9fK0BPX3f()SLpIdwUQJvAz=_1xy|Pr5=Y
zKk#%M#dEYD!4hD~FbIvMhHRK|)le{Wt7Sr2(ezVq>Ye{k*is5{z)H>*gHrZuiZ*}o
zw5NOPjx|MT6mEgt?@?1E39s%|QrI=X+vAdQeaXoG!_e)pchYKDmuUMpt{bjxPuA{s
z`r7mRf&XsFC#}8C)=w*y5C6W<Km4Hz0M-Zw-4`EL;&4cXTYx#=DaTvrK9(R<7&db5
z?KnP25aHj4xUThYU3}b#tH+IaG)8|+@Ab!x=J>c#8xl-@kn;NDM(r7HkCM;!*s$bv
z$Bl)~g>6yROKywuj_vqq@K8<?(%RL-%gYWD0gV8}e_;^b?R4`&ct`*LL3psP;uJa;
z<o{$4-tln9L3p>*9f1x2>%#hb0v!iJJ9{Uf<E$_S|HIa~W#CSobGyk68|NjK@Z2}Q
z0~jh~!}4y23Yn*NVZzwUb|X}{3zzy&cH4gdfg#`mmyrc869P3dm%&si6PJ<yFTsDz
z`OlBvPBJUoX{JFK8yf3SkgQHK(!$Uxq`mP*-+lUSXgBmxJNKIZx|AzX-)c?&3+arZ
z`)hWkAzSd8rI;j%?uxp9x_LF_U2TE@sksd%<W<u#(bfv14PVvjV3+rEP6Q40egg~i
zlrN-|mvmiT7j*+bVX=w|WkDN5LE?YFH&TGH-qP6yGXEz<cKVn1ud}vv@i$gzG@UIr
zMJ(#w+GD#Ylnjz0*!VS`_+MW>PXE*?rX8v0n!kRW;t!~dur4&piU79x`K|eW`khN1
zmR4(kwAPrh0Tf2#8w)@s20R#3&R-5vb}^Fp?xS_>=*s(`gaWzSA=f~Cjt75p8M#3U
z8D&uBwNgh{+CT+4X=m`ld*Hk1kb|UBklMn!=HRK(?sFd0phZxX2j%VjVx@jfzB8A*
z{h~3nkzMyoU&vk{B_X%*E$g82xj=+X(1{@(Eoj9UL%Jsy*>p<HLUqnoL;luEZv@08
zxKNhz!)h*Z;MG}OoED&}m2H2#ZH{yF$=mY&!3wEdC1pRj(#8l4mCrJiKF};;G}{dB
z8X;zXw?K5+p=nUX(nV8yWf_=VAPlP_av(6%I)VB~TsB4tNBMaSF4G~R<*T@SKkib?
zf|zyV9TxCKS&BJ2zT#1QQC}iL#-XfTk*l(`Dq836N6d_W>lQuJN#cLt6AV;d$Y@gZ
z?si(&f3~xD5W%TxCwcb{oy4F8R!OCBx}6a4IjUoJ_j7GQ6D0d?BBMziv4FWNb$L$U
zL^Hegfk7WN>>23G_vSe{-`B5PoJgr=oi|*&#}d2KWMzI1iY7+Q_e`MQBNf^nnp3l+
z!*zSS<u>LLtY@puS*w57$&$+Jy4IK4Z)z}?MIWKmR3ep+7D{){a-{JD04hl(a&s}4
z8N2hCrsUYtgymfk7CFU_Ia4MEyedZ(l@0VgW9ZE)NBfARHXYlixi0Xup+2GH)j6oX
z4RU>uiq;ZPmzv3T_B^tyge<#&9RZoAW=q!4;ejx#1&8YDQX7A@S{Q;Lro(j9(;$Sy
z9$N$AY(1zT2yysM2aX2lQad|vHbAsO`Ebx|fQHcAnqSD-m<9)NHnz)-oQ+A|jkB%$
zh0iS9#)Z}c*z+fJJM%WA-1G`}<ZVpi9lVXLe2}+sDf{y_ZsGrdw^d%*#M{`a1Hl<W
z5K<pX+`z>I#e;vq8Jras59Dp*4K~oW@q%X%W`E{p<8Zmbrl*LTqlsJc$Q;7kywKY6
zjh>-xiHR}yOw%@skqDmq@~Ngk`Ou2o%G$7+;M0{U)@I~z_BVI2Hl%n~+C84wQBS{x
zxq)yN_`aK&8%{Dev&!7?Vde%R7|Zp2h`Eu+ov$R7=+%F5HxDs4e2}?;rltY-pJ8sG
z633sa8_4Y%2lm^Qfr_Ll;8O8zVBz#=Kby%?2@`>eNgRzKu8ZLLMHN4EUh9Ukeu%Ms
zEtLG4@lUKI#9zy7nGOdsTej1I!;^M2%2350I6P@$wjua~h9_-+fM%W9!o$-O+hN1_
z#8$+Y1rmQ0<wCv=Kqb%S@uh@rPRCS6rBq96da5yN<%HF^wK2~z614rjE(2%Bg4L+p
z=U-6B;GTh_2lpOeuSJU)fL--ycju~KmpM>;pdBMR_Fm4WSq?;i;T-Z+xr{&EC~X%a
zrHIe?!K*8I@F?1M?Z8iE!+JR?Y2v@RST*`MRWE-RW1gJb50h(Wbmcu4?93yYT(HVH
z(ZBwHpL_5uBaq(qWTw>TSYVkTu*^}n`wQQz_8p?JU^bNV=Ud~Oq0%D3ki52(C~36r
z&Q)I;BTWDNd4n^lGdL1Yvw?~mB~~=N^1i|0<6Sl6_n~f^%dL8}C7pnvx#SkOr#yM<
zNF`=G{NRmSA)n)2nU*rC?Hgq<ZFA<B+~xxEG?VlnsmBkm4D+>Q10u~SFaW9ih%(!j
zeU@~Ja8&8GOHOL?gxR&g#}emT{Bz!mZwv-Ij@*owrr~k_>5O=}yzf;pcBT&SFFw;(
z0+*2mFB1VZm(lPr6qi*7Fu{Lw{^R49Cz+L%!kh4K_^Hp{Kc4=6>Qz@L$oS3}6w<ox
z&+kv+f(8=C`VNfnXu2PNbzh%;t)-5EalpVxV~8feC?kj*OZNi9r34JI!kW$27yh3C
z(R|3)^h>YF5TRK_|L4cIt{VC6%V2|%f;`r%z~^vGU>Rn30aDd70>FR7;1eX>^jrSl
zC4GsK1BrDdOy^N&la6?QoPQ3!g!l8ak?K_4=(6GVYD1Mn;VtYiR0{T~6f($Al%!5Z
zSn8*Xmvc^2UJ2d!Ol2&J+WkV?>MunHN-wl;SCUt+&`Siz7McbiJsSn6lu)LDQ@&-#
z%mw>m3Ml&sk!7krqUe9D^(i~wCg&5K<L2imc}5IZ!xbIlKa7~!9TKk4DaDue{n;=x
z=$w&mv{PUOS!}dBB;I4JCu8U|*~Cj=1>iYn*6PG1v*>7MULua$td{Lff5{3&2>MYv
zD>%*WwxVA#GD<vyJ&p98hfkvxzVYqKrd~2!8xr18BWcN}Wki2wKbdqD<kZjzl$-3&
zCN32SWXQ{k1>ziEam^unae1;en4{>8*I?w3IHFlidMl-EVnrGmuj#W?I{TO^VP0#A
z0y(qK?Iz4It4H>mRX2hGp#W>C`ekHcyYJTgsU!$erjh6~lGGy^J$QbGf<uoQ<p>&Z
ziq_~dntm7XAxD29RTp^jge1zIMy6!wOO@aX;G7ESlMjc4DXli6L-L*hQ*laxN`?77
z!M}MsM^xNwC0WE6yilBjrwe63xbQoPj`c+{&crImbv;@$du7d{0;=I;X_~E)g1-dY
zeccwJ6IMQ|+ak^B%IU3ctBgmZ)MD&k)ronA+v-|_gByPo@J={~^(x8_v87Fu)26XD
zkDx*&JTwT(OKIe8>`;~V(X-pQ^fSl_t*M@IXvMy>)gj4l@B`a%Ga=Mh?v2FEVZ-IH
zlDMOj;8TIn%<PkDf#+p?eQ$&W+eq2mAj3w=NG>KZBB8LsET#Fr6tpFCt_@5@Vp3w<
zX$XOspI(1!3DJh!ONBxmgJ<0<ypeg3T2$71JL@^~HXeCkzGf(x6}d;|Yt6+Dz&|Ko
zYvZ_k*k;G(`g}}AAo_ZRRc6di*S4OqEAN=$p3<pC2k7B`%8xTb<DH}Uc$S>_>$c7J
z)F95SoR=G9AHk9{r>$4IgC&Q!18NXjV96+jZ32JduF9dg%%a>)l9jFrg(A6-9l6qu
z2vt7YOOmzVO3-L`Xy~7N&ep4^%?4REp&75$+Dn!*1Ql7%qbp=NfnJkkvxh8m5JLmw
zO>Eh6_G%?9Ik$}5&z8B`3)KHNwyY&5KGZ*z4purmkPlWXJ+Q$*r<vu$BO46l{?HhB
zP=kMg23U^!iG;Ay;6a42TIE1OSV<lcLURKdWasj7+n5aOAylS%!ov#Foz)!`aSKTY
z^1@2uM!|&e{Kio|>uOF2oMsv!M^Zx0NC1s-|7}Vr1t{C1vOx(Gk*sDtGGj~2J4IcQ
zv2$u!e6_2b61+>+QfQ2au)p5&)9H=iE|Px@3d0<&Y?seTA`6HwjRCG)ZDCV54rk-4
z)m|=rq$z(UJgp_nk;ksAZi?%W#-_TdkX!Y7b*q1ORJ)L~?dVm<F7U#(y}r6x453-C
zcyx7R*Tq1Yu7a)ex2jpGlQlx`*VizK0?;;=Uygkf9loc~4Zc`H^tM{wSf9Sohz);v
zT_W>2W=f$8vATpFU~XP2qvNTo*|t0lN!gpH#N)2=8y;pF(7VaCuirrDkp(u>&Km_`
z8BGlBC~l$HGT^YSEL&<%xA=qh9J)|T8&}V4#fIlXm7dtkXOs80Iy32X&rAq~&wg~|
z)bY8IEfV^BiUuR*G+NchXZfpK9KwH>F{-vTG{#Y$8d|aV5zz)8=f>50-5~c?tk>m<
zft{S09+}dyPOr<++gz8<2Yz07^F#Xuu;G0i9M~@aDdFUN-#8OXkq=YaOS<&*x?r)Z
zj{wx1q<8fZcWJToCfvtQ-=<ppOZ|n>0+p`Y3#Qc8!{O}(aOu<59^=`*?dX43(QJ~X
z&MPJd17mNJ#|<CJr3d>{!E0{<a&t4Ca9gFEHIdsIqCQ?{V(pf10H;ECd;&O~fMc_I
zqS%^{V}a?AX`YB*p)_*w12qwIbn}Rw_XbRsxUqjK5RXOeYMyp?K(ToWeO3Qxqm<H-
zD~w!ej_O}d7ykfqU)tIVWp0<01uqZ+G&Gk`&o3qcHJ5q088MfQtt|?Fu|U40AGEhK
zvoEtnV8Ds%G=%~fmJ75sgg~O`M5!#Q6csrA^?hc^BgKl0X!C5zKw=H=_I74x=jD0!
zu0*abxDq8eW;KDFWHNzWtjPp%d44OplH1_qYAp~VIihwiG9)}F%;JaxKfyJVV#G9@
zA!)Wp)seI(Qg9S=G$}fNN~KMVaXB<y#ZifNE#jIMF{wDFU{YdOD+M}|Kq|?_Qlev_
zW>tlhDkNCT^F?Hg%fZ$a6;f;#B*#VVTFQ<qHCu|45N(#EPDlfEWOgF+yA@Kq3vCkB
zY$>>wv}sFlUAAjEJJH&t;$mQ1c~%#(Op~gM+@8Rqv6c)Sg&dxLZ`a`WQq8K8&2>!&
zW~uOQRE;i$=7XY;B37<7L(*oIxI}ETU;<2RE5}A9DFhuFT#C&;EIQ?C<CI-eY1gpL
zp_pHwBQu!=np~D>vtaXJp6v$|<<hKDby-upTV1Z2pu-kTZ5JvWM%ZMy0FzQ>yN1Dx
zFoFv(&h`!0;2hh3Hz+fy`JgGX-AVyJVLtvLJdvZA4$N1YPi+b27mN!|gH50Cm?9)h
zVy7~{D^8O=29G(_HeX0it3_Q3*;T>~=pa`m!dLJjBy25JY*~%mW>HAyzr?tkOnG9Y
zLJ3VT#+72%5GoV`WfK!bPNoIF5xm;eAh<)OVtSAdK@rb?5TPwp%aX*w@`~*-#$;rn
z57DyCb`c&)E!}Kv%fZ4ngQXdj+Gc<Xs?uhF3Z_(_eDh8B`uD4G_w4d=GIQOxlj+%b
z`X(U%!*Tc5?uYI>{J?C--K+6v_67#{WDymUCzwTveZ^!ck<VI=pL9nzr!$-I@ci=M
z-LvKNquKd?<gz>J{{HF*`~T}-X0xm7KX={F=d)jKPW@=|uKV@o-POU___ywFH$T4g
zzg~2w7n9R2k4Fl(oT3_qq^HbuG8&D!*0ZxwI8{zqH;pKC*V9pVeTL`0Iz*K?IK)c-
z^z`Wy{I|BoH3K&<nC3s`3r^|-(nTu%S^-k^R%&8@C|<Q5ch4NAKkB}pyq<K=-PZ}V
zgs^RzQ4~-cPy}JczD87PN_!(v(>z=1Z!v-dcd*9@RAB@l!kqTE0!a$3D<g{fQp<KD
zP~uSJ5u8ZL98ldj4q$|mw`g8kQH%crBM|MT9A;RFE$A6(gAKdKJyY~*xx9#bgrgTq
z4;}Y^znx6aPfjn!*YohF7!d!I(*Oy8?&=GAS4M)Pd~+mER#>oZ);s=29nO_d({d2X
z2*0%2^XG5S!VOux>i+T1e_;kxdvcdI7Z=AH2axS4+7)8hqkZo%kg6{`jRY3)gfe^o
zNWcvI<y#|N_ub@TGCjIF8I7IT1Fm~{GMkNmr<acBKmYQ6_WjZ9WHxp*zwlymIfITD
zVqE~`-n6_xLnG@S?FYIqaFNV^0A>(15ccZn)9&STGCCU1-gGaYzvy0%-_MS3ZRyLC
zpT{uhayGu4T^kx{8nJcPlbh*id_9LgcIn6Q+4;%0llN__GkRI6e2nL(c+xuS+>jQ3
zBG+>O*3>_F(;%2Fqam=l5c?~!T{+&bydky{z*9z0ky?Kx4=j+8>LJnx#n>JtHX@<x
z(grTYvu-y6_$Xr>3BWCyx&7IzuprBkP-R(IBxJ8hum>(IizmN2)Km@*YY0QM)Wvoq
z!dq*lntcYMA}OA7-q1)ahBJCij-g6_1^nT)qLRX`w1wMHtu|DawNZ;QfaEr;zQf2O
zo<sc<Wnf$8uN8<7Ed%*(8Au;h2C}TmK-5Lj1K%%%z6_+kFeDwI9xQSK<iZ=71B8I{
z$>ofIiSbT=WFuDq+yM~{ss!9S0Ok?~+5yl;s`!Ee2(AXe0TNVvQr!@wl@!r`mjJp1
z(HC;e0~mpDCL@!mPGHkW18-y?2G~~Ua~Z*g004o_6Da(kK;BrOm@8_zTN(*?p=zxP
zj>>j2Q1&+>5)u$u(i?$bM7_*gjkvdH<gguye45-|>-T%K_>%7Ac9D<C?NV2{o%@u|
zeM;v(rE{P5xh{kfdVzX@H>CZ42hhosyd9}5=_Fc2&GKlvUnD*_M12{MP)5`jqwEGn
zvmJkcT}e!L2_<+FI!`PBJqRTYB@Do#LV{9R8Igjw<CZNpfHu`4iXs3rlA;+7U_ugf
z`?@qi;0HTa-0FucRT%U-5hc+iVLu`lv0_<89LH=Pd1OQmfEz>(t$V<Kzyr7jX-3=?
zZwFx%0C>ZO0*a1nCsvZ|90L(ravEUQ!6qdyZ9s$1c9^xz1{5J|c_A)=D@MjRMC4^;
z97}HmIE2r%-3Y_>c7}qSDOylbtFASY5_dtLYoiu)HCpELE+ZMu4&3Yhk&JeTOxk56
z8fpmTiy^j^?{vKyKBnt`)w1e(+BY)o8<}!1NH4@*=v$=rEmHdyseOyox?dk;@wy^`
zC~{ta^Bz}Gl3=H+!a5)q1V97Ew5SYTMS*b9plB1UgW?iGhFd_*L)D=B@os?G3SzV5
z-I_gs<3qcAay2`DcmCh;^rv%V!1OIra_;<`;rk*3x}Fnx1-W2<AnI(dzvd@>cUcx7
zfaE}i0>HLn5FG*qKv`NDAyhr^#a1K8SROVWb+}RmhG3%oaDZfQ@8VWC0N1%Z+WoKC
ze?Z0|@ql$?X(!pq-q?oVneO5*B72jC{mEi!2SO!7yi+jr7inokExx@#-eN=qsbutG
zB{16rC`vj6$qrC|+M6IneCuZ0+`tvwQ-Bw_0FA~l0J)Wt(cf!tLd@d-cqb?U0|O5$
zdJ&pCty9}d_Z+g{8n8M$H#EdMjEr_%7t0PKB?FA?G!V&GfCKLz$!L&v4z<I`pkBil
zb8w$1{6D_wReTt}v3KH2d{cW|QWw<S-ia>+>ILov>4n&T3#k|Kez2w;rq>m&QSqfd
zE?9$-hRzzo5LAnE^cDLOv3@%>vE#ecxGjTX@Ft$4sIK5Tq?~iEK?WfCB-Ixw?X*zr
zn*yys{Ny4<zX4hqu9a%%3o775aMlp7ReF2}SOwCK{0*(l2#N^H27Ffl?yyTWWbjTr
zEn4=&cL0=sk#4<L0o$we;O|vNlgsPb^aRa)5q(7u4r|>Qd>;a4kXK0B6_|myLw;<F
zMV8sIMxa4p%p%Gu3L+~nqnzw~DeC79bK-;!9+zQGlF-8Qrw?;LGUW8?IEx7UsXVsh
zEEKi_`}bCuqkG%;T=$=#tdClN7SvgTi#v=2zo<liUkn;xxq}9nJ_ZeNTA~5_8g1Ur
zsc-PqH+aguupFbk&_f3fdw~FOOHZ<@#MvmG=pzf!(hfID(ETcLDiuBV)(Z_yUg^Q}
z@#O8>pT_U6rfsbf!6hXP`AR>c_}X`>ef<OO1XqwxU;iRlTagvQj0ywdqjGl+0GOHb
zet{)__PyJ!56}$U#fSd@?LRjsXRD<aLVrgIAM%t6+}OU=UIIuX?ddZ_78g6ASVfk7
zfEXT|zXj;A#Ij##VWCx4C!ttplm{nimls!UiOg6c$GDChn7`1M_-X!0-CBc}_%RJW
z(yGC8-^ID_;@o#}?z_132LRF^07!oTATr!Mw|q(GFSq)9O%+sF<EPKpl(V;QotB-q
z;BThK631exD*L9Y^;s>n>H?tCQuxs<(%7u%g5LcPd*7Inmyrc869F}saTg8~mx&!c
zDVOCKFq;A0w}BZjGXsBwtM&3*w#vmeP`g)Z_WCe2Dw;jY+$ssJ5trm#N<fJ@qR-!P
zDqm9DL`vRH75hi-Hu-xrVwboC?u9y-VOTvp55F;^TneCwyzdMnNVwtxct~*srcDHE
zJ|g6~YHliT&4rPrL2Dzxk2PUb|30fd2R6Vi5I$7!;ba<pIF5gGb~jq8PeV83hY1^#
zADhh9O);lPG5&UGKH&19&rt&IJNu?`b!Up^W0Mb08?xy3l(I%QaQ^b^$5j1ly-Z@6
z1X_|0AFuVOqo-Z*Fhv5DYCOy=d)Sfau(@emYZtq*12Z#?YwaSjERD-UIdXCgG^N>7
zE%}@;YL9{`PVRq39}Q-tLhxSpK-2=KLLew+^y260yUUQt%M7yOey%~*B(G?Ypo9Wy
z4077v3?+#|(Ukt3I1a80E@PaE@%bF1fz+TMwZCt}ugXTHs$59s`(tHXPurM2q*n|&
z+iRbakZsvc;v|-pMDqF8Ae~JA0vy$<=S7yd;(P_m^Adl8iRFo$pj<hH@6A$q5Xz6~
z@%DK`GRuonhi&o(IGr~zv_dQQ|H&KlqO0%9$w|2G>a!d=01=v^>avhtg~D?~dQI}7
zkepnlw(K)kX-Ljor81*)SE)%}?|4$i3oiYd<H^<e(iZ)K;)?iCy*70fPjjT5zHJep
z$O%GtLuY@+JT*pZwVb)FH2aD(&%BCq1m9oeRW%0ZT4GJoTH+=asI4G_jRpF~0__xc
zKvoSDKfVAetq_y504!(Ha*j{s1v8W;#w<E*(aTdv!sspE5_B9V)4BW=m{l=Ub{uCt
z+V%c+|5B~zcwIjmkr$yExBsS}?NMmA3e^Q&PtSkOsPt@eIgB3ZdMe4G+Rx;;p`EQj
z3**Mxgbtg#K!DB~yP*qo9N%X<yAyInv8A2eQ(doE48ADcYbut!aDGd@HVnr<qu3K_
zt5koDu7J<HZrP})$K#4ZCk!8=9noR4m;I<LpTlZPSX++o)LmL3&vloMYC|=E5t3+2
zHMD;m->Di}A=jLtsG*~}A!<ny(ob(HA2*L$;FJLC-uMlFx9W{=_&aWA{Fx>cAWxTP
zJ?vhbCHCv(_U?qw1spX9jdsTI@dl2Z3OSf@JYuoh3$M(~3gk^z3l;@}pVrP*imNIW
zIn}+X3W=|gOZ{T18mF{MR{}(MAzl8I9NKJNGYl`-N3O%bus=`E=bBfsH!~+Gl!O^~
zTHwBcoAWuX+?It`$MJo}?@mZbJ<zNFl<7I#;1v-5R!(OreEHv~OrjhQbCv2$h`L^4
zUVG`u6o-%g0oOGKUit)XYR7bux4%EV-2Dq#jN)&Xkp(Xkmq6DS6_*KR4k(xCATW~x
z6Hd2%AuzT9f0lz-$4u`*D;tNL@=*9o^wr=xrZ(q5kzvAAp<mjJ$@Y<di5z{EqfIjg
zs)SrG&oRE8S&CDx;sDaI!P#Iln>bPse7~#Z2OhV9J;;&^E`<kea-dk8d3{+=qE*b!
zD($g}xtd~iX15SHwu=F>UU6iYq=_&%A(_FH=xKMde<z5_XnHr<@4>VZa6G3u&frr_
zUzEu)@Y{b;BF7L_VB6fSAxYFFJQc@n^VcVEh^U;IQE}@8Udn_Hiz&g2+=vexc()ro
z?o6B=#+KZ#C~-Wn1@4j1up=rDGj65iPvH?qCD-{A$r!oHjo`*dl9q7dPy&!|L3aqy
zn_ljXf5K>c_!}0t8bMj_mY|1}c0#r9>d}ae&P5++$)Z>Yx;B%NF!}*9K1`B?+2q%1
zry?`ZQ90MwCa)a$BxrO%eq*vg8x*mcZkH^$JW*iANxtdXkB$*VBl0~RCC;t<*poXf
zM;vc(Gk1Ix^NPPjQrIK@qEuovDig8KCEhEqf8)(_iRFrgpO0&>9ReF<lvK#VBjmU^
zi3MF@cXj4*$kEaFkT>Egc!?&i-yv<2T6ni(&pE5n&gg78V8$W85nGkl%cUMQ%CX>T
zZdv0FpvSVsZE)g@-3IHr0Gxsy19@I%iqE<j!kDRAPz>Sx1FDM(I7oHSH&s{IpYBLb
ze=i-e^9KqSj+k63_cXvV{!-?2t4*h5vnZc*BD=b5Ku~D>PWo!c;6wCXPg@i$Gh}*^
zG}e)dA>L4RmeY(Uvr{yj^ix~y>r`hYGHKdNA3FnAEKBr>e5AoEvPz9&pJQ(N9!dk)
zOy{K;ES20^lro^G<$QcDXYtw*c=!;*f6bpKzya#B5}zRF^U0>3?3$QBilcT6{Z!d&
z_-HulG~0(0(y>o;3ng#O_FX4uwNtU#PhftKOZ<+ORVXXvv6VyajH4YTt3fVV%?$sJ
zq_s**@QRD09^koexJ(;7LqaKyuxU_Qt+-KizDWK`cw4Qy*>tV#0Tq*-Sq|}He;C?7
zBB^+~M`t?b(Agu{N3$x)v|==q15F)0w_fzR{X`ak>nVbg&FSeKyFmq96Lh=T_wF&q
zN`Y^jJ&dg4(x{{fC!?<CY6q_LWee0W>dTcQSDrBsrBzWow>Eb_1BHpy?M&)sRq7d1
zUG)jcqYqq(ghk0yrW<He`WU?;f8%Z<{Y5K9&qaJYhFznX;-(9lDX~wI(NmeR<&JX2
zSNJ%qZER82jCGZ$*gWiE;8J}CWg}m$Hv*$F5@|a~IF>}?+)lgPom^VG;fiJmi;NGC
zVJ43&BfYdU!(1M2dPh(zds=ONB%E<0AeW{^+U2UBEQM>|wRC=Cr^0|ye;(+DoeD_V
z%I<9>niV;xHLV%7t9=13q+{zw`{HxdoQBJ;cka5m?hHLaqqc_iCm^|J8sG7Lw2uMS
zvKzuJdg7&HOgi_DlVwdtUQ4fIxtCl^O0aQqwA~bX*U5gLfWAx7VWaD>a?5$X19U^n
z{a7^(npm8EOizjzMDGi=f3Bh->ZV9UZs88Htc|0HmevjTw{6Wno)_`7yVmW7uT)=V
ziT9gDk(F{a5+4?O#8SXy)@d5_hxC?b$Kieu=f=hQpvHR1U4&Ls!=S{vb5ve5h?gwl
z+2JmmP_)YFp>15rxs<x+`DE!hYML1e!N?hupJ4gum>h66x|LVBf4Ye5T~oyS)RRx-
z`EM|t->$`3IWC8uOSz*b7%L?oR?28!pKuvb^KIyaT>rO_p^azauzFJSR^_<{w^TL-
z_fA3N{btqd%D*BKhI$aJ+%<X^ZusH&t_QN4afIwm=L;N=EGb~mn`v{&u!H>VXy^z}
zZmSY5Gr3(W{pX?RE)27xvmUyJo2)eNN4Hv!??AvtfzIP_oc6bWV<E@M(3)%hp3uL!
zw)h`xfMyk!kp(Xk0X3H~+yfH>F*Pxl5vVCCmnbPPkbmH3|M&YJJDG)jJ_orasdZWq
z?IJ13poxnij)piI`mN(w{>E2&N6BT@#Ex@m&3SwqlO%XPd)VY`jIesLQ9~KvR%!Op
zj5C4yPD2Pn`c|OBIT{z-OEmo+jHVxkY;TB<_Q_Mp&LY@o>iBwVzP`Qh{@WF`Q3+$i
zV8IJTmVf!@*Dg+IAf_%1h$4U<e*ZiC*!|g3&Ax2kDG-y9r6u8KMn)@)@q<MzEWPjV
zUMV!ODt~_e9Ga2uZ$)<`m#cuSG(Z&Lbr!B=2TB2?QnMq<73GVv>a0tW%a6JxeNK6v
z4jg@^x%7R?dfMyd4^}t|o>qS7%1%Ox;KQ}HUw_xF7ii^puK-bwHsig}L0u*T$Ggs>
zaTRgT5S7-K1Hhx2Aje>e3LfEDIhURNDC*KJ{#j_^B)v-@WCtffq{&R8`I34y$@nHL
z(TB1UquARR6+n1&4bvMvxQaj42v7|#bI1|PmQ12gw#`;AO3mzhPF9?&Tbl{_WNaEi
z8GkE6Fs%{h!Lg;7a$aC)U6&%pXQ$ZdY$mCICD$Ao7{4q><>YR&$As_E<xGLbkjI#E
znqN{Jty!qmIC0E2b7fr$p?*{vFRU~*J|R)Xpk5|w$y_duMj6ToWXaDZ?(>TFq~%dH
zY0Vx?@0tAyZ&}V|wiC~Aq|7XseNv+5#DA;hk<(o~LSuoV!PeS+Qo?Li-fOeI8O0fU
z@~mVlRv3R=;~TANsu&n8JLeg{RF@}blD&;<K=%EvEn#~(tb8U;IeZh0rXXij(sScx
zsB~Fi&-)S}0-LRj9N3w(_S-u2&eQigz|!=ohU+mD3;g^2=-^}Clfi})Yo%e0!ha*L
z9t1Ls@&3a50wmT{NyEb635mn_#6p@y{h-p*M8p~05UVV(DT$>@8L`={bmd@ol{7|D
zl2NT{rLH8Z%2U2481|OlHb@-Hl<v&;xAknsL5p5ED7Ah`1wKdCgy@%c4@RQ!(p{Ax
z^8rn;bH1ok&|V9QgIhX<Rn9BrQ-3f*w~Qav5oqnG*8QrEKpQ1A-s-Z?Rihf5Ks&2O
zBN}Je`LH%!uR_~QTh*P%DaaEDmDy5Hl(pUR(a)f90l5tF+H(#m^lg)4b(IOZAWc;J
zupMF8F$`zxkM&VoFFh1AR0u+;J7R`%F#s=)zi#i4$5W!Y2tpL~_QdNZ2!Gj>BwaQX
zVb(4hG@0Kw!aNFAh@@~<e-W%GYIdc*vsXWdGU#d2+54kRr`JvA64WuD#rfIhVlHWP
zxUd_xQztSNT3okw%e*mUpx>aN4I=0Xa!<KNQ&r<3D{%SXI9X5LUQp#iKUI1-S0XkL
z@H#%1g+;ly>qkZ^UN_rLNq+zXW+5c^expTgg`lCrZ@0)PAKieQ*mO*PDy#V`c`9D!
zo_c;>bDcpUOVz^$7#X|DM2yLNl_t=5kMR<V9<PiV6To703;Zn#<Dd~?-OWt`STCxq
zZHYcGb)h0ma`&uvQF`h|+*L5gWr_Om!iBkj55hUPIa&2MZ`QkW#($4WOhEoO@KyCL
z7;1X_tuFgqCIZsQ;^numD&Af(8kO4%tnNHs7dMN%Yaw^OtFb9#<lDr8bin)49hg!t
zl*3e2KW|7fuAgPYu}+n7X1s&hKP3zg6PKE2!jP!JYfYc&mix*Ty|$ZlvN%u_HI_Hm
z9S5zAaC+fC8Si0Lds_E3>BL^rhpH0L`Tlm&t*bfKH*@`bK<A$Dv;03{>R((%xm>)Q
zrrKlP^O)29(u6nExne6>vV-r(WXc=;xVj{VpaJSe791a#L;?z@%oRDLeG>-$4~@qi
zmto5p69G4uuN*2B0W+7;PXZ^GS}!n_0vyk`@GmgZ1O!w%{h^nU1uqi<HZhk`$37I7
z&;=(cml!oLmVa{o{o|JxnTVCboA5Jy)pze7FaNw0(oqRAelTPONErV3`z8FLfrPPs
zKolPI@bk~%$IGwvs(G*KI||AH3rZS8v=9^;LF5Kwlu*_dK{11x86;*<vp9(F={hp5
zSzwwOlq^W{Ba=e&a^>UaUqdzW{oBN!<(<gXJHpp+Eq_E-W;lq-tZ953X^Awtk;Xu?
z4odP9-EoY6>+t(FKD*%Kns`_fW!J#6me2~TO{R$rN$XRhhYf`yUFT?81i{E`_*Z9h
z3$zlBJdjCqDAAkbPc}czD<PaF%15rxh^Lt(O!%F5itAyS&v*Kg|39|Bz5Dd}6;6qU
zY&maHvVY0;8X2-Tu_Zt)!(>xE$tz<8S>4PkoJLA%!R4wopCS5p@N|vpG)97;EPI;7
z8rcnxo3eB;CtpO0*#OICb{vfgG37Z(&lY9!jB%$+A}$G(=?UyNh9<r6<cp!1WCm_B
zs?ogTI+#ov$vF%z$d^;3M%Qdczklm03CIiSRe!yIrdVTFymMFXDHCOg&#qow-kwH3
zN+S$5@*^`j>7zW55}pI~f5-zl;XF1xxXBft#ES|j1C2hCO@6sCQfvf6@+iPgXzP-7
z2O3x)q^1DOp0N^i@1r6~CoH-e6)gzMFu#q8LIyO7ACy39G4eejrI%Ys$y5mICr9`?
zDt{s}mXHRa5_SiB$q1!<y<!v<j)5e)`0Y5xCd8~(M!Al!p5rh_)aq%~&@p1ZWG>-7
z(R5#=cjc+sq8uEMr(~7$@}LF?8^v80Bj0z(e7P&CfIC|`Lq~F&gG<9h%#7BgcRRlt
z2`T80cUWax&z3F{dx3hUwJulOohe>snSU0CyXhnHwHTROa^`~_ofj)Ua>=sW5$t;5
zfCa(O6LJvaOt2v@F;TPj&lX6g=%lS@Pf~M`iJ+geQ`1;d(qFfL(G*r*<mJUagJq>0
zxXy2JUtYJBofQfcErNQopb#BG2?g*dRcAPL2<zZ2IzhUk>+Mz&G-b~~Aqfv=7=JOx
z6$CqsH;S~-Z0!}^7$Yn_y$I7%b81V^S9TO}7?zi1a;a6^P<h3$Uvg=^-+GRrk`aLQ
zW?O|{3a{0Sz;DW~8b%xnE}RP^JqcoBs#wTzW)f&tJ8p8rS)5b#jG*Nd@EdM8tufx!
zYANNPp9<dio%2Ea6b>=;PPo9TsDD`DFa*d9OGpWj=OY_Hg~+%~4X<j2_B#ND%5p$Z
z^-A<Uu4d$M39$2nuNSPcsE3$0Yzn^J!KYFV6Q<eJQnLF5!D+a5^`f=Mq?i+maI2=O
z<(h$&-g2`b`H(Bg%X{#_;~vB5B`}UwJHoB+G1ju*Lp=ji{A)r}y59$WQGX(_!AHjp
zBkS!hD-T%*D-dO)qR+U7a{|%SE7ALL&XaBJ)*%xqylX@i(}&Piy&fh(fnxA&?FiQQ
z2gkB*Snpvg9k*-6$-i}WFFk~Uqw`^#?$8^cP~NHt_nAhf;<O&5o6|$^ye<+aYa!UQ
zn>_Bw`<gU}T)n88f<^aKEq^<#rLB@7HSaifHeVSZ=RH+@H={;xgeFtdppWG0owki8
zp*%F)Eon;>$8=#-FOT4?TRPpXCQWwn#5Lc|yUJTDl;?45piZ-^Xx5V-NySmD(c%6i
zZ-lPxH0lhedzn=6w4OSQ=G?H+MyIi>xXw~I)mM6K3r9}uqL)6M9)E1sZ6UA#HS7Kt
zm1D0V$^nrpw2FeNLtoB7R{CAJMSh*P(4$gx>R()^eobj9F3(C*8$QF*(ZUZ6Gr`qL
zef-A!`a6~1j!F-m7G6&#tY-?xzDeJ2EFEzMt~&Cp|F^PGicQqkLg;iSmGT0E<SUUq
zL~`jxHHxt+rYk+g?0-!y8%2$A<u-5o?OtsAfO5MhkID%}?tV{x)1}L5T&>aa1U7r3
zODCM#Wbf8=nw`$R<^LkCnwjs$uHE8amppkK*^dyYj4)EyWSZ21GgDXyysK%e&2W~Q
z>B~ha^B&PJb35{F`WU!*9vY~(d#T5*4LWH8Cm;Z(=A_CJ5`QPNEaB4}Y4U<(y&-NN
z@M@z4a;mnR6k6liDL`)T%4H*S@?0yRkU)Z~7M;6rqN;Hj>9~lcMNhcn=rlIA^4a~o
z0Ehk=KFxDgT0e&VJPra^H<ANlg@f7)2kK!Q3@t(AYUqhaU!lHox?Z_=^u@^%c~w8%
zKl)1a+&_S7IDfSy(5XicoY*&<w)JW|51byLp7&L#=hW-t#aaCdP_^nFE<Vx4B;lK#
z-m50ff%Q=HNV6j7YQ=m;CyR_1BU^1B6<+abWS%<tyV^Zm4jKoI*6v2Y>43MB9s#lD
z%=FR?rm{H~Dmh4hO-YvC(CQ<vT>)@*u5=HKHQ{N3FMmyH2dyD=rY9>v97pT-deXDA
z_j=Muma<A$euL;vbHI6FxVWu2yMvYJS;JIMI)Ig-Mgz_{>TMqbOT}h_wxMfp!kckW
zJ5wQEQpL`<){;V!am28C>nkpIL8iM-7i10PK@yiV&8s=n>n`(G$R>`)K1x7rP?z$m
z+2S>^d4HyS+cqJlL2{y6(ek|F@>nWbT)b~3z}dQ~>FxAMoE6?rcz#Vukul{hy;<WV
zr$WZe8xO|hsNj|i<U^KHKo)@~1(lhjAm=M5)7(@HIl&RMX5op6eeg4M)dnkk!<?Vh
z#><MtN+O!26UrT9rwF^5jC0=1v90F)P6>dFYD-WH>_`cqI6u`p1->(=X~#KNO;c%6
zRTh$iMM~(cTiVZ>fm<V_YOAv81y7&Ub(>BH3-FZ=VeY7v7XmMnlIydn^v!wyztozP
zW0#QyFB1Vaml0k<6qDPiD3>`tFqZ+#m(e~jw14C|T~;qaGNL1vj2D`0!?#3=K2u1a
zNZC9tF!8P=%+%Fe{7ha)o>YtvCQ}_&zWu!K2F?kq)O-tk-3=f{CEA&?yl(3doj~oj
z-S88-qc8}1rn;fKmA1zdF^43tnZ2U?Jvw#H9r0P(bIsrJ%%0TR3Ur=ti?<}tagR%E
zD1VXYFjG8~xP<9=B4)s(d6qa@To%cEiGxWRP9<Q3QK4G;vMsYhN`2MPcZJ^L>IhP3
zhtp*sw>1&G@Lo@w_e=?QCkuAt*H~dVl1O*YR;t0;Raf`q^xa5i{Yokx<jL$)rrlyV
zE8UsMou=$*XNpkrTFRm^FP!oXSYjn9%YWaisHB{vt9$iY6}acBdkp>;N-nN$0#I_(
z)r&b*xg$7G)lCGp;0$8|GX!LWA4#$(iBI%*0TENsUKre~yw9jr;`vp%=pbjN-E`kl
zef*nGq$lU{#B&LaPvL-G(MfT#70+N4B@KPdbFA}UDc?B)-J$ZdL7|+Q8DYZlLVt6q
z%8M)n7Cni3E+V%*C2=H{psFdUM(H!t^*2?7hESnioAqX<K9m<J>nV>j)firA0@vV-
zT_!nc1yphS;*u*DjDZqb66y{htE5?MuS7u**4xEM5a4S9D$srN1R8<f%_ky^Cv_0~
zcg1+ot8+78bmK4b)Tgkh&cttlKYwU@8qom?4cQ{1LM6>~c-JElmv2ccA=6Nt%&&z{
z%+XhNRu`;1mD-hLsH<}>X#8`L;0TY7x~Ju07mVEt*2XKfRmXOjP5=@I-Y^H3;u*GO
zd^`<1Ry@|Hr!FsC>uSxNamBuLbenT${mHlkXHBLv9Ap*L+e)AJBHv*)(tkl-oo|d)
zsqx0=VxlPfXFE@{>pt=-d(*3`bh25u=I};qPEO4VBa|!FbSOwgHOd=rN9inrb8r28
z`fJVEVX6a4aUra3{H)hgLxU}0Dtp`O#}lxx<!7{*N(cY8=k?|3sq?g&W8+i((v_FX
z<ddeJ@4TtpeBAi)In|Yh!hdmDz0^HWO61;6IK_hRq|G+%&Kl0Bz95pG#w7PniRCkw
zL6XsDSC&J<V!qWe-P%8VFT5io%<cLz^bpsf;1lHZQhNf31hdJrV%ttQJ0a4|CtZgF
z%7`_8gQjY2(1oYIdA)3UELadYQ)S!YP6b8Ra(9hZ-`#g=HYg{G4S!1YS=BDCQb}Kz
z91pFkv6NGFJ$3n`<Ux+hPw06!dli6#00f^@-?|TS+v-yd_otp>cZ{in72t+3CPTW0
zC32qEy6Z{R?TpJeIQn*n82eMV?sOy6gFDdEt*2{6H=3G1lY+1d?rO3JC#f3Ab_(Z7
zDp@Je>hUCX`c6~Roqv|N^naQn#D}lgYOgmu?(C=;B%QkBdl_X8@jr6F(ZqQ7twgA0
zob0#I&5+M>{JK)PJDMF@!hu%8HkHPzCvZa!#bNJl8AF@U<c!fH`I~tsTLg@!o<ULu
zB!r@k&KY90dXOCmuu$J&S9a1jIUF;^PRnME(6FM;T?%V8wSQ`mHB`ds*31C{2p6#O
zTYQ)MhfeSJ%pZVombdHZ=ujzVXk)gkfa?%5SI$J^H1!6xvtkUq=DGEkk0upGzzqmE
zPDw>c6ju7X-p}x4sg0R<<-YebV5pPDhLTs6k@S1#IIoO|CyISXkYe%r>31^|<=t^>
zp3P3X^@@)?7k`u_rF!otcZVvysZUL}gMUDuR9764S9(!(UO23LyF=&Wtq<mdraAIE
zSjiBzWVW=|M$P9hPd-6Ap`2dn?{fbOAq*mDyVAu&wX^v|o#s+TZa*@n>sNX|Wgk#M
zXdC7cIivKX&vXmxPrk?BfaR49*N%JJj>&_*?zbv9fqzE*hk0P8*!L)K%Rh4_UwdAf
zv?e8f-aPT;Ee06IO0u~%ZZwM7Ogi*1hFetnyq9MO;%IgP+7~5Ty<>eq4svj>U{TJ{
z6THLCzI>C#^zNq*%55jT=k>&KaaZ6)o8alwt0@)uSwsI?E3ZJ0x&Y=nU$9C+jjl{%
zeZiacX@8ouZk%Rkt25<#YCw&0>S0GpINatk#z<H-uS_^KO&7DIHzTCdIIU+ih5KD3
zTxU_wsLf0Mmy}U&n_Y6g?{;n3L$D@dOXnstmjbfq8%B8{wVDp0>9Cqh`q+9^pcI9B
z`zMXoDY%!u(@0*1N}pF<%rF7f`CU(^(qW_wLrCLt>Hjhfl4f#p9>sS9|9+fihxVs;
zpu1X=95kyz{?K&s;E6{n{aBFcnzxycaj1fyY!?65F-q{l06Z{7nZF6Q{|jC<xZRhL
z1uqi;HkV;}KobHnGnesK9w(QNMlh6rc+f{?|M%w~J6*s&pQ2tv>bxk-Jw$4;Di+cf
zNachFNG;G~B`v^c@luZMZ?WAsdDZ7rpw-o^RZh;lykWM)gKl{tx&wlF%di#|wNwaR
zPl-YDFNxnhS-}r`EDMKhj!dSCF7cU~_v*iIpS%Bd{XW?!>tf?5C<7|~`MFDfFNCnD
zj}5bmB+ajXoA0|n>sgDz>Na~@W>q_YAF;|#SsNP1q8!>U+YtTVT5r>*mXVk8Gn5v?
zoi3aRQ1pL({b;J5kN2JyHfiG%poAIeZRD=~2mxc*b-lF$*QPD{vW?VN{@>sr^Q<oh
zbTOz6K}j@c9G%+?ktA1ck#u2yoqEz2iCQ`LB248_2&b+$WA)`pK8RqD*Z%FSQTHu|
zLyzW?d0p?*lO1}Ew%efTK!<MXJk_aen9qh&3~o4?98WOTw5~Ae{diwi2gk~KsvxYo
zareIzo<b#h3nD;KG3-8Gs1I^j%LL#x)&+;wMa}JQX5Xyss3X;~a@@OrZ^vDB3)>oX
z0)1VtEn?o&RIh0!D2fy1z@YM7$tuYOTjH`~ZqaV5p=rELyKh31)n)`}*@`;UfF32~
zB2kf;TFOP>)_iv<7w?rZV~3eh7pLiN#&1QR-Sw(oG`hmPrs@_dDz+Z6c^|f5zAXzG
zq+fPh@TLIAaILeDwX=?YdOrF<omkQstB@Zido0Q&MOc$t+D_^7-YXec@DqVodNAmm
zGH}B(yKH1oxb$ZhueZ21icl<8tJB-E#Zw_Ht{%~QTPD=%mfF1bsAq-eK4MVN3HL;V
z;@ZA4=q*WKQo2a1mYnFjwB6^#S=#qc^ios5uCRfX*REdg3r1akluk>iOSxVt7bGc_
z_7XZ@xU{RD3zoj1(Jj(kx}*`O{+sOLWDZ4czz);EPHm}_S4_L(Ri03Vf+sscUrAc#
zg6B}7<8X=BoQ3A8#qi@Qy*VSL_!t9xV1XeT24!={0>gO36Y=fZys4Nl79maUF2%$)
z*Qbbm)eAqWXfSYp0FMgu_niL>TF~}Ilij9~93l*QPIHG6bEi^Txbyj;SBrrLq@e0_
zfrt9-hB{lcp&LKa;S2gBeLh$m-RprZ$EhI}S5h^1^VEm4>^r*L)f}&LT|N$G?GJMg
z6%6*IqOo@r3>J7F_cL-Vdpuk%3d|m-rMM~zYqKwSDo&Sw##NIK0-xjFnl9zh`7YsH
zncc1f6k#E>zefisIuu(MEYapbdwuao4qg{cy~;}QM-E3PxMhv&br9!aa9QCl-%puo
zGQLf>(rzU5r&(k7B3;b;E*|!+V$f8>w$)~jQ@ri#`xjJSTn-<TYo!@2jf1<MFnYrm
zBN??NBH7n}?ArARHNnHPq#m_!QANVa%J^VaDYq?U+aUWKGxotw%C*s;PR;uZ-Kfx&
zfjqS2P^B}Qh5IF}>H3oF)?2mv-Kg29bc)sS`W<57;)g{Y2A3`u;|H-Esh(J$?gpJv
z(WYEhLm$HBSvTEu11O>d=g|k^m#74(c@h6MoX%h}bbKUE7cle0l6@Au5^rS1;_KN;
zSUjJQoz}!yBS8c+xhs*G+g~?6*q0XB_kZ5TmG+mB1uqi<H8htoiAE)t`wcG&0yH?4
zaW^P`+in}l5q;NJG_Wt52laGa`W7q<c-Qe}11(_K4v<6;78FI>+|Z;zQHK5ZIW^>n
zqDjXiQMPEnLd{HdcU7_a)Ty4yk*f<XM+uHZO&}*pOkfu)n}91$CS+G*t5aNEKB|s_
zJ&%g<AxEc~)e%uT!c&uCwx`h%7h6X$LW<3QrUI$9PqU*W?Nf|M?dmZ(DjXptjIFQ=
z_pnCI#UygfBv>P|V}%)z#Bs_t0ybhXn>iLILCw-wA+;8y;Dp%jCAdn-97^njgIR^e
zg|#@fsf2658c-3jaBwSS8xiqT$wuH(VrxOcy`-(p*+~v&bAkm+>r-_~(e!d~%5D6A
zGC7S*SR*r*jOKYxPD`;7IXkUntHEilv0V+b)M_z+Gcl-n1mzft&4)jc*hUnAlx##%
z7#@nPfqNWnL`@FACRR&u8u4$fhh%|(lpU-!KT2_!vwd0}bnR&kE>X3aE4ietZ*V3d
zOjx6eOS$!}s!O%a8#9+uyDp|8Tqd!9fU#Mh0{*aBpDbWv)~A3=G3!%+ks69gc%U&H
zAuKRWYdgY>M4F`>bQwCVfqRLv{lVDWu0fcfXgePsl-hb;TnWq7sw))RVZjwHh*ctT
zC31T&x{_i$UxHM_a{2;OO=3uv5eZY3O7@hoJ%Z&1;shzSPlH1++Nb2I){HfOkc2f^
zbQ8i<v4~07V--##H6kQ!?+K)07GUF1AU5-D(*$0QZg-(Y%61V2l~$J^rYuK4{nWhv
zbUA9CPN%a)^XmG1VOJ;P>3?1G^K5=Gnx6(>Ksamu-u%}5a!SjKm!sk0loI(i!)x`H
zA<V&B>T0N7rJOZS9j1TP{5pGoJ!_u1A1UV$uoY`SGGl*Wujju!!M{x_b}tOu9d!#c
zdrMd;;uSw6MfMeel~O(S#M#k?BhAyB>4%H)YzmA1`SLgW|Kq#G;_~XRP4jlVcz1p7
zhqL$1pV#j%k1s|antxpX{=)w`Y0f9J^Cpdk1wZV$3Op$1z?Z>rIO1A=FD{1R99R@3
z(vU)PH6J$9$t4E<a02)|K8cdx<Ix^YhUAOkX6Gb{7mf$xqzatWyvxZu<`ctXLMj8%
zeFa{!W{<t+IEcqC-OeXXrRJPDq8g-x-6vtMd_sbHgQhOjRMdz1=%HSX{NW_$<CDVc
zNIyFTz(siVviaA)|6@CU+7|;xpvd%kGCBK#7NQC`VT*F@qnN9=di8Zp)JqST?V%Wh
z2S%+GUGvLqGMm4;91KTJOy4yx28+dLK6Sjj{M*OHudfz^#mLd}#`D>9fhC?BlL#ne
zw#ajoJJdK?UI31<Qn8CCPns9=+3?k9aoW6i_PlvL`nWi|l_@WO25(2OVY(Pi7gt6$
zQai=V(bepFJ{(;w`G?*5eRMG%{5<>ErjtQDD%u&w&oR=p+U1s+cRYPlaAwigZIX^{
z+qTnD$F^<T{Ni-baq`8sZKq?~wr$-$=dW|CX6=W)9(UEAW6e3xbijUjCsD#s5$Nl_
z{XpNRqFbORB`_!f)qxcJ>ggzeqsK5ye0K|Tc48iL#`{Cansfes{o`)OU6jFRQ#Sah
z?Ux)+72^WqogV;|{HO{%;#L4AVUJhX*pk4O=R$f%K2Cn7D*^iNl16|iP>eW_(D$i!
z?l(+wqSkXoW-hP3OsA9D;+eF!tws%_Hc>>30`=7~37Xe_c~-dWph{`%<!;8U>BDfw
z>c^*H-3LEJx3w=wq);w62d>kT+hGr+NZ`(_u`LTYW*op}a%sg;-0+u_4@wAmwpEu0
zDl8fQ+Y-)Hu}0LZ;ZHpXohHr3eM-pTg>WicO9mS^T9lu{Z<gkFqJHd37+ss!c{DKS
zg3mLmF<Xc@-j=AaR+kODc@1?bjp3B#PQM>@(3}6<xmAp9HW?qbs@2tRlo%)LV9*_^
zX-OBMok;^m^jOCKRj0w}UbKLrUYlsVy({xYr@F`ka)0)xK0~&b+<kz9Km(yX8fuI}
zy#RPhO$-vYp-@EKYV?fxa2?2o7u<vi0tbwvt_-7cym~!WY6EevmmXhHK9$#lYddb0
zA-)kPR3y@@rVf#eyQQ_^AT+a2-BU_!POJU%BEYh^joLqoHVG}kp+`s~4qYD#_Z7kV
z>%_^Qt5vsdG9{vD(mCyMoMHSJ)}Sn@gS{cZ^?5ehl!4=U7XCu(ZGxs&rJ3%KadM*D
zl1Ps4y&?FXFx`-dWS&H0tbKER@%_QwJ-}auMn+q$Z!US_|7tCH;y?8WXgUXcpH+Wd
zS5w)6>PEz~jfB9H;k-)T;MxZP?Q$AH-BcJ@LR5e4C8>Eqk!~zP(_0Ud1wiM_-oW0Z
zCU&VJvvbrV$b|B&&o+t5X|p9OKKD~R+-!M{w_~UiDf3sVr;?U~@c%ufz3J2UEi2cc
z7*_Y^;rVmc<{$1zN8!LkYX2L6kRj}CNrlw6mqnPcT^Q}Gv;0B%Z}HH>-xFau?7CSM
zB=b^uXb2sXtcQq%Y&m*H=ZOH5+CV?;MZGplNQ;QQscsJCgEn}gs!9q{<+yUqkIwQ+
zdM=5E#dLQPPuS^oKYX%(=<CI*141rYCzr0`lLfo5RRY$*{^$I!K)Mp3YJSv@eQ*ms
znJ^tU+R6z@u8xi3%7n@pUMS_wlC@T?DV$Oges9%cD$-O#!0;_eQm{l}DCtT#r*vb6
z%7<(FE(cwjs!}ttE7bs|QILbwH3w7`8gA`0sG&aXf(}fzrhPLhg`Pu37+Bc$7fk*h
z<S+CEec<SJY5=U>MByfY@-MksZb3YC(E7D6jJF1wA{}f(4V!Rn<7c%2b0B=IviN$c
z+CsIZpr{ygAA|$JCYDX`T%ZzL<H!kq(}+kdgpxzfJ~SR&GgGQBt18%BXlHOHj8*iD
z7x+4~Dq_Yl<qinqd*|WzQ%wtoeJyL?r^BNoABQl5X~FVG_dny}t4PUCw+E~3$&DZJ
zFPNlC;es2xNGent*SU#Rt4M38*zehtd88yu>Vnr!jBYG=qzW}u`+cU&e&-hVEt8gO
zzPU%AhTi<C>>5!DiL<xs=eL8UI12xM{GT0z@1sZxf)hhn!_70dhn@o1{=CKf1_LC?
zB1<}*M0Of6P|s@sa!h3MJ*bs&=|vYSL0Vgz2+EJ_m4#oc$CvUlh1QD`sLr#g4LAM{
zW!6v9HW%cIxzgcbS*7W{Q`+N@Buz+%-SU!LO}zOQ!$)P|l=+u5=g+2!c#`fKHwo<!
z`U%%R^wXDj1_?Z+p!DYgfsRy?RF+hK<t34N8`VFZL;<!E27T^1NaF9XHpAn(jjk_H
zk%NvZjUk=Sud!JHm9c>6R~CX5FHvZ_4E(!nVtnKco!s0N-Pc>;T^HmOZ{Pfi0pK@i
z91cD3R*e}G7?eVZ4y*)#(*UF#=-Uvw-!&GTAOjyomFaF~0yf>DMrd$eFb|O7S%dM9
z;&={^!<Ch2rJ5d$6%ZHgG?FDF$A`tLN(`v@KVMHD_fOYDCl1tVcbwi{%)tC}U(wjN
zUES|M#!zI5dgK7#)f~Lt`me0oXt#&2*S*_M53<NF2(`YqDgpBiT!6Tx8@`ZxZRdO3
zy_TDa?cQysQ-PY;wNp`&yT^pm%&*&9AQeNuHHO|Bo8O=BF2K*v<Fd$@!pf+`d0`{w
z`RN}1A}Gj8XMwwzq+^87ubNX}NzWfuI;KgPBczIwZ9rBSpJ2pYzWJfmaph%F_`fSJ
z<#B%ggQa@5*Zf+2Jzye~X{*zSgTyuC!yTqpYL-_FT_Unl%ucH*OwF40FfE-`YU(b2
z@cOFuCD}Q^_9T$?OuqWfeIc?FXS+BV-#2(u^qz57&$PXIuHubdALq{FM5SRg?dJ@5
zgwoO4=IbeB*0XLrm(Bjf<XUJ$9jVmHY^D;ThL5Cnz0O)kCLlXYe%Hj{x|t?HW5<f3
zO@H-~v7EA8(!P*4#L90S7lQ84uSq0VEv3!ZDBuyDWh&|7JX&k@r_w|Uf$o#BmA1g0
zemkFCFUD#gv+SazzueOh&)q?d2}q@f*D(4Erxup(-F2yw=!$o}=LhZs1Wy&xSnY2y
zPOf(gWxFTO7Qme|%Yr>_9cIQP<zI#LM&fOX=#5ORgs}N!R++haIT8!D0lir`GUFDP
zVrSG?Ns#0_j9#iX2MZZ>M5s)n)*fAD^&dGPb2+`~h2!_K%q);jlRjt}PU2RWhLW~R
zv_v%&&AWV5by>fW1)I{*jdPq6%s*hCU=7<|t>$cEh5#P`x9jl6>E(A_Vbk0sYT0Db
z3dKoh#UGPW=fP81-FR}V4yyHbHBg57H_b49rYrKI`>^PCO>d*PN#7C%a!tyJ9Avnh
zi%1oJP^lcK(3J8jZq1~7`}-Mc!fYctWWcYJe_*%uu2(}6b3<@5a&vA!4Yl%;=*@>e
zM$MZ_xB{LX!h)%=1(gNXqDw+@_ajy3%+WYm53^p|4jAOK<UR#>k4{=wm1<r++_n-(
z-59Lf`M6BwQ<US&kPXP@w&R@LeZqniqJu1T&z~`2QS0;i{I26L>LhoL=qz+?HCDIO
z3YR5q@+voT)!fYn@x8*C)DmblM!rG2(t~7jk^r2m%<42~&&%pprt6*k*+<WAZru)8
zSd`Tb_{Hq?Cn-T@^hTSw_6CN{>qvl4rIekcgpnd(a~6;Pi-lJIVqRqSXtPYTxI6`Z
zYsi>RCh+4u4$ei9B_w+s5ZhLm&3fTzRJgeZrcbhbK>!pafD0>wcK=Ms0sNHS4--n;
z0Z2fl{QSJ^cXBy+ijy9xbEb#z3kn6(Sl;_x3A<s6GGH;7p<&koyZUHsa!=Sk*ml)M
zLU$pJ9<-<LNB-vS%_+Ig=nvO$Nl<%)vE2QGZ9j&NdAu!U$ZlGAJ7|RXP;a+O)z&18
z8uC8#>0$u$T-_q^@}NVH&8ZlcvYMKL4-lwNd|ewC1JBG7&2B&KK5V|%d9`m{Q?@>s
z%5Y1HJKZep(ZIi@U>c=2rG@kXj#<k8Ks6+6qw(|k{&pAfw=msYr1q_ye6Co1Ym$8%
z{J&vFr<^^kt~-k}Vy1m*zWePTl35Au(W{K4)AWRmCo7ewMbY?+K^TwOqfGDfc>r~N
ziZ*l&mkR}HtGdXasF>KUZR;r%b6vHL%9`uM2<~g_mwth?03AM&7WT15{`Zb0Vv_^=
z*%t}C1HEcK&WH^@5(M4v(jn~mo3QiEZv~Ty$FMuzpXPL@Fut^B^W0Vq?^05e8NAW?
z9=A-^(%KB$4hR}aoNLiaTopK}F@V{1d<39r9Yhm-Jyw46s(SmgveD%gr!3?s|G*zc
zdNGqzhgM)VRqJI|fjJc<1{i|uOeC)(PU<!b;-(9n=wvQ$xg&>zU#FW0{8*ck*=8Zd
z2K%{am<E<1aUoNe1Ul1kIyW-oj}6w5(X_XwD#$YsYM-YcSzfBWy?l-v*nmdgF)Gtk
z7RRyyy$0t<J^Z+cFQGEd)7ZwTUhY#C_qJ{;XP2-?y<lr=Hyy|0BK4Nz%Qu}nWh;da
zbGKqF^|ZY1u0T_;<%+J_;ve*F7rsq<pvH=lRBWGP(<WO^a<C(QI@oRL$`-$#thDLl
zPlq5Oix8*O=>L4iRW6qtYXfQ++f=gC8-BWt6Vxqwt+u=zx%hpldulM6Ji@moKswA>
z`ZW;Ly(q6|^2X}CWT>yNb8^Pm@YNPHHvo4pC_vu3*U$1@C0?0}*Jpn|`4MxRnP@n{
zN7lM0cY)L@g`Ql)f+*^K#L#`;fb8Ta272shgBb<_Wv4Z$TUp3W{Qy~`FhBDVKvht8
z=oEQ4bT;{^2Urp!{fmxyUlYp@AuW2$XhoJPMjQp5$GxjNDm$M`04?A-0ZuM^)OLyB
zCZXR?zB_YGHU>pz1KRREAZ_zhO>I&YDct!VWBbpfV*!z%*U{9C%+;L@^_w|SO0-;%
zd#k6Twug%SfnJErL4cxvNkqN5sx<9N)hH%T9`kpd_`F?{JFT5Lw@rG#4A4yi9nO-o
zY@#FbQ_w~(LffUFj@@LL-y-%pl*`ll!jE(tvKy0GqlWUrH*jW{ypT`qY7XneWZKSr
zux2ClF{}wv=!-PzD&F<gF+1xHAA)iJKFe5RTC20)a?>9EQyl*paNNm<1xYblr}23G
z6pHrB_%olC_BECJS4Cg}`_v%KAWD_xlhW*U5j9H|r99&)#Ux*I5_x}#WWg6p#|_qZ
z`T2hHTt!1POfV`q;1zzC^ov3u4M5;wHgc9Zmknur&zNXS2bEg%2k&Tg?Rj@1fwbju
zU!YnuJc~HT4gqkNt>blWq?b}kZ23u`erf&p5Nljfg_MJMR`wwWhqh<BAr*>4`Rc4_
zBA(~vBmfj6{iy_9jgm{_Vn8a?N+#mAF+Z#P3Zi41KZ~Wvx*DnoNhmP$d&hH4#auUs
zHdy)<hkW&bnL3~(e19|9j;psJ&ulHHHd9QyMaPr7z*2xEbAg~E`Bu%S<BS^~xv^4{
zX4+TH%qxitvj{6u8j~O1S0b9nUv<F*L_9v~7Z)lX;`Z3{`@eJ<khL?MsQ-=r`+l-v
z0MiVZ`$6YgbPbu8puwU1o^1aADGfI|aEKm^7WmtWITJAO+=~zJ{65c`=ufdpMwa5P
zpfYLN;b{n&PH0zpw37S9$M<v6yRamb^mu_h^e9%O|CEccr}O<0j_k+~gXLFXAr@UN
zWjYG!h{lp`!x1lWA2-VOyze58l(8$!yvsw%OafK+n+eGLh?*kjq^1xlRUsEJXo3F|
z0BWUagajb(mZ1H1mwq010d2jxfY${tGd{)&umdYnjJr(~Mqg?p#fIq1j#88;Lfrl4
z-fxan4#1;=H3v^2)JJ=bdU#8sT1Os_ukAz#U(MeF{n0@enJ}XIw}^gNf-IO8o(?d^
zbthIVJAMN*(s1#RV2N|Z0w8U?Gt+QC%mjhhKLcD~&rv&fgn^hdDkbwY1j{Bc*rQBL
z<dT~t0o_Ch@RdFw@B~gPrErX%;-{VCTKehK(r`6TgBu#cJFN>PoB(EvFEYKdN9Q&a
zO8&|gT6=mb!&uGh-WDqRe(Pv3gMtWmSgcgQ1_wcR(;ZCc5tUhix*PLuVK3zLHi61v
zT0qIoEo!R^0-E+V%4dJdd#_%(?tvhChkOyHp;;nTz3F$n1p~VGSoIb9Kk6xzTj3Zx
z9Gg(uF3o9t%Y0Dt65O@U^?W-E_rWPY0V_ERNXA(2ot)q#N%^lI2xHX(Wr-_F+(4X6
zbK5NHg$}Xe_I7?K%_MhFO<akpe>Q4e1OOtAv;1fJqKUPxfBqDi_rLfp3Vp?)l|w4J
ze>LrESAYm;)TnuU^&f<p&^9zgpco;YEU~HAU@&aNSL9_(WS2$%OxC;ReQU=Y;H@|I
zWQkXq2(y=%VBQ+*%6))OBsAhZt}W`e{3h$r`D1PrV*K}jc@p37hubc5M>*U8X(-_0
z*VHaGl>mxsTGvY1sL~WTmnxX-K4r|urQ(XxlOYOQQ4s}<v@X3C-s)qj<@7y2jP=4_
z!g1<mMLe`5Oz^DC=WW6{Xd4X);`zP@vah~>Di6kBZDM7?v*%B@Ms)aZV-inK)upX(
zGXQ@*zBUxi_xn8HNu!ZC$!WRVTnZ2{RH5o(#I=gCct<^DAKG$KN>l`)5%n`tMIGqg
z+vUA7jqE3x428)FBOcRtlimfQ0dKT^e>i?~dHH#@94c{Ot-`H1AKSALO2l50G~_)s
z(>2o>?*G30aC!Caw5F^DGiowNlrXyK>d}4lwko48hbyvh%}n;Yg%zSkqyZr4Il0Sj
zL29&3i~Upf!C)L`lv(1Fdw)*LqS7_)pM&w9hetQ3Sx&I#IT9OPn?AWWO`Ocwu9Yqv
z?sT5yZxXIiSHd|LPiHS$_;k9PBwLqkLdT$5Mct;3`^TS01Ld&C#$_qxJfBvb*Qh)U
z(b^}-ifgcK;zKJhx={>6FeQLjg@xjd`L!lPa|9b*KGcVwJ1##ekgxsjxJ+NU3wp8_
z#o<aWx#uw}>?~zdZ%?uJoClFF7qs_awK~zJc7IjGmjS1-rG0us-Gv6Ci43E}cc)Hr
zyhiVimxM3gLJx2qrWqhmI`EurQg6{V&MbI`t9DN_IJ%TaT~jmk*JA)>rvigVn#|{q
zXIP;D!Nuhy#1GAo(y=7w*dz*m6p6=TMD?g|`#%QNkW~Eg6h|!};}mJ+Zw!6g*Mmi5
zOOgHm`i`7x#8tn=Eb{-%n1P-u3KK4XxW^P3k37e97C`Boxsi8oagjW4BlJOc_02Ha
zZOlXiC%PM;-d<z!A-Do0_!tjAXtn^SpbwR25bWuA`zH&qlgKr5rith;T59gCBPZ<)
z1%^iKx!m`ejoz(bILpDxLV@0KJEyASCyK?&mOG7dRspe#(6G@n_Vg&N=`Dm)s?Y-*
z>Ob(}sS<hE%@Cg2I>sTKas*TKvd0fs-kgejt1sUbvKeUDT5<uwz=671fOVItYvw-L
zq30oP$p}RLK^99uyu8E`K|!;w@|sf!Ex|BkVX6I>fHNn)fI9=WUK@8#01F*`gMmxA
zec5@@YitqhDj6bMm}Xm^ofKmjiqDX`TDJ{+7U^zz6&FGm2rd4V5CT*Y13p&DW85KG
z+amKlp&?H;0vUk0Mb?c><+PcGFBq%rehHtxGDZdpnv;d*ZnsF#as#p&Vg0yScT<5v
zB&Esn@=7<3n@*mJe)WplSb%LK_V+PvI0+7M%~}>Wm*Wr3qnX$DY70%1*0Z&40@NlG
zU#IW7(a-EKv~t$v9n|%LY|$_U>)f%`-1V#A;wzpcv~Ymz>9zb&Rnwj(Z(y{GC4ppO
zhBaW_?!`*cmUKDo<ZGD~d|BHAH&wjTso$#Mfo8e=`*d3Hx@Le~M6mMy{foEM)1RZV
zp$8TUj^R_S!V;Od&fxHPGp6JSL`3M57T~?o1h19n)oep=(#c%-uJ+Yd6LqM2!rRMz
zuWnfbHUY@Hfj+OW1sPHc;I#fx>(<v=K5ISmHN5%7EaiS;DemC1s|15hHL=lf@0AGL
z3d|~)4`rGKfifju@r*EM%bQ|ad?+vh<1fU>ICl{hRfxjTkB!ydy+qpUqd6+WU4GXD
zb4Ka^H^p_{kIY&hB*D&f`=UeC{2>6;usA|y>jwOs;c@pEs_VdKycpayu2$pRvixxv
z^2I#4))#Mv$FoYg&Yj`B7K_#umUt3qn^Et{dCWa<c)H^6HWESA?sOMtSmx^bs)|??
zoczKlHr_)JA1QX1R!PdX?*6QX>i`~dGIQRL!Kj}>M+jHr0@nJP9ia*qRC8r&wFNZ}
zTjS6*o}APSq^WANkH=?v3w@ej_8aBDAG2Gl>IX(Op6QjYulpe+&=5(#ysFh5cE|c4
zTdX8rmv*!?BFH_{4meM@)D$<qTLyESgs7R5WC_SPOkR0mt99VDIH-D2E^L%_<SKCJ
zZ*N_HK~0T=i~S!8;R5zzQeglqod02zo)&u-W(`R2Q3-eX|D7_QQK173!WgK5INr>e
zpoG9?Z)SP`GEa|gW$PG9yjAH+=d5KcL~K&%id5t)M#4sDGLDQ>yC-=lk7s5qSRmmH
zo0<}`b_xis%i8|gf7l5BaYgi^bxDrQa}UN$p$Xcbc2M|&fHt<?Af|8u3%dp&{BuP-
zFDni`zZO0RtJh7uAu}y9|Ls15np*tsZ^Og1mJfI;(JE+y&rBb4f`^_S|Mg<y_B#%W
z=Ajj<ofzY>U&Pynj^U=OQt_A`K!_v$au{-Zp<LkrN($x>uYY1TrlYW%NzTqtMB4zQ
zR4iq;OGXlJRE#O?D|kpDeCrd{A5rkx@$h}hqz$V@5UGJnRO2jkE4R!j2}8CZBF|qP
z6#?vN9LnE6@M;zE9uo!@j0n`_=C~;(7)vRO=Z@2=4tUh0CoVDNVEM%zjAW*dFsk)K
z)|=k)RGt?}JBoZ$B;=*SJlfh6!gWlF=uq}eOcpHO*HYrxjl4kW>9Y}u`9}uLnikFT
zHuGk3F!GE6;o9<S2*DWU@uN2M;Tp1T4*>2=Tz+18keM6h)&+J^LW`Hg*Qy=$Pi+TC
zjJL4t@)#W4hjr}YCj$u+A^q!qyT6=?ACvrA6(80#{mAs0<xTdG?n;#lQkZDMLVOKb
z!idG_`-@NPBd3haM`;UB8ZAg@^UTNA*HVpprI#=QW%^+|GheBG3tE%h!kN+iq6Ew;
zrTCI-oL+LPK)!8RlR|OKcsr*Tc`C4OhCN_r@30pr_7KX~jEU-%3VR9!o|Ybt+l>s%
za(TBz;+dh<=LCjBALn-9=Wt-r=$1WCM$sl41x0&K$O}PDK?V_HOU|SD`UDWCk%4Zb
zEHs=*Ihu~OOE`4Yt7sa6w30DuvjQ$RVp=UH@ynB2vHQq#Rv;V@5EK8X1&=dmz{}$f
zN1p)s%+PH<l3f#onO01ATpy`Fnu38h<$^}}1u0AGW(vmf@y8V}&R4p<Ja}705o*nA
z9W}hcSvkslOVNgQO`+Fyd+?v4mqc$Te+LHXBwSR7<8Tdev-3`aimqw(XaR<D=TZx5
zB^cwo%p_+Le;eE|b^`LtNyhU~HKbR0znpN>|906#vM>mKXsAqRSX47%<rFL3S^2(l
z#rzT<fDXz5yrL)42E3Alh5f#b%Mq|!+!{nKqVB+O2B}J2ps*0B<jo*+f+u$}2r<~(
z6m&M=QY`<Av@s&%t1t6Q0~O!|`qBYNSXDb%cxp8FC2;F;vfx%^9W+A2UzrxWk#k9l
zR&eoBKftty2}A;m6vS1G{DO5v%Y73TdL#=tgNbJW=PS(b4-K*GS?Df-Zt?{4qC1Tq
zTxHLJ-(v4gByOcrca~0fEz=C0KJ}7QAFmE)cyv6E9T!l%JE94&dju>dCRul2(pgwK
zZD#4;8C@+Ec$Fx4r?v@)$ye{iyt^!`eo(uWynCkd(U!j&&Tn`eASSS$x}8@)A6J93
zl}f;Tyh%PWdyJ+znWH@u98VE6Ob+cfgYfQkn{31_)ms1N3;7dch{dsFXZSQrCjRF(
zOf;Hs?eJ7ucAds<ya|wi8W^F!@d{m6-7iaVmb$~AXO=e?A4coMuqI9A`^$H;doTrP
zEvxkVwTe`({Xliw$L_ZCqm2r4iA9&lMO;mNmE+O=Rh^}(FB#_t-m~+-l*!jeLgleK
z&f=7V84d^TT(Xi+lFF)h6XhRuHy_K`5Q*x>zv_I!oYU7xQ`P`_QZjFsrnc?O=QTWQ
zTQosmi<lx#k4F{J5BYT&0-Jk?KDN<2w;f%oT6%HvwqFSpw){NXJ=%VYmu=`?pu8YM
z3|Fw5BZg6m&+E1hwaEvYoOs*IU2#Ir`|gqzZZ?;#1b6c#YFqhQxlOujB!ay<g+E1T
z2PMbyd9QV3qTB&+HadE>-sFsbk}~dZCx1SgXM4LJRIW;ohHyR~wp_|K9~M_gOc{cB
z;r*3P%}M;7@%AmFaWD&a{@p$a#o=AM#MdaI{+4TQw&ZLT{(#r>j1Prsj=!%UAxkuy
zL(vSSWDLg%B}{&HrQ+vHv@~2?T7Kv9o`T8!&4ojtj_3@KPizzF@S0vX!6S)1DH9%7
z>P&gw%p;U$y{C7$3QK;?9y1CRrJL6j;6y)s7=|M8q~jR0ed-uqB=a!@Zzwd}gdMwI
zG{G^X_X-^?Rnp@E#$iN*hVTxY5U+`rH0BFw(V1n$h%Y*@p57>FA6^hz>$mzIl=4}t
zy_<KPA;keq5g5aH3}kXJY_^ofB_@q4tcOGr2U=`imhKRThx+HwOibXy8%Wa+56IF|
z>rLNan-=qaAF-Mdxj0ng>(kmmkBDxb3U97(99h<$GFKqSjJBU{594w#rYhT{sD-OX
zz-k3`cw4%`wLrvLjXqkhJJ}$XNPCDT$6`lnYc(RaZ!Vs^etq3OF!AI=`v0$2<{@(Z
zN6X9!yfvqw0^p9@Z*bmp8{`o&58zDQfYM{$@UDV%E#6d}lvz5D<-LB3grkco9as}J
zUl2nvm_=qj^UV%C3nLkL3<Eby67&8AZ|yjGzH$cm&fLLib21<y3Tgcq7?0hp=~}Wp
z<QWsu#v_YO^kAYRB82zk0i}=ih2Z|l9pou7g^hoW3*g&gafZz?h0r3}@7_RKJA)<}
zxyN1}*k%G;xKgTD81CEXuJv?xUi=e3&&^Ud8`WN*TVj8`)<qFPYieOZiedzW<$NP<
zy91gPsdOjP!4HMOi>l1@p<%0xpMsmjagHN%LwGQcG*c;1kojZi?nivr(8Qad6WIzP
z!A)OZ0k3|R?%6$4KY2^&g`zM7hAl{%od<LvR^oZ$A=tdg3sn{p<0MosNBIVy=RMuc
z_mLMXb6ByC!P~9P6RVl2Ev!goCL%0zq)=g;NiXBzqe;@!qYSKA+a(!IQd%cHxD$2R
zS+X!Cq2INj$o8TB`cc5+ZBxm@5T?8<l?<Q01N>e#7w5yFRxEJu<h*4QCNWn=6@8^~
z!qmf7%a_O<Dpf8@JJh)4lJ0`W)iMM^BpX_N|E`>3DzZ#b0^LrZ8Ws}SxupvX{iVY=
zr0}WG7#drhxOQ6(@edmQ5vEQ4x=uV;c-72|1!9C5hjFr<StR0J3I_Js-i&OSxGH&d
z0*u(&?OWbI+|)vQg-;D}AJDo&u*?j6-qV|vbj-*znNUvV&GB;&U<R6nd)=RQ8Crk$
zMQta-ohRqTzzM#eg>9ssc;z*<*{0^mC#W!j%ew1Z85O&io5U=WqH?-qVm@=DrPdrP
zptfV%We5^bQ$h3{ew4XRXq|?;(CS?L1RQ7LJ``n+(c4hR2MffuOGY1tgv^Y1Q^>%P
zoBV=fiGUrpOknqH7ew=s3|0Pt^Y}`SPp&SrU?-mu6H~`J(@)_$%C3iXs^*0&P1~uL
ztvv<j$^1i96z1Ic_!TmZsV+aF?Nr&Sj+7@bhyba~pO>m@ez(thB)%sEn&q6H1#mDS
z*X1G{CYdW)uQ@)Mx#>_RVyZk7VQRx6)9;4a?3zLGw<B`hL=9KEHywcL5S6u7MXBG|
z3LM(29WGVh*v9u=-b;l3`$B>#zqD=&8Dv^`vH_H7^MM5>@bxk&8_<dsm#?`OXFayA
z`JhBBWHBiVs!`<S5R9~Hz0G1I0Nl&FT&Lv3mIjsU4aJp*u*yuDbFqt!m!WRdt(Avs
zi7hGigVDviDNb%x&qT>`xtn1>k{Ys^cpk2zEjl%X(+grxvE?S-)nK<pyY76Y`u}9<
zUChmHXU!o~bi+*|1UhQlF`{(U&xtnD>zy!?*_FM*x;&yjVohs)hN;hK0u)^YgXhRd
z+dk{(rFUmy{C<7$KDw%2Q~EU?CS|8d-`M`#d<X?*!v9Im-NXB3B~}Ho5*hs08D6Ja
zLRh_Pae?75m)uQ$k_4S5nh?fa&_d-^r+clOtI|H3$}<7#NY)@XNq^0Q!deV8H0q9w
zOc=M!i03H1KRd_HroN?79bjDn8$#z2li5|hMgU<(p}P1M;vL`^Yy}cQWMysCH5FS8
zuC6@gkt%(#As~nw9692UO7y1^mGB%SomhPl+Bp<LyrB9>z2)FnQlTMC^<lq;k$xR@
zME(M4wKHMZkBoArBK*~a(L<?G8<kv92MW-$U}a)wMP;3C{oop+T)>fdW!}yl2#4ok
z*Qw=}Y0%ryEe$ChQNIJ9z~zH}x9eJu-S4$oU;Zk|tRiYhx^Wv5RVoam^rxjUPZYPx
z#G4lE&|kk1n7-?^EQ5Q{E#`$DWGf}#y9PJdwb|PTTMu=W-ae6Q_pM9_J2B|5N5g~=
zyG;r*qU^_0>Zk*s`2nwy-<W&+u4iROQlWz1j6YFjf-bYJhK!iGm3KaU5Q9S*ZFaGy
zf`rKr{!xk-tcmTp;p~YG(jd4>1T8V)=fu2nk+LDUG`Ih~&fR2%>jq#OXM$Pqj^)jt
z>nRpp!YZ3{DGq;glf?1}n#W68ick$!F^EK>Gck7Nl`xQWQ?yS2)~R{8QQFQaN4NUe
z{~=|u1{0Yv;>}symWsa@-=Gq&0uldh9blaQc$Ml1<iO}SCOF_PRBUR%ruMo${_uZz
zmGE&`$AN5mE!Ahe$U8Cz<Z+;V&H?0LC^z-Zhcmiiy0~}S#|!k~;|RcoNZpJeP%ACA
z`FTA(J-0Je1s{wn>0`A*lREUq#5LvGsmMabj*n}#1|tFGvwTO`jZkaYIk9NFMjQNT
zV$EfICe11W3|yTO(=`ANj)?WX;}**9aD|w}1&?7eV#`pVw#3+KO>)c~DES$6x(W0{
zzZmtkuuu7_0Rw@{xCfn)_=%Qr($E*}mLlawuy^J7SEmWxYxQ!B&A3mXerFTD2%kO0
zHph9P<<x*R)j^qBXq~BVWZ$4^(oLgxa|O@KipwAQbW!Q5`8_}xTXcagyvdHTnQ!E>
zDo~+do;RH+^1KyN?mj#AvQp*PiRZ>%^9;<zx`Kx>+A?&$ID4oX!USABQRgYzjP+2r
z@n&kGzUTMUPi3DE&3$~Vuu{{4!nhko)job~j%y4_#euG!GH<`57_lMRV#rB8diAG`
z{DI#E{wLYeqg;R+wx>M^Seg_l)PDw7;4puJNhxGWYBz2eb;Z%5m54c7i$%+qoH6j>
zkRcpY`xKJWO%zXqX38*vY!YG$iqOm0ys+|5ws`otJML^t`go8oci76{w6Lnn;EbXV
zEt!4jJj-tkB{{dqBzYEZX--RpLz9*qLsS#t(T0PI!yW;>A06hWaxq1gZmVp+D4l~4
zfev!>jGE%2v9Vkz#9~&R>qC-bj1YlJ;x&#TV9B{oBP_j*0`Cv=yXyv1<^vWbCHG=n
zz5;@R*;zd_zhCPGKg?dvZy*?Nw!UsAFD!4vQhkEIYk>mJLi>cb(Kgqi>t7AKkB7UY
ze7>(w4}yRJi{(ushOH@u@PCM*=hf!x`yhO9?_y73;O29E5D6ImIr~xaPv4(2^%P?v
zhy>{PcvJh{A9eI(6~sTgfF61hNmiyuOgr5<u?#~N5D=gZ?=v@c6*X(+7Q~G?D#41n
z4{j-mf^)qI2a5VIjefm+ys~}hWSpgEXB^9oM*$8y8Q#EZD2fS<=9TWd`@SVh!>_NA
za@<b79&Z*u)?aVTo?&3}SGY@YCiWaRiA!%K8Cub!wQHIToAlMLmhwSru&Z&`v>~`5
z&Ji1DdP_U!@@d6uJgjE}0<a0ZiBmGdqb`La&afxcrP)tk2~|MAF;K8YZ+t3$&yJxQ
zbOUa|3+DXTKK66I&rl6eYD<7rRPT6rinjtFCm@sEpm)udrwbOH-91H84=^)2jR$K~
z_FX+i;6Z4dYxzYsM6KUnDYno7Rk+2mYm;Rj16cQoudn9~!3A1WufMoDc{S<Hk4p%c
zn7%*kZx_!8`;3?VkuRfOPr*z&K~%x|T><!q5YF&gT8OWtAj+p8`(n-zV9<MeulKu)
zuOBCmwzuY<kB3pa>#H4uUkJxOZa1Isuvv`eAb(`H_H`WVqh1eRXD^ILLGG588}?hE
za;CTc$ZMgA<6ZB6x^zYy6v~K+TXbvICpo=XIEZH_%wt=PB?~3?CNei~1E@)LNCA${
zIdr1UaRSjfxbXl1Zh+6fFPm}Gsz%J(x<V<t<zH_oa--6fUD-P-dQnE@bF=C3o!!O9
zXE4>YqA%|WwHl6Bj03_2t{2D@v8d1Xz7)x^74i-{B0*P30wMMbu{<_hHxzU9jfz3G
zlPcgaOS_h1f=D+VMJ>1fk1tejEI=udBC#>M9y?+!_^<#GTBy=?hExTy1?01mq0RgS
z_ikNoqF**4dMgaD4$?c~L#@_qWd(uLU_YzOts-qZp8(!E*27SpFCSQSz6e1LlJa;4
zmRXxr0jCfodO53k;r}m+S@|bOR6%9!iYu6IMu<1X!l_ag;o`Z301OKZ8ld<+<W#_@
z{T`C~TSyQqh>948!cCDr40euxtO5kW7B2{aLWjjMh@1zN*!N$k;Q=VSoQ8C&3)3Yb
z1Nh(aTwFVZMRLVCTc}12oHBT|!X^P3ma5q(3y9SKkvGMS*MJ4s(g^Zz73)P<y1yzG
z85LadR^*B0=Zt6gwUD4K3IJ8PV+ok;AA5zMrd(BA1sf8J(*3=KARSD6nhwgByROl~
zN@=m9UAc%r)-zWfn7<4{`ij6GM>cEK7xFNL$(3w>9Z%aZLiS4D>?nl%<FV<$Y9Vpy
zSoaYV2wptZ*E0b^KBbSDgE7?Qmk8Mojw;xo@)e2$x@7LTqM|uIFo4X}**d~7V@mC-
z(nod9R4{i%Fl10@a1?AetYXE3jt#@K##;3o;PnDnwpLdtCdFSHgtZ72c4A@E#Z~VR
zXkGM0^ThaQ^slP^AI#*XQr0nl%mNFP6EM$M#vmCb+kDI|X)$kL8P7k4&wXgO@etIT
zs7eiX?X*#2$DinCk^mjF-5g-|c}WGg3}drnBGeKM7>AB&zxza)jY+XtP@+?z#bv?C
z*7mjjHr41-+=Mi=QU%JeAXb$Uf;dY7f;>)Y%_Fi1G*01ULDLuK88EC#W}rN6*>7};
zck~RgXTUl^ON}H=caUlQt`e5=B~M!lD6NJ0)4aJ00*r}A_5sOpV8n71%nor)AH)IF
zzoJ17bt-xU;!%3l59BvhIfGSTUscf3<`Wy#R^e&)?WklSZ=lT;r2Ve;8k#EzZ2B_~
z2ORH6hQzI@=c?^C@ueQh7(TbhgZGom#Gdr6D{Zrow=6TPJ*yT~WAjx@>*U74qT;@r
zX^g5g^(qwnDgjo0BBvetm({QvNE0;r7vC<@{?^_Ad8hoQ%mkvDY8UB@JF$P&K`Ysn
zB$X?*GknRQ?aZW)A%tOhcK2Z5Gv%+v%X-;`4%kS^hmD-TyDisNgz?i|FR5$}&uK3B
zrNBmEx}q%-ip3uo3|%lXx84Tb2p)o@C2ei*$80wN7$b^{gf(1MXH0I3zZ}&&+w){^
zOuOm|0bcjdU^M9rA)*TT8y*ME<u%sDu(az9M6lbP(GQ9+^eik%;Na>;ITU`Zc>~^}
z1=DyE`D-Y0e=<5c@HjX0eG=_)Hb2bytQRZ%aIkEa6A3$h<J6-?I{`Y(v$pV-+Rr8w
zJa$h4CS2%1%LP_FT6aU|?rNv*e%<#ICCJm*7pasZNtJX;mw2_Cq|r)I^${BQJEdTy
z6CKA1v7HG2a(lI#N+QC3M|x475?|(p?u@I$^q>(A${%qV2MaX}Ic^>-K&JMXi>#iC
z?A2g)eAi_D-5Ug4U*J;w?F)z7xpk&by|#b~IFF~A*xF{DW_MgQcO|_Z{qR6OaL!;G
z59V&D7Wh2KD>t*fM?bJlRHW*oBB3<eTV%bpPV;QKVQjWck%&^<jUBdD^mtM(tDsIq
zZo%t*P%ZrfZ8r0!uBB!MMJ=4$;Z_8$DU(gvDg%>vSVzftw5u72*&$s1Q_Z_5Sn|jk
z(6NrJ*+{C?>MXn>k(w+4iUGe4)Fj9Q5xim$XeL7^-6<xgjOG6&{5gq+4hF9DRw}P3
z`KR*S6nOl6rz7q^6u<F3VHa?c>~I|%RLoJ>i%Ox8FHW*Pn8uNmTAHgaEt;Lzt^)&B
zDN{-(6VNV}lw~=~{HSR~Qi?p}fa_t!tMtPc1cC^pM4FI#mH?kL=_h;82CG$G+o0n0
zESMDYM!AkK0d1@C_vfP6dEV8MB9UB2-uDmiC^FdpL$q8Rz+DXpG@wr+b1wk@+xPpD
zII)&VgULb4x)N2EPiCxNYEl=&MTI`EUEG>U1B$E6k|DbM-Oq&9z7X>X&N28dlCsGR
zL0RI}6@d>DkGx4@CK6z&{<}Dyx`N4G-xKYLNPoK>;O8g!;X<t&7sBb!yaRrV`Tgbw
zO@z*@at<ZBD~#~{#s9(mu?0|(gT30G<;!m_G}@)|`9;>~9xO}&-=pC+_2Yu!!3`fN
zV}4g>rBwF1#i{xT6l1CWF_{-~Sjjw-5O}5}*jt$N)1QTPN--i$F3OG6Rm!A9bc5WE
zHl^p|_4#<_kMFlSdgK(0=)42URzI7QERE*1@`%unM@cu<yDsi)*?7RopV|v90;~nA
zWzCHW=Z>Cmjs~ZYdc_9*ybEs8z9#$@WFA>@`-|`(lYlE8QMxmhq@?qHZ!z^jM1p>E
zX(|1bz5E$%m_Bs;bi4AOO@_sfC<Rn)vl;n5?fB^zzfXZ-2T~>QuCrbwFY4MwbM%Qc
zAzHIR-IYqO%jlY87oh+UX-VewD6`<y`eYeX7qneTSg<q9E4cBZ)J_<GXzJg(!|bev
zA0p7@u@{Lrq*~kHwRqciV+|1rbY=IYdyUz<#5H!zJbr4xhsA|>(^8pD_l-5T%vF1X
z)&9eUO73ta+aem&UCN@Av_J1+O)&;kj2dQ&|CUx&6R^k!bt(YQ2Uh$|aKlXx9k-Lh
zF_0ZMk>UKj%^2k%k4Y`=Y2@CL1=S~x?`3NEc0JJEgXAj>e{XO{ZPL2qbdlVAnL-OV
zR9pDS#;rfi6DwmSF)r9sKm75naFu*HwWlN7HnKxn*!W|06Pg_pD->q^wPgNEC0H=>
zx0D{&kN0<2mVp9-BU_$-F&Wj#r$4Nw@SwxJWx&Iqj`livr>DHKN1i^)=OkGts&n9g
zx*y%zposa-d3UkpD~j|uTc|gCTl2xLs!06g<u5BDOR6;?{1H7Qdg~P^ERSvC-RAv!
za8a)0dKtXIfsAwMmFzF4%!w4sk*DI+g|2b5vv`$Wuwy-d2Z8ZE)$l<kGRIVzqu5$M
z@h(j&1;a}|KQ%=;Xh)bB*p(uKR}p!xmneNeFP9-CCHGfbY0^eNZI`k<Xf6?Sd&z}I
zX<iB?HaH<1H<d|a%EO`PpC*SyrU>uM!B0Lixh$lUGrVsKHKA3J!sHzlbp-bCYBA2F
zC~iDdo9_S+e_d9s%Shy`SB4L}Lr<ymSml(HludlWsBJA~&rS2l#t1C$u|Ag@gyn|R
zzJS>)A`wqsfh5>ei!++B-uUVcQWu*ZOI2PNmy6-YVCVkLkR0QzVVAO+ji!;kIH_+#
zQ+eht5ju^Ckm4N!5_OjcmNzqXupY-hw$kF6efbIi$uCaz4!SHsC%ZA~_Yj;6GWkjj
zC{~4%;T&`0B`SVY*tj+=d<1q1HtP1HdnS61t{M5lfCv<8mEr6O;JFpc3Lx^LdFS!q
zS%wu`zsMa~`AaZnBIO{A@dJ_0OUNYUXcK}{lq2(;7ah_A<Us>?PI(4@lA2St+rg0Y
z$gKl1@V`Utyi2n!kbd-+T4aL~3{w4j9P(KRars=r|DHAzyfTs8vCzd{@%DRM=^<YJ
zTg+!V6Ui2u22FR-geE<i)OvctvmFp}xra)%YH3)|*#`6RyK`UZ_4U#ZBNJpd#>Br6
zqxzS(a0B&YlfN||rw6Gj1XkRNN`@)s_$z>01dNC9=zcTJldaEm$i?$?{7-+&0@Mf;
zA{Mw_bFNEDZDYl{hR2Ctp)c81iAhHFU0fa2fk#L6ojP}wD>Xe|dr-{0h2CWz>F#R5
zS55{0f^xvRP2%vCnQ<(-%`uI8fsE6oGN>TUxed>?euXMazU&WFZ{@IQxclaByJf(&
z-4WgFf_*J-OIM%VXL=}K;o4>v7mfQirl0smFxRi8vnLoQB+eI2xk+UmdQl5o>qgDh
zjp-uyNlbVUgnnKZv>$}K?R(RiJ(`6vHo;IQvFXk|gzchiY8tWq!GBt&2)MDyD*bsm
zL8NyNg5BYyMjI2c?7g+5<Mg@3rY8Ube~crWOFA9H4N$U-DK@G!CxmLEI{t-~X~e#F
zo+QM4HD|ih0qy^=9aNSGO)itMr<M|)cvC;Yb6%@X;n7N!ow@p7x?f`j_|3T-Zo||j
zylEp->rE5jA~a#Kt*$<N!2=PhJJ%^OK%F#=xOmimw-eH8D=@)3iP#WC)kGV)Zyhqe
zC+-+(FFJTvakKfldgAJ~mdz_ThTz+%1kuq#*@S|gv<Ysb#Kq4$B*Nf82BOp_juxsa
z6xoZlHx=3r3>6>hgfo@Zhc5F>kV79$%;R;O0tO!(1o*#xL|~xm{%-*wtcLY(0l>o!
z1XY2d1?+Po{}WK0Af=A)!W#$~vbz%sIcOcIKPZAvKGo8+uZw*9hoeg=6&`R$WPwDI
z{+s{rlVR}q;vG2ltP0w9Z;&9Ub0erD#7_UXfg9}+zRR5rhCqfi^7O^=??_O_>3;H!
zPx|ev^|gMhcijvfV-au^wGDRBoI+f=%Tqek1h5?97-?7tyzYHIXP!F$Zx+1oE%t>+
zkKk8(ywnP5+ONDnM9~y|jrFGVpeWlBMY9va;T>CIY3$&qww^M#G;<izd?k=HznQ{u
z!*eFeh(5ii)tzfCDMW(E-RnmRZ%I~+8ZqmJ*2q^Ms9}nZEkiqI^ZU}*N}fH0V+l+F
zeFMrEi{%InN56{66<g}XvIC(f4q=|g?!fU97&!0ai(rlaCbQNXG{AJIQ*YmzVLEY`
zy?MlsU7B2L-t^bPW$SYaLN49;|1Pf|nBn99;0iH<-eMWj>?Q+!+3B1F0=oIX?;fC_
z^=}+BNuZ3wgFFUgyhV4l>;!!krp1W?cvw)#*#&Ws&cCFn^t3b~v%?VBP(!5E=DWj{
z(3k2UDh6}4pp>~;O2g{m<Bf6TKDw8K#mf6hVfFq(D+Eew;}LrK>0=NMs0kjIb)MB2
z-T$;DjO6FPL)!k6Y(APfflj|1G*;*(i9;e+il*e9ZMYI270kwQFS@iOibk^w=xXw(
zu8b`9DxBj)Vkwg)LBeQDzs0!36O-_X7A^~Y`jOyhDkA!e_0)3d^tuGtc!9^%UUNm3
z#*mBEo760xI5P&NUH4Hh1B8`H&$c-3ckU9&st6=ZOhr8PCqSchnOe+~uMMM)x?xr&
z6<2FXtVSwH56>#($TG}sPSG$22qVMFv5ZS=jNulpw$Yucs`%2x)oI6%bI`m0P{E5D
zpVjm7kv!MEGOp8Z(<QmL)bf*byJKEF<#tc>62~<f>L$#qZuSV}uem(n*O~rRElH?q
zGtg|2ReDbIhZ>#flrILPVNoWK7h>>NuDlW#<k|LsuevqWdU+21c<Y8Az_b#FqA`eU
zB?33`-BsD3*h2CsZ*ERKn9FG<rs6$mjzT(NxWw|7rravbGHEQ{RQh9#{akS=T{OAa
zvxd|K)27hkTKY<sU*{bW4WG5#Spy9i*`si$?#zXAN)?<Usll=M8`C$t#$H^SMcxT2
zy%_n7-WFdcH8O%lSVS5IU>5_1z4IlzfSMPrhd9cX{Lvu9ElZf;Dt%6Ng?Fo~`0B1b
zR?LsdC+!nB)%@#EXhG_NtdlflRQnH?WIYq$xH3A1Ba&7TEqF}1T7Mo`qicruLZOyU
zhzxa`d^niGVW|vazW5q1njcwOPQpKr%zvV?Ua|#}7MwlLvZ1*)z{?T#!8A*;m&spA
zQXkUsEDAQb!`X2Y9fnEI2-#KZP&wgT>`3RvkSbaQnqD)mT3Ux<MuGwsy|Xp?tf7Q&
z^B6j}nDdNhhT-!L<?5N@hd+K?5VlUqTz&ex?%O||!e3?&d9_Rf{^YfXW1;Y}1*k@o
z2ZrSU5T`T3Q>>jO0Y0&tB--&t6-zEPc=nvp=T*nQ(|Eep1uvV$`_~CZHu(&TKIkBi
zZ4b^vGSK-j+6txyLG33U*`nHzr$GZop;5*28{%sy82Pf2|IvEf2FSTYr6YdM%gS|}
zlL{Ra9*(fVGUJssv2J*GIbd^QIFoiB%q@E}vApa8@A2WB09MXAKd(yJ*<#Y$uZ~Q@
z9WVOuTC_|A|9siQX|NF%<7*tC@Z02a_XW-sA}vC$V|Ke2ACfPx5>qVyN@-}gx=Z=y
zFvx2Hvz53^3XXzVNv(LIA?(FJ6HwW?*c1!k)M7}wHm;6I0fBvpINZgkuww7GUUcp-
zgd#tDEfO$+0+37f2iWob!(EqTT$cfh(yF4{gp?oo;=QmO0V_ER+_N}&=$Uie+f*mi
zs#NDexpXUW^!sYd(B5VKc+Oc9R82Yy>MPbc-;@>}kiG`vLPc!~$+@cRKf2lEVXUuh
zO8GIjoPerB8vACpU?vl;aWADyn`J-TsIbhPVwba$0UkbGb&lCpTEsxHL+uW6JKOe!
zQSBVpFJc~3vtcK*``@6={Ylr$lg(iIt3P>-2%1z!)H9N9+z!k;)wK_-%Vp5V+GBV*
zGDgf+|K*}&c3fJsjo$ac>)4$pjDx$Ym&xxG&&L*QW1?fhJ*GR|a=H<D>6U_U71@_@
z_?&U70p|W`g$}FmpC?f@O8S}kI;Hds)WSMJAf2Lj+X|Y6{kjk4E>K<dB=wO!sHZK_
zrXu)~X8W~0U^W8D%wtIo6#DV<vdH*e?r$Jd_V4!xp%|{u%TR8VxkW%NO5Vsb^Lk;9
znLvIPRN=i>OZo-Mf{KS2Uf{4n$iB$FQ1Yok0Zyn0)j(<T;sr86WimBtvzLRkhk%h6
zxy4O67;7Np+;;*e_0Pf<P+)2ZnQX(i<o-`ckY|rKGkITzB9HC7B<1p<BDOsu%bwr^
zH=v*FMZT|d5z!SY+S2#16Cmi>cn7i<u&=Jex1THmI|8J8_ePEbeR(IS#rSf+sd9AU
z$YAR7kMbW3U@tDfhZ+2k=YXLEVVXkTA`Eudy6|66K2?9P|J%5MaWOpTKree@8lY$a
zb2{j8%XR^?00aQcYim<2`A6W^#Yu%pqyXfw1b&LNXE{w<rjL)Ulvnh40i(`V7fbo`
zi!D#u;oayvL8&MCGM&Iqf_z*PWrv>UxMk_nU9SaK$i{A4%0-o9bjKO+e5>LMxK+{R
zICpv4MHVR)HTCnPj|Y!c%j4OQPRNN8iO*3sIr>QNDg$s&4U`wi{BBjQ!#_G1Ml<sM
zye<)X+0>IJZK~;M&BJEfT={+PZ!CKe=AK`#JUHFb5)JzYq#J?xKAxzH8L}-EXauvk
zwf1;8U|ASi_}-CQ%5W`k#$fxs#xm3v5%wI;zzqxCYC$keL;8+1{e&B%q|2*F-GrUd
zl@oX+!VOrYk(Phxd?2C?Ur3=A@PUDP?p+Ro1_j`~;HF>sKk^{%Ywz0!y~%uw=N?u%
z9?}oPDa=MI=^}M&KMmo2Xn^`O9M;q2Wb%VyV}Qf!1DrFmhmNIv*3^0Pp5SH4yvgJ?
ziPFQ^vC65{tz%Mt8m1bCo_#04+2;ji2qtZPE1=jeL03(L3D(`d1D^)~jEP*pnHFXY
zP-fsQ8xuNkshqhRP>0C8E-2`~eqwvEPjY6G!mAKzh9x6bZ&sBe!F$s*{cOun=eP94
zNBwNI@ac1YZ<z=Y8xA7Dp;Lj6AmF{5yoa`F1Jb&fUe)SkJmWlroIjVxaq>z$1;He5
zq?GdV{{X8%RKLcGe$xx(NF_h3=)~1ByfwdGKQ{kt>bB7+>tg3nD+_<N`|HO>UJ$TS
zA3H=9fx4f6ci%R@w^DN~+c70#a<X=8{KU!Vl(m7iIprgKLU^?zR#pG}{N6QB-rwrp
zkwb1Gw$=b!056kttt;3|z*}lo$N?fY$hzL1Vk70>Z4$}Bk#}j%q3mlZaP_#X9PAuD
z8)Y2XQ*vZ%kyy}MRE~dldV}?s-D1s6)Bncu+vO3;dXY|4*Pe`Eoq0NrQLM;`Jlm;v
ziZF~JY;1Rum?TYY^yg_VHkKWiT|DR5pbY^fkN674nm)f~Pv75GVFuAF05jW%$?$nN
z%g!m3uHZ*7Y|YGL6(D$%Ca}fEy5M+pQDeWK#WzRxRAXbKLe78NtSRLlW^dY@@!_H}
zV2x3Oz%fkJ<a!<jUBVSh!k&fp40&uios0HoYh5|6Q7ez8(fV1C&TAEMAioCEDKI1g
zT^B2MH**BNzX1Zt8bx8yFFJNT5TFaQ*p{bBKV@(%HYHfzK2$Kwpx3XET~VyR20izK
zh^WZnY)3jXul;}3d3nm0hED}*YBg8zXQ`U~f<AdJfn)W<bN0hNAM~BPiBk_1pBL-&
z$iD?{u}F<YhpX!f9LEZT;GU+1ic*{s@W-JvB&m8KW<_#=DwgP<=A|!{!&2UOHR#3S
zL<GuWEPfl6k5Y_chC>+0!LaLk)E^IgppNA_5)+q-i^6}|?QSR*2{;a;oV*Y?(4Zna
z<AM-C89V_TN(;|Qe{7~-lPB*POcL&|gNZg=C2e(NKDvcj$VL&GlwN@+>y+Ut*hoxn
z!kmgp^2O{?+#w~Tk&JqgISTX}Pwi|ov!g&EC&O#D>9k8v`O~ni5d8)I<f`vkJTny#
zMM>-A$@_n4pEp{C;}m#Jd?0HU<~&Z<ndeA3ob5rpWS6|h0OfQ#^gP47Uj9{Jn4x*o
z>H}*0bAd#L1wucQ_;Q0pK9cPGR-#@0YN4p2akDJYqms)dY53E#!10rD-j|&6T-MCM
zE2UMShd3nY)YAgm$Hip$E&$jAFJG_u-D-aHt%iTJmgfm;nECL6eHbpqfSnQ6>MLmX
zS<|;YNZ&4<f`Dc#+%5>e!!hTPG=B9lpK6t3Uh63=d=&3%LM$e$0HpD))N_ve7%pvg
z-QdUa<cJ^Zv1<I-Go5qF4f8xjWLpW>vEkc}Cs%DEQ!RLHQ+p1TB`;HM@*8GAL0Spv
zE|q_(YVWS+qKOyzo|E=5EzB&lG%BBc*Kk(xL+X}^S;ZB1JT53*6LfcE8I44w%UOjZ
z70Aydt*{Dj<SV_MeMQv+@09~+CEZZ<fIkQs09%3fC_QRtU$K!GR0w7!$z5M;bDOZ0
zxPo8OW2ae2<W${}(4l87I@~jL8;SP%)SZ7yI+eQ9*|((bR1&Pv$xPdu+4_nPfReL~
z-guP8wtl{IUNF^j@rwz-ROy}S|5WJd@_z-^mU7>yT50`TV)ER~t(@Zmr4CuZ%xSQC
zP|(}d4r)ScZ(;{b+?5~x6F6rF#jX0a)PF|qeAlqp@_x7rB{0|{Z@Z4h>qZ*<*&2W1
zf5A}_7g?igiyURSMRNbnJ8H`S0e9@Ya|{(YyGUS5Gpw}iu4}!61La32;8g4VT&>4j
z?^CUx)q0m>=9qkcP3s+u$c@*<>JF{<$iNih|ETlcZ0K0y{k*I^xA8tz{HyPNX5T${
zE@t%UZFkRrJb>(}u6uKi87KY)Q$GSOn3vJFFcg>a3N{m$kpC|!f6ZA-kK;BBzUNo?
z4~$lnNIif-fJtUjptmmOR`gWt?r9I*|9>d;v?be#^O$K5jvZMhDLy|@a`-XG;kVu4
zefQ<0aj?P>ST($VAAny-{sqZ{h6W2DjE3R;=kVvVl*;gR;0vYHnlH=Gj_<U8(%^Uo
z5MTTjVE<$h!Z$oef2;JZHt8G9v(%CA!1tFh&fj}!WT4>rIs_{Fzute}$t<iC-h{_+
z>;JvI@BZ0U(Ulft{D9sG<7xQ&=PtZJAYrT@z=6<e`1Ni0zWbx6T6szSB>j*ae*GA_
zk*`^KUb6C}F+`2>lo3R3FvbfVUM1z>tCYmb!V_P-cw{vxe`pr@<>X+IA{n8z?<ukt
zC9!IeRf>^vq9!oNS$?AU%?OKMDYH6)p*zY-HgeV3UJ7lopRH$d@Bo`o@+#pi%XC?#
z7xjr&pte`&Q?kjKJm>)eqsX-x`MN>pEmy^hC7-wvjhvCPByL4>D2$q|A!vz~w0GJ+
z85zQdc!qwvf7-4}TLD)0{ANLQ1w{mb%FZah-a(N=l9k_6q(u!X^ulX1^8y+CEbE8#
zF#bf|&B)a6i%)4I(7KT~UvjiVxKsX5{DN$;i}rbz38Tq^!5`=?zgzW>l>XwkJdZ!0
z>+p(oXek`1!6*_0zGjB#H7OONyq`KC=hPW-S|EOZf0&b3!l3@-Fz>jAFps8H+3>~8
zl+Agp4zcL0QZsYgQ{58o({xE$b)z23$QeDBu<G>~CEiqS6gh!NJ%v!Mt3GsoHkq=`
z_Z${syp(`D!jrf>8HZJZx8Akqsj~QG^i6)Fo~2WmpJJ(sp&8^oBxDlTViO`sHqC3G
z6r$1Ie<qCq+@gm)<lItJ@yKu?FRU>SL>V1Lh1a7BgTkSmQCdM2Rx_F~{p=g69#|_3
zKu_t8st3*pPJlf@It!2b**7eN!K-t6M#<b>Y#GsOKVQj8{tCDsXC?_z^*}&J3{o7r
zN4joDk#2Y0meM8HZD-%(x-BJGfs+}wceCXUe;a_}vmIZ57RL5^ez0HgUd?|x0BD##
zs{S`bFPHxta&0a5EvwgMxd;XHYps~$3Z{mUf$s^yX*E!kC4olurltGPfkfdgtfPaX
zR{2)yKWFaz)S=k&etdGpFe+u<{yUt0H{oD6bMXHKMu}Kt4X!O>l;su)hop8q%<0dW
ze-Lk8s%*BUH)90On@y!-V>wHhvAa>iuPoEdcGo$PthTQZksWt*kclGW>SeyaP85C|
z6{pQ)=Qna+T}R={?#Un2RFd;oFl%U<Mf2PxNi<;<X1Rl>c*~|aMp|KBPNCBVHYmgH
zfiUR#m}E(=lY{ZfW}9|iFihxrZAu9be|VwS*4<-YZ6#CZu-BeqC*ZMGT^4V(L0T4W
zxW0^41LGbmYq&CrO@z`R4bhsK1!bhfISW!oN<^P>=^NjK(%~1rOiSRMn^PfM^W}8O
zqqEU7!Kk#4A=54_=s0#TUfZt6G>5F|^@1j~bV6F&F)W6wHTReVh`O|mvwk_Te@~n<
z5o+Ge)|L2u^!ir4F{;jmQzfP-ysXspUeoiKUduLc(hbbcFgI!BS)3wvi}nh#+rV=9
z2F=_VtZAtPL0Qv5QkSvscLQKlfzEcEVh$^fdsNBh@c;nI^Cp{vW1DN}sq?Cfwq_xL
zFWLE{4(sS~s>RfYtN-K((`I?Wf1L@wi@x53a`yOUmYTI%dJ8IN1?%_|EZA~N77WKX
zhlg0Oqy7OF!~~@Ne-^YLtTmru!G@DmK3ObU(WH1(`9Mvba(sqEw&PtERw2{z7O&%e
z&cX}lQAN+WL*8aRDj?O?d53KA)=IUkKHj<?fMXT9Nx^vlj8zPzCMGxwe|X#k``dKt
zTFOp2en7hk#wy-jX?HV<y(maRv33)RZDjc<5DO;cH2fqK3znAz(6~tkJ5;^8o;E$k
z<ky`bhIP;6414_9c@--!H+U4~?#py!!(m0oCFHA^^$VudXLIwG<C4meE0<EH=U9*{
z+6}wkODGW{h^9#gz2wXZf5wXz&mEwY0qrK>TeF)fnQ<|bQj?i+f;tP;<QL8SW_zM0
z+sub-Ej;qR4SVg-1>csqZ6ARwfiyk!5ntGe&QfgbpNcI)eS(jx2%tbXcw}e88v)a8
zNG|O;{L5B%W5dE)Vd;Fo;u3?EcrYCDj9?lDThn)!QaNS+3z6Dp0kW6Tw=fl#qV6dZ
zm%%*-D1Y5rxsD_^49)v1>IX&^De3|S4NT9mfHN;9GaMOSM;mzmKQMKcq^jy`YS%~5
zUDGOwq{v5#q()AEPICHmIK3Z!dQ&u6;WSvB-oH%1FC>41<Uyk+3m}Yw>HX{U>!Xxt
z_;cb9NvRZnEdTBJsgDl|9M1sFAASpPe6R#z4S&y3nAR$j)+nBZ&-?^_K80~!@1>D}
zg6GdEP&)qh{_{a*VWsdU{23nOuh;j(Uxz9>Dxr;^jJ5(KOuv6Ugck%_80#lQ;en=a
ze@tHvzl>B1ui+UL<$x6>jnTSO6dIwmn~YIHSz8npm1EN8!^Sl$K=WqFrA@7}-84qo
z5r0~Npk^LNr5`hG=0U7bZS&JdUJuHd$NZMf#9EJUDa>-9PDQG+GOn<^f_j<rXc!?T
zLNAG4oMr3VK!FWM)Q*NZaRlm!?r6GMKI&*`?4>zA1ULnWw8=Acq8;U*=~JZ==27Ai
zWza_qG#bswB5$S^D4jN>=;c<IG;c3x9)D!S!otc7RAfYDE06Xe6`Z8(m=v^{FX-a+
z4jEp=AE&gMz6G!?Mn8WlxL#t;IKh(#VNmtZST{s8f9tzjz(ynie!1mrl*oxqh7o#4
z2nvmH#l*N0)ai>t8Wj|!Y?wAHumIJxO4QHEuvw8?%2DJ6&l}|q4j(DV%-!|e$$w<3
z>)KGBgy#@4YNPI~n0$S|#V+)gok|uP)iu6#+ib<6nU`5x@i(>rlcxp97O<5Z0A&qM
z>%5pA$tI9QnhkDB5J<D)jOY5R4N0bv*5$EQ`tbWHtb|cl9j&FV_>bpKN?+GxCye?`
zbX^VuDoZi%ueCZAd2%3ntHG$uR)1wA1<r<F<#mnM_1Y`JYt|)Y?p2FV!`wByY7at@
zU83=<hk)|JGKPoMywQ{kJO@us$K-YpO&+xjpV2Dt*lhfHhC+YZrgUS?6wwUsyhDN#
zs1}>3F`X&Px1d5(PT|&7gE60q88i40Mj+v~>xQ>?yY0+r!R0ClAzXlk^?yQnEIQeO
zmxAsbjVa=_5VB5ar$<iz1cVl8#AWCZ2*=TI5Wa;DpJa&wyUCfPPa6<q>H!=Um!nJO
z&~X{bl^F38ZJW-wh-mer^@_c+J;~2_trmakLvLEkYan{Zs?4l}m$iKOa&nIb861n;
zo|6{u;X-mRB(F(dTBgjX8-Mwn6MiqxCH%Z+E+l3Lcxj@04mAiv7PL`9Xd{&JBgLpH
zYykU~RYgJ0@)gDJMp*nx*+H!02FfTqx$0~$xrCu{wpz%M@&Oqf@HNYHS*4fsNj}()
zZqTP>n=^SXh#)YET)UCamvugH4f$;h9Mm5gF}%%gw$vO6gNr@Htbb`udwggZ3E3?;
z-$**-gmV6hyXi(+yB3`}C%Bfbh&c`V*}*Et2mrey%14THJE{!{$H(zNX$#es^HOcb
z+3Gde02{5{B0=_)f#%S$SfHbyeGgx2g#nmIYXNhGu*Elf|1dk$K!B1;RF?a?vErPG
z_QXq1P~Q)ij?g8vbbpi8>;0zc2muWqdP^*4^x#NO-%qayh$zs(9u%{L13`QCuo-6W
zjYi2_%&u@8D4LQE#*FfAl=M^wa9(rJjYb#<-q*LOeHs;-rVDjmKHu($?r1gVyUQKG
zrq3duf{12e8r@2=-H@HLnGc-A%6Z<0o&!dTvxzsza@Mf_KYt)epB!yIfXeH6qMrSr
zqTmLxloXC(pty#jxq7(7vpE=6UIz=8Uk<5^U)JpISqVpIE@Rw|ZoNUVmMfTx9yj-&
z`jR{CwfC_jJZ`yJBDvmAZ%Eb1@^OHYj)hZ=%E7d=%hn3n)j&`jXQTU|U(+>EkS%;l
zWf~h|{62KjN`H0t5d<qvo}kTL7PzF3GvTVzXt>%<KFkkLn7J$P{rO1{R?%54+TFbr
zt!M6LM=u>H*0bk1yHnekJ6(O^#e?d{dShr*INh~HYZvd7`diVDZ0d8>n34O6<mU^L
z*hQ&*uoQA`ZiSuegB{DQ+jc7~=_s68i(r4xJ!eQa8h`gtPe<*_tsT0ijW&<e8$k49
z$106r4&F9v*5y0OL@zh`w^rAdy=hlfc**9n3GyvwPpRC9uF!X9JG}YMbEC_9o%EY`
zoU`Uk-RTWI;+bFU?Ivnlec*QPT0a?4a)`S9n!yRL$Hn)zgxCBa;(v#hFdvY@a(X%l
zo7=!QcYi=F2=t2%C@3q)(GTf>f_GeM;7uL7!a>$TrW<KPWsdm+m-_DlbF2BP6D0RS
z25R^=z&e{UCN^*RqR8c&NZr&ksEgD^f!#=3&kzVgJ93t$oV@7g+tHkIU3ZVX=UENb
z2u}TXJgY~F`xMFp)sGj-Bf)(N<$>y%LOCgd{eL-|i(!<+Zb!}-5RKsos78)ALTJmF
zN@{_m*7kEpX|KTXmdr>E*2{Vopb2^bq$bO_BAjgLjUg(~E@7JTh^us?QE=Y4(ERsp
zG=K>f`PVfXK$AfCziTuA`?H#><~_pVI>rA_qXCf5ySbD$|B*femr#QA_ii+JFX#u>
z{SnUb3w2Z3f6c$caz6SurgrC-!K4Kf0yj99ak@Jcm&(5|D}RpLI1s)2EA#_lSj8$X
zK*K=Wi2}@Rg1!yrG??U+!{q-DQ(PrcUgV|wAk(JED%NA|*2B+158rl&x83K_(BPD3
za5%jE7(iZV`2;PKmInu*tbyU}*YMYi)@bE9$U|BiBaf5sUPj)3GT>zeusre>VE@Sx
z#67acVBTwN-hX3c6&_^-8DHbP><`-NNFm50JizY%e*0sm%VD(&Hhzsm|Lg19?!R3-
zIvT|`43<#=4*&ex#d!i&)`bB@uE4|Ze}^BtKYOB$m6k2}gZ1$H=g`f3FW3bw*lBB-
z7ucay%zm&oC=*|0?C6WuELhPfkH}7f62d~p&!dNv6o1(&V?s|+K-7$dnv5*yl5k)_
zSPOz=^LvK;F3vM&&T&tgcl&rrX?rBa`Jj;u*^U&KdAT5cj88tiVlzx139ZG9ptXXA
zy)uG%5kxd89V&98Lsl0y`8T$M7U4LbAPG?f2FU8jn1o3<9LttChP=#Fe00IgO#)Rs
z+CHB(wttd@LZU)gE-=l%mP_>+t7rlho72kMuKSGfRSlLAJT~Kt{KQaXh$o6%ef-r#
z_Vbx6P<vmH&F|BG%<7nw!qo3=ebj|;dUv8e7Wr-66=^nw(*@*;P}HJPRw16GRxhY(
zT%u7><}WkdI97$8OEEiVQTBK`KAt~)lGWrV>wi?EcjkVZ;Zj|l=$JaQ3A?Bft2-74
zi|FPiU8QXZ9HMh|=}z~b@@==74JMwDb=5&<eCsc&Ha*VJdp*6jlcYGSmG+_|#iHk(
zWZ6ESVVv1vHT#?yXZ%g&vZy+&a@f<RMqm{|e^!9%*}OJh%+|2=Rz@2%77uso&G&C-
zQGcoT%|S+SiuJzj%o4eJv@R&5FmpCUvy~CU9OouH6+Hdie_p)3MwNmh)6R({ctuYg
zIaH6kiPIS`=Oh*g_DfE$W7BOTH_a45I!>nAuI9#UfB{7=TBQT10??Iopc_-@KNF}+
zd}9bo$Zo!p95}d!wXx{1p9R@_0tk7d0)N^mfK+Ic5gsN-8&3kdxRVkHih~Gs)fqRS
z6|$Egr4`BJO-%t-vzLq`I2F9_&zqDDa|&>S!d*iLEvR5@&uyMP6)pQb%3WT<*IJQt
zGLpFaFr6*w>^MW}MxB)`Su)sF3JaQ))&4_rSRzu|Ju!t{5^Nc4b@pRHisC8jM}Io}
zVrHIFdPojyNf`CwRFk5K$9B{$Hmg%)dyGjgd@U!s4(C-Y#n}Mk4PZ)GaY7!<0FhDR
zx|sRC0q`YPG?Sm1Cpsw+7+1k(0#0-eS2858bqYO6_H0gn4s-mtKsMo5XD;A}#wl=S
zg<no^w<sS<h)EW|q_E8|->i`sP=C=1D&UjAVzA1IpXz~qtiB@2XCO>j#-UTI$CcUm
zpwK<Dxa}yW{G8{7>IHSXA1ULc5``&5)=Gq?OqHscj8o>yeSLQaXUW?PPfn<W-gO1u
zOetsMck?w8hyf+42;RZZbVKM%UFKtYkiI>w9xv5W%Gb}>7k@3Fmja}+O@Cb=8HHji
zsv}u28;k4rNZt*KS$TS5@{WNO*;^cvX<e92c|5vka`adpe}e0o#8cv{L(_4c<@6Z#
zY-$A#Q^h}3_V=gwNaBeDe^82#B#ztReOFgp+2M|C$Ht~+#~i-I=VIXQcHnC{7R;>p
znD4_}Ob<FS9w|>yqEv*kLVw76qL&G_Y*zbRbU^EPP%BNgwKr?o!h5WZ0}8ERNF6Df
zX=CgXl_Kd}_v2>%*E=<(uLHAQ$LmG=mT)2N3wjKTN}WSK#vGRXx8Kzyb*(4Vq-=Qe
z_3|sg)II5MTk;X*C+CUfcHPT#53L-X`di1`ODiX6&7O9;(k)E<ZGT$_W$jeP`oif@
z5Vng#Z*^8Dg&oX8b8W6xB`gukP9gajbub5|C2d=>ujeJ)bO)PCVDXI}dO7v@G)WL+
zvH=}~d5OJEnzC-y7}+GH0SWC=;J=%cCWaXBy_D}Pm4pCB?wD@uSO9xT3*GgOMKY(m
z!muS}aMft@C|TBrw0|tXNPf%o#LA?j_{63~NBK(EqFenzZ3_TWWb6mGEjq3bwk<kp
z8`>7;X>E%hzuVgu9fkL8TXf8>v@LpIA8T8HHj3QT9W+mGTlkpwgeUrjCBeI@0P_%L
zPHbxi*gMwp`HygZW-XscZY$@zQ9{rEfhzz73T19&b98cLVV9Z(FA4)UG%=R}HUcD<
z%O@ua0yQ(2@nSY7e_G#f9JvjC_g^8PFV_caL{XHe;$R?mX|5@pHb_#StzjJO&GzD}
zvumu^hqQlvKWaSwal4Mk-fiRb!CuV_B~ldi@uOx+;p~Ddg^OdhUv)+7w>qNYgoGSP
zwL9!$AT%UEv{)m%l&n*VE>#-^S2^+y8p+vXG{kBDC}Y5Ae=UiQIH|Q5EyaTxp)SRC
zFQHX}%WC(8qTLg;@Ib0st7g-Hr`fD6SsmdTtx_DRHf?CtUaz`HvDsZ*jAC+vdr8f5
z%q(UyLvwJ^riK=|SgSxsj*cnX<TE>t*=)(l!T8n^$Awyt>{yevV0a3}o)&V6g1DRl
z+=MCFJ@|>#f5t29Qe=V-7@lG=Tj2M^WX><>!7x`ClSy2g1qIGdHV+Jtq}X^BJ#w3e
zLR)D*g=wkx0Uij23LOg9#bj2&f5OdIaJxuoJ)kJJr<I&&ds-u$-F#m$N^M@Nwg5Fd
z_={W)&G)KkY1k<mB9`7fsf4E-TVDusHtTC~Swg#qe;$R+K1?es9E-lV444dBP!yw$
z0@syjfv4~vI*{qWq&c-7P|n(VKm-mny%+`VuzPTm*rl2oIoWu88G4Yj0h$g<4tlHy
z+^$#|=)e?|EnXDiuSrap)TwCeQ5~Gy1_%gj+Xfnwt7H+OQCy8N^uyN{FACEt+q4wn
zT{$6ke=V{r!*y(PtBG3=LaUfw?8usJs)|imOEZ_DqeL48ZpSWYPob^tEwyM5p-ss)
zS0Ii%-+tRYKR}!%Z0|SSU;p-ZXM;n_?BewFaBpvo0EOZeQJo}t+wLPI6L|}w`z4vw
zS47na$spdM{aVs>-_K9yi??S-lc{6t;JVjGf6L``F>|rM{Qalp4{w)8%c-OO#*6uE
z2^}w>fP(f~USO9{Ox#~!-N`n0yMP<XCW4E-z3%m5K6yJ`9(1ptzv$jgKP@jUe)q4l
zDK!5yg+a6BbhbRVm}(lkH`DX^#bPqGTd5!Q)%5+z(RcGt2LYGac2kXq=)XWuNKxPM
zf7xs{?+41g`kjM@s5Xs;sF#g2G-PNr`NJ#2+Cbggi{qtT?Vrs4>ALUci}%w-Gawvx
zf9PIz-ycwa@n$+%9>f}bHK0h<7X+kaZ!j!Uy=Xn`o;i&Ew)<iJZr(k2-%v<N0ZkbI
zo)ByTQ2g!Q-VN_O#)t$XGNSp?2mq8Pe+28DMgSIwuz^y{Qe(hX+1!W-tr=J)Ox`vW
z@{*6iJVQQ9Sz|>7a>lBKVy*zPtm~{;uNcwh=u$Do=%rYGNuH4*NU$VW1Lokh@Zrv_
ziFD7dhMz1?<}+CImp3o%@0$<H<=Oe~y6&fw<%f%7Kbe2*e!lp4w)=kiOZUf%e^;;l
z&!^q->HN6M(}`v(#{eFn-*Jp|G?`41^WMLogkzwXu+W4;cfObmJ%6(wHSF#)1x#}1
z@w23we4oYpS%Omdr1{CrQZPwL*K;(Gl5hj0GC(S?1JYYkJ(Oh?fF6&LEbLPrBgq8F
zMJBjk60r&GGC~sB%M+I|lDT@ze@*`daJ~V-cO&&f!8z9za7Jo*1UQ#+1Dwl%z<Sw`
z;f`G0sh5o$h6W9d0o(I{?RmiVJYaj?0=9!5dyrAY7PcD5dr_}QR$0l1L@R)XTx<*w
zBonLzJXB<JQ00UT0B)6|wy*&Y_Q5zKtB+|FQ^Dalfe8|4Sd&A}v=OF)f8lzhhby!Z
zDL!DGzcAX&PR|<HsM79!4QnCwp`OM}gBT-&hmp3ph_})J1Qx3<E3O8!4e$gqU5UtG
zAj(@<Zj>ONp?omHGP!M%yNnn~C&sTgxQlPK-{pE^Y{Z#<>GhtUEl)n4{Cm3i@dU`U
z_y?e}tv_nRUHp@6tUo6oe;M^EJYn^E3K5$hwJPXL&Fe@MQdkwPsMiG%fB@JAlPlE&
zN{zxZDk8yhfTkg5i4U}9{jtTP)+i`j5Cwv=ZnO2rV7?-UT(OG^J{#(G*8kPf^249c
zmM?ECVIK>>5SXH;w$y)`&rVNf)1yUa>dxCT6Z7sqrA?U=$OOv8e_?7=!2>cq6(~cz
z?c!hBMkr?i!?6n2BYmeFG3|)5+jHcV5Xu3Rqy~ggO#)L<Wp5Ng$eDJEF?bntr%-tU
zPX2NF>1@$QQ`B7xsoNYi>pgY;4S|8o39@j)WJgFwZgt`Pk{GY~afD=G4>~<UGM2LL
zJYt<>1`6gfLXtp=e^h0RBshcCujf3Xh84=I`WVVv<(l&9KwfnquR4%d9muN=<W=*~
zC|fw8umomDkHwBAtf~>8aEbC9TYpVn;|eTLfI<<;uw_FAnaR_QsvJZ5{N=0i0d`wa
z9$VJj6n3fFjLFWM?brpo4yB$%kh$t(jH?V1>|By4JT71#f2cOMQ7{1eu?Gel61;Dx
zs(0Bi4*ehMh?G4b$8qIW>qjiTfr<eZh|+4vF@be~c8XOkl2mmIR4Q5~s(NJw8yo>W
z?z93vC|CsqZ3<Zhy!-z>6QWw&-Pd2A)@U7bT60V4v!^xdBunrHfcHz<ktPzz7)fSd
z(^qq{VlP2{f0btJwadfPjP}=knyEwPsKbFu9S&6L<$;R5EZRJ&#O&Ma&4SH@kx>VG
zd!Hi8p4Svh!bx)>Z1Idu7LfBcJ(o~VsyDJq0@r;iNd=aHVb*PM`G^WIVFxA&-r|F?
zBLntW$^lt?HGL=KS}rA&jEF@>A=<DY3<(R$*99TLe~y~Iw5LBFO&0UMI7IeD25&57
zTh6Y<2dP)SHWbXO#WxZ}QUYIxyWwYqbiHylFv?!&ezu~xDe!p+T`LLxo%0w;xjaK0
zAqlD&${0J1>rfmch|(iw{c2D-uA*{$6e@GLLS=hl-a{{?p&>&f4~;T3bZ88K9oGT&
z1J3lafBGudSP2mJN$OMD!Z~3J*KHi*Mk~pKI7Wg<Z3o+z9*m-+)>Zl&CEKe<#rDV|
z5<=o_(BI@!wNGHMLBapD`Ve75s6Lfd#I{drs6aEI1O&VQwYO&j(sHcj%7$2z$Ie`n
zpGky5s;JH?>JsGZ?b%@AXPE<(K~Z?xYm~z4f2sfuVy{h7jTF(MP&Z3msd(n=$}`#_
zI`jYVEEA}<I61yp4#${STJQp*tpqyCn#$PKoV|Vd#1{DYwc?9&e(X#C2OOtjn&27N
z=cy4$AL^5RRbXexnaizO$3Abmzt*`eHGJX&H%lHPnd}8Z93dImJ{uaNJ;L|QW3(e(
zf37(sTqjw@$6IHFPdia_SxM9^k4n__I#KsuQ*HXz82*gOH2d6ekG!^5xY4gUfx-R_
zsSrn1Bx)^+0_hfwY~PpOK_j2^L8|s|_irsx|H;)umq_CE5~a(}0YJBhMjRRg%^E)S
z(*9Gg!~X!j&oY_{Wo~41baG{3Z3<;>m)^H92bWPUGZdGpO(!Uq711!50k^lV(J&hV
zf1NF%LiQddRwS1<B0N#SYV`sQ7)g-e!}FjWsNyshWTpEB7_}zSDU?gDrB4jkp<>b(
zzZ53FE{gH_{7A+t%}=C9f>@I0D$*?4U2K7IG+nm9nxwZ>O;KoqQI6;y_WY0)r$dc8
zq_MrX;<MOQqN_*}>i=O^V^KN#oKO}@e~Kp%N_2!JtrW$fqW(#_+^sBT4U9)5vV&c4
zTn{jH1Ij=zzV#a(;nhNdAL}9};u?7FBHI@vSIe%<Wn9QG4w2<<B~DtdrqNTElqz)H
zhDs>%Y<$}H<a?>6P{wr|{Aeki32l>hrZ!~qaJcfj5Gkgy80RFRLOyite1wKWf38Nb
zuk(>ee#>3^TBUS5Pg`6O!E#({QYzMo{FbQNr3Dl&B1lHK7-{r59%Pz-rJ<(lP+MOg
zxy3Q>N{f~!)6r<mH8}{_Mv7O394?QL*2{1|3a!yuwfBpxEj($3Cwoy3_T#e&Xyr7-
zFev-PZi>Mwr=NM+K$#6)m?zxDe?ZoNs=$@eB6hwMY;#lpL`LD+%cqZzXg!lkLHt28
zX#K2yiP9i}{;$&rA}9iPQ&QiDXd;5Q+;bmlK_JI7<TjDcDhk$zV!}?sk!Y28eow*0
zp=@Y;!d>N-CWne-SNByqOq6Gb7ZZm{g>`>y80i$#ftH&{4_w=k$+~KzF1EgUt2RHg
zvhgrC(&9(8+N7x+2s5L+unswUUy_;o&Xfm?x8u>N2hYsCgWSoN^wlu11P2yiQJ2xT
zFcbkemm!)L69F}s@mC%vm-yB&paFZAnb$CEf0h=$%G##F@~Wj_Qb19drz5;_%yU$g
z5{W9$DhqU7`312ha83`Bnoz|R5NElUi)#q$vU#ofE4nmCfkjyJOS3kqtd+Dlq{o7S
z!=HYsi`HL$&XQlOSZc|vg>l6mmP$pw=~cK~mwX4YE!hHLMsPJ!3Ns0>vkzMnfFl&i
ze}xtkzaOy{>zDwyx)zfjRka>wHWkaXtv!_m>bPUl78O)wC*UpNa(wc(AwtIY2<j4P
z19;-PGF4&cXAQtQwZKxcd{46!mQ@0@I!aG3w3)(H$7h(W?bmJHZY4JI7iou4&W}pM
z`FbIHFxjHY2?khAg*z&IUfoV{w)uuve+C%YwhI3vsOM{2*j(Sj*h@ov>)=(m0L84q
zgF=E8*NKZ4t_Iuv{;eTUF|=l&6g-IcE+~_C@t($S5u3czIkth=wLOG0L`98MX!ReD
znHIBE?lzjOfLh=8d=w}vxHc64z=T%!P(VkCOBel02nlo)xKvRIzC9=gl46p}e+RQm
z8PW$IdO`7y%D~p6VY0vzub6yiYlg1rO^_Xycn`gGlw#<W^_KJn=cupSbb1uex2YA8
z8Kp;t?x@yX$_%YFEJ^`d2q|ix+$vI!>^`vtLGP<Fo!T?_Jw3O;Wu1*A!aTm$%MnW3
z4#;sq&Km9kxsFne_&?EPNWp{@e<S64nv5K9xADlQnGf<+6idiE=(4ADU6<#i&vVN7
zvT4ydW6b_Jn2#{cIeo^QSng9fWnPq(jSPB~Q(QEtJhe<zo_cJkJbuk8N1+>)H{WI|
zKPR=Hi-XVe=q6|lVX%M~IflY1wGJa>GEs|l#=vo1p=c#jQjYR^L*BF*e?(=-P-qIa
z1iDFTGv0(!Od30mydT&$LFS`u@>v68{6fOVKEX|37+Ooqs+h~y88C5Xix6E4Ow8d+
z`fC@o3Kyv$kj8e&XKB41<^r1}X@M0j7(!=Wg7dX@KeB~(OyP2_{n_(&B<eyrY?yfm
zqNSun!m73Py&usL&F42Ef0{B-1{y}jJfcUTCyzF&BDIY8Nd0DB)lR*t?Yycjfw-;v
zC(a#4$POs|ul2pQ3O~yC(CALFx1xJTmWGtcD7Z(K)>E2C`!``RM*Euv_I<A0Y03=L
z$b5+2EG*CmS8)a@axXS6peotE9BqeQ5n(Wiuz88FeNUOAe^pUXe`5M`&I-!xZ9&vx
zM{lofj^irAp~0!(>!_w^Se3Hehl80bror6<%pIi&V4F@BY&u!!YCU_=8Nqx5KvU{E
zFcbjl-01K*zXnieXD&UAFNQG8Ut-PjEokoIA<)<E3smE{y|LwY@8HFAGe@r26XRe>
zX`r;#^wy5csM6Ihf4@|~(1c7-(;&4XtK|w+-L8SdOXXt#BU;S>#5i}i;o`h`gPNM2
z2R+29_LJ7G#vN;$2^{4a78l%Ki%B?h#MdzT1EXyzX;YB|1`0c46TVp3=k2cT^A$^(
z$jC>CM5JDeMC2lqk1ZRqb^1Enh>^}#L?TVQ5#ulHsjfsMe?9M*DXsH9L{X~%w1OE~
zw@;!gvsX|}Wyv4zDQLNlXAD;jM%)YvL>q)F*bjKAfYF>Iw;vFM1E-mt8)x1DbP$pW
zTuAWakiHA%C@9hv3oZiA2MH&f2c30Ti8>u4?$~dJZzX5=F4$<0g6kNh6w4Ug!eBj0
zHfz`ycD(|(f3<Jd>TCNDJd0aMh!lC;qyE`bng-#wVKG|P4+^>sHKHDweP_w^TF^b5
zuUs!{mxJOxLHEWjgk^E|YV;)RZI!tx@PH)k?oq7mDaFH355Dj9z!XjuR2h8X0M#a7
zn&wc|WK>1qg|0UR%mYK8T90qfWXy0Dvj#m~Pi4w=e}@H_)Tx2o%Q~P`x?Y3d6Ezvf
z)eAUyPWK)*?kLFTK1l6(xehSO_uM_I3mu^dCzEc5B|mu^tZu^;trjSq=$UwX4(4W%
zO4(CV+YDAIdnq)pz$*0`VYQLvUBA?_2AHzC48|@W3@t9t+1TIOONDB{Y9PQJb5;6T
zGR;9{e}AiNxy?6j!VMlGtT|fSIDcoCG4WVd^5$H16Y=SqjkJ_s*1~@|4$j_3FEZ4x
z7sqE;E2Z+GRCi-A>ZtcAm~IUgbjo^B3S^?*AVZcu$8>T$O;0m3l%i8yTB7xDacP9I
zkg5vtUIvc^sU43EgIk;35YBL5@N_8`$GlI9fAQ}*Zj06t-+E_Ma%4i@>B^`S1Vi@;
z062>0g?!(3n^YKNvK|hAI$|zDF4^?wkl0p^I%e}JXR@sE#IuD(IY8~f(F_Y2?zh%Z
zm#;&n(HyAuhqJ1i)Aqq$oy*@StaHq7cr19R$IQMz)?BT<&5?TKleU!^be4>5ieTre
zf9rl_#l$Dk`U=2EL#^a>S<zZDS&h+#JDeZgig2-XoatqLRIDsh1h8>bzSRk?(OUX-
ziO#q<bDh6*C=jJJyK>AG{8lF+&n%;fX5^MwA^)bO#{Q}S83fsL`$ay+AzH2uG2~-7
z^h199?{RAL{0aSvQcg{Iph9U{_#iJ0e`+b+d7VXQW)Q|v!*%EFE6ij<?SkaB!$`rD
zk*g%zD^OrtsEZ~hI1TZSOCvDHV%Zp!>{8X*FNWhn80*p4)!!k<cg>vnI&#hnb&mB)
ze>VbfZ{0X=OhryYRtwDQwW#+w<!E`$+nh3LhLmp%%<a7YEI^Vgua!YhKkwR<e}*>&
z|En%aAVX!g@1AEB#8kft<WJ2x@2M*<T)WYitBn*mt!;UK>PAsS%h{LzJWdS@g#<`Z
zPB-KqB#yal0TREq{iwelr@Vos4W>szoX^=q8U+I(G3QMIlY<O?dVCUEU*nK$tfGMF
zMh*>)5(9^lBq-*@EHW^-p(=aOf6N)MsvqJ_s^`x#Ze;5NmR%@8O$Qp)O{_EGC?<4r
zT(!jMs)76e;xw*$suy1h=eFNAUw-`cI6t2OW(Ux$IfM<0{`3zA8UNUY1yCmZ@3T=V
zYAzLRL$$vf#pf$DKykkvS2+Bqhy140;s5GYJ#GqRZe(+Ga%Ev{3T19&m!AbM1p+xS
zm$4886PFI*Fv5R%j@R>FufIP?R#u8&<88c*-(FrH{(Y#bYa`Hx2`vZ!KmGmlAwB?w
z!n!aix}tT{uYacR4}Xke>t>SH<Uc5<Uq7Z{<Qt<GB%vp*MeNYiRv`Mx+8|7Pl+lB)
zQX-)RSNf3kKsagDS^4Ga)v83a!k92hBp4+Uwm_IdlAM2~35a@@KS}*=gw(H+y-J=Q
zBx~8}Rc8k&jDvBuS?ECn<U>hTsu$L2AJSbs!3i+#4R}iOIf++hQgNfSYd7-k0OtL!
znvBJecSD?a24qeA267;*Uf2*(!Zq}+=SM4J{$Rq9F+?-Xf08xbNbA>vnXv+#uMzV+
z;ByDBD4>6g`fR6kn|{<Mov;HgaUXMuG{Q<<UE=QU5~QTI9zNM8!>-XOU^9+Lg&u82
zE;$IYCTZNQI!6;qiV|gzy6+wH!X8Q^x};F^?af6}J0-MS;Hu<+;aeo40IL@gVI2z>
zgi}i_K4eqS0`6*3u)<M0F-pW$lPEYu@Qa`U*`9wE$Z<a>K^eJ`1V#n{VVMLe+zy?#
zMr)z18YRT8B?YtNN>jqm<<2JEHvtda`J{m7x$$1xbE+a`nx7BL4^$P>MmRkm$z2wx
zwJhK}a(TIf-5Ho%l)rfmWjr<j6_RH*+vM-M=J^Z(aIpr4O{sgpn_<P=YmV@$rf9GC
zly!fUYNhcsY|wfe3t^yYot?1uWG;=%KDsocrafV>s8`UiizAAId$Tb|>4lx0Z8H?{
z$|Yn|BNjj&5E|!V%|}FZWbEvdL)N5j@CDR@`u9QV(>3<>%0&&`Ss_H*V>FoLsqA<T
z_&0X<5H}1kcK4LHLBURQKU`BIzj3m0ZUKJ=q1_F%*LT`Ps~AC7CGCdXM~ylic;`pO
zPh(1zG^jn!HO@4rk__jB_grtzTAxZH6`3WS(s69d`t<Emb}6H&C|v3s&0W>kCLvub
zA*O#yUsXnvw*3gUJESzQt&UY2-O9M#NYt57C4^Q-UXM4k0}Sz+(Jy?sZg~^H3POKr
zv8va<`ZHUaeTvm1_{7Mu&vJgut2>mr{lN}<j>JWb5rbD)8T?L{{77BF=&?{*-Zb~+
zY@q0to+Qii95wqW9vNxuYj0CRB+o8Kz?GX*pAR1$+-ZK2qC4qSkA~;@7)i04wLtdu
zh4+n>|7q5dkxG)3$?qqqXq@u<h%A5eW@H!(LecR$;`vnejr(OuoK?4C#snv{T1cDi
z@8V(h!lDK!^Q`A%vca##!7;+N_gj%%(WTm9PkpaansmZ)ExJqyLWMczqb?!J3~e=_
zJxS@^=*#x0mpI!N4Q*X(8!Nb#F|w;VQNpF2pS+<P^5cqbr2_F>&sIijp~QdVsC65u
zNz)WVzK@|I|E1_vq!gl(R?Fhq9G5(U>lp94kcqWw*A;g5T#A}nIp$cI+ojwd+15QV
z9hcj&@Y#e_*Lw<KxSa>ImDPRkz1f#oC5$$Q#O|h7B=(?2VaZ<-oAa+UUGM1lD(g|i
zdcp|DN-1CEbtd)~FfI<&s|A0*SYxeebJMc-OOA50o$fVfw$*J;ruB1=Q6(Ot=o~n#
zO^qm)(Pm&;Q#G#<=-P-Mx)(ck0X4g1SD`OO3v*h0gSumz_sF|(?ci4pi%=#qu05{d
zpSxE)c1FZ5Sc4wr-Gv;@ep9F2cAQ#pBFTPK1e><6Hs1FqI|d!C(RY7jJH{vhpjR7Z
zJVbYlGJ<wKOn2}Q-RY(Nk8~%#E^_3A&!IcoYkGJK?)UK>4Jdr@n|voAzu>#cb<!-e
z+OG^;C%_ulN$VF}Cs=;Hk+@Etv&D6)gpTVZ34P~`E!WAbMy``qRj!a%9N{{7CR7RU
z;5q<Q9Ng%YJ+^~%P8)x-Sk+rlz1j9GtmzihN#2?}bzCx?wc4$X5dlo|w(-$;IMi#F
znXS)wZNqku^)tSlGd!nbJL^m*ui0WcklQSd>x9F#O5b2RtT8rcEi;{>ANPl8UST>`
z$7lM|+L-or{JD(XL#N>-cMvl+(TCZ->a)g^Z1r(`T03o#0mXm+{<i<M62aFvtbM)H
zE16U?#5eYaZbiY!ZS^&I-acQ-$fLK%IcBW#lq91~@8y`pEsa_LbXg~bY<@|DgGR=g
zs%bJu*s7$1#BWZC6zooK5kGI&h+p1)o%pQ=kCK9pl11=%eTx)&y+aBu_W%tFIXPQW
z==B~c^j$#u1+sr$kEGD+ffU+Pe=IIZVJR!BWrQ1hE%o1`eMR^35@P2d%0EFEyz$Pk
z2L~eHLR#aO`;k8gAss$r935kS<OE?!{<ylOAD-#oTJ);mg>>`JA>Kj_^HE%1*SL`y
z=9frsp@#LfU8RP>Zc)Shit7k9%rn=hVV`4{PK*BA+$4Y9riNvHOTWV#=wEz!9GoBh
z@_*339ePp%#9N`X{$%>6tQX3MJJLS|#H(Fsc|+;hQt^c4fBklTY_LZcqb=T>J-YZ|
zEjH-84Cz;Qy%yW_@?+eRf`t0RPokDV6Km*WgYm(JVet<+K92A)y$P280KFAZ#R_F^
zWOH<KWnq_^1uqH#IhPR!Hx&Uim*F-9Cx2NvkK0BR?)O))v<r`C-n_XGV*%}YWguOQ
z?UGbMoFZ@$BSn6GzB$PmQk1k;AyFLi-S?B6{+i_U<>v13=HpTG<b*fM(e(IjQg|To
ztt1YunVeF>YBfE6pML!$C0Ts#@L?&n#>e5e$JN|^(8}WtW$?jMrEWhsRuKm8Xn)GE
z+GZGyx9A>Mh3g@n13pMABZhzvXsE;O?~gy<$ii&njI=N|OxCdguF6@&%79`IOeWIh
zcmhsLy2ewAr@(H&jWO+{Ju~2rZHD4wfi?}FEs!d|8SxrZ-o7vAE=z96_B=&NJ|4Z=
z(lt2}#5_urIQ1DX+k#=+S`@ShYJb3>^c_XoMHaUyha32}Ypu}d%m>U3n>SrW+)?_^
zg+|gyr2(x`q80hq`JU>%E!RrF(B?yf?rcdbT`b^|95c`54cEB5%mU4)ED7nUTA*?2
zcD1iq38`EUz1BV}dx7@zTT2b(lcfC^MfdHU#$AcF6o0y$&XCoclh4uyg?|)ff{X|6
zQZPPV8$JC%+^tW)&XK645zdWDi(!|X!xodt!Q8wu-&UAw>xGeeWs+j>EX3fNdqx6P
z0q+{|$-@~a0v8*O(bRa2qpu|g7E^;MY`Vx!qM5l^B(*5b%B_tQidtfc7AbER-edB1
zpGrV;NrDeCBxi{dL=D00a(~%zaH#zhwQIe_)doq~c{ZQGfzF1ZCK3EH3L9}#nDd>}
zd^`rZSH&!@78b6xVl&T|^Sg4d&u5K--S{9#H-3AkjEZ^!#S@vv8xg`t@o==&fZE+E
z!%=DX72(2C!A5<>%a`Hq;qm65mD?DDY61rS_<j=~z*U8HVIoSk=6}bZ)3=*nMy$mk
z<27{}NsH22V|t@oVT_+(W}!Je(MH>BoYZ5LkRn5-quJvwdaVJPdz_PfLV9|6_`Yfo
zA#K4<FBwJbSr|uWEfnV8QQq|2ly@$RD7IE8XP*>Nmry^^MWI9}6ttfDYeg69*fqRS
zaItoRynCMMaQ={C<$p0ZxAJ%!3yAjR2;f#0b+DY~ujI+y$)b-S^1@mknN+~2K5sS$
zgpP|*&;|%9?5?7&++W}uPs<}LktxxvG(KA1Sa4#NU2@JNizb2R(?jjdN^hI93YUpy
z?3gpmIm6AlX5?La3&CvX8op}^y8dyt$rT;r$#_RwWt<v$=znCTbJ7ht>_luUR7}`(
zp!ACQSKA<*e98Q)nT58rvecs9aGA9e1y^iu^6cH-xhw~3La$-Z=iujXa4HY82$Ba5
zUMmk%fS<luBBp?c&b^QgpO&r2YT<(3=V3YLqqoUpM&qVU+&xcccLRDkdGhkKmZLn)
z>pG0gILBPsaDT;_cE+*ad98SuHw=Ex&?&Fn&+_(sCG15jDaXrB?$Eg51`(Ee=ti|u
zQOJ&vJyy0dcl+b2NB)&|1=X}fon8}{iwcpqKPww;*9|fX`Q3MexoI^cV&tOHR6km!
zs=Sn{*3Q+!cB<&>FbvJ1)}@^;e0S4k!26o@{q+tCnty;)BUD_uFzMx*@4l!oi`L(9
z$M#)SSP`!7N?H@pttp%0E?8xuKnQkb<aS;ky=v-aLnyOqLsuCJ8E`K#X-0XSm(7uW
zbwXJ({rNV&VN*^R=FJ6hcG9BbgcA;`sv9517Ry~a*3~sF>L;czOXzBso<vF*MIC4F
z{(hf9P=6FQL9F1wJ;-lRM}a~+umz>juUOwh{r7&CH|+&s=ott}xz8T^w20`S)IM*B
zK^k=M)B6v~+H)LQgA!`4<<J324IC0QsP>6r*42(rn95I91gE{}K?0h@$e$ImEHd8F
zK9zv3AkZih@va6nIy5B*k^X(Z7ZX}R=~9g`v42>Y?-GPN@eoms;I?F1qtDhWza_J4
zEX@uo?1ALIU`RMxuLNr|%9YJvPzl@63o2ZVb!VDEOn_7QGG*fJPQI1~F*W%o8$<<{
zSS%B{wB4;GfyGD-W5bD-(^v=-MvQg)+nSr>)dRcoDvbyOW|G@~*rCCWp)|a$SA5_A
z`F~l)3iVKUI<L~R`g90nx<Z|p@Eou!J*h}15$#UCUZC=hDCw(4*jaC=hBQhLRb;kQ
zhXQ70Dq%^9b*|OZ{FUd~<;l{hRbl8IwN*pMd5xE!HA&pwuzqWs^i-c~lpfm!R&md>
zN=(6?hnDq4ts;F&a&K&vo@%tpW&88(y?>z8Anr!2=dGh3iiOe%E&TqPJthVPW;0S(
za?sip7}3nzu+oy^ZRst#61dMO{j+3pK{(}JQYNp_JE;{$vn`WHgGqMfopM<qOmX$u
z@9W8q-^bp|p{W=#->?3XKIOw&2K45u$a(r`H5b5VWQyQ~o!5@NW9RZv-2q0dze9L%
zll2=WDLvo$xZuv`v5L15{jLb9^F1y8?5YBDSp386pF4|Yg(*8LtU2U%i}nitRmbAY
zS8RF5EBes?0Yea)+Lw_9FB1YeGnX;yCli;@Eh#9M4e>CSf6p#H+dp4_+vsBK{XXa=
zORXa%va6+06NzP_NRwzH`ORVli$-?2FMao=Keje$yx)6R)WobTO_+SivjvWX3BZ-w
zkMD?#Qs7*t{c#_W0~c59w!kD`I10yXfNVnoQ3V@ujLl2;>-lx_?`B-1LB%!<7Q8}%
z;jizT_<}%`f4VRjqXN+I<L}|y=J#G|*{HrvUX~EmR*at^YOG@RgEgeAv{&t&3>L4U
zed&9QGPcnHIm+2@@2oZtWiU<%rMFMVP#PWYM2Hlq*n33*M|YsQA<-n-6>~V74!SHS
zYIZefDAZB4CoOZAmXIwoqQ;16B#%eN;YT#|p!e&`e`)s(P6h9Kdz0Im9FAAI?m5Ja
zXT{RhS-@4J8~ybFW2dDZO|IZPCDnI^?Hl3FBfEKiG9hh5Z1>|>E^p!pYDdqcO|t74
zS1i$`&rNW~-_cK#uRe`-U~&MHT(-<itt;>>XD3OiN{i#pp_+$7Qm~X{(6Mr4dLfjz
zoZPYnf59j=)J;B&w?zv`5)ZguMgnsLYKo0^>#2b4+i`6!j5)wqKVMMYw${tTlo^r?
zd)z14E33J+F<TYws5xMama>9_=&l*S7ni7!=j*(7&a|4b*Uk}coj=K(atvo@J7vMV
zjE&=p6Ox;3EHQe&myErX7fDOanDSz@ayRW$e_F@B$#5$x6iLm1g8~IwDv9=IDwNVP
zCx5WcnRM3kwN$Ni;y78gSX@<r9HZn*%^fAuWdMsnbibhny2N;*4zg8<p_h0Feu0Xj
zo%JJ`KkE_mN68-2nWd9#^vtfQL}-PM`q6mA31Y3)v1B|Ko6bXe4o?bJ$>HMda7O<U
z>3{=7J%35*tmo@_$mdE8mhh(>MpbdD0*$jC;3i&Y`V~Zoj>S(ti=;HaYr*4!*92be
zSpd(VlYEIicq#3?$@dHe70`A9E<b{;c2NXWj!kC(&{3o@dZzBJ)$0QTqtgNRTKTwE
z)Jl%tozkb~;3!A0=_4tg^|TRLy<HAN5}jkL?|(YieusFnsEoyT%uk)fd}5gwxsy(R
z1yXDj>M2G`c}_8E%UWuz?(2Ch6}vg27icC>NZbZRLXb0>^;{T>bP4EO<}cPOe&#_!
zL0a^-@r`_W@$qtzuQAx3D|h(_pZ3G$qx=}C<sd^PtC6i2!#2p%J!wsx^YdIyy0l9h
zQ-45o0SRAtXs=z+T(d@Gju&UqfnK_itzaN2J!n&KJN9Pj%2Un?DEZkK(_T2n#c;)B
z72AMT?#BQp(`B~<)E~c>83v3w8F&@La8BciORQv}5PFRt#VQ~e6;sP?ScO<C0-E)E
zbc{)8X$1HP8o_v@oOM49jnLV9mPT0i&wsPjfs;&p0MZ+n17k^+M8BwmH&6#1Ru^&y
zy_9w2K{xv{_MjK?f20qLoNBxzzhCshiF}Bp0=bW(4<HfT^qKsDk!pz3e=F~BKJO&H
zBY)6IKEoe$qi@6?MD28Q0-@7-C4<mOyD5dx<F}qe=p?K=ajn0?Yl%yLumfO^-&)}{
zD?<mXyo5R^ghj8{FESRr-s{){a6u9C2eJpiPISK~yV*&+5qkh6t@EG29so*BfpBy7
z0OG%?pF$oi#P?akz_wWp{3o&n>EB7af-LA;IdJ?3T(G=Rmyrc86PH0~E)<uY_Aoww
z8<^WPavRKP&^@O<O#lC3sI^q8#dc!nV3$iJkx!3{$g&*%9pvzBcYNPHpOFSDoCd3g
z_n!ms1IgbYdC=&=0tkaJy#E^hdXZ8Y{v7#3QX=xl@^8miy?;V*JOea;_$k2t$r6M$
zJO@=;izcl>o~2HF1->4_IIs88$RNRg^M`f7Y5({8A3K@BKA*kZLTW4tmR*nrO~lAF
zYtZ30YtuGg{dliW2~C{KJJO6)>+mdUe?B|d)T*^GH1hMxD+3=g@dNN#B>O;N1+#1{
z1td_r4$R>kNCg|nuzS$-l5_;wK?hqX@+-Uv6o<>V?(Xe<_usB=S)tIz4+flnKxXBi
zU%T*tfR$Q5C?!15@cYN`bN6R2HQTViMhA=pOT*wN1S=zS*pnh*_<CQnF2h?5hP-p_
zoFI=q7IImp;t7g{5`P?O9DWBCv(fM0*4>Jvu+S@_bGDkwo4%3&%9~#Is7pIn=!JJZ
zp6WU5>gae|sEspxHD0Q;>BT31TN^JFTV895qOidT+U1Ryk?kwff~Md<6O+AVMh@vd
z#)M3tIBv7DQ&JdWnP@t$k6G3+vKb#^#daxaX!dQT>0;9OFwkV;8P)}$D`qDvh+GHs
zQ3-|EyS%3_N+{&DQ1<=~5jkJs`D5vZv=<=F1nN}w6H+*HLx@lS@|qKWl8H{hPknGs
zUn9SaQ27-z$lD(EOQE33yhubVLsN3$h)wTR42luoGGS$XK<_sL2N8wkJk&8#@*ml}
zZ<T+a8q95;3@pH==aW>NfbC2Z?@48>!A-O&dyD&mptrJ{lUB^yE;E=>gsMpA2+G!U
zl?UHiZ#=Rh@?zPPwr8Gy!tn~aRw0eWqb@D?Ph?tb5syviqkKqvO0#r>jrn|Om~1`3
z<u!3ew&JP;UCyAKEEt=?aF)nM;vuaHzDvfvbovU@hwN+T>Tg8{mXR#uI%F%}nV!mH
zvUojbkIj)}HyE=c>WDbA9`6~sJXM<<n$z1=51rW<wE$l|a@os&SSG|+4h)>?-R4|*
z9P<N#Db_9#U65tmttUiz{9K-SCC+#lWvzKm)L-qB?&cECW8C$gQxNfbNdD5wxyp0W
z?A~({Z*$9(`8mfWT+6Ia=LC!At|&fj;(yWFd{wnV*mkCJC`YKKG-4%Xj4)C%cwq5H
zQVW#sj@27lFVNe6N0o0HYx&lIyHviR0l_?earstUbOLm{e47G&RQZN{<xc;L$~R_M
zNN^9V-=;*T{3uVKsN<AJemJ&`u5@u@;O?KI$)SwsX?D0vYiYEw#`S0vj`_XKm$9Ba
zF9ht`D+op)XVS~r8c(1x5jDFU?<Td0Y0Oh=l#<$F{R*3Zl_j|5WXX4%yPTxXip%A!
zo^0h7E9P9oT;X_7YESm#Eq}3EcvTJLQVxjFWp>Z=T5EART)``$aMo>E5EkJ=xN<fb
z@|Tv$1t%lla|p+zz8b<xBOepBC#TouRL*u$(0ScTQqV>q`d(tBNi}`JIa5UeXZeco
zCnF5MN_G%`bA2aS$qBAH+w);*sh^EA93`8uFT7<;mnpr-CxSp{Z{U-Ubj#uahj2h7
zt&@@OS8zVndSf0LnImD;Y!9&}U1E=m9V26=4sW2}ZnmpJ?vyO{`hYrqZjeYINZFah
z*H1{~kaRGUs23Sj<b_8wWAJvL=hKAC27OO%jN;;d(chCBV}#duEjOIil$=M1Y-3O7
za2sQ<jBLc7;=(k%C+H=gYX5dAcPY-pG=uhK^-NRKFrDL9Y5>}iHY=8lqu4IWkL*k7
zeY8S}&XuBDzoq|<!W)Y4DTiBpGr;p0;@mrKz{ka*12rh=1&z*u+>*K2OQx)m{-KiR
zgq-Amr}cq%LPPBcLOrme=?Rp?glVdD`MkY<j|Lo)NcVih4rm0SnR#$-EtU>J)s}jt
zxz@>D;(F7(aG=@_)jxm2w8K~(cJGz0>EQsJkk`F||Ij|hx51OGd|)3d^|sSO_4{~K
zhv%fQ?u+_6$ItWwoGCT>Hya_BB6aTVhiX1MD5m>@*IAw$E$f6*rYAz?KhaM&BjBaN
zJKrOKMeSU@3PspWR=JRywPj{4X6GDlw$S~R-SB@lDwU&`(YG)Z0yQ(2QO_?W3o$QD
zWo~D5Xdp5-m%)o4Czt(%G6;XyR}7Gs&4YTns`^&J!hm-jZyd;2!*YNmhOnS0uC0ws
zD&$D8|30Uh97-~&m=c*4W5`ks*_Z0No~oW)f^U42OQ4218^MjF`iV4AD-oK^!HC$D
zSd7p{l|e+`kQd)XPXSb}pHV|Z;UHx;DkhtdK&ef|(j;Q5WqccgZ4iGkHx%1E#)dMB
zQJ7Kc+?X5oN=B82rE-$E;TZV<lz=L+i8-f+Q$0copi<o}Hk`@!Ncc8K<1b;otaTNv
zSH$*63M$)-2uj&zfFbcgum?65!kNJ{gu~hP$kIrtE9cNi47PHP5FLJjN{tj>$6*z%
zr^NEXXFFTgCa|}0#W#OJq+pLiO$fQ}%S}jZycB6duDU3p2}NzL9o4h(T(pTC>@#R7
zG8T|zs*ozbl0f13S;1tC5$90KO-wjoR?iTaPez%RrBb0uWLznJC6P*{CPjRKx!j~I
zl@d@8BDTy;D&E$E_7fC;fX@mj<5UT&XkCS{ikj5a279s@#w354)KtjKtJt>~B6T;9
z$AH6NE~u1ju1zZnS($o@7n=(UD_p;2*k(}M4Z<l=i~|@0tfzifnv$rFQ&3d~SiXQ6
zNX^#Eq0V4XC1MZ3Ts3Vem;`5a1}LhlV0r9FMukR0sZ?w<k?kgMk5boz;?;}+$}4ag
zJB-59RFgjlSPOqX<9!uEGSWOv%^N%wJ3P#sJQvdrd=H*hk7dr}Lp^oK=5<wRkRbUg
zOogcu`BWD($K#?0-+$k}`*JyIpO43ru4#XmOwUHs_vZWlr2R|#s{IiU;OC@$I~sQH
ziP^<$w!tpghLN2~F=}zx_2i&EzB=t}!K3r>zuM=^<%fUW`DEN4w|{&4%Km@%vFk2p
ze`?zg=iSGvQ#YJ^YJa=>ba{9-x^Dk^_4<wb?V>%sn4Gq0G&B{RLhU}5Q)U_rha*-!
zI~)4b0(X~4L-OrxI&5cWnD>XH91o93@*kf)JHXGKBi`X}{dsBrFh8#*2mL@%><TQ6
zlXn3&5Y&IWQp!pDyutFv?N5_;llDdP9mI>c=pkwnf`e1?;Noj0HkY%ocOEX%%UL+#
z^YY|!9(A)SX-VPmNDA*l)Ui0PIBFIr7RV_<<S_3djw!5)nlrwpsE8&BI^pQc8d*P(
z3ns;`VabIMxdivWp9^?7Jl0zR2?@!k#O=92TETy=bi+l-F52^bTx2bRTnkU95MzLo
z-`L%Y_eiuEtG{jk{*Qkm8bGc{dzOaBR~Hv2_l(NU(qzo)@Wm!0letRVV5H!0piM@y
znjG@2kxl#K<YF>Cz8nlkjo5^yeKY8~(RAGK{PEA9yPuA`K{sk>{=v)1xWgVV#Ud!q
zV!MC5L~<lp%=`g?nk;A52fzu$Mzx1$&)PTB$?$m8y>H*Vc-g)ieeO<f>GsXw!w3wG
zyV1Ct8D6M6LWi@-)pR(T&4Gh`^m=r5KKNnsx#~PoO;RVAKgCQ-{l1Q>wlbT8pt}9R
z`wD~X7!?IAh}09gC+a~?9v#l1JLHgMG|zvCc0x}bWv&O)^TFxGXdY^Sv;<*2@-Bjn
zC^>_wA;Q5T@Kg$}Xhm2KnR&&*#KW;M2*Eu8Q36HsR&hNDoCOG?mv{(*d>Wow;{U!H
zoc%KTd^xQW2YNlkueEt}^P{pxcLNl}vjdFTuY+<%76bU-?x29J4#!&_6}5{i{>guJ
zI-Fd7ndfRks=lR6i@C`!V{WE(DdE<DWuS-801s_4QWR)ugMonqa(~=pBzYHAHyFu)
zY*yKAFftWqi2vIokro{C>n0<CUaa`|9W&QMlT|4z$*QEsBrCm@tTHb$QkI_R(^u}(
zSMJkS?$cK;J)sMt_@1Di=vOZN%B6o_xx_Wu@1ejik+a%9!4zYeBkxoz-T=t8nh_Gt
zjY>KJ?FSTKgc^AVxKD}`h-b+Ja3AWm`*@&&`cDgyA}uk;9K!ZIPzVl{VhqW#K_OIj
zR>;Pf9EzKD6;C#GX(|%3#}mu$m=Ny-l{FK)lz>x$cfkFt%0Mp77#JB6#6N$^es??o
zi<?e3M8p%Z=8B_|Rs~TD^$a^m(NhwX;$>${1k-{bR3X<`s{#J+z|fj<gi-6G3V`K!
z02}taJ4sj(9Mqa<K;?kCL6Pvnsw9vMIci2_mLLcSg?A_jSVV%GvSw(*nqw%)b)9Q~
zh2$AhD1dD+4dvZ20T7cB^?HAk2xguDHD5E4i3<ea#Un&1MT~nCYP3Vbtd{!%hB~sp
zi+vxAFb31l=htpB{m>3hXDtO-6Rb7vqKcnONTLUhwk|6$5Ym$*56)K$SELDOGa+hJ
z>G&kO3mziTS?jFVlHlVKWlaRj9L-`8&XqH%?l6oBDOAS|1Mst21E+sDC0<j(o2f=i
zCvcsTBo^AL!a%u#zq&-DjEcvX9RcsxgYM(cm)$F1iRrJvGfQN8O%8dJD)0ohnVz3s
zb&L9q0abyreeri8XxXSWQPwmEGC4$=O-4oqTl~KOWA}9qzJh7i!n00weX#beHLK?j
z(H>|q40`gCAA#nqY5ac`nzOF@!_l1G2J~po?gJ{qvg?2a5qcu_MB1+PAgCtdt=<D_
z2905FI-C^q;St?7+J~HyDWV26AO~->8!N=5P}J^z%pxClN*wayayGfR9=W-ae?<fz
zx$&rLql@kF&&Td#_i4K#79q%h#ZA4QkaCdY_eLEBr8;SR+$n#<{F(wzmVsOh+ZBhD
zQ+v$_`}?O3N*Pi6R!<>|6$X2w4%jzhI^XRo*wx~H?JC?Ju<R8q>a7h%0<Cxe@^6on
zAWk%IG7_-T?tQ;K5;(=NgiS^U<R`z+Edb@Pk~EkflQe{GC5_x?hDB9x_nQl1QQj}I
z-~6U_>(U*imdk%|mK3$75>(|-5uTG809?UZvr+XDRF*Dmg|QM=u#QLXovD`Zd(iP4
z&${X0e7tbdn32L_3EPDtb2~G&K2IaF2|aluclL2`*<TQN;7g*N+(jnBnL|z69`?)#
zPOOZyEhr&$UA4(M!yF`rgmM&kLASg2O~op^n>3uGl@Wh?b|a%|fIe<#SIMq4Z!QSn
zvzhY#;0^1VQ<S?2j%9<I<?g6?PDm&RW6@1UT2NbQU#}Y(G`YxmlaclMc#|~^23q57
zgOSLw{&%RY?xegsh1T5)>fS<Yi`q(7I_36a&Lh%f;F~lleQhOuZ6$qeB|GS6_iwJm
z0;pS3?FWC@i+E9^U16;09HtCmC9f$+NTP^k+l4U1O0qX8`Vgk-62<+2uF8i;LSct*
z332hxtd|kQN_|E!<2C8RQ?{$*98(Msd4{8JCF`o=vJ|ytsiOPyk`N;?OT63nQOVyZ
z5@fZJlS+i3Ws{L8lwidhj1(x%-WA+nq~S`I|CfKaQFQ*k$epki<uPF^-3nWw@B4(*
z6FuG!J<JcfgZL!|1Onn$bP1bs@J><n)%dlp;U6+04x<*J0)~;2DR1RdG3njL6d>3h
zG^Q>tf0>QOv-9r!`n>zX)xi=Vir)@}1;CA4ucMH+8}U7Pos&Ho&eWX%dn*;WcEcIf
zFIF~n;<W)c0QPI%`w<v($S^B=%^gK1r>%_23JaE-uDtbkp1xpQ$XTpR{0iyLUp~7J
z>HY^@#g}n+Dii`YIhV2FI1`u5H$5qT)mh1I<Ten!`z!PVW>^b1z-Ykk84r+K4BvvB
zf|;D=koo^ZimOCwX)kR9IjGg5NLJOWl_fp>GwJEu&BOD}Z;!;2Q{I5X>G{V5@<Pim
zXc@F-asbK_OwT{3Kku|gE7yZutTmBbPX2qjo6{|Uml<H>lD7b-TgMRB$Q;6dyq0WU
zLoy4`atFEJ$8lL7wAFEkAeX!YI{o$h`;9KlHqMZgv0<`~mF(3x`=Ygpoao5e=@uc<
zjw42Ba=PdJ6dLbw(u(EsWN7*&6xO3p-vk{Go9I|l)|-(B5lS18lwGFCk8;Ilo5zhv
zYeL~hCL7mDxTgz5tGvb)H5Ye(t&%bqX<ebE%UI;)e7a3@Y#vK|$z{$BCyE@MF#X3_
zlC;`i4L1@=Sg&-KbcWd?0J3$l$(l>@n8J5?j#kvk!g)>__opPp%b;a+ikMc=R+Dkb
zY$`|%a4b*!kO<b1!cv0%dZ2MYh2&Q}h8cjU_+)TJW#*h!TDiD@LUSp9i%JV$g+h`g
zIwrV5-t6jI<}Pj@DXw;AR=}Y-YvfM?1bM$O%Y~zsH#5!IJQIE2adcj;>7SBIz|3Q!
zabB2|DJCTE^yBH4wT-Lp=NX39+_aHqCo}czY9`~b?+8W-?Gcw`z+Vf8lpSM47<Yro
z(lL?~tbL<Q>%^<3k|551Oz$|03yLg^OjoxkkkG8mEk$lXw`nkh3IaD8MU9|~fDeUA
zRPB8M$1P1YaRh$ba0`UVbsB`H5HW+QIO7jp1z4eNt`c*gd6?Zlq3}55onWIM#@n~y
z>Fe{&zbgnhIAuea{{DFrUoaSDU69Po%4$FT`ab=*`D5I+c)C`9TGr$rqjSvZMxILc
z4_c(u)*91GX_XNGOpxr)wgrTeW0Pd(EG@ziVv@nHcQjTJg>lDq2q{=`<g@-H_Pdd%
zMLpmlRVu^IsJE+!pcOeDXKy*07JcfNk-iq5E<<_|PaBlO@CrP&=v=l3njjKNRJ)O<
z7hv8~MVgaGiUQ|<F%<3~Cw7M`86rwsi6913t5Yca032Cd6X3IP?5Rc^yUK2?R?cPI
zZQ_ZVgt!}beF>gR98H04#<7kZ;D{`-M1&o2Jh#Mk$)BZwQ7j2;*kP{ZUhT)|K%rjg
z7GN7aXB12P9d#1pZK;H7aYUse)(@}diF8mL);#fGTd&xE5xnTeu6D$Dy?s~G%B@TZ
z%qDb1QhOk;4DTGT0+Cdt^G_n(jV!}&bxD##ifIW^j8x1Ob09ANfO2^ceh~Bu0YTDn
zApY!t&{MruK<Ib<+<-9P_=bSc?{J-p_Eb-}udiq`fbWP)96KPIN5OXDQmoSa+90nS
zmzaa{fmS?!y*(~*RG-Gxj_-?0EHNGI$|!aM6UUgoVj;(+_r)ghtl-RHY~m{HioLxG
zTXZR>aw#W`^)^P2Aoyi|w%#1IlWN6pyNFzmDW(`Q+l>=_^5@<59W$uzbh$Lbv}b9V
z)mBk02xHMk%Y)Y3P1u#9(u~q*-fqn04DdcqPNfZhaCBBhF(MMx_srZswsr=&6Bxc*
z_DW`_e(h0>yj93N9@P+40KRZi8}8{;wv%>>Ia5Go5U#A&ZT2p;j3}l;W4mzQ?|n<j
z!BWsSIW?!iZ%f-Gz2#6$-Lw&o|ERh_C*KM2uG@Q4(vEUCG+nf;*yAWoZ5Fj;RTI~_
zHu>g%M4!fR(fHZMjfj*cv!b-@z22A3iEe#eDBX>W%(8poOL0)Rh+TJvaF?@7>w07y
z!V>w4cj5fYqK7hS1X4>(H;$I++6|?RZPaG^Tm2nkL0r!#4?pkM&ZM;!$O|pjr=@gz
zBYRen-lne6c~Nl0b|lj-zHOb!ee?2kI5(AlZq=&u{AC?*?whzv403lq7kR|1E>`CA
z-IygR#o|g2%*LHlm#uIl>q}29u4--YMc<(I;99}Wt&_OYI-wMav(?iau}tgUZE7a>
z?K&@M^Q%TS=9JDu4eQrW*(7<!$dTcCDI2|)dXu{7UgAxQIZF-Lt0!=DT2A8bVgi4E
zGgmo*Z!xy_0n97I179DN%Ta7Ju=}K3juaH)2WWd)B)u)Q$MdK-;bCew;V8AIw_R$_
zBivSA6jx@ZkHUvVK;COF+|N877XbvQK*K2aM-%}bHYsft0mhF-K*sPw5nxi%=!$@h
z!%K^RQFpZn*b3_qqCCxrFBS-Xm5%&>#e|k-w1k$Qx*f27Ch9+EwS?Aoyw9uhH#>ZJ
zLaS8Aq&0B1E!(bjX34iwo<A%`d{>NIm`h_ZGUL5bjKGeTT~NTjY5AyR+4*98sNtNx
zD~>y#?p|H8WV>})Zn%4amh;+GY2yEHr3qM22=<Dl36N2w&9#zM#dyKVrP2g0z^b6b
z2bU(Bj2)IHJzpPHnur5h4xdz+c%uS-fUZknQopYr6e#0eF!>J&jL?@cDl!uRH<z(2
z0u`4C<`*b`-CN6!E4dBb_gDA_TBG*^xG>PJ+a4ghT=Xu;DwxS?7McHFgpx=VrK)mO
z<vK=?h1<3)krc(pAyq9e|Gdb{uWz@nw_onYUV?~D1%3JY{i5Ip34fJ@LGLa>DdCN}
zeEo6x+czn-ht~~WR!U>w72tD(cXxd^D#8rq;03>bDs_DiR^>G?$LPA&_`1fxEIq-y
zg7=R+4(pTjGE+$KnlhE^-@pFyCQ;ZXnKr`tbn(Fophc1ykir3V8?z(<M(lKb&-|UD
z5eJ;@t;{Kug?2IOixd`K(KvJY6L>90fBXu1ifUC)-ihGjD2)c)IM{~;O*ofjsh@5B
ze4izMz`bpqf>t1bTu=uhm734gue$6~^Igv0${d}wQ1w%Ob+~>a?c|fQmOc^vHaE|^
zwO@gnP3_ymU1pwbW|OOMb97eaw=CIIIT!A+=bm=~7L|4O%iYD*lOVfrUTEdpNn}bl
zlrd8`{RxQV^C9H7Ux#m>UvK|fn4Jq+1R(!^k3Zh>4^}zhLjpe$0o>%z-!8wu{dJ^@
zr!0kpCH0S!)z-R^ALx=v!Um=H&h;Dg!ol9XF(Slf4E8&0D&nT_0JIv+s~GQ+BB;uk
zG*T>T%7+~4LD)O0%4Rv~PZ7Tx`N5ntsuDg1%37|wbdtgZJI*!>IbZ;eQs&_Bi8Aef
zt8`bND$ANq=u^Ux19@dGDl;lvyOAFUbl!7488BweS>g3!4_QlYO?&t2yO%j!m>j@A
zPoJwcL8xFxenY4}MbRo@by!gR{u30zB|-TkMYpO!ZIejGFEn`G<{@neYuz(-<F)W9
zJySQ{i{#8v-9S{+a_J$G`<_F0n|+UeMjrT{4hf>@;?yxA9f@5!X<-+lnFD#mtY$Y{
zYbdi`vzO%M6mjH)y9V{mf6A`A5gr98!br#jYF4(CDHWxoGwM5S9bFZJKt@@``9E;n
z&AD5tW4m>qFmESE8;A5$7a}(Fbe19rQC<tPJ)uYgI7sP>i{PwIx6>{`tl~|7d9Uwa
zobO6Yv-Y!_ZTDw{2vMYDHidBOLbwM)tfK|ULMFnnaX$K-Vv^J-;@U5dP%42JYlLPv
zlCWXKlk_Xdv^el)eE{PSV&;^M$tF`P7D-l(c2vG28a0SUn~foY*PbUX#C#k3LMd!Y
z0z9H6Q%vF6f^(HpeC4MkE74$oQ@9QA4roRLEa-*%xO;-;CR~-X-_VCml@C$wX<TX{
zSAe^XxKwM1Jaz|{4(|nnZlD*8Xx2wzw0r)s*RUJ5@cB-?)RbYga5L((Smp#x1%X1o
zGLITw)RZ~-Fp9h$i^{1(VzdT>U>+@qk;<L&YYu&dhHF&&@^j0mHI$Tp5#md~iJw>T
zLt`}ngH#-Nw)H%xIHVVvy}m=eOcLHu5>K1nr_mHParIF?E9>)d1+K<YN7hx~o2BZ#
zvD%K;wri?i*o%f3s+u`dP2Kpn8oJfkAC;b)R(cDpPA-N-r$eXqoldf=egj*9vA~8_
zEw^1>jU>rD(;`{J$K3sY*ywz)8o!x%<J0ZjqnxDu)=wa#hr+P0?Vr_Eu@b<YREu-j
z(WtbCx+>OMkpQR<)DQL-vTPyQ=9lSOV)#lRc;s(H%|wtT%wWP?jr+C}eCmmA1t09?
z^k+F>uNl%nsDZ*80)|MDQq)bl0iQgfPX<Mqh0?y~RQiIgk2uwTj9uQp*FbG`?GcuF
zp?)ttNLx{JdBxmgJl0Qx{ThHugA|~hIVG2#;p{7coo&Y9HKe26p;whOQuR)qrn&2-
za7`G!o$xb7(wy~_dVEfxMXmiS&Z*<Heoo~K4(VM1WOiXemFs<sPbYMq7>ZS0)NCEi
zQbFA@%{v~!D28u;Mr_-~If8%pA@*YC-8(;>gbgq8vyRi;YftDewxWbQt{wT&v9pB2
zJ3xpWI6UbrS+}}m8j5aV%!n<c+0M6A*JDn74qryn{WD!=U_vUG_5A*+D^ps=FF5s%
zQ@weKbF?hc^Ldt_l@}(B`>U2vZhksTC?kg~!Rn)yP-rfHETN8)C0ME`Tx8x{dAOCx
zQRi=-cALTBeJ@AbZHXoAerC8Akrg}JWY9mU`(kX%IRRhON~DoJk|hg}N(9dV{2b;g
zd8-qKMyT5lI7yu)Z?v5;G?va7&f0S&<cL7uN}s=dvLyMz=Mm7yRd6JJ<w)wDx{%Q5
zhcVsG#SrIzLFM-}!rPi0nAH3NX_`UTOp6<0-5)m#ocGWynqH;`BF974F^MQ=SWM=-
zFy|Rq6vLWJm*#R~5Ugca7i6gvxFOX}ODhG+TqeXJow94losg;Xg{ON##OxTA;Ad7K
zrkN0mdi?!fTcBs4XTC`HBf}gkwH>pjwntGX&g6H0ys1@}#a&wOION%$xfOAtPotyo
zBWqZ7%CjSDkx=DHuBd&9j)b+Cp&*wB7>mjA0Ap1yU5=pZy3lAa)SX9^x#<?;S6wQ%
z&r&;=7Inc*HLtf9ro2F*uB*W1(%st$NnT%fE~>lkfaXC2&*5VZf(&OdC!X)Qd4<tt
zPAs2)wE!^RJz(YNDMTy-2MaafG0Nt@I(EA;cXp6-JQf-(pbq<r(7+k(hK6V2DkGs<
za)SUlR9L5nP*F#YgbJK_FjQElcm{4<N_MW_sO7YWADmeY7et4QoqgLe_Db=Hu`@}-
zb~ZIunCSSW2zLa1+7WoL8C!EhS^9(mg#t%^JnW+TPPkoUh5yweORFfnc|};@j1Iaz
zq6du^qq>%5Wu3$Edc{<R!Z~2!tN?H^$W?$v7^8&_N25iZ`KU3Cv>Zc-cD~JF+$K!J
z^&3Z&Q(w;M+V~%APLB<I8reb7Iw)3aFj@9v)@qcHUii?rqFbszg22`1!mFmv*9b0u
z`Ccv#!ImwnrYUwh1275ff_D*%TEo>eIZWR=?mRd3my{d6n4jlL`USq&F=}bKT>I+z
zVbkV&|9JzG(0kAD|E+(y{-kDyf$?^&yN3$>lc&rLS%PT=;eIGr!3pJGQT_Ft9vaXy
z@tuZ!+BkAA)|C8cP3iCrL|J>NZ<f}7+7)$uOYZ4Na|qY$Syc*qJ+(?_fG_P1v;NBO
zCb#S`xm7#7`kY!edS=QSp=IiBy-!-K4Z^xt6eT^U#}<34`&;10Swpr7gc`SBv4?Xb
z_Hh0v_9$zQ7;>e#Kn#u>GEy5Fz$nd10~qITbQ=qns29Gh;2-2@yQN&Jw=hb7;k9A*
ztELJDXH()2D!eu{t0aPbiJ^C#9vk|pu3a_s6ZqO%5u6_dD$zQcF<O_AL(w|g2jzGe
z%5f9BCtvDXOZ<*@?r`=-=ME-rbnbB7BXm(DI#j5pG{fgr6PH1RI7j1^9v?45*UN4W
zeyNYAcKdi?(Z`FY`*>STg8qMhZMJfm`foXI(n7wZYt}KU`KJAk80H&-;P`9@xp4RV
zM>ur@*)y%K8TpfckvBc*;VboRaG69-d>FvTwD<VdOe<aIN3rPfueiAMraa|W9_)&7
zyk@NY=|SxqGgf-KF}7=QGF8X<C;sf?ik9$U^L5}9rI2AxM(3}2ycI@7zrrdzMtu<&
zf64;sX^5V|a|*))j|Fbol?weo^u@WaIfG&3V0cHS2GKqQjh)sO+U=tPvZaV+zkuF9
zzux{0XUf;qmr=Me69F)np$<0`m%RZ%D3{tAF`Iun{Pp_#6DMKwwMYn>IjbgP6+@aN
z;fW7VF{z!C6K4uP4m3=^@Xmbn<vZ%aiA7x~&6KkPrzm?|7wJHS?-yDZr>pIxS=F|e
z%xX7s5MK|^$_1TBmA@&}oW_qr9cz?8p+*&21i#iXOTWWhluEli9;gJHgqaFuRY#f;
zD58J7+9qn6=b751(KrjzN4ZQZOB6>8N#N2`$`x*ra+&WA!UW1EqIStL9HjixP`5rx
zS|rIUJP}1sym+P@)|f`BR7f>Hk35u*_Ly1LNX8Klr1wZiCOt+P0iD9;m57Y>tgW!p
zG;*(XD&3$Bk+tFCerl04+Xc!p>EIE&pyPj(b~GU$9C?Iu8Vft*j$I*x?=druVOK!(
zEYRv-$t!W4W^Um+QQyKfaU)#QFTY^V#H*7tVzIHVCfbqOum@=N3a*KI!gaeYNxRAs
z?=wjz(*59@mDjBX&xG>2F0;H-5N{w|*(r&4+#5RY7Lj*L)mV$}eh(E)08uG>Em?o3
zJar`Mc6s;NZtaKXs5jr}YJp~F1xxa3l0&r5w7XouiI9~rRD1(*i&!yfozYsC)<bOD
zr_)i@n5*8TMOe|eE_U7Y5t1`4{Udxt$0^y^rEByHC@-Av^9@`wDdfm0`i5u?q!TgF
zBjhUoQ0=tlABwME^$!bHtZ|`?WLtmQb|miQ^ox^yd6ib54<c^TJ!5mt>_g%~%P56{
zklcMuCYPoO3I6UxNeaWj535U7w`z<CR$puJ`0S>|6Rj<aM;gWAY5r0~Nb^*lv+EDi
zeBF2{+jzA^Bd=O@&r}{fd);~FM3P?W9a_VZ_tzenMQ~;BZ(5&hbN+K)&yRoFh=B5Y
zu@7Wz9rIQ{0LeY4lHW&f#or~rA2_t$YL-bp)Ke{3N|dZTML}NTVdtZIk5(jfjlZ*F
zhTO@*M>!(b8uu?WkL1XCz#eLhV#1n7pr+%L5M!xIb_qRqA;%&YJ#q=ikx~!Hk<wg=
zoHTO_IYf)HqDnnyP?=ZxHurxZ`H76AN;T&(H9xDhj?F8}z9jw1`D-Hy;jqd@dI)-5
zvzk(pc`rtNE?X#!6SZaQC>sVxv=zN(^mks*l=Z2m2K3jVT6qTc5xtI{Q>^l0g%{9$
zO!|ubA59_uNIe}#6`{E@i!;oFAGt-p$j#^%`PCU5-$OUi+VYE$Mu~stalSt0?7r~G
zkzEbBKJTgx-l1MJQC`&jMbA*DKz(as_ObATx?3XtkA2M;*IcE`t3M7t)UMWd8qnX!
z+6mdHv6k!0Jp@b1jC@31+H*?L+%>fbbs%F$QzdjqFF^M)vO(XJm=0V&PlUg<qU=3#
zQ_qXx?#!1j>;|f+n=XHbtLx1&hpN4iF6SMDZpuZj${0iMTB=82tm9NV+Og14HKIsM
z9+USSV|lsFOZMD8SFO8{LGr$f+SoUDo0@(QAxPCsBKAqQEv>K~;SaHN&uQP%vUxvJ
zx}i69JqM1T)su8FF^vo{D%$nj;J~&Bt$nt29CZB>xH;x1th#?N<$454I!+ZOc8Dvk
ze|sixs|5xwpQwu5s|@=HL;js8LEA?t{nKNUUtRVsZ2+^EM^t@1r}e7uZP`Szbk8ML
zea??npYzvNeR1Yy)t6{(Sej_$+N<ymmZo3p(Hm#Du_8>*>{W!$wH0BJp;Ux4LUGU+
zPOY9izos@z@<e}>Qm0=OIhHbJr3F*AydmGJ>KHNuhbRMWoWZN{TqX=I-r2&T1g4Fj
zl=MvSt|M3SIZYOH1Li#6dgHG+upmQ5wrhJq%{$SI5u1EjTz7hdwN-y-F5on)ZTrwI
z*7*ei1clwfcx_9exdy(+i)m6Qp<@^LTz5pVn>Rsu;qrfd`1S4e>A$DZD1hax({}p%
z=To?$gl5M22}=Vn)b#70>BrL_Bh}%Q18<LLL{)NQwC*8bnAUDW$S?u$c7vhPWAm66
ztN2v^L~wQ^FOA#v2v~Hxg9PGhbXALv+B@bP%w$HHIHre&rx&j`OCnLl{suXuR7k-u
z%5h#Y`ksHo&Xz|Tqu(gFRwbYvn}%rPbhYkUp9OqN4f8$}>n!K}K}`zJyq-&cYzoDp
zV1xY9S|vn<K$j16=qpQROmM--k!wsSOW;cSOO7}d5ncOYZAAcvp=={<AqvjdqylA`
z_j-J?9^TmIYtX&S*A)3~`zPE-%nCDq^20(0aFl-)-ZOMai+)e&&J%T5Pwxi%S7MYb
zxpFvXFLX$Q0axtDj<;~$j2%7WCU{7H(9m`R&3cV;Yb?P;pLw+FV`7N}iTH)LTQJ2E
z?cW?r1_`c+CBqY&Ud!i3D2;WEs$m`X#DO525oH%++OO2YK@-;<G>I>4lYck;ft_?9
ztcicwqCZ&BB+E?#Er&*5(~9el!7RU~okgkgf_L%I+Kn>H-Te+$GThOlF16qBqlQ^C
zK1923W+o>v1Xbv)RxaZiz~+obQ-zQssJ5yn2gjNxcR7huIe(RuwkB(M$0SVqMn<`%
zeqVCr@0J`0?HTtl@-OENOsaJCf~yGQ3xI#QDmef?a^#oyOL#zGke4He_g;fD$mJuK
z;RVKq`7%uPG#_9Z%-PWIe&o8q;2_=_A!nN$(A8Gb2dq(PjIZ-eM_co^3x>)q@2?DI
zDz0-Yr+VBWA9^MS@usu@<pQb3tiTFpLF{A&jECqO|7Z%z-N*}N3NhDP%@hRYCE$Nc
zuBRvqD^1(b6dxPjE4c^OU;uukxbXVm=9~iEd&n3E%95|-6Db23X2VR5l;iO5FACpC
zrhS#}>XQRg^%eLP`ov7%B6)$XIWUS`yOEbGbl!87P}}dzIns*e6f<(RhL{O1X|GiJ
z`gD7x50(iFBQJ58+4Ux>3M*hRGpa`_)e{1iy9nqBjh|`posoec*4n@0M(YW_ImbUW
zIWqe5`X^PT?MS;PX;l`%TDJNr00OfuXE}XC)epwJrous2`9E~`us)YjxG@s~FgBMF
z@h=mXQzS9Ge;2<4v;Xt+x1G$)N}-)ZH%NinIKy}2r5b+m+xa)&zp-=<W~EaowDE(n
zB$%Hfoo4PdbIRnTW)6w}CO6^Y%h}5{eHn5>>q3x4l0=!<Dg932_j4eeR1!CYyt&P5
zDXkCB!#c_;)Ubzag}BUR@-{C|O9H=(k{3Qv3eZ&TfBKQ6W|Ky{0zArmWb~5%jm*zG
zrLrKTqmKE*IXDm4pl`e9=V9YO5Yr}*h)Us2Fk-lU>;C%sx%+q5JBRiph*kC1&s}&y
zpoOu1P)c~9;n&~8kKOM*)#9t&JqB7dh%_d+N9&4)9mWl;6~XcHe9wu7xIy_+5OY?4
z#Q2&#f8n^Q2GJzI_q+r12p^NMsUH|4yvF_um8H9oB>%mxjXCqya9x16Vq@mVjsSpu
zZ-XCQ*=sSlba<$HszIs<C6uWY>OtvO?5xdtM4%SlUF=BPy(33XT%ihxQ*n{=NQVyP
zZC<;k2}BZ>AI)hZNM%>0LI%;`CEeFKkS}aVe_o)+(U<m7qJTMM8+i|{taQw8OqFWg
z$BJgM9XVJswWO;`#pT^>mbCS1m9C_9SNyDUKzYr`@fb(9f)^nQ2dW1zkR*(BSI|qy
z5b8SokQo6cPrheFi>YuC(X_I{YF~ZQH`+1t&{Ly!0V@p&N4BRtO<p&H7&wS2v7-tw
ze~Vz?j9|U?!cpUJ8^T&1r>;zFHOnl35v~S{EbdZ*@pjRQ!dklUu?Wv#G1O4g^`Sl^
zH-bi$$t0CedE^J@dD*09UR0HC3#8v5skOoY^t-IdnhiPq;5m6W=G^T?&!WzVOAzNO
zv_;~SJR>EC%P({Akwf8Ct?%S%p1$Q`f5zxpo2Z6kf44ol9h6+NMYs6&pqdSuU<=mL
zo001IeuwNnim+#W0X$H8LEc^H{ec1@Nd`0Cv1D_j7{kqylDCu&;n}>hTYh3Fxe@Gj
z_9%7B4ik?|Jw<iOiSSQ0Ysd)-du1(|mEBib+piw0onVY=kS*zTF}D^YYVL$+f2^zY
zxKn9fx8~Fg`8~yJQ1E~Z1)_*ox|(Kcim_eWe@|Uc(!@@&(AHTmTzQeSo!f^BjHGKT
zLs!;~df2~3P@k{S%j!dR;Yeq-LRQGNEmy4_r?G}nX*dFKUwAfs-~F(;J-8C1Q)gc-
z?}^)Zqb@|Rv0sWqH;NyoNyg)Qf0`i0HzlZVp9&Q|G&>?1)$`Ff!Cbq6oRdMhOM~s-
z6zR5%H`(dr=-iPutrox4L8KC3Out_&qg8Uqv|sDAicD#JPJqW+vF#PKA!f>I-86gA
zl@3L_sm00kXuh3FZ!h=X(9{;*w+BqHl4H17)c>OaGiDgSVFIv@Kj_$ae-_&$o3X8Q
zjkMIEPH`x>Pknj=x_}bIopZ;>HDx^@x$+!v#y(H-w&T8DF4XS}3}GV)HGmQ(8mB#w
zYSQ@&Cd(K)FXf*&Sw{2%Z)jACYW>w=P2E9pmL>DhOn7eBg-~IKv77~_yS{3es~p$q
z+Wt?&T;)M%#QSW;*VmNif8;rxmc)Ci^-E!G<O5SrR)o-_b1Gt$Yx_QpFI`Lb@-|jc
zr}9@fbt!9_0zRjnG;@Qdy*U>;rl8t!OgWj3nWn8t(kzIf?WSat$_E_}qp+X-B^ol*
zsL~Z)Q{c85h{&Yttj{!))a_bwSu4jzQ>BPHwRl&xWG8$qX=lD<f8kuqUBh&Y5O$nQ
zMB6B9rlKm={F`Q^RoAmTC@0hy5n~nCriJr}ly&f_v~wynXnZ-;m~i>x=^7?;Tc$0{
zu`ttgZ#iqY5@%r}Zqlavy)fA*U?n*7#AE{um+1Xg01L;samO<9I!ro|Ey)KA<Eg|)
zlFueC7Az)xJm#KqfAOkh1Tvghm_8X1mkg+$o{Rt)?WdK<48%7xo4Zw<GuQ3YLWLm@
z_pGj+BN7Nww)^0*W5o1d33<p7lu>i+4Ra&pqQ+jABgA$YnJArTPM)mHPcY1YxVY$K
zcPX7vQvly{#F7V0!Fihh`*_i`zT4&cw7&ZQpEr$y7=-F2f72)j!kF=pdD7xsmN0qH
zbi=W;4ZzWF^GGkpx&W<uf`|6N(#{n?sbM>n<I!})S&N6u(R_38xi+uR{W10G(KEv+
zu?Kg|cVJcLnsbA%<5lg!`}HXoN0OzUj~7+}+;QS~;<KR$=<l4X{sUAzvRw*gZe(+G
za%Ev{3T19&mtVLs1(y*fC={2^PXZ~I@NYQ^m(D9Oj{)qrUMw*p1b^Bs{g17c(KS^m
zuGiDr?3tsp%lWumA=~dWlH0awE_2U0cO#9}6(lNR+@DC<??`?Gmr8F|kYWpz-dRAB
zux*yFE`f=2O#@AwbZJ#9?PQQCERQvW4PKS`)TG$I<yzzjn{&!wFUKed=GME&?VojK
z>a^1{0*u$ZCI_7yf`2>Grqz{V4H6kn)l|<fLPy@_N&Ln^i_u1D!yD8K3>)<@vr(Vr
z9i!&CSU2qtWP0O39PLG<(JB8eNt55)o-`y$AnIlN-<vcEo$uebs)1s(aH`!Hsizut
z#jO3AM+l+%bO*`+xAV&Oa@;c5y}Nc{v()RyX+JsfS?!tX{C@&RYNQQ~fF<nyT~0uF
z{M6$c484$d4?Mp?1mRt~uK#e0x#<)JCF|nx<|)j*QY-|N^zq*XtrVYilmdopg8E%&
z-K?rGvW<>xAziKcn8m~$2M@q)AyMt;RnsKZdQ*}s92v*|bc`!ievitp`M$Znw{dwC
zH_g3iWm<(q?teTa=w73CANK~lQS?|tfh#Qzuo59IS-_-&hjJoXs{hdG)HkbTx}D!*
zFU%nPI<{LK9oKg?Jq&NUk1^B|OqyQ2y{}!pIf8V;Y3Fnc8v(fuc5deZ3Mela2AauJ
z2cR{5Ak7#*z11`@@k!Iv#Z$I%dmF<J<c_d2APC&7!hg;BYCZIv<pxMSQuBNal%N5%
za~M{V{BcUFRQM)OSBf4Fb73^So;OcS=k0I{o(QZ^FbyL)0wXgzCM!+rg3m{AmHyWu
z>&PWH!)fMupqc2;(41R10QX~BsXgbIA;r<F!ZjbQF$S&7*CSXh=b&kZd>NK&@gOek
z&GV=)Yk%ZR<yW9%Y`I8fY~S&TNp0-YTV;vDcO0Xo3#+m$ZJeZCyOLw3UGNpn^PZv3
z@R}*M*54D)HKHhNV3~=h9<!j+5(Pj{ykKfF`^>rDQuyoe2QxnjD|x;#OAF9;B88a`
zS;?KR9a3vgouQe!krJ!?Eg>xfHrDzu_KAJE;}1rR^o};Yzj8PEKcb4cIhT<IFB1bf
zHZzyuOE4}0IhRrID<^-|T3v5j#}R$^uUH^2>4TO#AG<qS1O}YAPE$CDVW&k~!6+n(
zNt8OGN>PUU?|Wv+OGzdb)6z|?JV;z#?#Jxx%$YN@x3*kE3PM(ba;Pm^Fbzd=4#dSD
zv_Qpzw$=hu!jqB$XYX4ma8vIiK_vT=7PQi%+G-BDR{tqsGVy-~g<ATA99ru0XrLrM
zzW8F*d!YyAgs!bppvui#j8<XHLp6{k_4x%YEqInN*37&g=4+B5s-d4xfcY$NBA>Zs
zh_5YJ!s5)ii(3QhYBjWSm!)Dv&CS=qhOPNngN9Pqh7oj?eKq4C+<dHI_sYGW1}M&F
zwj88bZ<QcMeMEl^;`%@f`b;KR!=NxG@7Ex>sQ1I9s$F@l1}&+#FrsMhhY`j4GljC*
zcLc%JphW1#x8HZ(IAl`S6}FJM7W|5ZG8Xn$2|0_8g-&y+$y*jO=JHkxxrvWp9A7Fv
zf+-YgK7#!yguEr8Xpu)~!De!E2{?yRxsS#1H1+uy;--Ht0b_%b`YbqFg~?$>s3iAa
zF=?&+S1ewwj}?ryTCj$=a>!+owlc#ETk!+!vv_A!{DZ|p;r;{8*Vy{Z=pxx=hOJsQ
zmlXDA)%r8ET3-W)*O+{&TtXA|vG5SMMW2rXEWm$-IpU0cY!S3te}=IQZcqfL(A=c(
z4ECH{pGALg)Mys+RSZq)s)nNwH0tY^H^n7&jicH63_h&%r;wErHsr%nAaebqiKlQ?
zm(M+gq`6q`Ae5;*`r?bp(?fWj!p&b!{{D}DVvZW!T40~ijpOX{&70ScAOBj<O0g;J
z(KF`~3tV!Kp2{)f`G<RALs9ZRa!RdH(mryMT6BL!?a@;-%5^cr<g59c`Qp|2(R3EX
z`-jPkqvdk8I19Y``lolxuU{>Xma{;s2hZkbODyqB9P(_n$TN66?AKOb{A^ie^~LRj
z;2`+&`0?b$Vm^H}TOLkcJbgC#e)ev;7Wn)Bp3kuIk245#ww#?UFC4y7pW@~0Vt%=p
z&Mtpe80e3_ot>N>eK~)3=sgNKfb(mNUtpxW*lOUDv$OeXqP$xE(P0O1-$n=Vb%WQ9
zqC+DOjpB{h8wvZ2lUJ9=OaJ!#^z7eZ^5uMSGFx;>((B2eC*MrII;7Q?m$T{ekP62F
zhjGIre8NE@4gw}iF)6*CJPDZo)#U5>_w#?r)9^Vor@$5{4Z#C00=NFd<Hy(3bQd9n
zb5wZyAcR7QQuD1s!1Lhg1mq}OYZE5s`%9s)eg;ksQUJhXPUUtfu(oDX;5QSbFg4jB
z#qNb+k%5LYU^pmXL`#VkSVfs*1Ndl_$Vp#6y0MYTlg;$g<>~wk68-h%H~#nePs@Mh
z`Nbb6lOIo)KV2Tj>HL?;&zHZPADqnIPX2QF?Th&Ho5}H;`SGO8rdFsPt0X|=F*6-a
zr!#Kt<YbzT4KW)KKBY9dSWG9=`Ps#CadZl^#{{1}r>x}w|BGR_kM6!*A&HvGX1hWy
zwp2g$b}vr)E<slg@UPx)k5p2Gn{0pfNX1dRJ&-+;?~$(W(V@At4KzEy;VtaG)?vVI
z%@PfPe2<>eBIv^&J+TqQE!(3fm`C^25BJ1oBk$Ew-1pjV5shvOKsVyl$3~-8wxUtf
zkB>&3vu~nNsq08nHw_yaZQZaTV%ZR}Y=~GULt_YCHiRxKLt~iPhM8@cxeb3aHyR_(
zX{>bx?oP~cw$bVFZQw4_w1M<X0y{>6C0y=>?M}kz3fePR5wfg@f&+H&Tyz)GeQXF#
zu3h9vRHL7<YVLGaxk3wdJA^{DBo*0&un8kk$Rp4RBx4czr(}c5C9sspNb96PDCHe-
z+_ImrK}sYiS&$=*Cxnx4I46Gy!Yb&&0LZpAfy~g5y;i`Z0?xp$+ITX6P8m)CIeV-i
z-U5X{;oFKJCZ{qc2wRH84xDp5At(V@;dn3&ahVb*-VTcps~M)Fa8RVGpqI%;lY!y<
zj2sqxF&3ng?9NX+6p^`r4T9fQ1=ZD7WIP!ZtfxT+As7UCF$6%gEdzh}FXIpJ19Qk?
zBWi(i#H?63LmSmgG%(Q}S}3*T0?5yRV$+(CLXRc{nMKc{hohi|xNF-u45w7&<DxmK
zHG+GLcNA1PdZmlIe;`B~e3Q$zBH)+g6^q)`5I`CdkMV@SM*(+%$47Odh^Mqe1_)bl
z4Dgu*-$2-^X*?MS3IKmdG2j~b=k-V=B4rm`mn1|P$#3$&2zVV$2y4-6qicL<7Qtz@
z%>k}yUIs0QZ-%})o=FzOu39!I=~!Xg5>zU{aTZ`?6LGx0ItS;HEnnl3P^(vBw<+|E
zUh6UqgOL>3rzei<=9~|in_80m9yjeZ#5-_Py~a)3z%tuvzrBCvx^9%AQHO>Njka!Z
z8X7b-cxVh0vJQ=5W*%ndVP+m?=3!<TW|lE%>)JHeC5gc{izT>6+vT8!z{w3kD*;us
zj0Y0MKO}Duy8x*G7+W$Ro2uLpyf_225y2CU=%O_O`oP{-*;xUv)NM7ChQz%fX0$?}
zmtL(JPlUt>R6Kv+RB>&z9d)t-rl(5CkJK_l*zryRya;isZmc1Dd0BRGspiol=awOf
zHSo*v^pF9U;uQ@bHAoVKf{BXlFi9=|YD%z3n2Q19$Fqm5Nc`b>pfS;}QM@RQ%xy~u
zQ0i8R+#CR$<~kWq25QJG^I#AT<1~F^a3;~(ZIX$TiEZ1qZQHi3Hxt{oZQJI=wkEc9
zbI!S6RsZPfU%R`jR_}h6)?B#rWp<sE4C$BacmN)w9-#P}6l_5?(O8h<HMX542m?@u
z#UK@l)4%>%Fj+{)>x8iKSXk$P*oixcTtVA_76-59DKX_3TTEA)SlcGvbY+jDfw0Gv
znV^NpzGN?X90QGk)%M=O+#Ewb^<1LM<pPoICiPu}5O?ZdVo3ZG&6B;<(80nk?Yp_0
zy91P7&!p}U6wvzt4X>hLZ+r2fmaiZQu%nfZrcSPPNNrZ&@SH5(uf8v%?lvMyYS1J8
zlKlmOlNR^V?VysZJ2S#z09FZMNCyV#LVpR3Gev(|E44JH_nMem@$KyCa#y?7c|Tkd
zv2ZPTRcSjhHjT9pLy*I3if)ema)ltz{TI;8DqI6=1F<K-05Y<(ligRR4$cc{sHfDS
z3n`=osv{UR82$@1ZZ-<24p!w}Xqf;ElB)pFqYHgws*0?Z22oH`f+!`a6esSV4=?0V
z-`zXzScwqL1f=$stuj}!vj4R(_=PrDtNLQ^vbuOlIa=WT2r4}P9C^X*-GRN*F$>6I
zs7>?BoZQ2voQ=-h=hlbAZhKanNW7$@oUH)8rj!RX8zpNV5(-laha|8^6RM62SIp~k
zd-4BX!+(<kJJdz3kp|oMdmDJ%C>%bWyUld1vP^)>Y0bKFQ%~h2zNM?vCPAIaVIi!)
zC2AQmaR>6;(eI+Q>jBFXNyo{~r30ApPdS_|tQ|J?{@&os^WH9h!*v57!)WjAK6)U1
z`Pu>_Xvas{8_`0dH1g0y{_+-YA}wTGjXv<I!NS0S)9NE<g02qh#4K&gmhwL<J-m)C
z3JS`SPVu65J8(B_ZiyaAhmQL9hcb)$q5D1IjJOXLPfnes#$ZPUe_3n|g9R|N{>c6M
zRk8(98FBrn)vIuPI`&>0etA_srY~5>w~5>!Zs+}oBCNFf^zZZ$;Q3A3yu~ZKhx+<0
z3C73vw$gz5d5)KQ%LdsH{*_W&0KlY*HZ5aWSLmn`9<R5{fmO+$=q><+L|XK?*+;{6
zn)({h6|U(oSw6#je<6kU>H)Zd0Cr#$SpfC=#L9pw&kuQCVF4r@>vvyU!TcvN^?t~4
zICejSxv>hb7kkg3o9kLof?CrE@HS;oeI)>J_<qb8*AZEfAAQ!4vre_~BI0LpRZP$q
zcn&xMOxB+rn1T0=76kOhnumWx+mW6aVX->hxF_xd^LXp;F~R?KJ)!no-ER4<;Z|S&
zk3X)T<N9$IFs1e{5Rj!d@-wCZ@3hJYFwy`6Qbl1)@}fMnnxhKRC1@2}@FQRgKyOzZ
zn3!x7x9fD=kesIlUe!f`psPsw0CGSEzm0(Gu$=^DWfZRM9XuN3JkmxAnXMcA@(W&A
zRZ0tZ=BF|mCROgu)Gb9UDHofw&rRCn3(KuU?>vZaS<}Dv&!HN{4@79)>alebPQQu(
zmf3^Q-3F#;8uhe`Cj4)8-X^xzs{t+d$p&o#`BTw?l-0|)AuT?GQmiU?E!7*Td1}Ja
zzoT`5ii#r91WZ9A^o)RL8|X+II8pSG)G}FSf1EVk!y?f=MrlQq-nNGB{K_<bZ!QS#
z9Hzq~)DN^A!$g(f_hDvtZ-Z)9JlBr^iTa7Yxy$}QCXbQJ<P_3Z_;Z^zas8fTg}8!i
zGYW#b38%;tl`JM(t6cU728??#2g6DkmSf`xC^8x<nIsq$l%py@!i%`R$VWR`Aj$#6
zL|N>iRb!wDtN4LUV_YFPCu2w}(SoPM?4BhECHeN3BM<2_7a_b^z!-%i@7o>VlIQrB
zI*aqJjApw786VpUm{XFDMIM^sKI;Psi{|g*9!7igb5?M52eU&6mR)CHQ4rC8Xej#j
zc@%V%yRj&qsDepaWEh=wAX}u$k{tKBI+2h1UY)gUq?QAmIi8b+=DbDz<*^IK%adxR
zv-YMo3w(EL4Ase=^>`Wv9{qfPqJTU9L`RBiH}EmF#M$y9`%*hnsXA?i6Vx;J8)sxz
zT)vIj*mf~~`_jaO#l)QUf6hBJMoU`Fh{twCZ-{{F15)&VW;o}m$Q*3FSPoO7GvasA
zNsbX=;?Ntz7q)z1dckVCB7<CPJEj~}C8<;LVnr>t5=ZJ%;^UuEwkVZ=ah2OP9aXdD
zY3+Ole8PK9WADm%`>iuxS-nv~=TRlYyf7C}OM@d0DLX;SLKX)yJde>e^&&b4Ck{{y
z%B{GLK5f*(Re?D9WsZXjYBr><tdr@0gl;}#Xj0@9?zJ$%TeCzos&G5!(Fz$U53nJ3
zmaTXqecO&-#mpt#Q$76v7YM03MdXxq@3iXLvi->E7@xs$EtjF!88^ddA(z~&VTc;{
ziTz(o_nIFO4%&@RW9P2AOk`k@mo58yAT4OgzomB{N;>n5MbXfqk252HOjIo7Re({o
zKGp8KKyU7>DJ0HcJ|@+V4B+|>xXxBwaW+FagT4#V)<^|LoJ>9eQU_npOOSdhBko<b
zA^uKS_s(-fvxYrLRo$!u^UnA8I0H31Xm_qkXM1ZOxW@cfaX^EytV4bAO!uxI>&?%$
zAZCVWOl}}{I<^yQe4tpM7k0Cbr2C}AG$U@8qFfmCBe$B>ECfhcdtW1Qj{GBU1##gC
z2_naW0v@_qhbO22kgQ_I;8Ps_Ev7|l4K2Nj487PIkM!yeIqiw^e<3{qX-zbBy!*Fb
z9W0wq=isWFU2Ka|D+Y58o$4z<LNxUk6zB9ocRGcHptnL8U{>s6N8shU5xFtt{SuZT
zxyJXI+(P6HtAc2Zm*T=a51FvVUy_$ZytC+rP`6CXtUmXEDLNL?NzSB(zpk<eL&a-w
zN7k)fe1Z|tT622G`szsT;gD3a+-%X$x||jp>piS~G(Y0Ki5*?`lHT=6?xS_lGlnwW
z5~0HLF@L?wj<M6e_4Y0<6bd~zF#sm_#Upq_cz<Ew@$}cgRWnXI>^#_T-%Co;JwAsT
zlg)o8f2nG1jBvn=seg<xkyGQv82f2(ggQS@jwZ#->S7C1PvR0p`>oZBqJYKpG|5gy
znS_Yf$R8ZplivD^Ja-FP{lpo|K|zDp9FqTsw2Y~HBt*y{jQ_vj!&`<i4e)}^gZt(A
z*^!8OF*rWSS~QZdTTeZ2FicwHk&xhEGDp7Xh(^{ep1qzQsfI5Hr&euA3;Wy8Gerr4
zf9w$5|2IOGIXEd4eYxO`3p5^RCg;43bR#bAml=-8>zbHZfqpQ1fAikcJddyE)@N^}
zvKVR8Z8k9ZnDO)VtnQ0IAK<C_8X*|s%<psMXZ*hulhJN{t>`h=KYznR1}8?`<d!Xn
zlu~UcT#7O@xbp4y`7eWgM=#{ni!!Mg4VWj?nrJ*f6nUn+MI3>Pk#1V!CAWw#9$~!h
zl@qymL}ZPmL_m+|9$^`AcyGp0U43joYX;QnBxq?6Iy{4-f^t`D7!Vh(c7F!vFuqdy
z_liR(*ajp}_|-|O^6=l<hCTh{`J}dCUy}y^-(>Wx&~2Du4Qv5K*u#EPU6sfMDijU3
zKa82SN-#m*NG7t@(Mz4`hY{Sd>8mef9FZUar56&}*%{han%XffsT48FcOa$_V<9Pi
zG_evQr}*6=u1zuu4FLU}GaF*>bSrx?Sh6UmsTZa-E*k(!@Wmu-#qgR<bHol}M3HOG
zXWy~jFJ7|RP(;(d64~3NX?6Zo{YR4<Y5q%0%}h%n?rp)!Iv+RU;_={o2#-)+$wRd}
z4YK&?$^N<vY3l59DiqoL6t(f_$5ZM2u-a0`&VOdG@>Ix%vj7Cd66rAGR{4<?bM=!U
zvD{-cb33h2wRg|u`XkELoM_e*{ri?TCtcWl)Tu6&U`}K!P0%@H^HVZnpD@qx>)ivx
z{N$}R26Ru?#Ls2ZTAaq=@KmWu!*rZHnsp9*R&<*>RcDZ#FsCm0Mx*S#&Re7fI=`pM
zbveD`$eco+CBSP&{HRMb{fdc~78P(!Ok11clV{_F^UPZ5Z?+ak%S2f~FXaL8X#ZCG
zQq;Asgpo-{(g!dja1AiC^cS4B$0i>z$^PmTR+~JCgw}1FmvKt1G^n!H3W261Xnb=4
zhp3%b(88?AQ8MhYCud+eREVefvbe!(S010}@`}OM1i(I3iD#6e+NyddZJ&{j5SV=@
zJ4G(Aie(8aI$aN%<Y1G0j+{B4G!0g8@R*)6>iCHiv#wbes=|+lPSvp8k-!&ENUhj<
zeKvNWG@X<ZA2lUO7#I>egj9^+#z}>%_&MG>nVOjMNc~ARUV^d*c?%_c+sIeaQpvwL
z`jLAP7l1+5SaFTwmUQ?qwGYmshia8&$l5ZZE)7QlJF^?ofpv6m$D@{=aJ@1+guRIN
zLW0fZ6GFfo+z~uVH;i5xT)cb}y$8nG99P(}`UHcUW!}atUCBs_pL-lr<@><`w}|W&
z0sP6af5YjCg1rnkxYjX|^kB}Gc}WaGh08v02_S-yqW=pY+%Aop7rY28eU|5J*fVSV
zaB5+Ar|>jwo}&Q$=+r&&NmqxsigOLJ8hljDnXrL&kv${Q7ddq1jXK=#kP&L<olo+I
z=KU)}BZNxE58Ul?RfGW2cH<xxj9Fmc=FM585&Rr?P}BHz!~>kkgzxPuK`#q0axTI4
z4<Ps^^sgEM;;zukBrV#2K308D??=)R+Z#A_ejWZ<Zn>A1cgCDO*{uTDS(tEhoeHvW
zy*8(Q<d*qyyBJA4Qiyxsh`v=mm#<z}Wt2s@5K<Fb!^(1<C0y0^dMvUg1~JqnSG?mH
zo&WO-_^Qx&nzJv)U9^;RJk*p3XHv5^2;gibn$$b{9~CqXOwA|lzhf{U6fQE|PHop|
zX`ved<&y)@eOW8AqaYJmxe?#MlyF%4){6f7(d4S%2e%Bt&3VS5AfXi&)Q81`!x!WO
z@Vrh<z2Lw9Y|ae64Ioj655zfb6K%(G^)eVj*S_+WbUPD$VE5;#`TQTTld9Rq2=jyR
zj5)x_sU=E`?0|p@kfO1K%OT=RTjCuxZ!J>Jqq^7cP+GpN=&hK`rfghGj7=BiM-SB-
zAYLErhntplJhfWy$Ux<szVy8tS0K;)BYq1@6D%BmBbdRx_nQ@n>fh39Qi9dh$@_<`
zsTs@kXk9+QB%<3ORr)wW7sNXniSJr7zEwV^Zc2fRnSfQg&p<O=T2RV6xR&!?ctDj4
z@OuaR8<LV+*^ryCR&{6;1L5(eUBR=jjXa*-Gaj85nNpNE9G{h};}n-OkV9cB<_Isv
zV9$Nanz&gX>eYCaK2a%24+S4gYS_^l=`yl8l|vD&n*G*7>EPLps<TxU5!|mpbv2Eb
zlXt2f0RRFl9qdRM5XZ+0laKD_nl`QD>3Nb$xWY73$~eMrY9h=6!C|ExJgRC1nPvcg
z&u`tylh%V(4vN1e&*1pR=|yVids#;QJ>oJNC^{q>hw=w>$T?x|M{9>{+;9<-;Yhz!
z1JI2_$2Z$Knq_jw<B(|X^bG5wdr-bJNKtiiaRFoF*6GEFbRkbDy&EwKn2v@-`AS`+
znc;kvpC4@e?m|Tg-c|5LcjSlZ2udOD{zw=UBd(bOje`}04q9iO%Ngmb5rVPVx{?b>
zs^@oI^okYrX#5l{j>;N^0|w#br&xxE#I=;jPo85M`+u%T(2>roQSa83tAgiYoJ9)c
zivR&fX~x6mkF3%5jYlh%G)=Vj(u>AhYRsp0-6q+jPc*C$=_{1vvpwxesjU7cD4A@-
z)M`}inr4%<Z^8|2oNbDLp~&$w3{=L?r(*o%r?(5G5pv}#sx=~7>kdtUmSX`k0dfxR
zIOYY!XeSUm=sS_i6m>2af`JE`VqrxTYryB7WPBU{;4>|VrO2^DDBdC19~QQG7pAHs
zwZ|r;{!o1s2@!?e?N9+X4M4AeoBtJxMAG{1{hgniN<zKZp<e_!l=bd}JL-}{b6E*}
zrpj=w4V|n^BP|uJPJs#$%UugKQ)cewzjs~oS*Yu99&-}%LJ1EG+q`3^#1k%%8o;(^
z09Z}j^97+};ut?v4-QszDOn9^a6vR#0|Q>R$f0s}t}PAfh9loym}JgOFj{ll+G_K(
zsHqi7HqY}0x*f@z>qq7ro0aK=?uJ8@hv<&26SO`2X3MI9s*Xm=MvI0U^pBuX*_LxP
zmG7y;H+(n`^wj7*ffI!pCIIeK2w)ts4}}d+oj4#_m)m$7XOo;N9r9RG=-$`Vd8MGU
zu4z%(*U=MNJjZfx`m<SaL+ot}X245=|I)P~<f@W(cxvnpAU?Voy3h6o8&nGrj-(T0
zo=fJ9G10od^_a<lw2aa2U1b;3&UHY!Y{9j7&R511&dTzwMcpiXNK<@j2ZTyR3#tuV
z2PhdwsTs&E7aia_QTlnc`W!i~SZL}`t2Q__kG5xODG@&4ks~SDS>h8A_c>V{n2$?z
zr8s%mkJ%-_qPw$gdUv^KW256k`K%m#yJ3JuMr^(-B*2|t^GE&Xm0<v>MY%i=FuR*d
z8*M!<r*<}Vd-P0!IGy@Xc#^r!Ou{!aq*iKBrCf*{S1{O?kk>#+Dcnr-(?Tz<Gz5G>
zFnKL){dY^5%CPwh<rgCpW2zyH069QI*8bo>nE>kl;Dh#HL-hp=`<~Q*z0L<(3rY(*
zOL<+szw?EiL?m5`R&T}G1TeHvIPW>^$Hb-FceuB@ZJtJ_o$br+sWfg@aeH0+=vlrY
zrlV)v@B}!2ZQY7Luh?r99yFW?X<-iz-(UANZ}DM5ZvLV{_k*ePy#uv`eFiXUj9{sc
zU|a=5)ciTuwK};~<j$rMp$k=M0pNeHS^&Rqig$-BdMep;5}}^kuoe#YG?B6lSz!1G
zk}~)4d3#^|s`9E=<HS0+6}845%8+dA8Mp^Qo};qDy?#SsxDg6BbxzS;f8cH>^n6MW
z!x2N(jWDUu%P8H6&HQFt!UtfP>R#Z0nsQ)#f2<avgI?^6!YPX8U7hr9wNB?&tL9f!
zJLep=S&iPOO~Yk3u(`lVa@N!u%TpUQKJLt3>n_|qnr4~%5B%F?_K`GA@gzpp!j!)Z
zOPU?RpgCdrw&Y6MHAm{=fe|#5ilc8&l%OWDTX4#^xni9;6Vi}CAq32X<WXc`O=Q6c
zx4`bM%Mhk^PY}sDP%f7)+VcwrUw|kJw1XPULOn>HucRIwp>#X#YE`+z9%1(nVmP<3
z`!iu69CXl?3|sz2@tA&G(w+mE?Wc-f<=D$YjhL25MMx78xkz$D;-k(_%289VAT4W8
zr6e8p_h~jQ9+vz^dI3m0<h;!#N>!HnSKOl|O|f}FDatL!fIrdJ5nV5g%rPA0+V6N-
zFp}V3+oLHx@P{^X!cceCZB>C3(bQ6>Ks3dj75>B;QEu^pVAdf8PjkAHMXyxcdK+CW
zav8eIzKdTVY3!HvSfNOGm-<K$_Fs>522QY5F8w9aiPw#&wk?2}*DpmPkeReuLaE=&
zvFOF%f4N1&+r&D?bDBtG6g{nQKe+}}Xw)OQPt}jeH=4CUC0l5OL&I%eGwNzfuVq?V
z<byBs)kU|9rf`F!n^eVilvbloqN1D)E7b+s(PofJNEYIBBvU5R8i%Giwl5zy_T7F5
zkp3Y?Wo3oLBX9tOSog5;OYCw)NC~$gW}}B^Z5CsJR#Ks>K$hU%R&Mn$VEMgr2G#Vx
z-3qM$d6oXnZ>8Fwekv=TWhxr$99g4qBfc$sh}z2zX4#fSq9j*}&pqoVJ1~JE&X3y!
zRkzxy-Z{JDd79UUms!F?z5@4bd3t2gkN#^?*j=bo@JIy+(Vig1Qdy66T8q^eI#2&G
zT#<l<(ibT5fhuTPyy}=01309)5Tl$7b-8MV5>sDb{|?NDl(|>{PbW2HC0*s0>}C`B
zBxMK*%rrYdw9qT6EsFiQrOiM!9e#+;)#2W6c^)N#TA`pu2wn4mGqtVr>uO;0-!8B6
z7i{mmFq0M@b$BxW+_YcTsNH}uwd0iL<z+cjN8zwmmMqy4k!;*5XjG=>e4JPiDAAxy
zm1^hG>PH6e)hu^Rs4Ij=amC3!`yfe$(qY6iHp}(o+{4GYMrolstO>u$OK~~{Ys}4}
zmzzZqd0f%1$Cl^O(6Z&w<?6`Qy+V%%+gn{upFJWYhH8}m^$VG0M(*u^_C&$<XJdRR
zJ_6(H&I^2tIvLo$3$(agHSqoG28~er|AvqLqTpmd&6j$OG>m{tEkS!6cK9CvgM6X8
zB*Y3Yq9~^TDU#iK5{3qKgl=N>sayF8ZEdu!S`mG_elL5xSbrSKS$93a;3L_a;!`9M
zRg}+4ife?a3DUu7Ow<`aGMQZ(P9sd(HL4(sD@2er5FZFZ6d&$GB^LgfgFK^{!lYd!
zusJxkW3{QMmjnLfGmt6k7DlSbxXY%Iut^eD`%SYsdaLWG{5gxmR3en|$2TIY=gg<C
z#EQ|OnT~;+R=rSXs#IpcDQd$4hM*{yyb&;<E;^K$4!~IKiqMX0D1>e$B;-QbSUuBd
zo9>TMm1XubU?ty6J(wCN2~Lt_E*LAejQOiXLxmWt^9x`M2OUSkNsB|ss!c(ZrE;n3
zpR@rf!DbpRBNsO9MH5h$<_juW&>a|<FmxeYN1AHjT$Z>#qSW%p_+Hi~GR-HGq!fu|
zCOu(B87VEAnWR(E?~ul%YI&Vktkwwq=WpLJM2KbGR5(&r%VM|&lWjY)VNuIr_>EHe
zTGlFMWi&v+k!o6+CcVTgH&`Uow?NLCB2BnqQjliffkBy#mVrXLF1eq`+}2TY?x>25
zjnqV+b;qXt*9H!4dxI!(@=`)_ay+GROiNyh>l|9dkQfD7fe|qZfUz0TW7HPzB7kTq
zWKB<EiH4z-WrJIK3TaBo|C~xqODCFoww|&r0~Fve9FT^hJ3Y<9p&`AHRZkso1U6)2
z#EjvMrnxm>wZYn?i)lLuZgR7ns_!=I9!V?ZP}`B~_cbi&8j`SfIybkyQ2Gm~sEgie
zEzL(%ELOnmslYgnvRvJ!5m5pWfA{WD!stnLWJ(UDd+1(b#g~^qgWfIXnOu34t<RPI
zITbK@YIJPz_%i)|)XF{id9*mZ6nXh-6VBOE&#!=7G~HkeHqEBsO`*0>RPO!O<lUYo
z?^ZIob#-)qdG+}4c{eq-0>G1JT&q=HWzd#0pG2nCr}nKaA-`W3Ht}kxuy{HPJ>A>*
ztUbPV4{V$cM$)x}KS0N@EOuqRS+ikd7X_#VZ#-ysRwHB{RU_m=J<62jqhtBm+iEBu
zcGMn64xScFV?K3UcU*VY-U59;8623jN3NeCvmt%GWK{<w)~cyX-6Xl%S8j0fYGe$c
z+ne{g&n};lPqyA`wk~|WoO!?BuVo(Hy&Oi?ygy#vO`rbBo!W8j&F<m$bK>|cd<T4=
z!qb1+i*NPMedH~if<7|iLx#pp057`$n9`N1LqiiC6evjL9Fc_G8`50J?Vn;x$uZu0
z(aT2U_3{?|V$!m-{f|$;h(ohuDnpN1sA^lGtM;uM?&)6wB8W73pjY&srv%^So-aQh
zyS%UXCZSaE-ONVH3Cnte4Tg4XF#!117I$j>ZyLz;D=K_$RXdmRX!ojB<#&nbIOn|x
z=NgKP4ke)x7=tSj)lml+p^>kCl*=6SaaRH+i+uzf2KMOIrt*p_`O9a{Q~2ET=ESt6
zl5vY1P3-Y4>*}vMm&%KLxD~JH3hK)rgeEaLjzD^`pb(%=unH06c*bO4w}4ey;<$PO
z`#XTvU8n7#C5zPK>^XnPXZIKOd8gnfxz-B6xVw4``;-^C2wjQZg8opCqFVtXCM47L
zK&1iVQ`i1LB{4PW$_Kj?g*U;~m#cXH*RXeYtUb0>{BEF~5bdh2!JRvvRn4I!n+(u1
zX09l{OI0+m2}2p*vI=y?EFe2Hqc8rFubT{H%qd@<bn)PG=xDOhu7#dX9+G|sn_syn
z8xT>|U)|~s&ypZWd>!ddg+3Pis9Mt{odcmWO${Eui<)i$%H(*PG5Sx2L~j5iJFD^}
zW0Zi^)4EtEmW0e6w*_%bqx;deMqCV-sf1BtQ`JiH`DF$>o!${I2Ee|@^Z791Sysxm
zsEm774)>@M>Qy1kyOiJPUEkdQB|B57MDKuy1LQvI38`M>yfQSj3Wook;n7+|PA0#c
zFf*(^Bv5x>4?{Q16#&dsX+kEdx2L9eaH%05&L^^6hezWTaaX4qyPl2nz28R!yf7uf
zku2R$mw;$wS|O;k1SqR!bbwhOT*lh0!$a6g!e>ioU!lG-^es~)aQxGLF9Ar{f`8|5
z@_jxaI+IFa3Idg$V-h2mv$V$-6C)75Ibtfr5MX+tM>2+GVw}Oj{L{3}p?h`$ON6*U
z^_;FyBrWUmuHyVlD8GRaW&R%W)HCRTN5~70kT*6iN4)4S4B)mt{BBjm)21NGqTs=$
zz<=-)B6PEKOSGLZ*Wl@?+4qxEw66JRW&x{lZDq$5pFC_7%J&*dbl;i_&+;NQN$_xY
z<tOU`f(uW0Y%gXM=<XM8Ea9QUs4>v%pO{gEJEhTs?fCG<Z*hkl8IvBUbP7vhvZfP+
zb!W&+&Y&k8AtP_N_@Bd={x5O52MMyVFfB!6riAnmBQ~B}5!3_aB-y-;dZov6d`Zx`
z*%|}XW^Oyr%}fx9u{1W>AMDb_2MN-Ua*-6q?r@kg>enFzrx%!spGEi>L3+l^e}Iwt
zn)&_rs|&)&`M-h*ZxhB&fWS4uN8|hZ!)OS(seWustz|{DNt4^$gGRBJ?WgxY;&P-!
zva~myO|T4t+eZsS)dVG($%t|6lL_NKVlp6iDgIf=Xfp#Da$ms1*dNJF3any@p1WQs
zbIOHIs3=gL^aVVnoi2BKeYs2kyJ;sy+PvLggLDFsuh;fle3CdHzyY|(u>f3e&+i>}
zz*8mJ%o`hq4p6G~`Y5Q*q6{UsMUsrNH-B`Nq*lkr>!a~Mub!Tj?w{UU@4cJj+;G$s
zs2H(*8D0C01Qi97orl;-;UWGTgL&_Li6)$M3*nsBsX|!m?1QSTu;9kbFy&>YQ;|PR
zt3oX)RhsscN)Y9z02T;d9nbrN5ix2K#DD!oqtK*8V~yLj*OTTGm?eVG9ei-3+3te0
zN$7pTCjrgw&|sznv@?^2TAOMmeGXRm`-y06EissDFhqsE@e^hQ#HNb}Er?{2AcuT{
zWk^|;Mf;{E=L$MxCg{`V_&dep2_QdDHt8!SC#B=>2_@elz*CB(kXvPRIQC&_+;)-M
zXUP-a6tARR6U1H|g_oh)@T7x9>8{6CXaa>YuBQkO!FXG;$a18&LB?6NiegfQd&opp
zY1UPd3!n3pNjfoba>33LcdT<3_I5-I*~`&PmmL4nD-Rj|z3FQR3WX<O(O*)GVNC!l
z{i5s)UwuLbAa481+j!1uXAR%XZ=&Ko#8cI?T0gnEWih0SfU`Xi>C2+7?l2>0UQM|Y
zcl2M|jGv~nUb?BIvFt5sMSX5w{7<Crav~;J0%r6u?p>0Tr@MLMeH)c#yc~1>qgF#h
z$6LV0oJ6HtB!B?@^e#|ZK)Mjqa{`2=c1EaWWDocopcMP3YgahB*hfT0Jyu+sjgA1f
zx4r|BlhO71Zl8WJ8qXNH`)RIs7AM($CnC@m9{I2pEa8svz;@=Qd_-<L37F^aa<KsT
z%(RVfl^|mhMjSxj{Fwmb@`kFjm&9`1hQhd5#E0q>3aTA4U^o%U$uM5CD?U5Z$^OS^
zQG(nANb$zrt8Q7bxip|HiYApi<5spMWQBS81TS|hJuh_VOj6gLPcSke{|xkO7MO(H
zpXM{B%usxUD9~RrpQ9AFZ1_iZ-ls|Dly=BRNnAt6wab~~{9*24e>vd2+{*k~kooTr
zDaL`W@}tRLfPnb`4+G(Ef-u8^f1ED+NN1rAfH4^QEN<lGsRJ8_WT3bk2gfeC!$2<v
zNN7&dS13=w1@>0+4@j`l8Qi>G1m?&)Q`&?M4Cbw$dBwi)6^2{;SI2t?_{VhM3$Kwx
z>G?N6cY*SuzYYX#?C%ZTY@>_3umJaxbQz)8l<*WPI&1S-Og~w$6C?GM>Tf(?4{qu^
zb9mcvU6|8x*Zv}!+7Yvs>E>Utg77UzRuDzK=PyGEpiOz3`+pE2g9uaEq6JO?)K?Ue
zKsdyz^kUi8B-N=T%98S7EGA1a{M3Ins01#fH+_l-(#P6=8kx@?(~l%}kjTG*|3v4E
z{|9+cfSLa5=K5)jtEXclPd&0^Ob38bzXSBY*T-R7h=l*cPSumFs)>&HT0v6K=quva
zyzG9lJc}pQgx8LvR%Iy0*Q)1l5%R@;LNCr3l7rpuC8=%Ei!t@S9|d>+Gnk9qhW=GD
zjMbH|dc{bP{Q8j<pYoojP#NZhp(vQf+lVF%7e!FHzFkS0>UAsuW?gAa0r*-L<EWA`
zcwr3)f-*J%N%~$pgYJz&1gdC;o+S+^v^aeEs~m&j9TxLwcaw-Ys$krm@C(_@&ke>A
z|2_bF!TAm?T~X!(G;!81s6;hN=NOS)A!e<h$uVShL|R5i%f~4~o)m2YhFJDi%Y@U6
zq@%%S0+C*R+W8lWSbs3e6o6xTP=)udUH*?4zn2*j6hkZ2Gcl%0&RsgeL{5nYcz!n;
zT$$YI2}_YcvjO66_t+C5eszF$x~LQsOAGL`3rldhMM(Pkapny^G<+9SKV)g@@!9Ok
z)~sfLv39^J_4ns$Fa)R;#yPjh9_C1G?N;m7PW2Mb<r5CF??X3>7~q_;IO^}ZqrI5M
zz$Ip!58q2V1Zn3U`Apz=Hob6-I9{WDe!ue+vh=#YgpJ{o7>2-<7a3wi3?-8l4sGU1
z>`wrnTM@XlpWd_`Y0w|;L7vZQ){6d);Mtke#3h_!%~|!cL^z$>)|2Zq3lAf%rF{-k
z3$1HJywpvdom*w#2!N#+y)=GuOHyPIU*xJIk;Z3`9I5NLWM*&>nrX0DQM2+m?I$cP
z1eLtPyvL?%7O}htcm9%L?OC@Y)|Q^!IJY6KEk&u08<gpms11$j4HotL)D&OpjISya
z4QuL5vPbUp!z<=btOC(1sq6rEmzLiw9Lb-Y=%a`~Rj6GV#efdnF`0^Ic;L}xC)qw$
z%!7_t9b~7JKbNn`4O^~;o{0_Eho`5Rj#&wAFf>Swe*j4h)UiDMr^*!GfsZj0+Ydg)
z4b&~*>z{fi`uJRq1~xEUi@=bJi@-WNIpx3ZxJ%YB81mpaEUX)Zn@k)v<s*;E9=Wue
ziNWyE>W4e<&j7I>c9*qhU33`CJ&ad|UFk2cTMYvGmHKZyAyEpAQ3!n7O#}+-(EL^S
zY!r}5Liyu*1A88qcRlHo?`G+7`6RU5bjkx{w!ZqndH!jQ4z|D!cpjkbM<xfZc%v1-
zexc2rTI4v+r+h~EXIPe>Plu9FH{fS-7w{0YH_CT2qyX%fH3@#HZN*>m>K`a~+Xmf*
zUF^L2l-Fev6ixV(S4;U;bnK4cciEF~%uaeK6Vhlahdj#nAe{J@I`XZ^7I$4^bodub
z58WV>&!O|Bof;qZ6)TWxKz$)zzwv^=c*#sR;Ti_f6sw6%tI6pRH6gWs?aqc@40l(P
zFpW-<9sn}7ubNQ6hmR7w6FETQ`z%de^RKrHEDXAU$cz^@phSOR&ofWmM4KHQzAsL=
z_dX_NN0dKP;oe7w!{rK{VpXxb9B~3a=uwQU19d0?bt4_tO0!beH(w}QU6766F^(MR
zYv@luS;JHAt$4HPpDQW_69IkI`^4e${gng#hhVRxsC9PB*CI7aKnA9`b{hk!^LhB$
zr$?EEv)(u%F3D?RJxXOA7Ry{-U?iG~h!-M1NNIz5P7w#DiX_dy=VPmFT}HIuGn0o=
z&2~YDBCpy&W*W3b!shDkr^C0Y7A9Mku-y>9oueuL1u^{eq4!@LOwB*!M@fyeXY2>I
zY?XCj-1-HYZCPpbKa6Bc?bqQ(PE~VZ>;?$h1a5sge{{`(T^tbp<1AXnByG_ynllb%
z@S4YD#Y<Qk4Gh(^o|?$;+oIQMoK-7O^#{WvBeKq(-&V!Tfnx;Om5S1|X?)rBpm~a>
zJfRb18x{q_7Dnz|IlEYJy#Fq74l7y>84*$l+x&QUMz+U5_@-kCqw=@>5d`?`_yEeQ
zQ&v}eiR7U3S{h>J?*3Al?P!<`W1*J_Qxh+A0($*emXd30KfP9|3KYEkQ&+{3PKE7~
z*snm9VXH`zgQM<Br=8)Ppj)U<ND*0Nf+kJE#w`;~6RKO07UtXo;7GY~OtJ}Gf~2AB
zDd&hPPYrDDwD7uvh+WJbH~svf*8tv8b4kbHl;W?4XDchGp~M`f{Xf$!^ikbS`jC7c
zWFA<-c!O#nQU-;n^Q-W59koo5P%3caQ#hu0h5XVZWNV@61er5rWhK-=L%8`Isk}DO
z>L`<~XjoGb&3wo1=&OH}+R`q&e;z$9Jtfde1Or^S8?fh;-W5Jn%P*b?+X0e#|AcM2
zlefV=N08dLyO_o6*pM683z}^&v`BO$)GPquTnylX!j0|GpVG_Bfts1~9I3O3PsJap
z$AvAlO2}xaVAphNiNj&@fzBXS`}y$%-hI%E68>)Rb=}dBn2MI<9#_OdBY{XIn-u4%
zVQA3E6b8{2VoeG{$Z~2o@c@!Z4C>XF+PEs{^LJ@AFpFXn%m=CU>6YejR-kb#nM+>^
zM^A`T(-~bd=XWH#4aHv*IP3|fY_vKtzB9wWi^u)tI!_*p!g;9s7$)+mO;cuzl_r!q
z<$f!VH;Z_UX&r^lNX?Ssk+*<pibFP!a-T+2i9*p|6;X@KLo_m(q61L5V~9kpW1WX%
zD8!_Ef^RD(;@!LX5#i?vZgy)CVt-Z@DgLu6k34czoZ4*<GNg;5v6h+^ykX#Rl}C6a
z+xL!NE%*Cz;;lvhs4%S9zm|n7ynON_PXlv!Gk{$uUW9s_Qj|KSMUwF9w;I}n84wDY
zN|XQmofcn(b5!g{1yI-F2^UO*QsE7cFD9IXoKGle){*c5Uvj?}uBHVT`#*&^QL1K5
z-Kr-^uTh#<x0?UN8~K6IRxT=_6n=|3$7{)PW~ooc4|0c_K3rC}Cu#@@fSH;0EL$9{
z?tzq7A^0&AgStV#p>N=8QBqf3*CwInB}+k}pb%GWJ0B-E02=(dnXKd}rSnxmc=M#N
zzDKl&2X}e*?#mFkDW_-34K*QJuZe$ZdSN8ElNp6?i_umXeYv@B-)ur;XiRYymuF6I
z2;`rYwFDFdGpVU;>!cP>aZVRXdGz@%UvZ9~wuIh)O8e+sUisEUg)WMOJH^6Mfu*n~
zph{lZONeEtVVXPU<>yXz0BX-({{sh()IYJT$f*XHLKLZHo{Z_h-YH&;=75vFAp%Gv
z*T<9hmHhM953Nt}9Im7KR!F{JYoz%%@X(Q29!RuCu#D>YO(g9K+Xkj4k@6)^nKZp=
zJq3^Bh9&8K-C&(`_9hTP)iE$$gk96u7`Z(<ZPlt>Zze@+aKom-l}?7M_Il_aL)-pH
zs<KwG$BY+$Z;EE_##uFk0l-V!9<kBAAV>{Hnt580+GL!mL_e;@3J~7>0J4=W(0K;J
z@E6B*0y9%S;?Si>DwdDnBGqV;I0>HE$4iQHp0I+8M#>SY0OtfFp9h{eAp)(SqIPPM
zQ+)CGHmyEQ4A<Gdsr9VDO=jtbs=<+}=8DBKk!d2CC3s$?YV<gW1WYNSMHdE?C^xfu
z?+mdVNu-tGp=QigE2O5UPl-<nbqAcV@RMF6RS5E`Z>{w_BsD}YC1z#Q$Z*aG7DqF0
zV$n02{@YKirlI}Pz*isImk-y^iS2TeX`G#;)mqeoa#}t1l|Yz+@+YN~<18m2l`fji
z(m3LL+st4L|7&Y*52zQiB;-UgE&fRIG4c>~0)J7Qgp2Q8iP9doD+a~dz;x%y1i`aO
ztk)hH{O)qmoq8TBbG5)5de%ZxM|T$xi1I`}i7ews@-$yZyCtp%?LC~5&-Y_e^EmQ@
zI=i;<C3tkts4eo-^9Jj6007l?45pthPQ@ueAe_dyJHa#K0%{h4J%^-l<8s6x)4fvn
zSovTY28MgcQbyue%FYND96S#%h8*j2#cD!R2NP55cE;Lr4x|>t3P*gF#j8iGuJZAy
zBQ5qk!P}@dIPwv_*AC4cn-(UDhv;4~+5F_rWN-*k%^P0{J*ddBTq<OiN`r389=cL=
zGwXhpf);iu1L6)^EYrpNE&pmiy>#6YsF(-ygf`f9>?boV&Pbhc*dfggDpTs7>g=ph
zbtFbD6>PAMr(Bm!{D}2l#5C(mQ|3h$ru-%vdDkz{OhQ4Jf%&}JRW6W_J05ruay)X(
zB84L6eGDb?Rvay<^3<{DfV$zSz3rjpbh@A~BxqLC0I1}xGD5NgUQ4~N9&OZYA{ps~
zh__mFggvo7SLuXg(HlaZ3B4hAB(4$Y6;3M(e9i57K2WsGMIh>yD(WA@q>eNxOd7J<
ze7*=1&+DO-_MlSUgT5FC;(4{)d2xid4nzdEY48*(VuZ(qBTo#2J{Gydjf19-%XH^j
z-P9<<fJtE4B|d}iA`HC{ye8lMvO1mjKxFsvZAhG=@hq$DRop-3VfPK28|!#SV~+Yv
z`~#k~Z~T2<;F)ADT}sv;q-?U;)jv$2w>RZ=u#q-*V9lm<K4Ssr?~YrmT^97JRm$+u
z;uZ%BQ!eghM;}}m?PiGN@@-|b%)l-Vsp3jJzyhhJmq{<io7bYc3CXyr4-ZPQ3{UxJ
zbRyTX@lrrpx79hkmHdGZ%QZ@ew<Upd)!Q4^O<SEKPBry5dgmh{X$)=Q;rkR$1gUu$
zdCSqkLFW%8LEP1Gq(~0J@R0}R4tmR(`$&%pIN2n3w&oXUGz>HyF+O1}=+=6lkRm7s
z09kWi!tZMT)uM2FXAFV?Pusc$U+};@#R6WX-Z|8v6LPS-nhx6s4}uvxT6@WV0LZpw
zSM%pO{M;$YW}S&eK_-oqb=A42G~WV>rFwaeoAGR@1$)e9(H4nyq}xK6Q0M6kI}C4A
z!o$?ecnbSu%>3JR81(bP8l~FsrF{ZG)t~3BS<vUmCyuDR;%chlcGBf8AMbau>=zQ$
zeTigehpaWfWJzp{v0F-4rGf$kL*lekwbZ=!sMBV9Tzzq=>lV9V#X%xZYcAkJ_981g
zYB?PY_f@!p?xq}HY4R1+G*(_Wo3+*cG?A?hP33eZ+FBi^W_;AfdBtnwj;ZALG}w{M
z4%6co#2tp)>3@(K`A1arLv4|!>Z{SgfH3?hjI`Lu0a`ou*z8Dt&-#M9dS38HP8@@T
znkeA&$Mfg;U;!KuI<q%{v@qq~HK=1w@$c8y*iuPJ8g<WEkPeT#Khf}hjvP3CgIhwu
zIu5~7E=ZUBokUOiIHF+jYw>6ZteqO;$ru&JoBrh7b7U$y6|xok?z|w*4k!tfA&{)q
zDuu-P0234!GDg&9u8FlG5E5w7LITAE46tyeB}+qv<VfB?8`?d{Bn}6ba8Rktpbm*P
z|A0;s*#iHHY4i`QHIlN9F{mC9IJ$_0PiS9CR{4B+m=R!IHj4$JLyj;{XmU=pNYHxY
zey}7CHAY_yf|IOhC2~R*C{So&2-I<?6EVVF01}mr2zc^3Gq9?xzKk4FCZ9||Q3<_P
z+xWvUh*YMCYl1K~VvcD&hJ=ICB1xW#mD#F+6Q#zG?Mt}UP`WM>EJ9LSlFb_;GK6uo
z(v&5GZ&tR_Js5Icy(M^~RC607x2#MZ@Igj{XRJUUZz6&W9v&7Lt=2RhnW%Fl_(1s@
zz=@bh6-B-AkGb?UqX1+YKvM^9>6lnFm2!hwM&;N*D~^h@S#vnZ%+@n*(R30=K>@YE
zGvlH_7R}Cq6)M;<n#|Cs0+9{nr+L9CIDn>P7-yoGvah(#Qqh8vR_$1U^e2<0dYt4U
zbe$+Djv&s^vMVH|leFEcT!+Z1+G#QYV54$RRasf?85~~gSzg-)?JQ<E9rQP$dX<T}
z-v~;w&IndxQsb9SY|p2ghi3#SjmExx$d4k(o@chlz+Y#aS;)iL-PhO6=L5lfhEL$&
zAWH%W-y=${sCu@edg1rZY>oMS;`2#jxiPZC;Q9Tq5m&FHqvv}od?dac+DT+NpztVA
zy)hHarc3Bai9(vXw(5qY>c!2Nu^x1I`u6t5i!U=v2EBCfJhe8YbpbvyInn<6-4ELM
zy5wfhjImpDQwvI?!h^Xxduy6}VZ-#XPMlw!zILWJ)0g9Sn<`46&H?7(Nt{SUHIApF
zGc&WO8g1rl;er~&^t~E`x`-|!fYl{{3z2&c+@Zy-xg1aKiVn@~d0}Z9uz9m(<HwoJ
zo;t90^Sn2@_pYq9-IXqfR_d@#yjyNE3#M9UM8r^DPDW3e{hJ=ADHoQ1YI`Om^@@54
zhh`^!7OZ@Pb*L8)-#)nVwQhrktWS|0yRbDOuI()-ycETE*hr>;uPg@_;L8%VH>~w(
z;>g<hNBG<4zVOrc@!aV4>nYRF!@c?C@oGTctQWp_>+|dDeDV~3kpIfh?-J07M^|Hx
z9w<jHN~!-JLJSQ0>eYlPx>-+8ZXh@rQjR5`VRMFD_l}+)asJR+htnWUTppjCsu{k;
zx_X3if)C7(?O?IZ+D9HRN!oH0vjvqrZv5k_AjiYonk1DfXJ8whDk}}@2B-Ye91pTO
zZMF%RD#C+2kWY}2hhaO5Z?)~E*8#$X_qv8jW${{;tpWy9?9X+tvRx?m>^?P+p7Gbs
zy^zv|{b?dd@TukrVNRT7jM<e@<OQ0<kVCS6A2`Sg2o4ox4M7IH+w_M#aF~l^pyCxG
z{{t!ZiIPJ?@x}T%pU3v`>G2>KxuL{k82SX|=`#$3sCb~|qlG$WR|9n=atq)l%+`>$
zB)>xsgibjys$Qt*vWC*Yb^SilZTGOI^%_?!>cr$byo-4n4z=n>v*%S#*d^YIbd8UK
zq!xcbE~m=J{s2@d6mmX9keUCU1fhyd3508Q<y)S4RB*1&abEMT;-sa%z-k7<GTOp%
zshV5%?oacsnhWq2`0gIX0z+iTTB(nD;u$j}77-rCa25{}(b!0TjzNeYY5A_IbY<<j
z-@MJw{<al>L?esX^^JRBT!DTA{u^Jym?e_qgxOCClmnPz3SowrO%Qj`a!(P$8N_t&
zZ_>?T%p1fu#^FH2#k&4AYb)3m+qd-{c@lKB$5|ua?;O7DF(ap$$`QqFaR9g87RYv$
z7yWa(8Gd)d|4;I8p+pyy?EzTeI1Z*3s1j`Hk&Ila?&|)lAG>dTSrQp6chFFrYx&<L
zM-MqG4&W~MDR*%$Z=-}TJY;jiFf;@7k;qO+HheDWl7DeIF6g%enYBa!pEn2L-rhlp
zZAfTrrkm3hkKOuk`d{u}57B#30Z)EWGh4M0KbL(}n<AZ`D+{u-X1`Pw7d_ymEIgo-
zG<qntg~AsRMb03JTtVZx0wiz+i(&|rPDMTg016<#or-BG;y+{Xt|p10xR0u^Df0U(
z@i-C?^f_yZX(>7gFZdS&mKp6;bp_vX1`S8o>jF`>2atwKQJ!N-Sv=1%mkkE2xG(Tn
z(t!M+>+Sl2BC_4CzK>Kv$!a_U$$p{@Q_N$@l16ake1;wgidKiy*O2ob4zeWw&gHY>
z0s1kQR#3zpoLKYXMjTRB^}u;%!f_X<W5bc&Z16?I96+cwsM!M=8{{831Ct7#1tunj
z1_?-)<pn10RF{y74r4@|#EII8<2U0a^nCY>cBrbu{G^n)Mkw{KU{c-q(AmUiCSK&<
zuSb;WwtLRB29t^SQh$w7Th+3#LEyjWTaaGSE+Q$0NBGjD|ABR8#$OLjDC2$$Vm`Tl
z4Xxz=u=%p%=xxK5s9OgkeYocX-p{COw{@rly@U8lDXnX@*I(Hw$upcHc@I^};_({!
zos{s8vt%$t$zr3G3)ILaR!2Pa4X~NN@&Cmw%&C27<S40s;OQt*1H%|IfOk^y!Wr=a
zbe0}`o*KzEQFUHJyP{GDjY(m^ebMV^2+st6-{dx~P~w3{SFtzwL}h-TvfSo8SWQ3E
z<>xRU_lvFW->h%p?y!H`RVb+ptTJf)6g5D9f2<EdV2y#JgNq;v!1wmT^Roe-l-B4!
z(*me~sc0`v8$_E1FxH8ze!fRrs!A3BBvpRDZ=8<w))hy4HpsYD>%deGS)s}r2RRLY
z&UcnIQwoby`7>c~9^mU2y!$Mx?0u&^5%K!_f@>Ef_LEYqQy9d78K!L*T$Sk2!*3+g
z;Wj?5CRr^`>B4rd#m<-Qo_c6WFfoDCyNNQ`hqa51bEm5y$l^=ePZ0`w|4>3e5#%+j
z)CdqvD(fG#R#lx{l4$p(46vO9fm-f%l#&Zot?av;ZWSNdF^l;^_oxg*(6wAKHF)Nf
zQ`+R(HrLAuY)Q3Mvi1@$%S9E^di0le)j7Fwtcb;z4<M`9>J*aEhTjR7H_G_2U%5la
ze+q}$QDai5<m1bt9#h1yDOQvLSu=c^kxTn3fdfXe-pV#d9!Ibp!HKY_JP^d>o@QOf
zc*N8dx?ew%fcgUSln2fbrIFpStZ+}<$p-b?)OeLm$LwA{6U~WX(F3$QMaly16-;+A
z_t;XdOaHXf$}&0RVu~Ytd|=Fb#1ri_M9bNHcH3a|_!ouAM$|ct=_3LFxC)Kv9P@eb
z`%m`<E2P6WJjjKOFQ{2++6?C2L;5jyy4*t8+x>z2kCP}(@6j!4YPGF~W5EWoXS!c!
zB`Cg>ih3D2$_@s;(*(96o}PR&qH2)-Znfs<IK?&7eO2^rY@MMt6Y8oZR%t6|%ZL9l
ze5$+{QHn2tGr7=>R^=E#<0~CinDKWddpoJvi(Y-rUiD3gvVkJ$G6(?dO^J1?@hN&C
z3IxR!TI>DCKdPHd_pZsefhk*_^6GH!7b#v~L8xr_4j*9=B)n|S>b;F{Ukla0tY=TQ
znrhyAN^@p@#t#y)v-)Hq^5|v?Vb}rxvrE?ZfFRgg?F2UsdKWPO)Zd;|6vkq`aNDWJ
z;`LVQ=GqI~HoxZ1sANX`lnHEK+KKs+bVXMNvm={$l1!rZ;W`yXvBi5$BYrgso63f4
zwNelMtL9qs_5>T}<e&mnCtN=i({lk$k(pCtd-eqL@^&bwOwZa20zxN@9qKYsJ{yI$
zkb!9TfQ;(x1gSMZQigTA@Na=WE51tk<D2@(!TlO&&=4wS%)hDxaatM${*Oo^!%n;K
zgR*BP$I+8YM4`hCD9WOe%G{Z$@rTl3L@omKC|t(ng}=Lpf|M^i3Jx~_zISo|N)w};
zzEVpCiq2iWTeupD(X}NzgNQd=Fy|vZs1PPPv2Ti&`{F$SR6g<F4a?(G8>ngW%>^AQ
zHfV7Rr+>F4rIT1fw<Q#F{R?4}T4G@*!~0C%Y~2l3|2aaKn%P2ZL-*<B9{h9zHrd~Y
z6p|n**C3O~<Fg?gd=vgtSP<Mq=%`=OE;s62e6M_>R&f}O-9&F?woPa}AoMu<nFBZC
zAzJNP*!Yc@e|S{6BJuODdIdj=Ym-x+a4h5W?o_9=$)1A2SO>CO-aTbko><=8iZk>{
zp80)%f`isD8nWRPs7wna$syOCrAE5ggKD@w=3ih1OMj~sno+nwlS8n6#jc}<HeUh#
z<f{Nm*iQdDutEl5W=h>t2SZL3ieVfCHf#ltWz+=$Xzm<+!fJw)+2nCH7B#;ZY&<zE
zt}oS!R%d_alvygwthO8tlpF;Z^Zdkn6OMUcve|S;FVR@gaLDnt{}qXXC{`ia6_&P;
zHnD;+jBThinbE`*2Hp4Mz&w$y*TRNpw^XaM4$Y6++w+`B+|0|8`T4oIfmG0e@V4O1
z@MCHL&|iaP1LN)WfV{~x+7?n>Fxa~(4i)#H+HPc$v4BZ_=w;M*VpANLQa5T#%%>A1
z%>4b}Y{culH&x9yti|cf6+?IPApd^=tw2)0$`Bq<hMl!v+Hi4g*yrDes>2S<jr!2Y
zIf4@xW>$!Zq7Z>UQXv`$${Y@&8G5YF;*1KhvwTG%cC&vMQiv$YPs9r=#BSAhD#Xs(
zhbTmJqFHdxzg%<{=Q-c=+5Uq>rT`|XqaE;bvcU!A!1c2qkt^OSM0@MHQef#QUhhyV
zXekxPWi%#|s4SlIFA<%^#q;cOnOfuCI&dZK?d12w-H6HvctWMZ|3amLR7C)uQKf<t
zl{&FF&6H|g*W%r*OsA<aX3ZVS6j)RqgqxM=-9T@ULzf$Til>wBfG!|M0c6*U@LWcY
z##J%QpBtI}WfZ6s6bI&~5uh#HgZ_lw1NyTy_Dg;)E<d~E`%v}SV!2U#%ISb`^M5D1
z3L_QlKQbxVv6oS}F%tqaGnc_!I}?{Nf+Z=Jo0Bmy0uou5&XX}4e|$soCrF+&)K~yv
z6f~C~%^zQ+M8nsKFD<1~d|CePc%;K#f#VfGeDPg?!`^BL&+r<B-LuMc&nRAnXC8sa
zW0>dXy)-gN;NPIac=+@3+fF89rSK+v4MY9w{j&RK*PD(?kns&!5?b2kuOGW`Ljwt8
zJ$t5wk*@jqck^xce|s&qS0?$H?t^6W^Ltaxd`;+i$>>RAhzj)32qM=Q<An;hI`s4}
zQWB%3TfW$P%{U<xUcR*sW+jplO8HtM!zhWd)r_f*k~?Yacr$!N^NSfaznAPF?sO#?
zWnovp+DoCVu2-vx9X!EtBH2rLPc~hL^eUd(3a#uNcuL-Le-;m_fuJaAEoQ#nfO)AE
z*;(?PHljhZreuj*K#t=MCq4v}Xb!#hus1T~4?O`#BthV}jLTAD@P-&CjTsaEZMrMW
zQSkJXFoZOR2~SMx^&z_N61?s3Xjm#wcZWI}6vj`C1)AM3G)8z*^=f&4ieack25;4C
zy|7>>de2EOf15OXM7_n#IJNV3gXW6Z?~Uznj=;JqTON-s@(i+88^%^8@g=d9((dDX
zT^W64udCG$>@{&YH%|9pokAJGU_l#;dMYZFJxiw`1Ed^JD>BZ6(&;58JLc{T#dO+G
z<WpIfg~FuGacP6hBq&qBEw(5ujSuwM($j5wl(uMBf0iOaAoM$cDQv^R78)IZ70C-N
z6j^SaR3-V?hE@k`YNnzbNI2LbXv23%X>xDlqm7D%t7?1@g$I45@2SSv<1A!n3OGKr
z1mX((MLOj(?K}$dIYSmf%|0s4CJo(+KI{>KWbIQhK^IIB)N=+ptimr<jj(xE`%yV)
z!>06ie<<Av;rTw6Lbw`Uj0}gns_pI4G9uCD1f}z8I3Xzk+SQT*rl?f1F?!znL&xN$
z%f3g1Oz9=3+dj_Qydpk#&M0MRWlHoqRy(0FhACw-qV5orAjCz87rukt{HU6r%LEyG
z4XYuqwz+&9A_M7G_ulFhWk}9pDA#;BmCNa5e|DvWX)e;9H>ZxhV@5cQ841(zo9sb8
z0rR?yTHsG36w1;!9}DEThqSh~Bo&OemeX=AcffG_i@p6xf3!!Y>>c&7_m&^@K8~Nr
zyOd{Bz;4)sR1!DjU6tf3ysKLMBHpDe2=e~LysOgsAn&TAeG2bVnp+hITgGK6asKoj
zf7?&V<iY+tvu1ye)k@nklUTIdJK>zFCH0N=$8AJz<dnw=#-lTx%*}QpI$?~S*!l6s
zTq!3k;)__A;u`Pdmj66e5}zGum9_^Wt&}8(e(Utueo7KfFR|*ryDK`p(<hNGG-cjC
zPP$MlVeNd~F#{U84~m^++x7jCgCEoVf7}PzQSTIqc;x2;jlYS*Arc>wbTfy$+tBV$
zgNdy(hA#B*;lDrb;Y&kbKncM)0YERJgyG0l{t3a+1;xyNJWpVlK#qB>{m(GW%HDxt
za-F}AVOEm<XP7T%n04%Z9>c6GxshRlX1~3D7Q-}_hsldQBl$$am+(sf8&z)CQ(aXO
zH{_R<<SYELTKyt^=}A}vFQ%8J(g)dPDeP0orB&RIZu#D&6u8YP-gQv4bc%oYm0oEp
z=+!rR<>Cik@>E*a-|<vh*LstB<SO_tLxjxPmr=Me69F@qG2sOhld_8`mjszHm4D#1
z;raB}+wTXNg_Xja@EC6W+w0rm-$NB0mC(iyMoEDZhrfRu!UTa9#`=N0aG-{t{|w&_
zfAp(XURn*)&(H5eH}W+b&jA}x8l&~hcxZ&yZZO6R6{eE$$n(Xfp^KE74MDC`Q1a*L
zN6W-?Ho{E5T)e!>k*pwPE5{e+<bM)J8WVSf8P`o6v=K`AenriW)a)F>q=RN>YMocl
z?@_-o(a<^1yTm=tyHkjZIE;DdIZtyuNPgCQ$x9McP|&~pM6=9B*epe8v|5nCUTDxW
zl_zR1EZCk}j+yU%3@psAj3Dc(v7S#)M&dCVJwJ7_Y{s9H%JHz$#u>gF&wr7eWLTT|
zS{=FE$z?l85=Kw5ne0rLJ+TbP9BiApL9_j|_jiZJC51(#HEXdDrP<W0NMV4txvCi|
zhdc`sGZer_C5yqjDcpMcM5P28+X;8Rgfs^>C(<YuOd^#H5S}6L(p!_BMk`>M^bmyy
z+DUrd{T&HTvzn@j_lHG5vVZw9A@C-Osbg%m>6$K^uvW|@Au3RUhE7iikaKk063~b|
z7GDLbu!(1pV9RKm-&2mSGdj-%C12ZBG+qsIMBZYIR-L>lu>^VHQWA?EuScX-0?73%
z1)FlPW*MXARUduMTDqcpw2+x=$w}iWj@v>JzZ<J7^J?>rS$Sle)qgOdY^@p^l`?oh
zSBo-+*kg)|5OO?&X}k%ialXeUT>~`~YFm=vDHbq^x@t=~%1}Fp0dV+ossO@b^hzQo
z&zq2))5qSqSUBIC4WnSvpkR2RD^?5TS7U{F0$+~>?HKT^?InSd(|G90-XTX^K~?cH
z`zVgg`~Naw$*l>DtA9g*xJY0S&`Y|#vTeVDidpS?5k-as82`W`s#`JIDdaz!U3q(@
z0}dkkbt|J>t#rV1y<n-QRtq^&vX!hQyk?m`UX&6cZxp=2o?O*fUY-*l1V)kTY~<@Z
zY(D3TYA@MrSdQZk7dk{pTC?5+QJ-K(WH0iLeqdbAS9}b9Cx7go74)|k^4z6)WqJXD
z)(d&fVZG4ip8mlg*VBuYw)zKSgy)>!^S!@+Aju7*tqo(bfAsR>n?)61KtHls>{k4=
zc)Gtq%nt4Z-EnY7{Q(Cn%MGGmX%MG6?tg0#6>+Mwk8BVXcwt}yr(6Z6vyar8MR(`2
zA%;h&>?!#eUVqZHT_MHLSVgh%UA?2?0tivvF5uNt8w{7)9=p|><}XNGO3x73Mf!W2
zKJGLH;IC@>&~no^#qc!`KBCaEmSYHdt|>OwA=??YM*qRpGJ89MZUZe4g!Tqe3(S2;
z)(^D7L1H~^d0e>i^hD&_hS$}Rjp~QoT7%40)-ks)W`9Q4Hez|jSfxci?%1}PWtIJw
zvE@K32Q9j^<KuMT5?a2bS%Dd<a<7DOOKUr1tBv&3!e;8G#m+022G3Tz;$;6QpUn4{
z5)0LunKz`Gf8kr?<U8fI>k_8(dO~$fXZsPA&zZ-VbV%p>Tdi|!#-U`mHeS+5^4JN?
zF&_$NkAIgqZw|c{ng9wnRYkQa&eHSV%93wKQ>(%3=)g7-ERGDS?6x_)rY2869)!V%
z<YgZS!eF^$GVUtIjnn%gmG1)ashkRZ0Ej_zU1sq<A$P!-q0HHi!zY3<fa9DBzX*&0
zNjREn_#IC^@Z_Q6d<~HE=|lI$-H3b-(s?E1w12IXKUsTh*@Ak;CQ(S)pMTosIsDm~
zn*27+rBm1ys7SO_(x`;3swvXa3Vf!uTlma$H%eH_<@k5(Sl6AOPAh$E9Ak?$_T;ZK
z)P4!OQsSzVyLMay3XV)UC6R>g>xkrXZY`&l^HFVQe^wUUpc#r3*QB>`sv`1;nd4Di
zuQK^Fp+QD9)ilkjemdBW-^i-XYL$=<;@e80;b7lY1AG>)84?QV23YIwoNNCB=D?YT
zmr=Me69O|fmoal569O|cm+`nICzp_)F%tusRZ*9wpD`zY-alWy-V9y53Kl$GK7YRe
zKhXRQnr}LL@c_yhxP1P3`PZw~==k%_AJW<w{&@Z$_}9LDF%WnJSpM)+fVVH6Aicw5
z43>9|EAJQ{g&+J2{CiFJ`TeM!P89eWm^OU8{rmGDFS;`O^%nIWr_PgN+MQFLftU}=
zGg3(W@Gt*=x6FxU>Xv8MED3xopCl34GTt(Ww!TmG+UU0=OnRMsAxSqTQJDOC3$V+a
zwaO73XRcD+P>w~#U<s<%TLKaMYht$0O^nm~#5F4=L#m=nsHUGE;_Cb7%YR?`L}-+C
z@#0V`5B~C>KVQ-V0-o2$3o;u(m%o0y{QmM!NwunfP5x$^%V258{10NGQ`QEa&8Zlv
zfwiVJ#!slC3Pwz-fBp5xMcn!MW0K891sdgS#U2LB>u-3k9Dsb|zU6G1%{pY8YOmU`
zku9CD&Ck9^^g1crreFJ(z0tB48e`K_O8?vf&vrL+82eQ<918u;I^tV~kH)ZVILYfK
zm34_v(R#KBlvGk|A)n++F~s^VS-NtaV^{C!@|$qaPG{#*TCAEt8v-UP;ZHEu6zf^s
z{rO{4W)5-+aUrv`U&3wsy<-6!Iu#hZm?6Anx3}+?nxQdve{UQ{_acb1oGW_(>FUkE
z?Pi`vfV_&nhrnfsd;5YASAq?qo;hk6GdHjDJcrUSnMr<X8;1p#N<SOWzM=>fF}8x|
zRF>ot#Fi2!mnVK-TOO75MpBYG*GYQQxXZacl%Qo|t|6|?NQ>!KNvKP|hUN-x9a|Wa
zPeQIGpN>|ce;j6EvAA8+!Q?p43Wwwj*FHg_m8JTqX>!i#<6+TqA;M0vm}!O&==A_&
zoYMSe)8qz<#r_zp5MC$r>>OjNIOh>9nPT`w*did0A<4ga3_mnM`6aa@_EnLI(9$xP
z86QKMNY_08S1*ZXhG*kCCx#&ILB;V_<48dks|OX-f9HPXXa8**U*hUqOC?2)g@XMy
zf3H)f>U(8oGfCGa`<Nl`*X<2CaXT}_`4r=|-SB&(-BC~fn8xV*W{?R{w`88ByY@S)
z^V%kNE1f28S`iT?T`s2D)m?1oT#j*acv3ZXl<GsA`#I;O$mn$L>$#1bipho6*;_$R
zQD!Jof4tO5DO=Wb0#o$q<9<PG(NVcn*$A%hV6W*|tl^5srXqQyo=Y%l`Z`?5P~<|!
zPJMO4u<2OWNOC(|)etv$G1^Y8A&pULE#FDyWY$sTIU{r(vuhpM3-mSKs=XWNZJK_d
z;d0y>AW0d{)}z5GLklXsvBI-;N5$?sRj~Blf9X(v9$RZxy{op=gYGt}3~iO+h$`iN
z-<6Zt`kc25Ip=k^;Biap%xkL%{64RqTL!z}HB)ZYP~7J`21O1fg`9b<pEtx97kV3V
zBCJL#a$EtduDoQmghfjNv+!BM6YYesT?5ue#RSg4!W0>IE5uEErocEYh)Q$Z>K&J1
z9QmPcvMu$s(wnXm?Sf|*6UgQY1NZ&Fp;?IYUYDn&F;6Gk6xk3&rrcN#wH%90loX$_
z^cf(pN(;)U*OxsbqsLv97`?iz`JR^@r7=u@yTIrF*NS1A(D83p$~bA#$%&jSb(7K!
zaNrq<-dBwbb>#`pU_Rl9=s}^DaPr*J7f%rloJc(gqF~oy2qCXo3m6`^J;OB)OdE!x
zwy7Y(w;5>lA4^>S309wCuhuIB^TenuC?2d+PDN?1^#Ug|G1EqrM(ve}xRCVF=GoJK
zqqZ?tA*a_T;yqH_sCgdAf`am$EGYR{!P0y`9|*WoS;)At4T)U8fwilM=*RsQiE;M7
zGG}O11o4?RAt>IjLeKeM$=+g?P5-O;WE}l0`@o=_BT0SBLQLA24|nl-+*&XK_n}V>
z^6ND@TZjpK+gQgf3)c)$V`Xve$<DHWR__sPV1pf({ITs|w(jlJ$&<9#jktUAB&CH`
zIJV`H;q=KzIU_UG8U4w1)>(#G`RUe6a1P>vLK>CN-_%7`jI*WDP_Fw1Tic)Yh5s|-
z4<3a<;5Nd*IkfH^hw;i8BPjpE<1m2SW&+AK(|hH)hI;mK7zi#mwX1-gk7~q2GI%Sg
z0<32}V!VwUyl&0of>N7?+k3=uAra3rZPWHRNop%+by6H?X^97xK4X4oFHt#;D$&SU
zif<lEO!Z8WprCKJDW@@e0R*2z6>*nQxG@s}GncX9I1`u8s4**lOK&4L48HqU=nq)Y
z`vC+-9cLY&w=8lidMJV>JtReUd+6Wqr{`$onThQ<*~12K<dG<fq)2|0#*91@M($QO
z538%aRK!xJxfSH$iEs=s{O1gvMia|9HIkEuSMu96W5VEbgO6fNN_;H7JG6SUmfV4Y
zYkV-u`DSeu4|70&BSo4kO`0P?C3a{z+HL5^e9sIE9JZN*-2DFV>xyMzulAlTNouU3
z;9VqzaB0<DkOki%rN+t%&uX9A`W(hgqSJbKR%H(OXt4>(XgN#+WomCTY3P7)_-*<G
z#bQ4n#^Z7(iI-U_n}&l_Y!`LD+B?3etJc&g-s-AU7VIW}>0Ja3`Ue5DZG`ZNUS%B*
z?UMvM*+fhBN#@g8ziF?N%u&yHr{8uTyo1}KsT(v6O9uPF+oG>fZ-Tpo=B|7C^04~0
zs)iJjYU7FFj!M}2@pTmj6vw*O6CtSQihTc*Jgx3~uGt^^HU?ANHfA9IXxoILT03Hd
zq6Y1Ri%OS&6;PN`|9yWZUC-CMK?emLtnRrMQ)ICQYS>%$#OH10;Od%{JxCrC-S}ip
zgtU~~1L0B}O==?)HVVe>X1pD7li@bK!Lz9``;K;wDlgmA7=vnGZ7>_gicu>>&tdEj
zA02DK=+^WyM83R)TM098%;{a}CUtz-NOj#<+y*Lt6S-{G(I5Px9VEa7et96n!A>nF
z4XLPwYj<37931SLWoY|Niy4w<^=(l<Dm$n?+Tbjjt)shQI9GD|F7AP-fni$Sy^=M9
zX@|{+FG_8kL2En{DSReJzTVBdSy5RFYQzxpChJ)OB)Zl~SgVoHEPprF2Hk@M#hC}k
zP7Zv3F`VrZ)X_Mrfk#sOD4d-)&ysE409b`gt2)I#5MY(%z{Af1SY<7h^3nj?aa{)3
zj`+Mi?7_9>^s~S<*!sXWi{Q$s61Jc5so<(0LnsN(2iK0n6u5psfC(va3X%Qlmx9%|
zA(a{I-NEvwA}XWELOqumL{(^o`l%3AF->uQR=FUeO6#fg??=>(PI0XGQ}DEWdKH&E
z!B&=MoM5E^F0dbltg~iXvQe!d-U3-JW4$%j@dH6tqP6O?KsIIverb^HxGsZiM||FP
zPJwNAA#*y|X6Wp-Spr*6y)^xdPX$}yB?jLEwjGB_u>H`O6G~9!Y%j?Ban~M+yNEe|
z2RH85KNWaA)x2jigS*0N>cvlmyFy#)-D$XsjO7uNf|)5C62igagQ@H7b@;wnqX}2i
z;x;W3{Thc;C$AHYC=&3-v3KLJ)h6#bB*9|d8aXKmY6>>!c9|uYFZa1KFnx6)S;Au%
zIb;Wjg>hJ8TU;nQR4GDCj_QH_Ajo2WWV`Hy`;y1XXnB<&c9s6^a0_KN$F_AEjE#0N
zdxcXo9JE_3`#BK#z^)u&tX#n3Zv?J*N1{I>B&2tm{6}Ip<w7eTyq|L$qQFc1g#cMy
z{qy|v%i|v?AIS5+ffS%lh34Lo*Yv3I?6gzlB|k$nskkScdL|7{_;598K~N@t){aNA
z0z^?RwQG7`LN5ZvsdzD2CQDAGvw^TI&~z0U0>|20)_Z9?i&VoBQKu?KcnVzd-0+C`
zWmb3$pZ2+sV%A?|^+Z<p*{a$I8%P{WjJ?6%lvRP$INb2ovZYHR%~-EYv5L2FlbPOD
z-`XiC5}U#?5b4Dtqrg+kYE|cd($nZ&y1=V&C|#xP`)BOX>587}q#8L0^E8fjHn=Fy
z)=d^3zCXU^dC!=2*R+YxIwWO<Q{tK%g(5%)Ua@>co}0<xfxJ+ckS(F;WBrwYF0E8@
zcpG6I6D)a88j+H2>p{8uqYc72s#|r0LHY`=F$DabBT*DX%vW7e2UNa)$Vju<SBk0Z
z8++9*-LDUH3Hs}-T*B1staAn~&#nV>TltLgdgYfgGM;3g;p2D@R$8A1R)tN4g7p44
zEhqd&>Qc*v_slYit*SVzj%9@k&FSh#KE9F7H-|@T0ow;kbKT8XBOZf!`m!x<dv=ok
z<jb>+u`X*XPhH=wOAFI~^0L3S3}+F?6SLn$CC2z%_HvBYBg|KscQlyG^|%{vW3AQm
z_I4+>sh}OgV=8Ee08ByqHyM*<a9ua(dc+SHIO71=A>HgZLrJgNBELEDEL-Y`x@P*Y
zaaif7=4vCnGn-qf?t{+K!}-K@yC-#3lc~#<<{<61xevJ+&Z9?WUpACqga~9_S??Mm
z#G^Jc@f5p!*FafFq&jBPe2^aXe9iX{@v7|b{(%G3;UL!6`oFvwEg5Co)&4UV8Pdd;
z{sW^ff1P}Xp=GG-8isS%7dXEd9PFBvkFQRTyTHMo{SPgCIY^gLxG@u#v92H$m;0|V
zDu2t4D>n|^_gB;pTt%cvJ%Eb_+V0yI$SxPX8_a5uWR=C_{})3&t4b=D?Xr`}qWn;)
zD3Utj<JfxoebUqSm(%CVPalRRr#yqh>GQV<<c5}4&@yOwasbL2m_GlQ{`FOBv~r!~
zBCU;)%gJXi@BI43fR`D-a>-qQ*DsDBK7S)~4CZHz&CeK_g=cvOd4G%J@_f)%M+t95
z0=@qG^XC^`l<hUMj+Dr1n#mLNp*4$i5{;N&!O1ww3PfdGR^*WHr@8W7e?-mW%rt^|
zHo0m~97!_L{5Be;ouZ^RqRZ2Ahl~8h-M6xOc@EyQZ7SGkUVQuBef#$L^54;DVt>WX
zO#k`gCEg$d%6Oc>s$lH&=daUmFTeCsi~qFJ@`&R2BA@>Jed<Pj6hH(mfY8=5uK)pB
zP(qVsP@Y0{don7TZi=fo8oAoIpFTWnN+d1N;$9-YX9tQvfmm@9MIGx)Qhhb@V*w9#
z%ChOD3LH1VWI+Y(1|96Y@oCGbSbqu?Tyoj1Eo?ULA1f<6W|5i~smOK}j5EOEodzS5
zfQ|83kq2{%I-VMU%MqZDI-l%vH0itv$Ye>`r$J{#0U1ND^Ln46hi~;;+l@<J5^2hS
z=~s*r8Jo3s>5Ii7_q)~@;}yV@t&!`a1T%V#3Pmw7F5<qSM5sXfUYX{ebbk~tf=EJ{
z^1(L~YUG2gR{R{JBA<%dwtVM&C5N4Noz+%yC}9Gh<*ZTD<g}ZW-F7CjIMemYzJ1t=
z7~@`n>1&Sr@hC+dYGAlI5~ERU!^DPFKyHy3SXme3u&4lZB@%b<De7K`k!TgOpICr`
ziw!um)v{ZsoS@oSp}XTN3V$R_z+NMGpOackIO3*~W5iKO>T2Z2_lwyE#hqcWdK(Yd
z@(da$xG=pm{+3ITSO8d^4y|}o99KrL#^oA??ft&;ULo4SS5U9boD`#9@PqaGT(wHE
z`!pY`SY~7k);3|Y=}4xkIM-5JIJcP@ZHdWTN}lvxPMbE#ckvUA8GpUqUleKcj+VP!
zeyXF(vMb4EUo+>zAEmTw;peTbxyF8#Vdrp=GUhpF*7SY0hpT?H)?_!^IIq0t-isS&
z@M4M(G=Q^%gkA5g>WK}{r8#fwf;d$f$+nF8SEIjH<mYmpQsofSS~?R{cf1wHd!ArJ
ziV=D{%zlEbD5StjBY&=}UuoFtCS5R5v6$;1uEn_Ps;GqJkUdc`L~hj<tZFmg)n>_q
zt@+7Ce?{Fqb5VC~>K8{F9_-bM%t%CI338oRQYMJ{yk{-OXrH3UZsp}Ax{Ji#-{)F|
zy3OW?YE%$J+OOHWxIj<q4(dr=p3LlJH}?{&%ywSKn^qc0Qh(gTixWGDThv-$wstF}
zK|7TErG1qI27U4ed1vcx>ceX5HTPk4<Vqh_Dgt6w((AMxdSA4i=WHp$ZG9M9WlR_(
zUdXoh^kN=T`PfNDz1@pRIcppZv+wD}?#Rt1R%(MEFXZG}L-e{mP=}dsck664E2Ke-
zjwrcHA{gZu2Y-pYU73Ivl-gnT0ZSr^VL%4h<Cd5hu?F;j+qT36g>J;WUe$vK?2@yD
zVf4`ZwiZ?nyo8aBlP6l-y|y#bHPfwYON3-Q#u08_xO@E|8Q$9~Sh`Q5R8Q9ox<?>s
zqjb0iyLom@A_GWt4TNl_`6aEp8d(Oy8nlTN%pH5;N`Lij$uHA56`Dr_PizqIJO@*W
zxpG9}c4uwUJIt4~4z;!D3i3SK0&&&n)S8r{Rca9I#;)DrJ(8>|jXfEo(1ca+-St}4
z5@^*Dear1Ctwz&G0{`QBqBL$E4(Q1%jkvVc_P9nBcL1rZTQ<_Aq%Hf{mx+rOc&Sea
z=|ERcE`Q3_gtBu+;3bsB=K2vF7_B^DYKe?Bi}d#eFm*0Nthryf@j-$HEkZ*R?mFSu
z>$q!0oUu-kgv{p^alOQi7;qr1)k3b5ej&Ahw<>zFQ-u#@6^J>&n_=ngkc|@H!MI^<
zeK=&tjLvwep14%v%Ptx(!}7Ff5qU3-WxcICKYzYoreJH!5m|V>>h)6asGNtLyP0j>
zu6H#UP<c_V?`m%Nap?vM6GKn{^l0f^DXv7+#aNTM<7E<}u&m-9<?tvZ^E>sgaCm4@
zb~f}+-NNA+_W)am*2rlIcd>N~mZR|zmBTnM-rXFOYJjRt!_1A`zO9uhSa4UFUFey=
z%zrTW1-r6Lq%N0G$+AZ4@3SXXa80)NRq^{pX2i+i7(Ki$fC?XXdNu7=sreT@TySlC
z`G@39UM}C+=H+hTLy)8|0rOY!A(o`PpdoO$n5W@yp+gW!+CP3`Cha9Yj}GZ1U!g<#
z*?%n^(yexd4(W9}9Mg-w$A2)YKbrI!g@5(`I_7HhCM*~!3Z3hP&Z%cF?;K$dN(bL>
z!hQ$80n|g?yTHxht!tNFzhC<}dk|X`t4G_9b6YVC`7ty+LLZ=(Y(DyEn=K=mC3R>&
zTb~z<+2bt)>i8FeXY#<k#QN(<`436j;L~<%abVvu?t9F}>s9k#$mBBH0Z4pGdVkRM
z0cr*UD4_p0HFGgsZBaC-@SRy2{AaK<;N_6`e?ChCHYhX?lQdLF8h)6h0rm>r-%HZ$
z;O$9d4UkL!<r>d*<5jkg`k}>)gfONfi#;BRrO#7jo=N8P5}!-vbkg_9oDEeTAanYq
zd6&%T*Vc!VOeYET;Zgu&DZrx>{W{}Vu5c3TY5t?cMaho?SMb0bo`~B{;B#TS7QT<#
zQIQ8g+bz70*>2_I{G;^H*U@#$zYL{u{~zqom%;r&6azRiIhTP&1|^qYLl+E}QvN0;
zf8AP9ZyPrfe)q2+a4)@wa5+QHkUSIt+~(5jAq9ds1$t)~g{^E__Z-QP<s!Xbf8S7E
zTdtIJZP^#4fGx~YLvlDAe)A1ig-dk76)xE^R}+e3WfQ84Awj6Y#iS;Xi})G>yA-Wg
zjV|S;oSYDpt%t_ES0Y81Q#76Ia<!3be{nUE3DptV95Dn(Tw6bKB-|u+q>@ciM+Ig;
z%8n|xwgO2jCJ5DWAdD9&K$2(@L8{`U5Wb~oJ%}OYiZ+f|BU_JPl3EW&-)QDV7RM6A
zdL&4~wg|LoX>IDT{x)7VzgQ0{E^@K)nBgL_6=9B;l39R>Vk*`H%ZkFy0_aI@e>TIr
zN)dCwB{8@0qQbvov$09F*_Ry#Zf(USmSR$MDPdR81O8Vwy_g^=*&G3T5?Y%;O0tJG
ze4Cps*tS$`qa!vaRke`?-9obp4ih4qBNiu7+8i)E6&o)F2jgvz5_;6sBymdFe1#BE
zY=c>VHcmDhK9t)Irh>Sz-H%u*e<H-j!|<BOEJzHg?YZDvX)9MmlD3@@e49lu2R2!Y
zg<zEAa$vI$4$IVbU({tTY7$~j$RxNV=e7<6T+ZmL$)pln8-lbTfRiF<a;0Q*6iCwg
zV)9UJS2G@|+Pvm4Zjq>HOU-MF$E9EFv9>+3Q&DkXQif?YM6)lOlw}WXf5}y9(UeWX
zqR}R<BA_v-LP)KHHMQ_lJ3DP!L~>0nbg?I}(;SpDq+$y~G*$%H-j`o?uYWu5yZz~O
zws757v-z8Tei(4n!%_E-?%VEbJmAP3b+7u#;*glVz`U$pus=A79us3cALMASJGeYq
z*k~_Kr~m2pug9J&PG{5Zf1vyR)wlNl#m|ey`NiM5?x)km&zC2DGJDtka{2E3$(#Py
z?w^<6z4X7Fbth-DlTQ0dLDZZ8M1qtPoXO+Kqz5j%c{2$oh#!O=O(=91^U2WjFE5BV
zJ;7%Te|`FN4}UP^*<mpU1^24^;a~qoARq$BO)t;Rjy5V#$i83?e?`S7MEeL$Ax}C+
zGZyxYSRJ9M+1E10PT-ZVh`te;q4*qDHC^}h>})naI6t2B&QSi9>s}r&7X5tcc=`DE
z_ls{17RQSovCtZx&!!8^@f>hM0ibTPJcp%Z0ofjaJ0w71JUo5cy`0Y`2mRu(d-?2n
z_qu<-xH7T@$jjrOe|p$3UG&q%g`w`!>R0{6>~cQoFP1=TP2crzPLIEuy>Ah!0McCf
z2>s{iX$fO#NDF4`MR7U(-eF5GW_e35S3+G1mR+s{8icm)qua;hR(NkdE3|xAo|UQ|
z2qGo>SS%~R{{_z#5hZfe?K_Nq(0wy|J?oyiFI00Vz#Yz1e?myH21fX<c<GKu)Z!6A
zHXKor;seu8AJT1Iol@`taLYM)20ErxJmtLRYJ+12rh~C0sY*(9D_5t0`K~>kDByto
zNj;q^`HGwmj^i0@K}zunE`UY8$hxaF5OQ+>AE`-TFd%6wR}&X+8G3Lw#ROD~pW4~)
zjwkckMRT{1e>{01G2O~NMt5(3SYfhmNI<u>mT(PL=_`272+d4BGmp@eSiI&jV}fb>
z6vm7h5fDbZMrh{jLm4wBk$9^XMra0vV@i*a)K>D??=l?t8_rg}PljV!%Wzchli^sd
zGn|zp%j`um2s#LP5X$z6Eg9Ncz_9{ejp0XcgjA#We-hH8V*u(G&`r7l+>CbN9l`(=
zdh!DK{6PSB>1-4&o<QseXQROISsx#7F@3PA`pVZF&>qSz`I;+yoyr6Hx?s&Hng&;c
zuw}YiYBQMP@bPkERbjxfBykhsZPEmW7aB7rp*pkrbA+a%G`Lm5JWgq_0-Rgb36a-O
zG=}@2e`wSj6y*UkdB99AgHShs=N%Y%LNcXQN*;WrUE(epGR_j~0}{m)k(rTx2BMe}
zfuq+)2QEe|OHMbymBoiFyF?TtfD2{W13H@ogsI**n~)?p(w(6z(ipaSOKV`3C^|Kb
z(WLC{9R9H;90*>kV>A(4N%XT}fhn!wSK|BNf0xh=e#HT{;s9H5xaUa2Jx8LgfJ_6j
z)CvJpRJ>uQ2$&QUVH(my0$D9sY`s2Phz5i@KRS@b1ZUq6u!y7)@=p0$BXF$nCSTu$
zXab+9t`JSTzS^BzloG-deB|v`sbG{S%B~MiZOn3K*2qZzz>C{g7gs>{VII4paE<Ob
ze>5U#jAjPeu62ZFwA_-%jETC}$F)oQO`67y1IIC%ntd++*K6~wa*Kww+#=;Zxkb}W
zZV|ppAQA^*xO0%<&OrwLEP4B@hDg=h>t>55f@Kk`E>+7wV3BD<`5QH2okS9us0ggz
z0KRw40LtE8{z(#);OYQNU|FE=$=6M{f6J6rYpJ!QoH-{rVtX5~8eiJG*63rYj$`%-
zv`mr(FmJfxmJk3j%>?4YEDvA=Ql27;Bvhn87=db)>@7NWDXcq!7~4k3orzT{2R!BN
zIK5!e;f8lG;1rqmQ{(4kcK%z#PY%e0AfbQ49P#$_tpD-tZ2s<eaaAf}srh%8f80{X
zlnuAEK6=T$Ni$pdC?ho4PV4$L^Crz0y#46mb1570I<)J(pgpf#d+31QI^ehF?QT1E
z^8MrNd~y2j^uPW5$5Z%o{uUd&Ji1meaNY9g?&OQW4=p(wqm4ngTQUm{dbtJ3vFz@e
z39uF0>)sXI()tGsS4TsZ*0Jy9Yc^<p;Jw@?&CT!SHfVm}z1$|v&F|%O6SNj@Ki(Rl
zNmx<*e4w?gLu<Ylw3eG+p4hLhwgJ+}p|Py%8vs_XRX_C+Gp`+=PW$;q_pF~@^!x8m
zFZSp2<0TaWSqGky+&*aeqXgl9T2%jvmr=Me69F@q!G#4B0W+7;o-HST#ahX(D<=@$
z->>K&c!YvtlAbJG``VGR%hTOOSw(qSWs&^<A_WxP4fNo)on+w&(?w038bA(z408DK
z?fm)n!%gGhgg4;S@cDfJ`a{w?NE)<eZ~($;7(V|Te)%e;vh+IB3#HVWUXH$d`ZT96
z8a&MaLofOl;Pk~I#Aj%Kj#l|uZSymlW~mE(0)4*6ae6*TD<g$>reIFLe*W|(nOP}<
zjbGzO`|JAo_Sf6Uy3)efFc@!zws!dK&$svof)Un*A!uPS41fPVe1H47r8=%9>5=?q
z<nZ?&Lo@P?l@}x_Pg-kCr95SYF@CVd3r&PsdH5lvA!Xr>UL-t!l8_22A%D1eSfpr3
z79Bc@M8TN+O5ZP;nlO5p-V*!O2(jPsicX+*g>o{ZB&sg+YT|d|-O!-1Jf`yRxH8^O
zU#u+Vu4ek|F6PWnS^iz|O|L7Bhpe4QIgJx7pfR9NdCOd-Pe=|Hmy2JmjEgJ!W!`y_
z$i%s@ltzGsiL9M}$N9`)oFBi#N8@t`6p7?^&J0#tfzFRRW><lQ%jom+Ryfnw$W2zm
zG@3@D(&Y1eDkBVp2lzA<N0_<lcIyK!2uK@uWC)AVMl-s}D{8eMj_eR~AtZK+BIWc{
z^0!O&#XWx`*`IO;Erka)SV?#z)wr*ay;q;UfRY%8{5bS~>|!P;0yn+27_Ka~41)w$
z+S9N$)?50tfkeryEdBFsM6f=f6jV0V_J}T)+{?3d(uQm>F`?{gh%zKRF;OVS5JN)~
zHr|avY1LK!%W9xR`is4kF+F-&42V&;%1+o=e078%KXDc3_S_~^JH+sE1XrZ!Xp@@Z
z_gjp#;G&{`+zCwdiG;p7FA5xwjs{yaN`!!IutgMJtDRume5Sy<5(XJ5(1QS9I}va1
z6kyZu0vnGzL<R}X!6)V_@l#z{MHrKT@|HG5H8R_u3P~L#Qo}dg79fU{$$5q{nQAri
z?Z?%$Lm3nV=*=RL?I=RpVVbiKe^=I_To4Xir|xTiQ3fmG36_Huj{LV-p<sW>XLk71
zG6)QvTSNr6^02YAQbO5(+gMr~L2@ohj;40$gdhBS1q60A6bPwlBQ9BT=H9HWHdUa&
zE0h{jg*px&)-PDX>!4ezDT#QUe!&I-1b@vcCCy=oxGVRjIs!UEDSf!I>fFrHN{NuZ
zO_f7`8QEgFpqmLojhT2!#?ajnvG<}c(kDneVZ&w!0p$hl(DtdLbpfeotW{1a^vt}M
z|7{a@ex_7QzF8@y>v2{D_>{eck&>H7w9S~V0_{PoI<2@(t!PO}R&ec8NrH2U@F$h>
zQa^H4baI<Z9rIR}!E)#_^P*(_bS+_CX|AV#Q7@e1#(y*A#sUfoXEv1G>_MN^l%d42
zSil>f0%)}!tge^Vs_>(D9T4gms5CXRHEP)wX_t}-jx<+2+fmVgwp}ApaHZ5sb?I0n
zN)GyU!Iehp(gvniJugMSo%D-P!=hi}%fOCTI>KnvJ$Yq)aJrJDXe**^o;20?UiXxL
z&rHP@TbNLGFMH+oKGHsj0KKdYcd?|M>@pUyYuut%0^{zisiww~1<q{JzH-?&nRhpn
z$i-3k+R3LfKBSkLb{x7n6{EI2kG)$5iZoMCW;5Mv(@U-bmZN^I1bFCvvfFOyVn*&-
zrlWf+#0FfHvR@f^nb{HI<3@ciGdAFVpv=G;6qPbx`5BeXG@EB9I4q!#;|w-TnQwNT
z6Hy86Iy0G%Suv*ac1o$xw*Jr=a>}r%zbuoRg@H>eKh*NuDTtQTobA^+nRkM+#lrkt
zagb1c?iy)S&Ef|Cnwu8XdQ1jXywSAx^8j#Ro8~2D4lv0+C1IcbZMNYW3zJrVW<l^)
ziI#=f1fW=M8fdjtIzO7xG87yy5f@AI>f^aGanEU}3=Zg8_EeI7(|4(&i~XGX!6-+H
zN#|EcaT?)qUy#uoX%{F8Q40Yi_ZF%YC7lZtM)QuIk9O={N)lJi>gxN_G@2ze_1d@}
zY8_CfZCiOVuRGygQ#H^%nY=iE6htf>B|O~oU<!88MH+g6&Af_M`$6AQISbfPzEt*_
z0SfE<V@hT7F3d0iMWa-xY<Y-rLYt{g=GpzQM!V98INENa{78gg+IACA<%ClGCx<&)
zqj1xK$qNp5T8ay$>X9<rIbi`>s!G1D#Lk6ryTI`ZTS=;<Vp395icT(n&&Qi>@yL=Y
zf`(etTbO2(Cv~dFS$FHe%2Y2~i`~j*d1cCPW7gbCeR-h02m+PsP2IWRA*#8Z*b1|J
zfG|E}tq4YULfxj~>$rH_hP>0Dgv`Ogt4zub%64tLxHh4h2tK^Pq@|_#T9cO6X6K-r
zHZ85nSK2g|qR!uIQ;H;iLShFsXdbeuwp0U2bFWQp5ERp2VN*incu@6Xn{sQnlV;PV
zNRTUoS(WPe>8+;F?D})>F;}}T9YgTA<#JeZ8LaaEQ?jluQ-wJ{pq<UwJjz|aAE2`3
zOLJ$nl6~?pI5rwx%JTel<{krFeZ~)LIkE$D#NIJG-hTS5n0b+ZAiRZ+7jri+_GP?~
z@zPo4wqDEzFKfrz`>Ebuo~NuQOve}Ic`e1YOFRd1JrC({x}n_UdgJVf7!JGHcO6?F
zkD#39G^+uuBt=eu)9vi1iu2May;KdgC!kIlPXW=qg1Pz@%2bP27oV1I-aLk;Y6DF(
zPQw3>lM2)J85MkgUBUvC<E<7Y_b<3Sz=fvzM!F3y>?#+p!)c|A0Q40X=1wl`zlVyH
zKPdf5P^<|qY!!Wd92YC%D@i7ovkfDUrI--B@LsnIFK3vg{B2g2591RAO7o{XN~>=0
zU)}kNHlA&y7u@-Z7v3PaW1)-BS~SOi&G~oT1p^@WlJKg3yI?(MV?0=7ZlyD+$%_a#
zYUbwmzU1sm$#Qq^a6c{`!b01QQ!(=6vD;tdRgUI>9$>scQH^}N>xP(*7a}MBxZma-
zD)2p$nqca`rsK?qF{$BD&d-}y(zc8#4JXsV^vA=kvctTTPFKgv@$4|?do`iG)iH>m
zTeh*P2a!8A&gqEPWNXcIb*D2Nsl&NT((3Z@!am-<(w%BnIwagrX;w!@S;uLhoaMdu
z>_*z`{{Sl*8(f!BxG@s~GdY(bnimt7q{=bKe}cRJ+v{HrvItux(?)=cgY#A(tyVIC
zdrcfIN@S4&c)CxNsnyb*raS735`ijxu&+JlXBrEnO(aV2P~K3l$2y+-`g~9H(zq<~
zr&CZ%Hr1encB+@$9?PiZna}_thR~0ar*&RV2`}}G*V8&L=Y-X9f!Cdtk>*D6@BYqN
zfB0PNoNQ5T7K=qscps`|*mi8uSl%ceXVE9%xT=+LL4dV0t%weKC73jKYZ{^}2+*#l
zkmr}ZrfMQf%dyJn3pc~kv*x5*_Nh9WX@>fKW|`U-N7`8XjU!2ynddt#WYDAkob$#s
zPbr2(bQDH(dFh^>Uk|^}Sr{Xt3B$Lye*?X+3WW1<P!3z&@W<bW?}slv)i$Z5>66X@
zOp=Or0svjvN}(0h&^4EYB+kWSzNYX-Q<~@82Zym!s^X*Mu}&s`dulthZ`J2%D{^ed
zO?)adYu;XvK{5v|d5n6>d6Q<I#-uqXp2wv*D%L!kpLwDr&P=SPr2LvZRu=D4f1YF1
zafVK^LXkgmzrYa~N(7{qi4W&Km*mde^VoAGkECDn(uZ-3o$v$Aa?iPQ1SK~(j=E4p
zHN7~R@{^j_WHrA;Y()EaRvn#gHMdxWap`?Oy2M~Va}WXsL9MK2qAP0;N6t0Nx?M3F
z>=~+^?ep`>wKVZNv4wfA{Ftxjf1a{xVTl<$+Ftt$^jOX6U=XFusXpHu;8;H9YM!{j
zeb(96*}*J0n;?xP6g=g?-v<oHmmx(Z2C9hxhJO#KStt!QV%R#Ryrb|n_;2_FJAV4-
zkMF;H`&9`X_Aly+iBup6h+%lkkAOwUT7*|W%d<n;8ex|b5o4JY5;_w8f8v3);oESq
zh|R3gy6N_lT8;1l<G2}2={s0$Mg~n2%aAB?EA#u=HnLPBi=<N(gPw>=8B32io=4Fm
z{n}^lilcrT4&R5vmts|L+EW+e5>isce@j+H-pV%ydK;UJ7tu%SvJ%+A;jAt;PP8p;
z8kKg|q^OS3Q}tt3Rbg&vf0X7XB*hfCiLZV886$LxVoUW<z!(}V2<4M|s$$t}-SPh0
zTN(BP2fZ}cOy*9RC=^n<14o7360KCeab}llPU+6QIdiEJRlY$inG;vBHo7_@$nwk3
zh?G>j8+H3~sX1^`0PF?k=ob`?84&dtOdo89Go_a`Kq1jnsu><Qe*xwsi*wwSFc*9e
z*TQ{H^^dhbgEv3v@9I+>Q3}JU0e!qsIpxHhDJ}u-^2WZqWHFcvWmLVcX?gaPXsxDY
zi>^*tY&x2-TS>#MB#Nb8(v!+jg=GzlD3E(i2dimOqkAddksUK4&T>D95TN%J-JF@E
zTt6GN+(hZDm>lcrf9NWReb!W|u2&qD0W^&Gif9s=HF~U`rl8FPpjAbjXIFvJ2qi<$
z#Wv8pXi}!6MFgn_PNQPv^-$j~t^8Wlmr=0Bgs(APQ})NgPDy{E_3O*|wQq<bsJZ-?
zVT|{Xw=r56r+ezW(pt%QV|H1{cw_p?8Sfv%v{wNocMU38e@oSjVI&q8Z>pn|MGeE@
znnA*C%fzVh;`>!15*HTh+3GS86LQMR^dHDP$&;3^Whoiu+EUU6Ev$rbf1|9dPviPP
zt8QM;EKUg<nb=d*7c5Lm5=E^^cK$D^G->(T>|(_rLNGnI&WjLLdK{_<E674|($Y{&
zQEU86esi_we^-i@fS>9qQWnWe@2j=e4huq@+gNs5UqR6;q2<gEl-)oPgvI#oDL!qW
z7=>4}nK`q+se?g;I=6bWqj9y}(Fa5fd)nQ|m|}cCyBk_mz@B!;;({^aF0?xW`m{US
zXB64)7v13V#$v8Cvq+cE%Tj>G<(OW|YjvLptKaxvf8sr9X|oJxNsJO}NBD;MAjV{g
zqL-25jydnGB<S^=_vtxvC>n)VCl_Ys3b_ncafMnsv{tCQunGa8($eNhs<(bS9~%Ye
z(0D*YyxxZGd`w7nk@-xB*W1|fdA|_8TLqO94)hgOa3n8lWtp^my`WM~x$5K1k*b^(
z!TK4+f7|mDF06$C*Hhe(pKv`&dpt9Wx92AnRQkr&Q*^icA^Stxfi(ScZAN*!=NYV4
z-tUROI*K2RzdCL^;ICf(?f9#sm{xIHFqjh&W}44p?fv}FBc`6>aapT^hI*~G;2RyC
z@VH#tPbY&Bq^-G7Cq_ekG;GEWSXcbqRA-<Af37ppxJw}^(b-fPwMWo9OkCji4EiZW
zc)S_v7{F2-$5Kzn5#;!Bl(${#w#;!1&HQP$ez@~p&rm+!J@R%>d4^`Q=Xf46$y1gl
zU7OF2PBO>-umQVGyI)xL3R9qrdxhm=?*>3dAQgRvd?=a<VD!6Is~=Kr8^UJ`&J7aj
ze;~%`yh0$@TOf8YT2<%~qcrCq%7ppm^vT35bONQwGeO|`zlXq$q^L9>j=(8HD``I*
zf$IQ5S@_ctxOPTpy=O}7D1I;w>bUKIgL?V5<Dib>4*O8OjNA92dOO1AeW;$|_I;?H
z;`%;RZ}mO)q4u*oBd=bk?fX#s*`4>He=f5-??d$&Wv_jxem>Xtp%(JmhJ7dvq}%_&
z*=<lrTP4unH-j_S4V+<87Y4S1vui$I5^$2o%^*&zC`?ovK^**#fH*8XbRgx!!Ig?$
zSa1JyaHX&m0=ON)RY&oI!Bxj?2XNKPza3n46n6kuy^PzzRd2`bANhKU+rd>&e{mgL
z^;X{lT<vFf23Pypox#<9c4u&PncW#&^}U~~-EY~X%lmFU&C|^GJ-AxPYujO-_X6B(
z&d$NS;L?%h1xYr*Jo=mi(YJ@v(@7HsWH$r0e7?V8wq-bx;6i8*x2)m53PR45XIIbf
zFC#!KckN#csA=)9e}}Q9$OG0ff1ILp#<^l=O3ov$W0YrQt;2e}ovlJ|;7&*Rd|crU
zXSuuW+>H`muP}FKnfvX`wG`f|33s?q^&Pll4h~YUa6hkm*vy>(j5QPH?n3503|u}0
zhJkA6e7C`zMW^RoL|wCJOK(<A{nfvSvBKzzQGroWL2>zyqo$&D*XP&627dru9CGNF
z5ezaE0Wg<w)&&y*GMCY)F(;R>)G?g_bGJ^_G1LQpAM~^=<Tl!MF3XR(7B2XVa+^Ww
zfsC$m0>E9G^O+I}Pc>9$GzZ=plxbUENps*hlZ47zS*A4yj_OycI-1r&bH~waugK8M
z8RJXcIzby<^ZNC-PqB-d?qaA5)Lnl0RIf0gJd~d>+Um*6^Xht&vWkIKW%Qgo7qk$Q
zqhIuYOE4mb+(Sf&$C56<_c9(aPOOBi%UP-k&Mb(7GLlkd21Y=D2n1Jyf%2u<&qw?9
z2{W?(h<LuV`}rU+$xEG65zm*d;j@kk)*yirj?p}RJw#9wd}BaRGh4v}%}_rI4m2{a
zqk%^5%fUcRXax)Oj`40#V3So75~v$r4+xxpKhr7#_FVvgYC`rLf{t0&Ir-z|1d(#N
zYGZy53Q#$$c<ZhgtcWLkb3MHnV4(4edhwAdHmb!qD&M+TI*wAp8Oc8-VIW<XZ;#>2
zA}@;wxuS&#tVUn0h6Rt5nuKXu+PIYdkd~9;J;tz5aam_(vgW5jnsA14Xr7nVn{|PI
ziDeKwROv(NNK%F1)cc@7f}q?mWl9aYs<kAg|MpsvTKE$C*ctv3i6kXl1bKl(cx{QI
zhe?ZcR@LW~xXc_aW=5+)m0tWOizrV8{=~<DKZ4~{CcHiPlUh<r*KZE~BvQd>-XHu)
z|A9zf5&YEz-x&PW%vQi(Gt`d)e~rw4>)@}E`*QGC6Iua(y<@x^_}gUF1peyAKLPk_
zq{rRhuOYY}{51rZ!C&Jv#Rv8+^C&Bs6K<QW6-5Gc;eMq>9`u_PV4G4940F1Rq~mV*
zCxL9u|A&7Uf-O$df!YuM@M`c+I1?tQo$#*?vO)9HZa+Y?<bgY2qOeq}iM=p?5q}bx
z=qFuQKAyfAd=nZ8b?Niq+ZACP&De}vdqo`AP<&$`*Kk`wa_wk83d*%ouVcAZ^2@<o
zLum!gP58&V!MSa=O?a+pon=TnKqG)l_PoG)ez8C*x50X6Q5Bp4^ggj;kxM8hb?5DS
z0HCl;i{2tL2vGVxNuAG*_s?E`0FnnxyFGH!m(=}{lU5<GsL!|zxly0P8soe=M`4(O
z-^8ZpbFeK<56X1vp4o_-!3!C_EM3)EQ(T@k({A#b{DSF$poovCs(k{nscK8fKpihj
z<rHL!yM_t+ooA=%OfC%ruhMN&J)Zpmctm*P_up}k>wo?gf}Gss(nRWi(Y`U<Q)Gla
z7NyRZM!?dP?P0?Ptfg8d&uBA<a;RFwY}MkyTJ&*SC?vd^x0LO{rGyiGP6y5el${<=
z<60zDCW>e1D9UBGqHsQUaE=L-3*2#*yWY>;V<D#&xa(Q&W<PgBv%kcOJ8tA|_W+VW
zZNGD;IY^F{FrH%FwuO7U_&d2<Wsp&@NK20?(^?(=Pr^l%ZR$-)0JB>oeT_5n)s*Hi
zB15Go{%NxO_ch>zY&cx<_qDWO=i@iG#cFfr+F#c!rP9$x@tyx%>jfJ2j<qt)?faIY
zeyn7=@Ly|9S^4yO^B-KYbXb=W3^Eh}F_&?_2NRd5-7&xcaJOmRG3W{e=!rs^mk|sy
z6qk`7Iu)1k&pjuXrRg!40m7F)>M`ekN<ZPY6z7ORfm_ymE{zt8OUY1h3JjhX7v??m
zk>|&}5DBHYxVTE0n?#X^Xj6GSN`UlMX;$9`eOeh<>>n*Nl>ry>7B1#&Wx;u^0i*y?
z!!bwNmy;BcOSY1DkpwTMiei)`5HGL<CY?rSuC0}#qH-bBwX!d6k|dXlWs1gs5|U$l
zmX)_$K<7iW6j1h7$!|qNI_B^^rE(KBtrBCgB(qf#t_{T-T)3g+R$P_}nMP&MHM`FW
zY%d#SIn5~Nmf}X@KQT_lZRr}IpaR-@_wp6iQyJ{wq#&@IcC4JX=OubWSu`SpftuKC
zRb1mLue`^GDs84XxU-^oyn|wYx{GUyJnypDmQic;)4XdeDM!=W^ALA$b5RNWDb*Pt
zHM&ht;A`gXP?(tf(es|Hyv^ucX~d}|)eF`F5*<Qw>a?Ce#_rM2^5<p#B^{@yqQK^L
z-eZi)`Jvo@L?0$4KGb}cidV7UboUrXL%(?36j8vWOhf)sK;^cuHq#-0S1ZV7I%~K$
zwP>t>8ZLZ_&wP7myxe-N`;O~*$6L{G>!8L-Yr?lQakV2ahc7ZHP0*aGicPt736zI0
zHi$!z-_j^t?Giw`54j*7*d<gHOZ=QlN0rxZ1aCu}aWsGV=yjc}oEOxho3wZ_s!ROa
zlmp~PNjQH`{X$7QLCzC@IR_y5n$UK%(AXiSbvTS&hmrHfohLHd5hM)#hyX2F?w{tO
zTcIsVb1wE4@S{SY^I{TF|0u&))JLn{!%qO8M!v$2AIHy0(78!J6lWth@|$n8#gXV`
zS*{3g<!yt8EX~|6m*7un{Am*XrFwFSZaV`I`wU&27)9(Ls;=vQa1-d}g*fi_A4><i
z?{6ANA>AL(<(`M58Br~$o(07q?XDhM@mxxTXg}JD0^KKkFK8?OjlwU0NN1GxR0;ol
zvIQbT*fW039yFskW2-RYfVC(Q1v4Y@Cr#AQ3V9d;>9*jwX51dWn6Do)gez@cCHS==
zTpoy3r|uf8AOuB!?5+=1`h5C;;fgf6y-rg(@J1Kn1@}z{G0Bx1W-%Vb*i;L6<17XR
zg)q=Z4PyX=w^Z{g%_Q?3nopU<#NlYr;elONP4Q!9F%_39vzU7K$IfCZZdYb8)$Rut
zztu=Y`We?4>yFt8qBIDSJvGJS9TXF+sVOeN8Im+Ztaqn>#tgR2C8X_A5E{2#nM<Gz
zN?bc5$!h-6Y)a=h%l`tf3(_zmG|8zS4%>Rb9?T;PN=5PigKZ!ZX@oKAv<{(<z5=H7
z8NZUWFI(Xmd=5N|WlOu;;n|Y`P0dDlJgx#HWei`zRSSag&4PUv05ib}=zk#ol4Yf7
z3;d$;FM@}E9E;^b;hazL63VE?=HqU-6R*K7!>BCLqu^F6?1&2|@-7Iw&=)c&<7#0)
z30@f^rA@Hi=fEo?bnP&nxXt&qEpu&!GxwyHd}AM%hSF2X%yFikiM8_m1Dr@sRxaE~
zWtKf<0^yQNZ@QQYj++ubn~i6bP};M*0G_vs!YKVl%Qz+MFOr0f3a-1XTmFAL{Vpiy
z;tJ<$i_Xwvzhg*-Jmx4A&a>B6|H&XtHs*HD=J_B4q|l^e)&JV2gD!7Lt^WWhd1sfQ
zoCp&FI5d~By*d*DGC7yg$1f+BGVn2-e|S3n^W*Oa$;`ff94d2w!q_lbFhWb1zWg}+
zInl)TDa}4efx(;Ue^VT$MRXEt9X(Mi6{f#v%5Q&s`}(&p|5Ab${5P(UlNMl9@Y9d{
z)o7aTl%KxycPnf#1}7x~@vA4}g$aQsl!bNR%XFZPghqg7+V?n$!g*<$xmp=vf4v5r
zH)w_m4~b$KE&F|M2Mbs87U@itjrc86;kEHAtuL$M*Z66_m#MVsKhxpsbof;93`qac
z(>iI_;G|8JBduv5EkkI9Hl!GtdzxTLkgJ2;F?kH=v{pxmK!o62nCe#bq-CTzZ5bj`
zg+duB&vAU?^us|j9$6_3Np#Xie*pTZk%}SBu+D#f`N7M+(W3X}naMmT3_=m52Le}+
zD2P<DQ<CVKo$_p+%w~#DM>O=*WJZQe&5)Vt6F;+Am3`ODtEWi}e3}~m(m5aIgMQkJ
zjiTct?7*h*w1Nqe#p4XhhVv>1&`zKa6%}$k`k9f44oB*UQJ`<0u_5<le;m(%I2LCm
zB<mVe*2LdUr9wz+T4@5K&eXJz<7HjihewlKY)4<xhn%W@;~A2ls^Kb2kavZbOp_Lo
zJFc|+QMtI=Gp{Q&Y<sf&J-TVF!ee&LiCdrDF6qDbKw^pBz$5u3GW3<v&+jA$uh!(C
zK?$>zLqZvvbLr-6jEshPf1Y+TSF9OZk|)6kMb<m+xR6)EI*)*Qtn?Qe#QIj~6f_&I
zqqG>wt#CCETQ{iLI<i6iKJ~LW<~=k)@0?#KA(G;_wiO6qhoGNW83z?3h1>(PqLX+R
zV;c%(nfS9b!{T+VP~<wacm3il@M0FG3MUfW<ierwIt+x=%khEMe<zzHX&bZ0#*KaC
z(Py?_*NsM5zaBAJx@Wu1lr^)sepVnBl^TFbrk2n)3%9&=R?=DlPK}@yPvZ(d<qSoy
z(+Qt?j@vV5$d4+Wl-5956E)+FX4zyd$fIuQkrFm<YzL=GmpMxxm{DB#u^xF}*YljT
zv+v}w4c<b-oDv{Se?cqakXEGKh@RH$Ep6j16J1?@Ej!E}B>Pr5oTq3~T9tV890Af!
z6?t2~g+WTPu6~;1R8OJonsm6X?Ra%{?$fyArfyWemptVNVUo|p8okcyJZ&nR#S#50
zjayqMs0dk5iI{|nR6*WXR9!(Now#b>0+pn}>sn(1MY-9;e`00hnvju#z)9gVG~qqf
zKHJ9B($EZJf>LK+9LI(LvhcMU1<G)%v+Y)A9-i5}E_hi>MMy9sh3IB=K6Q2|=Tt&@
z5Jurh1NL(Pl6*ZZxLCMsSzQFObZjrJoN_IkP8!*hm?Wnti?o%>@{3RU<s1e`at!p-
z0!Lc;oN>U{f4JZA!OB!O%lKE+p;QgiJZ1C#c{I+Bqsd|Ute5kz@fU5_Eqk6J+Zhzl
z_^)}V7@1TZ&%DIH<m+;kdP^%;O$e1hQ`06Dc<&P<bUtS>l(P`;HCmcE=`wyUX!hnR
z5MpUJY0^Q5uJWW87I(>p1`1U(+-y%Vi>W1msXT7-e-?-$fidz%wq?=nNH3%psbkQu
z$1v-DLX88udF2dHW!Gutd7HCjb4%os4<r|?L9>fZr<JpPX&8$np4(U)&*?BMWlNpk
zAX!rl(-cT-58A->_NV~_gvIHKRmJ_Xt)%kcsV}1pmut~-<!NDe(T>k!8+9DLSx7#0
z&2(F=fA&Flp_5djJ1WSEY)V>kDiNooXNha5k;<^&<G9tNKzLM-ZC%s^tD$yGt}t)%
z#sYffzQg$pAu5)lkTx)1T%mkk!2iN9pDrZHUVvKAEm;k;nHGnxdIH;Kvi0}oT>A34
z(9-^hTjwe+e(QX*+s!%I?2<d@xsfmF*il?!f4Y%W6sKr3l~A?4o0_**I>nwsXV9m)
z2Xe^~I^y0fsHG$Wu&r9MTux0bV&+VIPPTlJb*O1s8EwP<Ol_6tqT_z-Ayl&|eU{<^
z+x`tZpnG&uFiO{l5g*u*u;kuEZnalyc5})Ar)==saQ5nOK6Y!A!XS(p-V&bk>QJaD
ze-K^OeCGz0ilvI`xA}$LDUgAjz8&@pW08=(lILor1n;*0kV_6bxb8I_EY?igL-kv{
z0x{9<rLaLl25j}0bD*(sFlZ64RZ8e7xD|}-3NAsOyKIX7(;0V*cJGr~()1O<Z638S
zW(XcTn_bs-%?$h;vzf7dTDGZO1}LhIf6P9;*`ikvjs_pkb;25bq0A^M2R+`0Z?7MR
zKNnYVWOY>4*w~efKcssp7bfR~qgv;$e@|ZzpGK;z72`8?HIqD~wZ`<SXDQ$H6Dl$!
z=YGy)V;~-KDmqI9kkSY$ZEwi)$?L?6jPk!^Ufsy&CIovc6pWjtGVi6cnrM;ne|AJB
z=az|fBcH2fEdh%*3Vz&Wy+Bxt`wNNyQYf<k$-FWExAO|138FAko(q<J{GlT&z>$iT
zH6D-RhTw!YVP^(&m53ciuR88<Zy9@)oYb((N^|8Bb}D!(_;94Og&Qfb8BMU7*2WrV
zgqib~(z-BuZc>c-No%;<27a|{e~PvZ9Nk$qfFR;6|F8@C=1vxB`|P4_Q}(A8EZ@WQ
z$a`zO5Dzw$uI9F_e%r^YM-iMFDb|$uCwTQ#WPu$i-=OX&jR1`!r8QT<aR;oXbg5>y
zT#~JawM)4zK<|>9tKpd5fon@aaJbq2&pnqs-*Edo&+p8pxi&G?_1BrYf9^OG!n|{)
zd#}73Y<AKE27Ar?0>VgXwo`7Tp02u6=XQJ%fX`_8dW4hMgusjtcT9Nh4r#Tg(J$vO
zAem}^L3Oulct=S-4UK3QO*oaaa$ZyV$$`JyrvJLH=W5VWA`dGP*wg<7Md$M77u{K>
zysPGp*WRP_t9RyxC9It5f6S%|;F{(v#R5=jWN5toc8_xpWXSf)eJZze`P)i6H&f~A
z)C;$^eeQI8NRG;G?ATFp``kcKY|sY%db6}pZyTa!wIP~0H=8Y1<p^n*yNT`FH8^(`
ze@&-U0V%B?dHCs0DHxaXlJRV-ozlK1ai319qW#fro$Hr&N|CZ$e+WxV&tJbPN|C^;
zTir&t`fSIsxQeSTM=hZl#$7hMr(L`Mk_OkS>+z~}zVZ%@Rr%(<h~hu$`sM=S#(r+{
z8#nZEFG*K}K)m&Gy<dGz&!$|i4;1p{x@>TQAg}g4WmQ3uX}qoDoszwvHqyrCoryvx
zjFel6WJ(Lo@a^1`f2d|!rd!p=sd^8U5-lIvL~CQ;9TBAL%x}o>I^CV_T*6k7jT*V&
zV>OPT6zJ=v_!w!mJKu_Q@r7GGQn|w=H2FYbq!}q+vKU46GZjxaT1>b4-hF`0lM?SJ
zVcb1gn|A!3Th@p^y+;VG2g6b8hLO|!p4hxPOM^#OQ`x#tf7YtI-5uE={Nm7es<q-(
zM*If@Bc!p`GCnt*Q$kQ(Irx>@BXSV0?2BP%Ko_q!SC~^*0*7llto3WD;^6GvxspIn
z-@ez7PfxU*?**}KmME;$Bdrv`_58vcr=aJ9xDvrGxVrEb14t)~S>1ko^V>5^weY>3
zMd#zc)3}yJf3L+lLTDS4Gk0SXO%T@U6*eC*;N$T6;fB#TOGXwp<?6eh$abnprWcmt
zNhzyl>oBc#xcXW`*1+x`3><CbKG{hUZhot+SZs1N&Rz2!Tfot6*;Pvk@lqFWtAM)K
zC#zrImp`_--kdx3ncsY1t~1LUx(!#l?&($>$Pnq82silu50uC;wwDnMG86$emqE`s
z6qkJFG$((x8p&=PyYW6>(LXRWi#q{f5Y5OEkedzNf*b-Pb`Exs_j1VZr;5$u*2`!{
zvVfjevso)^W3lD*W0KR$;rx2|^sLR~MX=!2^!j}Qd?E1{Bn~<|c>v)wOs~JDzdcB)
z9A0O<jFejA<?;W3cY8W&2si^QUicQ^bo2%Y2IqfhHN$E*!)TnPE_esLKhikDN9iPC
zc+42ir@z1c^&m5|pI;A!Ii-bladO%U2X6ZI>+r{f6JMq@`yhpm!A}31Xc&P|66ZYL
zwE>9JpE%{ye|~)b%eQ}kK!E=egq(~}7Gjuw<wqkxD`Tgh`Pm5*q>mGbi2Rz#1fhe$
z2@HQ&2EI)PBqWShs+snaMz!!k+Gef+lpz?UD}iQ+c}N(`Xr}kG9V}cG5{c=;20am$
z3`>s)&rtM8SNqIU5$a#l;rn#>QuG-${^8wu>DJVVM8TugG%DpG<%BZIxtV*M;BX;V
z2Rku&B1-EFssNOT(WlkOq>7|G(n!lgh>CwP@KE7v7~hb75OmhZY?M~mIH5(NbC5+s
zR_UzP`v13I8SDoFy)^e3%@K4IfS8V`T;UR`<fpizYkrz%=gHYk_2Wrfyc={Qi>_wT
z&CDg=ACp@#-8J{>aUug>riQ;H=dlX<<*mBG74lxcK{w3PicXA2zllXD8^MbO$g_Vr
zcqI)`pz`=+<>^S_$zo`-7?!N(H^ybn(LKy}`OE>%UNTH5BQ)>Kqxk<bUOs90({q3|
z4wccusW5WL)QAIlp6`U!`1zf@WUT9Qs}O{9ORRQ=HZk*PPf8_fMHjkCoak3u$ug!9
zUnKT~X;(t+ya0SP@{%PY?1Y7=7HWTCwpOD?$#<nTS_|zKlrRBprH{cp)(yN`>uliv
zp88oi2OXOvwCAUsvcO}Y(ZwW5<a1WO%nR{cM8-(#h=01h3{NP~AB&7=q!K1cMB^zt
zV2o5!3#G&K=Rc?Khc6@6tRVEi<b?ENB}MgN<ZCV%qC^3ubJq3(I$=?E8!>+{pbheR
zKQ%sanFUq-%2W#Jze<=#zP>bp#0jnANQG3FEXn0~r#cGRMwdJXNJY-bGy92}z0fxP
zV^k9te2ooG!o#M&7J42n)w;<B%<HoH1lxaxN(Mo6ViUo8IxWN{N30Y(ve_6?9seYa
zb|Z~Ol?{UGkbx&9I}MiE&WV3$^<wrLy!sB_w2WxYVWc*Ac9~6;k5vop+H8Rrj?7uR
z*H-o1@NP@wB=ye=K_T=il1mo)x`}T(78?!T!4`|R4I$<d-emO@4M6(=;cr?(EM76*
z!9Z+d1(k_dT8^ULmojJpd*_Q-|9dl8rvyZGhx4ssvw)#huVU*Qa~pqykwSU%_7b3A
zx@8>-H5G+!7UF4jAt0Jmo(!{T*Js)2)KVgkxeaiwAtZ~<P#@s~9QJWns-y381*O=k
z3B}UpgntsHZshAuO2}!<MBI((k`k0|(*lR#lpL0+hhxJdmrYb2@+g?{URD?8@{v}V
z^>ssDzLu1YH^Qo$a&mv#F9Yp|c(J;XJr(A~>WK~-T(!C!>3YDVWW4le;`VOCAGnc&
zxVIpjeWOQ+%X&Y;AOe<z1Hz}f5psYZ5Zd*an_NnbO0T8H)9o=|p`D;j?5ey@HV8gi
zi~gLcdOaIm`bku~k&Y^De#V6+FQc#}H>5YX-&cMq{ms;K>RNxHZnvppDspj@ow%jz
zeQOZ3h#p4N8+jM^jY^HoQ)no$`+7uO#2iOk5~${fm#mqwL_cV@oH@&<cv_@aewZ0@
zCOT&*m!ET?ma;aDf1dk|DO=fgQ%ky?IZAinD;0N6ZU=B!Ef~uDq4-<4@C`~ds63A@
zWYI!o^`WcEV=jNGfm+K{W&4f>iK|sDp|r2Nourn_`#DF?a>m@jxotV8s9$}!eN%ml
z+~t0L*<3YdxX_n6b7{fro=Z<IK`SSwq;%pv#uRIre_5f55uW#rlPbUW*bYh(03$j?
z;gws7ghvR{ZHK43h{+(l^CPiB5OY#PaFytKQvwSWP3?bbEDVC9DkP#i%q*?FsddR_
z3r73Sr>v`;yR;InwqMFlDwS88ha}oKq^@q;O3hw4melL8DKNF*$zd<KUlev5afM0D
zk17^(P`L;s1ErIk#LT-GKb3Ck4)`kL8{JHYb)@wJ<B!jdyO!FYz?z2^VclBZ+Bs7Z
z>ssv=tuucmBA6?kv*Dh)Wc%9-6|w39wY9N&%R;PZa-F@+E;&o?K!7P*9j#BsJhc7b
zee*I(1S~cOEFUD%rfn!IWQ37w)0f6gwGBa0VMGUjngiKlbSGjcZP0ZVvR!XT4AEml
z29KL-X*Ogys&sfdH-qQi`#i$iMe5$?IoJ7VU>ko8GrS$KmD3EaXn@`-p4(Isrt+nD
z-a41-w9}?H?#Q;^>0H<{mf;XfJI02^dB#WC1{j~(HUmYj3Jtf9@;=-8zDf07+fiOP
zZAa=Q_<elVR2oDrCO#l(Dru34dvd0R+9tECZ=cWie9<F3G^Lj+s=bycAyA9kd{mxX
zGJAjGYO?2dgV~Ms$9)0#k=fss8vz^^d49}|)(Vd?_@i?pAl};JvrUnloGBmG&PDm#
zOkq@QF28-+>~&!zZ(hJ)dI9LVx_j7e?~C)%(?AKHc6zdNBWr-rzJxUHrE-_B8(C>M
z4Va`Wn-)G)sHc%x-yE*<#wX@^!0XcQ<5_<%qa>_lOs}%(H#A;Vx&$4@`?cItTG28E
zLhh-e-sXm9JF?eh<(28Fs9o*ce+0^Ig3c+M#Iw~4zo#UmsG|DsCicyrzYE}llU#M=
zuM3*2-h4mM8CZvf=bQGMgZVJ#q+s#o0*`V7ox=pzC;6)m;`CCn6>6&?q(IegCKyTb
zxt<PNU4rW3pg)M%?6%sBu@TWSL9?HwJfMx60sVbpa6iKsG(11b@f!;&<><q;QL>HI
zc~gxPF-ICdARLY!o6kf34}_IHm(j}`69G7vQ0p%hm-7TNzX5!=vjs9)1qA#+5m=WI
z3^Ej#aWMfCmp2PCHGcrtj-A+%P$j5ZB<5_06(;_F#f4OY;bQ50dcF`mf#5PwpXWA=
zYj$A4T+9lz2XJgC-c&1Aub{@r;?Q1jI}(XZsl-$z$~k1yLDaTQo$l@0fd!MR+8>Y^
zy)tMjC{%V}%3qk4&=NXza0+>d!}}cZEj@Rq_QaN?UdLk2-6rQxnf;^VDhh_4+0t((
ziK0{lq#`HTD{)ldptAfymNvWr)4o9ymk|sy6_>%_Iuo}!3^IKP2(Lr*zqAGGFqcud
zF%tqaIF~WKK_i#RCnpR7G%%NebO<Pa+in|25`FhqG?16<gL=BUy803X1{^!K6DU~2
z@&eg4gat(n9C>BZAV-0-U%#h%G$LtA5ifF<25hKirutT0b?Q_%GiPs{@g-yvyvdxh
z$pL?JHr_i3g$?_tZ^Uq@LNrp6e!}<KsF2A?jCed1N{{dcf)pb`f+6Qfn<YqpfmIz6
zrGm!J8S>~D@1V!7N@6H*R;d_@sY=mMqNpvPm&HHIfzY;yPS<%zvGx}+YhslWq)=NB
zq+Io|CT6Tw*h5NQ-?8z5u;M`YDoI;-F+S!x2HGZ_OCYo;PFtXRCS8X%V_#d4+QwDA
zXjlSiFV4W4s^`hT%=Mjz6ze;G^irH)43Az)NZK|*sxxIa7*S^x9QK8)p%bL2Up@9J
z)VcU#LKKyt68g~?8{vYegpG13V+>A4NViiBD->%BY>#(32CN~ytpeVuM`YbM3H@Bi
zh=}%#-IbVirdYo8D+<snqU*qr5!;~eqKWLaEtVBSR6QXRrD!jhZj4-i-!Y_Ad(I{%
z(m6vpX3?(^ZHwq=A|^Q(For~v`kCTG>YWuXkgzzFib+ZID@>enoqY@@l{(XiUP`rn
zgo;yN3+mXe+CIWKE*FfUZsEOt)ve_^Q#CuQN>De8PJJ!mF;zcE!IZu~1VO1j9^eAS
z<@yd1c^w1OEj}{F$j%ght~Sm1su7h1uR=9iN|3bofcNn5Diu?rss-#$=}m>*F1<75
z8f3nT86t)B==<-RXKxYD1>ya=`SqXw!uw=1tPq{I8G}!*E-z0XKmHR<$=S+^rfHr|
zFQ>DUkE3yGSR0t;)u`***~Iw%@yAcyi<55DwTAi^&!>|PV?1Ynoh7EKqdW(8kRW2}
z1B?J~=l%ivN#yHf9zSkg&8Fj%wtL&WdiK0|(|+o%XZ+@mk1a-i*J47GuAOvqg_wQ`
zuiN?bYBp}?J?iMIU)uAF(GSy4Zyg>Zyai5iiuN<K#CDf{R!=6AY3-<RKm4P&758)*
z755;V-i%(gP4mNlbavj(YRB%hdD;BbJjDZ6bc%_LySL=M6?i#<$TFOP#M^?6mPj@y
zfBL97xjO5#!{dv|f0`#Z-H*GA>7+Sn{_*-J{r~N6UH5VRUDLd~=zhC8v*YQ9=Ka-&
zk4NY2@6FFwzr3>VFPpQ==~)xoagLPE018g>84z(a9=E=Kl=JhkJIfwhDl{h7%xB|j
zzR|^`oz0tP?PT6Q`E)UVGMkP5KzH99d-g}iKFZ%<izUNvxpYDV56dPjTO1wGvf<3L
z&(|G~jyWwBt?`lbO?+S%6ENko09sPG37`N2c-}<QXRA<gzb2ti7;_U%^eIGNP573)
z7QQ9!5WY%(D{jKK|JL)0_3&MB!rL91`iJV!;1L$d0ZGzXFa*QQmMhP37DVB+d1BE0
zN%La*X4*V6-v*EeMs{+pBqST8>Hw$OJ&+<>4~8+sQY<!6I-pScFdEOMHGtSV4WYOi
zLnSRaX@rx;&jRVe9P}=qTyfnk9W*%k+(DH;-tVA)!If19rG<kkk=rM&U|p7!6%Y8z
zVxyA-a*FXF_8CQvj$>JI3nle`3a!fuH!LQLErmDHOl*}A-mgg%Eoy*GG#%UI)Bg=x
zo97^yr2Msi2>OtA@o8G)Oj?KR;L~0fK27}sYuR)y@F0XikU`juS21WTxT$!u1d`71
z`vk^+uF#xf*vtQazPh{^JsWkS=FRBrvYo?m_ai_aNbTGLxP20kphH6LZdm9y#{|k3
z@@)&=GFwstr$Lq^iMH0+(MCNFh(IV$YsnH-3qlG^#`)wJ4u>7SC0lNtveiGrFbNdr
z_;dFp1l6#QD1?JcrHC&n?>`6Bl-8lHq=3PH^`J!Qg2t?{jWMoGLJM}EljuK2e8l4R
zq76id4aDnBj=+0Q2b&^5zA3C`2+H>mhjpzUs6Dowz|c0(45-JzqVLxvEuiBjqZeoO
z0O3pXo0XKYI{^<I)^Y*+U2;KK;x~O->W{6ig>o&h+n<t?0Aw`Fdu%ia8lu-m1~=w^
zU@6J2D6@HR1uix(5wHN5<AWk#lXl*<=l{NdmyNn<A7esC$B7;S<WZC8xgn$2WSBn#
z(zGy&7Sv+BAflwms@dO{C?aLQf`|&hAn+owf|qd!2W0%;G2Q9x;=kB{kF)9d)wug?
zBjV~tD3UPW!-tm?2B-lSv_;}3n&hp2o+aL|N!Uyz!A&%^6!r4eAf321qzic$q?52f
zIv!YW7+7x@SZ^3uZx~o_h=Y*!2f^T-6@(p#i`CnQkh4vQM$jz;<t<bTY}t9|OBO^7
z0>)?ApXP8#R*E-4lwM#KF@cH$A@!uw@n1yvi1kz?7;KeI;sFV8QMfzc2sYY(*%q5)
z5*Nlmi+tdmb&(G8l!me%?~;3fJm4epZ_-;gls3sQxw;D;Wm)ik$CI&b!vH9xS9sTR
z5QhJSeCHfUgeXhQStO?DLYDLZQ4sZvP)Ro-8We7klHd-gQ-hPO4545f$|lS=tY6+B
ztgr)q6HRQLV%8S`@0BBqyO4Z;Tkwv44ZNe<1-xTgfVT|5ECVphKwM=Yu2S|os^bI~
zD2uBQP6_EX9waiL8Uiwu2SEmAo55qe=yil1et@y-{X7DN=&wNr0x9>3PD;*$Rteea
zC3GzhOIn8p@(yjd=nXBWQe4P;B?aU*)S*j6mFr>1!PelEu>Jl)@`AU231u}fh7oTb
z$x(kjkpyft-;#`D_OO!=iLxU-DVG&EjC2n*UQ)QhH#lAlX%kK5PbF@k=_zTIw~3|;
zHYNFL&^7TIx+dBMT~l13YaAdn4z+)B`(Y_-;4b2c1B#jw4!r6{trz40f{F<~1x~94
z8dx+qCKxuNiwdkEx4>C{c14&ZvY;l3<Y?a$#i-Th(yhMUu}J)i>TX7U3vRXUPJR=v
zwW4*68=h-g*+3J#IMbI$gS&2#wnl^X>!Gy8_0W<n?)mw|vck<t1(JnZKhxMqb04#=
zSaQk6+Zyls$4AS?>witOEtKZ6fYLl1F67~Gp`YB=P>NE(7AmlBT|$BC4cx`*&)CE%
zHR0`J`MqA7U5+O2>~!{S#|txXL;azZ_YQxNJ5jl@7L}N4+OLNG@%4+pA3dF(x6R9o
z7cWm9;^-4#81$E^;I*9US)?PZC4<-cb#60n>ntsha)r|Lzbm^7b^ZqhkY-7j5ezaE
z0Xdh!;5rkRxE?a-e~@_Rf4u+UPF5|vd{%afB0BAWt3W#^aDmBz6L^nQ0JwdH0(r=*
zynJ>zMMQ6c?Wm&>Uc%&T6f$Ju`>!u5MHuo60hiH8fqY|8`TdVCHkpJdyh0`hL`Viq
zB)?V?AI5TNVU6iYUMXUuN0J$7h19y!#Qld&$&8G`pldm>e?pmsWGh^h7S5`k%kSG<
zR1gO2n9JC)wa2MQp^N|%$f(vOkHv4FJ@=4dwH1NPbkMR|i-;a|i)AIAzW_H`mXjQ{
zEK|q*Z#z2IsKBH_B_hBv%US_`!DYv?S_$Rc3YXz|HQEc^TlW5E3<%qz883Y1!>E4A
zB(o{JFiK4pf328Y`hfJwN6?v&#ED?y5$SbIaU%9yP<%@^_hb7-;W8seC3n=M^G-XG
zu&pH2Ebg%<`E^?`3@>U)^6Ev3#rHcd3AiQS6Mp~VgwAmPUHbCnt6hCxzP^ra=8Yiz
zA2gX)c&zPBCSy!AX$8jS3Iby2DE_`{tr&Gjv1hCxf2{T$#lEdP23$^Y%v4qit)U}4
zVyf@*s&C%kEvpSHozQ5P4`1&7Ghu>$JLN2WCvkZk{%y*`LTL+ek^+N+claF*{!RbI
zi7)^C>EmBN{8I@e`(K(WC%~YfQ0HIjrAD`sR!^VnHK4bJ;3kzqGwu(wj}00p5Mdkm
zFx>&3e*|UFH|NpMX*3E5fN^vtHcFV_(?nvxhDe6A;yhaUeddErs>Y&hg2YHyX_ZQw
zE|H$7(q;PaYnqBw|C;VTPIvFBR;|-MO$o9r93B3Rw5kN)(N<iuG^DX~(SUiYY9)G!
zsbW0J_d8F;)aAh74g+Rcg})lNwr3h)`=%jWe~VV2p(fg5d_@n7z2Zo7QY&msrv-gp
zB^OmyyG6tE?+;%p-6y1a%f8DxC#5mhST~)li2_|dt(8l08-}HLoldi<6vrCWV#xv#
zzxBC9zru+sTYkb8iy+H)6BDWI?KE%h-C>1bhPD<CD1^isC_?$1CrJ@2Jqhih&L&0p
zf2@GoKiH%e>>q5PUcuV4PJIYqI{>cSg-caln*BQ6YxKU9`1y{bHHwn+TFRwaGI79h
z>`nkG6>(1O2+D(pHJj;tQhD9;W?Aa6UdEud*qgj;Q(>~Q*j+5Bth?Gx(+0e$J5N%Z
zQu(IeNi*=K>CDV}%lfi)uc7R{*2r6qe>R(R<Y=?g6i53y_0wuXII)aypj(AhG+0e~
zNDX1$U;4IHFWxTpm;EBrtTY3v=CW6n`IoXnwz04Jyjp3bslqd@m}wQ!Wt~44Ps|e7
z$^ptuXtG}WUR%-5UY+gP_S<N)yxrBT1KHNCdT8sq*^@14rk)o1T|1?djY%dQf6vA~
zwDoA6P$hn>^2wqv(vQS0-h@!~%XG@@26S%5^${|}!QS%wRlhY@17qpXQT?*nH(KL_
zPl>&EG28;LJDVeVhb4<|m_gu*Y=^8K&JwPwD?F|r-K3H{3e_C+yE+CWLziD8mno}O
zvnCVRr*7p?o!JmAxjL=vn`_jNf2L-cxa*R8urQPkMqKWj1tli0FY<2IovFGHos#b_
zzi-yKAhASMrl)ze=_G(HVbHNB*>)08K`41;{u(W>oWx=`;GjjT%h`qpo{QQ!q)g+R
z!bT_5*0WxeT<NSm9{K%lJZqA1GwXH3Zn>;~=p1X+WpiX~u9_wG^+P4lf2GZZ##vRF
zFI*PN*NbcyNvay-fbV<L6(JaGKU8PS)@hD$8VK0n>SX?Ovv1i~+$w^Q@ycP;>nhqd
zewqAO=|2cxg2sgH%`GSw);-t+*;`O9kl61ycMHnZksL(1m?jRwmyqCZ?QVfDIg?}P
z<&eoHdTBfgVDvV6smDetf4~S|VaHzy88JT05SiSJwh0+=*r|0~{<5PCz?cq`LHio=
zo^6NA;FnG5AkG2$#~m&SN-&dg7^uBJ;&7NXghqpQBzq2rr^ZX&?|%nY#gZhT(K{{g
zcpR1_%C6XQg$pPVz>c}BphMtA#5CA(d9{NJ;p))4=?bW7EMSrif43lX#$qZr<`#sm
z<8l~O?Wm4{svXHOP_-jD2C8->6{(X)K-G@$4}+@4L=nPGK)5j&>cUMxc*o@!tlE(r
z!&N(yW58-hatvARNN$2wJIdqO>RzLqf|{u)_d|K4n~Sf@qF1rCNvp*bheW7-1nLo<
zJHUii^2_;MyV;kyf4n_>Vr^~fLP(mh3e==EsMyS9b8fSD%Vx*nn>O3_piy2sjC3&A
zQ!986%nhvb!Ml2Hq+M#Gm7L3R&3m}8+FIynu~KXWf`+m$$PeU?Ko0<V#m2noF0&d4
zrYf*Tdaq4=omK5bGBh5(?noC)&7(lDfp4Q^r_NK9dDOFIe-`wCIPxx)I@d9L*Ce*&
z!}H0Y4t~T%-=y0?&r6clZh4lw@)%kg8xSj7t2wk-cap=EcYX`n;8%8Xtnar3$3FA(
zc#dklY1S?^B&`S1fi-lGU6Z&6u2oq_9zNq6thL<n^2+(~Do$4KVQsSHSP6$*sR=FU
zN$c>Iz`0T`e|K8ic%j=n3<|Dr*+ziHrN1K3(Dvq8gu9l^Rp|(7q!Tws`(wI^3?N+d
zhz_#tbU5p81jZH)e$Hb>et#6AWw6F=UNpw9)=fdSQ&>#d_B5xyGPP;Eo`P9#6$~o6
zQ4gfcc3PP>*lE4S<-S1!+fWe$YW@U>=oo$(IxxxKe-e@EUHLf5>|jTrQuBX?ipr<e
zxTdg)m9D*LXUrMkfky;f*W|dEq%nrW`Q?f{+VRnr^`7c}o>)y}t}?Ca*7j``St&0S
zvG$GHvAigYujM>r+};C<!e<%3+p0q=6dE35%Lp_|_Nc;Ej*J`3J~FH_c@0>0zKk6g
zRz=8Je={vt7`^a$I*kz7PRF@ic`wxbgjcc`Ez{Fp`Lv$D+o|!Rvz0xFDp}uv;bk3q
z9iQ1FGG2iKkJ5_iuFIHtEyM0g_P}+d@gsf3bzNN}I>nGCM<>tKZrlv3w`{v|GHuLV
zH^tr8g)+aWYKD1h^!PPLc7fAOdba}Oct~%qf4C@hdc|Vp&fi!Ta0$Z<OKFT`x_m6J
z&XUgct``>A5MJkaFv>j|i~tpvT&Y&65wCSTbH4Wv*Ac(kY5By+QJ2_jj75r;+jHfi
zT0EGospcKAY`#&yZ$<mg8}eHj*!V~9%&3!AMhWNQ{1B(_TMPsqsq@vh+Dz;7kse<q
zf8g@Z@|2`8Sz}UrMZ3i$#rK%!wQSO^)@{Fx0*Y6ZPdo?9hdC}#FW6q?Xn9_Bz9$D!
zyk$bq_qLGvuuf)PiX#x*lVu<ADCZPpLf#%LhjA)B5mDVM&12uj&Qh5+qg^dK?lL-f
zg`(U6-%+b7o982R;=pq*j4H62kDa}~e;#bJWgoM%KHBR9{jRQR2WJRL+303K>4!e&
z)<|&j*-EqLH!Zl?Z|aEcr0HK}V7klFFJDda^klwNQIM;snv+L1PftGF9G2;Y0`BVF
ziv_MLy$L>}vTmuWR#HYv#eLgc%Dpvk=(_x0KSr&Ie98&MKv$S7Qm(Rg9Ut&YCcqr?
ze7CU?Mt3#3R3e1#?p25EhrRo_bN%LQ-Qxl5^{f8_+|-emF*+_212{D>m%-pV69PCm
zmw|K$CYONFHUxiTxe>nmSLhFz&83P9U>J~OYk=I+=v$CekjZHdng4%CabKUF7JG6~
z%X*J2*7~uE#pZ|q9Qfh4m(#D89~T7&!z>CT55Imt2ztTkU%~02#)A=pX(bN7{y6;O
zjdQ8#bD|G$t`vO?{cY(R-(QumG(uqdi1$|m;w_Gq7y5r3<5-DdymI;7+0*s&4Lu>E
zTuFM8D2JIeLJ3bHuOct+uTmuf@fCgh;qV^F<i!*bmd2p;tQ`|;`u(53{`SI?7V>uy
zU1;V=8`=OP0kh}=f>7<uGDAsb<{8AxtJI!dB()(41tbhQ`X<g0-?f*3nCCzOBK-j-
zy;JGCeMWzOqsbJr0p!^~@v1~Wj!XlP$N8uJ7hvMR_*K)h{99*=K!162c|OCsy7JPG
z476+-acWnJW6HWFeX?}NZfuaPte&Lw8Vai}=LOwClHy#YE3@=^h44gI*(PnAvwq_{
zehGG-?5ic?R%ZL6uDp)?0#^eU>B(_#POv>x-jIJc8nN+t;_($EE>}TF7Ruw#%a!n5
zdW%QRr5Ts11s}rNgfMmQ3bSvZpYP+zbC#YYIz`I7mMbGdR0dg~kv!O=&?~>7_2)uW
zwX;tVCbD<R#)4UDaMjm|%F?39nUGDy=BSGLjgZvpDf)GJQ+W)yz$9Ao56*x-R-rol
zNl1S&KmFNYhec)pMoxS@1jI;oDPE9eoVB<@CJD%-5F_<19Lvb3IMY}_Kf(D`<m+ch
zh?Z0U**~+-%h?M&^2CvyM;>dOkCCE4WgI%!Xb!V@SvAK`iqc@UxQgt!vkCWd%$zIE
z6kJKzn`SJ$I>*jzmxxM<V{cJc_2U1+JFtJrM<j+=86z@Dwus2M@|_X6j1^&-cDkTn
z)?yzC=Ok~~rf;@qq_MPTjzs(2kw_a6lORSpWwz=^<j$urW&bf1(OWbMvgLjKqwM#q
z2vSJeot2xNR?lFW1kSD$71Mxadcy0*fVR4GRc5<wq&YGLYGReq4`oSJ3bobrtsQ@t
zQIX3xjWWS#Cb*sC%A1=iN@?VY(iPr(o|s+AEtQ8#F-KYIu*wd+p>qU^u8h{Z@HWg)
zO@b1FzH#ZChA7Q@sCZTA2D=2C9EVVd^G<1AebQPhJn;b)ZJsn2pC$-s6RL(E7{CF$
z)JRns!aA4}OO%EFGL@|f1O2^%ry_s*1W%EJ1w8312}7dS|0)oa06is1RHLk+V+C?(
z4U-y*zkb0650(jfG*I@HR7p-wtq(j5B~sGrVp{3@up~vzj|<v^V?gWcRSk(~n6s)&
zCqhc$zsXiaJ-UoeS!pRl!e$;guuZy+<!SazxKd<A4zE|as0gjFB=&0D%#eTY=`oH_
z)z88k*K89R7r^DMA4N)2%64Lt9m;w+^(z(#$$S|hyYa2vDut$`Y-AaV#l*|0qH`@l
z^PK3R?b<nsDDq)TKUk*z!p4X6kSy6NGE8`uxt(R$pwRnH7xmfx5}0r$J)Tk(+a@(#
zu_$v41&fwwnK<qvD|5URZ)<-~&Tq$qfu_8#yF@|c5}}r%Gg__uADm{^mi<$8ev{+N
z7e(TuMBCN$+_wG#c}EX6Pw7(eduWRQ+wGYF0d9$3FqT2tp5;eWQaDN?_Nf56F*wu6
z#voG2H=$0lRdamWSL)5v6qNB7q)Mp%+EYTJve?GE^N>jI(wO|%fb)N`a^D2NH2;b1
z?Z?H6l}QxUv!L~8618`D)9q+m_6BJg7X$5dSo0Ok17!h9Et!{z*^k90;BFVdIHtLo
zC3ux;I57GunTTnov|Gq@Th)1!TqO{-W~gfF!JE^@HufqsO%y0yvZkISjEQ6fTbrua
zQ#~UK(U|J>dUjJ1<eGnSUIi4gj}Rjh{CItm6_jt;aTe;9iIUBWjGhhNPy5aG-T<Gq
zc8Gg=lSh7*c>~yWhorplb#7~}9ySBkM<1lZ$~zr>{Wkpi{PptR$%0bRW-9gJUw^##
z7oY@F<gJn`CaGrp^WTTxU;aLl&1dn9)13GZ^62c{lb@+;;W&R;f@_V~Swc`eAP`Py
zW_WQ!_-GDj67tBt;;So!zv{yINte-6y{m9sWdXB&=!E#3d8UvNT8`94Yj~H|JoHmi
zW*S-?r_kbvcG_3nuHHB>Q}clD(%Z0#YjPTW^;mOLhlFiT&XjkzTl&bom?GVuLP-KT
zn?a_atw!;r`MG~bG%9;EVY70I*?RVxMPlGJl>Z5e>z=fVOD-}{l6-w=VSs6Zp<#4;
zmKkz|1TFlOB-`NLN{VXRBKnZr-Ytz)jBs|Oyd{kjOj|pSR*U?|OR^apxAR9J#lV_z
zMX`0lE|~!_@wZ#TZ)@BL=D1sRnLt*E>CdL4^sB5ftwMjHXGm0`4x$8V_oTH@=0p=b
z<AXU|H6StvBe6|xOGNp-U41@oy>2-H18UttFdMih__5&)nmO9h`oSUO`2C`wV+P%5
z1~J9TT||v!DCH=QT8m?7Pcox2OTtZAWkP$x-sn;7lPepJq$+_vqi9=6=Nk2;qiO&~
z{XFCE;K+Y^;)%S!utrI@KTsGb?t6Xdah|%pra21Ac&$ED2tMFgv@;k76L)K$R-B3=
zK4~Y$=fJ~$GO*<lN-Bm1$5omG;aglnQHp_X;uX$!>&R0OTuwgmc;qSXWnv`w1TrZt
zLc%>#+%Th0B7~dF=#O8|&xMV0j5fR{OUoKMzyN>6NcDydV+azTpgj{L8T$r4Hh@0^
zT0i=$gM?FgjT2Az21yRJfyx6xqWpyzm^|Q#AjtvTQ+y^$0GNeIXIoK{!+x-%Xd>i+
z2p!hGY^J-)C-2fuwz-ln`ZF;DU0eQ2;z@3}pDI4?p@l192yJQxgw3_5d}e)PZD$Pw
z1rvY8=88tV>U4jEb4wMs9w~1n`a=4%ieYe2<>_lwm&#)w<N|9EYx^U?5(1><`cG`i
zw@-dvQE4joC8g#~+v97><qXFmPdPUH+oo*LrEb1n{5Q3BIUGCXrvwW5eS1glR2@58
z<&M*(y(1E76qCl?aYV9dg=FuM;+Gv-JXU`{Nn}P&at;7oM^iFk>^PcV9<JNQb@6J|
zJ?f6PS@R>(r;=;rLZx=~Mz)`7sxQ!jr6fy<d)H6-{8}4#Z(1qIm^>i2>^J$yiN&ag
zjE(z<HAS1I+Cz3X2l#MNnD<xoQW@p?a;@|eKYdFf?HaOnpane>^zF~VhRonTwh4bw
zIk^uCMtlEcKJ`3h@aF0@vZA~?T;FTnNKQGdgsoc;67lZKkA4=sO*C}uoQt=nb<<o*
z5B)H0#%tVh?m0DE3#Dw!azEn|??_0TZV>6FAyl~O+P#RkwrL;K2DUBYjrCVI2-2Hl
zH|{vJ)l7a1?QK5O$%Re(+unRbKp20+oKX8W1o~S7w|D*fyFAy^f=u_Oy8(hAo!)IW
z6#3ZlSMhSZ=-Z6pzD7NaX>QeoW+TV>;=xt!oMlo>-0gH4{jriVYMd21jWrQkHe8FO
z%{@4erOVzUmy=H9jsN{ZP5<#B&b|G=Uz8c+J)1<iyuZyy^1MZ)`umCyd!K*zwTb<G
z8E(v-)&p;Ig;%V19PS74#U+i&zrirna_jO->Kf}7QA+<eBWF%0>m8?@&VIhDd`Ujv
za|+m1FJ1Syc-Pm!hlqK!v#f!V0T>(0MleGu(_hh77L**Tder<){n25SC;Q3+wf{?B
zS#ZU`nD5rxu*%o<l?TcJl{!(Bd9++BedS!S)gH6xdd%|rA&{d2MA$EAG53UaI?Vmx
z2OG>(@gTZy=r8w#ciPMS;J=`|+%LG(TpqZ4ytiy9Qog*gyae3goACU9GaEE1mr=Me
z69P3cmtoc}69P6gm$8c}Czq|i2ML!PHZryW^S9GBGMof|RZP7!DN`K7V&})C2BBoU
zShR<+Y9ph1hO@zVKXvG8jVy6qYS?WxNDZdx+D$dJw<*#nov{)3lKyU`(Fp~;LuqPe
zB)aCDO)EO!Y#8b5fXtWQjH4r$Y+FlKBu*HI7A~=hP+fb5kQBF^t5@8Bd?KAFL=#93
z;xKTmWg2FG%%AGjP)JCT#Me87OArvfGmE-4sxE;T^QzsqsUw-SU~{+kHa+NxIGUTp
zbmrvH+LX~zy8Ow=$)A%RO%)|-))UAlN}RMyk=3S$C?lr!#K07%*P5HJ9l2%Isl4E{
z(7}+d62}a%ZM0oltma{IV|q_U&Q0{W6zkzeIdIE=@2g_N;AC=?7Zp8(FDolHwWGET
zW{_o#W@!ryvfjpAYA{C*BWC8XnD5wNnK{mlB63`}<`W?CqyjF)w8A<M%1+Ad(<C~@
z!K4yaO^y3wE=@pDZg|j^8jhj|FG@IzN1~nGlhBCD;MVU+><K~hU5fQ6RQn);dL#<n
zlc?~2$du1Q*=1I`9<gd)9S-9P<v7QqCZ!z04x4_e`!)rK<1xPmC*st@`7CX-Sx(S)
z%TW{j<~u}h`3})(8ZuVIst58(8e+4o=Z^a{6-3lC&cOU~R%2Q9%Nk8zM+cn&>eSy^
zE8k2tw$k^hs-V}VbOha!k=9m74Xnib%+)J@#%bf~6{8+cKSco{sW+DjTwi)TRAkjL
zns<JlE#@=Mz12{mo>3pDZ0dAogW?=}g(xapG3wC~i|gCUF>jRvhb>QgEpaTYxRBJC
zbNCY$KcZ91v9tKmm0Ie$ZP=}Do>7EO$e@khDQzj2aBaD&_8!`ja4tOGk6$DVN=P$*
z*Ao?^>x${tWkR}M$2fKTB5e*;b&sCE)yX(*KntTj;*V5sW*mF|NMOLIXZ^5^8>Tt1
z-}Kw63vfn_UIi&s)x&$Pm&&=EARoc;jPvXq-(%OV{L~#~vmCjIN*z?rvDa@MK%Y@}
z`mM@E#GNzUyJ5KH-3{@19j}FZ8VmP-1W#xkQZsPZqK(G1&M;biotjwnNt(vyJ;cWa
zOo`BJ(v8~|CRxWi_M@fFWpN@1LVJCt7quL>+^D7gKlY>5Wn*>{3*2F}n>jWkE7(_@
zubn`5ukG%FYLfteX8tkS=s1h@!l+Xrr{%ciEG_kqi|~e4(>`HKtBEUX>mN#g7A{Cr
z;8%|N<JYB)P5-~6xon?sUD`O63t@bwbMH$d%nIXYx?Z_2J)$a;>%<T$HMOS87f)Uf
zQW}JbaAGpWIBq0UjQWQsQ~RVX$rRVtjvJ!JRf(4C?YTn-!vNd67GjJRGT0@vZ{ASt
zWL*kl$?3|-2{-N;$Bp}CjQWUw1DB}0<=8ESP{hsVcT`j>oeqNH`i^V~D3UAa^~zzn
zm3YKI=)cbPEvvNb4=&48F)GL>Vp4Pk9rnsxu(qWhYkAGd))YDROJvg(EH#&CJxsQ^
z&0LtdbZ;w3BdkBOu*EoTVR1|S<FR;`v?CTXZSBBfz0q?`*D$~j7fNq`(+z{X=(zAt
z5VzrG#zF|SeTP440V<qm>z3h$saxXbv!TWCJt%~b*e{Zg8<77HSK>|#`fGL@5q~rD
z{~Uyu+gCdlR7$2t7IYN6Qx#RkJ<^uAx+cxL?0Ob#r%IP(vTR_s@*>?SXk@BqUu>qc
z*Y}T?e*hs#MRW>fZe(+p$7weW12{D@m$4886PH?}Cn|qQjw3hnUa!yxn$0DG3s4Xs
zsY(XsW}|M#oCecA`7r+fVS;3k$z*QIl-ig%WJygigRyUnmZv{X^7PBw_3Q2X-N4BS
zkIL!O*N>B;A0+)%k`^tVoKnIXb^7}A^!vM%+R}5S2c$HH9!K9jed76>QJ!`v<VT&q
zIZ*F;ufBiNbIE%(ru`<A-)<3ZeBaRu0hLD63PVnB(hj2{gmQ|$pTB9t0P;`t8MnhZ
zfhmU>0<4`w9Z|b5YX1DsuYbHr=7JQ##;@_C|JUc&+kf7Ao*)_lKAisb=Ue;%j1tD{
zlhRh8wx_@TefshCYY(<%xmUz;X)WT2<qm{(>STXy5GMYVSzZ}(%UBDb?e))Jf1J9N
z-+mbo_D%rFw}VKcY3Z-K1Bu%Ees>fA5GT>*CyE?M1%Nn@#CaLN6IUe(N!}CW0E|F$
zzX}nZuxL&ey)eWEhA40IyC7ogUO~5`6=i#fOLP~$k9Q-Tjf4b7dWp73V*32Z{E(1o
zkLz(SxV#R3_e!A@G9riZk7S%jD}?j+dyuo*P`F~O0Ou=K$kL$G(C2^Sn65?rD%voH
z_ynKnEun(zAujM)OAvsn-wl^S(f;$56I$V;=W8XvhJN>=49Ue<sh;GJ^@5&9F?5<d
zh@MC&u(e^<!1z5{9I$0ZM_hhOS_(|TwOS=pg4fr7=q5-C3dDDf0#&Gz^w}eT>-;OJ
z@$e^|G+Oy>MQ4?y*hD(3Kwxzsg-B#qG%;1h-`PNb83>le8%6!e$Wp*~=htns?X&PK
z`L|0$oo^&z<E|K6;t^PO*aj7y?2~ulg;x{eWmQ4_T@Kb=2#Y#dWL7yUWh1UZc4cKv
zcFa+K3|d$)Rdyd4|D;o_WBNhA!rIV5{9A;S#S=4<2r>ozYYO_i4V6u=BVy4|t#Hou
zyUj+(h-5Zy<#RWdMpFi2CP+d=lpZsStReYen7qZYde`v~gpyMx;rjVj<+Q2OIQc9y
zoELtE8-}s*^Vq~J2kZ;D0Wta^?mQoq7D4$(1#axdfjqD!vs}^$T*d?FjSwb$DA8ft
zXHPtekNG2vv_cN?G#P*#{3ylyBKyF**l#9J5%U@{Zyoa6$rG|SQM5)Wo}}%U@k25;
zf8B!(=XK+-qR*|xR>_VK&sx2=&@eRq8+n%<VF5KKC@U;YWdA-r4@MQoX!VvNv%xVI
z6WaUOQ*{pp=L{u~-0KOU(wX7PCMLo(rgx`9)@jDzGQy{11f0c{xCi6?w*eXlp6_?`
zCr|F%Su>{v5Uh}4sF`dNQw_m}R4m8xe>t21-jlu1YALi^9<{+sZ^q{WJKJkLX+98d
z=`uJc+@Q;`rjK&hDrHs>zo~a+uDWCcS!(%r9MPO(iZ*7f7wpr#G@cC2p!JqMZJ=P5
zEqU&jQAHU<kW)-2^kxf>*%aL8QwJ5B2eyY4F>i9uY^zW*&&+|wPSx1<k`yY(e_DIU
zE6>WBB{5?kncI5C4)~p0mJr7_&LJn7K*^qvlz#PGd07sb^2Y>fn&_{oC>ko@dbIjB
zM^R&o!a;TSErQ8J*dZ3jF{?b171|7r8E#qip}NA-eT$+i0dQ{<OGz)JcYPDf0MQ^}
zy@~Zm{gE17>m>+)5o+iyK2U-@e^RveWl47ma$sq~&O=3xwL)3lLtCsoo-fcqD?!fU
zXLF}JWxb4jh*Ol72R((%oJtPu$}$+0c+2%2*ijSXQvRjiOY3pAp$NoDjWr=H`0r^X
z7O9URX}Q`+>~`05MrG=YGpm%%V{bwbI$-Mx$1SP@?O@Vf7kbD`)Nnhvf8V_%M>{kp
ztIPJK2{lil_re&v1sUzo6w%lukN261TUTdax@IZg^aC;ry}ry$7n6C#e-blCD{aFd
z7~ul+M96|*>Cu|udty@<^2tc3#D$tYSK2-F`_sKU6Pd4iz|b<w4k^=yt(OwKNAq><
zZ$puno{@jbDad2KC|hI?f8_=E6<g{dPyFN~7P^RkKoMZV1qHiL=v($Z#E;dMeYcw_
zoN$@EB`Ck?DS(a7<@>~9%k##WY`Jmw@dk>k#)CZ*?NQLvqq|V|5B5&^5`_M(9Tr)Q
zMp}XBChRH+NtpKtDhQ+1;`zIqh!gVXk2RXoM0f?++=lGQX03>qf4HXH_udy8i4=|1
zOGFSgR{JM#Db2qgm)pE$APTULgwXb8wcI+iMlw`uxuWr$Q{dpgOB(@ik6doS?SKLq
zfet&WlsxQ|ibjs%!MZ|@HXmwcvpzm_DuD3d`cq#R=jZ?&D$Iqew8c14pcs-2=%b1j
zv#Hb){1`W6Nk>%Ef9*52>EhZ4Gpikfky;ozA?6wVDjv6kxFG9TfGOSYUP@xM7U<2M
zpl3>s81ujDyu*lNHyW0c|3jRu(8+m`3N3PWM;Xb*UI%d_metZ?{8ryqVQjSz!pR|3
z#fEkp3UY)*!gRE<W_SfUQQ*LhLcri-A2L3srh}BqyOq)}e@Xu~KQ~iwvrafaOiDCy
zpg=Ar%6jw+N3`^b&)um9IHENQN7~y#syOb*)pQWS`5w|zj%Gk;qbKJcnA<R3fOnH~
z``e}`xDA@LMZJRC@s<h0l-q||9%W*sr~J70Em!M3L|DCApU80CQqT&o>m&-di01sY
z1{B_@4X++Ge<`TVK$?2*ewc$>nFX7iKuJ`<VOrvPxH;89+GtL7yHnwR8+=N&(H${>
zi_{LX;ijHhng(NUGk)txJ;zgE9VCUTQ(*nM3sQ==^#L}c`(<cRnZ`@k&d+VQx5;{v
z6RO3mJPm5?uWPFDbZGCWu{inaARMW&IE{J&7h{DnfBqG?aQglPE=~%n8Yj3cs10$E
z(!5Qzam_`yV|Sw0D!1GXd0p8Na~J9C3#Utah}V}*8!AsFw{m~Ow4t(rihKXu)y1pl
z2i5=9{GbBzy70pB!4BXP*F^_r$x1XDBe5}9S3=47r6UoHv)u0V^|88=Kz4kSZMXCY
zCpeqze{fkJYAY$;^PKSHJ&f9UI$y=ia$<k?A@>RvG+}w5u5`sZ9vq*iQxyJ)AOCmr
zbc(7!V#ZIHr&9`qvG7yo>8N;Dp%xJ4az_7N@4*&d*0Ot1UN&CPDYA@T%8j-2Hy!Yu
zg_+tzamDg~ktpfourGxS9?8qYd-WF>zOk}!e*j8DdZky)0Vtm_lWP=w`oW%z(`byZ
zjO68!Ha`2-WH@Sm@NPIEPRKAO?f6o~ZO=F#L!G6_Hp6+mZ)G^I+20K3MF&e-J;I#%
zJ~tOcb(#hk3ww2Daw8X5ojdf+h50Zrba>Qg9hbRFhqD4bot!V9pI>kP1!#m~ehOu7
zmsGeh5Cb(bH<ztSGDrgyDan`kN-`>cQy2Th5V5=c?m!Nl{DKo>;4T96!9#cZ+x@&a
zhe6gXi=6XbR+N2`FOO#rNj!jU!FY}Ym*R`7tf}}SXMaD`vqf?BeiVy<I*(#8<a|>+
zJSakC!F4>NH-K^^Uv;N3!&EXaz!)Q=N}EjDF#fvzI62G0bI2i5FM-vNHAIqskSPh$
zpxC6A1^f(Zb&h#N4K(l779PCWPUIF(WnVnAijOtHQ323lp9r7Tki<)p5~@h7{AOQ_
zThd2Y(;{}3oN_Q9{a%gvxb!%<9`||3NQxN0)a@mUKJInTrf({H{(qf5a-UNCjwxrW
zW!5@`S{>e3mzzo^sWPgUo~G7+pQ<&!Rh5lT+MBWT5V#!QuKxn|F5wnWy2p?;O^TKK
zZEdO|%$$tTRFo*)#WHJMs_Z<Nu2>^?78_$l=VaALB^KT5P!q6?X+~;$zarZWr$aiZ
zcjg*4<f*bg^HAJkqiy%iw3RAk{lRj(9_a|XYQxJrW6-w+oSl197fcC%)`}`C4Hkn<
ze&$##8_>p_Ew!4P>e6q=oxrXKjBO{Bh>fAiMApB^^XTJSIkQ?O&TpvR)9#ju^Hlxr
zXfGzh=O;X8Z`cs1L|x9_nTv9-or|N6H@q)9e$%j&-Zy<u6Kj06LTSSet@>0g^CKgr
z9qz{aqL-@wWXkMb?FNoA$F=vV&ktUyJ;VK<gW^+eAU4xm{vaPJK3Uc)dus1&F~F-j
z06hF(*6;Op`UfZJud0_(xG@s}HJ8y>0uuu<HaL?}A0~fQkG%>4F%X3Ndy0Jl<7R&o
zkOLJ@uudnoTKWF18_j`N%ud(|-y{exK)6mg{M4rXq6IQ1HFtydgd7RRr0PjbK|H6-
z1=<_#2SnzXEL;RpHRSV-#d}f_={!H$Z9pwi5%CRFb1EEAvczaBlsteXVD;y+#F;`o
z_S#FzxY;KYv2gpAt)JO1Ea0hVcKYw+cnBH;*OZCZ%6trI_YLIIrVk1WJCc`CxG@s~
zH8huT)&&!na85ERf6K1iHVob8E9M8HO4J(|3^3zl0(4nqS9II1y6OKfDvP8<$(p$i
zl7%y|Es+!-hYw$U`RAf9-#%&jeSH7)+ta6uvC4-S!{z(m7gOHo@~_cRrQk0?gR;83
z`uFnZRci=ZyGOa4*4~yYmj9>n>u;Z}OJ#-e<qEgIzW+|de}jVLTwD?NJ}MWTB<`VH
z1_}E@%5CK;zdjMO4d1(aXOWcy&`a@MjtZDQidL?&-?-c7@}qaQ4`O@gGgRilckYZ>
zX}uS|;F_W;Z7qpx-K9w9O@yV$(JLD=3vKT=U!^Ei+*3%S@tJ-WUM3RfYMWg}VSifp
zMO<I^rM?eOf74c_8`5~7T`MQo)s6J(ih)1x_jmG%fp@*z*S2-L6m^}EPflsKov~`y
zEVOI9+pTQ6XC&ow40fqzhrg5WxzU*WliZ9cM6GH~Bz`<+SSXGv3_F`Ez?Fi^-@q{U
zFJ$^a=Bleug}W5{6HD&lGY6wky`l5eRn@ePR!}rmf0JzD-56!eaU-K4w~eJ~ckKl3
zPMn)GHxsTb3Cp~nubOykvavo=)uvWOy*6F6vOc6xenZx9`L96N`pqBTKC#A&APTgZ
zH9olu$$PLoO2d|~cOO`u0hReXmsPSb^DmzH+1Fr(c5--Ym4Z>%SC*mBk9G{SU7myI
znW(&Je>jW0(8qdc9U!2cQP^z}gf&Pl+OBHt$!pURl%fcFN3>7Itj$_!J=zn5Z@M)4
z(G1|4E`mSwVzLCT%(RlhN1uvp8;uIaHo5XGwwUn*nO%FNF9hz&bv}nlzcwCBv_<$h
z%XvyK=8axvzX#`3p)m6y>mIb`tqRE;@UYm;e_6BUTHb&_9J+ID2V)PQ590eFp#|mu
z!joyYLtDA8?Y*N|Pi#fv4LvQg41{OEX{%8c<;Bbu*dKyC_p7v4Uhnma-EIbVQfehW
zBDD&co}98@^Yeq|;jQK;jj=w0C@@0R%hKY`s0gi?tho+&M^F3JR!RC^#UDaIk2wWk
ze?YH}XKSko?2)+>InXJq_XPR0@@=}95wn&mPG~@8O{C#ksdDdeRSV9YvTrgm(vv1o
zdVZ@P6rXD;V2K*r@+?pG#X@d-9hqQN%pK31un%X+=-5w0wX*vOy=k#D>7~B?IIajE
zQ=SnWsVBi<lXUhKeke#APg#uCHA69qe{|0lqsW8vVBLGDE|3S?0cB`A{FTo{R<VQT
zNC)*4++HEi6Np5-*e{x_BAXm6ZcM3IQp*g7_&=SaOS7k}wLhGr<5<HsPqlM{>0}E^
zHe$>_nXe&um2GS?ZdmI<_QzW4C9P81mC|k^rSI=I*`j85*d^f%g=l?O@2@+&e@#Uh
zWG4!A+E0*guBY4SVP_9?QVKaP-i@w~q1xB6VIlXjUzC<j;#FFvJp;G7JQ5lTa>p+!
zmRxa?=wY_e83*ke934~uB)lx1TnZ``i6*`^@7D$jo;NDg2Q4P4haGOKy|Lg}g*ue7
zE2lOTsSwkoZAs<Wx$~MlB<HbYf7QaH;!xrqk}g8qdM*$}anK{-%~au4;ml9i^XaT&
z?!hr2;>#^y^(^>7P%b73EH!ak0aJON79D}`^}J?k1QFs-L7q-ZcdHL`Jj3xv>BGaJ
zNhDkPOxmukaa}i9qwmBr+O}l9Yup<yW`F?Q!|R0C?Xi2(#`|kY<v%XYf7`0@PU;)E
zpJ{?Tyhw^wD`706CBHT(@v?2<b2q6%_s@l_PSW^=NLPHNq-r<&#_WMcFLA`=pbYdJ
zcFs>{)HA-w70pboStW^qg6^ekyBi`*;B4-`QxOzu=X5Bd=TQ||ROjnd(uV`r$m^O_
z^@vX7<M{N0vJH_^&+&xxf9G+Hj;leT?@egUO<P+jeO}9UfF&_{m5o{J`6{!Vm;jr0
zI?UHgFDd1qiLJ?hZ?E1CkzDKc#Dt*0=GEYUN@(f9p_SXS9<fZm_xw=L5k(9~Sy>^K
z-G_KsSDTyITPm?f=DRoCTkE-z)nbz?9h%CY9*ct8-_81ik1`H=e{wcg{B%-B($8fy
z+n5zVo;lRPXRTHavkRPGj-8ml8vB4*fyXY;hauFUO+k|EfoiI1Ca5m{l>Jzz>5s1S
z(~~!LS??62K6!i0X>JXU;Fb4z-u)^pY$$S=xvR0@SFi+!QkAC4RTtnSupxy&t|7Hp
z-|Q*9KA|bbp7{N>e^VlBKBh?kw8ml^XF(MVAzVeFfNSOVjQ+$AwC>blH<=HvEa4nv
z%hSG8w03M=3|1MpK~CvMNb}u<0uB=j80HDkK}^0I<~rby7RtoSt^H#U|J*<2t|wCO
z+gGo2bTg$7o8DN!z|OnGCq=gFQ+P-(?vp6fW3yD&+#FR~f9*c_dd4<1Fh3vF;Gzm9
z&L^7YIm-!oh^5pz*rLzy;xsYtElvO8;>PbUgJ&Q_mvVXO0lKs&!5-%xae)x{^c6x-
zyV;uXy7t-9T#X}lpUWyOVwSi@QRoGMjEyQ6<(MADx5_6=5(#1OiNFwTrh<wn-h0Az
z&-RqRPqwj2e;h^;EK9{dfpCWb#D|(=U2&f2h>B=HD?|y=WsxlgJr(nOeA44!I>pK-
zvQu8C9+Vt5qMMI>bl#lKirW0y`|S5acnnud;cB_c<W2^BDqV<Hu3X7_i@|t!Dg$xv
zU;w5=zz_XN!94I<Mcm@gI1Z(cDKa=WiE<pgH?~=%C9gI;`SB6|5b#H#g3=}6Z7$7K
zuOII8w=j=k)7S5x{s%$FSo4=rxG@s~H8z*AO)3)tI5d~R(;y~)<y%W{961iY`&ZN-
zaE%_+qX5I0cH13bZ&~EF*ux+|GKXZbkHh}`r6f|KRBF3+90!~105$GnQ4&Q_<VQ+T
z%jZQc?`|GG-28H}u!J=_9nA9K)1vW%!vAZ9MDLcM_3Eu&K73vNa;KE>_&nePN?D6f
zz~2%7y8BzJBXVeehYxaTeSaH3QyXjW(S_e4hdJNh=5P4hM(XEWc+3g=ipqr4+lhKS
z-~aW)pKcTh{`BEy6x3U{#?3Duzux>~LB@AW=Dty@-J)Cmy=3yLE&jKtHFyZv#cG|8
ze)%Ud{_^+FpML%Ln_jWmzY?ojU^Px}arsKG7WHIYSiaDIyI*ly8_KY5%6eEJu37;y
zaKlRYxZHpVIJ%31`Xy1f6(SfT8e6^E7!o57AV@4j>akI2yf2birK&?@p{p3_E%ice
z(<`>;R_Qf8@|{_+)$hyAr{(57Y1Mh}1G2<uti!)PM1YxA^$I$mISI`&$@mq`N&{<U
z3(C1GR5{;&#=7ECg2freH7Je0>S%S&OcC>D686PCFsX_*(68uWm@87Mk-F94GFn&z
z>IF?ylXi=S+wYHGsoiI6^-laP{oHC}S8#s0^#Qe<yULr`#pxy{ZepfAoc0>vUUN3@
zDIR0CxBB)t*!xAuQYLlVX7&ECCeW)mcSI8`)kFh-5b#IUz!f*fGbqgL#c=kpC6ASS
zj7z$}edZ$jFo#|UB0Yhr0OgB2SIPtNT#`1GQkVM0Q=ckvRO(})X6@@bTHv0Fk6T?t
zpIChps(w=@ZKf+<{@k=<yjxW?!*n4OG2JG_HAUo1cT(&u3KX&Mi!pb^ou|{eVIJ0^
zql$5V8;2!*6%A%hpQJYO+qsAXmyQZVeX7*`8j;qLh?+%AlS}O%>WEFZ+|KrG@|2Sj
ze0eZ#U5hlmEnV3S&9cjG_}kL6QMkGl@rrkF#2$jTvoP#c#5iWE0dGQeX=+dmwEl9e
z45m0o)3l_j#1ZmTUxENGqK8FEd2HSWMygkTS`jZ{*(kgW&g?daW-FuFkkq-?ZlPOS
z^H_M&ZnL)bn9Re^T4?)Vs-9lGy+T~Q@}ovNZI}6yf$cS(<(%#z1FP0y?3>JSw<f`?
zDwztobVZxy>E60K(w@pY*ON}GMD@$LXen=mOtpjLxyMlX<yOkP#f>FbO8OSm^ak^P
zRxWc}mA*dI-cOw+30dv9{R`#P?55gF`Eq5%;s>)$7Cu8k;F_H0<`nYkcKv1jY`n5N
z-SV1y>X|ilD!>@@mR9Y2uWfa&t$WYDvr@<(=I%(*#jwf0PEEuk+EuZ6V>ZsZ{j7{-
zRULgy%7LKUW^KV=yPb42s|LT5bZ-WKvkjxSZH;ytThoS_WC*Y!egpHC8G=Kft^!6v
z$`Bl4dgqO;C>mwOFpLb`uoAiqp;EgHp*5};zYZWsEW=NiA=rROe?x=mVE>%@gJuW_
zq<m`j4Kf60(Ke1|2*mt{GlVwS3=_3keb)@ZDU(C#u@u2+SU0ElIfCv3(6f+#3ET?k
zi-x|`>4jDS_eU68G59V}sDx<(O|zh0IAj>zfit~B(&kS1`Azmz;)R|;f=lIb4Re>=
z$L@k(?Zt4;W4>+9V;w>vx6`@I!16=7W57@%U!KJ#09w47<NyLR3<>=>6op+2^*Ov}
zec3JtY>*eK=Jy#;)$@6`eqD@zXPYGpA0lVTN3Z3#B=f^0|GbtXIfl?>Hf62DU4=i4
z511!h&)eWhqz|XL47q1DT2alx?Oww2PTB;SLIj#mmJK-Hn<p1tQmosU^;ZY<;|oCl
z<xEmIsMn;D0@xf#CBfTuvq4sL4<%zDQ#a_SRf{l5(T=2;fPT|K^S#@DeJ~Play{Tu
zEK1nl-o7MyE_j!GmcA1Rx<|r?64Bk<UrtZ8bI(FjR6ZKQ3y?H{&`9vfg4WwXFp9cf
zjWwRZs1A^v>E<e8)4?a)=NcSsln#$JatBC7`-Um&D3V5!{0~9Wq}7d}z7NX(msnIg
zz{N1p^a%3``3FU#8eN5dHs+haQNSNA9L1x|&4Bbobqyf*s%1`n6r>6frtY@Q_PfGV
zlWIBYC`?5Stk&2aOby%U4l@PVJ(cgZZkLbPX5g>1iuc96KFBIDIY$OkUbbK;Qvx|t
zji6?yy}5y!4ij@ob<6={>A3Xt$dxJz1!oG(b(YsE+!cQ5QH~CObYMRtn5%Q+Dh>yE
zU&ocfrs&gkdTbJ21oRADMRHtW8{=wy?5g)5!2hhCWD@Ysx-?v7Z`yE99!(6eQX~+e
z?yw#8((1jU&bliZanNKK#Bi(>uR0L7(U-G;W=>SJ8z9p{dvq`(4qDE_8ZTrYorY#d
z0SpDi3|lQv`xLW(ALm6U%yWZVY%0Z6_w=A#$NZ|A7^KCw!w6TlOh4)sQ9)s5zP+WW
z-3UU|&Ze|HN_)k0j9QIxZ#kWA1Ua1};wA5_^jnJ8jZgt+=M!7f!wJt*vFiMO>g@io
z=bY~c@>v|i{$CDQhxvZD6V`FupY(`TK%&<q#v;8saMj6wfLReHSIlX8oMUn>C$(f8
zOd=*xDZXX<Jj?xAtbuwPP*b5w4_YX4okyNMhBAI!mo{3B%uk4RY%9aiI4ue?<w%Mw
zVr}1wQ)L>Tuqe%7ILRfw(Mc|~@I+BPQV<&K`rOZ%XH?4vMrfT8)FTxsuH+<aA0}nC
zb??~@6>{-^LqqQG7CZ>Q05p%GY;qb7*@j}uwp+8xQ;MlaN-IYB)w{m5&$~6+8Rm_!
z!@Nwb=j_o+N%32E9Ld~JrF@mo)vyN4(}$VtigoeIsqM_4CePr^u21zKN}AD5M*+t5
zMNbsTaw)U5V4QlcEjHQn?!>uJo-Mm3E;IV+((WXGVrP*3^XhQni}nCTsgybRQmU3Y
zfi7MIJAD4Edv@roH^@glyhdJXZ;#AD={u1zm`b67eTkNmuX=-6sdTVV-ek53<@h8t
zs{lSY@x_j*hhZNQRp^=e9y~qqiQ;lz0eyXrwWt2=EN39?LVO=?B8<J7n$e6c6Alxa
z8V-eji0^=xucd82U-@X(0JcLWGLJM~`iM-4v$}!lu?X!LBXNb1MSmmaoowHP5oKw{
zycK=iLXI`&)DFtVI7a<!5~tE`XqcMh;)5~gxI|sQ$ofF(k?ePOfWh?_6K196GJmQ9
zRP0udvIrUHJ2f39G&0|2WZd*L?kE~(*tKqde!)ZswWg9S@AdIU&MpH7-)<a1e?T+8
zZbDX9R!-&)wT4yrPP=0#RV}=Xe3H3F5OsrBmG3u9e0_M!3ne~aGix;0@KNPjYedLg
z;Z8tVJJV{smCzONE48cNx5l;pUI%ETD#K5Ap#-*dvk->5{1xhtS;BWQ<C||$!A})`
zh?vaaPz9g4|L~O(G2AQ%Y!myg#d@vu8iW#_FV<`0x_><bR8|k{tgGW=U~T2~LFuH5
zgNO0VS&6?+6dSCB2j#7nrmYs`A$HZm*)zWkB)wc0dGQiGWD13LS46N3<?3bN(yw&a
zx!(+@%eRYg?pDhpv(=MZ9krONMRD4HcK1!Jo&gFw2UY|Ra`Gn2=~_)o^;%nWJgpov
zE4nKuk6{W>lOZfq>e`eyAi(WJk(}?Alt8u9!=QsQjuPNR?=JF!l%NKDBOq-jQGd-<
z@5)O<$gg9ydGGG0s9i#%%n^{Z1IpBHgiX)Qx4+S~TUpxY*Wm2&`vcR1h+~i}<2mox
z{LpJEWDUVQ=hFG#0;W@Zc%lB`y86d9t6oelh>T@*obi7Eu4`Pgmr=Me69F}sG28<b
z0yjCA;awUgmt<lxq<@+@{q57A?__0s=F)i(NTYFyqpVIGIefD|zvc152Apq^!dMp^
z{&i96<yWM-{pBxgE`ySgZk`Js>kR2ZiP*&LJA8C`0C|B+`JVgYY9Yx8R|+n1hL5a$
z-<9v_J<-B064JQT7E2#p``~BXfs?ZKp%TLn@U$r=?TvtFR)4JHCfYR8lc*9y@KhMi
zzu%`X{-LuEA<wX<Hby(4AkH&Mdu3D<-l^3##*;~_%ObLwn?ow#g!S&Zb6YK(U(3eU
zt)Q(<Jg6q83x8v>VbaPY>LzCHFI5?`!f=U~^_>bWRkz!oqJAi51ll|l6Unorxm}yP
zs#o5%Cz?M~wSOm_Oe|5$JMD=>oA*9TWo1ZC3mJ5ig6r$Bm$tg3s*wzSQpnnxWv7_>
zHV0uZmFCc@nVINjCiFG&pY_qPv8wb&36=e|;mDCW01<+rl^I78Nvp$oAUm&YyeM%8
zZI_El=f1$?44Z4ut*K9zk>g{s1dlYaajLy=jMFOr%zvq(J<#up1I=<Gi-orTgJox_
z+b%U+VR44hB}b9|e3;(ef4ci+tsGJc^u){O?|10~phGmME6Ryz-16%`mM?c7XR18X
z1cl3%&RIJwTS=4!Aay5nY?>!e_f94Y=ngQosI}BL)>3D^CTZ)n2&zkZvM{amhJ}?@
zAPYZNNq^IxLy@{xYVr(a+^jLRwP}Bwm054LK3(*XY=nGbvdTVJ4Rcl0t95H1g!l8Z
zovP7F^$65esz#>cD0J*#qrliX&zc-tVLZkC<WB0sn|nB>8bE+sQI+Q_qnvPh9+YIA
z_q6R2dHcekIoG|obf4Dt9`{5htuUJJ_bnP^S${O$-Y*Cdy>XcN)fHlxr}~7d8=>li
z>)+H<g%fFhUHOO7milXT%1fl~J32t$xgJp4GjzcvAOdbf7t$znIx7p6AY5`dzbt6g
zAC_#|XzeBjyZmQK%{T<RI_DF11Epeub9X=e>+6?4eg3-=IN;xD1o|tXt%_m!US6%m
zVSiDZ<y(CZLMx9!M2U#uAvq+WaH~uk*cv`BcYsb>85>yZx6~R11kjod&R)`|U8D(E
z6Y-EJwpN<&8*e18`V@Ihjg9n{sMN9P701)4^qL<2PF-=-&&%DH<?dsXXEgX<!oVH@
z?}j*)0Oir9nMkS0coBWXJ+9)29(8i3m47d^H6k1f4p49!ot0GhtIgIHOD$}{)P%H%
z0yWX!xqU|f(^`v2idI@-BaU;3R+CbyD%B#{{`>RyI_xVB`oR7cGWVz#0<2r^t;Oxl
zq;i1VK`^y<?M%VyT)=TQW?K!L9MrjjyD~3@LvCLqO=RHX!uTueyl)Hrc(ZIUU4OK=
ziPwcQ=O|1l2q;vc(P6{BR8UQn+JFr#x6010L!v_v=7I>?PGrej5)t(6heTo!Qu>L=
zl|)2K%rDV<@j=Qw>h9BX)?g#W`U~O0<CqWViL6@7>5v!jXnSUsg0E?H<^dQ6ZwTMX
zIUmX``rQUnEq&7p-e;AiwaJR4Du3M+U5y{nR7JvDve@NeKvb^f4)?Yod0+8#iTp>x
zL9G+XjZC|HzA^~)O6rwFn>5EvE9Z8qmI1QXIhMiMu+GR=c2TWp%H+WFJT<>vrd*L`
zDjACc-6QmoC+SwZ<x09=D8A&M%h;{P`}-E~j~rs{48aSK>PkxIhm_8B6@RJ)4(%RT
zQkgacN(z;JMV^wMTY_h6N>4r|Q{BTbU^J#B;TXjvxL_fsn5&d`Ee79)Wk3rfy=_<q
zb6w1`g72uoq4=ZKZD4B{uuN+muuLD^Aqq{vnuv$r5SEd`Vm6(+1X&12M2gRaG#V3A
z&~5@R+pWJ-NMjN%e-lUp&VQj%h8r;r^(GTm)f7GRYga)Gn{vzbw6#48#dvRCg<=3r
z#Di*b*g*=kgxZ@jybG`OjP=FrfMWILPrdR$S!Mx%HE!10&sa;yqo|@5=9*Ed^M#%I
zS95obms$-x#2~>4g=RRf@?4AlwdR-d*mNL}^3-&|kPjQ^@F36u!he-wgTlsnB0JRJ
zn^<2^h_pkh-CC!l`g*iv+OZQYtNtBFOBPygILiSf?_F4xyTY)Ja4OPiW$OGk$JVYO
zu8k&m*h<huT|KmD+Y9=E(=VmV@Gw>wATaedUA_f&wh^1rn3vAGIfY>^23^Z4B{L!5
z1$Ftz;GrGL>@CNr>wg?duf9TDGQrl{4ARw{zpLe@3mLb6qrxsSBQO%J3cCr5p<jJ%
zW!FRy#y2P`J<rjgy}eN#ay7fxXmWmIc2;yPh2^Y8p{F1OxNCo?^EV1FbYOM1w>&v~
z87JE(io<zg{oO^iDPoI2%fwqmyawQ-J1FB?!tMqm-JBaSm4DOvrtLLMsh+l}Nrwmx
z-fkuN2-os;4VBnXgOff?Dz+_eBh}Ex<>R|+sYWgDc?Vv#Bkvni4TFNxXit=CAPkl=
zVI~6gnBdA}a4prKhB4Kk)-ly+gL?)+6R;-Y;Ww6Qlp=Hp+Spt?4Tw9Roob-HDp;i&
z8?C?BR3jmuw|~hsv`1;{TkEIPo5P=6H@Y>?VAE~p8Es+z({gF5)j8vDOf;xBLGUMr
zGs1x8IOw<acdsuK`WTV*Q4)}fqG*9L2hv}uiXvu}ejYUzWREEr1UD-J*NU;YnaPpP
z(>AcR&dn$>7_+Ch0G$V=>>mg*)wL=ph#2FJT#u}%>woKO3WY4x-B2pjD%^1UBk)oo
zNt#N8d1qlrZJ=k{)p%~X-;oRbj2EhToSTO9nadH+QI5XP%WxNI(7rM0b3+=e8@35^
zn$v5?b9Ep?^x^C3zSy=?Ks+jfjI&IB@RbCRMx-*}G%K%$>!}*f?pq8M@p*Gqx|a|d
zfHb>XnSaI3MIN^55|q)+_a4!fl#LigqEHR|K2FBhZ!T)#Q37M{;d<RzNZPsdCo$q9
zxWTZLE2_qta&vOk%5P12E3^;u++={3u0?cZJJ;>*Dz$l~Q6nZpw?5I`aE;w%zjO^6
zU0sN4`E4lXcxCZ4m^0;iESaHI?d~riM=IXBzkfq-M2x3?$m54SWk+B=;;&^#VF6sf
zmpUB#0mB7DfT04VwKMfwT~~ORDGcW@Z5DpLakQuAkT((~2f}Fhk$F<O+97?2?0Si^
zP8)u`1br8FVAo4Z!+gD@w9eN{>R^7oq)aegFUdUo-{nbvj9VrS@zq=8NeV%g)aJ%K
ziGO;(I8S2J?c6G<3wvgs1h2}IFc6`p-<T!^{ARic&eO$%TvcW_6|n_&JsN!-pjV-A
z1if;P5qj0SueP_}lhhvVS?Xy)S1v2)fZ}#)8q}^n7v8Hk&X4D7%wV&;ZCttN`DE&)
z_b!JNdGaW-Bg4Zevc;3%KxBndQ07sMuvRahA>@C>vbuh+fuDdB3b1(<qiP=Iko86f
z;b&ILcO0o7+&+KxKV=CB7%(i1j2U|Z`@l|LF8!?Q+CMNsGet!i>0Lhs>n6FI3BsHS
z&XDQ9XI+@&mr=Me69F}sQHVDa0x~j}5%>frmz!-eGXuc!T9?snGBbZZ2l{}N#?Z&n
z-=4m4w>8R>ghKj|lv2B`16BGMO&^uz4N2(ZZkwLb-^Ld`$1J1wq{~a~qPBg}2D{(x
zf8NMq_=^+zOIBM4f^@RghBP3J05lRB(hkYeKrJ*6^=h{T3ojd)B+qh4b|Kl1@d<J(
z*xU-KPizgP6%JW1(p-OF&l;2c*Su}f3MH}H?RLsWwWy<e-I9r8D-Y{<DQ1Q%f7wLt
zw=ex@ap1L-@>HK-(WYt(kL<ZczsfSSJ7|kc<f`4|bXoqUHaPP0YGWMdSZ5b)F{QsM
z^2zdkOgQDI096yu(#_}4YfD=pZ8H&U+-dRauKV@+e)C~7TQ`4+LI)h)K5pUzFiIG&
z2jgj}VfgsR@UVH)Yi-Y<SF|gnwTP2lp=9cBjS=7j?~1g5<SuzA#yUevRr&8@9AVP(
z%iU<NHo_|3Yn0>?OD<44o?FsBb7y`Ayoe$H?v`35Rg;6t18WHaP|XHDCh6IPwH7#n
z4#?fsN?leH_%MGTvmtrUr`?m?0i>+{m)tRFY2ATNSTy~31YJvWsMZlAVm!aSb=}!q
z*hxhr?V)}LA?+0{Rop=z1`@m*k~AqW!Q=4n5SytJ30v#v+kjFbPH*$#?{^PBz5QJY
z>hMqOK^stLRKX9A<);xrnV^RE<+l}@=8S_9fxr<4@IrrCBmwSN8{Q5Z(2lkd`lj`L
zYz?%MpmozZZF?65d4Q&gZHOAj)-vAvwy|1Ohe$?OGvcqPrSy$IsXzCMKjWwU&7#!n
z&tda0Y~ECV21Ebo+d65N+^Ga8XH@G_ZqimzAXLEa2#&zzOlv#QQB+uO!O+yaa1`Ur
zP)Dni<(7YD!rVl<SOYgz)4KLE4(R5ZT~dyQh)Zffc$sib6}M*Bj(@*>EdAb5ue;_u
z>)A%g0kp#woe=~xQCGseGZDOqvouG-{mvl6InKgB29Z4S(jn`Vk_m5y#>70=?Ks&t
zs~r@vfH2{@gL*V+90Nf(#OjzcDFV--NV8*b2PuDbS&V2g=Gnz4-N)l5`pW|P``o9=
zCtN|`*JUw0N9ZQro=Uu)#OI3Fvb8}TO4y0v#mN21mcM;(OPZOVS}l(=q+VL`(YZ@Y
zJ}zHg^1p^wOhy61t;?t!gKWkSXTe2GF-WIxB{nNbXVszf49=AQ5r7!b6~UCWB4{T^
zxEz05CipxyPXM~<KVobhrEOzVz9Nj40!zg6UK>nE?E!9tYEMTSD8HKG?{e!m3e~|x
zPdyJ*tC*CywNPE8Ytd59tXU1q<8Vu7be!0Yu#BGs%V3b^Y*)fE1fV*(!g8!-A=Cn<
z7GjsHKtR&vdtGvWr57c!%Jsdd;M4l$1oD3ia8AR6Yk~;p_DRC#-&_wwV9$kMH>0+!
zgtjzHq2g3RjJ~CWJRh9K!mu4P)hVZ0Mo{Xina2omgo=sl)MK@`Eq41_-Cy#*2~KD=
z({^(y$s5U{Z~(QGlrq%?iXhzBr<anHl>!j9IZd#Cea26#E-e*Uz8fRP@LBP%yE1=#
zw%*;oJ5yAeIMaT-<AuWO>M1T~zJ;;_pNE?4%y&9Y$O|q?Th?i$Un(;A#d8$N7bY#H
zsaHb+(b-nzye1;$%9(Ok6F#o?Yocb#iSo-S9Iw%}^s52Uv*{O}OTV^U9B)B`Fu6Lu
z6p)HIP`+bqSi%0`li5F4*uO$yRj+@jF-@)>(`99zgCR8D3ozpjols-E*7{XuS9=bp
z!SA|1CcCCNlTW3>3>`|V?PsFG456df>KUo9PUACDVJueni9L^Gnm}LArgw%OGe~)>
z)nAMr>&Cl6kD18yNb|X343I6MA0!1CTNrCw0*vHpbs>>C$5PU8!OgTEZ{vURsMHc@
z^baoY2bcGqa(PCgAfR7aAhPtca(RsjUlEt5J>{Z)ZPCcmTr}eNJ}ku>w^Xdn%cUnJ
zBmXPc_k8?b<M1c(dzO+7!i>*I@F9hC=_qCuz(#^twAxfqqpFD_H3@LX+Asm^+ByO3
zv2%soIzZFJHhg!)9%Y-s1+afpFkWu`Msb@0unpl^aGMZDBokKQHcLO5*lUJcI-KLg
zzAQ$QDw@PvjHX>6Wtb6r68bW#7Uh=<IaI>+TCY)jXs9vty;2FE>`0g_BTmO+kS$3u
z0lAQmXj|>xH@7Yh^Vw8wOQga0q`h&HHMDd_lH=Hf>N1HYz0l@XzZicNaT;fSsrJO&
zKgE`4!M@sb=vykiLcNqZ#Ak<6E=zyIECL}0=v!wILLxWL&SJcev&(PuG6_mpm8-EW
z<V!HNRQ<XM{Y`Mq%L<d}rpfpGNNQD{9pI+dXeJdYAChvU*mO<`e0znbVZByf8sa-v
zJeySB9SaNWc^<_flk0y=HHA9ObjdzJs{N*^(c<N)bi918yXE5$*xT(Tt@;wy=QUbj
ztj{y+?RM?hKwr1(%(u<8rgdChgNwJPQX^NVC8^7>zJjY$9@E>M%Us=l!qe5ley1&`
zrchJU#InuS37FqHvqo#UH$};!C6ShUV`jrIFuK>?^~j)mohyId^-vPg_ewm^c-GiI
z(<^mndB@2GzjfAYV+nY}jn-@9X$$$8^^-SMJ-^K}X|4mh$@?VJyxBUON0=4Y9w&Nr
z6C&e{*R)@-;H@?bS1+E*m^v+j_dQ`sb&U<5%$KHZrfLc&U8M?bz7v^$Jp<#pL{vs(
zQBp08UY#3T)XRTn!&aFBSTmhc3FXZyojv5&r%Aue8BzWf1vFG$P2K}`3_NMGwdAU=
zcmJB#BFL{L_bj2B5ZK%mJhU1v9#CBLg$qJkT;q!Vlr%1=2E-ko?qHW4AMa&HN&<Ag
z@15>`^IdMw_qkgKMD&0f`%5G)NoaB*G>Cb)K&tlNbZ!Rz11n5cSeH?_F%tqcIG2Hs
zH4~Shb~3<!?)2O9&j-zl&szF`ig<{YnH!RpNSqyrW=Oo8$-T_kh0-frm5ZdsqLp<%
zdiwgL&GLg}?a%x&XyA<U+Rhqe=|w_r<R_PH_)N{zmUY3k&KFwJ6tm~Ze)L&0YMN<3
z*ZJJfyof|&VCzz)F-$|#6qge?cgM_V0v~prXA&xZr{O|%%udT9#8sH8LikWyQSK~m
zb<Nq$(QD<9jV=ywY#{f`yb+oUK3RTfqn2rH8;P4<*aX$o6;t4G!!3#>ZPcAi24h);
zOcA^O1m*0%PY*6$+bZ691==+binXFImhEw$fp9va$Ghs8CdqcM6yM{<zS0|^kz2Ks
zsXS|cd3{5B8&#R&4UH$>4yU8RO&r!59Uf)vT5O3*lLw8rvj_F?75iM~Gi*ERr8}9I
zUWpubjCAWnixHcBO~dBeJq}-pBA84eXp7Aj&79Vz$C;%yz8NztQxM-}W_hX2UysSK
zCL@x2O(*fq9y*!!Had?qd{+~5-J3l*RPUF6SY>ovpuu=$pznP_n=R;sZh0M%cySLo
zH#q*iRt`m@U@N+W_F?4Z1~2mb4KK1rJR>-$Gj1<?RC>t;k{2Fk{kwPbXu&e4X32Jf
zcs!4?x|igjDHhr$c|aZ0@sTHKJWDo$j5O-C$*3Fe)U<y=ha10MO5WPJivvJqe9H=d
zEgH}>c<~8m@|MNLFXNHQS@q&b!n>AAXYnq1Wliem_X&L6HM9$2n0LWWwqz>FE^$7Y
z>d1AJaYjox6brmg5Kc7sW#m6doAP}fg*bAR%tJg%tg2>V$ZRDiHC<bDVaQP%J)1<&
z8S0AV``jAE${8xA_pM1AAK{+=MQ*`=Iy8+gE#Cz~L}df^TBKnM2CfZp7x7%j+CePh
zQ+_^=Yq@&nd}Xplc0GY@o05}r6;-O;R^@J!L@zY@_7Jr;2byOmNFlXuiyGeGI2>ai
zD?Oa|3K4FU9_v_o6kQ&+>2$NDaVWMvQdl^QR8GdGgKSPemWJD6@zSqSbTGAlZsh~`
z;h07)J~{|yOLC3e3&pT=<Z_Za&!E{P8c8y~7Z;`PBzALZD$xz&tQ5~(=4HXm;I7>6
zMVs({5;4;-UbIR+SNr+%IQ70fAHHAgI}A>R2+PacAy1$MWusZVR)M1Z<B#R_@HBHB
zFOjq~DDAy-8)$$gyx`)DQ!!b8frcjYyfzZL*7Lv@B<PnOIf1id!Q8y~Y;uZ7*H)$A
zcBBu%_Zr45870>bplR#cwjQ^c+m(&cD5Cj@?5X<O2_rk40yu}yCfa+>=kR@Bi12YK
z)Wm>h-16^Ih7}lxDPLj%LcYu<s(5U^=N>-&{q^;ym)}9rvVSruVL%yw2PrOZwd?4T
zh=hyf8~5qNPKIQIN|bQO%~6?@p9*AQ8+cg`Xo3Rk;xPIxk0yl?jTsgf<b_Nq0)`>c
zA#-e_BJbNcC|Y$9#ipi4?qx2#Hg{>gtjgWo#=el!sy~;*>vDKvuR3o+CW*Q-PM-eu
zyb1~-3=VqhLpBgZ6MFG~N(zLicSc8h+n`ZCdbDI=R6zs%YSNm`GBVgLL&Qrs$dG*J
z@pt?%*-9h*iA1?X>m1!XLxw1@cdLi{@0T}T_mx&X4&Tb>2*xTTlphHLDiCssSIIu_
z9F9yb)N(MjTg^~~R7ra~Y+^zppO!(TYTL0J_w-?h0MMRwA8^Qj{BSckg3=5CG)KI1
zLW>Q?<Q(0u{wrU=F0RS8T;5u_D3_BmT9RJm1a{tDMK2z9>I%6QRIcmx3Iy{NsA@Ak
zLh#x%%U=6IQ0jnMeA=lF@Eg5%O7AMG6q%J~ejW<;gBNgDid<6@`n75atx-S+!W0XA
zzOjnRVz`K4^^6&RpHAGi+N{>pvRUR^<yoyC%Jq7M<(egm4-7|*u(CPg1XYYd(P?~c
z@o{)mF!f3|KQH;zo)T)6Hyo+V;o`Z3)_}d2N=&XkKJ7iC_}h;tO;a}<{Zh*_ikMHg
z?pnO8p_2;i?3)`=w#HKyfqB)|l2VdDe}o<Ux3ze@0{XZ>gOTq2`5}RM`tJ3@yU)1M
zC%JO<Qn|_=#Njp(W$ox=c^&6SC+M#O*V1rLczbEMr_`E;|0@55pM3uV%hd~Omr=Me
z69G1t!KyG6x4eBa$_5LH6B0JJ{{wCZwzZcL3^Eh~IWU(2HUcD<9##(u0yZ|6LC-2E
ze{FBu$PxaYU$H>HTnl))ud}<wHAND~i4T{?h#jD9U<9(w*i2+mCs9t)U*BhzqGg@@
zO3BFG2OC3b$eo?nXJ#3_6G}uMg^eP*=tab(OC(HkxUZCOD&Nt<Y0X3wp4nVVf;2yr
zyddJ2C=xNaC((m3weK-PE0?*g&^qK7e>kBH7688%+Ii0u_<Yn%fiGyw6!@acLPVic
z2qKaZMrn@~R08F3%WDM+vf!Q<Mmv59o;N1sUvbY;rm&^F6|eCY=L{%=XMKoFIbkUB
zV%U<Q#A_Jvv1UNweKCuyv2L>Y9to=gKVhIeO_}nb60c^Ju=trN1FF4eg>@F&e<eKX
zLf|Emu>8*Gu~AEgDUSyNvw4eqkyY$*>6B*SEx`_+`D_r*vPry$ZCu_YBy(g~do~Fd
zlENAeCQOvq(8BA$8+RnIHl7GdKrY3013|o&3j&<HmV>FC&-@8gb_kCLSP0y+phCn7
z<b|im&te%LG9UEoV`M&f7^$o>f0!D=v2>mUIhUy*$R}O{nx>E?!1gF<-VfX=sO(at
zMKHwggA)XowSp}NA2VMRfeh=+TTodSut*4s_l>=Un6(U_2r2Us_!6-MP(jp&*I@TF
zo_EelvVr*mwdx0D+l69wR0zlv?uG2P?C$Lv92_aT7gUmz<0nD_<rCI`f4Iq*{DcDq
z<wSy4CY${mKA#-F)JHtw^L`V4ZT%r}SY=nwaiv&Z&w-^{t#V{3o6?h?e=fE^AesrW
z{jT`;fBuX6K}Oh#eS9<<4iBF`y@d<TDHROA;1tD+dRR~P$AjZiSbjhhdxNGaCnI6n
z%a`ApSNqMNDTQut?9`(Me{1Z(uo@AcSJ|;FV|48WqEjcBGGD;1Iw2z43q-5Yy1jsE
zz&3vY@iS)r3k+zKnm+?Gb9QJiU=$cNOZxO_u{WuY_six(vA4Zb9F*Uh`Th^Sk4x<1
zw8ZvCO*v|&oS;H}#k+D^&nCy^v`uk*>rHu54W8HEKJbc!JOXDrf5h_>JP9t<iui0a
zs`*8}6$~qV$oUiW;pL*4NfgiP$w@iM?^B1x>teTffeXlfi1m+~56CN0S<a3mIY$_$
zC96UcNp1b%lVX2%)bI;`twvvpXP2)(ZmN1z>=*Cf?egDGXH7Gn{=FzptLAKWB#-O!
z;`8i$ymeAu6mJI2f7#n{vkMJOKEc59^YE{bdf_5z;Ua<KQ2*YC(y9+3vr7ZpZpMdh
zaht0yfbcZR(%c{NMYztpuKLiDcP;tQ6~}#`4|%oXLtLJE_)r@Ib02E8l#A({a4o^5
zbWWI#upQy9@w_Fsx!w(Rd(paKbTS~5k&D@=#*s9NQ#TSte@utPGlAFd7q9Ary4V&!
zP0Q?m(j=+jN>>!sXLUnC8lVW$Bo!A=uO*5z62620H!c1*iE@9<;&%s~3tqzTm+PSr
z%Y*{dE%@M|8%(}c7qXt57K5W{0e3YBYUjct%KBTz1*uU)QBEXTMnFkg;2#tN=!kAs
z4(i1Ema$mXe-T3skDwP0_EtcFz>yM8y0SQc5L`AGAfxC|jIE*dWdI10NDrL|QAsYN
zH;F(#8chKvwSbMzBDrorgaoAnAIxO{2$7a<mO)!c2#{-HbihpkU?N-!fpp41dXY{^
zGGJUyy31T`(gj?R9Pszx)JifKy_*xs5jHlDk;s>jf5DZCUF-n#XvY0LQm=L^saKnO
zq+USdd8(c3LfhM`-c^Q9Sr%2#YX2C@x=BB4)v0+ATvArLzp@r}%DBrq<=fKyaI6!e
zOp23<lfc|aWkz!Q<2bdE-3-=wUA=m}|9gPr=fL2OfLr34hmlH>&`tem?7fOhjPDa1
z<`zi!e;6uw=tk|TW1G}xqxN2|VOL!PnIBCQ_2PVtcI%-UbbdvHb8mcLYVc-}8f*tx
zt?vlxh%hI7hh=T-h%_gh>WF^rlqDcX0C8tvNfc<WWHiVymQW1!W-`n#vp0K4FvZcZ
zJ}S-!Qy8sy@pk*=$G1BlU%dU-_WKtH^8DmFf9Q9g6p54wBx(t%GIBfUB~KVEQl}3+
z+Ci3DvecjI$?UwJP!3@c;J$_cdlctuxOFtbu&^nuA=L5NpqbQlPq{qcyM<T5;F8B!
z@E)`gP?~0?*)`<qOw4poL*`V1J|3aZ0|)IAWXlaJ8mjn1WwQo$^gLs*fYM1=Kx(uS
ze?kW<%~~<?%WPN;w(H|*fq%2}k_Qe;wYZlGfwpykc!+5pF=~!+=QkV7&7>^5U_QYH
z`iabKGZ`*wS_U3*k(V9~gXR{dPh)vj4#)C}+=EF|9S_TCfehVLqtmTV6|T0b(bfP0
zo|TiW*%be#Ta$8JPphV$d@nZ8gc^=Ce|QKw1#~k4{io!qzgFBiq?>!t=+jc)<J=vH
z`wGL&6V2<m&=p7xU%s2u!F`;*&pRDrm1os=bDGp<#CvceM=5}`KNb`qT7d%O`al6n
z3n<{aNtN#~f$uPZ?=XSC!~`7_@EsJO4hm2Q7s&L&1cpbfF&{_(4CuCni%F1oe{clz
z!^)QlL1J$n)ztFP%kWUj37<%&KDpwy^j1x$Gh~fX`SqVoc|IP(m;P}yJN?LMV$$@7
zD^W`{!!V<Cc^V721-$1eNJZ)K;MP1P7Q6fV?_YjAc==}U*JlSWivW>8Zodueh)0i#
zt%Bav8UrAS;)D8)Mp`Cqe+&TQ8DMr14}VlhT&PhA33-_HY^jHyXVpnLT`DYKy+BEU
zlc44d>#eW#WO!1nDX&ea6MUu%PY*mi32^ZqFua+3?}V-Db%oLiLI9DGl0jL-9fXSg
z;%#ese4f>JRM%gMa#UQH^@QU>LMV;KdjJSNfR=`Oq$q|k@-w5c;b3%{C(|sQE`O%Y
z$*?*q_P(4Jf7vcC%3(cb7=hyR1vO#1rF-VGd5$&2gVJ!BV-1gdDE_z{3#1Zqivaof
z8vsNQpWX0%(*7}$?DF7={C%biOI&l<!wJXThm#L;`H=T<=eG}q6k&C`5PD4)LRdfw
z>dw4E2M$694njY0psnH}jF^)St$!`Iv`R7PR5GQ_R<KPR1_dWgY)^~%w9ptO9lA<B
zEmqOq0>q=;z%d;fv6%ej>Q<xr(|s0qy^iS^UBEX4x1|obBsU>JL?feH%Q8)MepU~^
zW0i|)x^f!MpOZZP=<}wdk%z`vG%IlyRi8O%)OTd6Ki<|?yO!9iYl(4@Vt+#Fmq&mb
zNzo#a@W@x`M@}T!ZvM~Xt)y0`_NmWt6sar6QL$?oQ&$yIR~5Q1Dtz}vg}?lwB6ZDZ
z>YCB^yP(6~z>*s|>v9*QEYL1Da%N48Q4fBbRp+P0^J>Uw?UqvV0720<@OLOWT2zlD
zEPoBA)|-wGntF`1I$(#JjeiE~%jg)?@Zod-5QZ}OLzF;k$ykj(RimmYA!)HYr{9j7
z^<_l|!BYzUh=%QM`HPM92C4rWTmZQPrw<3y^Ll*=xdB6Cu47^-fXa=&!fa^9u!mB*
z5x+DDpn%Bd_BN_^s%F)C=fd?#{k7QN-d;%ee;mFBtqNssWOH<KWtUmFF$)1Umm$ak
z69O|amtkEwCzmjjDg>A3k1~~iGNi*lKmPWn3E_6{SXxlVFS`((LfC}mN&9J59}QMT
zBy%3E0rBy;zPsR+{}1^(e6uNI;?kjD<dJF*GzzhHq27J0&-}T~Kn7GehTi5;me#=M
zE_T@Yjp_69*i%`ty^nQfUdyKj`S2}swL^F7RrCcREU>R{3&xiU)4<7p5*8PN-zhCC
zE%{d}GyC&USMa`KQLq^_<n{f*ntj7O?{nr<9$&Cp7?!C@hBJ<&p;BB|VU(|+#zwQ{
ztCv}s#$_5;_*P_KwfDg)oow5C-&dZv=7~oH-B0+1X|Gm0@A#7pY0yEqzBefI{+F4i
z&Cc)bkGeK86Sb|c!r@(i2Hcqr+u;)(7uvnZo*z3*V0)BDpB`N;-|LQM&RN$e3Zy%M
zbf?VGOos+9yW9mPcT!S2K1bsGxnsJy&t5BIyN(x%vwbmJQd2D#Nlp93_PjnBFw(58
zdHwH~WVY<6Mf>TzllT-)SlSJu(U5&8mL?L^ZR>1(P;Oa4oorQqWL79}`_Q=1B*BFZ
zJLO2<&sCm?<~8p(d9!V|n9U+9NO)8&C>C2;pHMrsa(E^1u=^8j`1<M3VbvW(WfDkU
z0JElq)nCA!O+VwoDH~S1W6Uh6<Y9xgcz4~yl1%_tH2661EO$R_df6A(_a(JC!(La-
z?N}d&rvle>y4i<+GK#M-!)(-#Crw-GFRm^1i5*_Gu4sjiqbERi$#dV@Fl=##J#)C(
zTGU^{LK^QXqucZ|A3PIbHrx!2(@mh6aFDsAZE+pK!c1apEF31YZc;a~m*K#N5#1IG
zqt&8QHyD;6q)A9!6K;DP;Ybe9Dobbeq$_ndytS@hKv{r)E{-Q!y9wel1ToX*eRDLK
z&t3<?=laBvuz2l-u+R)97WCw50MSNp(Wr3ssqRNv+xNAZ+OM6^M8yraFDYv)_`5RD
zsb*5)-Pi|D;S|AYScbkP3wmfw!;@rz=sd4!GkSZElTkJ%?u@p~v#aLgOmFdxCU@Xm
zltEAq0j$)2GJ>E7FMVv|I+uBXu@GM_!*DpOCHGiBI!tt5jmZ){VKO@#`^sm+n+!{x
zBSS9%+i^;yRM1!osS2I%2h&2lW?1W}<~ZqBGsIeiWM^g|Q(u?qe0IeQ;uYNvivrtq
z`@s+<Ct2Re$v_HP*!E*6j*m5x@o|uX*~hJD#w4qM7K~=56)66@jGVz+C;htTadq2T
zu2kSVqbM$~bGrAc5T;+9EiI<$Ca(VKB6TJM?9G~)TA?&R!gvLINCEF*(Y8i2n{QY|
zgZ=`?tT>*b7g2**?P&cl97N4cM+993e=}BbCMs(exz^yfT@K*)!>EGScO!dpdwyZA
zfv^;RszOhNs)SipHbvP-1!0)57HI-qmZ*+A?&sMsxlN(w6zASC801v_FpSMT4)@&P
z7+3YdK)d((l<(C@cr~8gM+Dv?Yx9GoYzcWo^X)hB>;2>HziYPHgi%G<{p*jn{KA1z
zfRVKcYJsr(^WVGQ-~KM4@?uf35IO|!Ck4iTSS4YS<H2Jn-ji#|+@OnP)%m}yQdw%Y
z*i+w{K5mmGev1Kw>b;_tTNa!E>r`QsPrw+S@bD143kQzU%cc<NhZ5&JKlLb1@7Ws1
z2K$yMGCX-?TOkE97?yd|*r2*F=pv<~d0k!W>QY@R>^q)~)@TN^LSkvuP(l5K&m*XR
zTxlv~QhOJtC~?bZxRjJ;G4+%p_Hs?5cT&wxZ|J7bguN`M^r*!+7#ZL9i%+|x{1UUw
zZHQw)ACNj)wh&<8_}&Df<#{#jNK#PhlO|TTD+hf|XO&&-Pyr;Ta1-62)6ghO)rmAT
zT#Y|n`;VjHw42drqgg~=$ISXd5oLvc7e+Ez%xZIYcZw_lQ)q}QqI4}T=PWSWjhrz!
zFXBfx!WIZ4DlJk+X+G2(pnCb6+=bwkO->?0<liCzVy@`eFO#s$ky!JC@R)SCi0%Ov
zi3L}>x`kWG#+*jn4k?=y0!Bfj5^w!%A`P!iK(Du4m2Gc}a}!Xx08;&LRT~<AV%xR$
zY2Ruws{$KWz7?W!C=)lGt46!Dqd#8PFmZbFlN|7(Z#iJ6lR_?Z%6~B9f3Cc>W{Lx^
zx^#y&3A}`O?JS_WcIf6{mfQ!WQY)9KZGQ?ArY9^$wvdpkQQ~JVBQ_F|uJxGot_T`v
zK953h3geX(3TAuxkQ82W^otaKdZ<8D!NU%gsLV1_hd5OBO4;*S>doTS;#Mn}AFB<>
z2~Lw_+<*dgdh+wPtAcH{Qx20)v5GxUW=3d{C&d|B)ShtsNx&fmWx<USdN*|&kZ<az
zR7ODg025xQi~{0pE}!4QCMz@4EAnfCLE{=%R88i&*5Lf<d0Y}2Nm%iJY(tt>@;A@;
zya~$Htxwp1s?~C5^tKH^Cm-XAhn-*dW$c~<0Ce22q*uz^3~_yM_55CkpiHn5@)e$$
z=v0VCj)si`v@CnMLTTP#$4`scKf_LBT14pR7ECum)03YMuzXt70;_=gu`tVxw)uvk
zoRz?edTz1q(tLQX=)n7b^NL`Na%k1(<%|=|c?O_Y(aIfIxt3=DV$vUz1i0NQV)GFm
zSh@fD<pO3GObkaU;ulGnST&1SH{@m`7}<8W(s&D~kH{M5<4w1HH9(+Um3(Ap%6AN%
zMZ*%Ekr2M*NS=3qkmbQxiNzK11miOyD~_L;6GP8F{?xiNf3gvOuvpy3pnE6;@68(2
zl3S5}VJB+8SAOGKxh`4P3U^=QVTG;FLi#^b@Uxg1LUNBDYC@ES3-s8m7&qg62&LAe
z=PWY+KVYrS6e_%}Bf)chZuhKASXP*_);gbcFEK3JX-06H=K`g`SM%WWb=VK23~gjR
z#H$6g<5tH9Uxi_R#>%clHN&~fLE&N8i=8qON-uOC*F8Z^3efsFyYx`zIF<fF#m($R
z!Ibr=jH6yZAFYov;UfAj^!1uUubWWmupQ%3MWP-@-Yk2$s32P|LD%U{?~p~?k>o7x
zws~gJI{IM~&Ibd%y<=jYbJNj>?#*uH_=qvu@iC-AqXVOVGHVv>u3(kD?3mAxptdHH
z(cYD&ZSnH`W|eXDc|6;UbhWH!a{d}58()<5-!p$XYX%Gp#Om|GQ@mNu6V0Zt=a7p=
z`M5|Nd*G0F+OEpm_bB%ovxcyO`q|Q+fknAxRoX+;BbZ-O*+Ukfr+;N)Cm=pi+4~Es
zc@*pmoNiTr^QitOW{9J>r}6L}w`7gSm6D!cP`H*A4lh#fFGR%vK>?(d)V!{^mD9`M
z%?ssGK>XrPw&Iq6dZ)N0Mqf8VpNjEY$x7mMy~x$dpJK*y+Q_5!5!F4f8&tfykNuYl
zzA2xet)*c*1HGElVP1s6f`syk0w3t0ykAj$rC8N}^0Hg|^@%))$Ksuauh$-o#~p=V
zU3(C#zNq#fp#IX@gXH7$wFd#=Y5hKZIrldXJ#hU0%>8e7R~Xb_R_aRY4**;{tj2BL
z0_0rz*g^n$tSo!4GLaHLeQHjxw=cT>VNkezppFhdB#_|zB#qb4%>k*lVlvATZmx_$
zuwIORT=^{e0Iqd@Cb?>u%OAY&o4Daw9x^_};Jj+NjZ1rOKnUWIrSP?n`Q|pVQjYhl
zuLxGzuZyBRQb-z(2wUx<FUX{fRR)7-u@lc7A6)GW$^Dj#l-Hi+2VVBzG}d2nKabho
zS=LRoRf@p)!^VX-Z;ke#)@sJTW30_}k8Ey#`F^dmncGtrj!}}(acXB9y_@|V-)>;d
zLuTvy%seabqJ&Nlex^VTZ@ejSg=ep7Rj9X^uIYQbn|(<OX1>VvnfWEfiS>dwI?-LY
zYxo?d($UyEyUC~(E%*26;*wBGO}c*(nx>C%Hb1&?3asO8TSbt10WZ@zJ`udpJbY6o
z;HJ_c3winnqEQq9dk^B&t|Yn+Su@ie3`7#t?(ZLO{{v|Pk3g4ExG@s}HkT1zLKK$`
zW(p~nTbweNe|oR{`Cp&EzsQ0x$Idzw1hZim&JO4k392DH3_&ooqx?93Lz&*h5p#x!
z7ZM(bc?Npg+eZu$>wiY&Ip(d66nF`?FVG6D7DAy9F9ieJVA`^iY9prZ>0awQdaebV
z3L#?fHo%9xvoW`EYWol*Lu>GZ9yj2GnX>|n_i&N|f9dhs=+j@s>-p%<AEYZ@E$NE(
zN*L$3G;}DGnM_UmKhgb>`3jZ75OEyjyrz|iN@zWkd!jHx{$8Xy@St%NhtbM67&Rhu
z$P^iw%I}D$OE}JS(1MV}tAYy-R3+F+qY2ogZP)TnW!=m=U$p>TwI%abF!vo5Rr+~G
z8*L*uf4N>~HT}#xn-1@?)?6Dz=y$C1vejh$EI5V8L(<i(SHVVM#*ZKD+xzFs|K@BY
zN8?GP|NiqOzCcZqCa!8vT7aj&{&V`{<sTes(H1ErEy;h3%=$Zw{4B06NKz_ktuceE
zQ!?#gkycEuE-F(psE>(}pqQE_gP21kNH(hJe+Y5SRIN;I60?cMC%Ta&{A+O@epEf<
zge99KpCF`LQN1Dtj6A4+5^IK$pPf@Bqq0tMsOo!smff(|Lb<?KK;az42MQ9q@gQ{*
zD!3UQY7qyJDA3HXe#H4SeM$dKYE(y(3tZ8mDskn7Qo2cGbRCnzE~)WWC^nJXJ~Rq#
zf9LF=?vot~SY7pj28^EQPHxCo6Fw{iC~t&C&vC@6>DMU)A#6ISdqM#i@)t1MaiYFC
z&uT*#dShgDnsLYXvo@34kGrmhIMOn(f$x@WquA@Ua5_SJ-?C>Ztt>k((-(v4g47e&
z$Z-RP*D0yUbY+g_L@T;wGn04a53q1kfB4deLDv$E9KO`zzWE!43f1OJ{7ujoyjRs{
zrvwo3eGm^frJ$4(0Rmrnhv!_iEn|d3%OS>u3n6MzOiYt?xgp5%lp>GTus(&^KJm*V
zn6*6d3EIj&C9QUSEq3BtnC|uUn(6FQjGQ>QP7~Y>O^6sg8my7n2ITHouvTOxe^&Zm
z60HpevJ1T<THoW}xj*N-gf1PlJej*MH*lVGcn0~&n5lvh0X2uZ8Z&c*<<vdotqI}F
zVs>IEcLQiP6(QwBG3n-`*$IQXke!g^LKQ7BIzb&qZpHDE(Xd_MdAS4mE>MO;CV#xh
zJD}NQXvb!D4C)=39b0iwX2<ckf1TN}+50lP#^GK3s!~u^XlcJn*Q^!F0^eoKCGPVn
zRazsY`?3(C>3vxTr4brik&W+Vb>;nA%bIraw&FTMC%0anWqmCE7z0wI@y&X!i<l$r
zfz7Q&MjFE^*N;+tUkU+E6Q-*CNdycdgR51-5(?ucF}y+-?Q>1XqrFW{f2ib1>`c04
z%fvI&BT+|g=u*svN!u7s?zYs%eQy%K>eVRkr-yAk^(pda6s$Chg4HpcE8v)4pRdTO
zE;tbArW)Mv{PzCH?H!?r9O9OR$;-ScbOe3XBCHZd!IxB=2v(TM+J@$4%Q;JX6=1RA
zES13*Rh=$S8VVdw>?Cqjf3>M8zcQZhlqp}UHgn5)rP}P@mIhvFQD1LXQgz*SwWL#$
z?l{g*R7*No%3<`3-yv5p6)z*#smso|t20BFNsXMbQ<E{Snt0t;u4V2@<?5%(b*TbH
zMR}3dLw^oBZCz=_qsp!RGgXhTU9i4R1z%FFkNhrmc%|b7qO`9afAiD)IBH%yhpkF^
znMYkI8D)Q_P*xsgQsA_3QghaPfk|l%!h@adH1)WfNHf!YKH<^W_oOAJIi*QaOLErq
z>Lf6opWHGX1S4eNs6AmhXerx~I3soxVi9y}&gXBX%Ad5bDNj=fGl*ze*CkL5&}`PY
z+CQj01D4z0b)4{Ae-}Z8KtXQS@>c-_P=@#`x(M1+6b1XmT?D?%aa{xk`JgTWgSyd0
z;MDDM6|3C?ZG*6OI^vVX3hJ@b165t+*$DY&6`V6Zw&Lk3IM=ltk2NlPa>$m?)-!&r
z_7b#-5T*`tjgPw1b2}V`Kdd+7PB%E)crdQm%%xW-L?>Kpe-N}&O!*iwrCBlpy`xt$
zF3{a!U!Op55~_>PR=A)x3R_Ka$?GZE$NkimrUQcna+$Qs3pV>%H<K%926e6Uugb+~
zf6mQR$|xXvr`O%gTw-;l)FLQ6nkOUFEuM_C(~)4GQWSpZP2Tiw;hfU0V>|-{FDTyK
zNtUJ@g=X^EfAi|O>+0=j+!Vn9<rl%2sdIB!7p6uSzJuG<#pA0l+@*E>gZrtkafCrn
zU3pw>&j2Wac8(J>39Ce)L=HTC2D{<Jz^VT74w{FZ7=yIo#Q5xE43<INFj!9eF-{CD
zWu9Gbf4>uhcmkxt(L5QUZt-NC{r7cZ_mNed7@K{xe-qQ5OhA356JtB<bYg7v!A@+1
zj^+N}mJ?$*tvIoG+$7-(TG61afbLx4`BpT8C`<n%_B+Y}ggLa?3^KO+@k^S`%1NQ8
zQ?PpP;_dd05b&u3JT89bLI{;vT3H6LzsG^9zrdTnrCJJ02eIDYaP6(@PHhip(IF`+
znHZy4e;?E6Ce4a<$Ot^qRC@cZU77Eg%xPThN0Y3K=Afa~N4HzgNWsE$v|zK7r^)t!
zuY>Dw6tTLei++P6ochoxyC&qsfpp!vG%jnaF<p?o@L5d_R~nGys)C=<yAcAhTvTGm
z$7}Q%8GF5alUuipeCZ#Q&^I$-s^(Enw*@ZMe}=nWe{povwP!O~dVj^(4L7D-cYVFx
zz?%A(_jM{~SD3pKhP+FE<r%|p0a~?I^kdhXQD{Kk>hutA3;c>kr_xqvg9l!2W)P2P
z(3P~FLEcE~+3dx}*d^y%<)XtqEZyX%LW;oq8pqb<ghsXaJB<9igXiO}JAFRS@EEj|
zf1@pz4keq2=PUX>UHIIcLxT|xW_)9^-3KvH81x*X!p9+M%*h0U`27ip<Pg1^L0rjG
z5rjiE(=AzL0FTFjJy>gHq6%)vY6|OG7&q6=?N)BhQQupdw+l<sV(<06SW;f2!<@I*
zE<ZN<H2q0?4<o&&s$PI1C_2~wI&S9$f1omwvLKw2$9Q8#m!z`)HCwdN*orPZtE*-5
zHszbbL{r6Jb{B_f^M0xW2&4SpjA--4M91H_nW^@VzjBn5k6id^ML|V}{x@kgmCBd%
zG(U<dKONX=b?8D?>?KTgv}aZ1YP$OO{Rzexp=3Bw@;HW&5+tG;MonxT##tB{5N33R
z>di8ee^R#fe+_Hhq?b{+F%tncm(lPr6PH)zD=C+tjx7Y2KXxbxm&&FxkOB}6w_K+(
z1_=cG3W6<{QMfS^12#1_m$7X;6PK*eD<_w>urix}%f`>+-(UZF;3zO>!Nx!12fcfK
zJ^X&4;#<X;<Z^oZd5B+VC7AMZ(t&xQra%5UeLH-mR4XqIkL3RhpZ@qhk&)jJz`y}O
zZmlr`fPw?6da}kd1%wem{h4zE0O%LEVDTD&l!7eMI^Y2EiO4H49!7xirx!04xoViJ
z4CN|+K~);yGyt^$s7?GERAEe(AWD+P-5^P&>K)T+wiXVC00a_M4b(KQi@zKnERU);
zr%aFOh*|kBt!Nnet?`mFu;BcR*Yl!eQhBDmqqBi8NFX1%xBPYqi~;q|RtwOAQNyK{
zC9$5tHIw^{{EeF0e~JGv(($jBqa*6BP*Ns;BW#{fZ|AJTqC5SBsK+(u)hFE=Tkb$j
za1mR(LoQ{R1`QQlOrc*<k6iFCp)Wab9hJs|TY@Iw@~TKO8*P@q8n6HYCTysdpb!O<
ztv=bP-0BT&ZW@n1`!cL&EqqGA%K9*qCQ>>f=Mn|n@j;uUt)$_2ryfxXjw^UU-E;hZ
zEYkV@`LqT(Jm1YT<|w02YD6K@HH~+~NYp!jOkvoZa2`CH#}O@y^kx54iUXnWPkc{h
zMd|5)PnetxOt!@}jcV{Hr3J+>rjI69jzZ!y)zT1!WS-5l?x@8WADD#LS|T)Bi+;-M
zOg@hndh=@1)CRP!D8-lxwTzg^a8&w#v7+#hlU<K?;!d<KK4T2IG5JYnm~(+9KLvPt
zh*ZBK`x$Fl@b0GUS2*yR{f3b~B4M^8T#WB5G)KnKXIFA3P!U#AY7%9R2Ba|dZvf{i
zGGHsbS@lODL%7VM-fhT?)FL7YYlog^n$1d0vec=IUYNC#H1_uVtU*6utV9}rlG$iB
zr?Z!|dV>s3G9zb*2(c`Kv*3tsrVW|-N36v2ewHN_6Gq%%S=XvbLUCU6-HczgsFCuD
z4u#WD0Sh`?E!Cc<7^OQruh-HoXUiA^sYo%+gf1<dsj;t^oq+O%65CQS6Zv<r2Z;Al
zsgI5{giFBKAkfd<P%&(>KId6~kTc5f^jr}P3tCaCYo5!&RDz?k_pK_AsVqyzI~FL$
zF$+~MV96kt_hBO!ba>@r;$|Q`m?l&QgnBHcO2N@Kq@h><=p#E(KXS<kgMgCAQ!euT
z<{>3iH&Ro|w8$$My1<lZ+we0gPNH}FCP&5-=X|b3_RT3GhZDkhXLzN56imZp!bKLd
zkPADO6*zGxO(^j@Vtau+*6+rBlwn~8AzW0bUUf^1Aymxs{q96peepzv@{PIS+_|{M
zSY7rtvY9AY>#DKCrwVn4o}N5qgI{|BYAUVTKGm#0f$?4|1NK6WIc8*~hB8U?VY7Nx
z3575DZ(fdyoXY7r{g)trA4Y~yj<IFr5>ecISaJj8^_XM0%ZDnF$fBZ^h$Ww$$+&`-
zP|Sc)7raDUMUrn+WLd$LWRweyJIZXy9BjXdOsi#fs&~lbKWT@cBy50ZXs)Dqr6l8C
z&#+&!AfL;)K^H69<dRw5N+N@nbuI@#`{;>lrlG_n+-3)T*Gsy8)g*(Nj4bDzPJ3=j
z)pH6ToP^8HLB&v^^|h#s*?`PW>b|19YZV_q-F*&CrEx0`7??9*QTD}xG4C_($nOfY
zs6dO`<Wk%+0~u~Xu8ZOA+=sYS>&y7M=yqviogikotOWzdnQ}AC<w9amXr*^(R!5iR
zVIzW?duFX#M9>g_#Llb1NcngoZD#K!?hgx<vIWbN?TioimN^xXAaot0^-2vr*?8ux
z-|ETZ?Du=}9bTLgBNba*Jgl0DeFs|%2ZYi}@UWV3D6W>^+B<k)IM^hYibohiF3I(4
zf=3Alo3y8*iAqmr?;bob9Be{R$J{OaB7%#-W8cuAp%CMLp$`p7Z|?r%rb1a+)}mQ0
z)*RNGX8S1<?!*YeO*M6IWj9rK??4u}UkXYpAH+6fFCHJ#<3dR7EI9Rx8x{!Hof;Mh
z)eXn5H*CNYwL2oS2TYWh$+ptq-~*d0C7lw#qjDRmUFmEnTUaXa`CZ)&GAKwn`sVJ2
zoi^<6?t+ki9=46f?AQhI{5pGsD+1UtlvZy)55G?^@ynEE0|6BvSp9EG!{BG|FCLUw
z4HXi6X{@(#8HZ2*`Tp%MZ~p*`o4)@yPUVw^azq3_{mi$<G7jlqO+WD7$BgG3cy8dz
z>dD6ARSy%&gLU9-I)HG%-S}?wM;r}Ido5-6aC-TF>|kPL&<zzCBE>pd`2FYyi&Tw8
zT2mDx-bE_-Y`lf%%Tw_-Uj2QU3QzrOI((ZBUlCPfZHi*vO9)Ugq<N^m7f=;U3kM$E
zAzOx&kFO7K5vj8&+RXB3zur0s8)hGo(Gv1E*inqA$gQrqjCAWRLqMxgC`09wj<4WA
z)(RtkEfq5pVw(mO7|BEhqMH$Jf4}|2=f1;JU%Jms=Rrutw6W8J)S4;47-*&50^=9E
zEO2_6T_Q+{77F)2nah%Q5dy&O>(q&p*Y(gA_H{EDLNEs*@;Vqa$aWy`z+{DBrO~9U
zKnf;W5K>R2<tDe~sKA$MnJ%Qlo-(z(l(`#!b1pt-${QQsKVMpja87AEDu+A1uc}~{
z52RTshH88k=f~PA9uaSsj+fm_=$dWvTCX(Q#99DK?6HCAVw;kNttKQgq6_nkdgxk}
zEWg-31k=GD6{VC{c3FlH!<C!K@OemjqNb?lTDi<DDp_AiYA<6FySf4a^*#@Bqyj~M
zLTPl0dOw#=VcM`o+jaJ63JkFO;bc|sdwIL9{m=SXwT)noMe|ZTDxSX-jap>M(dggP
zI1MX?hpuGU(B3WFaCW2IQP}X9eNYPV*xLV!50``5;%j*pc`1*h$@}oxge+g8n!mwM
zQu5*Ku9Z*<`PlT9CFqO1cDL|^F0UAW>S$;-m?DiI=$+A8>S$@GUFEW}yO2utjcI;2
z3)do~w-!{9k;`*Qh|8@qvbXopo$@$*J|R|{6N<|y+xFV(skul}sj9uETigUl#P>8r
zc2Q#8O8PIRYhXLlcI<Mt|Gpfl3Ka$Z<^7@q%)2vdFKymx;|!80b_rZoF&DRg&!c#E
zZj0Buls%K&Q7*|vG;*p=TJg}BWLrodVaM=ZU75GUih=)XXLFsYE46yuw;RUFL9*1d
z)$egWd<A!{A<(;J+<o^p6de{FVrIGmK7APK-&1+32Dmxf=`y8V=Q=L6OiQDiy3PB-
zF;P$Y8K{o3SL%HkMxV4TZlsWZa~A59@E)FCZx$*mt@Is95J<(6VpNYk^vapux-*qo
zsZhUJvhE;ea^!Q20i~)9JJ-gdp1EN`>7g9FMN^$8HBB8K`ovnbm1jIPOG+9!RVt-u
z7U&m+zioW*8%NG?&~vair+Br(Z|s)YSra)Lr|$fPmFJHA{o*j{z(b&a>{RgaAwqZq
zdjU@v8{UeDRQf1Gwpi^1{D^V0%<3#`4`kxzS;Op6K#5Hrv(@9h+z%loq4IHrf2y^N
zSTSO$EZL(XG=6MD!4js-$DieV2#i|74ki3~C>q#v1W946C2{w=pW@l(ycngNg_Jh@
z(8>5#6qdJ!eZZC8bbCU7>FsPEf%?qzx)hS<ZrI6R+A~Y#nBbBK-U{Eb>b(K-Rf82e
z{j!lD)w*|A3l(X=re>0^bj<7^Da?e*!V-o_7ePMudN7z;P$RYvPj%0?o39wngb75f
zU+66LH-?%i*@E&@A3a)`+@Es{KYOD;rc+(Qb02pOLc|_2_1?!e9|A(Nr8upXhzT!#
z>P};B-$*qlgN+lXDv}zj1~+>}+MLX%qoVP_^ivG-wzD1J{2ypiyw#UcxG@s}HkYyC
zI1~Xkmw_n-CzgT%1eeCVGL!_+@9?y@UA;2n2?PWVBYBrmxG@s~HaM5T3q2G8IF&&z
zCYO}QGM529w=BpqIR<~^bvO8RD?5ORXV{S)ZUJ44;t?^bd?mvuuSMGOijK{`GjBWs
ze1HzZ_g4kb?8qDMY-hcqksZ!G;{Yg%_j(Kp43L}nV#(^O@7?!eO;%vR0?;JBT1S;D
z+9jWzMdydOR!#)hbKz!pS2~5)wp;Y)*QX*)D!ZPGC!p!*P6vOa(syt^_a^B(Hd)48
z{<aP5r@2(Y3ex)c3%vSGQbP*L>s4K$wi<QhxL--tY?Y|SChT2)mnwaY2N{w}Qt2*H
zk}#fM+n0Zzme3m^gP*)SE<jHQv>NB^rO1@)*hv(By-OEx?mDRYXqxre;z}JJbrZ(n
z)qzo$Q9LB-=2w3_jPhD#SYkAnY}!{o<Q)bCj9|`PFk$r?zzI^gnujMaa>u@EcX<6|
z+zUcmO*W%S(CYE$6}RgTkjW=HkN3dZ+W@kXlYIvjIn4M&Lk=5qlX%JyaIS#b9bd4a
zTM&50h{kaTvz11elg#_JXP~hijzj0H6HaTcgN7dH(MvrEe;W-&;BwuC-jU-9n#;N9
zIGwAp+83U5#U&2a21-q#1@F;@*{)tzkt3?cw4V7KkFDkW_Hp_jT}xwGmr=Me69YCm
zHkUzX9}|~>qd+H@aL_Uje+EJ{BgsH+HgpSecraq;U<Y|Ghx~quWRWZ`J&Qd_42(z3
zCX2Oi<nZm=#<!0T$M=U1Z%S{Dc_W<M-oI`FUU2wda2T{^b3!ny#P<GU`^$lIY2kB(
zkLFw{_&E66!?$^QRKmjyVc-L!LOeY>Ee`np_$2Is%op&9uukwTf6Ft#2uipJ<r$%<
zr@y}c`GF(CAKr}LJTt*rwZhgq&0vp;EkMv-sOMMC4eX##`27q!>4<H50(?vzkuUHR
zG31fLIuD)<;Hh1nuixO=CfqJ@&>F+6ZX5(0bco#IhY0H_(QAOG24N?h+<<MKeHz||
z{2iYo&RR#z>TFv^e=5H4CQ2mkWMt+nlg4|wai-zpl~LOt!0UPRbL74kD@vvG3}~7t
zRfr+rU<_g!@iB;#A7s1cqjU;nIs#8X6D#gVMu7QhNk(L46o1^&SY)6<a!@G7T}6)4
znt_^5^qh)-II3mZ&()E*E2#2OqhL=>St2$(iCq>*u84Yof2$C#&7TCNiKEQQ>Omp!
zyf)9v&u#araGK&D;U#5zPJEfnLMvvyE%eU@2C$qt5GB`TB>!bAS27I=>l5PX>uQ<I
z?QS~4C9`%!UZPe{s)5gHChjWDx0(qd$K9f5H6eM^)00pVsWgdZO$^_N!l&W}5`wbw
z5#8faHi;qEe`fjJgL~lpnbPiN_MLSiG58=`!&PhruuMSw7&XvV>RrdBh=qBQ$rjMb
zW>?+u@dSJ<!Gz2vo?)^8C75fWwb$QbwXW$f2G}V;rBQQ%N0GOnI1!K6_X!na1k-B!
z^5fz64fywQOH2frtKiM{?=23CC}3-?+oBMK2vPnIe;@w&?dzYu{3aMI_*Y!PHw`)=
zg5Q4RM+;j>ZOryPKZ6c{K*A<ifRGOol8ig>hIL&X__94{>6lQUVr}$$91To+EhSkD
z%v9dRi3q=qR16}(I$E~-P6rE8ZHGir#UegMD7ZF0!uql*KE@AyE>mIE-?xXa+ry{I
z&7j~Pe|%fV?TVQo6A=}JFtiQ@SgoBe+z5_GCagyZp><G3)d!Jlh~170_^X1h#Fi04
za2Wzv1wt7r-{bfW0|r_%OR`cLNbtmF05M^rP6}jeW^McX%MV=l4OV?5-x<!Mkcw$z
zw?`{LCv<|9QeH?RR|y+VFSF4u)$@}whh|+ve<GPJJMUaP54ErXD4_6ZBPeEC9~-fs
zu4WKq=EmqN7*x=j8Dp$u!RyMHolF*|8*iBpy1H$DP2MV&E~_RgqC6!iN1Ct*Obbe?
zXPSGZ5Nv(Y6kc+GjyOhvfzhX=vA~za;Tq!HPc7k#@fJKmRt4h^B~=m!8nUv>FKhHO
ze+JxBO{09nA%|-c=$;GA#EBCL=#@7lAsdp>QUMISIpVA#{VKAAN-03HTY?YWlZP>h
z$VnZBkv>sOGHW9xF?0=sBn3Oo5HqK%Y%&I?ZufaFHH6wgpp^icAeka}JMT#_t6b)X
zmVgwUtUFW=8M>G0)|Hgmj+I-{my$^6f0p3Ot}Yx?v(e!U(nLEh(xg>wP9aw5bxIsa
zA$3(MQD;r*O0g7;eS)rFDmAmC>rtLLy@V+$MpR{DQU*~maG6ug<(tpVpUGe#Seuax
z)g}U|**kkL+^KTb3u}%a-X)eTopnscNHy$R(@|Uv&}8)*j(CmCWlXat&zWXEe-{`q
z`dWD>i)Hx9bJjHPOfVvDe-(^cV9CMgA6q{Slg7uEHD1w3GD<N>vBowC<R63k$s?=U
z1hSVwxpDPE6=OXNq)xB{GBqjoTQ&rQQXCLEA)Jd&0kOKG7S8Kpjf!i|y6f4z8kMF%
zH%(y?SJ`oI*XdF|P(w!?9nK|ce}@olp?bluRV`yA<h)W<2SnYnM)N=QrXXGE8AvCn
zxw_^=3y&kt(#aBpM9;y_oEAw)hF^GwPb8e=kuga@tWG3UWf`PKW`a$s>g4*k!xm?T
zJk*VoJIK<3G-`=YqWx5;o2if&!fm57N|I1>L4V0iTlUemhPmvP2o04xe>vS*Y6pk1
z!&cj;GE5j-qU*lrw|3Q`jZk%vvBao(I^m$!npW$yUxwVwGm>WOoTlMOv3G>9*Pe|t
z)Iw`;ouM$#1oJXO*Zrw?%Oi`>O>iY0FNB4XThfC%SX;u9?o`sfnf9iKBlZes+A>qm
z8WwHC^JHU<PP@{^wVXX?f7-TADaC8r4T7TSQs(ruN4GMHmJvqYIWge|#_UKzuQTgH
zs`96VdLsv2C@Iy@MsG=xX8?ZU-l>B}lWMclXEKjiPo7&7yHlHynv>EE_K18oBNTa2
zd*UqW_Yjh%R8R=iPy5bKm)Qe_OO3Oh8E#h^cUDGumZQmJO|vh0f8J6qw50s52Q;x0
z)06ez>RN+a2IM9tp6ZF&ub7nLf<bjL(UR19aIuKD!kpivPy%Bw6D{p<CO}3jN9Y)U
zb{j`UZM1|Gx5XoD&Nr1GPaxgyd85AQ60Safgo2D|r&nQ8sbxc1`IX75ur&EHze{u7
za5QGi`gC-Y=fBnZe<#;q&lK8T`ipOiGf-i|wCv^&UCM=X*R)<|;3U!WW_kLe+}JqD
z2%3`fbj&?Av@_ush{2p<V7JOSz+Qu(D1ng7h(OO8ni3>pI3C8krx5+E;4K3!ZL`V}
zPtIL$H|0GqB^)R!_am<I6$So^N`=omro8KxL;#EiPGpBs9V1Qd^Ogvy%)}|XR3Wn~
zH#1ZaDI-yKZXRVr-Z__!*fK4DC(4?HcJbDV6!6_%nQ7)Wrd7q~pomuQGY!W~D@$fk
zPRh3Ha|34%%HVESs$>A^1q}!C!e}M%^>KK5eSi3Uvfy|P!a5hbE%8M@vXYyKO&jaB
zKmN9TefTtjMVsJ|2}XlVrvf}%XE*X0?*au5RB;;*)F@Q~QVgRv<sd2-k}jhs9<sTf
zW{62LGc;uZb1_0o*~aOYQQ0yV8yy;>$tp8KJePVF9E@ew_E}Roxp>3ai=5y`mxI|d
zB7b_ipJ23Mj@Qg;zKwI6;`PvyFl4?grhu*T4X>z0mdl(l#rNg!@=lcqs3RfezeHa*
z($#|y;mm}AT(_4cxN0!taq?17x1*vw(~ztV(1+e}oei0-(-Y$7v{TEyauw}Y7VlpE
zeiB2n%#X7>;HS0OC=CeuJ2R<jOb>mBNq^Po``I+KXt@qAXv(cm4?Z6vi;{}D@au}2
z;`wp|P%+=Gp!~8AzJayHY4jNEu{IbMwA#a0EpBJ-Q4IBC{Qf^u++%Qw;x)e4Ux|Vp
zUYDGR*BygUT}ePI)lys#NQwI6?UX~L4-f^19S65}QylXS>U2Fq<YrCBHPv*Sw|^7h
zqzTNKFzxoXbVbqyNVN4723LvMrTffM3nqBGiegEAtcONX3Zg4mI$lTeBP>I@A|=yF
z-!GV5=+^UCL5<CK$leC1_0~ahzE|}pq=eYQ1EtZ((j1G&BWkb3x9*AHimQ&G@=WSI
zPMEFR)m@Odf;QE0Jw3X~?^S<>+-_aO<CP|zx)`l#h&_&7sJ@lod7eIyR-?@$Ll4Zh
zS}h%-d41Tw&TPgz66FBxr|87vt-wq3!tY*%lwdo~oIOF$vOHh&DZ^~E(;nK@n!cl_
z9{jej6ON^lm11$%o*Z30>QuYRkgt)|ev}k@myp{sJ{yjtrFE>*dn4DK9hGZ$FG|h_
z!nmgRltQW6)tBkpG9VVtk!`f4h+D}3repH+mMCqiSq3Ql`IjHuGJ64zmz&%&C~g<h
z7r-h;ilHjF+&&r$@L`vq?NeIIgq0(-sUZRJ#pcTm7+d6X{##kv06WXNi+lqcP{G{5
zw@aPL0ixMC>%FDQEBxvq?=-jKVUKlUoV~2%k7sy$L&r-BrfnE^-0OJ3EGH+<s|!9Y
zot>O&moeQk76J2@NZm3lU^|y@v}FMt=ty_5DnJu;0GN~ILr2bNg)|2AX3`3|_aoG`
zl%@3yJ`;ZT)m$I%5r_Y=;`W<LMJm&m#jgOuzV3R(gbM?}%c`q*3pIIEVhYx=cNt}e
zZwTW50jwk3*_UBmGZX<ZmtlA>6qf-i4JVf*-ZB?S3pn#)GQ*K!J91$Ef8gRxQBoh<
zTEhk$s1HgkzA8S}K}t@)PjdS5a{74rbkR6j;RvjnK7N~kZ%F<G$&H343m}Y!>En;-
zr<a7@GB!gB9)AA#;YDV^Qg{=7hOhD8`^U?lFC%;~UU=oEU;cOrH&BDnPE7<-NN1<N
zew}`M`Ee9m2b0_<eUO~~`h6N&zGj#G-ZGDW`ek!QmDj@h#f&QIw7%SBTOGIoYLq0;
zS@YZTzsWkX!jPT}IiZ=(g%82T{?2)4>w+02T$!oX6NRxb8MFemo%M8$C0h*IV#}UZ
z(MF(k)hc;}gz!%5WE~X=0$+73dOW4;0d^(<pdiz08~uSDRs^#?i!E+$*7Eg=HOfGL
zQVSzh<*^A@S>qfgqL&LEfvwD~DzhS5jb^RIN6nr~7A-B&p2uuURU~A4hf8<<PUi2=
z?U}ikDkzn#BibZ0dTGAorQV67)qlr)|Cs0Z@~~Y;JCawzpkB9Q)w8o);TQRi&F5;V
z-aoq;cAM<bS^lc!3WkM;$A=@=Ez**I<BY>$U0KvtRMq;<2<Czx!iB^e9FcfW<l32q
zk48L=UI4GSK{sacUh|>De89qZv&C^=unR7ebFe|u9)2z-KMYD(MK>KCv#Wi?k!CH|
z99>T4h`g}IJe~{{+5}?ZvBOwgV31<!!o8lp6^C%lt#L7zzML|bB`v#PbZTUOE|z~F
z&%Kr8lIDCH-2+MPrlrfA&Lu7a>%&W|U(_+s&T@>)^^9~wV4?7BVl51&jBw8XPo&W<
z5{0+0Cz0k7_woLF)h(z57s(^W9lf|B2wBHygvY(OGEVR{N}7A|r=}jA7LstssUIZ%
zE2n<Ixc}6TX7xE6f2{6NRtZvnYLxbb2Sk=AeGS@SHz-xm*<-DGs+hB4Kz?}jQpfp#
z+#x0KviPg;RPtremmBiooL~XmfB<d!EwNnsziJ6G=Z4YamaH5*+E7QIbJH_YDeQn5
z2zWssUaeBpeq#FhG<uQ(Y%R{!d0ryakV1W;`-C!Tp+TqD)5>Ts6x1YtL6@DF<9}l+
z6uVUNSJ2nury$NnP+P#GVZVco!HQauhCSaNR9&&YGxf2$%H<Bn(`cj-8f*9os^%P}
zS}!4M8C`n@d&H)%>^nK7o`V(`s%-N#e(fplK>ug)I~1ty<AaO4uVoi#E%@RtoY012
zlQgT5WSy4<v2L|H*(Y#+c%hBviWvl<o#C+A<&@ovwqU=DK{61|f5^GTB+W-D{=JZ+
zRn66J)ALZ2=i<NmI5l%%*C3!v2SGbP04Gcfc*i^=I%W_uE*>^Y-3W^rpluYYg%!3k
zs!?0z{>;Vn9dbywQW|{&ZfS)Dcq~Dp4bp*==Iu%tyeu-ohaj_mSM#VR<&Ch^yc_k^
zY6MRyv9J4>cD{f4jM)kXN5XnN_pr5MUSV4Ho4%PdCj77u8jHed+(6SMUjW}5iP6GY
zT1P@%?(uz4WKt_q!b9_sV=0#Xbj@VCm8I8KXyJN)Q1pzE5QHb!j5Lm44R#~&?SVy0
z;k~~Ti)w3sZ#-px>J(R;mB||Bxm}EGPMK$u&ODa%Lb|pgatTqaKLBb+8qL==Z<zCU
zDgE0&L>%x=fKnT)%XC{Vs?SDLiGH-(IuEBFVDvVUlKDilZJ2Kl9M%XWw>;k|<LGuR
zsf!0OHkcJ0YcFbiy6}p1g1ilCFW+H-aO&&rtYD`u4%)eY@6qVqx;@p}QnK$_E@Rg(
zm(01Z%_o_<u8p9@t#@sJ6BUq+^mzj;WG8|AoYi%E8xaj1(&mQBgx9{<*SFKk)Q_Wk
zd;^Ed4vNh9F7=c%?c^xV997t^;>+nW+l$tAShCPZ0nfbE3w?BUN7(ID<7WP`Wi3V`
zI5!-nGgFFxrrA2!aXp+Km%Di`*0FQp%}x$S>37WiwHelEVx>PyeLrX_2Uh1?`=wcn
z?^2=dbqcK^C61?#uG^J|tY4h{<lkz)#)HU|WRuFJ?i9GSH0IV;72h5yGVim+g56xX
z1Dbvdy6)yh^scz}%H)Xl)Ph^lHsng^Y<c(QT%Rw0`wWF;<wV@3T3I8mt*bb4-5}Is
zUD&PP4Zd5=<(3~cyvJXR?;fbUHmYAfU*{ERPFKc(^HScVIC*~B?gH0#6wDJ8RX8=5
z{-8?elwceW>cngi3ZI*THPU(%?IoXBfpvj%CyjwM27~jlYPD&vQ%_2vl^yziEAd=)
zAJ_1ID$4=hz<kX3x~GY+c%BcS>l-EPLr_=GgQ-+St#U-NIU<pr9oy$V+iSAODBdpr
zbDf`{gV6z)EPes>Ni`#Mqb#F!L72Tu{&!wR&AAVr<q>;D(?UgNiPAKlI&9BzG6lfh
z4*O={l}ey=0}z1Kg=3I;*-o{53UoCkNr9|NBfv&VQl30jq$!NKdZz@86fm*=x+hr!
zw-$&}YaOZ8Q2m-uW7;u@roO5R*0vwd_Q7;MdhK0NcBycwGRK!$^Y{D5%iqvWIV_h^
zxG@t0HZV1pVdgR~0XUb@+$<=6U2j{t5q;0E;N!MHc{wv2eiVxU*-evX+r<WP3bYOU
zAUj$|i7mPEb&&n{Jwy4*@yd!VSu(IeqENd0I3#DzoFQ+~sxG*qH95+~3fW0SAq93y
zDBxGs<VwN!8ts9&B+05)m#j@9b|sJ%va2CK!K31;w#w)?<Z7kth*hkAR2)fBD^*8r
zkVMf@idK&lB@C9Rpk8L{g$SjDF%!O04g3V9>^OwhHlU=|A63Uxvem<Hf)A9)actu<
zJJ!Y1<T#U!#)6XCZ=#M%u@(_+Ye_anaS|e1z3POrVUgfOLTf=-s%R~QooKc(M4V(X
zEP{nB$?yq!scn4(z0}x$>P3`_VhjL}#P(FHi_F$@Bo~FPEe4M<*;6!&TCHu2F6Ko&
zw8bh~DY;k|Pc!z74FR@MS+O0quqwsos{y;Y7>%4JwYg#XnzOC0qD3K_B{ok>ThoNL
zwOWgeO$pV;)wVr0u9`8BU<`l+GV3MbskS{&c$&#jrRY*wJXM!}YHN!P&D2Iq*<qt?
zzL{K>VxwVNIWC?Sl+^YEW|6fTW|&V#H`YR2&SsAZ+mlPOaWjC2VD)TTB-Ts82R9os
zgTuDkZx~1#D!4(31$LEfqqMZ1ZK{hg@PQ^)LiaVbwgI-@R-+_Wxjlt(RaxLO2PoDv
zmc1@)t)fLK74yJ<a#9G*cTk%Glg$|0M*$XPML0`{0{{w2);$~!MccC+f)$%T;)6o-
z$Q;zXEw>6*7tZ_S`|rEAznzb}XVdAdcij)O`TOzwU4Ykz!|w0htL{g90iMI|&G@)~
zM=^WBGI7d2V?Hd=2dpP&{GQ9<lkVX1sJ9OPIhlUyo-Mn7Kkg^9X?M{5>&+|s_uYrS
zKfm~E*PTrI50^*&cy`vEjpqGidNdiIb_f0J_`~R;pB%rQ&rar}vkU+6^iNNpKEdB@
z5#xV278oFU#rDGVebo3M#2+uu&j0d${25)G{qpO<|1L-K@eYu3_7$uE*76D`W656S
zK|nwMa^4?*%uYX#=dJj1JRQ#`$2X8Ze+Q~CY;U@M|L4DEGgvOTAlP7fd3t)drj{+L
zn)BYHmJ%iiGrqf~O+A2au4}sP$Jyy@esDfI9y_rHu6sS|`|*70c=7dbU;38^{iq*f
zP1_GIX44*Hyg)}pplBQAg_u78Zu<fcCyuSVr%yY7*xK>IxPRBZe*U6+JO0uiezl3O
zM<-)onD*moe_=GyT4CodW|#Bh@x>wz*-t-@-%mzA%)Ydpw&<5DAENynEv;5+wVF{}
zETUo?{>i%*7!7#~j8}y`6qb+LL=JCS*A3ga-PCCDrI}ik<Sh~+9QcSqRn#NuftNiA
z?D3g@L+=OO%h}sm_uPGVF>W^YXY5Ue#cXccU{Q~Nfsm5CF;b-B8Bs`c^1^v@+dltv
z@_u#<U%Nb=j7~?>6F-}ubf2k%wV#jr$<gF=(*M@o=7<|E6~dQC3}C9hVppW9w;0G=
zeWtXsR5?V-!XRzIX93ICTUKn-?aBem$4d}@3iq%`xekOP^gwlFilBHPMv&bB2qI6W
zD=;=9NmRoS5yw1BS%Dx>G8H-mTBt<m2jR4YU`|iGbb1Twt1R<Xg@F}fjyBdkY7z%e
z2*$hNk9CysJ7z%uCtbMdrewmpLMANRkO^Oti3#5YV^F;+@bC-MP~eA?i4?O=lX5<P
zV){|4XA}R0Je*<GL!j&=Ac3etF+vUJ^l;;(620k81r~umW@TTDU@8ou%Bp#?=|c?x
zLJfL9$H+SXk-=(ET9L&=ZKCKc&50sjZP$K>feIphHhruS8azWX?|>^fv(Jbc=B^o(
zD}rsoN<=1)tmCmz#NcazVuembzJc6-30ekP)?{Le8?yz+JA}!8-1`9@EosH0C2r%<
zF<g5z4;p|6kCwrsW$<X>p&$=;Z7c<R8YvzHk62=d03N9y*cBVKYD$ujl@L-><Z`80
zwv~6_RS5<Skp!_0^@tdlVBesbB(N}gED-P_OM)X~^lH8>K8TSW1Wi~qXaq`sm^344
z=YZTC#jryVh;obseF*}(8mZw52zYixuFnNDi%?Bccsuomm5;GHfm54V*`$$kI?(_X
zw_OSa&-9bC?kCf4$Vir{)D9Hef+1ij6@du(I)?&P%T7Y|foP+>GV&3<L#ljLVVUYg
zBru8ZQ4?;3%=uBd?1paHl-P8CmDqGEV$*s}Y;j<(II!0O_To6}s-Qz*$c^HV8^wVa
z<G_nL@S@TVSn$fAke(r1xi+X||L1M}Zc`AIe6_KWF(Hz2s$L~EIs`+mV2GG`4wa-<
zR>-gtcaeu6ZBvui2;;%hMo7Mbypal6@s_+15*)B&)QUWo7=o<C-30`HY(bM*R{?<{
z-!8NE5)cF_8Ns-&Kwzgq&EEh~2+3Qig>ytEr>XERR^rMqA+Ds@Hh`yVT9=aT2LlKO
z(m)3cB1i_(H!wi-tHK-v48e)A?f`>n+iIts3k+&RpY@S3sK5~be?Wl0#bxpBKpnA?
z?XXrn7kMNMjUz$?`Fb!9IKp(Ae;uO(!|EwPO{rGNkTVfZUA2@QBw|9`yzvTxu$@@&
zItX7KEZr3Rs`ojG)YWiGNqvp`h^99J=(<yYO`~ZJE73HSZK7%B>u9P&5Y-`wrXjwj
z;Za&X8scjj;%ge>YswFgoy_X9UG&5VUT~bNhpD?{l9C>atfLJsG=NZMe{gqM4`MGw
zDFFn*=A6;n-I*PmfTWPy!eP=QM-02Kv5T)EWWYI*O+6M0Mj}J;8im+#uzUl=KFHEe
zax<)d0D%k#`=RSm>?g5Q*dpXV(vNmx$#!~X`Yt1ZxJ^>+Wi2o#<oUcrAqg2Cmv5lh
zi*aW5p}-ChN+F@09U`&qe@L=CH*OTbEIg!oN8<KXg&}?j1uwFf0CMB7cvt64W%U}W
zY;_G)^g4j#AvokAIOHKX<g|l)3Id*p9JbDvSXxeFIh-IBMxL<I>EHtblaYVD24P2A
zv=+mDyq}XLPIjBmuyiMi{0ez4%*5lT)q9;#YOP9z1(`C)E6@&te<Q{YsSFsR5*&cn
zVn~t%IEfUgFcsvW@rKhCzK?P%x7_Zq_?pTe<{+f59)#3w4npjel55W!jx_Rcq>+as
zjWQf*ltG`Bp>i22m!WbUD%U~z)j|1H*@dcum)OfaVCrDU`_T7V9poh(Ss%%P3nFE_
z#vn*US!6AUJ$Y~-f8SnYii8!(iwGL96dHW0t03?PdaF`E&Kj+(Wd~9wRWC|Q5E0oI
z>3R^HSTe{1YmwB};Z`2G`(T(WI26VQIfC)QUyuv1b$mzG7t`JkhVOweKwZ0N91l+8
znqVT_&Rjt^yLtELr^}yT`;VvH(dq1{I~!eK0$o48=-UJEf75t+fzH;v6|{xl2?;h>
zeF0Z+wH%^ig6fsFwYGcQ0E7Sqwj9<D{v$7keaJFSroT?6lYTtIR=>I`W81fXi7E@i
zIJp?AFU78lGEC9#NUGPpZ}~rFpGKc%yV+|K1NM#|BR<+o9}#L8;=j%4wwFE-DWb;M
zE^`Bfd!ZeiZ1^I@Xje^C?U=bbGQnc+{RK;&ZebR!L{-Ywbqc=i<g68g*LL}6Z)Zk^
z(KAF7k`@kiGY}>|>RlH)Rd0F2Eqi-A`0ko^Pie0n++CB^URwCW<Qh4wD4ra)Q#{vK
zKRI0Kk{lMg<p03r(3fEu3=;!5H!+vtG&vNPI)5H0m;F>J3zy*WGL3)H+uvV)-N+=c
z6yAid;a~ad@n!REQ;LtE1bS*-KQ`e7AqaG;L1Tqc(0u=+dD}dfuw7@8+q55AHs9Zy
zqU3X8&`V}e8l&}yL6A%<HO6>>;Z@5ZJxHmU0owD!+7mOWV5H@zy@N?al57iIB5LMQ
zGgn$C<oJ7gV`OCC^DloDp0<xp`uMF^DwlkIn%Br#L0D_VeQLHJ*}_)G>8rS+(1SgJ
zt0nqW{1v_UwwI_w;xhCWSzBls>x5Soa+obGN<yWC42iX1F_Y{<LE!um^%#{5^iT=}
z8DI7WV_Ey!T3A(cyFfO1AJ{ovKC6UQs*>*^JcadGjbt#UFQR{rN)C)h<v_6Cs5rme
zng}`+mdJ7+=9Ib}@RYDaKan#NYn{p$8i>y7x_q(g6=<t*5InLD`e4vu89(d?&{xds
z$Dy~8GLRv6mwO$}zFK2r&<L`w>rE%lZL+ssK3lUNn?dx$+1Y*%GXWo4UlC`;?(|uK
zPLQ)@uk8BLGd+KMH3YjBjaHylX)mnyB{F3L>P1v4YlO9xQnZ46P78i&<i%!I-U-%c
z8ew|!(n>=%2O4(08vHc^r9{ELXw9Tzj2OIs7zdH1h<x&Z2V(>nRU#`nIncq9Fc!+%
zSjo{&K*?Id<IqwE`8$@RX$~gF(MgGiu?!E`KsP={C7*wfb6)hSFgYJe&M5UvI!WSY
zm}`@d0xhtlr`X%))KEvKc6Rzv^aoV%a|&XcqLhda7a)4!dCr%JX4=Uj35BW}1#%Sh
zk0wl~GwHmj2j9#D$?mMIJ&3rS>*8?Pe?$>RB@+&XHDVL!o+y4~KWL(YvUg$q`T|q5
z%(+$@JlcP301po4uxd<geVZ**20uj)EkH6mA*^xkmfcv2+Yp`m<>=&uleR+F>!2eq
zTFF)DG=aW?%i~@>t{6eafM+@vATNMQ=?dS{r-sOhs4G>G=g$I3>^5N0=(*ei<WNZc
zqk~S->xDt5KwA)Wb{7SmqQ4b_PKkJV&?ylQ2c3VC!P65hMM_wLHMt=pyf<rd)2?Gm
zKs~eQ;k;Sa6Xe?Q7{kPb$|N%Z8Nmxw740hpWU@*~KMKg+N5*pCm1)x2e&Pm)^*;q(
zGQeJ_54^kHb;^AgB5y!RoO(^GL{QdIfVGtn`mhjsT_Mm51+Ef;4humCgjgZgfE8;7
z`w@TMnLa(o`VrQVP(P|G&smg$@SMRd>N721$K0KtU63>a364yWwJnd11S$N%Sw_+8
zg;_>{wjj%RNc(kX?9-*ZpwvKKoEH>`m&E6kJCukQ#OJg-St8ElE-B+EYd?;5$j!h)
zHt(CZTQ_|uhbx709*3IO1c$BI*Je*7D?Wc$3gu{#bJsqU>z{OOjYhH0=|lX(s%vZH
z3{jJ<(618@k>{jQe#MAbAhz<K9TN+*nM659#ZsH|V-$i5zJ7@jrq0YCU#cT`!+;XP
z3H-N)>peBGGPh^Jb@Oi%in$HfH8IF=gajXi-F*7{{q2|6FC223_kThQ*|6KwzzTnV
zv_F+%?jg+Buem_6KuX&tgn2Z?B^02!N^aYiyfzzBmcyEM!{V=@*t0t%rH1X}<pCS3
z=3a(nS(Jd<73WUB4hyqveJ@eBPL%K~prp3pC%4b7!q4ywf9I;)>d$8L)@+_TUOn<(
zjNm+VCurq$HCTipwYf-16+$7IOo@ND>~H3ZhpP5<g>i<MJmrMtaTGF$)L*A=E*Yqt
zix?{-<f2ZvqI`yeLfe*<G<i5eFb=Takr)OjBxYN7`|qzGso^^}dm8+2fjE_C2d_<Y
zqZ}tmGF~@{H$$|Wyjzp1=i2$G<8F)kw)g5W`sFBi7bjotd1|^9pe(|3GZ25VRrY3}
z?ek)a*q90y5LCLSo*2&K!i|8;%gmjltb#Vt7v(6ti?<MNNV}9Z$2jP6Nz2z(#Y`fW
z+F07XzZ)FArE}dFE%zO!>rZ;)y@&9|<798K-hD1j5yN(GA-x%u#5n9m^<p&dBs4<}
z`epr;Hk}rCqt=JBr(v(%S$}_fN6V2;9LM(g#zc}L=4~Atqn=EPly;P(RkkE{pT0U?
zj;!qy+Fdoy!(9gFf?Evk&0&sSRq1GD$38B2oN<phvhCIcJ!1A1<p#W00~@h@6v4vV
zG1|;;UMW{;PTk=z_o0qhJxM{i?$GLk)<>+K+PdCaiR1jn^liH%@pgYU#anniAF9fa
zMi{%X9AO(Oj4kv&X3?!*OV<jrZmzu%80yA0GL6fWzvg^(@r}PU;5dWV-E}|7Lq>+A
ziWfH=Dz19nPZ{n*$J{`<Rkm8#4Ju_y#5GRl9SEyci9B~Q@0mpdW$kO9%x6B?hG_JU
z+P=t~k9$3L`@-{Jgi3#h>z&MFP^s+hqH_RFS%E6V3-%d^y`ymz;@m!i7micvauCS#
z*5heNf#bBhKgIp|&bl$&`txyjj2503pMh>HFHZd0XWbc{Cz{4<AGBtJBk|fyt>}{r
zM7=v>BbtQ8O19bCQ7BYc>=gYTxw5G5FPkVFQ;o)-f1&^{7CkMkeJLcPONf@d&P3th
zYpq!{$0xIz`ITb7CYxF)c5A+NnfO=yEK&ah+R+ZnmtkEq6aq0hlYwy)m%Ww&CzkR7
z1ead=GL-@RxAgiln+60O#4~@FVO=v6m(f=O6t}wpGx`b&F@NTs{(k{_yGxg`)GZSO
zI5#twA(|Hxm*@;LEPrFWZSSu*KM;*by@7#I$B75%Zi?9zU0e)0)5WCdy<PP8OG%WD
zr1Nl`#BBh_J~|ZNq)16}`Z>wz%i;0c;nPLw$q8@3;q>jt1oT4Ee~>h2&Ex=tRWN;f
zng04&O0@KOq>q+TDf&42?djVb9~5|k07D-d0XRN54LnCh7+9O(71Bo*cqDi{(+DtD
zeoj~|L0J_{#CxKTi@1+}`}XI9WD&}D<r|d%6DI4lAZpC_qcq*D3awB$61<0&>U^X(
z^urLV#uF39#V2Ax6N`peHLi?m9e<bV4KpQw-_VT3-!|h<&$i-7)R>=&IdgZ!XdXSw
z2P|<T8$E^mop4|Hx8dKu<<&rWQW5sVAD1<ea1|aeg@PnaBrXEuFqO7Hf{mC9_mXA5
zWEo||VQp^#%UI|&$Yk<`sVxfWv6wnSCXE$s6b?0W365=(tZ`v|c;+;@aN;P=c|IzC
zQ#6|9;VJshj6LB<UQ#h*iz$T)$%q#Y0FqsvQ!sV1aBy0sg!8_-ZLu}cN2eOCCJrx5
z?bLYbDMIST`;rD7Lm>Nvt@s+7R9onRFG9tri9BRV6?^mntWVYR1}wJlj;-hVGH)%u
zT93~>MvSe#lfBM!tY}xI;rL|ix{Tw06I<X?IaUpa&qpY+k^H-AdyH@%UCLqED{RqB
z^ur3zTA!s#9Gq)5TJ;13;gu?$P$NDtU5`U_B#F**G)3*il&v8`qGsA&70o0Jo#kX>
zjgp5<j!X2R=%Z3E66cYmXU`swIn^_l18w%K{LW|Hujhmc)#MVU%~9RKpXX?QPgXcH
z??D#d;B3q|T017*iv5Zon_h+{zpi1ZeCd*fs?1CNv7MdXta^bA!n>Le*0WX~=Zx~{
zV*KQV3C^mu<==P|%2mIL7HeZabIBP``$5J$(nb@*oST(Z`?Zy|;)+>A>Cnb<k;1fb
z&C=`_GsY?roLkG6G5>;k#92dsiH|w^S3B3Nf97=-I$ra;#!q`)9TB?~!VEk#S%E#U
zWaM@ondcX+Y1hh7qPk{E+0+Hl)fV7ax#+BxVXovrmgm&TGTL=TSG`?XZ>G=L8|$CZ
zRuKN`o#B|*pL45|buAZ*=n{e5nzz)_8kRxVvu=Hpaq=pgz0<>*ccP|$l9gg?kOtZ+
zlDX4iwm}+0*3r4<gsm7*QVWZALFFkI#v<5Q8^^CN!_)J(!|w~#Yp(@5JAHpS#1~qD
zP*hcEN<g5dKmI=bID8$kwp~Y3g|ya~E>J-d=Xx?!#cK?8IeX$zC}~?W)0(E$e}DX(
zhLOMYl|DrAd&~%ZHz8zyoND4f$r!s4!uaVD<SOr4Qm14_3LDN)C~j)PS^*Xp{B}zw
zJHS&Z@cJMrx|ARcz&&)?V;*$5(8~aQU6Px4q7KeF7^%PpPtJ>g3!dH(Zc<R>N&);Y
z@z{;DJhDXytuUr8sxdnQ<{r5=fn}LSPD!>&RScgJ#L}8^eTc7ruL)Do1NUZ0FqbQz
z)NIGR8)0llX}C}f`4eBqX%IjB9H*t}E{~;x-88=gqf`lHsyPdIM57qYGAdsE;<TZG
zU(YO*sTO$(oS+07^VIJX0e+bhG$ByIo9RDO8YV;0Uo@T7<V5t+SZ|{>51;<^^T(gQ
z{|17l{X2qEvJ+@nAo%Hp9~D^@jhlY)vlXP)Ks<;*u1)s#0!T1dC)R=Q(?O#n>c%&t
zzv5^RlwuK^$tlIgcCQj>h6)c6<2=grezk*@t4bm<UD=3F5ev_ak2Jr`ijVQrmtGPx
zYJb1Kzwo@DH0w+Ao$)-xT`z6y^q>tH#=D3sEwfbJuhqj0UZC50_?&9oypUunYtc0z
zAaY-)hMg?e!z|d>S1SW3Ba_vxD`U**jS`}zqa9v$cQ}g-3_UglwY+yt@mn|Va-k=A
z^jynI(-oOXp;RWqu*u}xlGuG-D?5jFZ-1be(&A%NRrs#0+cVj$oHdip>dz+o=hTC7
zT778p>MM4URS2TlM1CCAQM;L5>8D}ua_+n4C)?~(xpp^-74pXV799xZ!&u)!)Dz{j
z_ZGOySz&{@Ls2$hzEYG8(55W%Z9Q=2eVxY-FWOMXl)T`1qi93GsOM32H`Q!$JAd_d
z)fS^!L$#f)Y`UQ&xdo%4RXr>)%5`G_MUPb@wDwK&DZ8waCA6WmFI4I#e!7vpUdE_c
zAB_}M=r@UZOXF4j3|HI(M)RB<scd-6U6Pbm-tJKBB7<6=21DIava_KKEoJl0k!)GD
zyqn<X{fb4k`UZ-zqyyhkEZM<)41Wd67~YNq(w4_5C(n-{g0WJBmWr`WS~sf-8zT>C
z{Rm#&AzcobucXTX`dYe7d}6a8bydWeQd3X#5m}Y`gAf*?ZDmy|;cy|@%~|zwb2qG_
zZ2?EhR~)_0R`D2b;TiF;>>U{%eTaQURhfQwh95ERXZQj7e~{q^iXV{S2Y-zB$?&6E
z-=E=;*q7F9xnWfbmO`6Pw$E4c@>+p-tz4$gwoRI9tsPKy_k^cHv@J>+8P5iJRPXOm
zliIb^BWLr8?C#z(QZtUYoIP))@u|q@cpcw6m417k&CvA(=BS{>?VH)A-M#1YPv=q-
zcm$gdovI?6<IAct`ek+wbblFc>sm<0TaFmg(^z1qOAZ<C)Y2|JUpdg`(R{G7c2p8g
z5Oh=M?;gSLxdJH4RWvkU1ndK0e12U3U2!8?*r<gaWf?El%pG-qO=HVSuyc6_(iNsx
zo|dAdDAC<zsrB>ra>AKhNbfSIcbI$4nJ0KH<gBdj%riwscBSUL(tp)A<GE*!q|&2j
zm~xh<VE2YWguog!up2b6XAb<#foc`s9;@5LdW=$Bxn!^^#Hm@ETq87e@KAAfGdtg_
zDs7Z<)5Wb4^QJv;?6@7C{Cu^FU1H$^j%ua(K&qa;mRn`bx%Ig_6>iVz^ZRk_fa=A3
ze`&EVvvSv@obtYPZ-2%)iZU~9&3KH(&wDp!KnJ1SoLd%WFQxKwxsrf-PS}mCbQ%L@
zRJ^bcCp%=>H`kmtF0XE+v{iHR3F81wY^A7D6!2EU=e8B$y5DbJz`41aQ(B-}@F4wq
ztD;dj4I{KZ&6|SEv2Ctg<n%^w>eV!kaK5s0E6++=Va!s?Pk(%zhU@)GtnhJ%cjnF4
zRq!UBv=&}3d0KDz6hVPpXnhXZ+v$_%rA#x>-_S+$JO~T=rOPqR4tUA!k2mu~-YLMm
zmM1E1Y5potbJcP?4+uizYwe=Z2<!L!d+mH56jdZJ@6kQKq5I7((10mcX!)~wcRJ;@
zepadF+AEAXDu2szFs`!N-Oh0U(v-iEV~jR7u;5s(>lIy!^E<2rJ?49^1U;o|D?tza
z|5*uo9<HqfJ#^BZtGRv|y9*lw<?{LL?;~y8W4@=2drH@A+(WO}_<b3###m2h%Z_<B
zXEZ}O<a{nmcj-pPZDYNW5+x;46+R**_LyFq68l+o4}S)7<Ijp-O_cbtiL#uq-6jXd
zWy>z4-$;}}Q!c(k2z7Yx7(gAJE8)|@-Ys}KD*K_+D_?TZ%o}TivuU$xLa&5FxPuLQ
z#CJ4ePw0vjd*Hj7u_v={#~wBjS9KxV+(x^dHDiR4qro?ihJtV=jJz#3?be-ay$2t^
zsl)#-3<5%4m+`hU6ahDvaTg90m*@;LD3_8LGn#*JHN?Z;-~Rd}iG<g;rx9hRv<PAP
z_Uq}-3Fm#8^5iEebPRU--;{=N4bFAW<6RqoIQ_;cpZ@dX`(M8O0|YMkZ(1QIBb0?0
zreEdJT3i;Do_^AE5DL5v6NrfDn#oB43UC51tOMVsC!@RoZ9_NuGmS=Jf>EkltU>7D
zabkbnpc`Tw62v;n`~B<(8LPHMrm3Qlo)Q#Vn;vmJuS$>U>YsTkuKL&X^nH5zTKO3@
z{^8wu>DJf@!hlCr>R8G{+KcF;am<eBrL`NQJ)w0_2^X9}#;G8@1pHNZtHkn1M=%c|
zEQY{Cd>_X*7%<RUSW?u|3L9~q6W%4I1mb^c;cWZ;?H4Wkfs4L$_X6e_^+Fiyrf1}@
zKxrVTY`>@*YxYaCX)>9wJ|1*3CWD4%(a<a!ojKF}m<@~fT{EybOhn-8)Ztg?e6ECk
zy_q-YW2`kdFo%&^VT!iG=*E=|uf`M@j}{VpuV&FJAxEUug(c=PV-1oe6XViM^h|#?
z-94{LZ<D991ShSv5$|cH<+sLm5J>=erwp&oI`flcK|y_oPsS5$Ly%m7V0^_w5fy><
zFa_5PS(R1EI|r*0?Y>ZCLS3KfVhr-iQ~%zK+`8nq;M43A?c+z<Bw=RaCY&YLi$V^v
zBT19&DlcqDA?>OQPE{(0XL#u)Q{I2cOP$%#O3~?afrjW|yyMXs=92b<M!7xLnP<i%
zJ=J-JwBge4#eAogCz5N@rMk1N6xMmn=>+pw;O7~RVf&X(TYOmu`jqCJRw?RbvCy44
zWpo9Wbk?k`_&f-3!+$hqP#nm=r};FB^3N^I-;l;QX+<=?8oVJt7!Q^8)mVSE?p1-!
zQJb6DL}~`Sik(H?K^KKHF7I+A!z%MwbjP~iiD1Wmo-y@TY!1gX8j=7$T>`9eL?eGa
zxRE2ENCAnm^lUy8<a8;5&KS*%pXMhuaEI!H93iqU8UqG1GCd+v8J}6YU_8(a)BHr;
zXPsc=OZN4{KJ00X-N!PdBqV<<RyTXv+uKSfz%EyxJKudF`^n<l#f6pGMlE7g6&fp^
zJ*ezg1M!RF(-3FK*Bv`ged>Ks^MLF<S>WueiuYjAH~*IB?Mnkwe4#LX8fZw_ey$T#
zsT0gh{dT#cd=LN&uE6KwheUMA6{Xvk;R$`ypNs0j8Z98G<X6&*l1qQ9NeA57Ouzp*
zeSi8oLhZ|`Oty3mlhR&vAoF_b5gevREIJA=RT&K-ZBp99|9=0NhLJy73ItkV(cnhN
zYVL;XY-B(B-*I+Uk<b#1rcx;N6XB^klhd8gJEb{<w@(EC(7%lLY^1{*>zu^B0TF3C
z{iP@-?J6k>MS|A3fN+1g9~Del;|4C(OKjueutalVvHF~@`^CwMg0-N5P_^02_P!6I
z6(HjRqFh%HhyqLQaW-cn+1@4Hvo9tJT}*j;+FjX#&uj$o`E-eL)ikV5mHD`%j%%K5
zhV!Pep+|_g=>XBb7T=Pg^|?5DLMQK*6f5w+*3W~WZdb!9D1(2+x{X_EIGP6l6$R&A
zbzOH^xqJ+mczKlhTBwmx)UcJD^DKGhz~bknk~V>|l=3TwkdE0k?Je^<oNKvaa}s?^
zFf<X(PPPiyAHc<zl7=Z0ITvZw7`T@K^-yQF=(iWcz<FU|+<99@`{7F}VV87lbEV{u
z%9$>IR8~^{XhMH02Wgbb3T1P0@H)>kY0hz$bnWJk3hQ#r9|wHj5x{Y}@G&kpK#2!u
zpefBL*mByEidZ8PRVhZPbIG|2o!I~s=iYTCCo`@UERCB5?u;krN&)lFFgDCp$)$a`
zk_u|!ocbWGvDyi57tkp9Eo{W2Q#aM4WSt6yR!Z_%BYc0K%4Ir_#IbYpHWbztySUUN
zbIlT45!+JJSwXrO&ANKzGPtIxb^o0**tKR!H*#uRO_~oISrK3YfDFdc4|A!OLYnOr
zn05K(3boi1HAK51u`VN73S4z{tt6UHt&*+PsWG80>Du=$m?XvMM?B?mf0Y3&Tim!R
zTVUC}D;|H+aX`*tEx*kz{7~BYslAj+UPVU&v_6*ln}Rxy+M)-=Id9aMHMPU-bm~@X
z>w?RvhEyfCuD<DFtaeVIH7w+I-FuB1fcT-d!w9K*(aNh72T6+Sad5YIHCTSiu)udG
zI;3M6OS)yax#v2+zTr_Im8c7i9=$Ndmfi5kwt;_<0z>6HJsr4o+_8Xm&5|VuWtt`1
z>y|uUk@5&@+12RlIYG)|<pRvUVpXIziH}i$HzU+}H*dXjY4yA#P<nMM^LMLtFH4vi
znDa^}f2N$@8M*IHq7xx3sd%L(T;Psr9|E$fxSjr};UGpWw|w|0x7yR(ftxLlt&zlC
zyp?~pEVa|e7J80`_cj?w*|f$K-<#TGkjX2hoJxdLB|O?>&{lYtPSb*L%70{&!QysW
zkC3;T3}~_z<vh;5!O7l(6WWILe%xi#%!xOcyE_+pN19KYZE>9W1Ud(`!!u@U>40QD
zOlV)a1~?j_ld7eGUBg#VI>Rnn_a`%$yefaldFQdyP_<v}88voalIlzQa(!71O;j*;
z;b*}mi|M6|*SWOL4(%Yj9Yxo9emqjfD}!s}IxckL*(+tc!-o9&2Fpdp>X*A2xylKH
zcU!-I7<D8V75ej3K8#`fIz{N(v{mr2V~+30y8{#`sjo-Zt?GXx?68{xjV9KD*=Bz>
zKhMYgYCg_2*W97+R0!wm&14jkLy87GAFae)5I)!uT00&b57D|RbIFm=Q5-u5e3=~&
z#+CzjM})5Cx|>zFH$r=TJLQ`Szj^SXvChY;eAt29>sb4#V1@Ew)+vG(z*RJZmXwi2
zH=@kztDVRvWBnlnjdG2&wQ4vT9=(4=TbC>OXjhs=eyv#)HT4`;_GHRKg*Y5&5#^zo
zeXbQ^?Lff3<YN%X?UCd6N*A-~s9=O5xw{m@Oo!iHcNyby0x{}|W)2@pC?IQLkDz)^
zx&Dxu!^CQ^w=x~tjGHw<*F#y|mFnkm99-Plx{`a$@s(~_c;7xr0?&=7V;z4T4oMz8
zk%AC~({4PG@*vXpNV~}*HQ=o*t@tde*~lN1P;Rv5=`aH0KBfchh&GJ1>}_p%?GSMo
zp>e%Bm;IXD=R#rJFqQVE3E9naDSI>Y<XRp&$pVFPLi1AT+p}5QKIKi>Np(fEz5^jB
zCoJ_qT^0~q2T=-B6=g;|E{}f`1bA=XeoA}2p4~tVyv(h^5#l>=!;&d<LE7FM{8kP~
z?x{UMxDZAxlw8zAcb!8@BaDSL+CJ{{jO+vXV>KA64Y*Q$Y>Ag_5uz0lbc-gB<eu#H
z?98t9@I4M?OGA&o<({*TAxvNW`|7&qA2Q57Y=UK{xzEz-XMlDYzvzEhly|z}<BZn)
zSa#b=P`T1^(_k@Ib0g0_#{KlZ=gJ2?E_dn%ES4XdtT(I{pZJ9K`=MS7D-uKzVi?0`
z@*~l{`-2Y9>GNXUcgC*yNa~g?ASsu7eUGQ|cX=>}yCq@GcUAK7<7+HQoQDzO$9S*9
z%y!g=c(2F&QpCKewF51>+NF7qRFkaL#o=Y8CF}c=pVz5H%Z}^N_0zAVADma*%kL7}
z=b8Tp##9`smtkEq6ah7tan=PCmq5=dDSyk3BexFS_gDA_M$v=R185lN_KXAE-3@vd
zWECV?WpVTWi%Y#Jsj6(xv=>>FU8Pb{6vf9OC3*VmBu_uTynnp>bTx2t!lQEf^zr9O
z(F;kxN>Zc6lT%7qqfQ@xpMHInQd_#-=>jQ@q07;|r+++uG0Ia9g>+GiQs*xYRDb!#
z#6C9nH<~V4`)lg2FVvz6)$*LETIyv@z)1IvE*JmLzkU4YMV2W@5p2ATfAiC~kC*?w
z%oDOk3N7{N_rG7_3ouF;uTKz!SLXE3AE!TG{yS6U;_`~bC9Opq#HA$35Kh(xVdASS
zF8C@X(nR1&7flKvGcSNXYd`<|by=EPe)+lcZM0C@&Xl#9#319R7=L7;IJ6MOpRPe}
z@(z-g80Qfp-)J-6NQ(ob;CCfqtx#4k{Q9B@Ix7O`ES)5oYoff!5}W|zo*~PT@+8Ye
zE>rK@GkJ)YZY4850(p;@t|c=se{aJR<~c=1;9GTGNgBF6tro-jL6nNGE#S(;LXlO`
zY;>&$69Jad<3CcHytZCn`?&1eO1(%vjqmGR7&lRn#gZ3M8BMMc^eifU%}bFa0eOrV
z3J}?he1A@ou;o53w`y`rS|FR*?5!DSBWT@5MwEHiOtn;@fd1qK24~IIf2yN10xYyu
zj~d0;O)OuE`1m415^7uk)t<`Ku=d3w;@%hg+q%q<m+5G>jaNkSuFU{VwE)SM^unHF
ztaO;hj(7I8C`FM{pJ#h33!j(Pl9utTn{@^z4?oh6cWM5<P{M_N968jdJ#|iPjepzZ
z97k;<da}+)@%?PnEAt3de-2Q_zjifCCgQ_FugS%9I(dc8**GrK+UJ>a18%ZPl+&BW
zdnl5_&O37)?~`nXuvga|NZZLL*DyS|j^>K88~xH&ifchPa?f;GcSjMi8EAOTNGs{|
z=%y}}QqtvO=va)nJ0E--T$*JzGCIW8M{fu5JzFr+3r|64l8JmffBuRvE7{8;Nqg+>
zloM8)`Kv=8;ki4z%a<9i1X8%)SmhpUstu+wYXGe})b(JH_3Do*jiLXpPij7N(MOHK
z=&&u1Vl#Vw(;Lobemdf__uu!O)C-D?bfs4V=w*^!)M2lSRc8xm-&cvJF*hZCVLI(N
zhl5RKYDI>LZ{KB#e}0~!o;_0fzTq3yEw|q1SpfH_J6uHTis4ybTOTvP7FGQ#t<EU)
zMhpfazcrLVSrd9nlnB4l<qZ}bxwLXDSyf0GVO$Hl(VZu?dGEq5PO~j(5=~Rv6f~OR
zZ6jmrg;j1IYe%UeAfbc?X{r}=uEfQPHIpdZmKV(N%Fnu$e_K?UM4d=(E$f@6dugLh
znpcI8*4lM#%8};*_@+GwN>-++HuJQF5#Pf5u}gL2u4ZELdILaeVS}4FXh7h2M9M$B
z3+sqqOqguwkOxZ%4rK-t4_GB}%!6};fX)1|TRly)cfEyb;;PB_hBKCRtoq<5<H&J1
z_mr0$x2=rke`+pgN@icV;<-5lXQitsJ@kx#2HUg=Z&JR?G$rLgs#|CE8QLqh1clz-
z5-6mQM?iL_U-O7J7Mgb;v5SfXGljy{;n-e_0PB~Qc{{wR{hDBii;6tW;2}jQuuG|$
zZen$A-_8a{!PaHughNG6v6p^j3*+jyn#Bm6-TK?;f7WB4F>{I<s#|2{NHx>DXOdBS
zOYb?3L3cTMq&fcW^SN3G`?@RRrYbundg{b8)jDn12RwjAcybvvIE1`B4c2G#)GfGI
zXgkz}IxJ)6yipU$VA`!<y6`)Q9ZE>J>;c97;K3v6T8+3HzI0COvZyH;_+G3)fMZMg
zz|!6%f42)?!^%;XeNKmp{I@Uy0F4pCJjR|LPz2(MK;a~<m>mo-LR~a`4fE3s1JH2b
z*`8XZ=2}}f(j<BV>(@NN$}L<kQ%g&fxzWn@wizcg*lSNYkf)Do_p7Fzjn4{Iae=Ct
z!fyR5`$^O>bvJKkbXE6I&k?Rsn&in-*xaW$f2q<(R>!+kw-+7IAQW~_(o!z>Iv~=V
zh;OAs#cH_I<u;D)M*NaO`k4)@a+NgZo@-gWR8VhZd%5y@R(GVhW7Y1yG?#2gx<!7i
zOt&-JDs^c&hJONo%2|P+=8;+mvirWWmUya_WDl@a^*u`J)YsxEAKbn&)(YpA?JG@h
ze*+$GUq%00cZ5(@kIY(rxoe9nI%+BV#(<5ORNq{t*h@_nvi#jW8!BxI4{WSdew`6C
zYEwbnP9GHtCBxjihTL>v*@cW%A6qV7^bYx5vcPzi=lBgJJ8|-912dHT1KE8j<V-;(
zSo%U~!>mn-RE_E3;k_p9QP4F{Qp-umfA{Ye>0k&MFG7CeZqcXg{umRt&)Sy80BA;J
zx$CM7S%GmgC54}<Gf@e$-h8gk?0RiYM&Rc;)ljOfo*ZIs862YPy8!GhgQXFA>`@w$
zChk&!IrQ!I024%T3-8Cir9{vhkrH&xg3cw%-?M!aWeLuX$l+$WLd9aM(eu}z1!qrZ
z^}3hUD>EYj!<YFhGbRm&d2b<5ql`eUc;x~wmn<wZI)CrL*~3=VI@dRrP4+C@c08Y~
z<~wnkrud-Amp<p1k#@1jt)6f+W}^IrV@A^mKX}Y&=J)YqMjDj>G|!ObNcqMwqq+BU
zju}m(+&E@5%jWJ#+1%Op96iz`5eV>a0h3yS@UHn-!>dBw(+u^pwcfX}RlZ$Myzp~A
zs}I|zwU^~AGY<iGm-Q?&BPl(YO<@qyH^tLKC7?&Sx&_GddvLtsYagumI#v;D+mC9j
z%MN%VkMzEmK`k>ge?Li2j~3k+N{WXFvTjWCvt1+H_4B?vEO4VMb=@#?oK{*e2n)zr
z0M1=1tjHO8WSbXGzxESqZ8xzgVLu{xOrP>Apr(objgvSE;_31EBW6>!d&cQx<>M(i
z2~QZM77;KV-3X%li4O`N+)S+>jdHj0k%}UJpxS3a7!5OsN;E!&$sYxgK?_v=3Akq@
z`~(O&4)B8ra^m(_TV@$LV(1AH<YaHa#-Zym^N-<0YEdI6yI?Ux%Rv_3gA_N5*3&<k
z$^QX1QA)~}VO=v6mvSco6PIKzGsb_5M#EwM*X!?3vZx?Mu<>uajZZJHPyaoQ3TYn%
zNW1y_+f#gDGzjh0Mg<xq!{+-xo3BrQjASb_r>PCf&-d7&Yxz=a>Va%3twro?3X%+M
zHr58A<Ev~_87-!=jgV|);vMt<*p8~S$IIk*v7N|C^O>Hhzd6x$jPLi+<nezscSToT
zh1?d&#5N`$*=57h72jj-Q=eTTY0A8&?pUs4>26BbQ05*>f4;lY&z1MjSxV;W3OPM>
zWl2(7h(_wBeP&YQgwc*ox?q+xS{NEuf71MQEngbSS_x2gq^w|B4kKxq;MhCbKN>9A
z-IeQ@+vOoI4Ys#YS?(nB>1uy@l{CU2j1s`Iyh?iE<V?GBK9gi@bvRycSDrC#g>k;h
z>4#-~4Mn6C9cC2eipS96vwG}p*js#BkNFN!>!P6z#8^h|3bQ<23(5286y)NvM=f;l
zBS#$yS6+{CnfG443M9^pKuiaU_>w_3QNuDz5hMsYt-Ri&Kt=c)&H8_Nn5grRIGZSi
z0wJ{cXBu&brT#7sU<BHZ0o%gnc~0oy@(kRAq13`k)p2C(!Mkw4poBGg<Pg0k&TmBX
zdvNg-?!9Tbq|kGiuwlVs<Jv|H!dd$OG#)6QLF1wKIW$gks>xSS7p+~kK#$ltD&7d%
zMhQxiMnTs=+gK68%m{y1gEq@~xqE171)>`{pINzew4&5Y&jufFiXUja@N$ZC=H~b@
zQhYmp484Cq{1`aiA$|-L9}_=Dv)&#*P?J9y-+5t^NSy;Gq7w8Yu&E<_4w|~o9~hZB
zInM!8Z`vP-OC94g(RYNUe(a5?L|J%nFriYcU~WLBf;9ux6=Z)>l!JgDdp{JJoUk~P
z?-pe8T7U~9#k(Vu6AEx<*W01V1S6=bUh~AZ-izfs@ZFS37kKj|FAY5M_*#SCA;~jj
zWQpclA$x)DCyi4=2E0_7M@eCU-q=f|NRFsV`W>eIn&k=2J01S9!Xv$L?q0%AQf9Ba
zqo6v_!t=KZIyZk~67(JdDo>af)&$hN6?eRLEhhmVG4%vx`WA5?<&qScwSo6Y_8^IL
zz_iobmbNjCddXqL6n{%$B^1uj3!IlWhCiLqzqO05(Y1cBn3pUvu(p1`)yAjXss6J$
zLB%$h%KEkqhMaHZ^yS>nzSfkNt7?z9|4#hDML#`cUX*{4ZOLS@<+(`b<gRcld%@(|
z+?u3|Pq1t3xuj$2-_QGgOImiV+^_C?Lv2B(lQ^Xu67<1L)=H?cAep>KvXpWkTMBgp
zWRl9%e`On{;TLf$)c0kZDe)@FS8d}hsbhJzB?V2EH$~Sp|C21Crc}&NPMLK&U9rgn
zW;3oV-KT$YZ+70mXEIz?<TrbjtQ@VAydzo0)+fWG;s~~D@(gT8M<3!bZ7XZf*5M@V
z+H)O>f5Ip?p6dYBipHC_&vh<1v(c0)DxA>cFr9p`4XJF{S_!-56C0%))N9()F^r5S
zV)-mg>qtfI+Ul*;N<Od5G?IO{_-*DZeI1KaQ5k>jURt?)<O<Bx`jC4}tFQB*Vg(hn
zP*bs_UwB-KD3I52mW9!qu_Ag*?puB2SXQQ+kYtHkl1nbj<CW0b{Nl61ma=)3Up5nt
z{M*&6Ih_-7lqlzeb$i%OGacKNZHmHEf2NqZ?is+-F(a<zHL4Fk)9`hnEmdM*f40zX
z90h+2O60Rto|5QZ=@>dxCl77-VEA6v-2`FK{cPQvoN9a?;&Iab;`%g6v8Xqt87BB&
zL~otv{PObddV44*P=^m}k3KiA8xPm-S*+UttiX@v66?jlR0TgX900=sR=mCf3M2}v
z;6v%(DHXa<-6w<Bs|&#hYxPf7cj*|jNBe)78Dfv)dldFsIB7nx!tGhYptQR{;rA86
zV6a8L1Tzi3^1f~k#a!Sr5b<vBP$oA%jM(T}%c+Y>VIWL~>1<uQJ=m}-HDj(=YOPcV
zrfOkMP@|9vW(0L}#M7wURe*&9f7MmaY}Q0q@q+_0|G=qMx>@RK;{6tK-Bs$Z!PI}b
z)lOAduF9yUR;3b$3(d_HB<~l{Q>MaMb*f{5cQzPU(*Kt#NrjwB52g%z;Ka3drs=0j
z*c{HujmO#}x!9X?Ydi`^72sg;GoNs+%^!jn-pr`IxB9i1s)3>SrIgdh?XeqcU!d9-
zm4CpI=|K6+nd#X3ejm?)%Z16nVda0glkGyqFxrY|>Qp{l#k}Ct$E@-5SaQQag=AmT
zZ3&9g*RcuId~+h&_-Y%QaAiN-sPQ;B(bl`mHe45mFNHL!vn`z0v{zv&Xd&!+ZI+}a
zBf)MOY4{sfJ~lp1A<<w{WLu?n*V=7SNh+x=LJQjM9D?-Pw%d7O=62g)^=KdXc0UuO
z(q{L*Ij*6ccMnfB8m)Y+JlqL4{s-HLIOmsPT{9GyK~)zM0x>n0;n+JUmy9+u4S&=N
z2#h+(CO~gl<W}@h1WkHKithH%-!G}5G_pK?)d3n{Y>PwjO^IZTdQ(O{u5O=KUmhH)
zsFR0i)$^+YdSU1j3;}~z5kLnA>iI+ca$}4Q^xV>e#yCe$q~D3Y{bubT5e9gA2m-KK
zM+Dj97$;2l&eCHlz9oEnM-ULIvVSgA17QW1J=5LNL%ZQ_^Xv1^D<dixodf^muW|MD
zdG&oYDioa2)>!rOvEmo#K)YlWW}U#PAHS>D)ze6|H&+5OmkGi99dm&p8=?v!Ysar9
zbFqnvQgJdHlIr+C4=L!$96pkfk`084S-@lmjDkvAy=S!PUBMw327y_@<$rzbfdFjq
z$^rdkeE#Mw1^rgYRyK^AHtS%l8Mb1#wFoQ1dzK<ik}*l|gd~%@E#BEPi%;5-EyXNF
z5Zmc{*Xd(7>x7m=1GmDcz8Z8H7_URH_JsQ9$cj;;i%r%GeGm-fCE3pxYK;lPpa<s+
z`+(Q;_GS59&=^yb--{rfynlWx-ouG`5g?r{|FSU4x;O*o=KSpk!n0)jVoshqzlg4O
zfGr`T@L<)To<V*)C4qw|9xuFOd*60d2uDH9hmsRw)fUm)ey_Jh135hQWVIHxhcwD+
zg>JD8PJH-4@#0bJ^DqhDFngN<qH^a`fv|{#2(oN6XPu5|PY}3Ya(}?-qt@G~eTw5z
z8;p*`_u8Y@$EZWj&9h$@qUnD25Sqno@geBgG{w`$h0#b9+Q*MKH9)>S&P?c50UViF
zGu&X&(N2($LU0Gt6ZE9BcLPCoEwHpT&1H=X0mN`CLDEcvbx#SD-Uy6dZgh$zAT)6D
zI|&Ov$2<&2wDCK7?tfX{dkVHzSzLH(U*v`WI-59hyv7fact98_ex4)B=?*s+IHEHj
z9K3G|$+9ghlRD)yM4hZ%^4SHK!o~iJs6&L2`!)VcPD9Ki#n1I0+n^=XHU8T*?QU3H
z;Y;V}ly*zLyrS~E$ghbjtvyMu@oV-vqZ=uHj$fzbu@@eYM}PBj$CvJar7!UD@?}bw
z^LRBQ9f=7%U(MeYAwH^tO$1y{Uj3sAScCPLi5&G6F)&$2f9!(Qm%rX#e|q^1G;RBv
zQR#ly)<aI}LzY(i43u2&vPRmTeNcQDyMQEvPUr}sZft-TwL+>L!1>fc-x-=UQg>jx
z&84c~s--dLKz~&RrZ_`|-+K>iQZ*Kn5>*UdF%{X%OWI#|@sgi@T~TS*A8PfgR!?G8
zA8cflWG2En(BG0(fi9Uy&39+n3bH}lL~>TE<Vdk{0NU$2qw0_Xr3$rLI#OIY)wngC
zE2PuAf{0e7P(e+!0e-*^gRTUbs~q;4q8!Fl8g&rsPJb4jf4_XlzBk(S*nLlQ)<E?v
zQb4Sc&IOxGC0Z&=pB8%tL9f{LJgPN)Zk2rHw3J~35n}nII*}&b*0XX?XBz~73TJnQ
zLB=D$GemTnJzEWNpn0k~2zD*8K^mRdw?eHs!fBo%O;niM%hujwH~qNpu8upQJ2zxz
zemilKCV!mE%7=y(PPbOUYmV(flTW(r4)#%)_BaY<zDC$S^gRtL>XqcG4W{PPhib_)
z8b~yW&K)?WWnYqAe$gD&7Bih75J&C+J6F?Eip;^dlp=GuPbu=J+C`JAY~9^u$CA6=
ze2z*4Gf(L3t_{32G+1LC)6Lvow`lb{Z;y_Mj(^&&Nd?p@o4J{d8wfmxA;+(68S2&A
zh1puL!&G8|@Tvrnk^03?fd*BDZKYOwwl)>2srk?_a0)`<RjVi8c!jQ>ij{AB+x3Ob
zJa>)Sc1P-N-f~d0Gv+Px-V^EKQq~mQqtS!(upII{Iq<1VA(Zm%#5AzXU8Xkrcw~(u
znt!wAGakJ*CXIss*ZGX&35rkSe1@+7l>PDfjG;T7VBSn0r6}Kld^>A~<1+PMd9Kq#
z8AP9vmcc+571P|(-W;0%9qol2^@G{XW1uAeYq*^b!dC3pb(DfW39gr?UMVPexg4G#
z{}w^teG3UDReFVO$1(~zH{(??3M^mTY9ZM~)d&^#gQ8coS~fc!^PcJcrxD0mebD$#
zN%1dafoArXVO=v6mywnN6qnUJGc12(Hx9o0SG+%9wH9gv3Waf1QVrzRN#BATf}ZJ~
zLwhjE9P;-=EtDudy=pt2z$llW7K;2xaeJ2AmrZW(A7uT$e0+F!vtd(4Sm(XlK7QU5
z&Pe>Nq!+>2%}Fhc#Hr`)=aZD$NeLJHfRxtY$K&4tzwP<aK)@Bs;wPSe`S_QBJu@4B
zQ9K6HOg^$75Cv#0m&`*%`ijUG&hVq~1LsHll6Oyq{i!sQ#8+vwjSvv7u1|Xv&1R@<
znaTU=sx)5COgp<wLD=Y>S<2*%h{1jvGdRM`z=ou(@_I?%T46(M^nF$MHFNl=qdQe8
zLy;q!K$_J>;<94E!*%JwHfeFCKh^zzm%@47>k2jS46&!Sej#z?%2tVQ{0zs>NP6=`
zsTpzAg1z$fR@0OIuF%D0(wMfJbJ=-kmT~WF`!sTX*6PGCO<2nE$&xgRHm!<vg(Yfx
z4l`BBIjk}kE8A%ujBuvi#KFE*L%m#dMI7r><)d*p>Z|M|-m+Gb&LGW%9fIk9VxAvC
z1Gl=+&abjGYm0<vddAC4yr`o0vWp{C&upi!?5M6;-TN}b@9(XyBJRJ-$dYqy2x}vh
zjT@jNRG|WVk#@b<XyNf*4vV`*s491;T<@^%x2xD%m6W^>j5iiszb93`9bU;!o9Ei;
zH_WyR6%VMq+_(-$s|r=Cd2OYCYW?`Ur?a=4yQrJ!gp>A#u#c-p$$Zten=mkD@?pjR
z$tuVDku_p$wN!|-w&5y?f+hPj{r;ET(CoZOg1%asut{O83l6_slzRIe<y+7E*rnFN
z3Wt$ok{ZZUe}B$=WZRb2lV3U-(`{ejYS62Ig4GGh*H)xLzl(#uuk6f!0dXKMfyG<b
za8n)q#XfF>X7Z-Z(~xJ`S$x(uzmH0e{*){!fN<TAB3C01m(3v`8nClMdiPQT9#$6@
zr774<DM%70*x7jVNCTH3A(Nm?FTm_9LW4f7X|xB|<;OH2Ukgt(svBzOX9F2%kip@_
zDO}AiFu+=xP(#+G8`VyKv1RW|9A%~{FDv=eNuEw32f$K1dBXrFRpq_71}<eW&Or-d
zZC)0wL$}PfIhy9Mn#*nYD|2C?wmFoBWOG#Zz|FFN02DHBJjBp-?BRqv(0m83%svW#
z*Xjt%dXCBvkjiOf&29ZYO?X96N6@azWXNcAjIP%_*QymG=>*MxowelJbKRjOZLh+E
zTqJCa_RM{uze-8U3{Qwyf(B2_W_2Vit44|hG{g9NYfv#7gu#+-#&DGczFy{KkO@u%
z8RipnGU+nbz^t{ONx}Sbti!f`w)wK&4%{*?)9(D0B@Dh785MQKQra9;*Vc&TxmW5q
zHHV6A_iOd8*Fdd*)RHdN5?DxvU7^=*%+5w%eZzXvT@xtXy9Ufl$SzezXP;IcmPEV=
zI4l5Wn^*q-Z1bsP20tU3n-;qIucsNUHIDwkQypXpU@#6KGt<dc<KKVWD_waZW86MH
zKm4{Kx%XS9iQ=2oRQO}d!x&pE#3ZRwK1m$WkC_C<!@K`~e);^5ProXG+y0l9$_<3J
zDu(U3JUS;d;J)o^eGaLgV37wArE#!%DIgl<+Qv5UX?p<eg)%m<(XVMV3J9Pz+ia9D
z!Q;Yq02?A662&%3=Y8da#8p2c%hcFNPl-z1n;vmL?Mjd7!=Gs>?)vxk@OgXqP*Jr`
z`?Ms;LO8mA<gXT}D$JU_Q5@Me6j2%A1ECT<+V{~YleZ0O%=sM{lrRV{75>#oH<@XK
z$xTB@izv_#eb3_uI83tQNHbErMp11xv7wWRs)}wwxc~j>x$gUdyWX?!Lgz6h&tTp5
zh<FQ~1bJ@F&Q<HHE~nK)+@~C*`p|L`-9vuECL(x$<cE#1RAhVX!hLu#AW8{q;RS%q
z?tuY;+IH47AP#zfB!ksxtgs9X`Y>qg!$hSx@+&1Hl0w3*hE}MlpKEKOo%GZT1guCo
zE!EbT(aC2XYJaL#mz0pT(_0WNG+Hq3Jg|VTmq}Y$H>b(vgk|E&Wwd)#cm2zn#+W3D
zdFu6lH<Mkys5lxD0Mas`gO_nm(Ta2xEo)9=3DXm#GFws&dueOARAppkD`h73u$^a)
zLU~zDJ(GLqytYGW^-6=8UMlhC-ffwoA~GZ7^l3sT4qWZ{+@HpKKuwQSij7pLKrEx9
zMJ#n(?>!ZYLxIK9xl-=)WHgmWyeF%NQmyHKtF{Q>X)ZfEzf&>NQo?<?N(}^Hebvm#
zZ7xcSZA?j$<;hhZF<r93y2cKsXE@5a*GumL7~ec`zszT+yJKwWoz+C_UbyIvc@-|Y
z>o4KrKijyhrge`jue%rM>T(q^`+?)y49DUb<calW=B8K?`6Vxua9BiuwVP=w2t(a}
z_127BVM63EBR@EMlC-sClQD<Wp|nw*%G<+KabZ|kW#l@&OhtW8Cl|LlHo^y!3f<6D
zDh^Y}^;Mn8R@AhT<91m3DJ)oYZy%Ybvkl`t49v?fS}ox;7IXXJD=eCSLuWq>CG7Q0
z9qo6wpjIZRjT1h?EN@cb+7!NJ@MUj*Ls_1)`x-6b=suWKau~N~X00OQvCv`FIR>@d
zG|-E8Ta$In&A-;}y44H2MoW2xRg<5O&c58=zv=<Y(74;#J*Qt<(ApKW7(F)tc&qqI
zF~QINTlEG)kNfm)4oj}2rS$cFdP3j)+fEA(gwfgkrtwlV>dm23c5(atU)$$@hYvGR
z{=XF&XXzZ4T^uciB_KL=r9cGs&J`n%lbW?l(*1C`UHsRqwHTI=xkjkJ@E%Pg*=?qu
zG?jmrBO^Z1VE{>@sNDG=fUx32xeyB5;Mr4p2}F(l0FgHJuI$)-fsm5`7Tq(tX|1se
zwn>1Wf05fiaq@>PH$%2R^3vyj%?dQ3Emi$Lp*SN33UtQ66bz0?$W+1^>vuFzedwU<
zn`ksqzD;8kh4df=B$Vh^_O!zQkw9+0$d7k*m1Kx<XU+-+uUJ)KyN0Hb?8c+KdtoHz
zOx`#5!v5W6@e=aTwlFA|pk``sxndKa5vynR_Wpo~yi}j$4I-fGuMEsEGNd&aV)Klh
z4oFm~y9!k59=s=XXtPew1g=XlLdrek-zdg-fV8z5u|<qS5mK+s&zsjGcs%HD6o8YL
ztazfL>Kj@ZwqI8fYWj!0!_-1Kv~#nWD6p<*HxP?1m0sSPxd{0a+kq69aU2&D12{K0
zm+{2`6qlApGsXe-mt99Q*?%52#w=~M#xspDzG(<?5e6D+q)p>Ha+r9<k*1^|AgaX}
zp>Z2fMOBq<k#PV0>06!m1!uiy-v!T|(neTt%boM6!#1JH<_lrdd@FPXChJ8z10|(J
z`vjfsiq_e(uMr0>_+er66vpm4XCH2s2Kp2LdxJEHstqaFfSa$I7=L!W7_m^HNe(u7
zAdkPSOOhy30rWt&N(e$7Pv4IGRePHt@r^|v1?carWwSH3HUsh7bA?X7pXd$5GMeXK
zlsW7A+*ZpN&xE~U#3%clJ~Wd3Fbp$gh0Anue7Ni6d*2oAa57?O@}TwNFw@jgS3a2=
zzoN6j7yU}SJ!GszPk*Ms6W<Rp`+2qN->1#nQNXTDMFD?bMz9(s?pT<(A?`VqTqlsm
zm|Y&rUQ~V`yN=jLl4+Beo3&)vGtxF+aT1=sVt7z4L7uB5v*z(t^5~_`@ADoD=7h0p
zgs=OWd0={di~ix>X>~tfQk8=WSM(Xq4@TZ|eer}LkUoyuk$-7)T+0(jZEfIBq`@Su
zG)Kn=1NXOSpYGbX>qe_KKr{k8bpKIlaeOCehmk(p?Cnymhb-GR8LBE18@L~meOI#H
zTvnc?(0i=CNUKmjHtltt3R|(-Sfz8BbXM=uuH$IibtKC-bbZMh<r}>1J3*S*W^kR;
z&BHp5nIu7Wk$?BUywz1!lT8)fY3pCAKh95EI+HS03x2lpG8;OjHdL+loC1_1<MdD~
z-cDL2a!-5iQY3D%o+N3t*|6m@TC}S@1tvcrFG5%>?L9&5c(aO&X?Uk#)~E^_yGJQD
z(B|mw1M7(CGR0Wh{7I4(gQ}6P;)mf~7-LxBta<J%r+<O6pU5v^DW0@v4hX%u<Di6g
z&C0LIrh4i~n$NQZvWdo_Sw#_g<5)i6dxldz>54|o4~=onbDmkG3}Y4~pLvAXx1>2L
z%#voA$)>Fh8Qpfi;|qm7TWT5AWytOrZJ8tBJ5cCX?}q%Q$w#@I)PMvbZOcrRU1daV
zhS9>WXMgBo!-rA91fj4hn`X+#<5?C9XbtnMEITjn$2di?s{4%wMOmj;loe3qEALz0
zQ(!jojsiz^*D$X#kplNf@u}>=5Zf9AxAnSe^YS$=p2wKka{fe-i+eFL<FW*sMopua
zbZ6wU^W&!MzTk3Q8pKPslwkNC_%TLMY-2gpV1J!emn`NrGBoQXkKbyCbuK&95N(&B
zWTt6OB9EZ*lr&&DNsr9QWk+1pUEmIw)Iu<THFs*%M5$9JVdu@q_o%u-ZK<xdR-|Hm
z<j(AP16}=3{^;gy_v}JPCyb0U_8OIEtHniyi7V(m8?Ro$dV?0*y#ieI3eyb3^M%Pb
zYJYz{wl%05vp4NE(_l7#&?p)u{m|a>+?Voov}t`mT3)eC@-?(x)_xXAhpa^ol$DH^
ztxvXBXy94FmHoF02l3S;@hjfcVm%q7nP(LW4YPBq$E+jz49hetEY)kK`%Wg=w%-lD
zR$AnxDr~RWHhqJ`>_YlZ>D1?NyoTWZVSlQVS>J%PXDe$z>R$L&?YFbeWZeL~y3U#U
zQODvS)7!~zEXpb>!|TGrN6BJ_3`;Ytagt^8v2&fDp(M}8hCGRPuN<%NiFbe4pE>Sb
z+$5Kp-H7OtV@hA|r>Ccv(;q9r8SS$Z{q*gWUX<3gUeHN88<#)-zI;A?n6XyRCx3;^
zJ;L+{WC;+%$j5rH5hWU;bPixhYm`LmMDxXO9h-XFt!BdLa{CJTS6`S$J~FqA$&)j%
zTIJa3^3`f=Ths4@h<b%ND@meTkESNiMUmu`0Du<`%z~i_)AP`<T<Uh(#1j3<oDL(L
zQ|Gcnn=olP$?g>xeLDP5Ge%{S8GoklSMSQME|3&L5Pc!m;ka!>D!C&PlH?Ci7ff#A
z%!pU2a*J?xsgEF{okcJmy1i!iP;f$_UQP(Ns~|0{u&|QdGiydXb_cen%d`RB3#-+S
z=rT!s#3Y|q`n(d~v^G*{YVl#^k81+yDcYLt$wQNo|9(G3m;vy03uFL2ReuIbk8ku(
z0VRPs(2Vg7?4kNpgV{Y;dsPa^R{vA#2tIOT6g5T|W3cvsV;}o)=VVGilXlui-Y%rW
zo#g8il)fSDE>LrcT2?@-r1v|eOtp+IgRKC1t$B=5h3j%_ie84s$rV0!j@MWb%W(BX
zNwE?Gcp|e-;xZnd8eMvGdw)8vcvbq@noqnjv5_|VjdXjUvy=nd6{?k(@Mu5F8qUW>
zgmglpUQV<2T1*<f8N@}qD3qqs^DH;R+D6l;lmX-HYZNMz${bkDBIz5}vH*VI<f-{3
zdmD|lX^&*7(ONyMC~daLlWa<kkGasG&?&6A7>;d<e$P6HI-u@%TYvU@93V2G!)m$B
zFgR`A{<(MFdK<BZ(>q(g+Ufy%I`j1RojqP-cQ&{CvCUMj8--R@1PrDV9oIHX>*T>#
zvu;wkR%iOnvo_*rbjQbAsfD$(<mcpdj~AmIhPZk1*ck5`IQpamf_KZ_-ahRDE5`t@
z*fwi|0B{@IR-=`CqJLjEiotq?b>IqPEy6dhfT=X0TmtxST>%4&4m-$p1q^Vfyfgg@
z7=+ZPl>G`=8yK#Dwb9`USfAW3akT~55Yg~|cLfYp2+M6ZUG~}@{U5yoW<h8b-{cAy
z%0vWjZoC6V&_8quj4io!38ypnQ*VK#f=$bC<1MgMtSNRIj(@s!&QTJYf8E@zEAtM!
zbqNaQTN7m!?u^~)o)AALEASh4buciLqDrHBCkxMJXsFy-jUF7C7~Bo5%`78ByaPMH
zVO@4plk^%QuH_}EfR)kDtB}0vZM(zx8c@^ZdM{(|WYS*cTgb;JqnJdCG_*kvokFxT
zW4b#NMy=_E@qZiag7uh*f!Wg3q(SMEPpi(2##^MGPH<CRHDOKnULc=jYR_E3WHZ&m
zNHW7cFTYb@jVI~sEca+<al3?ys})o8C6(6dUHK$gq=cxJ#54-6_}$GD)1B(LV(6Bw
zX1UkuDyZv~E!D!SN^n2_8dK#%#rF<XL(TMSdz?@rbbmy*J#)>EId|5JR)AcucYHXO
zdbS!%nQ7MbOP<V=|HvgzmI&s6u%l{}Hi8kJRo~nFooiPsytXS$(<OvSi`?dw&b`Y9
zt<o%-w0)#A9bg_@+?}s8AD3hi9kx?5e<n`Hvs7F?9l2sD-LW)Nc!(~{t9|bwlTrqA
z7BwF-VShdtBK(v?CWT8X>0WWjwA+%b&Cdzr(Xo$tdVK@Dw$i6n%B`f;(aYTOoo2P{
z<Q41JcKW<N9IlUk19N`&2J7q8h*2vkD|ftpu0q0Uw`Fh0-q4I*>5j&DXxjl_`9D<~
z$6w#m;!RN@w2B~iJHrsv#~$i%ec<{fWJqlsJ!dU)?Y0^CX#(p(%w3AXQw}swmm;<=
z2qqlzdE-ATFt6;7Qu;Mj?VRMHU58&^p0f5h;+lx+6DBcyXJ%`XB}X&w^jk)e!=6z3
zKW%2#fy1!D>Y6KI!p4wnX2d-;!C9*DAM5CQOqXF@GZmNOG6ocvvQjg|e?HQC`s?eT
zcam9_BG~wE{4>A3yzYM6%>r1X1gPcn_g!p94MKahDEb)8^84SHFS}1O(K?l+A^De)
z%kN*8spVJJV34FiX{|AX20@Zqaj`ZC9Y19a;xAGfk_$-_NOB`-Qo$hOhogs0iYRF<
z%uJCKE={DQ=&#MxW1FOPf6_0RvcSl5`IST&T1b>Si$h@@%(DOvJht9m1hwKx9xtEQ
zJArzouW9{j??86x{T=2dtqBNjHLu;AS4k9TR@%X+?*q?m^wNH%xrdfs=3*np3r)sX
z=1%@Yk3MJn)mqMbuvJPZttzL6EyP7HbdT8z^IP;uLw<;4ImSGxf9QuMTGc*UR_&iH
zfvt2suXW^NcAl5OLZa{J6J2EVFoqwM=FQZ%rBVt+T=q$*X#SxvSf;=VG)Z8~3{<Ex
zI=f$Tn)2126;1^$9E6o{HqRP`Nb01;!~I!7G*OU*L=z?Wg^lP!m@^Ipsuj*@pX}?8
zz&R6wuwd_qa}&kWe>le~bJ4+oe<d-7mI31k1&H!ALnIgZj6Q-kzMsXOd35%F=#(du
zI;WNm|2N(X!hjlOQ$*!Q<-cSY4_3lOo>WGu$83^D)wfL=Wrp+>`Kx7TYG%ze<V%h*
z@JoJF$g7=b@#utPF5xnds=W(c%$yeVEV{l%i_QrfLUqa)e@0P$P~HJe4FX{});V0K
zorsXq^}u3Shli}k>JQlA+$j{b!l<yC?_MZMV(S(4QHr)YPZ{@PjF6TAHrf*RbD5_S
zS#Y1%Tg&inU9jaGD$P*kC^`j}d|z7zP|yFjd6N%&wqNXf#~!|qdzo6%EHpk72k>T4
zKxc)g9COm9e|$N+2Iyn&)f#XHb8;?k1zJ6iI_-OBc`y)!F=~T{qD5rQyYtpsyh9N^
z7y8{oXGP~Y0cC}AI8#WRfUEdb?4X8lRRVu{+c$5_xHWw7Evv~ClBKbIBwIYP&A6Wa
z`GleMP4<6ycM=lPE(cgE1peprDz%`Dv+A{Y9Jcscf3R6&mG@A%;5kzK-b9+1$xd@z
zj5OO4D4@)ilHQ4W^{De`i{4NSx2Mn+Q9pJoVQ0^cJ+L5DYz<ZWdF}T#>0??6clyj%
z#xH7IPtNi^?OiE+Xk#hcrHmy=#o7MWmwv~gF=hLcbZV6`^%5q`Jro<QlI?lSq2TcL
zt|d@@e*rMh7`+P}7C*OzYVnrr6Ll^ew61!cWG~@<w&;$Zx)yqq<9_BU`^%Fyk9W~F
z8Yh%l+mdD9KV(~1$4$+l3$))@trb*xbm!+R1IMm{C}jsVZ46en?r3Y_p?OG)rp-<_
zW77lS>|;gQif_t<$2O~RExy%=?c!MIB6_A2e{pW`f-qQEp^Rm8_8RWf&vES9Oy$!*
zLCPYY8BzVlXgkc*WdqDFOq`YeX@)IN+LkO?BWL}^Ugvy5IVjp4^i@`!r7Bi$OF!Ag
zC)QKJYT2Rl8&+XyzpL6FXoxB$*Q^#cY;U{RomxOY>rwkjH%0Tu##0x?RKKxK@y4;%
zf3aVwJ`Qeazs1i>w9nCTr4!V`2i<2GB|3G{*0#7+xoI_zO&^_WAdEgIivcq3L~!Qk
zUOBo!fwY^MsAnB^cEcE!Jc;ucE+V|Vu2iSJ6W~*s;J1aq?&FdWj{sH&Z<hZosaXI`
z9wL?kxQMAi(@%uk-G_gD{qm>Jf1}OVfAaO;v8!B=;1z;jzVj=wYm{-zH-5K*gbl=l
z2!y(%E-!%0&eegn;q$UXMGR`%54YBDu{8+vvA)p;rwWIQ9m@n=Q{f>}Y^^-sx4yA*
zRi{WyS2p5Xq{3t4D~&Is;%j{R_tKR{{bSjES$3aTtHvr9yLg!i2TMO&S_OgNe>C)(
zElrdrOND?%E5VU9oM`RkZCwS0Nz^2v7fu5GS7)m-mzGZM(nPchh0;`6o7zw4VUm?v
zs*xIDXc;Lxazutq6j-}i!|ngi-+A0u8uhXJo9R4&(gKZL9#BE5P7|$E!8(y;bNAAF
z&Nh9?sxR>Dl83+Pt&DJ4f|7mof1W!rN1}XMI#ILE9-7iVy<5>hn?8f@Q1qM>H3buD
zq_n6g_|)bucQ3sQaAOTK+x6L#5bpc5-?xDG23%``Yd>>wH@J3!HaRo(x@A4~VRa~P
zWNA)tk<0EwP{OTpeHweHnmD_ut=plREOwix%qOJW7RGw*hE;a!zh#xpf5Ng={&N|-
zaa4Nf!qPj=B()7fVcQLxmYG~zKG*aY$!w<bIa2Z;Ysu%z=ZWDHg={WNY6?-dE3>JM
zwA`?rlvPt~C0#;1r~a|{YMw7YXbVOJ!itHjT$r6oq&Ae2D?v}9I<)-y@#L{UIVhWE
zQIp)*($@DJ4aBW;JS~~1f7YY73y%Fz5n`_SliM3;15w_o^A*ac4UKRntOR(O6n_Cl
z>jgcl<;001v7b(J+9A)-vQbah+6e2~*{Ury{5~@BiNOTme9BNEIs;j5jr#$nb3r`i
zv0>G67iX$+>^nGxI>!vB<DTs%9V-gl-5FBVQpy~-s$O?e_Wi(+f6ub2T$=}!yyZIN
zdIp~^O*J;w!0=Xg->$ZVPV<L4>bcdCQpOWnAfIB(N-v<vPu#HUxj`Fow|fOhl#{-R
zn^PdcHwmWRcZPfD@$Q-<EafOu>9)^%o<iLgLKOvlukZ(j(6}28&i`Z~bmHm<h0tl1
zJB84R>J1@u=I?9>f1Ntt5JG1GZV#dJa^4<76{3(3e(n%@J-GhQ!lu%+m-|19n&reU
zQLQnuRNnb*s?e88XyO&yw>P4vQc8e?`$f+yZu4mYF3g<V86iu2o@TifAr%O;E8ed`
z$W(S)^LGLvwWm#}n|t3c)}`EGNQJJ+AHY-a)2>%`cSJB;e<rLKO7dHGp~OD+<@kD`
zq#bFTIeuz~5fN*cwr_OfST!8p4;MVj;XM|^7!FL_nFD;sH1&z=2dfby0uHULi?Xot
zxLMJGZQAaA$Bgr}1=kx8>k5XIv%^u(k+r1;d0(#25xo@gAYn=ogBBLU%_$;SEwqB4
zEJd8S`avn;e>BUTQpAbs4JqQx-`Ny#>U=|rI16xlia0OloFbMxv*p&})N(GC6*D$U
zXxpwfpHINvcb_&0z%}I`wl8f}Vsd*5B7hW7WvhQ~TvWa05-;V}p4M2?TA}Ih@T#*?
zm7Hyh4S-dGK2|<->yQ0y;bY0^pbIr6W9!CGMSN~8f7<jXQ)gf9=cb#t=4w0E&I|AL
zN^3JDUA_zu&ripz)cV%clZ}c=*<riUZnw?4ty(nuZgq#d%jGszxqlMx2cNj@t)mg^
zm3+r-OP5|4<yMG!J1hhzjP-*HwR@RwA^;QdR#+K+lA_wk%o{bevGX5YQk(x$a`Rti
zBp^Ntmy&Ta6qmxPFcO!DT{Fgi@J{*1zrOwXP9n)4UR*pyuMbuLwni^NwE)Eey%um=
zz~mHs9{`ihG`Itf=9y14YJp~_;p6J3!0>feSr5-HKUeC`anU<3VpNw<WtHBbs*ZN5
z9#wcK-@1(NQ%5V0!_!m6O>ivIlXog#V3u)tl<m9BMyNIt9R0D5(jKgT&2J3H&~RqD
zoF4X<-yhou?8~M`Vd^{e`K@H!dZBe$PRc$Fqx4wdRQ?=i-m1cj$9qK-v7Kj}1&1dN
zHrfK)QD`*U$|FadcYQs~%5t%i18YUnp5&mcI(GUCMoar8eiw{3&?!os|AK2Ie&cSk
z-bE!Js#cXz-}hPQ3wX+ZA<{|3Docesk!U~(VWW;{KHymvs>(ST#Bs!DHSA{QpQY0&
zb2~<oG}Y#CnX9l4jOIS?fc!r)I_T6Z?<|glhV8f^1Ga0@8pfoRn<s3vNTF%EJ@nIQ
zII>iFvVo-2q=V`sT9h+(PB{^kB|K*H49J=(Tj1VXZzrBk)=uGn->CVOGtE+NNUO1F
z;w?fX>1-n%D_T|G=H`(nb`Ari#<>GwYZ8xAMcm8KLHaMT$1X+Tfx%>tkCj_GMlP*r
zGj&WSq&#bqPOj=mwK`8sLyz4mm6d3_5L45x$%-7C6~UQJRt<8pzp>p)W@&R9Hlgk!
z&>3pl)B4;QJ$UwiohS3zBVQ)n9;j-zgK;)a7AZ7pm=qxtf}03R;(jU8sTDfX=ueA2
z*-bp%w`^&!i(Uk2ng(C(bE^%uI{QJL9*ZwU2%z%;CB>tp6dEU+jy6Dw<a5}Qzh0-O
z=eN7xSEO5;gj0U`{QWM!uto_J^`f*F4!q7Ee_y`beVW04VqhhOgf;a)Czn6IF4M?w
z7%wIXnkBt=ZqO_xLEY@)yel|~d#eZmc%p`gu45>Fb;L9Rh#M9wAsTfnED7aaP1d_w
z!7+Y>Fl>wd5WzFAAeWQLbZb0a(k{Jk5%Rliab>mxPk53xZ-)LWdmKjC9<@_qO^Eu~
z#iB9JC*l=<EGR&Sk>6f-k4w8Gf=e?NlP1CLYI2T&IywQL6$77<0q>wSf>f_0ODD9Z
zy6bt#JHkce8kL06YK98?mhw2-1sA|qn8i6OK&jitj#r<o@?I#fXGmI3$9e@s$aXwL
z8xurjvJJ}LS8Lj?eiIk05fDZUns>R7G<qi}#LIVo&=-wRA*~^=EWjoF-K9S{k?e{K
z%Kq>P0`G)MVFhti^*$TU*{%8a!Xl?b_sP`)ttFaH0b*4r-FqaOxIRLX&nu~3ad5&7
zRGMe>5%T@gK+>uen2wfjvFP%LJrF?#f9GPWGw_JnQG|P;=ZQgB2v=u_H_!?6mNq5f
zbv(;|1=v>quV5NRPU!`De;XmL2D|=l@A-J|WsV)Q1fR;GywvCYKf;=8Ds^qerBCD)
zs6E#NO6Q^^3!Wf2bw!l4%XvEE5#oE}7V~_4+@f@jC?`Tn>bZ7BZIN~&=oLdCsk$1y
zu|gZY!tYMww_c<*)jxvYIXL{)PLw~5-$wy|pOrxLFys=oT!|OZ5>gKb(oQ(Al?R+l
zovqfwhGw`8rS-h9T9X`)Wwt57)Kh|4%OIP@g_>}AmdKTqT$v}$XAMbv!~#H3m(?z`
z%?@j|i5<<?u~&dul(WYooxnGMSDv63U2}{@#I74;d<GnFQgP1sG`vmjmn3{vDxsWz
zy`?D|A->O)AzMRM7KQ(ntqvnYKo%kW_cImJ^@49cK<ReuSrRG4)oKjD7=x=bpGG!x
z8B#I`5SuX7%NG3T`khOA)x}*d$h3-==1;l7=0qjaM$ZamsQVqpd)!aWcI(v<FN#rs
zyK>1tV#!RtlnE{18j6^zmD@ur)bi_p2aP4?ZhthlK~(J7w``FW;}E_Z>>+W@;D;`w
zSUC&VVet}arckH_qP+-*Gh8h};_3pgTGX=p!H%yAuyy2A0rm$CnRbv5_%Iu?WQ4cr
zml!f*z=1pelPgy<mfKaXCdmK0%GE^3PpVu^ShQuAwYH~EXRB6VfsI-*@3~%oSxY#4
zqgq*rClXIA8HvsAm33D$&R8p=4J*#Q+>~?1838tL#*+8fgu^xKbx{!wL@!1&L+o7J
zETkE2g-hWEg|rFMUdaK=9W&mK>@rGY+QZVl8<Zc60`q0!JBwvAl-m`{R*)|@I#|Y;
zAw+hC&Le}+ulddtrHs_@oo-Nnl$s!2P?VZ+|D2+fPD%v*6KYZr<%!_c4Qo;}#M{-R
zv`->R{R?VR+JTX>=KrcD<<8ZlGUYc3f^z}J^^O$HI!uO;)3RQPm&(C>^F1futh1md
z@4NK#0JI0eH5D1c?&%KreCWIiY2z5wQ}<7!v6l{bFi$6XU}mhO0Bx3kVaDzmZ{TeS
zo#_%C+Yun*o%yb8fE2V`)&9P~X}vDx$lw%Bif;MOQig@n&Sv-Sl?pzet$}}85AHtv
z>+6?4ef}GSFU!||=cy1hLja&sT)x*w?;w$lIxOGtISSxAXtNRtGAp})3MLRU*A}*c
z&&!?F0WxD3+32@CnuLviR%^Dngz6zrBm``TXvh%TDB=6Y2Z^eVqU@=Wk)JXYT$>+Z
zJ*~=*`NN-SDy;g)a`$Dq`-EI|UWYs-$_hAn__xJXC9Db%9x~3hA)r!%og}WL03x$D
z%J$m^!PP_$cFF2oa8U491KsvaBW&L^1h5DK4bk^JzQc!UuQ1Yo9SQ_W`EC@xQWjB#
z+^rn$e}Ddt>%PLOSN2`u+$(K_g~0GWUtB?ym|>MvY^S*ir_=1AJ%{2aj2bE?GILzR
zCLn<F)51uqd%N$O`*blw6ui>v0*1^1MrVWqZxmc;jM%ZxI9)|JOW(pPL%k>x`(bKF
zT<k^C>X+IZ?5Wg$t)K5izaq>Ed!0VLU(OG(nbm{X0A%=~HousZHkw}$3~Q6xYJMpV
zRbGzq=w3Y+tYJ1VGo#+&<WCiNEr^s?ZpPLrb_4sziV?b<18YYbn<M+WxYypv*YHEa
z`-(j-HE{|M&Z*w*5l%PDfnkDJdR|rBz8Rdb8Jw`G&0KAN7V_V^nQoekVtTS}FQd>&
zuXeDXyY3e0jfY^_&X7vWR&)<F2juO4xS#eEA6TpK5!DFW`fxv_jHyesDY{X3Ky8s7
zZWwrla#<y(&0`JKNCP&?GB|Rt@o-q<N1fPg8#NoPeW%XHf=1-F`|BjkMsSx}&8tq;
zK3l{68A(TfGkaOCzpDW*39t(@G*&4-NlWm;$TY9cdFW8%#&+zQ5DcTIsnejJCfmLd
zz8X`!4bRAY(3WTHUaoq^e=g%Rj9~9s2z$z>Q^0E=?&`VqnTBOIG%TBba`wP#e6qC=
zQ&l`MYe~|dx<v&Y4S{T!jhn5@ryEK6x~8G_a^1Rrs!drbGc<}(YrG7Mg5JK4epOBP
zrI<Kb>3RC&2@Y9td@}vcV2j@f<zjx*4@%_Rg1?2ErCb;@G&)+4LqGm#bQCK6R^xM?
z6}R!8IpdMT`HcI>WoOKHIQ;ZMhV^ar>f9xS?mfO3e<Cp4!q^go)t~FcRr7JX8RF?`
zOX-h)oo{8A><2+R?Pp4PNM7^;hD~=>t}Pzh3yc<FwLo*e`m6csTZ*pe>0D@wYqoxF
z<KxyT`E?UJi$%A!)_~y*YTBzjoLOVw#N)K6k`P!&KONjqhlxFuM1t{ZZ`xFB@3$`d
zCIj5rTYinT2k9)s*SQ3(E%mdV9(<G+52>3x?mYz;WDHo!*^BMR!p+&D)L52d7Kq&j
z^BY$l>08IkLiOIL-|2E_Wp>yvV(nWtI{201F|Yq$&Izq8myvNZ6_@d{KNOdpX*0nA
zzqem%GcE@NZrZS=myvNZ6ag`pK|&@Jm%jZSD1TbJjw?3|?e{DE16NlPMLmEE1KwTl
z1=1zhE=Uz5sZu2WUxbpV=Zwep*hz}<LnD!tctjm?Mjrk+$iw%S*Uy))A7BnncvMaw
zK7T(bx*_RPNoursa7qaa>hSsJ;pcBsYD?EET}Da(x*YxX^o_?apgi?ZNLQds9ltoE
zsDA})?gwww_apW%ef5c2X|#@a9N!5ljZ$<^`FI_F`TXHUvdOc3))_(4i+xd#A^QwT
zL|gP2vIj{PD*8><pyM7ow#8nKdf|y%Xa?rokLP~Tj;|kHt*%$_!b4ctLz<0mT8n%n
z`8wYF$Zf{t`)TFq+e+59?j3D|aK_Hs)PH+kD-mp5uK4wR`gi_(`D?j|V1!0}`1Q}1
zc*B4a;PpY%-GGO`e>?pC^5ZO({UL><A^8L4;qN~VQ_D{bvmj|A(ptn}*_5Q&nS(_o
zjQ3?C<~=(sE>Cn+Tpq0kAh08r5{)Fl;w+KYTw;_I9bg#_GXFP@oy5+HFo}H<M1QzV
z5~$$J%WLA?%14^=(DGA*sa8&CwXlYaqzytBtb$wXAcFJrDEV=}e()s0E%f8Ie(J69
z8S0U6ieaJNZmV}%pw<gekO}gQZ9q7*k7UEp(%DcUM!-NFstsY)hIN!*i;`9v!g`gR
zcC+o{N`lwN@k@+s@jKt)ew%T{cz@v>%o;$pn33XDgz9UKD}e1@i*mUw#hY`ExC!R&
z7)>I7Tn#y72<?{Hki8)Xtr1R}|1spCS5fZsBF-rAIRPhr-@ofL?Hd~=gOr(oevd#*
z&YA@3c<mP&4;|36P+SiQ9LL7aH5%NZ)mRZo!X)T?Ar2zpX`c5Uk8m8E=6~w;FCJ|l
z3SCe32*=$?sJ#%REd{2F9w~b<{c2mZwDej@fFy=F?Z>zk<ho`b#A5%d^YM$7dY^kh
zY5{Vwnql`!g0uqBJ=Z;;Fb1Dl=sG_J6$EJY_=bTPlHs=f0$>Gze|*0{JgdXPA*cOo
zXN>UX*$zPvfD4D1MXap@tbfXf-=lj5tQAUy+xHAmqG0vxnFkpaI@-BX4HG|=RptK>
zx24LE;SHLYm+$)7e98CKDePlpNT*BiYG7oM1G&n*-V*yM86M}2h)1nT#&_yjT$WHX
zN}u{>e%Ft|f%UzX^S!o~{Iur4nFEd-0b$ABx@^`ooh*HexIrXr-ha$1B2QOFQFtf1
zH;ux&eu>NY%xF{Qx|0YRA8CM_i6EW#5n}%d5mZ{3R^oUD6C5QzHTQIJKPvu6QyyAY
za?i;cdCGH=r_J1R9`$j*{=(T-^y6jyIrn^m`og(Z)Z5GI3kRA&(ITblFqY%=)Up;S
zB{6a`v{cc2&eGDQ^M5z(AC8pZyL=59h`vP6TtzDX;!H|j7xPN)^8~=D_k3NhV5Q74
z76em3Gr8I3h2p6s^L;Zj8DjhClrYTy0#E0yOgo2czRMxF8>u)@dN!)2xz&Q3!s`fy
z9S$bL1u9e1^D`G0_%hEp(hM45Le-(U$e>lfHd>Wgr&QmyqkqbNYwJ^IUmQ;BMY-6w
z44bqjzDUMWYIv)<*ypLAZXV_Sid$W6VaRr)8)UbNAGA-X(a$rlvnr)ry%+Kc!etSR
zzUwsJIz0`AK5APSeky?Ps?^(cWn+Qz)up6|cKD|42q9JcCyg9|7oap0FCkpvS*mlE
zXZO3A0hx1UV}CVF@U|W*@g{78hL}rC<|fPj#TC;;VGJxaGVFOBrbx|mA6O{kVdA98
zvbVx5{mKTwR8M6aX}@d*5oLwN`ousp*-5oS$G(7mTeGK(FKg(#(DIghSE^*!Dvz6M
zSL}k|g|_v)g;KNp$bo0ZMP1cf?N<-jUq&w&y<cVYrGFl4xPM~DvO6Z#U=!mkV{kx1
z$M`Gjwj82mK3vm&;JPv+V6n0kM8P(XpIYEH5Ab#%o@h1f<L%dT?8MtIuxvha0m%~O
zuT&0|LZM`sdt)g|5ZB}syQC09OMlkKN(iBMk|b?mB=JN}5GtVy8Yo91r`W|5$~szC
z!G!}uGJkVR0s*kxkxkkDi0U2>n2|4a0b@_$qk6zt`VJR7aG%6LAca|?d|&rz@x<7B
zk54&coAp+o&JUEE@38hse#Yqt!LH5qoVD}$Wj~JcJ4SVp33sF=>U@TG0mV~&+snJ=
z*`e=i=i58<7T*rC&im(Kf>35&m-R}YilxEeCx6{<YoYc=_OC+_?OCWUoBDwWu4DNQ
z7mwhGAOx`gNd(u<31~_7l?Xod-ir!G*evY<T*RukSBu$PKnA07*1wI7ZBnO+vTNH{
z7y7&<@u$q+F6Hv}CXS{xxy`l&G#mlDx~C3>s}NQT|1Av5UwO(Mh&k(71GsIwFut}D
z=6_LgJE>Gi+gO^VuAH`;?Z@ECVRqj*y5n`*ZwYDPa4~x8Eg^7RRC%o+x%`OU(ad1g
zHazNdRQTeZbVzYYrsg@Om0kGlfa1$jDFSfOp?O$WiFT|@S0>$bmLf&7V8SWOU0sjC
zmYPNP(yHx!ZF-!??U=-vN4h&9iHc%sSbvOCb<zOsp@IH3bB1m3lx--z2;Hk7Hv(51
zfd=N$TNS#s2PI=-lH8|zxA*C8eeuLLj?!_HYf(*k5K_DT)k7?m9@&HR<^ZOw8D3_b
z0+<2Qvn`d=w)L7>>}`7vSIQdaEVD0NWVwCnm&c~p607bs>eS{{GXbt=au3_RNPlFy
zmwy=8v65%3={>N%Ud0A3IW+00TQZtkOWK?>iaHX9J}^B9gNkXMzb^ASwS+q%;<aDP
z-b&NGvYE>rRofT>*3w>uT3Ba?APCJ)-Pu_a_AX#CzG3;|DEjr+t1hJ93XE_bo}GRx
zNyHM}KJs=wKe~auDW!!5*q1UzmM}1-tP7ad9<DnpE+n_r83|ym_+MoV%YI$=N&>Zq
zv^y)bf-TZs#Pc`#`X9>yF%_3#T{9GyVu2YZmmXFR3YV4sCn|qwZ`()`{qA4UK)%c_
zpk`my$6ye^xn^gOV2n76*;&H~WQkFj$fQ6~IsW~=YFV}v*|AL7gH3?gMY7mcUG?f}
zItFEg5-}(%w6#nGVZ37^3X6vzoUo4XXe(S~X6*$<&x96~0uv;HVP<VaAUXgMMNArZ
zj1nno;Z0CNE1iFsDS-0)(BhuXD_DX8iZw7ARN#A9QIlms2I~1lOgW*WX9<RcQGtaT
zFO0GI4QTH)D{G=K6nQ;sgb5M%pt6aO9g_k<yc)J_VwP``Fv&)~7qD@!A;BhLHL;#f
zfr_lV)502?b@!mMd`Ga5;&&b1pe)}RP&QLmShfNvtQ>z+z6awb6CeYYwaN3!<b>0T
zwem_hqge(q<@jBw;jCv{kP*%$mP=kZPrOG8pt89s3Ky~`DS--XDPqU*#)HSZDZc?9
zlX96NP=Qy*@}${u0nQ=qS+0+QOyZ~TAKC2ofTvK4p9WCb8v%S2`CT}heBLFDLu@cm
z7C_Owx8Q$-SBd!o0TfvVgnMV10=&G3W7$gH6JscXNd7=an86435J8p?8Hs?2j~RGS
z3;^ME;oYDzgd@PH67LZ}03ujEfB==)7El&Y%~BJn7_kemz((v1C=l2*puh*rGNnOf
zE3^3lVW#lIX5AS^*0VGg>DZG|;Q_G)B8QT<KvjQako)ZMl$J4sd&~@YfMR^wOkoQN
zMvssrQr<^IcqJV(WkoWE-v!)~@%bK!WECC!@=NjN>#Qse>$>TR<HcFWS1+sjqbQy<
z?YL}D6rx_87XK=KE1sU{<;ClA)SVbj(!e1aza4gMC`lHk4N0K$v^W%4{<!$HdD9fn
z#7}?LCYKxnRKc_aM;rf-$N0CG5-_cdY$7c+2ePu#WphX*i!Bv^_h=CyN+pCHU}M~+
z3eQfEzz9pfF8=e^f5mcPFF486i|O?A2Ce`uBqUN&Ilsm!il@!AX^&^aQ7J6HAc~`5
z*OhH8%<}U2m+sf&ZrGJVFK@hP>JBnqAQ^vX1^BU)7XT?%kIM^0j5f|LFMwgK0YO|m
zeq0>2&FHx7PKu*vFN!zim+o}Uv`52935Dveth+fwX}S38a^5W3Q8{1cW4`sO99P39
z&6jL)&q+F&Q@r2eO`s78ow>5B=F0@1#UGqxe1bmR&z}GhN3(H88u-<alO74gkcNMu
zw;VdhJ5^Kv6HZSXp49fzM}X$06aUX$leS$Gc6ZI13F0nb-r!n-*~cI$hKI1&nq6)-
z-(4+c^_ImhM10?3!MaT_Te<?XS#Be@78itF5pG4$itsBUtcZ9`EL;zNSV8Cu_`9S4
zcNBkZC86&?hae4K8!NT*rw1G1R2+X^E}z2yf${Ixzw!T{-gn(>{&P`Gs_y;bOpcoK
zqC7uqx^ncsC`Zj4!E*3z>b+e(9hKAwbE~vOi=>q#))l!o>AiIOusEL$#^tBt`O`m+
z<%elM^_)>Mp_$PDFn5iPxD22oC6A<H_tNn9#k3k84M!h`ld7H+e@!Oib~t~pdTL7S
zUxxgJreuTEeEoo$7H8AutT-Re5w^wiZ0X``QBB8hyJ34aoKEMx4X;C#4&Ad3PR1B>
zE8rtL0@)jNc<=2zsp?_-_0M)Vo0V-(@LJf~petM{=}eRwZRrCknGO8*!*tr)s+c3B
zLMGD=afX&(8JR~O^cp(!z{r2><7C{7;AK{JI;<zMX(z=eT@2f<dN&+(^P;Ew$w<v(
znfB5lJ)kQD9B^$sve!czc7^GxE)G|lkX>`WW^pcZR^8kUe9LfE1>Fu@q=U3bF}HJC
z2e{dLDykcGbfd!VQ%B=g9kqV<I=Z&|uuJE$!F6HHL=v;g4xBuEU?_ij=oV$8saxw;
z?rK`+w@m9?pBBZXs}{wrI+LY>cWySU$KZy^7Xvj0IX+!TXgaKL$Tu*@L&oi+S=)>k
z2-cp3-jP&5g^31<Mv#KS$mAwrXt3!%tLIT$zANDi_-rGOGb|n5l2zRHph#%sjQeqb
zOSA=CvVDN-)fI3_T>yWVR!ugUdl(%rjsh^=OO7t>(S#u561qZbzcV|+QTrc6*k~ag
zT?GbJ11vH`V<O?al9A%~q|-IhcZgmf$|Y=*P;Aj2Sj$Hs=oY+}%F+F_<4?~<Td*g$
zOLpJjy1+g+yV~x+iE$!#K09#2%*H;*!OKk{wAXa+?N&T_-KT%XdUH$iS!1|tK@bt>
z9CzT<v;@}=W7h3M6CZk+b*DoUzkO)p9RGfBgY99Xux+Cd`lRxxuTptf^&nw&Iucf=
zBVlzq5`3@r#PD2(ArkGiOjyG@E&15pDa_8k$Gle7@^kgEn(=P)=)+gV%i-s@&AYc<
zTYbV*RCQl_4r_lX>^8)FQp2qYPn&!&Z>^Eu#Jg*I&8Sw^nARkE9ZWp+VmK}4-PdV3
zm@KMs*>iNW<k0hJ{-Px<e+RNo@<$hoVKePlL?%r$nU+f(vx+^3OUzB2kpil?c4+*K
zR44>GB$?EH8jdk2F6!a3=X=q#=Q%uk&+EpL<ZnPc4pV>9+4a`)_eiv*=AJ6@?~7MQ
zYlCk-%#q)0V0!=#@s?^4;w}U5HC%i`sa4v6)A6ygr5!l=D~?I@ch1UtHOVn-W!6Z2
zGHY6ElH)3~##Lrrs(tr7>THf2K3jHtUaPdU$uFlScaNEAm6guR?tL?!7muEm@2VP8
z)Zo?f&UJrbb=Ni<%{?E8Ym9pQ(cF3`jmGGQe5L(%o1KDk=!j#o+Ip2Nnf+UR7f^`k
zN(bUb!QH3=z^7wU$!p^3XnoL|->RbDrRUh-x`<FkP~eA|CDZmSnR?BVDPDz1T8)s&
z^otH_C1L~9o8~`gNthY2I|tYLbe~n*Ustmq9e#gms`21+TXh^J&8M=RejUtb<*0gB
zjRq{Z?+FQ}Lc{g^Vq-314CS+)z8hPsMaQG+)yw72W8Tc8Nm-XI^c^+ha!}0&vwUdM
zmgB*qo|g0Z<%5rP^SK^$&ETxO+~sDWO750Im9Cj9HRW%_Kgbu}sJ6dXrqp9t;lCNV
zJ#k5Fa;-``pPt$u<qL0ATgRu|THkK5xJp!)wxGY^&jRi#Uq-FBcb)p|TJMAty^Rz8
zM8?bC&dGnDVKnW)sgsi_Uat8+k5{#dmyvNZ6aq0hmoZK$6PNs5K_{0`gflLG<b|Rh
zYk6c<k&?sTgB*T2y}h44Uo{R+cmqxi@81TX8<IXjQlmA40}xij@cw=H{Y6S;>3X9J
zrPP`(N54INoAW0Po_c_xE6@e!PY!`vw9WnWNPVxdf9<PYsg<5q@s2BZ0;H#a?tzb3
z=Re+mJ;@@n6v4)y@r(ca@_zb%=foATRtxF;@aOkaynz~o_G$ntg0{ntzlLw8-#BU;
zOBrf2GJpJt9lDmU#e^Qngwk4LIun8<yCO*tgpRke38g)MvQp76#P-X?e{{Ln64ggC
zno)Ggd?A^_09|g)9HsU1)%Z21TCGF?-<*0O+~`CZok}ji=$tev7D-Hhl&0N@En}4@
z_cCuuu19pTfWj!%Mh<6fUW%rPLT;`*<gsupW4g%>SY(jbN~R<$(aXFRiPBT`ZqzSf
zHS!r<`kAvzKJYV*-pYSFeG{RhhFl{PoY0_~Mz2M76}?GMU9!J^mbQjSW7S65T1Rap
z69`#o+qFDLTjPufve8z5k-1E$Cd*DqCO6saWY{?|8yW@yxImL{h1$hT(Z~@cx-(CJ
zLMSOPPPY1yl<8W&&c+I~R(ZdWoSdp8n>Np*Dv1IOa8%GY^;OafC+$Mt@fHo*bX}h`
z29?lZ(x)vJJLpC8ke2rUlU|B&z1vGKqscU$qb;emhgRMR)Jutf_tJ_i792-qwAw}(
zj1oJ4Z<SRRH>|SCnUz&mS<|nwIT|DmEaX(+N%w`ET%s0Ra@;E(l$RPcpWw?qOKD=E
z^kE?ZR&B7_OP5Pc(I7_YqOVJFaZOiktuVAGv5_9gzR>l7ZPs)}V~Km3OAr$@l=dYj
zXDQ;1o%S{Ue>rP^tno43)i0%QvqQ29PCQC|&CPg;v@jag;$^X5iTp7Z!wfH4y4<-D
zET-t5q=9)Rkfg7+eJ#vXm7%Stoypzbxt3?Y<hk>R(bkB#&@t_%3~$ET^-M|mAXEU3
zbSL{j<0<>Vqq7f;Q4ACh589`tdo?^fvGk*+FhenNGSW|fFUdeDwbdMH$8rpcG$&6m
zMBwT6<Jh;*R#7H^U7)))nN5xA<IO~W3V4o)ra9_LKD9NAVk3av9qPEum(>i4_1+M;
z6*!^5EHt_93*0J*;Fk8m+uCRpN~?wLH6Pr-CraId_eb-oM}3^1Z<w`>2W+edJm`Dx
z;X#jd6b~4G>H{7y-S^Z5+5-jjaAaMeX{Gnhe$c|NEoEH}4p{gZ(g6!ULp)&N*Rk{k
zb?`)Gs2@^D*NMu<y6gx6c``G^Jwc^RYL0l!q~@sWlbYLU`{J%?rcq6MG-rC$kL3&-
zYkhHNy|)CF%7Y&1C>}7>2RvZ9?^@g?3X4-_Un(4bRC221Ay#kIgR!zNcS}u$sg~+$
zDwiuDEIXkHq3k{<E{u7nr1nm`zPi*xSa$~MD3zCW{Ky_eS9VsrQNt|>^B!xHvW#Q=
zjNR8ZR_Jv$a&8F>4W-+&&JyIxz3Q6&b@D)|_R<|!`nD%$*fWiiw0>@tT&qr&srw|g
zPAwRJy=Uw+gKLf2Y<;_WRzaoYyv~=IU0RiVf3I3mO^N>KVo^ygf(bllpm^$t+W*B~
zqB2OaTRp#*n0&IfpU4m&mRC~NW1*@>*YY*a_aJZTP<9vDh)GtTc~N#Z=JMXPNXcdt
zYKlmDN74*b2vdT|k#_ES*5d$%dS^4eX`>Z?t*dP<G~;${dZeSY=~3@QiKdN$wEJ1L
zxop>_N7@<lnl=p8FvJ~Puk)3@m7X?$>Ta`plH*!f@Em1Vv;}8{Q|`!8D!LmKZYiY>
zS{A_=p}`+?g2&K4^aPJfe#{9TLw#_9$MxNw)RXnZ#M8ZQ=)j)v!noZFz1#;mp#!~t
zmrz$9^vdpVfz~7s9OZ)^?FlDbwmYF8Yj0gsy}(e-x*M3jdsf;Igb$M=FqbvThkVGA
z)BCscbUahi?UZJZwG?ho)FZ1P=jnebRC4PqNv==bNYD#HO}A2k!f{inJvQ8SG*pg3
zU#Da|N$=WtIHl<AFiQ5XoW!&}9kT|1!%G7@*Jk;8N*yKj^;7vajbF=-BDd+RGdt>S
zWzhVfGxr+jR-eB*0}o%Fxmk)()*EISKzDvOuW07gZ8-^SCx-o7Jh$HO>1sedRkcac
z@eHW%AFVm6U(^(W@Z8=_TF{c!7Qpt&$~R>R{{l25?BNP!Ze(+Ga%Ev{3T1AWo^dk=
zm%*eL6_*N&GsFV_ez(wzGtvhH@u7limyvNZ6qnIg0u-0dlQYAA&^x%l{^#T0Pm)=d
zBG~vde$9V<`*`}_(=34TR%mN)|NZMJzCaB^dvyz1SPZwnf4}|l^vg`Ot|e)TF!3My
z_4l8*>B+CG!XQb7(pqB%6@nza;?3G1bbOUnh$NH&`gW6G=ZkG9qZ6oIrA%r!g*qwS
zkV=tMYgBwTtoG%9-NTY+3<7XN+2~9eX>;*CQ;jR2y0+>C+4i)kzzGFrBkNDpqB6oj
zsFZEtojysPYnsz!XY^{Orr1$THgg@=dDwB{`{-0Uj)UebX-xc^R~HxsDE}n6h9{j|
z3RJY?i4i3yMSX&x0)aNn6v0L>rw_3Tj=!VFK0duAr!EA4D2)ACJ2VhWn-ej}FgiW*
zhWry7BT=B48TgTn;hi4nrEP$2OB{nT{`74zK@p1TQ=eRc>z8-qj{=V|MmS#sPu!xo
z*BC<t#^7NR_SC=X+RTUH@x@jQKeu(p8{<@*ap&Jq-$q)bJ(^bLMVZdRrO}RTtYW<`
zw25mBDdc^BWh5;qisg-A>@*L4Tj$^#&v_L2>KyZ`fHhwwl^xftktj9OiZTV(UZMBk
zk#CjmsW#iXCz{bn)};Nc8EH;tQGDL0Gi}+uh_+M$GSk=vD|#;950^tp#zvm`IVKUb
zP!8u=7^;$r()}yA$mzWqRSc;c?h<fq*RfuhtqQS!P>UP$jkWSP<~Udp<W|n(^^K&>
zuxhjZqqrn}Bb#W{`?0PK!LT#xlRQ)hqZrf>r+T$ymAmE(nUzy^fN%yVuNVmvH?((H
z9HxQVE`KfvAN+Um*Yc<|*75uMA>WGlHd;*!#;n^X0Mk9JMfbOI4xOQa#FVr>qX}j7
z<XuI7ghVrqGCj!UCwdgOUx6*8@YDNBo?FI4rz!@2!)i-#^r$opQ~vg<%}X0ykRsv1
z2y;=f8_ObZM%S#n?fbL#?xk&5;zY%#ev}~V$##Jz&Fl4(=NR94j;T*wPj`m)&YYKY
zUD5jh&0QGa-=G>gZr@i*CAj(;G+^NVVsJ=*ek5_)X{LG=6?d@AkjGIaH|^UQW?LE^
z^gJs?7F_OHviU(1Ci7jwavy9t72VjDgg$eglZ!KK(VpVWOeC2-PMcVX9%CMXl|tCC
zmAB`Rl|VVDvS+-Y`V>D(Qy#k!P^ptij)2?GvrhJ((s$vg@SkQy+_WyDhYA@*Dk@2T
zSd<(k_Voyt#yr^OmFf|rSXaSfsRk4no?K76XqJ3wi7V3x4I?Ywnl7cxqqga7k+j3T
zJoZ9?KifO{Kotq@*?aOb7%(p*K`~|B$xvfXo%2?;0rsUNxM>gb&L0;RXi<<Hr-enP
zl?tGxN7A#Imc&a%0-}k@Z87sjMWB>_9TL3wB!{5-?qVR%^9r07DjX{GSS3sWat{L(
zTH0JUiE$#kGMM^q<D|0vEE84^k4=upPY6@yK8mb%3X)f8CyjB&mh}SquA1pGw_H2+
zyK@#Cs-ZsXpri4<y&4XMM8jOJ=+;iBDNPTPZfUizLp5G!DJ{SRDt(+VzL@ZTo#Asc
znh@I2j*g>Kb0l&0sG)B+au_WrSDh)Gxr_y&<!0Tv5L5v3UV~a$NP60Kj`jS`7L(^l
z22DyQu>;$@21{Bf*_N*G?iVUk`oJ8ntjcyE*V|#^)ecmV%e}5@=eH$!!j|sEV5GF3
z5-_IiGk|_?S?Z{mq`YL>n_7Z@JFaUs-MR+sMTTlSJ66Efr-SJaBd(Ta7(}27Z<<|$
zjr(-I0^U&xi!;UO*9c+qYn{_yw+(G6t(2zh4SRm}_%d%u^3#)5yvS5m9VDi|=pgYM
zGnN=#QgWnI?LY)7-X{&5cr(RXA*J2yBO_H!4YK4=ipVOTp0pHZiST8A(8sDVR7x(X
zH$A|r6Sk{ZVzwuJ83}Sk;9{Si5ZDa}+&sFKzB>D3@!v3}GX;NUJm_b|zXt1Il;AOT
z%*JI})6Ye%#$K{})lEw=`?4@muIbvz>Q^!oUcb)W?0J4t_;`$($}zgG>HD<?+U>8Q
zMk!C1-qdDqqr`IG$s3}7rLz;<*Y!eYe#f08B8GY*s-_AncKEYD^eD>Z2AmX5u@!Ho
z1ulZALaIfzy1*TV&c?B6*nB<3S1p6Ee&$d<YL35`%xFU!BCjSh<LJH;I149K-^{bm
z5*g!!4mi&$CaQ8Livc1z;Cp#mJ88{D=?yF#y(E{IKucBH@G=K~o_2&=jUNZfCCwSv
zsnl}VsI<FY6S~TGN$&FuH&x9weQBX7ckRYCn9dth_k&DBI+ffr1H)rVkE(VH`0;#b
zUJCLuy(I+1zJh7V#i9ytB|xorOP8(I!Zp-GITGsA*ZlOXfZ<W^b651(uPr5W2pc)<
z)O|c@<OD6G(=(NSi`b$;3ILEmZ@+zohx0^(>T@SIe89R}p#g=(iZdGe?lpOn_S=O1
z3?Y=qneJ(+BL#Rp&#L`<?WX{VI&iwf@5Vr7VqlC`UDGSOE=h;uKV3@XzTa<##GtHq
zkUAFL9)(jsND^K~94LM5A`$age5aJss;p5ZerpK`ADQ$tPD}C1e;Lm59kl><%5N%D
zDQLO95>@+BTo%Nwy4PL0?p4RVJ>gz^Sa6@89-Rfm`KjZX^rK9_wz1p&W{UNELcg3@
zRi(Drp`R$@{tgu`AA|OX6O&r(`}QVNJ2}5pB7`vSw|qO9Whg^&8!NF(TT+2=Yf&KC
zsf6xg=nl3irwLQre}#mC5iOd`r&{2av1pkn!`{UOroAdV;WHN(wuMk1Kf8h;-^HQD
zw7*dlG1go9wn3_VwPpFHyGoN{kKX93FE{$Ge2<x{Sbi}^Uqh)?4jY?&QiYk{i#>3K
z^sTBDs&YnPx+Ck*WiHztxYFlViZbQy?&z?}Yd#@atrV>!e>X!{=LWT{o2*sYsbyzg
z1`53$rzgeQsy$W;;|WWpaP4+Uq0<+zD^6{@f-`byN+cM4`WZ1{v>VP__V2L-VQ60`
z&7WWs#yh^`71OLNLv4td7LofFUd?ddxnZJg=PN}+X3%#3)C^}`b;VUfjCon(95{h)
zXb_LSyWNj^f2QBrBplkL?Q)v@m%MYe)}L<YaX;2vjdiy4ewM3RN0j2^f=JBhgPSRE
z75z*S&!=CX)5wg-+heJnbJE(Tc$77Fuj181FaBkk?>1Yj$eYJ({M(hD%&<r9`VEAQ
ztw}>(Hbv(e2YLj^v927D+)oZjRNAMRy1nH2U<2hBe|j_fVvF2cfeQB5?(>dgtE1}0
zQSCEPI<s~NR7Gz!`?zN5nz;XN*_F|;$+TGds-rnd>0&6JM1JfhW#GL4js1VB=o6rA
zl5i<(KE};DIrjsrD1k6Lt+M^AD7oD+&dzJ5<mK2%W9tm(wdXW$cWs*G5cb#ZT1K-p
zJNy(~e;_4O0N<;NVfJN*ExMRye^M8FYhO5PJ5CWHhK-=drbvH9qFm5RErw-4ucJ?{
zr2Np0>kxztHIvnEtD}sH&E;OA?Yi=F9tUoDO}e7_Rng&+=&%-j7dku$2cPX&D<h;H
zrw*mF&+QkRE&wSRe?^qhn=v_%%HvDM5Im)kR%%?YT3?q8<aBx}pbnv02NBTBk`#V_
zoO4#PMWI!LXP-g<xc+vhKCDY|X??Okh8!wI)!R%TdTQTfxHVPtqr4iQ7$ho+d9$0d
zlJfoS<LUqT3S-ijk#RE=0W_EK#Q_tSdY?1I0o%7fpfl(Me@k7rT7bw4YaHL7qK9N#
zt?ANSwV_0R56T!;xMW*rm<MqMmu-kqpiX|Fmr+O~tAIlu>EUb9Zb#C+q_>jt(?~<K
zU`oJRdIi-BLABC(f488`x|>lZt}B|2$tUz;*>qmE7WoM}<I366T2LLgCrjlRsS3Xk
z<1pwuP$_SPf463V!t!7Uv$_nj_L_RH-Zql-0KOBUo1L1sc7ti#`tGzk(K5MPt>#C6
zL+cR35O2jWpwNtZ#94@Dy7)8`L5zE3!Xl4d#5Txt+_lW^)svH31z=Y@?W$r&J=cag
z+aKqIwPr)1zD*&*fr1jsuv!>3;JZ78$yM>EJ1!i!e@R41C}oFHoa+eLaK|nU4>NrY
z9O7q_i<AsDCzhy_Om6*VPW2B>e(qFpd<xG`oinO`ke=`SioiGVf#2g@JJ0V7G6-oG
z9uPe`4fwSc5`&WW-}`oaiw|<oYA4)+N$t=nxtC=dFxBjfn|hG4)e?Q53N(ZqKQpjr
zY4?G1fAuJd+A}~l&b1&~OeBV(XTr^sj-uY&RBf>njMfe{mb4n>=Hg?8H#Jd5MUAwx
zfMyKUAjWDiZnKK-V16Vk^Dm>k+6-Nei4V*f=Bi#AT4!tw>sfEBBi#$u4;oHM3Y7O8
z@>iO*FNoiC091U#vF!0X40oEob;)u(rsK7>f3`Zil~7%ZBL(bM%J+0N$qP>n4o1Hi
zqHVGMe^wYEDq$S(b91vJNMdMqHl*J0*42KILClst^g)l_NNWz?S(Zx~=eG5Yww6}I
z%G%RVt3W#pC`bdjKrhv>Q-|oBFhTK~eMNN(^I)sZLNi`zarUIrVqI=VUq`NyX-({F
zf8wn1J!A_VqG37gbD<7^6<TkR{(p*XD#L(X&z+TR5l1AkCuL(t-ApqPvxCTMZm$aT
z3#m@qa0$Xj@64`RXF_svqdiL7$4iQ{ZGt^h6Et37E^n66r)uX+==o%;NU%nm35KRe
zEc5ImK(4Kvd{u$|M)Z`mI_n-cI#1n`e+9i8_vq_)J*={)b*1WBJ_HiXE)jVWQ+Pj@
z8dkLOmtpaaQF$t_ERE|NveoAt1al^JJA9bNBn}$PHA*m3I$P3U*61W1pXEGIBIwbf
ze0#N_)gX;ygA>O1r>kFmHI*deKT7=qYbRj^CP`Z92#GOaw)LN@EXQ0?V>HtZf7jS>
z-dB||y^=BT-8ByT6kZmDUI6Pi(hPl@S|1B)L5b*gsC_q}*G%eDWl0Iby0{TM$gauG
z^NL48G#OgAFJPudP5^#VDPmK>=|USwb?CkocOa7x$A!uyDCzI&sn8wMsmW2t>2DU}
zRohxk>Am`p3e}Jyz|MVmy;UK4e;N?q4R5TDH(<c_hP2An_NTPSOIpKgMe{o|bDqpx
zFz5wG-Rm24<bkfu!PE)m^_dCrYSn;+<I0PumN<?C7)5kQj#{$V`^Y?+onJ^xI(;X(
zq#0HC9z(jvJ*JGKA#s13*KqrC;ih9p3yE{t@*|dhF8fUbqIMo#`;-|ne@-Z)7~>v>
zZzD)v+a+np!ZU!?>+;1)8=-8xXw1kk=Bul(at~S=Ej$K>^J=N>)WanKE|38Cz~pxf
zyv|1QNergd(`Zy179Jz9-CH|4bbG89fvS1A(=@6!6@WQY6v^n_`Hig++UjR-Jgwf`
z_+XTNxD3w+sb{(z7t~LqfAt%u%Fx43A0Q*3gg)7=t8m!Iel!?4#+((juK3XQ_RU=B
zNqM!UtA_=ncQ&_k-Wd?;Y1YxjJ6W~Fl1)F=pt1FdO+9Rb%$9kCGnJn1!N5DMk#wzS
zqmHD-HCh$^oN5Y03{P8`X?5->?xlfSp3!9TXW5+!G`pK^S3K1pe@kavCf`CdrSj;E
z9x7T)we6Gim{((<%D7%@RlT<IHjmPliW88kWUW;79b1ojQtzVmvF>ln`l-5RN$s<|
z*I~>MQEQZgu?71kZKaCjUzjDivmmORccZnvr4HUJZzE$RO~&@!P4%tYOy~1)<Wcde
zRr)PnJ+g3fFxNO9e~gGDUFb>B!=DDEm2}Fq?6W-FX3%JoRqi&P^#9SI_(z9))S5r2
zXNDjgrg_u-w5hiUWT^R|YAI9W+s*s)%vW@kQ<>tRDaFS(H{YvI>(!qs&>wcCXC?jg
zTgoiGtfR)D6W%h6r9I!HFlyvfs*-ZN2LvR`@}6_FMBFTwe>T!vqQeZZq&<a_ToWf*
z)K6#|;om$fw$%);L<CspWKMt7Bt>+|xTZh9v2Q<qe*5ow3>Xo>UjF^(xAY=$5t7_F
z5skb2^`Fb{-~P^_(qHjr@t4jet4%xgR(MwiixsInAQ2$_diTMwwXvjFo-Qd%SThvp
zwN1s#+wehKe=Nrjcn|8n%~Po~MDwiYo>%i9`g|l?I_?g}%nEj5CWWLki|pshM~z~K
z+zK38BYv~F#Wx(!VCJ7y>ypG7y~J6H=NVH3YInmrJpFt01IB&K^RUWq(xjc^+?Mje
zpjAMw<7&3cPVo|{4@Bbc*-ytF{iL$nU(&ZPv%Y3CVq>~?;lI*0IJ~Wl+%%Yg1cFbS
zhUkTt!@QaCIS_3QAfUg29xF=Md)-VnOBcK~$mZY6j9N&YaD_4ToA^^9&cie}`i;ta
zegEs)0jMaac|G!*<$mWU%vbx1jQSr9TS+9Bk#RE=0X3Jw7AF${HkYxRLMN9>turqI
znHraStuq-0exdmjG;f!hturWp?^rj%*(1ZABYcK6d*0>KCcg3VW`uMK_JYUm{`~mu
zO$*DP&J?ck>pUu^R`_MCWVA-+f#Qn4nL}iX1pK{^`yGV~8ZlM@je{dn4zorMnwrxE
z+8X5<K?|L9MJyGDi4QtvU0~RO=Wk-|lAX9t5PpJ2A0ZP&lqbsYMcyQTxiX$4Rmf+O
zbeRH{r<+<?mnmz&lJelUYM_Bmo`dJOvOH08Cz2z-Wk!+)5>&_%B*?_yb}4G<BLp5q
zuR*XEohE&TNe;!YZ5Xt&<bvaGH)=EdV(Hf3e2YF4V9jTYAC(Ti*Fmz}xs$+WX+@IO
zD>r*5Fs0FKwM_$EY(sm0vBa^~J_&bN;y3YaRxIXgsO0lvj|6ed)@W=aP(=?g;h#Mq
zLGI<eZ_dYfp`8JVj>Jofu)@0){$bX3@&XiGM>z>sJ)cJ6EE6_rw}qBmR5${mI7hB8
zTu<?!D__Q1tI1<4O_JZJqiq@x3OdH)m@`qGr>;e-p5q!X$L_=lTb5ugqV+N9wwFn-
zGckW(pauSE2BWPqoHu@bAAY?3-lNR{uDs<vkdKsZnw@V_uT!Scjtff@|DZW3;6R9K
zWKR|SOrpirW|IxXX5VMwqBq;3^%Sp1$*`4ol2o<i2U)k2GcoJ+w0n0t+38eenHmD-
zB_8k)&siH_IRBRXX6qV2Q|wLrT+cue%7}k|<{X`3NU9CCwNs#rkB1`lxQxL$?o!Y6
zti()0fh(4>k_~%GD4Cu;J@bYlpD)s4k;KjUCS~!1U_4J9jKH5O{QEfri-v7wZ4`^k
zb9#4pg_h0#q-7VBL06(I<|~)!+mhK!^?IBhDbVA9=#(|3l7P;%Oj8AFTP9B~wxNF|
zPrd#yV_4VfT>Z0E&6Tc7_@Krv*v0&eJLdfHd8-&y2(WII)P-E`FrTKajl$8E4$oHW
zX1Dh=bsm~jm7gQUVvXY@vw>_CC~>V%`HA<Q<{@x@?5X%DJ&)0i%7I(z^m)t_lp}w$
zO0E^^G=T<1qjz>~0_7G%g|Sp%>vw<L4}NZ&`)<&>HH5R!iga-<TQ%YcD5-jSN>Uhc
zw(J54jwGyO3o57TA}47$ahT6lP3L^R9a5%-4{fN@PD}s-UyWrH+Qu`sZ02(-#XK5{
zG|Qn#Fz4n{PTolQQ){t%_X*l|&F4&=q}@$H<D0jVDiV*Uwh#$N=Q+G5Ns50IW0mwI
z98+p+(@$@86WGhJBz>fLxT5_$<abSVXB{ba3*4TESD^db=+Z*x(w$qucF;cUEQyEJ
zLt!bw4>s7&t@YP18pV_Lt^Qi;IKjsceoASg*~$I&sIT?cW;V!$P;2j!E_Cb2>bF#5
zvz_tc?OJj)xr^Dr{!QMEtGj>UQqk9MeJ=8DCnI#&>`Fy@1<xAme)f8xZ`A#$02i2Z
zifSf+I?WyCiR2l%WSWAJXF^SlT+^JrO$Ou|(<3RPn656fo?w-u4cwB|<91d1phiR2
z`ONASex0AHHWYrFovN5e10q3>-Ne8#)8fZ80>!JBbkTgC8o}xP>g8pPAy7+};Z%3A
zQlut^F={e@B!-)vnqlvh_kO{!?z)c0O=bcqUJHfSC3stgvfR&FZaF$kuZRHg-C*hw
zD%Fs2n^%$N=FRpPN^gbp#d+FmuEQ!%+uk-7PirZzU#cX-Q;jk2m)Wv2>LqtK2KC0N
zws%%Af^s2r%q$wKJ6bi#zb!)8azE66^C0^#&C$1)!Ak`b7&tgKG9WM@Z(?c+JUlN<
zWo~D5XdpK*Fq2UqCYL$1Gb#aTmt?dvWPicUkJH!5X~@oIF0bV=emk6|*J(_E65GIR
ze!fg)0|VHQyve~bX7l#TJWbz5)b7j46-(tq-rk<gV0lyuvl9v(Gtq)VaKbAZ=4?aR
znkhs{iX>?Qc^7I(5=S9`e6cN(y+0<nh>Y0+u8=EIL2jsmh*Btwy7r?t=bRMN?SE&l
zmZRha!9Y)UVUcuu1YXz`#V{osgtR7)Fh(&H4-0ecqk33a&OJ~N<TQF%$>9@-3^uzs
zBD%xhBSx_R|IZj>vf0y$S69{owcp6nplQWZt^IDU1WMCU(5jH`T~*<ri`#ARVGlH&
zV6+%wS1M^VQKO)h_W#7|s?!2njDM+f(DIA_jtpptZvso@6owRdVR^jL(v7HvBgdON
zcD^v?=Gve#R%(i=?A8%GbbcAlc#yTdtQpOUvX?0snzvghpV2YO><Bb^0Jt0vqD^@`
zTt%TwkE5+T7ZSZRzi<7v5iR86i$B_U&<fYld5~WuPi?sFX3~R?50_$UJAbv2G-~q2
z%B_U09b3DF_fnR-5O%&<G|OYz$>sj`Y=rN66xy&e0$-7ep6`R;lE*V|yokMJk9DXw
zELIORzxS5rey@(21!D^i>*&!~QIV<-KF_F}s#fUt=vpNzQtY=&bhA!Y3J~Fu?}}@F
zDjUc5I%)%G@2>R}$mWbIv@iYpc&#;0El)-JK%EgRr4ejZd{_}$|57ak++>s+eOf#i
z%RcmFd&(cG-!9M#XP1$2GZdG>JTnxR`}8L$w?elwJ_!uJ(XoC1@$G+|zmj5?LC7-|
zm%&si6PJjYK`56H#x4Z6;lMLf2My(4Kv>A_{plZnM<f)NLC7-|moe%m6PIboGra;W
zytgRHGsXl2%1Ig`myvNZ6ahJx!Bi;}lTaTgmqN}np9;|)a;d`yAbUMYm%;Kg6ag@o
zVb(4amp}<BDYre(GbsuP)@_^jKP3;Qm%#Ef6#+4qVb(4amyKmOD3`z4GbsX6>zC@;
zGkJgg{;$u!y$BHew}%Z&SZ(x9&!l$VtZ7XywMna658p`|b4yP#S>m(}VGP{8R~f9X
z|0@_>?xlU&B)^GuyveJEBG)r5I%`*Fnifsn&1yr^!Xj2}U2y5!MWL3z(jMn`{_*pL
zR+e`4T8}GdxyxQO%A4)t(;b65a8>fMleB+3V+WlSE21va{@_YCKhL}6Ids>+3ZAaS
zIvzD^t28QGwXTDbS(QiQy#IqW&`VHP+No6-HR`JDnxrVt#I^GN9`M`3Z_-4TH|_kp
z7Bot(LhYq8bzk}p62|GCjFGEyVVo_iTs!~Me)S@f|81!8qW=x{1Ur=au10Hq(<*<c
zIrN`UN96y_gFI&A9@V5mFD_Saa5X>UBMJ53cj}XhH9weF`)WbE(OB+>2IK(QtS7x-
z3>dr`Cfb5MBD#c=2#VnA50FQJ)cfe$M{$uz{s6CwF1thi`Z@gf@%i%Ko%yX+s}y$m
z_n$BMhS74>K`l`y*NMwtzb}8h{C$61m8Xax6SxVNe60gFDH7tD#U&^bQIm7&=EMlN
zMMV$T%t!m3K?L2ZBn}PV`C4(gcgb{Vnb(=8URS)Miw2d{kpS8Wvr26VGd;5(2$mjH
z1kIo8zF&o%2$H_C@m`%MZfEi@g`kQ{)79#RpuqreO%PgL3g$-)2bwzwTU39;uc0+e
z4ZSMSPO$>MfAli}`tVsAWK<j`>+HOTyz2uK*yMZCyAJfiU})RSpeFtzNP1=Axe{h(
zqvaVAzPBn%4q~)G1Gu`EOh_;wi~a!u>r?WGVs_WG*K|Hv?bc!d;UWLeWc~qqGYo=?
z=D9>TFQ=))XFXsxOXN&y!P0+Ocy5=VRwaoznI9a4;mM}1Mz#TZry|fdTc4`kxYzwE
zo&S+vDb&9J9}6ROJKuv(sWI_G4DXE*5$01P-kh!I#Hu2tPXoS^<)PIKtR~@iUiA(H
z7RI7BvqzM1wg3hff~ITaz2}0NxF+Vw$RBya0|^ep0ei)2`GNhz0Oyxo-7{cSGkxgF
zII_G7Ir<Ht)*y?Uvu&f+5dHXs!qmJNnNiXY@J-tiUp9Y%de8(}VUmhK-@w#)>O{RB
z%mB?-c&L?V;hFU=ORQkINn({^%C=3aH-e+&c9#;~GldW5MfoLo!1Bz$ahUANJ$IL;
z-ZT7v>6U>_(6%`cnxVMiMRR8u(|WXJm5ukCuBrPt_NdE*#Syy2jg<`nnAj8rx~|r#
z;R-K4gKM2~qm=3*T=$~~xW?duPTnsef|9F`{)>pvO-4HPs;S#>+VpFq95V&xeIa=?
zLl=OqyVg{+kyyZ4Nu*MJ?35a17Plv}bW)Rl9A`BNaYxx@$eBg!<urYaMG&$G6DIgH
zqrcQMC=7*qYH*#=;(cd|M%@_d-Wh53d;|}`iN$t>B4Cg}-crYmWdgX2h-)yaW4gYP
zi&_=6RDvF!;jajfMZM(lY!!p~t>63*h0o$aMZLzaMFXGOVxccV_Dyer2LeprdOUo8
zL7kROYNoc*=Qufzg5w?R+ZaEUVRkUEF+b?(5eH)512C4XGe+&-Sg#TyrXxabki>J&
zS(5=f9B`-LnxowYdl6jw(dC%8N3O<vIYXWxP{-J(`fZ1Tpb^^PB~KVpgU@Kkdg2bH
zZrr7C#?%iO!+x(AVrneknPs^?!j}>+qOu90oxEzZZuHz{ENkd9o7KKP<r;lrvw$|M
zYY+Ui5|Z}h3tX~}#Q16Us#X%0PvJ9xe`j$kW8p{_<F*?;?WvGoV+PQKN4hkIZ-WI)
zJH{-#(qW8E&3%ydSG0c{w5ub>HXYc>WcN~sqvs&1@c5cb@EEcs&#jt{^ZDU>UFY5o
z#wq{tE|Er#Kxd`<m3jMFAGaDHT8>(<I_$V+no&;a4+1rJ0#)m5x;|(c%$~ZKmrvp|
zFMm&IKT$U7{PT8*>%e0NJ?L;$%=o9oOS+_c@2^^_8~w~umYs_Y8elHmoiR)(v1h%z
z{yiGQi<!<szJIynnGQ@?1s(G})j>D<SO*pQSi8J`r%#N#HM^5H(Zg#G5IgPG`A<@O
z@t}ie|5Agj?hZ+j5l4Ng?wPxHl}JZjTu4?($I@f2o38BP$(`n7{8!m*XBEfGnkf!S
z?N-4X2rcG2x*VTlp>(VCm%XbS{S*s9Puqxd)do<VyQA~DaQyz*?!B-lXnU7L<1<?W
zD`HZYLH#=u12{P{ml5$V6SvFbGo1wse%C-q8}=`1T16w5k#RE=0Wp_>MFu33&ASVi
zgxEMJf6ZD=bKA%fz57>8l`oqE2h-E@nR2<ZUdM^I;)=`Kt$e6-pd>6sFGa9OP?G<B
zUJnI{45f$y#gq@W8bHtBP4i9nYXB*ktc*&Dk|GW;tWnk$gOv4IE=9$L<P0mvQ4FuV
zEk;m2crl_1rkt@Vxcn+5HYz5$BqMSeo>Vqgf6iD%<ch_xV)1f{EXfS)VU-&L58d85
z!*UGE;hc+d15b+l>JGpjysU!ZCsV->%P=O1g>_Jr72%eaS!Tvm)|Z)`Q#M2y>yol5
z$OJMXn}~b349m#~OmA~phVuvq3l{GbQz%!nVq4bXEvzJW^Fgs&okIY9!QvCFg48pt
ze{6A&K#HT4s{!Zahk#{Zd6@;kIY+riGRm2<-r$s@Tr7`YBuLakP|iESheH=hVi1yY
z7$^70%9|(=$5OrZG9r@l&dSV~-UlNqf{TwOR*{v@POwC$0$ITvlL`n&`4t1!3nRb?
z9(TY(gDVV3xae8@6GjG%%w%9-?V03Ef0p^NkPunsPXsF!e~>5_94W)fAV;LEqOk#Y
zKv1IzG9p`r^RiZeLFkI*Rb)74#4mCvr$|L7)5)CCQ?YP0kSrP~Vkm*g5lpobri|H5
zp=gE+g)#zhE^7s|!APt|b_a{AEFT0iVqqeOm8ZgRZ=G-+!vhx!R{`Y}uEIGcf7ubj
zz6>_Q0&0_G7LwV$%uJ-Rv*nyu*{3o~P&oiaMsRZo<rI#hpvbaxY?D`S-l*!QX41@$
zX2Y{uF*f%XRec_|Z9ShVyE^>yP5YOlcG%X6Rwq6*)Ar4q>Vxb;gPY-G^#O4u>v(kl
zd?Xw$2Y^UqC2@Rt`?mT#Z_bYDfA+Zg{Qg7rOMTOxynPEC)i2*>HSRpB0chIR({>@r
zq>QS*)Qe_0KdTq3{3545)#u~kyXNND;1G;cG20W2pJODvR`*`b{QkHkEE)CBfBhR|
z01HS>my^lKgAQb(mc0_3)57d~&%*?3Ej}BOdv36-+X|MoeS&3qEfzsbf3|oaVjEto
zh#?DU#?E`%)3VW$lMfx+mm(t5c5L(Rv8{fbPMg(4@ix}SrNqHLxw~m`Qq{X=eqPVZ
zeCDM3YxS}E2?to!3GRQ^9<#}skqm7#I+Emcr;+qfkBnv$PhM3=%hOgS{M&f?d-dbp
z^k?n3nN~;DuU|gOzaK8ze|EO`V^y6uXAAvxIT;Tp!|6yj^HEh#t1GH5htuV7QZ3qV
zlX@^(j?e4r^*`q0wjTVpoR--ZgX?j7F&NHA%S$YHP@s>eqrqZ)Ii3vXgSHv`EHe+@
z4Hxwvu<m<Wuz6xnt~qqP@gZs?>>WH^4BD7o=7fCENa;Ixf;`%xf6N_#wHrizEImAz
zwJvV47J8%yYdzhwmdmx3d|DaKbOv>X?F{Y=*N^Mg>L@zH2F)mQ?43ph@a#2e=gU+5
zvN>N)>P7YX))>$|XP<-H`JgSLfS**0t|INBV?8D9QjZ;aeSNKG-`b02S|Fd-SJkYZ
zE#$v}fpD!}k1uD<f4p7vLkkV6g>R6%TBH*6r#4Va?2_z^et6HvSK~&XHJ8=tq!~%%
zONtnlAw6+=);6u-OjhTx2TlaCo;0(1zR;tl8BOYfx8%E?5FsJ82VpyiE-(@n?-$Bz
z$+_)CUDzuGSmolL2(fnvZ4;B%qJ|xcc4;Q8$ekVtN2mbHf724VlIq_BW%d5{=dS1a
z>!cr0LdZZuB1jOZZi-HOD0)IFa4A0@XW8~8gtv&t%_hw1%4l0sLACe>rruPq|8!Zm
z!vP3<_WOD&koo{MzJp5I@VeEN2)MK>PkZ`E*q3n!PoW*f!&jyat-4mGo?Dr`->OWW
zdsU`<Uzxn^e^p~_OvrZVE19)Hw}t*qqwj|PNA0n62%;_8;VDv#Fb#ffy=WKJCcC_Q
zLL99*JVC4nJelpGgv?_M*{ih)c8e~-^l5DZJ+WyIF6f@S^ti3zVlNGs%>7OY?pBrs
z-x;AZVrO(ZFLXLDb~-P1IxpT`>I7fx1YhhY_{b+6e+bnm8y%4PC}nNXG&zIZ%z3Zs
z^mTbT8;ExLdHLzHY{MCT_KHD-o&^eWBovwCv_q>cf8~*jh*U%TvRKE9#S+4lDEU3n
zS`nf5U>VtO&$9nG*qrtROBTvG-=jr8KuiZgi#o6;TAPFR8L`czB@r-E480I%KMmW9
zzt7r_e^NqzLvmUJ4neRd?u5-I&z*rEX~SagS(grun>fe5z@Y)HwC+hW+AK=Q<?{GF
zM4h8QM3&&b`;N2I^0dFh>YYT_9R_PhWn=qh2_`Mi^sPg~omQ%wL1l}G8~dGsWkcw$
zEuk?mC=pro8jjJ(n4AuS)!IEM8>OH{?6oMAe~nP=@)vXn?jVIN_RI_HvStc*SUbud
z1nV6Oa^94{F+3rG<3qib+ri6&F_z;C$-WK6``DxC=gPjsTe2@vAH|#OE^X?D*Q23)
zz{?YwrR|>M0Z&(Y4)<LLJ#7a)>=N{B{wwJ53F@vn+{Fh<_dvbKOFd7{+oa6YCq8*6
ze^TZLNR-i|`_Rkt5OjNH1hd(CM<=g<caGp)H<$0W`EhwUz3RDg*SqDecgwk77sjj!
zd)P6mK8XxB@@!B!=XEf?&m)n`nSM+TUXiv&_k7a6;tx6m$SIa@MUvA|v}_DITRxbG
z?fYK9>VV2)QT^G<TPoJ_c?pDfrtqSLf3+t-gwObN2hdByCC4v`5F6sEzS+QtT`V77
z{Liep?(le(ddTbHD)o@3gQS%2GoJRe!9&3Dca~siJHb$|1Vi^Ps^vSBl1}6^lw0HW
z!H;}|a%)h(Oamz<FLVQ1Q3FsMjP~e~Lcm5bJ+|*~elxx*FG+@{qAHEmhMnuFf7%@q
z6WWtIc+OcUVE+Q}`r)z3Yx!aen=;D-C)QGw<^5tzy#K2k<hYXSK)I2;u<xBlYzP|d
zy>h`6rESrmXJls=vQ9RA2Z@dih=*<6Aw+AgQ#pP?gpVXT+T{bAO|LH7MeS4E!BaAx
ziQ)hGg#Wpz%9`y|m7b#}{{!SD0&ymn!SXW|mm%Cf6qoAqGroU3{Py(oPGm*}?;L%l
zkM{2FY4>f{vhA$mQnBZ^9l?#3TnEJz9(I;}|AW2k9$TVyD}g2HpB3!;D{COXWW99)
zdK1oBJJOpJP^MwdbzIY<q&M?L2n!t;$ZH_GA=%|f#;f}vXDJ#13)q&Ta%nX;#8>$j
z*&RW^ZcnM>(oBDNQ=-xJJ1APuoz*SH=mQd$Z72h^{}(BbAVbPUk2Y`(11(XWGv(F2
z4J>fmd8g-m+sOdUhN=hH#CW>k%Z7Z}(jC0rgbL}B?X2zXXl%AC>}!iv-s!J4Xs(5w
zIIQBGG@s(IHgw!#%k3Ptp=fVE?63``#$j7$Y_O4wV%mRE$M^5=0g<kyDCPYA-f-vT
zM3dF6aH%ye1cu>smT+};!sCH5DX`)a`z50i;Qp|8f}U0ah!1nqdoK|)MD&<0c{k^y
zD`UpOBthw_T$lGe7c(jgn7Jlil2BP0;bSplOyFMoHf92X+WF6k86~*KUK_^DnBpbA
zYmrg9&{BUq<+~+z#@M!+m)t0>oU9OV2vL*h*lSO^>>MT21s+VdVca$p7vi?vf6;Bh
zTbC>zLvHw7C78X7aeIL%${CJ+YNaU31&(VXic9RpRCC{_pIGs_d>ykGXT?fsC`pz?
zaa{=;Ws@i)$S)5kqezq8Cu*m?iN}~VOxinkgnoZjAnD-Y_7kd<{^+7y81vu>7Yx)P
zZs~y|f=doK-hC6JC4*&mB<U~lMen%xew|P%NG|lHLeZI_qzx<H&9SBiLa22dYs#3m
zwOOs6wa(hH550?@;+<oDbZ%<(EMHI!m+<n2uNg1SMZ~~PomNB4TJx9d-cr#s)*>UB
z4U2zXLve$m*HBz2dJXot=&dek**h+YFKxJj-PUohbl`A_iS|~Mt=hH8UfMxsyj&7Z
zCvCk_5Umv#!OXcq95dXezL}VgqXyZ37soTLESko)Wz;yQdJ$X3Zw!bJEMkKIyvDMB
zSp;*b@yDnWGDypUvO;hq@$TyH*O#B3e@B06W3T@ZD*8^8ISHS8OScdrK>eP5&G&)h
zsR0kX1K&8&3Q9>pC<`m#ne7Y|C$$X&=vRVvFtt%?usC+R9*A{=0jTgm6f3ms_jL#?
zT=iQ-nJOD}M^y4!y2bjkif-u|-WOD?`jhQm*zPfF)jI0SkwkEQu+%5`fV16<RwaK*
zr>D}<Yqo&k^T;mPDQl(k2)`51p}ZBSI815anhJ*pc$;Q*w<>di4CF2#&?*!PsC>5Y
z6MAT51*A_$ICfzjiY%-N7nNDNNyGj3=eNA>6{{WwpNY<1D$NZF*&ZDT$0><u#nTLq
zc5|b?j9#!y{FS7kbCF5A%tcp20NZ~(vH?3uu6-}q<7#C{$t`NmvNGDF(knru5m8Q7
zCB#~EQVF`#T*;XOSSryM_Al}}kvY{=L`|O>$tr|(2W2!Y#_b&F$SzJB@J)$Q#!C6J
zxd^!2Bl?0x3@41d_q=*E0w@?eQP160DoTB-P1UuihsejCLbR$HJvq`7H5PxSK<cwR
zG@95udq*izASl12IinG0%^{8AnrcczK_2!-=@LM<y@Aqve0--S-P8F!J7ij?+dak2
z!Yae1lI^O*g}+Yy0(7RzS)q=svKg(*6P}12@3H}-Q<*s28_Fl^otZxvlQnyc=n8&b
zLpJn1is^YUS#QToJ5+q)zdL_OS7uS6@Z$&5siCxy&dnfL$~-I57<55<Ic;KQ(%>NB
z%%#CTrKL3Z7aODHR%JivRZ8+cGndowMkVxnX4%m$s_y~`BMv1^AD|KD&U;tk0OAMH
zr-qw@jk|>EWka6igr{WtyX*}n_uQ`cA#+zLt>S#a!({Itaaw8zDJy?_l|9xKt93>B
z<74wov<qDLuAOJ1<@3<^XUsFTkmx1lhDW=G;s!^%hT_7}u9c{n#jiE%iP@*>L_Dxv
zKKoR@<I0;2Cr=H<h0#;H{|`HRS|?T+K3yo*M$@NtJeToPgB>RjHyh8NT8`EYpjysO
z&rxX(t)<=4oyQcH+<AYDtp=~w=&GHW>3vhVvokdX$;di$Gqw5mXy>b?QlrGA%pp}a
zKkRaT%sNFTp~BSEDICJ7hI~ey!m*lL@7__T`W@o8e&Yfey>(6|Z+97=lol&s4uyt_
z&C@J%91dj9@^w-h+1Xn@Qv<$=N2CLz&6LO4peAf5{f0r^P+Wfq>IVDcwR~8qV~tv#
z>h@+UP<7#iRCZh|xl#qxTU&1@CK&KaB5Bpvt~$@%z|l%Tpawfz2SoL^27ZIz;z;fI
z=+*1#M(WHj4MaZwm3(>s!3>A4(K`P>{u(E8VzzwA(oo!>6K^Okbm9&6Qm-5*JF8y4
z#0Wn7BB1?*&Hn`hSt2>zm%;Kg6ah1rA(|Hx0Wp`+PXZ}_&2Hp248G4(_ywXB^)Fx$
zG?`=;=xvMKiXMtZcYD|^`g7>(my$?HmOZgInQjgqk7bG?De@yFSx%oPIlVvJKONqk
zwVAvK7QC9CK23lxB>sZLMrS7vAe@Hj>Ff0KQA*|Ty2r~%sWo06{|CI=+Zzo5DF7(|
z-vHd+cmv0OLZbuSRiec{r@}*>@!8$p7zt9P-gk8A+h3l3JV+As-PwjI2phfAYf?II
z1oCGo(`Xi?G&rJ3|8Yc%a@q83_#H}YlEfCrGpLh2)fQh`OsFlXH_C2MUK^R~?5!^p
zR+lSL;224ZEgL&)g)`O0x*J`Ea0Gp?dL2@$@1;C{kw$lgo@L=D<Cq`)f2?oN7AD?8
z<wjZ&f7Zs+7?SP}WJPXIbWnMw7FQ+dY};h(wYzNF?(M2=QJgbUMV&^CL{a-8|DxAD
zNh_)Ex%BV)k{-s-bnxpER2th^H1Q9GveQdt>sDLO=8f$cvWF}@!f4@wV~@x(_cP!*
z+F0j*^3-dhv-Sh^pb7PQMiZi#ajCWrT_nLGZkl|f4<)ZVvxKwR52IHhLa2fSBDw?y
z>GpnjdVD(kwKSpBB4{^#{CY?)j0T~DniRf`cKZ9b>C@rE$d%AW3W;6HkCoHkpQoYa
zCxXr>F|4J-_??9{NQ`weIR!%c8rJ5aD9Dq4narSIfVlvx9F0T}wj)QhYX($xA`_V;
zAOB>rXDujp<!H6gF^n8xE>ENO!dbo0=#GLlh2s4NU=R$auyYb#C%w`}D6Mu%uXGS5
z_=WZQ<RCj3HA^Sdee&=vM3UC>beq(XtOnYt{bV&}E&U|0qu?DtRl!u3TTjJMoGamf
zp<Mfg&%#}jUvcpH60~JzluZ9}IyN(lu7}H3db3`OXhJO^&IHVg#(pdbg8+-8<MCG8
zBSVSjR1&s9s3iD(02For5P8*B31W>C#^~3TAZ315i;l2NxLsdEHY_c+#iix3VOElS
zGn^U>IUVB&r!fkr+;@RfbT1szUlC4!2aZ>O(}Cj#oQ@j3HJt8~p5b(~9y^_D^O7WI
zh~YqG?d0iIm}w=3o2sC)0IF@5+wwOPjIeOc^gVFAg1-$MH~ekX=&k*2pY-f+qxCca
z-3-*XQuV8R&R`<5w&rOS%i1@ug&nPFZOPNA-EbaPl*ggu?V7`95tH*Y2hG)g^7R$M
zz@`lVbImZAbG$|v%sFm_!JyHbhrurCYZ%O_6?w<^aQ`%a-qv}8qIq_6M^-qut)A@i
zs?e>weDCks{a@trdRK}1eDQ{3iU=*2(JHtWZqMxuDOy24v{iI<omf}+TSYP!Q^Kl-
zZeu~%R`YJYD(_o%$ikGdxeRN6`qGuQHVXaX4Hw$lMiHgGB18=wuK-a4#|?-YHF|4^
z+9y3j)M&j;uSA*^f4l-@VTj_=v+kooOtN1IGqj8XLU<8otky0&mwkKC1WOa)W2;p9
z1!r4K&+sKI@MUE;Z!t?V<cZroS?kQjw%u49wdj@Oa-o+(k5b-^dfEGbXRV=}K_t-&
zYrGKW6a+*JI|7|U<~Z*606OJ@0DsLT@WAm3KsRvQ0J>45w+6a>(lgMF*4uFMOG}mt
zbX603|0>v1XkGNyszo}IlLWpTaP(Z3<kf<44$+nmo*v6`CkMW2MOp?_yr%*a9tKeT
z7P+ke87F0&z))|C`U`M>{Q}O~3jKfZRYvG`ba#>S#RL<DQA)Zs5C<#?9q8FCYT|Ei
zZ_<n&OPM{MZOtclenMDsOgfF)bF#@jZMThSI19Ml)h#5D3~F)0Ij?8$2+mHe+ShHX
zm^0zI(<eNiuX6AiI?YO4-FOm~Ip~o~Iy>z$;<MIW-*(<%x5apWo*=C5;olcHV*pgU
zdG@EFEL|Os?RK}i0p2_yLIXg|hl}&IfY~o6!@nkU>GxCiYjkx?0(Sa$%FPLlVoGzE
zEbWvr`0(z}&!2wy_$vsc|0mI4D#fS=%pG6LqmLME>Fn-Hc@7{#a4t@OWu<mz3L5h{
zR$R4XZTL7H46ZzXK-<u@eo3tuX#r!c>zos~PlP-`YpOCNj<uD?`_ebERBec?QxzjU
zB`(xAJz{_El^)a8KQk%z`h7ZlnhqbxoKbm`NN}ZS6DPZ7Q4j{ahTf{&<WSKlfejKp
zPV{<XuW4;m=}_8e)GY{)V5g@#R@IhUI*qvrc`*lW;``Wte!&8Rrr6R73ar$z)Ja<u
zkh~JeuEn<P|HrS??=$v#-`xwJ8&Fyp>+p|`0%Ifb%B2|^d=24<oT<BQnR?7~S7WN2
z^_LMj4fL@!FvFogOr2;&$D5|)huww&5Z2lqh8b&}W)KXCXsS6b06uGg!-lp5{MIC8
zyz3y9o_4^0pYgtf)Y9*@kV>O>kjm24@j6H??=B_PP7}wy3m<oE)m2I=#D|tuG|pYC
z$kM54`ot>4cr4q_xo&GxDmb$I_Bfn9+R@^<_MCl)!t<EEqNvkYk6?8!QtKd8Fe;kl
zZesI_Q$wGfdBv&I?w#dzZQz!?s8K9jrE9sJisB%DbwWUG(|0l$5K$^B+1B3lrM@*H
zAjMoY#U2n_oy@C`S`rljU9U(p1D|JGxw>Z1+9&uPhMgS+Ynn1}v;?T(VtRW3=kp@Y
zx!Vck`l!y4vEA1@g`O5At)^PBF}GZ*4KK}|nYrGYLq*8jf}KNIF4FkpuFflyhZ|`L
zLGs6cR89{=mzFvsw9QF${icjtp6%xk)e7a{p31E##I(@jw4(zWqrrU(k48ybL&K=2
zUY2DcOrNWHbDU#nEpD3~xaDy%M#kJqNA5JzygIxL)>q2gVWM}0*P6GKl*6&I5_oDm
z@W3H2<LA7JXJ7cV#I08IlK1VpaF4T9%Fb0B&0<?&r2?qAb}j$6Twe13ANjW~HtGH!
zH+s_sm%;Kg6qmu|Hxrko12o2e;O*Z({q;c;p>OU!PEq+K%x;UyIe<d^l}bQrAW`ED
zesE2XYeTG~X~fj0^zKEzXYTARQ*kDv*?ToZm64P9wCnNFNTL81e!^E8piz>f<Xw4A
zlFD3U4)EKjnFStZUIh>fB*>X@!kuvXw<e_LXGu|hgIFO{iWY`1n9^E*5XwG+S)P~@
z$w9(BKNYw_TY76rrruUqq|-;@$?@C-RUtZ=0pq+8UBq|7lgASF9a-kJY~jD3TQRSB
zkAxe)jV8H58%y8%o}d%AzDP1@#!3^wsX5N937+;9-oJJJsc>gwrs0Jo428<XO?lI3
zNh7}Tm+e8}YZoW2oKNe2Xf-d|PQRm-Y`^*CL!&V|g$6m&nB<(+%sjs@{1^F_y#mrd
zc<wBcv0A5<g=NHAw8lznfU@6(r=qJxs&+9a`$lt$He1t4vRQbQZ7f>pUx)+Zjx7i(
zBx@<nyOkZwbM8cerf(yum)c6SEhNP>@6H&T-4mx}wyD<BNkdY9#rm=47og&KCVuYa
zWAwI8I-^4K;Kf#rd$F{yRU(cxmUsLnG`RA_whUg_N0TVoB0W8idk?e*Yq_s0Cwzj-
z)4doylzS4sQtmbssG@25i8MQLA0<zT&N50XhS%u@tn>l~&(HhEd+naY4v@uHOPHHC
zF22Dvecjw;v#1b%4n2>{aiH_GIbv^H9=~PH&XlrTZz6^kDB_CN5E7;}svhItKvKSy
zaZBU4kDHunqawJ?=vr_yuN<EW4Pr(1<&EaRoLR|h^}WsKF@M*+B+uYzY`PqSEa6D<
z-Mr@(Epaa;?#4&n5wvJC`i!QZB$H>NprfsehmG1*?jSLLy<*J^v2|zXJw~jhrJhSs
z<+f#koP8i?pV1D6YTj*zEn}TBewlnq-3*(}i}u^l&Z2SI0U8rmqDGTC#<M^sAokLs
z3{ll-9z;93KyuBPUOy2og3c6p>980Isv$H-nOBcn22P33GGC%QOWz9huwuhU$5A+i
z95712mT6>vp%sp#+9>N(xZr(h$<>&w@ouJ+Wv=>_=Sw=5&TSs?-W+A`PT1(HbLh<O
zPKox~MUSaC$i1xCP1E?nkao2lzk|E1?HzO}O8V5~5qvn&c<^zZx*Zi&R?iAd99Jh=
zcV`)`&_j%~yhw7g%a(yY9AgAfVBKr!@K`R}74^q|5(*muqR8z$Z1Vi#35qwMxs-|x
z_+x#4;w=bo5LN+oE+x(cbLvq#Js>+JOe@LAHHhL(>ou6x#%8^)NoCXbZ3Rw{fmm8;
zZ)q1(pjqw$rKJ|Sl!eOHmQq-0&0NNa<Y3q1gx7jo6QBO@{s|UL^?bPKpL*ym|8&^A
z@j+pKYB^##v2QaQWzA~s1N%P278j+Xe=QA*SJtg4AIoD`Y(eT(89B;NY;6t&`NTS%
zBW~7dAp34`yV_*ko;D#pq_wO$1=EnaL+D#T*|J<5Eo<&@a7*o0M>rPTyU`;nE0iBv
zKl7Nq@x2sq1D~aFV||Qv(hBesop+^avl{(>B{e(It!?c)>wQ>M>g}-Z`9V^aX!asI
zrLj!ey-`)$t9zOj)#6txW-TybyL6Cp-|=jDw^;McHGda3OV{doZcW2#bFh}X&C6C9
zHJpdOtSiMj-?<m_S|Lhfv-I~IyQ_AyJ{6rb;GWK%q{qz&mz`t>or<VnWq~1#+f$c+
z%5_N{6*tFjEGT2mxYnr~)0iT;an^UbQfnf*bN@;jtOd2qD;n69a*E|(ksbBh;55CK
zy{;Pt;g|X*Z1iFZMyN1nr5nM^6|0&#cE#QP358@O6(ve3xEkgg8Ct>2XA5(K4^oA>
z(~;*^6DypIMR9@gqO;aI>RX8&K*Usk-gbLYT}&WZ)sD`*?;Oq~z$p}WKpXbMo!D_J
zA`#?$F=H=nr_+}E-(>Y?Ctc{tmB;dklO87mz+Logg?-BtowJ05SJ4(Ny6%DH6`J%!
zSGpz@LG&yKO(7^sMyZ>5YaQ3p3wPY{;$3sCgP+FsdbsYehigT$-C(7TiVie?v>snh
zV1y>rB%osL%ClMY8rgS{A0Qe)&Ud@#Lw40kb_ZQrqFq=lQ^Q4#Ow9>9rvk9qcXQ~O
z&D>(fJ0VdDky2X^^vKPCDNZeeZ4lk=8z**r>Y7Iue){1vQRup`<)oJ_fqZrLc(7{T
z6xgO=S1iVXSMB&~Y*N#4i*JX29-^znOt^KG8l^$p)@H@-X9c@5p8dA8aASMh)0)Se
zYY%-OfqV0ygbp^$76)P9fgWw|QIlO9J8W$DPVKvvNobBEhs&KdwX;Djw$-6Os<<IY
z&cZC^6N1QkS9#}FZ=%dtr7rc2RQymI_NJO+Nfd5_qzY9Ra^M<VLy*;ffz%(}cPv^T
zrOpx(>;l}oonA*c?n;&l5{rL2GR#gVgmp;T+->7ARkX+HYU=#>sy)Ibo0FadCb3~$
zLO-9hB(QX}%Zjj2h#=P!DoNyA%2U@CEbnClDN+<<=S39)csr;L%{@7cDRYg;Hw8_a
zoW#M2Z*G^C86ZVmP3b0o3cBI`NIrX7hVm<Ni>}r|^1SWV<dYks)^Bp%H3zPl!=_?s
zP}&69y3UdN-B2oaM-%Z<gF=?<KkS~am-&9o#$V%{GcHSRVB>&g`z$+2izP?y`u(Ss
zs4WEL^gMn3_VDisOXm+$uAd*Ya!d5nf2TaGjKhBuR;e=tQ*YLP{EmQc{`KXzzkL4H
zD4h161gfW$=$Iu=-{{eyrCJ}Ruk<`CBx^w$)S@vhny|_-NM@)L>%iyf0WhQ(=c5??
znno7|5il=rlbum6h6Kn9h#{sSVQizc-d8zTsj5q4qAMBcDPf_x=@IAitn`>Z@|i(#
z)<33)-=>F;Bvn6u&*<aOXGQCLx~CK(z?M{v0+TMT2+8u0ov}m&11yy#pjPaRN`4z?
zx~v3P_}i$0HuzUvt@6wxf^Qx|T#SK-Drv*`jvNMFaij$Sto{rplq#THP(=+%w@A4C
z{rMZs`+~DRi0^{uF*S?;lkb%DEA11ieBpB^WhZC?mWgbCzT-eXtE|D@CTUy!d|d3~
zBxos2JvOdBUY3M0%6or75@QccNB{$CoJ|$MP9$X(2%4)FP7%htEr(GV>j6MX54g@^
zIXb2pW56)0>qTvN+wOo`9syiT+&Av)>{Qb`B}qK#gzUx>j!f+i(f&P`;Xv?5?n1g4
z5yq}|I`qwd>73Nr%`$41$F=>@W9ckxc4c8s4Rt@RJ9K=vtg0#SmRU6uDQ5N0sRyID
z_b3A2kt^UZX)!!3t$JTX9@8r4cGr6D({phHSwL*wG=nH|u<<}6=NNEQ-a@!kqg%aw
ziwtbNiP!Si(i^Z_MLO87v*2XW2z;c~q?Dlbi|lWI*e7wsQ9CtscqICb`c_v1NzJ~L
z-Jz82)=f0xhVLG>6?r$-hjC>`%;ApEG2IgA7G%a12Cwo=eR6Q4PwOOKGsXV6q~EzB
zy|MoVUFEWscbf%63HVkwiN?U(`Oup<;%i4>zW0CiuV<0NHqYtWjtv=g--@}5+3)wS
z7=gTh@yhoQ{oxe=84cQ1i0^LQj9uK*MwIM!ul6E1WsKdh(jUJ4o=<vUT!DL=sVZ`=
zznVe%wsY-nknZN)cpa6st2L|j&FhOO+gejQ`*T+%@$Vi^?vjo%w(3^j_Ij2T?{?L8
z{d^?r_FbOStd3JRFU9O10w2|L#>@(|`LmvX`?H?=W7l);r_^)C8)Zrrb)leR=<lxR
z#B#eOo%@cG4t}VlGpXzD{l$`wp?^k6H{iNh(hakIQb{K-8(jG@M}J<TMG6{LcOh2^
z=G9^?{^VjUNzg8<J^a5HYv@Zb>Wu527i&T_cZ)SRFV;j<yHKnxTEQZJrYp;+{^ze-
ztz@e*d!`2YeUxc;kb2x2zEXk(tMmd_?iDme-#`59bEluw**>{{u-!4)I^d1+_g!)P
zeT~TVz2tc3GHpL&?k{}rcG%B<*(L8c)a>ox#Gln(sQ>=}2}GEe@gF-A0ysIB!8|h)
zmsb!p#{tZj9}zTDf34F@j`VDeWtz*X=h!ktohB)7SxS?nLru?R$G1r;TkG!4qSdcY
z>sfHl@3pts`n8TI<0@;vJ5OmO^FogE<ky2%ngusD%F{}U5k)a55MAIZs*!b){nKg+
z*=)5`Q3Y38x9_JZpq7aAl%;x&HL8p@_pEJchl?rFXnFbvf99<?bKLwC=55cJvrvrJ
z{W#eZD=T$H7>)Jrot^t8VyfXb7Up`OmYtpR(0;LXF+Y5I8nt1xV@{jp^Y^EJE#TBX
zF3Gb!ai)Sd%YT<t%me^tK$*XodssNLLO2WGLBYS}-%#=0KfiwY)8}6WgJ%DURr#Ws
zG$Q!rdwzuN<*f3{w}0|zm=IFvMX&%4RxideaH_sG!{^V-la`JNWqd3A77GK@UQ5|p
zD#4U@u^M(xipub@yIP22eWQgaP#Yl4rpgeXVlSm_e1!JcDn7;!ea5QL>hH_bm*we~
zqEBO`i&Z?&9R~~lX8IIN3kN}t=oE_-cpd|B40RG<ADWF)I)C3<D4F$EE102&JuKi?
z9j(q8OUS&j2>QYvun51$@(n!<bA^&>o`#tK?G&>n8K)@P%^GgMKYuUnzCx>y<U9BC
zETjT`*yS0zV><Sfki~{mE0uFy67SL;4QV&WeT1VR2B7<@>jH-RWg)`kSv|LD{c>6r
z0%i+a?pPHl)PF<;uwR_>RS6q@5@%oob1))$b{fZ$Gg6oFsl~in;)?t77?Dz|Q0f^h
ziB5`t$R85tD%4ON)TsTu7IzxD-gRxU2}9JOBUPOv)q{)~aOjIruM!<}H}9eEnRK$(
zq978OtfM$LU1BtjI6xe+8Zm_zs{e)tbc>R!)h@PCR)719L?x)~ZGvWSvu)5!I&Xu2
zEPb^SqMk{NIzbB{u0jl`yVsPc+)NHJA&KnWQgt!eT_`4ikewA=)EYTMisi2PiEWkb
zB7xH45Xn}V=A{(_pt)CM(L+F#fI`74xysK+5Ksbc=3y~hgPT{BhZ)qmCgz_&S{kvk
zk;t*+{eKi})80}8Mb{vm7|cAE#LXpL<ZT+AfTQoxapgK`#!+5Fkf1G&<5t_CtI?&#
zz5K>jzV_#f4Y}rYNFSogEy?PO+S+!UDzCA{{`qB)v5aMiqmUlkPy#_$1Es0@YO6tX
zQf>FcOSL+ErII9lG%g9q#si~<c%%$KfQT3auzw=}HXCpL(gKwRqy}_qTxoSHV_pkM
z!kuGr6*@Omd-I-oAZSBxDY$S?16tFDAA0uHt#>XHEfSc$m(}hQL(th?%AVRGcxDw&
zC(O8SZ=1d3F+8!S^&ZcCZTYl~&=|yJO*C=;Oh<|1^I|P`)Y*mxLK`!UD?9%*^=M>g
zKYwgr5N{ly?qS`czlwf5dK49jFOU1@f(fAmg*}VR`llx74B%if4!BLmwD1g^<31S^
zkv0HZaB-?yh;WjI0q6`u1%W(FTPCfPrqLeHgiWZsG^_<36RV{i6>E(f$HD<f5z6or
zkug!}ESQYrZ)IW9GA=bR&YoILOaGuOOn>RfQ-25x(?&62)JYzOwOevYnd&KuM{Az#
zBr_u-%s^?I>>Ie5V8-Z&xLL9X@K@)!8P+Bs7t>J5xz1hwIy1PNrX`^r;qP%Mo(D2P
z@`{|m&l?>j%aMamWGae;lsby<!bOHA>&)k)p+Z^9gmf!CBb*B@8Q4KMLxXi?r+=;m
zHVmeQ&m#u7(GXVa+w)~d^2132b%u)(1DkTOYQp<mjD)BYTnuK<oLC(lNHu1vbIA&-
zD{4!9odi1gK#wf7rP#t<n&!%=Bu7z7j=;b2r@()K-bPR7W_*L_M4D^cIcWxWDrZ{Z
zl2UFui-4*;AsrQE2AY7YTc<xkz<;1_Y3!|)ID&^N*tkRS{NV8&XSijG>y^b^8x4|W
zEp_o6B5LUUCILVjR6qorsR{pL4Uftz;yws!>^*@=Rj5=vSbRN}GtwRx#-1j79$>dU
z2fHQa&G0O#huRI6Q|e#ybolId0r~AsNdjdR18jwn+!J@M@st9$R1RC6;(zc+2nE-E
zmh#f8128-oDhEb*HY{(aat=xxrlnb>Kz$5RX5(lr2d_S#g7wjrmDXi8<*Di9QR9?c
zvjBu+kpDo(Gvfpg3UOhjiC*&5pT|<-mK<@WgKH{t9#*jFrPs5pI%_M4P{B-BW{1m@
zv5~LO)<gHJGw}7}_1t2+M1N;unpXfS!gTWk2W0!FAnOkstd)@~4pUF|YO?(t*{Qco
zkD5a>caP&iu9&rcl&{!BRDArcj^$Je@RlDX6&r>C;LNM$ehBi>o+Ng7GjEbU?oLZw
zhc&t*>;Y#?kFLvx2CSV4SnG3&Kxn&)=Wh9|1f9Wn-(V~4HDN6zrGHru;P3N9R~hr(
zYM;E<AOi>P?ML1J*4qkxbz7mk8-m;+*Be}P)$6tH*fYUPI>`*-E*)RuKb9{E6_{14
zAy3gI{@dx21aP68AII&X6UV%@BV^N3=b3JT-Cl%3gqijX57fSIwJU?Ri0OQq{%@Mi
z7R>v=jwP}}2U3y|+kX#U84aFG@|B8am#EeUSY#2@oUa<#B|sa8-uI6V%C5=!?KIT`
zg-%-@Pgg~2HS?JeGni?fwf7z^kBe+;%5J9pYRLlOB_VslHA3Huwfz40PO~YSH_C%d
zOcvap4l0j15=hUC)njiRSGA8vcPSY3wUU`fuk_+v0dVQ)Nq-hs6;oPA!UJf4UA+>%
zBR9>Y8o}O5hs~9EGFkS<TUjS2-{O>KCOGh~GcOMq!AjA^sZB07-ou;IrGEEzhkj?`
zjtGuPY0gQsVwbDyq;B`Xyw3)<r|rw4b-@mN6v5S&q7;fZevizfjG0F}YH{b9Vj*3+
zNLJCHXxW|~Yk$o>IjkPM8^_ffmwmf?s+w#Fv1=pRfVCPoj&g9-RvRc3E6Zy4SWyQp
zrtTcEZe9r4^!>sG*nre+<@3Ur1~Yq}AtlMqT%}3k<-RE3$n6nu1Vg<@de?2<Dhr05
zTxw68<*T6SCfez{pJzG#G@W#gTr9p9^GQ8&Oj_S2(0@0W)tDA@H^Ce^fvV{|+Ptsr
zbs}7TY5>*&yfu_qCQVvGLNL!j!C;pP?%<R!x_`x563LtCV8tq#oZnB~L3%T$&~pz}
zGsaB&+tf^Lo$SJxqey2u8_hG@y035bh}Dfb1EOx&UR_pgyOEn#Dwt92^su{|lxywp
zlBs1xhkxd~D<H3gV!TohRRtiIg9z)D6dt_!tnN6PZQMEVfpMpQ*ZbcZV47C-;(@Wy
zx3ulzf$>ZROAe@OL4O4tOZz+EwQ;-gjRTM(l;QvH7Pw=<1#O30;M3AS>Md}?gmXXS
z{((?Qh?DmZur@}1`X@d9?0o|=<n%6ho!lGm8-JLG?i;{VfH#Db_YH)QDQ!792vOD%
zC!fHP`T;B#Y)8p)+<Nv~P6ONcnqqWvyFi=aXu@pX){-jT3vuI^UA9r<O?GnM+jd|{
zW9cdfYaP7=XveeY{C+qD+FBQ-0SN;j8D!;Vzm)-3qgSgrS-Z_1Y+zQqw~p(8)9977
zHGf#!6&<hI>7cCb&Nc@PmuQ&l4AIbza5t~nwTGf4&osLsk2{9sxGD_lvp}}MV%cBJ
zMcX{St3fvB=3NbQd{?7|_EUdN%d~1Fr%#7=Wc-Lrv_CB*O3l}j=CpAe^vn;THt(}K
zawT(;YLH0t-ggT?7UF+!-ZLx;Wo~41beG5SGY*r{LlXlrI5U?)uskQ1ryMksf0sG^
z`TpxcW<#a$Cj1Py<<r~y;orm3Y-F=p!SvUUL->Figm!9D#tH9X`uX?t{qWmTY#S8^
z92IE{Q5O}E94efQ@j{1BDJr0XzfvMwj_et+r^rzv_8KXAQY9!hkV9?A3C(I73B<E0
z+&Mn@#+T>olHY}Irk8{$oSzI@e}UQ-L1>Z*!5j*v$RK1mQ5%8Q)ro%A9w%Y;LuH}t
z54!yOZI=Q?g1{FAVflm55F|xenY}Kkgd0#$7poX)wuWFFU8#kUs`^$(=XH9Gs-9Sq
z=FgG=Xd?Uqyn|gBY<1;ru#!}Ldd^D}0;Jn9DMYiZ`YR`kIuE@guY^I}e?h@y9cJhr
zIZjH4HL{T1dk)EzJqW<@`8{=R4<%YDtfD<3PK!(Ag*A0eg1P`Zq!i_14sDndQ56F?
z!xdjoaNGhDp%YHpPwC6w0?^{6qP_(vhSAY{N7qJ&?T%y?hz!*d71^<<eCUlb!nrv$
zb_?4!H46q3zCx*_SjpWQe+i$XcggR~w%h_ziNagxKx*gNWQT!6kAlZmdnj>Y2zfvI
zx(!8|(?Sw{-7=t6LVEhjWys+|0|N9+S<B*ZPR}tAh4S545=jBpwvyCqE-fVqLLuFt
z?;7T8bK<hAY>V0p1r_bFT)SMZZHiKOj?#h@JC>rHP*fov3Vkg4e<n8!khp4(B(S1F
zI`Amv>Jj6gOH!>QYyZYozT7t(Af*Mc{hGPf7tFm#>iUv8SSw+9em6K(yqCJNxv5xd
z;uThMp=nP~v1z&wfm;eJ1wNvtqJrBt1ILCOOKM_P3$gLwQF2k`(!7wiSgO@I=ZP(I
zpldA38`Zd^W|zu%f1Fn|jw)7})Kn7bKDK3^%c##CP(w|{*5^FpT;Jr%p1mvi&fm+|
zB)|7P>zFm4TYqPnU0ik4+u^%_hMP(6DhqWn278<%kTjP@JI0kx3d#ss#z!Baj&r9Z
z<)v8K#5Mw~>RA=f90xjih}<&<f5{XALC)+=m2ULzTtUyJe*l)V2mQ)sa##?`(o#Z-
zS>>Qj{mu#EUTvU%?%D3LM=?)2Zf=+M470VYe_iFg{sXLBGR`9g3VPyY%e^8bWQ(t3
zOU}9aeXMCz9Xm{=xtW6%Io^OtfqXeldiohCdG+@987av8AGSz~$c$UOYW;J=MQ#0q
z8n<(s9m)(*e~L~CkZYXX$N<fE=k{hmH~Q`j=x2`*cxQ8T4WUx5G;Y?D6y;m>B&}A*
zb#Xmby-h{ZVsF;4pz08?k~zm=g_3n7b}=3AM{RB2Z?1DU`tEh^XKys#+iYie?etR0
zSFGJ#>*2QQPR}^qJU#aU_;#wCS9mq_0;pmyU<$i^f0%2uwY>ku)mmlAkx5+I?`QT~
zZC}PbqFA2xTk)S`o;IW-o~`kk=YTmb4r!V`=>~R{H_WO#q1z3<tDn_+G5G--mrcS#
zD-3_?4cJ-Xl_8<b$og?7zW3}dHwAt^clh{0+m_>D&*gAwdpY@P0HjK5nwnTWN@s&T
zP%BKke-&zzIm)4Oo*;+Ku5u_)*JihwpHi2nYsU$y$#B_~!g#4t`VPM|DGcEt6qX9c
zkUMEQ&W^d`7W7Ir==S={WKoi%zVd9QC$(DCwly~&Y|Yzt&>fegTilMNPoIFO_oHyj
zimhukXm~#=HI-SuUC3SgmjUnkxlgD&c9j!1e*(<qNV~qe^c;4*q7^MDnNjPxh(4bf
zV#mIapz=2hNgX9;?1n8PYT4q`<S@87L-L6G-77Pjjajtu(wEv61<A(9I+0O|jI!Cx
zPtDeJ{$84){ANd4wd5CD>|*iu7CVn<8me}4S~b2@U-G88^P^*ove1Fl=KnxlTag*U
zfA_A)Xlt-&ORg3T&HQ9nw(dMS+pJ@jOSW1HZHT4z#<#l%@@oQ_cYyVTVigwGo^RBC
zE}n0srx|jeC2*AC#BdeFoBXk#snLXh`awj8zylBt>iJn%xJcBP<E`5R^fr=dFKq|O
zW){gd@8+p%F3Q{RZ&f&MfJmnHqb6ise^LA4Byd*~jCH<V?JdDsW`4ENTz;y!=k}r3
zkV<RW@s8Ji(^liSe&8c*?KH%b&Jf#h`Wb%f57^9|wXP?H53#w^kWT2$Ra=)?1B!$t
z{eNbSviaQdHfzL|^%0q5Kbc7qn5p2AE8S^`>=JEkPx0K{2z4x6J(2)A1qqufB;hli
z@|;i}@3z{Y(CU#?&?%_Rd2RCat}3>?zr7#+1Kb`z<(KiaG!p?Zmr+dy6PNcRG{gb?
zm%Jo2Cx2bu=N=cFwZcK;LZSpN2@+8amcvb>KJA8-08(kMG%xR|+bZM&E=2tB_pL&Z
z^;;37JTpO2!#c&NNxR1~3mnI5+|S4gsp6kbBki4@^}{gp?yl_+%LqU@F@X#>VUfaE
z7aV@ODE0C?p5ykLUp_n$w~g1rt+_p)nMzdnihtp?Vh{%Y(2Dvd@zxTpBY(M+5gy>8
z@M?tzN_3jNk_{7tQm`Q#g&*36ppja@PcYULUaUrcNm3rvrW9%r)|=H%vYcQ#Ii>2(
zT?P;3+JS%wGsL_y6lu<KsVk=g$>%1&w30&l6|^KZnd>A~z$6#tN4p}~?boeJr+vxV
z1Apftb=vur6wsQ6@s(z@H&ya0><g0w(?%V8lYYPQ)%gt!yIm_qVSo!HX>IhbUe`VS
zYQM^#L_6aVak6p-zjd;fgm+$j%XUVAJ`h!%lrS`79$$<-)nmRfH90o}wl;L<&hVmU
zZx-{06&mfJVTILY3yW^VW<ekW+1|TOQGYWHc3W^!!6TZY34^|$ukea9X*-ZhQDKuu
z8%~C>jO`$4ZWF0t)G)*QTGataCEEGEep(%|1+=fI?4Q3s{k9;%AC^ot9*~K_F8^Ng
zuuum7;hA^`UNw`9NA~pYpI^WH>GQ8j;Ie<Esd524Mis;Iy*xUP%ZygbxAN=}oPSc*
zElNamy=ZW>kjml&TG$3YFHfL7?w+{aM!%)eC?J5=Y_U<o1fM3-27RdEkRY~E+V2}5
zB&_-tnWlzDdP-30+VqI)X;pemAO1{Jan;|Kr!ULX$I4afobyT2gElrfI3EDExT*vw
zZxrWj8;T+t@PW7zJ<7%2D5JLxqJIw!4h&KlgqI3`HPDS_8ew$P5W*q|G(_L?_zoW?
zT5+TasTDS=#RlOlBoS4WyM@F3_vi0*-B(=ofqfS^&q^BsC=ky!Mu9<Tf|Y~WoiMRZ
zr`g<9U$ZV3Ygu+7E5WsE0)X3nTo^iquIEnJ$A_7L@wI)38ME<XOt2`OjDHC`KA&jt
z7EN)FmF)`v8MPEac{3X|4|_(<TQX|QBUz{}f_uALje-lTbN9PtPq-7cale={3meVg
zD#n#5GL$!j8{(<^a7M>Snd|XMT3P1WVQ!&NooJomad5%@T-{E?ltanxq|CE?@9ee;
z7m3=AX@VBp%z}e*%>NfxQhz9zT|S+*GsztB%xI=n8&~m4Csdz8jWjMVszpk-daiz?
zQk#PR%d<M|L~vz>V=<m%j@8kwL~(uEkht<jUPf2ian9=~CYvHKl!y`&Z%5h2py%-`
zj%5rk8yrBooJ<sJnSQ17Ft0LiJ7W&Io_nP{2T!hy%POk{!rp2gjenDF&}xu}u|E1z
zp`4oOs@zz~OSL6L{Tnk>%5}5sNYar`b<>Oz=E$C>!-eG=ZV0u=hFpu3y*=!raEdUI
z8&D{Jr1)B$I<;BO;t&P{pWh{Ktu2jmJ*=qQf*w@zJxx93nyXjcK5|Um1#CH{?qrH%
z`o}U(tHI+liyq0H27iQrq1IMa#400{jV*Ae^S!e3f_i4a^B9&m+^f7`mMZf0b&qPL
zOoTaWN~axdJ#B7@)2wvjpkT`3DiqhHuVuQPpv@UxS(EK{+igEoANH6lQf2*_aqKcs
zHpnukf!V&UX;oINAH8K+WoDTxv=QEp?dCWR*>6QV>suKK1b>fJjBbU)gRVCYY@}sT
z?Wd~{v<R}LWk&JDu;tlzDjjImabyA|&vzDKX6oa-su+}(7hlPMZ5P`%!;13e%!*fj
z(=e`$8?AvV;AWGB_X?$8F$D;5r&#sw7Ug#6=#U?)#jJYRn32BU4@vhfN|@;>2s2bm
zzVbb<Z#-rz41Z=k(QH;}r8y~cVewKv4r%sx#J7VM<hl7Bv(&Acn)lE=DwNGmlwEjj
z%~v~QI(9Ak`B_y%?HuO2VdhGzx`Snw#dbSj{lll)EA!iDVgN-3;=Fr`-9bfuI*nhH
zcdR)x&e@E#H_ob2sHf;%woIklp&Yx~>sBSM!Q<zC<9}ygZEA*FaoDtHNsY`<X6j1r
z$#d0RnpCo=YtK%~D2!4oAF0Azg^g2IcJ1#<OyuX7XzV83m5m|l+ViXBtz!8O>!Gx#
zmVtDSZtcYJy;$mS=W(CB5_&8O{xC}bg*m=o@!Skiz*wv^(+;^-|ETTUiRqOW5ms1G
zoG7A8>VKKOK1@#-JN~v3AY;5J!DAXPr8B)aDLiib``?x?PakKXRsE%qsYqdSr9u`X
zHH`dHyNubfNaw5_EVn{ChJmRGKF6IKSw}zj$?oLySwqAK))_lP{L#)=$@EGN(XzeP
ze;KusXmS=q^Rc&5?X6{Ara*qw>Iz7FrT;R!hkp@XUJn?ZV-Pdpp->mp`^|E=iWa5H
z%qu23--3gdsRrp*IH;XIl1|b!P5xO93!Nv)rIqAP*$8KyM!i7cF*dcyY>OJ((Q9gY
zHseBqp_+rDxxIk2YM?!$J$F?HjIh0&2Q4U{cG}Y3d(oniP8fJYyhmO3&4{Y{23R#h
z@_&eK@3u3A)~nxWdgxRhS9h8osm&2;f!=^p@xgWn@;53ITe|}T8|^fTPHQX#p<GI{
zX<-`}+8u3lXm|9*y^BE`upz?Xf2iG&_!(;($t5L!Tb~1CAyjPzQ}beT^#8Qak+Mi>
ze@dT22M}KQ2m2f}n-uq4`n)X@RYQ3DV1Ms}p|DkA>cqaW_hEmi_n{r0lJgJtK4|uw
zAzp{%zWRK1V%_z11#h?ovmLRps;K8Pb5&o>Hwv~A{fFDzJEesd(@b}Vhzji<sV*`m
z?^YM>{pun!(+}J{(?+97vrLFfpK30G@J@u4bfd(o172yX@9?UuCtj8L$wO7RXMd_j
zm+EF6<rQ3D_flGKtj%PZUman&R>h8bA?5u>O|$#OlD%nXr);QF!8;`dHowQxFv_=O
z%9CW08>ougyLbf(_5E5-9Owi_`t<Roj6FGk7=%VIJ;}kH(jB6XalfBeSvSze>-r^I
z-pZ@B#(1`sE>=y;`G(9<+NjW}M}G(t$RErCRYW%(W}u@ngLAk0)PicI>7hVC2RvLV
z3UlJz3k&)^ORj6F|G_GxmO8~3ys$f$V`?ZkG;Uh(2>kuFTC};uTDy5&id_rMg(HXq
z{56e-by6LaP+HCOwGw+%qeCtk<PyXckjN_TZe71=tUtD@Dp=GHH}`BTKYws2Y|A#q
z997b^T3Iut?;J~+L78-ckS#WjmkKSgicOtOc3A24m8m1O(>`6saWmCFsCP!0q;C_@
z&aS}(cLI*vsE4g-pG@J349#20wXj68wkrAAyJd6$%puH*E=J8^Va%p0S>X{JTT_>G
zf-ZdMKa9HtFKmE$_nm?%Mt}7J^u~2=9w?2WnLl+MW^+dSbm##B=5Uy&&wKxvC~8c8
zqF+HBZ!ogf{GqZ~OgzJiJG<6v2cw`<+O&dlXuek76o!zQUF^V+Se-Vi3Om<T4=hoX
z2LfWP2oXjr$=R9Ril16|LnS@1I0{~044j5azialZ_a(}kwb-qUM}O^BtnOhH?2ZrF
zd66zBP8339ku-FN_pn9OLi<gq{kp4XJ+pH2n)7MA-Ljt_?RVhzPtycpde*0}t<Scr
ztO|saab9UDP&w~cshSb9Yy6C<#w^#ziXby@kghe!>#=rWnhjRXh*8W}9pvDURl8hP
z#pYc9gyJ?|w`<<&j(@IBZlA8mm$e7S(;fa!$gtb`YU}O?7DK3@DK+Xr#{-@$1PmJN
zA3prZ4*#)3SMSTm7}g_rE$G|WNlyBaD7qsgP42ZrQ><4);#N6Y9_iQr4+NylyeA;z
z4()F!<Yv}L*;(Ar&S)*W!)2?UQR*@NO@-XJS+hrX7pkftQ-6M)1<0$nMV=sCs!Gmh
zU>@k8K41nQ_2xjYPWaxr6dtd-qqWxMSCXK!@7lDfDjA|~fG9zFt)<4W^2zlkAoZT!
z+M{$JjE1>Ki7J|!AcsB5Tx;6t7S<+8)WWLpUvJfoXgtuV8yEb9M%{$LVV`cC`;*#q
za|<q{+G4*P)Hw{~+j(3Y!y~|Ad8byHi`0kHwg4SQj+$~1g#ws1F%#DJTG%diU;Iq^
zu0Pn)C!%HgM$P&U>e3#U(aReX0y!|3@M0(xm!@t&D3^&DLJOB+F*J$+vA66oG|d5j
zRoQ*gamXC$=Ku*`66N2IBk`1hGZT(Oz`o@*LHKNELNBIvUSPqT&e5c3D17+k_@p^J
zCN2u2nK5RENXrU~j6M_Tdt&4HZu}%EH@CBat;cPt$kyE~@O3^-Ml8wflzp@j_fT&c
zaX&jJsR0L-5(V32T@N(wY%Bx5k2P+8!4*>o<1XJ}+(H0TwC%{T2N*Y{Cq_*AaIbMo
zDsQZr@%xF^ExBNl?~!R);cWpXt^HBfjfUw-8MchuS6na)`~hsThRPn;X{t1nvYy#g
z$6{@*oPBped#%D(&b4`tNURQK(Ns=3OF=`GzOwI|QO|`_<p}CLLnVbv_$B3kpD+=W
z;##9eOl~Vcf_lA!p(u|Nwn{qJI+wIMn9{sjklngz^;>b$K6@R328Myjie*ztVscAG
z?GZ_snMFbVEU$I7xCsT0h(%8+{hrI(O>u5oE(pm0-J+0^3YU}e6UV`K3$#{v6QGZk
zD5GgiqrZLF76-*DLqq*eg%ibp5YC>6e)nPq8zRXxNJ$RW{R|qqouVXkimwwrgWtcQ
z>J;gW=^H~uBZ6}7khPBnQ5&;+3$s*;8%h6E)lRI9fvjB-MtG~jz9KYYY|-=4j6~ar
zC|HUX=YoywN~U%$l)0yjOW85(a(oI=uHdn4S!rgmF0*c3=RBur8@NP&hmK>(rN6vR
zO-d;gsh;#r%D)y8@`oiznv_jC%69*^gkc7f{z$)!R#X7fOVn0-^*g-#^ULR7KK&aQ
zE&H#Z$`?f{3C=Fx;-faSObS@O#%HRg008ykn+uX$k)d1{kS64XW#H3tP?S%CBo(it
zU;U_~EJF#=EDrru<0s~S4eC(A!9y&gdB3miV8*I$5ojuC_$Ln~t@V$zKCkkR{?k6^
zskG|%<?wkqd`w(Lt+nx1YlTFgsTr%(8doXh10$j3Y#DOQN#A=)TycgpU~9CYw+vD=
zu)dq2gk^>U{VJdf&3UAuI}Z`83?UEY?{R#C4+E_<Qj`>_tfBmW#4=5F3aKb0?nVyR
z-=Dsvbzf-JhvqxNdGZcckwQF4D?)`tutNKMBP~hO^Xw*D57s7A9Tj)Up0TyA0RV0H
zacR(rbUjtVKJI1)07H`6ZDx#tvSEUxM4#KjgzcadXh6z{pyn>_D6w&j<Yij7bdIS+
z91V^)xkwd~dsQ=kpFAg@NtAwEb<_#=jX<Ho{JJjfVkX=Q#Mt3YrKGM+>za^cDj==*
zZaX~{x8IbWx(299Z`k`e?X98{Ph7B#dp=#Kv#iND+llh^J<j2&EBLcS6_p>lNnnmK
zw47>S%cnyjEfgh)h9}ns9a*Ip0<MHX?s5KzV*{nmkyVO+MaHqOo#WURIzbt;5;Z9K
z13-l)l@!XYZ=AEA)CS%%2uV)WyYzv5{Q3GSB!i_OGOUWb+f3V~IYq)4_ZxQRte~>W
zcEWf`!pPZ19PNZ)vk+`2l5;i3h8KJmM(nf{NY$iP6)ZWJ?m*?1xg}`{nG3+a_RwuM
z7xFwE%`)YGUTr7n+~!~<FO=2UH2Ixrp_Kg5WBbzOk0k?(d@b)tL3J{oqDN+|ZA{`f
z=H8lYa!4HOq&O1yDtPL?*b7oqT|T{fgmiQ;m-K6qs>X!1NL8awj8uOt{j`)#J~ip&
zj)Q&4H4~y7?920!>XTy5S^Wd)r)uPZbkq8pP8|Y&gdOGTwaQz7X55$&!M-)>**akN
z9FN~(qbaJfFMw)uE5Es;v2@L`a0MfdMWG|_WN&IgjU$It{uR{poSkeLT{YAK_qi$Q
zpj#svPUn)ULY?qQrX8W48&{6KYe_029A%Ljx4U4S8=+i9y`PM#iFZMoot~_9v2Ne>
z5GtB~(Qwy8+R$PDp;tXp7hf1)`k^;HDBOV>0l)QhAYC8nfo>fRq;t1^@|wX#((XS6
zO2=6PHji%o`?_kRBD#j1$7S)FLJuJN6!D<rFr}Gu`ZgoV0;IseN9DMyUH<mdaiP5w
z;1MVLz9COxYlOJst*W{O8AMu)?~OqE7ij5!M%GlW-^sH|Hupi~dN9vNB)NHNV*QZx
zps91xh^&V#eT!(&?tpF33vBVvBUfZ0fs$g1J|jK*?h#1JKG5K_`t+Fd4e!DDq$gYF
z8z_BSYgGr#pu79!Ey;&Tqh@)p<KS<Y2s_PwfX(DEK|C??_BEUDImHn^OVsxMQ=EK%
zd9KYjd`mZ%ajEp%bJ7A%vhXT-B;H)jJM|t)+-#hd)ZNycuP;P)V7*#sAa4pZ6-%I|
zr5pSw$P*NMExnDt9h^h)YTaI%(xLEOIh-+TRX6t1O4c&**&q|rZFPlmrq%a3_VVrO
zK@a#?rSK;llJq+z8I{`M=79)Vno@3m8%lTRHn;V4$%A8oNiq_vI)JXJf@}+;%6?;y
zH%fxyFfWzI50yl>Q?<qYl0c)kq2wAwmAXiE_cP1L8mCKz5jY*6T23dZw|zyZTa1^n
z%*`Crt}l(6-4Gk9bkn$qBiP~J@S@@|)lX91te<k@%B=FC^Q5?f-8CPN{t}IUKYHlp
zc%uP_R3B@crC)g4Wg4+fn2zgu&zE%(kisjr^>)(=dx+L68=5G?PIBXpu!Jgy=@QX~
zML1-N%9de=tvW2(J1qkBGt9l)D8@Hin`TiPi{0*`*GR@Kj)2>+T8|&1>x9-lD|Mq?
zM46J+19*W=Gx2g&M+Fh?Z{qYS07~1%Gz}Vct-n}WGV3SqQvlfL8Mbe?8NYqsW1q8I
zyCL!NdiXz!7l4<MO)L`wI5{$xF}*<}135Q2mqCI)GJjgzZW}ogea}}An3vfF%qEL0
zUKWD@jy=u=31Y-qU}g<tAj`du61iK@C7b>GJ;iqW5-G7PDVqRE*e#M(ELNR5MRv+$
zl~LJw<sAM?Dq58?k+RlF!zs&*hE2-mC~t!Dw)9DgO=)DsE~DX$;$)woVO1a*?ns3o
z!--WiGJo9RA1^<7R@s+PkQ<XUya<&UJzA7V-eqNd_s*adJZAS6I+-BvpeB16>qXCu
zg~+Sm%FuvT^k^{#Bx|IF>9Uofj?oH5k)Y^WY*a*@h@{9{(HAOGD4MfV6hoQCqm@N0
zOhd_+cbFv=l^L^SDsnMP4kC?>a>mGr7?guF$bU$z9Emw{urPixQy}HoiB?d}MJn&G
zHK`Z~UA*xk1Cx3(A|piQUD>mclxHsevhvY8$cRRHJd0e5-C`LLNimhRMTd<`A_Hb(
z4l*+Yar81@#J0v_zf|^BW=WVOv5du1Ipr+tz%03pNSH+)F_u*@#Rf1_u-Msb419r#
zt$$Kb!IeEqQ3cOsX7qy)X=N26%dgna0RM!HY~eoT*GwvsFEcw8y%FuRSCNDGQihww
zPMT2>b4d$h#ToExL@SneFq6d(*;q_KCW;JnA@NEhhV@ih_?4jxTKF}<Bp87uBm@WE
zVW|o3gLhC3f22iSAfV+sW20Y`NB<-n#D7eN@8VIsp16#~q|Qn|68|mpAt>QTA`Ov<
zy(_<BMDnsW7#3@hNg)W!FwzkojHZuIe*3LDJVn4}#Pz%Ck3at<AssPevlx%ho<3dC
z!J4QOyenV_AMV9AY6t5-+|y;&79Rg_PZ(Z{p>KPt>YyFB)04?ySSu$3RCPR<&wuM_
zqo{lQ{mcCM$$T)cmF->}wapwdjt~XbQYYm|V8viz<-tnW1Ix+-9NuOay?a0aleO0m
z(vA8D1ituFd4Sn%zknlnq$TgV2Y4&CwoLu>X>~kphbQ&?v^qXKs@~UM=GW_g|KCZC
zbzIh1UNf(o`ApC;mMPxVvvx5Z)_=1OFy+;o`eHPA)_yq^@d17iGo4}l6eHns*>vK1
z%L1HlEgN9juxjV;^}619N7^D9AMVM)X&}=YJsq=-yiQNpO?x2iZg&6PhTqwhulDPX
zlRS36Tw2;M7uMgeYQJgPGLh%&{W~prP?k~hVAt@aadx9d8K$ad?ewCa7Jmumta?$s
ztPb#irJX^e;rx`?=_Gm3SPhq!+@c)--&&)SrL!m1$>Mx26TTibpQ`<v>4)=C+f*mj
zKi|ESzrTK%&nL5gRMp4D)kI&^pR3`A!F<}bSJmI3@*VrEshCfskvOq3P2x%9bwEnW
zNh7EHJhd#52I@wj8eRsgg?|X7paIRhRZ6_y#Rp*dPKR4qj+&Q)b45?)tn)H7ujuKm
z$D+d;Jsr#w-8$0AIBi_`5n+4}(Z3bbj}2p{r7&i*Nf>j!6UN*}Z|EDbZ=}AFHwtWv
z_=$jw;H@@7GaUk4o&lf1r^h>M-RS>}K8+@_THQ`BD|xC87sF3~VSk!YbGa&@?g?3~
zwCx@`rJ-a52k?k5A)q$HMPUhFA}pL$`wG*aRL|S@ZFQ)AjUdGVq>Y5CVV8X%Xz}={
zb|q+#)H!GvO8Ee}s04sKIuZR$xN%y@x{XMBpwWaxEll4gqfPW%DE4ac=2(9mSF`!x
zd|X$f295D}P=@_-vVWLN+UdN$*qgQE&-HY+nfA^HE%lWp)=p|cN?J%hIKIv{?jA7N
zu0v_=_4+PquITBZy9w*`bP>*E*O5*RG>O;gX${hk|M94~1yIIiej0fbpiFo8PO#T#
z7^rXb3@G#rDD+?&dN2(=n1&uqL%wcA(>MB=V?T53XO5duDu2&bOU<3UwgtPg28t`6
zo!%s3FGuqai*r4Mv%gx5M}y<R@YCQDcw7B>d3iA%%tqr48J5Wa+X%1)od*x_@Yf*E
z2_T$)jHcCmTG!RpU<UuJ4&XMUX7BC04bcVQEN3yEBA6od_fdNk=H5x<`=1BeYyfwt
z;~mJVkrNMZw0|5!cm#3S(gj4vJ<;*`V%naMfZ!V@mQt5PM`Tt@2z@v$N?j?!qs_Zf
z@!<IN?|*Epp%6qVcAW>2!M2iPz^Ox}NE-Qid-r?^$eOm3>R>!*E;m;bH2@*pokPLo
z<%CaX=!|sjXh&u|A4;M=AGhay<jjWC(PTcWhFhubkbg`<w7u$5`pK!<PL+Q#dj8_%
zYxOVdM*eQ@ye@%Pz}TQ^36W1^w4DBe2BJ>+iIMMNQAzo|V%{f&6K}5AH+*(OvjuD{
zIHbT)qC&1sIZhL0cLYl(_2J^p=CC9?Y4}!0g34SD*tgg5UeLeUfB$;_+2)}`l6zo6
zuUS%#>3>cn$#^*#d%T*HPnQ?%aHc<sjwPt2!uo8h;j{)PZ!Y&HC7IT<O$!OYRFHcN
zBbANs%E2J#JQ)TKTZGJ{Xm#15OlStpu&#a?w$plV2u3&^oiFC?-RePTZ43{bYQz;s
zj)_;DRjlfXbg_I?v7#f}pdE<e!#!POZR|R83V))SwbUfa+QuKu;~z`ib6%qEIX<4c
zm+PAw>YjQwo_aQ(dN!VVHlBMno_jW)dp4eXHlBN)pLqo>f8;@fa9?CRz;iZ$>Ey6&
zpc<S6ZI+H1W_5{zzY8B(alY_K*-{rqP%dY|yQL_0nvlF4fp)X~j&Nr;I^3|R!acx>
zz<=wL0g=Mmla-4{D->`mOucBY#*1dqVe3&ly(%8NF{}zhCeCuRstHbnoC$Wyh5N<q
zwuk>ifNX_-7FTt%jiUWQl?B~nF8AJAu9~yde14a_Z%h3G=)D}Ip;iKhDhnu!o_EXb
zwp0;iTV(vtK^wU-V^k(S?ud>YY_A}yIDemQF`;F-DJfOmR{9~skM{=6Y*g)!C!?m`
zT1Pn!1X5Z#x0IW2Z1#3u*naspM3*Gqj;n*WCvV<vt)x(pEWRkEXGZb|#S1&K<Euq8
zntT<|l6rGm=S1*cuj>R2^-yR0hzz}(n?+>VuhVt;df>ryEuR<MPuKR&Nn7sHb$|J=
z;kv<nBW!dY=AxX3Wfd4%7P|Oow0xxy0+^a{ySj;ZU(e=y7t_(_y7>#CUys{~fc!3Q
zC6J*R$OQhZeUhsy&;~eC@Zg0pd^=-X8{GDM!A7SOLA;H+d?v&mYT%h7*h>Dcc)z`P
zxdZW_V{)u!<@+MwwS36pz1-OIl7DzBpC^Iaq+}CkqR8o%G##Fn?`gir|L=A9F8^QI
zlR+J#No(|!OCBWaHF`RmG=X;hA5Y!4_(UWdS_TH%2pF8}fuVeQ)WJaa5lr{t(RG9R
zhU*)Cdyal{8OR2On&b`Y&U-lnKm4-eS;mGz8Hz@}`LiJTJBYqs3unp)5`PwG#sC5D
zk#gNKU5^HQFp}p)o)W5xV0Nbj+!a+j70#69Qy7Q(B`#YcfT$San38viImfs|OTa~u
zzwM0@4oW0iQX-+4aWDf)kB|t}w%r95LM8~JeEV6j&}P&&vMYqbJ1k=%aWW~oU{F4S
zU?!%tJ40mBY;Qe6?&hpt1|mdn4U=<-`54(FG(x(EhqMMo`Knicf+1%*H(x^i51#3e
zmqC*#69PFfm$BhE6aqCgm+|*4CzmTnG?oFsx57s>WC4H5jbR0K>Ku7UFk`7hM3@=<
z`56&8m_?*GvtD(G6vi}d<Arffq5KuzwHwD2-WA3`?Z>(5X?&N*X&vcI7Gk4BmN*vV
zHQaWD1V93Vvf@M*@%|9e(C^3eCJ+_g<dF$!8H&4rr5O*2z!6$z%;HSx&ynS8UZ?06
zr2=(7iX?yFAdx_svo1JFDQL)~&+7#w<qJ(Ra|x}EY>uNN3xTMSF&-IvyhwG9jOBF!
zk-i8xXUR`&Dq3JA;vl*}I_1A0oNK|4@D4LKBN+11(+~t2M6}6Pz+$)*dIH5qdZAKi
z;sX+A;f)AffjN)T66ekgo}%K%6;<>JRFXnXsuX{+iRPyu&1;%t&l7)$bNNH2B#s0V
z8{tG<2IHAsO)HhpE3e}`ehmh;RCRPFun-zcm>WamzlU~OtX=M^;$=e(h9DV0))v)U
z!H$EqXFj5KlG~ECSGv2VC)YcD8BwFug4ei4KBU8F69KC#PFd;mz66<qq1ICIjXVO;
zeo}w<n6ZpN0gqaIiZFQ!Aj$Wpd<FZFky>4>TA3<3sD?h?QXk}*p^rJVvATIpYzt+b
zrTvl3L;%_tRfrxFtr27-zx>LFf8xuZ21+^>t6Q4-f>R~t;(<!Jf1EzPW=q0Zro6|p
zQ)0nX>H~Etezr*OVNpy<g%!E#saS|&doX`y3<lpD=rN4R{}>F;V2ferFa(Loc!^?l
z=p*aeww_t=JG8xr_+T0HI>B0=Ici=EMj4YkJz2+-Dt=3d7S1<UC{pUsL-rU_5>H_S
zydnP~4CMXFTVEcJGC}NMQO;)2*lh`}SJ`b+SCbZZvEE_(dss&k!_0$CB{emf79D>T
zdcP!veVn$y{}x6iM@K;wOTmOMw7kJ`J4h3*EXcWWWrdm<KTM%sViS&x3eKdeO*zEy
zrzS`4;<p9(Uf4C7DQwL()<e=2ZP-#H5JuWl%FnU7lP1-v`{dAJdM|>|ntA63<qcD%
z$rZ7F`_R?oRl_xv%`SO#2;Yl1gus6chT?FFDXV4!n}}K2cKm-R;a$91w_U2naI?5N
zg5Hh!KqxE`4bxb1RPix&4#AF63KV1QD^pdhtIBs;-K+|w!}@NF$YOFd19}b9)dpH@
zige#z#X?@(M<a`y;b~41UU(PV>b=&}Qu6Kur5rZ9zVWuYtLh`q4F%Ova)5uJW=nLX
z1oN0~ca-ScKeQ+TOq$Z<`-{})y5Y3l4qPf_>b4j@L^XRunCbT1JvE#q<P;G=0dYnz
z#;*!u0PUILV)*l4!<W!-G+=WN#c2muhizOw+EX)<+7Hi>e{PJ{4gY{9r_~_6XU?gO
zTT-og1Vf!dRvwv}Kt^$ekS2ewQC5oRO#cP)n~|#Z!{!AI!1fTkX1>fe*zSl4XZK~3
zA#7yvw9$|pXg6JgW1i74q3c7jh>kV^g0$F+W<!?b-$5Rlk=1?=dDto9b)f2m1I8t6
zRGn~|;WfcOTUpcIVWel>6BgY7qpHKmAf3ZlQ}&p3F&3TLP?+ptBpiR5r+8kC{QQcG
zXpH3;?mMg;gQE%GV^b}ZJ_HyJaYv4JP{%n%D<sdw{v*y#`@0eT(!R_aB$)z=we)2&
zyIl$<`!P)*V>6HNZV{?l$VY^#7UEi{YG)Vi93QGWy4FKgCz>0Fs_o>ZP<5E<a;Vy*
zx*V!H{7a$gDLJwByD5K@LNaaj2I+8SeDWlWD*<XV@`?jXDHwMXXAnBNlqiO`;#QK}
zicie?$Z)MR(wa-p7#O#?JW53B9zxd2dQdNrXajfLTbFGJhjZEbs6w#H;kWucQntpb
zk2|_a1fqE==_rlHMzN%Z;Gw#FM9SVuq4NH?ke{@H_KC%Q9rAxezcy_t-jp<>yfH3e
zY?={x%GoA~E#>iAeCaz?R7r}c@#A~?K2?<VSBsJZ3Pc{6JtJh{QPf`Ts&Mw_Xxi(R
zaxMEWsyIQx`bfbc`4~;pAfyiBQa(<yYG=Rxjfryfchm`E=I`lRPIzDyGZrSYxluyK
zwpd~2*iN4ahs1x}fw~je_ge)UbtjZz9t)Np|61>`zd-0P2!2=;5YDry`FG7gi)?Od
zVDYS4+IA4=W;Q#i2tg3!{bM>95O72yr%)tjcWZknYfeJPtnyJ7sKRv2f$^wBD3NR=
z;PtH=!3C=-uj%0m=*~`+!$Q^RDJ`1=4>QMPY)%v)A4GqK+aJHUE*%Y7<~Dz(u8%N4
zb0lN+<3e*xVzC`+IGY{^k<|twG>3phf52x>ztJUo1p(E_!maJC%^v1h;Rblq#O{)K
z!*zNyhRJ1ZAGVF_FsNbR?__g~_iX>LL>~jthdV01cOTOOvtaveOym?Z$X~IE2}WY3
zus7Hl=pcU|urts>T-h1uW*=xKaQ?o+C6-8_(T?4w!bKW7Oy94#QQ?y5Q^1_?H|mu$
zPc0Vt+A-p#tw2eCInhJXk8d+Jiz8qrQh--W%bpuwb*b0>pkj5%J!MkNf>aH^MQf63
zid!qKLGl)zHGqy%Su27SjkT;)u0s((igMDkQp|rOdIDDN05{mC0q&UB;#NIp+(qxP
zyjCvn?`XZnwqm^@pRdDL4j3yYMfu8sG1-J;@ww`(kIw6JbHH|TY`Z=fn@-GCadf%%
zi#W>haeZGZQN3_DShs)Kv3&GRGWtWXj?*N>HYYEhp+yy7q0#{%?0s-32R;7}SAHnU
z3T2n`v@{Wya8N@P0Wz2IOARNNV^K7of5g|zUqAi%iDyPRn{D_Q-uiE^pPqhy>e)84
zV@jIo^Y^FlKr6wNmy?&w&eI=%pT0bO=!ur8IE-K>{0E=@_&RkhAG6+O4tnF(8q?@a
za8Rb6teu$(kCNW>%((#_7y+L_Lk4uAK?g>Lwno->7cYjC^~{7`PzF^;t;%zvf5LR&
z6aS0sH!Z+^p=br0x$7y4mY>?&GGS%k+t@6b%C8`iNfL%<T4sr5lHO$LLm#cP7DBsw
z3C6-tw~HZR&B1!D@l3&5q8Pu0*H{!~!5l<CBB(KnfQ!TPJ5bU-vCJ=T=TAGPmAk?p
z57wE@4U9KZZCXCQUpG9IshuTke>!tca%sQM(2N4D_@0t^z0cSjv<SP<!iupu&8%BW
zGbf{o#&gaF6h2?l3dw9t+T|4kR|WQV*Q$YDwe$b$RcmLtXVqFt=Jj^TjbqlRh2)pV
zN{;TNlN_Dojn2^rJJCz&@3L9RHPdFXIx99Sxo3{szPHS_86BTZYpYM$e~v}4&HH#X
zGjnF%%cIo{j^_oB$GoR}=F3%&_h5ci*jH*NyGlvM{bE@I;xIa$JJAJP@|KZlyOS-y
zIG3eiM$AZwqX657{;DCPqY#>za+aVr<58D2+1c`*Nf&#vog|;5Q4jv@#8Ed~G0SDa
zEVvcjH7u4}0fj9vm;xU#f6IWuexIP@`ziJY0^q02^q(m<Goj!gkaOVcfdpU+>uq58
z>D|A+e)-eqzX=Az{vEmkkOQwGz{`ElUllMejhnvZ@0NiH%)kQx8)fyRJri(3xH7Ob
ze4d^(tejBBlh$vcb!OUYDaqgnA6@8(6Obn4AyBNXdA@J7v2azUe~3&MHo~_+C65hX
zVLXlsU&D)j$F4ByAJfy9>FGn(s<GBO=;FOpCb(ti0Xq(~D&SOh0Pu-su}NA6AYEwG
z%Iv_}R$8gNwJU?A)CwdFo>9Q3GFy=uTZr7)1hkL>HWk*o_7y#JvO-JUP|UzGD$N||
z1x!?A?IsP^&!4~Ne{o-7)O+%t=sXLl;Eb@-v$2N3x&&G=J}`b5SLN>5n;iJ@$Wz&C
zrCCP1t0Mr@eV7P4Nv`KA*oWQ9fKbnvJ<90aG*N;U%<HOzmEaO($h|547RjBPa|kHi
z)c^xSPOZzQeSrq{4hV9_Zw3)Z^wy;{Q;a;#cq$eOcrMz~e_xOj^o(N~YOAt13>0NE
z$2n2!j>wToROOFMqAqVTiGNOQx9ql`iQh&;_CUc|HA)Z|E?Mwx)R0eEW*zxQtbQyl
z8C`_ciqct%_ae%C)>a?tn#Fa+6P4GaDQY<a0HhtB#`TK&7wY($=h1!84KPARjc%nt
zz*EDCRgOg^e=>#R(sGK;iY%dz$0zoS5T&eZr-k&r^gW^;GvBA)_F8H?c1fel7by-K
z`b?5^`aH=seHKm{>ousYYQYQrLL{I~fn@SlK9#+dPlc9J@UF;vSh)kyx6b-gp0@hZ
zFQfX<A`V&V_@vBSW{cdqYz=AjMl757!Lb>L5&GMHe~!|?-hpxe@nI(Lv2l$B2maem
zg~?-eKtcH8Agfv$QzvG2951#30LyiQV8Mt14;<R%)`)A*StYLYA`1b?;?SAJnbHY8
zqIsy=8qI_%Iz<c5+L%t+{e0<I%5#z>&ixt_j7N)zbDs+r3(l|eg_N};hOA4gCUdyz
z9_mvJe>eaf0S*kISU`jd;)oazELuO70DIvKHN;Jh&x`fjNSSfnh3{imtDR;7@M=dL
zbsRNv$vI`Cf)l=yb=<N*Et%29-TW}OIC3}X$UkTI!E?grvcf@?i-LvKtTKfqk!G;T
z*B7M^HClXtWx%=1X1&yFad;^QeBIscvl8x;f0Zm}qH*ejtUaWYHe}ifH+RW43kJb`
zKjYewDK+%AW6)&}b=MHTP=Uz^o@#@8!6J<UFe*ixU%2xe?b~v&<=OI#ooCB)o%7Dg
zGZ&hc?t<2K9%0E!F)Qa@yASel%JZ1;98ouqlVRp8q$k=*ilM56R<$h4yUREs=667?
zf2Vr=QVW-O;KU6NPDvVqfOU(fdb27|hTB0%X06!AS2A%aF}6(LCA9bh<rG_@@y=xg
zG4wp5GvP`nY>zxjR&Jun1nSk)T^U6{MV_7B1e}-J4B#To(LSks6iU1pTLDLTq=y4c
zlSstZNICb|SZTtmEHhuS?sNiEatLK|e_x7OC8bw&T@W|D-!Z6#?PH&F05I*f<ub0H
z#k$i~9wJ?*nw+^r7!cqvPg+fY4C9CXP`B6<?J9p8@`L@w#zkub4Px(`VNK%VrYMS|
zofC^nt>JzkUh#*-ylN!4DkFs{hQbsOT=|k54!2BPCg(VOLXz9SATw^Xc!@c9e}4^2
z4wL;BRWZ7U0DUlJeNSb9JnmrAJLZ%f8krxQq~}eZp(|Um9K5f#5-(9sG5Q-?JBaLs
zZ$us0vY<Y)W<N?UhrbzOFRTTl>;^oy_IK|xNGl)!alQ1ilRo5t<Ry6vb$p5E90FSc
zx#n2Xa($VUR4|hm>iaRKrL<t&fBAlf5JH7=p76fGW3dS%nfIp2)WrOfrD2T4mg1#s
zg>D4$4>O*w<a6EC?5k@sVD3<%*?I_7dYYP((x=9f)Bw2R$w+VrzZ9trNr)t_nrvW!
ztt5GZ>5SZe1e8f|lrrH+2lq*vHDsMra3;{Ut;3FO+qP}nwr&2gZQJaQZQHh;j-A}z
z`=0Z3URKpwkF#oN))?cnUt}#!Ept(K)gxqE-pV{O*PB4G(n0`?^y}s4(4olxbX<P8
zt-iSIK2?TqJptPWP)=TVT@LOz^jwlG*6uu268Nu{8{YDqd+a<J-TK_V`zv<bSajk%
z9ws}#Y!hnL_}u)8%->bwrOmb2<KM`}i#V1&`jV8E<(10ObN0P@g;P=}^D1_eZ2Y;Q
zEXJ9iGP~Pk>|h6YvXiX7vrU?x-c$);D>MR!?J&$Umm@!9e-D(Pisp8@>l_WyB9?#7
zM>LZ&I?47+GdmI7`AQM5suIdg)(*`s3<O@tJWCQ23-e@TQ|n|HD92VPDKo`!CHo2{
zu!+TW#9-CR#?jXw@ughfBM*yP1VQ;_g0o;$DTJS-p&0;_P9wR>n&)xGmuL~RO^Y(G
z>(<FY^+{^ERZO^K@``Im){FG{q*(H)p6yM!lJP7@TZ~v$o-BUB@J_N_B(!SPeVVoc
z6jY5-X-{%0Eo*}{(f^1KUlYe`nAb=AX!8ND@h@sWHryYMR4Do~F|e|#z_*wXC`b1_
z>C7nA*)IWH7PR?F>!d4djg8$oF2r?E08w+7BWbDa2SP6K%K39^$mXRACN|4SR2mO^
z{04Qd_)R0`%Rlu?L30;CHCEnBse%=;X?=FHPktUC%F=V8x{BM68D1|^%Lcl1G;LY9
zMvIqQhgA(>YMrHw&)R#`U4e2|0PwnAkky}XVl@C&f7)rTlg-WN$?s<QXB4%HFcouy
zA#CrYvCyN|c^fwUiU+CS2&UZhvuSMyH74X*AY11Nxi^Vgm`jpFn>>zVy06&wA+e`y
zOtlm?t(nnOW6C}Oe|z;qB{7M;`#vD3{cA>UomoC2RxYQyYa!nJsi~NT%j4!%2A$`W
zsT%;K^K;5LG+oA=jQEFh*L<y!b!_=F5)26VM$t^?c$&Jz$JO}9AXN)xFA73i<;23(
z5br-dVb!7WIU2J=&kBX=y2@MrSreKFR)4fHIJxGOBGxA(kOnDunN0zg0DM?cZr?b~
zXE6<AD3k!hb>ze3fCa}JdksHTPt7!oy(mDHE6p#&ne;b9>FrxDMsaZUoiJU+3|rkO
zHQdy|*>l=e*X~d8`8rKm`szW{Ior!P0?jFCLQVS0U4;v2imnjroRYzK5ELgxNg`iE
z!VIh;Xo%gav&3EHjDRogTRm&okZ8z+v_IFZzlkeD_f2EQjJ+ut*|aqGqJ}P7WEKFM
zXw|e9xjy(JV58o2cB}3xlcnU0@uA>cHn>RSs#Ogu)!R)?=PO>N(Qb+*8SI1>px)3X
zDKBN!VCA6~<bm^U{Vqk>N6=vsn;k2J$ax;hyBXk=S^(@=dcuSef7ctQ1!M*qN1*sN
zZi2hCp#L!l0mKUYAO|0mb+598BF(MsAM_D$s2SQpxGvu$a`2ae`|m8_(xcTI(<0f=
z9v_-hAqM2mK$*bb1{*_sH~Z3?9pJnlj7nVIn*X|aK$zIm15x?V(oq(G$<r0x>A=!G
z^;r=B=0`DsTRBs7e~B{sX8KV~xsF~`w$`uGmgpuAhXS@PkkdCV(YS*AjDxy2`FO8)
zc&Sa(c;M6<%Hv_rAcTH`35({ZKpcw$__}}R--aZF@gqD65v3Hxm)^$Zz5mwOvjm%i
zEHD6q_lb5)AE8Xo{|Djb8f^;mE%9(!D+VG03|@VFdU5)Cz8g_67n5j=5SKquYcyfM
z_XU=5$T?&^&o`CF`DO~TMMgChKR@PtVN(rp8NZSV*p+27Bip^es9%`q<2^+5fzCbj
z#P!MkahqklA*$Jp7fSYr+8TWs<<=+vcGWUCI^B#bCgwr4LzO?sa?b5S$z5k>?us4*
z#CO~94zb}sP$uR(jany?TQyOqZ*Yx+NxXBGqJ2bP@s@wT#(ggKCjqRDI4wE6L9*l(
z`e{s4yI#s=^c`Z3f;_3J-MMA`Q9f(eY39auG5FM}4;ER9Sy!b`(Fj8FVtPWw1*PmG
zS+~z)#K%-kQ9QHxrA-lLn14lwoxwH$4qqbCZ5>1-&gB~8RJ0h@3oKz_H72^lM#y<`
z;~NPP?B}<L{yDPY9Pr0|X4?plDeeDs6kH^Ocg#=~2R<pkx<Q#TA2#7g!1n^Oj&<%S
z<h}Cvdau4sNe6h)j|pigZ&^RW!I{BQaZG6W&_G@d6SuPk`29X_-Z1u9_iq^iB8;Ro
zfgV?6z0@~CUWWM(o%;)tl~|4n!@yYF@ZyUB%zu#c*5ON!3&t*D-?<d|B*K;5lhGWH
zZ80_3&m=>`e;XH{eDS{wO{i@XGwZEGIvAz$a&#?xGShU0Dg5Y%E!6C;-H>c{*S|kT
zea|my8l^2k;r3IXscB0tB#a{hepd|YFuy`sGg#tG#IjUKvX0gQy{mgv51>m^I1<fk
zn}rHj)xbTXKx3N9QA_ztQN_iSCh!ynYY3~`DBFE}e<q#!6=~geSZG@iB*9q^K5uwV
zU{KefDa}XjO$|<-A?Lh6l-_ES#Tu<63PM&jxIq2zkA;avhQ8F8FpYNr90qy}Y6(Fc
zwCk1TG+Vfb6`L$qQC+yLB(Uz!$}nSq1T8{i<|JS7@o{11kS`n5@&u*^rDqFhfsqnb
ze88ftc0WEl<%g*S)l(SFH=C@(O#X>*U?`_-R+kSV4n;v0Wio*>&L)|U+?@B%RxZ6H
zQCWan5*LL~${55Zl{x;U#Aqo8@>~&J(zf5VT>|C!g#fF}H6ATZ`=#Y{a9KKE?5?|N
z)(L#n0o7V+N5~^qROdo4DxtQ+b#`D$E4U<AY2WcldU+HA1?xnT%<f-y-Nq>J9J1V;
zWba#3ZpC*?;hl3wlIMy#1AF?Q?=B$o3I^84uaxmL$4#2sUUtR-PBH+iYQd|7N=<#W
z5A{Twrsc5>!=^6(a8t^x6RQah4U4ipTFOneEOvifa$}^C)cw$Ya;{uUV^%kj6M_L~
z2Dw)$(c4Q3e_VxD|5-=IDpS3|waR$pxZ5!`o-eivYslBk^+m$Q<?EM*=^vvOyy_zJ
z8@r442-)7})`2tuDm=Mvc*R2YaFPbx;v&}(f5n&Uq7s@2yPM$akje2(O69pn*xR3d
z-Gbv(Jzo*rVuEZ)>|C)E(Vv&%9qe!`O-2l}t-{+RKLIVV+AhV%avS2ISu&9{n<LB}
zJS}kvldGT;Xm-Y}$H+zY19p}uQs!j66>=#%E+md-*%fDi%EA@N<~8Gz6DYcq7~DA?
z?c78Ut1DzOVguU7Nm{<#)Xka<G&P$q<|58i11be4I-?^k^DDOM{WrcbR3^S>Ikah=
zp$(?qli-h`sRoeLMBO6%OsT&88q5i0p(0ndD5qU{1PToNj?17sVgpLH+MAK(%1o|N
z70E>LS@UtgXk9P+=4-)`+{(^GIGgIXMdQG5+Q>IG`CXF96DoqXBJiCI?-g5xv4vk+
zF*aqP5WC61S?tp96|FMDrRu&V)BX*3aX6KN5*K6GU0(vrxd0I^uZY0VXM3!Lw97io
zqO<JywA4;_wp;H)D7bPKdVLu*fMpJtl)Za%39J|3SXOQGp#2Y>#e|e5bT*ZmoW<jp
zoyCS*y1KSx9p#h!+psT1RtnEnU#z`)gGZ*1dYL!BQ~%X=W_V>e)f~aN*u+-t0^j{`
zacSKVEk50Hk^8r0<zO{UTv|!0?ivwswfMXKE$0ikl0wi1o?B_g)lviAuLIR+jT{X-
z9*6}%BYr_$qKrtb>!|LRE<l2xhm5%TOG}MjjB^(reAx+to$~V(A?Qn5;9Jnhs=M@r
z5Z2EqDS;R{3QF*`eEY*CEHQ7Q(E^zcyxn2>O;KT3{=v61C9sdI^gR%!i{a~VS<gU5
zI*kVI5dI1O=NFGBG|j*X3=eH_**ey6L$ea#ZZ6sO_cQuUu|*D$JtDR-a)e?SIExU*
zXj(&xEK$Q6P`&d<#@uoPnLG$rNnNhuabFHGX~l&RbE({4@*x9iq?GCIr|T#68yb-Z
zaV9YS$EH+_Oa%;=usqeEm<0y4UdVLj4z)&+l;{a8%y;&Q!j)B}bSe>W#CP|N^|B~{
z(Mu+^ZAcmmw{~PGyvL%+fez&;r99EEJYl<JYwI>aDdX%VCfiqn`Cnf8vI~2t?Xo5a
z&cJA!$GGYpfd$c8*focH$7QtL-|>(_*=x<{RG<ADoevsVz;sU-HfYZ2SGGQLPsej1
zOR=yCze5}?ngjN4iI%ZVi@@OA&I48emkwdFxNH#Kh6ibVL&Op(gTFY2$V~MIPW_F7
zFAb#{nq0SVx`!89g~<k#`pZ+G{@tI<)a{PP91@h3-MWwg(~zOcpks;i4yqx|Cp(EK
zkaALx?^9JWG$ozM)u|geyvIe<VcWg2)G4J&FtmMcKgZuHS$*!XTHnY<#m;5qpSG*h
z${E};M3LD=u6}CFV`?6lz4!VpsN}EOp_MW^KOi>V>81a9!u$sYz0b;pnm&q5Ns(@2
z!I}Yl*$QUK`UC;`${v~Czs`yZ!pM@&qz*>{&&0^c`M*eXmVdwuIw}gla<-m>ZYxsQ
z?H7$_rv4*sILjwB2&lYrxzOUkZ{^3qzoCb;$mMaSW0wP8FMcOAZArC_lC}hNikN9_
zox1mYgP9&XM3wk#p_tyXY?=oi=R(%SN^B+O-%{{p+c;#VJJ(D$6@R=3g<0;z;k&(0
zb%P{nUo4jit-kf`g_>pn+nBtgSBpc0X5USaC@WqUs3G|~#y4$0OTtNB3rQjiU*%+^
zh)fcdM8|KPu=&=H{#~+ar~a{N0@0%U*Ty^4LMHbR_SmXXFNaiWhoAcJ))&RB@5cwT
z^VV}6KTY6v7mdyMO&)+9{9q|}9oK$YEBjKzm8~$nV6?l_g2rXQ75iC6BU`J`Xr2<a
z%OiHfm4F=)MM0pdh`FT$RxNLDy4BNLH>R_?If`vUw2tI&p*pv`+HmA;^?#eqvQvU4
zG$`_u4{o?f2Q)e5i&9o~;(HNU+C?U6N%8%A@dSo`=c16@K&U@LZMinvbGC{x{9Zp(
zKn<#SCg1q?#>p+fcn$qD5v>==fMgz5d*HOA*B1Rx9p=~lXXU&@{KVO+u`ztHfkyt|
z*4pqL77CdG$z7qQZ0j$9rdQjZv6G4<g*dP7;rJqZP=l@wJSzW<&Ucr#eDN2)o|t0s
z^TyJ<rHYIAgF5V5PUtd2y%XTqAVdzAaM{LbyeyBi?R+GFf{EBxC>VmF)Yf2$*VGy;
zfrSZ`xh^N7&Hfs=hY8dP{40*l9z8WDZnWabsqt@8KYM0PX;;bzRI25-Z7$5a=2?MV
z-=`2od$6y=X@T!I1_@p+#U9q<lKs!=$#jL=-ExaHU$49SgZr!H#?#G$@6g{~kG8Fw
z`g!3^fVWXVYiIqe`>gvW;Rgr4e$H*iVPOZAr*8gnVepKT!13V8Z@CwS9PBNCr_;~(
z$^7x~L*w<$_{qEm0>`f1?#(kN`|j7va#|)}3*^kVm2bEEnKS+YF}}5)VMmvLw`T{U
ztM@7W8i66t&*|<t@`du%@A{l-xA!RfF^R7V;`9`N@e$}M@YPMhvifE+rQ7OYiU)@i
zJ5NFIzBF%HJYSDZ78e#HjvdS7j75_E-@)ss9D-!3fWyL}XZNkz!64^{$F`@f?bFDI
zOTgB*<8ARIdpcdg^ylF2`RyjsGJmT-I+cojO=+^^hLB4^@A~{i$ndpsdKF_=&$b5|
zkdGuc6q}U{k`sQ}^4j_F;y^nrQAo)wMj2O-jSLHQssQDJp`y67Nf~uC<h`^at22XV
z{^5$s`$EbIsLS02*?fugMRbl=LFPk?+A1HaB2Zl7W%^V?jh!k?lWNx1o(Nb#6J#Zi
zbp|q`VtG@7{WUH`1LjtXmDjSsj|UnF@V<yE?Bg@F4h_!XX89HP58Eu+&eV(Uo3x~f
zt=j_vsSY}y&?pQ3&0`3p67_{1b|6(&#FoT1CZ;ma20NdDi$JJ8KDbH<lzJ~5gDyuK
zk~rr>Gi;@=?iiWz?4Nm|x<3sHttd9$s${89QeY01HdeB=VF#}B$eKusD5^3I;E)BT
zY$oft7@isnoa@@h<Er0B9VQnsJgPwHIK&38Semdy<-*FU0tNCHA~V$+X0yAXC+?J;
zwqFBL-8e>%+PKpmDv-Wan<vm#2TBYf1P2qHe{ZJ3O*s#uiej@lqhKcU-+uGg9z^|%
z>@_b<W{_foT43CwL+LG^gbf`HpcFDonqPICRwZ9(@j4k2L%{3<!=Y;<&&w=?I-{n6
z2EppUhin`-n*atYm~e)+nG+>S)+wZxgnay0*37Ettu%7xbae{H5c)qftz3)UK#AKr
zrQZF{&Lq(txO88H5|yabR%hCwJ-hn_8faOH?J(Wm|9np68V;1}2usxf<i!u~<>)qq
z)oT3#{6}jX>;^cv^{_Bo5s{Wc`Ax-gyNX5rB`bo9CIq*Q(J{Z6t^XOi{W@fuO-b<(
z=-cX)=`lo;+!zWdsB1&_CS2?1D>^DCl;*u+Q7z}9XD^Dd%#8m`L9HW3&8-Kq2Hzx_
zVWC);rq>i8Y@-6|AP@*6s)We^uwbYOv78hgOwdw}a06-w(ab=4XNfmN`V<WHa>LCV
zl`F+hm6L4H9CD+;@x_E+q(9*h*)}EMu#=0(8$u({hF+&Ec24(SkW2K7$pL{Xm6<o)
z5182y;rscw`noy6rw;MGcR%@GO`F}`U*XdyqZpm}JH7w`zTQ`}hlgwf{jcxvYXG|e
z;Nxs|ah3tt^Ogj$g&Bzq)Nq8qq$3r$r@ia-dssq*bWs+owbU}xmq!GpeZb&QN^;nZ
zk^l90d9v64<7E<K?o)z9E!xG(S1{^bWfxr+<$OKeLITL&l4^QEg@*J+>1TSurm*fs
z+r2XBn+`y9*KY(RmvqgflKaB9Uwz@zqQY=4x+F*}3K>BB<}XgfluYUSAp=9CBp)(R
zo2Lp-sXRzta&<_#VgLt?7M1GTw>QysI787QO0IS?_X4F-Ec@V2UM;%rPn}Olw{*d?
z##t7&q=mcNWv>zg<6fZ2I^;uDv9Q2OtmJT^ECAqWOUu2nGZV_`*|cGhZ+w+r+MQFO
z*55q-G22wDd)9QASo^X+4xd<~US)1EfKnEyYqKtypPT<ZKUS<HUK}Tm6U$^6Rm`+A
ze-{_m?mAU!LXYQio*^@;%l^K0W>OrF_YTY35=1{U|1zLVxeQtrE?Xxpt3USH{gRA0
zMFgnPML}K~R}%*gE~UDRWQrh;;yJIML9;CUHj6_+iD63t0jB3e?op-54Ox4)x(Eh^
zw2W&nJ*9esr)T9D-Kxb4*nzX3446U`W+<(J3<4Vq4g@KU8`OCw9+p=r!4;%bUepc5
zWab2=me0ayQk-gjOtCJhn;zc#>>`F1iUgQta`CA|_{;~9cf<-eI3TVhC0vt%jM(y_
zZn#;aUM7qjky<d(S@a^ST!=zI6{m)1oJHC1^9T>zB9(3UYP_hbk)%S%6R1Ya<V48(
zF3n<;$hB|jBa(%b8UrVjCde$Yu{NNvVL_n}QJj8&CA+(Nt23A>xUHyWu$|hV)&NAv
zvG7HlI4tCHLHY^4phGZVFVdOg0@|_%$(eW&5i8~MF3@HK8xIQSP6o=Pr%D*c8Z?Gr
z-sjZUWy1e-?phI5c8b8mQodv@42)RC(hVR^V}J&BgvpTZ9!6QwuPS#KLBgua=o#8i
zOo@L-AXQN|%vkry_xMpQQIB7S0t1M-6-q*j1ysGQYGvp%Y0>TAdSm?H;f?c)=<ozR
z)fp&w{@(LRBoxe6YHy*6QMfEB?j_49h@m2~56`hdv*sDNaDg1ed4gQ;g<hil+6{A=
z;Qe)A?-?{PRn2RiP)XK{a{>|XA2>pu9C}7b`l+Znug|7to$&<!O)6eB<PL}hmQ8cu
zF<;h)!)aP+P$l)!JuHy=8i)WdsdMPZXAH0c23}zPa8`sTW<lzH%2oCTBp<&b%}&D?
zvx^rbj9lt1jnJeHo=v|@?z_6pJs2H3Vuwqo3^L!KjJGL}9`Yx~e2D&^2`71ylPu}=
z-+4-a_9#<gj0rK;s3>z>q$xo5@ezfgSMW1sv2-bNEb~S8E=>3?>0#V{k1^YThA-f6
zuB`6|W)zPRPqC)c4guk(=&A>OTwohXi%cUil~#naB|#)QO+6nJ-OX_eaiU{osB-zc
zny}8TYp>uVUOayRkXoGEd54%x7B?BH_ckTK#9J%=P3Hb<0H<Z{*ripKMlIUanyU)4
z_wx5aE<)18khejwU`vrO20gdy)590`FoXE9+l{a8wgh!s1;GQpV1Z6FU7C}=uz7Hm
zv2adl*MwcSJbpsJ^{I}}Qlz37RJ)=Hy)Fx4=}tp?UHwun9#XThp4^q(a=gxlZ<H7S
zN+Ez2uN86iFZg8P13<Qc7_vA1h<W`P4XugQY_Sxt?F=@M&HMRd^(>gpqxW;64A<6Z
zyB<Jm$0V5EbA6pn|68iDg*?jGB?I3*Pw{stgI*d+FLAC*tfRQ|%p4jAx1;!w{NE~S
zq&hWn)H%D%wZHlX;o^=S(3m<X!Q;LFHX5$T4?ZUU)zVI9Em|*OJO2}+aR<e^hyjME
zHv+u`*)Pdp)zAR_DCH#u(pY!N<RbI|%~RN^AI`xCt!I=Eas6ajrEJQ(M^c@oNC<TX
zgDfTfU&z?7Mh{Hq2fv<_z|M?*bV3aTe7sv%m)DbL<%wP{`QY?xvj8HqC0OB5(#5Ag
zK-!=0S^xRW5T?tGutR_{v2mvFvxAYRWBL7y2Bc_*R~q<UO`qk-g`<0kXXxzp7#1Z}
zw!JpwqpCJ~z5ckltoMG^;}$^BZr=+7EX>`4de-($uJ(o*xC=PE#mZM3dj5t_HLvo2
zz|Yx=olJl2I{=;Be*F{sW!Wfv$QK1pn@%Azi15dM{rxyubKhth8l*f1rW@@OwOQ28
z4&Yihr^&`h1<|r<c=g4)Mi{rB6^6<*fe|+^tJEC3@?@)1FWQH;-ZSzV67!Tp=yM!Q
zQ1fR3)qzs7%QFlDpQ(S=1Q*J=H}7v^H&3WwP(f2`Zr&k2C4E8)t`_57z{<utQGL@c
zOopTo)8xIOdf0Rhx-u~lMH?*={$SF^Ho$J*l5K~C@bZ;disg{)HLJ0baFGTjr<9Go
zS(`hyR-Sdg!aeuU6h-uhbm%Ix(Ik=(8TTbJW?fEE)+-X2>HRB%4*p3a$72jJ4%IP&
zFR|-%T}Ihho#kv%(m*`RWF&y$-x1A6|6ncV#5TD!%e?Vm?7!eQwi8k$VF4}A2S_%b
zQaxyx2^sRbJvt|2N6UY!oXBvEgVHv@qumNYd<UZw0#FYkYng$E)CRD6b^b-Y6((9L
zF$cz_cs}n4o5H$}vc!t^#zxzA{6mH=+~YVr%K!S?J{IjJ^4kKQE6xl1cy>O)jx)QR
zd4V=<Iihi^Ga;uW4ej8BUxKd^4Pan23|sDHKO>*@l2hwxeYY}`%1fNmd~IVJ)x!)u
zn#v#DY-?;=QtJ7!^kr>XPa2$)>=z%~gu#;`n5YV#aE+*i>Ho>(D^uEpq~igheL{&M
zaHK;`Ehj@&BLs8S+&seC^h?>pbkb~$%0^erZ%ivNrp4W@wV9-LwA2(H8sJH!rJT-Q
z<T>@{z|!JP@`Qt~Q4UU1QK(a&xanE8xR%yL*Ylb6Sb=~!&lz5&EWf@5hK;7MDpZev
zT-3FN$Z*P-AS+MWtp|{CX3#spi;}{!5AqQ5BOHj3rDVOB#7=Q3I3U@(!r<%dHNQAf
zW`f6g7MoO^Zpyy{hQClD30Mr>Bcl5dXbjkB$D=@JrUKUiLVsgGF5o4jJG=WQ9OXr7
z=mxPak_prjO}5op5Qj!BmNuPjoRE?hGGe2m`&C;oy84-EwVxSt5+{v3(g0sP`O9nN
z-YnQ&_A%bOc4K7BMBJDPiVGuqo%Z292z~PJ+F;Xy42<kH>ZI_jAt0q#QEF<HuD?0U
zdmLBND+B@&K=B4Yv@69c^4V<>Px8G}*EqBre)4Asu+!y(2)JFmjO<Nzr8ww?<qdu9
zoKNRB5$FO!<?h9Grj74jq5lK~gY@B>`9wSD0&&;sfbTV&R!G0RaZTacSXnZC1E?{t
zh|LyApu^^T{yrmt1Bz39<Adg<c{LchXS0Il`Nc3oR1~qx=`OCxN=J%PyZ-dh+-zbY
zR8cLvJSIv06l8&z9z$Vz)74i{8A>MFLhwj&tfah}&KP1SaRxGsPIQEFv@WGs3WTKy
zjhW(4%F?q{D}t8`uHqfVnn@B?`JRQ^7qOle<~lV%yUC;801_`?pCIKFM0Mj7CyCCR
zM(YEql{$31iVac0JP+?zthx=nXe63-O6kOrA6a?)DLVCD_v~N@ywoK<TmIJOmge$&
zQ0UJoiRI=Ju;=lU)=0E%-@W6-(LEGCex?pq4o8qqQkJap%%sW<2xULbhcH8T&VMD|
z@uENNHF^8w00=B8ww6V<T@?1cmOOv!7daC(JZvE|aP;VMp<Ut^Hkw?nEB$TidGT5~
zv4KhoXT)iX-@y#dL{s5s@M6+njv%6(-5-E$DRvkmbcv0zi|m}-0OacIc(XB`tlAx|
z#=hx^p22-?YPX)57Dt^d)NmBnPjc5cr)jW9#!oyaQP)lS4MNmMD6srIKW%s4V`;TS
z3X2KFp{~W?<naE(27}FfE8H>IeG7yB|9%u}djB3bO8N{nHf4HNFe^Mj4Benla7^+N
zi;x4y1;$wu^N-7KjLvM4Z~v-SYxYi`C2XssE9?ZHA@b;_6Nv`W@cLlfJde@zZ3P8v
z2#PM?^00IU&^Z{XpWqU4n_m!wHDUY5AAjQ^s8UO36Zt7iS00@nTg%J2uGl;;fR{ke
z;QEYi5%HA?t{+sK46t1Q(yU7;ey!nr(g{%#0t?}1p!nZ!_b<6qG5((GPcene!c=U&
z-#s6G1#z<?SKZ0N5DsBK@LRIM6^+etdDy@Me<9r7&RgpZzMtPP5Lx#Gc!l99Qv&zG
zZ_lflDPy4Z;dZ5i(RRFkgGLDAk~Dh?!Qo%5{W*U4EpiNIdANB3aM;2H^Cs=dzQYb7
zKj_IUwWFzFT;9+n|5{;3-x<*=vQ0(%9#Os5{FXkPr2L-9fA%?-F8=1b>VfTySF0CV
z6!6xjxK7d*BIJatwU$X<P}gkep`wP8$8KENisF7oThUn}#jQ--5<_t7p|-cO5~Y_D
ztyZLzOKq?769dl$G)Ym7i^MVpo)r*5=i#9?m?yiB{v#U9g7z-_K0V2VycV(w9D(;)
zV6qC68ctyIXEJrU-QxPE&EmBbQ7|1=KEf*KRd{>DEUuaAM72}N8tNM@+-H#La1i}!
z(-JV10>^1mY0I(!Z$`3xD@se1RTmYM;ObNB&VeuBI>wX+NR1FYul)KJKZl7gP5mwd
z3ys_1mN%<udL?fe3$KxuBFO!Cf8^sQdX9;V$r<yBlZd0Kk3C7yviy3K4N_FrL=-(I
zOnK-g%t<0qNND4r$I61!hn=>=Za@*eAwPGMb7K|doFLwg!Mu+-F>kOkqr$n!OdL3C
zqm@Y+kI7O83@7TXygFs;3c;`>TNmTM;W5|;<mm-z^(o>VEpaY&<CS=G4~N2Ez{7*|
zLyJ{gbC?_w`qcYDF{caDNaOE@z%0|}!+#k1m7gZcAK_2rBnHXsXSY}2E-7U$T4t8Z
zJ~$;2W&imcsbFr=o{yyhB`Hs4Y6ct~r>IN~B|6*!fIpXR=<y3Io7KKzZvW**mxzfQ
z2`uuILYf87+0(C@NfD3Z(<+R_pWfiqHucV~Xp15u2GX^n7!{Ism!fIvV1B=@{xy5~
zQ=U~?XqLPxkAk6?>B?*MmLdd5AZ=<&OIRB6s}U@jl63I~M=?2Z4b`%w1%}M!F1|ZS
zp@S|5Sfo%<jL-#go2d~!XnjXtDQNS@G6LnGNdB%VhE(ylc0d@wT7HQmf286Xjh`T_
zB+WTI_9fDquj+pl)IrdDd(RY@?7|Y>f`3P<Ngqqb!AN}lY%iJ1K-`_me+YTFL;fx!
zsr#e%9F`Z3wyAY-EVBDyCw0x8t3?4aJlZh>fFuh?c9|fpN>0nzQ!lDN^Wuha$Ubx~
z$4I)@ZZeCNqsl~Zh+w8e6}r;DFL!eoPh6343T*}301nQ)`+Yzc$R1L6+qJB9QVm{x
zP5qn-Gmx}tZ(`sj7oC`2d3&ICBNtb<KEDjUvU1f(y*0Oy<s1xWUEIOy!#y*dyUfA@
zaQm*ihQus}ZnQS>33FA)$DDtlPGDPSNFLFIX3-%}6<d3-`D+bL8TqR=p{giu%-?+S
zwIO~t4O$w=0~N-cjpQVC^zsu|5R)(4YI?RGlE2}?yLWi@^ba0fs~Mf5Sf;t|hZfwu
zD~=1gMB#N}Q_n$VP^oUKdrrQUxK<$$0BZ-h(k;ayV2L}*B0+m~lQq!`jvAD%PZz|S
z9CQaKS!zQTye*{rJVM8?AugOH!*=5X!jKu3Fhw2XtOmykH1~<OtFE!`B|d~)k_Wz&
z<9?O{DXBoMzL!S`1zNP)V#5b6oAO@nk)x?~g%YGs+|LRROp~Ya<@)zHo6(^afaidI
z{o9Sib<&@YPG?wp?3BucxSL2__%M|HIZU&%{MBExGW#4>{lfc)?GN&OrLUvKpM`_F
z)pu}p7mX&@U^Rx~g|FrM!6uC!mmQAunPcJ^6yVWv_J>Xv<ltAA<cM*L>5-Q`+B}@b
zh*SF`wrh%5ob)Z89R^IX*p|~iKn}1NGDjZi;)vcG){}2oU4JTQXVGAS{JqMykWWj#
z&XFg1{2R3c-6)l~S71w{^qTdbV%FTnZNYGDC<(d6O#8CI1Cpb1dtPZ(Gb*nM{1$f~
zK_P7O;@RMx)vp!m9`2indon5Z%BlT?1ts>KFq=lTL5KFGaXuy<&BlabKy!PD#mQyN
zCZ)5U!vNlvIX8-3&I6qryQ5Z0wQf&YD!)XJyB@yT0V9|nLMA!Oa6=I^_0h+eIT;ZP
z0Ac;#!b17j&tVp?P2Yu`b5YFeH@vgFxwP~9oj;j)q0OE|E2kcHI!NFR{d4vd>g>TJ
ze{Y!(rrjF-DbA-`VGSI>SS!P?yC&8*irvJw)(|+Sc35*WYEu#F`cGO5(XCPBmL{LM
zRq=2iHv-Krg_>Qdlk%dvM4Xv*sOMrz`Huj#1xbLlmyL)&5z4;d`EIG2U0Um?zFVku
zbS+>Zi(L+f)WTkQVI&mvUk4(qWuX(iTT%VJR?K!fMaCyUsqGpdW~+-w^QER{duJ@L
zH$vUA;8Pn#XHu0DY`nv#Rsx0v`JsxM{GlFjeXBQXE-H8?b|4bc6u7Hi(JzSo86&L{
z)b!h}VzYh_zp6)icDf=#ykH<`aD?S~xr1dea_XsTVsuqP%b9rt%cEGfd2h>HtLo(s
z4oGNXcKd;+aPK(4o@snulrMD?9gb*K)?4$yt^39V=VxZ#-0H+*8)-%v9sAIa3MLlI
z?3_{<-e;8U;D!%4sI(0>@Yx!RNORUL$w)T=4+pqH1WOZ#9J8=RUT}c3jsDRssI2nu
z-}GQ8$;obWSuSv}$y$JXhM{f^R2P(N0Z>{S7WvVRnOg{;?0(514C*jN2ACV#14)+g
zSA7svph9QsN{7lE%y#awg8cOTfMuI7mwIzrY$biDI2X~zhD|9WfA??~kg-%vgI3st
zliY=tC9nF#HL(1}BJM_jZGa?Z<$=oN`hqw&-C}b(Epp3)_@0SjV)ThGU{Mwd5fdEY
zGW#QuV$2yJ?s!M^BR-hVl|Q+1EUgJyoBpux`ze<;-dky0HRWTuv;Nsxy%oOMuJDPM
zE`7_2yP(jbFGebj1j=bbnrg7x_^!4k!P8eDBET*8gPt~fFEgi(S^k05o}*h_D-g#e
zx+n%ASqrxkZSHN$*hJG^g<~{iTvj@l#TV<ywPy_=;%#!R0zJ4l@_^qQ$`ugA^{`FY
zrP{^IW9|<5DaUXgk96jzDzmwbl9yxaw()JT#~{w;<O->s_!r|I4?#ze)`}x1n?2&R
z0f~iR8QDS|R;lRkd}J&{pvIF7u-yNd_!1m^@YL~?Wa*2!g+&>qX@L4{NrqG1$uj28
zs)d(X+L{qN&3-T%{p6wAgDSQkF9@(kS;57KMXSivC|z~v<~1pHv_Bv3rh~4Xxlvd}
zPqu!ftgx&1ztF$Fstx#HY6XeK`-XrdK9TwF3o~8mk^wbcg_Ma3fHiK9!|_;GzlV^q
z-vbn+C>06!m^RQ_6h08JD5C4i_4x83*)wWqjytSwYRHQX|Mu($_AvnFvzCkRL2%P-
zlm|i6Z5%%<I{*XxQyL&<k(r*vL$`hi^Wnx2;_R5a$1H3Kmma<geDmyBZWe%t>@Mj^
za=br-)6U@I+Se%!@Y$bg?^AsFdj;<yE?Ky6PNEBBurcTmFK`QnEd3&#yP&%<mcxO&
zN9;zylzT7{R*<pxtFEBd{HHJb_<{j}Fq}~1xB%_tH>>f4FXrB$ZwQPhOJ6hiH|$?C
z6RpK_cwX$i`7%DE)N(mE#P}6QszW+YT*g5C0Q=BK6osMUe^w)j&)!%LyEf!|^e(uJ
zD*GC`PgB120h<$<R*%fpn$f@xh4{{Ots1wjjIlg@cL2-UYu~-Sq3EdLI)IkJrE(v^
z@@2*osK42*pKQgQ;QRglb8+!DY48tVQN2IFmd3=8;TVZAqUnQ%|CfD9be0sp5LOr$
zP`RMqldhg8z?i3)mk{E-dAZ(-%^4*|dE1Dt9Uh-u9c&vpDFJ9jO!%eQfV2qvUs7Z#
zF&`PFCp+JO7x#ae+l%U)PY|1YjoTI}m2NX{rU*h-FdnDDOg7A^eevM|E90qYu6u~d
zB_xHS2<*TAY&_^-ph2`_CZs}d!B{0}BkCMys&fN+te^~$z_g1#x_XGhBQ-QD$*tRC
z9cS(=8v~_N>=PY7z8-^4d-^KFZ}j)J$X3}_{_s%jg+oHA&ycX&dU<cicaHy1#+_ZM
zgqTgYk^=?Yuw0H&`EIj~@Eg#=O4{vQ!h|rD=aYl|)vfERlY-pbr?(5I?JKR963q;L
z{73?D-i!5*rG+G-NOy<?J+9m^@s{G{!*5*aNVkg@z~<@v-j5ui0h^$!q{Qi$!@y)=
zhw=ch?Gy+HMl>b-pe4PcJ8V$f4|6*_7pTtvi^;u4GC1L^N6+TOYg!}G>$won8sXwg
z+|E!z_@{IAX}~VwRJx&|m%Ui;F%3+VW{Ut&I~iQCUQ%agj|l&YJy8Hb6RW7~NNymc
z0Xw)=aHL|*0FOM@Lm4S_b{^1hFX0=f2@Dt@+SI34pR+)gbOHoWl1bg<rbn_gv=;K2
zwFLq({Yh7-Ry})OpvbhCh<j|&WI`Lxgz-rzArbqq+s^l;S@_m(#Ou;H6diV>M*;;9
zNevxc`UJ1oAx?ZPRQD6NNaspD;Ow`Bt{}(#m>#mzX54f{od=l@3DdWz;Z;S^Rv&I*
zNnDFxNyBY+UvCeuHiGAoGMV2U>cBeK%!y#|qUJJ)uc_LCw`p@%X8*2CtGju1Lj2Zc
z8DyuFv{z(Ch;Tk#eNcm&XZ`0tvQ7ak>2!dgsNi`=*BK76&VP%bfMqs0MTt08qYeeP
zRQCrSWuJ4>(|Rq>_`uW@3Uy7s$E?h?R_L6W_qit0FA)oMO@Ei107qTp{N*&9NJf&@
zNBs-&9oJUZJ2~PSP;oWoy+!+%JDby_E>C$ttyWTJR;^785r-+&YZ&{hoH!kz?Yt=k
z%68v9Uexrg8%%EBcN;+E1D1N*Y%BQfPuja6qTM^JYd=KVggYtNOzF`(>7p8;%MBQg
zDqdL?Nn}s)F<O+ju@OX#a29_)rgRoq#k;-NKqNsR+7iqsDVwZ@Z!k#89(*gkl|D`2
zi5f5IwE_{$unCTS$rdEv-bVv4zP|V4uJ~AcSO2oS3_btzoVBc2Vo6*>)L)NLdCP)s
zR8Y|o`3E**hqF2Zc9=eBs8mSd*BQ2hQZszSj%EfR`I2`o*qz3^eDzaU_!PY}Fz;dz
zQ@f(F|2jtfTK~F}=4$D5Tpg7oW33Kuc+xoaGk$=iu1;y~Q!aJ^JJ<zq{thZ{*rC4(
z+rz$C7?$_Fwx9N8-7K{mD-dTA(@#k7-cpcvEz{OOQntM|>=0)zwZSabrcN9>lO`N&
zs}_yLno5n$j-hRsR$1v>eOilms&oyyD7J#Un4U`{XmR(e+Z?+8O2==LnJKIfb2Z4D
zzY%K*1|OEfigWea0F3~U%T#u=Y$92Pz4hEJICz{WzWtyYTie$oN#@E9H&=jWUs>8$
z@$^fXu{2cPxD1BQUrP>qRpy&nMM|V2rr4JNQw<*#Y-WSKnQLwD5;@iF1}D813+H%2
zx~1}i@ne}NG5f(cnx&BgQ^2d`!Vf^?o`a@O_?%Y#^}gkR=!y#{jB<&EYc$l8t}x%1
z8SGokVZv-aUprFJt3T^Jzx^^VARQ~SsCe72ZnP{7IC+$EA^bvn@u4d#aX_BG#9DL4
z{iwn#_MJ=P32j!uNk)<vIkNG4os++BONbTbkXtQX4o;&Brmi7u@TgS&mp`W5R?_xs
zYD*-D;5@5rX8ZBi8z5BjoAf`oA__1|dY~~D3dlcU58*pBB|vvPj-(x_`$gjnB5IJX
z4JZgm&D{!U*Y<Jx?|8<Y6@1>00B0LSrmR^Njm=)Lb}8yO<j6a*BwO$2=>S8ozt7Lz
zb$a*$#w0Ez4TFGxFXU&?VQ<i(2q*>(7>l&7|EJU2X{w9W-={c+>2~GUhS%0XkG5{A
zeaDbKB#>UD1HiL+b6XkH#IXm~$UzQSEfK-%`qmCBhHluphi`Ydn*GoFU8@5x%ljng
z77KmsGVE%-E-_gwDx310q!0-k;;X*|z1tr0&6^eGn|{8tZ#yf6KZ<xoS6FqiuQnE8
zY=o*IE?QoI?4{U78-9u2_4pPLV{!FVVxTC)eU>R^n<aF<*$&nbo<trqV1v}3HQeT8
z@JLCkLpI3eejsxsU8R5?P}H@m@Pov<Hxy$r4xYSG@99rnJkm<<+l%m?CnJe%2#<wT
z5t=Km72PyS7%Qp8u`QA$AT46>K%r#@%7I18=3KO3p_RUq#)=LIwyqWKKE+eCnf|<9
zGsAz{q7U}fNGn5QnNj7Mu|wzE)Q=8{&bQI~jzX^DMG-&@?;@*hL6vrj8(Xq1PP5>>
z=Xkl=XP0*F;IPU_=+HKO&MUfXny~UGOfg@G6Fu&;M-0k=)H=BqBTDT<3+WB{iwTbA
z)iT;83Pm1+of!!b-RGTM4GLtZ;ldzL&fmV@|Drz5SRo;qU;Dh4^9I=Uvt;P{3+M8b
zFwE4)#u~p{*|I)AeibS<7FBhYvu<VVG(%03wd>?P@Pf(W30ks({%o~#Y(sPzg{{sN
zNIOD5VjM|JHuNGhHmdBmQc<4lolPdo!#IU%2)}TGij)B+yk}TM6xHa*Xvi^jPHA03
z8Fc_##){EdT9g1Hx+AS+ba@wSGStNajfJl)*GfcmFMA2=AdYot|DBy2GxWj(LAf0t
zigHCUr60ub=MttX%6f5XtI&$-)Y|4|g|*U_G^&7J$>4y(X{Xkey)BDe^Iu@G0Th>-
z5r606QjP$@qAU+8-e|xJT6}d=1mONdug8>>gpB)5UtY4-zYh+o_h>0c-lVe(j`ERs
zA%1XOIjD%BEf?cWoWe{_mT4-L@~05*ZG4r9Vw#^Bl(B>C$V)$?Q_4b3#`ZBdbHu}4
z%ugIE#r^c1Ks{Nv(T|Ehu`gtx+E)hHJL{~UJZk{*Hry2?F3#~!h_l7$+JSg=MM5s-
zSxhOR1nV8qr4Xunv_hH*TIux6N67LScJkaTgPvA-6q1vOQ|f=@OLY(nq-rC04f1u$
zbi(nuQKaWZmSmpOBBuTP(HNlc8`~>Y^CAu_NJXRLFpz2GVPG()y44QL_?C^SqW`wf
zcvAvGArV2Re|4hw)vErqbMvPxwVUFL0|QZ5vTFYt9?`)v;C9uou+falu{q&A@Ham8
z_n!P}vH3-WDPgM~6x2&)8EUZ~=BauVuqYh{RuBto2evp6*CfwjW=i_F!LM(iJuf;j
z$Za+L5@IfE?RsF^lR|Oe{VS6Wd%t#GYI+^;IFnvphzeFvIH9zHv$3krq^L*Gw{-A`
z>PQ~w#8#}L`zX}WAv2jEj2MhDU0Feb8^VnY{CmOJuxKe3>v-<H=M~?y+CjgPDlXRK
zb{QexJYL$u*0_9GGB#I(XZ@$MY>Th^Ebov-`TXgXp$_1%Q0a+iug`k90ZO>B=VJrV
zlim#7n_jmRo>+5Y@4cNGm{^H&QU~Q^3LA$dORwoI_1|>MVpeSu6@N!8nBQR0+vI9+
z7p=@1PkLo#k5dIkiG{MO{6jkUq_P-DWm|O!F_!Wc&=6EECCjE_{aE~VpYw@}M97xE
z<!sGM9=F}>fMm84LPcL1?RwSXDme*&qNH@ZWY@E#w}Rp^WS)#AeUP<UUew2tgGZX=
z?V^rGf)MXYq12>?X{Ju$AXK=Ws8w~@>>x`+?LE$IhE?>OQyNXL^*tuvPvO8m@wUm&
zK@uF5q1blqU|z#VJ)h1=88a_D#ZUgL9mMu0+u6G*l{!?ylzOx=a-nRgtBC+$L-93l
z(UdULycFvFOVcZFlK(f{Z^ya|`F_*hMhhU?$FEq_Lwv&k5+1cIR{DW9r_{>Qm0Y^h
z7PvMe`_kyefTKI<9GD?g>m$-UL~R_Iv~0FjJAxIYHNJ4X>Gy}TDjzmYx0bMef--cD
zX1g0qj3?S}pe=yPUi3|R7Q6zvP>d!mB`)7zRCpHSrq`WG){E)<GL1?$&^n0=i~K=g
z#`9@OG5mDI99BGlV9Y!|Y3%d)hsAtPT>L!#NN~mD7x_ZuVe!%@+@t%0p4VVo=IEeL
zH{BfceH10?G?4vNR?s_4`|`ZSGb+>waewL{J>a0y0h3{LTUl`SslO99$?8;{W|=#d
z{d+ua8+H_b`WV!ENh(N5T&5M@?n7<15?7A#nqL3Ro^uJHmjzF8g~n$|p24<#b~ta&
zf*qY=vQBhc>saj({J1puB8}jRG;)kx?@EbvUPi0tZf*Cp^|x{br6gAS<365QpP+3%
zMFROKIX^>`=ALDOG?tSzImgEYKjjTuXb0{|#U_6^wrw{mteUq7tvNHh!Ex1nQl5U(
zr_SAeoPGkJ$6;eYby0iCv@!ub_tMU^IMvfcySUqe!hnBo!c3LXi-)?e!lgTNRyjtz
z&YI1xdf@Tf1s&BTp>CDlR?p?UP1%c}Ol<S}B)^1-+>hGMOk?NLx-rM><T=AL!K5^e
z(zR0!!ktIYuFB~=1F>k*N|lBU8@*F5ua*C1^PB>SH&#fV$!vK{#%|qq8l)!eIK4BM
zl@$Qx_5%r6Md=W!xKUJ3lxMNu?4NQ*soEiS-aa1=R4pBliXgz)W<i0qf+xCUjBcWo
z-dEi~6=7(k_Ul>{8#d`c!*E7`fz0rVF&9%e7#uXQE~pjq9(LUh*|c<AQ!o&mxqi9L
zg`r|TAZezt=J4vYBaZD4n&9I&%2rynGXcm8JwKE8_YjIfm1<!+bdI}YFzj%EZ1`K7
zls9d2Qslh|!J#z8*R4Xa#ZLic`YK;vICt+<y`w66e_5&Xd4zTSI2Zqp<!p7GlV*mY
zp_|CbQST!JsI$tMS5C@xZ(F;a!t4KL9!tT^oCr@Se<X~$3o@UJi*aw2%c*f4KLmVu
zIM=tJi10<-(7o=Z8sRH&{vuaE4OZ=|QDTTTx<yt%6I$T?Y=TSxoIJ^bJ$`?9`Qe!H
znjdQ%JGIpoXUkrd9~E`ShWCON>kwgojr-i4(*S{&ZhtglQs#D=$<nwxdwrBElhytn
zI;jtG=ZV0p+PXL4<JlmKBwEKt)&RiMNejK!{<0Wx$-J_AhuhHY1TM}TLJC%Yd}ffZ
z5L1PL#JB{NywE_}WdZDj{4*x5?k|qUJN)QnM!d*M`n<gm><+lpDo3jmA|;&MpE{)P
z_I~;4rhXgJb)c*i?r*?!XGYOOyc36tlW&fHJ_gk9v?Cs2!i;sBX2R}E76aOL78X7i
zz9TQr^AMXQspg)(#;QG83*o5~_D{()XJA=`?O}N*j@|HG70a~l2u~h`3>|SG08SU?
zyuac03GzG^`6Bz5A2Qht>1M$U#;aTW-wD!i3t2ZZ|3cG*x#NCF<M<5*fzEn2-&45a
z#t963{TL0y;lYFm4vaam+5s4cdBkZ-d+dF5S>TL^S6}QtFqlPylvp7ck$<3``$)(i
z>wjC|4z5P^stmw%cd01|#NUbs!nMM_aG)$mF!??{wOPI#W%hz^BGDwqE*shP4iLpF
z;;&iD^{7Is^P+kN8XQ4AfktLRD8)J=x7)V*%i7&zTQ91uwf|Bu$pR3hP*|A9V-U}$
z!C%nV8iH&%k;F(0R7X6j+|4?YcYQ>SRn28?X@ITR@3S0Ox9;g{UbnjabX07(G!v-$
z{PA^?<(txls^s&q5L?=hQ**-e)o5r2Hi&NlVQO$T)|I;Nr_+bj*Je2#1y{L3UR)#7
z_*1R5`QvkiOv&#Y(IZ08$)3w!!bhtyiZ)#S$1GL_SMkl817rp|bNj~~l<Usjdxu)u
zkJn+P&VOYqpQDxiwtN3JDNjN$VfRne2Z1xojKlwa1m^!TqNMk5vs0x9AhLp`{}i&O
z0227m4tsY1TytZTWa{ntH`RdmNB`IBxyD*jlD@Kf0fag|N5ZY&k^3?<<-0$8emuT6
zkt-RPg0dCL%Iq0HVg*AbDP&vS&>01ssztgW$0(HLZNGLR`;6XZ|B1Zojwf}DvS~~a
zy(ePWDUWu93vfj&LgWt~mBrov>gYZ<0*d~Ct1957yQHDfMJ#|G+Gmnc2f8Ncf{1>9
zm;{X7Njha(?KuExVh=c}A+upqHt>`f<xn#64U<{r1Qp=9c9%x!UiLk;?X5ZT^-x03
z>OkO1^<Cg_S*Ris841D@GqBz(J(z)3_3+h2D=ufPuNjm!-z(pw4Ppnmz&TfF0k}BL
zLR5BM%Ak6=qQ4wd%^O4+#g}tnR0<-LJEn4QQ%(=Sdn;a)B?_L|_DJT~q#|elm5fG(
z|7yQW;p!vY`->H&A*n+hg>7}w`Ea)1%=}YHK{IR8CL`Txl1L_cWEN-B(!7E;UW~q!
zp62$|^-C-RDYxZx=Sg)u6eiIhAmwE$U;br@&)OXm=WVt?o0%2MIo4)f5wx|};8Yni
z;9y1tqs{VBwt^puA80Q#9q(^GS`S0=@EXSU(&3-G%KwNi9_i1HF#&d1PxR<DF0W*-
z^VVoHa+@^A2-<>AwbC_f8C_zAcF|9a((tiD;sfJA6ed}XmcL9@3Ux{gDE=s*iYWQE
zk1>d{fm8FQX<2?C4@L}IqreV1cq*b&5_<*d^%)#sP^=qmFVZeHyWEK^53nlo&u6BP
z@21WhwABxD@lQU}x8~}q$&b_Rfxo{Xb+Io@&6=*zIN`n&$)ln+#Z0v-V`7Uf5|E=#
zP{&{)v2p&!GKW0@`kX2P@Th5R&nt>eaq&@?=``;Z4B1{5IdP2W93oo1Dv@Wl8}O)<
zJe;Yek;op}G9Ym<Dma~%t+MbnrTJT%=i-I4Tkn-ZA--6OWQ4bBnt7BCP;^upUH2Pb
z7&-^p1h0t8$V$$xo9|;~Ii^U&bdf{sbZW)h*N8(V|C)r9b^2=pXgeZ`fU9)_0+pgI
z6X>RG-P7_}p;kR(n}z1#y+r&5Tk#fu?ApN;4FM+(^;AZ*N?{9$J~}EIbfHNrY;1p&
z?zVGc`F{X%K#jjmR?8A6Hw71=Ie`Mjqvpv%(9m)5k_o^nFAQ^Hg(}qXLQfl`Fo&V0
zALZOZ;G#E5kKrX%HM!T$jKhC9wkTyWyF{y6Qx*ckjcdwENLocW&Ce^x8s?r0vg)Rc
z(QQunka9Xo^SXqMP$~CsF^R5Y&?^O4-SNCCNi$6oKq^yxAcqDtYd{Cp&B;E7S$uWP
z=j&m`h|snl8Sp8=G`snkx_HKk-5kbpPgQyccbuKlr}q(BR|*@y_w#?-gd)97UM_@?
zTO+4GC*Mqf<3B`8E1@ByhkzbF{O|j>e|-HF2oC$hb)}O)9}m_{KjR~o36e=Y{fN(s
zpx739AeLeRG+{~#SXk>kFgJXi4uWZ%9b-mm{oz`%gf&>iR)hVvm3AG|1XWYw!BH%&
zalRjQW9BMP;hD~CxF>%{C60BEI6jSXkM2`Hr>;2aKhxpcbod-mmC9Ax7zPdF(&?f$
z7!qbs1tLImsF6)gOcBZih=hvOs3QxlwY;S(MlmHgN@$3tfWPu=S?1JI$(@>zR-%xa
zGHYG?0uG(5*pii0VgPlKmS`k{Dhd(ZfN=T!^=BOS9Y?)Y-+_P5i`&=|QcW*{Tf&%h
zwEWg32I;wY7Q3h3IY>t<<IzinS&|>9t*aw|(|w*Qc7j|lMX=8|g8`T~NO1!gy?d%a
z0Mrhw0TT!dK|RS3dsn#_X1vMqHm*1)9-_*#*26VEqO%-k*$CHeBb6fx-KeVcH&s=Y
zbNbGSCywJ*Ulo6XbD*lSTvg@mvZ)kKRho8QsULKybY!f>d^i`?F8KX{i=(blFvEJC
z%69CSC(ldP<2EnhN=RHYiK2XSCQ;Non8bglwp$EaFIC*~I&A4$V#+n9;=0tuRj8HC
zCgV+ME>{>3bva5{uADDTm7CIc)8Lx+96}3A)|EA8Z$W<mIiau9RT?GpXEkH`D-}jj
z+e>yQPRv-e>46%2oZFV(KAmz0IM$5EQ<M{i7O{Fhh5pbvMd>|b+<R|YzPwfXvv`KZ
zxyE3f-HfT0qHM;L9rlF-v3P9N5}UK48<^hY7!68^yh*uU;N8Z)z*Q9^<wYF{2O(~@
z>bOUwCsuzS6+!u3-U`_~U7L}nr!&^~<AGT`u;S=Dl;P#r=k628j?(%GWY;-YZV<pq
z0V9Jp&pfl7XeOR{oGQ6X_It`p!YG&8>b^IhJLNlXE{)51Cs$s$3?}l~w0s%Q8<}BP
zhZOhABW2NrRGBbdkON(w8mMibSh^?vDb7RppYVS%Oox^2=Y{Jc;tbP7#s^N_CVJpw
zKMz<TS@e@VlC--<u!EumOzsDv!~+B|A)sQQ_;DU0X^EBtsO~`bo+ZMZR5G?_iHs!p
z8J5V@s`8l9YaoW|M(H<?qZckAgewaj#cjpsw0Qc;FRMFDz}22+;NWz>ws3>lpdKhL
zXiI-m|F*Y0FFRd#uv)eSy%TdThE>Pdr+D~Z97W8$mwT})42-?@<C;)txn0uPghJbU
zjLG-gPfOVyC!=tqo6$VzzW?JXH!U?~%q%gv=?K$PAlxZXe@mhtwUUPH7MZ(*0F%+@
zi_BegmTFdJSeDsEH?V)J%);Hd@q10rmRWy*CNP?vli4}p!BO1dPBED^n3GlRH1A1Q
zd#I`SD!Z_&_qsZ&H{5;_F+i+4l-&lHf@jOmR`6_l-xD~MA@~_^SeREUqP6aO%k%zu
zQ~2%Vo!7*{FLZLt?KOEcm~(dt<JrpB5Z$4B>tX6{-D}v{QoW}0BbwKOu%UQO=SP3^
zu9X#nYu-}3R%W@q8jorne%rDM_AIq)sqWCb>))eyEk9d|*Y<uy^Ew<hRIly*i0*ar
z!fK(Cm9LXqZm&tDeS>CUyrz)1gyhD3W>0EKPcDydfhDe}BFIPPN>&@;>1OYZTVGNW
zg_Jsq^xUEmwt2uxLanIt>H5A;y!n5$|Dt&?w~qSKCXL~kK#jaqhsd(3=H~AAcomrC
zy<WT{y!Pvtt+&U;2zQ!TD@m$e)Ar3hn^y^!miiVh%e*XI>=vx%GhpGBc~+4<E-P5L
zYbLt6G9Na{_RAl)<2s5<^2gf0pMQ{RKOffJhB7_nrT;o$-90bf_x4yWSK&wQf&~LJ
zTyuYXw;cTTm2K+#avf#i|J7utv3~Xhz#AJ(c6V+scieFfWl{C5`9C~{Y6W`&Vi@uW
zR_86<E;XpmEdK(lj@y^PO9c}GIWw1VzXueTX+sw%f6ZIXa^p4(zUL|Y0%1u4{F~YF
zOq|_xrnh!->-5y=_S8fB{)d7hA(4<|S(cOb;MkU_5C0$tDEjp0NuPdvy1YOA@CJAa
zDmoMF>HYi3&<~pa8ci3yI|XBu2XlJ=>-6gvt*ximg<hyN0KEczj`Z%%&tM|0FtozZ
zFUFjoe*>B?WevpoZu45Y=e1ziH~Q|+&!~;I`R*5b5x$A%-`;<E(v?xIlFuLWoBa0e
z{ps%~$u@#j3C-#Ezn=0B2u4A)CyPPp0H=TcIDLQmS;CIQG~J_I{zISs`ST>M{8BZS
zG-*zI?_8%jqe-23@;)iZKNZb6(tsn~AW7~Ff4#qS68`WO%_16j;@n6y`?VmFqR9>g
z=?1npc_#fJ1g3(i@}2o5-d(8t+kEXd{xK<oGh}jsf^?ZQlF=Or<XoYE<O}*zDC@WL
zv)9?1ve%Sv*<X0AqqRuBgl_HEEXVX~Msg$_VqKMKzZUvfEj8gRAjjDiU%7s($gv>n
zf7d}zqdqJ-D219QyJCo}52)Hyn2p7M1{-!)$c8h|=#i}M>@wK_RZMd6AQ&6?r5gCs
zsC|Wb>y$I8HRd3CC}{Y$Ul<1pQk2MR!g!;S4stcj_$ZQWm_{ofh#><fGX_p~rR9?~
zU{E<fvI)I6WslCEvxgKFuZZE5@e*v{e|Slf(Q(-w;aer8ckA1d=7L2*+Ptl*+(J|s
zgSa7wZ{cD;2hJ$#q$0IQ`+M-n7xYXbJOG$fa7hAwT~?1rMcbZO4CX%L1g}B}JB$;P
za#D0oQInT69)Dn~;0Vy5DD!n&MNYNu#%-abBLVUN_BfG9<k^mOi$zx1=g94;e;n{O
z6X=TZGsI<gGXW_@8M%7gMGok%7{(>5JVL8JOt)Fq1?3$`gssi?NG{V+1??q8a>hgX
z#2STurir!qulZE+;5yfn`<IJUTt^&r<f}e%*0**0;Bt|5_gZ&|%i|-vnw~r{*s;>&
z?T4A$;|kSF1>#4#0-K?yc+?f@f8Hlhg)30ekXYNasFV@yl`E>ZXcZlnW_*v8@ooBM
zx5_)0_p5Hz)Z2qaUz(EWV_gfQiaNBeHTZTAirgk7cOwV%j&T!tyV6@#E>p5z#hAL%
zk7M;~RcL!uO6%ocP5A<GyZ!?`8L>H}_&dTSsYIs%$-&F<P!WhLSrffVe`%(H?0^Sq
zb7!1N&>9FmpIPf+Ak;b?+i1AD)Apd-Hkuzy_ge>0S&;J3QT=KuSjuXl1H7X=FV(M*
zd#(D7ie6%|yWCy*?h`qI3JE3krL+I?ah8q^6b56DYq$%V5aw{J;XNe9H3xWHfsH92
zDQqhbXK`1v^{qXaT9OGwe|_n-QF1g!P}+4h$)E^qNbk=y97Y0t?R-zGOqlZ$rj5#J
zQ5$uwsj|^6^WGZoJRGMx`wc<7q?cfoPOfzl&iP!>+|&19z4Yi?dRgaXdH<>U82PLr
z5!W1f(90EFP@!R<D=<obWxxEjWD$sK^4A?x(x1t-pO2I)@hFjve_7@uQKq3PU##Zh
zX(i2&iO70m_49~SQQNi%Nm;X9{Va96EB5LD0JZ625UY-{_)Fig9q4V16d=+W$xRK*
z3rF?Y4&fN{C?W8n-Sf9GvLlZghP^fO^EzI!!{LO#HQPLI{R5`V6;n&T-Z^wLDnG+2
zT_qb5>`XH2yfQjTf5{H$D(`8@dWpzKj=kd$BB)?K4$1KZ-h9e91UxN9`=D`%K>MI^
z2v9X^nxpXJ#vuavx~D9SL7*FhJZ>oR0izC}cjNs7MjZm_`lv%JzH`(8#w)AC2aGxd
z^2Vq`tp3DN2WY44bhJJ=>kw$4I_wa%Z<uz7JC2V#q%x1se><d5#|It~@%qF=tUfsM
z*kPhF^N<$0HuTu<`<6~j_My!lf%qRZdn5<Pn>`Zodb3BYzNbr2U0>4ucE2RmnFV5b
zj%DF#lpV;Cd<)F2Ix0^602S4q$m5I++A7d+qactgC^HKn*x!f(lN8un6__{u2y+D<
z%K~)lhDSFPe|ykRZ7BAnZ)+$D9o*kg6mnm0DE5mFG!*xV%)!%BzrCSo1I5b!uAykD
zl*ETN6b0gPL(w`iih~boC<^qA#+l^hzicR4ND32vP(x9mZ|P;7mrrRZS{s#nXhTsT
zZZs4n{l_&F1@cBiQLNr*C|VPg`;dmB)n56#8;UNTf5@FIfh@KE4AolAuy1-^m|o%h
zoXcDuee3#Q03Y4;#LDQ7+_%G_Z-?Wj+jeMP?V7`E;F0)E-AsddMlqd#YFmMOmu`>-
zdCy?{$iECc@^JpGkty>1=F8NqnC?UFYt|kbj<m)k73;3kp-_rq1mE^to1{lHFr})^
zdM9)Af6~%l&=~(Z@1OE7*wMeiLUnaK8(&Mg(j7Zn`(=7)WPCg&jNF&!Zq_w8J+W5$
z%X$j>V!h42@8|<yl+n{BwB09M^(u6fHb5epW^*zdJUL1slmG<+j1Pfar%#njPFs>b
zb4`v;@F{p$)Pn~4ajN!Vsxn5c>iIKGgAYw*f52A6MLO}f%036n=5~`S<;ocbml<~<
z-v?kjRv5oEZc?vIisLJ6%4y9@Gh_%YRt~p1%vcPV@_!O&p-mleI_*z9aJ@0KhhMu9
zN$AETAVLqC8Yi^M9$Gt<_JZ7X4==PkdNnPV;AtwNUixlavoOwiX{2o>cMG=Yc0c~l
zY7|fw!~7wqafZ{A^VI=3!jVbw+^G4M`m%=wEe44#^RE^4(oyAo3mXjIPNgPQM|RLx
zvA=myezut@T%URMhP*ztSmX>lJ$g3pc*S*7xY%dkMngPnyR389KAa|sBYyw({`4<v
zeVtdA@w7A)0Wy=}WhIv@wKNQu#BDq$f7M!RZymQ0{qA2upkLAt;gXyO$-*#@N1Z0H
z5yMV_wuWHnapNdebPee$P5*t*aK)9ByVsKCwa^F<?(8m?JUg7n8Im=H<btayRToJi
z5V^$3)mXDjrJ7J2Co-Wr5o;5IQ+t#`bR{KwE4nIT0<KDVgbs^~fvjs*7o*yHe-d0w
z+1jJO$iccu?4mH4Bra;Ub_s8k@o9FkavOnN3b_p-E~QX01_0x}y-NVg<J01jhGJ56
zDRV;*@G05aadZ?ylaix|*C3Ijlv}&vh{vZD5=V@|0;$+N04I%xDrZMI*=Wp;N{;qz
z<TwPoC+s+qA%I4B{9dSyIW!$)f5$@Bi$JfM6ULD0xU{Dl9M|?#qm$4uYjToW&q_|H
z^{niK8)g+J+Au@KlG_|=c2e5Y%uYkFF`yhYvtFQL4cWdNz%|+zQH0c{op7o6VC$tg
z6|r`Tds<t6HuIRWnsBMLJ+0B@P;5SGaycc_MkSX?3{_>9x!AZ>T-Inke^+%mi}eR1
z%cW@-R#rl6?W!x$5MZWYUL~3YL*vMgASGAu>2{A?(QN1gyVC51pr+F7g-o5zUMLvI
z@u_t>KE-9T7ubNSdDH<uD@(>01|!#Y54~Vuld>5$nS@?yYVYC`su~t3Hil@^VD=cp
zI3CO&Vgm~YiGgDi6GSoNe<HHBxxEHcjG-C^^OsP^x6FxZX??*#F@QmVIYZ2CFm#JQ
z0|N&!$zFr<VzfD^WG)9|q(_e*yWy+#$@=WoyZytJBY6EwH@w*2Y*uHdE{&Ie{cZEj
ztIhsq<>Gkb>H2i@`0?=6IuC{I=I|6V63uJcg^Xj37chz_&@wKbe>@ppoUIREtv0WR
z7hgXeepvms*?aN?IEEko`)-Ai->v}YbhA3$oSV0X*2T-!`TFAQaAmhr>u7ksIy&C}
za{b$DYoQS~R?{BdKf{}t<2a)F;`DSqc0}Xwk6yPJW}new*riZM!D85@&^~o<0@gYj
zUR@k)?CRO^>A&6Zf93k@Xm!>AguUT!!?(j%uj6>}a&@?Q4Lk7y<510>;klS&4^wiQ
zix;82;R}cEUk%@^e^?J+yU(IVxHMuxDNIGA=3M{$<jFPl+zSO`RP*1CLW(c2f?X&|
zmPjnHZ-P?`K+ysq6oVq%br2kiSA?crAQ--YIfHc}Brv>1f6K>%fY14gNTCvah0`G#
zc!W_^@)CG^I!F+!5lXo-0*qG5VF3_SJi`^0s;?Nmg8(K(_H7Wb2t=NYsHb9P4Yf=l
zzJA?Ov>Ez8{`oJ5z?~6iPcKeR_GTL(JotjtpecDM_*%>V?<l}|D+(Y1-y;f0n4*A@
zjnUN;cR{+Ke_gnY6vi_yh2DmG8=BgbF7#-^J(_UrLh1s2e0(ueN(HvV3=bdenZYw)
zv7RJ;m@VDkwUhW<@)e3wy%D}$C_tM+h|1t6pasaebph9ouxMmRwFnDuiZysO42XRs
z{6L!MAn-%RoI3~#@c4(CjuhB~UQQ@L`@FH-T_`@ye{>+474TQe>Z`Gls-D2_s30TK
zOy*<BpxuF)QlS(8CoX`3qR-&9pgTSolL8uJb|e)cw>&Tkuv%n)=%PdqY6?<{SUjT&
zFjc7q1e_UjD4BUd3`n<DaF}T#tD*kEe8langS1?Q6fPZ5`GCj7upkHl2LSgsOI5IK
zv(UJ$e~4O*-lBf2C^Aqpvg(2ukbQHsf)c??Tbbxfno)!$pB%uLGps{G%(5Ve6ugzF
zk@5YwGGefeKoP*e#)86&Q4=%i5+sD-m#wmf8X~47dMo*2&WM-#K+KdQJOQOzM}e^A
ztzzDV;*&5_Zb_7{K)^*%XYB;>K4xlki}f1ae^R)(&XKYktwvI2+Q2t`3Q~51huvUk
zA3@48ZIQAZ??cLRz9MDZd6sl`CexkHyreTP>C8(y^ODZIv@<X5%u74-(%!T7p0)R^
zy=ToI&(6A+YmSJyn1KqyZ^XaF{~0=fBwVYl5h30NDF}ZH23-~MA*LzvtHR^ccGS%(
ze~5OR;T8-BLz-ML;3)wFO3z6Z9DSOtyx?0>Q^^4wH-I0NB}j=SJ6MRovu&*irsXsE
znH1!o5+4Xf$`w>-0s&Qf(f5G(q*j5sM(Rn^gn$e9G?RoXqFiLoFbc%rP{0Y61Ht%o
z<XPlFTO-JhL6lHb%%~zprb-}2DWE9Fe~c_>n~dM<ssQ!zIc=*Vg#xn;WV0d!W#t7q
zqU8Y4fr*yjLp@Z!tqKrBEfwqu<wF&em3dLV3}S}3dr3@485wehF55~Pu}us711~P(
zq#XmPx47Ohptz(Neg>Z)3snLTa1dl>S&%N<0-hz(UN<C^R-w*NG<a?XLjpwwf3u1y
z)Fs(6m@4#Op|kB`fGu#HixXgI_+zlj6f>8!B#1yh!9x>y&@W6%^h9Z<fFdOziN=cv
zwjr9A0}%^(Tj=j7V1J%x3I|&|N0C`EILrZd5lEI;!3@44N+EA5@+aKhXJmE!&|E1A
zhL~ZNP(mf-UR&kAK@)SjW!dgpe}m2tV@B>qfV@LXSgNJE2VCI$K>A2A{hmJB#>t1%
zN86g3=%aQbsGSJrP6X>DbVgV^Bdnbf*3JlPXN0vg!oFva`kq1Rdj{0^3@F4dq%Kew
zxC_z+?LzKC=|Wv#Z7>AH&ouEq&|cUrD3>O$EDy8{grL<n8{x5Sxgk|=e>-x<1LN!h
z3__PJl0x0!8f5u(iDmx)XP7_`Wgm*{20?_tB^j!qQjm$u5_=D@cla(CRwZBCPO<n3
z^FaRc`ar94w)&PyCJZCR(OnoaJ{;acj2cz|WsyA?wp?l?{1StZnFK$m9+06BB|*89
z&kV*V>TPcv=|p9Qb_S3uf7C-katSe$qVin;Lp7r;69_9t><Gdx6rV&9D1m)3gSZ4i
zwj#%}69kIGqHw1oV8GdSzZcq`%rs%(0OA+x!zzMN_k=4wpz~%R4{8w!6=7t2IO0=9
zmIZb2TH1*Yc*aa2l55E*P7|nUKn(?!y1;C}EJwE-K}ayVLpwo$fAmU7um=Roh*~-D
zBKs)t1tJuJ0E48OqIf%QA;@V$b0wLg41!q{W(jKb$O4+M_4#0jc2JG79pg#fj$3e2
zXh<JsI#AwtL~RT%_$_t@3kC{I3=G>v0IFsbYPmvD&;{mekzI4!$D0tKyut5I1Ql@|
zoJ(_vAk3iP%dA#ce{ZXPpwgbQZD|E>M_L#;Bg~ZOsGMQ1jG9aeK$fz^I)SYZWT}lv
zLB-8OFouKgdx>erC}Tp!0^$$Lp&GR=V3gyEa&7DN;7O<o<G6i+ssxl}hRcM?C`Sv_
z0n316L|JC{AgfL&Y`vh;ZWJ1lv28Z<1j~h8#AQyDu<`_ge`^7)t2;q3(gg}Z`~|68
zk$>kjF+XlVMHHt92~KMX8%iJ@H!Y%V-!KS`q`gNszGL`e>i%$Zygmh>zrXy}{y+O^
zvw3&^mtlB&y!q+kz#p!E8GgR_<=y8;t6zt2F3#2m$NuM&;oxL_FdWsx)xnBahe7s{
ze2#uBHRd0Ge}o^Tl^#;rKRDVy3^ferXNSY{@4tQX%f-p&_|5v@XCOZQb%l=p^ele<
ztWEx-sdBXQeYfvb+-!AufTV!EA}!qBQc$znF_+t0a;`1P-`<jtrjZ@*(lXj+c>1I4
zsnB?+<VI0Eo;<mwD5mjT$yHIzSI6CmdVNa0KBZotf70A3l-qs4)}*YCL&En7^{Xw6
z75IXJEg;^0ln;D&adN!>{kzTZe1G#3l=t?_{c|(mE6crq+_I!4AT!|q9PKPGYGF`t
zWc09IdYrEsT2CmX?-eOf&3crNV3Wlo5jsoeNJkD?8R?6ci4@;mJb$r6J9mg>gqR8U
z;7ILQe}EN)wjBqTfqUO$9u1bi25UIq>>r%0hRwxheRjNmaz4CWZ8pcJZ{Og-*=BY0
z`|CIV!8g9Kw}+ddYCLzfd%f&>gDFZ`rC;yTQcCbhk9TPa9u&4C-%LxR(lD(%w2TRT
za{Z$m$0M#=QR?!9-kqY<l`govFaP`xmqa14m+`bT69Y3gIF~_ZA3A@{9J#I|xiO!w
z&<{K=kKzS92+-567I0`xW;nH@P5l3Xc@)Xw(Jg9513S>@shdY#tee9x|Gx0cpWnW|
zfBWsB;9{B6!ph6{e_cd)!Nb3ThlfUAtPsp7aryq&<sYA%OA|g{;RBp26+Sln?ZUUd
zekkF>2oXkz@J5L1hXsG}na3!H-!hI3bMAQ!`3S$w^#i!zGS0q*Pgr00>p#E$?Hflx
z%XfuNDkikQ7z-?{i_m>>-OZ|_m1H)c*N+bShkbp7SqGRQegZ`B$lqGWw?4cDAASY+
zwaBA%zykvOTE=H(RQ{Fc>wBI1g{zrqoU3EXdOSYOZxclKHadS&(;FM9$ukw@y_@fZ
z`l>5;``HXdz(+=I#J72Ghi`S<M@{o&#Jv9c2@2)e;`bmA&cxA;?!K**V7ucE8h3Gt
zpA#RAWZcQZNB)($&ZKtXDQ6HqPHA=dPmqK8>EHhhQt!l4>XnXT#vN3Ja|L6T=7hUN
zDt>ixczDLVWzv7S37@bUe_TJLQlGeQ;Jy?x{R(`BCP7k&4GGP2-$*BE4RahOIxVFn
zpY#Z0`rMn8%0sxX5v>6F4jGjdOo2HkV<LraWGsJa$RH$;EwUIIC}93)L=NCE3)@6i
zPcc~${94W(qOn78Z9N0Clv^Rz8o20dkm1|y>g#TF(${|yRM8u*v7s^-V24p@4XT;*
zyrkn?TfGnv9(CmnO`X?MNfZhL>U`7}dEfFzbylI$G1x>t=zE0xK;I)42l`%uJka+z
z5B2@eQ@H7S<!ndStBm?z@U>P~;dpRWxIJ+|KxZ-5nmFN;M!wT3c+R+RBZO*6qs*;H
zA@v;Aozs6NgBnoQkGnVZraLPEODnN;F5XQy1)g~WyL5ZqN^EN^0AQNia~d`i0o6pn
zwu#uF@KWl9V-oDI^kw=T4yE>o%tw5eZS_7gWMoD(1_SCv?Ik6lP>n_L&4EQyx==(T
zK5fo(5ci@8iP|e*<~5Qky5an^dRgPhxy2oxD!YFIVR~xns+JfTTe0Zy@=fd#eAR)h
z;Z#w}xWl0V$__MvX(?>nFzKtoE;y&dZ_@f26^3!2y&;St4UWP6qX{X(ldwXC6aP-i
zzl8?nV72+87{E3U&`hKQY1_y;uZNVn+;@VDo%S=mVbVmnD=fB*ey94gLB72isV<*G
zzNLTCWPkw1#j8fMqp(Hv&}Y!Z`!H_Huh}6xv1y~=X%9@b$U7awFPve?-?|>u>G<!m
z5l5q~O*7o_0d&Map*w(me&iZ$#^YZhpy~n-lNjs(=2^5cj6<`L!7(>kI|PjgJZZG>
zC_ec@_wCW^LrBS%<oZRhZskdqtN2R}r?-DPA`~@SM?zgw6DKC3b89CRtSSt3WoFV0
zR1ceReK9Zs_aPW2q3Yl)bU=;xwo~dtl%?&+0C|9Oa&1LUggqC>#O{b%w44bI^t}~I
zP+T?+U2nG+jVTilkRt*|a;71~Y_7=}GNH(>(}xT%Z;K&LY|iKX&<6>Tq7`D?RRn)Z
zNRBg)aw8%3^W(&~I`uk?hj9n_Xnc+zMXmUzf1?!Rc0!upD|t6%r!fT2@ENItrD0ai
z_=u;ZR@^72_}9@|3CW}}6Kprc+YLrU{}A1kfv^j_gYMZAHN7X=9}&}gZlk`d-cdFt
zN#Tv`EWhYf>T(%kZ*S|<95W=UTK|8}-uoJRS3MJ-Ql}EfOtY#xS)6Jb8$6l|9h^du
zG=B2ilkQfxt9GB=V(0-Tij@#OZ!9~Y4!m$&WejQ4oE_gh&yUHNVx@3y8Eb}Q$*t~<
z4!{YaOiSj}^uvP0DVS_nR;D)FhC7_54Cva8GQ=KE;t|4KVVm2v5y)ouYbSrf)8aK)
zof6>=Eqcy5K05ENPAlXhzS-!q+_N8>nAQBc)ye7nvp$qM_Aql0)<IUnlLr#ej=Q&@
z)O9(#YKMP&Xm@c;du!CH&ZfC!#y0oEU#qjDi)Rd$THoS@Lm$V-Kuv)|H>?O#H7DRH
zwBQQ0;Uym%+_9H%YSj^2a6x}nH^Efh!4j)D#fu4}_*8aBP1K_%UkUuYIPrOKt$R7#
zZI4bv-d4N9pjxyoYjTwBBu#qE%MO^m%?umCAPgC(|2|h2b9frf-7pk?|BS0EECcW(
z0@?cLSqrUnB3Q|k>`W#ld>|nnGVPl9DGjt_;LBL}B&z6vTGf>C)y#kAHTzUL3PpKa
zt=@yqt_mtYA;N}hKCu_o?a-~1h-xci^ol3iLsha`8IB3o!w@2`f_b<e<6@T}$4wtv
z2@HjG@{<ai?t`Q7$nRn}s*L;`atdayBNtIHoGD}1qR`u;fo0j{Big7Kw{pre>d1iK
zskN|$IQNr`#3+=n*)M;WT_q^KpBostuvq<S#m+Qqn5ldaQ--LYK{&CKHqZMezI3}p
zT1rxz!W|6}T1)9aBtf?mvePjZG|5TwJ3}=Matd=<n-qrmpp`%X6BT6vQ|a?PDg2y1
z>o<F+zAn0Yu^h;46S@_h1qa7R-Q5!|c7-TTB@itw*qx#~j(UHZ)GBMrG4W*2Y(Ksw
zY7?J~Aw!=54Wsz9j~pVsgDNNlG6@Hh%YGkrOww$&dT1t1*5@~03yROqgmN*=udLO-
z^i1@d6<;ca-9up_>6P~ivy_qrm`99*Fl%ns1}NCL?P`)<Dz7wubSE)5$ZArYW{6Z!
z(=rfupd3;`yQhENyE2OT5$P_MFDmezL*S{Lag@yxSHXO$5!Isg^-dQaj_O<-PrDE5
zla+H=geunBx!#egatdYgIG~ou<WzT&Cn@yY=b?*MuPaomTTVAx;J9+G@4(?cEg?~#
zZs^AqA8<DMS-K=fN-SjLQ-OS%A%<9`o(NyiPTQKNf(d^iu+hP&jH?h~c#ScX4O*U-
zwt~Df)g(qB^_s6Y1`W-_ll2!Vn55aB0OY2c*nRTjUgqs{U9KDhS@9C)*DCXVw`U#>
z4HSyMcC6vA#fa0Uk}~C4Ls1$t2L@XnkJJpD9g!na`MyRQ#hmlMj*QJo9K3>`AP+dr
zELW4HMUQ__97J&+VAR4E9Qaa78U|f>H(6S#KKpQFVN=7XQFq|HRlq0}HC<9|Hm&86
zo%za7UWw49T1DtR^!IMRqbR*XQ{<!6L_qm5#JhLU(=a*jOe(+)uc`ohEK6SrNtO**
zty{uXh$^(`40kp9ix5)FRpWs>G1MfeT=w-GqzHegaU5KE`8lQq2BScI$D>36xlDj-
z6bX0dkWFDPsY$eeJyD@it?14^^{r1E8>BN#%6frB@kIs<#I9!ya_C!9plC=B+wKnP
zyM!x@ol0b#!%n3#s&Wm~${o@q3F9fglhIeCfMLn3hPNMP@VXY23^?;^6wq#X97|F>
zHBEor3Xg@xaK3i6Xp;okSeMnb>uj@gq83h0N(mC8|D}gaOfVFHz4BdvW2lBN6mb*x
zr_=Vc1rI4b^V)Z<>|TsEq)Tj`H&y^J&ds>4UgCQWVc?F@Fs@Wib+sAAk4~kFRscCJ
zn+V#^s7ObqrKOZs0~S+-ReY`rBVsD~%(H)XQ=!9r=n6uE-o&EKV9jtW?Q%TAWJ;)(
zH>oylpC!1@<)O%#a6v9wsu0_b(!Ll3i1182kqO{I7N*s*U8o7{uC4C1<a5%-*p`qI
z!hPc&Mt#i|4k_FFpvP~t^xN4+J*Bz&p0aE+zpC{Vh!<}spp+rW^_Ge1oTrYTgQkC~
ze33(+>LOwd^M$)H`%|wICMrKL9m&gS#64(=(6^3gx$20n*8@{`Ag<JEGPoj5V&*3s
z-fUZ~CQc7ld&R#U)_{&Ou%hBwq89qOZT+Fhj8DsT$Ptf~JP7Al2&AblJxKm6XIsvM
zQ>)^*(so(1?8-xZ0`A`B6nAA%TiSn=k#`;ac<mO~jY`8pJ=wGYSC1E1^BRTz<d>i@
zR4Y=>JZS()CtOb!e3687PttHYc{!;&yJ9y`Pm!xz6D1Z{l&=08(JG`5mT*(9B|*F3
zJJ(LOVnK#-)>b)ej7DmQl|7F_;53J@hB{ae>Ge{6yNIFTVXHqpH%^0DHwu5vX|&^1
z+YmJgTCjamUhNu_)`i*T+0@znJ(KgSykdf@k!tdFUo+l=g_-_GT#}M0qxxG3^;Rrt
zJHp$qr@h(b8r76z6{k7YyjR~&TyN;2^q%`YflMs0X2j-w7m*Q)Kd$9}GB8h?X|-n@
z#0Ze!Z9~D&1tPG~88b2Hv@w4<g17}ANX@jI2)5lTC1}WtqR}o>ubUgsu+>2vO@QP7
zhJi*HrktG_xIRUKGvGDHxXEH6;zK*m-tULrX18#L888!^`!RTkr0uB1)#|?@P)hBT
z>GE81`5K?p_#u|DpTrP~w*4tJUu-@vIdR3bkRwghc}xU6*ka|!S)yU$w4l&Y4klCT
z-eL2!nP_MvQtCB3QzX_q23l?+Erev+t6@@$1N$g#=f>mfSGl%VJ)*X-8!a2HJz^1k
zQfd8>Xm|1~b|p2)nv~`>p2Y?muScGZ#XJ1|<NLS&0!nS4mth$U69YLjIG4fTIuijl
zm$Cl?Cx69Txo!hD6Yl*KdI2*W-UJK*mSP7;m%v?+>PM0)Me_bdhR5()?rL@UFV?bl
z$>H1_ezW3-zYhHH)641W<%dtH4wg9$Rvf<mJODo6_!l_dH2Poxn2~V!`up(P8|T8{
zb;3(=E+t+T|9AMQkFOFOjsP41_ypkiY8AYtF@MtKPZ7rAoO7ZfKJmLbzA6q}gxT+S
zox<nw_phH{cxIF{Z^FOfWBm5<_43cl$hJ|yz`)^uf4_tWN&=Hk9Hc`@)#2Yi4u8J<
zGE!}o;sBMxjnTSO3h+4J7|&#Ql$4@TWR1dU6ilO_Dt<q{8Xm>RZwkd#D4GToS*MZl
zo_|gE70%KhK}AVyrf|XX=ik2$!<}D$suFuNkaHtXfQp452PGPh(|MpFbfVhbL8?D|
zdOnjJ&BnmnaTN(kVibofo#oKos%yFwLNhBCx-|T14G5{2kYa}h$_!<gN23ojRxM+$
z=gAso?LuD|6jOd2Fx2IOpWsI#c}?qMN`ECD<NfPBMheNa)jN5U0wYQ;SX0U13%2t2
zxhmrV#c=~+Ogct8Ceav~vU-ltVDS!^aW(jH)k(lQ<}oUU8T*WqGF<pSN*|o)ltbJk
z8I?MQ?4Ul=)IcIC)#r<*<5b~ultEQ+d`@h9q~waR6G-yUka|Kk74Mlhmlq*kbbp$l
zPWgUEqzS^988lBhBI>uhvym`Ls#KOKX^VA947-)zOX|>WYi}~#Ght=1rB1t!GHNxL
zahFn#ylm7=QEBAu($1(oN@$B@MW48aka3Movt<nn?Qma0-^9yFEt8fPcYiI3bjc6;
zr@+wzKkLaXXlZwHfwweVt&mgVx_^)@rkyJnYYHZX8-<BfOi`@Kv_JFZqC@4U`kf3>
zBS`zQH_@o2xmFw|KkYd<_tKUi$|i1ut#TF}LbP0962SjF)ethx0Ny7`a=4g~M4Kqp
z8RRKi0O>#o`LGj*H7M-_{e*-dq$J&(N%}IU)kVCcXdLMc!V|dc)HGj}Fn^~tOr~U0
zkV@yrps6$U5p3q7R2QylINBj{6TgWZuJiDan1DQ&3++9|pV$0Y36HC}#i~g<5hZn<
zZq8ZGbGs+B7RquyB4sG7H=K|&?xHU2!{b^jha1LXMfA+6ryH!#Oxbcruf+s0cx}3a
z|3$yirz4a+;Zj_Uewt|nDS!PM#S~k*lsRRc9X^Uh#j8@1az&k+^T=YwO&{ixS-#Lq
zh#q~Wys_zz602=+excb-`PWZ2S(cb%vX-D`yGByDR<|XiM!V5ne(JMlwVrnEo$3ME
zN<3>P?-JSHR<}~}Kdl*Nskxq;GH&;&t}Cln3kz@YBnRW_{9J1Di+_Aeti-JQrW^;j
zN86A^`c-Rw%T!&Bmb6-c3T)XW2sl{VQ&SYvsN%7SUh7q`>9MfsvCw%exC!lv$HLNM
za|tTseH5ly5j+D`4<d!zs0nd=jXNU6L((wDgV~OURZXtdLz%K^*4*}?Zbr7sXQy}F
zt-Ea_l}p%sXuB2|l7Cn&TV+RJ2ylPGbiIuX?Kuxc^l1A?q`YCeLTY=C2^X*wP{A`6
zlply^&|!F3VFfb)CmC43;O99<dqR@-@g_oLhS_Sg!ri$t>{8e3HG?3^X{N2%fF<rh
z??4*#au`>c)DIUuO-m-d8!6qsBbGfm%8)}zbE;{ZJ2Q*-Tz_ywnCsf%Kfxlonx=2A
zo=(>MnXF_LGf)7B<yq@``%~ohl(%3hja34c_rnZ1Irjx{J!;^itA4K1G*?V%7*|W0
z7Snaf27{LARDRAo`l+=B?L6tu&f0`V=ZZRiBT$F%vv=aBJKebSdCA^JqOMiV)Yscr
z3I@WAl(y`vFn_M49%oe8@DZ3{gmv0Lyu~##h(0jd1mOj!U>c0nM0l?OOr_B!V`+g2
zv1!WOZmgct9%J=0#An6Y!nK99-k^E#nt3c%hE=QPWdU0rr)4T$6fi4IJ)A58+!6hf
z2r4{%wWMTQ$zwZZk$%Js$8~KA>QhSuv{bV><UWQf7Jn;=q=B%mNlQcnT0N^=_MRJV
zE8XT`yP6VLHq%Qnfky@hZrNM8VWU#a`H`UDD|=L~5{Bh(!k5zNe+AdPx>ls5)pG`O
zIfRDbnxh#0)*j0Z1)>BUs<535OB^KyjH7SJh14V**WGLqj>{7ECaX#Unt@5<)<ht^
zWqwkH=zkG!pe+U(oS~@~HgJX<f*EJHE+D(wr&iW&ifnL9N;G3yR=!Q6Q|`KSUV9-E
ze)}`T4H|bCF-k0TzAu%+15#-xk!s(q(C_1hBi5RGeabzLB?#4~%?5Cx>pFzOUCmO`
zwZ)Cy)$_&NtM{GL(kWq}xc7_}@z%yhC1USy2!Bh12~DV{G=!xv_^{-M_$pFq2nX5Q
zjHC&RYrLar4(~XBq|tgrht00`vE+_n={#*8YyQ3MV`covkKWK$JS{)AKZGJPq1?zt
zd3mC2su@S-{b?t51eG7lP}`QIq*uMv^jhb)jBj&!zPqWPLRppnx{fO0rsXMVlHQ{t
z*?-ie!<yF5`_tT4RyxQcU|ziL&tp%lBCJ4|t6lbPr<ARW<-+LA)a^2RddC={gsQj(
zBTsWp+F9aQQ)@tdrV&whq_2c9Z1qe89%L{v4vas!Wg-QhFg=W`Mi|~g3t!};eg&tB
zBF?Q^sHC9nKi`GSX;q)@Xso<IQCG-N$A4-AS$U?U80C0y(zujDg)z*Cw(kU!zLTqc
zr_COk44G%3seDybHp3=kMlkM2#;8!Zwp9U`6H$^ytAZ<LFo?gSRUz=yhmt>`RWVS0
zoafvH!wV@HSLVqu{IJR$F#JI2{{+JiTs$0xzsZuAX}Pn{wq(D`J(lb#^>2zk1%L9r
z)#%1Cyher_NbVx+3-<t$H#q|U1L2<S4E-wiI73e<SnV1}`Buw<WdiiLdV{iburfRn
zR*t_fF#eliWpExPhKUasxqvn&s?L{Ps{Brcb`74iqt#K~7Bls9-zUcj-to|0nZ9@;
zo<!dbTE-$#=m~<<MaX$smWF4%_J0#epPlRjwr2;Bo$^mUDGYu(MGL#Phul+e9xxbb
z{u-QT&#EK#Gp^FfbvyMOv7j;Q$>-oW0ItmjU^^U#H>Z2*C{WC)g5l5)ROlVS4zDS1
zIz`Mtt^Zi%(h<aaW^DuF9Fgt(HkMRsH-bT<x6liz^czX2w%*T5<C@sWcYk#Iye+rF
z?05v#ko6mWrYFv`vGFwDaH0txI4$!B<s+cC&Q$XbZ<2wc{k_hv2zk&_S!2kDTb%V4
zRvlKIke4N1T}IB)ywUE?j*YGIy1P{)$YkZkMBThdM*xzQ6R~I1x11qAZJeRrX%EaT
zOddmxy9tbw=aSpjzGZsJkbkcLR#v_(XtbLrZwkg{8|Z#Ok!JbIb(+1<(@u@hR-C@I
zE}x&f=C%Cv8p+HJtj3Qn!xq0YdPfhTH0B<F;~EjOe$UdSUjfZtikY9akENll`eO%8
zxzrDeWBQsHn%?B49|q}wde}j{Ch#7mM`;BU%6_jNwVml_w{kQnEEzUn;5o#xHiozt
zCHZz6bpHJKdif8h#f4ax@w7A(12Z``m)O}f)PFhs^~;~1I3YS>WvpTdFJq<kNragc
z&Z2P^?P6QRM<X6m$?z8EQaG<R)XRi7D)d3DlrK*if}}Anje!q{r?oUsgoHiSg?i{b
zUa9h{n{jzIKrR4Zo3%hPX`BX{RS{%pzRxWM39r)P(-^H&|2SUfnL;m}DZdg9-TeGA
zynnqWoFbg(^a=0ybClaJ&kmNfN-52pG^70Fhs?u7r6fubvjlBfN0md}6mF&sB1&Tu
z4=<QBFpd=0hQmR|=OWuJJhaeE=@~SBZ6!`HW}0jS<4)u;QFkEf*z@TSb-LA`_@?Mj
zT5tM>{={CJ`V%O%xo4+4X{;y6KyWOL&42LfEQ)$nXr@_}CR4yKL>sLS_3<S2v61((
z)TB@h&BVwvX$D;+JrDLJ;2u8>tzwg&Q$S3AOy<h)Z3UEqIblYGfNv|G)XZo<juK~D
zo{mVcby3JHl*r&4i^3vwH$vyvXp$FIP$65DFL?MQ3{u*I6y+(KIgyOuDXC+|i+}gZ
zho*pfb}E&|>c$k=ywJj1b4)E!$+HT_)~ITN7DDQ0dedLIR(g;^>T5TqC@v|cURe-n
z-c8e)pd?!TxCGMpY|NsG+|~?%vhy%X<Z%j~lpA&{xR!<~n5hUU_)L*-Ie=HimJv!f
zmIFX&wBzXg%V9S);0)7J%<$Zy5`W2l#TX#QCCr-;nuM>H;pyX-r(b8RQC>0&YWn)~
zDZEe;n8d{i0x?1Z^4s61Z%>~_uoV1o#&HbO;UE6`?fW#ee5OdwbEKLZqjk?;j;v6V
z;ht$*WVmhByXcD~H;1e|YQ+q)SX|t3DJR~kfOq~E$>~}?H}Rfz%nLov0DrI?9XiPX
zdySMT!Xtb2@GV|r3R3@S2u|5CkG3Fon*~wP;k}l0DI^jxC;dZJ!lEC>XO%!+MW|iN
z=a*IGkTQ6QnFwV^!!BizK^cN`?jgmcX^30r#MD&<Da$y0LI7oWG+F01&`uWK#Drm3
z`jK8*8JalUwY18G*4Ckg(0|@x7+D;p$)IK9wh|g^ZY%X4-4-F(xGm~umMmN8zti0i
zWAvEyq>P7VQr(lQPC6-Rc{AiGgbBr&^t7H8^~!kLSgl&BERYbw7^vCZB8{(I0%aB*
z0wg)kB4_mNvw>uxMw~yAI^MrDvo%}7^L0xT<%s^yj*x;PNnAcEFMpjkNRE7rwuV_=
zvl&-{q1Woj46;_tI{gmbYk>R?-fNiQhW8pNeS`PXcF85BAn-n~BrxUX5@sYl)+-5z
zK4d1Ddz=!-_zGjo0n+d8lpbIcm5iz8JKMy5mbco(0p|zW#05O>w23_=s$acHdAW+!
zpa<Y6fGIPhHreghhkrYm!#M;NY8J-!;Z7{m)}mVzFceqMt2u8%1X+upsU_r|p&V6{
zzf|C@K|~Fy#AZkZtDzDF9xG;)$VA?G?h2WgN2Jju?k|S7F%C5-yDxQ3H5yXD+44s>
z=xYN0K$oq&XWrlR*h`7&t-S+fI=vO@bP%u=M+@kbL?u;At$*vqo)m3Y$VG-E-w@{#
z<eRZ|>3OatulTm$k$vn~`zI_j-hPpyYdc5s=pZkmspI!?Odg*i@y*(3aGB1IJb!HN
zdnzf%?%a2~yrhB-&J^YDA`8c47QOvYWWhKOx%|7-Sd;=L;2nb09`HNFsQm;tLezoG
z|Lgk7K=}KVSAY6R-cViX0p&P0a+w_TAFQ~dWa#sN>dH_?->jvy^_Mk|K(z0uw!A}y
zLxE?)%Xg0Ehgsej&5t-gE}CD$^S)?)fRrm7G5TBGV-7)A3cem0R&mFqRCmQy=-fuv
zCX~j!+XvV7>x3O&rq~++NRN5#f2PokgRrx~;2xV5fPW9DKmGEr@8ACP^>3KBPv8F?
zMxbXz!GL$u&-4fq-LLeB^^e?O7&wdJL<pj#QP|=b_kasIkF70Vr>C%S1yZ}V`A2B>
zOgSY)J3e}48@uLN+O~{JfLfaKq(9omj9Z;0(pFFjPXP*z4UafJj|z{G*5)~P#Zmv5
zp1w^_pMMHpy~Mwu+m@Rpwqm9RRzl0#+ytOQ?2f*MSmG^e;aYP`YH1YAIHPby3&$)6
z{8i?woTrwC<lKa?GK7K@eUI%cEEuRwOd(^#v=5d|F)N~i0&zBRxBmY6lLmgrVK2>h
z0(1cM$V!{(AdP3%xBx50T-q&W8Pq-Z&er2hXMYK@t?y6Bq~t2iytLj1poqe!sX;Lj
zduXWrw4X)5ERt}WLu1%gIU}^eSQZLrR`Rb*gfdK7-JWiV=SF1od^r$xxuwK!rp=FY
z##P3(oB0y9GP5R+k&;IrPUk*+H+JuKyo8-O9EM$G|9LxWl4!zF+Xqhlv9l)S(6yME
zcz@#=1UWnYB}L0-eKK-k13xQ_RIJ8o-Kjc5`$nSpF*puK&$QmXnVx&=xvO6jmB3Is
ztjHcmc|#lWiLNY3ec5EK@Td=boFs0Vx#k#F!o59@l4z&knV#CIo8xljaGC9UZY!!c
z%!3^ed)(eY($Sq+I|f@G)PPGk@2R8Zm4AJ;V^PtBVA|Ar{6e)wX;{)h+sK;4K|LF#
zS2|`4E}uUueP~CM-p@v?WgxYGuwVp>cpUQ)-^7+?247`7^HR+FvI$a^Vi{?O7oqQ^
zj8tXB&PA5TMXI43#iEMUD&KCL6(y(}(#06KZZmEb?gH84sWbv2JccH=9ueimf`1i<
zq~&;`Yn-dhQ;(+}ki1-7@BxM$oMCFY<qT==+bdTX%PfZC13mxqqRF<4K_Lh(N2EuM
zPE4rGOL|2NTv<!>cDtIps1S~0%vEjQYF_VlMhS-dV<{c2o=!u<YtZcsv)_4QSzttO
z^$Z#n{tw2%aD*^Xla)1L+$t+kOMgmH?NIL9+7{|N*p}R{%(c_DH1KD(g=$Kwp2Dy5
z$CKnU3KZE^qOdzN<BsE5^KFS`eKD9TndJrd3LhEqCF4^T>vLReX~$<YnROp%v*d6g
zv0rZlBq7!tY03JchKQWJ%1leBWe8)`jMu4UI1RKNmUv9p?F_%mJ(1aN)_-lYOoL&1
zCjHxny66q%SnqQF;#7l~8-~InGIB&_sLL-Tu1$uxn7Ec4H07=ToceB5ZaOserX81r
z0=B@Lt?m<~5c&3Pkps4mQAkQzc5I$7p`8n5pHRMW68S|n>7n)W%Z7dTOei~&u)f0*
zgplq*?U6o2>`!nUW0WC25r1nAbMXLAUlLuGjdYkXU}`%MHXc$;%<hbUT-wN_emMk|
zqRDsYY241F9H$W{-6$YfIijm(W~NuDwxFSP`RhWV^@QHlp{`pamoyc^5jb^2%;DOO
zYs&Jm^*R^yf=b?6DPVjkimzR@NH>2}xlgl(Ub(3ggh8Wp_s-S3H-Gguj-nVTJ%x}u
z*i}I*k@{ToT_d-Y`W86N{h*JH0-dSeQ*c_IuZSyF(bw6fB&$XAd}J$C92!@$q8HI=
z$VN{2T*`-C*SZQ4LNTvYyFF2=D~=yY<8{L%>yJ$cbWQylfyF1)J*9NAyrY8x!?*}{
z^VW;a!CZy9#IblnXMat((#YGq^FV2T3fhLWzs0>;t_FH60`8qF_dGYNPqy=v^E6`N
zT4s{R<}9f&xpGMlKrrvR3kCI>n;n9LdL3?lm9CV~OqESeWJ=g&+eK3sxf@Cv)G};~
z4y|t6lqC6KwwEtI+RhBri+xQAuw=#8(V;IIAD_!Q<?Xo-SAUx^Ff%{4Fa31&#!b2!
zpD;?(V`!sIcjq%+*6J3X=;|!IaaWMvMqy^3-yo>6fCRJxjPN@Fidj8f<jPXMP?uWy
zvB`6J+c&nD4Orvy6MVx}6HMP;{+q+)n-}=tc22ECIHG3;LF_w_yIr=hf&nJ$J>YaX
zua>I|Ro#_FPCZu|&DNF1>!L!1ZEOy0F|)lKUZ$xRc;|K@Cgms))s(1L0#@r8hI9R#
zckoIe^0@E^KtqDjBpUw(4#N7E@l`Mr0y#97fnpgHmuk%}D3_p)Ed-aR<}{Ok2kEQh
z;jfS1UnDUqMRe(Bdh!3>KVJTQ;cPq5Lh0S#KVDM1HA<MEcN#xN*!}!x_wD5mPP9!W
zafpEQ2Y>zizGE$4NpDf2-lTH?gWi-xW!jx{QJD0p=*^;50A&Y-Px|c#`97b5TBK$i
zGOjo^gA^=k(7qDI_o7{SA`KINbzu|F^-mUMXhBihOx_CTXEQ0JsJ;xmwh_S#Yi5$7
z>9eF3`LSqqdn#7W{JxhqC$HvD8!NQYd^RnYc}tc?hjtKzLGv=bC7mVRCEgN08a)z#
zc=Yrw6kUQz)}nEe-d_&Mu~U~cEUTZHWQ}As6?R8+9X>gkLqykBQp}QnXI=_br*&3Q
zRZSK&g{9w<2ycmp8mNOyW@8<cOFiGRd)9Hul(CM1K*~D~eaO_svCBKjMq}#CX^d;L
zFY2d#9r^;%&hF%gK59uC`=#U5b>`Fk2moO~p1*jlawPt$EY^|*Gz82T=!W<N1E#p=
zto3UP-kDvD2N9Lwc1fwQa%4{(e-Fn=i<m{GKyoE;=FV80F#BV5MwV5U9mWQXb>lf>
zHO{ZiwaBor&hTVwqX;Pd*<_m&=}PLOL<%tqr&<WZpq1;>oJs3~<5btuh?S@);Jq(f
z@P_moMAWmHa1mWjy+$Y>XMA1s8163E@Rn!bi4)yiVDfyX#LfxjW;3lxf9!EQO1i{X
z%}24z$Vmg17N0o%&JWQ4q$T-Q<!kZuAQ5!{fzL(}b%-MQJCkr>Ifd%?E>>`Yu9I0A
zT5@h$hABotYUHzJU7ct~TNI$MO>KiK###}q=T!4rhgz0LdQPLy)kyaq`Hyv!JTCV_
zzwx%=s+DCn|JL`Xi|VAqf23>Kf*{n`d5+cYHlB*Ouvo*ny3cl6gYcT`x0()EY9ZsS
z0Ts0U8K43>2=EhPR1{&axCq;`;Glesw3I_@DT_AgYd&vU?*gATgHl_vAslZ3fy1JE
z37w(sbIz_pW3~a<m#Gu^JZ(!`C|UFj`rt&)ru-p`4q6gxh_;Pde@Yq&PQ{u2a{rWM
zl9>9~`q=zk7noc7)zmqFu~=ausK-GkX+)5bD{U226J)y1q&eMRu<Au9g<!{8%^6L_
zuX8#_<-yo}uW+}YOjZ$dW}47ynmwUsyNsS=ZEcPddLtlKAF?KSpN&CiZD%N|IY<g6
z*CX%z-{b-`XVFaIe@t~wV4;TX@nKG&q83rkc)#QX*}X%z-s>~Vgf337EyOQ5gV9mM
zs5!&Q89)Zw2`NQbjiILKwTLZ7an(gy3-qINCh*;8&*=`aJQ-_Ijr|;OY?La`Q?_}S
zp{tVyC$!rf?Ss^-M&4w(>}m)u7zejjMAiUSM7$s}Ap@9ue+wckwGdoQ2&kqKHNoGE
zd^Y;@4Mhs$U=0q@{t)Z^gi@kH7dF!x&Q4M4!;G)z0BsE*K+br(7pzfb!Kh9^dl5QY
z(8zZKK+o_HFalR7KGT|qIE4w`&w8(CrmOjh3X$K8pS2>Ba5d|F-GZ{F*Rs$ImG?ct
z+p4I$sz5YAf6iCT0C!e_N$HsW7}YhWv`Tp6o?(32y<sz~89&vok$2v<#!!v4ZMV4F
z_1(N@e196M34>|-PD|4+0;8JK(8U4~i?QgPSp))7<rIC^CtC4Gs?MqhGDF=9lJr(}
zXPRoY^98C2MnIT$(n&M<seAxIBH#0pzXsMb5I?1`f7?<k8V5oJH&m+m_~c-qf;bme
zVI%Fn2Ym_VdATP*YOBr?<+z``VO?gTk$bl)#nQPzq0z(E(o$h-o_SVXaE(<c>%wyZ
zsBvCsuejcuuKYImAFMDir&4P>|1f^7)Iw^_XKEV-+xWE#!7f4*MkSXk-VfzUvsf)B
z++gQEe*!f{VOP~oJ=+P*;5}@!VX1yag$Qk$pu=I7r(zUpz(d0?^{8h&#ENxFayLt3
zP90+^jHX*UG;^KX+HNtQ9f3`O(^Ia&3VL=_*5pcsc{}B4$7?HxbH7L~T0mze!lFmV
zz>goTgKBuPiFIy29yexw$bl-iNqdm*1OTP`e~ywO?HwC6mc&Zy<Zi9sCOKR0PEUOH
zVwb`JtaIRteTm+2F6X#J_2bAB9+ot)48@q{b~1~RQ})Ob^3|I?V^h)rZH4uYbMb^E
z+5A#tO=2})Osg!{%Bkz}Q)^Z?p}TD=&?*WUW{hoHac|$0h+behG`kX`EH?XJawi78
ze;(^}&+Dp{WBN_^V{8-%+-H<uCs08LoL3bNbO@SG<h+~C?Y^wvg`s;oY{ZS1LT={Q
z2NBdY><$ASm<e)fd>@yIQeSD8BVy1{cUW83uLQaC8At2@Unm93+-IguT(zi7&I34n
zO&Q^J0`4^ko9sQXy+FKL!Geg!48_p$e}QMZ(-fz(HY}O2Fl;N{Tel?;YRNdwPhN64
zFju3p@{1W+s%cjU_;&v+8n^i7YQ%yU4ih(<RarK}9dcFeDkIPrTTTe$3UkFwoO0S#
zQYJhhk8#TLJjSi3!f^0(ZV&v1eLP0}TuPz6JT~GNXF%nm(E5rDI1YFmHkhuGf4J^$
z++s#l5QKL6f6a)>dSRq_Qbx?mwsu2CR2tAM@IPlng_y`{cST0Ll<gQr^ZvL<rBfC*
z!|D!q*$-#Ew^o1l0XnXhmPT0TFJ|YPF2WAZY%my%ZZ&ADo#nKp1mmUB3l-Y5r%-E4
zqUh+Hz|{+-xu5CuxUwU0egzpSGxz@PMrjG%O^~;KCt<iCz-IMV)WS=96C1AZc01?Y
zRbIco-?6r#$$YuKvB>!k?zr4Em+`bT69P0dm+@W=69PCfmw|K$CYP-5G@5@o2Mou*
ze);o(XF@-IIgErEA(?a2*Pn;qCxm;MBJhDT>7AbbJ%wSMgRsU}d`bo2r+*^IFaP-d
z?N48S1%?y;6=v{BF`>b`>1X<?HBR$FPCxQ@#{^jICSV@PRg>Wi1RwxSECXMsgA$g3
z)UFx*5k@^zP6^RWmcXR52v~m>XoiS}fU%6ydOzAhqN*+tc`9UtZvji58@}Rvnial=
zPy3xfan^6s;oEfhROIO;{==uS+$@O`m;#Fm(}<LYq-EZEWk?*}a;;ZJTSm*EV8$7R
zf)kEe4)|ADtuRX?4Z}2qxCjFc(Q_PM5x~G}5=l}^&9ujPhFKGo60(0(lVt1PUw`Il
z-*M7Q^Gs+SP%oI$W_m#RGTaSB6|EF?W64T0bOz2wDo;nL@u|?jG#Z-*{QI1rr)XHT
z?vjD^IFW!)Q$t_k^Pw>MX|rt5lW46r2#1JT5Q@}H$tsl<-$fiKix%RG?@~t{Je2nM
z5lJM<qVEwwSXv)N@K%2cJ(WgJrCl70K5Y;n&M;BD<*(=b-h{bj)K{!wJ<ZEnh+jR0
zwF<0GTa{=EtF&(;wc+iDM4#%`(Wkt6#v0dh9VI4@qF>hMMPBm!<hA`{^4Ni&D33s$
zDeLi!zvotW{HQ$g-0>0QDq5<d6ckZ2!e$vIg&p=bYeL@J!Xba;nXdjFD3@SKj-Db_
zy9;?_MF%YDfJ+*au#LN(*SN47X_Y9O=Y~$*@5mZ;YdVtQ7j-1==vpKlM7d(=)ohcg
z|9$Ggay)*h0eM4OVYp#Zi(;BSd8{ZbOeF+BHdOLqk?BsUlHmC>LQ6Ra48!YVR^M~o
z%81M6v8XHrnYMp%HjgKhDeLK@4|D}1kuC^5q%lFE^1O4i-MZjj*|x-E+@?nD3XP5^
zf=2oDsjRhGHmaknU>uCNk;BL}Z_TN!-<v1Lc<Ee7JVFBDq?RE!SQ!2Mg1X^gNjH=S
z=DZknLqevMlaonYOOafxyef&#0Iq4K_6k!-TO12HC%b<epCB{?aN|m%b0^#R6XZIg
zW*<qZZeU#qgRI*{)@}aK;+ni!RkL6YN(2)W<nK=Tt*_e^lu<(O*(t)%n{6u==p0Ql
zR>*7H@}6Rgr!}IBk&QT2ips6euko&<IU4}^eDyJ>SQT%>Roq5h=xAaS@IIQ6(mBDD
zDk@JZKs0|+bZBCTX<Q8qls6ov_}A`{6qOTwZlg`uwIokR*$Z)PM#o9};D}lW@<~<A
zgAdtUya~z|9xuZ;j6}cB1cAj1vskBL6bUa<jUy&d4&{Tk_P6QV;nRpVt2btlA&t~A
zm*mMVyOGbi;`JP@fE$AqMQfITri+0N*})lQFD`$}VX&`g@N>+{Gf;EPr>b@QDa%U*
zTIs*YLO1gHr3v3CGBjk3*fq)~V}%b@`a>QK5*;g&ILwea+y#09mtrws;n>Gi!w17a
z%eb3f>6muv7JB;`uMY@?{@LCmjYeo%iKBg(crFn=+EbISx$q_i(jt{Ck?Tshs(~aS
zFwB1#Q@nv5Gt!ynZP|xa?bOIKMyz4g;RNPYe35jy#<IP}7uSc@8&(=ID<_%tZ(y<m
z;(aVuS%FCc++YF2jCZzJWgtFpptql~Pihl3^H@Ulx)CD==0X0KjaW&~yjC|D@ihY-
zFy17*up)9Ukx#V)k0fc~nD_dsK`%w?Yk7aT49O_e?D(AOHEEhUF_qyZ^$Bkk*}tHX
z@xUdIbjKSlTug!DH#*YDScfBx#Tk?WnX08RQCgwmil=I@X1Nk3!f#uN#G{Rcm0|@b
zQkq+|<<uNH!~==KXM2<~OGGo4oN~ql*_vhE){ZL@g>;<aaw2aAi7utdm<lp!iW`62
zQmy7>wsG8jlHaAfq%-LtXa5QpOJ~MGjC_cT60g!0x5TxTQqN=Rhjutha{nq9YaiS*
zbuX{7YWNnuq(h44lHFXw-dV1I^UNACD_0yo#k*81C?^(feD+U$M$GfvXAF4~XDU+y
zZ!}F&9kjY510h;^%$G8d94v<!=plb%)*;Tf(_nD~%!~$DUf(KaEG(_x%c5pQ>SG!k
z(;1R#sD8LwR_Qw=s+jX5(u)F>;YxL!@fv{|!@$fKTNP*)e-FKn&s-NeKDDZu&2lWe
znRAk5kNuMLO}Lh|<BY8Zz)FT3e6x0@$&mNTdL@>4M%Z^QU58ckA=|1fd+2{xxrS`)
z!}@EQjM=i#BXsXrm)vH`vr?rpMzr)s?5Cr`8fLBC$dkZG$K^}{YNK`Tp?O)S1=b73
zxHmXl5LkR${RR<sKzu#G3e;fh#Tx|J0rQ3cJIvT2z>YLr4zLUB-x*-hd15vz_YJTE
z(hUK2n6X2E9cj26U<c^i6ZL;}0ZWfh^+v&Tz`h}v4l`aEOhq^Z0X+oMnGC=(uT6Wy
z)0Q`PC|HkB@XN~8L+_znJ;rOw)kFWU#XRE}em3$H+_n}o3A|^_Bc<Yod2e?-V;nH<
z;njPL*SvZUEuPzRv~)(>whY~I2p`U1{G6<1^9frl(6&cG-*3GlHZy;^jbaf}F(GFg
z9q(@2dv?8RDh1wPQs?eE)?Mn<RZI%?Hl4b_C9~n?{0DtU(lN<arJXfU;XOL6(hW)j
zU8`;x?OJu~<l1aoR%pjE6@Q#wU8jR8DA_mQE|317_UeX$L8@JvbwOhepxwr9U4q`V
zUAG}&sf({NdsD-%;qHG`4ZBvO5{ixex~LSWDz5hHj(w6%CuZBq=WEgs8r?#A06N|H
zQL~G~Az|#$ITJ4pY<y+fwj0s*<i*$HrAkJYAlGkHGU993JOtL<^Uwm#MGrxpOH^O&
zq*K5E<S5W>iEhTcyj99Obmn%%teT8tP~Ckx=CXp|)9PBHtX+S5$A+ivB}{--`q*8M
zHRzPsn%r%?f-&)8b6}=iBr1_h`-r>kGmF2bS?r#Rb1q(TsVj}GgrmJ`ER>7}ktk7Z
z&)2!{xsLs2-jsf`g6(owH$W)rYn~+regmD3KMsB4bP?&hi^$y4o8YM{J8~Rg9XvLw
zy3~YAvbBD^-gJM|KuZ^n<OX6tU80Y!6Eg>~^Ii{~d=A=++eBx#4qq*Ch__N6nF|<B
zO``zm?5X92v^z<s?;kIhqD79)N!OZ=hfW`ZN!+lX{B&iOjYp587T8u6E1Ki6q^}K>
z^SgS{-PMc7zSgYGpDk=rGFwL1o9rL7wvSDPQZRL{!}ot?(%<#Bo#;yxcP*|)T!>iG
z!0b+5Zt6D>HNuRNs2r`{?9W`t3f<9Vs5P3C;Vf>mL$z_XI{<a`OOoS5j_`T&+h1Pf
z0N?&ZAi)+Nmb-u(KNbaaCP=c=5mWUP!dz}O$X=0~_J;$;F48r&r{vLvriqb{(H9qp
z3(+A%)E0k;-CFv?rJeD^w!LG&1L;l9Lz`Gzb`RzaA6;j(>3_#VIb(j?^{}U3_w3j%
zf~BU0fq3bvB+5^aY_si6_EY+ip4z-D2W1)61DZyf`20|`Zd2ohbny_Zr}6raj%s1B
z@|&Yvd9=SdDkl&5`5(%lHscCqZe(+Ga%Ev{3T2o5v@{2oV0l0l0Wz1-o-HVs_xd!K
ze;<6O?D>zkU!G(U^wTSZTT(i21=1%ek-Y@uWRasq&h{0hvr#(WZ&qb4U+@zpwkQjv
zvHTtsJ}${{&q5ABI9n}b{8_mrveL-Z$d^KVDU8?1JidvhiL4Vllo-~+9$DK9%}7FB
zqZ)BUk1CLIdG3YgMk;d(X(YaPWggT~e>m$TQyjH<mMt^Eb81;Ll0VV5Bsb?{?M6>U
z^di5!j&zP=GU)<k3!}5>wQ@F1#zbo&Ym%JVWrT(h^hpYfPcUFkKhTM`qrXMB3Tn2s
z7J@L~$F1$12cNj7>)@z!XR*Nv92p44Gh&^I=L(8JNTEwrs%(4pQk58PywDoqe}W^*
z$*`}**qY8}GPTU6jBgVMNe)xCInAt2vlbbYe`%*aZAzWT=ohYubKIg!eN)}7(5@Tb
z&Dgk7)*!h>=I4rIrrDftmNGDBvq71wEpykImbZAS9Xo3f9=(U+A?e^<Ir8;@0gcw3
zD>g{lESe68_mjAE)q8QNu81=hfBM!a9nwC_wH9fc{5(`tF4ZqDEI`>3csic9k2lSh
z*K%|;St-#7=j3BR`!V=ww%R68(F~?(4WN5EE$n=RQQKxywsPtT$_T%Ndg2b)T?fs$
z<Zp53bTCY=YVSw|$O3hK@wh^ZX_u*3oB1azv{g&-N>wBsa-Cc&McEQWf0j8>1x}6c
zZ{2fQDEdM@<V43-=o?ZW^q``R>%E|!l8$GT8P++^`hF5~qo?Vx(2S5cPdz{?5p0bG
zaA_9Ev8kTS8oTHm-H&@E{z$R7m|Nn<*Wufjx2JzLLRrwlN^|=2`%`>jjS>bEpE6$9
zU{62(I(>WkbtGG|mlP7$e+Y<w@V}HdyOG}tW?UpDIq5K+b%;n|Y6W`&zX+|{$++oz
zL;6e!c}v+YLfH-Fc|wj6rr$4(rorqBE(q<G>FNxmvc+xGmi{CobR%elja)rt^*KJB
zW_1O4;V^#;E4GL=W8<XNm+F^Ot&m~^=yRujzhYp7!Qj>tzonN`e`1k-SYSGr--kJS
zs#%9POz{vqL7N}et*3rOxqd~`&N0|ig14@zAiJa`q`FB7d2ZI}oOAKMQRs9TK=CyR
zA=kO-v=!Ev_ns7MMOaRNsm*%3e^9o<m4#9ZQ0~#r=y3$84%C&JZx+^q&Z~_q$4iij
zVs=piUHh>+pLTo}fBO?cdW^Gck5<rnRA9$;+GCmG{W$wJGO0Fuk6cDA!f3UTOHSo7
z+6eG4&VG=MOpN04sEtevAK*4NQe}Ii44UJ?`>r6@-4KY=1T>UcPWzEWy_=<pvRi~R
zEgM3S$y5EZ$<9Z(j?pTX(yl=Pf&qRYJVPV*nm%Y{e!%Cse}x)9Ja&<q0|>d-=&=}3
zRE@i4Y_<$&&+duSD)Cma&V&njoYlU(lIbeldohi1TD$t!7TydyXFbL$py|Z8?@HZw
z2NLp6R6mMojg=|y4wx2?Ef!n<3#2s&JVDx@0BJ|cJ4THu;Q@dZR|~cHk?RfYqh%0Y
zx^cxp!1|0ef70WCfDTGXM3?W7KH$uakq^VtIkqy0-NC}gw(Mt^Md|4DgIIehUU{Jv
zY?L?(j~<tR43-KI{7+6m1~I$S{~-e{!fK5NV3HAxjNq`9ef8=C9}JVRuTqXmMJQ`&
zWCsQu{m{3{2`l7YSZ;f>R=oSrErkvScOtY~ilM)oe>)<VB3W&#*Ed(z1H2a5+4q`d
zYELWwmn3HHLw!2ZxdYL!xnDyxiteWMc!+DYRPLKy%IvzfcBNaNhqQVv`6<6)Jx97;
z&@VS_&8734U(L*Eq`SVzD=0cFeI&q2wC^-A4za_kURW1@rXBU#W0d(a$T#a+X|IG*
ztupitf5#+UVTvw%k625G{Ml+lUGSFmCEJe~ZXoN@&N^0Aa6ib@*ku)^UE46x28Gcm
z#nsK}Qemg0Z78XBrZc9yjpSTUNfDUD(18(l@y<-oW4y1Cf#0jrDZddD?sRdF#sUqP
z0`zVyXx+*%g*Rju&+?t6^F<z8k>m!ow)5KYf9r3Oc00V4`+Zw*eqCQS@TgqmGjt8>
z=~jn%J(T**wwo-z&o^wX7OZone&0UBw7pQhAa?j=>+nC+Hx$;Ji*y^`2xAv}bB|an
z1;5cC`$W~^!2Zqaq#17(Wm>2GPQQY{;(O~5Wf6a3=|9vV5zj1es16~t(x3~C9IHbV
zf0omFX&uqe3`!J5_0jr6V)+jvMYw*r2|iHL@@4~kn9UOAN85t~#XIf6(UQCr)(2W%
z^LkA29*qWqrnKYT2uJX+2oK9BlytFD(0g$sP#wWzY@UDbF0stI4j>r!rTs|~cOw&c
zt*jTu=<POU7;NsCks@u)CD&|P^v-I>e>)KUPCQeHc5dKaYbp1vgh=Wr-3D7Y60g#l
zo^HgKj_{=)@TFwPBgQamZbVlVksWk-vU0eMV5>H=;+m3GiD$)m9Z!<H&bpgl3Z8}f
z109AebN1TWnk0l-&nasN-JWnX=raZdnK0;bHnw568S=$c%wSKgYF&dcq+I>Me*r8x
zu{`RRMG7oN{4j{Jd!1nW-ZAx<_3mDjOQb`;Nz#DnOHR!a_XH&6R8Z`fHz4^Qr}L4%
zb+uVw**?)}NUHdi<Xq&HILVhDY1IxQRIa|Akm^66dRgxU@3pO|?SPHOvzAEx#RK_X
zI&G%^RnpA&f=55`JlixNoeT*%H=9RSugoi)Vn6<xmsCTE%ZjnUf=XWkO`%<P5`5jp
z0{FQ+vu+WD&9>#nZ{MAxPXt;20)%y$m*Igh69YLkHJ2fp7ZU+Em$7~qCzo)c0t}ZC
z1T~ZaAGf9iH7fysTKFe2l=>&m*R`}(z#c-oS(5AF`Hq2VTmj7i-E8rh;%C@dVNf#q
z?5n!4e^ENCkxTue+OWzTJE|vqZShX0<Agz+QDX5a$827Yc|KoNHRy{38zdXJ!iT$P
zjE-B`1T=6=VS~^n^-rzQszYB_TEdL7&J?NG+#Za?l%BSKr4%F;z4KF+m~H#1V>btL
zC%DyeYcbulqO>1nRjYoeIw5$@ntNfS(+(-N!8?XJ?LGMrbHh+KQoKIY4V2D=x}mqe
z+@FFmGZ-8xktz2kc1h4oVQ`>$0|%s?qgA^<i~}N8PdFf&cvfDvKsEB>sv9V=^O1~F
zxrNy_CYzprs;tQ>@!h(VxJJW0&lxv4p0L7mPE<>rg}0~iz)#_hdiv_JFuJec+!Q;v
z>PGo#c9Us_DA^Kv9XTM{4SG4r+PG-_rTBq`Jdmu&5fIoV4{ARvi5)O-a8T<TB`w8l
zv0yAvh|2z)1;YX}IofGcg~TFLWg{OmyHC6gCF;a~MRY>0n|jQPd(59-I|?TE7fn>_
z)h9*3u<C^xq9Es4V)>?#<k$@Is2G&)jdp=%$Pg8(u(7le5_v05009<sB&QmYLJ=wK
zwhcIM8!Z4y;KBCu$Z8RhH=E}*dE*i|z!<|e1?Kh&>`I8BKv4*ABSi3$(6;|wAcFFC
z)@Uz(fCzg+<T)5I%sON>*T9K^qIf<DCx+f{fD=3C%}`-$-pe6^a%R@1hX`^VB7{I5
zfB?v!0}+I<<Nlpj&9i!GNMg4v(hLNz2d+J#bBeY;>;4>=D>K8arp15jY4FC@T%l^3
z+>wV))$9p0s(&IH{U4a|k7-U6S*(?}w5Nf8;xz`PC=*@~g15(lp5h`QyTJMk%wE`$
z;zj0{6{x^;e7ivs3Ur|UZ;<b1kk_Qa1ne3QzB3E<L)wCEqHv?!R#Ip|UDOF<`lC<I
zuJ_>c)T^XMrleN;N5=&0iPNsqYL=x-F7Kszv~Nn2F83-aA2MJ1A!PxSC*XbU#f8d$
zjxW!7w`%i-G+SGWOdLX9oydyjTFKSXMM97>7RPC2yNWHf<y1`z%{h2UEOVi=$;E3L
zs!u1GB%m3+KbK@;+R+`^QII+s=p@xwGcZX)t+n!)uD;=T{<zvE-^8*bv4qVfXA_>Z
z86Q>&uW0!GwUx~8?X{I~;uA_3gZ$-xDv!y*1rFs|E!kB-QIZg=vE%IJu;}gUQ;H}H
zzju9?9UER$EYi}<R^!H67xqAihs?eJ6s{21n@I*x&miSE)8@`!RaJ`>wF?D>Qc|NG
z8Ow*K3qc`Lu2K>O*|<e?;=T!X*;N&XYq>{TaYF~qc~GWc21mflu5+kDYs<KQBi(ZW
zC{xSfD^o1IKvR5;+YnatQu>_;P)(1GQ>i@jj?!?Z<(jHuf)6({S?Oj8a<Q|nQMT76
zL_f8GX=?{^|5%aTXRhdGKsRy`tCSEV7x9~2*Ga<6O4oxU-mjMjsqbrWkU}#&5htOK
zFm2IMnAHHDfrfCsyI|TaTu2mu_$(HdC#0pEFU)zq&Vi~pzc@A!M3_~MLRgKXxLhhq
zfR@YRYWiNo4_jniZySsQhR^&STUqMHrlV~fOF!E{8&`wx;LXOpNAD<~EBX>8f~R30
zTbCN)cpA~4O_yTx4*uFQxo_zWXV~le>qT>A<<cTEQgjK)>riAif?Dx^cPTQX)w6TA
ztmcgrAFnRw4V2EDLl3=gaOG;Cc;?F0IR9BkeEEGTz@P*B*g~)~!eTql`~7;gEBLG{
zZX}^42xskrPknXuh_8MP6gLXI=;dtvQnh`)WA$`(WgjUW_vMVh8%gNEiI6^dW8Yjm
zLM>@~`P{&VoqEsf1%ZQqx35V_a!7u$rgy4|-DXYU99tJ9I5tY~x}z9HfnW>+K<LTZ
zcpd52i8_9qVs9cKJt^>irqB!l6q=9+2EY&C1Gk3{|N8dzPhb89GY$B67(r2la3pv)
z{YYP}pB*AG{5^elIFtC*6w0U4sL82WSVhlyTxs|+9TZwhe+ZU;*4FQ#)z8W)#Aa{|
zT{d*g^lMAuAP_9AJl*%Uk(sJ9#1$1T!nZ&rj}2dGJdX-r!>j$yU1`)mro-3i@To}C
zOZrFe#!_m(B21ttU?sEy<)%=4Cd}(t8s3sdZnU-wEsZ!EXOyI+x!IC-%Bjp&7RxQo
zg1L!kF$Heo_pyC{Ljc2ClSpzvowcVBPap&$G=VJEB-;A<%a1(n8;$zdyeB$$fHDI7
zJ0OotyBTOjNU~9l&E2^-2hzu!4`i#4*4^h*JoB1~4Md5BPgBEUTI*e1?5EuV0+==J
z<aq(L#~XXoz7-)#HA^%_h+FlFyk1%n`1eW+33$X?aswBC1;TT^Q^+ClQ(fXGOX8<a
zCRknFH+?&Gm8bStyPy-F?yN6Tl55J9+<XfsjW;#Dj4$M5P2E?n$|Zvvlmkqt;OWJ_
zUmPxX?QhXC?PiqbXV%{o&KmVN(`|mxaACnMBt35ABwge$I+Jx$H#hwHaujnAxC&(w
z06?yjgad<rR0@|(Wy<vB!-&x1q~(ahmT{t~z2!=jR&IGaE*(B?s1m3U8W$qJnG1<O
zx`;n=ddUE7sP3|MDMPRyHURvlZc-kq6)r|+R*HAA;(oAA%Vo~OGVuolPNiR1xh<ta
zzO?U}Yw+o5TPiv>CrwuL7x7PR26}>q&W@|Op+ARz`<n5ADoFgf`+dqQWA&Yz5DyfB
z4h@?y8bm3T7qM_1Cw&_OMr<v5>}x=}NvS`pT@JMG=xQ)r%T^;Nr*6y+Re&`FRPn<a
z0;*^xA)tRweYc#d+%>rhV>w3@tdVO6*q#!1W80JU;+n8RxlX*SnTr$m8=a|JGdo*<
zo0<K8u1Ziuj2E`~NjcE*OE(_0SEaz-{JKW<K*01z-3s;KQd9>G_=d8sw1E#;yBV{+
z<k^}Er*>=W7;j0d0<zQ&;8S`%ofWqd;YV^fh+F8>gGdVz)i6lyP@Am(n}jvta!SK5
z<3>}Xb)&}k2VM61#0gN<o0VPI{&B!swLFx6ouEfqA%e;B>w|bJL)7VV5xu!0AN$tZ
z4eV-jOacZE<t9ct_1De42kB4T63D!i8znWi7S2o^$wzVn&2n*Hj5F~M)=u!<$z#D;
zFXU$P8dMhOMoe$=ef67RbD%~-B>gtNLtN41?5Y32$a+}mD_J$H{1?!WuoMbqZe(+p
z_Ovt(0W_B}+yfMso^1;$mx~WIn}5$KQ91nm^Ivz87#+EC9t6^>Tp>XTI0|qyAa~Xu
z9<-PH@&q`Kjrbi>YvRZvCl>o!&07F@wDzDJQMAaHc5z?&Ci=P8aW>{Sqw8okIcts7
zYKcqy4e9r}%qh><Gh}Yh)Z_VJV4rCuDSS8wwWcE@K?J22(m)PfmZp5Jzkk-+6cF`w
z3ozh?v2LWE^a?VNUd#Gv<zY>lo=dan#3jk&C(w$09harjsmM8%tld!i%Y|M#Rof{#
zW&?_7oa*nq{)KHhEo9J5noP^4tcKRwcqH1n=ZV(mT5seDV)c$WwbTNUcvdu$sN}U^
zn|gZH2DFYsyOC*S-=uPFy?=1pEU2k<K>-K_vGy{L2ODU?+0>tBqORf?Txk?b7cafh
zWEM}Pd=gO+aJb<7Wv*&YBxpy1mi0~H(^hjNbIS&<fr4FAE~%p`I6E=9t9_z=Gu5?u
zX<ezeH4Dz6ODT_U&#jj|=xoZ_^qH}EX4Yv9rzT@xLK}^LYHpP2;D4JEsEbym&^Dm1
zr2zZ{W6ksj2Hj=rr<aRH*epUjp;ZeNCNQTsGrga>rWMIFYh8EQuFL$81#zC`hKt6m
z0OMB}nQ9S%j47EZ^VUaZh}<Y!enTcUd2Z`<;wrCZGgB&%Vw(cz22jPus9ZJ78IO7%
zi7BE;5nU2t`g$3jo_{~@eqZ>z4Ne&6rY}Es=>-@i44R0>x6w|2{A2pM`#4f19Z4Z^
zNcm@FF|b+7Cjui;qAI0x*3Rrpp$%G?oHPOhSySZ`OCpq5wHk~gTkMh|NQ5<Zq=@dB
zA&Nech#>J-|7EskEuR=Pu|}8}Mv6+U1DO<Q(W%DyuX@bwe1AnNBAAVQ;o60Yej(}2
z9$NU8$DH(T<yAX`8fw8S=C4&=y%tvM1+QrG^3-Bmv9E4+vlc#gumV4exn0Ym#|t^t
z=J^)wy}x>;e+>@=(`t6BIh5cI^@UmAtuW8eq=)M^Q~#G`GHc<sLw;wY77lHQNw@_v
z*I;U&VLb%gFMrQi>je04bDp^#Ha8`f<RjjXT%?e$_dRu34uI7x$t*!&>I)RjPQ7HP
zB|<?5w}<~y_H>-HOivx2`k#TPw<i<j5Nd{Bm`toO$cedQG8rh|A(;#muOyRk{`X5J
z+xV)<WR&zfCJ?JqfFCBjls{|~-o^!cZ_FQ*jmEiKyni-!j14_5lpJxfYP&-O*IGmg
zjjLm#TJw-v<gIy*>V{n6sJXJCp--$rNh>`PRp^p^HCKE%BW@4EMxO+^jQ<b}R-UeH
zO3@sty2?38civj{30tBWxx}?iF+iN+ZJQn_$a}s+&`Zh|Y7K>3R_@6jsw*4edSRQ{
z4;>!#B7eG<eNa99aaKpDML*aY>g8HhF{@ZQfZ>HHlfb+%t$QbCUg^BfZ<TFUbX-(t
zjw_SKT8iv#>!3xfY-q13+5`k8eTyR=EbNDkh$;v0+YRiVU<fg}05G?Lp$rH3Mh0`}
zvSn}+7#g_P4u-n@Rc$ULI{=kx1xxN5&Ffm=mVewfMv_h5%GY}VBx8ec7-R;31W^R6
zkp2vSWFm5+!yN(AK=BR$X`pxoAdU0CA3)m1R|QBz@9VBf0O|2<NX1%gS?Ko`q>`M2
zD$WbeL6n{ePn~=9+i=EHt7{VtxscLZaE=KMy(jX0a3%ZxD+G3|Ks8laky%{_ex8e+
zwtxNIW3kzs%zei?{m6sW$FA-ur$erFou)%D3~Q>2;ncIcz81xv=hLfLK%Gmvl$m4=
zK5vq!6-vg*c(iCW1CI6;;0S*jaD2m35Vl3D{!c(KW`!{)=zu8Rq?GXcg#BKo+!@DX
zD74dmr`#-*wh)n;Q_8zU)!<)}(C)*(zkhxG%a`Ajz+wMMT`{j<m{1r<f0RdyMT`T0
z={r56#FX}7QdmBq{3eV3r5xxsur+*{c7Wx(!o<Z|zo*tHAV8{(+TfhceCk*yU`>^W
zM6tEj`M&eU%2l5tuc@(-o)Q%rn;vm|9hDx_hd<Y@IO?C%?(4MsNK&;<`wYuuA%7el
z{;f$>2~gfBF4@{-oPZduuTqH~&F4ZZUvF#F!iBUE2RaCkrevlzTU&E&Ve76<NUKt)
zP4#<hzaWQUt=Q6YEY@o{&IpGY4NX*$bc=-h?=L@S+&3Kc!oCZg2c?Yw>!t^z(-u*t
zmC?!ts7~E$@0#1|=7Mifw$8NXV1EM<obuzuSSo9K=!^TfSs<90q%j)=@|KO#rO<T5
z5aCP{+J)Mou2&}ayL}@yumpF!h{iY74_XPY;>gLC`T@MDo8sxn)&n~$YpEFKu%z^=
zTcCvk-}EFZHPdpa;KmxFz#6LzKR2xVSjRS(V_UGX0n7S!G52_fgm^Rziht2QLOm!a
zp5eaLC0R(dr(aZEDkHR4BNqcuTybRP25XMoow7Lczw-aLc^>8Fhn}2YbG(|#6BLQv
z+}p(!rfN^;ZC8zPlPFa{92b`}vx8{vXko*tMSj1yQL$dIZ*!@wrxNS(x-c`mcd4#5
z;!E(q@GL`v&Qsp6R39kbp?_2#C|)Vmhu%NuAS>na?WLtgwSw+3$Zp-pMrf+FbZcQi
zg@}lv?IicRlc}0JJS4{V<&tEpKf~<C&z9Y4<<d8BRldENoo#UkWsZuUFs@v4d-c8<
zM(Zfm$f2}XMtNbC9!J^143hBb)^WM;gt9$qcInGgr^Sx)>(fV{H-F}CW7<d~d@u{9
zO~*Rzs$1KrxNwJ%@3C99jxgIxzic!rU(3C4t4{$I8~ws>yW~mt6u>yT)7IdKuW#y|
zyN0`uwRHw_N=x9-=-^CNP<m_k?D)Vpx^)?m-KHbUozuv(6tNxodj?Y2!9-b46>xd_
z`Td^5gAYl@nl5tDEPsO(MssXytq{=l3&P;szGvCigWZ=Txw`HJ^*pTAKZB-?$FkD#
zOVn0(NtfFlvs;YpBK2aET=tYxJ+V4h+gH<Fs|*zcW~`_7rmnOfH4#o3580aAp86GE
zc2ot6Ip?=4mp2}m<8Un(n*NFw27F>?^8|J%HIJ20E`8gT6@PNRngky&9I4rka4z|`
zMsUH7i$?zjO<^hX^hmaqTW5!Bez?2Z#}0>|G`3!HjI2m%Z(BjuX77hG#Y<YTL8}K)
zW1PUT<Wa}5uf7rx9Yqdac8>ka9BRSe7__S_6tgv27x#bA`}^!H=bO>zbzRFKWLU@g
zyofvvd)_HUm<lCEWA)bm0Q(~zmod3M69hRkH8Yn1HUcD<UqcrPmkC2HD1X(OUvFHu
z5yjvADfrmFa5*H0Ljr+;B(Bo}EnviXXlwYvbrwZXLkeUWaKC+j!@F0Ov#VN>t^^n`
ztl7Q%Gnbq>=M1@{sDu=TrciSjMRVwtztw`|hS4nK=1(k#(z-vYL6f?np_Tbf3ufZI
z>6-Kjh6YLVnYJLZJY>j{e1CB5L2B_KbOg~h=Se{<CATC<FFqyn%2=M3kkc4$)sS<^
z%+WODvOFCj>+EX@s}#2c%X|)+<R;#(V_ET0Mh>}Gf7<EMhtJj}l$7UpysBttj$T5^
z!@Xx!O3glU?}7K5?FQEH??z&A<{QzNH8X#z0WaokSlxS6oGhdHcz<*&!~8vlg(`DF
zN5`70#;aE8{zjv+F&Y1=NT|MaSoEkY1mopU)qB)bLT$dzl_jn{&Dk_mMtFkN*t6xS
z1<lLT9$2d8)(BczIz}!*Tbi@6%dO7u6syng7{#*BD+$Zjz7`2O67Nw<4J{@21$SGr
z;#LW5nT>9(Oix+jR)4221Xyd!Qwz<Cxz$7K9%q$pY{U1wmJ}?NIY$pB^C_EW+1!^r
zu!dW_jQLPJrH88bt%5b?cR5(|-6u+|b$6f^0@pXS64Yi*eF_WKlXoCsjn7zwUr%)o
z*Fu*%`&#IW1GSSgvj<JO&P5}idYel?e)Yq6<Q9v~9Lj@bfqy(>mIay-HMe8>Jr}q6
zP;&`9W^~T$$*6tyrAhOEt?q_-lWH_(;wRsHvwQZ6%<LrhkGsGB<DcZC#ZJz!$1$<v
zXCL0Zd;Rq3|7j^*V<94L(NZi5-+yWucHf`BJHL4O>zlXxP`pFfy?AqZxxY9IdHMLK
zkC#8ZynJ)H4}Y?J@%;Skk~y9eNRr4mpXE83C7PBgpR;L_Ygs-xiU^y}ho?_>FD}mC
zzT98F+P!%8eD~A-<K@BQpZ@#no|*r>$Dy;!{n_PvhlFLkANTLiKU}=szh6+ozxvz$
z=U?7@cmDBJ;$ss+X5`oOzo4fh-SmC;?b+G+9LVxA{eP2J6Bqq8CN3U~b})2(r0XNS
zdSt87S7RKEmR3VnBd<nTjdkQUM}B?v%G1T&%Mbs$^iO~N<?KIU_ucu$&-;t{YU%av
z&%3|uzJDdlhadNEFJDQiv6CT9NAwt#(IYTPl8%^qd%gQMF#OBi59dFf@1BJ})Lv4O
zMj$2wlz#!}{>P_JudVxCoXDop^JR5{HIBaSH#w1Qbj$oJ&Mw`=+WnoVOlAc1>V)<b
z9dU1Rf<yo;`5(rx-l5T4@9%|lX3GMET2$a7o5W(yx)yUf;e|;tBRH%G_yqpj{hc7+
z;~-&HCqUk)u+2?QG+>?Z2AtDymWN$7Scx?<rGLc?4P|$N90LAuVjQo8A_y{AtPLgy
zVQbl7CCno4ol0!vU(JF`xB#AJt;fk>%x%`2lly3CY_b_fT*AVwM5b1@z@wCHiK*K0
zRwN`95NVRTH1MQu5tO6yzNiQlQlq26YD}m^ZGsagwjvR>28S!ODnus|Y=eamiTRTg
zz<*yv<lX25+tAT*#ByuM8m(;uq+&s%i-g)GDgs2MN7p5F;9EHrlt_n8AQW_)bRu(X
zk{S|`F7P_DB4i8kIou24Q<5Bcfp~zj(Wo-mliJZ}+d1L!Sz%4sAgW7d5DfKtVm}69
z313(pIy(Ac!<8rna6p2WoFfo`Zr6xPJb&H~)SMwKQI&{h(2^wC0xy}>)&0Q>LN20X
zZ}y^zBt&*<sRLC|+4uw$uD6<9=`B>V!3Vz)WY-$3f=nGJk>Ou%nF-<|524()5^LR#
z#9G(8h_!J|to2n}^;KK-Ra^B{TP-hgPf4Y)B<_-QQ2Fp`GP=}fqhXK!IN1s4#DDH+
zX)Pvd1$9a%Bo%>=SUGZ{WN4eXq@1f>@O8->WG5?8(GR><Y}Ayw0u5esi|2(t5=6ic
z&FeY{6CcoN(Fod3$~Jo}fI!7%s=|X0%KO8OH~R1;c<Aj~PB+GAr!eBPR!inL2qNcx
zi<Z>w5$d`{OC)#{O45ghlh3Ncdw*cdXh+yG>K(Af4|cAwWt0tY!%HuMggS!12%KIc
za&A>yM`5CXiI*Rl{Pv>-{(KfCL%%>G*BV?=Lqb6ey+un=<k$FWw%rxVZ?@e}wC=F&
zVprSlXK|A^{3z~V)YZ_{XscnX(N|;KKWd8AIK*p`5&ami0`Sex(O{iiPJa~el1f7h
zXQ4#uD3-PX#+C$<`9UJVQAw~`a`b~corgrng47EX%6EWXa--60t=kr5ip0U1LLSu2
zE(%i|z5CI%quvQ|jvtiw;5>ys#dFTOffo~z@uC+<j!+(m^LZ-EejG1FsebB6h*wIx
zR>sgR_@<!+4HR#g!kq#ErhoC1N<UbUdXjN8CrZ5Ds~=kr+k$y3V3KmX@`BWf`9>eQ
zDa9HUgcPC^H7i@lAVMIKc!hu@$bxzW?<xSrp7rD*LeB^`gDEYc-HGPyLc*&PbXH(&
zN~I-0OFBMB&@sdaibkTURD8M)?0zfoW%~cT)uBYFDQdY*OVmeH<bUIrTfz;DX<y4Z
zFvxVo1LaOUQ0gZh$SV)zl?U?519|0vyz)R^c_8=u!#~VV>DlHHm0xT4xsczsNQprE
zvBECHsu>kW2b;Bwlr{)&1dC+96rvdS%fk%*`WZ6$Ur!uIFu$rKauM(~qFXiq|15xw
zEXNUOq5!$XG;AJ6P=DN<sJceUhOSEBO#*ex7Pw2=py;oh@V|bYa+4F}c89374gx(>
zdVxt>><4JyNEhf283Kq!CYeIrq&`ypS(c+?)QD?YB}AHEeAU=boYg(ZX5?9AsTW<@
zxvva(xOmYuPUxb795M$L_2N&P_VF3$7P(nXZ-#i+aP4Lp{C~R<ZyY0@-39TMORfXr
z9d4}r0`~qOrsikc1R*Kx64nS*zbJ;Z(}}DpSvuad>Q_KEP+D6Yavx?NNLH5K&nnb^
z5|QTh!|%F-n(@+aL(=8=VzrSnW0YI8oGROHGiGr_&bJv84H!lHZCd)_v)Wga26}V>
zT=gy&z;(HZU4PPGz3%m&1{{pFxT|Z1Xj%<fja4q_Di_w<;;I|m>E;Ftq4xaT5zR|k
z5!G*N0KikX0}peN3523;{S2JC)XSmL=N(#6sPI%$Kg305t;DGF@4<=n_UtAn9_Av`
z<x#PDq*nZ*#4k4emb-#_NgwAT)A_i#JQ9=%DdoX|V}HxxwCJgzKpeJSyZ=l`I*YQQ
zGHHl|#!5(ynFtj#KXrN}Cu$=oz>w98Rh!)8#e@9hL5NIsQKkyVp_vJe)NsOYHS6R2
z;$hv5BC%0IK&lXr26wHLNBYr)K#kQe^<$<WRxewKPLnwD@1|V+!mZi$23I9<^OKQ&
zi<ZS@bT86dw1miFC)d7~lGyZ!8Lixf8Et%G#<teIZKcSzQe;~xvaJ+plOkXL4`cm=
zf0yyJG!p_eIhXOTC=>xVm%-QqCYOyQHI{$in+k9L`1IEoiHN>?xVS{&Qt)OAOCPL&
z{Tr45!NTqqaIqiYfn6PpIe_i}6Q^LTLO7$}J14wV!_KgSO}LwT)0svafqt7Q=p%>&
zP;KuB^V(+R(Hnm8dDP@lQ`Zz+LmSl@UVEdJkw!9vGRGrzvO>=g%WR9kcf=XS<s*Nu
zPI*5XZTmT2ArQChqi~C$CEQWg9n05y$Gbpz&++RJkTQ?lc?@)2ac!Y*pW2zfc{ie}
za#CLg`4HX@#tG7ay!PH^6e$(0SZRSZ!bjh<;639^p`En9Kv2jp6)udESDNFByy}}W
zxBTtKzg2oGhm{)apy8#BP%bTAIbna2M!as)ZO=AZ1W@mBcD7k-2TdiM2d-BtT&1??
z*|lnFp~@pYzEMzQ(fG0A6qAIidnwgqRbgEl9=b_H3fRWPMGE7xB&6_3>g5lhbo<NS
zI{i9&n6sk4&;Z(`i3g-f5-ssQY0Z7aXyUT5K_um8C`==MmJZ{cP|>dB=uv<BH9n^L
z=}lVVfcu_ax#WcP;Yqz3XFsP`Kg$ah;9H0**ngfEO<Qyl&a8I+GhX(@dR=hBS@SHH
zhuP9x-@LsrxE}px;j{^FI+0J4G^)`T6-&M!$G>ES(ld@yUQbqep<|d=fp_?68<twc
zU{^Z_=2Jd2Z)Gul79XTHd{BSWiRP)4!@MQ`+UKk!=LRZ0E5Bj(iTr-4r8pszo#9ii
zQFElX#<%Q|nct^qnR?77yBaN^`Tr|<^O}1N*@PM5Wb+H7gx+4&WlKt7kW!sBN$M<5
zbUF&9EYD^OCb4HFsH<mrp_54j6`M49+~zr9*T>Z;$#}?U`d5kaxO0E{?F!?{hZ=ta
z&lW46kFuUR<Ib(XNH+gg^YtR@P;>U??J$?4Y|tY2)cb84P+lZpq>V~g-T1HUzK=w!
znO11qIPqwFRsJ&TZNZbgv}<x<IZEeI$zSeh5nTgzY4A$8Mu9u0fX~;D!Rj_;*6b>D
z%dug|udXObo>5aZVUB;Xw}4IP`1^DG#}yD+lyjGnWlHuUhf=~qrW=cd%IYS<LG{L@
zGh)(7B}H*WqiZ4sxHs`u!vz;`V7L~NC-2>$P<AjC#~_t2yUJt-WwHm#bmmS>3PSsI
zSYVjOvuy09Kub&s?lfwzJpn3;V}_tuSNe`)QAVDeQwhh*R&#%-nua^4vd4qtX<aYY
z3dFXixp%YD0n^<RI-y7ugbhyThJ%f{kkgSF+3klYUZ?tm$BiiH{A}NTtZ-P#K0-R*
zgHqcK?WmOMtq5y8IIx-K=p&YwVcYSwQr>f?-sZaAEXnrWD3Kgq({ncz-9reYpo3D{
z<be;uM70j5AX9&3IbDcr@i?Aq70oPfo^gRUCUn~m?Fih#)Gom9w(Ym0^Y}FkySf4A
z$sQ<J_thN=4`+deGpo#j2ddbUR69mpPDG{MXtAs)_1xw=&IpBg+G!u)&e*Suq;1JV
zT57fD3H|6l(XzFf^kSWE9_a?oKG6-95_xVVt(Any>yv+>It;tI!A@7YF*J?rfboV*
znEkJ8@(ANSJ;OwV$#~OaoXeOcw;OWY)BqENO=i_%-AHw3oWf71tcTHAYe<3Q3d55k
zSZfP(nR!)D(TjX|GCsErST<EWFHiKtu#4^qaEU%~FY@p>;;yr{bWI#{vss7TudqML
z1LI-SWTk&4DhcOT0Rl}`M|0tC<k;{p&hWtc*Hix{laD(+m|@-bIy|&LDP&!{)o;km
z*GW+tqG>K?sJife<VaIX9`=;ZKt1Y{pV)MwsYWRnoOzp746!4X!bA$Gy$<-1pc^-A
zAPOlgSbp{r2Fe1u_Y&sxt=Bia>yVYJa!}&B%I$w=u%RxmI_vVrZMB16FMIX#ll5^i
z!XP(zYrN#Ht-f({D<fla$VZ849HC(2(q*g<VfX4T0t-52R<RP~ROJv~-z>m&BQNqa
z4dzj=q0v*EN!@it-71`U+4N+XV@vgHsLNbjYumFxsB@p~@OM~tQwwy0qS5_W>ufp*
z*{XkzAN$^V#|!EZL{R;p^Tc4*$2qf;Icrc=|Gx77EF)n@KUbz^t(_TpGYf>~1{gmX
zy@|WGDK{DO5Y4NV=bS;4z*#>*JMf`0>^69|dIWn}qy2#f&sj%78zZbzdw5yN8&6i&
z3<B$ut!k&vIK&OL%h2p56~ppCJf=nXSJr<vRWw>E61nauw9)&phhzusv5m$#cV6$_
zg8OaFkHFj>Rl=4XcG6z`xR>p@;@_|)bqgdni(iTlfSO&+jpV}$N^2!73*VAKja{Ex
zB~_D@qb%DDI%`&z@96@8rrK2VF3V=U()veh+>W#9Xi5$~Q7PTYsvy7tta!up_oRO_
zR|9|gLAXX45n*VCJM^`}VJB_i_RzB%dw<ji7`t>`k66cRIAWr(Pck&~wFMJcn#v#t
z-tVFn?XWeBcx`JcB&lvbet*Un+tB>+hfk|*m6albJU)Ma`Ooqqmyb(nX1_>bQgqAz
zmNG1qfq#n>!P}tPf!QB0@!h|_e))gP=iilpW&h1n<zfXm_7s=z_0_rT<YQXC;d>Oy
zd2JRYk~H2f9!6b)3AC^cd|qCx4glRnHu^1(CSjx1nk_a;m>6JU1Hgs|hrDAOC3xTX
zAYs)(6q*_u`7Q5)Yx66tr&akiKm47h!m58RFJG3I-;k@$dmms*jM}(-{1|_twOd>T
zgHXXJ&RH6QLOF$GJ#i(0!MmMNPH!93`F4&4X6XpdLfI@Agw|-L5k@x+!Ctfj4bksB
zzQc!!Rv2kY>V!+M&O>>YvWhC?Zsl<Q`}22P_Z3!sWWRNv*L>=nUCOHh-{wG*VXtz5
zZzn&&=`@=c&OM7DD0KM4t!sbU0KomWFmx(iubr^po@NG=oerdS&W!m?iZQ`y(Dl@q
zu#*fC4c-c4?QZwLsmtmucMqJVkVdsUCM?CB)>{-fzXd*WUk?oCy4P8_bHobl#%5a9
zkX)An`F`Zi%^C>gzxKX>0U;@n&cw5)-St)gl|X90m!#q7BBkW)Xs>o&w$31bA&(R~
z&ZzMICKr}s83wOrn0TqXMiqtW-Ef@6VrRSR<C%Cj&T&Y|P$hY;p@s`|>`JPVJ+j&B
zsz-Wck6rV(?6KQcwa0%g<1~z!uPkUjaSd2$4Z)zF&ro@vSobfmCL3!5R5PjBhU={d
zI^7ALUo-LD%t}2FhOE5I%@t#Rt`haVeV6PyDAgCiZ{$%pe=>i>p}n;#p^m316US_s
z_q-7rUJq+}L+4<u1u>k3z0X&qC2L|KY{EDVJhJV(8~exQET?0TIGxj3G^ds7+o%pD
zucXL9m^HvE_v^(?6T7+dQ+|7Y?)*DI%A(>7WtN+E)ar9|l5Fy*+o7_5&aYa9XP}C^
zVaIMC;9Epl-G;MJMd$ymJ?!jh2Z-Y!Z?nIep;0rsg#v67YQJ$2Y)@K{oQROq(MejY
zn=-Vs-%ocYwxt8)2|!~dom~Ij2P`)_KPUt03QQC}aWa^qNJ;UscCWCeh8wmo(z6FG
z?2vCqRfaoKM*@K4S18DTgGQv8uR5$&a}O!_Y4NN&NE6E*c2wsG&s^3@Ft`y^>!0LD
zqvx?xh;~oa>&8R94zYTo<pe7adAso^SlQvKfA7uoYjRDOb~W1Aa!Gr_?!Mb_jgXbP
z$os9+6~9C8rahEjVcO*E4lUF$-p-KO>EZfpcu;|+w1<kazIt{a)V)%<Gnk%hPHx^S
ztl49Cnk__Z-y3A<&;J2NoUD?U@w7A(0X3JwsxT7+GBh`rfpiEamwPZZo&oTe8Zk9x
zf5(26-!C2(DWW7B7*0_~if{Qn>Q%iZYJ(ic3r!2Ga*%yYl))4^()X8dRwi4$=3kI(
zV|0F@FQebH4Xy0?C{j5&k3}a?yUGViG6VUQDde>Zj9RRJl5)Ezq}+MkK>#AB3_s$l
z)7&(}IAM+EGqfwb_YJ#iRc!@2KdsW2e-*l8maJ5i5eC9!6IHbBir!>-|B*RJvWBEx
zXU^B_uxg~|xTQmD1z5#rY<Y%!yIL+KQJ?|NMae~(6sWjIzfNt+1mR2=)CSAqSri0%
zeDt^l8dW{O6$#9Y-SI>zl&NG*VbneOx#6u^2~d_Xk~V^cS_hzOaCC2tCMaQ1e?JgJ
z8O3Q0JH9l**8|q%DWrOw*5orNS8u^rrp0})T{obmum_wY39{zN1h>KmKH*@bA>9e5
z%)v<W1q%0CfD2r%8!_8NNU(NCCs$!}*<XzEnrW&r`_Z2nJF5Ped>gSNM8ENau6AvG
zTMW@&pm&@biy>E0L@PoVErO2+e>U{OOZNrn%{VQ<_raa`$ki(=yblK<a2MWsYTA~t
ztLEJi>z4Fcmdxd?3m$`fU_@AqI+&u)4O_j4+7VvKb}9&WV;_MxC9~@ZufYw0CW7Ts
zY-NGx8^Y-|DH%lHKP0>|l!!@g$3rl5j!-ng5qAs>G1F7iw*ArMA<?Dqf94>-D66&j
zoU<4<dRkH3JmCh%01*fxxC|Uft)r=w=fbq{yC*+K@8J}6NekHaOuB1#ShR*-HZA4Y
zP7RiL!1k!%&nU0DSaL^hh@;VXfRs4`a+6orqi8J6l#{8++_UpeQN-rYnoE(em7YV(
z{d#vPQJzV}a)RrsL7C2Fe>g>5l2u(^q&YD0bckNzc5WJl(rVzQ?x2`6kt6a)*<Cy*
zTXrjFg>wU?<X2F0=OeaPp+Ra{o;CA~x;wun_Y&F-o|&l+dBa*1*WL*oVDRr6vCU0B
zit&6whQT7sn;E4S*6LLmhOFVH2!=?dXE-cVgd;P^w6>>>y<VQ^e`^|#%HyR7evm$%
z=B86!m1WBTw0f#}EmB6QZ>T`Z*&;e6WH2lFI8U<JXsjA5yQHr}(|$QOwsq}(N_@Uw
zkTN$BoHM@9W-091p5LGU@=KY5%Ts8mas8A({jzb^J?)<Db9&TWrsK+@d_9&YcU4de
z4LfHoQ4tzkJ*~3!f19Zku#WNVQ{PgY=De2ib{enh>DTn6KvAWX!_Yv#=i17SMfSJ+
zT$ShOTFGN3kW<r20|&@#Xg07JuPrc9r3~U*8WFnn%_|92XGJ2M^S~VLi}NJd8w?l0
z0D@wMTS(;AERU45fdodN2H|TjOmwh9j`t<>BYEjpT$dndf1$xVs3c_6;#E~ThEl}?
z8X(v8d*t}Kob^j3^n3z(CAC|Q>lw$j8m;E-{b<w4kRdtAdFj&uWW`+Ec_}&VoiJ{K
zrX;!zpLDD}!<;9Vt)xJelfs<#*C$j+(I<EYeb#$IPgMF?(@{#F>hjG?2E1N|4opbp
zk)N4nR2o$+e?>(JHjt_8HOoH2S>e=xb9u51id*mzb1z33C)4X#2IUNNw=OyJxi2-*
zjMI0o1rL?|iDaCTNFFH0Nx9YQcKyr5oPIt}t9F5=*qlS1>404K;Q%+XzT_FugG7}e
zLXT|nv=^`;#Xw>dWOQYfj$!Fr=Nu)5$O}atkZqjFf8$F0gf`Mm6f$*qo=eG!nrWjz
z<DQ9P-4cbu@_kBAe@p!;=#&Dn#Y?j@%L?l>pOuvU+N52}vo+;V-kB?TY6!xZF!1CV
z$E26VEG7LqVPrRNf`@8L8Hyuz&ed51I%`z8NzVITAgDN92CEFjF-Nh8JkJZN<zbr`
zkQ3fTf2UO_G=>hA<7~=(lnafI<!{)1>pgfHQ@?pR7K3YW2A-0bQJyQ06IcBWr2vYL
ze6{`BY?H%XHl<aTqy=1+_HOX4b6*biWE@t~O!6_{YbeVK;JF?unk1q7{B70#)gNhZ
z>#%h<TCWcu;~C(+P|9(>xeCxR0$vElYn|Y>f4xk`P;5CZ`MNX9Q|^Yt%I8X~NshLv
zq^zSpsLgm{+Jwd5UME<xw8EiIAW#MPnnFSOAl%%6^MIBBM)BZnw{l+irWxh4`l^(T
znuky5t74Q7^i{DL5A3T_#!@Mx?!2Q{NWk2*iyQ#nn|i9KxD1|i+w7^zg&Ldj@SduC
ze~Qkyto2k8@HlC|NN44`gsW`(`i_nM<bJAnEfD6WmruNsXPmyYqe?-{*zF_Xv#0n!
z=%`YX^g6dVd_hMQ<G9~Z#U>n#G4ZAi(Es0cR57`3L~HPbS3Z0~M-`*I(^17ne^%6>
zd?650aM?DZhL2s$qn<vin=0RmGH!Que^V*paM172O|=iGgXh>lvM%6fQxBI5C|hm=
zpePLU8B89ZPjIJite(wJrwhMwD41&GbJs%DjGVg~rVFNynrW_V&D->}9&(yKQ=bsI
z{5D*%-=I1}6^~P`?m)?Sm>_>l3(@Ve>q0WfmE#Agx_uw#Oi$n2CXP;(o-2SUe|;+P
zT6NBa`DD~mxR-I5siv`S`JiT2YkGPq)ckl-%S`X@?vA^`B<O^O&YcxX|J<D4$>t<S
zH>X7Ye&!Jb4et8GvfEGAs1a$>--5T8<_<P4;C<VVD$TTY?YgDvd2Q)tsZy!_VozQ=
ze}UFd=0mD^{LE2}O<s*GjUR6Je;D3nBxR`LFiRdXs$(n=D-vD$uB%xkQxVeN?{bx6
zZ9?hzxMoA^=JMQi2<V0|{D{mC04YSy{{(UAPHL{<#LI4wTA*{?$>@2*s8?_(<s;|N
zZ=+)wc*-q<5(fMuyH!i~UQc+fTeajh{oqIR<CSu@C%jfNmj5SWQtY*>G2YP;nsx6X
z-6}l3Nw+7w)1=!EK9s<2G$7uyFSAxita(q4u)p8#!r$LNp8gB;Z_$@w84MExIW;hs
z!QeU*mq|J`#Q|}*4?8t#3IjjmpqGJiFcX&uS2YfoY}PIamo0%ND3{GeHJ*RZBfY-<
z>+{zaNsLMnUHUb>@W0+aU;gvL*>)g|hTFgYdP(ntQNjd$bIJ-A)$Q-!Z-2b}gQJ$Y
z0LPRcfB$)7@BBuB3kn4%owK$QoHfG4aC0sSlm0adZdeo~u_W=Z#mR(h(+^1#R`1}1
z1wD8Hs4<3~#Ay*3VuhSXq0oPYihjY<`$&CY!PDxzaMs+M4Y*2IkW6D`Pap>)l#dJK
zh)-BWU#yOeys$b*QuaMp=Rr8H7RW#*pUu9#N~O`VagRFV1o)5<Kn~OkDs6>Tu~6Zr
zv?wQ=Uj6fOfhZIJN7Ztj(=5fB+MLjbQzA;wk(4D#`6RV<9()i{TRwkMemu`Z-f;~c
z?oA%F1OZlc;6W>$d=zIPyOYq~8aj%I#?pZW69fd`<Y7Y$7Dq_v!2^1ATxnAkrp-f8
zLYgkB2vJyQis8{@!AK{p^&MC+7{xH`oY)n{Xdy!nRT#kV6ql+nyXIh$cj%!B<D5{4
zu(c~1ETm9l7OH5uQU!m)mZ$=eJSjeal7{>W$+M={=RJ(aca&ojwVZ2qY!&et-wGfC
zN>3W$q%P9)B*6=zE<0_8oXpc-51=ck@j}u;&YhvY<veyf;b5aqE5SxB9GFT8wu=%t
za$1zVtOSmF(O^gkYp6gz)oeysKJvUGFitphkQpdY0@^_l4B3BnNhdk#MfGMyFjOFi
zZuS+yMxIv$`b4j7wXpj|MPRsadVzf1C2>Sp1fEk$V5pYd(zZ+Dqs}Y9M&X^SEvwH?
zV93zNC9R-#a7hgHaVReEQRkHaK?Lc)LJ15xi7Q-J!U}2!cf?Tdt5R4ffsHz^1cP}X
zgs)J-3Nl<*0z-f8q=Yrpo!vk&fsH(`2<=1&b)s+j!~}+Xzpey^m}|2Odn~5V8djxX
z7sv}Guu<ogfW>F5^GVl72Q_d971V&80>TpXzT2|1sD$X%PMrxcbvH^i0jN}q4?L*A
z#NZEMEI9G)sQ^vyooQ>0?N&A`t2&Rgrhoc~a<vgo2Frif^;op2TdnkLyoV%H%XBnW
zz(q=1(ou94!vTbTobC)WM5WdhPY#|@19j=gE-dv76#f2~^$<{{@=kpQGyOlQeR%yo
zEI6I5Bst(_{^o|$6Z=L4X~3PD6I*tKf7{162}$s&4otG#g-jME<YKlOOiTyB>YWV}
z%Ka)nLri}rPg{UwHk4`pkXaejPC6rEs4`o{1(VNP*(2YvWZ~l$9rX_35Rch_kyT|}
zPXn$I_YAnE{vreJDS3<8pZ9ag59eaXETf>9D6}dmOAmKTr-BT*)#S_Kuc@!7pMoMo
zY^k4`jC}quQ}xHp)jUYj?!?K4#~JKFzO+^X(0PAwDZ0`0(ka*A_VQ`y+bFNL36lNY
zSA&NR=$>NFy;mdd?L9XC7YLt9rDJ@RXuQIgB7Wi|RGXFZK@OXKA09hX*!$=!F349$
zBYcFT>xe;M2J!>sYm5`dgoAwLBb61qtDO)2ozwRVp_NWg*lbrrYsB=JW1^A=4a)~T
zAFF@c4LyPBx!LA;dt0rV`wKVP0iMb>o6`>0ZRS^KB)#Y9Dveu;HODfEMhJ>!Ih-mf
zF5QsnAb&JtXDml-X<$@7pT55ea=mkqEh_gI^^ix0oX8PRdN0q(L1vyv47pF<&Zs#|
z;MLPbUt602ZAhD*+N=?Gv{@6+rBzE|O$2{e;s9-(W6Ww9gqLn1EH|^eFy*(!gMD01
zR<%{hJ99C43Y_6^m7pCuS{FD{%PPFV?5R{G!@|Ne?|hRD2CYt?WV4xej<Ol`6>FuI
z14C?CYYJ9iIQ5X2_~~@tQu0#vB~SwMF*$>!S&tjMShmjlVCEIlAaHfnls5bzd^>+u
z=F95^=k7|NVvL7{U|>BHNyUO*h;fYH-AXm%G?FhbR@i3>WCkKB`=8uN(q0dysea;2
zKc;=M$8ZiQ3?a;RqMX9TqLkj>t!icKla$cmSnwUQD~7m}UomRF!^-W%S!*H7ptKyC
z!DC$aw9DfBuPf1n?u;E<st#7*!>NDu+mtoTRhr&2xl2b~lw<y2bS%sznD=7$&n^=%
z3_7>e=(=lwbQ2LPZ>~{Uxd(23sj>;7r}@>0d*)YD|DT!Pp`{k{tEu%YMy2PR$<6I;
zG{0*mH&%&!VRqf%Qt4zD3u6Y?B_8RBQ7>L>9U6t($yY}e3FvY5ysgG--nf4ryNfmP
zm3>Bsay4MHTS!@7z4R_=6t50ebEw@bndhu@Jl$X-fX$1wHSwB9%w{6J@JjWp&}zh;
z3autS=Mm4>XNOoV^;u25ZzI#D>*b2Q6W;ryd}qg9Q$Lppoai-@1u`DZ=7&m5G6z_!
zY{PJML|NFLjUtA)ZMcm1hqr$gjxn1}mkW!IV}YC(xLFQQt#jjMHG!L)oP6|J3AMkE
zt;9$zj5KR%Pkl$DMu(BO9zF)Brr%ozgqez;u2AHw0M8gsYO4puky^2l45udUkb^SB
zog9?U@213Sh=3FsQ&Ts^v*F3dSYhnT+nkZO_0sR+AqsuC=Gd}w=jngqvdqj0T6y|_
z<rvTTfQ3=#9C-SCTr-$lNq0Gq+m1O`Wt`W>p7Lnp26{T>8ga)d*Tg^GDIa3CIAtcR
z15SB+_oef=QW{OyX=u@LN>Y>kREJ@cmBp^5jCKZJ?X}h#0~r?ay~@Q^*VY<m`uWxx
zNB9w~HEmNv#9s+#RJ?yNq@AlFjJm_s5QeyOHH6RKEuOs=(_*(W^1=aMZO=O~3~_Ja
zZ2lh`@iaIQqdUkm9W*iGqc!u14;7Z=lCk%FJGo^1m6~nv$+B}eE{0?ziZUJy$i-Od
zQR{aZv__`~t)NXbGGf72O0MDfI3F#i_yAc^=k2<lVb87^Kb3!-QhTOL(55?F$GFoP
z^B)bI`r58LRbUs>W_-)jcgl<TGep=f+3C~YY}Bj+t+smc3YVd@?pOyg#GN{bQGaPK
z9(woUk1aL#;u&?Si7n9iKG|)u%d~_u4IdSS(Mz6o8a~t}Vxu~@(o;gFoVnOaSKbLN
zdsa0Jal5Kv#Cv~te6quS;+l?fXeeM?E*CVrY3*}}I1v&?C(mYV<(QODGwMr~+IRtu
zA?{p(W7PRF)YsRQ)CxJsEww_pw752NID%#UXjZkH3&nZ$A=B{!$`Y9?yR&-dx8=u=
zlm+xh(-%t^n#j#cbvrfmTaLR%`9fo*Z@b}a%ci~NB!Pd8xMBhHZ8r?DWdWdt2ep`A
z&%Pt=F<si!>lZZ@U>&GKPj^=%?zp>}_<Ueq%{nmb7#G_gsAy%<DrJkRXSPxS?j<Yr
z>|fEdU9E93fDaC^UyaKUcWMDX|No}OJ;pQDxQzHrf=}O5#i!j)$BXx8^G|8#dtJ4b
zs`|qHQ7V5$*qn_t>xc}R_xW*#p!7l}7m6M#Ulvks6g!mUvXwS_0v)7l(wE&yeNei4
zA2M9wyxd(m-@AxWK-5pRF6u$+{x$?*jj76lRXXp5^*ujWTLbR+<(hc^lGEvYm2nlf
zUBQ^Q)MF||uYTPqdwu_W`7couDd!4hZe(+Ga%GqKv@{C?H8DAtvFkPym#>QmCzl^l
zHI#q&9US-ny#2Y8Nzi=%^0sRQM6ZPMYWV)X``8IWt5l?zJ@Tu};A7_`#zE4UUq1iw
zu3>onZ}+XxeJkrc<%OPRD_t<0vLw9I16nQ66ubGo4f<);O|y=~_8{7;IOc-wY-KyX
z*gEjamQnfz@W`W@N7!^spLkFj3-20x32%SGW4N{d*oGgS6~VK#o<>v2K_dusY5?to
zW<&k@clZ&kQ?s2%kpnh{G=`{(BE{_4K8Q+yb=ey76s7H`6bN=%B4Vv#xTuj>3eD8z
z=hx4n8Tsq$#868PT51?_M9ytA5o_UmLrZG3E7{R!(eB~?I@({d-Q{r{5474Qznp)I
zb}WKr$p8lsG6`Ib%mvOC04bE2vMqbvjF)z!*U2txVnBFlTP}|nl7rY-XgRJ~(mLT4
zwbUV2qvZC2&CdHLBjGV=6>_vXd!w{UE5QKOS<BAsoV^ao4>=mtIEE&TrFSu_&8Ve3
z%_8A;9YUta03S^x$Z##{fgyhNXL^4f$qI9;(~*NN;&3A{tItCWd}D!#1;ZrZrxEy-
zOQOjj1BIWQDAfpy_I&p862K_$1<EOXX$m>Y&V&XcudiBGVci_*K{+Ro;V=ibOQB&O
z=WLA7oI06vC=<Tox(>0(&T<`x^Gv=&3eyml$D3_lRv_fs(5(<X8a#Q3C47Gw5ID5J
zv@_*BKr}*2D?{tk4IG^n8avAaE^CtGj7Vrx%DSc$Yo`Qf5$d*U7ZQ%J<-U^ez!HP$
zWvO6xyO$kuiT0fg*S*XjBf#5sCPVsW0GCc!X<AM|%j<#)8<DkTEsnJ|{={Xgk@>P*
z`nBm)9}Cs;(#i|Z`<Zse{WX89<)l0JZ_<c(Geb)XF&xAz;6f;4;dwcVSCuchz=el+
zyOA*$CoYk@9Z@Jemw3Egg}j+*RtY+Uc%+1F5#LS|&heIc!D!L&G+!}~T&OKdYM}`_
z3G7X{Tw=MIy*)E?oY#zQNz5Fj^-TP7N0e%zuyZ-+<#lsR*VJLlTrz*;dLVp?3sWB4
z=RwXMlDhZc!&s}V5!Tv!T<TF+-^;WeYk9ZCJ0|R0x}5VUjUckC>)xK%2oy}cl-DbG
zgS^8b+k5e}DV*93cDMfW+Gqr#>({>fh^LUX(Qbaks}7lN8FTX<tezcX@Hn*wVk;o+
z{B%BIo$BLPh_2XnVOoE62NWn0R_RufwL?~hpqd)59IrzinT~|3_2~gGC?$wb{dany
zp8Y$$P)l;n3pMqpdZDv$Y#eqbIyIoVrK#fuXx%Vg+plL&Rm3G=9yoC31Dgy%rmh|Z
zml|f*dahvVdVSghs@#++%|dNb#6fPohLO-zgd|XDSxj^E_|$(l6yrXyZz?9v<dff1
z!Qb3o<YHsdx9vstsDmrr#ao;eJCK+C1sa6&Q-dINYNoB@ZHMhp4ji=HVjn1NsB`Bq
zbjss%N~#eM>Qu?HA?u7KN=rJz)CAa)C!yD>Ov|-9O>6+o)J|<VV(Lsbnn!Kt{k>r8
zl{EDvyvCOE?*M;P;{;!=Bx`_b8Az3Csc!_-M==P&3skTaH<XS>O#xer=e)gXIDH~c
z0)*0|KP&HQZkUdov77AYdB*SjEJhB?c_+Q#4ey=Y?rHEeE3Wm?)6ut~rb>o%5b;UI
zi{FM)-@V2aX!p?flq${xepe9TnBzPbp~!lbi8Aq=STjtKhbxnuFuGltOet~4ghYo@
zyi)6P-X<97U22qO^wMAUNV$00!tpdOvIWm=D4L_5;owGY_GK@A7bsu#D1FtV@>Or;
z+vfkZT<@Nj@w7A(12r)=mm!E2E0=wXH8YoxEg}S$ZdWx40W!CES2fu)0Wy~%h!!ii
zeTy|X0+-Uv1`+}@IFnExCztSzH4{0Yrc&o@Ro8av|F5;dB<bC$gU^m_o*xe%m~R8-
zZCLl=Wpm!dDG^3%_Ai560RBbbWFQj}r{K`+mpqO&9)Ig`NxhX`K2sWo-S|7?(F1~&
z(+gV>%b)xE07~?pRfv9Kyv|CHJ+tyWg(zE@>Qf%AbrSjKS&6(vH$A!Z3^7Y`Y7#SJ
z;u$1e$laxVi?1ic6|0>%GSUl*r7UVkp9<<xFgrU~HDjAcxZZ(HJJGF9Z`g5If7rO{
z1&4Z-E`N+eyNTb<n#5H8+#ipxL){}oinO$fbU<lu?rt*IlByN|L4h{EDs)a$oviZk
z%C2cux8h+Nf}J4Qz6}*G^nH|fb{OaDeKUqx*{5E;SGQ%KzJt5#4qDe*RYqr>><<>Y
zM=g_78L|ft$YV$DnpGVRPIcCzo0#-%gQ{B^2NNg_{c%6|jjsQ!?qB!e7a4)mBbTwp
zH58XH>L(Sqm5(((2My8`Sf|gjMf)3~bqqU~vBotN0x~(5_Leno0s(xNXO}fNf6oK{
zfRx7IkH_Bwe%iZRqXIsm@Ck)~RqF26gTim{vFUk*KVkY^Ki%De0+0lO?^A}L?!JHb
z=T~wjZu$KB-K+n5Rg&d`5l*Vh-+p=Z?f3HX1OD;P7b$%7mrv=@UoK#jF!+CY`T6pn
zul~{;tTG~ixj3(d#Ls20@y-H%3|b`GBJqGjvp8UQ#Fr|VH7b7-j5U}4N6ziP{bm+;
zP)#0KC+xxj6OnA#hC;ss{r*J5#bH0robl8D?E@3xmrqV#qKHmA$Z=^6!l3jvTsObH
zCO`y03PkiKSPp=1iO3#ng_Ldr`Dq3@mZ>%%R4m3LpGZyp+gTtFlh7u`O=7ZlI*Aa)
z1Yb>KI+FVQ4n2R3p8jg}@*aTojc$e_4&5(YTc30D8+5{(-UvQ}fEjOC@J(D+=3&O0
zsPpUlP%ifIfQqIJ*XNleCu(MC)}&~q5r8U|NwF;wD}zwha*?RX8qR2OSA_wD1+K)I
zASZ_7_ax-lS90qLrLHL4qSOWUs-o#3MGJ?Dmc%0`*F=AGu5hNOX0EDcX{7VJ<wIs7
z=?SJNE;l7o7njs!rqIqj%*T^VV1-~NlE#2AO5SO=DC~i}<l}X;#b{xDSmiZ)?=*s^
zO*1HJk+!6t$ISqBYzBx^?l7a|=D)iR(K?SHE_t~G^6quZT(z;g@vd&y`$G^g!rOwd
zFWOivf+&CBAb)LpYU@P|u@pq+tM3;R+B`HWU6YJ}{E7OQr4=%w7sf(K#-#5DntQ0F
zoBLhdX_HK2G2BOghFit~W(qrQ*+aFxi|R@)R!iZu?imfGe2eo+-)Dxi8rIH$?$q1Q
zx1o%%o4>cU+tAnxknhl~cFYL#3T-)5$BaxALB)R^BGuMMG(H!hWcBmR!ds!EJLXb_
z49v7W=#q@8S#%}7nn=@eUPo&&0deRxJBh)+!_{Y$+}!uo$Ato><Ani1060j`3mXwh
zNjZ<6>4mMZs55-D9WQJ#soarKRX4e1Fl)W_aM}xFgw^H@w~U*ObCAsAU=ihYyt9oC
zse*s5cZNCnf+;0j61PRK3FfSK#)x#7uAo#Y&Pge3nCrL6{joB5dCta&l>(Gyu!&(U
zgHc<(;>CR%w~a<<J;zwx@$u;E9Oq-NLd8VP*lC2?$DxD_v~Z@t@|TV;z=Q#m=P(^#
z;DRuLLv)f$kID`1)L1-ziI#rIvC2>aYOsGfH4@Cwd@MjZb}F@yE|eDWEU#t^vIljK
z%dJfY)`SvtNk-LOkfA-!6`q`l%RC&ow*N~5Yz+Z0i3$E#Cyqw|Nm<87P^;mdPWomm
zv1%!q-IlpkNb24T?ntF~lyTVx5MIU=#%0C#67gXx#;@VKy~di$bek7f@ZIgcMnQj=
zhSh|cl~C$%MAj19mnn{Vs8s*^tYMY<8X<3UrN4FuKAK<>E!PmBqke7Z3h>`v8s&ry
zmPQO@xHMu=W2I4vXsk3USp6cU(IJ=GrBOk~$t3WM?lYa4W*d$REI(dg9nM2oYP2Ia
z&dfs;^EO#lCo$S6d@05dgBH?wsWyKe(|<EEX`_{#vH{dpA1q6D!_=`<DkjnEwZD0O
zW@aT`I}y&1{zCII>o67SC#?0ie%gjaKPGS1s=0w1sA*az=!bTQ8iU%;kT83<hd^2h
zMasgVP_tgQ^QV9z-Epu#yuMY=-F#?z3xj2?l%7L=|N5mo_{wUDl|opHI(>gXYfLD*
z<@T8EWV<KXo*}&Xw#_^((*z*{`5BoPk%etxdEv06Vt}g6Zq#T;#YPUYI_H<<I7IX%
zW>fA^#QCR*(`2JbvEu=UE5n{RoFiDhdX2*xs~*ikK6ZHvC=>?r{^hxZMV9fKUOsM#
z=%SB#{8)7nBZZUd?dLdz9OZwYW-^dPGtszx&$zLyG9~=W8#g$V)z-%>nV9BjXJ!%u
z_w0feyRw(x{MXCVj!_w$TSMJfnc(#_p}C|2hqSj%Bzt)$JzPyB2az4^KqQtAVf9!w
zg;1&Q$v{-w5Dk@l1z}e(M9e>}QoEExZnalFWxJ$kDH+rAryDlVw$y(zQbwhggUGI9
zOH#{0-oL)pTCd)q7OK&jC2+`&0|6(bR~+Pw1F_^d3ivBH5L8QRT~+uUhw^NJXXr--
zo`VbuJO^0?o<2I!69VQ6mcM`r@#i+pP6_&o!GAGf?xdC82`vMMoYTrT6YBiKl}dHY
z5Rh(959PScQGh}y?K!tdo;7F)e`(6zdcAxOQ1Jgv_ox)wmB%^RnbvcN#g$C!B?DA7
zB|%5z#8@_f5j4;|1~Sk4((J0sDk#`8+CFv5x=1cW%e0r7VNunkUXj`q-1wGtOXi`y
zU?lo!eyyhR)k@8~NgdD|6eByFssu?ZQ>Np&!cwsQxG&4MR|EQ>=DcG_e<!#OG&+EG
zB~giuvNOx-vX32dkuQK6LTe)pa6#0%swnPY>h`N^d#AnSzVl{v9!()=^tu!jk94J0
zy<&jhOS{{R)xxR5#dI3RK+A*d;i%dc0G7@|b*Dz<tSCKTe^BVPeQBJX3OQh9Yzk3b
zA&ScBz^J9i{4k%6X|i*ee{6Aw0+iQoJoEI0HF##U3~pm&AY<DY8Hn!80ZO3^xnU(>
zn@(^;dmkB|vAvHPjT%nvePl?+_C6l7tdgQn=izJgPir9VWPfN$(VpURnw%_U**VT6
z?Um#ZKevZMmpM?iaX6F;5~?;L$5L*-sPwv{pgj8sS8eUjRQ5Amf8Y1Ze^aPH=q@`~
zt{Bme-DT&d1o`9C?o{0sn1Skm1NODf4EdqkfK{>Z6)*1=RitCf>#c=@)&aT9&HWwL
z4wb_eXZOc(-qW<5{kps~Mn27l5U;lpne~9v{G0sn;<Sjy&bJtSJvnsRYz}JrKhf&s
zddb*VcekL^-CUmGf4I-DzDeKA&L864dhZj_X`Z&rXHuW_DLaN$W=$B{&qznuK32~U
zK~W1#q)we5l3&1P114@ZdbHDn3)mEv-Elh_&(sdQhfthXoLybiyW~v)Emm-iD~l?}
zOtUL{>c^I5=eEr6^F;{Xj!RvRz0NK&Qg5@BvH8(To<VK8V^!#t(c5`2M~CJ?+%7Dl
z@06M^W#WfbugklV{^c9{Rz6~i?j$>}W?R*}KM<MUL$>f~cX5--d9?gxN^o+1rV)I7
zXYAYj&alMkdv-rr<adrGZo73%)#I<c!5j~EV|>_tw~nbbz5xXCuoLB%vBotNm$BhE
z69O|gmr*P$CzqhBHJ*RKn*?0G`}E}lPm{~<A3r_(_o1DZaf%r(PT&9h@XdGp^aFnR
z<%u)v?dgB<)}K#W38tLK&;L37>*1f>#R|ooRwrX6<M^Hyt8ghACk*!TD6CgL2*9Rm
z;X8Z*;QI~VRk$yGmiQ8JG3%2zj&X2DCeN7!=Oo^x@#Pe#(=UH0p5@iAl2)DQE))#o
za$YE;Ny}*wWs!E$__UzuTF`i2H0p1<{*14e>3e>?1f^@djo&JTKE2@DJ@ZByO(!Ld
zWw?xLQXhW&ERa6ESi4GsiUQ+nl5aWiL}lbUm0?=zn(EiTefB7f`S5k0p;rtzEEtN+
zaCK#96HI7RQ^|kKNUxK42GB72DwCj)n77J7MIsEdAdtkPtGbCqe4b~fR%xU`dvP_*
z4}CU@Gw+}#2{PYfr8N`I9Ac#%Q%WsZi6`0Bk}{GRXcYWK*CrJDqHv8R{o2>6)u<xf
z+Je-HO7yK9#lej$0Ttawp?evB>Dr$YOiNQy4;Z`yrfPqhJu}xR_$3?BRhRURNwhz0
z8OG&ge@}g*;D!lMHO&uwO^Fa9YXN>}oOeCE2V=KjThYf*Z8C1f;C&4-EL^eSic4W#
zrDzEB-t8Dz?n1R{pa0W3e~P#naMYGNc^Zd|3)f4}2r<55(640iQxw7xq}YWCba|uw
zt6(DhuP1*Y#-%}qhu2{PnaL<NP?B+Mpi~F3p<y_P4U}pS8z{*jHYm*`PUH|aP?{33
z{YfJX2qjmsL0KlPXcn^`s}%-b)u?GFR%=Z9Y^$Tblz}s@7RIm>tI>EDf*z060UwlC
z2MthJ45G%?RDHu!&N9W-rr~XvgcnR$+nB^oxb}aZ3At_&>&wwxF>4xI9g5akcxI(;
zg+E7Ydyi;}iAL*LcZjDV0HrPQRHc|{(j;^J;O7_`<10^4lj5!R={`Z|Tui%<XX~#E
z$7CTT9B&2I6pks$*IlQT4jl+3SdsKYV~D{Pg<p|;wUdHi%nE)W1u9Su<)s|Z4RlRH
zJ$rwH5za{+V<aS&e+{G0#YhUQFyQ;z)ON!vGLvpyLmL?tph1Ig4Z8NxI-_#$KBMwz
z+q|*5JH!C4V;cJh#(+0Y#mE9ieV>vH`aUHY^nFS)==<oJTIit^C;Pq#mDo^*gH9}M
zBd{G;6lgg@t_y6dxbaMd#I?4oyTYY%9#?->^SpzrfXBdO3{Agc1bF_v9jajW(Nzwb
zKAq5D6;hE5S0NQuu0no`?{+UK#Kg;B3Ewi_sYbB4mN;NVwNs5?o@QLbXr~$pl?9ps
zcnA)bxa?FT2xvYElF;n6k1)d(rY0HbQ~|~wVGpqa3?N1?Sna|znuo=qD&c*xTOfZZ
z*_cLOsTfO5%Y+m^8#b7Y-J!RNw4~qqy)CSGJjjaVoE$VX{jWQ{t<VF~HACVm8$7X$
z0dOtP(Hae-x$<U{=Y|TnkL7Jewd0PMO(@<ldg^nLipebnY;ci|c`p{r5!CVJX1<68
zaz#1TRRhV4_Y0-}MRLIPs?{3=Q?Y+4jo&_dL^S_!=_&$3@d}N)dDlxpB8XynP?6jL
zObv!FFAsz1+a`~67NZDCNyZVBQoRyE8^(hON~s1Bl#&c0sO8LXdkjG-%}%nL5!9k6
za(=vnXd26=X8AjS7K%s?jjH-WHk=9DYGxGB02n5RhX9R`yaLcD!*M{Xc^-cSv<;p+
zfJUhX0gaLj0@|7ab^xte&H+H9l*a(AQTiKIQtN<&HKLL0vmF8)jmzb_v9&(nXmoBa
zp(eQl;Aq9TD}WPDe?(w`Nl+13QVtLgQe>g&3GrJX3V;*7=P#E>!{ZSHa~yuioJQe?
zl8nO-r8>5a)G{2~MyjaBwvm4-lCf<h&<0(^9@<8#Xnx5gX6uN-g3Y&R3;XzH8%E#n
zfK`B=MooKNQqPoX>|t;737BoUAY8%Di3G!>Y>AZWE657wMQ}O<5tkLTQ-Wfy>*PTL
zS1v?@#@_a}sSsqm+&1iWM(7UcG&R-O1`-dORnP{YMI7(OWeH4jQFDL)r5{$6jOSdM
zWzNRC&PXGbZp`ZY{`xS}@{I~!yX_TDvNWr_0}?3ACZ*j(0+GrKGxz9nsc|0^AoCeT
z0ZKBC0+i|?3N#D{QGiknq5vfsL;-~rYwm%Mq5!2CL;*@Njslcw5Cy0R22r5W>=+8v
z8l6$V^-w@L20m>_|D%5|02>ltv;i3oR-~r<zjuPUU{<JhY6WXH7-4y%Lj@qi={#98
z+K{wIl}dHcjg-gmgn(XA_1#Cu+FfnC&&FYeu=Uq48p9eaZL~3Mh&zTg3O!NFn)?pc
zV5$Ssw#(eZ8o;pB*zOM07_8hH4ZIR+oMX;6i@XEW2*iFncY}Ygxyd;1pynOc++rNx
zft?OE?{2~#*^!!t?Df;u6;q(BUawjlV${`C)|;ralrF-Yw<J|sx?On{4y*gtLS5xz
z&$?4RYcrp=50IZ`@J#5l+3nZ`dbkv2D>2W#C{m`+w=|E`=qYQQPxNm-L*OJ++ATQ2
zMT#WxFz*l*q=$bqF$5c~7^16?1VNHro`smWzooBek2$Up)zc1UF2JK)>|~kB9Jb7M
z^yTqdGyNB(W4l_E;rN+bnQ>Pm;f!&>sfG$x*GXpua&ahFmpQ(L1{8)>ua`1%#OGcb
zA9UFch;P~1a;G!7LmCR;5&EtT_Yx^E3fN@1@;q}celLF#$7+)YW!BYTwhE_W_mDI<
zO8}*5f7Z=&x(;)5NQlLXSD&)c)-uD@p{H!dpNrDk#=YV!JnGC2nw>A{%r<JKZ*$5p
zQPZ#uZFh_c%$Ct7CFIc{bB1ixDYQ-DExTPEYQr;ITB!4+05Y2d0BLNO3?w@SPo2Dc
zzTzbWv}k`i&oz`zV?b)7PNNAV7d)x>HjNzq7f@#3m0O0g^q(i&R0+DGEMBr9cg<9X
zJQWj{M!Tiozu{2cV&<*)ZDL+E%?<>Qqc1E>0aIpfU+ltD4r$0k+Fe=nVKd0Vs&)<~
zOmFxG1{C0;rHb~TAp^I|^UFavT7@jZm-K|=tbKp7wp+N%WGPD5f@^vxA=;9iAKXhc
z1O7DKZL{Y3mJ$w@3$u`r`Z$d}_b1dfJrAk#RY|YH;wnrp&C^vUb@~3yp402=F<W3L
zZsuT}z@eXRxp?%wp?o8$=)1{kb(6cFTu8yN?rRpBvb2RDIWTQ?5y$fd+&Q^|_c^Cd
zcpHBw=4pPK@6s?Y2lL)XBsxz;NQK)xXPKVD!Zd;-fda_U3LFV@bJYa*Te_>sN1`iE
z<ITJ?KbpgOQhk{`-xvh0bXFz6-^x8=4*3*7;;8doran$mnu@lJOe>#>rk*X$O6Tv(
z)p)NL=M0TDX0jKdV+@DcvtlGBil7C}Mtpw{4Q?#o=Ga=t@+dIp3zjF*=Y+_Sf@H%N
zG37$RwS0o#<kuU<3>h~SMazv~U^hJkhb4(SG}qVoEGC}-t8+|95)Hvc=e*L{H>E<+
zOEkv$KOk=VT$iqzTVnjBxV>ijc?QXaC2=5=XH^;O#Md-)HT2t*9j)4*6AX-6r2l`S
zb=;~1rp5Yg#?ly2K3)`@$&9B4M3&Ng`pYz7p|m8(S+LbGQciv4SWd%&q-V;iqWS3s
z+$r=?ZCLIS=fb(q;3?Vac(SUO*(*y{F_?iVT{S^+xinp^+ZX9-kw2e<E)9b@xRp$<
z;*#?jpu6dA$SaBzFq1K}J*Z_Md{lo{uC2<dH7}a(_010dq-CRGmVc@8)ZF$cjfb$p
zDO+wO7RQc9F(U7%cb$X(>|I%lX>NI@uaR_F`L?8I{|#UP(O(r@v_8+pZHyk%SVXpQ
z21UNPxQ(NmrtV1*naG_$aFPZuR~l`U5oQ{r&)W!v@zQd)*7aS92$D7lCxCwsnyXHC
zmd47Qa}$qMO%$}tQ>4=@ax-l87eZ{<J@RpX%gXLAgoHCw;Pucigv8Ai8y|HSX7v|B
zJWQ4I@A-j{H23frzOe4d+~4X4Lh?FYIG0418RCJ8zwI*E2GmSR(+nhaR;gMm^|G@~
z=d0x0kuCwB^rYsV9o51_n|ptjdD8q=OsVR8hg=R~WqQlg$v4W|+cjLXP{RRCqgSO0
z$D~@1^p_IreME~1q~-II+BcH8-AJp!R%Pt4xSP>+-NU9Ls>9q?L&KbNQe@6_e}#s|
z&fN2zV@-`Wm$eSZ05ttcs0^pH4Opc~wlq|RW!`XkbD1wv+3Q96$_anw#!+gQ?yQhD
z+j5=!=V~gPh492~)7)PYOx@BJCQ0|_^&Hby&57*xbj!Z@t2WS5Ycfx|Gq*F>&9YAe
zQ&umyJkN!L1{x;qpQte@u@IiEo}_&$CArb2KhkvYn5JqDM>SPbSLw5y6Y-5XZfXci
z503?TJNkfzl4t(GS`vSFuk$nlPANPoJ8s8)s2=sGW@>8PpVOu$gk_jsFGhQhz9KLe
z_ew2rh(u(Wa8uG_h32qk8cqi_(@>WZN;bzG&AxV|z?Hkji0@K6ZaR<G(+)K4o4L3>
zq@9}6QSH>!rTtYkBzAm8GU;?X()-kk1^UB}t6H&5YwtEU7LR}WP0j77W@>7^Kd)WC
zX-0gP8ZxT-eZVyhJ<s`1i6$v0QuM@%IHs+d^HFWp)Ouy!Iz4YXs;y?kcd0E14@%bl
zlpFZ8`B<NkJEWDG!%?l&)Ovh=+qoK!=;dPM-=Q6n!`(qc$}<7g@9mToUyo^`8R?)V
zYHC={WBVK%X2m(G9pHVsgS1R?S&3+Q>SS2pE2A9{cufs<;{?o>&@BU4tmw_8zS{Q{
z82R8(*SGnt?EL<JnM&K5m+`bT69P0im*7h<B>^#)uP#C{m&m#`7=IGC$;4x)ty7OC
zBg+?CuB_9XPPX&!`+#?pbh)!F$s12Qb%fpHk`F=P!OLz<nuj>lq&kp@3pOw=Ae14+
z_ER0Q5Cn^ZYH~pbtHlKyiuhnFL&3kA@sHykZeT8D{8HmUu?@z5h+})6AgRx7JWxtL
zuH=D~e69j0R@*%rXn!?JDFagjfge~%utXI|$)BhRQfX`Cfs^>mRUw%_QR={%fG1-+
ztd2KF%wpW0#jobO$`Mc0HjZeJNOceQ1d>$$H4Q>OUyfxE$>vgpgbm_}I7kxqdl{1X
ze3(jBs}b(C)iN=pwi;oUQVeS_o~`EdF+t_j<}*my=i@Xaj(@INOpv%XNrogov--Fs
zP1kuyxgllT%i6}_2rZJYQHN9{`4f<8wRSq#9u~uCWADZ9!~uKvTR;NT>=K4iZ7axt
zm9t9;lC|~KZKLerUKw)s<5#F5H#^CKdo?$y4jPN^F~NwKeUFI-mF)V40jL`QB!P*M
zV^3*NYZ^#-(0@`gc7ds|9p7WlgHd}n!xp)?ZgC`*+OvRRIr|#94yMHsz}&33ePtzB
zZ~N7)HWpvQ>QI;x)+iAs?N*fw4JECNnunt8T#CTXldp?OaeQqYfUi!?VZfYS!d7|p
zteYkI8l|?)U=24JZNF9+R3W!2H~8wFkfCgf(M?Fjoqva;pB9WCVruF)7kvqZZ&Gxf
zVm^ON9l62=3EGyu&p#i3`t58setCL2UyhG1o-Y0B`{UF94&ztz#k1MsNrd~ygYmoZ
z{&?>Rtu7wUj+Rd_Stytv=N<%qnK%^aQ$+$Z9gJTNnEuiD?fj?t`1SBvDiP|IgulS4
zZX1940)PLT1VqjfO2TGq2?}4yDj^l#0>M$UiNsh^f>9F@PX{7V4eoXwL}Cjm*%k;&
zWMW`Q6Sxj*hU-m$f;r%HC34_Khb2_doe+h}J2U}uj#(H4TB8-qIvE78BrE|;ODQmj
zgKG`2zfYxy^=YtjAUBDULSuV_B68hE0Zky-D1Z1YgRSATv0kG1`bl-Nj>E(8@z4Jm
zR&zU}&~SQja&qvFQHD|A+p0lv1UDF&W3U^IWQu`6Vpk)F@!tGozIb$Ycr+Wt9~j05
zhs)(`aXRqo@|(BIZyzlWm$QLZH}>bJORTXE#s?9sR@pBeb(5?vJS=-8YZu^4I7;=4
zFMqxmA1vlak7mm!<AbmF$Dd|zmj~BQ?7`vl833IwXQ#_^51(yCoXq+BVsSJ(U&UL$
z^~3Dh@!?nVx9ubq42YP1fae!@(jT-os#(DKDm!S)?>%Yxfp4Rw2P>iEE@&653+=&!
zO`F<nM%eQ}Yct}KgA+E81w$+Z=ZA#&fPdf=B&UPD4YM4-yqx}Mc|1Rbef{&{egF5_
zi{<j{{4eA9^5W$9@Z|9Hd6+Mrk55nL&qt0-W2EtHF*};SIy*j@{j$7R&KDRr!UXqz
z?+G79Et&#h>m79h+Tx+|4?ul>I-j2%|MF}$JG&!VQV=GBCxTljNCucoh%FTWkAHBz
zJ=*2*s~7W=-@^0b<%^4_;dnkia*+2{$_+11#`ERj)05d9@p1&QAuea*fIvPp8}VpU
zI3TgRJ>EBO-h|)g7t4#Mvj7Vi-yANFUi|I#Ul;%S>iOc|KmX(V?|#0cW-SHJs&fWX
zIGk7n1>+i|n9FoP-#mW!!<)Z9`+w%;<8()ZMT{V14an_-L8ZuQ12Sr4_=Lau3Ej*I
zx5i7t1;^}Q`D+6@BWtnX`P_#`BBKqea)XhcIly~AJP_qd^YcF-99$#)jo5I@2$17;
z1gN)&0GHeM>g@e-c(Sc;1iy{2h?;~F>-r-3h22Rh_IssxVR-wn=%r{6i+^5>wvgww
z=yu4vWg6v{MFvwOB(bQZ#5uYdCU|8>5>=5yqWUxx?jRb4SL{gTZa}ffnGBVh6Ik~2
zH6bKm)1JjqxdR`pLxGdl%ZpcMe||Q5J^paGeDUMi@;;7Z@hgnQ1H<w8`NiyfBq7=0
z1TXvF%L&&9T>5W8md0{}k$)U(Kp6c0u;P!&^fIL_lP124pObtlKYv%od+V*&?XBCL
zwq8D9Yu5${=TN=FR5S?XPpqHEKR$g4f5jvBUnTEPkiJWE7y*6Vd!GV1Zz+)FE(+xI
zX>8|~7NYBOyS71HDeRVfuPaTh_@37#j`#7_Lj8E+7gU%P5n8f_=zqjt3UU!0L>2b7
z-AGVnDk!*xMJrK4!S(S)JX;M(X%Nxid?X@>LaQnahq?s<ArCNWl|)%3T9tCA@FR$J
z<}GdzxO$s6mq6YOt}!5uynXmNT)aJg-BQuBXTKs7MP(BuB@4LRh%>rojCcC}^#ND8
zXS%^CkWr8-?-`jX2Y*H3<$Z{xHySA(;cl=vBV#B@ZZMKiyT+UBn#d1js+hTv=Wfs+
zNR{f^bN8VYoL*dCDE75Vy~cg>cM5qOKsEVAe}*vuub)BP$WhkA%->wB!^XaEsWN?R
zrmegchLr6tBqMx9GSaz4>RL##3)BVfLeD@`>cYA)<GuN{aDRD6yU=+o_qB3gOZ!^d
z*Rnpw`k2zkls=|(Jf-6)ef8cb%DqvPd!s0qK6B|a*FJOYGuJ+Im5*my@{W13K{JI)
zp7Uy8tbiIZfY5ey&?QEA3AAF&5^J4+X)i&DH)-8qGnkGVsGN4ROO3=#R1BqiGSurf
zAqbF#H?2J@_J78ZgTk?HH?O^E%=kaD^_Wm}m`(_S5cuH6j8lNuYpcb#Lk<ohCNmgG
za(y6$mI*OLa3PHLQV~wOynPy2RuQO;Emleb->lPt@U~opQ(bBzN<d+^Z4%Mc-%FYl
zu~H;6rwLIg*xs&C!104o-bR5eGI9aqDAHljZkq5sKz~+&DPMt5grXhE5M{HPqQOb3
zfihuB6C8+JfmKEDMOKpQs_O&Zf}-K8X~%&uTmm5)Y#9m0If-p{6seYA?jQzd^>=ST
z#dX_Bt!VKV<_QT~25TiP6M{h1W69s#*AQ2jkSl&-43_ZsN)e7pcCeKxq7A1A+aym-
z$tJ|0-hVoGJtQbV8YZ!AMZgXVl~_O~h|*S2UrmVuc_SSE3Wa(De>;Fh;ZlgRB-D!;
z^}>{9DJ4`?LV~(akij^nyo)DT&EGc0stM(Uz&i4TC>T|eK@;Fk@P&iQJ5B>$nF_*Z
zRwPe|B$lUm4=ouL8;X`fL3q_eLz?5AI|PSq5PyXD4rXBRrkh7BHX#PBff2Ym2Gmo9
zw_gEUMlz37+C*>|#8#%%L$<*FS_yT7w-OOZY{xeNiI7K_A&60qwE&S$h#}?RpT~3z
zh$ALD^Z;uplt}%hJXDgXPKW`;GbMk$2AdQwqEpbiBX}wDDM~DfGm;JATqXp;grdzU
z0)P1>1B7tU9Un*Fw`!{Xl@=#tcQH+A5#DWeO?Xq=zr^B*<JjREDYAdpRU=#jRi9^?
z)1klyS2iKux$`f`V3@We)kgjRL%<)TchSolixZ+KI0)3EIP><RyLH}oxCi_br4iVn
zRCtGfOf;pBf`3*#QT;_0;#dJ+&<<UIDSv?gRKd02*os=75QV3A(b_jm{^1ough%!*
z6bKfIpoQuSgqfrw&P@m+QD6@p(LbPaPeEv>vZMGx;GpFB3|IJvl?k~5-iO7VE3|#h
zg|r2>DLbs9VDpG_Kv{^Q+LR#Rp5*BSG}0<D_m6Gf7G&5(W6W7t7<J=>@e64+l7Cvc
z@(vVWW&12DavAu87saSKR1iI;!+<OWnM}Dd2>eBswOs^OwPfB!M=|Uld0y+ldJB!q
z4L)Gn9JOw=A$TJ!{o$_`Hu_{6jkb+G*+vI#qiJSdDO29nw%h2(U2ofIn@vCVy_Z_t
z>bhyUOV`chvg@XC-KNs0U9c{cE?2DERQ~1ZTA(g)7o-cR3w`F?f9jn%&6hv*MSEvX
zduL92XHI)(PJ3q#F}n+WX6-X;pILR<wpmK;Lu8#(bSE&o^=o5lyPev$jj3&WYPbK|
zwr$(CZQHi{&3E4Qu5)qL%F0!en_MJM_RepM|BP;SnA(enOQL_F@Ru*eiT=&5@_Gau
z?DXsYe+3B*m@Oq&851>SW0+MP#QaYR>j*20UJm&uCS}ub%*YM!&-#cbJg_IwGmsbT
zt=kQ}KCBD%7chUGml}P{3q)j`1=utsSCrT1Lka)(nABzl22L|Yq~d%Ph2u7qb`HhG
z#Xe+NAAdcaZBilxG;3?Cd*4ZY%Gd}i5ukJECbm1kM+&vjMQL6Aua4H$Z2G&nhZK*w
zwZcrJLpniQ?l!1NS&^OAGA+xZ?o5*5YK`%5T6my5I4pGDaS%AG^ysy5R$^wOz_CBB
zc6{fvAV-1g1F}l6abv7tY5boH4?3)hR@3@U;iX~6V>~a96%(sguX`#1u50Yc1B?zb
zxc72qQaM*PlS27so03Y{DJPPPofO|u)@lI1pjOux=+{N}Q_7(|Xk~Lzyn=5?T1JwU
z3cbq54LA5yP7Qlky%@#nl-zSeLx`Z_+mz$ZvR&<!0L+zB0dD!?r+LZRa+9Y%&YW||
z_9gF&x~0I)rhSHuOQhz<8u57N;qv(Ck@ni}fBY!=n;bk{_wo;_2XqVcnJn!y*9rg*
zRe_7G`KGiAhPbLp<iLa0m6PLid35ENh=7&DL*;XP&Zz_I8iXpkvC%KFU^ap%H5~}a
z_VUwODK!hgK9m9%jh8c?XTKVC#Re5iwFb0c;`!Kw<^s00S=vj?TJRa>74P_|A@`A1
z5G#Eq>@ct!ok|W5``AdgNo?V|-7~<yd&-C!-g;+gNl*PEUUMCkq=|X49Q|l~++9^z
z5qhF4VQ0AT43yp|0A-jhLrI0C{nU()_k=gev+3&}5O=L0hDXw8fmf2ZuPQCiD9<Q3
zI1hMVjgz`_*gM~)zV&npHYx<aRN{M5BfknowZr`h*dCQruABNF$!>!60~}C+@J0Ki
zerkQ>h}fhdKd4y{d0y0H@M&cDAc0HHI?0fkV$kxiPC0IY&|s7N0_}*EWD~CMV&3r0
zacu5V&{R7-qkUG92=6jiB@gY(F|!N;*0Kr!ceiicL|Q*QveKKODIi`#)@4Jy`(Dts
zVIq9+`1t&~Zh?pXsWmxQFh}Bo==idlm;(8QE}EDCdW-Or!ac$2C_9i-Mjicr;qqeA
z@u6O1nRUPRZk)!!_1O|n1qG@LnnJ^RMo3BwC;X3UZF7*D^ss@QWJe_01_4}?;=8kM
z8D(Z^;H88zEy^s9@|BhHF~Ld<Xqj|Ggoy#ZV1rwT?EOkO9$zbV7V>LXy4eFO($VPB
z*9ddr*)dokZ9N=`ic@y$xS90kq5TD9py0Ry&csdp8!Ct!msQ%IW0V)^AE?^qG1C(}
zx>DXj=5*LyX;@xC_C(gkg0E6M^GQ}#zf#OH`XHCiI=m)<9)q#lL2A$(VCsTSag1E7
zMig3l$pb4Ym2^xAtgOjA72P_jUfbAnRmT_eFDQI8#s!$yVYBRn+z6ee=8&scgSk1z
zCk70Ggo;O)Y?!B^amJuR_`LH^^^-~=|1KueQ#+&iHo*d%I`t#Z3U%JJaL1x52joNf
z(`C<LUOT<Id+D`Jc*y$(V7No_GB*wzT#Zp=7*5g6j3*i;G&+Dgj7WeeK3dlT1j)BV
zg=P8OW!`@{U{U09KDu<7X*YD62MC(2+sLmO<}Zoj_egD^BcDN-uS;xfI^ytVC>c;+
zuB!nE)W!0T5{=(08ta}OxC|x_luRUfAK25enYkQVHaN0%+Y_4&Aag5_G!u?Lh}217
zt29BNpW`v27qQ$YLFCNW%SmBY>zY~9zi}%(P~46cDM_bz-&HkTsAN`2tAdb;{5J@h
z8&vfon%k+0<|UYhA)>C$0bxkvgRIP-AjtNqIa-B@_%oA8EWs%^dKNOk1kdb(eZt`}
zkY~4A1?>knFG9E`K*}<1{`Zh(6v9?>!E|(OHN>+A;8bC~BIJhlXw&~KbDH9Y;to^8
zjX;NK`tfAoF|gybg|`f*z%E62`w_$pMQYzFM1eM29I)2C>78j-wH$*(d<B!y5=>(e
zn0J)o5rIc`4RWR(548g>hZ)d~5`zu*$>9{o&5e&DgJpUNARm-hqhyyY)ktQBjy!;A
z8lMnJwijjPfQF;xEE5COfclcck0h-rKw91~!H09hrzv0^hvA#sXeQDvHDS=2UAZ^P
zebL%C?9`7Fgk|3HR&lAY^Bv>KSBB!Aro_;qFf?y6_F(A<B1P3>(k&m(d-K4c2rqqr
zv(4=S9B(E7;H0CYTIQoDkD`Wg77UujIFHV&4!jq+kD1Q5tX!J(TLW;<+@)XeXd{{%
zvK?k)q1IGHI{e+P+0Qh(oWfK+Amq$bzjkAFak_LpE8d_5jwL6rP+WYt$`EQ=V7*V^
zmB`~tHAo9|>Q{&K=o}?9ENd5xR|bN8ZV5!-(KIRn`N5v!{QDLG=AglmG7_&`I>}cN
z`V9NRdfXz)Z@#xQObs;bIC3c~<|g0*Gf$~L=6Y#(ODnc(={NbyI;MKAJsZ(I2HC2e
zz#TX%E+PLQR59LDcAht@UzT}`2oFe`yK}TjRZi3CIoopr;G%gv82m*c$YkAj!Qmh)
zj%PuDoaW<vh=LUAxXyJO;G|g|c$SU;s=%ykL0BFst>Apqq<Vs;N&1RBkun~gO6Ibc
zf9^W=8;$Jv6V+=D&?hQ({j=Yo?1zH|y7%hT>apbBT|UQ)@R-%ualIDKk}dw@GtWhr
zRUOBsRkIu|+8<s|f3$4=CB~`e)|1o73qj-oCP79Lc;K@8S{(T^Yy0M<`MT0&1#gA&
z`pL^{)*|uSxKyHy;=cyn$=<x&E~@;|$Ai4n4)3QYX|zZt^fn6C@zL{Hx&lL@JLz+<
zC4M@NbPTw3fH~!yCF%^`$t_m8nvn5D*`9TceIcVE&n=Hu@3kMAkP9_hd5J>Q9<-qW
z-jJ#89qFPvCOwGGigeVU76FTCsSGG&+~dYrASj%rFCFI3Y<9aC#m)UygO7o`E)d04
zvCoy>chhFbOYAnmKiy@V)c$^9Q#ms$juX_$&53=S6R^NWUhH!eo>!&=*+0+&Bs^Ht
zTc@{NMZy7tv%rfM5x5>FdLY{TO(E%k9cI(B-Aaw*^UW2uiO{&^>=kO4GCAT@)B|mJ
z`{^~)M<EDi6e_K{=R7rQ(Wy0#l&3?CsDub_iqq?&CA2_LY8Rs;72XA|erP7<u;DeI
zC3|!n-8v@;%%RqVWWHyy#HUkk7fu1M^-aU0HkkXwB&!e4%m*aA4Vt8j?H3rpI_}Kp
zeL^5X<F(UK<)@5(^EGiS$;I+O4d3HbUKG9%j*V;4JZ)^Z^%~0pDO0R;24svrcfa<#
zj}xf8AA72@_TI5#8+!j>#;rgkwOjTCqGtYm5E<Qist$5t<wOD@<2|W_4}cLGIfNi*
zQ0KU|lAGNAqqLV2>$={sGhPM|_r@+=rF<V&WcD#0z93xj<J$;{^;Cb-)XWNHp|2st
zd^-z*7{8-%?3}n$)Nd*=7RU&_S_9s{w7;M;cFgtT!R&9-y7(kytJ><`W_tee(H5C(
znY@~a*_}NI-S6I$VL}7yfu@PHA=a>9*AS8LxSp80?dZKP_S(Bo+>b4NgHZSTD|Vq~
zYU3fohCR#tis?m2jHvdb4F}Hg)p=Q7)H+yd`00k^d1>^^!cM~Um2CC4qjs-dJBCq8
ziRy!OD==gAXrKppl$|)muN?yggp)mGP7MSl#czR?1498ZmtnSW$`>zWGgzl(f$rlc
zL;;b1HDz{z6&K)g`wOZIpN=2$5GVX6e_v2MKpf1>uJseYyN99(uZ4uYB9U>EM-TGa
z!ix#h4dNc|up{&!p=FeWr)pB%Hc$KF;njFfGbwxbh3JnE|97hwZ>i<xLt6hY-+Wio
z38oWThHqmHA)=~8lL(rERY#${@4+kY#cQ&thNR^lnrFa;((tBIACg}~158*G@A8Yu
zGMnn7qs3XeE-V-0-;Z?D-^Rz4qbp)deAkqTb;jInZ(SB_QP|`cDXxR_Cnqyc>F2CH
zpgV!-%p^8Iln8Q=@_nje;-ZGlAhJZKf__KnaA|vpJ8k%1$;*_x#V%q2Cq!W^HPCJS
zE2sIo_mzN3Y`T|vJ{JYXD2u$E)S$J*SiP(BCa21wK3kf1(l{9XtQ(>@fp)$8v9oye
z65_ZF-U?BeW3awq1VxF$jv@)}r1_uUH#wxD&u~g5YJbLf**sy|pw>_E@1Y7a?yVsP
zwt5>z9e0L_aQX~lR*B<mCy+`cG%Wz3NgMO1ocI77B?<A+I&qmeiL3cgbtdrmb4nZJ
zTo4owidu;AHE~@T73g50cRdm*602aK8;?M!u*M0`8DPyf53t6?fEs~uFJ^)yyrwZ1
zA;K-A1gXOM-;y6K*v%CLrauwR!o%&^uX{evTWsABM23j%R=JS1IypB+vT+$;0`Oev
z_%Hyd!atJw%Wq*F`>krf)AWhXi+n3KKBxw(rh^h`rc_JUdt~!SlH~Vxq$v(?JGj2Y
z$r2fX4O89)h$*jNSFCT^+$4^9&*t%);l&tu^P7P*@9)MA@MmO3pt_}_>EaU0xM|+*
z*>X(0mUCY{QD$<1>PC>I3i})G^E4d~Eu8>TmDd$@udMn;F`Wv;NfIgv#hlpt^}+ib
z9=Sc5&8@Iotg-Wrg#nJ%RhN#mwTed4eyYvbf*ttewIc1&MLC>t2Tc6@8_CY*LA#X~
zeHw&G@%m5y&=K6I0T4x=t(79{^!Aq3jNW%+Rx7&Wc?<e=;@vcHd?^eG+R_`X$;sNZ
zr}0SAh1nnUFWn>G2@(&IoJW~zR<D{PvN-f6zNI50nqj2OSzBC&HxY-t;5rPyY+~@J
ze5;kZG^b@?r5B%t1HlVAK9w6FsDe%`p2$w0=NO)DQU>%0&Rg04gEwPOsq5lGN%0~D
zp-hQiVNC#TYx%v(Y5)o5&T;YnAHj+J-<V%uC@Bp)|D27REe^Y^vXEdpdPoKTV{rUO
z6lUxaBu)7|VTA)`PBEB)rvPL~$NUSZ2zq%#b4{~eHRgQ<f&=vtk@E%nMMEDCrAcv5
z$Y3~$Ige@g?cHtof+u-REna6XDG;$W_4^`T#d(@ivEaD<$Yyb~IUhT!Te4~nGb38C
z;Bcx*q`<mKlf~GoKr}#a^wLOcOeOzFi?ZO`R(S=J3Bt9z3A=!M=N<4_wo=vj&?2)n
zv-4aPiIniMS}0MX^6*=+K#w-#H1sV^LL1xhmN_gZia};Q^~)=P$N+XGZ_TJvm5rA7
zopnu_6u-np({Xz*!z0Xxr*`?_??$m~B4@kpI%8&X9jRmDbw(2Mm&cY2P35@_v(25g
z6C=HWG(m<+hU4LsW(L49aZ2o;;$WR!#Q-V2Pcy=X9YISriIbH)w0}QI_h~uk(Iy$8
zy0v){48PUcgwKDpfh5*6DR;zBZn9=+VJ$w<t}@k-W3xmvJkeeE>A2p$lx$?I>vo7#
z)orl)>0(_q>weo_N5u#u#%9AP)f!-Cfbf~8lJ#YMU}45Tq5-(SRBOp*y`;9eD4O>1
zzCgBJ9$#&r`b+kAYhVuhW6{XQd~CU+UU{<n$^JVz#p^pcj}*6V2uSlp)8N%9rMMFl
zR*b)0-C7|7cEUgE*u}@I`>3d52-gYA*(X@~m}g%Qm47PV@$yx9U<Zd<r9NnZZGsu9
zA5OL9kc~Ta_7XsXk1kFDhZIo9S(KQbHc4z~JjjJ3s4$IVJ*m2B#9&s7j4)WzxQ^RP
zmTto*2)A)EtQB+YMcPgL>!b#oS>iVZ|J&i?6u_sI)7!<<(-q)xf53lpu%*}C@%7^J
z{&{`W!cL!KIxpzs^1g63Q$nAe{E$)eS9g<tXn3)%TnZqY)@0CIi1#v|ECD|wmtzC>
zbNINl#qL@{-?Ag!L@XD`)yJ$XP6>ET<hP2XNT>Rs;w2{<Eq5a>uZh(EV<;^_Utrz|
zr)Mr=>aeM$DTh~MhNf7d#+F!~Kj`t7<X(1W8KXMzB^jGiD!fQ2^IczSJqc+%0Twzk
zmN1vAxCIczrYt$U8BZ%3PF_^F=VBr<V@&42+^3UCrjQgQX8wl_gq^Dc>5Q>tv!<mW
zDi5JO&KG7blKLb_>=&+_HUA}q7+P1oK`eGTO18#e2d1QB9{l9hAQ;1l&}m~COAixB
zDlt~vNn>+3z`x}p5@=5mJSm*L9t<d%>~3_WPZ01uG&)TO=TaFnhr#i?3=0yqvX>E<
zjQJ0PpQ{!yFP4zpQ+AQFywg}P%O5|m$W({ik<u~{NYYAUmn$?<W;h^OLJq9+LQ~kN
z(n6d0rvRiXvWqVd%1Ednpj7l0LWW}MD!YA<Of({AkU?t*TZgQcD$Z7+Y$kJYJB7cn
zQ49bg!dEphn0~_QvbQHiibN&gZD1<{QLj`Q79mg`b?7EzLDwqMW6seuVH_N>Ld|KA
z7>%>ELn&xocrVj^64gLb@2ncGL?{?wBwV+86Fqa`>cQWz3IYnygjQvGA*KC6+;6}T
zTmp!0L;cPa1IRGDPo~CvFD|iFy&mxr0aXAPueM%X49z(*N)h{(d^r8a2|sBG)zu=`
ziP`RaRSaz4G-#lOIC+y*<=Q<D?oD;STQFP{&7g1%bN_%tipo&wT8nh2Us&LigiXjC
zu?(duD@mRPKy7?rVyR$Te51*Szpbx`AZdgLhlOx%K@qQgB-ad3GWXV@l6&>vpy&Wu
zSpKvc=HM*EV6Jr5!>M9pg5jsZQ~}b2MVYep?g1O#Z>Wr{m>(AlJKx`kg@@?l*Uuw2
zJph+g;Il-o``D>J3bAfmj@;h%FJD^|<orAU`v!|oeQsVK9{}Lt-X-B-ga6y<XDQ+1
z<b_>}6L5Qa{hZB-Z?Q;`OUyV?fDb@4_3%+LiYm(7X#_fFqT2z}m+JO1s)O-yUvLL4
z&FR>q<kV{y1vk#Mn%UYluuRDlW({iP?@0q)e8?#E$^|}{D#^*l@9FfsJ3RhAJG)DX
zk=xn)bo#lB>hk*h)MKygzxtZG`l^`yIU(xqc6WX`n80AKnFbGdJ*{Ei@d3PiK6K<9
z<L5-MktU4H=-o>@-E!dYdmk*dJu{a4Y%OSfg*Igj#Yhi42j4`sv*UBUbTQIJ#l48F
z;W%bwd&gQyUH6%noJSBNZYSsC`z^~rx`QBCN?gVDMV=Q3bMW!-6b2m&nmT3G7BOTD
zYdNspI@yTGRLNgP^ALcnXao8jq3fOIw77~9s@&+-C~gnNgVjC{J32NB$=>qY3LGG8
zJ6OJ(0+ook9jGAdr3;Gtj9#uUv!XkRwO-Qmv4g_<h<4>ucNHE3mJbVLefLO02%E)<
zMSQD$|AtBm96ky)QIL6rOTiC@18OYDIKj4u>eFN|NlDQ$72vnfHUYpbiL!B+3brui
z8HnM`qJFf3U|J2gW4^O!VnCa(B#KO;k}{C}C6HZ%;XYB>K))$x-$2LM<F8?K!vdum
zNS8^l89<&`12<!clR(kn>W*tH1XQmJ!@$P`fhRHx)FIR2L*s%0&!zoEUcJPY_HL-H
zYqU~R`fniEz}J($UI2!E*mdb`6Oq3N%>$IMhT^yxkF-Dp-w=Fiw{7@&uUcKx6!^1u
zJ)UJPiCtgl>OT;z-*B&fCHrkZFsnA2J_sFGOhfbown_k4myI-#OaoW#+2*58_VMl5
zuDjWh|7_a&_NnKA|ENN556g4h1fRp2INisn2|?Q>NeaOGus%lnInIPY<A1x)0Slmg
zo}oA?<w9{(lzar~cWCU`tK45!UFh|BgIJ;#B+_~c>#J40an3Mr5<d2q`+-}|?=OtK
z{Hl$bPlnxv=Pq@9pok{R9{Z1QZgO9xEIZf^(C%j<coTCYQaC`s#F5VC1`#d3rx$>!
z*-IdK^I#6VdxezH)h&pNF~^Kw)`79yFy)U0`5?cLmiP==q;RFDDO^Yj$wO;P^L5wY
z$oh@~NWvk$lP>+m2kYeMZSvsVN%7&pmJbzz!9!z7R`U8f<wHuP3z}tnqk=a4STavQ
z8mOvq#zD(J>nQeZ2WW_Muk<aFjQap0e%z3g6jXcqA&)$n@-{{W@eHszr-;;y)OZKI
z^lLHO!6a1Nbe7HrzV#iF{z*ExDC2P+9dyi12wsPi(zn1<(h_3<NF1h_+Zd@I43QL=
z&UVJA<e@|~;&<3Qj2YF;Labi;%dJ34qlgR+zi_;Bi$@CvDq(1*(76J|z}Wzn+_g%|
zlA97hqSY&T@r62la1+TY24=8E0j*}hbSJK-mO(p9_BMl<_R-vgrBEN0_VRVJp=_fu
z@=e#qkPfbYLzd_ck?S(V1uzUr5-8EJ_l&5{9NnaBLWH;wngS?K*0pdGQmeHgMcY}u
zYk@Jzu<*=-D@Cd~Q-in(b-e(2TEl$;D!c^6{o|thxw`^ww<9dq8bHURmiJ5JTce!r
zn@zsL=9S>>-S`qIFwGdF<%45bx52eKW+JDf6bRRS3S1ud>*T>W-|Fgx!Tt`TNBvj6
z1V!lMNAa|KG?IvlY_O>0rj`MT;B9ryu`{R<{n1@){S214Zz54|5iSUyOrGQxG?OZM
zy1dt8+9|&GtQgcaZ8~YJwE>t-wp}9fZnf5a8kNlXBX`}rV30ul#RS`YXRoCD=DBMc
zDM=}P_pCU8O4`A$W<4HUw}yd>dZB>%=9g0}6N-=0q@bUc)ZMr8RQ<0NRFqBpaA^Ci
ztNbR{DY_2Rl`j<US7Wui<_fnBmCh@wZ5On^9pNoqeEJ{S%joCt=;v2{G%Jz|2(N1t
z_9OJU5{)BO13|`ZVj4Xc(@1?7uOCRVS6DsO?vn$6vu@=jwV-H91-4_dme$TR^;1w&
zFgk%KiBH~ZHc6e)B)`>IDnt?1a5_lj>U@qRO$HPW%cN^+Y8tG&_jGEtQ{N<EMjRHb
zNK0_E5||Km4WqCO*_xyXah<mHK3?XS3>CX4B#p8aGExP&-np`ikk?M8UoXO4U=}=7
z=k*`}Q6sf*<d(p35E9cj*GVlE-s(ctoHR_f>`^X)5vVa3_ykCj-)bwaI0ib75V3e5
zl=)!iFgdl75haDw%u0JCq-T_6{PFPiL6(EO4-xR#u4=<>Dey&ZU?LUBXyl9Dy2NJm
zK4q+OyZb`?9Y|(7Q;IW?YUos9Oen+<rbA7Do^~OrahYGvx~ZWLVXa1JXpgOiPJ@y4
zqYP1oe`=6_()9h{VWhJ!Bemi6)eEO+HuUijs*kwJ#tLpBO|GdxiyzS$>+kSG=>EQN
zIpY5Xp}xFXV4gkf@_@l&_}=OL>y39ntNH{i&T{o0TKrv3l73vHk#r+7;)-y@9de&V
zD!i2Sf13B<*S+ZHPZ*)n4>)DckmY}Cs6I*#l$38pXiC74CU?vM+pp^{YGn_k)Cr;x
zdSip$=PX2tlp&WkRLNNG!d7CnqZ6~&M2WHFOO;F^-HiO;2XE#~$fnMZ^}FkZ<JV?)
zKIu!i2AIgx?QSOlb`Ecl62U4i(B8(t>np973D^^%1?NIo&XjmM;T_s9m5(eiX=Yzj
zpe^v#8#n++)NCUM!D;<n2gi*oROlwZ;JxqfK7*f&jXyZqi3gJwKTAIr0B^62&n^p<
zrA;P-x{rg6;T-{J8BIZ`R3f0ApUZA*AKRkjCJtU?e$<<8z*TzOib(Q-2>Oa<Svbw^
zAfbyU_wTPN33Q)bZ6tl31?T(;&=8ozhqeAfkwbt;hb5T@RS_X&2}Mxe3OxX<sRBu{
zLqa)T(o8E%{$gQ<O@?}HT9u1fkx8d8LeVMB4e^ZqoCU$6JIs;CV@3#*-SY<P6BC_)
zww$J{d?KF*Mg%!9OwK3U2N_=2VmWtW7V1=WpR65@$wt4iJE9i!^AM;sUidXD_H=Y~
z@eaVmRECA!@w;AgAGhA4nS)fBjc5`|ITG4U=3sl3AR&7F2#-v8u@2k)%gH_fIxiK8
zW4>RMF2I$P)_1s<X5Cnba(#oO?684^`tVA^gTl`J_snI2-+PR2B{o)l&cELgN?j2p
zyvX@MUFZ<7q$b=GiOiaEK{leM4ce&r5LzkfFaKx<bEj5}J8oo55Ghwe9PmxgA+Y5M
z{OVR(VSZqH4;ffvJ2$%~0%n<hoEUTk4LhDKraVrKoJBUw_*|xfe541Ilxg<0hQn5*
z)v<TjVa!-HqAX`}e9k&DGTc){SUq5&?)7GA#X^(VZNC?q23b?9$)*{%2)X{<7^K1)
za+%oy%s!qHzxO7A(Y+zK;7C?F_L(g8XGgs2Xtt9%vP@G2CFAPFBzrv+oWxKkANx4X
zthI<!q(yQ*;3^Qh6x~h>wup=BH|624L>W@5j9c+t)SJqaG`kYpS%=>}Zw=Oof=CNm
zqp^{LxiC7!tFZ%WV(Urg<b<E3V7CZe<o&1td@dtxo5>7PQdls1z?c_F$wrA?1mKw#
zu;Y(1_qnMqHwC>Zjr#b4f=%;=FkVf?mt?0e2dafpC^5Y{RgZEDJ0U8>L+l?=l}l2t
z>#ux~s3J7u;`)`Un$e6=L6>Ruo<|dY0j)_-I;c?LjMa}uKV;xfQWKPRlqf>u%D=Y&
zh%?(EY5w7kuTYaqQoxR9XDH9HRn7$E$q&d=-At^Dy?5eZ&z5;dcd{q=h8o=ImaIf9
zwB;e<^pfMXr7$BBZMll=Sp`yIYUOKt3lXp!CXgST=>Aqu4<K_G3-H=Jn<DA6-<|7`
zg^#FJ#np9_cX95EHe%><Ot#{sH@rvyAlYci?6;pRD5z+-$8Z}<TDu^+CUlO4!sP_d
z1G+QuoxBR50kBe&b{ODA*;dngZ&}g7p+5!PTuT&_uMpVt;=S@wa10VRNQBr;pIigk
zv^`Ab`OZi}y1Cj`A!suyy)eJ+rW}*R_F?chBr@&S(C7C0_&Q`ZE*qyVle(S(-wT8c
zzqWsGA2+{b!2kzYklFGfKxuGh!@YTVQ}tA%XBi`HD?{y3{_xdQ=A!420?S{Kz4xm>
z>7pS<Npv$EdDVR0ha5`v1YVQl=AR%CnWJ8V(K{c7O~TxGr=AZ!UvElkrs7>^Bb#bn
zu$C6>kJgDQE-99rkL}H=4$6xG=5dtXX?260)&I`)Y6%i>mk@ucW{Am-u;FPI)UMpT
zzwYxzhg-_UwEM6ofBAxfDm_bEAVYL*=eF-aQ^xNsZ!0oQJ!nKs&+#E+Q}-~VZd+vS
zd29f(tJ}*N8(%~!3|&SGVUGXqk<LG}oIDIWv2&Vuj6XU~06$Z9)5E|65R?^bI$>2k
z(Ay9F*X`Khvxp;1rzI-R#VIc3ot&eKn8UC1)Y{@3vzn43+UA{qQrP-vxYMxOeABR0
zW7GLhEFm|2jtO&I9R9gv5gPCK@2IQ+__lqK`UiF+bYQ$lVNx5`=ve6vzy;6w!H-(K
z`7J&Cw9pgF3Un^C(a0Hq*36~Fnf{24ygvIpo#JCLMR#>-l3PndqC2=M8>GWaBqx|P
z5+AK#Kg9P<uYR}7B9*DVhct?u+HkvMd(87DZ`o!-Em>PWCX$sb*2%)l?k@w=7b4y{
zLqwGghL}^#H6;^gNj*I-N5-<6<%D-bn%&@cpW>sM3K`b&0o4NqrM>3*4D4?6aFT&c
zG;xRJn4a7*``FIA*+Vq;M3104JpYXNjGqk3kWwkzGE(s$eDTOLzBBk_8OV+|0SLwS
zWAFKi45(tnj?$hVjKyUTy#JhY94Q8N@F@Qt`Qm=)DN@otSu+4!-<>``cNeZq%xXS9
zZ+Bb#+bYdo#v?kFEB#WfpRW@gH{B^IIj0jf^op&!bU*3A-wp~agBJolKUeox9sHNJ
z%iIQi+LcN|`E(gkkth-Kv%lj~MGUuZ&1N3!&i;CZ0kuNsUkGnDD?H}W;$(>7B#85!
zoJ0e~pWX|l2Uh@2{PSVd0_>4<APTZlWKd9i@Rz9Sn$y(Sf!h&spZrbv;d50|2>n&i
zIelTE1+Xon&`!U=c0c&NLn~0iy48^~pNokYrNAu0kSGOyPwoI{NG~c<341GCk^mUU
z4N`_-jPd5e_u`6uUqP7|`a*xegT@Ea(aZFM9E$U(=1T#Z@l7M7%)PPv8szzl%BvV1
z2vj2Q?qfpDX9W@Q`z+S3o}^V2G$ID5xM=7$5fV_CHZjl>5xLGC<Lc^M`qD!EOrQpw
zNV()7-ZH9@=kUfw0%sHzCDz~BoQOHuS&19QCYsPjkCfC4s=iMpm60`yU?-c^N%ryd
z(P^0EoVEbNf0G7-#l#WuK{NVT8fEZ?OcuLSQ7K80kELJ^u7pU~6ZN<S*XNa@cvKXi
z<(%coP+`dS5~Xfc5V*D-Hz+md1YPB+F_&<-wy}H6NZ9gBfj>CXDa-2T6BfBJSAtCZ
zP_uyYoGQ7td+>#`X&rE1gUx+AR0U66xRIV2gaN!J5=a;cqPu3FwY#nnvzBk;sbaI5
zwHU(f`jQw3hSF(O(7}l~(Ly+&$_=A)KubvQAj*BoO5Vh+sDZc9f)p3EcND}EBhIVx
zd}Ycd10TRa^bR!HcHsjslc9V_M&X#}Xh<YtX5v&rPazG*=7RzxEWz29#%H2FndiAc
zc!1cPAZEE!(!ZjicnFY1qoulMA-a3}Nf>nRL{wJcO77Uke-Z=6XXCxtVu|F2a7tbc
z2oZCsV6kCTTKk<f<QamJ_cOd%iTF>9Hz8q^H9P?H5wAMPROl@2E;k>2*>@^|S0}B!
zLzilU9=Z9uj7>C_csY`HWmiC{^8U;6jDSkg?y|u9VcbPHTK^k!v2sDuQ?2+oTL~mX
z-+fY}quylwN*q*k$yq5>PK>2mmlP1%q0RHTBA#Nr$fln-kJHbNDd9imkU>C;*M_Gh
zh<m9Xn1m>&<`CzCKE5(_d9Mb)!l5OLG~?;{cw)V_2@=5zyg)TDOKdWqUnbQm!T~Rq
z4-w<sz<pU3K9bu^WGfvDOk@y--D+IL^j9v568Z80bt>#O=BpIO>kT_qqJNwj^>w%n
z?|6cme5&%3W<jOU*`7>Eg2qr8ja6Zo?S&nO3z=?A>meSMq8?MtRLX|ZM&2$Kaami8
zDYb!%cXalc%0%)qIBwV4A1i50@c}b7wdVr+6jKH4c5-93%xY0DYN1?c8C@1N3%D~y
z2$e=9#dhbKuqHf7qjU6%`txFX-;&r3Uhh+C84$v73h)74s5jap_xK;ld|0xqarV6A
z8n76e7)szu@R>vpzQ_cEJK>AgLO4b4GG2dp$Oa!4G+Z9trCtjb-=KrJssS1%9{Fu_
zrjuTBw=jHvoY_`N0+Z+mP92%}-JN^1x!2S-kLJ<m7lKJfNEa;h@xKQ1M{K-nQeQL^
zv?fQ8yC>9bigpC6)iv^>dqx^?E@ZhvW=$a0$I2gNWvt7i?WE~5WjBu-wUz^r54p;v
zh<&zT#RMe%$alt9D$B+_1_7Xo27lJKVlPKjpqL=Wg7=xj-K4rgzGS34{CGVp&(}h0
ze+4GmdMjN#OPVwa#K2z0iH1iFvB;h*VS%9``=hTQZ36i-p;S$35K?g7t@c5F2&i+L
zkmLOVx+urStF%I3MugabM4q4LxjA|zQme)T<jD8o+h2!gdgd+`YXN9OT0E~PRh#fX
zK_2WQ{*Ll%G%mq59!`9@y+qSEjD55rteX>Pig&d9ZAFxQDsY+bzH}`M_TqNnTdukN
z%5^di7}9YOyj}&<zBJq{#JU1&Ld{j-cf>)6xKcX{M&)`?s_X_ftzyV_j}MKAkE!}o
zhU+Q`JL)`>v?0Ya)DK7^fhjv?i2Y&e&@G(DOpY|kQB(w$QA0kjOeZ^5ig%hY;4S(y
z!&xNvz{lP>sFC9+bh}y9MBk<!7lEA5x7MFBRh^z>qL%tG-ZS~x!+`n(l9a-UG!^6*
zwMa^a4ct(TfuLSCAH5^gng;nQr*2<23sH+oujySSxx08vFb?Qit?<cQm{kun(M%>j
zCa);ce{+#-s9Vjif{$fVi_VukzoV>W-2ln;nXoqAFQO;bs35bN-};PE5c1e6s77y(
zgT0m;rn+E{!A-r&?I>e@R67lVfxX*V*~+(9TBy?!rC0u#YQf0}5_`)|h&q%DSuMT)
zyjK-7RmEL;W(34Tw~nMubgL>STx8%*EZ`E}#NC6-B7XT_uuHv36q8o@u<|4UeTe(i
z4bg7~>N$(<2|$fE>Rfd+i>d1junTtZQ09?Co+sJKdo)Swx_AIpw4ap0*?X)3j}g(T
z3?`Eyb($r2n|v%*sGqy5sen4<oU2-$BNr~{k)g7?odUqCFPJ+_kLQI~^ZAf%ok+^K
zU8Y%7q7`2ZIW%{?^A++S$_uo6&XYasb-}FTS17PRwV4+5Hp1k@=EgIl+4f2vtSppe
z!;YPo74sXfP45-YH1cC}tba>zL&wC$JgnC)PdUgEo>a_GO!2jO@xR&5vYU7GI_7jl
z^X$FcTLQ><nW@OX-rA1t@;%~Iz!_Y^D~813F2OBJG#n92H~q5mX$xo{Ou*}mJ=bEW
zTZ%AvV2TxR?u~c&QhLgJw5R=@C)r$F@1X0-<_mmc@{v%<4Dxc`Qst&7Y7;*ge4b;s
zHtEGF(<8(ryhjia2DKuVol%%&uA2kSsQj;8?EpJ`2qthKCWs4Vo8z~N8fnK~&<)#|
zTV}$=iKVQX9IB~Bciv?tQhjRD*NUb#lRX-*SU#F5Kl?tVVl`Ib@4aRsJ!M@%;v+c9
z&Ou%EUu~Qp>Q8-|u$()^73A{v{I+#~5S>iPKh2tCO00WUex637><vv1%=n7gWLU$t
zDu7_yWP^LqDBPCkI<gV>lyb$tqjPASTvNJslnU*)Lu%T0Hzqq>UMS8aaqxrB-fmdV
zv){SeuABUeD|aB(LrsfJto3|%B7iSi^0Oc?R|Vjpd6%1PAsL|}?&su$rhbU5_mk<K
zXWEi3o8w!jn3l|5kD8gDW945?KI;z}V1SjFrhQKMa;FZ)?WG)qQlD*pAE#*l087bi
zv6Gm@w%2G=yBcw1XhUhH)@L%KKl<MEI4w^U4@3b($RyU+#FJ(rrWe*q7CY5U-H_$}
zcEuZ0F&`L{EHD;VPRvo=+>ffDQh9a4`c6bMut9r?#&B^0E#rSaFiJ9!80+>(T>)Jj
z&Ow)>z?t@wiLDRe2%0Bkv2E+aPj7u$;-poL3t0#6&vP{aEBX>~Q5KTL)Xxi14LOj=
zWR)a#nQl&scsJ7R$;SE+cT;SNKN6OXTY6La7Xpk>!n$cS^&|>a@hsw7y{WWfNb8=V
zb|1q*^$W6#r9+VP+7yAwhbxH641k`WnQ98c*mde9WPEhq!HeEJGW<~dN`_pmc)wFL
zD7eBbb1MSYfjV5F=+o6N7?V??X0Oj{P*FbaJ<|mOTOd6zG@44^7C=Cv6vpWwbxau(
zNm~L{!bFa)Io(^+MZ4hjRU_^wtVo^$U2bi+oBm;UvG63kuyoVUJzkq5FkoF~Z`}1}
zuDU|ID3rYSXbOGxB%+@jq|pmr5#cH@*p~Xkj~4xo%9JwmFyyN%^_4hm;tmhtYDj%t
zQM*HmEgE<+v4{iQESC-y_uQ|+Q6T(4PVnk9kEJB{cgDkVmVdewq`;OC@2YgJHr67!
z0NzcPg~cBM9ZHTxkKH|mRDkXd#94a2*H@zA$Rnr`ggx$=o8!)qlP^gXq=j)O;2Pr^
zA{c>BSMqPEGEB4{zBC2a(Z=t@1+)U33yOjJZuc{xEJ=#715Fe~4F%nJ%u*l|ZEWoD
zJfoQWNcIMfIe`U<2c4c#iA;eb@d=IO<CcKW1k~vZXgG|?yp-Z_5Wu8WSKMb*y3p&c
zNvdZ1L?z6>BN{E|WOk=qqPuFCDXeYCX{w~0gf?8KR~^_Y3%C<ciA2spJBDM4Nc23J
zvs5zkJd;Y2geQBRw0U`>G~QvseI%nI-iBVdIKc0MLc%TQJjqDZ6#Z8T2U*Tpb11**
zIVZoF)mFv)Bz@PHDWKU~ok15?*-eTm^Dxugs}tN#ipjc)J1p3G=(D-Ge%(823RE2J
zE9|<0=UXqqZ?1omV_7)mSm3jL!+zG7<S^t-hBaxOt6c=iTzYH&=n}yh;ElY1HgWyc
z@|9US8DL|Ym@~eSv6-i!iX_y%vqsI_c1PxRZ*5B?^QKMT5BLjBuWL&#*wVkzq7j7}
zQxFi!&c&Uvr9bma^ww->ioW28@d&N*@vAydHQloMEwv`y+NY?RQBBTdJP6@gQ~inH
zLcBmAdv^7WjL!vxMQNJ+$Kj_(FmwvozbCE9^c*J{O>gfGR*Y&<zNT5&ogeLN1U}8X
z@M0*)S!c)SN@RNeVZkBzqF!J%2@n_{pWNoMm;PpB%VF(T(uE4t4Sv&HQvwVI^1b7?
z0v0h$PX8F=U^1c0ig!xTy9_{o)01R-^8asfH#Z4^0?N$z&%+T+O97CK!EHh7o~bD<
z05dY@yO#MKWv+=2+)cSmS&!HZLJ3?*D5pR8_VU(>K2}{_ew|ha-e?&<HPUi>W&Ciq
zFj&IQ{&Hj6?GtNRKWR~K=<#At?A3S2Zx4sft3hl^DtpuYFPw)110u)M3-v^C-{`Jo
zK#9PXjB}E3lF&=qsTI%^&fsrh!MArXzq7w*S0O=(Pgsj@fQ*O8-}&(zY_>oCkm1VC
zzoT4FIvVf8B*!#h@qT-B{DtWCj??|=vOnp;{lzZ##Uv>$<HyP8o!#Bx-Z5UuBm={W
zMUAOe&wv938KxZ|2~La_x^lIOM3At(B#b2RD-QkDSI1^vrV8*-X@uVrqIJ?ZDM4Bs
zVWiVb(Aclk!J?6drK9<3Cq2JrVs62d$0p^d;W5le1E_nf%z%4P<=T>-VZ2AKaQIyn
zJ?|+<|EDg+eVq}J67iq-%<Q6{7hz{Lmg@oC65871!(ezniDRG)>0dSv)Xz{P>i^m#
zFe*yAF04((Gzy54CA?XV!<;I-jO(OTuhapv9}HDSd|2W_{}Z|4mqM`OekbotMz<$w
z*9Z4z5=Q$#9Lh{PN!AI~m*xPPoC6c83sEN{Js03L6Qo-rEA*Chk>>_s-*`w6s5&Ve
zO)J6JWcjm3-#cF<HH<TYbMe<Z_;#olu}_?Cog;3!j{wj)F5PO1m7;*QQb;Uw?75|U
z8|eHugEBKeq`^&*Sw&twJ|(qGQO2N!j!}v5z@YBSbe(w3s+e|Cj@d@gCH>4tw>*tb
z(t1GV<w0?(5q!;}m=<bM`HFb~-LfQ&jj+cr4d!pZ>poFNkz|gQjFRC>)CQX@w3q?z
zAIo%*<pIDKSurK}U?n}1EW_v&S@PQcs#V&*#V4!<T{MqvRm%<N#;gVyUK$kP8^GTz
z@T3_h(GUO7{|##Lq^<J%gr||UM2X*)w>xzi3o*9QG^XZGIg&me6<Q#o)T74EvT#dX
zc)+%>Qdube;u1dnxVM32M@H{~l^+?P3v(Hi%mIW^ItRLG`dJcPD-vOZOv>1LnO-&(
z?z%r^`cmxF`edRcG=sq0h#WO2#>W~iuSrWGZ>!donyQwoU$GR3Rc7oDOjD^9Hfu}<
z-N0CpmyR)#=hu9@gG~J`mUxwQTEU(iG7ysoLJWpS`a}xx9O*gRNqv3l;q(4{2qD-j
zEe8DgfmV7eOq~TY`yuP<$D=+$9gNpyCAF-~40XTEN^dk#6-=Sh#(QPhqTm3=Otuo^
z4^fC=P5K32OB!6xEJFj8o+l%uUW<V4l(W(IFLR5?LAea-uR#(VjxMlXZok!Q<BbVb
z?l<3-SElW@X__U8b?UyG<|@voRy^fncqO1vwb}&gyex3B9BN4G@+WOBd4~&X=LW7^
zhYYs?m(RR>m1@SMaIXW-psHR7i)zOHGZGg<myt=_q?mUXm6^C%kwThX((U_WAO-FP
zTK;dknI<IVuV%fBGxTq)J3`I7bd}>X6PG%v@n-ZbiN+HDvNP~3Eij55Ie5vimJL8!
zTT2JH|L#T87E!FwvOi(%&+-^mC^J&ot#3V)yBNIWH=*!eN++*&ToJDSeP}4cR?^>a
z{%C(CM~;Q!e$x<oo7)#h{)))PYMP{E$a>ecfRt<1{v&EM<M!UweZ}tXtlfFf%)j~L
zHLQIw0Q)K2y!v%Jb$`dc29bvpE)T$b&|FU^Cr<3%!>`z(R`x#$`=~yO#@Qqe8$UcH
zEZCZsf%wo98TZPUYD<&hpM}t4MynVSoMTZ=3W%MD<fakY_b3fh<otItWjV<OZ#F2?
zu7`Nz(u=&s*e(jYl_EtI3B0;voXW{@RV0XOmAbSnSv6xy$(|QM<`=1gyV?NBAw!NA
zPtA+f$I_~C!qvF8?j~C!D?HANt?uV`p>9`aPo2BaoBnH$I09m1$#!L*36m_5qe{-r
zxZlN9n@ZNwiPMLIA!4QVZ5g*QOX#u^RsV7qlIOvEsc8H-tgR*PHDi0KRinZ`)S+4k
zhe}G@s>8X<8a&~TL?{`h>|g-SDMt!K!L4ZUgDT=~s-=C-?XmL^{+D!lRE1u2g`Eg(
z?VDuEq(kei_z@CEPa{kHxMQrHb3kKi6M(_K)v7Cj@7CxcO$ZrYsH%oTuz-OxLB6Yv
zM~FlF3_87;uN(c^`sDPB47%KdKAKn;oQ8{CaTquVype=SKI!<|*E|4=W!>e*W3GXz
z?;#F0`Z|3%<Qtf=P^Vi@o<Ve(0_F`n3*0@<*~jO7GSA~>a-{d&1ea8akm&NjPwgL9
z>{x2&l18p!RWTfP8x%opd{<V}Y@5!qUu(S>iqI6N_+7}W>?803@(puzX8xN|;4w?Y
zYelO4Qojke5gc@?z>onz1`yHZnnNQaPr8LK46IA^vt{Ea&MHKHcBG!KHOQu#YKrFJ
zzA0yfBytb?h|ta-3<c-e4a}96kUrfZFg70cM6!sY8M>diZ#|a}lV90a_0<rEVO3Vt
zJUL@n16e3pQViS#Eig$Didy&os3;js=hy4YMprZF)QArzJDC8QH$z>8LO2xg%X7Xw
z7>44K+VxK+C8Fe3rDCagoQektp+sjg9i@X@%Ux+a-}M{aZ9S~YU(sg0Zq0!xW&7Lf
zKl0AhXwXm}c)iA*4-MO1Ar-JwRA-c?{VCTLgpVrgIbELVIAHF2>@UhtQ;CguK{%dN
zzb>O{Nt9wXlx_h^u-Vn9mE!Pq9r3D6RTAxn=d1WS=Ik&mN{Q*#8hRYsp@ZleGnwhE
zEGduY&fgm}Sks+&Z6e$~WS&kgv8Rq_mttwX3j4i`5aQR?e1ShZ?K3BKa1@T*v0jB@
z^naV!+_Yv0_9BmbA3Q_<^4jL+BhStuSbpIuSq?aHDcu1)<rs^QCfEHM^ja#$<5Hdq
zj?`7kfe$-UI+TlnH5Ah9J{a(|-@Z5HD%o%l=Px>lo>^(bvHISc1%BxhdIPldd3GIS
zJR`QW%~>R|);*sHRXgP-2O)xKWb@bc<Ia@U08Q8Jun2)BB`3m-%?%k|QrGa}6%(}7
zU*!V}uLFQpIxll;Czt0nSGcH(;S=yqi>L67BLPKQ?Zetx4_`V|)YgO~Ib(4QwpExw
zY>y|-S=On!PSq=t6qd1Q$O*|FOO*3M+NLE+a=w?*40&ptcjt7$nhpqq4N<boE&(L+
z3-~q9>X7WKG}C&xZeh4t?a?oXqZf-m!qzN2qlo~gab9@DGV-6|Pc&k^_Xt9@zKaa0
zmB=@H-nT#;#{Q^As;$<9Oc#n4ycsw1<roKU&v3go$8g$5k1q2xG8&i}VG^W4Gtxn|
zHF=8BPIhvfR_`N=iYz^}2@2C1wMvBdH8km2Bh-qLr@HW^nrCiG)jxx*;jGrXCThB_
z)M)^i6@hBidW844q)2|TF%M0G3ga)!o`m=(5$yuw0f=fnl8q|mQk<tJsU<;v)w091
z*%z6-`I6|{-hrm^-?1YF*qExGTqYaKESIIiUF`C9Va1Xt09w}T^RcIJuQs|YTirq%
z21!6yv-9lCbw9F~vw4xg%2bsE?@cLR^9i5=nuOohG7oiBvJuz4Llst`C@4Z_z_O<Z
zJ)gmq0tOO>ZcgYR4b7>317i%3Bt|RK>BKrOp(9*|8f}cHly7m2Yi(mo%Vyoqq`(rt
zsh~eqCP=qHu%1sf&2fyrw3)4>WAARwwTgHktFNLd+Xi<vUiq{S4Kctom(;u=EDqRM
z!m+7Q%E-EUT0Hpc?)hpa&N6Rblocd0GzYDwhRav^5saz?L4&axuk3Gs@~7~}DG4#C
zc;@Nfi#oq1wH$Rum*(8UoqE-&nkQY=X`OZfTEHXySJAe|KvaKUR1QFAF4Kx~{z}lF
z_bE4AxeeQc)cwyFV}PaMLZ@zfhAKecS-P&uQF-Wk#oI0Wsr=2p_p4K?+dS#GwU{!k
z^zxYOCQE#M*pUB)R5SC0T3~L=1&2=B-h9BMJcGF`(#$o1L32m3Qbg*?60xI0)Sdr<
z8~5c=T=r>-R`=3|dB$P!5Z$H2Z^srk-vhBHr)N6IxmV4j)-h?uXg>>qe?v(T)P04E
zEyXtHIfA!IFQP=iRKOuj3zU>#&-JVNPWy~b0b-7#f30|>;`tT@=A)BXiSD!b)FX_N
zz{i=9Y7AuvxC#!P0N%Lud{g`Ai@(WgY4^W{#{W3Zi}3|1Q`#}u5&_xgFb}USDLtEG
z;hP<1ILQ<OR%gu&j`(2xe^Z@XU!h}OUX$3^Ijox!ZzFf0Hi`S~sEp80)KmOS^#xZ?
zv(~^&i|nM>^#`i!u0uQeJdHs~*$6F623wG<?#c{f0XgRn$M2CRO6hf)H!x&a&1R^X
z7=_H5=0~CaQv9tTpnyW+$_=ICI0K%9r;n5Hup;M@NO4b(>P<fpKw8EtkYb6>&dOVw
zkpIx>)hyARb{?4wbOMJyUWN4%i?pL<tH`@VKUy|)3mEhD0a&yZLIRl+7J3?C#e8)-
z1^zSUTV~GuX$bQsVU+Nak2&rx)2N;Ea!}G`6Y6>#(uRhoFo1bm@FEdF!8GHTft$0j
zNs8l{Pd@U-Y3{^knP&s#5eIVnFFj#`!2I}PPA`5BvZ-iH6AO5*s<bV)2>EwQy$$&~
z7m>j>ZkM_s!QOALyNiRz$GxNL7qk%@Wn#@9uWz3eU0;QM!o2Y?^UeE_ofLZ4oMW9f
zrD@#QX$|hxdVmm}M(<X@tb|yDiN>Wy4plv2%D3L}1B3rRm7+<o2mKfiXAP~052sAH
zS+;>*8i@Z-$fu;DvHSWyCqLRWhow~k!{k&KJxmr_6UQ-3Cg*Weg$|YkBg^x$t3xB-
zp)V~V<rmnVKvU14sJdrU`?&EUi@VIqr_EE?E8SZz2w+4dV)`S?JjO4V1v_1BDe0C6
z(O0#!y`z$k>(+fLV!0$-XBdXdF#}Q74@8~y{!o=~z^&1;Vdz7ihuvPo#=b54vAbtN
zxkDHq+*+cYyT91m4OL&m8rh}QO+hbl^YKfSoQ7uYd>BjDn?E7pYF%kNNun$p8i+k1
z=AL6G1AsNLB3&9-?uK7(`mndkxXVf<qh4pjj7hYtLZ1+HCZ(>VfrhWb)?Pwf>BVxd
z@0V)XR<h0=`SKaRelLxs>kPh>SCc4Q<Tz%_MFtKZQVw;!7tb(HHNcQ2s@T86GQsr6
z+I=)xI2e0(*)=HRwpTmqub_&zYT(TwLIe_PDWG>MJX=)`LmeuC@ZY6Ct<F|`c&><5
zczTCbVq<MDt<BP$2yJrD1eX)>2B&IkY@++#or^{NmuBAct*bN&p<%g+#8%qs7g3$(
zZ~SmO$;-n0K`)=l<1J7nw`};Nw5VJ8J9oqLE3WQ0dHPo6V0{^zOHg)VrY??f>~_u>
zGeCJ&!+n_e2Lt#FIm!P6Z$Oa0QP@oqVmEinK*`UNS(nJ2MbcYJ;F%Kam@@<=x;1MU
z0GukzPP66_gWKr)Ga|eF5v?K~(HP8)$p1-awDut#&LiwYXLQ#kN0Ni#DRyVwOg=fS
z9P|AAV`+uKz}s6rrxjiaZ`7;P3c%PoXfvdVJ8ZpwCdI(`#~C~CuH|-$0Z}Me;qsak
z$6+ca{k-E|isPJ6PS3N+`}(7)O!2)=svCtf393A?>_{e3ZTe4E4h$pN;-Wk&ZwRPW
zl*VN>kCWN#XiQBTryXvkG1dsw^m%tJ_qC=;ZCvpht;vqT&O7dDO)x_FxPq@)mGsmr
zq57GByl0b`mCbXqkzdxSHyDrFOs-10q+`0mf)8BlTG^?*--;d6kyqcc$ghr=s;{d9
za8**di7sj<B2eBAt-$G4k^2#@?dxsjqOu|ezvg19ZTCp2QB54xoKxkXFY74ealJ@8
z>B>#GB(0D-a4xZ!u0z_coSd&upppt@J&PWHO5hY9`^n$(4M`=xCl>O3m_j!h-zZ`T
zn4_r`#(!4G;IuH_&CpAvEjRgcyAo*B!s``mv(p;TEtCwcG<}Fd1AtIg{LO_LEqR{U
z!k;YZb=Xw1pgD9ZdO?M$vE1N!xk0Y-;IekF{8Zkyh?U?_S!3NgnPC8NX=NWa2rc!0
zkRmm#+}Z5hWovn20iyGbUsHmRf7|YN%34YgoIg{jDV)m%k0dJVbxQ(Vm>E7Wf)Hxr
zbW|t}$~?7BQOmD7@{AOMVtUNXlj+osYc)wqJzT3`Qc0wSO(w`eaR{ftdq<qbs~XuE
zY-qmb0$K}n!go_CzzGe0sIJx8st%fea^m(xZprncL35XNew+#zGQF95LUp$@s-2Bh
zDOLg}73=Fq*Sza=JP1_&nbr%z&8~BiQ8*0MX3kw%ai~1mcH@mcMKbPtvL02DQYqBb
zu9{?Jx>GM$P5dfOqDhRNO7F8eJznjmxt46{vJAKN&4AFXn%pc_n?nwzT9>weJHbrZ
z=VJM;Kx<A(1Lt+qz1-T@u>=vlkD#ZbXokVLPB<E>)ewxK25wei7vWEpb2wbE+bGtR
z)j-hq+<SpJbXCk{H(hyA__n!b_R&OJhgvJ~D=pC&q8JLMw~L}FT**?<Rb%5^Av?~o
zM+J%^1oJkn9F4?z9cSWh9Q&ky=uT^xp<GcCI$YJ}_3&`9x;-9ItHNqsGnNr|GLlO@
zR2?zPV5PXdLv-CMJ$J$8eD<Ne6D~x8$tiU9jq(UkLLm!g*gGN2YjOrD1*QxWpd<Hd
z3h{^3&$%H~bQgPiefQh-TeT5Nuf@4zr008gS2gZHmWyf=<aTc*E0Y_4xi=)MvywFq
z?nD^%ZS$mitG7*O%V%qcNtc1mJ31R?wbKE>hze6S;Met9ho#-*VeaILj_+3A;w*0i
z@2JoS)xqh_Y1_eFt&wDEsT3F0m7Lg<GPPaltP9%(g~<_+J%k{@gc+OWl-IY__9d&M
zu)dcTYk!Rj9_QM|dk$QGU3TEB2GP1V_XbBjM1gN+l4DcEs%BGYYxa%pV#PzDn8J$5
zQ_u7!4$@fRamoZqAG}!j3K~Hf5rWyI@$Atu#11R9R7%A7V`{m3!INKcBxiHsl4pg&
znz{v#|BeC&D{PYT#GQ**TDz7r3Jz(o#l%pD8R8M?9*hcAzyqp(UlGuAG3T0S@WabW
zkrb+>R57(Se_xQ?k4vT*#bF{^=D(LTtP<~?a~{79D4dR>xqJ7|uU~%p{HqeU;9qHl
zT#y-7#jt#@k48jg@KpF#pPj%!ALEqN8Z-_T2p9=koX{51z~|)-G&*mC;_sv1(r84V
z0a~-gxgYD(#0G$W4K*Cnj(xPk`^E<eD*};`CN$Dh+NG{dkGQ_AN{{KopW9Si_1kjy
zWx4xA{EP<w;kWbBZLw1cP^r7jSlWiP6H<eM#*XMw$#zCLp?xZbY7TV@bP!%D{6)}+
z*fzolZbP_N+o25+YSZ`*114GvOKKBRB#{~rLE5C1s<O3zu(tpH{JpOGimN`d@3PN*
z>Xm|Z_`{(C#?%t`%FTE*x2>e4PH(g6B5Q5O)B`ZPu<?#{gWOyNH?nC{MB&rIP%O0G
zcVa&sW)S)tYvBNc=Fo^S1|*(Hh8VMxDK;AFZ1$%Z5MC8B<{hJ-cOE_5Q10050eh9Z
zQqwNOHCjb~#c~y<S^>CzM$tJno^5+Kaq0-`cAvXSq&B_sbZLd%D!itkPb+vqe@Y|N
zkQg`?D`c!n!-1+7ez%{xrQThUw$!_;YEkchEaS9#65X?-=)hk{Dcly!cICpzx2jyA
zzn&q_`ud7`g^GxWQ!|7^1&d7cp@Ic*rxZZrp`vMjym7%IQK(l!eEWKZ2~jwJ8S>$J
z#jIM5fqR*3ncrN<dUX-oY|v%i6x8k9G><u085%m5)7p@tlpjhTcAU@uVm7OU$IyHv
z^=QVY*G<UF*=)Ma1QWn3Qi0XqkAX1ymO?u4r+w@^`hUs#OuC&T^~A1~TYkeNFlu3g
zn-I7SLhy8hsGqU_0mslM>X-4fG!p_iHkZ*@DioJ+BR0N&%;Rq#ez}tcxi=T|lNT7A
zg*D8qGlKT#n4zYRET8zDZ~zmr!uXVMr5)2>^l^+oHRt(Bot4HCoHRRPca2{bQCQRX
z75QU`KI4gAgLSXChh5!+W?7nLu*|xYr%!dgarsGt6JcP6kyo_!r*U1PO}FRgvSWot
zs}bVgM;&E<md0GS_MB&=y=>-@xwIKNt=)vqpmMrtj!77r9T^pvWzWjL{3LNCjQeBZ
zI^Q?*#uP}A6Y<HYEd`>@oUi+_J);gRWNtuMZ>!8BEE7n#f~-{OBFVtGZSxti=?ri^
z$?NEwa&i5s;2{hDFJ+bw3$7HJY=dDk5A;I^`COoXTq=(Xzj4s~f~|~Fmg$_R!YS8a
z@l!YGPo}48b`LXA4v_*_<xk_@6T=G%2O0<4ELmop+X%xY;TH<suH&95BW^zb{5%#N
zUzIs1;byzow0YLWEIqkqO`@%I5B!%O&TJk*t`*8}YSXS|(i(+VX2e)NU0LNsprK(^
z;(5k@ef6}?XS3j2=?6|zJbn3B^bJ7t=yONmCfuYs5$~TrEJT9N3N(89`1S6OiQLG$
zDfyB+DIj>9{xhXvp){H>Nnw?8mi`8yf78Ec;+ubd{`Aww-<6<c|Bh4TWQ0Z){PeZ_
z)PlSo>gh}QZG?56j=mCsG}%msQ1F^2)P-$-;Nx^>h^GqBv(YbcGzjC3)@(5`9lVPZ
z>jrG7;t(Ouqh-G@e6V5Fw#Y<OG~%xarL2uVX?<N4f5s30U8mBjzfX6crn~ort7x^0
zQ@qT8gQb6KTvfuTs8Ecvbx2!5!2zji;Yx6%+Z&^t-g#;?!v=^#FC4`aFtyzp&2@x-
z(OriKR)tW98fwG%1|J4mX{0GBQY51$0HJkKQB}B`IXr*=__eJ2Osn3pzZuTG(jW}l
z>E6m99EpWsr8thFrM0-EOs}(RwA9(FV&U>mHU8o!Yyd)AzMmLMnYR0xxc7${Lg6K%
zIlz#L){+q<RUw0IjPPoY5p{MogKzhL!#W^C;lj8ooBL)8mu9*cuoRSCMx)rAh3ms9
zbPE{O`8#e<u?B|KsqnBM?~tA-IH{EYOjURZ_fXwjf!kwh_}aDxilef#W9O6>6TPEI
zOk`_{G%H0`q<>8Pv|{k@nZ@5b0@Xht$r5x6&0!9ntDA}xD4F?J6M-1y<}{&y1v`p4
z-&2=j+~%4aW}t<Jrd|5Nbi_9)R1FwH8w#M)S7dRb?Bi=D%%pZzker~bh;0*HHgb$X
z@$cKMc*gI>x(x@WZi%tof#_p{b^y*Dgzm6J`f9|L;y-0#u)Ea2KbGNw@<LN2K+5y$
zZ*uw@J^Y0Xjth=Lg*f$Vak}Jx&d4?V_3r&{zHF7y<PcC=VZ)sNJ&lLhJ+KiT(N<x{
zj9M49SvO@G$l+<T*o)_KTmx3JSy<1t6xx{o0J2=oC-tiKrlr};l4!5w&Vkz4iyWAO
zsO{ZZ+nip4z7;u2P0rrFlpN8EfMGI031!@DEJ{3IvZ<C+cz&K%PfgW-_F7*!@5|O|
z+8-ACYgTVL&ovc&*)=dgngR+(Xr<D!g#$PWchmw09dfNF)pBgcLt{2qaokiKb-^*}
zS>%nC=G&L=+D}Vz)`X}FLIr)Buz1SWtn9*@?F-9LoaxlKf5m>E;$wpAAzu@mx1{%r
zRf&_jaM%_OUve?7DO7xaMuuErV%b#^&V`I~1_9>`KGg?(wjdR>4#|R~ueN}4S2>`7
z4kCpJ#yawi-9cDCV$E$lv1I<Vh}C$3qLz40k02kKest+Ranw^nhA6ae49$Lug?7li
zYhNq6X2S>a@|m9V;v_CtME6V^(C>8A1`PQ9@7aK=Ck6!QAFU^U(b2e9(UYFM*VL1K
zr)zqGR%na2;CH(cB>O>nK>6n`>}+RVLC2I6l<rSP1<L>DIK2j3ZYb$DKQo8jO|NRG
zhkYDn7VDa_oO?}kPMBEmq4398wJ17oWu&XY*)i2yQaxYe6p_B5$Ln`Hs(u5G-^G31
zdna#mue%jjMZpYzrKb=m3zm%(9uAkR>4?x)pd_4h$w}%Vz=H2n{+@`BG}uGJ8g?OJ
zQSOcaxv;{&Z-(#SgYXh=3Ew5ZzTD6G0Vxu~rWQ~0jhewldXFm>`cw)6W<I9r_u}~O
zmjMpnAA%v;vDN9I=5d|tDL4$bY%vpVw<mL<rYjX(s}s0??~uj2BU!vUn8kypkk`6v
z>SGS&e!&U@Fp*{pEyL^P@XiZ_u-NPQIeZM^)gp<%lEEwW4GDbB6`q;EgHl3k^Fkst
zLIyMG5JbFy_}=L|q{}w7*#Fx?G9xh+BL-+!e4^1>gar!U8o3u`dKn+5zz8H!Ywv_p
zYD+v#Z96MQ)K1>9u<KfSe_jZd*FLcex@bgL4RmUetn+-+e>Iq!Z9*=VU$F@&2*${r
zW5E@R{jq%aWnJ;{?YGMTjZa?>>0ghKf-ARXD>eZG0~;EQmr=nq6aqIllQB&tmn*e2
z43iNVCx69SYi}I475(mC!H;Qya(G{S6hQ#z;nYRa3Xuz>HH<)(x3Su_){xeS^Y42u
zcW155&f2mwQUwXRGs7V-d0*#{(~9MgLMv7RiMmh&=i)*Ong~Kp!NgpkkaKl`L)|=*
z(E3|h1EtL~3&e>YK+5QH^Aw#cE+N%G)*7V3fq#nl&_y75PX!Na@{tO<tLj?h2C4cB
zjRGh0QCkU2Rxm_?w9yvtvZmf$3v7MZnlP3AwBp^`pH>dsgk6##Dff07gqkZ~Er^JZ
zQF9Q@iZN;lVw<NmNa^o3yj!zt+VE~mu6)D0P5V8(+p3S-64sEReWe_feY%w?sE98H
zrhlb<3|CDJ(`vq~+Lxs!4q9Emite>-o@y}KJhfm9J3z`X4~B$Pg4yP24VHaLYihx;
zU>Jj7#<tM~c1$oPBo3LH_d+#fVef?+B=dHhLhde5$snn>V}%cxx5GfWbPY%#RrWFP
zw2?nWS0!n6sX(gzUJZq~-)oRcUkwR`$$xuM+*5L$(S0%M8gPKi_&pc}R`hwm09b2(
zR|IZYeT86dtgEjO92I~4X%1DZ_besUOubzVwP@GAg<6a2qG+7l>#Aa^Bn@NuRJr)f
z6@%;MDLOFe_i(vcUGK_Wpmk?$?Cs#9g#KI^cB#2C=Hh1VCN7qQVa%guGkY@^t$%k_
zUF;Uk{+hbWRgaux(tGHlbrIo_iU(BfA=tXZcrat+{+hX{`rz}=hbO<jUJMT}FIVf~
z^v(0NUwwCR`QI>nxw?9>xO$q93DUFSFT>;EtEaTN_+fFreo8{IlpHdbs8|+iDJrH7
zKZ#cP+3+x+`_tju)sxlmb@+_L*ni-jhDMS&H!0~4U*KmREmDmM6K58T4lO80Du+s_
zWFdm#6D#6YEtmO}uv%k=(W$lAU@>bsB4ti2MJa~3t0l1@&XFYR@7Dqoif(GlDW>EZ
zJ4F%?`WCavp%$5p;a*5#D)2SWk~3&gN^G8BMlG7g1gAfyMMYG@42Uzym46LN$qWWz
z42n#k%rPN{Ae}`<PE;7GP8dX3Ofhs$Nv*IV2hN~G%CVwUiE?bs>zP3rM&aOyM1DU?
zAdy5yRv4AwY7u4fs1n2tNdmc8GjnQ&EI&tpXpLTyg)PF&h*FB=*T@xoUrI36Uy-^;
zC4?|I67Ml3$nZo^&5U7_0)MOMe4dH0i|m;q$L5#0AW9B(K*`9B8G{2wWJP_o8FWB2
zG%qj|y%r)RM=$zV@+ebz;%;1GyAZ{7sgB*E*V@Xl@ZmxL7BPw{0F*X!uyP|+ef<=f
z%g6a)`0=0r0)}w#5H8;=m(L~}pf*M?fw>lxigJsVfLzhx#4TDHT7M$O-?a?GSF7df
z>h$&5`67tF5QdYp^?Gr28F+K~&D-_2r|YxzBGBf>qt)daV>}YC-Y8~1$|EoEFvaEq
z*cA&Xn+t#mGNjsmfi)!MyuDDrD1Pz97sJWb>il%Eemb0d{b+cyc)Naf!-gkkKQEx_
z<$7_szIFi9m4mIWSATD=&KK7k;Oe)&U%a?D`*QWR+cuLUFGSDD*%jVID}XI;)qUc6
z1I+vI51w{t?i1|Dd@HoAP}3;TDDWuADCj8IDCAKXd#=6bXS4XG88QQK6q!b;UehEh
zO%&3^hz?=c3h2yXV4e?DpfDmV>kd>Xawlwhs0E5SQW|(d%74}psj<wbg)~gv6O{;-
zraB9Zrgor(XRF4*gXI4~v~rzMi2`t!-hn)+6!S#pMzTikZoqaZsRV8$Hlq^egBy5f
zi$IY<GL&ff5ZJ(T1P==82hgcXn^6Z?!I|{dB}D!9;X)u|gMEQIf{7S134D4+9pIM*
zd}{OvfnS1wPJh(V49~<5z!-Q<LCKkMCBWOGb62Jso#F`mvev^GMb!|L(OeR6CW_ds
zA<FPe5Jyx}gl6(>1p?Tr6T#7HLo$Ik0R4kJSDjG_bNG)k-W<+Ai45#36VU;%jU)vc
zhN8i47Hjj75BAjAct<|mJBpNhCLcR3w&dfO7Vje;UVq^m-Kg-6=JOP`*lV@H0nP-G
zdT!BDi$<iYTeO7XVk&kQ*kTVs>|u*N)X}|*)Aj<z08zcUD6o7Y+g`xzNy@facC=5Q
z06{91-$IbO9pd94i0`iLAV}Q;j+)Xa&?xXI$SCM2*eK*tD5FqEVeFa4o@wlv`1gYn
z6p)yKlz+7}BpbMxWH32!+}fhiVY1-@w9NQLqE1>1l{qqAL9zj-Ma;&;p8Sz)NCS#Y
z%;hloJ7Y<d0gF>X)lNu&^W+-e1>7l7Q4NKUDnW@cdj)p_$|Te*<m(x<04YdBJ1s<g
z!{fLXz^bTxNH^8@O4W1@WaPA7>bs>7BtXsYv42MaEL4IIA@x$unr3hcaSv>Dr-ZU2
zqjG#DvR5EfK~vwS6q<ofBH1F-V@y)iovjtzM*tNu-b57y76DKNN-e-g0ChY=Eq40X
zd<TE4g*jGaJ+1|SoSMNXS%6w9)}_wF?|qk88Zt!llaCz5?UBH-ClJA%1gZG`pZJl8
zMt=qm1wHCJ9Zn>$-1oH0AfB&~#l;z%Ihg`&A!q?62SeS9Yu?-kbUvOD6u2_Sh<s1U
zik!LjB;`5)g@ebIqdMG4s&a7L{G7m21UvwT(OCT&_hn+s$0HQXt+iwco{_wg-2`WT
zdiLHupPLPKxQw`?xDU4!YLV__Bj-4Za(|H5Hels9Fs%4iA+?VHTYJ#$W04oz0rFz`
z6yzmohrIkmaEpd?1~inAPHI2pjN~Vq&PL%?rZpCw+!OF)%jCHpoI{Lxmfc*06O`Z*
z!6Z7(AU_8JG%Ko8TQs0t$Sp|s1O}LN-=kS?K{xhWZQZ&Rtb*O9C2D}9mJhcSM}JvK
z?;-;BefBu1xasuAV`Gn^+=z{jb$Uw&POtS-oW6fOv2}VLIfaZ|Kz4V=c#x50$jCBe
zWEnED3>jI5j4VS(mZ4+MI`*t%&pP(3W6wJFtmXFyp&u?O07gbVaRBaz8c8-s!T=Ie
zTMvZ<^%!6P>yS`dXpR~Wm&O_z;D0BoR=^;@#09AUjy!P?oioq?IDg8uAPk|*9K<NM
z$Z%L=U@hMjD=8v3`A0IbP43qE*j%_|urvQ!WR}zLhNZ}*bHu&}<aEHlEveOyc<yN_
zDxf&RR&oLO^>b*6Qn+&y6#M&80*(-i>~sLij-E=7>fn3UB{hMA0+0iE&3~|C3<rk}
z(xF62rSFLqWLg0~61^WCfTA#A_eX_iAcfYxdC)0yJB-j=YD4wPs!4&~u~aj7rJzcB
zg5K%?NEiJJvtv3~DW(W&_4CdK3Z--G3|8pSFmb@|gxn|94_@J%AO=K-yZ6_4>0)*H
z$LOSFjx7rhK;It4u<GbCcYpZ8Knei*_w<E`LvH`5Z`4A`cMs&K7I$%k!Qg}D)=vlm
zhZ*K4zO8TIPwkn*I4Chn(35hq3XNv@Y^cSYfFh(|C4fz`Z=(4gEAqtzJKeeih)Sfn
zM(>CyI+XG3?4Lto36T8`m!QH}zMBY5L5{7=@gagxLTT9PAl1L0I)AFd-NYnvQ|Vvs
zp>!fGLX(D7ss69t&tVp7Z$xl~eun2?ZTc<}Q1=l^iXQ}lzz&>D8{*RT0KB#jP|_O>
z-hGCkw`rBt3)d}LTA8G{6D@tgQG0LE(#r<7|KDc~9}DHP96<R@p91A`eFx=jgy7aD
zp!2k^VG4ov$)azeCx1r+z9$HJKs6x%%d-c2t~)&3_1-fNyzn1CJodlOUar@#um3y@
zzx?t`ym<a9p08dFFIMN*!vd{NPaoRJPsr;z(r~d{{Tx>>UR}rY*UQD*_;Rs)JzOpR
zeSWrFo;_bKmKV>j&aQqPUY(s^t*(c!PJaCU@ySnrKY8-_`(lUR{q*>6kG^?$8egto
zE#c%3ej8~Yo-m9T**}S;sOI<_EV+NjiUrX8@e}^}V*T>X^FH6@+2!(2XP4I(!(ZQA
zUc6ph#sAz;(!DtIZMgJatM`2tG}Y2F-)#9mJ@d7fm+`bT6PK{UITV+^IyNkSk|Vbf
zz4uqt4>W`&u|-TUqMI{Rgl?ASrqD^YeDNHti+|q<AQM{^tBXxe%Z~7Y(<F!`k?VU|
z9J%~sk;{)yuV0^jxENTx2v&K$eEoe<_=UuOmBc}37q66XMlD~zEx&w{QagOT;tQlS
z245cE2mEVK&qf8Dp>T%6Z<RWKJ$q0%g44=;g|ATF>&xjG6o5=CUTK9r{rdH%CrQHb
z-!EUE{`EA9<rWu=a8fP5eS7-lSGoKRzx-*D!biXSJ5T+#fKkE(jqm@t{PpQCyO)hl
z7#9}jwUA)=vgSofK;ckKX??n&|J03`@&de)Rpl=vb?AJJA1*;{ouvbR2vBZymZzO8
zzEELsrd{*d^jViN<7kY&vuT|>Ju9oTzAbI#%2#kH(~|A-79=oU$+~o-WJqgIX>C46
zmaWseYe$6;lQ+=F8?%!-wyo%bz)6p?qQTvyUYvEp_72N>vypXtDwcd<tFdjf(3)Kt
zfZy5aV!bu;ShCu=&Nk(LTecx8de<Xj5EidI%AM4yDd3=d&SSVI=%VWUeHx1;2sNLx
ztsg+Gn@K$ic0~gXLeG-%X;7rlD3P`#VIb)YNq45O_a~zuFV1fEj`e#tQyODQqMQK2
z!rJzdBK1P{_r2YlrLza9W?3c4M=oSf%dTE6?Xx%Ku**A!x{srO5$I;5vDbW)mU*Hu
zHZIO+^Z^!JVv4aZ<Ul?z;^Z^Bq$c6Q+AZbznA1%v`+jaW9(2(0Ch7hQUC)3(^J$4y
z@<B-_t*f8748+czTp7>qcwwh|PvY@WxDZy-zu^O{nSE;<@`>@_gFZ1ydq|ph+16?d
zPJsDD`FRXsPChb!pf>-Pa!<e%fC$7Q`tcfu<fD3*+)p(%HN_ELBrNn+5@H7&RrhiJ
zH=L`?A@>{BWG~VixiJtnK9@0rYElr@ZA<p9pk)-%`Nblh^e*<{U^^_iMl97{=t%?Q
zJ0CzQMug2H8xJ6R3>P80U1&`5nj<!v4#OPr>|IXi7>Cb)&PTZsl2HqYxE3*XP}Q?F
z>tQP;bE18z`njgb25PVA$w()n5A*qK`J|+zgm);#AQJOf5T~89W$hu`If>s@b!O*F
zHv3Gm<T~5MmO0;PyHUzr+iemAYpqb$-9iv#&ux)gWSHa37Pqz9VJJ8A0I?aSVdu#$
z;7~GzTE#toC+SwNZP;+4eFknZ;posZPFGk`%GXNIQpN5LuFfDQ&xPnxN4ezqG}6*R
zNex5cl8x$ivjffS&pdBRwV^B8W2GxfOq0G+N5O=^WcjrQaVvqbhJw~KCevMwrb1if
z%Eb_7UF1(mLtjqDmb>lNgOq@cTG>Z7jG<J+d+8Q`xwJ~s2?K#kk<m%DvD$3d)%Y%H
zhcYMc<jM*1+!fw^--6WmFtrK6%)d}s#->TrtmUlTmLdyQ&0QOzZAf7#0AImav;2X9
zxE}ra$EzAu4$*!vvs8w?j`|tgcqx|dAyuh8`*An)n64IAr-ajbmZUEl>|^9_F}+u6
zEJLh+WgkQY=<6j!)S=AzgySL95pE_iOq8K#3FXsEyC2<)z5&<!SHF5_Wy#7eT>S*a
zTWd~mEn>l4k)uXp9ofu8{B82jE=u|2Kssd!C(tZ2=#`t=8bEFy^;`Ps^SG?AyS%$}
zeqyJF&Yt#|snH!8>`ma^h6co`G_U-^AxsE=wHD_a6Thn~-5iP8D=0%|4VFlB6sMZ)
zHSj`B9;bFK)M$mKEYiqEuXh!?9NM^Q);Yx_d?5-0Q|_o*THT9%u`$_hUE{i?I?>wD
z*zg42-k_2Q&@dOx2Jo%$XW?Ck`Wk<_3VI+4>0#bgeWnB%UCKJ?&yUlm4eVnVgd@U#
zc$=_K`hszeQdU^R{eS#@`8~m%aa-=4BO}EzoBqLfITH^fKNCQT5^o}%vvx?tDuG12
z30N#`wfUa|mksDv<$O%6S@G+gQAY~aJ9kAA^tOxU@u~D}7u4M6p$d3J6Ia!s(GqB5
zFo0-Quia#PUUA^0rUnq`A)t*guAz8;+q7UAOdZs#2jRS$cWiKilmST={wIq#jPxP~
zFYw~?j%t*rF)o#v{51Q$;OAy(wN(f;4hpN0Ram*Y+MQVIA)cm?xXRjTx{`q2bOwv}
ziea@IyOh0OUh-e51c0jt6ku~qL@!**z|#&*MDmGJ7D~BxQs3z%5(HQ^pL)%IO%T|7
z>E_vwgQF+aSjJt}K(rdAUy!g1K#qSrog6#$iU}#?r&A3*5;OBj7@oARmCOygk={(`
zw^+`mjisv|a0S4S#}I(2k3dr?e<{GNtlzU-BAIl`w)*&#>W>wu0>iEjid|Z+D-voZ
z%g|^sk=xr1vm2boM`3jsLaQKuuVaO}Y~li<jkx7fg%&a;Bxqmbw@ZsR7@p#DI(2RN
z+KlMIBLP6N4?p-OAWWKn8JiewwBCd$`?ep9cjmKt5>(dodgyv{53w$l3X?LZdc(Rm
zVG;s;1{|j5w|Cycop1FUOsYu1j7*q#h6m=L@zj_Ba}(G1ZB{*>Ko3TLoL+GUWnOjT
zcx>KsqM4bEGoQ-*%2Axz;(z6wX7>Gu9H`G|=DXiXHcWcz6x_UPWl>E+C~T~9M&?P5
zdt!7)eyT_v063H8D$nD)n@<}BukV1C8O%A|sAaaNp>)!IjWR*|>Jzht2JX4Vt5G)0
ze16c0Tyx4&Ix3emU8}!;btuYrvgx2PYtKE(^UR5vjgP!^)h{U{oiC5S*r}3L_U?~+
zU3vr|70kN0*XsZivkmjE@1?jd@hX@ZuGXdLzsqnld%aS+D3NI&L{PDBZq_y+jE~PL
zkGoDmfjKVcbUhoFFy0Z`V80aH>g4q3xa4x}l0MWR>BwkrW^crQv!|Wa=Vq*#!4V>#
zR+uXiTZfRdQQ{p$l@rThPX7)}Ur_oa(Z)?BreUssh(PkL!^lq6)yfO4?0kym`mURm
z-U3#vL9imW-~Zj?T_dwTHTZM8GWnXLa64|3f;IhkU|Nx!9RVN*)=U4qW&mVkVB92Y
zPy$)TN|AC~&VXotcqJyc+8MuI=}+8C(2Pls5yI;9KG%dp-eVHA$J{vW^oSM&XU+73
zruI<ufxr@>SUUDQbQ9;DAKOj5*)dUr+UQxTy&wWDJg9laT{+=#O(`w4z>2-o<vF6A
z>ozPuu9{AfuwA7JEuAJmqF_2Flv`c)TD#NB{3GsJjz^b&!MmFPsY#<Cd{j$_y3o_9
zS6f2&`=$y~sP?So5!s4&b`hia@jZWKgV0_-u;)MR_<)}OwBs8+|7rF|_xz6u%RT>D
zS$}@dUx61!nFsa!=iMIO^H&MWSh%C-Kkx7{J^$&{H+ufl?C;a_pZoU;v9yWykyR}i
zp?sWoz82739R8CWe<Hc_T3D@Ti6)GHk5@B`m`;GcMekPd{HDCm)>Y_;#(nV9nd#K;
zQHm)qNF%`e`yMw$G`9*L{Ip>_^#2bVHfj_GiM8=Na6-h<?p;qBHvX6Be*oI7;Fob6
z7ZU?HI5?MK)-Dv6(GmhDmk=%<1%J8i-d|C<Wr0|u_XB7cu|4epa+^Wlf*gWDW^$Mb
zvYSJGKct?Ns>&{R7Zz!t+HDm@krZE|*nIjv@#)jk`St0;Rq4qwZ-kT6>$ge33l6^p
zhe2y5Cj_%fOs_wuzdmy=Ej(v<H0Mge<KS-(pXPW_!UKc=2mx<}I3Ao9Fn<DI30}bC
z)8F!XJZPac#~J55!yf<k`j;o3iTiLhehN%W$?Xzl)@cUo7s}eY@Py+5J`gz)X$(A)
za}A&R3J(xrI1sFnU>Hnqffaob>huDifDDM^CXS`|Mnv-N3CtSFBE?x?mpJpFoO(_;
z<2nUzk3U1<sFAC4fHQr<w|_dnjU%qYyToDTbT%(S*|I7;Hm_;&Y8g6?SxQG+3D-6u
zkD4?m&ZRO2yc{-Kd6(6o$`rO-;p!@Vmt@5_6R~9MaW8z84btlKOqF>sBdz_OBCeyb
zIDVnBDAeJbGKy-0cq0#W^$3Z(Kih51nKacrid3~1>Re`DGEY>Z(SKBQ(GB6An_MEV
z>HQ;nlxEO9iX(HZ)e=5tJ+VYbLbra%vK9kW(bk49Gzuy|bPu9Ff>Az4$FL|#DH)(B
zl2n!M#5<vXu-CI(v)3XS_e>xv-BAXO*AvS2C`IO|=vbuORV-BMlFHbe=txJ8&)6lg
zl+2a**txta6RDUmVSh3KT%6QdCTDJzp_OC~oKX+`MmZrLQSO%veXd+)>1e-XA?sRf
zZ1F=Y8uKnKyKN&$<I;|G92Vbu?lJld9qB#Y=D?VZ9NgSt2x4OE{u0>TcQ<__9^_R0
zUgxcDddQh9>+W#ktp)XmxAu3VY+qhy&ctxb0+_<rpHII|Ab-$LQ<UlxXDWCz{ddA)
zm;}JqI`~vth%o&IARqqy{o9|v{#`Jb@E@GPCyg#X_~~c<s#pM(2-A=J-7*Cft4XlH
zmDQ8>kO6q^>%=<nb$Zg$F`<knqd#yIG93@O4w<Y3gUE@1WdSl&Xh4i@G_UuA4pyq_
z5|QXi27W^<d4Dc`h52O`e#M7=FQ72%pVQN~>FINoXHf71pVmPNv>;Bwv~ZA45|Nf6
zX&JYeYG!c+2PU>MN*QgS%zBGblAbxp1#v_2YL(eCLKrSXfU96ALzT2)d_@2QuSq0H
zFAXz6OiDAa9crb>VojoLe}Da%=Y5A+pU7vzb3nhLjengEMhkEX0&&GG9&+WSq%?yU
z=n`B{M`fN{@)nJ&rHCm!r<IE{qDD3#1rk0_gv7+_p)vb;vw+~RVT|4&pwUMX!3fW^
z^Hne_*?Qq2Ll?mie+3aN<@;VNVn9O$pD|*XWj%P<$sYHGU|pdrDA27L*0r3C<Trgr
zA~GM2#(%KoytzWCJ_DMamgZUAowy_V*teuYZL^kCsHsCzp?^(1SoZb@;_izvc);MS
zS~{U=2aXVSZfVCZIpSRE>5%3#dX^37E_Ue}yY%+_ebcE@TQCBtD$Ok`juHl)@g5yJ
zI%uzqLf^YE`sLHQK1;*68y6zX_Gj}Ia?vID8GjaXEZL1*6BrLO!N`qACiNI!67-dO
zJ(Er`Iw_x%kyPU$2hV6WmkLD{N~%$z(nwJ#sft3$?T7XXZOXdpUiwbU1Ch#<AlJEL
z$~@53rOfHwrKhrLfbBT%_*{{U?%eUrFf{0a_1%tK+R?4JAQmicF@Ag+zJXi%b%s`k
zz<;cajtO6+Y)V2Z?l6a%e*4Gt?dkJ~HY*W?sTiy9m*GjZyOA%s@)kJg8e}(+yHL#*
zp#28(H48qn+0e)2!1c8iTcjAYH`>Y^qlSb`nHk1hZ(|ziD!(GR-N+Ym6ai@qJ)=$X
z<%5fH%o;U9qr=7GkGwp-zJn2AM~q_SkbeaSChNqEZ(2~3+siEc+-qRr4x*I~%g64#
zUt{s8%;cIG*q0sU(qp_QmtlR|LV58HC<m<9#$sF`o&xK*i$}9#t)SpR5$k*}?_y1Q
zaJfosWKDX*;CU}=iqs)-L-W=@Lbt+Ao|9cF)M}A@SeGfM$60u^wme1;5GY6M=YLhJ
z9w7>*F=1xn5$YyYT1XazpHX#L)ocZ6tZ{Zn%^on`(d+>lZzzn;{xhq9a!F0!u(fbL
zXL!CHX}Z0)Qo`}khH=)C6tlmEhyyC*8f1qTLJ6L<%dbdrH$tVDV~TPD_NP0nyuBo!
zxNR9%viutkw4f}m5tZ4z+#(~Q+<!{RHX}cZ=X)-UgoBp^gXp4D&r}K+1I^@?&o@s(
z1_9Y4P{V8pyNWuhabjRWleZ8X3C{OzgFbz@YB3oWoI?en`3N{*@MM&sKmO^LbZUWI
zU~!6MuyjA=55M%<dp}9Kmmw(yb_iM7ei_n$@jZUaD6BP(7}xw(@tZw#JAWCZU0ca5
zAU25iQ7eUGQkad@R!`PLKYsQ~TxAcKxzLuzyLa%UL|{sajXc4D1HAG~0y{^JoD_^C
znN3N-Fe{1;9`TaN*ukKLh0~AT)1Z2cdl*!Y@t#2q9G!k|gW5!GHK-nXM}vBU`Ebh&
z`tX4!zIxoN2DjUuD<xyuWq+F#m9h+MWJdAMCS0s>py`ecd&PvBNwe<D8&WwAftQ=E
z@L@8Sen=FTsucGsj?xC^nh?)f6^j9dma3$Szm*n-kPYnG@{(FjDix<xe$Jx_lFl#c
zPDV9tdRVw}-f)E*DQFj?YuAsZ%v~_jDq4T7+*O{s)&P0C8KX7@P=C2-$}i0WO3T2*
z6)hQ<4`V?Klht5u*Rx`+YQ@cGWt7@6A<cT=c--BC=?iH`%yPxu=WsJX?+|VVjCaD#
z0R6k?Z`-IX;bxTW|7ZTTji-{ojXa$)558~6!Liz6q)V~t4Z%RiGVw^`y<p%#w`QEB
zgswkK7W;zGCTv&FP=6*rGIn3S*}i(rYYj=ozItf&@`ehsMmIDrE6Q`*y!u(1FS$b6
zK6)d*iQpkMmfg*-YPYMkpYcj>^HSjH`nL|+>g8KOVm`&BN;oWaMPoPJ{s|YD?R!dN
zuu_h8t6@Kq@{e2-DMtgLY8{02olbp>DWFyuQYjrXt8z_bvVT)hy1(?I*z0Idfvx>j
zDX>DwmCsG)T^~`o1F!gEuvKnz4=6(0F*K{x??%G%IR{@pbV*eynxZqV!y)NT#k4{_
zcr!fH1&g-OI!-TLPvf=Uom2v=Dav@me<B*n^EDgkj$P^0#7iL8IyFm&vsSgwpmk8t
z(G}~`YR$S(e}A;4SYrxFQcv9byh5q3UAi?_Iv#Y|k6D*4?HA0eT`KfahaAzPyeiS7
zMn~}sy3>lxxu!DMVz&tzLvQmZ6^GAOb0+($S|>-6sVk{D?Re$;UCtLxQI5qXjI1f7
zFC3pf0B>wC@Ju*AV$PPoUvq(vXWx<j&B?^JrUD4=#(&JD6l*$YPw@&|hNIG5$Eh3r
zsU3U9m+UCbdv_21VyBP^)c1AAs%8y)?TzLhQZBcs0XL)VrajS+_G3rrh@F%Fr8u{^
zvgRQnb@giA%D!u}jxu3bXeSxqm75ECKz_GP+s>NAT}@0{QyQv<t-;WS@w9FqN?>f4
zj=0|arCWdQ;N!nV*x5<kR;PTWFo_Bh65dEj;h3Mvx_UOx)~lpd)%SuqIqtf?okMDG
zW2;?1iC_A)s}w{|p=?PfYN~ar6_7=h$4}QzzM)SEiLunORBH?TA9R2FS(ov&G!p_j
zFqh$tJrtK@XaOgeo=`TG0Wg;|Q8sUXs3SH*&sY8k>{cVh?nKEk(sWBn@QRNSS{YKJ
zY@xjkr68>9?emlV5p+utV0nInOdFWVO`1cKF5&e+nx$_=KBfF`8doa8UDG&?@ux7$
zOgd$mBnL>;2{BL$)RVpV%Q7Y!hSK!ZOHFZ4FU|Zq>LSSf62HA3{2D`-uA~otFPU)B
zBg1{t8(uGdnPzozT8{GhA<={dM$7%6Xu>K+*Q27zWIPY!4Vp}U3*XWkdCqo-W^~C#
z_N9+623suhpR?w|E0z?>{p#nVSKY!Cp)pW1^?=;u3PCUjcE;6p-Ls`;Xlzf<t|{($
zb`33Fk~wv}f_a~k8^B%W6<#NQbGEk4D}-W3V2ib+$Q-f>LxGzhn{ADBJ$(d*tC3QX
zd0PQwN|%-iT1Xo4R#GzFiz4>r#Sj(>rp08Vuwsm#aqVw}&`dH1NAJ$+d5eue@+c?A
zmU-lIGl&%%!gGZ2<U&|aT!QmS2Ird3Z6_#)pbSWJ$!OJfRdsOWNzDs?zI=q?z;w!#
z8kc(af-orNj8V4EN+=BbcvcAY$3Ok#Q3(Z?Q3Ru@Ym`6yR6Fo~s<f9UDcT^k#g2JW
zLovIG`jGR5+oPD)476w|ujNfCiZ#%WL{aM__e@u7QPg=TJJUn4sB^C5oir?!aid(#
zfM^GU_tBlF{p_K{z%g}yR3zafGj0YgJ|vNBu{P8C>##{kL}t;<cvRa-K+GH(q_9cQ
z1<#3%4l74FX0-0Pa%zf?c&w6?Roj;0n#ZQRqm}5IywrQBEB1wJ^v;9cnz_GyvsGxB
z<zvQ#1rLEQd+8nxg=W^eo;Iy1_ONM9anGiWL=W3^9?u`MG0@(BC)*$yly0|*5NY)J
zbyj>p3LDaf;=^wP0e&0;G!Y=}so_6E7-k?8IwGe5jP_(D{dt~G@Ba1n<<Fo0MmzTK
z_HPf$2V{csNIQH>x3o2rQr+-1-J{R!2_2ZDGztgQOd=<Mj5rQ_9!^LZ4oGE-(XW2g
zF|;T|fgI{4#slVm35uc2gQqx+run{>gCket6j6xGhQISv60W}`{EX#q{Z-y)P{R7h
zaQZTwJ|(RxEsO`TJQ8-A{+6@~bS^P;FfBBjhsa|o=LF1JafW<iWwexc0;R~Wk(4DY
zGnA%zyQJAhne#{?cOD{InL-{KS!>2u^w7vkBh#{E%F!l&C7B>Ikc&dnZq)Gf`}4Pi
z_eNMBi|<J13_>!blgJqjEp)Cutxy)DEOM!cn7}i14o8p@5Hqf~Jk!z&YlzU4pN4{^
zDBF1w?$cp~P=2h)a{CII!hs+&dQT4SCUd-*jsCKzH07g?8<iGf_`1>(^x;gd*&$Zy
zy4o^yGwS(&1QsZd=J7RM<DZ6fD_3}{(^@*EqsdQ0gudn@(};uP^K>*)6NZoHZoJDW
zoUJf0?htX*WPM!YaV#FSE@u?@KGvAao%#lwt=UA<S-=`Kv&kgO5#38G$=}3G_t(0P
znG;*T6E^$o(z2MGYH3+aE*_V~KZhDDZC~eN|H=!0+7bn$<x*Dw!SC;7dNFR@*W{L0
zG8E1X8b)yQ{66Akx4_YXFuUJ=>Gu*(x!cVR+M4~=G|JK6aX`>%LYQ`@8yyFbHdQ)}
zH&pV1MbnR8Ul^yovzbX8EVpl0W)8_JP%(pcrZGRGxSWL(ZNWZx+$9EutA~Kb>mtsc
z*Con-$?Fmhj=zq1U4~Yfmq1fpL^zn(6s}&1Dp9VkeaM_6;YA5X{3fKfyiLjR_^b^n
zILSR}B*|LU!Sc$mnki9jK2uOORZx!HZU*M)oj11g1cfj#Fl*=0aS>@P{D^j*)RH;q
zS*B_!-rvq^DDAZK8t4wi{f6RBale(jZ#z$aD#6^;EWDweCn>Tzr+ez9rnsk<X8!-_
zcHS}3ay#!q(fYOXTDFh1^BSz~?Yx$&W9___tMzuCl-_7P8`TZPJ<qP8{YxSWQ~XRW
zAWd}Bds=ziAF}5>?e+5L2(4X<wb9EXl`&p+{T}Y+Nlx(_-2a<--@lnBD20T+ZPVa?
z#gYF7NB+)zyh5IXZM+hre-}>}%>;^$=}|**R`qpvckvp^J6*g6x<ePQp}5n<Yv=Eh
z7F0o^M_NG6MN=OXjn1!hhLw+s)*~(OMiw(A?{)adkkZXuqhDSChWVp#%hmRd7Rt|3
z&$#N_bQB7i6g^*SY$$eUP&E{H8dME`^i+>I(u{g!&5)JpV0)0;oUb&{eX220QXY5X
z$&Kq?|2zWscP=SNNv5ZB)f9U;S50xxxf+Q+{AlMo#O>%@HT3s)u4}p@<FC8f-L2gq
z7TUXeyL&Zb2*=y0?(&pG&PI0A{^aGJYGQR$Be!mH6J?}P)C22=N`x%&vq_47QNMfY
zjaEugDJ=Nds>R0!P_9YpQKG78GqrbOMHb{!o~f*TLfKTR>x3oGzRtTUL27pexl4Uv
ztDP<9grD!LNdhX$=44xw^ydwywo8&z$8R+fQl84Rw(F7b_V`6>8-#8ti<S75;%5Wh
zA$~R#cj9LQ{lhosW8CKW+3NOx2X4+!;}LS^N(z|auB^UDl`awfUFU9BswwA{wB|}R
zw0yp!-i@@lX468xwmqrFLcX-@$H2ec^Yo{1r{GxAbE%*`_Sa3-8XXTaBkr|qgHDZ4
zwRluI*i=2z4s|m0iob6+)W5h=Po$PLd8jc;iLnsa5d0o}QA7T$TMey$on1GLAD@?R
zsm28HSa;OFh@efY!Q31LSZR*x+#H2ja}>k@e~nO+zovF8&lI!ICTiI4%?n4%REVRR
zzpo&2MDs}Je-N(Kd`L&SNsErmJM-r?X=PbYFJBMJhSvu=M2CdeTrq|1!fk?Y@si4a
z0hjo=cM4^0WOH<KWnpcX+(!cl135V{m*F%y6qDaqCzn@RHj)AGx9VCp*a3e88Y9J&
zRxV|yD1o5#$4V4rvRzr7w0I<7MUnR)n-I<^5>uy#7Y&0E!8zmWbcFnGbYwGve*Wb{
zTuMvy`Vu8*>w3vgA8O@o=SZyt$xvjlUUy887{puLmO)AiJcqdDwzAt~v#V6%I4XrX
zjuPBRGbl{J<ES>Lpm9s|)k=T3s2IEGP}+(>sSxMmmSwIaUg@{uFzje7<5*IB{c$As
znt)7x7jY?)MnsMoFX=E_mAp1R@SO4Jpw3iVIBR>Va}BY0J@YZ=z2S1TwpM^u9P(EF
zpQACGeT!N7n1L*KwPjWz%FHkJ?~><MVXsG?8@!;ccanxu36yv%X+3}P+$dzVU~<c>
zNRSv7s?jge1t($`;xeqX3L>CgVAa{skx~Thdm4HTu}9phA-3WcqbNJ+v1o3^s#|P^
zq^!W~#I_{hY_{52{<o4vnKX8E2a_77Kzz*KNcfX_Z~5ZOb9$q!@P4Wi3i?T<&UV43
zQDwYWk-7AEEc-4-|CoOiakImx$>V}neFJ|{C}tAty+$RqkTk?FDOixdzE=Q`C!Es$
zDvl<3nY2^>mc*SiaP6j1H(#+#(hlI___6F%a{d`ym3R7-Vb9em%xCKJv1f+$xL<;f
znO2*zFAo3XG+$S%kL!{cYmW-&qNTTMj1E*I$gt3uCTE)Qc#nUU7)RmL@y+2=Chx*#
zv7%rm$!YRtF_EZT_#w?=ZM|^dI>r<Ru|rH@5F0Ut%`RqmrV*>@GgYD&H)1u_EiQzP
z?J5Sb(XQhA-)sc28*sQaxLWe0fICT1BrbXuq;$)*r^bybnHYugY$nQ#q*NnkGO^s7
zWw>rPPj(qjNoju>)Op+GOvhBC5wpsFksfR|3ANG`{^2|2s+#JIV6mrPsUi08D>cNH
zU#Vwr_gGfNgeh$4D|`&b_ZZrC^f3%#!^iOb-<6NqM@sQAoRs(MV;HWTd<=uS)5pZO
zOVamz-^d`gRFLieF4}^V3Nlh8E1fQZnAMa)Y`h-I?>T?js@$dLUKC9bp!|38`iQZv
z?&$PufE|2(4Y1+z895)DnIuw4tqfe_<-rIQR8JGDA+}76?f<^K{4P=^^I@dCYcF5p
z+RMw=P+MNU*qYP=5oZo$(S7Qnmh~<zwXA2q?~^9`NGm5zZX~VWNfVCtzLO?gTY8@~
z;aFWCRyBV?(lT5~XL%8@w2Eu{ttw#R+_nAIhZo}qFHCT;3R(706@@roXKBj<<#3PM
zKdzr$JLO7cmSIX}?6m$kb^`M^J*^0Jz({+TX<oYn*E$QTsNU7#loCvx)@~g}pu!j^
zRbV@<7ks<AhG4W%N^^))ztr`^ovhQ^1(Q~V@_m2X>DH3C+=7Wy3mi6D;I`2MZz*nW
zI&Y}qym5VM=yzj3hUhfHZvJ1Jg6tB3j{e3kEGDy}7s*B<l^w8`WSpO-o4qhAB*ytm
z^2b{3vR1tX5MD0GFH!l@UOlCvNGAM-=eiYO{IO$o#UWmQuA4z>oa<(@J7kp%Vk4{M
zboYNf*R7-yq%~I*fskG+^*sH`AU62I_wVAiR<iAy_<Yy~JSBd<$GepHDpwoe5V!L2
zd?=Uie<AT5^6^qfaJ==|I)csaa(WM?6~?%o9K`6LKrJ}<1{IA*0l37wQPKn>!ZsgE
zL9<pm7u~f%DD0`a6~LC}*1Nx<;?`K3^Ll?<P426<d$>1gZOy$mR9eQpJvzI=!*eQI
ze7*5Pxzjlpxjj=mfpFr`^A*v^i=YgL)VMt|8CSjPik^asJ%!*d++W4pE0!->V^j&>
z@u0BLJFdBaGJ+)7Gtk!%TY;YK53__PD|V{&^s?(46QXNeL_8ts`~uGnAAwXbqvC(F
z&onOq$Z?|Wc;yACu+h%_Pw%|!3KBo+&P!0jfbYraQKL@{`yEPs!xfhq{yT5D%vbck
z;1Zia9m~WyxDK!2=HMDH;O0}`(WRCh6qFk+74N#V;n;iYs{%5TWKF2~o9@h?NjReP
zOij=gp%^WUSNi8WyT%)b@uEPb)p8K!D$Q!|g?&E1-u(xvU-T}Q@w7A(0XdgJ#x@j}
zuznXPmltC;mI3^iu46WOf0Jw`D9D+AjN%t>Sp12zF#-(MoKY5XP1Kfzu~1WoHB~%6
zUap_tSwU)LD}e3!J>$>0)%-r%ff#L?F{oJ<ts}O`r{^-;3w>OP@&9Se_<0xyj{&Cg
zc@$upSX*JV8Vm(-?KJUANkt?Waz+_J*7dMXp0?z;XB`8=OS_u?e@g3wT|A<SY4~3@
z#WixHB7q85a?_^Vs2p$AmE4lOf{}Qi1yOrpwLfV=G$bUf?5#|+Uf8kY?xoLeLB$d%
z<<=Gqvh$-h<yYhdNZht$fVF1tDlh4I-2^Xcp~3eAZ^^L*-fI4L1aBp`J_c_|-50#4
zEI5I;vfw*{x3=IpfADTuvIX9ZH6Wqq6IbpNN+o!GDo1|Z$(aCKavCF^eI=(8v>cVB
zyzb;w0<2%nZ;$8qT6kLNzh$Wub7+sk@Q5ra8y~U&qd;80wc24)D@NiyEiCGeF;d0x
zE(fMK%5XKuJqMPMRcrK=wv!E8DQ>CY#3*gEpy>?pdz=MWfuK)Hi+>1$jIXIfL?5Fd
zezMEe$$2Cv5#Aj~P*GEPiILRg2?HU&jQ``s&+m7`ShnW;V^qF)vs|hSr{FQo93Ut0
zM@YG&gkl1hc7qezTcfrTAkAiDLodNKE#@WqQq3AF3ds)<s9=Yv-}E+_m|XmH^v#!K
zfn;87ChF1XGspaw-hcG!6Wwuu8|=_Z^|VCca7y(;sy5D<tSC@BPNx`z{5X}@XZ7~y
zX~V8(Vn1)as%O&MHZ5Jij_kYzIghM4j<yMwu>EAWS`e+Rw0G7{YhIqYTWcR^>3Ljh
z$eZJ71>ZdrV92M?t&BQ`P9e+In<jn)`(1=C+b_|NLbyC;1b;KlU`}tBV5TjFra9xr
zp15^e3Q}ro(XS1Au)Sqkw$~~+2eY#M7_#<*UfQ6n0RJ{EeHA75M{&KgH_!ToG26OF
z_?=guRcFUp%azZcYwI_B3JS#@y{_M|gC~w#zC`_o&qpD>?HHTX9L@DHNeGf@CEH%T
zCdDKb^*u2W$A1gN7_T_)#Y8FEt@Tb=93Py$NGs~tDaw#`0+ic4K;{T*b0pa@hVE#W
zl4Hv*)%;z2)?(krKuo>d7g#zFQo%{p&}qTXy{h%dLc=1j#cotkcpa6cdej`m<|K5H
zF<m1R4-=qQ<^gpm6#)tlZY9-zB21iuBlZM#!LWtvf`2%)1oZM4dX4KFFnUXUMmk~T
zN}}U+&G}9_XtNBKTs23hXkezSvWM$)6An^@lEs=#2k~=UAFWahxNv%T7mPb3hgdEu
zJEv=L(;sDQd2?l!As2l)5@Td}R+4l3dyZ&bJBJL?1)9Ka^yf)R@UhBaHfk2g!2?Gn
zBsp`RYkwA9Og(})DzJg$Gs2Dnr}mbtGaBOfQFv=P|13fg_PxQ0Efo6tn-4L8<1k=z
z2nwFM9P-tN6212qZWNyi)U|L^aqLjIsW>(YHx>1E9(tyTGuAn})PD(uWni4s1IY-_
zad2(NE;;E4{Fs(5bJApv=bO$#Wgv$7m3oabaep{taJ9I(Fx;^=sW{ir()EOR$+2Vq
zw&L9A-&WKe`nMIwM*p^+zi+M{`?vSUz1|6)`lM=nvh(S#?p3Wv+?zK@C3j<6@j@!U
zf=0i<7eeT|0%|MpMK8QEE2%nfbY(;3qT@a<81^{H3WlwNtW@VqHy}>Qn&a`3_`lwM
zP=9Ea_oJ?EKlH2*qGyMM_VbjcCqH{lX&M|WqH4jiP@&6#0J%KG%QmGnIDMjH@fiVA
z9IjJ{PIh{0^9XE|x+BkJsXL9qG}HIx-GdQfeo5Krvt%lr=Nq}Jq_0WRXYc<s(Ni({
zQtpQ-05dC@eYYaBmK;|26^5FI=~G8{Mt^9H_3V%vSH~?nG+J;}@-Jx7ZC+jJ*rE#o
zzxAKJMdwyC-PxiG2U^APbG7IYR#uYp=|0(9C%Zq@KJGdP7j97L)r@oRjhekTl@OF@
zcMz05ojPi?*`K3R{ke$bM4gbFIri((F^)KjtCynhwartm1(g6ksOu#v4BJuhX@9FK
zhK6chSpDM7x=OBFn{|fRQNL@i4nc(vJI3%e$J={#6{kk8uA=Txy017kYWKC?eS3At
zX)>echF)E$9RRtGcBwcv>{8F)#d&S?>PqEP!33P-pr%JgiIN0qs5$PXX{pyz&;Jg$
za#I(wmYWW{kQH^8s;rVBJhdk+sDFZk0$f?}9qK~i!l+uy-Drr03#0pXJi3#cv1`0Q
z)f2h}!xnPO+20YkrPTTuw*_%u*q*W8gxb=2?+Dt;dS@`ZW5XlJW?!$JKzn#QI#Uwo
zGobFdM{b?)sB2D*!sxWQ%278IW-;7I@b=32G2BQn^`L_b_Yl4lz=2x(>VHB=I(Y?q
z9`q&0)<Iv*4~u{&0eegeX<hsX0nsfkB2>ApTgkcijg$EOl#2S`E&~8N&N<5OaF+q3
z5@hZF^e)4xL*h%_W$-98xSp1dh@Rv>1L`}jE-d6fd2?YoqW{GO_6a@H#3j29jh-dD
zMw@3z`Y*bya6*H8Q=zuq^;&hg;{HKyVZGe=r8P#MEAg?DUF7e#ElF>={}AehWXl<k
zzdn}kG!_$ImJ4`Czqk-X@7%p`!Z~%f;;GkM%y|60wR~@~)_jaO>Z>1{b*Ad~%iF_0
z^IH%}m+`bT69GAwq3$UY12-@<lTjZgf5lhHj@vK{z56Tl1L2SosRbwkB<Z9;Z!K~w
zdMY||nnUOR52Y;GsbyzFQVbA!kty=&`H5!Gyc#rLy6vHR+B-G@qEN8r@M1_GpdJZz
z(2EH~;2oL6oB8nx(R#J58Y4QV7W6vmS=K8@Q9VdKNS_j|SAj`K)UEiFS}fOgf4^Qa
zF=L*wtur?3pNFp<R&WT(=gWL*@176cpRR=};(%i^zuvn1fE|I0){yle)|>a==B4}A
zh8>fk&M5~$^Zsg@k?)nQgvu6rFEU|^P$^{NC4$qyl`Z~^D7s?atL@bC%4xF7r+uWw
zh>Xg+v=K!)WQ`~rJzKR@q5)@|f5V}NKN(T@V`T4vylo>}oZuS01A`CLdgpiQG{4>5
zo<9a)H{*(EIe$|9#W>5K3J<Jm6V;$77LlSYPStitzHBOQqQ9?m`saBbt+PyV+XEOA
zO8_s<c+POcnD}|>#SAJr?>umbk6>p=JB@gX9Z!%j%<|vJPTs}6s?<k+e*+b_;I@vz
zIy=M7i@2?ex>aYmQLBPSdmn>V5(G(CGH5*m(u{BC3wwFT_Pdw=JA+OUI2UKQb3N58
z<QDNePC#gvnar{lt**GMwqF#l6%a~~6z??RCGj@-KTy04ZaFXBsrO^WdliGDcpC;k
zP`nN9aGiLUn7kz3L2JYyf5$H_o_0hB`hJ>1e%qI`3U(Ta+}TBrU(o5Sit@HErvnP<
z*8E8z@ofD|Ox2!4k_E3b2Bb<p4QyR*lP>j0x~Cynd0>y!M7*njIqfoH8}U>DtIBFu
zw5i)E<gHY;R4!s%w^?54QR4R$3kqOKu0(4s0sGW?=Z?N~L25(ELzP-QYLUoe=R2sf
zQrwkJYOI9WQ3uPNzWibQWE-n&9uNKECnNv9szFY<V)|o)T;cbUis$}N45Q6i|62S7
zXnLYlm+`bT69GAwq2xamw<&Kn*aixh2K4de-M`2K?w4T;CKCcVIhVoYHxmOhFgBMV
zumvZVU3NC0e+V|{@!_9e|NcU61Ag;F?v_;Y>}@sLXcb3y^EN2sIgi%jDHay&&&yNZ
zJb82TDg}dEOLPD#GSGy-c0vF2{pAV?>1I`mWNLaMDa!)UfW#F-FPm1PQy{W3^6u^1
zx9_qgOzFfDYf`XQTl&_t#L*~l;gX7=s|Lg~`Z(ZLf0F?5DzaZ8zjSP9<%Bcil-FT}
z{5NwJuZCr7z=xeOJ{?zbDyGV-KmFbu7D7^y_m~O%6Sm2JT_DeHu@(<c^~lf1VYFA)
zay^jm?SWw37!H773O*kA38z*F?7L7~dDR{mi^I+dI(LFGq;zuixby}&PX&(PF_Vsq
zOMfaCfBU**^}1ZkRl*KblT{RKMmoPDTv*aXGXer^JFcGa({G(H(<%{$ORRiI^$gFM
z=NJ%^2P>~SfrhCuc!UQ$hoM!Lf^xXz*?`YDw8~Oc28#uQQ<l=mDwBbc46KB7)|nxU
z2&tl)|Ht<H8F3UbGK$E&;dG(xw~)d%GPKZGe?rXE1L-U@!-{RJ-Q8Je2CSNcyFa{q
z%^{2~g*u?0ErrK@DfDLQVp7_tnVC97tIN4<=hL-jT3O+BrjHrvPmMY3A&6s_47MM$
zB7en3-5#Q2>|g=*uqw14ak<3;!VX;zvF$%AMCb!|oB!H@ZfyeVR3qQMe2x8J{S)$^
ze{Y>Q(8}9-1ZQy|3dTH7i<`D`9t$MIU0Ye`AI+j$p722o!Hl+0M;6a<qOi1Rqw*Uk
z-)wj6D5O>QxZjQOI9?Hwc$aV8hS;9uLOTf~VHs?H<#_ibe37^~A^;FR-YTC=nI0~!
z80^OoMwOL6+Iia|$Atr=y<;*WlqlAaf8U#nh#{m%f~JK03R&AW<P-8ys8RoFMimE-
z=zJCzi<kT!8>QFfn^br|?*81ozj&M5naE&z*3Wp?Kss6C8Sg^uD<9*Gca6i=a%8Zh
zhkQ_pdB(dYSO_9>^sd%zL(sBd<=nd$hLVOJKk&?%0ehORAXP<VRFE&t?cXA|e^^Ga
zB4}OC+pn5oXR(=~7(0fYev%+^@3-_L38$SVtU)Hn4m=S^R*E6(W&+7Vg?gr$K(fv$
zCYlK(ivdOFW&+6uLOl_mr3u?M6h)@lnKWUs>5%5XBu$)dOV}id&92t_DFg{sK`>Ov
zj^{~4FyyI2J`Pz7w>?jcDmD4qe@ff)WRMRu)sVL53GsC_(sn#KtDFrrq?<uVLS7$G
z?F>SQu*9i`v@?iQU@jgA_|jB<6r&137?6ebJRx56i%_b&1Jm!<E|B$-L`-@JAtw{J
z2IUfv?jDqgN$o!7P?l>5XT!rPW?KM;b~_Ke{5#&SlKGU=*$+Nl5c?>De{`Alt6}Ro
z-JdMB%=Xzvvsf}U3k&i-O1T=!g?dnS%Uyt1&);?7$Ma>4wE$&Xr3=bky9NKW%{Vhp
zVH>SY?d9)i)#HVQPWedK+kKEhp`Z$;R{a+dhIE$_ALp?(5T4DJ67uT3Z<!v9%iOzt
zZl%9g9{k2d+<3@Q`jqcjf3DghK3CkzQhu&3ya9hvdF|)I<BvJ+GLoO9ue$b1#lo@)
zUyM}>3+bQoW9h{lRXIBB{sf!$`LEB{`!{(`dps<wvSuZek-CV1jNaIw++{uQ(Tcuz
zsF?tUXg8BRT1L&6P=i>JYx=P5pC>X#pg`AFh&@}nQv~XXhy+_ff4;QavsU6wtVkKV
zH97v*){eN<A%IIFdV(8iJR0E+97bR7mKS}X3<*)?arHE9gsw>SxH=L&Vb*k*N=I2~
zhru4Kw0o8o)iy(n$yCy@qt`Cb55aa?pxb8!Iz!1unU>dz?k_(k`j+Bj^09ito)WJd
z21SGzlgw!R;f+Gue_^GOKRQJSVy~W9zm%1lLL=%bRL^C!^ombs({Wi%`l-uYzrU6p
zV4b2EW*lJgCvlxGIu0m}j5(l0^oxY-fD$$6fV0_j98g%fa)2LqfWhLmX2t;w0cyAG
zYsMT<8tJDjaX^XqFvflf8D#%#DjoI<D_87?%a)?mMr|+ff3`QTi(X9bphZRNxyMQ~
z;x2pC84K=m_g}v)v~t$ZrP2tcQmr)4rl)tSC^GX%3q8IlFu#9u@?_CSD52lqM`=Hg
z%}g|Q**(2Ksc3o3&hwdF>k3`{dAD8G75AB6b$%-h=~L&Mgs9W|TZwLecvX5AjNaA%
zUMt;|t}D2He=K~t-cc#lQh6>Tc-o%&@xawOnuAArno=iau$tw)gYmx9l=2QH*tVwB
z7N~eL)+-Ka<h?)ZsH_%<Q=Hpt?f3pntT=_*%DavkLf$|{Rhi*V$DNf+G-n3sDW$_~
zIvE$D8cwYk@4b_^gppw~&MO;-v};CKj6j<qEwt$He^ig_d)tb%o}5`)(=l*u<*Fr(
zs?r>0-)l)DL{pj~ESRM@o**nEO2RtYCE%i?DZ)^FM2NYHFz)QO@Fb-f!mK9Dnq~--
znj$P|V3e>OQ4-dUe2aCA+I)sEOSnj}Cd{;Z)zTN$G)0(^SGDGxu#16V70ZF2YG7j(
z)gjfMf9jL?yrM$hM0LF$Kk?>ADk=^taIButJELHb)|UTL7^|6AfV{gL%Q<q$$k#*a
zEFtTc7tn^&@_KzJ=NN_ku(s06Ie_Re#Z{0(IZd&ZA$PvG+xQ5m<7fDav>dpo`0At_
z)C7ikimw!x<AW1??FnR@uN`vdYcS)C^R;K3f0D0W#<7arA-*~lk+IlrYQ%4ygoDS(
z##xSQFyYiG<22!<ZaKPh34(~y6qkhAvOQ!3*W|LaQuJ4@&a<p^mY_jy3dj{VAIG_o
zqhtAC(HJ+Jk~%KS4mG8bqVv(FrUVUAQ$Vh$`E*%FbHHd5cU~Xlvi4Kaa&NZyW6|Od
ze}8d{|Gb#{^DX{3)8dB`fg@S=QD<h$v8~0woE^#)%utKJLX5TeE67-jpA8}_i>0v^
ze@Vn(i=T}gQ42FIej)DI7H{{<s&S3A_$$a*i@$=5wfH51f}6^LhhRHh=#Q5zenxVS
z`p>lZ%O{LAIGN0Hs&}HrUl|$dFjt7-e-?j*8tZYM&ZgtCn)GKbbJelQo&yMbP;_y^
z0VKz&y_sq8S4M^$P$7m}{1s}<0jINRJD{|3<p4YGfQYO~cE$nnQ;;lm##;Q9k)alU
zg&1q`SCB#W&!*C0zp!$}zPpqr<bq1~y|rvq@0#MSM%Kdz`Dll8_tt+)chqgzfBov?
zm<qr^gIx}Vr8N6nEcv-pz$HRf(d5Q@vJP9iKp<mj=(kwXHx6WLE8U(fqMd6iy_T$A
zjuA{*VUKV+OvbeH4V1OjrXqac1S>=XIjT{ez9jZc%V9aYM|!ZMvOz$e<22S);idOo
z4XoEy3Ccww8gekjkxI=tt2odse`y{lW2Q*+Q8~$96QpHGNm{S9Vz*^8PFj!y>NRKW
z63#@w9P^yy4A~fU>I`Rc?pfYtOmMbCO3pg%6%7fvFfh!SB>mvHCau$6!PBidc&4MD
zQJa_)&l$Vwukdn?t{T{wem>;s)7{9QG<4;MdFhsr>r?!pelDX#;_?CAe|9d?Dds}h
zA`<af`dJ~#DP!DF_AwHz)u_IA-&Ik<dhz?a*Iq+c>PbATw9)BLV#|+~X1Mb5V`DSL
z)dnfK>LndHT%2aef?|~8SS4A#ZZ1*)#M}f~@{2V}6J+hVW1OrEDajhkHsfUFWK)vW
z&owF5k@ca|l}p%aT!rXnX>@seBo%iF>a>5cr^{==aw=olsm?E_$tm<3oE!I%VaU#K
zA4TL3b$&PMOB4OZ@5X)!8l$B^T+tGTX^|gI(F`kIG^!u!`fjuoCi*?!jg<m3M#>#<
z#mW6WLH<8V<iDT4zW6`UQ{<P?M*|ZBIXE_#ak@1mmtd~}3zs<zGAMuDSwU}HxedPO
zSLoPYG$JWd6tFO`yQ#MYS|EsXXlvw<%?^r11AF02fb`$@qh{WCCiaV8ykj5sU=QDj
zlD;RBA4yGjvs6-dYpHU4Qu2z=Mu|%cL6eZgNJ+wDNK#b~p{Cj?gpyQ3$0eEd7Z5x(
z%*2o-m!X5jY)tn=rH_C4JrPUn>LrnkQKv+;TS%J7I3$Cx9fZ^h$NCLY9p5n|54Dcv
zdd0%MD8Ors_L7)5HUm&*?ICeu>9K8>#L5A)*Tkmrz8!O0FW3VpxO2Q#Z%K-3tV}o!
zAA2y4gDE6|WJAxB2p)UKiAlAXL71y;=vflu781T|MlT>?Mr?l$si#^3X{VAZ-aI6N
zQTkX#Q?1)kg9M{NV%VX?cZ|6PBlIpXNo!-&Q+0FfK|N_Hu^z)tIOmWkDO(STlh*C1
zB-QO`O={awP3q%~U<vIK=<y0xw!2X>E`mK^MdL9Set?%8s63hQ(6eOKV-LZ1+J>ZX
z6^hYXlC^N34iA4cj}wR4%r%y)5PHQ973=n~1I%tc+^JVhtx({s2Fc=-YfFvB`@j_%
zg^%>wA^AWDiN`rArB*#=sH6r9VGp=Wvl8pU*v)i|Vr6U`5>{>!*3j_ofdCHGd>}x0
zG#?1?uk~?eFk<Ti0o<d(k+BCH*|&Cvd-(PoVNNFs%m{x&Cmk~!QUpSWA#f-$1Io@G
zGvH7T_X~+3*}#X<TI!4MzFWP##Ab}Ze!cqVzy6JQYuT}#G|GhFxcc<r!`rj7uV@n2
zT;)(Rt$w`zaD8*}+q?JcBrzeaUcK9F);Cv)w~v4My!q*3^KP?FwEg1z`f7tc&f%3*
zKr6A!ISzjh@7lKy5$;^aa%X3&S2x%1FV>sO)vK51t2gV<n;pnE|NCu?&Hl3nhO5o`
zYV$GJV~hIr`s4MdoA>LFTONw9{;~e$*LOc$f4+=Iqc8%$#rzx0Og&5(THjw?U2hlC
z?e;G&2gyNxyWeyutu8+Oyosm3|9bV`wEE%t=9ho<%~-s=UHx_Sx7ClAw0(HJe!saS
zldK@9+9fym0b9uwVOAs~n%}OzPgwq9_0#p6>($HjCz7UAObjLj;e_+ipU=*2?bB#)
zABYXH`-K4y6ea^ArZXd`p1^<yHv7Vw2wMt9dW1E1iyvCk`7zdXzqMww@fsuf>;#{L
zz7v1EOah(2^az_0X;D23xQoIHU?i|}?ta(w@I4YJSFp0Pl9B8!|9BACydW7)Adsc9
zf`#Tl5EKoLiCnIQOQ}ecCqPjQj);^jnhB0b)v`s#uQfNm2SwqAl#V>xyikiwd6Khe
z84NywV8T%-&4K`fJ0p1`bDCP!YX18%RD^$|AyrNws&Xa}{Tv9SCSg?<*h7?|dV+I+
zV@8%BYMD{Qka>>k%kqe(<p_dqTn^|atZes@(vjAdlnhb+uvYDyBmF(B$))E=%g>ml
zj!15GUc;Bzb`KHo+#W%Hy_7rj*L?x~^-_=MFG}A%{q>33I!yvkf=oi41f7J5G<$zL
z1tD@qZB|vXps+$6pHa|&ieTbG&*J)GnuGTz4C0)N@+^j@;bfK^eUNm3L2((0p}<zP
zXhu2zT#2D&RPYl92LDu$;T#4Ox5B8ZQDO_KVnu~=nu7=e>Lo`72Q!0^!G)#Zv)@x2
zF?p@vVrUDBX_tBe5Z3ePD7PROUP*t^J<frsD63%+oP&%%c;^ZH(m`I)-)=!5<sedM
z76h77sKqIe1R825utp76h-PGq!NBH<rfT*eO0!z<UtAgpS()omX6XHaO7%ctPxn$X
zy4(et#G0R^B{`z>BeI|4_oD>ytz=JkYTfANA3^ri_GBNo3xl@fa$zU<B<z1`-MCrU
z38$!acLiQ`2aR+jW66jUq;Jo351l+s6rpI%SZ$9&8(2Orvu02ry|APD*HU19qmxv4
z6RELqb{zY}YFv^1RGyjDCKMB`&7uIIqclLVZJm(~z4{4INXM?I)g}~C%{B5_6o&8;
zwT4WCz!3DUCuly<$K>L=2Z4W3!Y9oig)*i%xDSQ8b}m&8c`dH<9<F$gwRROMdaf$%
z4?tRSr$UVjzDJOj)LYV`iK1wtD4r;aX%cu6WD@Em=p;<!$L%ytd_$26ia@^>(Tu)L
z<Nl^2S?Y5|b3=7j`yK_n6J4Xuqo~N*2x9<2ON9EMw-cbK2%ripO(=gro;fZ|=TJOJ
zm2#MAP;{V+M^o?UB07SseOs0CyFoC@HXQOE1oxgr=$Y)HsMRU%bap7zfui~>iYH2U
z2oU$tKv3BQNE`{$UhB^qAfSQ?MR2>KXPrf11a?v6<$|=p&G-~ORUH>Aw<wfRN=MS2
zLGdK%Ztt5T-5vO5C4hfu-<<CFfinJ3VSnK)a0+V+G`V?hV$YiskM46$b!#K-PEOoi
zM=#2Wx2GibKz;G#C-kkaUg_?u_s6_?Ef4tGfx`Sf;(c055t98Yno62Y;)hqy#tRe$
z8Hih3qKVHUzCe?mvg;B>wcM&L(4>Z@$d<r_FJz-rHPuH!cI1C{*n)P|-s;@Y9hBXT
z=Q&)RdE1Y!&Rg6{>8@BEtg6!-PP8iyq2|$T?$lro+SQLxi+y-n4Q7y?qn_KCV~_E5
zz4J?B{MPyiUsqbvMjU8OW-{X}#g=Gt&qV%JO_)4>O)qtH&JSzGEm5VmM6<!`n=a9e
zA4P?hXjW7gzGQ88^!<6!l2R>~vS>+(NLq48Lg;&`FVIBpMGd!PcZM-rU2>cZu8Riu
zVNF7%=(@x~NF;Z$C7NijTjd3ssMygVEKqb*GFqP^1I|;?F<%_9&n<a;!9K^lG)qOt
zEQh5F=^e_O;idlr`^7EUm+`bT6PIxbIurvlHaC}1iw-9iTw9MNwGn=wU(t`dipFcZ
ze1+XnW(JqJf;Jw1kxqI~AD7E@|H}3tr|%{?y}i17fA#8t%;ZI|%IoR<x0AvT692Cx
zUOGE@rGz6jy?>m(crB%N_}t-Rq$I+}<KF>Kdvi@HV1&X5W${rr*WM_mu{{1pxlI2l
zKDSJRkH&}T72bKw5lkeW<u1)aH($N~>`JBufBOEaU7*u{Btn=zJYM}YVcgqk9(*MQ
z#bBpjCVmYXjCIcANk*wS{fr^6{`1|pfBo>a5+L|5Q^?5(ZB-1@<MP#7kVQ??59xal
zT6r5LC88u}CMSi`3Ij6X`r^ZMWwaNHY$)FTz;B~4!Dw9&lM)m>2CfW>S9uLQ<N7x9
z`=NZ9SGBQ!m}Od*!QXh6M9W`6pQ-pOf6DK9D5(BDU41)Uy<7De3ID^>dFguURKh3^
z$0<~rUy&ElN8_?OqL<e0ye$P?UuxlkGf<ok!b^pJt=v{&^P588{0ejB8S<<AY~SxF
zppkV^$>Nq)*a$i&ykn<SNvn%yw|{?lOl04I=xy<TT&B5(UkGE}bPfFqSOZhd(F)(_
zMQa+J2B%=9+Z(cY8U(9FFj)kdxlhmAIj}O{UVzneVgc`_VttF8*Q=oKPU{94#aeSh
zIeAeXrH~bdHm&UBuq=V`h>*DEu(+tjYF5k5wGNuUA(I(o9TWJC;c*X~<MKD-k|Ozu
zj3pm`@Jo~zl$^chG54(GMcLNnS(vAZ>$nS^CN8tDoOl+P&ODU!8_?M_$~>%d$Ia9;
zohfO{RWSYhhSHNA!mq2+=?&=YeYtY0_E(<uvdCL$W0l!-wZ)oEwwP-Ko|-MPta(}G
zx+dRo;l|p@H<Gny>ZB1oR-KXZ%meaqjhg3wV}7R8Y@+gu9ukGw^tDvHSz=z__9p2n
zeJdp1exEJ#NwT&}qwbRxw~-fGoh$2JO>9)K6SG?YE<yX_R1Zs)^?EC<IrRig;2rgV
zf{jj)w`-9314Ww9D$=aNx`M_m0h)eIhMbj`XP*@}t78OYte@-4ni~tFnCK6<r(bD*
zaZL5Py8?`IualH6(=5)a=q@`~qF)WV(ZzapD_6c3>s%9&D#GHvT>DC!f8A)~miG4M
zUtP}6vdpfADi{L835ErMEOEu}y#4$w>=0DX4iqedk_$WJ;C7pjY+-=k^J-+0GNrCL
zQQj9d&EI6zYFCQBn7Yh~OYT^6Y8gv^dro8tWzOn$@u&gMrJmJIofAOIYZyHwce`Ce
zDjVkiAft*}M5Kjg1<xsCg!TaN8V;0#ddZ=GRL*4nQ=;nNfm&3%kd#<2Ok#qNachC@
zmvTnb#t%`IvU6T?dIE5Dnt@rN3}b}{C@E)uasM)r(B$A~h?$DS!a`UzFc;E)@IvwZ
z%fsY{h9dabijXVvNQ9~`Ej8__zObsCi;ut1<$jeY?YK??$(1qfEII2YadcVIx3^j3
zx+HI4rT(SN)!QtlaYv?QAV`MVS=A-4^-Y=SZ|k)W43hwiEyBRw9r~GZQFz_>+3lg9
z8Du#vpX6vJkWd%qEs>r76UZ)q2O_etBDZ?6UyN-MrdkTElG|LfTiq5sa(SF7wlZAg
zl!TMh2va)3%E>QgTZ?Y>%UVpAH*s@r8>(NDsf~&?<)Y@DA_`47rCLp7J+Fl^7OT5&
zBz1>yi5xiSngJA>Cx2M*T2jyb;^!B_KUcvp%FKAS5tm|}o)#BdHl<2`@pF{wxF~=A
zzLU~M06=MhKJh~FC7$YO;lhQr5h^@KfTVWhGM6k{Z$@)Q%IN7CZK5HG_ZN~pACAi)
z?nb?Sc8o&~t&h2|LSe<!<P{KP4&VC}!fr(gqkag$UO4ajm2T~UO3s&EPT9Q6R40~E
zw}rOu*0*Kznw4cXTNqay@!gGxB1!CQH)R=pKvZl#aGF1rQr;FLD=FpOmuiSMK^A&Y
zwu?YLXIG6Z3!M?I(ZJb@W&Gskx50=uL;-((d9A%4Y#}fJQrbL4X?&#SU9E!e50z$!
z@UrrS7jKr;sbf|)uWlp!=;gI?Zh83p?eU|RXHWo&4FKAT0{XSN?&(vV&iJRVe%^?&
zU|r5X+`Hkb1dF)<_fThytdQmdW~;b-DfL=0&0bZfnNdOY`SIEUP_gF|SfuZ?w8wuz
zjSa#fpAtZTR_4@JR5Zdz)7qp!z=vf*2Bb<?OLgppSQo%)LvqRuwT{AvM)4o_{4jm{
zczHY?DnL6vEN|tP4?RCzkN*66a<F@tq#ufn4%@UwYpb;4$8j&bs}?9UpnKvPq;2w(
zza85^^X-v_<V!&2f&NjgIfa%~zb1bJ{C3O`ximoUEkm`~{msj=Dc5KcG}kO*=%o%|
zUmAyE;EjGJ*q7!+RYqk#2m7orvXN~7`@9k|^2!6SussL+0Mc6xj$oe^+60gQtXqE$
z_94Vn=*zMP`*>%{)g&X>=LO#dsYph!PbVkuXlxJnWfCR$8qO)?7lk7A6!CuvWmr;p
zlK2D?*>D*kKcfUZV8@g@Sj3Cy+K7{Kh(j!=SUy;jrl~oh%Lf&7#(O{)UM1-A&tonY
zWr}eCEM(e^>>UMQVTKo0lDs*t7hE?nQLTdCj#uUp{%FfPZng~xaH~+e)p5z2&sd{+
zV!9>7BDuzb1MD3gdWk0D2QYt9N}>3g*LjyWT=boOipv`ye5q~U<#$dyAOp=&(vf9n
zVJJ{e`A(HN2GSAMX~a1PWPxLZ3bdUHFB@WKju|19mA;zc#qWpGDP$Up=%`4BUW9ap
z{*7v?<6^B4{4UxHM(S86hTs(!I$5ET+D?+!)0BQjQ9jKSNjRH=@b7;Es7%OPT#V4E
zL|x{siy?d@7R{NYqaX^!Sd$Erj*1My8|58Gt(F+7{8D^l=p~(P2!x1d;#wn7Qrxlo
z&N_g3^z2|pxDe0uLK<!yWNckU4p0Umq`j)CzCOk>gaksqUfc23hc=fXRw4U8joIDB
z8(UoxEAJKE7AM1gJ5PUEXx9sD?ilF<=%OIZV4C77ppT-Aez2k#<9y%~LJ_Q}PU3uY
z;Er9WVc5s{jDpd?YQu>w?&JJ*90_`+-Q7GT89*_63g*wJ@3GO!aX<wvqU441Rv(>$
z`N#pvH(UlV9{_y5eNDg%#-75t5-?Aj9a;KtQld)EMyM<+$AEvDV#Db7;Z~|?y?hLo
z7_c>bj$M`UDsNX`y6PdqB0nE}Kf$B~xAm>&22F)ke7C!z81xjt$4*jBGUzGLw+4>Y
zuHW>Xz$2x)11IR~Y)-&%P)kDB#d)`>K;+C0Y7yYM4;eHSe8}Payh|j5J=IiTVLY@V
z88sEC@4PA3^|XH-Br8b-F5Z_uQw}k1u5#qhZT~7MWZW_M&N`Ht&-p5YkGOdL%Zr-J
zNo)*}(#c?RueWoS>YLjM)vnj>E-WAM7i?bq{Bm?MI>Sn-XLqcVULVYPJ_%A4ljDNR
zC^NX4VvN!GXp~$<bh6%pLNY?F*{$Ukk1&Ese2&1Wv$B6zMm+hR$cjw%_V;+_o#u;;
zl<whxZ&CDj5u}p<y}qXWd3b2bJNM{yjXq;W?G4bgt9cB7Q-H(AhGdMKfKGiQ%Q-_5
zx8blt%TF~r+&!?=QPsr9pn=dTE9Yx1j!JA4j#}vkH5SMYnWiEc6fy>bvf*;v8<6F8
zf_7x-7czefG-G|M?olD*#xx)+$oY!!nK}knH7X)oA@WXDG0d%?8O3)zgF02kkk|OH
z=t_!F5#tT1M*6|VScJ?9>v9?mMeJ^jVF+0^HscC9b?_U(ivK9ET*#n^0lM-mm9f7+
zRgsidz}CJxcDyTYWNt@~d&zGneOo^DrE2~`_$hyfQjB`4fR9}`6@JQvxz(<?9ETs5
zFYRlJH{(i#dS8(YwxuF(vo~DEB@dQoP+U=sYMp{2%>|?5Z7D6`jV83lFYaRc%?<2!
zpDW*de`xN5Z&!_V`)6b2t;EisnVDi!uz&MQ^Bo09VgJCrFYaV5Jj%th*S;7AVnr3j
z@vVO^2CzoEisJavmo~_M@Bpta?|ki)Py~}ik}Z@lD%D6yoQ_M1aiw;=l|mKKNyvrP
zL?WshcIp(^i9jyFG_-5g*@Ff_C><wmiE>))?2_bRK0MnUxI85?&&exreEEl@t<^d>
z*o*VzKAyb*$H%?e*E|Mb2eF~7n)nLFx$A##7$%%nT{(7i#pPGmUN}HHTsC6OW7M(B
z`N~gMjy}$D^X;u2f7@%Mm5^|>5axRtXHbsk!*Wnqz)*h<=7jmw31v|FaIROEbM`MO
zhOIg`)pfLuP!9ix3tYz;%MpZf0whQ$&(}EIbl1vz2qy|{Yb!52@-_P6PmKQu$j1T=
zEtm1MG!p_jIhO%H1r(R|nKsP<$hX9rHsK2duJr&cm%&Ik6az3YHJ8AqHWZgJHUcM?
z`=&Okf9CNIAAfTv(uDK(GE1c=X~NAqnlP6%%V$wjEq7}*IV`ijTM*V_Fh03R_4%t+
zw=e~(EL9vaViHsl-~a7hqR=-x1*g5KDa@oikLM<9^i1+EiDZZER!!RY%@kiSMawL9
zal)C|;Ltq%=X8%q-@nE;*lHHB7%-YbTA(z&e+6oYTzN24doa~3W3wpmOh=ES#cE-&
z9L!upVN}B&y3rIo_+L|m$p@woph)cbF_YNtAW~1lP&J5p$t12j3RFY{2NG~pC7N%c
zapSous0v|jDomPOGq<jOXyvDX0ZR7W|2C00*Co+ra-t#m7E&A!q{C#Kt*QQJsEL}O
ze~I7$sY%&TE8@-|1mT!^j9<Nbkk&3z-C2{t?6ymi<I$|1bZCbAVr(hdn4v{7sI0B*
zkXU1*Rj@<RwAc@f<CU+lJy4i!B3j6N4=CuBKVvuG17xPFaTm?#vqT0vN$+ctFW)_A
zVeiL$QSiR3yCq?g!WaDIsxLeL=I%cYe>tb11>U%pb0V+X`JM_XD1;Ia5lm3ZB%*ot
zUwyp$%iZ=qn5fsc?>^oA{tx2zANcV%HvuX8+h60YpKcnXFxgMH-`#zfgVDj*@G;IL
zYgA*-8JaYnUa@X$B#2Fuusn0{NBEiK<9HNn;U1b%K8#0!XIGNBkDw6z8PB2+e?JUl
z*<DLt<&-y*F}5c8Y&=RZi1dwO|8eYAgn$PRJ&tW|5q2^moN}b#jCAJ8a<w$L)m>Sx
zR_qcRrvCeQxd;Y^LSJ-!TVoF&DJoN2!ujlYe7cfdmoME?5z3W}fqeO;c3=Dix%{>~
zYeqto@xWPZf5v)?ZyOMI{tNi1fB(xjD{3Mz%BYQ_WyF+b5J5?XA{1RLV>q(u4!O>w
zL1YBYR&w*`606uQRw3^u67{6+UxwHLz;aTmQPvU&=suO2ECYd_{`O@KPE6F8>-!}l
zjdT*6ngTU}VvI43L~LmmHHMhrE&^veJBI7z6hrjzSgKcP%sJ+o0I@abe^ky9HY#2Q
z)OJ=~j=JzJ>COYmO6MnmFi?)MO16))J<x+9kRUveJ|iQ9V}P~pLhN{YpdiGRYI@&~
z9jj0i0sdDc|NCw<ih?4j(2SA;9-LN}1h}Lo_+}g;;LWUUruD^mC7Ru=zqAqp{MNQ6
z`4;)$cy2}`eAV)~7vq79e~&@3BKctqgn&8?piH9$tGX^;H2G_YBzro)s`>njDDO|;
zbG}+BeIMXxg1n=Tm6No)*ec%!M>IL*-vnE&&(!cB>I}BxY(gVZtXO@JfEYOm&)@=Z
zAg97(Y^9AuOln3MV=HYD)3xb4g1yYcZ#C@zY&c}PzA>Bwu;F9=e^pT&K&ybrtZk2B
z?JXt{LKK}^Eiw@=FRhjYO_J!!Y6&TV(T(>SLo1T7PhqdPa6)3w5TNhFvqj1kgY=M;
z^5I-)CI~&Wet)d<<dFe!K2eHr9GeMY$tY^qR3Sq-h7iKXL92NlK?pbt;d=8dZ!xB?
zX??XZN>0i&^FYpve_Up9<U9C8TO@fKxk}lGB7Uy1wdsFdFnf7?h!S6}6!<t5$Rb#I
zO}b#`PxRcjz_rVrQKt|Gj80U3jSErNG7Q2+^Rg7-iZSAIzsl%1kGUBm)+LMl;AOqu
zawJu34g4nOLb+ml0&*dH5^S#p<q_lph=eFXGFPz9KrZAFf7Z9tTg=s(RM|vM&8!M3
z$R<pglb!TyLYDkX4A;&!LQT(eYwagK$=AZ2S-CtlIfg};RASrw%s8`LD|(4#<NV80
znaB;zloM8KsAl=!4zf>n_o2`7ok#^2=}oN?Wqol65yV0!s4T#U4Ety#-)8_My?Yxp
zvOH5y&dO70e}oIWjOEKk9-t8o{Dfo|Y7oQW88kw661ExT0F7V`LHoYI_A!rY1$KZ&
zESB$@i5;L3r;ujqn&beEAS+WM^f<P;#UMzCb8y7dL@b93%MEhU;?i<qtCMt%tsTG-
zfF3eeOpV}uopl>iljC@gxayQ^ogzCTl^v08Vul~Mf1K>dDo40DOj|fE%I-8iMVF{(
zcAI8#zWpe}I|{TGH_Ymg<DC2{v;tr`X-U=7jyBaO%sD6$OV{qq>_i@j5z%rp=XKw2
z7ZvP<p=wjDjP6?=+qj=89+hi?hw-a4)A*FGQj&qm*fe921s{YI%;Y+^5*zSbWaho`
zH^9o^f56br4llR`B(-e!DOb#aF%NCFODM0loNt_~FMnOa=t~O3Uh1cec(Tr}6k=U#
zeAdgV?@LW)yjM*~i?s_IEXY<MRRtSTRa6-d>?kI}$HQOG717PrZ@$LZUiP=4z`bBM
zW9!5(+hDh(PrA#Wna>$=6g-k?>wGPz&dXaEf0b4FDZ|mK+`3x;K?ccngl6&!CW}mq
zm`uuoeXJG{Cxu5tQMn73k(8V3iQ2o)j$C?bu_LEqxp+|~pJ8*sil;Xd7)d}Au@P9w
z-CW1a=C@!#w=s`1Tqehxn(ctucg(CMdb<U3A;-9XPK{hlnd|$C7L7b5oy8HC;WNo=
zf7zyxJDn6`N8j!$&rY~WG85(2O8R{pnb~I0tML{Jfx&?gH#~hig?P+K(=VV9Wc{Rp
zRh!B0<i2EihfXg>{ha2L5J_Gd&ndP7s7=&YIc=<Bpt5f-7RftH+_O^8`R^f?e&{|U
z4vlap^oG%kB}DcuMK`xxv1#K{WiZ-We{fWb=aOcH97%w*S}`v=?pTrS6@8X5ON%b^
z{WOA~o6@*@s)E@?&aYKxJ9VRWfbWp}!i^>tE}r3K@dSw<GIJ|5rK57Lv)_8EV5wi%
zC)UDL1!#Th-I@)c@>v}Z=HIWXludT=iEc5tuRd-TuYHK645a^mxj})xzC|s%e~T3W
zbNlNpUY?}lt#dJ1DIIS!2aW&OYVJP&&mX@3*YEx)Cy@O&-zshdOJb1!Y5o)VT`Di#
z?WgoN@4;E4ZgL9HX0De`+6(Cx!sWtuw>wfut~ASMx%v}d4ahwy?tElojoi{ZzA+C_
zT*`CcCoWg>ct4dF^Q_h(;+mFefAC-Yltjyaf<9C6pZqER&bNZ<Pq(}8Z+9OOSB>Q!
z1~?lrKR8(Y*W;?3NIES`&gM(nPhLr%nGzfnpntJk?Q#)b%v<)*$|G5p_}4Plvf_ND
zSarUHXXPj4OZmIKzQcuPrMR*ten@*bh9Lb5te`UScINN;@9#b(x*tIGe|7OaKj(q_
zOv%{W18Fl^7rUBb6$`0Vx70b`p6@P7vC%%GXG@-B9x40X(!QF9V8{=*VoLMc9@fQu
zIL#0#Co@~@0Yln;je-$EPD=QN5qmY?v<7Y(v%0BC1NcV=n4KNxX<EH6pjJrPn^td&
zk)`9hgKOvY4<37g<AQ1De@rJ6mzZc#XsbiB1j@p9DwD?wg<J0EWanmF;%0v!n)%*8
zjmu#`7us*sCE82E^z+8c+a*iU({B{v7t971lSl^C?eMM;)Ex1Ls#%qwI%St<)h2KW
zKSgwN0F%e9iG)ZltJN$*-fK-HfB;vM9JMAA@Kt}1oknq?Cc*Jre-PqWQohG=s~)_f
za_T@ZT`?pew}`@RM3GX#u{cWg=>>w-<@N%>>eS~z@aJ27TY_s31ztOG{zE%CxgvzU
zCIIfs9rOOKqEfN5qI|;r?}j^109cSS56ZiXp;nE(HgGoq0X5U4T#QtDli<;|H8J0p
z%egb=w<$D<B44kHf0N~%=d5lSX<Y2gv&=1>y&PnJ*U3bWrk$OvNPXIW9&h+r5j!a=
zlJXeBPP5J2J#}dyKM~awpUGvHq~2Xz)&baOX|%hjEPOYe+FrDFGEftqh}OQ|UfgjD
z-b^~qqt-B2oVf98q$rE<YxzKDebVJd&LY`G@ANsn8}$J~e<a|Wug&h8aRUe}l@TU4
zHtCecE|To~R|9Bog0D%wg?|_iq&2*;=?`?^hmfUtxp?KV0j1#Pse2&A*X8G-5A&S3
zx@z<zALdybK>-cRf#<-51xTzJq-qlPtq;@wm)Xov8gZ6vJClEo(lHyVm()&v)dcmj
z)qsVWdC&4Se|NCQd%h~5*i-JW?-YcGt?Ha>Xmv%0Fq26~r9Z7i#8F-Au@c1sv|iCk
zLM5-Gt(<PbOA~)D&amA-j&5!#X}e9p-L2p935&3Hb6rFm7T~CS!u`J;pp7sY?PJdg
zd4CRVc%Y5Y&U1kqT5L>J_TYw=rQ6r?EZorYD%{x3f8Hv%p+$9qM0VlEPRELwTxj@y
zgaSA+fdtXOgmXa{@!HcyKQ0B~SvI{OTuNgUTFD^U5mc{r99qdD4+)`O>o~NMb(2t3
zlN^Rta+<4S{ekve1}8kSW$LNWYSlKhb`4s3EauVy3u$LZR!|-qGS+OF1n5qmi@7>M
zp6>kif7?NRfd&HLG(5bW$=@U9{(nkQdg2q)&7}9dIyYdEt;?Y6rf8b{PM2%3z43Z3
z5?JmV*43!@zaA?PS1}+175v0i|1%2Y`8c_6*S180L6>nA33`1*XF6f_$}x!kKGbU;
z*W8p3v0b29$WjKhb7NVvpyP&l_A1g)e=2*&f8p87*AxexjO62?mF^%rl_6u}xlb*)
z!&KFkp}t1(u@Z-;Rimi6mwoJ+v{#cvrh7rLkQYr(hE^U+GuMvNCsR|NPx#wYUQ-ex
zH<O6iNq<J^xu00qqm*7#gtm4m{1H5iG`R;Nx6iHno68+9<rAxh{&|dQ)UfP1*L~Xa
ze=p*>d|nu>U|f|1Zp5XE=LgV-`$~!8@oewqDd-~+A9UF8{515z4KLY^igGW$;cGaM
zWWKV<sqCev!d1Pdoo<mg8{!VWFQs`UeF|J=qP}MD6eqY)3OrD8nNRu_r+xeKiAQ%?
zKI}(DgCuqix=0|2pz8U3IrJw6X2gZue+$M6lR?1sSBy2EqiGapCq`q6QylYT=S&?A
zHs@>V#)6&GHq>om7`X!Jx2+IQtJ<w+bRxy3`Qd`v@n)Nrs^ZIm+VL9&gymzU07XJ>
z4w&zqc%ayntdJE2rOv1vV|;#LPBpi&%4s6#PpC=8poouba<NV!h4De2id~lbe>$jD
zrtJ%s`g4p22vmQLJ1qBe>GvYbqhD~BSRPD#bIt{p6Ths)o4y9^)P3Mf9Q(TSEGHQz
z4IJ9Ae9y`W8IoA7JUqIk8P}{^ovVcxn7e8{Y7?RY^X5jbmxIcxhkb==MhEuAuh9_}
zVl!NnJt=E0t||h5U;u(n$yy;ue{=LGXZGd6aeWi9T17IpE~Ox1?s1{{Y{Cgq^MQS$
zjS%-ltpGXKhROOEpxQeINF*!D`{%<meIZagSd&H%D_f}sX}fWyIzh5~?B7EFrRUS)
ziS>UvcxjIU|8!0AlwPa!JQlR)S2={5^v8qkZsb8i)+7h&6hhn~zpy6Ge`Y8&qZ}kc
z{IU_~xeO98);Ece+6#)-H_UDs*B^}W3v{PHSU(&9zxd#;@LH3OM<g%sS{wKU+<JxA
z#vyEkOT0z~Nvs4Hc<qonDm&)2**jHdrR+0qyvrtGO{79q1W=FTw}`-x8+IOUd2@vt
zYd6pfw|<rs2L`JhfPt$he@@yfpR2R=SO?~6`861(qByWt-rrI3%6j-fNU%R=J%Nvs
zs0-`ij1!KzB=~HsJrV@sig;oX&^6*YM9SO^Ecqn?PM<Odtzy}x9ORreAn-fmudocj
zmo6v(?`H&QuW*bDN<nrxw*Sr%a<(tW-B*qf`@-{52@30wm!!gxe@ffyQemVk`+B_h
zP7y{LwOxL3%BKNnDoxu%fpyzsalmmoJ9N|>zzLG91~`yAA4Ii2R21h7Uqb#C2(zU8
z@!bO%&o5Ilhp(UH_MVb<{0feUfT5o?>&i2rG&AnpD>Id5g{mk2eE8NVersS!q2}d|
zGbqM({N7yTg^o^ve`*KgHKAQ<YEWL6na@?wTE&!Wtt;Jk{HhJDYy^<!b%v3X#;ujk
z9H1aEIbDx&;49eUDb*cguZr#=Ai4q1cnYUn1$>^Bz8d441LtUofb?@zUn5@nhTUV;
z=X3m${9CC$0qi)fFQ`5tyVJ&ASADXC6oAgFK0)=RXjMPHf44gx$kn#kO0<Wa07$7~
zK5<@A4WNd$R}|H7iOt9hifVuxjM802J$UoEU3yT<l-8f?4V}|jVE*=8)=ZyEnDMW=
z&4{m>YsE^XLws#WG3A2}BQ85AWOBf+F4v|G93?kDs6(zSO;m%k;WKB%mJiLbkHm+n
zl$*AWJzXs4f5Iv}w6NsRK{c1(b%p$kcLD;Ptt`fu=z$1dElk_<H%Q#IAM>>=pI4kV
zt48H5<e6tbe+!v7MFpfQIrVh#zEc_B*;6lJvw=JinHp_ra$!@Q@+XW94ZK51G4>~t
z+gK%a$Ji-CYJOWhesWXL;Y=Na-*LhXx#bwAHjia|8`qAzaWPn(Ylg?HNLLJazd2b@
zoG|!54MHZAm%&Ik6ap|Ym$4886PMpz8Yq|Sx;Cc)@wb$_Hk1kjCJ}CzQA-{Z1Tio$
zGnesR4HK8$ECeW*ea1GS0^&}$AjdWf2?Y3w9`={PNH-J#Fqc8s7ZaD-2_-3)fY3Ie
z0dcn>(KZqb1h&0hmzTjvHx!p4i9!^YU)(mz0fd(--8SI^(#OxY4&FA14Ff99x0lgJ
z0}}%=Fff<Fg#{Cr!s|A^e{wnh`SIJE%z(e$$xTj>w??m_H5LV9DrhzDhAf>JC7CE`
zPhUeM48j8mtfM`JmGgVhj2JaC1+f-r(|V!^^mr{R!7gO76xNDd=CkR4qO*ci#~&p6
z!dvzT`?YeAPXTXv-KO7y-BIBwolU}!sp=U;+|0`&l6Iy=8F=Due}V5FG^;p#Ms`n6
zq`FkzPG|d&e#VTKRSkVcm9>i0jrQ=p^ND(bN(&%A8D)jW;saMn>(p8a%P!DUN54i|
zv2y%=epdilQwpf`Y(E9+R5UB!T8reB>Q9y9nz-U>-mUsWIgMN0`N8>QS(H!vvOrQ+
zBBC4#4%Q+;Yr`=Fe@|LwE4H*HL9noO)D#N-<p1qTW6v$T2v#VCqx3m)H?g4xfzpiJ
z^3DjfZhSaIPeGY-8C~Znik@&!J2G$dIHAZnHA#W|I8A!`mwj-#`uoqvx;Y~>=;zKN
z>u&Pez4+9^CUaI@w+$I1IHRuE=BxCpQ3xcoSBLe>8KUM;fAx}n3OHkEJ?;*$nP140
zKyx&|^wf;}Z0%_c{n><^4f3pF{`d3g#W;5*58EMI#@x~8@2*}Q$m&hhW%}dULPh;(
z)Kj$x-F9{W6q>zkkPsrUuOf5(`h^jwE$0VZpbPy<@w%|PDizH))$KM{iWPh8c_*S3
z(z+G9d|7rOf5kg2R5Oix7RWZWYB3P)T&!MDi<A=HE_U4ux==z;9}T*W`qM!Gpg>>0
zr@YNBT@oo38<-q`n2T+s46ay8cxes0mM^pm-fWd~?`k=3r50p0A>PZcT`~2g$%}HU
z23>hgEBobDJ1@Ap4kmpfzEL^>I38h&ZwX_C%|L%$g*7Oggnxz0K+|(V(^X;9llt2I
z9dmpTq|9SGGzYeIAl1o@7Odzih@Ut;B{lZ|e%x&@T#>oV;c)G!rIs`5=;*iP2_a0)
zR+w7<pf;vo#@fjD)~ygszYd{HIlq_Np0efU6KW=@%2#i6vkY}=l16*K!OeTl<I1Fk
zh0%EZ(WY!pn15Bv*=sLg!y$&k(@SnL%w{`|lEdBX;%W}JO1_j3-^S;Sa7)c&%kIX8
zAst6ZD0WCF>`q~qg){Pnz{=|A|5MuGgea^THwzPy2wWwPP<Hm%7Yw-D|6-{BFrpX^
z`YY5`<b~G$Q~Ex3<UKukXJN;^)0{-N4Uc<T-@1cQ?0<^1&~G)9t#<LtIx^pH!|i0@
zD}7fSom<wGVVZ=EUCYcUW>2%9$$@uucr=rD)DP~V74Iu|6&h!1^DL>M&;F6_5Mwix
zS617pmS_FmVSmxRgP0!7Xi?Qn`)Y+x6U0N2;Iz<kA$7iM9qd6cbOof2quZUoBJw@U
zjQ+Ogoqy<oMfro?*;B_UL-ad>5<RZW+`_6VxzLnvv68Dimnf}2b?5)iK0nSv6Ytv0
zm9}Pj8Ns8OE4BSZjqHNvHp{KEgT!%o(pvE0txUTs;)Qltu2iy}N>|-4Gm<Vl_0nz&
ztK>;(-s4uiXmPUE3oonJP}5}vWZX${t~X;$D}Q-v=I4zS%Ur*nDfm^(m0QZg>5Gz^
z8O@jZFII_pb=6xEUeox2>3D4FK<96D*Fu-@q<0G`Q!yazI3H;;&hV%4Ql)IQ`b$&u
zP?BI-qwPc84)ZyEU9V<*3sM~416N~ZY^PEFiMJmhSFYZdW~wf;J?FO%tG;Bc@L011
ztvtQ~dg(pMXDzN|n4|j<p_cv3Nq=v?AG@0>?F2eIN|u-HyCJvw25D)rwGQXNd#e^{
z`TX(m_8+#5(=eCuv@{a|Ig>C=CJZ()FHB`_XLM*FGBcO)R~{#S#adC19JdXAzh9xp
z^hG0zA}JMt0VnRI0ouZ_U7+pZP`J(_HR9eH=QW!C`+U^wj(5C!cXPh=HGp%RS&E`W
z@#9DBwpLR}p|xs(l0q;Jq}hcMauXLMlx!|&5G5BZXzg!msMLHoMu>k@IkZ~NAQTwP
zZn27+uCCq!5&3j~Ed^rpPg@RDv`d&-Qt=VclcE@w@Jy9*7i*xF`y@5wM2cA&oTTbw
z;A&=af18F}io4e=WT|~Vj4<+LwH9(M(hVl85KC~80>iZ~=D@jd7d0@Oow;a%%T|FQ
ziZ3r=Eo{lUh+zifE<&Mlw@U>gJaAD0x7_U#>^AM|BMT*erEyq=DA{Mjo|e?QlR1S#
zqcN<#U^BXy5tH0~fXR}z`+ygPsn3=XG?qRaf>1DB7cm>;E@HM)`)ruL$Y_i)h2=hR
zfkNfc2Po9ir(n$1%r~+I@eT5K1!Cz|p)aZRa4}8`jpNxMTJ^cG63Mx*0e(o<K5a=s
zc-+TA5T&nwp=5}yuL1Fp!hN=a*^2eq5Jr*F7*31GXpCV@llp9ew^h0>4z4sEhqa)-
zQddYpNk@@_=5aR3L9wS@<e;$D#S&E8s9TW$<Bq|)R%M?JYTCvv5rH^v2|{BuE(n=q
zc{GL;GfsV@l!F!S#;^dB@n9%)we`5eYpdLKEkV?OzBZ#^Z9F-Mhk5$B2xFcVE|ySz
z>pg&w#w~+!q*J$rEUG;Eh=?I4Mq5x28*N#rmgMZ)Z--a^d%qf<zy5hOe80Zhtgbd6
z`~(o7<Hhh|^<n+-`pxRYnBo5F&#Rx_zW#3g>xF}hlDn5LG5#7Op@pP&t*2L4>(P+;
z^OGlkXBQoRAdmR76=WyqPOzO&UAUY)>=uWYAOE%Sr$4>D`XvnCt*?JtU3X8?<?zSh
z$Km@68XsP)-fS+omRPVCViGgbMA)L?@C)7p&Mt?i0p?#0KdfJ^hiBm%D@c3Ih<PLS
z|26sD6a4oeCtxvF#95Rha0Wof5&1915t({_-~(gQthF@cH8w#Js}uOpH2Rs{`H*Xk
zCU&C}WB|Ql<cO<`T*HxkxEGm9#Ei&si3M2#C?CTk5^IwKFFs(<+x1oC98WMipf);`
zFh>C%0n4IS7GO_I?QjbmC0Nw5rAEM8fc6P?0B6C>!I5GC0<fltj1i_tRqk~jpCO@t
zdVzm2{O#}m2o4#5%d3y?-d!G5pkz3Pz%E}m@a=9*p%gQq@@~!2pvn810bufOef{$N
z>o=<~radrXvsqnVg*-m~;n&R%FE_6@t3cz6bHESgIPZ8_M;ti^924acdd|H-dj0MX
zI6&YGY5TyS4Qcy;a!ift_7hNsOe36soda)j5!-%J{9&sG+`*XbJmw^XcWN?nlkt<{
z^hCCwbTu6TZg_sZe)Dp*xfq^5J0D)He%)N|9hT4WwH@MXy$8O2ibmUyh69ZrI|uDZ
z%nGs!Eb1%7Y`*{dk(q7zEHWsi%q1NQ(`^6u!!$bq^O${#%x6bvFM#)pnUg7h`7CH7
zW4w1&90~7K7Hi#>ytXTgZC4iCt}M1)S!|PXU(lB8?j)L|e586F$0u->Ks$`gK<Mls
zQKG36Mu~z3K&eLXuFJC--VIMD^Bt&xTz`G>qyPKn-<!?*55FIV)m8lO+h5+k_w9-6
z>z@aI8ve4r{?F@o|9rE)S^>|0&K_Fa80Q`sq*1*;02Xydh-~1;s_`fe(fRe(hhm1p
zmT&Z-6|ICm2K>Z;vY>K&0v}pNVt;Z%i$<q|kGc^|8k2c@fdhn=MX@hv21-Mv&-O@!
zt5F1;pz{Q)K#gUv4d6yGMd3dzvZRdK)pE7QDESrL(FxcFfmwmq7t~OH0TePBf&nLT
zicRg2oG79=hctPCPB1EZotwP)D(5mg;X;G#z5vF!A}m0CR*RbRBOM7#hz8ntFi`W>
zAAbpLGG{PPru&8C*~NHyaU1oTYfOv^JdbDG>tr=|DH~c?fWLqZ-Ne-HU1o`nl-VyJ
zN8=^Vo$P3?=A>8sN&-E93G6S?EGd%dB2^r>%!{lRqz4f%QsovEBw~RoPIAm;(FU-3
z&uoiS1=%Hwe9b6Qnad(o%aPNfWqD&s;Li(GL3yL<q8*@`qtT)rAm>eOk*@@5rM76Z
zdJX+541Jcx^A2f0*>7gsw;;@yHbs8Dmk%?a)7_cx7k~B{!!8ql!ya!rAAFfK-QytL
z$ZzjJJFWVSWsTOBU0(D^Gu>4>_B5upT{-$?+5%On5-rz7s<|S^%5AC>%78vm(M76$
zpR>#=`$ZHSS)vN7Vzx+?6DVI^vI7}ZJ*T$BY6g1yRq_(m<an^iDzQabw5^UDeEX_)
zs@7uG`qg>H93S?7VPr-IZoUt=<v!+S($So#(8JMYJMRo_M7|7dl=jf3Os%<0t+`CC
zx#UUWNh&9}>^ASw{Z3b7{|3n631EPR>=2!@D0;+N8;OGI__r`8VA-IzX3nlttR7k~
z%g9D_8+zn+6Obl|cS%}}NlIZooezEu(7!jqUv!^=6MRa4hh4B*ZNjRhq=I&bl+p>E
zFaq83!`os*LCO{H>tQs20Vd+@M#`XQI8L>m&<pXd19{S7N%*Ve$ammG&8?y5QOlAQ
z`OfWxE&z}PC^v0HVKk<P`#@S+^Y26vh$IybrFrcXT}0?-YTE)@fwM;n=n+*a6md5H
zkJs6AV4dB6pe|u<_l82Phft{cWl(6c2ZcI;olan<Ws>9sv}}9CisFc2N4KC;6lf^)
zxS->mm{|W=gMY143NkM1;W4kRpj#kn0BNuaE9Hbf0Bt3IO0|Ob>5;B_rUKw|r~dnm
zSW#yEt2ZuuB;Sd!XN-NwsDaI|*ys7Kb{ALP)7Y_pALq(A28&d=qRqZda|a5HHHg(E
zMl(8sk`~SHmt#jr^^y6B1S9I#?*?aRv;RHHoRZDD!5I$iW7iDgZX$*EhO6yYWiz-M
ze`s)jxN3P1uC_lz7=IwKlj<Z*ysl2Xu1>tJPEc1TsH+pyRhcAxIOr9$cuW>FmuP;m
z>zAoFNDyVT2uDzlxFzg1ik+h2X`CtN)1|uta&s*^=mn&~iE;$?*hs`0T1k-p1|~dK
z_ul}xEJx>;!ALh10Wg=rR4EeyF_$rkIwya<Sjmp$HVnPjSNI275h+p&P$*!^-912V
z8T2j4DVWJ=4w?Tygj#K>Tx)mw;8NL=$tRMJq^zfZCOv)K=`tSQcE7!{pTH=#&e`eh
z#{@E?<rlP5-tuHMDx+oU=k(`ZYqVO^Q7+P&ja*Ltdl~ugWXQ`3u<<(l_4a!aZWVue
zBDW=+*(q|=go{ukLUFFb3}^8as+ux_j1h`o6a__ydUF#LoWiOl1C&c-03DuW&{35w
zyP+b!{*NkM1r&iW=BKO|<z8`c)@&+wMmfiAli@%?MqX_r>79#YdB)X#9wrgO;BZqi
zvobDJ$zr*T*kvCXTzrIck?dB640(TjHe@zPbOFV^B~%C4m!-4-#c&CH0d8muI9jU=
zy8-x@S1aZP7q=iPnWEA;W6!vfP_2TEfg6Xf!`;i<?%!QA55{;76M<I%Vfy>$E=~}z
zvZ54G1)%BI_vy#(k5THtkd7(I8QNOr&KYPGO9T*-GQqB6c`@WeR@{bm&bEJb*T#E=
z25!Sq?Ve`+V)7K;ZtaG+vL`_I<or&`s(2T1&~ssQ#bcX;6Ok_M%xQ(ew{WkyfU^Fi
zv*wFNSk`4#Aw|OBRusEpO_ua$d#}F6t6J{z?vtKs0bN!$%O~s2WyTcAeucRZ7M^oF
zln7`T4J$9%)1X(fu3$WB?pA-FO(MjKyc^!uSmJ|MNg?g4CX;!Pe#U+k8=!E|K8me-
zdmGza?rT9ZC3+*?<6T%WSOqOnk60!4*n1Q@l3d=fnSMobo;>|_eP6w%#lEocw3Z=>
zl$-WY;uRf8X>gCRpH5ic#cyOzxLBA_3X|mc`%t{EqIgbOchJbS6wZHbk9@#b%}z7M
z0dZe5Tw*kp;$Kk=j85~fYgV7hFLgz{!^U^+vgr-%&-rXs1_iqD5^aX9w=&wGF~{QO
z%%ym&QNu<B)A}&3@mL#$rPDgeL8TQzW0ZxD>_i%XXg-rWk!ojLd1q}cdv_<&;a)W&
z(xtJv(xpAxmq3WRKdFB$LFwNuF2v}pdap@k;@Faaq&z(>KPdpaC*PJZL2VQmH==>(
zv&9VolqEa9@xD|d_~&=DRv<w`7MBL(K(t@zpXqYQq?g7=wy@7f3<NA>zmh<~l6cLH
zn%Y}u@JVA7$e+A5=QNaY<bqOm{{f{eNScu-9&}#wPdsIjtx11$Ibm)4Ewg^r5npi>
z2I;{Kj-a$;Ya#l6>0GlzYI=(KxIBRZL)T1Q*JmVhTFmZfPqygP9A5Ki$?&a$tg?G{
z(2e0i<dyUO&hUt*EpXve!$VH>@teZk;Sq+@0iFww6dL1!=}LI8lViFZ(g#rgV_^kx
z7-M)xVFhW=aruA#@)4D_Z#Of?PCL%C91}oH1Flc1oFV49&#Ij9n-St|J^I6i5o}N(
zvD7QOP#3{DX;@%*;k{+iyre97cUfnzNE~t%ePoQht#wKxS!nkpaVh96f5cWTNVe&f
znEva;#;NjDke4pkji%$5EX+n`>C_=SS)EAnrIv^KVId*TVE4|2ZsJ-tGGGOxnT$jL
znr1vv2pq?5PeO<P0L$8AmoYjn69X|YF_(akH58X5&Mhc^&00;5<0ufl_pk6DXa<TZ
zKBQKpPI@xZ-geZt)t**6dz!<}|37Sj0RwhyC+!*S!Es_k@d{o&D2N>X8RYQg>HPlm
z>8f$C!jZBtynh=MzmWW?BtJAXSfzx~YIy%X{P7|s7{1PYp_E$lW%;|~JDr}ja$G{0
zbg9#`MfEa&m4!<#qiL)KKG#lnm0ECI(mE@HJ#J0$dv!Ve`TpCJOwbEX(1MH~j77m{
zGET;W$aiF`#26{u=@eThZiP(1VqeC$cn&2qI%EU{xm+rN>9@_L;?ra`+L6T~*F|5f
zVX0w8+rbuIlTA7q9lVq$h;{w+K41<lpv{)iCr?y=LK!ezCbaTe+Jb1i#(JlZ1&y29
z>a5=KbGg=meloWL7iJtFBb21V)l0+|N`AUJwa9X<ghJ|TmRdJrLp!03o7K8BymGlW
z`3i4z?d8~G$wRx2(R-n->08rw6(dL^$omb9I*tse%osnGK%22<PQ}uxWFJ#@im5d}
zz>z?IfF2RO<v1EK$n&Tw^0Mk5ccCCPvzC}41%uOWkV2DjhVRBp7=EyJEWi2s<#zvP
zcJZjkR|$8HpoBd}wrey>3U9*KaO;14et-IRPGIQ;xZ$twPvJQ#62|&L8R4}Ve*8Ur
zd-}aMS}S?PFO4B;@iP;M*~;dKibAd{z&H+nQL*A^r(Q)iSw3QKHc!4*x@@g5O7*R(
zw{Ug6#GXqbgAJh3YV#zHjhBnQIUarJ8_`)hiexkm-s&jeDNp;EHTpbGPJ~ISl5FWD
zZIZvkYa&5<KTQr}IXs_K74fmkH5D=ArL#s?Wx}jE&JyV^r|4s@#jxdPTsw!$GCOX6
z3R&VESsy+B-%Xx#`yC~3ij859SQa|ZX0>mKr_%yi9J<;`Ts~bnn$~*LD1oNTLZP|y
zG=6a&xK3JeuSTF$OkFh-33LKcud76gsrW%DfSV^@lakzswxlG)O6U-alDwtBxTe{I
zJq6Xpcy1wnn;OnfYUN->*V@loxlH|kH?AlqUD^ojq!OJ$Evr}!lMW=~c3e7SA<S`%
zdC4*g{j-rJq)qq*$!4$vIU2Y7d$Y5H{DZxtvu}1o?%3Ai4D%f#&rzm$_UtSwwY*ff
zI!wAVk9JH%AHWEJ{KW6X2;hV<*q7ec_e!|W_hV$8vC7UcV!+9lUdHNyRjRLlBrnxu
zKzf)y#??>#q#nvlH%Y7+ZUe#*#74&%8y(2&*5CoOZVhg*x^w5(ka`!w%;~o9=KA9S
z8XIOJ)2c&7i9&7}UE+RQ>4Pjr9u(=-282UcOq@YleI$)ri%0OdwYWv(&ZA$%<XspG
zmD|#*%kxKg44i??F_~pdne}ad>5noPIrmX`qwPbO%NaCyb0l+Hi$^fGwYbIH&ZA$%
z++7$8bGtrT&G5$O4=pH^0680DW!I&c(;jbCU+bex4r#S^8yIdSEXVKTXxqf;j(wb%
z@Hs)=8ojhfu2J*Y`ZA`=`@GK@-HyG@ja{_TCXj%3mUujw8Al}S(<9}7SSh;;xmc*a
z?Ao}If7_|`T-zhh|GsnR^;-5AG((}V9Z~P1mVLfEvuAKlNX`Yj>={6K7TfjYxxeqo
zX?G;k!GCatJlG6EKLU>@)do5DwS8;t=JY+ubY}=wVeci*$hkRp?vlPFVjVzroozf$
z3^c+rIq!672j?H#jRSIj+{3gFN7*+s35~9{)7#O_J9K__kn{M#)aWH6b`VB_FQ@aw
zFnhWp&S0GIoXp<k`@S*Q=yi?AQwj7flliHK3GA3q{+`2xegXjH`R<OK0JN_bBbTyc
z9R>hXwwO)j6CygVi#q|&wLRFMCGN`cx?Q>>#C^wkisGrnxoJCpvFW_sJ@=SIDDKzk
zC<?;StFtfj!^V)uQ;uw(RKUfUf2|gsy4H<stI~<pop#`*rm;PwigL<_Ro^;K&~o^j
z&PbzFt{keG+K(9Rs=Tj`DA5;>0P<-@(h1Ci@mgwb3|~C+SZFaGdyhRT)9FP%JiU&8
zf-qVqq8lF%IHsz9$^13B`MS{cxzWmK$q%}df0Xo6xt2=g08i`68IxFuteIp_rFNS0
z_&AbKAK=8Uxq0YQPXBA9z6<W-MOAU6JN*$+<(!jOPsfUCGr3O3fRtu8pRUW)`MNSm
zay@Vir`}{c;CltO6FnBy{{V;%MhgmMZe(+Ga%Ev{3T1AW&PX>1mqI}-69O?fmtYG(
zDVI+NH<EwvnCR*D&*$F`GA&&SPvL90)xRE}5C0u{L6jDTd_%?xkkI_|^$=bl7(v!I
zN(m3xeEYZgeE7Ylnt4_C4mJl2n<O%(fK8Dw#x+D*XlpZUngzgcF%<X|@5V%E(5LIf
z&lmu75!P@IWB5;+$M}q78xWPpNY1D@Y)qnKb4q{HnVhi^H-Wsu{^Q%1rh4+}V-J}!
zP9U0E2>1}LLJR#2BH@jmh|NouPwgE^#lMUn1n*WDgf|Hgw|SKsw6I2PSfyrVD}u=?
zSJxHnHYp}6h0&L%AXU(XdElqVL0v3KmUmEElx@zo7X2@qt+GOa?$>Hs#QVfe2ac^$
z`8R)!wH|&4f3x@TS6pNNq%g?#E=UqFXerPK*(w_8EMQ?=vNl`L?c{Fb?<i?<j)_d{
z+QOr)W!hlL33K^jgH0_VtO<EFSg(fTR-DCNyrviFg>rpZOlHw7MMl*WX1y_389<J@
zrVdt434XaAtAq@$(tViZLB*3#DSmmm7QcUtBf;-_6@5tRe;Qy`PzB`7q5C#4&QC_J
z3x<T&e!}H6pcf3Hi+;A?vKV)n5~!LW(}^lc_ODeUZU?V6qfm?4+9~^S?-!IUF3=2g
zIGaV)p2zzv*ldXw?FO(HVaULi6wPsSbOom1m_kKK$=S#XmLsa6dh`UV4g8ZrOM-s@
z^X$FkfQ}RP*97+MP$~4n>xuAy^QqFHfQg1=Mu?N>@X@g?o{SO7)?+viIyc2Uz;!JM
z-ST0yunpnmths9t&cZFty`;Fu?8tMKsA`H!c9+e*;=i-nyUGSRLoi77x=l2Ks(7vo
zy=qo0*}CH9xX&~dXD9aJn}p+J&K-ZYZH~*=7Smzp(bY66FR3HRfkZ-AvLz?z@f)LZ
zI~NP+&iL4TVrM7OjvuqdIpRBcHq@wxxi_CaE*4_B%s^MBiByQ0j6V$(Ol|pTdx#vy
zQ0d!ixGznI%5+)7L7OI3iCr~m)BC_4R;xhe4DShQK^~s^+F7$@m{1a%Q>uUK+z=18
z)^*6WCXjxmO(<&3?tN`kS_ZY#NvFzwhZ7Cj)SgCHC)0B6Yg^A0!}bX~zv1o#?S$($
zv3ax7iJwuytGC+PapYyQyI5QuN*%3gaqTN^Nwwz<lQlo)mJ|{65wl`*!wD}55TqKm
zJyL{4fbd>OTT1=zulH#RH6(w$)voYAEm7!Tl2QuiV_JRJzF*^_0hmk4<a?T7?feYK
zoU19ntTRzO1<M(KzhB)Wvy~~JTIInVlF|xOYZWBLFIX>V!-WPf!E(f-oOq7XROR0M
z`hklabhem%MirdEh2`QzTMmg%U7`vwR-knym+z}E0dX8BTZt7dY_Na$5j@#XV8Kr+
zn+`$gH19bsdxRS-JMNwPx1q%G$=^9zG721p_;Nbeag(PjwS#hApn+P}eF3*p|5C%o
z2+8WLIOim@WNNHVspwi$8_hoBD_~RrvSL^n0b;L~S-@8NXW)@*6+hvynv!xh>GIhn
zoh{BT!oslQw<&b$(H4J=MrEGsRd6<O0GFH2Tu!;YQ%0We^6@CRADNdO(Tb}}bmEBh
z9wnQN!g{F-h0S77N$f|cYbr^_DLeLy%JnaeVnZJTVz5<R2Fywgks*pajs=Mt_5KfZ
zKh4H{Z*2rr_x*t>pfB37F_TW)uo==(hE9>JR~RML%6R2?mfnA+7=;gssx6X;hs)cr
z$`ij-FV%+y<XpfT*s5L!&5s?vN+(uXc6x1JxD#7*+1-h)2>v9USj911gTG;Gs-n1~
z6I)R_(}}G{@9)G`1kZG0Yxb|(iLKYV+KH_O?`U@^1fl8YX~ZhvQV!KEt-YjpgMI*n
z4KkUU@{;vu^DBQx>&IOlprjSrtD5S93#jlP6G)Rayg^ggY2FdRWZh?)!kkjAo?tjS
z$3LdF<<nv$-p~NydW$i*5L>n}-iP`h09y=NmeY09Dz>PQBZNJVEz0YokaS#}=tA4M
z1WTbNmt^Isx}M!erl!%&8@ppI#U}67t^2ZXAFitXunB)g&9yt{JMdN*g)h_DjsoB8
z>>>aV&oAEo?BWWfxxg?)KQYquCP?ou077X==x$g5)D->I<!(jkOu1W)-e2xk1kaSa
zHT&Biwag!{boRD6<w{OIexm(g(JQGd|MOfmYe%^ZMhkL&hJXGyug9DgM<vUo<hX@*
z&2Xd_=J+-?C%ynhR=eWFH=G1Wc%=hhjVU5w#IYlRgM)-=Go9q+!`@(vu9V@@v)=0b
z_3`=eKT>-hnwP;yHx-xR-#ruoH<$5O9w(O|)*S?YB)1K|&#&m?E`pIIiqr!bG%($s
zS-`j5%+2;8c48wz_7Gp<{QFR^DoOS3X$Q!mx2LO0qWI9qk5sKZ{d$t8PY;*3hi_gr
zo~&@BES%ncIVrv&`CldZqM?&jN*Jw9Z(mQ}KS>FOKNtR>lv?x0^52fX>HMgb<0q8K
zA9a3zw5XoOXW?3&(UeAkFQfCf$`TwuX`Pi}zDu4D&p*C>_aO7qrSK;F8NT{oFK-Y3
zdKeS|Ey(zjQCa{i^z+w;a03+yWBIZXmQKI@?expT=YG;6s{4*sbBe7djUj5R21XD$
zo>dENZLyl>^<((3CGa(U6YH$Qx69XUx)$$$@!xn&V}gazrF$gL!Jzb#F7sU}FVQT+
z0+6+3g$#i}<;nuWL<DIf^!bwo#P1+L1;NS40(rXOpWl8xbyvQ88YOzks$wrueH%WF
z%f=B-IH&s&K_mG(uMH}JraVY~v|wC(upVX01_fb^mLrNdBNTM9R@FdZz;y%k#^T_A
znk<dQt(Rr3wG<@PA+_mdwOe6(^lnGoL-DeEloy0hW|oW_MnW*bGmpKD`vG0J({71u
zfVqvlh;<nv$hEtc&WKN0nhp}r0Ohx-H%<6`egsQTdGB7*+Q8bU{On}+pN$T7mS?b)
zbHn|J2;pO!Iit>JWu77q1TDXLHE{BOtRon#^#St2AT|1Y`1$vbc?|D7;jMNJa)IG7
zq<D;qn=_92`^Qx=7WkrV#cW8iP8ick`9~fy?;LO>I8pS@X+z-*DmrG>ESgi9R=rtP
zg==Mn<rwOOhxlfXr;#xeOh<AogPtt=AP_IEi|q*7_Os{xv}gbq9OjQWOuOKJFg+g@
z3u|M1=&ZcYWGF$9hl!{6!~N2dkTVS@+sT@+K+>w&&^R`XIbt@njFI+8m?J+bwY$Y^
z&I=xWli6t1Y@`D@nmB}+T^hwuy@sZS1TqR0=9n=*u1O|DO>3D$>}@{Bd4#v7@A({u
zINx07b3%qR82_Q19h)m(GRO0Oa*gA0h7yi}sd(z!jNGrIy+L)HwMdA6je^USIxUkC
zI^ahq<vLm~;7DYJwvn>PO;e)pCggcW%E|80kSDie%BG5fVo6k{PQP&mQ_^&pGt9$;
zKe=|>-kZh-cA|m<#$Gf-F>E^{U5+y+gTkThX977-n=Ef4GX2vNTODS9Zry}REwHvk
z(|hr^u(so-0%Tm(6Xm;_ZsaeVK_(uO#X_V{<3CYsbEURWi5|(~Nl$Ga%yp{~;__9s
zumkY274FDsm-*)%!@ZmFJz>X8QAy}$T#CCK_PzYORm8W&Be|HHdL1YQgwpeIX_Wvu
z@eZ@DQlDJPFI9fs9<*qG*`3zh2dHzPN2{L;V!zk$q38-N9DGe}2^fbk+(<RFC~C{r
z7P{X?43_>wXTGj=pN6Y!LFRBWM7%TnZ9H$3<n6gTUp_6ex`F0ko2GL8K*h^uWKjLi
z%N;x>dMw@DKaNTe4e_)$jzY{(E_7l+l<!5w)G8$f&t#W#ZmZOP^Q#H1MH{<FXGNiE
zr)%bzZp5ae9Fu`lAXg;+lb)ew9&aSx10h=jr_I^ecvlZG6q#lu@U%m4C@QE&33D8t
zIdnDKy<5RdQz~qvl3Ib%tj(2`WDD$0-^++V;l{X3Bzsq?h0q8DJF(l#z0zHBq->oR
z9XQ`W)W(W{HdVQQTlXAdbhx`Yt@H&usrPo8)Iw?7OL>pmAhbBJpK;A?uwco#1Q`J7
zY1>sy(=1HdL~*AUgv8r>^h_m;QoWS5CU2@Ox&d=7`hs3a!Wy^Lv%PD-R?8MKtIlgm
z25sl{{5U46v%%bbH7niMJj>=LEl=W}#Ae-mXH*Ky2~8`1)-7B7gNiMqCN+)Efsj5P
zXMO*pJpGk#ei!J4cc*{HN57slO9{jNdHOTwshEK_WJ`9sNv(kou+jTv?Bsn^yI|~z
znOP}LF)aG1A*d|nye@@_wzj&J^0|nycDBhWL;KOTWB&AUy)le2*y|C0S=B`=FEnGt
z0bN2vScN@*IA^g1RWu4_(uY)GJn^oR^3!_179C@*?RQh)(mDQTTkXrxp-DK4v2hQ+
z;ORy7XgsfgVjMwdo6#RWuPUI31(NRsmey%eq5%h2sB~$k$*i8ON(3`JSyAS4moDWl
zKIbdGwC{&IAq@<N-ZLwhPg><}D5rdeM=n?$1YGBTkJF`uoLpTgC^WMZP0(BEuFRzK
z@It3sD@GaHFQhPzrD!s|<A`pJrfT%Y*5*o;*}YDuTXxURX0#E=$l3_`eLZ9iR}#i{
zT6H0d*o3McokM1S82!J5C&Qy<`*s$29|PMLj}ka-nqKl&Jdzfu@YZ;&MsL8gA_#`;
z1nxqAx0yrdwSy!GhCrN1TIHgJMMk2#TDcSQ?9k7?^Dqe@G#$soB%6tru9~8zLMT@q
zo2`>bPz*;F_Rq`~PARZ%8C}N)N%Ts3QQ%t9z9pM8f#6HsD&ARZFZ#gpCSxo1(qKC_
zoE2au>%5_)T&5vhNGLrgp*r7Z%gU#!A!Vw6*E4J(?wP!RZqInLSDL+&<~4&2((4u7
ziyp&aTjlNe_w%rx;aFek3)aW^spzO)6KRX56oblprjfSS8J;G5#kG8;DpE2)N4R&j
zmT@jG$GtX@&cqztup?nRY84aHaH!CY{G*~)GEqR&OB-HxSQw?kYgsfIMH{0?NM_@I
zy=F3dog>6n5t5^3h!)(yK&ILj%+WOkeyb*<mhq>u&C+i6g8W3w(C`jEtd1~fXnim4
zTVOlu$y}3nD;Cne=Ay&RuY{VF;Py(a;?mhoe7E+d(88*CKPpg@nb!S;p=&wy?oLY}
zkRWTOdum;#)?B))S(&Y__aJE)j&W&!;;rL{E4zppg&iy>C{xAwBU+r-c7kz%+*Pkp
zbetxx+D=Q{5npi^<6w+^84$T-jMs){m#ZVLmz<e%7nz!p6V!CQ%LaNR=y>JK1LG_^
zd6gor3ca0ep<#TTTjztj;YxiV)U{f-xR&B*MT7Or<ilZ>u3k+FR<}dz;aakP*>JU9
zWxRqj2Uq2l&Em^}x3*0=p23XeOp#FGy;;ZmJbnX>AGPne$147yjU2hNSLKUU95g71
zOh0<TZ;b_qUw*XRuutrQt$Ljn-lF5&X9${AZ8Ua}!kcZAuvu)!jEb1y7Y=`wV9WHW
z>d8*Rlv5f>KY!+W9w=kgXD$YRNl-zV20dp;Sdb!h1pKlYaH006>fH}l`f^ivXhMDU
zZRPcMhn0?9+In?k$Ao5wu(=d;Lw;Z|S?}$FV?@Sw)X)!|?^DNqsEa4t%B!|dJ>NoQ
zt8&GoS}E%-R3I78gPI{fty_S2L7X}dTEK*g*w0&FL68m20CQB3<$V-?wq8BzKO1jd
zqc9%(Al+#BvK%mWp(oP!K$?fU%%fLps(pbh4YP-4O+y?naewwl_%Bb*!UkYWi&_Fk
zgc=9pvk^C{l4wFK(wGhM6fq&r7fA7tUNGZJp%eLpE3dM`i&QF3%Fyy;UF<CYx8l+t
zbz{g4+-54Pl<y1~js&wsVPpnVQ;er;ev^LL5p<i^mWSsXUay*{sShb+cix(U&oEV2
zZIhud><bRxi##h!40zCqzT<xlYmV5v@0YiS{{lvuWtUN(GZO<bFfo_m+9nj2p9C5x
zm*g2Yn}3gv|9JiJPG;82d*v5NII9oFYQbD(exGGOs`l0-w6jC+D2--%n(1qW+@hH>
zL1<C<tkbh*If7i8IVR?qY^3q|Y~bL9H(+$DNnlNqqK?-6KOfSNW#xk+AxBx5ZlG5=
zW8D!QA#f5n%$($JBhNlO>G;Il^Y{EGwM{eWAb-u%G&Vk;%ZMPGK4%$aD=WWD?Ugv%
zk{-zk)TWpIr%mb6c%iKt#V3=D6B#b-Re?1Xvqi=->#PE09*bh&F^S0nY_Y778*ek1
z)><e(ax8;7(ayOunGctxs^3rtz}O<Qn(3&*Ou^((OrAeSNfQWX05>3{gC0T|1TN3`
zynm?;6|BKc?Mzi0>>!qgSQ4JFAM!IVuZK-S4u#ULANAjDLS70Z^{s@w71sNH)bHC1
zQnEzOkLskSyq3;s?VxInoqSzXIUA8M_d69YlFiL+R6S!{g|-+dy{odq$SrzmKtYCi
zLnlbO>Zt>`o-@!|`tG<TX`KzEaNMPOMt@#NKH{#3S)5OF4cMbGp?Y#kc|?QmtSX;Z
zfy%BnJE!E7w%sadcB>tq_usM+tz8^}QUp)bj`*2%vk4YJEfA?6_4{8RJ}F}=I#=jv
zFN}9Hg6iQDSbP7s@aFDe6X6g%$|&YVgo9H8Mo|mGEpsHiMMwP*Zse83TGiU85r42P
zo=3smdm?{kp#bi2Qz#De!n#|9V#F}pc3G$uLcmFZ)3*{r<VD^O+O{-mDQE;qR)yZ?
zkSt>aZ{^TvmdmZ!cbP>C?DWhci?kh%9lJ|CIs`enJDb+4`jS#DrX8}a@b<&-^zwT5
zpS#i1pk=?$U;pdZyYPkP+{akY3xCfD#ZmaTe;>Zx{WMNnvOpnuO!`Oi@Y^qkq2+5%
za=m2Hq%lOjvonI&tCDxZP)&06mnOr=sS>C@{j3z?r-^WB`Pw9#WY~rH8I>k5q-b7K
z$wKAxnODgY-@I5Aj1!{F-afloggkl3(N9zU79BGs=c`Lj%P!eSH6=*=1b=zXFc0VM
z$+(yGeNoxg2#a`wvRBnm6x+|36B>5*fvNH@^4hhuCJGBm+i@n4%hdxT7RijgXS%w9
zel}=7p;F7_IqN}Qkt6wR&S=__;M0@~Lznxd<Uro8su0yY&V*NbkyoIY7!RE#v!Y)1
zLlhHn>?(FdR6iq~-j1k#rhiL_>U-~isD8prh#JU$XG9IW(k7z%&U+&YK{#*sMAR^&
zX{z>;`08#y;U;T)&AHf&?*VN$8AWnb=2CW;QMmvje&MolIC#%w-P10dIeP2&w97D~
z&T-K<#d4qkee8^Z+uRQIu$#t-`9O8WD*&flirF4|<IB4|1E3Y`Z-2&FS6l;VB{bL>
z4`$lNh}zF?0#@{ggYMqoCQzZU1T((vCL%E)N6mB?($|#XM5h(7BX>orcAme49g#ZB
zNO=RP>=;BRt)~(?T4yKJhN4i;I9Nk1uYRv)8o-x+gDOWnZJkhZ#~f;y@Iu5J$*;f~
zuTXXm@nguP-x3Th34d+oVr-)voD%>uZ6=hfg7O*Wxh3%jR7#hE_pYT~g~+wLoe`P}
z29*PwZHtO_=O^{kc{jHNy|5LX_c5CuY3sdrDP{9)*nVG2skC8tr}d5i>nFSfuz~#l
z55NX7%yj_kN38<v)B$Yf%J^NmDPhl!L=YaPe^H9ArL{$IvwxZ*Uo~gUha9WN7lq%L
z+q!Aju~y(O?=7U09@h;W4R@QZ6I<c5EbbP=)SqF-z9CL&WO|i-n40w6pjBSnE<hZ$
z0Au&rSQzyYnSCGjM|~XY#zj?0(uI6(U1y8e$q!`y-c;*^y0;?zqh>z6-=4s}a5@_J
zgWJ>mLG4K^_J7<V*_2A)o>kE>p?~xCM?cYp-H*QW4w1T>?^2-diN7OGLwBfktohGr
zD)|E*-jTqou7bI(W^JLpy$kU^*OQ^it3_yph5dufgM2A+G&$~U_j!*+y+DyZmZFqa
zFHr4Mt?9@f1z9?v4c8wQDx#en+n3}1UTttLS{*2y)qk=52XAGK?gYI5OFnzzoN{r9
z<sqsc@~4L$l*<Q0ytL49GdmV%_}zF38L~^?e63D#^++5wXGP@N`zYluO4OREhwC#R
zlP^<YD5MEKCJx6d2sLn0Q;Ml67F6F*_G_*=fM2;2;n^na;(MW<&ZZm;hC2hKb2SjJ
zAGpa2-+y{a2##5BI6EZ_dky-+A>r^I4+K}@U-nwywNP|98r-bm)o75a;}Q+tu!qk2
zaB5^+eOagnSerS!;37ZH+O=%t8+q@2K|7QWGlR>B^~w3Fd;@r0TV=-?``8wt|MaGv
zE0kOwYW!fel%2^@FN-DSReIqZOn{wfn#%s<JAZf$Q3hukffDsk*YU@BBJd&`W{D~K
zX=#Di`L=Hzv3nI>)6#uROc(hUSK`rxD}YKKAzm}`)AupK%G*k)=rK#Vxn>Kkt1c%k
zhYT=YsG#Luo_>+%h2J4OX9oF|b-8*o14A&QkxIInVQ#vjHEOBc0?1c9bgTdQS}zG}
z)qh;QTDVM?OsB~*x9vF&uOtgKckNQJY{z7Tbh%W)lQzQp_Ry^Yr`}&7n4(iKep`Ds
zOLjFoup}Iel;+-i75d%^;59o;P_{*F$0nugtThSiFMw8SP1i*us#fms=H91^XhsPF
zBrdVoyXIA(f)YxNN~-r7qo-Q;3E%}A`+qH}Ei=SZH(2m0n}*fwRgD`*k83!^lFcZ%
zZhFr-&5XZTr<IjrS8kQ+vU1!E!ZY}mUMA0r<E5h(K`zsp?xgnDHeKNmnp>*pYRysi
z$|HvWzwFU%gXNOkmYoZlxG7%Rw5{ca`>nE7kA9jWMO*ZggNti4Th>v%=Q}+;HGdRb
zYpbkV8!N#t%jH7;Edv+C0=-`OEklU(jy(_OJcn>KuT<o-)~(VeUlZ>sU6t2{_yo53
z9Y^?;NBDfsR)+ALYE6~u87$o>J;%>^ir5|U4@JsPLh|>qJ*3-aNKUqwy<%_ETz#K=
z)ETb7I~HY2<0CO?XP8WZF5%MNh<{5~aJXLl9dOxE+7FlM^sH@uKR(lxw?UKxfU(n$
ztMqYliMC01PV=AXEafhnP%vZA=mScZN<6h7pHdAqPc>?wu#ZhjX&b)d+3pm!>Vfln
zArdZjAea8V3bD;>DIuY*=oT&Zs=bd(cGSuV!^vCEM74r#mN8{ww%Z380e|5XCrGy2
zcW}u}oFoiB3(^moe6!nsuC0TgO0C6DO;4A=lY3s_(K6i(i2SslxxslsdeqNC9s;Er
zd6{j;t=CapziT&4K8d2*>o#RG{nX)mH&inZq((X6+%HIkWOux1m(gdY4Z;PtiI>4}
zjVcMPAUWdr%1qyAfuJ|)iwbck@8$LGf1`EfmqFJx69h3ZF*cXsG&vNPkf<3am&hbH
zlYeKLr(a)weBdN{ky9$Bjh~DKhSFp?gIs_!Ay9H2-)JA6@IqP|f7KE%9cg~JI<bg>
z@l1F((gp}K@&1f=Lljm}^nwXvXA}(a7fF^t5iay>io~rBo)^w1VZ#5<rvR=-l~9(b
zS|-<%&@xY-6O1!R5V%$$B2bB0X6=cWsDBW~(KI`l+o-@Zm|4Ub>x#r#ilo-Ws~RQq
zypS$wPUe0gV`ej4>ij^DUr&#~U7l}?6r5;8>7&_ZPGltGPVSp2II?|euyfh9BXU>L
zY_Lnf2qXmelr1Vr5B;1qu6hv7tVeu`RD-@{Ic>}78r_3r#d0c3dPH`emh>WO0Dnw6
zUqS@=PLK-D8L32@JYz<`;Mv<K%cWZ>=a`2#(c98f>uqVir!iMTKsIQLw2;Ne<?y03
za~38eIa5}Xj%TiyM}ubI>@XSwU_qg^l&RLLD7dHU-BDJzt*OqIbeo|}qK#0FQL+#w
z#WiT3f^c5<XsK4e)$o?l;lPYJjDJoyI-nCrGpQY}J<b3xr?i^>!TIyIUp}t~Y^|8m
zqdZ(KM2qNC<m)o~0Ywmf?&Lp;0t(8!HV0AYM}H3#1`f*5nk3K4h=lYUR!F>>_%Bx@
zeo8#zbs$;AB-o=Mb)&xvNIi!&kQC5Tgybe|(uyu+M!*tCU6y<uVA?XR+<#HPy3yYS
zu%5#zV7ryYt$=;FYPTpDr6H{pGZ^BsV6F+XE(OVWNlLf0*dc^naJxmxeyx|~5NJ46
zEO5yY+b-s&%(N)Z>HaYbQnDC{65pCKKjU1p$Ew2H8A5B!;$jd#7bwZ^H}#iT;|e$G
zpQ;#}i&t?sf0S5S1IGBC-G6bbZ@Fl6#p21^wd#>P3DPhnM`)9ElV9kQluu3lo8D55
zlIoU~YKS~4P!LbGwxODQsa^x`mRqGtB~|LWtHDk2w)_+upk_VSN`^poWh7<XD4ip3
zO%$IS?xMWSG8OK253wli46!s%x3X*^_6aL=1#Qip=Cn0;(&!+~gMWHfq&EAkGODqP
zQ0Ee$Ry!MTrt54djX@g(ON5FRUa6&Hg)W<!*S1rs{(GozIpE2;)@YYRh4Eg&L~)NW
z!3fh{aOQAm`6~>qpv=q&*@&ezC`q(wl)#pSyq5vNI1J&gZN&7MK<8b<cO;LYbr|@X
zPwh#V*hkA^j1nyfCVyI!;A$5;^mm;<*?oUcf*Lt_=8Z%PwHZJH)1s5}-PiYNhP8_`
z6b?20j9(C&_nT}KFr{=q>%VP2z_HtMGe+Y5qqO$j58M2(z!Gkd^4~xI+U95llg8bu
zn$=8zx=A$`R+C*&jb^#YFmIL9{VZTs1Y2Q^kTw(F$@iPpgn#ph!svueH9;9T_w6WY
zScT#Cd3gHt^6>X8AS%2|4%2Vn9>NcRso7XR3A}4#r+<E*zCQdklFg+%<2Wb%(L6<)
z?#&n4TJ;=N<JgJnet{XLaV3Lv2u9gjg^zWQCU>cF=ot!!_+Mq5-+XC00*y6?@-r@%
z8VrWYu!gZ1B!A~}WB3wsB&c}P37u=X6rOWqoeP3kB=}}~2*ZHkrPwmzNfOOKM_$Od
z`*`ny%~5!-ZUB1~h74>^F>e|$2twsQQP%EFYa#>8VedIml6<+vv!MYq<LDVyH}EHe
z_5_7kX79t!5zN@%ytjh#-VT)nj74guynsbnWe77f3xDy95Y__qPh{J@X%Qmt!RnG>
z9++bu;Cd|xQGC$|c0+jCb9WV-Mc6yK`#|w6ZWoSKxEv|2`5n3a$p3Ep4l(|Tr~WwO
z6)su?H4^Wo3JY7Su%en1@r-Sl!D;iAVVWYNOPzRkZ?@bZ%c2ESzHZIu2JLc7z#oXH
zPj9lm<bR3@EFcD|E7q6X;;6X{cd>r5`odw>E36+n2yCopYAqitD503uGqqyTOd{XG
z3$SX9oBnU}0tJ>?XP4X{b{9MZutwe+FpEoiCZAZ|;#AgRX*NSR^1t>?Zw^^lh6CIq
z{7W>ht2LBJ;kz@}`88>Qy4$79K&U2UsY;5!=zsjnkTjMnRrGY_@HIvQ>?wlpVB}3o
zFa5H$0Zf&%^oluYTc0i4I=a+#vz=SVw6{hrePbYvVIXBKDNY)m7EkLR(HRdMMwgx1
z6-_c<aJW9qeJa^xa+x=}YFf!VV_8tO_#wj9c}vp?uQW?Ov{!ym==oYU7eezMF|Egi
zHGkyZ+B!)gVgw&rCn=f7#k}1)jrRi$&4Erz3d=x1x4Wh532SeG*LPG@t`FQ6PGJuw
z)`5`R6#J;-tRAD>lB1o%9$t(Brjz33Jnap$=$tpIbin7A9cgd#DwI~%9228#PBmu+
z<)e(=xWfkpb*ppb>ms89Z2Tnc;<0`?Q-8V3fr!O%LfxeLF$?T%Z-QZ{$iTSyU}SIK
zQW$hMQ{ymwh@kVC>xIPbPwVmrBzfDIHl+`q(g_qg{?x!!j#*l7fSDI`$8YI_jLmH0
z0O46=_~yl)?R6>}=6i`ElsIY>X8zfY9kz<<XF<&;Jo&Sx?N4U}CI+e34zTZXG*3;3
z-#w@1u!-_`_Lh8naTz{CGV6!+uc)I&kMoYHT8$nkn`WP4IwqU?`wrDg?_|}9=+rYz
z*{fl9mRs1D<_>z<j}|6Fv4F+M|6W|CftSHZHx-vL+yfH=F*%p9nJFiiaCtohmvAjN
zkAD|>?&+_W-yd`lU8`W@ZTxC~eSCTN_n}`1k}@`Q);I+ky1&0Y#1{-kSr<A)6@a_%
z|8!p-{%EOIUd?xc&4a+Etu<x>8(L+I@2n+7&Q;h*7Rs_fmW3FB%%EJdXp(OZ<vr(T
ze9HG6@6UOjk$GBxFySd8dU<HZNwG-Oe}8=c+BHu;f9fGZ2F2)Fq2WWk8q@_zt#qV;
zCR3d}5vuRbK`%jr98@5C9W)cJYha@@iY>Mj&Btv(QJxvmHg*Ly0XJf-PdB4S?3C#0
z488LVy?%z(q9oAHz#^lmNSGpb`Mst@(Ta>O>#<2e;_JBVGz7|$mQ;Jy{HxWV6MyqJ
z{bohF&rytMpe}M+-abna0{w*j7|!5FLDP5_DcHxX|AY98vKIxINr{BYcx<JN6qB_)
zrwx5;b#g;G-r-1|4HH_!hO}1*w&kkNvZ9dCe_v5bSYR?)e{m(mTHGzQH1|sCts?7Y
z_?*(rvV+Y1UfQxhMF|rJ*S}4^s(<Jb8ZWquDahr?P&Sd-)6bt)SIite?@YmNw%f*2
zECXRQ3lKAS3xpMjM7v+^pz4vs%rIj7@clmNf=>}EiVLtf(FB)vddZ?7i6MFf{VtUq
zfMy5QPzB4%*_O#pH9{SfpV8EX=BXFlR--NG@d(ot_ai*X)T!`3<ync28-GF_RI`^z
zdcipfDJARZ{b%vxiJBg9@k<VHJ7b8}6C}Je83TAwgc}({4v6W)l=v}u^0{OY%b~Jn
z5k_(ri`c8{hhm@SVH**e09`qxyMb|iG74N^#5+)!aXAg})evLcRRI|UF~F7}>Cp~9
z=cFeQcY{6myKl1lu5mA#MSphRiGr;zJkpGGu*U0cyf?unqnzd&z+Q!+0z0Kxebo)Q
zMKov6uB9ZQY$>k*C;5c2?dTa+H}EeTIwc6O%syyIXL!c`PM3aD0vnw%Tq$QwM4Qq2
z4!(vJ%=Q_$G09s9dd1?=T8ba;r)W_!^tp^?85%1tPvUk9P1PKEHGh00zLO_o&8b#}
z5|9QT4JM|ALy$}xUVdjkQ0;v!+fCX1E{_rw<C0qDBi4mO!fZQkfQ2M8iWUVVhXy-0
z(JkTDC5ss^@UU5Xq|>N8`opC&3)1a(+`=MelE?Y&hG4kcZBGepc<t%n{Z4yIZ^LId
z%nvG8(8qil&9^suSAQ3jgBHccyY_=Uso<d9JDtyE<=l6l9HfmxrdEz+4@#0NbMUKj
z?7)M>I^C6UkT`ISydVbaUH+V{1W!{b21{~SK_MFN-TjWIQkVD_Ld_gQJhvR<e3Bk!
z?-<GM7nHryM6V^NIZjX#N9Xt_Kvcc~Os`naBgHx&A2W2!+<)AWL1RjBkN*}E1C1@k
zH4k3H$8>aZ*}k}M8t%DPDNaq|Wye*w?7mYS?S5B5ffNcdRepP^fIn!C*W~4~tV!~t
zSY8$=GQXua7b;BsIbM#VJVM5f&@ufna}nFp1R*gZ)^yB)c4TyYh|OWL)^oW`3AKKb
zhti0o(U&VWl7Gj`vKztQpKYzxkgzlynVw3a#19&FYoP?KQ$~DDQz$WVhLG5EbZdpQ
zRSq{>E7ze|&J!*s_GuH!DX?eCvP5S(46>L}z3-BB?LW^#U}Yu$Znz%ZkkO=oT+3yw
z!IB=fjkncij2!e>hnAW?FXQPYL)KwzleG|-d~gj%IDfUIqV~68nKRsm;{uk$46nSW
zb_T7;c(0;00FFR$zrYmjMserfj1Jy4cb)VU#)YvF-TV7INM2&jpaXol$}E|v6}QsN
z+GmOzC1|IDhL&UPWFTPfWL)ePAu58I)Hmwhs)WwS0mRIx*ZMx^;s!!r%%s8dR-U7^
zl)`!b)ifhkyGVZ~M#tH1-r=p*R>|5c$8!$m39)i|UsqZB74tRJC-?UbkLp4h@qs4r
zUQNdB9_H1S81rC5lf5&Ctrv)umRzXevU=d>f;yMxjTvP=rFnB@`*~ye$g$)EtzcJn
zf>6|w$i7x52tb^&pLK$KEUo@8jUe#iT;R1EK`ocvji7&q;7`&B0!u0cbKu#qp}3<F
z)KJ<eDK?|`*AyFq8%4#I{e5+Y+nYhTz0=CNQa8*koMy0&rIC-vX_PoO?T)NAG!zf0
zHZ&ABstqmworkb~oA>;^EcXB#$?iSPltx%X?H+Y>pd>ua@Lg%ZmA|yLd{5&#_Kw|_
zsVewz<!*o1%6GflJ5NG)ph(V1B%Y&ab2SBMS^nsVXpZKj5}&X=?XF(}fdcTy*NAdE
zaz}L|CTJ;r#{L+RYPZQ6%HJMqcdkd~5cv-{F=P41>qPzq;d@0`Eja*c$t=U`v|N)4
z!Q)Jm-kn(v8FDamq9RRXaF!=R{rcORoa-_#PjN9a1*TJedGljxs-6|Yo|3BMGlcwu
zE2?<SUr$%gE0Wq`yN+uMtWD8V&K&;%>^PIPm%&Ik6#+PxVR$bTmr6A^Hh)M_FJLji
z%w#t}Zw>ZV^wOfqDTn6NpI_=_TarCbXWIhnof%sqsgDnlqUG>)ki(~k!|TK2i=x2_
zr@`Xz`eguqAo)K?zG?Je0fbR7ynY*geUcIlUkAQON~QR+{N3@V-hEKucm$Yy!R~`4
zc$&uIv5ZkV%|c&h=TFES9Dk3r&dNaVkm<SIZ?8W;$Sk@P-h{8=qy6>#`taAoxDZrA
z8$TE%fk?w2-yXsP0xgX7gVw@&7{32Ge0lh#rCNA3pAj|(44X7Y>k2kBLTfh|qlB`y
zz@~Vi#0w-|h-Sz%@}(0r<nwjpdyvKjV6@A>^PKn`cLC9GTR$iC`G2XDhltH61o4us
z9*jmo;EE8Dh)az5Pim786~xoVf8W0j&7IGmCd!6{BimAjkKt<c!<b7U0}a%PYITRH
zKE8Ol2%=^1v}w64gIIB0Rf~7RIn^QZe7$N^i(=G6R7+4rRL9G*ItT=r!aMTAtjnTe
z%D?Sx+Hnwlwq~@oa(}#J%ch3O9lAmGMBk9BjJrXzc$ktx%P}xX4lYJwn7d~&4wI6|
zg(px^&Co76`)WYUaYTD9Bo8SIw}o2>ZZ>f7=@&e8a2N{ukUKPGTrjei<m(YWTc{9$
zI#4mY5;$o!E`)n=RgW(Q2hVFlLlw*%>71uZ{&)Mo-={eos#ccfSP*ENIX}HW$%btZ
zm93}zo~JA$i%ts-*doh%FNjte-C;!5byhuUO7TjA)IoYv=pAHhj>`)G(t@0=W=}Y+
z*oFZC<Qk4iGp6WH_m&bgsIuGZm(n*kB!3vJEc*v)=P-mBzMRr3MT+k3=h3-4WRwC9
zDz-)DzM$LaaakuMjo_ftaxP(Y&1Tq3Za0}}@)?r{-iH{*n%&G^>@}K8jF$07sf3dq
zqs#W5#}^jyR48QSx-rK=n`4toF)genFe)}2^;Za=iLpQe&|OY>jGD(3ccb0d*?&kk
zM>bj3f`^<8Y~WZns$_e&KONYRrzmWfliYoW_*98bSyVG^A7Q}-y0DxnBo=L@4P)8S
zM=k4JGN_T~6ncsk6tLp={m3O#AeT0yDBH0FoY7VaKbnW<D#*E<E6MpbSEP;{_L-`!
zLcI^P04}md@<u49N23W84p}NWGk+epwE3T}AJB8^Xk1%A0N+ObJY7Jf#47&NIf<+8
ze9lG1O3IZ}9<n6l<Z$#cCl19Wc%=wUqE|S?ZxiEkXP!N3#!vZ<%W=Z5glTOr%etLm
zyKf1q*}fK3t#C4QJWhr+=5K!H9AhpQDVUP;1)6~lmzYVp=kY!ZHiu|ny?<X0_9_eo
z*qWj~++2P&lu$13X}OGQMt4|UQ(z>8)&v2T+2=x`W$&kirvMpJmgJk02~aVC!&u*5
zBPtTdaMdjyXi-+w9HBc6Nfrg3x3ee}awpu<qH2m;6!^J>Y$&f;SW$63T0PelTd_m%
z0=?SM4r_wv?66^fXWc*A<bUm@G#ccwzLjhq1X*|qmqlB2N_Yz`wOf40If2+IY|VGV
z#^^e>$CR$5XnlvSIZt3L&4tQV$lI~$Ku~7N7^A<@jmo}4ZcjqHUTz&^=SwZhPtpoS
zok=7l<LSmBV|(mRMv9+F$I*c{^s}1aIX`RI|A+jnmh4_XtEt`NXMYX8`lO#VY&Z9_
zi<CY3S)=TK(a##W&1HVppm~v><=t0#reVi|gO|8vw!oAHX}e*z&{Fg_=bsIwGx=vT
z`u=#-5Ihr)TK3zwV&q{d_p1Ez`38Sf!s>pp_M}~_U_GGQS=oZG;;*z|P&)qZG=H+f
zy3Rs^Z^)u=t%m7elz*YRGuL<V1-3i#<p9MSwfn$w7zp2OQ}EtEU2}!6O4jeSuTah!
zl+PtH%y#gu`b<x5D{(@C72^#{aQ~VJIVx65awnE#{tru1APmB21c#Xsyeyr1+!ubO
z7go}i;kTxEPZ%^_fW(&KS{y84xE>wjUOsV)$1mmd4QHo;a({N#XXdS@q}>oJ?{I7M
zPFT__dC%HW8Vjl%-{v*(wjZ)Uh9dD>dUB#UF&^z1CD2Y?VKYNVj|WeS*ygt+kW`y;
z-+sh3^SE)JN_$P-suX2oq@=1yf~Ku0Oj+odtoPr7f?TtR{mXb>9@^VFv@Z2ApQF0T
z+O9oE4d!U2et)C)Fc6Rz?9OfeHy{d67rldl5`^>qBCqRW==G6-lVO9d<+|}J0HLV_
z%C=l?t*4MTcj;SdD=$%mtj^QiRhR0G0H<E)8D0ftN*`;63rM!_nA8*2DxxhcZu!Ls
zuwHog_Af>NaUlj%bxkK5lfb)qXs0!^HgwR_ojGDdSAQ+{u>q?nwiq^RgL84gp|#_;
zp>ZrG;|hhf9oH+JW(`B-xcv3Hqh30AQQis59Pje(F}?CIjwnsp_^WHmF5^mucMou#
z4A)25wN+}i6fQD44!M^aIqe%i%>KaWYh6WtF1fF5?i-a~8M5oI3~gqmGOSg%lHcQ#
zef1o}SZnjn(zy^5YGLC#NH1&*1CVfFepF+aYg5&~SE_h3(DWv4Vb7pU?==3Pu5iQT
z?xt`}?|;%0wwxnvSP{<2-O&x6lR0w~Hy?WcIozDunM1gW@!NKQD=AO5f9E5s-tQ&F
z2*-wh06I#H0++!^HxvReIhQfLK_r(CCqE3Aa&QVLf5lo|Z(F$$efO`pKwq{6!b{E<
z$zl;8yJ@xyWYHi=fwn;&97mgYn@E8zC)xh`p5c|RB}?AOk`gbn*650t9CBvPoFSQF
z%%zK6!e{Lgvr8p*WY(nAaV<@vPMMm-otZXCx?EF}w5udd^7nFkrJ`FBIdT!Ls0^KF
z3`5>3e?{~sQk#gUs<pj_zAd$GE1G2)L^P<{dVoUIx`9|bqBQg%9SKJt%^;a;gI-7@
zx1SZFSXT{Wla|_5f&OHi+Ru1sxxEwdQck(Oj+aUsG!fRM+R#<(I3_*v1`vscwpc3R
z(nRW5bL%&8$GHr;WUOT!+Llft4{P8emR>7ne<X>h{azt1W+0tVG#`w#6PDJ1&_Hw$
z6U1Qzye#I+2a!7|)U<(m8u6M$Es$~>7C&hk77ljPWUVi#q|Dg|W2hw3wB-n~wzgO-
zOsuT}oI#5WB18S5UzW~%PA-<t2!ASPcnsyCFYS!U{Vkz$Q+D@A8sSpX<{%JrY<<D-
ze^wW5smpON1cuD13=QCaJWN8g%c4Wu+GT~~!dhr*sjUYrpG}Jo1|#N@+k7p<flCvy
zm?aL&h;a%H4xrK%9z0aFE7DvzsdZ)1uja1gVVSVHrIcoMIC71(P0b=AV{mtaPVkE+
zB3?8%b<KS;4laz>Oq-<y(qKtgre<z$f5AU%O-<e0LS^t4Ytfd6;ruzpp}F@j!vZ#M
z)WKug+-;ajvs$Tb-pxXl+#%H;*Fiy8FSSs@AH+EvJ-1g8Ldk>r==<;c=f8iL^be-f
z*}Q*t@oMhh9-mJC+x0)p&R$Q>UPM4leA)lK|GEG1MH;?5ot(^HaE<})3;&@|f8Z>T
zL(FiJNI|H+>>qUK{#pOi?D?#J*nO8u5hKEqDix?@x&HY+{@Y6l(S$)|R6>bEb*H^a
z3HUVvlMsw*R2gt`040n=0f=da5{N^h06h?6_>8)ygTRcOb94s>A|jY=e;pE5KLdA^
zYKVXvk{F;5KmpGatuI7&_{gc&e`?0?nwZo6T3|T?pmfxNp@mhuQ42EQra}QNNFmHM
z9Y6`T6b#XHse~k$=MR)P$4jTP>0bhH`~ag%4G3;nHh4bbB7u=DHDPeE4GvjL0(@~n
zjzhkI<7`r5^SB7|E?qidlmM9MB<DZ~LxC4MAQXd7+RLMrY4i`q-rL-vf6zal{_OvM
z_jW%2@YlZoG<g+HK7Kw6C$slGg#PvHq=#Zhx7UoIna6_|R6#l*uEF70vIH3XcB-Ab
zZl6Ad_ovg-|D8@x!tCsg>vr$m@qBVV?<dnGEg!yc*5V}TY5%W(|HrY@7o*F_ywi(!
z?_S>2ETb8atmL4`?N?2^fBwhWyV=>Z5635yj{SwMe{wvZPtK+t4PXEEdH&P0Ih5(r
z@ZiyGI>#W7*pV7(%13#GjFX%Sv@f1{K!E(q{rmlsv)ReB$^1qC<l&?K`Q-Eb<<gQr
zIes&N4%7K$IzMlTr@e-SKA&BjolMRbG{~R*GI@P^{KM>Xvy376f8+}1e{y_=cRlSO
z_j$8s#q$M|YbqSQ7`U9PKTx?#DUVVar8bF|dxDn)q-v@fqNhZU84xuJR%9(=5F}L?
zjDQ*cG6Oz#3Yg?DjZ}ej6p9l;e0mBct%0fB-Ywsp&fi|VY60`x#rqHUUQa&uzZ}os
z{^P^^XBh14S2#`hf9+lW>fP*B|1kxMA@%2HC;jR4HO!2@&++r09_M?Hi8h7WZCABm
zA|Qjw(-R1lFE2Z>%p4Gl>rf<4(HMvfFwPX6a2_BzVl-z?3nety;1uO19at1ptbpYM
zHkcK8dN1pOtAYs7XkAc5eDXK6?yncq*5_#4M5_qp#ndJAe}L?50H@^^SKuVuG}Dv=
zSKg+n0PaADwrUFVfxU0hl0eV|1;-Xm!Ss=@4+Xb3A?!6E_DWY{*$I1-tz&PJ-LRM9
z74~|WI-sjpsh3h3B^o8(fW6zthH(bo$`~6a2-Vwig#*S$a^#cUNXF%yyef&l1G1PC
zXTzNGH=NV?e`G#Phmch;p(6DFU{01pMG;7<NR`g474t>TG~~>(4y0y;g3D0huz_Tu
zI4k>QNszWyWM5`UAHV3`Wl2)3ZQajV5~YxVcw=&)0`6{iHz_Xn9t&Sf4@tKnu&d1m
zL}f)Skjhp~5&LmNO4-m96$*;FZCV1=z(BqRe_jFfe<ciG!TnnUI&A=S+7;025}?xv
z-ZX+Y?FZhdpu!S%ONC2Gx;tLO_a2vn09`3TP{%rugWh+Hs<CZm7vIrIYd}}E3aBay
zORKejm}uOOFeVF<@)9QCGnKnbiYbCJa=iu$q+jBUq#GC>1pxOpO%cw)Hz~h%?5@zX
zl|c4=f9*QZB78AFEg0(BN50!x({3Toc|F8gcMEY|4sjlt8jr!wW3V%ilI$1u$iu$Z
zIL19lGe^BU_f+KJ)C$F;y?4$K@#^O`6d2Y=*${lopDK!tb%G;YqJgf~;`^yd4^XS3
z`U%-H95BaufkZ~}1k#lY7QOC>J-02GrG$2Re+C@%Q=J6d^HUYzBCv6NH@R?($;5YX
zi8w-$1MDVeX{De^@Y4og4|idIn(Q{FXz;?e6@gi4wg4=~BE<UDQD_xwFyR>Em+&e3
zS>^&xZp0`2qTY4d39O{7gO%8BV5Q3iri@@EBUs4@Rx*N>$SCPvu;uhJX}T)Qqy|&E
ze>2#km{2~L9AcYo*t|?A(EOB|39*vkY1ubk%7zVp6Y^t0EO;#l9k|*dj(-LvoN!kx
zFQYp1om7rU_#2MFvLQ^fk@;z&3J6a+416-mjhaI{wN5#tc7AHXm^bY8DPYTwy$$#Q
zeE?+#ID!IBsvj>`BoSW;uWP!3>NTz>e-W^pEXiVxP|y_iILJDb7!>QUP{GfgH0n({
z5Uh}qJ)&R*%t~7FhlY!bQ2qm|x76aQo}+-E{~E}-3YJ&!b`307P;dn-S73AnhyB8c
zEUK_aeyczz?}VHt>&R)e8*<vxv4~$bSRkT~D5)b#s*aM4QXZu;O7&8*5s57wf8Zbm
z@ap#z48%n-9D&6nv@p$fB>IiW!$Juya791X0ih|?FPbQb&AR`?`7;$K4jiTAbcj=j
zQHP8ako<-a5BCHRnxq?mAZ8Ok0pr5C01~NfL>@{DMNP?9N`O|WtRF<f2=KnF)&DZm
zL_P!_))>T5C^3>@>VzctjWF^He-rDcYXz<YT&@)k;iu7Jhv3z*v~=Wq7Jm1M6>I>A
zC}8$cHe?3zt4df5lmLY(!a>u1yO)(h_UlHMyO-4u>F<tx6tE8#Imo_n9kwfwZ2ZQW
zc<$v^E#wWmCEINa??^t9s{$U3vduN3w#I9=CCN@Z6ky5VDkaCdmu<l|e~6d%YXE(S
z;ybe~SG)HZ1>3d#8sn{Mpp|Be?N(T}X4<N$X+ROSO;g9>lD_Ts&I)nZiXPlL?`K`l
z`<b`N`ze=ec|R|2@7S2TZOq+y%-wmcqB8BD#X-kIMG0Zx)8_ZeB6u|@gE7i+-ZhKg
zqF6(^;&O$SY{(V$&LT=;fAv$5d;ql6h#C4NWYkBXnObYx*YLxvwBcqLQm|n$kO?E~
zfjTK3;L0Kf433f#{4fBK5(?0g{ApUhEC?Gw>7p<$Dx`KPVlUew1M^iGj*_ILyE`<%
zq+r>69V@eZso+^^adp<^Hr%FVL1lDzhf)VDa>|C|sv^6AU0DXZ5i$Z>8G)^gg-;m{
zuKo|&)Owc@G6)j{F)%SXm%-pV6PM~pH^Twpx2{PyG71EIW4VQw!ALh00x~g|F>@Xh
zmk{^_D3>)<H<N$f>*4d;&pV!2=gevU=x^=6^V{z4T@geohT3(STP7`Zzkl2L7a)*n
z<48VcNWAZVbYFJAv{cJjsJi~=``4~%`87dl1wqNRMy#NenjzXwYsZv-6`*uT!VBYp
zqDvk2ntQPL**{4yI1b<W7Q&D#>4A)I6j1;I7pW(l;xm8v=L=mn_Cx9iCjLi)l6yF*
zA3sfu2+tjna|4IHkQ{iBzftHq$RkA2GDWHyB5D^9)&D!9`q_x`kGIl11fQHEi-GJW
z$^pYbnXDXrJnTUeVgGz-sB`)>{veOks}`dE{^Izeb4g?<T#i#fu%JyMCkUjdMxAw6
z)J!SYQtW?0j#L<U=JHSAxoR0ALXI!EY$=MGpVHeX3KP2RjWSAvq%l<eGsvVxuTh?~
zNt#8S+GPCiwF@qoY2^P7jS8SpTG+>>fZQ97<;qesiDWh1wId|ra8qx7Es7bZq!h`~
z481-GTEOvEu7j0NLYtBr3B$N2-m5nFi-++T)f|5yHoc=@b`*>-2A|Tgt@xacZD@jz
z1*R}&wXB6~kKx(&Zt623Ce1|>L+u4&<}$T(jsciPsMm}3+17aRh#dsa&?(9lc}`|<
zV;Uufm=&^3qf#;_U?q(<_bXB+N_uH~AFM!r1mUP4GvOjRZoYcUy4lqw5FR8`!?^h9
zuf~602+b7YBu1)L%R<nT2u!%^(77t+N#mFirlp8yqN<ajL&6sB!93tSHf+wRsSin0
zRx?|mCW9nrWKKc>A^9-aZ2sutzaa++8tuJ8@PpL3UlKB;4u6?KWmVi<+`_UeCwh>h
z7LS~Tcq3ETu1EF(-WBSQpKw~;`EC3v;wpd9TIp?-E~|35AGI#TUZpJs85p6xiE#{Z
z>LX{Ht*6ZZjA26CmNAiIW^kZ+<1rMsT$$ZWIRC2oAu>|Ml=(>^=hYNx&m=mnDXd^i
z`xfr3X3m%q1>LJ9zzw}>h$3-Z!eTAMq?+?vuzk52E0AQ;X_NuwWXg%^c$_PoERBD;
z=wxyBsjT}D=anT{ALn__{1#q#8bJ?RLm6$eJmid9IT$#_tQqNiTL+6#_;izl(X@r3
z8yPx}CH%-8KvAiyiosznR7>1D4Wk)mUTkSmNdO%Ur$oI)hjD^Qu<ht`<$_OZeo-PQ
z{N!`ca)CuuCFqtqA&?_<m2>Q(c~yT>CB??9U<c-vr5=?0GEsuw*ek>ES?e^UTpZo?
zm20kIV|FRo3J<b`T|Ue^(WIK3lK+h|ZJ;GHKPd=!Dz5_reH~RoMT-i_1t`Y|%Z$tG
zBu9?Zy{XayYntAq*HMNQ*@-jS*&gj=)}NQQub&sKQ35dHqvgz}262hCpu&INRpD}#
z9~N}TA1%VQ=v2KuI$H%Sfqt6W@^i;}MqL$OiGuP8PN*wUsCz#Hg=U1gu<fHxm_52W
zWY16kVm&181IbWLRAz*$AXUAe0#ZG~Y;H?oFSMIvN=?{Xoz2|`n9p%#v3nJ;s`pa>
zt4G+b7+3@B{e>pq3stXDoEd-F+e%tN41@6$Brk(Nxm-glAt_Myaq=Lv3jw&KqLiRQ
zz)<mjuHy29sSoRBJ~^$euxQyintn<RRX~@(Pz>38_vIyiKsEN%2Lull48sq8BVb+@
z2T{P3!s?@cmq!6kdB(QGh?l)I=G~9}X5W2U9u%q38`^-xKY#wUJgk4A^rmi=j;3i&
z+#;M+Omi3ump-zcpW!H5FJoB4z<7g-6R7Z&TxN_)M+Y^CKfLT|3bw+aqP)jV_1))}
zC2Ln4$ni~M?R^yqsmjaM(XA_ULO8P0(FT{Wf+@MS!L_~sA6Dy@*Wf9}Fk#?^HTZsn
z@YD~STZXI9X<fb@<<x)2oKoPM>sI<HX$#pBd~7v1?*YX*%y`!)>Gtsr52-EX6tm%m
zM?*1JW;bngkcoq7DM`M#(b4pNWTPc1v?e&W&q8Wfr1p0ByEf$RSSY5LqDI0CnB+Ar
z_%@v%*vcU)OgcNp{Lx*>P+4TgHr(HnRH7-nrM1@tU+*{eiqL<Wtyi5lF!hSwnx)r+
zKeL*=%O?s*RBwS4!_5<zk&Kd?-_UEmNQ+aCY_SwxgrZVzFe}Q-MmJ@cnut0-q<YB}
z6I4hvRA-bgxnatUd3f%&&K=rU+%ce#)QUB|Sfr?z!I4;H*IR621yqoSrJTpCK!IV#
znkh2~vnHO7a$J9I$IM%)81XkaIy8wLStU0kegYrOvThERei3-E^m%{c@k_M`ef5^M
zT`RpE?@369yyY-^QYJ#2>a!lAr=%-?<U`9XtFwmU2FFnviYv!a+VMBril5k+NH%be
z%f-FYdT+Vk{?Ib5Z8?>s4twFOHvZW>eCA1bPv>rUAZ>pe3tv98b%lj*a->TrO+j}k
z9Y5deuR~D|S-G_|VfM7qQ9&V*X5C_V8m;L4a>YTWwKXj3y`!8-Dqdi!lL+H|qSYOq
z)g2w(o-UZ#)mhxJpm~K~Z>)t#6TsPRF@p>#i!>hvNtQ>m$Q%W^`onBV!3<7^)s+<8
zvRh6xZGREs%n0fVHyk)?@8`GOe=bfhm(iy?6az3aGM9mmH4~T0zc475oLo1Sf5ZOw
z*I%Awg8uPDZgPUWHF^!Vu_zc_#;y4TWC_H0$;6XBo}Aj!+z<(aMRQg_BF-luW75FC
z>-ldu@~2AQ1u3i#`guJj$x6+ZKne)L+I=uiT$+}J?G<POxqdx>AS|c><jctf6Wain
zk0%TQjx){~!)!z0O)z8l_&ogefAj0pKTpLJpamH}8RZ3->5p$u;XNvrY5fEsyu#`G
zpVQZ;U&lk!HI1K<anbJ5F!s*wU<C1{8X!U16vlNnyJ)a=7-w0Y6eKSNIdz@PY~uK>
ztJV#x;_71Y%=f?r3VbPELKSs>1G7y0G=G!zBU4tmx)1BVum*}GH#0Hre-0<w4o7+$
zxjO3BaF|TtO7ycW%UNcDt$s!c1I~EksnlS`h*?%|48D|+7SFtJTTFUwvD}v5YudFI
zww;AL8wN?fvt7Nd&8Xm*vT7k)UqI|pu&a){=p|Vu>c1%$lopz~Je&}Bo5Yx7y|rE=
zHuGWj5XrIE$QX@fc*0+>e`})DDO+q)A2uzc(IXmpDpo(HdnlX>cb%DA*&Z>=g!7bc
zN>=^`J=CWz;uo805{_k)h9v(WX~+#{?hr|!v&~KkiT%_>>Eb1N)+gAoWs~|eKgq7W
z<TE_Tn3WMmx?6E~yH~LXTmkQV%K=})e;9C|G9QA5D;iz$LG)UHf7We&js8wc4o2^j
zwtC;DpK&}PwBz}2%mQcACg&J$a6COn@|dd+8SQM_lvRM715PW*kJF^5?;*)7-hMfM
zi$I=}9^9mvsa_gU(K7E{V4y}N92W|M^mv+zvH4zl>Y%57vzmTLdvKGQ7}S&tdQvXl
zLJ2BkSkf_B0OX{Zf6PoNpvJ)J^6H4xrY8sPV@;;><wS%FM{!DkElwL`s$H8TV;MN-
z$@&%6D&uUfygEmt(Oj@m@zthsT~XpbIU3OjX-k~d$Gy_9Ym<W=@1-Q@^WajVI_CXz
z8reUq-4CS_<WkmndnPR3!zpqI?<LI!b&Yn=m4+RPD^m<|fBT}Pd4w*Vqzk&Fzn2A9
z*U)VcN;@rlZA3$>NRV5uOxmu{p5Nq_EXS8q<ya}&ucNd!MrbIWtY2own0dB;1~R$C
zZXFpLE(adJl$Dr&i|@lJ<-68yAytF%yXuNTUm!cQ5p`#dujFdWlGYE!Sd~B9veN>Z
zrnWQ)`J|jUf9QC>#9@FFhhEs(*Q6ICbIK&y*K_k-Zu)RtrU-53C@XR2UK%0$B;i5|
zT-k=*s2NEo(3myd)*H-bFu>%{jy2O{oZ(O7CCqxSezO%Nve%#!7N{$=AhD40cw{{~
zr$LN{P&oY;bC}Cz*V#hTDj^;A*xzeB10Lc)<}vdbfACItqI=h`H_dH;D;w|qW9K%d
z&)J$V%;eyF>{%n=mZR$gtWZXG=s$e<X_HR|+4J0}@H_e~Z~+$v54ZbmXtr!^wIAiU
z*EJHC9JdGmUY@IbxR?hyuJT40u=noBZgN~@6sxNKr;gjAFOJLajqp}GT60`wNPwH&
z>BHaKf4**kJ4QR&sMpE1J}H~Qoq;vvT6%%HxBCScn%RhIaaC9LoC=Yd=pH&34n#-O
zXQoqW%+CfqF4C-D-~Hq6arTmxVJeM6sf3xE2h~6z94=12FsSB|l&@=Hgyrzot`2Wg
z2mZFMTS+aXe?X(!392)k7OiQg4cCoyo#Spre+`E>lkohXm{T1vHNmE7<x<GH46B`;
zn=Y^w{;f@ra^tp;yhS68Av2A(>&D>fYFnP|!6e2BLMue+1JqU>?Iwks<$A_FfI^)h
zE~si#A5k_V!F<<BEyo{dOQU@a#eiE>hi(ua=K)%8Pz)4|(*mV-o%p;)SJISV+fEJU
zVtv%)%a?RHdFoMQg>S+?BGa2Q;j!W?^Mn<&FwQpWj=FzLC(XaIY82`a!Y1X>^Go!D
zx#83dZ(QwPN_NQxp_(JIYczL#wpq7MfEr}JV_6@{5x_T8_2VbkvHt=H((-4Q!ALh0
zm!RZ76_;kX0VtPRV>grmxVP_PH_rtGygmMPm%&Ik6qm5QIu!yoGnbL(GANf$&MgXm
zR24-%fkA*yW-`Fub}+Zao)&FSJ#7E~VX3DrI*F6mO|b_jjx14pe56Rma`-;T;cfT)
zvHN^PJviYFI5m8H8-N~2`UFXX)(j3nScKu@$MDyal*-ceOqZ4tkuFE~p1#fg6N0B1
zV9N#jPfo+rI9DC&95i{A3Uzh)hO&cyr<u+>8R^k(K)V3$|Ni)6CkyLR1RHPT7ytG8
zvHN$|1W_oA4TCjKfP~?nAG`QKgAvw+K`9YH4?n*S-*$g;s&y~+oz&)m)F!Pprc;};
z!WchTivpdiYSScuApte1LLcc;^qU0LG-h1+9p@bA!aOcXaI{%yp^1F^`1yT*U?cBu
zqr5@V&YE)oFY#h3B(+E(BMoIT)oz5Sem(|yQqJIo_sDm_PKIu`a~e}*h`{NX*PG>5
zN(vy-T<+sc?s0_|*3L8);i^rxPqK+Ab}PSeVmG>#@=$m^1)a0*RZ*#&P=HN;X0P?r
zNIzo6n0;~@1JNn)sUYS#X?QY!GSNGYMk<C-r9++5W3bI>(llCaRsTHRpP4hada?FQ
zZWT1?+|A^+Wz}hIsSpGP%~#1|lTBCd5<h0I;H(z&xNh<$TUQ$@<GqU5qRXSfjJet9
z``cLw(Gs4W<-A^w0L)WPYe7pCE+D5I1FZ=$RH_5F%7_wu5&=nmo9&r@HPhxEI(tKz
z<EFH<2Ygm1b$V!0%@9xiPWo(~zF@1w1Gs8qs7Ht{^IV~!3M%P7sH~NYDB9|R|C!(f
zQmCuGAE`c<y<|c3ag1fR9y;0v;U)4z2k6_77hFDdavV+{rUbuk<h`6kXCibtiHy_2
zwewrQ6bKx%k)ogq*s)N5H%NEmzI9}J*cdAi1K;E@YA@U51jf&tpWX^kj2@B67Ac5E
zRcO16`GgZ!F%f-WjMx#;R>CAkSg8ulNJmxAUsm>_ri~*1R5z&ltQM+Tdx|rjZ^#M=
zh#(cG2ypUb1V<;Lu``M8&0_B7TWRtbQqpyND;-3@8Q(Fz)DymcBhT&Brf5EXBwo6a
zsSc~1HP9)5+n_VbLTg?UQsJpMOH%A9uUrREAoYOR5}-ZBEdbh6TmmSQ=r;oND!v;)
zncfPI-BIicluSKr^S$bh>vceuUIfzOmO!=*$hQl#IpDRYhXm8@0M96H0C+}m0q|U+
z-v{{X_^yCwvumDz)iQ(v<_59DDAR_3*Uk#*x5T?j2H4m9BJq2-Y?m4wZIsN7o`wkx
z-lmG3K$N9F_2fdO(vno{f|02_Rk7`iH{K-D6wXYar>s@%G<wic2(_Tmb@}CjkxF}`
z8DC$Ht(Hg}`jNv_&$<!Cx90ii6@`aKHJ8wPVMiSIs^WTB$_~#<&07MMbHW1fUAJy1
z<;0(Dg2iuAMUh0juBSw}#4w{~UW<3?K5C|}m5TiYdj6$g(oZm4aEb{n&DkSkY00rn
zvH8MUcu(fU{%87a#^SO8a7_GGmuqe}PJcvq0D)?mV199XfbX)pKfnn7PZ|Wu(?RR}
zhUqAyxIsF~C@!R<T%s#ZqBW_f(lipT<gvE*+3e*2KZ)GWxAW+NaNuS<>YUcFf3d~}
zH-)DeK;&Unwi{WI5EfY21lZ-jwGDyYQ``dBJ;fzpGl?Dq?5lWe!0u-+rXhEL*?(FQ
z2wa>mQ;>AJDrhdyuUd;{O_5)#1s*m*Gg*VKEz#Ul+yc!##U(T|i5`UJt9Wc^?q^@j
z)9*pDqEckelReE-#eF|}ErJ1svgXX*TObz#2-I697>wcu$Ym54kjo`{405mIu_Bl2
z=whjt$L~rkKqxu0ae6i8LM*2GCk;5@t5zZzycFoe0uPUu!EZM|Z{PjeWwB$2vTlBF
zM~vz;a*fCI!$KtMI_o#huTgup+D!ASUe8h*gjSC6ym2$FcHR3;?s++@9E426uXjf;
zCiaDwvq}=lL|75qbBV<mI1|}}6^>l0pS<{DSeX2a=SJY&^uiebCg=JOfUjP}m%&Ik
z6qiACF%$waGM51=4JdzFOON9=62AAZ(B8U%Z4Rqgd?AK`?w)vnz3rfH*_?t&E(wwy
zV3EI{;!C6`TC%&HWDl}zNvtYXef5;(^wT7#Pj}C+cOOoQCM%o<i_`0m6YvempCI|7
z(US!bM#1#@%k<qrN;G^t^FdN7#fRnZj?en|sKD_PVDbUSM@xTjsL$d{c}D593Vm6f
z&yYDde$qNC13l&j%mwK9{p-K(WEOpv6P3`$PsWm9X^NaI4-ub<scK%W!&x6Acg>P1
z86=}b^TWx(PzTxza6M^cVKt_MPMN?NB|0g{^aSW7$iQctzLN=(Bsu4MPeDcu(mwI8
z=em>myPltzua<u+;Xa)*HQ7*YxnR*mq!Bjb=k=7H*Dv!G+fZ})t|^p`k8l`J+tRJ_
zR*=G8H%wzop(IhjN}(Y41gzU<ixnYF0}nG+M7$R$sHonp4QZ)_S{qxL%?+_wf=I6*
z@{}d_3$s$Apfivy?qE<zhgGdfOF-LrRaRKtFTPs6PA-4cyD2bABM_+PxuZ5@wRwV-
zqyd4}7bkrwvP9{FPx!3F>>t6rGDz%o>qjpc5NWN?(F2mFuKXwMRv!`0#+yiQ(;LVL
zkSom>d84eRf#6sx%A&A{E90Bwqe(MScDu0BbKXsf8uf0m>}jD*ok5v?WmH5LIb{vB
zC@H$Pvjl%WQ`T#&%$iyOT1x>~<3zC)<=ucSJ9SBxctkFZi${CxW4Sz8XvGy&_X04M
zQK8V*%!$Gul|Hn4bxriw1xr-16==)Twu4)~NUG<xy>q6F7y2+BVPoTxRnC{C!jR5`
zB2v;W%OeSw<JX*Q5=o!AGDrEcjbld~<yU*mez||0Rg@sNvP9H^&IsT3Ap$K(<BGS=
z8#6V6WRGOK`aTOMOS=>g;c#IlPIO8FN;GOF9wH8<FgRCdmmnL-G+gqpD*Gn(JSQtp
zJ=dL_4PFc6r@K|xW|b`IF4$_)tUE29Fh>kXQ6oLEOtvXfT%{LftYCLp>*b=Ps~iYa
z`u~4VKPjK`&qJ#D<C9N8vN0B=6UJ1Jag9dxHm*nsBcZ48H^#KkjOgcvEp!Pj#k7)}
z(JUX^i*JTC7K*-3V;96X&hQG<YUP!60(+z<*Hg(^O$yeOa+4HF8)x`zyu|4%_g&*V
zpFi!EquPT2(94RXR7oZGjr+UYhiVH#>zM{ZFxYIw=a)cqHx_>`g4TA4Y7DE%c7i{h
zOet2lJkaW}Ra`bE@h0dcoIZ8GzJ0y>@5&do(h(;={Bjp=grsJypAd!TTKnsl>BqZ&
z_f)gu`*WNTMC(dp!aiE-8s3_lH1QTW>lW*takNYQH8^DU1I{<hYW|2ATwM8Fk%HWS
zaRbv!10TaN#1?;s*#@_4-HI@;2>HHLXc0eD4WxXn1>*f2msC^5hOWX$I5eT-sn@{O
zJ5+2xG;qNVgJ<$`8EBwRR2NsG&w4qss!~$el^Uy4Ef}4HhT@Q}mFlVFz&J+U*#e?~
znp;JO`mzyEDfn#~^OS<VHl^iCN&YdA%valDF40&3w+w$gMvX&kL}V_p;wn(j*fc>i
z2X4wXgS}*Twf(kt@N_ojE5lamiT#NopUZH8maYi-A5I)goY3BK?LQbrQfO<5>WHA0
z<FWCi+-8v*x0HLMP6}lRr^SN!GC(s6yp-}h*|}-^Deo|pThT~i?Z?*cr9~Y|4Qr`q
z`<&Hvk(_@6vGn$q$q}FOJ-14ZG1e`r#qC5$VJZQgkWC^~_6Rz7a-FojJ@auB#Ukjx
zDDx?AIOo*q_(Ex08sp>Jxbtgs0C2YuOx|&Z(@k?g==2mjZ^uP7d&{Hmy+nWvQRT6x
z6ovpH&&P^2_Z3a2a8=${R)AG2jcnw7Vo*3(?aqI`GeIc>f_uT1-ZF`li9zzF>@}Qg
zJ@Y|B%Xbf20AFfp-hcOymKYo_vBE1@F8TT45&ZqHzdt;pz5nv{fj_?~RP>OFGr~Z}
z7#=odWj$)XwOH~WM)@YAP_H4leidTjMZ;B~gjJn*pC2{@5I5Fd_12j~E4%TQO*ttz
zs55`P+!#hLUQ9jT7OqfoE5@9=a24YVnw@x+JUM>o>?DgTpK}a5<KrcUAxD(v)x*bT
z68w^OiqJ)t!XbXEjLVfaUTJQFlwWaarpC3ezbK*vfCBae-<R`Bu^uUy@Cy&^R3IC7
zU*CJmM$ecAe}%Hosv#@3qqwL%T@BN3NHKpMWt8IvJ*SH9>6PxTZ{#|P0OMeu6HL$F
z)oyxz6<U6~w>6kKZCof9@l(o-^s=;1C<U@CZ%^20wcz@AWiww>Z1D~@n1l<vIF3$8
zs;63#_na)Xn(9aD8rmx@wu&D(7lPReEXi>3wZc$QTU!c(jpEe%95&2$Sp_$A*d>2O
z{d~QH8WN6dPwAY4S{0hN{^@r(XpX=g!83j^$A?alluc_>8>`E!SzQ$yU%H&?lvWQQ
z@g5AaL8PM*2qOS@B*OM@9JwqAQbOvMieR_Iq6+WD%q_9Fr1&087%xC#Pw^Zm3MRW9
zb)cYTuD+VaHA#I8j%0>Y$>z~Mr?Y>#{fa!`QINZEr^vN&s8_qEujv%!^$|jm71r66
zj4DAJ)KFGs7>OGCRq|zk<~O(-m%N;;Z#tw%|7>Vyapxd1^48db55<`09hXY7A1r1m
z7xCh>e3cfarj{!&K+e1FZ=2^_l4)xh0JOr#hZDgRIvdM|?;@GyhOXV<{Xc(m_eL~c
z!caD_T=I{HB^t-=(d~zt;oleJ=pFB|Vv~+^Q`4U6KP$Mbg~N5@Idno(C)b}foda&#
zLJ9rXOy>f0yj~7;XFc6+B9}8?$L-b=IYyoV=()UMA_ty?!IgH3H@9!x-#U>);wyfo
zk&TI*3J6%q^zMnA=T6eoO$T9e4aqR$lH0>H?L~+*%SEh)?LnU9C`k+H@4tC?<dl5>
zzu$iQ@^y0<7yeMGz$_MjzdeWp?)P;k<K7sKT{aBU+C_U)IL{`*&LkI3;kdgsO{23j
zgzF*&N1{IfN}Gbkm%&Ik6ag}qK~)zMm({*EDwmwtJqdrPHyf?Fd<<VmU|`pY3-s2-
z-jW=OBD<FaO&3V}_dBFWNgTcWkT+PM2l*k9GsAh$Ctf{Oy!v%@_prJ-D5xxRT3A^<
zJXQi9aQqY;ZyH@$A(&C3diYv>c+0soc-`RzoGXQw#orFU^=_+#!xh4$OYF85#M^eQ
z+~+k)*SmkD&%4vNNE{rlw9fKC54j2CBJ}Rl!*45|MCZ(#@H2eTfA1ewf3KQ@NX4}A
zmDY-Z^VOeUSK$E&&5ZT<*f1?k_58VdT>VbTHq1EgsXsKYo}Vgm=RWDubJQg_M(aYC
zfroaLF`lXLsMaOC<6PsZ;egj3FVvG1<dm61{MCOl&UcV;!$ljWl<#pt%BWth(E?iH
zg@3dyxZE6^7?j%?CL|?@bfTyljRI&KB@^-B3W1qO&s@z6e7zn_xVPIH9-JalFyWs}
zGy@%Ukto;^`kqxbSO&_TpzKjKWW|;gqf$|3NN0sZk{yDO`6J3+-07Oki5O&rvgCza
z&1!$onX*7v_jGqWf5{aSNQqDsVvxVYl%Txq@psAI`&PTne;2KMvRr_{iN-imF7V7N
zGo+W&KKTEAk_8h&@+PTkS&BOo{w1wuV0u~zsVs7kZFHc)Fm)Z}V~T4gK~bt0GMr<K
z@oZbIs`ewwUEJyVR*nN@7+BZ$mfCVcjA?&r>P$hhp+t3A#0tPny=1@^1ohqN@o9iL
zu!PW9cv^*N(vZa$L}yOF1-+3NXmUfTLr6vqS4h(-700xkbP77@*wwD42b|#&_rnx}
zIey6fqU+5_lnY9hM&H#^3fP4zk_B7P3pM6s<6uEIBor_42&}{cl}RX$`9yL^dLn<~
z6rnE1#vs&fQ%89;2c%q{>iMv9)Mhy3LLTjN7%l53lmIWWT|4rm(i>*6k0q<+Ay=UK
zQmJE!4kyBiP8Lfk&9I^E$)#&|7HUvFg;^cjS-Okw2R!)n5JTfID;Mg|>T}5G6VbMF
zZZQ<0p?D07QIkADlKBMDNLl2Fx*30KR0TK$SdBw~XV#cSRaH1y6o&fQd@XG&y?~xi
z+#Enunt5x{*d$|F1I3}%zih%*+ycW}&=HL&xBRr37_3^qbl5N8ZOkan#C{1AQSF~Y
zPwtz%*dHzPL?%ONXhut&^H5fHJueJqwlth-EH1L_Z7w!vcm^rw*F`3nEtr241ESXL
z@Wd?fI1cM|Db(%3gsIm?x7Y~{;iUUou9|1cK_`y0U<^a&q2sKS%nO*AX32M)Tt`Y+
zN-<z1lfF%?g_61&O0jRxZnX#}ft1BNk|NJ<-g8MLTmtS8-HxZ^5*_5YBJUk@PEm?E
zoO8z&L*KOyeM%*iju{Mds+oUxJ<T*blx29*-YIRTlN)l3Q7kC?9F7t~b;-pn=dy$A
zG~w4dKHn|+3tJYD=KD57L%N{qGSy~X&iu{O1Z%V<DFrE}2OJBHM9p=B)JV)%u6vaA
zV@jDg)D7;i^ziF{He03Fe|*_&C0~F4yqV<a1=ej<?0>?I)~-KtDbs&mY~2U;DvoK|
zm(t#Erjn2nhIy}usZx@;IbbMH*;hCwye*<Um$8H(CFP%$vZTe+2=gNmIOR2(UtSr3
zqg*aV;0VE2g1{+EwjpsXH%SqS3nFlY@(cn;Rxg0S5sEVi9OZvK1nx+=5&}1%mn$N0
z)Mm>daD?gt2%N&2W4V8R1qhr19_Q2rBetY?0R~4X&46%Z_2noWAvgoWQAPfE98O8k
zoOG`Thf|pAyjao|3xW%va3#SR5RM$@HDPc^oijsl<>IM9@htQW!^%>D=CC&v`|ono
zLeN_{wb$~0=xrW)lTOFU<fX8htHW;H104yw6<dB3>=v4}P9=Xh19p>GOEiZ?O<Am~
z{YrP?lNspEF%3knYawp}v!X;hzYOvg@FgwVggT(l0%JNCeXC0v;kG#XW|<L!a#0C!
z>lk_-bpvK(Pdho9bHwZ063aUYbpyQPdm5QS-GT@7WQwR;Tbo^q4RyX`Ce$2tiezbE
z8j|c1JW$|ZqdI@Dx_f{i>=PI`>ai)Tv8WHgU!iy6-W7b-A12W{fk$=5lv}=Rw!*Lf
z`Zu6%>(9>{{CtDAj9GvG_6`3}$q;gi!A`y)_-PvPQ#eUVS#Fud`ohF*6*=tvG$XB<
z0890=Mds_rk7!&mjcnp%V?5E-n`v!$fy0n0xBNVVe>#8WEYXH|nn5wcs}w^WEM@GT
z8v=%^gl>uipB!tX6e>IhXu5MGZH$J2woF<>snihPG=GCPu~5;S!^vTkzgNtw{weki
z;xooy@_<Mvx-+~IZ+c2UQ*i+chCAczQeo=uOMNU~KyMP3?hXkyy@Sv@85_a;N;c^d
zOpA%d^pk&4mVL2%`0azs?=xJh-|jz}-V~7|PipNu5h!Rt!!6B#u|MJMyALQRjPmU?
z0!Kprq0Im?uPq4q8(<bQA!(ID11i*u0RDnwCb8zqfIsDOG2l-Kz7oLS32aQ<f{;x~
za6!1gBs_!mmx~tw`%8i|SU=%@J*fXUsS?uP*U26gfqr7GWpMtI=mIdm#SG^C)lvRq
z=1R=P{{i%u5|^>2B@+WNFfy0HJTnuQzkxTvfB7VH|JUnpJ6TniBD(aLZv5@#b@%VC
zGejFfE;basFq)dbzwXi>R*`T%HbD#g`R$+P^X_*pv|CDIM)@JxeEZU{FQ2L>ql71=
zb3_AAYA1+7<0uFN!PS%Yqm%@j?Ge`j7rdm^z{ZDTP@@vb31cFcNY7=qut~4>C#yU7
ze*$$oNzRI3J(r~AK#y*;H^PwPqY+wZbdK-U>jx*4<^q)(Up`7<yghRy9A|yKpa#YA
zji`65rRr!nTmn#dm89#O8U)#@Pp0XAvigHBr|P$%wO&}`&mEF%!SjMfX&k9u8+tBU
zD!Whnr&a6<I4)vQlOPqh#Ntrm8cet-f0Wc*;uI)aOfriYD6A_1Pj064>}xn|u{eMC
zQ{X6JIPUMA)Q>ej-P;ONb=1O!3oX^r36EsWM(Y#pGkB=q0vXoVX9Z(OAZwrLWec2@
z8P(5V6Ch{vS#Z?V>jP7;_H2EcUmv)E<!9?3Mt0Eu*y76K7&(P^-qFeF&n5n3f6Dt<
zKx`WjSH7uw(2>s2T0-Smf}ocvHHye9G+EA1otA?XRU+GY%Hp8))HrDbcE+Zz8jLH-
zFqkKIq;euZ>wZZb8o7Lcw0xDkdIpJyqg4%w=sjZ15~g={VAO}7@6uFEvEx&)51blK
z`RQ(^c<EsoqzU+p4G<jeE@FOxf9KdgpSrYsUcR+ej|;*p+m#4y8f=}NojjgcYXL|W
z+bXylhtoc_Q})HT<5j7KSpfyZP^qV*OO>Th>iNAQn=CyPy~>(31zlS*DlI%&UNUMe
zB3wFOGxB??GV2kIA~eTz@a3`^5_wI?lWLqfoO$umvztjxTcj(7#*CZne;KOZY_I~^
zpo2_>iVba)sKD;NB&zOf+P1On5$vRGWq!>*4@Xpl5&CKD(o>{J9Ibkw%7;JhX0avV
z1?*d<eFe|M5$>O}*_8;jk*PT{Zxvagc<eXZwNONugj1eFQQ=O`f0zEnTAYB|Tv0Rj
zE99qs;r7evIa{&FS2^_lf6IxBGGZ=5cj+ydz;*d6)qJX!m*gfZ!1V>XMUruH%Vv>D
z94T_IZ{AcsA(-T;=CO8fZ1kGy83(LF*Tp1GmCrcOOw?!VhZ^rg@7?Y_<S1XQm&N8p
zFfP;!M_Dv(%rCq{rI)J9@>w#4FYQTKcjkrZ%IPAE8A(DTWiqL4f4#M$v$lzTvZI}~
zX{*;!Pc2o0o43>^MB(+8Ej5$4g5F`07U&%|dOf|v1TN4!+>vij?--9F`W>-5secZn
zw}j$Q;s)#vleoa{aEsm(yEAX;s$^anA-2NGuD~P|DdJK~B44+Ua4?2wevl|b5~bFZ
z+jrK(rfG)#_<nqZf5CYUbyn~ZIw~9!VM`F1k0-5V0h6iOq$;z-M+}v^6(6CA1Qg$b
zkI4PrM9#0fRXf<E`<y>;(y50wU<@?Yw1zHb3`$IZhLO%+!AmH71~NZ<uB|_g@v$jR
zqUERI6j`UBS{|K(RkRyU`DJ012*)zT)ZMHt9?h#^ihY@-e=(HDD#~-0R?M9_UIvKh
z{2uq-KhEbd>ad-^hQ7$oXn5+Au6=5`oSc&N>06C%F=RYtXq_qr!&RJw;rvELCF}yX
zmueE|IjCa!Ugsu1oPNz#x>GysOHD6Hg{Sn~{bRevaY`XWT9}}iLXs9Hk%;JFhm$Ow
z_`F7r4)cP7e-uGqxj2gXs6BUKT!PER6q-U3)sAjOp#cF#+qY!UQUbU+fyNeD&7TeB
zy%YLOSucz-8+JfU;)e7Yle|EmvC$jQXH4P(ea6?n9esA8UX?x@@yiYAGaj>T=rbmD
z1NuxEpqK1BpwE=DC|9;5yqUxm^cj=1K%cSE>*+HlNN|BZ<BoiL`V2*!u<waKLt-Y{
zlF1zk+<-kB3S1!1m^tr>JG;ovmIh;9UhG#_f!gEH_(5ddjL4dc-t|NlyQ4k$50R1%
z441)3HxvRgIG4eh0~42Jh&NaRfu=>5z=$^+e}dvKqj;!QV;Cb^u;$@Q^WlzCu5sPt
zl9UR8%iw>DyE@zoX7L5n>0*alBiUW|EkEXOgo>xQ%cs+wB@7l{8f&PB2W-sv>+td6
z=N(O?Q=%*twxL8QZ8h%NQQmyR+wnK=-+@=lMtS9eDD4_;Bmqa6R|QHG#|6q1D3tgu
ze{DfUEiJ7Cz-Iv9nMyFPK)wVU!UGfWH$URe(z}en#qT2`FG8G2|0@scc<}O+nXY_1
zgR(t(b)a)fyIg7Fg;LP_=A^xXjHlcjn37%w$4GJ{C-P35_U!xHyXx0Z54%5i1F<qq
z4DkN|yFf(0_GRZ^NT!H3uHl?GCYx`cn#bKQHGkBs<?0>`HYky5tyHN*O_Z{Y23AP2
z)j=V@bA^ZNryzyMump<sPcX6K$@`umI7u`gP{B^P-c|+-(NID?46Y*7;z=-AN5^>O
zR1gM3t$9_E*I+}Z0^5|-IH!c+Q3VMCp!D8DNm6-}t#q|SxOBt`J&?%TfkZmvv_#4Q
zO@BVr(9_{oQ||;pPogu<ODJ=kNV|eCpd!w?QNoU6zR?PQBE}{MQPx16st?bH3u&`E
zsu+201LG!nHo^@gtGSj1Or;3Jqdv@=$9+6v%;*sf8-TovBXmEDMFR`h<&n?(Th95-
zhk0a{5r$XA$$Uif@gst`n!exnx6IxA@mDu+nwwAG_W1i6kF>t|>+4tio}+;aLyWGd
z|G1wGm=bA4h+0)CB)&Q!@XRS#JHp$%;xnDPPpN?>6FzmhWo>+w0&l6P<7)_r(9wK=
zjo~KzQORG5CzmdZHywZfI>h-&Kd2&G3E>7<j-C8MkW#I8EQLL0HSD(rcwN5EktsmX
z{ufU$Jsh3`#In2>h0UcmwH3mi6hjtl2{8y2G|R+m&LwN|Al%#+L=|LC$IB;&V$I0<
z>w)Z6#P*aSD~ODZXB?Do!tnuq+tmnJy6x|-$Xb^T;ha#$nRI`p+&LnbQa`?4Nwi#1
zRTxF29!S9#>A4Qb-T{Tx1^JYX391N}qK29)j5cv9(R6=e8Q*`>x2^0448<$7Hn5hc
zxllciN@|WLPuW<Yy2@jvW5mI}xAWKvqPk!2p&I&-riL`-p+@+2s{Z;44^1Js0yyCa
zIY88clu5Pz^1**P1>_K%lVGLF*)<xR8Zn70j8G!bQKrU>Fc{zhMLDfxYJ6Sbsbn&=
zdYP~-ZBaQwo!!zFONiIl1nMM)wzPscWfa*aD~>9oC^|kJeLbi{_PReaWgJlrr<~4%
zeom9geRcD39;e))PC;Q>OZvjzmadN$Kn3ez1A0|7b0&Yr8z!$IHlUlXuP21BC%72k
z*Yo(%jVn8_5PBT(tP5OUTGSYB%V`e}8VokKWUgb8BMdD@$%r7=%$*hkS4(9PyP|4a
zQAxBYMxjMP+hqR}bFp&@fKUVk#}A=+U@fms5+ximk%1XhG1j2h?6G_EL+UG~96#Y(
zQC6JzZeM>|=6K(>{rFQ-MmXBy%g3MIOQ~^Bt>@~fEQeFmEj%ZNvx>;48hQ#Tk(JrM
zH1re-#l10;iD6P=mk&BqgI8hRc+i>azM8)a2A!$Js~~MT==3gZ#6Ti5gU-|p*4&w*
z;@Du&X&f=ktf9<W!nq!$?3zg@iUA_D*mTm_hk1YIq_a43-lVfQ(oQ<V?Qvmxi9Sb_
ztj|m@@rKjho7p+0^gdWg;QL4vn<dKce=N1-$wa}A{$}E>{q;KD`m@z>c?qfscNM$8
z!fa+a5e%n5YH@0w1;JUED~z#l2AMEy?|G$G=hBCOxf#c56HfV|HgsRHOV?#Wmcolk
z#%_Q4U6&VQj?7O(z;?;>Co;C-VF6KLnFwN&PU$T16+#U~=CW><1`d^>uC7l{gsbVK
zEKr%NQ5w{Nhjbt0)t1D(kd?<EQwMwO1^~HAjY+RBZ&cU#C0aiFQlCnZJ{SvU`x>+T
zv#KA)0_<En1&z2?68ZaF7aeN|ndhj*%y)lBskm!l#c+x5`CT{uj+C1ANqgqlhSPzi
zTJ%U#_GiB>Z7q(LCfnLZejD^XZ><(tf(<aOv{s+MfqMW&4k((XSDFs1V)GV`?iFqN
zKtd}m+&Ln-96-so7-=pNq0Nnztvw;Pge^C7g`GZoKZUA4RHsk0pkmw-!T577s5pPh
zac1ljFQ{rPS1+h4fG=c0h57^8e8UN232?*pR0()tJ5_$X!E&kuxUie5us_{u>O88_
zW@;#tCtOTb+S+C>RRY~$EyWolg0dH|mSWZs)CU_nY6bC%wNwRZVJ%gCef?Uh0=Tf2
zss;J{Ybhj+P<6bRwUoCYMa`DBSVDi?U@cWaTv$uh9KEEq)OnodwNx$X71PH{R#DOr
zW6glZg;i8qU{$9sSw+Db;L-#93#+KKz^d_Yvx;KK-Ada3S5eEWC?@<)@cDKW=kA{_
zUOx31JahY09C@zwlb=Qzb;b6{r#!pul2P<5FQ&erW7|JE-~9`L>e7_7ox2Qks`fzW
zt^Wb$QVsl<!ALh20Wg<Q&o3mG`q)4Vmy?h;Dt}>b8#faD?q5ORzVtqXOLAs7<e@0g
zHkV$36c;2t;O-2ga2&Nw>q>?#_j3RK-jMQIvgE5|sZj$&nq70bLvr4mH>6^SCAi=s
zzN3r8%}S0~n?;VbHcMTOsaeHkiOph{YkQ(*S0p#9x|+LXZ9+LQ6Az};p00$qh%^(A
zGJkqA6Q4N+S1IA|S$d~vm8<!Y0(_{+?j|^qLi00);8Kjanb9Rm^~6jx;GEiBa!ITk
znCz0K(oAuw<k-yYQmwr+bre&3Ta1o~Y#?|Kk@lwwCMC6}k~or@?PaMt($HLp<fz2f
zHsY}=&4nnAL+K5&<J8@a>X^7TvpD9`+kaNaxi`RE;AXpvcS@kp-j<xiJ~Ygeq}ok&
z!nJjV0g4dWQlY76c3SZvIoh8CCSUU6d3K_qO$`SpCAX0??p9TMm-P~;^@>@@QK^~Y
zGWGFNwm9f_5gd8Z7ST~sXq_deq0cB2%-kQt-57e?08>(XTZGwH3jL^cV~@ckZGX#y
zkWenoWN|8enivDj++U$iOAD-o=q!u3VG#-j-b{kYWN+gFvoDHNSEx^PW>=V+iGB*_
z{$hl7*50qWQu<>EpBlQqn9tgQlQO|ntZhKI6^nQx2s&7~O_LM^c%QrsQ~ES%v_NZS
z7MMM5;C@{MsoLKHKB=^|i-_ggeSeO9>}~O4Z84(OKZoW|gRaDGTYI7Q_O_PF&HWbL
z71IkR>_)8?J~ql2sJ9TaI|@JuF)rc`ln^6_bn@w^!?*vvnhc*$r?dI+=JxH}u3o;M
z{_KV?X4mf~*QdrY;cWQZ@b&QJY3vuTCl~WmqTmAnC%JfYfP>;25v|pSTz}4n&mFpd
zGki7sb~ZeBp8)T<)Ifk{p}+;R5&!%d{;hhk`Nfz>7Z)!ELcCxiuzntJL>&4G_$2c~
z<Jrl|IEK&1-Y@3wXHz)zkJn$@|4)9L&#!L&G7LYwpZ|FK)?dti8GgF`<?7kH$*;pV
z=kp)`el`DkI-gvBe||CXKYv{gZ!c$WhhL*l!N=j|`eOM0cG~*9!C(LMlB+&@Da07l
z$-_bIIJH2M2YZ2JqtA#ZqUfV7x<oH=eQGlkU;~1vB?5s_@`j1s)rUa7W`rS@8CU?t
zvd<P=@J&)&UtURq$P$(e2P$ZrMkk)1TFw#l{(AW5zy6I7%)Ve;3V(*2-d<jwJ=RRo
z44j{Xx72js)D2(GE@#(ouCO$Y>;X5tI-g^)rY`l%zkZm1^=5uPpSaj>yqHbrc*%<f
zc<_<E<%OkA<G6NF5oLmG(Jl<I7)8GTGaxJ0egQHtB-zDhpAE0BXBTfK^V8wg^B2Rn
zlMnN=rHOxa{=)>WB7cge^P7frZGF6++{|vTFD5sOY-#ttnY?>{{>AJ=TN2CITGJVN
zxW=>AOuJ)kEpHYHySZ?3+H<yzvZw4ji$<#r_w1=ilY)3O>RCVynb{ZRqd`q118SNN
z05wYLv7kurzJlO6Vx&+uOyb-8KaBFMg;b$<3W-A96xIV^4S(|smWT52phrhpJt09t
zet59zQ<-#ECzw3cPZV#FPQpHdT`7r&Fj@B@AbsP-N|=v=$$&0d#XFSc9gvYIH$i==
z#U0e86b};lJ)pjV?iJiKE;!}Ig=yQQ3)~lzth#{_Ku5{sX1_o(3gN(9kz!cminQ>5
zSFSMaJ#t0ro_{XH_nnnSi$;q_OGYb?mhQ(Xs#xiI&nbm~xZ_cZW$!HE5R?M=qA)-i
zfLI2(vlfra$Uz|3NY}GB6e6oiwNJ_jKH4A6ToWi7WeKL2paW46cq^z-UqPLvkK=@`
zEoc%%7QI67qNef%@x`I~BI`3@bQWTUgb0c{35vEH&VLCi_Hw8RF1{c&QiZ^Y>+A~c
zT~KeDjxuu9sQG|SK%^oWqpa|d@u)x%5}J2o*ZrLkj6Bi37ZwvFPmGD}A+J!9rXUxH
z$Jj&Bqahw@`Poa2=^pXVPy2?&7BY3U*&=r-0t}1$QDy2<g86@%bt#eUoQD>faU~BW
zep((%OMjZx_FWdaB#xOTj+rKonI`IJmC>qqRvNQe8nam%vst1IS!!2hB*ojV1lQm*
zk|J^N5Q<Jh@QF6)GpZ)YgYDR;qTmMr_D3(77l?{GFBo2!RDIM7P)8J0;|wmrSCAAA
zz#+_@YtzZaw!sQ}@OgrM4eK>ws;JRXl5!A`nSVgo>ES#G_!0|iP$-#lhbN(KQ0$P(
znx!!1s)?BokyI#0Z+L<R_o%ut1MtRnUzJd~r;n4_nAi)Z;?4;PCOqn+PJkzDUpG~V
z01%jzxN$wUCv;-<d1(u#7Y*ejQu@tg-q{>nPO?e7?XaK3vZ6?n#E+{;$HBxDpW#Ce
zAb&;|XaN`dc20mwv+eB0-~q_B5YeClu!6ZH0#zYsJs4s^6{zfxPQV=ODuXgM#AxJP
zlL|Q3b~vTRyr1X*Hrm$xVo@`gD(X=dl@b<}<$>Nv@*8FHa*I2uk&EqG=vGZZ1`%0s
zi>AaGwSL;BDR|VI2>2be&WfU05k8QrsDD^xL(RP;e0G*M9M|Oy)u+iD<|TPUqr9O}
z-q0v-Xp}cJ${Q;Ci<OwMh{1MJ#3SFdm5LPGo*j#9ND8be2jJb#pK2)@4gh%|7^OxX
zy&!`hwEtsXKq<vivYjI43JyT+ycg;@qNL&b9ZQY{sgjR@dluVRI0Thb@+QV9LVtVM
z(Wjk%AbJV!%w>H}ZzgR}9*;fopoNSq0lI2i|Cr3uk0B@|uR^;|=49&s^Zw`eR|YFT
zyZ&LYtKs?W`Q>*Pv+3l!^Xa?qu4b3#*X?q@Y;GBkQ6=3)fH7AMM@i?oJbPK`0q=9z
zs?J<#c1Ql=?W}K`rXYXLTQy}Uf`5!Gy+u=qNFO47T=HM4dUvLH*|3w!PuxkBWs2vK
z&UvJB*8NGGGJ`H%MpZ$!HmG&9LdqcNZrZWhs?6GfJgNAy1RaJjrsF`TW@x6E_oM=*
zc9t<x0auhQM=4WMN%{<*Tp!Xw{Vf^D>pM<NDEK06)6|ZVQF$J1b82VI=6?;Cl_z30
zFJo3l%*u#a+4C%;f)Il)osXIZI|^WuR<QsJx*ulj@FM2>x`|B#<Rf7wwjss3Ck<%!
zr1wLL#g0ghQwvG}ujPj%AlvOdOWk;GT37a-A8pEz=CVOPY--9-(Q&I@qxQ4w<86Mg
z?02u8)rwuj4MpT*C%?P1RexP)tE^9BD~(HRl}EP9c|VTAH(S+k$5F{5MvfBwYFWMR
zy<Jzlq2qqLPEa1DqcBuRAgB}{z&V%S%IQljMkFVz?&>y8AvICl<v+@gC01azdK9&c
zxyL}eGv;*tw^a6v$ERU#`CLlJ(<7z*FqhTmaG!BR*<<mJ5~4X<4p%MvVUFzJzTQ^`
z6p>QvQG^Z=QYfwIXw6%tP~Gs;BilI3+q9&B`bgp*=0{48qH8trJ(3@H#@)i#72K7y
z4enCO4{%p>8F%A0X&KLSOY{Ehe}#Ldx|hL7HxvRgIhVoUIuijjmobStCzo(}Jp>IK
zxe>n4uTX9?Kuoh(mtmVX7JvNy6rbI!?v`3okAXdCdPFsgRbM?=tfJg~+sWOhhnLrf
z_b1dlC%gftcCTM|KtD+O3z7z{**O4V5q7WNcR%f=RF*z3^wClx(#O&No}T9L4#CqK
zV9N&%@0^DH@UA-6chKZnD%9EK31tOOZ#wT}q(`d(tpfh=_t(EX$bZ7R6v4)S<AZ<u
z`1<hghbD+ZVQko0;{-_9{qy@n{6T{e)`gu?B7olg{$=;|;jf%(U5h;vY#s<UX{|9G
zY|08_{LWew=v;*@imwR<W6D#~lLR)dJZZpujBsWlg(h<C&+p%MY~=H&K|V(-8Ov9M
z6Ma4~$Or~0WTc@?rhnRv5Y_jmAZO+Bkj#P0MG7;%$YZK3-KtASDKan&ic=)*OGzp>
zAtR+vc?IpSMn)lN+~EY74-;hC3{6Hs>~bkdOY&IOpEHm}C#trmoKS#cdio{FL=NF-
z#Mt~Tx{GmjRMV(Y^nqe25szd3-k#6ZsZ2-JSE}v3jOFN8%zxDD8z$S{`xB`-kbKJO
zPG43D<9#Q~<JYG$Ma@rf6Y(Ga_w)|U<3E0VdZ*;$f4@APT{D0%h&_RCG}F1qze%Oa
zWT)pQqYT0Y#_;DdWnNzhR)m0DUlhvXtcg}>0p(|unwLmsKQ##rf}rG+FUE(%IljSL
z;mIO<qS>g5L4O6@7ggK%O*8Y-^<B)1CYLMjmt62zX*8pf`I+8GtXH~ez8ZfU;SvLl
z_bNIoeLNb>PCOa?{Am$GROEkBGjZbvVsy_sW>)%mw4i(1AEeid7?QZB!#fRXpJd5y
z`*LWGI-nONS7XU_9Zjg~hGZBi{g|CUF>Xe#n)Nmg1b?<5Y*;{uV!NphaH@kv+bo7{
z0)=MU4>mGR8w3!rjB0P4P*K!HN+=Q!zg8qeMJ5GHHI}YgA*{qKjoCIM{^H7~p$Q$D
zP`#+xgWL~X_cK-?jzJ11abv_aT0t@X8pIpl#ga109iC>4d&B6|a~8)J!lm(sHdP)j
zQ}$ppn17T87B<tqbQ-f!6s2+B!i|D`VJVM>@#y%Mras+VWBguqPmO#R)l)Z}aZ#-+
zPj>j$xSBD;X5_BQ4cv$rFC_qH{;RWB6K$i-VH!%=oEuli`g_kYHjedJwKdS!rz(b$
z*E^5OtN~D_N@;g9$EuYaN~hV2EOsz5wI$OsMt^A9KI%?jMiS`3#v*N%xsP;1`&QR3
z+hZQB`D_<>Dh?)VPMHwr*U^E#99E5ptA=>hNO^C~xu9{><QQ_l&9+qSh4UQ9E+VRH
z_M!a#U?hhcIA`qS($1_Y^Hg*MWz1##J@cG`)R`DEypiIMDjCBm&7ZX{O;6P#SLYw6
zNq;ojN+UqhN@w`$NwS)1;9!k;D=TWcN^OU6+@EUhX0;(i_}m1(ahS`lKZLIB%-K@T
z5pDu+Ro<}9aM6p20@F67%FfC|-?>~t+kBj5^zO_R45)O}nL(;jDM9!B$@sHbqcXw-
z-KFC%15YW`jjZR|`VCqmfPQB68=%0YHGf_x;N%iV`eRDfpd0yIiaHC$k5bezQi(uo
zR!x2wg^uDQ3Ruadu8{s47Oo@XLXM%V4Z~!W<AC2_ud7Nd@mL6o6GT~ao@{oghGU-^
zCB`*o>RO2@JE)Z~${ws#g=VCqYL(<N-j~4E9_UB70_;T?3b2ggG{3mM_ExH5jDNC{
zv(+V_WFMX8Z%?qhqQWIn0Oiz-BEWogt5n?h`ZZV_2Xx!_lyO|oSE-||<Vyz;(9Yq)
z^FQ;k@N#=4Q#r~a>;#2aicTx2!8=G{64tpIN+T47Ax5=UGh6kLwyoh$;=3Ce+g9}h
zCk03~wU(?ofF-2P$rL%83sm=cEPsXKc+lU_V|$7*O2w?NM4*%j-|n+L?Mpt}zrBUe
z_7pGqEW>t_#5`wyvl(6~!p|7ipny9-KvJ-xEqWaYXvqW(Oq8(V@FZWZ=1r4=WSAHT
znCxt<57lN-KH{fNAdiFdH&?qM2%r;csN_W5i~~Um0%tS!VQa6tjJX-16Mq#6x-i<Q
z_VyehGo)4);N}=hy|4)n*<P<?<2){j<SaC&{wx^NV>*p#Qi!r{n_Q|e*qL#baiKkU
zKA;cVO;hNRg<DMtKjngsOo|0?XLTps<DW~XMC<t}j;26A<xfwWZz)m>)M3jlrKfmD
z3`=!w+L#%|g&0=bwCwG!zkl$S5m#Ysbo-Uu*48O|d#y_&ld2CcwXm2+)SWvC2BBP7
zDHldJ#)({Sw38s6zoc^$-fY=PaGuBD1#KW~(%79ha4hsnXx#F2#VBszSd8KY$Knz_
z|6s>j$8B{i_V&h(b^B=p9SH5z_WRqAYwFwgx3sqn-UIuY<Z8bDLx0ANo`%U%yiKP?
zGO&!u+%M<qvb7$CMxg8(#kH24#I!SoQUcfu6P(4lt5z7Fa=0ZD4o2^XT66A~l)|BQ
zZ1X6m^h?*dln-RTEq_s`p~sW&o%m6Aq?9Mey|P+Ye;(aJ`7r85<r!gt>)}?4ABwK6
zytsgO<vdxY-qbS9(0>8cr);6;VN^gtnY7q5^L(;ZbaV$uT1!14tZUs?@>Hk=t@wkk
ztSL>QEerv?1E(};UrGQyj(iv<EyAk<emP__4rVTD-F1a6?yDa8Dr%{wzi@3W-iOqd
zQ@QJUn2KAyva7F$oxAlE+eSb`W18*O6U7}%xjgzF7^}#rLw~9@1f665<Ilf*|MJ^^
ze)-pLXD>{}Vg_v_On_u2tM9J4{rIZU^6j)@g2!o;^5Z5_+#2Ml;_*C+6UWi@ku+<i
zF)v@=2xPdP%fFiHLfc4#2ug9PE88%!=#OvLhN0Zd<Kz}C5UZ9p!ynOw0gJ*p|HfSy
zzRKNQ7)J09(tm}C*A@MyT^L4j!!8V?e4(Pt-rk_1%P3x`=<@a7t_!o0uhNB?!Q~CR
zFkENbbYU3P4Z1LZ@yh8Rp$h}ji?C@IhEX&xH=O9T@Y-=oOCDNI^z7|>Dhee8_XL*;
z3N=75AX&@BT&TC#e0e6o!M@d(ukvRXgo6o+VCVhWBY&lOz~7D)1NgYv{@~MZ8yZ_r
zaDysRPw+w!iHUPVmu<P;RMmce&9fT9D$%MPFkoko=$WIKYT#Py&S7;k(c*<@#Fbj3
zG*=kUwKjiny}k{oqyU(3GVg3!QL1jaM&y09I7dNCZrW<fWyW^lZ=0@MWaCipykFIo
zJ8OT3mKookGr>(emUEWVx760u8rhRhQU3>h{*~XC!ALh00y8j|fsZv5mm6v)CzsG=
zEC`oVr8kuU50~ksH#dI@g%uJlXGsNY1+dSW*m=w56T4xOFw^|TJHA4G)+&h2`;DRm
zA+LGUKeu>*+OFD$Q;|a;_B@ofyuNn{e4niH2e6dn+%@Jjsv5skbjh`Jc_)6oeZ2VV
z;s`Q`44JKd|8kKY5Lo4WtPoWMs{Z~^eZ2TZN^K)3u5~ij0n&fYSz89kDVtnEs|@t`
z7T7ca(~L7EOp|juKZoX7eJ{P&I!i3R`>Io?xbrXwaO8CIBrSk9=_3-8VbMw_8E6}+
z#hp44qaO7$1{H#lTKZfmG@U{VQ^}H(QuN(eh&x#)+CYx8jiA804*8tX6xyt^O>tb;
z3F>o#az#YegN}a+nbs=cTSJ+bAFrAv-j6PES^?Q|iN2cD;#dp1a>=K531h6HH=%^z
zvlbnA2`L{N;4c68{N=;fe?I^D^=i84K%oy3r#c(kHqur*<xGUZ^>v3MW(?Ea`&6Xe
zlLGT_lcm~53cc>7>AT?@6uV#0TasR@b%eaO?Ndoc9BO}ASWTyzD`8N;eMlg)Sn4?5
zyj*z^-J7nBy}IXSF_q2UaDcl;pSB)*n=#P2Dc<+Zer;1E%Pw0dn<8jZj?-(pD>i+i
z{<66#6?YzTr##U?a;LNwIhCnH`%SZr)n45-o+608)oQ5!>J_(lhTYIQWlTKMa@XV6
z=H|kl1;KxDwMv28HqgVh)2Wnd2`>G@EHlZ$wjIvnJqxzsz7gCCuusC!0$WO~LB(7&
zwdD01Sb%y$5t31l&og<u6K@}3w_HDJqkv~CArW9yKk__fxC2Z5p>I+xy4~;4O-8Hi
zA}&r2#zMP@cN*iya4F-%zu#tCo<0tsw~E{m3Tc08t~Fp^9UL&8k{9Ke(N-ZDDK*Vb
zX}jvi1Z%4PtC_2ZcZRvu)&UQLi2gWRn#k4T>?l%GJ|)Y>8r7zXrO5fE<Bk<8B--2c
z73z3iCod(Ps!-m-_<SSZ-cX^r1q*>E3ZZQf5=hFSO&`pY1P!K?^xU?)Yd^7%G`Q5c
z_4<E4hdwF~M<{|DujgQR3<UYMW>ewqZu<jq9h(SyW?Lz7OVnND9T%m<Q(<qjPIx53
zUW~l1spdKb2BUAfHy!#@8}<{DVUcrPG6-;EHmr3dPc{adfx|4#@us=%=A20-XqHCK
z@R`PjqsWaPp;z0&L016Bwv1ODAt~dI!jgZgZo}?F+&{_eO1_}$Ntc;aZbi0foT8mJ
zwsA8Wk!xt6*(5x4qwZ85-nT>X{@hffp&cWD`e1e&_xOQle<F`~w86NA$Rj?_^HFV2
zB5!c`=EH3rt&?Z(d6JL3A_}FQ=OXzG*D1#5BKZuTRIKkiRDs(ZXj24WgZa}rKumv4
zj%^kO?#uxq2IU+{Eo(VI0!2KE3n`m8K$CLLOmTpVLnRrkj@zQb%TwGQvPE8(7Ih3F
zADBj$1SNd9{CAQ-MB_Q!ST4bqrlip&#3}5cT=Fo<Iih7(c%m|v3)#Xr=QvJ^a!D$P
zq15)%BnsRJls9G#3_KRI7Alv#*JytyZ8MN2uX52eUu$?Ma8bbhX_9p1=|{nFV)mq`
z|7N78i7KS>D=phfdIsb}7f7iKNl(L@B5OMr>6r)~vi^wl<g4Xb?>R_MqXW;C=FFt0
z)UuiM6av2!(i7QuBD>?ER!H2D^c0e3NKf(h4y31$I750${pTY+7uuCc&k=uI&PaO7
znC(M)3aLAgp0&N<dmug2sF6DLjul%<+(3E?Ni(FUczZkPDFn`to)X9xLwZU#rb$on
z_Rgdy8c@#eNqP#2JCL42;tc62J$m7!=Q8KIWDwxKaME*`b1ms9om)*QtS3Zy&kWHV
z*?EExwWNTKYGz1?_H5bu9SVPSeX{zW7@Z_W6U-fS#I4cu;K+yHbYA|JJm4g!`s$vb
zvRuKQ11-9+=L7c8fy*EBX_{z!iRUFyF8Sf=+c3DmQ<dGs;MTjw`54@ssW>>mKaIgH
z>^v8P3)}+ivoW~o8a$4+TQdjRb1=BVz%wzpLdz2wTp?u>gBy`6dntc`I~l3X1TF`{
z`Jq9x{H?9G$l;1P{?;g`3#Vu4Tcb3TI_B70P6jYl<QlfPj=W7<y(@H0bGI11g17>J
zbmnzz)o{gojWk2t@`1>e7|ap3z>z5po8fI!fuIxV_#$Xqr!bH(Z1aCr4Jx24zJ;W1
z`WJG$_&=(OR?S<%e`f<^86TIyNH-J#GnXOA0uuu=F*TRLIyNYm>8v*kI0#F1t8Tp<
zg@8qi0lcyW?rZ}FY)g;@If#>mKR>6st9zzyzd21s0lb(QPEXf;Kl@0!dDrOXAD1kx
zH(7t^%ex;QzrNB%bgh!lf94PQ?e*i;zpe&Fj8)F3#(7rEvH9uaRsMm=DIZcp_;fxr
zzyH#_z4}&?9hhmH>E4`fet*}9D{qP|lZGv|_s&(eG+sFu8}E~{`KQj7{FT-YTJr;5
z_jtihhJ=<qhxV@%<NgX54_usA)}|2`P=<ebwSxzo!3$qpsJOg%h%_m8^ooom2s*8(
z4o+b>$C^SsheBpX<+s=|<J8^HK3}`L^*RSdOTp)FrqIlE5F)K$8|=MmYz~TT*kSBh
zGxUb7DYiz1%v^Nkhax*?A^W#h_Ug*H%ObT&CMX+S$u-=aw6cNa-6Pv==Wn!C1WJE`
zBGDv&gOXsA8S#(B-}fi$4&P(z<=t{Ik`vC@C>JM{te>(=?H_!9U1UK)DA8nGCrfo@
zA-+s-iv7q6xs<gW%qctIFqE!MF=n+>hOMNEl+hYFrhDHy+nnB7xvMMZvDN2+^@>9r
zM@#29qB%<~>69g#DpZ$cEHO&)kqLiW5p=Ki=O=nb<RA#mrDw1xlV03>L3Z}#x1u+b
z14C`7ZOFx``IXCbDaCDBE;eOXY}}@57j6@!E!S+lzrvZ2R3K?V^3#bNtxAE!%^HHz
z>%;njOGbHQ+HqkfC|=@mbXw9zQk?U72{@^daC+{$ZR4lT!8wcD0N_%qOTT|!cKlqU
z-X99vl<Q+b8aEZedPT}alV!?-hTR(BT~R$uIh*s(ZuOi^2=F-aTpoW1Pxa%PuRE$s
z=Zfr>lhG?<vLiXC6OtV(;!)1zjO;toQ2+OLCWq<AM?Pk*>^hlgW4(3jTDbQl`%upj
zk(=JM7#+zSlB@^n-7p3(#cF?7atG5fmYLuhP*lOVBTZKaU#IU%{Poib@hBc%JTP(>
z864nfQH6u&N`dRY-sGb=!qZ0rj!=E%FE<Oz;3*~}W*l5bl|;7X84ty^i_NKD-VC*V
zybkwUq^(AZzO?ps?N*%0VqNxWhu&uDA@wlN<;_mFo*RNYor1piqPu_Z^Q2tw1fF{k
zU_82<3qtTOAADZsLslW&k>i&g;VgS6S1C<9w{#935#Z&PZj4IAD@(fgY!~vG&Rr@H
zqsbUIuD`D9g!`=6<w9Fdy2g@8w0@(rG-_|(-#cwkX!lZJ`YOLfA--U>DQlPDa*D%E
zW07%DF*`|RSYf7oH~W7Sst*gSMLM3An363+=wctk4(JUk%lT=B?mkwhuYWAX#$!@P
zsM%OY24mu<%+M*9y54k6H%kdZ4tjbiBI#nZHlFU<OhIy%oZ+ebLv@%?0(YT*Qmw?>
zB2fs_M<6^lhw8}pV1Kh@Eu)jNG1RC2^S7H@^8WgdWrlDQX+M8r*c)y}GlPMUg?miT
zh$oL&49HH%_@?J#r#xyn*Dn~up?&D$(VyRc{N=ZQzyIfN%W6e*9t1(f?Tck2Cd}j}
z2_r6>AthueOXC@Lb8AAl{;<f;qCip-(ecM3KeQMXVWaBEhsCV0#-VwI>8zNxokhjc
zL2&tdIYQHd;Z}bqba~k&I-vvwzFYFMMW3QJPtIg~zibJtTPy?c*JXPLqYSwjdru1%
zf=W8=5)6~E5jExd*LRC$$R<)8M%HV$xir+BaC#uT(3<9^AVzh}IV3%_mFl$Qj5IUC
z4X52{)Cx>40e-d>%*=0nE2vLjC^<z3dyJ^ITXgK#kHLTTu6zF7K_K1GH+(B2fA>^;
zXti_N7-tkkEBO*HPjU{?Nuxu(ikOF%IxyvsJk11UV`>F1=t8H&)8;;d>KqM^&>7CN
zg#9GwjEzxw9iJJUkz6iDX9U6j6Lcn5Yq<Tmhj2B)1F@N!@Cr0j-+TZvQxjYPW(4==
z!)3NfRbYRau}z*4l@Z=L29&9Z9)QVMhrV$F)jwlQ##)Pg=*Xw#f?{9no`A{-N-MC8
zxchQkMi5*9W~3o^A~Vwu=PwIFW6<t-DhP)Mm<tk3JwNRp26j)??~ji~Y{KbKL|;(6
z03j5V)?%@~`+!)i39iMWFyv>05=u3eVM1~D!8m_mmI;;hz-SW`55Nfp#TA@TYV?`o
zgxi$sgFzbiAe_*?Zd+7V4zX2~YZo6~aIlAdX1+k>yj*$g)!S6g%TXLh=;cqZ_sfml
zBk!S7?l3KPGK%d;?t{=o<0De<PcU&mm}p!;HKET6lDY_7JXdHVxL6(eUxACp)Z@s0
zbTNN}1->3~pTU*(%uRccL^diI5<)%~l1M&xJ^P7}#N3g+nXSe17CT1`r=CNefcPh%
zhR``l(T>Ip4TI~5%U;Z|;G2TVMVKLB<6`fgh#8)r9x%fYmoP)JgdaCShMbg%QHppN
zW=P5Bf2C0Aum%~Da;U1+kqwZcQ%J1()JA{EkTAkpH)6O389Eh$&}<DdbeK`Zku}KB
zC>ygeG9a-JF~pW{Y|U&1F^u_N*NcE#LkwXB$z?lYm>=ZVXIB72@@>tmj(k{532R8j
z<R(*M1~1I0gcWl2cp>s}5@JNL;Mk^Y-!`wmS;pd=R#@4=1+_BxO<0_RKv87I3P@gv
zh0k!BvGELg2}7}|47LVdI+SL?WDO$>4%;uwx^pZ>Xrc=6hTw52Mi>ya!VB{lVf&wY
z5KK?Q2*v4xFhW7~lVOB{?>3B39@>Txri|aGtq`FYvM|K`7Zke`8kfOHHxvRhGM6!P
z9uomGmtkEwCx7i*S&tjX5q{raF?rhsvaLR<FIylGt+WCBWXt>J7zking5t<=oB+|!
z&)3Hc*)=DZvyuh)flG1NT~%Fo^+(C>Z6~|GUcGp@dUkK9^CDQ~b@%Y5Q}{ySuabD^
zZ0D5{&ZzF;efQmSDYe7v1zsejF?f0WJK)_OZjB09p?_R`sl%-&_58S2-}P&ZExUB=
zch`4SIRvcq!OKJs+>}r6hwmT0x{_6NDWXgNOmF<ttB0$9UkwM*Mpzd+Ym~4!;}7qz
z(ifzxa6Wd5f*>;8Z$EW!uD;=92WAp`${#Dc-`;lY$?K}iD4|Q~oVATEwG-Bc&bcT|
z`l{%XUVlnyaca89>kcpINh@~Bo<siEk#YY7j0Y~(31i}j3n)Xq(!v5(<Ao2_H(Z|G
z2Q?{oa6)NL5OlJr7Dgc$$LLBtxk6&5>UW_qqgd9HOV19sP9~?w6kPhGE6qd)E|LY?
zpzl>>lNV(C8OokjL$BDDVpb|-rg~Poud-7TvVZ@RWpAFGnk=XsWQ?+wja<X(D24IF
zSC4eJoxhewC{SV)m73(QkrGVQBmSY<`|haS;d97VK3gtAaKacH<)SE}aZ`F}?Ss#+
zsw_wd)tj^}vNTT?{7ZW)$c?m+Qd#DpkI?~xA$4u?F(a+eNJES&LPko&nCxw9O?Uj0
z<$rFTocdND2gV8VVeBo%a!So<YUxZ#vZ+LMS;Q(rrrtAP8-jh(J10IJi9ntanhQ^F
zkR}~ld_i>f>9?Ub69YqTXmv={sOgo`bWO!MEmxhAtvU|H)$D{bT;hI6G05>#?w4I}
zMWVi-^wQWDMM_17R7HBhcJxAvoa_Sa=zoTcqJ8mbKV+0v9FeWQ+!GnW6<6enJkuAg
zoAbwF`SYu-_EC5>(xP(HRV5e^z2d87a@<$Dbt%?DCbwtOV-zp3I{H3^&}9_o%-(w)
zI}tN`>H>2cs)~tIP021mwR5Q5<Fe`J+QvsX!=}Rx&cKu#t6=>>YJZ(%N^P&%=zrtp
zRW48IaPX~qxRmNh7LO}OR+k-_(kLfs<B`*V6viH!Mzmp!sIue)9m?otPd>}@GaWf6
zq8cfhby34=uY^MX%vNWO@yc+1>9QCNgY^5%J*C_g4P`^G1;AA0OHJ9d7=`2+lB~P#
zVHnevytky`L3M;-+ruEnqhN(SRe#qyUKVc2_xjFR(a0B`-4h%P@$6w_mXp1sM)>PL
z-=wYBLepD3wy1W?pKcZ|Dvtgb(0}7Ga%OOO{y?fNemV8c>ml=xuWI@m4ylE4L#t(b
zjr2+_^Y>)#u9rO&G~)58&DNUgQ1m-}3#D}smY7_Bxp(Oz$tMYsG%cmW-hUO*3fr1?
z&b%R_G<DMGhDd8<BwD#3V(<AfXr@lr>_S$7ZkWrf<6N4<{INBhaZ@kTi)F~i&&0}m
z;gGc%mkVd&{=R`xJ=Q|<$Lp=^gEgh*XIXdWw*{7jichRA?KIYoxEw=jn|kU+1Xs`9
zl=ly`W^Q}?5Rg&W<FHoaet!s3aEl<Qkt9zy6OQh}>E`Rd*R<(Ohl1`+X#IsS&m=X=
zAmHg`%da;}nLtq#q=`wWWK~&?oI}nD$TBrM&{U#9nwvU9bTEl+&hVxR<o(nREzwQk
ztJ58S1M{0DRcUDv{HodX-!E@&m2=l0mf5K&^oX0fSvr)~C|Q`>PJcJ)o(oOxFjg4d
z%J+KFl$x}Ft&lC#Rbn;ekMG|9^vi$V{p*+IBc-O=Hcof_#j^SWoFJN3eak9{rHgI#
zrft6|Ol>j}qZGOr%25%9T}5}pDxBD(D3O9%AXHxU^`f%imN0*^q7=T~s9(QZJb-}>
zLueiIAB##dsF}{TR)7DnES4c*$hG0PA{H7R=ghQN(@c;h$h3a)>#`Y8kEI!#!Sr4Q
z*AX!&(<BOGVV30d8ksy4;Oe@>NmKRHI_nB`O+Hoq@4lDakNEPd)O=x-{$F1DzB9t%
zEVBD&_bobrT-zb(4DU=+rK)S#A*U5-gNW2B^6g?KG#HUcvwuXGOKcnx-?qXy*H9+$
zoi&WbW}?$#qg0`?dY-^lJ3rJqWC9AFq}f6=(+Jl>#CBgOxC%DMIis2{Cqh5N4Vt$F
z<SIC4X?liFb#5xq*HGu@PNOXp77XXBXt`syX@DMW>W_$2xKRELx<!3*(Ot*REO3Po
z%OM`uU~8Bjm4EJ2La<14s5wuJz%nKm1F$W<&jf!>g-aT}<IN_cxFi0`D6inJZ1oQK
zE2FrAzjFRhhre!=tKzQ{dU-<pmD_9|{FPDN0e>}$g!JY!z+ZJl!8z;aDQgkSC|-cT
zGD<7R7h8Qf^2G?QAYWXOpB{zPNOTx+KOYoU+cfTmJ%4?%rFa1f%P6h+V!L_=Uu+4k
z`65%~?*@hC$5=*T+3KB9SdD4cIJKv@wG?+iVHw306qcXpcaFktJKW$5Ou3&W3cKyd
zJqpW@yowf_FIT$xrV*!^xFn02ZS0kg$Ok>*%lR%NzMStqh_A*wwLez+1M%I4_-Z(6
z8sa}Dbboje%6lsCMwGYN@~Kc>g_IeszX0V`E)~apP32B}>*(#tlTEcCosc*O0e0#C
zA26Bc5#Zv&7l6R&T^sk2Kwzb#u<B@vMvdp_u!O<!rgPB49+;@n8DM0>?hT5P5~Mxn
zaxN%Z@LNXYB1|-5e%F5DD*D!FS0ixBXi6S#ReyZmf|ExJ_S+dqiKl5@ZoG8AE<Mrr
z72YgQbz|U{X_cL&R`)DIYf%YI1E;4QKwR|QQ^;SmLaPq5t~Cr?oC)O^b1bzAorQ2G
zv@=+89!-&$ys0_GGK_1i&3M%5j#U`fU>cFC*|QDBHC_v^Bj>gZ;~Jk36W?rFhH*`R
ze1EYOa2dum(Fq;E!{?vCoPpv}`sq_>RlbbkQf(O;^(|@uJqXTYncB3yxz*8J|Gdo2
zS>dTw`G;k0=rIXOINkB{VknSz`Ns)Uu`t6UND-oGcr3$y6cS3JSo93IUx}8>aKAp|
z1udlC3-?1%2!&MK5}m>Q90{Wz3;VU3cN#+dTB08f^lRB}!~EDDhGW%tG(G<ZZHvF4
zm%&Ik6qhlR4i%Tyc{eGS|GqZ|5BpbaZ#Gaf9CG+-0|AL*mleM^EMQ(dzq@#Ht8s9`
z8*pm4douw2An6?>J+x+U0K#e*?%of-Je5*edOfEXN~txy9R2O-(`>Fac%lGXUa+}#
z2v2vkYFlZv$*WYTtIH=837#mOcQW!LX+Tn7yZP<zrwf;$zc(y@?&9wY&fb6~w}j#M
z_ZRU81S6~q1E3d(dieNycysYHXIoU4LeiT2X5{ek?ZBSADv=42ER@z7(^_bSA>*w9
zfl5^i@da7si&fifse;C5I|=Epo#6U}1dmdT6`Bm>Qnow)8AUJpVjPQ%hpyh<3Q(LO
z@yS7<5F-UhtHL~g0xJrG@|(}XFl2hPDR|dbMu;U2HhxouXXK*_WRCFnqO(;XN<6{Y
z2i;H|+j2}!C7rR@inm2~j7rwGq<iyZKV%+K$}-MMHhPWJK??1VQ%`(%ynH3K07^N=
z5uhtyQGwAxP2~F$@9oaJ<9DB;e6n6h$w|RDQZEc5XxsCDOB)~hep7Uzf>6+;OxC4&
zGJ`MWj6gf_LM)6)g4$&V1&j*eQ5e(G2t{Rxaix@zl5$Me*cv_T-jd$UlYMN}JOD2Y
znkiax<NywNYR;ypS+7u?m9ZvRVU=GoV;RR<_D;!Ai^8MhoDOBE(?T1|sn5vJQi2)7
z1Bp-z1a~BVme9smEY_tUr{Y{3i>B_b8E~cpi8DZGJS8+`SO<g&fD0PcZKY}x`+}$?
z;3n=Ra=ShjtrpI}npkW(n&<b&V#{d>i!JpESZo<CVUaoVL@XM62#-1Pr|-GS1>s=g
zMX>$7#cL59aB3NWd$U@vVvza(t=B|b%kiA3+aN%HD$emx*emLoMJ(*G7fj74<PgW5
zB|!j7|4@3!`OqI~9`6B$ql_EDz?>V3V%3n+Wk;Q2^Q(`<xV}kSR5XsTRg6nVAej-D
zO@{MK#&|rW4&ykJ<1xuk5vuK;$!b@ebuk_H`hT@gleU-Bb&if#6I8R_YDet_c+IJk
z2l-BaagVJ`OW_@^a~jIqU-nQ6{8B>ncEZ@I_0`BMPtar;%A4`<Hizb1mPgki9l@QU
zBRDncVAgb93&hcd$6~Se*MJjjJh?^6bjo)}L1#7FS!|5C{QfE~r6t&3;%Eu9OTN2m
zk=Va?plv{VdQ3Id+jL5yQk-$x`{mWRGaS5sW{J79--lF>R43cr6$`KJ<$?)mFGS^5
z4WlVqL^>Q{n<DGZDL0!Om;Q~J3P{yIFoT}m+IZ{bqG?o(4Mt_Q<pRp@D&?scP)NW5
z>-YtfQiKUyxPM}w>8|NO15nBoYL)J_YLFTSpX1#eoTi=^l5)|MNJ|+@&l2tFTeDSv
zgz3Nyffw<`s`xplc@RoY+_l5)QC(MU3>cU3z|@QB!E4+C7fpvaYEQ-hmh6f*Xe!tk
z1G7<MEGA)%RHo#2FSsB#*AVJ*vq^a0MY;;XUN%t;ph9L9Rl$8HprwbVQT*~vDU8l_
zmlzNMZM)=GLQz>|Reu%rBnsM9pMJc5x&~`6Kc`Q3`DI=(^?laCL@-><T1W>V4Z2;W
zsvztx2}U@j+9j{56g`m3oc8Q0C{_eVOM>trHyW3JynFxp!$0r-`Y>yjMx_0q#m?mq
zSJ%q9QO~|lwV|`6W5N6MdDb1;lJ>OYF()dM1^ziN(8p9^5%@AMFZOks=~$(IIN$YU
z&;r1=zRaGOl4#+jZG(D1_uy#zWw0&y`&95Z?>dZ^ax!#WHy2)1K@d&5O3n+?NM+Eq
zOWt11?TQo?LOJpRl#XCwtP95#{6ozM(#{*pE|YOI*S_XPQ)^}v=UO)GNr81<XtbRI
zJEu8yio#mYSD&dlcq$We!ZjU#eSJL<)UG6T+rCcba@ilhl`cGwSH=zOkTuGvbW~z;
z!OZtfhEfeCi>it9V?OFO1Oh(Pm@j^6si$3*5n$|pGUn5o+Wp`kc+AJ8oIU1a48M^v
zpN<z4e$9(o#&N|lALG0*=3~?=jQJSHg)twO|8QfzBlRj{zK&lWaLmVlW46wik8xdL
z%%>%a&<^Pz>6kAbmgK5~dB*XKF(2c!Fj`^M=Z{tx!-dfbcjVU}^C_o<?9ZAX=a^3!
zNq%>0VzK3D{Qa@ma$3S-OT7XXTZT(mWR83USkys)^{ZpiJQx<i#XV_7C^ihw7&y0#
zmVnq0uYkjb-x3U4A3mCYVKXyge%9O)uRLmoU<e#*j+z<A6-LdB<HD$!TXgSHb9+lr
zxqD2=I(QnGPP^(#`b1bU6BxsVRN&%Q8&Z$cnY|&kuis!gvhMoEDuKqX=$n?|l5d#!
zha8TaUNgKGN=<wO{h^!+MCCbg&-FE3<A3u8$7d90OvK}rn9=@M|B3kN6LE<D?&-oe
zF-on+)cg-Brw`2cZ9Q^e-YofA1M|?-(R<9xspiQ5tER`Z@_Sq>GXVWwifl8vmV1}M
zaN20xfsh_s-H)A&vuX9s-NnE1#67E*!ALh00W+6D1~3zsz{WSne}v!2?!JBZ<(1B&
zYn8m*%18U{&AY3AUCj$IRym&r?}8#t!(TpL$q!6U`H%)<R3aX}|J(5X>Z_J&;njSO
zutfrF+I#0J*o;@s#ld^4YzQ-KueD|@!WmpT-vc^_FZ_TDi?;bW11^_=7{}A}TmuLW
zJ3&;Y%w+Jh^ZD60e|*A8F(jJ*eE(r+o;<ysAu>#gO=yLtFXU=a2a;N8p<&!awR!@o
zcaKRgL1Q?nWc?8|4|xrt;&cjN;|(~bmWqDJt6ti_lR92h!$cKLDzrqWS-=)8bfT}k
zx_vS-z1sgeZQ?w?+R9PVyGIxo0P)6iwLYtrvz18K?{B6Ve^~C~45u7XJ>}1kB+v3@
zc~(+3x_Z_Rf4aFP@2@}g89EC~az_m5rn@i?wR2q0GKCQ@P6@ywG%I^5TmX_vLDo=B
z2`YFfnKg>pO;f%rT;k%CC>llel;2Yyg5*_jrk3oFUBIn|r%4TJpSqA)GKxI6KpWSU
z%|fefjP<O(e==AMwT&TMXI=QAFCx?^h9e2NOyYr6iexLhvrwSu1#tpfUvJV0KF+8B
zKkcFH)=Xx9{m*eq<?XE}h43?ubH<Ce|K?jg{1rcb3AbQV9R4#7{W!1(OaZt57`}$<
zAD#LMcnZ;_ty@+qEK7}=8<e$Kr>MUNffR*w5tHBnf4;4`m_Clt!dKgN!!%8sUGBnk
z*OeD0VB!f-mW`7&K^bas``gaQLbw(8L!Pqi)EP$8IyMOqk8s9boUwad+`BmNof!ja
zamr)RIi|_1f<B_jVKo%rdOfB<TJFrN@=WgZn{r27?hK7%MV6x=MY7pE2WPkIp+Jxn
zPU>hDf8|+7Q91$9B?%5@4$%_oa!}*Pygqp*2_27^QeJU5z_K57T_hD3V=i0D;O?20
zr^1Aje3RVJ9nU(7)Vo`HbsViTwI*6;8g;ZTPi4`*AG;G0A#=F0wa}d$tdXi2>ybM_
z34^O9{X0ABAN!GkSW&VqyMElZ$dIGvYEnOJe_IPSvBsevwJj1*<olQbudV$#QQM09
z9>G4CvJ$n$k9$)uWhH3)q+m)l)_u4_f_ldW<2fLn5G)y`T`ky3#7@YKZLI{24xI3)
zIL}$Q?MA+SO2M$%7|aFov3N>n8~M6|S&Mkm+JatT1I589@`C+ghqF%YeyW6JnGFsb
zf63Kof7eZMI3Qoi>S<kQro=@iNC2ELffpf{PNp~vA>dTUaX57JQXH7BgIslrv&_5@
zjqv<_^8D9d^`klXC!}I7M3r=&^7lTD@OP>4B|mnVz(cEGMUP}<zG1Ss!%j1dsp8Sk
z7@SV9C2As)Bn&*A=`K$g<{;<g%<DdTf5K3n_mcvH#LFSyZlpRlDR6Q=vU5@(^55s!
z3E`a0$w6E?iLcjZkVvS++}39ZuF>GLU!a-l591$~iyhUIr(CpIu1u7ojgzytpI69V
z<P0RIJhD<lO+BP=81MFs%aemD3pqqAn~u0V?C~3Qz@1$Sq398vT7qL8iTBBoe>ItS
zI4rglb(VB0Cs}?<cP0fKDfpsTLR09NQf67ldlhUQm1lPW*e78qz}6H;P)#<vk@Cfb
zF%pjeFVj%Y8G4ax9xV<TDN!@}2)hdkbJ{Ahp`}Q)$lu_wgiPXye=b&ys2*hR<M%ix
zwU-qPVVn}H&yy<`{uXel_XpqKe`Hx+P-YAc0saUDDK(EA%-%Y%VVce}HLIO6aDEL{
zn9Nw0oy(6%dG>!sxvM9Kx|QvKk;(b!kF{mmIigcY_0}YkEjg+eSu6wDfb?NIi)|<t
zw|9w!dRCD`OX-w_28)!*x1(QRq1gpD1Wy=3aS$z#oVQjkan{)(&*xS7f8CYnIw_o6
zP>Gi)p}?rqk`j&zxFS&rD@7W~3q(rfDRfYr-%=MDr_yA$)Ww>jyMMke)|A$Cu^zpF
zF4hFsbg=>QLS2;Xby**$oxeQhBH2@cj!-1O99tqzW`Ww#S`_!qX1NNG02kg;+iHq0
z*>x#V;P};0Jhkmb>zWlcf7^ao0+TJC%V(VPhnz#|`?^+Hs1x9&P9`>MMmy6coW-2q
z&KeqmE7s7mPg&psLC(*unbKx^rghjiyHbRt`4BfAp7f-vr_hK|=@d3Zqb9h9Mm>6)
zCidkH!F!buwp!W4vO@@@jE?Q-C6Kcz+7OUr$F`F^2dm4YsWgA2f0eTUnm}d0$$@N&
zz!Q#>7OmLiQZSZmaaoL6G0&A`-6nSP`LSmf&1SGwF1W!u<k@CZ<SMP5WP66<mQ0`c
zNJZ`*8Al<w(}2BiqDSz2xTfnr<VqCt@(k%-Q#IpvH=WmB_DULYT9E3Wle2`JmJkvq
zWnKE^>FsLi3Ro*2_W4l3qkll{_obfJedJe%tcEGtP4}{>v#qRQE~SM7eS2*8uwNRJ
z)_|==e*Gw~bx;`FQ#*1j>>g&LnS^>?vCDX;i2X10GLra2-dk0T;w2C^hu!DAYh!HH
zkgVRd8KFj^H+JkupHna0_Xnpp%RpzOa^Ct1&r9CV@lge)Y8`7`K7T~>0IjN~thf0P
zmK*|DHS1IE@%(giYyH&l3D-aOdrR6VXj;pBrRk@qq_Uc|()FV_zJ^pydFuD@49>wC
zvN2+QF4f!qVk|6#8z@&=cs42y4URwMDYU$Lb_zbTRaGcR5C_J)ev=V?Q(wB<7H7vJ
zDf_Qs!E}&iG=Sf&ynkZ?x8uw*7ALrjjTgRDf1<P}kF;*3%_dDv*NXS^%`Nlwf!-5e
z?^e2%4!K0bT(2}A_2elJr>wkZB5xSkBZqcoGsRH-hQ-Fslw{yL+)*zDpU#J0q?ban
zm)d;hUP{a5#l4h<;BTatLa=KWx2#Vz6gTXpG?Z6*Db45&dVeVm#g$%4%m4HBQqGhs
z_fi&c`Ha1kcF(ryr8HDG=%q*n)*JU5=%o<L`ypXVi>)cXpnly@TB%<*qhDU5Y6z~>
zs9GR@HTCOu8U6ZoGkW9tHQ_A3uSad6Y!$%`s@FBamEv`?oM+drd)t00SU*D%&RN>5
z2Q{!8)t}wuR!KKonib=-XYLwK4{-DCuD!enu}P_xaL|qWe{cJLsrLL*?f4q9`PG#7
zUR0tzSj(9u+QS+Ds}e2ZwKaS}g|_{~+3~-=G0cjW!ALh00W+7uzAY4&xyL9dmw**J
z1eegtF$9+v(KnfY#CJbE{cs_Z>Nk&`T6$Z@0@`K0Azm`{FwmF5Q7ind`|s)jpOo%F
zl7TiH`sdm9*C=X+`^Km1k9VDE@$vh_O7(0y7Ma~ce!Mqq|4e7hlJmh<80{+~zI(LM
zfi9Zk?#M9yFb}=(UdP@;BG~Ap`0;)9?bXx8e=Y`BDJ_hD4Lhry(764#_ZRU7lM&X1
zol+tY?|yl+dwcQcnrm8x`m+nDZegUg#uN*qtS|;Pu?BlR1XSVFqdXcHM1oX4%=ux6
zn}vAK&X8Vc6Nb1+@xng4p0Ef}62*v03K=<4(xz)!u|8T-<Z~HAiDy*a-flH~pqdLP
zgnod<%5oWhP+m}J+~wP=p2E3Vil9YwQ9<%oSemG;ny13bvn{DU{ORhNtiAkrb*=4X
z_r+a)YD(9Nq3={WUA3%4kr5utTTrPih-XPy3dOo4d7ViRts*$>O-bSe$K*t149@f<
zx^0eO#Jnk`ylC5Ep%~Q;kZt#C>qLk-R=7Wr(7YUfXM|R268DHJ7waY8H4O_WRBR56
z5*=7@9gZK{9t=TPSq0WjQwmEK#=CiE%u8W{)5C#L$Alo!lUP;U^gAm=5&LKQomnBE
zXzF(wt2^%&V+6~J^wZU~8dCEEzBqq*+jf<=h|g+4WZO}Jkkr+Jyt`^`gd8lY(4;D-
zo$ls;5wYH=<p$Ad_f&an>G}Yf_RYtkb{@QMvc;$3dfy~`8tYazP9Z_l|1xs-%e!55
z<#pOjgKTZ4#(LDEY#_f(o6F;VyHHfLO?v^|UwgUv3giw#j1^jkTFArRd)@A1;{G((
zn@zPw%#f=19oL!NU0WG<<Bo%k-&Eom`Ka7~39LJN)zgemD8Jp+i(bfXEje~RRjH@w
zI49jXDw%(Ex))c*3+AcoeJjpN7J3a&BRe#9#p>ya@3zZV5<LUj)ruq0tb8S7+o>k<
z{ixqb-W|UCZj}$#i;<aFjE#CRYL~Wieks=n-(MwNPz_QvDbwpxTxrCY!v4#4;)QsB
zGU_C#eRg2MByGzyElt<fR9ulVWru9^ZvU&(ySOrrt(*rM)y?Q>wDc<%&q?#1YS|Pu
zn=4btWh~*Kfpt7##+Dq-{oxsES$K9er=<*a8oQpYpFSc(a|JFLE+oQGASy?4TH5#%
zi*+i<U2)Eb6qC2s{y-FiZl=I>C$__X>zr~yIGT8o?0i0ZjXs^~rzKRSS0~#aVsqoj
zpD@}gj>oLJ4T7YqIW8@Gin`jOYT3u9&$z@fo4XL;RI}5*-Cp{o4g2t5*k)V~hT6Fy
zDV7aMjZM@kHXM3ujQ7R$WX8qwmDXEVE}pB5pRC+jMlAbHWGLB(Zzhl`-01LsbBZBd
zr*Nm}T4ZLHIVwA!yjoJDXL*1o>oIjVbgr$i>g(g+I>w>H85}->@ZaoobKhzZPa@^!
zkx`A2h7KW6l~RWz<KDE(Kc>?m^E`nA0gVa%a8<_S{Mie3YLCk(b4Zuitf4yg5j%J1
zbz|I|&w`o-`(fN{vNy8B?f#g5S(n>1ux4Um)kz1|I<MPzL-NK)4;e&jEfmhOrE=#Q
zL~G|n@Ly{Xtv$oGeScp#YtB=tI=I94-?XDv5(#NTFRLQ@{F6jJ{nd7vREBterd;bl
zf*0QFno~|`ZkB~-oX$uvVypJf%ZYdD6IyMlJ974~*^9ulvwt=rV_OG*&i#>)w<BK-
z(UtEUGQ~F}RB(l!Yh!#glgjy$WPOZpr66qC1=Z&?#{V)N<Nw;0?v0RSCeqsjRfRT&
z6qi&hV|;0mB2<#BjPY?CAcTiD3X9L1#`v)kl`>wm2i+PGhz+FlV<yqFYH3ubTCF#q
z|NXC@-~9aFwlIe>;VK`0d};e24*yA)!iL4M^CUqi&n3win3%T;fnbQ*RxQlQk>je4
zAZ4zujWU=2Xpg#e=$ie4_}8{LM#OR@Z$%2Dq{u2RQm%Y%3$qA8xOr+^bMo$RhUZFN
zJD(X_18JI7@yp809*L#RfhrmqkoQX}&I&SAC43tgFi1r+sn}hAb<Oj_`?7D&H6HmD
z0uK~=)PeJ;q7SWAs>loNpLK|9TXWs8LoRk)5-F*5&2b~*p$XX;^tliZ3JUvz|1A-Z
z4q>+D%#cSd<!H>KV)#1&J>qbh*qv~wRdGBa=uvTA0X?dxPXKyU99KY(TK@S!kBxez
zmU_l7X9PX!H9HM|=uvS!0qB9fN|5)z1JFaU66&N0TQEEV@+cXtARgtz#{(WEzZJZr
z^5HiFcT`5S(T?)r6N4Q}#R=Id!SIUX3BZnu;|kbOTl9^C9ovi>Vxe;GHx71eGcJQ2
zwQ(E9c=GTn#@b<*pQxt$;!pFVUB;&B!-VQY;EBQ^OYcvA2>%C9wu2{ez{%Qt^;!Ft
zmf$GpWNg?*@T4I5t>6=ilMqEmz$bBO$Ftz`AY$v)rG2<JI6u2u*>k}uB&~puuY;x#
zDbzZ*ggKuRT}f?QrNODum4t~@Ql5dXC@e&%`J4+|F-UGU6W=tj6>CK>Q#+5vR-8j|
z;Q7}o>@%T%m7a1ATIuf`0<AC#C`#5y%~fQDP(NKku@0;lBq6JU=ozm^D6gRHMD_}-
z;)7sSti%ecqGHUWo{uS3K$VEYAQ0lbIfPA^3h8)G8H;!k|Nq<-&Rl>>JTliuVMf~V
z|MwT=P$L38ims3FNcZ2`KmH_lf5*Ka;!YL(?o;=F<a;~`A=tKd|J?lrwXv&u3!)QG
z$AZ|;!Wm0%f<R|pUb%T`mPEkrIQHr`X~A@Fb}hx|tTd)L3NnRTKTh3>929J~xRv9Z
zL|nRcK72CIv0K}4RD8(;y?H^gibloPJS=V<yn4)xgfL61w_p@|oY{)STd@gULY&-<
zDb{U&LSThH8J1#X3E?pvoK)Hr0k)WgZRpF$*gNWi_R|W?W8lG(VnqgdVI>t5sbYR;
zi_#=G%ox^M;o&tv7FzKZY>kvh15^+U1-+#ny8HgMp%LM_df0Tjw}$4_n-^njSM|pf
zZD=j_;z7DG#S~gg{pykMnZg+)R`62PG3Z$k95yObF{~yC`4i>re+nOQm+_!O6ap|a
zm*Ltb6qniBH^+auhaX?RztKhXH!m(u%5AK2J~ckDf^>u2;NZZCAqy_k4)X8|Ua{k<
zlFy9uukXaW=hvHm-$?e3y>jH6zkj^R|6p>;htv>$J0F@~{@J|0`HN&*MAu4#ru*S^
z^UH@ueDbb%CTVD>y?3tG5LZKa??|yRy@vcPw2DQmr%r!ULzT_O8`{6di~CP-@o<at
z%Gz{t%hPB-XYj%s7ldabP0Ag;A|n}sPv;Z_62n2(6y`aw@?}*17rV<ib?4c)&b#wE
zhgi$O=Xa*?%zO}Vopaaiy=rWeaFWk3_N*Ct!`2*Iqe5mby7EhrowJbrk5=~TlVg`f
zYExElHoAY3YdD>>vVrC46WeXS-)O4{DJM7*P2O)%U~Doc_s8Pzr_s8@``CK<Y`GZ8
z31@7Si<3&$PuZpR58gi)Sx^v4G#S^)Qhl-zUnV%k{=^EoFt#_CF+1QeP1{n<YNrfa
z$rUN1HF8Y%zIC=4f3$K}pB%?lp9d6`V|N-Yo#%gu<}9_OQ<iM1P+gU=E_vlm+%aPX
zNB4MshI&?!gXA=qp}{KaygYqHhL#d67|tX@Um&C*xwtmJa<MK2xh>Acu<ZKoniIAQ
zx`_gpE4Mzs&Y6%@pc5;SpU&rKRSG0eYluqEyY&YLxDja2V5=Bj<8z-BX(JfU`MeLg
zS{;Ak^4xpd=1-k_b0)X$=Tfgr&t7)>T(iC)5Voi`#6fM?RK)7lD3~Tol%a;*8r@x4
z-Ay%{lh1DSnoX=+lwLET1R;0g+8(mK9nq9I@?`!@pI_71u_BfkK(=cJ6elHS?^Pn{
zjGU8wZyGwm`8~>ctAe9_5N4-FXQ$S^Ye9eCv+M&tM@DXc(_(Zi&yZw2U>^oRycDxt
z*&R&B7;A!SoHHsUv#aT<Fl>d}5{c`Gp~o{VOX8EzdGo>uFywqdOo);lJXg9N{&<^7
zLBi7{0VGtD{Nc6^)ag+Vs@sT<QB{$l85p7#cVRsBpLauNn4N4fc2+Q?H(c)nKTdzO
zvbvX<p3w?z4Q;RWa-QIOu*W<^b(<I^V%rg%#>8>`HBZv}51lWjA54G^@;(YB-oE(!
za3N%M!UH+J?-;jpejF?3xpO|dq@a+T^Nm6I;*~L9{BB43nWkSV7kj7dn0=K^uQF)c
zmmW(|DkD6{v6!`=wZ|`*uDW@tmYsjT%`Z{PuNaeDRE)8szGMu}P0-2&QbtQF55v%u
z4`-i__2Gd53y5v-SjnLLV1iQ)M&}+(TaRb`m{0oqBX@+6YV_!EYAR8X((r*dC~C**
zS`;hoD;W>(OPvd0nj~OS4mHW&5G{oYgX$C&6O2JM)vtfLy(8}rzun$he`tT-gu|!h
z+kJ$<r?=DXqM8ArA&5+W4^L7gLd*$y-}NMzanS&5tV1V7MTiW}HZ6eVKwtIucsoU8
z$snv<P4x4k32YlzLYuD_tq+u_jIqM{izc(^_K}=4v~fLEoTPF=u`^I->_3M;FI&I|
z@Il5|U_|eZEpT!KB@iLf9r=H0(HY1x71}60BKd}vtpxNB+)=CiX<2wc@GLZ=`Fl^b
zWHfEq<bZ{(EDL>V+sB6`=W}k)!Odsiq?n8-OSv>!s|*XpxM7tdSyNNwJd15y8s=lg
zsvbCUCgr((*8Rhey7>qG`F+MMDK?+lqaT}yc8F5m{HysZ`eEDY=A?f`=%J>vaViub
z_1lWHunOf;i&4R_2#{$*B8z0<JFX*Unh4^yk{wYEMQqUnNC-+a9VpWoh+YQ!h1m?8
z(JO-yDdKW%X3t8KpVVqO{5d#`1pX|oB3u;dT&+-BnbM3_lNUK{l`zAH+a=KGbW}DP
znV5N0ckOpn?I`ltW88nfXP0zc#^X6ajUFGCtV68wrG?S`g938>=lPv}9zN&(FGJON
zUWUjyv6GdO6%4%g&6!O}p%u?PM(7{NXG~%__7kz`XM{z`hC3oDYuyxXvSf=DkEX`*
zX*&p}{Lb8u^U6rI1#K1nD`=~`L7=Uk_G$PUViuHONI@SsR&#$1dPRpUpqgYIdPVpO
z(_=+-0raw%DN*K6ZeNC88Lj9b)Lh118Lco9u_9T<Ue+eeToEMb4sIj#a#jWFE1DJL
z<w+$YEcx}a`hYI%h0)?N3<+qdN!;?&rh056fKr4RLQ;`8)#LnJ@WemIIB&XT==?M>
ziI`e0;e-J$nc08fokmMc85Me247Q>r^y`@7e8SB1YO5$Rs2s3)JXWs8KEON~mS)@-
z%_8;*fPAzG`}8~mJh=(_Scm9^{{pyay9xR*gwyWKL}vEz-%0vwy#hyodz_%6b)bdu
zTVu?t$hHBkz$!C4BLxsl3SDY81?C`$Rxw|6Otk{E0R4X>yGgBny}iSfdU#m8(@}1_
zGgnTdARuDQ-ZjYz4$4qI%8V;*wR#}jH)@{%oX)^Di-^g4RONUn)0xJJDJ2K|$n)?`
zws#mh(M;<Eu5}zL!k<^3j$mfV_CBKc3Rf9(phc@2r(ty`^xg``^1cjzwnZkw@M7HQ
z15O}p16hCn8;eZVM9hSrdyz@ra`hsUVE8*(WU|_~kFNLordn{^agj-IURh)kr*~Ln
z5*$|+ndJSSZ;@%ET;;+1gk3)4B9rvlK8s9(>kf-dCS`2(dst*L^(N~M87nxtc<~~W
z;Iy*HBu-zy$Rrr9EHX($p1H_0z0ESLeeAeQWSoD=+U@A@ilIxF`@3Sa=I-k74&JW#
ztvS24;TN;CCX`s-S*s84ys~C2YP#Mv%xaGA?d5S+b6N|u`t%NwRx?}+HDSmvC)R`-
zE5TNu-aFcKR3`3;I|aubRwM<-l^sc`(Zx%W_26B&Qv#|qZXQ*i?r`;#h+*sfe%T-x
zu4I3Mynd$zOZ(g{6~#6x%Y2kmt9v(mhoj3H;f!rEO4U0e5W#TGHsbod7kn<L6dh`p
zYP_AU6s4|+(^K{?IhyM^bCkO_S1fmuvx?LIcRSDD?Ibp^DSfe9f8$tp!7}HuavPU9
zE0X`!GAHS(T3)csi8_RSv#VF>=#iN2Hg+BTxJt<~L0-<=-}KPfg<9^d>ziVw^7-}V
zKMVu!m%(j26ap|amoc6J6ql%C3n-WU<Tsmt@{#BMuP;B`$*j5*!Ny<XkM^&3FL(dC
z>kUy_unjxQPLQPCk6-TM14AsV3p=GmAhY}W_uZ$vKeb$otmbEOn<uzUTFYFyO<BS0
zch+j5ohxpOhOxpI48=dGVI;#P28NmVyX&4AxXL_tPDtyiTO$vdR_&ul`4J!Eu>qNX
zd5SDM-v|RO`fp!9@0w3Oyzkwjlo!^TR#AEzzZy@&ds4{ALvhEo`UG5GKL<JI4yECA
zr&}rNsWXjTC27g2$^;R#ZMn$j{ezY2O?IK`T*tn1adO>dKEeix;KPl5q-78$_zQic
z5y;BY#79rr>bbaH%>9xwh!Mh)62#GelZP0P`v*nI?>>hybTyAd2&c~uaaD@exC^1s
z@Ptd;twdFlAX<LtpRT4#zp78LJ=w&wNKdI#BpgaClsFknG1=qE65mBrqT+s47*aH?
z@I{IVnIjEaCVn}I^FF+OMoC35TZPmcc^rW%$g9fZuRh&(B-qVa5*B22P4W?cCS@wC
z=Ba{^+SgRS{NetAtbO(8b%DkT8Nx&$-LG24N}vtYRH_Ivvm{mt;+o<^p@8ZlIPJNn
z2tpf>Xo9fNk<f38VtSUwA?>uPX};b+X!+{r{}f8IGpGQhtMo8I3zQa6#w1rA<=@tY
z1t*j;wXmPQ6dF56!|1?7x~BPmh{CWci3$0yLrwML&;Pi8;8*|t^3PSn*q~qqcZy+b
zj8J~TuK74(G-88&t@ii#t5Mq*B(i8HS<El1aYwMhcwduzl#15{8S2`1O{Mi}_>sZx
zGGd&Ju<=m}zf~s?!(obAK^xgi%z^{gRO?ww7CPyPob{w7>jle3&gV>jvd}Y}hQDn^
zl6$mfS7w$&?isvSscQQDnz>41Mq^ssjrov7K}In!^=|yMs=OkEq?{<93b=&<Ce)2<
zRcWIH46K#j5N~)AtG&Qp{6>eu4872y@{`<!C&rZL^Idq&RE;nRu@KIM<GiWbOT@jN
zm|81GL*pblOu^CYqGD2iFH{EPph}dFh1W3t<<O&%E0c_DVj8O^`m%(cJlz^Da%P9j
z&B02V)DB9VnQWtr2PC&R(3a!Vg*GW1eN4>mMvLQz=*2)r=eEwaoT<~N^#XgGm5iDj
zFFP;PA-R8`H^Xq)t<Z<wCnZ<YUtMI321Yds>bLos>uKX|_J>n{F5{c&=*_!v;`OU4
zZT*V>BEM7yJGzlgiTk1D?&i63*7}g?!)U+z!*`!Y=`ZzmR3_qxZdP~C7%vnaHTxnT
z*Qr<v9dt{@b~OeK$I3m4fnnd)Wun>OJl|4p4dGPs0tvm6os6>P?QrGc^TEO7J=NKx
ze(q?seG#EG##{V<oLod;o@cvp6VV+g(>Kni*XolG`F3Lds9A0&7^8eV{1?AX;sjPx
zbHFVU(r<&{^2zAf3@g$I+DdsE{MX-w9Cx@S>@%sBp&q9aHb|??+>KAqW#Z_<qfcrd
z?+&OGUt4YxBG1T2m2>m9zc-z&3}$0{g0mOhP#jxxOinF-MLKal`KE(BrQ5!Svyzow
z{px^}$Jwn}J@MTMw^35#EZ|y>M6>c0PDUE%lUBYTQoeY%|J(N^=_I`{nOSVck$UBO
zsu{o3`-8vVWnC_~TXSgRcH)J2r<x?FAv>^Pl8(Gb!pyCVZn?%D7H?Ae`y|%8;ak$X
z`eYniB@b|aGIf}8C(+XF9MO`e)@+KJ%@wMPGS&oSgS%qF)(lJ0J0(AHiu1-bjF!?9
z-BF{H)Qd03&Rl+LehWEJYD42glQ-3M^JF=ul$k6&=h9b)OS?80E(kxc5P@he{8H#`
zXjB?X(I!p=QA;xEM!qO^%_wA?t}s|#F$`;t`fz@K9M+t+0<m7rT<Y;?oYlj|U(!e7
zJt~ZEBrer!)How2emON*(}2Geu&4K89qS~ki49@a22A=0s)>)~(|NKrj^~r@C7u_p
zSxB7nTG3~SH5iSxp2)THa*abHaZ{7euMIf)$l*l?VJAaK^7D|WXD>&(=5*N@dT#Ye
zRVN02-sDZx3Fc$BeJt!6MZ@vcOCzUEc_2)8*JSH%!#2!R@iTmhkRq~I_ocm!(f&4U
zA8*qUr4bauH9KO>@sb(I20>D_9G9k~(it0zn$BZJFaU1feU3_tJk%F#nY#{J47YGA
z?P)^jl~A6pH##;9H)h9H{K93PR$VdGNA_ob{jLK~#KXh8a^N4|6sARou6Y_+HPMNO
z90{$|6+Kil-11N({$hB^>62Ex>P6(Xde4sv%32XnFq_q9j0)w{+sd6E6_g9Yk(<b!
z+1OkW;5wo0WT-f&Bjl>3O@Iwoy0z-+a<_k7G;frKxe~oRwe$&c(%`b1oBRHfZ7n^2
zo6U&xAO;exy)cHa>AE^ql5(eJOSuanGl`Ga&xUp^f^*pTrh_?aDuZ7AF25-=W}XC7
zuBDpfH}@+TxFJS;gm}nUsJu-#N@z@7tj<08@P6@{zQ41U=p=G<xt~2!9Bqf|@U~)Q
z9HH_koOMgN#a2=t3G5JGCHCrV+<1?F6+w@U&X3Qwf9>S<#3nl8sn42DtMgt7D1-0_
z-w2whSmWU`H1TK`j$w)Em1}atrLe^ANZu%Uby#9_?Zg!)!xFp8;fBl2VTmdrS`fIm
z^-;M9zRXAEdZP8wC9uSZ*6`|k!x9(M<VZ-Ooj?$Kedq!}V)rsD`{6=RqD1F^i3MZi
z?=4UwX*8>vVjGi))tVZrBQc5EU>@d)zzs~IAt5mpR#Y39L`D`WTa#>n64?o>Vo4)m
z^pT)Mi>5DqO>-P5(PP4mB~78b`{O`~o(0#UHXw;6Fqns$W*d?y(M_q5bSxxMd2Fyg
z#Vkf~XB^Sy4O_q@qLe$TclS7dNMgjOdFzjYB&zs&(9|SjNTQ#IBqH)+gku)Tn*DLr
ze&Se9VZT%&zi#XCXUnj{5KDcn*apsy(vTG8m~<ws1TBLR<1w%iw1s+%cnsxgstwem
zi&N)R&<5-gIplOBXan|$)4QWO4Dt-1*q;e|L=-A#Puva_$Hf$oGuyF$Fy<n;l%!Vq
z0;N*MoB(@d5{)0Wq_aVNRIT)adc^E;OdITUoWtdJPfAm5Z7H9efqHZpWrCRx$0gLG
z5i%xmV`{IQJ_Gfb4QNo09j_~)9$S4cLp@T^n7Z|MMm@G#To3iwa9%(?lCwhT|DzuN
z*HDkmz8r~qY^aVzJ+@STr=cDj>MNlhk?(@}R-i{TevQGZx#95f4*8oP9!XoF`PPWX
zR?5YQ$A;mr1o22@FnM1S@z`*@BI2>(yn%RZR$l?}*l^rHJht+0hj=_vuZ(z{@yiVn
zkL{ja2l3c&y#nHqu*b%^UjgFL$N%<r%@^x6!wV3PHKPrnS-tok@|UtTzm0dYjSs&+
z;?W_qz7^c@n0FHx4E_u8ynvS>21FDAFqbehKoytH?>D=D^bE}5&-dRRcp{xMr~Oxd
z>wmqzKm7Ah7!i`8b`wfqNId@a?ZH1#fJ_@VfwD{~Ieq_o`ttC554IrVv?qQbpT2*c
zx|PpKmX4Dwxz>n{EQMx>cG75>B*00Q>J#TkoT)duUg#n@323<d_-0`dB5Ee3>mkZx
z+|87yD}Mxka+?(*w?NX0S!H@if;V)Ow~b^Hb^kV!{uFV7q&sztoS=+roPN_+E=90_
zYv~h$Ga%+Wi~x)D3885JAQw-MGU*~Z&?Opimb}8GiNuGeo0d3=v@dCjzg^9P1j!7F
zfd{w91Gxy<)gWX3vJJQ7%na%UTt+RCk-HA4H=tF2<aJUGc=a+~Wf-~p$a2Dqa+*H^
z-e#pX=j>_(QAsoN%eiv_|D)zUD0!0%CSYbWK(F%G_L(<lZ@S)`{cqvlgY?X%y~&)E
zj%z^A<rx#paI$IP@!-kwh6kUzNOuMMq-_giX;L#~Y>5lS|F#kr{lSZol=29I%RRxG
z`+#+SvkBcIOMwi*?Ji4wAExbnz&<a?@_bS}n-RHWnu|R%Vhizx%mdmnAV?j=mCREz
zW5-dKd3=cQ86|`eU4<W&LQo1VJZAD9Iy%8LRVz8~uIHy=r=>d1sy*alH6yp^0IrF}
zaVCu#(7Qx%7J-Oj@Fe+YDlbgX@-;+F4+36)eP|a?DZ3}&;iY5Do&=-LqJxACWlDiN
zN~i^RPC^}en5E*?9DXOJvJ|p}L}gU5<hs7hLSF{sK=X2)AT*hwbN$;J>I7kc32q0x
z%XNY<$dn!=yQI_<4iFtPdZ;M2q=OW)kPZx%o7zlntrepFq9vb+3rq2Ny<{oQaaGlS
za?&pY$g{55DZ>|ohKx(nnNrz3Mc2AXAnqDZ@wA5YrUu*i_@-b&=8Pewcd5&0)iCq(
zFCY9?+M<@XP<gAN-SX3i6UW)vJ}sA)mMK74O?T}mw?sQ#dh)p>$nij*&%wNj#F!<L
zg&c&t9H~-EKv680dRHy#xX<4~9il9Mm-`ZeyU>qlh<JX}zO`UHIpzSUJYz5O88nC9
zYkLMtDF%JVKnUhQooN_!_s|?zDO^%qAs?bVFpISjNpf;?eJVo1IBC0s4sQ#`3}QGH
zZxmVs)Mn*WIj6}Uog8?;+56PnWo4W(Oat_gvY+B}g6IR!b2PXRr}Xuiht%nR2DyCN
zRW7+5(aU8q%3dTNp&WB$_Z`G&9Z60vOU7Q{uC)8I#e-r3=JGi?wX)g2Fvf?+m|5~_
zDQocnIy8h1!k1z>A%uV+s??R!*mHp!<d7SLIN+6fcIOZ}UW#4irecM^q{oU5WolUA
zsmXgonOJ!oFRjYU3iiT9$diSCA|H#eiADLvcphuBWKw~2nh7(XHw#EJ2*y5L(#tBW
z1ag-}Y>R5s`J;$6#`f|~I&nl%l+}__Yl8?o3%)?pj0-BUewvjP%!*1O^RzA>W0_QX
z!9-);SV@11JV6Y*&jgfJcQI51ko$GeR8^go62$Y2rD}rwE4zzn%(ZiW4dk^^q3HD%
zAzfkt@l9RPbu6o<>e`xNb*BF{ua*@X9Wa0Q))DSdLn?N+0vBs{j61FjCy_&5hmjuF
zq9tiMB)Br$X!z5xY90A$;_IZ<dHbMr)?wOcbpn9W$pFnyS`cLpk|Jt8o`O@L#w4T2
zRj$}x%8H!>ds-=ytDlH})3jH6gAQgX@uaij;WK?bzOv7PJgO?#C%X!^F;kv>V*!e&
z%G)536x#wtBH3iz*U^@1)3AHtJd^DQDZS5?u=hE%O*$B3>|&ByN{A{I26GHIb=)*{
zf-2q%NhkJ_C|}bwCoScwBU6@ri6%9ab-BjpkbQ%y+Y=>+lvDtJ^U~82=~Nr=C%Z|w
zH`VZ<kpYG1e(r~kd`bL}xu2i^SZlB7a?TjSw;6F>X-fRq>gci5ewR<k=CUF@_OCjC
zw;lBI+$EY-Zq*G<w%c3E!U3JYckC#)0LPZ=hV~$Yb94)MZQ`2RgYC<=3{<&fAWa8w
zZJtU;e(6Myy>j?}EZFkfZaa-Ba!b;~WmKg@&?u9>8IG@sk{ix&spr^f%OwP2b_zdl
zg3CNia%4iu4ro5*-Z1wRWLqS*rD2uK&6Y(uonpyTw)^NuPypsm?&+v4MBM&ohz`=3
zKSOkH-#$3#0G=5f^n%>&zYd|Bx~Zq{n1MlxYdw7{$pxT)f7|!;;dQ^PAPobv=U;C+
zh&v1#JBVinjXg*E4jKi?O@P@jX!LzGfp$l4XaSz{h7SEsgU0mtj1%KS!4(tE?Z|fv
z@EqSA`mTMPhNat@q&+1nCK8Y<|G~0M>w%&zfh3r?9PsPY_;Yn|JN`5j?eSy6Z&!wt
zairPpTAMz9r*Hy}%t-AxrvxjQwDA9&^7=U?kOq|ycbss+(<L?6P4mf`i3{hG&6fXa
zJ_)i3;2Y+X0Ez;Dx|{m&q5)+~#l0q!eq2V;|Hq6deWiwr83u12P&TmdG@fkf-#(md
z;M{FA*-GR-gUMDRw~r+ocoz>P`#V|szqjSe2MT4EwA3vS1Tio(HkToq7ZU+EmtYVV
zD3?n3H<*9&@pb&`+YdJ~lV0frEzpL8bw&_1h0aKqp`9i|QjVUWP2VC#=*m+=PZ8*a
zM`a(TcYyfZryZ#C??jo2FrV}1#|qcQ6@@cg;MEyQgcz|!2!<qLPE{HW{t1Wg=`gN^
zHT~+!Uip-sMi}v3RvA~$ROTeoi5Yk1VwOW(X+nRc$Q>18eEeL9FAL#0y-2-L9Sb#}
z6<`Nih>Jg+yc*h?ks<)!jE|Z9D6q!_Xkx-<U@L~}tZ>dUU{!AsK+!~G3Zp$lnV>4)
ziaoeaA58ZsTd5&%-6u#P!^TGM@$;!zSY2rwx}NIO0?}k$^-pK|U(3__#3u_Yp9yJt
zS1o^}#L+rjZ0#-WxX@n`hqCn(j%#Q}#t1F3YOm!l+3V0as_iGfHf_#10e*q$okj;r
z3hfrqVhl8(J5NY5t2zo*o>h+W)W()Mn)q${$_kk*DfxVr2zF4iB}{TM+4B6lQ-c)<
zzIt`BdcCk3eYVx>&i>XQRr||Q(fH7KLt1|$vjlDT)FoL>7N?8lRzFWqu*0RAc=^(6
zDn3v6O^p^gE>}K!i7?!z)p~p?&&H@b$;6NxpTbDf1Fk07$K5aavt@Ch)x%Kt$_RA2
z`o}V_4zt%VQ+7FKVB3whRq<Z*L+{PhVW40is=vh=0!zG<Idz?Cl%XtRWBCx@KC^#s
zuWvX1STJt@g$?fTx1Vm}4@S{f@^Xbi7zu};e>{A>`4b153lu3N?MZ)79)A9IU{Ai}
z`YTA{P+E)Fb4Wf%%p=t`#?(OglAZN5yB4UfFq#oepJu$}$(N>hq8zDp;7Fcl=_Tz>
zw$dWWQFTeRw?YAfr)BZkr17cV(dmCX)JmkWoY30p@!rbnLpFYc3^T%^nlfp2U)Z~n
zMxZDM7l?XQ3Pr3v#1g5Bmr#CAw3|p|`7_bpJ!uu@)l`);$RNkHafn&T@YM_1osS=+
z7K*BD22rUYexM#g2Uy_GK0Zk`OSgXaHP>E5mm$w2Wt>Tu2_k4al1r~2`u=~KX`yJ4
ztqCyE(mmO*FK{S~T}UDJJ`)C*y@M1c)p?#qRfYhgIn>w!#GXJF-J(9se<r%SC#`K2
zKQM}dpI%x^(wtIjlG@@aQZ@=xAM;ohgf)0gkL@Ai)8!s&jqt{CNNXPIw9shz>JNBm
ziNHO;6-TH6f(fz|4e<r%6m);d=mr5UMs9g|oC6nxQwu37JN9R<MR1C*ma$EJy;^=S
z!3hWMiq^&;enjgw2t^eMgSe*lY;~L?ruOqr5Tfe)%CSHvOLW$+XTNG0i=7lPy$sj$
z1_p4)8#wwfAKs)UHi;p%LHYM{yyGc(!}B`dF?OK3BHum0J-!+G=2$gJVLoblz1)?l
zXbuZXOiCn5j~yDkuy#Swdjg4eg2cPPZ@2O10^thy(NiUi3;dYzj}JA$w-Eg|Sq20E
zg@+oK!ALh0mm#Y!6ql6)IKO{D@4#LD{rQhKo>=G1Y5&#V#$O+wZ~uKO6A_Xr?M}o5
z<20fF{PE^rASkAdI|<{MQ>Q<FJ^l9f&yi{wi&K~Qq4?>~-%rDnUy>*tCsA^(mFh$(
zG*il+w02DTS5cJsm2*WC!#iC!x`;yo-6Vg$TbQM&G?UVe6cth7_sf5~=3eY5A}xlV
z%lr|T@179zg|0D71B`S9@3<<ht!5y`#<b8}im=)b5AQ6qR*n<_W``GCGLU8>RuUtc
zbP*p&`lRTtNVk;#PS2G}?QZco+S*>Z_%5`$T_BTGX{D$sCzYh1fo$lOR<X+d1tCZ!
z7%{R_2$D2Gkg|@8@HKyO<tO2}l%h`5bI-k|+y;YE@>D+pkKL1Yp|VCR%Os7SM)j2I
z%stQay%XnDmLe}flB4q6U4z&?ityVg)<DcRzNTr(_&XC)rcNEx;9dv(Gf4Qdr4KE=
zmNE4)$3tNDQ8w1)5=hR9<{lQUECdp`8FkWV7?LeCHu$QXdKG^P47FP1*T%1RS)|*5
zTnHz{oH3*(0~Q*M9m7BV?I+)ry5M{lLtUWn^21NP?rsD0COQkn6u2YNd7$VX2Skm`
zq8TYI;jxi~0eWox3Lqzh?+GqI5jf7I!O8G00c5r{5p~K(1wrpzkHbDD@PZG&Q=jlY
z*%<_U=!_uG8<2lIeHr(oZ<M(suc5g(Ksd$!^b{Apl_97Ulkp#vj%=(}@gj<c_@;=*
zz?5&C=i5p$Nf=v4OcpP?I<A;Nxl@9YJBH3qB&ZW^VKP?y<`R?^m=#mLF(N0(dHx-#
zr`5}n4B^|Y_`EsoRZcKX5<5<N2Z|foJC6VTw0GoNt!jVo(E3pA-6z4Uy`u!bpZ1P?
z@6+BzN&4s{VeTW`pwBNkPPv)o!e?z(O{9o?+j3~u+TxLx(^K5CoMHU$XgLGl8<x|z
zKGt$}Nw8SXAi?iwIRoE2EoVuRuUgKo^(>atw_dZH6jqUXG9y&iW1*!Y4bx);_Dlg{
zpagPN!o`2a`(BC5XDSenjW7I4;AH-Dl1iz8LK=scd@UzWPo&^V8cGptqU5x^yi1gz
z{de<ng0JXqZI0!guZ=XaNLe}m$eP?(!UQE=DF@^<^YRXqU8umx>vJQEaPluOHH=gH
zKaQw^!5<<-j=@c2^}9CecfTB+qa!&buJFJdeV~7+^wT-|KxrdKA6mbjqYnf(a`che
zR#9%%ueOm%u6;clHK8KE13UQ?=*TuFV)DM-*Y9MiV|fDzy9;R-0#m8e?<2I|MN_bf
zb4UfY$3p<A7!A2|8l+*2!mi0Ql6{8qW15aH(*=>kbD;@2PNgRn0=JFOgw#P!l;@!8
zFdBcs2xN{dO2&4vD8EBj_YW6I&W-rA$_0`O^;*<GIfOn`G@Vk#dhMWBUM#-6=J7J@
zU8Etw`#uv*q&1TR%G~R!9!c)tDtN7;%THCDTaoXqICou+uQ;Pe2e3zav!3EBUPU_w
zT#OW#+)DOp!;GHTiu2Fr_t&yLrC`WwTa$l7>0=;ESSFM^)Gl>byFO;i$+HnjKDy<!
zT62gfxT_ey7g1JG5Db&YyNB!>;%p9a4|%vs@@4n1EEP`1jPMur4F!$yaRe^Qy%&FL
zhaYl>^${A;?H!}-GGhGvVGuG1Y3VT`p4D{Mj({@4%iNn^3T{si*`%{6k|HHvnk#=<
zNYd53q+u!NkvARFkxDXSz5#G0B#vm46HEUHZ6%Y^rlqq)0%>{z1}4AUwLcGJf-+t}
zVcG)md`2FU*Lv`vj|9u4){}!if;qsAgFeGO9O_wcxyfTlfo&za7uuS{q8a5VwtKG&
zdPSUlJlXQ4jgd9@>aY29i&`R|^bmi^N2x~t9!mLup)eCZU~tp4W+`qGDpKs8tVB{d
zsT}zpG#tN_Fn>=pC5&RI#YnXj>E<+N@fclFO5uR<W>Y+)ugxZqWc(hTdVX3WCQ4BB
zIow4EelnQL>Whmi`R`u}-Qf4~!8`I``8f|t7RS3@BtPe4P?NDYRJpA{h0%Y?F~7a>
zUebt(%uI)A;st&LhzIx~I7>aks$44?hcS(R8{kmIHLGnib_E>0fxL<;)zs=s<qWTd
zGi>Z!C~JkcCc3Q5q3h)7ieHgPn(NF&f~h_@nS1)Z7$9yP$r#LP-V_imF@+;jz^m|q
zCqFVjEuWA1lO<0o#zUA}?0tXJvbEn7sx`?Mbsqg@5zgX-7UwnQi?>KELlo!yFL|z!
zK&7N+SNjrL(?SSyu7@E*RA?K7(79f+oE@w#rQ{?vLeOEp!mY|YYz1mF_6iH15z!P$
zsp7TP?8H#?G|<~2<5ey#g{^$Dmi|DrQl?$iTf~}XWo^WaoPRZMF|U8*OSr5@984xR
z(L*Vi7Osm(XEnLU!qh7H+v3k4w`#SpLD<PQZo$RpHH^}D=A-aoCLYV=N2;*!^{~|j
z)Ez10jH&v8nOUu$?##2hZQBWj%*b(h7SnyBZSAF17jYM8Ljou*Lnp@3y(UX*8%@sE
zmr~kg05?o*%|?8y*%W`>78GF`xriPy`cXMLv*#YU%OL_8<1si_%^%k}(ZXcOO$=tw
z%{(B*sP}fv4INt5{ZQA($xN@Q4tLu?Sqh2)#@@korrgkT>+c)uErny29u1WGoz0-V
zU+ZCrsSXU=?~8m>j4df)82L^|$eXo@*1M{ypegn@Bdr%bgAadTv^7s@psgM|==?(b
zzzHwpZ2X{tq*V^Lt{35&m_NRK#kZ)qezImMAp%h?mB~0Y^opKB*9biEo1z#`z448G
zuCtDyiW!QW2g&TU1qwBtN36bgLhj6CCuOmO7bTWj*E~u5ayePak0eLc{vjg7WIJ%&
z7>b)ll01bV0M&o6J!yreo1&PVqS6`mR;QB{6UaEx*rgP`7f|ADak^WfxpU-Z<Tz#z
z1VPnus&*})+m=Pm<Y_fWmsIm!OQt2YD+zdLd2L#zEM%b!10sejVe>@}a5-1Hn9dv(
zLjstALMZHt$X(c6xwc16=!HM7P{4Lmw$8#hLU2%S-6?;OTR|ZvJ<}>DV@y0jcRH@V
z6a72e_1%+&78xsK5=}id_qdFE3L7$pXJp)u@Q93up0hIU*L?uQ-Ex$!`O`)E<L5(S
z><Fw&5$?}Ou^ZqKA$A?7dRXNL^SZbdhyqYJ463bTloUwUz@kf!?)7h;h;G0p6glxN
z+Xpj+K@@+o$M5tcV_GVcE+<wxoM;}n-|0|&?Z!Ge_D-qr>bRD?7p2pq)OL81aAcr(
zk+MtZHa&@?Boe@mV>Z(NQd9ZtN#ZKBsX+;MT*G7a1IwX+U`DBN^m-t$w}(~RxtiND
zqCSL9N^0Ut%*bReAO(NcyBbjzmu+j=Gj4V@q|j!HNJCHB&u`V{{1%(no}M33Yqq(z
z(bU4)?5~Z8Z>2i3s<4C~8pSZ=I>NPy&=hZWu8;&8(3oLh$hnEKYoKH;tA9=s=@dec
zOP3Su0yuwi+^+0-G#T_XSD1uJ`Tp_w_CK1A7MJ0HFcSnZFfuroF;)#Dm++-H4wsb_
zJtu$NT2XHtw-J8VuOQHu^g&z>ha8ecV8Ds%)PRB-mJ75s^dQUE*r}s4q;uf>`+h^-
z>2^;iYjqoAodSt9yE{wHkn_!aoIP?r1>+1kpPlh8SBXp@sZuoJgGnW@k(gD$l_L|K
z7n5+IO5UVg)r**t6C`#zWJyQ*xEP{P>T`eG)zMYNzk%xe#T$yD&Rp1#tQBx0=XMt_
z7oSWK#xx~Wb4!G|#C0-=-s@xx2Rw7C3IU>&s^E=}sTsz`H*47?gP{X_kzk;cO>p8M
zC4xib4O2Cc9UfDxuj^gTL8qc<IFzPP49B`+Oc)^fSVg>6Smz5WC9e`=5^9gJ$}WHA
zDi#x()$F|qA+)&!qLeyUGC@++^>{7A0W1a#F5Y1jEKtC8eJR5qS0}&@MW3sIzL?0(
zR?jB3-DT`-lx8cOkD2`u#!?9@z>=X^+6q7^mRMhcLqesAVuW229YkLZWJ0JkXBgpA
zofo~Cngi^OWKZf=a3)a=6~@9OPgQ@!yU9aK#jHWr6cA;CFk<~D2_l09E~kXRfP}D2
zsl8rw_v&I`9kWYy1*0>Wa*YNUg!!q8(d0nwZZTQBj;g969xI&iIJLzL0<i`uM>Ykp
zsUr4)ny--{hBj9;g%L>@0MEJ9@goKdp?al{;A|BG5W`hOfR!Ap-8CvSDph|&Ur~@^
zv=+=6Vs>rLI$gcpc30Gv*mkNehe(x0omfXu#YDzh6i)h5Y5P_cORBc$eOj-P;)b}H
zt4|4gb@c7G-OJxTEV{>+m&=vuzFS_sSzNtx$RzHp`&;*{`yLliKI>jAE>^F|dm9rX
zCwZ$FkNAUG;gJnYXGh)X$Jc)=9q;7b<$t@!H=|#y-YqY?)9xQHp6UPJyj`t6T>qu(
zetNfh`|-8CSpL%e{PC9$$8Q$Dc0Zo4-v0f=>e=OLarK|`i-rC9efRqP@^$wsS??@$
z*H;(a^<s662Y))jejJ~0;y*unbcFv7^d~qBth=>#2nocB%;HZH_)~x4G=EC6QA23$
z4!K~RlsWuCYCJv(K5zYb@=A*d!-W^!KmYY_qgfhJdHM1E`?LKT<dV&zIf$$v@sCjD
zk}dM!7}X-yhyMJg`+oU;d3E{$&Y7xjy65LBxO-`QyZq~?)eonu^VI@tUvE5JUam05
zQ&u46P$zk+1<;2!<5PdFnGu+;7YgIzsa+^A3zAoR;`NDZ7aEb|*H400TGWDe5%ePI
z#iK{v^Q+~>>0<S&d;a8U_j2)Rb=C`==jT5y;1Ra#a&=t`YJCj>bG`g{b+Nc!S7E*P
z<KoS`^Y50QYJ>@jWGem)BV6I#N)g)hH5#wi{Ggg}^r|rgU5tN55H=FsNd88uOL8_0
z?m=kok92AmHPTUQZNYMiA{F(CW3WijLfFk5&br42qn~y^EMG3WC+3@wLc}fy_!ASM
z%s4?r?gH$gHiTr8R#(7Gu=7FKg1=}nwjs+cBRC_Mqb8SJP)%ZVfu|b*%4Ja-K?a=F
z$^vQwtiy=tY(#%f-ZcVjjf+rnA&?R;U5wvjDWavqH&z5r)+xWd!-|v(;;?{ah7wCA
zN@b4`;}afpu>L0Bfsk0R+d1-`_xbq5$Fh^}iuCf1An(ZDq&HAoXEIM#Nyi9PpAnBJ
z`SYc+LH2?rUODpkRE7aSv)v}CJS?v%gWb(*O8F1VYq@`y*GgDtE<J#3q<uN;k4T^b
zBVa>J6A6?rw$kzqf#PVT%;Qi}#!<pFhbRRbw*F9dC@-`46A26#Idqs&^sxMS2G!Tt
zGAI-W9JxlQg0oo;qf{xe)zjS9LZP-fWGC<3ot}O#a>c~AlMvBiNvIjul92W$)~Scw
zq|_6cZHRx{+!Gm0Du!7?I~0Vr<Y`d#pn}<NvFg28=2`s7IJ^nD9jKBYQ~ZZ&4Ox+a
z*O<qsddesvU!?dCa5q5YX9{^AYH-+lqWge5K;!_$+G#74_dq&fza4KTtrv0-aM!jD
zftwEluJ=9QlH33{O#`k&%*D6`E_Ob}S-KSB0bGBHJhjH+kUl7~5ma{$f0BSR;7qa$
z+;U%y?P=U#);a4!9HUApp>TekknO{7EM&Lv!;_+ivJK{;7UDhleK=%qH!`%teHdg*
z53-&&ko92_WJC7Y;BBo988kV}65^5wkS#K?vFNRSZ`t^xT{f3zK`bd^gX|#Ik2+@$
z)z*J}H)EOsiM>;`;4DD=C15*%#z7b!*I2x}(oVr6hue5$e;6J&RlMSnH|<3IC^ZF-
z+BJ-_Euq+5>*i_nXX4|N946vXJC6DhaMMR9Hc^_z9l<Fz=UwW{_x1Vv8aFL3P+6g`
zvn4+YIiD+dgCco1kb8{x!8lx3-60OkVK{$mYqh~)y1`+dRu^#w<<xH%;RrWN9mzQ#
zpA<I{h9o$r=_mFLla*{E%~IzJ1vtdK1H|A)6~=+ux@VNb7`1?-s>cauger-%IevM3
z9|GwCitj#l4pPUccJ|;D{BTI;+mH@&7^GwGo1OlMVh!mKr=3nI+PM7$6LH)5SsH()
zB_tp~rb609oVI(1QzDyYVG1R?;}b%9*@bkvueSCyZce8ZEcrM_6?gGB{NM20VSyeX
zvtN@3pzqc32V*wwp~LY&Ivo3SI8G$N4$v&yV#N-{s>~AF+Sk_DmfhHnJgQv9J>=2h
zbQb>r<nc4%*e8#J-s#w{s1y32vWI^>#!=*vg?FM>LO*J6$YabC`=AbgWzYw8Da{kV
z+VRJhxQWGvk(IZ{Qqo^hL_dp1*++Pk<1Qk_-B)d74=F7AS$LExfj=gsxv!)d+C$pI
zA+(3Ai5v!@)F*2)jiPY@rAfZ`D@W4bi_8;dQp2p=3bQmS@k`qClfPhbF#R^-iT|K;
zz?^Nl6mN%P;se7m^};c2e%r^%PYX=_VJB?~Ny-*<dzQE?gu?@s<Xc?+4-F8-1((4{
zHxvOgmw_)R6_+8LC?}UG7&w@JFPC3Fet43Zbt!_4KjRnw_xADh*VC{OKw)g?tO22*
z`}NPKcwZ|ctP35`E`#a*{;m7{=|@g<ijr4^lC;*C3MH_@7~feN1ja`RC3C%eu@WNp
zQc8WJJ1YK;cY8$g9Yg48+MA*M8^S^pQu>{yT+>WN|K<BYEB?FnC^YMT8!wQU&g<h_
zp6|+v^0iQz@;3ak({&83uI2I$AnSnvy6CCF#C5+DnZNVyjLo|-1Ml-7FX~|t9{@ds
zvP0b?r|=4Ct$s}#^~*hHT;nc$(Ku6falK&FsyJS|GG5Q_=JG{7w>ye5it_5I#U0*;
zRWUxfKtGG9X*0!T81+(rcpFD2KIb{tAya!#SI_6mC=>lMjJ!@pX6)5#Wi=zJL0E4)
z6M~R1pc*xd>~=aMYkn&iZG`r&p|DHq25pk@8t!Ay!IDOR@?^7o;RuHgdEz|FK*oaP
z%H~+s!{YBVEKf|3fwQEv8D5V9Ee<Q0{qQmhkHM=RfVW<F54`n%%#k%FWw{?t>x7ef
z0K@KwlTmmGPWAo(ob|$c;H+mZ!l_r;C~nkf6y6O76}mNNlw94Ci;7R6xYE(l+R+Ia
zpGV)a+fw&AC6Iz#@)4(0Q<!2_sbAy;dKFm3&KuzynlIxKps(walv)IXoI(tK;?V$~
zKw!TZk6St#OK89q>u%$qe^t=iBfh4lu!!1*ra97EZ5nF#<D~-e69T8OiPt`7qP-Um
z+yQ68D9qxug%t(Y$~TnuU?qj4Ta3acR&ore_Xg5P8G#`HStPAY5L&`|C`|@HbJB;A
z-QRz7?9NX)+zxUr+@^R<Ao#(Xz9wJO2R^stw?MIUp{&sV8W%2ie}0<yArn1sXjZw6
z0b=jXV$km9G^XX-+n9SWf+(|#%?=eI6L<3&f-V_?O%68+wWlzgO_64#10~kp#(NWN
z^g^p0STCYbaIFaTkV1**BjL=sjx3gsp7ZL?RA659xoS<Aw0-uJA&Wt$+297Z`&S2v
zLZM>>nJ*a!J4)1mf3H3{C=0N||CHJ6=8Bw@B{|&3l@21Hox!A*4E=x0oRA~PS_CF>
zQr%h6D{$mKEr%d_tdiaoqaab0vw2ImP#OUm8PS-0%IhTaq*HkO=((=$Om(VgKRBVy
zjGiQ^DYb;t+Lj_`bMEPwoe~5F^_hlRQ|u3u_rdSLemixJe~vrF=`I;+cAE9z1BR*s
zh2kDL#YIcaX3b#LS<vx<N$DWRwtY#!<Z-vLI8nxzGEXWYT#5&u)(ZNAdq$^5><@r|
z1}F%HJ3&Cw1fUa~OKh`#wV$t``>2BR#qDGdMH07e2<l|CEV?d}bS!tuW|rK4D3P#*
zYV~((eFVzSe;OF%EnVvEv|}A9O}EOP=q?z=BbI7|P!vltiVI4r912p@W=9WLYP^?~
zmpd(VAgpoAVauQ;%}h%=75ph4EX6L2;u0=oQkKF^zWyf!8>F>C`8)Xnbb@5YNB0iI
zXF0hs#_suh9bF^7?dOj)Be+Hq2b>E~z6D?#if-1re|Nd99LX@sa=vDQ3WYP!VEKRp
z1`uG+#0W9wrVT;6>HtwwmWl25S$874#07&hv@kiO6AiAWxMgtl=zYDCha-K?qgf60
z+2}bmDO)vxB!NM=Q4lw#5Bo!EN`Zw3bxs<blqo^nM^jtwy^t)9JL*%6*5}zsBzJIv
z3tlWXe?j~4A;o-?&@ko+ChI|&R+UV*a|!j_iZ2?HGvpy+BNxs`yC`(p&x%f$ENs%h
z97769)(rV(M|#166lZnC{MX-plPY0a&rfkQMb%Tj`evnPd}rqoJP1_2EtwBx@>Th*
zm+vSS2K0Ol2pa`ypbT2^S-psKzviTX-)l`We~;29G*y3brjwdFqkhoSp}0RN#wE1^
zWYu+rUcD!cfU!q8_EhuJP<K*!4rKBh1R8Y01I=?WreSwJb8*eIn179D#$rEswO2c~
zDE1adL4$BQe9~(MV90rUGi~$tn}=x&Up-Imy6skeeCc9vHAMklCSKw;#%43^{Dwyz
zf4i%8mPb9yqiK$Jn0+PT&i(9uHNhQxH329F)BfRC6ZD<@&cArzkG^=Y=AKY0LEe^a
zkRrF)PUtA+EaI!XwPJm!DUBh8xg2U;&hpi5#H5~1%ia4hEBkN{?Slj(Jer0R^_aJ@
zS|I4A6GQox_WEDx#HgTz(x2Ul;j0|(f5b3?|B+5iEQ<i1Sh1-Ip4f@032$^_>cJ;;
zVrqgLofx+I=XGN4<0_|TEheAXiDA+@rxQ~XJ)sk$j2GbGKhTL$NakhE?8GpN50rNl
zofTd?PHD+Bb4itrK2k}^j<+VbSV+m9fdR=Xy7wdz^UaiL!o16tT@X$+C^FOFe`QBD
z`s6~sQd(#{vyjgyo>0hV6gLX_d`Ax#^4CPabWA^>40mmAPzDC>LcItfg&BCSjz3lz
zM-mX(QL_j8c-1!wQR#-_>j8?6qTNQ(7mN;L2to%BSE8|Ge7uQTx`g*hKE~y~HProi
z_kKPdg}E^tV|>n+hII@F8e2`RRma0cUq!K*cJBXG^c{_OX!Nhz@q7h0fH2CRiCnFE
ziMI{yJ5a=(@PB6wcRGWYaE~9Djw0j}Ps=H}6OPI$xs7wO+Q_kP_d=0?eZPG?{R7x8
z?dF%kNH-J$GdP!l6g?CHHJ8CUHYa~tOOM<(628x`P;NcI4$UgE_;8#C#%+&*z3uEA
z7D#dklGq54V7&%{tp9zAFFhzqYW1{{J?ufPZZ(Un7r!c2Np>GP*}c1Y{dDu<LFvv4
zM_{pgdfx&6A^8&|KQz=?0AUn#Prr5FzLF9RU$6PnQYyul<?oK~bi7mGcm#i#e8KU~
zYIrq_#ZwuhC{dv=)cFp11;-<;vog%btAPvm<M&U0zmb`BDZB|^!>#`H_UYz7H@zUJ
z1R38M=>$66{rKBW_(KB;V?E0!EQ|N)=kEQ@Kk8Y>YI4A8lEx5C)MTwN0G%;jsPI=(
z6U!qBrPQn*&CKfue((?lm{EU{6N;D1l&mMIFe)pGD>M)9KR$ivswa={h8evS%IbPX
zHpEzQ!)PrGy4m_ZbIfKSo1`F0JZYc**Xfx)rSF=Ogn^EjPO`~}FL_#)T3ex&y{6RD
z8MmaKjt)*2B@K2{BWYy5L)J!=Za0sp+1cDoYOqZnQG4N)sYkc8X{3K3uR5cQ(AEtj
ztF1z+8$KeZlP6<sq7g)_A0$EG$M%U<5B~f**(&HM{KnIupQtNJo)SI7q-Ov{$H#yO
zad>es%0oxflqt*h3C=`sm)J;C?%-tG$K#ZnZeAHfGtgM-3}^<N>uaufXX0(eK?7@H
z3xb}x!ixtD9SLtO+jxHrV{CS7^qBX<-);jXQ|LB@WeS*3e095m(1SBNsT@C6WSrr<
z@e;FBw*Gm1SHO23qcXIU^&G2u>@C_V8jYH4qd^XGOd^z`tJ*BM>?xy(OD;+&(v4j&
zE)zWczS#7bwBC-&NJa#uOc}m^Fd?$qz`8vG=~v^m2Q1thG6{dRfLX*u6rP{1M$G!Y
z7C|%OFtb8KZ^fzp&Q)G={O~<`!2L+~`$0QoKHAib!JBYZ0L#jfLWVhz(yWsw(TBZk
ziH|4+@S|##Hbhl>xvkI=Y^oU}e%g!<563$rqh4Hg?Ox6M_2f9Q!Y{5T&j`hv;(B^g
z?8s==^>pg>bf15qpx^Q?{DYJ6txeYRu#jcvou)krJCe$l)ylr61Y&n7)Mbod<l;=h
zAOrV!P9g#PXgufPeFk%bO%jF^&@?6a@9pX;#t9?Mlms)0wM`O+X+UHWc9@h}9KPe=
zR)}<)$O#D*YmbFBPJK8uDK|wFM}jvmlW<O+k1mIw)>(h^$Z4UPuJUqSXDkIUZnbsR
zWDKQZ7YGW_ykjtt{IaTMlqnb7p=JaYyEf*xtFkaN0-zHA*V{YrhkyTZdx!S$^QU#m
zv|+5I1<BfL>iBTF=GVNYF^Atje%xA+eF+KL3ohuzA@8-yVBfE6pFJsr*$IhT7mJT0
z&Q2*d`0{@YP8G#BLFx7Yt4k>e0@<bpUxu4axBl1H>GspA%t~=G<tw{<xLwOil-4Y>
zs!=Sb9WNeum#kqQ?{ILc5BxdbkHLL%9J?%uL8C-K{yK}{y@;G`uBS06GQ|(dVQ})~
zF{d`|m{n36%~>0&UPApQ#oZ`XK3gDEk+Q51zYTv2mnY*R8_&*0o2_yh^j8izWY3C>
z;0yzGJ>m1R;u1*X45kFn*|~pP-<!%-S}33$%3f4MQS5Zad2MHZ&B8%EcT_W@T&P*3
z3P&~i88wsO{Ge<-<3h~@`IMvrB2??UZ12;c-TrqxB41o5Ig_tkCu!)VuD#LQ%*!(p
zP)mP8o0%3u3A3Cx%I;u;)wP|)Bvf*)qE$U>Il`CY`vnyo&G@INQCCcsi*=Burq*N{
z(%c;gxw&H{2(8>TckFaV+F!Bb1X?(4>zS5zs7<=AMn9*jnIU*O;bKuU14N}rolP6A
z0M_T0Zs!0OkX!D@xsat<m}wDAdp|qHi5`E{qo1({_;e`rXj#7G;Og`GSqV~01;XVT
z<OK!DrX^Bw&$UaS!9KGi^$xA@LD#WUDYgnBw2)z)xn<54%8eIs#|KL?cy;dRI^2=v
zW3Re8cc#JJ^O2R$E*wv<nG01jJ_q597a*~oaS7rw!qyU15T{G+cca0OdpNV%Q44>I
zW>*(4Q-TuK*y|yGHQ@&G*UMk6iI<T7v@Ewo;p7z~UNzyC+|$v|DH_w8fu^*nF6{Q@
zs8@@!9`&lxS?8t5n0<OM&Tg(0IED^$CRe36+TyC=4H+HK^SMU*EA}DLG46W!e;;yx
z9}-PC9^wB)Q|p{_MLb(XxVfx%IhKF-%;x2U<ge;pB8M~_-mtl#eThKA%B!}VmC3*L
zBrDZ+y*&wDNLMrTrP`CgxmbAnp{neKVkF!2+WpDURj|%Y%#>k+df)F45EJzk-DU7F
z7r*0D@c<9A^T4<};hd@|FvG_T^z!dTM9Qu~<3`v1h<HB#=rNq!uFifFDZziT%b*{d
zb~JjIp>iOsW^D9A>q0|C(L2d@9eC>y07iw5M9G+432Db|xZ7%50&)&-dZzI{hfiPD
z;eD@t3}O6@!s~pr+^}8jv^gDXWaHHiKdg5r*?j^COh|CG(TJSsSeAZ5^0?8@<9R%<
zqL?%vO8V}~5<*&`oT@d~M@)Y^(pZ|O9XwF;{-Bu!D%V|pjl?q-c1BG{#Ns0={VvGg
z&kRGAH^unCv3oUyo7l8YwywEZ=s2+~L9-Lh|MT`vOSfS&Wr7xPkIfW^d=rr6e48oD
zyGu27U65bDE_Um+php0L<QxXD;LE9!`_7)OBnui0kC|C)y9S@aIgNktTjvvP+<A4C
zx~+lGN%x#e)n4n*+P}nyQ(bx6#@hwN#}AUK-oJNBDfWytWA)I)8x27>HrOiEXqnVJ
zqCR|HhZ@b1R+}jarv7bV)tKgc3WnIPg5QTl%li;gt;=Uas~%?Zo%)8$qUVSFNZO9k
zl(d~XzH#jab10uu=OAGQl1TBd_1y&^FuYzXV;G!vzg?R`3*JH#if=ah-XX9l)3GL`
zS5-J@v!(c4%GtOLugm&#-cAU0CICNKWBuh06<i0@o;5gA4FsblK5s1H<GV2PCF*>Z
z^W{GQXgjqHm%&Ik6ah1rvC%pcmuN#5DVO~zI5U62mL-iy4xb+>%jR<<o7cPj)9(3E
zQDcSEV6l1n)BxX*{0)*X8r@g`VH7k^Uz+zXq(sA?J%30_rTAm{x8tLJxL4qK1(^JS
zhkHx#qFalHvPS9j6#DXXK0@Z;c%^k#271g5_#PY{K0LkI$t-#+Cn}+hZ;U0u(iAy0
z%SC_uBn3)f!xa<G!ktLx`O(4TzO@oWrlzOD)8WH=p<|*%%tP+`l==LSsZVPaS*xCE
z;&cc^ilBikIZ%7#t58@K)dc}~ZDPXq@xmK4G+?5oP<+WK5?Z^+<=LmZ^oX_7k7%Jt
z4<GqiYI%2_krxQ8djdgLrKkRgk32r@{%k>QgvJVu7}ptA$gf{^??1@qBj0?}u&mxS
zf5oL=n2S}0fBoLP+x?XM)8HiULbEUNFJ+UiF~VEr^xzST!aBaCmu@RKRRe<^9GAu`
zIA{TUmoF?hE<ZTat}A?9bqXa=V|z9#nq3uZ*Z54sl&AUAf{AcOzCDA?R~Bi)+O{4_
zR(U`N;qbaT8-!m`gfgDP6+-xh8<(6cI2i~_%jWCnrn>T%y(~B)1z#QGa^;ugEI2oR
zKEq5i(7}xR7U+9c*_<M+_ZKMpq#BB1hZHM}8B%4_eMOlESNf*1N+@k=iYTX1uY-ur
ztJNc&UBI7{g29oF0H*nS$HiHV_!oyV*~x4^<{(!^KQe1P`$Y7I>?i5Dhv*P4wFA0`
zghRH9E_5jQiH|QkP%ifuZ=DS((2k6M33_ttia2?w(bi>AxmHd075NUX^i^KmlKDG=
zZGtJ7tQaCHalI}2hDanA+ahSO)6E5`b2gNL8!1#<hH^!bE)O9?Y)4yDIu*ht;D@W%
zgmBKyLxLwGm~p115qsBpN>OHe?s}zmxmbu!7()x?LY2^4EmyC{#sU-UL|=Y?=-g0E
zLJR&p<~=Y3$WDhk?O|}`j2ki@;bkM^g?UDhBEps&!mMZs%ZjTa%b0THKrH^LIYSpx
zSEL=uU_BtzY>zOrky&HMUT$zD<&-+TNrIa7v%GF#_mq8BnZ7nsBH7xBw|0!0ZM7BB
zz10nQ?qa0gK1r7h-3qckNtX<N9ZAk5>EK=Ps`5{+Mw&Z08@iNX^0n+SeO)~JJ~uE#
z(~q!A8yG(F-QA_Wdop<|DIDiKa+>>i)A}+c4THKqN($vLs&jTk+2-FimALrZ<VmH^
z*TVVBd{kkAneWqjsf^-O?`HSAzR(hZK)7*#GGx-WmTf>UY)uj-9umWUer9kv1aulo
z$6dV4*L4mP+N|9mU|i|9#3&WTHCQ3Ez7lJSr@%tSDmDKbfQ9h7fy_d=kURE~R&-9q
zZR({iKVH!tAbH`{NFw=aS|S~02cyNu)rbQP^4byQm52j93UW6PH_V8=8;ff@-w%=3
z)~Xnd%RrLiRF5>0InTp?<RQwRp-R7V3@PQB?$BDH%}7UJrn|$Cgc!Qj(j8WHh$J1y
z|20r;F%BAnpb1cLO_@^Oa+9HNP?i<<It1olR+f3Mg0!98@6<qXZfdKDxprkOO-U9G
zYYRlyW-y-p#v|M)sn2|&TOZ3t4Akq!xSYehr_mHwaW@S5n@n+kLpw;c<zTU*7^k(^
z!~e%CrUMBcRg_OnC-QipTD<~spk&CZY)$c0WtW^eJdXI^0C5oei-V<dWi&9Mm!W+&
z88cwWCuXFDwPOaU1(K?%mNCPMI;FxJt_f#JhKoJ#q@w1E(dJzZa=)lUp4}{9D~fu5
z_4u%&G&4S|R<9p_A65is#)q{cDVsKYFqZ}z4j^J_I`%QyWPXD_5WOIrt&eY}>TOKJ
zHFQdN3-#&%*WhupMi9AMH=4Hewu<7Qs$TK#$qEBdQ$6|aDf!88@v2u(>19$y@nmI4
zb)@Eh0_iGD>QPHQaa{&P9on>lc)d{%_kfaQ_-$8@Y>^;;{z)fO57E_6vN6HV?6V9f
zOQgm9?qjq;YM+?wdO;ZwTWfD6sNPOc?$aE&TM25m6O>y(*IM4}dwzA?ygY^w_q1hs
ztSD~qqP(Iw^P;?3y}^s}isH<R@|ypD`bBxY_46;vtJN!#D=-}PPW}I){E|@C#5eVp
zp**Cx!4SHCqBt{zuKC|?2)#_MG=#3!-e?G2aXCMPt`~1Ngsv!WGlbq=+%kl&#ocZQ
zUCTT<gsx{^Yce3`RFI|*FL^f}w}F?SPyh{3TO5$45oX?vd@TRa>y0-@|JhpuoeC0t
zxc5x)#U}6x5aK_xFWuU@dGgk3(Wxum22rvk0@43P3#@InwzdyA1^QsQjgsSnQ9P4%
zTvx(B@wH*l9WoA=sl<BfPWKOm)BoTmp^*CcH>RCZ9!&f1?N*5D`1rK@2iDir+LtkV
zI2D&La~>3zfHF8Le@Sm+xe>nmSG?S&fi#;-aRUqiw%TeSw={f8FsDhnPjkqe^7E&P
z+avk-7Flx7%t4m)L>7x*?PQrx-zPqOd3b((_;gV)S?08`a(ewX3H*WMU%~OA(UTQ|
z86~FIU#EXRaxM+Np78~oD}^tMpB=vI(?=y7ju0lj#Ob32fAKhvmFGN0={QUJJUhLM
zw1UHt)>%&Ua5aJBggX8F`i}>mSpRg<ZgNa}Yt)EuV}T)eHQ!p|3s=ze8Tr<@Dj;ug
zPR}i_3dEmQ@mHDwPxDB83$Hu_p2`fQ2`<q!fzwYHC$^9^S~CL$S;f;DNC9xqN+v!C
z;lg2Bh4}G_fBUqu5D{s(#Fw-u0-sn7Lwz;Xl?_Km$0{JfG62^x`gAcQMoTQ?5*ldt
z=g4d1Bg%<+L!zR$FWtA7*N6W-6gZ?}+W5%`#}p#=%dZdg0|;Ey`bkRW1x&yFcl!45
z&z@@1h5kLLbpW3mWWE7k8m3Wmj09%zJ*!6p+#YHuG$|Aiq;WwZ@^7Yq9<?oD0l0ek
z^V|2Sd-C;5fkR5p1oGT5C_dAxu>j5xHO@HEkUCOro|juRI6QxKByOpuiHa-mQBCuw
zZy(YWIh<k;e$efd|9{wA#UPg4v{Pi7VqrMNI>tq}%D?g|YBJ1Q<yyscgt48pmrQUU
zr~GG|G3S}$s$J##hb@yA-Z1NWAq0o%lB;<xg2<wQ<T`_E59%Y!H*$wh^%Js`U}>(i
zoai|7sB<x%HA!YEp_sK}L5(-3q+nhzR4nX+?EmGY^kbuHjq(*Y51I4wG`r$U!m0cc
z#!{fLn;`*+Ga-A<#vZ0|FCPsLA(qo8`B<$q#sHjRsxyhl;<Q6h-7-$g@x6DNK|N%S
zgz%D;Ac<a-UzdC}I2kT8&W)0pS!`J)cLXIdNLgWf{u(NB7Tj9{uS4p_ELJ%uG-L3s
z^ALPjpR4b<Gnd0PI3pP!y{Fii?}-IwM-^M4clEkxHtqVk%K22@m;W_5C4UiOSE1<%
z5(LkF0Y6s0p^VXC%%%!TOuwY*R|%=S2+8H9nSf?M{_F-G%`T6MqtNiDxACYjX@PYX
z^?g`$0@H~Fj5(9LHb+*3E6Tphk%jj7SnedrF)s{%bV@>-Ctp+IJkA(%;tU>eN2RWU
z&ne_xlE)*YM3<-L6~d=^;eY(3%6^45Oew!u<viPAG?&9Us&MFOWKZz6Vk@gf3TR0!
zh$hjIIsM(odlzggnUwwtuvcNoz_t_vs5xK-%2yXG0#%&K*^aX04PSGD$(h2WxEsB|
z?uvqmD<-he?<fi}Twh}Gt-M^UUxmGlZu#y?+Hw_NXsZT=agZ;-6n|Hl5iYHK@cT>R
z<%aUlq8~8}6zDi6tRQFW2uCl+OigjkB&eP$<swusm$Fc+Y5o)VZl2V-mF>U?ChbyL
zqTuM{V+_>RmWZ*DpSsOp6_#9D-qT=Piu(C_12w=Ly1tImnt@s@5HQE@MpL}I!AOhQ
zw^Ub*L7=my9Ii(bl16-O%CZlf>BU?$j?ZOU7!#xs4{P+CwvwLefhtil9ew@{ASl2n
zuG9_)x7f7;i{<-L)LlfQBh47uj@h-fn0+k75p6oYm*_V*H3I5@mmfGdGD3Cn#kUT$
zipxkR0FZrD&R}2ebLy`ael0t-=-=E-D*H4lVJI#y#Bl{IiZe%<wXh779tG!Z=GHkQ
z&q6g~&KPdtmy$R*Fn@RSh6E$u1`R5yaB^^2Ef23KC9Dspj>k49onqdK1-L6-h<GBe
zvZrhWK-g$Z<?psRMX8}WD!H6Bbfd4ozq^{x9MF0`*TPQ>R^8~BFTLGwWJ@%X8My1s
zaXis2N9G&aKJq&2p)8WTHkxhf>8YH(Y^iEI$tcKC?rrs?w0{Uy`5$@dK9|IUAouA4
zV$x{9mM&CVCu~vX^#*?Wd%Eue7*D}!z=(Fr-#)a;H$EyvzpflUMkk5JMg<5RF9y{d
z{W52kt7qa$G@-8CIb?L^gYSj2PW=)RgQci$hVLOdJ*&1-MX$n;m!{2Q^Q>H1<}opB
z8GNrk38`^4tbZbZOrL}_z$BmjYy;RDTB$$c`ZF_um8KD*Uv<#@&>in~6X?EhF4^_d
zn-3ND8M)~gj(V%WEVv(fz+)X5sEU}}^b~`8&VJ`S<C3qpvxh1-9A*3sogCC-f^K=#
z$ss*7JySbf9Kc*dZLfzzW4h!-J7vFzLxOCl?DcS<u7B!omHi$Li4H<4l)E|B!r{=^
zTPzQ>a7g^AgTEDax^3LPX_S(#p;b=_?4ax5Qr&4j%}zNaj87%VtzUR!wj^}F5}B~{
ze$at}u6EFY64ff04)kNoiGC>4w{Hfimhg75@9jM(OSlhaRh=v=>p9IVUl<G}2jp<}
zIy$&SQ-5E|JETj+)Y6GxNKBF@=)Eg?tWm8R>5)p_e&_NNFiYiL4-qkiGnJA6+ej*R
z1j+JLAHov9GY3nqpAU1b6((9*4s@}}C$p0ihViC=oI5R6T8Oeo&uvGgPh&+$`PU@S
zs*MWjmMqtTl%J&4<~|PlSoP=39U<-{OuS#1dw)9Ip<WIoz`S!k6x)qH3h}dK@y(=^
z6ZIeq@0ieb8A+lm6HU`rEDT0ZOKD$S?&b1+mPDapX17!Bp1WC956!q4ZXx~LE2Hb=
zDe*(uVN$d8rUf55E;~f_z*gv9HT1K1yWgfmXrQ-4?QDvzRaZi1?7ClGZ)epY(0f?B
z&VO_eip#ZX-=Z{|73gLgViq4A-<C_Fr>dMdr^_O@u96OuI|=J;s!LN0E9XOEJ*Qn&
zdsU84s*0wK1E{DiRZCYTRjSV`1&!`wGv0fJ84dJ%OIMDovD?SK?QT{jC)0O?7UM1d
z5TJ80cV3a&s739$Ys8;XY8}MLL|1UMU4Q)PZp&gz;dXn!rmmzOjf1gVQ+weO%Y#j9
z0^_ap@&~kM3IH@{|EF!w?D&JKY6aOj-z(~TtC+-Mv!{MlTjo(&V*L%$9BfFu{+}&k
z!+v*$Kqs6+{g5^?fd#e`;+@;b`c-ajBkKtMMcT-yL#yg4zy02(C3wdsvX<~p4>(yn
z_zo>(Ey0}*vTpVFYalzuRq7vGV)BmdW1Y0_(>>M_y+iYuMGvOn-J8db9Jrz{{vEcj
zrvR5RdpH%BLUbV&0XCPBOb#ZOLp?a20q?isJvhq&e^wPq<7lVAEwni&!LgK0qTEit
z98@6Ng0JLAX0zaH*7;glc_Fmm>p-FQd-*z}*kTL5o>Xv4JKJ@>-Ydj~omrn<Nf;B9
zv$ky|vAuelrFz4%ZuM9JaUW}eE~hMO$g{r&WwE*UVZ>HD%F^Tnx&~p*37!KiHz;48
zmRJ^oe>U7OEmagR=~!$~Amv9zaie3A0IWz#sam6|V@V4KOB0tDk;&tRa{KIru}{{L
zFsji9I;Fh)=PhXY3tnbt(m`=JCi^Ano`94y3s2VV4L_7NXBXm#N!HJ*74_3)C1zP2
z4wVk({WW~JiQ>NDn~+(nb59c^kvzTiJWyu$e+hfAcJ!k&9a~U9vu5aGJ1rF%LDpx*
zmCX40g{3r*2<!v{RnT-&Urq)6@b`~#s!ZYaDNd%qe9GTGo(jy(pqmqhSX=^%Yihay
zV!i)2W92yzOC&Gf-IzkGckUczWi>WO(h^1ucPyBKLhv_dzO6IcZ1Ax3q-es;nQ!aN
ze>NLp3{KI_m2d0HHoJ$crCnjTIrD9ud7aHx@~jxCVOLvGY|zzK6pwVZwMKt!SG!Nv
z($!X@-?*#YCwu8otI=y%FV8%a4v(kHWu&L%rT^XP9?vk#vhvz2Y<v3FCq+a7FZhSU
z{Jm9L8jPBMn?5_r#P-%_|22I!S(G?>f9*cIUMhQbsL$-jrwZ-t$7fTje!O8{P6m|W
zTVK8hl`KZ}{+*(7qrE?h%!yKakvW}mMPy=77H@s&9%O=7*1cz7n$v+f>a-V^lNncp
zr50{*Ys+@PGQHns@IB+wo{Y<>Qagb;n{qv5qBqKg+wRXEWIAb9h~G0Z-Ra1je<-yV
znbRp(M5ZO>ow*@%2Qp3E(0JFm+@FrjiBfx!Ih|5Jj=X&T4F#DYgogWX6~zYoZxzKO
z`){>IXYIdj?S&=KG}-h*7o>c0JC!SM!PUmy*i?x!#T7t1U;6L2_4*Z2cJ`j?UFI_t
zWlibLjD3o-RF%XF6yapwF`l3qf0p+1ct7r1pRvowh86H)C2YLOHO#VBmgT}BBQ&;O
z@@=*khbT*(+b5G54%#G>p5&?LBn~3GPb9~{y^&mWP};ZcU1kU!gj)mYm3K0x%I!@u
z(<yb5Sw9FSS%SRWmXPahdcjO0br0LMzl_Dv@_=1Fmj=!#8sGjjB~TZme<x$FhyHv-
zFQy=d<|`ukE=kbCIIrsblxFMPAc|VC=9>%k@M7cHKNrQ5^2%@|_|(hQ^UMM4#6x``
zu^>omRUPWXSUm8NLw(hEI>5J)<8$PRxWREtYHp;dg>rgv`S<k6K$+9muIHYSt0aN1
zD=$x<|28DXY!q;zZ}pt+f9PQctw3&7REFbiYfjlbjhg)(;tQ2yb(|5-@wg~cUOmaw
z1*Eb(g0Mgd5VNAq2st+;_u*R(85UP(SXtpb6X3%i9}n`7GL8i_(;4{j6Cn<)oLimf
zS)o{EIuQ;F=Yv8wSLcTxem*Q-kegvX=X;q!ouqk3zCuNO=UYtRf5ff*Hp7R*JYsyc
z$g4ayqDOpxMN{SH63P0gR(!p;q|N7xwv2}!#dvnrvl~JNSxVNg2q{HG5zZ7?a!QF)
zq5GI|kMLzBZoI^ip(KJ_3h-m0RGH$CViXTaA*@4o)x#95(jcyk4}i0W|923ptT8=4
zn)3MZAXXV<D9ok$fB!*qOZy@*4{CFl$FDq`cP=qtzlqp-DU8A?!brB3s(vm@a^8~v
z;QpB_?O3J_x!%mnwH)j`NKr9tpQe&298?F0lqH%@9Q?NtS1AT=Q?W`QOv$=fmx~=Y
zY8ev*YFv@ok^gd^g}U?)e5|_A$_W(bjB@o*+s|?ms${Q+e=$6yvg0v3Eh(D%mZ##p
z^8Xa1PB`xin4%4pTDy$mx+kDf-{D?<D<ID}5(UaTCRw&R67)29657_?Lh)p=SYU1l
zdd?4IhNdl5+e6bETflsBY|9Ojc*Vjs;hkV1FE<GfN(D2iFu#5>`E^`IX3l5gr)(-{
z7yJ9c){wC|6ZmSnPnP;Ew>){0zokU|6MKq`m(fQ969X|YG?yXB0uz_sML5QPK%f8l
z`u&~E&_BLtH;WsU(8iyPCBf(h+>C)<q|{7ZBU7h<gP!>tgg5;89DX1ELSbxNagjr1
zC2a&cEGq?$5M!oGr0{@71$rj00vQ93(mJgZM~3%y7(L9TJeM;5_+d_w&Db2wVMDM&
zO4DPDlrDMYfWqbPc6FzyQGii@6Ka>lv4oigJnSSq)tBl!uO(l2q@}`Jh<hkIyv*-D
zeq)T9Dy+WZhO0RjC`G>FFxO|U>ob_}*BayJFbX4{GG4OWw90dAO+U@6rJygvnuU{0
z9_0BUs(2)!#;I}XJY8N|xM{jh3O-RUBB9XsOjls@{HScMm{zS=kyWmLQk-h*d9>-6
zX%g!AyDsKZR*W-LM2e927&bl3meK5!Ds7l28lml=O2PQ`Q${T_hDbFTz3?X3eRzD=
zeS3br``?6Vq_xm+`qz(l;XQ#C#`+UzFG%9)r~jP3-u<<wnjMQ|!#U+2TAqIT;ncnP
zTlSk?@&=_bTDKeIjq%=pjU>T#Gl=a=){&P{lMot@Hm(qV8y7Bbc;R6cwnU};WR=Uv
zzNGhqZI6C4dW<P<X9QqRP-U$xdZsF^#!F=!Szo9ar%5_#B#h3KzhR~s=x{hlvfbwQ
z7PGZOW(Qrt>_s+Y#&#4VQ-ikKVo@181(4-O#NNEAMdly|)Q&QL$_Cdk+DoA<^`j@e
z+rK_aC4d8BM-gCledO@3JWkekl`||JrF?fo$h-;{v2n66cH$y0yfSlqY30M;pEDNr
z2U(j?CoIjI75zf11(^vKLfR5nz|lHlin7ntl+*%A^;AQQ3;{?M*=ltfe?;8Pn_9Pu
z9avzk(Nk?nk^@+O;?&xn0<t+rb(_Vq9EqO?T5L;EKV5I3hJ+*AQ(Cf6tAx^~AFZBm
zveLosTdE7z0G@=wDK{z0H+*nk@|lv#G`^%*aZHrPIlA#G)W=L|MwLRAmz@3$Oi;v8
zT&WEv++x^@tl;y-MAThKqt7L@?K5m^Is2H1BN=u2sORf{xxqr#ZlY*FvnrvyV<nzw
zgsfxjYKpPMG71raCU#b(*n3z+$l0u!FrM&sd<uoL6E@|1xmBEP<}cz8(Q4{#I$>NR
zUSdklPnG@4+bL=iKod0tGpjH>A|}O48uNxIq?n@j<GSHkRGmcfNufTxWtUOTE2cUb
z=@g%C%S3a3?3?(t;?$CVb0ewZ)54)71m=l2u3(Fj!f|S?k-Y=W#pZQv%fNzT0U9%g
zTfhO!&cWg6^fBl-V={<`qcbz;ZqIGVDPiT2r;(20E%}A<0@&I+ic9%L&T;x3y&=B{
z@8*++XnQGdz#ES^)(Z=E=jM<{c?%QfE;ylpDL9pXEoBP~h#O6*{9&6@Av08aC0CM$
zZZs|y00Zc&F%dep%LxmYB2A9a-RP7Wz1>F?tFtI9>0>Gv$**847ROT8hHB5bnB}mQ
zNM0McR+aKR6k68w@}(^?hCG{(i)4VaVRJO=Mny~et#qxNb!Vd2#}|T=X3Ml88EL(%
zHPnEAz5nxvRBR-gpTb$bfp*HDKD3%PUjt;Y%JFOVempT8EJEEd>fzj9#*A^X;z^7j
zts2g!rmp-B6|-3{oUdX?ax`e8z%d%0QNXB{nJkvu1}fUWeow%sD&oYm1^W5LlybGQ
zXV9_keHrV_#e-d2<5e8W7(rj;MHC)rqsH5RJ!3t0Ap~dlqO2CytJ;x{jN9T`jtKCE
zqG7Dcc8kN8nS}K1hv8C1=(jjtOYO;3O2x3FG8P2#mDN=N9i;$X=6&~%A&RkQ=~kKd
zx%+q#PGG(#ctk-m9M(wRDk=*C<wW4oDG%b(*-hv-h|ROBR`kwdD`nJPY#r{cnhm&r
zcvE`3tQ319G;pJ=l+uDd%dD&}bB=SMw|;A_i1EA0rFFPRBsCTP#}^Jjo|iy2l;(t@
znVhc4-G9wN+2HI9!!eXhGvntEt8180nZRZS<{6e|XYfHl$##vumo?PMaH^nOcm2ir
z1Bb3qdUQTzk@x>Di*WQ7-hpa4$qBiC^kM6q*5-1@J(MYKb`{S@4yW2xzF7svPKQ_V
z*8szChex&rhO5BX8wgPPDqs+93Zb3y>C>lzEu?BQ*iZ}dCSyzLeE*kK@gpA__&C!D
z^M?<sd5GdrWKE~z!L=#!kEzuoY5RDWopZ0nhKitr*qcqhWLU&u4Zl>~q|)AhknK#3
zY5UMCP@Y)V=F5!w*r9P1HE)+$*8@<nikcH!;h^VU-Q+O(hI%LRY1LD3ZV(UW`#pv7
zmYuDe^%V1JOj4yBL^)3Qb!(FIi%?9iTj%oahX*j`{!@6^^ccQ<n;t)}+6J-HHo-O2
zIM$7ployTy^l&rLd6RjL60rq;qPxN{LaiBLY8R`ka4yq(-*vfBXtPji-owxH%V<t>
z?N$6@gG8(6`L&=OR>d}^W|eO?)Fad?$Q`R58Xy*1uh_2U(~S-OXa9CJcwwMp>bg(Y
z?47MrR{=xTTi4>kRkhWM)z+9R)ON^OI8j&Flp#wWS#F0a-F;&V7UTkd(@S~V&Kll~
z-<Hn2^9q8@We&Ivw)_9ChbLse<|eyYydKF;cGwg7vfqL;9_gRzrmm^0t{cSSoaeLB
zkEM0YKGxx8x|*ABHmiE^;)<@uT3x*@1*lLIVGrE%Vv>9=xwBA?ndrsN7BmGiKHtAX
z6gl)cprsDcEX&m*`uSmhs6r0fKF)V<?Gkk{BzO3Rq7trH!&0ZH7g9|KHa&}NqiEV#
z-z>UQgrPr&m2hVl;ii~0r;TzvNZK=>JRXZjDfTwVr%fLFhD@q>wNMyz$R|77FAEzw
z<$O4Ic9Y_AYOJXZ8Re*6TTZTXnx%uUG3D}Y-tBBTi{%V*HHYPY+`x1G(d{@C*8_&C
zzNCjvrNz|Nhtw8kxd)r(LVV(p*3Tl*4HC17g!ueJG(l3zBiGhK4Mhjo_q#ga*S|=+
zE2@wi&^x!g_N(07?%EOjjkLQWsPM-FFPJ-u2e!L*lsDR4yU_=<yLJ>e+Fkqg->=<u
zC$8M?I>VQDY<KN{Wp+%vYe)5fc2|@rK-%9xyDK;;!1)6$wx`&JLcHEyEgwn%db&&Y
zn%rkM_@=&B$Hz$3Q(QFHWZW}MIIiz?xKaP*`d&3>goGPRxam$J{0A4u|6a;B_Pw$;
zNIMb7*!OyKb4R!WKvfeyrUxv_I5i>kGCb^R{jkoFbUlB6SkGfjS>SPjHS?V!pq)UA
zhqfRz<C4*R9;dXOB3E~d)PQ$(i!@XEZjl2*fno=)kL(WYDe9v_K}&Zd6tshn2n8+0
ztx(W0;r9{>R+(@}DCneoP$+1r9vKQybE>Z2D;DTKV!_^603=A6Un3Uix5a`A@ZMNZ
zPdPpocoZ5|j{GOMj-WXx60~I313^3XfH=?++z1051AZq_V3h?&1c6S#$Hah^=z$@?
z0YChEMF8_h1lSn?EGyct;r`8=-2Z@fr|%!f?`j$Ep3kp${|7)4a6OkXdpH#VGMC|v
zJr$S2Kp-ZUH&Hm5f0Pd}Pd|M8{Xr(urSK+v3~%+f%g4ii9|lGMg*Lv^&ImN7`|;<8
z@B=9=jP;#IVaRvC{JZ<|@Q)hmfJ^e0_(#j`m#<y*<hvwFFIkk*7_AFYf)QG~&S)o)
z0Vh$?XDKx^P4E0V^NYm<s$lZdyHm3eHS>bLhNxLo^8NH|e`F9l+58Q{Z!A_~Y}!T)
zf&%UNPtQ{EgjpJ}c_*C18XL%h*1Q^6anM7(8ZVK<34MA>U-d0eE2mHJvYBqMN`P$S
zuJk<JRtqu>-Aut|ybgu8>Q=nY2>{b*hn&jl4(?HE0o0sYWk%y_FUYq#Dl4PY*o}XB
z{Kb>0)?}?Gx6e{I+y?{<8V@y>F?%=_m%!jU6aqFjmr-gLCzk<RIF)~Z$ANwR_499c
zGNUeqH{oaaYVY1Z@BVW)Y($~Z#-EI~0wkP%`~5EbL7;`P{-l)fK&L<cb^3Pqa|^Ys
z)qF>?IbgC$W3;Zwri{?qos2<&)=tT$6<@qb&70JLY<$s~f17k4fAfEo?<H@X%5%P@
zIQKbFG)#$)ReTibB8Y!67(V{^erle4{b?YHQp8VuizK{-tD!PeP6`=dD4n2IPZ;W#
z4=?9Pp)ioPC&jstpjv5(po*AJ{8s}-p7XBf__Zk0MHG}33b4QmGE9NkLF&L3`tn_*
zb;9$@2P=c;b(EnBrq%-Z(G=$Ww}&)^@tvDuqyW<OlwX*~d6s`s^Rq~>BGj|~{nv*_
zFy{Wh>kB$wP(imKJgn}_8DS09vs77#W~X=~0?n%_e=k#56oj|vYW_+k94nwfDV%a?
z3jDcXhO={kC5u(Ln)A=q%_y)!lc_QLo~O)9i(=ED8G0iv@?$C`L7{`f(rdA8x=0g6
z*=e7;;R$4O5LAC#XZ0SeXwYo(<&IWO#W2UGRp}$+6(E{|ZJLv1IqF)sWAtnn8ukUa
zug`0ABFP@Y>vmSccXd4iwoHwLJx#D&SdUq0oaxr1OcT4lCX(>FTeCN`ot1qwGI}!^
zeyLzDsr-`f<=3$KJnovk4njvOC_84)v1YfV7g^l1TaACLj=qK2Oi@(cGx7oHmxOr!
zhb9bJH;~$cK(d{IJXZo$d>V{`sdW~?b2gq(1k^rF=es<=h-<vuCi4_HVZpZe#Z~5o
ze^tRX36$7E%Zie5MmbS8{>Haj{dTfWTNE5}47>~26(hkbXe>0?#~*e&<kZNYXE6RL
zi$StYGTVO;n+f~T`St-QYT?+PYXmmll!3lVJwKYl(zY2LNy7GOjd`HiYP0w@Gm~bi
zs9BA6t&+llP#zj+FCVY#I5-rx$3EEfwCZT5H|<dHLd?;2J3JZR1Nsa^7Iu`aWkX%Y
z=rYDFj|o~Lhm9>Vj|~e4Tzi>UjXuP@dgdbY^ge&))zBucwP(W^+%0Gt>jcL*<Q;?7
zgFKkID6^l;!4<G!sb_c0EFNIG)s~srBhc)a$jo)AGAvFU0=Jl`7#$oTgrkX7J4}db
z_5vXmHdzs3fmjC!(O`0%5LOEG{!&6zvyVxH*&z{nAjD?&l2x5FyArck)$7qmRK1zG
zsOo=am#R0=R#kn`{YxVZb%-68&2_osMAMiVgjZ2BPM~&+xal3ZZR-Sb*r=`Vn7S|Y
zMIzzIF5+l*;7Fsm9xcEqc9hOzSAdboKEQ2MU{b*$e&}4?sGfYyRhjuxM5)S9D16wV
zm_=_nR{Y3{5%vsC0i>)De~k;5CzG-YXd{0ReYeVajPa87+FkqzDLAMWU<^rV)Erb9
zxSdv4N=6y<J%`4*s*xC+(}B$305iZ*rLM9~?@eNJX0J6KAod~|3Sn!AeT;j!5*Dme
zsB$)tm7KS!o_#^E!1d^kb_XhPEc}4Bg$OXazLJ)Gk?PiWV^yF~ceo$ZphI}&s@s1~
zyvPfWW{xgx4x-;PE?3kQt&Zla<(bbbs8Ks+>c&elhLVE?!WJs4Of19>^+x_{{D^p0
zPo}b!^#D1&)Sh~2Nt^>%qSV@+0%miD>MD&z<Jlye8)|F=F<7RQYefsBj_a$n!T_{z
zm-IE6&t~uy>Ixi`JBKaAMRiTe-I9NmGNZ$#w(2O^v%7J8sdU6K=jUR}V+}EvQG&$I
zRRBiXUqJ-q_>d9VAi^~ojx0Tj{eR2l#<$VSP&M79;Vcc(cI-&nJ!<%X{{@<^Q+a9v
zd8+ME(<|onoGW`QTnmAhe2q*Fu~$U2V&dh2Ktf3EkiIFrDPqAy^*E$AR0n_2Sf@(_
zIZK-4^2elyLrtqW(zU{BcbkAI)2SMlwC-h`QvAfgq;n&a1Sc)Ljab!|*K$fTMfm<N
z`LzVm{1hH)pq}!lhnxEX#n~poEqy@^F~3zfWPVLbU>ZnE*57g>r%gK=->-B5DF$qy
zF1P^jB#i2*c~ch<ekXzaVJ?3l=o#V$5w3CpLAqVf>MLCUhY#9x>=<1@5Rc>3U?ywV
z30f{-w<3WdhUnyHNUFEf%_Wji$PTE}vtZ0-i)QmA3b`hlj_^aQb_IT@gtHx)ChsLj
z!PzGMBK{$B>8(w}j3ewN&Tc~8qJ{W&IYdnYs1rjlvjRgLRd80aC^3H*#3{gbbbNgj
zyLIGfBFHuGLG55Nd>&P7R4He?-^R%<M*+#cEd)rx`F89$K$g%(5b^|$*S*5IMD6m>
zE!F<62`)2s$}8aw-5fh}8|)k>1ybJ}J3E8j^}!7Xy+S&B!{DZYcuj!8=?Sp9HxQQs
zj2z^&JGvmiC_QI0Wi@}O-jQBY@3KB-Nc%RY94n-Sj<^d}2vxEwTf!Cy5H}jenIVB5
z_Fok#^dvU82?<C!4hdl=35wHfMqk!XcpO)@IfR|QG$plhwBh)!liV96*P+M=>b5yu
z>YN!GRc@001~d}LqmFM?q(?Dblj`|&SfLYaR&GzaB7oz5H)MaXPLmI%$kR>iOp&W8
zZHheKIJ(}&6PgvLUQSPfCYtPcVrhrIDJ*?hiNEDeO*)_w^#EVqCd7ADyX;nH5I7Rj
zj-5d-)yviHm22uY0emfiT-8vx77*qU+f9Co!&b7`Vz$vJr#P-q?VPQ5u&4!iXlEU_
zmFt*!wnNlyr*40Z3rkTBD$aFt+R#^AwZ;uX0Yj@^>FZWH#OiFC`EppyW%u;!axVi~
zPEbw9f?o8puKPvt+v-su4E<HQqt4Foqc}lMrzN``8A2OQtV7v=BckLc=KfjP`=A)c
z2*9zAV=5GScsDv5V&#(%p`*fq`4Lw1WhDNvL<WP?_Tqny1~uL?YTF<l8emFqqkRTV
zVbODLL39%y9DKbG(y^b`Kj?e~gcKInb$mA!r)Mn^6PCMx4@bOc1qAa~BYf+#7WsW`
zOO}IYEkdP%bPMIGvlf=4UE^ktS|AQbEp{EVAc<3OJ5E`kqiV*kLl#O`Sl|Da&shBL
z;StS!ys>|q2QHjIUZGig<Gh8}Le&)8IBuaN2~YJFH%?msum#m8&el1(Du*q$j$0sb
zI91KrIBbE2nWZbdHqKfQFtcnUEwM=q&D>E7r2eSIYHSc*{uyc(pp6D`kGzi-tHIR|
z@;kV*^BC#I*Y@W!ilQxgR;7wn+oLc_wfli#g<fK3mX403e^m-F;_;FEBvZ~z93G7k
z&u~y=9KGHm$nynffem5wdHw7|s85?YVy<`n@jZKPY!_QUl%iwHRJYo`M!Dg<1!o?h
zZhb`Lf@34$D3Ze#q4sr4{$}s|e@7;}e3vnMI2D&c1~3x>GBuaMiytSKz^fe$m%(K?
zk^$JaRAxBO0e?dH+Gk;sD+VYVqGTwXlV$NNRV9b<D{x)ind(&5ek)07#avIK)Pz{W
zX=_TJvnBU*o1JRGP{^K!+EDaxz81d&$L-V@riEzft{G}{noZ*y3{^PGtd%XrRZC4~
z-GG2e&^i2XCMYR(+p)b$PCT?ZKV=V&YRpf(#;OIK8h^3B0R*Xl5hE8sxW&GuPje{h
zw^D8*n$MeJWZP%pqz&B6#1XA~wD7|J6_&0DD}__aQ)>^FUI}k5xje@#HJ^6z=OAQ(
ziP1`k*N6Ye5bV&tDZMFU-bKxM2+SU3V;h%1auJ#s>Wd(d5==v3TsnqM^+}TwPF1^T
zb*HgQj(-zHNasc;k(?HD#*mscp4Vus7{33@L(I_;m%AA1BI+)GdgvYtgdr$R&sb1X
zROeo?pr(5v7SxS<#DbdQrC89w`g@55n;>+F1&x$D#e$k@&se}UX{UX!SfKuh1v_H_
z`ADXf`(MNYbyX~wGQ2YuR9*VV0yL5tgpRSGp?|2JkGo%u%%W*2t$BZHMV_Q>EnjX7
z0dXUsp}G<R1T8Ez%+c%`0{s7|kY6nXcsrBw0th#S03<E9b5Y+I0w{e@dM1uB1bFp0
zjs^)?*_omVTvQ3^bBLM#AfN8MZEiTOL56;l2xoTYomt#?+yxN=GO1_4mt5@z{v<GK
zW`8iH4eeTSE`z_yzu&={dz(fWN5qThn00ey|9Cw`g-qZ~4TWA*7*v+}!>Yy{&r?tX
z8P~Xq_b&<f=puqarR)5bB)JTiT4f)VH?&{+1ljd0Ao{nJ1jz^AnK%x>B000PPCzA?
z1zg&+yFPV`@ZOSKX6}%O?dsm#nLgZsV}Fk8?%df0?m8ql<SL!DJ(HW7;w=e=c1(D4
zuPLr27&*ylW^_e@;qNVH4Z-+g;s7_k_3&>2h1oedu#)m$F#~tS3wfTrSGlBY9)Pem
zd+{{l+Ug}w&TmmnuKXKNzO?ST5`Yq3cL9K4DSey)01Gdb;glw9CvG^o+VYYdV}E;o
zJoP3j%;8MWnlgd0wPL_$e0N%AOGkzXeOGs}P@&WXbXf>{HX_NRA#hW<%Z|^celIVK
z;~q$V<gzY~(><V$^*-vMb@OCCJD1JV9_^}mx%Pl&CN142H%m#rU&4@gj*7bUWXw%h
zOD&|LzG~bRokFO}oBBm}pX!99^nat<?t)C2pGq4XuM@K8wCuFoMJtjl-eJSw>9v&R
zNb@8VCPy2JO6=(Vjuw@GlgyHzX^TpLIe`~yN}AtLKe7_zv)>^f$S2#86<}*<Mb$<#
zR)UnYTBm#2?;MY&zFLf&a>JPTxR4It@a7#j6LhPce$lI4i4sUJlo^QS9e?|hj4suJ
zfu@DG##t?xeR>3ClVt_CbEOrT4&|vDe5Yv=hYX;mcmFB56XJETi(>?T-Q^!7-_%RP
zVlOQjx2o>-%U>TJh1T~!hF^34<pIony7#YbCLD%rs%1hvY|WMeGkL6LwvY$H#V&%A
z^R#uBA95EvX!5tx*76pHp?}B6GGWp#rObTE#U4O>*`~{Xw+A81l(g!oy8QXV^rfcd
zd?UHf51UR6X<=w&t8YFbgitl9)Dz?dYVf;s4Ijp6jMM7BoIjkrmYNTv*DkJgIeJJz
zsXn(Y7r4s}Z8~kKf$0Uxr%5GB&V*jGZS7uCj3}M*Jg10l&e`vMSbu9;&U^u6OvQ--
zzyF)k^^yUJTwxhU49l9h3{NdnzbZ${UPVPTG74KXRi=527gWXbtp`%_ljKEOZZKz`
z+tsLOn1uOUHp5VjYo3SSzKHkz6)#Jpw4=7FQ{?ZF{C&egDMsDSkU%0ml*dBtcKt9d
zl>IJRL%Fh4Ow<61Ky|;*jYWU^<OS5!QZdyIgAOvu|0eN<?gZNoX?uBYa^+>JhSr6~
zsZJU~iEn)3_t(zT5egMmjnv9E*HbTtBB_0aS_Q}bFAv*Z!nWvn+m>LHW|jPPDw*OT
zHn~9o6kv4C&X1eyAO#qwV1uR6R5Ey#P^U?EQi4<>n}j`W-Ka)U8D4)o_Y2chQqX!>
z)YJU3IZZ|3$}CQE(RWemYn>haWMa=1ubV2|ck6vf+Vhss#B1j#)mm<L*OrR<l~^)z
zvr90sR$tZIAH^50XgYyzl&G*NiPf6(^0n~L!dc8dvXS&lm4g;p*~e+OM!B;VAGl()
zWd91sDTJj!Va@;Taf*M2zYfWc&lS<)jPk;{ig9NC>6Pay{C=*M@LuOCC^3>SJ61vK
zVvX*roT?B$)i_i@e9ftfc9fe>RWt;DBd00^C?>}u#||$p8j2myRWy_@9IR+scQ{$m
zP`q%oq8<PJ&Q|Q?D;=&_z~voJSF|$gbG)LV+TnbK5XeyK-@t$Q3I|Myo^Kj!f>-Ea
zuA-_5T_~ce8+WRqs)=4Gp=yBqom5b5LeQgts*!S^`l*^|r}C)=ip}b&Mj84QPc?+E
ztDS0y^edfe2)|F|)TSJ+DV%Bu-d;D=5Zo@CT2YnbF^AF`HP#gWQ8X0`e+;>#?o~gy
zP$#hGB_6Giegy=iiZmaWF?%=`m*Y7t69PCmm+}1tCztDQIG=way`KN^<<AeYs4hix
z^p<|{Pj6oy{{1jaL}_7Nycn%SG;sOnFAwww16DX6FUoi!y}tbV^X2QqU%1#|Fsm=)
zk6*uC*f)PIwhVx_l+IZ@v!#|IfVwyrg{D7cTcSy%R%92l7QV+>8nl%~YnW?(Z|MCy
zzCvziv7PbxvqgW)NU{#LkymD&ZG{`{Y!Di!eYSRPK|6^<=O0#P?PlMgc0asFxoAt}
zgtg(~v=hdMs;&8qBZi_)(N4i^)K%|TC$x@@M4x*LLrT>jN!QsoQ%gD{g7I9Ek~2Mq
z)?Q;YINn;nO>KW%hbV>ihD!xl)*%|<RBKIB{75g<!=Qf;t;+tj^xl^^FxlG4?_q3j
zn^<`!Aee>=g$BhfiLD_+f0OUePfo(8yNsX>u6eCDj35QVLJ2x$v&7jqoiNl08wgO1
ztBWygozY8K40cy2YVQ%N8h4Pb<Tbza+i`24lWk}M4a{Ol=om1Fn}{eeJS@@lZd|$b
zn~up0W)y#I+I1F7Z0$3VQATLj7~ZJ+^}_@ad}F2@-z;#pjlZhD!MxL=4Rgj3_ZkyL
zh;6WczFnd<0!$Nz;1&bMVBRoNtwzj&OjNO5Sz#fL<Kjz>5Rn#C`J5Nq02#5gsu-m*
z2HAqC36p4PzGi5#I)CQNHsMymFp@>z^>bjxx0Qc1NExtk#|;S@a2=nYpZz)PL2BVJ
z3eMIpoT8nLN;soW*Dih+Zi8Vcys{^98z!-0?G4<9N!s8xto2FUhDqJvHr$&3BiyEM
zgfqAecl9~khDkk%+i+`ixebqq^SBL{d>gmnQqJQxT=MtfHhpK@!fm+3`?(F5*ylED
z5p#c>+gKZgckch(h9zi&D^vQ@6ZFeD^g(;Xjc~elujgoop%p?rUAyb|Ave;9=;2Iq
z!z9{irX9fZ0L{Qm<p8+cHL4SwAScWl-^FrpDFn?WuCN@1yc@J{jBqN;$p~P6PnKhq
zKt(o;a2LzL=uy!%UcQs%U^GbIxMO5Fx37QpvVIIQ9tmnH`J5G>oDA3$bWwKZ$b0jP
z+7W?Mc~j-u9G8^+9n&1zf*dzEgGjiy7ZbFQ)=XFZ-FBL_%SyeBT`1osbe=lu%3{+X
zcwzW?3>SvL-kF)wkAM6?U9pQ<?t)tZ2lFm}_;7YkFb)vJ+=)4XNxTP+T@-*6kV$`B
z+jN$VFQ3s3a7^DVV{8hFygJo=lT#kqDEOU|Qy2vVYTURh%#?JZ2;YJuGC+c{H+r98
z=&eqUW<7o)5&6@w=ay5jy3-s6c^>GU)-VQy4wko0I9FKw5X2P|t|hak^m$yp5Cnu)
zzXS!#I!1>Hu?gO@QjwJc$<nFlmll6Uih*+yMO7n<JenxFQ9ZDtZeXt@4jqMpaJuP+
zz{|?5#NITC98De*#rGzUX&0V6R<7HF)B(RnC=D)=Iz-?clS_XY&m`J9hI;*vgv<;J
zz*hXN_DptMM)Xn$p#w5Hsn7Hn6-7)aO}6E7Ifjxkm0`@aczJhJQsGG-{|tZD(A2J3
zkvvGjn1i(v8h7KES({o3B_f!OQVI6ER~AuVL|dGb8LtlX+_Ds?=^)J(pUImKuh~fO
zdExBh_8mpd{d_))K(dd`l&v6>XzMxY?Yh=&Ba=Kv?k@cl21evE61WW6IM4VvF`-QC
z^>_A|q65nSCwfeI5K4yc;W2-25x~^F9zzPDlw0)KbqOp6c`IdN6)}`d)TC<lGEX=b
z5T~S1`I6p+va6|iJptD1@Ry3tvS4b-uh!K=irO;q;yCmhWY?C~*Ky}9lf@>S+PZ?i
z%Sl8<g%l|8StjBI1<Rp}-x@S#5yK^<xkH_i@62IowkawVREzcJpaXx>Rb(EWIjGQ`
zgON&j30wl{pvu-cd#mUaCUcmcsHFM0!r$$6HXj&1)x)T?_Z_~Cqx_N!s()u5?OI}p
zSxIYQUWj7H$fNB@Vs)}>6|hmD?p#{IJ=9Jg)Gi`|AZ1^aw_7U(xW2JBlTz5|YMet9
zfmp`8*AbqAn<79U<F<d@v)^@twMypCemC>AkVnT|M?{2DxTM%B^K;WZ{vDlJXey%T
zVka+boU8o!+m9cfz`4i&q~HGWU-gThfBVqYz|v`fZB&4&f?e$8tnkQAXT4N^AWR2f
z+kqtXCmZPWaPHy9j~{>PkA*bE5m`7isl>QhFkvs*pO?YFl`(%FotGhQ%FXfd<6D0$
zs#t<>bY4c`c1ADE>j|QI{MsMVI?MwsMEmhcd-uQq=JCh=D9#CFBJ(aAvm)YR*{r_R
zF8bnde@CYsvo1Ia!hi-+Fz18EpZcQ&40O27bO;whyCasf(h6<hh~=!L>)Ck3auy@j
zTR37li#Y4Gnp=Oa#_BS#(j}j9Y5~woGMx}@p6FDhaL@}?32Ez?lBn`^TO7(E#u7ct
zttvd5LQYhuwFGYXWQNxB<Wc{=l*pAEQswg#eA<;RRlwDV)Qrt>AVVu@D#S$Sc(0nY
z)j>+35q`iyE3Z!q)5UG0Me?bpXCLfKO@vcgHMi70{epi?iJj}WL0aj&vF$Z_+K)9P
z#jmr}oxf@PUwfI@)j6FmO4sU~<20QX!zK|TkkVjQ=FlBKP=;+*PpM{E9S<Wb1h?+f
z8RgHmm{Z-JPF5RXRqK&HaUXC1lfQ=0_z`A)gvO7s=tl_q{0L-6*9;y;HzosY@mXTU
zurh@%K;VCFMcQWa8*WqzK|M5DmjkuNu#OH#N4`8Yo`xmk>2x+yt_U(Cu5A<ehT5i{
z8Pc&{Iy<a)Vy*_2<w2FDT&v9Gk%qIG3h{9~tITYBkYviJ??b<1yJtF0jL~o_ZGjKn
z<l#<&^_nEFYTF8u)k3d*ZK4$lcd?B&$6!T)FwK9I_MnPJ76Z++wopaeR1U#<bpxPr
z+Z<=MHAO=?(>kD!HC7@K_wIDJ4NJ`$5kTJvBVdqtH8Qrn22=-#vtd0qyarASh_2~X
z1K)=3Otlw@YL+9l;?I0_fmSpLf8WOCNES^9ZieF&w=XXc9R=IK*E~OTxA2*-)r!SQ
zMsa^brf}1k9FBte#}OgjPBZ=Nk>d#D5K3_gW3Kb{DyLp1QW^5+$!E9UUy?ZBST0?z
zB$KA)AT!VVwmTVziAWiKC6n$@l-qiIb-P3(-P81|G$7EqUp!Eyv8QC2S8ni8u9I`%
zTJpN@th-d^F@g78+zq7%d3CLo9Nb?LLx_K)ygIQYhDlr~D>;0v79d@XGRbR&lI7_Q
z*7~06m2!#2B(7FUdZD9g#t-BDc(v~EdCZcB^eWd%JN3ex6t!Ej3oora@fwNyES3?}
zxH{5P_xqNbmPo3M=2B~&!Em*7yv76!dJPtz+|u<LebG|AIgw_k7dC^k7++^TwZ(rV
z(tk%7Ycf^1xv-hZRr2PiDJsOy1+t)N0ASu)zTaFT&BqOQmox#;p|zV<*QN&~2eU6e
zka&Oa&;=u$`QHqfz&q<FO5n_e_6y?YZEtwAtA@*5(8`lt^<lqx;~PS&=9FEd)|7Jp
zsHu>GjcJ$1q}x~Y`Tb)mD)KQqN2NReo&sms*WkukHTHEYsWB%~7R<LleT9U@+VR2i
zaK^(|Q_?u5ms6gK@|f3GpQ77SeT6%T{{kU{i}{x^dpH)Cau*H~mkz%rDVK1e0tA<a
zdpI$F>J!B`If-t*G`jh9_wc&=_=2|aDiC<nynb$gKhXRNnjbo9Jb-ct&Fk0Z&rez#
z$JYa2R%=AQJpUc|PWwBAz!c!}1^YX1;Zsj*9t#apx5~t_I^QAh5SY>juM-~c222O{
z`@dd)+v&`@R?($D)2)8@{JQ&hcQ}YaMK1D}Hhef5e?~p*McDK&-@Y`}lRvU31F$Hy
zb3`LiqE-P+<6KmjKI%jXwmQ)KCP#=A6f&y?k+vgRuPk~=M0Cpfa6ohwqGs~$g>;8F
zdcuewU!p!KSK|~>XdGD9aM{2<X-#ZP_JyxUzS#N<j5eKA6gh<$D?nlsC({8ao7gHZ
zEw`g5f5Y>$wn3W;%;*s(2V(D)1E`4c`SM)@v~D{Fs?^vMI|Vg|`sMOXqgBq-il{W*
z69*LnZr{~#gu-Z+?P^jOvrb@FqZQDRfc@!RjpgN0U{^FMLRh#fdIi3gu+5fti3=A8
z)2JMqa@m%>ivzW`#mo3k`8`2$nEjp8De_~?e^jox5noJ7&nX)e>Kxzs&SV;%HisOF
zhLw$iL5$d~T(FZtYz3m%${UhIq^P``GLBV<mR(i*NyP1;oA@3=6pY;!BDvPzvOU@r
zVLV6X3U?}wcM-KB=GsT2X$zpsLu(f;D@1_Pp3ja7oV{zYvKXz;m&xvEvilX+q)eCH
zf3kJkOi!`<tfhI}LxH5;|Mf!`0}^bzENiis{RSZ*8O4mu2tg~Z^0M`@W3`B*Cui&t
z*?J5YQni)`F{(Y$eU5Up$R?4kLHi?hBw6XlbglQ$;y+x*MjQh7Von0eA^4!XW5tyt
z7Nt>Q?xEE&*v#ihhb>qle&{wL?B3V1e`+_(n<s-y&mh<LbH+Es8#0#i#xV~qvzM}Q
zUM?-<Nzrt<`=AHQ&4s6iv{li=(BU`BX++mK-F?3##kHjGgOVOAle8SOm?zIxQUzY`
z?1<r?fBBG>@)CP`iRUFSUh>li=~Q3?Dos`}6;y~xkRqv|!abh~Dq?|DP$ABwe}Y=9
zKS(M#F9eBHP-|Hx6;!B#sla-cOr*EU1$4^=3v+>Wz@F!>kqc;5E*LVrFc*xLyg@F&
zs0;+jTrfuL3~u#Ir$U<<>QvO5Qh-+IZH+jQ0?JsXmS&X#XeKJOE!2EWhu-eiL@yIK
zF88Iqml!cfxZZSZ=-hB3Yzj>>e=K-~$A+^doOjdtEeWUarrEVDh;!#fxWh;agAwO0
z$Z!{|vPNFB@dS6<!!bI;l#&zn3UNzF%pti#oC=8^6&H9%Ct6VzlV3f>UB+t}<5tx3
zp}26}Juw>#Zmm@eaUsvBNf~>@Gd}e}Ht~(7?<@pMU?fmv{rgKI({*l_e?(T5x>Ia)
zCGe_J_o6x%E$bl!$vSszC+$Iorjl&>L}Hz*Z-<O7tZzq4>iTxxFz5agj3#nr7DZcx
zgNb$e7wM)78>1FzTQ^m3I>R4ZnP6vmF5(>8C7C)1At>i>y1cF~g{nr){ts|d=5>`R
zQ)HXS@t@31nb%e3+w>&He_It$^4|qDV%Rhys?cUeL>0Bbh^RuG84=ZIf0GfB37bYl
z#-R$HWWR;-{~r+<N=>B~A0^z&#9#|5QV{3vL54fN#~8)Adz9hMa747?%rPBcYnc)}
z>Zo*pH(}DAW^#M(fZ!E46wDnJqQE?>LY$dr)hGJC^Q>jOmU&i1e?6Z!4jD#rp0$kU
zGS8~@`+eqF;GHsk1b9Cg?}wacflIok_dm}XGP-b{HCiG+&jPZ_1b6GSuG?wJlx)c#
z(1e>Zud7U%3Kuv<qq|L-udGa&h5!sGdmBz(S(k19B%@E@>XdE@eFqA)Z_;Rnz?zrr
zGsl1eZRQworpv63f3vE|oND;GUSFcaGd{LI>0NX*6R`bKdnMS3{wZ;k%nTRKL-Dzt
zA-7+keJ)39fAH47%hKwe4BOpglKoe1HwwjCQtrrrCWK;bR3Tz56de0wI$#n;@vWrO
z!Ix07*oLj}iWb;=?#--xgxn9mC?gXg_vP3ZMaZ*OcKCk%8CgS<jhK~|$G<Z<_Y?f`
zw7NgP?*0RT#pIU}G6)j{F)%bWm$4886PFw$LMVULTT5>vHxj=4S7>iGu;!A*;tO#E
zjAeU(y=8W8NlwX3kV6ieQ?S2&i?8lxHM^x&%Y)s6EU86S73;;X$g({Ac94gUch8^i
z{_&>q;Dk5I!Qu1QgQ6cK{Z*14T61tp39HrN^B;#FpQHp!uV;Fplv>lv(PvNZ=J=?U
zrx8HPmX|s{I#f^NSUA-&+T>Xf>g@8a3I|Ukop&<QBW^0h_v6o>|9K~i=u#LjwLeIq
zjrEq^ZIE#IlYZ_0^64X&@Pas6e|>nf(F`t%=Z?x?T<Iyxm3ihBieTd+<J(8}?e+8B
zzwc(sQ%*?I?=OGc#UD@^VO=00L{ND6^ULAu-4C2<wIzE`ZkbFatu>}Kl@-SLgC(xo
z45n(8G}RDU;5nNzDcP09A(0#YB&TVfe43cnMNXD;R4?(>cLuMrGqf{ee;A-CZsm%>
zi?QQ(J;+EhpL#|pd=^8tSi*QGqT^TH6z4-<?T*#4)l{0}WAZJ`+RLt4{L8sRAQ`}7
z!_R|o(l-3=e|*SOEcB-sfljwm{`0WPB~U=4X{UU*&LyA-oXgi$E~E_+0NN=(Nh4Po
zfkfegYiIqn&Pb7i1vO)&e}1l<On(|ZY$w6GXb^;!Y|5vHM|c0__pc9+O8WcXzx=u`
zszm|dx9Y@tp(NAfzt$O1-gVy2b@H&9-YCqJvE1EL2vx^UdG!dYJ|9wF&B!@BAJ-P7
zHKj1bE#9QH7e45Xn$t+}Wc?Oz(&z<RSkzqKnY1xvBE7|(H2H#2f4XngzI`+$P8(om
zpS+LNRD{`Dw3QWXP!kir#eJzhEwk1{M4+fgyCt_EnbvuhTfYDP;Srp>U-v#Vmau~2
ztCMeohkdANcj|-dD~Po8LU!z%5<^U=z)hJJu<oNq2HK?<vvmhG3Wev9@ZCBW%3HwY
zTlY>wb_Bx&`)$=2e}<ABXynGY7RnHs5PlA1OiDNFNJhwrAkrUGfJO7<Q$aUDuAv)a
zjeuZ=ka@|eEQUl11PU5Z3=UTg>C?!#KA9pKa$T*%j7!}0WdW#=@kzx?1M*lvW!?$`
znvCE+^m}}?vI3Hrx2=RiGtz+ykk0G-tg?|{nsXbJy{Lw&f7q5{QYvIHnn3B>b`V-1
zC8G?o4?qbrX@`9DOm~;_2dRak%7{@^s>>g!8qzWIGyVnKCao6j4&U?cyjd=!;3Q*Q
zC>K(?(Y7m>D)NNUA5+nQSyl++=k%I@cYMu7hL)*1yHdsEIFu>Fg-g*i%EYbW-Bqlk
z>w&&hF3JSbf4QFPwIM*?UyCf1aZA90$+9$0Rzeu#;e?zydNS5q;aCz-IzqTCn;|Np
z{VdqX(#g#7G^x*MpBn$<!-wgUX$aSJLzn}Vp9f{OUMl&oa>(OQz7?M|uM6b6#^|Xn
zk<6OQ>$j30!fanZD1xj*wq1>@VGjs(OPcQk*Q{&0f2iIzx@$kd5U5~<8}28JqIt<U
zv*-q{IMN`Pq!g#Sw43D3oekcxnW#Y3k=LB!;%1V!9<Ej_hK?^-I@4&+GwO*@9B+!_
zr7@>5c1*F9F9$~*pBi<)LkZe|AVDrvf+kM@LYmwNFO%c|1!1XC`e~}#5whsLFypCa
z-s{Hre@n5~sB^B%iubA>O_c6A=NBx3Hfr5_Yzi0n!<61su$nkq8IrJ@Pj6TCl1O${
zlOf_@h`8H3bHEt5fHyr}Zd_;pFx89drZ*Vr#4;BW_!7Z6aub#d9hG%cf$#rbYPeWI
z_NF+RqS`5cd$@UaHYOcf-Y`426wUMYglJ=3f9*J>rF>itq}k{_3AtqWmf%7vs>KCY
zB+JL<OhQ;puGb?gL=+BP4_RHGqG)bLR<k^M178FnXdQ0pi!H@HK4C*T#$^<joTB<9
zThSGpbh`d{Zqr|-PH^8lwOm)v6K9<nkM|YaFjiWE=VFDie}_UCys#E8lndgKHq{_(
zTy(89>Wx%A1xoV0jB%@92N1eh&Op`AjIlB~M3rVSwgFGcATZ1*%l;LeR-OWN<`*&K
z<3?BDNY9hB4<Z<SJ5OQ+&v+7Nf9BqE&ykOIf);(Dd_Xlbg5#s%LzkS0II04?SC>eM
zI4S{rmvf0YJON*qu!%S<0oRw^i8w@mdGM@W2Ty;6lq!V`GjA>NwkVf1!ZB}oD(m#)
z{+i&*htgYQ!&AjOgPg-@+VG@wgQQ<R46|U;;nH76JaUPdn6IOs3l}?AToBUF>>c}c
zNy|?RQMf8~vb(Kz(Nhb|J6vXYg5G8DV~^G`_sBetbd%P`{;aw&%d`GVUxibDX2#=X
zRSU|+yI?<?mG<m)epEM@n19=usp@#SDn{^Zj6m$H0Lyt!x{Y{@<X)-`F+%mTq5Yh3
zFs0m2z*bg?CmEr$vNV3Uqk^Yl_G+{Rm-upfOm6<hjxj<>xmopXJzfZ-NUM4_CX#zB
zc0Q{a+NBlFJYQAcBB-;ruTqVFs+-2)w;B_q2TUDtuLWH~&idrF^%oqDkk0Mjaj!BG
zX8ZSP8jI_vcG(i8UAWE|=KC_gx#OL-t}pCK%2P}`F5kvX2N=)QXR9nRUa_*Al%c0`
zA#xha&<iaV&r6*gvMH33I@zK$RW&$uYj-^Xen5`FgC=kizDY}(LP80DO22bU8t1aR
zC5;jMKWRx*K;gfs|L@$z0Hb)rmNZ8BOiLOYeS?-XM)6Eb8t4ChEom3ZRa(+IdU=O_
zDR3sWi*uXp>zAqrs*LK`FLm{eo>C}K+tpsF|Ba>}D&UOrHw@U8;Eqe>me83y<#zD?
z%Z--anY#_9!;KfpsW$0fUMf$Ig*Qbql?V}putnH@AbTx>Q*7`K->39y%y{n_k%$~!
zJgrzx{@cY3%`*!p@B5FV&t0P*>%ZCb?iQkTX2)t9w4!DDUq(mC?&cT#*tyTS<T#U%
z+J6C9s{>J&F?%=_0XCPhO)3)tIX9OfY7Z!XNt0W*5x&o_U~Za921277TV-mhG?XS)
zzQxha@hOgdiBgF#`TJ??%UdbYK5$4r0^N-dy#pgJzg*<yFAuMuAKt$qU94~#th#*u
z`2zfd<i8;Kq0tu$APmCg^OwsH&r&MGuUCGNl!*MY{MqrletAN0JOWI9!ON2+cy42V
z)wPU4on|RtX6JXv92}3d&dNZKxdD%Z>&uUyzk85bbSb<EAH&=H?bGLn|31`(C=}ZG
zi$)Nd*Z7Yw58(#_EsXUSASa0M^6Nh@KR^6+CR>?lytF^R{&Ja~e4nv&z*tISv>vcT
zDIBPaF<xN!DPhTwNK^rrfEIt|RT5)=r5O#nrq2{UzwyhZahk`GP2V*mBLg}-M$YIQ
z3%rm{JAu6HdrN0lq?61#{n0W>H#}kK-oJUdils6_Yj-hd1X)+4rH{<fWofc>3I?v5
zyQ7Uj^fN_m=QSp3f`X9cPn3Ol(j)62+R#K<$q5bC&b)S|6UE`l9vWT>w6x%V;FHW+
z<vp|6hU_Usl#&AULT>zYGXhExKUrW4G$C*UMO4CryC+2gLCP*D;y|&DX1~F(&7;2p
z8%+ZI0o$7h8pVopSFkMzC6B;vqqQKT?g5QM;Vs;AC0+?~cLbZ=qwvn3ga)M+ilfZ8
zLxX_AYIR3wP${9zJ;CN+FRVR(3k_<)DZ;{`XD4|}AS<CkX`$J&3SV}9J~%(~@?<3B
z&{o`H31=Z(=}&vEQ_2v@;N!OkS|H9y8Z0TkvMxS>lC*q;QB&)qZ)Pu17^#{KoZ?S+
z<FDBi4co%b`tdjA&r#Vpx)J{^{@<R-XzhhH=8l=nM6qRZw`VY%u>$LV%0v-h)s(>^
z=eK6DBFb4|cm=aE$-F|gs7+^vluZ%RvyeT8)Ha%8&MN8cC0MY03%0DjZE?!7m|#yX
zuEeAnT%QFi3sGq?K_>FYX>$M|F7!2~$#+{EhRhJp?i_k{Np7)oXqN&<Vb`8tzpV->
z*1+|cleX90;)dxpp`kH<eOH~XL@0HFJW6KLGsqRCRuS4Gm0K%fLR456Bays8l4E6w
zBoFoFEy7%(Bd&0*<_ecfc5?TGZd8CIoI4jN!z<Q*w9t$9Z^U-Q)?|6ZCy!_i4GVeq
zr^hgrr_lNo%Tr*S@`uNrPC?U6C(-d^_E@a>q2^*omvnUPm+z~8-(B`fN{c}bkwq6j
zR{%DximXWRa-7iK8eHqXkmVY|`n4@>iOb6qlFR(!?u1{>OM-2`yF*$F>+q01tkCDD
zwkqq3O1e>=>J3eg6hEesafYok;;6zaDxB*%$C#+|xNB17WDs>!w4&@Om3B@AugEWl
z#rrIt?(AxC_y{w9L@Hau!i(hZn<ietn-pMugV1o2W*6ME`(A7#MiQ<3e?k3Rc)vU;
zfR*i2pc-i9o5|qH^s3{FH)``%RHb;1?4zC(;&X>ouL8lxekimeo~n4N$|3u$Y%-hM
z<QXZCGX}x+1WqlOhQum6qD|D<kon0(DlRhyg<V0ALf6E9tP1k62oiPvdMt=#T5ho@
za$O)O+XM5KMu1J-_G+6ctZCPUIBN4lzZPBW&bp#38(VP?>mna=MOdirk$D{|Xz#^R
zfbJ?h>}&w-7-f@Aj?KOUE3Nj&6{qvMcl_Dbi)*UR%}!eqBQk$i&&_t)w$>RrU5)|w
z+EdNjQl1=tx!sZvTno*~d2UlKvKF_ihqbNXSc@1MMq|rwI(#a3yUu;1V&C$I{O(N)
zYaa@|v|?q<#6=mq$))~UJZ1dGvy!Z@>w6gY@!A(Uh3}d2CmCzyoV6!&*1gPWMn7^|
zQ>TmFj4Dl(LiNAv6DAhDyS1ZpEc9QS29p_N?F+$w>#T|S_qRIwWSN*#ommIa8?M6b
z;KaK$B|FE0wd+bu+q=DpdmJ0`;~Y^F6*Exkk%y`Yl<k08L7x(sjre;q=5bcwJ<V$z
zV;-D~dCD;jwF=jXm?xDWj(KZ>+#K^Tv}kc5$!^RW5{=&_Zb^<?&dpNZCvLUs&+f!6
zGwPLp);Gs1bR<~wmho!SqoOea)X*2WFRs8LQ~^zll<pw1IX>rYZl(RPkKk3s#_uY0
zTWn52s<=X%;fbDsa&#sCJjzvM?fglKB(jVq;!j%6nM8aNXXRKEXUm>Pjhc1#xZJKx
zizQ<(gL$5*#f)X9eK$igP9VVq-MM5U_Yr-6OK`Nj%>csLo2iw0wA^c7(GFoM7c03k
zrIrX4Ky_)Y^pe}Y9=;)%6+YLEglso3Fqj7B<$YaT@K*31EucCX7=VODvnt8Wfq}Hp
zez|F1k%8^CEYfxZzCz7#Tb2%kciePhMZM4nh3c+N%pt(E!se@MqdCB|=H9v5XmhuJ
zGzg*@gAZ>NePro%IA-a)IHfOr>#ZK;UF^zFw|Z2#BtQq3dgk|fXcIrRvz+pzT=LmJ
z&PhkTZaDndObpfzlXeAY2Qwrb*_qqe96NVt<f8wb2>qV<%!xYuro-oSo;)oG-WF%K
z&A(V-oxK@?lbya3ySk-Qq)4DyNLH(VEBkWX=-v#4#e$3%@@#)IRDFFzwsiv*x|c_G
z8OAiXa2a}%*7r4$oqG}KNj!pDML#!{Ym+&7b$PAQbYZyBYyAA#9;kI^d!R1nRUf*F
z_w{}`lvY9m%zIcOU#)0Wm(27f)DR(0B~Uz6IBOOh`$Xu&vL-LOrY-v#w~|YLJQ1pH
ze`2327kow1c%jtzvHU3W@^L~<qsngN1A;WF>@%fuaiD_hJuVXu|7Axp_mXs#`E36y
zkDmHpne&lk8=a}V`YE5uPm9j^*7-`*)={p%=45zpN7i?j^E9y7=X%8|F<#{zIwD0f
z>dO0)2fZZ=4EBR=Mg;AgmCd_<ow(JWV)`~vh|;_@eVuo^NW}AVN0Mh5e@nMBah<!J
z7S{65NO^6)GL^x)s{EyBSs9a#<AcSpT+yr7afb(U`A{zLnEYiyHv3tB$MQex5Ta1f
zJjAVD%czUz_@_dd&{5}F({nY_Q-AmG$K4ZB%Siz)1u|37J{KL>0aogNIXx8Ax|aCn
zut?W#$N2O7F|-tJLkK8iPr}JCx-}S)Q$;6ryfGV!y9ncdB8)?F9(rSzNrZ78bx6*C
z_)%Vd;vc^Y$Lii){uhUSVUiq1SV=Dby8MkZ^_X8<yUM%}?e`Gg)8ms;>fImHo638-
zt_GwW2d{N4;HqtH23o;?W#>5M?~GoawJs`gLTRL(MOC{BM}}j$hFRgF4O7W_ly25b
zG`2kR7Sqcf1R7$tKS=h|E(Dq_)4>9PeWW!KXR{l84)Re2LbbD-P34Dr)<E5Lhi&_p
z%(kap#aZ9OaF<QNC=Rb%%iVPcuOa<yj2sE_I8Mpx49xIIZxok*yi9q5?dYVq^_bET
zpuP3bPxaqcLg~TbV394fHfsBatLHt%2%Yy<r{vpNk*Lff9(N}SF>spwkX>nL7cd6B
z>*|WveHSThyml9^VnRM1$X;nNgnb{_)XsWGyE+mseS01n{#gkyQ-#N;03K6M?=*5L
zQ;-F5stA&XVt&_u>r=SuRLT~?%Fy&~9(R`?ZtoF}%g2#R8PYFbpK=W#l~A(YhMJ83
zE@=qzqJnTEr{A!kQql<ni?zYntb}q?>MIZV_mnFf0D2tP96cTEx;o{<&7@KAcYoej
zW68S#W}LDgNJ8fWIA6Glix>QCoPzsYfMUSDg$r;2%(=~fFopKHuwBV{`3GE7`0qp-
zr|5ky9CAtNo4CX~IFkmL5PKf?`|_NT$))PTDR<6cPuX_U7`iK9=Ulik0^7e4gs!sD
zIUnw>+F;j8|CFx9)e;Bn+VwQ5=YJ)-Ee`6s(9rDE<s-D$a>b&*z|*(Ex;u>rZKmzZ
zFLRq|GlKUnhZv?s^5Q#M!8u$t%Ey7!9ZIy7?p+nKtEG7kpKkjQX8Dgb`JD@k{{y0%
zM&y?<dpH%BQHVDamyDP=D}T$5+_>@Xuju0pf}vSt@s)TRnCbBt$SsC%*_?uG;sATd
zo;H7<qWBO?q$suF4U$98L(O8npQPmF?-zOb^zP-$yZ3JzFHU#^PF=qIbOHK8(oc}|
z(3*<_5LUzG%g>iTe~?mH`n=EwrPP`}j{f%a+gzVCcv=BAf8hG$5Pv?bYt^l+(I!Eu
zP@wZ~$QV4Wbl%ChA7O)M`snLlzWni>%%qp!m0oRxcX+W51yPmheUXMi)zxSOdbe3E
z4bfn9`e6Kf;mXBtT>Ouo-snU87@`!P69m5HXH^DBJQ$U}S@x6>qJB&Nn%lXx#KH1y
zv-ewm^4!ws@1~LdQh!l9X|KX3r){$){wZm5_&647QGxLS&1n^|)jRBQ6<nVpC36<t
zZ>)x~sr+}<0E=L&K$RB8hKtoGoK;1j(L?bi5*d9eWWix1GOPqfqL=5-?aD<Uf&TL{
zGuG||`b8yHO%$Fh>o8PdCB?~%$5je$5vF2k_f41zL=o_<h<~x}V6xcVYAVbBFrYR{
z>HV9BQJm9`%ushyiwcie<=gei%EjDXqtx?W<Cuxk3P)%*d#(JhhGAZqEFfuQF&UdV
zu~nYMh}=M8Gu|)+VL)%`!H-ie3`KqgokSR&K5KU2Vlh#boidDVQYqo(8fROLr=3F9
zq0Mqq+}X`*{(s(87ip}_W+}&<gG$GobIbOYoV(at!e#RlnXT(BQ34-W@Mxu;cFA;A
z?Ih=fOb1(sz6lL%mt`%q!D5w!Z>Y<1F4VY{=1MCxywqJ?>$iJ2h89`mhedv{@{(k~
z_`@vp8?rb6bu#qx)r?N%A{C6&Olu5_%9)1Cw6^Cd!+*Zi-EQk{18<vok}^_QqqY=(
zQ<P911az`WoMW=A%+b&%vY*|!Pb7)#_~<nL*;oN}oKcJ*j#RV+S^lav7!v0T_pgO>
zsOtb18RY-uYlVL;l3G?7wyNU8gBTmM@T!;It;U}BZxpnG2v%9-aZQS#JcE4r`j|n?
z7{m$Kf`9x-1?}S$5jbIyllbt@$7#$}DA3uCt2GAAbXfh#6S++(bykPBJ9KzErNh^`
z2T;znU^;q_b*Q`+<O>Y4igv6+>M;a{4alJmwFE&e2Z#DP)uEQ8L+rRZD3*-_VMy~Q
zBPLtTbJU_i7VA?)@_qcwYoD_%EHL%yZYJJj;eTomcSamV@=?Ziyv(Z-6;NpO9aIxF
zGyQwjmV|k=HII_iq_`gY?byQri+8m#F4xSMb>i~cm)2_=Z+uP_f9H8ebzIOBlbJDW
zlo8VQN_uS=W`$lcP!ohz(1#%SZ9~x1bPPG9He1?qI-Qy-sa=MSsn`Eut)uh;By=>N
zQ-3>>+Fj_?II{Dzs=dEL7#bE*Qnl3(Pe=j`ih$5TLWVMa2$ee)L%)Mt=Uy&1@R<9B
z95<I}io6eK`m!C$3J&TQl`+FkJ@v3QN2mpBGxjibh34Y{EOhR5<n*M=6X{8WC|!|k
zy6ZB^uyy5Lg!OBLW&gR-&EAz?5_~sHsDC`&*sf!Ke1QNq&4BULL1#(dvVT6WKHKbj
zO5z#Us29eV$%Z{oXgj?!jY*x|Y6B|Roz7^1%DrjYt8}-9KXof_Xj|MpHLVmA=BCpx
zPkN{WNv^AAk}UWNt{n;ogSfMjbc6b6mEuynQQd&W6Mc(Ue@r-+o_VoN2zm1D`hO@^
zbng{aF<4_!q=d~aY7DrwGX_@WtP_nqzbGD5W_XF4?O?M`3w+K%zL`$gb~?bK#X9*I
zM?XOIVRsaX)LdHXpPIu1M#`-jmPECI+MrjEO4vN&DPq_RZqh1d8f>R!T{~B9#))ZV
zXNsomo=tzjAX8y0DQJ7IF4R}h$A4?1FTYT^-aq~PX}_`vRv;C@8wu&6PTQ2`w>4<$
zqCm-|7dfq5-D(nojnJ1ZcE;0Lm~Hp5TVr4l+WRf>#b&ecu_`9V5H!VVinmtruhH^*
z2zSY=z7uXdhr&9o&QicN6Y0YYNNZT`yVex<(&)y07a?2ggp{2R)uIQ6-G45{AS(L(
zQpowh+9V)a9G3Q4Uo6!(+8Yk7ghLV>kb+g7T>M%ZWjh9;ZjNYRAS#16GXhak39B^+
z`Fzlh2VtaQY*oT`iY!XA9tcEI0l9|UC9LOgUrRyfhLZzl1W@VO?(py<nFc)0XA~$T
za>NgRn9nGDY%jNnq@3fS5PuTq^bG3Pf@sW=CeP3a90rIdn9`?#TFI#77EH~jA<7Bu
z^j7>@Uv-(?-WTNZX%h=@qv%E%X(<cl-nS<pG`?3})VXSfG}H%?DV*DY{{HbyBcRD-
zu*m}5@d!`A{jtV)%zq~TW!QQCn@1dKvjQoc8a2x!4ytApxEFD>@PD2ppgo0keJ<jV
zvq;B5jv|gG)mRQvB90>$5`pZCCur)UWBAZW`9|f4qYoS(gtlA~5C1M${E#qDK8%Jy
zP*7OR{y24EMKL;#u^E=%9Yjw>QS&asv~)>gEn)f~;1UQS){4Yw?6(A5+98tJY)}<@
zz*!4&k~8u>w6HOsaeveVBV@FN7RwKhPs+ImIG`^kaZNkOWd}yzUaZZ?T9Bhy8^Z@M
zkpGKVYvOc`F?ur8Qf=fnB03joW2Oyjl+w>BQpGV2F3~R?@|OQhz7CQ9^R<PY@dZ8z
zP{?o{*l`S^APonZ+i?toAuQ8$sz}b5iHg?~tlP?<3~VxwO@9_?B+}%b{B+aHv4Z-x
zHle86V}i00%YLrnRfmT^S_fgNq_{&!F?<>0?~#%=)S_#%m69Txa5}Z<c>uzsRI+E`
z%-Ho%p7LOmtv7N`7h|S=60?>!_t+C6rBzCwcZdqn3uoXSQBeacjPgT+vik<br<&_T
zGk-gu3XR4L6n}@QXj`Y%`*(w;K4_%?0Jd9|*-liUI22HK1hmI;s=MN(<=Jtoma}nd
z`nE(c)eI`rH)jW}TJ8;6RdZQunjjqtYSFmpVkn<?FxqPYlm!{sK?4)5Gbj#nkb|jU
zG7j&;5@OokV3AE)+B<up?$o%gIV{sX6Xv&SNDX>)CV#8TG})9(hSeyFci=7MYp&28
zFB|%gr9|xOIYjY5HS|Dslj@nZ8zr2DCPnTvyr9W(D00MmVfk$4;UDw57h=xW9OT%>
zl(LjGHkjtxR~jf8TMly2^J_~Iba!E)#==lb#X#g?>7(N8&B0xm40S)TRM<*xj>x!_
zZCj{=ZhywAlx-C?LQb6$T%2;LPH<0P?GZ9(DNhHEOdT|6)XT~6Eyx1p1_d0MKIqe^
z^yq~S2bWc-x3wFi_<AtXF}=0^@HlJPV~b&xVj$$dmi*SrG$@RB{Ix1R?8oI8l_fD0
zlkEqnqY`z{OCfc<Dw0bzLz$9D(g7jPw$d#HLVw_WF{iY&8t<4kYXunSXwymA4V8PD
zNeOg;Q+ZHoQfAUN`im2Y;f(R41aeGs50-ev*m{oSBrB~HPqIRGf&otzoQ0M1SPCmG
z5<S%_0tIH`y4UKod?ue&X6qkI^M1e+p>gQ$iGYalZ5ee>P(rfucwvKKDb4KsjzXNc
zXnz-KGi#c|AfJmpcEZ;ba`QGB9JN}uo>IrN&U98;a=kLSD!DZk%u*BR{O(V#3A$U<
zI$RyUU)@k?>S16Sdp?)4xy7K~(nP7b?cFw`p*=8cZ$vM3O?!{J6vEg%&$~1?q3wF;
zxSSrDW>UiHALE;uNuz^y_R_r6(NtOGyni`<RzBk$7tr~Wo6bBzqtd}k?JrVjW4)!{
z6uebBDI0z>IW4O?w?I+IwR5|*?yP6T!`q4ZT`@o<<U09l&&qJgC#w$Bt$FXHU<m7Z
zTznRF%+Ktu@ZozL+qGwQ*Lp{?Y=!H%BYDu&V>Eeu9Dj4(pWkYY9H(Svxk7MTFMlai
zHB#3cbw<BFZT|cC<=wx!N*yTK$5WNR{`@X}fiaOUX=;gwzx?g;)4M-ycip@&SASE7
z6RjowpuZN=VNX8i_XC3@zNNK>kDsWnj<p`d_XADa`R*UF@)!6{;9vV}6rN0zBio!M
z^F!6e)_bZw1zWY*=lh%sU+d+p&VR;E8JH9p881H;d9WvqJaiO3)pnc@h;i|C=%I0R
zrpQq4WPE3WuyT|sKi1@)8AS{By%~9STX8j+kzpU@GU^l5nk45u#}_Gld+kX#VSIaa
z!EG<gaM*&Sn#=NHM_@0m`0llr`nKWemp{@~%1jL#(N_!=UpP5=%5a0IxqlWCDOD+N
z`?1CMy2ycs8cKkT3b$8>P;K9j8onEK=F$Z%72eq`LiYY;Euy-Y+)!wR!y(`zr12}J
z<%I0|WS4OpzXdwKp;mTYZ#KH_W}}_n*tF}7(mcMdPoFoH49xkc)RuP)4%XW3Yqcw{
zB|zTRcU0?Pbg3EMk`YG}L>Z|^FVZIbjGC<?^INpN<VSCsQ$qd^X14N|AqGSg128i;
zmoZiiBMvbzI5aOzWo~D5Xdp5(mtpBJCx6{qZI2W;68^rw!qR=oeZcMVOO@*sA(3$8
zkcbWek#e`%56dza$dMhiJ4EE)-&5|M*`A$QcHYRK12I&)-FCVB)KgXNrjR7M;0XV-
zqm*4L8JApJQ*x2Wq#B%EnG{`SZg<s{P+FTfF4-gvPPj=mIZ-|_$*zO6Ms$zV#(%Dg
z=}L6U3Aa*ZNCGKGZMQK+hH5q@=9SbQQnc3Gi_oib(_D*-fz{^H2#rJEL0wGA^veP3
z$hoPi0<lPIB^S9ZGzUB;_({R;3%WwW^HTevs*5#I>&dv4%ZZ8V5|w@e*p0X!wYro9
zKV<ha^^F)XH)S(n7Unh<p;e?8&40c%H4$T}+JI<;X4U4xF61zDRhLsVo92kU$f=3M
z5DPa^Y=_#s5sy`DY@(Q4R-4fFIBRLk#!i$Fnp%-vi8<I@DY%kK-<lw{UC5ZR^xK^y
z9uu-rVcsG`R};0ZASK5k^&&f_+=~RUHy0k`#N|Y}qS35pjS*sQYA1Ern13c_w2Jnv
zKv@miTFGI%0w4o8N+|sR6=L5N8z`wE*G3e~U<=C`Vy_HzmZEK8cvx)jtphMCy=eh5
ztfq+oX{CmqToOdFeaFITjQtD>v4K8waTR8^$s{<yvvt6tEBB_s4pr^f3lM5(u+G>G
zC23|X>NJJc7kE)Kpo!$P^nV=~E+hR2oSfDqHXF8Eb?9J2hTcwWzoE3nH4THdb=2n6
zeV>{KRGHFcn9XFHx%6tAV+M)suGsjR_~hGfyH|gFm~@XWFQ>Ea<;T}EyL$HS^1rV8
zZhG}*a&;O&E#a*DNB6Y*{*?NQ7n6(GY0AlqIiQdy7!XGE43~~Icz^i8S@+0c_?O)e
z(^u2(vHJ$7Efs%C)o1`Kz}sIR;*V+Z_|(`aWB3={&;R<jgP#bt=JMnF_h;*SU@^Y3
zxnJ-KkG<*Yy6>m&r&ljOoL@{_vIkuE{CqZ>TwS`@U;h2m?1z`L^V!5v-|%F5Im0AR
z8eD2W%M&mm6%5@j%zxf7r~U%qCG)rbBHM-Z7v_}&{=>z?hu!n5>BY;*?6iCS_(}I_
z@@Y0KrWyVD`7aZwdO4e1&aN9_Hf3Pl>*>d<i^=uC(AM_j<juSD@1~#Hb}$G?G@fCE
zD?Dr6IQ26wELFqEQVdJ+ErTA+N=t@WGTB{L+TJjA)i9N}Fn>%B@arMG^_DvABb+U)
zk8pFTm~^)68|&Qxi*+=(1I6eI;%kxW1tdcydNrD@#mAD*PL|xOd$bt-V)kx&2|xMi
z#Z&wD&D+`R!}Z^~?w5D7w;x~oi|KFOuOEN=@Zinl_wL8@+1r19m_3DQuKt6K^S{3D
zUcaBd?tUj9gMX*)`s$*)p3JUs<1f!hYku%71_|)9laYb9TvANglX2>>Bm^1Hu^eMb
zk!!({glOH8wq!6wZ(j|hk$ti_tJM?2Hc}a^FRwBffsLb-SZY9CQIt3>Ggy`T7;MyI
z&R|6Zh>=aYA{+9?E??S>@8-892Zu(ak%o(C7d`Dfv43czq{Br<az|z820U59a+@u2
zhh4N>wZ8z&;Jw2D?P53X(7ehW8h7Ll9&v}#UOd6cM<if#pJMewauk+;Gd;kslpB|7
z_|tTh5V8cSB^)gy>;@B`1JxG^NO_oQ8B}#AP+4U&0+s9qszljaQ8PzX!gedsQJ6Y3
zrY2f}m4DvHk)s}S;zKFnOO=hfA|DmfO}e56@J38u0Mbh!ygNnu41?ZDnukNNtf1J+
z+}%*DHz>Aha32(tWFJu#&QVN5MR5!iBR4P<S6)W39U^?59M^Q~K}V8&!B&4I1<Qsp
ze1yQQaYktsXG5Id8fU#n3^=2(7tT<Cdb0HoIDgBDJTo5!XVHRi#D3kv8O3|pa@=DM
zu^f}vT#v|WJK^n4#<Sy^lxbby&vq7AcY<XMYhbx!5RY*TOCD-TJIC9rmMq05J2ITt
zl1#ozJPtsM)sj);ezl~MZ>M_0iWGpQ<18qVVnL~t+a9MIF*~iM6Q#iIgg5D06?qwd
zm47kdGb#7jK{v|3xA)04DM{QYCCLLR-S}i0o%Gl-_~4UqU2?MR*=LL4FjOUPN6tA#
zWGrSy@<Djz<e&yk!eE@Js2#8)ql1H;G9)fOq)~>1oVTRoWJs_D{0LNc%a0=87rEk^
z<OpT4y-oU4&ndqc{12F`r`2N&d)=~|V}Fc}G3=erUKM9nZ?8`l6(>TL9_MH<$7(=i
zx&_3PHX`(;ZbN;TtB)yu{dv@$pxqjcc}?n+>TYPv<J2kbmoVj!yuFcLWb{FOEyv--
zELJ&|csrv{_nm)?dn_{g5WQu<U&Z^I&*5!3y+2f1ka-nvId6ftk#%$V|9LZ(3xBOK
zjPB-$C6632FY={a_dDknxZ=(0Fc!p%bet2(lB|T`ai+3U$ivXlK2lg5va%PgM_H02
zi+M;bZ;?z1_mNBqYfnyDHtC7l$%j87P1}%XHf60%-a~Eeg1m;CIPd0yGA>z4nY&<F
z94Hi7NOd9UT;i~hXd%Ugl=eG}2!8?p-@LIZz(No?&RgJQwL42bz+2#05n3s`jOv}h
zfr+!XZ?9lEFS)F~1+Gb`6j20t?1zRD!<u4l+jr3lEwJM`*_ot$jE4#jO5}_LX7yDx
zmlVKfTo`;7CDtkj(7}=nC6^Rg6!*tZjBLu<m32i4)lw5o7AjHV0BWcGm4ATqd;7jH
z*$Yy!kNU6;2y<3)Q(XP5C>Bi|iKzzYs!{79c7p}!R#_;4HTnoHxl0Lx=jIqGNvOaB
zfn~}8v?!?R!B>$PWk_)atZU!Yf$x1-hfL()S{dXAB^2z!Sr4%x!&e|tJ=yn-u#J5f
z36l0O_=*-1qy+qBhQ5GBkbgrck*bHoIjvqIKmqO{``(QLB9?4ti5AqjY*Z93bsJLg
zD$qeU>55Xsm-N+S?aN<VEJf3$uPv5>?oFKCIx^O{8X1T0(zcF_b)1~aA~MP%GRh({
zO5Oi}FKY29IuJ(zJM4Kj+>y(cA*HyA`5b_vSP_@Ng!s_|xIcUdc{}Dp2j=QSD0k|x
zSxDICj4}9{ZccWnapF-jcLIhVJOlCaA-?(=SQS)mhhKA2tzderZeu3-AB&q09G5YB
zI28dom%-pV6PLNiC@Gh2uQ;0l!nX>rIR6L)gP@dAm%&Ik6aqCcm%-pV6PGr*IKKhl
zxAwU>*9im{M5Yv%!ALh00X3Ila19if`olQIe~LVhzrOu+qqFLxo=D2raIlV*Xfx^=
zj%4KBpuCc9PCl_ergtMeS+o+d+K6WS8o!edujNi7>*D7`k#&yuCK?wx%KtBsPtxh(
z)%rtF!J_544bhS!v<J(q91ptP;$0+J+=5nl;Ncd&8@M{R%GI;%h>(y~&SkD}caQ8)
zf4JWx;_Zjw&1FUw*BFs7An$8N|NOQooQzk7Y7(ww`fZtBE9-WNL>n)Pn#jr{&aEg?
z%qod<3M|zmzsLr53@cP<YN~A$Vn8L4Qj_#dh+YL~)<b#!Ol8Y(mMwS49&HII(d}w?
z>)3Kma=tC+RQ^0$&PjT<1XjqQrZJoAIw1tb%sbaZBvh?(;k@wLvmQqZLbD!wx0Pf_
z%;uvB=1GRCBs1PJPcp?yC(?gTt0w3~n$s$m#F>+q_{BIpe@zNZDHA;KW>=Dbl3>SD
zV2%inSqB<EWBF`nx)t4B#d+ehEHWWe*j#9a^0nzQEN<59)a}o(?V5%aND8WGe&{ED
zwlxWEOOe`U(Cw09oeIX(B(yEXp$cSoNs*LBa;YNWPOe2oY<5ZUvBuSoJ>{l#DFKu-
zyh}>33MAMne-gVXg*Bv%zC#LYqfCIBWV<vnXl2|!DbN_$cO=NgcIh1nQY*PxM)-9z
z6EaCzGJcUQa;{OvYx0uk$XIXX-3E<A7F_*iD-)KS8$BBAJQJoo;4|<{8D<>}XV7j8
z$^eQ$b-#9D;XpI!8uJZd&RY?@ztp)ZNeILJoO2mOSpGU_!G8k9(H+P1=LW<+q~GBh
z%@FdKN}meKK5iKsoskL!%>tfY$J9(B+HkbWONQ=#wim%hgT%K_&8LrVH~(7n3<iuM
z!o%Ob-NY}%{RB^i14i@n;rD+WzTW(`<(e(tzNbzqZL79c5?HmZR+;EP0z9l99jG?w
zd#x!u`<yo8l7D(B$mTHpNxoZM`TS|tuUJXY)Gh{ojIYWkqrVHe)(H;`T&pW75`tbc
zS#$%Ipl?k!`f5|>*A0)}E0o@<;i6yGyrqK~ip#J!2|bp`npbd)4JXgG74)<pQ+(wX
zROfwwX<JRi2^hVbD<?Q^;&W*y<eo~i+ppW;oJdKBTz`}NA}Vd;bHTGgb4~T{AMWnK
z+S~8x-Q9lav>(!>i@jUH(CI)V^V&vmfgX<6+m8r7D}v93KoGeKj=X966%&k*1aYJ+
z;zLo3R!M?WR?5$u<nNn8ND8l~iePSaTBqRzkM~HCJ3~eA0iux~u+!p@vS@0n{iW05
z1SOVhTz_J_{bD1iXyJR<fnUZdMyRR2XA&@{(5LgJjo6Y0>3JJfJ4xE$PKL+{nc{ZR
zY_&3Ds*g3-hHaTanpxTzl@z>m&5f^r>3})u%^Qk`4$=|~bjtjvu$*L3<YlPO!+l*C
zngkpx%93X8?lHjaaa%NtS8}7R-t<bAh6u6VxPRN<wsD~mDVo~7PDhBCGW=7i4f77M
z)Csew9GeG>>xj`1ONRxEs`Pw3r?+QEO0vWl8Do#@F6(h$j`s<Xz{X*!oep5@t0v=>
z5wpzU&(t~7455A;j(Qs{!q`27t*pD=ix0JUMdK~|lLLiPff`H3!|o)zV6B3~a(7R~
zet){bY+;5865WWiuZXdY;9JAkDxE>}!byxoqp@p%2VF81OqxZ&Q#4H4aJ%?O<%{t%
z6pD8m-@sJ2X?P@~Vsq7KcL4}KWi8SqDSaWS1WUF-Wfi~BWL$M$m%=e>9!73U{`-@K
z+%EOqE_QS~S+LaWc!JkZ#WXi)8YKfvjelbL#@olq^#?oo;wxpR&6pb3*PIhjvdhzW
z<hbxp7Sy80=Tpt|o*Xs&np0AiUF;5J@1;H4N|aKOto36#_otp9y{`?$cp@#<Z`x1@
zD9I&s`p}`BN`Fl5-Kr~}b8oudXqDcS#mi+Kqj&!}*eb0mXBdR0y(<#Zf0N*HWq;N*
zf}l5niVG<(ll_K$JG-?5Zv~LF3@<z9INpmN&FX^Hn0ZD%Sh~i(y?e$c4+_ZcF!o6^
zWWzQblTmS&248pg`5+wQNm9cZ_4>I}5Zo(uu(^9-yB+5Yva>+0<p{7U9|h4y+}+Ce
zQ~4>@9Y6coQ;Z#YNjkJ1&ecnhXn#DLUa?EPKl1rw*5&e3lXlO->{KF0hisUJ^FocZ
z5j0ILR|y%dpJ%5j5{A>C;rw%cW$s%g4y3OK=a$|w>|87+>z3MDQ`BrNQC$|X?8Wxs
znh{%bENO2~gj!afUCZfIgbLV`Cc3@*f(XqixM6rA5XJ=27?ShWx&f=1P=Ao&WY~lg
z#<R|~<aU`8lBA9`7fkNg(WexZ@sRI%&=EKR_oCJ{Jux0xd%UJ6)*Le&e3vKIoO+(v
z$i3VXYlb~fYz(>66C-+D_r_u8_g7qjddk7ViYLaN(31)d+Pigh+>cj7Dj1|iw0uQx
zYdBuA>ozFR*m69z?b++v6=XHrK0kLicGh0;)S1U?`nhgK3U@E9wKk`5*ytGxN(brj
z`qiSLVc3a=R{ZK1!k|(F&2(mT(8uA~luKS&5Or*w^x&(ftYPERQ-4Kk)C_ypsPDew
zle+y$0TgP#(BG^lg$%cg%s89^1jooy#+SiJHx!q^q!$$dGM5qf1Sgh)0R&qk$?fj1
z=-kHGOu0zbGZ_f5HTD2|%fN>KNlr;-PJ75={~uw!S}OL-lG@1*U|CYD_$a=MMLm2U
z^zh~G>Gkg6h0Nepu;6ic{Wbvqp!q*&e(3Dr0hA*cUVoR{&p04|oz~*9tsz^WFgED=
z4wXXSl|Fc#@G&)b<d;AE<Mrn|U0EObL{iqp!Fi*Yo5pA2%P^#2Oz1s7v5ebddFQ|U
zZkVQF+=e-{3F64_u|XwH^Dp)Q2}4?G8G@P3kQrm)>ft58oD`8Pa}^|a))iXDS01o_
zJ__u0UE#Y@n%5V9X;Ks?lrpp%Wzb0_OgI^Lz0FI*m^hCkwkdvnlCHHC!_%tFc)4=u
zCK`QO-M=wbktGSGUmB;0GE|+X_xww6O-8=SDKuMYnzmI2c45!<fKZ>>$BZTMIk8WN
zN(US*01HMcmgOo|<|*T<w42<a>E4D}Uu`suQHFiC`D(&{@>J!Yuu-CrwUa$bMVFi@
zy?qhio?q|&GueVkA;HJtpTFFt9}HOK_$#7SVBPTRzlU#kf01abyK1Ewru?&d`1SiB
zo_wv*C~DrJcFx*<hfZ1EyhBvv#~qr_MOJxN$Ffs|E>4^%<WDEU?a9}(SYn+biW4fw
zll+R_M!#==RH`jc`3_D2N`!Rg6v<r(EAkXfk1en?vrEQS)+aMbuH`he(B#rLtT+=M
z4zfkL+w$HNw${p6e}S+U#n1%XW6VOuvoQElc{hp4_Am3^JvmodfSjfTX4Lb^B5Kxy
zk*g=NdpmvPunQb+1V(^a`p6zc5hwJ+Dg89*j^D!<7*a0ci^HK~WxSCu6ICQPCzoD7
z{QbF?57Iaof0K8x!eruKvN6(Dp&^NtziQ1srkid#N1Xm;zPl&qwpIPWhNuju)`_(g
z&81{AFR8UXCC=so)lD92omV#O=&(J)7W6LhPiIsxULaa>PtP?BXOe#L4c=M9Z;x-m
z2U=_pGBj1GzEo*8wwmxH(_A6E6vZb~vopqUi}|rGe;mS``DN4F;HV{(#!adS5|>~O
zDT_45_OwDW%4^!w410{V6uim~drVt~*sr#&6m+%BHMOX!ioYd~pyndVN{UPI8oAKZ
z6S-V!tZ5&M277rfCLrZ4PWCW=aO0C88`VU{`gk2}(#lWN3wEA#I+dZBkiLSf;nJAs
zt#pl7e|)N)Zu$Dyz@@o2_C#0f7&HA^B+~5pt0bq}%Ft`0PmVgt@ZhtEJQr-AwH;Dm
z2V{Y|-9w|zO8E#}-R+6i&xG46Hm$kjMZf!|-TQ1hyi1NTo-0<4fcxbbf$^5A!$k!e
zCB~(?qvjg9BN5A6i=edxMu#@lsjIW<r@sBrf3UltqFp|CjnQucXuXON_AAB&!i{Q7
zl8=?)CV_e!CoJKK-k9;|IftQKjz6~}v^_(f7d@49WOSF7CKBSP&J64*p&sFug!<Lk
zP8*N>x%r|mZPjK!=^3T(rh1Jz)i(O_+6o==*)LA`o=<}499tsE)pG)Fo=Hw=KS`I@
ze}vE$JyOdG%s%~?5+93vi&bs7=v=5=&c~ynUX+qoedI8<kJj8oC?i!m-VO##?)L2|
zy3S+^a~E1|JLQ9!*vG?*fx)WieLB_knZ!HOZRo=v?@OQ6n0XMc-yHNHf4J|halZS>
z3{&9895A?ENxc$I5XDyKe);-ku|3|pf3d{*0!(+MkR{aw?M{te89D0J4%w)Tr9D-d
zE^;y#n7ak}_Gc{!`Q8Te^T)1Bz0~i=keoY+O+3E1)>9`YZ%vZS@pMXSvGmf`H}Sh>
zMkcLq6*-KocfQ4Y?s6}bA~ie_Ms>P!>t%YAK=XnB>8AUqixPteg<9?hKF`$tf46iK
zqw?Sdk(0r`sR+b~%1B1XbJ?6)7w=80#b7jwG|b&c5;=KqR8^a<rOcUn>%CL9=d72O
zIn1~_#%4eD<>BZ#?BiThI^-zV-ASU-J%b;<)wst;=vfFBnTH<a@61t!ayScRSY&)Q
z?Vx`9^ZiF~?!#Z#3EI*dkObj=e|2NdaLhJ66gEO;7R0BMc<4dCHjprIsp2X0rV@bb
z^hAX#Qz-HG1N_uDqgj{;xr*&z{<TgfaxF-q2RWDd0-?z6auK0&aU6S)pCC#50avv|
z7??%sh4-gbn;;?<gaG+%RjZV)gM*Q_fTlvhJd_1x(MdN<rAEqYE*@5Qf99+rXD&2&
z;&UI8Fs5}wCm$4gvuNlGDhtS|U)TmbtXc+Fh3xeUma!mtX3CTB!teL1QwXsHV<Gl*
z*G_pBB;AmgICzSHB_>)iKS1QKoF7QnOE6LW*8D&ytt0`}<|q1?ow=v-9Y38lg+pH^
z89zG<fJ}>w>u5l$TRIuQe>kIIT4p|fYTM1UKIM!t{kw4`ofT@&hmtP}cWwS4juQ#s
zI;oE<>Sc4mxX5wImmIzWrfPs+f!d*L@^q!hCE$F_hAbpS?FQcUm!$3FPE7||)Hm@g
zcOJ*mH21A8FG;sb0A`&11|;*f_M8|4XDyHMcRIKvxO>OJB|+&2Vjf(Q8hzW~l7RZ%
z3@)MNz<BDuyvyKHvbM9P!rp^RELp<I;SJKsF`dHy2XjjZlzx2Yxg|;E=D8(-@CTV&
z0<N}Gd9>f%fWWxp+>*e&F}EaE?=ZI{FmB8(N&4?MxAaDre%Cl<f0rxT(WN;dx8de<
zOCS|J$kpkd5}x;~RX7hDzvqm#z_`O4gTS~k#~^pK?;Jx!Wg%QK$H4WLwQ)aR=<#j1
zf*}5$LkwwXVPjyybIG%xAkMLEh@m6A_pR0y;h>ar-3AtB7HZ@;U)l{SzUAVNkEZ_x
znZ7fL3T19&b98cLm&r&s3;{Klp$<0`m#0<;D1Yr*OOG5i5WerPXl@(9I<~9)g4Ia7
zlT0LTM0-F2PC#<Xfr$Sv%Fnjl?w)4|2M#?w)6*_jm8-sb*nIjp@#)>&^UK}q1K?zs
zRl>;W<-;WC4^E$g(?hK$BLveRrk79CFOQr{P1iGB$T<LAhQ3?6tK9<#OK%9BF0p$s
zD1RR3cjZ3611gS^E{{%kk!G;;rm=>5dNiAOqRZ}nd-?H>C)PQ0+JE)8{@2@=yT9&=
zAQG6;Zqg{2&|>=i)1ChVg<{&c2`tHu)8{{@4|hNHRBJU^L2BY!E7hn;X$C;lwPWD_
zifZEfj#&lM{PFqY)IE9suHsiSrR=1YWq&|yB#NXzj-*IM3P(~2Ncs-(x7^cRr2j~b
ziqwfNzK=RS6G^p7?);jB*L55cUiYt7WHL$2Bxy?Zc8!fl@(fPE#qXF|knvql&c6>X
zkPI`a$XOuazaJ858p;nr%|&%cOKlV_wP6V6ma6j>8kqE#3Czhn#AMATmD%VPHGkvb
z`P0n2=bd}Q#)YlCKA@NsbH<Pn)RsneVKg?v-GBGW52YbE9%5*SXor0BYQv3Yd$k-a
zJqD$%rn`1rCc7Q|^Zs4-kR_00|05ug2J)h)tg)>0pzKByJrpTNY@+*`%j$n)WV+a9
zWO@xpr4fu<?3jN#Xpe=)i$T*W^nV)JNB$#kQ4{BaoKnEFbv?*Q$(%R>a{YY^^b$lV
z<%mWBy;hl*Xfis5($V?%GyI@T-@T4_BAXyAJPV|g&XBk)AHimuCl%I8Ey;deFNk>K
zuMWXUyPR=PL#jx%c@jLp@sYZ`2}r(B*KDJm|Bsa}Es%$co=QfK;hL@n_kV0Ibho?&
zcSxQ@CjCHo;cZgwWWe@nSp~1cLh-L%Fy^o}Idou=kErR#g!aV<lG~6dULZ-(R!a-W
zQ0s=n`wt22j2RDc4AHhjekGRDC@qJhC}x0^?Wn(hc=aH(zW;f9LQN=WVKJe2wJ8~8
z3WR9wD48e1;Sfu(>)PeIe}A8cP^v(N0k*Rif__J^Lm>b~%p&f^x0y2-j&WovWPHt-
zKk_tgLguW1#*lzZMp@iMxW!OHC(KZ5;NyTt^IDkiH`j=gMq;dAqfIRX$pF&zYa=m@
z65~;1LQdM=FXSOc7wvYNk`eft-e!*)ASlDM5Bc?|ZNmet3D)(h;eU-yg1nnhyR(Sb
zn-)PiI70_xi&zndU`jyeg`GQ%K?$5=+%1$J5`_f&1^VmsN^b%Qp%5)f$Ir)2Mi3P<
zxIjlrHXl$2lzYlJIU{7mO#L|LBk_sNsE94w>N&AJ&YF}&W`$ZH^z&x2p@0H?XBa7x
zD&3~aIRcP3S}aw@EPqYPrRcYOoD=R6cXM1eIgN>^pMTXlh$UYbjd=<4Q=gN?+<zx@
zqp3m{3-3P8dKe(6q02(9q!v#%ve6Lny<S_neR&%A$oWLhHQb!81lf6)m$)k4?Mn{n
z|2iXgJQl8zQ!tc^#Y#zHzog`_l<k#78+4U6?NbSo=RbJgDt}893bC81S`gl#wF`c3
zqUus{-dNQjBm?rCj%n}v$sVh@mId_gp%m99*llga6;jMkgLsWMRZEe*q{?=c9Z@R&
zh9#d|TNcz5rR+k&BglfjjN#CaO2|_A_%>9p8?B-G)s7NTX-rmWQoBz;^HPVR8hb19
zqwK8qsO+btynniU*jAMU#eA{bGM5eag4$%&>Nq8#uAe9?H)){JiCr*KWIoPHrWl$n
zh1!z-`C-dPTZ+K_avrfQinkt1+CkpsBA<R<A?Bhz{A2awj2}@;Hrq*FWmQ7-SmL_4
zd0GvTsF8BqQgEek)xiNClIy3rv(`mkcq*4S3Nb`Yxqpu=31{lK<$xL50*4vI&TR`&
z_33l>Wm`Zoq4bfyKrC^?hf4Ky^W=T%6^x7D+$*5!fT1ZpIee4g`7k%hr07cSJVW?4
zPn@6BP1Dqd?PuA~`pYfG6U%L3H50NYcwNtF_n^6$a~k?36HQM?=1Ts!zK<%K3%a{A
zlzptf-G5OGN|hA`9?tH)2rEKYucIvaxCiK{)~4#XdwZd~GkQPV7b>QwD8z7n$*G<s
z2kqy#bLh$1Ex&D*xt_rnp`rHi`82+eY97!dy|nA&+xLl=YdTqqp<ZagZ{vg+<m?@^
zU|gK$YQPnf2tC!@BQG7U^Y>7ItVS<wC9YfO|9=kkv?Y<}YXnnQ`73BkYYsQ`*N$Qa
zy>tAN!k$IbQ(E&+1Aa%KfBOpmOaR;wTycX854|Gm1Y4%;aAQv?be{(%ag7No!;Hj{
z5tfax4UR_paB<@(R6vV;#v9loCB2sGTiRkvF=c}n*<wrSm@T$%=Pibi4Cmwu=7ql|
zmVZc2iInPRT*GWh<vd^L<y>P88=z|jdwx6yL3N>$3mME%2H%n;DD2<n`^)AA&dg$Y
zg&I%ot1|Mae(DRR_4~yDr_PJy$<{1Iz8o85H0v@i=gx)sQ=$mN3T$g9{(^1zv(P7?
zjyJr;{@5Af(`s7}?PPJ*+M7e83XAd}D1XDdC0TRbv5qV>m&{3bbD{N@5!!x{7$Yh7
zC%$eNV>*i0<Q-vuPWBYn5|0#XdWkwL-|iaD&dQ(L7DT3QNWDs~S6|DcFnOf%n<_WV
z$2x*X^09vUqsM+u3wiLi^nb<MTt};lczEeK#v^~krix1o4WXqvP~A{NEx}_N>VKxc
z$r*|Z4nk<=2(WtI$B{4}Xz1P^DLK0*7$%Iq;qf?0cPY&O$K%dF9w&wJ8#({Eb5vIy
zi>rk?@mO3l<ZB&^6Q)7P6OP4&Ub16IMTZ`p*cIrQ7IWjMoX3rsrEj3Gb%^xbthP)!
zqhUA!sa&{H(#o8m);m2aiIBD?0DtXc>p?3W(yh&fGC%icTW)H<ajPZOUApYysvN#m
z+}Ofw;!y24<s(mTPB;rlm4gtS0M%JQ>2?b8E`pf-f8M*f{Ai#*{XrJ`@0<)Yngqpm
zBP(n7Qq^7-*izrNWoU4~uM8l?xbAhCPqOV#vGyZp^7=b1);});n!^fZZkL6+7Y_q5
zFg2I4!Z{O{#Njx^0e`oN;yC{Z1Zsj#QJ29;HxvRjIG15fEfkkU>L(|c&}A$Hm*ML;
zk$*W{PrtwY@*p$oBcG^*HhwUc1anjPWPXVFPE1uZk2>7-Q`}uc;KM$q5qb%;3TJfY
z>%=1_jFzVw`KcejG1o+aUYZ*|lg?|yK4zq5;;smpleo&}&$e(8*6Py^PrBMGj6TQF
zppDOG2W!@2Qi3$KS|OcRt=n873SffGDu1?QShdi&HUgv`HEyz8Zj_6XPNFU`Sx=yL
z4hJcSFQ>E`ezHB6PrrTLlnt~O(yPV(k}nKi7C7en%#vx0*mCN?dSShWXrE}dQ!-;V
zZ0&+uYD8-V@}q{qtj5o0?N-GqYE@3ep&W359TrWi%E4s5Lw1+qR;T2u(ZMOXdVhXm
zFOmS17S6K!fZ`;eGA}PuuZJLMLC!X_V`@l_TnEAE9H$Z}$yG*YFX2(l<gycu!B|G3
zHqP02l&$p2syUB=njEiuE7}<YE1KEW;aGw_j-2bTB>og6*tf+Zh)=V0aGIr&-as#-
zl%Lv~W|2WZ_ot`P7o}kJD#t;mw107i@5W0^A-OuT7na$7wev#j(d&~9N&+siH1i`V
zya^w}t^M};_VD*2^Ds?e*wOy@@el?Gv@q5WV1!p@`1$AX{o&V^YF+W70MeMiWi5bV
z#vM~<I<!nHSz{!J>0i{;?8&z(E|eDvp&?jQzX&3h&FW@y@_^(k@?}im7k_bKNI0@B
z7hl4)&WTmQ=*<GmUR2VluA(k{$5r2r(ReAxNFWF=;26nA%iqj$XpsHg_gN0ZBraLa
z?vQ<6=IG1*U&;`gDbp-;dYW>39D5B=DV8I7PL-Dbq`X`(6>JN1tlALoyjsdwoD<R6
zlb8v-Tni-SKsam_f|XFf3V#6Gl%n;hB*vkNT2tufy0)Y&X=KdLuL)L8ipok_%LbeE
zuOu2Jg|+5dP^njaVI5pjw_l9=VX91t#s&vA*FSX(9VI3gnaj@pw+kKFOp>5o1#;pV
z<N;*$j2_HF=jH&i^1g0isRF#Q6L`B@0dE2l#<>MQ`@ox6cCKOz?SHKUZw<&9U40nv
zZuf71_w|`CX1eU;U4Yy0L$EDloq(G>2@n4Yyv<G~jKog7J)gaaw@Fcjw@K-Yw@H};
zyq)KE;B7Lx!rN(GjkoiY25%=0TD+ZiJso51yuQKO6W5Wl9(S+eJImgcYwPt>4z2g9
zTl~5%*ji>gS|Vkgn}0=R@doXLp(*23-9p-*FYK=`ghk>Mjq0)%m=PH{YMHuY+rhdO
zidWvT?I6jTW?}Brb`XHDsBYMIG!zdSk7o7_JCBx+<JO~D@d3R@KhLH4xSjn<_t9s+
z(tfl$+c^H}?K`GpuyO|WG^RMS2D4*7Qb71BJjYwZb7gf5oqtYvjv7<x+XHaG8H)fH
zw0%yaMP-;Ii}Jijr7oq_mvvPdxr-OkqhOfW@EEmzDuph{s~#r~CPPl;E6pekNS;5I
z`tXD<Y)RHAUeJ7ST^7ozR<eWUgE(jb+C>-i9;6k=uwEE_zxSYU47FHv)Outs=hkD=
zP=O}Zc%9Pr^?x=-gi#C$RcG4)=mc`aHLEt-4n?8f!H2aS?f#wT{r%CIB62-cx<HRJ
zoPkZtXlEM|I$1jbm=EnklvG>=eP<sspZ$Ly^55=5&@+Ov^zZ6J8m8{phlH~TFy@|p
z2zU~teA|bZaCU_IeMmmP)iRyrMoDK+zU5wJ{baZFDt}1C_c6moAJ6en08*8cdX9NZ
zuh3AwrVFxG7=V`IRu>e`tht!ClFhoHd0EVT6Dq#@@f0TJ3atG7{O-P~q1@3oH7l;s
z9@jEP!7U3L%JUp299=8h&s8y1>5#Ci7Hy$a<Q7P&I*G(25R=3^6v__H2%@37r%>Ys
zaKJY!u76T!cCdzrEq6@npHDvQgZt`8c_(MmXts&YVQlwp6W$1_68!cd`DHfB8lO|&
zb^09O^A%Sf7A#Vo<g^Oo&)MT|Ztv+=Ltxj*?Y%!IRX|9)*$8gGU|<(<G`-X?xBPCW
z!R-mVB(}1wMCkVGF*?3o)KHjs-;jDev2OmH#D8#_7>u{OY*E?xd5NVBUeHQgfx6AN
zwmL!G-cL>SYbh06+iYVwhNM|&cK2tjjn>w)_Ebygj<5;Ha4_HX_!hv%_Ut~#w}8a>
zY4i9r3Zp7$R1PP;jE&NJD@wETUdGIIx><3QG1VdOsrgE0e^=IObWUK1+it1QbK9-%
zB2U%4N>1+swuSPVqTqN;xyx!o)G^dAhV9q#{fsU_0K?5WBbOv}Mkygn%n2Ok^S+4K
z<eGQmNL=pOMROzDEAD|jHAEYTR_>f*{sCf@E0;mdE)xVXFf=%qP|q(V0y#LBTqHFy
zf5ltxZ=1If{+_>rz`k@JO7VX1NDf7SrdhfIEmkBQu+8v?#L?1psU6r#(f#*(j;};Z
zv6ag9>llHhdnJl@B=4Skj$|llm5>4@Efhpus40}<LNk0KU1~vdmr4sc=ibbM)z+mF
zD)(<|4Q>4`z9F+m{$q5{t+#9xNQJwIe^*=gj@MGGi?E=jNqY@URJ&4znER(i0&_}1
z6Z9~zJ8-dPD9gzLdm|TH;9Ms$1vbfDq`<{W7ddb%wTlvjh`Ojjgt?0q6Oh%F6-XEz
zdP+gEP}h|q7N4_Bn5ts^N7RDUGKpy0cxpi2OV>pllx-3vs7@9T)EK2M)_@^%e-LX0
zQl&?bENESKl_5?ez^vnk*2NMs=SfTur(T&tcDr57AsfCDt1ji<7si>rZ-`j@+_d??
zqy<a4^#(P-sy;D{PlfuxBn8t+uTZdDyWIkhC|F`>Q22)UhcMRav>sA|HJW~eUR$1Y
zNhm4xUg3ddJuxMRBGVAyF2yDVe+tj6)D2g(ZQ2`PrYc!qT)0EUl6Fx;RqkWZ9BR(h
zwZZ0Utq(`xwRKu&^iwCl&Dugs-K}#@p)vQinL?A?+u{{^ds{?AGwRwfo2}$-B>HK!
zifil3k=8Bzl_rb3AMiA$KFvyfhAgekb=ei$GzHzU>g1m0Et_^*3{yd=e@yN1N}N36
zvOzD2OuvO*nnE`F+|m@Z+1CW2)|bP5m>Oo+jJ%sRMNRR*P28Ix+KFlMaPNS5a&dE6
zOdoyq)#B-q2R;E#o-h9SuYZTWo)k_$ynA>2`0;fuV0^4@d2SK9x0;5<H)rq8&R@KL
zd9n<SP|w5S$Cs<s^87ULfAsj*-&fzhSiM{=15F=1J3C!L$+He)a6#Ad4BE(HqCdF(
z4KXl1WPg~9dFD$3l$suV5*ng#Y6isdsGlAHFF>W~0Ta)Pb)6oH|7P6*P*OxU%>ZOc
z*3GOx*!1AY=pciK$B!33o}Zn(Sgwv1KR$i7_-Xn3YCUN`{qOw}e`9~M#FU(_mZz(W
zj(Jmu&zBcxAI?vf7i%2!&wg0Ge*5yt+3!bgv>{?k<74!2j^BE-^)SCaJw03Z)g}`j
z9Zd-DvpFIBMk+(9LmKT`8WIhOha^MlOL9EM)D2j?`0#4wPv5^i{cl)2IXi#7JnxpL
z<Hg?=-z~m5qUquJfAVB?Bw1qzr9zQlf^v!Y1ISQB7#%OZ4(R^H;@h*I&K6I@7o-IP
zW|ESs!yy1E|M~Ibt(m=z6>$IP5oAQs5rEC2VB(A^GRsvfOb{fEYiWW+R6PjpZiQg%
zh`)^yNZhdjsrDJs8nPh)M1dq?$)Wlh8?ILiM-BAR;I&vJe}|OJ5kvF*gdlFmk9LN!
z_<D4IvU+=V3WNUf{5${q;^)=s{l#Avi#KmqKYw@?PtJZ_{PN+~_g}tV{<iqx<?83Z
zzh8X^hd=)hFd+W&Zt?2f*{j8G7+i|9xHvyqTr5`?_~K9BBf{-VycWfX9^G|ui;oA3
zS|obH5XsR1f4E#FvRs{4Lpo$6KfpACt#GYnhPBlibGkGFyd^qq*=NMuiwj_b9vzFz
z8TutZ$l?O9u(+yuj>Sc!vS)oC9~N1-#s+Z4)Wt^nc@~%5qnw+FMDQo6HF{1}Of{8z
zUs_NA=*eSgk-wvoJxtsdUs|qqxwM@lG~40`2999Ye^QB^W)qbFbB+b0JfJCPL9F1q
z2Q+OK6W;EdpX`u`9kQ~+Pj*Pe4vo13qj0qOm{E9%a~nqCwbx)2Ic@lX7hfBRha^Lq
zfgg~xSjbD69KC)`9NjvUGr;h2%kVxtwsL8)I#*uD7I<?+HTW>G)daC6iROUK^|Wfq
zF~eoKe_?#Z3(*W3cK?{=1gqpr%<{B|YWG3rRhWHVv&=BBjKT+%LTJzgL{P$TeGeoh
zWCXg=v_3tWwm&{yeQmecfLksVkg10>W%D?adk4ZT9J@0r-WUk!8X%-w0wHY!Aq^m;
z0faPwkOmOaEI<fiqXh_5qF~Y(D~QS-zzC1Ke}-j{N^RAAXDY!3xtX>*LLtS9JTRaT
zUMBP~5nDTfq<US%-heo@6jd%qL|_F?ABGZVuXnU(tGi;5NOEc)YKvMX0_A&Z@Dsqy
zXDVgMpu7RGsOuI2ytf)q)MJo0v!O#59mF4Zn{B`li%fb*Q^JfT)%yU$9W2|Ae>>do
ze-i+z=xvQ)A8ki%Z*J{ud+k80`o2A{&B2Jtw_HBn>AEq_vR=hm=0k8cuWh-+*~y+Y
z&ho(7X>N{h8}S0nN?Q>x)nns9A|CV{5ubIA9S$Li>RXw`w_PQup=R<O4>&Yx{Ex$}
zXhIO7NRtNEftQwdqmAW{SjSe2P4q}He`o0T%_MgGcqc0MJM~Uj?T@~jht-WkMDo=T
z5xHH6$Tk(>IznW#kL#y88|iLgV+8-TOSW!g@Uw?u>)EQuO>1+ob@QE8aH!Ovs-Nhg
zx<t;P3aV2+bFNe`8jvV9$ufJ<Q1Wf9xB(9;CAO_1mw<};Kt(Q=1Ivs0eRl)NTTe3&
zApkqgCS0=Q)zCpreI=S4*3=7Vm4lixYOK`m9U1rI-wsU93x+!<WBUlljbXCzRhTU4
zb}(7;7EH>h&UApIj@nHxza6!k>X7n~Y)EBD1Auh^a2`7HmvQ?z6n|kuhYcMzblA{g
zLx&9=Hgwp~VfF5{s%nk}C9xmsDf~`RXt93kP!GkW6mW0hDx6F+&3vfmmsAWDbNa-g
z-eFeUpki)*CP9~J)6aaEKO?RJQ_kPM`mkC~=*pClJQ*-IL-O=R_PYleaGWr$eq31)
zxS29w@47`MQ}%zUiGLz{&2aFNlg6GYClgQ}L{RFzCLo)VXs!y{TNx!F$s8wp7R9gl
z<Ql1ZO#!C|1xKZ!pJfup`7I1#ad1Wz;J2JIW~E-`4<<<Xjm?l1sQ}h$n!yMKXYk{D
zMuiU>P-s=WtH9;URVGy{H8_u2Dy>SZnP)hKs~qt+Y>-wEZhyAd264bowOK0-nXZ}I
z-HkwL)v5Bv2;><+pYAn+wEBH4;$o5shH%w6C?&KZTPUEf`0hSMsaOqavGUcp5LQt7
z24V@nK;cM4vw4I71dHD)Y66q|?w!PXbqk2$g|!HHQd8&nC}?Kz2n^rv$!%<4j|e2k
zJvKP8Mnr-(C4bOi^E**9&Vz3~RMg}Q_6moP3eqj@YH<#?)q=&V75;!CFU=60>_^ey
zAR8kDHZgs|hyo_kw_6O+fwiC+a7t@M)mVoSAW!g0+iOHE1jS;g&XATB<x!h2GQ=zr
zNCAVKv)>ji8JXl7z&1D}Knko5bK+=WHlI&q-<#}rGk>tsuqh#pT@BAt{jyWSi)SrK
zK<^n=LlKX@1vc3L-t61Mdu<TTewf<)!fXXvAivEw4W7UVNkgm`z9NpD*I*?~VBuE4
z;QSU?(`IvA-l&}(ZmbY(QQWFMR^-+Ymj#ibmcbe404hK~xYVT$ND48D?zMqS0aO7}
zv%nH0+kaN)iBIwPTpMA8xNNf52t-D0{qRyV=!EJlnQQG5l&}c!2mrWWWOqiZS0SS1
zCO~T|$&XJP%AGkJLj4*ZZEWz(RrJ2s2Dm^qc7Q;Qjlx(t&k>#(>{>Q9WJbbDyw?U{
zK^co24)rLWwGz*^6o^Q`wBOhVTw`StXl}a(D1T9DTSLkO5eM_s3VL<7D*`~|>UXU-
zR|G)sz_TB)dcxwOdcqLrxPXCbA^FYremsRdgnFT0g-3GK>yGIT!@e7A`Z)*k9iPpn
zpL?2p&a;pw=Wk!XSzat&_4hu-%3prJeGj8=sg*yvfxg{h9Q%Q8RxXD$<pMuvez22^
z8%6Jq)oboE)O@?$vATVTU}IOQjQyZ8_MFB=a{uok8)@u0)v@PP$DUK2<KDwMP|PoM
zw$dw0-z7Jbh>1R!y8a)-GiEuL!ALh00X3Jg(K-{Cga0_ifBwGy<KwR{vWPB4u<_6M
zi~sla<K^EkgCa@`W5dN-Eu{08fByawKR__Tx^Mw2g0`1G{&o5F<!_v8YbFn5CTXoP
z6Ei6*jPb-h2pvDlW<nAeNg+e#k)#0WCyk;tm)G%IRz@JP8Giin+l8(C_&(}kQFwHG
zC-@~^6VnV7e@@e#mQ(0X7BV<pc%$oUUX$OhU#yIWhnTfUm)R|~(oT>ZJET@f2SdhY
z6w!&0%xG+tCCi58|5>djvKI{_k90EEq+V3^R@9n1`^EUEZl#UTZZxjBBEF7?pSCjB
z<n;!k|GR2bsL_C{Vn$v`zT$$NBuavS=onK;W)5yie;1_`IWce!$(*6eoFsWq(iFKD
zoO)4ChazCV%MzCbS%VkGS<Z#B&CD5Y$O@F^kk~ybD}WZx=+DRs)19_tg(=0)$qJjL
z9<sue;%-(5Mb;Wt<^HH$)Tqx8IV((}uT|WH-OB2CMAGa~R>va(?2hQ3GPjsj#v?TE
zCv%%|e?d4m8vB;>fwG?E17*YcJO*9~0L6FRlgFeMpgRW9EOrpv))*Yb4%zdGd0CmN
zmZ&DCzkaDPFq(fo^?4_>^jxyyOv|YbDY2VbePH<*cBW;Gl$NdQWZRdPwe`Zv?(kyZ
zD3|0kR|PYL(Y;n}T-D9h^9qgSqm?5~4M*j|e+t)9ZrAii!h-;v+EARKnzhE%+p79V
zRY~{ET1Gy5k9n&enb%6;>O#-bF5tS*OMrTE=BOSKdO%fB^Idm^o)UB<9fulwS?<BO
zRqm9xql~&-s#vLdovXA-o7)|LLD0Z^F2i0=mK1dgIxg)w!XOpa|5WwD>eJ+z>b1!%
zf7t>yb4jbE%Hm*+lubN8<?7>!=54Zjan4%(l6KfytwOkB?bT?RarSvT8%nd*s?I{k
zLdR%JMXuv^6wnzRM1b%lKxfbj2%pjBv7I&mI!5tR0Nr+}E1+W(|4o4IzOqt~M@k_h
z4`njf$;yui!h)Q0xe5aLv4(P-)j(=If3N}8%{#HCBIP@V=Ib$JN&BBfmp~;0c6ZK0
z_c?Ef`svWSZtJjfnJRnO4DbraNjlGWHpQx1WlVNWziiYXfgk-C-zPdO+I^tCwHE06
z@MHm1sJ>0Dl4w=Y=MErJ)CzmW&*ctCMUBG6V3<<flX)jdMQ<rvU?<zY(oP2?e{`qd
z(Mp3JDL|h!I-ssZnD7~+1GdwKO2a6AN~PH@b*(gv;$Educ8u`uo(jBNTX9mQSxPUB
z0qp*YQ)<C>r8OaXuR1erR_f8C1T+-D(2o)Tc|QZ@5&euL*VcU3UHuGoMA~;8YCn_Z
zZf6Nu?sk@tR%UeDbO!64P5_~Mf2$VHTAn93;k=&J36{@d_dLM~tqtF0Zx`ms!5nZd
zsq@$7e&=7bnamTiHcRsaZO_jWjP9&^ZlJ)K9UmwlqOep~K6jwNc0Djq;2eI&K!M3{
zaG=1s{C}Xp2MXLm=LQOl+fkUOof3g6DT8@d3P<~%hIzCLwLbGOZyLx*fAToWtAt*q
zyEolB1O%s?w*bK@#YP}Ft5tl632TXBpGFz=Fs^GF*&skFPVqR9c&!n*)`Kn>huU)i
zCa^NU%OF&gsq!;M51v>66)UX+-{r)_B9@W(5(?uOM;2$0Y@!d*+$5b%CljX`X`;bS
zpoFP*L8t+Q+#-}u`R{F7e~V+TEn}t07LoS}@T{qWQR;trP_hv-KWt!fAMM5050D?t
z9RQA~D4rE3dhn<uV0xKjy2<WIi5K7<V{*8J0<{E$fSxlc(Q%jHKVeeBcG^%N8O4qQ
zxoak(VtO<i<U5nzmGXqru+u4q;w=Yy=?d<p^*WJZoK8)m$ALZ+e-3h#XV0jG0-MV9
zj<eR78$WIrJ58fG>75*-C?BXF?7BD#)0*<pY?&+PyWYkIIi=dX+S_ljKccv{fLy9s
z4hvnLrn*YDV;qPoC&Tx#x2<iwMr@NN@*#^xdthi0GZgwxlbxL&(sBCE9?>_oQM@?L
zyabVT>$Mrm-;pBQf7q*Pas^wgn`MJdvxULh$A=_?A3K*izH^;k1f14%+~S+Z)K|Uc
zcZ#EAw(FlKFy_`mprp}Tg(PkgP36U(nY`3+doOo~NfGbMx}8KxU7f_v@sg{gp{~_G
zscThac!@ij)|m07hbJ_xjPr(?%qX5zlLyUpoc^e%wOrJIf5(h7-Q_?{t_?8b^te_l
zwOP3pns2qL!N`Nb>Xl-0WD>x*ZkPlxip6q1-!Q89jByN+Pj4B)c>?a(jeCt9hm5O<
znpzbkJvg8Bhrw~o0N!vakfY5&uJ3E3-*@us^~MUU6%H+@+S)Z{t8h3}R(HT5148vu
z?14j0k<WGZe_Mk?%9!>UPL{bH=No(M84h-bzxxqR9Hn*2q=zlmZ+V2Xpv)OG)S7D^
z4doY6H=v=pr!3JCl^q9FCy<h%K%gbg@gP$2F`0GT9*LB;k7@q%&=P1bw0qKg0^)NB
zbe}MvU^{KVLX6_c^NIWNJz_q=bvizuV4Tikp~ucCf6CNh`<d&A6;XBedB&UqfX>Z!
zudkyTpK+$AuXSZ(q0V>nLS0sO6lx0h*a0vp_Y`V^DJmGwVXaW3l1-s5GB<gytn4k+
z)X)XhX|ngr8z@jY&L#RPqTTT2jjUU*&6!GFIIv3Xno3<{J65TK7Y@`T3N;21jOGlF
z*J+8Ie_1Ni`HRynyI4e}Zl3U%@}#nPhEeVq?;l?72xEAqAqF22X1eBmrn1t8YTdUy
z>L32L6CUF?Zp;2Dz%VC(M!rH>Y+paYfl4=B(<m3%mg8i1XsMui@5fE{HsceZ#KSiO
zuTeTT>qb7lYtVh&rOb_*w^JI#G}fX9+ay0>JTs+?&agpe?i10Oe;zs`3u{j$g%F6h
zslrpgovPjJCqKU5ohMSyj$`cE@qL)2pY19?|4pdyPY3xnewV>WHxvOjm%tV$6_*_e
zIlq4^9p}@Z@4r3CY`PS|#;@_je|>s?`sZnIL>qxNbk-Uno$vno{uFP3LSbF#loi3)
z?#JKV*QejP)@hi$qA*Eo5vwq1D-b<t55mM-DNH~H1382g9!U<6e-brabLl!<k~0EG
z&G7N#TgR?^`8@hzL3nU{B=u`Ns?ZD)&d`68mP_bjo62W04$gE?;7wyEUAUZ{t&9^1
zkUZ)S-0gve(8qea2O7NqKrRujEVcke*O0|%(r<lEo}rs~7t-_5L-4!pHgS$sV+AGn
zoXwQ!!Zp(rYXHa;v?Y9_N%8m8g*X{8Vfyp0qZ_(+`4nejP;t&`{phjGg)*DUi{-c4
z3OObP1Vmd<Z<oPHHx!qDk2MtmH<z(~7bkz^TFs8zHW0q=Q|JYj;S7g=z`{W6W&`xr
zU~ffFMUztw&HEoZq)1sDO0vAR(*iwcZA+Zt{C_jl>fzg<htE&Px2F%Uh6b;K1&_nq
z*8%u}=D(o%rn7?wP|m>c_I>#MMQe0?9r+@yjp57l_rRa_@N6LP3UK9u!?P!NIj?`k
zQ(a^1NQJRdmrp1=1YYTb*NNY;T;}k{+pkZ$h^|$1>1+CszdpS^{rxm4f<{>v2d9I=
zK*OKkpV9*YtDKLlpYp8UkH3bmPrpg7b29~CGim3nZOnvDSsMoDqB7~xXeP@_5G!Qa
zJo=agpD2js@pf=FD8t)S3SMZDsr-LPNw1^2ikNKr^W)nf?tJ-t=IlrX@{%+BlCH*h
z6NNV{wUZR4lZ9Qv7oGU}G~Uy%%UY3FWPIgBG}<6v_8l5&WuuW?w=`lufJRA?LXhm}
zmr<5cHY&eOEVA07Wm%szPuk+7$00@Zv@EeKZr(DAaf*DHTph1B1gjH_WVC-^2dy~N
zwbF?N+L38<hnaqOje0Sbty9KCxl6W9INK^kM)hUcbu~Kz<RT&s@M(v|`JRPy3|cM$
zEWoOX(JBk^Y~m%E`8Srxj^%riBJG+SnRjyiE;Y-6xR%sS^r`r9sb;L!)l!bEZ&EG?
zm@9=Nnd4(pKh;a0)2LR?Uf_S}i9KM7&8mVmtqBs>$t;QbRJo4^n)sx($2sRViI*Zj
z+o1$Or!hvZv;n2Y3#DPrW-O-C!ATCIBPSb$b4t@u+G{z?CJj&3+A9`tWw+OJfHro;
zOuuCX;tY<KIvwJs=47n`?>pO_=FwsG;Tc{Q;bT>r$gsz0Bc;(RT$X>Ovy7nxAh`f6
z<;7)$rbx=eg8<sHC4GP^0NN76DdGbFT5#S0(1PLyfVRePNj7%|Xu)*{KnsdX0FAC@
z*}|Fv6?q+v0KHkR){TRxqNd3>?!7=Q7;V|3fUcm}t?vR%t|`Zzd?edIq=3E^*yDXw
z4EMo3a!n-P0rwt^UX*_t3zTwSfnMmGir^#%fnK}(y(y`Is5FF5Qn!t*>|-S8sBT3V
z<B$`f^Y@hqV?t^it>p4vgfZNYf^iRm7$Q`t{S^>H2H5!F0T3fN?|>LVaRbC~W6hIx
zyfefIt~($`Q2aL_hIHh5e>=jod>%9jwwv5x5NsKyIq(*s%k+OSnVN-j*(+&w+ZYo~
z<WAMYgU3V@Jom*o!^>x<G41!UqEp!?q$ltqRY^MBfU(9lsr)~fErrJKm@Nf|J+tLb
z%>IcpCeJtlp|i$z2SCm!mbs(4ea4K+QRka$XH1T1^g(hd9Y7cPJXIL5P2|^J0d3$&
z;d07XFxHGEvAci!<@2gWU^rH|dWc<2n8iRCvm;-nnp8v}KR-y^jQ2~Kj)q5N7o;R(
zT{(;%<DA8psY*Xnt~1%IF8#t;6xRICYfE)Fucx@(>1k4t@kGdSZKucWhlh>}m8v{e
zo+?$gHjdd<G+DL1)HP+?c<092Lm4D{iTBqY%I7AX7xjPf+&tEJxI^?(!oEIcwxZU$
zUD0TbQxG~!lsU1>Q3RHwF;~X*86e#)VJRp`rLEc0%o_8{`8d9n8l$Z=k)=X2u0aL$
zoZkCV*OGF^OEYX_ZZS*A+!jBUGWWcGM6d~cCKF%q`J>R9ASj>eXCDDO6Na>c)W)?8
zCJ7o_`8j{G)-an_j`Gwtoz5&a-aBO?!-VsXdeexU#-6Tu(}>)T`tXQ1je_%zDN|5f
zF=bviBNffP<A!M0dqa!d29l)MJDto+%F5`0j+-+VZlC-FqaA09g5o|SSFU~IsIsmz
z*Se@<AFJ2vUR7y1Zk_^^7;P!c`q+jnb+rH5)P{e$Ney&m*9wpyYn}3owRvXTdIuW>
zxc!6?4a2FnQql|aMLIz>7$ntsGF;IIias9~G`!T08r|<Fsn^a!(^+??;}ip@`4m{O
zG$S#Vsrj`wGYja_cx1(gG?r*_!hNb!cZ0<%SnXeCXBlPUuRb0p{CnaZAgM>aw&S*v
z5&wUh*LL7ZAw1%>o#4C!?F7Yx(GGZ!M{BM?J3+A<7-DI-VCjfr6%Ep^81L|QO)%Pl
zfP&)ZAs`UsA+hS*Snl0t;ezXqvv5IC6j(w*Fj^V!{xukkZOYgKQJs%Wsjl@TF4a06
zyXps>p}SKwUZtBeF1Qmy+MtMy*D7%dAh~}%O9y2wk*zg5>$bgXATG8+?%K!ND(Gp$
z5=gNDF5qlbJcqP27%!Z*I^{ralxuW7TK3gB83fC0c!pfP=<ds4w?8O)xXn9+*`bVY
z*5brJmkBUOtnJH!#7jpafvb}}|I_5o>QwQ%zVIoZjM4u9!pG*+3T19&b98cLVQrVo
zNH+=sHZqrSzXubSXcsxce>ng1^WRTAvHte?R9KaYY2ydO6_XZ*FF#Iy4A}AG5PF|D
zQ{L&}e?w?yLTc~>XGREX@aYx)41Z$B55Ir=`nNCt5)6m^m%8!+nAF0%;YWP6hJn@G
z4d2tdV@@higJ7O31B2ttO3Q5-SQ@?zCqP|-lCtXdomv&Mj_ax~e+bQt1F6Ods;10C
zpqN|peBbNF%+)wWkg2diZ$u@HrB@uEN6{;N>i66gNBw6weH~7pl2)}L)pJI(5G8N$
zXHKg)x3FkXm1c93;rIuaG75nmOjy;K#)-74O`^~yXi75gt@T5hstBB0Dxq@|lFAfv
zQ+~GX3rc7N#g<~Bf0}8O#4<yM@Z%(sW}|k;zrXxQ!@l95kJU4hc@oMqfs~$%MNK<F
zw1N-g%Fg+4>YjTiA0{u;5<lp>;DT4>ndZE!A%atW8Y-58XiuYTpZ2Olpu|e<P)BBJ
zMh$4!ikqT_g;16y0DRI{Neza?k7@|2xZo32121QkN>Q8&f6cE!OVs4C2~1KRDbb|S
z7`^~T3fhj{F~Vt=2%Bo6Kw|7hO=g_8tkBql=)cEL`06Ec4|YNXd|1y9@3>MkssJuA
z#I%}_dC<K6KzN4s-i}H<hoFoO;he-p3n4)15Km`XfBm@3n!-k_x>fQ~_MqcS(t}RA
zpcIl<&8F3Sf1RX>MVY*3N}6^y@pgWa0;0n2CR#B-6j2JhZGhK^O@4i*bemmUE?dkV
z4M}G(Z<?0rFqu9J`y8H9<{1$H7|$zPj&Iq8sBhu!B^rO#zw)NQBSg_ea1o$MRQAO}
zUUZWM4Y}xPL3=Ker8fSKc*H>-V>5-ojEsGQx@=)Rf3ZxX1y6CZSy8Dm`4scfY0fZ1
z8I308ObQ)A0vOl14m#4=Cg#Kzn5^|#5Uph9OBN@U!-TI|{=lp{x&+NEX6;rgQ}JMG
z&UW*fG(dVz5$Ya(rG#ZdbgS^7kf9KsIca;H?nwyIN|V<3XqjY(NppHxvo5a!#L<<$
zVM2Kve>H2ZDP7h9yjm-p)=I^{;#wudUh|V$HnfbcyM)&pcIFYq2oEBn_$~ZKpy{|G
zfT=WsGJ^4*%V;{Y-~3wRhEym#Rh2@(4GUpAb=MdP5%<;ZkOM{#-V}wmNF^6h8sU)^
zg@WQqbxEm;sHqgSE0cjdLsB)EtlGqdOXJiSfB1E=wpCA;j%jH4dU-c_@evVd^}GgM
z#sx;GPh!*KWAlcK{zq4)fpb_<YEnLZ5vmmg>&7svhd=*4d_8??i6*tS&vEUzF<MX8
z&M=LrSW_SbTWg-QjiqSh9SCXWU%`qdEtU0<)ha1kijA3<-hjfQJWcQEs<lgo%)L$>
ze>YQ+7Q?wKRCsdFoGssvOJjgJ<g1}9exc92L{^b!wIgoPDEXD_SXfY~N5@8&otb{B
z3S{LyMWnZq`Dc|X490U`q9Ke(40HE!g<=5ms-?Q7P=jF*BUCfuCWVR^ZW<mIG@8^u
z?JQFm?^0QMm;36Uyzet=G_8qF4q(^Ue}-!;1`1D_;NNj|Xvw&PhS<!CP^zGUlL!0}
ziB4Jq2Jw70(ByTWOhRB)U!spUxnVJ4SxZ||4I%w@y%kKweqT1k>{?T$ZyAEFR@Ejn
ziz`Y|tX<U<jKKr*qU=NP%UlB4Kw%||7{RO`6|T<!&s9B@%9L15=d_k;y2XUAe?0?4
zcn$3!sZ4R^eKoHN!icm(G^H>>T9>LmX|^2_@_NZwtHxa=DlKAujgG8lxlhishz1N*
z3*E@|WL3Cb-bPvDRw+c$%5<9QqwpdgM|o*)vtgPB%Jr!c^>E?3k@Hz~aeDL`qr3SF
zDmso+w{R!`U|(|phaSgfR8jdBf0|*2@VHc~*lY(D1DpiR0W5SMq!ZIzM4e-hWzn*3
z%eHOXwr$(Cy~{SbY<IECwr#7+w$-ohJ@331F@LOxwfEl{xia$`gY`|vV1&xtZ1h^8
zB&a#8A~fVcO2npseQlNrqu2^_JF*6-7AB^$qDxI-3(fVD3rpF^RB1i+;Mb$9uuu_k
zOT!DXZ$G6+8ScGD>5=Gyyy!=?+oIB|?4msDOFtD&?oin2V<+p=BOn{4)82<)l&dc-
zh`iQ_l-Zlksog=+P?wR`&5>rA05ug|omk~7r(jF2hm?wwG#>m2SC+35A9N0~?H2^5
zbd|@Dg<0)$NHDmi?6>(AY8Hm(Nf+kMvZf}O!*9Zp4dP?@pzIt`cM7L}eKl|m2Of|P
zC!RmA-js5}p9}lw0htu|iqw|~TS1dmpskX7Hk<8;kPd@ZzCEQ4=xwAY?tAf;^w|{o
zz5~a#1MkokM7O)^!$BRO^)7d4(Pe0-6%(PZYc49_&H%ynrK8W(B;qi#y%wGrmF*~1
z$K2)G>n^)270|F{2@XL5LR(d9t3g*XCcTsm4o-#iFqG3%z_ySoJtJ~byNs?3C4Qp9
zvOJ8%WcEfRaH-wN>e*UN4qr01Or%5JH%u+q@_3mQp2DvXQRSTm*DQNJV(<q7o@qQP
z4`B0<b@9r>%-dBwi0z}kX&pI?+;G>CW%X~8;!aZn(uD)>x7r6DdNS25k7FXvD=pcm
zaI#}K>a6{9fQt2_z#^@57rU>?_++B&HRT;ES5W806)2@0?_z`Q^+Ihuj=ZqD+LsqT
zQIMN~&V;uYO@duo=fbmdtTOn_q@Ke3=IPLpuco~iqgk$>bXiebn&rM7uScDvg6gdm
zd9~ulm6n2JfL$*7NRe{`gC%mUl7TL-6rGus=R0#VAVduAB=#S7vncCZ{6-=6ovKa5
zm)^<G(;sc#Dd}dLiIZ5a+Di0OtfU*{i!z;{cmi;@-)){XM9t<JgF8CFMXip@Fp*Ji
zk{?Pn1z6fQseL;t#4O(Sm4RXLnA=+svfv%y#ym8*r6T)XH0VS-92b^~yIVRqPB1YI
zi;qEo1tFVbp(eBeb+1xOnUJBEQX1Gkb#vk5*O7)mUApChlnX8G0(7FBWLJ+ibquqY
zq6_mMPpUhx1^>x$*=&1zk7IU2eV!fd?=$n9P|N)*j-8W7Xd7-N3Zt(Q@45<7OA{1)
z4R<sK)M}T?yAjd}u&FtXj9)RsEPtIQXik6J1cZ#RZDu2GG#jcIP#t{sok_ftAYk%6
zhFi8tx%GWO%v!!N(X25KJF%Db9`mg`I>{noxbQE%%}Yu<&bF1<&V$Xw($A{7&08lM
zN}}$g&4m_26bxG(Xu5}yp3ti|7kDVl)Cyifk%$EDh?M%z))wu@Bnm1{qSC$kp0ohK
zkpLY2E-}OD^&nn*YUDgtOkwy!w|jQ207v5%J^;3U>Qd0O*>6{)o2`Y|->E#$cO7Z1
zTiMZ3cvp+nXZiOL)9VTx4~}l<V@udhrTOxA<to2X+n(I(uOm^a9K1DlFsRUr+kb@H
z-(BeNP?`#>z7NXJ;7ATHAx<#Cs)e<HkiBmDYq32MVi~6Qaw4a9#D>hb6QZ_Q*yMl2
zBv{J^Pz>~SL=JeqR^<eu_-?TT!h1Q?<vcLZL6h?<FusVwdocP!O?q@Bx7&LKPp7WX
zXseD3yazfGo(>DPn^9gyDLW)SHWG$0BgpAA(1;)hB1B7?#}aG|Z6-!npmSaTQEx~b
zM~SV#5q_K?Cr^k2)gs>dSGPX@+BwwReSO~AIYZF#Tp`d0M2W*iX$PP~Rgut9Pg@s=
z^oTow3dS3f#E0rW_n3xV;Hp(K0<fhE>YpKV!$2|Nk{JaShxu_RXf#P9hGy+kt72@~
zKS%_&J?;Nr4*$<u`-~JPBnSsvdOs{bS~@X`FjcxXDR&C+Y5El@Hzhzf)-<4pM{7AK
zDL8QC54x-@jlS13sh!n4b&x~?=^nXMzi0NhB&wNo@`~rIung{@|FPkTIsFr0Bh4TQ
zQ_LYJJ4`WolWn3OCY4p>%o^ORLBYKzSb*wcLv6gO=}213+78wmF8e)fIXx|<rlwKB
zrmBE;f>rX(lRa=R>=%H`x);~o^C^-D?Q9wOe&R0c2&SXyYC_umBG6yEJNI+yW<mlb
zR9%%(!)TdoY?~W8ksqvVjrrq&hMX&wGMDYw^n(JkO=_ZoyFio8zfc0c2UC%ZQ_&2*
zkP~X;p|7K~d@t3!r58I*aN0XUdsKf%L60UB42r0EOX)i6=W_r6_s<~m>503f<K4#t
zE~V5oe@uMHO_;LU9ox!~H+@S=cx-5qqfGJGWbNTb;B6BR;Qq+3a)u(6gkM((GgZb%
z7U~X+!c-{=lgtqLPyCNwe3bdR2`GSF#>Y1{h0GXKW=aD_YUNcot**a0BL#~T?<E!F
z%_R%?fYP$^?F>M;+u|x8g`5ns6b!sNygFE|LvH^&+llRHrzE2M6r_qMD}1gjR5sm;
zGk3FoUpfv%@`cyQp(+f<+Wk~}`+<m9yAV|Cxf^(eI07`Y+L1_o(bQdn0{h(=+FBv(
z6O2$ef6zA{HGO!*JmyR`1088XC|{)W(R$I)&qY*(mLAafag@|hNqVbyzDNP~kdoV^
zhpMLI>Tz0Th_Hs`mtF!|jZ#t$U;iufa6b?nXO7Nb`Jybn=`@K&&A2ySBpzqd4%C;z
zqtmVSH8gpEsyZOu9=UW4(oLw$btRtE?k#UgWP`j$!`7VFR5GbvVuu5NIco0dfI#Wb
zwpKC~kr2R8(B!F`?C+^BmRqihTs)dp7wxk(P;Zmq<EO_=JAk}Dj3Iv)&WD{l2v<gm
z+hd|LqjB4P92n9ms?W*V;@}T&)#?M97=2f{_0pz34!pB}dD-&8A&CExbH4WKKnNdG
z;OYbltf|3@K>_^MSiAnWe$@<&B{Y=Ip8QQGpAKNN5^mKgJ*KQjwu6qPYHR1&yDeNI
zE6fDzN?YAFTFz{44g(WNi@|*nhh)@H$!%A-$#OJfDvdb-){L_GfNIe!=WpkoN-36}
zzbhy>$uXqapFvJ_kz%v-r|G~hYvG|%X>%7Hj^!urGkPDDJxO>?n{}He!xpExi$2P~
z&ia66EiX}O6upNnil0qdzdmQF3TOOTO;x@)Yq&5ng7O*Y)@%G!oNBDG?1ECK5020G
z_xsyjykACP?$3`e2@$s-L3csZv*P)Xd|#g@k6(XZoR{pv?+9Aci)u=l#ky<cr_4*d
z!X)YzpAF}o-5tFRY9P4S#}@A?dF);e6NUjbBJ8KeC>TAVw#&%q!9o<}qKN7?<&*YF
zSyfH$_KwM^c2aJ67mwild5wVA%QK}e>Klcv*x(Pc;Rn$`=V#t?W^jqy6yv}#Qq-yI
zU(e0UsUYF6*(ySp>~HGoF_Wbi?UobxjSA;_K;QCfQO;BB)+(nD`pa1!GF;XkO}_yf
zZaz)^_qDVVW{~y86>f&!6D`Pj6oR2DBuvkj!|~w8>1|h)zUFKZNn_qy&LpC4!E=Z6
zGcKl-xY9(Rj=MbORlfp6>C4O-L9u23lq)V<%`u~PCZ2@#FPXR-E5Wi(aHIB?1#^Q3
z+g|~>1Ipl;)JHD@O3Cu@s4lMxVcr0;?kK5t8lFt$PtCTp3R*c{Aunkj-1V}R>$ek=
z%#$3(m|jh$A;%wJ6Ak+8aoxK~LWLxSX`LZL!(K*res1q?Xi~nTVI)nlHe8}qPkqGs
z8?yxF0vBOvkH6q*)4AtEE<}FcT`<!wW@%^QhuT_YwkQfYr|dV7MXla5O_u`j2j?K~
znBux9+%_Q$mX|Ifqab3%ZBZqdLt-~GO4U+a){dww@>qPfj<25)RCe_;{-z5!N>a(`
z9t;wT5}YB0p2K~G-&$ZGtTg&81(t;T2*)H%8U1&K^AJ8t4{vg_V}amJ>u?A7Y|C09
zs;}^xA@eE~iA%;g<m)kGhCTr4f9KTPyB`Nket+TO_K<$kl=VekW1ef8zg^$S!>)Po
z()JrVU!h5S%$18=;~c&h1hXxvjQF7;(KHQQJn0K@p4(k4U0uVV;zeZQQ0`JQ^^<CX
zr?ic>eZ`2E!fa;5JA6#-Cu#u~#{@gi@hWN$a$?{tvrUxJUP=V+aAg6K$%w#=1XsFu
z<*bfls#lbYTpQ~))ONyFJD)Ng>LxWkDYn(No?P18xnT?$SpNbQS`d7^e#EQgqqUJV
zdb@!LT*^2S<_7A^ZcOX^YVaRuH<iz}PGyN9Y(?NeO}SwEp2$7&lvg#%P5j1eKtrMg
zx8Grtbw8HTrYWE!xWoc@&wzU%orWx+hRzz7d>1@|5hF96tXEM(?=sLk_izFoRQq~U
zlQ$F&c9xUZh1m*ZX+isl3ECkYQsU-M6{yl84GQxH_@!<%f0%m@fnaDMPE1I9;2dF@
zY+~PzAPVeL(mI<zvGq)JLwL(?;sVjM);gY#U5yG!*Kr1#u2>M~`5qKALk?(TDRQ*V
zH18lY!o$>DbEo0lz!Clfg|A&I{O|aSnN<iaeTJ5s6F`?7NMuB&3*i#O+)7kV^gJH%
z`VWyhS)60sV5wmQl?XCUVq4GVph|O@9&s1p{`T$c^ZoI!&E8&U>%`gM#g7mhH~E`J
z8grf?M+GE~rz=44(>Yaz)PLp?m^!X7q4*!a|K?(&a-Ia$VFC|7uhOV>bx$hu$gwt0
z>iTDm7vMiW>{`!b#Jq36uPf<>Q^HyU`|&sgFlo)hvMcgLykCdlI@4C~yNA2)BD>Ky
zNoc=Hf)YnTk8Npo0Dpa*f2gR-`}A|(zn9<sv89FgvF|8~)M0Du(x>8v`L~;}rN)%O
zQrezwzSEhYQAT->e+=82otj}GGi^`=_!9sE{C)i2Pbzh!aE+Pq<_er<@c%RIYXH&b
zvZ4<T9!%&(G|Uem)HFXx*rkq779=Zb&J|>XvHb$>hq5kwEp9y`5KRUhcVr{qWpu5r
z=9}|)noV$w_&bq4s}7JyHVGymS0OwokNLwBB8NJ1XF&noeldXpFo~)ww@jr5C_(5a
zvl3qn3cI8u1o=raHyc8d*321M=r{?f8;Z4$ao#tesXblR7)eCYMGbr?y?Aj4@yCBN
zXelX}^`L|ZfP3>qP=;CWNM9V7aq{@`xe()Isn;0n3=*bsU-$@3)<WaIwlSE-c&KQx
zE;>dyuF_eQV2<p!kr1BE*RJ>ikgeP<cxSmRu#oIr^VSrKTb2dV`54Dzj}-oxcNb5{
zSQ~)kvUl!0N;R5>(W~@(!7w+3NWPB{xyOGG^BQv0_pXgo3u~8~z&C7xZ#wh14pLLE
zhlEG9PUHW&0v^wTM#0)oo@9%d?0y3kZ<40H<~2vGiUsm2KQvZHQdsT-+{Cr7Cd)#T
z@$OT)NSTQT78~hui^&#BbsCQVvc#=ik^S04VM#**nOw&h$4U0iHqSdE_<`{{Rn;2$
z5`}rXHV0gb3_zaj${}z0n8D{7(Q#A9w8yW+XRrLsz=aRnHqU8jSs?E#Gqs8DJ?D@G
zNJfO$QirJM`HY`Tld(hrZ`1TRbtXD2ru_`#2+8FqUVJcAqim#gtgMC`Y?iPUh9hPe
zXW=xJ*QLdf?03M6sH<}+IRT2oHIz<5FkI;0n$VwvX+S;~SRts){+PKj|F)(F)ime8
zIN;5fRXJi@LSRBDLchW!&WZWcMI+7C7TBw67IXy?q?7uX7xcOT_d0WHGwRw9KzrSr
z+xOqFXG9V#W*}FrnbcyY3nc5oq^Yand_D2q)w&bmMd+Z0OZ^-`k=tWzCa4HRR6Ju`
zJ`81{O?kSs$H^@*l#<+ksqRuahMIFuy0ZS-IXYOWoZ>LSf>fLP>CwQDGqO9@kdzVE
z7D;BJ7+LJNGo1DVWFy8Aq0>vOtt$^7C_lV(c@7%i^85h;{HXEir$hI&8{-+DwpyLK
zureme4Z~WRE7Vr&BX^ITo4RT0B%Ly^UKT%hO#g;Rp?84DjM}C8V*_REnZfvxSAoll
zDmO2gR)v@T%oEMbcI;a+NHJR3jN`08qr!H21#G6}Msp@0yKB;WsBu|$DWC3Fu^+!H
zdHV&yFy$w#MK9fK8(RrHMXJ?3m}T1trEr}k=K=noe=v(`M@G@PSeiSj+v_Id$I6t5
zRou3>wmQYfb!>HwPCX5HveTDn=&+3`p_mpZ_!5tl5H4=4JLTwt%jhWx>a@FP!6!Tx
ztoP?z7?tO6(!X6W43E%#gEuCE(;=C;$pGMz+5yHl{aT9c{d{x2OsV=zp`2F&ZekB=
z>oxRtkrzj@i<I?Yx-V1I^!`_)oYS}MV4PUJ7%O`P54_DxCU3Z~`7dY2SW#*$n+!}E
zZU6kg_XtK#lb<5JssuV>_<=ZST9!fY46Nb~V)gkGC3&DL9c`CYD2?+%qf#r(<^lH6
zRdh*$YzG?~9M;C28|pK#rk8CXdKe3cY|67|O0?6n$~4Sp(i0XZeKyhVg4h0zCw<Dl
zog7*wEUYv?bjs>V7aJ-J0bn@yO%SrHIam(_qU|JMEjt%j<BxYES$KaNt8lsUl}(Oc
zbe^4g?<JCN^351@G51A<^|qp9-vOM9)N^e}4rfc~OKT~k6zG@p(>i+5z33*HMe1)+
z+6qnbysWf}=jyCS+c*(x)22KX_;hU1_Zeb$nvIJ1>L)@Ek5M=6$9^qwr0YX3WQ*@n
z3>U&xFKgA~C{6IddFGq?D;wx>9IBlrDI6PG`w0vks8W5pMZi9KN0f;FOaM^8%3L{C
zna970cufWm<>lGcjmC0CC_J!xG=4MN*#@=@JbW2DjOdnmL314+=<6P|<kq)*XUAK8
zlpJEB7|M5z^AGqoFAXu>(tA7F%E83mFV(<aG!OJc)jbj|@gl2+X;Affl9WwQ1N+#t
zdU9OsPDD=qDZ*B;0zlfT4FRx~DxFv1S1R~m?uVHE_CQ*;Fj;lwAA@M`?vmEG5UIUV
z2QdnMQH%xLO=J@d#+JT!<_SLleQ1nGa4hr)ujGvT=7pgL!VZe!v`@R6BPFXW5sf0D
zTnHIa_ntXF<6|e=(O2AO)-VZ5R<RV61-cx-(Ar5;Q^vN)+ExBxX9L(bY#glDdd-!l
zFNHtd%!w+0K^9ohx*;CR1T$(0lJ~bXcO-JrLRTQ0L2Mx5cs*Z<4i9Hl`BrdrDK;Gh
zjZcI1ju&;Zvs7(c7`y%Q_v+)^Yq|xtpcUHOBDhk&M6$1qFRk1>wBMbNXI2qKqR8+!
z=N1eT;eTq{{r#z|7OLEq$>n={3jv}6O$%l8;f`7+4gi>aIJpW|X*eTIsJr5AJmPUu
zui)@nK4jp-AM^c=lj~`y>!DRx<6`aF!7G=fZ{3{*dL-#P&B*P87sJI%_h92@q!@|c
z-%0YW#<Y^~PbUDm^uq-Tsb65|g-cwrpH+XONd9Cbt@RQ5-vR}M<7fZm$B3Q&EX7Tg
zj>pcO2z=aH!_K_|36{HnP5-~Y0cGcAN>`p`BuoFt!Uqe&#grc8_JbYLb<`b24tS~4
zB4RuXdA{6wRsn->sa!;Zv00;@TS0=6wm>b7w~|szd|Z9W=3hcKU!bRFDY}5=j=;N~
znR&|IE;TC+MT5FwXG&9OCX!HCMr)EOr%GLVqluu|qK2loEx(m<|LlNBUg5D0QaeXY
zgb&d=>!OF8a3^DyezpW8?%7xVt2VwZ6R=gl;X2wH-`|hxBGR^@ByL(^fImJQIEs2G
z>tiqfW`)_>;KG9ZtHzrdLX|!j3Xf(%1$H*n0mcn2^`Ol9_d~Bbg<YyLcZf>GItO;b
zX61y$iq_rToJu9LW3Plr5~5bP)D5DrpyLqSdXk#U3R|2k=*J!VKFj1ld)~+lUc3=z
zhEPVzD9bdLyJXZJTLO+%Unimx2+h_+2EIfEvV!>JhMfkgv7yZ#F_rmYMl{@nNJMN)
zpSzT(Pew_9NTcrp>^P~<B&Te}?FcV!WoA{F@fbyHO?}YzyHF^Hl3w_Wy;L9sN6JbL
zHoy?|6WHZ|bpRGX#iVj}5$;3dx{6$_FaD8%UEP7#6KXR_Zc;(AZMGp|zarU3ciSNV
z!q>z6k}r%Z-yD-G>=Vg#b{*euGiXM?`O2S~^*1bT4~j`;HHKSz_#{|gS5`lmfKWCp
zn>N@jwg2Rh2j*`GO`M9G<*Gnynsw+Fc^#?Z^Xw*<!U!#(biq%j6-*j&?HWFEP#~b6
zb{z``oI>Q)U9??Nsn~(QkRCN*a#BIJwdx`VJ6wF}b+p<cAU&{sg)-lt!|&x{bN%-1
z=*SN_obgz*&Qh}uC0UjS`<u&;)h`NC4#3itk;`cCS8|?-9z{aUFwof!3dt5HyCsEk
za5B3w4ABeV=loXp_rd!0VR-n{#f$s%XqfZ+-y$JKOzUbdVfYOrhTvAt@WY$pU;?LL
zH@^@6hxhBV+uhH`5<^eB-c;?kw~s)KU=Q}p&eUvpKDA3Uo1S4VqM@FC*LQ#if6kL<
zy`PKS+bLK2F2Daf*uQJ~%RIGgP3eQdHkm4BVnb|zGM$s*4132}tWLGGV)!g*IDp%a
z|DS_^AG$yo1`TX$uoR;-cGeh@sV21=X^z3P$z7~9wa7shJxNN<WMC{9A)BHrlmFk1
z2?*>K&Ogy&9{+T+v`U4s{j^Yj^w@gd2^12~QQ>&~%|pS5bXxW!G`s4IFzD#Cp-VwI
z#Z0e&9k^Jg*u3U+h~x0N<j+mq4A=PaYZEMYtlwPK{#E$+t#QO%UmZrdkz=_lM>kOj
zmu*Pbrx~F?AN$L+kh<7a-prp*j(`_E{3=tN)RD<-PDceEU96s*-b}-8znJ!5E?2C)
zd<Lcu^S!$}#9q3dfOanZDUdiqzHCDVrhZnyoeAjFF)QDL%rMPCh;D9M93s37n-B^g
zpFa;@Z&rS94!r^VY_AW8ALoa;bAw6LN%`(95fTQVz3uwe9`9@T(kH;xU7*u|MEKw3
z=K4j<BI9wdTL`xV4ZQKw=Oaw7I|o7fPMmLdXoh|$Wt~J~R76U|I}l2X(S$B0RrDu-
zKH(I&er|bqecHrhyN0nua#ZDlcg2xKn~?mQ8Ll^^#<hW6m-jdzvZk9+>yJA?ivyxn
zO;=?HV@riixhvZ<QDU@J_6CR{Uh4eAIYDsS@59d}ijTMMlx7C4VCEcK>E`V14X$ST
zV)f*#x&d3K#Z_k8^4Jj*Q&9iY*W(G05QDz^eu%XV>;%8M-B2VgtgS6<NApMe!wNH(
zH$&E!>Pe0oMv3v(Moe%pN0|Um6Bszs{&yopSL{rOh#|xco&Af>Y^{SEnv%@W6jcv3
z34(eOb26V=B%+Y(^}RLtbo(r&fS!z+z2vBkH)oxNYx55M<nnK61icH^pVThE)27=o
z=JQ3TH>f6X8mUvrM(VF_J&iNnZDT@vqLKk4SqUO7nc5AEw1PVglImMQQ=yt4m6GQd
zCT#GY_zm9N#jxBdexg=I4P+hKMK)~>HYb??>jr;_6|FK5szcaqsKV2iN=20{*z<#Z
z_E6T#LW7YC8?^{K!ysZ6{D2go##_`h2j*K#9AhyHs-158iEcY={5zj$4uxg+L5~6P
zL9Y;fwiJ2;NilUxg-@6$@6~JnfL9=wYIFNyK2+jWL1+|Lo?9!p*m%Q^TSePZsX5v)
zmAIVJZ6R4_Gh$FWF7%rZ_T3FDjdu8$#Tu+$%acn%nx;IMY%p%?Y~mUq0%bg`#F~1Y
z4Js@nS`N1t$=CPef{sF5w{KkITq@t@82CI29Of;~X5U6mreRVR^llp_JI*-m%Z`Ln
zM{`W(2U4WQHUcvb@s~+;22v&yHFJs{pUB!=lA&KMyUo%lRX_L|QWReaM3vaZV~9@&
zd+Y2kUK#vJX<q_VQOyBj{|JI1XLPxYrq*H;#i4-Xgu>@9AP#Av+aU}Rb%E;}!!H{V
zv9V4uCnftdI=0u~!7^P87;}S7q}z#ah@vnV#N2vqoVQ^nwoP?K+N1e?9ky1uyl}Wb
zYux`b-J<oDL8SHWXc$Zxbxa>sYh6<78=^RmuajE>do%;_=|BV&cy<aGKq0=`m7R{t
ztjj3MUHlD};1oUszeo8@tGe#jX3qW{9_uq%Up}C{jCeJ+co43k3=CSTf?lOEZ^p4x
zvw;@hp<^!-ca^vt&W(@nI`3a{sAnQLSJU1rn|4}Wv*B_`lF)=*5S^;!mJG5^!9jnM
zR#cRaBz`VLs<I9sO1*5s?!)J=!Mh?8{N&0BtHH-VX*tjUN<DdVYNwEvdXh<KD}$E{
zMD0PZKcIl@^RRnz;KWTL0A@z)VRJ1>^==(DkNTPzzA~_j37hCo|K)F8)tS(pN88)0
zt$&a~6CnKF(g0s&>C<K3PYv(iX2&kb8PkAIi!SOqoZt=keC+okTjU}5^UTDTY!9Ar
z8p_}%g2__|o2}#Yb2hKg2KFxEG%n~i!V>yks0rVkj~tBY!*iNlMo1!E*)SmubAV(L
zuMd(VZz?`&a3M;lgdU?;WN`LtQmrg-qq&%!aQG^d(3T}Wgix9bbN8g7Ks_EeWp7YL
zwBRg=i~zESWGgbA#^ME$r@G3E-@L1lT!L+xF?~}X?Ns~dm9OLtTAz7P?^d{uKoS)$
z^8Gc<YEunGG3Z#?V%M$l-1f^nJEpjejfTBj!P72rBVF)S8A0y23PMgqBk#(Z;SGPa
z3hmsh6J>qNS(chqu<gzj?f>8TO~e6EPvrCebt91dTPE{Ahr=W0a%Ua}C205F4Qc-*
zuJqY!7=%>MT~u;13oebNIohJZM%EfJTPeU2Ugt5_RE^(^(hR=0J9@Uc*05z?3@%Yt
zn*9&k#@2kmDKuT*PYAD7gh=znR`Ujxgi^3G$g<&22~6DO0TLv?Ul|7g+BmxH-ryg<
zh!sr^Qkg}-Dg-&{IspgPCs!4)Ep68BG3S?`?&kD?f0)36_sSBGmreg=H95@yk!wsE
zr*u-*jjRU!M04`4({>1hN_6w1J$G;FeuTAm*UY)qk!Zt5h?0tU?Mhl1x<J%`-|hX`
z>TSkLQ1YirVYz^GhT10pUkE%6GFM^YoeFm@hwm7#LR(A)I#38%Q3^6!feru*+yyG?
z-hnPe1oXyyZ-M-LWj1ws{z>yrFCp4iwJge;S+N_%x>+*|L|5hJ87e0IZu|=TLgQa`
zpX(oD^3QhHD9jYZa(VXSf}>gvrTvtndWq(}p$6>s-75m&A7~_~=SuE>dwl;J*n5)V
z?f|}Qt(WGe00ofH-mI$!7PMJQXd_(GP!+*KALvT2=m$W_4eU=(gORB!Ut|%Sw&}os
zyxQtvh!j;=faGJw)oFOBi;-4rEx=Nio)<dDQ&TwJd%IvR(eKTZM`UdeGKg2pk=o-}
z1JAZ6=YKRMc^RG<kymW1>`Oq>mbh%3dUCY|>N<~PmjS+w5g3-GoSaPOKqF8LGLtOO
zE<rnmr6s^BudN`h%b95R0ILWi)$J9EoN$i#t^!b=6MkyGkrE3xmT|RnE*A5A4g9gc
z;;2R;<@=Om8U_+_0`+%iprq0H_C=0EVoL(&5pZ{99xARf*Nd~eS!`)MU5q<2KH#6|
zEk(3x{Q$dy-b&Q5^Onw@l=o8S0Z_ul=mcENxAk&C&rp6tiOw`A5XSLA%T19@!_P+Y
zopWQsiDeTAZ6jp|3m@C26l3)6zsu@&Tjx`f2;1~*F6vnhmcImwd0cyGNC8!0eBEcy
zFQCiX8k0!_8iF|^h`$$S5T1@pH>W?Ic;2m-g8|)&Mg*}t^~Biih5quc(U&a?LPsSY
zJ9X>T#^Ejxee4ON4)Wv!iZeO*s?aFT?ujzBQuqVsbJ=FIIgWWyAnu%c*-6QJ7wq7}
z>gG#r9UX@r-!iOXoGMe9zK3;{$u3;AGr4kYV!3H{!UHD*%_;GzgqRu}(a=MZRx$B~
zfq<m_UoGn!U+2ekCc3B_mRW=T3osL4LSf_X3&-pCRy7faajU-S&S*#0gNEfW%tacz
z^lI(g3r!D~E7a#*@Er7+70}k?db?m;(!>aoMY^55H^Tu>{9Z4l^e|<h!}MLak6-fx
zVCAy|osL@l2Glh|gHrop%>*hvqYM57lYo~Hly6cTC1N_zw&H(}&i!kI&UjHBnck-d
z1SxwraSO1Y%KQ{q_ffCaq*6>-zp<TRrP5b?g;1fB6C<|A@zS<%<xA?F%u%teo8-;>
zT6sew{d`XO>P*Y44M?a|737U|R_}snvj<9cg_-vHQFO}#<DTpUvT0M*EpvT9?f{7#
zNggqTy9fmKt*&z<igdk?qQ@xbD=(+UhLKs;0}p-t>SvP7@Kj~io9J!=kD&?SVm}>P
zeDr_4yRX(X62OwbA+NBp3YID~%<py`b9c#OfXs9{(6wDaHwaMe+2RNCMc0!gzCGxq
z%8Q3XWqQ-igaHVAy|ONc@s(l7zyO<DeYcs2T`YZO8wtX2U;%AyLBSR`+sZ3;QdpUw
zsfxS6df6-&(>^j;yiC_689IDJ^RtSv!smKRZ$O;uGTAV+MPdH;W)^G*Fm(EXQ&?5p
zYi=H10wGxXlnj65UH=%;ijq-CP$ka2Q=&oL=~u;RJXGxS+-dOW>lam<EFdvfReYPZ
z*X8(yuD|)ONLgT?Iy4kDfy-jRE2eT1xI7myUyR#cwJy(e<g+ryY{5$eu#R8DFnIYw
zI?ZUFw=2(Uti}&__HL+F%hXlSdSY3pM|2Pu*1H7E0<&x?zg#R+)xIek2kl5@4m2}i
z1}<_6o0SRPtNE!h?|zkCk`M9jBc-@FLh43M5I8w~84;nWl%mpMg5`?+ty-GH5Ph8w
zyQK#7xyMN%<@UWs@&$Wl?M=z@iBNnd%k)2u0Z!)hznb8v=?%)<-5{iP>CP(LihyAl
z^KoBZt1uTYBCMVm{n28>JaA%11!6O7G}6akqNyY@e^uwxEqTKDFun-f<KNj|QfA`v
zXNMt)XxX}N$&3i`nLF}|swS+^9x~L&9GtsdhohSEdFetjj)%)V;pqy6e5sILov#ni
zU9VBi(Z=UMUVt%i<q)uvvdMEj&;h`-l9I`C9*@dIjp{nV1?Q|f{v2>M^rqCYu1XqA
zUt9y;xqLbU%gdo9Hc6eQthV58#32TwPyQC5)P)5GD#<Pj?;{HNM+rPbOaMLlyTIz?
zGuf_sk$qw<^jj(g88&G-UwDIRAd^{_;{yAZyTe>qk9g#$3bBoh13q!(3BV=vG^q=N
zcql!!5f*=`rCR%B9_iR>%U;^GIyCz%Y|BW~>|TH^3@W9iK(58UuIL=LGn2=JAVFI=
zPFoFe^*V%lYq03vX{?veb!~w6`loB^^I?bx_rEuLRL271VFsy@N+YM&@6}uP2i;zq
zI}d<@I=}Gm%qSfxZREEgGC()p0-qbDCByTf_0x<8*0KwS$(vT&AEyA9J1Vj~ck?Rk
zXU*@dpEkD3Z;nIWb`Ctjb{%7gAeHQydd|edWu5Bog&B7cTBoxL^o(dmebkzKL=@C?
zq5-Y;0V|ocvNp%BY@7GeP#nz^A~V(+2wkiye$2s`+DaMuEdqs!YJhT(eHfc!`A_4-
zI|9^N$-%5Sx6I-h$Jr<m&1Y3w<`Pqzfuqxk^|bkIy4MiA;|`=_Y2U&9PlwM16{X2f
z^(gq}gttB0ol!I#uj!*9>}zt1*>Gc1&)NRj5xqtlxex5&1_`Dsjosf?xn2`&)cVd#
zgvDz5VkkBzd|q-L2rweG-twYA<31@abrDGYMB_F3M-f`ngh1fW_=PfH&1*erFwRNU
z=~8aotwv*azID;1FWe<=U0roi{DHt<PF243xM3_jOZ|biUbUv~I}_fnI_pirGy?$^
z*l@{nWWis0^p+h`graaxZ_g-<BY;A7y=WlwQruOjOPl8=4;Xx;99Cq#HxFkiHv4r!
zo>~6jZs?*^bSu@GVCNr{OM*M|*z>VytYq@fV7nSxp@aR}U4PPX6>BNx7lO~(Rgcli
zsDTAKk_I!+7FN2YvYEz+Q1AKdT2MFWAgI=`s=Qv`tr1zVzgF(w(V1&{xLq1g$ne$~
zYB)E_@Kys4e89}k1~Q}3tHJ^;R?q4cDfW&@;&%!FD*VQ+n|0800h6T*T_dTWo0*}G
z*{}6{woTOc=l>AsiZ{g=OkGK}=yNsWw^uWa%%20(yUi~nHTRwsK;dK|WXm_Iz;G!p
zjXUn8_)qlcBwk1Sk6C&ZNE-Vv!B_dOb1=sH8c2U5bSjW?Ydvd`av#Ic2cT%JVQELN
zw71zSfV$`Y83O5tQP-QmJ67N?AcTylb3!&6iEgaiz&ZQ8x1sK*C%xkreG?9S@{TMs
zk~O=rb@V>Rm`eUtA^8SRr5C>YPbtNjK0gJ4md>lm-3c(r`{(!ZMAmq!;)U81uKVv|
za`P--RUJ&p>HxC1Mo9%8@WmB|S@J(DNT2~)oPYq_p8S9dO?qGV7pQDixNJ6w3kVJ-
z5nE2I5Re-BdhlU3+A$xQeMFeuxA?wYY_P)^D*j^GcY@LG<3DiGW=MzRLG@I}zo}tC
zjKHa?%K#-t2p_2>3&JSb9>uFB6;vtHno07O`unFm3+3!??xC;WeE7wf$`y{J7*DGV
zB<?4&THfBP>^}(3h7bxmZ8A4F3Y!DC-MDn7Pq<r#p0_XCKi@nH3$RY$r$${+@*F&w
zR69%A9~Fx;&~@_egxpyqRG3>5rtk3SaMk$R(Et`N>+%Ac1m9^TaO+D~p}mTiG<uTY
zbZ6}rVI`C4n?CQq{B9FnU0EjBX3T8hp$yXbWEaUXH>0g9ITf!@ru(|CKYR1Br4U;Q
zM7Z#-jftxTp(zeH_hJF$R9f{R#Q7^3;eJ`X&^?D(lg1!@2*!Gh`e2jfFVUT%$wGj3
z6u=k}wXE@BnJ)>A@rg<*TdI3Wm*lV|FS9z@s%V)O5UC4OvuD?+R_({hA;+NSYJ9)v
zz$MZLIpo;wN`HlUpXD)YPFGHCtv;%Q)=BA@k_cB%;UPYBl{Lvqy)xWqq(jbChe%=<
z0hcdOC;#11;&617_p=)*HK<J?SDdGhKcI_@6Wecj5{(k3&az^q$UJ>s+PQRUaj4u~
zyqn&e(c9y$@xgC9(KP$c;oRR{f1(rJvB8r*<uGmU{q~zhS_As(DrABT4R|@@^nitJ
z#8>XSrN+vb&V=A%KUu%tv5Xo*wxO}LW2Y(_C0WQ%4c+`ORJl|A(PuB4hR0+V6aWZ;
zkxd3uRyCe)XIHsGNU>SfT<bb#Q!QE6iI`@k&8%(QKVL&WP5ZNr&@(Nm<DlVbcEA(D
z&jNhrw3E^$m=*p0d((F2Iv~8Rnrfwb)uKUu91mtsW_NSSQFecw9Pg<S<+xZLw@IdZ
zY6fRiWM&=ufkKik4_SM#jNmY%9q@R+$AL{rabPyBOmCC3h8~rk0qSFv?ap3%AjBq2
z#uBYvQ`B_<<H&`?-)e?H^=H*Kfg4X&vl#)*hZjCpPS-OEJl`|$xBwGsT5IWrT9eEJ
zl|^TIX)b<5jP&bu3xYVMX9=o<&{;^<Pm|CHp+XvHCzZR&fmwsd-=BDsAE3qR_9JZ@
z>ZSsw8I;ujAd*2kML5MB)e&QRKc-u85CZP|BQWz%@ZH7@uYUbAt$3n~sT#xRD&wXS
z+k_XK>rrKWyHoowpDvjOZ|V?EA>1SFW$Kg==G<9;26x-wM8E%CB!{H^F|OxyhEbQ2
zspuU(HtLvjf9St8?ThXjX~3*qq+R(^HwzV&K*t}Tgz<s`t-sKnUwef%Vq9rW5Qn_x
zYQtcwOh<Vw^yJmeon!>|x@Ad=E#`hMZT8k-`!1W(t;;Qb>9{8t{odGH-L+-F#&V3N
z_BI;THma=r+!ET8r!~hGB2DWz(J=iZ$Q!f#Mx7#wmPH7xHFXt8Isl%T1lC0dOXTS(
zDHXxW7Q1vm+vBsGbo$t4^C;$w8V`mML228#cvDT?ma{9B$(1qd$(6_H<4?1*PoqP8
zofbRAy{yGTS6&t^5n73jG)P7h%|8>51LLybZK*%uELX<i=WhM0lg4zP&h3u&R^Ir0
zPcaLJ@|t=NxAoCye_3>&S|vj69_l0Ty6=jAl9ZFf#N>5;1nHq)@2Uu&T+bTuYyrCZ
zH^a(!PN4$li6?mZhA;p4x^bS93fab9>IsdgzH*sAFgp#1_5ORg7G6`Mr=zRWQ>I7i
zbEg7EFWzom1PR=fyU(F3*24RFREp%6RiDk~fw;yfA=<@2G3?KXJoT>~UDEwCM>EB|
zH9ELvN!lA3tHfF7nP%X`9rUyFdds+#7e|XZ%M-!jWPBaiM-ttMbv+`gfdSLy(mHFI
z7HreQ*M76euSdwRR8<j+8|Ouz)~!VLsgwYlw1f$jj1Z|&uPYXzaQBdSJKnLy-#qUs
zr6L2wxFxRGuqIFymi6XovNPE^^5ZS1kTr=ltuC#?gSWbRw~r||yE9qc*D@t=VQ;n2
zxhA9y>bpNSG%1~=!+wp?c~l$Zn08j7p|z>RgUhz!haqOvkZY~WBs>~!M3Fq4VM@RQ
z2J(D;{kT*^i%7@0S_L?Y<aED5JjXFCraN+428m9DUrV*;ALV-F3a^h9zxVg=TSH5a
zXN_yV&xZ?Q0iwZ)4kn-qBdgah)7Q6KOWJ0wqVDp~%&D3hbAv2h4`V?mAn1&9GueVV
zVQxt;$fhx5!np|5HME-9vQK29mS8|+zRH;u#C0kj>ie%IZHt)&3xf454sl*gy{c_>
z>qf!ZF-^@f=_Ui|ww8%_YnDN7#T4*pS|9~$XLuo#a=IoOdhoiF0LmfoCHx5sb`=5d
znx2NSDU=#V%TwNp3_)z+!W!-QIkeqn+u1{P6Y-s9lo29Qs%L`6{X(Ewa6dpkX&HE%
z*c>+<P?yNxev8#Hu3NTYlbQ2bY-HDYu|5NE<1`JMdFsN;)h8DgDg3;5F3*k%=yrEs
ztSYGAaqyncr?-Z+8RAcVC-U4Rcbw*VjHIYZQw<P<QFuSqnZxiY8TC#ddZ@yTluajY
zVfxuJggl!a=?k3;X)i7`h67;Mn>K)wY}o2%=q5Pt{+Olr27>?L!$$cDT2E2MM_b2*
zKipL<R9Clz#dYSX<i)Amik!l(sWbXQ5r-(GYw(Wou16BQP_^q`pyf&#*ms$?1r*Q@
zajU~!DU0*vfd*VH9CU$KP787?=IikrT2wlZV=w2G_9;)g6J=LoA`xJ5R0C~zN*^gQ
zUAu}Rz&#w#TgeM$NHe|1gJ;QR=u$my$tRn^S$YDyo^U0v>)=+B(;ZxpB#pciQ(&w%
zmm_7#dPjJUxVnfrR)8?=?I^xPqO$Sri_AVzJA@hKBlrq^YVRlvqaLQ0sLQ{ibbsSR
z19^k?N}gt^n_G1MXa-Q7I>~gUnN?|3;id($YVHi5NT-;%1yG5#DEo{BEaDZY_<%R4
z_z<igN^ZhCXKsipQM_u*@-HE%OuQ?Ih+dUT;o;GWUHbIecMH}_-=GWZd~l*%FMV)!
z2B_TZ%4{pcNI6G<;p1%b1PEzt9K^@>YCD(51G;liz}Q-|gaKNk@-w7wy`n;jNj8t*
zo5TGG$W53stz4b@Vz|&&8m#;TVY@3C{Zh+}>c<tosCnCB7}#Z)CR=}|&&_ksJz!;N
zZ5@z>oXL}<rnT1{^UV_q?{wUfRjU^F2;!`)e`I{}RyZn8wqiJn=iJAd`k$dK!3oha
z5%Zg=zZi09fBnG6pFCR8;_FoZ<}8%s*32#!{}#L*;6*@PaW+q8(jax<rVVJ2=b@fW
zCx75Nmj(PCT`sPcnicP*soH3La3UP93n<h+qoqT^!OWdBuvA@^onKO)_(W5jOI{1H
zbwBdQfT(Qjf76(N|6nvNEwSt<0%iq!FeLVq01Xca*8>C~gB8n$XHH^hZYQ#=cx8gO
z9T7V$t7FqQXLPEn{={@et={1WK^Cik;+`X$wSDA{FhPf<(awzDQ}kS!@BLj0w5qga
zYl@C*s$`0JCtYe(l`JY_k`7OW^MgT|JQ~*ejbo(UiA6JdBD6GGy2Wq}PeL_J9B}_y
zzwwZlev1qS*pUsV^fI&UPt&|!6RD;;1cSoUP|9n;j(&?4i-Bk@=DvHi$oFgI3!jL5
z^y(SdO4?%T`uoX-%7DBQ+P73#{r6&@=f^@z|1$kyY)G#@eZD00cHIuW4^c{$HSt=9
z#I}w5$rJ^PDPs$kdq4b}$1e)ed^faIKsrcqs(eK|Q(b2pIOZJ0W%5_Gylr^nzSS+u
z8Vx<MHW;#wjk_s1(eXLs@<%ic?E>1yj5;3T>|k3E9A|x9AJ|O6%R8~1woP1CScoB`
zbu<G}1PGo!rDvnNdJyFrEg0AB#v7C{IZ9InVp~GOXxj#@suL)|9HTv=4SfDqQeJfz
zNHyP@@TNJbxNwwS<*P}SN-gI1JrCKYo)(g~fHW_pIda66^;UR!wO7Xv{sl(orb?)V
zT!22m@(6}v>Ghz=Z0JLBfc`pYx7B}H+Yc3D3P7Qj)pHJa)2=)`^K@ChyPhSCNe<$e
zD!jr9(XC>Eo~P|fh**}n1W!;BuQ5m?P*yWY-PD%jTU{t|Pw{4_(HbLj2Nj5+kdXMR
z<nhU<9p41koFjMQgMpH)`g`0XnB16Um@byvNPB(RTkV(OHnfPDj(w+%BN4lBAQqOx
z9w7D?b=9^z4+rw8rjXsV7I`8=`<Z0|l7j?MW>}w-<Sf@1%PH{cV9uzJA|efKlDW=K
z0n|VMvxu-*zPs6ynbQiw1D=?KZ3e`;Dzdk6YHnDe-DqDX2RO*Kw(&kOR<?XhAU8~v
zozDeL&Z(=>FSDt~sxWZ1yfR!%&?WV$c))JgbCt4scw%1p6p{bnnY^kKMZ0W(B&HiQ
zL`2%z$*dzPr@9|yB%oA|QuQB3zJ@0;kK^CI$*r~zvT>hqrAf8N!iKn~j=|)YRA%vA
zpHaUWyC<DeJW2FKl+RBZJ>)hnakp^pJ+i#al;Jh`u+3GHg`r=xH0gR>=zLGo7yzzC
zR-7BMhl+G8Po|vk8&jLC>SrH)a9#drWQ}MO@QKr+I;zj2z==P^MMPY_LY)0#In-xL
zCZoH#a+>k7KPaek44z0?v^wJPsUic}P+3p2EcI1CHCh&1&=D3(yC?|SD)Xvkm}lFW
zlm=OfM%R9osdfoT6i4^!8D<U~Pk_EjsQ98h@VqoXVKWAvMXFT=j|ao2h(l46QL2Rk
zb_TDkdZ7x+s)p3bMuqBul*%ANs*KE<AvTaWU8Vc%johd2a{m+m7|6?yo*hPSHl?O$
zWA$<A&Ax*ACXuCdX7X%8Zpm7;eP)V5+F@1Inzp~s*qk>n>s7!|S9Z9%ACQt94!j(w
z{0*zA&(ft-R$!8<?N&Vaxi!jSxClr{PdNM30T|B0;i;#lrn;p{M`t#9Etn3RH^4m5
za;3|&E#K$b?t0*@EGjnGAcgpyH&j7oYQTf3D(rUJOk2U;OVe6LcAQrTnjZ;Z(<C&>
zA!Vd?V7lNmE1eXWPT;!p0&J=sT2RAv#L=8JV4?amnO!oT3M!r7-Hr`ZZ6zR8Szpmy
zL22-Xgg6Hq1NXIN?%piQ@mQIlI?X+FSR;NDKCN3o{(|P0^U-EhEoB6ouFra)hSM>6
zbRrGQyoml3u7tAz@!fprPM(O|wvIK&C@lX3q>sEp)bE`uq}oU_0*X?Qj(9@V4#O9o
z8vXQT8jFs~OW2}KihfvISeA(-wD-QMSsrdF3$HuY2GqU!Vd^+ZAN|?^XB+!D1q`V{
z0b{;wAU-Htv^?hysL1ZnyKadqsE}rOZj=U0_%%#N;(sa$r@_cQV&YU*wtwJ^i{ibg
z$-6tqNl5Ul7(O!`fJ0#x^kIF$i3?PAye)2)U)Ge82vVCj&^vE5oYAl#@_1AgW?<^+
zjA!PB&Pj1fF5xsmxA8g4Qxu4*FGA?~u>=8K`F~iuf%2)m)K@_4$}o)*S(vsH-VYx|
zPz(yt2CLJJ*F8xSglM?`8XQ_nAYwD<Arv6id6cWI&vkqn15Qn)+18`5<|WU$AA^AX
z_j7H^XBx|TT6sx&+{2xq%1O5Ixy<n!^&VOJBzVpfsq4Qg0#$=%LAk``RK`yw5Tg6b
z<p1_Jwe9LNu7AJCi=j?1k&h;^MpBsX4yUW|E#N7Ft<KSQDJNvg`9b%S?P?P#e-Sw)
z`W6c$#}^Vv0N{ZzGS@)rrSo;4kWZ$xnWuc#@mrCo47B;BbnX2gt90#$8Ll{z(^R;-
zhn8Ee3Z3DM5@qeCY|&G=#v0D@`yQ>1-(D4)8hsjT{|vd^e7NCR&b(2f!=<Tf47rI}
z5hLaYHK{~ks4*taIJ&nMZt%CQoPfD>)#Y(gg5T%10qipnG1rq0j^|Qqrb>Bqny5r~
zat5A2*>pF@{}3~i>wIfvbfo&fs7U@&#h$VzjQko+xZeC2o$MV=!5kyoJ0tAve*K_A
zTaV%m+SMrOdUyfk1^B<&Dzc0FD2(q>e<}0jIayzS%9WqvQ4+VW$TW+b&mBH+cd|4{
zTy2Gq0Vtq~q?sq+{`Rss_dGPxc%>oCm*gL94F6!F?e$4m@9o)1M{l53+9O-tw69PG
znl~_Jy(?(jV_#tOJxtxN&Aqbn5K0aLcptkf?Scnxty&u}mlM%9m8*S#)%TA=Jbr=j
zcfKhm7PYG~5@Z&Z#5QJwzWO;1YFpG??#{SB0fM}PWGSfVPt7l;Z=u9R*##fb!aDLv
z+RqF8zXhtWflB5FH_u<Afs~Ld?=rIr<%Xa;E3qf4Ci8{<w5#gMx3~lJTGw5r0@6D1
zu+uI!ke&`>aKG}TsuTmt7|a4@Wb<k8bc5V#?|2}s5#KnJ&K)MW0~J9Q0)AaB+GR_L
z0ER=t^Zpf5F8^}A!);n)QD#e)F0I_B5JRf9ZNU>Eq8*sOiBD$4;zi>Lx0T+HKL@o+
zM5-ztJL&A(;|GQOBI_m?4UJ?_Or<a$P_R~F3KjNP7MoVbHGHhLlic_RRPd;vw(Cn^
z_r)+<q&Qj*=OaStiUp0yPfrSdbop)P064H-?`)Yz{-&Kr_OsRU=iu>Z<U_{CooMuL
z#38<kzsv<I6>qury{aJR!&rb_YjL|CJ^ZD!@rK(lkWuT&xC`vMC~}$2L0-JQ*}m)}
zdA>%Qz$nzI2%BGg3cbb~wJegH33qBNAHo(ZXK~ZD4o0}j#&cY<GxtawJ;5?G4^UHg
zq4CCY;t)lLq-c9PR5}89kj@1NSD^f2HQt8u{de$JCh;gRh!W!*e|YXIaKT_?b6aRj
z*??}=lzF$CpN%`7vKxQ6R4L7=DSubvYDRbUyBgP4(n*Zje%PP~<$UUnyrcN+_ov(F
z1I)5<xRW}r*(-s+=zTr2M|rgWB!Dn>lR8>V+JU}TXm}x;V659u8KEyOleK#V{By*3
zU%d0OK6M4mAd8olA-1PuKcp~XxT6$XwLH-;IL6;zvV@89YR}F%<N>9Xu^eIZ&LB$2
z1%X(yJhtj2WZ56Wd&Yj+F2&f=BmVQ!WU7)rF*Yma-uBz6PU>~{sq)z98PH(Lj$h;!
zUVF2)L#jo;`~VtVV5V?x36c71A5@5;skj8(CmMDqM>Y5Qs=+jcXO7i}5q*Kwn9`l{
z**QC9-9T6IZP<#>{wM(okMY%S5WNX&D8sSgH9wE05Q5dk)BIIc>+D$L!no8w+wcM$
z-rH!e40Lc*e-t>%A^(({9?(z8r6B7x=(v&IyJhcjm(^XixRgI)%|pE6U)GrKh}bU2
z`Oq=i#K_1k7i~~_@U8d4oXW#L?``9v=#-na6@IW(7tp?WxySm$Su7Ncn`8s2=9M#L
zPLJ3s&2-<^KG*c@lySAR8CyYM^K~ow@cZZWhA7k02=jllv*}|}obW$PHDP@^^z<rh
zTFUfWM{bs%T`~2#=;Qwbi$HY0(~pTye>}Z@JbgKXnhdijjGR8cPlA5n^eQ;r)HoR-
zm<BO@{G9$ia4t1nuXHKS0q8RH+0r{6pFvm}A#}RL@!2SGsAJ_Rj{%HNNtd5a?;@>W
zX{51+`}t@!aiR;3|9t%U#FOZpIqg6CxBlDv$J2jLMG*-MwU?l9IUIjKZCQ-dGw}ua
z^y|me-T9hq={VVvYmM00Qfh{1CykZ?O`L40-Z)3nTAk@44Hel*ibl&XXDfy?qGkZD
zpAlc<(@c4}@=Fl6xkKWXnY3b7nSLh88#&5o0}N32qsh4(Wbp9K>s!r)>?agryu9NK
zj9M6y45v)GNDOrG9%6qrtiohyGB^^%IKhf;ag^z~;E1EgD?i_?g*1*Cv9O}Rp&HTx
zb4K)|dA1(HRA0in&7HaRGP0Cm3QS#>E}oT2N~iBgKcG22YwmRgmquxSH=4tn4d0_z
zgBiVfUZaN*;&*w<Go@L*d6U`wT>hRAY5t)rf~ozmJXQ7r(o%okLE^33@HdzRII|vA
zpqVmu3DY&<E=%TdBUE+PBhpD^DBSMS;Xu0e__Z3mC{SuH_Y7X`jE@8n?U<mMx*0D;
zksQR@(U@TCAw7-`AvENTbHB6_By+-UlvaXaB=N#pGNIwE-N4=#m^L-&*_KbAy=MHh
zlP2J?A#kH6!~uUO1CQM9q3m<N58j|F7s#JWWPBXPXo=5QwIhtX8FB6y|JmJ<6Gh-&
z-OQJAa!QKb=tlrb9ukyIx*0D4l$atoxmaJPOrVl=Qz-L(4OvE1tB23Fj`6$@18`%O
zm}5IV%5SGSz#UU=0f8>hl+gm1!G&5r$rXY?v@(HDVqAZ_xpM`MdFS*lR%y4MHbIc!
z$B23?O#dFU0leRCBb1C_yReZO+laP+Bm&dhi0Aks5BE!A4Dt4*x5ktT{Fae0k|aoT
zKeZy@F+<Lr`~P*1Y#q5~p%defyayHd6gq&)GRw`LakQOr2RhKrxIzcI(Z8JzY?C(A
zf!<mh=s<sss^gG~?*Wpb3^+xq9}FNX>X*qa0?hOEC9@ch-Y=EYyR~jl2(~kZVh3xA
zx(bB{B8))hV2=Be6Hlpb^Y6j41#i1#*D(%uD7=uXP~mn#(<9L*Cm-=@k_b@AzhE1%
z?~x&yRZ*4AvEd^D9*bR4WJvYzLjQ3~h$aI|w<&*iSRGP7g`6na=$Eere#>54Zc!o8
z2Ce21DEK7T*C8T}V_53eDG<z9(O~Jp3SMRuMO%p+Y+N;Im=b2qgS;=y7C@FMOTfV@
zb(vBJy)ShuL1|_vhPL{@oVD*b8b3ctj_S}xwj5$_Z6$)!gl=545`7$3E1ksdJ#-B(
z1f_qBW8a~~QZKdsc*GfzTMq)nx-)XLzq?V^DJ9MqdAM|vsihSX1Ikv&V&-@TgAgsV
z8fM0%BK**1LyZtovd4v1d4>fsMWt(D`w}-@tiUr6QKc|sXF>$`<mNcOB(H=`o8*yb
zt@PHU%R&x;i@R5X3>kxJ1=CBZUNX(;GUb0*S1A-z1j=p}AIOt_c$B%b7YuFV9uL12
zD=3w=^95OvoUIoIz1FMLX>0R@3mJW``SrLV1*P7_N~+sqWs|%VHBEMIaoc(X%Oh8J
zY91LhDW@ibH3uj!L2dv1L<l42-;kcZoCzwHp)(prhfBT+)-8$6>1j!Q0*;<R*+YNJ
z$i~U*D!$74+%hTlO_xnziovCiP3vl2!>lxdaVl|J;Xf<aNve(%|6RGxrw47fNBM1*
zyr=ioNpdYgdHUXuALgH2n7lujXe6;6>o!{H_^c#$Q6~eqE%p3Jo!EMeN%;<5s4jg>
zOC#oCAL4g;sK;&f<f)CON~&~#=aYX!TK`mA)?vJ=wMvyTmZjyHGDFZ+9-qbG3N}<<
z^^_-}>Qb~WCun7qX%lyn$i$ViMCCqIYE_=pvtm^hq%J%SR7Zj7L#}@+SM~Iy6R}_L
z3{D;5nS96-r|2OGh3J)Ujpt8YEqRVvCq1XiGvn2-#wtb8L&{zc{<88EG9P~odgU8c
zL_Fdw$C`YWNQ>|{!BP!M;bPPn|Ep1EC(frOm}W+nS2K|kg0`~mb=?>~E*VfEu{5=?
z`uVyjXqmO9e`==~yc*MdMyc`hTEs~d@*JZ<G<4x**$K8`TMcS&;Zuc)<rq|HdPn51
z^0c`Ob_lc2;%t&rb}|pN6oh{@RDYR0IZ_Lm$dMMPNd?rYE;X*-=Ed7;pFlcC8^d%;
zeHM2T!ZFKL&z%sNabkDwBxwISY+ao!g^OO-`mTM*{zW#5|7ZwA$}s2Le0Q&{9va4a
z)}0>eQ<%zdRUmVzc{jHlqNudl_1%XH?d0NZ&^NsQcu*L_v=H4Q;{tzpQ-?d#+Rh~J
z4<BUa)89_gWb5G6jS@=AS==oo+#>^J?AG&*3-G<;uV3K72GHZ-yel5^v244@t*6EQ
zuYSDW7QUC4x+L}VC48!fGWpJiLmB_hYX>?LEM+$I)YmwI116oge&j`@J2JGFmdKaV
zXI^|1E*A;_kefnN9(iIzaLP~2NJPF<KDBN3BwHzRZk26Qs7`;<)%q~Y-NT|Qm3npV
z!mEoFb<4I3N5~~o)aP@0n;9ThFo62E&<iS+8celK-!&8(s!Fu!7WqW#M%xwSEvG^K
zJCTzA0k+;i1DC-_HxvOkm$AY*6PF0v0V$VYcsZGWfrr1p{`Dja>YrY`+QZvA5QH7$
zP1{Q#Jdp4Of@n1T+vImm&{36dgar~d5b}Jmc}&Ic%^EmAMA9&Vjmv+@(_}_+i>cZW
zQXve;J-*n;7#CR;9?(R%KBE-3&)b#lRbp&nga|hBBK~|H-kx7iznv;av=L}&Q)7j5
zZu<6rc8VWBp|CDYT8p4y`u!i%_tTdVYGL;HoAf5}Us{XU@n2hk=qC$CNCf78&$~6o
z(ere99%3OLXuJB???0wt<m=~3zb1_^L5~=zU*gm0R9!)=3FG{z#X#s_5lPY#m_=kR
zVon3+HPKa#-mpZLpR~gJP&ImYKUOH#_dF7RZ7;3C2d#=!B2Dm8aFo0*QKcKZfC63R
zncx^+9xZCWZj;(-;Q&Vp4$69qFGix!e!-Vw9wYbAdlAr9X=0%x6g&z?BA+1t%ND9%
z%EQz@?$WBf5-82M-(~&ZcO04=2DlNbd?9SX4{`>r56C%F8MPVM?tGcyG7M#&bk^H{
ztn&;xL8|*?O?kLW&{S_938I}nD^wWC!P$<h{AEE&onu=$6(0IQM_;P^E_!u_pZreo
z^E8?*R1>Rp^d<CogC_f!#pu_aNW_}w$W6)`K@)#`k&j{RD?E!>=#k<rYEuMC0@H`c
z?ZfAWqFEHin-OVwt`0>vPWa<Y(qt}wQLp>$s`rWfIcEYBfAk?Di~{2P2(-pc;=b{O
zn?dU>{n{XP#(ee55fNjcNEkX17lTT2Sn_{NSJ3$6BL9@>N{?KVt{B!yH+vCq?@v)s
zk~Vn!gHjYE-aD(VO;LdPo}J>g>4@@HC_QezlZ~hV!s-ROH_1oHTUyC+Agk(s_4yUW
zB(3cg6zM>jD5>1AB5D(B5R_P?OXP*;B82nR_!$#~cHsbi(co#hl*x%@sNG0%1wpPU
zl>vh8-GU}Jd;1)H*Q%H$G)oQz7dYedZ2naZpBi^zl1wv)-cJj}nolm|6LY3yBs#^}
zBXofFq`PW_WaS6)U~rUS1vlG&!LAQeMhQbseOM*&!YU8}7JG3<oLsCNvM09>ktAxs
zT#JXJaEy9f)#s+<jU~336Xa|ykG8QiSyFOc@v9x{gAC<TeUPDseiq8ztk_5|sH$J^
zv-dii@tm?>E8T(%O&~0p!RG#;DSyTVgrzIr#K9VpgjzrVw#J1NQm$5iNBx{*qTR)n
z(pRCZ`v9&$)Ph^+THzC8bg6zEsHqyWs~?;77oZW=)@s_|1Hk>9J>*pBAy=rXXlSlP
zp;oR-U&>!19-&sGP%F}#q^P93KoNVOVBZfPOSXaXDssgG<W8rU2fOvPBi&r!*L&-+
zu5Y{nT+n=ypp5$1!{;1-Z_HK<3h#ilnEV}b&iGXP++sE$C46p}ouX#Zx|>yFmOMb@
zb-&Xsq)16??5host8J8g3rY>Gf&DILY+N-borWf{w{dnVx=MPVR-B9P$Jq)p%vNvm
zx~_{e=84sN*7Yqf&T}y_<H25E&tAWwr;*sUZ=-ky3sxNq0{~Qi9pIADi?p$0vaU>Q
zTv;L5)_OMAwuQ9nW=#|FV_746*T%}LoPaSX#fC;PxgEB4EwskX9bjc_Qff8semqEM
zV^eC@tv}F`3o0A`9q8DpHn(q_=zt<<Gf?)ymVk+ExiG+-n|~}T*A%qHNTQLUz3lK}
z?TX>3!|P@zH$(w{Y`3vU?$=2>uH10wrL3b5^u#xPKeXlizILb}ykG2Aii=Ijk$GE^
z^RVimee2k8mu9j0vADKqFCpyEOq1@9rGwFB>#*SE73m-a#pLt{2yN(2xQ^*y#<3>w
zcK#Z7PYqoy@B8$=CfSdJf63gqvOQYxYQ5$;w|W_?{=}(&;yml<sLjR)vMxB%;+zi;
zTxliadByYbh~)&P1i{`1?b)MoG4V79?XjPI)E+_Nxbhl%tX;boE)Fgp0vJj}haIEN
z7^gW^wrW+s;uqbf%4uH5@%()ynwGh(9p<^JmcG32^MgUwKhu4=d`x7#6br#!Wt1TZ
zOI7;-AqFddMCkI$-c;ICAWi^gw$>Ejr6}tb>sn>?OEYR2(d}ttAl*nVXc5)EzD;)4
zOf;L5e44-3T<=`W%=ejPNj<s{=vmB9O{(&AyTFc~s&F;Zl;|iYAGJny<yZroaKB%%
z9!dj<tCp&ORSeoQ(>YA?;fL?$)jBiuiNctfH<hb@u5U2I#_RwU=-Hz^zb>lh*kdIq
zu`&MJL{xsB5~(B=6Fg4;ozk#S8X@-htWqwXUj##cqMS~j{{7?oU%vgS1TFhdoGK?E
z*{R^CxBQB+*Y84@ewKGD04yz6B2bnG6A8~ipb5ON4t$$Vv1JH`Jk&h;GmcV}^b(u7
zsi*IMDR-PVXoeVv1d&I1zn|?OV>NG)XsT$$w+N-IjjyzxSH;)(wC_BXR{e82eV<NW
z3Rlr;m(-hf@u^!$f7ZB4dn;#D%UK>WBtObQArnCpEJ<~Bu%ot&X(HOj=AqIe1n0t3
zx2gp5NJBIa5h#YhL;O3AZ!lpXlt$(-C@^+^cZ806znxU!Z07F#?{9Bq*$-Oux%tgt
z?vx2a5u`iPSV8I@VWl%1lI&`OHc#i-+&HKask#K-iCsf$T>}u>^2^kqlt{aqY5Q`N
z9V&a5%j_7_k(muT#xjeQ4O@NcLP6Y7tCjxAol@R4frD#J8m_lN*bi~4X8V=;VlJeA
zk$g{d>^&MEy!2ocd5GW^P*)EfXLLD9PM_ytmp`Se3YqrYQOTu0JQh9-Nptk}1p+&N
zer+NrGt)1a*Q7l+>0%Xmvv#n&n{1=rRdH><KOYI}F)ggc#CTq~dR&Zn@D70%l8_sa
z9P;%QJ8X1*#SUkq%69m#sh?Fs<y{khEB71-M^=}@=G>Q_&mNa~)_mfV#tcPt=;Hh0
zE>0P-J+HZ6u4L2KTh+i~AnBZL<?@6Tj-uB<!w1c-at5_fL9h9gzFER$4BsC@*JhX9
z9jfMxno5Pfd%xf8G9rK<aeCj8J)ixKXXshyZ*+`4Z*s-4_Gagk$X*a%{L#CAE?OuT
z1lYwz3rhtU?Yw`{BA+0m%xB5Tg>@re%jJvpvt_<~p>+`GT-S=*y!3kLtXWwQ9^44|
z54zS7i*t}++<osBgo)Q$7M%Fg-0Lt=QVOr2$QTra|G44`1DLB{(o;u!D^JCYUo_X7
z)0$Kn4ktF(U&k3X(sK7I&WP53W^a1k{jB?oEfz>8EQXJ|yn1owvh{A7Zox>!>pK-6
z91n4i8*DtbMBBr=EaK*h%lm23nnEUyv@HAM*Lc0r#VL=%!F;LF*WPds`8}6l6!a!`
zWn)YG(DZ46ub-Yv-;xjgJoMrDCQYCi%8Ay2f{(ksW4S5G8$FJ@Z(W0b*4{8D&qFUC
znLmupd@W&-#4#sblRVmXv{okRf6&OeHnE$v2_uIacQBKfgPgDI?w~vN9$CqZTP?g4
z=XkY+uWP+^yYqT0f^j1bmt1X4x~__n??}}yb{aJb&<h?m2Sk_dE7@oyg9Hvb|6^RJ
z`R=9c6D%QPMr8hv_o<aMqC(Jr<UTc3lE&WjI(5A-G1m(9@`jtQWi*|q?Q>2o9PzQk
z@tyw(Y`FDI$^QQU_&`n-m%&Ik6azLmF_%%#FC>@x*gy-Har`wXf8AT(ZydJ~e)nIo
zKwi>^a5+Co7KQ;QuG0Vx8rUh&*3g43Ut^_?&XCT5^WXOyuC%)3F3m}I_N9R#X>M<p
z9L@}9z8TV#z37k!%HCukmQMM=(y7XWWb2d+#@s0#inLDIP|Nhp@w-kB3x;9pWF1J1
z@5PW9L@`FA0<IVde-yqpx53FFfaluh#&~Y=dLcY5F&s!pQ(NhSwt@3Bf6D`tPp=U~
z&eW%>0~<|G!Rr>QD;d0{awoytNIMzpQ|qJ*!agzRAQI{TX@fXTBlsXir>A9*=J{Tb
zb<jNV5)UeQlB|FPZ>d6{pdL{8TDb)!ph~Eq;#;F>{sz+ee`;9=_1Y(83{+a52@>n1
z?t(XLQg<N_yG&2ZU@FtoI$*5nX&bCeyqpKm-JAsma+-MA2N&~@hCwyE*O>4%#2!H5
z9fg=p;}zpjn-#obKFKPBUnxlNH0B7h4e*$01sJsDE(+J75KXT^4<Z`t0$Z`k;!suz
z{|b_!sEHTqe=1G8tPrV^hpeRxm0?s!ok2m@jiJKI-4zMrLE+Oo@KPz^g+=kyx}O!*
zp?dd{B0)~VVwhUITT?M?o#ZwiYLm_j1GTKv2!_e0l`A9??!qb<baD-_H80cCHniHj
zj}UL<iP>VQjFGK$KAucwVIqN%TkaByO>@Wln!BdDf4)?^s!GsS`<z9uHb(Ho@~@T<
zI}C2xC2;V^M(Q-aC7w1}tebQ)pko{|>zcq8C5p`+G)92ePUGM$f?@96zdwv$EZ;0I
zp1pl>x)=oU<moVe^J2AHT$~R)ef-P&)tAp!FIJ0zrU#Fg=d1ho$Hx*P*9byS$H(wA
z3S;&Uf3a+W5%J-{gYlb-<>|A<>iPK1qsQa7i}$OO2M;jy`0Y<`7ntn31!Op1EzVb$
zT?NyCPZyWVcNeFN%d6ND&%R!~I(zZ?^8NFSkB(hi8lPZ<3%na_rO!A%JU?GvjpdUF
z_nuE7F~pxj;x(l+Wi#b7Rc5LNbuvrcrx?c1f0q}p78hNdd@}xZ{A&CHAE1d7$bY(e
zj(D1^V58-l2!9bjQ^RI*(uDARa&LV0?&T^5d~$aFpYh>5{ORg!c|JZH|Niu=`1$np
zYW4Q=&*S*r+3NMXm+5r*!}$HXAKrfUYVqUv^^4W(zr9_31r=WW8)i)3zZqY?S-u>9
zf271Y(Rg`rI>MwEXD{Ea7MJ+?lP9oS{p?9N)}NrAHFMl@VJeielzCw)C7IMCE{wBd
zVN7k?!hR;z_EpmKp|4UirySWs@fW_WttLY#5~JjzC*#8b!#^9pTz<P89}S-h8`oNz
z<&1p+BYTAuch@d+!TdaH7r8xImSbv{e>Jo#^$^-MOB^X&DprY58oq<XPg;_1=#~(F
zk>(MBCDInc7wZ<~-ckmUoP;Awq~^%u#Mqa}2`!b<?r7gzig0c00xbY3a8}sbk<Aj4
z<E=z^^n3z?5d;76&woYSN)>TGFvIz~H*ZdUjSW);nge?QwA+CdXehbzF09O!e*z)i
zuzv8AGd_+q8gStb`3E>>BmV@%yG#BF^t+aS+I2Q*d8TNl>P%hp1@yh9Xr_3kWTtec
zY^Hps%1mv`Vz+cxAZ?gF03#*EfPx(i$eUXGNc$y%ueE@*EvZEclGH*7ZwS&KJ}3iT
zCky;{oy#q!Y<9$;Oe6@(B>4bAe;J`2W<`}^H323GVgexUV2;O38_H#a+JTN*)Pn{U
zs~jLAswUuWjFx(}hu|Er+bE~9B_txkC}9r>A$(KCW~~R4L8~DQmstjFh-OkYvlD7T
z$cL{H1ZsA&!|6c_){Vr#h~^rlWQPB3N`O>mC_PM3`Y|YgyS@Be6)~P~fBam{69zse
zW^I94v9JU9IW@=X261HD=r5#(B})l~(_OL0?z$DgW{dQ7rCSLIRQdS2&3ZW9vch3e
zc3*23u_CF~53gM&i6Z0%He=OU+8;~1sQpt7ym+NungaKS*DfNUMpN)gxfl|;-+y1q
z4KNyA$ZIfKXd4J#=a@mTe?|qsyRgESQ#7V_V2v^(aNQ2993_bPF04(H17UtdWPNKe
z&9(`qS?&g=x!i#1=&4?TTACp*&5)O7$V+vmW?0QLtY-RywRK3WL&0_sD5wdL3OKAJ
z)cu2y_nmLl<N8LVDbk9TVzpL7)RC6+egLFpM7SI6YAXpC@0i-vf9QUovOJJ>C3;Ih
zmerwMz^#Hv(W0u^62?))uy8BRfpxiQ^Zg)5kzPK{5Tvx!?5NspC3;JF&z2;kROyZW
z>+BCucii%<L`RY;9l~o6^V!=#UQ$Lv(G0JNJyHM)wjtL;yG_81IM<oezgl8Bs(P(g
zzt(<{{#kcvg#DcDe}6W@wy`R9uhLGeQaEMXiIpAJG~I=j1VK^S2L=JR&ihg}^S;>K
z^1igSyw7v{i|5SFb7q%0v&-Btllcvi{j$9Z0?SP|-Vw1)kJybb6`o&e-SY^iQOd+W
zOS4ADJY>iv5<#xE-@+opx!o*GDkeN?hEvX;#cw}E8CVfce`3GFe6h8V6%xgT1OSMd
zH@jrb;-DMbE0cI`=HmNzsJNo&q#GfkHyg+Jzaj&YMRf+sXuwdMv6-wLK!^-|$Kg5<
z45Bszo-&XW#R-aTHl=HUcS8_`9x4OOlI)ie9c$K`BTxeV5IyD*D4;jo%#44L5ygM2
zAYS9RDTpaKf0nlcYa~P8k_@-7MlxjEiIqG9sPg~M0h01&?$dm?+^4#h`$P|MeB*%m
zae9K_2oF%!)Z)^Q3k<TQx$YbtZOw|%>0W8AsOuI}`VE-%JGYj}rnQXjW-VLqQ0Uy3
z(77+6b6-Md^rAC*vAh>PiH1Gd+FcAY<iBGEqFNY%f2dhINZYfeNa84b3KY*|hzcm1
zOG)s{%~a!1LuLa!QmJMQFYQ+^L#uHya7GhxK&Hn;6V0R7(@Ng6Ue1bLyLQQhD2abe
z{I`YalP3%eC<p16U#%19kK<k;DrIvWfLyq39!6NJCT8xaC175ybw6|vVKpw@UDpgz
zu9uIff7`S3%hikX)fs~8R5JuJ_ysO7{w-Xv#<AZ0jw^LLu!;uo_MKQ+k^)HXz#4ul
z^@Ch2S<5E}?1h&txS-q-7nF5eU~>lXx&HOJ{`I;3_4&ex&)svMvy{(S%I7TQbC&Y9
zKlZrZ6IK-dAid2YSQ@Z+%3F`FP|%=iZw2}Xe*|gG4^p!LPp(NJ$RybCzb=K610~^K
zZlYXE*()HJmDK#^=ohgF35<^&ZUuos4-sw|+Y=g#+wr|VWsTDV#0$W9;K$9nw6(Z5
z?MUV<;;79IadQGSS;jU+sh#4Qe3Xb);(97Pj>^XoB6_tg?u<}0pt&Cc=J?M7fZPI$
zf8WW`ZYu-4Eu)ACv{K?CtHvD;ps1R*oZvWggUEyA7q|TlMRGoXEu<zw=;4OS0||qX
z5;HbHTrG0na%rq_T*BjKa33)s*hBOPh~tLZKnfv%!ZCW1e){44zVepaG^mh>Yo%(9
zxmca7@n15?h|3CPZe(+Ga%Ev{3T19&m(NHy1_3vh!6`Ttm!*z5!vSBnX^%P12?S)Q
z$?un;w>cF7FqhF-DioJxnK{6JMx2+wefaakTu85U8c<}zWF0D|M(B)mAwwc#G8EUu
zIC9nSqT}d16J1QCF8v+Htju<Cl*$)|qUZc~o|&k4bMtyjYDPO{jUK6q5{MYd7K{-)
z<~8Fx7=A+j8<|)BoS(l5tr<sxP4JgQ`n8}fgfjV=xa}<kG>QmE;vyk`NdPVjSh2jk
znP!Fj^FT2D!6ILe{^i}S7=VI`Q@&LU%r=>_RZ-x#k!|5Qu|Kdanz6?|T2kQLiv3$b
z8nM}`E!VWy<&D<bRP-Z_MiN7TF==-qJu3~hM$Ih3vTkBDyqd^00NW{yt0bDDW}Cv{
z+%%1@r{t!cR!Ae_`SyB$q&uq+^@{WrxzQ?4k!)tKG?*2}*VlAoi06&#*rEuKsgA-=
z9GG*|lUKOXm*`K?K=N%DP4tv;$2F4r+G0z${G;S-T~f5mV=|XR2P3m_Wp5dJ>1VR3
z2t;$L*^<T0CmHCTo@6>N<;!!O*O^xtS$TpAeU-J-ydvqeLMU^8T@qH_VI~<o?<0`)
zmY+7vChzj*mm@6|Ge%UJw@1ua_PV&~5=J&dRSBt<F?nNasG6J9qZImJV6It$RqAnt
z(7a|G)BGJ4a=6sEMmeS?^z+>SVZf;na3JvcR<}B79`>5&J6Q)C8IF&4!`s^r55KSc
zA$iMBnm&Ggh#ydYld>+bMX8{1`r{wdr-%0=)*{jInev#lZL_sR%WdOT$WIiE@>s12
zHWX(qE&h);WWH`Oc+YDJmt>hO--o~$SHv{Wn5Ew|%T?%?nW09Zee<$QOvy-}dlM7G
ztIBsx(JE7Y9iiO*DjPD^<`m~aKmAlgCh*E}VEcYUCe4n2+UQ?p$ejpNNSAa&9~(6p
zrQJyE&oXNCT4}3)kx`>}3cI9#&!~4L_X-A%q?K{-e`(<R8rfw4jP;n#)s8JppVe#>
zc`Rhmr^xa9#2fW)O1s4y2!V(HJEdWz4RM@ihoCvbjN1zQFZzs!H~;?p=}#a3t`&p*
zCr)Ke<+M+K2S0tyuYq_vlQ4ZL@18f=qn)%0tY^qRfEVzFX9#2)`8Yiwm(2*q^b^0t
zVNlNVruOjBMmRJ<TcjHjIucG9l_kIQgF;zbBk@%UiEj~0fsL;WFHrF{{`z;B%25BD
z9zIPE?~7!~8ke$TUd1Pa;eT6_<+Zf~-Ah-7T!=n@l(US)1W&NMvzvnpZp)ZqrY*80
zG<%ld*u~VPn&it!$FvMFD#0K_@*Kx^gfKA5BUMXDQJ8#mTPCWgE|SgCZGV6KTEIRt
z(7W!L(L8DhN;4+T`hy~L5m)kwytA7HV4g0s#V6`wldVvemBhpK7B(Q^CErgSNr|<`
z#@qXUv*O@QiLHnG;uvFX69Z7J+d~z@j$0NEVvL5CcZCbkQ*)e1<{(3f1x7c+7bgw^
z-#PT`BEL6BX_>cC2K`7i*;WQi@migAqwyfTjWiC@1vVcGc3hW>`$zP4>9UXdUI4_t
zcSQ5;@%S*4VFB<Wc2h_%EqO~7uoL0rMooNwgA`>K8@mx9^E`=L<*kcb1#64JI*DmM
zq+FB|*q!^XheF3t9Oj(UBUF@ROf_SsVSdYLCQw%IltRuq-Hetm->iIti^Y_sTwKZh
zDAcy@aURJo!C3f49-?A<juj(RtmjRrCK%rq83L9*f^kywJDDS$j+A8>*X?opF6zI3
zFb(M?RJ^4pBXB6=QL0swuIbC6@_SGmf<_*Xvd-j{#jvU7J%`$lUNLDaNxrn=0iW};
zO85B|%aCw!_URlgdS13JMQ<aArnO9|VVjS&Mia5t(1_ul=96cwaU)dRO^umHP338i
zF);>;Qv0^`Il^fTE2wM!mCSS{CCkZwTBrpjJYRCFIBI5crYQQBU2BcZb)-L%z@mfK
zJZ=`4;7(uwpR7A%D?bq&ghFb4JS-HD?N_d4@+_n7P9QCC3rx~!O!I3=T&fS<mC<e8
z@GX2KdaljO*Iu&S-@TGv>?8~ZL*9Yx?_XM*D3#2&(k^1lV<%ZWYC80~m7}eHmw0qA
zFC-xc4s)4b`?qo=<;p4~v+mh)lSEHT@v+AN+$b*dMk%~<1dqPU@uy*xbij6%-gT8z
zZw*cwGnZU9YMB+icA38u(aSvX<!>*5Pj?d<WXlI!Q5rWqc^>GVkiF6!pF$>gMh1-1
zXxY!jyS#TJ7jKDm&c*+g{{NAGT=Y@O?8mOoZq1n8MhF;WQp^wZ(_rRe%lOysr&ZPR
z`PeXl9cL{%sM(JM{j8oD7qtA<|Dv9maloYp^%mX))P(xOeX)3$Ft7UF#^K4fSe|*N
zEmmfBZLzY-25V>s=a6Ui=evzUaXcV@$FJ0t2OW*y!ke_W9&t{$A)aM_^}F{AAjEbD
z48pC^#NDZ%g5m^s1a^0W1F|`Q)mwmNEAAXDA;iws0H5NMH(h>Tgj=Fs^wjj!#YqQp
zirGa5G4{TnS+8<9spY!f;?~i8JjK<9#&hCyW|^x+(<@2oy_SmdMBg^g?CTQOxRCfN
z5vpb~tKvT4HC@DYJ&oOe>2_K1em+?;w7gQ^oh8`s&dOKGW-~Ml9F>SIdr@X-fq`8k
z%ceo8CW94C^+Y(d3^iww6>I2nOW$?9hCSDyv|HYjIJ5#CQzURBV85EZYYDNkLH6}a
zBqSa{4)R~>;kqQ6?Z{h1m|OXNJbag%0;$ZuE}HSKt<vz-RNpRt8uN$E{#`}PG?yy_
z-4%f|sxPIvmXf|*8Q3(xlgB2Dq<^=0T9iU_f@<KXd3rDGY3CHD4K?ohp>(-Mw-ipt
zz;Mvu%3l{ZkbH$Xv0EJ-x?)?<xNmGl%k*4ICG1Nj*KSaCB^dV3S=_&Usz*<<CiZ-9
zxu1b=dm1toL%NiIfs(c5jDW4k<kGoNe`^<Ka-JZgQO3HHFy1!Zi;lO=+BK0gdYalc
zM5FbgFS&5cf72aGtnwYoPr0>FGh^)ZORfQF!LX-S-Pe-!bSt*+_wxo|SvXqBLgR6f
z2wHIhzLKDL1oS5~leGB1bo_{ZQ>u$noYtJ*jS0v#nqPB&xSdoVr$~<TYtGmQOlPt)
z&n2MS1U#3W@hpXI#l)66F4SjvG_W!N;tD!of6X$PWXx;~J$1qf-5sy=WmhqcU37bC
zR8vFeL>BWs!`1yjqlUX3@Zj_amAkFA&lWm4Wj;j-&C0Zm-BQBJ*S3q=xaT)WqJ-Wo
zEv(<yd69g7kyH8%abl;6`$yXw*&cuQ);-!OnvI%irA^;qV8rQLaCW7aClWu=XpASz
z;d?VbWLE(rl9g$}SdZqB#iqcn`2db40o14J8`$Y6(a&p3z6i8Q+vWym$kNzb7)7RD
zy!jqQqfx+E;}6tS8u16?f58CKdR^WJ?3&e^PLEw(1bXc6c-Lc{y{ZD$9NY+bCq_ms
z0y6?tPgDEowJ+?q@6WC`=%-$})2Umjf1pJeI5%hqGOyAh?DOy3AvCukC3o$_1&`00
z3_6AAmvo+O91hLAw;vw<2YyL<Vwa(}ITZmhm%$b%6qk%YF({Y7q&brVnIC(%R;4){
z1_VKQlSr4Lw>cF8GB}s9Z9Ef~psP7Hf0yD02tp*yw1M1g=$7O#V~iXGc95Al<nO1r
z6t}mw)H09}(CThJlC`i{{Hj>Zr!NzqKAaw&PCuL#OqMw<tel>nCxH(*{udk@jh?I!
z%qTHEeVhJv%egf8dcYTOt`xp3es=h-?{1WE*h8575_dNi#BJ>>AE}ShX_WLde>%U5
zOu=DK>nsoRAvN)cFL(Ekr@x+fW}P!{!pHD7e*5X^^xLTjA{EodPexlNxR^eFJB0@z
zG&9yuQZg@K`u*4G`Sfv|m1qs`QEg73+PE=VcWRS{Y3(Ls6jRm~wV?_Ch14jVmiYOM
z?<lZF^+8a?J-%ZfomEPG!ZWS&e=}4Ln7g3S(9BUU{4wcdk?Yl!C51@O42YIRjcW#4
z-6jSCGm$~mzUFr~k#hL_BQX|w!)<2I=9f;rqPRuG($nX;%h;e2aCLm75JbH%yoA<=
zZkA1TmR?1vh1RBOxA}ykA6V2+qZ?@2Wtu(g2B(5FilhloO_{y_cAQpPe}`ZNv-K60
z8K-G|_tyxgJtIxYZyT*xY~h^N=^=CDvJT_ZXcszbul9}eQcHT?=~X66#+~l{ZPq{J
zdC#34Bpa97If+IN_;O0C>Gxprv(`@^XmidiUwVyOWTu%t{c`9$&)U9H$}ug}ID1%m
z6Fr%_Cp*royPzvktTfVGf0IOc$L6Ghx}q?xiB%P@F%AltGIE@a=NZ7atVVND`6ZJn
zk@<iPE~z!6*gJ2mD4OZ!qOI$<4Vi0ZRk34REh$=Qw#C2%>UwDcG!X`dCQ$ZH7$4U?
zoJV$8gVX{*XH~V-Y?tls_Su5dn8x4avAg~zDJ%ERE#R(a@pue5e=*a5-V)~pr_r(f
zp&go;7<kD`hDY|eYTykME}YlDO*r|7DN;bSpz-VJzf))yK?lYdi|+~qN{T8v{qUbJ
z&wu&+vtT&jk1zs{dsLh7Zu*wKI{fDpPV+T?mkgA|;evUt3``EU1Tw74z|!z}Iw7IZ
zT{hjmhE~O_Lwjrne+Q7^K&nchX(ArtjMPf>zP63TRV5K+Dr|&rVU|2Le8urJDtryE
z_B(aOQU9Dy&(rB+(Wh7VhwtdL%#u2hsIZ6#8<kR%kxY2NVUc_6U?R>~XzawyZ4#w5
zfGQA($-K2;H8QCpDYZ1xQWH)^Gf)$~kL@ec4}vCrB%9PsfBVUzax4Uk6j`N7t@ZQg
zZ+X}k9Q2`iPt)88<(a^lZnQx(s^eS{P2!R+LABhSdei>Bzf+nHX`T0C@9T*&ZOV3k
z4U-}ZAE$=JDC$j>_2W_X;37Hg4ycEiT2Kpjj4M~vvVveq0#InuRV^(}^zpdijf10;
zm-kMu;tm>>e@+;o9%Q514$p<2G9w2(9=8rxX;31sMqatz9%Tiy*n1eYi+dCQ^&Hxw
zp(G$@?n2I#w8+^^_u&-N3P$AUOUz*N;h*$ohjW#%DJOK4@sQ_vJ+!N+wL}MAjL|`A
zIgHISC=oOMD?07)1lgTJR|>wsu4O1=0@@E#6JiNGf01;hO%&vE<3!MnB#eEZd-!_r
zMquRxacLSz#s}SiEr2k5bGHLDr1&j9DR`S^v1N>C;-J*`v8NpM-aH}NMN(5^29h>Y
zuj_BisI|2QH?DP+sab2mmsd=9JW`6a?nP@s&(fRRGq#$f73tT(Ym#2}RtYq&>;#@I
z;oSJ6f5zrBg&|7J;}68ynf`r#dQP6L;oFkdN?+MUz-jjzJ)z2sM7+?33S2M^(CPEt
zzuxB|a(M`%AtCx9&+j*>hyYbgKjiPIr)@$~FzNfXe)r2ZA&Cc<)fb5)5ygV+hkT(S
z7?R<!89-eZUnz97L1_qleR3P`e*Wc;_nSc(e`qGG9WA%9%=>O9xBhCCcWV$l!?K1}
z*g)XwwH(=~X32+bnGH2Z7<t<)ZpnPq(~P=z6ET)NZ;B)EXm!?fGZo8PK@Qe3gyA#d
zS)3dVHsYknY!HW9eRF!5x*OBZE>WG#tvQ<jW+{G*J$=o9B+uGLht0XVr4*%1vYS>Z
zf7|w{0I{A7D`Mj2K4hOt<JRHsVomNKnB1!gKy!e&1!OB|TmZ5al_ntj@A#iHq^d%w
zq%_6`A>8p?rT}~K$)^^s<_PtshNv|qyqa$X>(HysP(wNG%q)@~bMc^Y!Artc4d7Na
zN1=Qi58IkBp~Qm_V<zCPoLnzT1~mR>e+AV#jgj6U$NjmOOUTTmjOSxTOSJx%=KBHR
zEk@AaDc=t&*ol0z49gYr{R)lMYdz-sm6OME%(>?g43c{`IrL0rBiFConCs7Dj^y8*
z81rvV9?bVI@v52}`~m%}<oiuF>K@5udi}RouEmIYX~mn=@Lty6%Q1G-ZY%4re{Gxf
z*KW*YH_w-|{<+8Z$odOJY_8MBwKD#SIo$NB&-OczOWFPf9l4{~e#0a>Ep>%dzeJ90
zJ5!hD*Y?shzgJzFpKh`2Whs7T5h8<rMUFq<5o-%Tl_XI@R<tvN-3dRB@^imPLuRgj
z8Y0kw`AR$-XX;|o)N;UM!GLoLf2Wf8UY;9hZS-A2SV&9W4q1~$%$jFh!d*k;3jl9(
z8xi8VaDEPJb^NeAv@5|Vqyi(ASdSph+%_mjlJvWj2}zn0<H}PeWX=0Rx_fEy9_34%
zZu2G1yd+<e=2**@#HXG8L@Jk_V=*r#-SS+sc%{&-dC_awjY7N_ZdM14e_q>e5Fsk5
zm)33Z1WfwZtwt!@ll3#?j`9fTR{dOjQN2TSL*K2lp;(*3ILE--Uh%C0d*PT6a%AGA
z&|P?h?ihLv&^=;pX?CzXT2_eRc>0%Zt!epm2+u1K;=h9DIWCytF+DqjG-<&M-%UN2
z?TMQCk>+awL2l98M@OR6e;yrq+_B*75=D(#k=`@q)m$sR6oEgXQiLgx1MZH?Ca)pe
zjPG#ete}mYL$oX@oPr=S{fnCDT3Ssg9}`H!j6;&5^hQ?_@{2?7HVg{ziyTFPJ?G?j
zlO|umt>lUcWe;*Ixy99$Gi_bPl_WP}+E}(E_%0DianHcdmNb}If6Nx}P<^;4o(jew
zW{Z(y>Y2${%fRYunoK|Ap2_sR_al}KQ+g~X@x>cHWcjpRaD$LtU_~bvxj>qSKkVOS
zWrqj#HX}QQ`yTi9trSs`^sjMy08D!~*Ft;aoTYy|2S;NcDu&23ODtIWn5Lf~RFSMA
zI%TnLe<U-;w&toHf3ur#&-%O0#Qyj5mq&Ol?y}d^8+yyj_pu5ELwe+~Ik$C_kaM+~
zq#D<TV|~$9lA#ty%sFYyvf1#IjH;WYZnszAi5zCH&YLf7;uC1xkt2S@ZZKV_KaxHP
zDNzYZNqQ9rTGD^zTYN?-5Fq#0l!?w|c=Z}#m~Q0)_I)i1f8(54qps;w`Wg3}O5gkY
zdzC}n7Pm5J>jJ;>3Wp)BW?*L22c*9c4)@1j?@v10mrcpTJb9UVI?Ib+*z!M10pLO<
zsSxS@>`h8Z5g|F_rJ_<&?U)o+yHajxI1^GdrJbdrD<v(hT`6fDT#7*pfu@Og`16#K
zD$Bg}&{vHve>VPKmXd<O?RAw(Qel7@sSj3?Qtzu)k~S<XT)-DmFDWF|8eUaMDkO%9
zxQCAwk_r$^*~LOqrBf;||NQBKV$wA7NHM8O*Cj`aNfmQejQs8ulNt;qog7CUx>!vz
z94sis-Dli|%qfBH%+Ey%3N&Vbfk{zqJ!x)fz*Jj1e_6+LQ02a<t*JEHUnHePzfxT%
z!Z#{z&~0(UdbuK5FCmrTl2kvAKO0#qM$A7<ad#pL^iY`vv+{G7g{1ya1Z5iqr_0T)
z1hsT1%P3LF4(?bKH;h|5)5ll}wQ_BM$G-wSqYcMs<6#rB&Ex3Za-_!fESDo`4R2G9
z)Zm;0V)aO<0^|rTs7KoSt*{gAg!0@J3)`+7&R4tJ|50MHV>etg$K;dCo_2OMTWM_<
zQtV%^WYO@1QxdLPdeuC~<E4}bi{*<rWok59(6!n95dnA5^M174YAAt1JO3Z%6OH(n
zp|?2}m%)(;6qn1lImiKex0JX!1PTOi0-Jo7p|?2|0XLUnTsIV#{HHr7mr!;s1efu`
zIg)?1+y3?X`-YR~`CjBznmG%TvA|F&l6TK_zlurioSZlVyqwZ%`iZ}sj(&SV%{XyZ
zGoTqbJ8+7!E1C9!o8^Gt<uU9!J;%H4q`9bjFPYVD<RG4d{_IoII-p*_WsL&r7?2z6
zIM>KoqZEJ~HO<sK&(tQ3##t&|fVu};XlZ{MF^R5rX;tN>Pxxz*`EH|xiC_GNqu^UP
zwJ1+nLPruqe!eIRALWQ#Wto4Zc_c^BM_7+&Y&Cy^8toYOW)0MIoDyOzohdpAcOl0j
z7d>(b$dOVH$dS@qiJUZZ4mm{2CbU%QX@2EZz8#4R&)mw2jHF7{(1_1S*~;KoDsO*H
zNYRJ<6z7Xh62f7Xi}VomddX@4l6fyieV(^a7$<7ytphd;j%X`-&FHV^E!@lgfcDsE
zVhPnKz0i6MRCk=XGD$xX-Mgf(7%g-fle@;}n{Zu)ZpjVv%8%TlU*u-=i`-u87x@yp
ziPoH7j5IQm(|o<7d8Y7igi5kyl&pWqn$HT+G8M&9+5>%e-5O5RzSbGAEBsVFEfN2_
z^emq;x;p9d?stO^wQ*Lk(23m@;-%oYiwZXDRHl8#z2|+<zLxMS0MY2zz;DMX#d1ey
zJLe|{%1t0)JNf|nE#q40>u@X#Q-0jvCyMU{$LT4fHY3tCs66A3$@Y0x`|E#+CqHu*
z(rQ9_iuZNDS!OtP(&em!uuYkW4}2FeroCfGXa$V*oKnmPIP(=#5o4D8B`04=8MZv$
z<Rb@eEw;8_qCL8k`r-K_Cr?3`YV3O;A61_OAqZ+75!NkP+DWGD8kIrMY1z`Usih9|
zEUxq<Se=;@j-S<&bP+L)3=w}S+V#-jz_tiYm4ltdL7|+bzAffxV;HD4DuJF;MTs5a
zit8(Fd0QngaQQ?v<WgPOM;P+&L`h{-<t!uGe%EAI$7M@vw3lBaJMKAMuKC`UO%zM_
zTw=}V{8;nFk*jLHICHk<OSEPzO*C@lRd@$W)6?bXjWgU>52iIs^`L)qZ9Q0IDD@zX
zP#pAyQ>!P>uc^$EJQ1Zr?F)}6$GLo2E621d8Q)w@$B>zFh%%*(Gk7<i%Y?zRzs%uK
zDy9wQ1R>w7y*%CKbDAvZ&Yv0T27*`2^V;*`P0wbZP@8?OJUPQb*R*Go_c}(ut$1om
z>gRMQVvrQ|;Kakyq~d?^s;=>3uym+8R9wG_-O*%6BHNOyH9c*km<DV2U<bUt3{PKQ
zH~(!$y@NK)fSvySu?atb(9BprNy)r`>E}Pw_st(8)uGMF+m3CtGB+laz^#=H(;ByD
zkgkNz+qBumeab~2!>9TY@z{-gtEiEP8s+6Ek$8^Y=^V}^s>Xkx?4UtV=@*Fxc};Xz
z7fx(8Z91#x%(&!bvn^4}w$^O&d{#1m2WBGQ(206T+`))ogc)UhoF3yTZB@CHrh3m5
z$~p*?3(H*S(fL|(W56AA(!3MEFQ?iH_W&dJ#}!s_U7s8`p|4N8QhTO4*S7q5hrofA
z3v(Xs2pp(*rdNN78&FE+i<ZP)IU#$Qx-gIo`5vJ_DMXQ2DH15<(d6?7rK;-|&n_^c
zA@z~NjAs`Hn8O%2%Ji9hyW)z0l)r>;A7uS_-#!TRM&B-7NR<7RkiBTokZ|b}YL6*?
zgEMN|piR*W<=c*$-jWKWO#PUOk_#;gJ~lV6ZJ@t{&zFD3q7yN9_<RW#%MJBNr=Oc7
z>BWI<dAq<+M@{E&J|6&-Wgf#>FW?G=faIQelw}?FAO2>L;Kv42e(7RHaW{nm<C#E6
z4cz8JfekK9j4R&Jg}o^n2rL)5gI^*V46=TFG#CVWV>A#Z{Lw4lfZ&V9#heet8VEly
ziElhd!~%ag?!KShp#i!CUvzJ?P){h81=Qv2{wrG!67_#F>ay+ge=1~e^R8h;05B>1
z9bqI;a!vR25n;rpzB;86BEQmcKfa_1ugeWGuMk-rCOk^76j*#9TVDz6lzFozZ;CZo
zhbjw$0CS<nW3lekuwd;-Gv3jH{{!8AVQ30vZe*9mNH-7yH#e71&o3pHFWNr~m$=C}
zDt}#XT(=Q@_pi7>U(yG2Ih+r2iXcEf>NJ3Z8g>e_HS{1W+eE1&8PYmP|9#K!t|&RX
zLDt$EtBsM>++E%~!{N+1GrXi+txb6%{Pl^rgj9KArK#mi6MLSVC{I#6e$zHdok!(K
zNkej)4C6zxsalt+O-=bxNOkg7`*&?3qJRFFiqE1PVuN;W)<xW^rCZaW!?pwP*{t8z
zWGXp#_0Cf%)Vhdoi<VI43K1X9rA~!;{yjsCXXl1{i&E_(hFa-{HHdXK42G1ZT?33R
z-uet%n|Nz^^NI0@uB=Wh)@^hQusUnQfL6GxfZKAnp=E&wF(1!t;mZQ4>Lkn>Qh$Yn
z+abZf(p}KteZk|QsS)nh?xTh&i+L9_o)pU$A|BvvR@$&G+bQ~_Mfw1&pw6~7qDg04
z8<k0$&)?c)l>HG*a7e<Fg?EfvSSdPL!6kmhxPrBJJbp5!`DTn}4Og|eN34sC0b08V
z-<e4_3X_BOa9}G49LL;3EK{W(Vt?VgT4qy=OqJ)!7M-eccMOKrDiS6TBuu+XX%k{*
z9*H1qvs+XU#hgQz)(LLx_cLOfd#E>jVo+DhJmIdcmT_A+Bt)H>&o0Em^+|^aPZXM;
zfK_AZb6}rbs)RNua?5?bLX%?>T`Z9b{Wb|jIrsbe-NfBd5}(DoZi!F!u7AAyZae>O
z4pyhjJt)2ODaLr%HK|Jc=<|y7Eb{p=NNuEG>A8TFgvec1JQ!KnXCJ9dI!`|`foyh=
zM*7w{j}+gKQ+f39$E$Dt^JcyJ^y1=jvwHFN)h4c<o?rZXT77nT_3irVWd^?Flht2V
zU#&iWN%O_?_1Wen5<EHFLw_u(gHGY(5d$L0ik_@Ko$&Y<t1mCVxm<lQeMACNO8m=k
zd1<x&@e};_t_}OuVtNzZs4XRFcSK7%Lk`ro0HlQIIeGNXG*+LE$DeJ^FE3!v-=2RJ
z|9|xTX7lFyPpj2;=bP`}zDj48Kdydw`{SF(->!dJeSNz5{;zK~Uw>U})>r>JJzJ+A
zUawxgzI?U%iBidlR@YZ&`v(2tsno|$8)0UT=Ba)0GVn6|_I&mCfBZ9+J~aeL#VTIB
zef|36*M?OLL)ro}4;ZFuN%7PJh9U?O0UU3KPOHx^UteCmc!N1i95+m>XQvy?>S8MM
z<)44veEDK?x>-*&e}D1h@?rxePk2s&uH{LrMiHJb0NbRw%@;lR5F5Do<dfC2tIM+&
z>&?s6voD^kzFGggIk^)C&rZKv!yq{IVskxbM}Nli_4Vc3tF!g>mJ{N;uh-w6pMG}v
zbB`87Mq}^^o^XYG!_&2EoFnUc%W&O<M=xiRi@D4scO$hS`G1Yn@{njqWk@`v?uL`2
zlB$^ZWKuzLRRI^}Bp@wSO;E6#RLTvsn2`-RE5J~5*^^f9X@z7-0T|Gnz`dX$SOt6%
z_mq9d2=<a|tgmJkRs$&=Z$l|9g`2@fiILx2k@qd?7fn=mjUZS6Twt#VK_Ixh+3{8Y
zzmuVQ0p}-&c7NrXDiB=Q)86HP7cf2xxL?2~sBDfmf~=&#dn|Plq!a@~46X^|ihd|p
zP_<M7=keE2p>QTF1ZvD?iL=}@Dh>yHa7OBExCML`Ppt9*0ss8sdUJZQIR_4QtOw;h
zeoE@MXn+5V1Asd!@q#?Za-puH)R6ZD(T}p`cXw_<#(zg`d*j?fIU&{`(z#w9KLxP;
zf=d0C(GYw4Sk!kpkA>2KLs1qbuJxdyKq4mjzhP3;(mPBVZOVO^H2SYwCT%)6aBylH
zoZ1Gbw!x`Be)aLI-~1Y#*U?PM+5it|cPe$z1Rf{!d#I==WucvnLa3o3Wdd~P!cOK3
z5Gh*nWPfNkOoK?V!XPmTBS7;|v|~igimWT>7Evan0j!pn0B(g+9Y&O_Dc1ehjEMFj
z*IdvsW&|F($-D@k3)X>et-xPtzwU~K+XWMbnlb<mT~Siu4+wC;fvfXZZ#V1fj&JA_
zXfJadKBp)~BD!wor0BfR20q}NoF3yt?>8+pvwtXOQ1Kx?OO!dpm>y3;s7B=;o!#~N
z$Dh_$>GJBk<NO?*Z6wgBJfavBdpJJ>AfwBa<EBO6c!}MMn+}>#gNQ!hw8$|nuJ2NQ
z?D`$hJ-<6Vol}rz!P2g$ZQHhO+qP|+U)#2AyQht5+qR}{P4o2L`;T)|k+JSpRIXK#
zS#LhC?IvIFi^=YJK;cA%{lISsShH7zM~{BVi9oxrE)B?0RIrjRF`zplF`?qO=psPC
zhMf}Pdp3uwe)v)t>sa1@vrz%)7Qg?Ltz-Kd4y#J1@jf7tj?z541&kXE(YOkYGIA}h
zM?#f$Y-7zxJXhu7N2b2qG4xERC7!?6EhwkNTkbi4=-0TslT(awVb|D#hck6sT!hl9
zo++qeE0M@$p8O{M*WFKvMDP;?ki_+-TvNYH;uS{8wF}IQ^5*9+DOfka@}!w}mv*0P
z8>6p6z=~sYE4z4DcAR#=H)%K<qik@rZ!SigfljpOKafw_H_yC_$MiZ^jLB(Qq9_%M
zfVEx-J?G!A2pnYbEE3j9PCI@x)p#RUU3#@(-S|FF{g<ajI+$~kaE3Vo(40bamtZQ{
z*$>|;b8lH8snR`VCfFM^4fly>UCZZ}qka&D9xrp41Si`1>w758GV&!KdtemXC-5`?
z5!Uk2BbL$n)V#Kp)|?SX))N;*4b-65j-piR;3EL`NDN=VZCpkb6W<mar;G4u?bPhi
z;0}J9s;Ua1We~W5(T$c40Af*kvkMxgnIh0d297RX+o3jm|GP-K3I-D?TheKdkR7qo
zYqWDM%48Z+ptm0FMu_==e4K9FvOCOE@Z41u)ZgQ<=#O(HM62Zu>}C@LGF3JKbBi~_
zxqu-{tM@{sWtfI;%I~qw=rz{apTe`i$D6*#Vz%@Np6iwn7+<s>5De*w2y~Gko<346
zb2wjmM~H)EN+J>ti^uFbHb?aT1Ao<ei_^xZ85<%J5s$9`H_h>_gXz#)<eO0EOMRIf
zsB<e~W`a&d%XWmYVW$$1Nm56qBq!=djf2KgcwxfL;Gl?GZ7d+=Y>ne8H!hNb?qzPI
zy#kg-OaLVy9q}Lml|*kYN!m)?HB#;p2M1c6EUuHYd}C9;Mt*(K8>30M)niAPu4~+x
zJpkG|0C9O}Ko;8LvN2Fc%>jM$RQvE$-9CD~-<*#d#uACoAHb$-xodN095m6lLLDo)
z<;jJk2pEwUy~MZaj!R1`F!4Yr8jPrNas@FR1lS2N5OVAvd|kH@@^KcC+*F3tW@W9G
zT?2)3Y3NNw1OcafxKkC$&id?3kWiF@p;@M#SQ`3ulv1uJgRP|-``JLRBUtVPBS^}a
zf-nv$ZK>YZwB;aCE@S$Jgh-sB{uM(x%ae|^W}*#N3m|huCO<y5C5$`i_*E0=e=E^H
z1mLe}i`i3>pd7j*Iyhbl${L3G!0{w(d7$l*nNP^7?QDXQAkZ@FNEjMI7RzlFlM?p)
zeMv7y29kgQF3s)o>)Dwm@RtI5d4PkhAO}&e2i=^4k5hpP`&=2FB?xuW!i1!pu(#=M
zs5JyTDl6O`wJAjbH-p4BT!AK1MkbIFKVZ1Btj))9Y)&jlMO4Y6IO<nS+AwW}Bk#nb
z9WVP@;|*CiQx-XqME*`o|I((}sa@dLiGJGkHKLZil<HTx!5`IiJ1FZ80~y7e9Oszk
zflXl@iT%1_$5o}U3(6<EEu%wt9FMi1DCLb^+7G9+H&%H!4F2KpCBJPSb(FgZK+-2=
zYs1xX!v!;n4_SX6_1iiw$+07&&G3VBHr7|z=*L6VJDx8a1=yNa6l)JToafifWNBOB
zuY-ySq^mf0dU^JMZTA>Kw98aVWgM@?L5j)s0Axu^Q;w(|)d_oV7xB+fVkEdA5x(Lu
zgQ<|-Cy!AWrwBt2vWzXY@%OL-n6`p$`vCw@uL^tH|GGDv=}hMoDCvM@u3kWM-_5u8
zKflrBi(!QX*w*Y8rI>Ci|0FrooxH-NEWb$lbDVLLQe!P8t@TSZ;8b0zf9|8Vehrhq
zYC|Jg#L`nK%kz)Mo1Ql{2A87LWOI$QPv<q_3ybDS3Izefn|nV_@DXlJctG3pBT8ON
zh|l=;W>t$&H7Y{je=|r8Ai@DbEW~azB9wj(6yd2N0a;SvWS~dkEljAlvOPfxfv88Q
zuL;=PD<C88hRLlck!PqX;_)j(7|=OQSTiQjwiS1VR5t@eK#@>OOUPbUl{8C4TzMqW
z)S;4Cy9=#nNWh8t2o*wSbYPSl46PZqq$z4%S-j{QFV3=GrG~}Nbix4QF5VzxCnoKb
zeIG=wQl{8m@duz#@0f~r{EinmH}Q!Q8~ri^p@l={7lp72I%i>}7F8+QbVF4q$KH$c
zoi8D$0`-p!h{Tu&>x|hBv@*qG#HYFwb)*qx?+K!!i@7!!#$bkVRLEvzNC922xxt;M
z`0D27W~R}`(8LR*Ib(o8EO(2qd@Y7w>lw!qI_)e)umc%7W$KVFsdXMGP~@_sjNwF@
z)EQG93SIpOmorgan~sB>@NovMJLPejTh^smUW*a0b7n)u<oz%foT#>TIh9FmS(iKR
z0d?=&bVn+0yauItD(&GioU!JsS!z3SUy+FVn%cejG6gK@k`h4Zq#=xM&{r7^i0A<M
zJRgR)3Ya8BZCPX13GJE2-tyq$UfOo1u2ZnOcC^Ip<+jE}GLZimRfk0TyCY=4Dn7K&
z4^LWwq3au+E7UTQ&FSrZbOule%51j<+MlDiI14Dd=@J;xM_D`MJIzH@s$(-IP?{cF
z2+1Noui81A>jpqdFM#@1eGBKgvF&ErHe^9z#zr#)uBtDFYbyR3+}@Om)R(9!@<}Rm
z+3ROI+*6Tk8d!}_Hw_kXNHL`hfa(@v&W~v&<5;#Fi)IQf##A|Fk7`_MLNYmZ79?hQ
zFK;G{PJ<4npegl-oTao7j`;hbt7%s#H9Pe-15?(&Fb8l^_J#-l+P8ywuhKuOVWmY*
zU>=rUp|2DlBSA&#7#TeRM^Pii(2|Jyg-^3fuhr~mN5gqBbSz`0L_CyjD|bq^!|-`j
zsYl@&^{4Q>wiB{KoB>tL6ld5_da?s)inv%G)#9F=I_6;6XsmB)w{=wn#Y?v$Q3UN@
z1m7{(<TGGfEuu+q%PkgK2|J^w9kyd&k`R-$PwWOzF0&(%OsNcmd|fJqT4_DaVUq;o
z*dto}D_vHS<>#X>3E1UP_Vzt!BPK(>WaAb~0vG(^a-G<_7`;dtDC9UtwZg3xxk=cg
z*#7&d9B9sn%bqw`XVM;tYRzs(jO7$)ANgs%tqiCQIf^d*X&yhz6>XQgQ?WBYtSs0}
zZN?wrKBz0OJcv8Uv)p7UCvMEaoJ8gZC#K4k`(o$7ho-7k8_pP0v{f}PIVbCT>LJmh
z!>CED+LG?r!d_iRYHgePYQz#JqVxL2iYF%3{E=gmZjZa1)EYLXlB$<LRG=RG>#H4D
z$qz`MZo5L&=1XP%%(yPl>Hin|<#UD(4Ut@Qo+dZ>j9w#Bp}yBEITKKX&B5*U14Bkw
zG}i3P$^nXn9r~NV5?oKfU;!S!LunzHNXh&%`QzJC6az*90SKCK@>b=Vu~p-mVST4I
z`zykpMfe;#9oJihWL+~I`Cweh2b2*m(hk_W38E>ch*s)$KS%PJEXR*!&}m7td1hWo
zft7@PU(P&Z&nb&2AAJAQq0B<y7F@#bL93C$PqHj(v9`?w2g1ObQofeZ)$}^Qm-WIH
zb>Nt(xhp3UKp1OTH;*`|1%6+KP27|J+H0!oKFD5Wb>i}sJCwr+7jMgLfA-O{=mm(L
z3&L_^h$U<>!xf!WN1KUm@crX8Y51Szc8I}khq#+Ra8~H!=S3Ahev+QlUnn`W>FCYW
z|Ejx`0~M1mq{GCu@!u%Jr#|5QPs*L0w;O}6X`V|tQF?cET2G>;bq2+n>ORM>rlrg@
zF_#;&yKCgH?OhMQJQ*J_HLOnD<>YK0#bvp+L$E{jr{2E(R`&v=qRMX|ITKFI|Bg`t
z|M=Gqs8XS%=Rna@rsr*Or2?A2E{<PP>E~LZar;K%?^YccssS2yB;aUH85udiaErUU
z8h4>!%9n$tG0=wP1mCg-8h{w(>biZF%<13^ITjSUDRWJvf00;d+^bXM<cvJ)1!<#W
zf7V@BM4MA$F-%BZobWkPRwIQLPEyg>u6}Il7a8x5y>>iqCwy=d`vLWLS{Y4sj0euJ
z=DLZF(DcZ(_X;`#VnM6W^3KQY<0eQR*fckkb<sZiZZRR8)8tNOJmEEdti;tMu^Iy}
zX=D+B$M_buONSAfTvT-X>4qWCG&7s71+ZQYMSRD1k{MUs9B;=R;sg9kd#E4?iVh}Y
zoD^8i97hVW7Q%{@ia)YUWOT|VH||EIwRt(UcNy`7t6lb{c;eLxV0dma|L8uaXT3Y~
zAF+DJb0ubSl+mQ@f*twniTqvTWLd+o+{%@hq_a3q=us3kU5Xkz5A}JAYQt2bhO4@s
zOgo2`f3CYk_I(>Dh|@L)ADu_4NTpR_Yv93NiGcA&O=D~aIbd|!)7p}LR(3^SScC>B
zR$u-XvQL@<{KnCP)9rL6i}GM(keW8K)16%_q8e*A#xaRlV)KMhG9pPMUZq@|^x((l
z(#IU$@DUf!4TfRNx$72sAYK0HU#oWL=LC+Xt+~az4!wm_k?G>_1>KJwT{m(8Wbb3z
z?MQ<0qy=#ORDjqFv-|xrT8mYTI$;JxK)Iz#5_L14Mz>L0Uj(~Yys?P@m;=1#Zayc3
zAO>?eTt)x-JLu;;9TtV5CulO~^1@hUn!#kH*NF`$o7MW^QcJIsb(;D|`qOG^q8g5f
zVCuNQ;dD3q65i#$WtW_;5+<t{uiHn(W~T{kj@g*E3IG_)^?p>J(Nhup#JJ|r>J!Op
zLt0~R&0t`0MO&)vE04El(pW`7tE{Zi&OiTT#p_A6V{vp80<A<!+x|ozmkv5rtQ$o_
zC&Fy3x@lR<eP-R7{YmSQRtFxWDR~RGhn4;}UVaHn*5o-&g%)y;aKD1kG|0+J`3cZm
z=u(U=DL^8#9pU+~h4z;tf@F~yGzndKqQ*S)MPM40YP?xfOf}AWf`jSe%0^W@`1ez<
zoCy<nn;m;v#iepb9BH2Yld_SaG^&R8>;*Trq>UQ~2@&_UC<fh@>3klVwaz6xGD{M!
z!FFArvy?Ip;05wGliq$J&T~|OL$?>f`c^*!G$4_-_|lD5Agj+6W9u5=i4+hqfoz~;
zSYYafw<QM?b9E5crDuVGM8(mcgRziNHz`lx8HaHJaw3Lz$Icewq#wxSQAHHf4lUA^
zt(Em)tCQ#PO-DreIUTSk+NkI-6^=$wo0!h85ILlvL<g(6+oz5BOHY#ylNEW}vABZ&
z9B@lGTz#mt>HE&t)r%B?v*5vI{rfVm+ii}`?cnD|!}Mn~K2ceF$?@_A#F}Fl#RM0?
zjTqrpQg$mRk>DTL@SuOb<fl+IDJWo7NWa`mG(P+VR#(6YS&|c~qpTaPntL6$Fwhw=
zGbd$W3lZS4$5K#$`;hwzn~|DPu3*`G1+X90!$p-8gxipJN{@!ZfZu<Y+ybK%i!W6e
z(EHh_zjZ$wwFRc^I6>4y7!=RF8mV+e`TJX8!<U4;+}*<fxk*m>PQ+^#)9Jhj=x^_|
zy-RxgWg>RwGW|5cML5QMN3Ksh2WeV=lKU0f6fqQNPY+e2U<agAJ7F4FFN~)d)Y#J?
z**8!NoJV@194!h66C+nT<N1$>QulzX1Hi-Y%E=Ds`)W+W>O|cMn{>p!@dxOAwpW1w
z(apV$6GM?LKhDOhc^aZZ%lHE$>Svz)?OSYL&M#gnvf-3y<{CCnULkfj2cbdSHmya^
zs>q;&l?6ed&7KUo@MDB(pY$zb6*SUR{e1uDhppY~pE+vI7Y`+FckaT;?Xt`f0SHk$
z|3z@f90RkC^pPBPM`*%+$z9@$P${>$m6mVhIZHlDWE(UphngPVY&u`8q>%$g+QPzi
z9z=~+6u!d3X&VUIz-~RwguaJua3kDj7Ysee1qpuq$7*SO4x-ke4^@M}_-P=;edtY4
zrPqk}HZQOONfETT$q|dfJi*;n6EF<60CTvFhUB}{F(l4bYzVLVoEWc9uK}L?$vxZj
zl483-xogr^PIILO{lW@6-+hFiQJZf9WOk-SO!d5Q25D|%YGmJtCG@Ns!;wHn{rAfr
z#RRQgYoph2?N^P?89gzbazl(`Hrd3%@2FNMGoet;{FRD&zd}$YLEXFI3P8*?HeNjL
zs-^3l0*lMUN7p)!nK-yjg)y9lhzj&|9p0=^?1~n@&TO@APGPiy{PY4d_5EN`ZE+JE
zeT~iNK~iQWV&WnC4>ly5t4A)GWCXr+9MNvK25jVSM2lIC)z<}YVoUUc-rrRbmkbw;
zDHcJRAT{rP1|J<<NZ-<x2LLlGeaU`j=YepR7`f`Lg0U9psI&$8#gtbf9fPr+MQD)H
zKdOZ%aZSz;AC9u$Acxopz)}Fn5xj<8fkXI!I#XQ{T=HxDdEce=85m}WEMhPlJy|XE
z_cdqq$dNn>Db{grv7$-|qIuW97-J-7-2LmyQ0D7O+A>Ucew<=OQUG<RBQM<qsx7=Q
z)Dz>;A?H$U$MiFI{B*pda?nE18)hsVm>^cwjVg2+Rc?h@ZQWGOIUKXq(8p+c`1|`6
znH5<Wm9~}^lNY<DeQq!NN5M?(3<<3Hbuc_>2N%`S*Gej-74*ZQGA40_tM~(+Te;NM
z3(_vHJL}?VvWv?_2>=<%2<A+PgOoVzEv&lds;TG+v2NIl3iF`e({(55Py*u!oM0E}
z5ffB|drd}OYM^C3m{B_F3_I^+0EQWUhc`h>o;06w9rIu}jETK8yqC*;VIrv3E8_TQ
zz|&`PJkVU~c=(DwXD3-gJO!SuWqfg%>dRt=>yh_s&ct?)7NDc6Jt+>%JA-DEy`_O-
z0^3&A`q}nMFq~RuU|T#h!HzdUij{Kz6Hc_K6xnsRh&@DdZSE77!&y!L!shL~en$`(
z=efvdOu-Pfq0pB_!5Eb=P3d!f{oYtae~2!xI9|5?+f?Zy#KpV+J^G3yn0-*@4RMIs
z&<|jY^-E8R7Er*sCbUge6wkedGxK&?L6-!Jn;>h;6>Vxfn@kGwc=F8HN}7C~M!gwR
zwdh^dcE&2<yQFF<X!kqoI#){qNAu9#<-99D|IvifjnZA+a6zZaM2MKd%hOk-qt+Iy
zFFpQPZdfg-NGC!?i>CX|!%TjeREJq^lSWe+r5x1!7vL2%PqiN^Z$aTes<bv?CVsa(
z>FW9`)NRgv1Y-qhmtiEPMqJWC0F>oFCo~p>payd>$o%cWzq|L5mt}cd9kbF2)7P|@
zC21R@V5Q}oqK>qxkU238gG9BEy2{mx>N9a#=OLulmFY=K5SByA>|%rx!a8c!3>ZvH
z-Us3jFaeQr4wwJ2<hauJNr_R??X9q>(v42J5&?4NAoSnsUmch%j|?~vqmDY_WT(*f
zT85$nLO~c{ee?I*B|db>koOKjL7;GNtoU~4Ad)Tz{szRtV>m)*E7?EDc9eSD0es~;
z@mft02<O&DA)%aQS?)hgSlIH6VjW6Y6PpF`j$T*Sz4jR2bXDptN~WKNu_};%`ufld
z2>^j9W&)PAynDCr@3$s*8hw^49)Q?76gft1nWYRt9kC<u=!%+b*}00Fl56qMCf(21
zPnT2g-X8f)lVaF)ec}%!?L&6Au;3$dypc>8MaRrb>}9u=WWA{^dS|k9TQ%EoTs_cj
zeibzaenh`veMaiGI&lNnzyne+-5bJj3xG_G7-jE{*>`;fD_422a4DAvIhA+$%VuK4
zLk;a3{>7DQ=F%9-Dayp+4h@MWk0284(eNm?2t<3ToZt@SkW9~OL13n-+qDLU!}Lh%
zKLw{<hT)9)BO<a$QeCrSC3RDFn%Lzztx{FwTq$hC`Y94d8t)ToC*M!Ji*~s*Wx$l?
zYQ)>p13qo$B*p^+8}F>ys2^y#W_s5FxCQqWQFI{?Se$q6b-VnjfHhHzI+I8rHlE}1
zZ;j2hNNu^L96FVl<o=|X;~L#s0ypTco<3Zdo`Ygrv{UrrnY9Y?kRC6_MY2Rv=PP^Z
zboMho@hNJs#+h<(>1}VeLw}qNC%^`2EvX}S15ZmP@ZXgsDeMq8{NdaZvdVkYc=2=#
zntM6iEUd<4-x!e|PwGSXm3OkQXiikJfXt8by*c=ccNa^DQ(<_qaRtk|r<33?chWP#
z7iHpfW>hwG{8&DVJIV_c-mI0PEtBSZ8j=(?YKxJ(mWe~#E2m_wL+k8+mw<^0sf}*t
zy-biUbNQLwSGN<^YU)|pZd^T|)N8lK2D`b`ZH%oJ*}^|RwzW8@Jotu6g)Dp`p}8a5
z9y7iYb*S_jhSO-eksDG6BiYVomh|}wWed!*@lT9p0g)1!zNfSU#HlYzGcASwH;e)W
z3#|GwjQ7E(3!deeRZ2Q;y#TFrhoWiOAiSXBma~*(#Zx8#;|42o<K>;ooO$@ur&TkE
zE0G<xl-iglNj<-%BSU4T9fO^I)Xb!!#|85vre7J7+_;u&<ipT*$(kRS1s>bCTe<4C
z$bi-Jl)KxJ_h3vmxE}eKaHW)p48hS(;@S~tX}~SCG0QJ&Sg2?BApm;x)n5C)6vK!k
zb?fgF-+P>gN(KSxmF(uIe)P-WD8cEH5NU!TYi(&}{BqQ-(|cPnePvMhq*+SfE-ZrB
z=@wEB_eZL=nm=z}wktrL3DYe{6fLf$9QoA1JX8eX!<G{Mb9&xN34iUeN1CVcO!EfM
zoM`X${)q-F1-OFchX9g8V|-ZiPuHu*-Fm}Oa*WmHq+x#Vr%X|JRHOt9@^F0<$^&s_
zY6JXH@zqZ@J;`_ru0sh*(Xc8#_I92>w^DO}ldLnY%J)+KlMK>7$si9jt|a*kzT=yr
z(QUky_^4D}e6nwReun=nhORZM8UB}{_L<Z15BI-PQjYp-`Y92DJufM;5q%+<srD}m
z$tNCr{N3S`4S|W%hcB!J!|pvNJqcK0-fs{DH9@8S{>0J~L0D1Kd#|{9fd_Eg{ja&0
z!9h<y{7wEh81o}`P?}{Rh2!L4kTtWnaJ3|4W@1UdGiIj*{0}s=Y9Lg=a3=Emha3bz
z4g4Vuser&+Yr-d}&jpkIkcOgW64d`o8oI`Ffa^FNOspS-n%nm95pgz`obk=sE>=`_
z96;I`M;Z+)ik8AQ9#z9ZPPcqh*GoIoc4m`XW!C2i0^3Gp)5MD$LoyukfFemjoFfc3
zj0gmTz!^2@CJ~LLYSy}i8Tvq#7QAa?*9;6R;!FjoX6I$<9%XDu4vUG^gVS7&RX+V%
zSKeF}juXkL6CZU-qH9RuXgbj3XgcR~Fs0EYnhpX@d4i<mfl|HH2FkjumWKpS6$FxW
zw#oSwovCKjH^eREMqG%cGaf0gAMBJ=k_iSN1AQ<yACRw1ZoF+4w{}raR>^~{Sg$-^
z(sIfoCom$cHNvL*jY`TMF&<;>Z8C;Bk4o7&glve$q}b?1G$z&J6}T^DYdw#txOtn}
zBtyj>v6|8$^h$x22OcCSw_)g%3C*UdXWrzdG_Rm;uTdy1Z?~>?nlwo&gTn_qSm^{H
zsjhQ80I#OWnnqt!w~YlZR8nROVXrjT3?<WPlC}O*q^oUHp;hLlq@`zPgFKQDY%X_&
zsTZ4rZK@%bt*4~BywDP52@Z!d<X^FM8O458vS{X_vn9EydP=&}4VKTbWv*_gGv?j_
zbRy@|fAVfiBRdC?sQ~TGTw$Ic3$_J-Jqj!DuwQUe-qerWOsh3V881OI!IfUBF@+Tk
zrAXJ@oc_qXAWp4Rc#O}SD|YC`+WGkMXXEYZ)sO#kxBbsDUteGUdlPQA&S9s{UY@oy
zWO3i&PxO8lN)~^&zFrIkBpm*PRh|BRn&do%Y-QSc=Ed3VQsK+t{`*t*)9n<%FBEc!
z%TO^a8`LNKPk#R1jgYJ4B}^J!Pz**5&-7*E@XeWGCewu!lFeQkGUxf9Gs1O>aUMKY
zpLtgW2vF%lRO1nC(Gt}pg^JrVC8prGw&IafqcRm^1<jd~`Aun+D*VR%$5RrTu=1Yw
z=sRaWI6r^i&%v)gZ;rSdhi_&8tbjq>&GpRZ@Z0KsdY8d2;M4tS^Y+jB>-EjeD8}mH
z`H6uA2F>^6vZH@1KY>39L2&v0^NBGh&S>`JLzk0Vbp7{ZlJ!Zwf%uTpn%kp<pAJfm
zP`_HdmP>@|PEar7JMLMp!#)xx@n{aFPziX=Z-FDd5#zR@wkLX#{2u`6_hy%;pWqE~
zws*+GwO-j?$5{G{z3h1OaC@xBd~3b8=FG7B3)|moT}9$})Vt$cnV^2d7}G;tWKi&f
z(c^6k3h@Wbwr+g?w^0GQeDC7YwXN&7>+|!+zQ}=0NgG4m2oSeO%JiTV#Po=f68;KW
zB(gwYLnR(14&U}(?Oj|zU$};@&0d(s?a=wj#_-vpfk4bILA=G`#JubNzc=q8!dVJr
zBD<8o;`uR~2Y-?&{#0i_4c}SwV(Ihw@ONwZVJcYA94@cV(!fTK|HgN~`A1nBzgoTd
zv^_n1?k-|<wSiFGo^G*QC(r_BYD=@~8HH~@w=!fH6WKcK-bHi?kP9tNj%>;vK{44a
zT2R_9S~!+Yr)ea)mgH6zDMx5LP6|c3=7IOxS=H|Qi2HMWk45X+b=Y6SSMk?qm?d_Y
zLUv`PYhv7DTcqh%xp32(mj<F_HExDStig}rE=uk%`^@J(J#lx!ZLjoY%+)|I4)Qy9
z+AZ(ZSMXy$L_BE~K<65AZp4EWH{%U*GDFJ)z0>eJes}bKss}Gc!zj}O(tS=64?ytw
z19do4$`(uf$9jPQ8I6=XO{|D^SZCNM(313tiw?Q=ymsU9_m5%tDj%~cTf;1oi2B`T
zV&O)m+|^Q<yR`~WYvqR0Ii<(anacJ81<ey5CRS8NHm`{YphhzLZt_ODb}e50Ya5;1
z`YGvO|C(ZjzLiHO$g}pdV`sAIavL6&wB&9id<^Vy)ZIcIJ<te7^K~GslPKZ#GFe76
z1B7~d#Mm$NVs4CuBbx(1N1_p@7z<rW&q`-YUCMyu6fBIvN~$A1N1{3RE}Xe>L`?eh
zQHiii_Klz-!0f1Tb`b)y%v`6aD)K*Cvq;L|9_VtiqPCzvMe^0JxW|o1#GM9sO*+>3
zVi5;5+@>5qGiT%;s@&&~VZstiN|zBlNpbiM+^8r`l37T^LJhkj6DzviGfh$GG%Wb)
zTgg#+Deg093H--=CALV_J*jv^%GEu~AS9w=^X@v`0F<JogGqcyq?(p0I|@-w)iu1L
zKgOc&Do7R$GZ7((LobRPWl6pnxhWNG7vaa-+P-u=f3?G4wVxGJEC{ST6bHGeRQyAH
zBfd18Ea~Sf6-Oc|X_3(eGRx=_O|6-HA}LcPq#R#S&<MV!q7Y`O`JZ*?!xDz%HH^z?
z8W&S50C;ZFs61p*xyZxvk*`2B{Dp5?(DEsXST8~jf)?h-dBXZ6c-TFWlKm|-FChU>
zeF*NsuuNS%B%T_4Eu)HvRVwSt3!I0yWmW{2WpB?*`V23W5#C%a-UvI+nae(ion$w~
zJHdx<HKa(DnpQ{s=X~?V*D*T{VB1aU-U<<B0QH;33LOHXzJ0SN{9Xf5DY3K&s~-mA
zJ+czg6Vi+h;Y6dSPn!Qvn*A1Yi?j@_x<9o~vHGo$UE+&xx4>B}w<U$>fjpE8>O>*5
za3Q=9ndVwVmTM_Rq5Ts+GLin&5vRpI7RD4qep@`vc<eqE7|=(KmI}yVPplLwA;o$d
z{9~$36eSifGtT#@2%)m;#bl;^jzHh^wgRa@VrD)~5&d=7`5=xVszx6aRJv+f=ODbu
zv+h%E`TGqFQEs|e7Z~%biS=jvr81m6$F|{gEy{2y<$v$LLHhil(+Aj?|JQi>53L)7
znd^rsvj$9_-t+lG?2d<a>HT^*bWq6A{C4*+j(0E#B=X2~$O!=XDWwerzfy;8sP$w@
z3c&b8=vkdX_+ufGub~ou$IRa4xz`(6IhIom)-sk(s)(7lhtLQy(};~8ZEk}Ct@qdY
zp<&-f>m%n!hLiprdx*lXFU4JNUw4oB<;2jH8SG!?DQ3E_fX%)yPiO=|OY@q({|JjW
z^ZT~_0dGD6G*TVk1aRcgc{s&S(zgqYwcFDH#tRq3uc>p-q+!^?!?$K_V?~c}Cet)x
z$H7X&nnaT*nGs6?s<1D%QR|ck6t_{p5OTi5^JXO!anY3)CRHW}(z-<uUdL{}d;RG&
zpVY;0awWxpXfn~SCMJ3-(T}<_AGR0=57h5C*OrH=v8L2R^f@*1L<P;e%sNSx#CF=$
zP*Wn}dY?A^3Y~!(w_1C_GrY_|W;lkkb4?v7cPn#7xb(IO#pq?<rzl*Zq%oVO9K&P}
z?ftD|Nrud}KhRHM$HA9WVANgAs(YCSb1H(pc>|&VtlmU2`lF_uXxKE`L`t~>ifbU9
zYjCRx=jAms*I+RM#yGOGFg(UZjV_Uhd=a0xB+MEKpOccRrtM80s@H%N1}Z@?`1=V2
zSE?Ed?ass<v(A(WCYZx0&ts%wiJedH<G?ZPvnm*!+m-k1rfgcRh+>;Y8jw;lmYHW<
z=(To$67H4Qx3V!87EZ^30@pOkiOWO}dN~&+Hsp!kv`0XWQx?obJFV^f5v*V!8Ft=E
zsyXJsMY1|DXSU+aQB?H3Rah-r2oe#;!OeS|l!;_pk-?(LNDTAd=)|b$;CUdjK~#5|
zdYlc^7`dPZR7(5LB#D-Io3<KW2LsJrQc^S^XB_M*^jnR>%%*fb>yq(d&dP&>a)wUE
zDa=Y?9=H^FAtH>wOw4J!em2Nm+<m|)q>`v!^IMp`C#6HSb$m#pslk(=>dzhwmoz*R
z9$}Uk2OU7D*O@}AP{{$#;Kg~B-Rn~MpJcImPI6Z@GgVvWZz+7ymWwfFp2RZNk0CUG
za8vxXePLhL{0lPGSwS^^T%KlUb@t~Ydpp!VX1kk3oo}(9RS+91yF5^y@Q{%<_Z>qx
zzVjrH^Gr#N)Ru7XkJ14zb<*cbe!->%?=#R@f=3pwJl9M0j%FfPHtjGELhJ@wsRy_e
z{I~E`&d>(P?O}>~yMPE8^Va9nsJt}bv|%Kvr(0FUXatQRmyP4OO7`%gN$iR3HjKkv
zvzcd6g+l8p%zri$$#_qPefcY?>I@O<*pGwyS6YaW;<l}6t&guDrB*7=<yD?%BqmVF
zg{OnmJk1y_Kf(*7PQrWX?-R|ss6gcwWVRz3>!xh;chDS@g*-`(c?(M}sc%z2&6f;N
zGM~Xss9JK+XuE21(Q5U<>OqZ87t6nc7MZi{T)nIl?)y%v-Z5SK=SH$=Cq+mU2CM<9
ztyIfIQ#W#L+V$jookDIjji|qvHT^}O(L2}SyqvX>zRt}P?2!)5`}$Fjq%uF33L5+J
zb<(~J#WL%jnc=F!%zK^bN)1VX-v<Vi*O~e*Ru&j`T@%Yt^<#;d`ZO&%lv(E5bb1q*
zhcC2|FDPOwi1Eqf?8Y>fqY3T&ZPJM}2vkMoigM=j+BV)<%@dxfEQwDUO=S|ylaD8r
zQ_f<eDb>l1T2%~j)0iKbILDvK#w?Q4IpKM4k!!Hf|I}GnA{)HmJrdRcFfXNLh-c4#
zTK*hN!_wwLE#h-DVrdFTQfiR6MoeKeXgxz2s_7U_p&KR0zaY(Kh990Lv>GHKMg4k;
z&hug0p3ZM#v?6=(Pf#hIky$xZopR)+c0P9EPamdcm6<a3@yj1G7GfNR&?4B_6L$Yi
z>X6Vpv6LuX8<PSwx}%T-Fs~f9v<|(xVZCxHhOLUUuV>K<0?EOG)mDJ!#%|9wmj{Uh
z{&2t-A)IVaCKM4iHX;G?EkSWk!OjhZvDBmZSN>Qh*XJ7FZx6n}Mkpm;Hf)DOn8c#X
zp2mwSQ>NH$M0C<2i{tWUYT<$o2lK}^n5mtQBiF29BgPL$&8uYz0DY>rAg)*HS&uc(
znR#sI&1a{XFp;%iZ(RxApo8--S-Wb&aHinnuk(@W%h?tFvpfTc2yoxBz&yEzhktDv
z$^5^d7QV%U^jocU?FJLy+y_kWg_>%f`9Z$?Nbw*G{meY+)`7SMrf(hT1aLSQ6C;I4
z0og)u%UG>sTvE_QKrUsig*xp0hibkq7-d%Ww^9DbT*&0uJls^*8A)uc@R=-7cfP)1
zkQwWva(9CS<N}Ks1fLyhTww}&MxRG038`lHRezkI*XFKVErg7M3&r-4xM&dm#P64~
z5*nBAkAR9F+f|*y`CqXamTBfzw3cnqP76Vn?0OQSaWolBfSF#Mt1-u>;ivmv<(mvu
z95Wy2((pd*c>R^VU01r`TF{K-ceOj$CHd3?Q=d}yjeD<<tpyuwKb`3vscc8G9<!<G
zf)?pw3T^VM;YmS|;|%5CkC}XXZjF0GzF4RoSt*EcePM?BGxJ`yjHS-JE3p>zS1bw!
z3eTshCDZ<X0NZ0vt*(#6U3V%p9=%HHT$U7-&s%m?mRq;m?xxI*6g7>rPajofe3Y{+
zW+`R}ay3a|*|L&d`3bpU4T@jbC^6^hr6R#baab@59cTJTSD8HbEyP*uio_zG<)J9V
zBZBU+HdXrrvPL5t(@eLjVyE4=Nb2O$uUa9%mBz0KKn0z}F`-7wnGQr+Sy#B-*rWZ9
zIQ)!oZi`VM0t7KGE}mt^oUI%hE)gj4`?#qUd2WRDOaK`y(Nv*J^S_=f|D)7XxvVjG
zX_n5&hFyluv4TLUn6O;@zF*9X!b}bj(Nvitt&%}-f1k|*s5J5XqL7(l(dGWeypt$(
zSuEn31A+^DxbrPDZSxh}?|kn~#UuhPNhDH%?heo-#zwL0qUV4)SEN7+Z9rOXF?Bc}
zRaeO3uL=rdc(c)m&bK9*K|8qjUi6EPYQ`Gl@rdb3G?pUjPY@l2Mh5(;ILS)?Fi!x0
zbCR~0=%!NlmazNc62EAk(}%V7BkX|;z|^5=0GNi`_b&a^Ms&^S>^Q&xtZr(n)!^Y3
zkvZ8@)57R0bSE}Xu>@s?Du!S6x{$WU1&>g-)GZ(2bPh%uJB-mf1vsBnm(n35WrP}Q
zYP_&K!*%$SHr_`@?y73GM9PuDF_LmSM;1y9TJRX1S;z4EMYt+*kFP0<M8&CSnO><h
z0Mz5Wc#52)A$sW`t?ZY!)x}WCNOa(46fVUES6Y#Kr>o{1QOVivW($9nhQ#HFfW`iR
zZYfOol)>vjWki&t*d40`4(wGT2UiUBhn8@DL3>TQSci*Rx)}vc`suddHn`>+0x?ur
z6KRG+ntAryDR7IvGDYgcWpD2mI~nC4z5fdxCk=|dWSILrJI$iPue{+&@B?!XB#!D0
z>Lz_WB761Lh-@{3kl(|>nLUPe;9HD4FlOG=2t@W!^6Mg7E5zC`AF`HB)+>1i(;M86
znhcs=OjY7x*taxK-@sK;I0%e2UQroY=?I3_(6Ygml|~kexqPDH|J)f$?|z;EtZ#>^
zyp!Q%<$whGmlh)MRf;p?nb?vh1G0}jSsqM)1RIxQ?Q+#xypNB@Lgn^aS(|>F+Te$*
zl2Z&KU-1RF3O~Shg9kbE+pvy;ccbUY$36ax-`!$YO=)jdUJioj8U{mz$MwTkq>=SO
z<*@ZdWqs$7@2Me^3zV-9FQ@#$b*}Mw$?9|?0E0wpK8F(c)%3K?rT+a5CJU=q@_&++
z9|a^G2kH-Nk*EQTJiQKzI~f2;xBYf_^`1IbqrsngG6HUn=Xg|hTZshY*@=v*G2$}}
z3enxop5-UT`>;$<uSB9xZw(6S;VeEzDhv*VP3#%!kv_yoU5h1UgZJ*Uj1Q{;EKZHc
z+*V;OLDLn|?nOdZN5TV4c!&5h$D>tHM!^hX`!G>fBU)jnsuRd)BL)!HO`2c(nS<cr
zvCd?0#Dch1#v-UK%q9gPGQOIJ-OzNnvkz7rmyC9R$Ex}r(zrz=x>9IGC#>F_>=PrO
zn!9V=T^%LV-YG_BHCWc%<1suVHe-sjfv}EMD$6K%vm<`4WYor_A*qvkwhfW=e(8b}
zT9l;j9mb)tEVxly3Itdfq`%8~9q&X)Lltbl*gxC8I-lfS2|3sR>GIK>he-(2!N^eN
zl{7Q}SiDqcBiXFndd6c>W+`df$U2Bm#@<OzyZq>Q7I#fj!^1>;BDa&eM~RtX>==nf
z7}=17pPUmb&=A6_u?^NW4$P3_hkTAEQzh@3U-Kl4w-(K@H2@mBW_{64TVgL;q3F&1
zIZ0L@+~;iXpQM>w)}fuvL;Y$qcwaS44l#%DdtgisEu%F+6tP|^yumxe8TlQT<jM2l
zhq$O7((tl*G&8Yty20M1oE*fCG@qK9$qVpSm!Jc1Gq(Z9-#B3HOFF5+EeOqJuk;s%
zFlWOF)3ZZnSAf2Pc|CV{*8n9d?gZtce|?Oq$86DWqG%U;aPR5!O(iqq+830XLKA{_
zKPAtN!i&JxCovcBhX~>+-<s|C-nxi)i)zI$zI_*O_n=}s;AG9w7BLeldR#B*z-kx|
z+fY|d%hD{nd@D5*#R4E~LuV}nX^oFX>0?4-+Un4)zX10}MmD7gWHtKJ$OcnQZ_!s^
zv3S?kSu+~p&jIxpTQ`+fEY>f@370{7DwIg+a6GT#xlU9>Y~2%l4$b?k;bHZu?lMJf
zm<r%XAp9H>6Vq<Bg{B2bois$X0mZZlKiyrDPeRKD31_K4a3>_P-U|r=6EzDZD`z)d
z;jqt8fq=SfA7K|Fw>@Mgy#OG2`RL!q*2Lvnk_cWt>w3zI=CyKH-Y$D-{=O)GDi!IW
zD&XpYsG1!AV51jTU?kUY#H%4-k7tZlSkU^#YBxY)GgX=?tk5E#TFzCau~(ONqdgcV
zZ@a?;OKoAf*cMPMpN#6^Gb6WEil?DXSt4~_QvtvP+HeZvaMc${sj`?z(?=HYgYu=h
z<mw8bgzd^lsIMVa3NcbjQ|+xq?WW)zl|9wcRr2Dxb&v6i{+2pb2YG05j8-c-PYwhz
zJ2q6sWfu3$B}I5vXc8Wg*r){AU8NZwCf8<2CzEqW89_HtP7f+(?<e1s^g>a9;fCfs
z>jB~mm1cfrc%qi&CA^j0eEB8|JrYB<abb%Tg8X}$7NVYT6hQok70N9uQO$Ygq4*Mh
zW{GPXs-IMLaY$ZiMs-BM5B-eO>pN?VHcBk7Lx6dmqKQ{&{~02u2m9qG4E&;*DX2|<
zU*V;i8W-h&;4CXT&L{sXS~aaX2|Up42UN^2=LAjV%sw36^(Sg7%(#NKx`Dr)6VJV0
z+l{jt<<jU6%r)$ku#JYUKUdDknY}ii#c%SEjo?5tpkF#fwhcZ&*9g!7PysZh`jq{~
zZR5+OTuV{wAjqoYT2vVPc1y0?6`9q8jQ04H!z$}KKZ~AR>1+M#JJ;ex2v_?W20$*J
zFILiL6!ofs2Js$jv6da^uZh>fSsRV(g))jOPvq-~=n8(OBOb&pNb%n+N4KehuT*l~
z+IhAdtJ#WCc%!}*dB)?afJ}A&E|i)Gv15+yf=xCg?s5==nVOF^`G2Jkb^xxJzSKI(
zt0ddTWz(WE_Joz;I-cb=nLnOKN`T|Anbz*kq*f0FUW4NIP3;_DCGJD6Oy@uyS^L>p
z(h1~N$8%X`wqdH|lwQy5`4&oZJR{cCfy)C&+<QhwK;30g1A*#Dud=(+8Z%aH7lWY;
z%1AMrS5OQI2v_uXSoq1mWVI5HTU5HiNh^;l3)$7k5c<%m$1eE=yRKQ+QGj2Y<3$#t
zEk)m7QslL4(x;TEBH6aHEnByjOuEL#*2R{3VQx??ei`9XWzY4xR!f1`NU63gZjRcz
zZRGfDaX`@}@ps9P$d{}^#xVWt@%!U)*=^N@Tia)MJT0GT>uenjqynX*#a&wF_bSS6
z^-OU#r_H5LToW<fjLb7xzPT>hF9@dvg}Mn=2{-c8)NN4b!KMWkKwT4Y=xF;}>O_?2
zW0gPrh1)+|vAlw+y?~9D`S0=~V6`4DHdktY{j7>n{H}FNK6t|Bq{Wg4CZg=Gx%-^&
zQ6Sywa^D5K8%(ipAe~p2^g99`G+;)?^hP;qlyng!?mnQcb`WH42uRSaA<my`dP8nh
zU`FP2q&o)GbWUvUE?^PP_9txaU{DZV3$6caJ2Cw*if<Wz2*w25)PV80Eq2777mZVh
zuzlrWFicj)?M(IrotO?VF0eKQ{-JMxWFoCHG1=22e`at^nkc$~8sB{*XWE~5A$SA!
zuJ&mM2#{bD{M5xin=eB15K`aNQVewP@>E8#*gO9406v*YIz66FLFpew_v7XwUb}+0
zmvtenf%W^5AYZ)Q6QH++DZ}SX(fdK~4l+-h`^L1Ty*%2xI{j+b0bolil1{w#%ChrT
zzS=(Bmwg_=4s3@>%UxK4{q?US^1sG%u>UcZdwsHd?fuu>u03*oO=YXEv{up@A-C3o
z7UAEp0(qzlrQ<y1LRr)d(V1*gV50fzlhM-nuSdL>;CC)yb0lJ=&OOFpVkmF=@*hqL
z?C}~DVaij!fk4>x^eO|J?S{3H$SIyu-)Qx?=+%D8obxre5y)Uf2*eru5p?IJaW}{j
zk8^ee&NxTJF7$P2O$~E9sjiQa>M)W?AT{&cJj&%EN-;I%DV8mEs#lg{$Q$bB6?p}#
zcfHBc={-0=F7EX5YLy|B-lb*$s@gXG37yG@LVBankr9V0wrZ_OwwY5~XV}2Fwxj|7
z+rZVp1CNra4nnTQe*t<!bP*@U!LfSzL0Q?FfTxTsagw@uu~ecFq`{O)TiVY<M`c%S
zGBGK_cDZr*zJ{f8-FuHi^L^!xxzS}dCh<<rDgFz<eAmICkY*`h1oF&r5}HU#fJt=S
z4*lq27LSU`aA?Jk*|M{+4PN<RtM%b*JHmL#h)HDS>~x=C@#}}9yDod03Wry#Qrwg{
z6cbo*lx;bEWTB@XZS^W*Zvb4qB)WQ_`xM!gm>ya^Z(8z2*KrvPDdaKJjccoQ%KE81
z8%hOu0Au*7v*K1!p4Z&R<CRoKic#d$BOX;WK>6D}q+5}6<b0Zdmn;cqN=W<ro(MW}
z0p{#ax#Af71qfOb&OhPMl4Idtjjz{_dT!xmr5r4TN>9HURQUWwuuC;m<8@HQ_8Ksr
z#toX`S18I>U+JvC^^R=xt=TT1F9_Osr8>p{6Bo~s&QH9J`wVIM$ZQ1_h1p`YBi7eM
zaXtYdw++;yhCPC{KAS3N7F}ZQ8Ea(jV`gdTjcO&LIZ@5Fd7=sunqyDG3M}g;4U8j5
z>E)|RJ5&n7wR!@;64#V%1z>U^SZ$*UxdkC-Mq?ng-g9-{-TpxrWhm`7L0M*85N*!@
zuBK^I92yjQ0G;a*MTPscCC?LOwR<#bR-*Z9APT?nGfdlJ*lj=m80sq4IxQ9V?>K0M
zl#2Di{_OPigM`R<D=+FHj*1yV|A=ZI1ThYJ{UHxH)3psWig=*~$U(5h7hm38IG{v#
za8f4zW6P0O{`(K}k&LZt%*p0YwloVsHJw4hhMtL6bVq}4BppkcGAZK{SA6v3!EY1e
zwv99Vb5Z);YI)#|;>EJtduFH<l#P(=G^y!3LHcBJ3eZ)QUUhv6b)~Q%GAo&lw+&S#
zCi|Rz91W$TBF8T!h~$yQ(Q*FAF{x=Uud(P!Ja}R&zSbiZh!4$Fv1Iko%yI&Nn%TUt
z2-(|T&hX0y{)UNj@!p0BX7G^4*TzeIH#8+sJ`gOL2H7xiL6@p{DYNT`nGTb-WKm~x
zr>m{RO0U5V5}kZiTC?txcc@nUcHY*Nu6#pmvrv{*Tw3ca*^t|^D6KEfXY8BIn8acR
z5`~N1IogMR2~wH8X!a~h$y`2wg#qJ>sbF|?E9FE@zH>vRMfTOzkoLk}@)<5UlZsgZ
zP20p=$A-0zd!?lESmbLLN`088z@Sxl-@+X$F&(_ao7wFbc{CnE4eJf<e;RB>^xVlP
zjj~l>!!%snj-J~|Vf&M!XYP~SVL7Cp$Mde&?Sj|i-u!ej<>j7qab4DcEa;Ynahoc^
zV@Wi=E!9O<6#G@%hU+3TRBmY9i=mqCuAVrm0$r;YLOd12(A2~8-JzHBy}VNWA#9QC
zaQJy9AQ|sfpIfE3(iCLKHx)?D@TWlNl~~`x5HOKsq3y_)YI;f%*jmu5J*if~8Th_q
zpDMG#`C>1(C~G0Usi{Mqo^tIDwR-J>vMR89YPqUBv|5*C^E-0X?BptE+t9P`Ak>=6
zPtr24I9j}JZ!+MG_596`mP@@N7ma7~bQh@1aQi4Z!-(a?E~tjmD1$#(Z=1VP`d=Wc
zuQr1JE!IG|*jWB&<Ag}@LzH06!uXGkla7P#D4hRlJ?4MZ6(_LaBJkyqz(>mEK*&oG
zDh0KL5L)Zx@_4mGwQw)HFNZnX^Ap-6G*)C_?1YhjX<uN+_m4^bW+>Bel?c9024?W6
z0jIA}K@B^Xvr<ZxTQgq2gxqd2B)_&QRY=|PP2j3lsZ>eO@fgTIfWVplwFjVibUaI}
zAXjKofR&tVNNrUZ=Pn7ZoQT<sQO$OvGuL8ik9CxiBxDJvE#()WHl@|&o+Q{@ol`+H
zSW^^Q9nfb=@6kT!!A-FX6oR9l%j`|oFv__6)0!^OYVAi`XoID3s`aD#FfZ;+*|lq?
zk`p;w84#gEH9j|mA+z~gYY3ohWm&?GYNLlyr;J@g`(q)zmEM3FB4g2jELuP*YN#le
zGstHKa=fc_kuRN)6|Dj--R1o%8gnZt4VB-X6`x}(wAg*s5M)lD+fq>K;Kjli@zg~J
zR=Y%Ig}P*J)~I%l;E^N}D&g7$Dmm8<Z>(N-ZY?xR-R#kVU6uCF1|48+_pps_?cyTT
z1Zwf8j7w?&(!dd!UbCY5B*n4MijXM0pL`(XxS|sxv%l!e3976^Am!GJ?z0dJmi*yC
z$h{9_ypZfdW<Whw#ez_?)zKor`!>Iq_8HHI++N%nq$5YZYlW}-cC>bQp;o>ZZkRh>
zCV2_1{RCeL$&B!m3Jrh+tn8hlM)8RSt^u*`!R%zPu%aZ>8}yx)>7FaaldA_gBwVhS
zoYX1CTZZV(5?`%mc5=5HYyR%}y3mb*m7TobC6VvP{`uOs8hsYs+<a{9{d_V!QQzz1
z^}NgI*TMU3<@Ci%G+c6c=CIhls4;8R`)cIW>79o&YK-qdMhuus@UW(h0Hr5885&*0
z0E!2&`!#<!arxo#&!@o*)-<Y86F204{B!x}q8_?-GrASLa3k_UjkjzXtBX8>d=Y}C
zoPT{ijly2UI6Nz_PM>5Mdx$6Keactx*Q(Vw_%O@z_wRT+o{skH{IGj_ns%rc__Vrn
zfBQ0hn%p(0?*;Jb(9L|iICn<$^=9dJYxidCAiGAh>gD9iQ&cbio5XNXkB0I6D8ES{
zUo77>dxcx#_BN=;!A}1zmNEWdhB}+f$j;1~WFA#@l{V6I44>@e>acam$fn7A)^o#X
z6*~^%(DO%?i`awz0G-6qFM8bF?(Qp!YV@+wv^M=se-$wLxOHGnwlCA#>wmpoA(Tu|
z2Kc(zr39t?g*vm-?d{f`4N^&ueV?KkJr?5Qh~BEXppRfyg@olqpnf!AX{8gu=-2Ao
z$_seLeulg>fLv5rGj#GtzxVyO>QU{GS1Yd`E%*>iG1-iU?zI3zd0tw7Ka}i8&udWN
zn`DTm>kIh$_xR}R^=8}X&CmDwHvV{cuzl`1$(NV+(H#K0mfXB777p&G#W_5DKmG6Q
zz&MugW9KD0Uq8>c!*{huQYp`Ra@2E>tGBC;Xl8})+Bmkq6H;Q<0`}x!7SUGZtl+%B
zR?$wx3L6!U8X*dTZDp=8ERL<E-pO}#NTry&Vh~`kB+!Q1V-AEZzLEHzQu=F=v~_|M
zEcNrcBR+C3JHT^{3wmy-Rcw#W&{}+Xd^JSylTJ`-Xru&+=~s=@FDmIca0?;!*eF@0
zCQ(#-(YT?(XU4K$))DYhxBm}%K!v|F><SQgrC0%(fYzd9xh{mm9ab$R6{K%cI)EjB
z(xW4vjSzTle}uDtNCvbVMF3+79g*fN0Eu@hMTM{dTPeHXK0Rj6CKTCE@!QG)&;~Q&
ztaxu#qc?xtC<4}@!Kx@!{4i((?h4-q#4*KR^n9AIIIbIfUUv9;#U}u|f=rPs@P9pk
zue(2Gv@zMkUcGCh*gI4Y`;;HX*9Z7364{%G4fu0se@}AnxC^EP=*+{}(M1CPJtDOB
zq6=q77$3sf0jaaMoKO<oion~Fbo2l_Kai`h4Q)s&zOUIJ3K~&3IzN=7Lxk#MANYCk
zGM8gQBMmo)tKTS{b{tk|3x_FinC^s&SJ-B_rT|xauoXg=2{g0^D+(lHxZXPe^tOz0
zyE_^*f4+_ejrKu<&E!2aXu5?4b)Y;QD6bBrRtFlagRc%gOLxa{n?wc=jy@Mz=P@;c
zv(dfhF?G~bFDC@5DBAXzX-s<>t9NW41vyK(4!s2ZB#^*kDrUtMvpXur6*<`p-e#m@
zC_GIbQ?W*lvJ%}*F>LwOii*ttxD}O?x?*%kf5iwPl|2z|Ml@*Yt3IY))b+~cqngk_
zq2_9Fd2d2v`S&I?TO6b00%?24z6Yxyg(55O!^#zP-}Vv!ja)Qt`>+C0qtBCvE#|OI
z&P3}Ags-1o%oo|GpXnB5o>J2H%X~k}xJ8-o=RB8X-bNv}QOIo+@;1ggZH#r=80)k#
zf7WSZtkcF=rwy>s#u{y`(bj9w+C1yCo$IjS2Bob^w5<vtX#4q5D*$tkymE*>JHwgH
z{DctNQ4Ne$y#a9aEkX~|T{2r_AFgeLu!<JogEH(lNx3O7EG44r@cJ%(Hry=wNgA5n
zSax*TXenwnb3n_bB9Ln4!wG?srGl29e==Hr2O$tfccIggD+7~eSP=iHro+hqcSrvA
zhTkV^jIEJfGe+c#QsD&(aIBn)+TlcCMdkLh_&(VRT-0Yl*M-QrF@hg}vnCd}a6X_N
z5D@10NDKA?4#5bJ>m4yNMV3|tMGSU;P4eOFkWuUXJlAJo)!hQ_V7x0XAc7_|f8=Zm
z=Y=%+a6%M{<g>7~Lu5&4psr7TDlj(;kT{u1BPo<`)iEK#M0nL$w;(VH<zln$jnnVR
z`MZ~I<MRCH>Ey-5<mc3gDU(B-jLsH9T8d0IZ;!wK(i$Yc?kZLh-&K(^x^<HB;WCGx
zN1|mJyX5DjWOYaSs#)A)k;_}Ie-QP6Ih3SEf0rR6esi;aXqwCQ!&$2Cs^YZ-p<~Ko
zbhp#l?Db>obT+qG=rK=clk{UuOIJX0e?R8;hg0vdeP}qflg%4&B;&nU5m7>Wu~G)~
zrS_1}_r{oZhQe1dE!pibZO#MJ_IG$Th>|xH4U|bVsKG~<d}UBFC>_)Re-0?AX+$fq
zo@?b=Jmfw&*OFSLFn&-c0A__tyW3ogYwWL5ZMxIKQB)s4^7IfOd2L%Mtz*NDaRqs<
zrgHDb72@dc1&$5`O6P1JmvM`a$VG;6i@(FAdhf>NmLhMxvIS|Z(Z^+5KVNDcnasBJ
zHwjX1_io$lZ%b{@82xQ2f8meIxYS?67`7$vU0c3)+oJo8mdX~^(a7l!mT^0dpx=3A
z>k7w!IqU}|`)~@HWahjNE1Kl!qoh4pvtoVOO8_Et5_Hd)=*BRJcZ~M<mZLqkEf|ys
zsL2D=<N<2(05!P|$_6#Yb3A|(58%WDI58g%5M?wUsa+x>I2PZ3Wq6|XjsD*pM$Ptl
zQmG#qAVTZn6Y>FwNMhRY#=@bDTfF@+4#jgUg`yhW_57W@LhW8h0ylUs3l3&-*@TvY
z{3PWgLlo#4%`T&grNxx)zt3QQ=`HKl8}>O;{sV>8{{e%=<1LqAU^)~5FqgsPHxri!
zF*?3~I353b`~AR?^)F}R_xQF>GvJQ$Z6sVmsFyS68fMk;nS+&d;wk-tr#jya@MCOR
z^UK+b-L;HiR#zPXvvn2?x>9jnm4C}Q@__H+3Lv=-nLhFpp9z7y#c@22Jw`PqhH#jc
zghb#Jomi0vWPaVi(g_xX9hiSP+h`;gRThqaMwN9%n>Vge)&(Ocqp)D3UE=L^`1RY{
z;on2mm{iQzFj?W50;T@_^AKNX0X(=cNeL|J>DNEg_ro9KtZ4Z0o_sjjhg*=PvyZgQ
zfSXvUnT0#rr`PQ*iiPVGX|i1u(F@1>*J57Oj|PtFpI<+w;mNnxsv@Zc)6$Jhi?8v2
z>O!E62rMW&j)-BWO>|;(Xw%#*{-$oZ&O1Dq*X&zEa05Ef$i{o7gIa{(irP)|Ivc%B
z>-e93Cj1D(U*!K>Xp>~VNttGHb$m2ssPht|E;K@dGIOHo>~ZE(T@!;k2l(Qtr`#tF
zieojdUvNcN=GBWC6!PDmCSO9@>tv~a^O&QIewyY$u1r9#DpeZS%Z`4L-2a^XU}0eg
z0j64D%_L|%>d4|O250gWH5RiO&n!0;Mv1>9tt3=3_%zOyr8qvNC^X1_qDnT?m3K@#
z$+yyfowwvC#al8ZNf#&8hkupwWl4R7Zy_h~Y5A3sf;ndSz*_RTSciCEr^AYW(KEqj
z!a=qsi7!EARYTmEk(w9E2y=0bc*RkC-nO3RfZ>W-FUe4OE%wM3L|l;o5;9^STwWQK
zv2pSA+v|1`Hi2<KnPGi-0mOIG{Uv;p(lNJ`INLPE)V*Yr+N2_8N47nGyR3tpz;086
zf^gN|kIbQaY}RMRT_D#xUA;wr5WMrdbxWOqYvy??@Mer(x4>h!iO1<pMZP874ACzM
zQf7LcR?5-p2sXQ#znosfAXji6y?@i{4j%G_gdZ<kcJ&&~DcraykBob_s?w5_BQpis
zGguSIlSMfzs;)+_Pb2}k$ok8ud{kdCb5R3;r+qH;D^9QmSUO11f?TM7cLnleFJrC4
zqAKwndzlaxW=8qao62kbRrQxgmI6(-RMy_4(Iv1E{1+utp1JfJm8UM{yQZUL$C!3P
zN_l~RvL9!SE>x|zAwhmmY*o@xDbuQk)-_w_=}ID`_LL%pRio9-=CwQ&nTL(^1i~>N
zT$i3m(VMKNQJQm3iM$qn4r$F&hKX8=bWo8&A#pU1MZzL}@~jI;Dr>IO@9MEq^rc0L
zrZ$6Xr9q`dzo=X|($;|ypuFQu8S5?FZQwGevDG(wX}Z|KLXT%F@RxGVlwR1@cpclE
z(rs%kS)1S<&32zBQ%-zGxv*I#7|WtkI(|uit=v%<jsB7_)ZgBJP-||Wq5h<PTE{&h
zV?S1E<>K-~4W@Di8cf+5-s6*0Tgg7Zxs0rplHecC+UJwI6nGLMi7zF&QKMy?c8e(5
zO#2X3TG^2U7@bk*MDn{H8gj%psT9G;n1vSlDmvXkFsV$}{^Zgtt>HT>SW;Q27CG8g
ziUF;Mwa4<QNwSuIH=sL)7Psb&9+lnYy5Uhcs-QKn{3AH3q+!xO3P(-rEpBT~-i@~k
zDpvt5MJLyz@K$=yz)+uqw=!rZgZs4h?+(x`V>h9CXcpofdskA(YvDRmmN0fr$m-7&
ztP*tx5yKuEQc&>9E4+*I<X~7(8p_j5-jPzCN^uPj>rIM(@^wGS(RtIa>h|4g_HM}x
zFq?S*vC;OI-0)gW&DmXXsypUDD}CB>9~gfidP5t@L<k>1Zz%%v&i*gT;8mq>pF0`m
zg%|{Q6iy2d#kVk@1k?huf`3dIcM}Q=Bbk%kiBLk9q+VHe&IPW_%bT(nE{!RhEE1h`
zWhWY`HZ~J~^4qSM&h&jTy;&fe0HJ9Yl02Y%kPZqeH?DNeR!l@*&E<1G4o_0eo%Gja
zu}s;GE1PL$XE|k;s=c_igrhueJggS$)v7d`7&B}Wm9v_x)zwm10BAW^y_**Nw{BUG
zBA@Q0)-#%|LLX<ZP0CSKIqYg_UyAjaZjNkk$(U|`Tcy*J9ZyuY&zs0z?>+LX3Ng_v
z>~N=j;Y*kT27<U47vI=z69BNj?a`=XZtiwa1U+bLOR}As5nF7WhYJN61{#6jp>r!A
zHk`N_#N5cx&{t{i`oRfrVoHr^HhuQu+ovy4$MwCyk6O-nZe|e^@SVz^43tOdTNjPw
zXZ$9A#$8CCiU|=0{y-DkF&Or~;kci7LJ<XJ<-piyIj0aTa4{}k&gpAy(H%MD7@fzH
z;Jjr@H+%icHa{ZKp8rAVVZx2>N5`V)T<>Y=Q>s%twN<&<+jT1i#jG>^x@%rVBd|l&
z8l_ps@7T091X*>ff(R-vaauJ#x<jLSdcEX8YyYc+T}xWatiIKh-Qf4zG9XR)JnjNK
zNZ?reVbMK!Sbk7JldvY8Nm7*S&Agt<mmgg0xB=B=xkD}gAG2pVm$B3>69X|ZFqg5y
zITQgjm%+6NCznY%It~JMRhL;gIx~L}aInIWvU+&?HYk1}`LB{ZX=t!Y2?J_)`#$`>
zl~Nmi_WYof0Q^|~?0BTzCs2-8D3c$x`(#mV$F=%U)&Qkl+LzsVRGEU~mDX7q&O7h}
zVch-k_UlGw*2`G=1rpBUU@QveD)YOSG}-lXbn2wENCKVhc1K=7XTAcj#~^=2o@dEX
zr}?<1@NXl<CB5pPPcx&WJ_p@c{)DV8vPDzK6w&2b)4n05Nh-XpYfYI|uxE|oA<x;_
z3YSkF-_YbFKdy^3l1x(Y1h_#8Ama>=#!Ef?VBMa6^Yx6zH3%|(Fr)-~NYMi|91<mt
znXlG{y@#@Q3Ztl;z=CHT^0-P*wjoI0gYAMm@m-<d)A~dy_-(Rp)vw>FVu-5H<j|1}
z<>T_ZFcLH@9LA#{??sG9hL2HBGg3rb<z4b104ub?PVU#EDINrNr<c__IwF5X!dO3O
zExf|v$6v#@&2RmwU5Lsy+eB}W#t==t!3ZKZ7^#B5ZTyXP_=pjb!e@54XoS;}Ehf#P
zG-CxAsA2><oDM<LHYK{~gkh9rHq!O=8)R-xdZJ)d(m`EST=$qmI?7Sr)*ue4S)1(`
z5>+-$&S@47`EfvyLU}fOCX9cz0ihkuByJ2xS=%w#FdHWa;Y9<R8Yxp_by<BOQmkj@
zX6C3T6{fOs<H@#$%O^zI<~+a8Y@4FW%V$(G99pB0Z1TCvxjDTFC+W#9m0d9)po|jF
z9E}*aq^pvd#^+th2lQIZ-nu|`jFj)OetONceO{yH=yJ}a6`ih)%;JBtBc@fZ;)!~K
zUx${X#+?gx9#0`<<y_{#YV3^mGJ4kJ`EukGbuR2VGkee#iTp06;S(WxF2C2)G>0IW
zdK&DYB#OnfknFvb7P76Qj&%ez7b}mG=$h*(^j09^LOr!c3MsYMlv1pvVqI10tV5cM
zpU>81SLFt+i~!%u*VKPuQ}23z1fb1Hed^&epT0g)u8D#Rb>LY`J00ta2UvPSX>H4@
zU{Qvoj>>wLy)C!h%W8~m($OQVHlY;j?AX}_>*W6J9&xH!PINu>Xibg1CKGmEYt0dn
zgT)SQ@pW}XHfVhfM{{hE$FOSVm+7QZkRBUErpo7QLAyHBW|x0N8BZ3DwbsB=Y-U7=
zyL~MYXFkC7hN^{BWw^Git0c}+#HfYl)a??TeH!EX0(BEux1|bhg_EcGv{!`S9ok<6
zEtIk!Xn#>@VaF11jrMm6y|Ds+XH+jnvkA9J9nVecda7?2OXb_USg<DEx;&<f1x6mf
zUM0g7{xJTD?lym!`Pw*zXBT8t=r~ZHm+L(5gmcjEc-GDv(6vQ#?YvWbA5ME#j<Ti+
ziJf>4)NA$3p+GUp%y|9zD!<fEBS#U-)-jHg3yr2*I-mMj9^u_eL9$#7qCjgpV*iIK
z{|;aUibqKw04vSrkbd2}+KHn?*S{5<kWopij>!k06DNNUA?Z<ftlSd`XnxVTBN|x<
zKGst_hel2%Kk0Z}Ej7lvS8X4JMp}DrUf?OvsOR-*(5PelMl{k~l7g`}M#Z8dSmM)d
zfBLiEMce^f0#G}zKY%Sk*XZ=BjnL$dBQ-*9)%rwRC8WcS`3GQ*0xuBNgD@xDA6Vtq
zzB_tXXOw^XoQ{(-SftsnfIYI?u})%4zbPLc*cmCMgnkYv>R5dmB%1PlBM^BImT=`r
zSxi3?golo1L??nC8RE><=#>5ep}kv8C%&f-N)&l-D%h^!YT<s>=-Ixr_p6Ccd%v32
zw|!lr+~2QO4&(hQow;A-w|ahwdg@nRE{d)9vs)Y26Ze+?-TT#_R91vO;$8n;)Q=3F
z{{joI=eU<)U^)~6FgTYX+&&bSJ(WBsm(XP_1eaewI+A}=P;mI`{r3l%SpRY|uJZ!5
zmNdsVSrp7&=6m2O+z*d@3&g=64DucM0K-h+vHy(W^NBBJ%1}Hw9_WPemSFZkBLyj4
z6y*=-`ECLQJ`MSBWBU_~?w=LPPOX`MKntxXQ)W%}{03@l3JBDo|E`aTvQ4}PIee0d
z`e#7}@w<Obmp%k58Na56#OS=QXtV;+HH<!=DA?Qf$}n(78SF37zQR*5ak#xT-(KGz
z{(UIg(I5=@&Uh5wQ1|zbhwuX`BgpzrYvC1kKmX~zJ^ay9O+M7#V>m=xNFrk@TWAu7
z9e|_~8j6)`J)Tn{8;<S8U!%~NY=oyIIPy{D&(DAFUGwDYTajH$5Q?3<;YR%yt}0B*
zju6I3?GBjm!p1r!-d_{X$Nl^n6w2>mkF+2^*#T$NncY3JZ(innsAP<VOqU&r4BOq6
z0LDY~-QK*}lM!<cGn7IK6F-s9^6PpuC2AoCj4pdpK=;4LCo|@er}C>g=n%P4a+{*d
z^jv>Otz_*y*WBAp{Dl*+((gf0*!faob)rGe0}9MYgS|DhaZU#(;md)6-gzcZzrHP}
zLQukbH$7npzjMA4YD>&DHtjq0SA6z-$ckiB*JWlQ<pBt<sbvVqCSddEr*uGY41%oE
zkQfO^CeNnqnF&fCuScKPSNUda-kzF2TgiVio#eQ&RusjQUrlm?{+k_=^yuj^bBV3%
zmJ>{)jpD2-p9U$A;{III3XSOce&&o>=K~{DiQa`IyXaqByp4jbL>3YWLk@*4C#9~j
zHo&p3gCR$JCogp$3A4YfMMD-@S;4ipuuI8Mx$sj_R-11#v_Mbn#`vkwVuXVhV$FYO
zp=`_S2f+(8UJxU`kAWA+HOS+o=ps14%@Bqa;B~{$D!|}{&GJp?g3B3K{aCX0b4lzG
zTsh8i8DffeOBw4<Pqcaxqs_}X+h}V9Qmcv~9GZGDZ(!}5&|EKSB9}lbuLZjMO3Zc>
z-c79OlNBigmzdSJNDslim1zi?*Gqp8o|Gi7KzXRWu+}V~yww_kj7^DJ3dwC3&qc@f
znw-0g9;fJxyVa+KVt#iN+2f)GAX$L!)GP9@xq@#5#S_qZ9EyvfMc|3t-8h6h0?)|_
zpA9@4hJFFhPXnIdy>RYY<hhl~I`nL)+!1?%Q(St@71$G~nk?Q4dm@eJZ7F|8nY1)J
zvfGNWR*e9XjbO16?1ONsc5F*-L>rvS0EvsvtzjwvJIc^_v^e#0otJVm9h}v%hb%P<
z28z-7*{IoruM{EYRohZtLB7gDh@N)GM({f;m2|>tUtt9|K+QeU7!m-!rMM<dmcQPi
zlRYNpOy6M^&!p!A@2w+gZyA3TsVnfXVpnuZx0G&-h!uDNTD7M@#!FBoh;wjQ5yz^B
zmRiKavd?__0!>&eNKJUm38l3&=kJh(TZt~f<E<Ab57LgJWk&IEtz<RDbCT7o^@gN0
zt~Fx`ue3~_Zg1m1H?^()F0)K-u7kA4<f$$roHQ<=0sYHk=&!Eqv9y1w%EM%m?&Q$5
ztYahA8>5qk-fnE<W8c<Li7kv%YrU9LTbdTP4`vj)xohk9n<C3&!Q&ZYkJqp13%J+x
zXeRoqbMPX3ArPFG+ugBK^me6FbaU?;ij508cWUvCLRp`2cagZ>y~7qGUI%<%x~z%E
zg5wCSUhDN-j(hC7t6qPUq+eee--X?qveBN^b(6j@3E0$t*YkwIGaIiQuU+*@?*t&W
zI^O7&sI)M_E9>10g8;21<J|$Z1jG{l>i|mraYN@|vQ##CIl@?LxnG8;1-}N{%L#sz
zMNZZJ5njUA{IqDg-@W9wW6MGrzYKhh_qe>~tgU+vo)i5j_050ACg6-fMXf#d4YW7S
zLl99|k8>xL2!=Wav&AloEGV(RkVzOfJ2}iaODx%2#8T)rj8l_Sw#Ab2!HQEJ@8bhn
zdiDWmPrIa9G#h7Q)-bNt?TL4sf&`7a5rW@81yPnyNB94lg7m2A%rJzph*XVyA2bXh
zD-5&M%t9Pj5VC(aABKe9&=7v$Fr<IaH#<5|nuV0+4g-;f@`iy(v*U_^h*CEG_tGTY
z6*t@`QQv+yRBt^pATJbWk>-@MgOtnPvgx32cEna&ulpNT2(P!@L_??L*T1L)YP(z7
zvty&uicZ2#k|1YqFgj`|-Dz~xoO0#psG(MDd<CB=*1T*HR=XR#^vwZK6TEiF<%pZq
zhUU^Bh>J5=oRetx8|RXmZlPkwkd^!mRa_Iijf!iUXH;A-{>)LERf6H)HtzF(cFfpg
z_n>VA?5j~*EaL5yEf(-HX`7b4l;XIKx`HiNCvxqbQPF?laImeHVPHBH0Wg<g)-Ds5
zLYWOHmmx<wlYf`=*ZE%`e}Cc$^@68BrnEb0BN<^+@RVmT^ec_O&d-uNx>IzgM1b?-
zYGF#ZLdKow;Squ~Gr=QF<uyUs;hu5KNi#qp+N8gvhY)`iaZZA#RJr_welKa1N+YeG
zokw~Y9vcM_W}r)@8JGAxU}*~MNO_*GG8G#f`Pz*=;(roK!-T;peKtYRha5t!8Q6i(
z{7P6Lz68DquRR5x4bOu91bi<Q-bK3d&Ce)w<}oI5RQH?=GIcfOErdH|QsE3d&_C#@
z7K{_!iQWL)RMH9-6P`kTB6`f?m?PTaoJ`vugas{1l1wm`^sbi6&dSGJJwJyrx{S`c
z8*$jv-GA9wn_zq1ycK+kmbSb+)OhKTIkdg}?$CGzf?2712*FP}^%8w);8W0vx%Mc|
zKon~Uc<X32Xl;7pWsXMX_r~9;1&b>9?|4Qln1UXK1p_%^iV%i5CkMhd12{rSNW~1-
z!|Zj%KE%x#f<EBJug9*DBNOWLD3Kn#Csy)f4}WR|BgZ*aiL3y%=~XI`-G)7b_M_hX
zTdZ70QkZe}5*hc5uH40;@?`@+%fhQQ?OS9e7|XcE0eW&K7(<C2#}h-Eg-ufkTE*yL
zmBxiE<192V&;vApx#JKtdW(*T6m=<t<^3l!GZSf&XDI{*WylH^p87e#k3bT4+KnQE
z>VHus=D9ZWRpp`IJ7(rFx}Dvr3@3B5<WD)1bXin*`jcYB@#(L>?2E=(CY<ir6c#n<
z68D5%ar~&AXLxCLx;moW6SM@$gv1WFjln^LY16DXyZENouB02Bk;G^#;|!Iyntrw8
zA~J3I=74cUzqllAw6#tL3~i!b59D%2{D1aK(ocW){_*tRL4_4%BHYN+zyEmhFC-|Y
zjXTk<L@)pR?ezQ8KYE~bniKHnS}WD!ubEQzq%1NZU;pirKj-*j+!)Ea-dbaoM5$3H
zMV=j`x{-U3X7khbCy1|k@@thPX-Tn%=^?}4{58+340<K#XW)Pfd6Q(5@;yFF8-L_L
zm3LUARD;{QpBkkZqU#|ZQIqL!XNV+|L8R#(U7|9-7jX<$M%7cB6^gcHQ3VZ``>90o
zTOfK!6Dqx#(sa&)*Nae_C&7RnpBex<h64IkFvo|u<EwVeNs|k4*h5qAiTW?#nYD=U
z6`P}(OK1^SM<y?Ni0VCRIf_uRxPMz}mh##GX!g9lS~~Ky<kDtD9Yi&viQ#<w#<}+<
zdxl4G2Xa@N@Mnq3hSx&eqb@5$J6bG;!ZRwrC>$YyNm9!-{C#$zNq2hw?KuSz10B4X
z5qU|axt>Nmm#i#-;BW25%`bBSXI}$`e09c8^Te0gR=67hr%hH`Niws$Wq&krM;T3W
z)x*2DjAluKpnE73M?Fg{<(bj^^V?hi!R1m8-d$e9ro4uS<uzO_uaRpK#tz(Tx6Iwd
zkYX7*FfCNplsTvam|0~_43#y!Sy`Jge}J%9I2aU^;t@^JaGh6I^z`*K3<+*sPwTQ7
zTe6J3#5nRLC&uw^oV}VDFMp4&rLc$dp8BV^1!udc{?qJhH|;fR%IqbE9DB(vDzl{h
zR9e-~-br(%B9|>;$a95W$#cat&lP4R&y6e0@?2qNd2SeAI2~v2>2j?JWu0cE*9f6j
zr5rF)uem-L(#7w}q|s8AM)^V~i9VrPwEmm9F>TRY&lW4@UKkxsr+*RGy=E9idD$(`
zN{YC2mEwGF{m7?e%EGY9nj!gEQ>YQg$Ebnej&^SBSNlv}ofE_xsvwM_?N0P7KCtE?
zDg4q_-cj@Do2!yrN|T_D%!|>##fZ`;V0AQ{H@C2YtDK5&Ib|-U#4SaZIY7+?#=!Vb
z7dT?DdeelDBNVBb0e{=k|9aKgR)jmNP><eqxs_z^tRtZ$k!#)8k;sJ@2gik^{&%6U
zgvPP9*1JjPE224<a^F(nvGuMybtlx884VApJ9Us3>Q3E^FVvk-kp}W20@U|YAqfKZ
zfWEsBlXB}G-I67)7hy6BTS%n{a|hwZFbyzJe%%*eEPr$pZhzL<rspEsI-BOx+}FDj
zyTyfHsIzsD9#m)RRtXWRNV<JJDv&QG@CcG|x7m$4OS#+grCdzAm{1=E9vX;4wue2z
zeLs<x6b=j!rdi}9>XqRnPzY7tHd@O6>bUKu1!1rHweKuN2nwG-iUG=@L(iDseQqMo
zl7;dc=OxmpTz`Zxhee2PEVL`tXCq9L$8YmmceKZMg_0?+@tsB<H{FXKbHM<&z|q!W
zYLWZ1<g#jpNv>|?QDvm8Y2$sfJVLt3_c-rCAv8q}O38r}k8VvZMce4&@vRQIQhy7K
z32@zcE19%H`y{(_Vtzx|xjy$u5oKi!&cVV@$a26^IDaXJ6?e)Zf+7Jhwl9YWfQ531
zKshOgq`5cBAp*osIV8R5$|3RT${7j@C>8XCeP39;HR(SQR>zj^bZdcBfz@23^RMM>
z+BHqQ__tx4?*O<wc8RvdmdYHtNnatU5AFLx;Mh2dspe<eRFMbp5@Wj*t?eHUQxp=b
z9*+_STz~q*4(-{LgDP#VcGr$W=L+q1CI;|LRJ-$Vwxv`C+q!Gwu+nwScgA<6i|PKj
zQZReuVx+7~1tRe?@ShKT+FQG;KXxZzYiZ4NOed#n8#f?RY+z-M8{&JN<G`&d9bEnP
z_9#lS+b*79T<Vp*#6z@VgUf%fBgcAn*vsSHf`6$}bpLFW&h98fOtT^et2Nckj~=VW
zGOMUo*{WEi*Rh`4DJVvcai^sBj24ru(tkN3eBI=f-lz(zWV!x)sun)>6Zvdu7~5D^
zd@|@jI?21iH=C^m`b<@ojxZo$VCr>35p`Q*s!7KUH3du|_ME!kgB|xOKrm-jzY6e8
z?tj3KdnM%YEFUB)=K6Vv3uWLwK<FSX9;}r{QramV%rAwSvcd_!=336NZo@Bf6wuG^
zCyUoaL>DoJqC@TK;6_?&nB-zW?T*8~A<|M%9o818CeAQMV+Ai=#wGKJcL7GT9mIQS
zn<#VFOpr4J135n><JfQUXhwquq^fPVi+@z@ws}=Ndae1?9ox@O$G~Eyh~R*s+T1P1
zKq7x|P`1D7^9hPzI+vNA*a{a<i59g?H8kumPqLd%4nB{|^Y^y1G;Xm~y{(?Nv|*je
zM$%zu``IN!m~f<goFTK;mT@;650vCT7mAxh4x$c9lFb=WeJ(LzGnG3f<mie94u7(X
znN4LZRlanqz($a@?N|6_cPo6MQ$WAU9ov)ey)~(N2;byQ{r48gO%HuJxd}NdE_uPW
z+_q((ZKf+;+7;$~w>*-kxd2-??#8zwlH@M=AossWm*~dlF`Co~u*W^6bc5|^+9|7J
zJ9hI<;~|vqIASw9#L~P|n{Nxyuzze)o8QQCzc!yssETF&?Q4D9$aXqY;ULTU2lCw@
znLBj<fn9D;oFoTd=yC%BbH@EYcDZkV=1Pf&lsylAQm0!>hC=?NMmJD=CXM;a`LNw?
zxAUJf+6OedTfhsw?smc#THThiT;;yr>2BdXsL|aa`}sb13+M}N?vCu6T_f&x{C9jz
zveV=?#NBV`aMxfS)ZVU_xVyLA|B_StA8BBMvzK9DIu!vim!UW+69F}sVF>~!my}XE
znSTfY?8`sC{pFJ+)Ek~k3+v*O^F|;x6;H&+82mN})gq2IecPA3x<$;iQ|YbAFn@Uu
zu%&^M0<@c9E5z2S@-goO;ed((M2L2VU~UNu{PyLyaS4}#X?ypYu+$0PSN!X(@1^~=
zr`RhSgoAl$#JwQy3y?b-P{9fr{Q|`iNq;qc5;-i9+BNbzdIL1Wmr-e}2898hX{*bY
zkui1ik?^eZ^GfY)+DRP78d~E<>!jcBL@?JzW=L%{TWQhpW#k{ncFbI5$%IX!=OUg?
z2^mzQUxIlnv@_vc2#prnE)cXk7d$J4PNF-`%DcQRV)R)B_gwd${_>T3_C-OR$A1O1
zjdY9)^jCvjAsz9M-j1^`@Q~)c^1T%0AW-OWw^A7V1a0-xAE++Yr+@qPklxxzu~0i>
z;pwC9Nj6DP=;PO2y}<LbF|OZ^7;78|tyY0joRJxg_yWGNV`TH5L5D=&3p5HCs8OS(
z;{~UplSxuX{UioL(5Q3Gg5h_t5Py&M6`r6$5YjIzq6OSiPAj358>|T`Lj{>yl}eQ(
zRW0Fj`HrKsA!?JNIi!L|DyT58c28{tq%|x!5_eD63qgRj%X+p;K;Fms`U<o#*33`_
zbp@4H?J3Egf^2jmnI`B8XHZ#r6qio!ANc1g5vRM=poH<T!!w(~z|{JFr+->&A&p)r
zZK`L`cBxi%4gB<w87k7DE4MtSN2NY-pvqd6_5|T;qf9Y(XS+AX39nZ!ojL_{>Mota
zq8u;l(pixrG2UiRr+|FVblRRug_7^#SZ;A9o>C_h92wIm{iUw%`;xZVA>Od2wQ&Kd
zRr02jhQZ%k<66cf-s>ym>VMG7KF=iAKolWqyOCneORfIw+skJIlGz3`p4!mfyJo-2
z`Qw(u6<IbiHth)dO)<xDJl|X1${;lKGZvW}Sym*9TUxC|DLXz`*Yx@??>1~TA+3sX
zDh$}B=M$~ph!>JvzxTU(qCxXYSkMjHjp7(E6HPVP=PqMipHwL*G=DBu*i24(;a!En
z=tNYibF!3@-XZ)SvD{7Dn<xPJGVhJ_anoA2>M(C3K;=D#CSismZ$++&9xH{e1Pg^i
zhRbAo`r4mnDuwFK57cF6I|mVsJ9&**b$Z?HmMaY#J6VXMy<OEdWaSoF!?@R?aC4i4
zJ5{uQWlHRf=dj|^pMM6jPUet@b>w9nOD;jjNOFJ2^HElzR4u!B)DuCeayWaYMfElE
z9x{%KtZjLsI7N7fO7FSvo!{>`O=C~&N|iE#VL4^TH5w#E`(`p%*3vr<tMI6L5@k#^
z#AJ*arX<~j&8=uSkbCJeK{(mYY1e~gmI;IJeduqd?1ou@eSeq>(y|or{QWKnd?}w~
zi!(aDgOo(5FP$CT47Z6xo&dkU?p>3#UU5$nJCv-yxbs((&beuiDX4^*s(GtP0V!S0
z0yoj2zM08#BmmkW>jw3*MEC$3#Y+w`KXh>IppXg)Nrh;BJmBu}Xd}{`s^~MWp1aIr
zc%s-b_)Y=W)_<eoftgw#E8)c}zI~*GcaVDt-%5ls!o?<UXVMSv^zH)Bm3x8bS-bMX
z0>fc>AbLyIL1VN#AwFbkyWpO(S!xkOm?2%;-eNgUBTn>>TyXPBEpwyf>`N5iCNN){
zVrbpGo}7HmGkc6#-;b89&6=6&_#0C_WY6Ia-dDks(SPSi&0vh~b>Mbps~A$`8!fkX
zK~t=eci57SW>!A=Ny9kSj2cfG)0q3Jpu!(=bNZQ$^IZj*JE}v02+RIIEK1FXYB>Bm
zEA>!V+5`Sc$K<nmpfscuo+ZS8SXngrZi%6;EKZeywgRk~5o%V`875rX*%NzG#vUc7
zsCadrn14z*=mnR@`~a=JP+_|v;E2+uPC=y_nqLr-uXkoCp|zT4b#e}Rs^D^xIZejj
z=23de84@%-5k2y=dA_a<ES|HC!L4XCM{@_-%7h?-8VhX8=|gxX&RCQuu4fk5)APJO
zjap$RJC?DxvQqYj(H?Zmrk+RbMUdQTZ8Pn2{(nHfj&2goIj_NLbVX4;&-D56tk=0C
ztk40rZeA5+*Mr2z;~XgngRf@1@2rcvjb@RCCapv<=H8L^%9#%9_p)(~7K=oBCxRa;
zOgWDAaj{eKw8$s7R+9S+bFIkKOhI0&m1nDa9pr@?6XEHq9hS4B-r+xn#Z)Bjko!8O
z=6~kn7VFIkDY;R~Zq%`s@yrv{=8f!pBf_+LY!jcWoKVfLb8yRjXGDZ9f!90()OWl#
z(fmON&r21f)4JywTaF3MMx%M1e?$-ChY`2-7e6OMivd)P`94-a=*^`zf*v`Y>#rb%
zbY`aK9X+6f7f#uQ3xfr2*6JkJMx(Ch1AjSOU|yr$(s1<Hw{m2ah)%7Qyq0_4@u9dn
zmol_ro1yhv+BC>58ygm!!a+qsF{iNkkg$<1@DJhMqNpQWHo9cO^4^&&qfhIiht+C>
z&aHI4E(WY38Ion!o5;;hRf&;1wj@m?jlb3L+(Av;DX^;!AefH1#jcHE6&)J4R)3?;
zqi`>LI&b4{rTIUqYEl;CLIp2n0MNS6yJgztH1k>|1HVDKpBP{Mj~7C<ivr{8i4NJC
z$JA$|;!Ez<sFGQpwtdbOI#$QG297A|sB#)THKUbZ`xNHz_PAYdd+q^5FJ2DI^P(%c
zb4EITI3+XEyK-%waiQ^uMm=J_XnzJXl!O}#YiR9FB8tY6O1^?4xo}@v#+@5Vnw3->
zLB3!Xc02Gq^#n|NyiR?2@z9~uF$##CqgQs7bLL^lnGa1g7c=P>v*?FH+A`{UhweNd
zVsO{<=6`bZsH4fL`)SojqDQnx8a4Ec_vtY#N^3oUloPq}L)da$a}rtmaev!|#t*eB
zB<EF`$0CQWeyN9v^6<9xLA8eA-Whc+P7jocnt+|Pyw-RRuEq!q=A~BAsROIdKZ-jy
zW2czzI5!c5o!InoqSgG8pCMY_3KiOUXtZ%X=T|MH2DkXr?y_g@u0?lBDYTL><MlDb
zd;!+J`+>wv$VRDVqz`k_NPi5n;g+{ZIUb;wzwWmETAmkgLE5e(ox$H&Q(~oRmmn5F
zaa!nJmjj-Ey=OjKEF=Ze0X;k^v`v*WDO{pk?rYk<IqOQUh#4!4ii6jzSd871^UF;a
zx{;%m72tYbi)4Fv8+#2D)32@1D6&{P6xX@+k>)_8$0_esJkRSN3V+p!necjjZ!Ff^
z3KpYsbZM*U>(}Y8uirlXXW^|v#TBD^`seST(hCDtIFFB&h+v-n_}9~KpZ+>SC0fqk
z`2tjCm~_tCkzp=3{wOr&%Ed7iJH~9CPDgPIZPdu8{!bR8;R$PvgWv?HdGS(k+e-(I
zbZS9YZ$c<l(f!&Zj(-yL4EK_9zv9D)E2r=*#0M}}9)yw$=4!#E1!NE56HBl1eSFdx
z#_3Y#uEhd%c#=<dlsjS35>=_u%O2L7VtuupeBQHv=}wnYaQT)S_zSn(Y($OZXV?$j
za!b9PFhjU}%dOuA+23-Tb7Qw5_czGNOEC;@4JymrnW?pB&VL2gh<eOEG9}0}c1e?d
z8GDpDKF9gI!wYyvz8cz^h4ByG;!9~{g%UhL`H5yuA=+B~fR$4U51x%5YUP*?Bo~#u
z)$=PUQOGbO^b^e+1SLYyf0B8FfDvYIjUPzVN%~P%Ej85wOpBgbwPFUx@qfNevpI=y
z8S82T#|ftf?SFfW0az^n6mOefdAC<Y%QcI*FYVXk!!GR&%CCgvT><*3KQteD+2&qQ
zhLk{6yjf#__WE%*JvE5n{WUs{r4Pj~R4FoEp?S)8FL6ThDwdV$T3@B({Z@A*lL&DK
zW+U;3$XF0ksRfGbfhEtS5OQ{?Q5}$ssli2fHF`_{ynlOZ6wg}IsWs#-09)^Y3(~EG
ztBb$UvM{`E$SY3lsF+-d)OM;U_AZ#!>fd3OmIe(7%sW?)aZ76{0IkCWh27Bnn4=Ly
zTyedZ7d_a^-QysW#$EPwWD{^yx_^cyJ%@p1^qV~R{~=Pm9sD%<$~_XumFqCieDHLE
zNB2Rx8Gj1z8q3O6TJAf_C3n>e6N~3LmSW1td)40DFC-&xOzarAwjS?^t9Rt$g7r?}
zOS|2{HM|1L_X2bej=!x~yA^B7^BT*lY&RAm`Rxwsjlcrq%UHKY?T|G<b?OdeGFk!B
zTolP+aIO~G=lN!0>qWX>!%qL#h{F>;f*lEd?tkgZ6l%Z(;DM?|zP$~00>J*CjPov9
zw25&+|8kFmztBsubvM3DI_lt7r%k%)$zuRz918-j=r_Rd3&4+)FR37pEJd1(6DO8(
zKi8EAmO25xPg@_~c?f8E?2OE{rU8X)HE46L>;xIauP{AmA5%6%PK;V^=p6Z@54O~d
ziBO39PS8amLvhiqGV`dA4U-=cvT@==LiY1YsVZXZn*KA75xHwZmx{mI!8~~ts3BD3
zcjvgNN(RXOXj`!rHp~c3!Z~CK>HF)qPyYqXRA85ZCoB^JF)^3HXGIg120{dv@wg=?
zm$P6xkbgyi=JmI?pPnGIUhb7&;{|zjF_r{#mHC}+*v+b=ofM$Q(LlTw`8eat!^u@#
zoa1%$V&;P{mO!%vbmKeYLW~pK14Vu=-y$;nKffcEO6$BT|4-rn$2C9}N%2r%$P7df
zAza}J11OZ4Tvr~WEDWxSr%h0T<tL4qB?ddJ0Dq7AB{LtsB!vaG%lN=o7!g^gOf|f>
zb#z@PKG`Hsg{VC9UB0fXrzq2a00#S{GP}{|lxc-A%bB09OnF+7C{W>oLCOr+UGWsM
z85lRtjtUGsseEB#Jt~#I5oV<YHNp%<{5Ca)gHgOfDo&wUzX7c{CF`8U-xHNA3Q1<c
zY=5=Iu;}#}rCfcBgWcvggJvW|9u}5EPDtfjc$5BGTE(g}2l|RtxsXBe#j@)hpGIQw
zH`rAe8pT0(!A@s#yC9+?L26Z?Gxh?&%lmQmti%IbHNr|-ti27J$`Dg>r5U<7B3R_K
z&#&d&1Pf`m4Dzg9_lA2du`xL;(*=bgnSUkUe?P(M2@5M!Mo33R4B15cL+_E^3vX~i
zL+-~{<nTiBY<%rTM-)0g3<|pyh36RXs97KdIuNLx!wU%F%PA)NFR!+I`^&d&FHuTJ
zD@T5CBhQ4QLT6q>DH6Xw^1E9#yH@gT6PIFvLz4ZPGH2`Sy5!S79~R7tD-=G0rGE_i
z`ts0jRohw%WB|BEQ{XZqw}&|;gft|iw8QLosB^tU*iM37A<;u|5JWS4egF#&vYr;K
zs8POmL{UY=@cNO$EIk$Hcj?}k`LtaKmAo4xTP~u~_{^DYC3l7AYD$sD*J^MXk87&B
zv*8FPq#Qeu(uOk$2;%{hi>-dyaep(3N@&go+9?6jb=3oAj&+IL>wzQp5Za)yP|MH|
zLTg;UXUkLBL2EHNTOs*AN7S%eo&8#C0=J(Yr5E5vhJxc{#VTlG$2|!e>(m0TH$ld5
z`!+nizCHc(w7`ySkqlja|NAL?AgP71{=xyobEf?FkISD=KaW(01_#<+Eq`hXjM2K+
zlo6UEg;rA78vD2E?wE~WmMx9R@)2C-KhbQuH@`H>wOl*1oAryzui@HABb=IW7_SzL
zunjzN1bLmKPO6(rt5K3}u?TJ6%|5aT81IkS501jK0b-U)M7uZ9g*>b&2Dv~%7!{*R
zge#3y^6)-(uWK4T*H~=ZJ%2&{8XNHULZE=N2&oZ*$EF6ARKn^}h1e=3D1+@QbUSmq
z6VphORzf;Rl}+M2j9>GwqePaSjN5Tx_NxH)Zw+JG(qlnNeh#kq()55$ahmC1PHtZZ
zRhmLKnd;8E+eWZi_jt?Apzh^8aEZ+n$2AmKH_%!p+Gj<6o&Ocb4S!ocIwFT94c4(S
z;7=7D-D)L%GfiU(E#6(fM@<{DjuLstR&9I9l-@-{vnN=zE6AFe$I5%N7c%yf#UeRn
z8k!@q8A_dElVGxJdde+}ODJ-y@{%^!D(Ya{m|lvljC{lzO0h&ul}oxwp08z8?M4a>
z*QDL4rx3N)uRdgyh<_3+%0jR!_8IMdO3!iSxKcOTHU**S>FjGBxyV6Wp#L!sA<b0e
zf?6!~z`x8GwY5T&C`P*$>7rRO1G>{((B%$fVM(FIb$v4pP3!E2&a-WKr*_Y-7Lctk
zreaoiWjxeal<l>Sb5h-@k)m1i6U)z_2W!+j%Foy}IiY==wtw8xCN7t7ridA@x~Dni
z61}(Z-F?JDIo`i(p^MZu45`p<TsllwyPemOfO(5@KU;tk9^@cjTQ@TqO)rZ;_MxJg
z!=5nsql#u_O_Z&%hPFZbIM}nU76$b{iXBD3?r@G<){Zj6hHq`{yQ~vN+OgqGZw|eq
zjNGk>lwD@=vVVdp8<h1;iAafiH|Hy|EiG<-%hSAF*KpHFaT~wWIj_&7W6z|9>S#c-
z#>BkUpSj0)3Q@Q^u$~-e7D`7a9Oyoy?QF%Y*Jhs;p_K&T?El#yK+mgx-?-WPX$o=!
z*kNNZ4uRv11G0k14UB$P07OoC>W2kD)Lxi503HRn9)Gpu8hLa2V-XGcPEAC+J~Kxr
zHWbmo$@rA$wXv<tE&p}#v($^XI&*5e+qO#fS81}&`&n*0jY_K=KpQK6&`tj|uOE+4
zIx3}tj7KI(o>oF#`nI^YjAwa>>&$cOyc)NCCs8Y7nq@ktD&q#D$>+WcQ;<^VCeGda
zU3%?jpMOtvIGvP8&HE6g!?>5Xr4y+&_0;5~+NsGEDQ3!Z4$Bvoo~-M3AtL!m=k0L}
z#;#21mnOf(QHi6vvTIJ2l9he0(-h@V5Xc4gx9-sI$e9T^m^d|3i54AmQr!_qh!krw
z6%}grvEAd!!2?qFTEF1onZM?ogN8;<Q_|euC4Y#!$4jK9oZ+EZde;7$g@<!Gt(PaO
zoJ;PXnxw-aUieAK$CI=)QYfi@bdpw0REAma$~4`sK0KU8$Ivlwy%3+3z$KCZ`-Aeg
zRBR1%0zJ&%=K+;3-yyww5;!|cNbic|ne#%DXWI|Lv2`Ii7RH3_R?swVb#?2RIoMJ@
zn1AgzS=tq16B4OcpMq8by^438vyZ}VQ|_J6)(!x}ZGTq_xP4T5m=wb>ZH88oYt0+s
z0^_zd((Ml;rtajj%x$-#s7rHhHF!D<=Pk_-gCh2+3Az;=b4dnabGA8WovnJiZOZzK
zvio*3mfCo?v2hj?SfV4FZEW30oxUD#^fN2q`k&)RFK_cRhRAnc3T$umbL;3;Pb;>w
z?T=OJ4^uuh2N{ZO&G{5G-Z|L*2R)pr#g}1VIurshGn3&MBbT6QIt~Fgmm#nPCV$;p
zU2j{*5q<ZsSRgOygZJ*t?(FOofdMD3(*O=?*e=l4(1R?qu~J1eB+79Aeb4L_uc(k|
zDUmXR8ZbG$+PfdKGjq<&E>X%Uq(JzlK$*M5As6jZ4n?g?8mg6`&00`AhG0Rp2DOwz
zBXfa5t78u(SdvGO1idB_e<<E1cYm>9M5aDO@nO9cM7&#C>0%C~X%z9<a_gc5PBQjT
z15+EHYTz=CR$5@my<f?JYwZoB1lH6ihL>5<jZ`*B=s+s^=9c>m1id!x!%7Nl;}n%b
z?o)=SA(Ny&mW9kZ_6o7IuBVzpuAIAw*HlIk8f~q15h_j8=T%E6ocm~1LVwA1Y*0gy
zHom4%KlPPiCB??=sTL%Sde$6-bzCo&lBxHA1xS`Y9mP`1xOD85l+s7j1hMtkFe$0j
z*JgOC+}Eb)Ol^D(qZ#*pTMCs%?KTRPwd<^LsAZflK5HE}8{1g5^kqWH)u0#HfkCo9
z1!zlQUXKhTK*HppYz|6jV}I=CL|CFRr+_()T?)kBB@7Hk*ewO>2wKKIF<}+$Q#CX|
zuU&-I+PEx}(A*katT5V91*V~4OS;%Xlib$=^KP~82y`1-gG$_2VB&5Sn3zkxD*PQ2
zO<l=E8SBXHk&E8ydb%lK@5Z(itjIWXG>|?8tSy*Q?;&CRqrQo{&VRCX+5KoVH*yWb
z9&?wg%UPjF?T=d@#`{Xr?U1}P7+CMd%^b=eyYv+`_djy2yCV9$#))>{f|KI&dMNh6
zYd10KJu4jHn{S3U|Nb}|o}Qo27sKnTcMJdZ;`IDqVfc1_@qTvk*5gY$8vZ=|Fnsrx
z#viX{CyTdaHDZ%Fr+--eTT5j3>)fJhJ{q0|^!|GIe*R`YJPTios$iv3+}ung7yH8#
z{JTAe&BJ0zftJHEM}g;%NzB-05{||Q0*Rs>J-W4s;px=*$>MZ=j_LgM)ervv>ko^?
z$ICws!%wG+4_EKv$^4h$=c`{nK7K#@H2iqH`0$sHiyzJxvww?!9-qwO&u7ECv-!K>
z6GcvuhRchS;drsQIDL1um|fz}U%h~px5xOb1*>}W>t)?fwJ@PrvP`PQwnpgh!Btyx
zWH*#+%~p#_`ukGtv$q~b1S@$p{O#}ma2JVic!U@V=T~QEN59of(F{@5Qf$K8G%dMA
zh5K#MG*gL^w}0xDti+7Sbkj5p-_6hF7q34;D<S&}VR#8m%r4FYk3awPdGY=0#qnYm
zXngQ|e!jpU&mDz4X8S155!{JO@8QXl;pN5r<n?Uvc6j;h`S51;d2zI27cY;0nqgRM
z#`)s1!%!ay)^It$x;U9#UPGxr`*HUE^!VHP=Wad~5q}C{`3OB+;9ZYCZCqqGwaaV5
z(5LX|?MNAX#v@}`N<2w<lFB5>B-KgMNop<~?dE0DqQi$|i-J92u8|#{i^i(!ZsyTk
z8J<>E8n*_P#T)$Wj<euWVk*-lfc()})Nf9Lv$W*g2>wt7%udrz&R#0P6>hw})D|(s
z7j*Xxjeov+(d_XHIQ1TaDd<QPAUpOBu7_`g3IVOijbpJ58ZLlpLo0PAoULTioiv=W
z!v%J>)Nqy*&D!Ats;}^|f65-z3(683_#lf!m?MZK449q1OQgK6-U7`Q{85cqEBG(b
zPTFNSj&sL}2NcZ1@&UEms%sZfeh=Es&0=;ckbf1cE1pmQjNz>a54s^dcsPLNVhW8B
zU|X`?BNXx^@BsncjkS6H<B`f4ClsCKE>q}eC^VrVQjkr6z0zWdm$eCeCU4bLb1ZeM
zrlo>lRo<p4B0=<2VT-1Wy{ho{2f%i5T94d!XQ0;&K(F@%dc6howuESzq&k6roxs0N
z;D28y@UIj2*9rXV1pajb|Jo+WHqd}O5Zz+2R|%F(S86PD$jF>K?jh&j2k=#njL0k%
z3^qFeB4&V|3epL1nEO(TyRxi~egR27!m^ir6lq(Kgr{jecra4xp0XmRM^+Ewy}ubK
zfXPFrH83m+1)&v`Zv)NqWE?xe3D{CwRezKHO#*v?_|Um#D<03C{jKkXHq+`QGGty1
z-6hStvAj&tizG{uhY%%1hXl!z9VA8fmo8@@f8*M}Nnr0_Bt`7O8Av9ENw4rbr0MOT
z{O(%0y$rA?<#Hzhc~UNCH(-5m<&x#7UMrl+JPAQf{=bw9fR8tJ>t!C!<p7{SU%$as
z`#XOt^K4RVr&vl`m3edn6)Y{9+BQw0+~}-mi>8?ftZ|!`Wb97zz*U}GwD8VMu<1Gz
zyry_}W`dShn4nH9O(!PUCMMV>CfFt>*d``8u5!on4t5lD#*a;qn0jHFqf<<gg2Z$k
z7DyCG&wzU*JIElD<hql83Drn9kxTHnmM?#axu1?NPLJQ6UHey-$1lo0c;!;Do(MW5
zEEQWE_clPn-W8j@<WN=sw}DfZb}-E{;#N+3$@W1KxpQPAFjUVu{d^PHt9AWI1>}$$
zWl(jatfd(3fh{wz$Z0Yoy*EEm)f&a?J0zIIhKd8ShT;cHDHZV(KH}W~AGBQSjoW`X
zL;~~<wz-WSzcBd~B(?g#Y8WdmZfY21t7b?MpwF$E3P6i>o2EiY0*h?ZG<%{Xd7Gw4
zP>@R6mn?M*&w(-|soe+dD3vv|^Qz8$(9X}UZ_v&r6tfA%Y^rD2RL`)fo?%lx!zQFO
z+a34V5Dl>a><lAFBRN562$fqTT_=B)xn%_hKvHbxXNRfmmT1<L0R}guvKI>v2YMu4
zPtm%U;0$Lg$epn?#U^`n_Nxn}I<BvmGRQAWfcr=>g<hN0TK5;8O92KtHe842f=~du
z`{3cZu*7~#W2z(o{Pl3yOP2cC@ILBg+M<XCHt?DTT45!o<hO?mJno*Jb60<EwLH~u
zO5;~qEcLMMPECa$9({{p+!gc{ftei+eG5pd@@@3>`$JnKgtwbrXL+z*zg@FMQ$p_8
z9zJzj0kfU*IHC>6M)yQETB+#!^`{Y?{PxpQ(n)HQw9NAS=F?K5-SI4Eiw4}TVqO;M
zhc<pQu2zfSBpcEVI7)>9H_(4oBd$&Ma7<5)y{H#z(a7q@ofJLgucDZR-P1Aed{+P<
zj*J#7Z_Teow&<5N#R;3PBhclndnb56q@7Q&Gt;dT?8bb!pr6=JNj6sC?sjs_bo=C3
z=}Vp*pPs*mTchuBS%zR|`zR}B&(J;0V4q6!OdZGuPlYui{B5{ECb@qJ!CPvvmf5K6
z<Jj1bto^Pe%uoEhB~iq$!ch8#;%3XhTJWfvlo^m~?bfle-_88*Dc5R?n^RkCEu!ic
zP5soi75kF4y4_M4UN7AF{p@=M{7fqWpQeybQ%L72r1Jy<%#)NSX%B@vuO}f@LCTUk
z2Xg+^67`!Uyn$$MXT5)SsM|G)b`$KRIpe0og#|beqMh(=(azJB>bLKg#2s)-q6b`-
zAly;iaanF0sw>b$75L1DJI2V!NAg~-{uucYdGL&{c_xoQCC)sM*_Tbn_s)3|i-H=!
zA;eU7ykU)mS5^*)2sueQ=m}Bbemf#xXM6tv-tA@X3T19&b99$DIxY+XF)^1hjtvz8
zH8PjM(;y~)?ORE2W4RH&`&Z}>n9ZfQ0fGR_X*G~r8omWN1(`X`A@lzaDU!wgZL;Og
z9DH~!>pilnSbKeX%}>9d`02;f_3i1~O~J`Br-hZLx8F_zA8`CvaJ*^s$qK=Y5~sI6
zPX9b}E)8B+ynu71@Ur-~!%uyAR>I*5VdxT<XA9zgT-VCiyhiD?OZvQ<J_T`bxY9by
z1HD(gkb<z6f4%+viIeExZcc2;Z7eX9F3XMVIUp5{@&d}QB>t_^cZnRdpl2wBMrpvO
z^<dIHfzK7)i_c##;b?J5_<Bpr!ZFih{3Moei-NByT57{QmxI7BHxra)gXs1mG~gfl
zHOZWR<Vo6K-UNk&j~~Og?{824eJW~^ifQ9dMj2+UJ^lNSr|<xTX2$xHl*|h_{rR8M
zZ%_XisglOW&*(hS2;3O0J0nQLw00*0ig69a2%Q3tY7hUo(*0l@G>CRd`8Tqeq%CM8
z$Tqx^egGvZ|NQy;X}I(AM^TXk!9dzkXz{myaCN@N8ETR<9%x9NsCIXL2LJSYNi`BY
z^U91|st*pF_T^tQf;8w@T1}kz-5d%<s~H$MX$U%sb+u<!O#0-nOP*VzdL<iW_vMB;
z=gV2(@+?L01-<E7VU;}B3=prpQU{01(O*^~Jsom{&UbR=7^s<aMM^JZLry-J?Ah#p
z*Rp&wM4z`5^>X^iPq)vwm1LCB43o&HKmM|_%Q7X@xJ#Q+>oF2tr<>@{MRI+SF=9&q
z=jbK=%efq+W6qk1zh0uWnFrQR2oCdKNWMLXP`i|K6qn$)rSy2y%TJUtd%yO8xuZl<
zq*BU6@g}UY;#~ba&<}k0Ew7CGY_7$Bm5NKTobQLYC>1+Pn(JbFBC17`v=);R7j1?^
zO`=DrDVm^(h)C7vN}Sh&rKc|YU)&?;+_tQ{l6=0U4N&;5Mi{L6vzqb&Oeau>auASS
zqb5z&WXvFLK{s1|r@Ua^!>H#+Mu+B1VSXMsWRD@uWPY!tp-6w-&c!-FTDmQNX0pkf
zFoU7aY7T0=C1>wOg6PpU<Q{&dcz8*Y#1*9Pv?8u|<PW-fp0cWqkNPcZ@U^jGiL|mZ
z$B$jGY3LH1lcB=m<rJ3bSb3+rKi4~5#TAocRE%0Nzsv>4EjAjfB|tkSuw5Cr>=)8d
z=qc$qb56w(T$7fvB3~vI^$I_KP_k)1ufe44NHv$BsDqbmdVq(Sqd2O@cE^iddrYB-
z_jGTgnDk5<jZl6;(M@^ii20<nD3lq63iS<qP(Yl6if<DAmuv(wWH2Ha^8S#+6^J3W
z>CSqce^$!3(UjKbgYM@9+R^4=x&v(;XipMW@Ik85v&qz-q!nSnSfHGLdvZgRi4#nE
zDS7WkF;9{$XEK~KLP+IMvZSY6iHzLMZ(&uNo6??rE!LovcB)}!r&G&Z3Qu+JT@{>A
z*Q^QZI-;j>9X!odkBVMaX_zK66F(q+L{ZnUrjwg$>G=2ehXpbSU|Px1PHV%dYQ4ke
z6m%#$!7A*p-Q*Fv_FCkBkz|u_`T{$c>WS?XLP>Se^Q33O)lT79pCOdPIda_mzFB@7
z3E`F*X+};*YRy^;Dg=|@ip-?02Nq!g9EaSA(PN9S@R~^uBbP2?)y`gYPlP#F1B_Xu
znH)sv^sh~kjCZ;`pLVQfiFP7M?ooZW;YM<m;)ko`xyu^KKuB1BIW&XP#+it+;II~r
zwD{yec?KV1PLF-0&bXhc?lIsQ8Ce7?vF{eY7+DOxK`vgoIB%q$)fjWAI#gVLh|fYB
zL1|`{_(f<VfW$Gy9nDa8XjGh+K*+}<kVc{(`|%);Mj$O_5k9HwGZ0AUn8buWa{2%Q
z=?yv*Od487Aklk&nRk;uKZHPHGs7IH7J&qPWcp95j*K(3vdQkm;aI(v$}pwWB<W#y
zuXQT<@-{{&umE~Da*MltJ@uVhBJ_~<yc0d-Xmw08qhyc7M;v<N<P)L424~OBBo!^U
zdLC8#*wEpEZt>#AZ<d=Jg%!;?*R@M~vlLZgLPk{688n!G7_HiwL@HW1de=%*rmV((
zQct*`obV#Ki3`T36D}wxTrjO0aY32=lAcfUcl?JKsnT`0poHZU?BRk^?cstW<vm=`
z*ip?7h6M$-#RG?`woO=YMM>|31(ks>NN|J%8>M^_9N2L9f&y221*Ne9f&ObjfrC(A
zP~iU&3PhHFjQcMn@Gp)8_GbD-AaL-9&%yzf3|U;i4+k7JKXA(4#4QvKtwxUAKY$H}
z!9{ZU`Pksd=yR~afy);*_zrAPdWlZv@4yDX-LyM-22ygf^cG@3bgZw%w!9k|-)DTr
z4}NY;-&7c_Yx>R*Vu=xC>X`P{s70pAGgXff5WQr7s|L|4q_*Fp#B(uk)g!Ua{;-$w
zs>;H7<$ZGr=0t`X>1AGrHZR#r--~Vvr!c<NFG4qk^UVA9J=`NMv(#Ja@9E9kQ2h-T
zl>7|;04j##Aahc;sF*76e{@JPr&0jq9j>f=sV9g&rJCk#d<SO}{}NcIq?u%Dv?6?Y
zpLo@OnLsmKH&W^IYUz7Sdy<S^(et+D^m9%tZDe1wS4TW7gqrK=bbcG9Q-(?GWQJXW
zSH6O8^99fFoFfIbUA8d;GYMkS;_-yHs8gu^#apB@RhrmoM*XU;I_3AjO1lEhyPCDy
zY-s@nV*vI}?c;Y$Oq2HOn|BLvz!YI1n6qtv%T~h*NpRnQuJo-ve7qB#w{f|-<nvb5
zYg7&zFTHTs&Sen^nW;vuA2dvGPgK>QrX${IN8J{)tek&G_vtjbAulIBrd7hFb~KwD
zoKs0Np0HdZMB_;8xpJ>o&0q89<Y5vEO#NlFjVM@v)6(qd)&Q`K#~OET=0KO{IX9wz
zw5FiQak^P@w1Sbh()d!tak{kUq2et|)1mUsyUIxhSmVEJjzV>Ci*%P0eIdGDV|C(G
zI0z~`+F~ca08VDyz_{y<>>eQ9@EgKFVJCh-W5<K5hh6-&=w!g*Mk~7(HwcpEH;Y~8
zUuB?M_FK%ZzP=bLs?h{rBOh=XzvH@pw1V2#4m32~inv#(tEk0YocsR*pupYXP&gtk
zv`)QD-SCvCPl2DFigN{Ev^e-ud-xH-1xM_ZyVqrS96b2AiSp^gKpF0sl5$|O3q!2@
zI=>6dNQOEaeXPY6MYc+hfll7#XxzRQdsV$mF3k<Yhed;~@!kC69JuC3!uqXfO5heG
zutfsf)~uD?dPk|rqo8fHp!4^F@VW`Fb^VW;tA2>hyx?xq$ii5YUXyswXjCS_JRRNd
zO=W&s^(3!wo{-}~Vw0rxxrS>W%<(QzB3O174-9o%G}Hgk*}Z5%pE%6^4{T?}m(ffL
z69X|ZF_%GA7ZjI&emcei<*v6!e>%Jg1B1qzmw~z$69X|ZF_+OpJQD&kHJ8BCASaht
zi8_-3cbDvmI@^CtGd%`_k!iuW6=g>+i_4<wDw#bB`4U}2Ft$rHjcEEoy)j{TEu-xm
z4+|3&{)?;D*lcilcECbR`7F?~5&Zztq(`(&nb|pbqw|dB$y-E*1^y9+oT$=#)p&LO
z3WG+|m&;u0pve*Ife4lM2>kFa&!ZLeuejSR`^2w=_9B1tyb>+BsqBI-HLr{Ha7LR+
zd*VzgO`6MzuUwpdl2cy)^5graREkz1GC)`Ban|9$eqHo&aQ-PsX@X>hF7}f-kTg(=
zsH6V3s04|52UodX8wFeopPXh9n9I!bjGPKgNXdjY9dfT%o!L#UMR`M+`gdpu&+l(n
zWv49V^yYs>Nu)UgcP}+XtK>XOG%3jM8-kx5g834DqJ}u_vGlL+-LLnLr~d}0qR5SD
zIQ{wWlOIP=OdEG1ZFi*YKmR)Y{q$Q8v`$}wZgH)ZY8j!KA|4TMnYE_zSW_OXnRc{m
z_$59ve2eB@eHPbbH1Bb2F?282yja}mW3rm&9h84Z>OE#m{2H0=vm);|DwhIj_G{~2
z3hI89r|t4t7zH*oNGXrhnkE$0-AFjII*>OoVo_anM$9omrWjCcmExHZePU;Z#Lmbi
z{mH*>)A4Ckk))#ndyH`&y#-2Uoa|S5k%fH&CG&kL0;m-TtHPh-;$p7VK?*UP+d78Z
zgM5E$c)o<l5TEAMQQprCWmLzwDy^dUgIiq9*+$Y02o{`iyjHo;YmN%#tE#DQGLcK9
z%bTCDqRsm9{E>gtgnl*aQc^IZ7a!K;YB5fpV{C_<M}>HfCFU31sPnL!eEgE1jdIKx
zN2|M4UY2R^_EZT1QBq1D-Q+M`W(gad0c3xc6r$wK$!lMeq*8qM9Kqm4>f8uXRfSzx
zvyAG*SDBt1A55O%d7EISxgis)yv_pOmy>DLja*NVr_t&RI%jjbUJ-bXJ)9_shDO<D
zK#gR0UOUSPTb}v2S;5H^7i~#c@yd_KFu5vOE|@iLvHps#Y^PZ#y-uUfww0_SV8MUO
z7o7SWaKI=K_PdJfQrGxL*4%}i<PV4r)kD0gR-ua045+*&LD?#`O~N~Zmeb4*+^tiE
zkQ$KxC(gd)pr7-?reL~^$*T8+QehW;qXe3aDz`&)Ma<#@({3RTd#yBgsaK{owFXMJ
zWlv*Fm~x^dc&^c6&N8pO95M$3M1FrMv~5)0iEC=Jo<<*kMe$#Ai(J}A$w#_mk+|Y<
z=i08BK|So=NGM@2uStb4OmplZ?KZ8}y*V1eoP)Y(QNdB=Kyc~X2k~1_yAn3qkfUs+
zjHO4pDX^tyXsdREX&`QgwcA2zl_V@9xlfr#)&sYx1D1`Q$<o%ESA?h(Bb9$C{={y(
z<em0ffhY}wrRWwxW#)=UP7zS`D|b>mkd{ephgFv@)tT<BkRi$e=Lf5?Q2GXqEZv{L
zP!*frp|9YoNR?XNGuy~T9%ZXJjAa)IGz=!!e0$-k235leU~5o%-o`*hRgM|No#OW4
zYF|3Tgi(FDPmaVo^>De5ln{R^fI4kR&Yd%)spU^-ST!R*s<x7JObY0s3BKNhLc(44
zlOO3cw5@}spLlx`;9(yoTUbew9>@k6J7OSL$@eLRA=T+t`8*<ZGzTI1=9Cg&%BdaG
zxQ9FFLPx5@evdm;t83@_RTXzQ6UsMFdHfS_Xhv!rjAE$808N*UbFP1QRK-8twh&is
z>nr-YQX3|%d7!UL2^85nerjL0sqBi6jm2G3K5?);w2>THVjVHACF0Kh?@>1^w4$%J
zce7gi<Grbu?cX<eT?v|TTT97HMqKW8ye+`42fp}SB9*9VCvR8oM}=Fw#IPY(6bYBG
z5WMekT9rS?!5&&{%XWVjO9JJL^XM4@)Mg~uqT{O~(meDEyIo|{UukyR6t<yUpZt3V
zNxdJZ+(NYOj`Hh3`BJfv_b~&nLXM?YWSD-7<Oy@9E_e4*(|k>~f@(d`qC5RPL|se=
zIzXQ-R!YjGbp0-OiB)>X-b1=atVuz7$|qvAn#4H}@qqtumf3$5=C3;tK~WT>8Dzev
zRXZ8{OuHz<qFuXl{(H2m1@u+g)uH=<cD0Zm(XNj2eeLRq_yq0hP}tF~cJ`ij$*LE(
zrb#K}3L}_6ta?$$V=ZUa^QVRM*{YNxdbf+n`Dy6Z-;kf~H6#34T$O^<hlaxu$;~|j
z$55CK^a|#ixcPsxX9*pE<|yT|(ge*d?d53Yt?EO)4ZW<fAGHzomJ2rdP@=j()7tgs
z;haRq-5O8iJQ<cq3$R}}J7!5TTBBv*()=QsWZDk6?%d}B%P3jy7Ws-X?+Se98W}Jl
z2Zh|Z(IsuD>pD%*x86IvxeF+WJPILo&u-}c@$QgkwLyPh4jV;Pa_KH6Z1wXLvy{w%
zlVl#bBl*VFMwaid<(Y2w%H^`UNuF&YwhM|~nCcpVU97ORi~2CI9^JOL&VDv+Y*}S0
zKM?f3z6N%!d%uytlfBiFj1(#HW=3^n9eUli(U^hXSvtmd<m|1|X=Yc~2HI<82%_rT
zl7Z%fRKS0_@18Kuo%wK(3Hk86?*3yHJpH{J&pVW+G#m0dm37yak*b59k5$r_{y-SY
zMTV)RWGYmrnF;EKW-fG<V^cVC&0^WJy@*-%?GJ=Ojt519G=1{`+bf1cJZbcJl{4w)
z^ysC&qS*xK@2TT~X4B!Ginh#gGqhAVGlbHkoojyp$2hppZ)1cO*G4A;xFw8M4*>0w
zm>o$fH8M-_r3Ji_Q}zWyyFjkAdROk(9WS0D+0HQFU)FI+eVulbduk``2!iMU<wKpS
zHn@iczBWSf#Vlwyd4L5ipnUQ+DUXa|VC5%NM=hW)DvsJsCbiKvpJJD<1=L=ih#ooJ
z6MZzlueIfnY0-0cQiA6ixFhhv21~qZdi(&#IA*{cz}{LWgg%IUTfk2)!uD^9IsXGY
zN!hKJVPHBHm*Fx76qoz2F({W%gfj$}zLPqWe|S0n_V&}0E}}O%(I~dzWF0AyrphVu
z5XqHfRTddJUiqAMXNap~d`fA}A1)rQVqmqx!1LND(lX``UCIhKQ1qz4n0W<Sq7-s<
z`G2xemMm#|_7bHuQQjB%_S$u6|MRZ&$U$lC=DkJuEV;|89B6=^70a!86<aVh<rDGX
ze}c8$mCGP1XZ!}$uq%f_r^s&3LV#+Bf!^Pj<TU-(QtbY|<WwNLIYC==hJuC5`o~kO
z@?<PTz>A!ux6i}rrL&IbOVt|kxIOtv5OBfVlq8-)v?VM%%T>Q#JMS!!)lkoo^D27T
zMM4%)Y&Q5&*(+IO#&YAE7M1qcYzK<Oe}VuRH{c=`jY+EbwD`M1_A*y1l8-lRPX9^-
z*xvo+%Z?wjvgl_yj33VLhAk(=T#ib&&-)BjEw>6%H=W9Q6{Et0%joB@@7n=Yux_(L
zJI#HY206(M94nJ{lPMp#v^WXUX2m0yW>CTLJnIpcwj?gGF|Z?-b^_jk{s%6df0CO|
z$t}8E*mqoGB;fIb=#cRe@bDezkl>5nK8OxU+92YuN2fY<3UbAv$;Z3)xvvq!5HHum
z1#1Q9u7wMe*En?8IMTI5yP9Xpn3!@GFt3>0rH;2{PL;dvq`THhFDPUE5aLM}!T~9L
zCo9>tUQG%SXMK;nMtey}8<ZaLe;NekoS9cR@)`^ZXfx{(yU3@|$PUbBn@)hUwe+#+
zxh_vCIa-}^3Y<$#BW*5?>#t~2o#%4O%r?=i1e-^1EK__k!Sa~y$vU9fqpLJH&Pizx
zvz0}kpw-h#2-A~Gd(ymWO>W)uapI6jt^KS>6+8*y_xYgY(%?5x6jP6Re?{`D<YdEE
zLXTE9Xv`V=>XW^!T|pP#fx<h+9EVlu>V{L=mFVsj2WQQ-nuY_WfLtA}Fz<GJOExz4
zJ)|5Yz_zk*BhzIydf@eP(5CV<mlH=zwN;l;jK}ss5~WMhBy-42gZnT(Bqx2F>x$+j
zo301UL5K^_dcIt2HE<&me{v+X-LD4Y2sWNjynLCyeSUlTXX9>c#8{m_{r>k;d_W*H
z&V`fI4*}@(?;oc>pMIWKEhaEuQ>B~EPFpJqtnADx7K33jNw>i&I_~cQJp;UJVAt?D
z;E)X@F+B)ypV;S5m`aW@gI434qhTHyO)}@#cIVJt+Aeth1U07rf7tVG%?D-FR1}q6
zT9#;cc(r!PyfqV`<BBqlTILwpC0c3nGCt?KbQ>j4<&~MrmI=|cLS+d{oN0QFsk`2v
zL$75e3M>JV*Ug3MkUjIY>~W<a{J`){&MBkK1L1oPs>D|6gl`*ESBQ59y#jjdNu}-d
zac+!)3My-XKPoyBe+1<qe5dG0D0x5d*ay7FWY_3rmnnUQuM2m^2dVdmaM!5ST@qC%
zxF?8}kwXBib7zXiJoWAJxTQN;Gs;{pX4hQiS=Sx*-tbTiX@i&p<%4gC6$&c$(t1zz
z8&_AkRO#Ee0flnsk%Dlwdc(V<s<fM5jVxij>{OFOD1~Gje^YTPckIej<*qH|<7k2K
zS)N$)JW%hN20{-Sk+M@R=pNT|Do2VIZb-S8D}CRt$0YHMuy3SL4`jw-<Z4+3tFkIv
z)Me;ke?FVqDW^6*OS<}ISJwKidUGgcrMGsYO;62tHPncdjGtHOh1XazrM0P-Ovy(t
zJxz0q*ELNSe{a9thu`>aYHn1tC;Kg{Z4<M-l4)+Vc7%tMKr!ii)KU`50Soi2M_NiI
zR)0LNaMV%~W{rAtI%~pt*e+>yPpCA!FA?p(w2|@<iKmU)DK_iqd7x3L8sQ498cO{k
zvX&~lKEcBVB6_VvsY#$r<n{@PXd+`~O_plaZf>u2e>?Q6oRTW)y)!}aKi6AHW7}n{
zo)2=nwhE|m46~fUVCB`m3B2Dm*@nKQKxGCuFDYW`+usgSQuczI1^fWp!ywc#n0com
z{qA=5Aim<_4Uh+k7suC;5uMkA{y9m#v6b<${`p7M8|b}4z<1OeXyxkt|5b08flull
zHO$z4f9AntZ@ssbSQ~Is=A&v6G&;Ru`#!Y@npkJi!)|zpZ0oM2uI+nE6%k1}8Y>m_
z^gyLjM&HwJ@tanSr-@^fG1b+Vp&p^!T$adD_p_bvMYfbA(qknv?vNTTo<?WWt3B`2
zcQ@pCl#_=EQvGr;iKIZD=9E<QEvTyrpd=d1f9L!zYrKo?lJ!#2xmmNhk9NZZc!*zZ
z^LYJn^|O-MnnB{K^F02HB<-|pii8!#jm9z1=R&(}SpnwvYCLA{HAQxVfh8IhdJex)
z`ht^FWZXpZ*oOn?WzjC}WC>zK^IDrXOY73;_78%bWHLXmmO6>U&k-v9(n~I4B;6Lm
ze<`W0?(T>8Vx<hpPbaqUP0BO}%8;GqK5CN3&{_-A#5T$q#~*%GeW|_YI=3b&+~o(x
zu?S^0U^c*7Gz4t$!SmX5T0^>xg^z(y>NTMk;b92F)Hugm5GMGN9o`Qi8=uPWNvIFx
zXUU%+yn~-fQJ}cDBEK^~!_vq3YF-w9CPn2ZB|eIm1#u9=--(ui24y?*zWW8)F`)2#
z2OXQQ0?2>+#I+ljVPHBG12Qo+mYh1)0dKe0oH}6w1EmgAw>zIYDG39?S_qdx%`OuJ
zF)=bTm+@W=69F}sp=b>!m$|4qn15jWzW(F;Z!eNruY4*kj13p-P!KnjPkIZ(^$Vh=
zzYKAS#5MXjf^meX7)2v@`VR34eVZ~L9Y2kuw!Q;-)VMk?u8C!ZG7qsnH~gHa;cZyU
zGFO<tjUQGBOv!BT0C@+AD_+eoy+jKJPEFoTd@IrP%vF;}qY&l%%iY78U4J0S+Xk5B
zZA_6C@8!-bs>V%}YMS%T6xJ;dy%s(!G}`3X@5kJl6wqz#Wnn8;3F%c`xoD6j7kN*=
z4gZeTs9?4l&vb6R9?DEf)XiJwB0aTb9aStTzAw9-nDTDfY4=x=-mTLhOEc~|uNqe}
zx611%11Xg8v}Y-hK3*Gr`G1Q}(sJ}~-!^rxwHI11<fk`uuZo5GYK{E~-qODgijZej
zNv<>V8*kV*QAiq-_bQGzX3KSoih_Jwh-eh}A*HHD*$xtKZeeOh3Q+BsFhx>!ULlHy
z_)*?w;CPm*z!;&Fn(1$<YH>tq%#7+wplCBt6aT=co^k4oW|av-!GA(1(NiGAQ>|5R
zx4iyaQ4lMc6s&^glMyjqc{d>>kC2d)LKhpZ0%yhZ{Iby#c#z-dahgZ4_Y?h2*0?IZ
zA>&8kDq_dV%C`~g3ngrFSv5d9vq=qXfCVe>j1|6kEdEt|QQYsnjPB!JeYYGVUh_zb
z?ZBd(5UbGL(cl`*-hWilgMKf&^?58>noS2!`({ze%F-H!_gYb#=^@kO6-_<zg$vs>
zoqm6P+CfHxLXj7o^cdr*0#3<`VCGb>#1*u+GpEyl7QrnnrbZ6UA~U>h_*P}<jJqDT
zlSFK%MLU5pxbyJ`cqxddnT|RY4P{UpBeFG_WpJ~*VT$q~Jb%F2hL8Ldnc=GFq!-F#
zRWXa~i4L6NXyIT%=(+wI0=XWEGnF3Evc{_ov?pL+n>{a3CfiB<%fEIf^{tEyC<6O_
zfX_Wt<gPOZYRp%4j`ajoRLO>7{Hi=yy#$6rn|hPdy_fKZ=2lm6N+9@Z_u;C|>nzz(
z27cov!tz>2jDK$b(Z}*CSl<doVr=3wvft}WEVmm*&fC+>S$QR3ArgJS+Bpa~4V#*%
zOKE2=s!?p5#G+1uccf_d=`fZd?AWf9GQz^l$9w@MMNAjqJd$YX{HY;@QEY-n?aDv}
zC!E2V_M!eByH08g<ktLRhbJX%IF(^kh`imdpljgn%YSC;oeZjAI&7=AJ2T(pdG6ZT
z%yYNSW`3H_=9()ps05vz>XV-5K{U;Ab|UbDDm(ThD7-TBZcZjX7}m?BI0M+7-;JXC
zI)@>?`tNmT;*Ubo9x|VH=_a3Tq<&Q5&jmRowQHE|iZPD(^Au2DqI<J+!lbZ4s=U+H
zCwul)41ZXmFrlKU=|d4y!pZf7ASW)3q`=v=Ag2#1-ubGj$J=jp#fA?qw!jr&M(1m}
zv#bJi=x!r+5zFi%vNI{L95?WGuRd85xSMp{xiRwkP8Y#OfyA$G(_de|zx-!`6lk50
zfR}&%`4S%>kUOADfKnpR=wJW3{PFVpyedn4{(nx{<fK+<tudootuPcGq0>TzhrQjd
zFyn<e=|bMtL~w*LMGo2t;<B=a6+lkIjX#?_$}@*}*q^@vDY~Ee085UU(#)g%utky}
zgj5TP3Fsn~U)#;MgTd~GjRT-kSWslJxTkLe?^%bvak)YW6f5e!KF~RdAtH{6!;%9j
z6n}bn*xivCc=lcVd+YKcjM9_sR6D6__W@TOsab~pKHd9*`{43!o!^C~_Y$+DEKj{P
zw4JPIVBdnud%8Kv^dyWt;D&2nakWE^<{IF0T@FeRopC?zakMAQX$6aCd7PF~%pM13
z{e1*TueHQ*1_nF{b>8P^IVO#oJdyoLj(<sG5TQjjquf6t^d#V82lhelr9lhl{ouXa
zL*C1|v);?ineXy$Pjg>*W#$k}>oK!ZPd<!M>wq}U_BjgYKB?|y2~8*%T}b)E9``uz
z_9Bqinka92UbpIb-GJ^{M{AyUO@0eYBb4-7KYESc@XE?-5oGHw*eluPJl9;ZOMj!M
z>^9FF-#hE+we2OCIq7Y;+@{Qxy>y&aD=X>kL+?56uUtddxIP<I<k@{xk!kc%Mc&yX
zQAOqGFsi6K;zZCw`wm1vGZ#l&_Vs@#;#lx*knT;*VerZdpvx2&m)>WbxlXqkjnI88
z)<~!IV=xf|h!{8kjK%HLYakgF3V-E$&cWJLS)1Q}o_p(2F2@p^?yYACIW56XZxj9~
zXh#LJFt9iHdrP3@(i@j_0nTM1(VCRnI1sq*%ASelt^3=TJmH@6yO!pHq?`92*6FG>
zq7wAp7V}Sw5BGI>s81DT?e)xVY8A6FrKxH;heZ)ue4Gl!R<^?bo>p~?6MvnXDJ@01
zAMwVhFf7EIYz!KwsVzyv3VctazKoo1-paENU=UMEx;98v3I{>hsv&79K&mAqbAtj@
zdx%XGFekrxgp1OpDuAWcsdG*|6U(n81>L!!TX)9edcQU>=Qm_KE~O0EJ)Jbc^or#A
zCX6D{dk^79?!DnlZ92&m4u5pT+`V-O(lr|_rCKqrg+fnjEpRAMn_1@{N4mkn4%-ea
z64x^mTV9v0pnwhM#txgN(nYUA=^I*f3I%)}BdF4enR|@(&7hj(Uj0`gq{86D`yjp+
zk?m)O$wiTj=9Ns(aqTIxU5{ZoI6-%DY~K6)b{#JerTJKduI5zaHh)M`pE3W?cfS@L
zRlOx1D}?NqQ0XjZXEI9TL5Rl?-TuO|a!;CJUpn(yqs-=-o>pKhmM*P;@bID3>$^X$
zwAt7gG7(+2HI;TbTwvJqiu#Bmfc;Sc)K)=tG8Q{m<TsG4V`8dw`1Wq7b^9$&Na`;;
zJR$z@gtX4LA%%F&S$|v@9Y;F5L$4h3QSff+M@*&v+ba6TdaTmx2)32yfxauUWz9?P
zX8Y>^`ygmS(U}GLcJPu}W4vo`SuHcv!s*z^GQ~mz>uc5F6?d^q`P^l9$1ujWsxyW$
zy<xl52M$G!jW(dkNdT(P&C!FiLQDKWl0G&@x4QCAt_qiyzE>kZZ<GKx@%mYLb|A+`
z<bO_@9Tb6kc(`k`$?XZba_`9pssE!w@67Oi*7>ZB5uR#qrp3>ebgy4>nrppu_=#oR
z&j1yQBjf)!I*%#yH=V2h0-Dk`+m~TrIu)04QUoLfF)%nXmtN5WF_+-UIS7CEuOQHu
z^uatN=fYtT7)atO4d9@Ll>%)IJ;?GlR_bUCX&t!#zTfa@DcM@3YiYegeUawu&LM}x
zxqLHZ&be?%A>+4&lBH8E6k(?*h0<zhgfEFx&}=o3Q6NnkxM*jWz}lQ*37Yy4l0(h%
z$cLa{rS`!%H0gW<p43VwF+_hU^JL&6=|hr*ER{NmH)`siLV#>E&y+)MHFvT=?!-P9
zQz%sCPirVF6k0K?fvroxSfaSlsen^1g~n<%m!O&zlyv!-K{j(z@sd%OuwtsleJUw|
zl6A2#3zgQ#g4RfKUl_(A<(WYP)!GLWsITQNp;+L=eT~ox=h`RM6c~R4cQOZNYCfBm
zz*76TN`dRFWaYp{GcWewQXe%1$@6;gffTH_<;tM2-j*@85Jbdh1ZrNcR*=nnstQu(
z)j)GnQzunH@;p9XYch||u!vc#+JbUQp_w`5dz}nQ<vMd}P$}Kc8eURSp9>1ss$EA|
zy_)p3SPt4|>%a)L%!_}8(LsQjfkPGPgPDX{wNHV~s5WmIR&R;Av}VDeoj#ieXseG4
zn>L1_z_1%A^>t7QCNMOZHog@Y5GcGY)-}*z$4dTo$sDY9#gz<J(^)`hH|5%Af*#zi
zoXl>5^Q2HZ=e}C#w<-DF7uVUmf|9`7QtBk$Na>OTKU3~%Dv5vHc>5~5UXpY#EAAjk
zdHx)OrE0m;=AM?h(=JfrK2;Z^z?NE_2Xikk{T=snkp8KAx-<LF+1X3is&ISs_1Du+
z|9Q8VzBxNvuEO-~^8EGU{AEHcNk`LPrXQy7@Brqc>BZt?^%71V-J_LapIl1BB-rR!
zb3J-A9ln3H^5K7;ou2)B`eqD%vN~O!O^4IpUi{$yzk0J;y}S6+H2r+Kdh`BOJX!ua
z{qp|TcaL8$ew%(gUcLG2yVVb8tHt@hj!zcxm$%cax64=4Z#k9}^K@~3G99m0=cljU
zuND{h`VY_G==rh#tbcs+<PrX1=BF<`s3}(RV*2|({^@^;jRy5s7|hxGw{MSbW2@0k
zK`cdh-8QW7?r3>C)*?~sR;&mg(RXeOR=92Sz<<RWrtg+-m*<D?puj->LYSUIX^Zo-
zQ0B)!e^`BgxH?`fu)Y0*gXP%@lN@-EOvZhd1GlBZG(X5}WAg*NBMS@@4^N&<&(D`9
zhl|zA>G^-tgXyQmht;LXegB^y|Ga=e*wM4qMUTt#8eS|emhaC`78jSX+CTeo@%r@m
z+vSIDgC5XpD@PdN9PjoZn>t&!*Ne*>)kS#pa!ymemN`wWDH)Uwst(Eq)z%cJLFGY}
zLG?L~K6agtEPt~f<@ure5v3A60Ew1}oPl(s(IbBWvO`RAG<_2=`r-8b@~7qWG<>D4
zAOfdGQjwhn@kwCl*ED<=8P>jx@ggHjm0}U{K&BK^+aN?ONErBwU|(t~rRx2K;7k$e
z3waBZI~aii3R1DK#mpOIuqLF+$hnBCNUmuIiXabP*o=(u0^@HKJvAf!LX>jH&DcJc
za>IX;J;GlGQ*MU&+e1cENghpAuA=F^EuAz+0+Lx<I_Q`Z@7L6c8{AEXsgtyrxb4@}
zi7NbUSUPcqyY1Q3DK(#T*gEBkUC}RY>!s!BFnDe204!0J9?;e$D{LKDdIz@7$j8(H
zmPG0RKi1k4Xl#0Ohk^PN5FsPr)>5z~&4zz+d#Yy{sUE{@0xs7$)?h%9y0>8CmH=94
z8&=d!5%gB9zOT7$#R@k7?)~DR<r+@;2*G?dXwh~Pv}nB>XsP8IXpsReWWX00@I?lE
zkpW+1z!w?tMFxD)0bg{$7aeoxm_x@LDjOip=TrkgEnZ}-%K=a3#(VVY)3evuK8$}o
zdo|nRXG)tqdizBo*um<rLY2U7HPOQ@s>$PLt?eL8!F&*&7=F#sE9)ZIE|k>;$-qj!
z5vPECIf)ca%}{z9LdPyr5%lWq9R<qjzY%0m@t2D5U^itz2CT+Hl^Q9+b>h4m8LFox
zPLU9cJUtoQu>dDDyBV>8AAr^ylCppI3{p!}?=OU}IC=Z3c)^EMT%va|3Y;jFo7jXa
zOX182F-Xb~CG7y45HeP#JY-NWC~udcni_OPY~BN6VYMB|06+w_u4UkyV{?&ikpaG0
zb7e|W4Y-5=)My7F4c^}nEZ0J4=_&kHMW~p<dQ2F34dT{a7$uOhAwj|rpjv+zV?_Xh
zL>^=}cm}7E0n-YwR3#&!!aLXu5@_Fs5Y$@lM%*UF?_rsHYa%012{PTKCUQgEYGqT5
zAb?c(2NIBF!`9gDG_e+9t%=*D*iIAM=b_BrU8*Aw0R}|Mw#PhlvOK$3ogbe92hEtd
zh?j@$lU|$JR=u{5*e)S!Zn1x$GTMeUAyB2QMrCYN-2huRux7T%A2s{-+d!KRye4X8
zhKUE>Y<Cnre<S`)o*%q-&zBv*17?4ycLKB3Zi3mWcLTGQu7Oz{AhiylS^0kOk_id&
z3W+H^LYN-V0}AEG_@Vbhm?-XJx$66P{;QTBa#;r4O!=O`83`#$xfXx0yunfS8L+-R
zKYji4;=&F7a(&aQ{?WY3f<=CMP*CgtDz&e1j9|n$D%WjTNg1$K{(oBZO>PE=Mt2Jk
z&8`KAI_g;))v1l@)JD*-QJvb=8zZ>bsDRDx-;PJRj(#er(xM+X7nNxJgwy+jeAkNT
zGeR6Ppo9Jq;{2S59x{K#RiyZKzd*oMqubGFY+H%i^x>L)gz(0+g9_TK1HVFnLK?}q
zu(*}<c-}Ug-eiRa5~?*A3gD~&r?>lcQLO?5@wF~0MF69R?4pW8LeX8cC#a>E{BB9J
zm@`T+ZxN&ZK_smkd=Z6cL%pii?}Q*IQrksI?)`uOs2IKNM=*b6>Ni*r%tbeqV3|>S
z<bvJxn+|Cw=LJMCvNEI)KW28Ec$*X`ehh!a03`9UaJgTz-pemP*S@MeP!s0|ZPt4a
zENh}qvUlUHN{^4&*7byAFy4k$T8UY<;pB>BzYQboCDyIv%zj9l>pt<r=RyX*cJ?V`
z@JlCmfed~l=L&x^*l2CH(b{eUJhTBG+M3U=<E*JXs4^%T6c0)UrGu)2vO%>Soc^~O
z{dSnv965QHNh0IXGV;jH{%d1=!(7R@3id4#?JUGOgssr!I>ae_o~!LS#I2=Rhqzm%
z*ma0=oK?lHom}{#>_eWtBOrN*%e;$5f8$83K~i&I0$6{GRc8wVv?;Vny|n_K0A8d*
zmQqsNNdpbmj`#xX6RnWnob1OB>`fQZJ>38hj>WII7C(IV8=@(Cz+%#fDfV;&KqwXl
zq{$o`BBOYV+i#5)j>5X1D*#lm`G}}MqP^u$MzOawA(PP;zX9NwnXcV-Aw>vy(AzGW
zz1C4KJG_7GQlMl;I$Ga$DdhDX?f13|k|@lI$88rFam*W{>X)(g(z{uJ?b6N7ue*4W
z1gC=|ZfM@4-j-$#epunDfcm?;?&6SH{95hhb(aj=N3H+GjU9kBHZL632pSb4^F!LY
zpNvC_PXM%=?C-a~u5%3RjIf0au?=hY1KNt!s|*^`w$FPZpN({>+(f#R?uK;di>gna
z{1@)u)|YV{7ZU?9F*29I<Tn$SL+U3fm+ieem4AF4^!|^RU!P<${o~2OmfgmpVC%Bo
ztX?A9L2NIw{aW#NJ^xm$fgoLDD~PqBpYKOApMzi-B^lw|aUWa@U$se(Ch_Fv^T82f
zH5kQ4<JW?%ObGH|^WkJdylk-E?tw7q3#RI*dbVngiSQ<bB^*8vzkYdn`sb+xM`=OE
zPk%;RLE2A$etQZR5J(v7C#8f3oWB1xeSP|El$zo`9wUTeDx@()T`H6jL~b&ug#%L{
z)P?cp7%rQ&3ghgtbfSPE0FoG<sAx9>FQz{nWjJegNvqHIM5w6QrffF`KTgrE{7eST
zX_gkjWb)MjsVIYr)C9)jnIh*DyN3L6nt$~4Jve%H_RBi_m2tvJ8vIRBj9#WlQ_2E1
z;J+?23HU=bE{9)R5D2z6CyIiQ)VsN@oo7JzYIMp*^z@j#EjhadhgZ3ZP6^}Of~&TP
zc*0GTo9Tvi$`fOPR4nU#EO}yFfQ*_)ER#3FB4r%`o6)Bxo8@@O6w+WVkuWX$nt!h*
z{LjA-%eO<4?vwRUuDBXolKE<oU1a8Gqb=nu?NuhUXp>H|Onx@oPb&XHY&BCu=25=P
zb=di;YGsx?Rp-G1A`-=ZJTZ}N@C#Wlq;~~-kYtw}UxgviB%H;Ghy;0`q&-WzuJs`a
zbar?;+7jBGPAx39V>7qCzznF;Ie(kJpVIw@(%LtBmvTbmX!BU)t9->a8-TaktX&#M
z^i?)%wi%5ZwZ^<Zr_5zE7X(Ih)aprSjYzc}5+UKNj5`hUFV?eG0%RGn@JhChwQBrQ
zUCn3v?y{Nz$*4gyTCp?~)&91P2s8qCoEaqMXa$L3I3(+eo-><|$-g)P#(yCUtAxOo
z+NsXR6z?cyX8p`6MK)AkDO_V}kEqWVu9>yl$_9luU}nOIl4C|rmyhHH_E+-kV3TE0
zL*d#nZ`q#)5o<j`NdaiuSTEy-Wy;uRa)hPc)ls+j=gWZk0q2oj)d=!o@Fdm1EbA8h
zMqA`_9kN4I!%?0#QOk8MtbfAMZ*}T(t8*;pezD03$Rj0Bse{BTc}$bz0a%~eb4b3s
zX8Dn-=?2>nBuj@;x;*9xt}<85Nad<cA2@O1JOJBNylFPyEi|dKkRE$*<ltZdXDWzA
zp!0YIzLLsyBnR3FZ{7Ixt!+kr+9p293n*R#jW#YW3sP}u+%-_hMSmHo>ufM;he)$9
z#LRcw6hGI=Dg#M4qcX5{9Umk8&G*jqcemNx7;d)aEcOo`&&deoV6-B(k{FLg(pIol
zT=R2nma<K>pT_~%MrFb@@<6Y&c_&nY<HbI&!3Hg@-EpYLYIh{{j!MfI-YpAC%-qn5
z%Wly5vZGW&p~AG_fq#Y~OABg;uN?nG$7R`37ffxNZie@ajd9FItAs84E#sIyj@A9$
zb#R&OJ2iuWE78nR&^6vMf`{4?NtPwx=9s3G;E8FpBH$tqzj-Gu;^Q`^R@Ms1rv~ZK
zR+HE!4}ld6ZT8`OsYsg@eqm7vdAa8nv)?XN3u`t{My6ZpAAg9RtHAfYXnm3Vh9x0t
z;KhZie7BEVcncduo7J18ffl|l`nBF`+HG%2C7)I{rD`>{OaPxYD~3s%7Yl0@bB}J3
ztF70xhUbe5YU2ecHA-VIbKP>BV7Ik#5$`(AcUA0H+_)=ovmIB9!aMbdxOSSZneHye
z*lD5V=tCFdihs!%#g!PZRB>D7cZ@iB?uzaa5$`nJYdhvrR+M5t(T&=Ff~>U6(|7Xo
z>%AR4>z|S>hjHdum=!jWfGD^_KLQDzrh9F2U5$iZ`~L?CG0(bg9KRk3$_Y;EBaqN(
zx^;@X$jIKZ78<pT8Bd+h{eX;@S0L*dkM(WZP5%W8uz!2KP`0R6USgbc$nsn{uYCEW
zc2n5FFwXGNc&X-$FJ9TodYa+1Bq6)q=AxYPzT`4M(@SmTNNDZu{ZhL^EWX>jv{nFQ
z&)r~5^R`;0td)I-cGV7L1Nc>Mt`#~avj@GnRzw2e;tVQKPW4YjNO$H-yO!jD>QOL!
zS|7a0aDR~fk|_^=hpv<qvUx**-2(Wyoa}_^#e#f?_uql~445-p-x#k;E!KjNa26r#
z{<EN2IBwHK*hbWZ7ht!VErytqG)h@&_w)Zd!0s~nzjpY32NDH`!b<xF76pTGoA!<@
z+K+Pk@?TF^?!JURje}ihgnEB6h431b?1R`8G*Bl_<1H)CTbR`_q`V%rY*B028(_>0
zPy7&S^`2SF{wJ7qF!eFS+ADn}uR0Lk!X4ggfdQeJw0eO*%ZA50a4r**R(J#Djz^}S
ze*qV4nYEW;U^)~5GMBO8I1>UfGnes84JVhC#X6OLf%bpD{;|^;`oqb?7H)Q=1YN-`
z%WD!GR`8PGHz>bNdPiBoLT(cjR#u~s`V*%*fP}6e7g5@M-5RGUdf<e)&gI{yJmjgG
zso%t!(b0j@o*M8^C+-dBviU7$xSS!=0AGwAjidticIu0l=I!})_usBs3yrcIIyS7F
zbKO6G-*<fjfmQ56ho}Nj_v7F0>+a8X*6fJ3jGT(T(ahFPe1oj4^&OK@+EVq+#I9Ex
zRV$KZ>YpFqx~AsK%dqd@mG`Da0-yV5!n5ic;gt#4j_C152lP0PsWfB9NNIhlFV3UL
zvn5AWZm+~-@s?&mA1a1OfI<;EF4H41xkO-p#1ZOSj*5y3t@qelO&)KoN!Faze%Xk_
z;1mPTu)fn?XDk#02xb<FQ4M)!UBhG*d$zM|R|U&A=IJVSZB#JrqC&^wSeb%O`%h3d
zq(>Wb6Zdm<EaZIQN=>52g5yj3%t5uK5NSHsw0(-hP1;Wx%Tr0S<27T8uPY=PN`>)%
zLN;UBvy>b(CMqtrY|*i}v?mDG@U=Wz+14OgN&rPzq?`I^SusAVuX&TKjG^B_R4f=p
zbSF)*5R`+fV|B%%*Jw(E<N{4;Mtq2-keAZpFQF+d*86Bmv+9+aLM&kb;l?~Yqbbdz
zzmKLw<D2w+n5Ga3Ej_<OQwSvyu(_#!s_;M-xlTd8bwxhSc(%<ULgf$|kay4z6SM+x
z!F>e81XjikfjP0?S>wG)D4gjz)~>iA6yPE9q#5xcLLmz%1o|a}qQ!b2p=egUQYegs
z9eTWykE>pER%UnY>|;g9YH-?Cgw)H{Bydwzu-RH0st|7z(MKByX#7@E)EW<e7j7&w
z4T-HkOGgZnG8EpRBM&D@TtN|{=B1Bd7)W_HeBP*^=H%)9EWXWhQ>lIg^Yok-D-v`Z
z=l5;$mW3t7X}YPP*$K)hXni+(+!il6T5I0|8jDxhTk|;bG0!($1gsI2X#EX;qV0uq
z{N=P&Nt;8Urp=zL<P$6ko3O5b(^Qnwqi;18SdNLjt*IAiD4JfehuihUWyN1od1%^v
zisz(&4I$GBHQtI(d79@`t>xg#IuaqLkTSt#wjQ`w7P5!gtcQ_^<qsEi70M~1T_L~s
zU$?htWfsg6%az#E$9Kp!1i1-JRlA&<Pce#07M<m3+oXpBu0gCLcsP-NC)earesa>c
zqv%~uC&;8&*g6Hj?3e8{5Ue6e#?y(S({O#~wCTVV6}c?Tw{}gEL<21DdOO(>$O&Dl
zrUP^Kni0aW(%LoWxSO5842$l~Rxv9X;nILLQR8Dah!K_#0_~^8a@jj~U}IFKXT=m`
zzDOog*gax7JoJ~#C>Hg9IAZOHIDKDu9FRnc)c%nomKM|E9{4dCqHLC)>evfSGvFD}
zQF|;(H9WmIr@ajdOtohY(Pp19K;}D1iTaYGEl61F;Y-TuEGrpJuG^?>J4-!QZ;LVz
zD;Pe7V9xq6m*^oOMh@_G8VeGJ_|r5tu5!J`&P6lTV7_CN%pQ$MK#3UVF8H`2l=|@I
zyD<ioMBQ!@WMVM+;A$4H$VSUKILt=s<5}I@=8Kn<I{c>+^R?ykAZ}F@LXM0#o!^=x
zrSR#0gSue&mtkN!6ag}qaWMfC0Wp_=GB_xgJIgwef1+@metZAvAhYUHcoTjNxAw2M
z_rqU@YKTHX#&^b`kk)m-e>;Q+8b}!HJ21kd>AwHbeLei#N*%h%0lP^WLo{-eGJ?pl
zbT2SGN^W9(B`GO2yGOGDnhhns%X9OLjn-E-?sK|u^6V5nc+Fzo(y+XqUW^P+>cX&y
zCF{?5e<6F3^5*2zSnxLSouGNH(pWDwJk_kK)_L}8l2CRF5e>UhmO_-~7O7i1;f#CY
z-Q+!6%`PeSv(Z`1;1YO50lv@RARp5eTu+>p8tpOf8mxA8WTYqHSuN=pR)oxs9>G2a
zbyfJ=7P--P$)=oG6YW(<D41wJADJv^-VZ22e*l<PQjD8gv3)b6ip_d8+9P>>)n*Nv
zi)o_hv?kWfWt{}LiMg>IITf0ZJ%RDZV!c`RYKiH@HyaiQ^(I-UKq#bP9OG=+XJRmp
z22CM0jBDwcCGX<Q>#85YIU%XtGNfkOV%2|2YGIX|D11rgmXeQ3>J$}NFj=BNiKD`U
ze_Y6loKsxMnlFch-S<$}lhI$kEE|bdpf>F;xX^j`G{|uL#Gr&0reo(#-p<3XwT_1;
z(V`>C2~PluKy|;mRWOEtP7rH?PD<t4;NMF&29-=(nPDy*dreYh(Xp60-?QSAwAZB%
zqJAxV(GIutx-^<qchwP->mv8pWY-{JY_*YY=zqL!JL7Z8kq1r+duQtaZH3~(X{6-z
z{&eMYer<1yf>A|H_uz-pQ`#ywbLZJMh9S)jQsD<-7><;1_ac}cJUj!65Ca`FP1DmU
znIib`DX|BQX`k5aclM<eQpT2(TF+=CwjD3$oQ0%>)N1Ukuy5%S1sed5LE9<j1Aq>M
z<A1P@WU~743J#Z-k%N~UV+NKM>+^{gTiODuljrOe9n3F3qE|*E?3H@Rdcz)TyZ|g0
zFw{%9k1Q|i2vArUFN?;l1H6sGW7uH&lPaD7<p?baP>YX`Q|)G;-Rs*j{0uEGtxCyO
z{+e@mg~H{$v~{*W=?FX467<f-pfw0DCx7I|NMYG$k2VJFIp@`OXBeXZ&8k`IG$*$-
zD`;+w{mfFI*<Gv@(Ce6>cZ$kh(t>{6uj=8&J451OQOvHO2ezBToQ(-*L9FKm1IP>h
zeI4gNV$i9AhicQKWG;!kgU`+Rq2!tl@Ll)K;Q{8SVVJjb#1El@jkS&e?j|yrJAVf3
zA%q#+?;wPZ2R$`LE$7KPCu9$CVK%}FCCpwKfU>Nm-|Sw+>O}Si)MZb}RHYZ_A)_DZ
z^_&MTn-*(E<I1+~TVO8ec_(N4%y}Yf0l^5k-C>Ef>}@vBW-f*ev8|G8%q68v;RiVh
zR+vjazfY6B%q6egpHw9c$J!0p%YUF$H`q(wd%IjYf=r{K_2G?-BtRRaNhEeM60}AD
z>5pV2iqBmJ4Gr$_D2(<G;|{JRLFn%U9;lTdw}v|wj$-5*HL4;OeCwM`^NKVmv3#U6
z5faAqZzkTxvlyI_#}1_zR@NYGf%~YtZ;seY4&nsp*Au#)3y(7(Wfaf#t$*SVav;SR
zQ@d7ZAwUtzWKTqEGiE92I|sJzpfwm`LeOdfdX~(f(ZWe;XJ@i@^WN5SMY-wTtAvUj
zBM{Y!4sE_E?}XL#<K<hU_;d2D5&AFXTcbtusjJPmX7+>m*6v*hyj#t;WgXfs7;vj~
zrh<rN=&$&=a?0+6{_95{gMa-|@BPQ#35-y49KjtQQue+fxbKh#V}(chptl1GoRACe
zT(?hYC0Vp{AN+a%JR_l=T<3Te$oTk2)Sg_!vJQ>}Soq&Fe6U(5P)}j_fTP^|k5K%^
zss}lKBkczv`Ex$l&+?nq3p5`Xpu^yc%`<7f5_Y=7^FB`2@4!Q1eOwd8*WCPFtKjO#
zNjLfg=(Z?+KR<q)bhl9aw076OAh==jalYeO&!f~{>&AxRXH2*EML@Q0yQpijNylW|
zKlo{~k+Sne0E0Hw1FJpn0m^?_<vYWQzX4lKzW|qEU^)~6GC7xlk2MpQ#nC#ye`Ah+
zz5TwC$#f~a3BQJ0|JUo==D$sGM1vsX2XaP$F~i^AHsJ*-62|($gD~ia@Ba*6H-Ge_
z)-XB6FiB&G8kn>ZL~by~3ka_!nAl!PT1sSikR3pFl=v>U(TUO<*)~gqAt!*RT<guS
z;GQ4s*L$KB2sZbG(2Kj{r___df23ps`RQ2zo%l|wd#|Tw^vdIy=^f%2A{Y!njv~<%
z&EQVFc4?TihMnRkI2KY!c0*_yz6Y?ICtp4@*j}!I4bltru6H`j94w5G;WV@2XUaAi
zb0~xWUTDsQZd7(5UXrt2?hXe4j6my30AAZxMhfyU7)^qdFRco^9kwQUf8kn;W82;I
zJYs35wzw@Jim<<=Cf7r)jpnw#H~dPdwbue&KWIi7{ES5O7E?0sQ-<YiOUr_`WkE@S
z!KeHqy4pNpS4+G@Oi1-sC~RDqTcdE2`qouOTr{_}T#Vdu?ocoJf^{3b6ae~&4U2qJ
z$n*S!>DJhNqik!&{HSDWe~eIcJlC4W*QQ!KGo(y&$20Aa1&i`0nLnk@Em1C8CWa)A
zhC!IeOdhQfE@S2-IDQQ5f{BU5lSF@4-+U=quoN<x|CwWLnVwSWN5-O*^6QvnB}B{<
z1`5S9kR>U@!U)t$kQU2sUO1R%NC`L1G(|zDq>I+rX<^aaEDm*oe^y~dX_Pgb)@suw
z`gln70fQx-?tFdOo1olm@S3e8e4f-yaCdTLEtY3UGKoSe-Agj9g;#i7lJ)m9BpRb7
zv@1;AVt)<rlSEozzY6@iHwQ&{0tzi_S8Hn1c|4ZLapt3rgjPByyqjYdTE*5Sj=5#j
zs=;Dv`lgYoQk7&%e~XvI4%z!{(-&qnI^i_NWTIX6A-2HTs#1~nWt=9@6>Rk(oEy7>
zmAdilm_#+PdOd+N%f3`&6h}!58EuKg61h|)@U;@lFI|01G3MfrS~UeqR-C=hOOWLF
z6fwI5jo33=kmDJx$K*lWn8h)D`HMy!{mOZ>uS(eLO|~l)f5M1T?0*{X(y~$-u|+j=
zQ~<k?8RxkwHMb`SWc{yZDtiY*pmr>F>@69hm2P8R@3nesB8qE@t(I#?%fEIc^um#Q
zm49GG01h8i{=qw8&X&aIgF2QK@;V)dWQCbG#<4c0^?7*ci}f9ES%p=$^GkgS*J_qB
z_X({{4}?9Tf5lF5QxL5dD_Sk}V3jpWZRsoM^fcGpW&Oc4iYsy{S2ZkEWIf~a+g)PK
zGSghG?DNU^)cxXu9hFJdIQd&@XKNL$rrKy?Ig_J~ef|7Cdx^7$C+cWgfQ!h40}bsP
z)c{o`uawHX^Qo&jM`FqgO^!t&XCPk=2&T+jJUv!qe|a~eZEw=SRe13kQ`(ev`@#cN
z&Y}S6dSgn~*qeE+z2aSxIoEB-IgS&vXC+q@k?A2gr0-8bz=I0t&AKDE0!X2hx`JCl
zDc&pGboT|mJ|rEprn)n?wvOD3TeT$t9v;N4+9{@#Ilp~-`uu)WKbl-S$5z<&s#CAT
z$Y;TKf8JTP0%1w_BR1$=L-v|0hOG|hes!{6kHqR`sUrOKUiT+_%~ck`{YZr0nRKu+
zJcRH&W9~@!XS)4p!tczwBjIZ<L~8z2@eaaQ9tG#cgE!NceF{0B{Jj^5wsY!U<gdsv
zpQ#6tzv2y`slI^T>|@g$pyp#!Cj^du$9V_wf0IC?d#8?Wt3i0H7BIZ$`_g`VGO~PL
zTah=X!MoUF4FBLE@dA(8AYR)GjIKUOP+Xm?p<h5;>}wXAPHWSyzmp&66l}4MV>;Va
zwOt2=Msp5Z6S{joxiVIL`2kKX6f*~+`H@6SXVS->T6D(TQN*0-_M=5iXVx7>j5k7Q
zEO&*5;q+mla#IN-6WIH5W-)y)Nw&SbPo9tw0h7kwbJa<w7+8_~51=C?k(Xf#I~4&j
zmoal56PHfdI=_E@uk7Ksm!Ixr7F`N&!q@OI{`&NC_xD}h2xvjZPX<r`;Pl7$yYK=P
z31j_4B&=7{k3XlccR!C*3$Nibs?8}@n>2=~QyUmT<R)XZ(AE~U@$Mz@2Jx9cS!KwZ
zBHk1MKOfSjk=2T<Dq<xdk0H;ePf5l6TFR%WfXWGZUzUF+Lm*JO!ijvQLWxWGj+8i~
zODYt|B<i0Z-=<;Y%jcPkwZbSha#0_{(;Y}%;fzo)F3`JT1>49Qw|V2N4EemiVN?Q5
z-LJ|Eg2q^zXp&aL82bDi(|G)JX-zl<G8gLuf;%wxUWgTSNWnHIg$yKs647piiN1gK
za?OYUq)>k}?lh{<<sI5=JJd1VvPwWUc78Re?o;0W8ze<9ydKpNEH+p>>E7;6Nokb;
z>24sUB_;?1;bL+@m&z{yW>ZRUgk{X^iOJ{LQKPdCwp(C9OF^=tsboYn5+1YpK4tTy
zs4{+|9$N^=dRRlKO%%`IM;?XN^VX5^HBm3dL>Yge;_vV_O6AgwXm;~xG`{+a4iG3s
z$uU!Lg>vD8Smxs@)YW<MP+Cq&D|PjAnUkf8n9|Zt&ad<nZaRIT9%lNww$n-F@u8Y<
zS+x(ag<7|ddH1yL#6|}ckq591hZrnetNo}UpU*|jR3bM>V;pyT%gYvE4MUl(E?QwT
z#AJWl>F)5L9#7~mt-$v5%V<1zJ?|2aObeP_#0(&4jE+&#>}U(y1qz_0N6F<=-mA;<
z==pmL0PB$?d($Q1K^FPeeM>tMjw9V5A1%Ny>169%$O*J_eua5Qdbaf_+h}D;*U1wT
zE&w?PAz<<2H0gwqi_tHiH#HuDJ=jsC^|*hNL~T`AToN`7^boyE4KTZ|U9gN;91I3w
zDr(rP?Hat|yuzHjZCx-9lZ{0jz<HZtey@VG?aoclQUMbcdK4?05r+{cwd3t+GS0+c
zC?Rcr_08U~7gI!DtLIG8G>%5u497^$G0h9)gvG2C55(D$G+1S?r9p_e3^VLw8H;}k
zdTru62Goe2k<;dY%vJX?Ju_-&&t>tPAa9Lc6Av9HxATUzJ*Wsybeth9=ZKrSE2NAU
zfY9dq)hHN%G-TmzpLf)7Um!mR+n^!HE}sKwG$NX(h2^ncdd}6ATZtUFNEJ8zjnNDi
z8~N~D(LQL0aL`V*#tVxv)Jk~v(4l`A0}x)})r&FIjuFAP7&|NWEn3Zo@`l6=B?#&9
zJkbshpiS{i*VriK+4W??Ok0wLbohCsO{wx6GjA_G4Nv@3*Fk>sv4zD2RcuG2JDNIY
zNG9i%A(_m-+5|GC2<=Wb)1JD1C3OlsLjnL4k}Pb!?%1Srw)wqWHHV!dk2`;)5d2hN
z=>YK9q`qc>_v6`fgb_+tv+Mv(Cw=ZD4%&i<u-jBKwx(vslBl`0G(M$Z$2k&RC3#R`
z{wP=HOgVueDAvVew$VSHml}X&@W7GN4I`r_<irRsh>T-=&{_*m*9;H6ViOWU<0Cf7
z8D7Vw$I~HzKqx!NdaDQ!%4mQ7_<sTb*IrCx(<V>rEtMU`cBypd=<@y%+k^HnN_6c8
z606gXJVH`hy$>(%d*mL{KsA+CwOOFssC2n@a--(P+Hq{(R$W;?T4`ZvA$u+&UzZ(#
zF~Ujc+BS2Mz1Fr#$ev5&7|m^$7MB@VV?&P*Wwf+L((I+(S2}ZC%#wdCt)MOWS3E5v
z$SDj4bd01mT87%8HjVu*t|>%5Z#_~NO36+BhNBFsarL>ptz$%U3*M&iuMwL{VZR-*
zYvZkn-}4Dqnzo~)x;2|8S+c~2|JkP)Fd(0P+YB5UL1Ts$WjP)c!fk^$Xc>FduMgg!
zn45wdNAEK-qs_>TRv>@6LC{-;ZUtVDGGpv!tm5S1+M!!nFAVtZ^j3f`b@yJ8Y>u}}
z$~fLl>!+nPv8c(~qlPVfd&Lb?h~2#ySHE9EW;!x?)BRGq2I>jMn~PVZTKT49aWZ6u
zRdXlsyne?;mz_>Ue;+vQU3PK_r@1f~R@|n;uuuq^m~j+^l=^?$Bzs)1*jo&~TtS|r
znA0QYYnx}zkq0i-d}3xZaEoL~CI*pPH`4B*0E-Sc|GG1$rm?*KZw6=PZN9^-n6^2y
zm$t>X*tR!dC3Q=)BDInP#x<2fP21A1bNZ@@=9w@TTMEuDIkoq6iQ?RNQgMdN0gSVI
z#tH+WE8HZ{2HAh}pjF(vzwRHDfA{-+UPr6ztchLAcHOV6@cUh6&T>B3Gy9fTSi8Ey
zYato>dNw5Y*>HvHsh92FT{fW^XwZ*r)UFDUV?H{#Oa;h=wsJkEe~PL%ZX?)VV6f~t
zg-%HzN}Utanij9;W@t1gDmaq69Kmwnhfod3)jnD1@!Nm?cCL7zJ5e@(Kt8CRz?|zg
zHPktdHrG`rc9iq%{eI`(;exS3g`1CA@ONK#4G*jqO7<&G+BdN8g&uypXy3lUv3*-J
zvQhW@7ZU-m)o=_OO;O{8Mb-9su5~V5KNXr$%6;JS?*O*Fpe+tU$IC6CgZK8*e5?_>
z>VD|)U#l#;tGNme6~>_8Gt>sB4f*6nD3Z{U2Ie(MvW@ns`DcKp#2M-iE^Gb+ROe5Z
z(MJOl1Tir(GnY`$FC`2)HZM$NZfA68ATl?Xu`V_zf8ASKk6X79e)q2+(3kYVJS68r
zP7wsirA`Als1c_?TVo&AYbDv#Ss80>;Qaf3!?U8+kv5PmdDn;y3!@{RGZ)U~n>lpO
zg*c>;bIJIW(y1IulTMYuxpk@qk=!W@+SsX*Ld&&NIT%}~@PyR*V+w^_tUYiN>m~O`
z7JMqte|KrfwRPqN7c5gJTPTU=wHC0bKa408t~Hp;R3vM#YWP{i{2G4MRGbwhkn`=3
zw^Z%oONC0clT-tx+__Q<l(};y3sh>qjZ+{_eJ$re!d-1Hfwayf1!{?$8*?C|`8Elh
zbS5=$nT-HVnP<(p1y-7`THvaku~1-}B@+&ke{x?S0&=#uC`e=HCMig!ZacLgP}#{^
zf>7<G<{-@d2{4<;tPmJUwfQPoSedU%&^+s=8no2Dr&@!uW#<NERm&_x137bpy;ZI6
zH7;m%ZVB6`6)uCFV0cIA>cH~aW{W{T+|wDf9BNLT1jm|5J1+^<c{{10lJzIRH>%M5
ze}oEhwyt4qP5mn5s4aK0QGkx!r0~v`xl0R~8g{Ug8X7loZmou<v*)%JTANuk3(aP4
zvlJ{@f0)6wDfP85p)qyU@taAV*TO0591xJ&8f1VOV2h^u02F@4v-}!-)(~Ol!DJx&
zEx0iyLL_9us+_Vo1HxjMRXd%aN}|41f8=^6?t4YaO|M-xO4Q@dH;&x8$%OiK=W_kb
z)VXNCw;PRJhSar}GP8A+Trsh(l6Kdn9+@hgQ}*lCWx^AEn<~?TXLABc;k)YuvZl(Q
z+MsMrm2pQIca(7l47sLwP%<bTR2|gdQ96&42cLdAegE65#q{~b#d0-0y?Va#e{YY^
zFa8&%FP4`t7MIUlhIBIhWBO+L@>!n0e787TJtL}-5bG)y8P+7rk&rFn{xLN@nLZDA
z{L|^{<@d|!tMG}MXDGOh)XaiNuJYH<@aG^QvT$^*m0L`vp5~%^7d9heFSdoKh>wcD
zS`pF-#?ixtfcDXIp^`*Fz)Q~2e>@Se#9Vh+qBgjnDs$pQ@a3G2U<tBVJ-wvrOi?c=
z*5pY{BEtA?wgc%aX2h((2cTr2N--f0kzXU1vLytys+O2M-FgHm*%2&JEFy0S3}T)S
zU<iVX2wN25Eg85WgP`ULEQg><X6<kp2+RekXDtL&jFfK|Vn0Ld=8IGlfAY}1M5!of
zg`y)4DbW!2NMF$`pFDU=jp_68^t09Z@&fYw>$`9K-zPt<R<B<FZJK^OU;T9TJf1Co
znSQ?d<<-L%i(jX2Pgg(v^VRB`i`C-thtsn~{Q2ed{N?ib^lOeK#XP;fJey8etIPA}
zSF6Qq{QQ^48Cl^WK8>K(e}nheixU`l>b_PCkRqz^Jt|g3pcv4~F2&x}Ydn=ATu^?Z
z7bQlDP0*6Sy~=h71Aepyn-)?-x#D&N&t-(HS8=4q$nHss`~jE-U>H$Ow}dDFb(Ii8
zjaibSnwDdP@Vwn6A$9>Gvk8O&vdx72vMZxHoQ23pHiqJ!T0TNDe`A7=4H<|%l<ig-
zxFw)%tZWd1)imotAq>v~pE_JUBe$xx`*ihO+|?-^7Q<c}9ByChRW$-H-lJOBuZR9A
z8@0Z2z(yT(P5=JSe;r3g;8lR|94@Y2zC5{YS;aDBd1&frWRInQ@(As=&r(=EHe`p*
z(lC9we7U@Q`U<Lse;E!=PoUo7@*>PQar*n4)z?o~r>jMnA>{OEd9ebMN8~8Db>uib
zLf6szw*EzWbI}pvXP-?^E|+Id7prH}ldm34-!I;*PHyb{lhYp;kOAhtSiSD;PiKq$
ze!aZ9JX_4yI*gsZUA#Cy{bKp1JBR}L!U<0BgiGApLHN9^e>>Cb>!!4e(9tsu6iWle
z(m={QkTMUX%mXR&K*~IjG7qH811a<1!Gi}M0V$(%!^Qy*!i#_j$N{o<k3eQMZDRqi
z|0UrM5K}^f#H}0w6;W%BJk~-WOChV3TZQ-_$c#`+*jw`?y>(!qbZ=cHd#aN=f``lw
zji?#g6+onlf4qR6iwyNH+4gQA6q9f!QbKzJuuS=QLU>5~6ryqTsR{Z9-rfroSZ%F9
zd*E~~sik^62?PMD`Wz`XFLJ6Vv^%WO61VKrC54e1DLV7x$-thSWg~>-h(>=)2&j+t
z8Z{P}2yGMKdOMyB1##d&qeUnh$XmH>3E+>&zy%E(f4T&?Bud9YHJ*SiV9ikrp%mEP
zC!9sHK{14CM}V_HNl_iNwb;x@km!AKpIJuJvMcYEf}Jg}1f^;KK=m*L)hLQ>x}a>(
zI^}z}?u^rqWL-F{4|1<<>p(?z#qY6gs}y}=w>GU~x?Ddl%XS<k8ytm_y9>X`Hq!C|
z{vOMae}|C0_gb1;L6*rMHFm3M8@p+HKkQb^4eX`^lj*=@Ixv|IOr`^q>A+-K5655E
zk)l8ywUrSoibf7e{KLnJPs8j+u%Mhr{@9}BQ0uaLIEwe$x{PEinLoUBnS09HjD&<@
zYJV={`d|pDWEhvU?wl+i-MCcd(e?l+1&N;Nf6roF^5&=X&YEiLZx6J?hLrP;0S|z8
z<wQwM$c#7Wv!{Kvz2*gPExWzm>{up>T3^aOOW9vN?Dtgi=Bw3xmKMZxL#g?wtC($9
zv0C4+inV;Biq%o4>Znsyk5HXRBf$p}e9uRoXvoC-a`fk^ZaHDKgf_TjE{wLYh&4UN
zf4d{J9o*mGF>GH1?K$~y9etR+Qy?55LXOojdqx}H#LqBzFAS$ZrpxV}?9kn}w%dW<
z2lv|Pk@<0v)ZWX5O}Aax`2Aei?1l@s0lM07<~E$U4baubAgql+SR2DO8^bmmz|ID+
zGujan?|_x0qTp>hVo{DT@f{YNUzRV<e}6bXJN4(zHAkzr#0O5jJtOx+4FjflN1%78
z*bi41XB(v|*~7<`<rrnuNB9mFm?N@BMOOsRC*jCM+b@i&wF(z2<VP_Y+QuVfFt1Dw
z0%s9Cad{NoCIaf2Voj>lQ1%<4sPcg@$zn=3<~Tk+N_>Bt%MF_=$J|_hAL!e<f8Jhy
zAGl*w$>{7vE1F5m4ErZL-sY}FSa0uc5wHi6kk%OWA&Zs7bkN?GTw*_VcD=Vg&XP>`
zx3`+Gw_1<5w;5o=k1DSBHiI?G-A3+OhV|y&BE!)`*CeLR=4No0lJ9SGE4rIn?RQfw
zlf7Dtx0YS^4B|8Ovd>blP<}vhe~+b9TJ!^Gdn_#(%F<r48l%_z;4tj~WJY;=L`G?^
z5t$$Uc?+5O5ypYc{0!rapZo;lnks{$LGhquP{(LOt}}rE`Zvxa6Zq*5rgFD=el5g0
z72GDou`|6N)?(TWJH#Oy_}*JGFnaj-t#b(n>~D%beWmIxotM%U4LGG5e?or8+rDgQ
z@Ulrs3g0Zn4znx?#jKkNsL!)h?{Wfq@$%I?Kaz)!OF0b67e8QFHWSC<9X#EGG6IyR
ze2jx<>YaZHK32>KJob|&>PK^G+Hq>L2pYp`lKj+{X4bbSVSZBkk6*gnTrnz@ACauO
z*HR>g+}b`%xk;q5&q{ymCtT}ZqI&c&-D#=%-FU&z;oE#Z9ACu0%rBiD)&5^a{%Cgg
ze^fpyml0ea69X|ZGnYZeHWZi8Eh#CN{op#8e}v;dU;h3e3+t6nrG<5Iaoz~zhWRAE
z4DxC6tas-Cn^GDOA<UBvvSN`ngN&Oqeh>M(#YL^c$CgERR2PB(RN1^L3S!;yV7#K1
zsoz|qlpb5!)c*^QBD(3951Bvu)%yA?i)*-7uPBlf)|J`AOsi_k6lrNJQnRPZWh^TI
zf734OlRlu%pS4zK%JcDHfsDcUwtAW$n{I1+J<Alzm2ccwa6}QMw)#kea9b2Sw&_XO
z#v*Om%4(bvGN`&m?+ar@PG^<cV#~&nO4W%}aoMJv5ytw}+Ppo)mD=0TUJ1v8lj;0O
zdZ@)4QB~|Z3*R<I<pOeJW^PEeTASW<e?s0!t$|I)4=q<)W*C_^^9a7%fpSQ2gzQW#
zkKZz}@{68U(A}t_nY_1kWtDb==eIM@JvZpMNO94;SzmTE#Z8N{Qh*u{J?SdxM?I~r
zlh*^>VTesAOPfQK`wgPDIu`@zjTLgcEyyHibvGp@p)MJ+@+gvWlSmrivcUx@f4o^K
z!pi1>l3^w8(&hw)1@=z)_t2Y8ZhffcR*dABAt!V6%gbsvY`{-xmrvgw{#Xz@o|gn5
z4^rqD?DD@Qbt9h^IhMi!c!!@+<KOZp4t)3fm(PFu^e+%N?Y}fsE=DK|F)ZJ5)1s3P
z4yS)DZ6}O2K`$Vp)XpFeLhFD7e|Tc;__RD2gjb`L>V|!#UM+k;eRY$;3KxvhjX>AM
zJfs!5m+5`&JDICSBFR+QpcbuC=28>q^DJu8)3<pj&ic>t@OgRoShQ-L@-&2~&`Qza
z-<DR9Hh2WgPP5#lgOCb6Y8C<qINa6E&OzGJr%-5TNK;D0=zUx&sVXq{e{@1~7cRvs
za2J2h{X0sS1jU|ep;lOA+6o1c%oB=cvv%9}Pv6S4FF5JB`^?K6Kt}<%(!r@9Jj#P>
zrLYcnb_<-!a30O!5)@bGE1)nX4!HW$H7$Z9KQ0|hNwh<g?c=TLz)i8%-k^>-3T$ey
z(F*CR7Iq>u765}>3@+!Oe{lezP(15XYm<LcDZ=U@C-;o8Bf9j*t@tqmiVEtnkNs#3
z(?wC$Uz_47&kX?c$UKU}p30SKklf;}795$xf2B=c*A-6ZnJ31UUSjmSDHKPW4;WHN
zkxsd`$z^Wz{`$O&F*<_K%Fa8HFz&=?fjgrzMC&V>QOs}RJw=7ae^sUR`NN)RAQiMk
zGc%MTKpA@shqiVB(%HO<EwXjVEJsU&QlP<sP9NN4<mTWeV;{6&#GJdG3DpXC<P=5?
zmLmrWD@-0JwVuv)QkW9<jf|JRYQhz^FF2TXm^b`Oa~9dZ{FeUX?ggm{ZnQ-gO`sm7
z%7(f5%dfo`DPhB*e~TNE8`+8UIm&QWgjhG)y3SQtwvx&E&^-qf!#ll|bz6iu*W9yW
z=ixhzT6RW*eed$hD=R#pD)dHK)32v&pvi!IJ507u*<$B3$Q+<#?dOu?+kEKG#rARz
zdof-_5W)Op<m>&p8mab2NU^XzwUmHO2V)A-D1AInEi|V;e^&7zhE3>_Xj`d2MMW4C
zD2xukEPwuc`TX#4UbO7k{MpPxI)`RBn1w^j49S&1cm4-8%{sz;K=sShmRwg{Ru#_L
zKH^rUroH#@)Syw+O(Mw>U1y!9<xpf_NyuVDB=o*DA8Q&HZgSSjP@o4hQ>$u-G>;vN
zYAzMf2;CCff4gCu)WYDAo4C7X2g-Lx(%hs6?a<QVN)Hrc7-i?uNWllyl}Z@U?1cGQ
zbE#HUb;uc7=_x2xS~sASk+#P^*f5^X@pO6W%HE9@Nf9O0?rKC0LD`x0$$VdJvki^S
zcVO++0z>DkUJp#yfCo31vaDAC&kP5+86pJxQ{f<+e>exl=6o3*j2Fg)iP5*h2koLT
z!T%(H(B6s^0Nm~vn$|L4q)p%uuJ~6xoJr}bvs~V@r=k}_=@kD<HaPTjL6OoV{EAul
zK+j-XWVe*0$>b;VszXn*=qR^Tq6}uz=(#2;&E_Ub*ZJR5K6-F8D7&I0UrnKvSe=Ln
z1+#|Cf9A<k9)yxBef|*gYa@jZ=FN~F&DFUrDt{&PYoS2TOr<(*tv>rzRoFjXDWIlN
zp-YbPi7H1)H$bSBX;&$!l0?Hk?b~6G`A8D=1qn&NxN-G+ZM8vD_N$$~?3ixaE{k}h
z=zLGxbsE$)U*I%{b}_Z#Q-GO7P_d%nrfa<De;4;}^iYk_&J<w#Qb}WGN|mq9hh4ie
z(jb6*(m3;;5@0T(=?sq)>cvirmt?A(9QZgg+7q3036c~3Q(qq_t6aQGut?XgBrHC1
z_mGpX_F=^}UmFd@2aZL0O}%04rJ@!NejxOs>y@bWUmAMhP>xs)U$kA`5kxPE=XYfO
zf33nXg*m6y<|l<?1StXFd&9A{Chw5bW2AsiTfG^wTM_YmZBV%%vbTWrR%BP{wTX$T
z9@&*sbMf8P7aoA!Q&)LB^Blf`ueY!7%gK^`63pvya$8yhJ#|AQQRclOl4x`Z7I)=*
zMHErwUPlpI&;`+n;N7Z_67Fbn0fXh4f6yHw4DeB)c6LQD{k`)LnAN-IMDg6cgVOSR
zcFwh0(B7umxvPF(YX!B|o^JEk{`OToy&<Ay4f;Elu>ny(#&)X{k0w4|e$nE_GB#-B
zImUZ}=jc3+{N76~7M=J7gv$VlkQJku80-KNm0<N1C44VlIwQ1N!KKASGs4~Be-fy?
z>8CecwtR1|JA%Z~^C0M2Z<nZqg?cSfUaomf77oT!X@6X2qe{_koSo1*P7|&|U+<`P
zUlTmGyVP~{vIrdL`f45OoAJfDKKwT264dn9$)QgZUSHf%Qh=SAxp!>`vnJ=vL`282
zWZJ!0+1$%s3C9S`1||&_nD`M!M@wmFV$_vjRu<DxwP<;j(ST91hiO5oXs=4z`Am6a
zc2yC2#Tn2Dz@Ua@E*$|K(O<wYe0US&X}As8k7DOfbWZ;V)8h2~mthJ!6_+ud0Tq`q
zi8?2jeDyk<0q>U?_Btbfr;8jvKRH;nfcC;l7&$AOsSeo9*flDB-^jAx-6}A5<t<9W
zEjF~6@D_flEyyjtyeXf454r?u9s24W20KSOQ+UVG;HZXZjbKg-S#{>6692?EOB>$9
z6A#H!<v3;~G9Z9x969<7%Is&!GW-eO9~1flwR1R0q4?uS)AT!k`@^#Jm*<08lQ+Tw
zE^s}nP1ZR4=Cda~WN!-Jb^h);A3dbKa#~2c&|c-yf+PX;8P|ZRzRQ*<Ui`GXg^Dvj
z?~i+PnI{H?WLuRn&zUC&Jl>KZG9%|`Lku&%_3NNZiUDy(?2Eli>lXh`9Al5hLi0G0
zhVlm!inH(5kwWW#!%>?N8lm(;vl9`3h<)49NCjywKE5nNm(+qI?6;pczfJ5r&r@_D
z_I2`3Pyd}lGX%#r##nxm0`S4_Z$AI~`|F>-{R0FO{3lSzN%33&@1~#mm4pZKF#X8y
zMxa5|69_NKDE37GMstTkSR1}gH$0F~gxGZZ5n4%D$7XJSh(QbE6w*9E(^O=L9BXSH
z??>BMrK&?jp{p3-EpW-S;g#tNRd@}b_Pumvs((&5uhY%fYEMu6kDta$vt~{nlzkgo
zla?lS3{GpHlJwlcFpjH@otmX1Yiksn08n^ut)J>pRnyYaXj__?RpC&Ys$4_+0qqAx
zlRa|M6WULIqy!s3cqC+#CbQ1pzx~W)-<jySc_ucujJMzfI^7y21wiB#s||*9Er#Uo
zrFWh)biX5A=Xw*V=6F{|E-|;AUjtJX;p^0h7*)NUb^W^4JPeZ7YJ+*kVx=()7@p6n
zn&m)fEgIzB#Vn<4cqjoe22@!}hxBVK9Bd7YG6zO~(}i2zhU7lItOd+N>8T0H5W{Mb
z3ELtk(j1jG$(KXrF-T|gtw1{ClOX-q)c4A9_jV4v$6Pb_*3JsDbAjo1O5R;u09CHW
z3#FFG&Nj`~Zc~1=>U}pAn%-|opRu%aKzq!bbGZj!){Ae_!dW$GrMa9lS&O~YLm70S
zBw;Fl<?KMsW3hGbvnADrf}33eamj^?CZwV_jD=c7=o%#@N+-}SG>N15W=>(oGDz`d
zE^bTVr3-qYPj7)z=@n(P?M!^C;;cX+8_VlUf$FFeP+zi$C*0jAu4eR-!-s;Dr(7??
z8b}e8JO#?ir-JcN5XgP21q!$`SgJ0;&IXEq3n0|$QlL8O0Vs|##m`QFW}VP>g<Dvh
z%MSvL1HG0s%5YuK5@5OoEl*a;bw^1~b;531=TQjmq!+@M5$yHWv@W;}yq}qNCGV`7
zsbNvsmc^HuV_I*{#B$xDTc?TgoRonsoF+YWnkM^Zzr$=wJPR&=K;ju>SjD~5Yzbn2
z-OIRd!Aw)*{>|}>iwdru@KT^U>J^~2$23QRq*`+PNYD_4vTl^&dZ5LMuuno~#k0+z
z8iuFt4WA8aN1Z?os_thw+cGe;sr_+qJqjDPB}UF42`+)~grf|b%f8M6+p>(5!pMcz
z?~!GI8?9N4bWTROKtJb%b+kYg^7_AjUr;FpF)DZG09>$mdDriql80u=Tld}hDfw3=
zUAsGK5}hu)-|fOH@~OL-?)0uZSpfmBUeb@)bw?dG=f;uiHiEB3oy-UY{twY2SM#BO
zqNjxn7cG^x?Y5H=mKL7z4)a54j0N)s-L}n{P5No%5?QaRH?Qwkr%G4|nfcg%=j7IT
zoph)6O2RfKQY}MESdTJg*II;LZ-SoT@jSdezTEt_5S|QUf^pw2{1CU0EjK%sN=?82
zI=$X}9l5rro)JN53|p>KAlTBym}9`#Z5bAtFrHv0VO7}XD}SP&bS=L;cRN;uFmfEW
zxaQo*+F3pJ-gZjELhchxb-uEHp`?&9`f-x8Yl*(><(ehJY8l8bEMZbah}(Cil6YFn
z`%8|<oQzs<d}I#HO+E}xnvt-Usjdk`$M6&mUFQp6xHS@@l4r|{l1qAzqrB@IaTreN
zl^HspE~k#0-k5lleMQ^%98Vd)@BIerUoJUvT=5&_Mw9~dQtqo9EC;=RQcWLJd#IqT
zhd3#C6bc7qq>VFi-b+lXuKsc)tJR20OiO*bz(yUj$jIYe5(17o^C~sKQj8K@k#Md$
zu&o=*2!7tQ32gwQaqE`Cg$MVCQY>0t@Bj=7+&|Ubbo4t-Wo#Hqm7L~_*PNT_(4z2M
zcnq127m*%LWV-RWA@iYs^#akqoAX{U15;=l1PjZ>L%}f5!+`>`NQ-hSzsJi*Px-3S
zP_pUP1Zlr8kjwJ}3Q5oPuiK!`Te3HS<8BM1&6^}em6cW4x_?jz%0wea{rKiegOug<
zon6R!d8I+&Mhef3fuoT}BIc=sI%NH&nNOzmPN}>32s0nFrK3%MXb;+ScckZ<Lc2I)
zx$iX}nTGpb$M3`7`hni@5ZCa`$d@tK50q(MLe#ZbDcfZo_n}|K)OA&j(yNOP#BfxG
zE!z2h(6v@$pro)c4s^zar<1PXSrA_CjU#mCU3Y;xnmGdf>RIgei&9O-xqdMas#Ea(
z4aERB6E~BtSqyZ4Q9C7^?fVD^lfO11Sfv;EgOWce&x<AfLCK%9rf?qLdVZZM@bgy*
zUI~K>DUN0Oyht4=u9xXkP`@jSi4b-5{(_v(&^W(Au{Y(>Tv?oJj5pK-#tW$zdg;e>
zk;TZ%DkPp~(%LiJfY@;~hp&NxY{jhIoy}LC($a92^$oRu0e}~l9k*R}A8@JWrs}dT
z*czt{PM#&ia3iceDpx{!)k*WG*H4Lr?YTz6m2$WM-jV!`ll!Nm17Wz(>3XXFofs&D
zF<cA;pN>XeBB27Vy?w)6%r<NjUQ>nHyTx6>S|znrSa;z~brsQwpL^GI3KE2!@1$Ii
zP91P90w0hXP0!f5K8ULc|EHuE`NDiU<T%cu{|6JfMr@a13Of~-0tp2Zm+byJ#D6?J
z{`KXT8<|;`!kh3hyp7+!yxjcfrfx)`(8eE(BB2x={{H<Y{6L_EvHqa7upSOS{(1Oz
z^QUpvvexh(#pZy;CXLa$6Pq$ZYj-dP1zKCgM!Y%An<W+no%p5r#T(W9u6ffsd}dK<
zoyQ#WbABdS++@?{1@SWV&yUx`@PFjjuhYK06VBnt349K(PLzd0vC7e_ks_#%^a79h
zdsNyft9aO_cImUWkHHe+c9~qBA_J9GVij3^Fj@%$SB3U;`KF1j7RQMe_ACXPgcLGN
zp>&$Gd%}}`I(fOMV{HVYA6MCxlTUU~z9(UHT&-D^Vq=}Hr@tik<LOofN`D~8RD$O>
z%d?y3qr}-Q^6}E_WwRm7;?gLEmNgQ}<K@!i862o_rVW<_^a2%(6v5@<hM8k}p27ut
z%)OD()1LA#E{o2EUsSi#%>PovTg?hNm};*DDXhwR$~<JHXpa4eVKt)g2_KRo<8QV;
z2|>`;Z_5fb;Wv<v%04WF7JnB*V_ECWfT~U0`KOan2haOfhC1_bC<s-z$J>A3WfHwi
z;)PMRC;1E8@ZHuz952~Gu%~*>BpiPiNwm<Q#)Y)pS%|I3nV#x@+Y7OeNH<#tHpRe}
zDm?q#g<?P@tRA@gZzfqbLqsP?&&-e=mrcPe<l89^C0EEMSKOyQIe*WDs=@g*b9B8t
zI*y7nY#JF^-Za=tbWfHgq0EK4CkMhxxJ38lLLES>%3szAc@WBM=tOsp-3P6iKKcNh
zgHCWvGvz=1sG#WDWF4QfejtR4v_kpTX*-sodOe5a+%elFk6^vg$V)C&)?}mDZHZu{
zf?^|_UABU)(VT=;4S%*_S1eeKxt4V4JGm*uRa>8|@g6DexlJ9e35GiXpPT@%YXWN1
zOPE_#dGB%7YF%}DIYw^yY2>re<zHj6u-TrJ4<pLnTuY;iHP||T-ErlU@|bKjX{|a-
z;V0FtTh@SvQef?+vttctxuxQ&YNR?g+E`i7aT&nX)ToU`JAY28KD1cTE7W8az|UYR
z0aiU)LrR4GYLf<HeO6L>l3k}0Tl5&#J?=<<(c{qwZrO^%_ljwq+*X$^lVW}qQP^GK
zf-{1=8y7W=9TwXyzVjSd-}Zb?ZfJ)&;|8F1b*r(4xT=^FpC*T*ry`YYChWMY(42FS
z4$;IvBGuNc3V+o-rkGj8J$QArfhF;`+Kx|1cM-Rb@T+$2xzh=D-mAi=<>KzQU5o%Z
zSI!Km6r723+Q|bRrWnnB=Oj9=lTjP?_|V=+mCF6aUw-|%ZfM22!3iTfOH00!^P{P9
znzaQPoT1R;+nQ@pHj$sDj4C^U%~=DQxUW<FQhKWJ41Zv(oMP7=z=|lkK=Qd4Ra^B(
zP~)j7m~5jjXoBgAm3q%f(hi$?I+>y{HdP(Vgw>L(QP+j&5ck#m?nZRT7%olU9US5k
zKs2?(0ygwEaVKpZ`-Cf>7oh>tR#;U#%C`Va2^$zD+G=KVqd!pmJba0W)oIj}raR$X
zynn;4?|*Te^}W#%;{|X$8CLm7cto*vaz`+%cm~`=$c3uJ>+z;xq7MR`hE+ZsaT>+5
z4>^${;fUS=avEkj19BQDesjnPoo92LQ0cu6U+S^vljdPJLFG}kQdpT~KR(y>-i{Fj
z*hWg*?fmxj?jH5+kGp&2ZvXa=Z48HNPZ;U{xPOh{kP?QQ_m1SpHhiO5k?e3C$+vCB
z*a4j#NcK@Ij<+Mx4v59+o7jZMb+&;FjL?i54KB4U#Ux_!m=tUq12UcQWkZ06d21F6
zXzBJ~ku~$%f<6*|1^O6gSZcE{AWHyUoqZWB0>KF8Gc9D;RX686=eSQ9W;hcf3AO=N
z|9_?sN$?%bX=M!}G4vLQ#ONi6B+j)7kytyn<CUg^NYeZ*M3VSigGhRQ&VficAI^kx
zfnlKIcEG@FhBMF-Z-wU^Gb-kT&=M}{J!kh}wWJI)q9sWgoqF`xKX-}%x$L^(V9517
zWUH<o`J}d2e>dPn0rNmhtXu<}aE>4@Tz`gH&xKAb#{O-jz8pJ2sRXxL<4nJE0EK|1
z$GL0N$pr`s@<R)=8Gc_73W5@z>=h7-@+cITcZE<sH;sczO?gi2)qGXoXXsXRQ~kUR
z>gkv$`wjdgxFmR|Ww{S<iDGR4y*pl_3<^0NDfRMwt8gPS0&jypRJasHW@K#ZZhwv9
z0)=-sP`nTJQOzt?mX|k)q0tD)_x^USX}dX4@UWTCu4t<<;0}o6jQUHKz!kib71r4q
zuXak3DUzSZC)c*Dx{u(K^Bl6rjb_zEs#Lz5$EPQb#4`<LaN;UH1E%|4Y*O$gE)9m+
z0*-PO!k`q_i_?`L)Q?0DQxTIw$A2T#7J(xzr4&u!lV4AR<=31%siyMYm{J_b;$AIn
zkc6^k<Tl(AjKNMNBqkIX374s1@-)?kf&UyR*5P<9teQzB%u7)&J>F9q(y%+5rB)N;
zbCEz-2p;pV)U{egR!!5JpOmW9l%lC`-bNYws+q3MUP}lCeLtn*E6)baeSeV8{a!id
zE>+~fax+bz;?^3ydRL{i>gkq5cPAzE1Zp$gRl25dsW8HG4dAPnoiqsTq2v1Ymx9CK
z!3b~B^<bv9?0$wxNI(V%cdo7|1H$TIm0iF`86M4XrK8$Le3S)_hkmvY+}(4Mo4_x(
zc8d!n=V(PtoM_cutK&LChJU6u`NXYR83HUNw>2a%2?U!%xcRjv;TpHJ4c&30qF5?i
z3$nkuv!J`sxVfI9D$*_KK3oe!On^4&EF;*tN#=!<(66!&_Nmao>7Hr_?NeT02;;*-
zaGuyC3GeX|v9%OuicMlHv>Mae6<fQ?`C@CS&J^3Q&<;q-Le_vW2!D{)16(&B#U-cl
zd!xbyjIPDrR-EMg<PKoDXi@$w#RWhedy3c1SciaI2P!7vQUkLee&3NNhalZAa)#b?
z{O5#@YDaI}n0s)ug<vzrA$nmgbukXR&{IdWG0v_GX=2>+kS@lp3~6JW-WC&a3t>$k
z-v6*|sG42Q!I+j;ihq~Q#T{3<i2dR_=Hu9>obFTQzqa{hSMfBfA`0+}dI<ztx%wHR
z4$s~(yuEt32H~WhNh{%GdHH%tRyIW$ER{CX4exnrV}|l77D>lxrdZ7A<r2#%L3Tp*
z-sj$TIDs%6oZH!V>YCttPTHPi*V(~_>mSflZ96-#AtA^0aetxhARRnQ=}}xm^sPcA
z-V3_6)vfinR%^9;soB#A8ajdav^`wNPuk-v%Juj$Jy#FYwSLm@Q1K1^h7BB=XEHrf
zJyEJ2EG|#XR5$1OTH~>;(NvnTPh0F;a%be)iazuBG#%d5Zm8$Rls%2N()(S;-o)3S
z)%;qHf675%#(%VCIz|=$Cv0>aIdv7u{nqbHR~qhw%Tc6ik0RTEFdk-2o#D#SSB!BL
z_HID%a$3JQ<HTi{v4zTYz9ZEhlV$0((DY1B968klD`O`G`!MY^S$yA<PF!#IoRgn4
z2LYzt@%#WT|GP`GzvW|s*?|i-2Oo9_)oHd0?7NxXZAR05H~$5(oh*SZO?8s2F>dXk
z&k7xqdoTPCJU#GM7`Wo>ARJX&{}XOso}1yWyn%__0&q&}_w~jTg#qRN@^bTk56xHV
zmthJ!7MEexE)$paG6pD@vI;w#e;yzH`SQz+&PUfOy7V=D$X{PxZvJ~y2|-fU#go&h
zNW0U&zTcz=2CQ;Ep1fCH$J390pMJmjyIi&GB|azE0x)daIcq!E&?#%f$vINweSys|
zah8cS425~-3;AL~Eq_`j+NSRerL}p?Q@-!>*iYrzLqgQ}dC0VE9;Rn}f9gRDlkM%v
zSyTo>J&2!CRECGu|9*Tsi95f3ZB}AQAh{5HPFKg0I>9<c6y=O02F7mCntxBsf6k&J
z@Umh4Zv9061RGD+&)YN`P{!qb15(28^<?wubuj&6TV!j+(xsJ7OQ4+>?e6fRAD^S%
z6c?wJjU+ZXL@Lg#FTZCvf4W2M$tR*#U<E|lewj6y*W1OE_@uhp0sDBd9Q6ymH=srK
z`dzMxC>xU1e&+xRQCWjkF`q65v|s32fcCp!HBe`P1THGjC|8lt6yvFbvP@YSbpbTY
zjsPfE8W}B$iV)<CWRxahNEY{TG0bdGJ~%zuqEGor$YK@J6YPW_f1iKLx9s}g?6#~Y
z<y>_n`-XAenQXjoPFwFR4rI%F)OpF(#5(6J2Q|@_`M#n8omH${l8P+sd$c1<i+4`W
z%F^=jnZuIED55jnEN>O?t#;P!kGnhtH>U(*E7)&z`>jl2)l$(@=+!+0h3wR<M+JVC
zliUB~DeUmKc8Q=Ye~E+rLt8yQTeK|HJ#@E!3^yVTtYx&?2<}E`lQvAI-$45Vxk(O8
zmia|adNYYrOfrx@3yUQ2KW#r3eVON4Oe0fdxRUKO%2kP+q;4p2h8<f3K};J1?7?6x
zC=%k+!{92x;O|?)Ga$6YCa`E(u@(z9GqnP#-VSEncc45}e{0$5plzL1MiL7b*2h~$
z&CejaJ6}dm&av28v>{-!seA>pG~3$M-Cw`%>pgIoVL(!FS;)H`?;4!p9)eP#Ju<h%
zz7#RY149uu_DUloZH$86Hzp~k9PEercX#??v!4ng%n1`(UGY`=M`Bx#kbEcBV4GU6
z#A{=4MSqS?f7Zw>27fPnpR0HEB3tR{<U9PVFc>L~D71yL|Fm~e7hhMZ6qh>_Ixb)J
zjCGZA9&_k_h}mfk33Sde2gGIoO^~=|42F9K79kHSj%<dQbCi(s;K}0QA{oH<V2K{~
z%@%SVyE^C01uU8!I=nl+R9>3N0*pDZX{Eq$KwBdwf5Dw%U!KaZaNn#-`6^iWLv$fH
zWnx(E>KbW$YAkTN_m~bpjyMQ8U6+eCj&B&`n5f*w?B_+XCSx`MGd_l25thm*29r8r
zd>~Vk-wJHVCcQu|)O+$@TP$c;KHo96xR0=QO}oM#ImhP=Lb4+^2K&J80;zy&F)E?x
ze0EdMf3=;dMOdsgXJSE9U^hk91ve!-oO?`8Z1c&$krg2~kz>Kod17pwj9WPxSw}P8
za~`!=1Ks!lhM6#Hf^T!HCaZ``!kX>@NN+lO!#6op=}qTa)3!t1wVTK{Ic)9e{q*b#
zD_M-~<Q<A?IOz!+M!vGf$D}crRy96E4hhY>f7kd-iW%G$f}0t4JSVtGb-#<MdK}Fu
zW2_`%kRhXE72vQ>%y^%&URTr0ne_!F5f0#4`p%*1Ej5u88xRYWMB;Td5n@lKIqwpy
zTu|-c(1LXhuT6CfDy)cpL*>k%;f!kqfc=sIxcsOFgK`M_3@YK(dFN#R9^y*y-&EEy
zf4~8UsaYo32V*vpwVDu2DRA|KcHjx~xTTa6?nOxSw78L^v~z|VR;8>qt#|nw+&6CM
zv7RaHr4z&bO_$^ps=DIAbf*M}uEa1RZQ7!DLFveZmNzobITo@(6fD|e1=eE40o_}H
zGVi1)LYGQFP1jqx=XmU>WH(Dkjg!J_e@k{dWi)aMz3WbUR7Nu?bn6K`Dx<MZ5zRIw
z`m?HNDbZn#={7m8qS=%dVLxkJMPn6L2yV5}xQdp%qh9ei2QpF;D4_B&R_$+i&-vU?
zU+W0d>*##9O{MdlwhUyl4m`C5qqz)ZEHvdQGlgg+BgiYcK2cj{q9_vPJ4MH#f3DX`
zr*EaQ&hblPIBC|Vt(a|=SxOppezjmvi;WVeafK!lTGvxK*@IX$zj<1U9kyuywze}<
z*<G4dbf`0~g~MTB0h)@VyYG26R~!fBMi^VDY;$Yo*W`i4-fXV8oHDK0W{Wn*wQW%-
zqt|9<E1WE`H^(hq$UA0AXNm9qf539;qv9xZ#nDpAS}0x@J4$&pv}N0kOnFsqpeFD4
z$2QxpDb9RdTkK66mkLg)G}|IsZQM?l7Cc*QC&9mB0!_rZn0rG4O~3{g{JSR5T{=>;
zu2&^;pPOt47uC*&@94b?cq0h=lQLyAPWcdJrfe<yC3yEt8G}`^dDs2ce{PrxdVS;4
z<9>NoOn!Q+Ck{NyiVY(dK+d>AR=m|jfZnLMysgI1mo^cIg|5^@Adh?Pt_-o?8C@A{
zS{0w`?QJe%p=$x!?{a-W1v~>li|Vt7(QLlo-pe6YI-{3E%=qcO91x>2O)urJm!kwI
z+o9oRP*k7d*el6b07X)sf2zRmrc<fw!BR)b?Jwg@&->{G6-2Khxjs4E$-p>0B0QY~
zj3>bw?Pqq(9ox2T+qP}b4*!E4?AW$#+qP}n#+&=@yD#aa)9FriQc0z%PMz;}5ZZBy
zJ$or3yRsjL4b8<2Jd&|{HzD|*F1N>g4cpg=<T`0X1u-WK(YG3<6heT)9a6$~o9(<L
z&iedjA$N2*ks?C~sCDIvCpRf3$z?RJHV{rGIW_qT(byR_BgTdSHG^*`_k{M*2&nI)
z9?%im8XB&=t9J3iQ^?r%#m4NXIm%>-_$;HgJ3J#UJ1e7I(DiO(sX2}t;kAin!V_z<
zQxKJzu4Gn5`GH?VPGcDCqx5v7^0DxA`7-MLTiAs^`+1xB6=NJPH9chSl#f2n2;;&i
z@A7meI!naZyB>4Ey&Twt)XOK3f&#_>ow>tQ9U}Z^1;~^pn*f?RcJE4`CK?T{CyaRK
zA*HYu%yDxF3!0smi23eQFbmpL)Zvt4gh}E2W~q>$@vxYkLQu@6UTYnZ7S2W)p8u!L
ze00bqRZmVA|AokoA9)@LnK{^dqO=0e*O;dGq2+l}aNrTZPPwXVFuG8)HzGexah19H
zo4I`rwsun+X1oR5;VONjer<s!+}hw4ot&dy7EUSCms0X8fv+(EeV}NA{A~g96U@iz
zo@TXpi0vasK9V8Kx5QL#y*O#vKJ5H8gWJc-BXRkm6jQWU(gmI+nTKMywJ60^EgEZ>
zG{Z>BPNy6o4z~iDzhS#8u%4e(d;W*WYa#f=KF!@=8+kK(BcIIOCmD~m_JrDN;oxH2
zb76oA5R1n;8nygfGK%`u?YY3w%eK~$JCOB-R(E>wBCWDwU5;Y<IB-b27xBIFw}>Bo
zr=)@Svk$AIBpl$;`iftIirE2eYG>l?;$&)Q3vHTg3(XA$%$e*MiGiG~fy&(vIJxn8
zSwEk4=Lw%mW#65?5e0O1Y<=EWQT>Xu+;<NKiQl;JeRo5*qsUFI^);uh-?;Aj_IW<M
zqEE114fhtW=kzz*AF|j?g@c`6yNrmH(bAqMQs9gnqhbl-+K0Fa472)JDrtZR<Wn|I
zS<Ah)O<HH)yE;Rn!I?-Eo9;RxYG5T<rfLkl`)7`??`(fyTsD`ZaSKC%F7B3c{ui{u
zlI*A;j2yqIPL(V|z?}%7yYCO#&H{AAYd%ns=gbVr55*Z+hGhuS8F5lNC?%=0+|ST}
z)v{<6Z~ybK-YP9OaZH}%)*zzlZm)R)63k9DFJKR1b7FgidN*AekFMMRAuoA%-93b+
zM|C4FqbSE-QrS^#1+IQedd^`y;%%40q<QJz>N)sf;FUabaFGHs0;l4xzkISKk^(_9
zQP<LDUoPHGsCv}|*d_hfIQp}Pq*J=;K$5pVK2qhKtQc{^(;E?`&YYB3SW%uy+-rf+
zKu94P1ZgYEfg_o?>C%6)YUn`SSasL{9KFS#rv*9Ij9!jkfn!?{HvQHkzK{p_&N2rr
zVO_T#a7%mibrk@vwh-BDieWX}NAHbvB(xoT7BpWg1h6@+#~k+Fh~54BoAu)3b^yE`
zAw(bpQt|e87Yx0GrC~*O=wIQd5B&M$_H41LRr!7M30X#r0>%n?bXYNqOEO|f4gLNq
z_kL17i6=c-5X*A!Ft#!w#PpssNiP#qUDPo%X(apd?*Rb!Oy$Rh2vx1Nqe%f@usLVL
zs}jLmEYX4PoR3W63|#lHP5X4~swwv0=l2Sh$y$hiI#*tq^~4PzyDQVL2CZJLQxrkU
z3{F}KTnEr($^H;01C84MGzZ29dj#nb7XCoL@e#x|ZI+hjT5vRsM&Udb+xWwnux2rD
z`is_x5R&~2p;bFox#`xWIog8<5?Gu8fpqRyhtdm4r(fo<00jB>@IufSq1(B{0uoY4
z_>QUF{%NJ7iVwB4$~eAJ$Z{*!{i3iFKJvy^D9J;;!lgi`b41^}XIY|i<JOKzT9C<0
z33O>nO%t!`MI!;E9#ue$8sDKerkr<#4j%?)xFMiwj<oi#2-3l{htVX}PHRA~zv?gj
zFgK{;g@SDs^Mo|!WtFpHBmb(&wZrCxg+I!+i!3T0r0goT=N#&j!r1R;sr!|O8J_ux
zZi(q><#k^x=VO&7*^$mDwSX1gLZH^6R;SWv8M0Dc<7M&S=Nc__>+4k(Ma9IP+tMPt
zD-OV;(lFG)3s$Pyj=R2dzh!F&%#Y@Q9!;GIHwxZ5T1!c?)%#xAgcgAmmj3rChW-$t
zUB)mdFKbVI$?+Yd{4eN9=+`FSP;d>~KBHCr{t*}L!^z%8CGLPfQ(y0%gb3Esl~5~)
zB?KHhb=6WzgKOc{ZZ_rcGhJo>Cl7e6+9Uu&r1m$hvBHtbkzM|VWL+jtnKq*{gPCtp
zD<j)>x$Y{W<xod+3WQUU<B@McwTJn#NwOZ6vP&Ohyo=UQ`_*}+-B$3YY281jHB)?N
zm<yrMgK<5V<khOde!EXkb*w6yqpoL7%(QS`>4Jf5{Bw+sG^4g#jpjhNhcMlJ4ru@;
z$J4&^3j6W^*|%4tFZxPkU%tKeK80Zngo&gfqiqoCV$?Ea-LtbP_12huTFOKMXVV#E
zHNH$}pu*i6iY6sKHox*NZS=;Bx2Hcr9GWbQd`0@$EM5;r-;V0a)1um@`1mQ-ZyU;c
z*_d}0E!@(x0mO3NNF6L{yHNoI0cb!cMgK5O|5XcYK~OLV+51vS<hiu6@x1D^+z{bq
z>Q>6XbnjJHGU-0<GO^kHMO%5BdDXrY8`0b|R&K_{N?q*9&1yCKlC{ASE$dNa#m3$j
zc9F#wHJHK;uoA51mOjtVu@WuyGx3CcUNc0^bj2*>Sk|T_*s2Q`A5s&zR&@Z|TI5LD
z5#nfhIk7b%Xr@vzk#+Gq;p#?4WECO%q+I{u$y(0|sEcjRA26576PpGHyz;5`de<vV
zU2Gj{w=;Q8WFy5T=zc^F2W?;iMdwYCwqHP;4sD1)oEc7FKu$Y<M1h<V9+`mdz@5x~
z9=(5F49!0RjZjOb0+0Bw0|tRiUiM=fx1qjPV7!=L9665Qd^W`J6`fKuX4kHPd~Hbi
z5CI2!Q~5X2JHuBo{4h{-czvYP$~6T?MzIuMTmc%Mz-BWCpd&~@M2pI}{WQ4jNG=XD
zj5A1o6CrYInVbWh9gRE#X}`N`OnbUMVP(5}ldRnZxp#Rxdy~}t0JIR=Sy0;WI(m@W
zbfY&$D^4bNmxj_iuS#qhSIxn+z<no1TL$CdhT}_d<b6qjxAu_Q0ef3xdtf!DDN?wv
z<$X!GOYsKMEknukQLxk`6?cI`)qiEFPJ<R%jN*g6OnqD}lq{#K&bzi|5~J&FbrFx|
zRx1Qz!iK@=FAot806(1z>6G(_7g8zrFxKSDY<GT@>p8*>R4{4}uYUgZMmr!p`2L|-
z#aW2H%S<84?E2_>H`9J5Yqk=0E!)W}5dLVD!9o&d%mL=7yIaf~Tq&88){%3p+N*pe
zBpg(tu?w+mNnEV-fOc2uc#WvVXuuuC?zDplF8U0q!c}Jgj{>TDi_KUqOFFG}0PLyW
z`(2y^2?91%V?SaG_+WcQ4|YzJgl<Xz=;UrYK~Kw7#3g+^;qyg~jkqI9s^FcF=SgUj
ze<~2Dw)v28lOyJVU|$NM(H;=P`|Q*YU)cS&=NTe2EEOVTp+pGcxza&n8n=8talVfT
zKJ0*~kxRzH13~Z9-y+G5x5?OMY%LD=OP=IN``bl}0NQKY<;v6JOcRK+3H(%K9?Kgz
zAHS#q)U4p5&zZ#iB?IEsv4rt(IB+pi42K1BhiIYvqwM>)F$ER1|CAhj*VPYh_i854
zDf}HD2!n0rzr-CD#$?B(pM%*A3hn{G>{D|yfcGTtEhG>w-ZNiuX)yA>K%-sH{-{r=
z!Ibx;Cwt~Ivs<(5&<7#$x&dw8c_|>S%L6Q_`Cp%@uAGs&YmJ<sgDOP5@tIHJXvb_f
ziluZ~r(5@YmC|AH``u7svF5at`W9K{KfrQG15pvGtG%I&@sImace;Dqv6>u^$XZn~
zUFS)snjK*%NZy0OH5AJf5heifT%Uq9?S>IiK=q!Th~7|V5iLjTrV)GIFx`GAyPD12
zE@K{hORWa{0sFFks>wSVKoqF7=J3k|#<y#@iW6CXnsL-eSAgHr$;4FzIs}_qF4*3{
z-nDCMIIEm8J1j_FB>B07BA6IZYG?9q(ewOrmHg0LQ6w;ozS<zIIR<mG>@UiyDXubp
zu(W79^h5{Nmxb|YC1RRMjidERVZb5?cGK&++Q{I~28qEHbf0)yJ}&nF27D3>=6(oI
zQIW~b5vMjj`MEot-GG5=rxvrY9Kma|blLh2%a+(h!!)^arAcPR!n6XgPo0|>j>WFO
z8Y|!Z`f%IOs!A+AEN5b4y9sx8k)-omB@&&fit`dh=%Ujf?-dZdYS0SRH2{y{@l8VL
zc3!%|Dd3@=jo$<X+!s4ezs#O3+-O2GVu+}y*bpR3@3|#stRkXkW99Df*s-E!Tc@dg
zxjm}D&Q*DWWOjO<|8fra*${l=y<)?RX73+8)BipN^H-f3j(i?FZ+U$iKDzG*kSse<
z+;JZhzwR_TFEv42CmhDucdDAm0!pzb?14MDtHGU{{`s66{cFCs9CCupm*_<5A`tJW
z_>R47_qek(2_6`>AR&G+f0J3K9Y!|JE0AbajiQ)y7~AH|=Sl#4NE;+Z<cfg(!OY<E
zDjDU~{$f-6dv5)-!IZY*>;xiHi45~(nreAZ!W~*dep#?bvY^{%xfa=odE8)1g#8iQ
zksz9qL7M@d5Yzn*+pG86V%&mOK%4R$gldNqM4_LO=!KBA18XK}R?DwRX-KhXoOSva
z%*r$u(5S4CN+&s>E~_az{S7)aLJe5KDeYc|j4F_Ta-yYZt(i4yew8rtyf6-yGcuA<
zCa40h_Vp~{7bSk}vbc&0#-=p-u{^}2uWBR(>1us$De8EcBD8vTOB{~UX!GU^ZB~Ra
z?<a-zC}gMGKXcB1zaivYi1GuX(nt_FMYJst{(QVG#1<p~)J5%ujLn8HP&qG@AZ-vf
z3Hgi&9XTimZ<JUn6ZonBr6%;TW4@3nGe8^1-l?lj$V=s_075%I#zrR2%{UKTstVt>
zI|#a()^JDti&x7QnaF)fZD`SfwYnoV$fM+HHUi?+jGH0piPs@Xv)&u}mgat$ab&nn
z6(ia2as3_uSkkFqqMZA)sf53^=KkSQJd6`}TlCe*D!`#bC-=$XOQ0^4CXz1IvWlc$
zSSBpTpN(LAZk$^Yx`a1n2{~Ik5~{kFwE!_6z7T&!hTYD+(2vfVLgBSgBpfEKJ$|O7
znOmd##-^<`-pAOm$BvOp;lXbtP8dr%Gb*(ePg21Fum*LBu17_vGZZ)hnlICuG6jw&
znlf|Ra%{=vjW?SzW88+Da2*wb9geFq5#H!z+mnuic)10=b;qkr?*d&e1{8I+YLH>O
zs(o=Sr$u{c;kc1s%-U{vSuWs%OxOqJt#b2mJ~M)K+e^7wFJg#U-c46)_9wPDzpvoh
z6WHSc!Rc6T@qX^djn>K87Au%_5$sz(?L|yktU!gmvtC3H4hAh;!g2Rn6uO0~xL<z7
z#kh+BMho5ppPqs<4I0pLDia=oQExJUDG0a=(k{}svUTAt`cEpfSCudTT9T8P6;-Yg
zv+d=HHW-y|XFngycP6sCpOF$cBu#<sc>x{>z>1kYV*>Wlp9x~eb@79D6!J7>=bTzG
zt*YehP}oWfYBtw(IR}@K5TC;jajeQYSfJowX%1*JtjTt?8^t+&RV~&k&eqe?kGiKa
z08@6Zt!W1o7r+N@kKR_9Ykvn`t5)pGASKpIJ-gOth2~fO{t=sRxp3d8IFmIjvjl8(
z4yNirKRjXXIRLSt3g?iIE0P?P_2Dn(Q@SqchB=ABf7rBXH>?zAUCa^v8BBG`AK$Oe
zz~Y^BZH{&CLG>19+1?mp;jXn=Doxf=Y79DJOIv^sQX4y71$WXa;~m+gaVd+1V~?{g
zJ>p)pnD4&8?KRsy#T5GUA)bg%_78yD{^mXCI>d+k?bT!*z2&@ftR1(ngRYF<%@bK!
zrr+nwfw8)h&6m)lN|r1EH!9EY8*O=x4Xf4J<^FBBM+M4g2f<3~P)<gH<4lwx83S_|
zYmo3BiI`bn49u#KMn=I8M%vkko*W!IbWGbhR~IYlC`w)5Aernj1cB_rX$b&PoQ{_1
zBC!b>geFE&1hZ(2zs8%2nd=gm%9Y14x*LWVCY;ztW<)McjDpVj<X3?85NP?ihA8F3
z#Y>h73q(13XB%<7j(p82bnA|pYp@d{p6;0}C*PdJt|h;>1QiC7SS4A0o(htXkZbwZ
zo=d>r0=4Dk8OYShk7$O|*$+TeLWhwvGOyw8M^v!s==bY18{F@TM=(}t+w?C{iXMk>
z?7}jfB4{$cNf=S|{vBi3!Hv(PVvpnM)dZX~%1*#9)-ML<gxzmsnbwTK*58P3TmD#d
zM4X7W&pyS9(oU}uPIG%W<Qvy~7Qj_#BQ4x;T9C(b2L}Y;4T=Y!mw~Dny(bY#)j2KO
z1Z1d%`KeEf#(_!GQlI;3BWIWUY8#l$fj~AFw@;Wd<L0Cc+JxnSlr@^*^Y=-?3Ldlx
z#&p!coM^RN5pYz)1X;&5oH!r&Uz+F)|5F)y3(%31;h4BN0bRhAK$ax5G0Jr_w(P|O
zZcaCZe4wwdlwD~iwdtG8SR5g@7M?E@YN*|BWh0FIKHRtS)$Y^R?)qB=T$)D>H4dKN
zo=7iX!vL`15V7nEgb4{p&u`ziPk(2I7tf*OJsl};&5!lncHfTVK0_G8J87ORZBjfd
zZH{C%rDDAufb{hT-5o5vwwAdE#&yPj&fvVh?{TN|oR4H*m^)_WtFPU^yR^~%qtM8U
zL6Jppe15*`z6e20Mm&<AI$WMT?BoBSssM%gO;%3776*R3Xv*<@vz2_9`>C+{qO6iS
z-Y7|wHccP*`|JOBUN4uaiiLutnGn$FcNZ(YPjdip@pJQdy+j|hPZ)m}q9k=R(c<`Z
zX-(K?(qb}l$2j*5A!jUSuAt!G4kzzbGn0ZI>p|I-PAIT41^0<qzbgCcKQ!~0Db@L{
zA#J%zn<^krVEet_SSgW1uu-6cEiY$^X_V)Z@IJg-g-FfifN(9%-2NSb3wr&nL{^VL
zqTT@HKsQGYnV+5BZr#+tJt}_<nQpe;z`0OQ-c{QM$mUSIQv(Ivg*Rr^!LCwa?Ft0=
zNIb!?g*ACeUvu?9v=9|a`=};qtW&5rZ4m%FjPg(r39bfeJApKTyND3ECy<>?(%n_P
ztHz@9Q|j{#vc-sDe6VPXtiNgrPoH{{5~Tu^f~(l=k1HRgOjBx9q<ibH5i_(`Z*#?3
zmgv0V!;w%^!vU>#pkeZ<HtWudc{|b1d5ZCBy{#VHGH<Y}hyhq(eR1#bAOWjvQr(jA
z##Yyeo(FZ6rZ!Lt1$CCCmIW0;Amq_kzr_>{prh9ie}D(hXGWJ+==mv)dDnN+S4Myh
zEA(<q4lgDQ|H@I_F9BSitT8hmSd&c=fWZcB=Yth}NYAijHNL4Rq{D>vsi(_R7_j(t
zzb?7QUA{cSkJuH_*p_{Gu43PV%A9lR^q>RPgG=o5#hooi5e$P>gqVE|yfdF&6#Q$z
zPJs(cDh0)l|Fh5~)_SZTBbZI}W(aVo312$?JYBh-M7D-)6DR)ThbKYUZ@+H;AUZ0W
zOKVl^H8{G%_V2`baw=iphH1VsGGIZL(%&;OApMomIWX9xbrf`I{>r8w4q`Sjq{X&D
zYJ=`xtIoBNc0|<tfb}|<rY-l0s7=iR4;Ix*ED<H55zqR0e9w{qNkS{7U;>ED6-hwQ
zS8OFGrUW_Zlkf~k!Lb-alG4e8VAL5>oTrFKY%=asqflj9*_ur(Iku!wg^qEsSffUc
zBK+dmU>0lo=or(50tX+1pz?z6uzZ~6gno0IB=i*Ryb-FmLZ}iknT;u}g<qCnEqy9Y
z68;s=C%uGXnNg0|9x;-A{|;!4QGl$?Lf_*&FHZJ#(B$t%xV3%XWdj#huSF0uc|y)B
zXDl$*(G9(7RVos#TEc3wXljH$(-YlnZrEg9u$OCMjpHon43}?Q@FiS33pKf%l5uI4
zC7`;b{=91r;vD4t&DTI-?QDy$>*icIKOTn8sY)8{8S6-p@xDbl%LHH|u&vI_+@zsJ
zGLo^ftiI<%9Nx?Ufk&{e*i(?I7aAna!&0e6c#&Cnu%5SOo%4lqX|nQ9!D*-KJ+Uzg
z$L;Sp#=Wd2N)<UPPD+T^C=#`lMkKGZTRk!2pG2HuK&?rgu)>n3+0u5C?*U?!e<xv>
zNE^+OU1AU75Ay4t^8%no9T}|{8s-`l34lx7Ml12Mq?##Jc**SMm1NjlV)_1q(uyz6
z$!)LE1ese{<t1pJ3FP#ao9TNy#w<ZX8r?q;BMp?RE6;Se5)HE2@Qspo-;=Ql#O7Z}
ziBTO68$V<3{%pIn8yp2|%RJejwat|=oyaTAWr~+Bg0|$y*$1F;?8pKM@Jf$p_a5yK
zV!ixke$iSEJ9Pw~uxLy^fPhoB2&aKQ;~Vj~T7Y8AhSNF)^Yzr`fZ)o5hc69x#>N`P
zQr6LkR$XJJ4ECfUXhh<E)Oh17pFWbx?4uFkUVh!ko6aAb_&cPhcX&^jj$4qT@+u@N
zO)wXO*UqXCD+Qpe-MG5ba>}Gd;$OTZb<~|j`h|j{IVoP&)ADPg@gv87wv#;y6n=>o
z^^e)7((m{;#nKQgqq!~s3XCQ*u~M(Jth!RI6?qNm<0EoE*gINgKc}d;#?Lxx+~S^^
z^GG|rLScn)ooaW4$AxJ)4T~0x&YS(w>fEbtT&RtqCJwN9X}3z+@hTkJIDDUoa3Z<L
z_oa9n65Q+1CwXi~hWP96!@A~9;v0>L!RT!o5EYhKv9zD&vb*Ev%2P*TBMBwatOGUa
zzRsBbhm`8Im(`F@Rh^mr%;WkXw~Mr<T9XE5ey-@dr*t-Zv>J=EnxR-vs-M%*>%^m{
z^#oC3eF4BV;Do?bA8ypm;&K6pg(RG%-Kj$6d1n5MlNa?)lMTooTk5P$GiMa3j6^rm
zU;@V-o`gg>MO<070(2UI<i+|fR|OFvuj}c~?J9-?{vL<qT0$=sZLcPQep%|2vEkJ)
zi}qu#$f<-Du#mER*8jEAc=dsp;aRd8N0mf2@(UoxuX3hv=1!#?O}ffx%9D0n{6#uC
zV#~C4AP@I;xYQLFmry*#8|6g;=BsfYOzSGv%^+<&pPa$imDr|~bLO7~p@m07>+S+%
z{c2v7d2`=YtP!z2G@}aY(GxA+H13;iDHgI;fj^|;EP2Ds*I-;dHN~R1!Es6HTf#8^
zbN~P{CYGQIsn~Yc#alBlbjNi9u2&(EOF)a^5Xz~3aZ(x|5C|!buV7H1y$O>7w|6}k
z*@7EwrE^Mgp1gIUG>NX4f|nozw|v?Gan5l_azo&2TY^(0Boyz{B8aP-`L%3w#72T1
z_L|i1brBmDFi!4?Vt^f)#;R2JKyzS$Q4CN>ln@ZT`kn(FZpgwt!{~iuMX@J*H>ar`
z9ZG}Qa67kdv<RzG5sp9`i(x6@_Bra)vqTRMjx8w^4s6>~O-mRephnKnK^$8`Pbqyu
z<;%0E>DmoHdsIC0Q3<1relq#P+MP?7VcLf|N2x!!nlVajJ$ySNBW<uESpkbIqz_;t
z1|5LAD9)}$@i=B>6i^CKYjYS%d4}RLCfvU#{_CJ|QQ6@j25RA5w})%xzdi{C$1pt@
zgp7r4UmAnik0fo!Fx}o0)EA9wK4@T;_Q7`rrr1?m5RsL^i-cnlI<K8gbWH1*lcdhB
zpq)7RDa)%R={5mM+1Lz4=)BH$Y6&RyO-RLy4ujxIn`^wuj<LptW7yKD9S9&kS;6@w
zoNjmY605Xv^d_sk&6oUlL2-a*nWQxpw~R&gSo-^j!Q~t%Pd)<P@Vj;nIL!~!o|0C#
zd<~!H1_W7PXsg1Tfz*MHM5-vgUQc7C1%(9TZ$O>-+>te{kuT%ezE$T1IVIrPcHm7m
zPC<V;_^eN$YO#w)?pg)NyhHuKzZ|>7H#(Y5K@`Y5@M`w-*U>U>L7&X*W4(9kInf4E
zU7&UjmW=}wt9sDVgjb6S`r7bP@(Iy`9o#Jp=Q5btqz-)EzoVydv>oF5=Cm>s&Y>-u
zfIc<E25N&tqES_G+}#O93SmGr7pRp?T4sI}i0LcHFg^jUGEk#RcPvA}_(*|kYb2rm
zI_&G6aEy0l`dD3TM-HZIPe|9JE9UDVISwS+FI7oX)$<V6oSN^NIt5^bkGodVIX&)*
zjP9!G$lNs!L2Y^mdOI_I2Y-&Rz$cJn48K?YRe7p$8YB#=bI&tsRssN22=*%dhLEH8
z1FlB)x>PVuVw2r(JbROFhb+SuE{kp;B|DKQ>kD#y`s76-U|d+DiMA3n;woI#k@@iF
z!IpD5DB#qZUNTw6=FZgEd8B0=@zR9p@W-}cRYwvX`(LboWBqPxu*IL`jf`<tUnJ$C
zh<^oEIDGXOUTpm?8j%2~U)-9K>|>5}&LsJ^EHoEBre(zIk^U8Hsh`OTvIlB{G}uSE
zteyDu-<Czybf{DsiN>IW&}W)s6@xY?p~1Brm$Ja?^=~3Od>^;hcQDC5D}jBj0#7vD
z&)KoE5-dbiwi49(rol=O=1WwS73O<P3+QHlaQLU9RIdH!5C0rM6)#Kt*XH6uGY`-8
zPtVfEl#CG07@K^r%cNX4bq&*AsF>_B>7jr7WmcGqL2}Kpx`icovi|&6ZpuJP4tu*t
z1_9t}4j&fln<BO5w(+$;MD;o}90|ttk#+{&^eZA#3b#>}f&rVq?9^D%xVOug3FU&A
ziKiK23O^Jpv=}gP)=S8q7mNZ=%P(<XA4mUlD%~Qu6nQ{>A&s$7UH6nu6bDfh)DBGW
z2;u`2{pNq<u+ZE|h=q$p<c}DeJE?3RB8tzZ9Gv%CyOdy-5zLg}40{_gBuX;IFjT_9
z8ilBCLa@aCBc3r-!m0>Cq{CE%U`)6+d#UTZ<_lj`V*~M5|2LB3fz|u@`{p;0HWmHC
ze+y&~9au==7#W!VKYNcY*)f=$95AS*XpcP#_ru=P$e~|`n4WUrIH=(O9nn3r2ExZ8
zp`2m65#B3F*6~T&r3l&Rz8=>6VZh(1Hn}JUaBQ%;hcJ`;P})}$m^9Jb6P(mOY?KzF
zLUp9p4wceW+KDckN4aV*kOr&b*eG2ok9LdrDK0Vl;Oad0h}=s>?A&j;0i*rKbEK%&
z`TKB1m$fpug9|RVQelOsW>*x0E0T%O|JDs!1uEuBtz6S9BM<Y~-eIziHPS=dNm&uu
z;!hYXH!1^lSGnpoM51(~=3F>kQ_QW1Tsfyvosu=fqK6_#>)0UZbeA^3%}$+g)%<IN
z&HBf!hBdaxw1)ZGw9&Ya0Ii*+#-*~r<5P`l9Xvvdq24t_2qDWxs!|#9kpxLk2}M;3
zR@X|WV>^*FGn0jyQJRTF{krF4-jA&AgExNNuvd|#va*;)mcl*EalxiNeWTv5)>kCI
zuS(aTq>)jLB>TOr$(7Qo(%P#!fsq$DV0Dy>QPwxCP80SW(D4op0gwY3PPR$AhO{gc
z2qL!h%%NKAaTs{zOCmHGT39soRb*HwE#(fm(ZQr89vUt@ikxuJP)6yD#&xP`zlib6
zmIToW^-C)YEX^Hn*gC93>&djtC;G9S%td)gFcHn-oZ7Vs&@A%=Y;yY{ZItcu3GXJg
zVroLHBQ6JN!;Bkx0R;Ch;eqhJVs@>=qw@+JuQ7O_iizt!u;`WW)%KQhwwD9heIT+=
z*?;m*Up^Ozex{&qQ!BIS?bVAWw-6m=W4_m?^{+>J_V};C^=z`w=lS#11J9>YQgzUg
zjthM`K3V>4J-IdBgVODH$Gfln_3I-2t(LGJ8`icSY`a?<K!5t&LYAJs{PDF?_wB*x
zXgYAY1LJU6kc#*y8q%PmRK7@lg7`o2(ulYgmmdRk>Rk8Rhux9-F@E*;`s22&{kFW_
zI?kTxck`FW_h&kbPDh5$)E0qLz_bPtM6`Vq3UnV_DT+~D_?KnKa`zb^r7P=cy3Jli
znv#+zRn#vWpc4P}3iVZiUF)OsCtv#B5m}KUYpV}<6d0w*pax_Wd4x`Gsc4oTciWfq
zouf0D_xsQ5#<$PCAzS<FLvq*m>HbCkvHPymoAmenvA&($w!H5pz?=L1<~e=Ux#j(d
z-=3a)_qhCPbMZm<n|dge@G;wt{{6fhd*ge)b!grhFrG3#x;e_@Ac~kkNR*GsY(zrL
z7z8O|lnXR%UqPXO(?~?D$$`YA8_`s8BD{&k$dO}X2r58DVzBEYRuD#2ERiLo7}HCk
z=|SJk<Y+)Y>)h=W)f<M>TdN{B?aUQw3{w=WrTvqJN+ed4LzNpt6mY(LOC9E=;J5XI
z$X8zh3_N4S5?R5X!34}GxRPpZj8~ex@Xwv7F~Kk%5x213pK{_4RpNAj6W%aFb27pN
zW672NO&kAd;}7H<2-pM%LW-u6t#46Ezq*69((htY=;AR!65QTOo8sd_n$5R~a7$}2
zy{GSw<?c-Jzf=S&6v@lk)m3y?%JVxs%?3gMDC7-MR5}HQI9wG$zJT}%TotItgnPNB
zJZ>nk1|RgrI0a=F(MFKU*0imzw%%XvAl5+YsfMG~?f!BPfKR6P-NlPbn>jd{e0~tb
zJ_7SC?twK(F#z~cdg#5wEXA?+2eX5WLp4`S#@FEG!`-6EQxV|yq5{DC%G9m_)eQ&8
zQvDNej_QdsP30P7K08u%_kXWHCh9zO)Of3?(HCLjZiB?_Mwo?;uz15ejuA{hbZNpD
z4NFCeGcath4kZN?hteAkd&&(BEh>Tpimjbh3(0|}5MrSW|8)P>m}bw?0GP8FnGVZe
zs8xyru*$%$3T57qFh?ITBn@h=07H-!)jC<dumeXN5qY)n*u2CAke<3_zh6|php;<n
zh0>ulhFBfK$Ix`-l3;^KPz8}Aa->6coPhSju(}!Y7LYog+ovHTPu?*>aU~<|1WEB}
zh%gSJP)ttYuCe0mQvBGsbIXs$lx7Y3*gb)eAr-=eCMndwJ`uQlF=m^H0YYl4NXcsX
znBF;y+P+CT@CHH1t2B|4$!XXieM0?I(5tcMbh-kzsul4$D&dl9d;)F<f2m5?uwu)>
z9?!iZ1k8UIg|e^)b)Ia8<Mco_*llSuv``T8aUx5AL>R<Ctv!U8)T;0@mO}m%zjFf-
zijmTkXu=SOJ>x#=VTZNB0H_?0V5#=$XY~f4200mM|0#<_8@yr+P>QnDP$ma57>OwV
zdL7Ub+s+jVNVPXHv&S>lzK5&`SMxp&RwXLg_th19K=UjdmuRscK4DU-(pXHYBlns)
z#|ZSRihRH_2ovQU^44I?j=Ms*h!YPhwWMy=o7IS!k)in2dXXnp0zfuu3#Zyipuvw^
zeKPNYvh8ou0Tel;!L_^oKI)K4&rsjr=cpcoK2D5{9vcxKY;u@}u<0j-vz>RTg$scn
zzMw$Vnpp|txkz^8EL8A-8sDKdFdj~cB9kc|^POh|hMS~*i4~6UjyP3$I7Ye0{g`P`
z4(uvoyKd_>`N%MW04&0Ootq<am!vTt+=T>=e75ZJw!v|~Sra%x{28#;)!|Y0X%|15
zZh-MYxmlo{bn`K(*H!<`>H>;~>n}bm))E!zz$-v3gV>Sg15w;p*wqge0I{S$6Y$UO
z^hcl*k)RACcSJ0_@#S8D%qtr&pgxi4-HNUa)b%H{Fnjtt0GI&A{XjzI4AO{V!Uu*D
zRCWU@8|$(>F3ufXQwy5^K7(jCGKkS;J4CHi64JIoTcqk8^#X(POSMT{d9|Bl95ag=
zjpjVVHK_VdURjidKGO9C)4R+G%K))2X)htl#Np_sBdhMa;S#7Oy1oW_$Yof)BO}xH
z#35U!!mJ=h0xW-05R?Z$wT|x%R2?bYD@M7E7^m+0{#l;VRd<S45EyfGif29|b|Ghv
zbR&|rwz>yZwJZj3mo&{+H0pe<saVkdc!}(I=a(?%ndX<AZgMS$9qS?XUs~`Y_Nh?o
zVlP%C!fa$j*XdCOb!U8Nh1E$o$MEu1-hF{bv)=K$0G#Oa&<zYF5_!fG<j_aL9`Y7~
z<Vd1wPsKCY|L~0Y4(?aG_#YMebZ@?>Ao(UHn-g$3dN4n344uU}Yj=J_HZNg?ixwS1
zH5kVNil5#mj8%>O)s&oKB-%@yk_gg=8`o7RJ;!6dAYpZc+i*Qbuv#`lqJE3Daq3hO
z>W)PV1KJv#`o<jnt-d_Ab)*~<<|t-;Q^a>d+HB+zv<;rbbMevSr~RKGCMd<Yjdsji
zlL%I)En$5EY&mI!(y6sR`;qu3X*<k+pKmwJK;Mu(kh{R{h>jR2o+9w#BhIby_uPG&
z^FyHY{h34b>}|WmsUdRGTv+L;FUh-Hzc$9)1qhh_`;s)Cae{p~hRvIBcfbFX-oy<(
zY+J(t_^dat(3RozDjIEyvi1r#H95L7eCX!KME7pvruX@9P}tfHGxFMTM_Cp=?%z@y
z;gQu0Ujd+~Gxc--etl!KA8LLWPkJJpL<)dmH?^*b(=Z0+?;=UiG~BySeuG>JVln)9
z8jTIOkbqf|d%I{+l3}w!DFB_?lJ>@}h+XgMCo9ejA<~W@u)uX~n*w&2FG&X|1%W`R
z!r+iG+1W&zI{4c}nwNn#ue^0^OJaA-KVs~F{C2k9cDApNvP0d3rQzY}=dk1sOcekB
zw0w#mE&uZ&8GWj-cbeMQ&$;l%__n=&`|YFRZ$Dz+jCe*ifHpLlOHJ<{qi~+)^xo!;
zHZ-t5c1aWOM3;H;etYZtVL;Mls$qXDlS6rpBd|O2`0fFsE3^CclKbpzKt;VYh3(6P
zhGBu?yezgn=|ymqMGEqTy+tl;R!fZ949o<&gau+jFFz5evgj9U;#X2hGTH;3^NpqV
z9*Rv$mpEk~K*Jdx98P)fhM|F!_}>C`?X2QTBWV&D%_YJ8;pAV{I>XHm&2scPercnB
zq>_KjaV~p1pva)QyM6<FyZ~?0^L-0_3^QjB&)eC3FbHXw=kiKS(Rm-Y(eu&8KlvB;
zW52uek%xm0e{7%h6^2^v@q%q?hpMNvp5dI1oVVN>fH}+<=~qPFYOg!(ef%xP;{}4y
z7c4&Ikhl|tv2jD8P?8zs($i}UetQ&K!Mh82drQ=KdNTaDf|>Q#)tej8D)#ci7BeLH
z5>z7u^LqiIGm!=_r2P5&RU#_?uTu0Pj6MiD&Iqm{>#SU$Pld?1du|z^B*UV&QL>p}
zN!Wm13D9wc>It*h`m}ROipGA2UYex(WIm9N7){%$z~hu&N^Hd<ppjHvUs8E5>Ds^c
zO3+%P9db+##Ks=N@cYmpFk;ZX?ZP=d>d84W+$ex#rt~<@mb@-*nESa=8dG8VIU33O
ziM9nG`1&1IY{mqClx*azQ7o}69=a-)nb&{A6@>3WA9vOIO|YR8sjNnZ0_k9N9*va7
zM~RW!&T^EC-xhX-ZMWyxX2-PAEb!7QkADE*TO{<lXXi!nC{|dmv|F2W`FEYG@k@zZ
zQ4Rn~QNG~=%R8LT93*sX3%5czN3}6JC2n$IvfgNA4KQOS4W7*tje>^>&sl30ch9=^
zV)J`6LwaF}>eObCZUrut34}Kxow{X}%)s_x)^I$7Q)gW_q*A%W*41bZFo+m@-SjRQ
zlyqf%)i;W>Rw)G1v{mbW%qd0O*8M!ME8YN8-A7yKp`c{SVVZ~1k!6`wQwqfVbh`Kg
z>D0x}2FiX~rD@#TrWn<{YN*wL<T?69yOr#2mQ!hkRG<8-gk|@b@fx`LgI!da?Lcxv
zA8Zy9E|YAFWa~dvSmi~GpEQmubU|3;@97^b&Q&P)QZxF}oTF3}s*<a{LXIl+XDk3f
zjdr5|zgv}Vld(?ZvZ~%Z>ha3IGrMTzLyA}58u7=4p?iaIh<X(4u8`9t4=C%dVuHCQ
zRxYA%+->C@`@2q_z-~6TDNcN;B?-PZRnNl46vEv3v+?fI2P54~Sw}xFf?4@jYPQ#I
z8>T*$3J?1(dVY(r#`z{c7?T5Wu{3}^6ZB+f-`e`*+cq09Ax5%WuY}DSaA)YK_N1RJ
ze7u>#<MyPQIAy4)jRTKwp;FN>s0cg*NzOw<f$Pn^CIm9VBtA%$gB^vQGiRoA1l({V
zWt}}~qfGxaS3?<n`d$^osz<G*QU)cZo!;8TElXnfI>MAJB2QeDTU-p=`9Hu&&N3CV
zEo{75wo=~n!{Vi3oaHl2|J4p6ToIC?HW18u0|cycd4aoVv+-qN9TKx2ZtEhPndDwk
zlVRlX0MRYOrr;lN<ZcX?=n$&7r9EuE`pLH*!p>eSvJN~>+zlLH3bL0y4g!0=yvZ#o
znoU=k;*1`woMB_Gn-$iwOk+SBNC*W))RjYjUr(!N4HG{=wKN)6Y`6?>O1HbPe#PO7
z%AS88aV-b+LkNvAdC;M#6HA<W`<W2PN1S+l6N;7;(5b2RPX{c`DFO9pI08u<p=*XF
zL1d@-UTycqt=5bnxV}`ILW}I8FG*-y9&sjU_E0RW1v+k9UX>f1p%&oD)sJpv=R=)x
zU}yQc&dVYW^DH9`V9|P;jgCqZOv<!46$iLAz}8)fj2krn7L%60M4V-^xc9R-rIfl&
zvA{g!L8HxoypBRU#4Xp}%``ZL;O9^SmuwO)#QzwLWs+#ar3jqlLnG&{h4?#~Gd7%G
zKpu*B8%ls#z!G9)!UFjCB;2E6e)Myo7kis^Z%I|a@)B&{u`agSMze}@gn$_v@6+Fh
zO9>33iH7tF@%ir4u{ONsRCXhLF(2f8dc$ib%depJuLfpkf#GjqzKJv$qA&3wxr;Q!
zV+Z6j;Y0eAJ62Qt{<A2nl4-0&7kR}Nh>siR!Cw1ibN!5L$O4dpnb2wOvDlMCXW!H?
znSlLUP+-v2s>?1*)RB?hpNeD{&$LG9i{&D+k~2^CNA+U@c9KGX52=5CpKa<8vfJOz
z2Btf3znAP+y$}5#kORdLrD3l&>rV1^^j74&xm+-OzYW5Rz#}Ly#LOzMxM;|ul2D!e
zg4>m}mM0QtToM5Dz(E&~F)K!kDEweWNDX<yCxnpKN)`yble`zB1R~Nh7Rf))aO^|)
z+xX~d{}@81?di25-%}KHx)9{jhHjzq5=@i?>+YN1%epZ>uuRD5$^e8MSKlNd8XxfN
zG2ll*c>=&Z4Dk{pv$C7p1PfUSx(Co>oJmz(PZM#|nhgL1B0JH!--BC{t8eKnou-&p
zes0zNo=+e3NlmVJl!a<@(*_mfhHNWpqoE6bY;Q2%gaf8(%d^ngLDNs|biK&uzGN`5
z=)#Y7S%`MK@P`Jn>%EpWv{sa)&hLTIFvw+=iE){eUWeT}XPZn`1H<pOhsOm)`Y6Hs
zhxi0m%@;tR0Kxt&<MCbJ;JD&*)jHAW914Dg*1y9PQa%Wy*xF)|0P~Vj`ltKA;7CEq
ziZmi+jXF1TnVyDp*yTzkS|R^H;a2QjhuurSg)dUoWM@%%3*I;y1|9}_P|n4X-0MLj
z+)x_|#uW2!m0?W83>5IPYCQ37hl`Zo?PgZLC?$ZZEkj1aD@65s9s^h`Qs~xNv1XKo
zske+NVJ69RoFN?=nYq!&?l#`lOIz&ytPAR<mcbUcy`DD|0SA|W!cuRtl3IX%m-flg
zhdf1=Pc$-^OVqtDS!?(N-kG}O?DH))r>uE!bMyU=_C^t~k7Z8B@9_gZK_hVPQe5?k
zbSi*o%zIjqRy0QjexqIm%I*LRnugknBB)N~4(Q*<+oq<6UzjJTKmm=?;|Q}KVx24m
z!;oZuL3C9z(G|(4Kx)TRmYK{>$h_X5NJ`hR@VA@4z>~qTRGaD=Df|kdF~14PE9B!O
z8wsXGT+MBfF_@>rzc$B!;l{`!J!xlFnW_O6Ez7`#x3o!IlYxQO43EIZnnCc9v}P4)
z)Ev{K3?j{u%~_-YuYD7-&zRAfLaVqIz3~m5HnajhM71Qst9T!Cy}DJpSXgYhj&9Ce
zzs_F-fHfpvQL1TC_1DQLw(^8g^<AJpPKqQ^FV3w41xYRok!(+i$`03~tl69dsrUe!
z!R040w@<a+Px894z*bD*mUE^T9-C8xbKq_wVt1@`)YyKLaPf;WZ-Zr*sJbKRYj_;>
zNi&U1w*Y@|I%?efo-9er))S0e>T=t>450Bf3R{Ks&2Ebpa9zHD=@S1FG2W^HtBa3T
z9M(h>Q^L#O50c+PGe_&o&fqG?#zlb6vU1?(37GqNB3={qLS82Xm_}!4XON9%;17IG
zv2|fve-rqP5g#Bue$anE3=N^^3>9?_3y;lP<38Na255?{Amv1m{o!qlP`?)BFOcs)
z;q7!+ljPZ}<xdU&mI2n3|2nVK*`}{?wcsz8P8}Xzd^wc9pS~UdD4u?}|8IHyi-j$@
z7mXYxIk=6RGWlMWI~tH@{?5Pc%RM^KhO+gydUxsT?al4&Sp`&4E9N$nkXk_Kd3jp-
zkDoxF#x^A>kHPc#_0{m{bCT6od8>FMr$OUS0w$T2vKT_^6b5ZWqo{4gFB8IlV1#Oz
zx9+gUr_J#0$wv}h`g1(#M?tUZT^l)nMcLG4Q(1Eg6&xZ84JZL#a`LH+bB6bZY8gJG
z#;ZjmO(dvh2eCKQOkZ3TlYMcN%mIJs`@8zHV&f!<6~s!m{xv0_4vAm&m!w^?DaHL8
z@3_PIbW>oo&&560h<6OW9+F=Cgx1pczWda63v&hWLz`1>5=amx{Zxuum+k0?6uYBP
zNRts9uV@!!sssm=|J|mt028Ym!2Ph;X8@nEs{aklO~Jxccccv65mRM)@@iFgmUu0h
z9q+(^#XqOkuc7!DnJa_SezGpXCZ^HmMG$XzX0x?s2Gd2ZpGRTJ%2Yf7hX@_mR#A0<
ztZoKfy(%V2IASs;{?Ix)@;t_j1DnOT38mNHcp;s|yKDq#H)0pPyqf}9Hl@AU<VZWe
zakCtStn`s~l-3ID%e7VCpeL-Bu1u^c8=yK|X(t<_GwVxdHQ-5RTvUP>S*}B?OO{b@
z%rMVZC#awtl-9F5j+2fASyWo9i+>VwcC}Ii6+y@+JHAR-ym<u)LdZlmhSzOyk@BT}
z&0q-xXD<ZQ8*Z!m;y_!=3KuWXwCmK{1wvahQQY-03Z!1~Mougq*av1S)Uz!pcK1Hl
zZZr>7i|z0_b{!L@jXPN^bw5PnDuYgX4+RK>xPIm6bfIt-3&GwTvYZmKndNIk3Oef<
z<J-B-mi}3ypwY7uF#_3J9oG=nwO2ThYoSoNASwfxQc=1wFh5FVTJMNs@{5odP^m*Y
z6Hfa~-N4U#yJgDWgP_xzYU>>V*{$zXm<_awuqisGUK4;NE+~^|vI<k5?>rfj^w%qy
z(2a`FrQ0jp9aNDfR3`01Ey||OGAY%!6%y!!d*m%`Y~o;cBCoZiX0qXMp!ee1|Aa74
zu*L$y=4o^|J7e#VeFQhL93(bq|I9)O*gWgNjI<S$sxTPMN;t-9r3fRAfoZR<POT$a
zkt3eD5&xu}rLEUWZ&qEdJk18dlCrCA__J_vXpF_YW9%Pc(I9fot_X#%9sPT5DF1EC
z#kmay+L3$u;baPX0_tJ?Fc*t3^knVSRGSIdr6OqPv(BsdbP#hMeo{+COmcLkFlCT`
z=K?pOBU+1)!w6avcN#<5*R=pGSu?ct%27BU3XFe#?p~wl9s16Qr<pnuQeDbAx$hi{
zfMZtiZxDlLro<Ud@IByN(YHp*!Nzzl0CL+k`34o{ldb=);+9o9l}~5X$vVHPDjW|m
zTTXSth*jMurXQWw-RqID6eT@^b*5xN=Fnd!l#)$cm?b&19K&pFOwHOyp{%tEU?QLA
z$ufT{S{ixZ{oRilhp1<K#9J(DgWk`A+eLTwPfpwNL?V%{YS_JoMKH5!&g(89Kt4^Y
ze!^QfK0N!d;}7s>-kI>MPUcD7CJqt6No6-;KRwVDJ5pL~-(X!=4XcJB{j+=l{0xzK
zi8AdXBn*c2u7!M#c>;$03y%JI?BCb}C&>(+{d@%qk{ORFze8X*AA%u;T{iC?L{wuS
zWY|gl`PXhqcF;!|Ee|wI7vhN`hd>a?u>U=$fB+%8UIAW~vU2CUjX@5{e#RTXkGvy*
zWwCdwj=(U9Hxzts;!QL)^HGeLRay1<_~5Z_>gMt7kBNmWF2G5E21K=9@en!DmXn>i
zm9=j)75?ykifRgbsIlDWD9gf^M~U`Lvpm!KZKFJQ39mA@f8rA>?Gw5KvF(iJT3fgm
zvH4fo=Ar*8UIjWdU)-fXBNip#P2bG}q_pVQwds4QK=jyCPRr%c^v_LD11uZVaVxR5
zGu$3DO^vS8H>)Rmlpkq~QH`^LNfz5R6KA_X%oNii>8A{>ng0<$Tm&Z%UR<fS{;Z7;
zQK{u=m@|vUXZ{b~PzDzsE7LY~V17(GYU7<TqIXt1>U!Tlq(F4?wt_AFQ+Z{mzZS08
z1vc(srSOGs=)000{2pH{CjRY(P2xFi%8Spd2ix&%aT0sg`<Ubb|0=0%?|)`XMgvNe
z<OLn>4&avNSY2*pFi@RAw8#IJe6TY9hg$N3VbH-RPe!1_155rg;*JNtZH_bMrhx`M
zf|6wZZ!@gPvo{RL$!E6QoFE9GD#>Ja+;RZTpCH(_889U*;pF`7tgDy1K?-^1ghT00
zZ^xXhb)8SGBa8|AQ-!Ug_B~XdNtdlr$m_)2O#QFVuW0ho5_O(NM2r4WE`-0bz07-H
zf5{O@VkC$K9Ab8aL-Sn_)hzBgX5Mbp%+`-TKwa@f(;}k?PasYGEXTPKi8Q)+`gZ`^
zJ7e4<DTG}oBblI#Qyjy$GWjznoqUqOE*^nL59ErSxSwB<2gi$2zFeUGz-C<{S7bcf
zFMaUW`v2_xIRhs!Pg>&-Q?ffr4ordBA<gAW1>$M~(qhBPj5=kFycA<z<YR6d<JVe`
z@z1FINvF$ET=(5Q!%(4}yMMB+m#PFqaJE&MVw+~!%Y#TP8K9u3ZPrv=hs?broE(=@
z%}K7%tu?F$UJcW70#NnZFpZJb88gkKy8x3K)<@p4f9isno9pelEr0<T!vsp34^iz#
zlY&yE6?L6Q2Pn6`lcwSff*a1-Y>=dnJn!N(!{rGZ@(9l~!`iuQ<K$#tq`|_obHNS>
zuX|%=YWy_6=_DQQ+HW0AhCY*iO8NE5AU+YLOWMvmYYu$EZ-^uf>)0c~$j0KtT#2`<
zH6TNJEm1Aar;((|sUMTm9J#-sGPkYn37;|%vXXP1xUB%r%eKyy$YS&DdL>Jx^h3H+
z3&Sj?DD83_(GQ)MGfFhUdui7)Gd5RYcmYi(3Cs&hPKl3e@e8Lu8tlJe=e!%|TDz;R
zbzysBnk<?v{muw}|I8<LoD?Z+oET#FA-oH1LwdoCC2GFEt=1jf!~KoAm9@)~3^j3^
zp@r3QDKP+Y4qk1e+Q!$F`pr?7{#>RMt+xkl<bUzBg^%8DqdF57Se}@b?$JQTd0Fyh
z<fSN{>Yr3)Io#!5ZQZd46q@YXiUekh)dR|dy?stEMo0OYU1J+Xv6We3%7uf$Zj-Yf
zc_KyH&ehhmJ*1bb<THzv54FF}X(bZdSbB3oiCY00uN<s#389kg_Y*6U9-5|e;36qh
zrduq!>OQNirNU>%zYgna5Od;e86P?3{Og3I4>uape6ii7%7p{_U4U8^#4-u$A@PIx
zwXi-s;<#n#^73q^Z3df~<RJPqST&A~!OCTHEcGPZ-~Uc79@PFG(eRKdy=su+G9;My
z__YYQd>rESHuc;$gUz!BHg4}3zn7vldU5amKLDmcS-)prA(44KZf~B-5GurGt`!yf
znj_mCdGbwbw?=^9tk^ixJ3^+}Ogqx7%;0+N=$kIS;@QnEt{tn~>Z+1mw5v@@mGFCw
zGt`LbQspTug0y`ya_zJSl}mHKAnwbDnJ0PPQA#_gmlJJRVpup}(49w!v{yIbI&M|$
z@V6RuI~@Uka~EoBpxNN(dtu)S{f%`;oNsNVKE72Q50}~C?r2q~gH>K&a8Y#IBd;2%
zqzF~2x*Q(#UAf+6@5xl*wz}Qyr_}Y07p_tCh}QW?LvA%}$T;NVcUzfG3F%=%{%5t=
zwTwrN+$Xizbrg)cLm`eHzPW#ME}q5)J7c6`T=;r__2AG<!vjt?erPX1xky1+ozphx
zZ^r95Ym<%Tk&=z2|N7$t$z7rEqf4H%%-9~7V>w~HnJGS^J?47cI+SZsmh7#XS6}tv
z8*iE<Lf>R7t+4*gbcx#NYQX4E!|?g$#^)<;G3I{=5snTo23z;0XoTnWcTTtR0mqG|
zVW7Z&0Ap*iR*_veg`rJ2R;g7BYsIUcTPTOtGMwzEr>}4$&z4+SDAn9^uEX)U3)5=s
z#r3sP-R`frm$#rh2*H@v7VDyZ?kF$#*Bb9qpLX>d!^AGcs=gDDJxN!|`noDticlN5
z{6?Yi@?vCH|6ZqkEpKqbnYL%mNlb0}_9tk6W!X-Ui?mI<RobTA*b^92sr<BVwqyv3
zcjsK6_tqT&n5a9va)o%qQ}V0<r#-#qr@l`;xsBg&D%S1yaoXKm%<wPWrdJ+eAKu}5
zJG);>-^2qbW_vhu*XJyusN_nQ(aEaGC)($xVS`UhGk1NWRTV$3HTJ3#tNJ?3>I2_@
z80A@?^o^0Bv@2%m6W4y)*~()Z0xAd}Hlj>HCT$vC{l<u;Q}To>Jg;*X-x#qJ_P;S&
z>YhYG9EZ(e_qYDCNXuM$O_+b?mqkGf?bVWZe}gZJ5YjQv3A@kuvdF}fW`17z;Fm?-
zQBH5}`LJlJsi<cAKL)kXEDB|AWOH<vXJ9%G12i%>mr-m%C6^r!LJ9#nmr?#ECx6|W
z(QaI~5r*&m6a;#cUWiN1aLAb=Fp$J`8bCpfI0f1oy0GPK;?&U^@;Y$dzW?wXQBt)^
zSJsgf)PPsqvs`k>8P0q|a&s;uA%&c|O5lP*AquT|p`^g6KPnPRo<7V$xb~8UR=Afe
zXzR}yEv2NrnEf?B@*;)IWqOJ)lz%oA(O#YFO31CGUPL?EdRN>6oi(O`&{!9R0<qLC
zaYfPa-Gl}RmAo^EL`A2nC6Lh6mZ4bts4AW|w1Q$oTPnRMDUj7(S90LotQRRT=RQbb
zOi$^floD84^WFrL)@*8_2DYg;X`v+PlH;{dn_93?iuF;}6iTK36oV+OkAJ5?zS8Cj
z_?mKWx8iHs#CvP_nq>`AEu&QXC@Tj^lbWmq$>gJy1U(T?19p{JA5WzqGBw)-&D7p6
zbSKTaxKO+pQ}0YeO)v^CTBvzyK~t!+DPWSTl-`9>sCAk$<xth83MI5eQw0evPx5LF
zjazS5TWFHH+7#AQC#@JNsDJdY87ioCAt7Hg?p;6<h)aE#DQKeJ7$nhD`zVZq=Ij$G
zP@a-o9f}&13-{h&v93F>ixxzub~EgF(h5|n+O;AHMeACT#K9;{PfIW^(^Cmny2%PY
zvzjIav|x?xkry?XNw1KlFA(WvG?a3-UT$8lef$Ma5~ZmQdAU!raDTUS?PKwYNIp()
z-pjR4^?d<0>)mEdce1W+A(u3Dbql!~XljRgkGZ$sM=p|k*`=*>9dB9l^j&|ozLv!`
zOG&5tctVur(dVD9p1g3M!0;;Ht^WRxe?sx0q0f^KZ{Hq0e!Qy&M2?C9NG>r!nDeS(
z_0{I>=Jffym&fbSzkd)`&t7h~>(i4!)7w9P+<yIh`*OPud3x}4bFxJzPu;;J*r50F
z)Ww90nBKTHT`T>KPdc<Zy>Z89P|5TOyiS5cn%;<4%X78<Nt-`m_fvBNebv+4fNR^=
zE8aYQyn1%JIexz0zF0kb@^tn6`r~$<jqiVbx5l7;SYxaw+kf@R_I<a4sjKhS?>8S#
zkJs;KFXqp_UB7<w^2^P~7v5)uiNFILp@ma?*ZpFaUhRvMlg-?gOd>pbF?n#Grpbe!
zmF%p<X_Q7ircvTiDx)N$R7a_eQt#tvjI#?BR$p#TU$0O50MgOwFRO1>U*QdObc7Kf
zZ(qQeqQGs|Mt{+*T<{0t!pB4icJyfV{KKoQw|DsF<ln0=M(fAhH=C2y^VQ$J`^Nu&
z_TzT@?){%us~_HMfBf(&9&dhL{q*7IcL%T6zpTD}x&864@3!Ay&QJdZ`-(rkUA=m{
zdA0f_M@liT-k%<?PTs!5^FJKI=^q>xlRsh>FQ}~#=YNtgyT_KNH+^hG8XLkEd}s9N
zgSZn7wkFbIt<dBb0nI;OeZBd9vw9Lfquk*C69N}%TIE&>q`0LJoGdnvGNQ4-S8?M=
zaF&QXF_kMa7&oj0Mi_z9Au-|wWGESZL&;|{Ao|4Yk!`mO@IlsE$p&v$IVB?F9YXAH
zTr4TNlYf|v3se!cJcx0b9vs4-E@51lLqt#DlFz9H##k9k9!7U{LIElniUl(v9AHv`
zJ_}eDd(V6>f%+1*dz7FrD2Gfd(S08)!V2h%68_p1A+WH#f0#G|aJB>lSU?8YR&rF}
z5}henB$~5oDG`Qtbtb^D&Lz1r0&ntYzkn2sm49UpW3v>9DzTIJby7eV3cezM1kj;E
zF$?Ge<QqHZ%u;|r5$5eT<^o9Sd;!CCf`f0k)j9{gM0S#z7dk*7YUX=f=pdHUH?KZy
z*YEp!x7rR4O)mtKPDCYLAR_o8;My*5x>Rifr`mFKKHQ@!7m7%F_Ntm&w6d1~u-Rx2
zN`G#b!0R%soip8yDXhv>3ah*ig_V~7i^3XRpfDMnC4;kMaFz_tlEGOrI7<d+$>1y*
zEGL8IYPubJ<s1!p*?=#1=xPtQY%<dUUg`p2GP6X_i5)dmeD?s35m_LzUW{=^pdw<!
z#48k?$l%4YEZ`WeMMRw;g)jgJ_I{ldC4VPmX^geeniS-PRc|i@3^O9~fWokZ5vAdy
zk<6A_;ZgXC2!aTumEZ$%D^wIo>#clqk4T;|&qh`pijzK^kwL)+hZx`jM&@ykoj(xo
ze76NA$CCp_Oxs~%sY+JOO`L84EUhk}-HN{92^s8To@~I>d}n|&M^N+*BJ=cK#D8DZ
zQqrpBRthGCLit?51SkU{KAeJEtp|rN@m&hO0aIX!<_mV4!I<?S?bZX%fxj-{9yw{O
zxdfA_@K5Yc!6QMED6+#lI;|xV)Iq`z<7v$nfY&nN`z!?V{@C}Qt`mZhnHqc|EBkhv
z3p0f1+)~T`+LO9aW1tnHVA-oC(tmTLt#z-ezR?9P-=nIpA0*uSRL#i!gzU~C>;h`t
z7^rLqP}!#emG5z%163UuRR>13flO^6Qya+C1~RpQOggYkM|(Qj(}86=uuMlEI{MJj
zhwj?`x^aNUI<U_88HhsgJXmI-YKFPxtLp$MobO$0B0!6#TA3C=6}SLSG=HQJP7$l^
zdMWNEn$1&D7@Z@w7=jp;P=$doaz%({iIw0LY7!whjhq&k31qQoYX|`li{PPag}B?9
z$j-6t+$cyOjO`EtbW15U1%4Fd_d>eF1|D+BRE887X3X|mQtUMq`$u8X6#Hr69EC+4
zsO16SLK5Jj@)9YE)d*Jx?|&G9`T#frU|``Tgu(pD;C9m(3|cc#dE?l~J5LdMh9itv
zOn0wegn#eoJUhVOr1a4GmAagMeTy#pHtDQiM&d=3?@7PDf)PnW-?ZgAtzfKZE*fcP
z9=x!Ik<6kaJKu!QH8%TyQZS-$#C#0}6C$VW+zY>(1q^eN;>VSUPJez?0503_NI>zR
z<hxgJZtx&}gg+7`@U*qxlZHD_TiI*#)vjW<_@QL8eX15j?W}E|s-A-Bjuc@>RnPH#
z2WOwE1?Hf?=L|t|xsoCHHHc5j5YEotdxoHc@@-JQ4a&DO%J&N&XC)q`GD<Q^b(GpD
zEy4R|`>z7K609Igv40^CuGRC)9UDi@%$Z*-n<VqC!6mY9q^O1=R%EpbDGJt;U&W~m
zskGh`=DOQz$n$E3{rY__p9{ZjL`Zcx-_KbdgY4O_L(M|?PK4jMx=siv5VM2~OtFBG
z95Mdg4ukw;x}G0kl?s9CLmprmd}yBbes8@?+Q4)TAk;tt)qh4l#c-?0-*%l*HbhC%
zmOQ;p3G=*wp|uLK1b?9SegjZlsy5vb*d@iCoYztkAcOV9wonb^OCI?RB=8gLtLCm)
zYPrHbFoOUKvn37$L^`+l9#uhVS#zv@fq=38?NcY>2Ebt2No8yprxLNMAMiw=_=&(G
zIi2g1Ko*J_PJhxj<^4F`1ZndQms*ma_f+Xue=yISFKuZ0<;LIM(7ae<FbLF(1+-7q
z3^_d<?o+kFC*CVqTsjQCF=fox`Yut%bf@j+l)-jR8GajcW`UeW>5MXj#xpv9Hh6;u
zZ_wZk8oWV+H)!z2xWXsn3ZINCd@`=^$+*HN(-pp>TYp}#(TMooHv<b4fFOr7z1{a_
zpdSq6+&Rh5atSUQerV#P9fWc>Y*56HYKIU7ZW1<ftq{wz0TMir-}x4phYEYZEs~sy
zIvlh`j!@^#)&e)<7Iy{*1t>;*UzWz=7lr@_usFY@UMg5P^-KPd+I?;LlEK9M=vFya
ztTvZMi+}&>05rJc+GsOv<QLjA;*7SugbW3#adj-3WhlreOP6J*3^@_{Eu5rj%8fzH
zyu^Jn4BUT&F$*CueZRbMpA(V&Ouss%=Ggp>y<!>quR-7fnXd9;)2d|tGniWOJz3=H
zhF^4K<Y;F?a4pe~wXPH5ZZbu`QlNgt5t(9zDG?M1L;k}Vi->J|-|qVlrqyScQNc75
z1Tir+HJ2fp7ZaBoJvb<rkWoGam;Z!2kWRz?@7F&bWF}n-Z^F;;>i>IrefaO8G6GtV
z@dH^BDDCji&xg=~iiEMA`9xu)8-D#ee1G_}pLQxIr&vzX7@|f_U<8pHjPXK;t|TYQ
zm)?atX@Ahpoa$(+1jv~>WfJF1>EwiFOSyyaHLnM%pK(q2MfvUMAj~K!drqh`i9Q;s
zZlfMR3PrS$dI03_eLp&cl}MEv%_VSDs7I56Lo}gNeklJR+Fuga@;K!RFMJV<G#V@B
zmGElrwbeo^oN>=&fa-#l44+f@BlE0H3(xxH<bUN-pwZ3<X>Gqs<cB4OPg!E^imPF(
zSP)u)+D<JfSPfq9XEh+SGp>a%_Ae!&AAO4lITaIDH{+TIu{70e`6$FsaUcW2c)yYO
z8qzvZUodrjAr^-mT^&zGSBM^TO2W9_lX_pILye@jctW*ZZ3J3Zl^0m@oxcH<IVrEw
zw10d%;~^y-wyQ|fM;7)tZh_F^iw_1#)K;v&E=IUAa`eMIx)>nLJbdPVK!TL79_>f1
zg_%;)z23XL6xw1xd*|I-<UP*lgmW``Wt9PD83a>|PFUoSI4D~95Sc1u40(?r6;cJy
z)K0h-qpTH1scvv&!zK5t@#y2@RgOSg_<yLFsu{c`-RrX|D`6=`soNvdslj-_1(5cf
zBBbU50nYo=UE>;gP6nuJTyr&|T#F@oyh@xSjl+vGFV#Riv6I?yGaIlo{5GD&64zbq
z8Ds0bxGXgp1}x)x^ifSB;@M^2$mJ=4XIVAnPu21vnRjsr$dS3m5Xup|Sc(f4+<zNp
zGOnv_GZMX4TL*`btcI5Or;`cwu?=k4y&?<}IVSoOZP~)EEPT;pOkiE%3m`m$*0XhQ
z|K3msvrvVdRIi`y<x4WF*mz~@qX^D0yfXe8R;KjPasSALQP3jnA63Tcg|dnhT*3-z
zWVA<@c`$k23emzSbup8rtrremyMMGjlQQJCZz~<gL~+QonTM6MQ@knhP;`_L#+v_Y
z=Y%!AfH&<Gj3Ij+*sLj*t%DnBw(ipeM$jH2_;6<#KyQSVxW+-B9P~@(NS>vy8Xg5m
zgiTd)%(G|kqbbA*ny6QkTcn_lg$j>X!?F+cM8EtNwQOV%2nFNneubb^ZGY@;wE1FT
zagDNe_NmOMjeeE98^}2bAcY^N8JA?_*|na@NLq3|$RWAsz3!-VRrNA|oNk2XtklgC
z{E@Kp$GcIEjH)y&Q9k$AZCqGl<DSjf#<yx#>@aFKmdzz=#y(Yy)$AraE4TBEnlT5=
zv~`bU6Z>sVHkBo_Np#hN1AmkwKP^oY&}vgfT|D$XmF(ZdBiqELyno*KeQMF8DyZzc
z9`lMhJm=XjB!E2MQw`*~rKZ~A61olk`8#OnoEQh@8U)Ssy9#S&5@bAqs-8Pk$2lgN
zFr}$U367xZc3A1(lC*G?nHqykp3QesmE5CztATqNHCPm2d&Y0y6@T64^BkGOF;BO8
zJjBsGk_f_8`T$vor&>X*`m1EX3)Ij4#=Y6q^lmtAx!EGE#bDL=)DL|1Fr)L<ZS2>h
z&Ehe$Y?keni(4G4+w-hdU$j$*KlAR$cJ;sG)Ena$86=h@HF{y|x$|g|Qfdm)0bNAn
zl4C(gwxJNt%(P+qC4Y~+pzM`6pc#{BSxZDc*%V4t*hO7t_)0|%vs8sSG8`Msmq!LG
z8L6IbLRHXX8;xD6U`|bZY+`z&sV)k2x~7R-F(=_o>yw$3kM}R*wUa^_bwe4ijf#h}
zzgrn!@W3J%SVg(s3q;#7^<L$@=5Pn7KB&Cc8W?fe4@6vVDu3OB3Se?;A!7tELF!PV
zoJ$!!wdVS${HHU4W~C5MqgKFg^vSy3N~M~6BgROm>3&Fp(=Dl93b&L5oWRk!*^Eyk
zrozMhN*XSs0A@<O+156(%8KGZI1^q#LMSPD?_)iAI5w&*P<~WtN<6ZR(ih-;Hkj$I
zjiGI6b}4y+vwsXJ&HQwC4f@;9fc{@j99Ygdv*q*xwV2X=AA{$Wn?H3n;>hU6!b0yg
z?an)UmdlD8H;}6Gte%$>o+jZe4!{+x^CgZRB+bszZgnDs#4c;QmD1&{aM)X}T^*Cx
z=$u+(EdjYL91flG*xVo5!Zj^~eYx8Orrg?U?}Wj&*na|63#EKdC<|~*m!rdt7;Q81
ze5Z={ta=qEWkVmK-e#qHtG8a%d(Ln3cJ;P|e!hAyN2v2Emf>O~e1wXdmF}+Ma_{mb
zy<_)N_`OZN#>HdOj>w66)wF_KN|9z)?whQoq}b!fgOjo?+y?2ff?BRu?Kz*uOxL7V
zd520#S${7GZ#nP`R}!qh`+?_ZP${gx*`}6KLb9_qY^)+p<>JN5md{D?1g5QPr`jX?
z_Bu)d7f$d|$D~Rt;U#@yR=(!82QvBx+Y_yxS78pI7hansHh;%*(K}%#mzuDNJ;!BR
z%f32^JE3}<#hrWAMB!}CDASYr=_ejqI;Op+3pSe)dxG@OheyYDX4U68#rnBrxWS}4
z`mTpC<xstcG4`EN>N{jIgF*vdk;wWt2jTw#X8$$RmtkN!6aqCjm$6MM6PFRv0Vsdf
zTFGuJw-Vj^EA#{AC5y!gFa$`t_64$gfp-sC1xZ#}B>%rikt~iqNb;34EZXhv=1@E}
z9Tp`Iza8Y^+tc~|>C2nOgA?8;2Z#4x4~jlW`l}?pwC3QH5>~6j`|pSUzeov|t}|UI
zrPg#g`rp&t9G|uF)IwRl)bZJ&dTD=c;ZoXYlY2oZz4@+q22U-WcQTH5;qflVpWc6X
zk}Ud{H?Q{OwhjeJSLCMYC6XM5<VBKS1O2b_Gtdl1@i&sfkTe)_d$jqTQoQacWnOPu
z8g<o{VYp=B5%?Y5;f<T~_xLU@GPNlneC4}MLl#8ArS)#<hkz8?%<DY8*@%C!V@(<0
zm04aLww0285}qR12x+{1o1R|ZpZ<8Nz=9UWhJ)2k=->}O|Nay|pfbX`Z~zcN;o;9;
z4!=JAZyuF(Hs6!^B{fQGX)V1P!3tyi!J?+&O*Owk1>b{z#(JTG1^!YXNmT#*`P*T7
z^W(P)5ug$XZWda7jn`nktw(<lLEE)uCQB^Vc<FR_$yTXZzxc-O_*>C=`tH&)i{Ync
zc1`WosWyT^(Y++2vrWuoPQ~ExvTkhIg^KGZkwblAcckxGDlpzd1ePwZjn05C-u}1~
z0D==%*@dj159t*UH3{VPLa-#1;*n(a%gyDnDol{V+JDe7ww4J-36Otk7LXi(?L@od
zbAs?(?84QjUO&0jsH4)_>}sl7*NGKn>l<?f#h~L?zS|c1u{b>qsy$hoHr89Z+aMv^
z;_8<@dn^ib6d@ZPWd(J;wt2x6BuYv%Mhp7?-<;A)4u8#e&DL~^k5|$e1wioA8vllL
zW|S3Z|E_dqXk3S%=Cywd_y_FoN@bMgmD97pyQz$jqt&l0@iB3?SbJLPcc(J+P9Qb;
zFQl?r)_+YZn*{q{DnlED)BliDY^E`!xZrRWu1jNz6BNu=69a9+nF6YP$xzflJtQet
z;4h*1k~Nn8H~jx#Z(VY8ixgUdJ6mazPOn<<XC*l_K)aE9lH7lnHwue^XfsH=wvWRY
zPAD-H{Oj(Lt~0HXuzkt)5+_I(@ZIhzxK1F+5kt$Wr;dvd*Z@$JRG(tzO_q0wZ}bZ<
z3|V1swb!K9$V2j%qDt<K^}S@bEqsJ}$v7TXW%dllHs_BD=UII=J;!{f&SwS*Y+h$k
zPm<i#F-$*VF|{!=cj|NM5izT%C3%6(fMxB?n&&D^uMfDMk#jv_G1zv}Ul=dV=x%7E
zbIj9jJTfu&mu!(c8h@1_;XACN{HMayj?XAH7YB;d=YG*r5;nG>SCfLuV<fHT`C4cy
z^A020i6f^G^~94e9I#rCBD-rgmE>ykx;9=Kr(^il*78`(6<)LFv0vs0+oMfZgT9kr
z^*mz?);!+2*F<|*=-*ipD0mkAulf*;>ssoFPyVGb^t3|PVt+_$QSG1e_EO^9CRvVB
zz2#gOwLH@%wnvLG+{*SowbdQ=Zt$#5GIJR{*~mIy=an^z*nZtD`qdF#PO{v_EeTJx
zbKk|obM0Z8bsF}l8mDrLJ_5N36;-6Jy7FCWU$e*5N}d-sxoRcWP{LGw!d#2WoZ!ts
z2l}?BnHpqTvRfnyzLMu=W3m=zD@|}i+Y*_J1<Y39haS*kM)LDj+pIhJQ?pBCz417O
zHZ|$a9bNK+%tgPjq+4kFdO(pIUyqG`^;+}=f$Bl4dP664)7#7%I%-$7Qa2T$^OxX~
zI~#xRQ=W`WW9a#)jSK`Y0QHO-Y$DRW)y#3eZD(R=M`CfFX~J%k&(z)Rwc`a1L0AnB
zt;Gyz<gM9tK(_ZU1O)m?(5|4OY4>Mhj8D;8SEliyDhfCyq%(6B1(8ex9$rO(5CqVy
z{Vs!PkEulE_-v)R+sNp8U4I`FXoKnlX9It^(m%<r-nI)W#*+rk++$5PEAN+w=qbQ0
z^2FUAb(b<ZeI9jqlN>+DHH{HgM1sO%4J>yhI(k94R?hmC_JCx4WOZ>ZldrEtjyu3n
zmJAo&H1JOGOF13)%2S7%%_t#hWX3c}HoVyZxVnU+cxTjHi7EB$s>+X)$^dCo-*$g8
zQ>ko|4K2zdU3qHY1k4{Nw!%Dqv)HkoEB9hIQF!6Kp4WqS23D_yRAzM)yC@*81f3dV
zA76d}K%r6HRessu#ezhEgxN!R&SHnjOxL(8+P#AbO&qwNvD+c1=P(!1CI;oEKBg<F
z`jIs~j4^3L5h!YO!r!c+wjOD~tTTUFV2^!Slr*jx1zIJvRkKhJCs)c_;XFQ)T#dk0
za`iFnZmtDEK;cxF7W7#pOesf6c5#$tjBr6coP;TgrU>db681Q|ooc+VoR2ym9W(65
z$Wb7DkxHjEKAao{fnkB<gUOLXLp!3nQyC-L2bd9Mr=aIi_o$H#LQ(pfdhCB|ZQOI0
zm1rbNRKt$huqlLANvhS0`+v-%$amw3^fc?rO25YjKx(11n`dcZZJOyiQ(X9Ij1GvO
z!VKdcVtIxNAg>TZZgC?q%&iv%-ke?f0&jlPr+O|8RWs>|RV>5R1izcHw2Y^OY64=#
z$2RK~;Dl0o9>A_0%*axrgdTr^c2uKoB03rGbf~05S01hC`~E;_1LOX5IaFP5$-tK2
zOF4DWKrP$B==5x2&I)jK<A`OlxL*Lf%O#nVq*fKsQu)W}B+b?F3UAC$r)f6=ZvI^n
zxSNfR>+u+2hj7%U-z#M~cYk=mTob~)BLcUm>JICvMo@lJwOp;cjU|8YIRhF``ztr8
zOoswD4>p`_=U)cb11s6K46%T&a&db1xUKv&k#;_S!)e=MpHiyN<+QLrR-Y^D*MD8s
zXs^HN7{A{sv>mcvm3E_Uzc?LU$};~Dp`4_yoxg6?{Hw*c%DEbLQ&)amqDF>N?nGMu
z0=sTzcM4^0WOH<KWnq`wM*|51F)=omaZp1O0x~m~QPc(}m$;TY4FX}Ym&cYnJby6a
z^6|$7=mDp{!0Dz|7X!dF!sX-7%WqfCg{JEZU6OM|x(t1{^r>#I2$tReoi4b&8VT3>
zu9)&WsA4POa_jU7NrI&}jWyi&H_=5>0B^s4{QSa`>P1h1Olfz~Ml#YS>xtBhwEprk
z4c(EXg}igF=!p{4-+Szj31zLga5pCDvZ(O#7fBfJ{`!ruvapDi7EBrFb&y_%FG&yo
z{VQE0S*dt0XkwFACO(riT16;Gliu;|#h193J0pKCA`FX9wxG!7P?~8Wv!o1q&|M`s
z?6N_V<7qSuO9$jW!GZxN!UTP17=Md|naY@X0WMiQ`2j$0kBI5hCRwsa=rWmP%@oKj
za~$Twd%NrGro!D6UYlNZ(qu}~ps}0N9z6f?t^4)$<K@3&jBsF5>&w6XeDMz?D5j0O
z2*H0G$je{<zWnj>OW$j<PX8G)M{ohIwNi5z&`c>y(hNBr{?KS3FQHfHQFxY{$$P>}
z@=N^_{Cjri*Xo57{!B{M4+7q6;^_QVXw4kuiLM0>x{!_UacBVoy2tOgSF+8ux_G$!
zFs3C-KZy2hZe@^3TSF*?P5_nsjf`-OfwE@qYyK@*F<$LMz162VKw}`5_DMpvm<%mS
zdvz1#t<DV^O=bHg-Bm8%&A~p!cHz59-)+%h$|DLgi=HW5m-87qV8YoR%M}k>vxo%C
zlpHJ)uooGpPOGH#3>%rS$v9={W^)YE370>aI~;!vObJ&sn2L?J^as==RrIYtmpC@Z
z2Tn87>j(Hn)R^@>c}in3P#W)P&if$tR#S4>Khl$sr$pAg^blg>_8Kg1^4-v9_V3zv
z7%yp;r?;tzFH_zkzY$w8RW$c<lWsEgqEk%g7>+&8Mj2YS{^;}ZlzhV4V%@%V_UGXD
zZ&`nrps75LkY#_9j)+;!fGZv&7N!}oGQWt32_y$h;k%232}4ThUa1pFm_oro!zW1C
zqiR5bVAd1@>6yJZpa79vJ5DZXD#=jmlBVd*CKZ3y+?8)z#)TNPd5<#vl#7^UsmvTP
zUR{bBrg6fKH17NMZY>bIq#ehd$D1pRG*MDzRCFb7IcW#<Wz|I$WJ1_YIb2E~vdDs_
zxAzjFDWnR%4<_|>%}NR|voRkRno<(3O7{5TB8IU-+@iu_7{d*&^I>3y%zuAxvX%Ip
z>vN)CqL-?gJ0CuT!Z}w1KItee1wOY&7EN=*RB)fN&5~haDqXhCY7s@L)N<F2x2d9#
z+#pUnQ;K@~eAy(0Vg`n?NlJ2dahC_1J0X9Gq?6{KuDFoOP(hPbH-5Me6_=j?sX6B%
z^~7FhcS8Dcd<kVlMN^0DrHEWyu1E1aSFG4jlE)YolAKebEq&L~{xnix4ATI;l#SXH
ztY#qk2AD+$+!T)p2GO%S8*mn<MPSa=TE8xd%nkQU8H<hz>bepXI7dmUY5oJ<P~Lw$
zJEu2VayQj*Ddvq_pi9E>jp1%xP24tVZih7_o4X<w4@-U99kX|OYifV(({O=zY!;4`
z>4bp60`5Vv>zJcrv1<jq$_2hHmw2n8TU`&hy|N~e<T};2%F7*T5+3+dD!PjA#A*FR
zR*sh1h$C4!QrX(AeTN=-?dgc3imZR4g&m~1>f<54W{WRN#nryCQAw_dA9m9Ad39^?
z67SGh-z-xw*&?T!{E(`b8oz6iXLCIu3;y};1DtkJJftb6Sy(@U;`!ysOg25OO{MKp
z1uXI=N~)woS0>V|yya4o%GzE8C6!?yXK^?5tFh__t#~6yA{VtDo#FaDpca3Q9JrU?
zT$3&T3U^e)x}KS4g>lSiJ7YvD)j0jcYQ$!};@^ONT{zsuYh4b=Ymc)@*fc9aYoQ>J
zZrWbqC-cy%n>xo3v6u3RSq!Pd1!V?3gRvfzH<BsTeajKf9bnPyMD$UdjFtp{Q+)?o
z5_t^G1+?V0Z0T2Q27f3lPFH`Va-7vA^%`sCtsI3N$_b14lKT)uniK?_EwMwp^3~23
z&FYcew-9awLq15^fe~0BY_M<1@7waIY~XfTo_esLZkXU^z(dutGOljH41saIOAa}-
zJXy%3&5G5O_I%^q@ru@6yUnSoN}Fvg9mSWB(v+5}$3V0jImsi@>|TG4qF0ICsufEf
zqPL+ES;L_B+kA#<Sp~LL6YLs?6f}`4cj7pFO^U?K`*AF=qOyT_->18F<lT6t?gqm~
z)I#J1oq4Tg3ma~x(hw6lHufD6DH%*DdRXV2W`E<58iE*-Grg2yT%C<%js9I@@p+3X
z!A{DFHJ!kCT;TcRE`oom=4P+vkfmGO?0Wn#UOykmS-c)BGGTwzQ5LX{q}DK;0sH?>
z?17>Hx%}r4`%dALh<zvJ31Z*txu4i~Ex(J{pNm!|_Rpi0iTx5<`-pwl@^2ybeq|q}
zTO{^I^~B!j&k}p2Y2Nfwb`g8Zg9iGR4~e~9l=X<zJEp}TYaoC15s#MC2Rsrv%;*d`
zw7j=Q>}@!sGQ|bkmfTCeNbalD6Ue>dlmiC8_+E1Ftts7#H4o9-8s^O4Y2x0TjL}aL
zdc};BC+hAqi4r*IegvVn{xI%|s#Ab-y%=;S^Lg-9kdr6#`A*77e7=)%BA*99=IoJs
zG)|K6;owx-qS$|NaB8;ccbNGMK?HU|xsZhW{gw1vv<D9EX=2T_$Jc#l;iRVYDLTPE
z%cpHRQ%@8wZ_?(BdH@Q7av~YLQ>OT+*_XOGQu^!r=#F!TSV#vsO1jehPIWv;8>hKQ
zOX~lyv^!NBHpoB?M~;@c{Y|Os?j0y&sPZ~=1FCI911~72_NIEqEqCSrk0>8Rk#F50
z;O+fk;w40~V6Ok(`n0PVsV4m&OX$y+F*+_212Hi+myvEj6PNLyJG_7B^IxyOJ;_YE
z6yAh?!>j-F<@M>`r^<*zLB<aTNr1!Q?;lTLKm!S5{h+mQsD_{a4Bwu9?{_U{b?+!P
z2P`&e4ADev$_OGi7=r?>En*{PikJyvMu{2K@$>nUkwHMH_!FiXn%U(^>HWg*Ag1s?
zBq7Efc)os4|A+Codw74oBw@#s@@YXqbWQtfmqaKu6I=iM{62Kw{PDF)PgxL3&iH{F
zKQ9x7j4QmbVEYx3E=aWWGzMOEmQ0iZzm42YE20;C`UA8|!3H6P40|X_yH390T|eKv
zT+%R-5>l%fe}VQ((rna8tIwccm{724Kbo!8(qQ-$4X*62ClY_#EgAMlK_2^E;bm2e
zV7vgS`V|jZR}>=2B+f|J<eLrIR`1auw5iOyWIEbzAwU)E3Z4}hkR+6+iQ4E4_7vYr
zowB+Z&lJ)rEAf&>p^QF9h4Ks&==0sEtZVN$b~+{O5q1oxBwiz**+gcXC?bZeo)JDJ
z;z+;>fP@b^`;&jXJVFY8aVv<VJ3Y&5CTUr?$S475dYa<?ri#g=ef2FrE~>PQW{e9s
z7V#kBKp3@!p6o&FpU$aZ<NP=zOg1+A_K&ZtBGhbOVC>8objG0uts}+xv+6l$#`TxZ
zzn3LN;Dn)>q^kO_MK*D_-gw!E1@GI6HwaKuD{O6w<w1YijIQO|Bviml87(T9t}%@>
zzhBr&dY)rD;_um`)DOusI0Gr@C3DFpW2$sa@y!fYx6wV46jH{*i%tpFiaJJ_76Xo+
zzanL7Psaj`lQoZfx>h8tw@Yddq^SHZY^ZF#*F1Jl!k6<KT)mp21A(r0^(}4ymB3&J
z)0LC7$Zdb0O`MEq>UV$w47>$8wn!x7ObqE9!J@q`j%?U$<&uNECN)XpR8CZA>M8f)
zPtK=_13ai2P|=<tnMKiD#zFne29v^45ReL=Tp___P0I!Z>fB^pb^db~oaTFv@w==E
zmgJ8E0L=>EysTsW9@^p}r<upRw!zAK3P3hLkfwj9Vm0o-2gDsTj?HSXAPxvST9fc{
zW_=+sh}lcYc%u55{bU>k$2!amX+jChhlz%y5x}Wm?;oxKIZCofWx_tMLk{>B?<qZ1
zrw4KcZ2BIsKVRLlWM6VYWJc$lSwW$7)e;hnKW+?)YhDjMO)^&JwnV6yh@3ygt#5T?
zF&BS6lRni#VpS4F6M&52ZPr=>P}sPwgjK_MQGYPjYWgl^x^<o(g>f78y6ruUh3S$s
zZ^j_yIC1Owc)pn;q5$Yo9wV$a5o%}8LI-D%vz`Tov;Yq8S!k!&xF76UOr`!_k>}op
zX={*x|8M95l#WHm&B0v&<xqGF{V4}`0d#-h#gDP-fk3dQD`2fK0R1UjT>-}+t@;JZ
zOpBF5yy1Kc{luJhg!4U&cs|MM{=_Xtyc3R#=xzm?!&(?xd(Kfmq=lga^njUm)(ZaQ
zwgQ-ihDM%IGR$n3wXCtz>@x+!no;T+bXW5{uw08#Pv1_(uL7@CLc)yuu_()rN+^Go
zd8Ib6Z&H#=;hX|2CkR{IBNrvkx;6SXe0DWy1J8IglV=Vly;Y#+z~sgh!=J|%)TsOW
zsoi0z52FiXMMu^<^Em}dSWe=*U4OX|98wG@#fqbu;ttCVYk9w?Pzdm4W@`Ju5*yi0
zkPfV{kr<Yh87u6uB+eF)v)6ic5;cFG`Z+n{^?k<MR0pd7n|ejr=2kjwt!34ev9@mO
zc$-w+=Bo-Y$Kqbq@fOHkz9Lf16^knr&ZK}NFS_+m=>4Mj6J2Kk!nM@L7DBef(lK8S
zE+Y@6;9jqVC+nL0I?R5s{s@;u(&Lc@2)f2QUG%!r`m2V~v>8?2s@&w|`<#DO$TNoh
z!*+y!RK<akqZrOA4u&cZ`tXVar4s;?*Z?C!zd?xmvc{AHPKh|WP6*fM+H{BPEEC?A
zCWb=D>jy74TD)r;f>Vzs>QT_1x^4h1^$JvXZ`Sk~#!h(czDz#3rdKapRGOgTh-#DQ
z-^TlOYpOeK;98_tW?U;PbANy21B0U?%}Forh>ktpDMVqMKNRn(jDuB;4VM<`-->nN
z$0i-k>ri(t@$0Q<2i^#+{4YW~B}4K$-q=pWxLbz%piT>;W)vKab!>9zRe#Da0XyI*
zNg5soJ55Vqm#*(xuuB(z8$S{lcUt?A0BXGic4_AVcDnt!Ky^MnlT?3MS3obVdj#SV
zZ}nF;b*Ht`sm@Q5snlXqF<xi;Unl&u5L!>81!+!TedNiu3GHIl%W%NpYT8@nt>k?U
z;U5a$fJPX?1B)8~&IXlVFK$v-hYJs=b~~o|Fc#WJRQQJHNIRqHfHdx28vvc~I6Jx6
zu|AGRMMfw&m-{=h``$8LrSihaW&Wx@m#y6HD!{gj*@ufN&ZN<s|FfpL`-{SdE_Eh+
zJX&IPi2x@W{{h^u#+SiuI~D>oFqctmK_vk(mnup-F@N1!ZEqyE4gT(5F+jhh1*{Q8
zN)(IWfV}k501kS{r9j(*Klpsx#OcL*u=j%G-}j+rSL=&+i#>a6w?!^v?krcND2jYY
zi6|!<;y{!+4vC~uG_Yu+(jb)@#Y0(ttAk2)ifqt2@ztTSk0SV{Dz|qs7?mSX89Xs3
ze_8z59)DTz<k*H3{2-?$NARNL>1`fJYm+2pAX_u&kYdh_tV2rLlouP4L~hbBkIcp2
zW*Sl{Z3Ku<H8$SNLo#lIK@gKnym{cHZ8q63f1AdIDROG+lnAmkhJv?+Vv|-t<~Cu)
z2wJU?7@=At3*MSGX{8K8vGHOYVQQK|Kat#|6@MF~lr}d^OVk<zew-tbOM`xyWKxGg
z)@d*qa!N2~WCOJxpbl743=j>9WzJ0n1uL5mP$GuSZP7{uLy@L}rXkmonlMN(YlGD|
zlthgI?oe2o+?s}xbJG!uz|c)GJWHi!)0zj3JpHT;x?Xs=jHdOt6~2_JOM~ImZ445o
z*MHKK%tmmr)CRLO)F^EP<Drta5x`d4G(qN}$}|frL-n~jsc^_RjR4=Y)aHh@vB`@H
z7R0EzS%Ej28`4M%C>YW>MLZw?6yed#C~z6um=WI#w~?G-vPVfe7z}dc4B)La*%@z5
zrl0xBM4iOKQ6gN@X=lVNOQXs6O|wZMUw_`R=D{=8q>UC|;WdwI5lvibTLAWvK{uzR
zX-t=>&7%wXw>7u-$w#)y4E}A3)A%VnUD`OVRU&Aj1xpXI?b8Aafi^w*^waUXU*9dp
z&n_>Qt6}_ndG%&-^(rE~#MAN5<JaRC_<%W`j^8fMSFb3g;7*n*2W&Y+fM5w*+J9n*
zMyE&P%lEHWKHka2<-f+y`snAYi{<6`a{TMJU;F=`{IFWRyZ+-iet)t0;r;7yzWizY
z@%>Nlj^8YP9=|zT{qUD}tFJFti>rU0oiD<VZ^zefm#@d4DWn)^yuLag&sM9ei`Va0
zi)(!SyAzFae4;7+{@Jr5{QF>5AAjhydx|uDXj4S=3B^NXu@I?+l@JkA5Z3~t#_9Ok
zfZ<<`UoF2|j$aO+)Jlk+ToJ@rV#KO2w791dh;YHC9S{<HR50;szLpXo>6Q|{-Uuze
z!wSb3v4zV0wMdbJM)znA2|nb@tnNmXppD(qLW+nmc7=xxvZ!eeQ)CY8n}2St1Y&TA
zEeGvZ0xnwE5Ta`}!NM`$SdcGYd6vr<{@d|yfB%Q?&rlKL6*j(n|Mu<aJ~jng9>j7W
zEW-h;NOA#j@*vg>k3(?TvJT@H%eTv`m+#>I1N#q#@daFdadkPQ>EoY%S$*|#b+%dz
zG`)Dfyj(%ZbN1+hfbLSBJAV*J!qbNq<{EUs2hWumLh<3*v+>2%^8Dpu^=f?a<@52o
z#V@PV8>{x>?E3{2!hXD5UAJW0eum|_UcSFNUrg^xd-u)a&BfX0%U{}hdIVHUrx@W1
zzx61rP20&4uGi?Isc`ga!WTZ530`a{=~V7i>6CV=cFHzXj-8@Tb$=PT%g9|uZZb~q
z8W*8YNJxO2zz7+^E%0Q4kr2X;@svD6Xygd@N4^ovuENETUISwRrC|0{Z-++B!~%3d
zjO>#=3(gP^1gQsDu`8jBI6!-qpaQfK3nPq10E`TvDfibxl7*bMrpOXQf~)S+B595>
zr3w=OKx60UG(&vG)PEBolO&h~;F5C=4r7El2<^xctcVO|Sc({pft`eQXmPl^vm!TU
z?(U8(7_gb1&fQDR$0x;X4jh00?r<}iN9Qyktp+bGkTDYHM{+YJL;jbpghUEnQS8&=
zLBI&GbMRs<RRAIURSAHQLt><_d^-UvChz43jE4+k#98Lkfq&eyb*lqBANl^a9UL`_
zvcVF-)c_=MHXQ)Dfr79m0Emoem|_JMmK{I@m>adBMF4P(X7OHo&GDzmHnm`W!Q=^3
z3u-0owK@mpCU<YtaEjRJJ52+7Hz0aA<2%OE!!27JSe9;BEUS_}wOv)Vc5T2hYX^=1
zQM&-^4UP@81b<ROF>??rH~*4DSV{4h>A?Z(4UGB$di`j?s^vCdRk<5rRl5OLONXST
zL(<YAY3Y!(bV#Zll1?L^&arU{7A9rwJEx#*1mt?~or5_b1=KCvg%2@O!hg};xvy!Z
zmcVIiS}`O6et6SLjxecq(@NF=(D{F1+Jr!#P`RNtz<-=3+-e?zvcSf-Q4KaBS&9r5
zi15d1xtDDlYVfeLX<q@%E95?&Z7(k0AZB6gb^8jYt#eq`n<Z5fFp1AJ!)pu?8=r}S
zRJ~={;$2<Ee??I^z<+PFjau~A!+IrpyQ|@?<pbCd5JTC~QrW>Of?6x;A*_tNCuE1P
zdXvjz@qhnH_r=SA-zGEfXx}B7`6<FSnQ8BgXzz^Fo~~-|pw!M+w|u>WQhQ3QJtfwj
z5^GP1wWmbuGOWw6F2lME>oTm%ur9;84C^w?W=OHcOD-7-aveeGR4J&qf0OrxZ*3`4
z6l~Q`e^}IOp{(~qrwERjwuzvW6e86pY$<uPqJLJ++&jniwi&QzDXqlxjv$IQyssY>
zuI_|iyqA2Vf00qrK572~pyT)i5or$F77a0#AyaJ07GBxrnh7c4wrQ^yZEk-E6ypp%
z15~OF8E&K_kOX7hC!GPSNUap1<h62Sy?0}(os8dTOesBPhCg3kUazjsE)cXQ_~Zvv
zG=Ims!e<MkN=9n;vqj4xdPk_#o#99-$4EsoY6Re(XjW%vOEIvqw_{2Uger#-8x;ki
zw`D5n4ge=P3G$C28WL7H&Tu-#5Eeawuayw*Pvx>#39Y3jLsSNc$uNkN87v_j1J|u3
z1aOAT_A62Rs9s{tULzQj&Y*+B>ERI@9e=zRtugP_ff90=r4p?OfD(I)Ziem{b0ELQ
z5!x|-RX-7iPQYxm<CHFvN)<sU&jMBSzV(uKBbm2p61Xc=u1~n*lVHoIJjb5*GiWN<
z>}_QQ;Y<)!vtvlq+x$rT4VADtq+Kh4WeaM1b)c9OYs?Xe9E;6J*gV&{OW7*nO@CPL
z74FgEq4pbsY1d(M_OsU#54hh5K~MeC$$GyLdk3aG{C;c4@4C&fsL%uJQ)bvy96RYe
z*vY*t4dzf=5KA<Wcm5;!eTa{{a&Y%93HBQS_^*{<l=SUrd7lymOA)c(YZX5LGFAJa
zmiOCZ|G9&FlcSgMS&-_}A*^BmU4OM6!pg<FTjdbe1gs(DAO(zJsmJU}e{}7hcGT|a
zj<tKzZSC$i)~1@*FRX2-+Nn*g?RV8S6m=?fiaRBp%AM*msLP-(gJ!6;1(B36(E?w0
zwNUpW#i*MDzHg7U@nQ{CXvh|WSZbKYZy=S3_#V@3Ed)uHkqLxT<rIM}?|*l!juJR<
zOYiHlw75d$>B>@*e0*Ygj?kJIp|x^Lng|#JqntrTM9B}?eA~F3y=oy8G4R-evNF$b
zlFO)#+ZEik5+Z)DYrhf?<7e5g(|EUIt;GgM?A1bI%m|05TmXY2pa#tWBTz<SKWA^f
zi08W8ZXRX;3DO<K1LTra1AiQCK{#`s<BEYI*Cix5Bs;&F?fq(Oq(MCC?dV%Sh$;L!
zlb_jbl#t}zpt@g)!%XopU5gTu+`LYcg31%NqcrL9iMC$M(LG8tpHAF{vy?5MB6-AV
zEv~MXSJRA+Pl)a}9iaD+Z$xJj05o};=`iLav)sV&gCegZh=aCbT3WoAPRM+`tD+I+
z>IZR*yf3xXJjZq8oH56`)&aRRkbl~z1As7cf?w}JiHS0^YWfiGCOYKo>O;2mp_Esd
zatNm%U9#~ZtSKYB93&oS?>A>+PlaUv4cYXHmqC*#69X|ZHkV;}Koghxr#rp@Ik&l}
zJM0Amv{lrX@l`Mr12Hi+m$AJ%6qkj%JGy_%MVc>vzW(;43+n@)XcXJfSw~8^8Tgbv
z86p*yUnI1dU%)-5J4UV;(|4IcamF>ym%Gc~Nw}a$!%hDC6ldG#GJ5G}47=DuRpzqI
zvzYU&o<<t8109)7p<kc=i}Ot6I7$2Q=3&h`7_16t7E*_@YTc$ri!B-IRl?T5ALxJG
zeGQNz$2A3G_xD9quvnCoSBycvH$+mL&q_!aW+f(-`)HtEdHRxPr_rn~t&qLPPAihj
z8#difahK)jx38Nb3RjAj4z<Tu;sRS8VCKeKDx)A=`-RT*b#OK={0#aSTo%fO!)}XU
zv8c3Yq_+eY3#0%^+`4l4e3xL(`jvkzjTbE%w%*F!28~(AtN-lidEkXenM%)CWaA>0
zUl09~qAUspKRTnHgs%9A!D6wPswQDb{SKkXTzm;2(@4RY!v7_?XvootjZtm}6Xa|}
zw|u3sPnl%)=iSf&qb&ML#!olfTg$i?;1oDR=9ri(*r>C3`&#|_^7{1e!Y_Z=pu`7s
zfBkrhe-N;;E_BW*r$hJi@9z84?={t8l=_}ZOR~JSmU(V@tC)Sq!KmmYQ_B}+M(d|&
z67RMeo@<Ld+8m>&^OqrwN*>VN_>sJPHu7!sMRY7#18bQ;c7V^T-iuylI;dA9-grfp
z@;jEy6dN#(B{_=@^li|XJuH8Pv824_15y}gloW-dQkW8J@w;$5%~9w;p%(A(r~;W0
zc-(K3Ii(hxRC~YECL7sg6=#$sRv#9ZB2v?IKwOGR#kzV$Yp7sJpy|qJYF2DWl8Rsj
z7J~*)L$PTL*i=xbs4VVwIqn2eD`aaf;PbW<pcb^U;JY=^;DYBqsmy<abFlG3Ulx??
zHr3aFH4_S*z@2tIfDH;HKr8AydH_lL7NA~H0&&4LYsvIu8P`R6Ti_Ah&v{HSZ0jBN
zIQNi%LAAn6tu|u%%#V&6bNtYSpV>$PGC}W_GZQuJ2fOtNVGj7#tLJ7H))!_SVL+(6
z8+2jP-PqFI;H0}LGa7%pn>@SjZY0j9qrM+$gu6Y}SrP<u!L97K-+@TZ($U}M8rg}G
zO=!g=oh_|WTEggFlRVP^jSWG{lGm|G8MVf^(Cl#?W@@EhT(PPz_P^|EkI^X%epTk{
zuC~}%<rwRgCpHG;S@|*l-q9G4WfctlqQ)SSOYZTxIKdPNggk%tc4^Hd-)IW77qA7D
zd}lQUCRp(qTQ>#F$_03MQ$R*Q8^5n7c#I_`{lGogGo#qKID6331?9k13m%6&vtM1$
zL7ozChC;ir6d@L1wcx?HGoo7xa1xHHs5juKYQ+YQB5C-g4L=b_RXa4`sA|Q@IEpli
zCd?@~O1d4T11*0}iSxnN4vs1_T2G17Dfi-(cs@EDxbd6mYFPUy426oMvR2x-3qy%d
zN$0pmX$3<SE%H>RJq$H!EMllN{rM*Sor+T=^QG)}KTa_y6he3;PQm{0_BMcueo0%J
z-8DN}D{68T<7WX7WGRgneh~l>c&q|6B9OV7kLtLfi<N%^1G9V-bPxKdWNA40s#dJ=
zb_YO+r`WN1Xy?a7Ysk!M#e*RNdZz-9M=kF^Zi0A%O0K573lDNxu2wX9#X=%yRS0Ju
z!&Eyoz=vwZ$?zfOrkKdjfDdSd75o1}4|kjbAY19IO(o8T_ixq!$rw6D=$y!r%D<63
zPk!KtfXaWvd(IUg2nGAta|J&+11j};^9gu~AzymK{_(vlb4t%xO8>4614+{vDuK?b
z<MEF8s}2&tq8ih{8wUw9?LSKOxx8=9pC|MC%Bp6Cg1+aX?p0Orl0|K`;E42%g%ubS
zA=EEV5Fal)lB<jfHO0NMQ%Z2PqRkHL(vVIM63>4ts?`n+MYURSa!~~V`mE3C$t4GA
z<hrO5y!B|K%xEmCgnFf@W*XC?TIL<iuN#F{FrDyQiC9I$_9-tZHr3k1h#eT8RM<J=
zd#_LBnMt(kLNTKXre_VMO2_AyL|piYb*t2a56-Np_hGMOIq_3tH`KWPMpn~XRtykd
zOsh2L7C-AD2~S@hgn{;3FvFvS)A)p?ulcwBrL|q#1378nxFT}uTUg5A5lmFyJe^aJ
zCcv_-o71*2ZQHhO+qV5Trmbn)wr$(CZM*x<-sheZk&#gm^;Qp6Sy{EdRkZ&Dp_d$)
zGB<(qAN>d0f00%wDIncc6e(?e+=&W%T;EUcAA)AfH`DNRcea)2le&8uDJki11jU7E
zfZ)%{dB7nW<URLsU@Fel!pnLg<L*mDPIwM$54k67{+V(*Iwtw#VSl4@`~IphQC>nz
zc0YFwFo5x8VZVo_$E<eWqqKlVe(AaGSRtuS3;#a2Dyv|rJi6ewZ8upuw1eRzlXhVr
z^2Y*=;^(SEcOEuzpEjYl!(pj{e;^Q-_u*VPy_kopQ7_F9&}#LsSU!A~B>`x)L?;{h
zY4+*cEsyZA^+<a<jZUe`CZJ3+&q?$=kmlqD4M2$NU}EN<jQ;k_jQAf>jTSs?*}Wj#
zSyA618<u)SSVw-RwZ+~g3NtKH@dcITSfRjm$Km75IYU8_J;5DZ7^~R+KFA63c0lZY
z!j@V44O&hhMX)r~DSJ?#Txu{8u26xB{PTx|E0a@RB0A7o#iU8cMCLR@sc{*5x}r%H
zB;e*Vib&YZlMd)J#6T3Bqp<n3Y??qW+)MK~*k!0P_^>;Jex^)%O05kjkhMh`7`uDG
zSKa@1gW#{(U+Yy+Z%EXIfsn_X1yf2+@J4nPkd!9<60)juWl)FgVU9>3;-nm)^3P75
zCRF2!kR(5CkA=~4u?vibdT|WmAEHa87{C>m@`X^khA5sA3@rS-B8O~ZxqV0;OMz$M
zniru0bbdHzJ!&DGD7DS7Lf=_2=7+`?D4QUI&<tns2{lGZ2Xl(Qv|UwZEpaWvB<d+`
zgYAqg*}CQ#Y5~I7mZtjI>f`6VQ=ar&JBy&PxQdvb(_P0VswD%6NihFAWzf6LUjPBQ
zkUD>Li;?faMvW78(~8JkvKKHc9Hyu{*iLAhRGK6t`{#fR^K75;Mu6Vp3#%#hjK;#+
z?{Pg<VYPpz%`*AMAaJ`zWD_yh+3E}&2fq!lP`t@AmA_VgV2aoEgRl(p1JX*VTM01H
zEX-7}7j#_c#h9?gbx`ZYEi9qMC;%!r=2D7F(T!x3*N2G$Y_bU%{nbt5>rn2;T{cGB
zowS~jSg{FqPC-~W!0G$xM2XdEf8URem~{5_NeNnq{%E1Co01nCjL}{9Q{w1M+OO$T
zZ22712IX{)>tHYqTr)lJM9y_MuZAFPYdvStacdTkcYeKZl&Ih~>17*vLIY$<KJP7S
zg!L~b6M-4CA5aIk(u;PgW3t~Lzen?H<A7}0kw1HMwCq-yKEtM<!53Trz76~9n%G`;
zr_oyLLiP$aJ$oALHbw$#!O`P*2s_XwAnx4huFH4PW-{On#!+M)w4md<7V!%-!p+$M
znBL!@3ann#jUj1ypQ*wyZUBoJ!`dyym|WKc2gZG$p49EaEvnD5tIlKJ#BK+c1ml-=
zNvqY++w*0JvI6E<h_bedoto=!4yw3Kz^@8aXeBeMdvV)NLU!)q??PYp!|joav5;{m
zuC<Eqe?s#Xe8}+;vzcRBZ+wb2NtXpW$D>{dU?(#h_)8B|jC&hARshzb@dDjq%Mpix
zl%XXCY0c;CrLVw*JTPinyfqWg(vl2D&#{|0Ln)T>qzk87nk|%!GpnijRoP)zmRQ8d
z-Cxoh(rH&m<~1IuMrQUmPf=-ZRAk5EWSyq?-Ih(5{c=-Bvq^+<EmGJ%oW0p=YHfV4
zcCh3R!y0#>=I?3GcQF1wI?t3lOY;9;Joj^=rreEkQvs&z4me=DUnoC8mckbK6~H8z
zWdpjda$AP7(Yc+|2c|#plj9}FlCwA@7tZ?XXeaOiqH!*dx5l#r7eU7l4;uq11(Beb
zM?R2#9v^%_ymOB*T&1U5yR1|NXLB=qHRM%u*0wokwOZW5;<*<R<}KbVZ1b?WScq<b
zp0B=LJ%C{&TV((@Mk1VAFI?9-<RllDHP_#2(AeK;f9+=_HAT9{n`IRV!36Qr@<C5)
zJH)xpXrhf3+FNKe$#e=QX7rB13Y;iC8l)IB(2&&>!(3BD*nNC=KYWD?Q+|W#h`Q*d
zsj%9ESo7U)NY->m>kZBJ_*CUu_Y#8AuR3M083R;|XP~W(P+TDROf5EfiFy}C&F1t*
ziC+i(*sX+*!^n#r8>4=UheVW|7lU9NR`uH<tKjtUXsgZT<-`)vF*#Xv&&a=2%4z;K
z+XsFu=9`x>uK%0wlRRLcE$@L?er21X0i$O^_b8JWLnQfBuYeJ0aRLHSE3<N4!RM1u
zqyTV_0vxWb&4&J7l>zY=5`ZP|gmbwI3f>yig=+R%z)FT4+=qUn%2ajB%1Ey(r@j<#
zl9?moR`C>*gIN1;3Zqy?*G-pdU3j-Z$-astQWIU01Xg1oX@n6S2|dP7Q+r0TRw^1`
zc0ipB#4xMduABD?j+Vb-F-jqi-Vmz5qzB9(r`<1{XD}XcsG$)@%xhks!x21uiM%;i
zn@-97o%S;Ib??`P;}l>1gMkObTY4IO!0Svwjc&IqP>NTaekY$FshW|2iV8-|#((wq
zkX0WhPz_bv(tEmO<@);QYOe<z3N{zEDi`3#cL=dOYdZ+sdnD`o`gC*qwKs#Oh715M
zI{STekD&YY&geqC!y!rG9sRuBrg{5I7HJDU<eyuo>y5KbK{!47Sn6qt@gG$dVH_CV
z^VvzO!6HZI@GyFxt{R*eGFYU+I*SG*WLXN7_CQpx@$}D#`JBlhk8r~cS)#g?-o8?=
z;fBr=lOrb$Z1IX5Lh*%&KMB9XZ{z?HGmOXB!ggA7+`(mU=K<_4#22lXK^U6zR!7kn
zNIM-T(doxaWXW^H@OZ};@+S3SX}joSQrZIrHLI0mZ<Om9(K0?oUT3&iPcqE1C~8BC
z^4F=|5j>`o+i7@p&Y@fzK~h{asa-rcogwUtYMKxj_aR*60%IS=;bDyX@{xdQJ9n4H
z(m>r?^O31Xq7ku(w!~iSpa`R%4&g;Kx7ZN2_(f$t(-#R>G}2ApoCI+4iH^)3w(Oh2
zi*_KSQF5!_YF*d;0{s-J@lMZp6A;n&h}^x(S|-*r<bc`zyz|D?bEfIi)kom=F=P0Z
z6ML6kwdv$3zStZhufq5g8)tx>Bc7(3l(PlSQ4Z0CjR|R(>#1h+@`T-*Nt~}>hpQ<`
zNxW`upY#@^&_sRYr5=+-v2z!G-V>D~)zaLhTY$AoV9Y(M?TQP?=;ytij}QW;pd~2G
zALA)fvb?gy3yN=trq7V&xB|p>(w{3}l2E$e81EEb4wu1q?F7i97qtLW_s}Mv4x=o4
z^;}nIqezK5D<>tne=owpEu`5~6?k5qMn&c^^y0~EHtfnERbb=B-i^p<ZbfcW8}gd^
zy6k_bH^$T?(i{r80$&S=vCT<Buf=-0vJ>GcBXad`oi)LJXSkLyHN$aZT_tK-37sD+
zYV~7Ncbsgk!fIG)o9O}2AG|lT=WI`qn4I!c2CJZA6q@#f8=6}Z2{g)exx%0P4rs+c
z2x*>Fc!l=s&Zq+teBlQr)M6J7OT}~(uX4N*EDAeP+?GH^5!@;(<y_g#nilIj2$`*F
zL|%M_@zO{wJEd%dvEK%(Ar#w%aYLop4G&!d{1gFss$RYT!$CkTt7W=`u1gutPjm+*
zyLLSNyL3`Y;F#Xn*?RxHKojEP`?XK(@iV-5z6A<@Q(v4iY{IS_Tnzlx@%i4@8Yiu1
zIm)%>y?a=W!hMkV(JrynH~gE=*p6(s@YoKaUE(Ki1HSgX{%}U(X9iij_t=hNXWk(M
z#CY2b#T)lH4k-XIb8<uf=1k#m)<w!MP>9UFRd;-Y$R6=+_Y9%_g>gS@3Sjr<#YtWK
z-VbWF)qI*zyMnviJv)CE*v<p>Q*bwc|DHbX#l38K{B=xnGmEW3bvQmP0C5`^9RRnt
zvx9M}bMxFRa8KpYDy`nDZ8}T$UF(?Y>&Vxnflx&-39c*pG0f}dkiyTh82>m4p`AU~
zZ7YxR;AuzhObP*$z^{#~-qR!RJosG$5jh+vf!EGtHl}VVGt0|2#b3O#4`znH;*G(s
z5@BKgE1{=u<q-w`vn8iZS|RC<Ki0%O1&Imrzg0d*N*@d_Y6|7PAbAS!EO#bwSjze=
zcNt*FNGkvin5-&{ADC9|r+%hVyvNTIagx!>R2P$RWT_oty+z-ULZYPN-pcoP$0W9T
z4!Uh+bA%fFY#+_OD^5UTQ+aC=aeykHQo_LnvX3G+rvjTGa8s)pgSu`)tfFv`Bv7eC
z5p9-ia(|z-t89-;DvvX3CL?jJ{IMg#h7-V3_*qp%e@l5cQ|JO>D2IMbY0Z4$x*w;b
zBCtT!Wz690*7Wt|`TKSzil2a6HIrt8(IeCE=YuedU;t_0!J$4P?fc94{b3JKuy~Gi
zz!-XJgLB1kPh$frrx~NV{2=LIl)<VHbviLGANKU<Ru3ERMr&e*DYO;_CJjN?R|x1A
z^h)o!SJRGi*_Z@N_%cr&`;6y{tja|LElFDTJU}mtjH{sp7wHNIvp0KWnFhH__Y6I(
zk7fkAi`bX?X|d&5>lX`czM0>oU%H@QqUwgxCU<xoy}lgAx{fZ;u;zqWR-naPhIE1C
z-}b{aOtjdc0>NKFLuSlFBcdJShXH(#Jc=mH{ZUL`D4_*6K@Gqs4d~8D^3RX(Xi4%e
z5U5d$!S#})@6y?ai(_(5p0Cjste<KZ_9Rg{&xKWzr$5<TK1xlDF_!ugIaA+^a=wxw
z{q6HWIPXk>LbfQ}t9TH^=*%uFI>FH^RelQf+x51-yFUN9hEQnbAS;&`l^kG9O~Gn+
z`}{}EC<SHFnJ?6HSm<lh_^z3!BwKNsG4NysBiXLsSRDT3ZFT&X;{h3mi_7x8&C^$!
zo+ev(j`Cn&9X@_n(jzQGy@jAuvq4Nj7Sqz6n|_i^dhzL1T3ml3De%$f#<`14yCH|z
zSo1IEZq^dBI^=6EL1SlZ-ZTKUnwe{UoH)jPXH>Q0W0iS|iCXuV^}S!Ip~gXk%35!R
zP+dv5jC@}1(`3V{#Qjw+fV`E<429YImte^GwOHPLRV}wHVw~ZF4Cb%E(Y)0IxIhno
zM5X*3mmoy90PYjq3hC#uG7CnhXa)uqwtWs_v<G}iexa-|R|LaiyBR<c`(NtUmIoKf
zM=dE<vxax37!7ZJy=eB9?=J_`i8^ADut-_W{$oNK(FiiMFWQHlgsfjNLkyGL4B2TF
z2~G@D;YP;Q>F20QL;tw=lE$M@o*K5>`M%#*Ry&PO9c5^>pv)t9f9_&>LS2Jv;reIn
zkz=GHvRZdY)UZhu4t4<-5Ny&W#}EaNT~kWbaeL<l8Q|0fV^|<3J7sDU%~H6BZ#);x
zf=E?J*qSd;;qa+(BWP5uSNuBoV$U;s=iDK`U`toW*@Xt=lm!_lF+EagM>M&Xux)ND
zM~>5K^q=2?3K@({9n(w<o=2+l+V6AMm4)=uP6)mZyrj}dHtGP26e8J$C}-^g_vxbO
z)t*Thd40G?$=088SLBe4?xhfJz-5C)9W6OZPzh4zj*|<i(}%p^VVt^Xm85pgw0kbX
zIfeVL4!(S)U2~FH11}B`QtX3nksi}sdF7PhyjN;JB&x6_P2y}s9L_@CW(qQ_v6KM4
ztigs$sVvpiDNexVVhv}-#Hf(bAp?eZc(A8E{p`o)7(U^c*(m9;Hti4z2#@>V*Q*vK
zYbHz8G2{K(VvIj(JKA@SRl0w4q*<qL{u`$vG(&ar=3oXH_kIzYhIo!yV*hy)tCaf&
z5!qaTVo}4CS5!nGR?(wwP`)2LE(t<xGnG{keQoKC2_-;|AZ=!Ov#o_iL_vbOm3@E4
z4!yRNry<Z$x5ZF2^wQ7I^`tJWz68P2O{k;av!n+IY|?a?sZ|uNea_6`&2U6%G^wg}
z$M80DQq0y0MhY+RX?yD=WzEs7zUyMFS8kNsMdLXDF31S@G^au}^3U|Lw@yGRWV_*h
z&{V{ue?Pz_?gfX1T4on9q{0lmR&CC&T@!fS!FGWyg)X{rQ)&yI=tL?*X^ozItwP4y
zl-Z(jh4+m52j#JVuUSK`e>&fgJ75&0)DSA?B9;ZYanVQ|)F`Orw6pcq!@Y2iC7qAA
z+1a`7(eilJr-WgIc0l)?ZGcdSC@VlvQ@23fvk#Cz2z>SvneW?Fzv3R&n`V!??U-UO
zK&Z?NtPb-IYu<p{3fWA+r0`T#cls+7hg1g)Vxy-)dE>Wk!v|_C=<-hQ)Dxi7gd9@m
zGtKeWSSDMg)Ik@4q~(R7&Yn{=uV19mj)t5f?j^t8CC(v<5x}DkL07BLPCLLQ`<UH?
z!y3@SupcXuhoi#4TO9}-db5bQOA{urbLNU#kk;<i6Lwx=uru&k#*3y0{xj9W$kDi-
zhILgK^udC52%dP9s9x@5sYh@BN=OUt3A|f-y<2cUj6eV7<2FCpe^5GQ9-jb(TfIea
ztHn?FR)@7!!W23;0-{WW^AU=`Nn~h_bSHSg(c`g_zf8Wa@idfP1hP_@j7JNXvNL8T
zgqYILZ^*psVk~BgqT~&P<9U&bUdmmZgm}dH@Oz?YAg<s7fa61u5->>ipA(jyA;s|)
z9VKOVm3sgX`fx^%|MMt-QV-5d7dCG<Cx+ZRlL`z<c~@1N0b)bQ%ZBVRZvN93cHbdw
zKsbQw1Ii3##eix?^>g{V+cnJ>uEUdOn@^kV0M$Gml2ptmF{bs;@2Jo;x3mWwSW>Z4
z#}xmhkwuS)P31w&yRZgitQ(Y<%wJJF6AQj_J1eaK@!nC9@zmDaL9C6r3hnHn68|po
z9)>6mWA(>|KR8{I9S$H&&HKB++tSdQh}5g|g-d6ZpffW2IKzvs82FU5bw09j;q%C)
zR3~}U;h`7IvsV^+>K^IlqCQMNroA?#7wITgo_X>J@l%t^#SSU@GF<xtk0w3+73k90
zkD_9LGbrO3)qJ}#mXbj!HX&mhxq`&htB9hw;OdZ2z5NNwbSC06t=Vw36y9upw%6>2
zBr0>uhf$rH`Ss7Jq_veoR>u}+<`tFir8kIaF$YPK1!DaBw#USh21g{bxdU$(_#m61
zCt)u0Fczbx`(UmTihw12_pNv|yD+oz(TEbjSNv{SzUYQlZn&e(CaK1^v-ACl!?3po
zbI|WY=$s^3c%}Av66erkr)z@WZlZ6%_GI^zB#}3J%*k>?Y$siWl(+T~=SWNw6^U6+
zWI3n>A{uPjD#`P?mP;_;0Qt;kx*b3lzF*hJ>+OXUCTdiA0O<SURUiLf86?}yK*b5b
z;QfYv$_xVju<iyB1w*D)m5RJS8FVA1wINgF9khgg!Mkm%hZIRO#n0EXDVCU@RUp>Y
z&{mWJM^-4G7{TH!b*7pJ(3teQ&dqT6$6xvCTmd5rbLGxfcHPCQQjQUz8Hk9_tuGG=
zF%tc|5Zhe>$4icJ!)64wptH&2a5g(YpBQPF4^^j@Bv9gJ-)_^i)A2ug*(Pk4Sg03O
zoZyKvuMjU0a-iK}KFeX^m~nns=uG%gZx(X$Hw?{<n3iK>Is+d_{ut)31Hyd8xK2iv
zc1KU(if9z6leL@a?JPXjqY>RvU8X~hI5#lP(iL9Zf6juM?8Zn3P+KIJI5vm?y~@>d
z7DugcxN%jrkkh~BIgadaLfnC0cJuOIN=H8hmLqmx+}*xzE@ouliNa@Sp$+8lOqSdh
z>~Pidy=|{)M!jawI22lkahM^Rjo3Tuig=W%d~au8ZYqDbph8C}*09vGd5Wa;+zKHr
z26&0Lml|-jh<f5=Ked@meEzZnya&>CJmHCg$TGQGF#R?1Gv|z;2X{*o&o1@;<7po8
z{o=Xfdh#3X)1R(dyOGdYqS~E7jyXb8Ns`thv#fq@13?LfObD_12rOZX0YPufaH;m4
zEQj4;4zUc43P<FqRA0SBnJCM8yo<R&3T*X+xX-c(lSyFKQ?`Z91h@i#8v++<U(Bfx
z<oztfT0-fHtn9qWR|3^za^S9$kW`$?a%^GPJQ4DV%2)z7ViH7LL&x0aCE{Bk;?XdW
z`M|hS1A}_^h1`hrOOT@?89;A{*^Kkk$+|LMYXoC}hqwKGFmb}~_a2Ny^HGu$(HrM5
zl&RZYlmjBvo2RJ1U@r^s$DaQshKr_GO=D<~KnXL^TXF;(13!xtwK!MfhicZ}ufP%s
zF6?jfIry&=HO}$+`@$mdrRg_~1{xypSzyS|k-Mbzmwe?+#i#s_CE%n?<EiCn@8SNI
z&{5U03a(gVBKSUeNrTc{C%8v^bX;rtm=M#u@YMk`oO;?LgC8GYdsd~4bY)2?kVpC}
zTbu@5D~~gl#pQ(XGRb{Z7`(~QCkO$w3X@W(+@N?5o0M@EMkJB+S`2=rhu>HFqT68U
z<Cy#U;Z!$Xr1OUYDV{s`PnXbmeU|gvnWvjIU6)xns;U^pns?JWs|}C?e)6v@i4|kg
z%K<|(1M#95h67SSweJ*1fK6}dcq`mScJaxYuq0>7Q2L)PFfdaRbCCQ*&=Rc7r@kB<
zLe^o%8}W%z4Kao#h7;TZ_ATME^LVl8D$AY6jY&HUhU$dEtma`p3^p~*b&=8Zf;1>i
zv{VWcfd0GX7!nU#wt+lr{f>alfi(ZypP6v}6(_^mF&q&9{HH=AQ7(SDm{Ip%$~4Ta
z2D=GW*<y=Nn>AF=FUjG!!%OR4g9l|rT(`6i<#?G7{SN*5N|UBktw5lpyhDHDDlSZr
z!nGcHFb|f=2|{=ByzKI9lJcwYvy!X)c=?J#OySD$J<t=R86D?8uX3TQHQ8-}4B8_d
z%s+b<8I2+U*_2*WX-J!v!x_W`c5nX(2SRSabWHWH_>@!859Hu06e_Y^2u}$PPHBI+
z4~5hqoeR8ifZLeiZ4U$XSNz5yq%_T9srFM1d&fE2KJhZGCsh+NG<megpm{GJ(;bFW
zpK7Jp3zNW^kPW}U?Mup5v^3T0EBCBIfV=NgNQVwUy^1u7YFYTW)fES_F5hhM&m464
z+>IE!`0&+@2ZY7vdd5?{*e^cZf&odQBHb*kNm>Lc+C3+Y0|iiktY^9mwz*F$1au~c
zk(^D7K268bmy%h52Og&xV!YwI;|ha>{;KLRp6?SDPv!ydJ5@B9+wKjVZ;~34;FgsB
z5+4DeB<v{Y?XA6{W*28}yfeR>iYjdH^68Nk)dHcs(g}T7xKg#5{>Y4Zx)+yI#}%Mn
zWHu=Rd`Wos@L9eR8~z;lfw?5(9a@(^LfyJpZEvvp1o4-?+_q%~&NUz#nNX!QXFnZZ
z5d;4E!^r}|R9}}vIKSGPZQIR$v*gGJsPP7{Ed&Z#{~VW)@?|Z^0d1`Q2>?%N^5lh-
zr)&MqnA%!C6=^?sC{smoGGp6kO_i0w@y1PXwXB47AZfj+d?*?QfV0I8F6S$3jfbl%
zlSVT6d(E3-VlUPzDz3yVI9S*Qi&yn4G<q(kWo433{0@nYR*^?lzKI}Bo5&{wCC3DS
zp;wQahI{oe6+zWcUua`9crd)gS3<`T=(yC@R^4s_-A=))<g`k8qlF1TA@L%y`==u5
zYre}~-zca5+o~nu+;ZO2;SUzt#PF^s0OHB=n_|6%QOK#k3P*mQyWNBkNf&m}+q;q4
znRSt-DSCqNhM2M+zkNPZ9Xf$Do_GqNzWD)fSg-QSuwE}~qUrpJke;f_e)*EI6U{{n
zHZV9)i1)NEm$~ucN#HrFwEwtUh#QilsKWvEV6YZWmdPwTM!0t<R}$dyK+*OQ(9NmT
zxwjjNbj(^W%|U!JR1$jo6RyUoI;bBGJKCuh7)}tu%`0?Z`$Q2(U;fe|8FUYD$FI;$
zNXdDZ7paiIwlc)!<7;))lgebGvZrv?-H5Mom*5PI%2LyPF}$vFYBErpsPSC!#5el0
zy}cvK+<L7_;|p|Ur8*_-!8I0fGSt+(sN+`a`%`m6pErY*ZSm6nIK|4$OrbkaGnFyr
z#A-XNJ=bM)xr{`u0d)1gf8ho&Y~h(A)Sv=<^yu(uue)!IW88NOID+Dr0+gyTG(vaw
z=IbU}bstPcLK$mVidB97G9<a**(w%YEW(<8Ur>+}7wbZ_ZSMz2=~nhFF}mw3y170}
z<PFXSIaR4MX(JDS=gEc|E%=ZmI8te`|0z94*osU#AKi#Vqe_3|@AU!5X|1|NEHoS7
z1G;1|64aEL=bdS1(=S4$bO(Y7pUQm3IU<`ox?qgY!`j!f-JNLIzDUE7WIwhDr{!mG
z1-TB|r3<qgNN>D4-D^6EeKEM#*YvN3+*%gl@?R<&fA4^`p;$Gw!=;$|xf>X}FL(v%
z#yWQd86WaH->nx$2V4SX-;(#vGt`sUWA4jsscZIDr1+X_U8Q1gfFzbqj|M|ZbQ^w_
zM|_0>$R@b9hJVRo#C2a};zWo|QdM&DnLYbO>UB{n)@BWJGbWF_opJGd7E)AumJGv(
z=UDw?O!^)_>2qXl3s9@C`y6haA?^M6yW*}T<hGM-f)DCPyV?VuA9(;6^h6g{tf3rx
zEcEKB-^fcdk?;B*wfTcNI)OV%)z}x=+ZR#|A{Q9rbduBTZO&)Dqz=nO<X|Ua<)YiJ
zg$M4U*4;jrR_28o?tOE`YkTvL>!xo9P(9Wwg}K_)Y}U!8H~OpJRVgWrSq^<rU@nE_
zH&@K^Q`&!5mbn1<7WjM12IkB{txt4f$IXc%San;onys0}x!029_)SC1?gZNwIUI{z
zeXWE>^Cz5FVrkE%D#+;=H|-_VOlHxEE!-y6wy{&vL3J)^t>@@Q-A14R`B560vD$Xs
z;C%#vW?g|hXt^rq%Hth;qqH(ru|<)87+q23`~n=vm1ij3v~s%Bu&?(^NRf2hvQ|G{
zpI}7%00$$IMhERHP_cU}kK$hl4*>wh|8WpuPnlB(K~2#)=I#TYXu&(-zJvyIwKmfK
z?<^=A3uB7n3_VH8`>FtJin;|I1wiY@KBEopd%H)dw|n!KNR5PIJ_jy*EB?imKdzgF
zeQz@ne*B{Qlr9cNjIztm7v8z69(QyDS9pfr97m(DG^!-8PpP^HYs+5{MYLh-lvZ%7
zO;VV&!J<qd!z<)qtq#58Sr&JwnzHB!57YB>5ZbZFKT4!Q8j&*wsHKJIjR3Y3Xz+s&
zehi$R*qUbyAN-ti4%Glo*%3{~2v=k4>VSmg>I_h4Yk8f-UQ*YNLNK))93mIm7fKOf
zxvkWQHD>B%>8PZ`O*nHJK64oxEveD@5#*l^162H6L7U?QRj4+lmg0ToBRcd`w(A5m
z3XQXiJ_8{xszn0MXflhxOn`L_EkyP)QJDNHdXSJX#j6HI3dwf6BLy|Za@Itf_2@)}
zT_=u}+q~A1_z@^OypeH4=^S;&<d^`Z(`Lr@C4%E_7Mj-75%#_|);A&)9rXnZ!s2qv
zC@dj`f61(J(JR*gH@=7)_D<B#6=_;bLkr{1(sJK<59jram>`-3aKtA{jW{CEgpEqM
z+*6Dwj*}i)00)OPQ~)<hePkCz22Zv1OBTJ+!M%W#gW{#{Z}hsAdjVk<iu(d>s92FV
zYyrF7)-US3u~Wu>ZAL~!2cHUKg393K;V&cf^zfNtsPIIu7)xZ54xBWv^$QCPsAO*t
zgF3_xR%r9oM_5G~z=p^$@Ga}zZ;^=Kme)cUc@@0)dj$q$x&gQ8!&iB{9-Ch!*-(@x
zcg}aY%Xbg^WAX2o*Y$<I9gnz@RaPmw%thB5c1%Z~hZFRnIX@q6Y@aTl7dEeMe%+Z_
z^z`rFECLe^dux|Qer+6Cx|Po@|Cr>LM{9TN4=?mew^4$rfT&o8evqB3a2egJi-(6d
z4s<sjzSr{#a(b~p{TXONRfU;k;bh_Dl=AVdE-ac`C79Xz*)(5UduwkuhOcJb9-P`P
z8}oCwFMFYT8yGCzo^0H@-H3UR4!z&Gr*ZKiJ5%gnLp0(8E5HT(_}w`_Jow!(`jGom
zKq-qVkt87lfS*sa?|%)Vk46XA+@;1r+(>(A+PJtt6%G)dI%CGJv$+K83C!;75Pse4
zsTspr{+zE}e*I1@x*fewS4=-%?e2D-@RA4i0s9pDZM@FRE*w7|z3%ssjUxoegZ-3)
z1<lW^Cx?44Kc>!tQObFTI*#8lDHB6$V@FPI0sKE_0EM^RTGGn1vCzo&0%Pa8RooA$
zn0_GrhSO9M))A5SG2Tc{V=6XJbPa6jqL~v-X^7C}W>2>l(K<_)s8%f%Ib_Zjah=7E
zgQ2y%v5N}U-?&O^!u15J67acd6?H}!$?5qimi5@c@9gp5h*0XOtZ3BBK!(q`CVBJu
zHY6|t0IKx@vWc_LPop*tKG1Eq87FRzNu27h>ArfDPV<4D?=O5*X%#W(lav~LeA^YE
z^O0XUM!s|Na^?MPp@tBu{^+-H?uteEjJi0ixy_x~^w~YZ?ER1@6qwSlCp=Lc{GV4}
z_a7${2S+qfqy`=lR(~Y3y-1w{&l`ALx5wf~0PMRUSuxN1OrMRZIea;^`mp@IFD|d1
z&wiebad_<K+LG-5%HWTsmU_({h0L0A)vkDxas#3Sma{P}d3?PcTsL3x01`0}Ptc<<
z*p`3-QRa_BZkWjE$&0LDwSUBNEcWEZpn-V&Ujy?b<<%8#?&W-DOMc7%&*zD?v%1(0
z0M(|yk?(5#ER?9*u&1gnbTZ{z1D$lQxRjT{63L0%g#)8u))`~P@!Fd-0pc(o2Kb^t
zW9qXKC#@GfNOMH-uUm`^c4nxKjwN}y`rf0|<(OdFQ${}gA)+xkiPNBZ%4Wp`Dl4Te
zE$)9?U=d80luSpN)WmK|bEFkIs>os;;I~Sez@N$Q*`|i%jHZTTT`Tjdcfo(q>H@dq
zGP(Z?^sK_N3>sy2%@VtMxh<0;UKO=lDtiWKDr-9m=t0P|7Kc^ucgU)OM=`4{uKap*
zs~pD{GzYa;H~i8N&Q|u=z^~9Ivlq*&DdK+LftFLO_};4H8mtbr2SAWd22}Ubzwex2
zJIy1P5Poeu{JQFWI(=GB2fVGx3RFNWmx<sWz3a6`LgZ<GyV!YQkW;aSlu?e-EB%M<
z?k<1Z-X05ninyDK@Ke`n+X6KfAEIIznoW;Y@@|Gi>4|6I<`TUE8&0FcMP-zUY$<VN
zppPr1_aa2YsE&Ix^aEyl>@oLB<PCINu<s=E3By8;n~=7B5}LK6S7~wGExJ`1XRt>h
zFVi>7!RpfmW_Gs`EJTuGU3YM>QGzV$Bejqsk~}d~*@lZfWyBbGu1Y=+fUj;BxxoNn
z4Bbe8f)^tAarW?_;twQ<DxAT*^??a)=b`1o-dz9hOwtnAC;$i_fi%?G+g>$+BFl7J
zV<0lga5x)K{)m&S!Yw1Vm+aLu$$Gw;#2omcuF5~BWz&+eFB>|Hb>)8Di9%~mH*KnI
z&MK|$G_~mK6tyX~c`0(YGL#M*O9Ywv0Nql4V`?NeG?R3GhRVR_j8m~<-AOK>Z_@XL
z)zCRyN`XzdbU@Hf?D`FWnVaAD>Dc4$V<@E6>Zm#guzjP#E>7~^Uh8kB(ZE0WUWkjE
z{45x3`B?|ukXQE)SIxeZr9!P@%dSc{=ROZ#j^(MP;->f{G21llZa;K^U3*#4vgCpy
z{W*5!TRVuaT~jw$T)p_8FwbOj?f$<2{AXmCplg`GU)>!p>u$lYh96+ZlVJj;xQ=i1
zLD3w`yW5H~F%IXlV(-=%$hZD!2eI3{>FYTX`C@*v?{jzI;N!&TAaz;7o&RC6{3)_*
zU-0O+Iu<sV+^N}ztNm`EW<CBI`-@FddEVE>8eHM;#5yF0RPTo!vyioA5-A`6%5X>f
zdSo7<F|W5Q!a#RTVRAQh<O%c{T>w_3HahkaZOf|1^^c2e<B=pZ;;g?LffxAJ{PRP0
zhY~o60A16Fz*l@6sCB*_s-$n`vSz7hPY5H{p1gdpJ?p+Q2^i_({bax#4H-}gdyU^m
zFx;Ueo~5zBJ@c+xJ+MHU;uFBGPEZ9J<{j)He|x>ii~c!;b@1>dTw$tjmBAOZ3aQ5s
z<BIR`>Hz5YY17`|YPLs5y~9d|3HLvXZTy&iE?4e{DZyQKe2%mxnIsh*%>NbsaTWb{
z2V3~3)$72WBu&k}_U>O&xx=XUo=d;eTXNNEs6U!+B)dpoel2POpuG{KlVlh2NHR&9
z#-Ed%ku*uBTfoUX-?BjbH{I-cVHOA*`TL=PbgJyy#$3LT!icWVXDC{hu8ls?gqJPJ
zn0W)GS5`<{jd~Q;;iJ5F4V*8pq7%z}QBmJ^1G{EeYzHBca5nSkX)2olXNcr_QS8dM
zHU6v=UxV8K175)iw0)?(=kgZMGW6|{DA|#AUUcoC_zH$4#0z*;`1n!#kK35?_q*O@
zH?^Y(cAdvcRjI;1-@w#rA&~vO4iNgGu__@0dk?2q;AY3P<*LJRj<k?h!y<rKo_C?{
z$eEwULqHxMGZ<Rdc5R_rBhc<LR?9T2J`7Bf^6fox^nM#ds704Xd=XUI5fZhT$~uIy
zGS7a+n;@9K@>w2?h1jfCoSxY=1VrFx=a*P%&3y!HVQi0}pxquw_08NX$M3Fa)d?D+
zwRUU`meJUO2dWPtzfwwAvF`(bB7L_0w<zvAWJgW0{^A}7c)X5;_Uw!jU2Y(-cR8>H
z{@q5v@7exjB-nfSTWVi8%ecA+5Vk_ktqc!Bh41C|iTwx7<d3_Z0`LLeH%0(NDSHoq
za20ql3K&LM{Rd;ne$Ozo*Z3SukzgVO0wO8fkWvv9uJZLdHXmHOsC$q`xPoQ+MeJZG
zu`)$-I3Wmt{2v|x>vAT<k@fKdUo7@|727bD5~^RkNGbCvcy;_Q`Ql#DIcccgU)HT}
zESsIwT1|4EL{Lo3G4MahjnMu6;wadb$r+u<>zdXY)N|6->RKZYRnbzdyHUefzxGIX
zOSrVMm|LQbr({PXIUP`XUaqLUPLCsKmn5nLK5%^iBC?iDiWKWp*mgPmezaeF#x*B%
z-p9@`QZU9_4i`M@wS;Px2kW)1jxer-$G(x?<AI4}`yEmr5hI%G`pS45yB&rk^4Y7l
z8EYL=>p+>c+xc+GR)R3dU;<11&^~vtGQC&RS_F@EVSMOFWA6jDQPA3O`|(35oOpIn
zHLym2*tQXMf^jL+u=8AGQOp)Hubea4{qnSnI=;J?I*0rjc^c7t3F<^71X)_J^>#~u
zWk#=6-qRN|6+v%22#Zjmd$+QGx$;$sq`&TDL0Gb$Rb{qgj_`aC)%zk#0uzMhHa?6s
zx8Pk9IP<VU!Py}xd43dOkB}aD^bpa=IDZu&-KXE(L9$Xq!j|!InygXyv6d&~s7Ua4
zYG}P|%u!l$3TDAU2Tlz@Q&jy|;3wyp|E}+dzUf@Nl+clp8b)56%kc1lFxHGbg*PZ}
ztg}q)cOeuyZ$U!m1RighA<qg;;MBYT@==~coab9YQ={an)tqRYylF4Vpq-jtii#%S
zF<rY#$8|ysTjH@8Dyu0TBms4xbbjijn)2C^Ge#57E@6p!P{BAa0e{MZ9=t4UL#nTW
z@>Bt|Q%`VKLuU1^S@dJhCU_{MK?6&ZZlx$m5G?AwSXq`0+EV}0&Lub4<yp0I+D|-U
z2~!t)L!N!;Wz0D)F$$SX^t0JO!kQ5v>eUYt7$zi6TX!>r=e76&W`rK`>upkYIT^Yj
z*bNzm6k7ZNsO^%@sk3ie2tC8dU40RuHoa2jtrAR`mSMbau_u&LbIij|oN}g&;4cb^
zd4jqrO4T$-L)D>b^q#!rL1mivuScnt7{?PPiAcQP&IyhrQvdF8nenD*zozj4Yf4fy
zjpJJe9<iEABf8r;XK2CD)i<Kwjib&x1jH>N+}qwylqw&aaK)tSkdue&^g0I`w1&O#
zd)P!#wMsLKKQ2awzbl%l?z;(dmy+zHocTHr{YD&nOjaG{YY8*`lk0QLhbt-B?M4VI
z1kXe;U_(j!2v)JvHsUa9$5FWfr%p6=H@!`k&z<I)Y-Z^{p0=~*Ylk>9s2bJjd_&-p
z55+o?s<`K<g5?XgNx129^b#CuC~&sHVjS0E(q1sZ4Q5{$wry-9*sq6G_`hvYy}Ylp
zXF)$1nek1NavLZ_?B9uUHcACKWxQr#KHolW&yICxnr%c~5GjR=0cvJ|sI<mTdI&GY
z{mvriWw~4gUf4De=^NL9^bw>$+=elVQvw3ojZ8uz|Al6AkF9b-oq3yA{q)1aBVTT0
zh4^V79diG8lrw5ZkG$NJW;}5`dVJn2!QdD}j9)57wtwc!@p)N%Z$TDMaJO}6lEguk
ztqN4Op<?R>b;v0h4t(MON-Zs-$QWz8iBs`1TJe$E&HS`Uvxkhu6=)Mc<-w|mXP@o~
za8^}+V7nBR$>@;WNouILcAn`ac`dQdV%@{TTic<J-Qv}0hfX8CxL8T!ZJ4TPwcj`W
z*!T$A)*OZ1arF&Pgl0i}F2%Lh+WiS@3P5itpIE)s(-Ym?aaWoFl6`4<>raJ;MxHK+
zWUp2Cn53O)l0CFNojT11j3&+6rZ9?=(eX@!UJ@z-a?wh9Zm4YcaT|PpgXZ`oSd|x(
zQV-5D1z&WqT=cGi{wb3^Jb)cIQP);UcZq_!K3}bIc}`INIL)`1{df(CK#@0Cgdefa
zPhC-+aFEMbyPi}7^4*QnfIIPKnm+N2xuX~vN5u6H_4-VEw<euCJ%xa3G5tMoiuHhJ
zjSPt<=>cD?5QD=s?t=JMUyT#`eqD=*>Rr${#dWE*ALpvq;&;;n0d}quL+N~FFiz=0
zzP}=C))X+?17J#yvhBT+C#RD$d)~5*p;@3zUtlT8V6O%MM@XQ8ZD*v~FqoAX@aoP|
zKBAreR&pcE$|2Nrh@E*C?|R_}-q=gGrCL>#Z61Q^Z?%U^=SbJ_D<0vF(uh=aA#G#0
zGmx70Nxt0PWPOfnC*7A`4U2rxS5Gl+YmS}6B9+DY3E2nf(Wo$2=8IV2Yo{<L!%n3y
zF<o^$jAliEAuWM|3F5)h;8?SRMb*LDvBB(j9S#cm5!kA%S+8yqE*$pvbka8ti%YL1
z?lzd=648*g+vCSuNY?e4M9SC>k<fL+eBcDsay>I18R3>9@VyG-gV(<<xHYC3+=j){
z4t;UE^_Z+&n?z%5ja<FD5E>JExK((%8<%C}WW8R1W#{R&78zq;)0yDl14&z07;@eB
zS_cT36?9wInTOGs1H4qeBM<n64lGBC^OekZ%D>&ek|)v**v{S{^9$56HIr<sRN1bo
zG_h3^3M&YFca4yVO76o&^lIjdr)^-}=>s@T|2#Q@wAG6)TC#BBHI;KWYY}ui)V!73
zvx6l8CN}O+r4N_UYd%8Q>3Q^k4^u@YAG33Vt-*;AK7bOZ!>@-3cLsG{1jlF*Jxf`^
zU?|?8@5sf*`(ve~t4GRX4&XJG9C?CNzoNJ*+xMA5rSGd_>1<KAhoe!Hry*ghyw1N2
zDwp~EKrxjCq3yK}EdX4Y%J3NMsW~Vk*BOog6}{l>NE+!X;rbG|xE)n2fw@*VqDJHJ
z9O(mjTk7H~&bj*u-jR9R&$e;uqQo#W@1Y~*xFg1!rWE{Y`|>fR`S}}GGgi&*$0F`+
zJ*5P+())U2ziN_+aTko$C#feY&CG5R>Ts0`jr#@bL=#TJg0!;pj;h=hF&Ib&W&R|<
z)gP@*6f&vIDi&`qI#POR0_i;^N-3qH?0O$=aOP$#WzT9fgQ9KfaY|q7VG|4!J*wG2
z31A?Zm<#eJA8Jw+X{AQ*nxlEj>;}7TM-_jz^*Nng!&p+X(jt|N@xwz4_&o6bG-Fy>
zEhtl)Xi%{t$BwV@!8v<>p_-?!S-Tp5mnce?LRTs^tKTl8(MrL^evrD)rCo-T_>-^s
zIp)_m+}ttr$YUf7d<$I_=Rg?d!(R6xY$@)@puLjQFP8>L@9d2i+#cUKP4d39@`f0%
z(E5g$gS7Srzt0tFo(&51G9K;O3F_qRC+lYUE6owt(0yhGH3@4!|9$3nTTlx?&&hPQ
zlh4q8gIyu>kooT?_<w2S->9qc10%;au>>{e>Hadgr?7&rb))3(IcWw0YPvXwrohww
z7~S?xi&1IH<(*d+OBYf5dpTk$3()?uSJuK4>`ty6c|L{%zm6-?*1gnf<zc)``pDUv
z-Z@H7T)u8^ui}@$lR{*iE%ZxF+`|trZN#3||F{9Lv;Eh!2ZWI^Wsi`QJmnLTCle4a
zKJGu=!}IeXXKvU8&E^+f^Sx4X(($wTl&?J^1;+FBenj`(2NuDT9R)p|)BWxC?(<U9
zy>OKGtUbX*4b!bbtra??MYE1cfzEW75?v(8A3PkXqgBU~4V%?HTxVXy|3z`&t$~|U
zMTAs^!>_eDI$)$2Q_o|fxsMy*a{|!8!%sugtK6CFPY3@i0S?&(Cg#P6$?A?@e1N7s
z7zcYe|H>OxAkGa(OV{+A%!l8=YX69gGANDcMNdf2b<fH^H&5b=nPpZ*`9{etJ8Q&b
zTg5n47=Sli8wTBnns7H#`A$krMJ8}cX6O*)8CyxJA-iJ?*Ir^n6my>xD+stO78$JO
z#a2S~*&%Mck7~-Q-!Id>a%Jz-#k)a8xk)02^mt2q{MgisddAFB$CHZT;V7NHiPeN~
zj=?M$60aIw!NLhrPalvwk|-ehN**Q?D*DnhMiGTg$KtNW3_>CtI`fAllNXBdJFrWc
zlJxglr_LY;uSb0hFkCoHRsclS8zz;>IEazTJtbdpa|FC4F4gAw1Em*=DKi1RF_G0i
zK|7-NVb1T){Z;%+f%~U*_PQTDCZ@=Cr^2Gf)p6lM!3}Ms(E{T$j$`@R11zSAS8t%M
z`B{5+mCqA8#ufd$<6xEUh$^Nlt&Te8DxqXo<HTpJHxq8y&1jOr&>7HTkm^9Xd69^A
z$Hg5yNs*U8X?*PxuWsuF<bYMMx*zXS159U+dM|VGt2H@wjWye@QbW%;OMCR$s}#43
z>1ABky~;QUkj#fyUqsIvD%)kpxE@Ei;u)pplYVao-vocC#`hAxLw_`gDp8kqmDvk+
zXYff@xXM5Uxz{^XcmbH>*@j=oMZJ|_Tg-@*Q2$_!Oie(p^^LWT6cfRKV3u}^6x(B4
zg5mr)HMPFOM2~d)O8rx4!RrT0#Y(E2H0}}^NjPJh_*~{9DwcUoRKO4`X_xj+wFr0Z
zxy0_O@W-XlzL^(Kz@;6p1~E&oAY0>du)D2>M6VLV$CLnjpa&2eItQ;rcPqC9BY9?}
zU-0vO2mVNm6ro2aHN92TW>o>n2GXocUAH*(?;Ygd@zlxNX%Kyzuy!jCgXHSnIsyl^
zMz#&Z9acdn>tkUl)G8AuYpNLQn3LvXHz1p`+wZfEY<%ieL-H|8MR<x@J3pzWoL4#2
zug7tX+Q_-6QUCz5w!@pMo8CeO#nD)08utSjVj@>fzoH8Y3Ls(^ZTr6Pu_cjpZi<X{
zIK&l?@o(ctpV<WWc)1f4yYu4tpXtV`Gjy$#I{3D@3v+;<t@JY03H4{q6TjuDspCk8
zkcB?g;=p(}NK>NN@7qf`rX&pCK^m9pDc@fiu?2Ue{6nG`n$EKStzhPzF)(;^S+u&I
zwPS)nlNBq=#I6aa=BEUVA9bg7UgscpntI<75wO>D1*0%i8dJALv2lasQ50IXCq1x!
zd%?<F_l<%TQ8T>#0!u_98Qhbowescxsa(ga2~aIjgivEmG-NV;jc(AY|8OO{RDq?X
z-c$3vIsiD>OmYE0af4DJ`+p2^3kVXVKLNNMz$3{h_Wj$POnsVKQk`*2Ryv3q792Y$
zDN!qbH}<cXcbxO1nRdmBV=)t7d``nQ+C<lAp`6opfIY07g4_rrqmGjOGwk4Os!gX~
zNYDCN7p1QxyYEZ6!6^ne{-p7(T%#I1-(W=??Es8F93dKi(t!ie@@g@x`Vce8If|r1
zfq0Df{gwtjhb#~!T29x<j8RZi^lK2Nb?Kr6mq+7_uXq%AhtuZ3Js<Xi(L2cFc&8`t
zabdAvH@gR3BwGuCcqf|}DSivpY^knnw@Jhr)@DYeiuPX_?nYBEWMrQ!P`CUUFI;ae
zf(L{aqhjkSne+^MArR+`#p$%zlA!?m+-ZI#A0!}1T`sJh?kR1hil7eQYsu4+IJ^kL
z6x*WY&murS<%+I`xz=fbx4ONRdLx)qQIp_<vw$1_HQ?Z<xaMQNm!{Q`2VEB)T8!={
z@iKh6O3Vm?Hu3J;V><95z7jJmH!MDP3<Zo~DyU_rNmxsoanFwHWwN)U2=8~$T_#5F
zFK1bhZCpQj$7oR}I)Fzz!3r>p@3A|MPmRWUQk{5vK0Evw<6`A!gZF%Dj~ZKaV&Iwm
zz8h_;jyA3+4W1yUa1EV#5dJ;`AN32B74sY+cgW484*wLlZ4VY>kPn<P(vdBG3jidB
zWHkG|aFW|(Y1<eL=9F6B+ba^bXkm?FiUQeI{MaFw2!@|G1mwb+Bx^%u?5BBHG!HrJ
z1=v`os{uN93Kg;w#aFPpyQuw7sst7Pebt?{Jt&~L19K_3LlOw6J;RBL6&mac#fgq(
zJg!4QXn`QogoWB5<F~IMrn`ABYX^khUWeN5fgys|ZiLj*X2y~ABQICh3SGk7*WvJ(
zc;O&9D-ELhslJrH#YnAV*gbU4^II0(-GEh92?T#2m?V&`5w;-AHL9+ssfuWKP~{+1
zS<=Fqdh?qoaGv5EUF{mSyaso|-fhRs?}W6n#kzjL5JNSzGfy|0d#U&z{9+KKpntc-
zt0*=vaBoj@mupVhz_Zev^SC;wai3Mo&U6#}tR-03<e$(^{m%Jas4M-yZKhroO4JlX
zB%VP){m19)#V~s9SF8Bx;vhjZf{}j#L1zP3z1Ih#rJHZQm{DmZZOQVxg-W7q?z#k!
zMwzp<hVkJrb{|?+`!Tn_j5$B8z=KVudsB+<sgp-34q5?GVxR+Me+5)d9cy6id8nOt
z=kL%t3z5@+fwEY<0{LX_#0GahIHBq3Qu$~=iU~*wUYb!5+Gt}gVh&Pt=W^qHW|YS7
zYT5(46{nRdW!)EFyR7+|1$Sp#p2p3A2V}Q2PbX~|AJ0MvDR37GvUz(W_7%PlhuW!>
z3%xQQu6tt=-Y8L{w%p-B_=DC>zQe}c+ivHYc;HR<yt8Coi4>n0XFNW&9X=kr@9axJ
z4jG$-|M84{hL5mBG(wpnTX^Bycai7%r&>?bxDYoDA#H5cOF1!K1A95rc0Lkc;$KMv
z`e>_`kvjp`418&bapJ_OJ$dQ7xk>-^%fP%{tJ**8^YjfWW+Z}6e)XMeQ!~=M_{sw5
zim3+qo;-7DfA&=lH2=|K>)JZo3aRS=d!-D9^?!ZB!lrwvb};TOvNOf0t!BJc)RrAs
z$>@qXn4!zzSi{^9Y{t&;{?Zx)=XLYzet<yf7cm?vr3$t*w_QHiZJ?KL419wacxt(h
z_%1&tr(RsIRsU(RKaafoO63)I%Hxkj(@Z4!%^S(fE`j||zB#?0_mokRL6@2fn92-~
zM^aDw3?N0VAdXRqCdStO>bM+D=$j{xDRnP=Q0wzZ>owL4!N@1X{Uojc*rDUZpUzP`
zpuNr+{t}|$G`*0Kxy#&bkR_AiVSbf?mj%s1@gd6$Ih=vzD`3&A+P$kMzE%GO$vXbs
zW|FC38%(1K|7X8;(JEtS%Vz8Y@MVV|0HvjfTe6lhGXzGhN97yl=cjqra?IUamC!t@
zKuKw+u$W9;LBiTMyi!$(7Pjn36J2;=t*fsnzsv+>q9cWI9#QusrK^agE2*xcN7im+
zcC$__KE6G55upr*3VgBfeL3?*V_x4>vm}^<$^<v?C*%YEoCNLv0fRt%ztKXIzBYdq
z#~|8{W_a|<!0`$yT7j!enK(3FOT>}QwRkvk@)OBh7xApnGvqDmt6_>iv%I}6_x5Z&
zs3{#}NtHbpN!uK8sbX?1Y3s6xMR)yUq^(1mv)do5)1Y1)(YA%JWNka%UM6n^{W6Cd
z@RGb$F_b1Nd_i?v^lA<1n7WMyjop6)&z00|r&rfObEwj2?I+NFw7TWwz|?dFGA^EY
z$gmh|=P$2n=M-&xrlLJ@?6Qw0c)C!{4y{+J*^$X=H9N9er)FzM>(p#z#kxw(j;y|(
zP_xjY9IC({QM2=s%u?H1R8Z{+X$nD0#OG{TroHui=9c07YkxC5W6PMuOc#H2N&-7F
z946>kZnhl4ihxKmS`+f?Dxt|DM$iCWOKE0xul!0yZshx!`jRc9AEFDp`mr_Qu%@lA
zjIPwzS6(KK{fRAm0j)_Pm}K4cwI(T&C%g_B$G)GbF!Sg=Ur5HW^-7Id8C{8tm6sR#
zvEwCFCK0u|=yH{raG1OQ>Aw|F7gFUJ8gm$twF<LReNdJE0jj*cmtbHz6#_Romoe%m
z6PMc-Jj4N=x1bk1I{|;XPEGIcg}VqBvDnHT)+rl&W&PdfcRhy)k0ld_-Dw(=-vGZ@
zU6{x@Wr%94f7T}wQds(IA{r)g%T`{Aoxx+f4$lWgVrr|u->t<G1yKg0)x=LWa&p~y
z4JvD+LJTcbz0M@3^Ulg1moa(eqU>}TGaT)Axs0)~NcOsn4S0XhGkEOY-LUr?7;&t0
z3v_TE4!)H%<ly(jOmUtWGIU8XDgEVrZu9P?vd{e_(X6LA<T8a+`!;E=nLJM#`ZSj>
z?g#GE+mi{#8N|E0oM0{xQ4&mTjRmhRfZ$O$v@z<pRylOog}~`7#1)J){TVrH{j>s2
z7#rBMvy25`D)N75*=ITb+Y%9_@EHo7k&q`Mj}WwWJb;Zq6Q>}{!@5ba;#2tfS@?}*
z%u^v~BWfd*lW!7&&E&b9n4W>tiJlHYD={aE>q4-$Mv=ND1RJA+La>=-7J@6j*%*#m
zVe6X!OvXtN+$C!65idTO6qsWqB_gA*^2t0|E;_JiHOqfakvG$Ij_bMW*PJtx@7F$@
zqr+NL$uv}@sV#F;#b%Zwk9B_OvZ?Ow`x=_`rBU6aA%~q(bs6%gT(wJ#M>USnsCl`)
zm!Y&wX~Ee<_J>lRhR4!=uiP}jLs*(ulA9)GO*XiC#_P4Do`>Y|qKXq#U^zABRlyk)
zwnJm^qJn?3myivj3wkp$GF%JhD7D!O$W6eEiHZ?o4YJo3n+D#(sE5Vk`bo&hsM@z6
zUS4#bZ}NMXneg<-3QvRT-IKpkcuuW13eU=9v+%5}wh7O6qHV&nvf^PPyR15qKb_Pj
zp9T{MLlwOVb#@IF6-HVPBR_C&^8LC{jRr*qHI9GUB~%HeOq?<Qq)-i6sP1qvv(0jB
zpNpB1jQt)Dvq3%*yVJwSTe(Wx;bEi<g;IFg!y-!_X6Kb{96aHAT~!zzk*~vHdX{5)
zCm?T|NgEK))U_<>orS+rRMgfRb!}sGkf>;8X+^~)-&IFItP7Zp+LkTjVLV>jrhF0k
zt;2t@2JuXd3rI>RTpNy+^+t``7#$Uk^(?F5IPpzJvmTBwp>4^FLh^iuwoQdb4Dw&m
zwwdDo%bi@<A(I{S3~gKUa+J0u6dNAZ_3LO`f?fwHi+N`fyZc!^QrtG6H>lgj;sANu
zVp`R=Eu3Qo?lx$H!fmm=w!{UW!GyyCa?^i7SY!C#2wZ_C;dt$b)kov4-&H3T+dr#L
zj5dB%t*6=eQFUUpP1;T^)^*RIY{e*T<{v{ZYg>z^%18U8ZK5MIbZk(zzFFMU!q#iq
z=iH=gUD)H?q-)vlbZ(Nhkz|K+ld3iB7M;8N+9HWs$F+0&w-)W`i_`sIopia*3T2lh
z21F49F*q@o!4@YIm--w$!+)<JBON|J{(7U6v9C`yOj?Shtc#8FMzKA;rJ~icCd(sP
z)@=B@39QA4l@_|U#<x#J&+Aiw#qq6EpzSbz8lfGD(r4FsL_QbqLA3C7?7Yz0^N_S3
z58|Z3i0N=3YUQ-AjSz+H1SvB@6!`8PyenKdOo^-t-ZUIo<)cfZ-G6IaIef+;#wrth
zC9UzeI0LY#Jk&zs1q+>8g*F6iv@(1Jvef)!L&i@(E|}&}k)vA6fI2}I%*U>^#;1Gi
zVrF-uka=T57&Q9>pa9O86!v2p)(aH<2#SNkH8|Q3m9_se94jM)W7Zz98zD`4nAR#B
zt+NV^zYvb4w8ppvj(?TVmEc%uS%hO2o56kaaMX(FPl0VA>w@FEiyAviC*OR`+}<O}
zXcZ|GIc1zUCj*xj2`g{_o6m2s{l1%Pi%H+jxt(VZ3(*j}-*=iipOm=lFnP?71l?LR
zxXq=cXVXxbn%2e1eWw2%wnO$B?7Qyum<_8`APdTv+6TvzEPrU=Qwp#~ls$BA=%$!J
z#z{(I#kUXDxA%`Xe~n^g0&zS|^T&@Hd11gR=Ogn`MKI0JKbx;NztyAmP^s?i;@{#{
zJ0~2@bBtqJ2i6wz#eeL?tsdlO@4H;2Pc1dY`SMX#>Gl!-%<1G4qnq=e7Rtqw48lk<
z@sXi$n$0!H*?+?EtSgCGv)z81Jm({aBb#8_9v?Xm`H?-Fu)W(hVxXXdjkb}%fFWJm
zCN|g!yVCas5=|p>ZK*Ss*rV?0FkD^{qEnF{RodE7bz<%*p+cuS^jocd_uHqucBk3W
zUy@t9yq~(Owf1<9A}_eTY`${FeqM)ml-7cfM#-0j41XgltE<7Nz=qk?l_48BF6hkE
z>Y7Wx`gYqn()&pf%WDq@xDffBr;wy(6j`H8rRVQuJx(idW~2vtJ$72ksh~&Jp<TAl
zd&T4_rM}!Q;wf0AL+$rsW#Uz6X@aOxxk~4!MKe)bMf;+Kz)2$dozX%tig^FDBV4j)
z)Etw2;eSHTLBPALN=mPoM=S+yI$Sg68DQ5{-u}F7o8gQBmFuwAVkSlS{Kjdu6#3YT
zgho;jg<>&A$5{}?)m}LnO+<dVEy+6r!Jhg+3c+7W{Y-i*^>dNU)X#-BQoqu(k@}?+
znr+n2r8ZK(@cSjy&vkC3{`P1M)ZZ1Yf%@CSHGfclSGWf1mr`4)KNYSO>gOUGsh^8%
zq<${aQ=j)6OVkHY9;x!cR`yf$3cb6K{kil;_7@^s*<T24V}Gp&Ztu|hS8v}g>l1P^
zehcXpK1bxUK2@IF#b>4(@m97_x<Y8i36R?p0ED0znL7i3d?*0i69D9;D8N(z7CA$t
zq<;iaNv-kOP+Mb|4A&f=d2yvG!Yh1c5IR?R4m&)3T~_9z;_Y`Ssh5?dO7&!9XAqc<
zLge|t>&~^BV}IP}jt}@|y`0S?<r8x?5c!pAndg%v)lw}1hC)Yq_6dDGP|iL9=)N!T
zJmhDm{M>Y|j|@*mHZD#Ft+3gQEMv!Ts(;g#Fa2=1b6QSTpygBRB(q%K6v9WSM@{GC
zwHh#qO=Pdui6HU{m`E>k9<H=KvE@L3%13iyjklEE*m9_Zu2kc#wA3|TDR8v`sLSHE
ztBbY7F$oT-R@N89s>0pcdXF5qVCg~^m(m+uTnSyt#g&$ki%;kJaNqiPPPYclTYo;d
z)UAc|R=1Wy*K%vAW#m?IsY&H0)0Lq5uF{N4*Ak>`g`KABH_4EKa{h%9WvQ(1h2dqr
zu=bSY@KUrD(p!3A0c|an!EFgbHjX68E$egLiMo=~CyqAZ#I^=DY-^$vRNPKWP&+;W
zo5`8G!5Zw_a`J@AeaUkW5e>6FjDJjjzAoWx?~2b02Fyeto@;6(M6|o6wznTYGd00O
zxaQQ1=8<uAYVXeLd^fF4<{Li37(asZkpAeri8j7!CS;%;0Gn^kyZ7aWTWlI75skd*
z9}-#u{rqPCAq!wAHutVi=UL-(f8Tm7*@F*C$EWnM<(T9Bf4YzD((tD51%EwvAG`Fy
zwffkA9A|i`KDOL_g#18OI`LWh*hByNk{)MdoTGobg`0LNrLzA`htoO5dM>PUX2L5K
z9}uQcDY2o-Y5CVs2(48)LGiR}HES&nra`1iXr;s^9S$jB<w9SU)wB(cSV$~azabjl
zX@x>!rSGMIQ$I)5@aD8=dw<RD=73jiUEyH5N;_8x4T{Vp6y%=@Cv2pymY>`BeYspn
zoRTUX>lIof|Hu(`fJgchQnc9x9+Xq3scl>7j7yIKXRtHvGZ&N<i3y+0#MSdAh+-Yr
za{}oTQJ*bO%Tea7A?cEe-*2Z%0=#tn18??0M+#+bWOH<KWnpa!WtZI?JPDVf4mT5*
z%OpI=0m`?OB|Inx1ZF=>$(PX_JQbJ0<Tn$S=qx<Oe{6q${bi%GvCjwcla_*2<U->Z
zmDtl+8nu#WB3X$<Gs<`Lq6MRsWYS&ab~fsHK6o>`J~?GH_3MWynbsxFx+v*I@d`2P
z{FFRzZFtEVcLS}{_=9!|!PQ#HWshtgS!-)SME+xeS?_}-<ry2<NyWkC)%Kbygij!V
zq|kFEM6F>r>wzIqY-*th0hxKg9VqK;Eh!Y4Rp(ggPz&ip<RwSsBIf(FXGB|w%sjXA
z!7CT+O20vq3dT)3?s{u7|A^BLD7SGgJWm7!XAUiEmtkN!6azOjGnY}%FC>@$GCU2J
zgt-|fmwoa(4u5$$oZ(jy7_j3i4d9@L?E-BD!I0&RtwhlZt82sQukRUhm%H+=lGSRZ
z%77(aYVPHbGjrw)NuuPPafa}l4Ea=B-uT%1*l^+65)-Isi@k}V_A#57>l-X4aeae>
zNvZucnoRMLzFSNV_1!?Glu%m{KV5D5`m>FUUZyrQI)6lbZDvCui8ckj$;H}4Z{}Q^
z_;^XAV?i@N|85X8rM?ydnoZT<jQ35~n2hIAhk(jHxZ1?beQ2*m!a}4vTEv=US0@o6
zTZDDASPZ)+kep#^swYBozC;k9(geI=3|GGp(ad!;{A9^hHfO`3t#%Z{G1XbbU^u&^
zqoG4w`+qLRl=+0Pa>C6tDH{o~jaH1rQhT#A7@-XTE4pT$9AHptIz%)>9Sxs^nX`_T
z$pkN?y#>AE^tFsN3W3@e3YtyH;!H?wbqZd~&2CuQ=-M<ft>~%Aiis>$b?6sE*3q(a
zCPwe;5QJuIveCrk)W(_cTB$>1Og6bTV>nYH^?zKv#s;n=wW;n{S0}?2V7vMSxDD2g
z>S)P8{ZM<e!K9i%QIB<X2xqddu91lh=2K+}E*)cC2TCR<*97o|Bu#+P%=KdkiQ=5s
z(Zre3mOA5S@m!Te+!YwPHW3!V)q$}DC32NT@D=q0ZQ^6OG+7LU@79ITaB+VA4g7TC
zoPQ3e(}AN@MOD(Z<;?2#F17D)4Zt=MYmcjr(;|x6Lf0=)s4_tpnXC8Zy2{cPg0(+{
z@aWrbhc~}nOopfD=hJa`{psDPPcKf-e=)<e>FoVvcBn|>j)uPvKMda;QhRtcIT;U0
zywz2Z=qwctHMXff7YnC78lD<_{`K(v^ncBCcy7MA#7n2s^S@Y3C;#~g{@rSL?~{$N
zolvwa8q>uVEGx`zi_qHe?1Rrasm)xLh^5$v63N+!I1wjM0(19l3szbPzjIhqLOiiB
zmw>1MY{0Jm_DTRRtvVFPYy(^eF<?F1gHyKTf7cG__M;1tg~tMagAG}c#z7m+Mt{Ky
zue$*S@+ATmL?ntN@ji4do*ydy>R7LafBpMEzzL%+b^ht>>}azDCSfxoC1#i%me@ix
z1aAfSxrL}m#_zO^C`+&&`E4m`hVQ0l)7k5b<CBSDeZdSbkH>K`J2$>P{_V^7{p;~~
zoET~y2h;NrgB&Q_z)kBY2SLLdhJUROSyR4ndjN)002<=q$&=ybY<lu~G9C^upC1fw
zCSS&*72Cc%{xE?;=i}sjysSyJ>Hs5NPCv~~CYSR>t9?IC-k%;nn|`TQkHA0p=@Gs#
z!@CfRSP>Pg?s7S2233Vehm9NPTpBf4v{G;7MN4|0Ec@y#k8b6W?4s3t9)BfTO=SYh
zAah1IXt}9F?jDpLPKHc~UyN7;53z72WKjYztOq54Ul|TWlHx)sN%nwFg{|hEg%VPV
zRj0a6iCh4RB_eV?;T_L;f9pWbvhpUzV_g?#olk2jAkk^Ilc3~=JQ(CYP}*ToAd~t7
z$gz;cN&p7n78_RKZv4T!V1Hf7#yhsiz|cah5@=7}mUK@_)C_^hphA|tIs`HV_?mYP
zN^DpPU%)PX$>3|TMDVDbu3~kd6(EcN!nhHzuZYc2IN>)a%NC*_Usjpa{|j1&1Eti`
z&<qcv4{0aRjd>k(W84jN6R&`7bPEWL)LXu{`1Y_$fKdk1jEEm~Tz`#Z(qe$%Q)QiR
z38FY*2_-PbM+q7C5Jeyx0a3(&53LxO5=aN0S>bhE2~rax(fABT>_XtO7v_nm0g!lj
zFscKTdcYNwP#(WQ?>EFU3L&_9U~#Bufxe~e0aznqSjz|iiWX(y9yP;&@B`&GAZw1O
zD?mRHqajLI<-axX=70Y6JG5&>qO+c)DW+^C8X(HsiGs^V)K6Q8dL_B(PO#zTaPSqL
zApTGoHA;1$n<a%7e+l{*+6h83uR}<t-5@0ARS1y|7oEYSWvg?!)OlO#ye*YQE4w{F
z&D^(R0@Ub+4z~&?X0mQWZ3l%j{KcWjRrUZ>*brZvhlN=Ut$%Qp091_#rMj<yLZN`K
zDsv>yM~|*dV|dy>UkfVe^Uqg5=>Km%j^oAUUx(qt>G<)}J9{$yb@=(yuNRNsPd*Pn
z9*-aYaWVb?gU)_}i`t*hhIePvcf)70Ks6dJXD7q6+3EWalS{Sx<v2S&Js*d6^RBb;
z!}xZbzCG0se1ASZ!~1`FA?fjps088Bo=`=Lc+x$^`N|8zjN|&s6G$9bJG%hG>;9on
zygzIMURHy2Edc+p@m<@WQ_<43+n=?z<nVy)U*h8zM6yQ*K#tNm_q5^x$*}K@(#q|X
zU`Evty8U&_33lGq3%B<GLb>8mrYt-F*l9r{9)<_}bboew@%Hlc!+AA*R@1u`)7MOy
z{Z_pEir9Q3MUwSCZ6m51Om0I)-yn*tYIQsEuG92ZDfdd6zH$C78z_9)jl%2B{4)RM
zo<mC~^QDvd(#d?`-je<0TCh{=kkA*g76h?J>An#^rnZMUhaE`>yqI=`o$Ui2WwzCG
zo_V}WK!1Hv=pn{CaY4X`v`4J>QLG@GJZD?rQ$7ij|9UkgdbpU910@TtAe<@g5$j8`
zCFi9QPD{IW!$VP&Qxu^`&~q9pw?q0AV1y;tLqTVg;|FQ^49jC#zJmJlhjm*?8#cBv
z-4eDDB`xt>dVtUkG4&G)^M|^-HZR1G@Vuv;Du1jz|KKXZ>nY(*2v21N;me|$(u2=M
zOL}Yh*VtRBw{mYS@`RpvE?Tt5s>_4w)pPb1i4nik&Y`$#I*N$TxWuH8JVGY#h)LVy
zLEvKR`RE)EbkmsGKSVt6;(-Uz?J*NDcyFN8(=Stj>BIf9eqRKph@FYB0|1fSGZWE`
zC4U{=BO2fo#%JUw+Hj<sM6X;3fFL9QfVh3^bLt*z)@w0jW2*O2tdJs_<qh@Eog`+~
zh^Y5*nhxUNq4`=J?gR}4rpI7Q9Yk}|^$U?gy^f*RoKV0MZ0K*v4kFNpR@ebsDGvk<
zK10*#?DW5riU#TN3+AiTK=0pdQ3YNVn}46q0Ct7rHloU)LW%c#I<qBN*oq(Ef%OQ?
z?L-!M16km$$bxQ$UnL9TmSjP%9nF)w?`hssZ!Q1Ysf6!Tq94KMegvQU5q$1P@Y#1N
z!F$}c4H;+^0GWX;yu*Y7LgD0o>j{Lfje07xPy(UrHyrogEsF_AB3|*zS*N%6P=DS^
z+!3~R$RT>wQ;Oyiy}W4k_%S%LvZ1)Ula$U(Hlr$;>!8=9^k&9&I^0c0*BfD8LRX>y
z!@+T`Q)0V>euJ%rkgN~Oy+nx&PW3?c5-(4V&yKI;5ZVKsRL<nYR}Q6jFB$2Z^`1e_
zmN#5U6TpunI9+r}ufddk9Mf`CE`PgJ0w%RVk8Q70;x33>XOjeHC`&Tw^@?7;Cmlp=
zy={$oqsH_eOh7a-M#Mqf(HJ-3P-3Z+6%OU>L-GEz<C9<Jdn;M@R`OQ_iscXXR#qkY
z-b%e_#@mP@0RkWIggfoRkgg*|*E6vj)1?i!4)b=m4%g{YkQKVbof>hcMmEy>WZBp0
zl61Nxoi0hIOYL^z(Oz&+cT_S`v0hVQ*LGCagP#(C4RPeeq*uONXvdiU0~HOkmw^%}
z6PNHkJOr1)nFA96G?!7-1}B#<K0KR$xd&nU=ku={osE6iv!ApSjba-*>qv<`#Zu$8
zfYBflQ(08gPKHd&PE?VTCk!&g7;^cE?G--kJ<KX%s}Sr#QApBc<J@V+s1M~n`=q5j
zqo@2F^5>69HbTR^jcCQ&y+OUX*`R43p&{k`$e>;sZxKIa!ze57mngV?tZ+SlEYGm=
zjyh{lA#$GWQFHaSR><CCrxnTYhE4ZFo{aB)`Ft8kVnyCHnvSs%{SPz1%6cf|@sz>z
zL%<V|6PWX-6tY$~p>+aa$I1e?h9;%6v4Wlz##m{Vm?BIovD0{g%CPlTt~O{)Ku$jS
zGH_FgLhY3VK$$r@wqK>s(QXfa;WKh{-rC|Z1;-g8{D)9^Os-&)zGH^@5JsHgr=lsj
zEX8pN;^c<iAo3t(kb^slXE{$bP}4_1_Ik`9y&6hW?vK0LP+D^@KA5U_K8n>=E!tYf
z2k{>^8142kO@Yhlg((96lwe8_WcLOeF^`we&9_g_n}0^WWCj%t`0M+BCO#mLuydh9
zQ~{{_@pt#K`K_I`?8oNXr}RE1wYAK3{<4ZChcY`w<d2E61^KlIPvo!TmeVoQ@6t7@
zDqB#I?v8)@El}MVgudh$F`U~4!2V=TLk9}=XvuvB;feUm-_ND)V+{98=5X8<amwn|
z$=FM<Ou{kc<Z~5?QHpebpqqGfQAKJD>H*PtpX@ULI$!gCXInaml7K{4-fS_|^o4q#
zMi1_8ko|WDS*d$a>Ru={W82TtTi^#!G+sSZ7b!U9v^h~1ANG<hg18MWP1U-g^e^Au
zB@msP6k(|X;{QtI4>3PWolwN2sL}K9ydJ9+SQ_bpqKBuYtn#{lu@2`=bFn|+TIBb4
zbAbd>KTT*suL=a?c1|g3$)2Xx`7(=#BnsvIO4t!lIp#IEBltcUt;1cq2&Kas*yCoT
z5LUsS6AC$_r1L*C7rU|}hn?kovn)XjzMV)m>a6;hrp3}$>Zzm$SNeG<8=huU`5`S#
zX2Qc9v*~`Qsei+N&3Ar5YB`SA5attS@XC-qxAI&gE^omNNNgas`bI7It1C!fIp*u=
zq86M8m1b%Uh0=b{p9zU2cZXy5J=r76Y0!^OgNsgrWW3Vue#&lbProI6ZsN`H{FJKg
z`FbZIqMbik@8-{TwRGR=q(1F@_jL-t6iun)Ivr3aF_OA}1r}PvmYn6}6zHJ^q@3l>
zg`k{bmG(EFC5)4!rnUDIjg3JU(h{;s6*{^BEm3Q_!<<-XiN??sv_!ka>(LShNuf6v
z&=N+IbPneKKua75s+5!l9Mv1s5;aC|MN6P{3K4EZOQ2ME3S(O01Pge7v_!Lj1++vX
z@w3qqXyo93LFOhNT~v`8gMJEH0zn#M##ho3L+ORIL?yD6mZ*f5(GsnmWwb;kwTzai
zM3&JKHR!LPB`TeZXo+>O7g}Nk?uC|EhkK?aR={3pi8Zk2gaT6joV@}q(M+_EmT09e
zrzIvc;TzEsC_UVG<>zRLAV(iKxAH5}5|Y%Uzijk>E&b2nq6!jHalDk4kcfm)L@uNy
zK<fX1uD>=d5pym;xSzV)7<3^q0Y>&Px>phtwe&h-V)j4%ml6|=VJnD<MsguB0b2AL
z`}V|y_e%El{X}DJ&!xYXASFlMTl`pJsOX9r@!F(BZReVwO4JLqHIn%%=hXiIsUCVC
z3T1AW)Eqn#0XUZt@h=mXChY|&mkvZcnST!ZKTp5xWENHma!k`np~*PIpT<koedV{)
zZ$7=Ub`B=BQ(BPmov|obhoT*^q{PzkagtbC<nQPzDO!9CY2hT66M}xo*bwW7Pf1tO
z>APaR2=cqa!KeL+Q$Q`@6yNcFwEF-d9x{o^TZ*8Ng|?Xl6Io#s_68G$y}=BT4S%L1
zk6F@S6&Q|->-mVlvh1m)Dk4Q}=6pyyFO*RYMu%gtHQ0a_YeyJ>e@-wZ2om7Jn}B)v
zcx}GDoOb{0&TI)r7=ztkUv}XI0tsV%r<Cx3-Pganx81Mpsl~OL&lnfcbEGjuwdWW?
z<T?}19n5})@wEUa+f*3y@=IbNLw~d!(q~|0WQGb^xc*O!h-&52;J1pLK%_=~-~-|3
zNk>-}gp#eoA^wpjoG*V@K1UG@Yf<S8!r;JBVe&&rgd?l1la7}_nFM3X!j}>OIm6n)
zUF^9SGfJbd?o6P_K;|GhW&mXV%)w5k<lvHhm{*RpkY|Fxk@lE{Ka2!F^nbLpS5am+
zPz5!@DNzD9f}*kNj0=G#;VgD!l&~hN^dza;{CMO{;n`}6V|Z*KB!SM3iGTY1Awg*E
zBn5FnVC#=5{$SZN(F_|xfi;(Zm$?`zz+lS-3KyCtGMou((bTeB)(8buQT+B{$%DX=
z(7r*tstTm`ODvV-NY@p|w|_+_I6VmGw&AUSHxq3DTtUcOZX4iAk&u1^-XZ?b`z?5v
z6r!CVE;yx^IFuMnF{M6^Vt*zXvEzbk55mJ@!T_?I;`JHTb4ds>{4u859OAf`y!%8j
znfv-P78CrKQ`}(}6;~bv!+55HUrv?Y3618G=D8|x;d3gKLw=kloqx*Y+3G^msDWKW
zsg_^g(mv;Y>*|^jYsGTqY*OYlFo;S>!dB4SSUn!OeOdBj75Zn@t2f;fKWL0jl9h_p
z?Sglg^EthYJLVr>YHNXFRJh!>>rwzqiM?3l&^lq*m*=Y73^=*U)zq=SNxj+SmTL~c
z<WQ;2n)j($kho%jfqz!8WQFCRwhRcgfP|ag8ZX39k8$uqzA30aqtK`e4EYM}nJdN|
zu*MxxhlZZsk2^L3MSJ!2W7K!Kb?(S@R$H8L$6KVu-gPTZ?=UR}hcw(uHG-~jWp1fd
z@RdeIlWWWN!9?qY+BpGUf5RX`8!J5dcRYwVH(?(l=hwn(?tg^K!w9d1=kTg=`d-5b
ztvPVO-)R`3HNW(IufP8CeF^1MHR!@b!rx{xkp~TpGE1Wg$0oIKrvA-l69JxmxeWiX
z*@TwB>e;(l_d=EOg21#w{XgatE9#}<`?iN`hZJM#l~am}WbK%uB3d`6X!)!gR8&;!
zCKVOQx=}?9`hOL(ii+o|VZ}Ds%d}zx?qyuD4fi~+*Z_MOSZskkCl*Dll^#0HA)0bj
z4sn{(hwGzIk7_(sv9038ebty?T|_Fr#nL@G*qYPo4mQ>spPy0pIwa~;LzcVN@qpfQ
z&)5i)W8uNb_><bpa+0>Zh~`>_H0CslsVtNNTBHv&%75=uvovqq!!fuH^OWHSL)~0s
zLeucPY<Lbj!=ZWy!&jVch|{gFE{&k01bGWM!ZpW-Y<eXeYp%wrpIQmWC0L55fahwZ
z)Ezvf=5!8C?zf6wWp;F%6fr-e0fD-O;pX1K{ka2mEd<{qSe8dfye!P<H@#aAGD9zB
zhwS=3`dsoxsBau*0$*dvUOUXJsrf>5;fD<FMxZ5HF`WUUgf_aRTsde2X}AZy!=TY|
zEtd4Yyrq_@j2hv(QKLbj6<+Bp6>%sND_o_f+2AQPVHx8!Md#(T`ww>Uj)|Ai96S{P
zIhVmyDHN9?N<7AY?aeQbKWub1Rx1gR>9k^T!QttmrtS;g9)9!j19lh8!QBxnaj|pW
zD(qq0oxoCoby!XdthM-VLt#bjA177?n*5XYXW<*c{_J+xtMKrhurD$B-KZfB^9N1I
zYlENqcbISeO>&{V%_1&O84WerXrD#wAU3K=bCdg%3HBj>zZ;gEme6~zlZx$dB4P(W
z^h-VrC!OY(){Un1K3a+;jzM|jDqr0`q%$`@c?~9sO$Tt7jp<!6I9^4Ue4id4s&Ds?
zo4+@+Gc~b4)89XD(hF-uIUhSB6^-q_{Lwvaeymp=L#%rC_oI&>?VLy%K~9N;&XFiK
zJe~i{+2?wHdN`zm&c@jNFiofH6Y~@AoX$S8uyXp*Uy)+u<1UcNITeNLjTo$G-X9Eo
z@)5`BYl%f$-D;ayM`h#8xno3B_F#tcLNj|#x+Cjt6Q>AKbE0hy;3Q^dTe3k8*p<IG
zF4^J8JqC0VjWW0euNkIIOdx9&S*{K1jj0XL`I$j~P28AkMwErFj3fz9(<XY^hrFTN
z-GSTNNmg#Cht{x@!dSmbZ$ZebjA(wh3#WyPVOUh~wnGw9md8D9wG^}8J~-+kj*pxz
zj>?L%!57k3Pj3b%dVf$v45n5hKg0YiwV+5+ESgg4pSB`r6t+H7WZqm8(bacteIEEJ
zt`z=%)YcbB87FHu795a2uG$#F&odv`SXLy5r4Z3XwUp+jd89?5G%BV3{M@vUOd2dh
z>gxNp8U|#AQsURP67~sA3W@fSl7&R=fZv)vBP4%oR{IQr=k$`#7NErgp#xXfu&9fG
zTTQKzRWNbQRiucd=;F$VWzEW2UK6n?H}ZjhJD;wi_eWxbQMK48hl69Gcw+#y_?b6t
z22|hlBm<#@?3%lu60AFLC69PAC0ZR;c@mWmX*$!Bffc*M7XW2pD-{FhT`W`ayyv)P
z+h-}0^O@hvs7F#g1;z9gX8OXezVZpVIpQ(>jtTyFiUpR!%}W_}!9!z)-4{mP+%(^R
z2%*}oWCkqB^%zMAl<Gu$0;81f>&Mw>s~vTOVB$*2)GOcRrmxYJrG}t&ta9HUUz(0t
zQ=c$7Y+>&?q%6{Luv3%U7gXD^iM=6ucRV;nD--I4j=J8X_~u%wcmC8nT02&BeHEAM
z#IcaxRc%w<ce}d?cH1<X$oBwITfOLi=k`6oA$vq$y_>w64TiTP6L42oyi<d53|*{4
zjJw_SM`ul8#B15>%6x(tKZBU5Upodf)&`}+SBNu}VZDF%Ykj-Nz#In}n;JLEXCh50
zjaK~>sHuc5gf*3xwO}(K$vPC#gj$YY07c4$j(RntNGVmKNZCn_z2%OMnnu%qMTv*|
z&skTQ-DYUl*Y;!k{b;_HayMDRd!xS6+j0&YfpcW$oi@gDYA{fiS5g!C_VS!72T_EQ
z@O;iS6>eJlZ;Nx~_d5&wIqp2JHAIC>Q~BspTq~p3B;r@%TAA@YH{mS%cR1|}GOU6g
z!Afa<lV3F`JF4xzD7%WI2(hw%XMw*&MVgXoY0HI}RqOi#=_I?5Dz&A`uGeN&S*Kk3
zIw9GvebA))Ff-_zU{lkd8_<29$)@g0pU9>bA}6z{h16+mYOUx9>FB(2wla?fuHxw^
zo2u)L>9(=}97d^Ch>W}qsz;}yM<OptQR*l#;9}@)uKsHBN+zblBUZ<MTeEay(vll%
z_v^z25)KU;hp_SliA|p}SA<GpOv%t%MHc5?$7DNH$U0nXGY^%c;X){TDV`H9DgU^z
zDqIrf>RenCE)<|e)<g`Uv=KLtm>m<DqoptUbTd-eI2E*6d4Q-aQQyd#NJ)&VBU!5`
zariN=iIkKHP2@8og;R`w>3Bw@Fo4XgiIcu6t;4w;CyZ#+ZUI`_TnGscIk<~4t;7h(
zy*}hWPydxc&*WGMcf>n?CiPKDpUARSLKmVwDlIkjF+Z18G>3^+Y4XpfIjV(DBugu)
zS0_trtDi@f*1LIGvecxsWy9skQX{D=k6#Z`R0fSCDV;}>n$(42Q9KhVO6e0x(n{z;
zNKt8dX{1R1xD?_GAVs;*iG*k+_01BZwRcV<L@TKa5~9(my}zxkr$fC*;b;GcZ0NQm
zA(J>u4wW$_jm7=#4iTU6AE<?^P9;SvqVoQ6^Dh;fLu8j>U^)~5IhTQA855UgQ9Ln!
z&#&-t7sV<fDe41B2bfJV1$t|+w<MPqO)kx$0Rr^rm$KzqOZLdtY-f`myq@vSv#7@>
zillV+xzpV*yZeXTizAuNDbL`rd-&9W{6ouM(DKl7=Kz!?=pMdw?_X(+R<3)wjMhYQ
zIr;46&igw8FTVi!1;|r?{hc!)Pe`7Bko*Q&?piLC&n`VXCcG8Aw_e9Lyq3$y`~Ks@
zuREQL)hgKdG2Y5=Zy$Dl?@p#8DYl`r>=bC|{`j(s|1iMHy3ir20H*u;XZLCMTlv;;
ztm57i<w1zj)-q>Ap;gSjvz8P&Hxd;m%`#gf6Erd*b^~$zIz5*OawLR@&&CdazcE5A
zgWL}Z>-Ukv$X)z-e3fzO-|-c12OsFj<00LH)>aSPDI->%{(Sx16;Iy%a-t2bRe`CT
z2*?}ZP3DOc!Ug5Pm58&p+Gd+b2W7&HT+!B&Jp`m2<IJ8&Ee^DDqPe%vB-g~}uH81a
z3Pm*=ZL?Zr0yEp94ZL7i`o7G6WwLcixk;w;^*k;Lbg~Nkv&ytL_gtVsV-y-H4eO23
z6XX2wy}v^u%Z@>(H|f3~OP;uh4};Tt>@f-8@aS-JIAM8@(e{3N2_?_^xAYeH^cIuG
zdHpD-GYbQVm3OAIS_Fw9&FNlq_Uji%i(L@+p*5MS3@hVZhQ9f9JCG=U(piEqC_@G0
zgP5OCX9i+c1dyZtsSjj9$<${9!8i~vqr|tSFb{8QvH&NsmTf`d8Ex*3d38>oy)KhP
zYcoTVU?x1ZQ7Fol?Xj5IU$=J<ZeGe?>u$bP`9ujKf$Uj6&#ajkMD&cDeobI~mQL*T
zh5z2yr|7IQj4g6Rj}k0@o4NE7P@Qd&BL+~xHL4MV7ionOrR&y_78FA1>6&%2wiFcO
zysQrra}@A7<Xd$%6azbP>uiW*V{M8tBn!lt))+&wKnStLsbqqZG;;3Mk98v0D}%fo
zWF#0bK_o7D%H#CSyAdL85OI;u(mir4GKfw^g+hA|`Kh{E$x8}<9e22VwWdVj#Zi*S
z%7)D`$;$U@95$tXqqS+%GyPTuY14C)a97U<*v1uPe!P63Fsx6I{9r_X;qc}m4_Mua
znd0i@vpew_lG09`qs9Hz833cC#U*G+JW}R{>^;VeQZ7%DboZI&?CHCTX%-b(Dn~ih
zSFwhIi+z`qvm#J`FiK0<9xZTtz$nK4wLD4xU6iSt$@EkhtVz{`xIg@OO3m%+$C;*-
zW7_C93Dr39E5e0B&Cya3ZV_q$x{^?fQA(i_Oo}u})E$bfJLHArsWBjdfxOrvC7HK!
zw?UGR-qOYYOp3UZwv3mu46|r653`GoON|nP@&=95`JHTkzaQq>$1OL^d8}bZna0Xv
zEd1iLe(`3S91_{XL`kr<woMezR?MLM1#5V;=6KP{AQu@o1t*829jmh#Bsk{m)73gw
z&aHhOk|a=^{5)+Q3R23jtwy2C2#vnT+Bspk$07Rflh5Rb&rfLc6cgFd<|z<YX!Ddd
z)!RI!J=bl2^K7&&wRuY0p1aLMF?(Wn1({ZNL|J>@mxKR5S~(l`W{SC%_7}Eursz}Q
zc&1+x+c`)%<b$ZF-p+}%Q^-*(eWO?a`A`nD);DsXoGr{-=^Htx47=GtD}5s`sbvVQ
zJOm>}H0P*q^@fU6a~?_$XFKSrv2vz)fpN=g;T~mwN1;aMk+gzBiEAyAC|>IunMZ>v
zhOvTpu3N_>wTN9-`WfP-Xt!a;`<SD<I_FZ02lh>*OD!I>Z>8O{)Zo}eyCfyX*;d+v
zR{*|I@Ub;z$t(%}F*uzm0mjfK=@Pn&L)!$su~VHvHg~Es)V5A_IS}05p>sDsuS7WU
zLyT~LQ}FNghPA}1KqZq+Aw+_ZX(AJKY<V?ui6FlY+`=V-Qz35bGcK!QMv22zg4SY)
z2oj8KVIGaFY49`VHwF<73OqN_ZpB98R^DxFYc|pDjB?tv(k>y6%*?u#7!u=^x3I+f
z<`|--L^7A2c1_V+W5^V;J%&tC8)HZ@(6-Zmt_+1k*jbsFN2SY&>6H{oiq$CoDdclK
zQb~SfVnvhrt0_`Jv=du=HiQst^ckY4l5LC65S>(DjV?nn3IdictDPP6Cx5z-Q~m5f
zDk|lA;#p}nTh0z*bP9UztE~x=lXQ`q{NzaiQmdc5OgH!@KLwP`4_h4-f+C_8N5$-a
z$QfF8l(mb8JW|I7^0}iXXKM9^cs%{IS39tg*@=kE_o?iQ!Jl|2gGMWFE<KdV(HjqC
z3ec5$-Nh)gLzxx-j9tL7yH|86*Y;sVNl)NnS~K)kT655~Xw65N(Hf~Vsmqh7B$rhA
zL`t=U9e**U8G0+FIp|uH=A+C`W!_AG6pFKxm$gVJd9@^J&u}!@r(E=XVq!k_6OaEO
zW)=R@<G&oeQD_C|N<u3}nXTR9BSrC|u?w6MohPnekfbm$UP^0*-b!l@x)!bZC^K3Q
zN!6tic{9alr#UV?ZKOteJUJa&?lp~jElCw!=!*=z(G@x7N}k9E`7e*RH)>oYQZqtx
zA*~sBBds~+O0?#KRB0U(Q2QA5H_$rX-#+aA1<LUdmod3M6aq0hml0k<6aq9cmytaI
zCYOF$JfDB@l2_k6e0ktm*e@Oqqi`gdHhwT%Fso^J`*`?yz{dL_?mloPz0<?*LtJKp
zwDJRIMhI*0?j`;WzhL9*pWeUw``d2?1KF=(D<2diExa2Z)1$S_dn@GdAw4^st+cX(
zU>;Jd!4V_nv7r!-3vY*mg7%aonr-_KRwc8J>)C%^B{VZmk(?MzOQQ_&#IjoWeV8v6
zsU{ZjPNNv%DV&mN!z1VmRd@`q`E%I{s{ag!cf;XL)~YsS!<N7$oa7Du^|T7!!U94w
z&6XttPLL~K0vk-QYHzW+ZC)c>+NhvB3JS7*7+H-ME-SN3mn9rkPAE&|y<J~WLUU29
zj0k_#OyedV*z*CNB5O8jcl!MHF_XOq(f#Z_9di^CN8qHRvCy;=;jHMjux!t3^Y*ek
zd#x0lYj~Z8*C`jhs&MLgS91thz8PjB#j72syS>?~4j4siwL=|6sE!(dYst;1hK1l&
zZlLfUzRJN1fkpXv{nvN6SdUU!W{qemLzsUP2vy=ADfB6XGHsGVH9o7bmY)fsOWh@f
zhjnc;$G)sYhIwIIj*${(Q+8EeQks!CmDjBE)RL)F{ixe;t95=x!WHBRzY^9%=oaZc
zNS^?gbfkUOQ_#&9NvDRd$|<FVBc9wYo_do{Mw(G7B%a8R!jr$8ETFH81Z5w;iBEqN
zQd9gU*iEK{<%A9=xwH~qMS6Zao~-l9ZnRK7IU<0vBrQ5djb!(0kcyiUP*7XqsAxQQ
zCUHVl{v5cI%``wao_wXF@syrMC~4&C7|sqnsG}9DSGAH(?`b7{vyCcAmf^fs3sHZ!
zGmd=HW3XaC5F}VdkPQfO?a3<`Sd)KKal?dkyW|Wuy4fwK3LYs=#Jo)gNQw+~RJ?vx
zLojG$qqA|b#_zCOo6BtG$w^`*1=n?s5NTuHmO2(eoz<`Imjh%0EFtii<p8kD74n_j
z=oTgLF9%B$Oj5Vpi@Tt3x=j4inMohGQ4FZzd$ZsOnUczr=>mQ{p#%1G!p?sXP&gEz
z+|Ex9E}Ks)PJTDts5pH-p^|tjP~=l{vKnR8SHfm)g7?EyPP=n`hkIWa<Xs6BB1iV0
zwW4B5;GLpkst{69@nifSwWdS20CY-?x-3-ko>RhZ_nWWgr9<u&j<aSVmJSQ}&fB$9
zkk^eaZe|wOS%0p&Y}+|$qnLk2(LVE#(H=L+TOzt_KV2$6vy{tTrSo^U(Vo%F*HZ*L
z$MM46*qBkKu`zqbeofb3)m#h9^{}f^IjdF&*LT53Bi1CO*Xpv4n*}{+C4x@F+kNu{
zME!iy25NG!l?|2+U&L+*%>ZJB7i#$B`{CW;O?zrKTKhgBgByd4R3U%E0Gw{nQlYm{
zQ>oP|ajyoaGKhyR*hl5ZkJn>mgrFQM<MLmGe6=!zI@%qq_*TGFw|1dQNH^$jT8c$+
z&G|AJRJZzsLlY;)-BvK-Cn~1>b=5ht$<QMsAncP(3ddlU6WO-9fo13EBXTP+DetSW
z@?`j}I<WAt8<rYwrTl-09(y;dcH9(MVj_;Cpm`0N+&%=g*caV3X?RR1Bkeqm`8rWW
z3EZpzviXFJkp;91shP8CP!?4j0(8354cATSfv-GlLPph{p=Qe2iu%<LKU^d&q891V
zAPjP%n&yu}(r2R$vqCShv8X*^7wPKAWKb0a5{0;zn*LX`&INzn`^jbg4KS}?$Dv2(
zvZng&ViiZuFb;gzI<GPjvQEC7?J*@TnUU0rs`;v@ge0uF?Tfm=C{!lG)g+&-bAeA1
z+_h`pzO4wk%G<_RVqGwa-ax4n4Tx<mD8YnZFjyZMq>%df2{XRI2&nbOo}~zY));Y?
zA`H@%Jk`Vcg)D!fygk7W>5C&-J9Di*OXry*c`o`qFze4#2Li41V!?^KTqH-)Obz<}
zc5xKOq&HihK2J|$<dHoIWI!gFT7rMBYDY?gFmkR%qzdJ_JuAeab79U0D|%*KrJmC*
zLHSf`aA90pa;hfggq!E|n9ofv`1w?-&+|;Fhc=qlG4Fp6g05XmJ4*w}oOb5K&z1EF
zHIDNQFV)dTt_E{fAM?D*&21<9mkCs<6oNEq$GD$1r?X*BBxR#Xu5UJmi0DCdtHE((
zkm{03@*?MWb32zYO7CW~q77_?nGsd>TBngjYd*~|OXI&jsm|w1=MV+_3)O=Sw-FP0
zOp>a!QU`x-R$|g<CLBG{C2DGPP3t#+k0YuzM5z$@bki{;HHq-raA~JwW$HDb76n7d
ztUfiH1I(MBX5z2~J&E)WiKjneT!}Q1EkncWarg%O*(c%;T_@^{m%?Aotqi=VGw|Aj
zKMs?6^;G;-)S-djgTI>Sx$swWIl-Tv{XMXn5$=D%=G8bgUx(+%jO_}#P@*xd%twsj
zNL%K)_<tV5>DF!Hk<oKL%^0rM!E=q_hy-k<UuX<Rs0AENcGU~9=T95MH3IG!!!;Do
zFotVoS|7u;@;vVtZl7#r4A;u`(qlN`sR}DkhTpEV*KqqwjN!rzN^AMVG2EPb?-;J4
z**AZN>$6!M!@-l$*7{zHT6P%3HFwUOt`cK5r|Y-nWlmR7(Z*;`G4E=ly!!3G7hB*A
zy?AZN{%mhyHX{@RBA1LQ`vWu>GyMJzYF5@XGcD66Q4HeLB=eb8pOHlE6;v{(N*bxs
z-y{&>J^u0X>u>n*9lrc>h`&S}evexp0eyewu`~SU?qA4-G0QE9jA{{b3Cr=G!Sqv)
zxn~ks3i~`Ns+PI*BXjV!qUOCcqReL+V4@wQEDXFPnq22=Lw=ryPk(I88x0^J&T=$@
zKRXnK!XOuDzsHeCd$^ojbfg6c1nqm($J&By*y?yg^4SI#2u#~J8->|}Gbhb1oZ(+E
zrPVghtOFMOr^Q)0vxl?w&Q<_5H2ogolMhwUCo?1HrLud??IVXOf`{M6=qEO{HTBM+
zN<;KaO>M*F&uMD6km*GHHRrRAT;MN~)C+;LW_ISlW%fg${{gg9+F6%jU^)~6IW(7X
zjTjS``y(nSmpfxTpnrJ!+w-3eJRzIE3x-lLYjDyA8R3yIeB!EsdULU2NFOSO)^yMu
zT!3EXiUMXt4}^rU67gCF%37E>L;6^$<n${cvRwUP12qjyYT4gXSm)UzB3qGIhxDz$
zV?ckKxJRk@xBuP$4fP(|vXv=Iq9DAa(pJ;Ab{rzs!(V==M1KI1DeWeWju~(nQ<ZN^
zx1(O%+N+rc$B?U<RwNZuo`$DHlZ?-Bj&!JsK6G(ECZ2fycy$ZEojos3UK8;#f#|fQ
zVoN~UBoNF(jcb|Rf&F<Ogd?Sxk7$l_=_-#sMqY$tO^;bdYg6&0lgwGy@#*p6%|;lH
z2T`cdGyfb}6@S@~l+tO>cmMUad;0$Q@XJD;$^jz-ncjas_#aRxrj45bm=iet`qT8`
z@UB0#vr=BWn4sreYo!`J2hEhSbem-e#l<Wc{iR$E0$MVf*1nIPOY_7Z<kS3LQ2l15
zQh6Z1p^2UjX+eHWd^c2aGJS@`%no9#1YQqs$H`g-pnn%Qi7`GTTwHMdI#akAHAzmI
zg2k1nN$D6Ex4_NgpayuQ=M8%rE9t7|&H7Is7?gQ2jodVSAf_mxd!&L;c(V0Nx|FBy
z<hPX0Q8_zOx+^&h?=n5o0rrvzc$|fej|EnpAb3wZL(y6|C17CFDcND4TUA=IiyXg|
zG|Zt}jep-19UO@dG>+nT4gKoppI=8#Y8b*OLCOd%%zrY-_!^}t;H#BujGZ)dMq1U!
zZ}1?Rp~46P+qY=9B<8QA_uJ%H&Pd5xsNY_1<}X$=PNvZEyoq%b*V3hSbSZF2`5^Jo
z^FNvBt3Pkz2^2mocVo7XSqLDQ@}wbs>#KMOV1F5r*Yl9`ZQLY5p&RpX1vkRUpuPz=
zv=qJNs(g(ZLMhS;*Wt!D<qO&)yp%8I^zgKKp+yGRTI4IS%|A;X8`KK_#<w^X(%J^(
z53i2^P^cTTQ3}ewedy}gs5bAD<kbDr4QwoNK#;Oy2?Mhay<4h}lvI<Ubr+2?eVyV+
zc7L=g#YLAs*=_rx%B9>}#H?5Me!26oYwD+&_O5TnZBH3AS%mj46~vA%XhR7@Lw~Od
zD~e94YSDCz3lz>VFJ<~6ZkZDIVzHL?WVXhA<7gRH+|2e6yHiRMBX2qFDW6v0pgt|c
z&AeAtMUu!EW}L36N@5igWJy_kue-<{v472{@V`tX>$g)7IBBhPwEXYMFAJ}tT5IEL
zJQTut4L-d2=f@9!eE%cG%IV|3JOot$q}9UN>2rK^<N>9E=~H^7I+ZjXC&3&kD^6tL
z&XT|~Sbly#9q1N@5@6o+$uCM~EU{vS^d*Nykn`cyQo24ImlyM0pX$Yo(0oClrhnkz
zpZqBa);|(_hVqa8Q$Ode1oijn@L@W<E1Yza{`o4|a6MuqT`G(eH8b;)k)+KEWyGE~
zXp-P?u^}_90a2@IACiA`&QJnft}2w~m5SQDL`UTl@>2fp*H@_T2#x$m`4f4B$<jGg
zpO7w!#L~#t`S16i6WB)rdRu+RV}Fi<l3+n69nn`WCY%*YOA_hGucqyJcXq6gC#fE0
zOAdB7=eypL^3B;h)|S;BN+`UWDii~$$C=i5*RqF%q?EdXJ$(vYv4s*Q-xtViS&3_z
z0v)4vAJ~%nL~R~gx07AbDE(ii{r&{~FgvfpZd0Eh&Pt_4fa4hsvHZ2Dtba}-ocuK0
zv0k_4JLg2GhOY}8tU61&#VAl*)qR5RVW<j*r=xhdrqOgdTH~9b(&jsoT6PjthAQO}
z_bzi^E5shXh+=U?&oy5uRDDF!u8&?}|4d9(XP#5kJV(aL$QGR<nM@zZA}?pns*~@_
zY&KOR({vSR!F$x|B<382dw*zka_)PUtDEj;Ltst@;x%2pUbNq2vJ2`1JKz`J!sa#O
z&y6^;t(?aarGiR|Vw3zSmg9|uh}1B+#5Ny4`x`MO6G6;Ycja8}kOrYtdS00B1!PrD
zNY44GQ68(elqcSq6{nmPPOn87Pr1T=Oj0>-4H`_7dat%40MeS51AlTcl0(FF+%2kk
zSLi6%nwri4HS=#MO(~hwP5ou(=27o90GU8$znkR#62$=4%tE<Wy#ZV^3${bAuQ%3P
zlks<SRny0Rpyq0<IVv#!v#J{SxpNi1zbis*Ym${VxCX)>JN&(#P==Sfm@(R99r28;
z<qCGu($cd~>IPaR;dDl5k7a)?6h%a}MpXJ9f9-T|s4FRUXJuK_Pf?;Wlu`kkLSvi3
zPAogGqx5D@B?jruoU6q2<{wkLD^(xIx%}8l)6#B98NCy>?zax!9~~vy$|j?HQbS|A
zpDT@jd!oLrer|`<dIY_yiz?Ak;h|T$sQ2kr#Exrw)3RCN+Gw8X@Oyu}<B+SB^r^%q
zeU&o+q#TIKzz(51zm4^65Jv^w9@+PK8@p>MKv{yGEybJ!*s7u_uG$AIGz)Vn%-uB^
zZm4qn)=Z{FYCCt|H5ssy2{12UGVM)Q_xI#}J}MsScHgMiw|v{VXI9oR<8+Voz0Jxw
zhH8&9&=fXR9*CQy?U8?2G(em1&fo<!pt<SmcX)4?x>2Fq+aY>ysawUscdWMA0gp-9
zd!ts@tH2$`S^z2AH0FO8YXJaCm+)I2Yf-&(Kjm`s#YS5!GvwS`UVW|67U4C}$rl=J
z5tP_ga|G&Mfi(Jk^%bUD9CK7ywRmq0w@^W*o$CPZIolGaRXTr!)+SptGo)l4GGA-5
zCC;i`g_%UsJ}G<5UV5%YD2hg0zo<TLCwA!HILjWf|C`wsAqC^keBo?M4SnrwO9OM=
zY|HqX&Dj<qkwBrIedu?9Wg2dperfR7osP)UFAWsz49p-^+^tkz{r3CzD1i4VR&<iQ
zX<|koBekRxY%qVqYbItWe4xXzFjuBhBv3|WMv_^bN+E=-5k23_#LUVodn71%6W>8e
zW_v{|d|yFMlLc9Mqf&s;O!{zCL+wox2u9gpCtrTqnUtY>OlNI7!x7~AeRkY3VP{f?
zvSi)y?aE0RB+=>x_x2`b6mbu3;hvO~MjJU`Uv|aUjUInVMR`Q8Roe`e%23798OiV4
z1d>uIDZYfERHsroa4SQdWJcMW80xd(K5*o;ViC`|cTVoB4k(3J|BIVmctUB#bQNhi
zIo%Gq(&1X0IidjN14jk5JJ{sQRn>y**shUG>lVUmnzeiWh!S2UWKYz!b|k30y@@vI
z8~Z}EAUjfK1W3vWUqGCsirS$X5?8&lK>>sLt@w&3zp_nF-TW2`>@vc*Z+m_#z3PuA
z=H7y9;GsP=HZcp#a+Ygku8!1IV$;36ZOp5$K&it0FROy+=a*q%Iurppme~gtm&9#6
z!hbsc^7!2aO~iitcu|Psf++0<Wi6ph4lmwa{5oLb{SanfP$HaF!|#J11`s&W+TboE
zV{Z6}DPR5T&FjCtc+Lm_{_P>?APHB@+2LJ$bc`6G`0zG8N<ytra$v-PTsb&JoQ2Rb
zSblynTmTD^g7b0GTR$j>u~dySB~wIO$$!PTcr}#Jhvo8M#`U%y%mj@H0yl*Q|KwLm
zu>KM78OlHUtA5T?0rijJ;`MOxP$cOD{^70-)pLRbQf6Q;l`!*=5yVO*MWh}RG*mb{
z*sz(#IJniACmAQs8RLe6s&JY|DsJ--mdY#Sp?vSh7r5^jjq-@*sEBfd#X4MjTYqFp
zq*1Kn?-%b9*c$-dSNFKgfeA+#RyshqL<h4D9tw@ld1#uQCMO3KH%FmxmvB%8hf@r{
zPw9RW94h0TbF6kv5b!Wm^aVZ-Bcl&{WdjW<C3nas(=?Tdat=Oki%=HAR%SpNBCIP?
zLAa5(NfJKPD2Y*JqKty9n16%c2!A|^`|%?)ehy9VO@j9CSXn`*H(A=IUO%j*j@$>q
z07uIkS$^rfCN9jE_Q<rKf9q$6eui$T&YD$bTsi4FjHnEssF1ee#%5YYp~%k<>m_`Q
z&QoX#g>pBAF(oMBvH8#+sJzQ`1X`kXcA69l-d*flWlh@SflHNjz>kN(-+x`EG4Pyb
zg!wXZ%cZ`qrHo4?I!=+3K(T37w{cduc-C}TYqXe*FH1MwrLOEJ`eA}?ti!arkqr%H
z9iH{;wF9o*b!)H%`AAL;W*ae_qA@38WMnm{t#%EQg)_r)aT0omx0==}I%Qme37S;o
zS3J$D8fjgc;<VH&=cW@<5Px4%G}fmlUpE9JRM99<MO?RL7p0NZS2;p0Xkk@aL!I(d
zZEA5<l_c4lS;Co2Y+Ng>UMW4TvfII;%V=)u5N11f9(BQa8uPl4l+EZOQL2)5Psv_x
zCbBDh6lzV)XMh^zH>*t_^m%bynXD@W2R;p~!S!Tv#k;5nZq2BbP=5;|HCxM(F?^wd
zib?nyWM=f)JyL{ab18yTO)R%PeWfz>)hCP9>QqM22K@UVcRSK6c`er?yKce!>%g|C
zVgXI}Z;D#$f_Ev$3C+QP>>2Q;f)X$l=`kZ^Gx4IVRm-it?#xJ!Y6*1NG0v1|6wUXh
zGDKFCR2xMd=fd0041ed$CFl2SE=wlL@_Ihz$Ce}HuJCJzgdK~rT52sia^_2*v7U(~
z(EFjCmHLxI)u60Za=D#B)kSUTwM14)ieJG)ya=SN{4y#iu(M?GWqrj7lVee&Qg6ni
z$z;zIRea);*c9WEXmUlc_IZa%gw0zrm#DQ-d3S^@=iKCe-+wgP8&#)>HF8iKsIpt!
z($=Y-B7!i9lEj)8(IkKf*Jl6Bg`eYtM|dwQnaPwi?v5ox>0JSVquAwW6<c~za(}Y0
zsWdu7y~VUzDrTas&6P1CRT53{(wbhfqfrnWjzn=6e60SDjN7UyDzV(>G-)Q@R3yLU
zQPTQ&@F(=Bj(@QA9$?d>q-KOW_61z3#WXc9-4zN+p<=sq4FXL$JD8O2n>t+I#yU&W
zeST*hF;4eHKegi_6G^ot8QyeKRk<ivS=(Z<SOD}R)R!=T8qse3`|U^{Q^R!{Tb)ti
zF}ys1&W$3PsgBO4Ryn+v`|z?V41SXgy?1D}U;PBq5r2M9P@ve+eqYZ~9toa(^Xz`I
z2`f1xRI(;{Ppg*M@RIhJK_fBD)O;3xZc~0pMIE2%>tsr#GUb<P^&XYq74UrbuEk7z
ziA#pqUd8)A5U_<}M9N+UH+MRhx;}Zuy{1#ztFUAJN)Ua#u5i2Wo>0^NI?<KT=&+UX
z^{9Ul_J5GkaC1%^s??9KhSwJl{i<E0p!L~7Fkh;*Qf+qBD32|$AwG^<H&Y?*mEhEV
ziS*ls`A@LdW~8!H4l5Im_tdbv#k9vcMGv5Qh~<v(e43;qPUs0uY@Qw^33C&2_U#Lv
zO`6jI%RbUzkHpz2ZK;->(3PH79A)16whA#J;(z^<O>Na#8D!%y)M1lzC0)H69DCLs
z+iR(xtuj~xb(eQ16GUC88Z*GAH1|gVZHZ#NQZ0&ySP);bD-Y4chcZnflye^*e)@$D
zFY)fXAsnU+zlWLc1_^b0sJFup2o)g+tYfwkaV@s>m;5Tys`)HLJQu^=ch`y7|7H*|
zcz@Xo+e8{c8PPU^A__&2$5ku>;lQa5%R@#p&x?SS8Od6hzpSHyi5Mzdkmu{n2#dk9
zctL%@rCyMjngvmwu8#em;jrt7G};O-TB!g0{c6=&l527eUR75emo1#9Dw&vMgi+OF
z`+xPP3&kUdQ|O_4k?gEQ{5!w38{<icuYb9T9pi>+_QKe>nVR>?#(kA?>XA=^B3!$_
zw`Lap<ki@gpUa0;tB|NQ7;1w#s~AIN-t^IJz)QR1nU$anCT@9k%xRheX=01WyOi0b
zV?o5M>&cTEt)lE<>v5kQ?wVkvTi-K)UyBI^a=o3W9C3R>cT3@xv=SdH(i-_tvwx=y
zWsT}&BGOQj+#2Ts^`Mqh>2FlVh@e`U9x87R27TNX8PxrbSv6W=IFTqW>UE3JF2sK;
z2<qbo;;-ug(}8%^A^CAN{7d9AC4oh8qEjig5&xxcp!htJ1@}Y6Z|Fo7l_6K>$4Rv5
ze6<s_Hws60Rx9x^$6Z%$#HK5jcz-i-Tj9G5QJ)P|X94w*-pyFkDO0tHtVB(jI>2v_
zYiW{O8qQd})2f8|@}KMF3z~L|;NF1zcy(aZZRBLX!y~MnM`rAN;UfuZZh$y+aJnw9
zrc+?NUHvs*8u(AWdIqY}V5v4jK<B)L;y+VG*O3WS_{R3gDBY;4EHT`)P=6a$l@!D|
z(MoEgtWq>=zcg4jDnw-oGaX?Y6`}@+=&+QUY5|P_44X(zv0arIhKZnhlKu9bD6i(W
zP}s-`hR@T6aRJewhV8KB1HV$**!C!MY{zVWx*tP>8&CH`(Zo|RRG;9p$-(K&w-DuA
zSkCM>Zg00Ui5A7#h27Srkbl4=vMjOzQ;e*%dgC+{*4D1|dT<=m_j>E7rG%mJJ<Xn{
z(xN7)dEIFCm@>qFDd<5p6MH<FX@MrR!?MxrF@r!sMGLd;%FO!1V%37Ihb4wq+X{XT
zGo1Ps$1aF$KC@64J--Io?DqWr7quNb8{Hlk9GE)Ooon_M_X}W;)PL!oR7Kf3tq8on
zk7&TYUKa|CLYRqAHc}5akUqEP=$IP~f0AeE*neDU=2g5PH%Eyw_8fEdn|4aFUcX<X
zR?)gIS(dWCB}z<`#Wr%jLXzz8@{Ju1M3j!XXIBSv`Qf+czrA|-!_`6A%U3Txynp%q
z73#^$kN9gYU%mnl>3{bLv-#zR_pjFLV%UW1Oup}hOFCTdiM>8t=RE7}1#Fyf(HZ1=
zf1!NmtvW30{e@zPYv+Ye#T9XV_>5Tg9vm{n1-Sno&1G1{r7r?klbL{-Nb|JG3?{BQ
zJ1d!0%<fHQ!XLr#vyxejWp6U8VRk078f0fO+s{l%X0sw&Pk&}L2$GrFm&|IYw}=n!
zax0?&kgLvu&q`@<8>LxK@ODbm{(mC0U@16nBpj6CjIp!y;A?1W&e8*VWA0SX)>i&>
zenZ#H>LD;^c4i*GgTeSjs2iyVU26ExJ2lis>JbRBx^ta9x#wee+L?C{IhZ!Rk@b0p
zHF;;2Lum{##$%qTmGGSq*Ii6mRe*FbvQCd1$%^`j+9BCWRtV@JeeAYmG+3nf^NeKu
zb20xNf|2&KeRmm7!t@t9n{1JN;-{1v)<Mk|`0S@a^e6u8r={~VefG01j8FX1tNP~o
ze-DM+>6g(QJQV^pHJ4FrK_r!0Jqwpdb{Qyt<y&oU6G;;O&aY_czU+OVr>na9?X*II
z<=8`FwFo%r?hryIcp<x4JL@=z+h2cA^^Dtz><HVLM3xSLQxo@e*VN1NRP{7O*|Twm
zC<__#R3$bXrAprT5~?I74&MfoqOVdkdH6Qr(`OX}zT&8263pvedZ40C9_l*!=}Vn|
zK&J`0&Z>9B(k}pUwL&p8ED@sN)e^}tr#d!d!{Vy0p{UlnT8_?eN~O*l3>Vjnn4w6&
zFlHz@R*{W&TnFF|PrVLVjQ6hBj35sBVnFm&o#c#9v91P9d@lVAY=mGw6}=HoRWp2w
zxKzbpBuKSGibi6pV_^cxuFjh=gBn|ZMa+<UB~vz1xaewyO>n)EC7Zy#k|{w<RmmJo
z2-Hs&O^8x$mN6jprq0<!*ZVRT6RCG~aVB!9N*4G!h}6{r#8hV}(ZtC8!U^JV8=07#
zDp?3(u1gdwAIqR*G09c)73WN1u4AEf@}X3b5l=2ZP|=%IqXwdwl&f_L43k}dwNAl8
z8NpM<WHJu}vdQAANZud<>uMwr`vK$(+*oH|{5Dh6)!=kkvsXvOoa@+V3UPm1GG$Pb
zvMKOD6^kj6`^lUs8Xqd+wo;^uFrsr@XVXvT64mvfmLsZTG)CZqItvntoD4H!G8bw(
z6;;B;8pg%LC{A@#Sr>HD|6WypqFfcLkL{^Gyw_Ev`mof14XM9ZKa=b4t<LVvT@$PH
z)>8X2bDc*ck$un^5^@c4ea56#VW~3|&22{fs;tJ~81sYgzwh4s_k7wtp3UaV?)Bw|
zrCz-_o&9FIC-cR}>Ef*e?znf|Kf34L4{xczcr`s)zV)eCfh7=+Rs&am*xQ1nk3QS{
zu6t}S`Rne-`I~w7)O?qM8h|NzFJ6dK=D$9|pIfOwk*o(wAbqR`9J^pOCJ|7U-F+z`
zV&Uc>*-}xU3pN2>Lx|SXh6<F(rv!XT9ycP5LfWFj_Eq6KL#tJpZ2*)6SRMF4fvtvd
z@Xq=|?;c!xi0*Nl{bYH6I-fzGf4+LI|GxXYT%KS2t?NFWE<ay>uqX51yI(JVKY#df
z`lb76vi$t>`SST}IbHlRIhopD&$<t1^AFt@f_W`<7mJhb$>Q|mr|Cua!+dtJTue@9
z%kIPQtBbSA$#1ynFE0St4_^SE@u&~RF*iJDBk&1GWfh6Nuua5&8}^p$odbA==zwb8
zm6AtuIa0_f6@Wl2A?rCsOVMxWz^MQv@fT_D1Wc6Tt$dD@wn-^+a5@KPS>!mJ{>$a8
zE_eFpw7Q;4gzG7c=lU3|kZ*~WybT=oO0=hMwGz>(Uv>Zb_kR??tri{$a5lR<J9~H2
zuw=DV3Gjn}b9Wej$_PsjZeI^IFxuIC@%kLLHC}Hp-OI@m)}9&eFaQ2^`Qz*5WH~j|
z|Kizvw!|XOlrJ2L9$n>`0zH8p>qV~U?)wXcdI8n-7or!bzd#vP`LMqTrcznGKxtA*
z6BmyjbuSn5lh@Pb+wSGlXWg6W*X6q_u6sH8G*!<pr?cgMMJ>2`jb2SJ=9i0;>BUg6
z>F<7;emtE#nSZTLjf5hiAV*(L7Pwc3`Ck7T4HrX&Syg!Owy!bu`TG)ckV0%yYEl~)
z+PGMs^4%x_2Zk%qWtHdA7mH$UBjh5C?gOetP&bK31zB#D=YT9rTKWum9?9Ft={WP;
zvmnofN62%3z@*hmOA=YF&;)pg$`Zv3Abv}xyA;LY(`TPhk0U*0G>rEWttQsW_o%Wg
z>V`v=*xmUa#7=4Le2*ZE=~$f~0-RrI?e@->TF~TV9Y6`yv8{bli53LDX?I{hVkCgz
zK{CKL@uT@ZqKYhQdPg*Jj(cfPX(9SX3ur>Oam<x}5i8<|C}p*=-(jd{g2nv(a?CX*
zxk4V-Xk~9E$!?QLQa+4H`ezuJBvRvr#t_LhsePNcNxn(aq|l_eg<bYQb~GquUBE?c
zr92&JZfE4SxHuxP2an(`aRuy19UJmy-3<P7lF-HKrw6;H@p4-T*3iACA-RQ$^;>X#
zHy!qWfOW;Sd<E7E%GZ3Pur8#o_i?=1UE4K6<a0nt1*l;}LJ0Z5mgztp$l0j7ldHB3
zu8)Q9sjlEGwv^Pg5b#}OzEO+2OadxmQ=zpt92(0k2Nt&b8X`HHb+2qtLY%XL*5xK8
zzR9zx?9|X*8#a(Fs6FIeMgj~iyl@v>yFa~uUgPJzxwiOCt}S#J*OseYTWWn-ZhV<r
z*Ou4cwyrHVPA=~MhF1~84`0BRV}R>wAJpmNE2**7xC#WbY)K_a5*WY(w}wP9_Ut~A
z5&~aD-3Ud)5)dikbqgdNsKh?Z+Qa#pSpl$gB+Pnoes;Qik15a3reEJr|G1o-E$20V
z4AR3FTt<umrT|`9?}an^$i(G-H{=PlE=p>H5+Dum4IQY&?lDk;^=_U2<j@&8ImQt9
zbGn#ctOCHhbpTZ6Dg3RL@ydubi)q%EU528fv7Ev#L&44hHab4S!#AiSopXn<Um@OL
zKYamt2>a=K)HVB+6_*v~nnX?FCiy0RNs~g8Vv|yn@(4ScE2cn#dOblP8AVsFEtuPa
z62NBfGV5kK_$s3Op%IxPqU7~p3ZJdMl?4@8-qHsOp$|w2<v0abpaR{FDMS)_ztCVp
z0w6NRV}gm#Sa-32(v#VzvuPuWI3`iR>LIH21=hedtdeEROVb!~WO1mxfeFQbQyMP<
zSl^_iRw^(`1qAy2)+_9??&z)wlj7ET<!%X6$m$81l`|0NHj3ZUlQMc?@h+t7ph_I4
z9b9R_s2x&hs6f!B(54nC$ISRb&|`%tL1G?HiIl8Q4HI;`WXqoaZjf-G2KW(+G!<^x
z_f3Lqf!`w32=!!Exz>;BNt+#i`qAsddd?M~vF;HZ3F#)?8R$t47n&(PZKGo}9!v*q
z{35S6>6ratbiC?^m5%ky&_JKmD3uzeQf^Wk$78%WMCcI8lc>Fr2h9Tc1L@fqwkXIq
z(^?c`OF)f`9?t1GquaeY3|V(pX?-pv17QOn2gRN0jbA2bD_Rp7k=A^F6zdejg1$dP
zsf((ZWXpC^WKSdjTn;e3=P)9MaVZucX@DEG((sztVH3u{e+88X64HU+ygL4NCqq)S
z;@8K4IOFHB^gr{H_X~X7N!Lpb&A;yH4h9D*A<@!a2Crz$@{P{x>WFqABopvD>@pNx
zmqqt=&oAq)qWN{uzl!F6*BM~lT-^@ZWpkfV4<jA~?Q*5l8XjmkgE>gP^<2Jv>(}45
zp3AqM%eS7(H-hkuAbcZ;UYBvYuVI^V1yM0OgikpTqW0JwMOmk>8rt?~2$sM`M*$BC
z4ezhQ!zjIu*FzI*P@h@UYOpw-nDhh}fgaQSymr#YLeK95b5I0-upUmgdVpLUf??Zx
z4kU4mXhlS#o?K%NBOYhbrmwWab3Q~H^MR*S_q79zU#O{o9?K>aU^l67C;U9T0Yt)O
zU~xw0r?27);&a<s-Me`KSkPPrSt;SsUU+!oW+m>V<6E6zMT3VDd3}NvKg&I(MMkTe
zxarUgv0aDVdd_`+4$+cT!dq2JqPqZeJf+fd{(d%}oy=!nreELxaXI-|fie4sFNj8Y
zX9W1E|Gxp&M@gW`nd^TRfmdAED+Fq-BfL|$D_i!hbvusJE<;%{^V+`k9W}<g6-=>S
z#4Xj+Endu)xMgcP9=J6fhl1&U{~j953T19&b98cLVQmVRjS?samr#W~1ed}*GZUB3
z1t%z%kA*y(e-A62{`K&W1JA;i_b8#%%vm)Us~F_S65es$QZdrb(ZHF)<D{0u@Azdt
z`r|!x;l!dYlxE7=j?=W4Ag@UKRN;Yc)Fr*OPP$S)KbWgASB1UfTf6Xm3Tg_Z+U6;g
zrs*G&iVbfzwPrN3ni-*cK@-!o!ZWCYoMvSFHFeN!f7*W7YE*?ye}}QKCv}&{xAv*s
zqCrS99VAemIpIbL>`73-V9~s~%`7d@rR+^+1|d}uCSh#V@`aHU%nLIzD%-kLTMJ6|
z*|IKkvPq6;PUuIUlNUtU&OwS~`6y-<WW}<inUuO0Bw|LNh;A12931icMxQg+fs-5m
zDz4LTf10~aR$TFcsIb4rby`7R%5|=&qhQ9iCJm^KQKqOW?=_A`l_hmlo<{XF_bBhu
zPBLXjAt|po0$Nw@vyw#0N=7qnC-*7de~v34aRk>?jbn^nk<&`r&q<h@3GtN_WsN}V
zH(i=I!Axo<liKz_!5PfZQ?)xN!6k;9D0Oste@ea>=Qe?2+*`DlkaxvS=gH)vfDdq&
z4iJ;tID@zGoDu*Rf7$ZMVig4tA`0Za=z^4N=a|aea{S1}fhNn8kTr})xZY(*%Jfa=
zDI^wbMz2<dQ_dP|;V}m^(L!3ADayvBDmEo2N6#7qOI$LRE+&;Bw2ZqgG)QHbMc!%m
ze}8)!kq|PGRyQf)2Ng7?^d{&tyzaZNA07^04!xbACPvZ>?>-;G4@w|XS&x)}OrVC}
ze;M8%-uA0@H50E3zK9BLjMgnG4ATe*%Hv(xbf&;P+MB{Jzocl*FH!NKC@2w`DLzd9
zMJP8TKTd!n;!Hr^PB?K{O>rcHdS=^Me-+m}UM53uH9wO{EX!jI^GFW8JiD;`XxSw9
zOnJ9cHYp^tnr5=89F%*x^7n@0ETi#SK*GXInI}STJ0W>uq;$_IYMN(Hy<|l2{4RCe
z6y{6%ngY1bRpli;(W19i$#Od;1f(ueiZx$S*eR{cQ}CTtl^GvbiaCYS5IR{Yf9Bp?
z5U)M#K}IbVXf^dWA3uJZv}kCNMu)l7()_QGEVI!eV(Jw()(!9uD_t9r;9!`l$TOW!
zpx@9o7vtxf`6qz9y&Z?#&SgvW$G4j}a)xnWyVZG<eUND~Q@6yFxWu^{I)whYl{-Zx
zp_mVfM*!EZNP0x3(~|+rk}?S>f1zMAg#w}QJ6&NW0W6T<>>*H6WHTQHks9?wW|0)!
zq|ps&(2v;G3DT-3H`9sUG&*VMM5~$6Z2^~Eq(UL5I;8MOfgIciCMjWVr$Pq)GQg$o
zhv-)bbJ9c2{~f}x2=%`)#)eW%3E@MCKD_z&hxgyT`x#t@5B~`iNZ>%Ve?V~loIZ`k
zrlFeQQ~I{Z_QqK~;G~CG2mBEO{4k*qjsx$80~j$SVUcO{Qy4`=cL5?3aVQ&{A~`Xb
zhDI6Uie<F$`&17WsU{Y2Poo&&E3A@f!zbtqRrm}~{avPl>eu1$emJ}x_30)4g_72C
zv!qTi6;_H`rDaHCX~}`?e<(dBm{9UA4K~}%V@U1JD5yZ7jPMra*2rqyX&I?@TZXVy
zS)mM-cR#+e`);F2A32mV9S&#A2@wz}vPzR$r@!BQ&SW1zbYH#GGDjgjYAQb*rB}>(
z9oCBV;mWBnPcO4elD<7jU0u11=a-cBf<qO~B^A{U)-4JV-VT+Be{rkFX|Hehswdo9
za<@Z0eKxJs5{^d*wnZ%~nV{^z29C6P#Px3Li0gfQn2cY|VKV)$GSzq+(5!nN=h>;g
zh-S$(MYU3z|7^D!s2d^qOb7ji;A-uiepkM#f5Jnx#-^K%Ocs9D)d{l{aq<2Acw?~V
z@<(Z;DQvgffYQ@de@juMHpOJC2JO+`0Y!Uel&w$Q^Xr_`k;HVvk;=pARww*AUVBY<
zJDEo9JRS$AA;sT}<x1gcssi}DbvyBO91BId`yEGgvtqdw$SxG+@e}MPngIJO_f(cY
z7qFEk9E^ICEuKqsJHB-`x&p=#C^`NVFs_|3N+g&r(az&(f8+U860Xk#1n;El_y5TA
z+V$I6$wGP3apjlsOm<s=<<gwZOn2(4PgdjEEb2~;dNUdRs)|P0^`;qnourau*R55f
zk)>ogC8mXBIc2_-EPor?S-GS=RyFNb)-M&Xa4|wwF0gP{h@9zWO_p5^7Qlp&6R&Xe
z?>I$MxST~~e|4v9cXd$6LW@SM9=f-?A}uwi+wC2MXe4@~XYNphL$@@t=d#%AT<t<U
zTMGiEF~h0j<r{n?CG+YP;MX#GV*ie~XO_MieGN4yzYBdW(MzGP<uXIxW?N-KXB(YM
z8!=XreV?TTX)54H6RVOc#^loWNgv;v>G3BVqBj>$f9S;sqXDqLb1P<6A|yjR-M1v7
zgi@UfUaA{|0EECOU#S~|Qj1CT^eU@w$VAo0KfksiqcGw-`8NA8mI*66f@f{Vqy|eT
z(MmT42R-7wdS9v;liMr1BnyQWEYeH%+N~Jx82M)^?$=XdxTmI;*Zx+Fo3&!>-=!5}
z9Ck}zf7ywti1#*PYLb0@m_;U=Z5T^T8C!Q@x(V<<Y5il`G_?>_>tp%CzI177BD&Jo
zQOvu>8e`jHMKI5qg`c#7ak0(DYi4b`n?E<ZN1^xQt!ZKMsOAv~rUea-nDc&b??__y
zU=?#tO`Rhp%(c!DAVY{q)-k=SZ6rkucR+>DM}9qWSn0*!K&udqYl>~t5{gmjd_(ca
zpW7`dgM)EHv7;e{G{NC5ou04sqpXj26&<>-w4<CyzHdR4+EIPcfTHC8%F5hfmtkN!
z6aqOpmjP8R6qkWAI4GC2jy#(ImA6ojJo*EFf`LUmBZW^1n-4|#85|x*24%hck2pM=
z1BVai@pT~;BQ<-!6kj(OJ!14Sgx!Fx!6{1#h0T`d4k@!{=Gp1zkaO%ewq~tSa}$(9
zCDTYeCpAa;C1NKwEl_hL0v9JcxD++lRvF<VZDSpk#7WR!oSRb=)Nyn^=+^UnsxeW2
zU^iNxmxtfp<fJmo#r<WuIC;2xnTdxrCY}f>m)`#M&l|Mb<6!Yf-4=CE>5*wHu+D@)
z$P^ii5wO&*;dTo&9Ip!1M%@_ikKp0~E{J9<y1~RF;bL`_zfa@gC{|6SaD#=T)CwWa
zS-8Z(QIqM)8)F6@=GpNO);T1YMOr+6#@+S5>kL-}H<&EMb%93*rR%sB#=fBnAWfCo
z2@pyNI@H_QZd-WBh(8|M2Hv$oC%GUX`nZu#c0vAk-Mpp+U`E%%X1&Y7+O?@*efvzP
zhV4m4ajtW!u&lQ@*pR{z8w;#AI8tNUQRU$E_69Su0yW@T?{6@mFUHq7<H+t#28jo`
zjTh@~kij;bzx1tJYx%}b2BE(3_FSG=OCQ<EXoN1r6B{jCos23t&A)-qY*_?@jO#An
zvT#+W??R}nm7Lg|7)H>s{{RWCnN^dqIurshFqa|RJ`|Tdl{~})K`y_y1eQF}1_LJS
z!k5uZ3KIe`GM6z{3lx`LCod_NDxEx^f5c<;+vWEsG8-#Ju<<gU%3sefPyRgFn69*7
z8!F3AkfiF(^+|kSh=p~bQc46e)rWQU?&PO()K*t<PJ!}-KuK$vdx26`F#F0{EwpoO
zprX?Tjv1U84m07}HO@0P*YSD0R^4}e6V&)yQ#`L?4C8C~H+VJ1G!1v2hdj;0e?D)9
zkukoArIk%Ag@M(ZKOf###hq6#Hepm=i{Pkqj4&d7+sn*GB3R)7r$S=XMmyUk9fM6g
zu!*+X1Aq#=RGQNXA;)W%%`=UO>Ast9nu8WH`<mt)Fgid@oK}Xn*p+{84`sG>8;~<<
zhts>ABgzRy`l~=~a$~F>DW`;$f3`F@B@7xEg~eQC<77Tl4x(*qQ~YJlk}lkFQ9d-5
zCE1H?<>@^T%;`p!S1S6H!jq+NFO8Bc@Oc!I^_=%%HZo^Ov+xeEktj=&vtMVDdv8Cv
z07e2*tE}mz(XcSy^`z@xVb7w8rB&Odg3y-wQg=@mOR~29H;axa1YV^df9RuWeb$PA
zs@apey7_srWLK}>+}^zXaj{gs{*eB@UB|Pl&lgLl>f4+3q(O>3DnghU0cDlYp&#}7
z?X)6`8SBER>tnD*5jCMmZ>P?rxiYCY5K_!@9Sn>VACh4$gOC$`Zr7&?Z?i5_{1O^v
zsBfE$UhkLESV`qJkJYpZf1VKP0@c~&>C|e3nAwcFjMaUjgt7wyQFJ>cG|l^5CXLPY
z#-u7jx!$-=#hWhCC>{dK#N#Q^g~4ep^G%6PM>e!6%{vPtYb!L57sip`<-*vPMk_Sw
z0kIP3cnYCrvgSn=^~S5X_~Y;zh)+$y;O?sl(_-yGcnU3>&Nz0zfAlGNE&{Xzy_mQd
zYh#65k%~Y~%NTc>H5Vc51`yt~al2q_f)RxN=+VHqFgO{C9zeA{K)cTSy<lu?5MJ5?
zVceJA6UK$mkzibC8Nir2GhR+YRSJ{|?V($vHMTh0AXQ`8)+e3rdxNqVu;6;NXIPB4
zCMpwNqgasBU4J&Ie?H6j)VUk)tl}+JV{6!U?D1ZVK_7$(ulF1%zp@u8842lmGX*GX
zZ(G;%Jk0s78MG}stG3rYU<MgY2@?HKo?h6t>yA~1&9la~)Vlk7rZku4Ywtkl^#jp-
z9jijvpkt7VL-Ww2P<<h$?}bvf@NI(VBV!`?PzZcZp5HL{f5M<Kmd^sjxiscjEWhFH
z`&ovL)GJ+D_Rt((gCb$nuLiR_R$nKy)}_(^8=eo}ur|8`d45-Vf1clyn#J==ef#nJ
zp45N@eg_g5>YI1*uES%T>#Mtui>2m$Ilo0LVZyv<C^RY#oVX{ogY!Ehf>NVl7=bXR
zy$Y&G^*qk+e^EI6Ob0dStCQ-GbZfH0OgK8CP@!cEc1OPVB>f(xcD~Er2wl1y0p=6H
zST!2}ZbzOCK{?vpG54n_RNOH!m@LQI07yGP&OWaCT^Rj|n*ee14QV`(J@uvcWKV_A
zk=RqAWhZ-@mFrm>0m>*m49DLHP%0MH#|G!V(d}jce;u|N0Gos*e<+-L()+@>FLW$8
z_qBZC%>Z&1t3&vN%>ZHs(Xn6gW`JsCRW{EzLz-^05x>u8pyu!Gn`eB`w2132>oPva
zV)#T}IB<c7VfduLCjqE-dJO`6Xmp#zOi=f9KF#N|!>Bwaq79*RV(O#4Lsda{Rl5Bd
z$-HEte~cd3G(wtVHx-6yI1-=dxc#m6wT*3RcuU(Jn$dgE1MQEF(CL*?OUxlv*s?!Q
zcLSd8AB?B(`WBDkroy40;_roLe2MP}{goNpKRXDz6+Y-v-G;F7Rv`!c=$?c1^hZ_l
zD6*!n=EeH<dVSZhyQ?=pz*>^^`t?mDdek9$e`D-e31Zg|@;q`JP~rxj%U$Hy3pTzd
zwFz|e<mog$MgQlenU`Iyd|Ie_*_~FSZTgnnmH1{_hspo_tRwMpS_hNxW4J+!yHzBv
zxLo9HN1lLd6x*2y)z+gP^9+Q>9a+Qu%)9~;>-F!*tE^w&UH!dSYFn?LFP7}<n~dDs
zIzK+9+vWDe$R`cw!P9h!TGM{GTuqs10!oA4nxgGdV5uPxiqI*Z%qx=?f&18L@L4uN
zXtMqTMbnv=v8E*x12Hl%m*FHL69O|Zmr>LPCzna2JgI;2y1D-G&39*VH1_h|hE+>R
z3+v*}c_U!YWGQR4&}4CU3(W?9n*c3_t+XnAYW#4O^<3TuRlUD;LP<M&;)W&DU3gXe
ze`Cfw`|-LRuaB#oKeNdjgI#ym?4-evn}sp!VIeJrbW|E+-MCOaz9`*%P{slIOXkXX
zY1)e5%?^J77CyT1Znx4Fj^A*MvBCsjNCO_#6(~!>t5V3^v3HVMK{f>1Nn!8`+Ung$
z7;=2|b;UGCLOWe*nbF4xQ+W*S7{e((Y!PFoj4e&LQYmiXG>{^#ObD|<{#p^^ak#sk
zAi!2IQvDH9Ih<W+wIK>?|4-0b7{hM0_L}HwVaR{Hye3+$vx1C&CR%f8IHm`!h0s?*
zYoVn?>oPvG%h-~bu9q<zP|9R*$EY3RmVkSN+Jo;lyfAs*M4{z#c_NpV6uUg13w<rm
z=UR?DuMVzD?i=C15`{58dkMjk5iuU%IHiGI+;ypi=JjtnZn`t5Mmlq#8r<P_+yj-r
zS}=bK_HK1m)i+8EJQSy$-ez|kSj;ZC^FR^Pr+0&QroT+~R~+wtpI(FBn~V-F@4eqe
z5gqZOmhg%~2D3BPi|A9Ovi}$O{5(m4f7YY`uZYlOQont3exaP(e?G4&^Z-|gM+()X
z(5_2)p-@O=QdGAwIAM#0y?XfY{^8Td^9z6S`*%P6752)wf56-R?c?3~g;)Czj}KKt
zC<N?@wBj#b`f?qsMrj#EAX_-Dsu3X;6m6-I%p?_7r6`D?;4p86lzx&jb6l~#<90gO
zOv*^%I5-?6O6cD|))i?dd~iGCus7?pOYQ$z*8&k2fga-osl*%mcN1#Dx(N0_<#T_L
zZyUb0N{}lY_qlGF4MKxY3<^Z!3(Gdv?@%NulRNrP^~Ja-oaERbiwq?Zby>Fz31E%R
zWgh#WN(!aD-iGRWqRvoVipvI6VWiik#sgGo3#yVqx0L#=ZW|CJvEyVSf39o6J{n!2
z<*B|Ls}YE>y_^j|7-1p+p|LC|?96{W?}WxyWV8l7x~|Lv6b!Ejb)p$J!B*x$L}N2I
zh%Nchm3hE5<pQm|1Mx;vxUG?Q5#1*w6tY~0uC$9jW$Mz)^|Uh*E8(qf0FUWPJEOE9
z^0~}ob@_&*;GJr@`f+Eh!_8>aY$#w$xY=qd7y}EPJ|#^jV&FhI^K4?s#8rP6F%%l-
zOxjRH03G6?BWf}gj%$;R4M_`Skrq;-HcUd_H+DTj%@?322upq$<m0bMd?j`z4hTo1
zaL+`VTb36NLCjX%NFC6PckCn*l5}C&x?dd=E*<$*HHbi==5vrj#pqn;t_7$d2$@Xs
z?8`T2|2;d3G+JGS!u+TAXFq@ZD0e^O!|%|DATYbn!&l#<%^S|C<nGtq-_O2Jhj81B
z?*X>K2-UT%fbe1EQlPP^I0z7B5>GjBxC%w#k|Do~4j7FQ1#Fo!4LU$_8n1GxPCDS#
z2m5R~Sc-SjL9Vf*q~$sRy_oQ%dN)?hw?Y7%1GGYKNz1(-M^|-Lv?zZrP%mnO92w7h
zn-wj4-R2@qdVmK!8^-0i?V|^5dMHa-G-m9OCm?ENh~gby!a^zuhz3~!@;wPS-4sPT
z8tuLeL}_eoF4D?Dh^Ru#Q=@W1fLY5DNsXF4=%Mar?sV4ov0qdoFsm(P>czv5^t{h3
zGo!D1P)eY}Im)R9U7&wzh@^6<PP(8)+4&5*Sc><x6@|tFU2NneK@kn_(RStiL@+db
zm_+_jSHf+!h7t*<Ol?^!@s;5HdMPpIy;^sGF#;{-a{c4sFQ2FB>5;+4s*w4XYYJ`9
zL7Lfn_0tpW^^o<NyrA*9P>X5;3LVWpv869^u5o@i*Vcr<gz$fbxMsSk;w`+diS!1G
zh`C6EYl2BEuoP)<4M;$<mcsqI-bA2y)g|ORd4RYSDRZSo1Bf=cQ<8F`0SaiuWPC0S
zWa51^P-vW&R;K|%3F;l4!UGhY@Lr$71B7nbT5|#qWD<QmuynvPct8S>*7fqhm{-Id
zx{0UcRVLDxSDAmv(>8Sw`h+h<93#;Y*voi(3el0rW*t=DamC02h4cGZAVd5qEU=X-
zbrv`wCubJ0KqjeYv%pfklLc~(y)1w(z<P5M3!n?IQ)$%60)`)qxl>qRDe-N?Y51-Q
zD=x1ee>woq<Wj9#+z3J|nTfP&afs+b%TwbpA`VCG<#K--OvtD0D!1(fDg=EP2@Ihy
zS1$BYVf1LB7(-G7*X2S!attQsB6SxU1rb$fS);<q8gy(9p7<hUXe8q2V6Q>d7rLsx
z)OR2feA*JJhYSr0KN`G}yT3t($P_X~;`foE;pRqdQpQH?_}i$3PT;1Q4bebI<(|!k
zrFbK?721ExkW0;vV);quH;rd0*n72GF}kW$r{{~ARD(}IL7oaj8+^jxj=4mOPE<;Z
zVl|WJvAwNIA?c*gq=qN<*Y!e2p$@Q{x;8d2R1M1Kx*CtNH8x}$iV(d%zwrBqUlC!s
zfBWg(|IRP8+dsTKzp!EdI6mY4@SmsQGC$iUezt$zmoG7Z#g7ot$QR2Ett~u1fv7%g
zK@dHR<IS@c^TfmR<}K2U#A3Wp&NC+E490Me>skm%BR6@=IGEaiJ98)>E-fYMXeU1X
z54jGGpVIs5x+?DkgfhXp=P4+VTj6+lKaZ;196~URe%@}3p^W?OG=Rc$Y3gbk7v-SM
z!Qg-4mb>lYmsxUiL~FiNHmH@P4#QgWx7>1GYZ$0aV^2VW4!l8Hhm4Y-7fEUKY4wxZ
z_|u496{S!rWYeU_qqD0sdE=O{LJj1jSq@Om>%UFIRKx0g8mgLBKaV<%X)t+x7|zUw
ziPG5N&0<I}o5dWBo@O`AG%1#@m9w5{0y2NyavXxx(-5RS3?EI_(fU|Z4V^=E7gXVw
zoBNpzc=FSlt2CX*uE|0i9s8NzUKM|371SI*^aeIPbm9-V0CU{Fl89)wKR)b!LD5v)
zwHW=A%aQN_Ez<EZc7_&o9yUN*m>v~1CQDCq>h+X8_GJ>4JctTpI61n-$lc+e0r7v?
zmDdMclJ?l5bYZKy-yAMQH8<1{#o|^Ye;D#R#Uc+RC|_En)~(^Qdap`ry5N4YWt!uP
zCZ)?Vgi`o&75r;}fgVev#=k2Gw2sOqVM!9sxUYu6`cyN4Sy5^gMhp><V=~{5N@T(I
zZhimot0R!9Oc)FvqYDG4{cXA19QJ>%^qEZDeY6XW*lEmHqq@_2d%SvU3NsD)E`&RV
zFnR129pX}m%LSXNf4~3z#fuk>SxhY?9jR@NVn*SBqNPx45)&1n51GXJB<9;kp2aLD
zF*k~yV21~&rO#BOr2y`-DD8i4G-ZA`{@L)+#s;Ri<<>bEvQ4$sS`fE_Cjx(hCto~j
zmZi||&9O=i?m+z*jwX3F&XIKbWz#~(*i?KPVb^ZZtUC>I`~a1Tg|!}E`QcP#De*U$
zr^Eh3YKk8id*RR-)z1Pn&vel-)-LNu7Q=acJOOQnb6T6z(8ebf!5^V*q;TV_NHb_+
zwG5{rjfY)T$TW`OcMZgu5CnhM)l0<HLusNG=(N`m2Y#EVu6jmg75XDi$A!{*96<Jf
zdEA64tqYl9;AkoI?(vxElw_vsV>t)p1y84i$7wT6687WH@sy0F({SgQ3#2b_hc0qC
z05$*4%?TZ7Zjx3TOn=m19ya=^vp;GU-J#Xu3TpXllA#bdw1ntc97{Xjj%CInLFSwJ
z%!~t9Pl{#H=hlDxm>!;oV(qK7lso!)(ROo0!zgZkQ`6bZrRHz-_e*Gw{s7szHII|A
zIurpgm*Ltb6qj&@Cn%R^usjt47)_Ucuskb&bpCjHy=XHz;hCJ8K7N@94<!DQ#GvKL
z5echl`uIHk^o^9t;&aBwNU1eG4!=FV`Sha6V+LUcVU+0f;tXL#V?<+)Vth+{6n?uh
zb}8#E)_dz^+~FOcfY15#^T+p3vM~F4VZX?!(t>T6tTO`qR8AvR1|`Ps49W)o8xJZk
zCUVJ?pBk4&Bc0car_KG@itG03EiTNPM4{quWAesPDQR367V2EiN=WRy7vgW1SFt=P
ze=K=W_li0U3*+6eTB@)q0I?E|#zj#HOqr}u?7f<#F!*@QdioWN+>ZXR!J1V<TRl>l
zeJ=VH^`_%ux}>sd7JF+_RlSK?a~*a~86b$68FD95(F@b=k!t10MgMd+{u#}A$VWMl
zU~m7wPC76V)yaHPd56NlEY_}<vh@Nff8q<hv|oMDONT{Y$x8<+qn9q)W%j}{_HDMK
zm7P+q%2^S!sutq13M;C8h^*BN9i>}YLvamTBC#a0ILCzY+%!Ik-zNUPoT^uq0NE2=
z83zH>$k2mLtdORIw7X){)tuF&DHYTRiCl$6`6MC?1r)q!_>sl84N_%J7oYslR76%y
z7Rrm@sH-No1#>Yr3pilpHFB<%Z|41+WleR-Wto$l?&{8_m^D)>$+=5C*}QK}vY<W+
z*fzP_R72*mG1un|n76EH{U*`vW;tz0gg}GrZ<o@tJYj!)%~it4>G~%*Za306tWpLe
zp=XaStEEF}2h-?5oHds1{rp&jHxK()3y-`w5=xG$J><c6^QB1(hvasIFMxFcV+?$i
zyn4e^`F|r^mVIy!uNVW4rTcQ1+b-wcikJciDoMk##O9??tT>P=d_KJ>QYlr$U(tg*
z5EE*ZI;DT0Py6|@Uz@%&W-L3TRb`iqJ|9+cG+7~?A7y<U_Dw;YtJI-NYY2&kXi@o9
z$?9R^8x1OC_r%v+M0Y(7jSi)4141guisMCeZqWD*7CWR81}w1ZMkOUeZ(;*U8OZ0_
zG~}Y1df4o9|ARl~9ny%coqIx5zDEB%+CcRT<%@r(a8=zr^S#^K)c<^L9mqz$&gIDv
ze``9B9{nz@<;^)4B%4%Pe4{DT+p5k>=(3aZeaHlM)LJ^>aC7B(y;wB`<UD0f7mZR2
zWhV`Y&nEYM&;R;VR`BwQ0N>z#$)Avs%PJIoBZ4R0M|}hWsEp7NU|;TdpY=8&bZHJl
z+5vx>b9S%if21WH0?{f&ppfQS%mV`X>1R3p1rNWE;4=8>-)ZV+@GeZiufI<}0*xe`
zv7(l`XDj4?s)T+*S#{cX&_rtkcCiB4g!2OWw5fWHUR&Qq!a%dibTrg94HKY%`-m_L
ziB6~`QVq3Iy-EydgFJ9;r{_E{$8PoTA>x0y^YtcI#E!B?aZ1BAg8JC4Cl1SQ6q|iU
zt#7$|4x3t2NNA*~U-H|gvnp60IJaAYpx~-%x3~!Qv^=mTH+7Ac5UlA@*Y$ucY4<$m
z-D-i7S9f!P70w~D(KOj=`kiC|AQ#1MS2Vd`Ex2?RyIu9$rx(r7zyH)GSEOBGErfp#
zB#h$T4lCX(8F@xy2@5VY>De5Sp(AowhHlkPU?Hbyo|MOS<(zsqLulbC+(iUQg-Q2R
z_Hu+qzSeh9L0Yl!>5hutrou5m>}bnq^xqxIF3Q0fXs)pHk%a4zM(6Q!PI>DxOg+Hp
zFhPNNH7Bhe1#U)=(uknw6%DVm!>)gw^Ppu!h91PLkTx^e9<(-MRj{+eRDc3yufRF1
zm+h+p`cmLN<%~cw?4xMqv_TrcdUYd51uIad`ldfTi;#-nM^HBxh-1=4KtX<4xD`&F
z6UyCcyf0IxK)%cAWp0ArTBmWhPQ#0CwsKCk0wiOC4@U14(H;&*O%;vJ6ik29GE(O?
z<=Aq!7)101ehB_)=@>GE*Ffjpk^`9(O;fD7@%Vub<#+|47m!IQ%-u=Jq&+YsFT9EK
z;wgyQ0){97%9{Zhu2hS<X@Yy}pcJ|=E=>dY*dU$2a!+{~EzDb@Of!w8ZMA@CHCUvt
zu_{Ci!V?9BVyutoUxgria9DqWY)7uGG3}ICekn-sJQll6@97w6`-)|ypn10g_dDed
zok+YVzC+-sd8SAOxkiRVR5Zb^cd1pQcz7FnDmz8hkkCoZ4V4x9I*Yn%?y|23a@Ky+
zz7E*Ijxb}}zAYn%?b|LoXy10p9x_Aot|Ao#qdtbrlyJe^MMQgHyzhUA>;W^B0`@(T
z73(svWx1j|YF(>^F|=u2w(0idNV8t8+cI+4y6vKa)@_#@v@Qb-Xx&S*!dEqKFAd8b
zr?6;O!ac@_0nDs>-e{Y**!d~lQ+eGs4wC>BfXO{tdR{MDaZ72_JcW4=sakc<mR$o~
zZU<O&L4wZ{;Ia1RNDF^HyI{Lyw+F8-*e;UCT(DhotcOY8-I;GW3N4Ms>BUIXUL;a;
z2RS3sd|@YnOpUVj`Yzpv#`QcLvgDD3q|X({&hu_Y+9Sy_XZIj}UfUhVx4bkjFxAn5
zV9Uy}J18>G$C~xpy>tv^4_}Uu&*g~LT+hNQPA~fG&+_Yf(`kRFyw*}i%8_ecKilDL
z5wlZ#^^ABvj9GmNV0cAl*}uplS^;m{U3Z$u6xfK)fsseQ58~oQi5;mt@MVFmQ!mRj
zQjQ^@%zXEyMpG(fuBLq0D-%AVUf$Kh7hT}g+nej#B42iaW5AsDHwrTcUr310vU2rs
z7Lip7bwBJr%*TH>Pt=||)up-eu}xoWt}ed!GEs?ER2IwOK9$UkH#Z6J?Rt(>SXixQ
zQl-M%kls0@P|~Aux?%QeiS6@i76y%3OKt}gYLs_$-vLzZaC@=BV7%^d1AkQ2u>F>y
zk1wXEulN3|Nlow9x~h-8pc{zX`GRZx2KxFhNJTC}3T2l>{W}o@GB7rmVR$bTmwqGy
zCztiPJfVLf-tWJ8{_I9(W6S+yp*F%h9;{;lJF<n3(o|9(9X)mD$40PsjWp(#q!&>!
zN;qnT7zP^})e2?2SA!Ic%WI>DpJB*+^z#+dtP<MlR?7*x2oK@@0L_5TU<}BPp8=~e
zqeW~-nK;fIEgapKo~Qfh8TJiU#&-=;S{NG!%Rzs5;*&w{o>X*sdVADbEOKxryEmiG
zsE=L##%6dIW?mF&5`oS^H<{^dMrJZ&S8P~!qIUvgt(J(FG52Z&CTlp^4b1;ifi(+5
zEEwzk-Y~4Br3#zEuok*j7}i>5VR$@46t9>O`0?x@^L8Awvb&lWr<h?ifrJZtS9dLq
zK-zx-Ya?_eSQ{-ZtlH$2>gq_0F&-0Y`g~f=d0geS;O<Kh$A8uIF~83X&WIq*UN2PA
zTfI;VUCRr#mYEmm4BZG{&M?Iz(W+48wiC3%c$7HUqK)+ycN?S{E$x2u`6Q?0JZKDx
zFXoi3GsThFv214!ZOgYoj$^78@5%PZ95H`-{9gc<Kxn@`<$1bu=x`XH=Nxe~%eZ5t
zJxtOh)e)DBE@g>}Bge54=n2+iF>QS11ldRsiij!eB3k{lEB|YUGA4b4&BHNL5OTS&
z8GbV=h>FIk=im!AVjXXvHvheSzWMjeha6Zac>C|~ZsHeAMpzdHrBHi#`1!lx`<uVD
zZ*5k8r<~HY6wlIHW9rIjg+cWO5~ywG5<LTVFS50tb556$y@h;#$oI!{t`{ta^Pd#@
z)yO2|v3LcuX(vF(#lou`2_<J!?gzwqdA`6V03m`8Gn=SukCsL#)-!uHt%aA{#!k4H
zTesPUoGMw(Y>PISj$QfxmBrPM*h|EYEep1PSB=iqrLPnVBXwg((kb&&FLb6b>Y^Xn
zL6LfM&Ye(&k)$|emU|qpWaq=#z5H;Jo$|{pg|U=%k_A09M!uePnJK+;<-JoTaH>b9
zwVe6Y>L+WDB5hiSc1X-V2PcQF5SK;rPy!7BZ(8<{6{(WvM~eD-oWf|K{Y>|aFvm%M
zKjw+5Gp{(+-lx0>kGl{+3PM`s)`|<=3!{F}GQYEG2w3ULa*y9lgb)9@9oHYYM^6F}
z1yf6WjkR2ad{-#On8TvgT39u*p~GK`2pSX-oc47&MC%+gH>1Rp38TpSBr<yh5kq$D
z0uMjtw>sZ{t4qY868VIRC4G#76w*|G03Uu^w>>yPpoNTFm8$&IFzUjpq=G?>7f(T@
z%brFbUPX~QK%aK0o))LbRVx)RaA2*hu}E|~F_;2yNjJ$iAXgw~;2s&#WMUO_I=xzX
zVVNRx`sy1w`~zQp7D;pP!*A))cY_Cs3i$NX@C8^@;zvv7BJl_0Eft%Ku>c-_e!XC>
z7sf?Dpiv`V`(Xh}M5re>gR+Fk(u!3=S=D;4PFZU)rc6L9>8lv~-@5m(q^#Hd80#-x
z<)m<Sap^kO74qO{NzYkGu7lyNP_s~$$8`YEORE^dn-h@=fK0tG5uqDQ4DSVr2#H@f
zvzmxbk8DpwI#xONvP7g~rdRrZ1wo=2hNX1Z36h1@idg`~x%aAyn;6?#*l3F&@Z5OE
z$W@o3A;H7G5t4Ep6{Pea-AhpM$RI&2ES<g*#5Bp;vX5`?$l42~K%*4c$Y!?d;?%R)
zo~~b4b#V&rg?aVe#Dp+Rkz2WmR^7x1cO|J3QtMMfy|A^w+uJ+i4*&gs*FWEX_x_h3
zzy0a<PRYa5?VVEx46TN!k!Un2G$ES%rb|2E5nbkUd)?oVxzp@to8Ox@zGHA1ra!Vl
zG6I>RH6%5NKEJ&1)>D6RkzHyu_nP+Zjn~o}k7yd9D{UV(T3)aJiPS&&Y?1x12ddUc
z#Nh)R)o8Fs3hGY~v9%$8<upg_JI#rVBG|nVTT5>|&1r<L6tRt#-l3u~9mve9pX?}S
zg>-vQc53O3X9|tbwIHswT&e$mjq?R%2-D!-alW9yYcs9=lO8Y7q*}@JLxYr7_2dX`
zDt_E|q+ks?0$amztrG<bELesQbfN(4M2w>1;u)TMz>LWw`Js1zim>PO01Y=e-5R@g
zdca_s+O`s}czU25T1;{?oBBKS=#s(7X=cxrP7k7U!G5>drZpPD*+f40;X&L7v|7Dx
zB(8G{ie0;IHE%sH7=I8oJuk>z@P_k(EW7)>;3C6`ck*H9mhV*h@XLig-&kZ+Ei&C}
z=_^Xb)VESvukX-*T8~T?C#g~Na<fRQO)Ok4qKj)iW_{a$P~utEARfM4*W-a4&-AQs
z4HDRzr^Z(cTZ8=_Ng0G@3%|uZS9b!RSEr}YqXR-nkx*r$t9Th@t}CL2t0khPw7zer
zJ-7<jdJ2rzc6Tn>35)ifm_f+$gznd8LjhV0@(3`nu4uVtBV>U~vtm%WK$U1UEx)uM
ztCg97Zq~OXtpj1B**bCSJFM|c!Zr)C1n+u}ND@J8uC|xq;ahfpK(?+G|IV!oWPYu9
z@2~o%!W(;(TIO26>8q41{_1I~uP*<<tlU9p{|}@V73h~?U^)~6IX06qO(X*`Fgcel
zQ$I0(&0B469Jdkv?q9J$zoZ|`B{?r7iy%N=>NJ3Z8g>e_H3UPJFOE|APL<A%^Y8l%
zcaL}H)5UsOU1iA9>>ih#of*#a49N?jsC2#yp|JgqRnji9R4H|oQk8NizEr7n%C(*N
zE+wr}=!(x(!Y7_X6{!oc@FNqs3mj`lf{sOh>Qjman(JLfgB0sy3O$lthhm-A9z)5#
zi(ZFep<YO}fut^yA5_Rtno1q5bWu`$DMf?ihkntZ44P@_$PYcB8AV@pNTDMd8l>0}
zj}53J9UJ71GMkt%Fy*W^T0&<fR?{Sbm}(EjcN}sxV$qIcsUo^$(%M-L9gD9@=Gd`+
z5>=5p&Z)M|42us&ML!Za*l1Zhp<L}yVkfv=MS{q+ZNXzwst&nyqJ##~!$y)dm<S)s
zs3dHUR~rV-oyNhE1$UXrMk}~0gNqB?o`SEP`A)Ni0L0L#y|#(5g9lR;sY@h7TkaC)
zp>0OE#9YN<kyG2E%antEWMY6=qsDuGfmmw4h>GlUy&JX5p;R$Ym!s4MFhPbhswiES
zIJ_p>u7*fmF21T0G6I!?jTWgZ5zaG_yFy_YLLgdrLi7+zZ9uU@tW`@A8sP0~XQe}=
z3~f^f*VG~KhNTQjT2#Q^YQvcyJRz{yBbIHBR5dA>EVVxiEw8!WH|zOShBhXD=b=k7
zXV`EiTOfsDR5ItsQVmTOj2zj9t+QISzD14g>}x;645q3)*E?D*$fim3bzHWm2rtaO
z*nl*4V-IntL#cT{2iw|=iwKcNFNbX6yXRvzSxKo$xk+UuNqYj>WF^HWQIohymbsoE
zJ^J#?{-@v1XZ_dn`C{3>c>j8TX;)8A=l|*YZx$DCW*0B5Z-3PPt^c9__GK6@p3hE}
zFJWny;a>IGDI%Se9KuyYb`tYZ|8<AnU-aKEep>Y3bzcO5kzr2Qi7`nM$0&dP3jgk;
zL|}F>ghX`-c>u}RCA_c@yUe_+L<-2t&@@pn9lWHw>wwAR5N<0SD0_#0X}4F07$efN
zBIRQC1j0Ozli7i#IGfcmXNOUc9PO{{+Q9=SkU$b;N;#9n$P*Z%WM>m{r3Izn2;SJQ
z1%f`o4w5ob4QWg?ohg`{W$2jLVID)NXv*xmBO#eb{2+~aAP5k20!w6e(K6PkM6<*m
zB~sQLOCZjwqOhUnE0nl@dR*j=F09@?Mkg4LLvh9OCdMmBtBlxWlA}jk@6mtVx<6T-
zF6L0^AJ2cV|6lyFT%KS4rSIRKE`NFd+MO)k^}oJ<cmDXz>_h+K@$#3ypD%xyFJ~7&
zAD_(JuV?-1v&HNFLvYc%&|hAh^xs^ZzIi*l?7v;iFP9g`r}Jfh|9ZT2xj1>XT)dhu
z<|m8!huOzh|9yY_2A%xrDbnrZr@(r=`=dKY0W<+vMrCK%6;+(~sfI%mIq0lJVvOrR
zOmI;MhmFHQ1Cr#x758cIhtu&pgC?!;(*N^c|L#D8jCIc6pPe0D-2&OgWuP1zAOKj`
zO*=%z+2Z2GIfAEu8_=nLcDzJ5&AS1%`oDf$e*a>5yqt9dqV-P}^Cbp(QprHBK-NEz
z0lJ5aivCQ)g;DPk#!rm=mgvU|c*1zta3S_d7@t9u7&E}dS6}tdE*2**X3LlTv+tht
zKg~Wak2ZYq?D*}>LVG!zFE1-St&hRzmy7ooC$r0u)7rg%A7^h)kH1-btX@)pe=6|^
zJzU^fV-+-LS>1a%685S>WdqVUfHV#ujRQ#I0NOs)_Nlf{Q)39JF@)3@LZZ8~fmNs@
z(Io`ID%iov9N5!=qkU9jm38+ialNd2OGAJifhjgS1a>OBtO)g>f;1<TS|uPugD*~K
zhd~GQiD|!oSB2|kzUyNT%;?0*hY@>3fPtDH2KE@#82A8#D&nNaPdQHjH=s9oNpRo{
zbTI~mcLr4lxpM0DKxCLD)X@=8&ejDan#Xai3N6Zx7J?u_)teYKcS<?xBqY!W7f@3X
z8{n2Y11zf|wxJANfjX!f6JjuCNvJhB^AvS~V(ROE0QXk+L^vcMA;%2I#4EH=WFl04
z#8EUBijpT(0!Ua*Km;VTH5gP1aknx^M#(j1B_t*LRpd@HF(40+6~E(53<Z@*c~~s)
z{Bodv93DSaUC+eICX#Nj2?d+bD@n&zvw?I7q7Ptxou<J;g6@N!Y%!88uCm2Dw&=+Q
z3Bf0S^+LxB1WpX&GtqJlrya6$W8zZ6E^*1_I>e<E`GmL>-5@ThEjHvvIdY>Mxh+cM
zbvtR2PC!V;$)^;6$C!j!9f|;47r>Jqr~nw10K0I>7D2}hAV^EBa_<>HIJ&yzyuxAz
z6<~gl86XIu_)Bq$;%_v&0iTGm(dcYeQc}Ese%#IeIkOrBMTt4U1F1~sd-BFY0YQ|V
ztua&Z_`8e7jCZG$JeaF^Oa{=#JNX_I+eK8qE8hcOfjSSv_4CEWyH}_4pHJtf%imW-
ztWuz<Tpv6+wvCPKxGwKUm%<g~H-UWxHu@ipe572qOWxu?SMqMOSfdYej+E;(jbMm>
zwlaRu>Ucl=Y=O*n5xrSABe~ZtgFl3<8)8wk<(W^gsBUN60*mTi%^HhxD~@w3j!U~&
z*7wGOdG(^>KFvysO`<08gh-XP6NTo(BnlPp;s#B`WN@kQ1IiSO`K%=goiItTWd~a>
zhbA8MgYo8^cdG*}nm83f6@iLP2w^mTAwGbmYfNi=C}f%8ikcDCdj>{fPG^cfYnpBY
zg|s*T73!p6;n)h-!sS5gHVNhj5nU-ks9Qa_;yxu%v7)pM@JN&tRN3Ec0uaKWZrqI%
zz)IER9}-j=O<MdmyI9Qv@>4=3@(h9Vqxc0wwrktQxfxd8U5SvukVBh_A&Sd?e9M`5
zeb*OpVPl^pfUD_4*(a%<zdWCvoF1ROIyt_awL)3SR-s%EDSW<A{-oK!KNJ-K0~N2)
zlsGv;jdzWvg&oCm?=|%wtnzO$oUJAQ&6;|HBHfs+#JI~=#^G<wR^qyHt_L(mb`iKq
z(j;wCYEqt#g$Rj@#+=6$GhuRn6w+Z=%xuS4!OI#jqi{=_?mZna+ltwG+f;19EIj0N
zU^}LnCqx6=FcmWYCs+KqgAvPNYi3LlCu=aU1=OoHPbT(0!X2Jiua<-yzys(ZyT1Sp
zg8vi_;%D}0w_3D)TE<%!xlYrNKv7Yx{r~d&{?*itmC+lgbfVpqPT_WcDV=09rMJQ|
zwxk_%lggCKo!+UDIgO|)?CgQ<NL_IpD<fg&>`_smhV;T9+*7W1V)T&O!kgf4l-(Uz
zri5IhsPcs(7(<!jJXTYJ6()$SuoAHq;B@cxbjWu6X=B$#>T!z)kwxCl7Vl=qb1Nt=
zW@m`qR#I^KB#BVG7N^{QYPQ-@N$|crm=hmcTySekWou1qa~UhxGXt%<TS^(uYi_5M
z;aum<Q;PVFltOKViW(!NmRzVM7iy`8+Y{W@w79Klabum#)9$_`&l~Bepv<F#rz1NS
zx{(QNSGXK@g~hgB^6>qD;fSh$;t(D7i0Y_A+lJbs!_*m3q9ulZb(w8PV@c(~rNpf@
zmN><^5DHS4vY3Oj!}1)|5p%69R1XEqOt$+`nie9d!t*{Y?q@F)HJp)aJ1(9>tuM8H
zB_tz8^p-Jr<C9TR4Y5*!gRh3zr^LM&f>Bn8efeQj0!~K7&3l#j+_}v9^@q>!{f!oF
zld7Iq=L7FN?FD8%`!W!5xB`oBf4m`j)m`{iz8(CgP59-uAR`S-rGcq5FqO81BkLy`
zU`qpRX@D&au(bj9aIE&|e{bFSx0A6t6ag@o(L+2F0x~j}0V)kAm$=P5qkqR9huZ!9
z?hjWy8~frwZq!n#7#ZIgs~PMmmMXgwG8m*`DnYfzD>YP+g(_`vhXf87qiw!pqQVyk
zCuS9qVbaK6QGQF%K|HxjD!oyMzirkJJM<yJ5P#oh&C#x`c{~^rK@g-`&GGHDlL6$Y
zO($*17XbKu1z_)oBIc~_jDM0$^Ag${P7T}>C}?8Z*+!a6KkI=Y$*gFF0t6(IS|-Uk
z*>R?EI0bJ%!I0CduV+j%k}0EFEj{M~uSDW)44BCl*nIX0x)8t!As+-Oz#H!0?nGuz
zFY3~=h_4j+6?PC~ov-gx{Bx`e&YY3MQh<Akj59bJ&t-Cc^v>%+B!3Gdxnrb@zf_EI
z*$8u@v|Z%@nUSgx>&bMJ&<m6fzvXRSIE84Gk<mp{Lb;}SM6F+eCO<MjjoOtmP!ogJ
zt}>AVkr!D4_PE$DSY!x0bKW;t><=L`37azt>7exXxWYxDuQvhPaQdP7_3G~G<4MSf
zkXOCz-oC#IH?$yTtbgyMWM1g*(|z~b)nD4T7REH^n8o9Y;KmTuDa$ZI+9a}ghnq}U
z;#?$g?Nvequj3)B`F!5zGiYOY#zyn*_)lbZb!EV4sRh&0wg!vCWHyteneeT^q}Aqp
zlbm86xvMovjGF`YNNG;&Sw;1;Z9^(!)~+;7A@5`3J~1trKz}1<<=-<)X`G)+`zL|q
z%E_-s#`(RI8!*bWS|jK@uj9iY;k#qrAybN)k_e$ICxejkV(oB9NAzHS45r@qvcnzP
z7hiYrEzts(-=YUwQKrhm4z9goip$9z9HoLsY^2d+JNQs~^K_j+#AYxif^6kT)*`<m
z97W2FO2z#c*?)%zmZtq_^XDRheH!aZxjJE4$(#$GVb>EXi98gMad&Lk!$fPFQuB(i
zs)>Rkc*b3=2uo^AkuyO3L=o|M^WcfqR7?FlugH65-KipRy+m}xsUbzRk&@ia#Y<v=
z8W4j}6)5V1tY`g`O0(ddXR65jc)b<EnBGfc#v&k%2!DBlIJUejxLeSKM2wXT?qVeu
zAh$w8)laiTlopN|(>g}8oR==`&@JcfZ0w|$OoC696rT?qurrQM0$1uryPH#}1h8$F
zrkn**Ng%KEISD{XY-=cvCKJH8BITG94fmfu%>szlj0vaDTxALg3YoN)`f*+nIp(z;
zS!G)(<$vaP^ZL+idhgEj3@Jw$d#-0-JF22u>Zf@{p3o+mGxLW64TV0*AGw%V>b|uB
zAd>n`eY^)j<Lyow?^R8=4z!?eTr<>4V}ayyC!_`j*j4Seh9}bb3)}5l8s+txcDoUJ
zrgpp0a?);}6RWC^K5P6F=o>nRXCMElurJWJE`MJetzS6$QD_0IzjX9dORpULG(ykB
z#YRiZ#n7F`0JM-%n%Tq}KkvJ#!98}dP1il?Zh=@!CUOOP0mPLw>|O<NE%aOv*INGn
zr)y~A0F?NG@tX8_gI52`<F#;K=gg&3gPK#zC8IQjngumU?e}wy(n2Glx&B6@G-)gw
zOn=f&aGL||m{d+jW)bgjtK5#CP3&b;G~h3+YHXbwqG^<o$}|FBbBJcq(hA#JG&Ms@
z6Su~iXBwe}xT45gWm%rLHHLlb0h&~jaj!4q;QU77u#uc7`1-T5-8Sa=lX;o_!g<+>
z(g)2slWKZjo*2pGeR(LoV)lb&rHrVho`1f(-iGPOzvJ01&4O%5JL*KSn4i~S1qn~J
zlw6ooNoIJB&*qOG-d%4MZN5iO>^ApLx7S-MHy<AF|9ZV8yZMNlaOBOWbbEfv0h6rs
z4gTZfsfo6h3+yU%W(*W3CX}C#!7uauw1Q$k6N8E22NWYqR8kL*laSuQ6y-MoOMkn0
zc=Px!g8NVW<QdwvrDq5w5nV{}q4au+S5gJu1!|kn+Q<`y3)Hp>n{(~t619=lG`7P^
z<TY^LeE9J6HK=XTbXe2ieh#&zC#Hlv6Z<eY6m7%`_6w9oktD@5^y7>co^jVe|L{18
zWXov$+$8c}#A+MH@8-gf9Yg8kV1IhzV+X2NUrQ_y3`{n@5!wIaUr)*PH*cQ;sods2
zkyP*B-{%_-Z~p{LxZk{aN<4e`oF2T-0v{t<>sY9i!QO2eoF?Y~6GK$g7qt_=q+#g8
z<DxgN^#<;RfPYTw>HcZJ23O@b;TJqBb&r63d@zsOmqH^I4f5rnzL^LVaeu6s>P7uR
znb2*}7>3+Y>yD`Q?Mz+rxR>>uFxxjMkEWW}LHCuP)}`p>gPEfkmR|jPLxJVz;~RD&
zs(tC#!UU;=jkCSR{ahZO1PDu2_GM#gWnrH*c@7t=P}|G}vPbBE@MPulLf;zk@85ka
z8^cv<HM-a)<Bl!Q_tFK=(tln@o#t3e({2A`T0=$aqMbvzo*%LlvZa76Y@nOj=xJ^k
z58LGt<b~jA1kVJzS7B+KpkkIOvyD>imkNJKSd;!^pVmGN{uI6JEA#@p+llzDGD~YZ
z2LyVQJ8_CkLo1suVqO1N7LFg-HPV;3$sqI{#N5|J4U4^IP7FWrwSN!9M4Bejo8K>d
zHB^X{>cSW19NO?1N@x?$QZ0Rs3Jt#=iho}C_7(a(I~0)f-~R{~y&MYs3iRfc7+ge|
zFpyEpAJvpu?}l_5$=P<0b3@#gKi{F1aEdCy1;kMMcqc0fxk^<E%^8{qtJS$mLd^%q
zpuTmgk{Cp13Tz1Jb6FseD5FEU32k2bHLStT5&zH4P-2lbp{i9dY$&wA8?G7hS0%A7
znzbwIqj`oVBbc&g9JFy2C?*xzQ61w6NjcOgYM75im5=iMVLIAhkA2g4b$9h&!Zm}J
z5mz}B12Ql;moe%m6qnA{JT?L(2$u%eJT-rDvci$FFg<;o6n`Q4UnP0b&}5YoMyu)R
z)AZXLDZ%h_;s>SFnjg!5JATvQK`X~Q6z@>Hs?_1ZqT&_JE1LHJ@mul({I|_(mrQS%
z-WewYhciD%e(><e(=T^23tJ@vEy(!ESQLDIkqr41WT{96No-i*^UxK!!Y85w1c`r$
zCBVmnrm*fh8k@eQW4lUmwQvfb@YL}(nM(0v@_tJ4;5o0`N*_-5*SSRwfl8()N?16a
z!)bM=%y(|SrBl?}>x*%d7v6xu)~?`CI8T#76c&^AUdj7Fve*JfIE87;&-aO|wX0mc
z7DnmGXd|r6T-E&p-&=+^0KEDsb1;9f#z6xK%ZGIw{@pxOqY~Pxo&?G%v`*sKcT7+6
zI}5pe5HS__o=R_D$FP&C_1mnA2A&N9taMgBK$T2wOD0(|vON_%OW!T!Eq!Z-gh52T
zt#497j`B77zNS2;Qhr^wZS5&w5pAQr2T;oIP$cpq_4KeUB%$yOj2%f8(yf2r%L|a$
z$DZn9uWo5$@0ht>LF$Z$<dyKUB6Y&TfM|yYFlIoSv=Wge>~k5bE*V1to}BA0lh7|B
z=M99*6x5*BG5uX|c1@Ouv7q#cxdg}vC>)V4a6iu2nq=&xR|7VNHB)v-K0yVL3^dl}
z>?G%#bvm+z!W4?IC(@q}Th4!j0GuOTXMr1mf#8HC)05l`4AfQ_gq~_UXc3K2Quicp
z-`)p(1f)<3Zy*I)&oF!C&~=wYM?&(&2bT;b0m`6Mt+G*V=K{qkuyv+rVmP!>QeB=!
z^GJpNt;9+1=5!f9Lvcy}j3l0RN4Lpt9gBwGbXGRSLgq^4hCVx7bZ#SCC=1ng$oNIb
z_~(X9I%P9td<g(70pRBVC>*+w@rxjf6j+zp**qeD*imG0YW3)n@tbOFS?#$)stS3P
zbgC+$DA#2AbL&jCy2QB#P5kWD3NY@?POMbMA99-+fMm?JSzEd|w9CdW_d-|0s<>k2
z@6;xL4MG8R{G04SS(iYKbJ^sM^@J-q%z?0r1EMlrR<bUhlzk_(8SmB`1Wm$OoG7#~
z%4BYTCrRDztCvL*=<Gl(!EaxH_yTGqO0co5TLXfN#gQ`hOm>&m(eYclN@))YYA{ey
ze{Mj~E=(UCbdUMbkRbaF%|9Fv0l7v?1Kn<y`H|ZLF!qJ!@2Mpm_n3bmdWQK2%FtG4
z^Gm@B=HfyMO~x618!ut{%-2}|=EJg8MrF8vx^-P#ErDTGnRyJLz2OpPwtdrh2IvYc
zeT>jB4mi&_2FnRPXChMoZS<2vwDPgBu1eE}+bKEkJ<Z2O+&o8?)a1pmUFQa0-)yq;
zfWxF=MUpm2x!<u-alr<o`i_^H5kj;QzR9xa4(=t_V!<5Y^?vyD_;mNr!pxdNrH0dg
zpP%l+7pO=W4lAxsJ+~5{|C&DD{W@~Z_8#AHG8A1_8bj2>f#IIhajU}YX`T!%^#4-j
zCnc}ye^frNJtnp?43d0)^LIvO9U@Cz|A|r4tu)aB45Z$w@nY(76>5OZYT5FmDz3?D
zo-LCO%2u?HNv+Dm3KQH0jN<GoY@aQEn$<!p+$);3?4wJs!91B8<;tHolF7Ow-%4T{
ze8l(f>9w1X0Y(c>C!?TP4%WF%ca0FGqBqw|%7DNz=U>uOhMdI#@K}q_Hs$(SzbC#j
z=ljcACv`I=Ye_;`cO@l{DY)O=m}-_~X)eBnjCfk=MbqyhpS2>X6xwd)l`0~C?y5@n
zbgn(>Ru*Sdb2qO>iGwsoLTnZ4UuxU>0Z|-`oN8~{c|d7H>Ydle!R4-Hh#fy6wmR5w
z>xbmmb@%_NSCWQf!HwDo-Dc9Q>U5jeHw(4wTVY@gpsS^t7Or&6TO;-gb0fTfahz<s
z8aZ{iZ3`IV&(x`g6hMw`&H&4QkJF?R@Ge$AY``0*e(Ck%{PJ_ld9CJJ7F^mlwS1W-
z(2O{IYxPx8GrCs%k!T-k&M?1jU+vkxG^$}||IFtK@8FyD&!DwnNd571&mia*_TeC%
zu&7?xi1pMv8?k|?X~e!tu;r$0IAsxRd+d7^P7fsq>s2hxJ=J-{)4tq)i}Y=<;h{|N
zqdQzy0&6=)ltY0UUOa8s6)4JZ)%)XzNhfBRR=~@q6$(Kp|I#4osdo;m2BL2iB*Q6V
zkm${N;+FT%-n=^izRdf%pLed74VuO>`BsU1*z-%CkH1AWqoBEzw_jAC*COOxnuT@-
zo_esl+W2nr?W{qXf4XQe{%6Ix^5(MC<mq^sVYf}omBPt^kjh@abX)zKt?Ga74c4l*
zJ_yVE@#*eAYA7STld(D!m+>wv6az9aIhWDgEGU<D-aIjX+tn_SB1Ilihq_lzzfE%b
zbocmj_sz5B$q8@Bsp;j{i7+7XFG;+#W^zQrYMNd?Pd|N>QdxW+@nI>o#>e6R9#3=s
zL6e6BL4xp>=>CIa!W)e@8Zye@De+PGzsv8Qv%ZDCw_e5>p743V$K3zx<@-CCo&Dz7
z`YDKDmF3#ECEq;k1b_b|=r?bE{B&1edIpft&T)D!*D5`vW7BU#=ry+8PSyLePXP_f
zDwYcXdUwEYU91M3J44XF?3&ggUA>R_rdg+^t@mv?E_%Q6-v@hY8>naXE8hMqY2E>q
zNBdl1t3jr>&dXh&zE%Em@0SZwb>y!O+m-~(`{Fte2)BvSh<_|2gEjN!Vm|#2&tQq8
z3P4yT;xK>=IxW>ot78n8{H4}kDU^wX7cj@i$2LB}fA#j@kW9z0U!{4)FM4qK8Az$7
z*<0-@klv0-?Bs~4w?<3+QG<^*f{C|bO?lop$UkUKe3|Canduznx6H3`4CwGDY_ynw
z)_4Mg&%9zGcYkI9!)|#ejmaEOniN{*CBToe|J}n*AlRgvkX9Rq)8PULt(l`<X<%Ow
z&LTA5y=K{0$7oRA3%(@_tV5MAHcWMb#V<J6y$%uat3GSopM!|bcq$cj_2i0Wo+CoB
zEk7N&j)+zpv}FVlozfp%xfb+vR^C{-UekcAPy>2urGI-v!=Ys4Q2h67IEqK%=FxI{
zac!r}ULJ22mRc~nyEbI+)sGO`aTJ{|(~gis^J7n|&-Hkh?C;Y4cW6~yKuvJ{qK;aX
zxsm7;{=iN}BX7#m%B7ulAZ7)pRPWC_wJJ6UWOURwr-cRrd*8lvs}c{JV8a(j?D27m
zCfNq+oqrqcWLSo-lVOr?yv{^RxzonX)~>V#vo|?Ri*KWsi}9B@Zk4vz9ijWApKwr{
z-b!wSwl1nkv5>?K^BIiBT=SP`tbVm}NcsW`hc>UXe$wVq2!d^|G}zG)I!p5OZXFFM
zvgW_sc8p(tcCNj#-HE@-Q2a`6+b9n<JFU(;F_(<yJPQOM%)|(ju{sn1F_$2k7ZjHd
zQ8*}<Fflp=e<U{&z3VITC>@Goa3?GmAeB`O3}C=x=rDlo!@!USK6D$?==uM7WRgsh
zxzu8*(T7r%m<&cR_9gl7pMeh_Zk}Fl?vGLpmN`vU3@?8Sga;gdiDS{|!4feeX?XcK
z{PMuLF!(&-qd1omAB*1(-+I53<gkOVgRn}p-&sXif00;`*dsK)IX(ivZCX1fcmsH6
z91kZP@Oca;?SFgu#|=-!-XFDFq*O?zjUS9v4D@75rQMSV1SycJfm-261ytGsm9luE
z2_F!nY<Lop-5(uoBGHBs*9D2mXO342r^Vlgbh=N^fnTS+5yp&kDO_L0z5BqaKw{uh
zB=?XAe|xNmTb7kjj(OYD7I0|27x7@1_+S$~K}bI|>@7EynmMZm2%TAP(rHgzcfwfV
zlhRB%+d*^TH!nb1FiSm9V$7ykj9}8p9#SUFVcJ2<2F|25&fwd4E{4yz!t#@!D>h_-
z!w!^l`Q*U3DC8v6C`h-=$Yep<Sw=<`+IS_3f2YwJwCvQzPHLlkQkA4WM7@p_)hUvt
z(NH6yl!&$oL}kW~Yhka{X>ZovbYxcS%QaFeV%}4C_s3(fa<a)fwF(S9jMpEpmc#~0
z1{vY&L-*hF%gujhGSbR3&(-kz$D8m#k!HsFK?vqa4WD0!KW_fnpPGc#e@BlPb&VUN
ze|75u4AXEyI+$5eh&k_0F%E0bqvPbk`T6wy7<nFlLMseUtWZE+{zT1fRw|7ZLSe;s
z^wDwE(uIU@tn2_*i8hy0WDN7VN06o5Q)7t<-m~{SHNKcKNs?JD&Z7)*jX@$kWq|^{
zK$SUZHOs2F>KwC{3(gP|r}#Gdk8-Baf6j?;Q+%8R+am{Nt3PizZ|Ske%rNDceOPlB
zmC9H0iowoSTu@D*i?PpjQDr5-X<S68p0KC@a4CYqnRTp-4i*24hsi6(@9bQY@DdF;
zA?^$pJG>b?rGOvO{v<nPVwp)dvgzAb`9h$bSIQLPxgS}Xte}9pbEI&w2$V-0fAq7<
zl+)~+u}#A^z0{}JCbC}=i8ZdW!^3EYl2@ru35t<zBWupOdNDu_KVYcIug-|D;Tb>X
zz?U^KXx*4FOGE=wc~4jg+8femB|?M@&tgIyF2#$N>ZwDXz>x43_mtHlYl|Cy1Irk=
zjTjI$jR?`X&cs6gCYsZ3zuYAdf4vA|h&g3jkYC{+6BI2&*)enL$v%!!HaTImb`jbc
z1PkhMYJRtQIqc?Ew^0AO+eG9&Bbp<k4CMyII4fO~DhQ6$!v#bLGfIw7bf1Zni!g+E
z01OSxr@PI(AAzaoeDVlz4oTrE3(e}4J;O1&yR<q%#XTju)Q+{B>Gq>Ce?Mg>oK=0o
zTFyNtjF08C3dfRH3b}RkOsd&k<;@vNKkau$nw*!rlqEmSZjs<pA_Nh*C1DF?7<bS~
z!%<DhqRC=YtAv(J!A~g^F{of$tukg&)><(QSt^B;O(u~2ro9N*i$XSwt1u^Iu@^WG
zB}V5UpEhkqcobR4U+WX1f6Y|tyv!H4Bc{2X@kMU>1qp>VJm51>Xj7o<rol-;&}G#^
zjb=Til?V{1P~je#bMj2%7X}ujTnqK(yAw<wYvCH^Ql`UEk)6{Tm|in}n}s#>wl>Nb
zH9LeKUjO>|`sMT8PT1Q&fB*4rN8I1O{`(Vt+#c?BmTv$0^qP*{A8xX!;Mq*m|5zx8
zcT&0Q`YSLLh5&GVb+z4RKsk!Fmj&!RvLx-Q)9-{PJE`Y=NzkJ%9`X)Jt)@1o`=sH|
z<FimFN+0Ptrt_EE>^vS4bmD$T4^uNcQl4+{L*>Q}(@B>L?K~S`E~ZqvGc^dUX}i$B
zD78&x*N4gWOF#aQt{L|<RaQzL@=zxadHPF_E&5#E=M+k%%N9eJDl?xinpf)LUc0XS
zaL6nw@Usd~EK%!fF7#81Q60yd*LR9)(KI*+G>}Xmmxk>;D*^YHvh6%9f3s4LKyYV2
zBkiy}0YT|TxLddTs!@<*24e0`+L{HyD=J}Ul)m9GD9v|zn%1U4NN`Z$_Vm2hIB1;k
zoJX640t-Vk+P?Nch)@OTsuq90fl!aB&*j!k?X2{9bdeVfgF^0$2;LF(%@2bhT}h`{
z`S*&LDK&oYVGuMQOT7r*e_<HJ{pdALXrLRL(WQGsorn)p-q`Zvq7s6Ta9lS4S{d39
zSM^Bg<j`himXd}x-D#Nt5-B=D@g>uhF*~TwJ7c;Mx*4cR8>my$6~l~}Bcjl$w71`%
z7?@ZnAr2w%QkZsNveMuug>w*LiI`5c>FrM^uz(jhOf!+*FS`A6e|z>r&{Xdmi0^5;
z&5Up@mEDlvW{Vl2n_GXOv{19fcQZODGTYoxo=F$-X94}S9Y2=I(*g;-TafYQXiIzS
zc0Eht5NSS^+PtV6v(5`I8b)f|E?;zoWJui8%%<!1z%)Ud>Z-c#%(Z1nvvRSWi()sG
z@#!29r*@kY0cV)>e`?t-SIPujMaljqD?ZF3N6=FXN~JD@t%_n>mC~VPwlfxtD^(vT
zR`Md1q_&j7bti;pxMqUxWi>%JTrNz|t9_3dzw2l7Esd(=m8)!)$e>{Y6+2E0&ZTtB
zIn_W^bMqB68i5nCse%5t>0P9Tgt_hHbf$AbvH(x%I~t2`e_6q36;mX8%{VXkNav6t
z&>F@St{@>n9oPVEhaSa=k!jJ~Z{kqgFk0dK3JwZyMwtF8bkh4N5xh<>v8fHTNhUHS
zR45&pJLyE3Ft1PxJqdoI2ISjh8jZL`J)QYQ0y}&Cz%IUiVBHLA<@JLl_Y=mtetP%e
zn|d=rpZR9O3*zM9`Q_&S@Y=dVld(D!0Wp_hc|a7Gj_^Fm0kXGP@jPh=1SaaI9h0#-
z6aq0dmw}Hp6PND!JjZ`qIr#3^=N~sZgYC|*2rVi0*f@_$aAbt{no21J#0-sAC<j}z
z%{RF-y!z!ye8Fba7f1@(SDN;=M7&162dz<TL9v&_K9YMxt*<cTL!KqmJ_qufIR3ti
zZtBF+BhWrZkFATHE{r-9H1MRHp&Dqbw0*-7+yP7qT$pnQbf$l1o1%S)rE8LXwHzjQ
zM5ZNnJ+z)pjLU38Xhp0kzsGxe*D|hh*_`WM*+i@;*mJO{&@^O|+q(>R*>&HCO(oD%
zv1!^JF|>rOiJePqV$$2fj8+*8=LmD5X}oCU(55Vjwowi_S6LK&aK9{ez{+B6=h0(l
zy$WI8#?Ki$wOoI+(9yQ?x&+s<=(Xh7$=L-BEvnHwUwVrHSVfUVCMd_$At9I9pS7oi
zDz(-N9)+L+6;-XPT6mwa+xB-C8}fVmFvqjil(|d2`1&?Iw^_vHM_eyWyY++et9ok2
z3v2tzs<y&5zk+)fWRx~P2fIBC$^|!CiERz00r(Dia^Zg=@}V_~wX0mY-NiND_~=w|
z0%26}(0a4=ROV%HF*Nb?RQ-B<-uyKPIXSB^K=a$zO}v3ds+JFu0V)`5zWv_3Y<{ZW
zT2%|(?}rlAwR1#8I)hV0EZTzdj$iws1&iFWK(sL&jm1<T_EkMDd~tE;s=Gqn>_1W#
z6<2!M1I&L4z}3bXd87F}nk^DD6*vssaR{SZjY+yo<^2GX8HXcfU5p>VS(H`lp?NRN
z<E7YGvH|P>TQ2kiT>1Az-i?#_;6l{piA5=!{$iLrSwcWuY^|+9r=aOvGrHQP^EsDD
za@|&>tJ$I}7oJ$1Bj_Z%%Z6@yjnwv4mhPaltCN3{Yw0Zs0r%Rns?R!*GhOdNvZVrh
zsKxxabI?fLbl`BTpXf_}`1huDq9n?g2sx#tr7A>z$Re`Tjfpr=MtD?+{32F$QsdYp
zTF!GR^)ii(t#ZHvMS5BhOXlgyqJO3p1?f$KKWx#sA7trbJAel?2@g!{P|65@`159x
z=th5HWI{!XV$IJj?m?wi@>>?G6~q8UWt>?<49Q^y7)IQ9E6U}Yo{k*&xJuI+ihvIa
zkycU!gjg2VFa$VA2<%FRK#)Fz2Iv59Zl!q5JH#5R0*q5d`izM}E>%Dll6|*5-s9d{
z<lMrgisySZcY2|aN0kw_Ue!WTfleAYx9WekH*oN|=bkRnh)xRH8Fq(-q6?lVl{il{
z0+@Ej(s@#JP8tHe$|q-}SIp3~&&9o?Mw~U1pmGM;|8$1PZU^{NZ=c_F#Xe@KA7trk
zzdW0%niR!bSCy$H=tR<fv(*PRpQ)CmVbOesR4p}KaIRKBPn4`P?l?xa?xs07k)eOb
z7$sR^<q+R7O0t$!Fz$aaN*akoDin+A=`lu0r7KT0O0pUhk~z~T$!b;})@{fa1)_EM
z#M*X-QBq~xf>BcKj-!u~YDm+gq#Dj?M@b8@<xx_N?cAfJxS4b0E{3^_`i=_pb&Qf0
zbU<-uZUVe<oYa?II8G`=mW`7Nk;Q-Gq(W-hIH^{&Y@Ad`EgL5tv}kgiR7n9Y8YdM}
zf6t7QSk@L>kz(CANun&}cm**a?%M*bAqEgA&p6`7rQ;+)Y1c3Wq;`h21c}Wq*AbYR
zCIzDic_l$0usl8B2Q3~ZIcZ!DFlBnPDvPzOC491J66lbs#&xX~CFplflZ=0`$iX@O
zY*1>t;5}*u^hA5q8Fzf&J?dOyrnS@2>8D5rQBm(Nepiv973g<Pkqj8wr=4TqTWY%G
z3@w44DnqB;aXB;JV@#BSw6)s4W1uA5XHVBZPfGs?hW2dmO?5!RlzQXBB}?Z*)vV_6
zdGjCQy(E|lWo~41baG{3Z3>gTItc?XH!+t{Y(XTK4thKcmxKEUCx6XbZEqX75&oWE
zL7-o*AHpSP_*xtexR>U7Kn@L(E6_GD49C_bx;m08t%J0`zR!>=N;*lYY_AO+a5i^6
zBxi>+&pbmimW=btC`-m$WtmHh6opt?temNTb;=Ka^(uy3T0vz)r3Gojt~8UfcIYiq
zIX^=)U=UaP2h0;<D1S|eB~xSrx+PcNqFYL-G%;jmE6o@cK|-3zDxzGP0jjdE-B^)J
zWwwa7k;@1XZ&R!zIK{@4ZqX~YT)M@qnCooGD0YMoG7clK3%&>h#oknbs5qMP?b#|$
zw#=3wn?ji)iM?_e7ph{cBAdK&JP2gvTva_A<pz_cfMzj?g@3hOl%+ru-G*^9#?6*x
zHgxvZ*SB7IN<~S`>7A{bMdd@NBarejb7`{jiDGGDgcQrF=Bzx%mA5vkU`^=;69l{{
zvn3Un>IgvvUl!hC$wI8MFic2X1+oe`l&`QxMOz%lTB{;e0ZS@!aSe+(V<^63ty3|w
zdnQe|RvHAdN`Iz~Kn6`N6O%<V)owv0OY-p+E1lSvw~iIOs5GH<^2H}DIh8^k7d;XN
zme~j+<Ut_9lf?zmOe%BL46B!26>vsnUq^6O<xoGK(Ji?0ZHS6ojcQ^0oKtBsej4jZ
zrV}?dR6N*naWY%R6WbuX%J<1+(4)-G_%^tDm4{Wa@_%)Hyn#*X=RyVpR23nEA5{iF
ztuLWxodjg$Dvw!guH$<%SYO)mIaJ3lhU8d`;pN-BEs_%B6iZB7337_L%E+7Kia{jI
zDJ6dP_1E1m|GAlT-_B<9zI$`~u9v5m)7igN_uYJPHCenhfE{zz{iXY<`~I!153eVe
z{aYANi+`JhNE*&dfpox9gk*Hi`K<d^Ve~iMkMm#V-4E(3o5%w$OiCH<3QNL4{`d|4
z8%GJ_wSx~6J1w9Xtkc2m(ZW!)(J`Pf8JIf&fX36pXQKnKgHnJN2>YCL^l3*60^-CF
z2zrpSf_5XC!bjGa0H#P-2muk&h}In?;Mee<2!G&&kLBbMs(X~!KQH`8+10#68gnY|
z4SvJZJ^TUT>3!4~{OsA*cXZ#j;V=8?d<KR7_WCFJ|Emvuf3y5k*S(+iA8z02%lXId
zx7&|5&#xw*x}Pul4}ZPsf134^#jh8a6aCwD_wIWBuKQ#)8Ew1e;<Ed0F}->}S$5yg
zXMf9naWS3s-Mit{`^o&{<YI<ye|Q<6zw`;8`s{%QcN}#D06G{xBJX*Zv;!QtXtfJ&
z*WesIGLS)P238~E`Z&U>!`hrl(v?oIJ~;#c(p1)1)Lr*?h8Q57jW+;(WJ(&*{n>lC
ze?n3xEs2$b*G%+8h8UEflyyMbCs_@EDt{pcGw<kt1eORn6{#EP(s>ush&mty%iNx|
zl|@MmgV_5n$MG>ph`}fsW{5#T3?g5gMhq?{*OOVlIPd4@lg~G+P^3IUD7G|UvG%Tf
zwoqOKBGWGX#+*S$fqCdgj7%O(v-%Hjg`*1Be%<~3AODm*5A+NmM^?>ludmMziGM|i
z0r3%r$4L?-K}|PA6^wR0U%a`2Z52xg)xEmtVeOfs`uOM1{f}?@i+-YPeeq&G>oLiT
z!blKuGRuo3N#1dNC`2E5xjYEv%4CKIFAsKjfU_9zm-+xt5>{6pV);v~4_5lvG>E~{
z&JPa&K>THx2b_(A8~VU{1O_7y-+z45y;{sK-%R?q-K!s7biYhK_h%bx|LWrXL|mtb
zqbv)nucN=7Ea$h2%gJ(J`tt7Q$<_4YyZPth6+uXA2GqQ|SfH=ar(A7b+-W(~BSnR0
zZ)-&&2CN0i(2DI{T)%UzRmM9T$4@CEbtIHDtUatAIBRG8uBSTSLI9m$K!26|$o@h2
z5a4`HIOc#4!7{*+PXixj^To&W<;~>se0F;&C~>pAYA6v#L<tmp+Ddggz%yu=rad_V
z91ScEu4D+@6*wA3PL|%fmZuOe;EGS2s95}tn;SAP2}Q`GyCK`COMx`FAy@#|x<7F@
zB%9!fypgX8w#V48J=(o_Vt+$oh5i9)1y(#pl&wcHi;s~gwsON(*0@)q*vtSA0sXWG
z^wT{-KW~D5Y5<=az^4ZAX&=BJ@6-<Kl%zr*oLW$ZJ)MSAFaLcD8@2L>#z0e_wCrM0
zK(WhKaSi}7EPI5%+TGeCWMe&m$BD;y)Sc0d$hZ+tvZu)rzu}=+i+|{L^C_wmYfx|m
z^a&x$Hqv-HSeBw|oz$(MX}()hFCdWOQS>wIVuiC#J0BhcE4Jc(JJb(i#UXsTD<a5-
z=MEy|d@n?h{n|Am<c0&e;Xv&%Y}c+@@nP72f+X^KZ^R+l6Q4+3_F=O4IPIq!yCY+e
zEe9Atgn`!}Du|RvbbrWBEJsvG`?e;ccZnI!js-v%<($Uc6>tt<jH-*APH+;TegJ{{
zYH^2(oUC52I}tF`*b~VKlm7f_-k&coFRrJv_l;`1F$;3i26f=d><0?+FIru&&JG;o
zFv;4KXlUpqBON75Fc=yC=)k#!m0OkdHl%K%=%I4{&M0bkd4H9h&)f?|i8fI*4LB*M
zPpejJt@U5?s%1@U+18@g;?{Dl^)PlOgj<dpTj5JOB|6<egk7G$tn}o;vB*g$IgX4P
z)TC@Yd2j?zQ5)J5iw52nvE(@fHa?hWACAKe00?mYl@bnQF)<xd;<$qIjyiZN2UWT0
zAZJ7bW^u?6kAHe##loI}P3ygKFoeC(_!!I4(qgvS$r<@-1Hf2qV|tqPSnrlcEvLMz
z6GTdWFK<3_LGF^91q|{=c9z|c6Uok3lfLdc<n;U{)@M8*uvlzhU*SYaKHp*S!!9FZ
zECPk(5Hlf1Ij<RkkYP{+p>ImKOj@ow98%(OkiaA@n|~84CGcrlPU;Wn@Wo@A!YJ%f
z72CGhY9q+$oWe1pg-WtNm5rdSqkt{;vc+q*wurX5(EYcFP-^c|nPJ8+QyH?3en(|i
zHK|Y;xsEar4r<hcd}}p&Lyg{0qc_!nXCv)VCmmzC;icu~ltb(~|3pFBl2l;)rgeIV
zjHH#HHh(>^LuT~3?bb7DY2LG2|Hx<COl4<fb-{#PIcmJ`1(Q1@sbmU4Xh|vs?UO&P
zB;|#yno)KYk(A{6lYytnt-<KD)F@W?dfYuh4~_!DBT_UOfecQNAHL?4u*Q*^5^}-A
z%9XkUO1S6)M+Bd<#%HNve^<S7`hV5yhD3w=h<|Jyp0*<0faEU${Puys!NcOww*P$C
zMSJh=hxTFPKtL}34cN`C4CPjaa?1tWasjtoz%3WJ_E*>b>ehd4nZmV9;aVZ=n+CpV
z;F|`XN608OqC$id%iXjgq7L2Dd(?<pDN%Q3=%5l`il_%0!h4<Na?Z*S!E2v)4?X4L
zB!4P0WD&TDL)^zOq9Fz)7-=2$KB$C?&Xa==W%AL6;7l0L5RPgIN?@$;;oyPzdrqKT
zVq<J@QhO3oY|bMbnh;0X30*6Z#S`eT5{4<sO=5Q529S~bu(TR%!tc1=5F)P)D&Lbx
ztJB%l^zx#gFU~J8?I+;*`FwW%-^pUW+JB^FJ6^7JvNtX_c!z*+WpA9G78zDbV0Cr0
z{-6?%i41Zyk=QmvaLUO4PYjVOCFFLg{JVogN*rg1JKV|ev7TD4?qmd~&8ZwwtpnoM
z>*>vM`hGTSl6vq?_9ODn#BylA=A9YV$xqEY2PIZ6a7c;Aj0FpPjBB=F<=|@1_Cn<2
z+M(*Of`mglj2#Jbk0Gzm>`>PrGlkO*gi?S-yL~lD{<X_7+fsh_c51LuyL6N&7cO^<
zWcGmgypsGMp%Encld(D!mnuI!4VS!4DhQL_S16Y90R)x|J)HrHx1J0=$p-{q>XeO_
zVI4gc0Wg<Qhc6VDl@vY0e;=>I&rg5a$wX|vj}$5+ywja^T7Zs3;e#|alwjjMcT#9v
zUK!Q>hA*e1KfZ@7Jk82N8=<{xFipHfdI`@#N)5Rf@Tnh%XC0rxqxQo*gL&Ew@8+SR
zbV(D;UMSu2*<&y?tZ)GySve8HHjg&ZE)$P<H$3oVT&4<_Pvh|(f9HHZ*ou|2I$gBF
ztYZTiICBdfSD0?D;~M+5biBc1tK-K)zZlo|>AP9f>)WI8vr=Xyj18StS{OYN;10ww
zc8rqE7#yO4cJC5?F63%^e2d!5c*qLpCPWQqk@C1In~BRzgLxdrM@_p-SHirYNyJNn
z8XC;KpO^#3c-YHOe^2s^{z2bW@@wti1QtPs8`yu1>DKn&V84?6Z}IqDVc$G_>v{5D
z6uXRbf}OpE?kY_4er=P^j*}kaUamvBvi_njLyF&Yplrf9#OIK|<o*k4c1Bn-EuoiI
zvp0g_{2Sun<K%wvihSsYBo_MJTb8j%oJik7?G>i4S?x8Uf7ep`x;XegMXxpro6DO3
z6ch~I!UQTz7l_)K_ud4~RkLPc*tEo4TFr(9xv<4DFD5N#@uH*^)~ObYd`JQetLcX#
z2O9kejdnG4P^^WPG%G)QG5IN_1ldjbT_6dR0%1gX(4>s@7IzyY4;0US`LRfRQUYff
z+NLpv8(WkWe<sB>Kc70K0@g^Zz|}4=y8T1i+|}^j+DwljxsFcq=xwV?3x_CqU*vK$
z)X8JCOcAsP(X}Y&^NoJVF&@7@300bq=upG^{V^8V<Wk0M^t&O;jqDGsB_j(q$}-;G
zH*eoR?S4Jck+w!CqVD78UHqU4(H;&8%>{(v-~QG;fA2oDt2U8TmIo5O-AZeXsZ)9@
z45|w42-f3AeJ&<zOOiXaS9o1F{p;<s8Y%5Ec;UTjsi5Ox3Lgsst8ELhNSl2%O$=>o
zb|*B^JU#Gx3lX>UoMo`Dr_Fo8Y|o_)iMc*eF{j<?gsk+w;vn-@w^cQ2Wd#D@mz&tC
zbCP!ge`76{UdS?e<FN?7eu@^x=2NyDzJM#abPQB|g&Y8mKyklekB5C9LL#O2MS6Fi
zXIHgog2dSHO%4HeOctB$Z=+V##De8g*7NpgxeL$-TRoS_l+btoeKQCWVSEn6C=sH$
zlpy?O&(7UFCrIaJAuY;H)j-OOY@R-+gyI_Fn{^~8KYv_H9<6iaSvU?g$jh5mYYN!@
zrER2M?J<TAYn-u+bn<H?&tx%GwUSTL(6b;MYk&sJpI2F9g{h`BC@GJJHL=GcpS4{O
z`@95@xH7-W_Tcx3qJBXF^MqL`WN1KsTDLm3kW0(Vy4kVA)W!|(=;w7wdyvMem2`^I
z*!0^734iAiMuoWL|JuQ~-*>%(Bi^duP4}O!A4cwBbcop<k&9VgK=RqAth=}WdU^ii
z$6rVQ?B8)J@&+gy+iUlEcr{ue`Y7r?rFV<98QMT<gRqI}3Oh@Pt%(dc4t(r(8oRDm
z%*C`%aa0NC5!?#kn5}pX=0^m@P^O_@F^^_^pMT21Ox1*<=V>G(zQt7%Y<vZLhKjH8
zD&O-|K>fYjJ$JhgNvg(1{UXzsag&UD8~)~`3fu@=R7y4v#gi}>U|BsAm|&w;=LRco
z(-?zks|R}t!6DfjQH>1ek%H+wgr)Kdc_`o8@dY6?jACTk6odiMMnsE)c~X+>Al>Qt
z<A3J__5z^y#eG<2Pbvt)N_*v@P%?3?K094F*iO^)?3_BJW9y-!=)Eap${S+oUCkmu
z`JpRF>RH=QynQ%V98iiuAl(*6yQddoFfn?mQ49+iROY}kJjY)tgJK!Y#I#uYoX`zd
z%#;uq`_b{~SRZr^CnDq5^Ag}Z`>~#x*MHOFj6N3&6BQF}Wj2=PuAa|~YnY<=rIydM
z(--(m&Ilh%x+XqJJ*c9KH&0p49{YSV8E~WPWEoit3sk1O(-xhbK9lX|KhEd1{fIB1
z%puTe`7q}~lex~g(Buz;3;naJXQf?kUu<)0zMnTmmcltzHk)S7D)vhIlZMt=&3`N<
zm5|b1xh!e0^ak$J8Hvh;LM}%(ivf1fM?`e>BhVVtts7Sj_A526ws<_RakXW~Yz(R+
zcm3{cWXm|ZvOzakerG3^V}7S@(aozd-Q1vS?AJ2rbso<=#6Pzm6A$mCyG2W`#`G1p
z<QhV+)RNl}$K1JlPKly;uGn87hkpuV`YFH{^cM(4BJ1!K_ZLpR=#Yb7ZhH3N_Ojtu
zd5zqumCv<$5JV1jr@m1SB4Twk{U~y!H}*Q|4vM)<t0Kl&)X*)lYpn_$L_jod*}Gt?
z0tt-6p4`&3&<;Qy92}j{bDdU&ipq^4@hsYuipl;7X}4}wc;A;h*EYoUa(^76kd_--
z&hKZq*}AXpXDszH{B?U7Rtv@Y3tJfl@R^;A5+WZavz_l?DC}*nCQv%Jo)J67U<>q$
zRzYkwC{)!Yo&U?6VHn=3d>t_ZG649A1!f*><L)2pC+m`gved}pjHV7F<~zd`;Ku2C
zEZ9}e&c@+78$k_m>n)w=9Dk&1s$MqQ9>Bqk?M+>er4|OzAb8E$+yX1wqb#x&#JAsW
zbR(?ry5|1Vn>~fQzizZ7tk5A;aDT{{{49hA1c-1XTo}rOyC2qZAjPyQ%(~FLX~u&N
zps;TIE$C#Kb{+skMHB_ZU!a*xm!Ab97Dyv1p!IH4|KlNLUg2a4X&8>AFzaMXKQ)Ra
zzHMx*+3Jepwf_URDkzunA3GESGBPoj!I=XS0y#IAkv#z>moPCp1edlQJ(Pa~AIFD3
z9)G@(N!a&ia<h`65@dX5tQIU!mNalB=0v<YF=x%!+A$X$q$KE{Y8&qcKJU*CR@WyZ
zK+<4|Q(7giz{jEg4ZMO?!s_b7crRXKPotAHqFkqkV5BI>p%7-?EvTfB23ujsBt=y^
zFfq0LGx$)Ma!^f?a8`HhU}1l~Nvj<tHHKd>j3%L-ZK$R4tcA|X_2)XApOfOra5@=|
zj_VprFSMeTi;k=Gatb8pu#-ab;S{OvJ4-&j`*~Rx1TUoX6J0V~10-S1>6by3H<xy9
z;9?&D11blz<z;QatDWFAICU3h$$3HaMljn56E-B<4e>q+qesDZV3vOj3DkB0v(JHv
zjPIJ$1QS={eZ<Ow)zAs0TRop0_sR<Iw{=_uMQ!yBod(}&FO)Q|;k$~{R4<bfgHHZR
zehleN`*fU&4d7Fpg?<B=1ldyxxnXa8?Dvw9``m0v7@$_r=XSGW_l9_|ZjO+NN%tyd
zljOLtsMmU^Bh(RKIRby>P2kg}Kd05z5J;O0O?ETD47@n3N~6cH%9O`s%TQmKP+e#$
zw(1zb^3(d=a3zy;X*-AO_*=%&5T^~~2tLy5;mi~WNvMfZ=n-Hs+Hc4R_r?3BkU(cC
z0)5OBekuj6hN_+1DRE5OhElj5>8ZA8#t*FekR%9jy7i$vY*2qG4wBeXjh2kE=+?f`
z`WRPkF{C|lVddMxD%zyn?KWv{ldhK1x)E{1ooLx2fb`ZMo@P|YWY_$rZsSUzFr#PJ
zazF2LAw5nsIjtvYYlJnXr3wMCFD3_>i`^U3!D)EX;xgHNsJwd{ZEK(m3N;aBqfS(Y
zA;1y8p$csXc}jnwhn;>`ZXtab#4`QUbafb;hv-5(C`pxj-2+6o9E%W^Ob~ha{I<R?
z3^J~E=xZQoYoPPsun9n&5<nx=OX>*9Au~nBu_vo=2{x;%#%M4IY?DWOx*SJlR&l$d
zS}NyqHe;JA$2m)pafW}5m)K{t*(YmBKAR*IF9_A1opyhIv&CfUa<$gNc~|De!4=0;
z?x_B{t)U}_6jysrDVJ7Jw0Fv(sBWQeruUdHRw|Gbk^=_eJ{2qFGUd)Rb!CF{aeZ}D
zIW0)F_BotQ#!+ZpNXPIp?tsD7+(-q#^3SK{>f__hzf)+lHw<_1e}26Q4>WLTWIb2H
z!h`O<|J{Fmx%s7CwW&O1IoXJ)k;a6cqePA2HpKDrDz2M-J~8mM)@^#Dv2>E>zJdhZ
zpNHx*pU3-D+^4bB5?XfSpSV@3JD*2DqjS6!OsgE67yBvpneg2CHWc$<85++^WK!J!
zSvwI4=EdO_fy?iSIQ#0Si$$}Rk>~cx=0G@96WM=sT^|f`rR$Y_HxK5mco(F$^y0wX
z)zsIr>?xRO2p%&DC6ybIHn#?cCSUi-sLXp~k<za#i_ER3E!xIpfqN;TH7riUSZmJ?
z!%%NdJI<2&_RG88h7A8X5(1%BLju3E{E2LVli}u%3%rW#zjaQWAwB2Rii8<KW<;}8
z7KML7S5x7-Y!2fPwj%j;T{aF`w$Ucbwb1W*Bfz`HQDs%XBpynR{nlLKBm@+M4GC^K
ziR%%;@!}FEA#%3jy>ywAP$^--Ug9LwTtc=gtvd;|7mlmySuX2cxk4{MsliK!yy?sF
z-}*F@B=?5xLPHa=z@97zdo`gNYw%T;R&#&t(Gn}O&NAc&KhJUqSM$Ly*W~eCISGkw
z?4;UY*3W6$02)G}Nuf~2rzUsyd9K{QA6El)gkt{EKRKnov#I^l6KZ<^?Rb@31E0!}
zU#>AzCT>lT6o9}N`>KOp2{iU?B-2KiCw|pI<Af)F<yW1WI@PV)J5Eg~i&5qx&(MFn
z`<AdTk9T1Nxj!*<_4ky0Na4Q0f_o!KsfBssw-l7fvGQu9RMe@H*@u*x@&csPgnh$w
z^?afAG*~A<K4qQ|6tvd9;duh;vC57^`2@rK(IZPz6z=&;0G9Sh&oF1DGDFoLNv;?_
zXvO6kC(_!SFExGuWQf3PjUU3V^{RhrA}fDf>x^jhd4Fs<V>+t1W(47RZ%*`Eoj`<d
zk7{C(=bS+FLxonRr4fV|X#GT<7aBpNp~6ntoZ%k9&Z6fvM-ZN)C;B%9-u1j0`cl(~
zy?<IUKSY0>X~mxZ-!VerwF8TScxX@;<^kONdEVM3#YPWk6lYso3pfLpssVo)ELH5N
z#lDmiE>!|;cjnwrsM=_^e{GBaIAoia%leE0L)uV{;?;y=ffu*)O1cYGQkEO!irekC
zAK%@9F<UbSb}2f&q}#&Cf%qnx3vdv=Nh(Dx4t5t3dGD{9PO-@iS9GoE;N2a2<MxMD
z6DqBQsU_S%1-G0WxGzJ6A`&X#8#7Es5(^Sq&Z|@-<J-D~oLW%UI{mb61rNgVt5Ykq
z6%JQW{{u9fnwLTSI}`#kGM7OHFcg=MB|R#COK%%D5Wf3Y=$N1g!+G(6rUq;`3eZ~v
z-`bv<rl%BWPWkyAa=EKr?pv0eCWp1;m1a4QZ@w9FWDkEI?BU(b{o~E+Bk{qTAmH)v
z`0)Vxfz`iY^`?`92QZG{@c8NQ+Z$`q>2<FcTT7&u*Y`nR<@Oap&~Jc#1N1Jy?JLiJ
zpm#{`kbVb6U#(uK?|r#<Eb>m}g9|o2;X$vv^m_Zp<1aV1DBHev8j_gc`QSVo)uU*5
zZ)G6$`snZwO?2pZU>}R&;0Mw#KE6z%uaDkOiU=5{O5}Gdwn)t^Bw`3)DD`D|AqU1<
z87&0uLNYDojpIr^EFv>O_TE@=l?!u!aQWC=DNy_XbPxKfffBjJ6x94)?}@)%G6fL~
z#JJOEh+?d#q11;(pCMU;5@vli?EoZa9LCxKqDBiY8^-E7yB7?a0w#nC3?F`c`3js{
zK_5D2n0fDl-YWXy9`vDi#sZB!xr#sx&Pe6W6$D}w6Df=k=u!am>QL4qQQy0N=KuUY
zq0+6QkT@EgNGJ<0Um@a)9~OP69A~i7aRrU!m0gf#(WpS3N=GCkPW`9qdi7&*2KAjG
zjlmq0FFEmTc8`9_KHqJguMk^$K7sMk&8NFwv0Uw%y+Dv}RvZWEHZj9;oSHI_ANf#}
zzUwJIgOY&8x?1nNKw6v1h0TM1bM~ex^z)z+jOvC~I1XX9Zp`?wts4Vri*eBa5HucY
zp~f@KvWdI`9;{)#0tp>Ht0LpO?-x{aXvk5$N>_4IS%tmfuADMsUICT_GT=MlN{a0L
zsczt#b%-iKLZ~0K>RmP1QZ3bh@Ph7j`gEvQsohJ(jV;_;RHpAF7w6A^xt9Vd9O?b>
zwyUddlh;uTek3>YdS38iW}^>=mEF!XJEN^dSIY#c@rx~SG-mh70t@)wHeBQECU>ya
zx)n$60Jz=4x#GlVP4CQ8F2h8xU8wF#Y<wbKKO){PYHQ>x+BgvKH%Gpe-uZ#OUNHNq
z^TaCoih4sMdnWk~r8km)ZzXgk@~u`Gk?(npO}z)2HMMoGmH_K`#=RG@w++Q$rD<B2
zWV=EGt*tF|?zd4N218uidFhxUDZ$278`J((xQQv3bZ*mhhQqAUFKbJw&AGVX^wmX+
zjfRt-yziw@ArjWS^fK&*#$DoM5|+I^D`uInZ&#Pb_l=FYukxpVICUv5g`&7!srX{R
zFAcTd^;n^w0&s$S7U2{0#TWP#_Ul;jp(~OJF_$6F{K)l0Jc~TpMcrUh<?6d*I_>~p
z3XO15;Gyw$<H>?yEoTW5`tq*&@9pEwzhg)M8`Q1apPz2h2Ml7IRze~r1;zhghmSYE
z)=#aN#rBVea#~n_=R}5;Ot*PjO<CJ<=2J&7iq*M+$^n@#bms)>)LWH=w8y3gX%kZ$
z$iwMRPQ<~TUIG%LK?k+{aGbAtt;bNGb?Ry-;neF{x2pu=3{k8WXkk=q<CJ6~)!J-I
z%iWrx)ox68z?(EfB_@hkYgTH;NUQa<oW}ZEi$;MBp?nj6z+Pn_QW5~EZORX}g>gzP
zPXG7WQe1nc+g5~p>X2zG*_WH&VtY2^uBY%_C);llW4ArYPstjD@)Vxf0jxKYq+qsi
z?&)t|`o-tv<5dykHERYU`TQ-N-SyALrdRY)lMaXCUzS%4!DwSK6j?dYjk*(dRC(p1
z2f7T#X@5I^cEjQ^WcK$^XyrUtwDVEF)a$IA=Ze-Va<%<~^V}H?bu+Eya}OfWd*g8(
z5Hm-F7L4FdKav$C$V$t9%K7o~u`Z!jtLSDOI&C9qqn+q`2L?&IX5mo`E`6jMwSTYU
zwfRVw0ZgcLTmd9Qf)?xqBx^}wi?uT{+2w7|A}IBLYnlYH3iq;_92Lh^z<yeGhUiVA
zDQ9fmZ|YvmSntkd;hywX7Ve9j!NPs<m8oA4g)>g9xVTU#VdzV(O#Px3YJkx~D>E0R
z5igZ`D>D}bKRBp@cF`h{W}cRJ`V19^>ELc4wNfRVb~TP?eKz9&Z;huKwVSF$P}c$N
zK`cgp<z0yDh@Ig%ZQl^WPFE5{W9u~8HfJPa<HYS<kb)7~1NZb*#EH<Y*Cb(6an4#v
zQkVe85#ldfl2U~j<^1-Nq#k76g??r-j^@$7eBzxj;+?<Rlrc3pfxfd{)X=6UP9ReQ
zDR}N7_)vP|A$TQpr9<#)mC>Q_0u=K3%;Lm<vN_6GIqBOtGitGlLYZZsJh-eLv~h4*
zNnL4hS*`N*2bZ<iuQRwj)9mLTTvm#199*K32?n|T;1adGk>Gd5=ao%Q9BM)&qcUgm
z`A~WzpI1Ux;`3^iC-eDv<Cl7dwsCKoe&h&Bve%!54p(~aS!lKTm1dzUHh!g9XwHUz
zC{LJ$0-<3K|KnLG%IRME={0j#fcS0y<o)FFWJS_8{mMFtoI{j>bz@ii=UM~Im^Qmm
z&o2>81itwE61165Y1<pe4gGtrbDOyICwfmGI;InJ7ob*pVd$vYts6?2GV?-HM+`x?
z@NT5j%y2JangP`6TxaY^O8(l0Y*dXeC>m6t)w(_R;L(X8q^kpR`R86_HD$2UW&6C)
zkLHEp+2(~+CbX|gHU9xItz&+dVI4gcmoc6J6qlpMFerb`8q01ZxACs8nB(Xmnk=&T
zkhB9xvSxvtVn>GnULS&WcCk(zAO->l`TA7x(M@u?*}d!Cg%7RPba%1dPf{M<ALQZn
z)#J<6H)m}QPIy!f4lnNxie5<it0X<NcyLMytJUG<m%|VDQi7%HkuD>p)^s`g-_spW
zH(Gi6LeYO0ik>QUx^YI)6HQMveFI2$Nf*%nE<ZbGcuVl!ddVw1(M33vKK=Oerz@F>
zE%%8+jlw%~u+9jgBUAV&v4Q&LY~c_@uwYHYpFv@~YoL$+y=G7X!9@_%A{gI-Kml3C
zVztoNVR^j1L3#Y}uRmYkC>@S}d;5=Tk{vyGpVxnKZfPx0UVy}Qtv3?okOBY@(uEf4
z?H!@E_W0Wvp(cbXTp&~#YT0nW7=b=?YLSMNii$2$JOUew<EN&W4cSKe=J5o9-;hxl
zHY3@>#^(>|jySDzp1+D;nDRERt>Z6XV?;>si1{lSO*L_Ze$f>`3$(<MAt8Y;9BItB
zLRx>7SLD^uXL?Mg_$j|HSXLaaV9}&7YbPwc!Xslk-2j^VnUfaxX_Ddqe_kOzvD130
zyid}O8p8ZlNOxX~*OX_1vPpAU+OAo3{*`sbvX6I_o+Z3^-_wL0pGrSwn$o(p1mZf7
zZN?PM*Xx=veVeVpSYCg^7K)8j{9%@xKaGDJ${_0p8a>T{Lm#XNPJ3)YoN#CfL<)2%
zmL5M0y~tQaZmWB1Nb<ML!9$kUQ_UNeH?yovt?Q(fO15KOCz}K*l*)wvWyq_7AgrSx
zNzuwG+w!^`XE=~MmS~OT)i@LzD}b@oAn(SZ7?l9qLi}^g3K{_#mRVUt;slybU^#!-
z-tx_*+w|)$-TpHs#G}wbtq`&-An%2foq*mCO-=<ZtZ!+*pMv9sNk|Z2reFx+<e&rr
z4um#2EgP(f0lk=|k)clqb!c`ufZ6i3yPv~4KJXI`bOx({pB)*W%G<GfM~4{ZOz+6L
z8WxZL8)^uG5<$`{Daq4K*3i%6L>YfPrITKeQD9oGcz9;s)LGy(@-0n{;+$=H87FW}
zW*@BT35#TeeG(o@R|h9TXxHBgpps1!JMoSKsFGAI*`AA-NSM~=lg9hW8j77&NvP$M
z`HFVAksKTffYv*XV>2*b3o5pk#LYP^lW|EH<hrJ0_>v3Hi|$H%2nc!-gUx^Lmf7m-
zgi@VNIer+wsnRom4UVH6#~RQ60vBXK&1&}Qhf=VlXX@Jx-vvVF4oJni4tAIGlDkvP
zMr4{)=ex_{6#_>H(_(?^8*pw)Ln{u+{gmw>$0fX0R#<~A$lVwVr1GHwQA3tASRt_^
z>^-$Ah(|xB&j49)N8dO-Y2!A9?m=M85*Ja*tgy69m^IXW*%st>DI5Qk^YB>fIv8Pt
zUZ#2_AV@>9ffi(3LB+^L`M0&lahKOIJs*F@@`mD0f%1z`P6@AxHtGv~?w*c+9n<Ed
zFqE{Af|tQAic&GW2tso1L0|(*pjr)EkV!S>!17VLP1&9z9rd(GO|>Ca>`{#;)u3*L
zwiR5gSP=7utxF{os1|A*_s9V$n__1IKW~ID6sUcR=2Z%Oa_l_X3XaP);Lf;GgY18d
zD>cZ@xKe}cj4L(Ba$Fg^2Bm&SsUIo8D=@oUgVLKeWSkU=rb=kTH9Q6O9t1YaS)+6d
zvf&zJv@Cf;U4$#%^kJXLO3H8cqb!#0iY(xxEVjau4OZ<k*a~i^y;cxA&9#EqX{{B+
zPGhYghPFB$PjQ7?MBj|3QmM%8ODcaJOTU*!aIrLXX<U?ta#cy2zp=UG)YZr)4}Z9~
zo4+dKs?ueaVcy(TPG8RhvYL`}ER^f<Amr6D&Z?Iw#c<VeZuE~DDN$l8tBo!|BgR=n
zC*HZ(UCFH!4H%I@su0z_-0CU0ke>n<?K@MLr)Y36<PV%;BgAhT@q>KM5I29C3MRh~
z33bDsS*fvXMFUFQp~tdO!YNIKyKT|DkzmF?M8@wTrV8rqMmbG#mbwMmh#!=PEk&vZ
zWP=@};&i_Ofnm}id!b4cTbub|ENHT4<9)Ol6}m*sQ@AA%s&}dVZa`kCF&f@138sa)
zyS=@=`)MpUY9^)bv@@=;HMM_yD%$jdfsS_p4CNk}renH9qui~QxfMnBYP6oluGy8n
z8ZFP1X)|q`Koxa#9T%Ug`PV~=!;Fdd*WfAbtMT^?pJp^CRp!4B6)`KLDJ|oSyCWtB
z$UM`Zf+pq=Y^P|owo%9W1?^Wi>Q)4kH|knw9ykuu6$0hS9w-l0B3gf{$~|eGESoM*
z_Imi02)TUv{eUzJvHt6@OnaQXkE!<S5TBDCt?UFdHb|`ll~G~N$R`E!l(5<@<D_WZ
zg;`eVY{N8xC+PMMVwx>6u`0k5K(*zA6MLJc$@3hAYkSC_#a6Nzx0o($MQ=>5;c4DV
zi@ZH{wz48M%0}Eabl87Cvp<05$?i6f-&WH-Nbjo6sD8hH)y0eMDEw3E{2IDzFkSLx
z=Nfj}fW>rHg6mKd+D&Q*Ih?BG0JfjQ4no3;nyGrU3VT97x0*d=K#B(&;^1Q)X!6)R
zi=@PY6~1<>+){lh!`D!GgXxc~@|MsqrSh>j;5<iMzNYTb40(T}Zpll^pOPC@Z|QD%
zb0AW=_~!f97aJTvm9XizH@d29F$3gtPQdv@H8E?-A&)sm<?R;ej-x~jp-O;$&ihjw
zcS+$7RZ3FnIMsGDJAYfYSEJ$hv}Ep(x6|zI97{`)yiF&QhqEocFxMFA9L3Xo!b25v
z3zlWfx7XdT=a+x0e_v=$c`ElQH@*GkieDH-22CvufC$PQe*OF5-PL#fs$IO0^?9~v
z!O~jn&Lvpt$^AhEvUq6Dv$aC~WQoQt>}=rjoP50)x-jxE|A~g(JgE#CG&6ods5-9}
zF~FUiR~M`;k~vj-A4L-d8J>6dG^u5JT%g+9Es1)^;O2j0Y13pVVAW3Ac$W()T=ul(
zGV6k_{CnU;%a*xSOY_+h-b?xXW;iUK$+xJF?~s@6<LJmGg3EcPhxypn(%jBfGqyBU
z%SwKzn}7*+<>;E*Ji4}7!7I&%fuDE^w1(b!3bX-%@$5980&Quml%A#1D1N-<?!BPK
zv!P-0D5!r5K*_rm)im{8!9$fiTWg1B;{>QQ6n}ee+k>B%(_ftrdKN3n2RyB2XO^d+
zO5WiXZ#p9MeA4!Q{@C27GmKM<A?vfDy}`2S*&U}QGs*c};hGyTaV9QTj+TaO>F3@T
zT%H_FWn-zCbk!Nznt&)FoypC8ZX^g>pWS(sRY^E2AmA69V)4Xy?hL+ebZc?{&Y2Z&
zo`CtpkyQh{b7IxPe4&9=i{zq;Uscy)vsUg@-2ESpbL=daVI4gc0XLU{k2MsR&Nn?b
z0Y#SvI6XIiouSSt&)~3o{Mdo~pye-Ud1$$F0LmKZ9zS=#-D{0jt_QhDYh&bc^4ZIs
zH+KfSyaDnC$WwsLog<JZMxGdX2U+e~E|kwMKig$`EA-xa9anfQm+%1E{PFnnjn2Xj
zlA%#-LuVZ+;a?=<t=1$|B&#uR5bxvXw8G1shglteSu61ELKrg90x;|pY12}X2+g(^
zSRs3lomM25H*C7EqQdm)`*+KtS*yT7%VmHsLIm<T^a_(e%mOK6RK9<QLjIllXHmlV
znd0Z&2JSOfwnBMHZM04WV<)I5hG$b7omXV(scJ{BjB2MV8U*_HP`-226?(NZS$YNg
z*Z6IJ@61*o1qW%XbeO;)PT%_u`&p|N*JZJ;=0QET0gSOTm};AC<0LA`-DeS_QC(%V
zMfWL;j92NWee@=!IfLNych@T)IYg`@&M`7$gvY2I+{YE!Rzn1VmG>?4#83jWI9^hA
zP)D$PGav3o7IWefU?d@#yxN22>g}_{9I+yQ?;1|?&NmxV+1ZNY72O_=3o!J`P^)Kl
z<MRcWkuVO2b$nA2X?6-M!BnGuK&N6m4#E}j3bv~`UE6jI_Y2wXfR06n5?ezu;iuWo
zg9;w5VY`ac)ooXEzm)CPb!=^y;xFqR^<3Dcj<%&#6`HPL$}yv<C2hYO6s?L?ueW`F
zqhn=(8#&KybI%H86J7@+R=-SnFV29Jz4*E2miktQa1DD`oW5s+*MeTk;_GnWZeR7p
zE9#M27h2U#E5XEgk;Aa{R_-=v?4_Lj<^8m2Vr1pQkbjZ{n8D6PVYA1#4IyhWxvfg`
zp^|&Zm4fNN&t)?z+06FO5*}5F*_(WSJ1PvJuUrSym#UxhQX@9iyS_Y6OJbI?pr53+
zb@4kT?WIfa@Ae|lHZ2*+pHLr@0p+!Apxo|a4-76h1Wb{^>mJP>sqClHc0`lQ?OpTj
z?c>cq6O)tm3bpJ0{CpFCASuqR3$pL20Mvc`tNVEKOS@{5X!LkqkX&C|%UpMV99FS(
zJtPegc{A=fCV)Q45WNj&ZLwyEeKhG^pIAy%Nf+Pk|D-6Yo{XXgy+DYzRStHGU5nUE
z_W3Y@V@XDPvP{yODd#34!Mr%KBM9z`)>)0J(?xSmJkK91o0LQz%|y0b-c97P`><Sy
zyLo3m)GL#;s|Lb%<GJfCwm@uujA?~VM$<WGlwC&Wdsc#N8P%1XExdB#iP<@g8FnK{
zx1-H@JJ60h`7jKE(ke*;KicY>z576?Pe4j9H5@N@7P}ypMN4y)$cXsGx7!{<*|p{<
z*r-6hqQKkL(VsQ4!Ma+&1xLw9Bxl(q#A=7h)|CjJ5e-2qMpGg2;&*|6wdjNMMDkGA
z`h0nHdk5CO`0#2K3N-Q%Lq+=Q?W%-im4jA5ScMM-@LIn6GQL|!LtrQNC!%3vfMAVA
zp)j3Q!5PnWNQreJQlMh6J;|qaCeA9`;Fb~pw$957d<-?O4~0YwADcu|qZg$93=suT
zjJ0$NxCN3e8JCPS?+QkLgMvUOM1=t<H?=t?Di(9#s{M^+1#TJ*aC?V{FMe7zkrAX(
zYOOKxfoL#6@(`R74Wv=o(}IDt7q@6YD-8gZ^f0dzNv(y(o#5MKP$9Sl-=0Mw?JT<W
z9^_TDiM&APXC+&}ft|w8de3M_DLYeX9e3-{cu|yVTYOp<KzcTR>)W(+<RIVppq$>1
z;^y0_J41dl!#`?%ZyYFRf3TXQ5t{QsL+T36{;;O6MGjU2hMY7>%jLlOV4|ize71*i
zS8IsQJR|^%%1D?tTqZ+;lWcv|5M=Nwxl8r){FBpWp(9UD8>$PPoHktk`;*fa^~;=`
zK2h!KpPV*=9-DA~HLKCl$Hfj#(aW~v-H(-c8=;?h;EIx`f(uVfyfyWaiMJuTP~vU4
z9F=(2E7wy8^C+o5&TB6}n6Eiq`%JCjexWn913Kb1>(_<=Uv=^>t%`COo-uhx=`tjH
z{(zJBm2`K;e>*Y3q?v1t)1}7h5W7<1D~;8qqEUqSwC;L;4dwnE<3eL|CMA7zBm2tK
z9Hn*(67;RM=bM^iu*!-1v@*f!6n_c*Gmp(lJj<A4Ws|JuB*V9|U3+XE_cUa8HP2kH
zgiJ{DjYg>d#aLVttO~STkIj#!N+B?&K0G}I3Q{UTP4%T2Z7wjh@lexD7_N^U#f}Pe
zt>mF;ya%xdb&Jb62ajl%w>~{d7PF6?6lvRi>WX}uBv|gUGye-)zkq;~u{sn2F*cLo
zWh9kZJq!UimyyRoD1TbfZd*4Jea}}A*q81@dC4JXI9vn)l6KQA&|-_E3+!%S7`CIN
zY8@$%m7?9R-$U|V%3fPt#g+sEtwTw3$-}wBGiOM8@*Isbo_u2Cg=#AsO0^}ygcxfp
znwU~;u@RSROH7K*E*VYDPK|etYZFF3$dl3|o(xHA=lC0a8-E8+M6R|Owh{^sa`-mI
z1_^&NH(PJWX&nO4h<xiT7z$i%)ftK!%V-K)@lwBqZHg7G!dA7}MiZP5waF#`L~V)*
z)Ie&&ynVA(6S(<6vkB4+vNIu*E2Cv^qH7>CnOJ%(&0u0^7&2TJ8-~m#GT@b_z@Zfb
zIrNaZO&&dD&40BS47<?!RWuW0G{O@cd>Nv+fFrfh5}G;H$Jnq?>p=`B@wEv9&9OEW
z@2@^ac#Yz@K8DFkbkzX8i?FMs37XozO-2$mTS7BcydfCL6(j{U$<=Tm?@jXKW7(wU
z2MmXBseZuMCb3pqB9oFY?+sm;r1f18MU!%kHv&{`uYa3JjJDTJK%!Br`$&z-I7CJq
zHUk8cY8xVWlU?&!V3SFxj26UXL_z&Fp;_NCVJ<mn9hig3-0;FjS-3VsHaT&93_+Sz
zTho`YuVGEa2hO+dA`m)CWf4&M#>KjHsLaxFY|jF^lT`6VjO=p5T2=-gs1su;?HavQ
zV$FN1bAK;S`OziiS0AOeBm9N+StnYifHmFvEH;^{H|r!^vu7QHwzLbv&?2P%8r#fk
zByp~oD^a9|LCRm{W?yCvr#yN3)C^xOFPE#=AI~mkCV*pJnc;`C^?J5i7-}E?`rG=O
z*Xy(O%=q@j%jIJI^y%<26jx~>rdPwu>fspLhkr6K7GC>M0wg2C+Amyw@$EyI4}g>p
z&z=oGtd<wAXX`h^4`06=ew_WbK6&;G2!|hk|2PAt_cK6UtY?e$RV5XveY~1oEkCU;
zW>+IsmG8cvy`7)ESpN2=01&+Hl<5RLtnh3xi*;hd^TlF0_BBFy@}{wAnPp?ssl~k|
zy?>Q@OM9z!bkb4RA%@qV&e!GBck{(BW_Yn&y`8NpHg_`oZTNQh>Wyz7Ud=AnZ-Ngt
zg1?x(6~q9@%tmy~;w@hO$?)8u``5!a%O97+*X9dE6au4!pn#)7z!LfMvuD>_cOMW~
z7#0zUaV;3Mu(L&^f(x?+``yAI5%I-^d4I4hBAD3ahhrd(wq&>oL}alM?`|82m{>we
zD#!_(k^MOB&k;_n2Sas7c(%xow_irenh5{lExM=)VY)XAAv>!G1gvzMKzq?S>-iQ0
zBw=_Vz!gZD5jlDvZUD_$aLNQBxNLJth<8KSEhO(A6;x*YG;t!!P!B=WYRTA4wtvuK
zNNj`3$&(wi7@qg;FV^$r0)YPU>f7@F7eB4nAFuv04DaXbpFW-2i{*#m=T9F#e))Fx
z>+t>A`lr8tTz?B{S-k@b+Mh3n^ULM=@GI6ErZ-%zE`}GY`P=uitHMB6>($wOu^!IH
zujY%l^NX|fa&>yOyu!nO`YwI>U4I}<_(_}b*Ke9`E1dJsfBhRt*Ca6D;?w2j$=w#-
zCAgXE(pSiejp<{dYnVkyJ<s>yZgTE6cW1p9cNZbL;qK%nCQiP$sJA+^Yq!&Gr`=9B
zc0B`V&j8vpfc6ZaJp*W-TDgyr`xv>8k^30AkCFQrSstE8M%IwzZIPGA`hQj-ZbZFx
z(cflI4Uk3xl69%*vYd4fv=Nd;uN_1PYKs!t2Er38A^lvI7@~v>)wDPM6mXc7WHxz1
zz#=Ih*%Kp(K+!sc+%^!<rU`{o$!RDbfFv>P%@dG^R>LMHID^h(IIbr;2;`Iuk8A~j
z@(TJLLO(KuSU@GSKSv;)=YO1gTIV7;^f+04k~b5Ez{wV}+Jd1JC<uvwbgqbvAe8p#
z2ZqE#ebxPdgeyKB*%2d%0I~?sH`_o!;VFR`qGItNIHTzP91#NuN};hpP$*=%C=TW$
ztqdssl6)o*c-OWht!*GsZYWZj1|>X_l(DoPjv@qVaX6xg9&(%i0Dr}9Otrh(V}kFU
zcD2WpEc?eH_O50h=BMlB>G^zdw)%Zidr>xO@1YS$Nv(tInz)%1l}D1}a&reUa`RRY
z2jph*2m{1jcLd%~ibwXtgkkc--53tD5Jx{i7TT0I+)Q8vmyVpICJ<AS+6E#Ct59^x
zD8f??K0SyEz<#yto`0yoqx6?RkAn;TGg~cBmy21i`P~21d^umM=n%wZC;pDY?}k|`
z{1}{@bSG0(Iw-~C!E3}D#ovvB@kWh!qk6nnBMvW-O@qWVK7%#o+|&W`tqU@K5L36N
zTL<-*bgv@Q$6G~a*%QB4k?HkXk(sBmuk7MaE$OYzui9HvQGZxA@E=^*0gZL2AB8dq
zA(T@fTi%(VxYFZt6XX}xXWkSNfC3eB<U(QsF%=TqKpe7=h+;)H`-9A4gFmu<9x()v
zWsJNHLrjq&ZzE!!lNX6j9>meHxF3sc;^>$hWII2W&AIjL{cKevTcPXgY#r~?Y@d^w
z1ktvcHiSC@;eU}#d-3VRIc9yaT>Ls)U18j5)}%LcR5_{q<9*5N7E{iNLN<ZqcQEyd
zEh_s%q^BDh>GSt9KRcm>I}%E`XF>_r63UqE#&g)QMXtB{uSC5SdMox;Pe#;}5%pw5
zJsD9SBlIzX-@|^Rplp|91X2gn!xZcK5MK!L_DP8%g?}Z9<S0ky@akkAY%HZ>Mjle6
z0*BNnVv+O=BQ;=w9_g$kl><BNAR?AR$=BOJ9OUq9dK;yfE=LG(uq~^bAsVvW-c*5J
z&q%1z;e~|kUG)3QjSl~8wp|$k;Y5$D5gLZRqp}UdzH_5`E2V~;a-*t!EUEGA^78cU
z{N1}*A%Cf!75zSTGRD{=a6ecHuN8U(vK~5x?^uK7n9yY`04#wJ0-{y+#{|S!gwb~<
zAaNEPcw8o!Kuk=q4a7m2z=M9^;=Tq@e_?6JvEIkHnE`zqk+>%^@Q4XWh2xUJ1Y+{U
zRuFq;17B`*lv_)aWMr^`>La9s)75PL?k=TVWq)1>yX<S+G#iB@QX<OGT}(aJ456Ak
zfN;FearZTS`Psse?+}iBPvOYvns5wV%Ed0_Vwa1t%f;B`V(fA;cDWe4Tx`2D`;mbG
z9LLOEcm$M%w;8XL!^x8a;Xx`D+TGv&aTe<~`{<~*EwCMV|9HeO?o4dOaL8@+q#zyL
zY=0j^f`oA7qrC~lv^B8}#BPjxPe(|wrC#ffNML0jj_imLMBhi>0%FJ0-raK#7y@(}
zdhWrhw<*WRk$cYPi&L=rFZ0Fwex#Q6C`60X%NzG>16Vx)mEIOUfnrvtIe7=M#v)<k
z@|vai|L&FW?L1cNV9|r;vF*F?H+SxL+kbeMSgslBtzGzsCBi%I?9x5&>{7eJH73H?
zcQ;~BiQH2n_ms%pj=P<7JLz`PQ)j;oO#L=6_1nPIZv#`m4NUzuFi9Vy-%F!@FOB-W
zG-~(KPWDpo^y^+@y|eYE)gkMh3B*+IYy)wadS|@t<;f<N&229#S67c*X><^y0ASb(
z;*duZ8CCeTGa7*i$t1_V>@|Xz9I*w&VNa`4w&|KAWT=p=d}L3IAf_wS+d%Ael78P)
zyI@1+&0$pvSf267r&S|}ep<B!#33aiN91UtYo~TlWo~y#_+LGi53!eF9X%D7!6`Tt
zmw+h+D3_^9J);58w@phugav>1r!Ff3Qt))ncF-Z*3}b_a){7pfa!iPU+BrI)#K7f(
z#r_IHE}#A|Ak82RgOYYwrlY5Dyd<vO%T6fc7%b!Yz@(K^uP{X*qT{yyY0iY}7uu9A
zN(C-^s2qA<PrN9A!X~y@dhtr1MOngo)=Ej~TuiR^C{N<pG=jIJPR4&Sk6gHydOCP}
zEw;&`LWcYLOt!;;X<%FVDi;d}puT8eV%zlXK3s2;BQboJI21=s;V;A?O5`6qRH-D&
z+xZmLo&?P)MR8L(*--Yvt5)s%v1gT}8430K@~j<EaX0Go+;V2?_IXw$kbXKI>^<wN
z*kbX*4W?zK9Gtk&c|Lyuj=VkXGd$FKq2BGVzp^_g8BQd&dHk;GN>h4Q9imQB-*Vx3
z??N-l8}$|TuEli8m&s|2s^$j1szV{Gk$CH#(AVB(L~DXy*bqk;7SSpn_hgY2U$DX#
zM>fx;0}ZINox$CN(~U^s;+r4dU7XN_?yAwIKcVUJ%tdFA2Wfw>Urb7>`I_9xQ&!h}
zVno&PYf)+Ct3e5uWua?Fm$v({@=kS<vsh@)XqFSAxFp&6h-cCyF>BVkC%)PZOVp4S
zZn%}z<MUMRhH#W7M9%Z<iYQ8mpcnQOw|CvIw=XyUyxdb%t{epX_2--LLJ6oj;w2ID
z2m!zT?eOvD`+k4b&WrGRkSgjmH*p4_j(3J>#9TtsjxuBz?><Y?X_FR(J(~o7PM)Ed
z2J*zi{9iPM=1FA&Fk+^Z?imoztMM-!DiFV9d5uPMHBB)VIDbKtSfnR(Wk|`@Eg0Iy
z^|UF=Z1C&*Xbba7iY{o&5&D8IGp9x?#<i@mTj`jv$oqes`l{+VPRW#to7NFlPSd0D
z58GpxUBTD{^YhvIlvh4h%sfu8D}#J?bC7Q{F%-(-ZdOa~m6QCOFx1euKZ2I=(365<
z8PP)hlBb@28;^b&AVeYEBOLeUv5!pfmdU9ZwmxtW&MsTIe1;MZCm~2G7^b+CpSZgx
zqT4@=j|zV^_D*!d_%M`TNE*ENtMXefXQlDIFq)YxnukCpb>;b+r&AnIO0DxWGpf3T
zV)DFGJ3K0Kl||+ba#>r>jf-eJkF!S#cz4#8f)wswdO}yA{x$Od6S}|m3Eg_1c24LD
z1ahISnb7T&;lDJY8%{3lRRg*{1ih=~b88&?#&dsbh+WgUHNd{%+$A=~v$-{nR~yY;
zBx5+4Ta&SGFt>tOpUbu3Snz*(EVtrt=TvSO%G6g4<+f7%LNmEFBl|{jTe4n#BDW@J
z-#~67*<JIvEel_19Jgj{-!$%}q+fFw7g?egUv?IEMzMDkw<T)-ByQiSc>8kmABUV7
zX$qI~RWK0(F*29Ib`BMnQ4>HZe_Bm%<Tet$`&abQMqq|4evh37+MekF7RVu49~Q{u
zl*t4MkN`WUWPkm>q9}=EOO$Ll!5rG{mh{MCu|8fEB@e$G<l+6*!_(E*XKfB%1gkt9
zo<1EEZb<x95*M94c%_8X>hSdY;fH%E$>H;WkC9Sqd_4XS__oJetpc7<e|SRSs#3>W
zZxpU*T+w)jEWRZ^g#UfMcFuSQd~iYX2#@$Y@JPoWpZ;|vv#?*E9UY>Gj+`kar9cqh
zADlJ94}xC*?}m{;=z|l{>tJhy7bXIr1t~iqe*JBP*u_br7vxPf3EkY1_t&2?(EBBz
zX8pMuD95R28R3L8wnE(XMCTwSq}LTfb(f>b32AGoe!01|)?SCcaE64EwQ#@or!qn4
z20{JhSdS;o@^Y4?az<$DD#V+cTjkvF<;Pzi?}nfK&6iwOJsM_CLM93duv-lSf$9?E
z&%UKYo)~aTTcPm1W53|x!m{?uSw6sRM^souc_o5|Y{m&?Xh8j^FPoRbcx)O$XFfxw
zjd9BmsJVsE#w<}-_NEdtZH+mwew3y<JQ!g#0t#tB+^($&*q7W^JtP4%mjPEjWC1jn
zW>-Bue_xvL%RCPEEuOdbNa+cXKEuC(HEw)5-BEeO6JODcxAPdr@EG_!j-%T+U*)Ix
z|AaW_aR%DB8meLS774dp+x1@?jsiObfLTG4r<U~xmMd)Zt^OK)J!JV=8xNE2%&ar$
zsB0kL1MGy=N>pDk*pxA$6)cGnXA$<PdfYDFf97=O-BH{7X%oinos4FBK99<yBy$Me
z4Ku8FTg$wme}nkx$t^9@s%deRV3cjt>U@Wvm7}0Zzzgqyx6xc0S(mLmFKZfwhFP`p
z<NmT;`|jQ<EZUH*=9p6Niqi%qA~<L(7L`lQ&UDJ6x#1dWwSEZqr~zTl;80<V$9<R4
ze~N@f)VI)8y0uM+w2Hz6SI@Gq#?l&yM!t<AO9yB$)R2uT%Ydo3@~O?JPp%PcU!BE~
zTs87-=noX!w@acfHPKNbg22kONqi$ge&}y3tW5Eeu)!fV<k*L0hz65zHRK!Bp9Uou
zm$Vh;?#-JwcOUzO7|Lz0s8+&OB?f0|e@%6(5<^js(G4@F7<T0l6h@Xw71x&1?5Yq@
z8d_>*`5?(`RAR6>5FrZ^W!=6?4914Ij<AXH>Y^51rkl=*z{@Ei)HMV01i`3=sEy*y
z36jk$FoKW$iMYLEVuRWs+z}kT(ZB}bj#?PGqJcFY5l~R-8qz1+T591O3#I{yfBOh`
zKpdsoY&HnD(n+AoI5#?Cn%FZt)=6uj+N7BZ)VJWIGHzuQW;<3r$APqkzESd4lkJPU
z6=YxBtspz&Zt0`@;_jl4_Tz2^*%@~$%roL{rG$NPw>G+6b&0_)g<MK=Q#u}fS$B-&
zGBL5o#JhX4E5uo>w;J{OkG>!de_YG8atxGeVtQ`mewiEkOr%*lR5NAm12&jZ=_wkU
z>4|KI@mI-M^A(l3rW~J5)x&s~60PqC3z>Rc)!P5stj?XozBxUUhK+g%SzKFlp|4O>
zZE~(FlMdX)s+l?aibv0N(x^Q(>zoTz`*zaFRE4Uy{rV7&#$*!gJJHd~e;U?WdF;-r
zhHSP>u{uO2#0`+IX!qo4-Ylu%oGDzpJz0wLhi@TMaEi;2n0na;@@eCCWhuoC8=som
zp>2pTPg;V6ArX|2IXEp(LlBo5OdXt7QU~e-cJNXO8!Iwl@C#V(-n`3nE$Z3|He3k_
z$|PGuq>ifI(3L90VRS?Fe=JRF7Rjg+Jat6u$~-rp&Lx-0wc*&Dfd&eP;?~Vnrw7#?
zAEhfW<%!`mes*s1bNxA$$ts6TKQx*c!>5JxIGhq{ePj>jL&T{a!O&A7*mCqCgM!%T
zR<JaA$Q8fNLfSF<OmSj7ew>!6Z8ARQBR+?}d-))3b6B~eB*WY0e;YJsux`4s#K%ev
zD7Q1xw1R8^=aa(V6JUym&k*GD>h~Sd9D+3Lw3%f(vE@@)^^*N;6w^=bn(bOo);TrG
zauBEcZ9BA5fWb8bZE6@bgVZf5XY|)ou@j0nvYw??v5q-6j;tYDrcU`h-(PIY9o2o+
z1l`g)IWMAi7g|pff4<tXxg<sEf9bOMT?1Zr`aI60328T5;jnDi_T2D!Hv|(Wcd5fj
zX0I3OnA&3-fmM28CC%OPqoQzMY}G064U-t-FQRMFy6#6<u2!x8@&arvSDI^LXd{0c
z7rGf=35kxJT8Xc(;*B{L*E21fW#_zsM%|8Ud)ViC6cT-Ie`bH>piUYN9=F%{wrCF1
zaBpz4TI*XueB3*v5KaHY)|C@ouF7RPGuDIVpAIPGQ7eJv)^fh~9tYmrxS<Sw%gUO3
z;*u3IhJlq;?cQ#UN-#HD|K|O!X8MvhycYED+VQ?R=sZVT{^+ujUSFFO&<Uz^$!EzZ
zF{E&Ou9T!)f4uYk`-`(Nk};LU&DmH}WGu%2{NkYssKAUb*EVI4ha92uSsfMU&0`cZ
z<00d8#MP#@VarcNLMqwqV$)`xCRtl<xD2PKsi`yNI{#XEISt8gTD~AJ8?8i;u2JRb
zJhiBpqF8f3bjeto#bKc21{ZHRzTP*#-aTFY_hQSHe^tpxfBOB3Z=kqOXCy=-Ds%Y!
z^WoFgckQT6ppoT!smb)EbJo@z$O#J`#)REFJ(<VFUc7W8j^m;AGDlk5h}oOuc?c|{
zSD~&r%zu&>RaXk{A#4<pYZs@^t89r|LNUY^u_~i`woF*E@cu$1wJZ)Y0;27OIJ>;+
z`J#Cce{L7{%Epa7(igI2I(H#g{@wArWn<oA?~0VLF@7<Y+o$$WvDUg3dVD(~#f<$O
z+^J8?3mxB;NL_j=UP|D_oF;5pLBy#v9lE_T)0uOT8;ab$QDN{a-Ep;lvvO9DiO7t-
z{@m>iD{F<1x&(ZH;5Ymd7XDE;{kK3+P(```e|Jwaf=01jN%+Uz@Y2_x{mEiX-c(Q4
zf5CxJFq0}F)IA6wP6{eks()!mX>A?kf(Z2%vC`DdEfHORwfP=ztmK>Clv|lBTA`#_
zQQF$uVMvtT)=H?}nWOoXWAX!;pLsZHS+EQxYizoHYEOcahZMv&pPr0k>$$m>yz(@M
z4?&){38>>YSit`Qe=t|sld(D!0W+7uq!$#IOJ6<C0hG5IU_HtT1MpA%m%;ln69h3b
zGc=dMR4EhzGMDl9Eh?8iYdx5M0uSZ4zh3`%CNtRXzDKB8cxM{xi~x>|@IrD)C5XMJ
zMhb(=Yu3#dycw>(zJe~i8r6kiVZ5t29b}1kRg%-h3%W78bZ1VwG9h2s%h+q8?09A$
zpSMv>k(6EDLTMYnN-DOp5NgV3<XBk6HPFN~ec=JrL7oNKu%wP?+9)P}+jK=a%tcnq
zv}{s9C)xvtX6r5PHb_b+N8h~exdds!5t#|XQRtAh2q-I?vyn>Ys@>jF5Jqxewib7m
zA<D@*JySMm`>x{<vK!d6*6M@vJM1?unf&~qqIQeznLfKiOJWY5!>-!vyY=>ybe>6k
z4{=6STOdBiPekdW6wzXTF{lqn6H3=1ZB;hqd`6M*!B?s~-^IvQE-DcS*+hF>ChaB7
zWWNkHnj+q=s;5`iXMYV$%z*^rvU&IEEWR)bp6^0KB!V){mz(DO*-!OXt3)FES+I3=
zX)SXR`D6vN2RAPa6<M&<{r4(9*=$tzeZ?Q5Meb7K=%NOyZVrEc+9)YjdeK2n36cY0
z6uziljz$A{ErL@6@ux7_J7dyH3&%9Tq(<?e!U%{yX^z>jxjQzCjKk_!Y}N=5x3959
z3_OA>KacBjEGG83f1*frfnf>R-2HE^I-4U<W}qBXgKUi?b5iKahRpX2nr+u%bV(Sz
za%PF?HiB$vciT;WMA`~{-mAFh-EQVBwRz1SD6y8_$)u`%{jLh~M4)uN2x}*T2@}1@
zvz>EIfH}#UvHqe@z4`@M)=8I^#;JnPR+Uump078`+SNbnjkc@y!L7dK?@VxaS{Q2v
z+*QY&N2$CNDu!a8isk|PAQ;_AL2|j?IJNro@6TVazN|NY<X1Ouzuu%@*Fdks>i4&w
zXYrGB0#XjO*OOtw8@&^#T9ws`O3SqOF|+qSnfF|5we0*TR%Kbzw(DJL-1o<=XZ5<k
z)Ftk}t&@LJ_F!{<NEwlDPCn0eC`xll0BJR9DR(d9{m(|5mZi$9TrYg*bBqZr94g6^
z&M_6pHUay8tYO_neV1DnBio>O>M$MI^CyRr8UVRKM!yAoRszK&P<1P!2kzn(Qn0qV
z<}&Ew?`7LcN%Kx?3_7r|ubi=6q&8ymIcgtEF1oH8xbD6IS#)1?sEVJu`<5WvexYBu
z`k;<*Uy1I7AK{E#Fd80_L(iO-xf9|;(eKVudxKsackU%Gf3&HjeDGs$tP){S!j}g}
zX`O<)R6jb*m`nlLZrXbhWOxL_m7JdNaH9fxqT$AZI^xWVP3{%5O1eBSHP|@T<PK+8
z>mS491*fNM@)GE&Og^uUyOe|z&bH9B3Cs7HZRwaT^;4d0wdyQNL;K0)dDYQXHlFQ0
zx}H)Ory5pKe^5fY#~M~qaPT5J^ijh)dpAvX{@fbjJ_9H+SP3=2xd9YvF3zVl#RCqY
z$XXG2z+qy@(~Zz-JSQ4J<>A<I*t+3ZjqR}qP-K-*-aVZA&9}?aqFuH%*N2@bO-+;h
zX{Sk~E$-82&Z)Ujs_uR2{Rn7cn8R8G<4cl-VUE@Cf3k{8JF)Srx4-?d-jH0~%+H-X
z`e{Yw7nY}9NTm)1vY`4p&nU3a-V7Km42L8baczWpI5Y=UE97?r0i_mJ>cscoPCDHn
z@i^_dByoaKC`gh}x%*t*&e~2RsYaN6CIt06w|YOT7Om0YYE@FsD_<L(Ox7oPKF`I(
zqghEYe?M77w<a`IxIU~m&ij?0hllKJeJCkE=85R-g5rblAXR|`Qc)u9=8otIN6pJP
zpxbt=iRexwZx*;9G*O|Ig~Jh-=5>Awm_nQtBrnvw9(1fypm71-6pnJdm~sp#P(yqT
zY{IbA{(DEk<VA8kHP1_;-!`CRheQ>Q(zQf`e+<V%OD6azO@qLrw^CXQ^6Dr}<6O@*
z$&#iOS}q`nIalU3a-WSRkdYM1#x#v`X)>FSxASZ$wn46Bvt&>O1ypQq7ZZYWw&e0+
zAt3}ofl88Z^T1%j;B6)Ud`3Wz2FsURj!OjvNylY=45{M(cv1Rf!}EaT<kcg$M<vsH
zCeFfb5Zcjqj=~G#BPZb{&4~u#C6mE0WW-JjgP!h=BjJC>I-%5)u{sn1GnYZv7ZaD#
zwkj!?L2^Bt0eqL;ay`+1LK*cHpN8C#Q-Zy4h1?aI&i!&74%?g^PkBt_U}C=d{H0vt
zyA5pysE4r6ujG)}Pu=EmdRv;kRgP?J=xe7rSjB1MFNY)Xm^Gb?eMovDnDI+1yet7W
zh;_l?(?!jlg_BRdjJ9&7BO6(Xj=fcB^S2r^;MxR*ZA_Xu-#So#jx&RwoVR3#agn#!
zt`0(NsdkdyM#Q>YYpe;%u@uC5I8j=whgS694L^{6PmlfGHZ~#4r_3Y|WGtIjZ%aaa
zJ@jx?I>y>Np7kMbuGOe#)Mdhi{mI$ZEt>(DMHZ^g+6F_7R^mc!JkfHc>=P<l`9R|h
zGt{X==0V9eW}X;-YM9OZVR{H>cMD%S9=SAf109Xu>7v*(bP~zSTu{3u)ok1<<M4PK
zV49i~Yd(}nq@KA@?GxDW@wxi;>G|fLGa@iilt<M6`hJrh7y>u?2>hu?rup%A^Y!M}
zdey4*=y7A%=2kmrZP^7mWx*eagwp2J7)}OrPb7PiY!lFb^W>zpYYiH>OEK~5nYO7x
ztF*SO>YUZw%%txooPa*!oPp~G28)@Aj@{K-`#377kAv*E4a`!w*D;>XjCUMuy>Y!=
zZxtb?$jMB*aeqYljmQtDf^ptLl+vA#$*Ye!rfyo=F*vi-ef#w;i}17v!0KHo^6%ZE
z7SGC;Cf<I3yIZVOphGt2UX94ha`tQ{F42Pu<ifZx86}aX;{sPEMWPaTQHaoDD4C(f
z%5*4aw4eiKR~h#1?maW#>cu1=oQk!b%EH!$*d%+sxfG#=beoBE0`+DEp*U@FDZPqN
zGy-VU3XwI08s*XU6(Va8#aSerd9??jQF(F~#)T<=pgVHYahbsmXd5S~Qe+LG&_Xkz
zoI@yTah<Sj2(<)e*_DI75TQm2+o3go95u8cZT>JSE;A*~>NcC@LMW1p>aABsOD)x6
zU}!zciWB<tSmK~Sa_HNxr*Xv^SF2OUGSN5B_QKX8l%}?f%yNZAHDDaGYky2X8d%op
zBa(4{u2YtE22}IJw3Srf_%6?)1e?%=(+O(&OqNeg<rr$y2ui>{&Ro00skHW|skVie
zzVGNeo7KKTK=?Mw6~-8)rt>kzhV#OPvqIawzI?hh?tV8;K84Vd8}s&)PmZB2N7sd2
zrS*lAPew^~b`Ewmbjr8Kle?Zqw+u%pOvJo@xARp-47eG>nYC90`tDI+x8wqPP9w87
z8Jhu+%4xG>Y*tS<7^iyU<R_hO$aoSt426an(TgV=q>QO!@K%!ziUMu!N|TMY0UnJZ
zCj+@Pisz%>i$)qa8L1G>>)5oyx`aJ*rtyC>&LD{@m`e|frS#r$##nmeIHM5xqZe_1
zjZ@J}u=O}Y(4LXOZs~p~ynR?!g4S5O(yNtY6osU@a^>B!mIx84wEVez<N^qpq!d{`
zbFmI8ZImiO+t0m>?T_oi6tiKJ>0K#KlaByfCZj){Vz!_I9Vb%^)2Ea;v=UOoUi<!>
zHe;Hz>E<3_um@lGJ-FZi36VVz=g|Fsma-k<9I@@$?D0gjC@-|f6(Irm%pQn?7UAC%
z!ncmSh>Zf{UCLyI^xm=8SbF2ws}NaZG911=b_V)d=V=p_1ys~nX<4JPprWpPS);O`
zs@SbEnJ9$36j@<1((P|;b%yp(Sx{u-g*j&9x(;A_j?;+Dh3VkWF_~|tX_F~`o})6;
zZT`}$RVu><<FE%9>^%s;?g0h|xDmr1kaI-)>dYRHGdfcjc7a@QSg~;z$P+W!3O#FN
z#*_}Jdr&@fw&{07?Or}3trpriBdw%<q8Vvr+1k9)0_K{|q_)i~({BdOzS+1E8&~+6
zhnG^}&Ov4WiKFO$3FB;Y3T1ASv^o)&fde=d0y8w1F*X7y5KE8TxDmehuh20*G|gA>
z0hfY#Jve{q&N4^B%I@*06L`V#UvLZ>>Z}mVDA7HB?|%Bgxiq-WxHRWV;j;McaHqSY
z5)NMod?7F@#NE+qff0ofg>NL`&T&cnwrT8=;0@rNaXid$j|(`&>F$@uKOcA^Hs1#d
z6)|UZXRKzRBT;zfR6{XI&dH85jms%g-S2oa9sPgt0kUvnRu-C=cDBJZ@)GD3{cjB(
z^+#ObNrP_!pBnGU_hS-w!kt9<{w|1#WIs%wW~Mg%=Ne+ym;O%Ft3ms8-KS|IntsX&
zrUnhh@CsKYv-cNrvvQD%k@1}&&Pd2X4)#vSp!!UeKJ`?Pg;KTskPrs(*!0Fkgo^^t
zh?IZCOEVzSJxfsvHsR6*Yf13I(BixMgfY^SFwUF1<S=tT!59TccT#>xD$Up;575oW
zN^ld?<esTv@wBEnF|X?DsQ+G{VQJZ9A?pT~7hx;!Oy>(M1}QP<+=SZy{bE=)k2qzt
z2D60qs)YfwcW_&fk|N^_?#6T3%W(G1kA;7hlTDL@ZScXuf)z!~E)i@7Tb<ZRa>-;2
z+>j|HKiK6t*^TO-W%3Dw%~J9eI%DcZTCbR1T&Kd!uT<!uolAk@Q<%P2>b^Cpf~9fa
z@O!$LydHw%Asr^Z?^*6?NT%*hkYBhxG*6!%55G<IB0Zd%>b`tGgcn*6GuFf0nHL<o
z`~9!o)8TVFYqK-)a$Ak7hZ{pwdp5%e-j(2&(|bKMJTikXOLjf_!C=AByIuM#uAft}
z1D*z0_->JXcY?xL?@f7f7nf!i$$ntlziVrmfOX9K_xBe;orfNz(U(JfJvV<DwafsG
zc(*D7XP7pwhWh$$)e^m8UYm)`3;#$m$$egH5`nPBE_s{Qq{K>aevj6qf(optLH@HU
z7^wxY)DVAPN1_cgV0edVNP^A`?KBNZ(3#q091@l;l-g+=(l{oiXrcH9dGJ)i>0QEx
z6~e4qL{hIa<qfmEkpo{vq+)+;I%n;pr&+N2?59jf;%@d+6?C-tkfK?=;ir%+$r^go
zPkF|JZ010g{FG)M`)LJog`aYWb5#ws<*78ImYpj+m1CNV8fwc^1)aHx)c1HQ#k9~j
z>P;#EiP9duNd>o$0IdSq@>I~9Zm?{)DrBA7?J<tt=j^l%X%NG6y~hJKq#>7jeLWg~
zq*bO6n=!JDhPE6PA-FZWtUbk_TEYMFAwrDN+&SkgIsaa-8NfV*YbTO^JwPRk^{HB@
zxY&j?qYglGNIK>m!)OQt{3?+lh%h84_(B*D41|l{-yIDPL95{5myG$%dhtSg#6uGy
z^2<6cq*;iF8iAeGQ3chyw-cB`8s^o1mcUlCNTr$6Vn#SQo6sq1u<Mk-R3@`1yN(-s
zppv_#rBh8nJE6`w1G_Ktqi4%CEKwf<@0;ZcVpu3~G&7|1d#i+qkeO+cWxhF61vx4`
zt!7<SIn-P->&+|Wv(4<WhT~GzMdd6r=dqhvq_Un#E8d`hU1NI9nq66txU#B$tcaM^
z=Jl(}3Kq*>tJG0Xp6kcrw)(LWI;L9J)sO41=jzA4kmh}2nQ{&0O=`y~CO<8utsSn?
zIaW0Xy`S%{$%Yg+z-n6TePsP&JSt(B6x%$~Xc4i7Wo8e%wlX{Ekute!Eu{YXkAA|?
zQu~G{4Eqf*Oxd0mT&qK;nBiA{>#+KIMu$0i$TdtVjzY#Pa!HSUTvXLvb@~0kvKr$N
zUDs<P?wRWC%dIfU9Q=4Ii5u)gnIfv1p1dr25^s?sPju0E^TiDh8=VZHv}KBXaja1T
zFKl86DTS~N^k$x89D-ooLJ2;agy1ixAmNZC7}sKJ6(ld?Mi?$t8l1R)N)b|n8s8#{
zOLij>(=MJ~&X5?$Q=+fx2K$KZ2_paArD1}3XmP8CE9k2=TtmIKhHGpS4WE+hX+s~n
zP1_J_sE9XMuTi&=v14I|-q^`<URC#}$H{47I8qVXh10^{IuQQQMd~43<*{Y`s|&*-
zNYWT%!?Cay2nrzR!~1`Ke0}=i%U=Woz&`^7?=*xjyz9RAkC@v@P1JqspDkzL2S|}{
zA*Jj>;mC%A#SAwLeCZB4oKq+%i_gA=QAA!YRDnj|aKh;{($hTuI5JQWC(NT6-M4Zu
z6E$Lp91T>2r!Y%`4Ud4&P~kCL<#YZNP=D(VPu=13AWg6E7Y^4DEjM$*1X5wKtP6n-
zLmtrDmz{t;Gk++w6)-C(*rP}y?ZrC&{_;J6eFe}%aqp)&3gwx=Oh+jZmza1G9#66t
zd|diGf1Q`q=}u9RwZ-^gJeLj2DPzvdy<A?+q#)sQSCAN4J&wG7zE(U?5|O?_JZ;!D
ziX~)tmdimbOF>=YfJ3oGk5pOEL~umh=Kd^GA6yeRWe(VsN=5zmdZ8sC1Z_l=mtKKA
zLw|)9JN!aA5%!0;XPACG)C%JLl0t{j*?7OYpDaf+A|AR|R0ws>vN0(TSA66OPofCR
zhjt2^QiAxlDltarHA_-TS1>7@?`Ik{@nYe;wakQ#WL@^J7|v@$OmWvzP!xu<%!^Wv
z;A5wmuC!95I}`g#J|q_ouj;9)c$jgPg@0q3ah9X!>z(DV@&9u=*^W5~Iu=pT*z>el
zcN5{5!i#5gz6t4>mEw=u`R42Qx!LH$cF*B6yH`~>oK~%TUaM2iD;5KkH))jfuo*j?
zVb6>B7Ws}Xp);*@&uXQMXFhL3@%nC}bZjxK>G{9yy(=8xe*v=b>AVVMZe(+Ga+hcw
zJq`jmF_%$nK_r!0JqrOhmjNmbCx6vjZI2tb5&nL^f<V8dAHpR$<d7@^19n`e1`@=u
zU7)RTFm!jaE$1S+x;?|~ukSP5T}raF&b>&&)qru_-C6SCa5&G*aAlq2pp3Hk%gT~3
zt*9t=_mT>xyO$MHlop=E?rF3t8!9cQtf5evxMA_0G$HS;J&`7E_|UzT(SOX{Yv?)<
zOEX(#qboyY^piRy3H@eU1QID{d@fC|oRvpt2Ia_>ffFm|eU#BsR4$}4T1pByZE0qO
zSIfW{o=oaS^NOr3c(Xy1MV1UyKuPZ6p(39<0anDO^qWOT<YY8Fqr?S6&Wdu8C5U-T
zrZlbcj;S=A@+1>PlavpxlYjNfGk3B<`IvoaW+=btoUB1J@`>DK6|5CKkRugHRF7s<
z!55W~bt;5Vdaw}SeimMZ<jUNju8?~HR@~`<Y*g%uCJEYL@<B#(L9vfqhH#35Eg(Ib
zv73;`Ybll{KymKo=3pb%73df<x=yK>Yz(G@o>gRD@RHb5^fDSjHh-J)8ab5=6<H#c
zY&T6<MWSMm1Z%@$cCxGzceC&iNZl+v1X4Fm4}p}6thXv<Z)CK<D#JeIE_9z=C{1*k
zsaOdNKxH3GXF!snsnP>vqwn+pfqwF0nZ|b5#m<fEz)PoCL|X=wT@uQUp@6<j6b;x2
zcxXe*-Z*KdZh3=rnSZ!UJOG8sreG0OSZm6n$QHA%><|jF^8~cCF>mWR%dnz+%ck@r
zy<u||8CgJ^%T^{9J=qRgXUb(a6PKR3lb6W_W9uEJPF~Q)PO|itaz6RthvxNvK24j)
z^ZBA}UR|HHa`p3U{+nu^ES4A3<r@RPV@{jDHP4$L-`MWr<$v_NeS=-9#iaxvb%NPZ
zh*}IF$X`RJ&0~e`Uo}51UN4%b>N_YqLW|ftF&P^%$w+^GgnyeTb`K1|%Ju_e+B#r8
zgY%l<aBOfo!)gOrZIeGe*_cT4xORWu&K7eZ`uof0^8dRJZTso!FHQ4))_%A?)8~ud
zn~&GOfBODn`hU6kWzv55^;7$N-cFbACg)TA@v=F)T%0wZt#(G+=4yG~JXy{z-cPTZ
z9~bkhb~%~N+vcplbv|33U(Z^+_NSlY_dk<N_2lch-csZ6dRo>OJ4K7_I7&K;XAE9@
zw?pF*G!UTnXfx=-h#BqB_^nX{tTRT18WOexC`ro}hkuLEHpCq$EH(h_IKsVo?8y-C
z4+3-29_BrO@Fr>_d@%xo#kB_(g}|FYVJ7cG6Ww|^Hd<I`SP#&!&S`r+l5qx{#UXik
z4+kFSJ%I37{4AuZBYX&A0G;4p9c;pQw}rvOGaJi3u!&CCE2llI20{C9IRgm7&NSe;
zF%S_%fPYv5o|!oS3s&-8p1?V)87^q>1c)Lu+}s241dHo=`*za4y_`<2Rt~|g9KutX
zNR$BYW%G}J{tFy}HH4>*VCDJs<>l#^7#uHXAUM(zHr+tf0Wcn{i7G(8Tr6LG!ps!O
z3#xfBX))Kia^2-$zqCKSYA5Yf+3v=(#k|EJ&wnKPlek1id6tB6Jaresn-<YSxezwX
z&h!^SE^K!pyqN7ECZVGl#9+Z5z2M@}qvpkOasF!BzG+@Oeb&65erZqFY~{t|{Z#bc
zPUr1a3GWJSSjg4ldU-y*>VrnP_sjHRHhHr6QcQ{wgT)N%d@)(#U7=Rry;{ugs*iFD
zgnyGaU7RcP@50=n<!dX{7S~p+tyEiOzNh<Pg5)$@{E7*()0EwjJkLM`A!$T`D<a^8
z-A_j*f*wRAf)NmhA%flq=d5E+)dxf4I-7iCA3T7_1}5RgK^R2(u!VpS94m?VkoKnt
zm?csJWR4m!EudJPIc-5el9?j*h@?QqCx1kTd$7tOEbTEMoYx*n#ux_JA>v4jY|e#f
znZ4orQBlY`NSZNF5jKSh4~dD+rk9KP`>TP7!q-HUZKaK9cXH60L?sq%KZ}s`0L1bQ
zE?H-cO)gny8F&YHKkhGBq6NFfGU%X)#Vt0wE&q^7Tl^#SfV=Pybx3>u$(2`f<$sl2
zc_rt*MV(6en#Ry_wMDh%Yb)%LlYG)lQsyBo)&p-V9JRIcg2xgOXuwGpZjHD5%uNX9
znAUO=cmc}C&P_THtK4Kah=WoR`?4=<S`eLf9F9y$1BjlIMnN2sl8`eqBI3$Puu7yU
zN9Lp+L?xvW5H};GJ1Rn|Bxa4Ay?@cPBTzc5B6<)*&1ek7-fKoqq9I<@jNo#ySN%ic
z=Zo1@JDZ=^Op(ndK}*&92ASffSkKZB51bFd(&NPZjg0RO#0>N#b>8lYTzI=9PPh+$
zJHO4}&fSE+$?;eG?c$2RrRLxBt*snvD+k+9TU=YQw$lC>+Dc4Ha6?Ae5`UW^2JFgR
zyHCzLGN=jU1g)o>_Lw$WZxJ7?>W4O1u*H$<hXaU#qsKrTlA|SuhkXqc4U$Ym$7SdN
zLQl|R5Dv)CFbdhP11sTA+EK?iC@J+2267&O@GWjGe%H}>{`CKy6pj1d`?%Y_X1L*Q
z`+jfjZvBS4tqGB<36ZM_k$+3IRpqG4QI(_Ww5d96s!p4#)9!S6x|h8WZCA_hHv=a)
zC2x<s8+s5!h&Bdd?;#pX{<v1m=MWxRF@Ffr-Y%!JcSD>O%^r{Ve3T<4IYE)?ox^p)
z{T<o|LA#jXkem$}MongKE*b7#%h~rpOx{|Zf<G=ZRS-Qjje<BJH-AxL<g32!N6JBG
zdt`PRKn(OW2I7#pBf%;;uN)J_taCiFV;VpZ9F%-bF$QAahm0T>9m2{l$*H>u56UmS
zUrrG?EZ?>tCiB59xwuES#1bWQIYu-IArfnfynWi_Q)Xn@(J)BC1yJl}h}4%k|0sjq
z5bH8n#v*JUV2+B)SAP;V9R93xGsYS{>)edkHWK9ZP;4#Wx^*NK;&vnz!rdY%Hfxa-
z)iji9)I>FE@-=Gm^{KB<Lwy=*OcmVzVkJUY0xSbbF=ec}JuXRA5Isqaf;eOXWkHm7
zl|iv1Y}sR{P(6r&q{cw(`_wIDag%oSDOip=8uu`Kb2~Y^oPX~1L$LCFi@~%#VgQ5$
zNxZX{4@z}NJ7Oa_@W+M?+2B7JXnlVn0ZBm>7zks_p@*<!-Y%dJXdtDS1KeKnY<IKq
zt@VISU*Zv`jg!xFC52ZG`rvR^47VsI`wSKEs+bJ@XUzThR`=`KM*y3Bo>rrKcReqo
z&Hh_W&!w*l$$x?W;G}n5qJC|JOvKxKBT+DL2C>~h)FJ7!nGRkm*<fz@0y*2Xt2(ky
z<C_G(UMg|&rMbTwS;rk^k9fzjM_ey^gj%KuwM@a4dbyS}xXSTLu9h>n)}XnPKUaF^
zO7GF^wd~<dDjzjr@x}@c2lXUu<gi-uZES7%mfj-7rGId6R|2<<q$aNs7<T!Ca3ue}
zo?I*^?T8<_e@4k#eqkV=&1CXjmb;_$e4s^lh*pqMB5IL}r<{iv7(O{G$7IKeA<2TX
z9IZe=6LB1awZUf>$;mhVoRdj$k35<g7+85Lq>k0WZZIhyXEjJFJLSVI1krLxo|DdS
zW)f5f-hXCu7Z^!q0`9<we2nq?2&OSO3kdc)z|A{E@)Fo1s3v#U+UH#{1fye?Ps#{B
z!j{j>dB+bcZdT~H?P|!zl-~v=5VXk%tq|GEPiqiav(0y0jhj0}i-3US*FwERgxTwq
z!h>*$SC^CX-)cbLd&B<$ZR0B_3T19&b98cLVVBD_G6@1PGMB(9I28gkGMCZZEGU;A
zjy;=y-OzRV;r>4do`lW!PC{jvce=4oGsux7yy2#xdUFw`2`pG?b)la&OnX;Ra}B;4
z^67VRmNyqWO>HIPq2&GQ@pU4R(<F{@XG@a*zFy?wBy**RuRbM;8LN5Dq#kPpM2)}&
z)QwD<&EOIob6#DEixirvAf|DFDw8CEDQHoDkcdSnr34qFrlFkSq9l{H66~ATM=9ki
zzGwu%$s{*ZR_J|f#v0i}jfc0`Ydy4}j+USm;`EA#LR?u=uX<=;oQINR%|kVFT$Utj
zE-IPuBdzv$Xkbp*D{+y+GimHJE;GI(6_cuha?L}*Ma)1s=b@H~+F;u})CfeDT{eJ!
zM#e%srLfq2hR~eTLPtP0<|HM->Jyw^OwMF@v}QSxa;HW1#xc!RO%;8B_5tDEww}-m
zlXn(v$v0?3s2klr6HOF1`c8f`dJ)a0hV+p(EUF`o0hGFyOz1^@Bz6NC+(;kY%B9^h
zPZwZ}F;BJ3(`)b)BPx-oj9)aoKx8t1rn6T<)<nZ6WC=GNaT5tUx9$#UeuG^kNZHLF
zS}R2hy2J~7WJo+k|6B9|na1evcP%Yt4{%D693={K9tbo8lR|*<BY})fj8(dlKv6hK
zVBQrWx(?{ZDQ1P93Fz2@jrX$QOylyjQ~iNo#-kta=7CJXzvR@XPn05w_HRvp276Zm
z9Dz%qPlFtR%s89(6I0LFEV2ygcb)u&45C}x(Bma7N|0J!ZFb9!X*99{ZU=!xNdI=g
z>71sfwXWUBJ7Wad<e^F+?NI+94=n|X%(@ty0B}97YaWX9o!dDN9eG*tb~JHR&Ue9h
z)z7*hTbU!v`3!49N(>^@)zbcd{cD1%AzBmhy3ODbn|8_t2a0SQt&`DO;S7Yz0~uwk
zx47GYTr1)>`OC+FWs{q0WLwVIBJPa?{iJWC0#IXf(@GA}?tJkk`HE_v<sgaEnL`Q_
zEvFX7cA?}|(k__uz_m9Tao!@QTz1Atl)1mJId58@K$y``iNpEQB`Oquql|t#%s9Gz
zm?K*wT<LaKJ-xp_{64mqG?ClG(}8}_*r%-vaD5iA3;gj@^LY4B&sx>Td|2GamBg*3
zy?z)3Ei(wj8Zk%fuoeWjBbog<8h~Hgg%<DU-n}QKUyIi&7>1nZ`9xg~%G@$7ZOt`$
zA5T`!n;aizpgDu}qYRdRT!j`x433xfic+5kG2`6PBJ_(K`a|#6F&Oi90cSl&A!U7g
zK*V9eU0acs)O&j-L+Xt^lSz{4GD;v};<6(79+)jNa+ZaXX1Q9UD%RA>fQ?u#mp$Cd
z`jTp;R11m)ZAX}Iua~93BF83soz<w4<v#4OnnX#$ohV8Ex6)*PK?z&o^2;(4ZIKw8
zT}+IoI=M0#U-k@1eH5kkxS+Hc^lGOIN;s}JZjTej%F&3OPAK7o@T!tZW=49F1Yywf
z(uH1q%p@rPF{*sA#VC%%12wS%7%13tJ7Y;O#qp{TQ>$ZZr%h4)uFSP@t)Sl9X&X{+
z?6ggi%*8d$0=Ddbf@DKn^GK^zNwpT)G&8k28zP%$nv0TZEwYi#{0Q|ukxjv2^p$p#
zNPYtHq8)al9$iqjBw34Wu$!*9tb{g$ddTl_4v8F9tDU|fEy@S3_xOgihK1mF`i6Aa
z`2EPE)&ehfSfoC?unwH@dX20_w$Sd#Lz1<~hMmU}+Q>|Q^`^iEKb6kyS{ym;Rh&6#
zzq~_RO(esn4sAuX)}g(=?@FhPMvnMV^#BeA7r#9nEvKAVaQaKeifcKG*JRugi2SnB
z1J;ZZt_ra8N(&eY_?H7rp)d?;4Y0+!R%zz7m;fgiE8E)F3a&DmW#s@sVarr<H(D=n
z0H9&VKzaRt)wn_!M$MRNhK-GRB`9suU#&4O1FUYv3zyLq*Dvh8OXP>f$QS6omz=(4
zr*YUH+1M(l3JBT7*Kd_m38=)2`%K9-_rt|?{cy>Vq;s{gA6~va)ej#Xehg1OUZfYU
zm_r3oOZvr!I9ZQV20A&(S_P&gpc?_5viWCITfv-v@bXo)P46VJ&t@mTBwDK`si@&r
zZgiUR?SJ|qN-gcb{a|skO;L;pDYn|QczeR8IVH(`*Qk(5o_9M+OQ!8QYb#W!#@Skm
zmE4_*RE4!sQ3>8lB>40B6LM_&Y^6r3Ho6$#lupOuW}wG@Rjt<d_oGu1Z({r8;py;u
zgZJ)#n)ovGQ318uUkwcl65d+t=#)$g5lFYgo8LY?e*5sFU?BLDD0rixuRu}z)IM4$
zDYY@p=k^&X0CWot@<MEJKwMHw8oy_{Zs4IgXzH~pDF?58rcwGvgDTUWm=1QlMh=|k
z??*Zb;)HoLv-><8%tG}XqDGw*^hC2HT6zS3eWs#Ex`xmBRZ#uh93Gp)hYro4@Q=<K
z9k)|pf~jyy3<2$iP!%HwNz&s5OJ`k^gF`jbn1r@g0|F$+Avj0INb`1@=8=K6c?eVG
z8S+rR*W)wVSBkcHq(YgQ&=}90FyxOSLE6IF|NZck$UcGSZg6j>ISPekgqe;)!;;W{
z2Xn;(yqO#@y`JCBtL}VKW{?7QUnsO^OSGyo=c(0IO$re{Gy@T%s>fc}4_o1Zk&MwB
z;HkrK4@<CE_^Lx?F2*MXB#*gBhsvO7B2GucT|Shz<=LCzc360(a<qM4&m4GC!EB64
zdqJejh4XoFcHc=dI6}`SjNNx7qw~Fgj>--SUpt(ylsoS^v<vl=bLCfz%jxUw5V~V@
zU(k;=n3bjI;?9&A*=pP~BmCSBLfHm=DIUK(x9vEU7x8{+@AM&sx^2(3;yb?GdD_2M
zdPO3h8x~Rsb(68NX&+a-*9uQDGs(E+?8K!EA^uvlF(UPnCn=>1n%3fJrrr`}&z4U+
z$1Lq5%h+$p2G!UHw$u&U8e}WJmSE6)%9sy|KaoB{iRcwpNl3SmuJo@ILUIpc(+4KA
ziAXu;bfoOVUK=TYjsJhiS3Sp^03G`bsQ=7k{%g<n6n*<QK=mD({~yz<J;;->Iu(~;
z83YrTx0gM`e{cVM`gNnTvhQ{_ENSYzQP8K?RN~ee+zm9fSlK*w&n7)%RTHqH@%P=x
z@CEcT@6%h1xpw_E4>{|U7(1++W_ajl$m_J8Sy%^O_(9h#C$DvKX*A7ZLzah}g}jxq
zY?|G|t0s0HQ*1-vMk|9Wh?V9We0VteIe9hEly%Hef9V)&>v-0N6!GNsC69OCa2{*z
z^4nx=`*|96or%-SRB)IMi+HP?sr3dLyI4+JJH%3ga+Zf!>TFGpX+m!De5+!L0_Arh
zs~h)b4AvH=wAP;Gw3y$A?%Ge=O<6i=);ScE*xJtby8(d$4cl!q23dBVR?#Krr`xCM
zw~tSoe}6ZFH(8W*z>a3@NNN81x=Ak#Au>J!S}KxhzWv>N+5A?|nxk9a+f3EQPCEw!
z#<63kEK-Bhux+2HX8lzG_hYwZoz6BI-Rys~+ol-7mKhcfP7zn@5@JRl3ZxF-$Dl%x
zTG8d~bBaRDh-$)qsS!<T!;->KyKNbYr)%V#e~MO))X0-!eGteZTb!0ECZRQ(12Sr@
z7lk@^8jzMN1!fd+^~`AoOw?_&yw%h1?LB9fQGOrl%1|w0&icFVO3(Ivo|lqlgX~aK
z_EH$@s*BFmWjD%okNF##2eVieylJe7N>C>{n;x{SlC<L{!v73djy+<mGQk(pH~-ys
ze@;Xmo)WR5q6wu4XAvkeCF0<BTdnf>b}dx8TIGK4(l#WVgFRzf)XIj^F*o0D@5J8x
ze0xXE-~3p#6FUHfS4#Om0K48)KPiR8?sP1r^rESZys3k$O&t)_;%(?b3iO4#D{&*-
z_-?$!(H@B-w1q`S3VS8&rZF<@1pz31e+{c&ZWmWE=o7Tm;aZ9iIw$3AtRtlL@<ITk
zXD+3FEJjFNP`-i!e5iM=-wyH?!+1|c9Q!^L`=9Se9-qm8Rx4D5Y9h1*i#x`8Bz>u+
z+R(ds__XOeC9gp8(6>yF<K#C^eEV)>qTQWsVXaj<rKrj=kJ~$RCaDhT{>gq=f3%T3
z0-+CFJT#Rv2a`T9X-E4Ze3C7<ebd+<5=?6sW!wCExbpnS=n7}2D-^71=6=h!EwJ=u
zq%TyfrNW>YNc(2GR9xcF(wCKTrTwAvanx^zeGAdsA>5V6`6g)FSJA#ZTRDud@i2gE
zg}QQbxw9J1I#Ih-vwKjO+%LG}f6muE%#0|Kaate_;{$fzxcdHJU1E83pkBc<cA1fQ
z4Lp}KxA7mhMrL%JI|9#|R3Mtc^8qG{ue+lp703LzLRL_i@M*8(MUMQ^X1Q9^SKB?3
z!gb*2OYff4Gyut~X4AeW?gE#gH}09^q<yI&)DobJSci0~m5Z7?O()-se-D{?QwT~B
zp{*RJ;=>{`3Bb&4h{dtW&>Q473fJ~^#dRqrm-;#HpOFm2S%v1T^2fC}5Me^wM=!*I
zq@Fp(FvNJ)v8Q@<zje>&OL}^uQgVSYVMEHWuO%)?nPkqP)@hERKte)$U%UCTq>1RL
zOgvU6p>!HK(CNi#6OODrf7z=+Z@EEdWKQjCB`(jfZ13oPie>TFWm(uooYnoAEX!E>
zNS37#IgVv1M0%FRxdy$dmr$lFkHQDcvh&K~fLouS=eoFv)0b4vIE|a}IPAw#)$c2G
zj8%ZgDGp&WNcl4k@nOm#j4BaXN{9TkOy3B#<a*_ew?p3uE3Troe@5RJt-O?I-_tiU
zTnzHJEhW~e7*zfkXJ9bbG#G_yRC{}dis7{J-1C{3a-JiJm{RBs@i3)gq+2Rx5#n&V
z<=PELvM!8z=$*eB>%ydA)oa(gQi*30H0cjo#`qT?Xv&q|06|kJbR~jj&FmQr4f%Zj
zh`F{K$gv81eeNa&Y?#=;?yjhmcqYw5sLl3Bl4tzwTc&tQJzt&RDb%?Vy|Zlc`Q*-=
zZ3d~?qyH$@&HchTgwWc~Lu(s6KZ76gL1~U}ksh-7f5^D||Fa&GBHRAo%h;bs1sSy6
zkxJ8#Wr8Yp=Ht`me~h#NPLr`Z6#+As@m>uRmoQ%sD1Y5sTaVqg5q{rap~p1>GaO!q
zWU^Ucck2zBhX8f`5+r$Qn*c@)v_b9k=l7#XN+Kmu=h$aAMjzyJj#ktR=lacXEQenQ
zIedF{|K{q;M{Nctyd|fGH@^&oKS=yb5;xfnj!3Yk;my0@#~UdX<GROXq|_Rh!*`EQ
zd;3z8$A1%qCkS_mZeKb>xTA4L;~8b~DRC)$clq99Ue8$X*~|EbhkWVVpWl3UB{Q>9
z1dd<hNBis5o2!3cO{^;|EQf)u6C`Q)+q<jy2SZk{3xiT3kQsja$MDP5U)r}8TFvJu
zC{GZSWVW^vlwx76ADFe!&J{svn>VhK&@47=Reu&5Cjmr(SX>61HTZ4ra6uUMRT3=A
zBmiMRYWnB5Ux((->u;yME2vEe)pA1+GySO&=d?meLD*8WVwX2QpC=PU(7fV_w)wF}
zP*{mGfV^3-gzqMPe~fIM&yx2-sgqd>N|=>~Rc65!?b^&%{=F^1%<7iWWeZ!oVzFoC
zgnuIa=@vHfu<Q7NMZm3wT6_auB&{w&Prt`c%G9|8f%wbZkF;>d#q}X^H?Dn5-|by~
zFGbF#wf0uUuo2%}R^i_GmXQOW-(oVIhx`of+Bj(|yfcFhPO#^^r+aCur@wqeyo0H-
zVpB=PSg7AstFPa_eUYTdYIBVxA(g7v{C~GB*(PIIII=rr9PAHz(Y_8otb~H^N)lS8
zAbqXI70nYi*h!oDiCsQ8thf6U%)M~pHdlEmV0ZN@)t8YsBKTS>*Z+C{;>bE3+@#iI
zvZVWLjg$S{_NXg(Rg9E1^{-#NB<Ab)T?r{2g!g_WA={M@d=XKo*{<yFRw0A=^?z=Q
z&m@oU@6y3`sZr(+{oQ#M<kf0-uD?4dhJ!z6cd*zoAve2|k0oYI@nMW6+p%T5!#OG}
zqH$Rl@7Gacjx&o`;-}cLCrJY!q<z?y<hD!<pa$7S<KGnF`^_a<K_&x(nA3XYdpoQd
zEkcAKQX_)+Y|5}?rFjlIycIwUTYu!17(sv=pr&!@?Ey6aC>aF-9$D9+ZATE+g9_J6
z`oX~<*;?c3CB-d~xi^=D6zb*`(OKxWD{u$a9Iqs9``mfBRcJK(RX=j&EN<Bg34k-9
zh?LpJ?7XnsWmX$e1-wGJaFeZEWy_|tDrR2g$Easg1XL*k;;zhhCkoN#U4QvYjv-}r
z$>!Ok7O^x_UwU7D1M<p!%)N5n!AS3}-NBO25etMR7x}F{JJQf_uw2RTz>c&-lF_bS
z^8U~Lzyx$S=CWE<RhW@804V^=-Y><yN<-V=&KM&cz-2=#*FJk38bE5}Y`ZFkPUT(U
zfQq6=J{}if%TGKOj(Jwr3V+Z(R1mL17saT7T_l^y=NcnS0e8y?bD+;gn4c>~xMPU(
znmP^ijc*G3$?&?l%pq=TU+wC^%#=f|Nwpx3%uEFt%Zf&J<Gv!tuWq%ynToF6T}KIx
z@|M-zuy8tG-ZE7<EB$GY9n-S3nQ0nRC0d=e#|vps(lvz}PEX1CFMkgUx$htk<S!*_
zZm#<{TS<$g=ISMVoULL<!%;)!Z6DoGh`F;~^UVvyM0W(?P<j~G3V4BW+yc2$H&md=
z6A_oyuX>^~5MTfm@`~DCapw1mv&womf$I)rkOsh&Yl;5^8vVR1QP+n?w=JmO2Wv5u
zwgjnbOh$RZ+z>3G0)G|fCd5O`ZY=h7x5_gpKo#g)4UDZ*1LFFNuGtyXpRq!zwMWIk
z02X@?CANgZlvOk@0Ef*!9wUtzwN5#vy2D&Iy<egxWqzhcMni4NC^}y%mjW(F$bJjK
zAkYWvc3KD$0Mx@r5!i8gZFTBHt?s_=w0m;9p{6u<PGe`3a(_q6&+VqN53Q1C?sv>?
z>89@GP5qCZ5M~@Upjw8KGv|KRr*B&!4F<`F+b-zW)j=j2h317w)QpaG4pKKW_J~qZ
z%g1~pioS;!s|B0cL)<MXf3pyH>6S)_W8BJ3CfOgkih<cVdM;OKsEu-*U|bv2k6i_c
z76Ez3RaWL+J%62aMit=ew}1;osl!p*GcZ{vGhp;Hq*+!|Ggw@74AzjdBs5(*4O6Ns
zHJrSC*;ZDaj^(c5+1S4ubU;>gC7$0s^5vrs8vt3?Q2HaI55t<ke#6kEqmQ!!>Gg2i
zv?!BGA$8aEqQhDH(=n7VZ|yJE$E`K(SyE+@{n%ni3x6o1rAVe<asm`6UKvIPfL(f8
z3K!P&>{*uYrLe%G&tUl#rO(0gYlbdt`6d49*l>%ub7$KHDsj`J%Hk-wceK!lMUfxU
zwoB*XVNM}+{!K@oOR1$}q56#P$e9Ru*M@d86JWTXv;Wjw=GEKVM1~BGY;R6XWcAbI
zu9~B;Zhx;OrHfQ0O{s5<u`d$zR6KDqD~`7;emC)^=Su1vrF(yxu4E7v;h>>@Ve{6g
z+?BFNdyF#(*5vv%_f7O%rMB|`yFjW~j3(ivKGkBXr_Z#Q*;4#dT8cq>)GJO~OX0Z#
zG<5-NJ}M9STJ#`QShTTu3>3{-%}Iu2#y(oZ+kd6)$f73J=12mk+7Y6Y)E;!Tjpqka
zaF8l-wfUFD>_RCG^f>BatGaewrEl6@p6yZI30Z@k3CP@M2(I%m!=iIjBB%-Tz{7Ze
zMs4wtLjT>e{dgQsN|m=+=OiO3w?C$@Wk`fT&E6xP$i5b9IcuUoALh!S42GOLf)qzz
zJbyjc2vRJer9U1);v)%OWCW?<=7bTXX2lbXAhkSoN03@^KFtWyi7cfNq*g-DJ%Tj$
z!Il1W3p=SD(NO!p7(t?d5J2W5N064QPaQ$3)|@tiv`wZvf<ys0OLr!aC}=^3PMSc9
z=Wmi~Hcps8LZPUe{T@sp1vySdRZ9*gkbmMK75UI8a%U!I`v6KI1xa(3GL>?|=$h*F
zhl434C0L!OOcl=y7)uZU)hVh{CLTcVt7>OAOt6JURloPe&R*DNNj^xk(rye38E&IM
zzH{cVP+rvx3r#JSov8!ZBGxp9)ivMP9g(`RifLF#Z=MSg(4o<z@7Z!Ll*Cw}k$-q9
zrlVG&Txla1`SE8t{2hP$F5=AKhu_C1-vJp43;oOCpTl2)Oh(MHGdNV<$$-v@^1n7H
zlYyz*6wyOik%mPV(i$|Vro3|+HoX(^%{IeCduMwZv>JQ+qAy#GjafI!mh)NI<Zys+
zadeS%Gn9ccVl+`dl?d0gArzf@jenQb$Mn*^LW}}{xU4?PD>u}!H@)J7GH%0{yJx#L
zPPa9|`ru5Ljc~!A<x@T$L-4UrxquL9F091qrF|tLx-f^!YH@n!h8lOyzGK7+wvGK#
zb_Hxv9<VycuWTIH7#+UvS5~r6AzW4^^Gmxbv83Sp=AtT@-+8iUSz4##Dt}8Z=EAzb
zsNhX_gHwFUj3>c814@)?Ngk0ooZBE|cdeVnANsh{0f(_42hStyI|c4JH_lH~RPu1h
zSMkH8kLlOf>ElD!pp`VqJeQ1oe=tr2fXh@fvY0j5fk1@fFNXNVs?&!$$O1u<f3l$t
z((%6y*k>vY8)~08)Inqgxise(>ZmDw216YULr*l+akTBvVyFYfpy!XD={U(tP}#;l
z)blt=P3+8Rj`E2#_dlFE<(iYRIu(~Oo&gk>&Xz7Hmz}9Sp?^iUUmkzF(plK@xs%Ya
z@{T6!NQoR-!aL11l|uF&C#^^>Z`e$~%9F#ppFWB$JalbAtjN2D(_WV-E6DetHHj`H
z`m*T9@{D<YheCc%^F6wszf=7Fa2w4ON!eDyuQ><Fi2~RP6c=TfS2%&%$W{eyH`TTa
zwWG>%kKFuDQh(KD-FOb4-8;T<6nhB?k;}04Rz7XeI7{7q@@WwYXcXHpSqjQwD3WsF
zmbsHX;M#_}#h^&@QrG0NsKZ^>$;^M_{j3Ks^rP%z$~KAiiDAEUxtm@6zPNi4Io-(E
zFOyJ&*@QW|3@79ykmW};x56svycSJ{K!pH}?(ZKWB!9uhwMA;CXwdsv{aL0QXu(EX
z#*dH9UmqT?{(0ad4yXiB(_g+_#RmketP2yO3P97Zf1AEu{jpuO*#Y!&`%1B^t!1tt
zi&imv;pP?LZu?p{;+6&F-iLX+vpOqIdfNY!;<UQ6h)<O0G67p@K->#IbmI|GD-e|$
zidEU{(|=_WtOR2ZWP-jtObWt*IfrrkbdkJQh)0VgfkX!$#w-FC5Az<83AO9^=m*{E
zp-|%4qU@^f<q<Z-C<NJy<yONj?f5!_GH({~@AR4XCbs2hF1wiU%E*rURg^H%KIGl<
zeK|>@WII~E#d1m&D!G?dNfP+HO4%*%_dEScjDME{oQ+d~!pZ^u_CdB~2~~@hdWZ6o
z`pqxDT+ddt*rj1m8nLGNPa)~Ekwee{uds1`iheP^-h7gigv2Uhnn`fX#fC#=ev)7w
ztK+qjsH&;{bv=wDZ<Ut#u5G^a_Z+b~5}oTYMQtP@Y;1@_IZ=CMp?36}fB!bL2+Dzg
z4u1|T!q+$G+?J!!cx75g|2}3Y@#I>P4M(HIQ$!=n3Cgze^ua3<?Br~1u9fVLqa+7D
z(hmi-JA@`Y_tEW3{^RTYu+XKPsKV(U=?gUP&z3b)8!HhdQg(x|A40~-&fUhtyW$<w
zv$XMY$V+AyE$toT5K1IfmSMb{Q_*%LhkuI+G0XV;MNfZ~hd;&yGWhBD^y)WB`4od>
zlhfa)KS|1wk_l`3%=4lqC@60JM~8Wk)HKCxP)2;)G8cIYb0b`!=G_mm{hCG@@9X$k
zF|VA6;?LT=-wn-6fRbF-*8S~JLDE5?4yU#5ZqW_v#!S&Moz%MfH9OYrqmv{v3x5b5
zbdFF$$dcch{>jsKBN<Z09iSv_e?Q5YJ9a3e)H7xiFBG%D6T_vfx#dqeSy}UG{kdDS
z<In!SPu3{!ZF|t@4f{H;z|%Hj9Lw%VCpK_dbj!kCjwK&XYvJ9RJq!0^Uvdt|O$zmg
z>l?6kcP6huKz1_r)cS^Dh)~vYynn>nuSRGtBi=s9qftH&lHbs>oJdD!K1xfn-KMxl
zHf~DFOFnu8S5w|`AF8H4SIL3kl{mU#MjeCXETLp4B@aoJ!0j;Sa<awAqa}|w8=0MR
z3XG{q#wk2{*$dINT=GUL8RanDl@Ro<As)w{kqp`Mf#r^jKe7UDq@{B^J%6klLn&8-
zT)(-6JID1Zkz`T<E>y2J+;&l8@|AmpbDg?bDUuWg&f^7Iy)~y-(CTdjI#a86OYPIN
zdbc1D6IgP42)J87|NX+ErR(J)4&3u_V5E4K)B8d=Xt*T<@-1TNsL-O1yp01R?n2oF
z&yRze(<{J1BhZ;}u%$L<RexRWx`c%xMrO4yHxV%gq*~z@Ohil^0BU%R6OonRqx-ZZ
z|1dv(!9=9i{ah1~lm*G&aG8k+%1&g#RW1Ck5w|S3pNQy(-5BaQ!x7Opv|<fMBVJia
z(i)0Wjz(C{AL0X<R^|{E1%9Dx*b~l1<REc|j+99vQU!k?)5V7)NPi@yn0a}y9d*by
z-2Oj?BS#$fX&6YfJ|Mx-fMiL%IqWb}G5eZiQ6jA{vSM~yi}~iy<1vK8H0i4~d&(FB
zq_}b-PsQ}}Sk8coG1Mgg9aeu3oN^9ok}u<GQ>s6sYt{A*Tk4VkrWsR3@@YJI09j|Q
z(Jj{<-j)0bgOru6!G8}Y3GQr@gz*%?NTOz;AFL~FZsXIyrXqPai~vJH(mLOd)D<pi
z{@0LvcfB<cu^xJHc5HtLH7N<C=OTVVeJtXaB-?|>g7_uXk%+G}dlurCOpiqTlH^Fl
zFG-F>{F3BI#4kw-;`>eCf_N>I`mjyCj(RZATK!eQ-n7=<hJU@6(ySuchP_<<a9xMJ
zZzE?L_AWp#*p7ug#2>&|d?xId)Eh=Du&+qojjfmv45p^p0==x`xY6rLpcgnYK+W{?
z#;*<JC~=K$W6+~!*(Yj>4a`erD1ly+;M`OGI_8C(ii?U;kC_Qd2pb3gJO+N8edso#
zUK8*`Ivbx5vww!16yuHIHh?dM93b>Vc7==ORPt9O8^Cv*e*Mu6xD{aJ`UtGxkn42p
zt~5_ddvM&&WJxl7yO43?f{C4!m!_pk9<}RHQCeE$!Q7Re&i%5WS{76L*2wN|>Ix3k
ziNAf(aVKmfquU#+qdD}4@{RAu<16RWLD{Ty<P-+TN`JZxXPd$RdZqLD^!$Z!!|jDr
z7?6C&`3p3kD@Og^^Pojox{<qX>)mmUX0er_M$41;nd9~xwlu`|S-W;Zz;oYiH)Vyx
zUne(>!L7{0%QF)$;M0oHrF~j6Ka)?#8sfvg(H*Z7`_X~&;S$)a7+u<rHS;t1ajaqd
zPfJWW-xHx(GgEf%u@B4VhsUe`0UB`fe3P*{6#+Mwu@D3k0yHp}fpiEamy)tQmVc=8
zudhEm@kH$Vo3e|P3dxjqlQzhRo=mAUXAqtsBvKU}jW6_}qm-ee(9jbF^dVx<q$fHY
z`2J>L^B|=ea8(X+aW!hccib_YM=Yoh@tJ2mkA>3$E$Ors1vt$d4Rxo5Q`A|AQ#uvn
zsgZsX+J{DHWjCpqH8_zfnQ=Ppc7Ne&c{VDB)^xLj;=Ig=@@D~FwaQiUy`G;T$s7Iu
zw`OK1>r5vT%33jThV-#g$?0#RWgPwK)2d#CAczV#P@AFVp#<xC@^jFo=UC~f@NtGy
z1CSY!%c_$?5Ku%<PyhsMLQK*MX>njhZ$d>%f(Ssrir<RxJ=`vd2LeyzXMdX#tr?;2
zPC|@eh>LvdLW_tn>3|fJ@GwXqC%9X=3yB2E08U8oc&RScH9`Ei?$H9Hu?1Z`j5kZX
z;eEWx^jjyMzKl1efuVUz-iqM?M?AtRZyz$2Ghc`32gjKb!|uiW%k&Lt?3#YRc;9AD
zS)SR`1u0%4isun^%{!IR8GqaFZ)6r6b6RL@Oqe5!;csw7*}I>=+q!GQIR*}UW&n~h
zlqCmCB9}^AP2bva5f!rdWJ@W@?<uF4k}{v<!4w3N6yfb^GAe^VNx_6NiV5y&wU(=P
zN&ZE~|3wmh@og$v>0p(0@*&WB)@Z3;Dt8E5hJD^<L{#27Zh72qOMj9UC&0J7zT9*M
zkq=B1o@%62YNKoAbwppEy8k}DKK=DH=tLUk&`rPn`Q#rUD5i}gm&hF8^!M-6=cgb0
zU6XS9?+^?I1aPgDY7wBBqUcK@G9l*3XH{su#Axhe3>Gy{j3S@rKLMA`$WIk|=-Nt(
zUAoiZwi@jSE*VfL*nh29rA^;WlaOA*gG90#dwNJqVB{dbO<>+nlL129VVVe^NY<#A
zQKCWRC`rwCF5JuYNOl<jia>S0X1i8}TLK*1kE!kG+4x;$zk4w0JYBnV{r&&h><5nZ
zvVr$^*IS@q&TsJ(D17N%MOa(Hta+6;Cc%yHXf?wwb*Pct^-c_rRx*F1#7Ydm=3=sI
ztyjPMC%G81%}Yy^B;#V+CE~khVa&VFX(fdNODnZyQg$o(_F+acDcTi(Y<fVjoWc;(
z$)o5Z#`wi`kIq^rEmUaF!BOyOjaFv8R=Ts>PMzpOaLW>3mYcAa;#O#vd><vwH-#ZN
zl99B%Fj8#RUpiEd0(E~ZuDr_+#D0DNiZil*-A3fwHZ`UU*(}+w$_c`nrZ7bs6kE$n
z_W#%V*4`0-t>lnOG4AhHOBu*begw_79EMdn$=fbiAVLWGN^Z3nPd!VtL;ifIPJ#ZL
zAn`Vqwc9=4rVzi$IlGP)b=Rgzie`mbF=>y?ESiJ9VU0kbA^LwsM^9BEct~YLY38`z
zV|p)}+J-TOX5Ns!ecerx_29xi9M6VPl+f>vCfl4;yl5!*yu^`QkAtBe=V$0_&oFv_
z9nzB-E2iHV>rx&$9a?uv$<REvI<5pVRaz9u)wvXROJ%uRAvrrpT@eTUdUMQhO6PH`
zdhw~trqIa=FeiWHu<E6%Ph2+7MPFkR(C@Hguu)LbYzHWXI!L(Ah4Z_DgIZTIyRT~_
zNnXJN3gicfV{}8AuDlm!CpTlIIJcebb`EC~2y^u>hqS8z4heOhb=|3a)OC~Wz;4kQ
zF9gTDMZ2GQ&}m9RsT4N-h(NJxsaM;D%CZ}GA!RDIYvzATICV<;Uq~6fWa5k?@)u55
z@=wuKUa#l0B&@%cnR!MEWaBR3#^aormGq^Od^}T)H!6>QXluc7_nMoaJKNc!V7{b0
zBc2{WLsg#yzD5(S(p03H5dgK$lF{$=ErPX@eKTbp`+9-?HYlwvX-Sr^RXy2!!A*T|
z7+`=|&^>?meH`HXa0-)*le}fb-m$KSW)cPwdnjBRV<Qq^R3)gvM35<d6#LTBC3kXQ
z|G)L^b{hd&Fzp7R7iVjReCOIgV8!qxGdTd;IG;Udt9YqO6Sk`JtJD?JJQp+U?E8D4
zHpdZ-y}nWsCMqk4YF6`RiiMULX*yCL6vatS($;^v+kCbSedt!%iOge`#Hz(#^zxW@
z)FqJ}WGs?3Pcd2f_0n}=pis7>nYe%H?AfwzncqdgXv3sdLl2}_@VQhqTeCVEptx`~
z8rbrh_t-*nsxUl6E-Lw^N;`|3A|>b5-N4dwtS$d-wEl6{P%^w$a_&k*0z}jA6bmt2
zH`;&9U6xyU+NUIOryAFBWHl9koJmDaQJC&!J9v}>J~9-QP3?7#nOC8hphq1A{)S3x
zkJvydD<mfg;{&_2f(Ff=gk+UVWOSAPs~%R4a|7}&B|^yUuDWwp2a{D$b*Mp_uZ%RI
z(K|Ho)xx!l!qlB1h;y-3Y#yycHfT(7S=E0VBRv?Sd`Yp2G7^Oi9eOW}k8AcN*z-Tf
zERTy+e4r+J`F|;B=E@6w$5g(&ppzOHShaWX<FU5@xMs-B01`EaQ;0Q8maQ_~UnRtl
z={9-ki1;RpJfO0LW{hr|F=)1n&--yf(;V9H+HcsJ7u?WV_a9pGU@1)1`hVD(7ngr3
zLgY2t3z|JOGI_^PM$bOG%1g@;I=#85qu*Jja;Jc085roq_Euef-)x~>zEYPTlxD&<
z$;7h!?-hO&?xy@7^1=WqggLUb9V1&W3{cu&zznQE?u7wgX+k#$ccnQmA$IY)*$;kU
zK%!yBnxix+!N`paGWyyt41l266U=`%h39rxP)i10c`c#G-#u9Ag8m=gJ#h3@UTWHV
zeYlxiDXL2?8)Xp>98Y_%Ec#iccxtQSsYzX8QPq|9JEo*)x^B=B?1|A_#BN6Z1>353
znr@WnHma*$v-Xn|E`{UD<(4|srEr?xW(fcJBEl?z4M@k<k1n&nrc`~P2;^B$T48|7
zl7&XHX?(M4pKxU*J)4^(-P`a!&zRc`ixP$0;%RPxjuEmxEwwF7Zn$*?($W;V-cuNE
z?|R+zOC{;&^46gm!b~-nETRoKI~dJ&{-%K9Uu@tcmqC*#69h3bG&q-WjTjS`8N5Bj
zfAq_5Z$CZoB<%AI?WUw48EUs^qZrAPCG}j9I3dlB#97hb$`Tjlr9=_mDifZ1KF>ER
zhV!Flf@4AA;z!20=^s}lF`D?TROH~%@mzu|*GLy=ZKMryU0*y%Uc3OPj^KE(;_D>j
zYbEHm$v{gc5PCsTClF-%KoZ)-cjCtef99gLL}raz$oR}Soo2gmY^XT{v=$c56n(5j
zx%@?M)=z)_xDT*GrmX2W-Cf1If(O$@;5<@j?s|9{<<fgqp+>)z|9y)9dkA`u+5u7-
zQUH2^f`vt#;VY;OWG2V;k=j#!md&c`_d#t;s7als;gwI>h+I^x8fs&phbcuGf8V)l
z;vC%aR$OyBRw=tEYm3h2MSy)@+3z(6i4R!wc>#;oF~t8tz;=QL2getH<h4OKlRfw_
zU{NC@5B>%L+j9C20=5(AKML3*wKs!tCK99Cr-Jb?5|5xOUZWh7{9B~?meU$Ooz`lV
z^hqA#D~^z?>V)O7acOxIB`gc#e+e@D5Fyii&YKJ}EwC5#YkQL=FY;Mx42Lg{P(>9p
zBsoj%>NQTj2#Ps(_g}oP1f@3MiQ}zGKLycYQY&02#dEFhDV}Dyq!LVoj7}*o$N8Q2
zOMExxpJoV8bS^v<-cw{)H{#POamK`&!X|5N>hX9jbK4i8y(E_;0gu&Fe+;ou-g*>z
ztrewmAz@$Z`zn;?-&~0bM*OaSmfTB}mVoD*wu@s<g9p0uElJqhMWJI>MPVM{sg^fW
zC_yRBMGEzl=7h~THIRmqCKZ%0VnguwcN${sLb=gMi|r6gAz^3;XVOH6dlX{GMsj4x
z#Thn|S`^__#|H8kv&jZ%e|qIwHb0CdqdLtBXovhWmP~4<skGacY=;e{Ng++k=Fd-0
z?<g7YBP0y?haEpTM#4&_!JI<Ek+Zu~Ksd{kQ9FWdb5WcoqBPtC(YUda%mCR$(gZ?J
zIVI&=MWjlSNC3S}aNGAe<0L}E=2x=EIARQqlCm8&#tEp0b;9lmf49ws3@FjIY`nvO
zb>hfjq!$reesJKRnu;ys&R|~rvSBmnmo=L?znsUr_~oWx2Y%V%z3G==XaaG{FF`Y<
zS0OU?%bLx+U)G~0{c>Blv0v6~W)oRX=`}wXgHclYDKJV_p9iC4^=;wCS-l5iCP<wZ
zQMUYK;kQ`<DB&1)f7q4)vhmSEQr7_ZG*jCt-@ap080%894BMYvY~f}TB;*D9?x&Pz
z*U%#6zjD`8%D2)V-EtYX9)(RgmgU%Q^nJ-#C_l3lqZN-}q!egbP8MA$mMJKhHAGY&
z^)?AArj;F%D9f^s>`?%@FAPb@!;Kh=LMw-nV@A|x$n5I}f5jylCMb1Nz08DlbHyc)
zgGz#p6m~dOT!Ix;km^%xEIq`I3mlfPzWF=B#pgTn<&Mq40!byja;9Z-knD7{NHgRh
z*=ff>>2m;3!g63wv2i-&xNQIqtclcSNOi`VNdB$ns-0}ic_zh}HJ78c%lU-bFqv4o
zC}^B(M;*XSf3eT4YD7Egl*l3|I{^5c{Z9y%?Z1iH*fL6odi0EClw!tWzmg73Q*)U&
zO+9MTG_`C8^`Bod{Fd5D;e1;z^(o<uWK<+IvK`g$WKq%bIKqsY1mYnnN5Nd6QPr-6
z3L`DJx%~w#DFc{OY$yd1{w<f|42U34adFgEg@v_de<SCV0ie9@&L{&sZ2L~7z^u#(
zE?Jp|jhtg6$f+q6oMFQ)wI^<_GN!bl(KZEHR~ToCx*#*#By~ZyLsAuFdlXecwnI`E
zWP2o4LAFCu7i4plKD8&*9Kg3J^>=qw!O((|Jg?hT1p#^Qc76CpZ}VYgI8wMI3VW9^
zoRR8Ge_0xYYkmq}xKo`ZNPEjkT9u?U)ltUl{cVvZ9dTfJSFkkY3B(2`{k40fR?gEh
zmut0?zkNdCyF74mh5q+;-bM5JK8)D$>+om-DZ^+Ky+Zu9?R0)~MU(@Kw&OHSH<`EW
z?ZCKg2imw)IUMjh<u~Q-iflp-nl5<ba46$gfAzo>ny9M|R^26xmqBXH{D^~-W}t7g
zduG_5`?b^)Qmt{WQACdw3>_GRrUY(#&UhK0%9H)A_mWQD6h^@U-ladR5o?d4_N_EV
zRXAoe&Z62-*Xb$VAJEN_+X_il9Xkc@%TX8QtoBwmcTtAuQ?Y18FCdOX6cEDRWUtnF
zfAB}5Ne)b?m25&zoevj7adH$;vxaJ@&l@UfkFVwYunvZe<k+lA_+_S~@?TNk-05&+
zOG_1ruq(}1wwwu~jU<p3hp-T+-Rc7lHUO)eadWtKC%~N16w)r7no(^veQRGZh5G=T
zZ$?)n($XW0V<_4ieCw->t22LU)Sc0Cf9o2j@Np&X*ErMlO}OR@k@YQrd`l`WYz@EN
zy9Jq#l;Q3~=CfIsm3uv_(E_QSceRwEa_X0d^!Lt{tE5q-<M6gv;_X`r+g!saPIZX|
zPjC|x3!l7Z7N}fRX4-UhqY*C_JFnF>F7eqK4x`Rdk4_ji`n=vsvh;WDP`yvv5Z<!&
z<?hnZhqs4+0a(=$p_8#X6aq6hmoZiiBbS%NJr9?`*a9Y(%)~tqe-4Knl0_iEPU<v3
zf&{i(v=t0PlFyD3MJJrjMv`COZ^&Kh6x}%|>2wufOWfUA^5bwg^UW+po>MT+5PnC)
zr8dN5UmJv<QCo!!yDC+p3Do3^iB#pXGZI2=kx6)6TX^C;J<%o`qR92GO64Yijsw*J
z^+t@Xt7J%W9W!S`f0?RmVdGuvp#(Hl3QcD`v0@1>8PBYv1%hk{RC_4K$MH2WUfRF~
zvYB#iX5&k2^D3C3B1LD;u&)CLhk1vPYSSBzsnjML=%$Y5g2ALhy{r05O}3Z-Xqa}w
zv*7F8RBZvn)h3z1oNLpgnW?@eCInw)83N)f9EQddbDbMHf6LN(z}-?RNI=Cx)PX6P
z=%uOzGmKo?z*wnR7czKfqQv@|Clgby6T+CW_=*8z3RJD5`Cx={lZ{wmU#IECNNhbY
zWF^;GFl5sR9oCQ<Na#Mfs_GE4Nm%;YLo&&y3KBOGlw3zECPi*Nh+0vd5JNVX3KHtd
zjw?vmDJ-LEe<wH?Yf~x~J;!Db!J8~nI}2=b8U=#M#YY`2L{nU>Lj*B}szrr}yFB8>
zU2X#SSwgk`fQ2mDBbt~%ktTp0R7Sk8l``TD=*KyBbvWE{Bo(~coAYgDBJQ|Ru~B8D
zEuQ{#smbVDt~42)61iAsqw^<f>K8mEuDdCy>Iq17e_RpWdTkApzS|5R_N47<maf$M
zrS*><sBoQ1Zcl5Ubrf7w8GY)T#Y!HnS^LycVxT8qd@+1=s1CxzvAh`m{h$Bp#4HvB
z>rUqU!`a#EgM;;A$*O(F%r@q&L@6Ze(est48NQsK%@;2(j!tHV^#wEhaI{>`7U#yd
z%fJ7+fBgF8@@P3T)NVYVpD!`UbC@x?9O@{~+Z4H7z<!BfqxAxtk6<(HLP6-}X%}8E
zT)R+bP^i9ha4`I^n4i3yEf0quzIr~qn*F*QXZGs1ix~jEp8@yza(2GFtp2D?^~LOR
z{$X)4yBr;!-urI$_Vno4{MSRh6!;Gq%4_tnf55vch8?P6PtVWiV_&gBcyid>wobXZ
z?Wsk*<$H^JEA&?EE$OY)TU8;i`<yC(;pK<prCxn|dj7K+p3N6;XN!v5y&nECd^3D`
zNbTaq>|}XJAy{mdn4B#bjUz6h#+aRjFM2&ZHR%22@b&!FeE7<I#&E4!x3_axHX#_~
ze_sy{R`hxo4Bo?75%HqK;Oc!PT5P{a!8$K@#ISx?Y;!}?)V;S68#83Gq7EZ^Ti_tE
z`7FiPPp-{kc-nhES)R_%f#;tuzR~}m{kU9ST>fns-k&ai{BUef=D!R-efZ_#^S85)
z!*@r^AHTm?egivOyu<#mKb;N7XY=FXe<NA%EDe{7li}Io^zHlE<?#LSPr!BhaW))}
zcRu2JzIgNQ!+9O_^mpyxFW;umzx6ujC-+qRxcFWYCXO&3i!IoZJ|zp~?`!eBMW6!7
z@Gt@a6JT3FjXaTcF5Lh@i8v(|P7L@6W-;9#1aq*&D<D#~h_stQPz=Q<_C6xYf0PV@
zw7YG9IDu`TECtsIZm>j)IJKwc2rzyP`ckYi01RT2M%`2_=-LesU_fwj$t9ua;6aFF
z2g{MfiV`drEutSHuT4(l0Tgr7C;2E+K#K~2X+>xw@V)#2yibuUcaS_h;$!j#muE*O
zKTm#u<;oA#kX@*UY@`$G;{S@oe-zXM+&!qoLj?8Phwe`ifuz|8#qk0g)N#5?xDPJh
z2=|%}+$G#=26Pqfa}W2WhwRcrcFDap$tlyFTGU&<x45@LZ^iouelSW)POY7-GWWXY
zpiIGL<4|sT8ze(sA{B&6c%z8C=>8jh3_u$2(KYf&bvCT&mrcU^@%_kHe{G?T0cp3@
z6lEiZhjINa2*N6?CT&v?gqhk5F0$7kIDtT#fz`+xL2T#xxA(i~%Q{VJM#syVl%)KB
zaJ~OJTAUsopIv$0cZ=DBFfYOsHA9>XAE}^4j=ERFpD-{(-7Fh1q>#as0|}NKYn=m?
z-5Eam1BlcOANFUF4(!+Pe<u)=;co)5Yr`*MYyD=ldw6BdupfoppPgR3xjcP;UX4G%
z`XyiI8(In6qUdm4tdR}@&n<<w5=B}nSdlG6IiXkYw-E)e7J3NIbB(fGBQw`H&owr5
zo!P9iA!yVi5xGC-;hQB6Egs&9^YBWEL(8NiGojd9(p#ywa&L9^f8#s*@tytn&VGDn
zKfaIQ`xp(?P6{9a^e4wF29g7kfc-k+5dwpYgH>#V0GC-37#o&OsVCAMzXKu$)<;+X
zBnLL-WV{5c6)8|sWIucI!#)LzBxJ(C@P`EB--O{oQqQgk<hyFklOi}L8`l)09z+qN
zO|m#!^fL&7nrl(Qf7zJJn!xPfd3_crj`c^0y9mtK2+9forL=D~AZNZ72xUzY9x^PE
zBt(IfXIIJ&iV=)}3~Rpvfn3~w0z(Q>oe_LjOt2gzbh}W5(jsLh#k#08DcjP1r8Siy
zpGr<1{}@{{<V0x~^^gEYzJdWsVM`M>fY?qEw_xT7>|Vdve@x7rz}oVoVCIX<`N^B*
z{LRJX+linDw?l%iHE1bU1RWZoyk~+wVweKMCJdhn43z@$pOuDIB|4vS!RF#0{Cm$Q
zN<&wa=a0;&K9Myv3Cw^cq&3v)8aEUD&h4ggpkfYmCu$YLO==bW-KbT}E7Zz&^5i>t
z@|`?!Ct2L(f4IwWmkXU_g-)_UCt0DBtk6kT=p-xbj${e^w;nARH<M(tF6?(RjSx7b
z8s0WS5IArG{RG<-U)aa{ePcqnvZ9~sM-M7MgSCG&9u)r{qP_u6ZzZkJY6P&9Wxup~
z#4s)SMhsgm`5l>DUG6JAV5Sz_b+3AuFic@&6Nc^be~Vj?2X?ip(#ZoS3u5;eOkXX3
zv10m6J7f;XTB3uq)317ZV2K>eF6@UmV_hwJ?I75LLS(-Y1olJr$R0u#iNLaxPl=-<
za<te_T@M9xBJh6IFhW4D$fsx%1acm99cCX>c7$E%6Yoq5Qml_x){wGhcKg-B2m$l6
znB7JQf1k)5Do2dGRl)C?*jUBWqeCDxTZn3=6Z|7(ovWnr8tdHjsq)rzGSN*sDdlc-
zvK}{`=w#?LF?5<3I!%n7CdN(^V}C0Bsr0APNu+cVDV;=0Cy^q%_+sg-=n+WR29Y{7
zLS_Q9LKd9&6H0?7Ls>e6kP+kXdp(2)DNt!ye=uViCkSwCR*x61hwvaDZ6=pS*v|Ux
zCKn6W2}H@lb=}0E17njUWLL<qe9HGD6cPSgi@=>rJBQzo;(JLup|NsIYxQ`>y-8%r
zRuAEJGObmaz13z%5;<b%U&%LM*lx{lT^|OM1e3t}$5c5@CaMJrgsS~i1372w^+0>W
ze*jNtNzaRIz(AphcS-kP2}O_i?t)k`W9w@&AvV<{3u~aK_rwOinb#DMIkf|Z*N6y%
zj(Hbe2)<XSV-O#$NZJU35m%GOMzGPqQDQE;SPac9pa>2~u8WCIYu5G85o}mjqysvv
z1=w9ihEpP<<^AfS9@bxvZI7M^-q4o(e<-s0-)D>YWv__(KYX-J?$0WLD-!*Xcwy^!
z8%d9a(7zdZET2~Iek?q+Q`y&<SUbIadnT4w3wzR;m~<wVdO0ukr|IfxXJYLj`EE*b
zBaP@ax&4u;sUkOVDR_7N3!L=`Bxn62jv_2c1WW68m!T?!Le|@L5cM<DEl%P51rN`z
zp;VUvTRs(+BDpve0x>X`fpiEcf6K1iHVob0uka5<ktpf~Tnun$k}1$#gS#ubDw?da
zX#RguN}^;-w#S|arpdzNv22U4LyDv<hi`)%etS8-zkEDv9Gvh*IT+r*4vKC_`c;w!
ztr?tB!fG|Ve;@vQl@cr+M><eSt?6*|-_zL~K4|6X2}MsR8dd7>!J(oNe@!Ero&iH=
zNe9q>$75%H-_rZudKqUp@!<}Cz5o6qnOP}<jep~*zxw?C^6yJ!9kehu4AwZIq#FMI
z{t|DXGQzqr0EnP)`0>y1_2rL#)<Ua43rSN56MyL6k8eXa@`aTdBq>u`YfPg|u)-KW
zSgVD0t|${piAYK@q##2Ie=&x~(sd{NHUpKKMd8u))Di_!t05bpZ)X2^_G*!DZ-s(>
zBDlzQwsRUIVo2BcHWJ(+#+niVg9zHrlt}0?DpI<HmMZ89+6!bL2n~8FsUaMLQrqUD
zr(mw;GHMQ>3b#66L%Fu26!Gn-w<OWqNV&{|i@HtHmnR}Oz_az7f1V1J$`>SF%2mkE
zRSGOIo~N?pS?U$CWx;q4q;RCeYooJuXQN*#+-6qL&I`ca2sLEHkf7G!m4!|BSIu-#
zua^*-$t3F(D@Z<nViKlvn`JZ^)WZ51nkXX1w`Z862pg<zyesPa93!f`U~^+j%#C~+
zUqmjWMS_IHGruESGOVdjjYO3<ZG?9?7#{?9&%7O_>3r;dn=3Q|z3V7tgE>AtpJ5-m
zr_Y;kD6YE<YJRMjNYXtpe~;}IEURI`=_I+OmGN$wtS71!JELTr`8-p61}VVma2b!0
zDS5h0@kppZ{S3u7lztAy$6PXjnkU<<a42zuqdwaE4G|;1RnIU)L+RJ>0N5Fo?lFKO
zd{VOi1Oqgber`|LtN1HCVk@J@poEhnK5d3Z5`Dhta>ip4zlqMOfBGHJyt<$Nj@o2R
z5<ry<+)+y4#<XOXJnhIMA!o~+g-w^{(<HZKJEZ&Pn#oP;1Y}*Ex_Y6@;d0`8_<*2a
z7i{cvq#N_X#KDmhg*<oR)v@kvr@>m0Ota32s8T<S?-#HaOeNZOHT%Vw$RJkxac!Y}
z*rB`V2`Js`sKunae{*@?Y@sd4-jeyfz7A+?#WJ1B``dYTowsVfzRNnSvqnm9Z4Jn}
z*jd-{yvR>XESR)VtkY`XZKB`f1#XwUr*N2Uw`gSxf*z{AQ#6Gowsf99Luy6XUM-9j
zP>EyykSQEDs~*%rIQ5LxyQMUb<9p+^W1MF%xMjgaK;hsSe+y<y>1LU=p2Vi)!3TE!
zmT}UH_Qakx)>}H;AR#lqcxT#2E8vwNh@}eN!XRfg4B$mTaT8U0ASE{<7YB{-p0*ue
zT9p+FVd^SNU63q#*9f_<hR##gGWzBkdJTBAqzW&6r<=EPO#&bsOkB`RY4F*3K;}i|
z>eevTi~_Xne_O*;)OZcowT4sMccp#GZC`Q>G1f}|dut;mqHUB$(uyOPA_#P2E)`RD
z?y3rm4chXP+@~3tx{V=s@z~FnbJfl{$rRiGrXoJg+KgQ21}jSr9CS}b-_YcCPFRY;
z-Av~5^^$v)-My3iyuRWD3O{b9tJ|_zDWhrSxm$r5e}h^HqiPT<S^PE9X^QbQLpRf-
z?W$u(NA*-v4ev({aNfDgzPugM%~<Vw4+eu!a;6c@tMQl*?&Gsy;qiP16c+p<pV#>c
z#N9x&{Y(*eRxX9&omC<PC(zXHuQmt7*Rht!OYI@&gsRqCSo6H5-mTf=g<R6TuGOFO
zH}?nTfBODri*LpiHhJrI$A6W>^nY4Kx^<Z^R@kVWP)7B(nXC$HRz(bH<F$Kpy72MB
z{qan<FX6(=G}E*?+}P2miB(M#NOsJ~;Ssmk-QLQO!H_ht)?Dqg-1h0@Yst>1e&43r
z=*U)evuavTcR70$q|s83PDa;jsD6i1tFWi?e=_UcD1N8v_X2G*CUng%o9}%m$>dAI
z8@n5Z5pX@1jlOHortL6{{hBT}VtdB39kF~>OmI8)R-?bl^VCnbtaf4N&onwYjgg^h
z)%7u}?CP#Y%PyvrnCzOEY?jyQpvL{rwW=E^zd&#I`d9PxhzmdCKKCY8I~RSAE_oMs
zf1m38(8@o)xn))Eq=a(!VSD8ej4)KIbb6SzpQpvfW+AsL9#W1XK+P;L&Bz|{Er2&d
zKUTWk5g#p;gLbCd+-<ZoeEB`zF`{KTi=E+Hd9rWQZEpSCUG(hBz4PD&jdUd9?!;^-
zhj(vkK#rN-xt0&BEptOcpy?^>T=W$|f7_d}CG2z4E*VzN)9Pz$e+5YOsp)HfswbV>
zs{M1eUA|ZvKVK>%`#u9x<UwSEx|VzzyYvOMp!U@a#tN-NKjUb|>#IDjX}THLVz<+=
zebzs{{)x`j{D9O#X}6OfkWT0@<Nh%%m%Jet9@BCuinCgP4v#Qg{#LJ>aLsO9e`|J6
zKhkbMnLB}XA3s1oi89R8v17Hmh9~1wSlY>x1*)*;JgEt0H2=3;O+$CT;OA|1=TSYT
z#MiOuhBNa+7`K&5VMMm?KdoI0FFH1~sEFxiU1~FxKtz0ipLLmNtsj4K^L;wfG~GEy
z#$8t!mkwfHakaFz$CmfRmpesKUCB$GnaTLXVRJuYp|5p?9NaC7XKkThWbbYNXOw01
zbU<qg;4q0m=lRzZE34e~dnT)?1~0mS;354TTPV)aGGKn89TdY;-mo*Gfxu*2k#F1>
z={74UA^Hz#2CbKY#3mC2F)}qWmr-;N6qm$)0w|Z9*gXw@A!qmi0t2yj9iX=cdn<Y>
znw)ZI{{NxFrzA?UrM2<^+5;_{CWpg$elwEYmriycPnXxzyQ{{|3P)g7_xia5{zLL7
zNS-v*SpZ=)bgy5#Ki*5J437&Blu~ORmcKiG()me);}yUwfM)^DPZoh^G|y;Wqlll9
zhvM%x&t3C>e#85naWYWo^Kj=sUw=Ev%&ZjNgs<Vz{(5;m{d20UD=o<Q&S(<gbobZS
zQ}_dcgt5Lu6i%z|+uz;i)9)?Sa#r(<f^xuulEx5C1f`51a-A_+Xlsk0kOhP`50{^O
zab#f-KM8n*?>tYG*Y|louZ{op=O9DUiA>lcFHtUkl>y}ITIvibf#9m9g0yXtpg>+)
zKfZnGnwd`@2Wm`3I&7)Ia~O?2SjJMwKtoZYI+<ar@2*}h3e8ACUbm;{c;Rx}$!V+N
zw^kT{mJ+?}zqx~G@$Jp8vzyl)Ub4z(vI@R|w4mPnJ7^}(_1P>wB;5wp%Ij3O#kddz
znsR`D?Dt<zOW{B@gpSs#_ChOY4$(e)KoLi<Y}%$=!@i2PAy{?$i2Zp1vK2=(KbOt#
zy4?he863`~M`4s(;&Pvps1U@NLSshPhdERnQ~d$_X(c?g7Y+VkWae=xO0VXwk^ne|
zX^xDAF*+l@&vCS<n$9t65rq6@W8f~CESf-nU;Mbt2J=N<e+B=&#Ea(rG`W7*y}Wa4
zA{gy2XW^J1T!|)I&W+#ctY9RSwVOmeWChM7uDuZov6c8J2d(J}r4wxD?ap!fWF(O&
za3H*)ZDKc+15gQ!%dsz~-|<;;EaqolM8a3cWDa@8R$)%<c*XD-aX5|yWcV_&B!r89
zq~-L1B+%Jrb@ZR8=6<3?Ia1LjD|Ik}!lxc?h_*F?ID3tlVlumI%x-Q1JnQecz%7&H
zfK&EV`QA92vWvn<OLDTpXG6KBmEnv4J5f=#fc;!f<c&pEeqxf90yaP_>yt=m($$nV
zI(m|h-KdAcBe^y}g?o)|*$=K><y0|$>2%dCU7VF#2%UBkI%zr6`w(8XIx%MghEvsx
z!#O47$$4hpTM{68PBhS#1Qvwl8s~ONkn&;tpp3RAGoNyqu#(F&l4D&6@9G8dIjbb+
zns8AC+h7M7p|x)*)(YzlATgt&h9ban__AE^kZD(|^?bCKA0@Q_Z8!3xLI&}FGwzQm
z#8MH~-!8<8KqMQ5(nl!78Nr~vA-JZG3`3jMN9sdK0c?cZ_)!XZzn}55N$&f3Ge&q0
zx*K_8k5`;>TU9-0*-+w9m!#=#$nkNUuemt0b|xsFBeoil3hP*c-4q^lD6t7zs!K&?
z4kYM2lfhY)$x<CY;>9L2V|&eiK@Kg0I$6OSFb|ON%Q6-_KN$(DwPFzRQgt?(A9-fT
zqnTRsPbw%DO5Pne%yl#}W&^C<9>#dl&nu$5WaH7f<R1r=Y+&ezwm?EIsa1w12AcwG
zB1y_!1)oPhvP(a}q$0Kf&?XTnpGYoRQ?-dAwFB<4v~GgdUT)eq6hrNQA6FC`H#m2=
zqT*c23N)ndh`q9H7E%?PU%t1LYe=F6(Jf;dwjAwvn)8-3F=TQ6Pl8?g=hMg460kxn
zrP%jtyrB~A%RS7jE6}!_QHE{+j`QcElB$xqU5qM;!b@y8_;57^;p)XIY)CVE@u&`A
z#F!S3s&xF(Eo#jG1@RDnV3Iq14aGHNk~`GR>Om>VFOlAWSW|<r<>Pt1XPZG<v*WEj
z+uB3hfmA}1OYPxVg{noH1NEm2ms#F#)L+3??fH=1T9%?S$w;f@gxU|;YbiiR2)&<o
z9jr;~4f1;klZE#nsAhID3OEa3>}`EY9lVL$zCPtbk-n&pmA>VFUENIjg`*x}4aK#A
zU`jKk-`#Av*;CXUwKz65(#cU`O{T#*teeSnXmG5^P7cAmz|`ywGi@ml-Va`o)6CE}
ztsp0+zHHx$9M`05%RHPe=w&v0E5Q=Ob3V5{zea(KOzwE){rW4R8oRmcsW^!_uq|U7
zJ3d;5mmKz&=-VoP$2U-BW8ZeVdg$A_vBOlOsNLM=B||Z|;*i<+M*X}IZjNJ=n($@P
zRZRw0eii9V0tC9Y_c!khCw|XQ@cy1IW)3y!S>r<;*m2`#)W^<upby(a*$AH<k9QTj
zrBd_&-ev3|IE?&6ygTLKJL27NpyF!m$KYKBtX6{u;@yUS;vKx(tUeI$wtU>jyUmWD
zfOog^T=4F3`YU*MoBj&kZDlrjAG{ssP8tFI;HZ=uJrx4K1v-K?oQobsj+@?eABY(3
z1Tosf5F=QDgpE+{0mjg;aL{a8!F~aQ!KgLyqp@FmeuMoUx6srloEqsS3}~%TJk>C`
z5d#8;2#fwNm%xAkENa~P?tZ%Ze+>BF#DLxiJ-iHf0|aF60kS_40Ztg$0RS6N#-Q(C
zx~1}$QNc7512Hl+mtlE86ag`pVHrXvmmT0engRZ|xZpi{0e_A@&ss-J@jgj{aTdrX
zHOTFC!E??s3(U>-y2$Ig#ZE;O^#L*ORzNSJ5`No)Qw(dz9-}#Gv>MtrwICp2oLk7F
zH?<(voip45y%cm7RtfCcgbksrm#caOTRCO>!EFDiox>0l_;8A2R`lw{)i0mV>IT{i
z>D6NW8f=5a_kRd}G1z4zp9hI`_|Wi0u%loZz7Ho(@EcC9JqoM+EKtepDyJ;I%;)TU
z)trt+6^5Rwt79{^of?4B!daSF9s#--z>Cypss=5{*<o}n=UPxaU<?Xu6r^I%I;NmO
zN=Fk3zsyrMY?rns>gBTW*gDoXlZr=I0nn$I=^x32k$>6X@;qJ8arFumQ%!D|Bq+uZ
zS3TBMFYZH~U-X`9=<}0xUA=SgtoyED+s|)7v|p>CJx>g_*V)#cNMYx0Zdt5>D`u3n
z*pE8cZhylW_a_*_qH%IZb(KMf!i*e^LX|Q~Fha_)a$Ima6RQh}N%hq`Tk01#R($rZ
zRtQDSQhy<s8oL+nl-ZI7C9Ivrwd&|EDSibMV$s_JRaWqwTeuVpTOq$0|51|gk_HK$
zPc_FY*^Dd55^DzA;dv#yBFpT`ns;JX<dh(dcVt(?6ECo<Ve}o@)hNehb~ViS0qknO
z%!OUuZ~qFr+P8m&U5%KXN~}*9UCXM9^HR>>WPc7dc$+l_5!qJZ7pu$z@69}jTpu?t
zF%Q(3lKNI{3fgg%iHm3ZoFEEhm^6#}yaiI162{9qltylc!cdKwQO1b+vBbZiuX<_{
zw2f?uIgKdXDvz&4Ot?Z<vaPizUcsxl;t<NIQL_uYir8rZ#z|MuD$<I5ST8nxn^rL{
z!hb9#U1C)k<;<!YLZ#^H?)M>W8@K8aj33foMg`2IkUeggb%s$9I6nV4Mm4TKFsj=F
zUh>FwSJ_q0IU_6+(?QUk42#8KcH@2uhNZajLA#e>%}4*wu>RW&OBv18=g&p3<}!Up
zex*1QS-6E?ap5kUzbn6*m&d7$dv2v%7=JSNh;*HD0IdZ?@jdKHAlE9xmkYg5$LtLT
z{W|u>IX>mW>{zAJYQgR8Ob(rniSB)v9C~0zRk!dqY}2-91{Pglz1Mmb!Qi}a42^<r
zksKAMuo`E)nV7xdy$q*E@B9v#Z+b5yIbit37DhcV*R`juceXp33u+T7`gQM2dVf|m
zPqKFl;x3R$H_Er)RCk7Dh*Wi4<=DTmNr^YTwb?9!*4|cZzeC}T2fJ8(>pjMqn8qnj
zEx`JY=>-)#uCh0z7v$R10lag1xzc{l7uXD40DPvzJLM2?C=m2J<q&X=0})>DfF)Pl
zCs9ZxfbDINW-C#Ik7<}n4l{1(;D42tl57p#=<=sGr;z>Hlyor6xS_YF*j~`ULWcF;
zo(k5$Hn+0=9E(PO0~2~{iZx8M^;P&V^WzY<kgsQ-FP-H+-_Yy>tprjtsKX<oDFo}R
zzlTw?xBZl;RGynNP8t&(s3(}EV}rIQg0f>W$DzNIsbhM245Ye~M)%U(v42CtrL6#|
z28hN6iSA;1Q6`>0W{aU+$y<f47wKWKtWm+WLoqJ<Tv0sZ=xG<s>`8WX`(x}XrdsIO
zk1v%_RHvcb(czAUr>aV}7Q?cQQVHh52)S}$GI6fsn(MTlhO1og*WPje{YlWXecv+F
zbmdNIi0Ec-vGs9k$mK6FsDHGt-u9i0tKH&@-=Ju@#9}=_pRb)bwFEPoKjlpCF`SLW
zt{<El9@(qdCR39sAzS=>XxoHjqAWaBOUm<R-?SL$aVyP}zMorggzjyKsn1HC?=Ae0
z9Z0gqgHCU#KLblaw|J0krdJAOvZIiZRv1`UUw3uo>g)s_<PKhb&VQz%&eql&Yx}-3
zSZooMtFX^CR*fmwC|!%HZE$b|^E5P}To<TxHl*PrvO?6%;v*?@esc?HF~xajt8aH5
z@0mQf3SFRbU&^O6u)Di)l@<g`AGC(@*ww9fs5n~tY-*OvTG*}#X6m>zWw(7A(nalG
zD<~c7`L5-cWowq^J#AAv7?iymS651Svcn?ZDeHsREYy6dGqIX^l9aRe7~MygT`#^%
zI2Zcs!>5PGK4IxIAAKYrb*pQZ<gZQt{NeT0e);aTp^CVg1{?sqy?wZAg9-VfRm0-n
zv6o>;sIGpz+{zfw%+mh_M&=loQNc7512Hl+moc6J6q8=ZD1W_L$!^>@5WV{=^aEyC
zMHUypXu$5I4a{u}-v)CUOmfO$^8be^ZjwlCEm`qF?lwi%#^S4mExR9`>|P&U-XFf)
zl<ur>1QxsZ?;Y?9$$vreprOtJ2&15T|JnWZBqbVtUii^cD#efG?~Xs|@~FV^4B#2S
zqX3sjtAR%pkC$cXJr;lG`SSPs9}hAM`{m|fO;TgEV0nt9fs<w_i8<+)#FbR~ADKEq
ze#hL2CDT65*CvetB#GlYp0DjCD5IbA5|gIrYD#{e$;a=QND`(sWTR3`W=H3R_t+@&
zd<*6Q8#cgSNbx7R%)?|gu9*i!(?`PzZ-VW@$7}Q7xA%wt9(sQZq7r0$XOt0M+wPyA
z58;Id62|%tQFzeZuYbGm4}Z2)^X9al(X--yNn?nr{X!#%TxYZrh`Q|8enavGzl3#J
zS<U48Kfiu-&B)v9ps$xgS=~wiU&B-R+_1t4=Ty5QoPsEKmM6-uhfTpbom}mUQAe44
zD}4PrxTL&Y9*uuYM`*W_N_wHayF*DSc<-f<fd*2dT8+Gg!{_DPX7ZWmX{O~8V1cel
zN$uv2@syGx!3J$9(YO7YD=(1!w$=Q$xV=hSVV#|5QNZfDUj`%D$zdV`%=0%|I7xfm
zOJ%4E-}!5N<aO@s_qJaE`vL$ArRY_!KV7tb8hAv^p-X>+d1821K7Pvp%ln*18oVp>
z!U6W-Kfih0`!yMq`<dAy`c_c5r)LKIo7k(+brHsklyNkp@qB1eN(t$wb}_3Mt~8sS
zbh|1^SJ{Xz2-&`Us2bropZ`_qJP6xziR(><%LVw+J#^L>(|ga7X1q{n1M@t4&2-)Y
zlRZYZmXCjXgJVy~7_k`QvwxB&p_J<_n)f(x=_xFDxJIF^Z7Cd$l*BR7!X#4ip|{L;
zHB#OY8Ofq>=sy-2qhf>lEVreO3<zh(gi|~?E0o1B7B_qv%4X#uwhj#0r5^;ULr4e&
z1;WR`ob$9r;gdi%kzUK0B6cna$??P!4yXUO3E6)FdMi-z4|JR)_Q&r-kgq*MoWZ*x
zKPI6S<ijlFLk0P1&N}3;0RSoxidtbg7hcV!Ih~b!=9^U@2eu&B*=jblgo%;X@gK<u
z;3(d#x+g-#G1KGH79MI#0zL`*QF0;Z7kC!OSjB#*u)lYS4eUn?LOR{@-GTk3rfMkk
z*w23?I3jRKK*pJv`$|mN-RzsU*X?~FE7T~5EjwRIO*4}f^SNFrZZ#hOZN^pjwkxrb
z_2Rj6<!!0t@w)DEMG#;WJ$!PNjdz}%d7MooH-p(mQfD~gyGH*}VXjzm3eY)@2S@V@
zWf|!icuuIqa~Vf66?D<BtC6>ylgt-*CFXx9#~F_uQis)ap(!u71?ahjLP*<|8|^PA
z!37X*2_4IFc2|F>8F7?pxvx~9XfUE{v*H64C^Fh<GqJ!z1<LhuOlVtsEkqGI-jBFX
z9>vGZ((e76>pL7ui^6et1K_G0&m!-dM@flNs*^Y(Ifual{5VA_S!Oc27D<c~2GoB<
zjy9)U+@Jds;}w_RFcx|(B;!#dWWe|c*3_p3_r|N`JkeE&(PX<qMm%zpRN6lx^vG}X
zb8S=qFbhYC$?#vvaP!EhSSOv`;hM(rX-IQ2&udD4qLY+nMNWqLn5dUz(RR={H43sd
zXv_!eCdL_Vr<4^&P3(D|+ZCTn?stF8?R4m^X=;TG*(@scG-RLC)0%SCv_+#6yJQXm
z&bqXlxYzM0vWYvuu?+i~A-a$omsNJtq4WD&dg$68t2<nBsr{`;IIH=<qj0|DwJ(yI
zKz{=6X^vXXHk>97*w+^-KbWflJF0+YL1)MXM`>dyi0&!r=Qlo)+~||!az208Tj%x<
zc6B&@qXa5Nm8yertzBd*{1!*EA*K5aKN~ZuT&BleH&igIg*A@BCL7|I@&*JVeTMcG
zr&ZPN4B@q;`zz}f50L|s^ma!j-4?6}CiZYnELdywxs+gX9{|moY`Jy8reZ_YN{p*c
zIr7k#9MtUb!ZY}ww=`*NbMb#mr{4gF8cn)0iXBj1D1?^cLUB<VyUpz5L1iwY%(jvt
zw4l(9%PVE~YfBhgVhy#EuXZ7%SzfhV4qWX5gRB8)DIK`l1qbm^x0>HpZtl}ve{1=J
z*BvX@1fh2U#y8FBUD|1e0M!yYdt1o}#VLNn<<PdOfZr{pBimXWq|twTWm{8wr&Zzu
z2$Fx{(TcFpE|E%rvXebOCq<j}E>Z%axq&$1Tos+)<y*C|A8@UT&__J0R(spAY7I{C
zt6G2!x2h)Z{Tax@tzozVM}o<%;hreVsbw%uP3deWDzUug5)m15hI4DU9cl4_G?CA)
zn@_IxR&Pnj3yrp=JtKenqI52#@6b}{?@2Ra#B_ArKPo?)NzKUT(!CmYvPwtJ&EMcI
zw|Zx9b7y+;%UvZMllU2{D7$&<!qT)Wmov3~pN`)(hZ`Q_rqZroHfwN#v*mWxj?3n7
z#fNI@m+P0!TDjQfZ~f)SW%I%nWIivB84bUJz=I9Wbm<sp;HYELGIw?UN`=f(ceq|{
z3*@%QttQ7{6dk|1x2D-Lsy`Y<N;T-GZF@B0Xw`A+bzV4f)0>3(V>Ya^BPfWD?}lv*
zy`8+Kx9$pW8d8zrXQ<T-KXek)EqmoIZN10k<o}OtTw#|1TRs&5H<z&x1QVC|@IAx<
z*SEd#J){T&)sSeHVHpe)1Tiu-IF~VV9ut>f2?8jWfE7Che>V)i=U4a-tSV6=B>{nf
z#EA>^)*`o}r=snthxY#;N<C+!d3x+@w?G$gY|9#o;`1Y=m8U;W^7Q@F>&Me~S8YyK
zI8+u+AHSaz{~-CVk~~OwvPuck>h$sF>E~xDf%x^xFC(Sa{IdMn@jITsXytf@;u(rZ
zl{$a1M)8Q|e-X`d0Qp_=3;46mW7qsV@$-(H3=}Tqb^hh!hbNhtmBLf_7~cAC?;lTp
zJyq603nYIcv_eUB`t{GJ@Pkn($odn2@XDP2{_XVp(~mvXvR3~tB$E&r{_v;2|2TCc
zKV@Nh$-<OGh>b9T1fn|;X`!tx!Za*I!$Lt85VAlJfAd)U-fe!Hfw~$Q;f(E9iz~2L
zk#%5PL;bs}Q?q<K5(;`ku*x?7?)jZ{ifR}$VPQyP!G;jd!8F2G^nss5CXj5VVkXaH
zCXji_bKlc4@|w(xWZI7~K^)~zlu|P?l9D!7ST~W9Vv6dmkxqSrSlp;?7RT%!GQZne
zU9u?ye@1AU@NbIX?V9scy)QU_@gT4Ql-{?~5CH7d&T?3Yqfoz5HW8g)ko9ltjF&-b
z)qujjUdjx!WE)gx#An77&1;???Cf>Sar~unYgSFT&Z7LBDZEWMz)Z)E<<PDbZN8af
zY+7lcWs9#sqQJp5Xm_G7=>@o|uCVx}JRY|Jf1S&z@z%~?ic8AtvS`MNkIz#US5)Cy
zQJ5nWeN&BMTcEZmj2a`cl`^R33Ds7?1Fu0iIjBvHTN@6>mO5c6)rqLL@k<d`CO+>u
zMk*m~LtE+BtV?rEy3N{SH+P%_*(o)WBYr0@Aw_}NUw-<2aZ|+t&?~`_t;YPaqa0tO
zf5p1l=vERwTHXPeLLnsVsOl`-ki4VEhRAjqnNV554>2K(sYw)%fq!t7JY|>Y(_BsC
za1g|LE~c@RFUDNQM`eD0E1a)tRa-Bdbe(KouGKd+?+VTHIPi_sd?Ri_t%W812;34e
z$wF?=m;?=HnnB$Y*&Y>a)k%!&?-~Btf1C3(fD=Tf<8viVgJ#%gb@#xrEFF*q|9edv
z%b>;B;?B2&jv#b}{bt^lF^9%zah;Y5ioP6%Jsk&Xesu(%1T(A0^kj#y7I1tN-Fq67
zmpR+t2|G|~;W$E<_3&&=F(zAr**V30wHoUcsxa77JZcQ{)}D~;zDNQD97&*Of2gJ|
zlW+hyUFw5tMlL1?7Ye_c4g8a5loWj_6r<Y0Gnv~a75kLt^RpWXJIz|gaYL%kd%?no
z>0rg$ZdR3x0iAj`QSK0Dmb8GHgvE&#7EqOOD7>jCSla<dMp;|+JpKkFV3L0uC$}+t
z%D6c_6SZ8>?rR-wL5&oE&OvZ%e||ZQI=TB~bOkE_%gqu^_Em$*T-KgR+1I>_w`7?s
zS$=2w-HH#|a>>K!s$)E$RfwH~If$Rdn8hF6=~C6S=0Z_*jiFXbn=X{pzV1S^re?M9
zu7c9Ymv7LnZHgP(E~v(s*^L}=sT?`rh+t4ChN&I*I~);0i$+hLtOeCje^j~=HoRq@
ztCfIu!t06DagyKG!6AoqwAfQoB{MO`TFG?=NyvcAvdqHXN+OUMOI}Cre<POE$T0yY
zJut5Y8NY!gQ!AqMBZlTg8QItKGK^+T#SX!1u#`fVv`<j`no%q%9kkUD9H($}Lob4)
zDo}-&(JDLk%X&{K#P$Qde<U!T4T(C|)-WpD4EcU9cfdNZC=5Z*!91)3E@!N=-Q?wX
ze>=$J^4EB`j~lIYOvW8I`)l!=IAbFMPdhow4Zn`Ff4AS)`O{LnZ`*e#j;fSEfJb_2
zM{!Ls`&F2m(j9S-^IZ*Z1i@r;_jsDLLi@@3ORg4<PYg}CeVng>e^J7K-OCTBYGSSR
z{CtkLfpY?Y_VVUIn5n8huOo3=cF4Quq?+9bFzvn>S@Yn~n9vi^`;1CqXtXIG&d_%G
znjsQiO%xj17O8d7;!MUU90E_g4#NZe0Cy<f*b-dOFz-t+h@UL)26dHiQ@<bdirh8Y
z#~0|(TXS<_=^Gsnf3O3t)i{D<zuKLKP<^eXTKZS=5(}lq7;>1_&I=maJGFJgRpjs|
zb*0B6Gbuz{mb$^YiZb}s@kZ}ZO<wP9^cL4oU7H)Q;j(_Rl&@pdS@x}S4}9L1wyBcr
z68izIkQLz+p&;h2&yaXlI(&lX-KQXB_C_YVhi|q|3wE9Ff69@nt%S_kOc|Gdd7OTB
zow2SLfXqk}1ulKzHyo$cvlcURfM#5GMmW;KR>M_GE=Ss9-fI#VINWSCQaQwY9S)n)
zQCIV8zp>QYNY(wNtkgCJshH#V^v}#SBe8U5{CcOtF?d7UZTN8zW8Y~Kr%ZyLi6l#G
zTBTG^8`2?Se`IiU5T0k4Mpm8c)+X?vjY+c^#|Q2vZ;mGO-E7Z_Jw%Yj4U|87V#}Pb
zcJGN(8mHxMg-=0*8(n+L$aKQU(E~naR5&n`13q5a<5y|Q083WV4W-^q_m48J=oXpS
z=FTffDjrL>#BM?EOOZYFdgA?|bgFmcXwoVHvX|ysf7>TFUZ$HlT8yTWsbgTNyyA4q
z_B|uXjl;Y{x2xFHweA*MD$;XR0{Oz;!>W>w;6_!_&3<H!&N*JOUGL)u!0_{a!p|qw
z=%IjQ0Jf92V)C7;zLg@qv~Y`Sr>hyc4pQ%T97+crdTp0Z0dsnD!zHS6yR)J=Y7R4K
zeCu)kf0o=L^cm9Wf*^)t5S)vC!~HbVe9PCk68Ov3_L5Kh-j3xug<fzT?{l5;f-%ik
z$#u#*p+P;cuh~(&KN;QMQpoL&j<b7P3b|3yQ7tWnyN)YTBJ3zH1s$h_w`wBy$G43%
zfB;fE`6{=0=Ir6|Ixd%S9;;&=@D=DK9mlrOfANNp3iTGj{|%vUsv)E-vE7^hTti3^
z#(4DKZU}wGwO<9}yUYh}{VM4g1=?@;)^8lM`4wfKNM@{JVAJ##`!T3enrmb*y4?%@
zjQV-~KeZg~1u&i_{`KqJ^|hAssBZrIQ6tg4;!L?=_<gt*Tk&TiF^4|o?bl`t5NY|J
zEx9&(7lz7(P`ik}{eMjrxY818YLiQeR`teOaU@rxRev32j}5Si{R7XoK_8Q`Iu!vp
zli?U6mu>Sn443liK`NIApe_x6LA>P5aE2^`0Qsof019f@Y0=gY0$JYJN@cCFS_f%=
zeb4ajUP%CmKzF}h30Y~^krO!5>=l<B4rk7snafOK7E;LgT|;K-QVXQ)QcZzd>5>HH
z+@%uK`ZuZu%l#YGf)xoln@O-{{aFrG(m|$@@RvlG9wL5bc2SzoF;3Jlv7|mzf5Xqh
z(nYACs$IkbYoiEt8ub<MfRc_gSfJeaz{MJfsVm*E5h>aiPS}%C7jvNMIzgmBjjf9s
z7%K;StdOuHNG&jN>VIa4<M#<hkyL$Wgr_C<on?X8w^MQnn51<Pvo_)`S`h5Ai!}&Y
ze@YZY>NrCRQu@wH&Jb6VVSr}xf7Nn>Si0gcA&#jJ5Du_q7fAvX9hEEri}$UQhC;0`
zf!T{|>S7H=WgO5#G3x`c+0vT$YB+7ITl!PN;Y-(-IjFmgi*N}|eFk=);J#zlpw)8M
zECoxey|KGW*Gw`*?ehu+qtTK$1j8_YibF82eTmY7Nq46raIB(z2`RyBf7}o@V2#Js
zEYzfZZ&E|etFN_CnfljCf<tQd)k+CfN6p|K)kdoq%wVt?4jvC{8o!0PYU!HeXRSRP
z{A^kp*Tp%uETiT*G#c#$w`^kF{S{BOdv{R^&A88i0BcqIx~_Ce;}N=HQs!|E^|%=i
z(MKj(je+7x$AEH|Ock4Ge*(XT*wO*?_tDK~$>!0PnvJtyOPIc|YA$E%o2(wYG`g(0
z%rx!|+T_whtoRIBTi*v_J0slsoRwU*a`5@*!}tGvHyggZxR@`8=kH%G{nOFe#ec%^
z)qL@4ws?^MHR*Wx>+tRH^@}`ycs4s-zJTo_PyhysZVsUulet?`e}q>Y4_^k%{(Sh$
z{QLRvP58{PS+{_4%j`c|`QsP(ZzmnHNQAq<r=sG_$RuLJ2pEmLsRF?Rjqr~%CjekJ
z-CYH$5*=mMI&h&VaQ8cO5CdQURIK565bcs~)8Y1Iai$@!cTO<7<B`=U4$)Y)xqm^O
zmfG>bjcp8HPSc+*f6wL@Q0VW^zV-h<d$U}=yZqBIygpmLdH*t=&fg9{y?^`e@YU?+
z;k%RNo4>tVetWT;Eq**XoyDKdhnMH`m&4CFQi^%FT$~PHEzVxOo?Q<AI9Z&XygZ)`
zFINxV%ocCYmS-my_}U+i?C?k#TRj-}`ppZ^H-zmy8~*vPe}8*)#EQ5vc=5&i^Yi1o
zh8c#z#kIiewi+oIX<LmXiy5J`&Ctxzq5EoR7`~pL&lk_%!Aybu3t@PAvV`F-0*xR4
z{LAt$&zC35S;*sqC-aLX7J1^iGa;URl_$m1Ezqxjs6Bo2_yHI~2?0EQAkV33{9yer
zw)z05Ky<B!fA!Sc*Zbm&;pt+2`h2#0F+Ba|$?*N`m*w#_$9sD6dWKEltS^?AJ%5%y
z$Ftex{Qcr|cDc%_{_MNitFx1@=D&3RM9Kl8(=ldP;Jba;s7v?d%N2>}Djd8RX@u{8
zWD#qLCW$9WCaFx4PLfShousysoII?hJV||x<DJ4Ce>ffeM3fNj$lU-S1*uZjNVROD
z9<nX+Mn&2d2JT`{`$N5|KpL<ii-t5-8dUIO_&^1WR<i^$0B$FekO_2uI#`Z^ln>)Z
zq;Sj~N$aK#iX0{xj4jn%8bG+&{&XmQ6hPie5mAI396=v8bZFH{0S0G40ux!@WB$p&
z{(uY@f6c8>a%8a$4QfFsYRWYc2qXczX@4rXKROL=NWTaikUCg_Z#fCjYfcPf!4(+E
zDDP2;3_Bn~C3}8AP9!(PW)eg>kwKCKhX4v~4Z*QLG5|l1idq3#9>5P)j>K$vGc>Vk
z#aBSK#I>lf$^LX8jtcIKOdml9p%XlVH&3Hde*-0JsB&<ANCDh-A_E|su>lg+WI&Lv
zz(pQ`42&FLo%lIjCH{vKttQb+6*x@<qM}}-2yezYW0RZV4thrwOtU2LO%tS=w)^qI
zS0cM_t#lAzjh1?s4&dJ4_|{O;7;2f?P#f=bz^z2OqIwQDi5~ZKb1jEQl+#US=lz2<
ze`i7n-a_iGk65!RcV91Gw-^dv0ET=F{^b?=imrUXzpgQ}dvmFgg5Bg&D~@(=E=7FB
zrMxq+VmG-=l1`FMQoV)y+yhzQ1I@cRYh)pJczaymqs;I<wmpDLE~s9?C>iU4CIpa}
zSiL`lsutn98QFy@fn}H?jVA#%wcR2ce?>V;SDOq>3Sybt;~H4#@Tl+}&D(#&p$Szh
zc~1oRne!%JG{7ZtGQvg`&<P;7opIB9UDuIQ9@n--Q4re)u(>Meex)G3Ic$ZUz7{OH
z#ZW3JVcIq$fwBTX`Uog{16XfB?k*_HZ4+fzW%0dHR<5HgO(;tf%F=|gG@&fje@U7u
zQJyMMo+?qMN#aSG*5GN4vD!c0OYCWCk@*VRGy}rZgQ4xAwHgw4;i+B(%7Mc@v|ijE
z0AO)nehmNxH9~9ok%@6SJn9wTzR*?tL<>&1S~P!4LB+dN8`dq@x{cjO1%beM@9lh9
zVP~WD`8r33H+}r`Wl2bv)A^$Me@0$;<nwiA-1b6FYKhG&ox(6Kf`=PO3Hcj8)cBi~
zEds15>J4S7z5l-(GeK>Gh@FTD4m$B8n-)6<^HbYPSy>|Too66;koWIv^w&mNllL%g
zX{RA_7{DmthM<coWe<%c*FybOPpZ-1ld&}KD1jTHzI@OD-`B?gkj-C)e_2eux2%z!
z?<X&C`8Toxi$<?wtVN(gf2CAml-gp-w{(%br<q&xURD!RBsbp4(-BS}oRg<(i<Td|
z#PK9SRX*T}Tz{>+zBXdb=D<H9OW9^917SpEo1x^5#wx!bl2~MuBo=xYNi6LeNtB7e
z$wc5}B8f7QM43pUOeC?Sf8EOSqHx`->QRt}4RzIH`R)(%cgwT4XaAcmx)(RfNQJEX
z$c1WvX%y<>%n{kC!2<mv2gjOjdGp{-W1uA98I%fua}Q*P+21($^%CvPcF`RDk^~Sa
zHg3O-5-Y|<EyV;kQ&e>+yFps4MK^V@N(wzFrB;F7O13{*?=Rjde_!>x{iS%KO~@{l
zD}VzD!=&Lie(g^OByPV&K{%A^@5NB(<hp5yRw{6d)pFy6U=-pV`4JAiwrd2*OO3}M
zNI<#E<>KV*V);S;y|=!v=a0f$^%-sq8{uB7Y%>hJAcEg-HI$HNb*rJo(%bF7n_OeJ
zw0-q~ZC|_IkSJ3te?h1Af?hqF+6y|h7j()_I%OxFvXf5PNq1`$E2;&{SNBmA!AbPV
z1Ow6G(S91nT>WJ;%I!$vo{8!~Jn;%Lz-HCZT1|dQ3?_$#gi*fjp=dxZG|7_^j1nfb
za;FkfIi(D$uE_`*x4q;-)JcB1bFGAAR4#J25(xFGB&i}{e<r^OxQB}X*qFZwn{@Ci
z07$dHts%UE>KEbQ@eZg+^x)9W%@7H;c^km0qHf73;Z*TGP{lnc;eO~$WUT~eihlp`
zE+syWOAwC3U1F_79k0aQr9@7ILQ9a14N+nYYO-5O=8Q_GT)hW}lXXt?si2VP;gQ|i
z`Mbv}4UPF)e_IANL~4{zW{b6fcBBvMR!hHDx7y}g2)|pUt!y)tbBfGc&FXC*el#XA
zJn`$tLaW<+erH3**9iCK7&UEj466_07~>vR9K$BYU=w37-H9C(6f1NUmwsgp3G7pF
z+RNd`_ZNLRRIL+lmVGoBQIDW3@f?NYW-I*$iJ)$6Lus?M2mi2Y)cpF2HU)gv-8$o$
zPzKpGpl}Xtj|>!-XBV%}XFuH9EP&nnZLk}-(YC5rZVX#By{hxH2|WG}Kk8S~mjPQo
z6#+SyA(|Hyml8xjCzsG=EChcWITF6>SM(n+!y;LH00UudB{HzL1imfyFbFo;Lp+$7
zJ?!7#D!!UzKP1bJF^4_KQqv@>SkJGDEvN63oE{GkPlpd@Z6+r?f>YDew+ZMUlKz6E
zL5q_E5LUzV^gR9J7b%sc&jWpolv>lr(Qi-Rc)Zi#X$H^?pizM1oil$xBbr7u%~43-
zl0J%lyF7Nz>n*MK(iq->_6A)($A3Qk{U9^Ll{=*c+AvvX1o2S0Ge>arZ5)4S$H4Rr
z?Pt#X;St3n#0k>JoQ`);=HJtYI6+%y@u&y=K>Q-4|E68ze@{GGo{;B0@SMo+hqH$@
zsVaj)OTFN`@-y<na!h|x$IJzfF||0~6h3$?Q(o7n)>;?`Bjvc8&k~V|3Ud(0&hb3W
z2~q8yP6sK1UueZ_oG$)WJ>58d-aUyyC5%xc6}C5PSo?a`MbTB3U(}{a&ygEnW!IjP
zC910~7dB^=kX}^@W<6EwNEUdCmb51=KrI}YaX-DUM?jxW->-k7(?>rM#E$lp6dJ9!
z^lgJw)1Tx$>;L)kxXoWVuQ5Vt+gXgt=f*5czwV_uhFC?K@}{0pCd(~8!|{OS&CjY@
zbJh-12-(uJl%q#<Hg$|ne3R4Uxgnc-Ry~Cy$XeuW5w?7E-{*`a4g2Bz_h(zIgmewa
zA#XOhMwP*`Jo$g?ad`UlbogUom*}04Xr`~v2mXfv6xM}FDG|uZ{`}YU?eKZrH7jiV
zj(H(kg|rs2w+br|{e(dat!n^V$nP5A(b%WtuJWALfQfRikC=Y*_VJsRS-B>j%=o6)
z6d9otX7(D+Vljp%pEWsFjmBbggoycVh!D+aj5(&8o5g<+Bb+WGN*Sf*ntpxVF;c~&
zigxH5k2w{yE_}zx7V1u_-7W4>cHGS}KP`Ts4ul}^74ZWhkk<av@nc~+$BI1ikalUi
zH4ZV%DCL$oL%zs)x0ln;iZkRt!k~VqIAgpQPKDNEw|(>_QjQARj!9WIhPbJJq8oH0
zUuI)f+6jMfdSn=g7<}x^n_Y*NOWs3a+O->rP8H;uSQtQBlGqEyQsQ$A6{dp4Ic%;g
zfh;8&)<CU<=MtJ<6_Odwc@O^&$r~Y=d~g(*c7**@%=CXuB<RN}N}Cj*gU9K=6A#lO
z3ZB+F4%r4Eu%R73{Qmvh-@g6|f)@OTSC9jf$w8MY6Fv($NY9Q8T@(BR5lHEzN=13!
zk(HKpb>QoCAUZT#sb<;_9wnsfDdIPaIcj*Lblra*sT{-&%V;6@qnB<IJ{$&5hi}v2
zbC-}4K3D~>&w1H*m+cci8zq&{_eYI6T*OEZbBZ?ixr$2^OIT{%s@af8f1Vn);_>e0
z^L@UTEfhYof0T?<9%jJLnH6BaIY;r<8#mHbIM=f3x@Ma)S}3_Vfjlv0V(v__IV`-c
zNmNVdrIhEIVwoYM6A85;*|aN*7V3-^07p9hMoICxf?rDNI_Vgo#AjEf!%aLBk&y~1
zqz_UmR&~lqEsc(ttf}0Ry!Gl@*hFCK_L$>wYmc_We|V5(d+PA3Yg(H<W>pzCEA(D*
zMZuU&Q*9M%g2m`gv8|*LIC5B<?WeV(cUkdrp<M0noSL&qK65v$?cb_=Y~8@%65>8q
z*Y69BjgV&<*}!;!7=lX>xnPpsQW2#VB5;>0XTWXMCMCATL+zHY7aa-s?*?c}7E~w?
zB;Pnce_})-dZbcn{%sJ{fQ}U)G^3gXgno;EZ|NG%JLjV2uBmGF;7+N5Sc=ZQqN}8}
zA$s;N+*Q(f>)6<c?4_oKzO7579hDPiIQP%$5~*Oxqwy`eM7c47GPOveS#={{awloM
zWl}mxlm#M4Yv9yQYoTx0TxOtBvl&G6)K>158!k0pfFc^GaigV8>>0#Lhds8`kL6yL
z?^-uImv$CDGXgj1x2qODz5xMgmueV3U4Oqcb4$cf*>Kam&@@(E*H~??M$+820!efC
zV^wx18(QZ|G$O9uc>}X@5^346lfyG7S7)x&T1FRJ%UK>+%plBF7vwTR`|cgz#95M9
zA1H7|rmOnZ5Uru<2h`xGZQCev!&@sXKkbI1Cavk}ywx=onkE`%o?E4gS(~nT0e>Q`
z9qN!~kC~_PuCU4PS699w<#V(5Xdydiiu21|?r7<Wx5ZUP+10i-l;)PJ6+ZE&j`F;s
z^t5y9JKBJO;#1mf`_!toAy8$fyX>yIM7MmoB+N1^jox|%S<XJsyySnW6dA5_=a)_z
zx>sMTZT50Xz~@^b2=5T@J3=qN=07NrA+#P7WfB#x-aHd}DWD?BpDCD*(0d5x0&cxU
zSXAC<W!^zp9if*Ct6c4U%zisxe!iiwnwy&c0Z;aJu#>Sm6aqFfmoal56PGO-KEQuS
z51xMg_}7a}2<Laf5=c0Uld&immxSfJq;{3i<eZ+QKz<w$On>p0_2`f9%nzqF`GF+R
z*>0zWo^zmhrjhtT{*6o{zVXBKk(s}TapE`mG*0t3bf>^7WKf7{={NqQt0i{+&MV>^
zf;68&k(=kBN%O)Q9YK<6criS04T67+pA0cCd^vA0GoFkepKpX0z`y4a{W>EX5j8U{
zVMgXbMCvV!m-P|lveqSn!#?kM&X}Y~Pg)!I(FJaUY5e|{Fg<U<&pSsb@gz<<ep@Jg
zOFM>{mq2G(Pt86uVVCG%=q^;AGtDafBHLhz4Tn_P*VgO9F7e1ONpfU*XVHIf4Te6!
zpaNsoq}6$Y6CO-H?<lY5N<JgEZK8QgvQ?5V!E6G@ngnE=;kWTp&w94`Wlz^KZdwUc
zm~o5z)%mp2nYu>0d+TMS5|uE^<;a6z1jP+34ZrodWo6vJSHkJG6ns|NPRDh|Q+4`)
ztum`D-MY33v(i!}O6fD|YI=VJ8%ZB2G!QcnI~KfMOg>NTPXw=fdRCpSH-W+7^}YM`
z%g4+ARzgLfnf-S9?a!C+0Tl^jd6<M{RQ~#X`s3x_J=D&^%i2XnF_Fd)wPIofk(<bB
zA$2+BHA}DMCyU2>t0c(uSenQv-d-0c=o4E|PRpMt>dnZHN+6Ms2zq}&>RWg^U&Z{B
zLdxi$W}wYT)Eh518A3(?_<on*qv@QihwHq-fVCEaw)T+dm)i-qRv4vLlwVeqmle(k
zrB_50M|(R8-U+wCe6_i~mLL>6`2Nf>nZRD5({JtfRf&hBylhJsyIduq?DTfGbhUs!
zOS;-!9+9pV(j(H<QGS0cT^$imkgg7e1L^9{{1c=rTSxFhVR54J4hn%o!oaf@(zC^j
zp#xj-ZkIrL2zf{^zFYZ_VstxiIg_Jcmn3pfU-sf8CMF=!V=Q4|Olv|Qga(#CY8bq*
z-QbZ9^Z^+|L|M3yR?wDMbqmop0-cerqjog-x(X%IXpYXiIlX@?wf!D(Hl~^9!m5Kf
zerr!!+mjKiQCVS)S@HU9m#4^zq4-nJ)09n=L(WuvE)Aqm=pgpJyzJ6IS_@jqFXiAp
z9`?SJCJjK(fV4l1oesp5U4x+&v}6NUds<?HU}^`F(=;mKX)m60abcO`a_DO6)W<bx
zp{wDf^od-K@socYe7rQEIfWv^6<xQM2DZLOC0nwL?G)h1+_A3}7Icr`=y=FBMzOLS
z+GEQ1Bg<2Il=I_oOxn}N;iYx+;q8j05B|JIA?IeeuM}FTm|t|O*)K8zeozpYQOn76
zF<~Z?QVb_9?0$yKI43N+6|--YnQ^LlDwT#R+dK**^lpFZbjsx$8e(>pIFbUPfqL5G
zChx0~%ch7Tv7K>cWm7P*uT^2kF2#~L)32?>QpR#O<c*(L@IdWq1-+u+@$-d4yq)qb
zN-dxm<RDtHcJ(cZdo2S#F$3MO+LOQL>Te2{hh!xrz#+fg7nfm<iA|t3WA7Pc4!_63
zKD4>0glB(a)Y4wnUx2ZuXnRVaaDTytk40FPv9934QG!=^Tfrp->m=u}J<4-4@*@>x
z24Q?kJAhDlTh1JJ6PzxTKSp2)wwb$V``ze{oOI3$J3xnd*^U*wA;Z$>V-dF7eJ738
zMzdA%{fY<jSk1u!niU5|d93xqsuknM{_19YKze_H-d%|guurk{-19pY!;unzwXBwF
zCCIM#8M!MTvtY+nH)^8Sq?Pz5gOgH}XGJi|I1bfdg~zQC%<zOxx!d$WsKL>lkZb`v
zB-`#tVa}0fRBopaG842Jt^hk>>!}TAIl!DdHM!kqtK_Z9FLl%FT7H$7d>*MQ0y!o>
z&*Oik&t}V>*YuNcgiDV9V_;+|Zcx2X>07C%i5viD2e9Jj^F%JoTj{~vA#w~w5K@n1
z%t8QiRJW<ADetJ;r2D0V(BW|_@w(sU9oZbtA4C<f%OwSGL+`OXr|*}PF`k>m{y-u}
z72R|PKpU10k6&;Hpxvn)XpCVYTb9IplbV0_R|KOtH|!4zPFvR@hUb=9Lt5aMT#smX
zYI3S#DqPjD)?<Sj*;e%9lzQ`bk<WIc82?YDzTRk`uw#F77p0oB4z0g})MlrfZE!Bz
zl}jy!AE)fbwpHXvKYF=Yv+|C-y|;s*EN`uO;9`Q}_*q-{ho;2a7Oeziw;8o-Zzz8a
z9jXL&bI`Hui*7P7H-v^Sy2zk74Fm^&i(eVueNf;8YOFInu($UR*)K>A9Xe>w?jeTQ
z%R|4*T`lUlvD`trr$z15tnnS?J1uICvcRr*^+;}?y3RcgqaL*t;PuwYKPk75e$v9V
z_#nr*dWAh)=|XR+fy`BnEAeu`+5LY?M(&HhcW;O&Z3P;7{Qgk~=#y@Ubh|uoL!^WB
zzzva}^1~Y<JrR%J5b06azai3}`RCmbQHtM7*WrO1A|0epx*^i-67?y&!pjpXYr$Zn
zx%N^>&U*JdA>^Y8Ye8>dKE%z1&|Gx@${h?_H8pX4KO@z+a{JTO6{K7R8#G+S#zsG{
zDfhYEZ8skPP)NVU6F9ipibgOp-NJFSw0ncO;gY+p5_q6;Pv8+R<VbVHqod0qp*u?5
zB9)`;LBlZLxklEpw)1{Rp>S>!E`8ZHQs^RM4!r*lM;kPk@y#<612Hl-m(lPr69O_Y
zmtj^7CzlK(KAeBx<<Ga@o@8dYJS&$5g?G?d2ZDIXT)xQZsyZ7`fOl+mk^x92xJrqU
z59Mf-n3X`BVg40~Yj8zMaQRo0pQ<10*T)%gei|hS@j4x?|COEncm%J7?syf1_NvHh
z;#~EsF!PM0b$P36RNmXI?2UP~%D-NV7V3;Z|3Q|?XUTtUvUOA6#X9JPb$+jPCbI|Y
z#J%vMBcgha@R{2Tpnl8tfy{B@_EzM#&%-Zq6i82rF6J?k30{l-;_N6T>WYnP)@HRp
zvUJu+j}q6!fPEhHnrwW_lF0!_IIO=DnekU0E(XiqSozmky04i>JF;<@U?~hJn9Rzu
z^nBPcchG<2*(WQ{tP55+ZHjdxPZ4)<hh(!aGh8<{)!VBeb&Nql%p2cHVbFR@PaCA}
zei7E%zkL5jKG3VC57b7W4V|SYNGLq_Sjbw!j27R<boQcF8sS{1n3hFry{@81OlLQG
z)9dI2mL0+fmp;ncd(%O4mjvH0<0IKWD^T<v5cGfZdI(&KhFd5|<fgtV%D)s%NaG?7
zP4)2QW=E(+^Sq9{YT;m`n&aae`|tJb>E9=AJ|yU&(cND^pW+Xoko-&N0EDN}Uw?N$
zo_^<4S-g?sfSJG|two%I#R^2<p(7Es8-wMz4aUmoO}uIYg4GK<b)*7^9a=4%tv)qH
zVwZol2{g6-N#|rT@_mG6)IxgT)YRAbnZ{{4TeMt88X9}5laX}lf@~_%(n$@5?_%a?
zE4wY4#5rM=;*{|9e#yN`=a@@=n@L_*1oAU(rYp1UhwtT0Yl@e4Zk2XS)<%n<{0tG0
zeR1XoN0BgjrKX>jGA0=mB`26G2LVZEvig5KdA(ojMhdSqr*t;nOb|vZzRMB3k(V<j
za7st;R?YyEd<Abd2qkB_s`>WyIOZ7-3%Q;r4?o*}6IL6^hLtUsU(K*kx)i^!tO#WL
zq6)mE@}hX-K73ztdfZ4~?n%jarB{A4y_#_W<CGm14&k)u7`OsE3Kn*pK9}@jyx)H;
zl(&;2<DH<0%C$QRDT_%!2m-=v<v2oSaDBd!|5&js5G4acQM#E>kJ7ENu3=glRc%fW
z`vS^xaRbf@srAm{W;*--P<k$4@*_&my=bAd#SQ6uY&T#zTNBhAOD|1Og0`Nemw{rJ
z*+?(@_>}%&cE^*E?>XgMh+H|+0>ytxTQkn+gG#xZtU_{XbyonJ5_}|v<ft?yxD-Mt
zbEZOpP1eg!@9s>Ast;w{K5jIXZYFTn&q?R|(#qYR&KC`-Ae_>t1o;%U8Nnq3<(!ty
zKGQ;43PqUQ$B))P>%&YP?~r3twroS#%a^49hi0ayjbtZ%&pa>cwB(Y<^Id-vYeLiH
zE;&~{1WPiwRBo5+FcF$+hK*KzP@IhP!l9Yf2Q)PN>H9Y0kui9)|1-e}7vRk9O6*zZ
zb6T5(&`(b}ZOEVN(E8glFTSQWM<UtO^N{Y=@X!t#=Xbb_jaA_T4Xy|(?eT*fTxN&&
zRjQQ;CxNOLE>3#k(B0eYa?yVPz;yCK%`P3^ZBY8bvJpf&A?oyImr*35si(cI=OaqK
zlsaXhp=G2N+L@CYT8!E|HMH0+4>YtGr3V^XT=MOP7MJ39?orY)&ck{>*0B6p$%k=*
zGOWG)fD)i*+`mWi;bKVLttW3pa5IdY+2G4(P*dR(Lw-#l?nfbz`!s)?(Amc?=NoL7
zL&vXAzZs<?$1fjbF8Rvw>l1Y@!~>as@WMBgzjXTgv&wD$aDQ*HeN1IKDH~5{A2Ukt
z)IMgrtibD7H&e$BlGBfNGY|I!AL<9y`8IQE@<-?1j~nIEU(W8TE_X=p=zD?=LR-)6
zkiFiL^}vuf4#-#KQ#OCIu_hAUuXhKxJIKY(DFavQT|q#Oj^R?o_&S5nFCQVSYDuFL
zfl3fAN%d31o9QJSq!vrFwK8p+tiCLrnqR(@6tQniTB&lyo88Ky`G9<Y1ujlD$SPdG
zr`OADWpR>y3v+#MOI0l>PT?KzIcP;|BtB3E#w5{x_&tZ+Wnh2XHzkt0dU+dn94H|K
zu7eLM`$#;23VGRQt)N8h?5ZyoG<LY+uv*dWFwH`VryvM>+MNYP?VU<Iw#x%09;5U?
ziN_`1F7db&kC%9yhxHP#Vfm9wyq)~y5|8WS{Yt!@eB~04>*@RwPZ=-V=q7_KFC(+$
zw*zm4KzMMRa^rtI@~Vric>JXYXJP%NM`Qb`D*DUJNx84}9cWXj6wqCbvh2;!(VHSu
zqHF5RujHl+PPiw+I9R<VGM)Qq6}eDBDrHmC^YIFjQCib(t_InPN$@&S{gIA?+;}It
za)srr>nW?LKZzc!Vyrgcy(Fu)N*DstlmO;LZ(1I%l$v6zmsgY~-1&w&x9+ZJpwrJ6
zCa~)D9Su_cVfn(gBVF8`G@~gzFx5i2{825>iuuwMC^yvkO?kgvus4x2zTbYRm?nzJ
zXWVTEj3tU@>vrA)p~ypK{E$mS<w?2vfAbVzxs$Ov6ahAuG2sOimm0S-D3?wvKA3;U
zp8oay*B4y?uI~{_!^%4vtRp3`6e!;{&t{FIi4}R*jZQd_;6z?vV3rmVNLd~_%M*Lw
zYOj@EmKK&CHmNo4^vlJ=8WmBf2f_C{CQ7(6bDjQ(MiKRxX_=^5=*2d|F@?9mM>Mv_
zQ9^Bo@|{l@IYK^T(J-FlB%55M$JT$IL>;O-rajIXhBhwt(<MIBH5*-5C|AdsvO1Ty
z&b?6A@VSreI74uvFkST~{w-)|8x12*7V=_IX^$fhY@?t^d6G0G3mQLYC5w40ZyPiY
ze~A38_k8=hDKj9PXB!4<ys|LeG_kS}gX{?ILd=WmTJWMAP0V#=VkTEq!$^Pn_#4kz
zNT%ebP?%cK1;~Qi2zyGh^j4<WJR~lgX!()bliax?F`+FLtpb&8(;AnA5WM!gv@NC(
zr>VNUKPuF_DrFoC8=<4i+a-z>xMT=Uz|jItHDTm%u#wL3^|kx;`Tga8FFpI%qasbi
zzkh#;9R#eb3!>L501bcqXZU~Z<sbd3ot4v5Go6G*TgyBb7OR-O>?XqEM^>-W_8!8X
zMspFTe$Gsn7W{_xk%WEL@~z1vqQGaVgA)WU_*fNUG){pt-Gu0Nl5!WJswSff*%pG)
zQi)s0s*f3|BI;tVZbimKq67EHd~70P#7zn8J(sF4Jl_rhIpxGMbr65R`tB^|i*RN4
zWJgCjWN(~E_gTEac}@1q@|%%ERRETEX#4RgdG~5LvmraBn1Nx}ivmMz>{q;37}ljd
z3B^H6g5nUoSn&>^-c_PE6UWGJrRSVo8G71uRkIOPZ1cIU$X8A5I5!kRLpLe~1$9&+
zpf5x*rI(vGpoR-qIH-SfynEQ}GMCXQ6dLx-y&fX5vC8p+3q=bS;wU;O^4rIJl^G6R
zr@yfNg0XKBOHHM%MiF^VV)D6?W^8SQ%KC(&ix{<99&gxWZC|wRL<uNlbJq4~jV5Ky
zUoV|?Qh3U8ALVk#zaMnI*Q|q)c-Cmw!5za$NdqbY3u98`(TIOo1KCTF!OWyR{c+Gs
z>A2EW{mD$<*Pm56LkZY5SIiPzp+*>4G-|3;#V!9jqJ?E%(ZVus8veHa+q4iR)CCDo
zJLr!jDlCrLEM``#m7$KfS32H|=N{Zfg5+QpROf|r+m8&zp0neA6{v1W0<1kpr>!!R
z7$xHN-cp$qNM3&?%70@ZNgBj5*glX9Mp+710!hJUC?8xxidoCI5?lto7hIAjg_4U_
z)_!CLQjHb_QG`zWH^NmGQVF}d6{jAYM}<6nu^MWmBKsI9s{OWO*gmsuxBB)Fm4b+p
z?P1%CQAKQQSYY!&T!)gPki`WI>wz7*pd7e<^tREzJ<fmBK~eAr;W1#6L+<Hu-Q3;`
zR|#nk!X;9NVZqt^;>QAO)u8mPDOPPso_k%Z2(<!Oz)pTNn6rx40To{8tH`B6_e_25
zR@|e^aW0C(QIQby#Zj+UH3BJ%3*;VAHtm!Ni~YBhEfzw_E}(s$vLS&YS@|G5;<2r8
zt*6KHl}&#_e-#`Yg-fzb#08z6Q<CVCv|{gX@(%@lajZp$kTd$TmOcG7lKi=)AcPW{
zVk@&jh_&!Tnif~HykLGTT;$@#JlJ$5))NPk%-B4oqEK~lzTgIr1urFm{9=3G_*IR~
z`1?+sD0QP_aG{CSHNQrdzSQvqL6CG!5ploUgMxoB>d$=DTEWuilr7DKp&Dt5)C*%y
zRfRwj$@Y;;($khT2ZG}Ec`d1odUb!5Hz$;L7bDlyJ}r4PGTdfOA@F4yb5@*9#bs5D
zuCHo!=h5$A2#BbKb{NoA88XaZ;<uv-ZXQWj+}9&8i+g2*`!HA8U>K_i6^CUCS9+eu
zn+|`+^(yCn@@E`QR0Yp{&R37KCg&@(tiu_~Wzf>%W=yzNXf;gA1J1(QauA1}uhtlC
zJ<@t_CroL+oAvgX<+?`B?9X+lc}38^E{`48urG7-q26wl!b`*^PO`)JdB+{c&XL4b
zT3kSpsmZe{sT~g*z`WP`RWpZm1kI?0lRAHio78_BKNfXnE($glCC!Yx^YY_Den<iu
zz82h1D_AgAXlTE1Z^f61Kkjz;A;(i!3ieKsUjTCN*`Cox%yC$p?a9u8e@aP4NTQs%
zr6hACW$h!1ewT%G9!nxUr1R)*wOhjWE%P2rJ^B(QB*y`F985Y;s22p?tNN6x%L;!w
z_a>r2e8>`njTkrBG-{2SSB~@UD2f>pi8rR-Z7WAFX-O;_ImgdSOS%QGq$S<xXQt?}
zb|cb#3?4*E0eMf3ACRJ#O5{PftW`mRwjQqMyU~=cf8JZdt-9TQ!{-AB(&$KtWVVg~
zi+`&ZkDwF-c@oLl*;$6|t|pDAF^qqsf82_Gnm^s3LOTWFHQQ&K2cG>3W$dMey(uoR
z&79rp>=4A+iM{Kb9R&$~%ujK4$nkjQiRH}sf{#0*$4d#N%rOV-(Mm$8Ps|ssg@lJr
zDY6HgFT7a~%+<q==Fcw`8AOE^jzLwg@T^F}^SN73E1t?-6Qe3L8{AE0qWgafK(Q2U
z0sdbB`pZ;+(1^S-{;euN2x01Q|E3k7A91%AJWciMj?2AZNqOi0qL+Kq(ylyM@4z+X
z$E$L;S$rW$f53P6PcHFdII(%%yS8(G-VfrYN%K#-(+h^u3E$*K&61`^2s*v{+E`n3
zL|^X6f^uHd0l0Dj_Mw=l9Oi#?yO`y_FRlhc<z`2>(k#(9S9>&GK&zTETFS-Ul6#Vn
z5EO^dYPGq7n)y2B1k=A#?A+*@MgD!2(a%<#RqeWZ;P#y`P%gM0<SSPS4`#QU`Yp0$
znYvBuov}oRiJ>3!fH9vc`Pc59F^QaHO25TzSH?&Soc$UbS5IZXU|~e{??#zB#r)^o
zQ+QYn^Lcu5FvD>ACz{ODNk&Tj4Z~e7i&CJ(=RJ`7_0)LnF8FTJkNh(XDUpmD{AK?Z
zd;TX@CK^tYu{sn2HZ_+animt78$CEEm+CY=nE~0igf%|!27f>QH{P;;qtFpW8dqKF
zE@eE25{(GY!}GA<4xY#DS9H~#^7<*$8kx%O9URvOQ*5m*GfkK2BOOt-Bi+$diabF~
z3e-mJ`$(wUi*umaTY7V1pc_^^A<_*i?u5Ew_78}4$LJ}+Zj{H5iFOBR%HeL9{d2`T
zP{B0piT<HqaB9y#^s|3JygNouiFczoeoVYONK=k?BTYYbyc-pIPP`jd-HLbUg|_}g
zJogLk;d$78MHgM~=s>&!C#FO0?+JJ8CXIpiAB?rZVHQH(a?SVnKSEYJ(76V+a}T-w
z50UYkE8G78EY$gmld(D!0ya07u}vxymybOGD3=iaE(({nJwB2FgqKM^K4gC^mnH3i
zBjv0U@_4;PDH|X4*qsUZn?l2q&+Aq;S%XRTs&i5N@#~BD@ALc1KQFx!EGTl_$!SpC
z>8~Fz@i7|a#CInm<vVlw`S<DD%kPqA(XRYX=9v_(og*44+&V=RP7Xn7!m99PDeo0=
zuN8esQs@`v1<A9efmOPr-Y<WmZTliG<e~kapWjd7$=5FfHDW3OB{lOFKO@W6iY2R+
zjx;P~s-q`gqjz+AR#;9e>OignQ+pDAHl^gfa>htX^m+fzgUKE%c7B`eyr0o&1%4tc
zTbzxG1t-<wV9|C#-TYeX+q`bAQ3hNuC%1oh$W958(9voorvSPMyN`dU6%>$@lpaxQ
zH4w4$6}1MZjGb(v=85Zm7P_FqjPT9u@|J^@;Rq>4-?J)kLtvS+q#B-#RxFZARX?yB
z&?-dvj42&$O3*2=My|4AO7jtX5lxiTKrtWjZYq5`r;J}EnX(a`6OJHLb5+gW%t65z
z(=dOw=2IK$+qumS%KU#g7rXwy5&bx8K1KB7UEvhU$L7&}{)}9`_M`8B19JVY(%h2Q
z=Wz}zL}`wJQjTsN4X{(%&dXV?6eo7xprZ_Pgnbu%atP9)HJ2pFYu#27X14}2T9qa-
znN~?@&Mm62A4k#<61%rSpbXL)fwG#{1Zv5v4uafBAAK@-Tho7de3xLwM4AagNP%V!
z3);A28v0U!MG7z<oUU-X1d=6%BX)|6R8LXMnmCYxvcZdq)i9z6?f_3@Q!|{L#{^8Z
zY7U8O(E)<u%-P4_DDg`XkuE+wc+DCcJw?lr&tMzXqhYd8qdc(wT319IWyby)<7o&s
zMr-~AAjt7{^7Mb}O3e~b5oaF3qqIHyI>6YzmMQg0?{GM9b|(@?gGVD5tsYwhvxJa+
zKcWo}c3zC<4@qb<;jX1|m^i_4p^Du-ae`^wwEE*~Bid`($6#dlM2eg|Ery9qDaWlM
zn<2>@wPTD$Ui3Aw<HCK7wKZ6!oxSCE^PqUsZV8-n+RA@bD;U{ZEFvRi<cK56MXp#p
zue)9@W@MB8P=QgSCtphmxqx{JM#Ii?b=380dn<wNHpBvJw!!EE$OXk`u&nEpVPOl3
z3s|<ocrkjZ0;2tOO-0Mw__5k6=qB8+AD8Us35`O#nI~ieWA#a`A9ipse`-)Q!^Mun
zBMc)~O=Ew0OJ-3D=keept%5^=VOQA1jkU<rJ1L5TZGv$O;#Q8F<v2Ra(L-UhDV!g#
z$N*vllsfAQ9JWKkV(n_<H^nE2=Rjlx(Yf-qnyb&wa-6-63Enf<@#yVH=nxI`aYb<P
zoDQrv$}%z*ROggXvx!6Ht(}}9FgO_gq@9hPw8DR{nT|kt2gjd|pDR2b2a@fl<5OoF
zil<Hfwx*X+@pQnRrx0@J3e2L&sNM~SC#<%hxfa#Y_kv5N%DF$QcJ9quaZ%xw8KpU@
zX|8V1O>m!VsQ@QR>*@d4Qd!K$ndFA!jqTO=`T<K9ls8Vu>OX3?fZ@>UCk+?p6mxQP
zzXgAotnKWm##w+w`{bYf{qBLt?0G#9aioNUdNG39FY3jJRi5a@2ue@%V&u;6_hO{v
zkMG3*Vp^&0PL~K@ary0mnhI4wFDcLHS;8SP<9RqFSy+cdGs}-fk=yA>xSxx|EC>}q
zahzzgLg_{CcbW`R1jRLBR`HK#JR{8N2!ely1!-1eACKPC@L6c~fVB$A9nkC%l-8tK
z8aHCbC<H#zsIg%b=j~!c<bH!@J@#*Flk8}bSn6VNBQ_vReDQSDvpE|-nO%^{xVy8`
zDO&B}SkGW#PQtLx9Q&DaZo*rJ%d0CArzVDva^`$2UG5@r%$k};t+=u(lR1cPT<w4A
zrFUl38p5`%v(j-Zr!V-{mI?iOlWyMRmdw0dh@n!LYI}n!$PxoRD`#%(b66KRpA6@7
z*O~r?F5eYS<4ofWLzSp)K=-yoYXiJ@P9~O<imp-cqG>+LAf-dv(cq!xj(Q?m(;~Tj
znz6<x(*?2M2E|wwF{0R3mTu_#at(jC19!<5^Ybm;<ur0~>c@2^IBG*8))igfW@*O_
z&C8bNAXPWl!_MQN37A+dCO2K$0xWr3o>Bdzr@Bjf6B)WehGURXj78iLY7%H)>KtG2
z#xrLh{1XMQ30~o^I3;<y;Du2N8s1qNotCu7y5$y2oM}%~i0;Jvr~wo*X$pT^c1J_;
z3P<Y(k5;mV^OV&fD_Mp%ox#IP*$7?H4_%2x;qH}ee|0dq-{*%5QfmX}dz6f>lw}sY
z;$lV4c&aSBRG^Tz^?2O{QSJ=5<#v|AEpU!v<|mqZyi@?p(LwSsUn)R65kNb){@uds
z`ULnkmc6bijn;2T9d;jqfKh)UPF9!q5f=MK<^8O&4JUH`5tkYTt9xSPdPs#1<VIvl
zM!+TkQ$4*cHzCT~<tAA(Kc#?cd_o&9e3iL7s-!%lLceBy15S*dy`*I{!%a8B>ys;3
z{@%T|@x>}?uVl@XfBUN<8;>Sj5X^2te9>$zh=~jL)jFM8It%yJ5X^t4_CGsRob|s}
zV~w^dD9RL)EnTZ%>fB#&1H^KG1nuRv_Im8CY~n4O|1f?C`3HMwJcGd?_>QQU1zZJK
z6q9h&r5CNZ6j;=<>ZU^YmC`0Rl}cZ5F&{WVj1C*tPdxmmx%-c;FE6`kxz#vvzR@j7
zY^nV(92~eTfDnYJwj3;bkXHb5;>@u_Xd>yJU9Mh;gibN`6#CtFaDvV%@0}nbFh~7n
zVEiu-eAd*Hu{sn1HkW}$GZUBcMn1&>@wd1~K7$Da(2g*mld(D!0XCD-Lll>2Q9i?e
zyuSSQ{>zih0{?V#uqUmtBv_^*ZQw>sO*1#TB+go;-+1N!ke{&xnx!M3=95Wd07>fj
zjOUYm2^*r<yu_p_Dle<k0*UbbwvbVoWMEgtQh|IVxo{}t<GyFsQOJYR5Bq(+1ucRN
zJ7-CCP^P{nK1q(GOyNz?XZZK?@aye={po*CRja6kHq1pCVXZy==g+6`fItgl{RvTc
zpwnOfJ^lXluaRmt*y_KKOhV}Jhrj*x$7vY(HG8F(yl-iY)}5Wv2(8`8kP=c;v2#&)
zFi22v)MmD$ww(tZ5ukE2Bplh1TjWsHy5*m1xt$S!BcZuZY&isvy`_MYx3o5Y1d6=9
zlo$hEVq`Wkh3%9-ahKglvrDvISXk{6m<jyoc2`$;6tpq|?iyjXt1Rw)qs&}FX(Nz*
zZgdHk!kb3(Z+Yo736=nmMmXo&B&ac`57X=@rjeItbgsC`YnT~E{MvEKD2BvSe>FZ4
zzO#K(BG^uR$caF!QNZLJyBP$3l^3XSTp3+(QQMSlR61_*)%$#n7M9GolDGMco_Yzg
zN{H&WTT*s>?n=_Fs%t_hs<_O2-QA1QTKKh%_tmEiCqJjhRnhldE+CiI6GHh(rKt6W
zI9fd+3Y<5Qqhr%Y7KJR{QhD{>v`$qsHE2m6^CS&s%ae2sz#15P>`SPBX-~o{I%o!l
z2*R=-4y&RfYb^}wO4y`TS02gg%D;}Q3p@xiE6ztEKyvosd^@bn#d)O<%Q;WNN*@!c
z6bQv=ZCLd+g(jl3P;8n)GfyuZAATiBHH@TU7^%;>ZmerVo;Su%T2o`KsX39p&CXXG
zZ{auO?XF6-X=UMY2iY}$wB!0+;}F9r@9max?zjw*V<4h7wV5Gos4|1LT-EMIUg_s@
za3bVzh&p|Z`Rh=?azba#J{qw7swE8VqjpMxd&Ew0<^emQ;+jf&l<`PWWw1D`nuRr4
z0iY`F)@DfuToB>BUGCD1rQGe?$!VdO)4=hbofzV2WyV!QVztYEIVjpJYCp@ZTN~LD
z<JT99=P9rRrG>L}(tNaVQfICgsRs;)+BpnBmOoB$)ryF5HTvt$fsr&rSUYf9j;GhH
z!{7Sk*Fp}$uE2B>-@}<@-=TyW`i^~foIbtpaBxlqKOK<Ubc{9Nj>2tn?0kAjDLH*b
zW+kBI0`V)Sg!P_(70FUu;<q}tBs<J*sUn#2Sw>yw*I+CLZJ!iKpSk!3>8y60gC)(p
zw!c#G0(B|fhHJZ@j^Dy-RwUEYN0Mfuem7~RE)MVUoVq@desKlA^i2YKS7iG1y0WU3
zqD%O!@ji_N3&@pC<Or4($hHY%`yNgz$$mIVq&3~)YIDthQi$Yv=^a&FUb7We@WO9^
z|AxjQQwAqd*SA)#$obaL9=&L1PjP;><SH7HY^Af6IV?r7^RqPfwP>H`y>0KKctP})
zT&kiM0HoO4^Z@OIp_*)`ahbfL^#O1zA8^f_Bt_$&?f#dU_u@2hk$I+R&k3oVbs*wi
zMsms8cS;?9DPqKKzai!KoE}d36>R!8ZlYGA$??mpA%PRho9|Ff$LTv%v+s`|Q<p?R
z`a3{V&*^b>Q6B%1qqiPozEfSw8KL1juvX9MpYX1^MjusYx{Npm*Ayod%|pZfCMTZI
zF~bIF=iHEzDy-5wmSTxnMAMhyl-^6?hOTtuc{wkCEyWn_mVvUimcvB1lN3wwyJU;4
z=+~24=|kkRb+gmg=A?ZqMGv|%gQY-T^WM!^)-V^MwC8$Ow4IdHX$Uf3+pa%W$T|(y
zvW)X$uXKo(dT=aNxi~R&$HEQwua{M<CFMBWls3-rf8!;lVBCDN7m!6Xg?uEoUDFne
zvC8d#-*(l`_l&Fy;XVBy-ZNJ3Ua2*a%&4T#3j3{Ym0AJ)o}0Ty^e<YS_wm(7uO5+r
zC0boQYAvjX$F3gZwc_M@E?+C61F^K%=Bk|Ok-uE}U{GEt40mi&Fs&L5dskt{Z|=(d
z<Uk#M%u%ORAzH>2OG$cMhIF$ziU8Z}JqIg)*WO|FyaGOcnTb*W8~2zzh{WstN*%j#
z_P9)Z>>)ZV6W2-OS2BI<cD3FgE|`Sl%Caol3+b`Cx{sZ^hmEgMNJk^JJqI+Lw!Iox
ze8vHbOAj(kJkVNU09O3(WY&_zYpde2)@;ck<OHbNF_qV?#CNa}h9Tve<xbMSD1B~!
z>aQ}T(UrbU6IGaD@3wr(tlY#YzbgIG=Piwd+K*1Oo90St2t^c9(NBkn(oq}wJ=k+I
zN3~L!7T=|Ls%dSN7LKy}*T98Ri!=}DrKO(w+{>P7*{^ywORsy%Z>IpPd_Zy@SUT_v
zHnp`KRNE4i8j7jaS4a7r5T&@pQV*klQCe_`9WuMNzZU$G$;tLAV117|bZYio6ArlM
z?hQU6hY74ywIsDEw&4WEm||S{QM3Ze3FJ&apcV4MG4WlrasZu#+@-wxN_dj10T?4l
z`7ehjUH~KVf%+~yN!fW=@tNw75yvRQ44;rN=*HP!icb>aI3^FG6UMBh|7v`Hk`c#2
znRu4hYNeyE#V3s61UcwK`9m59ZG>Vu2Z|eB7!Hu-#WOy+MylP&&Rb{&iVs{==?4_8
zV0{th?kj`Ea|cbleJUPMh8%(`ZA@$(Uj8MsxvK}!Qh1eZf|!6vXv8l;OduVjK|4iE
zSkVUBdu0g^AtvlfCEAX@-nM^#c{(B{C66GcVGo;#3A>_<Q^bUK5mV(+3pyS~Oi3Fi
zT@QtQ;S4pQZBf&N=L4w8b8?b6f+jRAXhO3En(~|uH1)!7Atug>jT<D&En>0_VoDt9
zi+X0<Y?f=sH8cHBASUHQqhrugpJeBlNeSQ-(l3q|#nR^M`i+fU%jVr@&edO5em}SA
zS~kxQUcbc`ORa0VqglPpg=oAwtcIbca?#tScR4QNWG9f^^n}X`F-zmte)2AJesAhF
z?@h<s`WRvdUYi=;k=*wuE|6Td2gMw8D9-=HC-{CbY+vgz?*1E`x<!qb0b4#60X3IV
zY(XUfG?)GUBr$*8TV0P^xea~KuMp6e?L%}WQ530S5nw->?E<sdAlVjegF!Ge@g`0u
z_P}!!r2X|h)V1$;VyAJ&PG-`F#Pkg<k(6|J$g9M}N-nsVSkz@Qp^A%AA(fC^(%Sw}
z9BDL3iz~6|sW|1~nYyanB-2jqCl#X->pc^pm4|_9a<PA;_N$6fgi{moZwf;y35Kar
z+Uyvh@?c1GDTG0_K+HB-iPfdptSX5erj2c~>{1>qta!?~v=P)Lk=#W5tcEs442}Y6
z6QiRvxE+(D9NV}Uhd8&N#q6la@H4h(gnK3yM>VJRR9p_yEQ5oY%zlUoBDHZ7K~(5Q
zfde_GK{tOxEQ1YL0v;wND3si0gMC?8Y&HrITObQX$00V~C^=5l1|xBBxS4}h;9Q$R
z!c$(_pb)hwP<C7f8?f*!rA?fnpc30`cv@Q|3x13MG>7mv25e7rbcF{S;82kvPzZPt
z9TdbBO@l&$SegwiPC{rC7jhCK+ibZ))aEe<CzOBMY}jB3RvQdkMOfPucpx%N%-M<N
z7CvUDVXzF_M$_PY2~G)tv#2MCLN*(m)w~>`KpgfNR!?o$n1~T%X(SY6SowlIUqjQy
zlGQM*Jlw*bZG%a0HMgBwB)W=Ku-PQJD#Nm1oT{~XQL?L*n!SY~%Af#WLWqMA*<^0s
zndg5Y4!^W55GXX+91b+>a)p~AYulN2FoL8urY(4=+#-qt9Ag~fXyJ}HLrqqbhfQFs
z7nkE#d!F0cOR(oWjAQ3JmSKE5;5f_i-nZX&KRz|Qr~sA6-QWN5Pcy=6Y3_J|bF(++
z=f@8oK-G0W%+KeGCl@DY6K5#&*maLi`hI`1n7K53{qwv2rzibMKXGw*@NhosG08)+
zXut`Z<)N9vIN|L}1B6V&7j8~9V;(%{9xdi)PbU4-?$M7AyWb}7`oWIww|`$uF!PHE
z44U<mS%2BkV;J!9<Z}LIaW=VJqK`fMW%7J_^8NhX(}1sr7gEzPMp)ptrUrB|weNps
zvw0iKkhGa_@96-h=F0%4E9vT)j#6DoHH=amr8G*-+2hfZHm&QvpD&(I7R|(P-2J8d
zx%&ZM5S?TAbk@U?<QcmMr{725LgeHtK5-H~_U-Y#?#Y|e-Ud9H&i>VXHx7T+Pv^7l
zN%z;sKii*gUiSUP<)6Cl#k7C<=G1?m&0lw~-n_oJ|9tYc`{ks6`RhghGs3z6q(EE0
z_#Nx)U!8ZS=kwF<ZS*Pl*j+Bpy6+d$=PxFg7X8b9aWb9tYa)K}hoe&NA4&Wp0=Twf
zpXeu(dS3IBGy8Dcex5EL{eC{Zcy>8`F>CJ1{iC?m+iRw7cRdbYn(HZ$XB=A6;41)s
zmP+&)TZn4#8W5oG9EN|={WSk=-u>vliHySnI13pmDQ5s?VwiFBm=lw?RJ#m0amC|!
zH$r~izv!R!^Jh~V;O)uz7<oNF<bjxoG*}G+j{qbpJ{giC@DgPs@QU;sM5+*C4qCD$
z=r=T=um__6Ncd`)DcRRpH|9sSRjI&#d=~JBT%{h)glby=CSdAl88;-ax@GT4q5y^f
z+)dyuL@Abo*&yPLv!p5>X$tAw2c=CDBA6N|X^1f(_o`4kg@-Z0jvhF|$^^kC)v{?q
zO0c0+0`XSc5IDl42`#l&8B;5wDxjKI8FfQrm3%?fJV?L}noI~4nDq{-#L*jnYy1ou
z{aO;gmfrWs=p}`XjBcF!Mj0K%`6{E=Jmz9vN+swht@60l8!IV}QW_;1r94XGc(II<
zY{~aG+?3dc5j=y4J%iB}!>hW94>2$pd|*Pk0!&xbUUjd;fVJ=x$`$DVI|O;Z-OO+~
z>7TtmxmYCvqC=!$oF_)blO_3o$d&?e{^4-e=xfo9NHYp$@NKcm3rL7lZ~zEOYM_Hk
zl(Pr%iP4gP6_`F}-Y~)HOtvxyK+F=Ln)h1kY-?*J{&KBR5Bq#aXxE!rn{N7_Nn3S5
z|NWyB`49}Ch_|Ck5ud2eQ$;kaiI4!Wx+V7j;sm=y5vE|}y6(U31lnPL#XSYY0zi|I
zgcwyjg^K+kU&e?f)S$xmiZXS@1MCyE_ZAL{oQEtJQ7o!jD<{k0j6iK;brpsM_b-hX
z(KftG2u}?KFqKGS6|Dy}g0cs_aS7)GWiyOHMDivhk_aYFs+9zQAfz=N&W2dQ%CH*=
zl>l}`N(xq%w*`WNRF;f?>8S!PRQ3V`#L`>CWCaE(!1>$2{a;UC!LI4s$r21!sGq{s
z;Wy@EZN@9+10*YD8|%8Rwq#uhGNrVOYL3_wnRimfVJ@}oq8cmCKRDG_!_cT~GBnCJ
zV`!vT49z;HOZG$~^P;q0-vpnj${K}HFIDf#R!#&4Bm1>XtY4si*3Dei??u7dZX<WE
zJC*}ZX$kwemXHGD+Ug<c<5b<5)ysVUh|7COMEsG`Z(U;qp@cbkCsiuJXu&S3DJbY{
z`g)Kh=S^hE>1N21uOZ7=wb5AT<q>Ui+z(@{da$nh*qhBh+{yvmo-V<b^$=t^wU+~q
zQ5htA*C=c_<2ajt+i43c0V9)CRqW+MK}z12m8dmJ^B8CA(}lpd8t2RD?8W)y*%0n$
z^NW7^8fD62#Zxlx2Y>~kj%x*YZH>v>ba<2on72wy-b278EPRZMT~k}aAWET(leB|s
z)CASf&zjcY>>AL0GPd#Nn<{&&xdXPLz<Ceb3aw#V9ubXyM?~Y1Kk?YP;t>OR>|8M)
zpi8x^p;^nae>>FBT-n$7g^Or5@&i0y711o>tMV5uqM599W7jf(SN5*oI<T8no5oO;
zvSKG^MPAZOhcf|0saijOSweGSFWdV00Vso#1~i$}YUzBqUOJ-$!dYIm5*S=gbxWEM
z6s#pl$S4MXa<xuE*{_p@C2!X&R-G&?LoIX*sQI^(#q{L#{QcrJ*8P%#1W$TZoQpN?
z?ud6Off4h2`G7z6GGtw2Oa+P^Txlm&NjUR6sbX*WP~>xDjcbYGL-hEQvxaP`wI$yy
zYe-$o8fAP_QDl_Hw;^TBDKh318FPxH17r;YM+4}82=Yw!@)M~aNh67mve|GHEt(7*
zz$Un4&gBqE1ErLu#gz@2)Z6v44;hi|^`}UXg7rjJiHCUkX$}5T>u7|%+q1iNA2K55
zT2ip<RI%oY)RDG^;uG*PBJy<&FYz2F?~Wi!AyhB2ZyN#RK-Susk$lv6M?9?UBFQ0c
zl9+#gf$X}Nz0F9DV)eOQm$WzQn{N`N1RN&Pf;~uVZ}js4USnC~Jhs>OcGr-=(-ogp
zENOjOk=RgNc&BPYwlr0)HB~>Tmcf-mm0eW96n#?udKxOqo9RR8X6ZxM(}#@3gpLV9
z#{{8=NDu(yScaF-0+3E9FAA2a5a>_U&);r;NTNQ+D_l%!*K%$Z7eCT%;62-|u_PbY
zGC@Q!Rc<#Etb~DGmitGP{y#b@?`49ITrO;Yw1kTP!K2qSM#w@nB%$o08twLzmam5#
z*Sd*r%PUZyif(pm`U>5~tGs2r%3H>(yrquVH(n_!S67OL+sDU;sp6}ze04FA6LxDF
z`GOY{Ib?7De_?eqkwKimH+R07_+MT{XcCukZ$1^5A;<y~moC9EE0;}XK9y=N=Re<n
zdy<)9DZB|k!&m?B*Y~G?o+|sOgf{+UzzZaG`s>G2ctN0rvD_?$WqJSo`}FPU_nvC2
z90#l%X^hr0<)9H-yOS|qsPHN&N3$Rrq$Czp^Fy!vu%Pt9kITPVcz(r_@^{ci61Pof
zJ|GALH~7ncmvL`C7MDTKI24zgZ$2r1Np9RY@a|Xi2g0z(DlUN0fUTqh%xwbS26Gxr
za>`-y|HBklQIwiiFYbffGDWiXm6V5{2YLAR^!oPn`K0k+g(I-)@b=>X{6X?xkUVJU
zU;%{DaCrN5`0H6pW%zmJ2c^`SAItBKf9d#1gX0;%Gk`|{j-M<7k7ypzJVz0K|4M!o
zzuP=^%IgiUcM>t|z<Yzv&vg9z?T;s!5&nE~Fv8YY6pT~BHb6t(jTjofCdj1I|7zl2
z$ln+PV(ieT`DoG@K*BnH<N0WhfgpUzOH7)Aiz)N>_^#5n<Px~Bbp#o}2Xr73cuovj
zq5z8>VK1k^F0er${z&?Hw0{(TMP#DzCh#%*eCz)E`u6nSQ{|@8f{Z^HZG^LM_~+MC
z_<%sdSbtDTc)-K&e-A&N{_J<ng6e;xrbXFFV~7^AQ$`TEgF!6-U6I{wOl85phCMRj
z$e`+<-#-uC$oFrRb|o1LW^{)Qd<oA&7wd{(A!zy)!9pSp8MjCa&bX?7GxW7O8tDl-
zJVm{uFSDx3)A5s$akMDyDKR9>kkiG4Yet{6zJPE|O0X#U6_MB}5fQP?!U_Tz;<}|r
zXWdg#Q8s)4T@BvkeH+F{Cfckl(b(B`RIr)o8iEU$!UztPF5{@kITd)vkJF?RPgkSg
z&$}uGhHZdWm>0)5C7+^yGp8yxS)Bss0!GDsVR2utoul05oAQns?0V>tN5vCu<E^C=
zX1#JP(Xl_%C>*Y+k9?n&a;Y*2)M4IzT0%ruocKKZ82uu>W;3FrQqQx$4o~BI3&g2~
zIERG1mkjI5+@#NS?mKcy>u_E-;&IVx@+5#3t}y6C#6!(8$J3C1*FZ)XTJ2V&0W(xP
z)>uu(8U8h1s-(}GUj{l|@wQedQ&q!cHkasXs(SRUq{MF`4iNIu@td;Oh+FH7G-a=0
za=0;bXj!`Sv}KEpI`2qW9fsz?A`3%HK0GMg6L_}GM}>RhUjVyMmaApZ3Lwd2D1+;>
z1N6dc(<4914#)|AWQW(SxKDPFvI3c@Z?l6d;=f_LC)hz+@CUqH${6|5u8WcADZJd0
zzBXRS;FCIbP?oH!UTubUO<usBKuC_s#gWhvYb+_b0DJdPutdKfjf-u0=$*+6RJPxH
z1NXDoxsV?~zXP_he9G<%8^aFYUMiN;yKzJ*#uV3ZXo#JE;&k5_>{pE~^L}}e)k3q3
zd&zZqks26v{J!5$$LUwu;jUB*(Fx^Os`!E77p)TD=!W=<1nl3`qvJ2-e8$Yk);!JU
zZ|dUwhnde-U^}876ig9NSYsX(Oi@8B#fs#wX6V(C5xB5$PWPCMA|vaZ_Nrg;sK|It
zRMC#w^i)lMxe~CDq6fdJ9W_Cy-Fd6$y%|lrB6;5elX9AaUmAhiR?r2fIASWUgXT&D
zi)$y?h?=%UO)H;g=g`1vsj^*7W*4|u?rrQ+=!aU3mAUWMT&ev?kqWkQy}`Mx`ibjR
za$WNUWYm_G9&oJUxz*e@^eiTsr_vOy)D<ZtewZ|WTAXym-<CCzA~%lnpoEsX$LGp=
zQ6aHGu6FytfTbx^rk4b)&nNbFg9fnX%ty@zddiCOCAFk3>$$Z8Wuc&Zp8MWv8mEc{
z9RQ*2qvPpqWOuE`alHzTh2GQIgjmfOQtW;-9l>j3s<J4Qd{j(ToUEhMAD~Nt!Lb{<
zZJ!r^6{I>4ze08tb`fF~)2*2E?8|W=6yR4Vz_1?P0N-aKk>i28U+>T;R)`F@Ho^29
zZiaTu%;K`IUwubi7#C!kiBvuG8bq8dX-9h7u2uLeY=pK=ALkYU9dl?*>&InaSu5vI
zRw3yUn#*3p&P)iJ-Zbs`7!uR&IjM-dPh+}&3s%sFUq~}@>AKPrHYoISXn&bkMlbw2
zHw)SXsfKjOSk^E;;#)piXaatuYMC;)`h684jumEVh*vB&tSQ^$ie9c$$Wxorj$st;
zXGzH}c^g_2+q^|s!F-T4adqrivK{o2bVREBtU1c(mQa3MBx{ZQN&=~Tl+kN<_(ey5
z6H3f*AQ{JM^bae7mX||{MLJ&+Sy`b6D^nFKJ16BuN~TRjA?Zzqu%W{l+kxdQ;BIP;
zRKj3oV`fDY-}?pNxFl{0y?o}PV3ET#**-kaN*9!k{b9)@VwM=zszDx|cAhc2U$2T&
z1gTU{Q8|(>Mlxzbqk7LJr#y^y^_nz)Bjm}f5So^yWdw{1=HwzQGqNjF(&)J&m70RH
z6jymg(|<tx5D}o!oMG#a5y4gpyf@L{VPRb@K`Qx$NpjiSEi9THYa3u9x+TD-Ab|Dl
zRD)17etbD@#3GkLQWND<d@1GkHBstp19sc+IYD&3J)=k0B3jRnv!a#E^$u!(0!9h4
z(4o+>_PmDdJ5ZP5&`fmz<mrz01$bkIy#!D?e6@|yboFkb#u}lxg#S2d9B_T^dU=#=
z)7(sf)}fn78v@U%<h>}-$fYmn9?;!Fz_E7TXy<W*736w+pj_%W(>5s<syTklS+8=b
zSr^$R7a9C@7s*g)g?-&_I2&YtbjEdiI#i5*DA%Ejk?*;d-#%_EwLGv9p{_R*Lc6p|
zDK`)mY~3_m??^^y?fV_?P|5?FM(7!JJ1e<;V&kRI7W>(^l=1oITesu6IzD&rdu+$|
z9xiKbSOHe=#V5BldVKERwYh=CtT^ob{7#a}jl5pOuE`!w^ZmB?|8=Z?)<N~uURG<z
z*{`)*s@E5t^flvh{fz+(*DK{p(Us2-)SOSTO-Qi2%otf+v|AZl=syG*%c*HMaU7Wc
zq<(ih(wcB_H*891?baE#qFZ{jXOFe$>k^F4-QsPbwCp>UDr@Fo;)9XhJ&eow;5qpM
zx9FuzbN!pHuYwM5?;o2gm09oHthxnVwgMu*7u}t=kv<-6%f_^u$2$%TFGP4xyPdOw
z{{aKV%4nByZ$1?PHJ2fp7ZaC;cs|B|`g#7>k6&M8QCNy#<Jb7Xf4%>B`OiyJUul6h
zoUHUhd2{;rUoY{70Tk8+vPw8I>hIrAf4uyIOKoGJJ}rO#{`16Iek%ZaAONMchy#Gi
z3PgXhHV7SWWdQM2O7lja1c14mzgQU|N-dBtLvSF*J1X`kz>oklF5fZNdZ$5u`Ieys
z$Zcb9`u$FQLirx=nLIj#{GzO65KA&3|3R6}9$Lukuh$@VZ8lCA?W)b@JB?VhiIzqU
z6bz$oqKH;#9k|5B_AEu1s9gP%bRJqr=O&Tf2S~*w!l+N{?KHhT%X>#xMl9AutY6D}
zvk&W|q!vJu@6$)owA5kJJ|dcb^oUn{#~b>|%bG_~!?k+i6UeGyX;zs#Eun!#@ZkfI
zrT8S=y})XXf}+~wXrW9PTwFJTn_SsHt`a$)#Bi@p(fMG~DNg33kI&a8WbZ{kii2hP
zD$M!~EjtsMd`SSPH4WCo$z7A$m>_)6Tq-RSQdYM+AP6ZG2;s;eG)g>wA_$EVZv~;T
z_eTVwyYy*6h#`C_-eCV9g3w)-v>^0h$krfqpE@-NF-QBLAatKjH3%_?{s%#5*5HUB
zG)f(U&}{-Z1syQNI~&<^8I%-v{2r5CIHxBvN6j?fy-grdn8-t~-TAE<I5YS{fzf%I
zt&2`JLx#S*FgxkSa{w}bOvosFiF+0BIOpR`MV)`EN@QInvh-QpH;<Hw>gX=##DUu+
z%U$AQEnr{wLrT1dof_sXiHuPuQS0xGR!2LxavrUyQ3hu3>ow*k;F8OoQGYAgLAEFl
zSB8gW8a<k+(DTBRm$OVoDG_M)?Pn_N^Ny?p$yDi=(ESwFJ<$1oRhBou9_b&C=>~*R
zjw=n{<F&p*(c})=@+Y2ISFEBL7bYDxR!D0_pa`)tK40}z<s<K9b$lEj4O%!<ci{6(
z7}n&9%EDw*jB#(-9atRMazf%#Lb(acn5b&kLU7>%DPg{J(ZsjZxb<I~aZQJf`bL>{
z<oP1dl3T68bE-jqq|scBlIK*}NO6r~0fI^?R~c2I_bR><y1cFn#nieNsmWW=8?H<8
z43nc*M0jh7)ti%Xh<djWQI@n-o@=oh21=ASuO*(Y+ZDMOy$mLBBOq)CWg}>fp418@
z3SoR9y1ufoI10YCJFCF8yGakw7q!%1Xi2Jfld1JrUGh?Yt8J~iTj1W$U~5L+#)%G-
zf}yR9sn@MlYN7@i(#cI}5f3)3S#Of`rkRf|y7Wm{ytWp9PRkfxvP0Jed2k+89X(G%
zQ=GkE<9g|f271y!75V8+M|3MbZ)MCx68dLZ5ZQ7nt(EZ3b40Y;_ag0tmB^opDMx3a
zZTh?}Z^c}Ho6W7y2IAXkL+w$#X%n$)^{oLNJS*4<O4VbFjN<##7>9+c>XVkjH;mz`
z>7K(<^b!rDS0#;_KXsXnt0C@OOQLJJ*R3c<YYT$q%2mEv8F`Xkn0e)@Wi9`DC4RJL
zHgOB-mx_3lP%~dyT)0on9^`>Kcv>7M9;iKObc+Xn-cdD4+;G2eV(8OYdtl*&Nj!iO
zf`^ovFv&`;Co;5lXlHouXXs!lm@tVq3MPDfzr@`(^IaDsf1;3M$Xz_#t5>1PE4a+N
z2v6tn${~$1q3~oLuO$j)G?$u4<Rz|QB<cOIqN0(M-y47A++Y&7vaV9uKO&u?kTQ+C
zF>@4uu`eU1-p|Y&T8Y7FE^*UCbLz*nJ+SoYKq%?&)u;L}duxZ~&zYrt+smK39Px>4
zj+U92axh$<tf?taTphU7VGgb*Y5PmAwq9s&rX_cmzwYwECK<^e5c*`cGBOG*9nY<A
zHgohL1*$!F*q@fBl90yb>T_86k!zCr-Mqqo?TUoV?RJwcN69sz%Pjj458bESM77aQ
zibTq-rDN%co6?bHa#Sufuv}LSc<TWhfi)f3=1;b#7}+lU?j+mO3z=MtD4PjGE%<06
zP9p^510Bc1TOIXhcsLs2gX7BG#lzQYdZ3XKKXcMx1o*mZc#+5|TnFGvB+AeT<4Ea$
z><b4l*<iW@9IhP1uw{X){Z#(~5~zc?nybfhx+yc&@OLp(F9xG;6}t*_D&`i(Za{Yn
zj`8Wo;J`gdax{TMAlZyYF^{sE8(|7fRC=8Gal+JXpbbA%UI_{rpWuh%NYDGB&h0Ai
zyBz6XmN>078<0$94ZYLW5e53awGB{z6X{U=$s4sFumrnD{#32u<*l7z70%T$Q@<sU
zeF;=!rK<K&H+5GLBU%o*)nzYtdXeS#%*b|aRe>^1@O+fC=w>@NZK-H5dya7ANC6OG
zvi$)^zJ*iThRBg*Z4VOb9^~~whu<KkNJ(eTpJR5iY!n_I0xT^-;MhoyA@|yU?bMqQ
z7M()?s!v6YW`c3yGf*2wdJMJevKt*Q?RuCpYCn(E_9{gg9D&Cam+K5u0Z=O4qF}eT
zQ0971d2>uFRv60==(CwV0@Po#B;W0MTw3eKPD{Dl?HR1eb&P)3!$gHNO;(i@phRny
z0&GTgo?CTW_QxN-fS`ra{)xkX7bG~H0QlU)7tbXTEorz>r`ena21E$JZ0`<Y#JGhq
z|LT{opFMdI*^5Q<NhdEL+NM8k(}ZoSU%Jyx57fvgnS;PohwJw7i*o#7eBha)p$o!M
zsAC#z)abkx!2w757J<0Ai=JiV?6sBX<^=YM!w4WzxUMo@UU-Uft}fJnvy$g8>E5#4
zPrd8@Yj&FYyeS7L@_l?=U`SGRSKi1HY6-<1eB>c|2n07-{JBRSG<xz7&9%TXbJ#J5
zLzuA784MfoL(X5_1%ZjVrJfXu|Gk4g9%P9B?bC{$?1)85D?IMk!Dt0sc1zo`TaUEX
z?Im|yRg98EON+-a?q{E7r7>2Juu+HRR=!U>mo&50BW1T0Sf^u4L~~toou=)Uphv=G
zI4SCh*H#5i1rh0w4FY954x0W=3wAU8@qudNvQu7c_9J~+>@HzwH@%jDpawQn|36YQ
z^VuwF)=#1f{{?vzB?OmoZ$1^5K~)zO0Wg=bTP-J-bAmpXfAk6GUqAl!!U^I0S+JCz
z8D~z$0z+IfmO{t<+c(P_pcyzjz5#D|As$-LCqe++AKufYuYo@Mn(k_quK4tdm!Z{~
z?zzkj@VHV2c;!xkn(xHt0isRN5J>v`UUAW;P*iDGOdEePnllfoNXVa+er<L&fv4g9
zc!v{qMjzfef5U(ay^(q4>na|ov5`OFeVccP|CzF%OPGYWB4-@&A>n8P{PJetQ^3p4
zck(XuDM@b7LJP8pQoPa|8EeyG8~Nf$f|quuN_X5R({4Kc(UZ8o!xK3#gYKub@LKs>
z^u}drB<v$B7wKJ;vU1G3s;>c>h&GdUiMpvT6%JyMf9|!MvHq1X0HXPJE?QkVl;QGA
zx+6q-8a^ZMe5C%Zk;A3MNwmCi8CurCXir6V%F|7(3Z=AP(W3XHN&H&i$PQ}Ojcmga
zx6y|6wS!;zY8pGRgwMR<+Q^&qDN5SZw;~SaAb+FZVp2ONPn-c>PHA=e3q9F<_va7v
z;Z7|2fAFk3(xR`Etq;$9$v75&cdgnOu!+7~(mOW2r$m1LEJkoytVoywGA%M&xoD7`
zlU@r((s{<kL<qmGJ4No9mM*VUqnRuJ8TXe~o5l@f)wB8OzIhio)i46)Yr>P(F^z70
zphUiBr<x!I;RFMyLRnC}{9P|eA_}pw{e;xAe;||xX0#oVO8&3Wq$~Mfm;F&oeDv|@
zzv3n73HhxU6-8k!l=Y-u&}d_{^&(F2<*jvf8<j<S%d{;KG(x_TD6_pLUpo5x645(<
zUwv1OR*qC9t=G}xJj=?Z<ftfSq@u=Gy5xFH49~WWAu0(HdBj?kluQXZA(S2G%$_Nj
zf0xuKt4sN>gp|a}&KlZ;Dk?%+V{5_zP?@58fvltoEB*z6Fm9@U$<{~8E<r7*3*9~F
zb<*lXX#+WJc-!_UMlgjb#DM#AJxcN>_P9r}hFO@9iX=jc%*3|VS7b3#<dv_P?eai=
zBXVq3)tkUos7oNcGM?-wNA5A(tRf;Wf3&D3#xWtPXEBYh3-8M|@0C)MGDceDlrh_~
z=9(0o)-+~8cN)<Nz9l~5j?K7KQ#gNHkYrvpcs!z`7(u~v@`0%?R6g)c6})@tHZ>DT
zHf5c}?JNV=?y`VQ3$dwwTH|Sb^O7to_cskobHjvole9UxqPz>zV0cWK__DU>e*`Bx
zm0T-dDMP$28)MeY9EBslE~!3Gc94wnHkg4(;P-ryxMj4bgnuujbb+)X9=}Oa0Dj$e
z519FREk@Q%tSEUVicZKH1GdBl(la+g-V|EK`)Pu?Un)C%Gtx19G7z^G%mA3=s!f(7
z>F#-z2ykY_w9(ysJXSS?)|Ig1f3C!YU*3@~8%2|rh8bU@BvmyLhcO<TWwv6HEIbql
zK$#63Yx<Ox*$LGsPfucEoYo8ue>3NS^_(X>^)**ZeDZwJG6gX3Vx)4Esa(yiHuX32
z5O-k4`L2$eS_l-HlHKP5oXYzYNEY*?rzke&xE^La2QEaDGEMW3a;LFIf6iri(fH7A
zjfqG~z!!6ogGsq%603v((Tr43OthCx20o2x$^vr|4l%8b-W0Kwtchn<fVK&7&^>@L
zOrcvCIZ|?@r_l>eGOPBLL^nkQv>m}cUeu)fes&a#vmTKx_DscUuG-=rDOJ{@wJ0Rv
zJ2A|guvPf@G5mM?c=_Mte<qd33c$$IZ-2go2Ou;v)}O3EG{otz-%o$M{Ck9&#{rDv
zD@=z!_$>;iyYrJ~9MOewW3=u`HcVp_0FC@mQ{{Zf+M>i~$ObhWtC|6Fl{m2q0-pI#
zM8CW9vzc31i@{8c5Z9(0PQ>ZWNZDJhKltVIIC#gLQxgu~tMCUre<Zg|dAY=gkQg3u
zf9uK6xs-#tX`3D=2Z1t==6IYOg|QC&q%hB^psB^O+=kFMA`cCQbX@90UUw&Yp=E6k
zhUI2$58)+-Q(1X0e3;=K(sTecS2Zy)3|V;A@_i#+wNOAGVNs94FlUQXid^b?hLdF;
z;Z(IP#gsPVE~+nIe-7Eb=M1CU1nJf%hz5!GXh%vjFZ5Hq15waJT=Ndum-MT5Qnk<E
zQ|$f_J~2^;3HxJniqGSw0Tb4vsA<Pg#V1j1jzx~nMAG<Vjq8tFyb4i>Psl%L@nHs_
z-6XAO@$bueE&+>p^GrSOA@1pUzxpfm{184x&j)<IR?m;2f0p%p5S4z0u9x)mOLg5d
z+<xNiqq>e64~#&UR6PeOp_un|OwS(=fjLR#ig8stV6HDyO4wV<llOWZ$|UJO*r~|X
z!P{-D4r<4uh1?FCOC{ePSJ>)Q9O)jYcr<1l0(X_I>BdPJJ+VF>PJQu?__(^Qr0Gt3
zDnm7Tc<qK&e;bns!lUsCx}AQyaqid;zsl^i;CE!|AJ2CX{tG8RByZ37JVtpp&U=*7
zVXQu>LOegCr}<JdvDTN(cCpatz0C+n!nP*rP%_E!tvU;x8DZPGrl`H`dW0;p>DqU{
zHD8&2w&OBb9Aj<H3^Ad#=+9=VsTa<_=Wob9`c^yTf1tO!`wGK3Ox#&%Q!6PLx|cRR
z@3}S`EDf<nZ<To5sjXY;Zn%xf^Uiqi-fjp+QwGwE-4O5$5!>#D#7@UbT|z-bdXHhs
zxAVx^-txdhE!s#553#ZAdnJ{76VqD~n&EfkGATd}_>1@T&Vv<9@`>UHa!=_nv|BOZ
zu#!+Yf5#-saab+sl<4GvbL(L7gv-d=b2(twx248xBp_kr;Ho~AZ$qf+2RIuI5}^%K
z$`3E7?(ix7zv||jre~*s=VHp2u{NJ{YhpyG1dOOn&AU0#16p9kBS~ywU%$(;QXFeY
zP$O`FIGN((`KdI9**azWWPZAx$4Mob;7cv~fAL_J>-;%vpYXZQLksjQ7*fB=Lkm<A
zZvPn%y<VT)@42(r?CIL_++iMo`Y^olfalI_+YyIl&#kb8vs#X;zo<%AX!1z$QQSgD
zMNiz^iz-`3Z-YxBL)SddwwgPg4@}-vU`&%xN>Ofdx)-X1bPY!TxO)*Hp88$yvQNyO
ze*uQ3k*Ys<IY8V!9!)?P3EFPtM8};{q}(n=01!tzLRn27KJ5<5w0W@n)uR?#Bl9MN
z>hVqciG$A=vXn3raZmMiJ7*HZq+}9OIJpK=rQg_C*)%ZvUkc?KI{g}k`UQ4%gqjbU
z-dQks$qa}w8eu`@Fh>0y(^Dbi1pEu8f2RZFCrnQVh&$8Mk+SDZPe&Z~r>7$yjvOE^
zc6--oG$eCge+pN$888#BUw*iFAobdvRlGDm=5wV_@x<IW2SiLToV8y(o*Z_0U_3cO
zx;@Lirnqr9n|x}69L>v%`}*U!j|;tZ9{B7`lm(Xwjm0NNesqN)T0)n{%cjIN3zN{s
zPxSWx2lcWHm$Anf69h3bH#nCO@h=mXDvdtDe`=3^zW;X6#o^DV088FFPl|o&yiL}K
zwOF=BFRf*Z?ce5=uTbu>2bR5~*gj9LT?1&n&iD9v^2ZbqJr@zzK9WUz`LGG9M7rm5
zAam(9-l_l~0(d^9w17{Ek4KAqwV`!8xtLu~VVHJ}Q=XHhqD%Qom#^ZlxA((8hdyft
zf0cEyap;uyzWM9tA>AOb%K6wBqaslA>+k0K;diOE6hvNgN97u6=d7)2WSp`#G>%Yd
zOI?jFOCA^UCEY6l607Wg{QA*|m2Y2%7`<06CAPK-U(%=QtzIHH1x5;_M$N6ng7#Im
z#o*?7i{;wra{ZEfAhk~t)2HJnr_EEDe+qwfTqdB3{*t(~;u346lMZ8Bt<}o6G*?ki
z)yzbrXjHN^^eG1^&eGRnKu3yDM@4a@$-Z43I07nfNA-czY(bG5*|<@v(iWDRtxQs`
zCJmD{S7E-b_3PB6^*NZ?QoC`}PK6MAKe~x7{?I9Bwak6Db+s#@+W}m)lEx<Xe}9c6
z=0?i8c@3hza~rSmprQ_=n1-gF-SN1AgB?j`UtgDPMyq15W64E(9E<X4u4}1&&*ZdX
z;8LP>!Q~OrrtQSZJKq+-sMOhpprUnsue-8ScVOd2(<9-G4Kn!RgW)XS4(-;J9ggka
zTcIsP6{8WsJP6t>f<pG^G0<j_f5C?O-nu@K+hHzpe|nN0EWun1JO?NxZi914Cim8U
zoQup&@RGpx;2dhJ42_bp1?URc$X3_@bg8Mpjv`+GI<$&mZB){a0y?XK*@Nx<=vu}7
zZ^4cf?`EU8E&)5I42~o(0=u9vL4lN90XwET5v9}}!HyWtEz`2?U`Gy>e=*_FXSbJP
z#_c}4)d4&R;K(o;a^XP$N65Py4QIEDlDGTpCakWX-Gsn@_3XAD*A32Y!pM%Zo6xl5
z>?Q+WKfB3_eiWo3(-A%FQTCYMSmh1Hce~xO#TDqrpJ;K12g~U-UZ~Cym;zR(#r+3@
zRslgMpt46&X=ZX~`)POYf4N8A=MG2DF*Qf?5lD+jj9*hZSpixk79c5@<F-eW0bpPo
zJ-5i{&{b<4lQEL=WTlZ6NRC`)UaE!eh{))J0`fZ$87*JU!;hahJkqN)^?c`hlc%Xx
zW!;U7W#Me;U_`7B0&mVBU-3cUoemUX*uHnAxgtkU+AG04zICa&e;yW1P|lI?^VX=y
zXCv?^jHLEOBaEzjK9_MrT*v^@h@PA_Ls3O^vOHZ0{Mp$0L_T;;CjtUbn@Clm1*=Iu
ze{U8;6axMCIIo*YfbamOtwZ3d%H)`N<hj}#^2!H)p*Pkrx=7b={mNtyKQWO<!S@F~
z#A6F})u1mNw_Qg8f2JXP8S*QgbgpF@fTy>bUgotUQPcFJd#gqTa(b3D1KK8S@E|j^
zhT6<KEsu+kx+82GZ<t(MbCA9oc37R#U?uE0tvvg*2s;<R47Wd@!TeouCdc0!CJu{h
zV=yXk13BN0L|~mlj8fx`p$HuDF5;z?8)FfJUFyM+m1!{Qe~#SZ=@m4ZtOldG(em4w
ztE#|Tb3mzR>Ga=~t(Mq<Rp2-cJh##Zm`Pqqb^Y`W(aLC4fc`>%4D)!5BkynjI%Kl`
z;JN>Ye;kRM>t8Fn^$>PFQ(M{anxwCOt!cw+l70w9<fpawUNVRsjbFYUk}k|OT4_sB
z^R>4$ars!~f5gtx0gQHvS!`!z7heKjfz=1_rJfb>@6_faTjq81G<rD^OLlC<te~=<
z>=jD6z^)c^zyAe}k373p=C|wuvurC6SH(!FJF+ccMi0^6lWn=6tRB6f_n<t&Zn6a?
z@@^X+Pqu&$X3pG@Yt;(xM73%ocVb$?$THEYAKt*Tf6~i3H)z<-vxK4xc$Sd30nd^?
zEb}b!^c{HCW&U(LOZxK=o|XPTLgRj#RqMGE%c_m+AX&Ad9UM!9y-Kk}1p675aA28W
ziDK{OSF>Fetnz1H8oI4rRqMG^R|#jgcU7%uM^}lkSG!6?u)nK>1It}iKfKme=!UNe
zr#h%G5))6#5C3_xNYdq%Y1F@FMA254@smClm%=GH6PFs4KE;2w$KOBy@glRpRXU{w
z+HkPW2$G>lr}YsWy^%PX3%!9J7bQ!r#M6-M#ur)9#&4xhY9ZN7DBlduv`id*%KdGn
ziANRZm|96*^qDiYtK4p?1SU)zuxiVg`pM?sV3I3Jn|K>Omr>5tH}=tKc2Q$LT|F$>
zRU0jWb6j&O<WYai-k^;aP8vRnM5?lZ+~#XRpz?#o>^5I(g+d(o!sHc#)UEwN3XRrV
zdfOn?;V<%)<zIgNc~!pZQD_B7ES#`d#b;~Ss^Q|3QUn`6#uxwg@%i$f!D-QJp~(dQ
z{PQLLU;u@6#90bQGX3?h!yhldaH{QsO%ssTl3uzASb={?3f5W}AL;}&Uk>zHWw*+1
zmVOJ_S=E}`A0g{Qp40p*WAZ3;WFw^n&XGcYWWp-j%24<EC)ucL`L$WyWCN`OrwH%y
zVY(-MN+BZ+g%j1TC0RVkMTJ=-ym6e1y2J>@xa`ZmY8I!X4-Q=9a$IvxSODI;#Di#z
z>W%bSQuKc`a@iUbw=uf-M8V+wRxq?oG$*uRgu!h6h#K-Q_lsZyEk-rS04vJNs_uKM
z^&oz)JW6T-6ozh-M^OsZVURwiytw-CeG1xZ@<pAc)mv6*3AXR^YK(2yW&2B737}b=
z{fhm2LeE}=#b=nfV?rz3b{ASt2|w<icv*yO=GT99&-aux)G&z$96xyB7nHx|XxT0M
z-VvjvP|ndr{;n(fk<zj#l9LJgAS{|fQI?r3<)&36S<)VMiRu3ttEFUV8h%;R=Gsbp
z3!<BORvf1mX1s~4T(prN^t*$syb;<XAN7sM3N?~XxjT>*--{q2C8fLdm8_6`r45*>
zg*$(RLzw+PwgK?qn}X`>I8&EZ1a+Ftnc~!WQ{lABp_J~~Bu>}ro&>>Bzh#90Luc8I
zs(J2&oQ6VLaGFC-LKy+fHss{*8z875-%@5Df({Cd*zz&t7|1W3;SxMMY31n5xI7s-
zs<`tucHpv%9F>v+hyd69HgbHm6268Gl~sSFuW*B=>j(!z$-z6G1BX(ahDTGBp>3}8
zxR;^~1eT(vmiTQ(ElDJ+%Qkt87G+eMCLBu-<$bEXO-?gcYEg1G86|-DY7E|7V(3Pg
z%->aRd4FiK+Q&U)tDjCAw$dn?Gp#;NlMv$Ul4H)lVy4|mG3d+lF?eCXCR<t)V$*-~
z?r6d18r5s<IVREJ2GytCGEGINy4J1gXQLVuWUjj=$f&26IxBmkF-cwUQ0=Q3N-7AX
zV6)U~l66BXGRv5AZ!~7+IvJ#qp}6uItxZ%nQ*jM9w`_1_1h|T1Gt%0^RiL#9#_^Fk
zR<4E;?J2byJ}M{fK`Ln}7S3Bk-ynaKbH%OD>a(KDVg>Mvc=_I7lu@s(sJCKf)p$el
z5LG;~L)RU4y%j5GPrsuA+K%Y)`7NPosb0EXP$0zW$%fj42o<uOB`)_iNJ!eT-cpC7
zb8mb5h0)GMw|;50PPV$tc1TohgNM6)MH2|uyj?jZ0C150BVvT)0g`RQ*ZhA!>&&7E
zLEWmeu6H_RxUClc_v?-HMc9PTK%R@<3Lv>vZ%7an;a0tMy%R>-UjhFEzUFA3fLwaK
z7M<aBJI3bdsTB%mx}wn>1M4h(#)??uU>8Aimb(n`b%@0!Lsl=uok~O*{c+UzHrHC_
zn3<^bng=djmTH>MoD+ugA69=Y`2jg&HLK~kU!R>ai-JWI4U4d(Yb}b=>MN;C%T~EP
zWnZa=f;9^SNiKgaBRjc@y3@#-1a0FyZ8)zSt7OH^0!y!L3FFcSy?B6XY=nUNRAeu}
zYn%~mfTamKs#RzPZ-d4yD~okYeJna>!>d3+4ClkMQ@0px;W@QJ;j@2N^g9=wHRT-J
z76P~Yqb%4$I~Eucz`k^az8vrpA34a=FVm&6#kG~VDdEY+{+IpWTl-(yVAANrm&4Zl
z+0HguS&T8*1~s3?xFgtU7cm%SHq~jKIE#OG<_#`nkM^1!z_e*s=)H=y270_kJ=GwF
z){SbTLMMQFvuQu?7<7L=INHhR8bLUPT#_|2VGv=i!>TaljEZoBY?g1QXCK;Z(F7Q~
z5n$35UwUr+0I4D~1*g<{n#a#$VvPnWmB?R}`qHVX>~=l0IvV-bj!TqI%yaCFkegob
zoL{#qPP&Ow+w5_svGhzO){A3Mo%+W>0GMx@<hQ(3`>a{jprL>Lo#OVAxHHNN<qaS8
za0UDQ_=<MY$Ff(&f|j25uCC!{+Vx+O6*j1Gs>}~aa5xu1HVflNznfQkw|85s=M5g)
z;;%MXchweI6X*6_7A1^TTsJ!wj5lD=@j*IWeVU4xrR@2Iqm%i%OCUG#gW1m6ATBGw
z=0dBQp0a?F`<H)wMcYoFms<%4C3}@I<!~_3DNCU&2gskxb*}iFSaJLLheS$C0TLcK
zBvLX6=ll;oB&vX_tfAACL(&9%W~qB1lOk_4g*{g6vAyBlc2FxEcsBnZbiM@ODAx_=
zOO*MX3{O8_0_TMbeN_9h?pwtlId1RDPRrMxz{YD5`1pTC8%F4ci#BY;kGWV9nR{p1
z#lf%4w<q*X7fTqI2QHQ{N)KEt;WOX8Si&dq_{9<~!uG`yZu6asB`R!3nYjI0^DcuT
zfP4Jf5+n3o-o?ayz`J@%k9ZfG`M!6tNj$;3m<T)G)%U*d+7eO!SoS1sM8JvzzI+u`
z2}!`6QG0*-Ra8PuL31wObQP6zS-FbpA>k?ByHq|JB=!{#Vfx(hkiPf&rG~{zY!Re{
z5T0_0&374h`GUa0so3T@Bkc__AS2!Q>#9tSyv7UzYppoPp=+~SUMw|k?Ci#ZT$rmi
z=(}I066w?GIG%z6^2yDyDer9VhInmEIl|ijuXi@9I>8BWTrMmZ23&V=yIjvS8FG16
zI!on?R9!3^Xy0E7_|AWO2qN~p<G&u*o3;i21*)N!XP3dTJ{18mmvO%b6PK->KEwg-
zmvWvyQh!Y}Y{~uTNm%fWL9gfFCa_^4{86iyEFY4}(Ueo32jUV$=Lv6u|H9Xo=GW)<
zr+=TS7f}gq{6TxglhWbuA5Y-|ffmNHE=E{=`1#M_+tVMd(7ZnFcMPz&WYQR|CrgG#
zXzdP$lmIei$*%Jyuk%~9wMu8Hnzc0w1-Kdzq<^zVh;OTZetthRBVWIaB=Pd7RuFs&
zpGj=i2ec7N`Sytr=`j}hNquVLY6z-vjK;;3`TJA29^)@OCeK$`l6`_85Zx|QP)JAZ
z6E%rXm+N3iII^7%Fh9TX0wi|xV~P*@lw)Na`#z1KU(XHwIp$DsXNVs7JRd$F1#(7|
z@PCN{E;qWkaVl-b@lx<*DP)j=I!T?3d<_`r<!nM3DL8Z6Spv=}Y-ZWkeKBpV5=xtH
z7CFCw4iYXmck7%B5oa&=r+31+&eDY-LdgKb5Zc_WOHkMx6fXVFiPgfC_FgEcbx@v;
z&qiX9^O|Z)tsyR4fdtVr@LMa{LftJGY=4wMt7^!rV}$*J7^QOQS`gwd%td9PN@~Xu
zqFdRcs9w~{a%@rmq{~{pCCs9NaWJa8<D{$FhDw_tyzqunql5qQNU6r>TndIO0RW13
zQ9FkrM)>2DRw;J6(XY$w0i70Jch(Ism&*K(a=x{?inIW0+nHhvor({}E82L(WPiWP
z@n{^;M=+6^qzd86P2*4*F#^6pId%=FvNI867-pm6X0NO`uLdlerWi8E=~dE77fRJQ
ziSufO%xUm~YQCz>$$}}B*D3Y^1s@bXnoH6aPt+1P6Y#-2QCGDGPLoVp=Yl3s@J8nc
zQ9<H*P1sREMB#yM&IK8Zg_RX#mw)|L`QByRBNfPImYbL1#95kILZcLDU$n)|H8s5?
zX9RSDB5~14U|=p>uX4SaxM!n2VJGPWwTcu8W2l;{kFk#{x+DlVCIsWI5CIx=L%^32
z+oh5^<;n=}7crc(16-%e-V_5HJD&FGtRriW@G<~eM%e(>i^iOXLLEk70DtQ8%m89*
zhAD<D=~~T@R_#%s1heesq10P}dc&eugsQAh()MOWl`jh|I~|lSe#lTsnaiICD4mD;
z6^Pwcdr32tbxU07ioKcTAz__0-IAFx5z48NiSU@x?ejpg-d^MZZN2E!%9WwZX@+<!
z{TRMtfS)ETuKseR{+J=GGJiYim>$h_z9m&&m-!&=a^Qp&!B~?qKGynJ=WH0)&C-CF
zN-604*>KCf!J(#jwi)vn3KzDs-;oBA<OUxuV1X?)Q;wJmm`88n6;XZ=Rbv^ir*7Oe
z5(Dwns0VR3MOG*-iI3rK0q^<Rq^Z2hq1SOc&!C2#7B}X0Wh<eX<$nrphn$$eLye^W
zA-Ur?jh_7=xr;fQX;h4XS+WN^$sUm4$ZB=lo0RB<m()GINcNVQ9#p267+z&nS1~>A
zcfD@a+G2W~eyX$Ou}sgf3f6X!ax<nk5bm8!&)B$E*E;16RL?14$#iRWQ>u3j+0`CH
z1JLNT!1a_2l|kKw>wmqjaV~cs&V>quX7<wB^%%~@ktdw*yn08@h0Y3l(bii_x{xzQ
z|LW~9T76}+4(Uz{Z0&du%|aDcOg)HZp)#D&erb!0HeP#OL#^^VoMz@*Z<~4cqNR40
zYp4~+wBnwp_0%G@YJK)ldd2H9*@^y@duNR<Q@?CbDiGE{D}Ui*l<F3nX90_)ZS~sR
zJ#zwl_tb;8&zwM3uZ@piN2;_JU|i>vd+eVvf=bmb+OOX~<1Sud7Z(ex3Y-&WbQY;$
zRcQT)RiTqJtSXI;SydicVO3eu9at5&O_n)jYuB}`N{!xT%VSA#I`1P{Tn>pzQlm@;
zyX;)Onk?0s3x5Z3^83<fg8VXH(aUHd((>)xlH8stlq2SG<E5K~#%iF<hM(o~=cZs@
zsJVeX+pwkuXJv4=&B-EVacko>%DhLoS}!CMu*MJ9hHHPG=vYyI;HR5m%RJRlzc_aQ
zQK<w&_N~mHX6tY+Lz!cZ5-P_b_JJu9cc^B=6<t&t5`Px$nEu9`&r!v>pEmMO8{7Ni
zhIP5+C@t#rc*gD8R1(M?USm^t?4+|X*tg5a$kd&fR5vqR+v$Mf==EKv14ejd);paY
zoedB&!T3ODQ+krFYk<q>|GAL-Aijt!eQ!CR8jd`bG3BS{5X-@!)JVW%8VQg_fckJF
z5wY3~34cD&ZrqVC`7X*;4`#4^F&oGU%em9oL8tY`Bfe$stS<+tM$J)OyK-c$*>CQ(
zEhDq5dznY7)ascM6}2Tgmu3QUZ60Yb+BYAsrG1%gM0CZ6WrhucGq!cw6&Y7Tw>veN
zg3?nw?l#*{#{5`4bV%XlnQ4U#3s~O@Q#q9vsehK9b9fw0ARd01vAw5yJyyfKl`peQ
zS2nvk*&6}PZ43~#>Tkp|vleHed8meS($Wq6UVF$d5mq(VZ$z7iUwXFRFw+3KNhiw`
zUAoo{3sgA%zi%&a`z=5?)!Bv5=vY8@=cK*To9&gk4__@Q<4_<j^0lr0f1ao9BMSfm
zSTx6}0(YRMh)xi~T`c_wBUQ=^L#<w=JY)*F(NeCQaB+7(f@{L|1YPhNp#5gn_a6z_
z^4gbiZ$1?PFqbjCK_-_yTRsbyF^M`Sf7M!DZ(O$#efO^*(3kYVyd2JOI4lAK`KZ$X
z3S!u4(bf<GN!~h2Wvw8s11G<}XLzN%>+CAoXxEl%z!r5qmmCgf=A0Q`%Gz>Bf$)<9
zG53*#tlCE<aBF>}p>W2irW9(aeMF%)?V}uwOP8y`s`)d5bV~Hdhj@TLT@q4Mf4L9w
zDyQ*%%vV_Y5VER$hJ=-^^{13Vrd;|E+Ho1577Wd03uLo!>Z>V+wd+bX<U-?XgkkCP
zGKM<N%NDYwaWc%%sH>Ax;3TaNDKI4ZFb5XuvY7*`^u1*XY@@!=Pg3d&XAMF;PF90p
zxjwWYlKSRz3sSntDvT~V3cv)?e=UF^CacozjCrKCzBk5X<}cKOhgSOvoPsp!6X0M<
zDt%g|P)PewFwCvbB{`Im$03KJ+D}C!l-fqw5=zU>7ZVMdwC^2?D7C&dv7ns$!Z4A_
zC{R++()!j41vMI<<^V(Y*9r$5bAPS0028;+gnT6Ii(wkrXZ(K+m2=+@e+&yXx?+{w
z@A?p@Uv;z%_F8KlWi#B7`eMofr#hNJg3;)an8%p99pT<4qhd7$gX&!X&a17<VoI~o
zBMp-^=|gyxwbA9_!!3_WR>4-Z?;ZLzy%x~W8i}uFFmfV<4{VWqsN=ce?Ud7Kc7L9w
zuWcSVxpfP==81b;VsjWvf9db~8mWx)R9GS@kMBayWJUY39&2d~Hn%M0G1?6BaL&H2
zM-LV2mayirN#k1h7YE+==b=Jv^zrT_X^iErVd|l19^#p5ee~(4!*@Sl%!bd-&*zKb
z#oObB|8{VC{@*ZsKEHf5yL_3DaMI!Mx8dvIi<dP1@_cr(cu67=e?BElQJ@U05(yD0
zLNT{?ID8f``-|bL`FHc-%kT*y4igTnG|SSIIg|eR3I6Ow2`aJRJP<e0U0QO9T<W?K
zNSY0f4x1!Ja3H(T-jqP{jf&(rDUn(f(sd<rh89TS*_y$35njAMC5)rO)0klqcmX$#
z$Uo~!z#rk?g%u%Pe~=WZwGB#aUKg$u=w{uVV}aW;`~dlvN{$VwlNm-JI(+orHiplp
z*-sXy^K&Tl_vc^x-%oy6EH1A8G7PUz7eBl`jwkas!;f#@Ts(d?`)T;*Xz|0h7mKgY
z7qiR%9G%SKk7vX2+5C9;32UZE!`0=<@cHHGtJkxu;U7nrf2T*sXS3mIw)p<d(FKJ6
zbWk53up9Z&xWO-9I{F#*^nCc|zy9qJ84Xr*1UY|uc6NB1ScO<Y#*}TukQSo>(3YZM
z_+oxGzkG3lO$7EA!tm^9fo+|KJpTIEcZ;uHERGhlK;wg_^YaB3dCEOmNMGftCn+4<
z_=|^jNqzk%e@}*Im-CYsv&GBd*_Tg;?`H27hbxEr?CA9jI>3<Ui>nT#;~dXtSM#@*
zC$p<-^z~=o%wC-yeLjEJ&tAzBc0R-mmw4CNgLaKaKN;V<R^idhkvq8hBXzhL)oEnY
zsNIan*Li-pTNGS-W04vguz;!}3;>v=M$NVs24Eyie<}trGiq>J%{xT_Xjf!s71v>a
z$cmQuKEvR+Kj=|F%8w6nO3Ogl+0a}qI+&AbOmF}&JY-&8&o@%wYc+`=?J=-T3hkC}
zlw)#kbdzwj#ex(5_?8|G{Qdl5ar)-;m)Ye_!s+<-=<Ykr)M9c%f4##DP}4*YWry&i
z?oGtIe{B=zn4QUA?-Iv|O5ovavL@f8rT4+Z3#OZN?_75sH&!KX0HNSSn`|WsKmwoJ
zLJ|OujI)(!Cd8)E2gf5X+$gu*!HM%aIB~igIPoet(F8g<O(U8{c^dIFl4&#nTb_U|
zPr#NZU>gmv8=PhE+5{G-3FHoN=G0Q4o|2R8e|3Hf8C3Det>B?xnaCct8=L_?QR^?^
z%yQHs4<|aAJUUtTfM__e==~E^1>;X?2tgDyusI5sbwn#hVyP9Dgve6fl|f6TnCcSI
zq*gp{55?G$A0J5FE!T6Am}C{Slh<tzWs58*6WTs}2~eH9?P3(YDaIM;+^hFSvC1Z2
ze?q=gZJ<Y@A{8-Sb}tD9(BM_gjTRun(ev`h)q<38_R`NhNd^j~Xx{(OuGN|rEX}!M
zNm4+^5s76FeS%DkMe9-t#8&qD;1(rFt4K*KW+_7-YRUGXg(a+-CoQ}{ETXq+VK@X-
z=?$rzO3~8xpaY{aJbhzuX3@57Y}>YNn;qNi*y`wu)v;~c>Daby+t$sw=e}39YSr3%
z{~EPw&pB&i3}bC=kpaH(4ZHAdCqTp$TWWJz!8T@lQ}h-Y#C7=A{+fs?k58km*eb4C
zs||TNKPq5B<clBUF*IDI&UH`q!EQj>GMH;rEZ&qvnk9}bO)6fFOqwE|3{^TloNQEV
zVC-i#N~A4Ob@<DDA`|TIZmD~~4SNYMH53^5e)*C<vE`oY!#LB|0a?WYT8%NfB^`NK
z=Te*VKtV7oAnL7}4Dg6OtFDMow;gbV5Z=)dbmX@1tMS&0-3p1J<eaPwok!l2+nC)J
z1L}@RI<c+5-gL2=fi7cF%v8Ky`Yjwc)5v!W<G&i1|0xzr)o)QrqNrP^H|5ezNqshR
zqC43ltAs?V+>Vn4iA}Tt>_Yo47D03Kl8@BHl6-&=d#ZC`#CvO&uG>k>LzKDDHg{5+
z-hyGjz`^4EdBz0R4m7j;erd6Jb{R9Vn*>{1pcTW5<;Ds@*S=2}lBO43DZx(7+FNc=
zDryK<T)1xWP=Z+DB5Rq1MhDfdl`Zs`lhBXzSYar2Z-@$Xd~vG-oDXk>soxE^rTjnD
z891(`NPvqMdfC)%38{fuXbulH-nC&ftqwkLwY{;fqClUnL4;m_^Bow4VT`-aVw#Qw
z%_DwA%Y1e=FGmc-cPHrsY=~j9@jp2}ze6l7b$)AWUk^2&(aisF^Zu;4aKzJrR+sqs
zqRgvn3UFw~dJ_QTFA8KXUeI37MH)h&<_=K(4^+sT?+Z4(B+`qMO(v$c7<ADKa`5us
zyQ%5!O2{fw*+oR9E}KZDb1((k^J6#shNYLF%{V<MWSV{}(q4jjs;1})8@BznvH=EU
z-<d$-jA&Hc*Py-r3rA69O3=dTYTp}&9OP(bw9@9l|Be6<g`0u0VSlcTS%UI+RBJcZ
zUHnKnf&u}yU@Q~R=-0n`UKJT=)g4tnt1X+r*Agnewl1GQuRnFP6)G36k~O0=P^`1!
zE&l&n+?4qRnm;Nt8csHJD4djek=>u<c76(z8XZQrzC^{B!B?B5-S<o0?LuFJY4lJ6
z@TrnP@=(jM=1FKYOuMS96#eoX&B-~vsK{`_fAO$*@p5_I)zBAmms@g;?*<Wplt9l&
z1T`qk+`gUF>MRm&Kz_648)H=?^`S258mP&v+HOXx1i7Kw9A`S?ofP`ns<Q4-uyN&4
zu;GHPR(w<}uW5{$ys`k+RJ8H4HKPLNEaJAL(xk$b!3f(lav=79MyWbX2x#d1lF{Lk
zz%aajRMiKE(k}|+mDo`DK+*N(42cX+XZ!gw#2Oy^@}a5tBg{wG_9+stpTN;81|x+=
zw!p<LWUb@)#@z+mM3V*EyxreaG7Y*2qN8g<GL@sl&6eBuGACgt$)T7RI*;w<YlJjp
zyiS#NV|6Th0YsDzqE`<wGoQuQo%P}oLiqDU)L+2tN}HD|pdLcsTpw$V4m}%Jm|KJL
zfAg<2h*p~0b!>1M;IR+iAi~y((my?GnnK}pCC+^)3aZ}ZID7{h!5DD6%{=c80IO`%
z6#tviO6fh|L<M2yO6e6MqD)bAqX$dDs24~DCP=BN7x)BBnM?c_LQ$qFvPbGeQfV?%
zE<r+lHpJddzafP$gqPp?CKyRJb%#@8Yr3ZB^qA%H2#nQfP1P$E2zewy*z$9-_##<(
zbbo-4pG@4me4tYCsHm}w>yLP8g2cyaUbR7VS&jqk%ViT9Q&`Z1n1NPhMm*!eiQ*ab
zNxFv0o#+5EQ@CunuTe(hC{qQAv5Z(~qx-T)x<L}kqj9C${XmSEbmyYAkZfrB#4(DN
zX2lpKiXdiXJ4z&TMn&<=X>An-$(T@uFlM}iMdLz+Z&&-`<h9D5DV~vnTIi{#ZA_kZ
z^5cKx5)~W@c;%Rkj5r$jN|K#MG6y_~r~mrPG%*8G1JxSS*pb8<U|}kVfuP|sZP4#K
zgtaru^`MMv5UT8m47b0Oo3%mYu=mk<Z6S!c2FVO7PSa1Q(F^`2S2h@w?>Vte_FCV}
zN*z%Z;ue)dkub>!ho)XaH;(0)$)gbm-G|lr+9dl+uuyv<u@dtYX|{uuQ)a#DQ{NtA
z70>`?8y^0#RQS<B8;1|1)@hdnL}00B{Iz;jvg^JuBdoC$JItnPbyL8(Z@b7YoP4?h
zmc0VDP_o!;XtoiO?376>rMhBYma?WY=UMis7oi;v+*9=kX~ssnziqRN+DU%NwEc;i
zPk^Zs3z-6D-@5o=nDJ!xwlbUiMu5yJ4D$fOd8mDAXNOa+++G#F;Q`v(Ac{1$u7(e2
zw7^@dyw*hO0~30uwMhhNMl%xbv{vhnG}_WpzMji}p^IanpR^zP+qTN86wW;3p#1ix
z&pgpFCp5K=-#@wR*B?2``Ag4j&0mA7TjCymwd*gSDHyvh(+1n<@cGN?)hJrG>-Pb=
z77vE9mlZ|lrAoc@Qy)6YcHz^fSL(4I4%3uRG0EuIJCNg=;A>2y$Ct{{;TheFMV-rb
z@jF{=`>lCG?dETW4tJXE7&$xWDPn%cIF{6^wvGDNshXwx*B?Dn(!)X@Jy0woXR3nT
zIg!g=Hha>_PNuoPfsI1Mgrs*&Y&HS;IpW%&@b2BrtP(u$?1#SQk-99ZH&ai646JQO
zW;5){?pfT{Fy_ZNblC%$MN4vA9w^Sq5Ln8?9ZuaA{0-4|U1y>1sq8#Qg?35X={mII
zNNT*}{2Oaj@jz9D?my!y3vy1nodsm+%QBB78xH4<k=46P0nqEGPN?n0+SLGYTBARb
z)b;1w)&n-ic=`jr8_F`%Md$?uu-FN(cs>D>^+Dj#*+Vt^Csb7RdZAb7(a{5vmyrq>
zT%J+uZfGK;&bjSapWALlGIM_0$6i#@l*OgNw!hb+3%A!xUaaVlwFmj6hi6J$$u@6&
z-YZB?9p&Wb_LF3P9wFmg<N-}$t{4Idg{VU!sp^5s^!bMxZn`J|m~OgrVU{;;c?^Ei
zq3pxDcT%RZF&?Yc^i4=@i)QQU`5i9Zhve;!s${%_zWwSTcqm$}hve^}tjJ@qIqAUM
z@@2RE|G3-FcEU4!YDhW!o$&9oK7{cz85rEm+|bXDopEtb7Q08Z(Ev<F+GZ6~qH}CC
z9bL3&4`1q_L4~!o3RdhwT*VdB)I6r%j7a~kJ~z}f&sOBbEy_{Ygd=0pmXk*lU5g_S
zGmk8(&^<*$i5){ScNhDT4h}zdsi!V<TN^k3YGue?Fw0i%DJz@}I9i!tQJ+&W-W@zV
zI>{$>_>!~?!@g$rE&(FS&gPZk3ehhaLOkOFLM#Pd4ef0lcyt}Ftlr1?Vxe^QxqFMq
z<~5US*N!Y7rbn{xIQlvO4it$Q5{;Vl=nW8ok2`mN(|@SH*W17i3XA8{&5O?s{>mw)
zTDj6gH`uB=7S#a9wAixDzjZzuw9G<F96Ui&{Gx!VP`z3DX^A_0ghg*z3w)KU=mSgz
zd!+oI&F1@I<kfB!=m92d!EF-|1P8<CJaGQ+x4Ky#w3JLM5UP}#?jL=p_xaQ7<!wS6
z<dyV;`(f|-?XFINk7uUv7}HFJtQL!rx2v2S#8*g&-`-MsP!uZ*pKmSiUpJw}Ja@{^
zTbjrNnoC~CG$whRQ-f1pENFeEbU)<JO1|;30_(^6X_gp&BYR!9@{Gdv=`dKiTVG`P
zGaRv;6^G0w;L)m#h`B!jua5;Pq<}9DXL%bwiAnUXBUx-6IE$!Bwi$TU3go=ESMpU^
z3}IOaDhme91NXeWMpF$mcXIjuF5PC5KqEoyxW91`d2j7lGex^Cqdk0BnFbNUpXP!2
zm1r`hU|eh2f#ft8i4k@JxT@8aIXb^r<e&mcPP)+ZHQ>aUm~h|$v>llIB1>Gb_N|4O
zgtEGk<^S$J&=}u!Sv9MY_bsN0V3$M2<)@@6;+D}G+#QOlnmwT79U~J`BoKV{i*)M$
zVy2;5mP4-Nd9xTQIrp*+Sk`bOCi^9h$L)PI3;gZEzE9t?;dD+lQH80j6t=C?{UE~Z
zi*Q19eWbhGuVYvRSm+j>mN5?2q>OEm5K*Bof^$&+!YjIY7H?O`=EgB0JmyDp009>3
z^u0V*c+o7^@IA7`J2h$asOZiaY~nQ#1~t1l9gMr<BgwIRiQacNxH40dTfDbSwoY`)
zWYyTPZ2Y0K31Nu1+7oRh;l}Y^kDNkYScw~m7UnCy2vfNKaiE?z4P)pjog(qHbb-Z7
z*6)04uJuWR4fPrIXQjsv7dMwoQLCP)wQ-I*MOtR?Pg6jROIS|Vcz;7JF?tPcd|GK`
zYvyJgV*@rFf|v!&;uO(FM_*6gBWWz_m(TYYHC!^5+EFVfP~V&i?CMSqa93Sncd63D
zRGvoZdm6U@IiYU32Cchzgvm)6+=*Pf_36&)12CjHo-G2Ce+2~NN1t{dnivK;L1i4P
zfjw&m;&+*4lrR!CXzw55-!Y)H=#l#m<xaX1(eb7~>g#NcC@uB=l{F8Ji#Ub4(@#;H
zhIrgfoA=YZab)HXM7KNtF8J4_^0Qr>*`P&%2Mqy~)D1sImL#Na7&@c&9EF?<s0cPT
z9d6>G`a<0~Wuo2B?9fs`+i=BiPp*m&nj@j|V?v1xDA-Q=wGVg`gCo7mA}2@Ou~z?j
zLymdHT~le07byl0<9sF?AV_r2%`3yKC40@*pfZ~5V@LV^{Yf_2-wr<5WL7z2{SsqC
zC=CbXq9?f_!bFq#U$VEj4}@XoXzt-Mk&rXu`*g5MRK`E0dRUn-?g{rA9|WL$z#$s*
zaC_2803B}hSu^sH0p}5O!y~4Vque@pOXXWdoEjHdM<&<p7@o+U|A9^aYpl62!I;Pi
z|Gf~&(Ihu$tWrlGd}I5^j;5vF6qSwVKrb24TD78Tj$X7$bTV%K7dh62R)<e26%~VR
z%^sjM6I6=^BTI<Zi+;@xtSisO<GUSy6nj&Zx`EP(o|nPA7fnPU>Qsfw+5;{8?(hwp
zM-i_jW9x?qbC~{`u37yX4}Zdb%vp+US=SEwI9)j>lGCII7sUy!=@Wuc8G*+k3fv8_
zx1^_s>F;vyT+maJM}KPaRYgI|$B}C1U2U%o+ouE6LY+o*(~4ne#7y5OHX8sgwk)=V
z98o~WSLgtWHoOxV_TH<Q$=Q)kuh@a{@k<epIF$_k(kGXz{$Mvqb~Ptsxn9y7Fz7A*
zq``_&|4nJXdux2(R7R25;(K0%*H8~guR|x8#?cL1wbe8kX+CEsitmWrSkVgYjn*c9
z0Z-T_`I{2k67^8fRP4@jpHV7%10haI_Ld7&lU4(L<2mF5;@y;};zvvEE^@LD;sdw7
zn6Bu1%AjB+j{x<+Y_mL$A~hyQ`ne>83H8@skZQ*I$Qq<G+c8JTfc_Be<S`P!zw*{Y
zbdB5lD1$`)hviafFU&PIKfFU&cgibpBf`rE`{Ep!2pR~XaRy8m{1!zv0FgZy{se_@
zOitcV{+sBJ$c#h3rw%CaC@y2PlB!c`^4C!rR@)X?8wIv88&*HO#3<|(rA1}^zd@8E
z`s9juagNa@m*wsh%8U%)<&z=6{%wGC`aKPl74Lk7!0pAl;rYGqJ)cjeE!88b(6omH
zV9PMQWf6ljDwB0)%{VZHt2PcVkGjL;In5;7eiagAIKcmPYVx>IukDUXyt59#BdFsJ
zEoHl*O=2%EC%wA&>$9?l7{$UZ=Q{-`F`B>i-edJT#F;R2s6T6Yv(^LP0PP2#03iKs
z&x=LRl@M6$4ex1|dD@Zfc!+LQ5S+Bg_02nm<Gcm6NmLLfm445Dq8^P^hRvIPBa0#R
zXPckn_qx%t=ccn$ImS*p{#!ra<31;wRZ1kXFoqt_OD;hEA#3V&S&oCC2{bKBR#x70
zw?h$WR~;W?&Rx3r8#D^km5Sm&O!Xk#KQoN~QN@X!<fBT792Up{u4$ng5s-ocb2O#j
z|3A0t!Ap^(%?}66o#OAyLzQweBLENR$V7A+A%TjGX@k$rmB<9Kh0lvUth!u?1GB_0
z{Qju{Xe8oY7Smc6iNovc>EAnr^zTk^i!wNx!BZpYH1vdfRY1gT=+Ga~Su7@sl(n`{
zsg`#R=9!(*Kjg6pE(pad+vj&j;CVEFjSar3d^9d9dKAoV*u+*(W+Z~BT#^Eiz%kBX
zkRa9RJE{E}CVkjzm6n0`9Tpj)WOmGbbbkpG6>w5R-G3y`(#<5os$x23%}fYfFJwfR
zXMpP<B{5(nB-w0akV}<a%wbKgHO5#Aqds@(1}dL6>|i7wBg6;LxQjxAPZn8I;5AOn
zi3KeE0h^6Yn5sF99gUC;q|gGC#?WH0NSseb1lPZ^SEfw{TEetC5(-lr`sy)lQPC!p
z29Z=C;K*b{&g03{*-2w{Jh%=-&bGMnqE(uwUFsNe%P!DESJ5AK4QU0=UVw2Il<DE8
zE|t2G=am~VcZ4T#B-u$18n#ZIEf{jNV|Vmf86?Aa_3^%24{ka<W%U7o^*7!zplii*
zaFa1%Ii4qh36HzTRUe!z9fsGmH6W|kJr-MnDH;~(d%KUx*<fs&L_)ggU9bC1b#x{t
zbEcffU%C_jJ=bnE3-EGkYL{Jzl4o<qRK%J-fw@-~-Y2tOCulqch+q|;?<pu}y{o2X
zD4*8T&<WjRRHG|M6?y^W`gIbW0axP)l4qZ|Y<cSM0j?R<d@UK*o^mN74Q6kBhPhwW
zg1zG}t8HFT?>be_s(FDa&SKKL?`mSLc1>#Iq+_msCH_jvO2|sm#;9Y{<5ux#i$cI-
zvzgAs3jg<@nBwSt^Ljjw_}!{0XONwh#aMz#s89bnKidvy0EC`a+>bB+9Zz8($5#Mk
zcV%_?O5cZGW0s7kIBc!#Al~F_?x;eM)nPSP6^+EllrVyZ0GK~dKSJ+MGj7t1d$npi
zE3_oYW|aZo;9uUA1D>s#heEWi19+?-kh{e7!-o@fb;Dt3b#KFvgp($3PH=#qJcTbr
z@YmaC%NO9fp=JDVaFMY1bh7<$9kFKb^i^R#hS0Z-%e(XC&)b#h>k}j9xczkU-*G@z
zCc|FLb&1CHw1K>!$^n{XTN__iSDW|q+vJ{=XQ#*GF1+W*6_pAv-`E8z@5|%Fm8E`f
zBa1o)0;bNR2hA*#ty)CU0`JkF+w9Qb#&1|(GA)3~rq`KY4}RT!wb}^#U?hI0G`LI^
zm2-c!<N^H(Qx(51^g%h*2qsKB<-H5&X*D|M%eo#pno+QhI@$IBHTCfmk>EU7KQEN#
zEA|H#L$23NL&e}-j$oFqi$X&`>73p_t<l+p2qZVwxv=DbO}LF7oVqZDee)zzl#wQw
zR%w7LS*<(`Uvx%YPpdx}*m%|k(7^0`*f&Fd%=K@n=@m7Ok4Phkohklz?A}>TH<W2O
z%U<C+@lY%#1F&-WBjQ8LjfowF=nxH@1DIt<p^d_R(pPK^cQ?a9s2ecK>9{QB#z-GY
zcX;4DWgNOVU@2fL_q>3`R6IZ*(a5OK%NoEnF9J^g8V5v6e?=J>D>)trx9Gx<9mOPN
zT#>|i`!{Y96})UOjL1QEozhZYsF%H`|2rr)M)asO2}vAJ=sERKwImR9fZRboP=m32
zuZdgiPWd1h(;C!OSd<bQNId8-$@pf9tCuOD9vQzkd9XrbNCrpzph_cJzbJ|WbtFL6
z8p(Zn7O|hh2_%L;eU@@Som(9%9Z7`#Igv6cF_HZoaEF@E#Ase&s-fD2@NG*EdTHbh
z?1_n~S9F&yLd=7D`u9;$;PWqt;_lB+!raVWp<h!19e^Lb5Xe+DmuSr%hvpNAH~ifl
z9ar)X2-%JNifh-IK-;!@$U_&e1tx%x&)3(--H&i3rnx#Rzo1rM;%}p3-g!cR*4OFa
zooOo+D{6XnRt`P!LE11X*d)WD*0C?mJtSrHpJKK7hK*IT-|7b5%>3Paf<AScU76W{
zdz&|-+wr@XWEFnFKMj5&nxhSeYQG2cApgxRzJ`v^4=$Et?M!#+bojd8ExrO6H=kkt
zrLpIx9aE-wv@HKJ*K5nsuYzIJ1OJ_>Po-7IPQVuniFnWpYIY-p&8G-~nE3yWA#m6%
zrX#V$qM>m3EG{E6|961YooDXG)R&nfs1YV3eG-VI0kXU!yD_?%3ei=PJ!viAI017b
zQy2{SM_n(5574%dQ5xZ}&YO%ajl|BVWw)kZWynS2VB55ai)&0cvA3vGtL<_J2hk=n
z;$lR%?O{gOem#chqEQGo!Rb-KmN`h|jaexiFac`YO$Udt5Nd=G3D!%)KXLpNqjQZ>
zmU{y6-=G;>C}qFq{$tL=N7D#;wl_OGStb;&PyTeW50JfT&%K$KAe<~)Hm)v+@qE0L
zwq{NWMHvH^i{PM{ZB$ieM}ixd8jyAq-e0IP2t`?iHOY0%pf{oxbqvDGH77y(_^1tt
zWlhovmShyW#HYsK=c~`+;&pJicP5=-5*pWaB9q*&Z7VFH<_&lfY9=)KlZBjyK?~0T
zl+_bk58z9x>Cfr;sbAjgpZHlOV8rZQsM2{-!$}WC<Uo@2Xxy6!g1u4RxH5`JC+mn)
zgZXt_kJvdpjqy6TRcc&7HsF;RyN!+mPa^)=$KfPx7*q8(j%|<QyLE@8?f4;C4~xWG
z8IRJ(Al~8o5jd&>>c>VKh46%tGizhT2g<3O0z5J)IKy|D|A~t6m+`CM{I!w_Pawh>
zy=AFatkYO6CYYi-tSoreoK{qQdL+r;qp|MtxG~whSb8yxD|No^LZ7P6@GVHUc)=js
z^kGTYRIEd%25G<0i|cwb#NuIs-BAa(r3qnA7s#31i#xtkY&JqADuK;nGs3jiTUJcc
z2H^C_M-~wYRPms5ng(05upV<<VQw-ji-Jj&|7FO{yB-^?8!qH45s0UYghpGl8fWMs
zjb2KKL!xFe_UAXxK{BZ<u2L_xN0s}usF3LlJGPBKU9hT6yP=0>)Rf7GMa#tBU3Vex
zlC`#oPEQL6ggRrj`a6cRcuvSErfy<cQ$VhMc0p{FDCr_hlk(>gFqES$`!xh=nn46`
zr<|`f)T~y6#%?X1_j4><YdFdrD7G?sAI&91l;4jI?lt;(Pgs@A4iWrpEq@e@ej3CY
z=rag|S<C=CMhTfOuF)j(rkl(L=m#yvw%xDO4^4er!V(vI1_g$#soyT)h%u#@tN;YJ
z+P)ywgA&)QS(Bnq^>CojBX&a|EzPpx0clxLGPx)MQ~lnM=k4Vq^eX1GtK|;6O+B2H
zPpi=FY;Ek6&sP36TP>MFosPu{F*8O?b32=H7mtBIq0QvAY<IY3wCgAVO-aM<x+QSx
zc6cd2>}{Vs)&H^4kEd}-->#l6fQn(E*>vE4W*OB1bkK^8@){VI8tKg^FdMi<jgpWX
zElf|6C+}N+?WcM@n=kV~?(IHwrl11re$yN~&iJ{l2s65n7S*6^Ex_sP2tp0D%Nf8D
zADTUkXr(HW;J@zM99{<|-LFNwqlY+UUbx$MJT`I%_GkXcQPk&zRW^>C0l2E{7L0iX
z06EGX+uQjW%HDBd&{ymT)B)aG;<n02;0w1U2A}r<h7ed!w5WUTU~8za*oyWPxH2vW
z(8}%7;oFrd-Y8(?y>rPrkH=_9wl%W4XS+(2-jIkD?4B2$de(IH1|B4-m=yvx#s_sm
zFA%tzQEE?otYfLz_cb%Z8VOZ0?u4%G)uV^LUe153uX^ZoqQ;^MBJAdy4A$Z)JGq?<
zRA-LbKMC^!H~^44Tcx>^?w=Z^hTiyj(KVzCx2{1PUC}c`%gAdlJR%w0(GDL5^9IGn
z{|U-LxtY0Bkc}zPQaCmQ7y#2MFW~C55etM!U?ZlKun$Xv8xh$NSJ_W9KHs9$TJ`j+
z;rytBD!HKp;dClFQfH*pT5f_ot3E!SzCLf>!e^BZIF)Dm-R?b{WPrc}i2`0&A~=!f
zw}$t9q9T@$KSzkt;Uk3O-S`HNx2OU81U>wII{%@B1#Tf~A$nFy2h_w1MgZ0q+@z;*
zeYd3ZiTuq#i_c$Qj?e8CSb)F0XRk1{O{AHMBRO(OC1j(ERD>{xD3dGZv;4$Oz71YU
zrLPJ^aZMIQx4$fV=3!fOi|<AlpJGyjIf32&c~^aPb_jglvw_BtLCHNYPnS3ZB7yNN
zQg@xZKA&G-z0Y4a0Tv1|smTQA0G6wtZ?kIj+wpasrjODljSLHr@vbsivo<#TSQ%Y6
zrn>2OtvQ(iV?+oT<ng90vrPlJ`MC@*XEE^!<y1Qp2FSj%eK(M!Lr_fR$(5%5)+dt)
zMsy3EkeqzoKHk5<cjYvI#`nFh4maSc6mU8pJQe@i4G`X&1H56fh<<bKsT6|M%#nX4
zy6uG=qq_q@C9c87*DOR6B`j=l*N}Q)Rcn<bOBF3D+#8ASlZDzBhnZgfX+PBE63R8=
zQuD5;$h!>tm@efE*?mO3L?oZ$jCdofw`%l<xqO%wY8S1VMX;k_@Ykf?vwT*CfR}O8
zm985Rhp;vv3M^U|q@4-Io;IoU#afO$Bn^bf8A$6Jy_LIz7T1tfGx<b4-O6>t8<@X=
ze_r%FtlOlfb!EY`&fCu`nZJFeWNiu10-vRfZVAu;p2Eow;CYk2Mhrnlu2xu*(Li)U
z;ze$>$*6f+pvZ!Inz)ojHi;u`ggM3H((Im~NOUSW2V&Kia79KiKH-o}8PDN?O;OwM
z1kF#XLl94%Ev}&P;{%g$$K#naMjr5nWd^DkRCtbu-;o^taAhSJF(^9(PPwiX@mz|X
z%59whJ|UHPki{wNhEIP`6zcz=c>F=3{e$Ah^9RL1*AI%P9~45DKPXx|e^9jk2gUQJ
zP3ox+Y1hCt>`OJuhmfs%93QbHn)MgwWACLXDx5xi0YKyqG6NKXfC2|Q8Y@CQ#depB
z2f}ZSYd-jsqby!N`3mhL>;k~cB&q_&N++_6d|}j;ZC$)`0bGiKD{l-DcdS&@t0-Gc
zKv}8wB6;{Q|LJ6n!2KjpVR=!Mfx1KJNCpV|d~curWH)yFW5V*F{MS63R1wE{0KPz0
zf_i_0nncNgYJ$J<fmFhhEk75zm)eq*1RE(+!gi=)3MWb;y)}0qWf343#mXtEY=(lN
z{%bF(DF`|Wb?gKeIzd{}+v6ej1!z=&&r#%%t8NfcE;F{#VTn^NlO`M<wM6DlW>59e
zE|yCTMFXV-=W-vl#P-E$uQ4|VAEm^WV<KFI7x!quG9&uuT@JkSUL&%BOeV&So_r=m
z#QFNQR5p?{U3Pw&G@#4nD)evl0VVd#p9$Kn4Bj%ljb8<C!$5441wW&|WtnD^i+3Y)
zdVkD+g$9b|D2FW<nTuSc72~KuZx45@b7$4-Qx)eCAK~}Dh@z|@dTL{Fb)kx~J!q+m
zL;&CSR$%}&^qRz1gsuS4d00yxt)3dQ^383YDSo8GgqJ2&U6|kXa2G<Et(`;9d~j>c
zid#lHez7X3@Y_wic-pq>YY5NZ9A?=cBHiOnvcQA~^batk<n9VA0GxTuQcdmP`H98!
ziI!(jt-}8h9O*5BV(q9|S(j<h3Sk?_Re5f!0Y8c9N_WVt`*dm+_?UuA;cYESc(>~7
z&P<Ea3LRWTbF3j1p-|Q3!6Uoq#;v;~o)3Q7gb}a$RS6`Y4}yCm`i7%qqV0TCGb(Wy
za$N@1(<jWy7zb)<0Hk%pT^}Rm=9r!!VGErO6V)AidwdpWt-F;5K<hz(In`8-N1N@o
zaGwspoFd~z^SvT<X8*R@ZM#w%A|N@WS~aO{@-o0N@NL>hS9=n`*7ag<=i^Wy_ZN%w
z4^r_{VUKv>v$WlqGJmC@A`l_O(QhADku_^Z7Y#HWYCdJ#%U37LvV0WNr0QxF^ZVj%
zz|Lt57&Llk*x#s!6=ug^8EB6loWFn1Vx-9tZFIDX>%_!A`gF&5kh3*TNXz=;P-4hf
z3t#~8Xwjl0-o2)D?+b`PzCAF)J88rJHwVP>-yBen6F60h)`378U`8VE<MHMA-Clu@
zc_MB8YsuhE!Tws|oA2?0b~Fvc;pKcd;TuGlsrHU2(PZRs_jo(hUZMNhbleCAVSau3
zI@IpadM+2eP`J8DISL}!BzHHLUGGyvDKN88Iog@5EUBCFawo025S@JQdGGN>+PQ}J
zacQ)BQkmNz9%iHzpsfW#rs*S~7uy(FbE><Tb8g$i_&zPsk2G;{;hgAtX!HKks&$-;
zjgv7=j(#0ap^_9qy=$I4Koy)qW3jkx5FhNyh!0m=DpVXH{QM8940^<i9kQ4j&#G4p
zB~nsYo47Ap#R;`Zhv<-O9x9>LEkj02MhAmc9B&_J%)#6gfIG&5{lR_)J-X9K-}X;u
zZqqMTJ+Uj&PE?yR6m|Y{!#tG>CRsilvf*4GWqF{J=eogq2~*p=6&o)EMhe!#`ZRFq
zkjy7mxX58*bcimr;5izf&#^p9%2k5Zdb1poaWQ2Y)2t>*zp?p=LowlUMzK}Qz^+qV
z?M`%XzuYec(2^MdLZ~B7&}q<)@@*I`!h2mlr{TJPy`bdj$_^J~k93b;SdOysjdtD<
z@S@d5y5l_cGCV_FS}&-K<PfiNL?$VXA6%axo3I;Te0>QD<S)%?4GKo?V%xk1^s77E
zK0?ynLfr{>=fw#Zi64v_BZ#^VNv0+%e~MNlV#rz@fD2og4Y3fJpC*G)##Ro^AM;9I
zyH&Z6T?y`tR)Md!>glv9PY@{b+%D_o-u3MbD>nLw7OW|-{WfG=tae(iYrcQHVOl!I
zpVXBh2t7hJg8Q0ah^^gFn4`!XU&WX6k(lo>bo|N3lgl-PYYU~zvcohJB6KTu<yjEj
zbL>nF@Xdo4ww6#kQ2!bk?xlN6T}CW)q=*#R%C4WozueUVJZqp@rKO=2^+`W0%GedJ
zh!c*v+8zDT3ts!>lz|;xiAZ{{BX6%@t|SB+YkOLCh;EijQNZS<rt+ttXuXZMYB3v^
zq<wNM4_UVS_mG#Ll)Z3}c@cum!i18V)d~$fK+m-hDGonr<u)gf+J)b!I*2q)Y2+%h
z^6K*nI5E?E7XO$Ce-M~qPHDp-UOzv@)T+VsZDm%j_l2lLS!z3bW2h8OAZ7xM+_@wN
zEAW7>mU&`qLE0qUUXjpQ0So6m;W<SwJSNqYs)XoK3Mog%j4NCta4K<2m)AE*?4>yl
zz*ezKM9Px7;O7}pc3uu`WzlG|?|sn}ud@_d3_A$st{M&|T6Gld3+p_ddCSx}?CnCJ
z4+Dg+!d1U|Np(artl)k;hjLCo*t^;or0@RIr<#5qz(R@7N$WZwJDizzU*W8>!@N4-
z=m4se3t?|F@*be`Z*BRWQ#ctXMy{C#z*Ur8lsYV-jg&7fW<Hk6dNXkTMjq43?MA(1
zPHH!SjNN<_m*r0GvqD3;(7KapjE73jc>7xHzsk9$xmQPqtQQ_1Vj6zM&5Put+<{I(
zfT0WUO)XAaJiDr;E$E|Jp6d6irhIS%OPX&<*(t)GW=^6j+vUE5vWqtxE5E-2U>uG`
zMD8k?GVZ@;-GL_YY=ci9)|fPc)6|>+qecDuwOJKP!aSsw5YeekB`JNhV@JL~-^z#X
z<f%2n4QPIQ951K%;3LXB-Ml&yrcMe!C+$gJI$>2Q=L`gN1lB#6TW62Ualm5AK8~Za
zEm{9#(l^ZRt4@RQp_>N}V!@6FcrrEKGbUWhI2Xe%sFq+GWjOqGyB4nCgBJ6s%ykIf
zPS=KaK!4+y<yVXU`)ZPxYY#omN=5IiG&{PBq`YfJ=pcd0>M#rXJ6^>K<XzT}6$Z^b
z3U@`;1uYO34C&RsmG~D3{vi9zL_ebTQ~UD-{SXV~Z<MM4b7bT5=@zvafGQ$z-|u%U
zO$wj>GrI#>nZd{b^3DfCEnknmFb^j(6v3OhO5IZ>ay9?fF%g!|LQoY9<gP<j&er5<
z2RtzacD`G#OPp{roTCHfb3`}#n&H;pdWFG|XzyOW)a<r|&_Iu9!$2GOi|~H!>-<1F
zS5$UD8~iPVhz{FB1`z)+0S^rj4)!j#?!c|kZC!TbUd4TpzOHD?#eFlnrB*Z#^n>R-
z&B=bp*>BN}J9%yc++uVUt8D&kmN<#!bDhyu(zn>?IJ<Ht=rT{+o?ao}z-1sV$J}@O
z!sdeGP-T$jY0-N?Xmthu{7Ifr%-)6ZB>8GHbvDjR_ZEgP?&yrj0j|`O`?;GZTBl4T
zhW}Pf2dZQ>)r^4Gta#?C*qY$E>PIiAq8%!d_mXv*+WS>2Xgb%In(VxMZ+H4(N8uu}
z3xvzmcx1z8CQD<y9EeL7dLJ-Hkh^A;D?yjng-apRGq|)}5;Zb#P1nMFpR123nrVM-
zgV6PBTgz?eQj&ucP^Z@#)xB5e{P_`oB*9uV5kh}$ZMWzA8ogI~M8QvaU3N_+JclIy
z`fY9Z&zS2|%|oo%k{ZE}hbID8QL@DgXkk`q!;k0fUJ&8zUsMg8E8WTAes1)ZsYz~h
z^chfTKe1@ZH>hH0oSghO10hl|elN3-q^t|*u>~ihAGdY@BQBNX@Y!%69;ahUMo6oj
zFY&;SHfjtJ9(`{wkfh{C$HNTq{g&SLR?ZhVC?J~QKVCe^9|RqQn>{5?4uK+N_d*~K
zFs%Ui{#Zt(4B&>v5ZqgTx$HlluPKT1E&yZgSxq6N{1EQfI6Ne2{H<Ma*aBcMP{@&J
zG-EDEe!Fh5A7{FQm7al(C2Cnp{S^=psQGAYYX}1%fj%pN948;0^P`L0(aqU4yTwez
zl9BOx1J7lX!V3}NyNT}hq2dDBWr7a^jCbGto=cUE{m29eYi5?{X-jt_I34wfz5Vtu
z?!*ON+U|t#jkCY}Nd(<j85j=!Lve&nl$ovKNh&-t3^X6mv^mYJWr`X&1jHTLfmtwM
zMH$n)AQzmRmd!+K%{-4NEfFFa$C7^h6mVn&#+O#urVJW<_rs*IJc_^C>Z-^9m2~q{
zEq79H3K#_}w*MrjN7Lnsp*0Yp1+-HoujaEOnT{#R*l20t)!dNS3g~PEILS)M0wy-%
z%*okUpQI|zkO+)a@%yYf7J)OK3)7v|)U4HCxXUl)=-$i9$j|V#V{!p;Z+3-w`c|Vo
z28dve4f=|vHDcX^*HSOAefCZO+Q8pHyd={sUE4Hru@eEJf>U%5k(bHh^ia<2KAPh}
zNu=hZE${=D=2=*I*L4apSJvbeY5ovUA6qr0_>o4uKzSW@#R@b~V_TPnjO*ti-fuk;
zyr#Llo(%GA?-`bA8!RRp=z3fA--%2Xk)%U`97G3}a;-^M*H8jO1wJPLK|O5~Im0Oe
zY(I!dNU>bt{UjtdA>27-Ie6VDZY9k42j+0m&Xh=MF!Nh`Xh{jC*7!-KcH-+T;CKI9
zsoq!7chaiD;Z%1BK$lkIpp0jvWU%K|+P+&b|2Re>POzrgdQ>9?5%@euCDZECys1n<
zC=z+E1Y-e<ZsWlWMfV`UKq`QJ)XjN-DR1&88WaBe#6Va851H2P1f?3V6peOR;QSEE
z^+q7MVIqHsi>2dkBzHBL5SivV^+jhE!!k;`J3E*o0R6u5z5s^oLt~-d`O~cWK{FYu
zaJr)`!6r>Jr&2mkrm*}1vzbX~lZE{|XMCbE<D%8}eigk`k^2|m<##;FnbuUk&+KWt
z#y+E^Jnyh>X?7ZucsBp20ef6|k%s#k3Oi@amr_EacnZB<1BuXZVkPr)rVCx?Bz8F+
zHhg5;FwL#ardiR6?Gmk4f1}I3%wml}F{%@Yy+%R<awcn8mZmOkbn-=3Kmqn3&`2e8
zj9m*<bh>0Zz0N-XKMi>#g)>V!{@qf#UYeuPJ-Me-fQ;gwY|58os}3n0=zldp*as0X
z&nv}};z#J+YFH^M<!vm?muVm+yk`iO=ywr?`5a=2kBas^xp8`oyU_A`lElm*Y0Ji?
zVA$kWyg1sQ=&2p%LlNxQzQ2;B4nb|P@OgfbjVhI?i68F)B$Qz8Q1dkI)>@PjL!|8c
zteRz~5<TX`(~IL0D&`i{NX`d1zxOh7k^0@C2exG8h<6Xwe!XiHbbDux30V(<1iksg
za_4biM(2~i;eZ*laPKpb#}!kXiA!*GcB3_%IhWbi^_SCuv7^xGOObJ6q$JJCfwtxS
zyVwmwXGlo~xF<$61>vcchZ}9W(8LNcKpFRNRq#^5>V_e|WvhfE$eo}|&Z|g|{ThDs
zNf5VDC$}&8OM}OI3KIg(#ce-gR>_%)C<dDfoP|R)9M$MTiaCYM{fzI5VF9jnPvifm
zb=@^NS9H;D;jUV<okh$jwT?YqaGSvb*|5C~d42*2(9rrc<Bn~}BLa{0J7OvQl?fze
zo+&G=CYbT7<1}$_Uw2BEb$k-}LuRn@hI9zKTWgE|VdQ2j(q_`*l&laUtT3^-S@g<K
z%?g%a^U9Qc{hR1CWeruhT#Z5hwS%#kM+KOy%9^>9wtu^tlRPcfZGdN;(sU8@0r;@s
z=U=-6%t~D<&>;nGH6rt#jt7UqzfQmm;F1q#ST?3R=DD3I@<3;65|oN^DKdFnCdDDz
z*bWIQ1;aGJHZ2dJrT&A?pa|Jha>*mommuO~UkK*pq^GdxYIZyxeZ2Ii2#GfGU;UbF
zuL|w@{jD5m*~7Kojc^3pp6{04*=^MAL=mtKkXBgT4D__)p}*jv2^*%>n{PR`g%PQ7
zG$&tpzvBolHdZfBUu{91kOml%FdKHnOTo=()JV%cZzs{B(st4RuFgJ--f189h`J_t
zi@7+24L_drC;D~OPGv-j5)J;X6(=}cWfXkBB%FTKhwRBGBzvsI1dKS-%>eY`7(U|%
z0KzNK`f3?48VQda`urX2fH6z+A_|?k^?AI$-0l4uYRGii4Fyx;u#yL|aTfD{9(_`~
zpQzb2?;a*7+_x40UQPGJoCfF7Pp?1+YLJ`&u?S=4&H9Ef2|BM2`t9n(_s$D80m~+u
z1~H&1aqqdk-Yci3M{=h+(qkVMe6*(lNX_}ZtKr?W%}O=#w(RuQmMIXY8kNML{>MWE
zkM6@(8vR#(A3MmKQfQ({FL>AlS<+6=0c8Pww;9rBT2zUZ+!B~KO$B=Py;Q$D0f;WP
zD8bnE{)-p_XbqF{KI;Ji(=^jAUV16fDFyWMw1MwThdtA|$$j<>U+#}2N+)3=V4){o
zT(=I0%~Al#BV@}Cn;bG#eQnJM6h-rHel9lA!>e?{XN?|-I2CwO4|C*%t(=GmkA>6)
zd7v;O_RkK>l86)CqdedjA24F`!%MamV_3xG!?}7SHTW#rZx^ay>y+cBDp*^Yg4`<6
zPK=desw~3-)W^eTT{`<VmnLIHKyE3RTt$_!6l{v@7g)Ozi2nWsE}9xo>#;hIM{_^D
z%5LV(<#nK|WBNl{li(2~uQq>^#%i-{Dd)cez4)WyzvCXisC)+bC!w?tH=fj%Keak$
zDi_)yS6!LPun!2{@6p(4*ZkV+iZUh}nu^Y!<aFM)Wf3Lvp7X*xQ!<sJ0Sl@AONsUE
zB+BVoJO)scKQv#m-DC-J{PC$W_H3x2cy9x>E@!saXUQ^VL<k{uI_1H5<2v&6+|QFu
ziu`GkY;r-16WE-}ZeGRzD%OdaLuY(-tCZ%Iyrny^;G5a*9*Z-?+C5Z2SX`R+=-(XC
zJB5~t7;#!2lc6kIT6yE-0a_&?IwoUa?T6v0TQ=PE&3NP|(V)WB_Kdn#%pk_*|2FDP
zT>n5?)a>k3X}nQwH68PM%4Gb>VS_8-$a2U1S7PX*BVlokztd|h8iM)EjDbZwOUQ)y
z#9K6OiQ%Tiu1UR&Sr)I%f``M@-4qH%09+%}arv5f{=ATH-R-R=4-oMLU%>NFv+eOv
zIFGlQ03Fg2k2*5|ec%rUf{S+=oUKIiTb~MEd>d|IaQ1Q_)KkVeD%^@6pSI8|rWh}+
z=jBZyK)~`SKd(cm8gJP!8DglN!CkB`{P&_bQJl+g8<dFed#}0Da}XE91qXF)qpwxT
z^AGs+VK+q8>&pMy6VOa-+extNl2Z&x4517-N`S|~h<`5~j}fjkyJuZj7iPwM5V+j+
z*L8Hwe7b8uY(FTt;qXW6&Ur!$L9d%{z6->#J~PL!pyK3-+u9$l#RxAOd*K^;(KOuH
zpZ97lqO7sXuHIfoZQYfZB{vKGgD<P)?(B?teN5r~@n`V#8t^<J(J{YZ8VmVvieMjD
z{`to!7Qc_o5`I#Z`=N<~-yeITsW}Wb^0GJ+FlT4n_2bHw4>x;qU0!nQLqGMy{9c1_
zG2tv{m#en;MJBv_0WtAP5dGq?Ldrt`Yql$N_z2Y!!Q)7qeO|TS2(FQFuZ(Hy9V&YG
z4P5QR-2h`|m(a+A{L`_-ybF$+;)BOVoua!T0GBfOB0vV{HIYn~iYOe3c^KI}Ckiz{
zIBw}Lp`9}fo;O^)c`^g?C4Oa;`k|+<1?nS$v)>Ncj!~xRUru9;M~Pz`pN_tMdkEU`
zfVhF%QSzHlAm2egKtAA>x!1cZdRaZHv(~rw=u8mv2bqV+hT_pJ5!~|SAEB#7+X@<X
z_+cgJ!v6vYJnZ3~205{Pyj~*dBLxVRcxd$96?c8Jdvkv*snB<jF-ij(@wx!`ZE@3D
z{`&PGMa<d?Xi)J@GJ(wch8{_1QjE%0TTHh@r+t|A>b<53In@Xl)bXY&C)xStkMZNn
z)^Am|`IWAo$;Sio4QJ%ryX=k|wS(@@lIlon{LcW+BCO~B_uN_qH9ZkUB&QElvz*A<
zi{N@@kjOJ?TJE@tiDt6MU|%=K&GF9<y<E0=^d1(#wB%H4TBCsVL%I#VFQk_Q?zoLT
z?t^Q~f6F6dB~s7=tFmv<cd?UQ9+SQ;RL7qzo|LIe61e$nBG&B%@W?L<)GMW=s#J}D
zMM4YF=mC&Rb;ACgeXH5w0AHriGpr+99(N`8)V@}{#}da<K)!cE+!I{nEl3p6IgBI^
zKXzHY?a=2u_i4$Z^g@Po_bHD0;6Z?vSMPncb75{HOErQLVqbe7<v@ZgDkW)7oc_&<
zUCE!U>{s`cbhoe~A(0Gz(bFKqg{W}=4xE;M+k|2xPfV-cmZnT<PW9Sm0LmJUe^<6R
zV2*f`((8zRt*ldWoOty|SdnQj+Hbn%k?xT7o*;KN-6U()R?MQ4VP-kimz8i~_%<k;
zq{EVA1=Toqw~~V8rc^L`^|>&BW3;GsRTPvx=a^N^dEy-|xz5b|=2W(b<A>?b#{P(s
ztcM<9cNnY!g#v53bf1|6!)hOHJ1T`HXR(OWR!-@nE2wj7K^DT5m`2D`Kbe}_(OntF
zb(nc#k`84*hsM+5WaxM~KVgTlYgVz|mk){c9CGJ3H=#>MK+vJ?JB{PRw;|=fQ<VXN
zOKc6f(q;nYI_>jiifI~w{b*aj!C&3q*|M4Qh9K|B_(qO(PMNXkLV$W2TIP^xJUCG(
zp<1n5Y4Ate%rtmpL?9RA7}CG>QHcfhu5MA-O{EknzK3Agi+?6j_$d)?>d_*%YM{er
ziM*sWP-2gTKVsvBWG7v2qqz*Y2TH~~@GMbnqd)mmWZhhCh(b4$<*~E@NU7Pzgb-G5
zD;Pv>w8A6Ep=JECmC^ZK<lp#aLQ5u+K230reHCeR?x#-(9#wv#sn@MlNkB5N73S#%
zvS7fba)(_80u@2(Ru$fOsq~WVThS|tgC?sR!gnA*fLb{r^1^~Br=K?Y=ucMM@4sdN
zqQDj{W5dC;o~f&?Aa!1V*M<X-#;uVYvRlvYgWpA{hgM}M<7ky;vL1wn=b(o$TBcs<
ze_EZUxplAflQdm$Yf;|AG|114bqIP>>ooF|#|}wV$FQ3ZPX9=wZo42+RJAq!;RacO
zc3`xZc+uK@r#CEyLsagT#yM}`&d~9QlxO678t_G<q$+6jsOMJ%09~$9E`}0;@4dmV
ze!EZD;1AXjVRs5uVYUg*+?@TZGUJ*47cd^_`xJ-`s#+^8=|XShhYn+YrX_=QmnmLH
zY$f!CDxMbUo$Wrsb5<lh%h}qZ!eE{>qNa=d{s<hP4>aqiKRB`-ru}#6DCii5xZ~KF
zqZyH4r>2StFJhJ%Kp@P%o5O0r2hNTA>otTC^6ri6+Han2%Hue!xo@4n+BupI$Kn9D
zO(EvOyii8>b&*1`&eZLkIeH+XvXS|1)99u4CS60+wr&tq$-F%7i+fZ4iutk*wozpH
zC~m>p)wDU`+Dz6rO(Q*BaJ(-2GQit5CF<)}o3MFGWFLbopvCJ#6<c7gdh=A9p5T(h
zvq-N2cQmvgHUMMj?JO-u#nkloB5zzRfn2oR*5lu&&)A|&lwale&bA7U=nIAACRGJ0
z=V!L7Z^RuHbW8;O(yVqDZ~yRTDn*_O*FG70*f>J6DgB(CL1_pSQ-RbpLrdqQZ9#Kn
z<J2TJs7GDx0A19Eu#vXwWmTWMHYcf=&VFI%&BA+{BK5At<>?Ze>rAOJu$IfnBT-Em
zneex{rx?aZk+@vWu9onwU{)%Qd1@!yuOPV;&5!ISgKj&ZwnAI9G;e7q`oJM6UJ{wn
z`PGD8ZlX_Aeyqa7mguV`3ag(Nka3+ZeKPYZdcjY10E82Y>)?y~l>6wn=n;Ks@)a}p
zpr>9cp{o*MFNcqz*Bp?`UXMHhf}e?b5w(MQRAsQAD$}#*)4bHT@CbnjqU!uN&}!fu
zoXH~qW)&1^dDUFf6fu4_w8>rM-FLBTWeYf^$E$gkCm@FO-JSJOv~;U}L3~AG<mnsC
zuJ7#&YAWoE(BatYf13?{T%@ob9U2HTOUmstJyjByAT1z}!%d@3^Sr|9_VB$gB_(uz
zy+7rz>R%&_-mt^Vbr2W-d0q$Yg@0H;#pHDMg2{r}T$t(ll%CX6MTIGO*U196H&nfI
z_(JYLRUmi-Q6SoJjU=Z8{tu(4<R|UY)+u5&;ag5ce2{zaE~L4~ZZG$jEs)t@1V!@_
z%<;ss$J_vt%VL|{H>Wx#{tTgfobbVAGMaEoVqc9f*E_gy=J4tc3m163k7_Az*W#4$
z?pd~sX>{5irZU5gO_lr`8MZIX(Ojlm6-UINN@K?GZqQqzjsVMHG0baOnHwkkf)5K8
z+tz{$Os=E%DYL<Ygp{OC)|><}af!q71OGk=m=^#J{8;7_$0H$Qf6Z}AC_p{Acry*S
z{o`@J*FX^^wJ~5L0sC$$2f(;Z;7iy_QL-C}D(x-W_3_aKM_WUUf=3zXScS}M5By7A
z^{>R*c$*EA_*Fo&5+uI5&~er1_Yxj@Norr*zAZxPLP^KPD?MJB&L3OjK^cglsCM89
z(jWg+d~nwIj-U{b6+bTz5@LW5lMX34I54OZel|kmc)Rh+orQF$yxj-gp7a#7DpIIH
zE6eh4)*!(`<D=<i4&NAhxKaT`I;lQxAI;4}3_U<XKtrBXy&J|;Rs@{>-lJwZF#@0~
zR68N8q?;k;1aj9MXlsj>I<=E+jDsjrtOK~^ZnJ%fPD+{<orNDzp(8K?ZIj`IJihJP
z$=b3%`YYx@GqFUh^137Q@TD)uwyG!K<SoM-Ezsa|WR&uGz+IJHK+@y%nTT+nqBaI|
z+*UjhDlS%%=IoIXW?KoN6+gK&_AG#wn1&W3ev8_jUbMuB+4$G&Br=)x!OtRfb`bE$
zYK7D_QMz-hCCQLQc3$_QLsQ&1jUXd@n!3<-vA-rX<pu4o_(7dAm7WvOaJHGR_k<i2
zqPXOPaDoiWf%1Gx?%<o)*G#VeGBer?s~-QSDfuCjn$<O2MCFQ~L)d&yx<4tPe(G>b
zx;QGnl?3}wcv$j?auDkc$#-m8pBg~t)sHV+;`lbPS_pkcp<s$sMzs`G1}=Uz@f%#5
z=0<fABWKaMuu?t4LJ6!YQd3L1S&>3Y+>LdwOyQ}K;!!1Osj`sXbB4OQcG+4O)J<DU
zNwZina_)Zsq(EE0H5&3%E05!jX=h91Z$rOi*%mTJE0<coiHFl($^mn>M3Lluq4FN(
ze~Xv@EgBeE;x^}<Tqj@^iI9SG9)giXu3U8=PZI>>Rt>@*FFGs62dTUgq!|p|j7o)M
zsNF)MGi*>PAkzf34EZ5<S5WEj84&0OUKE-kB|xQtVZ~%}hMrEn7b%j{kY$m#T^u~m
z@1iWKD^O8WL%kn=YR7<!kvn`{fvZA6fBMq^BSr5^TaNT@`Giok;#Rxyscb?D!b)}~
zTfpt9#|@P(q1^D@#^2$d=DLp~-&6C;+#-}S40O4o)j(gj-*1_$=YE%oWypJ6ZkQ;+
z2Cm)8>&tfkWHkdZZX&bz6^Y-{ItfZu(yh1!+zE+OPWLP9nTHWGbgKFL9fuk#e`i+e
z6C-fSZD#mZyU)DRa&&?zyE@$F!KNN6g<wLj#M<&OUo$KC_3$<f19F`|7Ho7^%=g>@
z@@pRVInQy;Ohvn;Bb+194aAl*w50N&WT~GL4`IfXD`hAORP94u_rnXTrf4Z&3B{vD
z2n}8fxIosTqD*Hi;vicTXt5u&f9zFgp*16|ji<RWBWBsR^W~NnrBY05HSX)fNpWbf
zbX^JZxZTruZVp$c*^sYo0eR_SK9p~F&#auHMKe2d(&MrARl(-E_)Txkl`o>B@SY*R
z0c{pcx5nbTIL=5$nwjl$rD|*sD+y&QJ#tL(yE>l&W(-Vl_pa1_8m8??fBgZesV65W
z2rG`N>N?vncf1kHc4Jgs2A?^eFWrC?a=+5#iEhk#B?^S490}8N{~w|N1o@I0rMuQi
zWKSSWMf$y=s%+X236%-E;uLKtMA!S2k($o)|JD}6sCzxwwoSgyz04&}NV@#p4d1)k
z5N)p8uD`;JkSlcUvPpgif6WAzFg;m{3Y$bt4U9)tR665aXm;iG<+}c<f=Z>^c~R-s
zv74-HJy{Wi+LV_6aWD7|$iJDwD4iQ{!-1OqP?`h8Db6{0lD)qn6v`2F8T&!udj|FC
z{h&bYFW%K3-Gils@Xuh%c9)ze+te9C!<CT8E|A<HyUdU*Z5j@pf5IrN#T7#>+@B;(
z{aGTjMlIk-ks@Qt6USM*F{w>Rb<YgczcTL%0n{zQxMP^PM+N7sa-Qe6ZVps3&sO#u
zW_IP#-4$GdHQx@owtEP8%WktCcFTqq!UXM38+O$DHq`NHwQ&12irO_paX4`9ZPnKr
z(bS0e;G|p`C*P+=fAjv@^UOYdB<yQX`%inX_B!f)?Opto{|Z90j2iZl>#pV^*L%En
zS1U->m8*NT(oyefrOf1S^r@cno@RDyvpY!DOttMKo<CEYl8UXoSD7vGbzREvuaTQ{
z*)hJR%Eb879SsQXqrO)+9reC)JZ`$&7Jfa~-SAUYJEB0|fAC)zs37<=^~DKv*t?T`
zN4>8vOn*T--sVg_=RKVT?X&};<2?bFMh|iISVw@dNOb?}(OZ6;DD#8DN_A|Cj&1&Q
zpy%be4g)u=<{gXQTVp!neVgg@7c%dS>WbYAF~@=VZW7SVQ}uKDlY@;$D<f;uQOHz@
zQ)e1LaDF<QD3q7E)o#+HunLZFteCf&DLs9?DU5zW7pWDa_WK=8<oHP~(!VqiEtgTB
zGZO?cGC4Mv!Bi;|m!=3m!~vbRYY9IE1q2_^&x@0>IurvrI5?NFZ9Ef~o(@06f6)72
z&;NM&%L7jY=WC%-DrOBHv_VF+Br305HBjH(EzI(WilH?fG*^UM5pKNE73f2>BGIhq
zgZ}<!`c6EdP0!PJ_#8i;PZi!jd-l<@NNl5{560gSBoMiz0HH0DPV~gV=kp1LoFy}>
zAP{^F7}A{}a(Y*Mj_L>^hy1<3e<emxr!{Vg5h!n5^2+_)LW!9u!b_uuL21e|?(|!g
z)<6@kNg?5y=<&P+<w2Cb;_11bQ<&FIP}-XmCYh!k(-GzsCap^1_2Ma_lIN2^@ibu}
z<aEd5AIUB6-lFK!dzGr=a}N<1F-e*C8cSH$6lir8!9;B|r$7THq1h3Ke_y@ay;KUn
z7VHf@u&S5>Gx8ecykP*S8DQg?crd<IErAvcH1^PyY*uj_$54tMTB?1x9YJw7;C9LU
zf)6#b8+c=7AvAL%SNyU8<rs*8o2=1DOK=7v?E;(Zt8D$+rjj#AHWfI!6pm_3)5+R;
zSL3q9()ZgLf^kq7S@j{3e>3q`4^qK$g^|5r{&|z)L|pbGld_V*rgyS0kr@2P)>2}2
zne$2W=zaXRuxFqBl5bs)?q8DDR29G0Z3)4VA8Vx(ofv-_l@|zb=cwcxZE4ASHy|Oy
ziSsE(#H?&G4DF*C@VT`?UPW>hVO?4;V!C>Wz{a3euuPz8jJXcfe^pEqkX2P8pLdNQ
zW+L5mcOlZ{H)_t~olW)SH}h5tTTEs;EBZ|e1aaM9jVpMEKoFBMX>&#pa5QngUO&)S
zw(Y80L?syayRnu1#~N=;rnc=@DNV#cPYl2Biz-|*;XaH59n)lyvI?JeUDtgwA~Q;N
zbP(~`9q$WNLK{Y@fA|NpBa=yykz{9%&#UZ6P55n3M|SFr2$A-{nN*ry;!Ha(4!@E6
zo`3RX#i<L)ly(PYfjK+Hvb?2o<}u}QLr6&#kJ41}C|hEC#lPQR^~Mo(g^)40I*0PS
zMtKI=-lGe9S{8B0OCe_G;-0>bDc!0GS;fS_$mOoH4M}qbe}%zePY$=}gi=?p0J{9x
z{d#_R`0v9YaDh~(n8UxmKKKU+ifQ90w_pzN@Y}x+zdZcZui8~}ye>zBdO0Q1s#ULM
zig+PvX?d)b)B8B?W5aU}rHyyb{`8*M&H?eBR?)?tIYUBR{sfb3?tGaHM-a=dr!3|J
zZShJ*Ou*c)e;`y`v$+$@+wnR61xXqCfTgQkT5Q`KEJ>T(Ce!q&$HqGhfpEnm3jsrO
zXN@0)3<h>`brk%QolEA%g$yvk2gBgEIcqL!E#&9Pq|qVq;Ofev&R>f+u#;0Rd2SBZ
zm%8+B0xeJ1&m{rMnRC@&sB`0Q=q=IEwld}&-V!O=e+ni4-`-NEmXeIEskeAdQo13N
zY#I7iUK1%faZdh0UK53n2<{bL6Q*J>m}FL$NOCVCc)hnI!|{Esj;4?Hm%%6Ze9ut+
z5#CZ721fSS25<QeZjwLlF}Nyr<vP^)_Ulg9Nhz-$H1RIiDXn9`)oWa5ZYP-B9^jM9
z@D*ile}GIiTv#FJ!Mbjfq?%|oGbU|0#DUVr>ActDOzE*#&+}_G6z7_+*XLZ)1{xSh
zo@h`A#mBHfV)GXJ43tY`q3Y&WU3#ik)Dv8wSQI!FRgivRDxTZ&k)AfKY3$d_#VZ1T
zs>K`I0|Mg;sUdK0Fi!om7=wAezMbb2CPR#!f6&}eKMRo!(M(N&8nXP#olQ~oE=}9v
zT%xd%7SuQbWulQzZ}O%{H*Oj;uX*K|fD+qFa{=o9t6Y#@XcRb?oaWTgl^gWvr`%r_
zi#TyXB8i%&OD%WGIj#Cue3hL}eeBTe5oHBa)y?-e$fBY^=k|=Ad7AK*2~)}!W3Z0)
zf1mkF7CftcH00DpJJX7`9bD9<s{9ir(FOZNpxS+rR^MB{MXQLzCpmVgcW5cSIiscY
zWJXIVuTIUby6apr?1|bsSm-9i3^?MvCHJZGKqz@$PzOhWPiSmIhh#{pi=-3=Vd~#s
zQz@CDwLEC%jI>>vK%SV~Hm8t^aXfVVfBi8*=y?bV(6WUP3BssZ2;|H-*MfY>NbrEZ
zHzGx!TUe<HgmO<Avw-8*3?k;_K;=Hu;2Jp`|M|GaJaS;x2Q2`mL=0NTmsKp}x+!wC
zNV2>ZI)6Tkmaf-NNNvkhI?k6<l2c2X?o&!cI8{>*0W?+E5`$6;#)Y6Vg3>zZf23$<
zrKl_;sDrR1O_4zUwkk#M(t#71T2oFMH}~<^Y!P;zE?`c`7UUDG>(?O3Pc+gvUe+nI
zq+ja=sF~KXrECF<Z&Pt-vHZS{g@ntgz@5jU=53X10(6VzXEJ~d+f!5*yB6xUiU>h7
zuq{N{tT+ondAG71lE3b<@Xk&Le+yTGh3B{+C@yGOlKp1PQF5R28l$v0lU@ncN*iPo
zZOnQjY>^i?9Rn<uZ#oCjs#%a{P1*IwASv5{V+qnaorj1)ItPyk1J$%6wQKOwHo@a`
zszB7q@U=~dC8~Z5I#DL7nbZ8r`-6OjsB4k>jDxLZ*{9iA!YS<C(j4y6f5@^O=+7FD
z^mQmsL-P`1$1Z9Pw`S5?z*n8cz<Yi2q-p-;-?mN8QZk?`?&|8&lsCG=2}|c<Hl(>$
zM@r$y;j3OIz7`D0!1|Di%!V0q$Ad(7O?P~-ab`fRH%2kE`9`RPx{)deW)w7aBa~sT
z=uU<BtW}l)si<p{ls$%wf7uSBVX-&Amy7j;6ifMiptXj{nflaBXy9v8vSpO*CTCPO
z4o)UtPUlGItCzSk(ss^w?Hnh~xgc!o(`x0Sfn1+X20mzEknxh!>xbxB`xO67b+~HA
zHBLk5CD)k7>$$2~8dGmL%JkMN9A!65htjD%N!wJknCUF%s<Nd@e_2a>f%5nDy4QO=
z{$!5AUJo1Rd5|g$>GQT0DvtD_J`7RSpvBrm<-#wqBb8?6zCCHJ@QqOVSgB-eUk-Oy
zCJtyV<zcp`t(*tFr52iNXCK+Vf?IXzbDOjp+I0DJS)59DCl}jJ=)B-G)joVQ`-RN`
zPBL2^-%8}9w?9Hye;_@k9?>AH-HA*IW?O;~K*T9Ku-JyBk6=*<!vqzCTRgtm9?r%2
zzuq3CFQZZQ!~y>%+NQ^Ib~T7%)ltSa$mk-oh|c9y>22}@qVtimj=EXrm(p+DBRVSX
zsdgL;s6w=~t3m~-n^ELqP+Uw~KuhMTab%P6)yEl&3#=Qyf4Vz&)t&H~vd@inY>(^B
zjjox4pj{hsy9H*O7MK}YV5tGVzYXS(u`1F!2xP!z-39}~YH+nt;-MWU7efiA61Ho5
z%=f`KDU~wT24tfV#vSw9t^rvuQiXM)Yt*q7jWDrf%xquKQP`l^Rzn_tCZ(?S7=&|7
zxmJwF+vCYVe-Xz#ItiMgh5;SyY8`a!4Xpz`Hpe*mF{E%rYS0Q6nqw!2T<w*YV1|TJ
zFStiuniw!G)J9i|V>oaP6yHJVU0o^PJEXF%!Lm~y9Xs%N62=@qt>dFm?|Z2Y@Om#*
zN)vjiuiy$nFr~1?^6M&B{)B?*K|V#6*4u|Dk>Xz`e~%v?PoVYjdw((UQhxmqFJHgk
zXTYr)4hC!%dtVe`jFf#zi)a^0T13~Cv}vT14>bE)fvS*xf{8n)b_w-dn22l4&J72c
zxvDnN@QPMf!dTBEqhL1bvx(ci^;v>w>gxHLdQ!kfV@f+_OgXX{!0ThK_i2G}MjfWc
zK#G2_RV(AjXQ%)C1*8;~p;Vn<HtI*pFexI*4(Lx&K1%C-Ua#oA)q+g_xFJ_<E^X^w
zX)Y2yFc{u9DiDw|1vlvB4=4wfA(tymZ$<an|H*uQdH5exY2BBBx)&1!F*7hRmjOQo
z6PL>xKgj`vx0M?|JPQO*j50uzu{sn3IWaSrQO_?V0y#97{r)5|m)rgq3V*Rczg!E1
z%bDTur8or0OMOk@T#Psc+6G49*h=E7Z5gtBN&kJHA*GER$vN3l;vOKaW>>T1keqqu
zg(S`;b-p9~=8h>>s&qjDsN#JWl`2JD%Iz6<g;J%YBOgj7@qn^%=B&7q)~D8-<_F{z
z_u^YictJvKiqNs-+7{uzAb-9x;J=WhK1Bo3*B+QVN>oW}JcZgA#g0mQjeW;Hrw2w7
zhQhHjB#^PSpdE{^g5eDgO-2@wTxGzORU0oRLrax`OUF?fZvdHcZ2^sC3f5adWA>Ll
zMaM;J$HJZXTpJVVgiU^ssT0vALk3yeYo(JQRR+wvgj@#{@LHtWSbxB4iDT`JKuVP%
zXeVusONbyt?E+URbrFhk7qAwUOkH59WXK?69b2%KF10uHU6}Qzp$j>9>kXoYoa!hr
ztWer02n;3SuSC+RZz4cX)Hh{`+$ptnO}<lHD?{#7xV2R}RjvI-Fr-vFi=orfW(18#
z&$ZncLB`qwu92v{mVZHN9YiFMrAm%IqxeeVK@4p?2@Ijs#u9fiF<WoYuxN8^WayGl
zjaR$GZ9sUCq^Vg-T?(m6hUHJOX%SX8C275Rkf}Z`W+~N>N@0+ZtT*UVs+!F<!elFI
zBzh}t{&VcIho+dB_%7qWN~R7gQRfV+kYh6;=toY?LSb_`w}08mFih|3OlE<NN*x=F
zws36nY6n$olBKR_vl~3X9M^S4pXjEVDlDJ8XJ2V_>dBTCG#Ifkji=f_r3XL$*uVbo
z#i)OLem<Ud{gd(Jcr<zAfh+!?|7-uee~KHte$c-f9Zla5vr|mJ7IkdusqD<+Fgl0D
zIe5_TU%j1LgMYgx=l|&+56vG<PsZo{e*gDZ&+Yrehw1d<^3Q$${$%>$>a9B(pY^|7
zon1UU9{t+CIGlcXc`<!{J{?Wo9UhI`FQ@(6)A8H>S90tf^_P>Q{>kLz`2Faze?K|A
z_yCSS>~iuCcR9yDK6>;3pF55$#N-S{=OcsexNxuI5`QlbaVeXPOVm4+&*KtJ&foLE
zSTQj1s|I#Wtkt8Wc2h?gG`kE12l6|DM^lWNVW=Fl17scak2^HK-~TjzJ?@`%KbS8A
zuz1*)7;0kGA;pG5kdH3IVWDG(X@;j$OtA@tsm5G?DMS*M2ovQ5Xi}d;<jsX(fxf~=
zvlpjui+@7Vr5YfJqqro45@5OdUHD8bCd=kBgcPA`G{~SCI#@mXN*M?U>LX@=Le>y9
zsthtbdt+gRqVZS#KmPfz4h|uqrSq%P(}UF<SbnU60;0rabiM{_h>#ZuzXq!WcJQ!k
zu@ZC;zHALv%)c||w+^dBDAl3>UH^1^I-cxbz<=^Ou@}02?{ErRJny*O{^is3r~T>S
zbktFMurofNqLUr-L{Pli%T5gfgR~pCeAA0|V}1?_YB!G_^?Q@?(f(-qrr&$E)4v{l
znjS3Z%iiJp5d;DhoKG)n#L?P#HM$&MO^!yFv(V0-y%-&z96lL;s+trDF1PRjTA1Kn
zi+@MC$x>DDau!Ke5gxp0v80Wqg_58SZVinQvz{yxQzvlZOa#;Ef)5U1!|kE*<HwKg
z{ncdr_N2n&@%ZTK3?n(L?8Rw68NE9@Jjc(+f4>_~&JL&U!}RR*Zn%y}2Z{)rf~`d?
z791b$>ko%=d*Kbt-2#~JQ~~?X!c~Q&ynmpOhIVnKIbLI%S%WXmBl!IWdV7<qUeoAX
z>n+D+z2$H>y%oQpx9#79%@7tSz}`|850JBv-6>*!1a>Z{+zRFEg7I%6EU!RV^cM3v
zoN#nbB>jFsOI9qt+;Q>4;^GU<K}&d`wxJO4OGsZr(CTS5ld~izD4;DZD+E#kOMlOh
z)*$o=h;f;93jt6Rmypx3i1YT#30fL0ONVgseWEZ9074{1b0LI+vm(GvLVPU>tNJXu
z7Us!{xFQ#qbS+lod9V=u<`!}d9v95z))o@NvV{b`qlKge3kh^5XD$T)f`$|tpD~e=
zaz#eWYv8z1p7U-pIvyj@Ilj8+|9`gs^5^}d52Ld~q%dDN$Jrcz1w<D(1`y5J36h0x
zu<$pz!p#<5WqIM1?zZrHZQ=7A5JQ<$>bC%mmH}a~Y8nC$t{mIUb9Rw*2@ozvRVV(0
zM6&o5cyx=L+pA+M_|0=}1ei(6cTBo>PEJQHlDk}_Yx*6^!`&#GRTTjEY=4!yc^L?J
zN9q4$d~9Wt_=SxsuLb7U*k*>B$YBB3VpXuN*($<f&1oH0L0;lx`2DzuxB@BST1Zi1
z`2s1@Vi2VRQs{sb(cdixoEJ1RCOLdAk7)^j<$KcPr%!h5TeVmdtdusaX#$5V@*HZK
z#4ZTm)97~h@M&Y?%@K%7bbn-<TC=gSXQ29hVeE^SJNB)kBLV;se~Cbme=ACe%eEg^
zl$=(usTBcqEmo^?sHAmR3AT{nyT2{)+!|hTUWS((?*=a=F2G9~a<nw$Xh{Yt4AfA7
zN^zjlK;;bq37Ng^Tg?Fp043?Z5b4F<zI_+<dc*yQkPw*1g?$#wSby&;bohA}pEfuq
z8=P)qPXa3znQIlCZ2!|+LaMgEjjS*)#huZ%b~PqWzEBViehZ~|5==LO1}fyp%7ldR
z7sv$d3d^x?AQ*bG9aV$?B`Z<fLPm~&yKW@UWRQR`IM9tke3OYljy3ZaO$cRxnr+V9
zAVdlt0V3xpg~Tpp3x5R?0Y2N=i<q1*Rv*aF4DAvUl+4RTnnYr|)RKg(M&or0<&jXb
zvpwXQ2x{eHVEYXsd`%y!yU1y+c={UKJf1Etq_PgH?ZNsKzQ(GgF4|!Xz>A&ISc8=d
z?9$UZtR$_tunw!0V#{|f<k)$bdC`3bVK>_$fOZo*W@<tAPJgm0Sj)NbPq3T$JB$W+
z{*FaavkK@=syFj@tRJ4YW}A#{69zZcyzi1Hh@XkK;K;@Mau-KlPhBiH@*z9OexL^D
zo(Jci2j`v#=bi`WUIv$5hP<Q9T9(;F&(2!RDIO>ps4!4EP;sErK;_LV9z%9+DJ#N3
zW~nu?2p{VMeSaUIv5n|xZf_<;FnEAg-6X^}0o;I@u&Uc!2$k&Y*xF4(aDq)UK7yry
z1E&<X=YyjH@$9S^ynq6uVg#z><^6E10K5|c99a{^h?(1l6BfsaL4Z=;_6d+B1v`DE
zj>C#AdALgz^2JsqB>-C4!r25nr6Dj%+nSF7s(^btK7VMDk;rms9tH=UfGjaWH-i-2
zf)I8pLrn;%0N}&U!2sYO6IlU%#bS5_AR->CfgCy~+0q_`omx})m24*jkzqyZG!*1y
zVa4_xK}0*ZM-Tz3A#oc4k$12%1VVP;iP1ULfE-Do&Cl{gloBHBA&JK{L|71QR2;>i
zFj)PZ;D5kz!t2~q2C)BZ?{U4f!bG$rK(>(t*sKP`yaqTJt-|n-UrMp9)8)t)aW9Mn
z1`j<W4PhG#JyfFsHyN<QzPM<c#VI>1=L*}0!NcBj@+=t+F6O)q{3%%GK?or`6Gb-)
z0bR@Hiv;GO2!UXMo0a0BWmd!ZeEh9A=5T-c<`Cuhv{k9e4)--p`yYOW(&v|9j6W0t
zFfy01O)3+Ymyj<gmme}epnrIL_{-a$Zg>)SlujFDBtw?YOM@gG+OX=ES1Xpffno|_
zS}izNB%7jlB)RaEK9hV%GAWg)?pVwSCiF;Dk@CtDAZ6%ZV1Oo#y#`V5r1NG%UD+!W
zS!NNGF0;}vafTm-%*9hxiE{pKgV+#O+DwPvCmLQklWCw!^F>`B_<zGBgG-|-kEq8y
zLchNFbq{x8+|zn&X9bGEO@&(3;=BVgB|l3xGV=8P+*qf`9^5vCjt0gp6hFyoso*1(
z{o|?o;P#;zCGy|&ebyuSFpM;o5#~OxRGFsB-=g))Dw}3MRXLz|EWR3TGL6S%Z;Y$F
zB!1H&$aX87lINntGJnO@W0qPi_y@hm=!;rR8s|g+r(U-dSy4YzbqSHBBJE&f+sW6_
z4;<CXqe;Hk&i08~(Ylu*#U1flMuXx^K-smO#HVCy5$qdNfU-wPQv)&(HL>WrqoY(g
zo|A-VyVwbF6Ve`|4Jqx$nN-?pde@G__B$oR^)KJPo=xbQYkvk7uSj}#5MqDYA#MEe
z=tU4d_V`nz>yx{b57Dz^1Aba2b6UjEr{i9-#69J%OxtIRD36b%9I0`a6iL%wtKY83
z6|_p7%?GWvIVU7;k2L~=giqy{oZorYN6oS}@n5y98MtQoZhx9l`^IY>Mb&EaoV?uT
zdRGq5F>aJYCV#;d?;M^Me%G0W598<sc|OF9cuu=bsLe5_L1P=yz#ZcnHl>iEu|+bk
zt=BQH)^gL#8EI9&+A#F1(K3P@VfB~)-lipbxrDHgGp=6(X^BzmC5F-V?w5SaTyR6?
ztdQMC1pwx$ke3<fIxhKFUII%?fMJxpT0?yKp5?Mjxql=u;pI|55+u2JUSgI^q-I(S
zTz;c)J{Ao*L)Sa0n0bd0Y9=VBjf}+Y5&Tj@5m0qwH27RpAt}9V$wSpgiV03DB{Cs(
zi`+!oT_x`y+rJ_3<)62Q(I+_rl(Ynx#)v|`{39<}hr1*u-1c04qXYPi{S7mK{i^TB
zvQbQEI)6--vhBD`X*NlAE;2M7T4s>!Ok`-yq@EdxW5XFqd+%DiS;-XWe${`D?PeVV
za63L-a*>Wi1KD#qHZ|o(%3H&dKi=MfF<-<XGo)OkVKBrWZtsNF@y+CKKQ7y8{eGEt
znwxf7IZOOWk76#{#?wOHr}NUE3Ke0h4!vSRRDY(aE+(W{5d6jGDY1Pm=n*qYEoXeL
z-(EZ1J0s<Mwmr5K@~MX}4-$*EfwpU~;VH?bTC9u70znH~5oP7wBk)?cB&lqV-Mj{(
zMaZWjOIJBIwmR)qh}mY@OF=%<+ie*UNazCmcB=SW!Gsq~ZZeEs3=)|VL1~t1<BFHH
z%YR5SZ3Ur|4RZxxmlRKhA|6qTmMe2j#W~)m^_QtSE>tR~XoXIP(IPxySE!h(#fX(o
ztVaDe$r~j66dN=dF~bl!!T%{ngPrWfRS{~-q$xh`Dn~5ZxymPeKBe!E$IX|_gFq7^
zEE+<ioo0X89^qHZ%~mr#vOa1y0Y_%+EPvU@noWR?S(vC^)NP8|o=PoDNJLo+R6`=Y
ztW+{qnA2#eF4tO0(o*SnXc^B6*~l9G&{c|iR-kAitOq04R5x%s0zCo$_I3F7@^<s@
zDG5nLQq*pL{c+=eKq&}_(@`qvICtCcf7`y_{CT9BB=k^yOMeuP_UlHzrEaU^q<;w4
zT6NHYY(UykIY2<tHf@OK#&zj10MVJIatNO)<2>@M>6#K2*HTSfrXz7MesK9LGl4Nd
zNMoROVPKQPyny*nkh>dc<R;Y13{^q{C*V8n1j5N;(z0}(PuKGYLQ_;y3>1Y}^CwtD
z<<<3eA&M_aw)4SsHj=nvj=L)&X@A$xIcj=}YG=z{Vb>BtJ>?z03WHQHh;jDMh&KU#
zcMpvA+?~U15Nk3|M98@Z#wefGKe40Nj<jY<E|r9lTyuOr+)<VGu_Q|r=Ei(Ype4s#
zBWx_DCDl*P)s!r=wCCGT^<er4e_!-Tx!^AwfQ`*j-vFG?;P0;&49@%mkAKqW-)r)J
zjtojL`n-_g@}^=h!}{p^ID9B4uxLa4C!+^p1p{pdruD+a<jbiD{&+5DRz;cJy@Z=e
z2kb`k8dzM^l@@9(G-vRxCWw2{HChD2`IqV1YTX)dBr=V;I~`-$HcMvKXGSPJ&lL~=
z%u+o21`97+3QSY2offxbv45sGtfxzepQ|gtU+!{qDR0#BGTv$CSV5wOvHo7%3%0dg
zc0{5VtH#el_n;}E2t8Usp3x6Z&3#u$dzYbl&{Rvfkz3Z<-*H;yhUzX^ytRz-qu6Ze
zG-vE;$VD%q+1$|=?VWeBvF>Sg8z$X*CqK3oyI#uM-<KagypSVOIe#V;%<wkl#H)Q1
z@rj?CFv@Q2$$<cq9VLjJAJg7}^<9?3Dw#kt$vx7<Gm;=>a606tRb8)RN<;bh3;maT
zpC`I*j;{B0d5;(W_9<`b+Uz9dGyb?V1xm@FR=RK<m^ho|2sp?F!?s!&)%Fo()7y)`
zjL)Tn(R0tG!2LvBmw#`C3QWbzaV$|8FUUrNHodKm<4O~FRAQvyGmmoWTFYGZbVs&8
zx#7-Figa9R7puu%?UNGr({exhho$6X&7YL}(duv?IdIFi3OUBQvAb;4I?Jv)&A=s-
z3cE74>(ZL;hQThn=dIPt69`jg=zuirXNK#a{xNgjIouBfn}6)2I5)>zfQn6J?6~T^
ztu-Mz(_qHdn&!+9-~UHzO<3Y3KeHh+P|Oj2O#=k!+`>t=y#|OfOw(o^W`CfuB@B%a
zt{W)sbUJbaX5{}aq;Y*yCbiW@S#hTcXBk>E&i+96MQF$BV+EfDxBRoN!(&mO(ft$X
zw;g<Dun!dH<A3hG-M@kI5d7DLa|+O&;?T{ivn_KlY4(oL6jm}}+{72(uZa^(TRX}2
zvJQ66nQ&&DT@DWMz73Zeo1+Y7EdvOKda{9PTb|c<+{dUB@3v@3AA^t7arzi^|3LbX
z1Z;TiM?UBMpL?+6+uWh-{Pr2wvmeb*LGvUh`5?*ZpMTGva@XnvKUa1T-|j#$hnMv0
z%XaJ@7RXn)YI-D~@10%!fJ5N|ZOOO*b?f}N3(*|GU4f$|t|(c26B7VN3VXQ^C}SzN
z8LH$lgS#ib-V8!QKxy6N7%`?Tsq<qeXCBw+{%b?qUClAuK6X2Hpwl~H%>Mx~bV_&%
zWo~41beCt0KMnyfm!af86qimsKf?kXcDEBfKhXvR6upm`mtl-Q6ag@oF`fYvmk>lh
zHh)$WNj-pIfwgyCptlBlD|#xLoN{RX|Dn`lUeeee+p}?VkjIuae2L^oaYlFFI^F%Y
z+rRIgj!2zTp23;!{c8vEhn7#!GHJPU0LmhC@87#WKWlBQjJ*uf8c~Lm?_RF_{0YI!
z3Xl~bvjESZ9D&TB%%H3>ELSbV$ah!fj(_ETEBAX%6e$R8kcA-Y;qd&|`|mql1lB6p
z_%&YYUvKZbe|NR`7PT_kbbo)}#Xkr^p*I}`<pH}N|8!q>f7Db5UX0!N@#9-p&3q{a
z^dJV*)-ulw2wJR4owY$>{3{zUn?{Bx4WUv_%dPo?b!Ir55r$94pwDt6rwCoojelOX
zX5mC)W;)0a(;Ap@Jt;qF*?6ug%ARPJ4J6Tn3T1qgIoUAMneqz5HFBd><oqftEp67*
zk^Q0@8>74q9S5(h3qAelg2dTANegjNcG*T8@*La5_fmj}dDTeQmU~?c2N!pM`oLhw
zmW$VZKhk%f_l1WL;)~J^E0dfPT3jz>2Ml|0bFD~*$9@-@*)PNXH~JYgO3RMS^}DD;
z-ZviXAXjrrqvx0%CQ8lSa@p)CZs{eUXJWx?QTxj5K#mnN)Y&1RvJ@_5hs(j5vcvg!
zW%f!g+lA|#;N_PeMn4^YUgsV2Xw>`V+@p_IYhavFY)q}e>!Jq6S>>Fo7VG`>Ts<h+
zkD8EM=IU9=sO^T=a}{59q4qQVR;ASS1W_A>8rP|T+Dd%Y#!Fkj&yaD01EP`O#SJwQ
z=t1-=HIgvnP;-yk=tCSuoBM&BMJg*!x?yIRQ*49jX7Fai%32(MYD9rNtIZW~fvqe;
zA)*7nl)K_T`zA`Sla>|4QUFLfJ;wbY?F!hvo41y5BwpZ+qeaqHPg0~G<KzzB6pV7-
zwY=x?*cU&O!XoY9{rn{XdNY=IIPkZlmvFIpZ-?Gv(wtpY2%az3h^#V(r{|gEnN1Ov
z3pdvkZqJ{Fs--D^-dy8FW%XhKm&!v~$%mnw>%G7(mUSmfHLI25Mz=xk5<3mGWlKfK
z@YGPdwZ=*Xn*iwJYAiCCB-t*NRun(5%BnbBD6^{7A5v%awol*2E$L?#>a1F-^KRMJ
z5V_!}sB8$eq(KGpTl&YG;)YO*L1M&*Uj%AV4rv<OjL++TkKcE(w?U4u@DQ{`DG&k7
z!_XGVDPVjs+9H4&dtb2u$g}Q9@;QIL&w{K(z>1W9xp0$H2B^9z#l>;p&CdqPhJY>x
zhz>AH*JmhyeP@P3G4ohZ9E-D!PL{ab3arK(6-37hSi!1*_NM@=@j*d=4HIWY@dJUi
z;&cI6tJVL1lLkhu!}$(Iwfc@0MosB_52Lo-?M;lD@BLkjb(-{TjN0Ef>0@lDRB2?a
zRH}3`>g^d?8RAfHOPOh9$K9~xqhO{7tt(*X60jJQ^TP<QQU~SZ)GU9Gz4vvX_Y=D6
z5T8Q;*7C2d-bRediKo>o_0$(sTh&_!BdM8c;cf+gC;%rx#tlGeQdaw)gHVG1hPUjZ
z)fDaZ2t*Qa)W~M^G6JzvR(z^j{Sn9*SPm@>rxVC1yd`b8#vT_PbUEyo(J?vmEdbdA
zMy1LLfRvb&daYi413(V2Qt_qRb9!`S3B_8rx8<A;L7{<qbuxU(BH7s2Yu2MhCC4dj
zA?2xmagHgr=;7Qsyb^G@UMeZ(ykG0G7>hY<q%z0(Aq@MfpR#-2U@bDB$TXawWO2Yn
z`3Olusxk5Oxx)1ggBYhKj1L6$<PfsYhEMfNe2-%ikNJR-k1;7$OU0X72t)444{*=m
z)mqf=SyOo?Tq>)W72?|<`mp(n67d{>&Tfc*3>C!<n?7KjB5=P*(+AipV*ilmD?qF$
zgoku|K*E>lVGSQJsMs$wyH?h|_cwdIQ(^Lo&9r&qscN_+AxaY@zZ?@+;pgHBl>IqH
zq#sIvlSQ=)0#3C6bm1i7NwCsXH$_+y=KT{M+sjV0-HSD1E-f4@bFB>J%NaGF0A@*l
zwRggGrvT%XakXK1ii&P9OKqfnt9@a8D3ynwIzmdYy-+>i902!b2*!Tpw2PK0H*U4I
z#EdT_S`%Gt3Ft<*OGav5gS1NW?&w;p53|G5#O(h$A&dvq;^&FpaQw8zfLMkZOSj|>
zuFccsD(A5O0QnAP_6lWgWOH<KWnpcXYm7e%0x&k0an=PBmvne9D3=ebECrYUN<Rq#
zC{mXSOFuM!B288}4OUHWzfZt7B!7bBL8B)NAPmCv_UH8Tvy{s4b>xerMC8lzZ^z&I
z@Cm{34B#2SqX36bmVif)N08?z&EJwQ#lLMHd(F=qe%?u<K!IrkW=eSY<?Y8OnHiSC
zoA5Jy_5Z%UJ^l66v#+fdB<=L;pHJZiAqaGeZ<8>8(oKK=HvRteQ%|-oCi$84q2=`V
zA5%B-Eo;zA)}S;->xBkEvRY{}#tRI$k_PFEl$s?bmYhy}v9t;nQht7Quu736?}Z*H
z20iISfrXvskI3?3gjufCGz#>tr>5kBmpa>k@T|ygHUdk-`>A1;DE@XyGJQK=Oih3v
zRsvXm<xBBSl7>Miyf)c<sidRArw+eu;4RO~JclvKw#~zFZhSWHlzhEi4)25{KgjWA
zRSuL&pu~Q*Jl?!6N+Yx$^+%m2-9Qm^Fvfwmuj`U16zq2W?5K50T2Gq3T{0r!Xpl_W
zbw0=okZRzAcUYeC^?LYZWQ<WIE}d@Y%Ao*%+-k4rEPLHxud1=ZU~^GRKHKN(E^U$0
z!nk22kh|@GlouY&XjwtpAy(I?ND1QyKiJ_>?B_zsA!32PyuLx<?VyKb1DOYn)BF*6
zSd6q$vqW|6cppuTgEGgvuz$y7OB&PvwY{aXw<e7R{^wYF2qw86PSaGiF`Sj9$tIG2
zF5CYxX|2eAL&Ce7?a&Ca?vdGUNpXzM`}dGNM1!wBNFf@iO<67P*vLzcC;BBnmF=O6
z`nPebIgYXa010iHaGsfs4Ctil$V9e>xO^}(cmL8787`@nwQ*KR{Fd3ss;S?-w)Px8
z2*<UHF-JGk`M*h5ls&o9-RnCBz>emBU{lxM=!4o^g0R&IF2it5>;w%AZ=lJn?*hSz
z!Y^BSJba2Kd6CcZgm#SvgUP>S_fpz1aORNAm&2mHVf0&d=s^}GJ%?}K%d*wkXkF%I
zC|TE{asfDCzjATPTZ;-J&O@a7qQbM2Fy8wfHHB6r6h1J>E(yK^st4<YSN%eNXFw7J
zNM(I5=Bfc)#Oyg5rwXab_xI;H9q>9lbPJXgADHLJvw>>*kK{RCfRUB{kVHp}5edzQ
zBsyY5N!ET?q9d}5hTO{fn-bloB&XeKl2b~!zyT)+PgMh0Qc=dtAd3Y8zI<&gH6$-`
z*+Sw@xn`$q9@jM$UrV;gkVBb&>WQImrBA$u#5M|L1Yf*9*0isg{(1rg&2j?u^fyM^
zvs`;CEDcoSPPuF~*ytcZk|Irj3X&R6rUE6>XwhPX0*gDr8-4xf@|@FLx5j5rO<h!<
ztrTtHBSyPMliTQA-|y6iBqg6y-!k5L$xrzC$&f2hnn1I7)Xc3Z9Jl;`M@|#G3cEAd
zP>5pEEpr#(vFx*}fa}yiv=m9ooy==o7dF~=fq=zk({uWuO8rMAVw<iaodI82KgaYq
zL)K({&5JqH2yM;6i?>+DLp_*lM(1n6g8C3U%je~<fSIb!w(`A%%`WexKA70cmTP;T
zEq~r+ocVy~Uh3}v%`p^zc+L)b+lUWJRyu22Dz+akZ|)`}AF(!<>>Z<fk-V~+v!Uye
zJb8}O_+5}Za#e3UJP=3r6n#d>#|Ho2fZKE5zoV-5cM-wz2$-JY9sTK2TXYo9^ryjD
z#(0C~ThF>fUkX&PlYSFmw{QxqP%sDw_W;gpdhGNT)odsQ@CEUgI#53&f676GUypOO
zj^aC(&D}~@>gI0t_3(T>@h*$(+D+bjHs-t-cjU~`nDT6=ewZD(IpyO$zl#mdBmxx-
ze9*;%t!{Q|GpG`3PivWIW0y~n5!#ID_ay_=1~6uTn!sKQP-pOBYDs*yAhDe2EcWZd
zmtMOUL6?$--H>`J$@(j>B60^X&ut-lP_?)nirEa^_hxne24(uJuK2Fk-zBRnMFP#2
z{ZKz0f2D<0!%FAg1qkPp)(zV4-v!vzPg(0L)^oW%v8F_iPLwmR0|aBmesblg4SPGz
z(Zib`+6>gsoTF<i3=9UiV_9PaFm~^k{ks3f1k0F64I6$gezTFg6{Vi;J`7fX%$fSA
z!{CkfeCuKGo(vWO3Fy9arcmA6IpdToO-KIjf5sWxqwR^-KG<vOb7$U54^(D7H-~Nt
z@i|xbdInb@>$6R_D~+_Bg0{O#w(S@+eHq^v)T!xQnMyHt;{q>j=ceAj+5wu=al_~3
zg1yW-*O%ZMTwBW!Yn<V4<Jm`&L#^Fu-}G=4-(Bg<$sbTZH_f1zy3Yv**?cm3;_ZNZ
ze`@$;K04{*fkq>&R^1{ww+x0If|sA6zH0l2hXl-Y*Fjx&(M}iobLDiw44N$SFBV&c
zvMu&|EX7(Xx@MSo&W1A|jopl$!^j<;J$orYZ=>z*1$W`qO-Jy7EgK*d-aoRA>?wL_
z!^^dIwQbm38H;z6?`qqyOIqkD4t<Cle<<c^wWBE45fS+C9){~@KdNn`K#<wWO97hw
zz7+>EE4<u}DgaSP)Jv=QpZ910fH9(dq;D-LzGIK3Tj@%Vrknk*?a}m7UGCBJQgy`&
z=(}pCo)@}QJ9U$rw?TZ@v`NACp2m&U{58j<@jRn^b4vRMdM5Bs3T19&b98cLmuZYY
z3;{5g!G#4Bmw`bFD3@PTKMsE%#TU>pV7t2mWH&+Y2D2JWvdUue|BES-MT!r}mSy!#
zG7Gic5>;e9d8&%$({B@>etNllzI^{sFj?lbuyXqRbrSf1<6pt?rqPoXf*B>I&)=ti
zzH%-NUKhN8bEWXI_}$^BK7CWd;Rt~v1l|?m^v#06I|}b893wS;a=d>eez$b*gWfmz
zzT*Ia0@4<WD8cDppFh8FS}<qcgs<VF|L^_t<-afe@{Qt5ayk9`_m}Vhlwiuq2|RN`
zO@I7n`t{`>J=HQ7$5+xH&8I(po4Px{kN`bL0dixsZUiVeN);w!JX7IO5+J;Bu90Ct
z#^H(=GAl*H<o6FJ7Ab!MM^)&7qURP>5iZVAH5xB|O%xe~wdOyOh2{>jP^eqHBTP?S
z@=gJdP;|`8VZB1{r|2vLHOQ2|H5AKeubBrmj7IqYnwCa0)aEIfNvWBYqmIRH4yE+W
zTX&$$q;d?L9VqfenMp9r`@sr=oPxD#@tnM@WLnB2<z)@C*4}@SR~hXD;~30(i^!a4
zHO+tGyv?2360>(}#}vwI@Ws<N!$r2XGhQx9FMewBeG^8>^j?e#aYV39OY#mjD)K0O
zrn}+kCEvvmnV)l_AD^eU!Or|W5ZLZUHDwr}ogNs~T8c_!?q{l5eu28yJ0cv#qjn}C
zmZOJLOjzHKj=Fzyq|<Bij41D{S4rzu)jddK;(bjX#It@*!AyXJ$3Gr-D(Bg#5wRQ-
zrB}@QVRT&0HLX6_V%U0jEJaDhv>hc3i)&F5_h{7*#%alQ!a~x2a(1opPP&_2cxfm9
z{-NC>#*vaa?kDYiRCP8zJ{xm!M4!!WHt?i&#4QkT&pm&oYZF&^4{vD<QHIZJ(Wi)s
z)=e(pE*nNbMqv8U^{2NSh51C)sJKa_00Ib>F0yh!n`gbvbs6Q_Ed<V_EvX%%F*GB*
zoXSCWxcf^9AA{Ufz^B3wa=RNog5z&>=jft0*lgq6vVuW*WK8x8&3^hQu*I1ffcX=F
zt=1TKygz?%3+O4HGuU`M)N6&?vg;`Zji+ll<Qv0;k^Nh{z%(#pMNdicn_EpcI)E(o
z{`lM;4a|+a!%dLf%bq{J4{^?fw*#doz#Rt$dXVBXU|7iD3%!Ehm-k937?`8fl9By3
zwMYd_DtUxjKr+n32cB2(ffg90j$#?9SncoVuZ%u6z%cJWf@MlGHmaGDf`fQYXlx_P
zgvJhF**k`S=s@RFTr$PgUFUl|XVT!9fjP=&Ja<1;5`V{3NlQ#o(U%}sKOldS@JvZD
ztV6G)Nhzu7Mpt>z`ELGmJLdpI49<<%IdPXUeD<-DF$C|6`VwOfLAlN{W=C6L&Z>T1
z@zzWCX2vABK{te+O0xcv6}giz6wQO3p)W{DMilMQ)X^jTzZa5&mKfJm56MH8>-7(T
z<cL;Kut!#9I*KJMpGifo5$%88Fw@Pl3X@ayw;TNh;V;*9yO~zPUq*Jl^nsFRyCkc1
z6Kq3XGU3<VAumUdo|U&-C7Et;Sy3JK9#6_tNXoPssN3bGXp4#cV1PTO4OyM=!?f@F
zp;5^0<e>1iG8Y=TTM+6FU7`ahFSZIW<bhyD35MMwU%Lz6pwV?jQ!{@nHKj}H@kK)q
z{XufbX78tZ6i~$}WaYCX7lU6b+sKW+l&7wsL*)NLCC=-u?&8+nQ)1rY=7w3Q^<7Wj
zz{EY0c+ULRbvd$t)m>6+W)LTL^f!jP^<7gWYiy}zyc{Whl-1`p87EP@xoVh3WPPVQ
zme5E>azNNTJQc6m3i5x*7~uRerFaMTYH_>vS@v>hsLqHo&D&QciT+wu5+5jsW`M!U
z8T>S!OY(}PCzW>qD6|Pb*n~eR{aV$6&qY@vTZxRa66!)Vs<3&($D(ILtl9LFgoeAb
zK=E2$Mz0{S7Dx_8EF{C+yTMSDYbBW_3|M=4m2(l+iux&7Fx!8rEq4sbT+x-Kk$uD_
z&Ux@285oIC#1~zrHRO!2RIhDqJ<S?`e39l?w0bk>eW;<|%QA^#w5b{ouX)QnKHE6G
zv*;~zyxfaL6@CgOdMt`5jRF5du&DB!iAS>Dd4l+Do;TmAcOinO-6Xe_wh8wVB5epe
zsj6zUo+}4)uWEk>E|J1ADS966l&CY-Y+s^%Y~DE#I^njhFwYW3^!a#L`JTEm)wDaR
zYjx!;$aeIO(pHB39i?Xurclkh=R`kUsqQGQ6<Jdix>wKMx^k*gZ&ORoY@Sz3u2A<Z
zuM}CS3VjgIsY-oYfAe6CA!Lo^!Doz%2gVNenU?CxN{fHbw4<N#nU>O8Vx9e=<9(-c
zvM+QzuNWkE%yM%8&n2F=l%B}*inH9*-Y%!<*7$bYX^zngIn5?hEKakcmit@vDn7`L
zhMyADlA~!#67>N90SuFFZZ^1*4?fz(SVasO?<mcb)GcKAKW}X)q0q>VtT%QPzp%B@
z&2*==(T#up%eFRpu5PzBdak+}pU_X09Q(ECTP4SCaCHjoT=hC;B>v2KuW)Eo@Ples
zn@)8}?J$fSHRCtsMW1i-Ds)_{-b|(va$+z09Mp?CQ(t$o{BEDM*BNmdK{Xv(A@t8O
zm+?(bpKmKw1)q;y<w{Vbu4$ElD_%zBOg#NZl8JwnHb2X;l7d^P0mLUosyPVP9V1&b
zK4=&>8a;0`xz+(Bsn|VJl`ce80Xym}oY29pZNHZATaCC53R0!sK>vpz8-9lo&QZp~
zYPPly;-LhtbPjIYc{izftTl9tsTzvsxnvW~^-_u=*SUx?tGX_XEEH586=I1$fUpP#
zp{ai@PEpEAje(v?Of-;JmR?mE`<kAzG5#&+F6!l4wy44Kq-VJO*T2wdDyPj!Py5=G
z0S~b{Km?CA(68e#tJ`jJfOy@iekBoBkr$_Uc^L1~pteBtGj2V7s+u<6Q14yiQz?)6
z+92_-JY{P_;n3~n^C10#MvPzvzQuvIoPB>v2nO(H@hLoFIqCXUG&@o5C0$!@4|@wH
zna6v1@U3t<Z?9zTL<0%AUVaj<Gpwm$K^Q9YcTbnAGDlI3ZQw*ADYn@d@~OP<`Gjh{
z@SeIkRWjMgJm1k30+Ln#C!I)v<u2Azgb(%r)d`o6I_W6eP)1&DKV<GbEc^nX!BT%_
z*i@mWPn#ZDTGR6(q*N^4*1r;P&6%9#E{GoD;OcHCoM`&ag;u*dn15C!b60ad)l{QC
z;Oee^lXQqCEY1X#=jT3ZBjG$#MyfQg$w@kONM$)el}Gp%UIfew3x$AL%IUrGOWSJ#
zT|^D4--TTAi7rZHFJ+Cg))Mth(-VKy$f@9oQrswVD(-u2=C;(UP1Jbti7d}Wog3Wm
z<`bhs9#PJ26{~FfeiarE?ZpO<AHP?IYN}DLU-dDOGKR7_SRNHAQ}to3RkngcJTPHx
z&*w_XK}f}f_J|hoQ|2oCh4;181LUtHqCnZr^w=I=9W9Ynhc;SF%-2sjagKk^)K%8*
zZSPqW`04l$5k)i8c6e&4rn#Pf9;UJ#`Eoj~dE6de+l1UrC?e^8p#|AJ`m29nN<!cK
zRxTDTe?;@!OQyw<b<vLCMx6Oj$0!v#+lf{iM(HO;?cgsth?&~m?L=$6acZt~EZmjV
z(vizjhovgy7x1P4m;<|wHw99j3BJhN-B?uzs>Cb5HoUzjD)`bfCjt?eY3;utWy(h^
z7W*DsfY;nzsS6_7wPy#rjq5QHc!nwRWYJlxo?dRvylv6chLec>C&s@20ZXe3PnThg
zKNSHnmoal569F}sVRQs1mmnlG1ee=hKb8UEmvUb}sDHg2ZAm?7&5(<c3P`ID+{w~p
zI_XKhk=6|<nxt1t#)(52i;klM44p9@#$ll>Q>aI*yEzPL&yyBqNz7X1`_Z1oTf=l-
zH^GDTLuLN$(qj{-R?wDq-Z{&E=Ea>oRLU{s#Khf(UifiV_}1^G5bJvcz?4uc7RHa>
z<h2B5oPQf<i;pC!m#p3Nztw6lUzY@V!!DC#!U#7rvPitGmc!dZ>HKQ5K8I1s_DN@a
z&wCzcvK&H`<Gv-x$MrCYl*Td5Cmsh|=Q;9`=xyzGUa3sGS2(#(Ho!6=CXAUQWZe$4
zXBwFlX2PUf-!o{^(x||5>jTnpz)3!$Li*j*Z-1?1H29bF-(^YAz5U9PKpGx;0$oux
zTW~s=uBB7iNvV`1Q)CNMW)*l<+BHr&{cFe}MO<@{*t;(ugrM<+%WPQR+l5BF{OW@b
z)hI&5R}~>We!j`L2c^h!nzevw7%SSj*_{&Y!X(x7>Zr62p#m3eb_fL@gfWK+s3Q|a
zqkqS}7#yh=#N7(Hy7nj}art{>7`bM^%|sYh<bf47d&%QGWEM|J<GFe&2Dxe3W9&;x
z5%lF+1E@PJk;0KHn+SvCXQSr%0NlI{k{UwtpfgaFJGOS(6vyd?*yQfR9_B@xKyfqQ
z8<jLS?6*)CQ6#%sg*Cr!Gra%>;*QWbVSlqpZ>o8ea?=KFJ9jQAxq%WEr?|x11Bvpj
zG0D0Kfs=H{QKEI9O21X^MLLcXPOF$P0CPZ$zcA9gS2%qVRw%<7U@9E#Vbs^$&69|k
zs%iIpEM<&Ae|th3b~S<*ZG&eWc8tViN~4d;tS#}IZnwHHJ__WmxC}VaHZw8AWu5W5
z8H0a;L?*afU!ffRW?9QEw~$tl++E^E2Sx9Sj8?^*x|lYujTUd^#Qb_Yw)L?E8KD7c
z;NeH!quVj8RumhF5klafGULjF!dJ)*7nUYYjRvGiIC4~jm(rWLJDERO%-^2a%=g{l
zkrJAe+uY_6X;*fhxA~=jG8!<C@avZOmYsk39@d`D0MJ%5y=Mn#ZTu{QjJjSnQ}#_z
z%v!)^YN#{tc9LRanUSy&vu|f&q^nAevk%F{B-uHbAm@zTX>Ta~*RR9BAD=J(o0bRY
zO%tB}^XH520410r=RnyCLv2of{eJr6<)0(iWQsZCG$(!_kG`dA`7NFGIZi9#S|fjU
zD-q;=C~H8}0cYm$N7|cf-Dx@I;DPHiL?B63I6>TbN<(}-)sL{8uBAO?Bey6>Ck9NC
zub&$_!L+sG=pAQrkJ2TjW6n&NgjhePQ^2H>(<<)ri?rp^YYQ5tDUv`y2C`;=oga_T
zMm@_ji5AfCjl2;KnS%+_7H>Fb!kK@GuH6C6Ry5;zN6)g&38_eVuweWV?FAv6W@LF6
zk|F3~q4W7D0jnQ@H(g81o2-izsi6IibI@1KOVGMx-{D*3TazY0<2#^Dtn>W!Gp8Dt
ziH~kKAL9Y}JFQOO{hcfsApgizZp3R}!sXARwA;Yd$2DKVqsd!{duXQ6;7flfNt_LC
z;oRp0EB-bAbMF!fq-4UDjl6pn6Z68H@Vg?q<v=rKy0*#m;w8v-Zo9{nUd#OVE>q5m
zmM%b6^WX0|A|>F&>WPD8#Y>-A7t`>uKV8!lswzEV$m;uO7^587Owyw<jM2;ylAjU9
zDCd~1Hu!)jW`J1kFqg+YA1Z&QoE;%<c)sk1#ue?_{OH|oi+wu3jglX2g8K1zsDl)u
z@`>?-1E`}NStB=&mP3NfuaDDVk<qi44u3T|5#5^LCrD6U28G;7f}+@-gzgp%s>!CF
zp@%X@!FSW+r4wg*cARU5m}pru89QBs(i8!&MC9+YNgnU0Rt<2T#p{1PN4{IlH@9hC
zwhQG>@VNUesVSxq<Jz7(V&8XA>?G~ngz;HpPCHNElQB1f86|#>F$Y1Z!jZPT1BjQ$
zO}JMn7%AJzH?sxr20vkAg(d*p41hwVj6Zu(>(~u2ZN$h#{&Z5J6z#GV9eQ39%DHGH
zCktxzP>3E%siaG;NmhT?BwQ>a@>cIli9|f8C2!5Rly}&+M`Z8Gt8|ty9x1Q`!~=ql
z0pgwDV_eab_k(jV5`&M8Fg)d`moMee(i&3L>h<2!wqp{oOR6(_Z>rWHwPVw6({DH+
z6_V)ldijl&2l>Qv#B}Jz?zz9)+vl8=qDC_k_N;Rfio!i7r%Zo?lhNb$r+6oS$R`g~
z_8#EM5yulgqX)X>sSHgQdKmnSr%Fzz2Nd+|hNmJ;1vN4Bc~2!NZ<c>UPbEpXRpY7D
zw|Oc-*%BP;sRoD#c&Y*79Zxl`=rDs9jos20Gu+RvM{fpmikp(|&d(jNwOTQ-N1lL;
z5Z4UT<4P(}GOT~+yMG%c6%<cnJu5EZ<dPJ2N%6yy;s!Is9GGH?^8bLiLU2tYf^8|w
zdf^J?z0lr<D{xcM%3N1|NCAt_P3=hOPp<HmhK$5WL$1hZi<XLMxTc~fFzuml<XXN9
zzs~nf;V>EX%~549QaxMgSqg^d-$U4}fswQiQm^fUe;a=zycd9t4#j%uebw`1wx<m)
zo79$?Vq0R+G3z}ocxmb%Ba@oV5qz~ozyYgxn?mSX7w1`yrUc%{2{(8Hsd@@-vozJh
z=XyBUC%&BMA`{f%301D3+Vb=zuhnjr4VB*EIPcqdB%NS}nkF9oMD<q|D6%j;4NVf3
zSH9HpsDgjZG?kWxuT?=)ePZY-^XuJJsq<~$h{&&tM66TYFWSEDca(RssNCUWwdLnm
z2#OX|P&Sod`F$4D152P+Ogk?caI8W~P<Gr#r;(Sk&;-qwg(l2<-k3tY^UVRTELOSn
zE?Z~uo6^)xF6dH|c$ht=%*(%P)NkfD*bY)|ra&E!QnM{yM3_*0*`Uk^?DYnB?Xo|8
zQ{Mi67PMoXmtl-Q6#+7r@nr-PlT`jFm-uKup8?som}x(C27edk4vc{*DHgnWy&Yq|
zgx$$(B^mcivAyA@I7U~>AIWU63nr)yX3x;MNZ#-JylTM|*k?W`_NzS}R7$xyC2kj5
z^;S8<Yk}`EtLT1OgsZ;Fu%K1H`a8HPbUwSNlFBm;aY@I(&68TQp=+Vzda<Ah%U+`u
zq7g~ecfKxQo_{k~^*|TdVZ7`jJK$d)P46R?qUoMlS6_JFL$Q<0NV%6{XPI;ET8j5I
zmOYnpscRd{_yNUIl1m#H52gH4Y9tQJEM`$8-Lq<6J5k0np>$(ks{&lX3^ot(P_Ife
z^4o<HEv*EuM~ado^SRJWJ5y*0mW63GRs^jGN4{FgcYkp>qApvEJU3l<yj0yzr`l@d
zE1Z12f}@wNR~pCxMeD^*qC}=ULxRUU#x!~+o~+vw;<GexHixzkVo9dMe7Pjk0UuD3
z=`e1TWcu~rp?i??bc3{yu3~Qobp3z32PuWvxerVe-k-k|tL~$V;?RNIy+`{Xwl?-E
zabE*rZhuPbaJKD(!veU9*ZslYcKbCR7q8o4*9{ere6!HeyJAR65ZXfDBk9_*5NwZe
zP122sQqV8|4$*c__;*r8+S-w2(Shp9e)!6{_O`_@b)%EA@s;>$=M`SJ$GEEsC{h}5
zq7DPH1xQ16wG(f5VqXQ_Dy6-EGgjKy!B2;!xqmd$;~V578kcBqCiW^`k$A;Z9NC~<
zO-#ZIW|Zw$E%$G-?H@^96h`dr7BMNr@Yp4DntMRx)A}jRYgA20M{rmHhlRM6!P`QX
ze*q$P@jcciEMyD=zQ^A|8EK#LZ>36?kR^VSOb)skRybY)-gC`i*|&$~xvrA@GwO?I
zJ%0j8Dp)#M-NRvExz>0Vsj|D}p!5Tc^flvaTlQ$(>gL!~T*43=C3(>1c7%jU2dBzK
z&za*yY`1)+*g@-_60vfU(laIITlkkQR)o`#PGh&2bY`diXUL5X9D*vn9oR+dFE$@W
zietA(M|oT;MH~6H0;3D?{ghhSnbH|9)_*2bH4&lsYk@yq>|O*xWE%WqM@6heNUs`x
zrJXmep>vW)Wh?zQ^BK9}8ZOG3uZteS*S4Rq!epv6(=F!vvc^EyGM2^%40M-#q~K%u
z5)ft2on$MXlr&gAKc=;gw<YJg4>0?;rc;tUP#wpV#@n22G*6;n$rc0{tYWjjGJn?F
z6zOxBt;q13COnH4rsI=RSr<yTVrdF4oKf}JVEgd=E$Qfm$feJ4tOARtdP;8Xi%fT6
z*7kjofIL9Zaj`8jwLD&~_J|m!>ap^dVxT#bDUgjRNP*~U{iH?xG@FXq7}S`Ash`Yg
zGo-!TS{w@y#?43Hxqj(FYXq|Rmw)!B0&%Wfc<FS?f<XZsLi#G@ii+-eReLHOQ|-M6
zLM3ipH0yIoR(e;pec~!FRekM^Mq8L<I`gMW^yi&YxKYhD&vVG&kvkP1D7ex>Jk9N^
z<DE8~s@ke?HG^g|@1`%WbX_>q6kBnla&~&#*%cYlhS|AUu9DD8PD)70Kz}Cs*}0M2
zJ<VrZ{l0*T7;fLL&?H{BE6HOEM_ADHdT}>0>ON>zH)|!ZiFNMz22kpSwddK_P<5j6
zLNB<@bKUixeUiVzBaVL)@XnUK@lo}!1Lh2|fw8}?Q{EaDmW}c{FxcZS`SS{b8^+y{
ze_o-(cuUhKc5Qc<){5U)t8DI|59rwLFm7~g_v>#oAkuFHihDZU)<S|6yPj5k@!q(D
zKKLinu|LQh`R5s0gyWpHWH(*bNZBUpd1Dq)a-3u$gyt7(>gE=_7TKT|{+=2*$y&Qe
zDn!!Fm?0x|=?}lqX!i~#>VJLeKQ)(Oj6W0tFfo@gRt+PUkUcjJ0XUa|bO<PaZEqYm
z4*s5BA>h9BK2#%7q$nPW0BLjX^^o?0BnRAGU=)tCN!&WS#(LAH|Nb6o))TwCZv3(~
z&;Z$RCK`!Dkq;koluJq+QposM@LRf61J&9k4aKZWEl44kG6j`BE|;K9yTqXt%#y@H
z)A2tgSgsFTs=;KO1ivZuSqQ6tW&bIuaL81;m@$CHDGAeY?RP1METbBPhjrAF@Y~h{
z@PwKAV$=eq(v>HJsQoeMMw$A;B}1&-MG8czkHwTkJTAcyt&e3qCft``+^sU|D-g%k
zun_eZhp2(9cFjr)C0U=BDHIw9aDlj9*g`3(za}hMnEPXb$7HlDG%3=5`f8Fw(QLk&
z<WQP)&7euKHa=DZr}5oHA&w@&FwWRCh**<}`@+=%7a7F_v2`VD4lH>bi^r;DUkx6>
zjk@X;^S1S44Wf0`@v>~(2Q+v`(mt7$fVsM;1fLmofKLc>?ji@Nc@!mxGWlwFSXv#&
zYN#pod9{U_OJ5ipS=U~F*;1%$w`@7owO=+0wT>?j!&J3CnPIGzzM5gIY4X*uz3T4k
zV-=$KyczBZOZ#NmoxF^m+BhhW`-UA+vGI*-P^CVD;cgpU1DlR@$AvYtOyg=TG#d8}
zTDQ_S7J(4jX!V?OfG78bv1ZftfD)Rm2XHX=1Q$K7TJ9nOtMs>j$Z#fO<_E5NM!4zt
zLM>R${V_zAje&z$3)Z^J<?N#M)eyoSo7QFALD-H%FcWU~`7-t$F0Fl*+{0ZTeD&4v
z;<uC8@Mu0?tcK?wUakD<$<h4hFnqmOzMd^#CLm1OAO10XH+=Ilj~CBohpU&QDHcO$
ziA4lQD3v0?1c<MHn6>Q>j{>HDK76})u^1kQFU+i9=S8eqLR53De|?BQw^9OO5DngQ
zri5^c)Z=}-615_Pq#@D(bs275cAx}sCaPP4H6q@D4-^?TCecV#HkDB9Td9>9C{;?v
zHs4-}A}M+RoGHP%#ZviFC2}(sYj7^+N*KkwMTza}!Yf06Uat#ijsl3FTwr;)nou-H
z@@!FA_a9t%kKxfY{o(3pF^59`e)gUJ{o>tfb#nT*VR(DAdiUW~JY2jVe){nK<jdEy
zUxx1wR`33Gvifemnl0ZP9M0lT$HS}R#jD|$44;d6I9(nNdq>B!y~XnV!RmDQ;mL5i
zIL5HQJSp;j<rDT@elWh=<CmVS{C%Dc|M~C#02@KES@RFa$NSqyAVQ;OE4CKlH&+_k
zBsySRVJIcR#$8E)te6$ha+RTiq51sKFnqH(UM!!Vz<#0l17Uc2u!2SBfyT?fe_Vb0
ze08vzg*@KaTg+EjWY1yT!PHmTa{wbQ;{_5Q0N4C~;^D*L>2h)Se71TyJbk=3yqJAl
z?Vod~rw4Cm&;ffmU!C@BUHcr*W~Yk}%fs2}I)nSY?`N-%4!&M|?0#1r(T(;o!xGP8
zbwEfT)_wGJO+>m14_=N`#MK{}$eENUi6*JP#QwEiK2_KlIW3h~ky=4dq9T9SF$Avf
z^0Rq={PpPPqZ5DWxLCd&{A&1NF+V<<&kmM9o-7w{mk00PAI;x>^5dH$1=JR26Aa7*
zAu3uUJ2wDQ01Gye0{H~Ip#UzNR3PID-yW~djEE!{dqz{>g<n&SCVWp?fYin8Ijcnm
zVJW0A4PsJTL^9q`f<YLOTv;jz5;-MkJIG;w#8I4>!i_pbQ=~Lx4*)BSb1fY@8i1L9
zLIXE)-3o9QhK-<6OGVbCKN>1oA_yF4pdrUyZ7BQ<Y~~8<a|5L)R~ZQtL~x0#3?+7a
z)%!%kVb58Z^S^!pGDc2{#|xAll=FC@T|Hg@k2454dj^csX#I@#!e~8Ay<p1gbsXt`
zjn=a{yOq>#gvFdUvAEdHu$a$dagj5$Ep?J~lG+vozAm02ZUppJT8UmLg5*YUUNI06
z-2x<FzF>g_^md{QcAs|;#7=jruL?M;V8Nie@IEL-$O%O@v}nYj;qV@<iaG{*%MOqP
z9MFsH0#@v0F={nH2&g$YVQZU8WE7Eqf-)8rkYr0BgtP--2TNkaAxVwmb+H3Hd;|rc
z6Vy#5P`h($j4B4PBnn!PF9I5WmaOq60V)ki8EoRN!~R^*-Hr_jZ$_Zd%_30Rxd`Nm
z4avlYB;P*nsO0FCyC7M*riC-gAmF_xu(T!cDydp&%{wL3kij(X!re4h_&73uA@~D#
zck?C~3PIf7;%)FBR9{vRm87{5?V<uio^?}z^%w8|P>m$0&l#C~Oa0dv;l?P^QGOBN
zlIWttOSw`8zc6f_!82aF@KuIVEl^hfJONT=GeBy&S%6eK7a(;Ci#invm39yysYR5<
zAV5Gs?~K4bU?O;b<Si+oS~Hb@hP?uWVi4SS2W^1_Qfc0k;OL2&3$0TEbh6F(An@-Z
z3|`!dBA>KUZqydQ;~=x#OIGn-qDEzGK1`6o?r+l-_~(b?ql3psZ{EQ9mcJe>U%QM_
z|92&R<<Dh}b0a1O>eWZ0s|+p3XV|K{h@T4~v$Jx$*=GDSy;=NNITt^FIu+DfCW$9$
zB7iy(K%EGnwn>`ORGZ?lP4U>KHCovPWduy3`uEhO-a_wf!-mw=kTuHPMFmszZ!GR+
z?9G3x(ZGwjr&7D)u()}S6+|s6QecL8ze94Yj2sIfL$cqnmLjdFWoFcKwcz0$P}|}=
zB%gEufFe-ar8=C>y2(&~PR1ReXa<2+)G8ngu{5vKIFdI#Q81_6?!^LBfu!ErUhCi_
zpZ)q09qy)K;$43K+;OG_3rBZ_%ao|1Qeg~*%bTI-YP)M4d46sF?Zf-BfE3DZk$Y0Y
zT_B<AJW~Q&f$gzNl@LeAS`EM05ktkd6Y|T5MBqzRcshK-5UBot4Of9H^}dt<c&UH1
zLWr_?h!XxKPhqIXx0Jpp0b0wzQK*<GV{3M(1d!ERSSTVRr~d5_+-p+_u&czmfoLHu
z8EjU6$H7~>5(mF9?2Lm$g)JO>%TPkTac=Qv;ShQKrQtk>$m>O}%OU(b?I#Y=Cce-n
zzR)JVVADPObZ_E+5H@iLJL3>JrAf+@M3Yn|i6==WsZNsKo@-=F(cqC)-6N4hP#*ht
z{MwN7Ip1;SDvDtAZx!p+xq9Z(OIN=ezy|vV%c<2Z0p^U$X8y_F8^iT}P|92SHNg-J
zLpXis8nqum$S}wzKv34El3*`)-CDMPAH8<=g!O?UXG$=CEP*h%Ob4qXp0-gJk{RGi
zcHt3L@Z#7FvIs(jSB!#qsKu~H+uC#DBt}KD_@N?dJN)=lGk8`1c&uP?u9Lz`^`26O
zcL@EIN=ngBP<bDj-4e-SE$vz<;}(^OH>(ks*~V@uGC8UZR0y`cTiO6#3_#4Kf@IcI
z>n?(WB-UPk)50&2MhHHFO2<zxMcNXfwZH%kyM+fUJQKNW2m3*cya`6dD9T)k!m$aQ
zLz+9sJ$WW@v7ggQB^LiYqkv)13hyc=yC1mp>d?=JtySRa<<{jYph~7<4BBN<pn;`b
z6g-H9rorG3AXCYYS@eTNF>U)c+p@s|*@RH;q(tz4pB*?Q5~owTQ3vnmi#L=m)8S5b
z!ZQ^3wOWU30X(T6lQcivaXm9Q1@u#Dm-eLYy!(1>vPWCUUcMpOqjL=-Kf1OiSb3sf
zd7@u=qF;Fz^b6==r2Sjicz{$9O@4SWIpp1wFHa?_pomwRPCIx}Pg}~9J1X9xg&Y4&
z3-;zyFkh~P)(WNq+So+Y4~dmsgtZ$!N$=$Nha1m$nR_FLZmPoPJ8$G6oVXuTL#7Cw
zi9rbh#?r1%>^xl@ADzDY5egmna~B)ZIolsql>Z0uP-nK6VT?Z&0Wy~%nimt7Nr6Aa
zf2LA;|Ihbd-y|_CMRe(F`j~%xe1H4j+swW*QfR49|NZMN-2kJ63Hk(41Z7Tt|9<-8
z?YEgKS58o<9O<03gL14F&YF{RQJ8eAC<l;XK;^&<Unm%$kQRjo6ws<7hyWzf-v0jc
zG+p`EFI|R+st;jaMg5#!lUiKF=`zvKf3{E^t{~N)?oqDEf}0^!^z$kv7QT~AHp!ur
zEW@f>+R@qyuVJQWKS%?g;8jNv{X!!q>$m7V^QoTW)2g)Ig3V%)7B|wGvVlew!f8X3
zQsn*fqC_KnTqsG$Vmf&>D5_U+aSpabGTI317c$8^J9)R44bgxUHZIIvjcV?re?bIT
zHz=S3aVy`!(m_d{;6|HBI!^jW+rt;}o0V#1QE%pVxU!>bbSUV<P%AbFgJ--R*J?Ji
zkCBoQ?<>_S{D>-vt`m3BI00ywl`<Yt8%Q>iRX=Bxjn|3dV3Du85~oCzex|DB>JFm&
z7NFc>^rKU8eVH#UgYa^(QgU(fe`A$5t-R1`p&vYv-<;IR&OUw&r!EM5t#JzL=RkWW
zy!Y#zsvpOx(HQ#7lF{|UYP67HVZhdrH&rvkiCMH}DR0aTzQCLhBCK#{lRHUaf1S$|
z31ML-j1S_FbEj&&Sqtuu(jJsw7zNFF`5xJAN>yT7EpHMzX@GQ9lI|=;e}2m&Xvjxb
z=b537hDaNpTm{OJrARUCQdg4oN|{e<8}n}9%;w%`*RE}j;p-`LO*HylGBnF&=<$<Q
zv6Ka5S#C{k*Mn4ElU$vpy#N`{M76lrO6^i5Z&ZM<mgzvZzCyFzy|Lh+I>!*-l|R>!
zS+=@ceXSgL$~d@?F6BrIf3=<LWwrzxFlgA=PDaj!kgl_5#((~m^<lcks$MB~%ABOM
z*t)cHDUGf>oo6szCoLN9Nowocde^F<&OuwfYVA_9p|a;)&jTgTc*Gg$pacERGd{A<
z_-0jPI|J3i%8G-tYLst-PWWDw`xmyd8R#`@#qU6;=T@%kdOb@He?{`NUld3eCmFoK
zAQtES+sSLUe1erqcq|$wN@b5+wd<Pq4;mtWv1(Ts9zdGr%dX7+4ebY1zd(9+?5u`z
zvUGIQSu0m8+uyRE@5Y`N^XOccUwV<Y?CM2>6*10&$jgICK`03GthBAf(uL)TRRD4v
zG{SM0xw@S+Ql?7ve^RQ}91>Kz<8z)})X~Rh<=nfgXuO1aH=Ro!oAb6fG6cyTT<5bo
z2)Yu;Rb9KID4Y-5rrY>>+#{!(9oLP?-`w`Lu{jU`J-qAeHnC#&i(^bjEZ6ewDCb|p
zQ7-o<dB%gv(g@AzEFA#pu+VbkV8=CYoUcr;xl3~E>;UDEe^aP2H&WeN3oC^i%l1LD
zu+Udyfi&x>9(7yYGkwOY8_xo}ZO8~;A=3<7;kVURi+uzqtZEgEuHmPwNbbUtJ67ke
zo$MT7lh;wZe%Jx#l$!)dHA~d>`~|jAWnlIP?c2R4q|Hs*!AhW?shaG*iWUKuBNB_R
z2`y3@wXWShe_ZrT*_a`vjT)6INsASH{K7hNr1v#d<Un((AwPqO)Vgz|UsIRcb8-hG
zl#e3~^YOZ@!3!|ce(cnxDq=7MGmbw{AZdx7+Nx<?yIx^Z_5A5Isit1Yx>eZ{!1w8O
z4G3-YbF!KsAo$O(i5@gYH_N)dW&sF?p;Kl2^73Aze=C_qhFt(*7Maz$M>vS+;~>sT
z-><)ub5eSZ6WywA;>4O$f6JT<S609?t27>+)?!dM%6>yn@5WlK;SsC3JVCh4Z4?-B
zftbI-XB5VnTnEHpcAUX$RIN}lo**iL+-{iv5k65R=ojtIK|aK%i}>ljl24mT0nhpD
zr|a|le}We(gpD(Hh+pG`);*{+K)sxuBLH_QP*S@~-aaocI%Vx29Oy%0D-cGhe?;sf
zP+fu)nHia$iSEGzN&_wf^X$s4^}AK9f&@tFd3Kh)!co4}CpEX7L6lz0Y=cI7GH9Kg
zdst||6}M*z(kZp_epQ|#o%1uLoTsxowLwa)e?D5z{j&x`=b-B$4Ap0N8k*azh)`I}
zT{eMQ?Uq1o2`|sWv_xT~2kD!h(t(T?-1#iK5uj4&)ZT1BM>Lm%VGQwTb#LQ_gTutt
zT(MdnO4BG=*RJ*JwasO_S?D(e<zB~tt<`JO$Yk3zZYPA#dR<?;dK11|bsajR3yXk&
ze_7iI*xtGw%liYr5Jfa$rZQC_+kOEA`ptQE)|{O|aSfK?9U6e$&!8Oafyxu=!$+$|
zjlg5Ncj35Ez1k2o7`XDBE6V}@OFz#<p=!8d3>BYpd_-+fWLc^V^uUD5Er3}4@4oK^
znJTTtK0m-6Q)#cHSy+eTny-jnk$eE$f2)V`O(%XB_e67|&ucbb=B_?A!cPB&=fiZw
zNLa`8UW+wLjBfM|%*XguTnN}6XHU3?3ym5YT4H?;M{_phS3wxx^Ai8ATs&9kszHh5
zdaNAN6#H{?2P*=t0agpf2SE?QT`N%N7W}$`&ew6gE3~EdID+QwC=ZeZZjLZQf9x&9
zkwhCsv%(rt;Q`YTs}NsPiK|>M>y^=1<LYVjq7B|9gBt;xY9D%ZU$7V4j`$&{eOd;8
zwr=VI>{Z>M$Nil4??CT|rBk+mg{+~lyEk6qr+#*ChS50`)(}FPJDgq(HoMR=m$Iah
zYue4<IYr){xp6DuX#hOXkj1o8e}`u^G>oo6?1)^f1KwK{BOw1&AAIX%JPtK=?v$ht
zseSMHmXK@1dMc$wz{*tXR6qt^p1{Yuc6`5Tn1W$#ShSrWoVi!v9q)8Ea<7@kR!@>G
zWxa{8n+{U2(Gr7gu(N-9NOdDkv|8AxW~x(LORMvGTS|L3=%vP|bSFGafB!2D&-=fg
z>vhqn(sQBHR^rKw_bboD(70!dt^Uk|VkwrdHb~Sq%F8mlS`U*cm91*hD=$ULs!#Y7
zC0A1s6+wX6syJAfSbzAS1hnlMSEXKg>=;FS_qwWy)kds`dRaKjDO;;O!acs`jw|aJ
zFDw3<Y8hrSb~RkFhjme%e;fr+!#LU9<<Pa9u%<4_V>D&Z^2^;prLAzm&)q>2GK&3A
zGzXP6A_V<Hd+?dZX6KgSyb8}`mbt}dBhERYoSvzrXSI{}OffP+hxv~Lw*1cD=??{<
z3(T1QWKjNK(C_E#Sdisl@_GZrT9kgF%#CMZwF4oQS%~p*_zBX9f6`X(i5f-I=)nz3
z_K@Ra`+8_xXwL}|{lUgHt@26Z8mVP$n*LRi4e79v730$7(h(!8W-~IlWn^`;CvgrK
z8TAY&N_!7f2P^P3&p6asEPrWOS$m_AY@`-OF0?cl_K}uAv|>!gds=H|46&D#;5a1U
z=M-y!Q&6HeqZe{SXz^Gy)2rj3kDN~;fzy*yu>TBKOL@j8zt|(iBPab?tM-IZE3P;4
z7oR?c_Qa57Qgi5d`Z9{xGCKmhA3fBxkM-=-Qcq-->$T@*TDAEg-uNu{7LQ;5xZV~<
zHFwkL6Vs;u0|(7yrk7!iKNgpw4mT45FgBO*W+f<>w2VJ7e~&|oA}Q<Pw?PlzzMMZ^
zzPVZ)yb1)~3?IJ^zz;P41<iww1`nW|h2i7(;m0?vjpNUmKd7~}{PFyM;BPv;S_nJ?
zcn0t&!0FW^@QCFR%X19zx8{%G|9u&|mi3O;2aT90Fm14fQaJte@!g9q0&5jr`Zs;a
zU+*6;f4oTXWhr`P4Glm4{*oS`1!aR7f>ACGfBrK3`trRbI<eBcBL5*h{Q27;Mt&&x
zMa}qW=ZG5oKr^T~I2V;ok1~GvRcm5>k@dz4e^^%oLPzJB1K0r)uaR*x!i>MU23X~a
zJlB%sY8aQujXxcKkd;|B(Pqn{&%Ban5Nk+<x4y1ZL9fvIC6($Q$uH0ppTpuS*E}}*
zQSqPA2bq(N#V;xSJ|)*1e-=_(0$8Wehld3Sd2Y|Ra?JwlFgBzJ;vVBQ!f6-PU9&|Q
zP?5Wb@CaX6CstiaTjjKwoZ*sPXLZlq3MS+lV<(jeQsD4^Z`o0Z@|HkGoRTA5POk~Z
z^t(7wot>VZi(AvljI%<0nrQT+)VOF3SR@>k-1NkBd=_iQQLGv?qoJVyl5XAnRJBZ2
z@zc{zt1*~1wyAhx<Vb{*r7Qo-xH==@8`Qomr5jORa2d7n4Prj8A=9j>T8Z_OF4u%;
zLt}w&)^vP-QuI%_CK_X4tYfaR<&0e{HoO)-rbw6$i?3_GdHaw+!*MTIc}M{82Ia{6
zPaG3uoC+a6a7^4f0tCew>m|;oSEr#1-UY>~fu4?`W@J2uHnb{%h#0|WClMM)5N94H
zs`byvtQpxm^E~qT2!a`Wo&7-`bHOU`Q7qU$zgFXaEN?v({gfMxGqD7*dfNH|=575t
z@mlID3h9)@6svHKlQV_o$exKJ=XD`SE;%I6a^G^cf&@lX8m2=63zdo`ogYWR2NhWb
zLHXGzgmfUG<g$UnJzGL<Lu6}ennHB-3tXkV8%1QuK+e{{x+jJVL2*B1n9;6_V>^@;
zuGoKnJS0~44H74_eo6yK>^u(khrOkx{C9av%OxdKxIDVdc%<EfyaAJr6A8$BAC8<}
zhtfu6V9H)8mi*~jOm+~?5`qNvzbOO>E}w$aGbQR!c`pbYm(_5JfUZNt#5oJ8*%QQ6
z0N*DsLE<Zj7E^!f0IoaQ1ZEI#MxP_eyp2YG={VPz@nq?3GFMIB!kp0Kc}k1>bfZAj
zb8}7sf#kMIVMAWtwZ!F+pnz@hF{kc((xm-Lv6BX1Gm33VWzZ_b*ir&}HggKyrtV`*
zkNeVjqk_SGtctXKcC#fZvuAtbDt>M|$ub;5*z>Hjs(SWWuWsYsV%8|nL(8>K><pxT
z|HtAycLw&&WS#P^Rbw=Yab{6Z8E4-X^+*EyeVUV;!R{L_5bt53?_GM#?Tt})le==c
zBR4;{ChJS38TY>K6n)yp-+C7xL>@$JMbA-t(qB}FW3q55zhlMEi26ArI-*(Ya++i8
z%9+qP9&fPV-BH(TRzI;Tblc;MSC-FzFd@Na4KIAI1lT)#&NCmhou+qb@UOWWm*#x_
zNTI946@cz=xGw518#V_Wj=v+pQHN_*ALs_|pMhN87|t{VmjjKVU6<ml!v^ehsXhNo
zw!4jsI=NITNaf56bM>|K3w^FWG`iL;H>zFATd7Pn9T&k2^-<F`tMBZ$-bh)0IKfMM
zv!(p}7+{$X7yp9KdKK~1ja-vR<4JjIyAI}!q|l&poIHuuV`8)2gxQpZqKY7>$yAI_
zB=1Xr;53YXg_ZO+kIFuCUYQf7Y0mPj%e1DZL!tX+?m90hOgNvwy^XCEe(E~Odipam
zi}0Aysk&{;4ZHS8sbP$E)gF(3rNyZgYdKl(D4kvr^h%jgBp2Fd>y3>Ix2<8x<brz?
zueDx0CoCKdLYp4%doegJ(yUy`EHvQ0$8uRm9B3s^TK3(SoqCE(`Iilv&ASDI#F-uS
zJ5wADak7OmZ{00TcdX1eCf!kOTPoyQA1y0PiEPx|2!zFrM5Es{3qN;%N07cPS3&ZQ
zz&J(5{Udh-hEFSFpV;pc1e=E5<2|EtdE62;8#m3KmxGg&?`kP-UP!0n+{~?P``*qf
z*91nh<>pqe?6ajr<MC2+E+=oK&E=By5nxRWoD4)MIOvCOP^NILjk4Ny6miv7R<nzf
zm$MIO!evcz#(VifRMvEV2Gf`Rs=AmKBIloE|9_?xf^Qlkn@!9<WC)HZ45{Qx8-~bc
zlYDm!;n{Q%pJIrq9?gb$vH?O=99f0D8#tKAr)}(JAHly`Qt;j1fRgVH{w^rA^eFx}
z3QRbrtl{+9bCD&40=EvbOE+;=&OXA|8$Oy%$9ElM&Y?M#@*UrQ9gz9hLGO-&yz=VJ
zL)xL^l-o%uwp=$~G>FNcF{WUP`^FS}cR*#eINSY?d2TDzEoIx;pYh!L_*Tyqd@q~t
zhk+#yD0A7Mw`Zn8TO?FRlEQ`#P%aI@Zsdl=6_#t(WGq=MfK?7*sW-fT;w3fTZtLYc
zRr6a&m*x$nYp!x;pkA80wfbi46L{I~>ROTayivCPPFiVQ%K9!zlq<4OYs641s!)1q
zWR~x!HS%1lUX!+d(GbLm`|8UAQ>`Z7#M1?nMv|4<&M#`gOYVs*YPv7;&Fk2!O8J9n
z)8g{YHQirfuoagvIxZ6fF*7lj!Q?j+0yHs~vHt@nmk^adqXF@^w3R<%41b$+nS<U?
zu1R^CCYG9`F>Pm9YH8<<bG*jKW<sW$Z9=n)C?wDBmHO!|3oMO2%B`ha7kk`0OvveO
zI{>3`eq036&N4q`JC?Nu`P@tW9m7`(ix*df+4HmKx9YNv7a}Lt_N$9c)!%%f*<d6>
zm@)b|iG|cy;<J%4(*ig;V}BpTmpYJ#XPSGehmu6t@!Dl+^_j_+Gs8u`VVHYY%|XtI
z<eD2c@MBun0$W7y#$GEKrwn8j*e7azucmYN^B)qRFxZ>XOYV=*OUp_flR-omSRhTh
z#x^Tsl5SXqP5st|4fgdGGmFa;zc*%L!zY<G{=Qz<$$9W@z23E+>VI`BB)5bFzwEj)
z`JHD1zNPi?y5-WV8$vvpQi}hGM7hP*yfv0(A$M8H?k*0{+5%pb<7z3Z7|*H3b2?Zy
z@Wyg3)ApF%S9|2Q7)S|w{x%CL%lO1LJR$M=Z>~M%*W$k(RG$>{v#G$|bh<m6W70(s
z+_BRK8ucm!$&p(SGk>>D3&V0}P@s^k=>orRN)w@0XGS$$f6k+fyAIgno826rt`?I<
zlxAy23$Z8)my_!@!rp>mMcD<mhg|Mfzw!NiKSc`ggw6s9EU(<l3ar{K8O33O?|(Rk
zr5#}!ut)o-*dXgXt_)+moWU5cCV}A&;oi!P53reZ(|y@%9e?mAF=S279MYm_ern^?
z);UFXE;%5>3^d=RFXyK%?+F+YJOO!Qb=;S6UuXN|o7Uq#epF|xlvp-)I;h<lJH>P}
zrMuC$mRyx?@o*Ap<o>DpGXOlA6NevbDd%6{o4Pj-%M7$}0iGI``hU1mnDQsAwZlZp
zPbD~(kU=+ZFMmkABJC!zoQkJZ6jPmz`!U}E;#|p0HogPc67dyJ!?;l0F$XlN=3kky
zT>5>L*c^Mk$a_}S&cvE^D{+HIG&?xaVk-5z3oq=t8NcXbS3yQ=H^5Bo6^UZI&BA53
zQtE392$Z^%$7$V$!mWK`P|@0>2XIY=TYp)fE>*TZQD0q@wmh%w!m@G9a5>yi;_<l6
zUge+X;-cuVf28{zgmV{OVy=jmCes=9k^j6(`z;5K*Qd!w_G76?)c>FpYw6$~AbZet
z={|@#R<e<uRJXFM-ayxh=a2=v{tKytB$vV2G!p|cGclJzRTmSNnWaC+f9ZVs$H%`u
zNhFxQ8<pA!<<jPZ6+jD8`6}Ho>dn=~C67jn08Qg)sn1e6pwR*y7U)xNPtVq8${bMT
zTz%^jyuH8&II$6^x8Vif06#8ZZ~C42vspe4X1A|6*5}{tjjU_t{?i*8Q{J{#k91y<
znuyf3wMxrtr6m8(v%0FAe=MI{x;N2_VB##7thYkjE3K7vp1ID}paWSzS<NC3?@pyh
zc|PMfPvdMt4m?V#y(w~3G-HwRMVVLCR@RF6Q%DuaQ^~^DPQP9Y88z|IWPGWk?fLW!
z_V~hD=q=N6rmtE=?wwcHZ?W}?y)3;cd?9Tl^D>`CdVTc|D>HZce^Plx8a(S4szwXN
zIFG!d?xI@iPHLf)nt3)3r8U?|{f*i?v<M3p5qE+llk|=FCW2w13%xH)Fk`gvM!kSy
zixIS5u@k2x_Oz@mJESZ<vw(CH`?;G0t=*Iz8)dg$bZI<J3Z<J9(A?z{=sVL>3>vw$
z(!X$Lq4zEPU8tp~e;L(HhSb)u)yZm43c1-_s_~Q=Lk}le4>+r2GbsM1^b34-&JCLZ
zC5xjX=!jhUGCAIrcn;irW!CH{;VK8u?Z9$VY@AYn@%ZM#IVmpiY>j;BtxN-zzojNk
z6veLBnk-fv9Z?snHi4=dkX|YLpR|72(?<cT&d@xAg2{CXe>Zg<vfetzR<9`m@ltux
zu}F)ByDr~Jw!`f9r??no5ysTRXl+z+xwXEW=0G%A3PS;tS|nR)Gd~tOYp}YJ*0mN<
z0w_>gIMty`^AeJLi%v%Ym4J{-RwV6yQ9xDB2xrVZE78{cf<+jUihWbENeDYqd}&d*
zfF1$jgXArRe{nu~`1VQa?GJ!L>o<Sva7qK%i$3gmJr!VflV6se&C;{8&fxoPNuL~=
zwUv_uA?bPL;^`GCgmL{6bu?PDtg=uPPn)!gFC~)F5_(nnp*Bkfu?Rito6GPNJ7*O4
z>Ko+{wAu=P0fa&q_+Yg51whWxn$cx<Ed(vDV>Base~elrpIR;^4iZ^*_fE6RMy3D{
zw*qt(k`SPi4+FZIHID38uJwp|8+6&TOUZOP$9nA;Zm=9BG&;}p6j%(ljgO2@cu$gV
zMHxhE;{o$Rwn3vLj-sex-;FWX$gq=Jvz;nago_0O7@W6KGwOi!w$yr10&J|s3uZZ1
zX^Iamf2!wJgLt56n}DsIqigF;xxYLu0qF&?4$tgm{b?-E&$VUBM9q&|Mrt!Hth3HR
zQpSFhb^4uk;%0QjOR-UO9mbOkFmj+&Z{&`OU%O4(Ya4N42#I0QeNN-DD$*$OUTfba
z0lI72pHW_D+t9GwFs}UFKJ4Lj=JU=@=?x*-e{ns-_?7tmmPLiSQSDKS#`0*@cygD%
ze_&f`it>#YqOQDg%q!%;%T7vR=ckRr%doLV+b@1_<vi~>$(dli5?NG<1kN*Cgwa(8
zXW9j=?ZF12q~++LTO!)HPe*vWzP&54p2uRybec1>BGu+O_gU+nj@BwWj?$;tKgWi4
zf6NlrN8Bg1H?LTFKugjq&q~aGW0k0@S?|m_1=}_a$rz$*wzt~1MH$ayIlh~01o@}f
z*JMssPm!)7w;H3$o@ritsb%AYvfb->Q9!YECZt^RR-B?RmRAh+qk@oRZHjrP+WBHH
z$Q+_E7LnB8qpD?M_%&Mz+wWgl*~7P4e;(K^;|h1JWb8?j`B}@uy~$38DhTIIL(~Sd
zpkq2Kv|DCJM(uE$auaWZ-%^=skz?ve)WXqzN2^@)(>sdqOa5rTqn#Hn_<7c9qi*j}
zhjYcegE70g&UZ%D-QYB;+Ln9RwwnC8SI0eWU7;(~CrA5rKHGjVtkrhHenOgce-_H6
zuaj<Rrn~#-zIuxeOUso`co~Q2GiTwzbIcswOB@Llhj{#N5y(@I!PY}!5@}`j+Uj^k
zu(k3yS6j%pLt4&G1^pc+j>kn-ZG}!d*;qW-Xt801v{RGYaS7gtl)_H>tfoVlzIpw%
zMWAQDtPLX1lJqjj4(o#W^g7{?f50_VVpG>S*|S;Nk4;8+r@8$!9eBt0Oup6LF5@mN
z*Ak8QGrIEJdKcI><vm-LhATbIu(IcpAS<a9Jh)f7MbTjYX+-^f-@yFYyA|s6yh4b)
z-o#G4KSL6?>=y<KW7MphpN#P~IuYz%<9BVg@mKY*w4$hlPyHzBr?NDxe`n+ziH+j&
zE>C3y!zSY%uFb;js*Zbz0(GYzT4&$cMWd`mx*Jq9zp9)q{R%U1#l*JPLy8rgCp$I$
zY-TO5F&ZG2yKrvq<Qf3ZU_oQ7h<yiv*j2G=-}%N7<PB~+lI*DO2EtA3WN12a<n`w$
zTnZUZ>?_9BYGGqPR&M7vf3!0+tY+8jgyrn!#W(8Qv<_q=nbdc9F|99oZrz<y9M-RC
z?VO*I<9@lTKB~ZvR(rfuVLtR8uu@=k6?PQtc3RR5>4b5;g??*|T)P<cCWVDh%wV$A
z$9Ju~Hl7h|v^$O;i<$d+w-?Epc{Zw+uMvE$Xk*1GwU(XjHkm4Ze{9X{_V}d3#3+<Z
zzb4k!%Ks>(6AzM)ZXk>i+J}j_^VRc>+YLBWRi1p-LH>F_J-z$*^!sq7#QC%tr$h*5
z`}FN8zpx4v71E})NXl-1{B!&K^kJTLx}_rdHJc)H>AiE~Sru3nxVLdp$nQ3)<?6uA
z%!>{gO}fm>?;CV{f4`O9Fo8+YZU5VHv&u&AJ;)fEtx_rb#M7JqeEIyBProSv1OA&w
zfGEIEREpcT`sf@COWJH-@i_{u0`R0nlE&N3=L2*gr44KipSCAhNLZsa>-IIb8WAGs
zkqyoYh!%de1y~c&P-di7TJI}wB&zxng{MYFe#*1p*!&3Ne`!>H%pd+tU18L}wx`eA
z(}yO{WZ+*uXbaM>#0iV4kbzhiDK(iOQb^WU>Y;<rM!9Mn*u*vs-1;0}l+ODIFx<9O
z<CIz$x6}kv(G1i?-*fv7?k7g8Jn-lfQs$N~lz5dbQlUtzSo`l!-*DI$81$Zf*J+-W
zN#*X>GXx|Nf1J;A6;Hc$*{gP^-sEcMlj$eg=w@RMq?K0=Hcbj7eApO?h1KWI>xW0h
z14?qvJs_Srv2?L$fa7(FqicTHTc;o#3qU-l=&%b(n1fcxLWf5&zaHx9^=UpG;ydX@
zg$1+Pvvv6d#Z1@1GM#l^H}8aXza@i6l5Rn?QxEDje^NMHJ(KJzAU`eCRHN9{LgrIL
z-B1p{F~Uz>^pMdn)6EvtURCvQ6x@LjKN#-^ZF9k}mwfl|sey^h05L9X>+N`<mpz%q
zokleal5+J>(otboHvUD$*#k%JPP6di?v&M!|6Tt7nF|-sl%bzlitcb>akT{#I)G8z
zAsXETe|5xn-#qMY%p%Cs27*%f^FbbV-R(S%!Jp5UkSCfs_6sNFO69y%CQs8w?Yprb
zP$sOke-if764=u77K(4m%4L$v0p`0I-jN{{l!*Nj;}G7}aBXkDapT2oI~GIuVS|gy
z(P<YEcmQ2oMVs&R_Of?i2S?v?6m)nDz1(^}e{zgGAM}u<9RMQRFP&X@>4IOFJuE!v
zJEfR*bk?G0z>5xV*IsV`AP0ba1`J2}oA=~zL>=o&A!8ifrSf*MX-#xhw|*!!4@VI1
zC}y10oTD)dO8oPHesA&iokcH?E9xo<9o5W4zDpb=)x4YF=`O5SyR%{S=&fC^9uiXS
ze<>e0u_<@q5+&zyH>IsWH8Gqi;tBJ7YsI(g<2%)=m?H|)>{v`^Ht6q`msN!9q}@E}
zT%~`k(xgFe4x;Wj*_6q5b}mh>%l)XG52J^_qBt_Z>VIhOX;e3ryk4r=|00{#$qJ`u
zcI*9kZN_l9CTM*+{{y#$?^N9wY@CzZe;rju*zFtpgPI%$k2l%onm#{M=Zd5g8xDSA
zMHqpuzA~}CtA8W)$TFJ$wMm^aOT5Iq2dB6f^HSdZn@J4WJWSB^=NgJ1g#mcNmJ*Km
zL9P7m(&_EvCo2@Z8p--Gv6U0>(InQga4P=^%Vbae3jTc6VihEV7rH}c2AS|Kf5+q{
z{60<C#k*f2NG+|9;3Fepj|uW2p=qatv3^>h-pQpqQtr5_3?2eBPBW~%YUNVL_N^iQ
zg*YpQHtZ@*e@s;Nx0?T@*Kpa_@*X`T)~f}_2dWT&y>c^`>zI3ZhH5OB+RRbE5R_PP
zr5KoDf5_0ed&~l;cNll!^JPP{6n<|7$a^@OV>w`o;%tlke{>enf0tp5KNSHumoe%m
z6PE<DKf!<V>Cd;{4l*&E?}bZE!ddK$MZvrzF3*xCZ+*Esn3WMpptIH6l6Y-o9*EMN
zc}C`uY`XIR1>RozOZ$EtO+SZJn)7-J;;ZoZ+7Cx^K{>uxl97?oB_k+H!T{9_!prFi
zmF91FPIQ?*hNVq5LuB?G7|SD(y~zFv%rV}TIwpUSAiNP~+VI5poJOu)nqkV*u%2ID
zXS4cXVx-2T<b?EQuSq9w_a@Q9`q>V}RAY0l&W$G3lq~+t*QI|wP6?zg+e|uTa4;!N
zyy&Fh*chNy`AzZ4OxLdvxwpkNf()=?Qkl%2Oc)#^n@=!F?K&wm8E5!zyj0y!Hq-o*
zZ!3SsQ(BPmozYrY=Zf)6k|awP`LEo8+*%eh;cg4Xka6hT>x?MtH)-aQR)eUw-j)me
zChFw8@JN|@$Z^>ML66y?!f$oS!8Wp14!2|%;O80!F7O(i4%4gA04cu0)UCqgrF$br
zl5}46RV|(H5-Rzguf-F!JpqN3O_#v|i(`MVcLy)v+t=#X>)YYqLv3{;C5)r)uOEl-
z0D%L=`c5g~0lS}nci#`c*Ho)*L5}yeXftUH(bUI`Aa)kMm<HVI5mO}=4bHQ*xVL7B
zSvKoSS`T|_ABA>o?qn|0+iR2qNR{}-GCuK}vD2KMq7y~ODQgtES@O}A&`96pAIyK7
zM#&*Vo*8E@J8V4eICA2$;oy)rrBLLYVpSqvPLu9)Ni({VN3`bL%9*FqZ;UaUE|u+f
z%jQ*{nYSoP6bhOZQz$fc=h@C+v3uS{eBO-Lvkp38E!9+_R5|P5Xgh0i_xk=MwNR!Z
zj@cPkSyPIWeo#a8c2aa^>7Bv0n67_-w;Fy_tuC*@&CF!;Ta`j<5LW723f8ZKWVnvw
zv{iz^UQKZbVlx*q)N3RD_IW!t%?52VwoCZL9%lkskJ5r2rG@8*1s(Dxf&8d;?#Bfg
zX(KGDdW8=JBIu0pL-Df#h}6&kjS&n!ei1ZyN5)&m1So*v`BbnGCQ$xFItPCn3{}xM
zu=hd)R9<Mbp9B$*D-omcecJ~c{9FU*-e%XN1C-tH5#4n%@@;S`6>4!bHDi5Ropr;I
zkb$ME-q166KBP$_;n)f$FZXk5K`80&;naFz!Pam55?rP|r!w+ITly6H5;?LqY$1Rl
z{7p@1s@c^z1H<}I4KFxK5^;a)u%^d<k4od<xqxn#eP9+>T%lWNRWl%rB2<pIJd%W7
zMIfGcT2M80x(@{DkXX>wR6<LOypIil<Pr<;HUeQwMXtv|TjPY;X`baMxrJ^~aYBSf
zsdXLDo6QuV^9UuV(*K&-<}y5q6$=i}65MR!(^<hFU@<02n`!_`x$1v$<M}!nZm41I
zs)r_RtXoh*b3BuUJ(&Pec&lp)(WNtAlS*pT^22dYXDHH}PS08;m)nMV1S4r;qx{}u
zt2pJj&fz_zmRr}E_~F%@HpnnT)hicEFZ7zSr73jGo7l{)(8&L_^z?S&AjMeQH82Qy
z5cE0>+5mergDk1^jnsd64;ldCr0lS!tmHg&HVoLku4m6;mUBTTjTGQm#L|;1c;F7M
z8Hq?^Yf4L8Bb$b+@@jTD@3`#Hn&EQR*A$o9=2>gi`hL`mzobo-1`lj^grL>?@m|82
zHLtjUN|E1hBwMm1mr@|jG;{8-CYKHhYHEwt<N(*r(vPkHBI<v>Th|Bzm93du!qp77
zjBDiHV`I)QUYmxJonM!@PW1KU9fspALm*OPi)k3#MbRk6Pz`TD`zfY!r&i7DJ1F+$
zP|f2)Ksn72p*8YlF{qR`f}MQa*(hgpDtT(~=$Gly=D1DSdUfeX?D1;e#C*-xjjFbq
z9T^IY;($<g`FVdIF*WrCKTlCn%`Un0GBLwKZ{h=5SBjV~##)a%LsJ@sv}#lpD|6g(
z)Vt{nO#$m~Kuw?ADAOCxtnckcr$*USu#Y-C1M3(O`cLYNfuk6J#T9mIZ(oBZ9D#Tb
z^}(3}SUEWBpXMS_fl38Z|84o;Y&bMaGjB>;W>)1@S~Gu@SBam6gECtklqK07d{Cy?
zYrxg|tZ~i;i8qy1p1HI2Vxg&`oxA@Wj5!mkVf`_uU}!aGB{ip*4aZ;@1oe$tgBlAz
zk54%KsxZ_9N7{><5YimqdH`nIj>ZCt_CYVxx95(=qSLG&i-9FU=o((80GvQ$zkQrd
z$6{{zSZwECp2q)Tty%iJlf9V^`cwDOiTgGW(ySRT{T+Hvy?YhEsd<T3J)Uul26e+0
z1t!jPm{PK$=Q1+$H?j{rtOM{2e)T&J>wGv^_y3c#x{XP9vCSTOwTwq6GVMI3s{jH6
zkpBQ>0gRVX!88*DF*7nSm(e0V6qh2TJSUgzx<8r$$Crn@Kkk1W3cu%qU@<+^_%i%3
zHig<f_?PPR`)5lbhe7%kU&^v+$=5cr%5ZzrCqrdci$OZ33^N1H38ogJX-P6emaWqf
zdF6@6!q;KSlRad|FC%S{5*k#3hJ2fIK4_L?LhF^alA9$O=S4yfN_#0%O3=Y3b8+&{
ziMJ}A2F2%eo=ks6Zeq&8R}6(<k{|OGkOMgTcQ^X&>!M(mb1+5_Ox@$aD=urJQQ%3c
zvYOCNjbvRJfi^--&0P$n%!$;peqnFQa&q(RgQqdmWH<3?+sq+z+&{z1IFntyJW2CL
zM;k(7*lOEKSZwColduPu=BC{Q_n9#54OAn$;mQQeXykvwimm!>ER;DT%i(3pW{_ca
zOc5~T0y;J-lV~Y(dOxmDOr|hdNz7c&8xaO4A0mTy=L>0!Vo(rV%VS|sU8S5?J5}AC
zOPA}F2%%HnI9Z9<r~=^SO5}>lgLsDvel6#1*^_{b(!ol~MxWh}xK>jXQNDe)emJx}
zvXjDD#YKNh72>d+X(ZN>7lf90wrm+r2=kX)g?hX}WlkYRL9n>sqQ>T{$BADbOLWla
zeSPjr3f*3I8Lfb%=wQ!Rh9>QZQmiEt60-BQeu0zE%{H7<S~KGO?Z=hgx!~QbUrY@~
zr;Xg%u=Rk>E822S;5ZkvmNd<hvs#+O?siSjDXo8LTC6KHowm+#Pbs@Z(fV|2e5kg$
z?(--)TWWMHC~G>YZ;7I6FG0yKtQ)yZaPsXNvRo=rA$EoU(JyTw7`VvnOe@^<mcC9>
z)VEVCJ=~Sb_ECB0jGWSdK2CesJJNma+N6xI(6OQN)bVdl^O6jw^(frrmY8;K+pUx#
zm6U&c@H~H<-GlCYAZzP*p+n?dENBcojO4h4N?kcm&m5k8a-k<^wUYLVz&5MZlu{CP
zuFiuf`WFhk0%O?#%H!B4o3gEbE{Ufks+^$Wtk7HJwNmB|JcckWA6kWf7*@*lxgR(0
z>84W~MO(`2WFdqxg>vNvlTtr3TlO)4u_}M3y3ef&cDebD?9$c?p<Yj|H!yKeSUUm`
z^b5;10Jc`hoRpO@9|YA|L=Ij*b+K)Fd^l7mI9i43#QQ5y-9(-n04n1Q2XEZMqX%Vl
zr|0I{HO*DOe!T;P!kECY35HPMj1nj0R#6s6C<}8dD2-aX7&XADQwNf>r`68b&Od(;
zLV*ytAn&-Y-D^%ewAArmg_#%(Z1-NW9wBJ)(;x_2F@=Q=emzrFO7T*@5s1M>EG@8p
zIm1XRZvzOeMjq_6-jIgNTU+K~(jQ&vVU}lE*Ui==HbWznL^JK#8D4WNy-?^ZV`tVo
z&aCIkIygdEIkLROgwon{e3*xukoJEe=;THj-KF<<oaZA&<$&bM#mb#kIX@C$%Rqh)
zE`z-ziZ_Mdio=a=-os<Q3y}dIJi8H{+Kj3gw_3j#uJe$qM8P^;PSA^q!aHTLqjn|y
zhSn<Y=sn>#;|+^Tvbu0udFJU}_zk&AvE<(YeiM{e;5YI1M)(Z@azk@J{w9AYtsn~G
z?FS%mg5nAUC)fX23{LWT8H5vWFIiHvWEkI}x=lIg7KkR7tb(nj(y=OL{`O5OE=m)W
z7BB<D#g4O`@(wq{e}d9I2u*>wqz3OmXbN~Gb@Sy86e+_}lEnpoE|8O4{<!GVCPYQy
z{al(mB~w0`Epg!4L5z9tsLFo_8b}XE8?we{BVANUwpLoK1<+Yg`Ebl9)gyq9#C#0+
z#KrXJ(lx;JmADT?E)eza!hNn4X1-SqAA%2;4-05+Dm4iCCJ<06Zppq(dx_gIAK-eC
ztDKL)e7Gi80Gl1+p$BFFk;80qVy5=6erjz${e(Yl5XSHALtp2o0S|wrMrl4%S_WGW
z4fvO9zt3#y4yUcnnd2kkpu=N3MC6K-H*Trd+rCO&caB6u#)`y>3!NdbONTIi#e5YY
z^9pKUgYq07f=ZUECL3FExO4UX<)%%e;<VpUAJLu5r>!{ITi4BYBR_XW*m8>EG=ZH~
zCT(!I3iU(7$-=BD7r$tfEUhmyl^{}0lA1WnBSWRO7(s$wJsQwlm`Y+`T4s8n2PbL3
zpMh2%aAqrTWe&b`>c<5au}Ls=F4?x1$JZU{<PVw+Wk+{HnhE&dRR`RW=a!?3%Xn|+
za^3g|NbJ#E0{jou0by2`VT?Z%0Wz0>k2MvSf<X!=f6ZFCjvO}-&G#$%2Sl+BUch32
z-LXc1GYgv;jtonY0_*<+7ujSV)yJGGIp{gqWU($5-($D!zIC?y{Pg<v^zlU81rr?v
z?B2e1AUCXhf|W+^x&UB2LHG8(`}5ga^m4t*#nuwZ736o6Z+G}W5Tyr550F-X!-v2i
zEhH@@e|^x&x0MU!cc^WrI^N6pXqmGFkquO((BZGQ-=Az%*qY??&-|tTeR_NP_i0cc
z$vB@n=c3^N-QVAz@(l)Od`NN|jAQP8{L_7X`a_Ftld13Gtq9^R>%DU`Z=p9bRp&h!
z3ZZ$6#S#wn602D};pE$aT=}=OJ6HQ1YrC}Bf7fF5Af3gS;5zTnFvKQZ1{P^Wy7-V!
zy{`<*u;Js!w@$Tu`8-PYY$69O8Gg&x%=5#DAQn2WN6<nSX*jK0!C=@AE=n>m_NHg|
zYz-o7oE_3IJ1<Y;!u2LQZ-|nXxMH$3l0eqjECXFHb=D$MKc13ZO^D{$`7G2dJ+?%c
zf6cg=UeO!Cg%du_UFGF&qRvB-+{S5S%8EN)@2VE-xkP}C%m~H<=-&8Jb3FEW?rMhQ
zwp=~<wYI}x(LIhwjK+IjXp#DyEIC@WfRt;@MFGHEM%ZboIbU_cwkX(WT%<=vVJ`7p
z6z08ioh@GX{q-K6m)e>Z^WN!qmU<l*f7a~y-6gRf`GI~{63_Tu^~On6Z&~kBt@x)0
z`d!ca7{4p2dw!P^B~l*g_qoJ#exLWwn8I<twKQMnoG(T95k41G#0wqo_PXqh7_`(i
zuh(r+E)%TZ>sgVBlMN~GfJ|(Q%Np?6#Doo0Vaon#p{7HDZN*^M_U0nK8oCt3Kf$2)
zO3)k%Q~?{Bc&P^%wSZd8IPA|JCAl4`!3t7r&5RZeMD2qqcyoC8fGHJys*`2=h?o;y
zu1GR<Nw!7nm)*rbQ-9nYng`tLMX_RG8EBn$=X6US<K<fvTfDe{u_1bKO#RiEE*3w?
z+D*fjFa`Y0^F?|ggM&_`B|KkBfrjx|V%Yw*g5EaL<8Y@+EycHKF=Fi6tI02g1}E8J
zC(!gb>Q}d%k0+_0JLQm5Dr9oH^|UzT6sL)I{94&PPh**rk$?S5(^!h4)fpCh0c&O1
zLy_>2yi(YqjjSc3UEjkihETm%IpyoHcP*^&-e4FoVp~hsXiZL>6ttu*t?M4Fdf$V5
zDm7ISw~$Zsnwme2TX9b2vasBkg3FpSoHtyfD@sWRx3C^~RaZoo@e9kgBI%qV7)y-5
zJ(6=t=OU@*y?^=iW^D}($m}XFQ!s*uwPqV<GYvng-nT@z;(skEY|Ja0AbH*Uh82j8
z4WvaNK%L86Bd{m#+JSD}5RXCu7W3O<3El^9@xBF)F&aT~3vJdHI3_f_SPkcx{0-42
zvL-CjrX0h<;s|zxW3p(7R(-WcE~IEhOHrDI<V9admw&}x!PPrE=xf*>A(;q%3gyc}
ztvG>4MzUD?^_ZJv50B1Ii}-*DUi*0GeS8Yqrj(u=rLRaT`;A~M7qW%I6oACBMgweu
zs9hc+m6TjiiocwE16r64j-Ka-s~Z64YB)jyKy6%{sw0{Nk3Uo=pNgrwh;ys#-f;ZT
zA1J6<_kWJW`_ox9VB<twiU{1cr?nmTa>d$Ns2XuQJDDu)EA&o6U5T&Io2)a)wD4$K
zP_5!M{a)@k%OD~t0E27ZmCf+Xp$G8eCb`M`MLl#{%`>MX$Pxpguxd~$j0dstD!V_k
zhfFT#GOncT7}-P_Jr{Bzts;<RC7AIYX;p?BUVoUIs$-mY6S|vLVImV!cn4agB%h&G
zs`r(&3L~0G@d30-Njd|@RqqGTDkbp@t<v-VTw0}d-KSNm_f{kWm>ALnXqA$5E|O~A
z{|~KFnw_Cls`nkVN=w?ORjT*xv`Po!ep;m+xKFFJ*^j1GTAN*3rE~b*XqDFLGFr9f
ztACAUEd?v#ZRmIoNq+%zLv*sX=nCex2^%Vt10h77Y4*m2ADLPXWp6*w%clt)%P5V9
zGiRD`J8tLU?fKUJ)1{_IlmI6DMU0Zp7(wlgjF&2-6y74To~-XaiUBenqq1>AK1oEm
z{4mBHM?G$3n;=yaAp9h@2@VXUF@J7yWPkJG3|ZYqP(aL$$<N4}pm=!Juha~jgxjx{
z817|1V?#8;Vf9S+S-#H;G9rV;jP2=Z)!t9%YZi}Ij!z~lyNc@7iiLOL^f=@=_MT$8
zdGF;crH0npIUJ{)m2<wPOqR#vt89`0v%BMXtYnD$MYRm;JEr;Vit^rReQ%U1fIH0p
zv$#ApwLY$h%uVHQ6;Zll|LSW0+Uj_}w7c>u>Tc}|+5bFu$@JW91<np{S%n2EY2Q-$
z{sU1G%9kMqL=*!vGBTIJR4Ege?u8*Jm$b@1n17-DU$4LKYz1sh<kU>BzW4IF`*%0O
zM`t|8fxQ^8F#P>}mp%}9!x9HX6M=>w{|sMufAn0NtnQta6+o3`nY~{q3)y%d24-iR
z$f&HU9E??Ab-_t#gr)C5X$O6`F~0P3yUr)vXFX!kdC~VYB-4G?Vb;~z)m*f5Qxu9p
zaDNSYhNyO?knAc-q2vA`p&$k4)uLY77q9XuHv4-;=b?_%MUkTPzVvfj8rM+gdY345
zR2`A&K1mu!Hbm59Cu-ro$Bb;mn(>9gIX{h?6iGDAPt?QMYu_90)IOGZ>d#5LpN?wh
zxAbVE!7zG20?|O*y3}Kadd5`O-PvI<PJa#AyE6w9BV6%kV|7mX@pky}ZRjXps*gwe
zmXG_uM!&Tc((|OIW@3bH(>4jYH3@DM6+BAutP6}MQVx^;CSxw7T-=+=39x6F;?U@g
zPh}5f=2Es_^5Etz*)OX{5EnmV3(fmD_3-m7wjHW*uj~7&vYCvK+@b7?YG{gGa(_I*
zObJ~3$(R?Exv4m6XvLl*Eb)w^MMI<K6ZudL3e&1`JB)k#UR;<?EaDq+p>ClQt!x*#
zThz`0{A%F#ZkT@ZA{Ugo@|%NrXC#dA1|IcMj$o?_kJW{1zqoaC-vm96;x6dEltbIs
zU6+F-7H&$m@Z9b1J8(N|LfVdmCVyZYajRq7m-Sp7m>@9c3U)nL)JUrD8P^;N%JH<C
zyB-+PyE*rKe9_q`_Z|rp)9FA9{9N&GBq0Q&o_t<I$s^ZG5RGZV3kfYu$MxJ<&j^qO
z_ha?>$MxL3H{ip0?k(M2FNMfJj2ETcZ-Z56G{eRKCmwOAK6fbIL5KD0)_-vC(7h?C
z_1%<;mWa4sy0ykJU0>&iNOvLX;kVp(zD9a3&v*(vZ{lT6f+6CYCDM9{EQW&IA~9)u
zi#2Mh<T2bVaf3CI&*}YK;ym$8QlHjKab$S8C8cIrk9BD|6MO5%J&6yR1u0pT6<FUK
zH5Du{&VBhZW~LW!P8b)iHGkr1ir|wi)7L6cof8nZ$?;Rh*ypGF@dCU^QOnS)NuyoT
zO7d6|!>jq@)#UNg{oYA5!G=yZk4J8IW#nmvz9xcsS+KbG-hEY7oN5&ONtG8QG2p_@
z3H(`GoHN9jxZh!m_f=pJbb`<Z6&Sm$T>MWfF%X@x1Rqt3VGY?3dw=%^NReEAN+kwF
zufE71T#4zWtggg#41bYI3{cRKVZ%yH$8m#7Ovmv`C8n4Eag~_caFt3-H+H)~{=QQB
z{S_H-VmwW`+~F6f#$@bU7h_T=oc(jOQ85OrCb6=(?%DMU(4#_Nd5eS9OamJ6i_00{
zYNi54;z=EZ;_ObsXASO_x1?Em%>4%<hn+K*VT?Z&m%(^DB?B-xHkYh%Co_LhZyUK0
ze$THU(3fk0<C4RfAz2&@<SxxMK&}mv6liPs!EuztsUsP(9OV9ezu`)2`y?Z4NsjLx
zoXoCf$>GfKn{T)qQzUZ1#S~+4oJ^qXVy;c9jxw2)f=fZ$J#~rOuZbNOX_7k0rAf(Y
z%qB^6nZqNKvMbW2NaSkm$Af=Pr2b(%NYamJ|Mm)u2c%cv=%Pv+l*vqsO@ys=slbkc
z6q}31QCyCoj?z*gJ7Q@Da&effAHi1Fri_vtmE7(M2CJbxCK#-8drUA`?H?0!MPi#x
zi)+h+n8hX1CMvm<VjEYJIjfnY?2<0;Lbq`9h{Bt-pEw7{F|-Ka2Dg8;LL`VqY_^;n
z6E$-(Tu;!#;2BYi+{QAox28bk(q_QyEVYSQ9cyn*hC+?~yv0f2wqQhEaBY*}VM(=3
zR-zM0ZG=KjRNDw8A?CGBT-a%d{ar!)`@u?enzV_SUQ@NT2Si1=EzRas?w1D7s<fpA
z_#45FP0UWUUt7UreGY#RT%@_JjRJH_zcyH+wa?WmF2_ECRb5`@034NBa+|HVETx44
zMq#Xe+~BaDe%$DAGc{3Mq2@C}{ucz+EJM1Ml)2qSb~935D#j|T7$w7x$oTZgekO>`
zArvu`me@on^m$AHVrd(XVsNGAW(-1u@HAuK*qWMeC?;2hdoh1Is5WDeIJE@xBViBM
z(r2t<gn}uuB2u-ceg=k^n>E<>F~pVvWF87cy&O#L)d?#^nA8SoV(lM=VvtlG{qRHg
z<j|m_0zzJP|NhT^@k90nQKwRTfRCq_XJ<zT2T*n0PqVYx{MGsK$;2glz;!Q<7mLY!
z>Zrf`+n2@9uNHsDi;0W<ji<Be0+T#VHakqPS)N+J1)*;j=1=0(U)UBm(89&RLHA-l
zJ9#x(9Ck0BJnddjzASn>y4Sy+PcZZQ2@IMpCey`5!?k|Emy?Ux<@{uFvBWyN_sit%
z>G6-VFNXn_ig@HoM;Kv_XB%wAe#XHcK+u&x5TXo!UAcc@O=Iv_QV%@350Bk&(hc|R
z;;|&_9!qIEk0rk54~0MUH*UN4?78#uh1QoZ;SS619#5yU<yd*?kHfx7n)!W|TuEh+
z>Pq4;NO6#cXZuwx%i~ItLDE4Q=HOutZgU(BYj1{juP)y#?CSaH^nb4VaW;QDnYR^&
zqweqBv+jSV!`NTEoSZBU0VKW}g2(JrwNfR%801s-aQRX9*kSlr-Osbvv+jxej%&#%
zK9Yf`0YD|O{`uhGS^%~(q7WuS(Mr)*LNS)$88|E)y+~d&A`)tP#g5dB0*uO6^B$~7
zs$S^I3fANULVlYSSyeH8C<VVN&^e{ujgTN-!<By#QpsCsxXlPA1>gf5z=jkAfvWCq
z1YGKk+|=L$nFA7R0`II9w|UKmB49F_0->0q7@;oZJ=l=5Z)6w$S5$8m*n}O#CHpLO
zJtq(hsyV?xszo`45O%jBQu5#t$OZ8gTp63rCxnADwn<pu9w2W96G|EoHDOSWxCdX<
z=re!mV1-ij3IE?_1sMksqy`)P0gUb=Gkd_V##p^^NzhQA?OIS6b6j@?cng;bKmj>7
zv=HGvKt)Cxq$?{pAU6TPw^?yNxd~Tf!zpamgsuvp=&@m2c2wPzo03J5$(g|^E0Uw^
zAs#>kquo@IqNtE)G(5pG=z4tEuuD?|ic)_;c@Z|WDM}lR*gP+av5Tt+z|&X<h!8d=
zwgPg^3nD?By<|RmbYmUe<6-!d#p!GcgZ}mMnf-nDVX-*B_-ogFygWNSetY`vouAF$
z!=y*MrU;NCM0J&YY{i6uoW_L{fJ5&D<Bjx4{2EQXrm>=l2)D=XAEU~ItO0u;p-O)_
zcnR^|$>%P7Q|mh6(!3qv%5sfx=|GxvAWb@uCLKtV>IPagBClgG1ZfqQs+}oBZ%rC{
zW?zclJ+I%NE<Rko@h7uS-KVq5H>Xp~^LaAAm^?wQBVdsH$Fsd8ykG~SR0d>v$*^3I
zd%(sB?y!L~TuhqDe?ff>+hG9c2KIk?ayB!t)y{yxR*C}|X$5H)Uwyp%bpHL@$>;8u
z<Hd)6o-dvuwdU`T-f&j;=4|$+`wXB9K6V%LlkOk0>DlRYay-A1p!cWe`TOUQ_uyLw
z_p#%WVzdm~Ix6~QL4WUKUB?pi-yp<{1TEM<fthFzCC>mo7nf1#>@W<zPQZWPlE9hc
zCI%9yRDhsR60oLF^d_EZ`AQSFL9K@$1yZr<7*!THDveV`M{NgAV^o8*S~EhG0Q>m6
z52(C`j=S}yWy4`c(e0qJ(lw~eL*JN(zA+DdV;*|bGW4cp=uOK2vt@wUGQez+t-y5!
z7@Dk|8Sp|=#qOd9!3&%2qG5jpx>%rv)IGpO0x1gC$;OJdy0(P1<2)gdWhv{sHAvMY
zV1x;DSqp)#?t!0{;u$fMfKrCkm1LByz$(DAmMuHk@lj@=WwBGqWtbTyHIY1Ym?<L6
zRQ3>NV7%wbE>++s${-uHL=yryWli7aq#$YbTGPq~W>7<1-ev>h0SA8os<P$)o&ziP
zad3mg0yx>xNo$DI8lvJ-Ry0H#8dJ6-a78bR?l_ke<Bs&BfU?Oo`hn@QK1iN@n!TNz
zwLBC0{<-cW%MziK*(pA&w}Zk2M%rFpfpyg?>tK>l`0T`A(C^exS$1Q0F`~n(I=DAl
z`LFq*w{Z;E#!;^v+$euQ`RZYQSwGBIz;sDzlq&WWF^^FN^m?W`MwJj+J7OE7nv$<N
zYFY3O*y``PO$0e!<39ElZX51H{d<BX_o)L{sY6Rx2R>8>K2!%jR9Ae+UM{Yrxck{U
z+cBNJ@g$G{MOsu3>;vgxUn*}n)kaAvg=1v0H^c$m)K2?XtQ3FPThrzK74(BOlm~`K
zpc(dWTiVWL80DuXlpnNXs5*8cWuZ3G72-E4!}ji?f)%HnJx42n!E1~z!8W4%FM&(|
zkSmNPE?(%NV@xQ49QVSf(Z!|q9fHrm9tNXjKu2tBXTuf(@K*sPLhnl%tp)Albx1~+
z0vjSaZi(RX1=)WF8xR6)*SmG#+r+q{r=TmAV+Wb&7#_L|r({Pg;bs}O&pt*RF5FaG
zR?V8cefBa&wS?AXjZh^tUf5%!RHMD}lrgHnn-b*@^9=w+T`#tgw=1@>UKLyWEM{3S
z<4`(d93&c~w2xQSf-bAyfA$8QvcbHdyHz{s*>)e^6_J0?Tie%WD=XNJJUQQL#keEE
zJM`A}ZWeAIdTS#W@qzI^7hply2>5p>WoR!_yw=SD^Gkf_ZZ5`(wI$kZZ3f|~+=TE{
zA7p8>^U3^pF`KVoTsFaYl&^GC*EI%M27q$aF{%Iw)bBDz6-3>KQS%!r(;o<xDXpV2
z#qCg;uDvxWvyTIJIiw|IYr2-L_C~q<F0wTn=Lo-P{fTH4=eVJ^OhcqhW&W0p@{Rnt
zC1w5(Dl*oumtl-Q6ag}qLB=)|mo%CeD1X&k$&PEe4c*sQ_y?XMN}>j!V4&>xDnND#
zY8PY`q<hti^#3nHYPuTkWVc^}EUIkF5=l|y;h<}I`Sl_%e}21tz5VG^n~M`3!KusF
zFBhOMB>e|TgBC9iKv)fzuiq|zf0t5O`n%Deky2~=bM)TRD_*~8@H7Kx2GA(L^?#c)
zKqHz)G|f>+uaf>0y?1%+Gq1O_-b-Vc0?`JYso?b=Uw?U%SzsxG<+uDYe){qC_V2eU
zzS071xL5-MEL{Hi+Z%sj0EKnoqCM@{Tz>!8<(Id=j#Miz)R*Po@4sG#k-udHdLRW#
zYY{sIYAHOZi?u;${wgUDzDwGrlz-^*D-0VvW;$^|FEokXa3D=fbAJtW<Al*}qRz74
z_6}#^>FKCxSfllpUTu&n^IrRgWXL+5jk2678KrW%$b@8INJd5EB$9B3J{#Z3HkvgC
z64Qi2LCfY;ZplPZr_U<tF@IM$gfu^KVq>Q@kfoqk^Lb^<H+d;fiR3X&jejI~>?6;y
z19_E6>1NDCk~}Sbl^kQseaRQ2)%E-Q_+&Q&$BrDXsNb{BAnhcIa9?%-*?JyzI^(Pe
z&W=tOg0RXSINk8N=U~bTM1MgHM58*`bGmI6OSX^<yK*wLFGtEWK}cC0jVJO7ld038
zCv_QGR{upq??zfDvqpI1Mt{oeWbl36fSqP{8CU2@78>|m-vxmm&8xeJx_Df2R6=?)
zQd^On-I|t5rlBxV?M9-*2f2F2mOvutkqbu@yk{W7yz|eF>!#IsD+tvFf?KLmM8Q{-
zh<=jYn9U3A^9SBjezn?)k@Q>^Bzs6zJ&`R{0Zp!mY{KK3p*Z99K!4}edO<6+4g(Vv
z_F(Xwst9JaGXvo*RY?@!YD%SlNU!vBIxG9>DZbnZv(5HgUZOV2!}!p9%H(2Ox--ED
zMMgR3&Y9=*m*Hig`f~jR0hHJbTwb6*B{i9w8!4~pF9%EGHA%Zz=zT51SFRhSK?(DQ
zfx*|az-vt3w}WrVJ%37;MLAj-@o6rP#4HD|HkJaigf!4M)g)7{uIxg2THS~9oF}bk
zl*)oY1LIzrptz+4%0Gn(^UAXGxW|3oK?!ScqOcxi_A))^+(@OQo-vE{5%aUg674J4
z_e|gCK<bw~7i*r_R1sVIpxT4%+fMsfPmi|tVIT-Mp9B=%WPiMSwy4O*a7l-XU$OL|
z(dWI%^V#~0e^1%`ULIWH{E=U16rO^BkqIx8EH{a3zC5osD2#W*>L?r0pjxZoQYyOT
zz5TqaSqLZ;CEf$&cs{=PD4%!qUMzkYKbiV&AFXH>ww`TTnx)n?Kdu@xLL>|8c|-&#
zCDTVrVjEdjEq@vti*}n9y=%V)2*zn>)X(j}UY$^OJ};ByUv`gI1!dQ+RH@h4EXmf@
z>gm?)`B3Wxaf_9+&UgjLA$}(7(u6ras7_ipVro9UJ)HrSc#<mcf!=}6z)RuBHQXpN
zO6UAENAXR*9O(uMZT7T@){-y^SGS)IYZs^qWazlp*nc*O$0ln-<`vd?lWC~;E)bR`
zf-C*BXztmB;x{3~SUt8H>^A|L3>>7HBnStA#R;t@lbK<|l@^T8((0v>UiMLRjrV<7
zJpnmY8v|F3OBdxZkffw;-0fySj81tkX~aHMAQe^$Ch7J8*dQ01Va4!QEBqWrG<Q)D
zo(v=X>VNDae>jxTsm8`zPd%r4q))^G=;_@NeBo-)SVq@^2AYn?X3$6yteraIsWj*y
z=}`j0q=^Fn)Pa&5=_mcQK_l&|B10cETpKiS3L3OFC(77(XxeYv18dVF*L<#ei;3Gl
z6r(l**%yYr?4wGxPbw<I(zG7Z=Hi)c=(J0sZ-4CMziJ=B3H}|GU?bQ4(^3F}63cPb
zLjg-EN)MDDXq=B7RGvs2C8+-P6JrLI<T9*@eIjgdtQj=@v!jL=LBh!O!Kk6Bh*{J0
zE24(flTHq!h7-=hiuLVA4ekrq>!ydJMz^}L(icPyvL~$u4o&0;71hJ3$Prc}N7#)V
zJ%5kw$PqRpM|ef#pu)lEQ4>ey29+6naXF~33moBWwSj&N999;E6nIr5<&yzqwd&AL
zK-*h6%JLBfQb_669R(JvgkuNjc>UzCLWM!#WVPdY5LA>G4s5^XWS2t&?J+=i>60&e
zx>NXXb~_Y7ukrdx!0A_S0u8|O_Ymj=5r2#wxlF;xtG4sub%BZW?g)zFtO<&i(F{eV
zg`&jK>;r`kRBV_@`Um@i7Dcd=4p!kUOLeM0xEt^&HMCYwk{>{vQG&p6lC;IRq}4OY
zaokIXac53*4ZRd4xTu!S>Pole`92B-15_i6BM6a-BFH4zV|giG%Cl>9G)}cr6n}{v
z?4*{Hy%a<Fur+IZ&*a0oUBAY=g8F)1>|T@?Ylxv~l;9vK);<QO@tl8dR?IoA)i`o}
zBrDd6llgJ8FG-3WB{bGg!W|^VE(jBF#q{@*V#W7-{@IRvzq)IsFg})8#E--oT<*Ne
zgCzxJbZE1Tm|H-8UIl-oFlW=n-G4VL<2SDff~k*Zhi%l%TR{#*^bSshyEK0ccg^TZ
z=A9{<%FQ;9Pi1({6ecAdUf~(5g`19vj^@CSqyXo~RXI5=5`Q*VlDSGaaj}QoY?d%d
z^tlAlh0>Dn$osCuXnedb9H>qg<=JR?wpQ!c5-xX%qC}sim*C$UT(Bi!u79IN`ty~J
zqBJ0&o@Dum7xxKy<v!l?0}hlE4-_`i`tJIq^g&9hRO~pTUfJ2%u2`aXT)L2>rc)aB
zd=MWTKbClQpRX!k3?-P-Pv^t7c=CL>M-c*y6prASqKXaAqH();>QJcL1AyE-^vUUC
z0ma+T0AYUal!iL#|ECy?-GA=O)(`B3?~#ThAeA4a5SAkLxTwV=mTbnOVUIOJ#P3!1
zjg<IXogWox_fk20EJbq86Ibz4N#XR;SSc+K+2mmGP!R-yH?Bs?s`0MH%*^*=EaY5_
zBUVrEEEcL;gBZ5e(Ie=_SLHa#aH1uwJe-vLRp_Q2m62Sh3|!Bn8-L%{a94TU?YT;a
z9}ocRH~Al*xS^^6ou0LQE-ylekGhYm?`(7INs9KM<YIyQvUn+|@*An0Y+OZOP|NeX
zwK8j3B1>~h<%h-M2tL*1+P3zG8(cIj0G<t3{pusI<fp1NWlY1Wit-1DW4+K0#?`gY
z{XOZV#7TKCXHN>xmVdx@5eu8cT(zK=dOtaPKfBe+2yN{sJM$pu$nCjOY$_LHLy3qS
z8=PLLr}Mm~50K{74A-_+Qs!@GyT^?$zN<kspZb%SPwWNw=|oQ=0D`7flTKR%s3*AZ
z{YXLF&HFmY^Z4;(aL<<A{pvHG<6vCWUzFBlAC|igxs(Ffc4agCU|>tRKw+nRKVQ6-
z=TmDR&@aJiD!l6VO@fMT@!R}7rJ4P`CO%6e%QnB(dHwILKyxSwf8h640Nl94|L5OZ
zrD-vcguJ9X?Rx4*-FW^HRIiUGf0s4<R_e#s+kXHj7SYz1VT?Z%0x~q0!Bi;}m$u|T
zzyY(jcjZ4l1_MDXQ<t%(B@+ZOGcq-o!4@YIm*ML_!U9S-x25bq+5`ms5oqX_VT?Z%
z0Wz0CRTmVOtnoj^0pXW&@;`BZhY7nH!aahF>oG^ocD2!32^klXY)$KmM#fB5l-Wba
z?31V2d&#O#z4utk#8Ju;zPn0N)&5E?v^!5@6l`Y~^R%P2%=@Bw_AT1&zBeEyv{OAI
zje0C%*E*r4DZO+PmHw=#oGqK|>6$3azQ*Mz+dnliy;ujy6Dmi|7%@(N^9<x{gISBw
zM#Y1p*QU7BvmdSBSVfERIU=UDl*h7<`_bwVbKdsWRFC_S>YWHu^-|eWAD<V_vLOlw
zfiEwI9|O|Sry=E}0~{WM9sWC{W<koDI7s1$yu)YI_#6I0$9Mn!_VpiM{vZLd|8Q41
z7@;i1FuaVn7C~bSQtkJDd=J&4LJB4kksl3o5GrymE`n3Tm*HTLr;S#s>h_&mweSHx
zs)Wf37mU*L098}bAt9F5GT!&Pv1m09k>)h1!FP<3W%(BJMT&3vs_#ozNd0p-d>sy-
zvsJBAp1bfUj1(RI)~t%L$yWw?nVYm1(MP1UGzoNY@ak$~GuzyMrZ8!TJZnga7`>0f
zG}JWV(o#*jGy$q0l&138w{IAsn<!eQ2(`jOX)8G55T2ARJ6d=C{^cc$eS^@a>T`s-
zBOL_+>CR~*ybp|(_yIh-#t(A$(z|3a${EB@b*79|YIwc6IuH!`d8kxMQ@fkz_W4|M
zz$w<6GtAKwi^>dtlJLb%X4s%fWd>LUL^Dl#%H>Vx<QQ^t%Y`eM$MnT(Oy)+$DpeYi
zU8)>^-zyF?!Zu0CV~yy!)Fn-ill<h$u^;J2wdhk9$1oRUTaB4|kK=r3iAX|sBUN<K
zH$|{3VR0^NDz`Do!OG)1yX=*T4)jcg&CR4RT~DQewWYa#cr*LHP=caWOJvR>BDS=I
zx=b_L*qZg3OQddRPC1H-H=A;Mg4SMBPC?<cw7msRl~dTws<5zIp-@S){c%yBTu?m%
zogkb}3k2dqikeNsTEdwZ_e#Q<v&Wcl{+0ft4h8I4kUg|T&gLo~!3a=>Hh%fFPAi#5
z+*L#(rImGmb4;|u6Lw*7GN_xwCOOniS$45Q%KJBKQDV)^GOLi_V9_P4d`?;=Ih7I0
zZp!6wq$<Z#THJJBt*0>4KFas?hRmp$WEO&*pjEw|M#$xo+6@JTMub8;-}lF9xvKbJ
znagf3**S2)FxZ^@@u|B*F#TAtLtqJ`l2iGIM3TUN5xf^3D*g56@b&PyN6r4~Kj)y3
z&RN?Ag+p9|$|;bAsHg4KRW@EVFxy=|wkgqIis09f5LTNDc1bV^m)i0#t&^suS|0>c
zfiw4`&<DQO<2yL8?Os5isCjx<?Nn3NCT}A0n|CW);fp3|u~iACM_ZXBClF>SjG^p^
z+Oz<F_1!Vb)hIzMT0}=T3LnDQ%u2Xkrsy~}qi)9CXzC7d%Txi0oaVj%F&8^`g}Jal
z(XmI|iwS2$Fus>yQn95i$DIvLRKHT&7V#WyTZuJo^TIdIAq=K!Ag@?w8oMa=Ws7v{
z5JZ*LiuzGh4!cW2uLN_8)ShaMlm{i9fB86nTF}Y0Og@daLFP+3*Mx1~xu=SIw1)NP
z6QK@~Pxq>oQFI^E3hq=Bi|py*qg5+zo(>v5Yox{KuQ?r5h*oIXW1ew3sK9*SoL+M4
zbZ}Y1&OCiR{{&EjT_G^~sl4I<P)6)yeUCWNipvS)DT2tSFLhbIoFM9`ku3k8_+goU
zgKUoxCts)pncz-SnqAHau4qhWf}cbgwPZ37eiCKmjS}yB_9){?v#;5FgQ{=RNDU68
z-k3x-W1OatP4_n<klG+27`hRG?9kqbKz5_No<MehTc!d<Q9<31Kz4{{P_j;9hd^p#
zs-tg6AUni!v~4AJ2&9%l@M-(&31o+V^hN}-8|4)QvO{`f0;#+bQL7sg$PV!*6Uct_
zTM)<&^8y0dNp2FzvkKbnNhP+)TKmQPUcoB0NAb2j;v2F`eo2AiCC84NkBKtNR4Vdg
zvS~$_jOJ$_)z3qC1SSzC-IIR*p<b&M4SiaBjskC)P@W4Ea{BQ>x%*j~k#m-R-n2&1
zonL{Ot?`p7@?P?2-Z$A=G<-l7x=B<dSNjLDraY<8MaSz{En(wB+V!j!yLZ&eTa#J}
z%qPM*S+>x(4OU7X6=@E4nUiC7VktG3l*wG0Tj0G~?lrH{PCzd-*(4%h_eeVvxoShd
z4*M%;bylq%VV77_S~#5qFZ#rP9IMn<bWDrqjHIB-_c$(^cu8K^Hl6`H@NF%!v)QNI
z?Z@rU+Dl5w;N?m0<?K&krNmy38?!2<OnO?@jiR#+jawv@Q|wBah>GME_WM}9=yy5C
zR7e*Q457CMtp=O&$O1B~&T7X;B~wp6eq_$K$(#u_*pOZ@=WEy@NfI%CG_O=P&$`cK
zhS&@uaWBwbS1u?T#8c=|XU12(zUMN!&FEA8=wA2Qe;pD9!z{*|qbC=^6RCsiDZfg}
zucWpX7|)6>GXgdW*fhi)BgF0{26=)KG6joklU%7Yn>3KLh;iZYH>82V)b;g>vqjiZ
ztH0~n^L6EBjF;7lb7{F3*^e=7W)#0GTAS!xTGnwTP_h33IW0k#AqGSg12Zx>m%*6>
z6PL&EEh(42`ahQe#J5)aKL-H?!W!;*AGcxrKQ9FYxxOG(mtl-Q6qm5kIuw_UHW?|G
zkQX@xm+ApPm4D^1|MTs)7nuak&myHp;hou8X9UTTq&!OOMtweeSe&ClVZ0kgOAS><
zmd&U%Z<FLbPi-te_f;o)hNKgm-s5k#=jkgSg;`TH^%PQtQ3&|T3U%}gNen|OH$JP_
z5CbH}#>G~!-0PV8=dU~tRpPy@04<e}WaNQ2(&SJxZGU>wgpA45rgvl$GY6mKV2&a6
z!|8n*6BS^R4lk1(thVBr+2-hl)BawYccYhWeok$NWK+9oNC-+e^dsdDF{DBxT+k!$
zYYl16Q=2&?<JmQfMT0(KG^2&N-)P<F)rHD!G0^CKQR@5+wpS!rp~EQa2N}|NVK^G}
zystK-O@D1N^>0=r=5H{q%6C|i$_u#PxZUVQ<Cg4kO79C%m$S^2Rkn?+NM_?lu3JXV
zAug^K-+ke=6SfWOrO<9Z%VtJwK)nz~4Ng4SN~&`-b}}NsQl6%jIet2ADnSSGbTt|-
z`Rj%v{NzoO+iY?mA134C$Uz>qM^R3s6)=uMTYqG-)xTq~8cmVk=CaGwCuxZl3(z_0
z%&kEQufx4=-Hl#$>j&IeS`mo)z-7ygBfOHytxUC~ao0t&s^RTeh^~0ZL;>>hDPu$8
zY}>H2X$uTa;N7<9M{n99CVziX|2gRol%JL>=6<}LJ){7KGihKcc$M(J8XY7GXQ8JC
z`hTeE%9(t|T^9m1*-%+EN~cn+;`JQEUWzW!FHa?B-I9pVD?wrHPA6#xFZ)P~E6fzc
zV-Zt}p>2FYC6yf+eGt6R!bvr-j$~8n=V$bn7Co3LZbr+v3KwpK2WTl)-EEb2@Um5!
zGBaA?m`c?AZl}M91g}7WS_<T~)G_`TD}PZ?u>b?>B`cY$)=ssQprudwl%NP&Z|P|R
zdBZO<!Sa{yO$1h2pba}DWa^ZvIXQ935Ym<?HH+bCO97P;U^@s@=5?HG6X)Y9+-bee
zgf{Od*aUU1lwY8=hOdsyU2h0CtFGZWkc|^LoC`eR${%?v15XuTXOtm=WySF0+kf!(
z`u6hgOyLGmka~B2{e0mM22fZ>cA(%KVE60q-H(^wN2<eiiEQ_KqR&Zd5xaG!tU%&o
zf)Ny;I`g`;+)Jw7f3gy83*B`+>$t3p*Fs@+Bj{mSaacpc%0JOjyO!2LtszA3M#}It
zuG(%FiJY5;OQxZ4qT025=Xez4qJN$!g3=Nhx#-JKW>xdS3-9$zBd^u1vsSEJ)}^Cp
zPSt-Nxt{88(`OJsF<_wnbbY!Q>jo|l=u=5eMT47Z<uz&GepwX-Fdb$ZQK8vENr4Ok
z?_XXguXY65^hh<?Ithky5;4l#wQT7&uW1WH>zYXNww)WSxD-NGlX{L$%zwj|c%j>@
zsMVKdO9J<5+6~NKqFqaX<W}A0)oBAU_bsDywAm~5rAC}Qr%2Nl-lkCP62ww=$UQl8
z$NYX=<}{{O-sd*frG0MW7w1x19!qv*UTH-{-TDTcL4^nLSz~aXT`1PbEm;%wi{*A0
zVokO;7-5wyt4gkLD@DQd7Ju>4F|`~v%JX?1$fHg1+SUD-u791+B#SIa%yd%SWbYRJ
zwLaM;_}f9&D*8=Y*Hkh|=N|xFn&K=rWz7eJF56<!3MnGV!tN^2)lcNTK$k*IX(^1L
zi{?T$#HXV!ZD>-~U5UD`SLTQmw6?-Z-(`X;OYd8dgSP_I$ZP>PRDTG>LB)?F9)Oox
zk-<!86*^U~w@(HwCC@NYd^QY9;C0}l+Y`2>eUKHmN(LB|@Y<`&k}nh9&>Ydjd_($n
zO(kgD6I|O|O}MY`tj@iabqfaIIUtUfav0e@h~NzA*^hG}@ufU3tDGRJtB2~((0Nbs
zlG6L$e88F4S4j#f@PEID1(X-onwb<28Uej921lwJR^V+G=hPFeFi^Z;g`szhb|wY<
z1Z#9IeV*a>K*3-p<HN2G!YDUV-O{=T;XyxB>pjIwTJL)=ft?OmJ}cske4i6?2C7SX
zPZ^QV20WT~C@&QT;RC))dCAALV5+xLK}U|fP&_IqC8f%5xqsluCl#r6Uht#>VY!q^
z{93So9(pphLnoPPCCVjEYPq5mO_JV%npe(TXw_LNIR-uQ-1*ozoe?Ksna_Q#RI!}h
zKBMJr7OzIi<b7EWhdlGH_}`M&XwA4~Ti8vZ7`Ahe87Q2l%5V~Fsx7C3tfKhp9vlEp
zn_{9g0`$x|uYcc=-<$&y;N<<Qxnh^<aRqFKQaLnL^41K%P%vN~{pbrO2jeLU%vhe9
zlf)HvG$l{?ei#WbL*}cj9a2`ea|-WPazmg9tHT2&H#jS#$N#D1jzgK;*PaaKYeMM`
zO&j(&6lRjg?XzLB16OSQ9Ghxfg+1yo%^g8IM%Md_bAMWUEYVn;D<uM$r`P6SxR~<o
z3cFpeS*a<KMKkVeFuH4lUMVSnWF7>3a%Gy)`oQIp4SJ;T8AgK3=N%>D-v=LnK(BV#
z-d=96$6-0p-ni^!`@?YTt{YwUhW6so+#d=Z*X(86=KdG<5My=Uv`E<Olc#_ZFZ<Xu
z;xxOsv3~_@8t{=vN4?qpRC_zAlbwTeqb*iW{8lqN!B<N_cc22Wm=%MjeB%}mC8TL5
zJKH?p&SR8?_`Y0EUM9^j^hrSK>qVr(7q_GvNnMdk6&T8ddKlR47GAhArfs}%KIa^7
zbw&MD2OAD<q;B9;(q-OGubOq~Dsk0!sC1?Ucz@xh5QQeK#yJiiqhHgyliR7vB2_bC
zNnsam$1~<|;TkDU7o{-L6DSyo#ls^A`S9pe14q>z_L@8tyj1jGe%V^@<YS?jXXXA=
zWRmaoMtOY?IgDFeNnBsnx_&jOoo4~`)E&kgn*Sxx9n;`&$kQI*<lBi@k9-cP_0)#K
z=}4Yj3`eP==9cKQuSKqSaJU{od%x{uIyc}dEQRB;H+EV+#>%ZZP~<{^p9`9+GR=ba
zJiyD-yP{xgV!*o|1dmSwaQ^|h<f50crX>>tF*7rlfnpgGmtG4%E0>QhGYEgsA6SMQ
z4qw2+K<teJ^wwZ+MNdVOQx47lKXmw##L?UKrp>`?OCpE!o^M7gyKkNBK0lq`o<3YP
zc20Ps9CUAAJ4Fv9{Z*1~TGKhDgw?8h``-QWA|+V*oauv7YE2(Uzde1M)3a8dRw!Db
z=&n+yXNQXJXu6|m4H){C^Z|eR?aIAt*>7pTmxz%9(N?-pIiLP~`|U{<hCf`rTI1F_
z6vR{IHlBt=%@8F#=aY@j*Knb4Rh|(ChPcC!*Q3q%6g}5*4dk_WJw<`AY>i72d6Sn$
z()HRD5(tI-xg2fEwJmEdiDW(@$+~0_t+cS<nT#K=QFF>g+k}(Rk(Pfi?AapSWspGS
zGo%PMsx*FmW>=ryp8k326$@Gz8#=44aL#pqeSeA%P#Ixe=m11e*!}ps`}*`dr&?^r
zzmwA?GfHcXY0L;#7(*hW7Sdof<868!gqE}hqarZtI{(aiKsveq^W$5`?tJ;&s{%;o
zz*$qT@zeNdzaEe%Om=@npNx!Q@jIjlkk`|*m1EaRBKgd#yd3A%1>uxC!malL!15ii
zOT}|O)ko|?eyNTnt?)kd+R>+!16p&JxLAmNbR@m!l%|?ojV|Sg9~~B{xF~ClwBCEb
zcn?wg^zqv0Y?aB~FDscVc%i-K=N{4fhNb~2R>`95ggz6Ha<qRZPPMV#(zgu~vK+Hd
z)?_I*CGkZtwhv!h_^nxXuxXzgj^-mHjetoya!NR=RjcV}NBS|=3XSS<=_koL_$!Iv
zOlOQC%R}#JZd9v|XD2GV!UWB!u0*`DK`1gP+p&<JW+gz%aG-k$8hzX@K}&Uy1nmyk
zYnKJ7v=&HMVm*J49Fq)%N5?78NETy^wo93^NT%A{NzoVNtoKp|qe$haM5LDZzsw=2
zDHU&p0#0N}RaF0tv~z)iegv;zgqbSbIv-bKzyyHUsSj2oNG%k>D|X5N7k8;Kg9S!|
zYq6#qg4hOo(0W6;C$ublv|3APEp&7c8Wi90?&Tg`n{0n>Dfa|awC0o!mwO{rSR_)a
zJtdXLR(p`tO?ku5eQ34MwU?a`5hh}&s!oA8$XCL0xN}JgELTMvYeu0aiB6sBb6OD=
zyBkV4%1<doZtcht^_R^ZUF0-YsDbxXyTUrRl;d=;>)e)>PPNrAT10=^lj?G(=mxSV
zR~3LEq#}R8m`ZK54kIT?<j1XDH0DQ^SxU57%HrS}AQnq{V)~Sz?8rO-FU!T!*a7?w
z4wPH|Mj7#27CNKWI`8*=te+t1$5*OVqI6#Ww4UPn;9NJXq}JFdF82(yT6DwM(OOQR
zVC9K=30Y$pIy-9RHPmK4N|po8R-MruQ=}er?*@MdO0uRs6UaH(xzMP#Ot#ZiZ#itW
zkuRf`C0~_!g&8Qs=a+JpFBZyca|n&EqGex;MXIH2FONy^Vzl0b(NN`DXw)<_mBkIg
zJmO;PMduoZ{nL*%O~t1@Cs?27laxMJxko-#>Uh<4qlbAz>)agHID$=2RN7E98@_X{
zz0ZGDdIHoIE0?E4sVt}bj_q9PuH=z+j*sPZpl;3U|7a|Sc9h0Oj#=|{O*?H*mq$e#
zMN3k#m5RytlSXpdcp<6l`8gvwO+5^R!I5v$X<jpo<5C!^mm9}%1f|e&##RS%QYey+
z?bsN|-PhEDwFHdx5iY6zqXoWWcT;=01D$_s@YMU5@vwj1$bIFnncF{ZAm)IIs-Ez2
z93l^P2LYB7Jb+Ds+FfgLLSbn(Sd;^=1#VpnPQ}%d0p1D1f2(qDWFvyL&T|S%Ep&iX
zp&SWC{jP_a-Otj@uZ~d1JGT0fh;oVu#}#{qUIV*Zq*(Dt!JCh3-vZTMoM>ThY{GxS
z3}hoZwT&rwqWT3`wqfq*W!{^e=5ALX0Q$d{SsevTFkD)_L)u$Sik7UQCi=b6rcQ}~
zxYwr5>YUiCl$Qdu3T*G&f=AeMmrY0GrN`J+(=8@({|@?6pyY-IZVp=|0K!Dy8#aCi
z01xPd)$jT~MmZ!T>^T^<QaIx2-WPu|0EVEhdJOoSon(VTI=&qR)vn`OY8*Bbio-Kn
zuTn3>PFkDM1~PRF?{)`l^0qjyPqS_L=odDO>W&vSY{%aEsXon0r2<e@9s;c5`AhKz
z7<%n--<`Z8j3y_2S9;^W1~5Jz{a&K{I{`Z-{!@)9QsQ;ry9rcjxWy<fyq$lDeyfHO
zbyt58OOS;BMSH<mxx2e)>D|y=Faqz=TdWDatF>qeER01JkkffTesi`lkXM>~3=QA;
zfJiP~JjnP9>a@=k<@pWEAmum9xmtXqmcbb1H>!DICm3s2Kd7Iov(e-n*B)G%3fufv
z4fV}*zi~l$Jw2<JL(ZqA6V87tF6$BZgOG9%qtL#Ia7=&AI}$hYmPWd1ps~bL@67sM
zPvxMkHUPb+?m+vUF4Qb$oV#l2JONksh4EG=;hB4mDjMn28-XJ2XdZj+cdxcf#cRwV
zADx*6*LNLhAKqipV>Vm0m!rB9+)jHnix0G4do*@ObA2gqsFt+V=hr2oHMtq{NR@&K
zd(Ll9@9V(wWd2+R$(M*p`nRimvQ5p(zo8Tum%*e369h3cGcuQPjTjV{Q7kJbmw6LF
zodR;$mlza4+JAnb`7JAJ7X+v<SVCouz0dO7Sr-&VC#|ow@ZMi){YCF2Ve(zfPnjiQ
ziE!P@yT@e7V{4(G+R`dW;RbD`{B=(LTdj;i#3a)x{6-^#6TzEHjlAeBF*HIRazer!
ztp10E@kIuX{qAbyfY8fd3eF+wnZBRV$IW3)mX?0d{(otUe|Q}#8X6^ngz5A5hd(Bo
z_GL<QlZtfkIQ>1vW{CQ*);fA>0}!IR9^U=_^~=9L{{n&r{1r#Y$&d;m`04xniYNkr
znE-}BdB48pS5H3&hZE40+LEV90hR>hVf_AidZ1Z^(Mr`$-(sT{&Ql4gBLRi=hRRqW
z-ZbUu6FF|=xxRn3jm%J;AW2hrh;NZf9vfe2JdcX6@zcI@R~q%d>EX-t@UdtzX!=J_
z>!e*%BuxsAAW)-BZqi;z1vV>#I#@E{YU4u8+@?rcYe)q^HV@8)sZ3QQ<(5WTZX!}l
zft&c*w{J+_2|DXzc1ILQOj~kl9eok9N@um!@1MWtVPAh~&`a~2$vlFlaz~UN(Nn#X
zCeli%lT5mb&gSmin}hgyr}3!<ZVFHi)aMjRJTnJs8(JMvV&UV|u$V-Btg?Q*R6S%Q
zM7%&feXefQf`$r(Eo#{S<1B%-WY7Y}wcCLWDa}w)mG&v))D~-QpkN{d#`gq&1PV@L
zG2_fL2$+Ac5r#ob8H<rjax(sk1xY&Qm=RhmG%}TzedZ_R;zfNy@0r`uHx(sYQ92g^
zDp_|!ncfr1)Pw1|vQ)hJkTra0xQcD>i}ucyB7;t7z`?f94fXv=?=a2^?QKs1-jC>5
zJvaDCH<Zsfa(4SnIgKtrgpl7F+%W7zY{3Cugxr5(ve)?~g$`z|Wr!M$D;c7O`<x+u
zOMj)Hd33Bn8Y$_<IufLs9cA^GPi-E^RsXRm`za4>ge`*vWvF1lVj0JwirJ<8I!esT
zdXz>uP^~+%+G6lT*GwjM6%n<txGZfgtCq$<T^td9Oj>8J_BD3{)VyI_(}%rXHEl}}
zwwZr70-<f=I-#DA#iLZ%q2}xe6^6CzIra__Y}8qNymW8mF@Fqz%~&T8lI7wL3Byn&
zS7c9uI)DB;eR=rUQ_V){pVjixTJp4o<t;g%r_hXm$hG+LIz^J9_xXFfJ>h42apWYj
zSsx~rGX9g?dePFjGvO6#2|bfBZ*1&hC>?**G>D}J_Ilptsxa3YW3T7bYY`k=!)Z{o
zyc!s_?S@D#c8T#OZq%9`4p+0S31%Z+!gkCx5rN25u4bx>mQ=EWT=O+;gOE6wvF3qp
zrZv~zEC3ue+ARPc?$XvDFTZo@o{IpK6>82-bv>H43T5x-ypgrj@~JHn_G!hf>e_!c
zftvC*%w;aJb?Lp%(W$blDSQ+Gi%vJ$y*F2$n#z*IVbr0lt~cj1kg8rk@@(*(=7&Pd
z5s>bv0`sx~{Mj!cZi|_IGmf`s62z!G7dItr<$(%vpRZGbRrYkkz92#C;DjRd)gx*~
zE61yY9#9GCO;4E+_G*ETDI(~;w;q4^m<vHpLW30EQj#e6j5k59;aVdp$@P@BxHe=P
zhkfsgl5fx&VOSk{id$Q4thH|bt2c|HR&S`Ak^*s%9?W_r%2l@Oy=6&<LIq!wNu_(0
zHRIU?^wc)3NfEAxsZCZCFCZ_hozYE~VKifvW5NJe>zb0DQLblhtLmG$BQ<}>eT$7v
z5JB64l3Smb79tS(&NqZ1Ema8`<6sNuYWNitts*2b@CNI<(u$Ul0S8}@2L^EJZDF2a
zS<eh?U^purRjuBOkIPdY?V23~xXIc&W<NE|l;BESYv@(}G{sE>Tx-hOh;D0AmTRLM
zZ;$RJwKLJ(P5pC4_ma7D(Y=4{{g&wND4mY(W#_j<cV~+`M0cmaGtu3NeedY*m^l;O
zofrSW=<fKsFuGTiwGrK@h4xqC@~Z-$R(c>+H_=wyJ0~na-}#cI`hE=vZ;Nfn>I&`w
z0y(iQDQ<y42BFK|mu<~%?W?SAT+DLZzCu^DwgRUIP1(FfF7sKdstbRPvMyg=zLdD{
zD1IOca@$h0_B2-gBfC+{9iNzznk^Jk7Xf%V|H*swP0Ob-_bxe?wqz^pcOmL^mwEDy
zTv&9X;*`30Pvh_H<$}$aSPE&jbuWWJWB!Tku$<{!dgyxJKR*yO66o=c`JtnDCO>q&
zf8zYmYwGR!p{IB{KPZ1sKB4Se^TWy}3OuzC<1P7sRkdVQmT*6v5Y)I%-`MNF;h5;$
zF_1!_P}5f&1EuY^Kp9?1yLQ`Cskqlv1EJjIq>tY{M!57eU)(I7e(XHI?#!jh$X$Z7
zdTlZwU$S+tMe%uU)^TVEKA+w^WP+aTZ0n|SyM=4~WsK*Z?ahD4*8Q-g^3&&>%#j6p
z`c;mcyp*HV^8${zr>^Ios2WYWo@P2}oyYV$UF`9P(u23h=aJ#`X<9o62X$$2&0(Wi
zE)TJ}B-NkaHZ;rKgTkF`W{N50dAK36dD=~>O_7ayTlE*mRtl8L7Be5tdIVBRw6rqq
zX5kBBAeRZ0{U?8}E1`mk?iyLt+_{Z)rlgFZ7%tMC^}8*p?{+hya3(FmjJt7m?f}Zq
zz^|LNz09~7Ge*^Gi}a?s`1c1Bvb*JK`uJd4C8lOtcf8vSigc?7-CpLy=Wn-qA%h!q
z*siQy(U%9k=c@vTPBX`xVUqlu;gT(MbGyj(`Y%^hXT5)2l8YMK<y<sdz&5$guEt=X
z1p}%jp3j$wpud8$ykt{O;qsZ_GNo{9{O=Tr)#nBu`nu%F=ZQ?2e}xk#mF-o2*QlIL
zfMa;g_j28E4%+fuxmn0(OKU!fO}y^g(@^9~A8t5JEz-94g6kb*x4hKWjONl&lftU`
zoE6Hb#epMzQ&UQ8`jy*9V}>bxTCuqr@eMLIW@4&4malh@{tp8RV`P_Mj6W2YVCp(3
z0XUZmb|^7_&01S;T(=Q^_pcz(m-N9r9L|Mg5g2gdIt`$-hV24v4L!*6Hg+m&1!-kC
z|GwYwtj?}1C0T1_?*g_odo+if8P0rjSrX^mLJGv0B@lHgoB!-o39S7SQ_$j6Dimzw
z1CJ?pz8b7`KFOhu?@9|z@&o77(3<o&z(ZKnN&l&T)X5r1i*~XFDy{YRCFGR*@}eMZ
zT-8FR)On?Vl-kJ(vXO5l6mr*vvtk9-z8hY~tDTe-vZlU;f-Ky(fF!w%yqW`3?UF%v
zmT@=yjfKb8D#+SZ(*^_O&S1%H)y08P*r>072V6#RiUh9GNdrTz-C!m`mUcG{!XvL)
z5Y@haSV=)jG2acQmr8xNl0hyRI7mx`-BSUD9dH)@td6Repec93D+eX(@8fTpN1NlP
zvUFaYuu4sRFEEtU<tjC3%Olxv_|nOmg8AM~!X>12eX9nu>^dYIxtZHa6WpVu@$Uw*
z8-ZXuE45%Vg6*vGxGbEij>|R<kf2Kr=PE>hqY@RQjf=HVxO5}56pFU44+JTd#uYTw
z#Qu~ZQ%yx4lmfvkQRyO7kk<EYaNL~yEyl^ghpm%XK)4@NE1@cno8dLHQLq+jt-b@p
zSU85-NeT_&@xaL(8k_W|jEHEZlM--7)Ja4~Suz*`rFPQgPei^mwN7j5&&BE5I*liP
zDfehY2!du+-_j2|iH$;GyQDI13fYpC{;r!f)$XJ0yQC)LzQs|X+mu}was`|QKWRT$
zc2|HK^)>woB9<O}`|a@Z*R$F1?C9s&@csOBF*{vc^pH-#m*>Ot*~R?w{CIY;#0!7+
z)9lT=qwnS)Upb3W4yENotZ<HZU2zzH$k`sBp3awRi9dMs=)tQIrto12Q=kqf8Tt!h
z_-=mwW_I2ePlv<bh98IT@c}9yLYm{nE6O=4Tn%s*dst|W&7q7en%Uul;l<_ag)eyU
z?(~1d<7xTh#k=|G@M8Gq^B?{HH*Xh<vx~nB!_V&)Z!cfR<N5pHm&^BO58upxJ`6t{
zE#Cg)Z1E!ua{k|=<5~RWWO#ite?5Gln6dA0aeh4feSUiK?sRr^KFrT%=SPeA`33m?
zbbw=dcmOl}87kx79reQ*#^Ug~)Q`ACK+sO~dq00OI~jE=`tTr=u5^3SPqp<k|JT3&
zgD?*zQ#idmIXS$Y1D+ZyPR>$)jiqeFO1VV_Cazcm-1TIB{^AVM2kt%vA6!76(~!rH
zfBm@l;l<);F$*+4cnSx{CQlu$Q|(?kJoTKBg~ty7By!;E9~`J@6z0*R;o158_{D7T
zYIyeK>F{#)adEhIwm0Hux!|?<5xQIaNLz~^;NmiVc#0%06ICaw-4b1Y*IegzSETX7
z=!(WMA!i}YNA+wi8U+EYRU>OX93BTO|6=%I{&GG%3Eu$KP#u6nDi?trnc*jQ4*?)x
z6wg4*5K!gVQr?3Q7m<etqOx6xi%@J{fi3kDsS&AWAM~>d{Yu_WKe%o*Ug?K~6ai!V
zpkK?-uazw!8gapE=ZE@#GuFL0KNOV@4^TCJ0@IS$kkx>!b}Lq2V;ccj#bQFGvjM9}
zik3E76)+VY7B^s}A`pPSW(N3d&{w?)`pTPtezs3R-|`yhOS*z}T~RbqdAmYN1t?Tj
zg_IFcsP2OoGf-BAw~JCu8jG&<Q_2zD;6CVQ1pP|d5@pq*(n>#n_N+~Np<e?ES+Xrr
zwprvg{c_5&QriRlOk6)LpP;PD9~HxE98)oj+Q!heZNREYPz-OxirzB9j+?Mj0}N!|
zYk;+O6JV{}4PeW9S899(rX~|LRoZ4-0W2aMo!nX(TfCFohcdP>9Jp*<#sW~iEttw!
zs{VTq^s8Q`SKTsyfgoC9YO78>TZ-CvAN2Dwy_UM2ew8E1$~|j~CcE)Y-)=cWsqjo{
zpUxSqZdAZtV_SlzwWwyBunIvq&YQ5Jc#T}^Myx)<*i2Tm5l!riBk|gNV>?)0u6_-c
zYrY#S*Rlr7WrE-`4Liy->?qT)qqHlkPW5r6?FS%z;Nio6?<`~Fw-gi7sj<MB@{IwA
z&sP;yK`D}{H2_83Y5?LBGB?_4#)))Jyz|&hh|Mb7mgPXzs%vFAP_^2Qb;}CeL2}-*
zB(DX<9Ir}p)K-*a*ScAQZbfg??K1;|8G*q7HO}ZYvVTPut+kGaQ2a(<6f+Z&RI6&S
zjIMaQ$skpKjcsi*NDU=O+I7<Mc0k*Gm^Z_JfIAJp8u|B@2#wP7L#{WiOPQ=T{M8DK
zOlXcuo6OsjOyxgnN!B=4^Z63}Yqbqn6+Kx&8rXo9i^O`Z&-iUt%e`~ofR)?<zruLv
z-g0KUC1<uf=FDbmIkQe#u})dBO<A$gcCCzLB&0Qe(3XaTv}@3I$lkKX=LSGo%@}|s
zQOd6I#LT<1ZKo$FRd}s=NfZ<Bx`e3EthMblGme>7T_@`MLc5v0&psTTTzXeFrrF&6
z@zdFR<BeLEb_=-nq^g^+dZE#5qgAPK)N7-my_@$i<r^UB_b2WGl71;<1(I!QBH9F(
z?Fue`{Z7agMH7`LiYH1YN+-(h;d@u)efG8kC2}^u0~I|T6UVCb#cL~68nH@SVLW__
zbti#!c65F*yMlP4d#J2(N~o+_0JsAd1-5Omsb9;=kww0^aj{mEwlXgI#2s6f>ca!h
z_h4Eo{=Ja}&KfoOz{AI$!CQsY?2eZG+~Y-mA3%zV8CecdgbEUy`)4$lrQW=;P|>ht
z6m){pIn{zw(t8l16l{)W2u;d(8Brq_zp9AJSIRBryELg)hy<vqB>}wdM+z=D1|&uY
zKUO42RIa(yi0y7jAr-hsCW^8@XHKncH@k4uQg6+!3J>L&a@(=D!#UyZ%=hT;IJL-s
zjjsG1?iX9lpXKkKqrKDm{T^+dW9rdvHeI)|333M|+3RGVZ(f=5dFh>!eO|jV<+mxX
zl5?G?DcO@v)Wk;<A5DBYZJ*dtF+{Ii1pr|gtuXp~h-zx+bLA_cQw0tb)XPk<Uy9a(
z3<#MLs*A6MP1Uom^m!)ZHzOPS;q*CwYrQKreePkP&MCPBMI)hp^;$Jnz&5}u3f*Fg
zN{J;kB{s&9B)ioVpHXOK6m)epB~(r*`*a^#7!{usmTkW`OHtOYe<2sf#{1Bs5g`fO
zM>#oma(*itKI8YV3xdl(JKoaGBwz?)g?&mwDx}m>y@zN5;6sQq3E>w)B*||X3PGer
z$gsOH3Z#a2>3S;4zk5&Y|J#_AJeOgNKNJBom*Ltb6aqLomr<!ED3{SEK$d@i6TyY=
z^WQ)J^~5v4oOu&|hOhDI>GSFL(+IxNf`OR6{5XYiC4osNCV?-#p8oj9^!4;{+_WCW
zaY_24`Si!PX&CvG>E}7pj~k<PM?c_5R5cmnnGCNI{nUHTHR7cZZ#ClwaTU<v`0nCh
zbw|aK6MEbciOMxnR3SxG+&O>k##Df_{E6UqBM5%o*%$^!j5`b7!BuBl#f*h<cIE8+
z{PUe<N-k&xuztQriF9spKH{!zrfQ~ceGc!;v*8lLGjH8NxP(+pNU=tjo<I2flwmQx
z2$awXmTu(JhfVns+B2*Df@eOTAd+ka3V;t9c}T{D^K2=!sK1<}((`}9S}mvcCxYFL
zP}UYT(2hyeuZ6HA#^JOLLR~{CKz7YMRAjZgX5x(Ck#UCwDZ?9nLw@h`JDuJKIKx`u
z&pH^z@*F=pKZkim5QKJUmrwxl<7A=gV)RpG-8ME{FyZ~r!6y7Z2b+EknM(F`&myE}
zz}*OFJQ%%z(moOle;I#XP6})<0PFH5qVZUSqK=w|WZW-}ihv$n8AoI}FAA5hXR1UD
zJ>QHy(i*13f;%fRFN$k;3S);zs)S(Lss%|i8p>8QWh9`0LK2_PveihInNWy^^1%~X
z!|Dj7Y0j(4)cgD{8JbZl@gOFZDv$+h5*d~CT0b<2Lo1U*3FUvUX-<Mxg3UAIJ~a#p
za|-+scAIvO>*qXf%N}l^78$W*?VMoHMW-BHKZ@fp>gq_*G}~~FbeJ1t>fwfhrx}K4
zBL{9KdFZX|Vt17ltWn5PnQGKAnI1BKM7BD%y4i&!S-}*vfYhI*z9-xl@;*YJPAvT3
zS$Z!CCVsh{-8Fx}v=Ad)hwbi0__k|A6^15@Q+v8clNWZ|+y%X|(RSR0KU`mBkSbut
zRgtTa`rBL$GgO%&5?`Wbq|vPJRo7J3`3s$OSS!=)D$_|$id^4-^9mFLl=+eKSFuQG
zQ+XHh&lXN*D^=ENdq@&Am&{`zIP&lkD4;8?3q3?mKw^IcCf79#`Kxq|)pVAnUYCP|
z9xN@-corL`n#33Po3%bW+w`;_c^vZxD|Y-=oEFZ6xOO1SPgZe>5x$p;QzLq7Y$z@i
zh+7GeGzT>}e=3xk*BuQZ>SKBf?=vZ%nSE>SnNem;zom;g!CR!?S|TJ0ZkT=#JKiw;
z9(LYIzlVR>H%PyS9e2|2@%*2cejkIYq~B|B9m;FA<IE7PzJk8!6frv22l0@hI3?eJ
zyfTO#_@iK14DhU46T6>K-H=|h1!S+RZ788pa2JZxiW{#gaA;vza46Rb94{8dlyaA>
z=jj&plelFnp{$fqUVuL!$AqkgcE?p44nx|8x$A#h7lzPhjUG!$L=fc$J=X6NPW&*%
z*|>r9PEY@x!Z3u8!5Cwwkn{u{LK2<c{rlV3zkK-(7%uovSb+`?#T>kwexyf@Bmz*=
z_x$MawMWAM%wzNy(s%R&Xo$<Q^895wDYOb8wQGal!=Pu%Ve;QrMh|O)c`loip?rMU
zt`2{ev%b^80#G|3VpCxVPhpq5Haz0`vMM}=4}C6Ean(Pk)7R<ru~HIa92w-wSZ>w`
z2^iK|36z*+2m=Xp-=-47TNI^@QHp3C6lf8Ql8BmPmIMB(r&VOj2tl|E;a+8jGE~0D
z@eKkDxF(V$iPTJcT!&_b#|8#uvL@2D-(P=z<aOV0)d%vO_PGOeKOug+Q`Rzx*&gl{
zBd3k3Xr5kX7lZbY^QV%Xz@l1FPF_>5VN(R*<3vDAxZX8nKOQC!VET>rXP-c$M<aw$
zf&urHFdH4Qf`eA*TxzL9s8ni**2#86xszIz7+veb!N}Re{Z672_f5$&saBeD&Z2)5
zl}1Y<ma3>J#os6zanwf;Bd#04?q&vqI>duj8g_Xgam=McSBgWGZdc0)T6Z+6BZtd=
zIYl<9C_wlDb+0+6sR<fd$gA(;el_Ddn^67$-lx??ZnBhZ)3Rh>wTN@Ri@DlL0WI}c
zRZQt+fo$%|nWjia9l11lQC)#DmIr_O;v^}c(XwwTu%z`7TAfFVym^xH>m<uVOu1$l
znhLHNhNk4nF#K!kr)BSbN8I-{6IYx)$5e`xL=WckLOxp)%8Ak$W<hKfFI6?MQ3gCI
zW|&5IB~4eB2_rOfr^T8)ckGm;(JZ(9zRH1oRHVr7iq_P1SM)MN;wet1jJSV>OJ5kG
zuy#$0dY*Emqm$;@467iRv^65@j%)hf8Xav`?)t<hlH%o&kQcPlVu!!&Y}i^E4YMU?
zN!{j&+PI>Z_I_*_&~jSTyqOe0TT5YX_CIDWwMdp)DO)w4zTs2tO0#V@E16+-=>&{a
zxPgxC*3Fb+UN26rrB;(DB9ecUMEo~e4N?ajntkZ0wAm2AJd<PGIIMn&)xaIN9XA>g
zfh_J)iu+bU4)tzzPpY=@Z45=?c&*Tu3Va-m0#AEsQZ84y?4;wns9bY%5>XIk6~xC<
zgbcsy8s5xT6#Tk(?G9zwNf&3P^W{Bh2wS28$^5u{YKDNEVVmiyrB8oT->&sIeTHEk
zLc5B64GLE2kqa{m(Ly10rUe7p5D5#<>|Lf6qk+%+s3su+LSgw%&n+3zL`&iLV2+2g
za<Ej+x~deyp)v+p?(q&+!s7X2KbYj2Sxl&g9V$;Oku=(Q$ov{%rz4GD6`Pq-H^_gU
zooT1^*2>JuYqU@@)ir+yOlavNn!wyQT1Y^CmLl_yx`Wz@kM*4z<?lo#QZNsaHs`bG
zU$t}Kn6%^K0%@d@yMDa1Ojy4-ey(mw>?SIpC6mUCdyKR^*zMlzcmk%apk7EQ5h}W{
zz>}89av*oTR|^;^FJYLs-4~!Ebg<&SN~p1BCu=~kYA0KVjo5#c9LCIDk8gRJFM2-Q
z`=C~}_6KqQmR7s=Tv}0MR{~7zhlfs`an<R^tE3x!45Xu7t|%#byz?^H@|N^Oevza-
zQ-s`VP;N*xI~tICDsOjF0nI;C%8VEb;!z@&^)60dLfr|wDhDv|v!<pvp!Ye@uaNc(
ztzK~x(%Vhan>T+Vy+#A#)QGCMkgFPosHmz}H=a%YP(zcoY!`21?~uP=aZmw#UlZsl
z`^zc-py^y7Om4c>MDk&XB7EOwdoELip*9=rK2!P?dphv~Zu|l|;h0lyfvQ|y2^1tC
z9xkRGW&y`<SA(tizXdxi=vUmrG3-}CejQ!$oarSIcJ6;RiFvr&`1m4bg&PT^U>uuz
z*O@h}c$E)1&ZYmq4q1yqp<AH8ieFACeE!VM+=)$$QyTFyR|Li~XiR+AXJWVD7LMpo
zrEut+Fhm%Lu!>$vZUVS_3i(2Z1qxF$X#er@xX-(O#j70FpGpB8^J0nq2HT#mfD7&0
zOXfpb?!|w4)@S9l+|H`8Z4~D#L(1tuDvUFoRZ5#(UTEc5n`-l$J1-|YdL5BA9#XSQ
z$5zsYV6}fzpq&B&8Z9@b<8%k)qV$(K3sr36nwqX=lxjECE|AT=yEt2Sq$ee#ysIQS
ztO)Yt9|ZV%j{$;vHiniInkv}is<2&N@yshVp0iCTRqf|7+ocsuM(ub&({54~zFK3E
z=m5dI4$!@8MbR1Nn?;*tN@*A-*drhtB^u`$2D~*Le5{%GT5ZL24I?;e?S_g8X~O;w
zsmLJ(mtl-Q6aq6gmvN046PHCZK+1p54}X6B<B2DPIrD}-(p&%S-RtR>Q;)u~hFNQ;
zPv1`TKuKWIiAiwDgmlxdf1N&`KJ;Mc%{Z=!e>9(d{W5hU9}@>XM-Fmhv~D;E9NDTS
zV?2}eC~{D}<y<3W3Mp65_(Ez0G*rHMaj*zcag>GbA<jH~j!>8){PLfGb~As1XwML}
zWYW7HBDDCF*_IhC&-O=Zv`y3`s^;(X%v_I`*UT$<0Wai+#ibx8AejV+%}50i`NP25
z9-=_zp;~f{&y|h%s+UFTh0x53GkNl-F#C*0+9796EhCNmkdS!tn8uoYao}YcnzBqu
zTlPFAb*qq1&_g}vuFuLYbmD(*J@WDWy6#)YJWlC$5TqH9w1X&qOx)1B^j~ntEH`J|
zAih1%ybHJm`z&VLjMQ4;9plQMK}2t4IAq;MjDbh2NrqY`>@Cd__oU(Y%|u%c7N7!6
z_0wb{*%sC;OB$xE8!hJ+>Y8<DnYFXUBhiv5Rz-F+M&4gNrEwS|iH?5*MH1hE`)J`B
z{AljCP%m-6olX9Nd_=GssH{d6AFF-#vtuOCap789NZ61>!YpyCG(a(JMhl|>dPx?|
zWD0izF^|OnC>+Y-p|KI{Z^_=H|HC+rQ4ua0&GHuNC7!plrO)<Uq9vu1ox5_P<TF%1
zjS)h}iJ|9#@-YGoc$<I9j<cO_OG$gd48}%drPM*aAf;}$T%<Q<2}EspJr^?(s~mVB
zCJ#DFdPW^SN{$N7VJhX+Xu-8mFA1)lEf-x!#EGcAWPaty34%K!td2QDPo1NJvm7IT
zJz8)r)JuYEXRitF2U}dnsKbZXm4Yj&IiC~4W@Ia6$->s2Su=khzyT+nQg*j5P0s>u
zAsyjs2WjRjpv>THt&XHfNey(g&rCtE-SnO2k}PI^^7Q_;ifzQQLZ%{wR*pGdQl{rl
z63EK(Cs5gp)CQ3XC0bhMl&e$bIm~@(aqbhzK%Xx((iriflqQ0jFIOiL%W_x;70BlZ
zZkb)taw?ALcTs=SQPNrOhtjZ3K=9c*7MN;_iYY>*-?brYc&u7aQ4>Pmy{||J6p|H_
z7KS<J#|tYWdWqH@=Tadw3Thv%mEiW&Ys3Wy)o{Bv#)>cDS->P9OUM{L`^d+XW@qNP
z;#WJ5q|og!Z>@H1DH704K_j%^{r+cKic8e;62TC&cFBKFKQDOO&HCu2;;7~d_lAMR
z(GKETaqpS~+ZYB$^pFK#Ik1!WZ)fjSDE%(scYE<o2w(~)_xSp>gD9p7o{IJRif<Zd
z-wtT7nx`S|UwOqii1Vs;^bEK8=<J}Q3rOa^&WYXy^ayPG*y|kZhCH$hh{%XFWUg|8
z8JXF??#F*;^B_=8qN7mP>iK2I8AW!5@4Q5S;a54!ny{VDvRTqQ%cU!(cdutRR?=Cf
zAlz8}GU3GcQ<#kaq<4Dy*F?j(2q&9x*&!8xClfop`Nx;ffB5tlM1T77Pg=nzg*FS`
zP2b{ILwgu2r?2rl#_Iw!MsdJAX7OsWf&n<3kQaZ>1D~doLXm^it{VMHqn;_Jgs2vW
zYc@+0^9E&T#(^Ntqj|rt^<c)TY!PUh(V%YxC9S2exIV9<uk@<F=c%~rZ`0}Xbowy!
z(@XppZjiWHVkeMTB-AXOhdk<pLCGd|c#Dd<Hd+xn4+_kEsZ^l^-Ee@S`>O<cCN_^$
z1m}Mtgq0!Wq5R&DuP~saHL@h-MNE4tGcZzm)JjNfjjS!7KYdH<zTm3&)q8|<2SPHX
z&2*;}o~Kq5tZ;pbV3#hS)AT$$`<KTDsqc&93;rW}v&ZD2;?7G()&WHnK1>ygk=DCH
z?1$Y9La(8<-oc<gU?_}114n%jGh@O*yv~0ZAuuu%*E!yY=*Cv={H<rg^?-wWb?I=G
zrPPoL%6Gt>Pr1c>6G4xk(_YI|2#uR?bw<V2_1;a2B#^$_eiz_HY3Q?@YvhHToq;<Y
z(IA}2Dp?qk?+F6xG>dy)X5~@<VTA!F`E!{grL)nG?^pS#s5HR+AU>9}ahabQLrZ_4
zv$5T+Ds3F4XVz|v3cax7E6gDw+(WAX%e?nJCPCmTM2Pv?Q`uqU3yt1{qN;u5OO^5^
z6-Ajfbk3R08Y`;kY8N-8mFVWqxS=lw#K<jhYF8~UHVsKGbnX@4sbrWVT3+IRZ8!Z(
z)}!oG7qSX~i_UG+TRJ>jt?hTnUO<0>y|3<Ui4XUcC${s9Q1r~*Q-Mhd2bFEt&QEta
zrFnKy&or_}S`{Aa_S56T3$|f}S`}}tYYPW+bp$RQ%;k`c2lL;jc3O&|@2VhrM~|B{
z7>B9S6O<*$eZ|E<hph`S#D(C|F}pPP@$NC@SH0rS4ib4PDNvOf!?KXmbCG}2<E1A}
zOB6NY$sSoLAB+r0KUkM4?bM(rlW0Oh<QLm)W}hPs!$m!0a&!`)u0JJ@qMkD8^WFDL
zIPeq8b{N$NUn%i0@}x5zM}}RE>~@X{3{&VFJ51UzW89ICX@eAJnjk7;Tz8shcP6Ve
zVJ>Sim+f|m4h-Z|)%8#r%X5E;3o8wgGyP|#r|FQbFI$$cBO0GZp5$#}=av1SU^Hfu
z(D#Fa!*}sr+Ce2+8>o>L&ko{%cF+#ug?7+>|KF}1bQ`Z*+Ce*rH?)Ix5LVhjCC-K_
zJeTX&C2RL_E3Kg+#Su+A$c?wS`Kq~xXyzj1YIg)&*JvcE{o3e}ICp;#3(ku-&oHuW
zP<X*4AKj_eLA}r((9Pa|796|xR|dVeIpaXN_aD`oCdff6J5XbKh<h|<7&R8-Jm*GI
z(?Q&6vj-bvg!T-hk&$$C5D(k&Q3{n~W7Vz&Sm;~Ttr2pPmT7>|9;$_U$wRfX|HB%X
zoM!ksW7J`%@B<o{gfM@YB#QwJOdX_Kw4U;C%rAOKGhYKQ$aqCb6*jVft%f8KiUe|C
zLz3Xg2dVzwko4cskOW?$(whIdhNMMWonRL1kEC^@X}IlqnrYo^d%B8(hkPWS5pnaO
z5<xCG=pE9q5UvXcN+Ysxv6Q$J!k2b2SZx;**SCw+CbhEzqFI0PQ(m&YOQp<vcd3``
z+@%uI-d*Y?d%8I;)tLdK3sHVp_tgEK$e-o<tCt(fGcIKZ@cN>{=KY2M#lUxasAN^S
z?5AvwzG8ixM#N?bxS5gf4%{ZWk_6&aPwwEWV!Gf`rrTuera02ReJ2xdecE*U;ZDcW
zXf*Q7!^dc;i93Hh@z2yeqniesZW?Bvkc;x!TffYA?`QfaG?QYplzVGk)hL9l`m!pj
zDZ>9QWmVu3^D28qS#?jW-N%^yF4tTgti{?fJ?bDINzdq_i28+Eh->LtDy??2x36$w
zyngvMaz6W0RrI=AbWCTwz>KWe<J+R6Jte74w_R|c(t3XjwvH|+{RXy&epR0+Y&U6B
z(|t<kBm?U^KsgY*QSWj|y<@?$+EHqyH1wq0@N3U1FuO%y7TNzBi>uz^fUs&X1G>a>
zYTjVys)oA#J}m#g42u|Ep+I=d`xZ+4ebdVQOYCxQ@IRovQZXT0VXf(|bWAzh?rf^)
z{JBkX+W%hZBG#!6U@AEk+vOfiC8x5>1x)+&J7ezBc=a>SBUqCbOS=ne)J*B2#k#I)
zh~&nrAKf3o42^)|+zrHzn3*DFQLULB{BKZ{aZKQ@HORWZeL2kP-Z6^%H}lrsUzcHw
zKNJBomx28`6_;_WG$)rSKR};<6^tu)`t!>l?<5f{Me_I)zvicpFL%G(&E$Klgz{?n
z^5YI)SfhlAdNC0=w#%=7UB2FZo>%RK6BP(YdhgsI94kR6u)0se;Hx4Wc`n2X3Dbnb
z|KQiJ-<E0Q=TES5lA9~LAiOh+cTpH1hgNnVIt$`(AT}4vvs063E@Q2KSYi+s^gdy+
z2|gfv3rcYwF}@xHOUNRoP%$9w3ckaUT50r>X?ieC9ZWUNbP5`AoCybUlDUB~`HUtS
z&>+pp8-%c@fF_%b_=(?Il(8mSeU<YrHPL{^!c2{Pya44yR4X1_U@9_Zq`=Telhy+}
z8uACU(Nm@vt>i<Y-1=gFEm;~V>yqITt*oY5x{MV$MW#X5r^kMk3?1oHY0CDgt0Zla
zCVtch@{JoE*hLK<yK-RVMKb=_fyp}AM4P?x$j{{DJFci?4iYX#P;jg!{Go_1o7TB7
zsYh456gH+6YfR+uVUa%mFOxlt^ls)1IB=X-R3kleaG7TBn_78)C!cPoD>;$u3X}TK
z3}K>8ODn8XE7=$_bTp%<BBx9v8?B^oGJB*O4ER=hwT}Wmx6m8B2qC<qE_F;|ZEicN
z)X9jDR<q5;D1dSDg8$!6x(QZzy%L|6H|LK%XEYFHm05}VV#eBv3lL{kQ3k9JeI26A
zFAGfkwB*^a0C<;wTmH4+FaW`1?*n`rt5iZ3?tc2mx37Qr@)sol;GYNqK}u++z$<^0
zM`vMrP;vQA&tCZCzz>y3Q23w{g-Z!0Fkl_{vfNo6fOFuxjef_`By6-=cfdI`2As$U
zx*?_kSFEFq?|VPUR5cWtr%DDs;VJ}+j{wh5_=unWnWqAO)Zdo7ugl%%D$iu#AAI{D
z{gybDuqr_L*@={gjE8^$6()-#1@Qiz(Vo#dsD+Onf|S)!1gJVuA%v|m%Of4bJOpd;
z3OvN$^Z1ScCSHq3C{lw`0nQ0+B3h*)u@=#`-(P+Z+&6&w)P0wA?v*wm55L?)U=%WG
zTr1~1@S&}LvV^AdYz`}rC*!!}a1=VY%qoQrW8QMIX;C2I^U{%6c)f4Te!f~jkfI^x
zT_K=3es+RU;QG;5!R*v)%!BqK+PWPwl$NqEU<(8Cd7z|ZDZ09oRu5DarSve4<`~WC
zEtWNCmc`XnS82J&x_&4XEha*IgB6?Ava8q%WTq;A^i(;nY3pdt7O_70v2Nz-j^3u8
zA6&7gD9a%ZG*Yt0q{6Fl7P8NECg!YEaWuXqy$fBzHE^5h$;C>Ajwj_2_KD;wNlI(j
zBAa5mla9$EILCOdwdte#I-TW5we^nty2^W!AFV>plylgbs^+$;y|#9rs)a2lZ05e@
zgw3IUiWB~Q8K+Ucyzdg`Q8=OLHpE;Gve(2$IhWD(nJg5l@7!!)KdN4qL(VmY;PL*8
zDf194RBLabq%~TQ`AG1&lwZTE2=Ho#e}B$S><Tw9r>znJ%9M#XE15YCSM<ai$-;Q6
zVEH{bjpBLdN&&sCu5ZaL>YC^pj*K)nFl;k_+J?Ok9R!;hSLR7oTs2s-;dit9*2rVq
zls1k^P^2O;wSB?{4V8xrTks)luqoLjWNQXhl}_P4>ykLhcx{oewrL7_Sq8cbK!dXb
zgu^xE(-;V~<3JcL1VZiYAb->@!ymOD2t&8I?`i>}%B5B<@Gd!P>bWI4!lYY@%Zb8&
zWjb;gS75P)jc$@W_vCYAwxOavP&c8R)J>zkk}Tz%a~an~q%(ZnHUtCJ<J|G8%Bxr~
zre_71F)IvW${E5{bE;34e{sg(=usf?t$%23+U(2)a>!6TIvY`ezQ%}%iphv*l(j|y
zq`j%J&l8N{oa)x{-HX8xY=weaY>+p9MDU}~Xd>h-A4eE!80_OoQ1eY{J&wR8!5$X3
z76XhOnxA{<EM#b%Cbq0U#Ah{ZHPS>Qm!UQOo;!j2UP)Mf^@dfVerH%*rn(9>8|`C@
z@rW0ll!}p+d#EIouPtGH9?B^;H9xczABRSy>2@O2quS}xDC`r0s7!*bduy|Q<`agU
z9+}Z1tRZq?P$-;JcN|)#kDVh7t{&!GSLXNeKC(;3q3o<9W2xy{h{~aEY7BGXnej?;
zz&W$Te%5UniyQ05au2Y@x2ekAU6%U@p{oRH8aOWcGM}y0U-iMZ|5T-P9;_^K18KXR
z9XAt8KX20OaZrC%D;$;)DyCI`E8Iq5A=dmidf^7*3+na#dg0TG8~$)}@7a5ZFM3{i
zYDU)C*t!gAp8b}dJRPo0EAH8SQo_f$!hdliaeIF`H@@PPbZG-w$gHLKrY%`LOPEh2
zpx}m`ig^#|L;SGH&nxappS3^X{|`uCY4phJ$kHkZt=yJaM(2R%k3@=pPwdUw6;g(g
zgWkHf5PI!O2zXAAG3ZuCsDy|6S2?&2S~#ueRq;Ftt4*rnH;n<32UOs1IRbhQsjv=4
zYThap3yBf!U|jV`DjX=K6Edyt8r1miu#L|DWlo2Y-bE;w>tuy8e(iL{0p8YuDSeBq
zik+n&yB&QTE7a|!%aVhCmXGvN$^f5xiCY9*bynuvE~b_~=aJ}bDl)N5NLKHpR&a|5
zaaH5NR)|MRFc<gITZEp}!?m=o2tATo8l}aJfh^555i&P_&W*tQu9`+BR-cbiU@e#r
z{L;wS_JQU}bvw{2`QB6z#4)YXmV($rN-uZZ8_z{E@{3rmknm=Il)p^Bms13SknSLN
z2RttAOF0E$Nezbg+u#4L>@Ka{2GJ0#!U*yk206ykN$*-LJxIZFOO)F!I63NhCA1&A
zbf^yf&<wU6$*`jG9W~kRyczUEU6fbv!JW&T{QaTci+gjJc2XFrb|@Z<rBZ8)Ib}*U
zQDqyI+P|GDlZ3H<t!{tps-!2TV{3O?=w9!l`OW*e#K_}Yc-0YhTIK?;SL{vqIlMg^
z-jWJWyeU2q@vF1pX*i^GZwt0ibXjfxpAG+Kv!V3DdiQ?W@V1=Qna8!+qzAb&XLUSY
zi=_vdbJlC_w&FIn3;NX6#yv}ncp(Qb)<nYrY&F{Q;Ks>+r4zxJ8N>UWToyCCd@OrO
z40wX6;MGKA?P#BOj^EGomDu7H!zI}?ckPDD7_C>!r(V80mUsODOWdp&j{DIXDkzs=
zQwMe<xe9rbZfo<nRvq<7M`&p{d|zLW>|aYx!r~pL5d(0;SEpB<&Nm*?v*w0N*Zn2q
zcHJWO=L6$^m1e%4aUOJkj5(=;z7Xq4gB+rMt)2$xx7Lumz9NnAVWzd{EUH;&QNM@I
z@+!kwO;r6&g7QALL{P3Q@f&yDE#7;R!hD_)b$79;d#W@HHWO=A-<@srK3`)*hrQrY
z8<if7?i~NJ)}d;IbU9xi>d%Y%lbLm8i+?!-DQj+j_#NH`=)Hzc>ZY@&cDze`m09&q
z9wxfRlaH0xAD@Ep*;Mn}wI?58*gxx`bFACEI6;P#uXW-9xbC8|uiEbKp<?-)ruXn?
zg(bj*H+!>RMc_R0I&Y>sJ?znbb^7W<@6kTq;<r)Ge(OW;x9!Q-wO`|<-fwxSU!Ar3
zxbKN?TRt43&^GKJN);#dM!v>LW7u-i=x=e-=zAB6D~4mZEeR4YXr{S9ivG1LGJkVI
zFYrTfzW8@eG2C<|ivC>Xc~hm{3*Bxsug}Pd{BnJiuKcv?Frl-vc6DZbDnHkhd4<Q#
z%bDFJ&3^+|`IxJhVT?Z%m$9xO6_?3MKwk%Xx|<^F#bTE)OF$lfa-*Ta0tlmFc=$a0
z{7On?_`2l_rPP`)%g>JQw7=2dxCd|#;8uYBjYZ%V%`KYyDB`>1OYyVKZHGMG@OUQ?
z0|llH+!0)$w*U9TPdk|f?xiyp1<O#R3lxdDp-ZELBInc~1@h%I>EREiGjDx&&vtOI
zXa_~iVUZE5fOmX<9nyV#&q4t3Gp6?V89!qXj?IT`0t%Pj2Md*<Y@l?1lgUt{{js@X
z!wTd%>V%ApRhSpcZ}_^rg;_!$sjJ962qGh=V^&aJv%O~=;>^j(6ZBnDEzI!f9e1fu
zyuTT}C2?EUM;j}Y(MR+-C8xGl7*sPkG0SJu^XSDcdEy3tGNVz#&P0Yc*%7@7a~N*-
z-P?~3yWeMiO$r5hH@yG63m@?03e7GH+6r_y{PD~1VfVHto4rH|$z#$#lF8$nmUnsK
zd&!DOV~84u2VN9}gCP*8RM7<44CE$n58kYxbI&*_bBZWAgiud$3<Y=DMi}HP|DxDU
z%e!j%lNOqP&C^i?SPmA%PQk0Z=YMt$UoMv?;Zd)Yi**;^?ab0=W5%g~!40>mJ@iEz
z;|u1Alf~5MJ>;x=D9=4{OZR9gPAU}n8)_`49;Sa$ho+^TmLfAaK}d54OwJGjc8Cm(
z(SvazMNfUw_$iQc8~u2M^y@rFFwIpol{rQ_k*ubFHjq|6SvlG!y9)TU=gPzW#z;7#
zTX;uGa}_OWDuzC_1^uIwLFIsVt<cL-L}7~{OeF{tCT!YBIka)iG#ui{*;A9CeI`ni
zGeyA<)9yMKG<o_eS!aTtvYKzL6`4}DrtbtPeW<luY%{fML}3G)Q!K((6<EZpQY<J$
z#goQ=N!*WnIjq)uX@bZbVU$~%pt3?z#*kcgoD?~^chNU+Sz+PXN?#|xkzIU27&!h~
zORkMrz(EHZ$NR_k|4f@X;_?tBm|Z*MyE5b?m)d$Oximu(mrJ7D(YTtU9yt=m_%TvP
zM+)<yUQHfb?8|X1&(wxF(^FzLaYAL`i13Yn<1~W1EWFA>;UuIoPZ!E)VQG|#URT+}
zZINSP1Iuz_zsb1r$#%=5)$u915+}A^ozeAdUUEvq%Xh|0w#671<>91>AH6fCpnLg-
zvlw_0_&BOxcE)GBGiqKE7Ps3OFWSFq_gK5L)I(P8UzLt)xzt1Y{cGuj%17S*z>DO6
z_#O|uzYpB;eu%x90H5<{`2P@_1t>z#<P3l{fvo1g5XW{u{Pyv~zuy0cl^8z$FLdQo
zvrq(lO8lH(6(|2*n&DG^N8vn+4Z=%pkWcW!N#FxW88<e(A9lzt0GeFY`YE*fa31Hg
z+cY>H)otjQ=~pd9f<SO;&FntajiXS1BZs6yks`bWDtT;p<?&@ycnw$mUb^zA{|>tk
z!|rXiDH&yBB}^v5-tga&P5G1xRzq#q(xi+~sEjE_@s^kAk=9z@sjD!~@W~mK6IKF0
zjoe0^OG~A9X<}N1LTMUB>)Malq0^OHrt*Ox4};c%mjU}HWaB0SPtWf^=V2dzdC+}z
zPh@UDF$FTi4SFTCjJ_W0FubZ33y-OL>0L@$CEEG17*;qfEw7FUvwS;LA|=&sM%~`7
zHV20bYxN3y^i_|_4we@(N26kgBZcUqLhfCHFGqE)N?doP${Ay?EBnn@?WB}s<w|Hu
zB2U#3!Eu78eb4j@)De~C5TLMs=iJtEY5O|2@=i!kGrniiFNv!v)H>F9X7=m~V@r-Z
z`N)#vPF|Qi_t&BAmUow%YI|9Md2mX=JBztKs<6ap1*#B}s8Oigr>PE38Q&uj1tbo`
z*bBHNqToD-^%;jRu)?%8e4=zmbUTSII6h``CezD&9TX(^teiW^V<829<#`9>kG~?M
zfZ#Z`ZCEmH32p!=v;@!K2_5(A!3nFRDriF6`FyBY3nYv+?u~&2?G3A~H}r~*;uU}d
z?T8a5d;t=^CXk>la^#|ya>;WomuRB}Df5>D5^Bd=St4nMtSiLS)?3S^9<n`<ph*cb
z_F^F6Y3b3E;xvLV)7kldnV<)(G@ohw-vT|*;Dzxo;K9|w11+7f)A#gVcEjhk8!B&w
zwdyN@2eo^w-A(Et>-MZ#$CX@~A&&$P6lmeyRlx&7K2$lW3V1+97;E@cUP|nISP#CV
z#(`r)gC2~n4SFyQ9sz)nLDf`b_{+fu+6g}78Ei3u<<@T$K0p$G3j8Mcfg&%Iv@7w0
z()&sHLB(>Z3_B@z5r9C1PXZ8>Vp6D90thNpy1cEw$sRY8LshSd2lUg(04Kr7KG{+0
z!+8naCHtVe5XYBlDosJ<XgyYkdNK7}i~MmDRQns{QtcpBVf(rdnX!kRQpen~kiJiS
z&(%d02P-*~?poh}4(A<%K=s_h?+uc~y0J#Ik>7(UdgqLiIlWq*r^*w!S4_Eaw8qlh
z9m1!+wQHLKWDm^#x>T6D1AVKe<bF&bRXVa{+~;*wUp$n`dr3~BC)~|3FIT=)tgF-_
zS0!=8$fnx9>pUb=m>oOT;5XG0J%ZolAaEMLsdo4k_zie}Bf_q5b^K<D(VAa{<XpBK
zeC7q&eetzNVK*|QTfJ}>VmGB9=G1fatEEWMc+km6p;3T@nc>PYYK!Q=v!_hY;jyTR
z;ml0y8E6vx*Vr!tnsC-3aLe!6w*)r;O<ICyK$DL9^+1zVQWemoGxFK>xfU>qVLFx&
zwgXInIB0Hv`oo-#;uQdsz=@200ZhIoU;=zxG7?`(M9;NE1mJ{O{(aIn4Vcu9x3VnM
z3|Uuvs;#${OFd+Jz$7FQ#+ge2lSgDjPl?T(C|PBn8AjqrSn~}*B#%(hqHwSkG7{Im
zs?Kd~HRE(iwZC&F{3nXC+1?6eZe(+Ga%Ev{3T1AWU%@m51Tix+IF})s7ZaD*Q$WiB
z>9?d*Ko$Xia3M>Aqz7fIJ^QmMU@*e45mSn^g<7hSVnLNt(h++8@}IzVH}bhb5~3uX
z9Vr6L<Bp<JN(Xsd|KcgVo-e1EL%ESJlASi$jBjs4g*ix|fwzQ_{_y$tY(Nh$9zP$S
zH5V_@%mcLg7vAC~Tq3MKCS_=sElH@TjBX@=<-B1LNxuy+N-JDsFL0cg0yps?OSei^
zK=B3yPSA;1mtl-Q6ah1rfD}Cym$5|#CztSVIRt+jHxRw+EBXgSvshe#V33hKGeB-O
z<`(1-7_oD(gTy)H_fy=LZmC;V+XHB%rpRJd@zpBw>C?ogm&57x@a7riWSCW8#Ps@c
z0{Vi}f8aEz)noveMwni|OuzobxzO}F(MxiUNH0U5ExoIUBZ8$FKr?_w0UnM<0*xSz
zAkBXfirzWB1bsGX>^ZI1wBB+lJq4l-G!d+SJ^c3i%Yi3`Idj^7^l#(S)9c}%!^pli
zk`a^Bhc5^Jg+yRz#iW=scKZ6q^zra+B)Ys+t1)G-X-rGOK)S3kuO&s2%QL+Ly-fWq
zAA@vjXP|{-0!*PudQqfPY13*8i>rkq(Z_$qYR1lW78EehFrz|3DB+LXEYnP22*REM
zt7$|!ztYct0PSZGAQTBC2?$a)>hW`+>>&Kg@-(9wsQ5Yln?#KO7@%`2WKb&x){c-7
z(lBmt4K$qA#D9uUUq4O5$om8#$4T2<Yo$7E1E(X9ljdY}Tx4y_w~da@QBMLo!$^N1
zm{XLT%vX&R=lV*zlAh1}FR<T@5c`F))`Sc(QWm`9tIQ@l)CR`c#T}e)-`^T0ai%MP
z<^3JaqC*|`w-s?&yX85U+0A?j;evD7#+N{{D-=t58S$*15--|iKnWFj=|<kaT*Q}9
zjv3`<XXerlqS_!h@5VP_OjyU}pe26?#W@)%Csjgbecd|NOhXw*2Pt<nje(*%-3~?R
z$m7p=yJQN=Fs*Q$eUB*^8Ua00ykrW=6h}?U1ULWk&etU{sH$11lBR7u+XXk`13u3p
zkd*ab5TP4sMaajiEi(k-nNqws^m&CTkppXDNSz|1KBdT9-QcQM!6-Ge-0Xh|#-8Fm
zpis!r*^%NUP-IZ{vtybOJ;L&V)4=n?k!=4yZYj5*{6cNyRtR8{oZhId6?7n^?|Fr=
zcTZJ<fR`K#jfW|ue}VgMq~RX9V900M^BZ}Jqd9T4C>5e$+|E!{tZ4_lQWj^e+Y_%n
z#Xa!aQ(VLAAkpB&q|TXS#BzVgTUDCZB;<2!2S--nEh;#60K3u57_(7SVS(~^n5p`0
zGEEiP#w2D3>^mGWAamHBYst<}X)}z^qAhF^Lo%UVCja4mw<u>3bbIYMmfW~{y2eHz
zxs{V`Q6wUu4J6&!->Z&m=a+M*?-aJABS!;!mKOuX9o)b`am5V`vwweh*xJTz4qKzP
zM1o&Jgw#zL>=GnhJfnaRJ(kJQI)XKwqj6zUfQaA=N=QzjP0WYLw_l_oO*8J)o)GIP
z?g6o$;u^&I**_j)o473y8??0nVwfd@D%~6>cSwRB8wC4pk{~D#hNW{4&Bv5id9Fpu
z?egb6aV#wZvuB1sP*i^l?ZJz9`u1=x(WA)D;`u4<%+veZ*i}ns{SDO?ydA>TExalQ
zoUki26Ux<se9WNl*db|yONO#EwLNaasm?5<ct;aY<1Xpq-e?7r0=L#Ii#p~NFaC2)
zC8G21vWdHo1C4hn>dJ9MC0QX$ru-B;_%(O-nT+PwdilkCuJ?b!5f#G=op9g&vQ~J<
zLobV2!rAEim$k0>e4DFX!kE%^J#1T5)%LvCTi(4hS=NS7bT-QN0FOX$zijqRd6}u|
zln&(G&X=3>3;mqjwm;`~ynhmj7E<1=-A^fgoXXaXye5&&mUC-2UU;`jo^<HOlMzvr
z`w_i(1lr2(naK<kcgSQ=28t`0%((ylJ#cV}If)zjBH0-@Gt=~{HLaOa8%eur?#^I}
z9GDThG-!OtHrdnRb8_H*pT{P!KPBg$f%91FJ<r$7JKoM06$#Af+75e8m)v4N7=L^}
zpT;rEZO3O@>6=vTZw|}8q`8~78100%%+_CTt7a~8d2;{{P`n(&m)Zd(Bil(x@&tB4
z#>=OlH_vgP(aB=0_uoXe_r#6V%%Ny;lM5DUHd$E4d$ic~Y=lD67?40C0~u}<{4>$S
zmnqC90_2I+^xqVQ86f@nma&$GNPjR?+G;Q6;mzNlKK}gScVM*aUq6+i8l@;ee*8;(
zRFs`k9V4dC=~**vH0lYMBdd^;vdkNXC=1KLhv^`Nq1`FvsiU9$sAJMfA)3V@mKr}X
zZ_o^79z4Y|n)mzM4rZ?E7D1-MhJW%@(pvvW>+>rA=wIz~o=U6!oDLtS!+*P^Ri!og
zH$sCI6i6wY8Y|_JRw-o#gV1U=520q#Ic2g|oFR-~7;WV(gFZ@Wif&R^W(e&#mEDTW
zd8Cm$4-u_QArIyEar}ZF23cvOhIpnNty5I&IS)}t+Kn2npFey_>psz{m*zdvIf6hY
zm7b1*TZUAI6Rm4^U()nEJ6Kz24#ya5E$^~qT>}By?%mX|6XiNq!QO3F200J%rW=$o
z<|S1LbQn~#yP|{(B}$qOmb@(1pMi7gY*cy@3pu2BpXfr8q8m5%yl`3QLg8Y`h=m+?
zzwwvJWI!K(k4=tw+O2RC&X5OjBj@|nZn17tyVP#pZAd3(Dt>I3e9o#Sr3)BUZrj7?
zsyL%BRdMk^&V|bvXcNRUXqPLzSq{Ho!i}*knQ%47Xu^L@{j?Njk4>E2aOn<we+%`k
zr;gaDHrZTkqx&n34SI90Z28F7nic?#A+MOkhj7t<F;LS@g|$cf!ZM+8w&U*lc*s8*
zd~*lGGDog_U^MkGc5@1vDh=-xhh16=IVU4OZ{pk6egZDAeMCDNZAFj&R*nW6=VM)?
zL|;BAp4a)&(7Xw2x?j3mA6zJo=UDr@vGhwbb>%Vm>!<R~N@1gPnOP5QhJ1R*T2A#X
zr_d{ZA}U$#wO{EYN*|6=U`{S_|5|sAs`_+l?$2R5=&rF9-0mrBtE|heQ>FQjeK8+2
z9oOeywQY0my3x?W{nbJx{qmCiK+tm0@U^?GNfY1HQPYOP&-{9;uL#u->8M%B4C0Or
zwVvRPcUybHD_2|l$vfO^?Fp`2Y#nz0e)n1;H*uA(wa#_&9dES`r?tza)}H3#oz_K5
z8Q#ed{|7g9?#q{9j6W0tG%=UKnFA97IW?ETIyNYm+GaqQe*%qzJN)td>m5%FbLO=F
z(|@&hkI#31-?i*Zfy_ym{`_|5A4maASe#HA#*LW1|22KN`>iEg7UT3z{7`)Q{&i|b
zJ|_enCj`0HN>vB~Cs^gAwPV6RiU`UF&K2!U($3|HF4`?ZwabrB7FH=rPO?xfMRkmC
zGwJ!tAA#L!e}vebDLDqR;|+PmN13e*(@M0ni!4u<uWx7sbW0IneSIUDHqaqS$Vrnf
z;zPMOrqqnAR0(N9oRi+tEmxS{2~s$C73DM?9@93qrM@Y0QZFgYNRihRZ-HJ3CY)_4
zo`XJ%zMidggio1su+qTO;e+OI8@C9CA*h|)A_9}ze|2bWD29OaoX?8lHDca@I)=N*
zJ^2mJAbkYxtC0owf+G{WL(eZ1D5}+62vA-Ty0=c_^b-|R;700aLs4Jvl4?p@E~%b9
z!X?!dw_H-Q@5Uub9qmFvv>9(NK?`L>I*fNV6kO86j4n1BZ5dtNR(xCqA6uK4dI4pt
z7+o!8fBvv;j!rP+I*}%0BGba#L?(dZ0m4p^2^373j(1E>Ac#lDj7A=lLjW(D%lXm}
zl8=@zwTgdv%c~T2QVWh~aCqW^5%?p`zor-*smD)T5uTD^*A&mW0j(KmY-iu(1}t;f
z?dY5vu!4bcYx^(Z&?8C#NS$6gR4URNE_ZT{f66hVI?hE3>VfN|59K~s$>n@G%Lsv;
ze3g*d@FmjvDO7vjf-^n&+G9pCXkE76#CLYU#gZvQNcKX44!l<*FIbQu?Xb4j&WuoS
z@w+TXIK?_lynb8}IwiME>lK^n?D!(8KAA$($?at?CWU0sZ|w9cj!ao9POTj1$Qp`D
zf6Q6qdaPa;aW+}y6T$ou_En86olV{c<sJsB!YZ#AzVrQqU#++b9ThH{L(Xu0Y3?0P
zxH1)AYP*O|xYxaTsK+=bm=?|^1TGfz^a_WEMx8F%t75?2q`c#y>^7^DDC(5<#UFEO
zuu!!$;!8a77@tUf11|}8VVIS+;I5PYf2gQSL9)@La~)tJHaFpr)+RihSS~$Im?44O
zCC`PX`M=L6Ly_bzHMw0yxMa@2E)|iTiX6)`{=%$Uk@CE!kZJVlcdM`^<q2HLy#*;|
zNhkN>BU&e_Pw9Od%Y_{9Y#eUW*f$WHG|0G{kEJ?@J)Z~_#NrIZx5oy;lTx#he`hca
z#W~51`wZsGuZ{syIEG-aQ~Eh=j8CpnFdx^vta2P%KXNTty?kse$%KuV-P`E^BG{ex
zPu8CmoBP@0k;h7>$0fH5g>#;#NMYc<&8ZLZ*{M*InMydG%3J)Dw_VoQRg{;tZA&kS
z7?ri?vnIN%$YSnBKNrR7z(Pyff2MQ?)@BZ#gnTAu%(-&$440?Njb*G$TT1WNpxr=@
zj+Y+R14^c-QfLqBHI-xz-sG^J)}sU;26~N#;>rQ^#U*MgdK=2;O5}7>)y%%7M7I81
zvcu|JiR^qvZM*efq0mG}X@lL#7gC{u0@rY9Gkf)TbK#iF?N(-IBl0dCe?#k7ue3g~
z%SYG)i!E|DWO8DpY19;N=wmgMHV%!O*#i!Z8;To;#x0WXwvQz!C!wWBioB*cppVs1
z+~{L9v){arwU67}$7<2~(S59jpPhZI_N{S!EXvRr;K&@mrFeZGtD&^zlIqzbTvAPO
z%Oy4Yeph|0b}QrhSPkWke?C?-dw3sfpU83_t5vYq>tnTi?dxN;*p5H`MMqK`nX5Mh
z-POil6WXx-dh!6%uL*8gem(j6&Xso&L#!dE%DZ-fS~60^7Q^wyMGk?xxDs45rA*7!
zX1#uCv3|&hB3BxCZX7k6+UW44zEJ(rYnZnY4Bf~!bwlxux;v;ze|3cTPjq)s0iyu$
zF1kDZJU}Y_L%KW24Lz~{w|7=_1RQxs<9?#{PKc!%fA01UXv%`Qcqi>0kid)`c{I^d
zbXPlk+BUPO|23@~WnXIL_~K{Ly4>5G;Ik5z7<-Ic^WF}3R+K4K+x#~L&2V-p#X2aQ
zBH;c00+yCbZv-oHf5Un`{&!_sL(Av|Xs&e)Hu^I@8v8TudeNv&2aMIZ-!I%RP{HNX
zP{yYj;XOUy6{OrUm=+2(j`_}{z$Ru5?l8|XdpQW^Su*Q?p;8X@%A8aEMpCJDNlB_d
z$md-vsZ}tRdS4~&H>kd%Fp9j@s8o@(_CaNmz}bi~F@MYvf2ClzkbF#!4~X~k7IU2v
z-d_2Q-9kSWz1@Vns`RZ;jJ@53o~hW?O&p1Z|63exE~8;dpXVO!Lf>jnkSnV|*ABC9
zL6yJqzrm}4-l;V{+Lsb=u_+H?wEdU`TxR}KQH6#rPoctDidjc6axFUm3?dtU_&pvB
z48JPQ*?M0;8HK&=0Vn)I(YY3!Q&-gI@5kr6e*mFM7K@i*j6W0sG?&4o7ZjJJZa~L>
z)ywnmpZ|H1g<;N}bts63#$9*@L*IsGFvO7tNmqEKD@7mT!_d5l`5NgOX%6{8BhjUi
z&)*`)7}~ou7C+)A-igl=C7r*OOx*qQ;nkuJ1B&20=MqBU!DQaj9Hz_@byE06m#@uh
znPu5h+IXQ+@mZpDmCq`9nh`sV#Cx`XmzA`plLq{BSHWx5SkqRtbkgPPFt*HzKEIPR
zd@WK8^QmH`O---TKPHnZZ)AHn8m3ZAb5tt2mp$q}>3oJatA+Hs>G+x{145FZ{%7pO
z5582(_D>B=7)Ybj(};e$k)OB;A%uYXuIJSA`VfQy;!6soj|T&ne`v+b@BaLMv5Hm=
zQ8veAjH@(R;#%amcGra)rSi(2^PN{ahU_`A?Vbn4FPpDl^lQKe8-#Mm8IFzzL-?MZ
zQw2~~?kxc7jMV^AM!xjs>>*a}M74wp`!=`XZ&hO!H!=g8kwWRVpc@ym@cP!E`b|5m
z;DC0O#>XpDz*+7?S$a+0_T;61>Wevo^fVj3UHNAfU1l^fmZhFxK@c(5MzI4$(o>~w
zdoJ4|&v!O7o@Hr{V~c_=yo4st;@iE|N4Y0XlFReDTkuR@*D77pcEv<q+Gj0M^g;L#
zI5HZlqh%v6x>^6oIx^Y_g|I}m;Cd0A;6)--MptyJK9ENHCs$WOTf&2XA2lBFTiyy&
z=+4PRZ`*Qb#kTor<RYT4T-iRXJjQ7{DM-SZHtw?sNJB}N=Tz&8&f03axfXxbWHz3y
zBH6Sji}tGSKa#&(w6B)JDp*Ucs-1UJtE_2aiR6(2jP;hjZIC)^d-2K7F4ePS1&=d+
zY{395Cy34pN;3IdS#C;yPV|(}WZk(%v{gh_1SRQT1|)51_N_=CY-BHf{a{z`pHKgr
zsU8Cer||N}-%s%Yf)YT8RLTfr-Q}M@FMmD#%E|UAeZs1=)|eq3RisIS7Zbd&Q+%%(
z`L!^M@-qENCK~SarqMvk^^9P_)6628gd;4o*>|&x+qtq<D66-Bxgu0JdXtc$ge(My
zJ3T>K<EShnsG%%xGZ{WEycP&`4K?L|%N$az0MfJP7IWtLPxCdF5`D|nEnjmL6|1or
zsH{}?Srw!fAmNx*L0Vyq<GP80V9`gB7pgOQ&;OGuPU4r$0}c=JOGT2^T!;Fj3;4E6
z9E&?Uoo7T%jxHR3;~CM4V7Ni_OTmxtQY@eEZSr5am60tI|I`IA+(|(%$W2G$ov>D|
zSDY-^t+*wy<lYpg1~Z5H1y7(vkd9N<awTMxWj^QQtNzggh3~*At~p2=g)r0V13aXr
zgwJU_$d}e3buqXWU#3*9;;tsbun-K*d*RcD3N>wt<L&Q%Y~ikk5DM;2PAefy!MUyk
z6=E0a82${&6_?jYwWKbyxh?4q;S6*jNe+fnTtf#UrT(+&9}qq(XpgZ~K|5E=v1Y(-
z9iGaF6R=bCXHYnA&Vo*iCQfmx96Duc1mU$k134A3BgHinRWHcqe}G?+ls^S`kS{G!
zpixVDF@7q4C;HAPEOyS;;TB;A8A1th_rs_qM<dgB;s{df4LQ|TJ>9`?46amPaenR4
z|2f2?fmox>c|0e1I|P}Qy*(VtsI&;mALI*_OLaQu>um~+5gFR_U7;Kv=gt1HbzD-;
zC4WMo@z$N`PezJ+{)El{K!0+GujWt2(Q&sbwqTNf->Al@c{6WYymYOq`<+EA*ceCm
zISWl?00mJ_bpt3m8qD{_PAnGB2&xr0Y=%3dYRxG|s!Iaaq>KuZKq}c7SxOcRl+1Pm
zml`xX&AHa4p_E63X|YeuqqCZeN+Yp^e5N~_i?N|*X6#pjEtSV<`Ic=P)>X}ST)!py
z()VG1yDr-GmQJv*Tak9J3zzmzzSv{4C_v9zY~X`15Vl)vM@r|^hTE>5k%EPmQ)@Dm
zyhn9@X;S{7g@ZxRnx9eDTbRimq!xa<y8R|koqaExS-njL;@&z2Sj%64%A%W4)9-af
zWKIp3#B4;`7eb5E!f1U9!(IsqGZB|c-PU4%xuM5cJQ<Qx*JQ%3$yhUnnl+>V<~Bn!
zqjFDbP%0coaV<4)UIYpmY;@kCC_wa+FGv8qE&;IVaHb!M3A(G_@OFtxyNOehTV*#J
zd<XDq$Yd0EaxFIchHdQISF#sQ+JpSnRlLkzS`VRouB*pE7VS(<5~fVLX3%NxMVEbl
zKHWLYuL;}GJ<hg<h0+<}yxjqYU?^_^PH{EC&FL^3{YWsx&U3WyG5(B-`QicgT3>n{
z06>WvbUAa|?kfQ3(#vRk=yzhwm}bb=rK}jdCu#k-GdS$|FTLa_HNDIGln^}<T3h|G
z<PM$<j*461&RT&%ox$EnX^+2gv{|ixM8FtIuk;xVj+FKo97kh+h5)7&v%)le83$)D
zJyP0ZdK}%mKoh2YwH2lzZq4EuOplcIm>x%`MgDe(VXZ|<XHSc;jN(p&Wuu49en0Kh
zE(M+yf#Sj0V9yk2q_`&#o8K3#!(o9^^<SzQBWwBE-T67olZd!*(-~B0YpNc9Q@nUe
zuzo1LqbK@%=ds^dPv%+=__UE>98r&=<fn0l;dq?u3(Gj;B-nD2r>XdJ&ObWmDBTS<
z+nO~4S<!nwdI_G(Fw~TidlmjIp~l(vgkrocKy;<yqiw9>Tf}Jv*A0NJlnkXf-L9Xl
zc!FAPxmIo-N2=|TT-!fsokrJx?ri2q{rcHPH#^>P$L@aB4CdstLry!OQp-?v^Q^oJ
zq5(zlQ>oZ+ct2$YQMd>6R~Y4c%Io5+8O2rCo=-M$$sXL>QBG)OW;|}RO3L{HuBU&k
zx#LYLM{nRbp12E(mR)J?%vOukLP_`ap4+nQ=bYCmJVw8((BH74_s&dzFp76In)2xz
zMsc;#lnZx0{{!}*a);v>TXm_h(P?_e`rX&5|ApOY{9zwZcPHo}6|#>&NAZ`+OQC#t
zfG1Y*7fRjX=<_@=qkMuVW)yckF_&y}#V`5|g-!rJp!{WDU6MXVbi?u0x{bYY&{Z!A
z6#J_N&!~IK0VzgtOPUIQi}8&!V8M9Ufd!lY5nJPV!{r!XwW+(rm+BMP?pRH!)^%XN
zp)k|$eek^;EtStLKkg#s#x|^?3;}{UA%9nj6ajXgzw2Qw*2vmU33VrhQVrz06SR#+
z@>FK}5k<KRqYp-Cd4KtV#x#z{_&ioER|4JXgt4}*KRcQ#hVI9IMXa(gQ)S&%Abo9B
ze^}LRyN(|I3s#UI(7Cx4Zu!C8RAqIN-)66Rrr!6N)KtYS=n~FZ29b_Zr`KRX3!~0V
z*hYdU<ZC0ropfy+d`5o!<t>dve3u;!W`(UC-}}=ruY@uCS=i#i1or*@`SgGI7Trn;
zWo~41baG{3Z3<<VZp#@412Ho+m%+X*6qj{*Kr4TaE4dBb_gDA_u0{{)1sDcydu|8F
zE`#0$Sp}I{%_8&vi%=3JQK~9exm`C4ce_<8ilq2Bq}1~A$3<TL_V)hy_REKXix<Hv
zuP>j!Ulcx&_)|$7boSzv63(d0=bx8<UZvCyKkxVfDUHF8$L|4u+uJvz0?tr4L*b}W
zw{L$Q6pk1iF*rwC{4MdL@w=z74|?C>`#}OE3P@YwM1|w({r0cVzrRUhSc>S<*Ys=t
z>&NHYf8S>It+xWCzx?~pxAXvv5+-PT48qB9`RhNI-{1Z*lWmJhd`H;y1Hb+C$7LG%
zjWie~YEU|7?Vv#=QLS)sE(((#6%9g^(xQK4R^fNlj73er9e?XIHc0!+M^LMjAqoWx
z%1yWTLInlC^dFi1VFcN46rB<g)ihC5BR_q%6Uym@WR9djMl4DMx=5GCWZB0?LF4wI
z!8sd*L8Vs;Nf${sSuf8^0U$!0iK<`IXI&6$aAug_Fja`UHIjb+O6;Y9k9C2-3ow7a
zQp$qZq?#`XBhm5fO!1L4)hxz0&M=FKhaIz6uAfYxZZ3Qjw@(^{A7yOTW_2Co1o%)X
z$@{v_wG~#yrt`&m&&yryVx+9)hTNp3arT<r@_kzh1;TbuxKo6v9`R28luQwlc|I?c
z#}VbGg|=k!K}2n*$*-7F#5@whL~wu0lA;w6%&e1#@nvn}vtfO6QVp)Sy#>->R?uxd
zy}#xdv4-{5xT$*EPb`~Rg?P153x_JIMsJA^PuayD&nDIjXU$CUum!##3%R)s$u^*^
zT(hHpl>3`hHax;F7R;Vg>PW%pgF=4RUtW6;9ogjap@+KAlLk<TYky}exqW}b)vjbu
z_NKF?B&}8+nmv(rwGD_@(3|iR3~p1_@M840w!ceW5dvBCqNjOH?s-IGH+3D9*0*L|
zdD?q{3Y;0rM%5GqkiKR05Hopk0#6vzt0Ku6Gm9VXojlrT4NhF=loezR4T-AlE1_7G
zR-znN_*zj+0==LTh}H3Cpl*LIf@2+G*3wP6D>)Mo4MQ7H@*q;TY36s@l2{$0bI@FS
zVUEghu-;slXcP_8*4er$(pCDp9X_=&>FIYqH))jkU?YQ$Row@8AKtuYeH1OA18A}x
zpuh#I{Hf=qnf5dV?;51DK4qPm!+2ynwW*D4on4A)?{Z+I(AAJ*;`agjml}OQM1P&)
z5&__6t3j*;g+$h-Y7iNB@o_Df>%5KpNm&eh#D{8tp@bM<HN&*3=qu24T5Y8)kVz;?
z#)#xH)nVjOW+q5Gt)hX9>3$4Zc&xI$nyJVypvPc@@^QhBoFsk$ix57YcadBvhm!F0
zY6}}d2UZMWp4SVQC?B)F8g8H74-up)RP^(z4~kPQgYeSLZ@f^K?tMTcBq_fyD1D;7
zS(27ge471A{YC}hL~*NttNuF9&Y*^H=i}<>6_+-CKso`hmvMeTBY$@q_or3UV64Ds
zw^L2;q8SZBJI^!KS`MkuJ$D9PXm-|kHLkkZoFZH?&VD7jRr4DuZnmcgD@^O}SO3rF
zxG16z2l>(lXBoYwmt#E-)<=7Y9dbM?aZB&C)~Y#SQD9{FneELAjK`IBdbZQ-ml7ly
z<w$YMglamS^z!9);~8@f%5uzCnrg<|Az8+o4V2+Z$wn19P)_@oEPp^x0V9`ue?VIV
zil+24m)n0pIDfx0|4xy}S!zR<Z#B|oD~-jX)Rkg^TWuw~_Qbb_tgcP`vVVOYvRua!
zPkX<%*^P5o7AxDEdl4u_2ve;8PF(ku?*pg2aPchK<&`@h_Bu+_3muBhnZF?>wRPNT
zTq;d{o1dneFSQSusmI|$K$C@HbmMJYXOu3_Q8kux&3{tFT(uF5Y9=m1iyKbUPdPj?
zGUKFi!J}tKyxwnUasUafGA=mbSb#MM*0)rkTD~1}^>B#f=M2sd1~M~re+hx~O7<iA
zL-xD16CsuWzuKn84H5{zrY^M!8A<;?t}N+*0n7i4D@zmQ6Yi!=6nE~X%#uwFsT1yw
zXbA$WT7U5Mpfy_h4zp^GT|%nxe~7lvGwsvb{u!o&75F)ELv;)SP3_=@8mX~xGHY6U
z4XvEoe$!Ye1?5g`zfBZhDp<>Xpo!vU!CK1OdHtWz^B{LP#8z#?O|$nC&sd%AUmu#`
z%o#>fqpbt{?SIPk7p<Leu^)HT+Y(^}x7v?w>wkmQdOp-~TS!%hKo<uZZXS=tX$G@l
z8sQ!G*2su#%W4ZfUzY%WJt43fvRBuQBTr}x0;{abJN3#)xv25ZYY%q0Ak^+cmB)r|
z96XBT%K}Uqu8)cbvn=n!lfK5>3Ok|l(Fh@YoVM-jM_|VayBLMXvT|IsYp3fyw2DNT
zcYoC|zDw6nlw8HZDmoG5;^vrbU6+pMslEt9?*kE@S-Gs!5uV@aZ}bjMu0N>vGIXI>
z{Zs9p=JiV_r*%6plqj!ey&T=*d{P0eYj$LNu`CwGE>_&AZByGb<wcHhRIz#}XnICH
z1!B_ES~byJ(mMUs5Pq8t?b{UIxS!UVKP+8ZqS@_oa>4ziiEW*u2Efg-Ekf!K9_dPb
zbZlR8ZLxX){n792j3y?flMk1=f<Pq!50~JAKqP;<^H%Xtz)wK{_z|&NwkKE%+tWQY
z(;5$}I)ES|J91pyv3nxRuichoILL<iUR_P_xj|wo4Ygtmo$}*U*(lq3b5QSHkmV4f
zt(}2}ZtqXJWE@HEU=QY>mv#|@X}{RFgH4qE-R!sZ4!_d>zh)=CH={F%V0HY7eOaX~
zVq||muXLW8n$fE(y|NNhPgI?b>F`#~0u|Vy2G3<tNX8B8#pKL-WjWI1xhv_aQc*3Y
zmiW@!*7e;3iIFS1Kpu7Ft=z`h0<_nb>sWi`7}}(4@~>YL%|_w-WvXSomQ6;AsvW)|
zj0AS0cmWRgr7xh7U*D%l?fTv{`V40BMooV|0vuuD%CX<D>l;UY%lx#{&PW{%eyg46
z7#0QXEvg5gO{S|+tB10L0_2!$%jj^RB^i_}@6?@yTCKMyBrj2c$KcBtA>)F_C%kym
zdxH@{>zQ&xIv`nTH=&`&fDsX(un3jcDDU5i*aYH;j8>>_9fky62&X+WvFG5^GM|6K
zqQHfPq?@-Kj#ZR^an&xUX6%~(L&6w_*hXwQ#6gZ4Nq%;WKs#(V4`HGeDvbZJCpUl#
zUBc)UV+A`D7?!8HOBngM3D0jR0KbdC!dG{5OhB5uG7Ql5&V&8qbCUHSRW}&?5U3W3
zYI$d!e*gG<`yUJhsQL<JZe(+Ga%Gokj6VzmGcz!kQO_?Vm#lR^4VPRhGA4h`T3wG+
z#}R$!SM;Ovke>di&&Wc;5Nu0{gn&p%5D$168Em|RmceoUeNOeAx$H8tcz3z1BVf1h
zrEhn2b=9d;)m?jUoJy)idZL~xo+X)#hNjwlv@BI^qg7L@94*q+xkigyA{nh-lhiRt
zZ%Ik-M5`yWGTsrUsWh4xqGEr9XkxBr)8}|WyYDe0(-5OW1er!4{IqEV!WTY`KrOXm
zH73IfTHAG?)cURir$#zfLY|t`u@o+7PE%kVO>F8Qsc9dx3C1-U9Rwq|PJKD`(k*6V
zNxkZ%184)UN7EAI7!$LnF4J2nscRpa6H490Mzf~wW%LzKRAk7Y*ouET$&2Z!`dAw#
zqTa^Zuoo2bL<dVmWypXph-J)Xu_Zc9tci5mGNnXD(-fS@ZFIo0xYn^YYy&s)F`J=j
zV+zy6#6u;H4lJXoc*{04X`EOmtwV`ylTu+syxV!#BxRTql@=y4WbkNC`{q4KpGI&>
zLesuEC*j=)o@APWN<n{<`)I=6)W>Z2DK^Fp3ldizO-)*CGz%xKEld(>PTIyQ2)fr;
zRWvOrw>CaQU*Rzu6r(cqKrkECKKizv)I>)U9b8AVc(UCxYe`lm)<(4^Yd+*{cuQ@F
z14l9*eikoG>zD%5oAMY8wr-R92>LYl(S#>@Ekj65<gJf!VXuFVjgHwMpYyN>7)x^5
zb?6D^JH~|<Iz}5!NK#54qlFumI;?#dFV->jfn0^AK5(>JScYB#D+;^a`TFbCgJ*#Z
z4KVU}_0NC(8~>r%;4;u=hH0hK_ix@jzkBynT^qF=MtPO4wdDp?U!iMtbgftG+BrKs
z=0n%Cy1#j|IeUNd_QlCM_4pvI9=+IZ*Jr1xO^<*5ZTsz$?ThU?)#=5<&FL1CJRDGG
zG-i1i;MS;558<_es^cN{lwo&xxO;c?=xlTHWW9a1di3Do>goEo?M~#U|9!j0%s;Ln
z(CK!4x_vi*YJ9eOyneTNe|EBdw*#+u_xJV7*Dt=={PusW;4#ony*<YWXZRM~0Dz<0
zz0=dpG#0l>ggehB5RbJ?Aii()xkYn}=a$SZom)1yeBYunx8^x$o`dE&Xr6<{9M5Mh
z43Sq)-v6+Tr@y^E{W-0^*_^#xpN+Ma=d15l->>dJtJA~d^~v@bLK}7%xa6J%zgw3~
zfilH2Ap3v$>R!U|PgdV<o^DnT(pMFjS`{AISRAW_$H<@V-aWS$H<1FyoeAi@mjbAs
z1G6p@qOt;eiX$!~`Z&O*E+GVE#4g~n=0H=bmKk2ht{Db(OHyFK%z#a(H!GY2AeI9J
zSv>=K52TRTirShMfKr7pwtx_b5;+*iUI;FfoiBeEq5>g1!em7R13<zuML%DPW}F!=
z#S9dJN)Xg@AT$QgB)6l9CXBhkNwj4I$RID7z=*&GvL4D%!D0|r+!P-m&Nh{{gb+}B
zkTZjYf-!Y$2Lpc)II|qHOmVAL3`_?Os>ZD^u^lWq=ni;cu48t|732ksFl(bjDN56-
z2)}=<R-hO`#|5O&k{#aJ2vUHB^ajWXX&6$tA4ma~gVk$MK`?O(GPi^(V3CLmtx?9N
zU_%g~&4;ssq8zG%LDkH`z8iQY_&7M)68_<U3Xpo3!(h|gfSLvl(6dKS`&bp$3s?-C
zvuJC*xZ4u@fq)SS4s0_xQ#gtT>vY(v9NK?oQ#eP%-(F$H2qe11emIOGcqmLWoD&uq
zu@lw@7&~H_T454^ED%<(X|UfV+(R`xPzodF;I0u)hGG~(s2;)<$T5;6KtzE(Ba#-M
zg%~`Hbmn2mZnPaz#dAIOObxD(E1dKST@kyod<jZ-j<{XU@2*Yhq#shc9ap<Hr4xTX
zN9lN`YCKalo~atoRE=k<Cgt;q7a^Fz3=`6x4D1@|K*eOpT(X(mj`Jc~!bXu=TLv%z
zI1IiUmH|`l^_F}W{=xb|2y9M{jPNocV7`LfwrDqmHUsE2IwHmKiD8+k?iUZx00dw{
z@H-^6L47?00HQhh^E($7vbr}9f3kmly*Y(M|9Jd;{QK(FcKi0-UskIhUvFQ%{~@1j
zep#Ko`uRWT^XT5w@1AVl{JK8NKfQsFcRrV*!dQT<dx=9u<UC7!9f-NP;Wo=r{hEa^
z#J*~m3eiPs2P1JS$Q|HC)Gp$joX<BS8<CH|?iPS43T9EPm>wt|^z$K0%MpLbvYSDc
z_H&RWvOwlnDQgrqu#bZMYdwe*mX0_E&UUzu-in4QxI}RTgyy4oCH{Ny>RgXG2*C*r
zPWPoN3m3NjLYd*Rt-tne+mYSY8}7E9uQ|vZjC9WZbq-p(ge|m4eS1_9!5+Y4$pU*5
za{XNB7zTSaI4fIq5I(eM5f*=^8F4L&v!L=wbEv{15gA~Sa;ywM97L`V^%kre!EuQM
ze5VWQmJ2v#BpP7Y2}BV(Im9<6F^Ye!I5G_^!j}^Q>c!+boD*sQ)$h6Rp%7*jX&HWs
zfrlbRC%x#L>bC{5m%H7U0a-gz>a#1A`ns2o+|KJcx6|3}Y=J^s3$}kvvxmW}v(?Kz
z?8{h{fiGxF&{>u^s;N;|PTeiirRyWUeIi5o1MTVBCgev<$ZuvsafI{1gnX{w`dq*D
z8N;2IL+*~ek9aDB$T(!f)|SK67i5`?UD?64LA~*33E&3xi*rLz!3N*MreuzD5=Tzc
z?RJEyO>>J=oNzP*$|!$V88!gG2u&OcAss116GSd?MIvFoL=iwNTO0`Qg>Z+v_e+I9
z?62^126ji8%8}*KMIUqh9pFCFZC472!b9P!bgcw2p{p<8z8ARGWxjk}kEKcn9vdf+
zH}cr3A3Qb=LMQKxbI^TD=awxH2uvb`Bj1H{6z~l2i$aQIu7ZD$AJ&PGTfxDhsfYn4
zr5;Upuj?#9sj%@H1!V9&E*bcq3~a;t?Nrt~I3~lI1tbrcNKpo07<WZ>dFUvhnH3C=
zxljkrLy!fuK<XfxwiiJP;rFE?R4uB~6wAl#U|Frz3%(j>>z~em&i9youW;52C$|6J
zg}6AUJMg`BL*IX!qW9kS_64@MAh2)k&qCvRz`j)%@WKw>1-d5;V$F!3S~7qDu>hq1
zu=-|Bz#JH%3YZW)pe|v4pkWRK5M&lGtHiBmcsOK29%Ozb->Cxax8Q_v*+F%gqoo43
zlg>Hr&&1JE+(1Dth2L~D8Lm)!<EU~%-%;d?_^9xH#7%#=S7xQps=fX6{+G9Ze!2d2
zb^p!VWy)?fHYV;8NsE)<6w(*2NKB(<tJhJ|K!Mi(=w$5LPCwmAxajoL-G%ErJ)L*@
zX0uz*?%!uupRZfceBFZP>lQR$x1jmD1<ltjXufVieSxq&EL29B-Z>6E<6O8U2Jb$S
zj}s)C&iQ|=PaJoFqtB>|aGb(}DIfqQwm51F+=xQ<kyFRk8KFwVqCSHpPPHt67*VHJ
z85t~SYZ?cD6d8wrJhUU%B)~8nk<WUSf*jI&S>m!r0e;UyAoa;{>*_KgdW-vdicGe^
zH>4u7USdCdoLuaMK+2Q#Q<}gAFmfK#(z-zsu|<E*a2&~31#%<lirb!XR|}~<5)MCt
z-$;wpNDTo-#lh;jEO90PE@N>LEr=QkTFx4e71X93DIW%lEY=)=3W#K@iyXj8M+oO}
z@*KZGnsc0cM$Js+$XyNi4S2P7Yy?L{NBmtv6`bRmDQuuKB~~PaafOmOPN<KB3L{9-
zk&0GgL*S^3oT-hfz_@!$Ic|qyi6el;*))=gL-#;bi|ext%K*QJe=Ol1*nNwO(V!IL
zKo3bqTq25t2Ppg6QFFQ=_bGOLuE%`K5_%^j;=8W@19c!&c$Z;}KNJBpmoc6J6PLx1
zKr5I1ggXp>XNE7JFwi)0fW2*z+hR|PZBIRH|Nmi!6dxibTe7XBJ;-usYWQ*Z%~O%X
z_dyP?4^M9opH4`F6;3M)!`rt(@eh(emE=LA2dk7YsD`&6!(X4J1jE-8UnC_WUzWc+
zzU#vyD#tSv&rm$7)Zx*R;t}K#<T;@EF8KofZqwL*DLrrac_)d21eUFMqT+FNJpBFk
z$Ae6OrSK+v4Y&TUm$!%i9(wrNTY;d5e||iK9|)B|2ZIA){P6SN;oHNXz0j#{oMPQb
zW3+B`qa-Vg24lRy@YARpI!dV(-+6Du*fr~kc(1%|-WcC;+>0O<Q$BuvAG(pRuXP(5
z1$x(iuLQEuN||kx@Y43Pi_VYNuWv~ht9priwu|dqCp;^=*9)J&4tii9`j~WL$1(Xd
z(U%#=<>~NfWMEDl(W3;J(aQi+hYy+1HzSPx)5*(==cXNN)f9X_;nU5o&YnInwb)oX
zDPbqzr|5Af)Y=N8afUjwZ5v|C=c-IE;T~RpI26kIUg9}kI4cO}djFEn-HzI8PP=OH
zc9eilkh2?60)b$M)z1zdD(X71Ghhjffy`=Ee~aj;dV)CLj;s&tCwIuS4$d=9PsERR
zRh1!+`;aG-mBJC&TPaT@oSz8^uy^ElC;bL~w-WZ?cPnv?-`x&AjNhA37Jj$0@8vgt
zOrzd0@|(=Wp}%CeI6B0z_m1B^%g_k2u4u7~4k3Cp%)2I<rTEmYa4wqWQ|<7qMtR=o
zOUB4ok}i1RWDh#wY(}OylOjM1t$YbECR2vvSh~98Xa_%`Na*UJUEb@}3(aD#L}4VX
zN?uCE)^b#B8b<5lQgcpP_?kK75|2ZFp8A~5=gQ0yIW<xS`pi-}zQg6za*O_A`gu8x
z(SarajZe>N%^N*tBi`uey<qZPCpnp9at*;nL^p9vhaBwUskQUOHI2xc#B+wuE^_fH
z08VnL9Cqte7oXq~#?!l|>BdAZbStUnO0QX}JSNuN6L*EI(OvZz&V($_Wu0?>v-Mp)
zUCM#MPEONYAEdwwZ{0>*fFuBCnuJs?1Qtf7okl#6vE=}D13orbiE1l8rv+a#d<&e~
zia9?UpHCK}Ut%Dg1<sD(*%3%~Fs@c2#}FtW7)QuJo~qd35mRL~3(LEun(QqEZ}PrW
zvlp^&s@Y4wKGp0b?2u}95?4}x&3*?Tm1=H7nN!Vf_HUPJlJ$JV{vA`zZiSUpv&Zvc
zsU{gGl={$2lh_uOR(qzJSu$}n_9c5MRklr8jGkHhEeS0lM*_qZ{A(>lQYcV;L<_N7
zX{CkO&EBDf*hySzA@+Lt-CBqwL7=|3h1f}X_ZDK0`i8jIO4uXrwG!8V;$F9d4~u)7
zP?osY&i?f+#GZJ_=|6hAUxAlRuz<T@g{PdJZjEo%CLq}a$z83P%N}NzyRkW;-Qa}w
zHk7jS<^+=DN~=&Iy*FbjJ|czQ1f~-VXHfI>dKRqH03t4Pp$CFo7K5Yhh1?j}gK30Q
z&PPC}v|4ln=U0=&1XojkHxLiZj`ap+S3NCn>|}pFkvy?BB54qh$=$V}m)k^O!Yj-U
z^OsY2<EU&vZx2d`7muTjjZ3>~<eNJ~fI@Tj28OBTRp|^V%>j^yL3l+gouTf2=_Mo>
zp_vn4IZ~QkDiQPoLG=>Pd1T_<T0L8T{|-WEsRhz)#N{~qgC(_pw3Hd^h2UCiolxA>
zV%-d%r_U^LW36AA34&ID#Hml<F#L#yGU;V<_NiqM9pc2h7h9|Cz4Ub4qrYgwpx9H`
z8Hyzh<-kQR6F+pu$U=(2{-=|gJ;I2%pfoAoV7eg1m9d4BH9<)Or!b;7#Hse#<iaOw
zmNLK;iE+0IpP)8>VCpj|K}vsB(;EM~VU!U4Tl++j9?y_!P%q~=^@ZVy=dyAetrK5>
zOAV^wDQN1Ly;sq2?RKZMBJPVO<zZn^o~fI}qL|d#Sd;N^6quG0IdszkbIol&RFM{%
zZ0|$K3*OYG?z)_ozs<->cRl%4<w(>7={fcBvO(}x;f@-AQX7HjyEP_;Ukzegc0T&N
zM%;ZdIeWq=!Iadi4=(GPURAT<l-a=*iax7^tkhPbUQQNn2|%6Xb0xHw0xk8k?<qng
zmr7`RiRX%tV4WUj>%T*5pN-TwwG|f|uSc__mw9APSlAOJHh~_oG$R-I1y&0(GcANt
zBCY_{I{3hUq}pa^D1#@)Ug{)+7;$s0T|Xl3W~6CaB|Ff}F~B)n#*<r4duz{gc~3P@
zxmdz5CdaGrFc+=PL@i55xR-wU`g*=X$PPSsZ0(!ZbWj^`lgg$kq%PDP&U0<;t}bwL
zx)6S!lvptxNC^h`gLX~AQOn&3^kq{~WcdV@`BgO~(WZU&dpu+x-)wU^!{d}8{#ReK
z&GThcTSIkq4afZd0f^3x+m~UCKNOeI=p+@FI~f8fmt>ego_`_!fBo0z-=Ab5Sc+id
z+xTz(^#1wuzo$k%pfEOEtZ_m~b@}%nPw@+_jIb_T07Ovw^5=gpzd!wBUbXHu{ZGR3
zii9PtHD(YNSYeF6Sc?LktHO#58YQJ(=sz)|<7X0rA%SROQb*=Pf1h5ae2!z@GxM9|
z-BeuV9D*<*G=Ex1;QAVANNW~O*0V7Pk9IZtF`V3@89<S-sd-mjRySlyprR?IHGTX9
zFXcPTue9TmDbucwFi>ab@kO3syay7frk6*9mp`L;S5JTbc2iCeoUnQ=<!atbJOKG~
zq(Qqv*+(|{MCp6eM%8IuolcB!boi##&1$t>siL=Js(<TMTwJ9|)J9FdLVE9uOkI_-
zD&vdgRH@#gV_R0*0%`a0CPh?SIWez|2O27uUD>o#qM{yklTDO{?^c!U#}pYXh2S@m
zgZ0(^=!se4lTf7z<*6;Qu0)sl#?AASz2^Bv`fc8lnNBnnX-CJZrQ>Hhu59_Lq><+&
zjgLal8-E|sEFkZBE&GO|<FZQqKJkG?vbNBq`-w9@*s05Nm!g(>E`9#2eU$*SvxS|~
zOiz;2zD0Q>0PIS8_ni+SMFcl5qC-`C6BkzNe6<+klT6F5Oi5B~@nK^f<qC04;kJqz
z+7T>_7m;Y&6$w_KHm#~=T$z_XX<mJPG@5E&S$~b~8_OvD;8dPJKv>;#b1zlqco(I*
zQd*^V?Q0D5O72VAOzfWLgvPe?ZUPF2-3YN8dfhK^vAilB)t%07^0%n+DQ#l#5kzV3
zbCVeMBQ`}MpzUWe?mG)W%pjV1mFigvRgA49a-CyGGU85q+%?2cc&FLs`P^*TlM}{J
z_J6Otz!ceBqsb(7?0OhNB!Qg6B#&>#K}KrfVXaG>{dK0^OkTgpD1X{CP_Kk@zOAOJ
z#nD{jCmdBJmPR0xH^T*G){^wRZ<|IK!oxEK_u3{1lIpxl<G=Hu;-|o=cRd>MSQQ;j
z9vLA7cll-hEaSAt3MVn!mLCtpmZ$?|5q~RwXKZ=IqtpYoyd~Y35-EU{bgTF2+z)mW
z8XB+eu01^y{e9NHT(Unj+P-AYO5|VJu%)~NkYiZpAL$G@o-f!UfsdM$p0bBDGWmvT
z^KtYW_b<Mj#<+J{v!KoFTRh89%2?m2VPlrjA69$PIc?5%L{rg|wV<&Co~5usLVvcU
z&2M(>smaiQFj|dXrO;@G5Z7&nlG>f(fB{N|gCD#jhXowa6iGzO1-v_lXgf->17-P{
zXrAes&qviWETg^E5Z`>UFFD#kbxtCFh435B%UV*Dn+SjVu+?ZQ3p^1cXP>@n#&BoV
zgB*--1Zk(Lhteo08N09SA%&FD=zroLdY3G4c=EH9MuOZcjgY8_v>^m^t`o<j;oT^-
zkO@n#w^p0eY<V3W3Gr^6p=2}&E5qJQ4GiYU=1MkKGhUEYvJ<5l63F<dELbcqW1ZUz
zYDz|e8gli8<zJ4i_Qjph62^eGQ|K99M^E>y2!<jE9@74RofP3DbxQ6jwtxE$;>VnG
ze;0{#p6yq3vA=urH9G`g0O5l>qoOaci4^z)`md<%i)^}>hldf@l}(T1c6N|DI-~k0
z`KaN^&t6+o!3*zk-WB9jXg>COw_F}jtRmH|_+oj1*^0IMwdGh7l<l~d^471FhLM&8
z4ZNnmT>`AIN0w|FTS2f}$bWv=_7rhY%{!j+y8CezQ7CQal}~|BR)@e<i8DXxV7rZ3
z$9?(+UdA+b8)(!u#?)5bQ7`H9YyG0^T|t~I&~sr@T={^uYfI55?zf;f<(C4y<MoZ2
z^qcAz7ABVt#}9o{1H^+|8pvjWKy9Z5u?$ZGnk9QC_3OT6Y6@1w)qk|Yw)cw#S9LSc
z?AvLq3KB$^_EE4)9Y`-wt%%!n!2@wkWcx%BY;lP95@l7&OQ%^U)V$lJAa5z%N<c-l
z*^u7K4#Po?d`ngb(6aP+sytjjTD5$c$5iuY<Cw~exjkEFqDkrPac#flkV9=<swGVv
z%6-GpP7zS`#|FB;N`KHo5z4!7ZxS>C54+OI!sV8er1pu`3z{Z2kE}O>R^dIqqz|v~
zs#%Aj>v{JWQ+gE$FHftQ_uL(^a=!~i;mJ}OG1FA*9=QxW1PA{@bV9HCneMVC`zTM5
zAwrf=^kDHCR8RG8X53YJWJ}KW4zqhzGQ6w}9kr{`Wto~T*MC!uF)#$2yx|+=ZV)v0
z{>JN|O<JmsnccWL0aIDU=51<<>;nxzO6v%7O=QRF-a#i2bRZz%=9#%lk7rqLMJ6>o
zY3vVUda-FAH}5fw<c!UVYTQ>Y(rJOPnw6#Q?w2|SG8QmfQpYIy<36lbKmbv1NI%a?
z2>Z5?5(osTzkiT993BH@M1T&JPt?3oQs~Bc=AE=X+vHU#E1`Gr(t97N;&=d!K?((a
zrXXsW1h(DLXYjv5(%QHh+IAhZa(ap%EH?_4($%u<iRSk)iu@i@DJLz-u-4KWYr%O|
z&$VNPa4(}gNTo0o1a(u!ym@L%TT(H^u$E1gyIIGQZ-0@V_F51=n0c7{4DqU$HXGf$
z^7*zz+l7piLNyeQoz)vc-2@U6Dedj$nN=^D^@WFqfa6Bmvz2AcNFF9;72+5@)4OY=
z+coAR`GI!An3ZHk_dXjXYe{hKSp>Ufls%$7@<^knaBOBCzId_f_A!oklKi~eNie3o
z71sCWaDR9;tzFI9rYSEkwTRl3sB{-Jc38PVBfQfDcPSUI2jeQu=IybV+G<K|hr#>;
z;Hmulf%fl&GG-=cR%rMZ!GV`TQZybnPqiG+0lb?VoEy!&5%rapCu`^(;I-J&w=*GM
zs$B0QD~w&_X}{V~nk}Z9SL{4nJZoRW=)u4sA%Ewsl3Dp?r|ebuddMwJ&_beGQTo@|
z9!j0+z|-zZc1)$@zR@TrK-m@j2Ne}^Pd1o&)?F9KV8Ae*BgBJ@K#QQ=ofK=SgoYKL
zthVwHyd#6q=ExXg4bkWl^~$kCx6z@zRqP>Fw4Zl*;Pnt~>9C&Qqo#6mMc=Tas=KE@
z!hdG5U1;DN`D$(jozU{;wHNqcX^Q<NUO)mJE0}kBATJ=TFz63?+2|aY)+Rc~3&j;6
z%>0eKSVEQIURK(@nUndHA|U3&H?k4EVo>IY5B$MY0VTUp=K-ao)FtAL@8g3I4d9_Z
z(6Z*&GI@WN)u$+2ZvokpE-u$xCkjew>wjTx&P8zT?&kArc4%S>6;{XT2sZ&x;&mFA
z52Huiq<SZ(CaAnxu-h&ifB7<AiGG+fz!*dIiif`lQ2_<lc#&z`z5h;V1o4}{8Jf-k
zzwg>^>^4cNnkMmEv2y>HG3(1*B$Ia<kxTM^?EXKbj-%2|o>?ak`t{jSYFe#YNoYDb
zZvK@~Nv&?$Z7U<l`NlS%yU_|hO3uAcjW5<2HrL*Q8SV$zQ>9=>nlQJs>C8a++0AfA
zz_BMa;p5lX$3Fj9e`(h!6+?1{H<tzje&da%p{T!eS(QZkW4!!Q{=a`d{U4R=(iN9s
zj6W0sG?$T%ITM$Eqd-}ISr*O5yl1}AuI&EoWcSnV__q6aCe>NtG+6B3zI4DpNd5%L
zgGP52Kp2AV?Q8ezUP?549r;pH67gmEyW_h)d?0W<19%4TD8S)^Rlp;}BgAu1^Ih_V
z{N1Lpb6RhBy^~6X9e8ixiQvQc!*6dt?_?ra3U9*KaBH7l-*$h0?F#uwLL1*{g+f#J
z``29<SD=Nlz5}vCIn{mpqx-V^rQNlhWh{>Ybs(hdzJ&?Z$mb+L2P{BojMkL^(Fm<w
zXN(sVeo6vl##jzgvk(*uK{4~{TrpX8&?qER+3X;Xadw%dm{Y4eqn#jTAfIzi3c5Ip
z2hw>0HM65-{}CB~tVTW$w2(FedH3cN$!aqa_ftIa#m^e}uKM4Vo8~ifpjazp2R`{a
zJw|P_<3O$JDbq}?<})-Us%@ANj8?PJD$~;*c?mq|vflLb<9f5q6&azd8*^2h;FLI1
zfrb1LIjcrSoUwNa?Wx_Boo$i^ql8noogExbt6mD3zu?G!L4PdmB!LI0MD{bkl14=K
zD19L(d<pVST2Z~Eqr`Pjt2(A}*otQ71fp3>wMUL17et>Ek*2?%LMX9OTW41iN+F0S
zyrA?nI6xMw_JcXx>v-OY5V0DYXTJj{&3?||aM9u5Mu$T!H2AF@uHN;U!_~9t(@4tL
zbF+=na+W54k1MIekjs_1pbipMkivD)%ge~^PRTahmXtdZ-ojeSvvV#`PJ?VC#Z#0&
zm`)nYx+R??DuRXU6l(8FC!n~n*)m7g8(wqOZJCiIc+r;JG&^pV@s%I;$b3ih%I3Gs
zspfoP6m1-onq991rDirw^4pj8f!lbUhLC_)B*0pK3u1gwfvuoc+2d0+=QwC`PTN>>
z^{&@6SI?$N|LbWA+>#%(890q{(MdKXh(E5XMlzd)!$&n{rLd^C)>yskHI3D?=QQ@t
zzSaop@z{T(=461;iNp-Cs}Uyp<Jrqee4Vwey|AVvFf}S-`%F}wIE&V^-s;^R5pJ_n
z@s22e50lU_rotu5Nv3-#7^kf)^%k_^a$64gHe2^}JA)NNj)$2-XF@Z>bE1sAO#bq@
zNb9mWl)EcH0%U0pMc9MM0(6SWGnhZMmy{<Vv2YOlw3MC4-e<FXrXWkvcB(l5`nFP*
zr|SJ=To2{|7^Vi0sodapC19GV#z|yku3v_KUJ1*!Va?FXIrO!_2G;BWbZBpl{VGS1
zTiaTSjSSkZ7NYQQt*O=AS2VDcK|!_{a%}Wu5m$+2oQ`8?(5b}`pR@VjCUbs%?QwKS
zScK{B`+Mj{8hJPNXgCAj5kAepeIqO@zFC2Z|1rne9K-VzKe0y;_<c?MC;kc_Q)-)k
z?b2p@U|ktofqR*(Ea!YVh&@&2sbQ;BHqnW`WSSEhbjcTK0F6L$zsO%DZUtq&lpRwX
zP!8yrz4J78ew^vcQ<Zs=%S}fiLeG)bHq5i1E%_-CTQxRoa(^Z;K4SquS~$xVAd&q%
zTfkAy1Zbs&a<-Wr?L|WoHzx1-RHxJdwHARcf2P$+NkdA;82M82?oaGhwJ20HjI_D_
zZYT#>LvN5?e-SomcgB8`RLL3P9j%mVJ)5J!!P@<`lDEj9=z3WW5(E%jp0#{VPEy&0
z^l>?B7%il056S-1S4>T4Iki4yw^BWEHNkAR)I`-^VOu5BDA_-WAST*u68<)mupmlp
zf7)8aY+;fuWW1QG=|v@g=4uX_6D7z&*dP}L2ZnP#7f|&EWR_3m;u{S#i*8qafHxc|
zAeQaNLkYk!ce-SWLx^6q`B1XzmZXhMo;YP0@9*Y2mbo(#5AQL1CB1O&Lj7<XQQ6Az
ziq=!Y4o!UYbs=%?GcIdkU+-VaBsGhwf6}T_l)WBE?fVphI*N*h-bkv;^BSM^$|FYZ
zhw6KnGs4eH^flFZXcVI@B_Yo5-{MVemLrwn;A}6Xn3B${<#Dj}(zL7OdHF2%Qx%-M
zIcqtlkktYh-&~%R<+m;k5f`3XUhH$l%9JFIEf+MG(mfUABTX(=i<Na{y+uvte?J2l
z*@dd?F~3f5qpH%@@2|5*aNH*aV;rA*3n|By(-X~lU3oO>3!bv29U%T?LV;h~Lt~~{
zXrc$8F_*SNY5yZa<JqpoAr~I0aWU&2z)1jI!{f8iGKLl&;djEzXpC^ue;2%bwSDvy
z{LCkEf#_RK7w~h?w`JJXthkDwe{bOLA^eQQ31xU~^dAGz4fpF1bhGFN2pwKpgK{gu
zUOoGt1fd(spA4bFgW$-vQmXar=R)X<vz`H=n?v3iLO0aB6NCmXLXE>aF?79pEr@Pb
z-GQQ;9ah5VX4MaYqgzg%1ElZf{AozKwSnhCY0w}ny8of{Lr}Vr;z}spWUN{VrQ5UC
zLFvo0?t{{;y{v=M&8i<7N&|8HvGi|&(tyC#yuJZSr@tAP*byWAU`yaEcUxh$oV)UW
zOE^LFUT?E$MRAGaI2<F~uqCX6?yMEb!M$R2Epl(rG^6#u0e;Cl-<M&GKNJErFqbhc
zHx!pxt3WD$j^s8Bz4ur62U;Uiq!wUMK(7PjmY{DzPC+K8Ib{C-5Nfq7FI`n#ogCaQ
z%Q8uE(<9Y-`g_vTx4Xx;yDzVXCZ{}u!|Cn&1oA@5U(hmWd2#^C8kpXGOuv8C8m(N9
za*@`?$mQg>mph;C4S1OWG6Q53;B@Z@WW>mbkvYhJa@TU9{B~vRwR~^od#{Ni1)&Wx
z5rSNN`s3}FJ6#0UD%kil-o~ewx4VDtn)qlG+b~&n3N%cA{<w=T2v}JcCPWp0rk{UJ
z-|v1Msa9UYJt@nBC`((*yigWe#q1|*jWW(vWyOV(ESjF>l0}jzf@R_KB=?ZVqC#2b
zF_d?IRNg~=&+<LLoRS{pwbfZ4n-cHZvMK|evU0Dsq-ZlIzv!_DDAr(88CQ#P5j%;}
zmeJnEH`vcj;_E``H9u20B~JWQd8WlCzV+MmL}lr2{x&^7G!B@k^-qT?Zq!4TK2hsu
zRpr|@Gx4lX(M#*xCQ*I%dBp(Pvy(J#)}_RMK(>}JHeSqS*m^5>8#GQoqX9SH{PwN7
z+P*Dykoc^o<myGmW8wm3>3!&Yc@27-AS$oJ#GaHk%{k(dk^-qvd@8=lMGa}`iNmg_
zC55oH3QY?NU5FO@#~_KG5})y@3xv9qB1B$N(e=V6DL8h`^6nb8VsIwe!muoLJHm{A
zSVenWF7@HnR%bWv1Wd{Q=Ll=wabZp)&x))abDFSreDjpXh7~)%aI}qFV0Fq+S1Efd
z;SJf2M5y*B#m!<w;-$)s6jU&FlpA6+J}&~IePL0eE{tDbk-cAW^~3mV!ZPEeHF4hM
zS}2xGRsyBzmc)=^D@IZlLa$q@o{c$whbC>t@t2D=0Zy_#z{mxjSDJ`fM?Tk)26Cpl
z82PQGkF}q8s{ltP+sYa8(YE7Z1RP{n%Bc#tUR0Do2`fqcGVLVr5zdp8KoPo;VsYA)
zvBAjZTPLNRHY(*|kzX#duO;Ql4TNuPUT5OgG%)#qc_Db^y%`7&0l(6nM&)#W1EDFj
z%mS)ht-K}W&(hwK%PrJ5Xk<*7NjTP7%p~(-<Vfa{_$h#Lgw0`FIy(a7X97`TD_Dbi
zASy!#+Y7O)5H9PgjPQ_rd$?F;ypLfD?-_d5Z%sHk_x1SunmaP=D<D^34d^5%;@24t
zHrV4Um&CW<5Xa4?%S;*3obG*pE~@;zVRW=IbUi|4WCWrBk2Yw9+0z!px&-zf0wu;E
zxzdf?(?<KWjQ#2>B)D|mmK&dH*IKr?PlekgQGxCPxIKbk(y~tm{-01VhC;^Y%bbhs
zx2=dFD3a{CT-0abT_YZAd>HXAvvL7$4tR|d7&zy>RLJ@csx~cE(7O+RiDT?vsrhRl
z?;!aZkazfE&H<Op;9^aJD;Z#$zYEGq7TDlFme5!m2rn(fMs#3gFK6EhOoq$R0Ev^E
zk<hKdfTTPl1@^#Y0X}2L%bf=V>f%{&6wX`e#qK6Lm;9rzFnIy$qXc@}<fs+!O5RrW
zJw!P!m%%tCAZw%wW?P0P0@#PyZS5ZiO@<_}EPo1UGLmcqyB;)&2@Vnv0uG=lj~s!f
z?i-gMus|pWo712-{@B)+POv~6f6I#-W-nWf9sm-GbrZqL8y-4u^WEy~E<}6J@5rGV
z7rLxfZm}FMRR-GsU^#5OB3>!3>`-H{VPFY)SR`rJTGw4@B+%C3{MN*2sJ%7WsWq+F
zE9jc(ebAJKbP3iPU}Y{8Yq&+DzN?LEfOnKijg#B({_N_btsLfLa_a^>f7H4~^5{Vt
zL?<af%-g5W^;TM0f9T*3!?;o7AlP!Qg1SQqMAN0uM6++&^hq<#+_GwU<9Z<6MtiR@
zgI8^M5$4&{B<n1fW8-~E>d{7ds!>*(jlQ%0f?~IcWIxck7BvA%tCVN%ICx;6&S!n#
z?Q#eU5V5qf<xQH=XXcjJH+#oX0{wb?K2m*9&?jNI9>N`ay7x+8(#U1ZDWt=YQwa1s
zIVFE{ZU)KpXuCaL`P;l{7T#_?m!Gjf8h>RguYDlj=A662j?a1xH}m)AH%4kMx8K2m
zPU_;%Nh`EIg|1cSJARHDQr$RkUSuLa?{|wsReu6odSkUA&~|}hr}ACs_n4id;={~U
zK-awUZe+@CPbzuM&Ux*sHieXRMv97|O97G+D0ZKY+7xHBB~Wmm?#>UM=hjzE3V#Ug
z9PY|y9@Ze|c?03zP*~63%kS>QDp!)51v+X}pk4nzB(uAeFJCvmga8W2@4EPNcV>vk
zl)TiRKHMf)_PN+F9Ca_4*v{{9&VQyBVqL>;RMH+pmhA3umt(&NKdqNiB}y^$G;~*U
z9$T~({aP^NmRnN>Mf3JF3SRs<wtrj<n7IZQfnT;~A}cd@PLDl*ovK?SZj}VK8&2I(
zN}-9m(zrUVj>?&?^IQLOCf%PG{{XaUd>}=}_TY8TO|+4O?LOS>-v#@p)*SGc9`WxA
z9Z)g&kIK9cr<!YTHpRAOl5&+f+~UxrW|r20Nv8l=rZDL6ol~LFJ*-wN6%dlsYRfmy
z>y*nD_b;uh8dR5Ij6W0sHJ5>6855Uovp~gve4zI5=j(5GGBYfNH{mv1)=w|5cmLka
z>?<wE_(@9;sA>A^>s@$(K*CrawoZ_nzWqIYx%+)hwFbulgCmV0YQQm4kk^wjUg+>D
zfinyD6vR&g5(|p_jOTQnpTj$o=kYo!b6I$#F#jQ7hc4kqI=v!F%i_4qV>XRP@~5PK
zqp&7i=_TSfGj6Rg0FxmnG)r3;XA1J%pNW<Mw6p%h9+37Nt<XlGb?*j6rJXQa4q^nP
z2#<DQCfvuRr%W!4J#HB+T2mwld|{9Y4rc<g70r_PytT=gtE`Wggy0?_sfCfM_9Cj}
zBePEh*_bU&<dra}_ZU}35F0i*_F9L3^-Or_&m=FbnPtT1WQs*I!M+rGvXK#dnfNY%
zvyt)}6(o}7ZmsX45HgE_7YiyoP^KqOD{P(;+Iq8Z!b|yluadK5QZaBWP!{GND`N|m
zqLQZ-*NMa*^9Cxb_$j+yFhW^xxZO=YjX~BNOwk<#`E&qO%>3HSt6#!Hp_rn7$O0^N
zK$K5?e);O0uU&Xn(<!^APS15fh^8EN&D>!2)#%QCap3`zJ$DX_hkP(%RHE<}jsc72
zdSJ1##$v&df2=Iq?f|@)->E$H_EnFWu4_FajOk*q*>egeVO_|R!SFIZC3mfHLXSd%
zx}`#spC*M~0H>$e81A}kB$8Kuo7SFP&0IF7)eo}hw2*{1g{o2REZ<1KZDi|*wHVr3
z-}rP!n26oMvcl_Pg>AqZYy|u1=li?_`%5e@fp*D<``W^WrBSgs0|NBHyt46zL0B%e
zK{PINA%q|D%Alq}Y+=HJYnmcKG43dZ>#tBwDDnrRQJ3$_5$Yl~gZEc|VKIy&V1>S^
zYea5S_8thN4u;@46vPs3_Oeo?#i`FrJ=w)@*YhIDQPhw|XoQ<*%(>i5@?=@U*H>KP
ztAf`h-s6V^ELCcJuX>T?K+SX$IiO-7Ga3QMib22Ve1Nd_B(zhvuStq2%y3+)NlNU!
z%)%zrj}z(9ESk7Hlt_<%f)Pn>E$%($wv+`2a@k1X4h5>4Mw`|Y=MU^XYitv3?Hg?g
zb}R)~QQ~L{u9tkcKe9{hDF=6HmK@xrQqR#{>Lu5e6WS<2(!3|NCT-Ec#*Y{Ky9rVg
z2txZmF0^;!!<@+9mXUh#VK3`yl!Bc5p+dx=Ku|M&)OqS9S2_=WIAQD$Uy6TMuN{{8
zS~9+{93d2&IG3}=<nP@h;-C?!9S>_V&hXWEsitpXV)xDGSy|fy_k3clY!gAKNLOsO
z;%QdDmwk##ILQ=|gjqht0FGluCY$qs=iDSv!A9jxznR-(;w7&1PKpP{5w;Da`vulg
z?)BvDH$^nd84+!N<b6Dc<^|Yzkp2J^dKmS=AwbWBF83~+WDaoMx&R$Aq2BLZtgm~$
zjO+*CIW*1XHFaAO<qNeO#5do>=`*E#wZ2IwJ%wz07&-WZqZk;=nYzy?BTt+EL|1G^
zK37Q_S)F=K2v|&b?q_-8Vodw>iX~=V&w}v+I54bvfd2k}R)=`95j`tt17v=_l!OP_
z6)du~FJD~d%>cKAjsU#c`UU{6c(?#?JNpvAItQ*^SFE^~(Zbceo9iXf#NKPyl=jff
z@V5)z1!ihztyvg=X&6Mbz5Ul!fGV}f%l2@sSR-*ZTUuIIX0WTI^qjRVu6KmzG0@)e
zOub$af?PR&NSF&oO3m-m+Il08F{eA&@YvUcLN7d50B_~*rQ6g51mP9k!mf(jwG@X=
zni|ck+1H(;Y(J7hl$7Tf=b-JbhT@HRmVt5JFN)HPe4dN2;lB!i4k^0nxW|yJ=e&BT
zqIVDrt`)VG{oPJ9sKB;sPPMCAg>rr_o!IYXzS;_1R{ENoq4|P9W!JM)ut@yR4O^8z
z3VOZ7r!gwh3zqvwL8ay95-RQNp5z;JD&%^o<dDlzP5Mp7ibhW|nSYe08QJrNoC_U`
zx%^8JmdeEL{j1NQu;s$}Kb#?GqL*QeKNJBqmqErh6qhc#K*|Bqm-)ItR)3Jr_V{Wg
zg_Dl2S~Q^(-&RYAt31n?ExL?P(zmdDt2&nX8Kj>^i)C)4>odLQ^!M^zRT}gJU*c$8
zb4tD{TjA-_-*NTAp~{==+<sg6`Ieu#D*Lp2YW-BG<lX6>%eqTfIQXoCw}p1J%iyb4
zllSOihe#7cjG9OTL#CmMuYYDvI%zA#)}rl2jxu8Gs$q(_B~}Cx0)LSLjP;g2ZIJr%
z2Sw2Q$*<qHJ-*0}C=`Y&d(oNCwy|~BL!{<BVu_O1w)$@IootVo6BbR!eA0L!77ihM
zK;ExXDSs3_;oDUecPK;IfXmcrI!J6IX+3qiL~@Lay{H3CgEFxFK7YU2Vrg(!Wn!8$
zui0gMW4d6Y{^G}X_U+q`hyM*eWhf&V((b>1e~1rI8DU+xXf1-m%OC%_{Pyq<PBj~7
z_!Lq`4~f6%uf<)ukzez95hU$fT5C-2DI|&U7ZbGbSiOAFo|!~)%Ezt8oDraMvq)SV
z=eBUDYVD|J4<F?w4S!N{W`yqZmmx{1r|K_KVBALewd}4NX?CZb5?T(rRFqrP-=<Nf
zJ&54cu#IxszTPfFu)>jxa~af>KQR))hZ+AR&*eyt*)g8@OL%@;ZFIu&TYd*fkpM9G
zov)d)_pw<hY@XaD%8aG3@@<uoVtK2|Sab76l~MC~H#br+3V&Yosx8bXah|2y6KqLQ
zoELN0X6>;wH&h0B|Md)8)-Q>d<X2TXtr}Y-&A1^>OPZdZng+aZXxhL`RYdk2Z9JW+
z<h}Ic!9en!VRY||gCIOvmvib)m+hlP!?yNOR~;>i(6=Z#&o@a`sCki)vP)MdWR8P&
zSti^#xf~6KL4V^-&GAthhkycNZu!$|ye+R#SB>AHT}yQkvE_UIi8UMTdFeauui0pp
zK<~e(W;<tQZsUCBz9+DT>brQ-hUvFVj#LUts>XUXwxytKilLk=sa08?-CGf&1);V7
zh>CEj_BN`fT<xI^bV%P2!8g|-rP5=ZSRF!9;b^IXi+`#WL^sk=3_?)|F^c!O+sxOJ
z3*PN+Ghgc}NnU%=`K4Xp`;jl180Jo$7|OTnqm^pXZ|<ENH&_cJS|o9_R{5Fk05zQl
zcCfL0orS>3nhgO2%_zow&KeF40HJh&Gt{Eq{0zH7sm#lV$}T(<<Pv&4YoiX(A*wAE
zm-h$`$bS-V<)aN4$$+p%F;>w}bV#V%>L1HUbep%<dS!;rGopCeN}E*$kl}FH-)p0o
zjUs{E+BtQgm^;!kcuHv0DKItA#xy`FV|^;ajcVW|?k}}uAENcV&k!@R2^Z#L+RsBx
z)L79<fHwm&h4zW(ko^EKo(A<mZ$@E5A|H=R<bM=8+$%}-oV%cc`Mty^V{cJMAyYAZ
zMr`Yqqwt{)lPaTOci$2U-o%b=qe+($(~fM!lui4z^1y>BXW(GBh`5`@HJ56v>Z&`f
zTs@^8>oK*zZv-ban9-{H#Sfy0^=h|@_qKyIyRNadXvS>!<TTO>KX}f5b|Q5Nqeq*%
z;(t-LlR4=ucS1uH$siCa<ChgUpEbP)ffPh=oMiNlgZ(I{M9_mQ9c=6dFraDEJ8270
zeh)n0M|j{A#C?O70G04sb1j8KErC%~F<G{71Au^_1cvpcORE<rlcS~=2QRrD;x7<v
zx6h*&On}IFp4gHK)K*H*S1zd$vwoAbn16yVwS#d~)fYAEoHAbR)!NCa^#so=rs+7%
zD-zhAL-v=~cHWuSK6!Rn>$4*X%xJMAPjqT_kfgOGaoTl+uWFLj;7yK7+y=0o={LO)
zdff)^)_7ebx=Nw^Bzh-_H_dx)nzWZYVo+Nr9D2U56RAo&I#wj`l|K;>0UMi>_kX;^
ziWx_*$HZS35b1HRy>4{#{Ffk*l54NBpA3G0L?N~LB=7?i`JmLF0Ddq|9{_$Zik}F6
z?7P-B%Aq`M#d5*F1myr~^a1o}mVK0iB0>f;?BG2phjvI-c3%Q>Fy1F|4!){-!@Zye
zHO|I4PB;1mSdMCdJwi3Wo<lXrhJW4}&h-qep#mLPgS;PsYgkLpGuY*;fDJ<jwHu`Q
zWMo5I>e(lEZtg`kIKA7E4MyQlg>3NMeI2sFB>Bn6#$eS`$VRv76Oaw=Ip>fKzPEdj
z4Q*n^cQa^12fQ8Hh($P~c@o<Ihl4at1Z8P<zzrQ>xCLl#z7>g0E?IfjbAN21cngq0
zXF`1Q&UOaJIsqN14H|S3<AsV}23O>r`cz|o09*m_Ob>8}+$g<3DxtY88@M9rmBSTz
z)O0(-YK2KVomae;Z)PLM=6O9_F^vZBLc1Y@FHi=!Tc8YZ9LiWmrY8Ulpd2{nck7()
zlz<LSfDSC!qB?^mCUkaT34gGMoF-Xw)j$hNz>NSZ8YNt;vU*#c$TDcDUyUbyzjBEv
zuiA*}HQEAgX~YZ~Eq9l4tw&jR>R;4eQj6Tylt&eLe8(1R;C|(qoQj$<56pNb@Cj0U
zeb`+qs-uEboL>v1)VpafcUOf+-IFHoP;RQ}1fjuGe<vJWd*4nsw|~k(A&dCx39h+?
zSie{59YEL8SSs`a61}#!=Za0RgVy0@XntH@z}dDxtn1$@m*e|=Q-4<R?Y^3Id|S5x
z>nnp@P&3n7>+dqU;Y~jgT8!N->i0m4%2}$xzY1Es5Bpi~!%cjxou1<7;GGQZpN4up
zG<cy7R|8M`7iaFp4Sy9Sdyk)m8!9wH=`JANkGri<?^O30DcfVUiK5MJ>-HF1ai?-)
z`J-?)2b7<YD@@C~|8D?%d}RoO2^70q-Tyz>faK*{PK?UCkUT{daWG$l<T2js1tS>h
zz>EsmY+IR8BAxH{Hz-!b*rFfIXzZ<Do^A5Dmdi<nP4vYhP=A4vb9_(Cx0%$nTSijj
zZPcKkTh#&$<UOh-MeVya)3RAep3qEjzChVZ>Kjl@1t=VOe!uWWz%mwtKw9mI#LQxh
z-N7r3EdPgSq(8tb)7%rW5HH7h&4u|SxDwDU|KQ2gnQOGX%kyX@cq4G2b{?yY-8op`
zZqWcOS)@s$ntv5OH^cj(N~29YSoz*oq;gh)k~&Df_BO@VWyb>$Gn@i3wTyLxWo%FM
zaXMml4*9~{4%khptXp2hj#3?=!|_GEW9tr1%<cCkD5L`}MB`cL^xAUn-TaaCh{=?_
z;OIDrJ?Nn2@n5SN7@`){*GKPt14E;OO&Izd1Z~Jy(0>lvJ;W%0nR0|uk|wtR==}5~
z08N!E+R>x;5on+;J8)ZPNM{zNF1{aCo7+j9c!NEskz255s{Z2tG`5V+!6X3%8p{@*
ziD*g%*70>86wt<r0B$1GQ9x5BK*tx}8_^OX>cgdIBl=b|0g}=@-mNvt0^jo7{Y-#L
z3k8}h;w)q)a4O_XaP%tF#%2&Wn;ZX`C;&ASc5dyW<Bf(OKV6F;-n3Fv{C`a3)EJjx
zj6W0tH8huTjTjV{W_txEm$=11pMO9vfBxs!FE6qP{L{t5s<cWAW5dZh6eLrXHgYf^
zB~J9N^8e;DOJw30NrNVThfxyZXv!Q(pHWpB=eUrb%pQsK-61xKEImc9w9c*F(6l~B
zLu`GLcEXsmO2S2gk)cky&(zknr7;MN`d)3t@x;|VRa==&A{jXy(3U|O^nXijO=;G`
zW~>sDo-R*C9{5pK2GOeBFk0;#ZKoxhpD-jgVp%$rF<WP(1%>vhQpEUwquN;@7OJx%
z>r*@X{)$S!<&m~GQ_nN0Sl5-rwIvA}wYK8ACN=Ux*XB*VOgkDf=S-2PC?;{$Ai7y=
z_?|g{{!oSIB|lN@da;9uLVuC3s&-r@nX>j3r~{f+(2TatF@Dx$Tz8=qjdN<xbn7l7
zvRHwgHJ7;9?8}5Q(K@9-dU<X1=`Zqb%hBJq{hcyFIIZ}~mCoFgrMap*=0VrbjPhQ{
zU^qp2d7R8a2~@Bub9a8C&fET|dTR3j`)4k@h<5DC6q8LJ)io0+?0?21l1SP^8bq%J
zIKI4exRQZCOpzFZu+ltFkyVRj^-D$e-esw@!y2>J2x7|@Qck%Pv;yr(A=+4P>D>mY
zvbt8k`RxEcP*xy@=Ig5(d|7@DHKXjt@8KhwA*@n8KB{NG`Jn34uttI8`Ks~B^8Taf
z&g{nS%GY=$yqtZWtbdUt^VJ;})kPz$D-E5TKI**0?jjX=jph#PP;rLw(d}i9tj7hO
zTf8#+Xa_Xu)U!3dUu=y$noQF~sMs4_Gn-)J-o>|{*{9F1mp>QL!=ND%@bs@gUg8%B
zMpze4N{Ik?`s?4P-(P;^REu%&_k^d@zDcreGg_b(hD^yAC4XSV)3+;f(Bisi*G_@~
zD|Ao~lw&ipNDv5?b8L~;LvxeylX@HRQ{e!%Mpi$gg=X=-(Ych-sm(0kaJ-c{ZK0fA
zvL+SRk#AHAvMZl@M38weV{)ZAhle_H>9qe@C23ljC4cXZ=f}Dp5BQn@Z=1NvNEA5Z
zXEjx%+VG(-^nWXoGaU1_l~|@<SrK-T4zpGY?G=6PqwXQtE%0)$<$_jw4pTvzwzH@u
zjVNvC^8#OY;bSqnz`_T1qVg%$*ibU;o-QQZhRC-{zVA;86hedNvyAIf$BnR8Oar|o
z7u|c0YrE3y!+y_Aqj4fopk;2FoJ!Dk-%aniWhCm(oPW@wfDj(z4s(Jwx{DI~%Gyz$
z=06G13?qbOn)lR>P^IPq!jJeGkV>0D@yya{*^E|snQoWCL>pTusRmrmpYmx-Pp$AD
z&vdT-BtQO{$0*x)wGxh_q+3>@(Fy0>4Vol$DHK@}Hx4Xlm;>kX`H+E3yRmztCRdMw
zqW!SL;eYEBbnF7Bos;~1SZRs0Q7fdMg$fYJN;$()QrMGJ84#LIJd+eNe$_?)(c02j
z!N1DxOA`p2I8Ky@Avp&q7JTn;`@4m@Kqjj+7wbOF(}q&E<r&cp2!KH84eCW{g;r=J
ztdx{YJH8PR&Cjq|t4lb)rti9tSf1{IgRAM<Ie%*y=JHH_x&X^v8}gK8Tgz8>uh)-N
z%&d=`IoE8sB;KiNcgQyko{;C(9knGQ&9kr7M->R5_|BG}v*AIFZq+Ds@N{;4JG290
zi4{m$3o$Aq#xk;r7~>P%Q1YBBRlD6yMk^e8E^YGGex-(SFs$czkvvKT+}}nC^ClN7
zPJe#os4&wNU}z}-n#E>K`DB(YSty*JXH?yJ-i;vDIEb&)vr*l=vfWjIcY?+^mzqq@
z5ITW+Xf)@-GxSMJGOxZwE4K^&%i_5|rWr*$pxr{|J2j!njRyss>UXOknlPT!_#}9(
z2|;Oq|6O>^1Uvw**^CFkYc9h+yk6md1An|8p1vJQ$6^+k8?>gi5d_E{Z3yG4lM&p7
ztTkaG<J422^$P7LV{687zOQ2rZr6EPaxc;0j<hsnDbt;=>UsgDg1zJWkJcNof46PO
z$9T@6W><ZUue-0@uIBhFkUHKPm<6{r3LAk{2ExypQ`i8|q$uHxSp(3dr?s#$0DtKD
znRyl*ttaKV8{pY7@T@{O2%c>ao?-%9*(1PncB1JRM*~VX4Ub5+FXyKrX+^;oxh^V_
z&O5M!q$4!g#L@XidJi>k-9X=inpNzXV3PgcVCL8gV)(Wc#9X#xdEL(ra2dUh3r%dT
zGRRc@V7y!njZmUTv&748124mW5q~dh6sSCx`A$tJM^KiK9;14&-W{DD1+zs7uwEX!
zcD>u{0ax^}>IMu5GMg487NM_m%2)PTYoWvJl&?S&4FbuP^I6Zl3=Aaaf@5u&F;lZN
zGqe(otwyz9!|%6N4bW|FY^qMw7Kb(!ZkK`L`X00ftbLGS4TS^!SiN$)=zmk5ZHE?N
z_d<{u**1VW5K#9H(?(GeSY~`kbF2@Ok|Xqv&3+_WR8pm*!Z^@s=H7r*Mx33IyG!+@
zC~10U*+*&J%8<3%RZO<tHHX2F3wAZr4JG3Kj&^ImAZTH&|K@&;v-3X4d7t$3n7vCK
z09@9zcgi`2j`PT!u861%dw*do-F8pze0L*Vsip3|^xO@*^NJu>*I<szIrTF3HkH3O
zwM7<Yjnv{+bBY~hWT><5wg<}T=0eosGGn{rm1E?s94x!jYc@;XDcJA&m~nQy?Mb$Y
zU8lOw-R6(5UoBHLi*Zi&s}<k-G;+87CMtvdYSr7XKGx@vfwRKEdw)*!&{;BnYMi!g
z4z_KMX2M#Bie6+ZT5a3as^uEWPGIYPt~0r7cn)YQrQRx@9RyndF&5#CV2jUsKG<R$
zduy7{dfcs)v#y@z%`SG;<<n-bf94uwqP#OuJb$Qm`kg%lmw-Tpaa@Q4aLIc?Y0Eb~
z*)Vy&x-Ko5$(ddD4S&*U%4q^4l>8Db{1Ob&;!=Crq4n1l#8W|Zv4XB5!8_@oo#Kzq
zyS(!xnMtrx-y2fr<;-3CF_hLRuFEGLE~G3Klwl0nZz&$u@{NAYQig`z?QY#;<(qS}
zfR#R*EA?~oGKA>3G9LKjfqGOZiB~S%ssOD0Xw_Ribn)^2CVv*NN7+)dh?dX!34kI5
zx8QhK8|B2czRP+z9;y|TD3rHM#)Oko>o!Hp+|%p21Y5ErSv%)di&8N{DgVZzPMCjb
zl>3D1oQjjmoRw$=ausRBdn$#)kHUtKcIXFCee-KCUUwgz%d`APSL3CJU;FV?!8>!i
zZI-AwZx}C2h7=}NCPk4fsAZB_-#@=z{syyq)2f$Yj6W0tH8q#9O)3)sF_S^rCzmPE
zK%IYbdG^1(|MX-<;P&oPI&V^>U5I2@7LoGJ`o*XpjtNfgk-VW4)#zz&Y1&(nwa8rC
z%3cpe2X*$dnoEgi13fEz;vFKsO!E-ykdn-ES~>)4oXaceSewMsP>@+zA`kq2ouu_F
zu8-1g<^8Fw#&5qVIkM#5qF%ADMDt2@w$OieRM&`Oiw`0N$vgFLUfb@=Y`qnxmbzAw
z@v>9T+wg}yRA(w~MJ^OY<h}D{DAtYfMCE&42x4h%0^3O?betx($4#Yk7vpTwD)1K+
zYiU!l9)rQOQfyVlL1dfcTDfLGV|ipVMcGjjWuwSJvs~1i<2XI6wr7g&MSpX_@v?u(
zbb9SLo3*<IUe?bg$IF?uj(SDgov!U-d!7Ar@qpqB(%Xu*sxy|>5?;D1=V^?zdU3$i
z)+?=bEx#Afk}(I7LuOv;**6Df3GHF>*Oak_74gib!!&zm-%mYWr#-=2133Fm^?-5~
ztk{Oi9C2<b|AAP($_w+cEJv!2ku`sPAJ<Nkd~(fx{LcMb2vxh>3-e;Fx*rr-4&dH!
z_PoAF6b(-+fV9jK1ktUN2PFiE&Za$r*sN-P8aId4aWC6m+2kH=EbCXhN>G!19-8FP
zcU^TI=NYCo+X%7J)@keZb;;DPB+MrPO95GTm`+*+BsYUm1*h`<)PZ>Oq(Of&DeJ`+
zW=qv+D}EZ*gG)OOi9V#j?~wVF{>!L(`kP-iuoj*1fma4H-i5w$7!H$f@qdBdnvwWS
zCZv84+!zAimKccLy_as>!&lwwg!Vmz9``UnT5S@qC|$oS3hujeZJWn|GXay@<Lj~D
zY7HI?dcKC4B*7x+gEd$$hoXPN^7>JI`uzU%&yzNd_e@;E?)Sf+N;`o!A?6*T$w0e*
z|JeQc^s}B+(K0*o%IXI;1n-xA5DYK0^MRP%gs%fG6I4uIJgK_HileB4c=i6LI`E?9
zm)Xq_InJ`oN<jEjDcmW?L`)_G(j)F8Y6U$`nXyyaGvu?`60>3D&6IyuVb3U+P&oup
zQ4|5MjH1%>a<oHsCkj!^rvp%DC}*M=1p^#!)GW;Hp3%7x^NQSu1-6Q;0~c!2uVrX`
zM<Pp2Q#1sZg2d8S;zq%Jm-65T1v#XTAN8v_Tf?|P`!KT%8Ivc(={@XdgDa8?ihQiu
zSGAu!26ehu0!M^7bgzGt1mX9#=(o&InfK*U`=!T~jeEa_-l2oZa|^x|r-S%O@Yib#
zDVji}<$qu*=Fl7Taiy_k#?Un37kUwvGU#KOh;8lEC1=<OtDQ>vTkTIn86Zum{8#W1
zHV&gj(koZ}<~e%xh~wd`SB$4=ur<#8!7e#=^H1vyxH;pINZfz4?!}*`LPtB(6EKTh
z6Ehx9rtdutk|5X7O_Bv~d=&uy+{UzQqERCDiNyI4V-`Qjc{gF%s&#{iA@Rw#UhlOR
zW)8x0bOgmke7}T4n4@udt<kKt(xUFnz`MTUoJqA@rTBc+RwqO)n-HzO&L?p?mEBrm
zpZKD5?#J0Etn7d4N1Jb2?Qx`;^8-USk>;DpA2IpMq{$zIiV~aT?+UU<fukA3Q@-ZX
z%?u5QMUGz2+Vro}29Djy>CWE~-y|@^dA;7v2H&Jyt)fO?E6c%<?8rHAD{gZqzvT4R
zWOOSeyj)!ZvL+(x8J0fr5xG_ggeCD-g3JLWDu9td;s<|WKJOj93~G;KSfacQwNPr;
zezpE>RM@#}Jm^W}EUc^sXb@NYrLx$e4s$B<onxu;?O}|<)NC?TRVa*ee_+AE<x+6K
zI|>dimSo(0qk@C?^XCqS7o*f(s#=?_V>#qA*IVxIgNhB#CBu9AplXAolA6y>@I^Wu
zBmeE7m6Lx-tYjt0R+zViLZKdUcUgfcRIu=Jb<zHoT{^ZO2Do9DoH3Xf^?(@-ZWp1j
z1GiT6X?Eb49mKR+zvMWzx~%Kv^7-NQ8)rE$=XlrjF*vH?zWw4=*RY5Fn|Pc!@1oU;
z;^OBoIio>t(xapnA_$^xT9zDoz?PO@<>6XzX6=8&9ZAXaWtMC6>oefXfdMG2<i9g(
z56P^JcLt0xgJEY~gYg#ua8^l)?Xxt1pH+9R?1Ch&KNx_9yVx{b#2j?$r4y`m0<4P{
z1E0Dg!5Y)(OU#O0IziAHzSFpb%-8&C);X<@FPcW3ap8WGebOzWGlAw@>ADJ4Vbvid
zJ^Fvn3RP!o+eDhrl10#Kug!hJsTaKQ##$97HV8+0z30c&FrCi^o!6_Kt52OxAJj2^
zAoE4<P4@W6ssSexhSn_ZE0qn<ITODN)}|!6g0-m;-}Ex5?1E-ILS0x}T?Em98`D(;
zy`28R!g9>=@`zfI{sc0#jM?|uuZYgM{Q!T+rL?+Op~c0D-@I6%(}=l=DG|7Ft;-K6
zjQdpYgw@J|t(%uTOh+8SO6JF2@<0lP=5+Zy+i*9)TL{J_)N?=bw{V;gc!1SzE^wH5
zFQ_@oi3qyYP9Fqnk<%XxdbRHu)PQ|Dru0ZqD_jaWgg*w<IP2h5(pRp9t68^e;i`Z0
z{kYnSdA;=dC3SEY4aAjjn{{wDOaeXP<~sO83nEgDxPDj8W4)0->nFF%5XdBZt&h0C
z$!h^RrPXg;5di>$k?y#0P&1smad=#@wYYZ-T88^C8|r;C%5NMlT{wW@2(Xgj)(Z#X
z0POTEXR_H`!c+cVT{uKvQ-&J=O1%gfl>P&6h7YWlVT?Z%12i=-mr>6zBn&w*FHB`_
zXLM*FGB%e%LLMi7)mqVRTe%T^&sPxW%l5&%<P3*H7K;G8n`XO!7Y&kKplx6*Y)9EB
z>sWy-2Wh{)XLu#rj^tFalsJ8`J(A~=Gvx5_9P&m^DY@Vx{)$U%23J?I8N!Z;npGY1
zXhjFtW{HzLTG{2$9u;*t;n_%x?@E1O-xa6rSw?%R()`GOm8fQw;HtD|3)%{~nHd3*
zG|wFIyOd_45gFq#H<j2-JXVTWPgxv=IQnKsQCePEkwqEXEsj!bx#j9eQZ-Y9qg-0M
z8Xc9`I#zO2Q)^7wanR8>Ii|?g8!}-|tuf86Wn!i}&aw5T#j%bKXdOv32Sma(rZygE
ztV$aOw4OqL7#(0nZjaUE5~Xat70eE43kQ>vQXNfoh`GgEqf3=r#~6ZCwIvj*ldyCk
zn9?4r470X`FtierjidxPP)-lb#J7mO`C^*WGMtPV&n@1nPHJcqfJiX=qi=ON=G+<!
zE>r75LPAlr3PAr>AEN^b1VOakP&Fc>Lxfo^UWqV&m0M)difI`nx+tw+1f#5>4GA)E
z8e2gLlGSJh1T#3LR+vH+MJQywu{v$F1P7SXLKlpgY91SdZw<|Xf-8YrI22u>(E&AB
zEGsh^YlU0K2(=V#y&>6BgSOt_P+K8dN2S)*E}E>Xb_>igi$tzc8&pBrRF2K01hb4;
z;4vA0(<C@#86DJBEp6Ty?Sw#McNlz3{n#A738?I5+d50FL<MUeHio_?Zss`jHq@at
z#$wAb<hBgOswb$9%^-|W5<mF%+u_l3!;b(Qc{=>#pZ_u^U$7dvs1K-*^UJfdqlXVa
z=1N?APS;)C@WbqEHh=c^_;ljf9&p2x<HcfsGCy}Te*EkE#gESx$BT)J;}?%-=L>Z5
z*ya?$wO$@qBySbnKG+yg;_-v!6HyxY@bKaAWIj86Hd#C$o;-Ry{4#mJScd$|zu!*K
z^Y0S`biSCJFD@E{jjcbOT+A-#r<02%tl4)zPhL!qzn{H-ZfWL7C?`5X3v)bc>%NYE
z>3?^AK3lf64B^4^5n=J+sQ295-0=Nu{$et3O@^c4Z^KW+5BPv=ju6A?;&~+UV&xZ;
z7n^k<Ur}6<q7Onx4~A!#Ckt!vaC-j7@Lg~IbTOTs56^~QpZ;V&-@IBZ-d_A=7=E8F
zUR|E})7hKh_2rwl2QMb?hM$iYum1jjcJUL2Vg3^H;$NQ)Cug&h;a&6;d>k(3r^By{
z>DhEKom>oWW-lgZ@cGkWAwD=Ptbaz5*9t-SS|MZ<LdJ@XYxQ!&ZCP59JYnj#ab>vj
zcCHx2h(X-K6>G~YKwt~kQoUisXDIC()qJDQcPZ@>*GpUL-Aa3mbXnTD{dc*4t*<QV
z7I!OkOS+Z2rQIsssw+!-AKLrS-iP)+wD+OC5AA(u??cOm<#=N)ZC;yk{Lt1W)$B9C
zAyf9%2toiXV$)-M9r$SY&Y|^Z!;iCHX2T=*O^l+TDWQT6F}A~!HDcjho3gtEfw}Ou
zkBC+ulaY+*3jjh&@mgp-3b46<$S85_-GwP+Ak&5@wnrwQl22B?OJovJUs>)2nJ&+#
z?<Vt$<Fl8?XJ;qJr++k*(u2cTuORB~Lv1hu;sK`UgHgs>y<#zfP5V+dOgMOqW&-Od
zflzyiXb%$(s_BiLt)h^UJSzOwAgtCnQYo=WB~mH6dk_)e2i!Me?%Do-q6$Tfod+z1
zwIC{MM5zhr6~{#2GHDN$&)O<}6$6;I7jWxaV)#7EF9HK1Hu{(mgoz8qKo|>C6ey*&
zC@^~=L@c@-l!!ss%kDv-A>Vjs%3i^9If!rSO0wsyUnYnYiZ}YS3IbbzH}!dI5DH?I
zBDNp`BITSy-a`;M`C|HiZy5w+f!B2VNN9mXtWAAo$X*ciJ&Z(5$&VTo@L=MVfmIe0
zzK$CZIgmz8m^3W`GzEykd#DN2fJu)Y1x5zy;<g|(OUZ;egi3%Y=C}utpq$%+T!nz$
z#!Hdg2O?L1R481a;aLl0*dC%V)UM<4!5D>NxBkd<q`(H2B*xi)0iC(-VKx}MGmYmk
zEeP-)&y;Q-3IaM+0$M?1gF>N1We=-E#pfD(6k@3TvBE@ImV~D=DhXJjLVKtRHk>dX
zTfrBoRRx9{jJMhr@`b>v1h0me%OE5JK})?gh{(A_E!f2b^#GNEN$F8Efl#i*FBs}3
zxl`8VG1NR%T-wTil>`x|&n`EvDU2K1<CEpaQZw9;8>QS$Zq(~?qjX|lI<YUhle!j=
z=34oR!$#C-P4}#>GsT0$T*4;mx_xdOvri;Fw}ywL_nccjIBaBihs2@*hm~tT5>d}m
z?>R9c%vt*2K%~^I4n%IaEtfsWtWw&>m7{=)=51WD_Wf3W-Ph#p;PeZgaeOMyNZ>V`
zvGeA;;7kvp;Y{r~Q#;Poet=N>Aw^vsQrPMEE+~{SD=Jqglrf=TLibY8vy;~d4Ql$g
zQ;hZT^n5b!TAFrSw*Vs_$6KG4CM2N;{lAwq-yP4V$0ui#ix1KS4Mn15AEyRjRz9Z$
zQI<gY+RYAsqzHHq^n1<m%V_11@AMdqcO=N7FB^jdZ}&`~P$+Zh+RS7E)aqIgV7}0W
zrDB0qaIeby5QwWHBqpz*!pO}>tYAt|tGt1IMuiH|pjeEd77&G-yw9%JKF<5DdvsK<
z0=|*9axL6e$ri4O*q2KGHz(2(X&t&_xEplGd<ETqcAdBj5R(RP_vI^uB9n~w15hvL
zlgV!#rB)D}>rH}Tx2jtZjGJ!D8#+;FK-$VRiMPwj&s|@3PYJ!b{9q!Pba$S>>w3g)
zX}2gg;rv0dv0f{-xZ(tM+q|+mC!o-+*e&W7@5UGsWNW(GN^^pIO?*FFDFU0#f1ACu
zt@L7lF~?@Q7z>ut!6ENsD~()yDNySy06KL5W=|ZQpfEMnO$SOfCTw9r?X?lEQTA|l
zz?D67=}{PM3-XaDpi{ApBn?GG1>nNFUAV;RHCzcgp^W*+_cH02=gZT@rb#!VC_5_G
z9K7vwUAboUA@En0DA#;&n1b%3zU}G^GfQB9$aZ2(Sk4-atgt6tXGq{iDJGCL1rT7<
z3o-3fSrN2^>FOl|yICnjU)g}C$QW?SA8dhC{}&OXb$yG4339I6xZ0TzZ`CME^{}^b
zjn&v){;CCb+90q~yxYc^uCp{seH)Z+<!<$DmHW2JeOpao5BK*{+99wN0RY-jsXkPH
zdsPh;nKuAAoNMM<lokJ1(87J3hgbAYsYj6tHPpK`iqEpCrCNNpxA2xxWQa1`)NYMp
z&zo8X<Z-%!p3H!r^tC|`Wk3&YK9T*@0Z;1f?apFXwe|+4LL8F=0u!$zy<JQu`@yyV
zJxW#U-tH-tA$FjIfm+pGX2f2z;XML>yAsO@zjGng6eaEBhOh)=pH?BHfb7BVZV4e`
zaTdrMDS%BBf5;zP!q|Q1HZmx!YiEEL>QJ|F6)crX`Xz5-J{hwmuVeP|)y7Z8Y`JzQ
zYVU`x5V)gq>ZmMT>_`_o(#4MKD0U*MgzB~9Q7hPcL5S1i(^u2;-)C<Z(>K!(|MWA%
zgF{`ilK%qgTvwu(VT?Z%m$2bD6aq0dmq9`vCx7Kx$!;V!488YP)DKuAiKG^w(SY5y
z2gof0--4Wi%$(+s`Ts+xwMweRy}2DAhjO)55=Bzvdz2)HAA=me?w;OuUtR$RD;z1S
zhqv#8;t!JlD#?w82CI}XpoX`f!ykJowc+Q9ACwY+AIrZTzv=J*%5e|HJruVpb$GC-
zxPJw>1-OqUeoKBd|F*g9HIFwu-butDfn_W1s60Q9hd<wb+sUl36yAiN;nn~5^0xbD
zSJl@b$oRn+5YoEgub;c{0Tl^j{h*BSV1{3R58rpc_fn_Maf+QIjUj5Bqm3YPEZqwT
zpOSN`amQd>Qd}SpsegX`7`m2kUq_vhb$_<hOJzq+T`>0M1%kdav5%4cNhZJ9&&VoK
zcp(4J>T$R9e$M?u!&7PxebFAm(uGg>dbTARtWb8QoO%w=qBR<MVc6N-2>qDVQ=^(e
zZ8XW-N!6k!HypFBOAERh>Lpfy1cIN2I%1NWeURf}E&-*mTK5v#cnB+b=>o;YAb-kA
zESnLWZcVR4l&eB=+I^P(6eZ?#MaotNeO?ZL`-#VG=w+gm!$XV*zv%~Hu;93y=`dTV
zO+x@s*bv&Sl$-~%RZ~tVHCwfye|&=dw%8x79c^z`IO}HX*JJTq{I9GF9dykMGnL(2
zS@Kqp<JIS21TxO>+jyx{{w#j^_J4If&Wg=a3Rk^rvXv!2Vq`62Mh10qN>~ZgIE_Bb
z(E*cK)W_^cJIO#liJ2VN;KtbH^;N}zr74HJQprI<ObPkZ@M$|_Br@P&XV~MbDT|~1
zfncAVq{`CeIdp2P3w9>Q5Sk^9(OP(g+Z@xsPv9IP&GCX|?Jjd>0sHeg%73FUdZxt|
zvP%#OY(Ju+tI3Fz;|w3$_*>Qg8l)%241}6N+Kzc$eaE4LQg@h%689(i_mz8Vlhc54
zlgGk-47O2P9BHa-XfdYpM9#rbRxwdFNLOSRbjxyA&@F%Uoz1dJDd>5uI+bL>5eG&+
zVD!%xH=yPeH8}JARdKKh3V$Ry1C8rvq{o6%5vi7am#}|{Z%7(}4jS}reDhAFK*Dok
zcTzwlYeV*icP*`t37zBnxt7|SWq>4xLetL>ee4fo6_EV5UKe!H^MP!J&U0kI8GLp&
zM59qcd;Fkn02<4M`;B)2t<ddFK->2^N|+WgiN-If{{#2+j@c~#m485Oy8(Bt{t-Z!
z1%5ic`tKwxRsSDQ`e>e&@uHvoR(1dTHGjaFnek_r{8Dh@_fl3$e&oCbp#N3L58ex@
z=!&8rtdB|gIYs{}Pp(9@3CB*I&UdQk0g=f9?o`hM$vEq3J#l%hP9aTvKf?wLbA2Wp
z@87!-H>=gpKS4TO_Y`u1bYtCNIc_iG7zR~_x3TJ@16G$U=s+QVOM5M()4i=ZepmC|
z8{y>4Lt&SJ8uB4>1mNO-O^#>bxZ%eE4~g+KYu$iWP+nMzvsGz8Z5`3xkF0cRx9!IF
zLI+Vt^qD*6qnJiB(*WQVOvAerp}DU9B&KOg*v+V8gbZG36}}VAXr>vX)SYNXGb}pi
z*6Op1DJCR4_R=?hDTX7H$+a8tDHM}7DW~gRb}2X39r`<$D;P$HEp7O81BNMkl#-S>
zF(aB+d$t)j<Dq5wH#zxPL3ZFehWA9BH{RhOu9et_{h*JfAfp9onl0OZi-KIZhgtB;
zM=_9M1~R12V4yFr3~?K`*vg_%uDS4t;h8qS4}fle#zuU9Mf)-rlZrE+b-B`-cJ&!*
zC<c&s1g%jMh6DyJiigC<5iTyJM5Pt;@V#zQl(vEtE(gWQo?FA|?GX3;^6tv1SINLR
zAz7Vvl5)=gKE@A*UwS_YxR(RaJ|nc3rzJ#@kt47KKs4mo9<_w~gw&fR_ZsmYA{`JF
z&3vA@O0VgEVR{D2_Yjfebk2xmqt3h(H7LhAReqcq8&1qeJhj)ckyO|V^YPEv$5tOq
z?qkyTqHngQn}OPUlRiS65(dZZlOLTK-Q^a^yQo*Uh@rJ$XIu!|8~~1FJ-u>TC=|wX
zGTSdvlIEsNR9lOS?d0&5vh+J#lcNBA&PJwxGf4df3qIq7v6o?tKNJBqmw`qz6O&!X
zDVI=$GX$5T>OhnMr<XYEKpubNG4A~3;$U+}GD5q$A(M}oH##L&NT=_rJ;ddMeVUIZ
zjR7RK3n6$p1PMi&%20T0(p0=O$Axh|5>?ukR62tHB)N9(v}KqAnay=Iv(AL3j?LpG
zXaj7}w?B{`(+N>(rjepbV~~!J!keI`aQoJMdwPEO@1crTX+g#h20VWgq=$cgJ%k4c
zB#iZgQo;iczyBS6Jp9@3n(d&!M?;Q^l*TYSts<2X#8zYY4!S7vWs;rqEl*)ily(@x
zycO#-i36E{`tSG8p&R-Btx7~G<PqP|2w%fBv;Vq42O;%LRp&b2PXV&<S%4h{uzosv
z;k8*geJ+RO-qEO^vi*NQ9UhI0_Q@l3%z_P`atvmFH{F`8MVTet<yxg*YZ8oL-6DBj
zGvf+8S70*RGI?HLvC86jKpHb=fkV9wHfV&ki(N>rTGmxstQO!FyClG4hmMYn7(NA<
z4dzVegkR{&^8h`h>bHz@8Cdel1Xp!Olb=I<PO}{&2p?3`dd7dgboL=aOz<RH1M0wW
zJVlN91hXkcjZW~yg%leX51&D;x&Jmwdfu7#2sSbd86#{G17wlJv2gfPT~poTplNJ{
z7J`ZT+ThPQ5iBn}!>F;K)J!FnmJnUzNha+8GP1WYTI&Tz+1A|9^+D|5VM2n%4h961
zQ_Y@ZkMU!&AO(LB8C%W6WXpl{Jfql{3nOaP?i1rm>)B3q{$hs`Z=>igb>6WFwWm#6
zQpO4ARnOqtg^j}Dp$I23dEPcdr-VlAiJC-HY@!hYRv~P$i=5?=f(5&UbdzDCuyjsr
z>=#F3bC*O~aoiM1HAD_cD9yq(Bf><qQ6)j0h|88~vg3c%xy6?9{HsDJGa}u-ep~FY
zg5xbMnT|-4EOb;PG}-xWOJxL`X@}d%*+O0#ghc>^KzqMGCCUka;wsaeN}1p3<O)?e
zdK{}X({nq~*o+=Cl^w*8>1J1~xRUq)hi=CZOwtKe?DQn5s}in-SrX{%d~xNKWTW46
zmh3qjw8<KDIj8r3d5&q4b~qF`tAT3WXcY%4)Eybj@S|0j>!o#qk}!kep|N927U&c9
z<V+$)<Hov2{uV>-hPv|_D(AUsq7#UA^;DTv9UvKWvz7|wv#^legZ#DKN*a|)rbP|c
z>leM@q>!qDzt)g5Ihwhe9TpSgCBnLw+Sq*c<odZFECC{a$|}2R9FGpNZ9Ni&m99S_
z*bgc_rkcD1p!x0vI!ZY7)qooan*fu1lcwi9WU`dv#BLd2%Q@KYT{W(28n{a3ZT%P$
z`lZ@w)~+X?s!xz_dNp{iiul;s-i=l?=SE|H()F;@6mzXs+Jrr=oCTj~V+9FL`+~$=
zPX)qVfksz<=<I@L88J;rqy`%}y+_p)k%B4EjteVI#u>gFFIB4UR-b&|P-Bi_h#bW%
zv`8L_E8Q1UooR0a3Bhot@7i`3l^#z}1iI1;u^k-E+BdseOQEVj37^$lx`9YIW4(Ap
z?sjaasT|k5gpR)tYAR{YU;{dQOjAj7nE(WPQGQ*2rfDqZ)3zE=N+s^^<AioCnx3$&
zR+5rDllm2}=_TdnR8PZ|W|Fo7tX@0+RyQdWKB=1&JwJN<{nT=@VA+uOOH4}=3iTUG
zEOcR|G&?0$dR{M)*lT)RI_TBcv`WQ>hut*$I-^f#S89P}Wf)dB_0?v*&Q?UWCvj3q
z#=PBs{+Gtr4D5E5jwp06=s7L3$QgyQR`SMNitK9Z&FvE8V!fLwp-7?2zk5Z*5_Pt!
z=?ED;UhOGq1wvxWbQ_~$`>NKrQ6)KY$<j4f<T$Bi58=yc(#dlcqg&a5VgIl_DZMRF
z?u|uT&`rO>=DbJs5^JTE@OdV%`I76Ay30p@t{9QxMAf^FhoCjpwo%<wVy9-oc7v2S
zx$&ysG#6OWYqv`eZK+%Qj>(;tT4<;5NbdRsOF1o!^q-Vql`XMpj#U!$c8O=X!~y6S
zXfmkKXiZK0rapOT%A}Ap9QRT0;S{l2sVZe#N(q)<6}1${R@K13tkAMee{Y_JrnY^5
zElFwqzM^g1n{iQ&1>?z0hqP-gsQULHodef*9*MGa>RaKWvTRKT7ZTYWBzGg1F{(f1
z9f5-B6NKg!KmnnZV)bWLV&}&E+g~ZzrUAKdleA@N$lb;i&C-%=D|22ys44@_i4?2o
zV;}~I4MOSmWx)ON?S&hlC`Enw2JicSUG&6`(9IYAsY>cryy}gAdWqlDaHUEk6yVp6
zf7g-&*if{Ye$PXP_zR&PK2!H__9_A1Fr5rP&d@zAJHq=dE@LMbOuzFY^M_I`6ib{s
zuFd<sUESL2y}k<jFx5Y8-+k&=y0P<@^8|fOFZg!Rro9qKOaAZs*#mN*A!pY?q?mfx
zcJWT4bT;$5`AgFSH}4v%ycKt~D;K}Uv(aZiCr+(;yPw*iO@+9b+&Tsxu0m$K?9FYv
zV8FrVf0^e))R$q5KNOeXG6ofwvGPDO3Y4sc+JHi$v%-~^?(#q^f4saM-klW;mN_k~
z9Ns<;0$*_aUvL~Wdayz;qr~v`ZTROi=hEQk1wX*KQuwj>+u^r9J}KdFhQJvDM};^(
zSr9m)a75u8sqvfRN8)dr$If}Z!S#*<1PBOQ;6&l^Gamo-_KyS41asz1_!++Xr`NZ`
zUxz}zR7@K`7;Ty0e`5Ib?GRpo(9BprNXfi_;qQMBpAR2<sD!Kgj>2*Rg~g50x)GK%
zOlvn7qnNUmgcX?8$SfeUfJ{PQ7MX%E{nmM0qhL4+6&Bwm0Wd&R<-fnb4Bg1D9}4f1
zD<+ifsS)qPrxPcNh%+8wNS&ZIBM9}~+4CiyWiSjdJ(j?qe-3mDq~Byd#v}(V8tiM*
zlz>XYR~=-2K0X;P@t2;-^?Dgu&A`YVGAb*aV%~}+9Y=#6ucyHqX3_h48pI-gM8v%F
zyKL;7?dUa)aiVR>%vS>dd62YB3RUzA!O`t=!UX85dMuWjLqO2Emgz~T7df+NnkTcZ
z*fQcXOkG(`f0$utMzkrJZ9tJ-Wym7qHZ=zDk^5|IGd`u445aP(T(Pk=X(?gQkYx`g
z>vASj?wNE&g#krye}`mCu_iJZVkvC9SDqC|AsE0>g@_6yLdnAulzAqv4$w|nrI8Bx
zOo})rWuG!$-ikUnWrlUiT#Sh_cqufGgk5G@B5my)e@EDR91Z9Cm~ce{Xt=H9mm!Ub
zzX<}Zj<9X!C-X-p1hz;?2h-lb0U3&&%r&lYmEO#v`n9)AVRX&<537j($l2-WMKYcz
zkSaLcNKRsR@R6~u89Z*Q7(iT41@RG}A(vam9zwr$a>mCPosFohC6-)wOUKDwO<F^4
zA7_`he@RYFs2C?3j&IKmzA^hm2}(6Ri(WR_;w0^PARzdV?Z$4(8<b^Q&EyR^)I|!+
zMVSMJeKWV6D5=6(*5Ij<j9U#t+IptET#Q1==gs74(FZoeZKHFjwVLBdmTj85SNiSR
z)y-Kb&76fnYt6WW%+m|k%d;`T0B5_|F%Ko4f1><4nk$J2iY?jqVTz-bpW3uJ2Dw4A
ziIMGwbsb0V&X`9oc`ec^#ejyy1kmn(I-8}8YQES26T)_@tK20R$&@z`Yf2_Fl^5AQ
zQjAnHo>QHm_4Hi%eM|gGB0{>Af=TZ8>b9AkBB;n-JXRXW@iTeib$Wh~MmaEP<*gcR
ze^!4FjrNiQGig%4ri7oLQ<Z1<RT@P-F{SiQjkdERKC-e%)0kt5X=2Ca9r|t!qGT|T
z-s+R>>fIV#%=;ULkCc1Oc}~(x`NLJ5YQq5ht(-Qizl*Kz()i8XgJhbETe)vm@7Hrz
z5k;=`lt^JcM%-V?_3PU#B~Yj#Z*=IEfBZEdXCH>h-XU11)5Cv<FpP_sGsal_Rv?5A
zRqo;4pI<)z^659haKV4V3O-<D(!#spTY5Ao94zhi@HIW7-J_`dAecweFi=RwZD<A1
zx-#%-I3OMdSU^<MzJ^hSi740A+}Po9m^K&q(@3Ep0;Eww_q84*q9zQHqe4Y^e+p3Y
z+VF_$X;pX(pZb}m;;P?=!{_1fu}IS^{1;j;mYXGEB2r<Ms0CVvJgx?tsmBQ+M_eAP
z+00`Sxix?y5R!RsZBRyrTR5eWidz~&Py~U7=(`_Z;l5)u$s_w5)0m?=W<|6|kwu!s
zI{p6iEie0mi#}A}3CxpFo(Y8Xf0T3^K*dJhil^t<c{2Ix>8SMc<cD!PEM;>}t`lC0
z2DNjwFp+?dLq%WY^J!xA<6haIhO~wqvgtR}NhZoThH5E7S=+ADbQL=#ez9OMW~uBp
zcIv_8r`?;f*;OP&ys?;sT&^cdOjU?bbHf!Hs4&bFm0y0ML%Xg3FuE(fe<=b-98;<F
zgsX#e#-mWOG`R|?YZY!1)v0QI*@ooJz7L7aEY)CUUPra{AgR}J2DVf~Nn};EN6HGA
zxNYm-JQ699885fBN${GhW(_q9`K)WesY;KznQK<`bybVaEMcOpQx}NTW8sxK>%3dW
z&Xx#o^jK*zQLv`P^j5W<e=5-JwjIrsM>CByW*$Y&V>C)`yFNku=&FDAy4~p^&1N^R
zI}a;I5UBy4GbxdH=VNHR2W})UE%+<CMGCi2E_*7vHuplntfJFG#GJxkikN>4?W~;A
zJXJZ(R*a(z2ytWXRTmlHnijykSx0RSpGWl6@H&d<lH`uLXI+|Le^cW`)~&N*TBlj_
zYHUiD6Jbof{$Q3_tP9CqIKaTXZ9BQcoJnd3M?eOXR#TFbeJLKMxM;LmQHlq^Fek&>
zt)x@ooXl%k7M6h~$19^vjyElyayLRy4G|4LQIf||0n&yk0gH=AF!=AvTu&M%0rR#t
zg<Bf@zs&V8OaMG2f7L@TLK`qUb3K|(E-$LCs@t2~RVWu?RblO#<jz`kS8^vk+PT`B
z+|lelxoZN>@{u2ftds~%UEgNaS`IMv{i)1X_09~<>4$rsQB-x~O13hCU(Rx>5MFRs
zT}iOZX9smZ5Z+%nfh(c8@cMG$d;oKQKG#6kAgJXN<u6O_f3$gS;5kBfJ@m`FZ)HBq
zyKlD7?cKN9bhccPwkV^+e$zDMSN2Zv9?iqRb1xWF0Jybd8=zyT2-W0W+*NJiRMFAQ
z&bUfStjmo3jp8jO75-tWIg42?9H?z3eVFeE0W~bRY$gJ8Esj}IwJP#cOCV_Br8(S?
z#oW7;w{EXjf4LJh4!?BpJY`ZF#SR(owU>c}!_5+tyY4a&c!oLn_GO?Qmw%+xOgVii
z%LXzxnMJH)O?B<%5#dC<pVeWNe7T1n&wQbueb42Ue95STxN><#avRSYd;D%vZrLU1
zc{)G8imrTlXeV8<Rc4mta?Qo9NsrZKn|Oyx^K_dkf70BHY+S9vTnuxz84r~kU%Eo2
zlxA4N%^*~}ujY5tDFFS-q77n>i8*Kq9J;@ou>0y4=8p^uwjSal-B<XE$1*{>9x8`b
ze<}9bt$4Z5XCy)kJADgv%{HZD2J4D$_QeB%vCV?QR+p~MiTgZWagC*hx<LIsR|*yP
z4E#(_e{y{r-scORNjsDJ1N^i^evE}+$`Rg_hwiia3`O{GK;$0~n?nkxr05~8*co|o
zH+wzb%$HTGlO?(dmTc_0o$uoH#O;(Zq&o`WXGt036IAz`_9cD-YXB_6x0AoA=UEaZ
zTlG8D)Lbuj>t@$6tJJHOPP)03FpV2GxW<Jhe^z%Ru+UH^Z0~PN*i7uy^!b{>iU!4a
zZ@wlFob{%;@R{v=y+eAvFB&YTxa_e#_%gPzHZ<&dEK?VEUnI&|Vh4ra5P`g7&Z!wa
zyyN)x*h@Qfuf=&^dcuY>%&>LF){OkRaV1@aKNy(V3^&vd@47}g64oA+=53v-<Qbd&
zfBxWX%Tb)Hngesn{wS+%v29$RFBaOaxM!CmYwc)Q+2wK$*yF#DO{p%P-{1>U0~5YS
z{76?Y5(C8y!^#EIz3GTo%vtw??5IVy!;IPm53r(+f^8#eC(E5pphwtJR&__f18vEM
zYm+^Pcbbw5!PEcX#z>3wLSLkv{D8JdJ`3<dSER%K{hA{CsLDN&xlFEViL3^UWBJ<!
zT~`eY6+v%DXq(=M2Np;zzzg+JJNYLmkNR&+qyGm`8l-WT@k2os0Wp^WKLrz)VXYS^
zf3;l6jvKuZzRy$i3k=PrxB-JfH1dpr+-&$3<PaFKbFhQ_e-3&56qn-K-D+v%gDf?h
zEV7EVd{yL_FFzK(e0q5K`tadR!D5-y!pi0A_eJ0fj{gddjYcn42xgR6zW!SN_Qbg~
z_<6w(aIO@7EdF-*t)Cy2aM(j&4}q;he>^{05ZI!yMPVPQ@tfmE;%}SV-tu^Z;~fVG
z6p*&Sj^Y%)pa1^#uMa#g_`{oatE-iYY2z1Tf#I6UYLT+W)dEt)$+<?VfHMNNi;vHb
zV8VQW>(es5)n{A>1U}%5Pda@<f4~)MmwJbHpovyDLg2U0Bf_kky~CSGPv_s~fBbFJ
zXw4Jn%$vY``1&+Fy?%Z8<6#s5(qxwV<=d}^@B)Np#`?uchA;p8=kopG^GLMS1}9Ly
z+!(EU`5LBC^hQf&l_&W|nFj7;=K7qtEIfn`FPRTOQstjNe=I}Gmrs=y<C)-coK1X;
zyVmLh4@_~!0}ZJY)vo1B&>7EHe|fT&86n15f@X(ZL`~>5Z$mPkGi70<!q3-p?HzMY
zP3z$U$B{M76O)l**RsR0M4MKAl55zlsOc{1XFnUtFw@p~wyx#Nwq&G|%!^59I9j^l
zGWZ3N?an5<FkIx-zzem6esId#B&t?LO)He5?m@P&w`^^YZS@uBguBVGe^>$tW*spV
z7NBW~WW?vpo;apR2ABgo5I$yM<dv@0If>rE4^ys~S?MkT35)}ZPrqS)Wtp&QT!C1h
z9Xo*>B_zpvi|Ri$?SJ;jp-qx^Cds`ND=k|9=t8YtjN;7667z~zhD<$-HrLy?RZM9!
zQXlNQ^mZb~-=X(AWGy@}K_lHraYNX}8}+@5E!wPiB+V=%{zi8E9<9*8gg>}YsT~8D
z1a(*~ntsv1-pILx_s@i1=KW9n^K;8HVT0oR-{F<cK_`FrXpMBt7#JyTsSQP^VSbV3
z54aX&@*1Hn{x<m?K1-jm>p8dE_wP7G7VenmP%MtZiaoZ=nTqz7c{GbAK&{0NLbidw
zs0lj8)vJ9=_Ey8JTfQx0rlI$H_C9OhKyhpDGY%T(cQtU68z|2=S_L5b#*I3-F7V7N
zvnH3|qR4;OhPBBZEU=9kqB|>Hi`%LjKZit8Cuf*<<{n{4d8W`$4-~h;P<Deu?>=SA
z>T6TmeptHB5E@s$mNB7SJ5(ENM#qY)3&{uOEs_7#Hx({#rm$KoW`*zEv#icOh8TA&
zDb6XpM#rdZ-8<Dd6{wP^sc(_TTMi2^H9QQcc9VY_np+4a6|{O&D7_>dihI}qv~E{?
z1LQ-}E=Ds=Dm__h?z+ips90<<SxL5hu9r!aK-ub%&(m8dX9)Wg1B1|^$`JE=MQq%b
z+XGsI#LzJ0_F6OR#eQ->ejiBo!Aa!JUe=pEVYZSi#emVe!3uQ9Rp&QmdWRzx=s$!U
zCz^k9>vvQWGP)jTupOK{62WDtxPh63@_H9^G*G>Ojz(_YlWV=@HQz2{WFLcopH`J&
zdyJ7;=@4RYaZKK**UZ|HXmI@m@Wt!8)*2LY#c33sv&+)1MBQ66#eRZ&m%lP(>nBlU
zcGndz2!Hj&qRO^Ko|mRbdI~C$fVSbcrig$2e3Lm}LC=Kx=twYDqQz8&E`zLQVB|0;
zEBfS35!`_jQ-N^JS#K8QDGmt3q^NvDh&Cd7POYD1{4LwEb1W!RuNF|Ag?f<`aYGN;
z7U7`cCnZ#PbE5KGR1WH|YPnk*LQ^d@Ei=EEuR9s!j~}PBqN3u^x`UJ?`f@EtAyj|g
znD(q^GMDJan(jtWS1Jb7VTLM%5nh(+X<Pd9H6%u8l|->6dQ&#1sei7Fi4^G#8Ix$w
zHVRxC#7|>PRgrQ^VGF37RE+~}PH~F>1Z>uVbX?EJcKZ{YY-ZbsltepbE@gZt5!b^1
z3vnpnz{nTksgX6M5EF0}(9|CbD#d@NC29uhsPaxP|64+{2nuP8u>s11K%ay9`tadD
zKfeFv+dl-u0sjpn5R9Rigm=rY^r#Kv&SG5onV%g46^<JOCig0oDUhD>A@JmZrQzH1
zfD>VNDVuITLo4c~<GSe^A)o`P>I6*_tszpRR+{f;+eoXbQ$(K1GQv}!lE;6BM;uS1
z!ejWf&(sx1{cCynzC3&`^7Lq6DugP_&6+rIQVfBj6)80-hi>1}>al|f8@bllv6<V@
z$2$Wk0)g-b#pjnYRpFFc8g8ixNf8BVqVKVNgZqKeB#)$&n&}XJ7%~zxQe=@PvDV+;
ze&u05aL}jbJCS(|>9s&gkIH{3+<h5nMXbM(Bj)bZo1E$CS?Q;2qPl?uF>+@|k!B`m
z+8WqE6qoRMX_gpSeXP8GJ}MsEBx}_H@r;qW5euLcFc~JXYy?{3Kzj^Fwff$I81&cx
z#S6zLl|5wsbvQ~8r8acyCe2MJU!_+UoJ%WDUS2${N0(u&qgAd&mxh0@o#-N-@+rR$
zK3SfV&(8CZPj{^}4kXl|EP`t_N}iuCjPfkHhkjHeol};VpjOpH##VwYVt5wjDMfF9
zAvSlhlC&l37R@f3v&vJmlRz=tt;(N}MvR1x0M%}}MIushh`)yw$d8Fa#6~LNqSeyb
ztOI+}#)ONZrySWWFvEZD!wl*HQ#hv!7^zj)ger-WsI_x+$G*LXx(C(_kpi3<dA%dp
z1XNk8r>z?j^EK~UJ=uzPt*$b8*MBd4uZ)2n8xXV?CI~VKhnY(Chtn2iQM5?U9xkSp
zSvaPqqMk2PLgDVVt(e#`E2fDq9suOBq8W5AN6EfImdc1%7s-DPDUI}=Nj@>qm<pMQ
zt>=Y=j^mn4@jn-^c4x$ue)V0OdtH=x++XPcv0wF4K7oz0a%CE)5e5BHS`Q)reP78X
zdM_AP(cGZelzOPlA*dThMsE#M?iSap_loVx?xoJ@?R`^c#nC&6iTC>|eksfU2E_*h
z=={{QdFN6U-t>Ru7Qv{QHgCMpaJlxpQsK#JZ6&j^*Qf62X|qCD%0Zi+(cwOOJxme6
z7f9KFuUw<)j5xM57o_qvFMI>XBp2@8ge#!^<a)EPw{JrhkjO|Z0Ul_hzNtbw0Dy2X
z^A1+`VE+I|B69N6=4G&2OEaotxfFt2hKMC8ptlu3@2`K`2Yntl_TT}?9vTh^Vc-e6
ziv*i?q774CtO#VIx}CLKa=MLoqh@GCB38>;GgfWQU;sC|#h!gt(!Qf`RC=GgK$$NS
zLO#f&O#)j2-6i-WK`|rM2%m<_Jah^>+wAU%E6srl4emOi+T0sh$wcyfuDcxn<r`BM
zY?1?W1_pmt0uYf;qr{%mPK&ftGL}#Q_H7RRnc!j|-=EZt=WR)P$Irh~{F{!zW|CVW
zHC#iE#^$>cfXL@v+m}ttampo>EzxCaIh#<fDkwz=FSe*`3d*lyGk=Y7rQY2cVcy1<
zGA|{Qv2Ty7Iz!gk0k_Iw@_J=6!{~!HR3ms!B`1H`oXNNG*W%?q?(4A8@v`bbDIvYW
z1>6tWA*=x-#`)f-Zlr5fo6T*gEnH|4lUO5uaps8*jEiajJyC)Zwo~x+NPuv@UQ#HF
z<87VAHWz6^i+CF9YviA5Z;El`md`sG(ki&x%dHBep+IxNqMIO8iY%QK<O*Hk$!&)~
z?JIxnjU0V-^^?tCbrN$hKnfSE#!dBw5?9LK+Q8P|(WLND-VzUR-MP^r4=Rjf(p6it
z=!>GeDpj!cqp!XKX7`v_+lVU;zj!S(c4=mzwI3;^b(kyZk<oX;QzgR<w}~nY0k}1Q
zat?8>vMXq|#b@qf`mfLz38z+P|GfYgtYv>1V4V4=X=p?(pxUBXG<VemJ6&(jfxS-Y
zi^16bs@hRgXzutmE=She0UJe-gxIb*BF-4^oZWaueWqf<2@=q~!nbzTQtIGM$s@~7
zc1Pz@Y#k8TeAw|DYR^iAdwsf-?Onn-1)5T%xi*|@j^Vsr*N|J@rSfhpQ($i4UXg$O
zLeAuMuSsEy^v?wa+zxDt`b)`pbKCxSoFR91?jrA;xW4rpnOL1&UvoL~4i{~dV;tey
zz}=J;y*0SoG!6i+sAmJS(?sRU;2l`;o>`C=SvfVgpP|2td;?oJgc<O^kuPO6&Tx?}
z*C5(#EUKw2U$S?_F)Nz3=XV%ASO$Om@;0qp6|X4#5Ko!|6Nog2nKn)=1ojU1_o-HJ
zCU|C7i-m3~7IuFhZkk-du4NMC)EI`p94`l><%U2Hfn@6yV2~1=8ZXpT`I{rxo1o=R
z$#%*@rp(7Q=kl|}bSIsSyK^u}c>Tu0=I%PnnzN7<6u%kwOOG2jZbIoTW;rTVdb(GW
zR-C*;#O7GQxZwibC43rOT|J51`=1vmLJfbB@cQ-P|6zA^JD2f8K@|cqIhQd`DHE5m
zMFuCATnj-l0@kvZj|)LO1f6^OG?&i{K^A{LB^`>+UA}hC&s%!lOT;LFWCIO^hj@Pc
z>-G18%nC~pY&?x0{M+;E;h)2xzS6?j(4kf$Aas9!JH#6ZMpzd*g-%GPyYHX7FNZ(4
z*kLkB&-8zc?7n~P*p-*;!63<l(pqCG4=O8+@tw6n=y<F5AU;WH$YLbR@sSR)RzZJd
z<;Sy!S&7C9tpk@xZk2|7isVyFl6z?aqn_nYQop)F>JO4Vq-ZnAN>+Leqk|OMAs;=&
z_VMcZmDB>X;}QX8=U0SH9n@rgvfa4s)$HBjd%S0_*9!?R6pW2}VGu#vIlt8Lq3_RG
z7s6-;lrp_8)s>e|i}+H`2(%L~sE~hWP(8My2Pj~albbN6r4b4>mr5xtCFQ*&UqaD+
zJv~-)gYpn5?hSLJmS|4b=YmDy(Q!#jx!^SU$@0-Vasj**pt!&V$PKB;9Ellj@gOx!
z$Y3y|W6@+s`#!X02Bz}y7%O<oh$|xugq9I!67B5~uO+QTydK>k;<doFh%<k)S;W!y
z(KHYXUtb1N^f4Dk=&wy9pCX=ufBib5ycWU1#FV(ALtNF_YQ9}ExmN@~iMD*2b&{4r
zV3?rIoRU=^tb~0L)7l8?fi{MWP%4z0=&sIXUFwh6OW`ReD#XxSua|Sw_Dt*sGwc81
zsR$`O0bs24t?NCl^H!>s*<XK4j8IPg)7NKV$?IC;QWYp2Q$2c56$mJ-n)qn340z$a
zp0j^Q<>*l={O0kE0^cGVlmnsUR?UaLg?X-h_ohJ$XVgZ$X!}A?p2R=ya?<>ei|08P
z)5o~$89kX$cN%-Hq`xYeeBt%b7g!vjZ`{?Uq4ES4SOM!*61AfieBytrw<`xy$Lqu7
zsXJ5-L-}SsHmn>>qS0e}w7fVDZBcKR(cSnb*;QS+%<f``lftQLZ%}4TpXK{gT8c|<
z+I~SV_$#ek+=a&sPe+mB`#pdun6*<pn&-Fn34ON@PYg;|*s7)Q)X*3cgtK<Qk{+;>
z?UK*&Ow~BcS7OVmpZhR7C~w!L_?_VG(GPyN89%1SQi^QyRM9QvxZ3B3GB+677hX=?
zCYY&^Y>wvz(&G8I4G%$`0!^Rk_l|`Um2v8+kF{zo=da&Z9ZMs`Uj#CXqOLD;#tTHl
zY=Eap=QTV)om`-NWk$QymX~GzAuodkv_mT$7esxr@NEhKMfhI}(jGLI@k2os0Wz1N
z7zh)WAQC}Z13*@bmwggJHC$fr@KT&hiI>IS4)6NmUJ8d90y6|gg?PBPN?=4{L}CtU
zymPz=f7?9vlGhuocU-AJ0kj1sibuTk!(TuB@s8&W*WHn1+W5&>#juCEJN0m{DC|OG
zz19yz;cu4?6G0t+zxfNk(K>wdn7;}0MVRk1-#FK#SNK-hmYSv{5hLv!O`IvboYZpq
z7i3(HeteHJa$@a_$V(=K9<@+)d<v&9N{N@p=ToQGEh~_UGwIC8Vb(&V>9V{*Ixy2g
z=KMME(?09rCGtJey{Wn^x_4zwq)I{GJ;$K&$SI6hZI=;$!SlJi<e8?W*__G3x(qRn
z)5uas?y#v~(@~ym6M>c#HjJwBSLEjs|H$(t@8@Z@iR3RG#k@n^SX^91$;LA6pebM1
zH(E*~=V>W=Xps8((?W0!!-DC3{(krK1RHrjMMHtjO7HaapD7HBV4%hri+8Dn@WFoX
ze)#9tFMs%d{F7i<@b9pKPYO0IyqmtKkJ><|*4pV?{_L2NMu0i<2m@*|peH>hl!bNR
z^K=JBm`>D8`xZtevkr#WEEbv>r%28WnxV3XNU@9-e&5=`va6Xzl&Nkbd<s<Z+VBzU
z%c}4(yxPxYDpvh{y8ANSeW?2M694dSEH`WF1XE#u5sVv^mLchw@V4w86AaFAb+8jN
zj}fFb3Mvr9Jk;l>%Bo6QMjC0$5J(jYWvIN5;~UZsf+l?=o77BW(}qEiV38uLG^us|
z`}6m_>?;<%Z{8D`dm%j&pmc8#(=?B?V*bOCuI0Zxz05AL_3=T*xTkWg6SGgL#~eix
z=Mr0gTmKq}VhbOpW{XkO`&rfx=c)&XWUV|yJ!2wm)WS8BTvoMgz=9%yupw1dwUi)q
zN&#FZB_Y^{(AJEE04dEyiKDD2fii|R*US=SmR=<6x^^eo<QZ)kO?y2v*<^WklTCs{
z%7*YvryRZ$-xA(ktC<||gE?)T1c)<Fc037x{&=uTBJ}BTRg$B^8Z}CnljN#4qckZo
zoffCy`ovnSC)--#PFb^o)?;(3F>%&J*K*CtlE2iX=nCM<@^3&UVF0V_xC44h6e4tz
z2{zU9m+6!gt?gk~+vEZ0)I!{%%!C=vL=|0Tms#0qROgkxsIpLNc~lhgvWQr>z@j04
zbEtq6ds%OYMV0o<>c!Dz6tfp?=g77aZRgOLqU}GXepZG=_cJEi%NhkhS&cket0uWP
zp{$7(^YvJ7&#Wq!0By8f1mSS*+-#uGXnnzUxV;p8^21Qd?FINVt%x)!C5jjp`63)A
zs<U+`hv!*Gh9=LvA;7Dyav?@sbvk-~^W|+?0_3;41aL_YOa1~zZ+97K6ba<|RsY>*
zq%E^V3r3<a(yObCVgW;Tt7w03Rz~CQmi+1Mo^44-vF;yit2;N^)jUp_Ng*%Myy6Bp
zR9>igB{&oEBFzU}?|?#=n!g7qee=70elmjCe(^86XQqI_&BCJifn2adIk5SEb)O9i
z8sh#E(@+K#j&PTlhO&e>Dn`?YIhgygC6b#jH6O2ULvvWDGNVrKrGE2!oQbgql4}x3
zQ+m`e>0L9>G6em>4EV0_S{h^@S`mJ&bD)9Ym9Oxs0IE>+lka;QDYmB3=w-J1RD7-h
zzY!q18Cr%oO5SCpkm>J4{$>Y%u8XIIO7H879oLc4S1Ub@HR#Uy+_5+cqrKkV{Inkq
z0SCOZ{Sx>V?T23Dnbu}pbrj<fM^VFei+<VX;{cmH^-K1{iv(EYM)XU5-RI&pBA{GI
zBli863nqil^hV0xRWx4?m^V=I)O>zwA!LaF+1pbhe;LsDM%=QNd9^%$!-M$f5W2is
zH<k^Np<B>dh<s`{KZY-HM))E?%Yp8}R<wknTeAoesdC<{U21TVifL8bk~^+`7KlV8
zh(&XQqYF&C`$3*kA<+*M%syYO;Y)5@)yRK~azp?ualYU(mLv4ibRkRwp;eXss4OAq
zkq1c8bIMTlX03iHVQ7bc;yh-g-PMt4>=mL(9@>L;<QhDNye%FR^MNcP^P0u>Oedgo
zvk>RNVx{uTIK89kO?C#MRj9?vr+t0~v*p4=W@MA?MsB6Q`O1Sl6&PTOx>B;1tq1G1
zGN*wLrl`AIYsE%@U5U<%@~sxiN2ROso&g+=rK`@MBQ-q{GF9wY$@nJbxWn#C<>3VL
zWG-tGqztvy1u|TeM#rg)P`-TZ+8?=bD>tKNQ?73o#19dRiM*v9QZd6Em^IQFksg)O
zR_3$vA5tq!943p<+7Iw(q<Sif;m_r-mv<RKAAfw+_AthM)?sv{bb!m2%TY+g5*DsU
zhRtQ)TG%<zrnmFu`Ki6e%YnFbH`lzm?|<gDO7XSPzxg~<6>rS`dfL7$T=C#$M=RHi
zrNb|3Q^w8FX?bDkx`t=h#LS)YDvLYV<!P3G{c_aZ{&Ms%TB2-GvSP9DC_{}F?M;Ad
z(0>h}Yw!uYy&pb3f4cj5VMM|M%9zl24CC=;5(Y>LNy);!{x*HN`!G_?R*b9{Z-2sn
znkQH6Mn2|eKs<+#+!#~^osb5_5<F4DbL|Y(o|?|(DG5uX`$XSfQq(5g&yEx~+Ul4S
z_F4Xm+wVp`Hcy1WAw%D4piF%I`4Q9u5`VLX*>jA!m_KoeQp}DNhwk(|8!-;$;mE&n
z>)NL;pIZ~UU|%ZSWQ%k61&HN=GgtI7CSCoU_W~2m28ARR_tK5TNWX4hWKjy3g%(1j
zjvQ!qbQ~BL_ADEd9QicdW%pF!>oHr0j4aCDkKN<YH@J%pg&X=tPf@?S+!s*%IDeGY
z1EnoruqZ7sYnXk7F9>ItppoE)$yHYz6sfYg*?jZn=M-NQ+dkB_i49LN)!*>MC8RUb
z-SEUmisscPK2SP1@nQD$Cq587IPp>J;l$6))XS59C@m;{c^)ArBV-#zQ3-Vl*@<)s
zWvCa3d~w?RX;n#*``8{DW(ROFmw!7~oS2lF0V|(*WUx|3cCq*U_DMLJ%lBn1P|rec
zp;z_a_D-c-7$860Jl=;}I7_#5{+^(Jz3+E~j$FT+d;`z#=p8wJH~G41zp3=E0B;$H
zoFxVPfLSr>`!|4YC|F>b=Db}=B1T4xySiETW2Inzm47P=x7efh4%}Nw3x9AutobdR
zTM={fqb0ZS?w(?LEd3~dRpgT7(&3TN@?Vsz8)==f$RSY-sSdtx-s8BMOs;CEER91y
z??L$l$i>r;PA`xZ@ZN>ULwI+!Unyj@q%7Cm=xVI2w7#$x`JBE{{PlaudMEEtfZNvl
zqWvr`Nz2}Je%9|AJrv$ZoPRi@BhSD5G;6Q(tOMo!qmb(eAptaXF4?PXgp6VHGrYSd
zqBc!LMH4CW_!%p7x8fxx<)tZJeeg;#(S-biLW#<um%E{NE2s9Ddd)wcbCvY}IYB8I
ziMe64P;uBRiB83GL&ewU2@-;lzh%)sQuGz$obpY&x&h7!twckM{eN%5HHXM5$JxJ0
zWVybV<QyY1<7ThzC1nUUP%Q@i8gC6x$IZW_>K=V}gQ2dsbTL(iADRHhI}<g;Df3?Y
zX77@`beQ9_VDp|#^N_u+*N(luHrnF(SA*fvTkkvkp_&T*l){drWG!_51EagQi-kF$
zu+n0UyUJY4O$#;D+kYF_{>l9j-9J>v>;QhA`YK^-MqHJ5<KmDoh_u$r`}k6CXQX`)
zT9qFe$ofNsS6xGdr%y?Z*AU1(4t4kbEyVC1ZyYwlVfGSFObgdlh~)&jbXIZH>aR%J
zh<O|POSw+QAA;r|!Ir;u`TL>%=Cw(Emy~@$2SN`z2=GBX?SD3B`UThYhQcGwipCvm
z6rSPPQ=Ipb`&iIWNsrIVABHFxnbngMIDX&#;F9J?=SyZOxBR7oZG7c5*|29So;DR-
z?s)nI$qH7*v53K=Fs+fzKeTzB0Qt_VWwW#W(tc^DC$n9P2pi6e*ziB?=a;JrWo~41
zbaG{3Z3<;>WS9R#K?eaemvN046PKqRLCAkaYUh7{{`w*d!%_qr|BS!J|2{rn{`*qd
z*G8ZXC#!<cU{3$}`z5{rg~B>|EQKR^|M}ze=gV&+)mAxPk#eN9h@EnPs5?Aa8-$6k
zvT~3F1ri2b{5vx)fTq+R|NM0tp8WpH!ZUG&!HiVZ&+*f#VVMzxl6stx6z<ZDiDQ5C
z!jPC{Bt(v({C+RoKf-z8qo=;YxfEm+DP*LfF;nfH5Y?ZqL9S}uIboF=S1}>;9c4(G
zCwL)tt8Up*OCywVBSriu8hD&nP20zWNh5LFprX&^7=Ky{g_G5+T(>Pa0VECEHm-$5
zH>4A|<h8-&zk$(e^LrYa=a~FS^W1;kSg1Vr+&mczT6<wFjH`SP77&CABSHMgek&=@
z<*mhAe9H{R3UI!fAvCIwTkTrZ&Fp#Je)n&?Acb)-uK4V27rgM^jH^VgB>p3`uET1H
zEXuq<TCIE1%Al=4^e3xPAnBjA<{7Me{=b@ZBZieAfIbL-$~ncV6lc;JCINp$>(^#8
zl(jLhT>vvvw^jscChE3vtJh^4#;u>Q(BdIRz>qPJ{x7gi@AWwbPgI48brH1%5yB)Q
zK&QSf5wthLko89qR(H|12=HAZ7_S6qHHl!$!YW%OF@i;}7q(CeSKk1W{nFF;pGhj<
z=hd+>g*fiLjtkXID3?apd2fGBP$Z0~gpvBz@!Q@DFy<&&0R?lG<~pb>p@MQAV;fn)
zEA!6b_|hUV%V|vFFpdoMzl9H`wrLL5CeJNgRLSyu?44(Mo4`aPu+j9y!5-KQxoqd%
zq?syv53UUz>?6si(9^9z8fEI&9Q@>!I8T{ruRFJaI^Il+73@>3&;);}2}PE?qVXFE
zCS}~KR?IlwvtpjdR!n*OK`Uw*gwtl^?10{O-53R5X9udY<zh_%+N4D4tdf@9lF*DA
z&K|gzc5S~e-eH;D>K1T`)aX|9#ZM;d_y?_%B)goy74OM&j#-G1F9PXsA3J;2K&p}q
z=mXYR#Vln|wr+33OkaOm$VR18bVvdw^ZCwHmcdh1%O<_1+?iF2o%);b4X#tay%*#p
zxYl+M<&KE7>!noOK^^T;w^dE-V~QyI{uL>c(v$wV;J!qI<)WLRUW1ZrT&>OHGy}g1
znqoN@izyeDg~P>wdvb%P1ly{Czbz2y>}FVsHWU*ov>M&FR{DR=q@m0hwr8fC#=Aze
z5;n9k-MSf#dzAJzdRgnjGU+>lsR7!eFzRWip5KHcWsp?wbcEJDFNN23Lmv@-*tTWt
zwcrL?1=qKar~`r)lzFeI<4L(i1*UQosqRg*6;SO)HOn~r`=X}uNF^)uRA<{Iusdk&
zVQ;%Y(j-_IIevfCovc^EXWqpth43a`flM6Sq{>SxXjgz;mpz;k2pM7m5zYpd<csEY
zJG)Yg>29}gp_r<uQeO9gN!)_bYigWJlC0XOD*c>LQ$Z_HP-)Sn0`HC1`=F!B8+TWn
z#V#^;(YqSn_PR`|SF;7z6E<{Qf)yCS(!0tDVYH*ET-|^0n_2`tDr7`;-IuXDaB2HB
zH|h_nI~0u6s~~m}hJD{g_yeL8Bu+_`Z&RchASIz%33gkFRd3ok?DZyW)2LV-#}ux-
zPkWwxj~eUc+Dl=0y%ioV#XD@Yw#olkZ%<&2u(tMv(%FDop8+UHZ{rN}R=PXvnci=7
zb)xM?BAtI?^wB(HzvMAWgfK27uN&3b+v1ptlVr`?#VNIxoZhR(|6c|d8Q1!GvU#@)
z+EjeWNpVKzZr$7A$+nf#p;~p<)p{+woizGDN`y)YO-_5np4vw$<+X0<cjIlKb`$Q#
zsg14Lfk^7xbA;l)eq(qH&AGs3DQ|0xoUpxw==*=ej}Bgd9(ha<$Ihvs+JZRBHHhQX
zyvgH^v|Ld2X<L~)74OTqws@~&H{l9ltPo@qFR#T03kWAvnCO784<7qapu$f_;14tJ
z2fp%NXtX1xWSPq%Roe^!t6o>mjyU4K#Sudy&vSOM$n*A2tzz`|h(mxZ^X$;(8Jhfw
zQ{8{lUN~0c%Dzx}@89k_BugIN`zP2XwR9Y=V#8LBR)Tk{!y5%kK_4X`-u{3i#O*LB
z)2Ug%@JhLPnX2~E$V-g$HN9KjuuXa1)G07+Y0ANEv%hd7xUH42qxYH$TORk;c%Yi|
z4xH`d%?TOZZKq0oApRdMH>hY*ue%!9^)7#mF$%9iLIs&yiu9&KI8`AGF8IimTa$9Y
zlood>OVje*ZQfSD#y;J4E!(Nx#p#-V<!U5bQK|L~P94xb0dlMStfzHHP@uGMYSQx7
znckBtPLx`pG2`kRWlk)ZY*Q7#opW9g<@J2f#5FwEoh@g1pTyC2P1A`+AH8^NErWmZ
zf7QR2H?lae>r!W0?aJ{k`o+a1o!O3OsQk<gQeLm)utjZdoN#BnjtC1X(@5E@wu=Ty
zJ^}2g>7+-@E9dJ1kG6J_L%6dOjM4lWC#GDfC|kiM$T%c;FSRu`34~jmXjYxuG=f_N
zhP$m$iVVVQGxG684ILxnZM;bf37&sOm9`c(Z7r6Gw^5b7bvX4&s>w~`iu<up+_7)Z
zwe)L8rFGkR?M(}1CkYZJe*B%S{NA2L(pB_OMd#4B=t(J~R5!L^9x)(hH=*(<tejT-
z|0>WZ_=qs8CBZ3mKJ|NZ700r&rhH!NgVk4=yHh_EQyj>y<3%kyS@ePA+IfFUu4ilv
z_%1>YfuagpPB;oq_V0#JZ7ICeT{{-I9U9TB+Jj3kr~T2Lt<=4G>$oXXI-pujWd?1j
zH9m9D*w5%8kd%I$oj|3`+L5qVuj7BoV}G<#wN2zk3llB5N~=3O_I&J&zVX(jP#Eam
zICqMjA`hdi5B85Eq>aT%@4A2S%D8b`;f73dy}`2U!5FP32oLT~6eu9g(4%2?*Ntpq
z1wh-4hTKGHBlo2_*D41&R0+T42%Bzw4UHG3iF#<qAdjsvO*PRd>#->qHtDX!%V8{Z
zvJI`IMdID)IHHl~9dWi<%C)utv;|eUj6B(0lUajaJ*#wWsEJK4Y|(#bmV~j|+clmF
zEH`4pa-?5qTuOHt?CA9XkGA-r-VC~MQx3=61niHH>9%h*!3#;{*`VEny9yMDpw#MY
z{UFa{i!<Gg9gpdfmO|U+#5?mLMwLPlnhV>Ho+nRTjWY+4rLl3N@uY<gW@JdQ`6v_E
z5hJ#?WzE0!&7;ag$FzUiOQE!0IrY1Bw;)zg!|e7P^Vt1|r4&rx?zz#oD>dA(YVQ@h
z)=4ejyB5e$Xpcw2Tc?Au5$fQik$%@+*SlQ3>c82D>Sq_Ji2>5N9KWEkVL9f5qFnIq
zQ9ioa^J%nS^u+p&JoeU#JV)i-$F~L{z2D$Felx&(9lv)#A8&uq<h%US)87xs*sb*V
zzMl?a&oy1-9+9k0xC(Mq9nuTs!&j)ITj^nSw5L|ExItC}VB?SYyBeYV&9_?Lmalx>
zsV)_w3sqmBu70KOL(jFjfoAKa#(FVB_ij9&-aV!sXVZ7UzCb<QN{_2&yqLFK(0YbC
ztZeT6|4?;wD?NYl&RRyx&X>0iej!O2OM<xv1N|_&yRfu*k+2CDcssbukZzX_D}8Zr
zS4s=1AB?Hp<RbG!p*>j2Z`6Zabd&QJ31RWto53B10MFdt$2nzYTAK$s?<TKu&IiM_
zIa4kWJ%RIhC0WB4c#3YJXZy{Aw|hrF%D+E8U;YQ^JM#e)m+?bE6#+Jv!4@YImlRbk
zD3_}$L7f3Jw@E8OsRaa^N&<wJVT?Z%0yQ<0;TR*AFEK$6m&Q#hCV$miU2j`A5`Fit
zxIkaJ58>quheOVyD3G@6ZWk#wNLpZb10%2%ZR6Il3|lGE{`x(`D_zTx(khnhcoRe#
zNps1W;e4DS8A@4d7dyhc;k8jYc1Z_{I*qwe+|{IwN*zTV$l{M!S40Jr;FpxDzt-V*
zd@F1RAQhwM<WuB=0e`j4mVK6zsF8TJ(jHUpm}6;V>R2$hlhSdn(#YI#r6v~yVS@@z
zmM&3gvlZ%+#6cVZDQ$$pAZvT9q%LJ{j}<WZ1140VHw`m5+0^qXQ>o1yh5@QI5|SoU
zZj+S`j}9}`P7*aGigl7xn+y+&>M$5}QY|%-yBxXk!djUJEq`kP*{tXC=furUwSp{d
zwrX9LL3c=!B{wCEIxL$TQUoJpX<A?@avqk#q{}6?!3?w2*k*$y(9a)J!eCZPBL%Bo
zi_d1IQ(in)JC(s?W}W&ffj7oZ)ufSdmdb4ejINfgk*UMH1Fv+2IX5zQB{k7CD*VUd
zfs>^x%1xYDyMI!eNtC!NwGJc(Ga9Ue!Io24S7RI)YFFWXPGX$O%`+6{s@yyp%5+sM
z9Tb8f74~hv7m%e11_i5HdkVkR8ixTCWElqF&SGo|bLxyzQ<$YQE=_WbYqDU_&T>*$
z4Uq-!Zd1VDt(Z$Lu$I;4oZ86doa$f1ILIsbCeko<@_)$VLF9x9BHkM9JYvJ(>e0!i
z%?Z(B8EoV>iPf5f?i?6spvmcBE)QZN(?;Vd#CD@SY)J;I3OK1nGvWi1K(scFX9cP$
z9enXc|HE$=)Bf@K`E1@le|s|bPft$If9d+KW|yy~moFj`PCV-W-hbDB{ep%M&!&_4
z3l<87)qg}>1V-Gk<$zDbA(|ZZk2?(hy#IFg!>s?N`<w)6g#B+cn6j4gmoM?>rUaWO
zg%TYtCMD+J`9P|M7H*k?!41SX)O2)kLx=wHIQnFMIy;9rKR)};|9$>?KEJs7Yv2EL
zI)DB4Bur*+`k&vvxp?$y`fLCF@%;5aFXrE!&wr<v|2>{e!_Q~^$=U3r|CIvAK>gL_
zr2lb#dUiTLonG~m<H_sO^PgrH^V2t{zfUj0{@D|b_0bb1g^~xi*D_fO;xV|;GMV9B
z_oL=N-hTgdR86OoipV;V6l1O@%yc}NOxf(!t4Tbmj6#xWLb2&Ss`~PD#YUe!Nt7Pp
z#ea5FnzMu|*GjWeNTLsCZAqaf72Zuv(h^X(mReGgK$RYlmX)AoExTz6t%~D>g_afb
zm-0~km(_!|$hFGfyl`NYq2;swU;qA3hoVuCiqGGkogHoFKn@QTr2~a9c#YYDH6eE~
z5XUuZ*MB`bn_WJ?fMi|r2fF_0@f@<AcYi#5{M)<vx6kLt^Jz!J7l*U+IVL%5wFc?f
zXE_9dA(l46mtXczFK3hI)A@`3={JY{57T$^qm_kydi>KA)50L<^Q(r5VjJVx^lJ9@
zax%SIAT9sy`{}FG<F97#+WNPWZ+wIiF7d3hp{_Yw^OdUwifkeryclq#OFY2HC4a>u
zl}1WNDvy+oRB`I)wvGu{6x5N<!ZU$v0-<g_RM)H-u4%MeG)k=@W?iY863`bsprm68
zH6!g>HVcr+b#4W~3li<YmdzMZHseOD@R67-QANJ>)+z;b10pFXVlOe?vCi*<GGv}u
zh=AN8`ay(QwVhBITMf+D{*_?(!ha9Rzr;|~a@vQf0ZTFvD3bvGT0^AXA~3t@_0DZg
zuytOANim1`pkPu)ER)9FVp&sQy7nlwQ0UV*)SKDlm-jR*NLJ}Rw195wy1*+6yl#XH
zH`oRofTledZN=&tju2>DurgQ$ezFBC!G*ma--fkddBpA*v-F3)>wuWK6@P%Kc^$yi
zbUOfJyaHg#2trC3sX9`&r0NKC>Iify?*v;!VrZ@Zva|^tLGvdGh>J_qu@(rL_|X%U
zx`(t%NUMxh@=8X&h3DziP)ZokMK<g!Ag97G2nhI0MU^+zo$_k(?ToV0b4M?!pD3@+
zE`DnUmV#zXO4);f8Ih~N@qbJ?Ahcmv1vWq(<J>&yjyG-yxLcClC|mo2-M6rAe|r9E
z`tIeM<BI`iH<Y~>H%@?@#gkVk$<_2ahR-lVDzs*0vy@6y^9^>2EIHSm%xqZHc-`F)
zrpLdBVrBs{$n;*}T;%^hXtP&1#x}dceJpLo3di?Tk1bfmlMTsMtbc%iNw;E6kpe)`
zHmnZr64^GamV?xE7lnAG_U^WmU>gdteM})<v4F{+z^l8#A%2dOW8E2)DAAfN{7B?B
zpu~@Uu0e^7P{c+kVoNCECqGMyMv6x&jg*X39w{BEGE#N7-52B~*KnDia<)K}K5dcB
z1Qe6^Q)H2rG6Sxm4}WiGk_nA)MMDlWLmQeyN=y|f9T=)`Du#%b4f+2gr4XV3jGhFY
zp<_jjp=3Wxl~C(F$w0|FHa{K03ZqT-cBH`b99C84^){JmObM|DMyv2v-h&Xn50R<*
zIW(#T!xC`A=!el1wBC~p6^Ou(#jz;Wv=i_45Jt>xA7M=h#(&g>eJKdyS9HC)2O-=K
z+i4~_4Cy4t;5iN;FT^PKBZR1icGf<~kp0}y0rv-FK%rBKc!dQ^tts<9mIEH%f`SNm
zw4GfRQ6zjrmth-ZVCXI>qR)%KXv!t+K?b-$o3+URTZDFk`~ewC)e47Af|`dQ&7yS=
z4pAfwN6apQ<A0uZK>Yy`q)28U9u%fr2z_YWhdYP@Dle^&(GMWrK5PZeRas)q0++Bv
z*X@Yo1HT#Z$0%}@W88T`SB-$*iq+dCH1YR{w{D=1j|UvK4mhaX4sbAC0}j8Zu?qx1
zLJ6iTEeZls^wU6qi|P6C+5EScm(#Q3cgH@@Vz(GwH-G&7W~1WQR<_vTyUDhI^>AuG
zL%Rd#xdBjj0PD&d{Dp7nFNs(EWrUGD0#6=+Cy&6BN8ri2!NWc}$1;$mq5A_Cf_)OI
zjNf|G3imU+UnB;|9njj3NX!q3Q!Mud#WbO-MZxu!wMauA1<Rd%i*3`+&)pQg#I0do
zN`c`6Yk&Hn6n9egU|J-60KsAbAg@C;YxflVfYj10VbI6i{ct?Ga`%ILVIETIE03P&
z`+1M+pOseLkn%Dh_xkZvonv!mQP*x`+qP}nwr$($+;P&eZQHhO+h)fdbxz*rId$qi
zRkLb+*gs(HHRm<Q7?RLvhH2Tp$`ApZqmO%)KZg4kEn2-rIS#?0JGQ}er@u-_pm*v^
zq@+Zl`HwgyP%owP|EC<6)?3RueI>{N$aCho{OgWyI3y5vLXQipA02qIr5-g!jXFW@
zQv@s<B4EV`$BzYNVtj9@9QII#G0aJOGV<sF%qw=*jx70Q@ImOOK|kZvFMBuCs|mD7
z-}lLt16OJ@cK}N=+9%4r3q4egMdP!6SR$+5j<2pU*`t<woX2>HJ7e^6sXsqJ5BSWk
zr<pmRYb0ng5f#VIZLpOS8%-QT+OGp`yz=*2m+xO8Vw5n_<?8c0+3pPysBeHH^X0oo
z)uqN(O=tU_dyeC>#^X$T4%Lx|2|<G%L6h?Tu3A#|SUv-UrF)ljnmc-1ZaBQQf=@v?
zwPJd0!8qY`4RhgqGVvj4<~EadN=EnVKm->wVRk2-y;r9rSr9srbQ|IR(l7Byt%o52
zGq`{{s83XUoK*PzskwIGje3Yu@+pBAKq?E7+!p|y&fMvG<cd~UFUh+Z1k^-1KzNC0
z=o@J1XIbyRzsFo@0~w&GX@-2lgP@W=pKWCP!V?f+)XV5+|MMLqX;Bn44M$9P7=$vr
zZC^~d3k2{qa+h*%FcMG}w<ck^XBhqO!Cn{U!B;7_?v0yn_Rv>fM?8L9dqgu<V`Z1M
zibjf{!lzEoAR#xvUsaqnZCdr|;!`0rs)T4JIknk7gECxM``jd<=8;eWx@j*}eJK8_
zwONq{0}3|E@g`OMt4oXY<oEhOKo9;4HTdE!g)d-_zN!edw{VlAY#c|UY+1(oLhP(G
z0AgY#Ng}YqaA+Z522IuqLxvkW5md<aH#hk}zqI^Cn!wLOog{d1O^vpcKOn+Z{?Ft-
z&B!kUHe+yeegJE0nOuUtdYje7wNk#redNA45#EqSG;pN!SY>jQ$!~&+?5LPIa=3hU
zS`L8dfbs)u)Cz*p&o+cS7?Onjd8f}|`aeiFjV}VIxr6QzIi@hTe9`CkL@Zrsk%=iw
zg20a9l2a{d;}h#-COFrq)53>D1fK6)?sEDPEfRhEqm2V=Oh5*YVx;-<@`H+u6M?=>
zPZHC8!r7YvYL7pMtY7Yd4C_MOlKy4GTOGjsp5ddN@P!aIOXeYl$xC}GUVQ4Z<dl+A
z*Af#+>FotsCbk)KUjCWG_GR<YjNDqb0z(h9dNNGlHt#g&Io@^$1kR#M4Jp%dn)z^z
zjLJYMx0Nm)=d$--<$S>kSl0#;*#?x~7BszhW+txUx|vy4XvysY)Kjdg75t|wLn2^B
zR_ABDquyZn6@>0Gu~kjHW*IoO%~H4WTe?1;+X10dzqUjFGyx<{`-2;ka<_9^?}`X0
z>e0&HHz=*PferGuwn_zwh~Gqx{qy(*w)SyQ+N)eDZ8k6|Lt);0hPM#8f&lACLtO7T
zW`i|SQt$n*bJX98NR~4U$(@(*T^qo}Y&J-@o+qPyh=(ig<nFBjX{fKga#VE%HIl+F
z{`4zj?`oC^g}#r`?G!m|5PLuX>sPy#I?@M`de*InJp$OR$F;aTcJ{lKc0~ivA2&5k
z`_b-|3W4c(p=Udh3wHQsnuOgKeSXXG`%M1gqDd~?;XQ}=guxvQXncl=`&__sac6Iu
zBh5xzBsv&5muXv9eLi?go|&C%Vx<XfT%-CWQnkRw6$+YFK~Dkw09@8Joi>af?<+<7
zz$A1}t3ta&*_Lnul**!qL>zzEZWXgz?a7*)n~jylh|FsXyNezE`u0dn(1YCQnru^M
z>65+5kd4D1_tVb%#(-5Fq6|QcK8BJESpU{744!p<=?B;czgmr`Ndaxd)@1LuT{9kH
z{fgG!lhTktTQzopWOJ{;LcGRLOsq-iloWR{GiYIag!<UvPQ0a8emf(DA`M%|ygH%N
zD#Sge^^MBN7JW6_wre*6<gyz<ZFS|E#;8Woms=au&w<Hc<HY(%><=LF>W_1pp+XC*
ze?EJ-+CZ=_7FvF#lwhxEv_V2}XDG-n?{&egg9N_2F*@7{UQa`xmy{*m5#E@lkI>MK
zjV0b8Iq*I$z*t}LgW<23HYI@kM`n}q{Nd+F;a%MP;ihh#{8%N^RTxxh<|Sq==mf81
z4P)g9c~xid5DXdi1`jw}OXvfF74T$8ryhJ+1qB@OYQFfe$2&jIPby#_<4m(qpY@7W
z?{vs(wrz7^*vHhnV3``Gw%tfQ&shy=)q48!F6>wd4r$SGa&x*C#QLg^>`mHg_;d5k
z*A(P(lxUa2Tje`*kvHq|P)X{4NV2KL$ZI9kZ^wgg_}_5iw*X3}2{AflBpHRz+>%~>
z9d~A(PP@L?0~#AwVr$ADo?3Wi?K`m$&K|l~bn%9GPX3ro5!6>NY`YJrMz&OWZu+>o
z^RvbtPeO}MO|;0xeO*Qx%YF~Q?L_<w3clt+J@cNj-$svIf1O#%`bwz5_VHce8=|&6
z{OU<McY5-Ge*@S`R@9dI^w}*x%>c~nezR&EF%D7f3(Yqv2B0zhFi?6ir8;1_b~k;o
z%s*)YK#)SfJ{W<~lqAaYzaN1@@4bQY>pB0C0);w1D+Qo51C^|DsOn0?rwAl3vye65
zI0lpL5}RHEk^Elc9ZRDjn%I5-iI8y*D3)sWWXZL3%LkY#!fFcrMt7=;-i+5KuTOgb
zBqn}kJb3TVdVV=B?KNAo)=!2r>W;b6S^_l&q}^5v?m#I0nxibang{Q@saR(3LYNcH
z1RJS2qUX8#mt!%%@)B4o)E0TKCN$~5MFbtAmm0n!F59U0jcE80dy)JndWdlt4<l5=
z3-ow@PYsg)8@Tx17b*=o_&=+c_l<&RX&-oelxY?U!dZa9=fAIygg=V<&x`95k~SG=
z$}b~2m{aRx3?Z50mEW3V|Ktxi>!Q=|?6IO~{djoS7lVmO;mQ+S;z?DYmt&<V7B38P
zC@#kvk*8t=W)QbS8v3>41!oW5y!_!RVFah*W#J@=XQd;gNcc20p<3;X=#>LT6zYe#
zu|VvYS^xmDnZ08ac{WtXqfWbHD}n?0dZoWUm$$uqdtaWM8`7{Wc9i6yDO{hAM+|&I
zM8?&G>yZRbf8Xx@+}zIP+IR&1<9&B(&S48He}*BAz^k(e@hrM9D3xt}ZOFrvZ{>Ra
z^l~)wt8Bda<rB1VY8~b?u1JFp>2ZjX3|Mo_Y3u|ffYT8Q_95<!uay-#&+w>2@Zk5K
zIk6Y08_e6`Y|X62XH4G4Zj9X5n|-5H(^io40(*~Wlr>7oXZ0wF-(9UN`52{#!gA#7
zGU9cyyt-^h-$@+VmqMC11C{w`zu>y_N(@9YEvv?D?j6fEy)_4IWYtY$Ydqv<D98M+
z*MbAsPQ`=#3XwSc+7wR}*~loPgIu9FlcIu~Wy6j{8)aG!Ur-E6AyA(lb;#ni3>EJL
z_5@Rqy2HHbKAs3%rOU({v#(HGXg4;VsKd=%=HFd0Hl7w$2sIYY&?boSuu<8K&dg9Q
zLYIu1jZ=M~8c1Co!y6TUQ|HPXtS%9XYq|zlN$Hj)49aR7s`KI5Y!qdW@eP%z&tqtp
zXH;Q&@o8pfz+ks0D8Pk^s_(~T)rv1d?N6>?r%<RNLH|;CMd9qJ@``XC%$(?AX*n;W
zfi6j0vjk&`9Na%9Xnd%e&a+fbYeR+Gp4|fHp7tU`()nn~jDPD?LPdeJm6-WlpsE1)
zrsD}>an8Ex6RC|(FBgf@s}?Otq~~R4wbZAYyUZ3J?Fv6u<2hsu#=>Z56K`;=Sc}J2
zp$CKYW>e(!57|9kE%~f25n=tCqf-Yyq!OLCf*1{iZ4o4H4gipwvIMxXQmkgn*p8<7
zoT|kX-KkM=oEZB7m}%K(!CQ{R+NuGr5m!6JJh`**{VYnLCqGK0=nhd+C1tk&R60f(
zy$kzzFwBPBpuz>|G6QqCh=#F(rIF^Lccj>sUFAEGjb>~%D-Z%1g}p>ve(1Q;ae2+D
z9c8V=k5Fn&X_SyJ5Zyw{hNz=;?}2l?@<~qhhB;FC-t@kPw)NNLYf(ewRuO;|cP}IE
zDtNN?wM}iCt!Ta-cXsOLt*y`Pa$z5W2rZ-t^~yb$4+Kr0hG&!IF}A8fpCr94l}TdV
z@d9j0W9EtyX&yDle8u#h`ISgjiQY4CRaF)PPrX`wMg{f~SKr1SowxBuo6(8Yn7X3s
zA3o~nG@6y+5lUQlXF86Euq!~3Lx7hAQ39$e(&zN|qlW6akDYq$u6NZ1h?DLvgEU#)
zIhnR@BC{mz74MCJ^;`qxMUF>abDQ#VN=@Lecm2)S&5YR08>-1B#J7nH^bgKWD&v|S
zXpOplrDGoVg9;dt;Iy-r>qD3|;qFF}-Iej8@QvMmhCgldPdgx?ww3@e&y*f`g9b&v
ziA!V1kvU8{86A$<^4f_DhNm8)ag{!qtJW=D>TRt2CH~NqSs_>K1+c!Y8N=S4Yx-y6
z8;h}F+(M|D>OYYNfh5gtg8^?uXv%pyMy=AnO}uT<7AD^ns6@19nHGCcBb8R6=NPoO
zho9ejkR#_$FHNZsyaNGJJy@`3O2iU>-Kk9rrPQ(#>Q1n{O6km_|70t6s=MvhM_jDZ
zw<NmCD8M=vl(FG&)sR%s#-9GzZEeC8-iNvk1Ri0~%8-ioYEy*7eP!}laH(cjs2CQ8
z?t8pECr7LXK3qztSY?e>z8XaRck|k+vp2m_FjnMVWGPRx?UMl)dzBw+(<8=<2kUWl
z&fQh*rJiYu<q|x$RBngoFH~ssw!`fi61xbY7<DVze(1f(aNtuu@#}Ri;Beq)-~pG^
z)%{8%Eu!jWH+{zEbugm7vcI6<ZqmI(v0!ouJ<2B629)+y#7~Pjng$J5N+<J3(>Hr&
zRICw3EMGqzE71YfgV~wJcJEiBEq`Q%DWvA?ui?y4)))~vAu|G^X-fq^RtRLpO-W(=
zG92Fi$xb>+7{rmd1Ff+N9CSyfqj}nm3V(-9a@$}ey%{jx0?J4WQGk^G*)HG8UA2b-
zXY#gFyNzaQ-H__j@?#yJxAC9gnp*8=&RQ25nroAG)w2P-D6^ONM`jT%-j5Kp6uS1=
z2qG@muIfrYHQsGCJ3{~Q#5_5pWsqj9NAq!#>hi_iBT7hRDyT-Ld2V`3=N$wPx8;F2
zL4bFXvC^JYg=g^oY^7NS3sl-gJicRp-Auk`X;{8k*8GV)+y=N#yb;c;DI)*xAW#~9
zurM?T3wIi_$<NSZ&6|@l4e1{Pc$$EQa0&pkS7EGg@0;tK?oXNPa?>J7WxnYVA9ZR)
z8itVVR1@l)(XaPD{F^22{><4~9`xuNNof{TchdQaq`k~r&fW+7hbA`IE2#$GyoYIN
zm?Ux{|Ge2Tv(zU?YnYsT6{EpVf;Lsb5o%iIJFeE!MhYl$gV~Ax{Uq3<1)Sck_%VQc
zFdjx1hlhc!2fBWvUuVy6L&IIfZvu!9Tyy*_LSb#T89_nB3*bQY*|D%5)WA%`z0A9C
z!ISchmWN%FrspJ&+0D%@1??#slB+#m>n*$M(WGI}ww$KQ;ovmEK7Y2tldG+Kd$JRy
zX~euG^#XNTj^7$QF#)Z(O*U^jNc?~W<QC<AsoGPis+nGo^CBXR(e4_g77&L@`{xqf
z6zN}IJ`Mk@x<MRos|1<dv`Tco2^zsGKkW9iBUhzx&3YW;7lK9TotVTzel>y@6=M3>
z#706i#&Tbx@%54Q;pb&52oLujm;JhizJSIcG`!Xg1FxeeM$D${#ovvS*`fk`E=#^m
zHJF2x2cYOO33@uo_%=wWSdeVh+<3%n5#yBYSv3FKh-Edd^|B*oJ=L=0hQW%p>6v|5
zx2QwW)rrH~${y#M8EBg!oj;R7*&uyueKhkPGDrI(RUx>CoEILla7v6P-WinWKdOOo
zJw|H<qY}oTaH_3P;u%@&-0=!Pk{`Q^Hc?}r`sAw7pS{RALgjnS{n@>Bxl6aisZQ(O
zCWW(yN}nIbEaP7)uv>h8wrBoGf}WJFo2*u`)x`Rdf<-99i@7{SUqLtWjcN8fiy0GE
z|6pWC+ZxN9tJHwCOpk)DS9H4O!F3On?sF%&TGg*0RCo{(uM!^}z8L_fIH{VmnY&U6
z^bjhbU0A%F<)}2A%q9mSbeNAcHPP+^E!HBYkWCwuv`VzGtqH#uuxx9Vf6;j+(J^LJ
zk}Gw)d{fG~8#~)Euju@SWLB@Ld60`E98#mtUnKQ&z;Oty>NOZp$4<>H1|ytaC4xW?
zN@z;QYw{gCk!y`>L$L**P^NwWt%l*9?F^HwIbBUFk4SMgz&r%fQh8!mtw*-{nluA)
z_4|2?y2=}Usi|aLW_#y8!@BpuZJBe$RfjpyQZs6<&g^ZYP3g3`2k}99xs@O;!p0Rs
zA|>DC6MAkgA4Ab+bYsr-kS>Ho{z8honM?tX{2U-fR{e{O0{{$9d>8OEwLY4kM+lNg
zEbom-g!X8}5m?bc6n5Qjzfo4~!h+~4nS|ML(Kj`4JWd#9HioRrMzh{{Z5kDdl6f9r
z_GmAp!KJHr7UI2Ijn$mmk)DR~dG6o7PBb`eu^;$coT;W18Hj^#9eX)cEjC@CD@PvK
z+H1rxHHQfvq5zxR8a@J*VGK`|cX~N(CI3p-wmqc_-bTI?rtM3Oy9#<sQ;Dox8BGcO
zx#4IU)b8sZCCIE~So@7|#WP4zb<OVlpXYdTk4i&)$@^kdPR7GOue5~DJ#O8)x9BNH
z>OXwxX-$?ScvO<uMyHJHLDqW(&bEZ~SMbRTT*g>Ymw+2Qx~4ce=>|IucwH5x@tEY-
z7MBSsx$T57Vobr=Ok6816E-|Kfa))M$jxqbSzUxzyJVSNNMuI2r_J!6!_47AvZ|<Y
z!2-pFJl;U8b+Ob<M3ny(ZX?mW?c8~2)Jt)s817N`X#6ib^2ydYgr_3MxFpbM$50ck
z*-Ppb0BGB|ElbKqKqR5H_qKlAsw#a&-pS)*G<YEab^|(yk&lTb{d`w!%O%=7*v~pN
zP`|~LN2q~R=f2UmX_(7o(e}=G%@54INdIvl35z^5E{nr0z9*Jpx3gf$a#va4{P)xg
zhRkN}QiMwCi)hr`QF(IJw|pcQrsbAeDW{eVAbK*K2)yH)dJRsx(n>wc=c^JKI*292
zwNaHLJ4RGkr(rZyl~`fZh!RQLXb34mnOBYV-s<|XUFIG$c|_C%4Wnw#Un9d6Tx8Q(
zBZ*5;;i)|Bf@Q!M7n{_#tS%Z#z7<mbrlF}6cgr-i)xVfuZeM%?SlOx>-=VANi+Rrt
z-~(z|Of(;Tb1U9GpembQOS#)`8=Uf~KaJa1&*SK-*FT)`2nJNpA6(95l*K_fG`z%*
z;H1;GVK>^I8XlG5`NwHt$n4DuE@S>nA&>o~QPtr1<OumXpWRQ)<+@L+{gAgYx80%2
z6_x)~T@lq$-+6oGQS;*U-Lx@>kYZ;XP+4f4V<cD+1WgMBt<C%I1%fsVC<+YD!Gck#
z0h5YXnah}7X(D8ae{99h)ceAV628sZX_R@|nv_Slp*+Km?cmYrcHMkwktBD{w`8sJ
zak6o>h42+=)nOF7520wLp%#cXJl2;ZNYlIHOeYpon}1?elQ%cbP-x*Ib|M=VzHMrj
z<s?US1<-<Lweg?#V;b|p54h)DUzje<pn(Y*5ONUD>IVc1+KLN#UK7=W&W|pwz|Hpk
zJ%6BzM(u51*QRUK(n@W^HRmXx;_}WsadCC~xADVol}q~osyfr4SFxXy>>F|-IE(_8
zC_3!o<m~AzDS7gb`!Q_Eawg-{9zJ04UlPO^Yk(zC9LTkxHW<MIU!iv_$-iVj>VFQ8
zfOT#`$G_f`$iZN#@O;P!emO6Ip3mMS?SnZ>$h2m54Dc(pX80;%SyBV|IBWC8i#DEn
z%46SZ;0BNLB|kQ6I3jE8vp;@(cfrC3w)ik@@|iMk4PK0N?pCJ+OIW98Ql}kA;-iv7
z^^u|jr#0hNC82JWP0vmU1Yo!($FIe~0J$_exK%Xwq))7zbIk*JC;PvDH^1H=8h^Y5
z_|s8|_el#9TDE0<fZIWMFqF}EY3c*uGQ+*nyDmYW3T}@tOSxVw1#7qklI%$+5>9M0
zRcKkRyGO@6y<X?(QBp?D@2_`echWh-s;@{?Io}mqu)a{+ghG}2tm1g#utG9-0P_Rh
z;`$VO7kbyyt8EBMk*bWIo9zkH<N`54(RXjL$*~~y+=rsXsGS2Jg`%+<zxf_P)}qt1
zLb6K}WXluCS3YVsB;qP%f5X!BejEsK*_Qi!Vw0y5@M_u`8P?hR2Jl<Fic08k)a*dV
zWV50IZJ+6)vWuUz0qI(09fUh5AVC&dhBW)fRo)y|u5C6}S@vuAF!?hYmII<T`8O<S
zClK~@t5|hN5Wrlc9_ZI-vHPd`K6pO18!KjloO1>*^q1<4Pk|AhDL!N(W0`ar5Mt5X
zR*$eXt(sJ~LR+Q;bhja7>@!`$HX-CeQ~i|E?DOsPe!x7>#GBp+vsoDmKq_YlM|HX+
zC6kx=z{blOs&$&nWmckfbd#h(;6V0W<4qogv#K<pg@^9mhL)|H?(AQUwP-Y~L<O<^
zsvV-sI|xbK;1xFdjveOdp~63Ec=ruUg4A>)>>F$&o?kkPC<Elx*%Pu6Qu#(DN$^C=
z@i;8{b?a0jqJx(Mv?)aO0N-X)TQ+P<6UHQ(jZUAl#GaAwRNd5&)+$L~nKEY4yen&h
z1mn#EfrwM^UEE=%QrzYIU(ksL7?a{?Puo}v9un++3S0b20e?FxZz-LIrDFadfuvEx
zPl!2~D9}mdvQFqy@`fHKn4tT~^L4Kq@>V`~g1EB>4}vo?Y8Gux0MNNeg`R^P6eFjL
zb$I@Ca%W8^;87yO&(i4z>^i?9A-WmF>rfIby!WUmx>hv*3bkP&U3H|;r1!-1vY}z)
z+!2$YfjC3ox1Jeu*SHY-{qJdqCWhhEXu!kp)zmy!Uqg@P+f!7$nP_~AGoJ(n%KTZf
zpb$@2UG6X5>AFh88DP-+(Kmx!nv(A|s^r%lcsx{h4aAoqh}7;X%!<?0EBLrW%Li~H
zs+*y|9nc9puBT~;CBdq-tMOJvH6{jxq#skit%}Iu{z^DVx5s#3(Z=B~kQKOV&VgH`
zm8tyf2-}pkTQyp%aQzzUELu&o+PSeSX%~Ar*Wio_K(^x<J7CC?i)TWIom;>7u)v<k
zalxiUTYMY*SIA{xwWD|I^SHz#PT?#%h`~zv+RIma41&zA$+<sy2`^%$iQ~WCK!oSq
zF8iV)c1@%gon?E|$k&1jYR;sjO{ty4@pcPdg>a5B)0BCv-=4)tIQU~DeSsoY=H94M
zgkr?EvJqr1`~b?<xp{A;_u1W4mGROFjm?Zde+(=N_`(+#P18dg_`<DWmK7SYyn8SG
z^xx+3_PU5ZSf3o!y{E}T9=(_Sc=)s55c@rN$?M7H;e3FSZS{S}MoOmhzb@b<J0_W!
zIw#*&fCTvew2vTE0_TPcv+G)OC-KbOQaWnp3R43Q6#_K#&(Byb@5v#H4ztkb2D3)f
z*VhR{cQqG~fYOYMtE|cfar36Yzw$v4?36yfI|I_HPS>nm+bqwRrW;JtD{9S^h*f^U
zR?MU=Mjgmh=q2f=gi&q_vYC@eLd*&^mg0I*Hme|dm4KOrSQxsYa0-EO4NYA6x6){|
zx9&F$gaX7*Jy{=fO}2Q(oLSxzR_V!mEp}X5dFJc|^j@{h;BCX=nysks*`rdMNMns8
z%JgJAyRW(iaYXOyZ-2U1<Q}WtCn;{v{Vt(5i#@jy0TuDB>#{drDar^?;4G-Ur6;|R
zbr4!=SNPqVtTRu9FTEsvfukB`Vf#0#F;ow+Yv|ZIw8EQo^1UzhwM8A<Zr|Yiwx)^y
zy{Sliq0rJ?*ZHZ^n3#FM0LL6iKUSM_$Z2DA&wi`^8*K3ZUuqMH=1EV#aY+WLO&?hq
z)53=_6MhFu=-;uu%aA8Q3^`N{5<Ilk27C@)*4J;XR5bv^L~rEv!TkAen%kz+!A*KL
zG=FN|QP$J_7mATJ#5`QSa==wmf)Htt1lWwQxF<jXbI}oXZ1itW=^cIAqAJm*5bD9{
z-P%vsr{@n9lcmB(-oI7BU*27xza|cgjAXimX@>%a{+z`{0f>c}?Y9HQ@x~9%42RmL
z9Qjv<tNK_K$kHDG=WBC!qMQ%xILpo~RBJjFqxt0-DjxGXOvZ97<z(jM*b)f|XxQQv
zTo`~{-jh4n_zvrdYBv71{(}?-mbzy6XT_pW@s1jc0Dj`WFk!VL=#f1SpBMjwJamP*
zNecJUyW%XrAg}<kB*2%s3*`)DNBWaZiJ<V7Me!m=jd?^TCdzonzH5>BdytsfY-ff}
zm-t@%$3L2yw+JRO_&4)A(p+4x->pH!KWSJQp|S{}AtbSdzc8b9L7gN5jMo0YM6&j{
zzR7YaXQI#-%b$9|y<<INb+xQ;5)vSsVeq1^%n)Wig%rmejfh1Ij=X3Cv)uyE!eXO8
z<(yRo-rHvTeo^m%IdDiw9TOR*viszP<ltMP1)KL^bYt6-v@Nu!_RHx7OR&LG<Yfci
zfJ%gm^^zM!7B^f8`15F``iPQ;(gZYTnF49Ll-}7W&=fZd5mh>?l}uoQHqO_w-A-p|
zxLHM>vX$f@v|yEgPjugVUKQJS>LQ?m$+|m^aw`cL9?7|-<0ZG!dBqQu%B1i09wXzQ
zsLR4;-pF?hCaO;;h*(!$)Z&j#L`DG|Tj?@Zr@1KbEPlDQT-!7nX(IWM<tG<KpFC7P
z%ADy|Lz(!DPTn{D(xbNXha>V)aM@s13r1qKrph|$ku^k5&)?dtghi4ox^eTvcGtNt
zwl_*s9Z4R4sKg(pfC<!gm>ZzjoX3;7ijPuYO{|}giHTbmD=^;{AlYW2q@)Chwq^gN
z50jnsRrO!R>(oL0JMDU4Je;2(GWP7^rTEj?y;z{4V4p1p=(_~uJZte7kux}JS|qT%
znWva4O!a`3!f`9ou5$<6Y0<Ps9*TMf*`3U{+vQ0-oX{tsPP=gYG!(v-+7A5!H?>xp
zf}QF8DpLyMY_Irl6}0kDM^ym81CpPZgY8Gdj`RlcW=KU4EObf9i-~-u;5BwZb|WJ!
zWQpv4Fwn*cJUkRBl~*e3#=QgzF!^k#gU@K1iNN-587?NH->%R|p<9WltMSJ`lrp<2
z-IU&Agfv7MtEVNCg8nml)Kj)P$OcM-*%j)t_qiJQ-#WC)tHlsFyv+cVSd*!B)Rs%C
z-G$1tJ@1VEMvEQ@W?CqjGMw$&!e=Nj8Z%T%UZisyg187ZugS(^bWe2-`<GEK^&7Oe
zlEFaQMvBSpEtqf8>Z~?QW^PNDC>eVY-C{9&s|DRq8l$TijF&WMw{sf0Le(D3{leB`
z@-ntP4>f)^Re2ftBvF7wG*WUJ40hw57oL`j@XF@XV*lBS2HHD9)$$i6;W$%7qQtgb
z7`l?Ik%n!@a9@jf#rCd&dMP)VZpb|3wJCHBHrJ9svTRH|?@f|UC_f-t0FBidm8&5)
zV)6{+u#s?^E<ZoAMtmm*Q5Z+Ls?ofRDCcHV*7>-0+qyI^{Ta|4e=*}W?P#ZNZHjA8
zv7*88@`m4j-lCuZpYhVtJWn}o&`KKUxY-uB^#PE~-i}taCe!$cZD4yUW}O2slW|oN
zte|UJoqpOn&~ESH(CxG^eDF?6qu39s6R0N)-G<O>A%w?c5ES}gvNt#M7h<J+@_5@&
zb_m#48<_oeZUI<_ZYbMrA}W-#=YVaMSj>I}rdagZn-p-*aWu?cH_Fqn-k8bxKA+3p
z)$(*?P)r;1M-bg;d1x($19)yl&_6Jr$5%Jfip;oNV~);J1g3B31m5mfikK{bAU`F2
z?nX1d(-D__^ZAo1x{Zr=Nyq<jK`*<m++<W5WwwEW%K`QnS6(B-rXEjaTQorZ@<X#1
zY%u9tb`jjB2d}HhlbM@FI3}6k=xaCMX9o;$3p|y=O!ku;%Y^xS^hL3gIM^qtF!n-O
z$UskaYxc$xAk?W~?90DT3V3FBc|e<cTU!k8Qo7i(+j=6GAT}y<r%-SrDcuxN#CJ4p
z<~}cXO8^-$9G1b-upR4&GpypmUFaiNzEw|W>m)h&-3kWiiY!aU%*R7-(OxCqR3KnZ
z)9Qa8fuZJAPLwTH0YnAa?@T~$$sE~#^MLv_((!@nzxKg^)}`y=fzGxGp@3euC_wrj
z?NEXD8MerP))jVhZZSA}9xdua=l7d?A#-@keAB*dg<+U^9WW}z{9#vN8d{i|l~Wtj
z-brCi0~QdP(~Lfr+j6?Q{AaJ->t1(vsGGh(B}6T={+oH>Ok?g5rhw!Ae?DV&t~BC$
za0<YVj=du;C(6HEL*f12CxLWX1k3Rsn@c=LngK8-CK<JpXCN|H>}LC!R#-Q}A^<Sw
zRMLV+9brP<Bc3%%K;3lQEy#L2njEU{gxm)%s9Mxml#Q#dSGh^$&OIWbUe<w?5?s~7
z0rQQ5%HH^e_YdFp7>N+;H=|1<KFd~H2|Zx)uXcIOq!CrdmMM~msIIW^>hiFvJ6d}}
zNkH|Gz~9tzevQf^vJsD^eJPtI@gmX^isDFFVni0TG!v^hX`l&EM~Dh}=g`Auzi6S<
zR$RFRl@@du0tJ-fo3IhFtZwuubQ_UXs3m7zxIz~yduFH+w5-t-d{-Q-Vdo~|@oPXk
z+kz~Vf(TtqG|>p1Hb|fz3mK(zhY0XSE0uC~vC%^&*<eEg@Se<%$NQSB%vhhV)LaV=
z0YNtGlEr_+fw}(%6tp<3!G&~UKz4Kd*r@siOAP$GKcA>0Lx)7l{+3|?9Ve)KUsp#!
zL;ok;w1ZX5Mh`_>&#{1<EPk1fL;z?CU{BPep3&A?vd#|-%t>UMkji$!_i74GEuhE9
z)==JpPqfkxQ+8p8JAhVPV+yfE);vPWB@ebc&_+5<(eV-ijbB_92U{sA$aglkH#5;B
zm)(Mo%E*cyHt9rzGz^(ftVtIJg^#4bB4%n)YRZ?~ZUx6)3@O>P*uyeGodkf;hmk8X
zcP8H0V24fPtyV~Iu0%rPS6pbMM-wDZz3_)qWR=JyuUjz_SZgcUrD;J(W--PWfWrG@
zQSkk!jBiKoG@EW0PENY-a`(D#Z<g-gP5}C^JAU^TtzOtlu7lFAakpc)A8F2i|JZq~
z_C-$<1i2^}7)hDjw-Tova|4Q#ZWD$s=eh__8BcdiU*X-9l$RN5oG?*9v7gt%|K1Ua
zNd2jfZba!JGSOa^oyig*Is;n?V3mte32+$*M<pST8y69`aAXHzAHrOLCpLnV*heu7
zqyqC@rf0F+!(i}mZlOm`4jnOlFuSKYrfp0xeJ;722<EkMiyJ+edj~wek%PUzZN$C2
zX0`nSxPP_Pemr*rd>=RSX7=3scDlQGd{}+=zT7tZzRuLYkIV&iKV5t{9{cp&>KIS!
zx?UGl|1@+NWV;K#o?)JSy<K<udTuh#3U+6HKU~`Rd+#tR%rSo0@;#;<xwut>`v|o6
z6{-be4%L<vDs6@*Qvgb-X!;kJ7d^9T4RN$q2C@2S=1~`BMo2y2^m{HX+(>Jb?$9WV
zC&9AEt-tpziq)#VONcT9<8*#ijp+C(bQ6Qw24MZZ=S2b~H{GS}G^Mx}eCM-xE*Y+P
z%O>$taPRvoL0*zBsm{8{m0Dqqvr(w)$o}hdkw@58KF|I8{S^Svv&Aax>h5Ug+L#5_
zdlCX^)&T}r^)!~RG?-GTD$Ms04{rjsoeqF~ns2MF{qXl^HstFIiqUI?)O&qu4qsuw
zhbR30SMYt{4cMI1!|U~8>AP;2KsB?1<GFqS+p!~47=mQWQl4);7#a%SSsMAgeyav-
z@c;TO_GSK<&jVD4PA@GPTE+wW2aV{e^3wga^Wg8(&Yr1Fi>dwfx^zABcH#MmNZ-h!
z)1z_$Moag{&US7qRus*Ek@t~eFgQ)r(cuuW<}I?4R%S?won7&ThNSp<0S_h-vk5fQ
z39S2hSezfo3ANOeJ`uY|L@X-Ct*#S5EL)z54paF0=SSEQ;?~41?DD5zQj^gvaov;!
zYe^T&ih`}>Qz7KjVz(}loGsX@Oj3X5fG(-ZJh~LD=fS_TFz=<N&VV@CJ=of%^At!Q
z7NFL~UV}aoU6|g6amQ;Xdec1E42t{Co_CkLuGmPj(tVu6G1g|l1TWv=wAQKJZrh=G
z(8?OUH3+CNO{o1T=VGhkMX7tK38|B>DIc{EA{-Z5`$+6;F#bX<>tIT2S?x0!F|0^&
z@)a`mLj!VlV2#L}hw4c=aSfyE4?ywm^UB4c&x_4?f2vEg0`(ebG;lTY3M-E&6L|wG
zF``rB5)-5I(kzqk;pB!BD;4^57bKPo>L$%JfFNbxMzX=n4k%&&cRO@0Gc35UmI;!D
zd?YvL&56z6--MyfYsj^+c#6YI8?Hvm=;VFwWnC=ljWsUhuR;QRR6>(NeVeFJ#l#$_
zBTtI5Rm{;DDijojMA&!6+Dzv82Lolc9BA96(GUx=lDIz|muIJy?6F(M6*-bO{npy%
z07*&SdL6$_;m;7_w&8Bz4iA-IjN>IvupJ&0g8ilVrM2ZODm*Gb$i$<+=;)G>5;Ql6
zz1czVia@&0tqn>~kpgI{Ns7j-8O<!90jkv4+~_tDA15fF$F|+zhD6C{7de{cIn&E7
z0h{1xkIDAw6-w&}IZ2kIc)w{1b*JrX0hW={=tAWkA)<Jr@B)f*yO9HvXB>tl>AjtE
zsKyczK3h&DA-}GM{S<(k5WxAxuGGiFMPk?C@VDj_oOp=WnbSWT49X(3Y4$95e5}Ea
zH9dW*H`-`R)vwS43b_?$E|7mK+$*|*ef)EC_J4mj@vzT)nRji)Dut1j9LpZs0Z0to
zCVFBW-!;}ic;BNp9Wqpt572pIYKTHcZQyI63J2u8v_&D&pV&ST+1iiggdIouB5mS*
z+`POQ@Jq?q@<YY^0{f|wgu?_JU-ZfFa~8W>6H2TaY~c|Z0(x}>58CDv7?~urh|qT-
zkGU!HIZ;(M)=rKWX$&?_)vgCX(eh~#Kcx`abIu{CYtg&h4M7@}=o;Ef_vFfD=o%9^
z7OIy^`rxT%#fQSn1g3V^pxR$;inBGj`IZntSbqchQ%8qcAJu0cj!UR(G1zZpkHxdF
zYxmf>fm4oE#r{(ajVe#)M;SL<T2naW0)4Fs2Ss<v7CNmV{FYM4hHh^FxlNtK4y*}z
zt*{-AFk1{!=4hZ`3@88x@*f+*Cm+J6Uj*z@a!(Hq)VX+F%tM6Op$X%McgoC0ZwLu6
zzy-s1j5b6C(dkgvtA9IEeqz@agkS(K>`SNb-!?J$PHqoI`=Ea_LB-_oB?ZgNxz<B2
zW~p6SpO5eRl*hX+r79QTCijVmk&S$q`C<Ha7nN~uO%yjk?7>-9%6Sz}xua0+GXYCv
z{q8d;Do(`qb#cN{I;nxxH0(q)TX%g+8{9}Dj56pSOixMAq%8K{8a+7J=?c+r)k-W)
zvIl=oS7r=0khTt7Xnv*S2`n9>b3K7eP_MG7KV5ct-Ee!O6e9?L(<_GR=vC_F91Z!K
znXN`xT9gUfYQQnE|Ld^9L+p^&=ZI3?5yi4>58<+<cCZh&rPcq`fuY3TG~|CJc>9BX
z+TX<_NRv)6kvd%c-Kq4PrzPv`HlyDl5X*!0l%+HriR^ObhHQ!yjK)O1UsQzkV4JX2
zYNlPx3dFp%tJH9g-n_(;YzoO_se0T_a>=h?xZvX{?DfWV)k_WRSv8Lj%ZGp*A`a~T
zwq%@{fzi{>J%q;qzXZO&ewg}+SPt9N^wm+pj}k<P{GJ#UM<R<P;8Dr+wPu5F%G$n#
zj%gE8!;bBR-qdqxhT9cNEY815e>a2dTS_YG{RXvq#6fS9D|v`kqf>53_3B(~&r_Xp
zz3@~u3+WytCw(~}?_cb-xeC>YCH0Jk{jC05xf3~+Y1WJkKxY44lb+y;-%<u%+(yQ1
z0DA#aR%I>7`l^}OpZX=gO;J<RN6t^}+ea<D#8DOkS1D~YM|06eEZz9hjpQMnH;@;R
z7OjELO+^SsO9s|k$<wZCqs!ZNtZo2Jxgxi_Ax(`ZR6qfaaLFW9Py>swr!PhVIGoV`
zck&u{)u=fPh(X@hKgypZkg`})s>Tp0tSyPVhynQi!2eAxMC_Ia`~SRsh~jV>g+4mZ
z181-d<$dG*5Umfze6KRz(q^@JPxhc~o$t_e6s<R@eMy!tr7w|Y>N?}3Yc17gVK-RV
zb2%2(so@w_{!LIRO867jlqBi^<syy3f2JBly{MA|svPn17OG#LJqVHq61a|lQ`a<w
zxhzvx^GuZaqy*09%;kEs33Ebx9{l);+g$Ndm<}N3<9U%qyGgBZn*V6vgTa5*#k;gx
zhr@@JnaaaZM-|4LLb(X-1I5)f^c*p@$wDB<1af9?LUbbCOF_(7c1x01C9)D|sG-n{
z9?uZ~Mm<hh?w{rGijWiAo&Ux)h8|5waU>}-WMHE=M-W(~t2GI-zp59q7nPKtPfjDX
z&fP5;mxQ=DA(`jEQBF78%~dcSOg(8%QOUts-4!_v>>ar!nnfSRn)!W6l~&ut(WuzC
zAo?a!)(m|=%4%SzcCQ#XndV?148?#Jux&a6FeY*0%Yyz{313>lYhP^ntquUCD^Kxt
zlLJXijho%rt>ehx;EjSu&gsFJSe{>p9-ZfEF7!=UIEEjT0~|)b!DhNmj(B-zvR=wV
zdf19!eJy$WuwOKOHCyH6{L3(Ad<Q*w8^l$<D{`hjS|QV#fqYO=Oa`@D>6hAjsu0!%
zxBzDtUE*!!_NCfW&u!gOm70pGIjkqf{xnLLm2G*@FRj$}qiy}z2%fXYDmKkBM%h^e
zRe6{aLX*vffe|YrV+W_0I4`Wzb`SYlSHOxc%HIl!B-e!<8f)fLz2q3Uwj6x5(KymK
zI-9Apu8cuyn|Z;OviKIfr-RyJ%16Hdi7)u`)<+?$oFrxTlYc<++?Vf&6v|xkAFVXz
z*kJThwxk3w5`B5!f34-MQNkkH?}jAmQ3)cuH4PQ;*=$+OY9&O<prl;s4Q5rRNFtcD
zRD`lX;ewo<1>C7q*GXpz=DdMUAz+w}7&%b$`gsl50g9q+REizOx`}h;`RvmGh%NVJ
z47Z($v3gy1ACk2-cX%dPTfZ~zoXF_&Rm9v)ewm4OHcl?QHL!yHdaLr}IkLigrd-O%
z2AfT5NlH-P9_RjUz2Tyvp@3qlY>SwA){i`yYH&;&_-WwUY4$ps)9Efi^PG9aQ|f9E
z%;bH*#PgUDaLDJ#J+lmf`TAG`wgMw9pRIBGJ?b5*i;dI6d1}3=;hTz~cuUy+`R6xV
z0Fhd<>pXG>N|U!rN33;g*UYaxy}_?Aj2?@4AL(E8Da_C2b0LA;rGTJiMn#mSDl@dS
z;4O@7G2fs24-od7(K&+oE#70_NpDn$jd^SR97Kcn<m{P0<{&a~Z%aUCEhv;#uJo2`
zFH4oUxom~_%i;XiL5HGY=ifp@iiH5~MCF>lexwN-{pl1t+U~ETu=s(Bl@}8Cgau9A
zm<$UK-i{C51r`Cw&A_uN1`cI*@rUYdWU^<(qW4)%Ura@9*o<)x<y*a2LsGEa`iH3}
z4Mm|>a0e~c=Qh*>*aHCegxGmG{TvBWj@7{rGDh`)!VCAMYl(z?Iv<C*3kTS?q*?D{
zwrQ%#hR&oa#n7dHlXbj~Zy_Z?LUb2)?`BoqfhwL`iJMU14P)s!4W)kzMw7=Sp=huf
z{lMi6BbBpa9K;%NOwG08D-BdVPOHKwO+)V@UZ)Kb(VXWzu+0IXF=4Dr;U3_-$n6?x
zNlQQte7?soIULv4I%)Y#Pg3f_x=1vmyyaN=+tycHBN~+2Dy47u4?D@C(v;mt*MEHa
zbF-#&t~w*C>=3tc0;{zY(7`3hv!z*Bu>cP@DgqrZXny~giJs#`4M6S}nb!{hqBvn-
zv94BvPFV0ha_azozrS-ojM2AGL?o_*K5K~4+6YRbmb;TG(eFEz!9Ov*q>lTAghFD&
zz~DfEjodpFt$}kxpbTWuqw!97U#Su;e}p%0H}IGti=NtrIMuwz(XOoY?{o#VI&jj)
zS=c!i+r&aoOY6A$W@QYQs=f^fN3_3j2zucV^Kf#Nj{pE_tLH`ff!|R<YxAO;VTJPo
zMi#jpV~~22Bp8ss%+_)R%?y&3Sm2Uz84^WXWHqEY&F)ar>M-u*y^r_k7a95BafL|}
zyYLu0OuYpXAi_FE_=13c3QN6*#c?~n#E(!s7ASESZ2M_qv}F>sBM?5_0{VzAOWIpY
zR?}N6M$&+*s4WKCHPwwgMhP_Szie)F7RrQ@+Jg9@l$KGqKzw|3gvQEPL1H&JbhM{!
zG64sEnkyH!M78XeCxbnM5p>IwWEt#|U$`?l;SFNvqY5lE@ND1WyW`LJ=2^t776zN3
zmM@aaFwK|-vzmqb%}(1khdwx;YD0UF<=~<J4#NS(wNwVU>b!wHh?V6-MJ5{yJugy?
zc=3)ck>}2&PrbNy&bA#*^*u)tx|UZt9i*)!eS6ulasg1P?9jS4C_@#vHOnrdQy%{c
z%oKj7Yu#0NCpUL>UZTv@oK)2ml`3TF<KtaTX>U{W3?|w1-wcfV3@?anN!ujL<5RPS
z>9_;7+$R+@&8zL|jo5fWTa4HwVYQ-9`yTrosjcFYsd0Sph6)-|TX8OFszRUM!+bdF
zic3}0;`=HSH|e4c&Xsy%%{=J}VUS6oM?&7DS5zagdI#nzb7`Q{6GH^H_#uoM2rLTr
zD0B(E9dbP1;o-F_Mj{(L?lP&>EjDOg%QFBbaAwgb$#-jUQZoi0f9nw>I1cnrmQ;{V
z=&Dl`WyjFNI}Kh8FPiS;-czRiG^4P9i)nV`@qFpoa`TSzOBywlSh1f=Xt#Ik{V!dY
zFH@gY^NKPU4QcO|A|XO4!E?_BOOx)(8P7nrti}*mq89bd_a05_JZaU1q`qV_o7@3;
z2yy&!Ds;DSV`2svI^XV=fA5s?2e=e%x=5e5vJ~Ugje=x?_1<l_Zmt}@1TPwuBNNs?
zd=rx6s2KcK`&&7#$h#m<C5lSL|FUOVsLIB}2pTSJHjF_rjX3v7tf1wmNED5=k9Uqb
z6l>fz+VId&7WH~%;H^Zz3bs@glwJYm{TpUHYAj78*41z_T{;O6W1Q7@XQmto5<lZI
z@1T~GT2<aFKFXOx8jH5hbHs0{t=r4rYf?8@Wf)Cyb}eg<_*^e;>!u|HqWUCu`;0Gp
z%>4gZ4=QezG04pkZ=2v|$c^w*T%F7lItyP?G}>r@lVr$QYi)Pw<+n<7cGv<Qbp)AG
z7|yvH^-I|m9FF+y`|;-Vb;aE=|9H03%E*%fv6`jBTk#upUZ*O{Cn0p8;<`Q(>V^{g
zG-GxXdFP{2iN}r34BeGzH63)VKWu5>!|KU$<`U(^FEH-5PcZmAJbqZ8%oq$g-xh1Q
zVb=OF@M`?V4K)e<p<~<*KiB~}jf%8)sH^*KjzsYUVkmo7rz0*|T(U4$f`On}w>xoG
zW+ztDPXiva2gTC->O}D_TKs_%e^+A`xIN9ee_bQGc@um!mJ2RC(OZ9=CO(>CE**E$
zoX5)X;&I&6TWun?Y<~4)$lZl3zN89%n%&LSA$~&+>bDdA-{@-^R`3tHqd}eqHH{}&
zSR2%=Brz>DSeQjzPf{<^mD*YR)1&=r$RJS^Q3O$hcEQ*Yy+FKByfALQ-Bk~y&)$Le
z>>lC>Uz0`&$X-Megm5wZTj1Mwtf@cS8`VEdHw`{SSOnk+RhM7PbHI7vu;d6XN<~%W
zd^T3DRZ>n=w-US%D;=Rs^)9Du3G=sR@Xoamtvy;VsUjU@;(afVQFQI$acO*o)*|Gp
zW3kLh9?nn#iUC`CD-KD!Hb0|P+aQGvVXm-2$nTrG13|FNmkCOcQCCd6w7)^?*sCiw
zf8R{eN)#Zy7$$x3f&9RQ*~k^dCa10G^RRRyGN_FIJ9jZP8D*BToRBy!yyt)<{Sp~M
z&>{^M^>8q&o+Rn*2W+k!4c!xc=8ny8xmcMIO%Ai|P<9!R1J`_5i{Zf2RB452Q9sMv
z^}}p1Nwub?sFs*QaUjZRY`nBygtJPpAZqv50|5k<!6PpfrahEM9%l#7W{0CszhlR3
z1t}m<mmNZ%kAthnbx>~$_05G6)Ha7ih2vk*Xf<UlgH%;%mdTL?#}o&Dm2k}NJMay9
zxy6id3JTWXGAPD+IDxp)hp}#4n|q6i(swy4gl+E6R=6Ko!DN+kx4SXPMtgbVG}~k3
z69S4%tVm^~mC`DB2!n18;ZjTaTnnmZbC)1C7d5*Q)U!)?l2XhBGdArM<Iv<}mo7fN
zI7aM41+CuVwU@BAilj(dRyMks?QNMnBl`6>F@n5RwyeAQ3=>A#NvIQYxkhI^3|L&6
zzswGL3>h#-nd;2zrbR%YX(5KZ^NHOGc>vbGZ6_Ydlc1|ISPIWbK2r(h+Uw-Qj&rF6
zv!X}hA3FD|JAh9(I1F>mxw;k+y;TNr=s=)uvZ0a!rli;TGo4pE(NnX{Mk@@eGjUw~
z+V|NX)mF8AHz)>lt2yu=lX*Fs;_D&e8`5}O^-oUGDD<AT-CE9QI0;2suRC_qO96+M
zT_nWBiv^1r4>0D@rvu<X=`YNw1h%<asJcET+fkgqd#Gu`nBn(#vE6*pEi@%$kdzVE
zxOf-EsCEmZvM!%$r6tg%b{&C_-7Azr=e5LtEyD;YtVIywzd}Uph)N3QLf0`|TCvb5
zY1dUdR#*A;M&{PWwNcO5g>FBiIi^X63o`@K=za1YZ1`GKRebbxY%*qxL1?EOa8J_2
z55Q3GkhsS~X$naJ@?ugYku+O;4Ljzoot5X-!+%&)qAke$^g+ohvr!-qrqRP@dMHer
zZS_`G1*m)WCsNBydW9aDOJ3y)gl}(#uM|mr9RJFjGn4M-+yIto@+S}4kX^;1v;qKw
zQQ-}p<!p-DNf7yF8JQ!f#xb-gTM)g`s~G`$yF!G#pbcAi7?wIkQjRjq9@rrx!fNiY
zipooela*Qx-PEbjiN~q-^ZHb^(_W{!{8Xyfd{nCCENhqb-Q4nI#V8c0Wnm6pC@E#R
z3;P{TjKnc<c(xhLlS^UM%O(;VsC57#)4{J~?sgJQd5p0QymIaI%$aHuig9@kLOsy9
zt5S&tKJloZJRg2eiz}fzNle$`BWsIY!lzrX;TbZle-0^@&C)Iq%3sZr$)`Brjx*~B
zXs8ZlpR(}2@bze}GIgp7xCD4#G=oFK@zkz7f56*IU0oj#5dtG1w3f;-L0AjAWG?Qh
z96c#sA8zlFm8CRQ1#y@!9;Ciabi?MV{`0T<y0Hs!Hd4#f>pVq<yWX0AL%N5|c!JZC
znwO+aNsH6_E(eCS1cY)LjfmF|HGch1#J#zi7o29A;%{Mj;N7&G-@=%HzaLj|eAnQ?
zxtYEsO^6>8U^2es5!_z%RwFa0)7L2R<2$l<i7V7~Qd`6E!xca+9_sF5{SQ;~5oqdL
zah<<9R`Jr&Y)8;f&)eQ5SfwD(gq+o<>x0T7A@HJhqb<SpCzG;k;orbd1y_}U-R?OS
z&fjZeTaOf)8lgqTwOVNaycN(IQ^abU5K&c2!GZ=@C+VuCgy>5`iFIM`H$mD)37}c8
zZo8J03^%{^TWni&)Ucvp0Ht0mT|QXM&J~w9qQ?_xd5tb-B!AeBEXHUY)=er@60<3-
z^yD7uupfI_Vx(v38*`48Dmyr{xUPRuAvY(~JelS}++(2q9{`O&a=)PL_J$7*>JD%9
zOsn^w=y4;FXuP+<ia|>O0YA45j6t{o1WD8fg_mKoK^2!#O$HN}EN2BNml}>i6IPsA
zodxu*8@>(ZG?<=KAEy8RFqA@4Xo=NnyXKIU&WNByaeZV+B`^QE$jgtX>)X>WcchCI
zPAdzSw?8k6Z%F=ENxo?G#VREXs>|D7mw!A<35GvcmxPW%DSz&X-Y~tBL}3ITTk(VH
z8b*8h=i6^jGBI4Z0~Ff$i?Jk_hs+&w5l0{%QS%HseCwCUmu7wt;#~@V>F4Sho{jE-
zF3$GL-Kj;HI1q}kpH(uS^vd7&mv2U@G_Jg*v6kfv3I)&IkHy@A$}7O>H{{<YeXD5(
zx=nx6YhGErdVh`UNGl|!(U53NTCol?nCZ3MTs5++S))Ct<TuxcV1*^>sd&0FuaevI
zl4Ko^5;EKK5?`c|>NU+9B~<Ibm6!IDY@&>ANlyBZb&xW`!eAY7#BEz3aKhq}DHPSq
zq((persQBmfoaUkps38O`rD-PV3w?n<>lHmK08T#LVp$k53XiFNC!pzwPD%TplZVe
z>}O6_-Y;lSy65JAkO7Rqk#gd7olo*9M5wgp?09kQ+k86K=faHJjsso_BXOXjIyeOG
zOdl==8_7>aT9vP?tTV4mRzb!zYt6I}*RZbH9HzFz3TYJ94(r@;Z&rao1B+TTg@YW@
zZT5*lB7e7L4JzH|HK;6$>G?%Z&{8O9!@mpuIHiH;mw-9wT(dm4Q;TtejP1wXu@z;t
z7#YgK!06JP?KUU&N?6YG-58jSDCZ2AToqw*3EXtwT{QxSIs?$p+C}M2=kmdP%rL@!
zAV8@(P7jokd)H-El?{F}k&JP&3#4ClSe>ryN`EKdgJ>!d-3Yc=bQd;lPI!jDjhArw
zn=#Mwoj-o8>oXfEg3vU3Ra%BiOqmj82WXXmpHU#Z^u-`#<xAJe5F$Vc0Pt4jfjmPJ
z0h!r5(9Q9a?l0=O()MkZOmrkl(h47#(@oTnPoG;F{#INpgRec1z~IY(>(=;sFHQsg
zMStV-R^+dg=S>h{`1;ZP`t|MUzeD)Y)(CCY<zIh2g&U;Uhpp%QB)p=_-~Yb+`Sg1~
zYSK~v9r1g#18I!btsU?zj4w4~)_{kqG;&_9S*TO6)0h@cJXQH8T5EIVkJ+d|qEHNN
zI&ReW@M+92SU8i*Fb3#p)aFXGeJ_{R41b|kA=A$S`zbunz9vl|RubRpMxGEq%z^rp
zOf-jl(g$QxD2x$aV^2|Ps@r)UtQS^|dQr>sI~3jyQh*fQhWaA?-j;u<l$K3?HT7k0
zV!ZEl0Rmv>>aPd3C9;P&ZKzMsw)$I_C$I%s874*lD9~a%QxVrs4tq|M7p-PU&VNqE
zj#PC~P*u1Y#+1Of21!FTKaQHTrd_p~wd{#(L&FFWCPOW`InsnHbuCIx;g!~6Op!cP
z@Mon$be4IsBQM)4@La|7t&S>|Np=9$Ll>bz2%7*GIRbmVB&jSLZFc2NhONiLFbXxY
z%IQ4;^(621nHr_=ebzz3bd%pMLKa<(T6k~!qHkS;9YWefo|XeGhuMCPm%NfeDSwj8
zDa!$8gmZQnrL7L?xD0cguHt7E`!ZET0cMhKt_%W&lBs~og=3~RDYo4D&Q-y|TrxrG
z6_zq|f{D|J)aW)=Bs=A$7fKD9siM-R*I5)0ysoAGM7*vQqXG2*UMKIkAo^K&UDquG
zO2fH?F)8IuyJN3fz_5@n0_9|Y!GG9SfpX$%->L$jjtMWhU2rox4fGTz{%9rb%o|`+
zc{XB8J;ftnGZg_n4=Z4k?V3~dFu$3>S9Ii5W&^baOz!DmW#b{Jv`u=Y&#!{v1Gy7|
zWQhsy{09m$l&@C9K|yjt=HkhGxF8=v?Sv>CK+TJGrTzR35qrdfbOu4<L4O1ZpacN+
z{8Uc1p7}$Da6M(!YTcYSpnI<t=hfTua-g!tY|NXbw}L^cN+CTt(H6R#c5F%*`>kBm
z99-nU(`?}>oxIE9rk+kTy6u!rMj#CeiO|A^F=E!U#guN1sVqj|@~oblp^@qGGOmd5
zwhEyMLgckgtT&}eS^=75>VKP%nL!Pq9>zTKE@j0by%(13FyaB%kzP2Z2Fe4j1EJ+S
zP?mD#j6{H((4;<4BH$cfq!o#b+*Jk7g4^V~O=G)$IzK^XZa>7of9+e*O{<w2+HBbq
zq`ys&xdZvhZYSrf$73M9Ir;%xU?v3xT(e*c+u=+r&j<^8%1;E^B!9VpbZ3zEX)uYF
zs|JqI9hg*xD@?NuCJ%r|bRdA>CxFLp<^$ldqxb;u*z>Rr9{c%MG7#3zQFSnN)$PNo
zqxf87Nq9e&46mYq@PuG9^!qr$VlXSE4+<7ha&h)Ka*ZiI$o3&<88(3MXAN3{ll@z3
zr%pi2>FFc+kt17l*MF9uL(q~zsj@xgQE15-QF&aPcL!QBx^Tg^R_D#qI{+=GCo{B6
zbnk$c%nN`Sw0aO)9yh?lFjI4E*F8TGWvUuwCabzNR<nXKldRTJ<`S+%)G@@>ZN0en
z2bNGK;0cszcx{sU86Lq@9Oo3Menk1YL&dtc!%C<1z|jJaxJvld!x%@rFtq+~kTW$L
z97L&RdSUi)5)DlS!w}<~L{TyyK)Vth?ImAXL%9?67RrtEyC}DvmzvS*7j(BWVz)19
z&`Z9Oy-L>WmnoJ(FadFwT9!d0f4i9v0Pl|C1AuqW!#42l=U>@Q_SIe6<ATgx+m91u
zhVI7%nY*?hEyzcpb%2&$ng5%vZO6}vu5CwovTM6E@3F4!`n*4{YujtJht}`1xU*f`
zj)|Y%wFTsyVbogJ7K}R6wGBPWam+=PrTmcvn%(!hbC`sqF1UV%qh^gHf0V|7zX$Ms
zILQFGwBkM2=5S|-2Fe4j?YYJX;dMa<XElZ#VAyUxQDbO*MB?w$7<$7}@*k+NxLHa0
z6^)J3<14UH4^+=LoAAn=^-5vU8$rh^uG)sQ-7B~2-R!=Km^$cv74fdV`TG!jNbs8a
zH6BSS9=T2*r1OagKA<~NfBu18D#}@5o&8z6%5PO6h<SHsTEum7Wc|HDOt7N>FNkcn
zs4v>FqBwn%{-O;lO51p5qS-@sMvm#2%=ul?xuEGd*erpZa1MKlM@pcyue|4Br37MC
z`2N|4_eeJdPz)>g2hqG@TXFNr-rcjahfWiP7t~<;u|d!KQ+IaWe<>#;Wj8{TUi<ym
zW9Z-7*S&IP=YfNMC=L6+8+hK?hpfW!z~z-zr?+Ykn-Ua9C5-L#YqeLhKIV4MsV&Ol
zd%KgfYGAqQh~I(z%s|2p?9S%6gY%U@HyB3y&OaW^uJUVjU`kimsKov>BgdyLGmr1s
z%1J0l0*eBzwyx&Ue<M<$vQZbhE&kgs#d*N!Il0++2Xcg<{=k7!eyRWxlRmCGmTB@6
z9lY5F_)2I7n=N~L=TMQW7T2$d?0jBmUehe9!vE^=;AVwtvea~?KaD@0v&i$=h1P36
z>O4isEYX=J?<h{OVdgbPi%R{oe1orqU&CJctV=WOC=O?oXPkqOLoOkbaa|R2%W!oA
zZV|(KD+hg<icOf0BFs#5N13*5)oPxjd4067y=B~HVv37aI#2gsXQ*5|Y>7+t!S(q6
z6)-+R{oL=|FaIxrX<airzP#5n#;1LA$hHSKjGz7o^<JKjmtnI(6_+!1855U>nL)$>
zD7Q?SLBs|Gr?m1vmtnI(6_?RH7!;R1p+UfZ2oLz<FYiA*NhY}bRw|Vd-syvNS`aNv
z<w=@uS6!`*(B2KJ6AzkjX&-rT^h6WI2J;hm26fW;nLfv7^bY8vUB=cOw8EGpMt&za
zX&v(KEAQo129IeC$!U?+!kO_x274`M-RhpRB-=<eQvo*KOXfnW!iCCLT*X|`Tl1WM
zxi`??v8*cRyvH{0Ukdaj3C61@Itpb`-l5)wf}N!vH4y*56mB97=B4mANPpsW@vZ9K
zR%ji4$HLGOiDntVCecOEx(Ls1Kw?r{FRG~enaZ^lpsRyc8nXd5B1kx^n-*FlZN}#n
zd#0$Qz-tY$St~IdRl`+vnFQhci-*O30A!so#t)EKqS3&T0~+Rn?^R%4%X`dgFk~*P
z!A7~9k5GsLCS^Gmq5~qu^rXnkT*y_pYhJb$E9^vW*-uu;5UcM^5G==}!j>C+E>p7R
za~WrEEXYD~90k(<Jp}~{^L465Q`Lq_7#j{&lF<ZQkFnd>n6ga+aX|o@5mS<XiSA=>
zaz<MWep}b51=tO!RR^Zkc+p6@eAalc!EBCo{jj_KSF9jP<a~oPltG~YW^{*5Zk2JA
z31`}xY&+RYC{|Xf4ArtWX$nq&#F;IuIH|%6S?!0>EuOS<LR&pzh2rBBQ<kiW?bdiF
zf(bR?_-y5g9?6;UIyx#uCk54iO%>vK-?AyrC{W{0bK1tItZ-6~yC_t5`8iprYUHsN
zeo>&CK_i>oq=AAqfu7ND+3{o$3f&CX>ozWD1#vdRMP7pdQ<B}p?7eW{M;Ln+h+Qnq
z782zgD_6NKA+$s_*c-G&b#bZ_P4-JvimPmSF7Lw-VG;1UNXRQY%qri1V?#+>^CpNT
z*W}#E^1dQUY30Stu9a<Rnk1i+ec-D|+^;3`N)50$)mt>zYJ4-djHR-h>ZVf`ua$|F
z*K&v5JLGp!&(o8Lsi-GENro`p+3rrxR?EDmhYEh*m^`^WRa*luR~SA9lS-Ol=Ob0o
z3;4V!L^%_k`G8oo*Mbs%z`$FnZUqsn{DDHOvEI_J4V0t)raE@{%}-y}!$+|tEd|id
zVoSCuZ%irEmg{A1vke5{L03J<#?ekWH73V&Q}CIJOxBB@3POO~UKf3ktJjF=nm*R%
zrTdYB_i_&_^>$SusZMrbVCv;kD`08}OzOc_O{Y;8wX;z-^z&tZxO;nl`fKLb+9O%E
zKK%OUQ+%Pp2urjeia@0N{oCR9ryobC!*oHm<B_C!X)QTRr+J&Y6HN)kPveo>1(_X$
z!=@CPz}G*?Cc7s;HI@^|PANeT6u@h|`k@KgF>RL-2Gz?6YWE~rc#w;H27>O@h(-B{
zx2AzA7}unRNPa?pj|OPw?Ou_m>FC;Mv)UGZD6B~t<2lEmWlP$t2{l#g*4lQeR_8=v
za_iU+QH<vf0aG2Iw?i(kD%}R}zNQkGpf|$@BP$G+542Tq6LL~*5>%PWoTuP5b$S&`
zQSybW20PRh+{4azAG)JOUKwM8@7guBI?7MRYRfY{JiF9?sWMW`T?yahF5Y8wM!LK*
zD!bF{=|<AfK?{_Aw1F%8);$X|u?sLbLVsZL(@G1c^bN`H_Lv@9IgMjX2P=ZI+cEu`
z7!D`^{I`tZJq3@7;XRhm#_%56Rt#5~-Am_|sZ8x^DPr4@n=6%tsJkD}Z$tE;*sh!r
z!RY^E`+qopwhuON=h!|#e_(7^v4ujoA-3<GCo1I#`^wo{J6mjRs9If}ZSQ+anYjDe
z^maafHO-u3ZDMQWq{D<oO?Fl7yXRt;HaGLAj=tt3P1|<ToN&G{y=IDY@VY(iDmtLP
z%+c?+X*SJDme9x|KbHf@o!LhaA>hEwS86gbHktB&Yc-h|r<A_)!s`T>3lsc6Vd5cK
z%7FRqg-I@d1}tiB1iM8ES+*i$7$x;~Gv^q;chGjm`DMoWSP+d|lrJ65x6`1a9nJK<
zay*L^QNgKk^?i;?b3=Xa5(#5-0!<P)k^rAOt<O#<Z;YzQ-1Ee`?e3dXRnwlbjVGLr
zN5Aub%u}H2*FlE1JF&YJzwN812Jhx6o3~{9eiUl?79&ADtwoLPIV#l~@ho(JT&mGq
zvv9A=#ZbS@A>C6|1tT4}eYR;3M8KI`9#}0Qb;8!`PmF;Pv{3fnmY8d^yOWSfawNBo
zhp&{6!BGtSO4%5^qO4_ivvE5jIE{Ytfs9OlzKA38g@imMS29MNk^7mn#7TCfVI?hb
zA|3I6`tZLU1f&%&O=D#kIyMa@Oi8ddVXmH#-68?%7|8_S3N3u5EK)WIX~Kvlem<G_
z!CAw24G(K(tzr?TW?Dr#U`T`;nOR5}x+h!w#QBpR5$N^<G<$WWr(?A>cZ5%sBcLgN
zzrgf;iaF+w^vc&s{3Vxvo-!JLVW8^?`l4Cax19O-ZaEW|Q~N9ODZ}CPpx=hcN`9Np
z^6T0B?CXGSigzbbdDXeSo8wrVIJMsjly!RNs{xxb`KYIrP?AQGHOz4u?7m2y!pRhF
zYHA!Kj+NVGh6%#ctAV(c&*A_>IvQ<%&)a?X>2c!{2JbeQ0-{m|HR8XaNZfNKlvIXN
z2{T}LV7Ud9VU)IyJb0isUsL>f%(+AC*A#r4I-<vYMH$g=xI+~&P_SJ@4A}2_oSR$R
zw>)>S@>`Y{q#NnnzH&ZQU64YJA4U40;v#yCpPBj{YYUCB*%(Fk(b__jf2%MK0NX4r
z#$Spe{{ZF7Wc-(5vq2V@GM)hxmnEt}zXCuwx8|xrApw8BBVCGM<Hva8-`?Jz{+a3u
zh#X^2zyJLde?Tz8x^MzASZz=L{&D*A>1Pgg95xdJ(pqDB2COiK>?OhJ7??BSWR+jC
zuH=O~ZxJub)9@#WwOjdR6c)V|pg3~)60i9<8c<j@A$Rb>!z3um8#+vBfLZ)1W*4**
z{0s3xV-|l;5~C|3#KFPi^Hb*^;=_#p%iH<c%K8MpmM03rM1(4US{N?~v3z$av>z3e
z4qkZNp{wXMv|Gv)6*ACnCHYg3bLNOq8yGm2n28fRk+mDcw#~wl@uTu&O)$YG8HPl-
zk(q^rp<7wwM>_$UBS>jr6k1tmSR~hYVrXKjvKoK=oa--^OC_opuLPj}UXc9H3lgrU
zI}{{CaLNH^cYOMIaRM>fOpI=!I58F}ADeGhoiO|lC{GxM2i7MBDV`{Qpg!S!l~GQh
z!EwfImU77kXLSCqu0x{1iXn`o=Zj(bk)_>+tXG$OCKTY1@E3*djZhOtUT?OVlL}{}
zU)Fz(y{YJ6@rD5e)h@hvQ%VHD_2P}+cNB9Nh(!m-EVmYP$BQ`UgazPO*Q&T&_OT70
zDB>X2CSFfSSSaG8rljon)2Tr-L%mdkVYP#$m^&XJ&o$J?<xiz(v`SbHM^f|<A7hHj
zbB^X(zK`=Y`IV;vUm2&iva*oG3)UQspP_%G=@b>z(Cm&%(6XZ@#6^4y>3T%*w#!bW
ziiT_uzgc!+__rIbotY(0zdVQjz|4XaPK~vX?0j_yLO-bQMQMa__G^dA9_x-!*>AW<
zsAMF--FmndUV9853$MExBNZcL2pr?5qAr!R!g=>S>N0%<cx*hO@F|oEpQu!|)@grG
z_H8QF1iFqZWVyRxGGhw*-XPe>XSI+e^^jCKpI@nvRVLnfFt%1WbN4b<Qy{T!m`Q#}
zvOU}K(ia0r+z{m|DB++rq#4rj&}x*B9yzo-kI2B1b5Q}ou&lS~yTWK4t}@8-c3cuw
zt{%uR%%dbx<=tb6!w7S&wy=+L8w7t)d}5H;+;!G+VwwnDl1sMtnj=C_%ZKK!Ks3d$
z#xUJmmVrXGLyqMGvaADj2Fbh~AI+CxJLmj%SI(7^0?b6dy86;FA5sW{O5!Q^S(eS0
zK0&WR0b$PY9#mav(Bu?qiKXlupQ=f9bSAAQik@%ZHM&%p9hAV<9CI|kJVJjvDiqqf
zDed)|<)NGluuVM1lsu@YW%R>YE|s;<R;H=*Yc1O9=}$cy%E~z&Yib<qG2YqZY8S+?
zXebJjS&^BmoiQi391LPp=SLa<rHqin;jyx#%0m}QT`8C?KW8W_=B!|*X-MUFY!R$1
z`$jRg?ycqd3~cqnT7lc@8G?VZByZ_BM~n4RIoqwj4=0W#f)(pSmg@$SZ>`s23G1xT
z8aPD}wsup?u2LLJj7fT$XN#QH<Oq$6tdqn^wLm{PEnSi?p&)K17SqdTRUXaeDAdZJ
zdI)EwURw5DY@f6j&MFRJXFak=7u6m*T22pjVWb9u+HXD7VXgfnS(|^>8ryRDkuR|m
zHf>t#Z0onCx!f=DRo`~`;u8}+Rw~T2rdjkHAJVXdbLQ4h8sc0TCh{2gaijNtOz4Mw
z!zq2^u_mq7q;|ZvX;piEM;{blb~C6mFKKAd%eGNyTUJ^;p|sSFB5-b7^tjYobsRt7
zg{@x4-*@>6c)Ab4j0Jy;JuYLY@betb(E2SBU{%cB{l2E$G6P!2?7UFYFc9)>!xvRL
z^=F&bl>^?1ppb2X^<!F~{?8gXh>48eA{UGNvhFGAZM7+)Zj=@+f89$}>|41-fLceb
zLYz(UO$(9DS|NwmyR=)#P6O4G)il0~MsjH@a;OZj<Xdbbx`ltSU^VmW$m@q+d^dJ1
z(RCpju1Qo==cUP)Ip5tF>>e|)d`dPOq*k?u@qIh%Ie|+<bkcR4LINt`Ml)}m)aQ%V
zjYc5#91Oyi%bGYy;hS(kyNJ$eim8V0yw3zIiKzzQN#Ufj*~Q1?TMo?gA-mf(c|lEW
zJRQ5SDF6PIw#|PXrtG-xIt^=jk4OH4+Rbsb_W(A+rvc%eXPb6uT?wT*xpx|Sj>WsC
zbdZl#`bd3?&Mkv*GH?N!O#zZPL`-%Mkl${n8Og7o7Zytbs*)s|i0+Xs9Cj%6K!=jC
zbZUn*>0c-gdQ=Z#qfTfp)$cmuQ~l1b>{Rm0<DBZane2aYK?~%9Q-SaBK^Fz?A0zNK
z_MYwe=LJ6Jg`VnfjgN7{4`}YX(N215Cp{w??WQWUq#g8Azwz5?;?YIKoTi->%7s0e
zo(YG`4N?$*fdOBbS#5PL@$KFoP-bi0CN$N(l;ir1d;29SOVO)2tQ)UTQdv5Sb6>-)
z95tgB7yxHIGXP2B-_o|!Gx|3ExU7#B5rlMHq@&}GMQ<^ZYHj4yZT>gc-hbuFp0d%_
zpcR4Bj&3{X8Yc{J$$ss2VLj$r6k0tX=H|nR{2!XleqKj!zts2my7$zNMfCpYquk%#
zpZ)`b^Wc{;IxZ6fF*7)qF*86Emx{ANHh(HEK*K=WuN@$_1brLKX)u{n4wL^sOmUS+
z$t2aS7i11@w<T4v_Kg(3{O7_iKfl~QUVeHgxEN+p7<u{l{UYcFr@w;JMU59D1k*}f
zKK{7;<CSx%>2sqGaIO@64E=BE8?WD#u=IqWCj?y;;`+^ipeu^5D0)UBeRKLq`hVY~
zYmWr43Epx5kAZLtdQi|h;_dp+k6&JRA~=5+Dy3uE>WenOh?Y!cT7M1s)5D5gJ^>k6
zv%w$(^i%1eWJa_=0A<;y%W1_FU<--FE|PeVL_=Tko0tF+4aoFc#ozupzf!n=Pxk}?
zd2xNedWnng@iyL9VDuQc^celm>3<%7T)%0aA^b<RQ19^?sW3e)2Tx14etH`Bz0uYa
zxo3T&W!&kaEqh$wH3lTAL1)4Cb+0wEPP7<ily3|}xuO0%<rf3!DZgg*#~XqW%jc+?
zK{A0@)a;I{xyBGg#v||WHdHgoYJ*`$&6<JOto;}Jc+Kix;P3aYI6;xk(|@H_C`1|D
zGSgS-X#?98FHi`ji!CPU`s*9cB#2#u(Eas{W#qXnq;g3`7_kI7`~M%28+nf^gQQ}-
z_026CtRgy#bYzf9*VvHER6q?EXtUAHgN?{Yr%LuQqP<AunR_aRnp}e;MZ&Ub<e2g}
z@Yl3Hq)>A!S99Re@d>iAzJDuO23&1c>g!rqdYa1?Dg?g?i8{y7Y?t(fNq1T4OC<{H
z4P;<97(#bhNiomsac5cVFH4?<ZUagFm}MSqJpIa{guINdh}Rr7%6*ePvbtikZUcn3
zE1N=^TI{QYGqZ-RB8HA}MS;%Vf6r?cdZDaSB0^fztZY`d;gO(}TYqy9bNMb`0Xts>
z*&Ks-9eA_7LbTJ?1JEhvl&<`>8^ZnCcC|LVUlkS({ay3Ac!$^1%2U26^iuO4oa{JZ
zBc4Qn%UttB$72nPS>m?9nL=03!W0)F;x^p!w*eHDxEr9Tki0lpp>B&o_WVOB!-SL@
zd}n1<9<u6^^L(ybet$*xUuV$hnd7`?HB%H6TC$tGC&zgYg`x}wEvAXUgpzWL+NffQ
z3pRS)u7z8{oGtxdu*TvOIg4DidJQ6!CfCyGzK;?mA_N-^)|+g5*N25Y7BsC>EtNA-
z6Ue4v16gDeX@yJ=?q<}5Z{6iv5BZM$GDt|4S0YtheI0@0m4BHCTwv=6T*-J9VX3fO
zm)31Dj%b)Rsubg}DaPAcriCa~WPcsfj8P?zc^@0C(?H~P+|`)ZBURcpO6=CBkG#pa
zVliRfF`%!3$n@>{lAo<mP|WskL$!n6E_&9p)a`xRIr$5A#v~ikYk&de3X<hR$}PCf
zrYCX(KAv&cnSTPb?6eRge#EL?dL$afQriY?4f*NwU<`BGc3RvG4-G`(K_<(t>Lgg`
zDc-Qt7tG3AOUC6gb$J`p+Ca3P+3iuT#=cJVDNpGI>Wx;`dmGc+KuplP=dE~>eN#^-
zDC1{p?qm8Jh#7fL`s|l^0&0$^9umMLud|36^VLAj@P9SwyEx&yj2=+-!Wk9dq!&({
zP<#0&(Yk-<x6WWH)m6&+$_3YTRah~Xsk+pI8=0iZywbi;Em=i1-Hb|wum7F1T&W{4
z`dI1DZ5+?}iG0!fBNW7?J-aV$m=K-1+V;yXBgny^A^z=AN-nPw^*St-qFSh~_Bx=i
zpPOIrAAc|Z-PE$uS_TF$|N7&_-vEJBp16zFj4uB9@5}EmzqUx5eFwhYI0*{oTKk#M
zTEUtjTIyG*Ew^Ua8v6F^R+rvD<Pt?nQu!opjUjOk^l14L46(ZM+sa%}Mv2$h(y+MN
zTp{BnaPDzPj99BH!K@wM<yw;~P$Ax|BHeUie19rW2LWS)ISq~uNg~TtXwv(S5=0hv
zwMacUo#UH4L{KcL5jFGKU9;g-Z3<D1pD@mVbG|$xw5-?~>*?l`n5|b)*>sMY48>yu
zEw(H>{#flp-VaMeQrK`CdF<yeDS4O~H&2ghH96=NbB;YiwUd-dw`AwfF8Tgx2^;0N
zhJTkLo+M8-xE??rHa03>tdK3H9b;P{98*%X5MNMlcHowVCj|;>rj%=+{+fdGh^ZbD
z#eqN^M@3UUZUj`0Yqz0RJWh3HpW`Lp*h47D@`x47=xfHIrK1$BHaaP8u$Qt&n->zR
z{H*XA4j(`&fnp(+1_C0DO`I;HslhJ4i+}VsYZR-DUiU>S>DF1=Fhyazp<o#`=E-S2
zU@t>wvnrV1GUwF@-RYTOo-=!6e4B=3yQ>782_jB)_t!Ub1NZLMo;4KTHnxzl@>`oe
zcXzXgmPSpbR`sTym>Y+#$8}|!CJygX!w@Fuxs1QBHMl<j?2l8yQoC+iidFwWgn#sh
zVGvrV6frl_kT%4ibySecz%_bZu!FZ%yr{S`>y5-^WQ^6wH0=e1!QMUYGQ`2?YX6>+
zR!?t&Govf-&CvmF3-)FRF*?&AtY(0=Ge#YqX=pI;_rXym!K|?@)YMre$)ws)-0x(T
z=Dc?}foKl9E%(_|2KAYQ^$CWuGk?=}M&FK6tG*Ef4#l8=N(@5!e<Fqrp2>}>VqFFp
zE~6IEV>?0WqvZzj907g);L4tKzHk;v_#B~o+Y&>+Z7}dTzpbH;_-ze&hCcEx=}87V
z^jTn>I@(qQGd>GU^B*LFKzF@a{DXaV+D3uud)eEX8ro=$fqx$!gB8}bP=Av)s<8z;
zC1iX*CuG$7|3|ErY!1{$lrvUKZ38#r?<|$Te>-<drbErU*4w=vGfq9!toX9C!7!70
z{E1q#f&9Bk>`RGRS;fHU4UeBVIY|Yrl+a?DI6(qZG9@+&nmeQcin?L3zij`I2KXGv
z_t%{}Gy`N>VoUkShc{6PCVz$AHN(MS!|Xi5R)f!{M+4HvjjK1bo=*zmaqz*gfZ`2G
zY>B8^r6&y|DyofD3wfUZos(zHO3QDnZNYq4XSPt8&cmEK;;+1o(jb_68Ks6f@GeRX
zcuI{VPpy&maU9eMb#HdH{dN3}ks84#UKptn9C%-(S@{G7E!3CSMQzS_KOltk&$n2H
zGrLUTKJm6kjo=e6do&2<-}R^g&rngMKJLMv_zK5N0ZX+xDBJRMyxvgOQd(?<GdR4j
z{C+Rr)#czl4GLojssTJX(}PCJ@O$dU!Am5_6yhFnUw==dJ)Q9W@$x^R#K^6e5raV$
z0XCB{O(d6?#61iGH8hv8izz36Tib3N#}R$kS2U0p=RrMPx4s010NaWZ12iH?HjqRh
z42oXcbZ9QITpG^b=Tr|lyA<ho#U-UNB$2hVHQiO!bvactA}N+b3WVP!Fn20*P~}cZ
z(2_f)!KemeEfy?OzoUhm)TvTK(SC2$P;FeVhL-t>-&+Dv?Ti^qXtGX!wm>GW-;Nt{
zkxpXe+Da!maFV_jzGKScRthn%7%Z?#7X?DHW%1o=3S4U6qEg_>{kDq5YVN93EY{Mv
zF?4R$Rc#g|Rfs1wr6AO5=gmR5j+&JqHmVCY$x`}mEsXMG1>DwB*P>C-Bz?CAW3u(#
z8my_KWKcj0b<uG_i;h=+E66TGX)S0Sb!!T;=o6Q$2213-Rq_Mr=T#H=*>ER0n9`_)
z;G9%D3ErB=tA;acBX15_DEGxu0Bh+afvnkgBM#ZP^{WE1jJK7LGr7E#akUtA624Ho
zTU30Pn@A^fC}iUbB^2q(SIAd#8n3p16Ll}C(519~LX|1h)H(xyTd2Z)G0q^Z3&|x^
z#W6L-5m!|SO>oW?g7>c(PB;p#_&)a&u8R9h_s9ybs&%vtY}`^mK<pD5r_KxUn{v09
zXlSFuWB1V951hnCDbaWtR}yI#(tqOQ`<=c)(&T<p{eY8N7nz-GBRe)u);*<~-<n4w
zyPBztXDV1MRU4&$g;J@FTKahr)2JJGNJfnmJY>1~wjPR9x>Hs^C~7?J1#y^0am2|U
zo)x)(n7M1{Mg!R`lkt;f6xqX&?P^KS2sS!HcK)LJ<kL@wum5|#7(P2YTdu?K`EvE$
zV)Z5=%cR5M@57hF7x(~K4u@BZ<MkVC63qiVm8fnw<Je$-dM1f2x5FpH>&v%mU-01M
z>_5Y2)AGmbljYg)didtmm;UdQAJ^;ii@yxR4=3v%FW<)F<<G-Ump`9B{ciEg@YT`!
z$A6r!zdTznR^J~TFXB(9!`sv4+u;|AoFWYutK;FD^~veU`ebo2tQHrSr|S!F{po<(
z(*xE&KYR9n1b^?T8`59&Fum3dMvId8<LjpR=>de=X!rb$r)F2^)$q@M{TtyM3c~U1
z^7QoZZVvcqM2H0|IU*bGz$s!;+it8ZQ4x$A)-ZgrJYB9{pF`6ienS{u9<8DKS>W;G
zUw>Wy?e+R-y}(KI3on*uYi#mDJRsrezRL@wYXm@l9X}wAptQ;O!M&Q0KJek$v*G1x
zdHi~@elxs${$lug@$33<%MV^2{jh*QF#Xy3qF0oD8&3aXdAT}XTwGNdzxLJQyOX2O
zm%ny*$%tFjhC{5d!o6V2d6cEQ*2PsV?IJvRGnP}=Wh|%H)aF{drb?P9nkb$qnJAqo
zn<$5W0<RA5>*ZM`rmYCh+M>-Nf{Ve+EA1D<@Re9R|6XZ_v=>=Fx^jRKPY?8N<!*>}
zx280H=$Z=WNbnWODJsx`Li2)hFC|5`dN_O*u>9-cZ_BTj!}IV7EMurt&4WdNcS_0z
zG4_;#lSK!fqBSaNmPm>K8~|Un=geDDsMlV9QDRiJg1b}6cS!Lbg~%v&AaY%Zu(9Uy
zQBL~%%d@`vl_MvGBU5{Tvj$#_8JLW+L;_sptg$)xE*t@wZ&?{I0dOa)<y4pj=&U}#
zQ8$jsL+VNh;G5WC&2I<+KLh9j*3^tvr5T`R*<T73m;&(sS_%*B<RI!pnqoo<aK40p
zb79gbUX>7wfV&^a6p~;9hjX6W&@}*VM_|&9R}f`)#2)Ox{wE;=a4lu8$SW3!H}#`Z
z%-m)WcT5QwkmVhiRO%L90r)}AxC<-YU>jkLHj8aHPQ<0+M0Q{WD36B5@)5Ao4R~||
zIPHU#TE2~yn%)a5dC$JVN;O|2nmS>Br8;4yI$@<YVWl=<r8Z%uCVQbFl*Wkg+~9>q
z5h~BYo!MI&$q!XVs!t^(GCDxco;y8<dLsG2xl>_b0bW7oir(j1qDa<0P%T4?$3DQh
z5<)~`QT|OKxU^c3F<D5G)<SopmiNuyW=NSl38848Jfwb02vT6CYAGR$o8+2*n(mdr
z9i#CEISG<i$ut9(2RZ~w=C)r8fU}CVY}$AhK+?i<K!(%`pdrxjo`VX}=(&9wR4Cv`
znnMjHF5((?h_PLToBiIfVvA!MR@Cm9yRbS2DqMD9b^j2lf0zo4c)wMFeZG6I3QSwm
z3ZHe2Rn#Za*OW|@PLxg5p0fvknBE6P)9gV?1SI_^j{6G)Tduxce*f(;ym!4iIysvr
z5vaWAG4`B9P(gOz=xWIFu<0z)Q1UcHB>QQo-iMTphMHQ;WeytFiY!#?o`nPzzk`FR
zP!hIQnS*{!pB}Wdr-w4yA)&28Nob{M4u!ILE7zzz0QU7J&c|cG`N^YyP!rBjLsK7D
z-Nt(<!!iJ#Rn<~PjPL}tp1&EN?d{M_-eeogPL>n}g<=O*qa4|GV)Z#3XZhn4GuGS1
zjCq%0R{7mxb}d3LW;7Kk$`e&4s!r5(W9iQ_uPNHEtdXLt*eYvqZ~*#Qq<Z#eN2I*(
zSiRjy^*()@N2>P`n<LJDwkM{RqLJEGS5#BXZ4UF~YEZCReTZa;tWgjsoJ%ybpiq{L
z1MqsJTHh-j-gjtEMYo3b1#UElX-hM-p3TYw8W{*O`s*c=k&{0nX^(DX7&vCzcrsNV
zE6l-@TXxuz+s)q4Ur_n~z2P>;R8GmYYrC+T|4uuxib>Q|cVR_;Iz-Z^4_SHNyVxq<
zF1DKPS8O$J6<e7~r%XMhOg*GbJ)}%Mq)g>mrt&OhzuB!c1lR_@np!lO#h@o8q!UWa
z4}xIfXgDAw2dG=1Hu8(2-7;^t9zC6)UypffT&O_M#PfL2Dn4vVQXkODg0L30$;ya?
z*zEC3qgalObZaVqEB;7n7G}=fnRD59E-U!G^~4^IluDg}hF0p+1FP~7k6Nt8<S)K*
z?Qd!L`*PCXOGHao<+dAXDG5mn(6hl9#rAtEf(Y&H*6S~XPI4A7o%`sUIq5saUuma?
z`UcxoSx1Cs6q6lT1=delcVT7W=&$KMc3I!*q4%rtc1J^hZTD=bt!&r$sb!X_W!9;o
z*6CZFzO@Mz*j)|vJwqv*`SY78l(J{Ws*fE?KAZNT!sb0WjY9KyGMxO;FjIcOgk^5k
zQYR~0p=h-|wv`>Twatjt@LQQ?VPwf}WPQl0pzxtY8AfKLF72%$D#cxI>=b`XwH7gd
z8QioNr#(P_9OtOEsSEx#)hym}C<#_BwC8wAx*fMB^{{v5IZWycnd|p%%WJ>e^4cAJ
zEWD?somj;OEs1B{oMyVCaJBmtuDtzNb9@Q=-fgAS8vFkrxpt;zfNme>>H7I<`SWtU
z_;z`|SRMH`qf-H`=65{m#%uYvaMb?*24b`t3T1AWWV1mMm$BhE6qk_4K`MXAjw3e`
zz1LUh2byM(!3_`usMMtfd>f-~gE<ZCIr%XD|AT{Ta?3>}^<WO!Dy0~Vg~3=J+5GgM
z6F>cUd;fg<<u37Lm{nlJ>GO{h&<~vc1x_!uIvD__C7eG0JpJ>`xzO~y(}SE#Nspnw
zEq$uXHwl(T0F3~83vl^n5a@r6q&Jer2t}Wq9zlPb^zNSMHPKs+=x3n0fd<06e=h&}
z{Pm3|fs1rPGNs)~8)PIymd?+tfOE7on=<qUX)dMHLwraCu4ryW{P9Mgfx?gWvWGW1
ze+y&Y(;Q}aYYP?4i}(2H=MSHdX-uL~x5%03yhgv=EzImLtYg;7K{J1osOGCBoM8xq
zJr;dT{^sR!NhcNGg}<TEF=>D%X`>olX*_Cgg)&}b-O{YsTR8fILK$`W=KZ3zZIO^>
z=J!n-p|(sS^x9@Axc8gy@$vpivwhRtb8Oz;>hGJ@E#9UD<5m=Y*X#+wLPDH(R~Hm#
zV26o<;M*b)1<0PTtS^5=U*G0!EE8PybMTLC##S&V#Xz;aLSPMZ0(&ap+q|4$TEU~d
z$oDl4w&CBmcMKPdTxVoejG4E<_I4G>v>03i#5Qx`m{fY8f<QI69V)g2;9Ck<$Y<Lz
zzST?M^CR{QNREMAc6{89TPS9^XleY$3%;y&7R)hadUihUZySHjXCkLty2H!l7~q7Z
z+GZ&&Q?8@qw_@kK{lYg@C!dh`{2O%T_d0PbZ){%AK#|jwOpk~GHIGD7(=I8cgiZQY
z=g@S^Jd3<o_)xfl^FkSB%jrN*(xY^<ggDVF%+9M6@LnaxDlKFm#rxdnrKAN@eP~}#
zFrK%qc;m-Ks-l0P|J9tA9TC#qvO6gKc&ni+{+nhaokBEFS?g3j<a8ftRkpY-bf(C)
z<zJG6f7xrxuRBzf_6DN3AelK@knf8_ZtV>z4Fe&D`DAiEk<=EI8;WB&KV)Klfpbp6
z;#{X>Q8s>^q~pE}`)tb!Xf=~Q9CTtoo|F|xCJ_6X4upRsB@{zBEX5KR?2?uT4;R3w
zV#{LddQ0T`?4o7Nnj$!%TYmRJq6CDPuv}HQ&9*Su$Ks`RvJ>)1#6<FF><NcDB}o%B
zvWvWF^A)Z2rq+5_>sWQ7z~pu%Oo8pgZ+vqk{1)0a{5BS@!Yh1V>0lhwR-aTzhPsph
zSjvEbf|7p_S~F0zHS8)MO4wqlG}JzumLPL+7gNfks((PW1VwSHyN64(*Za0~hEumG
z3pc$lUPqDTo9-juuTnhBzHdW$qwu`#JL)5Pjpp03E6o%yT}bMEH)ZcAc4U{BJRehH
zQDZ87r9MclhAQO_*nNFEiamXxDwgg`Uz&^AuwH+=`fTeU3^T3-WN`aBbQDPkk$k%;
zl5mly?1v$W9k(~7AC$ImzV6G~QS^IrB8wC;R{Ak^Xa7<x+maV4{?tv&`>wn_#Q}L|
z`&GThztJKyVu_!(efc|zBlbSHXP?%I+D}Kt5JkfU2q^&Cj#@3z5v>mHTy<dQG8#ZW
z%zb|&Osm3459%lXB2nkR{PttLp-d$eqGd+ux~2*obA_r)DL9xUp`9p+D^)fbb?X`#
zDHilFXSq^EsJO#QJ09X@L^;c<-$N~2I<wo-fdOok)V|Z$X^xE!AV{mg^<IiDzbx{+
zTP($;p|aYq1AYAHzI}YY{dXv3r2=LQp8kLR=bQflK{0JSCPJ%m`s?@8A8-HYCr$S6
zKLa2g7MW|URJ+KOG?hJ}@0qOL?F@s>d<)W6-$){pRML?o!b}i2&X9(de}Y3cZ+@%X
zMtS@O``O}FY8=fwF_CjWhfs0W=1p*K$ETWW+7eWZcU*)!oE}$g>O6X7`<WA7^pJlg
zvR#d%6}CUi64~A@7dgV~yP9jqfTBt_(-Llr6$TB7l5^JpP~PvZOX7UQZh39nSzqo!
zN!yl-%--{;*Z91F{cPEG{9YaN+#a@wpy=TSd(1PL5S$9nVNLdmHnGz*=4AVb)J{-N
z9mUSiRD8*)#lm6nd&f(GPb$l}s!)H(hkcE5g9Xt|G;y_sz%ePHBY0wM%l>U8SSeGe
zOOn!cgP&X5j4`z+Q3M4d9u>X#@%7h&c4IFr++Vtlg#jjb?+KwO%LuC~$k&mbEF7h1
z*{?@&|5?V~J0p`qQE(`Ixm1^Mri-ZxinUn61R&uiP8Vcq`{01~Nf5W=Q5=6V?zSgg
zWV6iDhDjf;TL~7RG2^B!yuO?WS2goBYu*6pP9MSYoSA|6=^T2_CCr&9BBnd%Z;rw3
zo9DCEP+S{)p+28&(`FqhcY)vHX}$}FwuFOYXlRYsl|!0{-ld));ODuLzjbH&69jM@
zO{KnsQ+d@%ymzKg!O|$~saAgu3H??^sq2!ctf>^|bQ(~0MSNS(%4oA|6-a7*!^m=g
zb1SrJnSVJW2&msJ@{5>hIwq02+(9TbISk*bBg{B=bA(yOH(JSl4-E0ekAf+NoaqZP
z9l74BI}b2Dtg0kYVok2ETRS1vTE~_x9AQ;Qs>7;V^kv84Q?eR{pBjI^Y=UwrB=@Do
z@yoJ9xiZt8iuEeB(0L9`P*YV+b3B7@uG-uhfc?-x$F^gv97iN5LtUHT;Pw690wVD9
zKn$N^(77i-3<|xB8}ok%Vw7{kyeKmN9j6pk;>r@e?_wWPQh22zrPcd?gna<|D*~>^
zKCES0o0o$hzTxrLO%8v6AM3WB89LZM+Ikf;5!{h(FM>?k$$k=I(lYsKyK5nCt5*Gw
z$su3@*^~Vq?|NGn>1Nhi)7uTc#m@VcGErB6*fWAT5}g~MQK||{`3j;0&N2h=lM$tk
z%M*xFH`4(`$zPT;s=W<(Xwxpos*l(;VSXLg7u!{E$v~Q(OP_yl@cLN4Dg|>e$i8M*
zL*1<_?FYIq_#mYl*-!Lx7mPSZ$K|Wh${3~r`q@;9CC|v2;$S!P{t4;yj>n<pMqo&N
z)1XVerJe&D6Qa_u@Cw{$6hkduV}Ugs;sWOZi5xszEAq>7Ooi7kdv-g~&+_b+%d?){
z&h{$UIR~uHA}fEMz&#Em`q8_duy#A8@r39B&tuXVULI811f_GS2@07tHQt;N(kXTd
zU&l^M=4?5>7(6ZR({gzRJ<TrD&iE933PLcdAdbPOmv6$VqtK~j)|p}c2cgsTP!^r`
zi>Rul#*GTiPPmK}MI~o&g|Iz-na0j<Q>O8I074qoa<6}p)9W_+>F;EK?6vtqS+{=m
zm7K-G)Lyzs+*n0#;>CqVlrm`#rXX(#!DWW+`g=wnk7^t{&*d7hFAaB)U?B|2qI+3Z
z5K<?Z&@cl?Df}LUjMK<;|2=+N0W|X1uFa0O$-TkdxdKaJxv^Z92=dan1#sN{uVYk}
z7~LuQLcf2LsE9#t$@ao)EpKaIrI$POmDRDfta{z59OS>UdNT*2`d4z@MyGLM6R3FG
z56Q~V?j5FWOM%UU5Gq4~ir036c;i`4Zahy=>H5a<_8CCbuRa5qfWcd50FbRgwZ+!|
zu|p<BdjZQXtC=V7t|;cukI%Qi0c(ctpbBMfWOJ8hvq2C6IG4eh0~41%2r(#^>YphN
zmw(bh2NoK8fZQ_ZTaZ(b$!QLm|39R-N~9Lsmz>f;839R`w$edJf0V}?9`7V#n1Sa8
z?g+dNjECRfe|eHw;Bv20Dld#PgRv-BmLg@G0h*DYP7YS}L=xz1KZ88R(VNyGB`oxV
za9VbAIirO}>PHfrgji){QxK(@$fo~M<>xN7g?{CCc!neY2ENDshhTAH%SoO>hWnAP
z)8X03@B}`mEBKAZf7znP)W)vLz16I@%1eYLmru3Nf|@L#N=hidy0S?<gNk^{s+Qhk
z=d%s#7q63NBUY!=^-gamkQe#98FP)xn}K)qILQB=#|7)B-5`ag;8`h*mumRIZd`xz
zb&dgAknw}jMri4(fz8Z_lqGOYc=T9jzzUF5W(ZhS-SU=je=EAew)Uwm^jSxf+5i&9
z#Xe)uJ-RA-uQa(NXy{5CMveVHSK+t9lTlO<8>&^}^%Q^=Y=CZmApMdJ0@FHiewErv
z{s?b^jl%6q_uujT>F<T@$$`+k_#fY&!UF^n#`-}i;Q@yqe-7WCe(h)NAYIODQp_Z2
z3{ji+Mi4n(f6WTR^A=B^HC)OIddtcoPgnoM7;9SQv1N6LyMDzK*==Rwxa8q7%%LdF
z+O&KPaOCBxUTnP(f88pf#v<iNwi8SgJ=e2v4FDG_Dg=SbE-ZPx+Cqw<)bJv`g-#!8
zAw?*__ls!A(JYgpQX*bv0P7jd7Y2x6IIiWu*^-nKe}VWeCkg2gDGe4RPZCC=6yd|{
zmku&QK|g7Z<4sWv%gofG8&y;LdZgVHuPDvP3lm4Y0C}$m80iEjV-GN3SXHwRFqZ%d
zpg{DG1eA8gdjU$j$|C@!v*d9``N!+Qg&fbe(4%8{wYO1hA>a4bexe+)=4LBID&nS6
zj-`~pe|-QB@I<j7UXBB~N*A)h36nQe>(N-B20rNPf2&lhJVgZkhFM!zRt&Udj}s)>
zTwyLy*IGy#XG+o~U1Zg&krHti^|S%Dp;_3l#!;qLTO<@S74tKu>=NB<^_=KfQMGd7
zU4%vFxf(5;d`E;u4sFjSxg)}&CWaP!CxqqDe-p}GP1zQ)X|6|H&*rw+OYqT%>z!~;
z_p96!aWgjdmc0>iE$@ttT3BlXaWjkRdT(y{F3!x7mT7Mb?QS)LbKYgrYk{F`URQRj
z+TH@-yeABV<9`6Z0l+&O{(J!53V25V*Sw;;CwvE+J5+aqZ*7=<gZ~%dyOZEf@ZGKQ
ze*pOIE&DO>-CJuPzPsLA@U1a<Gq&6AwdI(`v^kpK+ivCPgP>e%<QVwLxHz3P=Fbc7
z)(ESXcD)eZFAJj-I(uqg2K59^5b9podqO>9uF-S_*nc`P#O|EIDGvgaou9!?&hL<^
zzL#caqI;Pg9E>>(7q6f9-zF6qtQM9ge~PymTyc$4mhg2}!Cz~zZAx@_hL<W;3ICKf
zDm`gh{iJhAuxVqz67(f8WNz`jbUSIY8Hu&midPV4-l(%PtzBKq7SM3>5i9cVbHt0H
zBOg0_-hcgAd{0Jbr~a*T5tEHHW@D67uziE?1+|vQybn9Y9&}cs6k`d{gU(76f8U)~
zb2Pf%nE7?*6q{G`5~&4f+e@>;;_8U--2s|@#4X%NA-DL23>(}?IbjU;z0WejE$(C{
z!ZDqYrm~woG4FB9t4HukSO?eB9WMI?Ve*cPUN0mwZDefWiQ*VHoOYdr(`xzXyPs|2
zz>^uYGK<MfW*0Hg7fcS%3+XK<e~#_vZHvG@x-bDihRtq<a4cT{xPEG|!LXDU%K0_j
zE-;KniZ2U$4aSBmx)}!@XT>zX6`#tMoyyjp%$!;Rq&Z`^9Qg8BnEiSR?C;(!lQZ%+
z*cY8<+}9rzh!p&77qewxS)^y!qRm)y2Hi-R#+r(;e1d*c^OrphWoW&5e;Gv?8<syd
z#3UDWJRUSHNLaN*smHD*G-2?%ww$DLBE#C%@d@BKiMjqVRP*ID=>*}6);-#dQdq6~
zl?o0-Vxp(rN1;B|zZVt5L{M6CaSo3af$+2xtTI6RQebM@D6#ohowc$RT`<Lcm78%@
z>B9t%l?hKZ`C4jYP-ulcf3tDnSnlRmP!K-vnKmHBJh_m5sJB<BPyg{$My>d)Y<lbh
zH6qlFwVqEQ%jU9NAQTDh=>qu4YARV_^A=?`1-^otEOZHnuG-9DAK?71sMibh`s~af
z(Ka7iM(KC7Ufc6T?ep%h9QOuGOj@I#fA6;Kk)<7Sm~ssusOVl6BA8)bubP8c%pR@T
zUJR^j##PtM^-5>vJ4Cg=Gi3Y&llT3X!KOkK0y8<6FmoOh0Wg=bnJFiikWoGam+09+
zmw#}H_w&Es|M8>?!;L#MifuSqM@o38+{K;9$18H<WQ2qk$sR0E7&9$}>^+{eBDuU_
zbNXATFGqj>F52<1svQnW2Q%)Uc9H$rluIOKk(yn8Mp*8mK@#aG<Y#f5Y?%**P8_Fm
zRA3r%GevBf4Xq4<8yOR0#MQsOv0oLZs(+KM0?p0XQIwagM3Ewi6qOlO%(hV6aGy%i
zc$FR%1GysnWFT9Obp8@$m3yYUrFg3>^GNqLrB;mqjzDq0ibyY=OGbBtc<+>6gpD=~
zlQ+d>@>oOEiZE_96wlX}I)cTM^<`I4Mg(!<cY{Dt5VlQ8kq64YRh5;aEqNC1s?2y)
zsQrJ!6brEr%1kqv-Bk(6q0(r^{fkR#B|;f5F=yC%D|Z_-=7?N<vKN)r=}@BCgjrN#
ziA*jo=hyQ}$$A`u3c$UtlG_ybje1M*?bh{1Rc|F9Mg7KXm*!lO{h%7T4*AR|z`bcC
zlf3Hm``$=aIq-x0U*DQ1VmPyTF6=?stB8M4_iUC8HmWz?z7PMszCZnMk<8g*itYdW
z^%Nfvu(B?k5QP;w{r$)3&!>NmREzz_dn#<nNwke5XHH@jOBxiPQo-QkQ+LiDi$qg&
zLi#lQNnSo1`F-+5M6ax+kx}>-ubKD8$dKh)rz)1I&PGzq1ihvXG)|(@jJw1{&6j_Y
z^8HYH))k9P3c^CKuemJ8+$f2nqz(RB>p%$@UsGnDf3vcT4vNCUs_?c@R$BpH^kWx-
zx3qXW4@N8J7A(N}_J$=ZF032-D&i?_lX7T^v)bXr-&U{1*_8&N1)D+T?9@5&IrUzZ
zcFvXOTw~x$Nk#=Y60FR#Zw|D{k`sTHqAKPTa+E!;W%E%(yx{Y$&f!=uMnPGEHMzRd
zp2`N8m$FxCUtKNliKCQ>TETA)uwV4WyO9<;HTzwgiMiG-I4&-{a3LVkI{mcfd)WUo
zEs3nqEq46G7KVaVaq9go3^}QD&i||y2D%ceCs#=GzZFNp$q*?FFq5S6?~Z@{T3Wmu
zYinGz01Z0%L2ZrkB<anJZH;rv>G1HjhKLo1DJbr;!<rFtTXWS8t<`Cak*L9mKN?cj
zw3a&INv!#=;l5kyEm(<0awFxH2C1(!5{cuA57d0{HWt5;h4q>bLEO;M+V*o*<a3b}
zQ?GeUzpX1(e9SuAN=xWn_zr)CO2U%FI#AtHs0|7vO${qPq)^D|Yv5r+qecDm$%mRK
z+XVyVIXG8e>*Kp5uw@63f(i4cVk*)%$<J09!L9?vT^p1zn6LPV4J0>NfCUd*HYgUG
z&%W0PeTNIL>)O_!95;O5D7+OreJyaUkE_}gTuMF*d&&NGtY#ue^&@|0kH%^QiqMU_
z#B<feH%@7^?RH&;O*|LjcnFK#^sqx>oOKo<?JyhJlgB|inHudG=B?vV{25Tug4h_a
zrA!)0#jQ8emk$6b=_~qVj&z>-dY3zWgqBLm*(U<lG_eSQ`t)GmQ)yudH#eVA@kF5+
zXIvUbB|y?KU!DK8QCxrUd!>`^*A@6^(=K(tyvLQpN~zESq6e+66t>OdMr~B!kw7^o
z-8Y5r#W=vMXj8F)#a`>J&Gz{yDX?~=65pQ~A0Z!3KzNfAse{A|8lRLIuZ9Fk=I{gF
z-918(^l6M67T$7%z(zTO6$0&)32Lv;vv&~w`Jl+_oFkIbJW79j+*?uFb>am9TLY^>
zY3>=G+b<s;<Us#aP-;Y2NA3fl)M%0lM!z~Jy>^~^<q*Xv?QxLx2Z2qLyFovKO>3nl
zJ{)VJz$v;NYF>ROx7zf@c7!}U80rjE4?vw^#RH(u$iptw87BXDsDliO5g!P328suu
z&amPEP-o;}AL@S$vp*5)2x*CH^9w?qfwND7I>RoX40Xoc9uIX!dwvM&48%5|4q{MN
z(?h@xgg)b6d1>^;aG3X7yFH-w(gxG8#SN_#c+r#$%?ZRrR<RA}eGWoUUeqxsuz2CJ
z(a1qOu;-EegBVUS!2~(Ve^Wiph;p;0x~l2!ZBap98Ag9Fh5p+cTI^}_CmqJY{Jq;Y
zO4G{u`%Og}6~n0M4k$e0H4()x*I=~B_}R;5Ztg#~>8#fJN}C!vc)p%okH(f17xEK|
z%!Yi}OEV(~$-ubFQFWve7s}NgElhnx?!50)0g_W7q72MMpG2=m!0zD4S%kyxU29%5
zJMXCHA`XB0;{k`OyK!HsC89_l1m^LaiHS$a9$UxD_iFng7eIv-My`W%;Yk^3x<>yz
zj?NiD89|yW6gqgm_KG<+-Hc(mqT%BEp?m)9VBF%Cf)hvu0eB;YUlY7T>{-Jt;QiXC
z!C)Z06<_(b6FqeGXyNJoS)<3ALK}#m5PUkT^fiBy>w#K37U;B>qTKg~dyK88sx<7e
zH<%NxG^)EV!t5IN#GNjoJNNb2-kCOv)K*nZgz1E;ND!vPo8&I|qJDdu^I~}JU)#3b
z_>8dGlO5F4LY{eWR{KLx;?m8d`d-#|CokXj!?04*F-iQaPcD-&eD~4Fx=D9CK+g&n
zg4_;y?lT<5yuLsEAN<FsNta=aKNJEvF_+;qITV*E;z7%Q^YNcAf4kE~V6B3UALHBn
z-^Z7`|K2t6(I~dzVA(0qaQN46ckv4XR@Q|Bq6$EV-~V&?`R?!Ys&%jFJt@nBC`((*
zJSYpTV)h4XjWW(vWznO&<Zvv9B{?*S0$AQTj)D9y<1EV2qEPuNeHX{#9D2?NS0r1s
zKfnKSm_|N-e_GBQNLgG-0UzRPP-C5e+9)t?p3#hSx>*Jk1&5Ucu)RBK1)|7}-A7Gi
zHz9Ec`k2S`?b<<WRbX?4cC-@5(MrcPkn^g;h^+d<bI@C!fQ}VeT%9ReTI6RF<sWuC
z=TjI^(aM<oF)tRqMwD=Gb&N26x>+3(ta2WEgR_`_=;Xj}=q$zrh3;B+F(e^vRt7Xd
zR4RzUTwH{5phdTdu7al}*itY`u<0u6R@|RO#0n|sl{3r!d`zfS13d*d9}^a`U?6%M
zzoq!eJ#^2QDr1tRxWRf?58`q#?;0T^&E~-md>J{FHMdWK2+E+D*Gxed202HUJtw08
zy9%^_lF_VAy7=!RbN1$4d5sphrj-hwo2U_IpUMV&M0mI;vN5Eh_m4|vjqb}rti<`|
ziBz3ze^-mjh1Is%Z2Q|fQoJEB<r8B*anVOSM#?jfO(*r6Mb4_bVxVm1;n{Q9D(XE%
zmtS<9zo~1GEjcf`Pgqbqx>G^!RZxLd&NrHWk!I>58Rpf~DlXCtSDck$KXe=jg`s)v
z0+wgP+y$k*+mz_SeqL;|-flawpN&mGK{q?EQZZ49HH;U?GHks~iAiHd+0Ac0e>yL2
zQ9TDiVNp|+*LG&-Y|K^4+UJZcDs=Iy1qX6?R*p!a=;V2F){51S&CGsrB`PE!|9`xH
zUtF&~UBjUOK)*(Hpisi?xymwt2lS&!DlVjBkgUzC(mb_MTToNySIY`DJK6T6F7kHY
za$AS06<`T3xZr+O`vNk^)aU2APdsPL)hVppk{h|^M$Q^aIkadDAT#yF3P*V84bcl&
z#%<xdu1m>}-GQhW6DvOCB@1jcy~m}0KGeW3ORlo{1=x)7K$A`$pPR>EII?-@K(2?b
zB5sSl8*2|1!UPh7-=_9p6M-WPcy+TKwv3Q$bJ-4slWg1PCZaB|je2#Uy)q`^4Y+wm
ztT&VAcL;Wi96=CR+-jBs-mF-VC{)3RmE2QQ(vrrnDIBCMj|^p6^ouAj1t+S1@K;k_
z!WmfIz|i)ys1PhkVAGyA20nsNTmze{*eIcS=%uP7<1WXXKAzh1l_2M5seRgwulAJh
zXmCvp$;|NyS~$kCezq90_ZZz%UfxIvo1APlx|8Y+2W2HdW;?dhEn_#9(ur-CmT18Q
zb2hoPln@^^o0%<7uTzX0E4G|}TZfA1YUfN2V|xBlclnB}iL4jaWN%<ibDky=y6DVz
z%F{_rr;e6(d1CNR+b6@kBrXgtNwS+hHgd<XO-?WCo%W)eB!?QtNrm4it&=cxi{GlH
zNZ^$ZZt(7#&9-H;toK)&?W}jCW+vR;3o5yAI`12|=Zp@D!b+=OoN;r1A8T6mi&C!f
zBxkw0%Ty|zb)|LLXuF*q5Z$_2nAYVumwRO}n0Z!$$l`0cU>ZbD(08HD<P@wmGbOQW
zq9OX|JkO+7LyihN;%1GKmcaBBgrQ>&r^SuXE=!g=Za?o84jubrr;A-Y22z1xWt>cL
z_7utDV{Wa<3!uyB6_K5PXMM#SYiwlT;ik@nK;}+uYzP$KLVy}q5O*ZM+iz&ShTY~-
z-A7Jf;s2!pl13|CsOt=E@>wqCzM;`T$2Ks&)PykFDjmKTxw%AS(JcjXnokS4v0d-@
zOj7|&x@Q~D*s%}ej=-rwdRm&IucaGkLu#(S0o^Dml66j&DK>|Hk<W#aY$wl3Nzhq^
z;QJo*d1X(j`?3ISH%rQsn|I^eZ_>b93Ho`V*ZIhZI#(R$>_s6*H06~t{cw?R=HB~C
z4uOJpyxf!_^<3eC*=qQY>aL@HZppW2u|bLNieTOZn_k1F1Bj6zU;(=DcI9xq_7=TN
z`Y{)=e_|M$p0RL$>%ylAdlbE@w551KC;l=C_!UzvV!97|eKm`kWFhyR06&qpa$iLs
zHg2H`@iW71S2h#pPT_pjbw&JN>N<})`C17AtRM;SH>yX$kvP>tZC<7LEv;w!UQ~Rh
z$odt(Oeugn9RQwp+RC{RJ<WKN#PJ$92a=@1unp&1_qYXrW>?PL02_xZuN6zJbxx~m
zbXi4jLiqZuodk_TmeRUhx9#4R$GdJjAG=-&Q$vycSY)9Q0q0fvi}16xv+c;a?XJ09
zPnXkSsO!wE^4x~Rb^hf`W%8I>T(cI&gmN~Kg!rQ;o@ck4#52`WbvVJ0_=}tBa0#!$
zJbs)?-Bag(O60pndH#=HkEBf!%}RsaE~s2{)yx#O-PK+1uBTd34L7m%=Lo**6D^jc
zsXVfH5Z-Rn#d^p3yf6=`p#`#~d{4)D_wV<4hUEFp8D1HK!;F83Fz<J_2`I=lZu7@q
z>J;9Z5`SZ3;*STrDL?K?W<Y$(nprY<e~-?+`)iqH3){P)Q#c&zp;$EKk*Y&kTm@%5
z&vWGr7M1pyH^2L{9?fn}Fi?J;w9LRC_q9Q+$cN!<zy7+bbHkuCj_UYhe-p__!N8<I
zy>D7$V~>%BNq_EdBk2o_-8##h|Ak-t4^++bJeOgNKNJEvGnesR4HE%4ml32CCzl}B
z9Rz<Iw-LVYujr$MKoqO^Mm7WNuCxN|i|y#<_>@FOj0i{!0m;u#@!5QIvuAc^CFC#~
zO*e~GtSWx>AeS${E`0fLbN}(??W2&3VOD{`<>RLX=mn>Lfzv~+76ZVvfaT-o<>y<@
zQPbz1K9X}G=ws;rmcG^LjQ~p{fJOj41vr1bF%syBpeKUHprUV1AEf`A^z1RwYofPY
zN<Rb54KxrQ=%Y@*eEjZ)CxMG}D45c2(MB?oAxo$3IF}@yq#2buA>R7`hZ){JYFr$1
z8ntZJsW=1MO!Vr%uOr2dW|U=O<Sj{Br4ug;o0QrxAc|gUTC2h|1JK8uru>~?xWRv4
zMrxU<$#@&<Mc_<zGQ9eJx?n`kj5beLFrp)7YFMyNCYmo4rC<gIzTUAUireUX(cIbj
zg8DERAGH6?{Ts7Nn}2CP?(_!2m}G(=QIw37m`BPeFT9(p)sb51l@^*AWusbu`|b72
z_0VYsEk*XcZ6<;PGfwU>F=ms7?9G3jZfX9M+1O1IvbP3Yv{Ynwj3oaGB=0}o{JU&I
z+W$-$6~@sd%U?g={QL`F{zfmqTR0QWF8>KbKQEF%gyiGPKbJq>{1|a(TPbDITHa6d
zCDKrZ9m!wFu2by`=|yy%{{Q&;4QTDZr-sm^A&D*R_j4gNp$5I`g@ifQWSl+&!;ElO
zz`1IoqSk$!(_XDD?`qp~hRRwbTTA*_p~Uh#;q>_Qrw>z16i)P9MBa?VO+F^B`L|75
z5^5-+Z?{?PL0AC+E0?nFK_`C`NK6X`OKyR-=O#@Cg0zKm>Yrwom^XU*|C#x=QlZrA
zxCQF_RZ{^L%Bt4jlz2uBCx|GcYM_ZvbP5z1t{zvdtUBAYUDW1=?E>WTVp-d0|Dl;T
zaSlE+U56>_kU1>I`5w~@snU>lwS<0b4aPxlb{pU&Qt)chE9q*u?kazrBfY_{&%?IQ
zmp3J}*qCfY3GS;|a<F`;(uIw%60582%4#RIqWqm1TvID|ilGye>^T7{G6mhI`Lhm!
z{bMqGS4){J7~ja|SJG3^`km*MM)+s<g|A!vBd#mI2|LWKbC>^R4M3V(=Qnk-QFi}M
zmEo&rN^5G}9F{_oN+Ex2Nsdb)LA43hhz-@3Tnfz@DvA-H2Zk`Op9pDOYxGz1TtKK8
zC@#^3_Aq8H)dkgQ%61w&(1DQ*&BZ#foPy;*g>o*Q1J<kI%bXA8DnPkJL-Ol1I@1PV
zAcsv^ZLbWbh&`Cb_bP$I>hcfcc8wbwYxs!2N=3T~#WVFNAUJ<fgHl0H*GoH<S!?_C
z21w1h@clMy5kKD=%M_-nr_8(_C)nrEOmN`OK((shq&3B%JA(pB7iCK|cV|$Pt@@;+
zV>EYXpxp+hCpz$Epq-;_x27+1>1f+X*D?CToQorpL2(B@0kmzXX6Y>)_CWN+94bsx
zsrQvtQ#0)b#?*iLm3RA<S$J<t(y=Bhx~cC(8>-~HwQ?_<Fw10a*D!Td8P<VoCx}Zb
zN_-t&CB;5Cq7mJUXyjoUTSld{B>t-V7wO3`?@@}BMs1`>iQ`Uci57R)5p=EBwFK{|
zteSXY*2uQxIrZqR7=N$5mzYdVO9v@@i8|Qig-Qp-EpvaUdu}_OpiSSg6s-vgDoQnY
zckV<4sdS|i)p+XYyLVND8uzE!r#$k}&D434{>pXtaLO~du)YzXJP2~89mm-H+I%zX
zqGl~>-_LYS1k0YRiqxAYYU#XBO2<&zQe5*E<wW*IlZ)PUZGl5BdUQ_r`##(WZueVp
zPjk;x?1_I)JyduPrGVS6@9wem(x~K6;<noL17cQ|qd4r%L|-4eUrqlfn}rHOW<mJ|
zkbfb83T@*S5Xcl4%kO_*KHdD-Pn!(Xe}~>s@B*&2Qmq&0&}WkV0LNZo&y(@Hyhn;~
zIwQHOGbQo9>}v|Od|Ceq-rPL-sq#509=MhLZ18`s9JdW#s%dXH(Np--=z!Wa`pVUY
zIN!yW%8g_<2X>HxZ_11km_#$^EMfXQN~hh^#Ho0azDB6HqYjtpjY%KB<G2rK{(IB7
z!cncI#li9PMTVy2&`Q(KJTtTr9{hR=u~yB4)RuOA*R`@wlU8XagzIOVx@B5lOHnR6
zJY0V+gE3443^Kqre+vP;yU|D6Lg=+Cx%RWt^R`FPLyRz*R;)@|Oo3H%91gBxkkmW?
z^!sQ(&9`S%l2Qo!f}kbn+i7<t(CTJ81GGAdZv<Ms6m!t(b@uH*YX(?C_*(h}@T!@3
z!^A!_BWljWiDjI$c@lCq$2X&OhPZ~w;hukWt=8+K)N)X{ogNeqXw_A=<vz3`xx9jF
zLTfwYIndftTmxD=84jSe9sM*jod>M|j5~Y=v~DMoM@L#{EO~5m({H<q@qN0!evMBK
zt0k1BJDBv6j!jaNetMaVGIve;rIYi*+jD6A`ROpraC>e?1}7LO7-TrFN*0s>S~GvH
zz*ard+y#1;hU>b!)X4LzdJ>SCK`Zn1_<I*VNOV-;f}X^+12pebkhc>kJ<qCA83t~Y
zKiB10I!Umv+rf(45-U#Co3&5$0R=AfWjgBETW^j^*LzmOPSu*BE6G<lp4zQiOP_D(
zsMzCU&*@&^nmK{JH2FSIP7{3m4rzb;!@*@w3Ys{~w$o|!Sb$X$bL?k8eg>4Vc9aJm
z4*q0J+kvXNg_BV-H#O>BPwW>@dP{rv!!GYWUF$X#8bSgC{hH6cO?8|ip7OP%<1F!<
z6!E-ET9ma*j{TxuuodUCEtL3-Rvh}H2La!x6?bqeZa=<4D34m<zfzwK2vmRlL0;Eq
zL!bA`tGC&}`jgd-jzh1o2!iVV$#qD_>4gfBj>}i5L^_ThEK{m`d0LNBg{J4r>I#*l
zMB^)}O89-GaH6WGS7Smbin;O(m93ZNX`=~oo%MIs(8Duo7vGc-xV8z^Df}hT?xnQJ
z<*KnER^Ik!t7g3)P@PfjS2uq-UdtlEzF+KDAci49T({x`F8h^-WC~!Q!sE+kXJo`e
zN+!LJcC(*Z6=}-MaIj%jk(EEiDRNz|H%a>lxEf@fXQlom#b3a@P;bREf*%#B3zdD2
zwtf8pT8!F0IH9GwV^H!7=08B-f=*wM|6CsKDN%A8$7;SV(erSBPI!L@5A}F{c+aRD
z@ZsS~7!A&+G@_Sgg3-wRU({evah1b>=U}w;xB36?5E_zfyrP>Ujo^Z!s%h7IgwGHX
zkCx<0;OzhK%Rx8yxzMabv!bD$0`ukSl1QP69K4n90G6qkw_IPQua^a70v-Psft{A;
zkc<%ikhr0FK}4ovLZA)21dzLL0^$Dw!gsLWmtl-Q6ahJxaTg8~12Hr=m%-QqCYPu7
zL7D>;R@;|E_dzFr(k{;z2dk@VBls$-2VIEL3Uo|*X_bB>udX4#;~jO_-KY6s*DwZ<
zY@mxYmwhl-c&tO=wMkQNS7tarF1h|m)_ooaia>Y|jqZm_@E6!%1-l0w+@?=S{iIrQ
zoA4%hF5F(5U+1^o-@9sIR6-ly8I){k*ZuKz7hVu(VXW_e5QPWozWv#K+5OV4nxWC&
zBetVIr7>EM{E0?r29iN5fau+5IkVR6OC#r$tVmsmtE{Q^O7Jpu?|>sHHq`?pML5WO
z|0zjCwnQ5>vc*}LuotnZgJa5Uogh(u75u7I^dos;Nw*T$NIqq_nSN(oETuEz2hmU>
zgGyu^uiy`V1+Ulc-fI%=RU+$>y%)o#>WLxtMmDHWUDcdGvajq!bH6Ha04;iIzseyo
z!GUFnBi6MKdtGFX);kwXf}YV^RWG7-)-sO$rrI3>(gM0CKFYsm(Z1;GL@)1hH?MUW
z1<<1F6;qCzm`EQ69C@q|ko~?jWFjQSp#tg%g_8Gw82Q%augs8d=bfdUK<bm>DFSQN
z2@05Ebo=EM2!V1!F%33fMZHE~h=4c4puPqQi^8Fq$n%(BMuWl-w!6zq^<uVv4ooRf
zJBQ4z#FtZAb>9NhPDX!yU8XO`B>~zw-?g~BEdewI$pUsY3D2*ZU6*i0A?;OXKo%p4
zwtqr@z?Z(tNEOGG$Ol2_+GQW)W6jB0wwo8|B1FQd8W!ysw&iIcR(u|YX|)JUm^{G<
zl^W(xm7h(YH+ySx(5zWeF>sKFHnU>XVTN6m?{jyA9W`*ArPO&~p<`$0Md6z&l@5Ki
z8kD(0(BL$O7Ph+#!ugOfn}a9OwHr}cCav;+bc6Tv+IF_&oLGSS7md{C*DR46)?1NA
zvBynKB!^+u5WXw&RGK)nNhpq92Gy7<o~t^Z_evL?MwC9SRFbT9#AIa1;VAljD!>Pc
ztgQuMkRHRcD#^|+*;CfUIU(hD7WL%#uCXoZ!7xLb^`icUBA%S*jOIo}1KAKM%O|0K
zNp}68m_<e-oI&c!%wW<{yE4iV^Y>)SMWR;|Ihry`or9y2S&b}hG2PQbkY+<VyHFB0
zfuoZx(z_n`G+!1ckQ6=}6VT^nW*}iCwR+szhT(#zuUHS6w3rmyU2Dh6+bYM>D{Ywd
zr7&J%(rxz5ihk8kp_-F5gJ{$17b$*!5Lzv^^1%avXGH#R2M>!rT&${0=mbn-kF5?P
z6!R8(`=G-J#r&bIx<3k1QIWVN9<0I3q~-b;;V2FnQiD!%K6ebW)eL<MD}b@<KX#^p
zwF&zaQeyHa2@_pEuQfpL8Aqz!VM78^Oh9bCI|ztP&6b1jg@8`=XySwI-8BM#8rx$l
z0zxJrN%$ZFLM9*u{<9k74>1gKOn|jB3}bq8Miia;1Cah?bpMomJ~|g1^fNt^%%2Ua
z^ZH3B0+)DpBIW(KLc<BoyLM$R4@;qT8cDs_SIzVct7oH$PfMw|+=twFVyZlsI&Na2
z2OT|B6e%mLH~;s1|6b3BGIT?Kv{BE8ybeg%pyz8=zS8rxa%|u8fdkPa`@3!V#%g>#
z)wK0*YC3Lf1QkOR$TYHSScZWiuJHXGWf+j=UEP17410fX$mI(fY;3euX$CTMYY(c-
zfR}f)UAVLz<DG3H$CK?(gMa8Qu~Bv7|0%J3m2}Of8xj`tmFkbqZ{Vv5_RgX655AFK
zOqXGdKNJBumqFJT6qhGi87P;${6U`q>6cvnL0x~jtb;8}8BiidHOYRyCXFheY!*+}
z91B}XVrC<Yq01OrE9+TAmsFj8eoR+y?=Sy*sS4LdSQih@1cCn#|M~MJO@Ok(`FPM;
zL<NVx{(JcS<)1TEvgQ1noiUqNI%n-*UhRan;otxSuyI+_`>I0^K*IyMkpLuEzyA8;
zFpYow{Lv)zUO1)3J=JgNGq_}39~^38T%MvNIf7-w+vZ={2k&y2>>%>8W}q4t8$rxg
zQc}p^YFOA)4m2R`Wi8rM4quC5t5tHB960$3dY=zX49o0WUX`yRud1UIeh`@U(sp&9
zx09*773aqpP4_@tp;>}sJu&ELgmGrpz>|LrL!6O!Z3bun9P~2}DM6YLk0k-twDUUX
za-k!$&`}+&+A7xD)eAU--VF=SU~e!@X8T6t7aGSgd0mb<xvK^CMH?|=w1Sc?pgC1l
zt3uIlDJdum|68?7oxRd*PV;rtHD9I?JKROuPn?t_U$k0gE#>#T%D3o$G*MbiVH$tz
zoG6_X9`&@JD1}Vv$)+z!S!)>iNy#(Hz2s@V6UNrsm}(sb?@EPTQUR31+nGvAt~w@%
zIDK<cQ?;+j>New>Tzd>$>lR$+s(l~FB_U4)?yu1gR}5`Z)mtc9snvR+jTb27y!x@K
zHuC%owX>O>6(+`Cbry7M8_i55^Zb9t&XU`n_#CdI6vo4X>-(KWOM_X-FDOhJHgCPV
zb8`!lB+!=mwIsnX(sviEA&c!o?H8ouC{l_)1N%=gE&f8N2^hF5^bhZ}$S@cHR_mLT
zf2D;1EE`&e8piQt?6Ge@QX%D1#JxPtmeRs!0oD5ZWJYabZEx^OtG<+KdCY$l%4i@*
z(1lv2SY_lAQF)n48PV!np*#XuYH}zjqQs@X%uXe(jkd!_qeKl`;TGziByH9XjO|;;
zN8CFh^!%rCP4}F+d2G7n#8oKs*`{w4lx&R`P$R~Lm)lG%zzTHBo(p%t<gTr$eM@JS
zys-~k9h8sMLD}QV5o_ak8I^xf5awOS;-|*qx8=~N6XDSHCt8{FOHpZZknA#^yBh9M
zvU~a18=|b?Ufpv?V<owjsS4eb&|p8Juuq=2dMN*G_P<UGFm|S`D!=I{o;&+uwM#u2
zv#jUdy$0*5*00f8)y)J7)cbS*spW|NQ*?kV3Fnp`bgk0VKC{-65Oja)G4K-%-a_?c
z8vVHgig2uW5IWY>xv!zpZQR#|OqF%R-xX>#z|2Jj9B>;GP#SFZ@En3#ic%L@S7+{#
z*E$jWc);lLRHx+m2u~;yU~6Wr`@5h>*`6n^Ax|*r-a*!ySwCNdG=apt3HpnWCO}t+
zDt<!DsN*j~QW|BBGi!f4kd)TgT*}2A&jCL^2uT@Cwocl6A*teD4G3V6sRje}fiz1{
zQM;Pk06O|F+W-b4MhbU=P0k64Q<H4Z0-KC-cYsYcU=h<Z4;yV1rdB+JHoSZ)sZAH#
z`h#`-Er6VCKOTt#0Mo3xU%Yu0_p^T8-f&rV(>LvX?t0McI#7Sqp97XY+CfAcx;8<0
zXi+$Uevj$8YP3obZi9YE8wVGhd?M(Fy^V-rvE$zh^!v4XrSTVyJl88F0pl`!3F3zd
zLqzXBi1<xZJ{|F!xqcY&n+SXX;y3Mh1M!<yKc}}S{om&UgGOSFQ~ra%;6&xqfx(&U
z$AQ6_!e;}6I$D2$CjU}oP-A~DJMi;)$e_m9R&ayxt)5&vEmQ+8b#py-npOC38~Q5>
zW$3TFs>;rk?u7k_hpCtmoeuWv_L-U#nwbh(8|@&!j|PSbN(60Yojwp4CgNcb&*d`(
zs50-vjvo$f_6n<n?m*{#81ud)1tNf-DKp|m_SlC$KPrD+IqI!(P@AtQC~sG-%@jvf
zY}!r9-6}Tob{o)rxVm~P!082F>(#aTgO0+;_6i1Jixiz5wn$z&VT)v4Y90FwvHk7?
zQPR8<Q{ck{jkxA9oLAbeYNPq5d*`Qrm1>6@ay#}$nnoi@tX<_RYkC;zG;NZ{<6&{q
z9hm!{M}mJ@wMuGx!_q3K94z|8r}Kp&Rik7W>G%W<!r<g=W7B2oxXuSCVZuT}`M6gC
zU?XqhOleV$*ZcbCqn!%#aC-ggez^1d6%7&vXf55^OLWSi&3prx>I;4NLH|yVZ}O?}
zq>Mf`?9wtYfz$)`Fy5I}>YxQ!7$Aq(d@KLwI(L6=6n9?wQ|J|4t7N)bYanc}ic_A|
zMF)di+%=syLp^8izSP1jNOTIPS6JA6tb~j^H2T`zvr;$qM`!eEvHu-Xebe6Y;~~N|
z>dRY~*4X5yi{2i)gr})R$uoX<6y+G>p?&fb%d(H_?(EI&>V22-p;bj9<d`an9+&vb
zFgkx|Zb4mDj=Sf=FgzTzgFL=#R&7n+i~{xRS?cm2s>yRp7QaLnXWxoOc=@DPS}Rab
zukY$h516o89D!Z>ID)JNBq`O00omHgcTj1Ab_%#vc2<1-R%hb^xLtLVlQYNi*tqA?
zZ*W;VYRAt=R@LgA&@IWC^T~!ItH@4;wMT!S-LWaNKDX~tU2b}QYexf9Ra)^wF0HO>
zk@&cBsdrRdbKSdpQ#4G?ZR4+(f!?`Er_8JFP3fw<p0Kl=JI;0P!q}SYAH{}|n`OLr
zu5;<)%8!q@XX)A1a;*AxZ^sV|Xkwq|t3Ean$2IrBj>zctu5ovUvHj7X6DkZ|e77;7
z2Lql*XJc^(Si4=l*666|Xm*6K1Fp_c_g;eDhg8}0<U)Sg3(@b}`^*0UI=jtlmtl-Q
z6qoCdLJgO^O)3bNIXNCEmmt<11eY!bLYaSrmwWy5{kIoR0)M<}cS~Ah6+@Ydv?!@Y
zZu*SWRgg~GFjD_g$X%Q8J&!6??wj<yLMgSUxjIU8Pj5yW23%)Jf}YeRP{?`;wAI_I
z)SnXFz$4PPBr}AWHr&<ojIx?YLj@Wt&vj*Pep1ug!bsQ&WeZJDw93BeQNnY$PaJ<_
zx<~`{GqwJ36y!zZ)s%T*#1|P=k#Ep%k~49LGl2MW(t#!IJ6O`b;+h$5dhLt6T9K?U
zQoOy7N=oL9?<t{GM|;G3!6d`Sm+sfw`^&#CgE=@Rxj+5&{UwYm2~0Y1(pEF4;Pm6~
z)3=x3d#YsfoC%F%nx{WM!h~kzD>;91R0%gm>&75~<LcDO;DVI%c_ADf&TtWw!;idP
z;V#L*ILrdkTTmJ+7=iw;#4S4v`+?s-2Mfwnk6q5)wbbau8RvOlIPFw3@--ls=Qp=W
z2&QMe4R#+r#3pYhziKgYW#bh`<kme!*=hWSZyd57j#7Zk+|@*Dk{N#@ay);uo|fWI
zmR~#141VVQ$cQ=-Hl4T0E)6<7^wiqDI8J6`nU&^1op%(2hekb9%n05ZsoUhWK{Lg(
zb&57&zM@!^JT+1%Ff@Dpae7b)ui%OvOpdIr#<%WWOD)J+GN38RA3r~rj3})`1%dOQ
zxQ=E7*HJiG$2>xN$Eib^GTVPyrsS~7L66({<5LX0?J4G%*YOuzF{KB+WY5U>lJXP!
zK2>&uI^vf|xvWTFu*5K|rNaI52dt>rZ%c7yMWAq&AJ|vMon0w@ex40no@tC6LeL`P
z6&zrSaWQ0P+O<)SU+!So1N&zY>4n(1T44>d*6vv@FVXgiy^+DMq}zWet;C(wUJlX{
z7MJXqEG=l2?~n;S(7Az2jP-C>td|~ZDrI^u*;rzn-xlK)cg&bUUOvS=YJ4kWqQJa%
zo_8G9a$w%cqp-H67Ar@c@HJG9;+!y6*c}*)BVYy&D;S#*xt;wCB89zouBW($NM19q
z<e=T>F;<}*2vY}QY)60bHpces|92R>ZM7L=cP#f%jNK+}#@JrkpM$X-m!E{O9hWCC
zwxhI!vDvpAN@q&XKzD{4YT9U}n)DjR&(z<bnpjV{&fT5Gby5tqs_|K!2yl^NGjc<>
z0ySc}894a4N0ku>eCP&l(zB3`DL2Ty633MB47eF0+5KM0KuLdD2F{I*k!bg>u2||G
zS(>A$TNmS<sllk!hcq+y!1`(@^)ULXJdBAdhQ<nN%JiV?;LRLn3F|uv>ynj4Y~-yc
z2d)5z_c9csj}ZQ#oS>r?v=eJENR}6KgY)Cj%NPbm4Aj^10mY0RWO*ifQWe)xT+Ihk
z_R#ZR>LWaY803G?n`96mLW6Q)J4RzI>BmEknPF6DChUS=Rz0!ECV0YB`=}OGw9<2a
zq)gUR;~C{woa5?w)a@bv(bM|QWT8PZ38mYXaWF}=Bez17Q>v#rVEF?SwPi-jo?<w2
z2`fegAZFjT{r1VV7|11)T9g>5kI_R*^$0z*6d%z;`y_wA?lyGk1(RMaSbBAS)8~Q|
zBd+w@q=Y!0r-d=<ol8Q=IOa>$l&!ysAu+04=Xe%7cGuz?i)Uerv13@D1H`yFA6m+O
z6?v@asv4VnuDdI(j{6F-=d7ul{Y*D4;bcc}6;v$~pa=F#O-9=0XACYB!PYDooXn7z
zVG(K&tCD}4vCL}H5WUdWt&PsjfLzR=qy{aNm*?`a&xOqMN@+Yvpk$Gtc`hBa5|3t1
zop4q1=M+i}DatNgYtJ++gfU$o>(1eej($5?rY30J(O_|=X)Y%2xu|kX>$!=$s^=XE
z7BJ6k&*8vwB)4>W4!J2ZZA&wqWMF+yMZO)<QtyAL-f5}#GWaTmr}j=f=%*A%c8BDL
zv<`CXVx#1UTPVk1M|OO<A^zs8g`Jd^X+O3otwa%l{zS=}k%hueg2n(aCn`E$o#j11
zIeQE4lND`p<GN!nGnZC%qht?PT90soxrJ9srUt8TVndBRMdBtIjyrXI9II2oJK&bP
zjbwjv`FOidIc$}F+r`4VX|EJ{<hU2H@gf&W-5q;jfjMRF+Sc<$G-h_pxe`C_?C-Wi
z-IPpslpnF;XVyhNBv1-3T(d4Qw`A|GA-0h_BJJBk&x=aEfnL;UyQ=B@;HI2ey(F)(
z$?BlU+l+i2!$n73NHTS?{=QwhzYOIX*eHK8ye*)_-k8!0eB5g^!fA%rpzz1rzLH~6
z#$`|G2oE6Jl`g(vR#}zu+pLOe*N(yxT*k4k|6D9>JM0xpYdJ25(zx-urkF=I^ptxJ
z8+*s8*6JU=jIOpLpU92r+lw!=m0(tls=7YTmmINYyByfo#G%Bjr!qZ^HaU?O!ZCji
zHP|dgi<Ir{j<Y#$ZkZiwZ)DLegN^2vjkSIFD05RxNO5oJPhe!W(1t;I%lN!6!U5jY
zGqrP-@aJ}LPVnr$Xh$uQj?=Tz&L~uZ?5Xl=(-UrX^te<gdlWZ}d(jl%P~LTZWXrn{
z6J8$xDUT#f*&8^8)v`L0GUdBqW!!(}dz9}x)V(*NA>gZbKcPMhuS7`Myx7yj^)IR!
z0Wfw6_?0!|O6|AUYn!V^c>1-{wS&~Bb?6P+>-_G|dW-bt3F~rm^0=lZ+*b5(TKsqu
zaX1A}4bplmo0GJ39{sfXF{9T?ExpWKx=Nk0=^sVvj;dbi?&i_F@aEWj<=-iP@|Nf0
ztgP^GW^`_-8lousm#+`8ORJc6`YwDe{!dBw`S$+uANj!7m(ffL69q9dI5stxQEWja
z3p6iGWo~D5Xdp5*m%)b~Cx6viU2j`A5`E9FAh0jp2lJ9M!y&mS0;J8RTfoHzaSH5i
zU<8h%HtxETA<IGb-|rb-=}L-(R<;~S0rD~TYDCVEGjryUid>XkaFI)@PS^ykF6L@d
z7DuH?s*Yo9Qg(^k7jkqtNt4LsQkulBaIaS!zW)SL!Kfj&ub5J#{(svEVxcCgixRYn
z*+u34Ep$ujN3D)RX;YQpi06YPM*@?pDD0F%Z8CL~D7DWsL~R3L9@5_aim|d_8m<8Y
zl$4u7bX29eT7%=z8&;!Zs%?Zyj-@xOvg4#`vsH1NOIx*?oP^LOuIdECH!(Y*Ol`L4
zL~3stmO4>wZwcaDfq$K7OidJsz3-S_a)_9%Mko|+rV|FMb1&4T5c<KeH`F$#hJtNU
zZen#QHHRt|E|n0Q7+tC?O-xRs_HWfGh0tcp5L-mlSRfW{Vs+S`HXHU)xi>wMQ|V2|
z?yL3+7+o{>y6m*H4XA{-RhlmLDTiEbHd2>4^uA@6MS9)hvVUqnLbVtxZCnIdZ1GZw
zAWCpm5iTXRt*BUmLexITJVm6fNQ$nc)Ldh{RcV&9B205z2CQtUy$M_#eCnO34#u@{
zZ5}C20p?Y0fm|_8#T-pUFxK2$v%LwiW}8l6bs#o_q?cQIwRj6m)ZRB)+jkSj4HBBK
zHgJ&CRwH4Akbin(&7z<!Cu+h7A;;!$viC*Wyq28!pxK^U=G8VYGCfA-_FjX@wu=Fo
z9pii#EgGWWO9)$x8&e-N$&xnp&YG=JOIw5X)tvdsx8HWp-&pns6qQ%qzyAFnGtCz$
zBaJ<l(~a-0ug{-8g{tempIlF-XEzs@BWG3UmFr$!%ztL1>DW>K`1en<AI@eMvyqGa
zix-pe43oUD$hDGSv%Em|Mhn&Ufc%BIrGJ>aZ4XbMb}y%s%d^q!P51Koi|+O4)2z3n
zd;Pzg5oZ2%1cS!2(Rg;-s!s32tI_S`ZhASoomVw`_tWV8)x~#{Pj5^`5&gN+IYyY`
zyQW6H+JCe0crqVLdJ~?!>5HlP(ic-m=YtLH3$FWaGJQXqw!y=B_mA$!?t47I$#Yn9
zIeQa{e8RqpWUm%ciM-{CB=J$|`IGMK?%m7=JiQwKw|h1We>uCFjJvb$=T|@4-)}z5
zW;eHg>$+dBW*_d}`OC@g-EVil-<-T3ee8a^n16lv=gsU#EXed1?2!NMx_fs$dDnf6
zo`R3v?ewzyd3JSuHM<(!c9WaY6#E3NzntciQw@KGw=2tgPuEBl(GmyOP((e7@WEX}
zlExGKE3)*jkN7%7ya6K$^fJ&lG!{8w8%QYjJE)Q-Z&l^1NuT-XYw1(yVd+!KO8TI{
z&VSP<4~dlXAeBL?3#o*?(o8vFmzJ>zsPj1-K`d%Wh()EmU#=zi2Yed*OYjdp)It3F
z<!;<Yzil2;jL0_poA&b%eU)#$<f2e&@to>$cv#Fs;{7~CkbAAmJp@+!Smh`@WHAr9
z?&)Dnh|1+2qH#jh9>*T07NtqIdAMfYc7Jzc|IoY3)n{w+6E?6>nBc{_Iu(!R^1OTI
zF#K8f!{qg(d+xpgZ_`AfR<w1=8un^dtYAsu6)aJ2-N!a`T2t#GM_D{WeFs$z?2R<-
zplV1S*9<srAgf_ZA6OsO2O8yE0$GU*fU04TmhWhgc#vd}1`w<R2-a=g<bA_06@MQv
zkmiW+16M|+1j7P*3kbh@ukd>i*2e0i@$3cGQu4}tkSxC)&2AU{EYoA;cCNmnw*&Y0
z#-vFK`gH;TBj;&zalx;&8bizyDdiv#wyf?0w>ZV<tDTpSdTLgot*QD@EAD~Dz^05)
z9{P<yo@zSm9y3}giF-m6G5FXLp??_jNj!*OmjI{P^x|r~4B=QFA%rCcj~sQeMsG(H
zNrER4M*%P;vKc%fs{}iRD4YQYiRdvxm<b@6T+HZt0G2J0+JRy5QZ^HbaxIzgha(3m
zIp+PUR5VbwlhOH31Nu=Nk7A=+9P4M~wEihMt(%NfV{V<u4ysboDd7DMs(;qDk)&Nz
zBd~zNm#4fd&B0pRaPLkbZc<)58hucwKyleA*coq+@peABkjfw}h_0RQE~K~@<xLd{
zWk=%+7!Kk!m4ih1$BXIJ#k=d#vXDIn7Bf&#+c7%GZ%RWUOYk%cZIlpA94N7($-(xa
zn3TL92x`U(uTatq>^Kl59e+R&8E4BYd%~bv&$jI0*gF10_i(YD+wo&iGz0>A<LHQs
zwQWw|H>cQ8D3!R0J|tjsA)~BFu}dNedxZ|+|90rVI2%Pf35s}-^?iSJJG&TPj^55D
zZ?V?fQq;=3-Cga#Lf6%L6iC2$6v$mvMb+EY<qoQxjIi=fs_2Vr&VRe8Qd2$L9Y^;<
z5f*kc*B<OP)p$n#NwQC3|6sQ~cCFWc!q&dczoSM7Vtf$(F>T@>hlk-GuiTi}_2C@X
zNCqhl(twUMpd$_F*ssi<thn7qfE7pa#jcH!=@}W{*eNn^lQN0Im_(@r%HYURdH=3g
zq5_(b5!Iqi09_SCu73iY(Wpgzs1^6YEQPzYhFS553Qx4&w?0CJrwl4=zkU!n%ywJP
z*#Ba`C-9WF>^HM&VF7(s1thQ(+ONcOMA1vj63-ci=+UfnqLWkI^FTMUv?Hd0M4<!I
zgEc^J-%sX2WPO6}=bEx7yEk8S&lUNZ5AGS+DfMU6(0;G+$A8ptRbvfkm3L7k#;$&u
zd&@PfTf@gStXoGx<WgXrf6^ZE78HCmN+`Msi6Re!#I#BYd4NA2;ExCR;{pD7Acr!L
zLy3b#gT(v8Bo#DOh=GJM$!G>RC0{`%!3RM3H>xJ1ObX+ZA~}W%a4q|TBt`Hjuw*OQ
z0SuOk`ee^asDCb-&AH^9BnBgQAiXuCD(#1nsMN8p@H%0CnACqhxdUr{eswu#KxF91
zkAaTK&i<^2CUisKdjTn~6}*qU@RmNrSPw_eOjzI&bl>AQpsjDej9L&(tg%QsidD9&
zz?2?imFIzMmj-OtpP}6q*b!&dtu1E`#g6JxCqk)Fcz-{ZiTdu5mrHmCiq%X<vB)``
zoR;{Ap^P;KKL^zl&v?Wd=TIM5u)fAN;qwJ9_2jhjXRU&yojClO+WaWF00=7YqzX=9
zjBW>2)EdueCsnH$G3;b(G@zjJB^s8su41KOxp!BQ@@7|&(k@*^Rr;)}P+sXO($Ej2
zfpRB#1b+g(Kt~WOtT};4GZsJ!qHz;2>?gAce<_p_B5R>Nc`N#mti1B`AHuY96qyF_
zJUQj<cXRhB)|I{e;%?xJfMhPmfrxCpgIDiXTk0kKGvcwLu`DFAi<<oqwq)K#HKH_W
z$vdf{tqozfQBiLYxnIwZTyzUhQMpG&Sy>PLLk67qKdah&moX|b69X|bIhRrEFB6wv
z8A8PY{<kF>LdppQoevX~mtl-Q6ahJxk(L4#mmMTRD}Ng~620py@&}5|eE|x>NOF~d
zxw%od!5kV!yXIgwdO7s(&*Pe0GNsf~KpT*=l1awKV7y?Ebb6oY^m%uD+x_$^<Yb6d
z%<$>$+r;pJ;(tcbsMTZ`BU-TO?Z@<=XG*!o?}(qIR0#YG{<e6k{gYr8Js5g0v@*7T
zGLoT1pnpZ64_A0n{5bwLsqHn5*BEc9l%4`;GjwDp{0{s7zWsYglfb=n+DHN!vUK7=
zsf2VA6czXj=g_Lqs=!-<o-*ml@I36FzzRL2EPWxXf_xHkDVHc=EPX{^9{L6<L8>*|
zajJ=8u9uD;E{w878`F<cI?Lu|^4#UiUa03NTYr%ARHQdae;H)Vt*5AyvRPkmkwD2)
zd@b~|T`D_yO0ieulkDZvms367n<;Q+b>BeACd-=m^uD3BO}-Ii^JDEYeJ{D6g*|7_
znU?2i8%_34;p?o8iWKKJPemT01l4?|FWz2BVqRG?a`y8!_@3s3TAt@=lg!L?^O0Jz
z<$q3j6-gcWhkiWD9JC2vgB^M4Yxj}LF+RiCB!NuUhs76lkP5z5mQSm65<LqU1RPp}
zD2FVL$x|YentEMiokX^aBG>c>_xZFO6u2NvVcBs?IG$0fBUQ`21t$h8u<ad!(Yhhm
zr7C(Js)=hSjU~I#+99&`p+8HMWnPT4?SBiE%u%(0EPYKf&0l6~5vRQg{Nr=?_VTv-
zV=#=8mN@68uRnJFgG5l1Mo*fCuYdkJecOHMsn)n+8KNB3TB!zCgd#?>NjXOtE4Wf{
z;gp1x=wAN&^L^@Cey%1Gaw;x+ipk@LX;)r5H}G#J^m-YzBL*heD1(piYOsOIk$;FS
zyFER&YdC9SE!@7uA@5!R!OV;hh<1x3jUmj*62U4r4AJbA9Oz|vwAO?RRYr$}v$$xk
zwx-p@GTC$BblI9t5y3^@TP!XVw7(>Ii~wVn{tNOnEtNc+E5hg?k2iHcPm<?Q7yn+j
ziXXqI4|j%0KG-+S;Wcs#s)!SItAE_h{GXD{d{s`N8R1e5en3Zm%WGaK!VyBU9H*tg
zl(?+sr8zor4dfc3>(n6aQg#pn>Ky=dZK%H-<{L1uRR%VLai;9&$<?45`z*upI$$(<
zh#A;}e0lBX2x8VFS%O@Jd_t9(y#%>!?}tHd@!4bfQsrMIQ;yVlhSy>vk$?IH_+;W3
zRz&d^H^*kFt9=BYXFKQTLAujNKr-YDp}vD;rm@WR-H<F1c75P^93*!XFF<n7|1prf
zQHBJ`U2nbuiS+nTz~jQF<7VLA0v;vWVJP+Exr!8<mY;ptbaZKsxzt3M!5VztuGO5@
zJOz0&AZDDYm%F?LkN~eX8-HgC(Sk@edaKJjlxpLu(xUVQhA#W|8$eCjeSR*y=ZdX9
zCKT4G8zc|@C904@)ALi3R3=cp2QH;Ts%P~-0WKH9v|ZZZCSZ#tli-hI$;2SeU<vfv
zl^OMOajkEhEz<Zo)NbGy3ob3t&V_~9<0|k`m+cbXN&)7qC8R_n=YN+haoN|-YOkn6
z0;hT|E5!u~-;+$$FdwvO=e~{dwHMWetZ0wE#G-pk&5&NoKk+StZ{{X)k+5pZY1!*g
zh^dNTF(E2nWm}Qh4Cu*3!f@;WWdU@-h_IMNhy9PZ9ncxY+yQ<WGIVX$cZS!4ssN0D
zX-9c&XC@`FZs_dhvVWE@38%W!)uI@W1G=Yv2H&+A1Q!?9Ln!?oOY9vCOeqo6p5H*<
zkUas{v7zdyuNoPcEC*Tc$=<NBbJX+{*DQ<eAJcr`e`oe)v9a(u;Wim8w#;I%!MW^h
zWk!W{M97Eaz}OmTW_yaO8z}??1lz86uppg%)Ynu)GH_fe^na{q1)LNQRjMI)NvD?k
zosJP>$~FPg(r7!cl|C|t%Y010Q%UZW$~s*J8B0C;kSHCi;d}(Dyo+9)wvvpl)!=SR
z{xZr)MWmW5W2ZgR!sqG`g1P}iX*zJqky;|hXEbFrB9Cbc5VSf#YmPs1Zddrs1}Ni7
zVS_Hy9r6Ualz%5g4vXe5TB(fhSzc+o(+yo^$GN^<D9_njbsSi~u>~#8=5Cx3gxhYG
zf}$!4!+5`LV%ud`j_%ocv(E(Orz)qs;66f-BvvYmyf%kQ$ZYx{fLKBiCy>|tF~Lti
zPr+H?zrtBH{XK<d>?QHf7cjL1Hshnx$SSe^cR&69{(tRXUw?sNruV;mS0qQ+;xcEa
zAMq85I0C#B)A#g_HEyIalLs%Mr9Y^^{s0~Fz|!z_+DXJFCX}sOzx!54q?O2{=fNQ*
zGw4_tR83h1PqDOSe&6fHEY-{+=v2h;Z=OmT>t8WGkMghnSHI`381=tt_ifsJNmf-_
zBf~-Dw|~AYi(CSV$R(>XBv3}M+OoMxSi+o1N<d_1;5{p?wZ5g16HKMIAGgF%hQA88
zqH}Jk^v+G7l_}(={OsFT?9l0ombu<Ul;4J9Hy&M-CF@2D&wqdYk;c7a)aU9M={$Kf
zluA!n?jls+Kr5u>$}X;_)IImE0r3=~YQ*lyC4UF^dUbUK!0yXbi4*mDnic!9*%(NR
zjS(A+(PQjv1iu+KAeD`<5>!%m>;j09dTju2K#;$4=$u*z`p)T#wb;^2=1pR(hNspH
z7JBZNxygKPqt}mF<Cq?H5~XqupEj7UV~s!Qbyyl-gI?7#>VInFpCQSym<v5|Sok?!
zut{~5bNGLwn1wqHGs4?cGS6F@<+V~vk5cTKge#!c<(zA4!gaoqgWOoKUY}r1qyZ&8
zqC-id6cvO@cb9~}$FzLWp*ew&ISnz-^GQLMyDlvEYGExc_j&2Y<^J2$c1w`_RAKUl
zdtV&6v(s~s67!v-vL<p8cN)tiVWZ-0V)`3MLg#-ImvVNNvfyAF-CHdro#kZQ_b}a^
z7rS!jCDL0hO6q2|W~pqNW1VLyq@}toOQsZ3=hMER$Dy={xipmTJ1HZYD>k`rNj}pV
zQIq8A_Cn3F)Uwr~FYOjR&S_nEh{aMK2lCVEroy(w);lnlrXPLb;AGV^k$_pGi#@gK
zHe7%3vy$v4&oJEurclU6EN8j$OzZkzU8BXspRGPifAPQ$#yNY9<NOsstB96t$=r&c
zm&G9!Wo|IeX@<!wOf8VcF*;E5ZY;S7HN3r$O(olO0GOk80YZCwmDHD*Om*y;s-*Yc
zL`uaG&NR5l2UbTqiWkZwJ^zQ)M>dL(3M7AB?`tn2WkvjT6RNJch?GV$CyAR=kF6Jx
z@@R{6*qV>Hh?FX!9f!3_Y^q3iy&riIiBi7->p6VXMI^(C;)8u}xQN6pb`pYZmHXO@
zNd8bl*uf7LE+VBRk9{LH?)JWmNI5u3?2cTyh!oGADjnwAo;Srl?Lkj|Q0QYm^Osus
zfuXOXcp>!l{2vnfHj0o!U)TG_(C25Xa=j};pEI>5UK#l;KwrC!V1G;CYZ+XKdu`_r
z2zxDq4~}{rv)hB-HmPrnd7THWhdiUmoXDLGc>NP$>Hqssh?kK~EE5GWGdVaom$7X;
z6PJ)Z0VtQ@E<&Dv^LF{?ufM$rV)(ZY<BzztPH)Il<rX<LxSP2mbsf1gE{>GHH14Eb
z{In51QzK`_#@~sHd$O{|ZXitAUAYU`aHb8xooO@Qi0e$8M5j?eCXQa?wsj4P!9bal
z*uf)d6r2e}&E#?MQD<G_lS-E#Gd~Afd95Seam)TpWLW@z9##U7L2km!NdQIJQIU|1
z<w21u88#onND+-Hu9^g|{FVri+BNYnEuB8eFaPB)Nvh_Jo*<7#WqC&4k_=^>5zR81
zdptoh&or9lO*G6!$}^c5{Z3kqa+c<t^_!&sL7yAR{+i8tlg6Xm#HR7%l~Jd^(9q}6
z-+%7Q-ssJLIy=fB@uWzvw|vg@{e#Y_2glmcgAkk3a-#0Y#%^>*U*&u2q^--ZC?7ai
zcMLLX^HnEj6K^?8B<XpQ4Z4EzC;?@xxA<*?kXb&<Cn%%JwTDgEFxf??(QK1C%HyMO
znL?T=e~CB;PYI@zRoO*r*fdO%k5t92oJ?IW#-BTX36TdUPyZ;^Vn;t6<=6WT8ztYI
zd_vkyg1dkAZ1XZ07J=dp(ucBd@s&1uozFl|Bw2WjPq#W+qDr#LxSKfBPPWX^VXzVr
zpLk|F@%$i~7bS|!2czr^MOEf$g-SVzFZ;K6%TF7lb^hzfUWjM1_CZ!{hfw56-|Qq=
z>Ut@E^+fj(X#+iJD{avg{eD|p2k1@OK3Y`n6Q8e2$-{gTXDIIRc|MZ?J@?G!<M*xE
zL!u_?+G{y^;+xj@6bnx!cC&q0_Dqq*rYK%)h3#1UEX8_NqLvPI$1C5h^UA!F?oP|S
zvX0)#m|9{K3!@sokTA9Jn-*@Q*pAq#7rMiLeNYI@hd)#}R}-yUnwOB961H$rcFD9W
zEU}HF^!vR)o-yb!h*C~)t5nK7AT@pT(5L=#%!$l6?>Jp{9J6%314oS4n{v8_-f&T-
z!Gu?3I!C#YY>i&Pu%tgLxlyKh)y-=_DRdKmrkyp(udX2;I{4CnC<BI;LCU<pVS$Bz
zhY6!bjX^z)YW9;tA>Umvy;16qZi{Qq(=Ih=GAfn#%p=S)DjJ7l&(Q{f1dLh}*E_4+
zHZRJ;Rs+r~VyI}}+;y}n)Z{MtI3CnnERDq4=$u3};;YM14$SUO0{tkC6Id-$XS!nQ
ztGth4Y3YK2yI|lHs?&4zQ7molJ4@kz@|tG5iIc{g>XAZo=WJHF4qBm+owGss|6p}3
z)E2ZGb@feHU2CkFjGU;iuzIzCB){VDcN`_y%{w*Hzvn0+U`f8<YShM1A+Xxj?1Wj1
zQn#vF{C;It(n|d+Q7&jBbcvv|wo*%Xyb-FT>RdB?ajjegXw~YyriIaRGxZpMibEq+
zx)Gos>OdH&5mr<oPvD1V`|-F(a0Sq`l=CBfCHkD8m?29B(s>R@JXp5l`b1J<d==A<
z5q?(~eB=2&1zBB@L{*Nitq|5sdA7jZZ-B%unVPsQXAj5C4&gTi<KonrkTou^Do`v}
zH{;y65`NGDgN+N~*U#av_g^o6|D7C4V={9Xp8oacOMIXeYzku=YQhI|`s?4PKVJSm
zQste*BpYF^F`exI21cV#M!*)lI?!ZhT8mTrCuyS_VP!rjxq-AJWeH1Rfkr(lGx0H`
zQ{q;mMulu>fclV|b;t&&6Fw#Gx)IJDm<>c2DM^(>4)j6&Q|FEvK@%E(C<8oCkt8`%
z3~Na$i-rM{j4;6>@Grjj&Qj+mI5BRs7!60M?da6n4eTTd=6A>ud{3lW8~(lEu8qcr
z;ZAfn4@Gy4au3q6y28mZ*~MZS^m&Dz%afACGEA~7Kvi3D1&HZYc@Bo*_@wU}Pmg9w
zr{>%ce1Tax-LKZx`mj`g2_?PSI-|D6&EvUDS+yyY;vh|nJAiCWo<26^XU(qi-)pP7
z!1-HJpur;A`-#y`fQ}VnNeJSt;+z#`(wHfNgAOxkI&8s0nxSc6jfWj*U?3UzNcpkU
z2i7C@vE%BJ%$uoI{MK$lZmh_ysIHo$W#u&$wc_?LCe!#W)IGO<wGX+i@&OZp<+LhS
zgvTX}y)y~0+q6hE;A<;z4*C|E@Cc28fnnOHm}*!QrM5CS7Hho83X)y6WU7bU74_b7
z!B><^D<3HFBsJq&dydYcE5GcTi;0*u?Aj<b?7njCfZE)gP@}@rUL^`!iX0BLA<liX
z2jtuU(sJD?V7=&nRbn&>0XzEDz{<18&}s~sdIW|FFhF5oIz)yj5|sf``X>pp8~Obq
z^r$e?W0tR2!<5f}dG2pEq9|nyhM?|hYf=D(!kYc$Lq&U1xLj4>6sJ3cPc6kC`7^bG
z)7%T3`{}Y|sQ1<j%ZR6T`+$<pjauWhf!ukHDDG*A7F{WSwOP>Z9=!Lt%-`6A$;k1b
z2~%dMSvdD~5!dmZ$A$>e?iwN$&9X}tnpawNrO#>>==%w|#dj@&?c~Nf#KgU7`}R?>
z@_Qwz))N%{mKO!sNtQ1pG~`FNL^vz2Il30hGx9Ymzd&uF=PUAwm(lI*j?^@A^ila_
z$+{(>wGOX;wj>I`yhG639~rgz_C`RuX&eCG(Jd+zBlNY&lA1trd5QeY`6UK(=ltCm
zC}LS<;y<NE?M6}`BZ#ADE1lmwgbs5023e|~weEVWhj!w)#zdqi%HJ1l<Gjn-y=(bC
zhJI;ddV_VxrRCPol&|HOJ>?v9*Ci|!yu*qX;y%xRgA?h!gpr4x657T0M_{%8n41>K
zW+_#J^@@=ECSp!BPc#OwYzuy*ek557IMa<Q_7b0Y&mq@}`DDc{wAq4Ii-yOQ7u+SK
zBx|fx>n%cYMI*`$0vlNX6sfQ_@~zEM!@45JVY{&kqNPiz3Rgq5XPWcn^eBf4Dkqxy
z#PPp>R@b)+=;4?n-`Y4!uS*~>pTtZsLVhkZJtcdrONBJE?2qNyK@G{*6kegG)0L+U
zsavxP_YM%j;v)H3y|4ATxbQFgS~j}4Wu&#1wT9H!^#Lm8mW4h#dx1r2nE=`+TDfI+
zA+efrT`Tw44%QKu;&V^!D-qwT4O|#0?nQim?>4RUI6KM7Q<_x`*yguEa~kHJUEk?4
z%|fYP1m`*BYQ6bzf&iEw{eH%sV=pcAs2Lx=Ls*n>2a0>bq9iizAIkd|twYGscxnAr
z0Qt6iFdH_hzVloi!F$K*Cd+Fx-rAp`l%`zYjWzG(MkEw~iO!qdh>skgO$jpg3I-m3
zLhB0`9w5SR_YlsMT5e9mKy;PU$)~~iO}4i;OZodmbO7x(1?rok!$7U}fet)0_7wKY
z{?-`yf{96k1&9eFZ#y<7E#}TBuYUF;P`+y6^BE#tPMHN-lJt%rWe4C_9PL}wBDVFH
z$L2&Uh1ZzmVMH_68@%AQUAsXvskQ8XdUpqQ@t-ZFK9NNlkD#qZ_IyG!(jbCgU@6r}
z8kUmI4Grc4V&ZmR;+-^SG$ZNl(FMJkqrPF(UPs%Jo3S(ct0SSk#L!S1M~O4FJm)*K
z0UA+&J*p})FXGA(?fnMbz6A7bs0l)Q^UV03Znp8+Qx9qpJ?V7ODx7_3)gG*WwVY_O
z=&wY+1y6!dn|3{(oL~8!I|jKD9#($^I?y%xKykNYKpCo^t#1q`uB5!v@%}=r4{|f?
zMYI;`?Fz?K6V_dEavT2>(0+LC$~J<&tkmg?J$^6l?-XlYxAvH;VXk5?)|Z0JOee8p
z8Q%v>m~YR&5?VZLroHmOYP2Fyk{=Dbl->aI_dYWnyd}GvmFGzN+59#h2~ED;=e<4g
zrc!=mhLe$E=QX8N7Vw$hP!w=-cS-_2wOy_f0=-lA-1YJ04dL+p{_Ev`iRsxtmtl-Q
z6qj*_92J)#_%|q*V0bwMmkm2YnF6Bcm#I5KN`H>A7J%U8zu(`dVdR%D6=9IbXFINl
zTx{vD&>%t_fdLr@o*zU5knt8jt-^=C<G8g9cj9pzlPFLePO#t)Br`yn0Ln(b9v+>f
z5&N9q^lg-C2Lk1LA*I>{fzJr7k4#FK$ssc6JB^V}!ZVL6CXd-bF!W9^@<dM}SI%TC
z#D8mpFsKuE6mP%c8S(bOB*H}FLS$?5x<x_~1Xy(;vRVe=rCTtIFe3?lX6#r2lj6mB
z%w&pGrH&{|lDP{dF$f9qLQ0I(!br0wc8wSrHJcS-oqU$6-?tP9ya40xsX+aTXB4Oh
z?iHv<m?sz~z=twnUBlKUYAdXY_aM}I;eV_dSKK1hDG{Y#?0<psY5cUTeBJNaW4>fM
zl8d-T`>Yb$UPy_y2qj@bNn}T6XY7*iwzDV_73{~eh{#PNF|8S`iA#a7)v-J;;j#kb
zL5UteL(0qqBugT?^M8p6-AF4>Udcu+9GJ3_D@RFo*P$FWvOS|RTsh51DNXOv=YNwd
zLnf7t<*i%4jD&h7s?TK)&$H9786Vm?;q%H*=J~PA%lMc^k0gJx4oX_Vv^dx!%ha{h
z=VL5Adj8aUUHda{0|sdU?&xNQ75ClD0DK7}HqWnhKZ;8UukRUW4J)1@um`>afmd0O
zjlw&(v>*z~XANgvdMTE4Z1uhbMSrmyg`$&eVT7j*OOH4(2^r}TSq$ZrkW522vPJ_0
z5o9@$v>B~&T@2RXfgHx!x2S=0Lb(M#TWWyv2jxa+IR04ynT{g*a3QwZ046sJ+DsE7
z7mZa-_}oa^<nxe!vn-!-s(hRVqm8iME;N9GE4dzyVK8FMJ*93S3MDU<x_=?9vkR%y
z-X--tad-%n%3g2T20tO}osBu&DW}9KpHPDh^Nf-l;iF^|LmXL}a`ohljgGihCyGbu
z&Q?12G-s4Kx>PD-$K!^de4ahG=ODy<%5|r0=gqnkxA#C-@LD?&LR|8C6W3FXNM>Kv
zozyV%v&07{wwO~{ci<2BmVcKuzFCh`vOg{w05xK*QA&<o()zLCSJ_WLE#fKDR|xXF
z<kd`yif;Dy%8F>Nb@tula~hSB`rA|9e$z7<?||W-b{cR=<T6~n7FLg<IG0rzQ-3Sb
z`0HA7Zo3ivv{Pz%Z7qCF>u!mSP=3*?mDp(2zqE28?^vp1r^<tdqJP#4yS+@JTBBtz
zmiA`pSIf`*Oz|1a`!<$rs{byhuVsFj4odM38N=cy*@nl6P0Gy2Y{T6Ma`w|}lzSyl
zh*Z8S8R0sfAKta>qNE5GCJ08LZ!NZ!^K`#g4(8F9w`CbvVJ5DD%D}%(2>4}6(6|74
z8aw@WO2a~F3z7VqQh(mjvIhSV!VjPR{r>$g-+o0$H@*LdK;?u%7Ut<;`pK_a7_a^G
zgI@tN28H=N)hCQYikxs7e3_QtzfBK-tX0N_I{1SIqksTtx@Bf6UV8$|i#J2OJuO!U
zd9NSsU_q#CFr%Mwpf_5D=h7?A&$H;2p7uS1;;es858tPUuYb%>>$E43Am@dn!+&f1
zR2W`+qgtNkAv~^87x4@8Bzn}bjnS6g3W~WY6&^s7*H%Es74KGP&La)oc?j3aE99Yq
z+AzMshk;fcsYji#kpf-x=Sevgb9d%&{r>GI&wIyNUz*Q(okxYiHDIXv2&fu>1mRk#
zePP3DbBDn*bbmgePt_+|&58KD)0%A!i{O@Drv{~&w#Q1`*NYi~)@tn~3>i<mh7no|
z?<*rVx?GtChHg@f<f2~=#y-ss-zc9`<mBcKTJ^s);oIB^gWUpSo|gl8o&y}BXUE>m
zg90@NhV5IwbDEul{Oq7@H~Tb)P&kLg%zdr}e~gK#9)CRY$MdXpt{T(`E$ui<y%d^#
z;ll1Z4un$UGV*w4l0$`y<1ET?cyo@$izsC$v{Qx+E5g}j7C9KF7CTb}lFoWva|jiw
zNxDl1IAS7<t3eY3_fcD?qPdF0m7Wx-*+W#zKyl5YtL>~=bhXaeqW?AZV0`92Hizy7
zjb;<kQGdh|!I|U1hHPN75$2f5vGN=D)MPoF>F}uUV2({R>+_n9^X7T=Au-9nH_>BV
zxfoj=X{*U-+GsORa}=Fq1GUCGtdjMs?zZPJW+;xya|RE-#_+buH7En+8MYavwT_ok
zQ@E3{sGR#;t3s7uGpA?QSmM$u!Ba~|vv*hW&42n3Fkmr#^Amb&Rpz<Lj&r{H(JKNf
zMj`chLx4(t#;N@q0+&R#N%c2EY7TZW1fh7YnE1XLw8p!6KY<Eeuvocm0Hz(x+|e9W
zn45ziDyVXLV89FSN_wC^FA51|29@`M8fWyUQ!Y4hN<H8Sy5ZER8A}EF)VAC0#^|*^
z%zt{{ER%;w2Whq&q*pW~SXBMWdyL}_=`iD#YB<-oM;x}Q;UJt>i~XC9L0bt+?{lT<
z!<7!D$>1(SODiw5vTJOyx|R~nndS{06!P_Y{nu6mfRE^ZKSOU7M+Sw=zC(cs!43`D
z1y<6CsPB1^Wms{)e>QG^e@#)r2<79YXnzH|7Xc%5>mETj)}5=+`qJLtTF+faQfB}0
zyA?WUJWFcD`gz3c>h9OD@}9D?%?~qfQ6!eZh{J}&BDI*qV*fwD=Wws>GQ4uHZG|r%
zb~@KzZyIr>j?_j^i`<6N+j1nHUeVh_H)AejnK1|z7J|Q*vGIi<?Uj&rYw^+0@_!_U
zQCpLP+;XC_z4R+?o~Qx+kX`J**=gLSfT9VbTVHV#6^vKE>ZVtneeeKH`7d3OOIL>7
zObPW2^VMI-RO))oF7*s*ujJ*5t5@@Fsu6RT`;*G-UN#+Ok(O=WrP3xK+4W-6cJQb*
zMJTYiOm%xvUMnYby&d}p73H<Gn17<XtBUgH(tqU!r8)T8vh1&4S#BmlP(x~piSA0F
z03fau3TD_`D41bNYHgaPYu_l(4XO~RZk~$U#+uxu8b*KlvRN$B7Nb<;&_K6Y6chHE
z@_fy^z4jdz^4PA@q02x&JYYcOi;&7)WQ^46XHM}|BIeX`&&=ncv{CuomVXr(_SiJm
zR0sRq?e;p{^3JQw@TC@8r8YF&*9r`jMk!z8v7M_ATsP&b4|AEVajW_F+{|;kWkMAI
z^WC_%fBLHj)%*nneGj{3LdtrAU93j-QERPeDXCPFZQ$xsiAMj3$;MbKaUvRb{U1}X
zdr^iirK<A+*;4!K3KyLF4u8Av#=`TdIO6+%Qn)x>8@>(noVk^?kQQ%?Ct<o1lKxBE
zb|c+wEIe)EVqc<o)ewl6_M)R^qwB?59O-s{z0Tzt@ffrkNPoo?cdafj6_Ax~nYznH
zY<|5><J+4~&~)uX+i~a{J#bv~C*n|VmAkG-8!IAOT939P=4!sU^-j6DuE0fQ)nuv3
zOX=Dn%S6=#hcC<5zb+~_mtt25-|gXejNXhN@GNZdV>b@$d4}hLeEmzcuDCzXXigFv
zn>6xxh<X|LHZ7F@0Zl9&m%(W_6azFcFqdK0E)xPcGL>;RCzpXnLNx*xkC(VcLSuhA
zKMJ4cf4%*D;920;x5Fr)RfhRsr_bLHzfVx{;}mNjI8(u!>3>r!W<tTgi8E)lgD#jd
z*4yb1sQBTxuV4Q0`QL&;vp=yapEQ$31V4SxuL|&y#!cVyyJawi05Y%ulIux(Cb)(Q
z)UXtMo(@_%7~S}$^c$82roEQ3X{=-fQ{G`k8lWj68L-D%n#cRr78a?>5K-tN2EJi0
zxh=jzduoNR__Xg-6<YmwI((T9pE9b(TI-;S_fi@7w9ae2gsNa#IG|C3EEP%1fKEaa
zD#3AMR!ZCX)~XCtsTIu7GY8C<0Z2j}e?W~&GY7Z;MMXw8AzZ(I{+`=?g;pP%&(zOR
zNX4|V)6qIWId&BnuknkEH`mUU^V-F=s?VI8w5Di`rM1<!hrvEgjVx(W$2zM|yO98E
zV$2>SNU8+{w1Y1Nffe&bGf0)mdco{<KuhJ|%^()A&WtD9Wd!{0^W&&gpK{41e_}^<
zU*ItrQ9>5mzm#^m)OD!`W7BtjNmh*+4_2^PBLF*aiOXl@mtQI#5^RJ<^Vdx<PDkaz
z1Wi<HdW&M6vmp4y$rdqXTTi~!$)U-QbxmiENf@Oi3t^B2CP_eJzox>H#@4SLJQ3t~
z!=ij(UX8Fw!ap>cY@sT`;r<pyf6+-S_(cku0zP~Z0N7KQa(WmRw!pQ`tEcD60;`K9
zEwEpwzFOV^kIfcHM!`c5&Z?yy-Et3Gqb%6uom`#)1C1CclTiwU4iO4IZBf&z@yX*D
zP(=W#TI+MlbW7v7XQAX%yEO2p$Qa_LLC6#I+U2rKBf_@xVn8iL02!VQf4Uc@nMzr>
zGL;!yIE$mnD<;4VvZvK{&N8ba>oHP@R~TbUy*a)lQx%~ix9OZ5t6XZyT;+8mVXMR4
z9(Qe61uCtXS!ugokYj2<s#8^z232Ll)6#Tzs%mC3f&}kD^#dd0s<CiSwMSZqRU}Q|
zXDAWIs2*Du-iM19oAfY>f1&?E)8>s*P6=>SVl;r~Msc(zLLND*L5K`p#YjA*Vl*qH
z<JjyRH;OF|OR%X<ib2Sii*0q12z{=Wj7$!Lg@?L_kB`F}n4#Z0Lu70K_cDMjjkX4V
z&|reB3sZnn32yR#rZ0z2W3O4G<7YHM5qNGbxM$t2Y{8=W$w>69e^eV(%+UdDobqHe
zx(4)L<aAfk@EN>6W*n#%c<p3yeZ)`rbVS6x104<?S4ir$D}i1=TmoO5hmnkXJ&uCP
zugd5~TSdeM@c!l4lX*?OS$;bL-ZC-kQ1Zu3K4>6?pN$aezDlb|1dIiD0U@P^iG-Qr
zUbC+{sF!0TS2o(Re_P|&m)jw;;8+AT_WEejkgEGp%5JNx>}9m;b66!-;L#%)Mw83_
z3j{R~!_5%Dh`qRL{$^YySdlP8RV+#%CB*1vrdhPJRkPLVlkKHlCVJJUv}pLbc-SLv
zG`fe}F86*nZ-!p~BHj$6{KdQ(M!UkBvGyZ)Gorf9n@!EWf1fuF=BfP{W)=61lOEua
zHO?z`DS8CA<G>XY%8jEh@T(#R-judui|Rh!E<?7<qYP`>@ZhZI+4zmu6tSXMdBFu%
zi`bEq5Iw4-%$uj8!0V7!jKZx@8)72n+yjwR$R(wAA_;L2bACR)Z1S$>Bo9V#m`7dX
zt_APXhq70mFdIk9H|zAAeE8(EKbFU7J@;~Kv6wj5w|EbWm1DxGnG7o|)`Ds2X5;@o
zSPa2FsFz1fLNFUTIpsL)l5P8a-~vZ5wKNRy=ndVA7=HGHC|Z}2OhPYzly(2E=^D+%
zI9mnr9CS0|bB>3#1_SBJAsatGA1#+T$xOMYl!n7!#8QN~6q;{k$79sb2bc0W9>7sr
z#iZ0|PK_^TlcR%$lKs@ucb3`JZE4CtmlTgDN3l)15Bax}ArR^$Q+6OrH>8COIe4Zq
z$ezYqO2&5yqxJI?#7hN#Sh-Hm^A<wAytYy;Z3>k3EYeZ!AumW9>@^)?(idb=U|CHG
zm`;4rWXxLw<21Mzed{5RCd)SqHZlic>1Cew4u8sd-RFXD=fr|#4wl;>HQ44OtlJ@z
zG^X!S?o_KeZ^$KaUM6!w)@>wl1#4AKRu`%0xhBS`=D_JZZo8U)1=TIwrkruh-h7~x
zdmW?`eApecPK{x<7?XL5d3Cccl)_}74fj&2YdxdE-<49m)R9s_?V(hx)ln*y6w{Pt
zZ4E-`YVM&`&`Y#R?MGV0(nPCz^sq{;)W=S0Nr*~>eh;@WzvtAw-0E828;8%h)*r8I
zMOoWia!?5F(jX;&a!E<m9N5IIQ0ltPZJkyq?&Im+(Ca8Oy3Fz2)`&2cof8)E7LLqi
z<aWEw`6yT$M$S;zbc!xV=I1%JTnfzzYy34bnp`T8by)FG!<lQKo+GDUnjEvKwj;OA
zY*^v04B+etKjUu8nS9(0uAb2H=_;9}SMcfXBy->T5rO%CRbuapyS7Qe@p4iarALxW
zr3aFyl075`M1_%KueXwXeSR_I-Z@1;l?5M`6okC5lOo>dk`>-dWj1)9+b;8dk;hK3
zAj|%~x+dyj^cJbk3(OZgm=8Wi@b?nGZM+q}&-uI<B&TSo#%U!gT0dLWu<6@Dkhtgj
zIE>1Hkfy7DGU&Y9)<Jf4CykDyZv%^Rje*Xx<6Au4zIORQyPXThNoM&CSZ)#?wivu8
z*mqV!uyQ3szm^+hRN8nIIe2O29(Rc{qdw^BerZF*RN8n?hCI7-Do<BNme&R-7FKzI
z!!{Ya=AJ+k%wEyS;?9uQ2YBV3gUFlf3ZlF11TnOKd)kcn#^pNG-z)>Q8mod1C1zb5
z<K8#d26nAbtC<jPXgxE-#qYi}cQCdoCI#41ug9Kin7_!e1mV(XjgAE0@U}}*VTFaa
zKnhbRzVFS5wvA_iE;_RKnX@WfZ-Xn*MH~m&gYC6}e-yR?oHJ1_rQHf{-+Jyf*gixl
z##`5a6&El0(6yy{q`Hc@5qQ^l9C<ID>nJ}EVh<l$-3xKodLA(iX*Wu!(3(|rPQtHS
z3N2xz*@(N?Y@6ie;+<TxDQ9;NST~0HHlV>llfm2zbl3WBF5dz;ak(G)F@Wn`@?HV{
zT3%LY3w}#+)FpXAdc29*^thcF7cww#Jes?IqtS^chc4~>c)?qz&{H%n!==H&7Y~4H
zt^^5MngwSF+oX)Q$=fda3aO{!^xNw6>n;%PL)_AiQcJFsPy))g{kK6%BfHOitH%3Y
zl(f%Z>Y3#$bG~9a8i{0!<W6Sw=Pg;g0-TUgQ<O8-Rd4Ioi<A~mN5^ZF3btn;)8;i-
z${p7xgT}w-*?nvB9Ve9eoE73<Q@!bT-teQd)xOh}3w{o72iqTh-zJ@-_JgmStfxy2
zebsl}vq?8+`VQAFDCCpSLD0K0fX!c{?77_wr2jud>{~6D0ggfy0x>q1rcpv>1fY?I
zO_vZ-LL7g5!11r(IB4Ktg<wXB!`EMje;qlO20s`4XwH?wkHy~(zv1+*gu@vEX9ye>
z;`DB{z!8Nb3g<|`Z;l^{zil4-%<B!VcU<cLLFbPO-=}|n{rbi;!JK&$eugjl-^bV6
zf8Q$kQZX?80Nyes;qafo-ogVdfEnu#PBMJ>_wRp)Ki+<$KwH^3fwJMo0Ccj^2;9QK
z7|&FAlw_lk4?x*y<X0nKfd6Y7#uE=r<MR6F?>`S@<l8R``PMQc1jQgZkp?os84v4`
z#|W_-L5M$ndcMkoGfc<<0g4?2ey$y&sUW;#&M68-eZ<vumf;LM&H~32%DBubs&IWk
zJPLpD843|quBCbqW<NWqGRVwt_>EfysG=HWps#t{CRL4FsnfRd&vov!Wm?%+a@RxL
z<Bs6>iQ>r7HB%3XgP^EN<mWv7KjTUnrmcG=S3Sf%uKL-x?%A!}NnjxSE4k|-?s3=8
zPR?T4Be3h#7xyDH`nbB7j}xU~0PbM8VupXqGLskf*4!Qn#-V4%VPwad=uvQKWPbI@
zx9T3o7*d?D4dO27q3of{duJ9GN#&G-5P)Fbb)F;cZjjF<?HQ<St&JAwf+5>N!W4ET
z1+Mr!YSBp?({Pv`32Q+VMKh7eS;UkenXr`D^Jj=`!UpB_Cs4-)zRqx?;%22T3UhyN
zoJ01Q;)cdK@0pQiJJ8YZt-zNR3J|OG)!%Om4L9qus=pv4Q>yN2u+<=j#R~4XLZg|}
zmO>UkhtI?kliiXP&v=3qSSbk;N8oYWtX@Okiy~!8(UE^_ItAbYA=WUU8;2RqAV4@j
zAXv`?8asi$DcI!V$O^ei4?_7^<pzJJFfFS4FE0Ki8CKeOf?<WD=dvsj9T>q9gR@&a
zK$HmFlyXUQiONd~5k9g69;a1?OkWQYQN>$TFLfK8N-~d720BwcPy=__(R0=4r+6@i
zc~AonkmY?YS^8}}ILD+MBoaKD;dg5s8a2iRPoa^Vewpo=_JjOpvQr=H89;xrBa6_M
z5d6|gCEHPt;d|LpkL4A1)I)rw9T8fev?EIAkFg`d!;^N@LwnYadT4ugL^yfSj*MeV
z+H+n?thin_X^_dkRsP64uS9fvH{VsSS{hz#XHs$|fbbdmHj#e0BxM!XKK)az(1K}c
z2`<{<h(rTV<VX13$QCc!Gp~P$ylN@%nf8bj9#dMaV3NBY)=WphGSFh**Jh1bNtnZU
zq|&#A%>5c8HI8^n(W`Z)c&Hvr@y$`|O<(1N@;q7x&rJz8`0b}O%S)yQf~F24yOVx4
z_$+R#5TD3@A{pIC=a_&=tth0xrMNa9kz1Sjo5MO?L7mhHZN0?T8oz(tZ6*G+<y&e=
z5fpjZNI1o|Y;tX(m^H^@M5p&qIEU|4;;xj;xh<TWY%ZWD?E<H<$QbdzwAzoAKG^L_
z-T`yLl6yj$s6AsrdWZf~3@g`5+Zlc(1+&gjD5BuT{odF<<xPxuL&YW3`6ETS+IrEn
z1A*J6-LYvMo*PSTO8kG^3)xjE6(p$-SmVY<M55`!S97#+wFb0fOIB~yv{O2)VM@aQ
z=~`6OMxL|LNN{LM=w<NZ6lAj>Mz=a&!2?M_@hM^+K{e$@g}x(QiAL|mz*W?=uy;SV
z*>Pv0m$LWOJWU=EbtL-b@*lVoFJtafMq|!h|7*28aN$o0<8^<)?RqQ?MZVF-dZrYi
z`W7pVn6v&$(LaOETG)yH4f@v2j!s7VZPx_~!;qvacls4mZN+@qU;tZ#R<$)+qU`oC
z+i7u*aE;-jkzu6ev+RiLT6c5lwk~|x1)?o&+q&}=b%u{SDrKRCZ%Q=iw7pBBMJ_Tw
zHH)9C`G?q)J1T!8Ge%IHz19}lFUJcTTH5J44A#G`v#5=R`&)Ikn|+tg_A7UK1G_p)
zpgpIvCigo76!oJtRNSbcfK{!#Lwqk#P=y(i^%UnnrJzb+z9t?G>fP$c)lM*uLj+pn
zB;<e}Ck4lv`xWV^rtp`|Z`-*>nZwaioJ`X`2s{leJ?4M?L$?G5`Rwh5F!4n5MrYB@
zzRg*Lk_za`7qmd8Vo+FpDn~J9hqAdFDzux%mFuPd6u${AM=IG;e@$^4Of>zl>WO1T
z#lv}$UzPkdyEJr|GSz6N8Ex_9px?}D+^W<tajSVgNg2~<mFKlk01aJfO`WSZr538!
z(5fLkE~kHRxr(ddBi&R<yyvlV(W-QzC`uf<mheo%pbWDh@7(EXOB|DL<)$>>Yz>s|
zRH<&0j#6MeC`V63h@jH)BVm8$yfD+^ZeXPJWpwsj*X$@)s=>-+8Q(T-r6f~UP>k8p
zCKugiig5BWxxjbL!wEXS<XS3+FA4sb)gf23P1k=6pSmU!Npp)zck%w5@&yUDDDtK7
zS)u6N9)Pxj`JhfccTzR-Ta)Rw8-Spk5-8MSthJI44G&KGl1WiRSIjuo_9VI-$iZ=u
zN4b6ic~4x)-gaCs$IRI{itAq2D;0BGQz*flI65&S1Es3ppKT5I%;|xu?-;xkh9KaB
z%x`}PUP|L2R|Bnf2_ievQy7A*kQ74j37X5~jLU9U37nR+v7Uy6S$ZYZC}-OKK!T+D
zvy|4Y1-muUq#H0X9jn!Sf}i^aU|LPA9SPIf$Dq}YGb`O{Dr`-Asj6Iv%XL7+WX*05
zVn5_#qUM#Pl-#THm_kCZoE}Q7g*<O7sYriw^-iUWV^gQ4T&?Qn>19tAcuUDquV57@
zwy`YJ%5t^xN;0gsQ<j9gb{baJ)^&dSK2(ulm!2>1ItiFjN=v=DsyoIpYh~w4w{8Ai
zfyf$@ugGqBLc_&?<GxOCIKh!>pfp$%L@yZbVz=oJG^Z|cuT9mRNWG-JzBE#7IdXs3
z_4<6Vm@>Ea{L+6*|7g>zVKS|q9HjA_2YyEZr-L2*CGYc++Dhhwy%CBkccMt@vUE{V
zy9?i~r6n2aHhQk>8yM!AzEal>u|r~>KAk__>wnnF`)Y?adj8t0O}0Wj!FgM&9eOFu
zZ73%jdsWXfZ@Rj>9oejnB^kxeo9lnFmttRSyJhJl57}4arQ3{lr`(nrGuUZwZm(qM
zSHJp%cwSJuSWr{!#bfHKAv$Kr)AQRh7uHX4%X;dc6MUC5jZ6Ao<2g}dk4@ydc2KGA
z>rT_-yygODbD|Y7wB9_<O>#W-2bj>p9r@Uh&Z_0BaBEt3mOo0)i{jgjJT`w7PT3)&
z+x;@I<*Q98(X0L&-CVwoGGCfo@fdf(QL>9D&jhKB84+`pqiHIy#sA?fRrCoV2z8Nh
z*A+iD@~<@Is<DieMx}wtc~6XO)-*cOz-#?#Pko!w7)|m;QKSpqQ>tzC2});muSIyx
zWU((;s#)!7Te6%vl8$xt&mw<gUcA0)syvgLD{F1tPV<Eud+<$#SG-=EXE&acv~q5s
zalL^|PpE|twXALLVrkd5uiLyia93`tyy9fm$atnsXjpV`iDaYiz3o(%hiX$&3^$(M
zi+E41dJ0)$gI>-_(yl%Bo`BIzNljl95TU8S-QzFB<y$Bs9EmK+)T@6kpJTXJ%2Vi*
z`#j|x05bczTO{{iy~>xO2o;mG0#ke@tQu%uQNDX<0L_HIl?0rnH4H+Y+}95sxO~ZI
zPy%h466zlkyzt>J#6ZRL{fqPhBc=IC!J}FE8!0F^F3Ic%U#Stwba}?gdnA)vi$VfM
zE!Z~Q>3bSNXr`6<F7JP7r1wY&hMDoFmv6{q^>O!IzVc!Yfy80l4lSD&g=hjQq-T?}
zxN<#}V)$%njRJv|)-v*K#dLc}u>}xl5bHFrDMuLZ-<^!75ZvxwQ=}R&ze6p!V!C-l
z>Gf|-6AU}L(ocRyS9(Z6fV*$uwrx|sBtEKb>qEY?I9Dai87ybk!&I{KJ@A+WH817-
z=!@*ZJ@zEL^4@p8^-OpiTSL2-zVik39?(RhPGIo!{}-3O9WzD~f+jXl5|4iJ19zW`
zta?E4>T|j3tef9%Lzl6NGLiyXzx8Z6Gq=6fOujKU{vT@{mIapqjzSfez~DL+m*?#$
zCznlLLYIH2Gnn)5AHTdvW>^a2Y1|}*Hr88u+aT5SFSV|>eEUi(^su@@rG>F!vJM5U
zqpZ*nw}v<*a?%jTNWam~xEN(*4GJ)-P>5?p;@{93@hz{{==@t<=ndRfW(c`|o+(N$
z%Qdw}ZrnMFF@84OB@Ryv5?NFox!gT6yQNrVqe6cNo?ov~{@|j9!`bK)lZVVzB7b4I
zK+<iJhX^)`B)+~5zurGy{&`u58k|69ra%6Ei46!wSQjRxL;#%r{d4;M^6N;oXnuGn
z(@lz%)*3TctQE%i$<nN@spUS|k?!7jaY*FqpJZaQFW)*ta5|dZI3xTVpLNsmkWFXc
zG4g+QVcvWecr*|B#&2qq*Yg)EqeYPI<Q}{NydLIUVGuzd;anLn5Eq<R$G#lr!GTb6
zp|4_O<NVGDv}$esxhyicM6<>B{H;k2Z8PTNezDi<Va_!=;k-K_J+cy^!b09^m7l2Z
zHAiEbo6ybPX1o_VQ~^cup+lW33oIg!BkzAV^3*UZdAqG&Wpka_ng!jDF{<xd6%m&a
z&Iwn{g<)wOjg?WvtQ(l#u<%W6bDs180}(VlIIxh-S*cS(^}@4hc$ZYbC`~3Y$|a7$
zP`=PFI91r?mNd^MdhSXAr56NW;{w9&suPULan<+Crs8bjF_|L;5(9c9^2%#Ya@&9B
zg1|Gt$i<F_ZgFCy63VNAP|QP!Q1pPy8MG>*sftWt7pt%qUgWGmZ7N^Lu4CEgvvIE&
zz0F2ndh`ZFi4y2H)sQNVGc^R`h0SVTeF_M>DaRQ(JT5uP%!IQrQc6*W6OxEV@f|U=
zttd+%rhNrY(o-TWl-6<eopo{YV_1I@$&KomB-3jLTG-sG#OxRoVMBKmtDLIIt6qQ`
z+;&&$rAqej)>6Z)wbGhaa_QHZV5W>Ez=6|{eVB7z<qaLv3M|@jB?+k10M~F^&jEzf
z3u;NHR-P(6-p6;5A66;cC&|~wGqO}W4qiJ6)-lCLlhoNBn)C;-DPMoCdPsk($RP^Y
zhvPbuue~zQCB$)Zo-FISZdGW>TWr-uS#hyf&(@{QX!+W%=ApHvcvK#_A$?b*ze3-v
z<M)ETJvE7CUQ@dsW(}SG>Z<V6a^+g)_zbk@{3_5U#&Ev8YQlATV1|owSD2m)yHaLY
zNN}6`%>bFf)^-ec)l4UVOm%<HA*&H+6dvFK&;<jOdaIGr7L-xxB!k<6GK4!akJ4*4
z#z+n}s^u}<;+wG+%F&p(b#T=7H39Xdo1HT$Oi{RK*}T><Ljl(Dmdc+xvNQEtBfDIW
zN4Ekc!wvT<nV*|Ur~PWl0n-^1fJUd|tCyVX`EMf{46N&08UZ*f4EukT#z>t0=Rz^@
zS?zotqO;bjQ*VXQ8rW;5Z#S(GJr(lGS8FV5^_9j<O066c6woL|qe6-vuZ_+H*5d2l
zzKvDFI2d`D<?u_p=!>zRuEy{AHrgN@U@gufr8kzEWSL*vuF%n45M_<<xYz}OK<O*J
z_qRcWjAf{ae<W9@voC+&cxxocy{!?X!WAvH0^?@I(qy->>0$}R(2bNI*cJfige6=Y
zC~j;Elns?}{^iElAs&Jk-s?MgP*Mxf3mN>M8&60k<!IHK^eW9g8&6bcOONA-y^W_w
zbQn-r^3jpM4ILI~&wD?pLq+}-?&LrzDZrRD>GYbyQkGo0tOb7=Td7Tl?)Qb!FzvaV
z3k+-B1xncfQ}>ybY11Xe^xq2EOp-SSXINcqLm5FS>!Q4mR{4Hhqn5BH=#lE4#SxkU
zm8rvsJ|1g-oDXjWEQoFzmK56I8VgMn^NsqdmrGwsR3U%LNdX1Lxg$<GAMuDI7I+?b
zfPIw>^FIittg?UmSSTu+_ukwn9ZNNibA-rN)x99zP8}7aDe_wPEW>06nG@X<bXLo}
zb0090?+<p0WuIloRwdSevfzwDrE3MpNPQG#8RcZZV;uv-qfo*jd!W4M7i<*6hzB*M
zq!Xlo1s@wBA|8kmhOF>#7)w5LR+3LYMqY`S+ThoFp_G4pvdlD<CkpkQGSe~$41VnW
zdAsh%=8*bcj(6g|gFbGI5h-L?8YAXyPtdCii@Ccui-mn6NT5QY2?ECX_5S?YQN}qA
ziXz%c2(+Q{hCKQPC?X|e8lRjt3{-Ey;=Fg=lPWubs{LN3TAY@t`Q}V52~!!IZKpqz
z=RG6iYkGfm*hLA;(T}_<Ib3A8UB_(|abyxA_^^;pn3vtdrG<NDT-X4G(E#?DWn#hn
zIb~y7X3<B!{TVY7jP_eBEvEg(J<EQ%7#*nIC`O0g&$Zuk*~)&h{Z_D^GyhBj22VZg
zLPf6a;yhx!#Q-FV5d6Z#T)W#d80}|VL{Wt+=Lmn5wx=H_>?t65IKcj<&++Y)EsBSH
zj?L-w$}g$8a*C<}byITdyI{}12Bvg8)p_6+m;x)j!yj=Ath>gMwAUdb)+-&|3DAuj
zVU$zG1Bm$oe$bUL@KOMI>5sV*7E9200j+5HupmFTOP34ZVe;KNCvTgL?B2K?wx=H~
zZ)|^OofJ9Y^x!3<c!~v3lSysP=?>@mk#T>?VR<EBTN||pp9ke%-F8U1RB$Q1-ReQ}
z_H~5S3zQp2^1%yh^tkfl3fOf@ZIz&6W}GFh31^o@R%~?R)CM#Kf)F;ZL)5D*^<L9>
z-z)dnx(bnXU}o1mF5{WW@~FL@_vR7Z$sK=xsR#d6TiqGrn*Q>8O9~NUyxl&#>a1h;
zB-mK07-6J`<+R9hD(*YosC7;&)`UJL?1MHX|5URsW`95D6LHT_@mi(H*KX=6RlQjD
zj8nOk2T@Kp@N?)?Zo?BaRmQN;+uTn2OIYm+L-i5Yr8t~1mow4MQvQ#$A~+G8dLS~b
zaFi>1`vbBHrMsN4+{n80dE$*bWH#MKoAKKv5ry-FjqOTKY8%R^@}`90KL)ze(w6~_
zLKT<M@GuktFgKS0ngu79*JnbHf3Q*8{>$Z`ohb$zJ-PT8r}5vX%kI~1HXp5KHz~(n
zg8}vR=Pte=VD0>*5cLG4et%QncmIxB>sjM0J7o=61=<+(J8tMCZw1f*0;qA0xdRlP
zW)8|ZtF8AlpmIqTU`Rk<xirg6r$*KTFl%$91Nl*#5!bK`bIQ9mFLu@Ce-$(I8s$=r
z%juY{q4@l1|AOS@4g2yY+U-};5G64r;w~1aGsv?q_p5V~Cu<9GE4vT{Xr$3DK9*hE
zx52o2vg59Frn`NfWz8n6&hyzT$wT4x*-V<ijT&M~2vn~*1s0r&-b1x6%gkBubI>5l
zyvOuE?mkq!6Z%m6jT{1sf5S;euxxXfuB30DuioC0tXsimZ-uBSGh|b3eXerfh43}d
z2!dphUpXwnvxgK~0svM3I0E3RV{i_(Qy*w4WT}7{u7_?|g>G7B@@Yw1y!ScXWtB2g
zR|MqQW?y#Y1eVH}bV0G)*IF4}9noZ429{9hS73`}v9R9qFt04Zf3fyb#w=Oyt1T2y
zIn8^UdpD9$&Y!KJsn`o8sk<aj0iM%Bptw<Bc=FmJ&jlvrqLqAMG7@y|%ip7QpQY}T
zr&Zk-t0sP@?te_3XHVM0(`9~=_JxD8xrx>)Pev_2$>2`%)+h{++Ip|B_-%E1a2m$L
zug7eDms-#&YjQbre+Hg2ADz>a)UUhDM-Lj!=!wWiQe8OrRm-Hd*|g60qhRH7X~v#=
zpZ9sRi87-$fyUj>IJuC6LE|+q*BI+p(_yJNf7YiMv~zT)wP+)CTVZ{-SGO}>0@?-<
zTXw-kPfr3!vU>B)jrKVgASpE*e3vy?GKWi73fulD%$sFKe`nixnETkVP>2#?cv@~;
z=|rs6aXl3}0g%~x6OHg%#LqYPt4t_NdYW*!%;7vYMJr}-Wc_?9l|L-Lf3)gLY)Wd{
zk6`<ERCw}Eqp1H7iqs^TOOBDw{%|J6Z#8nGVeoex+u0=48sv84Rop^%7pBo<(E~OK
z6^3bL(s^Z4e@rbE-Iw@kDY&g9CEaAUJW|`jaLP<9N($6;8NNHNd|cl7wIn&F0Twco
z+4zgK)R0%VL0+x3l~T<TzRLCo%9pjD+OzsrQ;6rRF%snb?>&OtXln>^Z(Cl++XT5<
zH<IAIz9qp~a*G6KTRIXv@4JKW_IbB>rF>2tmy$EnDcMSs^U6$<w<teJ+X9~2ufv5x
z`)4WipJ+)0l1KPROB!b>y8i=xG^>}v`!N#&F*KLa`70H-Gi*ZV1q-o@px8hD1Fi9t
z`IiBXLKT;AzXuhU64C)Df8|=qZtXS?-sdU&0$0^Q4F)a_kmDo;x+{=f(N*!ct1jC2
zUsMArQL^mg+y-rdf8m^Xk|^@y$3d3MZ;M>Mf4aXvefMTy@gi8|_45AfqT&r1|0)?T
zI=gtKgfnV+|9$!CtCZTs;~o!?(wKPo_<xAM_V&f75T8)-2^Fs>e|7ufL8W)-nYH}t
zx0``@-RLq#d*9>J=}wx-xOmmBGF?}Ihe_k#et!SqN#Zi3SY7%z{fuX??@xa`jf*l;
z2{3l~<@cv_1B?<T=*9RTY*_yM-}39zkK?FZWkE%iN$0HXR0cAt$}Y~W!lYY8We^tx
zo0L|@-?&t(;^FX`e@^GOFHR;!<tr|o>2&stS5|tj-ibrmt)3YiFFoUv;;+nE-it;`
zimWS8G9m#M2R`b@(+Z?Z2hC@2o-`J`;h1zD@(TJftx$4ZoVLQMmBv)Ej`vzuQm^SW
z)>`!f-Rez&U)F*vb6Z=U0z5aUGdj;}wh2&8PiY+a*6BW8f3OBu6$b5jKL2AK6gzOc
zfG&2)vb6a-T@iv52<u2y7DcFHIf6+NJP&nGcdCGO61tKWm6mTwgsK`2DH7B^qw+e?
z(H*Pn%m+$l5Z=ZSdP$plx;NVU=^miG<O;wvQ4K?yh|*0JDvgLkR+mf=BC{!61r+Hu
zjZa#z4PNA0f6atbsM}Qw2j>49>h_sEdvEZ6OiP`dYHCj#VpIyi?V0ssiyoA|D7Qvg
zfa!L-WiX-0*=1G~-!(CJ=GaJiO(pPxQy;*@OX0P>2opf{`D4f5yA!&^p#x@s8H58F
zC{(HGJ;@+v`G`evCsbscvs7@_H5J{CaqPoJRCD=He=ShroP{8QJcx4|<;|<!xkv*I
zi{n!(YbQcjztzeBBduq7IQodhp2^9xZlBQ>TX_TShk1pY<3oMzouk|tuF>e3G`XbN
z3M-Rt+i8pD=umKm>ce<xqpsJ=w-g~NqdeECI!@|9+;Q#zgn)Dad){nLk|XF_GZ8fN
za@z1ce_b@=O$=@rfVuHK%L7iLkOZ-M^?CI27=Uxy$?F*2&CEMEyR{59p`GWk5d`%;
zazILUp|G~aQ)KO$lE}sraI7*|LBuKkO-8OtpuoJgh4!NWG|(|4ZCyoiRrQk15}FGg
ztNU;0x^OReE#ztsYK#Hm2Q@`pzEIv@wAYN4f5P7FC?T}Z9f0+~MA2&Mr5s6GyCE^8
z`>!SaN=uR(PibllLA#fjHBXnP@o><K(i1w+hzn@RRd%0C8O)7<d=@^DcEPMdtH61;
z0EB0=%|#9XuNt$mtCxeq<y42#hraHCom+JQ@FNLKeY+cz;MS_QIh1t}+I~}ce5tM?
zf4N*A^}0raYeBw5u;W^eqA0t604+PIO;{~5M@aAbhf&4O+<`gi!$cgZNIGFa9Z=Db
z0;S%R#kuuBduaKYeN6xX`eIE;bF>P4QX3MMF+F8Shp~psR6__xTFK<4%K<~JX0h&7
zirH1-vo9?jPGc$A6226yGEK=_75K8+e>$G@W0ip*nbDB(Mu4l4gcOozGi;_tOLhJS
zwA4j`3FRaQc^xBnO%H+6mM$mn(oe;SN%5rdA)OtXKciPpfo`l47LNLQ%tBp3UUK^>
zsP!tG{s3t`ZaZ{Y3(G`K<F6azY#)mB#W8`1p2hMH#`&q1R1nU8b5cBgV{m3cw{<+R
zGqG*kwr$&Xo?v3lL{Dtnwryu(+nG3D-ur&{R-GSRySw`Isp_u1cb~P_YWuM#9+n9J
zzF8RZGX*9d-v_TlWt-8QO>AnXT;UK=E`bD5yD@kk8L6_%Fm|fsl#HyeZQQIKYYR_W
zMM!dK6flXT)-;mFEw8Pw$#T@2aG1X9=JYx^MQ?I6dTY^Lni#LKrd`?i{j(TA{iWZ<
z+Mu7C$_r<2(^F;L($W$XSmyQ;e~t&ZmUqu4_iaQgBK3w;ZID;hl;c&zxLy@yc$03Z
zELOU@jDB6e>$`Ka&oDZmCmopcQP;0ouhPt_1WIRK#JjQr=>I9YB8BChh1(K(IQ{fp
zShA%i;z=1$4*cr?i)uHlPs|b+^sI|N>*QSZHck~v;o?@Z7)>VOZ@^)E=L-$6?%asI
zhMAttmyjWibuF}TW6u23Vq!$N6lIs~UZK;kt_NMzYk-mM>~)i!TE%kJ=mH(Y^0s^q
zqfdOzbfrgp*Wxq>7DE0|)csJF81&DC-3Jf*%|84Nf;8rDmw73e|7i<_&dROY&x$JO
z9g6zgCT36G#7Ywl+q&`gWD5<za>uF%f08OrFi{fHeuShjnLS{{FXt{>f;_10q|3m)
z(f}!wuN2eFS(e<^_-t8p+PgR_HE<-xY}5K!VWf;gg(%Jmf199TgH&bWyznpHijn(9
zAz24H#qNIN--<I3xEt2`D!-@Q83;V+0^@fXB2Z!5$IBcTBbI9O7|UHiE@-hQLZ-Ds
z;XBT8SGq3GvVj?F_bkY{57x?t`YY2wKhNzJV$lps{lI}vo4ORXKmnvztv~yp68i=v
z>NNI&zi#s@C7Ir)ppHP=fiVNoQwulx`5#%0v>UXP*lSu=w76n;6|{|>`UG2Njm9gE
za2@@2;rXPCc*G$2iB)HScBF?jqM5O@?BcX{Wh%=LRxH8AUH>%eB6j4BNt+`KTh7RB
z3Hu^W*E7R28xy7M!)>_5PFI^)+&=D=#rnn3^#@!omkitquDqB}NuH;~p#~Grt;$P5
z!PDlwnDc)+WAyPcwdW;L@BU@nrq&)ePsrNgU<cWfsXr9`*(ekN<0}HDH8_t%#%<Gd
znm!M{;BLtdBM3If^f^tcPC)~2*X6}J>O_s@kJn~7JDDvKJ%-q&*RGO(B&>XI_mOv`
zoUtOoHX=AFnXd$qH<G8TXS?p6-_q3Ky4q*fn7J)I&euKkFT-BIQ;;g{ZPtKb0bei0
zID#pG90CDXwT?mD0JemGU6t7FqH=Z{E7{m`ox6G3BM()VjM-#1op5hbJEuqK?GNs=
zlElv-hWBy-@3G1KhKT26VE+wdn6p8@gB=cFP=OF7(8*UM6Hv`15_<1gBbY;mFB)aE
zu<fQR*wvZ+$ybFGnkz2E{#{g&WI#j(YM_Xd^&7Fc=QFiCGeBE$g881tgrbVQctCCp
znoMCC3iMRHPj1~UE2naC29i03lA~^^)osR7`-n+ojA^YE6GHP?sXi7^_Ep<NUKvAu
zPm<z2SJDCl01U3rQ0ctQA4RbAaT~PokhDyfIpQW;ba-4Z;0S5zt(+VJvh5qf{ni^h
z76;~F-A6EB<Qt~!&9gjKH9~o4B^gZ;8;5F_RC6Zg8N0&IU5H~)e$dXom_~q4O(`iX
z6G?j>7k(jm6rD77kD9yHw=P5(4wocKop0!X6D!hM0Qg*M464a!w;$qV*{vh?pIZVl
zm~BXji>dMZsS1#o6PF~gFlhR@#4PD>L;F`lhM6#%Qd)*UWMo68+<rjVsKW_)t!aM3
zigBdr+LD`dNICM}+KH$BJK(8pv1l$3aM50x$2Q}<8dlhAYfbm@x<b8!cc5c|A@Q&y
zsW=af0+6LS@sdA5b&5OY9!+>Rm(tO@QH&(%&b{fjyn@uN)>T4BPxPg=33l}K;J?`a
z=CXpIurfxW$@e|#5pgG-!!z<#rS79d{HbsaK`8I=a||D|%b2&9y<gOo?IaW^8KO$L
zAV5UCttkE%{_;kobegQo)^|aV+0z1is=-ZQ4FFjj)j@XIS~Z~Q8XH|8_hHQ_Z;*k%
zuYW8!j6otO{AHl`gW!OxyUSz1!w$Bf7!~1D&X}LH)LQQ`{?i~rpyIe&R`njXyc(tW
zeMItBEroiMXf)AXtPuPHOVX&<&7r6cZ9B=4jT;L67B7}|hPUpvx?wMW!g7DNEPBN(
z7eEK1Zke#DuRd<DDawzFq-xr&-Y>H0e!Qn4M>=EGIy#am;yhBOK@%C1(@JlPO?o9f
z_<JMs6w0(ZNlK<n04Wx!3Ke!sh=m7O=Un{Hi}+bwK>qlgb;Z)<A34@<xCPl&9kyh<
zqJ;$$I21&0`EZ660`la02IjZ~jPW)?7(l8i$VFB+2rHTTsr+T<+Fv0cn?0%_!2q<d
zmHT3wuLW^-V)OJI>G9W^bq{E2g1#DyDqZp$#Zfd}#^vx>WHpC`7h4VZd1n1TN$@uZ
zf`WT19?J^u4Z`C69EV{=gVY_mm@`b|;9iSGCH%AWb0s<mIr12?lcy)0q+X@FGQa@#
zA*-y0?sn&x<RSWvj8yby097pAVIswPn~+)*iM*M*ALEk1v?6_%N0aBiOgD%ut00eW
z&u2a9@MG*{tMR3<aRDE{5R#newVp)7=V0`f-mJ0?9QN4b4-B_nuOK)^-8m@L7vl+X
zi6bPVP;FRiq;WXidM1WNJG|cXJiw`gp7+)n18tI831-CZSZJ_0rB@Lz_mJONahBx}
zL^}SXWlAB9!CWcqwA8x}cQg<?IE@BIp$B&o^hcST`4GR#@A}HCQ#i3Szm->V75oRC
z*M@#@I(0=RwgtIl#UbDLb4K^a0}7WW4690QnwoU4yDs<K3i3LVUMgY8P(XjVfH1yn
zn5>F_A*#nrp}Kc9^VkaY5TftnN;Q;D7UgTll2%s2SpoXVF09^`h5dBZ>h8)=p87|P
z;zI&Kf9iF$-POo%^j(|a8%gdtK%{L!P4aYqN>^#TOJ8Mu{1sMpnUHBev)*|&Z9*m^
z2@>di_m{(HShv!UpEfT$BcR}bwaG<+aREQzIfMvap%ZjlVym2@{?6Qd@JrQ>vGPES
zk~_Gj`2}sWPi6HP4rl_oT>o7-N0q^6sk<05B+%|swRPL&^QaB-Hn{w^fx2_QPU4Sl
z$zF=%AK!<rix1SP#j7k(?V|qF4@*KZQxS6X_J1nfzS@&#s#X}A{8FSZ4z(<z%1I)k
z(;@9=>d!<R@_WoGR`>!Bi9z!Qfhu3c&sVO&|8CZ=4Nw6E<COwj>_#^37b+7w`UrlR
zPu17VHJ`ZZ_<w=ISgdFe%xp~TK;m>vG$2Q$2n`S~ScC#78!iF|Ovn(yZ?Op$*?|Dg
z*^{HU7=()qf&%MyxzPVJQW1=k85ofvLfKLsDPj!)R5@ovZK00)&k$w}1X4t1CJv7O
z;~D1w-d;ichiBZ@5lt5LBinFp{sXK-M$uDKP*Zabl?a)djw<Nzs;M%0AAWepi_W({
z6wp)R^FA2Hr2rO9mznM{wfLsy&MJLMlr)>viII{DEphUj<uEid?N0jlvbQ{pt$lH&
zQ=YbKKP`}&z7ruWibQN393wfx3$+IFkK0l7jxxhGAoHfmpMtbIc3MV^N|<9Cqi)Ad
zRHU|rInhg@uo6#=ZCoNH8GY2xF-plGgN{CkEj9wez@||F%4~981>J&%g!?yQi*}I|
z*34PZpuIACi!r-OGo7}u%*bT!F41$CH07ueS)u{5as!SK)VLYCnfri|Q2QVmo+yKP
zAI=scpqv;(3f5OnM64d628AQTn%i&rB#blIjUy4dSUO00ABAZ=I4{zjBk`s_ONLcv
zwX_}sYDQk8flajkbQlRLh@;M13w^rG?zb`#wXCXKSQs(PvHZY}&}$`G5Il9X(5RkG
z;U6e#3|Z0;c-R+Xr6t88W#j;wSQ4r6VbNkwK<w?$)1T=ey>x`!;uzmDqH)v*<}oj2
zQrbCkPox2NWiZL=B(=1uV%;|5!p>Uk6cZv*Q2V%`!oT#uwa8KIOnLawGso6>kmfnz
zB_L`p2(A#HQ8fFAKcB$?N`ZOA{R{){7$%9ZMnZjN4SeVkqlSArQ8et)5@H1n8rcvk
z0Ksq=b=;~Mq#H!DNt6MRha~k|FOi2SdlWM?S2{BM27Ri3W=1i_p5(R-OjfBb&KsRL
zl^bs#|MsM#^LjxU6`}<{?<sw?BxV#m63$J$*s&Us5PNB&M!pFYb7FZmQ9bsdlsnHB
z-MI<+abNCEk6+icH_K%6GC~V%v_wZy0K6K+Hgt!h!jiFC2hlM|Dd})Y?H*EOdU$x~
zHN)!vDl2t|Dkvp~&hK0|Pn0`9S!jscG)3}^^%2jdq4JNgOAZU$aXcr+(A&K(#l7dQ
z@icODVSxmpKNGC?VNCyK{9&pv<ikO%IBoDN=r^e7lWs2RlUBNHLwpI_yEyAS0#qb^
z-V{(FR@{nnPba8#cp2bf#_P7XanX0UPxRLCNcKr1@+&8#(~fIL>N#S@Ll6Ko_$mxE
zhtO_&K{3YiPQdztmWQQ~kPuVw>Y`E2D&6p)NUVQ59CWg^DaFK0S5Y(f$$!{9b4C)C
zw>@X_^i8T#IvJl9#eDO+k!uEW0U8T@N2`k9UujxW>aUL4k~t1ET7@0xJ4hMY1_tmy
zX=^sq7`xSVTJFVT&buQ5T6koA%BJ(Pm$cbjqmGiYRId=rXa7LB`MAu3;>$--Nw`s<
zlS?Euy-U{Rf1C08GmQdg6&y?Zr(AZu@+?C!lhuUE(~N`~8m%-~`+R8u9_aB$@iUJ<
zVz)90?t6Y*pZl=ANGd<PKE74Bu2nHaBWVH)KBDhbZy#<aSikpg%x@KOCMCD)6tAFR
zMzoesGH?o3+mJ@9<CCw1y$|mRGID+>D9c7UXv{w;agu7E=krrt?60dGrR``hyEe1e
z@H_+AmBU?oi3|+90}3?&!Hdg+69JNa$|1)aGwxD}&Qi{{8=pt!n6X|~6NtZg2bt;#
zObEHiWH|peA>PDG0=`_-2n&Qpt^={~p6`X-5fdr2Xejm>Y<yHO*}Fe^%b;>n&5f0Y
z3$d!fZ^`2fJw&98@!JwMP}=5EUC3QABM!_4RhuO1dyyc_c202sRVqaY5x;ivHi`@9
z`uUHhtvU_uC&#|V{h@G?-`pGTmTkhV4y(3Q8jtwN<$Yzg&Z;Dx-~z$ybSJ~NOuRx|
z#}W=KPM9aP?XPP3-x9|#glx1vn{e5bV-h^qS%>VISgt2M7BiBQEA9UImk?+weId3Z
zT9e~*L{3Xat`A)Sr0zJzLaFK~-==?;^V6@0Y&`-nw>7wqo`>~E97f7}&zemR3DT9K
zd!TJ}iavbsZ8~)E&&2VMOn+NHTJx7oD!7-%_8AGSwWY+P;PocXu!HHSeN(eH<}|X&
zoQYQ)GQPiVM1Hdzjcp54TOD+>pTGvWqJq<cBjqrN%ZV@pDwl%Q$GA)WAToFzBG|k?
zbx2%h+{#kI&#a<iFItNE<A~q9zT%}uB*}xg6a>c$4;^SusC3xsEJ8JFV}PRCJ-fH3
z!JMm`{dG_mvuTU6(%`i)5Fef_8>)HINIsk+LY;zx+XbdPDALkhE__DLoifPSrZ4?e
z<>?ED(o&EDKxI@GHVSy<bx_Hs%wWAt((_n)uK%6nv#rH=aM;xe=~Lu=yVtAO|AZNy
z;B~43Rk{J~;J~J={r3lcn(6Nd;m+FNZ<FJ=r9kStt7GovI@2!0HQ&RAcUudz{goex
z4?#N@9F<{A3!$|Lwk~+7zCV_dxeqxCVC=Evtw!d40a7H<(nYDK@{&%TGWY~84!_K9
z{65bLYI%$bm-<O1cq(%`Y!2n5w#0E0<Uuxz89W(bEcp_M>7mH>^p6y2VbGLO3qAM@
z)38I0f3SY_Q^8BjXXIF_5j_+m_msP?JFMm*48%Dn(4GRzWNnwv=gaBOQ0)Q^Tlu~`
zci3-Q0fTj1>Gd16-1bFzwFnhcp$W~#OR2Hwe%_qBIxyocCZ6}uMF=Rcyb@Fn$*Z(>
z3(M%zVN!xDb&eO*ExTCbX`6h9)%;;j=Rcq{Oe0iIzXqi9u^zWR8aC!sV7kvdAQBf+
z7HiG2=UP1(bX=RFC=W{lTKN1uR(~!K@UlOa0t{PLYV2(wk2kiO=w3Y7=r0a7#+vd=
zJYebL`|u*RVWZa9Pu|b-SsSVp$E7|05-u%q8Gcr{YX6*>F(BySetR2>UO1~~U9|bC
z#nD5T3{h|-v9sVku(~p(Bzi*lDrwggzJZ~|xi8Eph>G5GwtVQk__wvvi)~m-_8ECi
z1&BMJ&fWHByFif1i)v`^eCNazza4pivc(ma{18qvRCbal77z#kb36`hl}c}8zgHNh
zE%)NN(7^mKuHotqJP7zO|MGTTO-o91Z%B?!!NiY%S}0uV^y>wezisznz1R;j4~P^3
zDDo{HoJ&O8l2uC>zrsRnyn4J^3{GA#0?=Ady`HOsM;Gc>au)X)_xcK=&pQ3IIuB4l
zilct8>Jo}%%vmWJ)Gw${ccwI8g^S-;r3eVnoO1eAG{p}i>Pt+EK`P~U{v4IHbJium
zHO#RZ&=-77eD?c^BYvTN8GGH#Ad?>8I55Mw;deND?cX`%1Xoh@!jB^Gn93a4129c0
z_O4`ppN~ple)H<P;xp$lsy5Y#>0s$yM;1BK_hb3-Mx006ko&$OPI=8LL-h&G+*g9(
z-uXzo!dg_^V<N-WCP<W)($-I>jYbGUC#Pz0R&p*7+Xh$Q?OJCS5Zt_a*(_(rowL5x
zGOEPessZK$1*N>H01=4LtFnD31F#RuQ{2Yu=KpZjI5#GQ^kqMZ;moIErnobv>)8sH
zX!w;sT%AirUr_o26*UYQm0%z~cj_wuhmLTTLiI||%eiA!i|fQTJLUY4(XB5!e@e^l
zSW@G^Nt~cFF*@((n!sR_82!evE!9nJv0HZet)j<wkqv9^VCL%PVs2~?3-rtQ4r}wc
zXwZQ486rBM>@DB%s}dY2XUkf#2oV_YPaHQo5TQ&&38-``g4W_xCbBCC%G#oGD)ImZ
zT&M>_1!`T0Bmr^cL{VElFGO$*fwv5F7%lX2qT&$$oi<cS)EEez_uc=jB#Nx`Ju^(6
zC=NInGZT=!P&5#j1}28q(po6Grw`<U5<_XJ027M^2VSs<p#i%q#J__}D6vF3pq%3O
zs4o^Vc|4#=*>}@WQ4CB6cr`1A(c)1i=KT$PyZE+(&56n40&O_O(OQfy#9(EC1D4|G
zEdiY3;!MDRu<z!IrFby`u)N~C`6o;qhaC8O?|ba`d*(TS%wONu!aea=JYWxw1bU13
zm$(luFjwfixqu@vNe<+am-yMzBqZ?*9+VZB_aD20yo5R|C?`;PhLH@OoAZBTdv-RU
zF$@?b;8JhHRfl^XFk{%auOLs}zFpp2OcUpNE0<~JqDVzhJH>TjbOV+x{Z#W6Kwq{C
z^5ch7dAI+n7hQ{+&Y5ou8Dc>td^agCEWmW+`FGx89S;h+Ii*P^(US5yrjn-`G*3uz
zfjLb<cR>?)s@jFsxjzBKNkWgL(b)+ydMw2zfO$Q*G%?rOUP8XvR4*@CZ{(Mf1b^=d
z3l<0kxRUf*-5AK?vRfvLxsqq*(;Yz&R#SD%Yn&{UEw$7^a-=ttN9QyiO{DU<G-&S$
zgn0j$6>p!JK@!+W;dt&fL=LRYg%}YsJpx8sC0LO=<ydfqJ*9P`bJI1MZr`eO9&!OF
zKsphIB7GQR73_(pMsrmoP0>V8yDEaLOu;+ZneIsBkYm8ADFKlquS_Q~KUTVs60#Zf
z&jMo9Yzkdb7dfqMKSFI)N&+&8TqODJsMP5dwkh)pm3;HJMvqBHp*1=MILsMt27NJ1
zUr_C`*gOO|vwUALg^dfmB=i`RifeNPpd@9vN$WEkHsxTP&9pfsls6dvj<W`gUZS<{
zPwF3Jl1wlil><cy6k{T;Lt+XoNGr19G@;z4REx33X>nIY-uXDV-;mX!(6!`3iblaB
z);_<Q5Q@d;<5oZyNtiFC%XrTFc&$*^`%ShaOMjV2y_D{brLpF8wpp?%he)8i0g89B
zgP>xF{Rg(ehQOpCez^DH3z2wvF}ukvi1f25I%A>AS-C(<ME+)<`_s&3ky=>4M^{SD
zBGs5`Ttupel<|`>64au0O@x``FWs+@1acCXa?V1>Uo@?Kjb=nYgOe&jOx=*Re!<Cu
ziY=ByFkiXvDkCv<G>!{GHzn%^=mB7n-0VYck<{$75RpvmH;#l6w3-^3h)a__aPcG3
z0bP_14vwZVNT^e5Pc7BtjxWORZ9mZOo`UrKjqZ^}sv?SSknS9GN1za<lD_qJsD1yn
zhwg^y^tbm18|2VLr>5dpPB(NVA^b$fr-kQE)`>x8!f{X+-awj1j!|1rbpvQnQi%cO
z!%M*rPY}4`v<TN0kQqlo%<Sw5g>IS6V|h-QM@X$HCS|ff?vL#fO1?9ilQ9EO0kJ{Y
z4q+O)y^v5TVliYAEEJNEsRzG9YjHzu4N7*V%mghE0!31(j%i32O-V822d(8`FJ^2!
znQ!qlSbDH*EqEim1tPQH9l%flvTP&UVsK=!(}@)sRim<R<JM6GG#TOedWaO%<Pk%)
znh|9Sw}{5W`oe7Ln%$vJ&GW`jNnn-Y{cO8$J05zGzW?$8<|8f+w6@1INa2lIRw4<b
zP|=W*l$ZA)r4D<BEKB3j9b9@vm-@$S&6o#umBE~Cv6W2m7=RjD9^iqLW3C$HvPE)X
zPon*s!?wqa8cCBRTlr=*4GdDdbh(&0i`+xo<=SfToW-<<uxpo4UfxtPWLv6T<6s)4
zU7|A;4`W%0A?~1J?)RIHlyviiuoZQne^SyO7<oJuLe)O|@)LCQ?-Ok&Y(-_E4k;dC
zW-@yV`lN%150F!$8i2TBMhb1y0O-Ji0h^>`c8<M6!?5ym8A74p-m)2zuyVE6>hEGH
zH>E4;Ipf&W5Y2>NrB;QmCs_O3e$-khvq2B~Od&4l6+wIRA`GHQ#-2~$FPpzQ2mssC
z2?*D^(eQ0=jtfs4cYWR2WVmy=*R3W;s5*;vvVrXSV)8Su4uIa~3GQuImiX4B9N+CF
z;-_E+0G@RZrU;NpbiAvS#iG%hDUE@{x9gyT;?yAO+0>V6LLW^PNxTK_--zKxXy46>
z=dJ~RGHQa&1vg>)YR{9;zt0!N+oz33A-=Sj<B^cH7dJ2d968s2;$JpI+57<yS97N`
zGm3_a)vu=H*MJ@Zy{(wnUaUt(OZ*Jx_&5r~j(^r~07GDbXfD71No07#eV<o0@1FA4
z>za|*i+LOuQ*gu_X*#jO(Z}4)`|h&~+lkbzg+QxMybig5hTme0-0_$b8`I06c-vl+
zmz&GG1=9?8Qh>qN-9!}M-qh4uCJ!6Q(EjbqfeBYPI)D<Bh%Xb+@g;}%Xc-dIu0Gv=
zEKax>#Ce`_PZ04qxG~moi)@R?@?)0+v6uQ0j&z%{-6q!OhAdKW@t#Wd%N|NUE9NUT
zkcI-u^8U1CT9Cj%ZW>B361WiEhYsfX*@yZC5msFz8A3bX`40J2TXd<?uEA|xVW4%|
z7pu`;4bTmWT3&F06vid6lxqcm=jCHdusYnI0(3X-C+STZ!ESrfF7fkUKW-;l7zIE5
z_?|VtKBh80y*zlZ6|V^8B``Cb<%Q`**KT&SdLpcQV}B#^u#4z~E&m>#b4OX24ilF<
z)=eo~HsKlL9e$u!X%=&@r&rFJj8m>EyZ5+z0319w0=jlAC=ED5_)jHjO9-(qWK>@y
zv{_cI-nMiezG^-UJ5<AzWn9RWJV2d&)Z$o+bq*9terhk?>e&rx+Rm5|a4f^W)J}=k
z@tdr#epXeZW3uHN`0i=ESvLiXYv1}RGn%nY^jrS?ThUtTL6hDHa_PeMT-8~Tx@=>6
z2bjpWIRJ~XXzC|H^2E&550qdO>cDf4CQml3SoDz|V4{U!plMA;aH@+_w}D}Z;zPuO
z4+{-;)7zt%DA;CH$T7_9^nEa+S9d6rb)IQQ*F9HxXh|w+G)URQ|IWeuGB6^qUO8z{
z^(#J8qZ-Ah&WEH9S@r29sNY9T4xA|R02_>W6V5KD>3*(an#a#YT@*`H?Gr^vGN`*m
zqjyz(##Pdd5q-DR^MAUz6morOcvcoZ==$bnMKhIt(r@<8QWCqlTY!ZsS^o)O+Pn9c
z0E1L_&QE;F)@0ljI=0}+SU>=|IT1~j!;V~hKu&ji(R@liJX1g!U^s=mxHr#c2Z$BN
zK!|IK4WlYpcX?unSj=JsF`RjWHE|qnz~0CLy*lAH`^jFJD7PJ3*FBYi=n8V6{mRK&
zB4)BVGx-YXv#~T2R;O`!ja&gHFZd!KryZ0v3IUGKJM_LvufMEvv@`>*`KkQN8&NN2
z_whsgqkq}Y5p3l1zvo(Q7oSc!0SpP)PSJ)5eP$a~RD^uBg9++HQF8FaC7`LdkCo)o
z#0b=N+M}sgB@UNW*)+IDbhQs_n=6&pMT*wuyWFdo1lBW=`4XfjFH9y6WLtM!a<GsT
zJl(p->8m`s<L>c&CV}A(eTEUlE5-v#`KkXXTB4+&R7HSR#ZoIC0xAbUKnY~E3=NoB
zD~o)Kv=o^P<4<=Ec4*{|;lnB+Y-S7RM2+hnFn=ZG9rXGg?2%(#$!zheshle%98PJ*
z8$FWFRC~3Zb(@K&{OAgk;Pfx*DdY)$j9S~#7g1Bg;38k=I>tqvd@HT{UV0a$mr|;0
z_HtDvS@2%9Kdl8n!Eq{v07(l#k4+Z(HLcYX?AUQ7FJ_#KYA_z_K7?r_166t&`#XXP
zKfw<E&xe<Tu{Vjy4}iVN@A#Fw^!D>LAJyiRq+*@8MVXIMQuk;d&ib_ZSi2`u#rf@q
zkAm=P#cjdQm$}pZc$sFs6dPI2oELf+%Rl@N{SVF7-ffK?z{`JyfLiza?1wqvC%VLE
zI9JBg%#V<o!)vRXk$1c9Uo~Nv6VvZJ0|3OMMP3|(YIn_7%Ynx$8^G3u`dw2?AuG8D
z?yuG!9w&~!Cp8F@n(eYSZE!9tAE!o>BJ8}eTWd~D3l~#BrkZ9W|Du(PY=jjb+D-2Q
zQCkH{OwQx`=-mq+z~a}iegq5V!spOUr(ph}1*fLw$BPq-Mdy&va3?k+;YVY{`^8r^
z`Kr#x<H`PQ%}jky*iTP!ygvO?Ncb9XjGm&7;*Y=V;~t)yY)c{w+I*;P^3AmccGug4
zT|W27^;y}C!}-L!Cqrb?Lu>TrqQ+ZKWug5A#a{oCtH7W(0Sr5$&SV6Urng#rD2hX@
z&LjaCJ(0(IYMsc8L$tRYUi4cLEqhsBi~<qydx_6<ula)QpYxcOdv;9M(XP${L7PLN
ztynR$gwQx2qIyvi0<ne)KR#ZJXu2cvo`KJ*F@vQ5caJ&Df<8M6331jv8NJPjy3~lV
zrJZ`loe(KcKq8#;hdEXC?Y14&Cw;NWt2v2<2}kE(zbo_gcEk%Ni6`kNDj>RM?_9|b
z?Ms_gdVP+Jk&xi)b61rka-%T<W{GIHxC58W{<akp!T)Eh{vREUWR~zoL(R}d8%bh8
zLP2f;@8r;j43T^aOyVYWPfV6}58}6l(}Qcq#IYQ7fR3$8D*mC@>%qg^zNB{C@?WLH
zTwvK1L-;zm4~J<9HuprSHjdXUbfScplAc}@<Uo=iU@Ir#9I$fvIIaD(U(uqdm^Zdk
z6*(%SAa-ZebIf<H+CdWUdgPT>WWRWnJWf3wzHpyWAt+H=dYm|CM#n&L|H=Jba4gOI
z{7)VLAp6AngzQpqDBddv2tclTaS=qo1YoXp{uv*7Jb{wI*F%^{Z|E}wZycyG;W(S=
z7MfY;6PgS!fNKOO4g#PS#?rGrr=BHyx_qCmA|5W*@luxT3cJ*&?%r(TbQkr|8`Uw?
zmp!OG<@0j{y7)eBvJ36tWT&K)wpP0eSVQRnDnCV=I|dcHXU{;ZbeKCVHp3YS1$~8=
zKBr2avOLfJ&fs3Zm%BPmumTg`j%hoJT9fp|)6O^#N;O9nKW@!gw^BCqe}A$+1sF!7
zy3TiV;bIJ~msxSO$%TItX}!IXO#jZgi4LEb@(Y5!TSG_*G#?)mhwt9Y6{W<fUoc|>
zq`&jiKAd$M7rDB-#_@4~nsOD3t{}hFoZ2}l=uT&{!(CaqNb4?sJ_q;RnoSeMxUl05
zOloo#yxjOa$j7`{7Ib5gKF;zC9g%M*r1$kgy4N^I;bkBHBN~+*$5~nEd)YDh&dSw=
zGB-1)prZZjQxU0*m-;#_3!fNw9wlu7-Ho<~zPG2FBl{(5Kpj*V4zPb{T~>D1+o_A2
zd}A;47@gL)l`-tb9`ph^*@Rtu{=m>%`=Rh(kq>*kUG`_!R59J=ya`#Z@)7#1Z;mBB
z0)nrZp;nvGTvEa#=$9vt0{$aA3LC+1&WPvV^3<z8!|?*6op2tH?{{B}1lkM$NGKjh
z_Sqe__FO(MS1;l5f;rvXJoZz;h*9-A9&BCEId1!W$C?!bcSqI}kD{cYJ`IZ~)+bHR
zo#Wk`?2&?Bj|T@Q))qZelRy9RqB~xm!a=F54?DXlKB@A)83n|;-*@4%Hg%@!bbPFJ
zrqBQ{6vGzQGNppEjz3`Y4Wf|%1JN%)?a!2i?eL3n;($aOy@K-h1UeEE>7@?-(rWw~
z&QY5-1r7GXC3)+D^^mJGY?s7J15#VL#7euUj`##kMfVJ|OB`)mpcyD*S+Rd<b#@;|
z*=v?N-n;-olZ(+BS;wxje0jIEcGvf<=H(lda>JEkGYY=}`(?Z0Wc3B0m^1QBKQheU
z#L~W@U3z&dg(`gu{gA*+^B;SLnxnKLt#$uZ!FKZH2*s3_Y->8jQe%9u=07-GwW`1D
zC~vLt;58cC^7ZFizZPDD9c+q=YmwHL%G;#fH`9V`556#rS6WPKUV$Ol>(&7oxUx$~
z<CL6>z`d5;p~?yy2jVHf3v5zri7bP5(A_7zKR`u2w{u2LD9d4DWb46v><)HvD^p8#
zrSSR=qgOM<t`eS0+c~pbh-&5VA$ceY4?aFAYvnWZSHAmeimAFdbaGH{<}bA}u`=w^
zc#`{{81d*^Pse2G<I;_t#fR+p@vSz&C)nIlP~{3F8*C{prvHusWMl1KCZUv5q?I>=
zo#aPk;$t9!<vt7bovn48lsYL$B6T8dOGS)dcM<v^Wngh*-V#rIs9uFw|3DPRHL*@I
zj^tW8?U&{r>Cr(B_vJv+CE>wS2-Ul+4-p#>Dehp40#KFMK^!dtok{k^jyTlu=*hu%
znG_oLirk9wWv#&gMioa*6>jXTWoc+rSH@)EXSII8k?zi_tRWm1l*2cB7fizp5?(14
z!B{g~pKk=`y4=Q!tU=t7ofc;8W%Er5p!&|0WefDKws4Mbt|p@FQL&vC2y*tSj}jvS
zJ2oN<nRZR8OL7XV^>N|Odt9)YheNi`pi1t0TKChel$An&5rK<T01$C8($brM;#}LH
z#_LzeN(N19OBCp^^ZV!vIQK5zq3!O1fSS5j%F0S`^z7~*oh<j#WN4u*0-5r^If=c&
zKlHJo@rL?Xdheh(RCk<?T*|;LN0AqTG7DuGTts&6%1M7nnVpNpC5?g-aF0k>@Dkwg
z{;5J+!!`ZRf6pmoBvUwD#3M3iNq%^loEPG==+aU~4lp@e$X=vG7Esr@qiX&_|0#4R
zSjSj|j<`3tNV3ITU+hn6yUtm^-Y?Yw;s=ShzbIm4wrQ2tZrPKZI6E%AGepXmOtbl_
zKo+PIE#IX9noA2Fl~y?c?)1u}kW`ULH6sgYLBx^_7=>RPw{=4>C^&zD5*nK0!5@av
zG$^2F9P&9Dw`IF82H#)Qv1|PaUOFnVKETDCpsAEhu1X6XC<BbHXTG5H`^yxz#xdzp
zRrte61&tb{IK%5U-6qCd_a<*x>j`f{qK^F-glpE6UUy_Ub%;QWr8k~0eo?=oE=Y29
zm!8)TP_^^sx~FAOf({-RRPI#bUHH>sAasLkWZ|4)kqbaY83c?{=h2+N^Dury(X<>8
zU-KFuolH0$RG~fn%=FTzIc7uUT7jH)4@w?TpZ}E}E?=t44w9U>gBX*gG$H>fsE84W
zpFEQt_WeYY29E2RZG9gMPcF<64u$87NO^!QMpS%em~DP!rlN@vUL03-B*<9-*_3f6
z8KN9RD749d&>>l!CfNXKrk2X4ar)qGvd8L1i);Vgo$|_!GA6H2fSRY3#`3(QIXfCi
zfZ9EApcgV_%1o!<_0p~)3D5iGA|gPrY@i0y^Vu~Pv4?bNSCfOsFR@LK@-eD<fB+!B
z{lf+0MsDc7pqDgAlB6N&S@|y9Ip17AD848qzNlG49VCnpo1iSfp=n=oNwM<jhmPb*
z7T<WNMPOO2`*ZfxGLI}s{NgXWHU}{1aAy<v8tn;M*g3FLiXAjskYZeKvO6gROOPM0
zFgf1N)lKkfTXY-IaMewAToX_A?v55nfh*46oj^@n@S3r$(L)R&ciD;HjdtOBKYHtz
zQM*k<-?znKm=*1FcyM=H*xdSW#i`!!cN%sl!ZRT@=+mBpW;#8zrB~AfjX?PFy?!->
z*L2`mHP#<+d-yriINbrz+VB<s1Ia@NXJ%*m?=>vP|4F&CvI6^}2`K?tIveHWBMAOE
zH7A5j4AScDfy}>z3{X+g?tb1GjUq$Epj|jieF~I2(A($SPT!^tdudsE>T6V~NS9A6
zVvTSW$8$-&r}nwkp4LIlF<VnayxZQ^rCzKK0cW`~`Gp5j(sLTth1u#KTzDfWdvwpb
zGbZnv>u<%i`{WVlli30N4PirkZdw5dyQ0)RceTz`mb8-tNLl$mCKg43hmH#kVdTmh
zN1kV6^lNbgVP^VBf-&{f*xFNCE5v5^wQ2H^$b`)n`EB)Khal$U()SlC^$)wG8JH_+
z1F|-*@BzoHMbgABqqLBzb^)yk1g=Md+_`iu8q1^OBa7(${LO%TmN+cx9r#C5v*KX3
zhY0KxooVl1V=@R?+v1t?w_H{+X*uegWE%<Q4Q#6|7Swo$qhv+nurGp=Hp#A{rsV$e
z{V?p1OZ*Br-6e9SnsA2UubI~oRN*zw3Td-4LoQBY3smGAi4x^0=|#W982;oJ(TSmG
zFRSRAjL&JsapeIr+rMF@HO7)psY_by?#R<ytJ;V;%@v^%HZ9@QFNRSi8)8~$5>e75
zb>gxg(i$E)!>LPRil8lI^-~PXDPklY5%#juf|sOzW{%v-w0R44y9EDMH_4mnPdEk#
zOY*I0aYvSw?e@iWN2dgDe=ezH#hS~expv+z^NT2C8ma>NS2rC9+{%Hg&6!I_%dW8%
zS*z>SXxHD^?=_1!!8&^l59{s{XIcq|jDIPS{mWt6Aq!Aj<bUrdgXDc|6{<^!$^IrB
z)jkz|Gx{-mvPW*NW?y8MbIDtaZggcWl)Gj%r_3st{(WCH>3Tb>`1WSofUd8*49BnA
z&yn|k;r{^J0J@jq+haz++f2^W_Wr~+;Q1{ckd;2c2@oibrRpsZs$PJT^HV`rfMoa$
z-X?^8<E-(uv0@0IKqIswcBfJ5vP*_T2hDjdEHxKrxx5Esz92{x>e*PrC><s>@97m`
zgzKS#)vZI2Oxw|wTW$^eo!v!Yeo5UEXcZL6LJNSU8pjgg20xLfP^K^Zz$mqrQQB0N
zU&vRnMxpwX=d`BlqtrG+#;?=VJM^ol-BhaY1V}uDn1t!H7*$Rla>a}?sn9+OmP!$X
z&YP-dloyk42z~>D%$c@$rt1oV45G=XN+A@<O4&=3%tH0EwY6*bKnEm;B`tLupI5P#
z{|o>L9SD;vRU8a5WLCnA$<#h#W}L5?ds{lpN`V>31?!AXk!0K!sKIOF6CcUr6EwDl
z1+o-v%#Mu#8X^Og0F}UT%umrQ<22CD#7Gh6eyKLjOEJVuEl|-eil^B&GKlcOvmR65
zg$2Tt8n^(*1e5AenLxXE#S9~=%FfGU!vJtg3ZGM+(B5@tM?(Q&R7yWVeD_Q>^NIb5
zCzl^E_H2+lDayKvNp)Sqri>+b!hl92n5#xCSN0Al{sjWj3u4YRj%Ku*c*jox59YK2
zYgrMzhC)Szk%S?jQ3RGFQ{|uKmN;hCsKmgpTALnR_WbKggWf50VIlMkZZ9d?HyHpU
zXj%^+>*-mGR`okc+CQ?ur~+0ADyJis#*XcpnYj`!!%7B}rkz?v>`@uqlLictx)V&v
zq97nXazLQqJyk-y_wO-AQ6PHV&Daf`hF((EYOZgCDKA=&PHKTrXaFvG9KuM6l@kgR
z&4jd4Xygb#g&GC$?SiUyHaZW;LJTNiff5Zc*K;0e<p68@k>D%$J02T1-GwbKpy*xN
zfw#1`pHU~Nu%0WiwQP)zv~dgs*^-c5PY86J_!z1bwkArxdHi%O!99MgoNP@)yMW%U
zg(i!Y-p~@ZU_L6a7!D-eYL=fuN(8<woOoeWTu+6_cr?q9dItk~k~mb;i4BnB5g#e)
zUBfn?ufiVfV)Zjv5#xxCMhBb3>xiYH_jM`eRVOot>DJharaA!s4Zma^!$*Dd2lW#E
z&p)nhT*oe8-Tz)%QjLlZC|Z5`7zXN8^k{IM%R*CBL5)DJAp5VF(}<XCM~)l#LQHA@
zV4EnXAL^jxIzfWvm~VB5ZJ7e<ZS<dm(N|@jQQuWgf7i%B;hobKAb>7Xemts1AImj%
zrkI9WNW^epf*gt9hZ>i1fH0Ae_KZN(aS3DR$+H>I4pq5;jJSAY9&q(H>K>wAd9pjh
zkK5kzD7gH!#o&!|!}En5$tK~1VCQvsKJZJFZbOmu*^xrxD;1tn==}xQF~$q$BQR1q
z?XzT!WM+T_`@zLLPBTBC93+F3KP`dkl)S4#?V4swMC>+`15tnYd%f*ft|2upH1%7I
zwgO`@TdB@yKtcXNh1?M<MMFZ4P&3{kGV0>RAk9Ri`Z;Z5uo_j;;A@nQM$Qa`Ev3^U
zw#9Ri2*B$wao1>%>Y@oy*^dyaf5qI#z-5$!QX5oygQLt$Zj@u4{9HsfMOUBcPJ+|Y
zYBrVrAeLM}Mq?_i21<U}q7o$5v?)Cp3*l~^&rtUF;Fng$Xh*mZlHQAU{uLpySHWjH
z4z)y|BiFX~G>m1`TRE8AHw>Aq88=UJmw~dBc1l&JgonMePr4DHdybvBtkBux1BkUi
z-HE&(6yuMQc1~QlP@W-RpTG9M1s(bwNjd*HWgOkWZZPFAx7nr(Pa&JDkj3qAuqj$m
z&{f}9h2L$p{VGJc)N!s*wQ*?>n;Xge+G3#1(YN~ZwS@rE4=Yw?-SDt6+r>qsX1W&Z
z*M%5V_SO^nB}OpdmGgAlhkh7W!7Il#YP`gXmx`#euFETMTyZ`H1)5d1buQOm6nfAn
z9&PuE6oqf1n;D|~APRcpLhPpglm++uE5kIN5#}zrDT6d8t7Hp8>tt+6wt66whQiXG
z4#AO^;itGAT+-L8nkXtB4f#E7g*!skh1WBaaADl$u{i<U)~@r*VjIJDY~^mm$}?Dq
z-KhLF5CQj`nYH}K2RAa<SF@T~?K!~dRK;s-IFYu%5+Xc*^7&62Py5aWc!HZ=J|;B2
z+F@+X#K@fyY|n(q9cnB~gSWW9SI#E=@2rSzsM@4rY!<6n`&zVfHjC_70*U5-4}YPZ
ziOuFfrlA9(A00PW@Ma-kJo7C$5qPFJFCreOOF~bER(UCHE;ar&wbwx6P87mipc|yr
zZEh-}9#Bx~Z#taUAZFtjS&j;?>foQZuP#v73_pdj)*WZon#=!kA6flPA)<FAbg+AZ
z)AylUf!1JK+2I;a2^CrEVMfIGJIDZ%awd`~hh+z_QF!k`JQ(^{ZkS&o*thgU0h%8f
zJuxmwWD2WeLu@n$!cz)22yG^V2ZogK&-7K1>jQj{5SSwM{D<cief|e-e%+$=I)Xx$
zWwa5-*e}L}xKbA1E1%6{i7tRElXiqe{Yi1`giY1T#qtjSg7TwgrFfsX>ecD`6}ksw
z7GyMFE4w4EFQ4D^5_E9gr1FiWl8*Vp8)8H!5{%vNSzckQAq2NVjGuR_%ZK)83Lp1X
zuFR(yd`tF>f370)f|~NL@#Ts;7IAe0NB=WnHc72xIyfy;Vh7~lhq^#~B6<-w(ez-E
zS3akm(Z>)7Aupm*#eVDG7tIQ<-@nVI*^LMQO#?<;oW<j=skhQzTLt~AoUGIJtR@Bx
z<E!v27b&+6A?NRG!t%(H{t3)`1Q^&D>k#`Bcm)A{L#;blHU5!12ysLy$9v^x7JiAa
zd-+7W=y-|K${DkfoG^*ErSTI$`M=s^V>Eg-@d7#HgLz-S0O>ehED@fu$ben%krzF{
zs}HQi$8Wfq<~@q!d;iDBb!f)C#oJ}5n)zB?gVkP|;zvI%|1M7_J_h72$wR*|fBinV
z>J{fYE~ldCiOPG(C-)U?%BADC^Y)*W8#6wQ_b+PGE9)!lSPqyLQxh+0Z|&&Be}$yB
zg?83KD2EE3a*6MY*PsZBvmD!NL>EQ?%BTB*^RHU9@quD*v)}NyN`ZNm=#A_vXMsI`
zpFq~y`(4k=B*KpQEqk8n8R(Xrm5l3k?ap01((DADoCr;@Eo;jkB5{cBi+wRo3)FVV
zGcCIqgS;57Pcx+C_z%4Yev+TG_Ik!I%uF$MPbgC_S9qg1-UT2cBGJ+3Bah>NeXQ#`
z6hGby0vwlkNw*JiQc2BM?<$J7iHs}3kB8e1q;1RVp0}HUc*1yEi?7{{U+%~12tG?M
zH@6-Cj@x!$E;*0u_F(z~^y>B!_wJ|!0u1o7{O$+7e<WUimnmZc?_c~DPJ0*x3}+_@
z7*u(ETMc{#q16kPhA@H!%Q^`F$uCRem@k6eUo#`;_+N|}`3~h8cVE`!^HOY&zm|Ng
zI*%&kZ=;Kc6N?F6Yi@N^f%*ky(fbp{N1`V$dRA%yCO__p9}YAxc3fz!PU}mXqxQ5k
z@&x|@4NoXP_$@>Q(qhl@d6Fnelfyq*46EKnzFx!t-Y1OvO~uclpEFu`x1xY$%?uxA
z+&eRlXJ*{V8ji>Qi?+#6z#NJ)ux^q58~PI|L5ax4#r3}c7G^FsVBZ1}1;9T|xSWg=
zKJ1@AtRYbh{aOP2Kriwr%`aq^Anos4d?_Oq818@r8eq0sb-DO_wZnPK!^G1{O^Mkk
zbx8soxx+q&iuHc%B#aC;xE%CvpKF++6I<D)yzZ1m83lbVWfcD*vA+gczI*E`n%va;
zO=Kwk<O*G}nty?(RV6ACg@A6|D1~eJl@JtNZVbFWb;^QLnzI9*er3uQlsF!5T`V`2
zSf|;sm{;YwF|gM-nIy#Yhtk5tZr1T}`GQ6|Dr|#%cTh9w{7_J~Sz+lw9?3?wQtj=z
z_1su)9lS<n$rW%M*G9e_X32AK4TL!`;X}iGNRg6EvHM{?tsmk)^Z`F~h-?UWu&xrq
zY=a`wTdjg3{NdeP7W~BF1|`&g@TAptiyadmw`0dln&u`Joe`<_8+?=960ti2)m(8y
z(#u~fLMp*8$>EA%bztENqo5frviM{O@|*6kP7XBOg?QSmO)5_{mJhVgYxX16yI|vj
z6s|}+&kh9o+3U?u7y%M<ANfPf8u)G>(5^NG9#Ap}lC4(_%~jR@_U);N-zyoeb7+!y
z<+crqm1v!|ZGVk!?B&p)q_2%#l7_Q0b)N^LUSXd7igLi62XkME5te6Lm)9Y()<m@q
zQhOx&eI%BA%Z?)_uUq)F;5mVHVuquu52ps>zrHw<_*2SM7hoOk)@%BkeFCK#(gl0B
zb}6hTo>1`p^YM+)(Ffhl_vm3Gd}d#5l;izzqNPQ!=i1WA_w(uLIzVh>fzk0RcBV1@
z_2D!TNb}C1PFO@FbqGhzM%{EIV*_=qJ+wo2-8c0kKGE7b26qr<iGM$*JK*i@%{NLA
zu$Fz6vKLPH1$dQI5BR;{^61}&nvy{~dH5<`>>c(rjr3=VGX1u4fj14_V9f3hCd>n(
zWbT7+0rbG@MGTKihQMVTzJRfo?BH^geA?ZcPQ3=@64-=4k*8+)O}U8%t6(WxP5n{-
z5!-%9R7;lT#+qG8K#j<S`g*>I1(iMy5BLO*RhUZKHGt>aB}k;p=cL)o=NJ(3m$W9N
zZRMY;`Wcp0HL>5f?TY69YpW9Ayl;E*O+NGQ9xY0rY2q*STG+7hVPCLq(-iKBK5vh%
zz>WfjP4~L*=VM2CZM)uxPujEj3b1bYcKiBoy0R@0ry~Zq4(Q;f&dgSW{i}3p`Z=l7
z`gex47VzTI>$t0bPxT|emD?bAUtIZ;!8*^0emd#6iEHL<@i^D8lKF5kwQ1I5S%Qin
z`g-|jw&j~wc;A>pCAd|&At%=x6i?J3eh@A);&_?vc$CKXjaAUheU<UgZ`Hpu7z4Rt
z@WYjZg#UJu-C9QC!z(HWTdDtK)CWt(_@cI9HURFPd183$X#^@<a*~5oNT|%ke`6t%
z@7*LNcNih<TWIWgJl{H{B@?TXu$Yr4Vqt|ca6V^`KG;z7JA@D-0VA;($?xv`62R4E
z^eUbQ)EpRvnFnRWDjSb9b>2({lcXj|oje0mNU0^{A_w`#(=gb2hF~18`I$Xr%#PRS
z;Q`*E^z(j}>!Ox1wWu((Cp%=cOM|&IDCK+HG_;aKTr__QP^p-KnHXx_o)b4TC^eKV
zwQW6q3R$;-`~(b}EWI)i7tbB9n^hRKCV|P&tE#yLRh5kx1yy3}5*YDY6MB`PC>4|#
zJ#9vpJ$<Sqe2O%mKWWd?SSw{xe3Ru2%mGD;7OG`PGy&k15x|(@1V{B@N1AII?-#UG
z$_|0^2m-Zo%^>xnG+4bdyC7B|me+f6FDg8L^H(?DpN&sR7S}h)@xxpOyR*3y>eh-0
zo%meHqc<UstvJKdX&SpN7XDJT|JjH110hkI>6!&=F%)eoX+aZ?>~crMOW!g#>Noz=
z-TL3)U;Txm0)6^wvtm`YuGi#s9bMKOTz&|j9P7unv>5~fwayvMw$vH*EA?{DuBu{9
z)~;pRWbD)xq5ImSHwTO!=l$LEw98GB*6W|?r&#iW0l0rP`kOe|a!?4sx@KBmV~$uy
zyI{R?7+pZ`vCqC2gRi;JCl?OzE2{s|?<0(UUps5xLWOtP25QmL`|-r3-+raRydcZ;
zft>SZQdN~Y^T&&M7A=0_tN|sS^j3jp>o{63#tSbJQ*9J2UUTmcN<7K8^|M(kQc{BN
z>?U_lN0Rn)^hVO<ngPZ;c}SmS^3UQAj2jwnbzxO+rxn(CjiW!z@y~!h{TV);G|6GU
zJz39xy%?@$sif^+0>ZnmS>bMi%90juk?CI?&VFsshSg)eQYbOuuicBVI4sl32YT>!
zTUI);94E@Zwxow}{r^M@$CEsKuj~=S$85F$y32^IDwA{{Z|RK@Gb!4!{maGKthvdy
z-OU>`A?ObY7SgSM5e6TdS!Tg#o9g!|uNJ_4pc1(*-1F}qAM%Tvjiw>+X%vZv_d(C|
zRJmOlg5_s9LE*zXgOU0H`t>99xYg(X-%E0^0zo;zND>*K;Qvz|7UTx~AGbPHjwcBp
z`tKh71FTPdzD$rQ?0?+q9hkx2P+~G8UgCH{>vmnurL(fo&2(SclERvRIA1;2?RgE>
z=HmE9^HJ91*m{9`i6(pV##r2gF<dUkqWCH@|CA~=|D!4uK+<yaFiZ9Aa@c!kT$QGO
zioI3#Ah)%NHL8`xE<M0fDzMIc90why9f~DYHUK@PkLU$v!l;yC%m{;)ETn>FqQ+JA
zkS-*H#uEFlI``O4-uS#aL3yP;xpLTpILkSdb&T08)an1@>l>r{ik`J&+xP{IlO}0w
z+fHNKZk&_Gwr$&P%*M9uHs<a9-}kO{KfT{(%~^BK+L+mUpV`kmQDxSoaB1vueQ`)q
zE<tmYaeZqTMf??F7^6EpY;k?nbC7eSZYjA^fous-bEMHF!mC`ufn82h0ESdxRwqms
zGd6sYu(^5)OuD*Og=tv>+o*VQv2aZVJ+ia<A(R%Cw73kRdkrD#^L#OPavUm#s}<=f
z9}!tRn-m6{pJOEpvL?H0y1G70WE|8?M=qv}EQN4<pSX(FM0Eo<UW7!4;ponkN|G&Q
z6dN&pG2FR@qKE-vA>&sH0AL>bF$R%Id+IC`S6S8cz+j0`X_(M=B40ligYKCEWA=Va
zYE`FZ6<pLe7XdYAcv=MRT4+rg80t#;m0Tx-opkg3UlZ6lkx6;e+&U*(EbLLWjNsYy
z2&!gWt?F>(QxU8eUHQMR@k-17SrCG8=a#4FUaSe0rggJ4n#;QYX5OBjZU;yVdOn}2
zSW&sXe*sU*1us#4uW$F4wb#=wk8e-+m)pmy3#h+(-l>1Z^M4@`@clzKq-L3)kTw2y
z#z5^)dBU1avRW5HfJ{Y6zdN0zaYVQ)MGCV|f&I5k{p)mh><gFox5M++>+h@Ia_=D@
zigG_bBq{tx#}w5ifY$Pe+%vW9QUagqdL&I_zwDI}?`SV4m}J+AyA~x{eafY|%JQ4+
z@tn407HWc%2ZE+8x9hyv=5^uRWPV|i_v}38$%QDFOs?EL??Q%$r6`w)TFlDh4YFhN
zEE}CNk15V(wS8UB{Js2TxNd?w7X8YkBXt=iT9_t!f1C$8fV%>3vmp;ru~n>N=)_nU
zH^)BQ)yTE}p9VLcF^ciqNkeI*RBH796%tA(;u#6`v}Y(4Q_1yMI-2ria~Q=tOkMv5
z*gv?^&P_^iemgN&OeNd2{cqJDSC^v9XnXCM`|s{p9OY??$xV6ZzSlIi-?>kXfv%Fs
z-O~~wTyrZtv;4i&5|FwfUF~WAD`CN}zBlA<2DJzccvqv3kSRoLN!-~BL-WudvX4?x
zLX(|L-pdyW@LM+BvN_3@jH^_sMJ$SD3*YSB%dS<$Re$1x7-2Z6y=G>23}Hee`zGm#
z@8lWG9oz6WiuRCE{#OgoU8wJIa0H5Kf!)J|S~@ZXLgKh~_kAG%{oe71)^l(SgMnq`
zx(TS^<Pr-E#0Pl@FB5_a`udPiy7JEd=QXhZTK4};8BV+cMcKI0`o4Vo|4NIz$>hus
z@<SM>z+rP~?Z=gj)yEnyowvUCh@FXpkzLN2&-TGUmu<ya3BkQKIf;H*xF$tHC3A6E
ze@TLV<=sb(vxcmo!A|>fFhUoCc(O5FlakZ78%d+a2yi>ZylSuH8;I%2lNRVr34O{r
zJBe0?lkbYH;vI`8OVFEVr!7*g1e+asI{cdi!0Gow9s5(_aCA%NJZN{xvz2_u(i3`f
zLGjn4X4NDMH%L9I8T|H+TFj76h*zgq5*MO-!DDL^c3fL-rf;yX6X59Jw$MV6V!t55
zzJlq1U)8Fwd>fHxu$|~@Q{~3kPcbK-#5~<)`g;q#@iT>Jt*~(2K_)GVe$uqjA=;SJ
z6fh0P3D2PyZ!k4=0jd-DS2ML6;wS&xlzi3J^sj6KTefo!lSOPgPFkW-j@>QR_N!by
zd!zV2G2w){G;9?VYd(^xX6f;}9*5<%aP@<cF2}P;@IRtCwgZ558T<RA?EiL?EA1Ld
z`n$5u%>9Gp><2J<5xW*p(<&J8AZZRP{2Tjs<^Py_{m(4Bwfxa_DkPuU@9K~pd<jc;
zznmg;6v5WHSKgzjZtU<Mk#g3hCcd$qWb%r$ve%`je8?x-+@Q0}<AhR#GqG^>fRWA*
zS2B^Dm3t3b$-x;*KIZp97_MvgzOpMH3(p9-{{Q&VUycH3y+5wpbL@rW9mSp05)R~l
zIWoE@bAx}cUbB8G);HE2YjL&YeL56>f(z(|YW!zW17}WU>K3L*CG}@Ugl6VqX5&g_
zmJ_7}vL|Y;+J0oe9@BnjU34T0cZUf1Nh+)zf=+P?Z3V)^J_fs`962P7&AmmASWslR
zyd11)9x}k-=H@<5(+F<#Au`xSGHCY;F7(a*`8izMsPXRbi4gpGREzg~{hW;Q-nv+;
z_U!OBF8x{kUYT`|YB6b0e0%-3^31dJ%I@bUFkCzQ*ZK2|@7r<h-<_Y&Z#;@QqpAwT
z>33(nzl@hZ|ICz;NahgS?<XacdK?wq=8>d|?pqw25%Ao{j<YJfw^?cVdD&2ZGyO7x
z8_6^E0LwrJV)wIB-lgA_t%b`^`o=&f#;+8KP)}KX@Rg!lJ`&+s&Olpev`X{W*Nhk(
zY~XD6GfzCY;@DC?|MXE?>_qZu&Q@eO4+kqYtEVYvw5kbI{q3Zr7EcU`flA_Czc!}c
za`#0KAvC2x=g#}RDoxE|R*JUtlin*<e?0g8E2n&j;JU`C=UX1YzEvSux#V|rBA72J
z^^W>h-8r4+W5Rwvzx&Hg_4k+}I`Wj-9?;er_QLK#b)jjsyp1dYx6kR7v2?ZE0_jDv
z?_lZM!1Z2T42v_hJ;O)bqd|cxeg1d5z-meG%^kh_ZyuPpy31{r4rX_4JNu>hx6Qkg
z(35JCX9vqok)3XW=zM0KcjQTL0-%w7<n(ypraju?0iszR97Vj!Y^JEScEy=!8d%P`
zSMyNdLLo|U#QL24c^ZBXqba=Z^y2j<^CkE;lLCi(*Z4jA-ra5%u(3%DyZYs})1dJ4
z)BdH?GiMhqGkM`XOZB>XOZdI|aVceh<m09G^J)9z?e+6R-n5qI<nu80Qy-xGywCl-
zOWpoBbtn1y`S;tecLPL-_kD+tBjE3Fn?0YOPoepc;Kx<#=OgdP>AEW0{5h@merzs=
zVD2AZMoDc&Zr};)$G+cq+R8`Lz8Avl^M2{Ubm`!+>Q3Zy(o<q-)}v}o_@yHJU)t)=
zC)#R10ZPDI#^#*%nOZSCSCRXdAZ2ywTTsV+V8`;{Ln-~)Qm#_#(R8U#DPXU+P%W5m
z|6XMOuCUorrsF5F`A2&5CA-5+x%HpLVPeNUuiHd{{WHAVUuw4r_p=o7%{R=mBgM0n
z>xIe4j%VtH$=rp>?ea`qw|T{~Oo8&uU#gr8XC=8xoW3fYK&id<g8aV^ycOBMC~P*B
z5dtEcPtu!r*&R;GoINUl`v%@GOg6@Vkm$Vk{mV>E1e2c6(`n#&d3(9*<3aJW!}Nya
zb3pJjllVqscS9cwXL3N|@A<At`tIv^e}OX>;IaKIeQ=XKYJ~n=CFnOw9c^Il%t(-X
zJA(RL^!S{;bzJZ`3p~Z`Cj!&gZ{GCU#DXP<rN*5l&4Q-r4^O}sBf-Tr_R4qs#G3c$
zs%5)vn)Zd2njwyEx7$m@MjL{o1{-hc)aJ{+#5UBSl$19@Z`WVc`Vfa+)3BT)=cc;X
zw7J(>9S$14&w<X{_Y*2^oh7ebr>fJw8ZNREmooO=y|BB#rVVM^^nTf)KIKiiAGWxz
z*Q^PU+?22MoHtV!FO8$ZKbFt;R5lgQj-!S@l~489G#3B<Lx?^izYW*wx-9DE%e0tE
z5hKw9obttzel2+;LuWsE9&((j#(q~sQMWA2R^F!!X4khqs|H!jketSae<G$C!$J9x
z(yQH9Y6kZdXRput#h2@x^Xva(>1jW`WnR3Aq!=*di(Ot&NqEjRGq=g%X6kdg8}{HA
zVdNj15|l{{evkb9Tz_HHb)*^V(*&AVP9P=$LMn{7P8_muy~M{La!2w<TE9r~>x*Fs
zuP=CWt5^y@mZmJ<i6>|5#oNqh8pE^yc_X;G^%yKM2-?W>a}ml-n|_6^judO~l(DW(
zQIYReSw=F6Xy7!l&9{kRJ{m=J*8aLMpFpfqP`B`fw^wjRWeJqX(NY3R@LPyk32YGp
z2Cp9f;W*FHn_bsu`A5)}$y5-yEfp(Du$xh<nS(i8cb|Tyq)p&sZbg80$yN^{;Z<lQ
zr?qvEmA|47yg{gqKSEl^W2=b6twODp@T52@D)Z5{%;6k+HKl0w<FB`u&=BLGh6&&O
zQPT=q$!l_sJvZ13-Wu9Zf9wt6-`v**Owbe0*xMr&-vj&@$*g~LAC!a_>5Ell0e!#2
zY;v8Kyj;5<b#D)&s1JG`XmQ^}_R!-IP%bxliLSVp@De=)2fuf6-QRh%A8QUpWtG$z
z6jRl_iZ>K6k{#&>b&aDp$MV2G-yy{`c~ln_Kp#1Ss*eNrNzLsx*Y5;Bo>~FmV?^-t
zgFOAl@A1<E4l~<^+4EbLiI}L6c2my}>5F9(E)@R8Y`u{%y)sThEF!mR1!|F|)Ei^J
z5qZ+=J-ppQ(wyaSWAnt=X6WSXNj<g{C`?~fk?){LWSfhw0<+00Td$ZOLUN68^t8QO
zGCfuqdazVX-ED1XM+lTw-h=yq*IYPL{FCN>M|^7&4U6ShG#vgdLbZ5t`HM^ncr=8%
zXXU9^=)IGLJ45FDA{4$9QT^#u|M`+gaZe!=vYBJuzss9snI)Ut0g{4T;><OVvOE!o
zgk;4*nx<JXs!f_|uT(JS!gFUQ`6KfUC5L{Wc`^x-@k%~4ee3LjUK?KlTZ}wrx65_~
zAza}*f9Iqhu{PE`{a&31JYo`S+oe^~s7d6qY0^*;(R5f#1<P#*K`V&)3|iE}3856)
z6lmvaL_!Q_qib@5&-fyrc(qRk&HZJ^mW`fZyxyNet!H_;Yi))l%LsxVMK&*9d0`M1
z(pS&m-HaG+d{^vJnU+6*xg1+#crym&q}g{eG@FaHz%MW{?eHMM!F;NGld6543{++9
z{ko{I?`U|E_Lg$tA8vAYfsJSQd#+(+jRp8S;lD<{_>LO~_ok1Nv1GOkiwle?g@0iy
z7PD`eny}H0TzZ~+E3OY>iZS(#A6Jd(3J-jIB?%DwOe|<4|H7OIoY*Q;U-Fjb-7@Y;
zTgp0}UL!?2^DebLu?^E{(J0e?=X0BD>K7M9|He$~DsPmVQ2fk|e~$#~l)z{NF6w62
z<Z7-hNfbt2yhet~qg6;e6GOhapHQjyx<DZ@bW!?)dwIy5r|gZ#DuM^qIMkZM`jP4c
zk4jl7l<VwYQ4new;B_CGOd;?j|0t=7SycyC8A2^~!3d)fipYe2YClpap)f#Y4+|{o
ziZ`@@8z-@7w2MNn11v>T=<l~<jApVJu_{wl%>H;EF7A_e35HV3^(mDiBsbONd6e_{
zE~>bqaR@0{+4=Oj>nGvv*q>m8$7S%iC->;mqNYyhq=l#+KnPPTikQfLq~@as7$?~k
zv{(^=mQklq88NKPsfX!|wHa%b{PiolD0o6IEmX^Q91jvQa25XV#|A8wcRzL5Lz{uo
zJc%&H-_$<gba6|9S4In&3(8UAqjgB~Z8JeFx@Uytq)qce#M*k@K1syOYR&c(40An_
zfziRd!U4v@0HRNyB)p5fBB@`u0S$SWuWLzBFj|C!Za^}L;7mKFtW+36w*=yv=JHGY
zwRA!fCMnYFsv){9gAhz$Of`xn(~Ip%h99KEEet7`a<wWo8583@9tq2VI!4#_N79$#
z6S!dvVHuhYNEF^FGq2v9P!3IA^rlo=wTb$rD-y$XAj~Hmjcgojonj+Yh(^^Sr5PQ2
zp)z9kcY%?3`uPHhT#s><C8o+7-ZE0t;|KW7olqFPTa*>~gv2}+GbJshtgnPUF-|v{
zK{N?kLonAVj2wZ(9t2U0e=qh(mS}D*cEm&k3NTyYHV3T~QT$swA4y=Bz9|@qelqV4
z{9kGlfJ_(AE>TMND|v6VL%+gxb*voBorz*=oh1A{#?9FKTl!*-3xdry_;A597BM>S
zilujtuln4$eFRfeGjshP6+jg7*p(B%id;dkwbD2k_xjXpmNWv+F1;}Ku7Lb}TzgUM
z?KpUX6J@4Q92#>K{<I4PF|3?gTzyIyx}{VrU`rjTS&rXlHpyUY76Hb-f!fP4<Z(W?
z(+~3Jc(1IvrI@8!8@K&BM_HW)QN1CEHaEkeZkV7T*$N*1)(t43;_I!Q&@t2KpMF1f
zK`avHWyvG6B7cgbEb$I$pyXNko<wo8U8%g^E|h<PZA824)yJ$XYZED&y6_5X8}+0F
zl#1<lxXUrukeD7(s_DUAND!QPlzM~Og69%@aQa9KP(&uqVWB&P3K}NTp#SJ|+;BSm
zp~Bj_2Es6JU6bo@!R=~}jTPGOg9ojajHXS9m@=zRRIa(thFB0?)1aiAKtEKDKdDS&
zhim8NjB(?~Yr=Z2s9XKprOm=(DR8ENFAdiuQa<OC4K-}jN*9&!d)jKlqGa9;;MQiS
z8iUboCDl&pu@5jO+Cg99Lw&<pg>kaQS834I@*daVa38C%!$MXb^g?A;L|^UxI_C_L
zew?NW8LRD_phw~3l+CkuYwXF5becdPviM(WlhzUZn3h5sv#NY^Z4B3Jg*0#jc$!2e
zzpV>sXEA;WQApvE;}99=Mmi7&k76L24T27lA?ZnR%*93vWQqn$q@oq=LS1Q63;7{*
zu=s~^L5!}x?r_ofd!B5?8g{m-Q4Z6a+367m?E|J24W!7zltj&1kodnEKJsK5I=*|*
zco>xrKR&_C=!<C)sD8>KVb!0%17Py<rSNwVeu)Bq5SO43wwN}47&ek!xbEHRv_+|v
zhYVYXtuh)qX`1;&O_<`6#eY%nO@U5M7=M4mc%nAO2GDA2N&2D6eiO_j{!2dgivEK)
zF*8EJE0d<Hm0ZdF;M&@g`3i|<^e7c0g^LXt=@(mS1fTn31p2i=q>a*E93Vi*by89O
zMk+WuN78U0PxjAku_&>>ZoXEk2M{q8;5Omb48N=TV}<F|kEiwWi)NE*?Wz>QDwJwY
zmiU%mf?YswLYXA(!j2yL2`@HL+%p?hd9DV695C3MSXlJ_&eQf<RQhVgI>-@RryVUr
zSQv$MG)uFT(MgyZ{^4QG17t6k_k_b&qb@;?d2T{Q`ZFn+LbAJ@&?7Z%*l@;o?Y4$X
z1ciO^n(V0ppcu1iA%t_V6!;+od`oWZ6(?B;<z{0y{Ot(wRg<jWk6cJ<jwRSi+FYUG
zoo$*I%gR#NuS$a)0x;aVf?;Lom&i1HdGFqTm}zaYe6L#?!L`7%0U9EnM@<gGQMJLu
zngk`Q$zJnS5_Cv)*`yvxI-(^Uo~UC2T&CtsrI+u=k|LS6hhH)7NzpKLUdZ;wOk?Mr
zVr~B5lZPG4Rs>A1|Dm+%`ku$nsCDZVG&W}kW}OAeh#uM>@2NrP_be`Sni}CTryK`P
zv6vDver=;Wu|z8(2!xMOglC*KF5yy|-3Y?Ro|ld_^lTCzo|M9DN6s&lwr#`o3}2?8
zQ7pnBes3~_bQ?p@n!*28tmXvE4Ib%8RO7_wnUQAS_Z?@j#k2mdeGv8h3{LfGna#Ej
zW#qjS<cSV@-70thCpZR<nD_*-gMQ(`Xm%hiE^3%W{NwCf5g@U}npBnY1vUDXMH9sm
z-V<GmTFFrYj>&z6c`09MpP!L5fKZY^Ri<21<^Im;*ZT)iEH&8hjPjlndc8llYnP#8
z*Sck=4<oDAarvIW9PzqMN+LO`6_MJUMR6l!U0t2o0EG1;Qcx84kr#X>R-(9i*R$Bu
zzU%LuCdCK7-+-jiWZhmP)|&8oqrUf{Q*RB4rFs}EEr*XAbZM2Tf#!Z&G66b#xXL+|
zPWq#C8GazR<`);;Pfnd(KRzKx5o*ZNa?`RT{lC)OS5BR5PAu>+JJUv~xEEwHDtW}$
z<!zmI<Hz2(x1oqD>#`a@*8~u5Q?NxKVDp4k&sQ{!dV!K5i|Bn}SJ4q&vK=u4%L<!d
zA(cH66{BQU5NJTBCnzk68}94`JeyR$fk$)tnu{(<Mw~dm^k>B!#^c;W5zIL^lw^=B
z7u;T&H>j74?num<8Otc(5$A3zJT=cQEwIo-vw7KLPP+)%^*aWs0O~<@>UPdedeF>_
zH-ufN9Z<LbBb72#dB<>p>|`tcC#5YLG3w-AuK{(RtztRei|#&89!7tc@S2!v9q%{3
z$xMy#8^2f4l(1Xq@!(7PimxIAFUD8#!<3Xmx+qd-x5LG@XYYX?1+hbv0hE$LJJvil
z@D~a1b1Q={<<*ho#E+;5`G{>&Fd?K=S8bG)Qh*h%37m(;&2F4JSKak(2&TA)`b%1{
z`>XXC!F5~Rvmgr?Y8~^i;;(JSPRI!nMCut&ZZU5I9f~RlJ9y;1)$<svR7xv~K1%7E
z>c|)oP>bK7Nu&TeDwm{?NEon6=_|<?W?(aiK`$bjVW2R~H^OtQDc8U7v@jYjNw$XV
z1Z?e#`M>*>tB;MIoviSK=Z}lM9g?zqxxu)M*5-?FEm!-#Q<6OBhL`sA@D7>leC~oS
zeLSJO;q^_Jg#&$?{I~~Gq@k?SQ>?{ofSSS8FvxHQ^E(sY?Se#a=8F7)kqM$xy#FBh
zlVY9DG)8hHq^%8n`c7$F<)D9+P?Q<(0)Ufv5Fw<PLCz!Ib;jSJrBo2uwrrIe`CI69
ze{RH47}jd702SQil`WNY@p{sL+B-0Nga6g7bI0^cfB1n=D{Lm?m!(%&dMymD-H3P@
z*4V9sc>W@JF(cV6D^$3fmcQm-VjTKEa8UYn5D@QCQOM}B*(-*K@=+v}K_$nbXn-d+
zw1c~#IGi6poiQB#{Qwt?%+@^lyl6*Sn-M&%(M$1q-<QL1Z_<@pyX_033D<a7POCbJ
z6fT(3gFYbg<G#rXR*vz;o#W^O4^_<f{!^(8erO!Bzabi)Q%YEzf=J+c56pAwlK2`!
z=tNd)s_o&zyrWSI4rc`HA@!bY0%&j?-EZGW=lL4;gfUa{6Dln6Hn}9ugC3)!esL}1
zXva|R<9oLzl_bTPGg|K=J+j6UWKoY))o4{x8=6OH_E&?EeSaf(s?K-C&ZL3vL5XhE
z09}qqup4;onVPMA;Rl!AU62SBr$w7oiZ4T-?muz?NG?3-NjVQh`0Pz74?scl7eaNG
z`G5x`-JmOZHsno2b8blOwZ6!|{zcD%_CR^n^lD=dHpN0;fi#m#^9qu%$B_9<G!zO_
zA7Mgi3v4ug_x~H-Kg*p+<OhlYZK_D<M7p9EV~nyVAtwDBnX%i5@j7BrSB&Eq0@yt`
zBpJ0w1=m>^KfMR$F<cdPz*vldYE4K8jp3J|A*)kB*yPSXBpeCBHLUAG9*ULYQL<Be
z<aCXbkKt5@NQ^bV38r-ifELCJVp8gt80v`cJ+<N6l|nG3cgtcyt2eN^$Za?fr|Q~%
zN24k;W@Z^9He6%nO+3)gTcKLNk~=uWb3JZ~VvQ2n_}C-sR_8+ka_EkIc|(zal8-aE
zAQ}kpsb#K;3d}G#p;FKYN((zm@0ez(MGLqk7w)yuT)uVP??Eai)NOkaJ4Pd6X<jGv
zH)H;0wPqT~$1Il6A{t0#$T^y+7v>oCh))SNO7gJ?O9Zc7S&e~k$-&k^DJ9d^METI4
zVdN;N7qV={qsyrP*}OBQk?SbfEj=c!3vnwjm>{08C`>(;^t0t`PHm2VRH`eCCl8xG
zHnn#15p$NpNwL3YFn&O&$T`MBM76>7h8))?nk#PZj}ka@b1iiTabj_-Sh7nvGm$S@
zQcwkj2ntpN<l9$V8qB#wH+)DO(?K8qz7*0S%{Z92C~Od55&$>%r_VqAiCdNri3i~W
zS?+;bAFf<Y#G=5MG{%&+aS(5ppyFG7&CBT<vaZr2RK$Z6MiLvmbIg<x=kOu2WVBJ6
z?3@&yC5&wYj>|M@YltL`YcEPV`%a%u7*^-An?y%on+1vAzi=CEXS?Ep!C=9XHZdVe
zj<8RHWhz6iq%UXO)j@ATpYK0e+c@noI5ZIL`AM4a*TSKs<)N;=K$IxeFkgZkV45VD
z9yAzzl0BJ;DUp5`I6^TKI}!^Bp};_F!xoqLeO|nh>rpgqEQB{d(nbMZ9&JnwvX?NG
z{6+0gu9gW~k~)kx8^SxPu!7;uZfsCMJ4$)x9w)#X=$~eUNI@5c750k6HaF?&H+aZS
z=)HcS_t%CaM9j*lC#1mvrKV;C*yqH`f{lo~QFpqA)2({F9K=;(d_r=B+Y64KD4Iiu
z7_bZ)Yc^$f&vC~IqH>t$N}v%ngn>w2m)1n#4MBjb3jCUZ+lEy?EWGf9BXl+N_~-(y
zgRt8A4mNMLP`ZE4bMyG#!p`z6M!|-s^iVg5heC3iYY6>?-8BTUa_}A!$Rf_eH)d!@
zr9oeULDWZcPh$R(m=XsUfI~7y;I7Z=v2TbeardOVd+of(H8~iXZ|+zIMz~8X`8)D<
zCWt{BX$cj=rWuqFz^N<3zt)Iuk0T)~E3@z1_c`A^bjTq+DniaJcIF|bFaU<}RY)HW
zZKu&0r7q=Eu4Y|Wk|6W~ur7sy8NZ4@;6by2;K}3h(iC44_OF<?U4notK^dD3k5b!R
zWXjUn4jRp*_xSbsMH~Td%A<cKam_bWeHr0$foz)<lT}v1dNDJlB^fN$O0JAtTNSzS
zS~Ea*pQ6gx6AIDQJ0mtTvfO!DI4NPM0FIo;uSzxMgzQ~2kpfi~Amk5-;mjP4GE!*E
z-+3?VOL)>v(F#YoZ49SKI&DHap5}*4YuZJ{eGhv@COMj84;};y5XSrP60JY1rL5xI
z*$_<}6M#%U_GrIZ`ikM70Tm)QL*qd{LG+i5+rXO&T{4Z}Z+(nUU7a6V7;HQq1!j?%
zWGXYxGAJ7}5^bUgs5zb0U>sW;l_Q41-#|Ix>e-sYcA`yJLnqA~$bqIlhApDftXM{5
z&+=$+A!rw}Nkj~H7FAE6`)Nm5E24-RcpxQ@wy_Jw%gn)uN`<naaK@lrwj)chRq7Et
zLY#(;&uWT7a8<>E5W(6Y2G>@T81B+!q`4(XS|UT28#1~CxEB1b=SwsFFi`+8WXxuq
z0$E@4Wtk9Pc~+PTQApcW_O?hGZo|Au!mxNnl~QE!adH{%=;zdoeT$w#%@X3;l*f9P
zjmkYxLC;j3P9H)EMG4+f%2JFF{VEn;H}{Jwp#NcR=zip(7whhiGEBw~A)l~2v*e;A
z{P%|EQ#A#^Yve>BR2H>3F*&KmKY<JH7keDDhj*eNI`Ym+66941`)rB#^~KCf{3AmA
z7LI%UyAbNj)--JFP9K?gFxEjn>M=3=BnI3mkve>h_}mHmvq@HyfHcpzQ^S5kki(J(
zUJW~mN<r5^yV7zSp2m0c5<{q&AP~I!3Ql>IOZWspw_^-Xe&Lmwoy=E8{=&?)p6wP6
z?lbM5mgNy2Ly^FhF*K^&UMfU^EJ4S`M#L@2_}hruUIGt^nmou+U^(J^PYBDJnY4wo
zRQvn(0RAf3mt}ZIToY|EyDlJ(l08T(Oe9>5{IrhVN<p7(3p`_M1s*-ypdQK$no+vP
z72^SLM9(TTwETyzitiI1)M96XPL^q^!rj4$r;t5}SbgHrqjd6%vH<x!tv2s(_9eOC
zrzY?4qJ&_85o1Vk5NpsEHoS*@;axLvM)5n7HAA%B9bu@-=X%AOdAWU)B0|jZgSZBJ
zLyzgOqDK}DM`4Fz=y;kya389){daB;iMj=Bb9qCCod4#_M(z@0By?5UW7OiG-9XKG
zhcPC|n?=jp8oiV5u%4i#?Lt}>Mj6Tb-vr%#>r<T4A{iSP0KcP^(jV-5-9+U!9w4Gn
zP=or0w+en+lUT-8O;8{No4jH`Z03oWs3vMK@<JA}H;aU{J^~3h6o(cX%N#`(=t!OV
zSRw%eQd47HMGF2Vrv<R0S4g)FXS*tN$28?B83zUB*N!qff%&*Z^MpWt9e+k|Eh<w%
zUeVWnKgHLV4j*}k?0Pr#UkOi{KP=QbF2Z0L>I8pVk_%wfSr$%#B0U#r)A$YJDv7tv
zJWyd5Wx6%NXOl#yF8Y}aD^5A{lbT7SOHPj%*niF;+2XYJlsY~eQ^%vG`8Sav0#wsL
zD3%B91WB87pby6)5rpt=g5uy-Ucz6fbwO|D{~{bE(sK{dHlF}Xf<Kf^NS+m9#{@`x
z`=m)keB+9-mG04_fT!;?D9N?BvgwxUCs8bxl+5IWUvozxmqm7C%(7Bo5l!6dH9a8^
z&$1=BvthzQ{Rapg_~5$cjN=MC6t|TEyTUsr-E_Gqi?ZF3$OtvjR=G$-oS`^4OdQ(|
zC*%swZ}kQVI+9)@3nHLr#V4;Wp@!XCe$?d+jTvby>5^|@Z)S`IgK=Q(g@~k|e<Wu2
zYBBtYOFpsABF-7&s3qHde+XqlC$R#A=crAC<HIA{$R%AS!J5Pj!?T#*z#ucXSSj2g
z{x0f^XAkd{t=A7rC(d>!G^b;c7uX*T*BrE|Yr8>@uQRO@ski?{9i^7s5`(zOPBjV&
z!Pd6>z9e=PHmNb#(RVp_c2g4PP)HH#S{!Ja`@|dwhFOS92-n1;Ub;2J6|MHKlwMwb
z7bdvS)E`)AvWrp^L-&wu+F>^v@TB-GyzeKvg!mMq3>Nvr;p+bSTyTAhJQu43v1*IT
z_~F52-C^Wx-k)H);I_RVV#FfgEZ>h%N29-aG-Ul*S+Dc`ei40uC^M<)H$X*oqh{|=
z+{^o2upC+F?kG#--Af+KfVCqDNTRilaP54uLKh;FSb1m?BejJJJ)|6eF(S#Nk!|rJ
zgim!}%tNuDeB4%7-3$5(Zazhnp+qlMgGFEm_oov_29X*kIQknV&L>8?J)&CTkHmC+
z^Na-#<J2mdT9g$^(VRi$i#L14doK@BNR^;+T;2r#RZe7+xngPJ1mb0Jpn{>o<@^;y
z#UFRT*`@p5v>JtLh=SW&7R40R>fG@H?!*=9#G|4k<kmGo4|hUxZVzE}gS=~qFQ=jM
z&T4UR|MY9Cdm>sWpDcXqAOZ8=T>uf*@(^eIF@393iEgPDM8WD81`xO@Xh+m`p=71i
zvj$3RW69FyQkFW#r64;FaBx}q5(Tq5YDbvK80p?k6yiqU1WrkKVpDSrG8c8ERtIT@
zPV}wQ9Z`L(j>El4oq!-sCLGg?C`;+rpO-y-h#X?gy1Vn4Jt0V`((dlJ@YBZmRO|Dy
z#bQ{2XcYs7n2}VpkSVgyGXfM-1uT|LUT8%864D806inQLVUQDmJ*HAOU|aK>MJ_W~
zHmRwE?d%U#r;TpZM!f$f_mKh)Ikr)ZNg_xj#e;>*g;61!gNxlvtdy(1pW)z`N6|`v
zpJyniV$t7sgQ)ew=8(mFRI8|kh`hc%f}9bOQw(o7AWZG&1;(pKBbM1@sLR_Zc0Zws
z`*Z!4{>>Bb59)s4@u%nCf-IX=Hm$Q-_?}cF9R+TT8#CBc^Tbv0qi<x+*%EttY&d9=
zpgKt>zqt<lAyGPJ@ku*V+##3o<DD?TIU46&ux=xqaLT9}+dOs8NiN4xTWhR}r^%EV
zbCRNY*O9BCiWt<V^{p+K<#1i3YHD4po-TsRO`wcWD&_@%g5}+ZH0sKJT=ph0CTsQT
zP!3p^TiB57yQk3kLBKQiH1r(%(6=x1f>%mJS5tz(#~E-wPb?t;BFCV!wcWH_<U)pi
z`O3wScN*^E$lQVEzNrAsUg!VRZFu1CP}vM!K^f-&JvBj*7@L%L?`jQ;D8YuFWod|(
ziXHS0LIVQ=Ziu?CQfsv%^Z0o&S0oaO3-=|7d)4IavqY4W?e>JsI_eL3erxvo!?M6#
zpyElNM7FrHb0;zgSuffUYJ1kmtANedkSbyD5=dnC)(`(nS;Ua%DFC>pmJv4pqG&x6
z^gBhrd*vZMKdWivlwBzb<(JSiI{zU<IyOTy9yP!R-#rwX0J7XqGkf-)dr^xF6ls&D
zs|V^R!a^fYD_1j%hW%5Ap^i00=}xu^1wzSHP4_fpq2SZqy0-#hthHul+0)^T4-C|e
zTQpbnn{)mxu7r-;nIHNYND*J?9LFh#4b>@dE0f0vVxr!1o+*w>=yO|P7h2g$7l9T3
zM=AWoeLMjVkN*dZaop+?#g;D=aR>(NHY91hkO!blU2&laOtEJP`mO6BD=-AIUm%z9
zuqkxH31K>mTv`JoJup~94Wgcs*lc|Iyj=OU)g_C_=1Yc8lZat&k#;}7Aee!ffP$T*
z=Q#(9kt4(rphVdr87@tXe{84jk?#?tb_&p_&7eVpbia2+7q$BvH0`63`*6P%x}kK|
zPe^;=55XI;ja#V_h>b@S!JHr(5;14Rn!SvY=Rwm4B<D|#QSrmm*krMJM?(3aZaVWR
z)R<~Yu($~aVY!4s+Dx_XN0qQ0-xyVqtLm}Qxok-H;&0!ctUwj;40v(ZThjBT)B;YN
zD%t|yjQ?#%dGkVNP|3URjS-Ej7HQqx@T`36hLD)!<L1z267q}mvnIEa#63hR$<t_<
z<%kR%AFe5MNr%@j@hJ}gGV@nD%&?^z?S@PT(wMOv{q_O$Lb1mEUPFY=d<nsGek5t>
zBW|LTP_rUn>G#(;M0BW}j;iH-UBF$#`w-lFh`(;9OnsyfQ-QzO5a)`q!6r1*<_f;N
zcTc2CQ=t)$_;pb*LPQZ>$$4p0hmzasMK`#Xz#$0dR94I)e*PZaiCv-pH6&^<G6EI<
z6aL0Dh9EKwu)0^4OhFwxb-^n{0$6*bi@GDL%tyn@J15}#s~pQFgl_&m08oQ&#H1nE
zF;aV0#{=*$UV8=(eo_%)g~<w3^jb=r<BhrVu~SU&(g&I{=s(5#Oy*^gN#@`pvS8JA
zECpvoIzyebh&A<sqc-CwIuv+LK^R>N=(z*71A$x`3bL4-rQCxcd&DE9T^}xKo<U;*
zPt15->yXkz(JO1`AKFfrfL563vmBP(Gpjjz@|J6!Bh;5~AQ^G&%VLl*I6FjGe|Py}
zw&sNkzpfMW+|yrz9}f69^4}Lm+oa4k`8gnIC030XHS!0FyvTWR7a{lh&WL8;##oRk
zA{xewr4aVB7CDpsC>$f~z;&|M{=}m;WDRSMnyJkSkyX5A&9Z4I0J^Yta(VOmcB!z6
zY~soAs{{wKS{P9&gt8y;9hWg&O;-CGNRzCyXwlI@++jL7T*&9Ag>8IuHQ;@R>tzI^
zC$OP<Yak9BvYTvp@l^XLf%h=eT@Z8N7OO&rUzBsNK<F+19@uCbTww9HQ`^6%3f!e&
z`I!09DB*D=Jt4C=K$BucTB9fg@jPt0ZiZO`=@5FJ?Zk#!4?Wm=FFve&!n~#}`jKl&
zqRUdJYUM3?>8WwCdyg6F<}6C5=w!_Zfda;~TB%qoR2|`NSHqyd;NZFw(Cv+jxlm4(
z5#j~`h-KE$b@6Rq51gI7iD0(n&!^rdjh4@HBuZo6JeZG|4_T<KRku)3kP}%J@nUk}
zTms0Fb7{A(yD_e)F-{vt!%}-L-!Cb_M*f{?UVlL3GZ8Kq1#k3I45s&rDWK4d^L!$a
zvqCjt9eGfrXq6jPxxzbP7;D<|8~dAX+U$odQ@L~Zd*q~E)=NkOQr3wt1RL`5{XWF_
z&UIr4?uvplZ;Kel&TeB@+ZeagUuqJGydUQ!{Rtq$TDf*1=B)*YXT8N(u^6ieB>j#D
zXWL$HBu$*1xg;kBSQUTyIj3@IU>n~?a{5@CV{x<_kjqP76!cdfx1_l2hw;oIQM8|L
z8}*z{+&~8Isg`B{lykE%mnrK}dXz<A#TpM^4N=o>-t`I1on3K798H<lv%mUs?_P#3
zXCjJA**4XhOl@j~rgD+pX7VBFbA>8D2Ww|Xq+G5<KXD#0`ReBJ9jVr$$PQt<Lvq=e
zL49(0Zu%c1P4@`bj(l7mg*K7KHfs9D+kdYmj&@>3eqWOUZ2D_TA+>IPro*F*4K*Zy
z*I*TE8If1}>@A2{2&ZbBbQJ7J{-Y-BJBV<1iIhEUm`A2%)*KjwllnD#zNu)zW}HsH
zc<UPLPjHB22bYO+?UYR!<@Ew6l`iPc`gCW<km703#`>|uWTYNG4{S>BsmdwnQ5_;f
zbLe?LVg0+tNQog8%LOJ$oy{^rSj->XgUHy4cRD;G2jiBv-UCDDyD1glTDwlGcUZ}v
zWH#2}ILKdJ6)rc>BBjFJe8D9ksS`csl71G!=(WcjpHg;JzC7*5>s*kd@PsAXnjbXB
z_d6X6-Qu2$I;8+Ix0EmRGV@yDay;zCXjy)OzEH>tz&oWShT;w}aZbQ#M94(b(w$O{
z_UHACiB-@)pr7$|l>7FrF=?r)?_C-bou1`vWN+F#3#SRXJ1}{NOBmTkvmnX~Ha_O0
zs4;H+%v2(#8LzS<oBYcmlcS>MK60yO!AmCB#xB_%53(RbXMp9c#j!#o#6xp!u{a8Z
zX0x*#&}lR2=)B~eRY!k_#)j)WFxEuo;wrgA8Ynv|Nn5xmo@vb1nT_VRc-2Mijzrmi
zABW>G@^OEpr%PH+vx8s45QDO=fyrx1_bzqa%TeT09hEVMzcTJK5ccBeUI=19+Zzu{
zY@xgr<TgXZ=BGm{iI3Nllm4jn)W^%`=H#gbj50j#j!e5Y>{5?dYdcxRuSyl5_~D#>
zp$s_Zz;#G=1X6f%u<Od_?%hS2^PnWYMMF>;a%Ulb4U4CBgsfHt!{kN)n@dIsf2o@w
zEM%GMMJ1=KhCGy*#{Q!E$1)yPZ`rDpD<&w~rz17mg?0cd(NuT7IpC3(c$aPrT1Kc9
z*i1JnRH}=$G|?@9XbOCdRzQjdQG3TNX;GF3uZRkd#OE`2xJ;Gr();f^(MhVz-p<l%
zKxO|dejZ&Y;Y&KxOOT+#E9txb=_^XwrbE$wVydvoP5GnFk@Ym6QBfDwJ%aMRPI1i)
zoBWaq`>TfOFJ<N#h1MVt^N!oXO*)k`urd~kcnpm*O-WtGhE3P&Alu}`rbyQ~ou+f(
z##xx}3a5M;6L@Z~-LQ0mH5o|E?O}?$l~05me=d}AdfG1tt3jm&9-sGdgd&eMwPThZ
z3`WKw&vm~QlvXnAbl0Bw7Bqpan*p<0nHayO!pKIQ7Z=M*6GEd<?{O6OE9YSu$Rtte
zpWDRvt`u8hq7m9MR>yb-T^ygZv9DgOQM#DbNl0=s2)!&l?9v<1+-4q#)yUD|kk~vR
zH$JF}7QUu7POqykVBKDVc9t}tIwT?9V17asT3K8h5<e`Yq*5noOCz7t8z-RnC^dVR
zkO=Wzn%6v<LZ)yE1uH``ahzomfJKSZCtl&-V4gIrP_)S%Plnv?;J^?Zw47<T%?-%;
z(ck<OkeR5{tFS72jFBO+^G0qoM=!j04^kNPmeNIarFBDUVTdI|78V|Bo2md0*GZl%
z6HEy$@@PgC|5L5eYLhn778RbT3*CdlVSA#`6N0Yn3YFid&`h15a;j5+@VxqE0D=FK
zMu;%AvsHowD2)*)A%T1LR5M35*HK@aYCL-ZcPuNQ0>w^Xip|vpC(X|DFDPe9T`{#s
z3s9LltE-^wuCj=zsD5)a7v^0QATZ6(+>jT9`B&IsTfOgDOwj+pIdRNByln&x`k|%n
zuE|RYOG68*_0V~pB=W~gP>oZM$c(fE<t>x`<WuY`;1`uFLeSEQg>t;q!EeI68-b<Q
z+#g@+tZOGFa4@*;`EMah#e~*e+XzRK9Gs1*q(09g4#o{m8id=y7)QNoVQsJB!jx?&
zOY_>rBhol0xMI{A#MXsZ>P#??UkWdCbuxuLbr=vfkS~UJg21<ImvD!1=Nv5V4x#Kn
znbO+?Y*Q>1^%s%Z3~OfD%NNUMuh^4g*)R0#8@y@U$kE3+Hrw%fc2ca%9KEPw<$iZy
zL`1r@e9?8ETN!I<a)~eQ0EKDN>Gjd@v#48W23;v)y(nAC7{8AuV7+BGlHb$B8p|Tc
z{W;-pSHrB=d8>nJylI48u-1nU2o|wVg+_A(qU>CU)4h$aCBNVjPRdC0z`uJ|SMfUX
z@K71IIn50WzTf2@yBYE>??uxm_r$QQi;v(E5W#3{K4ro>AzOH{V-xhk#zj&0jM5Rt
z{c+8j37(Scn>?~$sz4(xT!+*1`)Tc_#?Bxt<kA);Ov<-McX#sSpfxa>JAi$a#MJ?i
z+*uDyxuft{u|6H5N%a0~<B2e~QhO=bO@W-(efQZ*y>FMG0i+_itB6YNSxs6<{e6{E
ze2QH2n5<>5M5Nz+M#GjHKYsib6?2SHigF~HJwK@}-w>l_D4A`K&N-?_is3q2W256Z
zwUDC;G^OWsAB{ZA1|^6A4_5-QmK)eFl#BFpKavJ1VV7<pbLTJj1vXsUY?5UtTMer#
zn*}#H;%l$gfHNuuwqk?gDs9AI$hO-P-Q~;9Qr}n5lC`yCy^|xP6lZc8L<l(~4Q)*O
zkry<|o@;7+aV@6I83Oyn;H0FKZ_K`{6E;!lJ?JuhjT;>AR^nwms%~iUI}GQU4}Umf
z+N|r4kl&ZyewkY$up?5;7_t2e;t@A-7`kLLI7x&sfb7S}2}Jf%-Rtn)C{$Dvx-Fg2
zrUJHr*6+L8f+c5)j?()d2$Z7+Zoh2vTKTD6uAszO!bg0j@^sO~_z+z9co73Q(D^>O
zMtUYo6Vqgn#i)~{+QVdhETZDvlq1kAzsXJE?=&z#Y89v(Lh2WBDXDPixK1U@>J{Dg
zA#Lc-0z|1VlMbiW+tMU5AByhhtz*9zp@V(o(|urgI=TBfzXn^%MG#4eE8i`vBw#`v
z|85H|$zx+9lP$)_o>ofaZD=q5Uhs`N<-)(kWwZSik17>ocL20`N!>yj7Fa3A6!o&T
za}`IC!awxC4!3xgA;7A@3q94QWK$CZtfhsWZiC!f$7VE8aVsu@3<i}=q1{?v$e4{{
zR6sC2Kj5*Z*jo6A9+?;_Y56KBUSlYFTr!Sn5=DGV1T{qaiOYcYwI&rG3<@L8KUizn
zaG9O|VA`v<cUZ@j(4v=3R+Bq)1)=N0u=-uEg*ZPb{pMs*_t|$c9RW$nXbT}A2sYP;
zf$ZivmS<dl5Y`hJQn+0DM~gmwu3&mPcaiT2)ebwnwI>!I+F<s+CB8*X%B<Hcnr+2v
zgs@X@d>#2ayB=P%7)AES*F~Xn-t@JBkqocSh;THx9|S}3mKs-18{0J={IwzPo?6b>
zUqs<5lzz&J-l5;>t3)*UA*3q<O0FAtw^8@fDE%R>{*;PPvMxz$B$35ZQ(#kV>4Ai>
z$b#DhrC%TN&9GmMBcR_BR9wl!3ttjrH?r*a@uu;yEDKTO>GYJMXRKWfJ$TGA<}q1_
zXjrM245G`wVhwPHKNaC8*N5_|noKgl((p%FHruq4*-=QsJ_vyS{AG>9F1giZHm+{P
zrd8Aa$&H*3C2ZBr5*1V4nen;<28sDBTI%Mk!_P4b;=$w!Y_g#q%O#&$_>nr*Ex`)x
zeZ{`sA+)&8S#C9Dtg$K#r0)ry;`DZIF)^2bd-mva>KLFA9vmR+p-1D^R@U2ortyg)
zYm5Km#YG+8e}Bcp;B_%*Wfh}hJLheumu2rL<~Qxd(W$@}ngcZDFk@x72bN<9s{<1u
zrf85M+zyY?r>-_%8pA$!>eK!6^x6Q7HyMYtiz?!Xt{Nwu=mRKajH?!}7J?VhFyi2{
zwA6lHGO?Q0WOx}v!S~!xMVL_EJ)?GYIeHXC82h7j@+pkWJ)H^ey1wpAki&d&=)JvM
zHxfa9AtY^_(!H*`Mr%CvF`T1*$Q;A@6~~`LT3bsG^8o&>#G~{%&)=p36T-mA7?1I7
zV!D)HnM5kedMr}oSXufT3I>vt5w+;&8C1PJO!jWy1lO}_SHfcn>TJ=Vh47X|BQ`RG
z9!t;}y>W!+Mpilx|M0P6!J_}>tRz{OF&Zc2GihNwBSBmrg)Hy2qjSX<vL1T@gXGaq
z$Cvd3+&?+WFZUp97IHiDy*~hq5}A8RKt-mndO80&pFZl#dN=B`kb@RIP&#OM+iviB
zr^k^8RZYS(rB21YMsv*0xom9H*ug59F>rEB*G5GdVk(7(CvLcE2far(UJAGrOA>w*
z1d?&sG~eHgM!?;ifEf_lbiOG3)*~+_0oab<KH0}BADP}CZoqET*@22=Ns^Xl#{j|u
zvWUW`qCjurCUtjnDL)LWl7lRVClI-;$7asc9yqN!?Mo+;z>$^2R2qjJ1fu2O{h27G
zK5Vji9fepuYy1V-u?_Jqafv8*yoWnu^-ZPbS=vC;hSWopSnC&ehr&~2vXszQBzNv-
zl_CC%(ZkGr<8K=TQh;JcQP8Z0rvOpyg8CB0nX)n0xWhPzT5jhqek(gLUhgLPnCVtG
zEr&SNcG;0w4t}IH`soPAgQA_YHQheLW%E60%m=Ya$$Eei^Lq}1pK}sU`V1SM?Z=aP
zI(NS>mZLvGwiN$w?tUH{A4_B40IJGst-sc$O@vU1@qM*X5NNCq!n#8Ekyrt4EKLNw
zZj#IjPX$PZWPPq>B}m_2fv5)3E3ymM`ZgrD8&0)L!r2WJ;TYNC!7-w6s3!7{oo0TF
z;@q6Z_7{ayt-x1|H+=Q5<Ym+R6HcYC_<p|G*%XIP{I**sgp>QLi+%3p`0WA<nquC*
zm0B{;PH)mF3(&4<rR0QxoDlf!>HWi7?~$AC%8i9DSd~{<{QPb&W-Ovn?H}uyqK@Pg
zs>e_+J0StZbXo~BUd-^xRa+9dw&$ep^wl>yCt)(js}eo!d;7utDKWJ5XG@<~6Ay>w
z^)bB?+R9hee^o|jwoO?WCP}5(Me4uvUJhUM6G#?QbpoZ?b8azoWA|?%f#|TdckW|v
zZpoU<7td{AxeySagjKw(F8tXDE?$nfqU|R^%O>qYQVb=-At)>#DmL!CpKe>50?}c#
zr+;1V-LFRPYPmExFj;>v_IdiQQl_}^vhyU(;{w2hkupTQRWvWbSuDR<E!LrT&AKO}
z6^d(qVd{^52|9rH(4v9Dj&Dr8q=)F-#Yf7HvS<D|I<Huj`m&P*9|2QiXS_f)ud>B*
z6Z!D0U+2i>RGw$e@O1f{*Zcg+`jMSz>Zubd73%j6k@|Yky41qT>GGKpvn*nqfRu}X
z#c;w8;!QE528!(wTl*Ms2xG@;SO(r7Mz>xk|9)h_MVgC*c|IXh773eX>bV@Y`AX5`
zb+GD}M>p-}4h$Y`Tx5J_o7IyVe|aDvk;7?qe}%Ig^=<CgI*Z>o*C@T*Wzjyr9ZWSG
zkdOk#OMiVD057TA!0SY<!n?078KkjS&ysj6y7!jt5o2uR*2BINx`1E~?(F{QGZ`>n
zEy&ohZ{e#uXuwjx-eIY6y`L_}%EyQ-__km0^>%%qOh5B25522fz`vbw@1%NqeYX5k
zn0)Z|1@64#BsPNY`1DW0aU>|p+Z3xS+aH+befY})s~g-ZxJhCVH60nk5IbhnQdRPB
zUXcFRqiSWv2wHtf9dY_)MU>j*A~l4++W-?;gSod$!XWjJhqTK`{yV`oV|qH%ulHQw
zb^aRoI0i6cKfT`0_&&z)UT4q_M_g#4IxQ3KIZO>4ytc5X4W<oX<qbjuLUUN7c7YBX
z-+?pc!KG@9laaLROn9DHW`_ed9AcAMlXthMYHi}D5zU8{SB!{}{&UUymDiVyk-2lt
zM@>J%zklY=?LHP}4}W%`&X3$aeZ0O?@Q27B{*gJ{u}adQBQ&f&GK*zaZSPT8t!_#U
zy<j${Th^dkCv>Kt-Z80*?sE^gX9k{RENjWSQKK2jj=#DP0K}EF_a;nHeH#6Gdq+OS
z8*x069+aDC0wb?qT&BKZPu2zodh8~9IH7&(!uW==7p#HcnK}Q}caBuwg9J2)SUV{D
zpXcT|cV4-k)ZJT$kQw^$SAWwDQm^*Y4RDD4<V{dbJc&Twc<1LkD+5me6Zfy&vFpe8
zCoCI}3%`rnYfxuy%yM7-D{U7_%qEbyxw~vcLFAw!6KE1S|MmI52z%?Gx`L)p7z^$Y
zf(F;%?(R--cXxLV7Tnz>xH|-gKnU*S;v_g+B)G%gJkPsZUwySz`^WuNPj}BbJw19)
z&GbYR9?LfDPK34Jc((h_-YNyyxXiZq&3LI+2IV?Fa=rQmp3?Nhe__Mh8j<EK>KDGM
z290g~od1F@4iFK^p<Hx63*uG{1ONm`acLd?UuA%TYla<LSJN{`Dj%wzILXA}b4rA(
z{oEsGi8RkW=T8j%UQv=w;0f8=(Nu|JuJb*`HE9Dmz5cCK!ShEndDaxcc_Zm&na4zo
zV_vVB+#nZ4@baoo|LJGTLRM3)gOk9+YgIoY<d5Ud5=Iv&FU@SK23miGcey%lxp&8a
zp=<zEGqHzhg2Zn1fqh=+(xQnbC)9c<c~g3H198ivBr*WRGj&K#q<(axIVW-9NauU#
zNw^~M;8^KX*@Nnf_j66DVEDo-@xWy<6HDcf4GU$#;GD6v3v5G!g#+ajrLzk7qpxtC
z@!GW^-i>|PmZe(1*5iN3n(0U2^??+4J!R>SPo0kJ4RV2Lz`1$2dO0zKo%^LclPXX@
zu@Y)=<D#R@r~XrN4_qC#?bxsP9KPiI+O^6uZvY9_F7h(u+;3#-QmCz3Fd^Y$$Zr6T
zeQQ6YBtRl@b$oy6woK|9w_ADdbDWNyepmBLg<7OQzzpf;B7CmPO+MB426E@nOp`GQ
z))LZ>gC($nvTL_p;B60g*9YN!u09!U@N50tSrAf|?~Pr|rO%mcI#MEST>ICDi?unh
zyPx#_lA7?kH-CX+K%h^>a^*QVu|rp(A`p<_TGv$da(@`<Chgzg3os(R-)lNICCmKT
z7a5{^*(ajy>FR61<>|L}XBGhLngnV)8P&lq2=32H8x8JslvuiMT3eN-cz<j`ldf5N
zIldne-Kw&$iIx2Ea-1G~^DSp&DbTWH@Hl34rly4Eb&#?|AN=rJcJ1>;fU3iFqt5(W
z*YVee$8X$_Yiv4MFAuchK|TD(Q#*alYGr+#pdP;)iY;yK)wM4_38iCz>FTmv?cUA5
zzhCEEEv1$rWn#X|VyQ|!BFK9C{rkzgLD9*bvOW=JvW7+}fc5MK4b+vo{}?eVVX#~p
zb<t$mPVMsfTw7-BvJ;0qks{^qH$SeD72Peg--i<2hCev6KAs3tSb&KNuZ|Cq-e##w
z<a~s^7~)zdpO^EJol;2w2ae0ft1w6DXs>Ajfs6~GHU!U!C^o^g;X{&)r=T{!-ly*R
z=a5Jp0wI!diKH&lyNwHfquzN{(ixKcI#LRY0FHXWtmj9}b&0a28VbjX&=R@Dr&H}G
zS<k5k2XqYJ`7(WGtYG@5@)zz^mb9Ns$y#f-1g$leE5d7Dfi1fLq%PgJ2Rj`XeM<A4
zS4Z*vT%3rMY6>~*OOQ^d$zSu#9WZhh(pg0Imt>5m&HMX}S&v=cMmv26qSPEa45Y4(
z>1Zap&AfL8z6BIr9aqv;3Gl>+HoX>gQeIqC(QOEeQCRpB*;ktOLk^kv_H*}{x(U`N
zO!EkQBGkb|4G;vl9_<32xgU=EIw|S$v_^jQ@Xxc{GFiVP7g6AM+V2H!M>a*dV{|%b
zrSEK3w}c+K9V35@-TD6H5ar5IbOKd3oPuwgR#&b$!KbSj^1t#e2G?4`22ChV@`!s{
z4Py{k95aJVnC#*@>>2A$DxmV-k5A(bUVJqq3~l;et=SFKd;U!4W*<WT^z373W@~xy
zZa|YxfEKFm{^Z(=>2Xy$<FG9K#n(u01vn@>)rt@if503)Bm3-S`JuO!B$A57u>?xi
zj*wP`+cNXKu(ydm>eHfOVQ5FC*K)mBco&$lQLFOlV@NBV@H!-Y!_^Gd^Guw7A(X5X
zxJdQSJ_AlBvFZgd$7#MZN3>+NqconBomLC@+X|d+DAk?8pq1jTI=uy_im&N@NQ00Y
zT^g0AwJ72g(T7)rZij4^ztwD><lYf~X$J#ESX<Yt^-ZnqVW!%b?fIUukjF)Bm!0Yr
z``8N9Ga9Wb&u^FA#KRq?ev*E&SnvqjejG0YmN=KWgo5b>IcZ&@GfLLDfB+kwg%J7T
z!dyI9x-We-cqX3@yFIF2#mAWQLS$9c7kV$W7u^l)%Lm)z_FGRk`0b2Vr{Sj*);bsY
zA_F7O$nt$o3|8~p+@17yRmO+~$Y6F<={gPVy*{3Tx*8lyVPIMf6QG<5MjAw-T~3sN
zL}5X-r`}tMQ{t80Rznd_mb%kBKih8o(kGM+fzd<CH<SL3?(N-lB`})Z)2DKh!(Der
zFTWELPU6?u&&$Am0V`4Z(4X3e*0*UU6&`66sQhlBMNhQmJj;I24^E<rciw=|+N)7&
z*||HnF87U-@g(zOXz!Eh4vaO`&qiAS9k^kMP;$RF55u4oi7PkB-ADA=i0s*#m=|C9
z2OEW;0LMZO1(Y?@775$CqA$P890x#KOeL;!Uh!o=lo;O)U~N2i*pEdsKb<!>I8MAH
z77|2-c9H!0gp$hqWhQY`2N$!<|Bg@K$Vy=5wxF%iI26t`tZ2}<eXAF{1Xwq}ljAKm
z2HhTH!x3R3&rvOBv*f@90+DjFcLz<at?r~`48qe?WYvyQ0#IKMJ-Q9eZEX)=?p%~i
z`!mbRzP(Lf^yRyzQU>O3BQm?J86LjwQtPg)ak12$^k2<$n@<N#w)a!`+j@3$`P)7@
zS+F>!mXbr-*ItM;28JROKq&Og@`|TBIU_v4gR?2(Z?5h=I+@F{K>B5=EpTvy9ONhg
zrRMkn^G9~7fN9K5O*bP6Q@})!^L5*cH;wp4LpGu#r5zjA<Es(R+d2|2A?e9?T+H|0
z?mznuHaM`JH~MZaM!`A#Dvc~1T{><HJ*J$kD~fn`1pBG}kjADC1T0iK6=8wlImJE0
zuh%P1mJj<<m(=D0X$Ph297OEV@tpsD;*^Qzu{QtfU2t4azZMnltj@;0YW}G{IsxSU
zt-t~Kzh;+IZYaA5<b0DTHo?6afR<LncxQ*(=KD8PO=IndN0_l~VTP{ybeKKH&bFWD
zp+sX%Bkt?!G(Z&CO0jVO=_YW*)G*!3TrqlEjaOVxHr!nMNiu72sp>PN2`x4btwcDl
z!CnqTm+>RgFU#jjeeG`vKhAz{F@>68JoPfRJG*Tdzb!y~!svB#Kmdx1Z|A^`kM~<`
zXlm`_bNSk&wj~+J$=&trr9yLq@cY+{QPyJo@V1i~AX!qmdE1eN7-$?7J$vVGI1_Zz
z?;cw|b?KH>?TAbr`gOkPA^=s{Wb68*vQ)^+%XMz3={z#lUCVIMbBfEs^DW<B_nc_k
zt*PYv+~#bj!`R&RpnRR!ehs$tHMl0Iy%$DUwkGJM3opRYTm;xQ#rP>Ti`9GAVZQ>p
z_1`D}EDy+Mv6z5%d&m%Xelvf3D{zOk)@Y|_v>KF`n%QnEaHpUI0wsKOu%qdL494YS
zl7~+?GFGF)Vh7=QPpJ85&4pXLf=N?FQDX#T0>!EBBHZLh_cmL3GtBAo-5XdfEw0?u
z9LF)=gMwZ<zX?6_cD+V)^$#hmB`V&|->O?4m0y`!8}dBHJXOJwlx_TVN$QAv!B&XP
z^@=H?G+Zl3;Fs@a+Ga+kKRYpy9$SAE%xEXV!174jo-Qw<?0e}oyqfpoW`YcE!go|r
zPt+`q94;mOh?r}+oLrE0fs^|R5S6{p3>l`udB?`d&dZ&tFee?2#LCI3N6(_<Y>_!Q
zCyfIONoHQnNn1khsfENI!zbY3V3D?UbN3+Q<@!ID)UvU2^0B9y(ozBOffMzPd~WLo
z#;y@|7|Kk2DcQlQ!KPD;`Vo%tgoJ5zvC_~l&DGt|Z@mPA>fkejK{Ow?b7|&}DcOoC
zgS(XS4P`+Fq?RvRA`XN=eB%7bt1an`UB1YHOn%u#1<;E3)UaJ*jCZdKiBQ{pW&MEP
zi~d?ms#X8$eZQiGWCYNklD4e9@(K5}zF6P>=^G!MdB76ViR|h#v~1q(lJ8o-@2*2m
zV?zNN0MJ(BM@LKjAvCG>^k&_YYt<5CftTvFZ|%sA>Ffc`?T+uATFwKsIv0*#dtYx{
z*$Vq!8FDxZjWPm{P15~eDr?*l-2C<XcfN-90=3nSZIAalmH<!2TBf5vn6ZOpC&5hr
z4D9U?sI7F|Oi;&BzXa8OMO<TWZaJad*~A#;Ci7$iT3RtFY1k6hd1Dh^{p|2o7uc$a
z2yU_4)O#9tZ?j{ud~>YbY65rKEsd=E>OI(Sdll3?tU9!OU0<S@pb${e-y3PWYSnb|
zDNe9*NpBE50A!5go35@gIA1bYyzM~s2L`W;0SBFWop+sA2F6c^Vu6|I8AR{E98Z`1
zZfu21If2>nx4Oz38zZm1<MF~eg-4`2p|CsNwObQK_otwrbS|dPMhONZx2b!bCByi1
zptYZs#==K|FA;e6LVwqG&R_qK2HiJ^y!=^%1W$Pq0p6~=oeL)v`*m009#?_4d(B^T
zEmU9yc-p{pv~f57uv!B-Zu*HM;^v%%ca}Sv?I}zfpV|A9YRgV~MB7(SHQ|>YWi@Z5
zvuO!5^_kunO6)ieWR|pId&Vns4m>ob+zB+8a`}pKsJTL`rK;sc+gzz@;53L<#^>l1
zG?6N~0lR68>i7&xRma-FKbqY13diseX#n(6L~lm7JX3G)uW!bxzpCE)U;nCzy!|=o
zr~1&N{sK-E5e{rMZXD0z+LNg%9Xb3^mqJ)#+$wJu7c|E(mlCHHokv=0rakTaIb&gv
zLCcL_=Z76b$pE}7HI_z8MzwZPPJF3Th@F)#0Keq0Pf?WPTWu6#*OEpRX4kU+T(mkl
zrB$>_{NP-?+NvR4X2%g-w0d3Y#FR$u`AH);*-~|?L8EsS;+CS-oU&BZYTv0naV}>*
zkqX8d=%oM*9t@tc4?!wkmkXLleuiU+XwohU=}xTQzvIZkQ~G#~!Ba|lwLw6ueh@&M
zOL^kLC--AE|C|9vpd23EgSVtNjI^U(!7%hAFV<i*2tYIDBhf%nk@Bd<dvMk!*cm^*
zh1T_2a7t>k2o<JRusTE6oq2d?kuk|$q<2=iBi@`Hga4&DZKJaP&vl+fe6vQn*iv&B
zY;`(kZG#$a^T{-XT;GO(7oiehx^iBnY{YL8ma;Km61Fac%1yVT0Z)`BxQ*|idKqQj
zC^pqa;Gl|ws1rtP!Br84W;d|}0Vnw|wBwQ>)w+GJT-7?-rTrfkRnasWu!?@^Vr@Ax
zTqY%7Qd5zF<1-jFF_vvwQ?A17nEW{!=WJH42c%1=Sr;HNYl-MZA^|&J2Y7BmCev(m
zht~Zk#xOgSIVu9p*gR6ri7<%%p)VOkJn+)MUipjX-p7wV@2`JzYA%daeB33!>f)iV
z1_~iLJpwKl`(LoAnmM0m${=FRUTUpc##I=WpVvhZ7P}Et3?Zco6hwNH4XIyB3_ejc
zg<HD;Xx#KX$lmJDh6?Faq0JLvDX8Q1H1h0p-QjZZmOm!}6ib<7MA+Xp76Y762@!^!
zsC4$WIrvaAL)dU65d6Gm=3`i0W&tIGqC<lMMT|!Dpi-QpTg^T&*SPo-VPKAR#!{3k
zwj}lYyM9^$-Xo|uLXJOa(D-c@oe`|x6=9#C;`EwJ0mCm?ZvIfxOeeXPUz$r1O&0CO
zH5g46^~kLW;#hIQNrTnQ%U#_ys`ur%1}gv98bBCBIQJwFY*E1`rsm{@O29LiX%^S5
z{x~pqIa&%oF!y(TA7UTQ<3C%44iJ_o>MSAVM5S0N>#}n|h;U)}K{Y~!f4o|N5Q3bH
zXu&1a3d+Is0pXSe8nN*Nq)zg*&jlyD<>6`s1sIbLUY41|7Pi_si3nB%6U!zhvu&sy
z@TMAiztK^(?H|&bB}6(tF&!eT2aTmb|6%dc&F~zx%0L@+EBvUz<i=mAHLvFcC9OLq
zgo(32IG7T5NY{*5ogV%feioS1dsl82l2B{06#PF>T4e)8qANGx`wwN^AUIVG4S`<l
z!%@sX0m;xL>4IAS82KL#N9|_VE48k|VLU868;}Z|kpFA-fSOjz_W`vGwq>Y<=D0d|
zm$bH%t^u`QZj{h6x(X;CzlkxeXhGzzivU!Dp2AHYy7Zwh9$Z56>nKL`ji3V*<ibQ<
zwBK}KUqQ$N3S_Jmm41<WhS*W6NQ=spmEXSOZ;q3NpG}T85GtaaU@M09u)x5C03Sa`
zZ~~)@*+0?xrw|CK4k2}&<ccU;EEr(<<x%Xf$Moh=ZA4|PY?JLE2plxwGT#fW49cPL
zjx)x?!xLM!?53DzWREDpCYa&AX~MVc+QdN1tSFK#2A42KDx>m_3o=QwBGv^J;=(7e
zH&Bn6z@7hVd9-m318VyPkg<s}6MBh440dSPA{p-;yoW{dh&C*Lv(F+ywNxlQzJk4+
zdL%Rn({;ux?%^mvp@anQ!P)TX%`vQYxBoP$4BcE*hJ%G1?a;=^md{F5=4zY?$%B(+
z1|Ui*tHM_Q{+Y=V!J{=LKZ!p7zg<yg%56`a!LllJK&@pF;z>GNPS)R*y7TiB!6jC5
z_AIaoIhOm+CgCa4u&@bwzj>D4@m5>hLAZQ&CIf(43(d7XBxm_@8MMsQY71)hUS)`>
zS<Z5{@4t~G5b@CFmrQ!G-h?om2d93HC|f!U(`PJEnU=Dq#oz?I2$f?PlBo<Wc1R(v
z>f@nFrV0>Xi%M4PGsN0@qPTwtuRuQBS9?xCh7TMVSc)p+(Y>Q`m+t~gV+w2*4HvAy
z5(ecmF;X;9*DEnn7yjeq<P$#phnx~?0*9nKM%46zonizBRtm_t$4HI_8<&$2iTJ%T
zkHloLwH1bSpwf0_2)e||sFD!EaKM{X<Y4)P_S<Hkx)hptst27jxP;IB3I;;%b3jOS
zxKYG}%X<tLP9bPzSYExFAB<suT|ovCJcOeX0wuH1QUOg|x58WzW~T<}HF!u&_ycos
z8E<&iKj61jc-jP-6D?>(sN>|>upp)+V|GJJPxVZ*4^(nVRop_wT@pud576~r=3(IP
zY45saVhR3?LSr3QScPUCpaT@?gVhRNJYb~DgsSB5(-}peO9T^t0XVGV_Tukxu2Mjp
zn##eH4fHl#&<TPV4LH9?_1*p=x$N?p@&6wN;W6z%a)-sPoJ$(nAsp`|w>!kIf)ho0
z6B#hPdd(Ji!IU$(+mNhRR(B9_j~pcH^Cy=Fk9h%;7!3r{d{MQ{mi&;fg3jjVJyl%L
zVMqY(o{;Ay67HVw2Y(o;y1rb?#=&pRsNDFr%Xy+|!fxs?QoiVZ%0+afOc1Y7*45Ig
zrX{wD4>fjbC`>^^vXvHQLH0X%?!$5GHRq9Nv^X%BP|o=G;w+C|#j%FOuj@goe`S=t
z(rV=;IDQ@5==-5D2YSwcw3BUX_JFi7&;XfoI6B}PlV?tn2c!$M8F7|v#!}=$;sb7R
zSj%uAv$&gj5nUmrJq)Gd1o9A~Z_sxS?JKW2xR}eH2|<MTBt!36#7;1*l<%fx0c2vy
zKt;seGjij{CqHLbDXOHNxCyiKJ@ADTBf-_|SACl$md1qX48*Sm0(=9>PrxLgu+)q~
zpg95Ct=2YrRHSD#u0%$hcp?!D|Irs_fVIt{iWy3V%d|O~Z{!M9h>h2OGo}E)R<POb
za4$L24QGS0KE)EY*=u#8;=jgn${KVUi>K0Q*B47z>84tSkFFKUYT3tt6$dH@kOsAN
zLm0c^O*9q-qhO__6aYFC7-=bVZ>51s8`~*C=#sB{3=sZ=Av~*Hn;OJ-HwMfDYEzKv
zjVU;=2`*_HRndgzjUhhEfs#oE;p6<KyS%7Ea>3<WhcugQHUenkALi^3<z#=Bfw`34
zjMyRThu!z9s0W8Bh~s)=?2tJW=@gPR(hq?*=IRy)FoX;uVX6Ek8oO9)&?OXw%dP{J
zSsWE%x>kDTye1kx$jkqb8^uH;VGN?>!;HtuAczJ=$A4eY{*4$y585Nd#q`MR1gj+@
zkV8ltzKv4Gs*q@U3UYM>e#p)2MGcV97XXBmWg~`Q6yp$iB^N8TI`l^-(=mWv;0sfx
zAe|zLCW5%$_eEV2Oj{%zJWrESw>=Vc1X;cBxF<hm)!awFhooqA<t)5MZXqE`XY_}@
z`>XS~bLT;u=*TIP5bjgLciA;eG^Hk}9lcMNTqEheh|A{kk+}C*f-;A)L50=sSB4ea
zf$fhPE9d=--cLUh@({M_19pc2|F!QL6Ux+ANI2&j%~pZ@=>fc^6)A~UV2w-Nhvn~S
z2Sq8mlGOToA@@ZoO!_v9Op409l@`Lu+hOwD6M%3#>)V4Cd&k4F{C+)mdy4NQiT+qC
z=SYv%l^P)rL-q_)cRJH`Dn`c>)cg7TfMuJ6p1~by_v>}sUmdI~8#W0*&!u|_pPR<C
zqm!u{9E7T!j^7`(&SEPAp&|r0xtEqD&NwEVgr~~yS1r<}6Z?i{X}D1>gHz=baBu(0
zQD6Du6l$;+&fdsT%Q$VyQ8Ohr3}|9_dotMi&bl)^WF_az?*~$QLo8YCQkTo+Q(>^}
z%uuGb{`favep)b=A4t#h4e(^(TNpvLP#DwjQXS({)~~ysJwH+wCg%J(sf5M(6BY+<
zD!C-^BWGmAO(nwAA6w_2Vf&*s_!v1u>Z2PF5{!e9L;3j#(^3$@d@&;e{LQ9Fwvmy!
zI)7$-M5=A);y^<nC1m^S>Ys<1Vz04-AlRrHlh5+s<xD5&f&A~uD9Fg}M_=UiuniS6
z`Bk92(o+&M;9Wr2RxQvzOuAFRCcWYr-WT}{&KapDCPf-%4R(Z;uc~9j{ZZo7{mO)b
zhP-EWu%!auTrjK3j0n^BD_1ho_eW9E$h1Cj{Xh%Vsk1O48n!D-e?H)pvDcM#-z8v+
zT9Th&lB;bG1~irB+*@1mFwELGYZ03C#)yQ(YPBv&av_-?#Fa2Z(hDLc$zVQhTo$U2
z73ilRo~*n?4ajfCB2hH5&u=qF4UEg~*kFcOqF{x*pOD86Ne`%^_@`h4QTW2H{)G8o
z_Iv*{Womagh`tuas5k>(qD<4y$1lkz!0z{OJ4-8oT0>!(4RJ|cm<kh;-j~O&uKSp+
zK2PY|iZ0%^&4gl$253b68nr$RRju3AM-{8oa^t0k=$B@RVOCS;eF(Rso+l=NU?CbJ
zOe5F#UE(T>g!k{Db;iK0WKcDrbalcnD;<afJ~P*vhqyGH#Be5wS2bc#z`m{ro8d=x
zq=Dl-T|{d8%|Q&9#+<}jNKU3s1kB*29zIeyleXPW87M8V=c;E>WRQyYrKxzE-c_(D
zWaSqmxPO<E?1NdoloP24;kq*+E!in8Qz0(d3Hy?U+i{V;LlJMwnOQ_AUahupgA;FC
zf_9f@Sov56ai;=;1_1VLCex_lBgBKe#d->9z)mG32V(F;tv2N4$&D`mO!$cnXTn7a
z53$Vh)_^~qIB~veSN7%kApXjTTIBqQGn!6e>Gi=`WABEA+?GD;iRal+xcH7||AyW8
z^~qQ`dRLiQ;0_NiRglCB2@B{ly43v@D3+iw+nIh3Q2ZWQ0vEd2|1^S&LjmaDCOY0;
zG7I0Xf4&7Hud2hTWc9=Beqbw!=`TR>N$056!XU-hd*bT4cnrB{JAMoiYgb>%PwuWJ
zg%{y`F?pW78QP63CdJnwjyz`XvU&V~%JxcV@0-3rQJ(RF%x#Bdj3rZds=a)X^qy%u
zvKBuW=vmYdK=!Sw@}g@d9ybweCTTWPZMM}~@8Eg~YxrAQba)C*OzV{B2{IR5SdteV
z2Zyy|&R!CJM1Q|AH>GfT_kP<U^0QZ73rWn%;B<X`BV_?dHFMK+20UB*O+s)9@}1W&
zzb}o%Vd=78yg2jM!>92oiSIY>j|K0~6TE*(05cn6Gf&qy;7pQ-)}c(2SJbwtgpk`X
z9QRmXJ0k%?zrwb_m(N~lGWuPBl-44*0IwgZkKfC@x9rYe9&-d5bmLq8C;r{`4{MQX
zcILN4PQ5|o8oesQeF1-_uM)Ha&dn3gQWyG9ZHXo`>wh>)0$VUXFU{KL#Sxp!eeWHo
z0kpqsrcngOHQHPU5sww+n?8`!CJQH?MtqBoEVnuPBO7P4J-37d=wUZTh|7!o=S#sP
z(v`hbjYEk2pm$fZo}IjOcYNI-miGS2g4;HB!NZ3(!9`_D|08dF<W9qnIiU5}0f`#O
zpBD$cgU-ybNacZaNj}f^PDhS5&;9l(U{K_sT}i#LV2%4lwZkztsB#Vd*K1#QF^16V
zwj;1J{bWkEZl4(RH=!5m2mz{5n|ue#;`Z}yf>IBppdCXNtskiG#_~WcaHm(Sjc0q&
z+NVqY@ou4({>5(nRDUnOgzQNaZYpHZMY31|YNX4tnNNbF;7o|MD(9Ap#HzrN(ZsOn
z$udd~@BOOL((oYYcAf@(vk%0>w(DGUu$W!Mx>&0ujoNROFn_JEVvG)`-+f@p`*Yw`
zHO8ydz*2XZ+0?J`V)isq?38Nk=fAe}%rhh616s)Pvh-@C-dQ=F4^qcp6vi)da9{ny
z(#{9j>|&X#UaUvImRDSE+6r{1``wGXe;g1sUH5+w1+lmTbLCcWJ?UYN`0&n=y8W!*
z11kxJ*6SvRGMoEtBshL=#7uX7Q9HlcQx$ep_he>-y9CF!qlX04AP1sAu%c^@dkJH2
zmb>ow1N?|*D-%S{(rN{FSx3@^8xqj1BG(&a`xLtxA==j_L4*m6b_9{QwY-$|fF0)7
za(d4l;oj;<AJGQ^@QX?QvhkujL6`_CCfpmCWw8Y5RbRx1bMWo)tE1@Ot+TrhqyE`~
zyXk9irsIsn7WKXJ4*z0&dnf6T-;4NKT&i&El<J8teljS#5_7Dhn70^${_0ORD30cl
zwxP0pE2bv2K#H>$K1%fNn&34z#RIoW=N$K)?HY)sb?SO(YnD4wN6wV&?YFJu<tgU-
zPm-5RkAF}caR_hdM>q-*v^xU-sxwItZ*9M%#J?O|d~kjuYyx734&_d^8iw#8h+&!}
z4xuGjb`wG(wvn5Fesm*+e?mm5lP#4Nwi!i$*pGIu2p)$aISE%aI=pqr481z?9>%I@
z)w6i~!MfVs^tQ@S0C~S)wawmqU)PNdYd?S0v$}FUarT{#C89B`W3APP8q@DphIPMe
zBIxb7a!l=nv*x<se-lm>)vQT#*vP`ADv}V!gm<%a?w_aG3$qS^xO&Xr#0p1`wmeAz
zRAY7Xo-DoN+27nW8GePde#t{_ktR4>h60iV_cklDOHjtogsUk5@9Yo14bzrJOzQ>!
zwl))2a$k#moMvY<6*=gVe#i3Op+vH{<%s<uG}a*?q}aZqDn}puAfWhDL)hE2D{e<0
zt6U6TxX^?@>*6#-0&dm@UM;iZPARMt=$UmYcgy;NDZQ&{r`~embwQ3dT+@li5J^yu
zF46<$doOkQbr-ZbUDbcRCHT1?%+%xHqvNZ*s&qUA_KO@vIq#aPU(pyZ^Y!($v(rf7
zNh~)rHyYs^PV`B&JgyzGCsVg3*^J9t?e?iNu{(#D*b|R%g;5k?y@U~lf#C<J#1CD0
zTrWx?CnBl5ixkuq;9?vbs*K#6R?kMbI}jVGQcbabD-Sgmjj9TRQu(g5k@n|LQ<WNP
z!B@9<gUjE^_QO%uo(4POf^FOUS$~TnKMC)wAOd}8nF3<Uul3!3bp}m`QUxxUTfRqR
zskpmI)$LcUXWc-GywE*vt9%47ravh&Z|Zg)Y;OytKcRKO>JspJ>uPDh#~bGKvfoz`
zWu%9Qpdqpc6xHbvTqB2QrO)Li5%jRTVJThNrp4+R1lrX$xb2l)^o(mpN?$Qtw2WLb
zU^!=v+rsN=vuSO!!)j4%x}-0l&R2ZGa;|KFzGzo~fqNg-ab%?R4l=?S$L6V9N8@sd
zUh?FZ!B95jwT2^ghU+r0x-Gn{BFC*W-&{!#MSmX{J;j#P2uWUhVLp!S`OJKZ61N^x
z+s43pnW7*7L)q~a4Jj@l>O|9Iml|}f+<E+t12;rCF1NFmmjTN8lN}e-CUxc};=o#r
ztzV8<2433&YW-3%fkiV=nQx`~cETA(HikT~R?YYxqFw`&j!(6SL;F6sL$i$rlE&7o
zgX*cXb4K{>xTLX3-@&u-C7+BI02$&N&6QcblXTKVfDzr*TAB@uP?vzENZtEQjv(np
zLIa(wug-%Y>D2Yk06IArzMg^1t(!%uXH;Ns-XwUmuBsS18PLAnE`-oW4#-jKU0yPZ
zw{>7flwmOcBkV;x^Fay&rt?BC7<qWjgQo%JqP;$q9>PfXgm|vR&M!+0RQAu4U9Aw2
z>+C1G%z$7YLl{8_5A76Bw8v3#s{Awa67;s?t(TW6?daH+=LeJB^wbQfA>AcHg1wxN
z5x_#a)jMgY@(B2UfKI5~3=JV}5u}~ERP;lVGYtsIY<P9Ee-#lxZlvQ-R8rxv;X%$3
zvzAf3Q&J()Alsw{_P6d;3uIXctz+;<vf01zkZqPvY{Cz(U8{=2Zaa24IEeNrC1P?y
zR6%y%d1EGtFrC#aSA%_WNsp0sfNycW>$q;}7e98Dc3Hz_yw%veLh_!KtZfHIBcuFa
zrJayph*;g!qqc&R#q25F@Rhy){!3481}E!l)M2202#x+`UH+5lAM+5Rr0?;s%%=)*
z#K$7TqvU@=rQ?XPnQbI{OmhmXTp;!%M%YRAkgE&uAR2yu7Y{`7P*pFEzPK3XD3uvZ
zilQS5nJtgP7R!E1MiEbTp(K^WqFNY5QmY;xL{ZDxhjohj`VtzY{gpDF7;A5=BKg0)
zEJ^q4DC6=dvznqp`JV`_v<MU>m7ybE*f_<6rV8SSnpMN#<nGWl__*~N!Zdu{NFkG5
zBTYjSi)M2`XpdE5$@D#bCpZN3PaYf+F;-BkuEIA(){W^ni^?B^mP5`dG&#HY7VTq+
z3~^@fdnF^f4{dP{TA`vj$#d1Vula6MfNK2%(cZQuM0|7zasGQ-o)Q9!sv)L8!2c{h
zpR6nhq)&57v)rh(szf&T0VV79>l9qMlQ_WL2tnz_{hR_68j_5@tifR?;}a7G+W3qU
z`o9B1rpz0PriObKC<}6?91&GSPQMywuXBwQ_HxY>_DWagjg;`3OsnyZ+7-bn<4pOU
zh(4b4(Nl%gb=}!Nc8Ra8K5*oFjIvrG(~q)SNiATc=-a;TF3oR>kFYJU<<^>a0Yk0X
zKl>7T3%Vv0D<qSjl>l<%Y^d8QT^i-?qfrlYd-W9h){RyRW+Io|oXOnrv5Q4c;8JkR
zq}g`-TJT3XgSw4WSF_*ybmvXlvB*}-%$Tt^Dp}iw%%d;KN~U<lms9U7DJlr4kfr!r
zT%$jv%sUuhds5b|q}U7bh@@qu0zJfpYwNhyp|i5?+}F?Vs-`1G1m(QBaXXVQ9dI3`
zE<JH|i<2H5#-t1{y>O3CN=5ZAz1<b`I>;`%>&5#pgk)roTZ@CEX!|?}#LZgOsA)vq
zTGcEMsl-+**)x);92e^!oRw+VU>`)Uf0H3CL`<JhDe1dSMG`Bm%8g;n10F7ioMAq&
zzT!EPs#Kqe_|;Ms9S>JXu&d+C12cQbr!<r~WHn4uYmUPEk$!dMDXvo4Z)mWvb5((0
z&LZ0RkwDj&Q#6L)PzCM_y~3OqE9aEs)ZjW4qZKL#{y)wX?Pup14<~SZ58Y?tkL~pv
zUvQlaf2HQ;D>bbAtt>gy030*g$$k7d+R&mIo!i@s!}ti+A`+Y|92dxTtel9B3oRg>
zwTWPTWHCVY%U%w~ShB};UMbdl5%Q8X4S=jB_iW$UiU%Z?b>dX7E=;h=x!S6`X&{5Z
z-s=lGP8F4_uwd^y>btNCT%aStX!=-g)ZE{h0AmrnpwN>0PC1Mh$VF0~adP73jeqAT
z{mn->^ndp#nX_QmQR)xa-ph=x#XH#9+u}u3Gxv75)*K^W!h?A1^FGRMOu7~i={p?L
zFl9ujhs)u0s`4Z%O;c@9bBm>NQt_!XlW3dca0!8j9GRM65;?$LTm<`CG~|~`kcR8P
zUqSEonLirvPH~;SE=SNkIALeX9Ldfp7;&8{g#02kRGKFVa?7=w$zx+8evQgnOTkoI
zlyPiW&~W4j_E!5GGZwVdLK>_!d0~EfJr{**8aLE~P7F5`pYAn3k}kM5oG?oAv)wl(
z7)CD|83m@CywcCudX%a8M`+m1E4p<#7`7!Am;g=9MoR8zyoFXyfuNq@6>a7(RY=z_
zIkxIbrO$Bi)>7Kz3`5<w=;c7GOMaNX4y9W;S7hN&fu7;qmBtfXsNu5s-r`kByVBY!
zz|C$q=I7k9Jys0GzN4G^g4VpF3$nYNYP@lBB6u*NPU#0T=l2*iQ;~HP91Xf+XJl_S
z04tNey|zkY3uR$U1lN8G&tcET6tt-Y;m16=Mxae)>0V}ckk!&c*iNpl3PzN(myi%y
zAHi*emrU~s+Nrd~DiK0ko*Nr2!5yxd(=VYm;q1F>_%o}mg{)<(-|CWGy7wcC&$jSf
zL@W2#J~vjoM5`oY;}wI=Q<K)%U7IHX0HNk~m3_RCLx!-rGnsnfjdu&m|BR_yGm4)U
z=|Cw0duE92_t|*~Ym7nVJ0G*3%WH$VbGH_uL*s(fGx!BFf{q--too~3&{tlr%U`YT
z7yKV{_f*c;oy$cw?l&C(cN?y;H||!09I@_q*>BfguU2mh?{QTAd@@^S1cKGYu(F{H
zWuTj(kLUWP5G$=WxIY!uJdB}MVtjD)5(UkLM}gY(F)z9vMX(xB%xLwC915<c2XAh1
z*}FyMU&<RCa>RtB8>`@N*3W6MJorT0qv1B%3R=ri<#Sb4d4##EiXzKtpEk>m#s+A!
zEOAint`of3J#?F*;km}!fH8Opis@Vz)*P!Xx)1Kye<QAcuw05|oAx6jmL&!yV|qcR
zEYv#MsUl^mE~~of5A?+r{n1F4f5hFyQFOm06}gO+@x%4_5nRMp{++i=scp0gl-1E`
z6Q*LS)nkrES?jLE_y4yN?8l_A_jeqZ(A{%pm*^e)hJ@x>)+vVuun<u4G!+Y9wvM5(
zoO)65_E^NMQblHZ{cK<P*j7L#u;?TiRePLr-Dtx@aba4OQ_z#3w2-={7D}HBrv?hi
zMO2$1a%La?dZr>Sxe*H4Mk+NWVKz)LTgcOR&3!^6TMGlnEoY|5L#y_+o|Z@VADEgg
zsYRDYKVpJtl!3*YT^g0e5|Uierd#hdNNhj+e~WAj!&-LGMMTbsI(OCAKHQGz_cu+A
zO7%LPDHMZd@56iK@M_Stsq||34U!$yjI&q5V;|GTX$%*6#JDfa7H+A<@zy($@I4Q#
zQdy5zqLhX!^_g?y)H)2f$;~;^F~-}3Im1rG9Nb3A0i9sDc)xzXRL*0++&c^w-2;c_
zk@DHd@OUQ9K<fC)aiTrFlX8;x9QWlPmjYx(^dRBdl74mcT9y)-%0kxHNv}=dMj2$;
z!1rNyOe}38NFQ&q9%y_QQP!#a^1ALMn}4n>&58}tOQR4f-ww$=^F}ff%f0g6G$J99
z9VF$e0V$%><Yd-3V#4-gYh{_JD^%fm66J6y;>x`^N@k<fpcA=mel7sJQ{I|fZkM=z
zZ=Sb)&*`)FROE0yJ54gDTMi?QZ}x49vgGAxy1^m?m0$X_IQCeTcfA}H-%YsS7*$0<
z8hP<<O22<wv>{~Cx}Ym7!N+wX$}nCPxhwY<;JINH%%-UrD;-)J@f(VbgX|Iep6cng
zN*+l4REHl$yy=HmfSUG?`NfSXrECK!*;<n?+kKCWkbu6TZ$?^cysD$gx3jCV8|&}!
zb|GrIdf_Rh!ro#LR5n8e^Xr%vpQmwJT)Mh5M{1kfSlJSi&kY{?%BM_&D<pO}1tDPU
zZ(ZFoPjTb897AoL>_E-nyJkIQH0Gogx5u^a|2=a>)e^Ps4(IvFMM{P(cnVTU{u+SN
zWvMt@ylf6s<Dr%GQs+aX{Uq6JGM!W;n9Z7LP%RNv`vDk>;=5^v(q*d<zLv3?tU?KF
zHfO$5<6Ft`T&DzmdxXVuMg$}&_4gNX94#jEx~6IE_ZJ;a*sZ5$og{~eu-$3G_ETc{
zqJ2E=T{&!x_;Zq!)V+-!(!}Sq$ZPh?%vrO!SQf5J%eAa?ODhV_%|`z_&$9DWwK0gA
z*@JQjks8{hQ{C9sOaFNI6;gF+0W{g22^h+L_W2nh$%cilG?YLEID<u(&+wcmHG3g;
zSYUMe$ZsMPHD7_*j?cHRP5*xX@z>)5>hW7C2grz!Gpq}kMLpIo<z_+Nlyx?0U6OM(
z!tF`k?}5a0T<U~}8a+JH6eFac0hc|%?WtT{1u_qA1<01l{*7fUte07SiG_dJc%|wV
zRemjRxu;Rdd=I2L!*~)n1g2IhaHK#l-gLS&2X~!0ry?zK^@wuMQfn*B=CONUl)V=^
z7qwCaIl486b)7|~k}rZ618m?4!pCrr2@}ZlE5Vct7O0-RaZ0~bL2FAE=r=T}=qnP2
z=jDG*2~HO*usnO4#c9+)uvHc~hHg#l55Kr-#grjwpsMt}r;w?<m74SF@vbpQ;rH%`
z8!J5^M+_5vYmUB|fqJ155X$vEWd5H;4Bj1gn+c&ZjWM<Wb0>aO+?moo1$L5Os`}?$
zGS5Kq2+q>_5KQ)Dzy3vR*1S#EKN1rBx2dU%Hv|InDSnxiRM~Qw1~JrE-37bUB2{2D
zK>u@^ALvX0@4kMYs^8yRGu8@m>62^&g&TztQ%1nK1fB~Ov*DYKUzTa*%IJ%5&bTRj
zZ@3D!I_T%0QS+`{y}&F}$nZSL>aj>k`<#?cR+!9i>lB&t-xaqkO+&*#GRo4;89F^x
z-C?D?nQyPXO0_zuTi8MX^3+Eq;49@V`CLyQx#+5ThpMkC^By(@*g`C-HWFnt<d{%~
z46T}khEm)@h)I-P1EdWajXZulCZt!(G?@58aNWj<UDDN;A?>tC{ec7Xubnm3?<MWI
z(Ujvy2EypihC3*^pO!QYVT=OpL*uIIE6`<gx;!vFAzknow@O{xQuAXPaOb`8V;=qA
z+JJ?s4HSx$Y}k!L-YiTYC9moI!zp}}u!^jy8l2DSR@3#7leYIb#Xiu6H~;-TX5U9e
z+L>zbpvULB@w5?{{c1@exz1xkVsdabnBEJ!q3feI?Oiq4Hd3jd-*PAcwYI*3ME3i!
z-)e|p?MH;R-<tBjB)NY9ODQvh;m*gJ*)g)?T`4@tAI8fb;d&sC5vT#%`eV0t6(o?!
z)Y!9r;jF74#AR=L1nEAhTgwqK%R;8!q>ZH;7hxnL?Uty*q%rwNxD|avV^7ME8gPx%
zu8U{<W75bH*Rcc*Xo+q3RdSR(GkEmi_;H<x-q5hnpR%R~ya(W7cXw_6`i&FoR}%kR
zUtGZ}d(pL<5=dEMICuRt<xpj{reb9;Ova;bgv1VsiYuRV&2`QC&evF*RMj)g&DLmM
z8L|>Tm(GTKFiJ}e7Hlk!M3;R>)a67=<Zvi~Miq``c$0#|y{RtRgtNPQ-i3nccG$<?
zX`%fo_D^t<#2qm9uxqvCgceTR6{y^|JbUW?c_gAQTlClRENDoUxO>-(jBIz$VBSP@
zbPoL{oFGjvgsQnXu_#r>i!Xmyg_+)wrGx;kIZ4gyp|ZS!K)eO=yDBDtOG+{1zP{kV
zo{Z?DmPuHdT{tMsMiUBDE??Q}`Aj-5eYB1;K-5K`3{0=gPPxoaM_@}P`Ep;U@r_wa
zumq<7+<nZ!1zid%+TW9&4ZX8D^CStz^ohE*lwW6NTMBkd(Ki0<KQ%qhBUjl<424-#
zpluY;ia8$gXArj)qiD@9GFPNVKue<UB~$30rr6(@!du3ZS)RHvPmx;9e!(rAs9OaG
z--v(U2X-6Kgz%J**Rmo52qFJ3CX0AD$1I7uB;@$lW=Uk&S0mWrQWm|&$~r@bOE4H8
zbY(w*Rmh<y)JMHX3dQ?=KjLoNR=?tIOQm@WFkU~T-7fP9Puz_Cd@ay3u*h-5uveY7
z{>CX(_|2YTAH`dJIAEwQDof>?ZT8bU9;_>99)Kz3yu&s*APB5Wl?x`&EAbI|{^}<D
z>?UjckQ6&kL~5eOXDryMG5tXWeed!|szj&HS9810fJONq0Sof>Lmhdo>;dC);AUet
zY&K)od5VYNWV|tNp8_c@?nu(%4_no$e5C%or3dC12b#~B3nD4~N2KXYypkK^c*I|w
zEdi7SuSy1EI9`qPyQl7P-~KmM=JPxI&Hzg3mOo$$iJBphl;S@)Ny3;k(Ie?&X2)0x
zAt=w7pta{D5?=b+`FD_TDjcKDX5&FQRx5v|)h4;I#3T83EceWCxaaVUV&BtoD!iku
z@aD|oIs@6u@9ruNZ?^}Ygklo}NKFFWJ?g_fL)E@l$Y7rZv3AhGt~0;v+isb59W{%;
z9g{EUN_RqR;u@2!VUr8%JlC6I|D}n-%!t%~n&Gyy!M-X>8&P~uP@v(+x7$7@>&_)6
z$`?%r{9T_{Z@(`Pq9{B^;p;6LyPF?toqUyQ-5DF@o2IXRMdclpMVply1^5O(=!#h<
z2S#Q6G?Ew=TwamvnemLtIv^-K9zGC8qcbZAwWrvO{JrUc?B}QaQmz8`=}0d(@b**e
zU7ocH`AF_hL~TB2p-w)!{enuG1Nt(0vw~@QihZ?Qa<hWE*&g5TrLkEEiUz}iNUBmj
z7%+mvaF#QfUgY}`yBu7ncR(DpMBL)-Wcbe0RsI}Tbl}<FCy7bdBVXjRmQ_5FSc^hG
z=~$6ZFT%qb*Hf5#N4YaPX9u9xYg{+yBl_Lb)rzFMi~{ij901uD{X*$O6iB-->vkQM
za6{B?gydUflJJYg86?(rIKwuH6sGIJs_w83Yyz_xBlM3Pk8aF`Ip9=0PKJ@hOxWxw
zD~HF90*^ccy$Q*X|7$Ehmv5ILWjO=rW-?D$y};FQ^wXZi?&Lw|S4WS6%YY9UmtqgN
zd}rgrI|E57<81~N&$>=qH|OfRwQqD3)-OLX`M@Haz^}JX8^=J<Tc@3)Ip{zM%KIrm
zIq-~l5xf#(EWcEd1~9Cs%ZJ{8>8U&Y0C~Au=Cb5NJ$cHpo08VLTaIT|-UGoBFlOql
zo_Z?myj0JFk>n((sEs1vSt)jR!Afrz64iJ3Nv}W3H`-eGhlPKN;ONfPiB6b(^#KQ0
zp?QBH%e%`lRW(<a{%iU#p%S>^K@!Ii+?@1iPmMFh@e4YEz%4jF7~ImmoM74uEhAJ{
znS6WlH2mX+j}sFQbTA;Q@w~qINJSS^p!7^d6I3AjOhpwWZ6y*NRoHM5;C5L*qb%Z0
ze9~kM&lVj?#l`n$TwPw8mp(DT<svPmLm`IJJ~K_anSnlDL-R9XboIwFtZc!%FTzNp
z`pK<S6SgY!K$Y7^F3MKl0;*@KW?!eL0BjGD2S5agvu_%&x??W(^D2w;(~QU)WVbX^
zZ$RDc--$EhtJl4|hWi^3NHoHD?^W~M`1Ez@+}NAoirIYAa<V(lH}j7-LD2sdu)GOe
zGygx32-KvasC)EK@mU+UC^4<iygoU<9W;n+_`sdJ3OkT`ziihVKkM}y&zS9KcLGaK
z+TFm-6X+YT@?yW)|NQEisdmObZYbS3LFYZaH{H3;mnm@Ic3^Op2hMlzb+7lHzNruF
zov8FCxM42Yv|Q|tJ6QS}nQnS7@^9a_pi?F+n$cQj6SNE-U|8k#Y4e=@(NMa7;?SGm
zg}G$Q^13@ND0A)~bL86uzBj=Rv-!GZTz8ynCfgbN$p0|CGEEa@==y5)w+#HBmzKS>
zN6(N~42%8OtM0z3pSxCU?kM;E%=b$axcTUPEOfT`{}VjEPEUc~w*f<0y63Ldn>!l4
zKdX*!-mD#XE#4Bhf*vMVCPTU${=7Z+1FyG3z~d2%$jhrr|I?p-rir)jb6I<@e`kP~
z3CLOLoy@5B%XFdH4jg}-UndG*PuDQsE+;yS7u4@nL|z)4``_LT>@)!PdmW(13z4Av
zJ5nI;`65x|^_$(z5RS;(!y4%u<tJp`bK|8)fRWsDFcUgK92b_ImnRbpBSQx88nUR{
z`dO0wJ9@#(qOH$H#zDqr$nx3Q+2cPvE;3fK&n$9IR?h!Pa{o6esV~UO&cVSWE-B8z
zE+xUs%__|<DaFml%f%@s&BrOmCdMI5_WxEfWI<K5bh7raA>-g;<>da)*$t|Ee}hs-
z!?%M8SKgrkR@8&5hi{FId0ruVjoo=(kvq(d6U;)oDUoTUW;S2TEgsYKzM9dL<KdTO
zsZ~{(GYF-dU*e#Y6)~a#G|17uK7K)0uYzwq#_oX);6Cg9*S-dO^gB=Ejv=hq-UD&s
zc;{Y}Gic+eA*?gjNXkL1Z2(aSLyI$>8<PVQdmFKEhE_|34L2Oip3Liu8E+JsSGo?#
z582;TzhrAO%y6S<e#zBFn4v~d|B|ZBF(Zzm<(93DF@uq%;+CvUF~fGj>Q<=@HDi}0
zzYW&mlj4y@8+le69;fpYC2W#6E5q$2%h!3|gdIabma2o+WEP`KmJ8_MG*QIxk!3D(
znSH^%5*2PjGJA*nMcgcP_>$fZVNggWU{Fjk_h?KZnxM%b#_!LdH?C`um!Ev^k-Qnw
za5MdJuqKWIRe5Gjn4!X?6l3aF?Dvx<yP`B~y_(c~S*xQjdS*ws4WTB4`6bE29$7g9
zTxJWy^K{%{877QmJRU%lzL<U#Wi>~v4#N*Gm;7ZCGx-^`S+Uwe9poR%vr;EmY-P7J
zhv*|-H^K(E{-P{wLJh5lupLoHeAx7x@PlR^>6uw+#|W^)t5HAHiQy!MuY2ZQ@!`6q
zuf2GhbYkvA8}Ej3hcQ#fDNWLJ_?v8E%uVuiNZ35XkXYpu0|x=Q3t=Bo#j|P@rI6p`
z3A!xgcw(;bf>d5%SzTL{tO|;O<Z;{Pf2oGCazBbI8hSHFi8i*#Nb*(ADoaGot-z8G
zdyr{Z#ERmx^^ze-6)<525H#5n4`ZDsY7@F(?<ge-7LRtLO5%K?9I;1o35RJK_F158
z!R6};@i9w`z7PgrENCzlbq&l2n<8<?>;|M1*W6HH@Udjcy%Lu>r5un;v2)2sE>KL&
z$iCHlTk<i(pfhtN%g3LOb2)*%4i=?pqBl#$JrW9WFpClv--NmD5p7f$k(niz9-*JN
zhkY3S9F=?pl`o_!>yWTaC&drBj%H;?`?}{IyQl>rG+zjK#ZH#SPlENh73FI}Hv5JL
zVKYQ;(0FB1Kr$&Q{=`uECu@A8=~<L;h=MRNPfnuIY1xP|W>=1Cq7&TAU?HFML!{08
zj2>lzr$2s^cF0>MiTRhOg_fChAaS+>29h9Z!ZPcin_qz)oG$PgfnpW!N1a`PB^d^q
zqi;T!wjgvxvFRktR{ZB({nSEBVzs$n8T_K>_+%C(1<-``=bLUpQfe%R=xoHKb7wbJ
z@KNd5;G#-cHpvr6Q{yBV?jDAZGOVbCyN|jk=KR`Gwi}PVVDD5?IgEgjYM&zJCmAU@
zuZV5VN@`n7_va&~`R|W&f6{6ffZ@f&dbPAejOh1R>I{o=R?%Z&(lN&-Dp6G|bn#-~
zll;NykKKDkW@$0-X7J)sLgfECD4aldtXh%nF7gi-OtCnX>16(x<N-NFSW%`QbeKxD
zNjhjh@G%wQRK}9+VqD1z@C#EYHF0r7DcJg;<1{`i!Z4f3i@y6|ktD7NAav(_C4y8k
zwU+c8ETlqwvs9vzKgSp)DKlp4ApJm1l2e?=RiA`e8Fq_$uQmzYIgIZpgG|345@5z(
z{-2)e7x;kN7G`F4oVTi|@-aYj%-@fx6Zhu(M0C~AG5t;Xr)W#eiSkNVEv4GvETejK
zFQz6xvuxb-%1^_3QNU3h!x-AIUsRcT4>Do2t;>R`9n9c$X=?<wLdi0!UX+{`$+p(4
zv?7T>x|zT?WyNL&As$f#HqS=fnEayHtjn6kO}LF=PM>ddx4}kc0Z{R>vBEB~S@1`a
zF&%>>W|zA1H(}6bgaJGB=%FUoDl7A3bR?03hGuSp8QVg?DS>hf)G9$3>FCaADRpAl
zong1=Dk8;TIWynkJlyjzgyDJ|^$$5P3&Xpra6btCCN1N$dG^bYRd%e^1QAGIu=$89
z^8GK$-U2A9==~qwU0Avsq)S>tI$T5qMM;;=1xe|U5Ox6x0Re-qML=Td4k_tIq(uqo
z?vD53_xJtByfg32JG1kdbMHOp+_Q7<dE#@P=UnXx;-<dXDRZ0GjU$T6bkuC^L~qF=
zxxONJ@Hl8x`V~bOACdiieo@&B2-5hcoA|fSh{(V{hPM-?k3@Jd&I)u$k!eH)|Mhz$
z0_LQ!d~qzForrDRetI^Sl9@=P9lp1=kn(jEC%@f^Jmt$O*seXc^#t{W7W>GnWPfe5
zK|=(~isGkVN_HZ&{N3KsgruSf?jnVf{j5}nT&zY^>wx6rG7uHTUx#w%&j@lJ?oQbJ
zolJem(|2t3Lr9e}qR*5E<P9oi5a6x42+|;)HHm<ulj$-Tnk*v3(ee8C%h)13=Lkd^
zV-zEf(1CPVA<n1;_^|4;uF4DqW$oiA#p%~!2r*l>PS&l4=VjED<XMy#PFk;UC*JMx
zE^cIbV|kGM)ft&kE2H!v|9dy4By8dN@bjn=Op{YY0dkeAQ&KV`**3Q(T%L0rm|{Gq
zPj!H3h)Tc4Brrloh>6ORH8gzheSdDk*I@4iy0ppHX<sPjIJmoxsy~0zLR=mmRcXCg
z=z6zzFr2HzZpZy@?W$Vyvx_37PRuB+hZNVweUYOO2P&>}CBz%0V_xcY&WiTa$IA}c
zc_hbg6Ord(BXpaonJ)pk;I)bVlJ=iz#>;y|rKDI|;fM1v+$58ZggtzZcMX$8ho9Ph
z`g{|ULkIoxPQ>lZo(Fh@h|B7IxK0$-Ui;}yi=OiBBmB$%t_0ABcnUAK@A33dQ+-kt
zR;kNa(&^Sy|3Msd*3GXM#*=mXAU`8d|IM{14IQ)pOt#=>A{_#;&!6Fzg(FD1xxSKU
zsfv(dyQiO+;%J(Il`~`YrOMt!Oe?woIhJn-_wB^weRxHhn$FyK*FzF0|Gv)B(8!lG
zc@m|8Fq2SbDxLe#kL1mmRA@A2+B8HIwTtrdY3YUA#r`mYf4mdECyty-P~YRT<^0OH
zCM+K4(0=;%C`aoRj$r#K)&3e;O^^iQjl*7!_A6XP=zfgHW?cFmbTD1Uv!?&dJ_W|l
z(<!W{$Alc8&1gHA0nr@dJaV0_%u=iTERFZWf+N`*LS=P@G=J>^79z60cR$}tUtpzU
z|9bhf%YtmQY}I)T*^1!(gQb9A(9GV)!)v-)Fl_pL`h_<-&=v6}<=}QIn%mq9F!b;I
zkDfe(|C@}TXYnG$SWiE5u`$?=T`>n5AbbTas91u5x;|w%f}SQa7e1vG6H79A!`;*m
zqXq+*-@G_)vqtchclfyIQA<-iL9Xq+T4_ztLhkVeXq?z1zl|Ks8ReaF@1H0D3R5nI
zyOVHzsVdd|?O<?N@jjME=^uj##@`2X79~&TB@y>_{-*Y);_B!gJ#7j-R>)S>9WrTG
zeQ~dAj*(+f2bixDvd_s=27{AL&R#y=F^!&;=1#{o+m0(afES&Bl?4<?*MHAFOg4?a
zLQ2ce$xVtg^c^PO<sI=GtT=Q2q=7Nr_fj8{9pZZY#kUJkqeE+hD;|3X2=~SN8z>Rr
zV@?g;e=F9|A?)&Jsby=a*M5VxUg{<CHkY!wCFx>_^q)#Y^p-<w2`f>(D~Fv2k6|ow
z%{Z$*E7M|ETi=L>`vyy}&c}Ul(JE=Lk>K+mK9cB@)Dy^g|Cu06L%JDjm1PABW$LPc
z{TRHTk&wvo`pJ%4;>qJ31sfiDl{0rHS|(wpCrk(?KQ84dZH72AF4M=V)Y4R+`8Q9P
zT3~p1PzzSZM<~@fY%NxXc9cFD?t&sj75hmG^rIqyOYn<1uw4t`h$2omwgKxM<RZ#A
z7B?cWd`C>>Nrq`{6rD(y`KJgjG}Y%oH%%~~B6S{)IGScOkRQ!_AB91a-w$*^(^sN&
zhjFG`JzcOoQ0l`_ry!?cEMG;ITBuX7ESiM~r6`OY+G^rLdfI|V2emY31O&PZiyXEN
zdYDmgXf;-@g)!#~o*blNP6X)<31WVmD}w^7b7Y4?)S0tGvDE3aA4AlsvLA!hNwYga
z>IB)HV0CCVH%J|n%?-Xj)_9I{IhC=FbJ>xx4!$hQSchJwWh8@7DIbUv?oRmIlgyD`
zjLNml9_B+)4_0&{VIn->95m7AU;<72f);A@rbf!&<%y%OFWaM}k?H4>)EkOewFC@`
zl(jg?iiEaMM>LZ$E}-OAu}$W<S8?Rgw7~b8=Ok*e&Raor*h=PfGtd(hX%tFJ7^kxp
zI}!V&6-yZUwUx*lkJ+_4<h69kvj;d;%ZaBEf6fH*vUF`eYMPs_!~kIP{<aixTMEA|
zg=m^1O>Kg2i>TX^0k<VTO><`Gvey3%aaVVj6a5=ZE20v`y6!$GWGCofN&K6$qe-r~
ze9*tw`<QyqoSAg_e*;F5I1lRk)5<fwaPZUinK|wG(uBPpok)!IL9IlL!vWVNaWNf1
ze-^UklGy!Q<5Q#9MW2sy8pkUoNq@yV3R>QuZFuxBpo|ZA#AApKx`hGY3~2hr=p4{2
z(f0lfTs^$8A=1_92{&(-d)#l?E+^X0(I&T+t`d0Gusr$Z@q2PBX1nDwJ5$lv10T6W
zB!rIO^8PVLI1A~e_nZp&0l}$Oi$=gzFVd%!Woyx%uyWV(Eepq`=-tVpYrUPB7~BB4
z-M@kVYEMil$n9=0vM`@|T9cAq%4Dc4TxI`av59U3Z)w~;qNBgGU7k3T3&m|7Twu%0
zcyzYc@~`<PwFlBCxCa+YlJ<JF^ESJ?dhX#qMggK*HdMP}1gAY8ng<!fkYm$9-_C-k
zhNO<980T$<pnLwiSS}<&Ey!S&AhlhN-3N1Bb9d)*=WON-cAMwwSlJo>|4y-gGi+ZG
z%!93rru&T2*2L*mgyvzx(F~&~L!>5dG0JxsR2=kp7`r&odKhA;c*hoZqLp|UEEBBi
zf8w5+@*Z5qO41&T9gFu7MaUUUHHVYmf;EDo%?>7t#V?0Rbc5?yNf(15ns^bYL8b&y
znuriq>fP&y(7DZB+5*tZ+h$EHDay+b85ihiFpdlL{UB}^;`_lg!vxbXq--sT6N+RQ
zzrN-9FtIP}9^hNGIJ)2-27`t1LR*Au0e6*!4xD996a{!p_}js_<^<1z>07*)Y<P~u
zKA)f6KRU2mw`o5FaOLOdkoP~#&CduNAA)VzfR4ma=k&0vsKfw;=T~o%Z|jTY{ln`s
zdfBTeWLiL{`U9-2j3KoL83_Rjs#k)`r!lV!y2Q%Qjm|QQR>t~X{<jkiZx)8TEQx<{
zNh_xq%gD$G=qRsrtk-^$M5vQl&`{<eI1`~-*7(gilN@9NR-nH>EX8*2<v$#f=<*)3
zKK|f^z`YfWpySsq?u@p$hbS!H7BEtN4pc!Zd;wY_ec1n$rjoY46U?;0O0fH#H1-B`
z_Wy!&&<|4W{&myhs~>D6t7Nzy;aD>5K#P`vH{kQ{;C&XVl&WS!qhWrM`+h*q1WP79
zd7O3QtQIBEuly3q$puy-g}%o)ngm5M?}n?pAz>NPXg$VZHj?E2uNrW*yVB?(Mxb>N
z?jG%Se$ciX?6ovniP0mmQQ$H4VG`DPKiGd6rQi+&DmpM8CId@?TdQKYw^x4L-XPr#
zb|Xb6?6*Zb79H*0QkiaZ^f~ylZ#2?E=Br0)FnNU}ZN7;7I*Qp7wj-S=Y=pch<y+Oo
z^}t}gkjCph@;fR(J2MT}kT64@$t8+kv{{ZR;AkXjQV=j&BU`fryTMp&G$XcrGKz^)
zHV8NoE;%bjdmy^dOiIu3emJ<S1-dZMw4Z7Z_&TZ(r&t0xR$^~O8dZsl^O#BlnQjm-
zEA3+78%+>_BD7oLV%(t;X+X|{^B-*!ed@`)gt8|)T*oyVwT6$%C^Bh&ke0ymgvHqR
z2Yma|nH;M_8sueNvq3CztAtkH8McnieYg^L9ESvz`_fs-MQJlPr9L8=0WqN|JpxnE
zxRnN~qp@DY#6crg*|@V0m#u|=()YD;teLE{(!4n`*!H`Rtox77%0u8C8isHDoSp?l
zy#MAsZz-{h^~ns;Ei4Xms;EF$VK8^=eY$Eg=cMD;EOH>7IiZrLU1uVRWwGCVs{HT(
z6X>f5`7hF})Aoma4~p4kw!=`G+FEK`sUeqs_4clgEXpTsf{UH;h;PgOn=@TiS3HDk
zDY{T~<hJEzfLsH+$_2jMLg|i*`#Npl$q&~DJr6@YU5?AVAJEUuDok%@Z7FbXb8G}8
zhdjG96MsDCAN9Ly(*JtXw;E#J16=1=4yX>1OCO=p{}$760l=q=yl-as!(Ppk4^2<+
z*hlqVz)@{=G02OTH2dYWdDS>`YYGxuW>Dl2%|1r9VC<Pl>Wp(`tmHUvXyJtnnSRYI
ziJ`#a8WJuzy<M~SfYIkySC9YbuVyDUhvp(~p8A+C7c${^HM5p%JI*g0@2&pbayX%@
zM*ezuy((MK(c_;w)75;{02(+B6Iiq~gHu}l297OIo)<h-5A4}8?6oO5XH>rX7b1Uk
z)%#QZ?!sKz5uT+$s#3ssHA8td54zq+;R=&M9xBqs>Q&I}Yak1`=UE|Aid3<DF@pUk
z+Tu?jWi<Q0U8(=V<WxcLXc+d{;Mt&rkju?TK5ir1S>*Z$T<^eN4~?48`V6YjB-Y;#
zXDf3ehzR{>NsoANtqfLoE-?|8pOELT?#2)(7Kb!lxEFL|ouNUWgV+hLdK2v<74h>x
zFKK*Vltj2`V(|$c{+iV+BzO_bj~;*6oRJ?P<3c@Zw7R51<p4^4Gf7g6wagSPVYuqv
zoXb6c#-9O6Dq=#o0-jYd!FKWpD+3jV9WDMyi?4wjzqxNMk<fu^1PbS1&ynZ(V<b*u
z-`8FO_Jj*9k)i%lIs<8-#s4-fP)cA&5yEW){l8*->VfpTWO5J%hAaaZ33Om>iE_NB
zU;eJndFr!ScTu~co^0UYC^bt%U`7PvMc270Ys1?6{i{Nu{_W%hZ6tcHSk<R}I`dRt
zo{&z0uUJ(hQ^jJ<Ej0W!ezDUY1f@PJau<~x0={Z>x8h9nVtX%%f=b@;uSksd&;A^D
z(S*SsBAL5ltEv4d?xHG#^xfs9A+pnA>mojbJyvqgoW&w4gFQ2nHt6>6rUxWdN<!Sd
z(n&lkR=%lX=+5wVY^l3CGpT_Lgnl~Z)p@v8pqxZtzb>osWU*me3+IG;;OF*^(7f+&
zx$joQr3P-$9Q!5M2U+*qxr=@(+B0}-S)459^7FgG<sJJGVy9qJ!i&=j>Nc6L#nMUP
z%S1C}r5X<2=EcdMrZm*8)1|SLXpZ|+;DJid)ZIl7`ssHKCiTSSgzpUW2#DF3_kWM(
zrdASLRzpfB=>m;wN|6urpv5Ay{q(UewKR!kr91v6Rdk&K6?y8?REO&-4_q~0IBI4r
z*;BW{!b4P~UySDResY)PP^=BP8=y)}&Wa-?IP+IefZ$QEH<`n{kD0bIqbfuwmPi}b
zG@T0}RmC!mB`H@73Q1_x^y>j7v6BBs4k7_TShjm!Vjsz_N%T~ao(H$?)-K>jk9;q=
z(N7r*k|FaDEK+>_KlXvHrt|!j*|_xp8hU>hUKB$9Xh)keOPsQP;QL*Yu~QsdT**KZ
zS5otFcN-S-V)qpuI;_8Q%&V2VVV=`{jw3&b4y|!oSm_{bb3((PHvol>{I5pe#Xc3d
zgbHn4vn4`fd`KFb$gR!8_H75hn{m^7K>?}%<G`Mp+)A?l`|K`0bsN}e^ReNnvjQfO
zl662RXp?=w2?C4!O-G14ksOQm*!eOfw7lu|?6oDhVDqctDK3_Mm{NmOWh7U_8qM_&
zO}P1>ko4?8`7W|~6HZj;<qlfjTsAm0%gux2cxl<!V6zT5eF(X^I&dcZ;1!uPh5FLK
zc>LRq@Md}lLv!QB9~eMxizCavg_sl?*&NqDZH~vd%C}%lzD;*vOh$ioV18Q@-jvW)
zDdn5InIc|y|B*|#ytyzYbiV1Oe_Ej(P(vBJZ6YttzJ;_Oku%)%GJ3RlFjY@{4j$S3
zIBoJ`3LX$9z&@4hg;%qIHCY+?78ts$J-N>;obubW@(+_o(~ZvLOMo*`YC`nu<=m8V
z@o)Ob+WmV4z>T}+*eDN_yzkqPZ@sZn*2C{q+!R~MLb~X`-8qN|3D%6HW7HPuXa8eT
zq@u<89m)QO;tPV)B}j_Qk=*$N!;aH#*e-dk+3|?XnH<TY39uRe3_eyYPJ?#6lrbS<
zZ<vfr;Hy|H0@zpOfXkCHsauU23yo0BR1R2Hw%5k)g}qb09!Nywpb0)3Ek5}6u#7+x
z-(Imw`TA!?!SK5vBRPAJC(Rj4ML{iO9r-_APc69EaPj0OhM^|N%Z2HFkkBxejKF^G
zJ*|cm^c@d~ymN^($90DN1xl#3)^V`?MTbY+;|Fv&!OPG5n)y5oLP=-DZRTyL8*n>Y
z!lg?-^GFbE2fMnPB4ZGg<`jUzD2!{}isepnnKmuII5)HWgNpXeT&EXZ2eNCKamDEI
zVYN9c_FL_X_Ah_F{w;s6`iribiS+xx8Dz^xmH|ELCD2l}H8-qhXarI{k&tk_n|RQ{
zGD?G2lZG8@eaF9ArysC@tI3`2%vXPKxHC}Q2=cEM>#sA;^GX7Js4jE5vr+xyRRh`Q
z;5Y6~LPqah9q+VNtAR`%|4t(%(w!zy1nx~|ugB=>fT|4)=7U~E(z5}@(=sUSU*VO%
z%?FPp;KtC?^7q!j;d`ZD1Iw2vz30;~*?d^2<K?f%8IDB1Iy1NKEMIcySOUfREmNQ+
zlQ|i7<GFY6w4I)JmAv_&&7XqA;&pn_6*JF*aF<_`mHtDi8NZO)MDx`uEqTgQ7%HXg
zEwpjM10C|ZYAySVCTV2#*GVkau&;vaJuEz-4!=kR?GLp(RE)PdJ}?`NdNd+5S{SKG
zBCkK?GSM7mo=EeA29ibkzRr!vWTgN!WjNM6%~2Cr{2fjiV|xIzG75O5v~I5S>_39E
zVIP%cw|SN!JxIWlfZ@0pd2Bw&dq<lcWE^Yv!!AQlnWh^@8cnun;Gvale*pL{U}YMU
z6D~MaFoCB14@nk2UBHPE&jTN!N4|tLedVs2Bglh1N4I{-ZOWVmSc1wR2b!y(oxCsf
zl`h6O-X2UdAbB)lrNLoe=vXaQWv~KT(8=%+Gn<8+HbT?R;JikWcuCUXyl%y#t6Y^d
zv9Iuh(Z&$(UA1^;JG66y^xjZM86L_Gjp@@y{KnNWC+cP_KIQW!N_W}P!;7&NIKRsH
zVJ%fTnB}>Z?z&Z!MI)I+bGWYa^slcJUe)RYGfncg@iIGknn(JFNTC8q-R%L6`EQx!
zLZszSUkX^Vr(}4=YNV56sM1Dv74y)$0jtP&LNB1_g}H`@8aHan_l;C4j-H{oHuTG*
zl{VIBC}oPgx~qzBIB<O4e%hd`HpSL_DL;Y0nMHT<RGAWx*~ycd_sB>TO#~7|CVAwG
zxD^5S;j#s-Rs;|mmnCKxY0JJcRH^WMhBWTN97zm2iIn)1CC~NwR&;z@JY#n&ssSw;
z?o=k~1;D9os{<%Kj0$sjLZ5Fm(%d;0q$YP_N*Q-HdiC9py1}xHzQXsd@dg&n##v#6
z45t!Jf?*E(_oCUV+RBv|k7_;N-VPl<5_1vASG(XMpQn6%(2%jXO0tJHFz%U#od4Vb
z&ucE)&~B$kd2W<DMe}2ON@u?EwQ>#4(E9=7iPcjyP(yP3UQnWIdWz%-CMafj76Wl*
z{k~xRAk(qVLs%ZirPbV*(`XUQ{UrNtkawit{9A^*D62VWK5Nf{sc!x>;IImLX%1w0
zp!H%2<!DX{yu}A^xu-?ceF-E6{N6lP|DSVAv}iDM=VxUCLX%)$k(#HbcGg$7iW*HZ
z8e|`7bi)E41^B1Ih3mPH2ovDN3+yL4SY@!q5kv#;)8ojC)tjHzMXalKz6o9~l*To+
z8e5oZzzeu@%ihSMe=Rhfa#S?{#y+}1n*M2?sbX`7e@78F5(^e)g%^cFd(PVGY{}1I
z{WXU?3$i1t+<wrM>vhsfMfaBh)_AT0n<yF7pCMI1djX`CcefS2=`e#{QSqASm~`ZY
z9r4z9sndW3{y8{pDH?G&Z(gDdF~LN5I@B|W&n#piRZ`Z1;Dtt!b1#%1#KbxX4ld~6
zj6J$Jm|a4CAT<BY5L#q0$*^AJ0}3i+n=t&+;n!8^$k0+Re&0x`;vZCTqO=`&szQ|<
zk}2NuIrz|+31>IXjTuVAvc-vv+ET0dB1-}axuu#y?>kh{%+3_*GudJ%R*akPN!~o7
zeh$hR*K{r-Hvc*8IZSLQ5U_eC|96^Uq%d;`$zWRap-BADV597`M-RjFv}o=0cll|?
zO)?_9GStFBs#3yy@7@z#U@Bw|F`O6vt~$**!9e`S-!Qt$f6oLO-l;mB2zc^b)84Uv
zVYW!95*Xu&;Po02H47*|=()pWrG(?&j!QmG%D8o{{v-@<xN5*XUU<$!lL%_@EeQG4
zh1!^nhJ<l>?TL1SEwFJ&#ZA!{%141xWR~~`voDkjxxSq{AF9$y-36PBrxJOHi0QH3
z^;P`FB@mA(E4DobNveX3<5Cly-VGwPKMBNIyi?BNTMlH8wS4!dhBuy+0LIo0%|TCU
z|GDB2=ba}$_kIEH6<AWjE#Sc&LysV}6`H>2QxE^hEBic@gs4IfYB$N{FTgdOd674+
z)OY%?xH6bcu--3U&tP7>Dj$`+(u^@wtf<R43ylz*lc@ze1xct~RT+<Yp@fD>d=+bQ
z(oYJiac6M+<IkIeBYs1$%*luVa~`stc-KvTYa5cE8xbH(@_68Hul6Sq7Z&8LFG-##
zta0owo`gf%%IOWU9?;qI98*}XcgvTwNw*f7i8S)mPgeR>jF_|!_rYn-VdpLm1uq)D
z7an?)$;Ylp6b`K|V|bZ<x2K?^8NO+(Tmd)1$*@f~+b^Km&@FSKF46t*ih4ko++7@N
zb>Vf1Pkex~(dxOku@P}VD02J+-N9E}#6H32R)qL%AMK-?j#LR@Y}aCjhG1s%L9J|g
z`?~ldmqvmTLn#V*oe`OT?2~)~{8UBcao3A<lUL*)Xb~OhZh)?f5I^nT`&`e-7Zl;<
z)lB0hgAm2+v=b{GMG`4193A#EIKwu;FpN&4Y<ToLF!0ZYTf#%gT+=U|r!~{Wu?|&~
z=9BA8C~b`Yvd^-tebjy9GuoO6PYh3$j8%FY)fnP8FvLG%L(K`ikB5wuE5`6q_c131
ziY;Kxm+%p6uFK<Dd%{tpmqfrd_@z<WQn=;HI$cT#N|Oxddmy7PBK^Roas@BKJ^t9?
z5}rCGsUm>$2x*XfU^d3d@LzhZR<SUL3rHybq{x_%x?2B97M-Ebup2}ai={0ncp*D2
zK$#sZ5K9~+V|}wn7E}4{A)p%T;&sm|C4MpxDx>14+4m=db{wh%sW7no9C7hHS9<f8
zA>jlqz69h|XcA+vo?=zbZ~Y0HBb2k5bJ`s8!|@FMw&pjS<iPYI>dHWpVOzV`(^Y^B
z161boO*G+5_9T{!pyOr2Xf0vhmpe0%n!t2z6*sy=!)fE>YC2G`dH80`L0-)73HQKZ
zam}_Tp$2?!%(0^&jYSRd#|sYAaoImew#y87hslwkuPyfz!KAGmG?l{4&kd*fogI+(
zVp=KrUW^|6k=)rh`@|6drPO;UuvO!fAD+hl0W_fdU2y9+3p@mvk$5O1zexP?#r{h!
zns)Fr<JNZSXU0!TD<9x%Q33@pT|*F(|Fi}G#Uq71nyR*X>X;n=voqSazWhF{KS$Di
zni1Xg-|JcUC+mVd`NJzeI{zU;<|ebwN5r4Zp^8fLx(L1(p|=RQi%a#po)WjzQx_U3
z*2WPW#5WrFDSnUh9Z|K`Jzd_@Ew2h9=YPN4C{f_`9CV|*>kR!LAajDw>jJ=}Cj)jX
zIt)@k-RJixrT%Q#S{mnc;B~D!6zI#JR}=Jw{{+_T6IKuRL(&onYvBv(mfCgL{J~8q
z7;_jmYDtX~Ik=w5=HmwcohC>wVjb!I>|-W}He57pO`vA<4bUSUxQNJD`|ufoGW4-p
z`}@ozy!lavn%It!DBP;0p*=zEbA5ZJnm`?q*v|CCD0YsJ>lQ_rP_QLp9G}1#(Zj`V
z(wTB{VwT<c1eqN)@g@l-8$HdHZ1NTIx&E2NPME*_CYZO9Q1`RTvG_H`giz!1RpOui
zoOHYtA+dpr9_9|sb&3rk`eh;afr}1PIO{x@sR>QuiCQo)IdgQQ6pPi0ZnR(hDT&Y5
z?d!BnqkG%X-VO|@P-LeEcNF_bNNb1UNazlj#AG}LdE$r6<nH!-XPiI`$`LZ^zxe5#
zPctB-wO;;vgu@g~S8VcUPXEKSnpLCKX?=UD+S`Ul#eM2(8rt_7MnANds-=E=^t~_O
zZ?0qv-H1@x!N<LIx5*Ly*-V1M+nXNNF_s;Eb%W2%a<xrxeK%3p_gz_ek-4}ApP%Lm
z9$(@>>UL{wea;~*Tq7ShK}>Rmny4GaBRehD?k#D&dz%mnt~gX-c_D)R7IGJ_=I{&$
z4lHbu5q=T4LMjwR*pgY&e_+x60W54m!YiU;oiPs|h_dYRKx-9cR4bz%7-@fc5sUMU
zwMVq+kQZ(hOw@O}X;zR3aOMi#Wl_y%$bxPp)CWyRrN5lwl>dc{%3xVG1}=cL1b9yR
z%fHNuBLL6oY0_Zk#20v2fuMth-gA2xM;@+c{Lwn#5`ZUQ!7m%<4LXuxnqA2(F}r!x
z_l?G#ud3~hTK>P5#<qhb`h4|@(p3|b)scI7X?>1er-$x);Ws4#i9K2NGRML!#ZMzl
zw*lw9H^obsn7c*3?|g2f>cVvnOKgrU_FYVC_RFii!s&q^blF(Ocx947yqOyai;b1)
zZJ-Mo7@^1*j=<Fra5yoUekL<$aNPSqW|G(W>oGwdT7sD~*tYLnwxk+h%~<hkkuaja
z(>yV9tR3aou6pl(vr=aV-|ss=q0RqJSPm2FE-kz{59cdaB<aS6vJRC8$P`b%0f-71
zZyqEh{%GbN<_Ucsu&zP!!l>F^w+aYXFjhZM?K8F`#_-3d(G+v*1&-I-LyqQXY@q}3
zCf@m80`5!Z#Qa^vEqw53iN0dwnfzFJY+M#)WiYeL&vK)tl4&QGJBNnH11!AbxVKyd
z&T9d`Qr#*gbVlli_(coeutf&wwcy2dJi~!=(qHC8$ftPu6NiFYQf8Ej`E)sQEe0-v
z#aynMkeyVu8FB;fu-wJ&*HU{0o}h7Y1mN&Nchpi-rD>IIHJQZ}(%H8Jh~2^e0s|N_
zsXCMr;a>bzbV;6wn;@s*F>X4X+@Hoc{;Fw%53%r&0G1K*qE8|Il2E@U7H1z;{}S2t
z;scn>4(K-40bEB@>I7D#*%!QUyY(?MhuzrSUOdVULG-zL<cGTu;?<yx3CGYgZcaCi
z_8(jV6)I3JNJ4@9YLql#ca^54bQ5hJQ;$(gDRj{=4~kbl{7RaXa`aJO?))@_5|IJd
zj&+!?U%EsjC9}>zLIuqyA#Q>*m8l&=c{Fvv{WB-rOfMdH0F~mGzHvNOCNiqEl-6ig
z>Ie`vm!6H(-4miKPyH!QE#hjyI8!lGyi(KB3`LF_wNydB8KPALpXqPUm0<9b1?OhJ
zbGBqc3yts!M12aleVx5MLp=#A;xW!a?`$B+X_-^>$!Q|lf^ya0AB^F-g=*VKdC*wY
zsXaZR4^6X}G$0C#wvi<~*V~>i&0w3L=a`@$PZM!Vvlui;U2576^+H+L(0PUW+C&-K
z5FvwXh|FvzJwl~z?BXvr)>&-2J#Rbo301Q3iN82L$&cS#kJ?)=9Cn*Mu<Iz@CV~$>
z6*wU+r)tB9y_Mz1?e^5Qt7Ex(KE^&l|1@n(fBWuJ>b6udW7)Q^6~28}59|33b<4I3
zf4DUt;&&tqoMedGrj?P*8dNMX7Z&7wLZ-b5Pt6=PXjr;h(1T}IojC=O^clSflWMD~
z5bE>m!-a1`f5)F-7@7uScQ;8^E540A(2KG-Afg;1+b*2l7t?qlVtytrx<@oINLMhZ
zxgN!~MN~3)81rB`mDnI)FL#s<e<)NCIZw?83~)ScO>m~>GWSmb!Q%?CMVU_!B&j}~
zCUuP9&C>cqn^kWK8|4(Ds3#gBK|JQVr#-LvXmGwOVuxd^lR*|>mDCCc+~Uz-ncV9{
zm3P2WOVKXNjp6f;Fo-I&N7sAb0ARo+cOfy<0V|U54>+QV_0*=vJn>*RSUk0~!WVss
zZd_&7i3r3WbOf+gxDVsy!J<Z@CnMyXUo^OBLJAafsHx?U#sx5@ys|IT*d6x&$FexT
zE!GyA(zxu)_=NrK>Llw;`feVxExe1lrT)E!r@@b+2g5Xm(`BSH*jIDa)Mrg!4QV{y
zM2s<HAbdJ=AGe^W2QKUEcLFV19=_Kg4!VFDMg?o}pNAk{=kL)aLVm*PQcJ%iT*lkm
zv?vJ^au=;0z42F*i6_@6v^{$;>J**G*B@~spc_NBCN&zNFpfbaQg8L7T7H&^F8!kM
z=CSg9*=VSi?%@<tOMt<AQskVm%ixDQN5SyDAZh;SJ8<&(Xv(6=sV|6+)U#V79B*aS
zs~5IkQIl^Y>E@$<26jc@s=Www=~DHMvgo4My?h$O7rQG(h@KKLW%@fN@|`Jb)YYkU
zufpD#h-|x-Vs02XKLI+Few_c8AJxvnF(SgwWI5<sN8P;LHBJ2#s1CHL0U65{o2T^>
zp_$O5B2SrKgC5B^RV>VY%W9mPb)|)MSw4~(NA{v?0>`n&5E=5im&Gd6khbWuZ94zx
zs`k64Ob!+r0@9ry|2`3Z)v<G9<QV$AOA)_<rb4QxS6s)xE!BQ>Ln+^qQKmEVuL|fx
z^p8#bJ`eVhJ%9e|ETXL#WrB>8CGgEiPC15zS>CFw(1O3`teR$z2v?v-njT+2++CEQ
z6nVL9MP&YCphqf}NMFqTBUIw9Qqk#8mJw6^V0Y0d>et2l$_E6>lI9`<PXH1-iIr!9
z)~sOQTT4euG=ie3Y>TywF_v&2roaQ@V(p8#nO~Y#WaGhU4J_Q8_q2ZI!1mJ({Y)>2
zPm|alCYXmsj2?}MQOLhq3rTQ&%iW4ahLk(@yz}zK|B5OhDGt7a#`Orq=w4=&Or*=g
z?yO?J|E7I?>AE3&6r|vSJ5#w333l`%^uS_DU=t?lgf+)WX^(9S?FN#QU5dn3e1#h1
zqQ+5UL1lk+Db7e7MZnB~7rGx$X(I+ef+UzYtCg0*xa>xis=_~G4u#Q31_DN1?ZM`k
z#VCWGU)P2)mDT+0&`Mv2L|~`yyoZF}a7@%@pUjU+?*#KJ8iR}eB!eog9Vmti3Pp5B
z#)}-QT=>_am6!YyuF~QBbJZWy{lyiFQ8Z2ikSS1DEk%1r9IH?<wySGv7=Nq<SKGZG
z);t|tcmF8O6VT;ejD1W(3aL-IC)Uojb&RnI3E{c|-9+{&8XJ==EU=T&I&d~XI&?*s
zf2O;>_e^a3fMp&kDGxhQv&`lEyz$7mFn)6ZLP0z6t`<_IE1H(abRSaeEXTN1wK_E%
zruM{G0JXV5K~9?x4mjSE)V5_GrH|~2>0DiX-krF7bj0}V-$_Nb!N4@KUao23kwAyW
zumyyLB`s}WZLLguM-@jgtR}AiCh~ZK)ORwllY-jyy&6zSVgZqaNQmboAu{OcGs_>)
zpQ)q8KXc$HvPi^ojmx?-ghANON02Az@>rW762`g4zqAFSKb2hH+kz?^yr1k4=U$UN
zsm0coO{@5bG+Uu)TC`{_y#G=$hFvi)?$W6);$Qp{-CX5I(zpT%0#1}Z|5-Oj#{x`l
z7bJL{0--2c{G>O^iSHd44;};=Uh45*;B;JG>WvB!X73*#Eka?D=QLo!B_jwQEIkf;
zu#$e`j3fc9yT>$WQyyhX7<oi80g7K`Y9in(LM~=RR2ndur~R4tZCU@;27h&D(CoRd
z-~>#}7F>iT_CP)6Kda&0Dqt#oZ0XDslYE--l?9q=%=4?W>nAI<q@rXkRHkUL!F5`Z
zbQS6kv-#>lbl&R7UpEPFkTu66!djZZ-3|ux*IfopzekA(lWYe)K_`e1Tenc|wmfAT
zLaywyd1G6*QuV1smp?2EBx|`d)Amumu}PZ$J;GS+k5aOHSX@)mP1sY1CnwBJo32dT
z%<-GbhIZqr41dQ=1zbc=xes{A(Y?&)n;k}WJ<a)L0HMU}wpCYDxn0bs4QAi+DXKg(
z79sl82kx#4Q}8p;onF!s?0dXZcHgDs4-$fTzLPncKK`~dQ@#k>1R7CzIJi+s|9#6&
z+Lm+XXVB1UMJ`twrpV*)^_V(Ffn&y$5|$yKJ=I{JLuyvSJz1Hn)=}oP127?)_9c`a
z;1^SE;f)gfS9L@7M*UL`HqQjizl|6*5)UIt;m0_jCx3u{Cosiv;YU|CVUu$Yxo4-B
zSP>@eSEG4N1AL8jSoht-<VL`=_v@}Y0nK&($%|`ZM>jqLKbme!Z4`MFgP6-~*kqgd
z!x#yZMPVmZjZ6f<l(E=g`QFcJoUL|=`Hn0H3%>5yvc0++4*9`~(;oDQ%pzw;^~W{v
z{)8{Sv_0uz1^ffPUDW+K=o%{$%PN?Y-+yP;4w=-_K{!HQ@--3nHdqI%gSG3a<Vh8M
zAJs|d0T1oJ=rMc6cX8XBd9KvM=?;Jij7vPnXN&zD*5^NAux8e%^=p0ZK2{O&AA(}x
z6oamHoZ;`YX)rav8lyEa4LhTvYyUKj1W5G3^XNu|KWxKq7RPW5=|^^`zjnj)v-%31
z5R+@$6uGQv1>G6<Ly6ZLg%uvIHjgE~{ALt()@J&g8`S4sQ-18Vkr;QKYqqtGHNNlU
zk8N!bVni*b>wBy#?@PoinpcgI;P(TYP3cqvi|2Et&dQ({Q=<HY@3lUuq8yBbVnK*P
z0jI}Rla(O~MyKV-5H6#4y31!eW93w~C0@zEdP<+=OCK{WL3YQ?UR|1&Ou?DSr#0mx
zEP41UwuE-o!><U<s-J?^{(R3Aq?+WbEKjbnBg~$9m@Jq*_5EWJI}pzV5&%kqa+3vR
z<;#YSk@t;MD?;uYRn?|+!J93q%B%=KRI8roR>XN8I#eLNQ2%q?yJ`$e9~wX?S?yvO
z8S=Zv$#AMj!ZDK1D6c%^j!|v3I8kM(tl=xWq+><?28Zc~n1&b>8xC_LoN97JbjvWo
z|9fvI(&#-P{5`tf7t!;lXpv5i7;XV;B$2TWOP;*K4lgy>;iK9)z1Nofe#=-9cFIRQ
zXr$*TCG2fn<6>^Xd+>+D2wrA8yF)89^oJNKUXGNY{1oCikMrcsu!`_*9zG_`uL8!r
zYdD3W8B63EYFw*4fQVc0x;vAbm$^AN#We<~zc(0~F#_?ylH%mA9>PmebqQlVjp8Wr
zo<>hpTzBIU%Cn;}PWQMPIZh*>vh4^#!lTH26N+w63y^IS_wJ_C=DW(8ySnZWw|rdq
zqu%m;@egy0_WU1N*45Svv#f#h7RQ!_g+G4i+1~1F^{W5Si!QWLL4a)GZn?i}Fr`va
zxy4yaT1%~1A}aNreH*9pTNBM9X+5yF;ZGWX{6UXhu>2IFvi&8^e^H3sZ~d$w>3%hX
z5ol^&FWom}LVP`}OBz|RewyJY-rIZp^94Dsa2T(@)7PCky<pMHgLHeXkTpLi?)^Hw
z-@3;|DFx<RmvF(T68O%~!p*~#7lHRMrYJs);RZ9ku6yC;#Za><mHD}89qwJds%(Gy
z;fKeyWxe}H&-&cO;6nq*c3H`T-5E@W?s1mK78IUXxnez%o>CV;->2x_o3S?E_Xw*u
zM%Zw*M=Qj~e|yXtwvg1cV%?OU!t3p1JzJ-Dr%*0iyFrCvdACk)FXRmAWW83G)V5;U
zLA*R)msGN1U6Afq)z@puI8E~*v1DRYt=TF$BjwC@8_wm^-TNEqtSS2$v7DY_0GweW
zp3!Hj+TUDBL6cJM%*=>(56s9(x*yE+54ej-4mI-8?=C^yb$5(>e&kjPwsjW-R2{jx
zocoM+PSUrw_SEdlUpKj%&GkW>j2{71PD?_hCcm?RCgq>55Z$0<$yc1x&2wHvw0Lf<
z&*_8^Mc+yk-y=mfgU)=Vy<GyeX`}dPRow$sX``fR3pz3hJKslg^0NeFd7+eOSG`-e
zD22$@g0z;{fAz9lW+u;#DU<Mkr91+aXr(+d8(63e&UE<gqkvu%fL=2Fn9T=6wmndK
zwDrw_PETPA_evMP^!A#4!!X5O5dvS~JK-^>s~b*3;>RvOJYlvi+t(Y+69M)#%26??
zbSsrL*r|_F0Y)VajP{4+lJjL#_3(3ij~Ynl$H)Nwo3v&)G`JrcZ%kx#*4cyz_h@Bx
z52T?@bl(G=7GAdX*azLG<#?0Unn3uXF0p(+<{TcWO1uAZy!U3oBtm|<8f(>E#AYeO
zLLKP?Z1BieDfNW!^?~Y=6LbDd8%La<ceYzX^yn0t^=6k+{2j!Cm6;RuC>Gne1uH22
zjlYo*XRuT(i71S_8#2X;yBGwCwO*RNq;|w)$oWD||IrgX0Eq(`4SdLd;}l(8ggu7F
z84Y5OhTofQ3rdf_%5fSW1~If!^EV@TepKrz%G-KLX-^n@+moMr-Rjr66L=E%o|Pap
zBg^&fDg@B+tKdd7=?nxQ{B6tkSteF_jTCQ#G+P{pNkdzSF{L)6`mc+Iw*)*&2#Tty
z2W}FXH3<p8eR!8aw7b+7w9{u-Uj;ojBFNE}!r*N~<wKRK3lVgK?#y3)q!(Mv*OqKJ
zXS3GxZ7e7K%mytN(umQO9mULPJ50Ct(-=-h$uD!ql=&56x_hk}>&KL}URJtVO!;4&
z9*`5nZr~k7yKM0=%uQ>(g;zGIzZY5Hp_v;aFK3>g)e3?i_SSQ3BPKE2ye-7qU@ztz
z@MoFtJshcW|6t*Y8My>bfD>KjABW|t0K0M1U7O5ZF;j#8RWO$QUc~`GPu3d@UBfVm
z@G)Lv%<kbTK}%_tHHutC)%%4Vbp`%3L?oub`)Q-v8)^IUh%@%Fg+B5oMia|#Lo4Lh
zO!Kq<1!Bk>X#$zbU6C!&-58gByt^?npkySIE9O&{ABZ&iyl4T~3Wi42{jGvZqWfEE
zl|=9MfoIXe3#v83Kdzp17A>r;nDH(%t|@SfnFkH=Tt6`nX&%i|(cRm2#-ts#CUE?&
zwTk8@!5q5eMA)zrtS)n~>F(Wi))k;y<>s1))i7%LdSrc&LjQ{71bvhwv!g~H88`xD
z_2?tX`-w}0{-U!y^+|>BN6P=T4X~^NsW7L%YHR`1U|@_`Nv;<l3qe$nt;+v~@rrj>
zM*!hC*L*Mq7&sp9nM3!2%@tM5A;CHjX5MCCNqtjZ=(qt^YbP>>R-vKTM`vE~$;Ma%
z0Id%y(%r+Vu}A&_WkE65KKjc;V#9<og$ubXx}TVb@y``A;BqHQ@LRs0q@9kg^l4YV
zeOMq6b*?CkuOAqjb1o;)x<XG^)7yPBN^>sh%+QYtGl#I^Wq2?*Q+dCN!*>Q&Z}H~T
z^%}M3-ocx;l?&cu&y<B$CCOjAVEPI3AZ_{*(`~-U5$5s4<5K;~u#PYCg_K)+clBRR
zB{Deh%qfT$(rsCE*pl);_S46o{`=bK<yO}yiJ3lZx(zbIOcQ%til_`5`zkL$zGWeA
z7sH~czcJN@_`ysQcwEX@*_GiC_x#f|byF#-GHfYFO-gd!BF&ccb<$Dk=gO{<mw|jm
z`_{G9NEn%J*s+WIbcEgqDyIh<7FYT)HwAE#n=jLBSo~^!m*-(SGA(%H-;PSVi@Ij5
z<0#+TKN~cFOtewdGB+#eoP*w-Y)Gq?H+S)|DBDHKg4FbZYua@$k@PCawkPooN2S=s
zUC0SOGvO@3?Lb!)`87G_&0a+T>s~#f(>;G=@{yZ-;l~XNCS%f#IbVI;i8dLaW9+2z
zUOlPGElM(~eDiR>oYFV&4$SvDB@GIBKJ2$as$`7bf1=9lh%H!cdz!zQC}aJ{(wwX8
zogLKtBt=H5UsK3N5703{yv?gvtGFHFA2iIjR5X(ngmw-Y8w#vX#<9E_p+N!>hl@el
zEeS)RZkmJ<MQ#@d_Qqho!1$B`AS{-7S0^)*e$-$o<H$bpHgY@AVUEBSR|e{bwR*3X
z26H)zw5uQ)#y@>JGfW6r27r4E{jGb}0e1XB>J)ez!3`tD46m<7qyCRv1$3a+P1m1_
z-h2K=4H-xrfKi@1&S>n7k3o`W>^i1--Ldk-f=XzbE9DcD(3|;vu-7u@V4Uj=rL=t_
z<0B00T*X9k9nIZ*hckc!y!+@*ca<W8<gU5Wb41bpui4S`9|(?U$4_UpDQ%U3xOcou
z0mv)=U>`TzTr4OGa(ZP2?w=ZFc}Z6mNPZ^-gfx>--G8i^+;?ZXnr?BQsWv!wk=1Fg
z4=3mqYxh5RObp9@F9SQ497be~7BIf9dT=;j=%XGQ%~0?P-HRjPSP8niKN6$WKh=`@
z-pOUG;65prBi08AVVf*Mm+?1Pb`rw~r7&!F3C)*SY7(VZ07$LX25R52E&E{!N4}$;
z<~1{xd;{{A7`Dku(DTjM>GicstO%680{|Z|_PyAQrjHii3r5bXnpMncRFkEX^AH~c
z3%d(U+r4+*6X6=;GSlNP!1Ca6{%Td~Xge|yR+_hgxB3cokNpE6dG)m=N3h&r-{2-b
zna8;-Yd|QVD9EOcqmI%7S#q2ksd0+o6lHP2;VvgyG=y)gR3$vr#Eu9Yv|R~&stGMn
zB%aHdQXMo<apRURC;w~&KY3uUcZ3GN*bTOfRRKZ)UFv1ZiH&ie>x$3h3`oJreTY3E
z>b)vJVkWG_6#4t)HhX1w+gX3?RI(}oNYL>>HOw*Zg0DFv7?IHbm7W87x+OqYUXY~c
zFr;J7(EVNorSsbuK5rs=8-{zIwOMWP^%G{f>VganZOf5HPPKIXu?%CLP%gDSRxoQD
zMUT;lVQ+WoVr7Beo996K2Yy(*Fye_jv8FjbH2xoygEmZsK8)p1VSF760INo=4#a_s
z@4lu;7R>CRFfd_%ua@nX-(C<P|2QQsm;gniW?S}xK=ot<jv`WfjxrYdaLPr|{AeLT
z_41?o(+|Trh8SOm%;P8`HA&Ev*~n+2-*45NtD<kt4+EZDQs@ZA`xdR6AhR9!%-^-&
zoA3Pi@!@S#ig#us&B2@j?=FRi0spJB#MVeo|9dbOmHxxP95!LB?n8<s(}$rsU6&G?
z0l9@w`(Ak`q=VjAwBqfsF*EiC+o^XvSRi~{55L-YYfn!_@?kz!YcBS+M>toQh3-+p
zJ#ET<X`&TU9SWCl@-RBrf6&b)J-R~`8m6@Gr7A?aRvN~6sCn0*XPu1z$=GQ)W;0Rg
z_e$nbvQoA*0B@wD{yw<!g>^|=@kjy3p+QO?--O=@<8cXx{0BI%g?|;K*NwUhYce7R
z*|0h?xocJj%zv{PGJfd*!2X=zUx}G<p*8b*+Oe|W9>$Vo9X{X5=ENu}%J^l9?J?t*
z7B&LL5#$Le`C=^aAB>q1?Z8<8o+uElA5^_|7{YT4XASFi>GpP+?841HeirlKQK*uY
z)Ia!4WOd^by+dy6dfHMDU7Ie}s3wde3e$;d`FpuW<v2FAqa%3J@8w$mwGUtFY=)W`
z0Tacu-}^n0?<x?Eh0jXP33lS0#B?H)J*hjh{+pA=h8GQ4aUNC|gK-ZVx4-&FEV@+a
z$Q$AFKwqL6U!Z{Lz@R9P7eIIIZ8W3>aEeSkIEJQQ>F?`TT|L_>1`jaUV{O55eVlJo
zT6|%V@UCa}6+$kQ9>KaTo4u4<*EF@n+o%9x5KgQ7u*2AsE#&kV86A$lIaL&RfGP#_
zX->gBT6=;j@m4H)w}UyTO%yQ4RCYr^)DJ9<={Z@;`!q7LryY1^Y#$RfwZW$kk0=x1
z8kdiN7ha`;lFSNM(MuSC%i;Ao3lAgYpUcpNB`~EJ%R~!{WlCdc4`+&+Yz=3knp6z@
z2WgB*Sp945qWX2eX<oxkbVsy$1ILIuIcugYGxy8Evirg56q5SIZ2q{CUM`lnq+Yw`
zxiqY8BvaG8RqLU&tJ1}w!>J$iOjeVM`AlytmckC1Mt><rGF46jK+*iuddzWtnlOHI
z6?sO*%r-)T|64KiwsI)be#1}8vEL=5NbB0N;5)Ja5Xp9zy3_9JSme^qwURJz#-65$
zV44mpy-+l=$@QW4WcN|+JHW}i=FvN8Pya;0FKXE(IZVtSzVgNUPs$;L>pBzOeDjW>
zf}Me1?`5eQ{IPE#{K20W%Iw2a#>)Yv!n)&6xMH*eakIhqRB@rKwAuQ=$Cj`E1w^Bo
zlWVgjf|X0xUsqyo#UI#aLBE7AIHXWi-<>bGBTsbXpz!>r-2n+0_O`fiRu&+asl|A~
zgL}Ai_XUoc+RN+-!=sWWxa@5fAOCO>cX6=H|D+~51(RT|tH4`mS{K8pm-(XlZG0U5
zRR7Wxk)eUxq^)tE1bE^;1auW-n2wnO(GxE3nIg4bW^aVG1DSobC(1H9u!Oi_S4_Y}
zWYw0^E?z%GSHvc;Aj%_Z?xy~$sbmK}=0}m2XuOs=Jv5#Jb|0^7o%E)}{s$rn-?#6y
zoDp{zK`$7;XHHLg3u^H_XKqn&<R$+!82QOP!Ke3SrGE6%MclE<Lc)_v(RXHpV7&5Y
z0SQ30diV{z`S>Gl_mU{P;NMPBCxO4iKn3o-KF)xo_IsaZmnblQ{0wEyJJ&^<1U~h>
zjM5)(cQc`zflRlEW3Ju4RKo&aTIqLwT=K$P-}TG{u0IZaDW^|jeUlObr1>FVx-}Jw
zW$~0gt4|2_ziBAZrF>)1ge8oiH;eb&6OcCJUIG3<6nWoYRO&raqa#wJy6o`H6kQqP
zXzsiUun*m^?~#t~)y#oGNASxluwsI7C+)R)uT)2Z4h~9(!nfd3p)9#ur9e4U$4V-W
z$X!=#_c<KhVMi1W^>!(MQX-$U@YczV-E??d=;V%h%U03Hry_cI<5TUwCVq9WZ0EhV
z^@1=XvxS!+^-BxyngZ9)TvJTXmDpCqX2w^sOUp^#pj1(w)Q_z?+0q@sI&xB6J-H#e
zI=){We=JZ5s*AhRldAt%`K&lkMXD~2A+b}ZhpIoh<TXrntKT?`7#l@)+UeQ4M)2i&
zyF{qwa$Zx>tJ71dKk5C?{iGjoVREgGJ{P8cJou@OsXLdQ=6#$xGNnP!p-#_y`}?u^
zIW^<sflkg38}XvqF>4}%Ts*f3>0FPCp9W3({LNyzKMyeWVRv#h@-+7{$G$Z_id=n*
z^}PEXG*ee>j{=Ug^+HoV3Z^5&p+Gss0VV_Mt<|UsonjrdXKvKGI6&P;+sP+|=j+S;
z%T8H+ET@bnTGSsHGPA$efyVwEWDF#=!o+PMsafj|pe5adk2SVNy76lms2&Vzi&N+@
zJ3(+CTF2xOgwZ|VpPD`fm|BK^9xW;b_vkkgGGmpsLo&=c1OykD=pNKh*y=q?ShbuE
zm?f(vbPB=;a{jQm^$(v1+6iMZ2dSVqokJm89_{vjKBHA)0!8KR#m(k(J(jtBesTUx
zvGV;i>hy7XfCm;J*|Wn<@M{0>t54alKCV3%ZG7@oGU;8Fo$f)y&M5E*xZllkuB0p?
z-Ndl1f&@^IlzJV?x|wG+$)s0d)m+Xgi*G)8I9r1#>Yvn0_JP-?IIB_^_XZG{DTVQB
z;5N^k?ekeF>J4za*SVYdB%fR^!%=)GoIalx9VH9~x(+I7fA^tMoj-0rPy4wE$wqv-
zbRxkrG5cX5o-y;&8tDArwck3(kd%oN#4-2CtB=D2eyDb5z*I(&r_hxF!|gfbPWr?7
zG9_5|pe|uU`hG=dyYt6`=Oc~4mnjZYj2F;3?8I+Tnk`XQl9;wy4wMl1DfUxamsd#Z
zCn?I1-&|_a85PrSQwmfe#$rpy5R*yj&H7m$EkPu8gMmP2QKSHngpIt-!p8M3Tu%8k
z&3-3)Z!bHy9~X)y6XySJavNsBgC)*7Vc&Eiwv_{@Kfc)il`Y!RfxZ2oOl)+Pu7mhW
z#;E!)m~1T=$eLS)Uedgtg#g=c43qxL_ONWhtuI<erXzpWmq*VAZ!-f_AxbE*j+7aM
z49IB6ZB-IRROuEn^@B{34(Q~-?w8WAEp}-$&Xvf;Hy={(iB?Tsb=z|^XL0kChC?D5
zkCk8gj@#qTW*IhR-CO8HM2&rYpew-qy5|fbF#OUz+wX(Ny?U^yz&fB9F$2h_DLycU
z;cB|FaPmmHGSy>Rt@XEYQZ!F4+25HVM?L5$npGEQfg^mA{)*{+QrjW#(tjrqT&x$7
zV<1JEx@h|kp9L4j*a-x_wESQ2@j%(_|7q+kpsHxQzH#a9?ve)K(2az2b3jr$q(f;K
zx*MdWkq`u>Md1LF(o!PG0R*JG>p$pyKkxId@Bgj!tu^bcnQQjm*Pb&oJAQl5OuLdk
z*;Y-{Dcy<F3<53Ja;GI3EVIYuF(ui_9`-x#e>pauP4>Jhbvj6#61q+F^23Ao7Qay`
z0!A6{S^oZfH>0kfqXOOyk699MlS1qWP)N1S2~b)zB)5=JG_>I0&kF`bQx8Z5tj0%h
zoKl<3EfkTJ&5aO6YgJwum-2w>nV^1R|3%CQ>T^jpk04o8&EMZxAKr#f=07yp`BL6U
zp^#Sy!}u<3Q7T&GnV+0QGyw$4e$-St!`P5wSYR!^3ct=z?j+*#k#*4It2uSgJyl!Q
zRJvAngI4<4?62yVNmwLy3&8F3;9oJ2U>M4x3|Fkyci}GzpvF4^(OzY7amh=ik){k2
z7?^p<M?@Y$69`Jz7%|ce9jrC_ZM@Yq@J<ZcJYK#$UbZ~@ioR2#bU;idgy;rIeneC0
zaG-GNQSEcIKL-_0%BtikyR3PPMw2a7)K>uS9AM(bQo4@$xk-KZjjbq6|H|)O&8Q0*
zcK8Edw?&7$wH%H-dhqEda(pa<f;vf@(qQzrfr_n7W4;GI$>=8Zgg!EtcRU*--CLKt
zmp}b|X@Pw1(GN)g_t*N-@p1d%LkZ18r>{iAIHgYwv}o~DRNj8k!jMM}+~*F1E9%iJ
zbH+Yr(^DdWK<Q0szagtsE<YndPJgVU!0_-e?Cne@@N-nuuJ-eY)YZ{5V@122M$+xz
z*L7H4atzxDN95DCa5g86?{pX46~1)aQAk?YXiNCik=#K&z<0Je$J-6>5|bT_Xgy|6
zEks|o2)>F)D|V{5U)%4vqcAJNIYe+@n#Wr@BMHb7nS|%0-*_)lkmQAaGncS*rtTGu
zUXmg!oHnQtA$Usf($|4HvxGVye*V0Mg0{0P{pR~((pd{mP<qX0tnoN+I$D@bGuw6q
z_LoY3#8J5WAb=al=Md8tY3UYK9wtHls}s&>fUQa&3gmz2I3H(gCN32tcNw}{xN9>@
zzzL-@!J42m!6POoK7((j-z@i)%qy6Wz)7ROKl(1vpDL4qy*nn5S0tl?Luj2$9|jp`
zmO%}ymQlb4%b}>xAxd#FM%WXU`{;GglhMN3bWXjc0OBZjY<cmPA4%wTOG5J@ZEkqG
zZHhh^S+n(S6!w&1@{|uJ=u=R$`TO%g-a~WV-i9@u(=*g%ZgxSXN||Y#X~bI87*VC1
zPVz65!oI1t$)sU8>{or?v=O3EMFsjlc9QSfKkAZU#PRS*77ZJsupXh;L#69FNxz8$
zlz4-H5xWuk=cwww*{0Y8-sGK`GNIV1w)?Fk>w0yR#iP<Aoh9ivx!~JFx*Xk^I%UEC
zh5_oE=V9~IYRkWlbh;CF7L!xQrN^51!0|8<rw*eDrsDnii_-W%=RV{!K9crXXAoQH
z{W)j2yvCEmRe_5*skWZXte{_~e+E|Az8JflMfkSsQSX|d1ijcaLE7tl1lO`-Am<-z
zC7b`fScz6wk9=5#nJ!C_E{6h@CCLkSs0cBt2=j~1rK2pu&<N=~7JmebN&n%?E!Fir
zT6BA5B*YUR4T@O~7sx|6^iJ(ayMFy^?;)@r=u=3+rQ1o;^7kS^bAte@vA0}5uSo^E
zom()y;j|{9ePf84OBCU@@_saXr@X6)*#n>VU($CnqPOPzG$3buGzuq8(s{RPOZ*HY
zUT`UJ$+0)5zkMv=B%(I6++<~(t?$c>S^!NTc{P7)V9~%1)^L{DxgV7i+J}2G`O^rN
z4?F@XfklRUSn5V*=fFq$)sam3;THmWfkiO~Sn8f;_BmIw-{#43*4n|ZwneNp@(5@K
z6|wJNnc0P8)y6zmx$2q;fPmEBx3Pwm%+4<j22C%%`Y>H)4Y!X8oJG`bScP879?g;E
z%={Hl_eX%}_Py-$9D3V#JOa6(+ABKX{Pw`aW!DU0z(SVC@cK%^uTwD=7kJ3V7S^yM
zNQ-?>>lt`Ixq60~ebuGxw;8hbc04YKS>V!^;$sBeDY6la<CC5AHh|?PTX9*Bqfu6O
z`@SIeF+0h~f%(W2+eg+dX&!giy6-~-JbQ&k2b>cTJ&ChQVALpU;liw+M1jf?Zsk@y
z=jgyzh+?ED-%yH$3sbU1BDUvHHs^!q!>ujD+U9?F(k}9b)~%&jk3_RaW4v7a2Lnq*
ze<u8uX7VBjoGrpGQOWa>gy%$#)dDreT0ieU0yQo@CkoPT^X%6@u+KE{{EWu-w9aDc
z{zC$QG>pTo8AHmBi7)?W7#!UgRR%y##Mk1lJr)mv<%Ivp1C;~$C-PJ~^WMU!x!3F=
zeI0)^e`K9}n|HikD$RP+O6JmU_#kC3EdA<qF|izjwCmokn6q<v+C;ak*D@EF68Tq$
zPSpXZKf@a9aGScX3yOE86S}r6ketHR>nNIF`q4#LZRWznU9_`UJ#|d0QU{AA?bzF4
zjI$U~47(y?n7x=mB6w)JyJ#6bE`3YLzv$Nh_P^#5LtQJ!L;9HSv?h`gyiM$b71+Y~
z-AG3|eOn%oEs5o+w4Qa4o|uMyHKzf}@d_f1$FQ51Fy@t7H38FYceLHEKp%9-z^J+p
zh!#FP7g+rnz<7#(rX=B)q0*G4>M$A&RV)=_{FL&9{Y<rkges%c1HT6yMFCliag1Oo
zp<21sOqwnNMU3$X-0JtRMsreYU6i7HA}?;qHP|n7sq|8QpXorUo#~J!1=x#ztOzou
zr@Uqt_GNc~gxkpXGrgx~!BfCR)Kjp;;Brs;GNyHr*zYZ$!6Zu|n&$z0V6EbNXbZ}K
zOy^NCl|9ZOVD*x}WSXEOIE!vkbdXtNiYI5+Y?Z#-s7OL&A|@|_p+_mO@ydn}V8&~i
zha)o0m`mN@`B|AM#A^Zf+fRcLnLbc2b)w@b3~2~>NWS6<q-h(s3`b)UhC}8xncsF0
z8xLCw$i}6V?_;ku&iZGOGI3X054W+A8A&KJFK!5(evVqW3-BSSjufn;CM`>q9M-mG
z=L+lDr>%~7Sx4Ihi-mgps<lc2Eu>Jl$}BEQW$CgW@psH6K%q#n%^YuAiCO)sxZfqZ
zM%;pKTb((Ru4fvXI{c>c*W8*<4LUB2Km@Z7E@I7q#+)1_RuL&z`RlT(qKufMkgy8=
zFfX{<jbO6~;ocZ_J{Li<TcS3JP}NFP*$veFG!CBhMmL;y-8d~MnYsgUi!oioOR@gN
zCGA&kXgPrHfN(q66PIqemRyzg8kT4N56%iv@?p9@s0{sB%`ndS0&|AwYH2<oU8Q~H
znUg6pL-z`lkh5-Y$4*1q^D_I}XhKl-nQkHA%W=_;q@Bof16fw+zu|;q@|JfzARq6*
z@q+b2C+<q#^LSo<pB#-2QaD!R_76e3ng7ib`5LJFl@@2&d;_~5=dU*N0f*NWq=heA
z$9d+0LXU|`3+Qi<H(2F5j;{<SG~;f=Dt`?x_WGyGh*BeVp3MY5z8aVsO%2L8t}iWM
z2AONsn7dU=NGFcJRTmK4#Q(N4kT$y<7~{DC<qIR_UXH#=HM^qI?zAa<B8_tS92f`4
zkQX&y1<@XRmfR62b=(<oTy)tLhxR>A{xkiZ*bM1oMCXn>Uykf+txrw(UD0Ihi|<*V
z8KZKVD|1|Y{d_RzVSwpqZo_f$L$fzhuMT&*8+8^v*PP!e1aTH)uk5<mWV?-I`7Gfx
zZB2wv1Lx|ZV+><k5&kI*bc;Y(vhs!V9p@lTu;nb8BAksnj8G)~@X2V&zPkv5VSlGZ
zMR34i!MRR4X9OQxltYEuMeH4c@mu)ANf}9j$4mm9z7@(BAHTluW3Px%b)#6;!KW})
zx;T1UGOyQyO11<#Hq(}r!(p_NdFK`+;U$K7)eBZE%dJ442i*H?tJ1&+h>(O=@A?np
zH5-Y*)WgNR1VzyONksWF-F~^tqtIQnd-f(BI5Ashf36l-;5(3-sj749FGWIyZn=3C
zFM7bg)-js3CUxXtGkEvdk+>*{Vwd0c^XMn*5!JO%4^vxr`B|DrKmCfkA8r}Vx&tZE
z_Tj6&>30V^+H_^WkRK?NkblkYnGT$p;@PhHMZs+uTVG!05M(KE%_blqyF()DQ2LBt
z`xpNHDzl85SplRha1C5BDuASbE8Ihl1(4@~z4E8onSyQ)SMR>rm2kLZ4NF<YUdJ15
z^7A<A-e%@lahlz8>IJT?R;JUL(NbxsvLxoy^8lk|*RI$-sEpbDQdSd}_YAi0Mn8Q*
zkQG`F=zVdNotde7+i~l(t$_7y!cI-10J0U>+i*-#KWaB5Rp86Mk$E%`P|0=bgDAm!
z70?U+IrP;R?-q9cVW=$_zR0an)+kM)8FB6MeDH1x?~CpR)k!X7Q`l<KwU>8sNZg$Y
z7|w;L2wVMfSzE|Sznc^8bK!z;30q~|?~k`=r|1shkHoQy-x?;L$%RA;A8E{^?%4ix
zRxML~r^r3e3kOHq6*g?K5l6_Ea(&6XTkK!#J-pGlL_M?pIne)LA-<@CLJr`MytR$r
z9|lPvUBXuMu4}KM3E*abt}pO3FMj7R8oPJjHXeBde7_&0McB#>B(Qbeb^{M5zVkCh
zm_I)#4tw{W0h0Weza`=1&*ri4(L1SEyNOMAiS&ER-7RbPb4=fh=xzBq)TXrWz`;!!
zEomb2^8nVrFw{mKK*rjagBDdyxvKwJPEH#LXO6tMSN~H<g!^ZT6~8l6Q*2e(V!Ah;
z&hvr2FaC7{!HWjfyjEZKi!x9v&gz$>k`l`&n#NJENVnTuKHcE7U^Ev}xOCIBV4cCG
zB!Z*%(Sxq{K9B1dpnkPLsSXFj*lF0GP}Oht0hI+2F{}&f_tR+m>YYm@cU?aaGINf8
zRfHW2e{lY!zK}X6p19&epVuscxCPmEr0gt`KLgU0xioPn0$XI58~O(~?4~ypxqHj!
zxZR*h{&7udwt}u9GT&0q5IQJ>@vDe<u#g%w%F8g#k%C!ZXB!vR5mFFb5MB_f*_A7D
zHl>BssD*Yi_8?(Q!0g;^=K!FC^Y44^(=ACWa^HumA@y#!Bco&#`nyNGh#lx1$Q{@n
zhyZb;Nra;tjSEb@1!dTcY8S?wC44aObXo0LaGYD0a`j8Xdbjy+Xnv|Z_LK>Al=?D6
z+fgDt-b=LeQE2=4mere9_8}J?qgU<QusiZw<aWNK;<O|1)#f}_z*V5?57)}3n|9Bg
zvjnHR36Y7fmB-y3bV>QL4ZZfRPpNy^`{iQZ?{8T!L1BU|2s2A>M)lE7+?Ze8D&7QL
z1$O^%t!ch#PuV$Jb?M;ozT!q~2A->w-Mye6<c=b#!xw@n%p&JnU=hRgXHkDxplwGi
z)N#CO=ZIDD?s^n`1sFb+l^BLe&Eh?Adl0)sPaWFrhMd?15y54p$0E_f(9^<<s5~Yr
zNMPy*|6UooUV|5LJ+oWf`;)qo)K>TQGK$DeCB;sxZ$&Nt-@E_}c<e)U-6%(3h9YR|
zONb`npk;jbm_NO2;?7PxApGrLcL-iMdB7;yQfhvwKQEp4ceVoKrY=viq+zKj(QvWl
z;k_=&7CbyS#k@ut(jv*;6+W#TmPAPT?TEb55w|NKT+ntguU@cQcjL=(vvH%{iB;=6
z`I-0C&bg9=rGfCt$ETeRZkBEAbL!(Rv0l0jx#0^Bw^=NNw?3|SyuChbuZT@s$OE?5
ztj`Gf<)IrpxzZE!MHiPJtjOR9PH6dFXwlz?zqs^zks%Vbt#<@n&zMDTo;|2+8<<m{
zc4_hk1*R=LsXSIKi0+Etv2_x*YQyspW^Kx?Ua;O_aTKO$+n81FbNS$viBS77>CJ@n
z9MQ=ur24k(h2ia8t6kzBwkq2WEx^(sK1p)jzq(<xDbK2Rxnz5%)wPMey1iKP!L#%l
zH8z=JnsX8J;;+7P;z;1}xN7_UeR#yh%ZvYgZp;GL?wXmfVVkQQdFNY~XD|Eu*7bx_
zKGGp{e{?PVz)TmnIWVREZdW9<-2C;l+)*~!1)*t!1ei)9$5~)h10iShPJa1=(>!sK
z8C`tntY)PjMlS+CIW-Wtz^tmA^sKq7!%pfDEW4|q`p0D2X4GF<X0eS6AM=aA#~o%7
ze&`tPdRry!)KOuAfs>leR0oam8SGg<>``>pY550FX6;h8R|{8A2{G)R%I&qj<qa8;
z^uhLZ&6N#^Na~;z;Le#(8^QIf(E7Gkt5N!ipcd9kEj*ktbh5i=OJWsKsP^G3A}Hlc
zLjg;wr!Ck|;T$GNi+8aBcbtCsC$A$s-Dn@dp0prmE)51){4<Wty^oyV#qWd51K)zv
zbw<JOLh-t1x|{p1p8fBe*t7-h3|K;t2v^+qX`e`{YDE%TD(LctNrwTrru&d{k!vV*
z?{!&i|3k|sE6B90`ywHI7!n$A_x!}87XhNxb;g3!nl_8g;)ug?qR_|8&`xEM&SgaK
z&Q<5Qdb&v0)(9htwi#7>edL8E;r`OZReN5LmL^}?1XNTCr63Z(y2Axfm&%tGyGLPE
z`!}`yXB~`-ccfhVzI{)-IL7FE1xMTyJWY_OW?;R7b?!J$CUKNA#l4Lc?%WRZ>*O<}
zy(K3Ssy1O*E7#8xR(^aETKv*+&G;s)9eH>6_xp_AqAd6&?fjV$j+?g$74-~ipWsLL
z1Q(MMs+os<XL1DgK`T9g8+9Tq5qjz<yo@gJ(H+M<fHm(yjgtu{6%T?;LZ&Q5w3CVT
z3L_QS_m=+NRq<y?jfLY0jxSay7g`jAY0=*O*|2+TMLCnzEBMCUFX@|q-p>`vZ|CFC
zQ)5r}mYlhkH$t?Y^u2u^MdUNsy@Fxx!{#RZ|5*k8YyFzSMA@d}$Y1-$ihf39tGD*j
zR)A*)L0XKapTz(@&K$xOQdWnZ(T!<<@zz{0oc@U6Za1aE0p}B1(P$hrpF0-kpP&!(
zhq?F{gq?6x@Tx2#Kkbf$p3Zbop6rBtP7Vo#`+G$dE$dQANTNr<c>(H>5;yF`Zj63e
zTorn}GA$fIEsP_w^A2M|*Jr4!S_o)x0RdS(4zrsu?RTV`g|w7lxc|pVbfdJOXTteR
zLjK2^fo_n=US;(M?w8!f5Pvri?2;2f8C}8{qH-^MI{&!{wgP>YUL0z---W8_-<_HP
zy*&V&57G%f$LaVEulT6EdRBEjQJzWv8F|A-o`=bArb`Sivlj~u@_KY?iboZ~{F$@o
zcl;S)=^qkiB_0_+j0x4`XQ9U`9p0)O9j&69D+)Zqe*Ba~czzf&oE6W6PI?{Zf3MyT
zQjE7rzmmRAsEFYB4Bg;H3_ZS=1Aud_%JAU%`HoTbnV`t-k_7C<KyP0zQk}f7%vAr}
z@Y}|e$INd__u}s6JDY6sv+T2xzVpMGWx?ksI%S#nqRLcw^!zF(RA2o>-*n)(n_i4v
zIG^~M`K<l;uq>13bHK)`DyZllqN<<a#Mjhk8OI+g&mavC^!(&VjE;ev)8|X~wy4wl
zB0|-IqB;_#^4i}dG?wg96bXE-HR7$#g;~%;p@*~RTjuH<cT0P*ID4H&e_dK|Pav5~
z*bZECRW>f{5m)$0<#ifQaRzwWxfnmzA)l5Wjtchae*?IO{!a&2N4y~mOp*b%3}`lP
zsE3u;Pw(h|R#0;kVQR<<P6kU~U~PdXyO1CUPW%P$n>%^gA4{|E`m5`!8;T3;JM=a}
zI###?bIsh$hh{{Ee?C&&jVvEF484`U<MKLk-RQd7e?Z&Cw67Pi#Ep*k?7Yh|N#C$(
zqWBeQO@n;`a9}^K5~oiKG`#!p;~VsL%J@AMw~<_&UHbvGCdo_w_aTTi#P+KSTYk}v
z=4Epn=hIV_*N$}Z7PyUsrJxAM$AvXg4xA`Pt>YijkQX@?r)eq~Z}uvJ(P|{@Uj;dG
z3349kXEzUi((A85)9pE=OORw6qcq!i_$>$Rph2flZ>@%ees|ggs50;t@LG}KBr;xU
z%ZD4IEOMOdcQ#|Tzmlt&s<c#8-tG;Xro5Bs{Sr%TU-lTBkbO)gL?5#Ol)P5M=V~MW
zw$ym^O)OGwNqa7Ruf9Tq%16D4npBrpFC4GGF{e+rRW6@vZBHXYE`>BSu6-=JjW;@8
zwWB1(E-I{z>OKK3l$@Bme<UwGsXA}p@KN7)U9mM%FsQAVT~_FE%`A(J8!e6|=Z=Cp
znU+{?QIZ(%SDe>ug$TyS*0@2hg-EYRKcU3NF1}MGX_lrXsrw?P+@mkT(bT1V@FVL)
zb3Q|WBj_ukBMnIh$)}gmu{9P@C!7*XRT7epg;yN3cgmB%x}5Pu#q0;=o@pG8rU|VB
zV+P~-idmf}JsBJvO;GLkjWMx{hcHX|9@qNH<t-@^y#tu6LXZ9<jwT`P_dU_Eiz<s<
zay_nF;ERd!6)iS85&$fF3&9f0I06#m_=<Dq>%@n@n%8xDgq>p~`^}w;mz%BY0<_;x
zMa9-gL9g|Jc~{pru}QtJ#x~`8NZB}=+O^)FMBv(5$q5XgRW4hd96&Fp8N_}&uE>~C
z;a$cWTzg4v>@Fs`UPAP&^NPqSJr<^XoS8u=3;%BL6I!kX5|<{4^x**8ClDOQ?1+1I
zY7d5f8nam(L5Hj*`}(uU5p`&%-2v9p3&c-OyIg?#u$nklr)QKWnlTFgATO9Jz3V*`
z;vpUT3odf+)Z`4S&Kq^L9|jF@fXGMGH?1^|_VW|%QUuGP9#Ijz1BB7iSZ}=IeiRX`
z_q`|TEvX#0`x;K4ai4Zh%(k!59Ix)0pXyhgA8~5`diP)%Nl4C`0!u?$jii20F@=;D
zXijzYLG%AKUElxBuF_-RjSuC8l;$gGJsQo-u4QQJj-f=Oyv`^lzH^ej|F7wKE<M}n
zCj`o8Txb`k{S|hmRT4Uc#=^4KW}H*Bm8VV`uP*K=jLYWgagZ08=JK((RbIVZBz^L5
zfoUqAG1+0X9qPMFhx;@`QsTEL;)p^L6!6C_8_cDPVeEtR{*sxOEZ!L?8&tet(^*qf
z5OsWcmWRA_0(~=x7bqI5Rj9ZR?iWRQS^2^FsgpA@?Xgm$VwGgMu1%kv<rYN?I^vFK
zCWrK?gwVNNYZR#@@<t4h>pSL&RXWFXL%VS{k8-DS_N8NDdABu#!~gD4)v`hofck@J
zm-bR^RMEfKr%e$tYkJZTec}i^ABp>o6%CjAr*dY!+d>{9wDOV(*2=@$Tl)M7OFs%F
zzZ`;Z@maQC1W?`xo<2_qG+O%>r?Wa506AU#+VGzCn{57CX9K((MVh9Mb#X(z^02;x
z+JE~&X8d#Kd)BMR6W=7Vrc=uSEZtS$;@xZ*+BazS8(0HZxi4>^h~DkYVgh4dM74#c
zu1ya1H|wlNNW;EHq2Gc^vhuyH7?1Pc3<@y4Vdqs(evYa3HDGR!3Z(Hu$jUfwI4&2w
zdS0o)Z)tS-Z8@Lx?3jeUyP-*%=D9Kej!YC}+=3)!(%d)CH#AiNz-S(%O3ePos4%WE
z^JcK}t2-Py8<h0!$fj7$J1l5D0~w3qqcK@Gp#kPs`oq^93IUBQr_n<N<yDuj51@S}
zU;XR&^tjKDe4KH9S34TwQJFJ2A*S^;JB73en>$w-eWz*bF;UtwC;Q^}Dx$EtwhSFN
zxX6ZETxaT%E_MMpg=uuSRv%Z8<90EKV1o%w`r{W<wmMVOi}9F=ol+!Um`r`S%kax9
z@6$>U>-(MQH1MYCMw+@2D#!inzQDnM+LRY|RH#|M$}%5n+k)c_e`FH&2KsW_dNP3E
zQOIpC=}iK+=AT>;pXM$J4rx#l<7mmR8Z787VFmEtLVHD9QmUT4j=91sm40}`uz{xx
zC|$~d*>k^#Qij|I>eLG@6+N;aC5Y=m(MocM<f+G6TtxI64CA;PzbNNWL+R!a#AW8?
zmT3tO$377>4;Jfe;6W^v3gr-K2!AI(jzSg#Klir+C&G!bQ?iyr)0z6bRKbkZK%V*|
zxe$n47H@2bES8xJ^NExff~Hg%y|FYq)5gXKQo@w8)18vOOq|6x{iZ_`s{wyY-(0kX
z|5{3xHg$Zg3_9fn@k+Z$Zi#v*`NA;MEYE!j2g6m#M~)!BA@ZT*F?N18mOoc8GQs2x
zbn3R%ha1r+x$1V&=HLa&?_q9Um+HssE+K%s6-sXc$zPOKCSKqNf_UfHhfwZ1F1?eQ
zI2Gah?Deb6Q|FNO@1X|odw<u5TCyqWNgq?aH%4vhvKWgpF2rkwlmG7L9!oJck2jD(
zp;ptAZsk24h6o8!ho))axce({Ps+^iPdtdsAc4NA>U^aw!vyUFO&1xQ(F6#MUdH^Q
zPCWw`n-9))!^-K@YDAQq!o=3&<b~G2v>HQM3(G&kDI`MqD;+AhaJwHjAzCe97S}$_
zSko4pXa719fnRb>8dk|Ap4r9cmM@Mn#~!SK@IoOpNq4H&x+FvCBeUMkNskXl{>U(W
z=LjPH<ML8*<^}c*je!^MRP}(48`k%^aB+(94VPRE?;kG)KOE7c7wU$-CfrUpFGm(<
zp50)1JrY>C3EVz|GLLC^@4e6ueIOPw!~RKQ=_O{ZuFvbPE#hb84Pq!$hWS-*g1D<3
z5RG2MR_j{7X51n+EbkY4GR^)$Bjx3vA-tu4s>?%7c|A~kkR4Vu@Ru%Kt=yW6)Wb)O
z&nlH7Z79pS_aQ$i6NIj;N?MX~m9e64!}`E(r4m($B_D$0G_*mT_9JT^q+!X148$wj
z-m{IP1sTwtFTD?hUnlh75WLl$TI5OEv=M0PT|geW$tvtc7G$Y_$RuDmaAo+-<!ZdU
z&Yj(<1H6cJy9NWe@FJ}e2b_l%pjH|OSq;6Vf-E(V=LyP=YDQKQlxfp{hteTM@dM6W
znHnzVNe3AoRNpB3=HCmle1N1R#5k#0TQO1*eb4Iel@}tFlNv2g6PKcCj2<#YjOwEe
zoA3Q4L@FH7@2uux#dCes@_>*l{7K<bJRqsYdghW)L$Ffv;*g0w|6Qa(+=iq9))s77
zB7*9%rrZ+CERMLjjhNvCElG=G$(Sie1WFx^ekTE2xMUp}Umf15907H(r3iWyy-VC&
zwnqN-txr<78}U4mF1fez4q`J6$=V-OYCafhfPs^&&Ps(^T*XllnO7_6Ug`oAvP<CD
z!MQ*S;)d`j)IS?@Md;1ju`G@JyUU7W!WN3!{Xex)B=02#6IE8%K=#uXG{~og&)N~(
zh5Y4Xn}`vka8aekZdw}1d7xCYompJD>@L2CVz|WOwUjOmRk2)W-zL6ul)bq84~9A1
zaj)nmxdxi?n1^$?;LAuf0oIPy=HiSNGQ~m~$ZW6g?*5TH8j4uy9K#g2U#U^9ibiKA
z*&zxl{?AbK`!G_PiOv4bfd0jU=7WsGooN4jQ1n+^L$5D8?yN-*UvFP{Ya6uFgzmX1
z8b*l^ialHyM|U@~o+o9#m)Sy9(><_@Zxux@{z&d6Aa;S|SfbPEjo!}W;+Sp|#U*b2
z{f{v8S~(vYyvGeRN$~f(A_Hp5wkmuymfJ0<8mEBkr>Ii#-~6i@<?LxxYT5JVD($-O
zOXH6un0whF8ZquF@w%3Igey?_Q^wcV4Oa?@YRWGiWW+A~Z6i2$Vlcyn8%gFXM9WMy
zFaTc%R*@dv4|%Dp>0IU3qRd~}4K)Uw9X!P@E@FM~#x-KRR2p=PUr%h8m+~@?v8SjH
zc&lW9^b`8!0_*I;A{qk$Ds6dU=^BZ*wH2bfG!oT<y(tcPx`<hX*Xj1<RTBed07*oT
z<V&5UbNST#)Z_Hh@`sXt^nBi-I|uBgpNf}d-%DxG#o74fs;^M?lwP)I^e@>4yRxRT
zpUQ-8lA0Ic1t@lS<JaAJPX+Tt60PBm_hxG(5$k%BBx$Fm0V0?^ae;^*p?;?WTd+DW
z-Okg5A_>1BFg<p(pm9=~!ts1B*!d3#_$M{Cz}t=hz>4OL7IKt@)J?~&$7H|*&<#do
zn0bI8sT(P+TpY;`^;a0GW>TlS2tt|LLp`2zgUP%I!v1dUieQzBd;5x`l~MI{FDB5j
zkaTc~I4@Gwjh0gPB%@YgiQf%3=Wl|H^{907voujeZRg7t1mPvdzkL?nNIr<baOLIx
z*9tg8I7iB@C-3)z;A1pWze@qV^@WST7xbq-8C4xOxhw3#s4Wm!0)h*+n^iCs(mXl5
z^8p^J4=*eXp$MBxWh%2Bzp)za65A=Vr=9O&3`w2#$?)*wR|r+uS~kHPNKlHx!3TP%
z6@32;J+4C1p(z}EEK>!vUlsY7;o`X22MZwexM>;VU&Hj(g@ZnbEVDDAvhl*+Akbp(
zo6}=vhS;N#LtxPe3fR1s!30Re<b^+j?a{2I+z|SCd20>73ioKgw+^mGvZkS*cwGHS
zsRdyiH`$hc1fLt#9s=q~iyiI%cphn)HhB}a0lj&-NIq3>9Skrd$y17~5G%$@fX5X4
z27<31#(@M@fZVL$5t5{jLnG?Y8Jo4>8dbNwyKEnjWs}<~FLi5|8)@)~oO?1?&dmSf
zb3$EzV7Uhm>z`9?N4~1O6sLPK{6oe%S8c8FISDo5%(gm+qz5U^HPC@jJ8_t+Q7+gA
zLi`5U)ZC8A!RykVMY?bNiN1>D1j8KID?nFp@-gT)?`vG)D<Cp1Piwb?nWr>$=U_~I
z?>r4#aE0Oj33%qiuJDDoq(LN&obl`1%F90SWl1AAh6IYYgX_mPDCn)G{X4QQnGRR?
zWdRwN9S=t``>RmsjVEwxP2*^mQgwGownAVwS2p)(%B!+Twb8U{2XZ$z2fb-+)E3F1
zW14$?;I-6<RVx{UO#G~Ee=cI?(I+p8#<(_9pPe?{6P~^3s%2>onMN|%fY%xNSJ&TZ
z?(B(EeEo@2Y}$0^AC;V*BR%T%f;3@Oedf~xmijD(|IPyJAUc!TgI>F5_EED}Yopl%
z@62Yi(698rPFwakN9v(JR;Q&F%~{(Se#-Ul96D#ZwgvoH{h{&dUoCZ<$*kJQfAq}K
z3eQWUx#vo5%$BmTl!>uVDOYMZB&2oaPExYIz8ZMt)s|UxcShIi5?VlfT?%JKk#3bu
zxdL1!yvWmzWN12?5cNO4yjH7XXMTFx`Al=YTO9IrR=Hj6&?c22jX2BXRn?t~(9dZ4
zH011u{lBiKme2Pey#irGdmXES7u?yC%d8vx%8|-^Y9FjU9IFg3xHdhDveIuK=CZ^H
zv<^f16wOB-da{_Zu_OieO8H{9Kw71zoX&yzidd60+w)xO0gy6VM7couQZ3bSqRc!M
zEsZ+oSMNhTmZM5rM|tsqx604d37=1(*rZCO3FN+z%pyK~m;wq{E>&g(ccfCK9cNc-
z3bw7JiYvcU_E1lCOzlZ??-dmoC{o5%_pncOOnXOgzc}T@lYbYu=o&~=i2($Bxu>Xo
z$Tgn}Dc29QEYV9Hzl<OPeRCao1`NDdK&;0tFh3x}A#qDYB8a=#C*MKpl4oT@%-~O;
zA4*e4qF<hKNtf95r^Z1!i@`AL0=j1WSX*;C>PXAk42O{b|H<b2tKKP!$&EPi-jW%z
zg@skFO^hm{0HSOm66swG_g!un|5F`u!_|=Ut!llHfi5DK7b1>y{{{QN_&(Za&3GlO
zRvzUZiVmy4F9zM?$bVsEawCC(=qu5AB|)>ZjfxA5`G!|$-4f($oxFIFzPv+WORI2M
z$R|Smkd~+(Nv>TBcRV19heu~vcr!v;iaHbV0+&Zgu!Fh^TNN<4?4F=o>)jVh{=sV&
z*(0eIvea0)CVuyHTOrxP1>4g(3%fg%j`X??9`zg&fd5dEYr_H)drw&~Fs_8O64%rI
zJi3WK$s!U**E|cco0$9!j8`@CE5l%lg~5JzpU&_WuU!^Z_dGzp*2^m#*(x*?ZK3dY
z+6Hnd&`Q05xtUg7jE_{PRCOOyu=MlRMxsb_om+I`eGL$q_?YDow5(_;T!rmv4=Q3!
z&eF|G5jnuC7HPqUoo3PUr0?bS6}C%Rj&KLQ<^b=!ljE^7&wozpKZQXAp-@$zcmH~I
ziR=;A{tLlsgi!1brN#|!tM0$X%nSArdPId*(hqvAEsMXcI9C>SRWZKwn#6W6y4q#a
z+YHIme{;7&WR1t#RoT~J6>v7{uobrlhgz66TDbRBEH}ccB2-HHT!-5VW{#Q&5XYd-
z3;pJXIs$b4NdzcOqNv=6Tx5`6QazHhs=A0<_9=<OZ-(_133mQhWph~|g6fDi?{{p~
zyV9JFn|D6P#QB|`J_t{Box|F0HD0_gjbk?D+OSk6Q~hW#(UbbFu@3jk?2}iIj9!#M
z+ty07u~I%pPd@qli18&sX&k*N@&>@DM8@=ye7uq7e4>%a3qj^1`oxpuN1-nu#c>3t
zp?i#SGda+<GO?eTQl?MnH{P8EQ<+NqUX~%l11ETe(&?R2v}In!J@kYdamP}ZOz)%p
zG;{DHmly6Oaa5-BzZvDoC_eU&J$dy`Az7b{`=kAMPuRN;P-7E<z0TA}886-grE%e=
znZbyr;N_hBCQiV#bc0cy?D<FJMdtHIo!%JjpcpcpotW+e+E1IydSoe}h$nIH#zN8l
zQ6m6D0<x6~;cPbK8%njvz^=E>V(#Xmuw^=Gt8CgdwiI?5!VD4$?Z|YHzLgN#x5?nC
z56E8q%Ee?$0r?tj2q+94ntMro{pxQe96Cg3tsh_CkX6>h#clf>q8evtG&Er;)V``?
zo`G2Zj-dXB?#}DIxJz=#*G$8l;R(wO(*L3kF)X?la0zN4neI>!#JtHl{s^4p%XgnI
z35G619#$Dt_48S)y<Eb!tq|2nL$V0N0RYr4gP^{w@|5n*394COcsVp-H(loN-=teQ
zyr8D;qxtHMixR|*$+j72^#ZxzF__t2@PT_2hWIQ&8Sz;*U$(I_=kJ~M%_&bGKoo+^
z&>wJFQoh{yUv-OW1S=!D>WJYw1k&<3P2f|E(R8Pu0ito?gQn0sYFcuXD|58%@bk9&
z+6B7Mn>o5`R_#UF{(A&F7f%1b>a%}smvRNzWfqK7d3qD+vk3O6SXSI$-~SfQu=Bto
z8Ogi9ffph4%^Z-M;4O~Viy_^DuIGjY<owh1dcAlKD_lkclZgh{D)cJ5xZ;>WB9yZ?
zSAjNu>?C*U--RZwrJ!`~?t0O3%d#%&-Qu=^{ei#~_eFpca~SU2wgI*qOISX<`v11j
z{@k7&gkD7y#9S65snA{muRZtr+9k;27={qX%yI>#q~6wxh&Peh7m<m+)dbf#r!>+P
z0=XbCDM#MGX>;&3%%_G-oAnBlb|*A;efCA`LLkHcTC7CGg@A$cg@U*oPKLyE(T{oc
zVsZa0m*MrO+jcLp?fS@z#6{jBi2~yKxej+N{8g4Qx4*Wtj!z(VFQx2WQvI*xgvd_4
z__bh+*Yv}=?~secVY^$hJ9duMNZQ}kxWE1`^F=-Q+0q(~W^I%a$|bs@-fa0zr62p=
z#C@eP`X@BU?@;UiJ?3y<yO88kd$}b|O#;?mK@+it!}-05q^Vm7jKGYWA0s%Sgnh+x
zj-sg}^6CVg8RKm7rq{b9?;l79oXV)D-3n8g{x86ql{T1k!g<qS7$V7=r*}q^-pH{G
zcxCvCMkwE*-cC=_<2J!q4*Yh@za0<A>u~ET&bye3RWMAi&k3<Oi9!T~mq`DX=V5v+
zs52l+6?s(o?sltWjcN>=>`y}nJ`wS}^2_x*f8rqIq#~5vE>HSs|7uTt*-=9bNjv7_
zys}Zxy(p@qr;)~)CFq_%?MALFz`*HjAoD!~rO<P(=lS)7ej_TsMLCPsf0o0pICU1H
zbct=#BE~&QWq?sba!EVMlld#i!2{vym-+S6z9UBKe`wbu1&2KKB}WZUB<!Y+@*;*j
z_sB{=Hn}EjVptL!=jn~!>+o3<>+9R~YV7=aKA(}$by2MRbuOR$dbsaD8`W<4^_#xC
zfi4EQ5_Z`~c{U@S8KR?xtrBPL@EW&#!2eFE>gtrt9ho(RLtlT?a9-j}3a=W-_rC-0
zoFlEwO{r(pZIHg4-~Xp%(R}{`v^_JWmjwp2Z3JKV)<xsW*125r>sx$Bme)k%a{t|+
zTx&gBZ#)m_hzT^bbU&KEdc?H1I8_Zo5ca?<ScT_`u&JZghp8B#ytVk7u2o0j;MaH&
z$+z?9EF=Y<A!-VDlSUijIY<~KZ??};8he9%{eP;Nb$`P7ZW$5es3mmj2rD;gjmM<m
z%M1_N!LzPc$3JxwRgl!(5B6<$J^E8VpOm|A<NxdYK;H-4qhNnv%NdJXzC)~@gSAXJ
zB@}G5h-Bq0FjqDX_dsLp{aefQK^4|n;Q1M=Lf5H&<u}K2F4WnK(XPJl9O_CGU|f1B
zF_o}d%!@b2HpX%zw9)+Jl1JyJvm=r57PUdSL;ccO*2Oq?^R}bp{VQx|YkMo@+zvWt
zK-@yL>Z)5#`-@qs5THBf^%!BlFagb3!k#GL$2;g5a4WYh6DMTUFkYM>3UZY^{&Bq+
z+`1RzF>k-3KxH;ZKHeKGq|orHdmP%~W=PDUL?r_%&|4(r((oGDw;`sjhH<L)dVDra
z$ywfkNr#Gijxvj;cDy`+z&YODN*<tM`d7rjC-p*SL}w!V+{ZtV=UB#sFq$CcHSANh
zS0O&08cpsW64IPYpIhnvz=>`V{jN|T#MtOw!A;Gm^usgju1$-ts2#&dZOk&1l&B#q
zdMH=-X2dL!kt^GL0GXS`9<nqz<|RG_Y<WK+xKwLlQ;YWhLkJ}Aec6N-+<Byz^ni5d
zd`rc=4Cff3YP^=AwnRCL;Q_4Vkj`j~PR(tlla~qFBr+`IkoVsPg{IsHHy;xdJ)$5Y
z&N3eecl`3@77BQ|Lt&{H8ERv8U@z;)F0vj=1z%@E44+8GCeK~wYyt-nG<yt@CGp2j
zyUojG0c9W*kRm$HE*wQRV_6P9=j+VG$V|a_rrr{-7VY%+K)UIN<$bF<Oo{hYD=}2S
z*W4F(8TYoOScGB|S!dCABj1OHG6fi9`S(kgI=)M&8&P;JeD2Ey^zOKCIpJKm^Bcw`
zs8Q(sOwl&QvN?*Cw^UMe{-5@ZB>aHwijmSw9)k`pz!rl`?Szgqp*+s~eFh;#P3UT!
zz!b38ET3ZY-#ujSbH-Wf5)ZPuwH_E-kX0+4)B>Cw3-_%};XIr)8=~eTxs<h%|AKSf
z1hcO<flWOqW&2;0t`J`~jwm^TI{h)NN)h%TAd_LYX>t4EI9oblld}kOc`Q-1Uz}F3
zlU?K<mOJOQyN~^GaVl>;@1!={z=}VnKo_@t<PZF*MHEPWYsx0Dif6g#zyH9cOE-GR
zSm&k|a3uCc)ujF<o7~ViaO6q~9^OS*N^^YR?eN5;*7?nWfJ`Y2Nwr{|J21Z)nD+g-
zxpE<V7+4QJ1-jG`E9qmtaQW%U*b%ABs4=pnE6K@!ZMYv@=MGoizU?5q!u*-@+t;yp
zy(T8VnZ>Wiu>1xmkSjF58Qt$E<8R-%!u7>3&lhty?+%Hy<4PBm=89MEsTb?MT(0|W
zfph!hH^Y5@h63xppvAo3viH3mK7igey2e!%-+X*#_*l|)>gbcr$Ps8RJ5p+UhL2n~
zb-Ih`k|e*}u;4z8pB5Z7o{?}h1ywTmuDa&x5?uJRe<tVmh44^Qi|?&^gH}j;Ycl0@
zUSdv}E5Y$6y~!h2Bl-{N4UHp5(_%aeg`yp;1(bkLW}gesr?tdh94re0`F=c|Zg1{H
zod4g}K_{?bkY?>f^!T;5CxXGHN4Jo{TCd6!m962&|7jM~wpMgt<4mkN?bKjHtuJ*J
z!G2O%WWf<j9WLkg;7^|iMLH~H(#@Cmw7*lE2zV!o#(y9>SVVYOxpcZ-Y=VM8CBYHl
zz*{yK2|nV&QlWyQ`KXWf1m9nvOL<DXD%)r0Peeugu1l~k{RBk2I0OUEuS(jn+z2$p
zfy7qUkxsBBa@b8&@Y6Nf`#$cLWp1mj|IQFWR3yUgCsS$bYf33U9)JU<n{A3ncz{K7
z&nJn=*8U?xvrhLHGOH~@dY#eY$^9E?s#@-deHeu>VM~J@gtt1%Lg74v^3K=+=2|HO
zE0auxylb8uThy@%R82VH-eE8q#f2;2mBsIK3*I(Q1}{#~s!;43&(*UCUXOK(U`L*4
z4T3L$Q$Wl-<?Vz21awM*RQ+*FBMZ)zh@}vnHZ64V?&9zM!#hEaVKUcEjRX>TR9e)r
zb5tN=l3yCn5v3hK#?Kcg^3h-1Vjg<p74w=6IqEr>yH%!uI&t0Wlm*vW0aS6$xdq|P
z4T}A}CmED8#ttT~mF}O*B5yNJ^OSMG@Bfer@`HyId7MGXXyOp<*7My*RgCh|*x|Qp
zWl={Nc&MiW<&kDL_Ml;-%^+7Xxj$?}VHG<cUb$BkGno`2BT7K8UfH^#!+F=!Vw2K4
z3BinO3yIa<__(2XVp}#|xeyA>AuaPrM1I+P!)(f1c>Q}SCfj~qxyWn-qZvTYM&Xiw
zq9Elq{@AWmnrWB9eN^76prb0FjHAjUC3~qi9tzIY9Y#Srq~)Nn#^0yYZFZ9e)=T-j
za*JTtDE~79xO05lMvCc(!e~h3JOB5|{WH*HcphXC9LC3uV^i4v5@$NfelywPCp6e<
zJ;ZCNu+JX~6vRw(Pf=8kMEkD=ILx1SSm7~@qWeY)@Q<ru@W~}pgbqdfE6nq!=8rGi
zg-d+PqdaR+;FieBPHNkx((j}+z(J`aSw8y6H}q+Lk}G~O*_?qZw5JXk?Bz4SB?wS;
zknn?e86$9tR-z0^95i5Rboq*8QX1pQFM$^<*eDPD>6G6NZmfLo0&KrWY3lScK>O1i
z5BBhZDaR8xa<4bN{At=#e;aVUI{&tY{z*M#DJH>N$tO@7Y6NE4IEqEIl(3<U#W7?M
zFEojqoAf#Bc!C02QIk+HFt+Th&RV?BvH9g2yx|ZN<pfPBIz3<^;-BCG`RNk7MA$n0
z&3IOr?q$gb>V%f{qnojg7@T-=n=kg{2M+~<g4~@9FtDkZXV``ntb%?t(pP9RBw_kO
z`%5ONOfx7uc67Bvh7@h2IDTn+CGt5zLrWyc7clrJUtLuE89vg^PQrZg!r0;Ze5P0j
zGhPJ%pz|fA6sw)T3GhcNS`>REId-+j6KT~F3eMG_^f%fYNh$#JvII&&a%j&y=AZtv
zgpUr?h8x|tflKgq|Dr-Aqr{p#p#WmZX{OL4>4)2pq+MePeM?>jW0O%R@mHBBNu@1=
z{oa^7ua%C@u@N0i)y|0bum@#M(7n2?0FD*Qq^pz@f2t(hot{~l@x?bi$W<t~LdD;Z
zR!@*7bM`xqm{}gLpKKU$U0M}W9ef#@zIDPT-M}lDC>e|M2lIady2r;uNUfw`A~}m{
zY;NOlT{Eox74<&RAlVMvsYCMWdNhl3-}oU5AujFZd8rsk@u%UpsSB`(-jh;XA_v=e
zwQ!I4)Nb@9Q~!02xc7C^KiTDdLaD2jLfLV6QPGOPvBdSaT-C+=C#0N@1Y7RcJSGVp
zF-@YfUN=*Wz$E`R?kXn$X98aS=wQ+Ni*shV<VQ!}&|l+6lMUO##rtuS$ARqtO~#pa
zVG?3s>aKBCju*5S74ulTQ1dddYs)w#CwBSaX${zCaZs3q7AWyM&9|WCO27P#lNs?(
zbcx?3ng#ur)<0Q~r1*}VmDL3c61wC!IUls1K#Dcj0>yQ{T@Y!9(+47NSjp3sT}Ds#
zWwwXW_eyP?L5geH1Q#xMXEQl~K<EZ2XXuK2yf0cx;TU9%5A4(VmevuofIjgQ;nOC_
z%ydaU&U)TnHF<n;CxJ;+nnWLXya6`kSUEVD5*p_|E({#UBTcd&_m&d8Nt4tL=Lrnh
zV3J#0ot$jt7D|4TBoNrF`>ij?M#dqQHtn!oys}c9D#>7?FI6f7I36!fA`NW#?HIyZ
zxzUz6x_7le7bBVf=8a|mH*KrSajRB99sTRidk)&S2%b^~=o06_Jaqf3VP%yYia!f#
z%K-Is<BS}s?}SMvs!|5oIZ`*HYt2EIXDf|*l^aPbIY$U6QsyIQLl@?z4wsv?qnTII
zi^Rq=<+@LQCMlL)0Oq=4FIT`IMOXOF{Zo_Z>&ux%swDKo&gmDMtWuI4C#*@mjz1<g
zdfacNW>YkHs1g;$zimta6~0f!0wvS$%)P~KHl$+}BQ}o7I+$s9l*>CVa}VrY#dI(1
z4%pI00&c)ZZg^f_kj1Tx-@dTrB}<B($X;9|Ou{X##|o071xU*4#YexCgl#x|3-Niu
z_WCnrW)z-fx{P%Y?tSEyHPYmF6kG3(wqVnO_;l$Wr^CPFxKCJKk9g(fYi?Pk9sNFF
zZY=)(OL2G=F1@z)w<`6wkNCDwP#8>h5bMc=3J}nKZo_hEeCm;0n@$7jtz5i@h7((X
z?+{_}HQT=V?W|lH&%dM1OSoD4Ro^{0hX2C4=8a`*q;7(aLNr13&oy%r6rJ+V8kVQ7
zqi)r|IqH2<sl!*ZIvvz3ugwU48E9tMbu{H+fYAP<-RPfmu9L&L_CL=gA$<0%Z3^`l
zd*+Px)&27vfB`;Zc)qYb1wLiCCK%ZGPdS;lVS9?9aFEj*t3wR459Xv;3#N5LZw>}u
zo7RG;W+=dX{`CL-=cUx^o8=3Xjrrw0!~Hv?lc&EeFl}6{Jzsd(WYQ!kl3@w(@$q6Z
LGe6N(!2JIJY_CAA

diff --git a/previews/PR3562/api/JuMP.Containers/index.html b/previews/PR3562/api/JuMP.Containers/index.html
index 8189241e00f..e96600f6554 100644
--- a/previews/PR3562/api/JuMP.Containers/index.html
+++ b/previews/PR3562/api/JuMP.Containers/index.html
@@ -12,7 +12,7 @@
  3  4
 
 julia&gt; array[:b, 3]
-4</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/DenseAxisArray.jl#L194-L215">source</a></section><section><div><pre><code class="language-julia hljs">DenseAxisArray{T}(undef, axes...) where T</code></pre><p>Construct an uninitialized DenseAxisArray with element-type <code>T</code> indexed over the given axes.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; array = Containers.DenseAxisArray{Float64}(undef, [:a, :b], 1:2);
+4</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/DenseAxisArray.jl#L194-L215">source</a></section><section><div><pre><code class="language-julia hljs">DenseAxisArray{T}(undef, axes...) where T</code></pre><p>Construct an uninitialized DenseAxisArray with element-type <code>T</code> indexed over the given axes.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; array = Containers.DenseAxisArray{Float64}(undef, [:a, :b], 1:2);
 
 julia&gt; fill!(array, 1.0)
 2-dimensional DenseAxisArray{Float64,2,...} with index sets:
@@ -34,7 +34,7 @@
     Dimension 2, 1:2
 And data, a 2×2 Matrix{Float64}:
  1.0  5.0
- 1.0  1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/DenseAxisArray.jl#L232-L265">source</a></section></article><h2 id="SparseAxisArray"><a class="docs-heading-anchor" href="#SparseAxisArray"><code>SparseAxisArray</code></a><a id="SparseAxisArray-1"></a><a class="docs-heading-anchor-permalink" href="#SparseAxisArray" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.SparseAxisArray" href="#JuMP.Containers.SparseAxisArray"><code>JuMP.Containers.SparseAxisArray</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct SparseAxisArray{T,N,K&lt;:NTuple{N, Any}} &lt;: AbstractArray{T,N}
+ 1.0  1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/DenseAxisArray.jl#L232-L265">source</a></section></article><h2 id="SparseAxisArray"><a class="docs-heading-anchor" href="#SparseAxisArray"><code>SparseAxisArray</code></a><a id="SparseAxisArray-1"></a><a class="docs-heading-anchor-permalink" href="#SparseAxisArray" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.SparseAxisArray" href="#JuMP.Containers.SparseAxisArray"><code>JuMP.Containers.SparseAxisArray</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct SparseAxisArray{T,N,K&lt;:NTuple{N, Any}} &lt;: AbstractArray{T,N}
     data::Dict{K,T}
 end</code></pre><p><code>N</code>-dimensional array with elements of type <code>T</code> where only a subset of the entries are defined. The entries with indices <code>idx = (i1, i2, ..., iN)</code> in <code>keys(data)</code> has value <code>data[idx]</code>. Note that as opposed to <code>SparseArrays.AbstractSparseArray</code>, the missing entries are not assumed to be <code>zero(T)</code>, they are simply not part of the array. This means that the result of <code>map(f, sa::SparseAxisArray)</code> or <code>f.(sa::SparseAxisArray)</code> has the same sparsity structure than <code>sa</code> even if <code>f(zero(T))</code> is not zero.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; dict = Dict((:a, 2) =&gt; 1.0, (:a, 3) =&gt; 2.0, (:b, 3) =&gt; 3.0)
 Dict{Tuple{Symbol, Int64}, Float64} with 3 entries:
@@ -49,7 +49,7 @@
   [b, 3]  =  3.0
 
 julia&gt; array[:b, 3]
-3.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/SparseAxisArray.jl#L6-L37">source</a></section></article><h2 id="Containers.@container"><a class="docs-heading-anchor" href="#Containers.@container"><code>Containers.@container</code></a><a id="Containers.@container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.@container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.@container" href="#JuMP.Containers.@container"><code>JuMP.Containers.@container</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@container([i=..., j=..., ...], expr[, container = :Auto])</code></pre><p>Create a container with indices <code>i</code>, <code>j</code>, ... and values given by <code>expr</code> that may depend on the value of the indices.</p><pre><code class="nohighlight hljs">@container(ref[i=..., j=..., ...], expr[, container = :Auto])</code></pre><p>Same as above but the container is assigned to the variable of name <code>ref</code>.</p><p>The type of container can be controlled by the <code>container</code> keyword.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>When the index set is explicitly given as <code>1:n</code> for any expression <code>n</code>, it is transformed to <code>Base.OneTo(n)</code> before being given to <a href="#JuMP.Containers.container"><code>container</code></a>.</p></div></div><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; Containers.@container([i = 1:3, j = 1:3], i + j)
+3.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/SparseAxisArray.jl#L6-L37">source</a></section></article><h2 id="Containers.@container"><a class="docs-heading-anchor" href="#Containers.@container"><code>Containers.@container</code></a><a id="Containers.@container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.@container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.@container" href="#JuMP.Containers.@container"><code>JuMP.Containers.@container</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@container([i=..., j=..., ...], expr[, container = :Auto])</code></pre><p>Create a container with indices <code>i</code>, <code>j</code>, ... and values given by <code>expr</code> that may depend on the value of the indices.</p><pre><code class="nohighlight hljs">@container(ref[i=..., j=..., ...], expr[, container = :Auto])</code></pre><p>Same as above but the container is assigned to the variable of name <code>ref</code>.</p><p>The type of container can be controlled by the <code>container</code> keyword.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>When the index set is explicitly given as <code>1:n</code> for any expression <code>n</code>, it is transformed to <code>Base.OneTo(n)</code> before being given to <a href="#JuMP.Containers.container"><code>container</code></a>.</p></div></div><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; Containers.@container([i = 1:3, j = 1:3], i + j)
 3×3 Matrix{Int64}:
  2  3  4
  3  4  5
@@ -84,7 +84,7 @@
   [1, 3]  =  4
   [2, 2]  =  4
   [2, 3]  =  5
-  [3, 3]  =  6</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/macro.jl#L320-L376">source</a></section></article><h2 id="Containers.container"><a class="docs-heading-anchor" href="#Containers.container"><code>Containers.container</code></a><a id="Containers.container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.container" href="#JuMP.Containers.container"><code>JuMP.Containers.container</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">container(f::Function, indices[[, ::Type{C} = AutoContainerType], names])</code></pre><p>Create a container of type <code>C</code> with index names <code>names</code>, indices <code>indices</code> and values at given indices given by <code>f</code>.</p><p>If the method with <code>names</code> is not specialized on <code>Type{C}</code>, it falls back to calling  <code>container(f, indices, c)</code> for backwards compatibility with containers not supporting index names.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; Containers.container((i, j) -&gt; i + j, Containers.vectorized_product(Base.OneTo(3), Base.OneTo(3)))
+  [3, 3]  =  6</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/macro.jl#L320-L376">source</a></section></article><h2 id="Containers.container"><a class="docs-heading-anchor" href="#Containers.container"><code>Containers.container</code></a><a id="Containers.container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.container" href="#JuMP.Containers.container"><code>JuMP.Containers.container</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">container(f::Function, indices[[, ::Type{C} = AutoContainerType], names])</code></pre><p>Create a container of type <code>C</code> with index names <code>names</code>, indices <code>indices</code> and values at given indices given by <code>f</code>.</p><p>If the method with <code>names</code> is not specialized on <code>Type{C}</code>, it falls back to calling  <code>container(f, indices, c)</code> for backwards compatibility with containers not supporting index names.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; Containers.container((i, j) -&gt; i + j, Containers.vectorized_product(Base.OneTo(3), Base.OneTo(3)))
 3×3 Matrix{Int64}:
  2  3  4
  3  4  5
@@ -113,7 +113,7 @@
   [1, 2]  =  3
   [1, 3]  =  4
   [2, 3]  =  5
-  [3, 3]  =  6</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/container.jl#L25-L69">source</a></section></article><h2 id="Containers.rowtable"><a class="docs-heading-anchor" href="#Containers.rowtable"><code>Containers.rowtable</code></a><a id="Containers.rowtable-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.rowtable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.rowtable" href="#JuMP.Containers.rowtable"><code>JuMP.Containers.rowtable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">rowtable([f::Function=identity,] x; [header::Vector{Symbol} = Symbol[]])</code></pre><p>Applies the function <code>f</code> to all elements of the variable container <code>x</code>, returning the result as a <code>Vector</code> of <code>NamedTuple</code>s, where <code>header</code> is a vector containing the corresponding axis names.</p><p>If <code>x</code> is an <code>N</code>-dimensional array, there must be <code>N+1</code> names, so that the last name corresponds to the result of <code>f(x[i])</code>.</p><p>If <code>header</code> is left empty, then the default header is <code>[:x1, :x2, ..., :xN, :y]</code>.</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>A <code>Vector</code> of <code>NamedTuple</code>s implements the <a href="https://github.com/JuliaData/Tables.jl">Tables.jl</a> interface, and so the result can be used as input for any function that consumes a &#39;Tables.jl&#39; compatible source.</p></div></div><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+  [3, 3]  =  6</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/container.jl#L25-L69">source</a></section></article><h2 id="Containers.rowtable"><a class="docs-heading-anchor" href="#Containers.rowtable"><code>Containers.rowtable</code></a><a id="Containers.rowtable-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.rowtable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.rowtable" href="#JuMP.Containers.rowtable"><code>JuMP.Containers.rowtable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">rowtable([f::Function=identity,] x; [header::Vector{Symbol} = Symbol[]])</code></pre><p>Applies the function <code>f</code> to all elements of the variable container <code>x</code>, returning the result as a <code>Vector</code> of <code>NamedTuple</code>s, where <code>header</code> is a vector containing the corresponding axis names.</p><p>If <code>x</code> is an <code>N</code>-dimensional array, there must be <code>N+1</code> names, so that the last name corresponds to the result of <code>f(x[i])</code>.</p><p>If <code>header</code> is left empty, then the default header is <code>[:x1, :x2, ..., :xN, :y]</code>.</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>A <code>Vector</code> of <code>NamedTuple</code>s implements the <a href="https://github.com/JuliaData/Tables.jl">Tables.jl</a> interface, and so the result can be used as input for any function that consumes a &#39;Tables.jl&#39; compatible source.</p></div></div><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x[i=1:2, j=i:2] &gt;= 0, start = i+j);
 
@@ -127,7 +127,7 @@
 3-element Vector{NamedTuple{(:x1, :x2, :y), Tuple{Int64, Int64, VariableRef}}}:
  (x1 = 1, x2 = 2, y = x[1,2])
  (x1 = 1, x2 = 1, y = x[1,1])
- (x1 = 2, x2 = 2, y = x[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/tables.jl#L12-L48">source</a></section></article><h2 id="Containers.default_container"><a class="docs-heading-anchor" href="#Containers.default_container"><code>Containers.default_container</code></a><a id="Containers.default_container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.default_container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.default_container" href="#JuMP.Containers.default_container"><code>JuMP.Containers.default_container</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">default_container(indices)</code></pre><p>If <code>indices</code> is a <a href="#JuMP.Containers.NestedIterator"><code>NestedIterator</code></a>, return a <a href="#JuMP.Containers.SparseAxisArray"><code>SparseAxisArray</code></a>. Otherwise, <code>indices</code> should be a <code>VectorizedProductIterator</code> and the function returns <code>Array</code> if all iterators of the product are <code>Base.OneTo</code> and retunrs <a href="#JuMP.Containers.DenseAxisArray"><code>DenseAxisArray</code></a> otherwise.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/container.jl#L6-L14">source</a></section></article><h2 id="Containers.nested"><a class="docs-heading-anchor" href="#Containers.nested"><code>Containers.nested</code></a><a id="Containers.nested-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.nested" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.nested" href="#JuMP.Containers.nested"><code>JuMP.Containers.nested</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nested(iterators...; condition = (args...) -&gt; true)</code></pre><p>Create a <a href="#JuMP.Containers.NestedIterator"><code>NestedIterator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; iterator = Containers.nested(
+ (x1 = 2, x2 = 2, y = x[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/tables.jl#L12-L48">source</a></section></article><h2 id="Containers.default_container"><a class="docs-heading-anchor" href="#Containers.default_container"><code>Containers.default_container</code></a><a id="Containers.default_container-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.default_container" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.default_container" href="#JuMP.Containers.default_container"><code>JuMP.Containers.default_container</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">default_container(indices)</code></pre><p>If <code>indices</code> is a <a href="#JuMP.Containers.NestedIterator"><code>NestedIterator</code></a>, return a <a href="#JuMP.Containers.SparseAxisArray"><code>SparseAxisArray</code></a>. Otherwise, <code>indices</code> should be a <code>VectorizedProductIterator</code> and the function returns <code>Array</code> if all iterators of the product are <code>Base.OneTo</code> and retunrs <a href="#JuMP.Containers.DenseAxisArray"><code>DenseAxisArray</code></a> otherwise.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/container.jl#L6-L14">source</a></section></article><h2 id="Containers.nested"><a class="docs-heading-anchor" href="#Containers.nested"><code>Containers.nested</code></a><a id="Containers.nested-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.nested" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.nested" href="#JuMP.Containers.nested"><code>JuMP.Containers.nested</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nested(iterators...; condition = (args...) -&gt; true)</code></pre><p>Create a <a href="#JuMP.Containers.NestedIterator"><code>NestedIterator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; iterator = Containers.nested(
            () -&gt; 1:2,
            (i,) -&gt; [&quot;A&quot;, &quot;B&quot;];
            condition = (i, j) -&gt; isodd(i) || j == &quot;B&quot;,
@@ -137,12 +137,12 @@
 3-element Vector{Tuple{Int64, String}}:
  (1, &quot;A&quot;)
  (1, &quot;B&quot;)
- (2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/nested_iterator.jl#L51-L71">source</a></section></article><h2 id="Containers.vectorized_product"><a class="docs-heading-anchor" href="#Containers.vectorized_product"><code>Containers.vectorized_product</code></a><a id="Containers.vectorized_product-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.vectorized_product" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.vectorized_product" href="#JuMP.Containers.vectorized_product"><code>JuMP.Containers.vectorized_product</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">vectorized_product(iterators...)</code></pre><p>Created a <a href="#JuMP.Containers.VectorizedProductIterator"><code>VectorizedProductIterator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; iterator = Containers.vectorized_product(1:2, [&quot;A&quot;, &quot;B&quot;]);
+ (2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/nested_iterator.jl#L51-L71">source</a></section></article><h2 id="Containers.vectorized_product"><a class="docs-heading-anchor" href="#Containers.vectorized_product"><code>Containers.vectorized_product</code></a><a id="Containers.vectorized_product-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.vectorized_product" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.vectorized_product" href="#JuMP.Containers.vectorized_product"><code>JuMP.Containers.vectorized_product</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">vectorized_product(iterators...)</code></pre><p>Created a <a href="#JuMP.Containers.VectorizedProductIterator"><code>VectorizedProductIterator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; iterator = Containers.vectorized_product(1:2, [&quot;A&quot;, &quot;B&quot;]);
 
 julia&gt; collect(iterator)
 2×2 Matrix{Tuple{Int64, String}}:
  (1, &quot;A&quot;)  (1, &quot;B&quot;)
- (2, &quot;A&quot;)  (2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/vectorized_product_iterator.jl#L56-L71">source</a></section></article><h2 id="Containers.build_ref_sets"><a class="docs-heading-anchor" href="#Containers.build_ref_sets"><code>Containers.build_ref_sets</code></a><a id="Containers.build_ref_sets-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.build_ref_sets" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.build_ref_sets" href="#JuMP.Containers.build_ref_sets"><code>JuMP.Containers.build_ref_sets</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_ref_sets(_error::Function, expr)</code></pre><p>Helper function for macros to construct container objects.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>This function is for advanced users implementing JuMP extensions. See <a href="#JuMP.Containers.container_code"><code>container_code</code></a> for more details.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that takes a <code>String</code> and throws an error, potentially annotating the input string with extra information such as from which macro it was thrown from. Use <code>error</code> if you do not want a modified error message.</li><li><code>expr</code>: an <code>Expr</code> that specifies the container, e.g., <code>:(x[i = 1:3, [:red, :blue], k = S; i + k &lt;= 6])</code></li></ul><p><strong>Returns</strong></p><ol><li><code>index_vars</code>: a <code>Vector{Any}</code> of names for the index variables, e.g., <code>[:i, gensym(), :k]</code>. These may also be expressions, like <code>:((i, j))</code> from a call like <code>:(x[(i, j) in S])</code>.</li><li><code>indices</code>: an iterator over the indices, for example, <a href="#JuMP.Containers.NestedIterator"><code>Containers.NestedIterator</code></a></li></ol><p><strong>Example</strong></p><p>See <a href="#JuMP.Containers.container_code"><code>container_code</code></a> for a worked example.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/macro.jl#L182-L210">source</a></section></article><h2 id="Containers.container_code"><a class="docs-heading-anchor" href="#Containers.container_code"><code>Containers.container_code</code></a><a id="Containers.container_code-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.container_code" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.container_code" href="#JuMP.Containers.container_code"><code>JuMP.Containers.container_code</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">container_code(
+ (2, &quot;A&quot;)  (2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/vectorized_product_iterator.jl#L56-L71">source</a></section></article><h2 id="Containers.build_ref_sets"><a class="docs-heading-anchor" href="#Containers.build_ref_sets"><code>Containers.build_ref_sets</code></a><a id="Containers.build_ref_sets-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.build_ref_sets" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.build_ref_sets" href="#JuMP.Containers.build_ref_sets"><code>JuMP.Containers.build_ref_sets</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_ref_sets(_error::Function, expr)</code></pre><p>Helper function for macros to construct container objects.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>This function is for advanced users implementing JuMP extensions. See <a href="#JuMP.Containers.container_code"><code>container_code</code></a> for more details.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that takes a <code>String</code> and throws an error, potentially annotating the input string with extra information such as from which macro it was thrown from. Use <code>error</code> if you do not want a modified error message.</li><li><code>expr</code>: an <code>Expr</code> that specifies the container, e.g., <code>:(x[i = 1:3, [:red, :blue], k = S; i + k &lt;= 6])</code></li></ul><p><strong>Returns</strong></p><ol><li><code>index_vars</code>: a <code>Vector{Any}</code> of names for the index variables, e.g., <code>[:i, gensym(), :k]</code>. These may also be expressions, like <code>:((i, j))</code> from a call like <code>:(x[(i, j) in S])</code>.</li><li><code>indices</code>: an iterator over the indices, for example, <a href="#JuMP.Containers.NestedIterator"><code>Containers.NestedIterator</code></a></li></ol><p><strong>Example</strong></p><p>See <a href="#JuMP.Containers.container_code"><code>container_code</code></a> for a worked example.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/macro.jl#L182-L210">source</a></section></article><h2 id="Containers.container_code"><a class="docs-heading-anchor" href="#Containers.container_code"><code>Containers.container_code</code></a><a id="Containers.container_code-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.container_code" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.container_code" href="#JuMP.Containers.container_code"><code>JuMP.Containers.container_code</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">container_code(
     index_vars::Vector{Any},
     indices::Expr,
     code,
@@ -164,7 +164,7 @@
     Dimension 2, [&quot;A&quot;, &quot;B&quot;]
 And data, a 2×2 Matrix{String}:
  &quot;A&quot;   &quot;B&quot;
- &quot;AA&quot;  &quot;BB&quot;</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/macro.jl#L241-L290">source</a></section></article><h2 id="Containers.AutoContainerType"><a class="docs-heading-anchor" href="#Containers.AutoContainerType"><code>Containers.AutoContainerType</code></a><a id="Containers.AutoContainerType-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.AutoContainerType" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.AutoContainerType" href="#JuMP.Containers.AutoContainerType"><code>JuMP.Containers.AutoContainerType</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AutoContainerType</code></pre><p>Pass <code>AutoContainerType</code> to <a href="#JuMP.Containers.container"><code>container</code></a> to let the container type be chosen based on the type of the indices using <a href="#JuMP.Containers.default_container"><code>default_container</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/container.jl#L17-L22">source</a></section></article><h2 id="Containers.NestedIterator"><a class="docs-heading-anchor" href="#Containers.NestedIterator"><code>Containers.NestedIterator</code></a><a id="Containers.NestedIterator-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.NestedIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.NestedIterator" href="#JuMP.Containers.NestedIterator"><code>JuMP.Containers.NestedIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct NestedIterator{T}
+ &quot;AA&quot;  &quot;BB&quot;</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/macro.jl#L241-L290">source</a></section></article><h2 id="Containers.AutoContainerType"><a class="docs-heading-anchor" href="#Containers.AutoContainerType"><code>Containers.AutoContainerType</code></a><a id="Containers.AutoContainerType-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.AutoContainerType" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.AutoContainerType" href="#JuMP.Containers.AutoContainerType"><code>JuMP.Containers.AutoContainerType</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AutoContainerType</code></pre><p>Pass <code>AutoContainerType</code> to <a href="#JuMP.Containers.container"><code>container</code></a> to let the container type be chosen based on the type of the indices using <a href="#JuMP.Containers.default_container"><code>default_container</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/container.jl#L17-L22">source</a></section></article><h2 id="Containers.NestedIterator"><a class="docs-heading-anchor" href="#Containers.NestedIterator"><code>Containers.NestedIterator</code></a><a id="Containers.NestedIterator-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.NestedIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.NestedIterator" href="#JuMP.Containers.NestedIterator"><code>JuMP.Containers.NestedIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct NestedIterator{T}
     iterators::T # Tuple of functions
     condition::Function
 end</code></pre><p>Iterators over the tuples that are produced by a nested for loop.</p><p>Construct a <code>NestedIterator</code> using <a href="#JuMP.Containers.nested"><code>nested</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; iterators = (() -&gt; 1:2, (i,) -&gt; [&quot;A&quot;, &quot;B&quot;]);
@@ -187,6 +187,6 @@
        end
 (1, &quot;A&quot;)
 (1, &quot;B&quot;)
-(2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/nested_iterator.jl#L6-L45">source</a></section></article><h2 id="Containers.VectorizedProductIterator"><a class="docs-heading-anchor" href="#Containers.VectorizedProductIterator"><code>Containers.VectorizedProductIterator</code></a><a id="Containers.VectorizedProductIterator-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.VectorizedProductIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.VectorizedProductIterator" href="#JuMP.Containers.VectorizedProductIterator"><code>JuMP.Containers.VectorizedProductIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VectorizedProductIterator{T}
+(2, &quot;B&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/nested_iterator.jl#L6-L45">source</a></section></article><h2 id="Containers.VectorizedProductIterator"><a class="docs-heading-anchor" href="#Containers.VectorizedProductIterator"><code>Containers.VectorizedProductIterator</code></a><a id="Containers.VectorizedProductIterator-1"></a><a class="docs-heading-anchor-permalink" href="#Containers.VectorizedProductIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Containers.VectorizedProductIterator" href="#JuMP.Containers.VectorizedProductIterator"><code>JuMP.Containers.VectorizedProductIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VectorizedProductIterator{T}
     prod::Iterators.ProductIterator{T}
-end</code></pre><p>A wrapper type for <code>Iterators.ProuctIterator</code> that discards shape information and returns a <code>Vector</code>.</p><p>Construct a <code>VectorizedProductIterator</code> using <a href="#JuMP.Containers.vectorized_product"><code>vectorized_product</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/Containers/vectorized_product_iterator.jl#L31-L40">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../JuMP/">« JuMP</a><a class="docs-footer-nextpage" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><p>A wrapper type for <code>Iterators.ProuctIterator</code> that discards shape information and returns a <code>Vector</code>.</p><p>Construct a <code>VectorizedProductIterator</code> using <a href="#JuMP.Containers.vectorized_product"><code>vectorized_product</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/Containers/vectorized_product_iterator.jl#L31-L40">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../JuMP/">« JuMP</a><a class="docs-footer-nextpage" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/api/JuMP/index.html b/previews/PR3562/api/JuMP/index.html
index bff8508411e..c1f22a04fbe 100644
--- a/previews/PR3562/api/JuMP/index.html
+++ b/previews/PR3562/api/JuMP/index.html
@@ -15,7 +15,7 @@
 3-element Vector{NonlinearConstraintRef{ScalarShape}}:
  (sin(1.0 * x) - 1.0 / 1.0) - 0.0 ≤ 0
  (sin(2.0 * x) - 1.0 / 2.0) - 0.0 ≤ 0
- (sin(3.0 * x) - 1.0 / 3.0) - 0.0 ≤ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L3092-L3113">source</a></section></article><h2 id="@NLconstraints"><a class="docs-heading-anchor" href="#@NLconstraints"><code>@NLconstraints</code></a><a id="@NLconstraints-1"></a><a class="docs-heading-anchor-permalink" href="#@NLconstraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLconstraints" href="#JuMP.@NLconstraints"><code>JuMP.@NLconstraints</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLconstraints(model, args...)</code></pre><p>Adds multiple nonlinear constraints to model at once, in the same fashion as the <a href="#JuMP.@NLconstraint"><code>@NLconstraint</code></a> macro.</p><p>The model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the constraints that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ (sin(3.0 * x) - 1.0 / 3.0) - 0.0 ≤ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L3092-L3113">source</a></section></article><h2 id="@NLconstraints"><a class="docs-heading-anchor" href="#@NLconstraints"><code>@NLconstraints</code></a><a id="@NLconstraints-1"></a><a class="docs-heading-anchor-permalink" href="#@NLconstraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLconstraints" href="#JuMP.@NLconstraints"><code>JuMP.@NLconstraints</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLconstraints(model, args...)</code></pre><p>Adds multiple nonlinear constraints to model at once, in the same fashion as the <a href="#JuMP.@NLconstraint"><code>@NLconstraint</code></a> macro.</p><p>The model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the constraints that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -31,7 +31,7 @@
            t &gt;= sqrt(x^2 + y^2)
            [i = 1:2], z[i] &lt;= log(a[i])
        end)
-((t - sqrt(x ^ 2.0 + y ^ 2.0)) - 0.0 ≥ 0, NonlinearConstraintRef{ScalarShape}[(z[1] - log(4.0)) - 0.0 ≤ 0, (z[2] - log(5.0)) - 0.0 ≤ 0])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1692-L1724">source</a></section></article><h2 id="@NLexpression"><a class="docs-heading-anchor" href="#@NLexpression"><code>@NLexpression</code></a><a id="@NLexpression-1"></a><a class="docs-heading-anchor-permalink" href="#@NLexpression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLexpression" href="#JuMP.@NLexpression"><code>JuMP.@NLexpression</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLexpression(args...)</code></pre><p>Efficiently build a nonlinear expression which can then be inserted in other nonlinear constraints and the objective. See also [<code>@expression</code>]. For example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+((t - sqrt(x ^ 2.0 + y ^ 2.0)) - 0.0 ≥ 0, NonlinearConstraintRef{ScalarShape}[(z[1] - log(4.0)) - 0.0 ≤ 0, (z[2] - log(5.0)) - 0.0 ≤ 0])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1692-L1724">source</a></section></article><h2 id="@NLexpression"><a class="docs-heading-anchor" href="#@NLexpression"><code>@NLexpression</code></a><a id="@NLexpression-1"></a><a class="docs-heading-anchor-permalink" href="#@NLexpression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLexpression" href="#JuMP.@NLexpression"><code>JuMP.@NLexpression</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLexpression(args...)</code></pre><p>Efficiently build a nonlinear expression which can then be inserted in other nonlinear constraints and the objective. See also [<code>@expression</code>]. For example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -52,7 +52,7 @@
  subexpression[4]: sin(3.0 * x)
 
 julia&gt; my_expr_2 = @NLexpression(model, log(1 + sum(exp(my_expr_1[i]) for i in 1:2)))
-subexpression[5]: log(1.0 + (exp(subexpression[2]) + exp(subexpression[3])))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L3165-L3200">source</a></section></article><h2 id="@NLexpressions"><a class="docs-heading-anchor" href="#@NLexpressions"><code>@NLexpressions</code></a><a id="@NLexpressions-1"></a><a class="docs-heading-anchor-permalink" href="#@NLexpressions" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLexpressions" href="#JuMP.@NLexpressions"><code>JuMP.@NLexpressions</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLexpressions(model, args...)</code></pre><p>Adds multiple nonlinear expressions to model at once, in the same fashion as the <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a> macro.</p><p>The model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the expressions that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+subexpression[5]: log(1.0 + (exp(subexpression[2]) + exp(subexpression[3])))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L3165-L3200">source</a></section></article><h2 id="@NLexpressions"><a class="docs-heading-anchor" href="#@NLexpressions"><code>@NLexpressions</code></a><a id="@NLexpressions-1"></a><a class="docs-heading-anchor-permalink" href="#@NLexpressions" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLexpressions" href="#JuMP.@NLexpressions"><code>JuMP.@NLexpressions</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLexpressions(model, args...)</code></pre><p>Adds multiple nonlinear expressions to model at once, in the same fashion as the <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a> macro.</p><p>The model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the expressions that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -66,7 +66,7 @@
            my_expr, sqrt(x^2 + y^2)
            my_expr_1[i = 1:2], log(a[i]) - z[i]
        end)
-(subexpression[1]: sqrt(x ^ 2.0 + y ^ 2.0), NonlinearExpression[subexpression[2]: log(4.0) - z[1], subexpression[3]: log(5.0) - z[2]])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1726-L1756">source</a></section></article><h2 id="@NLobjective"><a class="docs-heading-anchor" href="#@NLobjective"><code>@NLobjective</code></a><a id="@NLobjective-1"></a><a class="docs-heading-anchor-permalink" href="#@NLobjective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLobjective" href="#JuMP.@NLobjective"><code>JuMP.@NLobjective</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLobjective(model, sense, expression)</code></pre><p>Add a nonlinear objective to <code>model</code> with optimization sense <code>sense</code>. <code>sense</code> must be <code>Max</code> or <code>Min</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+(subexpression[1]: sqrt(x ^ 2.0 + y ^ 2.0), NonlinearExpression[subexpression[2]: log(4.0) - z[1], subexpression[3]: log(5.0) - z[2]])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1726-L1756">source</a></section></article><h2 id="@NLobjective"><a class="docs-heading-anchor" href="#@NLobjective"><code>@NLobjective</code></a><a id="@NLobjective-1"></a><a class="docs-heading-anchor-permalink" href="#@NLobjective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLobjective" href="#JuMP.@NLobjective"><code>JuMP.@NLobjective</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLobjective(model, sense, expression)</code></pre><p>Add a nonlinear objective to <code>model</code> with optimization sense <code>sense</code>. <code>sense</code> must be <code>Max</code> or <code>Min</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -75,7 +75,7 @@
 
 julia&gt; print(model)
 Max 2.0 * x + 1.0 + sin(x)
-Subject to</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L3057-L3077">source</a></section></article><h2 id="@NLparameter"><a class="docs-heading-anchor" href="#@NLparameter"><code>@NLparameter</code></a><a id="@NLparameter-1"></a><a class="docs-heading-anchor-permalink" href="#@NLparameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLparameter" href="#JuMP.@NLparameter"><code>JuMP.@NLparameter</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLparameter(model, param == value)</code></pre><p>Create and return a nonlinear parameter <code>param</code> attached to the model <code>model</code> with initial value set to <code>value</code>. Nonlinear parameters may be used only in nonlinear expressions.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+Subject to</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L3057-L3077">source</a></section></article><h2 id="@NLparameter"><a class="docs-heading-anchor" href="#@NLparameter"><code>@NLparameter</code></a><a id="@NLparameter-1"></a><a class="docs-heading-anchor-permalink" href="#@NLparameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLparameter" href="#JuMP.@NLparameter"><code>JuMP.@NLparameter</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@NLparameter(model, param == value)</code></pre><p>Create and return a nonlinear parameter <code>param</code> attached to the model <code>model</code> with initial value set to <code>value</code>. Nonlinear parameters may be used only in nonlinear expressions.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @NLparameter(model, x == 10)
 x == 10.0
@@ -105,7 +105,7 @@
  parameter[3] == 6.0
 
 julia&gt; value(y[2])
-4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L3248-L3327">source</a></section></article><h2 id="@NLparameters"><a class="docs-heading-anchor" href="#@NLparameters"><code>@NLparameters</code></a><a id="@NLparameters-1"></a><a class="docs-heading-anchor-permalink" href="#@NLparameters" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLparameters" href="#JuMP.@NLparameters"><code>JuMP.@NLparameters</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs"> @NLparameters(model, args...)</code></pre><p>Create and return multiple nonlinear parameters attached to model <code>model</code>, in the same fashion as <a href="#JuMP.@NLparameter"><code>@NLparameter</code></a> macro.</p><p>The model must be the first argument, and multiple parameters can be added on multiple lines wrapped in a <code>begin ... end</code> block. Distinct parameters need to be placed on separate lines as in the following example.</p><p>The macro returns a tuple containing the parameters that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L3248-L3327">source</a></section></article><h2 id="@NLparameters"><a class="docs-heading-anchor" href="#@NLparameters"><code>@NLparameters</code></a><a id="@NLparameters-1"></a><a class="docs-heading-anchor-permalink" href="#@NLparameters" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@NLparameters" href="#JuMP.@NLparameters"><code>JuMP.@NLparameters</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs"> @NLparameters(model, args...)</code></pre><p>Create and return multiple nonlinear parameters attached to model <code>model</code>, in the same fashion as <a href="#JuMP.@NLparameter"><code>@NLparameter</code></a> macro.</p><p>The model must be the first argument, and multiple parameters can be added on multiple lines wrapped in a <code>begin ... end</code> block. Distinct parameters need to be placed on separate lines as in the following example.</p><p>The macro returns a tuple containing the parameters that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @NLparameters(model, begin
            x == 10
@@ -113,12 +113,12 @@
        end);
 
 julia&gt; value(x)
-10.0</code></pre><pre><code class="nohighlight hljs"></code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1665-L1690">source</a></section></article><h2 id="@build_constraint"><a class="docs-heading-anchor" href="#@build_constraint"><code>@build_constraint</code></a><a id="@build_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#@build_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@build_constraint" href="#JuMP.@build_constraint"><code>JuMP.@build_constraint</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@build_constraint(constraint_expr)</code></pre><p>Constructs a <code>ScalarConstraint</code> or <code>VectorConstraint</code> using the same machinery as <a href="#JuMP.@constraint"><code>@constraint</code></a> but without adding the constraint to a model.</p><p>Constraints using broadcast operators like <code>x .&lt;= 1</code> are also supported and will create arrays of <code>ScalarConstraint</code> or <code>VectorConstraint</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+10.0</code></pre><pre><code class="nohighlight hljs"></code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1665-L1690">source</a></section></article><h2 id="@build_constraint"><a class="docs-heading-anchor" href="#@build_constraint"><code>@build_constraint</code></a><a id="@build_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#@build_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@build_constraint" href="#JuMP.@build_constraint"><code>JuMP.@build_constraint</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@build_constraint(constraint_expr)</code></pre><p>Constructs a <code>ScalarConstraint</code> or <code>VectorConstraint</code> using the same machinery as <a href="#JuMP.@constraint"><code>@constraint</code></a> but without adding the constraint to a model.</p><p>Constraints using broadcast operators like <code>x .&lt;= 1</code> are also supported and will create arrays of <code>ScalarConstraint</code> or <code>VectorConstraint</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
 julia&gt; @build_constraint(2x &gt;= 1)
-ScalarConstraint{AffExpr, MathOptInterface.GreaterThan{Float64}}(2 x, MathOptInterface.GreaterThan{Float64}(1.0))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1443-L1462">source</a></section></article><h2 id="@constraint"><a class="docs-heading-anchor" href="#@constraint"><code>@constraint</code></a><a id="@constraint-1"></a><a class="docs-heading-anchor-permalink" href="#@constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@constraint" href="#JuMP.@constraint"><code>JuMP.@constraint</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@constraint(m::GenericModel, expr, kw_args...)</code></pre><p>Add a constraint described by the expression <code>expr</code>.</p><pre><code class="nohighlight hljs">@constraint(m::GenericModel, ref[i=..., j=..., ...], expr, kw_args...)</code></pre><p>Add a group of constraints described by the expression <code>expr</code> parametrized by <code>i</code>, <code>j</code>, ...</p><p>The expression <code>expr</code> can either be</p><ul><li>of the form <code>func in set</code> constraining the function <code>func</code> to belong to the set <code>set</code> which is either a <a href="../../moi/reference/standard_form/#MathOptInterface.AbstractSet"><code>MOI.AbstractSet</code></a> or one of the JuMP shortcuts <a href="#JuMP.SecondOrderCone"><code>SecondOrderCone</code></a>, <a href="#JuMP.RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a> and <a href="#JuMP.PSDCone"><code>PSDCone</code></a>, e.g. <code>@constraint(model, [1, x-1, y-2] in SecondOrderCone())</code> constrains the norm of <code>[x-1, y-2]</code> be less than 1;</li><li>of the form <code>a sign b</code>, where <code>sign</code> is one of <code>==</code>, <code>≥</code>, <code>&gt;=</code>, <code>≤</code> and <code>&lt;=</code> building the single constraint enforcing the comparison to hold for the expression <code>a</code> and <code>b</code>, e.g. <code>@constraint(m, x^2 + y^2 == 1)</code> constrains <code>x</code> and <code>y</code> to lie on the unit circle;</li><li>of the form <code>a ≤ b ≤ c</code> or <code>a ≥ b ≥ c</code> (where <code>≤</code> and <code>&lt;=</code> (resp. <code>≥</code> and <code>&gt;=</code>) can be used interchangeably) constraining the paired the expression <code>b</code> to lie between <code>a</code> and <code>c</code>;</li><li>of the forms <code>@constraint(m, a .sign b)</code> or <code>@constraint(m, a .sign b .sign c)</code> which broadcast the constraint creation to each element of the vectors.</li></ul><p>The recognized keyword arguments in <code>kw_args</code> are the following:</p><ul><li><code>base_name</code>: Sets the name prefix used to generate constraint names. It corresponds to the constraint name for scalar constraints, otherwise, the constraint names are set to <code>base_name[...]</code> for each index <code>...</code> of the axes <code>axes</code>.</li><li><code>container</code>: Specify the container type.</li><li><code>set_string_name::Bool = true</code>: control whether to set the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintName"><code>MOI.ConstraintName</code></a> attribute. Passing <code>set_string_name = false</code> can improve performance.</li></ul><p><strong>Note for extending the constraint macro</strong></p><p>Each constraint will be created using <code>add_constraint(m, build_constraint(_error, func, set))</code> where</p><ul><li><code>_error</code> is an error function showing the constraint call in addition to the error message given as argument,</li><li><code>func</code> is the expression that is constrained</li><li>and <code>set</code> is the set in which it is constrained to belong.</li></ul><p>For <code>expr</code> of the first type (i.e. <code>@constraint(m, func in set)</code>), <code>func</code> and <code>set</code> are passed unchanged to <code>build_constraint</code> but for the other types, they are determined from the expressions and signs. For instance, <code>@constraint(m, x^2 + y^2 == 1)</code> is transformed into <code>add_constraint(m, build_constraint(_error, x^2 + y^2, MOI.EqualTo(1.0)))</code>.</p><p>To extend JuMP to accept new constraints of this form, it is necessary to add the corresponding methods to <code>build_constraint</code>. Note that this will likely mean that either <code>func</code> or <code>set</code> will be some custom type, rather than e.g. a <code>Symbol</code>, since we will likely want to dispatch on the type of the function or set appearing in the constraint.</p><p>For extensions that need to create constraints with more information than just <code>func</code> and <code>set</code>, an additional positional argument can be specified to <code>@constraint</code> that will then be passed on <code>build_constraint</code>. Hence, we can enable this syntax by defining extensions of <code>build_constraint(_error, func, set, my_arg; kw_args...)</code>. This produces the user syntax: <code>@constraint(model, ref[...], expr, my_arg, kw_args...)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1371-L1438">source</a></section></article><h2 id="@constraints"><a class="docs-heading-anchor" href="#@constraints"><code>@constraints</code></a><a id="@constraints-1"></a><a class="docs-heading-anchor-permalink" href="#@constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@constraints" href="#JuMP.@constraints"><code>JuMP.@constraints</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@constraints(model, args...)</code></pre><p>Adds groups of constraints at once, in the same fashion as the <a href="#JuMP.@constraint"><code>@constraint</code></a> macro.</p><p>The model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the constraints that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ScalarConstraint{AffExpr, MathOptInterface.GreaterThan{Float64}}(2 x, MathOptInterface.GreaterThan{Float64}(1.0))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1443-L1462">source</a></section></article><h2 id="@constraint"><a class="docs-heading-anchor" href="#@constraint"><code>@constraint</code></a><a id="@constraint-1"></a><a class="docs-heading-anchor-permalink" href="#@constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@constraint" href="#JuMP.@constraint"><code>JuMP.@constraint</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@constraint(m::GenericModel, expr, kw_args...)</code></pre><p>Add a constraint described by the expression <code>expr</code>.</p><pre><code class="nohighlight hljs">@constraint(m::GenericModel, ref[i=..., j=..., ...], expr, kw_args...)</code></pre><p>Add a group of constraints described by the expression <code>expr</code> parametrized by <code>i</code>, <code>j</code>, ...</p><p>The expression <code>expr</code> can either be</p><ul><li>of the form <code>func in set</code> constraining the function <code>func</code> to belong to the set <code>set</code> which is either a <a href="../../moi/reference/standard_form/#MathOptInterface.AbstractSet"><code>MOI.AbstractSet</code></a> or one of the JuMP shortcuts <a href="#JuMP.SecondOrderCone"><code>SecondOrderCone</code></a>, <a href="#JuMP.RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a> and <a href="#JuMP.PSDCone"><code>PSDCone</code></a>, e.g. <code>@constraint(model, [1, x-1, y-2] in SecondOrderCone())</code> constrains the norm of <code>[x-1, y-2]</code> be less than 1;</li><li>of the form <code>a sign b</code>, where <code>sign</code> is one of <code>==</code>, <code>≥</code>, <code>&gt;=</code>, <code>≤</code> and <code>&lt;=</code> building the single constraint enforcing the comparison to hold for the expression <code>a</code> and <code>b</code>, e.g. <code>@constraint(m, x^2 + y^2 == 1)</code> constrains <code>x</code> and <code>y</code> to lie on the unit circle;</li><li>of the form <code>a ≤ b ≤ c</code> or <code>a ≥ b ≥ c</code> (where <code>≤</code> and <code>&lt;=</code> (resp. <code>≥</code> and <code>&gt;=</code>) can be used interchangeably) constraining the paired the expression <code>b</code> to lie between <code>a</code> and <code>c</code>;</li><li>of the forms <code>@constraint(m, a .sign b)</code> or <code>@constraint(m, a .sign b .sign c)</code> which broadcast the constraint creation to each element of the vectors.</li></ul><p>The recognized keyword arguments in <code>kw_args</code> are the following:</p><ul><li><code>base_name</code>: Sets the name prefix used to generate constraint names. It corresponds to the constraint name for scalar constraints, otherwise, the constraint names are set to <code>base_name[...]</code> for each index <code>...</code> of the axes <code>axes</code>.</li><li><code>container</code>: Specify the container type.</li><li><code>set_string_name::Bool = true</code>: control whether to set the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintName"><code>MOI.ConstraintName</code></a> attribute. Passing <code>set_string_name = false</code> can improve performance.</li></ul><p><strong>Note for extending the constraint macro</strong></p><p>Each constraint will be created using <code>add_constraint(m, build_constraint(_error, func, set))</code> where</p><ul><li><code>_error</code> is an error function showing the constraint call in addition to the error message given as argument,</li><li><code>func</code> is the expression that is constrained</li><li>and <code>set</code> is the set in which it is constrained to belong.</li></ul><p>For <code>expr</code> of the first type (i.e. <code>@constraint(m, func in set)</code>), <code>func</code> and <code>set</code> are passed unchanged to <code>build_constraint</code> but for the other types, they are determined from the expressions and signs. For instance, <code>@constraint(m, x^2 + y^2 == 1)</code> is transformed into <code>add_constraint(m, build_constraint(_error, x^2 + y^2, MOI.EqualTo(1.0)))</code>.</p><p>To extend JuMP to accept new constraints of this form, it is necessary to add the corresponding methods to <code>build_constraint</code>. Note that this will likely mean that either <code>func</code> or <code>set</code> will be some custom type, rather than e.g. a <code>Symbol</code>, since we will likely want to dispatch on the type of the function or set appearing in the constraint.</p><p>For extensions that need to create constraints with more information than just <code>func</code> and <code>set</code>, an additional positional argument can be specified to <code>@constraint</code> that will then be passed on <code>build_constraint</code>. Hence, we can enable this syntax by defining extensions of <code>build_constraint(_error, func, set, my_arg; kw_args...)</code>. This produces the user syntax: <code>@constraint(model, ref[...], expr, my_arg, kw_args...)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1371-L1438">source</a></section></article><h2 id="@constraints"><a class="docs-heading-anchor" href="#@constraints"><code>@constraints</code></a><a id="@constraints-1"></a><a class="docs-heading-anchor-permalink" href="#@constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@constraints" href="#JuMP.@constraints"><code>JuMP.@constraints</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@constraints(model, args...)</code></pre><p>Adds groups of constraints at once, in the same fashion as the <a href="#JuMP.@constraint"><code>@constraint</code></a> macro.</p><p>The model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the constraints that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, w);
 
@@ -141,7 +141,7 @@
  sum_to_one[2] : y + z[2] = 1
  sum_to_one[3] : y + z[3] = 1
  x ≥ 1
- -w + y ≤ 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1563-L1602">source</a></section></article><h2 id="@expression"><a class="docs-heading-anchor" href="#@expression"><code>@expression</code></a><a id="@expression-1"></a><a class="docs-heading-anchor-permalink" href="#@expression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@expression" href="#JuMP.@expression"><code>JuMP.@expression</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@expression(args...)</code></pre><p>Efficiently builds a linear or quadratic expression but does not add to model immediately. Instead, returns the expression which can then be inserted in other constraints.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ -w + y ≤ 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1563-L1602">source</a></section></article><h2 id="@expression"><a class="docs-heading-anchor" href="#@expression"><code>@expression</code></a><a id="@expression-1"></a><a class="docs-heading-anchor-permalink" href="#@expression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@expression" href="#JuMP.@expression"><code>JuMP.@expression</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@expression(args...)</code></pre><p>Efficiently builds a linear or quadratic expression but does not add to model immediately. Instead, returns the expression which can then be inserted in other constraints.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x[1:5]);
 
@@ -164,7 +164,7 @@
 3-element Vector{AffExpr}:
  x[1] + x[2] + x[3]
  2 x[1] + 2 x[2] + 2 x[3]
- 3 x[1] + 3 x[2] + 3 x[3]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1870-L1918">source</a></section></article><h2 id="@expressions"><a class="docs-heading-anchor" href="#@expressions"><code>@expressions</code></a><a id="@expressions-1"></a><a class="docs-heading-anchor-permalink" href="#@expressions" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@expressions" href="#JuMP.@expressions"><code>JuMP.@expressions</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@expressions(model, args...)</code></pre><p>Adds multiple expressions to model at once, in the same fashion as the <a href="#JuMP.@expression"><code>@expression</code></a> macro.</p><p>The model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the expressions that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ 3 x[1] + 3 x[2] + 3 x[3]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1870-L1918">source</a></section></article><h2 id="@expressions"><a class="docs-heading-anchor" href="#@expressions"><code>@expressions</code></a><a id="@expressions-1"></a><a class="docs-heading-anchor-permalink" href="#@expressions" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@expressions" href="#JuMP.@expressions"><code>JuMP.@expressions</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@expressions(model, args...)</code></pre><p>Adds multiple expressions to model at once, in the same fashion as the <a href="#JuMP.@expression"><code>@expression</code></a> macro.</p><p>The model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the expressions that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -178,7 +178,7 @@
            my_expr, x^2 + y^2
            my_expr_1[i = 1:2], a[i] - z[i]
        end)
-(x² + y², AffExpr[-z[1] + 4, -z[2] + 5])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1633-L1663">source</a></section></article><h2 id="@objective"><a class="docs-heading-anchor" href="#@objective"><code>@objective</code></a><a id="@objective-1"></a><a class="docs-heading-anchor-permalink" href="#@objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@objective" href="#JuMP.@objective"><code>JuMP.@objective</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@objective(model::GenericModel, sense, func)</code></pre><p>Set the objective sense to <code>sense</code> and objective function to <code>func</code>. The objective sense can be either <code>Min</code>, <code>Max</code>, <code>MOI.MIN_SENSE</code>, <code>MOI.MAX_SENSE</code> or <code>MOI.FEASIBILITY_SENSE</code>; see <a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense</code></a>.</p><p>In order to set the sense programmatically, i.e., when <code>sense</code> is a Julia variable whose value is the sense, one of the three <code>MOI.ObjectiveSense</code> values should be used.</p><p><strong>Example</strong></p><p>To minimize the value of the variable <code>x</code>, do as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+(x² + y², AffExpr[-z[1] + 4, -z[2] + 5])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1633-L1663">source</a></section></article><h2 id="@objective"><a class="docs-heading-anchor" href="#@objective"><code>@objective</code></a><a id="@objective-1"></a><a class="docs-heading-anchor-permalink" href="#@objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@objective" href="#JuMP.@objective"><code>JuMP.@objective</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@objective(model::GenericModel, sense, func)</code></pre><p>Set the objective sense to <code>sense</code> and objective function to <code>func</code>. The objective sense can be either <code>Min</code>, <code>Max</code>, <code>MOI.MIN_SENSE</code>, <code>MOI.MAX_SENSE</code> or <code>MOI.FEASIBILITY_SENSE</code>; see <a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense</code></a>.</p><p>In order to set the sense programmatically, i.e., when <code>sense</code> is a Julia variable whose value is the sense, one of the three <code>MOI.ObjectiveSense</code> values should be used.</p><p><strong>Example</strong></p><p>To minimize the value of the variable <code>x</code>, do as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -189,7 +189,7 @@
 MIN_SENSE::OptimizationSense = 0
 
 julia&gt; @objective(model, sense, x^2 - 2x + 1)
-x² - 2 x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1803-L1842">source</a></section></article><h2 id="@operator"><a class="docs-heading-anchor" href="#@operator"><code>@operator</code></a><a id="@operator-1"></a><a class="docs-heading-anchor-permalink" href="#@operator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@operator" href="#JuMP.@operator"><code>JuMP.@operator</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@operator(model, operator, dim, f[, ∇f[, ∇²f]])</code></pre><p>Add the nonlinear operator <code>operator</code> in <code>model</code> with <code>dim</code> arguments, and create a new <a href="#JuMP.NonlinearOperator"><code>NonlinearOperator</code></a> object called <code>operator</code> in the current scope.</p><p>The function <code>f</code> evaluates the operator and must return a scalar.</p><p>The optional function <code>∇f</code> evaluates the first derivative, and the optional function <code>∇²f</code> evaluates the second derivative.</p><p><code>∇²f</code> may be provided only if <code>∇f</code> is also provided.</p><p><strong>Univariate syntax</strong></p><p>If <code>dim == 1</code>, then the method signatures of each function must be:</p><ul><li><code>f(::T)::T where {T&lt;:Real}</code></li><li><code>∇f(::T)::T where {T&lt;:Real}</code></li><li><code>∇²f(::T)::T where {T&lt;:Real}</code></li></ul><p><strong>Multivariate syntax</strong></p><p>If <code>dim &gt; 1</code>, then the method signatures of each function must be:</p><ul><li><code>f(x::T...)::T where {T&lt;:Real}</code></li><li><code>∇f(g::AbstractVector{T}, x::T...)::Nothing where {T&lt;:Real}</code></li><li><code>∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T&lt;:Real}</code></li></ul><p>Where the gradient vector <code>g</code> and Hessian matrix <code>H</code> are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x² - 2 x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1803-L1842">source</a></section></article><h2 id="@operator"><a class="docs-heading-anchor" href="#@operator"><code>@operator</code></a><a id="@operator-1"></a><a class="docs-heading-anchor-permalink" href="#@operator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@operator" href="#JuMP.@operator"><code>JuMP.@operator</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@operator(model, operator, dim, f[, ∇f[, ∇²f]])</code></pre><p>Add the nonlinear operator <code>operator</code> in <code>model</code> with <code>dim</code> arguments, and create a new <a href="#JuMP.NonlinearOperator"><code>NonlinearOperator</code></a> object called <code>operator</code> in the current scope.</p><p>The function <code>f</code> evaluates the operator and must return a scalar.</p><p>The optional function <code>∇f</code> evaluates the first derivative, and the optional function <code>∇²f</code> evaluates the second derivative.</p><p><code>∇²f</code> may be provided only if <code>∇f</code> is also provided.</p><p><strong>Univariate syntax</strong></p><p>If <code>dim == 1</code>, then the method signatures of each function must be:</p><ul><li><code>f(::T)::T where {T&lt;:Real}</code></li><li><code>∇f(::T)::T where {T&lt;:Real}</code></li><li><code>∇²f(::T)::T where {T&lt;:Real}</code></li></ul><p><strong>Multivariate syntax</strong></p><p>If <code>dim &gt; 1</code>, then the method signatures of each function must be:</p><ul><li><code>f(x::T...)::T where {T&lt;:Real}</code></li><li><code>∇f(g::AbstractVector{T}, x::T...)::Nothing where {T&lt;:Real}</code></li><li><code>∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T&lt;:Real}</code></li></ul><p>Where the gradient vector <code>g</code> and Hessian matrix <code>H</code> are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -228,7 +228,7 @@
 f (generic function with 1 method)
 
 julia&gt; op_f = model[:op_f] = add_nonlinear_operator(model, 1, f; name = :op_f)
-NonlinearOperator(f, :op_f)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L995-L1087">source</a></section></article><h2 id="@variable"><a class="docs-heading-anchor" href="#@variable"><code>@variable</code></a><a id="@variable-1"></a><a class="docs-heading-anchor-permalink" href="#@variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@variable" href="#JuMP.@variable"><code>JuMP.@variable</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@variable(model, expr, args..., kw_args...)</code></pre><p>Add a variable to the model <code>model</code> described by the expression <code>expr</code>, the positional arguments <code>args</code> and the keyword arguments <code>kw_args</code>.</p><p><strong>Anonymous and named variables</strong></p><p><code>expr</code> must be one of the forms:</p><ul><li>Omitted, like <code>@variable(model)</code>, which creates an anonymous variable</li><li>A single symbol like <code>@variable(model, x)</code></li><li>A container expression like <code>@variable(model, x[i=1:3])</code></li><li>An anoymous container expression like <code>@variable(model, [i=1:3])</code></li></ul><p><strong>Bounds</strong></p><p>In addition, the expression can have bounds, such as:</p><ul><li><code>@variable(model, x &gt;= 0)</code></li><li><code>@variable(model, x &lt;= 0)</code></li><li><code>@variable(model, x == 0)</code></li><li><code>@variable(model, 0 &lt;= x &lt;= 1)</code></li></ul><p>and bounds can depend on the indices of the container expressions:</p><ul><li><code>@variable(model, -i &lt;= x[i=1:3] &lt;= i)</code></li></ul><p><strong>Sets</strong></p><p>You can explicitly specify the set to which the variable belongs:</p><ul><li><code>@variable(model, x in MOI.Interval(0.0, 1.0))</code></li></ul><p>For more information on this syntax, read <a href="../../moi/tutorials/implementing/#Variables-constrained-on-creation">Variables constrained on creation</a>.</p><p><strong>Positional arguments</strong></p><p>The recognized positional arguments in <code>args</code> are the following:</p><ul><li><code>Bin</code>: restricts the variable to the <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a> set, that is, <code>{0, 1}</code>. For example, <code>@variable(model, x, Bin)</code>. Note: you cannot use <code>@variable(model, Bin)</code>, use the <code>binary</code> keyword instead.</li><li><code>Int</code>: restricts the variable to the set of integers, that is, ..., -2, -1,  0, 1, 2, ... For example, <code>@variable(model, x, Int)</code>. Note: you cannot use  <code>@variable(model, Int)</code>, use the <code>integer</code> keyword instead.</li><li><code>Symmetric</code>: Only available when creating a square matrix of variables, i.e., when <code>expr</code> is of the form <code>varname[1:n,1:n]</code> or <code>varname[i=1:n,j=1:n]</code>, it creates a symmetric matrix of variables.</li><li><code>PSD</code>: A restrictive extension to <code>Symmetric</code> which constraints a square matrix of variables to <code>Symmetric</code> and constrains to be positive semidefinite.</li></ul><p><strong>Keyword arguments</strong></p><p>Four keyword arguments are useful in all cases:</p><ul><li><code>base_name</code>: Sets the name prefix used to generate variable names. It corresponds to the variable name for scalar variable, otherwise, the variable names are set to <code>base_name[...]</code> for each index <code>...</code> of the axes <code>axes</code>.</li><li><code>start::Float64</code>: specify the value passed to <code>set_start_value</code> for each variable</li><li><code>container</code>: specify the container type. See <a href="../../manual/variables/#variable_forcing">Forcing the container type</a> for more information.</li><li><code>set_string_name::Bool = true</code>: control whether to set the <a href="../../moi/reference/variables/#MathOptInterface.VariableName"><code>MOI.VariableName</code></a> attribute. Passing <code>set_string_name = false</code> can improve performance.</li></ul><p>Other keyword arguments are needed to disambiguate sitations with anonymous variables:</p><ul><li><code>lower_bound::Float64</code>: an alternative to <code>x &gt;= lb</code>, sets the value of the variable lower bound.</li><li><code>upper_bound::Float64</code>: an alternative to <code>x &lt;= ub</code>, sets the value of the variable upper bound.</li><li><code>binary::Bool</code>: an alternative to passing <code>Bin</code>, sets whether the variable is binary or not.</li><li><code>integer::Bool</code>: an alternative to passing <code>Int</code>, sets whether the variable is integer or not.</li><li><code>set::MOI.AbstractSet</code>: an alternative to using <code>x in set</code></li><li><code>variable_type</code>: used by JuMP extensions. See <a href="../../developers/extensions/#extend_variable_macro">Extend <code>@variable</code></a> for more information.</li></ul><p><strong>Example</strong></p><p>The following are equivalent ways of creating a variable <code>x</code> of name <code>x</code> with lower bound 0:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+NonlinearOperator(f, :op_f)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L995-L1087">source</a></section></article><h2 id="@variable"><a class="docs-heading-anchor" href="#@variable"><code>@variable</code></a><a id="@variable-1"></a><a class="docs-heading-anchor-permalink" href="#@variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@variable" href="#JuMP.@variable"><code>JuMP.@variable</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@variable(model, expr, args..., kw_args...)</code></pre><p>Add a variable to the model <code>model</code> described by the expression <code>expr</code>, the positional arguments <code>args</code> and the keyword arguments <code>kw_args</code>.</p><p><strong>Anonymous and named variables</strong></p><p><code>expr</code> must be one of the forms:</p><ul><li>Omitted, like <code>@variable(model)</code>, which creates an anonymous variable</li><li>A single symbol like <code>@variable(model, x)</code></li><li>A container expression like <code>@variable(model, x[i=1:3])</code></li><li>An anoymous container expression like <code>@variable(model, [i=1:3])</code></li></ul><p><strong>Bounds</strong></p><p>In addition, the expression can have bounds, such as:</p><ul><li><code>@variable(model, x &gt;= 0)</code></li><li><code>@variable(model, x &lt;= 0)</code></li><li><code>@variable(model, x == 0)</code></li><li><code>@variable(model, 0 &lt;= x &lt;= 1)</code></li></ul><p>and bounds can depend on the indices of the container expressions:</p><ul><li><code>@variable(model, -i &lt;= x[i=1:3] &lt;= i)</code></li></ul><p><strong>Sets</strong></p><p>You can explicitly specify the set to which the variable belongs:</p><ul><li><code>@variable(model, x in MOI.Interval(0.0, 1.0))</code></li></ul><p>For more information on this syntax, read <a href="../../moi/tutorials/implementing/#Variables-constrained-on-creation">Variables constrained on creation</a>.</p><p><strong>Positional arguments</strong></p><p>The recognized positional arguments in <code>args</code> are the following:</p><ul><li><code>Bin</code>: restricts the variable to the <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a> set, that is, <code>{0, 1}</code>. For example, <code>@variable(model, x, Bin)</code>. Note: you cannot use <code>@variable(model, Bin)</code>, use the <code>binary</code> keyword instead.</li><li><code>Int</code>: restricts the variable to the set of integers, that is, ..., -2, -1,  0, 1, 2, ... For example, <code>@variable(model, x, Int)</code>. Note: you cannot use  <code>@variable(model, Int)</code>, use the <code>integer</code> keyword instead.</li><li><code>Symmetric</code>: Only available when creating a square matrix of variables, i.e., when <code>expr</code> is of the form <code>varname[1:n,1:n]</code> or <code>varname[i=1:n,j=1:n]</code>, it creates a symmetric matrix of variables.</li><li><code>PSD</code>: A restrictive extension to <code>Symmetric</code> which constraints a square matrix of variables to <code>Symmetric</code> and constrains to be positive semidefinite.</li></ul><p><strong>Keyword arguments</strong></p><p>Four keyword arguments are useful in all cases:</p><ul><li><code>base_name</code>: Sets the name prefix used to generate variable names. It corresponds to the variable name for scalar variable, otherwise, the variable names are set to <code>base_name[...]</code> for each index <code>...</code> of the axes <code>axes</code>.</li><li><code>start::Float64</code>: specify the value passed to <code>set_start_value</code> for each variable</li><li><code>container</code>: specify the container type. See <a href="../../manual/variables/#variable_forcing">Forcing the container type</a> for more information.</li><li><code>set_string_name::Bool = true</code>: control whether to set the <a href="../../moi/reference/variables/#MathOptInterface.VariableName"><code>MOI.VariableName</code></a> attribute. Passing <code>set_string_name = false</code> can improve performance.</li></ul><p>Other keyword arguments are needed to disambiguate sitations with anonymous variables:</p><ul><li><code>lower_bound::Float64</code>: an alternative to <code>x &gt;= lb</code>, sets the value of the variable lower bound.</li><li><code>upper_bound::Float64</code>: an alternative to <code>x &lt;= ub</code>, sets the value of the variable upper bound.</li><li><code>binary::Bool</code>: an alternative to passing <code>Bin</code>, sets whether the variable is binary or not.</li><li><code>integer::Bool</code>: an alternative to passing <code>Int</code>, sets whether the variable is integer or not.</li><li><code>set::MOI.AbstractSet</code>: an alternative to using <code>x in set</code></li><li><code>variable_type</code>: used by JuMP extensions. See <a href="../../developers/extensions/#extend_variable_macro">Extend <code>@variable</code></a> for more information.</li></ul><p><strong>Example</strong></p><p>The following are equivalent ways of creating a variable <code>x</code> of name <code>x</code> with lower bound 0:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 0)
 x</code></pre><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
@@ -257,14 +257,14 @@
 3-element Vector{VariableRef}:
  _[7]
  _[8]
- _[9]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L2706-L2841">source</a></section></article><h2 id="@variables"><a class="docs-heading-anchor" href="#@variables"><code>@variables</code></a><a id="@variables-1"></a><a class="docs-heading-anchor-permalink" href="#@variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@variables" href="#JuMP.@variables"><code>JuMP.@variables</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@variables(model, args...)</code></pre><p>Adds multiple variables to model at once, in the same fashion as the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p><p>The model must be the first argument, and multiple variables can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the variables that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ _[9]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L2706-L2841">source</a></section></article><h2 id="@variables"><a class="docs-heading-anchor" href="#@variables"><code>@variables</code></a><a id="@variables-1"></a><a class="docs-heading-anchor-permalink" href="#@variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.@variables" href="#JuMP.@variables"><code>JuMP.@variables</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@variables(model, args...)</code></pre><p>Adds multiple variables to model at once, in the same fashion as the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p><p>The model must be the first argument, and multiple variables can be added on multiple lines wrapped in a <code>begin ... end</code> block.</p><p>The macro returns a tuple containing the variables that were defined.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variables(model, begin
            x
            y[i = 1:2] &gt;= 0, (start = i)
            z, Bin, (start = 0, base_name = &quot;Z&quot;)
        end)
-(x, VariableRef[y[1], y[2]], Z)</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>Keyword arguments must be contained within parentheses (refer to the example above).</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1604-L1631">source</a></section></article><h2 id="add_bridge"><a class="docs-heading-anchor" href="#add_bridge"><code>add_bridge</code></a><a id="add_bridge-1"></a><a class="docs-heading-anchor-permalink" href="#add_bridge" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_bridge" href="#JuMP.add_bridge"><code>JuMP.add_bridge</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_bridge(
+(x, VariableRef[y[1], y[2]], Z)</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>Keyword arguments must be contained within parentheses (refer to the example above).</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1604-L1631">source</a></section></article><h2 id="add_bridge"><a class="docs-heading-anchor" href="#add_bridge"><code>add_bridge</code></a><a id="add_bridge-1"></a><a class="docs-heading-anchor-permalink" href="#add_bridge" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_bridge" href="#JuMP.add_bridge"><code>JuMP.add_bridge</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_bridge(
     model::GenericModel{T},
     BT::Type{&lt;:MOI.Bridges.AbstractBridge};
     coefficient_type::Type{S} = T,
@@ -276,17 +276,17 @@
            model,
            MOI.Bridges.Constraint.NumberConversionBridge;
            coefficient_type = Complex{Float64}
-       )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L611-L637">source</a></section></article><h2 id="add_constraint"><a class="docs-heading-anchor" href="#add_constraint"><code>add_constraint</code></a><a id="add_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#add_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_constraint" href="#JuMP.add_constraint"><code>JuMP.add_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_constraint(model::GenericModel, con::AbstractConstraint, name::String=&quot;&quot;)</code></pre><p>Add a constraint <code>con</code> to <code>Model model</code> and sets its name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L691-L695">source</a></section></article><h2 id="add_nonlinear_constraint"><a class="docs-heading-anchor" href="#add_nonlinear_constraint"><code>add_nonlinear_constraint</code></a><a id="add_nonlinear_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_constraint" href="#JuMP.add_nonlinear_constraint"><code>JuMP.add_nonlinear_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_constraint(model::Model, expr::Expr)</code></pre><p>Add a nonlinear constraint described by the Julia expression <code>ex</code> to <code>model</code>.</p><p>This function is most useful if the expression <code>ex</code> is generated programmatically, and you cannot use <a href="#JuMP.@NLconstraint"><code>@NLconstraint</code></a>.</p><p><strong>Notes</strong></p><ul><li>You must interpolate the variables directly into the expression <code>expr</code>.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+       )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L611-L637">source</a></section></article><h2 id="add_constraint"><a class="docs-heading-anchor" href="#add_constraint"><code>add_constraint</code></a><a id="add_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#add_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_constraint" href="#JuMP.add_constraint"><code>JuMP.add_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_constraint(model::GenericModel, con::AbstractConstraint, name::String=&quot;&quot;)</code></pre><p>Add a constraint <code>con</code> to <code>Model model</code> and sets its name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L691-L695">source</a></section></article><h2 id="add_nonlinear_constraint"><a class="docs-heading-anchor" href="#add_nonlinear_constraint"><code>add_nonlinear_constraint</code></a><a id="add_nonlinear_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_constraint" href="#JuMP.add_nonlinear_constraint"><code>JuMP.add_nonlinear_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_constraint(model::Model, expr::Expr)</code></pre><p>Add a nonlinear constraint described by the Julia expression <code>ex</code> to <code>model</code>.</p><p>This function is most useful if the expression <code>ex</code> is generated programmatically, and you cannot use <a href="#JuMP.@NLconstraint"><code>@NLconstraint</code></a>.</p><p><strong>Notes</strong></p><ul><li>You must interpolate the variables directly into the expression <code>expr</code>.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
 julia&gt; add_nonlinear_constraint(model, :($(x) + $(x)^2 &lt;= 1))
-(x + x ^ 2.0) - 1.0 ≤ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L443-L465">source</a></section></article><h2 id="add_nonlinear_expression"><a class="docs-heading-anchor" href="#add_nonlinear_expression"><code>add_nonlinear_expression</code></a><a id="add_nonlinear_expression-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_expression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_expression" href="#JuMP.add_nonlinear_expression"><code>JuMP.add_nonlinear_expression</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_expression(model::Model, expr::Expr)</code></pre><p>Add a nonlinear expression <code>expr</code> to <code>model</code>.</p><p>This function is most useful if the expression <code>expr</code> is generated programmatically, and you cannot use <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a>.</p><p><strong>Notes</strong></p><ul><li>You must interpolate the variables directly into the expression <code>expr</code>.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+(x + x ^ 2.0) - 1.0 ≤ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L443-L465">source</a></section></article><h2 id="add_nonlinear_expression"><a class="docs-heading-anchor" href="#add_nonlinear_expression"><code>add_nonlinear_expression</code></a><a id="add_nonlinear_expression-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_expression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_expression" href="#JuMP.add_nonlinear_expression"><code>JuMP.add_nonlinear_expression</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_expression(model::Model, expr::Expr)</code></pre><p>Add a nonlinear expression <code>expr</code> to <code>model</code>.</p><p>This function is most useful if the expression <code>expr</code> is generated programmatically, and you cannot use <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a>.</p><p><strong>Notes</strong></p><ul><li>You must interpolate the variables directly into the expression <code>expr</code>.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
 julia&gt; add_nonlinear_expression(model, :($(x) + $(x)^2))
-subexpression[1]: x + x ^ 2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L309-L331">source</a></section></article><h2 id="add_nonlinear_operator"><a class="docs-heading-anchor" href="#add_nonlinear_operator"><code>add_nonlinear_operator</code></a><a id="add_nonlinear_operator-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_operator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_operator" href="#JuMP.add_nonlinear_operator"><code>JuMP.add_nonlinear_operator</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_operator(
+subexpression[1]: x + x ^ 2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L309-L331">source</a></section></article><h2 id="add_nonlinear_operator"><a class="docs-heading-anchor" href="#add_nonlinear_operator"><code>add_nonlinear_operator</code></a><a id="add_nonlinear_operator-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_operator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_operator" href="#JuMP.add_nonlinear_operator"><code>JuMP.add_nonlinear_operator</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_operator(
     model::Model,
     dim::Int,
     f::Function,
@@ -314,7 +314,7 @@
 f(x)
 
 julia&gt; op_f(2.0)
-4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L873-L939">source</a></section></article><h2 id="add_nonlinear_parameter"><a class="docs-heading-anchor" href="#add_nonlinear_parameter"><code>add_nonlinear_parameter</code></a><a id="add_nonlinear_parameter-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_parameter" href="#JuMP.add_nonlinear_parameter"><code>JuMP.add_nonlinear_parameter</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_parameter(model::Model, value::Real)</code></pre><p>Add an anonymous parameter to the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L210-L214">source</a></section></article><h2 id="add_to_expression!"><a class="docs-heading-anchor" href="#add_to_expression!"><code>add_to_expression!</code></a><a id="add_to_expression!-1"></a><a class="docs-heading-anchor-permalink" href="#add_to_expression!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_to_expression!" href="#JuMP.add_to_expression!"><code>JuMP.add_to_expression!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_to_expression!(expression, terms...)</code></pre><p>Updates <code>expression</code> <em>in place</em> to <code>expression + (*)(terms...)</code>. This is typically much more efficient than <code>expression += (*)(terms...)</code>. For example, <code>add_to_expression!(expression, a, b)</code> produces the same result as <code>expression += a*b</code>, and <code>add_to_expression!(expression, a)</code> produces the same result as <code>expression += a</code>.</p><p>Only a few methods are defined, mostly for internal use, and only for the cases when (1) they can be implemented efficiently and (2) <code>expression</code> is capable of storing the result. For example, <code>add_to_expression!(::AffExpr, ::GenericVariableRef, ::GenericVariableRef)</code> is not defined because a <code>GenericAffExpr</code> cannot store the product of two variables.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L411-L425">source</a></section></article><h2 id="add_to_function_constant"><a class="docs-heading-anchor" href="#add_to_function_constant"><code>add_to_function_constant</code></a><a id="add_to_function_constant-1"></a><a class="docs-heading-anchor-permalink" href="#add_to_function_constant" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_to_function_constant" href="#JuMP.add_to_function_constant"><code>JuMP.add_to_function_constant</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_to_function_constant(constraint::ConstraintRef, value)</code></pre><p>Add <code>value</code> to the function constant term.</p><p>Note that for scalar constraints, JuMP will aggregate all constant terms onto the right-hand side of the constraint so instead of modifying the function, the set will be translated by <code>-value</code>. For example, given a constraint <code>2x &lt;= 3</code>, <code>add_to_function_constant(c, 4)</code> will modify it to <code>2x &lt;= -1</code>.</p><p><strong>Example</strong></p><p>For scalar constraints, the set is translated by <code>-value</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L873-L939">source</a></section></article><h2 id="add_nonlinear_parameter"><a class="docs-heading-anchor" href="#add_nonlinear_parameter"><code>add_nonlinear_parameter</code></a><a id="add_nonlinear_parameter-1"></a><a class="docs-heading-anchor-permalink" href="#add_nonlinear_parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_nonlinear_parameter" href="#JuMP.add_nonlinear_parameter"><code>JuMP.add_nonlinear_parameter</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_nonlinear_parameter(model::Model, value::Real)</code></pre><p>Add an anonymous parameter to the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L210-L214">source</a></section></article><h2 id="add_to_expression!"><a class="docs-heading-anchor" href="#add_to_expression!"><code>add_to_expression!</code></a><a id="add_to_expression!-1"></a><a class="docs-heading-anchor-permalink" href="#add_to_expression!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_to_expression!" href="#JuMP.add_to_expression!"><code>JuMP.add_to_expression!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_to_expression!(expression, terms...)</code></pre><p>Updates <code>expression</code> <em>in place</em> to <code>expression + (*)(terms...)</code>. This is typically much more efficient than <code>expression += (*)(terms...)</code>. For example, <code>add_to_expression!(expression, a, b)</code> produces the same result as <code>expression += a*b</code>, and <code>add_to_expression!(expression, a)</code> produces the same result as <code>expression += a</code>.</p><p>Only a few methods are defined, mostly for internal use, and only for the cases when (1) they can be implemented efficiently and (2) <code>expression</code> is capable of storing the result. For example, <code>add_to_expression!(::AffExpr, ::GenericVariableRef, ::GenericVariableRef)</code> is not defined because a <code>GenericAffExpr</code> cannot store the product of two variables.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L411-L425">source</a></section></article><h2 id="add_to_function_constant"><a class="docs-heading-anchor" href="#add_to_function_constant"><code>add_to_function_constant</code></a><a id="add_to_function_constant-1"></a><a class="docs-heading-anchor-permalink" href="#add_to_function_constant" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_to_function_constant" href="#JuMP.add_to_function_constant"><code>JuMP.add_to_function_constant</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_to_function_constant(constraint::ConstraintRef, value)</code></pre><p>Add <code>value</code> to the function constant term.</p><p>Note that for scalar constraints, JuMP will aggregate all constant terms onto the right-hand side of the constraint so instead of modifying the function, the set will be translated by <code>-value</code>. For example, given a constraint <code>2x &lt;= 3</code>, <code>add_to_function_constant(c, 4)</code> will modify it to <code>2x &lt;= -1</code>.</p><p><strong>Example</strong></p><p>For scalar constraints, the set is translated by <code>-value</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -336,7 +336,7 @@
 julia&gt; add_to_function_constant(con, [1, 2, 2])
 
 julia&gt; con
-con : [x + y + 1, x + 2, y + 2] ∈ MathOptInterface.SecondOrderCone(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L921-L965">source</a></section></article><h2 id="add_variable"><a class="docs-heading-anchor" href="#add_variable"><code>add_variable</code></a><a id="add_variable-1"></a><a class="docs-heading-anchor-permalink" href="#add_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_variable" href="#JuMP.add_variable"><code>JuMP.add_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_variable(m::GenericModel, v::AbstractVariable, name::String=&quot;&quot;)</code></pre><p>Add a variable <code>v</code> to <code>Model m</code> and sets its name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1725-L1729">source</a></section></article><h2 id="all_constraints"><a class="docs-heading-anchor" href="#all_constraints"><code>all_constraints</code></a><a id="all_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#all_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_constraints" href="#JuMP.all_constraints"><code>JuMP.all_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_constraints(model::GenericModel, function_type, set_type)::Vector{&lt;:ConstraintRef}</code></pre><p>Return a list of all constraints currently in the model where the function has type <code>function_type</code> and the set has type <code>set_type</code>. The constraints are ordered by creation time.</p><p>See also <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and <a href="#JuMP.num_constraints"><code>num_constraints</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+con : [x + y + 1, x + 2, y + 2] ∈ MathOptInterface.SecondOrderCone(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L921-L965">source</a></section></article><h2 id="add_variable"><a class="docs-heading-anchor" href="#add_variable"><code>add_variable</code></a><a id="add_variable-1"></a><a class="docs-heading-anchor-permalink" href="#add_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.add_variable" href="#JuMP.add_variable"><code>JuMP.add_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_variable(m::GenericModel, v::AbstractVariable, name::String=&quot;&quot;)</code></pre><p>Add a variable <code>v</code> to <code>Model m</code> and sets its name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1725-L1729">source</a></section></article><h2 id="all_constraints"><a class="docs-heading-anchor" href="#all_constraints"><code>all_constraints</code></a><a id="all_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#all_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_constraints" href="#JuMP.all_constraints"><code>JuMP.all_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_constraints(model::GenericModel, function_type, set_type)::Vector{&lt;:ConstraintRef}</code></pre><p>Return a list of all constraints currently in the model where the function has type <code>function_type</code> and the set has type <code>set_type</code>. The constraints are ordered by creation time.</p><p>See also <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and <a href="#JuMP.num_constraints"><code>num_constraints</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 0, Bin);
 
@@ -352,7 +352,7 @@
 
 julia&gt; all_constraints(model, AffExpr, MOI.LessThan{Float64})
 1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:
- 2 x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1239-L1269">source</a></section><section><div><pre><code class="language-julia hljs">all_constraints(
+ 2 x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1239-L1269">source</a></section><section><div><pre><code class="language-julia hljs">all_constraints(
     model::GenericModel;
     include_variable_in_set_constraints::Bool,
 )::Vector{ConstraintRef}</code></pre><p>Return a list of all constraints in <code>model</code>.</p><p>If <code>include_variable_in_set_constraints == true</code>, then <code>VariableRef</code> constraints such as <code>VariableRef</code>-in-<code>Integer</code> are included. To return only the structural constraints (e.g., the rows in the constraint matrix of a linear program), pass <code>include_variable_in_set_constraints = false</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
@@ -373,7 +373,7 @@
 julia&gt; all_constraints(model; include_variable_in_set_constraints = false)
 2-element Vector{ConstraintRef}:
  2 x ≤ 1
- x ^ 2.0 - 1.0 ≤ 0</code></pre><p><strong>Performance considerations</strong></p><p>Note that this function is type-unstable because it returns an abstractly typed vector. If performance is a problem, consider using <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and a function barrier. See the <a href="../../developers/extensions/#Performance-tips-for-extensions">Performance tips for extensions</a> section of the documentation for more details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1378-L1421">source</a></section></article><h2 id="all_nonlinear_constraints"><a class="docs-heading-anchor" href="#all_nonlinear_constraints"><code>all_nonlinear_constraints</code></a><a id="all_nonlinear_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#all_nonlinear_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_nonlinear_constraints" href="#JuMP.all_nonlinear_constraints"><code>JuMP.all_nonlinear_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_nonlinear_constraints(model::GenericModel)</code></pre><p>Return a vector of all nonlinear constraint references in the model in the order they were added to the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L512-L517">source</a></section></article><h2 id="all_variables"><a class="docs-heading-anchor" href="#all_variables"><code>all_variables</code></a><a id="all_variables-1"></a><a class="docs-heading-anchor-permalink" href="#all_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_variables" href="#JuMP.all_variables"><code>JuMP.all_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_variables(model::GenericModel{T})::Vector{GenericVariableRef{T}} where {T}</code></pre><p>Returns a list of all variables currently in the model. The variables are ordered by creation time.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ x ^ 2.0 - 1.0 ≤ 0</code></pre><p><strong>Performance considerations</strong></p><p>Note that this function is type-unstable because it returns an abstractly typed vector. If performance is a problem, consider using <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and a function barrier. See the <a href="../../developers/extensions/#Performance-tips-for-extensions">Performance tips for extensions</a> section of the documentation for more details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1378-L1421">source</a></section></article><h2 id="all_nonlinear_constraints"><a class="docs-heading-anchor" href="#all_nonlinear_constraints"><code>all_nonlinear_constraints</code></a><a id="all_nonlinear_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#all_nonlinear_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_nonlinear_constraints" href="#JuMP.all_nonlinear_constraints"><code>JuMP.all_nonlinear_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_nonlinear_constraints(model::GenericModel)</code></pre><p>Return a vector of all nonlinear constraint references in the model in the order they were added to the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L512-L517">source</a></section></article><h2 id="all_variables"><a class="docs-heading-anchor" href="#all_variables"><code>all_variables</code></a><a id="all_variables-1"></a><a class="docs-heading-anchor-permalink" href="#all_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.all_variables" href="#JuMP.all_variables"><code>JuMP.all_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">all_variables(model::GenericModel{T})::Vector{GenericVariableRef{T}} where {T}</code></pre><p>Returns a list of all variables currently in the model. The variables are ordered by creation time.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -382,7 +382,7 @@
 julia&gt; all_variables(model)
 2-element Vector{VariableRef}:
  x
- y</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L2139-L2159">source</a></section></article><h2 id="anonymous_name"><a class="docs-heading-anchor" href="#anonymous_name"><code>anonymous_name</code></a><a id="anonymous_name-1"></a><a class="docs-heading-anchor-permalink" href="#anonymous_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.anonymous_name" href="#JuMP.anonymous_name"><code>JuMP.anonymous_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">anonymous_name(::MIME, x::AbstractVariableRef)</code></pre><p>The name to use for an anonymous variable <code>x</code> when printing.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L547-L551">source</a></section></article><h2 id="backend"><a class="docs-heading-anchor" href="#backend"><code>backend</code></a><a id="backend-1"></a><a class="docs-heading-anchor-permalink" href="#backend" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.backend" href="#JuMP.backend"><code>JuMP.backend</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">backend(model::GenericModel)</code></pre><p>Return the lower-level MathOptInterface model that sits underneath JuMP. This model depends on which operating mode JuMP is in (see <a href="#JuMP.mode"><code>mode</code></a>).</p><ul><li>If JuMP is in <code>DIRECT</code> mode (i.e., the model was created using <a href="#JuMP.direct_model"><code>direct_model</code></a>), the backend will be the optimizer passed to <a href="#JuMP.direct_model"><code>direct_model</code></a>.</li><li>If JuMP is in <code>MANUAL</code> or <code>AUTOMATIC</code> mode, the backend is a <code>MOI.Utilities.CachingOptimizer</code>.</li></ul><p><strong>This function should only be used by advanced users looking to access low-level MathOptInterface or solver-specific functionality.</strong></p><p><strong>Notes</strong></p><p>If JuMP is not in <code>DIRECT</code> mode, the type returned by <code>backend</code> may change between any JuMP releases. Therefore, only use the public API exposed by MathOptInterface, and do not access internal fields. If you require access to the innermost optimizer, see <a href="#JuMP.unsafe_backend"><code>unsafe_backend</code></a>. Alternatively, use <a href="#JuMP.direct_model"><code>direct_model</code></a> to create a JuMP model in <code>DIRECT</code> mode.</p><p>See also: <a href="#JuMP.unsafe_backend"><code>unsafe_backend</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L385-L409">source</a></section></article><h2 id="barrier_iterations"><a class="docs-heading-anchor" href="#barrier_iterations"><code>barrier_iterations</code></a><a id="barrier_iterations-1"></a><a class="docs-heading-anchor-permalink" href="#barrier_iterations" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.barrier_iterations" href="#JuMP.barrier_iterations"><code>JuMP.barrier_iterations</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">barrier_iterations(model::GenericModel)</code></pre><p>Gets the cumulative number of barrier iterations during the most recent optimization.</p><p>Solvers must implement <code>MOI.BarrierIterations()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L600-L607">source</a></section></article><h2 id="bridge_constraints"><a class="docs-heading-anchor" href="#bridge_constraints"><code>bridge_constraints</code></a><a id="bridge_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#bridge_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.bridge_constraints" href="#JuMP.bridge_constraints"><code>JuMP.bridge_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">bridge_constraints(model::GenericModel)</code></pre><p>When in direct mode, return <code>false</code>.</p><p>When in manual or automatic mode, return a <code>Bool</code> indicating whether the optimizer is set and unsupported constraints are automatically bridged to equivalent supported constraints when an appropriate transformation is available.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L569-L578">source</a></section></article><h2 id="build_constraint"><a class="docs-heading-anchor" href="#build_constraint"><code>build_constraint</code></a><a id="build_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#build_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.build_constraint" href="#JuMP.build_constraint"><code>JuMP.build_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_constraint(
+ y</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L2139-L2159">source</a></section></article><h2 id="anonymous_name"><a class="docs-heading-anchor" href="#anonymous_name"><code>anonymous_name</code></a><a id="anonymous_name-1"></a><a class="docs-heading-anchor-permalink" href="#anonymous_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.anonymous_name" href="#JuMP.anonymous_name"><code>JuMP.anonymous_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">anonymous_name(::MIME, x::AbstractVariableRef)</code></pre><p>The name to use for an anonymous variable <code>x</code> when printing.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L547-L551">source</a></section></article><h2 id="backend"><a class="docs-heading-anchor" href="#backend"><code>backend</code></a><a id="backend-1"></a><a class="docs-heading-anchor-permalink" href="#backend" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.backend" href="#JuMP.backend"><code>JuMP.backend</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">backend(model::GenericModel)</code></pre><p>Return the lower-level MathOptInterface model that sits underneath JuMP. This model depends on which operating mode JuMP is in (see <a href="#JuMP.mode"><code>mode</code></a>).</p><ul><li>If JuMP is in <code>DIRECT</code> mode (i.e., the model was created using <a href="#JuMP.direct_model"><code>direct_model</code></a>), the backend will be the optimizer passed to <a href="#JuMP.direct_model"><code>direct_model</code></a>.</li><li>If JuMP is in <code>MANUAL</code> or <code>AUTOMATIC</code> mode, the backend is a <code>MOI.Utilities.CachingOptimizer</code>.</li></ul><p><strong>This function should only be used by advanced users looking to access low-level MathOptInterface or solver-specific functionality.</strong></p><p><strong>Notes</strong></p><p>If JuMP is not in <code>DIRECT</code> mode, the type returned by <code>backend</code> may change between any JuMP releases. Therefore, only use the public API exposed by MathOptInterface, and do not access internal fields. If you require access to the innermost optimizer, see <a href="#JuMP.unsafe_backend"><code>unsafe_backend</code></a>. Alternatively, use <a href="#JuMP.direct_model"><code>direct_model</code></a> to create a JuMP model in <code>DIRECT</code> mode.</p><p>See also: <a href="#JuMP.unsafe_backend"><code>unsafe_backend</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L385-L409">source</a></section></article><h2 id="barrier_iterations"><a class="docs-heading-anchor" href="#barrier_iterations"><code>barrier_iterations</code></a><a id="barrier_iterations-1"></a><a class="docs-heading-anchor-permalink" href="#barrier_iterations" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.barrier_iterations" href="#JuMP.barrier_iterations"><code>JuMP.barrier_iterations</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">barrier_iterations(model::GenericModel)</code></pre><p>Gets the cumulative number of barrier iterations during the most recent optimization.</p><p>Solvers must implement <code>MOI.BarrierIterations()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L600-L607">source</a></section></article><h2 id="bridge_constraints"><a class="docs-heading-anchor" href="#bridge_constraints"><code>bridge_constraints</code></a><a id="bridge_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#bridge_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.bridge_constraints" href="#JuMP.bridge_constraints"><code>JuMP.bridge_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">bridge_constraints(model::GenericModel)</code></pre><p>When in direct mode, return <code>false</code>.</p><p>When in manual or automatic mode, return a <code>Bool</code> indicating whether the optimizer is set and unsupported constraints are automatically bridged to equivalent supported constraints when an appropriate transformation is available.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L569-L578">source</a></section></article><h2 id="build_constraint"><a class="docs-heading-anchor" href="#build_constraint"><code>build_constraint</code></a><a id="build_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#build_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.build_constraint" href="#JuMP.build_constraint"><code>JuMP.build_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_constraint(
     _error::Function,
     Q::LinearAlgebra.Symmetric{V, M},
     ::PSDCone,
@@ -403,7 +403,7 @@
 
 julia&gt; @constraint(model, Q in PSDCone())
 [Q[1,1]  Q[1,2];
- Q[1,2]  Q[2,2]] ∈ PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L435-L473">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
+ Q[1,2]  Q[2,2]] ∈ PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L435-L473">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
     _error::Function,
     Q::AbstractMatrix{&lt;:AbstractJuMPScalar},
     ::PSDCone,
@@ -413,7 +413,7 @@
 
 julia&gt; @constraint(model, Q in PSDCone())
 [Q[1,1]  Q[1,2];
- Q[2,1]  Q[2,2]] ∈ PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L483-L503">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
+ Q[2,1]  Q[2,2]] ∈ PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L483-L503">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
     _error::Function,
     Q::LinearAlgebra.Hermitian{V,M},
     ::HermitianPSDCone,
@@ -425,44 +425,44 @@
 
 julia&gt; @constraint(model, LinearAlgebra.Hermitian(Q) in HermitianPSDCone())
 [Q[1,1]  Q[1,2];
- Q[1,2]  Q[2,2]] ∈ HermitianPSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L643-L665">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
+ Q[1,2]  Q[2,2]] ∈ HermitianPSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L643-L665">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
     _error::Function,
     f::AbstractVector{&lt;:AbstractJuMPScalar},
     ::Nonnegatives,
     extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},
-)</code></pre><p>A helper method that re-writes</p><pre><code class="language-julia hljs">@constraint(model, X &gt;= Y, extra)</code></pre><p>into</p><pre><code class="language-julia hljs">@constraint(model, X - Y in extra)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L44-L60">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
+)</code></pre><p>A helper method that re-writes</p><pre><code class="language-julia hljs">@constraint(model, X &gt;= Y, extra)</code></pre><p>into</p><pre><code class="language-julia hljs">@constraint(model, X - Y in extra)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L44-L60">source</a></section><section><div><pre><code class="language-julia hljs">build_constraint(
     _error::Function,
     f::AbstractVector{&lt;:AbstractJuMPScalar},
     ::Nonpositives,
     extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},
-)</code></pre><p>A helper method that re-writes</p><pre><code class="language-julia hljs">@constraint(model, Y &lt;= X, extra)</code></pre><p>into</p><pre><code class="language-julia hljs">@constraint(model, X - Y in extra)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L70-L86">source</a></section></article><h2 id="build_variable"><a class="docs-heading-anchor" href="#build_variable"><code>build_variable</code></a><a id="build_variable-1"></a><a class="docs-heading-anchor-permalink" href="#build_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.build_variable" href="#JuMP.build_variable"><code>JuMP.build_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_variable(
+)</code></pre><p>A helper method that re-writes</p><pre><code class="language-julia hljs">@constraint(model, Y &lt;= X, extra)</code></pre><p>into</p><pre><code class="language-julia hljs">@constraint(model, X - Y in extra)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L70-L86">source</a></section></article><h2 id="build_variable"><a class="docs-heading-anchor" href="#build_variable"><code>build_variable</code></a><a id="build_variable-1"></a><a class="docs-heading-anchor-permalink" href="#build_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.build_variable" href="#JuMP.build_variable"><code>JuMP.build_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">build_variable(
     _error::Function,
     info::VariableInfo,
     args...;
     kwargs...,
 )</code></pre><p>Return a new <a href="#JuMP.AbstractVariable"><code>AbstractVariable</code></a> object.</p><p>This method should only be implemented by developers creating JuMP extensions. It should never be called by users of JuMP.</p><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function to call instead of <code>error</code>. <code>_error</code> annotates the error message with additional information for the user.</li><li><code>info</code>: an instance of <a href="#JuMP.VariableInfo"><code>VariableInfo</code></a>. This has a variety of fields relating to the variable such as <code>info.lower_bound</code> and <code>info.binary</code>.</li><li><code>args</code>: optional additional positional arguments for extending the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</li><li><code>kwargs</code>: optional keyword arguments for extending the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</li></ul><p>See also: <a href="#JuMP.@variable"><code>@variable</code></a></p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Extensions should define a method with ONE positional argument to dispatch the call to a different method. Creating an extension that relies on multiple positional arguments leads to <code>MethodError</code>s if the user passes the arguments in the wrong order.</p></div></div><p><strong>Example</strong></p><pre><code class="language-julia hljs">@variable(model, x, Foo)</code></pre><p>will call</p><pre><code class="language-julia hljs">build_variable(_error::Function, info::VariableInfo, ::Type{Foo})</code></pre><p>Passing special-case positional arguments such as <code>Bin</code>, <code>Int</code>, and <code>PSD</code> is okay, along with keyword arguments:</p><pre><code class="language-julia hljs">@variable(model, x, Int, Foo(), mykwarg = true)
 # or
-@variable(model, x, Foo(), Int, mykwarg = true)</code></pre><p>will call</p><pre><code class="language-julia hljs">build_variable(_error::Function, info::VariableInfo, ::Foo; mykwarg)</code></pre><p>and <code>info.integer</code> will be true.</p><p>Note that the order of the positional arguments does not matter.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1977-L2033">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::SymmetricMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+@variable(model, x, Foo(), Int, mykwarg = true)</code></pre><p>will call</p><pre><code class="language-julia hljs">build_variable(_error::Function, info::VariableInfo, ::Foo; mykwarg)</code></pre><p>and <code>info.integer</code> will be true.</p><p>Note that the order of the positional arguments does not matter.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1977-L2033">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::SymmetricMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2], Symmetric)
 2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:
  Q[1,1]  Q[1,2]
- Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L302-L318">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::SkewSymmetricMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SkewSymmetricMatrixShape"><code>SkewSymmetricMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L302-L318">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::SkewSymmetricMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SkewSymmetricMatrixShape"><code>SkewSymmetricMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())
 2×2 Matrix{AffExpr}:
  0        Q[1,2]
- -Q[1,2]  0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L334-L350">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::HermitianMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.HermitianMatrixShape"><code>HermitianMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ -Q[1,2]  0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L334-L350">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::HermitianMatrixSpace)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.HermitianMatrixShape"><code>HermitianMatrixShape</code></a> creating variables in <code>MOI.Reals</code>, i.e. &quot;free&quot; variables unless they are constrained after their creation.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())
 2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:
  real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im
- real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L366-L382">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::PSDCone)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> constraining the variables to be positive semidefinite.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L366-L382">source</a></section><section><div><pre><code class="language-julia hljs">build_variable(_error::Function, variables, ::PSDCone)</code></pre><p>Return a <code>VariablesConstrainedOnCreation</code> of shape <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> constraining the variables to be positive semidefinite.</p><p>This function is used by the <a href="#JuMP.@variable"><code>@variable</code></a> macro as follows:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2], PSD)
 2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:
  Q[1,1]  Q[1,2]
- Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L400-L415">source</a></section></article><h2 id="callback_node_status"><a class="docs-heading-anchor" href="#callback_node_status"><code>callback_node_status</code></a><a id="callback_node_status-1"></a><a class="docs-heading-anchor-permalink" href="#callback_node_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.callback_node_status" href="#JuMP.callback_node_status"><code>JuMP.callback_node_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">callback_node_status(cb_data, model::GenericModel)</code></pre><p>Return an <a href="../../moi/reference/callbacks/#MathOptInterface.CallbackNodeStatusCode"><code>MOI.CallbackNodeStatusCode</code></a> enum, indicating if the current primal solution available from <a href="#JuMP.callback_value"><code>callback_value</code></a> is integer feasible.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/callbacks.jl#L11-L16">source</a></section></article><h2 id="callback_value"><a class="docs-heading-anchor" href="#callback_value"><code>callback_value</code></a><a id="callback_value-1"></a><a class="docs-heading-anchor-permalink" href="#callback_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.callback_value" href="#JuMP.callback_value"><code>JuMP.callback_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">callback_value(cb_data, x::GenericVariableRef)</code></pre><p>Return the primal solution of a variable inside a callback.</p><p><code>cb_data</code> is the argument to the callback function, and the type is dependent on the solver.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/callbacks.jl#L30-L37">source</a></section><section><div><pre><code class="language-julia hljs">callback_value(cb_data, expr::Union{GenericAffExpr, GenericQuadExpr})</code></pre><p>Return the primal solution of an affine or quadratic expression inside a callback by getting the value for each variable appearing in the expression.</p><p><code>cb_data</code> is the argument to the callback function, and the type is dependent on the solver.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/callbacks.jl#L55-L63">source</a></section></article><h2 id="check_belongs_to_model"><a class="docs-heading-anchor" href="#check_belongs_to_model"><code>check_belongs_to_model</code></a><a id="check_belongs_to_model-1"></a><a class="docs-heading-anchor-permalink" href="#check_belongs_to_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.check_belongs_to_model" href="#JuMP.check_belongs_to_model"><code>JuMP.check_belongs_to_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">check_belongs_to_model(func::AbstractJuMPScalar, model::AbstractModel)</code></pre><p>Throw <a href="#JuMP.VariableNotOwned"><code>VariableNotOwned</code></a> if the <a href="#JuMP.owner_model"><code>owner_model</code></a> of one of the variables of the function <code>func</code> is not <code>model</code>.</p><pre><code class="nohighlight hljs">check_belongs_to_model(constraint::AbstractConstraint, model::AbstractModel)</code></pre><p>Throw <a href="#JuMP.VariableNotOwned"><code>VariableNotOwned</code></a> if the <a href="#JuMP.owner_model"><code>owner_model</code></a> of one of the variables of the constraint <code>constraint</code> is not <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L329-L339">source</a></section></article><h2 id="coefficient"><a class="docs-heading-anchor" href="#coefficient"><code>coefficient</code></a><a id="coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.coefficient" href="#JuMP.coefficient"><code>JuMP.coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">coefficient(v1::GenericVariableRef{T}, v2::GenericVariableRef{T}) where {T}</code></pre><p>Return <code>one(T)</code> if <code>v1 == v2</code>, and <code>zero(T)</code> otherwise.</p><p>This is a fallback for other <a href="#JuMP.coefficient"><code>coefficient</code></a> methods to simplify code in which the expression may be a single variable.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L368-L375">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericAffExpr{C,V}, v::V) where {C,V}</code></pre><p>Return the coefficient associated with variable <code>v</code> in the affine expression <code>a</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L247-L251">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericAffExpr{C,V}, v1::V, v2::V) where {C,V}</code></pre><p>Return the coefficient associated with the term <code>v1 * v2</code> in the quadratic expression <code>a</code>.</p><p>Note that <code>coefficient(a, v1, v2)</code> is the same as <code>coefficient(a, v2, v1)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L148-L154">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericQuadExpr{C,V}, v::V) where {C,V}</code></pre><p>Return the coefficient associated with variable <code>v</code> in the affine component of <code>a</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L159-L163">source</a></section></article><h2 id="compute_conflict!"><a class="docs-heading-anchor" href="#compute_conflict!"><code>compute_conflict!</code></a><a id="compute_conflict!-1"></a><a class="docs-heading-anchor-permalink" href="#compute_conflict!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.compute_conflict!" href="#JuMP.compute_conflict!"><code>JuMP.compute_conflict!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">compute_conflict!(model::GenericModel)</code></pre><p>Compute a conflict if the model is infeasible. If an optimizer has not been set yet (see <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>), a <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> error is thrown.</p><p>The status of the conflict can be checked with the <code>MOI.ConflictStatus</code> model attribute. Then, the status for each constraint can be queried with the <code>MOI.ConstraintConflictStatus</code> attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L478-L488">source</a></section></article><h2 id="constant"><a class="docs-heading-anchor" href="#constant"><code>constant</code></a><a id="constant-1"></a><a class="docs-heading-anchor-permalink" href="#constant" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constant" href="#JuMP.constant"><code>JuMP.constant</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constant(aff::GenericAffExpr{C, V})::C</code></pre><p>Return the constant of the affine expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L361-L365">source</a></section><section><div><pre><code class="language-julia hljs">constant(aff::GenericQuadExpr{C, V})::C</code></pre><p>Return the constant of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L259-L263">source</a></section></article><h2 id="constraint_by_name"><a class="docs-heading-anchor" href="#constraint_by_name"><code>constraint_by_name</code></a><a id="constraint_by_name-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_by_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_by_name" href="#JuMP.constraint_by_name"><code>JuMP.constraint_by_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_by_name(model::AbstractModel,
+ Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L400-L415">source</a></section></article><h2 id="callback_node_status"><a class="docs-heading-anchor" href="#callback_node_status"><code>callback_node_status</code></a><a id="callback_node_status-1"></a><a class="docs-heading-anchor-permalink" href="#callback_node_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.callback_node_status" href="#JuMP.callback_node_status"><code>JuMP.callback_node_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">callback_node_status(cb_data, model::GenericModel)</code></pre><p>Return an <a href="../../moi/reference/callbacks/#MathOptInterface.CallbackNodeStatusCode"><code>MOI.CallbackNodeStatusCode</code></a> enum, indicating if the current primal solution available from <a href="#JuMP.callback_value"><code>callback_value</code></a> is integer feasible.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/callbacks.jl#L11-L16">source</a></section></article><h2 id="callback_value"><a class="docs-heading-anchor" href="#callback_value"><code>callback_value</code></a><a id="callback_value-1"></a><a class="docs-heading-anchor-permalink" href="#callback_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.callback_value" href="#JuMP.callback_value"><code>JuMP.callback_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">callback_value(cb_data, x::GenericVariableRef)</code></pre><p>Return the primal solution of a variable inside a callback.</p><p><code>cb_data</code> is the argument to the callback function, and the type is dependent on the solver.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/callbacks.jl#L30-L37">source</a></section><section><div><pre><code class="language-julia hljs">callback_value(cb_data, expr::Union{GenericAffExpr, GenericQuadExpr})</code></pre><p>Return the primal solution of an affine or quadratic expression inside a callback by getting the value for each variable appearing in the expression.</p><p><code>cb_data</code> is the argument to the callback function, and the type is dependent on the solver.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/callbacks.jl#L55-L63">source</a></section></article><h2 id="check_belongs_to_model"><a class="docs-heading-anchor" href="#check_belongs_to_model"><code>check_belongs_to_model</code></a><a id="check_belongs_to_model-1"></a><a class="docs-heading-anchor-permalink" href="#check_belongs_to_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.check_belongs_to_model" href="#JuMP.check_belongs_to_model"><code>JuMP.check_belongs_to_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">check_belongs_to_model(func::AbstractJuMPScalar, model::AbstractModel)</code></pre><p>Throw <a href="#JuMP.VariableNotOwned"><code>VariableNotOwned</code></a> if the <a href="#JuMP.owner_model"><code>owner_model</code></a> of one of the variables of the function <code>func</code> is not <code>model</code>.</p><pre><code class="nohighlight hljs">check_belongs_to_model(constraint::AbstractConstraint, model::AbstractModel)</code></pre><p>Throw <a href="#JuMP.VariableNotOwned"><code>VariableNotOwned</code></a> if the <a href="#JuMP.owner_model"><code>owner_model</code></a> of one of the variables of the constraint <code>constraint</code> is not <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L329-L339">source</a></section></article><h2 id="coefficient"><a class="docs-heading-anchor" href="#coefficient"><code>coefficient</code></a><a id="coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.coefficient" href="#JuMP.coefficient"><code>JuMP.coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">coefficient(v1::GenericVariableRef{T}, v2::GenericVariableRef{T}) where {T}</code></pre><p>Return <code>one(T)</code> if <code>v1 == v2</code>, and <code>zero(T)</code> otherwise.</p><p>This is a fallback for other <a href="#JuMP.coefficient"><code>coefficient</code></a> methods to simplify code in which the expression may be a single variable.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L368-L375">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericAffExpr{C,V}, v::V) where {C,V}</code></pre><p>Return the coefficient associated with variable <code>v</code> in the affine expression <code>a</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L247-L251">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericAffExpr{C,V}, v1::V, v2::V) where {C,V}</code></pre><p>Return the coefficient associated with the term <code>v1 * v2</code> in the quadratic expression <code>a</code>.</p><p>Note that <code>coefficient(a, v1, v2)</code> is the same as <code>coefficient(a, v2, v1)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L148-L154">source</a></section><section><div><pre><code class="language-julia hljs">coefficient(a::GenericQuadExpr{C,V}, v::V) where {C,V}</code></pre><p>Return the coefficient associated with variable <code>v</code> in the affine component of <code>a</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L159-L163">source</a></section></article><h2 id="compute_conflict!"><a class="docs-heading-anchor" href="#compute_conflict!"><code>compute_conflict!</code></a><a id="compute_conflict!-1"></a><a class="docs-heading-anchor-permalink" href="#compute_conflict!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.compute_conflict!" href="#JuMP.compute_conflict!"><code>JuMP.compute_conflict!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">compute_conflict!(model::GenericModel)</code></pre><p>Compute a conflict if the model is infeasible. If an optimizer has not been set yet (see <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>), a <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> error is thrown.</p><p>The status of the conflict can be checked with the <code>MOI.ConflictStatus</code> model attribute. Then, the status for each constraint can be queried with the <code>MOI.ConstraintConflictStatus</code> attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L478-L488">source</a></section></article><h2 id="constant"><a class="docs-heading-anchor" href="#constant"><code>constant</code></a><a id="constant-1"></a><a class="docs-heading-anchor-permalink" href="#constant" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constant" href="#JuMP.constant"><code>JuMP.constant</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constant(aff::GenericAffExpr{C, V})::C</code></pre><p>Return the constant of the affine expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L361-L365">source</a></section><section><div><pre><code class="language-julia hljs">constant(aff::GenericQuadExpr{C, V})::C</code></pre><p>Return the constant of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L259-L263">source</a></section></article><h2 id="constraint_by_name"><a class="docs-heading-anchor" href="#constraint_by_name"><code>constraint_by_name</code></a><a id="constraint_by_name-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_by_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_by_name" href="#JuMP.constraint_by_name"><code>JuMP.constraint_by_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_by_name(model::AbstractModel,
                    name::String)::Union{ConstraintRef, Nothing}</code></pre><p>Return the reference of the constraint with name attribute <code>name</code> or <code>Nothing</code> if no constraint has this name attribute. Throws an error if several constraints have <code>name</code> as their name attribute.</p><pre><code class="nohighlight hljs">constraint_by_name(model::AbstractModel,
                    name::String,
                    F::Type{&lt;:Union{AbstractJuMPScalar,
@@ -484,10 +484,10 @@
 julia&gt; constraint_by_name(model, &quot;con&quot;, AffExpr, MOI.EqualTo{Float64})
 
 julia&gt; constraint_by_name(model, &quot;con&quot;, QuadExpr, MOI.EqualTo{Float64})
-con : x² = 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L268-L309">source</a></section></article><h2 id="constraint_object"><a class="docs-heading-anchor" href="#constraint_object"><code>constraint_object</code></a><a id="constraint_object-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_object" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_object" href="#JuMP.constraint_object"><code>JuMP.constraint_object</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_object(con_ref::ConstraintRef)</code></pre><p>Return the underlying constraint data for the constraint referenced by <code>ref</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L570-L574">source</a></section></article><h2 id="constraint_ref_with_index"><a class="docs-heading-anchor" href="#constraint_ref_with_index"><code>constraint_ref_with_index</code></a><a id="constraint_ref_with_index-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_ref_with_index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_ref_with_index" href="#JuMP.constraint_ref_with_index"><code>JuMP.constraint_ref_with_index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_ref_with_index(model::AbstractModel, index::MOI.ConstraintIndex)</code></pre><p>Return a <code>ConstraintRef</code> of <code>model</code> corresponding to <code>index</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L343-L347">source</a></section></article><h2 id="constraint_string"><a class="docs-heading-anchor" href="#constraint_string"><code>constraint_string</code></a><a id="constraint_string-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_string" href="#JuMP.constraint_string"><code>JuMP.constraint_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_string(
+con : x² = 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L268-L309">source</a></section></article><h2 id="constraint_object"><a class="docs-heading-anchor" href="#constraint_object"><code>constraint_object</code></a><a id="constraint_object-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_object" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_object" href="#JuMP.constraint_object"><code>JuMP.constraint_object</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_object(con_ref::ConstraintRef)</code></pre><p>Return the underlying constraint data for the constraint referenced by <code>ref</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L570-L574">source</a></section></article><h2 id="constraint_ref_with_index"><a class="docs-heading-anchor" href="#constraint_ref_with_index"><code>constraint_ref_with_index</code></a><a id="constraint_ref_with_index-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_ref_with_index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_ref_with_index" href="#JuMP.constraint_ref_with_index"><code>JuMP.constraint_ref_with_index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_ref_with_index(model::AbstractModel, index::MOI.ConstraintIndex)</code></pre><p>Return a <code>ConstraintRef</code> of <code>model</code> corresponding to <code>index</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L343-L347">source</a></section></article><h2 id="constraint_string"><a class="docs-heading-anchor" href="#constraint_string"><code>constraint_string</code></a><a id="constraint_string-1"></a><a class="docs-heading-anchor-permalink" href="#constraint_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraint_string" href="#JuMP.constraint_string"><code>JuMP.constraint_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_string(
     mode::MIME,
     ref::ConstraintRef;
-    in_math_mode::Bool = false)</code></pre><p>Return a string representation of the constraint <code>ref</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L792-L799">source</a></section></article><h2 id="constraints_string"><a class="docs-heading-anchor" href="#constraints_string"><code>constraints_string</code></a><a id="constraints_string-1"></a><a class="docs-heading-anchor-permalink" href="#constraints_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraints_string" href="#JuMP.constraints_string"><code>JuMP.constraints_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraints_string(mode, model::AbstractModel)::Vector{String}</code></pre><p>Return a list of <code>String</code>s describing each constraint of the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L414-L418">source</a></section></article><h2 id="copy_conflict"><a class="docs-heading-anchor" href="#copy_conflict"><code>copy_conflict</code></a><a id="copy_conflict-1"></a><a class="docs-heading-anchor-permalink" href="#copy_conflict" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_conflict" href="#JuMP.copy_conflict"><code>JuMP.copy_conflict</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_conflict(model::GenericModel)</code></pre><p>Return a copy of the current conflict for the model <code>model</code> and a <a href="#JuMP.GenericReferenceMap"><code>GenericReferenceMap</code></a> that can be used to obtain the variable and constraint reference of the new model corresponding to a given <code>model</code>&#39;s reference.</p><p>This is a convenience function that provides a filtering function for <a href="#JuMP.copy_model"><code>copy_model</code></a>.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and two constraints <code>c1</code> and <code>c2</code>. This model has no solution, as the two constraints are mutually exclusive. The solver is asked to compute a conflict with <a href="#JuMP.compute_conflict!"><code>compute_conflict!</code></a>. The parts of <code>model</code> participating in the conflict are then copied into a model <code>iis_model</code>.</p><pre><code class="language-julia hljs">julia&gt; using JuMP
+    in_math_mode::Bool = false)</code></pre><p>Return a string representation of the constraint <code>ref</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L792-L799">source</a></section></article><h2 id="constraints_string"><a class="docs-heading-anchor" href="#constraints_string"><code>constraints_string</code></a><a id="constraints_string-1"></a><a class="docs-heading-anchor-permalink" href="#constraints_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.constraints_string" href="#JuMP.constraints_string"><code>JuMP.constraints_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraints_string(mode, model::AbstractModel)::Vector{String}</code></pre><p>Return a list of <code>String</code>s describing each constraint of the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L414-L418">source</a></section></article><h2 id="copy_conflict"><a class="docs-heading-anchor" href="#copy_conflict"><code>copy_conflict</code></a><a id="copy_conflict-1"></a><a class="docs-heading-anchor-permalink" href="#copy_conflict" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_conflict" href="#JuMP.copy_conflict"><code>JuMP.copy_conflict</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_conflict(model::GenericModel)</code></pre><p>Return a copy of the current conflict for the model <code>model</code> and a <a href="#JuMP.GenericReferenceMap"><code>GenericReferenceMap</code></a> that can be used to obtain the variable and constraint reference of the new model corresponding to a given <code>model</code>&#39;s reference.</p><p>This is a convenience function that provides a filtering function for <a href="#JuMP.copy_model"><code>copy_model</code></a>.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and two constraints <code>c1</code> and <code>c2</code>. This model has no solution, as the two constraints are mutually exclusive. The solver is asked to compute a conflict with <a href="#JuMP.compute_conflict!"><code>compute_conflict!</code></a>. The parts of <code>model</code> participating in the conflict are then copied into a model <code>iis_model</code>.</p><pre><code class="language-julia hljs">julia&gt; using JuMP
 
 julia&gt; import Gurobi
 
@@ -515,7 +515,7 @@
 Feasibility
 Subject to
  c1 : x ≥ 2
- c2 : x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L247-L305">source</a></section></article><h2 id="copy_extension_data"><a class="docs-heading-anchor" href="#copy_extension_data"><code>copy_extension_data</code></a><a id="copy_extension_data-1"></a><a class="docs-heading-anchor-permalink" href="#copy_extension_data" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_extension_data" href="#JuMP.copy_extension_data"><code>JuMP.copy_extension_data</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_extension_data(data, new_model::AbstractModel, model::AbstractModel)</code></pre><p>Return a copy of the extension data <code>data</code> of the model <code>model</code> to the extension data of the new model <code>new_model</code>.</p><p>A method should be added for any JuMP extension storing data in the <code>ext</code> field.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Do not engage in type piracy by implementing this method for types of <code>data</code> that you did not define! JuMP extensions should store types that they define in <code>model.ext</code>, rather than regular Julia types.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L6-L18">source</a></section></article><h2 id="copy_model"><a class="docs-heading-anchor" href="#copy_model"><code>copy_model</code></a><a id="copy_model-1"></a><a class="docs-heading-anchor-permalink" href="#copy_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_model" href="#JuMP.copy_model"><code>JuMP.copy_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_model(model::GenericModel; filter_constraints::Union{Nothing, Function}=nothing)</code></pre><p>Return a copy of the model <code>model</code> and a <a href="#JuMP.GenericReferenceMap"><code>GenericReferenceMap</code></a> that can be used to obtain the variable and constraint reference of the new model corresponding to a given <code>model</code>&#39;s reference. A <a href="#Base.copy-Tuple{AbstractModel}"><code>Base.copy(::AbstractModel)</code></a> method has also been implemented, it is similar to <code>copy_model</code> but does not return the reference map.</p><p>If the <code>filter_constraints</code> argument is given, only the constraints for which this function returns <code>true</code> will be copied. This function is given a constraint reference as argument.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and a constraint <code>cref</code>. It is then copied into a model <code>new_model</code> with the new references assigned to <code>x_new</code> and <code>cref_new</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ c2 : x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L247-L305">source</a></section></article><h2 id="copy_extension_data"><a class="docs-heading-anchor" href="#copy_extension_data"><code>copy_extension_data</code></a><a id="copy_extension_data-1"></a><a class="docs-heading-anchor-permalink" href="#copy_extension_data" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_extension_data" href="#JuMP.copy_extension_data"><code>JuMP.copy_extension_data</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_extension_data(data, new_model::AbstractModel, model::AbstractModel)</code></pre><p>Return a copy of the extension data <code>data</code> of the model <code>model</code> to the extension data of the new model <code>new_model</code>.</p><p>A method should be added for any JuMP extension storing data in the <code>ext</code> field.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Do not engage in type piracy by implementing this method for types of <code>data</code> that you did not define! JuMP extensions should store types that they define in <code>model.ext</code>, rather than regular Julia types.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L6-L18">source</a></section></article><h2 id="copy_model"><a class="docs-heading-anchor" href="#copy_model"><code>copy_model</code></a><a id="copy_model-1"></a><a class="docs-heading-anchor-permalink" href="#copy_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.copy_model" href="#JuMP.copy_model"><code>JuMP.copy_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_model(model::GenericModel; filter_constraints::Union{Nothing, Function}=nothing)</code></pre><p>Return a copy of the model <code>model</code> and a <a href="#JuMP.GenericReferenceMap"><code>GenericReferenceMap</code></a> that can be used to obtain the variable and constraint reference of the new model corresponding to a given <code>model</code>&#39;s reference. A <a href="#Base.copy-Tuple{AbstractModel}"><code>Base.copy(::AbstractModel)</code></a> method has also been implemented, it is similar to <code>copy_model</code> but does not return the reference map.</p><p>If the <code>filter_constraints</code> argument is given, only the constraints for which this function returns <code>true</code> will be copied. This function is given a constraint reference as argument.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and a constraint <code>cref</code>. It is then copied into a model <code>new_model</code> with the new references assigned to <code>x_new</code> and <code>cref_new</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -529,7 +529,7 @@
 x
 
 julia&gt; cref_new = reference_map[cref]
-cref : x = 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L93-L137">source</a></section></article><h2 id="delete"><a class="docs-heading-anchor" href="#delete"><code>delete</code></a><a id="delete-1"></a><a class="docs-heading-anchor-permalink" href="#delete" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete" href="#JuMP.delete"><code>JuMP.delete</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, con_ref::ConstraintRef)</code></pre><p>Delete the constraint associated with <code>constraint_ref</code> from the model <code>model</code>.</p><p>Note that <code>delete</code> does not unregister the name from the model, so adding a new constraint of the same name will throw an error. Use <a href="#JuMP.unregister"><code>unregister</code></a> to unregister the name after deletion.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+cref : x = 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L93-L137">source</a></section></article><h2 id="delete"><a class="docs-heading-anchor" href="#delete"><code>delete</code></a><a id="delete-1"></a><a class="docs-heading-anchor-permalink" href="#delete" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete" href="#JuMP.delete"><code>JuMP.delete</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, con_ref::ConstraintRef)</code></pre><p>Delete the constraint associated with <code>constraint_ref</code> from the model <code>model</code>.</p><p>Note that <code>delete</code> does not unregister the name from the model, so adding a new constraint of the same name will throw an error. Use <a href="#JuMP.unregister"><code>unregister</code></a> to unregister the name after deletion.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -547,7 +547,7 @@
 julia&gt; model[:c]
 ERROR: KeyError: key :c not found
 Stacktrace:
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L367-L399">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, con_refs::Vector{&lt;:ConstraintRef})</code></pre><p>Delete the constraints associated with <code>con_refs</code> from the model <code>model</code>. Solvers may implement specialized methods for deleting multiple constraints of the same concrete type, i.e., when <code>isconcretetype(eltype(con_refs))</code>. These may be more efficient than repeatedly calling the single constraint delete method.</p><p>See also: <a href="#JuMP.unregister"><code>unregister</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L411-L421">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, variable_ref::GenericVariableRef)</code></pre><p>Delete the variable associated with <code>variable_ref</code> from the model <code>model</code>.</p><p>Note that <code>delete</code> does not unregister the name from the model, so adding a new variable of the same name will throw an error. Use <a href="#JuMP.unregister"><code>unregister</code></a> to unregister the name after deletion.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L367-L399">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, con_refs::Vector{&lt;:ConstraintRef})</code></pre><p>Delete the constraints associated with <code>con_refs</code> from the model <code>model</code>. Solvers may implement specialized methods for deleting multiple constraints of the same concrete type, i.e., when <code>isconcretetype(eltype(con_refs))</code>. These may be more efficient than repeatedly calling the single constraint delete method.</p><p>See also: <a href="#JuMP.unregister"><code>unregister</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L411-L421">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, variable_ref::GenericVariableRef)</code></pre><p>Delete the variable associated with <code>variable_ref</code> from the model <code>model</code>.</p><p>Note that <code>delete</code> does not unregister the name from the model, so adding a new variable of the same name will throw an error. Use <a href="#JuMP.unregister"><code>unregister</code></a> to unregister the name after deletion.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -563,7 +563,7 @@
 julia&gt; model[:x]
 ERROR: KeyError: key :x not found
 Stacktrace:
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L399-L429">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, variable_refs::Vector{&lt;:GenericVariableRef})</code></pre><p>Delete the variables associated with <code>variable_refs</code> from the model <code>model</code>. Solvers may implement methods for deleting multiple variables that are more efficient than repeatedly calling the single variable delete method.</p><p>See also: <a href="#JuMP.unregister"><code>unregister</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L442-L450">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::Model, c::NonlinearConstraintRef)</code></pre><p>Delete the nonlinear constraint <code>c</code> from <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L487-L491">source</a></section></article><h2 id="delete_lower_bound"><a class="docs-heading-anchor" href="#delete_lower_bound"><code>delete_lower_bound</code></a><a id="delete_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#delete_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete_lower_bound" href="#JuMP.delete_lower_bound"><code>JuMP.delete_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete_lower_bound(v::GenericVariableRef)</code></pre><p>Delete the lower bound constraint of a variable.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L399-L429">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::GenericModel, variable_refs::Vector{&lt;:GenericVariableRef})</code></pre><p>Delete the variables associated with <code>variable_refs</code> from the model <code>model</code>. Solvers may implement methods for deleting multiple variables that are more efficient than repeatedly calling the single variable delete method.</p><p>See also: <a href="#JuMP.unregister"><code>unregister</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L442-L450">source</a></section><section><div><pre><code class="language-julia hljs">delete(model::Model, c::NonlinearConstraintRef)</code></pre><p>Delete the nonlinear constraint <code>c</code> from <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L487-L491">source</a></section></article><h2 id="delete_lower_bound"><a class="docs-heading-anchor" href="#delete_lower_bound"><code>delete_lower_bound</code></a><a id="delete_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#delete_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete_lower_bound" href="#JuMP.delete_lower_bound"><code>JuMP.delete_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete_lower_bound(v::GenericVariableRef)</code></pre><p>Delete the lower bound constraint of a variable.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 1.0);
 
@@ -573,7 +573,7 @@
 julia&gt; delete_lower_bound(x)
 
 julia&gt; has_lower_bound(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L787-L810">source</a></section></article><h2 id="delete_upper_bound"><a class="docs-heading-anchor" href="#delete_upper_bound"><code>delete_upper_bound</code></a><a id="delete_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#delete_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete_upper_bound" href="#JuMP.delete_upper_bound"><code>JuMP.delete_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete_upper_bound(v::GenericVariableRef)</code></pre><p>Delete the upper bound constraint of a variable.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L787-L810">source</a></section></article><h2 id="delete_upper_bound"><a class="docs-heading-anchor" href="#delete_upper_bound"><code>delete_upper_bound</code></a><a id="delete_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#delete_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.delete_upper_bound" href="#JuMP.delete_upper_bound"><code>JuMP.delete_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">delete_upper_bound(v::GenericVariableRef)</code></pre><p>Delete the upper bound constraint of a variable.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &lt;= 1.0);
 
@@ -583,10 +583,10 @@
 julia&gt; delete_upper_bound(x)
 
 julia&gt; has_upper_bound(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L958-L983">source</a></section></article><h2 id="direct_generic_model"><a class="docs-heading-anchor" href="#direct_generic_model"><code>direct_generic_model</code></a><a id="direct_generic_model-1"></a><a class="docs-heading-anchor-permalink" href="#direct_generic_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.direct_generic_model" href="#JuMP.direct_generic_model"><code>JuMP.direct_generic_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">direct_generic_model(
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L958-L983">source</a></section></article><h2 id="direct_generic_model"><a class="docs-heading-anchor" href="#direct_generic_model"><code>direct_generic_model</code></a><a id="direct_generic_model-1"></a><a class="docs-heading-anchor-permalink" href="#direct_generic_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.direct_generic_model" href="#JuMP.direct_generic_model"><code>JuMP.direct_generic_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">direct_generic_model(
     value_type::Type{T},
     backend::MOI.ModelLike;
-) where {T&lt;:Real}</code></pre><p>Return a new JuMP model using <a href="#JuMP.backend"><code>backend</code></a> to store the model and solve it.</p><p>As opposed to the <a href="#JuMP.Model"><code>Model</code></a> constructor, no cache of the model is stored outside of <a href="#JuMP.backend"><code>backend</code></a> and no bridges are automatically applied to <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Notes</strong></p><p>The absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this <em>direct</em> mode:</p><ul><li>When <a href="#JuMP.backend"><code>backend</code></a> does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the <a href="#JuMP.Model"><code>Model</code></a> constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when <code>optimize!</code> is called.</li><li>No constraint bridging is supported by default.</li><li>The optimizer used cannot be changed the model is constructed.</li><li>The model created cannot be copied.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L188-L213">source</a></section><section><div><pre><code class="language-julia hljs">direct_generic_model(::Type{T}, factory::MOI.OptimizerWithAttributes)</code></pre><p>Create a <a href="#JuMP.direct_generic_model"><code>direct_generic_model</code></a> using <code>factory</code>, a <code>MOI.OptimizerWithAttributes</code> object created by <a href="#JuMP.optimizer_with_attributes"><code>optimizer_with_attributes</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
+) where {T&lt;:Real}</code></pre><p>Return a new JuMP model using <a href="#JuMP.backend"><code>backend</code></a> to store the model and solve it.</p><p>As opposed to the <a href="#JuMP.Model"><code>Model</code></a> constructor, no cache of the model is stored outside of <a href="#JuMP.backend"><code>backend</code></a> and no bridges are automatically applied to <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Notes</strong></p><p>The absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this <em>direct</em> mode:</p><ul><li>When <a href="#JuMP.backend"><code>backend</code></a> does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the <a href="#JuMP.Model"><code>Model</code></a> constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when <code>optimize!</code> is called.</li><li>No constraint bridging is supported by default.</li><li>The optimizer used cannot be changed the model is constructed.</li><li>The model created cannot be copied.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L188-L213">source</a></section><section><div><pre><code class="language-julia hljs">direct_generic_model(::Type{T}, factory::MOI.OptimizerWithAttributes)</code></pre><p>Create a <a href="#JuMP.direct_generic_model"><code>direct_generic_model</code></a> using <code>factory</code>, a <code>MOI.OptimizerWithAttributes</code> object created by <a href="#JuMP.optimizer_with_attributes"><code>optimizer_with_attributes</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
 
 julia&gt; optimizer = optimizer_with_attributes(
            HiGHS.Optimizer,
@@ -610,7 +610,7 @@
 
 julia&gt; set_attribute(model, &quot;presolve&quot;, &quot;off&quot;)
 
-julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L233-L272">source</a></section></article><h2 id="direct_model"><a class="docs-heading-anchor" href="#direct_model"><code>direct_model</code></a><a id="direct_model-1"></a><a class="docs-heading-anchor-permalink" href="#direct_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.direct_model" href="#JuMP.direct_model"><code>JuMP.direct_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">direct_model(backend::MOI.ModelLike)</code></pre><p>Return a new JuMP model using <a href="#JuMP.backend"><code>backend</code></a> to store the model and solve it.</p><p>As opposed to the <a href="#JuMP.Model"><code>Model</code></a> constructor, no cache of the model is stored outside of <a href="#JuMP.backend"><code>backend</code></a> and no bridges are automatically applied to <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Notes</strong></p><p>The absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this <em>direct</em> mode:</p><ul><li>When <a href="#JuMP.backend"><code>backend</code></a> does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the <a href="#JuMP.Model"><code>Model</code></a> constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when <code>optimize!</code> is called.</li><li>No constraint bridging is supported by default.</li><li>The optimizer used cannot be changed the model is constructed.</li><li>The model created cannot be copied.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L313-L335">source</a></section><section><div><pre><code class="language-julia hljs">direct_model(factory::MOI.OptimizerWithAttributes)</code></pre><p>Create a <a href="#JuMP.direct_model"><code>direct_model</code></a> using <code>factory</code>, a <code>MOI.OptimizerWithAttributes</code> object created by <a href="#JuMP.optimizer_with_attributes"><code>optimizer_with_attributes</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
+julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L233-L272">source</a></section></article><h2 id="direct_model"><a class="docs-heading-anchor" href="#direct_model"><code>direct_model</code></a><a id="direct_model-1"></a><a class="docs-heading-anchor-permalink" href="#direct_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.direct_model" href="#JuMP.direct_model"><code>JuMP.direct_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">direct_model(backend::MOI.ModelLike)</code></pre><p>Return a new JuMP model using <a href="#JuMP.backend"><code>backend</code></a> to store the model and solve it.</p><p>As opposed to the <a href="#JuMP.Model"><code>Model</code></a> constructor, no cache of the model is stored outside of <a href="#JuMP.backend"><code>backend</code></a> and no bridges are automatically applied to <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Notes</strong></p><p>The absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this <em>direct</em> mode:</p><ul><li>When <a href="#JuMP.backend"><code>backend</code></a> does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the <a href="#JuMP.Model"><code>Model</code></a> constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when <code>optimize!</code> is called.</li><li>No constraint bridging is supported by default.</li><li>The optimizer used cannot be changed the model is constructed.</li><li>The model created cannot be copied.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L313-L335">source</a></section><section><div><pre><code class="language-julia hljs">direct_model(factory::MOI.OptimizerWithAttributes)</code></pre><p>Create a <a href="#JuMP.direct_model"><code>direct_model</code></a> using <code>factory</code>, a <code>MOI.OptimizerWithAttributes</code> object created by <a href="#JuMP.optimizer_with_attributes"><code>optimizer_with_attributes</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
 
 julia&gt; optimizer = optimizer_with_attributes(
            HiGHS.Optimizer,
@@ -634,7 +634,7 @@
 
 julia&gt; set_attribute(model, &quot;presolve&quot;, &quot;off&quot;)
 
-julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L338-L377">source</a></section></article><h2 id="drop_zeros!"><a class="docs-heading-anchor" href="#drop_zeros!"><code>drop_zeros!</code></a><a id="drop_zeros!-1"></a><a class="docs-heading-anchor-permalink" href="#drop_zeros!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.drop_zeros!" href="#JuMP.drop_zeros!"><code>JuMP.drop_zeros!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">drop_zeros!(expr::GenericAffExpr)</code></pre><p>Remove terms in the affine expression with <code>0</code> coefficients.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L255-L259">source</a></section><section><div><pre><code class="language-julia hljs">drop_zeros!(expr::GenericQuadExpr)</code></pre><p>Remove terms in the quadratic expression with <code>0</code> coefficients.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L166-L170">source</a></section></article><h2 id="dual"><a class="docs-heading-anchor" href="#dual"><code>dual</code></a><a id="dual-1"></a><a class="docs-heading-anchor-permalink" href="#dual" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual" href="#JuMP.dual"><code>JuMP.dual</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual(con_ref::ConstraintRef; result::Int = 1)</code></pre><p>Return the dual value of constraint <code>con_ref</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Use <code>has_dual</code> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>, <a href="#JuMP.shadow_price"><code>shadow_price</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1041-L1050">source</a></section><section><div><pre><code class="language-julia hljs">dual(c::NonlinearConstraintRef)</code></pre><p>Return the dual of the nonlinear constraint <code>c</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L562-L566">source</a></section></article><h2 id="dual_objective_value"><a class="docs-heading-anchor" href="#dual_objective_value"><code>dual_objective_value</code></a><a id="dual_objective_value-1"></a><a class="docs-heading-anchor-permalink" href="#dual_objective_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_objective_value" href="#JuMP.dual_objective_value"><code>JuMP.dual_objective_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_objective_value(model::GenericModel; result::Int = 1)</code></pre><p>Return the value of the objective of the dual problem associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Throws <code>MOI.UnsupportedAttribute{MOI.DualObjectiveValue}</code> if the solver does not support this attribute.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L57-L67">source</a></section></article><h2 id="dual_shape"><a class="docs-heading-anchor" href="#dual_shape"><code>dual_shape</code></a><a id="dual_shape-1"></a><a class="docs-heading-anchor-permalink" href="#dual_shape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_shape" href="#JuMP.dual_shape"><code>JuMP.dual_shape</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_shape(shape::AbstractShape)::AbstractShape</code></pre><p>Returns the shape of the dual space of the space of objects of shape <code>shape</code>. By default, the <code>dual_shape</code> of a shape is itself. See the examples section below for an example for which this is not the case.</p><p><strong>Example</strong></p><p>Consider polynomial constraints for which the dual is moment constraints and moment constraints for which the dual is polynomial constraints. Shapes for polynomials can be defined as follows:</p><pre><code class="language-julia hljs">struct Polynomial
+julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L338-L377">source</a></section></article><h2 id="drop_zeros!"><a class="docs-heading-anchor" href="#drop_zeros!"><code>drop_zeros!</code></a><a id="drop_zeros!-1"></a><a class="docs-heading-anchor-permalink" href="#drop_zeros!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.drop_zeros!" href="#JuMP.drop_zeros!"><code>JuMP.drop_zeros!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">drop_zeros!(expr::GenericAffExpr)</code></pre><p>Remove terms in the affine expression with <code>0</code> coefficients.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L255-L259">source</a></section><section><div><pre><code class="language-julia hljs">drop_zeros!(expr::GenericQuadExpr)</code></pre><p>Remove terms in the quadratic expression with <code>0</code> coefficients.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L166-L170">source</a></section></article><h2 id="dual"><a class="docs-heading-anchor" href="#dual"><code>dual</code></a><a id="dual-1"></a><a class="docs-heading-anchor-permalink" href="#dual" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual" href="#JuMP.dual"><code>JuMP.dual</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual(con_ref::ConstraintRef; result::Int = 1)</code></pre><p>Return the dual value of constraint <code>con_ref</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Use <code>has_dual</code> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>, <a href="#JuMP.shadow_price"><code>shadow_price</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1041-L1050">source</a></section><section><div><pre><code class="language-julia hljs">dual(c::NonlinearConstraintRef)</code></pre><p>Return the dual of the nonlinear constraint <code>c</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L562-L566">source</a></section></article><h2 id="dual_objective_value"><a class="docs-heading-anchor" href="#dual_objective_value"><code>dual_objective_value</code></a><a id="dual_objective_value-1"></a><a class="docs-heading-anchor-permalink" href="#dual_objective_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_objective_value" href="#JuMP.dual_objective_value"><code>JuMP.dual_objective_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_objective_value(model::GenericModel; result::Int = 1)</code></pre><p>Return the value of the objective of the dual problem associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Throws <code>MOI.UnsupportedAttribute{MOI.DualObjectiveValue}</code> if the solver does not support this attribute.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L57-L67">source</a></section></article><h2 id="dual_shape"><a class="docs-heading-anchor" href="#dual_shape"><code>dual_shape</code></a><a id="dual_shape-1"></a><a class="docs-heading-anchor-permalink" href="#dual_shape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_shape" href="#JuMP.dual_shape"><code>JuMP.dual_shape</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_shape(shape::AbstractShape)::AbstractShape</code></pre><p>Returns the shape of the dual space of the space of objects of shape <code>shape</code>. By default, the <code>dual_shape</code> of a shape is itself. See the examples section below for an example for which this is not the case.</p><p><strong>Example</strong></p><p>Consider polynomial constraints for which the dual is moment constraints and moment constraints for which the dual is polynomial constraints. Shapes for polynomials can be defined as follows:</p><pre><code class="language-julia hljs">struct Polynomial
     coefficients::Vector{Float64}
     monomials::Vector{Monomial}
 end
@@ -649,7 +649,7 @@
     monomials::Vector{Monomial}
 end
 JuMP.reshape_vector(x::Vector, shape::MomentsShape) = Moments(x, shape.monomials)</code></pre><p>Then <code>dual_shape</code> allows the definition of the shape of the dual of polynomial and moment constraints:</p><pre><code class="language-julia hljs">dual_shape(shape::PolynomialShape) = MomentsShape(shape.monomials)
-dual_shape(shape::MomentsShape) = PolynomialShape(shape.monomials)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L19-L58">source</a></section></article><h2 id="dual_start_value"><a class="docs-heading-anchor" href="#dual_start_value"><code>dual_start_value</code></a><a id="dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_start_value" href="#JuMP.dual_start_value"><code>JuMP.dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_start_value(con_ref::ConstraintRef)</code></pre><p>Return the dual start value (MOI attribute <code>ConstraintDualStart</code>) of the constraint <code>con_ref</code>.</p><p>Note: If no dual start value has been set, <code>dual_start_value</code> will return <code>nothing</code>.</p><p>See also <a href="#JuMP.set_dual_start_value"><code>set_dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L71-L81">source</a></section></article><h2 id="dual_status"><a class="docs-heading-anchor" href="#dual_status"><code>dual_status</code></a><a id="dual_status-1"></a><a class="docs-heading-anchor-permalink" href="#dual_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_status" href="#JuMP.dual_status"><code>JuMP.dual_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_status(model::GenericModel; result::Int = 1)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.ResultStatusCode"><code>MOI.ResultStatusCode</code></a> describing the status of the most recent dual solution of the solver (i.e., the <a href="../../moi/reference/models/#MathOptInterface.DualStatus"><code>MOI.DualStatus</code></a> attribute) associated with the result index <code>result</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L563-L571">source</a></section></article><h2 id="error_if_direct_mode"><a class="docs-heading-anchor" href="#error_if_direct_mode"><code>error_if_direct_mode</code></a><a id="error_if_direct_mode-1"></a><a class="docs-heading-anchor-permalink" href="#error_if_direct_mode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.error_if_direct_mode" href="#JuMP.error_if_direct_mode"><code>JuMP.error_if_direct_mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">error_if_direct_mode(model::GenericModel, func::Symbol)</code></pre><p>Errors if <code>model</code> is in direct mode during a call from the function named <code>func</code>.</p><p>Used internally within JuMP, or by JuMP extensions who do not want to support models in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L250-L258">source</a></section></article><h2 id="fix"><a class="docs-heading-anchor" href="#fix"><code>fix</code></a><a id="fix-1"></a><a class="docs-heading-anchor-permalink" href="#fix" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix" href="#JuMP.fix"><code>JuMP.fix</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix(v::GenericVariableRef, value::Number; force::Bool = false)</code></pre><p>Fix a variable to a value. Update the fixing constraint if one exists, otherwise create a new one.</p><p>If the variable already has variable bounds and <code>force=false</code>, calling <code>fix</code> will throw an error. If <code>force=true</code>, existing variable bounds will be deleted, and the fixing constraint will be added. Note a variable will have no bounds after a call to <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+dual_shape(shape::MomentsShape) = PolynomialShape(shape.monomials)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L19-L58">source</a></section></article><h2 id="dual_start_value"><a class="docs-heading-anchor" href="#dual_start_value"><code>dual_start_value</code></a><a id="dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_start_value" href="#JuMP.dual_start_value"><code>JuMP.dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_start_value(con_ref::ConstraintRef)</code></pre><p>Return the dual start value (MOI attribute <code>ConstraintDualStart</code>) of the constraint <code>con_ref</code>.</p><p>Note: If no dual start value has been set, <code>dual_start_value</code> will return <code>nothing</code>.</p><p>See also <a href="#JuMP.set_dual_start_value"><code>set_dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L71-L81">source</a></section></article><h2 id="dual_status"><a class="docs-heading-anchor" href="#dual_status"><code>dual_status</code></a><a id="dual_status-1"></a><a class="docs-heading-anchor-permalink" href="#dual_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.dual_status" href="#JuMP.dual_status"><code>JuMP.dual_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">dual_status(model::GenericModel; result::Int = 1)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.ResultStatusCode"><code>MOI.ResultStatusCode</code></a> describing the status of the most recent dual solution of the solver (i.e., the <a href="../../moi/reference/models/#MathOptInterface.DualStatus"><code>MOI.DualStatus</code></a> attribute) associated with the result index <code>result</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L563-L571">source</a></section></article><h2 id="error_if_direct_mode"><a class="docs-heading-anchor" href="#error_if_direct_mode"><code>error_if_direct_mode</code></a><a id="error_if_direct_mode-1"></a><a class="docs-heading-anchor-permalink" href="#error_if_direct_mode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.error_if_direct_mode" href="#JuMP.error_if_direct_mode"><code>JuMP.error_if_direct_mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">error_if_direct_mode(model::GenericModel, func::Symbol)</code></pre><p>Errors if <code>model</code> is in direct mode during a call from the function named <code>func</code>.</p><p>Used internally within JuMP, or by JuMP extensions who do not want to support models in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L250-L258">source</a></section></article><h2 id="fix"><a class="docs-heading-anchor" href="#fix"><code>fix</code></a><a id="fix-1"></a><a class="docs-heading-anchor-permalink" href="#fix" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix" href="#JuMP.fix"><code>JuMP.fix</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix(v::GenericVariableRef, value::Number; force::Bool = false)</code></pre><p>Fix a variable to a value. Update the fixing constraint if one exists, otherwise create a new one.</p><p>If the variable already has variable bounds and <code>force=false</code>, calling <code>fix</code> will throw an error. If <code>force=true</code>, existing variable bounds will be deleted, and the fixing constraint will be added. Note a variable will have no bounds after a call to <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -669,7 +669,7 @@
 julia&gt; fix(x, 1.0; force = true)
 
 julia&gt; is_fixed(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1053-L1096">source</a></section></article><h2 id="fix_discrete_variables"><a class="docs-heading-anchor" href="#fix_discrete_variables"><code>fix_discrete_variables</code></a><a id="fix_discrete_variables-1"></a><a class="docs-heading-anchor-permalink" href="#fix_discrete_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix_discrete_variables" href="#JuMP.fix_discrete_variables"><code>JuMP.fix_discrete_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix_discrete_variables([var_value::Function = value,] model::GenericModel)</code></pre><p>Modifies <code>model</code> to convert all binary and integer variables to continuous variables with fixed bounds of <code>var_value(x)</code>.</p><p><strong>Return</strong></p><p>Returns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.</p><p><strong>Notes</strong></p><ul><li>An error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).</li><li>All other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1053-L1096">source</a></section></article><h2 id="fix_discrete_variables"><a class="docs-heading-anchor" href="#fix_discrete_variables"><code>fix_discrete_variables</code></a><a id="fix_discrete_variables-1"></a><a class="docs-heading-anchor-permalink" href="#fix_discrete_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix_discrete_variables" href="#JuMP.fix_discrete_variables"><code>JuMP.fix_discrete_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix_discrete_variables([var_value::Function = value,] model::GenericModel)</code></pre><p>Modifies <code>model</code> to convert all binary and integer variables to continuous variables with fixed bounds of <code>var_value(x)</code>.</p><p><strong>Return</strong></p><p>Returns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.</p><p><strong>Notes</strong></p><ul><li>An error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).</li><li>All other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Bin, start = 1);
 
@@ -693,12 +693,12 @@
  y ≥ 1
  y ≤ 10
  y integer
- x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L2273-L2321">source</a></section></article><h2 id="fix_value"><a class="docs-heading-anchor" href="#fix_value"><code>fix_value</code></a><a id="fix_value-1"></a><a class="docs-heading-anchor-permalink" href="#fix_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix_value" href="#JuMP.fix_value"><code>JuMP.fix_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix_value(v::GenericVariableRef)</code></pre><p>Return the value to which a variable is fixed.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L2273-L2321">source</a></section></article><h2 id="fix_value"><a class="docs-heading-anchor" href="#fix_value"><code>fix_value</code></a><a id="fix_value-1"></a><a class="docs-heading-anchor-permalink" href="#fix_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.fix_value" href="#JuMP.fix_value"><code>JuMP.fix_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">fix_value(v::GenericVariableRef)</code></pre><p>Return the value to which a variable is fixed.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x == 1);
 
 julia&gt; fix_value(x)
-1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1171-L1190">source</a></section></article><h2 id="flatten!"><a class="docs-heading-anchor" href="#flatten!"><code>flatten!</code></a><a id="flatten!-1"></a><a class="docs-heading-anchor-permalink" href="#flatten!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.flatten!" href="#JuMP.flatten!"><code>JuMP.flatten!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">flatten!(expr::GenericNonlinearExpr)</code></pre><p>Flatten a nonlinear expression in-place by lifting nested <code>+</code> and <code>*</code> nodes into a single n-ary operation.</p><p><strong>Motivation</strong></p><p>Nonlinear expressions created using operator overloading can be deeply nested and unbalanced. For example, <code>prod(x for i in 1:4)</code> creates <code>*(x, *(x, *(x, x)))</code> instead of the more preferable <code>*(x, x, x, x)</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1171-L1190">source</a></section></article><h2 id="flatten!"><a class="docs-heading-anchor" href="#flatten!"><code>flatten!</code></a><a id="flatten!-1"></a><a class="docs-heading-anchor-permalink" href="#flatten!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.flatten!" href="#JuMP.flatten!"><code>JuMP.flatten!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">flatten!(expr::GenericNonlinearExpr)</code></pre><p>Flatten a nonlinear expression in-place by lifting nested <code>+</code> and <code>*</code> nodes into a single n-ary operation.</p><p><strong>Motivation</strong></p><p>Nonlinear expressions created using operator overloading can be deeply nested and unbalanced. For example, <code>prod(x for i in 1:4)</code> creates <code>*(x, *(x, *(x, x)))</code> instead of the more preferable <code>*(x, x, x, x)</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -710,10 +710,10 @@
 (x²) * x * x
 
 julia&gt; flatten!(sin(prod(x for i in 1:4)))
-sin((x²) * x * x)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L414-L443">source</a></section></article><h2 id="function_string"><a class="docs-heading-anchor" href="#function_string"><code>function_string</code></a><a id="function_string-1"></a><a class="docs-heading-anchor-permalink" href="#function_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.function_string" href="#JuMP.function_string"><code>JuMP.function_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">function_string(
+sin((x²) * x * x)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L414-L443">source</a></section></article><h2 id="function_string"><a class="docs-heading-anchor" href="#function_string"><code>function_string</code></a><a id="function_string-1"></a><a class="docs-heading-anchor-permalink" href="#function_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.function_string" href="#JuMP.function_string"><code>JuMP.function_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">function_string(
     mode::MIME,
     func::Union{JuMP.AbstractJuMPScalar,Vector{&lt;:JuMP.AbstractJuMPScalar}},
-)</code></pre><p>Return a <code>String</code> representing the function <code>func</code> using print mode <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L562-L569">source</a></section></article><h2 id="get_attribute"><a class="docs-heading-anchor" href="#get_attribute"><code>get_attribute</code></a><a id="get_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#get_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.get_attribute" href="#JuMP.get_attribute"><code>JuMP.get_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute)
+)</code></pre><p>Return a <code>String</code> representing the function <code>func</code> using print mode <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L562-L569">source</a></section></article><h2 id="get_attribute"><a class="docs-heading-anchor" href="#get_attribute"><code>get_attribute</code></a><a id="get_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#get_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.get_attribute" href="#JuMP.get_attribute"><code>JuMP.get_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute)
 get_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute)
 get_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute)</code></pre><p>Get the value of a solver-specifc attribute <code>attr</code>.</p><p>This is equivalent to calling <a href="../../moi/reference/models/#MathOptInterface.get"><code>MOI.get</code></a> with the associated MOI model and, for variables and constraints, with the associated <a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> or <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintIndex"><code>MOI.ConstraintIndex</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
@@ -730,7 +730,7 @@
 &quot;x&quot;
 
 julia&gt; get_attribute(c, MOI.ConstraintName())
-&quot;c&quot;</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L760-L791">source</a></section><section><div><pre><code class="language-julia hljs">get_attribute(
+&quot;c&quot;</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L760-L791">source</a></section><section><div><pre><code class="language-julia hljs">get_attribute(
     model::Union{GenericModel,MOI.OptimizerWithAttributes},
     attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},
 )</code></pre><p>Get the value of a solver-specifc attribute <code>attr</code>.</p><p>This is equivalent to calling <a href="../../moi/reference/models/#MathOptInterface.get"><code>MOI.get</code></a> with the associated MOI model.</p><p>If <code>attr</code> is an <code>AbstractString</code>, it is converted to <a href="../../moi/reference/models/#MathOptInterface.RawOptimizerAttribute"><code>MOI.RawOptimizerAttribute</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
@@ -749,7 +749,7 @@
 true
 
 julia&gt; get_attribute(opt, MOI.RawOptimizerAttribute(&quot;output_flag&quot;))
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L807-L841">source</a></section></article><h2 id="get_optimizer_attribute"><a class="docs-heading-anchor" href="#get_optimizer_attribute"><code>get_optimizer_attribute</code></a><a id="get_optimizer_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#get_optimizer_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.get_optimizer_attribute" href="#JuMP.get_optimizer_attribute"><code>JuMP.get_optimizer_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get_optimizer_attribute(
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L807-L841">source</a></section></article><h2 id="get_optimizer_attribute"><a class="docs-heading-anchor" href="#get_optimizer_attribute"><code>get_optimizer_attribute</code></a><a id="get_optimizer_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#get_optimizer_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.get_optimizer_attribute" href="#JuMP.get_optimizer_attribute"><code>JuMP.get_optimizer_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get_optimizer_attribute(
     model::Union{GenericModel,MOI.OptimizerWithAttributes},
     attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},
 )</code></pre><p>Return the value associated with the solver-specific attribute <code>attr</code>.</p><p>If <code>attr</code> is an <code>AbstractString</code>, this is equivalent to <code>get_optimizer_attribute(model, MOI.RawOptimizerAttribute(name))</code>.</p><div class="admonition is-compat"><header class="admonition-header">Compat</header><div class="admonition-body"><p>This method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using <a href="#JuMP.get_attribute"><code>get_attribute</code></a> instead.</p></div></div><p>See also: <a href="#JuMP.set_optimizer_attribute"><code>set_optimizer_attribute</code></a>, <a href="#JuMP.set_optimizer_attributes"><code>set_optimizer_attributes</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import Ipopt
@@ -757,22 +757,22 @@
 julia&gt; model = Model(Ipopt.Optimizer);
 
 julia&gt; get_optimizer_attribute(model, MOI.Silent())
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L124-L151">source</a></section></article><h2 id="has_duals"><a class="docs-heading-anchor" href="#has_duals"><code>has_duals</code></a><a id="has_duals-1"></a><a class="docs-heading-anchor-permalink" href="#has_duals" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_duals" href="#JuMP.has_duals"><code>JuMP.has_duals</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_duals(model::GenericModel; result::Int = 1)</code></pre><p>Return <code>true</code> if the solver has a dual solution in result index <code>result</code> available to query, otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.dual"><code>dual</code></a>, <a href="#JuMP.shadow_price"><code>shadow_price</code></a>, and <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1029-L1036">source</a></section></article><h2 id="has_lower_bound"><a class="docs-heading-anchor" href="#has_lower_bound"><code>has_lower_bound</code></a><a id="has_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#has_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_lower_bound" href="#JuMP.has_lower_bound"><code>JuMP.has_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_lower_bound(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> has a lower bound. If <code>true</code>, the lower bound can be queried with <a href="#JuMP.lower_bound"><code>lower_bound</code></a>.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L124-L151">source</a></section></article><h2 id="has_duals"><a class="docs-heading-anchor" href="#has_duals"><code>has_duals</code></a><a id="has_duals-1"></a><a class="docs-heading-anchor-permalink" href="#has_duals" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_duals" href="#JuMP.has_duals"><code>JuMP.has_duals</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_duals(model::GenericModel; result::Int = 1)</code></pre><p>Return <code>true</code> if the solver has a dual solution in result index <code>result</code> available to query, otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.dual"><code>dual</code></a>, <a href="#JuMP.shadow_price"><code>shadow_price</code></a>, and <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1029-L1036">source</a></section></article><h2 id="has_lower_bound"><a class="docs-heading-anchor" href="#has_lower_bound"><code>has_lower_bound</code></a><a id="has_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#has_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_lower_bound" href="#JuMP.has_lower_bound"><code>JuMP.has_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_lower_bound(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> has a lower bound. If <code>true</code>, the lower bound can be queried with <a href="#JuMP.lower_bound"><code>lower_bound</code></a>.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 1.0);
 
 julia&gt; has_lower_bound(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L669-L688">source</a></section></article><h2 id="has_start_value"><a class="docs-heading-anchor" href="#has_start_value"><code>has_start_value</code></a><a id="has_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#has_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_start_value" href="#JuMP.has_start_value"><code>JuMP.has_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_start_value(variable::AbstractVariableRef)</code></pre><p>Return <code>true</code> if the variable has a start value set otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1653-L1659">source</a></section></article><h2 id="has_upper_bound"><a class="docs-heading-anchor" href="#has_upper_bound"><code>has_upper_bound</code></a><a id="has_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#has_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_upper_bound" href="#JuMP.has_upper_bound"><code>JuMP.has_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_upper_bound(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> has a upper bound. If <code>true</code>, the upper bound can be queried with <a href="#JuMP.upper_bound"><code>upper_bound</code></a>.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L669-L688">source</a></section></article><h2 id="has_start_value"><a class="docs-heading-anchor" href="#has_start_value"><code>has_start_value</code></a><a id="has_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#has_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_start_value" href="#JuMP.has_start_value"><code>JuMP.has_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_start_value(variable::AbstractVariableRef)</code></pre><p>Return <code>true</code> if the variable has a start value set otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1653-L1659">source</a></section></article><h2 id="has_upper_bound"><a class="docs-heading-anchor" href="#has_upper_bound"><code>has_upper_bound</code></a><a id="has_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#has_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_upper_bound" href="#JuMP.has_upper_bound"><code>JuMP.has_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_upper_bound(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> has a upper bound. If <code>true</code>, the upper bound can be queried with <a href="#JuMP.upper_bound"><code>upper_bound</code></a>.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &lt;= 1.0);
 
 julia&gt; has_upper_bound(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L842-L861">source</a></section></article><h2 id="has_values"><a class="docs-heading-anchor" href="#has_values"><code>has_values</code></a><a id="has_values-1"></a><a class="docs-heading-anchor-permalink" href="#has_values" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_values" href="#JuMP.has_values"><code>JuMP.has_values</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_values(model::GenericModel; result::Int = 1)</code></pre><p>Return <code>true</code> if the solver has a primal solution in result index <code>result</code> available to query, otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.value"><code>value</code></a> and <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1713-L1720">source</a></section></article><h2 id="in_set_string"><a class="docs-heading-anchor" href="#in_set_string"><code>in_set_string</code></a><a id="in_set_string-1"></a><a class="docs-heading-anchor-permalink" href="#in_set_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.in_set_string" href="#JuMP.in_set_string"><code>JuMP.in_set_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">in_set_string(mode::MIME, set)</code></pre><p>Return a <code>String</code> representing the membership to the set <code>set</code> using print mode <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L722-L727">source</a></section></article><h2 id="index"><a class="docs-heading-anchor" href="#index"><code>index</code></a><a id="index-1"></a><a class="docs-heading-anchor-permalink" href="#index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.index" href="#JuMP.index"><code>JuMP.index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">index(cr::ConstraintRef)::MOI.ConstraintIndex</code></pre><p>Return the index of the constraint that corresponds to <code>cr</code> in the MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L32-L36">source</a></section><section><div><pre><code class="language-julia hljs">index(v::GenericVariableRef)::MOI.VariableIndex</code></pre><p>Return the index of the variable that corresponds to <code>v</code> in the MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L487-L491">source</a></section><section><div><pre><code class="language-julia hljs">index(p::NonlinearParameter)::MOI.Nonlinear.ParameterIndex</code></pre><p>Return the index of the nonlinear parameter associated with <code>p</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L221-L225">source</a></section><section><div><pre><code class="language-julia hljs">index(ex::NonlinearExpression)::MOI.Nonlinear.ExpressionIndex</code></pre><p>Return the index of the nonlinear expression associated with <code>ex</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L302-L306">source</a></section></article><h2 id="is_binary"><a class="docs-heading-anchor" href="#is_binary"><code>is_binary</code></a><a id="is_binary-1"></a><a class="docs-heading-anchor-permalink" href="#is_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_binary" href="#JuMP.is_binary"><code>JuMP.is_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_binary(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is constrained to be binary.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L842-L861">source</a></section></article><h2 id="has_values"><a class="docs-heading-anchor" href="#has_values"><code>has_values</code></a><a id="has_values-1"></a><a class="docs-heading-anchor-permalink" href="#has_values" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.has_values" href="#JuMP.has_values"><code>JuMP.has_values</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">has_values(model::GenericModel; result::Int = 1)</code></pre><p>Return <code>true</code> if the solver has a primal solution in result index <code>result</code> available to query, otherwise return <code>false</code>.</p><p>See also <a href="#JuMP.value"><code>value</code></a> and <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1713-L1720">source</a></section></article><h2 id="in_set_string"><a class="docs-heading-anchor" href="#in_set_string"><code>in_set_string</code></a><a id="in_set_string-1"></a><a class="docs-heading-anchor-permalink" href="#in_set_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.in_set_string" href="#JuMP.in_set_string"><code>JuMP.in_set_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">in_set_string(mode::MIME, set)</code></pre><p>Return a <code>String</code> representing the membership to the set <code>set</code> using print mode <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L722-L727">source</a></section></article><h2 id="index"><a class="docs-heading-anchor" href="#index"><code>index</code></a><a id="index-1"></a><a class="docs-heading-anchor-permalink" href="#index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.index" href="#JuMP.index"><code>JuMP.index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">index(cr::ConstraintRef)::MOI.ConstraintIndex</code></pre><p>Return the index of the constraint that corresponds to <code>cr</code> in the MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L32-L36">source</a></section><section><div><pre><code class="language-julia hljs">index(v::GenericVariableRef)::MOI.VariableIndex</code></pre><p>Return the index of the variable that corresponds to <code>v</code> in the MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L487-L491">source</a></section><section><div><pre><code class="language-julia hljs">index(p::NonlinearParameter)::MOI.Nonlinear.ParameterIndex</code></pre><p>Return the index of the nonlinear parameter associated with <code>p</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L221-L225">source</a></section><section><div><pre><code class="language-julia hljs">index(ex::NonlinearExpression)::MOI.Nonlinear.ExpressionIndex</code></pre><p>Return the index of the nonlinear expression associated with <code>ex</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L302-L306">source</a></section></article><h2 id="is_binary"><a class="docs-heading-anchor" href="#is_binary"><code>is_binary</code></a><a id="is_binary-1"></a><a class="docs-heading-anchor-permalink" href="#is_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_binary" href="#JuMP.is_binary"><code>JuMP.is_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_binary(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is constrained to be binary.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Bin);
 
 julia&gt; is_binary(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1363-L1380">source</a></section></article><h2 id="is_fixed"><a class="docs-heading-anchor" href="#is_fixed"><code>is_fixed</code></a><a id="is_fixed-1"></a><a class="docs-heading-anchor-permalink" href="#is_fixed" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_fixed" href="#JuMP.is_fixed"><code>JuMP.is_fixed</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_fixed(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is a fixed variable. If <code>true</code>, the fixed value can be queried with <a href="#JuMP.fix_value"><code>fix_value</code></a>.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1363-L1380">source</a></section></article><h2 id="is_fixed"><a class="docs-heading-anchor" href="#is_fixed"><code>is_fixed</code></a><a id="is_fixed-1"></a><a class="docs-heading-anchor-permalink" href="#is_fixed" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_fixed" href="#JuMP.is_fixed"><code>JuMP.is_fixed</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_fixed(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is a fixed variable. If <code>true</code>, the fixed value can be queried with <a href="#JuMP.fix_value"><code>fix_value</code></a>.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -782,7 +782,7 @@
 julia&gt; fix(x, 1.0)
 
 julia&gt; is_fixed(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1017-L1040">source</a></section></article><h2 id="is_integer"><a class="docs-heading-anchor" href="#is_integer"><code>is_integer</code></a><a id="is_integer-1"></a><a class="docs-heading-anchor-permalink" href="#is_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_integer" href="#JuMP.is_integer"><code>JuMP.is_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_integer(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is constrained to be integer.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1017-L1040">source</a></section></article><h2 id="is_integer"><a class="docs-heading-anchor" href="#is_integer"><code>is_integer</code></a><a id="is_integer-1"></a><a class="docs-heading-anchor-permalink" href="#is_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_integer" href="#JuMP.is_integer"><code>JuMP.is_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_integer(v::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>v</code> is constrained to be integer.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -792,7 +792,7 @@
 julia&gt; set_integer(x)
 
 julia&gt; is_integer(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1226-L1248">source</a></section></article><h2 id="is_parameter"><a class="docs-heading-anchor" href="#is_parameter"><code>is_parameter</code></a><a id="is_parameter-1"></a><a class="docs-heading-anchor-permalink" href="#is_parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_parameter" href="#JuMP.is_parameter"><code>JuMP.is_parameter</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_parameter(x::GenericVariableRef)::Bool</code></pre><p>Return <code>true</code> if <code>x</code> is constrained to be a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1226-L1248">source</a></section></article><h2 id="is_parameter"><a class="docs-heading-anchor" href="#is_parameter"><code>is_parameter</code></a><a id="is_parameter-1"></a><a class="docs-heading-anchor-permalink" href="#is_parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_parameter" href="#JuMP.is_parameter"><code>JuMP.is_parameter</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_parameter(x::GenericVariableRef)::Bool</code></pre><p>Return <code>true</code> if <code>x</code> is constrained to be a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, p in Parameter(2))
 p
@@ -804,10 +804,10 @@
 x
 
 julia&gt; is_parameter(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1535-L1560">source</a></section></article><h2 id="is_valid"><a class="docs-heading-anchor" href="#is_valid"><code>is_valid</code></a><a id="is_valid-1"></a><a class="docs-heading-anchor-permalink" href="#is_valid" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_valid" href="#JuMP.is_valid"><code>JuMP.is_valid</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_valid(model::GenericModel, con_ref::ConstraintRef{&lt;:AbstractModel})</code></pre><p>Return <code>true</code> if <code>constraint_ref</code> refers to a valid constraint in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L435-L439">source</a></section><section><div><pre><code class="language-julia hljs">is_valid(model::GenericModel, variable_ref::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>variable</code> refers to a valid variable in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L466-L470">source</a></section><section><div><pre><code class="language-julia hljs">is_valid(model::Model, c::NonlinearConstraintRef)</code></pre><p>Return <code>true</code> if <code>c</code> refers to a valid nonlinear constraint in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L473-L477">source</a></section></article><h2 id="isequal_canonical"><a class="docs-heading-anchor" href="#isequal_canonical"><code>isequal_canonical</code></a><a id="isequal_canonical-1"></a><a class="docs-heading-anchor-permalink" href="#isequal_canonical" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.isequal_canonical" href="#JuMP.isequal_canonical"><code>JuMP.isequal_canonical</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">isequal_canonical(
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1535-L1560">source</a></section></article><h2 id="is_valid"><a class="docs-heading-anchor" href="#is_valid"><code>is_valid</code></a><a id="is_valid-1"></a><a class="docs-heading-anchor-permalink" href="#is_valid" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.is_valid" href="#JuMP.is_valid"><code>JuMP.is_valid</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_valid(model::GenericModel, con_ref::ConstraintRef{&lt;:AbstractModel})</code></pre><p>Return <code>true</code> if <code>constraint_ref</code> refers to a valid constraint in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L435-L439">source</a></section><section><div><pre><code class="language-julia hljs">is_valid(model::GenericModel, variable_ref::GenericVariableRef)</code></pre><p>Return <code>true</code> if <code>variable</code> refers to a valid variable in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L466-L470">source</a></section><section><div><pre><code class="language-julia hljs">is_valid(model::Model, c::NonlinearConstraintRef)</code></pre><p>Return <code>true</code> if <code>c</code> refers to a valid nonlinear constraint in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L473-L477">source</a></section></article><h2 id="isequal_canonical"><a class="docs-heading-anchor" href="#isequal_canonical"><code>isequal_canonical</code></a><a id="isequal_canonical-1"></a><a class="docs-heading-anchor-permalink" href="#isequal_canonical" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.isequal_canonical" href="#JuMP.isequal_canonical"><code>JuMP.isequal_canonical</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">isequal_canonical(
     aff::GenericAffExpr{C,V},
     other::GenericAffExpr{C,V}
-) where {C,V}</code></pre><p>Return <code>true</code> if <code>aff</code> is equal to <code>other</code> after dropping zeros and disregarding the order. Mainly useful for testing.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L540-L548">source</a></section></article><h2 id="jump_function"><a class="docs-heading-anchor" href="#jump_function"><code>jump_function</code></a><a id="jump_function-1"></a><a class="docs-heading-anchor-permalink" href="#jump_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.jump_function" href="#JuMP.jump_function"><code>JuMP.jump_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">jump_function(x)</code></pre><p>Given an MathOptInterface object <code>x</code>, return the JuMP equivalent.</p><p>See also: <a href="#JuMP.moi_function"><code>moi_function</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L682-L688">source</a></section></article><h2 id="jump_function_type"><a class="docs-heading-anchor" href="#jump_function_type"><code>jump_function_type</code></a><a id="jump_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#jump_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.jump_function_type" href="#JuMP.jump_function_type"><code>JuMP.jump_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">jump_function_type(::Type{T}) where {T}</code></pre><p>Given an MathOptInterface object type <code>T</code>, return the JuMP equivalent.</p><p>See also: <a href="#JuMP.moi_function_type"><code>moi_function_type</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L691-L697">source</a></section></article><h2 id="latex_formulation"><a class="docs-heading-anchor" href="#latex_formulation"><code>latex_formulation</code></a><a id="latex_formulation-1"></a><a class="docs-heading-anchor-permalink" href="#latex_formulation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.latex_formulation" href="#JuMP.latex_formulation"><code>JuMP.latex_formulation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">latex_formulation(model::AbstractModel)</code></pre><p>Wrap <code>model</code> in a type so that it can be pretty-printed as <code>text/latex</code> in a notebook like IJulia, or in Documenter.</p><p>To render the model, end the cell with <code>latex_formulation(model)</code>, or call <code>display(latex_formulation(model))</code> in to force the display of the model from inside a function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L17-L26">source</a></section></article><h2 id="linear_terms"><a class="docs-heading-anchor" href="#linear_terms"><code>linear_terms</code></a><a id="linear_terms-1"></a><a class="docs-heading-anchor-permalink" href="#linear_terms" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.linear_terms" href="#JuMP.linear_terms"><code>JuMP.linear_terms</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">linear_terms(aff::GenericAffExpr{C, V})</code></pre><p>Provides an iterator over coefficient-variable tuples <code>(a_i::C, x_i::V)</code> in the linear part of the affine expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L378-L383">source</a></section><section><div><pre><code class="language-julia hljs">linear_terms(quad::GenericQuadExpr{C, V})</code></pre><p>Provides an iterator over tuples <code>(coefficient::C, variable::V)</code> in the linear part of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L266-L271">source</a></section></article><h2 id="list_of_constraint_types"><a class="docs-heading-anchor" href="#list_of_constraint_types"><code>list_of_constraint_types</code></a><a id="list_of_constraint_types-1"></a><a class="docs-heading-anchor-permalink" href="#list_of_constraint_types" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.list_of_constraint_types" href="#JuMP.list_of_constraint_types"><code>JuMP.list_of_constraint_types</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">list_of_constraint_types(model::GenericModel)::Vector{Tuple{Type,Type}}</code></pre><p>Return a list of tuples of the form <code>(F, S)</code> where <code>F</code> is a JuMP function type and <code>S</code> is an MOI set type such that <code>all_constraints(model, F, S)</code> returns a nonempty list.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+) where {C,V}</code></pre><p>Return <code>true</code> if <code>aff</code> is equal to <code>other</code> after dropping zeros and disregarding the order. Mainly useful for testing.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L540-L548">source</a></section></article><h2 id="jump_function"><a class="docs-heading-anchor" href="#jump_function"><code>jump_function</code></a><a id="jump_function-1"></a><a class="docs-heading-anchor-permalink" href="#jump_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.jump_function" href="#JuMP.jump_function"><code>JuMP.jump_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">jump_function(x)</code></pre><p>Given an MathOptInterface object <code>x</code>, return the JuMP equivalent.</p><p>See also: <a href="#JuMP.moi_function"><code>moi_function</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L682-L688">source</a></section></article><h2 id="jump_function_type"><a class="docs-heading-anchor" href="#jump_function_type"><code>jump_function_type</code></a><a id="jump_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#jump_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.jump_function_type" href="#JuMP.jump_function_type"><code>JuMP.jump_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">jump_function_type(::Type{T}) where {T}</code></pre><p>Given an MathOptInterface object type <code>T</code>, return the JuMP equivalent.</p><p>See also: <a href="#JuMP.moi_function_type"><code>moi_function_type</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L691-L697">source</a></section></article><h2 id="latex_formulation"><a class="docs-heading-anchor" href="#latex_formulation"><code>latex_formulation</code></a><a id="latex_formulation-1"></a><a class="docs-heading-anchor-permalink" href="#latex_formulation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.latex_formulation" href="#JuMP.latex_formulation"><code>JuMP.latex_formulation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">latex_formulation(model::AbstractModel)</code></pre><p>Wrap <code>model</code> in a type so that it can be pretty-printed as <code>text/latex</code> in a notebook like IJulia, or in Documenter.</p><p>To render the model, end the cell with <code>latex_formulation(model)</code>, or call <code>display(latex_formulation(model))</code> in to force the display of the model from inside a function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L17-L26">source</a></section></article><h2 id="linear_terms"><a class="docs-heading-anchor" href="#linear_terms"><code>linear_terms</code></a><a id="linear_terms-1"></a><a class="docs-heading-anchor-permalink" href="#linear_terms" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.linear_terms" href="#JuMP.linear_terms"><code>JuMP.linear_terms</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">linear_terms(aff::GenericAffExpr{C, V})</code></pre><p>Provides an iterator over coefficient-variable tuples <code>(a_i::C, x_i::V)</code> in the linear part of the affine expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L378-L383">source</a></section><section><div><pre><code class="language-julia hljs">linear_terms(quad::GenericQuadExpr{C, V})</code></pre><p>Provides an iterator over tuples <code>(coefficient::C, variable::V)</code> in the linear part of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L266-L271">source</a></section></article><h2 id="list_of_constraint_types"><a class="docs-heading-anchor" href="#list_of_constraint_types"><code>list_of_constraint_types</code></a><a id="list_of_constraint_types-1"></a><a class="docs-heading-anchor-permalink" href="#list_of_constraint_types" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.list_of_constraint_types" href="#JuMP.list_of_constraint_types"><code>JuMP.list_of_constraint_types</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">list_of_constraint_types(model::GenericModel)::Vector{Tuple{Type,Type}}</code></pre><p>Return a list of tuples of the form <code>(F, S)</code> where <code>F</code> is a JuMP function type and <code>S</code> is an MOI set type such that <code>all_constraints(model, F, S)</code> returns a nonempty list.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 0, Bin);
 
@@ -817,12 +817,12 @@
 3-element Vector{Tuple{Type, Type}}:
  (AffExpr, MathOptInterface.LessThan{Float64})
  (VariableRef, MathOptInterface.GreaterThan{Float64})
- (VariableRef, MathOptInterface.ZeroOne)</code></pre><p><strong>Performance considerations</strong></p><p>Iterating over the list of function and set types is a type-unstable operation. Consider using a function barrier. See the <a href="../../developers/extensions/#Performance-tips-for-extensions">Performance tips for extensions</a> section of the documentation for more details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1301-L1329">source</a></section></article><h2 id="lower_bound"><a class="docs-heading-anchor" href="#lower_bound"><code>lower_bound</code></a><a id="lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.lower_bound" href="#JuMP.lower_bound"><code>JuMP.lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">lower_bound(v::GenericVariableRef)</code></pre><p>Return the lower bound of a variable. Error if one does not exist.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ (VariableRef, MathOptInterface.ZeroOne)</code></pre><p><strong>Performance considerations</strong></p><p>Iterating over the list of function and set types is a type-unstable operation. Consider using a function barrier. See the <a href="../../developers/extensions/#Performance-tips-for-extensions">Performance tips for extensions</a> section of the documentation for more details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1301-L1329">source</a></section></article><h2 id="lower_bound"><a class="docs-heading-anchor" href="#lower_bound"><code>lower_bound</code></a><a id="lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.lower_bound" href="#JuMP.lower_bound"><code>JuMP.lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">lower_bound(v::GenericVariableRef)</code></pre><p>Return the lower bound of a variable. Error if one does not exist.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 1.0);
 
 julia&gt; lower_bound(x)
-1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L816-L834">source</a></section></article><h2 id="lp_sensitivity_report"><a class="docs-heading-anchor" href="#lp_sensitivity_report"><code>lp_sensitivity_report</code></a><a id="lp_sensitivity_report-1"></a><a class="docs-heading-anchor-permalink" href="#lp_sensitivity_report" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.lp_sensitivity_report" href="#JuMP.lp_sensitivity_report"><code>JuMP.lp_sensitivity_report</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">lp_sensitivity_report(model::GenericModel{T}; atol::T = Base.rtoldefault(T))::SensitivityReport{T} where {T}</code></pre><p>Given a linear program <code>model</code> with a current optimal basis, return a <a href="#JuMP.SensitivityReport"><code>SensitivityReport</code></a> object, which maps:</p><ul><li>Every variable reference to a tuple <code>(d_lo, d_hi)::Tuple{T,T}</code>, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.</li><li>Every constraint reference to a tuple <code>(d_lo, d_hi)::Tuple{T,T}</code>, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.</li></ul><p>Both tuples are relative, rather than absolute. So given a objective coefficient of <code>1.0</code> and a tuple <code>(-0.5, 0.5)</code>, the objective coefficient can range between <code>1.0 - 0.5</code> an <code>1.0 + 0.5</code>.</p><p><code>atol</code> is the primal/dual optimality tolerance, and should match the tolerance of the solver used to compute the basis.</p><p>Note: interval constraints are NOT supported.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
+1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L816-L834">source</a></section></article><h2 id="lp_sensitivity_report"><a class="docs-heading-anchor" href="#lp_sensitivity_report"><code>lp_sensitivity_report</code></a><a id="lp_sensitivity_report-1"></a><a class="docs-heading-anchor-permalink" href="#lp_sensitivity_report" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.lp_sensitivity_report" href="#JuMP.lp_sensitivity_report"><code>JuMP.lp_sensitivity_report</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">lp_sensitivity_report(model::GenericModel{T}; atol::T = Base.rtoldefault(T))::SensitivityReport{T} where {T}</code></pre><p>Given a linear program <code>model</code> with a current optimal basis, return a <a href="#JuMP.SensitivityReport"><code>SensitivityReport</code></a> object, which maps:</p><ul><li>Every variable reference to a tuple <code>(d_lo, d_hi)::Tuple{T,T}</code>, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.</li><li>Every constraint reference to a tuple <code>(d_lo, d_hi)::Tuple{T,T}</code>, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.</li></ul><p>Both tuples are relative, rather than absolute. So given a objective coefficient of <code>1.0</code> and a tuple <code>(-0.5, 0.5)</code>, the objective coefficient can range between <code>1.0 - 0.5</code> an <code>1.0 + 0.5</code>.</p><p><code>atol</code> is the primal/dual optimality tolerance, and should match the tolerance of the solver used to compute the basis.</p><p>Note: interval constraints are NOT supported.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
 
 julia&gt; model = Model(HiGHS.Optimizer);
 
@@ -854,7 +854,7 @@
            &quot;The lower bound of `x` can decrease by $dRHS_lo or increase &quot; *
            &quot;by $dRHS_hi.&quot;
        )
-The lower bound of `x` can decrease by -Inf or increase by 3.0.</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/lp_sensitivity2.jl#L19-L78">source</a></section></article><h2 id="map_coefficients"><a class="docs-heading-anchor" href="#map_coefficients"><code>map_coefficients</code></a><a id="map_coefficients-1"></a><a class="docs-heading-anchor-permalink" href="#map_coefficients" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.map_coefficients" href="#JuMP.map_coefficients"><code>JuMP.map_coefficients</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">map_coefficients(f::Function, a::GenericAffExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> <code>a</code> and return a new expression.</p><p>See also: <a href="#JuMP.map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+The lower bound of `x` can decrease by -Inf or increase by 3.0.</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/lp_sensitivity2.jl#L19-L78">source</a></section></article><h2 id="map_coefficients"><a class="docs-heading-anchor" href="#map_coefficients"><code>map_coefficients</code></a><a id="map_coefficients-1"></a><a class="docs-heading-anchor-permalink" href="#map_coefficients" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.map_coefficients" href="#JuMP.map_coefficients"><code>JuMP.map_coefficients</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">map_coefficients(f::Function, a::GenericAffExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> <code>a</code> and return a new expression.</p><p>See also: <a href="#JuMP.map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -865,7 +865,7 @@
 2 x + 2
 
 julia&gt; a
-x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L305-L329">source</a></section><section><div><pre><code class="language-julia hljs">map_coefficients(f::Function, a::GenericQuadExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> <code>a</code> and return a new expression.</p><p>See also: <a href="#JuMP.map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L305-L329">source</a></section><section><div><pre><code class="language-julia hljs">map_coefficients(f::Function, a::GenericQuadExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> <code>a</code> and return a new expression.</p><p>See also: <a href="#JuMP.map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -876,7 +876,7 @@
 2 x² + 2 x + 2
 
 julia&gt; a
-x² + x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L211-L235">source</a></section></article><h2 id="map_coefficients_inplace!"><a class="docs-heading-anchor" href="#map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a><a id="map_coefficients_inplace!-1"></a><a class="docs-heading-anchor-permalink" href="#map_coefficients_inplace!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.map_coefficients_inplace!" href="#JuMP.map_coefficients_inplace!"><code>JuMP.map_coefficients_inplace!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">map_coefficients_inplace!(f::Function, a::GenericAffExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> <code>a</code> and update them in-place.</p><p>See also: <a href="#JuMP.map_coefficients"><code>map_coefficients</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x² + x + 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L211-L235">source</a></section></article><h2 id="map_coefficients_inplace!"><a class="docs-heading-anchor" href="#map_coefficients_inplace!"><code>map_coefficients_inplace!</code></a><a id="map_coefficients_inplace!-1"></a><a class="docs-heading-anchor-permalink" href="#map_coefficients_inplace!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.map_coefficients_inplace!" href="#JuMP.map_coefficients_inplace!"><code>JuMP.map_coefficients_inplace!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">map_coefficients_inplace!(f::Function, a::GenericAffExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> <code>a</code> and update them in-place.</p><p>See also: <a href="#JuMP.map_coefficients"><code>map_coefficients</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -887,7 +887,7 @@
 2 x + 2
 
 julia&gt; a
-2 x + 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L271-L295">source</a></section><section><div><pre><code class="language-julia hljs">map_coefficients_inplace!(f::Function, a::GenericQuadExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> <code>a</code> and update them in-place.</p><p>See also: <a href="#JuMP.map_coefficients"><code>map_coefficients</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+2 x + 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L271-L295">source</a></section><section><div><pre><code class="language-julia hljs">map_coefficients_inplace!(f::Function, a::GenericQuadExpr)</code></pre><p>Apply <code>f</code> to the coefficients and constant term of an <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> <code>a</code> and update them in-place.</p><p>See also: <a href="#JuMP.map_coefficients"><code>map_coefficients</code></a></p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -898,7 +898,7 @@
 2 x² + 2 x + 2
 
 julia&gt; a
-2 x² + 2 x + 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L177-L201">source</a></section></article><h2 id="mode"><a class="docs-heading-anchor" href="#mode"><code>mode</code></a><a id="mode-1"></a><a class="docs-heading-anchor-permalink" href="#mode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.mode" href="#JuMP.mode"><code>JuMP.mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">mode(model::GenericModel)</code></pre><p>Return the <a href="#JuMP.ModelMode"><code>ModelMode</code></a> (<a href="#JuMP.DIRECT"><code>DIRECT</code></a>, <a href="#JuMP.AUTOMATIC"><code>AUTOMATIC</code></a>, or <a href="#JuMP.MANUAL"><code>MANUAL</code></a>) of <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L528-L533">source</a></section></article><h2 id="model_convert"><a class="docs-heading-anchor" href="#model_convert"><code>model_convert</code></a><a id="model_convert-1"></a><a class="docs-heading-anchor-permalink" href="#model_convert" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.model_convert" href="#JuMP.model_convert"><code>JuMP.model_convert</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">model_convert(
+2 x² + 2 x + 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L177-L201">source</a></section></article><h2 id="mode"><a class="docs-heading-anchor" href="#mode"><code>mode</code></a><a id="mode-1"></a><a class="docs-heading-anchor-permalink" href="#mode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.mode" href="#JuMP.mode"><code>JuMP.mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">mode(model::GenericModel)</code></pre><p>Return the <a href="#JuMP.ModelMode"><code>ModelMode</code></a> (<a href="#JuMP.DIRECT"><code>DIRECT</code></a>, <a href="#JuMP.AUTOMATIC"><code>AUTOMATIC</code></a>, or <a href="#JuMP.MANUAL"><code>MANUAL</code></a>) of <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L528-L533">source</a></section></article><h2 id="model_convert"><a class="docs-heading-anchor" href="#model_convert"><code>model_convert</code></a><a id="model_convert-1"></a><a class="docs-heading-anchor-permalink" href="#model_convert" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.model_convert" href="#JuMP.model_convert"><code>JuMP.model_convert</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">model_convert(
     model::AbstractModel,
     rhs::Union{
         AbstractConstraint,
@@ -906,18 +906,18 @@
         AbstractJuMPScalar,
         MOI.AbstractSet,
     },
-)</code></pre><p>Convert the coefficients and constants of functions and sets in the <code>rhs</code> to the coefficient type <code>value_type(typeof(model))</code>.</p><p><strong>Purpose</strong></p><p>Creating and adding a constraint is a two-step process. The first step calls <a href="#JuMP.build_constraint"><code>build_constraint</code></a>, and the result of that is passed to <a href="#JuMP.add_constraint"><code>add_constraint</code></a>.</p><p>However, because <a href="#JuMP.build_constraint"><code>build_constraint</code></a> does not take the <code>model</code> as an argument, the coefficients and constants of the function or set might be different than <code>value_type(typeof(model))</code>.</p><p>Therefore, the result of <a href="#JuMP.build_constraint"><code>build_constraint</code></a> is converted in a call to <code>model_convert</code> before the result is passed to <a href="#JuMP.add_constraint"><code>add_constraint</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L1126-L1152">source</a></section></article><h2 id="model_string"><a class="docs-heading-anchor" href="#model_string"><code>model_string</code></a><a id="model_string-1"></a><a class="docs-heading-anchor-permalink" href="#model_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.model_string" href="#JuMP.model_string"><code>JuMP.model_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">model_string(mode::MIME, model::AbstractModel)</code></pre><p>Return a <code>String</code> representation of <code>model</code> given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L355-L359">source</a></section></article><h2 id="moi_function"><a class="docs-heading-anchor" href="#moi_function"><code>moi_function</code></a><a id="moi_function-1"></a><a class="docs-heading-anchor-permalink" href="#moi_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_function" href="#JuMP.moi_function"><code>JuMP.moi_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_function(x)</code></pre><p>Given a JuMP object <code>x</code>, return the MathOptInterface equivalent.</p><p>See also: <a href="#JuMP.jump_function"><code>jump_function</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L643-L649">source</a></section></article><h2 id="moi_function_type"><a class="docs-heading-anchor" href="#moi_function_type"><code>moi_function_type</code></a><a id="moi_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#moi_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_function_type" href="#JuMP.moi_function_type"><code>JuMP.moi_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_function_type(::Type{T}) where {T}</code></pre><p>Given a JuMP object type <code>T</code>, return the MathOptInterface equivalent.</p><p>See also: <a href="#JuMP.jump_function_type"><code>jump_function_type</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L673-L679">source</a></section></article><h2 id="moi_set"><a class="docs-heading-anchor" href="#moi_set"><code>moi_set</code></a><a id="moi_set-1"></a><a class="docs-heading-anchor-permalink" href="#moi_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_set" href="#JuMP.moi_set"><code>JuMP.moi_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_set(constraint::AbstractConstraint)</code></pre><p>Return the set of the constraint <code>constraint</code> in the function-in-set form as a <code>MathOptInterface.AbstractSet</code>.</p><pre><code class="nohighlight hljs">moi_set(s::AbstractVectorSet, dim::Int)</code></pre><p>Returns the MOI set of dimension <code>dim</code> corresponding to the JuMP set <code>s</code>.</p><pre><code class="nohighlight hljs">moi_set(s::AbstractScalarSet)</code></pre><p>Returns the MOI set corresponding to the JuMP set <code>s</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L554-L567">source</a></section></article><h2 id="name"><a class="docs-heading-anchor" href="#name"><code>name</code></a><a id="name-1"></a><a class="docs-heading-anchor-permalink" href="#name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.name" href="#JuMP.name"><code>JuMP.name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">name(con_ref::ConstraintRef)</code></pre><p>Get a constraint&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L234-L238">source</a></section><section><div><pre><code class="language-julia hljs">name(v::GenericVariableRef)::String</code></pre><p>Get a variable&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L529-L533">source</a></section><section><div><pre><code class="language-julia hljs">name(model::AbstractModel)</code></pre><p>Return the <a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name</code></a> attribute of <code>model</code>&#39;s <a href="#JuMP.backend"><code>backend</code></a>, or a default if empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L178-L183">source</a></section></article><h2 id="node_count"><a class="docs-heading-anchor" href="#node_count"><code>node_count</code></a><a id="node_count-1"></a><a class="docs-heading-anchor-permalink" href="#node_count" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.node_count" href="#JuMP.node_count"><code>JuMP.node_count</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">node_count(model::GenericModel)</code></pre><p>Gets the total number of branch-and-bound nodes explored during the most recent optimization in a Mixed Integer Program.</p><p>Solvers must implement <code>MOI.NodeCount()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L612-L619">source</a></section></article><h2 id="nonlinear_constraint_string"><a class="docs-heading-anchor" href="#nonlinear_constraint_string"><code>nonlinear_constraint_string</code></a><a id="nonlinear_constraint_string-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_constraint_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_constraint_string" href="#JuMP.nonlinear_constraint_string"><code>JuMP.nonlinear_constraint_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_constraint_string(
+)</code></pre><p>Convert the coefficients and constants of functions and sets in the <code>rhs</code> to the coefficient type <code>value_type(typeof(model))</code>.</p><p><strong>Purpose</strong></p><p>Creating and adding a constraint is a two-step process. The first step calls <a href="#JuMP.build_constraint"><code>build_constraint</code></a>, and the result of that is passed to <a href="#JuMP.add_constraint"><code>add_constraint</code></a>.</p><p>However, because <a href="#JuMP.build_constraint"><code>build_constraint</code></a> does not take the <code>model</code> as an argument, the coefficients and constants of the function or set might be different than <code>value_type(typeof(model))</code>.</p><p>Therefore, the result of <a href="#JuMP.build_constraint"><code>build_constraint</code></a> is converted in a call to <code>model_convert</code> before the result is passed to <a href="#JuMP.add_constraint"><code>add_constraint</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L1126-L1152">source</a></section></article><h2 id="model_string"><a class="docs-heading-anchor" href="#model_string"><code>model_string</code></a><a id="model_string-1"></a><a class="docs-heading-anchor-permalink" href="#model_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.model_string" href="#JuMP.model_string"><code>JuMP.model_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">model_string(mode::MIME, model::AbstractModel)</code></pre><p>Return a <code>String</code> representation of <code>model</code> given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L355-L359">source</a></section></article><h2 id="moi_function"><a class="docs-heading-anchor" href="#moi_function"><code>moi_function</code></a><a id="moi_function-1"></a><a class="docs-heading-anchor-permalink" href="#moi_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_function" href="#JuMP.moi_function"><code>JuMP.moi_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_function(x)</code></pre><p>Given a JuMP object <code>x</code>, return the MathOptInterface equivalent.</p><p>See also: <a href="#JuMP.jump_function"><code>jump_function</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L643-L649">source</a></section></article><h2 id="moi_function_type"><a class="docs-heading-anchor" href="#moi_function_type"><code>moi_function_type</code></a><a id="moi_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#moi_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_function_type" href="#JuMP.moi_function_type"><code>JuMP.moi_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_function_type(::Type{T}) where {T}</code></pre><p>Given a JuMP object type <code>T</code>, return the MathOptInterface equivalent.</p><p>See also: <a href="#JuMP.jump_function_type"><code>jump_function_type</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L673-L679">source</a></section></article><h2 id="moi_set"><a class="docs-heading-anchor" href="#moi_set"><code>moi_set</code></a><a id="moi_set-1"></a><a class="docs-heading-anchor-permalink" href="#moi_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.moi_set" href="#JuMP.moi_set"><code>JuMP.moi_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">moi_set(constraint::AbstractConstraint)</code></pre><p>Return the set of the constraint <code>constraint</code> in the function-in-set form as a <code>MathOptInterface.AbstractSet</code>.</p><pre><code class="nohighlight hljs">moi_set(s::AbstractVectorSet, dim::Int)</code></pre><p>Returns the MOI set of dimension <code>dim</code> corresponding to the JuMP set <code>s</code>.</p><pre><code class="nohighlight hljs">moi_set(s::AbstractScalarSet)</code></pre><p>Returns the MOI set corresponding to the JuMP set <code>s</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L554-L567">source</a></section></article><h2 id="name"><a class="docs-heading-anchor" href="#name"><code>name</code></a><a id="name-1"></a><a class="docs-heading-anchor-permalink" href="#name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.name" href="#JuMP.name"><code>JuMP.name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">name(con_ref::ConstraintRef)</code></pre><p>Get a constraint&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L234-L238">source</a></section><section><div><pre><code class="language-julia hljs">name(v::GenericVariableRef)::String</code></pre><p>Get a variable&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L529-L533">source</a></section><section><div><pre><code class="language-julia hljs">name(model::AbstractModel)</code></pre><p>Return the <a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name</code></a> attribute of <code>model</code>&#39;s <a href="#JuMP.backend"><code>backend</code></a>, or a default if empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L178-L183">source</a></section></article><h2 id="node_count"><a class="docs-heading-anchor" href="#node_count"><code>node_count</code></a><a id="node_count-1"></a><a class="docs-heading-anchor-permalink" href="#node_count" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.node_count" href="#JuMP.node_count"><code>JuMP.node_count</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">node_count(model::GenericModel)</code></pre><p>Gets the total number of branch-and-bound nodes explored during the most recent optimization in a Mixed Integer Program.</p><p>Solvers must implement <code>MOI.NodeCount()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L612-L619">source</a></section></article><h2 id="nonlinear_constraint_string"><a class="docs-heading-anchor" href="#nonlinear_constraint_string"><code>nonlinear_constraint_string</code></a><a id="nonlinear_constraint_string-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_constraint_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_constraint_string" href="#JuMP.nonlinear_constraint_string"><code>JuMP.nonlinear_constraint_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_constraint_string(
     model::GenericModel,
     mode::MIME,
     c::_NonlinearConstraint,
-)</code></pre><p>Return a string representation of the nonlinear constraint <code>c</code> belonging to <code>model</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L387-L396">source</a></section></article><h2 id="nonlinear_dual_start_value"><a class="docs-heading-anchor" href="#nonlinear_dual_start_value"><code>nonlinear_dual_start_value</code></a><a id="nonlinear_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_dual_start_value" href="#JuMP.nonlinear_dual_start_value"><code>JuMP.nonlinear_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_dual_start_value(model::Model)</code></pre><p>Return the current value of the MOI attribute <a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart"><code>MOI.NLPBlockDualStart</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L581-L585">source</a></section></article><h2 id="nonlinear_expr_string"><a class="docs-heading-anchor" href="#nonlinear_expr_string"><code>nonlinear_expr_string</code></a><a id="nonlinear_expr_string-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_expr_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_expr_string" href="#JuMP.nonlinear_expr_string"><code>JuMP.nonlinear_expr_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_expr_string(
+)</code></pre><p>Return a string representation of the nonlinear constraint <code>c</code> belonging to <code>model</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L387-L396">source</a></section></article><h2 id="nonlinear_dual_start_value"><a class="docs-heading-anchor" href="#nonlinear_dual_start_value"><code>nonlinear_dual_start_value</code></a><a id="nonlinear_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_dual_start_value" href="#JuMP.nonlinear_dual_start_value"><code>JuMP.nonlinear_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_dual_start_value(model::Model)</code></pre><p>Return the current value of the MOI attribute <a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart"><code>MOI.NLPBlockDualStart</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L581-L585">source</a></section></article><h2 id="nonlinear_expr_string"><a class="docs-heading-anchor" href="#nonlinear_expr_string"><code>nonlinear_expr_string</code></a><a id="nonlinear_expr_string-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_expr_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_expr_string" href="#JuMP.nonlinear_expr_string"><code>JuMP.nonlinear_expr_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_expr_string(
     model::GenericModel,
     mode::MIME,
     c::MOI.Nonlinear.Expression,
-)</code></pre><p>Return a string representation of the nonlinear expression <code>c</code> belonging to <code>model</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L431-L440">source</a></section></article><h2 id="nonlinear_model"><a class="docs-heading-anchor" href="#nonlinear_model"><code>nonlinear_model</code></a><a id="nonlinear_model-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_model" href="#JuMP.nonlinear_model"><code>JuMP.nonlinear_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_model(
+)</code></pre><p>Return a string representation of the nonlinear expression <code>c</code> belonging to <code>model</code>, given the <code>mode</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L431-L440">source</a></section></article><h2 id="nonlinear_model"><a class="docs-heading-anchor" href="#nonlinear_model"><code>nonlinear_model</code></a><a id="nonlinear_model-1"></a><a class="docs-heading-anchor-permalink" href="#nonlinear_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.nonlinear_model" href="#JuMP.nonlinear_model"><code>JuMP.nonlinear_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">nonlinear_model(
     model::GenericModel;
     force::Bool = false,
-)::Union{MOI.Nonlinear.Model,Nothing}</code></pre><p>If <code>model</code> has nonlinear components, return a <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model"><code>MOI.Nonlinear.Model</code></a>, otherwise return <code>nothing</code>.</p><p>If <code>force</code>, always return a <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model"><code>MOI.Nonlinear.Model</code></a>, and if one does not exist for the model, create an empty one.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L6-L17">source</a></section></article><h2 id="normalized_coefficient"><a class="docs-heading-anchor" href="#normalized_coefficient"><code>normalized_coefficient</code></a><a id="normalized_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#normalized_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.normalized_coefficient" href="#JuMP.normalized_coefficient"><code>JuMP.normalized_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef)</code></pre><p>Return the coefficient associated with <code>variable</code> in <code>constraint</code> after JuMP has normalized the constraint into its standard form. See also <a href="#JuMP.set_normalized_coefficient"><code>set_normalized_coefficient</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L813-L819">source</a></section></article><h2 id="normalized_rhs"><a class="docs-heading-anchor" href="#normalized_rhs"><code>normalized_rhs</code></a><a id="normalized_rhs-1"></a><a class="docs-heading-anchor-permalink" href="#normalized_rhs" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.normalized_rhs" href="#JuMP.normalized_rhs"><code>JuMP.normalized_rhs</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">normalized_rhs(con_ref::ConstraintRef)</code></pre><p>Return the right-hand side term of <code>con_ref</code> after JuMP has converted the constraint into its normalized form. See also <a href="#JuMP.set_normalized_rhs"><code>set_normalized_rhs</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L873-L878">source</a></section></article><h2 id="num_constraints"><a class="docs-heading-anchor" href="#num_constraints"><code>num_constraints</code></a><a id="num_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#num_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_constraints" href="#JuMP.num_constraints"><code>JuMP.num_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_constraints(model::GenericModel, function_type, set_type)::Int64</code></pre><p>Return the number of constraints currently in the model where the function has type <code>function_type</code> and the set has type <code>set_type</code>.</p><p>See also <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and <a href="#JuMP.all_constraints"><code>all_constraints</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+)::Union{MOI.Nonlinear.Model,Nothing}</code></pre><p>If <code>model</code> has nonlinear components, return a <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model"><code>MOI.Nonlinear.Model</code></a>, otherwise return <code>nothing</code>.</p><p>If <code>force</code>, always return a <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model"><code>MOI.Nonlinear.Model</code></a>, and if one does not exist for the model, create an empty one.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L6-L17">source</a></section></article><h2 id="normalized_coefficient"><a class="docs-heading-anchor" href="#normalized_coefficient"><code>normalized_coefficient</code></a><a id="normalized_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#normalized_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.normalized_coefficient" href="#JuMP.normalized_coefficient"><code>JuMP.normalized_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef)</code></pre><p>Return the coefficient associated with <code>variable</code> in <code>constraint</code> after JuMP has normalized the constraint into its standard form. See also <a href="#JuMP.set_normalized_coefficient"><code>set_normalized_coefficient</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L813-L819">source</a></section></article><h2 id="normalized_rhs"><a class="docs-heading-anchor" href="#normalized_rhs"><code>normalized_rhs</code></a><a id="normalized_rhs-1"></a><a class="docs-heading-anchor-permalink" href="#normalized_rhs" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.normalized_rhs" href="#JuMP.normalized_rhs"><code>JuMP.normalized_rhs</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">normalized_rhs(con_ref::ConstraintRef)</code></pre><p>Return the right-hand side term of <code>con_ref</code> after JuMP has converted the constraint into its normalized form. See also <a href="#JuMP.set_normalized_rhs"><code>set_normalized_rhs</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L873-L878">source</a></section></article><h2 id="num_constraints"><a class="docs-heading-anchor" href="#num_constraints"><code>num_constraints</code></a><a id="num_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#num_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_constraints" href="#JuMP.num_constraints"><code>JuMP.num_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_constraints(model::GenericModel, function_type, set_type)::Int64</code></pre><p>Return the number of constraints currently in the model where the function has type <code>function_type</code> and the set has type <code>set_type</code>.</p><p>See also <a href="#JuMP.list_of_constraint_types"><code>list_of_constraint_types</code></a> and <a href="#JuMP.all_constraints"><code>all_constraints</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 0, Bin);
 
@@ -936,7 +936,7 @@
 1
 
 julia&gt; num_constraints(model, AffExpr, MOI.LessThan{Float64})
-2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1192-L1224">source</a></section><section><div><pre><code class="language-julia hljs">num_constraints(model::GenericModel; count_variable_in_set_constraints::Bool)</code></pre><p>Return the number of constraints in <code>model</code>.</p><p>If <code>count_variable_in_set_constraints == true</code>, then <code>VariableRef</code> constraints such as <code>VariableRef</code>-in-<code>Integer</code> are included. To count only the number of structural constraints (e.g., the rows in the constraint matrix of a linear program), pass <code>count_variable_in_set_constraints = false</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1192-L1224">source</a></section><section><div><pre><code class="language-julia hljs">num_constraints(model::GenericModel; count_variable_in_set_constraints::Bool)</code></pre><p>Return the number of constraints in <code>model</code>.</p><p>If <code>count_variable_in_set_constraints == true</code>, then <code>VariableRef</code> constraints such as <code>VariableRef</code>-in-<code>Integer</code> are included. To count only the number of structural constraints (e.g., the rows in the constraint matrix of a linear program), pass <code>count_variable_in_set_constraints = false</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 0, Int);
 
@@ -946,7 +946,7 @@
 3
 
 julia&gt; num_constraints(model; count_variable_in_set_constraints = false)
-1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1339-L1364">source</a></section></article><h2 id="num_nonlinear_constraints"><a class="docs-heading-anchor" href="#num_nonlinear_constraints"><code>num_nonlinear_constraints</code></a><a id="num_nonlinear_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#num_nonlinear_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_nonlinear_constraints" href="#JuMP.num_nonlinear_constraints"><code>JuMP.num_nonlinear_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_nonlinear_constraints(model::GenericModel)</code></pre><p>Returns the number of nonlinear constraints associated with the <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L499-L503">source</a></section></article><h2 id="num_variables"><a class="docs-heading-anchor" href="#num_variables"><code>num_variables</code></a><a id="num_variables-1"></a><a class="docs-heading-anchor-permalink" href="#num_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_variables" href="#JuMP.num_variables"><code>JuMP.num_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_variables(model::GenericModel)::Int64</code></pre><p>Returns number of variables in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L6-L10">source</a></section></article><h2 id="object_dictionary"><a class="docs-heading-anchor" href="#object_dictionary"><code>object_dictionary</code></a><a id="object_dictionary-1"></a><a class="docs-heading-anchor-permalink" href="#object_dictionary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.object_dictionary" href="#JuMP.object_dictionary"><code>JuMP.object_dictionary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">object_dictionary(model::GenericModel)</code></pre><p>Return the dictionary that maps the symbol name of a variable, constraint, or expression to the corresponding object.</p><p>Objects are registered to a specific symbol in the macros. For example, <code>@variable(model, x[1:2, 1:2])</code> registers the array of variables <code>x</code> to the symbol <code>:x</code>.</p><p>This method should be defined for any subtype of <code>AbstractModel</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L843-L854">source</a></section></article><h2 id="objective_bound"><a class="docs-heading-anchor" href="#objective_bound"><code>objective_bound</code></a><a id="objective_bound-1"></a><a class="docs-heading-anchor-permalink" href="#objective_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_bound" href="#JuMP.objective_bound"><code>JuMP.objective_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_bound(model::GenericModel)</code></pre><p>Return the best known bound on the optimal objective value after a call to <code>optimize!(model)</code>.</p><p>For scalar-valued objectives, this function returns a <code>Float64</code>. For vector-valued objectives, it returns a <code>Vector{Float64}</code>.</p><p>In the case of a vector-valued objective, this returns the <em>ideal point</em>, that is, the point obtained if each objective was optimized independently.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L23-L34">source</a></section></article><h2 id="objective_function"><a class="docs-heading-anchor" href="#objective_function"><code>objective_function</code></a><a id="objective_function-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function" href="#JuMP.objective_function"><code>JuMP.objective_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function(
+1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1339-L1364">source</a></section></article><h2 id="num_nonlinear_constraints"><a class="docs-heading-anchor" href="#num_nonlinear_constraints"><code>num_nonlinear_constraints</code></a><a id="num_nonlinear_constraints-1"></a><a class="docs-heading-anchor-permalink" href="#num_nonlinear_constraints" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_nonlinear_constraints" href="#JuMP.num_nonlinear_constraints"><code>JuMP.num_nonlinear_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_nonlinear_constraints(model::GenericModel)</code></pre><p>Returns the number of nonlinear constraints associated with the <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L499-L503">source</a></section></article><h2 id="num_variables"><a class="docs-heading-anchor" href="#num_variables"><code>num_variables</code></a><a id="num_variables-1"></a><a class="docs-heading-anchor-permalink" href="#num_variables" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.num_variables" href="#JuMP.num_variables"><code>JuMP.num_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">num_variables(model::GenericModel)::Int64</code></pre><p>Returns number of variables in <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L6-L10">source</a></section></article><h2 id="object_dictionary"><a class="docs-heading-anchor" href="#object_dictionary"><code>object_dictionary</code></a><a id="object_dictionary-1"></a><a class="docs-heading-anchor-permalink" href="#object_dictionary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.object_dictionary" href="#JuMP.object_dictionary"><code>JuMP.object_dictionary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">object_dictionary(model::GenericModel)</code></pre><p>Return the dictionary that maps the symbol name of a variable, constraint, or expression to the corresponding object.</p><p>Objects are registered to a specific symbol in the macros. For example, <code>@variable(model, x[1:2, 1:2])</code> registers the array of variables <code>x</code> to the symbol <code>:x</code>.</p><p>This method should be defined for any subtype of <code>AbstractModel</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L843-L854">source</a></section></article><h2 id="objective_bound"><a class="docs-heading-anchor" href="#objective_bound"><code>objective_bound</code></a><a id="objective_bound-1"></a><a class="docs-heading-anchor-permalink" href="#objective_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_bound" href="#JuMP.objective_bound"><code>JuMP.objective_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_bound(model::GenericModel)</code></pre><p>Return the best known bound on the optimal objective value after a call to <code>optimize!(model)</code>.</p><p>For scalar-valued objectives, this function returns a <code>Float64</code>. For vector-valued objectives, it returns a <code>Vector{Float64}</code>.</p><p>In the case of a vector-valued objective, this returns the <em>ideal point</em>, that is, the point obtained if each objective was optimized independently.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L23-L34">source</a></section></article><h2 id="objective_function"><a class="docs-heading-anchor" href="#objective_function"><code>objective_function</code></a><a id="objective_function-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function" href="#JuMP.objective_function"><code>JuMP.objective_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function(
     model::GenericModel,
     T::Type = objective_function_type(model),
 )</code></pre><p>Return an object of type <code>T</code> representing the objective function.</p><p>Error if the objective is not convertible to type <code>T</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
@@ -966,7 +966,7 @@
 julia&gt; typeof(objective_function(model, QuadExpr))
 QuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})</code></pre><p>We see with the last two commands that even if the objective function is affine, as it is convertible to a quadratic function, it can be queried as a quadratic function and the result is quadratic.</p><p>However, it is not convertible to a variable.</p><pre><code class="language-julia-repl hljs">julia&gt; objective_function(model, VariableRef)
 ERROR: InexactError: convert(MathOptInterface.VariableIndex, 1.0 + 2.0 MOI.VariableIndex(1))
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L191-L231">source</a></section></article><h2 id="objective_function_string"><a class="docs-heading-anchor" href="#objective_function_string"><code>objective_function_string</code></a><a id="objective_function_string-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function_string" href="#JuMP.objective_function_string"><code>JuMP.objective_function_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function_string(mode, model::AbstractModel)::String</code></pre><p>Return a <code>String</code> describing the objective function of the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L367-L371">source</a></section></article><h2 id="objective_function_type"><a class="docs-heading-anchor" href="#objective_function_type"><code>objective_function_type</code></a><a id="objective_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function_type" href="#JuMP.objective_function_type"><code>JuMP.objective_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function_type(model::GenericModel)::AbstractJuMPScalar</code></pre><p>Return the type of the objective function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L179-L183">source</a></section></article><h2 id="objective_sense"><a class="docs-heading-anchor" href="#objective_sense"><code>objective_sense</code></a><a id="objective_sense-1"></a><a class="docs-heading-anchor-permalink" href="#objective_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_sense" href="#JuMP.objective_sense"><code>JuMP.objective_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_sense(model::GenericModel)::MOI.OptimizationSense</code></pre><p>Return the objective sense.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L75-L79">source</a></section></article><h2 id="objective_value"><a class="docs-heading-anchor" href="#objective_value"><code>objective_value</code></a><a id="objective_value-1"></a><a class="docs-heading-anchor-permalink" href="#objective_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_value" href="#JuMP.objective_value"><code>JuMP.objective_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_value(model::GenericModel; result::Int = 1)</code></pre><p>Return the objective value associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>For scalar-valued objectives, this function returns a <code>Float64</code>. For vector-valued objectives, it returns a <code>Vector{Float64}</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L39-L49">source</a></section></article><h2 id="op_ifelse"><a class="docs-heading-anchor" href="#op_ifelse"><code>op_ifelse</code></a><a id="op_ifelse-1"></a><a class="docs-heading-anchor-permalink" href="#op_ifelse" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_ifelse" href="#JuMP.op_ifelse"><code>JuMP.op_ifelse</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">op_ifelse(a, x, y)</code></pre><p>A function that falls back to <code>ifelse(a, x, y)</code>, but when called with a JuMP variables or expression in the first argument, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L191-L231">source</a></section></article><h2 id="objective_function_string"><a class="docs-heading-anchor" href="#objective_function_string"><code>objective_function_string</code></a><a id="objective_function_string-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function_string" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function_string" href="#JuMP.objective_function_string"><code>JuMP.objective_function_string</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function_string(mode, model::AbstractModel)::String</code></pre><p>Return a <code>String</code> describing the objective function of the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L367-L371">source</a></section></article><h2 id="objective_function_type"><a class="docs-heading-anchor" href="#objective_function_type"><code>objective_function_type</code></a><a id="objective_function_type-1"></a><a class="docs-heading-anchor-permalink" href="#objective_function_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_function_type" href="#JuMP.objective_function_type"><code>JuMP.objective_function_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_function_type(model::GenericModel)::AbstractJuMPScalar</code></pre><p>Return the type of the objective function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L179-L183">source</a></section></article><h2 id="objective_sense"><a class="docs-heading-anchor" href="#objective_sense"><code>objective_sense</code></a><a id="objective_sense-1"></a><a class="docs-heading-anchor-permalink" href="#objective_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_sense" href="#JuMP.objective_sense"><code>JuMP.objective_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_sense(model::GenericModel)::MOI.OptimizationSense</code></pre><p>Return the objective sense.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L75-L79">source</a></section></article><h2 id="objective_value"><a class="docs-heading-anchor" href="#objective_value"><code>objective_value</code></a><a id="objective_value-1"></a><a class="docs-heading-anchor-permalink" href="#objective_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.objective_value" href="#JuMP.objective_value"><code>JuMP.objective_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">objective_value(model::GenericModel; result::Int = 1)</code></pre><p>Return the objective value associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>For scalar-valued objectives, this function returns a <code>Float64</code>. For vector-valued objectives, it returns a <code>Vector{Float64}</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L39-L49">source</a></section></article><h2 id="op_ifelse"><a class="docs-heading-anchor" href="#op_ifelse"><code>op_ifelse</code></a><a id="op_ifelse-1"></a><a class="docs-heading-anchor-permalink" href="#op_ifelse" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_ifelse" href="#JuMP.op_ifelse"><code>JuMP.op_ifelse</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">op_ifelse(a, x, y)</code></pre><p>A function that falls back to <code>ifelse(a, x, y)</code>, but when called with a JuMP variables or expression in the first argument, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -977,7 +977,7 @@
 ifelse(x, 1.0, 2.0)
 
 julia&gt; op_ifelse(true, x, 2.0)
-x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L484-L507">source</a></section></article><h2 id="operator_to_set"><a class="docs-heading-anchor" href="#operator_to_set"><code>operator_to_set</code></a><a id="operator_to_set-1"></a><a class="docs-heading-anchor-permalink" href="#operator_to_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.operator_to_set" href="#JuMP.operator_to_set"><code>JuMP.operator_to_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">operator_to_set(_error::Function, ::Val{sense_symbol})</code></pre><p>Converts a sense symbol to a set <code>set</code> such that <code>@constraint(model, func sense_symbol 0)</code> is equivalent to <code>@constraint(model, func in set)</code> for any <code>func::AbstractJuMPScalar</code>.</p><p><strong>Example</strong></p><p>Once a custom set is defined you can directly create a JuMP constraint with it:</p><pre><code class="language-julia-repl hljs">julia&gt; struct CustomSet{T} &lt;: MOI.AbstractScalarSet
+x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L484-L507">source</a></section></article><h2 id="operator_to_set"><a class="docs-heading-anchor" href="#operator_to_set"><code>operator_to_set</code></a><a id="operator_to_set-1"></a><a class="docs-heading-anchor-permalink" href="#operator_to_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.operator_to_set" href="#JuMP.operator_to_set"><code>JuMP.operator_to_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">operator_to_set(_error::Function, ::Val{sense_symbol})</code></pre><p>Converts a sense symbol to a set <code>set</code> such that <code>@constraint(model, func sense_symbol 0)</code> is equivalent to <code>@constraint(model, func in set)</code> for any <code>func::AbstractJuMPScalar</code>.</p><p><strong>Example</strong></p><p>Once a custom set is defined you can directly create a JuMP constraint with it:</p><pre><code class="language-julia-repl hljs">julia&gt; struct CustomSet{T} &lt;: MOI.AbstractScalarSet
            value::T
        end
 
@@ -996,15 +996,15 @@
 julia&gt; MOIU.shift_constant(set::CustomSet, value) = CustomSet(set.value + value)
 
 julia&gt; cref = @constraint(model, x ⊰ 1)
-x ∈ CustomSet{Float64}(1.0)</code></pre><p>Note that the whole function is first moved to the right-hand side, then the sign is transformed into a set with zero constant and finally the constant is moved to the set with <code>MOIU.shift_constant</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L163-L204">source</a></section></article><h2 id="operator_warn"><a class="docs-heading-anchor" href="#operator_warn"><code>operator_warn</code></a><a id="operator_warn-1"></a><a class="docs-heading-anchor-permalink" href="#operator_warn" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.operator_warn" href="#JuMP.operator_warn"><code>JuMP.operator_warn</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">operator_warn(model::AbstractModel)
-operator_warn(model::GenericModel)</code></pre><p>This function is called on the model whenever two affine expressions are added together without using <code>destructive_add!</code>, and at least one of the two expressions has more than 50 terms.</p><p>For the case of <code>Model</code>, if this function is called more than 20,000 times then a warning is generated once.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/operators.jl#L263-L273">source</a></section></article><h2 id="optimize!"><a class="docs-heading-anchor" href="#optimize!"><code>optimize!</code></a><a id="optimize!-1"></a><a class="docs-heading-anchor-permalink" href="#optimize!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimize!" href="#JuMP.optimize!"><code>JuMP.optimize!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimize!(
+x ∈ CustomSet{Float64}(1.0)</code></pre><p>Note that the whole function is first moved to the right-hand side, then the sign is transformed into a set with zero constant and finally the constant is moved to the set with <code>MOIU.shift_constant</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L163-L204">source</a></section></article><h2 id="operator_warn"><a class="docs-heading-anchor" href="#operator_warn"><code>operator_warn</code></a><a id="operator_warn-1"></a><a class="docs-heading-anchor-permalink" href="#operator_warn" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.operator_warn" href="#JuMP.operator_warn"><code>JuMP.operator_warn</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">operator_warn(model::AbstractModel)
+operator_warn(model::GenericModel)</code></pre><p>This function is called on the model whenever two affine expressions are added together without using <code>destructive_add!</code>, and at least one of the two expressions has more than 50 terms.</p><p>For the case of <code>Model</code>, if this function is called more than 20,000 times then a warning is generated once.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/operators.jl#L263-L273">source</a></section></article><h2 id="optimize!"><a class="docs-heading-anchor" href="#optimize!"><code>optimize!</code></a><a id="optimize!-1"></a><a class="docs-heading-anchor-permalink" href="#optimize!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimize!" href="#JuMP.optimize!"><code>JuMP.optimize!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimize!(
     model::GenericModel;
     ignore_optimize_hook = (model.optimize_hook === nothing),
     _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =
         MOI.Nonlinear.SparseReverseMode(),
     kwargs...,
-)</code></pre><p>Optimize the model.</p><p>If an optimizer has not been set yet (see <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>), a <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> error is thrown.</p><p>If <code>ignore_optimize_hook == true</code>, the optimize hook is ignored and the model is solved as if the hook was not set. Keyword arguments <code>kwargs</code> are passed to the <code>optimize_hook</code>. An error is thrown if <code>optimize_hook</code> is <code>nothing</code> and keyword arguments are provided.</p><p><strong>Experimental features</strong></p><p>These features may change or be removed in any future version of JuMP.</p><p>Pass <code>_differentiation_backend</code> to set the <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation"><code>MOI.Nonlinear.AbstractAutomaticDifferentiation</code></a> backend used to compute derivatives of nonlinear programs.</p><p>If you require only <code>:ExprGraph</code>, it is more efficient to pass <code>_differentiation_backend = MOI.Nonlinear.ExprGraphOnly()</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L379-L408">source</a></section></article><h2 id="optimizer_index"><a class="docs-heading-anchor" href="#optimizer_index"><code>optimizer_index</code></a><a id="optimizer_index-1"></a><a class="docs-heading-anchor-permalink" href="#optimizer_index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimizer_index" href="#JuMP.optimizer_index"><code>JuMP.optimizer_index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimizer_index(x::GenericVariableRef)::MOI.VariableIndex
-optimizer_index(x::ConstraintRef{&lt;:GenericModel})::MOI.ConstraintIndex</code></pre><p>Return the index that corresponds to <code>x</code> in the optimizer model.</p><p>Throws <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> if no optimizer is set, and throws an <code>ErrorException</code> if the optimizer is set but is not attached.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L1063-L1071">source</a></section></article><h2 id="optimizer_with_attributes"><a class="docs-heading-anchor" href="#optimizer_with_attributes"><code>optimizer_with_attributes</code></a><a id="optimizer_with_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#optimizer_with_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimizer_with_attributes" href="#JuMP.optimizer_with_attributes"><code>JuMP.optimizer_with_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimizer_with_attributes(optimizer_constructor, attrs::Pair...)</code></pre><p>Groups an optimizer constructor with the list of attributes <code>attrs</code>. Note that it is equivalent to <code>MOI.OptimizerWithAttributes</code>.</p><p>When provided to the <code>Model</code> constructor or to <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>, it creates an optimizer by calling <code>optimizer_constructor()</code>, and then sets the attributes using <a href="#JuMP.set_attribute"><code>set_attribute</code></a>.</p><p>See also: <a href="#JuMP.set_attribute"><code>set_attribute</code></a>, <a href="#JuMP.get_attribute"><code>get_attribute</code></a>.</p><p><strong>Note</strong></p><p>The string names of the attributes are specific to each solver. One should consult the solver&#39;s documentation to find the attributes of interest.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
+)</code></pre><p>Optimize the model.</p><p>If an optimizer has not been set yet (see <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>), a <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> error is thrown.</p><p>If <code>ignore_optimize_hook == true</code>, the optimize hook is ignored and the model is solved as if the hook was not set. Keyword arguments <code>kwargs</code> are passed to the <code>optimize_hook</code>. An error is thrown if <code>optimize_hook</code> is <code>nothing</code> and keyword arguments are provided.</p><p><strong>Experimental features</strong></p><p>These features may change or be removed in any future version of JuMP.</p><p>Pass <code>_differentiation_backend</code> to set the <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation"><code>MOI.Nonlinear.AbstractAutomaticDifferentiation</code></a> backend used to compute derivatives of nonlinear programs.</p><p>If you require only <code>:ExprGraph</code>, it is more efficient to pass <code>_differentiation_backend = MOI.Nonlinear.ExprGraphOnly()</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L379-L408">source</a></section></article><h2 id="optimizer_index"><a class="docs-heading-anchor" href="#optimizer_index"><code>optimizer_index</code></a><a id="optimizer_index-1"></a><a class="docs-heading-anchor-permalink" href="#optimizer_index" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimizer_index" href="#JuMP.optimizer_index"><code>JuMP.optimizer_index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimizer_index(x::GenericVariableRef)::MOI.VariableIndex
+optimizer_index(x::ConstraintRef{&lt;:GenericModel})::MOI.ConstraintIndex</code></pre><p>Return the index that corresponds to <code>x</code> in the optimizer model.</p><p>Throws <a href="#JuMP.NoOptimizer"><code>NoOptimizer</code></a> if no optimizer is set, and throws an <code>ErrorException</code> if the optimizer is set but is not attached.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L1063-L1071">source</a></section></article><h2 id="optimizer_with_attributes"><a class="docs-heading-anchor" href="#optimizer_with_attributes"><code>optimizer_with_attributes</code></a><a id="optimizer_with_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#optimizer_with_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.optimizer_with_attributes" href="#JuMP.optimizer_with_attributes"><code>JuMP.optimizer_with_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">optimizer_with_attributes(optimizer_constructor, attrs::Pair...)</code></pre><p>Groups an optimizer constructor with the list of attributes <code>attrs</code>. Note that it is equivalent to <code>MOI.OptimizerWithAttributes</code>.</p><p>When provided to the <code>Model</code> constructor or to <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>, it creates an optimizer by calling <code>optimizer_constructor()</code>, and then sets the attributes using <a href="#JuMP.set_attribute"><code>set_attribute</code></a>.</p><p>See also: <a href="#JuMP.set_attribute"><code>set_attribute</code></a>, <a href="#JuMP.get_attribute"><code>get_attribute</code></a>.</p><p><strong>Note</strong></p><p>The string names of the attributes are specific to each solver. One should consult the solver&#39;s documentation to find the attributes of interest.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
 
 julia&gt; optimizer = optimizer_with_attributes(
            HiGHS.Optimizer, &quot;presolve&quot; =&gt; &quot;off&quot;, MOI.Silent() =&gt; true,
@@ -1016,7 +1016,7 @@
 
 julia&gt; set_attribute(model, &quot;presolve&quot;, &quot;off&quot;)
 
-julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L6-L46">source</a></section></article><h2 id="owner_model"><a class="docs-heading-anchor" href="#owner_model"><code>owner_model</code></a><a id="owner_model-1"></a><a class="docs-heading-anchor-permalink" href="#owner_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.owner_model" href="#JuMP.owner_model"><code>JuMP.owner_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">owner_model(s::AbstractJuMPScalar)</code></pre><p>Return the model owning the scalar <code>s</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1042-L1046">source</a></section></article><h2 id="parameter_value"><a class="docs-heading-anchor" href="#parameter_value"><code>parameter_value</code></a><a id="parameter_value-1"></a><a class="docs-heading-anchor-permalink" href="#parameter_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parameter_value" href="#JuMP.parameter_value"><code>JuMP.parameter_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parameter_value(x::GenericVariableRef)</code></pre><p>Return the value of the parameter <code>x</code>.</p><p>Errors if <code>x</code> is not a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+julia&gt; set_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L6-L46">source</a></section></article><h2 id="owner_model"><a class="docs-heading-anchor" href="#owner_model"><code>owner_model</code></a><a id="owner_model-1"></a><a class="docs-heading-anchor-permalink" href="#owner_model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.owner_model" href="#JuMP.owner_model"><code>JuMP.owner_model</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">owner_model(s::AbstractJuMPScalar)</code></pre><p>Return the model owning the scalar <code>s</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1042-L1046">source</a></section></article><h2 id="parameter_value"><a class="docs-heading-anchor" href="#parameter_value"><code>parameter_value</code></a><a id="parameter_value-1"></a><a class="docs-heading-anchor-permalink" href="#parameter_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parameter_value" href="#JuMP.parameter_value"><code>JuMP.parameter_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parameter_value(x::GenericVariableRef)</code></pre><p>Return the value of the parameter <code>x</code>.</p><p>Errors if <code>x</code> is not a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, p in Parameter(2))
 p
@@ -1027,18 +1027,18 @@
 julia&gt; set_parameter_value(p, 2.5)
 
 julia&gt; parameter_value(p)
-2.5</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1601-L1627">source</a></section></article><h2 id="parse_constraint"><a class="docs-heading-anchor" href="#parse_constraint"><code>parse_constraint</code></a><a id="parse_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint" href="#JuMP.parse_constraint"><code>JuMP.parse_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint(_error::Function, expr::Expr)</code></pre><p>The entry-point for all constraint-related parsing.</p><p><strong>Arguments</strong></p><ul><li>The <code>_error</code> function is passed everywhere to provide better error messages</li><li><code>expr</code> comes from the <code>@constraint</code> macro. There are two possibilities:<ul><li><code>@constraint(model, expr)</code></li><li><code>@constraint(model, name[args], expr)</code></li></ul>In both cases, <code>expr</code> is the main component of the constraint.</li></ul><p><strong>Supported syntax</strong></p><p>JuMP currently supports the following <code>expr</code> objects:</p><ul><li><code>lhs &lt;= rhs</code></li><li><code>lhs == rhs</code></li><li><code>lhs &gt;= rhs</code></li><li><code>l &lt;= body &lt;= u</code></li><li><code>u &gt;= body &gt;= l</code></li><li><code>lhs ⟂ rhs</code></li><li><code>lhs in rhs</code></li><li><code>lhs ∈ rhs</code></li><li><code>z =&gt; {constraint}</code></li><li><code>!z =&gt; {constraint}</code></li></ul><p>as well as all broadcasted variants.</p><p><strong>Extensions</strong></p><p>The infrastructure behind <code>parse_constraint</code> is extendable. See <a href="#JuMP.parse_constraint_head"><code>parse_constraint_head</code></a> and <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a> for details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L386-L418">source</a></section></article><h2 id="parse_constraint_call"><a class="docs-heading-anchor" href="#parse_constraint_call"><code>parse_constraint_call</code></a><a id="parse_constraint_call-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint_call" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint_call" href="#JuMP.parse_constraint_call"><code>JuMP.parse_constraint_call</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint_call(
+2.5</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1601-L1627">source</a></section></article><h2 id="parse_constraint"><a class="docs-heading-anchor" href="#parse_constraint"><code>parse_constraint</code></a><a id="parse_constraint-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint" href="#JuMP.parse_constraint"><code>JuMP.parse_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint(_error::Function, expr::Expr)</code></pre><p>The entry-point for all constraint-related parsing.</p><p><strong>Arguments</strong></p><ul><li>The <code>_error</code> function is passed everywhere to provide better error messages</li><li><code>expr</code> comes from the <code>@constraint</code> macro. There are two possibilities:<ul><li><code>@constraint(model, expr)</code></li><li><code>@constraint(model, name[args], expr)</code></li></ul>In both cases, <code>expr</code> is the main component of the constraint.</li></ul><p><strong>Supported syntax</strong></p><p>JuMP currently supports the following <code>expr</code> objects:</p><ul><li><code>lhs &lt;= rhs</code></li><li><code>lhs == rhs</code></li><li><code>lhs &gt;= rhs</code></li><li><code>l &lt;= body &lt;= u</code></li><li><code>u &gt;= body &gt;= l</code></li><li><code>lhs ⟂ rhs</code></li><li><code>lhs in rhs</code></li><li><code>lhs ∈ rhs</code></li><li><code>z =&gt; {constraint}</code></li><li><code>!z =&gt; {constraint}</code></li></ul><p>as well as all broadcasted variants.</p><p><strong>Extensions</strong></p><p>The infrastructure behind <code>parse_constraint</code> is extendable. See <a href="#JuMP.parse_constraint_head"><code>parse_constraint_head</code></a> and <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a> for details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L386-L418">source</a></section></article><h2 id="parse_constraint_call"><a class="docs-heading-anchor" href="#parse_constraint_call"><code>parse_constraint_call</code></a><a id="parse_constraint_call-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint_call" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint_call" href="#JuMP.parse_constraint_call"><code>JuMP.parse_constraint_call</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint_call(
     _error::Function,
     is_vectorized::Bool,
     ::Val{op},
     args...,
-)</code></pre><p>Implement this method to intercept the parsing of a <code>:call</code> expression with operator <code>op</code>.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Extending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the <a href="https://gitter.im/JuliaOpt/jump-dev">developer chatroom</a> before publishing any code that implements these methods.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that accepts a <code>String</code> and throws the string as an error, along with some descriptive information of the macro from which it was thrown.</li><li><code>is_vectorized</code>: a boolean to indicate if <code>op</code> should be broadcast or not</li><li><code>op</code>: the first element of the <code>.args</code> field of the <code>Expr</code> to intercept</li><li><code>args...</code>: the <code>.args</code> field of the <code>Expr</code>.</li></ul><p><strong>Returns</strong></p><p>This function must return:</p><ul><li><code>parse_code::Expr</code>: an expression containing any setup or rewriting code that needs to be called before <code>build_constraint</code></li><li><code>build_code::Expr</code>: an expression that calls <code>build_constraint(</code> or <code>build_constraint.(</code> depending on <code>is_vectorized</code>.</li></ul><p>See also: <a href="#JuMP.parse_constraint_head"><code>parse_constraint_head</code></a>, <a href="#JuMP.build_constraint"><code>build_constraint</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L773-L809">source</a></section><section><div><pre><code class="language-julia hljs">parse_constraint_call(
+)</code></pre><p>Implement this method to intercept the parsing of a <code>:call</code> expression with operator <code>op</code>.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Extending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the <a href="https://gitter.im/JuliaOpt/jump-dev">developer chatroom</a> before publishing any code that implements these methods.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that accepts a <code>String</code> and throws the string as an error, along with some descriptive information of the macro from which it was thrown.</li><li><code>is_vectorized</code>: a boolean to indicate if <code>op</code> should be broadcast or not</li><li><code>op</code>: the first element of the <code>.args</code> field of the <code>Expr</code> to intercept</li><li><code>args...</code>: the <code>.args</code> field of the <code>Expr</code>.</li></ul><p><strong>Returns</strong></p><p>This function must return:</p><ul><li><code>parse_code::Expr</code>: an expression containing any setup or rewriting code that needs to be called before <code>build_constraint</code></li><li><code>build_code::Expr</code>: an expression that calls <code>build_constraint(</code> or <code>build_constraint.(</code> depending on <code>is_vectorized</code>.</li></ul><p>See also: <a href="#JuMP.parse_constraint_head"><code>parse_constraint_head</code></a>, <a href="#JuMP.build_constraint"><code>build_constraint</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L773-L809">source</a></section><section><div><pre><code class="language-julia hljs">parse_constraint_call(
     _error::Function,
     vectorized::Bool,
     ::Val{op},
     lhs,
     rhs,
-) where {op}</code></pre><p>Fallback handler for binary operators. These might be infix operators like <code>@constraint(model, lhs op rhs)</code>, or normal operators like <code>@constraint(model, op(lhs, rhs))</code>.</p><p>In both cases, we rewrite as <code>lhs - rhs in operator_to_set(_error, op)</code>.</p><p>See <a href="#JuMP.operator_to_set"><code>operator_to_set</code></a> for details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L842-L858">source</a></section></article><h2 id="parse_constraint_head"><a class="docs-heading-anchor" href="#parse_constraint_head"><code>parse_constraint_head</code></a><a id="parse_constraint_head-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint_head" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint_head" href="#JuMP.parse_constraint_head"><code>JuMP.parse_constraint_head</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint_head(_error::Function, ::Val{head}, args...)</code></pre><p>Implement this method to intercept the parsing of an expression with head <code>head</code>.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Extending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the <a href="https://gitter.im/JuliaOpt/jump-dev">developer chatroom</a> before publishing any code that implements these methods.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that accepts a <code>String</code> and throws the string as an error, along with some descriptive information of the macro from which it was thrown.</li><li><code>head</code>: the <code>.head</code> field of the <code>Expr</code> to intercept</li><li><code>args...</code>: the <code>.args</code> field of the <code>Expr</code>.</li></ul><p><strong>Returns</strong></p><p>This function must return:</p><ul><li><code>is_vectorized::Bool</code>: whether the expression represents a broadcasted expression like <code>x .&lt;= 1</code></li><li><code>parse_code::Expr</code>: an expression containing any setup or rewriting code that needs to be called before <code>build_constraint</code></li><li><code>build_code::Expr</code>: an expression that calls <code>build_constraint(</code> or <code>build_constraint.(</code> depending on <code>is_vectorized</code>.</li></ul><p><strong>Existing implementations</strong></p><p>JuMP currently implements:</p><ul><li><code>::Val{:call}</code>, which forwards calls to <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a></li><li><code>::Val{:comparison}</code>, which handles the special case of <code>l &lt;= body &lt;= u</code>.</li></ul><p>See also: <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a>, <a href="#JuMP.build_constraint"><code>build_constraint</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L423-L462">source</a></section></article><h2 id="parse_one_operator_variable"><a class="docs-heading-anchor" href="#parse_one_operator_variable"><code>parse_one_operator_variable</code></a><a id="parse_one_operator_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_one_operator_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_one_operator_variable" href="#JuMP.parse_one_operator_variable"><code>JuMP.parse_one_operator_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_one_operator_variable(_error::Function, infoexpr::_VariableInfoExpr, sense::Val{S}, value) where S</code></pre><p>Update <code>infoexr</code> for a variable expression in the <code>@variable</code> macro of the form <code>variable name S value</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L2276-L2280">source</a></section></article><h2 id="parse_ternary_variable"><a class="docs-heading-anchor" href="#parse_ternary_variable"><code>parse_ternary_variable</code></a><a id="parse_ternary_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_ternary_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_ternary_variable" href="#JuMP.parse_ternary_variable"><code>JuMP.parse_ternary_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_ternary_variable(_error, variable_info, lhs_sense, lhs, rhs_sense, rhs)</code></pre><p>A hook for JuMP extensiosn to intercept the parsing of a <code>:comparison</code> expression, which has the form <code>lhs lhs_sense variable rhs_sense rhs</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L2405-L2410">source</a></section></article><h2 id="parse_variable"><a class="docs-heading-anchor" href="#parse_variable"><code>parse_variable</code></a><a id="parse_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_variable" href="#JuMP.parse_variable"><code>JuMP.parse_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_variable(_error::Function, ::_VariableInfoExpr, args...)</code></pre><p>A hook for extensions to intercept the parsing of inequality constraints in the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L2262-L2267">source</a></section></article><h2 id="primal_feasibility_report"><a class="docs-heading-anchor" href="#primal_feasibility_report"><code>primal_feasibility_report</code></a><a id="primal_feasibility_report-1"></a><a class="docs-heading-anchor-permalink" href="#primal_feasibility_report" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.primal_feasibility_report" href="#JuMP.primal_feasibility_report"><code>JuMP.primal_feasibility_report</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">primal_feasibility_report(
+) where {op}</code></pre><p>Fallback handler for binary operators. These might be infix operators like <code>@constraint(model, lhs op rhs)</code>, or normal operators like <code>@constraint(model, op(lhs, rhs))</code>.</p><p>In both cases, we rewrite as <code>lhs - rhs in operator_to_set(_error, op)</code>.</p><p>See <a href="#JuMP.operator_to_set"><code>operator_to_set</code></a> for details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L842-L858">source</a></section></article><h2 id="parse_constraint_head"><a class="docs-heading-anchor" href="#parse_constraint_head"><code>parse_constraint_head</code></a><a id="parse_constraint_head-1"></a><a class="docs-heading-anchor-permalink" href="#parse_constraint_head" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_constraint_head" href="#JuMP.parse_constraint_head"><code>JuMP.parse_constraint_head</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_constraint_head(_error::Function, ::Val{head}, args...)</code></pre><p>Implement this method to intercept the parsing of an expression with head <code>head</code>.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Extending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the <a href="https://gitter.im/JuliaOpt/jump-dev">developer chatroom</a> before publishing any code that implements these methods.</p></div></div><p><strong>Arguments</strong></p><ul><li><code>_error</code>: a function that accepts a <code>String</code> and throws the string as an error, along with some descriptive information of the macro from which it was thrown.</li><li><code>head</code>: the <code>.head</code> field of the <code>Expr</code> to intercept</li><li><code>args...</code>: the <code>.args</code> field of the <code>Expr</code>.</li></ul><p><strong>Returns</strong></p><p>This function must return:</p><ul><li><code>is_vectorized::Bool</code>: whether the expression represents a broadcasted expression like <code>x .&lt;= 1</code></li><li><code>parse_code::Expr</code>: an expression containing any setup or rewriting code that needs to be called before <code>build_constraint</code></li><li><code>build_code::Expr</code>: an expression that calls <code>build_constraint(</code> or <code>build_constraint.(</code> depending on <code>is_vectorized</code>.</li></ul><p><strong>Existing implementations</strong></p><p>JuMP currently implements:</p><ul><li><code>::Val{:call}</code>, which forwards calls to <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a></li><li><code>::Val{:comparison}</code>, which handles the special case of <code>l &lt;= body &lt;= u</code>.</li></ul><p>See also: <a href="#JuMP.parse_constraint_call"><code>parse_constraint_call</code></a>, <a href="#JuMP.build_constraint"><code>build_constraint</code></a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L423-L462">source</a></section></article><h2 id="parse_one_operator_variable"><a class="docs-heading-anchor" href="#parse_one_operator_variable"><code>parse_one_operator_variable</code></a><a id="parse_one_operator_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_one_operator_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_one_operator_variable" href="#JuMP.parse_one_operator_variable"><code>JuMP.parse_one_operator_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_one_operator_variable(_error::Function, infoexpr::_VariableInfoExpr, sense::Val{S}, value) where S</code></pre><p>Update <code>infoexr</code> for a variable expression in the <code>@variable</code> macro of the form <code>variable name S value</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L2276-L2280">source</a></section></article><h2 id="parse_ternary_variable"><a class="docs-heading-anchor" href="#parse_ternary_variable"><code>parse_ternary_variable</code></a><a id="parse_ternary_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_ternary_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_ternary_variable" href="#JuMP.parse_ternary_variable"><code>JuMP.parse_ternary_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_ternary_variable(_error, variable_info, lhs_sense, lhs, rhs_sense, rhs)</code></pre><p>A hook for JuMP extensiosn to intercept the parsing of a <code>:comparison</code> expression, which has the form <code>lhs lhs_sense variable rhs_sense rhs</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L2405-L2410">source</a></section></article><h2 id="parse_variable"><a class="docs-heading-anchor" href="#parse_variable"><code>parse_variable</code></a><a id="parse_variable-1"></a><a class="docs-heading-anchor-permalink" href="#parse_variable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.parse_variable" href="#JuMP.parse_variable"><code>JuMP.parse_variable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">parse_variable(_error::Function, ::_VariableInfoExpr, args...)</code></pre><p>A hook for extensions to intercept the parsing of inequality constraints in the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L2262-L2267">source</a></section></article><h2 id="primal_feasibility_report"><a class="docs-heading-anchor" href="#primal_feasibility_report"><code>primal_feasibility_report</code></a><a id="primal_feasibility_report-1"></a><a class="docs-heading-anchor-permalink" href="#primal_feasibility_report" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.primal_feasibility_report" href="#JuMP.primal_feasibility_report"><code>JuMP.primal_feasibility_report</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">primal_feasibility_report(
     model::GenericModel{T},
     point::AbstractDict{GenericVariableRef{T},T} = _last_primal_solution(model),
     atol::T = zero(T),
@@ -1049,7 +1049,7 @@
 
 julia&gt; primal_feasibility_report(model, Dict(x =&gt; 0.2))
 Dict{Any, Float64} with 1 entry:
-  x ≥ 0.5 =&gt; 0.3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/feasibility_checker.jl#L16-L51">source</a></section><section><div><pre><code class="language-julia hljs">primal_feasibility_report(
+  x ≥ 0.5 =&gt; 0.3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/feasibility_checker.jl#L16-L51">source</a></section><section><div><pre><code class="language-julia hljs">primal_feasibility_report(
     point::Function,
     model::GenericModel{T};
     atol::T = zero(T),
@@ -1062,20 +1062,20 @@
            return start_value(v)
        end
 Dict{Any, Float64} with 1 entry:
-  x ≤ 1 =&gt; 0.3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/feasibility_checker.jl#L74-L98">source</a></section></article><h2 id="primal_status"><a class="docs-heading-anchor" href="#primal_status"><code>primal_status</code></a><a id="primal_status-1"></a><a class="docs-heading-anchor-permalink" href="#primal_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.primal_status" href="#JuMP.primal_status"><code>JuMP.primal_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">primal_status(model::GenericModel; result::Int = 1)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.ResultStatusCode"><code>MOI.ResultStatusCode</code></a> describing the status of the most recent primal solution of the solver (i.e., the <a href="../../moi/reference/models/#MathOptInterface.PrimalStatus"><code>MOI.PrimalStatus</code></a> attribute) associated with the result index <code>result</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L550-L558">source</a></section></article><h2 id="print_active_bridges"><a class="docs-heading-anchor" href="#print_active_bridges"><code>print_active_bridges</code></a><a id="print_active_bridges-1"></a><a class="docs-heading-anchor-permalink" href="#print_active_bridges" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.print_active_bridges" href="#JuMP.print_active_bridges"><code>JuMP.print_active_bridges</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">print_active_bridges([io::IO = stdout,] model::GenericModel)</code></pre><p>Print a list of the variable, constraint, and objective bridges that are currently used in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L733-L738">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges([io::IO = stdout,] model::GenericModel, ::Type{F}) where {F}</code></pre><p>Print a list of bridges required for an objective function of type <code>F</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L745-L749">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges(
+  x ≤ 1 =&gt; 0.3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/feasibility_checker.jl#L74-L98">source</a></section></article><h2 id="primal_status"><a class="docs-heading-anchor" href="#primal_status"><code>primal_status</code></a><a id="primal_status-1"></a><a class="docs-heading-anchor-permalink" href="#primal_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.primal_status" href="#JuMP.primal_status"><code>JuMP.primal_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">primal_status(model::GenericModel; result::Int = 1)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.ResultStatusCode"><code>MOI.ResultStatusCode</code></a> describing the status of the most recent primal solution of the solver (i.e., the <a href="../../moi/reference/models/#MathOptInterface.PrimalStatus"><code>MOI.PrimalStatus</code></a> attribute) associated with the result index <code>result</code>.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L550-L558">source</a></section></article><h2 id="print_active_bridges"><a class="docs-heading-anchor" href="#print_active_bridges"><code>print_active_bridges</code></a><a id="print_active_bridges-1"></a><a class="docs-heading-anchor-permalink" href="#print_active_bridges" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.print_active_bridges" href="#JuMP.print_active_bridges"><code>JuMP.print_active_bridges</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">print_active_bridges([io::IO = stdout,] model::GenericModel)</code></pre><p>Print a list of the variable, constraint, and objective bridges that are currently used in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L733-L738">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges([io::IO = stdout,] model::GenericModel, ::Type{F}) where {F}</code></pre><p>Print a list of bridges required for an objective function of type <code>F</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L745-L749">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges(
     [io::IO = stdout,]
     model::GenericModel,
     F::Type,
     S::Type{&lt;:MOI.AbstractSet},
-)</code></pre><p>Print a list of bridges required for a constraint of type <code>F</code>-in-<code>S</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L756-L765">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges(
+)</code></pre><p>Print a list of bridges required for a constraint of type <code>F</code>-in-<code>S</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L756-L765">source</a></section><section><div><pre><code class="language-julia hljs">print_active_bridges(
     [io::IO = stdout,]
     model::GenericModel,
     S::Type{&lt;:MOI.AbstractSet},
-)</code></pre><p>Print a list of bridges required to add a variable constrained to the set <code>S</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L777-L785">source</a></section></article><h2 id="print_bridge_graph"><a class="docs-heading-anchor" href="#print_bridge_graph"><code>print_bridge_graph</code></a><a id="print_bridge_graph-1"></a><a class="docs-heading-anchor-permalink" href="#print_bridge_graph" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.print_bridge_graph" href="#JuMP.print_bridge_graph"><code>JuMP.print_bridge_graph</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs"> print_bridge_graph([io::IO,] model::GenericModel)</code></pre><p>Print the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>This function is intended for advanced users. If you want to see only the bridges that are currently used, use <a href="#JuMP.print_active_bridges"><code>print_active_bridges</code></a> instead.</p></div></div><p><strong>Explanation of output</strong></p><p>Each node in the hyper-graph corresponds to a variable, constraint, or objective type.</p><ul><li>Variable nodes are indicated by <code>[ ]</code></li><li>Constraint nodes are indicated by <code>( )</code></li><li>Objective nodes are indicated by <code>| |</code></li></ul><p>The number inside each pair of brackets is an index of the node in the hyper-graph.</p><p>Note that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.</p><p>For more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  &quot;MathOptInterface: a data structure for mathematical optimization problems.&quot; URL: <a href="https://arxiv.org/abs/2002.03447">https://arxiv.org/abs/2002.03447</a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L696-L724">source</a></section></article><h2 id="quad_terms"><a class="docs-heading-anchor" href="#quad_terms"><code>quad_terms</code></a><a id="quad_terms-1"></a><a class="docs-heading-anchor-permalink" href="#quad_terms" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.quad_terms" href="#JuMP.quad_terms"><code>JuMP.quad_terms</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">quad_terms(quad::GenericQuadExpr{C, V})</code></pre><p>Provides an iterator over tuples <code>(coefficient::C, var_1::V, var_2::V)</code> in the quadratic part of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L284-L289">source</a></section></article><h2 id="raw_status"><a class="docs-heading-anchor" href="#raw_status"><code>raw_status</code></a><a id="raw_status-1"></a><a class="docs-heading-anchor-permalink" href="#raw_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.raw_status" href="#JuMP.raw_status"><code>JuMP.raw_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">raw_status(model::GenericModel)</code></pre><p>Return the reason why the solver stopped in its own words (i.e., the MathOptInterface model attribute <code>RawStatusString</code>).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L527-L532">source</a></section></article><h2 id="read_from_file"><a class="docs-heading-anchor" href="#read_from_file"><code>read_from_file</code></a><a id="read_from_file-1"></a><a class="docs-heading-anchor-permalink" href="#read_from_file" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.read_from_file" href="#JuMP.read_from_file"><code>JuMP.read_from_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">read_from_file(
+)</code></pre><p>Print a list of bridges required to add a variable constrained to the set <code>S</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L777-L785">source</a></section></article><h2 id="print_bridge_graph"><a class="docs-heading-anchor" href="#print_bridge_graph"><code>print_bridge_graph</code></a><a id="print_bridge_graph-1"></a><a class="docs-heading-anchor-permalink" href="#print_bridge_graph" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.print_bridge_graph" href="#JuMP.print_bridge_graph"><code>JuMP.print_bridge_graph</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs"> print_bridge_graph([io::IO,] model::GenericModel)</code></pre><p>Print the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>This function is intended for advanced users. If you want to see only the bridges that are currently used, use <a href="#JuMP.print_active_bridges"><code>print_active_bridges</code></a> instead.</p></div></div><p><strong>Explanation of output</strong></p><p>Each node in the hyper-graph corresponds to a variable, constraint, or objective type.</p><ul><li>Variable nodes are indicated by <code>[ ]</code></li><li>Constraint nodes are indicated by <code>( )</code></li><li>Objective nodes are indicated by <code>| |</code></li></ul><p>The number inside each pair of brackets is an index of the node in the hyper-graph.</p><p>Note that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.</p><p>For more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  &quot;MathOptInterface: a data structure for mathematical optimization problems.&quot; URL: <a href="https://arxiv.org/abs/2002.03447">https://arxiv.org/abs/2002.03447</a></p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L696-L724">source</a></section></article><h2 id="quad_terms"><a class="docs-heading-anchor" href="#quad_terms"><code>quad_terms</code></a><a id="quad_terms-1"></a><a class="docs-heading-anchor-permalink" href="#quad_terms" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.quad_terms" href="#JuMP.quad_terms"><code>JuMP.quad_terms</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">quad_terms(quad::GenericQuadExpr{C, V})</code></pre><p>Provides an iterator over tuples <code>(coefficient::C, var_1::V, var_2::V)</code> in the quadratic part of the quadratic expression.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L284-L289">source</a></section></article><h2 id="raw_status"><a class="docs-heading-anchor" href="#raw_status"><code>raw_status</code></a><a id="raw_status-1"></a><a class="docs-heading-anchor-permalink" href="#raw_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.raw_status" href="#JuMP.raw_status"><code>JuMP.raw_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">raw_status(model::GenericModel)</code></pre><p>Return the reason why the solver stopped in its own words (i.e., the MathOptInterface model attribute <code>RawStatusString</code>).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L527-L532">source</a></section></article><h2 id="read_from_file"><a class="docs-heading-anchor" href="#read_from_file"><code>read_from_file</code></a><a id="read_from_file-1"></a><a class="docs-heading-anchor-permalink" href="#read_from_file" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.read_from_file" href="#JuMP.read_from_file"><code>JuMP.read_from_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">read_from_file(
     filename::String;
     format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,
     kwargs...,
-)</code></pre><p>Return a JuMP model read from <code>filename</code> in the format <code>format</code>.</p><p>If the filename ends in <code>.gz</code>, it will be uncompressed using Gzip. If the filename ends in <code>.bz2</code>, it will be uncompressed using BZip2.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/file_formats.jl#L91-L104">source</a></section></article><h2 id="reduced_cost"><a class="docs-heading-anchor" href="#reduced_cost"><code>reduced_cost</code></a><a id="reduced_cost-1"></a><a class="docs-heading-anchor-permalink" href="#reduced_cost" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reduced_cost" href="#JuMP.reduced_cost"><code>JuMP.reduced_cost</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reduced_cost(x::GenericVariableRef{T})::T where {T}</code></pre><p>Return the reduced cost associated with variable <code>x</code>.</p><p>Equivalent to querying the shadow price of the active variable bound (if one exists and is active).</p><p>See also: <a href="#JuMP.shadow_price"><code>shadow_price</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L2107-L2116">source</a></section></article><h2 id="register"><a class="docs-heading-anchor" href="#register"><code>register</code></a><a id="register-1"></a><a class="docs-heading-anchor-permalink" href="#register" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.register" href="#JuMP.register"><code>JuMP.register</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">register(
+)</code></pre><p>Return a JuMP model read from <code>filename</code> in the format <code>format</code>.</p><p>If the filename ends in <code>.gz</code>, it will be uncompressed using Gzip. If the filename ends in <code>.bz2</code>, it will be uncompressed using BZip2.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/file_formats.jl#L91-L104">source</a></section></article><h2 id="reduced_cost"><a class="docs-heading-anchor" href="#reduced_cost"><code>reduced_cost</code></a><a id="reduced_cost-1"></a><a class="docs-heading-anchor-permalink" href="#reduced_cost" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reduced_cost" href="#JuMP.reduced_cost"><code>JuMP.reduced_cost</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reduced_cost(x::GenericVariableRef{T})::T where {T}</code></pre><p>Return the reduced cost associated with variable <code>x</code>.</p><p>Equivalent to querying the shadow price of the active variable bound (if one exists and is active).</p><p>See also: <a href="#JuMP.shadow_price"><code>shadow_price</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L2107-L2116">source</a></section></article><h2 id="register"><a class="docs-heading-anchor" href="#register"><code>register</code></a><a id="register-1"></a><a class="docs-heading-anchor-permalink" href="#register" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.register" href="#JuMP.register"><code>JuMP.register</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">register(
     model::Model,
     op::Symbol,
     dimension::Integer,
@@ -1103,7 +1103,7 @@
 
 julia&gt; register(model, :g, 2, g; autodiff = true)
 
-julia&gt; @NLobjective(model, Min, g(x[1], x[2]))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L654-L708">source</a></section><section><div><pre><code class="language-julia hljs">register(
+julia&gt; @NLobjective(model, Min, g(x[1], x[2]))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L654-L708">source</a></section><section><div><pre><code class="language-julia hljs">register(
     model::Model,
     s::Symbol,
     dimension::Integer,
@@ -1142,7 +1142,7 @@
 
 julia&gt; register(model, :g, 2, g, ∇g)
 
-julia&gt; @NLobjective(model, Min, g(x[1], x[2]))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L724-L794">source</a></section><section><div><pre><code class="language-julia hljs">register(
+julia&gt; @NLobjective(model, Min, g(x[1], x[2]))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L724-L794">source</a></section><section><div><pre><code class="language-julia hljs">register(
     model::Model,
     s::Symbol,
     dimension::Integer,
@@ -1166,7 +1166,7 @@
 julia&gt; register(model, :foo, 1, f, ∇f, ∇²f)
 
 julia&gt; @NLobjective(model, Min, foo(x))
-</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L820-L867">source</a></section></article><h2 id="relative_gap"><a class="docs-heading-anchor" href="#relative_gap"><code>relative_gap</code></a><a id="relative_gap-1"></a><a class="docs-heading-anchor-permalink" href="#relative_gap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relative_gap" href="#JuMP.relative_gap"><code>JuMP.relative_gap</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relative_gap(model::GenericModel)</code></pre><p>Return the final relative optimality gap after a call to <code>optimize!(model)</code>. Exact value depends upon implementation of MathOptInterface.RelativeGap() by the particular solver used for optimization.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L12-L18">source</a></section></article><h2 id="relax_integrality"><a class="docs-heading-anchor" href="#relax_integrality"><code>relax_integrality</code></a><a id="relax_integrality-1"></a><a class="docs-heading-anchor-permalink" href="#relax_integrality" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relax_integrality" href="#JuMP.relax_integrality"><code>JuMP.relax_integrality</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relax_integrality(model::GenericModel)</code></pre><p>Modifies <code>model</code> to &quot;relax&quot; all binary and integrality constraints on variables. Specifically,</p><ul><li>Binary constraints are deleted, and variable bounds are tightened if necessary to ensure the variable is constrained to the interval <span>$[0, 1]$</span>.</li><li>Integrality constraints are deleted without modifying variable bounds.</li><li>An error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).</li><li>All other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.</li></ul><p>Returns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L820-L867">source</a></section></article><h2 id="relative_gap"><a class="docs-heading-anchor" href="#relative_gap"><code>relative_gap</code></a><a id="relative_gap-1"></a><a class="docs-heading-anchor-permalink" href="#relative_gap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relative_gap" href="#JuMP.relative_gap"><code>JuMP.relative_gap</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relative_gap(model::GenericModel)</code></pre><p>Return the final relative optimality gap after a call to <code>optimize!(model)</code>. Exact value depends upon implementation of MathOptInterface.RelativeGap() by the particular solver used for optimization.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L12-L18">source</a></section></article><h2 id="relax_integrality"><a class="docs-heading-anchor" href="#relax_integrality"><code>relax_integrality</code></a><a id="relax_integrality-1"></a><a class="docs-heading-anchor-permalink" href="#relax_integrality" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relax_integrality" href="#JuMP.relax_integrality"><code>JuMP.relax_integrality</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relax_integrality(model::GenericModel)</code></pre><p>Modifies <code>model</code> to &quot;relax&quot; all binary and integrality constraints on variables. Specifically,</p><ul><li>Binary constraints are deleted, and variable bounds are tightened if necessary to ensure the variable is constrained to the interval <span>$[0, 1]$</span>.</li><li>Integrality constraints are deleted without modifying variable bounds.</li><li>An error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).</li><li>All other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.</li></ul><p>Returns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Bin);
 
@@ -1192,7 +1192,7 @@
  y ≥ 1
  y ≤ 10
  y integer
- x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L2219-L2268">source</a></section></article><h2 id="relax_with_penalty!"><a class="docs-heading-anchor" href="#relax_with_penalty!"><code>relax_with_penalty!</code></a><a id="relax_with_penalty!-1"></a><a class="docs-heading-anchor-permalink" href="#relax_with_penalty!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relax_with_penalty!" href="#JuMP.relax_with_penalty!"><code>JuMP.relax_with_penalty!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relax_with_penalty!(
+ x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L2219-L2268">source</a></section></article><h2 id="relax_with_penalty!"><a class="docs-heading-anchor" href="#relax_with_penalty!"><code>relax_with_penalty!</code></a><a id="relax_with_penalty!-1"></a><a class="docs-heading-anchor-permalink" href="#relax_with_penalty!" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.relax_with_penalty!" href="#JuMP.relax_with_penalty!"><code>JuMP.relax_with_penalty!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">relax_with_penalty!(
     model::GenericModel{T},
     [penalties::Dict{ConstraintRef,T}];
     [default::Union{Nothing,Real} = nothing,]
@@ -1235,7 +1235,7 @@
 Subject to
  c2 : 3 x + _[2] ≥ 0
  c1 : 2 x ≤ -1
- _[2] ≥ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1436-L1517">source</a></section></article><h2 id="remove_bridge"><a class="docs-heading-anchor" href="#remove_bridge"><code>remove_bridge</code></a><a id="remove_bridge-1"></a><a class="docs-heading-anchor-permalink" href="#remove_bridge" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.remove_bridge" href="#JuMP.remove_bridge"><code>JuMP.remove_bridge</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">remove_bridge(
+ _[2] ≥ 0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1436-L1517">source</a></section></article><h2 id="remove_bridge"><a class="docs-heading-anchor" href="#remove_bridge"><code>remove_bridge</code></a><a id="remove_bridge-1"></a><a class="docs-heading-anchor-permalink" href="#remove_bridge" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.remove_bridge" href="#JuMP.remove_bridge"><code>JuMP.remove_bridge</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">remove_bridge(
     model::GenericModel{S},
     BT::Type{&lt;:MOI.Bridges.AbstractBridge};
     coefficient_type::Type{T} = S,
@@ -1255,11 +1255,11 @@
            model,
            MOI.Bridges.Constraint.NumberConversionBridge;
            coefficient_type = Complex{Float64},
-       )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L648-L682">source</a></section></article><h2 id="reshape_set"><a class="docs-heading-anchor" href="#reshape_set"><code>reshape_set</code></a><a id="reshape_set-1"></a><a class="docs-heading-anchor-permalink" href="#reshape_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reshape_set" href="#JuMP.reshape_set"><code>JuMP.reshape_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reshape_set(vectorized_set::MOI.AbstractSet, shape::AbstractShape)</code></pre><p>Return a set in its original shape <code>shape</code> given its vectorized form <code>vectorized_form</code>.</p><p><strong>Example</strong></p><p>Given a <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> of vectorized form <code>[1, 2, 3] in MOI.PositiveSemidefinieConeTriangle(2)</code>, the following code returns the set of the original constraint <code>Symmetric(Matrix[1 2; 2 3]) in PSDCone()</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; reshape_set(MOI.PositiveSemidefiniteConeTriangle(2), SymmetricMatrixShape(2))
-PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L61-L78">source</a></section></article><h2 id="reshape_vector"><a class="docs-heading-anchor" href="#reshape_vector"><code>reshape_vector</code></a><a id="reshape_vector-1"></a><a class="docs-heading-anchor-permalink" href="#reshape_vector" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reshape_vector" href="#JuMP.reshape_vector"><code>JuMP.reshape_vector</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reshape_vector(vectorized_form::Vector, shape::AbstractShape)</code></pre><p>Return an object in its original shape <code>shape</code> given its vectorized form <code>vectorized_form</code>.</p><p><strong>Example</strong></p><p>Given a <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> of vectorized form <code>[1, 2, 3]</code>, the following code returns the matrix <code>Symmetric(Matrix[1 2; 2 3])</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; reshape_vector([1, 2, 3], SymmetricMatrixShape(2))
+       )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L648-L682">source</a></section></article><h2 id="reshape_set"><a class="docs-heading-anchor" href="#reshape_set"><code>reshape_set</code></a><a id="reshape_set-1"></a><a class="docs-heading-anchor-permalink" href="#reshape_set" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reshape_set" href="#JuMP.reshape_set"><code>JuMP.reshape_set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reshape_set(vectorized_set::MOI.AbstractSet, shape::AbstractShape)</code></pre><p>Return a set in its original shape <code>shape</code> given its vectorized form <code>vectorized_form</code>.</p><p><strong>Example</strong></p><p>Given a <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> of vectorized form <code>[1, 2, 3] in MOI.PositiveSemidefinieConeTriangle(2)</code>, the following code returns the set of the original constraint <code>Symmetric(Matrix[1 2; 2 3]) in PSDCone()</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; reshape_set(MOI.PositiveSemidefiniteConeTriangle(2), SymmetricMatrixShape(2))
+PSDCone()</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L61-L78">source</a></section></article><h2 id="reshape_vector"><a class="docs-heading-anchor" href="#reshape_vector"><code>reshape_vector</code></a><a id="reshape_vector-1"></a><a class="docs-heading-anchor-permalink" href="#reshape_vector" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reshape_vector" href="#JuMP.reshape_vector"><code>JuMP.reshape_vector</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reshape_vector(vectorized_form::Vector, shape::AbstractShape)</code></pre><p>Return an object in its original shape <code>shape</code> given its vectorized form <code>vectorized_form</code>.</p><p><strong>Example</strong></p><p>Given a <a href="#JuMP.SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a> of vectorized form <code>[1, 2, 3]</code>, the following code returns the matrix <code>Symmetric(Matrix[1 2; 2 3])</code>:</p><pre><code class="language-julia-repl hljs">julia&gt; reshape_vector([1, 2, 3], SymmetricMatrixShape(2))
 2×2 LinearAlgebra.Symmetric{Int64, Matrix{Int64}}:
  1  2
- 2  3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L81-L98">source</a></section></article><h2 id="result_count"><a class="docs-heading-anchor" href="#result_count"><code>result_count</code></a><a id="result_count-1"></a><a class="docs-heading-anchor-permalink" href="#result_count" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.result_count" href="#JuMP.result_count"><code>JuMP.result_count</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">result_count(model::GenericModel)</code></pre><p>Return the number of results available to query after a call to <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L514-L519">source</a></section></article><h2 id="reverse_sense"><a class="docs-heading-anchor" href="#reverse_sense"><code>reverse_sense</code></a><a id="reverse_sense-1"></a><a class="docs-heading-anchor-permalink" href="#reverse_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reverse_sense" href="#JuMP.reverse_sense"><code>JuMP.reverse_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reverse_sense(::Val{T}) where {T}</code></pre><p>Given an (in)equality symbol <code>T</code>, return a new <code>Val</code> object with the opposite (in)equality symbol.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L2249-L2254">source</a></section></article><h2 id="set_attribute"><a class="docs-heading-anchor" href="#set_attribute"><code>set_attribute</code></a><a id="set_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#set_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_attribute" href="#JuMP.set_attribute"><code>JuMP.set_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute, value)
+ 2  3</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L81-L98">source</a></section></article><h2 id="result_count"><a class="docs-heading-anchor" href="#result_count"><code>result_count</code></a><a id="result_count-1"></a><a class="docs-heading-anchor-permalink" href="#result_count" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.result_count" href="#JuMP.result_count"><code>JuMP.result_count</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">result_count(model::GenericModel)</code></pre><p>Return the number of results available to query after a call to <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L514-L519">source</a></section></article><h2 id="reverse_sense"><a class="docs-heading-anchor" href="#reverse_sense"><code>reverse_sense</code></a><a id="reverse_sense-1"></a><a class="docs-heading-anchor-permalink" href="#reverse_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.reverse_sense" href="#JuMP.reverse_sense"><code>JuMP.reverse_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reverse_sense(::Val{T}) where {T}</code></pre><p>Given an (in)equality symbol <code>T</code>, return a new <code>Val</code> object with the opposite (in)equality symbol.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L2249-L2254">source</a></section></article><h2 id="set_attribute"><a class="docs-heading-anchor" href="#set_attribute"><code>set_attribute</code></a><a id="set_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#set_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_attribute" href="#JuMP.set_attribute"><code>JuMP.set_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute, value)
 set_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute, value)
 set_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute, value)</code></pre><p>Set the value of a solver-specifc attribute <code>attr</code> to <code>value</code>.</p><p>This is equivalent to calling <a href="../../moi/reference/models/#MathOptInterface.set"><code>MOI.set</code></a> with the associated MOI model and, for variables and constraints, with the associated <a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> or <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintIndex"><code>MOI.ConstraintIndex</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
@@ -1273,7 +1273,7 @@
 
 julia&gt; set_attribute(x, MOI.VariableName(), &quot;x_new&quot;)
 
-julia&gt; set_attribute(c, MOI.ConstraintName(), &quot;c_new&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L864-L892">source</a></section><section><div><pre><code class="language-julia hljs">set_attribute(
+julia&gt; set_attribute(c, MOI.ConstraintName(), &quot;c_new&quot;)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L864-L892">source</a></section><section><div><pre><code class="language-julia hljs">set_attribute(
     model::Union{GenericModel,MOI.OptimizerWithAttributes},
     attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},
     value,
@@ -1289,7 +1289,7 @@
 
 julia&gt; set_attribute(opt, &quot;output_flag&quot;, true)
 
-julia&gt; set_attribute(opt, MOI.RawOptimizerAttribute(&quot;output_flag&quot;), false)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L920-L951">source</a></section></article><h2 id="set_attributes"><a class="docs-heading-anchor" href="#set_attributes"><code>set_attributes</code></a><a id="set_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#set_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_attributes" href="#JuMP.set_attributes"><code>JuMP.set_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_attributes(
+julia&gt; set_attribute(opt, MOI.RawOptimizerAttribute(&quot;output_flag&quot;), false)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L920-L951">source</a></section></article><h2 id="set_attributes"><a class="docs-heading-anchor" href="#set_attributes"><code>set_attributes</code></a><a id="set_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#set_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_attributes" href="#JuMP.set_attributes"><code>JuMP.set_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_attributes(
     destination::Union{
         GenericModel,
         MOI.OptimizerWithAttributes,
@@ -1307,7 +1307,7 @@
 
 julia&gt; set_attribute(model, &quot;tol&quot;, 1e-4)
 
-julia&gt; set_attribute(model, &quot;max_iter&quot;, 100)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L980-L1015">source</a></section></article><h2 id="set_binary"><a class="docs-heading-anchor" href="#set_binary"><code>set_binary</code></a><a id="set_binary-1"></a><a class="docs-heading-anchor-permalink" href="#set_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_binary" href="#JuMP.set_binary"><code>JuMP.set_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_binary(v::GenericVariableRef)</code></pre><p>Add a constraint on the variable <code>v</code> that it must take values in the set <span>$\{0,1\}$</span>.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+julia&gt; set_attribute(model, &quot;max_iter&quot;, 100)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L980-L1015">source</a></section></article><h2 id="set_binary"><a class="docs-heading-anchor" href="#set_binary"><code>set_binary</code></a><a id="set_binary-1"></a><a class="docs-heading-anchor-permalink" href="#set_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_binary" href="#JuMP.set_binary"><code>JuMP.set_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_binary(v::GenericVariableRef)</code></pre><p>Add a constraint on the variable <code>v</code> that it must take values in the set <span>$\{0,1\}$</span>.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -1317,7 +1317,7 @@
 julia&gt; set_binary(x)
 
 julia&gt; is_binary(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1393-L1416">source</a></section></article><h2 id="set_dual_start_value"><a class="docs-heading-anchor" href="#set_dual_start_value"><code>set_dual_start_value</code></a><a id="set_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_dual_start_value" href="#JuMP.set_dual_start_value"><code>JuMP.set_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_dual_start_value(con_ref::ConstraintRef, value)</code></pre><p>Set the dual start value (MOI attribute <code>ConstraintDualStart</code>) of the constraint <code>con_ref</code> to <code>value</code>. To remove a dual start value set it to <code>nothing</code>.</p><p>See also <a href="#JuMP.dual_start_value"><code>dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L104-L111">source</a></section></article><h2 id="set_integer"><a class="docs-heading-anchor" href="#set_integer"><code>set_integer</code></a><a id="set_integer-1"></a><a class="docs-heading-anchor-permalink" href="#set_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_integer" href="#JuMP.set_integer"><code>JuMP.set_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_integer(variable_ref::GenericVariableRef)</code></pre><p>Add an integrality constraint on the variable <code>variable_ref</code>.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1393-L1416">source</a></section></article><h2 id="set_dual_start_value"><a class="docs-heading-anchor" href="#set_dual_start_value"><code>set_dual_start_value</code></a><a id="set_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_dual_start_value" href="#JuMP.set_dual_start_value"><code>JuMP.set_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_dual_start_value(con_ref::ConstraintRef, value)</code></pre><p>Set the dual start value (MOI attribute <code>ConstraintDualStart</code>) of the constraint <code>con_ref</code> to <code>value</code>. To remove a dual start value set it to <code>nothing</code>.</p><p>See also <a href="#JuMP.dual_start_value"><code>dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L104-L111">source</a></section></article><h2 id="set_integer"><a class="docs-heading-anchor" href="#set_integer"><code>set_integer</code></a><a id="set_integer-1"></a><a class="docs-heading-anchor-permalink" href="#set_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_integer" href="#JuMP.set_integer"><code>JuMP.set_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_integer(variable_ref::GenericVariableRef)</code></pre><p>Add an integrality constraint on the variable <code>variable_ref</code>.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -1327,7 +1327,7 @@
 julia&gt; set_integer(x)
 
 julia&gt; is_integer(x)
-true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1261-L1283">source</a></section></article><h2 id="set_lower_bound"><a class="docs-heading-anchor" href="#set_lower_bound"><code>set_lower_bound</code></a><a id="set_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#set_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_lower_bound" href="#JuMP.set_lower_bound"><code>JuMP.set_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_lower_bound(v::GenericVariableRef, lower::Number)</code></pre><p>Set the lower bound of a variable. If one does not exist, create a new lower bound constraint.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1261-L1283">source</a></section></article><h2 id="set_lower_bound"><a class="docs-heading-anchor" href="#set_lower_bound"><code>set_lower_bound</code></a><a id="set_lower_bound-1"></a><a class="docs-heading-anchor-permalink" href="#set_lower_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_lower_bound" href="#JuMP.set_lower_bound"><code>JuMP.set_lower_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_lower_bound(v::GenericVariableRef, lower::Number)</code></pre><p>Set the lower bound of a variable. If one does not exist, create a new lower bound constraint.</p><p>See also <a href="#JuMP.LowerBoundRef"><code>LowerBoundRef</code></a>, <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 1.0);
 
@@ -1337,7 +1337,7 @@
 julia&gt; set_lower_bound(x, 2.0)
 
 julia&gt; lower_bound(x)
-2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L705-L729">source</a></section></article><h2 id="set_name"><a class="docs-heading-anchor" href="#set_name"><code>set_name</code></a><a id="set_name-1"></a><a class="docs-heading-anchor-permalink" href="#set_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_name" href="#JuMP.set_name"><code>JuMP.set_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_name(con_ref::ConstraintRef, s::AbstractString)</code></pre><p>Set a constraint&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L256-L260">source</a></section><section><div><pre><code class="language-julia hljs">set_name(v::GenericVariableRef, s::AbstractString)</code></pre><p>Set a variable&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L542-L546">source</a></section></article><h2 id="set_nonlinear_dual_start_value"><a class="docs-heading-anchor" href="#set_nonlinear_dual_start_value"><code>set_nonlinear_dual_start_value</code></a><a id="set_nonlinear_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_nonlinear_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_nonlinear_dual_start_value" href="#JuMP.set_nonlinear_dual_start_value"><code>JuMP.set_nonlinear_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_nonlinear_dual_start_value(
+2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L705-L729">source</a></section></article><h2 id="set_name"><a class="docs-heading-anchor" href="#set_name"><code>set_name</code></a><a id="set_name-1"></a><a class="docs-heading-anchor-permalink" href="#set_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_name" href="#JuMP.set_name"><code>JuMP.set_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_name(con_ref::ConstraintRef, s::AbstractString)</code></pre><p>Set a constraint&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L256-L260">source</a></section><section><div><pre><code class="language-julia hljs">set_name(v::GenericVariableRef, s::AbstractString)</code></pre><p>Set a variable&#39;s name attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L542-L546">source</a></section></article><h2 id="set_nonlinear_dual_start_value"><a class="docs-heading-anchor" href="#set_nonlinear_dual_start_value"><code>set_nonlinear_dual_start_value</code></a><a id="set_nonlinear_dual_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_nonlinear_dual_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_nonlinear_dual_start_value" href="#JuMP.set_nonlinear_dual_start_value"><code>JuMP.set_nonlinear_dual_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_nonlinear_dual_start_value(
     model::Model,
     start::Union{Nothing,Vector{Float64}},
 )</code></pre><p>Set the value of the MOI attribute <a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart"><code>MOI.NLPBlockDualStart</code></a>.</p><p>The start vector corresponds to the Lagrangian duals of the nonlinear constraints, in the order given by <a href="#JuMP.all_nonlinear_constraints"><code>all_nonlinear_constraints</code></a>. That is, you must pass a single start vector corresponding to all of the nonlinear constraints in a single function call; you cannot set the dual start value of nonlinear constraints one-by-one. The example below demonstrates how to use <a href="#JuMP.all_nonlinear_constraints"><code>all_nonlinear_constraints</code></a> to create a mapping between the nonlinear constraint references and the start vector.</p><p>Pass <code>nothing</code> to unset a previous start.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
@@ -1360,7 +1360,7 @@
 julia&gt; nonlinear_dual_start_value(model)
 2-element Vector{Float64}:
  -1.0
-  1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L590-L633">source</a></section></article><h2 id="set_nonlinear_objective"><a class="docs-heading-anchor" href="#set_nonlinear_objective"><code>set_nonlinear_objective</code></a><a id="set_nonlinear_objective-1"></a><a class="docs-heading-anchor-permalink" href="#set_nonlinear_objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_nonlinear_objective" href="#JuMP.set_nonlinear_objective"><code>JuMP.set_nonlinear_objective</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_nonlinear_objective(
+  1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L590-L633">source</a></section></article><h2 id="set_nonlinear_objective"><a class="docs-heading-anchor" href="#set_nonlinear_objective"><code>set_nonlinear_objective</code></a><a id="set_nonlinear_objective-1"></a><a class="docs-heading-anchor-permalink" href="#set_nonlinear_objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_nonlinear_objective" href="#JuMP.set_nonlinear_objective"><code>JuMP.set_nonlinear_objective</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_nonlinear_objective(
     model::Model,
     sense::MOI.OptimizationSense,
     expr::Expr,
@@ -1368,7 +1368,7 @@
 
 julia&gt; @variable(model, x);
 
-julia&gt; set_nonlinear_objective(model, MIN_SENSE, :($(x) + $(x)^2))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L136-L163">source</a></section></article><h2 id="set_normalized_coefficient"><a class="docs-heading-anchor" href="#set_normalized_coefficient"><code>set_normalized_coefficient</code></a><a id="set_normalized_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_coefficient" href="#JuMP.set_normalized_coefficient"><code>JuMP.set_normalized_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef, value)</code></pre><p>Set the coefficient of <code>variable</code> in the constraint <code>constraint</code> to <code>value</code>.</p><p>Note that prior to this step, JuMP will aggregate multiple terms containing the same variable. For example, given a constraint <code>2x + 3x &lt;= 2</code>, <code>set_normalized_coefficient(con, x, 4)</code> will create the constraint <code>4x &lt;= 2</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+julia&gt; set_nonlinear_objective(model, MIN_SENSE, :($(x) + $(x)^2))</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L136-L163">source</a></section></article><h2 id="set_normalized_coefficient"><a class="docs-heading-anchor" href="#set_normalized_coefficient"><code>set_normalized_coefficient</code></a><a id="set_normalized_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_coefficient" href="#JuMP.set_normalized_coefficient"><code>JuMP.set_normalized_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef, value)</code></pre><p>Set the coefficient of <code>variable</code> in the constraint <code>constraint</code> to <code>value</code>.</p><p>Note that prior to this step, JuMP will aggregate multiple terms containing the same variable. For example, given a constraint <code>2x + 3x &lt;= 2</code>, <code>set_normalized_coefficient(con, x, 4)</code> will create the constraint <code>4x &lt;= 2</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1379,7 +1379,7 @@
 julia&gt; set_normalized_coefficient(con, x, 4)
 
 julia&gt; con
-con : 4 x ≤ 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L726-L749">source</a></section></article><h2 id="set_normalized_coefficients"><a class="docs-heading-anchor" href="#set_normalized_coefficients"><code>set_normalized_coefficients</code></a><a id="set_normalized_coefficients-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_coefficients" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_coefficients" href="#JuMP.set_normalized_coefficients"><code>JuMP.set_normalized_coefficients</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_coefficients(
+con : 4 x ≤ 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L726-L749">source</a></section></article><h2 id="set_normalized_coefficients"><a class="docs-heading-anchor" href="#set_normalized_coefficients"><code>set_normalized_coefficients</code></a><a id="set_normalized_coefficients-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_coefficients" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_coefficients" href="#JuMP.set_normalized_coefficients"><code>JuMP.set_normalized_coefficients</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_coefficients(
     con_ref::ConstraintRef,
     variable,
     new_coefficients::Vector{Tuple{Int64,T}},
@@ -1394,7 +1394,7 @@
 julia&gt; set_normalized_coefficients(con, x, [(1, 2.0), (2, 5.0)])
 
 julia&gt; con
-con : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L769-L797">source</a></section></article><h2 id="set_normalized_rhs"><a class="docs-heading-anchor" href="#set_normalized_rhs"><code>set_normalized_rhs</code></a><a id="set_normalized_rhs-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_rhs" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_rhs" href="#JuMP.set_normalized_rhs"><code>JuMP.set_normalized_rhs</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_rhs(con_ref::ConstraintRef, value)</code></pre><p>Set the right-hand side term of <code>constraint</code> to <code>value</code>.</p><p>Note that prior to this step, JuMP will aggregate all constant terms onto the right-hand side of the constraint. For example, given a constraint <code>2x + 1 &lt;= 2</code>, <code>set_normalized_rhs(con, 4)</code> will create the constraint <code>2x &lt;= 4</code>, not <code>2x + 1 &lt;= 4</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+con : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L769-L797">source</a></section></article><h2 id="set_normalized_rhs"><a class="docs-heading-anchor" href="#set_normalized_rhs"><code>set_normalized_rhs</code></a><a id="set_normalized_rhs-1"></a><a class="docs-heading-anchor-permalink" href="#set_normalized_rhs" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_normalized_rhs" href="#JuMP.set_normalized_rhs"><code>JuMP.set_normalized_rhs</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_normalized_rhs(con_ref::ConstraintRef, value)</code></pre><p>Set the right-hand side term of <code>constraint</code> to <code>value</code>.</p><p>Note that prior to this step, JuMP will aggregate all constant terms onto the right-hand side of the constraint. For example, given a constraint <code>2x + 1 &lt;= 2</code>, <code>set_normalized_rhs(con, 4)</code> will create the constraint <code>2x &lt;= 4</code>, not <code>2x + 1 &lt;= 4</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -1404,15 +1404,15 @@
 julia&gt; set_normalized_rhs(con, 4)
 
 julia&gt; con
-con : 2 x ≤ 4</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L832-L855">source</a></section></article><h2 id="set_objective"><a class="docs-heading-anchor" href="#set_objective"><code>set_objective</code></a><a id="set_objective-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective" href="#JuMP.set_objective"><code>JuMP.set_objective</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective(model::AbstractModel, sense::MOI.OptimizationSense, func)</code></pre><p>The functional equivalent of the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p><p>Sets the objective sense and objective function simultaneously, and is equivalent to calling <a href="#JuMP.set_objective_sense"><code>set_objective_sense</code></a> and <a href="#JuMP.set_objective_function"><code>set_objective_function</code></a> separately.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+con : 2 x ≤ 4</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L832-L855">source</a></section></article><h2 id="set_objective"><a class="docs-heading-anchor" href="#set_objective"><code>set_objective</code></a><a id="set_objective-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective" href="#JuMP.set_objective"><code>JuMP.set_objective</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective(model::AbstractModel, sense::MOI.OptimizationSense, func)</code></pre><p>The functional equivalent of the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p><p>Sets the objective sense and objective function simultaneously, and is equivalent to calling <a href="#JuMP.set_objective_sense"><code>set_objective_sense</code></a> and <a href="#JuMP.set_objective_function"><code>set_objective_function</code></a> separately.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
 
-julia&gt; set_objective(model, MIN_SENSE, x)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L154-L173">source</a></section></article><h2 id="set_objective_coefficient"><a class="docs-heading-anchor" href="#set_objective_coefficient"><code>set_objective_coefficient</code></a><a id="set_objective_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_coefficient" href="#JuMP.set_objective_coefficient"><code>JuMP.set_objective_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_coefficient(model::GenericModel, variable::GenericVariableRef, coefficient::Real)</code></pre><p>Set the linear objective coefficient associated with <code>Variable</code> to <code>coefficient</code>.</p><p>Note: this function will throw an error if a nonlinear objective is set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L249-L255">source</a></section></article><h2 id="set_objective_function"><a class="docs-heading-anchor" href="#set_objective_function"><code>set_objective_function</code></a><a id="set_objective_function-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_function" href="#JuMP.set_objective_function"><code>JuMP.set_objective_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_function(model::GenericModel, func::MOI.AbstractFunction)
+julia&gt; set_objective(model, MIN_SENSE, x)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L154-L173">source</a></section></article><h2 id="set_objective_coefficient"><a class="docs-heading-anchor" href="#set_objective_coefficient"><code>set_objective_coefficient</code></a><a id="set_objective_coefficient-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_coefficient" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_coefficient" href="#JuMP.set_objective_coefficient"><code>JuMP.set_objective_coefficient</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_coefficient(model::GenericModel, variable::GenericVariableRef, coefficient::Real)</code></pre><p>Set the linear objective coefficient associated with <code>Variable</code> to <code>coefficient</code>.</p><p>Note: this function will throw an error if a nonlinear objective is set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L249-L255">source</a></section></article><h2 id="set_objective_function"><a class="docs-heading-anchor" href="#set_objective_function"><code>set_objective_function</code></a><a id="set_objective_function-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_function" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_function" href="#JuMP.set_objective_function"><code>JuMP.set_objective_function</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_function(model::GenericModel, func::MOI.AbstractFunction)
 set_objective_function(model::GenericModel, func::AbstractJuMPScalar)
 set_objective_function(model::GenericModel, func::Real)
-set_objective_function(model::GenericModel, func::Vector{&lt;:AbstractJuMPScalar})</code></pre><p>Sets the objective function of the model to the given function. See <a href="#JuMP.set_objective_sense"><code>set_objective_sense</code></a> to set the objective sense. These are low-level functions; the recommended way to set the objective is with the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L96-L106">source</a></section></article><h2 id="set_objective_sense"><a class="docs-heading-anchor" href="#set_objective_sense"><code>set_objective_sense</code></a><a id="set_objective_sense-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_sense" href="#JuMP.set_objective_sense"><code>JuMP.set_objective_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_sense(model::GenericModel, sense::MOI.OptimizationSense)</code></pre><p>Sets the objective sense of the model to the given sense. See <a href="#JuMP.set_objective_function"><code>set_objective_function</code></a> to set the objective function. These are low-level functions; the recommended way to set the objective is with the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/objective.jl#L84-L91">source</a></section></article><h2 id="set_optimize_hook"><a class="docs-heading-anchor" href="#set_optimize_hook"><code>set_optimize_hook</code></a><a id="set_optimize_hook-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimize_hook" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimize_hook" href="#JuMP.set_optimize_hook"><code>JuMP.set_optimize_hook</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimize_hook(model::GenericModel, f::Union{Function,Nothing})</code></pre><p>Set the function <code>f</code> as the optimize hook for <code>model</code>.</p><p><code>f</code> should have a signature <code>f(model::GenericModel; kwargs...)</code>, where the <code>kwargs</code> are those passed to <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p><p><strong>Notes</strong></p><ul><li>The optimize hook should generally modify the model, or some external state in some way, and then call <code>optimize!(model; ignore_optimize_hook = true)</code> to optimize the problem, bypassing the hook.</li><li>Use <code>set_optimize_hook(model, nothing)</code> to unset an optimize hook.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+set_objective_function(model::GenericModel, func::Vector{&lt;:AbstractJuMPScalar})</code></pre><p>Sets the objective function of the model to the given function. See <a href="#JuMP.set_objective_sense"><code>set_objective_sense</code></a> to set the objective sense. These are low-level functions; the recommended way to set the objective is with the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L96-L106">source</a></section></article><h2 id="set_objective_sense"><a class="docs-heading-anchor" href="#set_objective_sense"><code>set_objective_sense</code></a><a id="set_objective_sense-1"></a><a class="docs-heading-anchor-permalink" href="#set_objective_sense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_objective_sense" href="#JuMP.set_objective_sense"><code>JuMP.set_objective_sense</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_objective_sense(model::GenericModel, sense::MOI.OptimizationSense)</code></pre><p>Sets the objective sense of the model to the given sense. See <a href="#JuMP.set_objective_function"><code>set_objective_function</code></a> to set the objective function. These are low-level functions; the recommended way to set the objective is with the <a href="#JuMP.@objective"><code>@objective</code></a> macro.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/objective.jl#L84-L91">source</a></section></article><h2 id="set_optimize_hook"><a class="docs-heading-anchor" href="#set_optimize_hook"><code>set_optimize_hook</code></a><a id="set_optimize_hook-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimize_hook" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimize_hook" href="#JuMP.set_optimize_hook"><code>JuMP.set_optimize_hook</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimize_hook(model::GenericModel, f::Union{Function,Nothing})</code></pre><p>Set the function <code>f</code> as the optimize hook for <code>model</code>.</p><p><code>f</code> should have a signature <code>f(model::GenericModel; kwargs...)</code>, where the <code>kwargs</code> are those passed to <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p><p><strong>Notes</strong></p><ul><li>The optimize hook should generally modify the model, or some external state in some way, and then call <code>optimize!(model; ignore_optimize_hook = true)</code> to optimize the problem, bypassing the hook.</li><li>Use <code>set_optimize_hook(model, nothing)</code> to unset an optimize hook.</li></ul><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; function my_hook(model::Model; kwargs...)
            println(kwargs)
@@ -1429,7 +1429,7 @@
 Base.Pairs{Symbol, Bool, Tuple{Symbol}, NamedTuple{(:test_arg,), Tuple{Bool}}}(:test_arg =&gt; 1)
 Calling with `ignore_optimize_hook = true`
 ERROR: NoOptimizer()
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L951-L988">source</a></section></article><h2 id="set_optimizer"><a class="docs-heading-anchor" href="#set_optimizer"><code>set_optimizer</code></a><a id="set_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer" href="#JuMP.set_optimizer"><code>JuMP.set_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer(
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L951-L988">source</a></section></article><h2 id="set_optimizer"><a class="docs-heading-anchor" href="#set_optimizer"><code>set_optimizer</code></a><a id="set_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer" href="#JuMP.set_optimizer"><code>JuMP.set_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer(
     model::GenericModel,
     optimizer_factory;
     add_bridges::Bool = true,
@@ -1439,13 +1439,13 @@
 
 julia&gt; set_optimizer(model, () -&gt; HiGHS.Optimizer())
 
-julia&gt; set_optimizer(model, HiGHS.Optimizer; add_bridges = false)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L326-L357">source</a></section></article><h2 id="set_optimizer_attribute"><a class="docs-heading-anchor" href="#set_optimizer_attribute"><code>set_optimizer_attribute</code></a><a id="set_optimizer_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer_attribute" href="#JuMP.set_optimizer_attribute"><code>JuMP.set_optimizer_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer_attribute(
+julia&gt; set_optimizer(model, HiGHS.Optimizer; add_bridges = false)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L326-L357">source</a></section></article><h2 id="set_optimizer_attribute"><a class="docs-heading-anchor" href="#set_optimizer_attribute"><code>set_optimizer_attribute</code></a><a id="set_optimizer_attribute-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer_attribute" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer_attribute" href="#JuMP.set_optimizer_attribute"><code>JuMP.set_optimizer_attribute</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer_attribute(
     model::Union{GenericModel,MOI.OptimizerWithAttributes},
     attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},
     value,
 )</code></pre><p>Set the solver-specific attribute <code>attr</code> in <code>model</code> to <code>value</code>.</p><p>If <code>attr</code> is an <code>AbstractString</code>, this is equivalent to <code>set_optimizer_attribute(model, MOI.RawOptimizerAttribute(name), value)</code>.</p><div class="admonition is-compat"><header class="admonition-header">Compat</header><div class="admonition-body"><p>This method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using <a href="#JuMP.set_attribute"><code>set_attribute</code></a> instead.</p></div></div><p>See also: <a href="#JuMP.set_optimizer_attributes"><code>set_optimizer_attributes</code></a>, <a href="#JuMP.get_optimizer_attribute"><code>get_optimizer_attribute</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
-julia&gt; set_optimizer_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L51-L76">source</a></section></article><h2 id="set_optimizer_attributes"><a class="docs-heading-anchor" href="#set_optimizer_attributes"><code>set_optimizer_attributes</code></a><a id="set_optimizer_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer_attributes" href="#JuMP.set_optimizer_attributes"><code>JuMP.set_optimizer_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer_attributes(
+julia&gt; set_optimizer_attribute(model, MOI.Silent(), true)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L51-L76">source</a></section></article><h2 id="set_optimizer_attributes"><a class="docs-heading-anchor" href="#set_optimizer_attributes"><code>set_optimizer_attributes</code></a><a id="set_optimizer_attributes-1"></a><a class="docs-heading-anchor-permalink" href="#set_optimizer_attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_optimizer_attributes" href="#JuMP.set_optimizer_attributes"><code>JuMP.set_optimizer_attributes</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_optimizer_attributes(
     model::Union{GenericModel,MOI.OptimizerWithAttributes},
     pairs::Pair...,
 )</code></pre><p>Given a list of <code>attribute =&gt; value</code> pairs, calls <code>set_optimizer_attribute(model, attribute, value)</code> for each pair.</p><div class="admonition is-compat"><header class="admonition-header">Compat</header><div class="admonition-body"><p>This method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using <a href="#JuMP.set_attributes"><code>set_attributes</code></a> instead.</p></div></div><p>See also: <a href="#JuMP.set_optimizer_attribute"><code>set_optimizer_attribute</code></a>, <a href="#JuMP.get_optimizer_attribute"><code>get_optimizer_attribute</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import Ipopt
@@ -1458,7 +1458,7 @@
 
 julia&gt; set_optimizer_attribute(model, &quot;tol&quot;, 1e-4)
 
-julia&gt; set_optimizer_attribute(model, &quot;max_iter&quot;, 100)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L79-L113">source</a></section></article><h2 id="set_parameter_value"><a class="docs-heading-anchor" href="#set_parameter_value"><code>set_parameter_value</code></a><a id="set_parameter_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_parameter_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_parameter_value" href="#JuMP.set_parameter_value"><code>JuMP.set_parameter_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_parameter_value(x::GenericVariableRef, value)</code></pre><p>Update the parameter constraint on the variable <code>x</code> to <code>value</code>.</p><p>Errors if <code>x</code> is not a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+julia&gt; set_optimizer_attribute(model, &quot;max_iter&quot;, 100)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L79-L113">source</a></section></article><h2 id="set_parameter_value"><a class="docs-heading-anchor" href="#set_parameter_value"><code>set_parameter_value</code></a><a id="set_parameter_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_parameter_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_parameter_value" href="#JuMP.set_parameter_value"><code>JuMP.set_parameter_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_parameter_value(x::GenericVariableRef, value)</code></pre><p>Update the parameter constraint on the variable <code>x</code> to <code>value</code>.</p><p>Errors if <code>x</code> is not a parameter.</p><p>See also <a href="#JuMP.ParameterRef"><code>ParameterRef</code></a>, <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, p in Parameter(2))
 p
@@ -1469,13 +1469,13 @@
 julia&gt; set_parameter_value(p, 2.5)
 
 julia&gt; parameter_value(p)
-2.5</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1565-L1591">source</a></section></article><h2 id="set_silent"><a class="docs-heading-anchor" href="#set_silent"><code>set_silent</code></a><a id="set_silent-1"></a><a class="docs-heading-anchor-permalink" href="#set_silent" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_silent" href="#JuMP.set_silent"><code>JuMP.set_silent</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_silent(model::GenericModel)</code></pre><p>Takes precedence over any other attribute controlling verbosity and requires the solver to produce no output.</p><p>See also: <a href="#JuMP.unset_silent"><code>unset_silent</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L154-L161">source</a></section></article><h2 id="set_start_value"><a class="docs-heading-anchor" href="#set_start_value"><code>set_start_value</code></a><a id="set_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_start_value" href="#JuMP.set_start_value"><code>JuMP.set_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_start_value(con_ref::ConstraintRef, value)</code></pre><p>Set the primal start value (<a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>) of the constraint <code>con_ref</code> to <code>value</code>. To remove a primal start value set it to <code>nothing</code>.</p><p>See also <a href="#JuMP.start_value"><code>start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L158-L166">source</a></section><section><div><pre><code class="language-julia hljs">set_start_value(variable::GenericVariableRef, value::Union{Real,Nothing})</code></pre><p>Set the start value (MOI attribute <code>VariablePrimalStart</code>) of the <code>variable</code> to <code>value</code>.</p><p>Pass <code>nothing</code> to unset the start value.</p><p>Note: <code>VariablePrimalStart</code>s are sometimes called &quot;MIP-starts&quot; or &quot;warmstarts&quot;.</p><p>See also <a href="#JuMP.start_value"><code>start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1665-L1676">source</a></section></article><h2 id="set_start_values"><a class="docs-heading-anchor" href="#set_start_values"><code>set_start_values</code></a><a id="set_start_values-1"></a><a class="docs-heading-anchor-permalink" href="#set_start_values" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_start_values" href="#JuMP.set_start_values"><code>JuMP.set_start_values</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_start_values(
+2.5</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1565-L1591">source</a></section></article><h2 id="set_silent"><a class="docs-heading-anchor" href="#set_silent"><code>set_silent</code></a><a id="set_silent-1"></a><a class="docs-heading-anchor-permalink" href="#set_silent" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_silent" href="#JuMP.set_silent"><code>JuMP.set_silent</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_silent(model::GenericModel)</code></pre><p>Takes precedence over any other attribute controlling verbosity and requires the solver to produce no output.</p><p>See also: <a href="#JuMP.unset_silent"><code>unset_silent</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L154-L161">source</a></section></article><h2 id="set_start_value"><a class="docs-heading-anchor" href="#set_start_value"><code>set_start_value</code></a><a id="set_start_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_start_value" href="#JuMP.set_start_value"><code>JuMP.set_start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_start_value(con_ref::ConstraintRef, value)</code></pre><p>Set the primal start value (<a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>) of the constraint <code>con_ref</code> to <code>value</code>. To remove a primal start value set it to <code>nothing</code>.</p><p>See also <a href="#JuMP.start_value"><code>start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L158-L166">source</a></section><section><div><pre><code class="language-julia hljs">set_start_value(variable::GenericVariableRef, value::Union{Real,Nothing})</code></pre><p>Set the start value (MOI attribute <code>VariablePrimalStart</code>) of the <code>variable</code> to <code>value</code>.</p><p>Pass <code>nothing</code> to unset the start value.</p><p>Note: <code>VariablePrimalStart</code>s are sometimes called &quot;MIP-starts&quot; or &quot;warmstarts&quot;.</p><p>See also <a href="#JuMP.start_value"><code>start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1665-L1676">source</a></section></article><h2 id="set_start_values"><a class="docs-heading-anchor" href="#set_start_values"><code>set_start_values</code></a><a id="set_start_values-1"></a><a class="docs-heading-anchor-permalink" href="#set_start_values" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_start_values" href="#JuMP.set_start_values"><code>JuMP.set_start_values</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_start_values(
     model::GenericModel;
     variable_primal_start::Union{Nothing,Function} = value,
     constraint_primal_start::Union{Nothing,Function} = value,
     constraint_dual_start::Union{Nothing,Function} = dual,
     nonlinear_dual_start::Union{Nothing,Function} = nonlinear_dual_start_value,
-)</code></pre><p>Set the primal and dual starting values in <code>model</code> using the functions provided.</p><p>If any keyword argument is <code>nothing</code>, the corresponding start value is skipped.</p><p>If the optimizer does not support setting the starting value, the value will be skipped.</p><p><strong><code>variable_primal_start</code></strong></p><p>This function controls the primal starting solution for the variables. It is equivalent to calling <a href="#JuMP.set_start_value"><code>set_start_value</code></a> for each variable, or setting the <a href="../../moi/reference/variables/#MathOptInterface.VariablePrimalStart"><code>MOI.VariablePrimalStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>variable_primal_start(x::VariableRef)</code> that maps each variable <code>x</code> to the starting primal value.</p><p>The default is <a href="#JuMP.value"><code>value</code></a>.</p><p><strong><code>constraint_primal_start</code></strong></p><p>This function controls the primal starting solution for the constraints. It is equivalent to calling <a href="#JuMP.set_start_value"><code>set_start_value</code></a> for each constraint, or setting the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>constraint_primal_start(ci::ConstraintRef)</code> that maps each constraint <code>ci</code> to the starting primal value.</p><p>The default is <a href="#JuMP.value"><code>value</code></a>.</p><p><strong><code>constraint_dual_start</code></strong></p><p>This function controls the dual starting solution for the constraints. It is equivalent to calling <a href="#JuMP.set_dual_start_value"><code>set_dual_start_value</code></a> for each constraint, or setting the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintDualStart"><code>MOI.ConstraintDualStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>constraint_dual_start(ci::ConstraintRef)</code> that maps each constraint <code>ci</code> to the starting dual value.</p><p>The default is <a href="#JuMP.dual"><code>dual</code></a>.</p><p><strong><code>nonlinear_dual_start</code></strong></p><p>This function controls the dual starting solution for the nonlinear constraints It is equivalent to calling <a href="#JuMP.set_nonlinear_dual_start_value"><code>set_nonlinear_dual_start_value</code></a>.</p><p>If it is a function, it must have the form <code>nonlinear_dual_start(model::GenericModel)</code> that returns a vector corresponding to the dual start of the constraints.</p><p>The default is <a href="#JuMP.nonlinear_dual_start_value"><code>nonlinear_dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L1082-L1140">source</a></section></article><h2 id="set_string_names_on_creation"><a class="docs-heading-anchor" href="#set_string_names_on_creation"><code>set_string_names_on_creation</code></a><a id="set_string_names_on_creation-1"></a><a class="docs-heading-anchor-permalink" href="#set_string_names_on_creation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_string_names_on_creation" href="#JuMP.set_string_names_on_creation"><code>JuMP.set_string_names_on_creation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_string_names_on_creation(model::GenericModel, value::Bool)</code></pre><p>Set the default argument of the <code>set_string_name</code> keyword in the <a href="#JuMP.@variable"><code>@variable</code></a> and <a href="#JuMP.@constraint"><code>@constraint</code></a> macros to <code>value</code>. This is used to determine whether to assign <code>String</code> names to all variables and constraints in <code>model</code>.</p><p>By default, <code>value</code> is <code>true</code>. However, for larger models calling <code>set_string_names_on_creation(model, false)</code> can improve performance at the cost of reducing the readability of printing and solver log messages.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L540-L551">source</a></section></article><h2 id="set_time_limit_sec"><a class="docs-heading-anchor" href="#set_time_limit_sec"><code>set_time_limit_sec</code></a><a id="set_time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#set_time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_time_limit_sec" href="#JuMP.set_time_limit_sec"><code>JuMP.set_time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_time_limit_sec(model::GenericModel, limit::Float64)</code></pre><p>Set the time limit (in seconds) of the solver.</p><p>Can be unset using <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a> or with <code>limit</code> set to <code>nothing</code>.</p><p>See also: <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a>, <a href="#JuMP.time_limit_sec"><code>time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L178-L187">source</a></section></article><h2 id="set_upper_bound"><a class="docs-heading-anchor" href="#set_upper_bound"><code>set_upper_bound</code></a><a id="set_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#set_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_upper_bound" href="#JuMP.set_upper_bound"><code>JuMP.set_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_upper_bound(v::GenericVariableRef, upper::Number)</code></pre><p>Set the upper bound of a variable. If one does not exist, create an upper bound constraint.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+)</code></pre><p>Set the primal and dual starting values in <code>model</code> using the functions provided.</p><p>If any keyword argument is <code>nothing</code>, the corresponding start value is skipped.</p><p>If the optimizer does not support setting the starting value, the value will be skipped.</p><p><strong><code>variable_primal_start</code></strong></p><p>This function controls the primal starting solution for the variables. It is equivalent to calling <a href="#JuMP.set_start_value"><code>set_start_value</code></a> for each variable, or setting the <a href="../../moi/reference/variables/#MathOptInterface.VariablePrimalStart"><code>MOI.VariablePrimalStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>variable_primal_start(x::VariableRef)</code> that maps each variable <code>x</code> to the starting primal value.</p><p>The default is <a href="#JuMP.value"><code>value</code></a>.</p><p><strong><code>constraint_primal_start</code></strong></p><p>This function controls the primal starting solution for the constraints. It is equivalent to calling <a href="#JuMP.set_start_value"><code>set_start_value</code></a> for each constraint, or setting the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>constraint_primal_start(ci::ConstraintRef)</code> that maps each constraint <code>ci</code> to the starting primal value.</p><p>The default is <a href="#JuMP.value"><code>value</code></a>.</p><p><strong><code>constraint_dual_start</code></strong></p><p>This function controls the dual starting solution for the constraints. It is equivalent to calling <a href="#JuMP.set_dual_start_value"><code>set_dual_start_value</code></a> for each constraint, or setting the <a href="../../moi/reference/constraints/#MathOptInterface.ConstraintDualStart"><code>MOI.ConstraintDualStart</code></a> attribute.</p><p>If it is a function, it must have the form <code>constraint_dual_start(ci::ConstraintRef)</code> that maps each constraint <code>ci</code> to the starting dual value.</p><p>The default is <a href="#JuMP.dual"><code>dual</code></a>.</p><p><strong><code>nonlinear_dual_start</code></strong></p><p>This function controls the dual starting solution for the nonlinear constraints It is equivalent to calling <a href="#JuMP.set_nonlinear_dual_start_value"><code>set_nonlinear_dual_start_value</code></a>.</p><p>If it is a function, it must have the form <code>nonlinear_dual_start(model::GenericModel)</code> that returns a vector corresponding to the dual start of the constraints.</p><p>The default is <a href="#JuMP.nonlinear_dual_start_value"><code>nonlinear_dual_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L1082-L1140">source</a></section></article><h2 id="set_string_names_on_creation"><a class="docs-heading-anchor" href="#set_string_names_on_creation"><code>set_string_names_on_creation</code></a><a id="set_string_names_on_creation-1"></a><a class="docs-heading-anchor-permalink" href="#set_string_names_on_creation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_string_names_on_creation" href="#JuMP.set_string_names_on_creation"><code>JuMP.set_string_names_on_creation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_string_names_on_creation(model::GenericModel, value::Bool)</code></pre><p>Set the default argument of the <code>set_string_name</code> keyword in the <a href="#JuMP.@variable"><code>@variable</code></a> and <a href="#JuMP.@constraint"><code>@constraint</code></a> macros to <code>value</code>. This is used to determine whether to assign <code>String</code> names to all variables and constraints in <code>model</code>.</p><p>By default, <code>value</code> is <code>true</code>. However, for larger models calling <code>set_string_names_on_creation(model, false)</code> can improve performance at the cost of reducing the readability of printing and solver log messages.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L540-L551">source</a></section></article><h2 id="set_time_limit_sec"><a class="docs-heading-anchor" href="#set_time_limit_sec"><code>set_time_limit_sec</code></a><a id="set_time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#set_time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_time_limit_sec" href="#JuMP.set_time_limit_sec"><code>JuMP.set_time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_time_limit_sec(model::GenericModel, limit::Float64)</code></pre><p>Set the time limit (in seconds) of the solver.</p><p>Can be unset using <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a> or with <code>limit</code> set to <code>nothing</code>.</p><p>See also: <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a>, <a href="#JuMP.time_limit_sec"><code>time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L178-L187">source</a></section></article><h2 id="set_upper_bound"><a class="docs-heading-anchor" href="#set_upper_bound"><code>set_upper_bound</code></a><a id="set_upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#set_upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_upper_bound" href="#JuMP.set_upper_bound"><code>JuMP.set_upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_upper_bound(v::GenericVariableRef, upper::Number)</code></pre><p>Set the upper bound of a variable. If one does not exist, create an upper bound constraint.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &lt;= 1.0);
 
@@ -1485,7 +1485,7 @@
 julia&gt; set_upper_bound(x, 2.0)
 
 julia&gt; upper_bound(x)
-2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L876-L900">source</a></section></article><h2 id="set_value"><a class="docs-heading-anchor" href="#set_value"><code>set_value</code></a><a id="set_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_value" href="#JuMP.set_value"><code>JuMP.set_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_value(p::NonlinearParameter, v::Number)</code></pre><p>Store the value <code>v</code> in the nonlinear parameter <code>p</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L876-L900">source</a></section></article><h2 id="set_value"><a class="docs-heading-anchor" href="#set_value"><code>set_value</code></a><a id="set_value-1"></a><a class="docs-heading-anchor-permalink" href="#set_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.set_value" href="#JuMP.set_value"><code>JuMP.set_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set_value(p::NonlinearParameter, v::Number)</code></pre><p>Store the value <code>v</code> in the nonlinear parameter <code>p</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @NLparameter(model, p == 0)
 p == 0.0
@@ -1494,7 +1494,7 @@
 5
 
 julia&gt; value(p)
-5.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L250-L269">source</a></section></article><h2 id="shadow_price"><a class="docs-heading-anchor" href="#shadow_price"><code>shadow_price</code></a><a id="shadow_price-1"></a><a class="docs-heading-anchor-permalink" href="#shadow_price" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.shadow_price" href="#JuMP.shadow_price"><code>JuMP.shadow_price</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">shadow_price(con_ref::ConstraintRef)</code></pre><p>Return the change in the objective from an infinitesimal relaxation of the constraint.</p><p>This value is computed from <a href="#JuMP.dual"><code>dual</code></a> and can be queried only when <code>has_duals</code> is <code>true</code> and the objective sense is <code>MIN_SENSE</code> or <code>MAX_SENSE</code> (not <code>FEASIBILITY_SENSE</code>). For linear constraints, the shadow prices differ at most in sign from the <code>dual</code> value depending on the objective sense.</p><p>See also <a href="#JuMP.reduced_cost"><code>reduced_cost</code></a>.</p><p><strong>Notes</strong></p><ul><li>The function simply translates signs from <code>dual</code> and does not validate the conditions needed to guarantee the sensitivity interpretation of the shadow price. The caller is responsible, e.g., for checking whether the solver converged to an optimal primal-dual pair or a proof of infeasibility.</li><li>The computation is based on the current objective sense of the model. If this has changed since the last solve, the results will be incorrect.</li><li>Relaxation of equality constraints (and hence the shadow price) is defined based on which sense of the equality constraint is active.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1069-L1092">source</a></section></article><h2 id="shape"><a class="docs-heading-anchor" href="#shape"><code>shape</code></a><a id="shape-1"></a><a class="docs-heading-anchor-permalink" href="#shape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.shape" href="#JuMP.shape"><code>JuMP.shape</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">shape(c::AbstractConstraint)::AbstractShape</code></pre><p>Return the shape of the constraint <code>c</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L101-L105">source</a></section></article><h2 id="show_backend_summary"><a class="docs-heading-anchor" href="#show_backend_summary"><code>show_backend_summary</code></a><a id="show_backend_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_backend_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_backend_summary" href="#JuMP.show_backend_summary"><code>JuMP.show_backend_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_backend_summary(io::IO, model::GenericModel)</code></pre><p>Print a summary of the optimizer backing <code>model</code>.</p><p><code>AbstractModel</code>s should implement this method.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L262-L268">source</a></section></article><h2 id="show_constraints_summary"><a class="docs-heading-anchor" href="#show_constraints_summary"><code>show_constraints_summary</code></a><a id="show_constraints_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_constraints_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_constraints_summary" href="#JuMP.show_constraints_summary"><code>JuMP.show_constraints_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_constraints_summary(io::IO, model::AbstractModel)</code></pre><p>Write to <code>io</code> a summary of the number of constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L245-L249">source</a></section></article><h2 id="show_objective_function_summary"><a class="docs-heading-anchor" href="#show_objective_function_summary"><code>show_objective_function_summary</code></a><a id="show_objective_function_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_objective_function_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_objective_function_summary" href="#JuMP.show_objective_function_summary"><code>JuMP.show_objective_function_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_objective_function_summary(io::IO, model::AbstractModel)</code></pre><p>Write to <code>io</code> a summary of the objective function type.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/print.jl#L229-L233">source</a></section></article><h2 id="simplex_iterations"><a class="docs-heading-anchor" href="#simplex_iterations"><code>simplex_iterations</code></a><a id="simplex_iterations-1"></a><a class="docs-heading-anchor-permalink" href="#simplex_iterations" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.simplex_iterations" href="#JuMP.simplex_iterations"><code>JuMP.simplex_iterations</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">simplex_iterations(model::GenericModel)</code></pre><p>Gets the cumulative number of simplex iterations during the most-recent optimization.</p><p>Solvers must implement <code>MOI.SimplexIterations()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L588-L595">source</a></section></article><h2 id="solution_summary"><a class="docs-heading-anchor" href="#solution_summary"><code>solution_summary</code></a><a id="solution_summary-1"></a><a class="docs-heading-anchor-permalink" href="#solution_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solution_summary" href="#JuMP.solution_summary"><code>JuMP.solution_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solution_summary(model::GenericModel; result::Int = 1, verbose::Bool = false)</code></pre><p>Return a struct that can be used print a summary of the solution in result <code>result</code>.</p><p>If <code>verbose=true</code>, write out the primal solution for every variable and the dual solution for every constraint, excluding those with empty names.</p><p><strong>Example</strong></p><p>When called at the REPL, the summary is automatically printed:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+5.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L250-L269">source</a></section></article><h2 id="shadow_price"><a class="docs-heading-anchor" href="#shadow_price"><code>shadow_price</code></a><a id="shadow_price-1"></a><a class="docs-heading-anchor-permalink" href="#shadow_price" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.shadow_price" href="#JuMP.shadow_price"><code>JuMP.shadow_price</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">shadow_price(con_ref::ConstraintRef)</code></pre><p>Return the change in the objective from an infinitesimal relaxation of the constraint.</p><p>This value is computed from <a href="#JuMP.dual"><code>dual</code></a> and can be queried only when <code>has_duals</code> is <code>true</code> and the objective sense is <code>MIN_SENSE</code> or <code>MAX_SENSE</code> (not <code>FEASIBILITY_SENSE</code>). For linear constraints, the shadow prices differ at most in sign from the <code>dual</code> value depending on the objective sense.</p><p>See also <a href="#JuMP.reduced_cost"><code>reduced_cost</code></a>.</p><p><strong>Notes</strong></p><ul><li>The function simply translates signs from <code>dual</code> and does not validate the conditions needed to guarantee the sensitivity interpretation of the shadow price. The caller is responsible, e.g., for checking whether the solver converged to an optimal primal-dual pair or a proof of infeasibility.</li><li>The computation is based on the current objective sense of the model. If this has changed since the last solve, the results will be incorrect.</li><li>Relaxation of equality constraints (and hence the shadow price) is defined based on which sense of the equality constraint is active.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1069-L1092">source</a></section></article><h2 id="shape"><a class="docs-heading-anchor" href="#shape"><code>shape</code></a><a id="shape-1"></a><a class="docs-heading-anchor-permalink" href="#shape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.shape" href="#JuMP.shape"><code>JuMP.shape</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">shape(c::AbstractConstraint)::AbstractShape</code></pre><p>Return the shape of the constraint <code>c</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L101-L105">source</a></section></article><h2 id="show_backend_summary"><a class="docs-heading-anchor" href="#show_backend_summary"><code>show_backend_summary</code></a><a id="show_backend_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_backend_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_backend_summary" href="#JuMP.show_backend_summary"><code>JuMP.show_backend_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_backend_summary(io::IO, model::GenericModel)</code></pre><p>Print a summary of the optimizer backing <code>model</code>.</p><p><code>AbstractModel</code>s should implement this method.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L262-L268">source</a></section></article><h2 id="show_constraints_summary"><a class="docs-heading-anchor" href="#show_constraints_summary"><code>show_constraints_summary</code></a><a id="show_constraints_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_constraints_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_constraints_summary" href="#JuMP.show_constraints_summary"><code>JuMP.show_constraints_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_constraints_summary(io::IO, model::AbstractModel)</code></pre><p>Write to <code>io</code> a summary of the number of constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L245-L249">source</a></section></article><h2 id="show_objective_function_summary"><a class="docs-heading-anchor" href="#show_objective_function_summary"><code>show_objective_function_summary</code></a><a id="show_objective_function_summary-1"></a><a class="docs-heading-anchor-permalink" href="#show_objective_function_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.show_objective_function_summary" href="#JuMP.show_objective_function_summary"><code>JuMP.show_objective_function_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">show_objective_function_summary(io::IO, model::AbstractModel)</code></pre><p>Write to <code>io</code> a summary of the objective function type.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/print.jl#L229-L233">source</a></section></article><h2 id="simplex_iterations"><a class="docs-heading-anchor" href="#simplex_iterations"><code>simplex_iterations</code></a><a id="simplex_iterations-1"></a><a class="docs-heading-anchor-permalink" href="#simplex_iterations" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.simplex_iterations" href="#JuMP.simplex_iterations"><code>JuMP.simplex_iterations</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">simplex_iterations(model::GenericModel)</code></pre><p>Gets the cumulative number of simplex iterations during the most-recent optimization.</p><p>Solvers must implement <code>MOI.SimplexIterations()</code> to use this function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L588-L595">source</a></section></article><h2 id="solution_summary"><a class="docs-heading-anchor" href="#solution_summary"><code>solution_summary</code></a><a id="solution_summary-1"></a><a class="docs-heading-anchor-permalink" href="#solution_summary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solution_summary" href="#JuMP.solution_summary"><code>JuMP.solution_summary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solution_summary(model::GenericModel; result::Int = 1, verbose::Bool = false)</code></pre><p>Return a struct that can be used print a summary of the solution in result <code>result</code>.</p><p>If <code>verbose=true</code>, write out the primal solution for every variable and the dual solution for every constraint, excluding those with empty names.</p><p><strong>Example</strong></p><p>When called at the REPL, the summary is automatically printed:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; solution_summary(model)
 * Solver : No optimizer attached.
@@ -1530,7 +1530,7 @@
   Primal status      : NO_SOLUTION
   Dual status        : NO_SOLUTION
 
-* Work counters</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/solution_summary.jl#L38-L94">source</a></section></article><h2 id="solve_time"><a class="docs-heading-anchor" href="#solve_time"><code>solve_time</code></a><a id="solve_time-1"></a><a class="docs-heading-anchor-permalink" href="#solve_time" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solve_time" href="#JuMP.solve_time"><code>JuMP.solve_time</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solve_time(model::GenericModel)</code></pre><p>If available, returns the solve time reported by the solver. Returns &quot;ArgumentError: ModelLike of type <code>Solver.Optimizer</code> does not support accessing the attribute MathOptInterface.SolveTimeSec()&quot; if the attribute is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L576-L583">source</a></section></article><h2 id="solver_name"><a class="docs-heading-anchor" href="#solver_name"><code>solver_name</code></a><a id="solver_name-1"></a><a class="docs-heading-anchor-permalink" href="#solver_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solver_name" href="#JuMP.solver_name"><code>JuMP.solver_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solver_name(model::GenericModel)</code></pre><p>If available, returns the <code>SolverName</code> property of the underlying optimizer.</p><p>Returns <code>&quot;No optimizer attached&quot;</code> in <code>AUTOMATIC</code> or <code>MANUAL</code> modes when no optimizer is attached.</p><p>Returns <code>&quot;SolverName() attribute not implemented by the optimizer.&quot;</code> if the attribute is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L232-L242">source</a></section></article><h2 id="start_value"><a class="docs-heading-anchor" href="#start_value"><code>start_value</code></a><a id="start_value-1"></a><a class="docs-heading-anchor-permalink" href="#start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.start_value" href="#JuMP.start_value"><code>JuMP.start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">start_value(con_ref::ConstraintRef)</code></pre><p>Return the primal start value (<a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>) of the constraint <code>con_ref</code>.</p><p>Note: If no primal start value has been set, <code>start_value</code> will return <code>nothing</code>.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L214-L224">source</a></section><section><div><pre><code class="language-julia hljs">start_value(v::GenericVariableRef)</code></pre><p>Return the start value (MOI attribute <code>VariablePrimalStart</code>) of the variable <code>v</code>.</p><p>Note: <code>VariablePrimalStart</code>s are sometimes called &quot;MIP-starts&quot; or &quot;warmstarts&quot;.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1639-L1648">source</a></section></article><h2 id="termination_status"><a class="docs-heading-anchor" href="#termination_status"><code>termination_status</code></a><a id="termination_status-1"></a><a class="docs-heading-anchor-permalink" href="#termination_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.termination_status" href="#JuMP.termination_status"><code>JuMP.termination_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">termination_status(model::GenericModel)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.TerminationStatusCode"><code>MOI.TerminationStatusCode</code></a> describing why the solver stopped (i.e., the <a href="../../moi/reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a> attribute).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L497-L502">source</a></section></article><h2 id="time_limit_sec"><a class="docs-heading-anchor" href="#time_limit_sec"><code>time_limit_sec</code></a><a id="time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.time_limit_sec" href="#JuMP.time_limit_sec"><code>JuMP.time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">time_limit_sec(model::GenericModel)</code></pre><p>Return the time limit (in seconds) of the <code>model</code>.</p><p>Returns <code>nothing</code> if unset.</p><p>See also: <a href="#JuMP.set_time_limit_sec"><code>set_time_limit_sec</code></a>, <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L207-L215">source</a></section></article><h2 id="triangle_vec"><a class="docs-heading-anchor" href="#triangle_vec"><code>triangle_vec</code></a><a id="triangle_vec-1"></a><a class="docs-heading-anchor-permalink" href="#triangle_vec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.triangle_vec" href="#JuMP.triangle_vec"><code>JuMP.triangle_vec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">triangle_vec(matrix::Matrix)</code></pre><p>Return the upper triangle of a matrix concatenated into a vector in the order required by JuMP and MathOptInterface for <code>Triangle</code> sets.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+* Work counters</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/solution_summary.jl#L38-L94">source</a></section></article><h2 id="solve_time"><a class="docs-heading-anchor" href="#solve_time"><code>solve_time</code></a><a id="solve_time-1"></a><a class="docs-heading-anchor-permalink" href="#solve_time" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solve_time" href="#JuMP.solve_time"><code>JuMP.solve_time</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solve_time(model::GenericModel)</code></pre><p>If available, returns the solve time reported by the solver. Returns &quot;ArgumentError: ModelLike of type <code>Solver.Optimizer</code> does not support accessing the attribute MathOptInterface.SolveTimeSec()&quot; if the attribute is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L576-L583">source</a></section></article><h2 id="solver_name"><a class="docs-heading-anchor" href="#solver_name"><code>solver_name</code></a><a id="solver_name-1"></a><a class="docs-heading-anchor-permalink" href="#solver_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.solver_name" href="#JuMP.solver_name"><code>JuMP.solver_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">solver_name(model::GenericModel)</code></pre><p>If available, returns the <code>SolverName</code> property of the underlying optimizer.</p><p>Returns <code>&quot;No optimizer attached&quot;</code> in <code>AUTOMATIC</code> or <code>MANUAL</code> modes when no optimizer is attached.</p><p>Returns <code>&quot;SolverName() attribute not implemented by the optimizer.&quot;</code> if the attribute is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L232-L242">source</a></section></article><h2 id="start_value"><a class="docs-heading-anchor" href="#start_value"><code>start_value</code></a><a id="start_value-1"></a><a class="docs-heading-anchor-permalink" href="#start_value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.start_value" href="#JuMP.start_value"><code>JuMP.start_value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">start_value(con_ref::ConstraintRef)</code></pre><p>Return the primal start value (<a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>) of the constraint <code>con_ref</code>.</p><p>Note: If no primal start value has been set, <code>start_value</code> will return <code>nothing</code>.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L214-L224">source</a></section><section><div><pre><code class="language-julia hljs">start_value(v::GenericVariableRef)</code></pre><p>Return the start value (MOI attribute <code>VariablePrimalStart</code>) of the variable <code>v</code>.</p><p>Note: <code>VariablePrimalStart</code>s are sometimes called &quot;MIP-starts&quot; or &quot;warmstarts&quot;.</p><p>See also <a href="#JuMP.set_start_value"><code>set_start_value</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1639-L1648">source</a></section></article><h2 id="termination_status"><a class="docs-heading-anchor" href="#termination_status"><code>termination_status</code></a><a id="termination_status-1"></a><a class="docs-heading-anchor-permalink" href="#termination_status" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.termination_status" href="#JuMP.termination_status"><code>JuMP.termination_status</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">termination_status(model::GenericModel)</code></pre><p>Return a <a href="../../moi/reference/models/#MathOptInterface.TerminationStatusCode"><code>MOI.TerminationStatusCode</code></a> describing why the solver stopped (i.e., the <a href="../../moi/reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a> attribute).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L497-L502">source</a></section></article><h2 id="time_limit_sec"><a class="docs-heading-anchor" href="#time_limit_sec"><code>time_limit_sec</code></a><a id="time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.time_limit_sec" href="#JuMP.time_limit_sec"><code>JuMP.time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">time_limit_sec(model::GenericModel)</code></pre><p>Return the time limit (in seconds) of the <code>model</code>.</p><p>Returns <code>nothing</code> if unset.</p><p>See also: <a href="#JuMP.set_time_limit_sec"><code>set_time_limit_sec</code></a>, <a href="#JuMP.unset_time_limit_sec"><code>unset_time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L207-L215">source</a></section></article><h2 id="triangle_vec"><a class="docs-heading-anchor" href="#triangle_vec"><code>triangle_vec</code></a><a id="triangle_vec-1"></a><a class="docs-heading-anchor-permalink" href="#triangle_vec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.triangle_vec" href="#JuMP.triangle_vec"><code>JuMP.triangle_vec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">triangle_vec(matrix::Matrix)</code></pre><p>Return the upper triangle of a matrix concatenated into a vector in the order required by JuMP and MathOptInterface for <code>Triangle</code> sets.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, X[1:3, 1:3], Symmetric);
 
@@ -1538,7 +1538,7 @@
 t
 
 julia&gt; @constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(3))
-[t, X[1,1], X[1,2], X[2,2], X[1,3], X[2,3], X[3,3]] ∈ MathOptInterface.RootDetConeTriangle(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L174-L193">source</a></section></article><h2 id="unfix"><a class="docs-heading-anchor" href="#unfix"><code>unfix</code></a><a id="unfix-1"></a><a class="docs-heading-anchor-permalink" href="#unfix" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unfix" href="#JuMP.unfix"><code>JuMP.unfix</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unfix(v::GenericVariableRef)</code></pre><p>Delete the fixing constraint of a variable.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[t, X[1,1], X[1,2], X[2,2], X[1,3], X[2,3], X[3,3]] ∈ MathOptInterface.RootDetConeTriangle(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L174-L193">source</a></section></article><h2 id="unfix"><a class="docs-heading-anchor" href="#unfix"><code>unfix</code></a><a id="unfix-1"></a><a class="docs-heading-anchor-permalink" href="#unfix" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unfix" href="#JuMP.unfix"><code>JuMP.unfix</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unfix(v::GenericVariableRef)</code></pre><p>Delete the fixing constraint of a variable.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.FixRef"><code>FixRef</code></a>, <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x == 1);
 
@@ -1548,7 +1548,7 @@
 julia&gt; unfix(x)
 
 julia&gt; is_fixed(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1140-L1165">source</a></section></article><h2 id="unregister"><a class="docs-heading-anchor" href="#unregister"><code>unregister</code></a><a id="unregister-1"></a><a class="docs-heading-anchor-permalink" href="#unregister" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unregister" href="#JuMP.unregister"><code>JuMP.unregister</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unregister(model::GenericModel, key::Symbol)</code></pre><p>Unregister the name <code>key</code> from <code>model</code> so that a new variable, constraint, or expression can be created with the same key.</p><p>Note that this will not delete the object <code>model[key]</code>; it will just remove the reference at <code>model[key]</code>. To delete the object, use <a href="#JuMP.delete"><code>delete</code></a> as well.</p><p>See also: <a href="#JuMP.delete"><code>delete</code></a>, <a href="#JuMP.object_dictionary"><code>object_dictionary</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1140-L1165">source</a></section></article><h2 id="unregister"><a class="docs-heading-anchor" href="#unregister"><code>unregister</code></a><a id="unregister-1"></a><a class="docs-heading-anchor-permalink" href="#unregister" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unregister" href="#JuMP.unregister"><code>JuMP.unregister</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unregister(model::GenericModel, key::Symbol)</code></pre><p>Unregister the name <code>key</code> from <code>model</code> so that a new variable, constraint, or expression can be created with the same key.</p><p>Note that this will not delete the object <code>model[key]</code>; it will just remove the reference at <code>model[key]</code>. To delete the object, use <a href="#JuMP.delete"><code>delete</code></a> as well.</p><p>See also: <a href="#JuMP.delete"><code>delete</code></a>, <a href="#JuMP.object_dictionary"><code>object_dictionary</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1575,7 +1575,7 @@
 x
 
 julia&gt; num_variables(model)
-2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L857-L900">source</a></section></article><h2 id="unsafe_backend"><a class="docs-heading-anchor" href="#unsafe_backend"><code>unsafe_backend</code></a><a id="unsafe_backend-1"></a><a class="docs-heading-anchor-permalink" href="#unsafe_backend" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unsafe_backend" href="#JuMP.unsafe_backend"><code>JuMP.unsafe_backend</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unsafe_backend(model::GenericModel)</code></pre><p>Return the innermost optimizer associated with the JuMP model <code>model</code>.</p><p><strong>This function should only be used by advanced users looking to access low-level solver-specific functionality. It has a high-risk of incorrect usage. We strongly suggest you use the alternative suggested below.</strong></p><p>See also: <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Unsafe behavior</strong></p><p>This function is unsafe for two main reasons.</p><p>First, the formulation and order of variables and constraints in the unsafe backend may be different to the variables and constraints in <code>model</code>. This can happen because of bridges, or because the solver requires the variables or constraints in a specific order. In addition, the variable or constraint index returned by <a href="#JuMP.index"><code>index</code></a> at the JuMP level may be different to the index of the corresponding variable or constraint in the <code>unsafe_backend</code>. There is no solution to this. Use the alternative suggested below instead.</p><p>Second, the <code>unsafe_backend</code> may be empty, or lack some modifications made to the JuMP model. Thus, before calling <code>unsafe_backend</code> you should first call <a href="../../moi/submodules/Utilities/reference/#MathOptInterface.Utilities.attach_optimizer"><code>MOI.Utilities.attach_optimizer</code></a> to ensure that the backend is synchronized with the JuMP model.</p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
+2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L857-L900">source</a></section></article><h2 id="unsafe_backend"><a class="docs-heading-anchor" href="#unsafe_backend"><code>unsafe_backend</code></a><a id="unsafe_backend-1"></a><a class="docs-heading-anchor-permalink" href="#unsafe_backend" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unsafe_backend" href="#JuMP.unsafe_backend"><code>JuMP.unsafe_backend</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unsafe_backend(model::GenericModel)</code></pre><p>Return the innermost optimizer associated with the JuMP model <code>model</code>.</p><p><strong>This function should only be used by advanced users looking to access low-level solver-specific functionality. It has a high-risk of incorrect usage. We strongly suggest you use the alternative suggested below.</strong></p><p>See also: <a href="#JuMP.backend"><code>backend</code></a>.</p><p><strong>Unsafe behavior</strong></p><p>This function is unsafe for two main reasons.</p><p>First, the formulation and order of variables and constraints in the unsafe backend may be different to the variables and constraints in <code>model</code>. This can happen because of bridges, or because the solver requires the variables or constraints in a specific order. In addition, the variable or constraint index returned by <a href="#JuMP.index"><code>index</code></a> at the JuMP level may be different to the index of the corresponding variable or constraint in the <code>unsafe_backend</code>. There is no solution to this. Use the alternative suggested below instead.</p><p>Second, the <code>unsafe_backend</code> may be empty, or lack some modifications made to the JuMP model. Thus, before calling <code>unsafe_backend</code> you should first call <a href="../../moi/submodules/Utilities/reference/#MathOptInterface.Utilities.attach_optimizer"><code>MOI.Utilities.attach_optimizer</code></a> to ensure that the backend is synchronized with the JuMP model.</p><pre><code class="language-julia-repl hljs">julia&gt; import HiGHS
 
 julia&gt; model = Model(HiGHS.Optimizer)
 A JuMP Model
@@ -1610,7 +1610,7 @@
 x
 
 julia&gt; highs = backend(model)  # No need to call `attach_optimizer`.
-A HiGHS model with 1 columns and 0 rows.</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L412-L503">source</a></section></article><h2 id="unset_binary"><a class="docs-heading-anchor" href="#unset_binary"><code>unset_binary</code></a><a id="unset_binary-1"></a><a class="docs-heading-anchor-permalink" href="#unset_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_binary" href="#JuMP.unset_binary"><code>JuMP.unset_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_binary(variable_ref::GenericVariableRef)</code></pre><p>Remove the binary constraint on the variable <code>variable_ref</code>.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+A HiGHS model with 1 columns and 0 rows.</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L412-L503">source</a></section></article><h2 id="unset_binary"><a class="docs-heading-anchor" href="#unset_binary"><code>unset_binary</code></a><a id="unset_binary-1"></a><a class="docs-heading-anchor-permalink" href="#unset_binary" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_binary" href="#JuMP.unset_binary"><code>JuMP.unset_binary</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_binary(variable_ref::GenericVariableRef)</code></pre><p>Remove the binary constraint on the variable <code>variable_ref</code>.</p><p>See also <a href="#JuMP.BinaryRef"><code>BinaryRef</code></a>, <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Bin);
 
@@ -1620,7 +1620,7 @@
 julia&gt; unset_binary(x)
 
 julia&gt; is_binary(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1437-L1459">source</a></section></article><h2 id="unset_integer"><a class="docs-heading-anchor" href="#unset_integer"><code>unset_integer</code></a><a id="unset_integer-1"></a><a class="docs-heading-anchor-permalink" href="#unset_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_integer" href="#JuMP.unset_integer"><code>JuMP.unset_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_integer(variable_ref::GenericVariableRef)</code></pre><p>Remove the integrality constraint on the variable <code>variable_ref</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1437-L1459">source</a></section></article><h2 id="unset_integer"><a class="docs-heading-anchor" href="#unset_integer"><code>unset_integer</code></a><a id="unset_integer-1"></a><a class="docs-heading-anchor-permalink" href="#unset_integer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_integer" href="#JuMP.unset_integer"><code>JuMP.unset_integer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_integer(variable_ref::GenericVariableRef)</code></pre><p>Remove the integrality constraint on the variable <code>variable_ref</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.IntegerRef"><code>IntegerRef</code></a>, <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Int);
 
@@ -1630,18 +1630,18 @@
 julia&gt; unset_integer(x)
 
 julia&gt; is_integer(x)
-false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1304-L1328">source</a></section></article><h2 id="unset_silent"><a class="docs-heading-anchor" href="#unset_silent"><code>unset_silent</code></a><a id="unset_silent-1"></a><a class="docs-heading-anchor-permalink" href="#unset_silent" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_silent" href="#JuMP.unset_silent"><code>JuMP.unset_silent</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_silent(model::GenericModel)</code></pre><p>Neutralize the effect of the <code>set_silent</code> function and let the solver attributes control the verbosity.</p><p>See also: <a href="#JuMP.set_silent"><code>set_silent</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L166-L173">source</a></section></article><h2 id="unset_time_limit_sec"><a class="docs-heading-anchor" href="#unset_time_limit_sec"><code>unset_time_limit_sec</code></a><a id="unset_time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#unset_time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_time_limit_sec" href="#JuMP.unset_time_limit_sec"><code>JuMP.unset_time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_time_limit_sec(model::GenericModel)</code></pre><p>Unset the time limit of the solver.</p><p>See also: <a href="#JuMP.set_time_limit_sec"><code>set_time_limit_sec</code></a>, <a href="#JuMP.time_limit_sec"><code>time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L196-L202">source</a></section></article><h2 id="upper_bound"><a class="docs-heading-anchor" href="#upper_bound"><code>upper_bound</code></a><a id="upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.upper_bound" href="#JuMP.upper_bound"><code>JuMP.upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">upper_bound(v::GenericVariableRef)</code></pre><p>Return the upper bound of a variable.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+false</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1304-L1328">source</a></section></article><h2 id="unset_silent"><a class="docs-heading-anchor" href="#unset_silent"><code>unset_silent</code></a><a id="unset_silent-1"></a><a class="docs-heading-anchor-permalink" href="#unset_silent" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_silent" href="#JuMP.unset_silent"><code>JuMP.unset_silent</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_silent(model::GenericModel)</code></pre><p>Neutralize the effect of the <code>set_silent</code> function and let the solver attributes control the verbosity.</p><p>See also: <a href="#JuMP.set_silent"><code>set_silent</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L166-L173">source</a></section></article><h2 id="unset_time_limit_sec"><a class="docs-heading-anchor" href="#unset_time_limit_sec"><code>unset_time_limit_sec</code></a><a id="unset_time_limit_sec-1"></a><a class="docs-heading-anchor-permalink" href="#unset_time_limit_sec" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.unset_time_limit_sec" href="#JuMP.unset_time_limit_sec"><code>JuMP.unset_time_limit_sec</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">unset_time_limit_sec(model::GenericModel)</code></pre><p>Unset the time limit of the solver.</p><p>See also: <a href="#JuMP.set_time_limit_sec"><code>set_time_limit_sec</code></a>, <a href="#JuMP.time_limit_sec"><code>time_limit_sec</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L196-L202">source</a></section></article><h2 id="upper_bound"><a class="docs-heading-anchor" href="#upper_bound"><code>upper_bound</code></a><a id="upper_bound-1"></a><a class="docs-heading-anchor-permalink" href="#upper_bound" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.upper_bound" href="#JuMP.upper_bound"><code>JuMP.upper_bound</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">upper_bound(v::GenericVariableRef)</code></pre><p>Return the upper bound of a variable.</p><p>Error if one does not exist.</p><p>See also <a href="#JuMP.UpperBoundRef"><code>UpperBoundRef</code></a>, <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &lt;= 1.0);
 
 julia&gt; upper_bound(x)
-1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L989-L1009">source</a></section></article><h2 id="value"><a class="docs-heading-anchor" href="#value"><code>value</code></a><a id="value-1"></a><a class="docs-heading-anchor-permalink" href="#value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.value" href="#JuMP.value"><code>JuMP.value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">value(con_ref::ConstraintRef; result::Int = 1)</code></pre><p>Return the primal value of constraint <code>con_ref</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>That is, if <code>con_ref</code> is the reference of a constraint <code>func</code>-in-<code>set</code>, it returns the value of <code>func</code> evaluated at the value of the variables (given by <a href="#JuMP.value"><code>value(::GenericVariableRef)</code></a>).</p><p>Use <a href="#JuMP.has_values"><code>has_values</code></a> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p><p><strong>Note</strong></p><p>For scalar constraints, the constant is moved to the <code>set</code> so it is not taken into account in the primal value of the constraint. For instance, the constraint <code>@constraint(model, 2x + 3y + 1 == 5)</code> is transformed into <code>2x + 3y</code>-in-<code>MOI.EqualTo(4)</code> so the value returned by this function is the evaluation of <code>2x + 3y</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L978-L999">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, con_ref::ConstraintRef)</code></pre><p>Evaluate the primal value of the constraint <code>con_ref</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1007-L1012">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericVariableRef; result = 1)</code></pre><p>Return the value of variable <code>v</code> associated with result index <code>result</code> of the most-recent returned by the solver.</p><p>Use <a href="#JuMP.has_values"><code>has_values</code></a> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1690-L1699">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, v::GenericVariableRef)</code></pre><p>Evaluate the value of the variable <code>v</code> as <code>var_value(v)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1704-L1708">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::GenericAffExpr)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L346-L350">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericAffExpr; result::Int = 1)</code></pre><p>Return the value of the <code>GenericAffExpr</code> <code>v</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L609-L616">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::GenericQuadExpr)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L717-L721">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericQuadExpr; result::Int = 1)</code></pre><p>Return the value of the <code>GenericQuadExpr</code> <code>v</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Replaces <code>getvalue</code> for most use cases.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L738-L747">source</a></section><section><div><pre><code class="language-julia hljs">value(p::NonlinearParameter)</code></pre><p>Return the current value stored in the nonlinear parameter <code>p</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+1.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L989-L1009">source</a></section></article><h2 id="value"><a class="docs-heading-anchor" href="#value"><code>value</code></a><a id="value-1"></a><a class="docs-heading-anchor-permalink" href="#value" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.value" href="#JuMP.value"><code>JuMP.value</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">value(con_ref::ConstraintRef; result::Int = 1)</code></pre><p>Return the primal value of constraint <code>con_ref</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>That is, if <code>con_ref</code> is the reference of a constraint <code>func</code>-in-<code>set</code>, it returns the value of <code>func</code> evaluated at the value of the variables (given by <a href="#JuMP.value"><code>value(::GenericVariableRef)</code></a>).</p><p>Use <a href="#JuMP.has_values"><code>has_values</code></a> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p><p><strong>Note</strong></p><p>For scalar constraints, the constant is moved to the <code>set</code> so it is not taken into account in the primal value of the constraint. For instance, the constraint <code>@constraint(model, 2x + 3y + 1 == 5)</code> is transformed into <code>2x + 3y</code>-in-<code>MOI.EqualTo(4)</code> so the value returned by this function is the evaluation of <code>2x + 3y</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L978-L999">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, con_ref::ConstraintRef)</code></pre><p>Evaluate the primal value of the constraint <code>con_ref</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1007-L1012">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericVariableRef; result = 1)</code></pre><p>Return the value of variable <code>v</code> associated with result index <code>result</code> of the most-recent returned by the solver.</p><p>Use <a href="#JuMP.has_values"><code>has_values</code></a> to check if a result exists before asking for values.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1690-L1699">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, v::GenericVariableRef)</code></pre><p>Evaluate the value of the variable <code>v</code> as <code>var_value(v)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1704-L1708">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::GenericAffExpr)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L346-L350">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericAffExpr; result::Int = 1)</code></pre><p>Return the value of the <code>GenericAffExpr</code> <code>v</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L609-L616">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::GenericQuadExpr)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L717-L721">source</a></section><section><div><pre><code class="language-julia hljs">value(v::GenericQuadExpr; result::Int = 1)</code></pre><p>Return the value of the <code>GenericQuadExpr</code> <code>v</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Replaces <code>getvalue</code> for most use cases.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L738-L747">source</a></section><section><div><pre><code class="language-julia hljs">value(p::NonlinearParameter)</code></pre><p>Return the current value stored in the nonlinear parameter <code>p</code>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @NLparameter(model, p == 10)
 p == 10.0
 
 julia&gt; value(p)
-10.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L228-L244">source</a></section><section><div><pre><code class="language-julia hljs">value(ex::NonlinearExpression; result::Int = 1)</code></pre><p>Return the value of the <code>NonlinearExpression</code> <code>ex</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Replaces <code>getvalue</code> for most use cases.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L338-L347">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::NonlinearExpression)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L378-L382">source</a></section><section><div><pre><code class="language-julia hljs">value(c::NonlinearConstraintRef; result::Int = 1)</code></pre><p>Return the value of the <code>NonlinearConstraintRef</code> <code>c</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L529-L536">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, c::NonlinearConstraintRef)</code></pre><p>Evaluate <code>c</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L543-L547">source</a></section></article><h2 id="value_type"><a class="docs-heading-anchor" href="#value_type"><code>value_type</code></a><a id="value_type-1"></a><a class="docs-heading-anchor-permalink" href="#value_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.value_type" href="#JuMP.value_type"><code>JuMP.value_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">value_type(::Type{&lt;:Union{AbstractModel,AbstractVariableRef}})</code></pre><p>Return the return type of <a href="#JuMP.value"><code>value</code></a> for variables of that model. It defaults to <code>Float64</code> if it is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L77-L82">source</a></section></article><h2 id="variable_by_name"><a class="docs-heading-anchor" href="#variable_by_name"><code>variable_by_name</code></a><a id="variable_by_name-1"></a><a class="docs-heading-anchor-permalink" href="#variable_by_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.variable_by_name" href="#JuMP.variable_by_name"><code>JuMP.variable_by_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">variable_by_name(
+10.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L228-L244">source</a></section><section><div><pre><code class="language-julia hljs">value(ex::NonlinearExpression; result::Int = 1)</code></pre><p>Return the value of the <code>NonlinearExpression</code> <code>ex</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>Replaces <code>getvalue</code> for most use cases.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L338-L347">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, ex::NonlinearExpression)</code></pre><p>Evaluate <code>ex</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L378-L382">source</a></section><section><div><pre><code class="language-julia hljs">value(c::NonlinearConstraintRef; result::Int = 1)</code></pre><p>Return the value of the <code>NonlinearConstraintRef</code> <code>c</code> associated with result index <code>result</code> of the most-recent solution returned by the solver.</p><p>See also: <a href="#JuMP.result_count"><code>result_count</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L529-L536">source</a></section><section><div><pre><code class="language-julia hljs">value(var_value::Function, c::NonlinearConstraintRef)</code></pre><p>Evaluate <code>c</code> using <code>var_value(v)</code> as the value for each variable <code>v</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L543-L547">source</a></section></article><h2 id="value_type"><a class="docs-heading-anchor" href="#value_type"><code>value_type</code></a><a id="value_type-1"></a><a class="docs-heading-anchor-permalink" href="#value_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.value_type" href="#JuMP.value_type"><code>JuMP.value_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">value_type(::Type{&lt;:Union{AbstractModel,AbstractVariableRef}})</code></pre><p>Return the return type of <a href="#JuMP.value"><code>value</code></a> for variables of that model. It defaults to <code>Float64</code> if it is not implemented.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L77-L82">source</a></section></article><h2 id="variable_by_name"><a class="docs-heading-anchor" href="#variable_by_name"><code>variable_by_name</code></a><a id="variable_by_name-1"></a><a class="docs-heading-anchor-permalink" href="#variable_by_name" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.variable_by_name" href="#JuMP.variable_by_name"><code>JuMP.variable_by_name</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">variable_by_name(
     model::AbstractModel,
     name::String,
 )::Union{AbstractVariableRef,Nothing}</code></pre><p>Returns the reference of the variable with name attribute <code>name</code> or <code>Nothing</code> if no variable has this name attribute. Throws an error if several variables have <code>name</code> as their name attribute.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
@@ -1684,17 +1684,17 @@
  u[2]
 
 julia&gt; variable_by_name(model, &quot;u[2]&quot;)
-u[2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L552-L607">source</a></section></article><h2 id="variable_ref_type"><a class="docs-heading-anchor" href="#variable_ref_type"><code>variable_ref_type</code></a><a id="variable_ref_type-1"></a><a class="docs-heading-anchor-permalink" href="#variable_ref_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.variable_ref_type" href="#JuMP.variable_ref_type"><code>JuMP.variable_ref_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">variable_ref_type(::Union{F,Type{F}}) where {F}</code></pre><p>A helper function used internally by JuMP and some JuMP extensions. Returns the variable type associated with the model or expression type <code>F</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L219-L224">source</a></section></article><h2 id="vectorize"><a class="docs-heading-anchor" href="#vectorize"><code>vectorize</code></a><a id="vectorize-1"></a><a class="docs-heading-anchor-permalink" href="#vectorize" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.vectorize" href="#JuMP.vectorize"><code>JuMP.vectorize</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">vectorize(matrix::AbstractMatrix, ::Shape)</code></pre><p>Convert the <code>matrix</code> into a vector according to <code>Shape</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L11-L15">source</a></section></article><h2 id="write_to_file"><a class="docs-heading-anchor" href="#write_to_file"><code>write_to_file</code></a><a id="write_to_file-1"></a><a class="docs-heading-anchor-permalink" href="#write_to_file" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.write_to_file" href="#JuMP.write_to_file"><code>JuMP.write_to_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">write_to_file(
+u[2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L552-L607">source</a></section></article><h2 id="variable_ref_type"><a class="docs-heading-anchor" href="#variable_ref_type"><code>variable_ref_type</code></a><a id="variable_ref_type-1"></a><a class="docs-heading-anchor-permalink" href="#variable_ref_type" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.variable_ref_type" href="#JuMP.variable_ref_type"><code>JuMP.variable_ref_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">variable_ref_type(::Union{F,Type{F}}) where {F}</code></pre><p>A helper function used internally by JuMP and some JuMP extensions. Returns the variable type associated with the model or expression type <code>F</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L219-L224">source</a></section></article><h2 id="vectorize"><a class="docs-heading-anchor" href="#vectorize"><code>vectorize</code></a><a id="vectorize-1"></a><a class="docs-heading-anchor-permalink" href="#vectorize" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.vectorize" href="#JuMP.vectorize"><code>JuMP.vectorize</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">vectorize(matrix::AbstractMatrix, ::Shape)</code></pre><p>Convert the <code>matrix</code> into a vector according to <code>Shape</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L11-L15">source</a></section></article><h2 id="write_to_file"><a class="docs-heading-anchor" href="#write_to_file"><code>write_to_file</code></a><a id="write_to_file-1"></a><a class="docs-heading-anchor-permalink" href="#write_to_file" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.write_to_file" href="#JuMP.write_to_file"><code>JuMP.write_to_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">write_to_file(
     model::GenericModel,
     filename::String;
     format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,
     kwargs...,
-)</code></pre><p>Write the JuMP model <code>model</code> to <code>filename</code> in the format <code>format</code>.</p><p>If the filename ends in <code>.gz</code>, it will be compressed using Gzip. If the filename ends in <code>.bz2</code>, it will be compressed using BZip2.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/file_formats.jl#L31-L45">source</a></section></article><h2 id="AbstractConstraint"><a class="docs-heading-anchor" href="#AbstractConstraint"><code>AbstractConstraint</code></a><a id="AbstractConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractConstraint" href="#JuMP.AbstractConstraint"><code>JuMP.AbstractConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">abstract type AbstractConstraint</code></pre><p>An abstract base type for all constraint types. <code>AbstractConstraint</code>s store the function and set directly, unlike <a href="#JuMP.ConstraintRef"><code>ConstraintRef</code></a>s that are merely references to constraints stored in a model. <code>AbstractConstraint</code>s do not need to be attached to a model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L448-L455">source</a></section></article><h2 id="AbstractJuMPScalar"><a class="docs-heading-anchor" href="#AbstractJuMPScalar"><code>AbstractJuMPScalar</code></a><a id="AbstractJuMPScalar-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractJuMPScalar" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractJuMPScalar" href="#JuMP.AbstractJuMPScalar"><code>JuMP.AbstractJuMPScalar</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractJuMPScalar &lt;: MutableArithmetics.AbstractMutable</code></pre><p>Abstract base type for all scalar types</p><p>The subtyping of <code>AbstractMutable</code> will allow calls of some <code>Base</code> functions to be redirected to a method in MA that handles type promotion more carefully (e.g. the promotion in sparse matrix products in SparseArrays usually does not work for JuMP types) and exploits the mutability of <code>AffExpr</code> and <code>QuadExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L991-L1000">source</a></section></article><h2 id="AbstractModel"><a class="docs-heading-anchor" href="#AbstractModel"><code>AbstractModel</code></a><a id="AbstractModel-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractModel" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractModel" href="#JuMP.AbstractModel"><code>JuMP.AbstractModel</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractModel</code></pre><p>An abstract type that should be subtyped for users creating JuMP extensions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L70-L74">source</a></section></article><h2 id="AbstractScalarSet"><a class="docs-heading-anchor" href="#AbstractScalarSet"><code>AbstractScalarSet</code></a><a id="AbstractScalarSet-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractScalarSet" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractScalarSet" href="#JuMP.AbstractScalarSet"><code>JuMP.AbstractScalarSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractScalarSet</code></pre><p>An abstract type for defining new scalar sets in JuMP.</p><p>Implement <code>moi_set(::AbstractScalarSet)</code> to convert the type into an MOI set.</p><p>See also: <a href="#JuMP.moi_set"><code>moi_set</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L249-L257">source</a></section></article><h2 id="AbstractShape"><a class="docs-heading-anchor" href="#AbstractShape"><code>AbstractShape</code></a><a id="AbstractShape-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractShape" href="#JuMP.AbstractShape"><code>JuMP.AbstractShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractShape</code></pre><p>Abstract vectorizable shape. Given a flat vector form of an object of shape <code>shape</code>, the original object can be obtained by <a href="#JuMP.reshape_vector"><code>reshape_vector</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L11-L16">source</a></section></article><h2 id="AbstractVariable"><a class="docs-heading-anchor" href="#AbstractVariable"><code>AbstractVariable</code></a><a id="AbstractVariable-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVariable" href="#JuMP.AbstractVariable"><code>JuMP.AbstractVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariable</code></pre><p>Variable returned by <a href="#JuMP.build_variable"><code>build_variable</code></a>. It represents a variable that has not been added yet to any model. It can be added to a given <code>model</code> with <a href="#JuMP.add_variable"><code>add_variable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L15-L21">source</a></section></article><h2 id="AbstractVariableRef"><a class="docs-heading-anchor" href="#AbstractVariableRef"><code>AbstractVariableRef</code></a><a id="AbstractVariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVariableRef" href="#JuMP.AbstractVariableRef"><code>JuMP.AbstractVariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariableRef</code></pre><p>Variable returned by <a href="#JuMP.add_variable"><code>add_variable</code></a>. Affine (resp. quadratic) operations with variables of type <code>V&lt;:AbstractVariableRef</code> and coefficients of type <code>T</code>     create a <code>GenericAffExpr{T,V}</code> (resp. <code>GenericQuadExpr{T,V}</code>).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L210-L216">source</a></section></article><h2 id="AbstractVectorSet"><a class="docs-heading-anchor" href="#AbstractVectorSet"><code>AbstractVectorSet</code></a><a id="AbstractVectorSet-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVectorSet" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVectorSet" href="#JuMP.AbstractVectorSet"><code>JuMP.AbstractVectorSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVectorSet</code></pre><p>An abstract type for defining new sets in JuMP.</p><p>Implement <code>moi_set(::AbstractVectorSet, dim::Int)</code> to convert the type into an MOI set.</p><p>See also: <a href="#JuMP.moi_set"><code>moi_set</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L11-L20">source</a></section></article><h2 id="AffExpr"><a class="docs-heading-anchor" href="#AffExpr"><code>AffExpr</code></a><a id="AffExpr-1"></a><a class="docs-heading-anchor-permalink" href="#AffExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AffExpr" href="#JuMP.AffExpr"><code>JuMP.AffExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AffExpr</code></pre><p>Alias for <code>GenericAffExpr{Float64,VariableRef}</code>, the specific <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L585-L590">source</a></section></article><h2 id="BinaryRef"><a class="docs-heading-anchor" href="#BinaryRef"><code>BinaryRef</code></a><a id="BinaryRef-1"></a><a class="docs-heading-anchor-permalink" href="#BinaryRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.BinaryRef" href="#JuMP.BinaryRef"><code>JuMP.BinaryRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">BinaryRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>v</code> to be binary. Errors if one does not exist.</p><p>See also <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+)</code></pre><p>Write the JuMP model <code>model</code> to <code>filename</code> in the format <code>format</code>.</p><p>If the filename ends in <code>.gz</code>, it will be compressed using Gzip. If the filename ends in <code>.bz2</code>, it will be compressed using BZip2.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/file_formats.jl#L31-L45">source</a></section></article><h2 id="AbstractConstraint"><a class="docs-heading-anchor" href="#AbstractConstraint"><code>AbstractConstraint</code></a><a id="AbstractConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractConstraint" href="#JuMP.AbstractConstraint"><code>JuMP.AbstractConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">abstract type AbstractConstraint</code></pre><p>An abstract base type for all constraint types. <code>AbstractConstraint</code>s store the function and set directly, unlike <a href="#JuMP.ConstraintRef"><code>ConstraintRef</code></a>s that are merely references to constraints stored in a model. <code>AbstractConstraint</code>s do not need to be attached to a model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L448-L455">source</a></section></article><h2 id="AbstractJuMPScalar"><a class="docs-heading-anchor" href="#AbstractJuMPScalar"><code>AbstractJuMPScalar</code></a><a id="AbstractJuMPScalar-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractJuMPScalar" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractJuMPScalar" href="#JuMP.AbstractJuMPScalar"><code>JuMP.AbstractJuMPScalar</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractJuMPScalar &lt;: MutableArithmetics.AbstractMutable</code></pre><p>Abstract base type for all scalar types</p><p>The subtyping of <code>AbstractMutable</code> will allow calls of some <code>Base</code> functions to be redirected to a method in MA that handles type promotion more carefully (e.g. the promotion in sparse matrix products in SparseArrays usually does not work for JuMP types) and exploits the mutability of <code>AffExpr</code> and <code>QuadExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L991-L1000">source</a></section></article><h2 id="AbstractModel"><a class="docs-heading-anchor" href="#AbstractModel"><code>AbstractModel</code></a><a id="AbstractModel-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractModel" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractModel" href="#JuMP.AbstractModel"><code>JuMP.AbstractModel</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractModel</code></pre><p>An abstract type that should be subtyped for users creating JuMP extensions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L70-L74">source</a></section></article><h2 id="AbstractScalarSet"><a class="docs-heading-anchor" href="#AbstractScalarSet"><code>AbstractScalarSet</code></a><a id="AbstractScalarSet-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractScalarSet" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractScalarSet" href="#JuMP.AbstractScalarSet"><code>JuMP.AbstractScalarSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractScalarSet</code></pre><p>An abstract type for defining new scalar sets in JuMP.</p><p>Implement <code>moi_set(::AbstractScalarSet)</code> to convert the type into an MOI set.</p><p>See also: <a href="#JuMP.moi_set"><code>moi_set</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L249-L257">source</a></section></article><h2 id="AbstractShape"><a class="docs-heading-anchor" href="#AbstractShape"><code>AbstractShape</code></a><a id="AbstractShape-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractShape" href="#JuMP.AbstractShape"><code>JuMP.AbstractShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractShape</code></pre><p>Abstract vectorizable shape. Given a flat vector form of an object of shape <code>shape</code>, the original object can be obtained by <a href="#JuMP.reshape_vector"><code>reshape_vector</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L11-L16">source</a></section></article><h2 id="AbstractVariable"><a class="docs-heading-anchor" href="#AbstractVariable"><code>AbstractVariable</code></a><a id="AbstractVariable-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVariable" href="#JuMP.AbstractVariable"><code>JuMP.AbstractVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariable</code></pre><p>Variable returned by <a href="#JuMP.build_variable"><code>build_variable</code></a>. It represents a variable that has not been added yet to any model. It can be added to a given <code>model</code> with <a href="#JuMP.add_variable"><code>add_variable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L15-L21">source</a></section></article><h2 id="AbstractVariableRef"><a class="docs-heading-anchor" href="#AbstractVariableRef"><code>AbstractVariableRef</code></a><a id="AbstractVariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVariableRef" href="#JuMP.AbstractVariableRef"><code>JuMP.AbstractVariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariableRef</code></pre><p>Variable returned by <a href="#JuMP.add_variable"><code>add_variable</code></a>. Affine (resp. quadratic) operations with variables of type <code>V&lt;:AbstractVariableRef</code> and coefficients of type <code>T</code>     create a <code>GenericAffExpr{T,V}</code> (resp. <code>GenericQuadExpr{T,V}</code>).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L210-L216">source</a></section></article><h2 id="AbstractVectorSet"><a class="docs-heading-anchor" href="#AbstractVectorSet"><code>AbstractVectorSet</code></a><a id="AbstractVectorSet-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractVectorSet" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AbstractVectorSet" href="#JuMP.AbstractVectorSet"><code>JuMP.AbstractVectorSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVectorSet</code></pre><p>An abstract type for defining new sets in JuMP.</p><p>Implement <code>moi_set(::AbstractVectorSet, dim::Int)</code> to convert the type into an MOI set.</p><p>See also: <a href="#JuMP.moi_set"><code>moi_set</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L11-L20">source</a></section></article><h2 id="AffExpr"><a class="docs-heading-anchor" href="#AffExpr"><code>AffExpr</code></a><a id="AffExpr-1"></a><a class="docs-heading-anchor-permalink" href="#AffExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AffExpr" href="#JuMP.AffExpr"><code>JuMP.AffExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AffExpr</code></pre><p>Alias for <code>GenericAffExpr{Float64,VariableRef}</code>, the specific <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L585-L590">source</a></section></article><h2 id="BinaryRef"><a class="docs-heading-anchor" href="#BinaryRef"><code>BinaryRef</code></a><a id="BinaryRef-1"></a><a class="docs-heading-anchor-permalink" href="#BinaryRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.BinaryRef" href="#JuMP.BinaryRef"><code>JuMP.BinaryRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">BinaryRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>v</code> to be binary. Errors if one does not exist.</p><p>See also <a href="#JuMP.is_binary"><code>is_binary</code></a>, <a href="#JuMP.set_binary"><code>set_binary</code></a>, <a href="#JuMP.unset_binary"><code>unset_binary</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Bin);
 
 julia&gt; BinaryRef(x)
-x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1465-L1483">source</a></section></article><h2 id="BridgeableConstraint"><a class="docs-heading-anchor" href="#BridgeableConstraint"><code>BridgeableConstraint</code></a><a id="BridgeableConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#BridgeableConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.BridgeableConstraint" href="#JuMP.BridgeableConstraint"><code>JuMP.BridgeableConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BridgeableConstraint(
+x binary</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1465-L1483">source</a></section></article><h2 id="BridgeableConstraint"><a class="docs-heading-anchor" href="#BridgeableConstraint"><code>BridgeableConstraint</code></a><a id="BridgeableConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#BridgeableConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.BridgeableConstraint" href="#JuMP.BridgeableConstraint"><code>JuMP.BridgeableConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BridgeableConstraint(
     constraint::C,
     bridge_type::B;
     coefficient_type::Type{T} = Float64,
@@ -1709,7 +1709,7 @@
 )
     constraint = ScalarConstraint(func, set)
     return BridgeableConstraint(constraint, CustomBridge)
-end</code></pre><p><strong>Note</strong></p><p>JuMP extensions should extend <code>JuMP.build_constraint</code> only if they also defined <code>CustomSet</code>, for three reasons:</p><ol><li>It is problematic if multiple extensions overload the same JuMP method.</li><li>A missing method will not inform the users that they forgot to load the extension module defining the <code>build_constraint</code> method.</li><li>Defining a method where neither the function nor any of the argument types are defined in the package is called <a href="https://docs.julialang.org/en/v1/manual/style-guide/index.html#Avoid-type-piracy-1"><em>type piracy</em></a> and is discouraged in the Julia style guide.</li></ol></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L458-L512">source</a></section></article><h2 id="ComplexPlane"><a class="docs-heading-anchor" href="#ComplexPlane"><code>ComplexPlane</code></a><a id="ComplexPlane-1"></a><a class="docs-heading-anchor-permalink" href="#ComplexPlane" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ComplexPlane" href="#JuMP.ComplexPlane"><code>JuMP.ComplexPlane</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ComplexPlane</code></pre><p>Complex plane object that can be used to create a complex variable in the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+end</code></pre><p><strong>Note</strong></p><p>JuMP extensions should extend <code>JuMP.build_constraint</code> only if they also defined <code>CustomSet</code>, for three reasons:</p><ol><li>It is problematic if multiple extensions overload the same JuMP method.</li><li>A missing method will not inform the users that they forgot to load the extension module defining the <code>build_constraint</code> method.</li><li>Defining a method where neither the function nor any of the argument types are defined in the package is called <a href="https://docs.julialang.org/en/v1/manual/style-guide/index.html#Avoid-type-piracy-1"><em>type piracy</em></a> and is discouraged in the Julia style guide.</li></ol></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L458-L512">source</a></section></article><h2 id="ComplexPlane"><a class="docs-heading-anchor" href="#ComplexPlane"><code>ComplexPlane</code></a><a id="ComplexPlane-1"></a><a class="docs-heading-anchor-permalink" href="#ComplexPlane" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ComplexPlane" href="#JuMP.ComplexPlane"><code>JuMP.ComplexPlane</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ComplexPlane</code></pre><p>Complex plane object that can be used to create a complex variable in the <a href="#JuMP.@variable"><code>@variable</code></a> macro.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x in ComplexPlane())
 real(x) + imag(x) im
@@ -1717,23 +1717,23 @@
 julia&gt; all_variables(model)
 2-element Vector{VariableRef}:
  real(x)
- imag(x)</code></pre><p>We see in the output of the last command that two real variables were created. The Julia variable <code>x</code> binds to an affine expression in terms of these two variables that parametrize the complex plane.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1973-L1998">source</a></section></article><h2 id="ComplexVariable"><a class="docs-heading-anchor" href="#ComplexVariable"><code>ComplexVariable</code></a><a id="ComplexVariable-1"></a><a class="docs-heading-anchor-permalink" href="#ComplexVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ComplexVariable" href="#JuMP.ComplexVariable"><code>JuMP.ComplexVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ComplexVariable{S,T,U,V} &lt;: AbstractVariable</code></pre><p>A struct used when adding complex variables.</p><p>See also: <a href="#JuMP.ComplexPlane"><code>ComplexPlane</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L2001-L2007">source</a></section></article><h2 id="ConstraintNotOwned"><a class="docs-heading-anchor" href="#ConstraintNotOwned"><code>ConstraintNotOwned</code></a><a id="ConstraintNotOwned-1"></a><a class="docs-heading-anchor-permalink" href="#ConstraintNotOwned" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ConstraintNotOwned" href="#JuMP.ConstraintNotOwned"><code>JuMP.ConstraintNotOwned</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct ConstraintNotOwned{C &lt;: ConstraintRef} &lt;: Exception
+ imag(x)</code></pre><p>We see in the output of the last command that two real variables were created. The Julia variable <code>x</code> binds to an affine expression in terms of these two variables that parametrize the complex plane.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1973-L1998">source</a></section></article><h2 id="ComplexVariable"><a class="docs-heading-anchor" href="#ComplexVariable"><code>ComplexVariable</code></a><a id="ComplexVariable-1"></a><a class="docs-heading-anchor-permalink" href="#ComplexVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ComplexVariable" href="#JuMP.ComplexVariable"><code>JuMP.ComplexVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ComplexVariable{S,T,U,V} &lt;: AbstractVariable</code></pre><p>A struct used when adding complex variables.</p><p>See also: <a href="#JuMP.ComplexPlane"><code>ComplexPlane</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L2001-L2007">source</a></section></article><h2 id="ConstraintNotOwned"><a class="docs-heading-anchor" href="#ConstraintNotOwned"><code>ConstraintNotOwned</code></a><a id="ConstraintNotOwned-1"></a><a class="docs-heading-anchor-permalink" href="#ConstraintNotOwned" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ConstraintNotOwned" href="#JuMP.ConstraintNotOwned"><code>JuMP.ConstraintNotOwned</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct ConstraintNotOwned{C &lt;: ConstraintRef} &lt;: Exception
     constraint_ref::C
-end</code></pre><p>The constraint <code>constraint_ref</code> was used in a model different to <code>owner_model(constraint_ref)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L39-L46">source</a></section></article><h2 id="ConstraintRef"><a class="docs-heading-anchor" href="#ConstraintRef"><code>ConstraintRef</code></a><a id="ConstraintRef-1"></a><a class="docs-heading-anchor-permalink" href="#ConstraintRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ConstraintRef" href="#JuMP.ConstraintRef"><code>JuMP.ConstraintRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintRef</code></pre><p>Holds a reference to the model and the corresponding MOI.ConstraintIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L11-L15">source</a></section></article><h2 id="FixRef"><a class="docs-heading-anchor" href="#FixRef"><code>FixRef</code></a><a id="FixRef-1"></a><a class="docs-heading-anchor-permalink" href="#FixRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FixRef" href="#JuMP.FixRef"><code>JuMP.FixRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">FixRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint fixing the value of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+end</code></pre><p>The constraint <code>constraint_ref</code> was used in a model different to <code>owner_model(constraint_ref)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L39-L46">source</a></section></article><h2 id="ConstraintRef"><a class="docs-heading-anchor" href="#ConstraintRef"><code>ConstraintRef</code></a><a id="ConstraintRef-1"></a><a class="docs-heading-anchor-permalink" href="#ConstraintRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ConstraintRef" href="#JuMP.ConstraintRef"><code>JuMP.ConstraintRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintRef</code></pre><p>Holds a reference to the model and the corresponding MOI.ConstraintIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L11-L15">source</a></section></article><h2 id="FixRef"><a class="docs-heading-anchor" href="#FixRef"><code>FixRef</code></a><a id="FixRef-1"></a><a class="docs-heading-anchor-permalink" href="#FixRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FixRef" href="#JuMP.FixRef"><code>JuMP.FixRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">FixRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint fixing the value of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_fixed"><code>is_fixed</code></a>, <a href="#JuMP.fix_value"><code>fix_value</code></a>, <a href="#JuMP.fix"><code>fix</code></a>, <a href="#JuMP.unfix"><code>unfix</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x == 1);
 
 julia&gt; FixRef(x)
-x = 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1196-L1216">source</a></section></article><h2 id="GenericAffExpr"><a class="docs-heading-anchor" href="#GenericAffExpr"><code>GenericAffExpr</code></a><a id="GenericAffExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericAffExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericAffExpr" href="#JuMP.GenericAffExpr"><code>JuMP.GenericAffExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">mutable struct GenericAffExpr{CoefType,VarType} &lt;: AbstractJuMPScalar
+x = 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1196-L1216">source</a></section></article><h2 id="GenericAffExpr"><a class="docs-heading-anchor" href="#GenericAffExpr"><code>GenericAffExpr</code></a><a id="GenericAffExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericAffExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericAffExpr" href="#JuMP.GenericAffExpr"><code>JuMP.GenericAffExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">mutable struct GenericAffExpr{CoefType,VarType} &lt;: AbstractJuMPScalar
     constant::CoefType
     terms::OrderedDict{VarType,CoefType}
-end</code></pre><p>An expression type representing an affine expression of the form: <span>$\sum a_i x_i + c$</span>.</p><p><strong>Fields</strong></p><ul><li><code>.constant</code>: the constant <code>c</code> in the expression.</li><li><code>.terms</code>: an <code>OrderedDict</code>, with keys of <code>VarType</code> and values of <code>CoefType</code> describing the sparse vector <code>a</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L105-L119">source</a></section></article><h2 id="GenericModel"><a class="docs-heading-anchor" href="#GenericModel"><code>GenericModel</code></a><a id="GenericModel-1"></a><a class="docs-heading-anchor-permalink" href="#GenericModel" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericModel" href="#JuMP.GenericModel"><code>JuMP.GenericModel</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericModel{T}(
+end</code></pre><p>An expression type representing an affine expression of the form: <span>$\sum a_i x_i + c$</span>.</p><p><strong>Fields</strong></p><ul><li><code>.constant</code>: the constant <code>c</code> in the expression.</li><li><code>.terms</code>: an <code>OrderedDict</code>, with keys of <code>VarType</code> and values of <code>CoefType</code> describing the sparse vector <code>a</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L105-L119">source</a></section></article><h2 id="GenericModel"><a class="docs-heading-anchor" href="#GenericModel"><code>GenericModel</code></a><a id="GenericModel-1"></a><a class="docs-heading-anchor-permalink" href="#GenericModel" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericModel" href="#JuMP.GenericModel"><code>JuMP.GenericModel</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericModel{T}(
     [optimizer_factory;]
     add_bridges::Bool = true,
 ) where {T&lt;:Real}</code></pre><p>Create a new instance of a JuMP model.</p><p>If <code>optimizer_factory</code> is provided, the model is initialized with the optimizer returned by <code>MOI.instantiate(optimizer_factory)</code>.</p><p>If <code>optimizer_factory</code> is not provided, use <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a> to set the optimizer before calling <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p><p>If <code>add_bridges</code>, JuMP adds a <a href="../../moi/submodules/Bridges/reference/#MathOptInterface.Bridges.LazyBridgeOptimizer"><code>MOI.Bridges.LazyBridgeOptimizer</code></a> to automatically reformulate the problem into a form supported by the optimizer.</p><p><strong>Value type <code>T</code></strong></p><p>Passing a type other than <code>Float64</code> as the value type <code>T</code> is an advanced operation. The value type must match that expected by the chosen optimizer. Consult the optimizers documentation for details.</p><p>If not documented, assume that the optimizer supports only <code>Float64</code>.</p><p>Choosing an unsupported value type will throw an <a href="../../moi/reference/errors/#MathOptInterface.UnsupportedConstraint"><code>MOI.UnsupportedConstraint</code></a> or an <a href="../../moi/reference/errors/#MathOptInterface.UnsupportedAttribute"><code>MOI.UnsupportedAttribute</code></a> error, the timing of which (during the model construction or during a call to <a href="#JuMP.optimize!"><code>optimize!</code></a>) depends on how the solver is interfaced to JuMP.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = GenericModel{BigFloat}();
 
 julia&gt; typeof(model)
-GenericModel{BigFloat}</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L136-L174">source</a></section></article><h2 id="GenericNonlinearExpr"><a class="docs-heading-anchor" href="#GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a><a id="GenericNonlinearExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericNonlinearExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericNonlinearExpr" href="#JuMP.GenericNonlinearExpr"><code>JuMP.GenericNonlinearExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericNonlinearExpr{V}(head::Symbol, args::Vector{Any})
+GenericModel{BigFloat}</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L136-L174">source</a></section></article><h2 id="GenericNonlinearExpr"><a class="docs-heading-anchor" href="#GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a><a id="GenericNonlinearExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericNonlinearExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericNonlinearExpr" href="#JuMP.GenericNonlinearExpr"><code>JuMP.GenericNonlinearExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericNonlinearExpr{V}(head::Symbol, args::Vector{Any})
 GenericNonlinearExpr{V}(head::Symbol, args::Any...)</code></pre><p>The scalar-valued nonlinear function <code>head(args...)</code>, represented as a symbolic expression tree, with the call operator <code>head</code> and ordered arguments in <code>args</code>.</p><p><code>V</code> is the type of <a href="#JuMP.AbstractVariableRef"><code>AbstractVariableRef</code></a> present in the expression, and is used to help dispatch JuMP extensions.</p><p><strong><code>head</code></strong></p><p>The <code>head::Symbol</code> must be an operator supported by the model.</p><p>The default list of supported univariate operators is given by:</p><ul><li><a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS"><code>MOI.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS</code></a></li></ul><p>and the default list of supported multivariate operators is given by:</p><ul><li><a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS"><code>MOI.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS</code></a></li></ul><p>Additional operators can be add using <a href="#JuMP.@operator"><code>@operator</code></a>.</p><p>See the full list of operators supported by a <a href="../../moi/reference/models/#MathOptInterface.ModelLike"><code>MOI.ModelLike</code></a> by querying the <a href="../../moi/reference/models/#MathOptInterface.ListOfSupportedNonlinearOperators"><code>MOI.ListOfSupportedNonlinearOperators</code></a> attribute.</p><p><strong><code>args</code></strong></p><p>The vector <code>args</code> contains the arguments to the nonlinear function. If the operator is univariate, it must contain one element. Otherwise, it may contain multiple elements.</p><p>Given a subtype of <a href="#JuMP.AbstractVariableRef"><code>AbstractVariableRef</code></a>, <code>V</code>, for <code>GenericNonlinearExpr{V}</code>, each element must be one of the following:</p><ul><li>A constant value of type <code>&lt;:Real</code></li><li>A <code>V</code></li><li>A <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr{T,V}</code></a></li><li>A <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr{T,V}</code></a></li><li>A <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr{V}</code></a></li></ul><p>where <code>T&lt;:Real</code> and <code>T == value_type(V)</code>.</p><p><strong>Unsupported operators</strong></p><p>If the optimizer does not support <code>head</code>, an <a href="../../moi/reference/errors/#MathOptInterface.UnsupportedNonlinearOperator"><code>MOI.UnsupportedNonlinearOperator</code></a> error will be thrown.</p><p>There is no guarantee about when this error will be thrown; it may be thrown when the function is first added to the model, or it may be thrown when <a href="#JuMP.optimize!"><code>optimize!</code></a> is called.</p><p><strong>Example</strong></p><p>To represent the function <span>$f(x) = sin(x)^2$</span>, do:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
@@ -1747,15 +1747,15 @@
            GenericNonlinearExpr{VariableRef}(:sin, x),
            2.0,
        )
-sin(x) ^ 2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L6-L79">source</a></section></article><h2 id="GenericQuadExpr"><a class="docs-heading-anchor" href="#GenericQuadExpr"><code>GenericQuadExpr</code></a><a id="GenericQuadExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericQuadExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericQuadExpr" href="#JuMP.GenericQuadExpr"><code>JuMP.GenericQuadExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">mutable struct GenericQuadExpr{CoefType,VarType} &lt;: AbstractJuMPScalar
+sin(x) ^ 2.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L6-L79">source</a></section></article><h2 id="GenericQuadExpr"><a class="docs-heading-anchor" href="#GenericQuadExpr"><code>GenericQuadExpr</code></a><a id="GenericQuadExpr-1"></a><a class="docs-heading-anchor-permalink" href="#GenericQuadExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericQuadExpr" href="#JuMP.GenericQuadExpr"><code>JuMP.GenericQuadExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">mutable struct GenericQuadExpr{CoefType,VarType} &lt;: AbstractJuMPScalar
     aff::GenericAffExpr{CoefType,VarType}
     terms::OrderedDict{UnorderedPair{VarType}, CoefType}
-end</code></pre><p>An expression type representing an quadratic expression of the form: <span>$\sum q_{i,j} x_i x_j + \sum a_i x_i + c$</span>.</p><p><strong>Fields</strong></p><ul><li><code>.aff</code>: an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> representing the affine portion of the expression.</li><li><code>.terms</code>: an <code>OrderedDict</code>, with keys of <code>UnorderedPair{VarType}</code> and values of <code>CoefType</code>, describing the sparse list of terms <code>q</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L33-L48">source</a></section></article><h2 id="GenericReferenceMap"><a class="docs-heading-anchor" href="#GenericReferenceMap"><code>GenericReferenceMap</code></a><a id="GenericReferenceMap-1"></a><a class="docs-heading-anchor-permalink" href="#GenericReferenceMap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericReferenceMap" href="#JuMP.GenericReferenceMap"><code>JuMP.GenericReferenceMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericReferenceMap{T}</code></pre><p>Mapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L31-L37">source</a></section></article><h2 id="GenericVariableRef"><a class="docs-heading-anchor" href="#GenericVariableRef"><code>GenericVariableRef</code></a><a id="GenericVariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#GenericVariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericVariableRef" href="#JuMP.GenericVariableRef"><code>JuMP.GenericVariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericVariableRef{T} &lt;: AbstractVariableRef</code></pre><p>Holds a reference to the model and the corresponding MOI.VariableIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L245-L249">source</a></section></article><h2 id="HermitianMatrixShape"><a class="docs-heading-anchor" href="#HermitianMatrixShape"><code>HermitianMatrixShape</code></a><a id="HermitianMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianMatrixShape" href="#JuMP.HermitianMatrixShape"><code>JuMP.HermitianMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianMatrixShape</code></pre><p>Shape object for a Hermitian square matrix of <code>side_dimension</code> rows and columns. The vectorized form corresponds to <a href="../../moi/reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle"><code>MOI.HermitianPositiveSemidefiniteConeTriangle</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L554-L560">source</a></section></article><h2 id="HermitianMatrixSpace"><a class="docs-heading-anchor" href="#HermitianMatrixSpace"><code>HermitianMatrixSpace</code></a><a id="HermitianMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianMatrixSpace" href="#JuMP.HermitianMatrixSpace"><code>JuMP.HermitianMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be hermitian.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+end</code></pre><p>An expression type representing an quadratic expression of the form: <span>$\sum q_{i,j} x_i x_j + \sum a_i x_i + c$</span>.</p><p><strong>Fields</strong></p><ul><li><code>.aff</code>: an <a href="#JuMP.GenericAffExpr"><code>GenericAffExpr</code></a> representing the affine portion of the expression.</li><li><code>.terms</code>: an <code>OrderedDict</code>, with keys of <code>UnorderedPair{VarType}</code> and values of <code>CoefType</code>, describing the sparse list of terms <code>q</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L33-L48">source</a></section></article><h2 id="GenericReferenceMap"><a class="docs-heading-anchor" href="#GenericReferenceMap"><code>GenericReferenceMap</code></a><a id="GenericReferenceMap-1"></a><a class="docs-heading-anchor-permalink" href="#GenericReferenceMap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericReferenceMap" href="#JuMP.GenericReferenceMap"><code>JuMP.GenericReferenceMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericReferenceMap{T}</code></pre><p>Mapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L31-L37">source</a></section></article><h2 id="GenericVariableRef"><a class="docs-heading-anchor" href="#GenericVariableRef"><code>GenericVariableRef</code></a><a id="GenericVariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#GenericVariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.GenericVariableRef" href="#JuMP.GenericVariableRef"><code>JuMP.GenericVariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericVariableRef{T} &lt;: AbstractVariableRef</code></pre><p>Holds a reference to the model and the corresponding MOI.VariableIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L245-L249">source</a></section></article><h2 id="HermitianMatrixShape"><a class="docs-heading-anchor" href="#HermitianMatrixShape"><code>HermitianMatrixShape</code></a><a id="HermitianMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianMatrixShape" href="#JuMP.HermitianMatrixShape"><code>JuMP.HermitianMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianMatrixShape</code></pre><p>Shape object for a Hermitian square matrix of <code>side_dimension</code> rows and columns. The vectorized form corresponds to <a href="../../moi/reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle"><code>MOI.HermitianPositiveSemidefiniteConeTriangle</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L554-L560">source</a></section></article><h2 id="HermitianMatrixSpace"><a class="docs-heading-anchor" href="#HermitianMatrixSpace"><code>HermitianMatrixSpace</code></a><a id="HermitianMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianMatrixSpace" href="#JuMP.HermitianMatrixSpace"><code>JuMP.HermitianMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be hermitian.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())
 2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:
  real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im
- real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L56-L72">source</a></section></article><h2 id="HermitianPSDCone"><a class="docs-heading-anchor" href="#HermitianPSDCone"><code>HermitianPSDCone</code></a><a id="HermitianPSDCone-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianPSDCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianPSDCone" href="#JuMP.HermitianPSDCone"><code>JuMP.HermitianPSDCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianPSDCone</code></pre><p>Hermitian positive semidefinite cone object that can be used to create a Hermitian positive semidefinite square matrix in the <a href="#JuMP.@variable"><code>@variable</code></a> and <a href="#JuMP.@constraint"><code>@constraint</code></a> macros.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L56-L72">source</a></section></article><h2 id="HermitianPSDCone"><a class="docs-heading-anchor" href="#HermitianPSDCone"><code>HermitianPSDCone</code></a><a id="HermitianPSDCone-1"></a><a class="docs-heading-anchor-permalink" href="#HermitianPSDCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.HermitianPSDCone" href="#JuMP.HermitianPSDCone"><code>JuMP.HermitianPSDCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HermitianPSDCone</code></pre><p>Hermitian positive semidefinite cone object that can be used to create a Hermitian positive semidefinite square matrix in the <a href="#JuMP.@variable"><code>@variable</code></a> and <a href="#JuMP.@constraint"><code>@constraint</code></a> macros.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, H[1:3, 1:3] in HermitianPSDCone())
 3×3 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:
@@ -1777,17 +1777,17 @@
 
 julia&gt; all_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle)
 1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.HermitianPositiveSemidefiniteConeTriangle}}}:
- [real(H[1,1]), real(H[1,2]), real(H[2,2]), real(H[1,3]), real(H[2,3]), real(H[3,3]), imag(H[1,2]), imag(H[1,3]), imag(H[2,3])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(3)</code></pre><p>We see in the output of the last commands that 9 real variables were created. The matrix <code>H</code> contrains affine expressions in terms of these 9 variables that parametrize a Hermitian matrix.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L513-L551">source</a></section></article><h2 id="IntegerRef"><a class="docs-heading-anchor" href="#IntegerRef"><code>IntegerRef</code></a><a id="IntegerRef-1"></a><a class="docs-heading-anchor-permalink" href="#IntegerRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.IntegerRef" href="#JuMP.IntegerRef"><code>JuMP.IntegerRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">IntegerRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>v</code> to be integer.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ [real(H[1,1]), real(H[1,2]), real(H[2,2]), real(H[1,3]), real(H[2,3]), real(H[3,3]), imag(H[1,2]), imag(H[1,3]), imag(H[2,3])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(3)</code></pre><p>We see in the output of the last commands that 9 real variables were created. The matrix <code>H</code> contrains affine expressions in terms of these 9 variables that parametrize a Hermitian matrix.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L513-L551">source</a></section></article><h2 id="IntegerRef"><a class="docs-heading-anchor" href="#IntegerRef"><code>IntegerRef</code></a><a id="IntegerRef-1"></a><a class="docs-heading-anchor-permalink" href="#IntegerRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.IntegerRef" href="#JuMP.IntegerRef"><code>JuMP.IntegerRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">IntegerRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>v</code> to be integer.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_integer"><code>is_integer</code></a>, <a href="#JuMP.set_integer"><code>set_integer</code></a>, <a href="#JuMP.unset_integer"><code>unset_integer</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x, Int);
 
 julia&gt; IntegerRef(x)
-x integer</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1334-L1353">source</a></section></article><h2 id="LinearTermIterator"><a class="docs-heading-anchor" href="#LinearTermIterator"><code>LinearTermIterator</code></a><a id="LinearTermIterator-1"></a><a class="docs-heading-anchor-permalink" href="#LinearTermIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LinearTermIterator" href="#JuMP.LinearTermIterator"><code>JuMP.LinearTermIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">LinearTermIterator{GAE&lt;:GenericAffExpr}</code></pre><p>A struct that implements the <code>iterate</code> protocol in order to iterate over tuples of <code>(coefficient, variable)</code> in the <code>GenericAffExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/aff_expr.jl#L368-L373">source</a></section></article><h2 id="LowerBoundRef"><a class="docs-heading-anchor" href="#LowerBoundRef"><code>LowerBoundRef</code></a><a id="LowerBoundRef-1"></a><a class="docs-heading-anchor-permalink" href="#LowerBoundRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LowerBoundRef" href="#JuMP.LowerBoundRef"><code>JuMP.LowerBoundRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">LowerBoundRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the lower bound constraint of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x integer</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1334-L1353">source</a></section></article><h2 id="LinearTermIterator"><a class="docs-heading-anchor" href="#LinearTermIterator"><code>LinearTermIterator</code></a><a id="LinearTermIterator-1"></a><a class="docs-heading-anchor-permalink" href="#LinearTermIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LinearTermIterator" href="#JuMP.LinearTermIterator"><code>JuMP.LinearTermIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">LinearTermIterator{GAE&lt;:GenericAffExpr}</code></pre><p>A struct that implements the <code>iterate</code> protocol in order to iterate over tuples of <code>(coefficient, variable)</code> in the <code>GenericAffExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/aff_expr.jl#L368-L373">source</a></section></article><h2 id="LowerBoundRef"><a class="docs-heading-anchor" href="#LowerBoundRef"><code>LowerBoundRef</code></a><a id="LowerBoundRef-1"></a><a class="docs-heading-anchor-permalink" href="#LowerBoundRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LowerBoundRef" href="#JuMP.LowerBoundRef"><code>JuMP.LowerBoundRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">LowerBoundRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the lower bound constraint of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.has_lower_bound"><code>has_lower_bound</code></a>, <a href="#JuMP.lower_bound"><code>lower_bound</code></a>, <a href="#JuMP.set_lower_bound"><code>set_lower_bound</code></a>, <a href="#JuMP.delete_lower_bound"><code>delete_lower_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &gt;= 1.0);
 
 julia&gt; LowerBoundRef(x)
-x ≥ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L759-L779">source</a></section></article><h2 id="Model"><a class="docs-heading-anchor" href="#Model"><code>Model</code></a><a id="Model-1"></a><a class="docs-heading-anchor-permalink" href="#Model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Model" href="#JuMP.Model"><code>JuMP.Model</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Model([optimizer_factory;] add_bridges::Bool = true)</code></pre><p>Create a new instance of a JuMP model.</p><p>If <code>optimizer_factory</code> is provided, the model is initialized with thhe optimizer returned by <code>MOI.instantiate(optimizer_factory)</code>.</p><p>If <code>optimizer_factory</code> is not provided, use <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a> to set the optimizer before calling <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p><p>If <code>add_bridges</code>, JuMP adds a <a href="../../moi/submodules/Bridges/reference/#MathOptInterface.Bridges.LazyBridgeOptimizer"><code>MOI.Bridges.LazyBridgeOptimizer</code></a> to automatically reformulate the problem into a form supported by the optimizer.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import Ipopt
+x ≥ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L759-L779">source</a></section></article><h2 id="Model"><a class="docs-heading-anchor" href="#Model"><code>Model</code></a><a id="Model-1"></a><a class="docs-heading-anchor-permalink" href="#Model" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Model" href="#JuMP.Model"><code>JuMP.Model</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Model([optimizer_factory;] add_bridges::Bool = true)</code></pre><p>Create a new instance of a JuMP model.</p><p>If <code>optimizer_factory</code> is provided, the model is initialized with thhe optimizer returned by <code>MOI.instantiate(optimizer_factory)</code>.</p><p>If <code>optimizer_factory</code> is not provided, use <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a> to set the optimizer before calling <a href="#JuMP.optimize!"><code>optimize!</code></a>.</p><p>If <code>add_bridges</code>, JuMP adds a <a href="../../moi/submodules/Bridges/reference/#MathOptInterface.Bridges.LazyBridgeOptimizer"><code>MOI.Bridges.LazyBridgeOptimizer</code></a> to automatically reformulate the problem into a form supported by the optimizer.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; import Ipopt
 
 julia&gt; model = Model(Ipopt.Optimizer);
 
@@ -1798,11 +1798,11 @@
 
 julia&gt; import MultiObjectiveAlgorithms as MOA
 
-julia&gt; model = Model(() -&gt; MOA.Optimizer(HiGHS.Optimizer); add_bridges = false);</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L280-L310">source</a></section></article><h2 id="ModelMode"><a class="docs-heading-anchor" href="#ModelMode"><code>ModelMode</code></a><a id="ModelMode-1"></a><a class="docs-heading-anchor-permalink" href="#ModelMode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ModelMode" href="#JuMP.ModelMode"><code>JuMP.ModelMode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelMode</code></pre><p>An enum to describe the state of the CachingOptimizer inside a JuMP model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L50-L54">source</a></section></article><h2 id="NLPEvaluator"><a class="docs-heading-anchor" href="#NLPEvaluator"><code>NLPEvaluator</code></a><a id="NLPEvaluator-1"></a><a class="docs-heading-anchor-permalink" href="#NLPEvaluator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NLPEvaluator" href="#JuMP.NLPEvaluator"><code>JuMP.NLPEvaluator</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">NLPEvaluator(
+julia&gt; model = Model(() -&gt; MOA.Optimizer(HiGHS.Optimizer); add_bridges = false);</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L280-L310">source</a></section></article><h2 id="ModelMode"><a class="docs-heading-anchor" href="#ModelMode"><code>ModelMode</code></a><a id="ModelMode-1"></a><a class="docs-heading-anchor-permalink" href="#ModelMode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ModelMode" href="#JuMP.ModelMode"><code>JuMP.ModelMode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelMode</code></pre><p>An enum to describe the state of the CachingOptimizer inside a JuMP model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L50-L54">source</a></section></article><h2 id="NLPEvaluator"><a class="docs-heading-anchor" href="#NLPEvaluator"><code>NLPEvaluator</code></a><a id="NLPEvaluator-1"></a><a class="docs-heading-anchor-permalink" href="#NLPEvaluator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NLPEvaluator" href="#JuMP.NLPEvaluator"><code>JuMP.NLPEvaluator</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">NLPEvaluator(
     model::Model,
     _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =
         MOI.Nonlinear.SparseReverseMode(),
-)</code></pre><p>Return an <a href="../../moi/reference/nonlinear/#MathOptInterface.AbstractNLPEvaluator"><code>MOI.AbstractNLPEvaluator</code></a> constructed from <code>model</code></p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Before using, you must initialize the evaluator using <a href="../../moi/reference/nonlinear/#MathOptInterface.initialize"><code>MOI.initialize</code></a>.</p></div></div><p><strong>Experimental</strong></p><p>These features may change or be removed in any future version of JuMP.</p><p>Pass <code>_differentiation_backend</code> to specify the differentiation backend used to compute derivatives.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L881-L900">source</a></section></article><h2 id="NoOptimizer"><a class="docs-heading-anchor" href="#NoOptimizer"><code>NoOptimizer</code></a><a id="NoOptimizer-1"></a><a class="docs-heading-anchor-permalink" href="#NoOptimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NoOptimizer" href="#JuMP.NoOptimizer"><code>JuMP.NoOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct NoOptimizer &lt;: Exception end</code></pre><p>No optimizer is set. The optimizer can be provided to the <a href="#JuMP.Model"><code>Model</code></a> constructor or by calling <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L640-L645">source</a></section></article><h2 id="NonlinearConstraintIndex"><a class="docs-heading-anchor" href="#NonlinearConstraintIndex"><code>NonlinearConstraintIndex</code></a><a id="NonlinearConstraintIndex-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearConstraintIndex" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearConstraintIndex" href="#JuMP.NonlinearConstraintIndex"><code>JuMP.NonlinearConstraintIndex</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintIndex</code></pre><p>An index to a nonlinear constraint that is returned by <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_constraint"><code>add_constraint</code></a>.</p><p>Given <code>data::Model</code> and <code>c::ConstraintIndex</code>, use <code>data[c]</code> to retrieve the corresponding <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Constraint"><code>Constraint</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Nonlinear/types.jl#L133-L140">source</a></section></article><h2 id="NonlinearConstraintRef"><a class="docs-heading-anchor" href="#NonlinearConstraintRef"><code>NonlinearConstraintRef</code></a><a id="NonlinearConstraintRef-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearConstraintRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearConstraintRef" href="#JuMP.NonlinearConstraintRef"><code>JuMP.NonlinearConstraintRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearConstraintRef</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L402-L404">source</a></section></article><h2 id="NonlinearExpr"><a class="docs-heading-anchor" href="#NonlinearExpr"><code>NonlinearExpr</code></a><a id="NonlinearExpr-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearExpr" href="#JuMP.NonlinearExpr"><code>JuMP.NonlinearExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearExpr</code></pre><p>Alias for <code>GenericNonlinearExpr{VariableRef}</code>, the specific <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L128-L133">source</a></section></article><h2 id="NonlinearExpression"><a class="docs-heading-anchor" href="#NonlinearExpression"><code>NonlinearExpression</code></a><a id="NonlinearExpression-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearExpression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearExpression" href="#JuMP.NonlinearExpression"><code>JuMP.NonlinearExpression</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearExpression &lt;: AbstractJuMPScalar</code></pre><p>A struct to represent a nonlinear expression.</p><p>Create an expression using <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L280-L286">source</a></section></article><h2 id="NonlinearOperator"><a class="docs-heading-anchor" href="#NonlinearOperator"><code>NonlinearOperator</code></a><a id="NonlinearOperator-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearOperator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearOperator" href="#JuMP.NonlinearOperator"><code>JuMP.NonlinearOperator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearOperator(func::Function, head::Symbol)</code></pre><p>A callable struct (functor) representing a function named <code>head</code>.</p><p>When called with <a href="#JuMP.AbstractJuMPScalar"><code>AbstractJuMPScalar</code></a>s, the struct returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p>When called with non-JuMP types, the struct returns the evaluation of <code>func(args...)</code>.</p><p>Unless <code>head</code> is special-cased by the optimizer, the operator must have already been added to the model using <a href="#JuMP.add_nonlinear_operator"><code>add_nonlinear_operator</code></a> or <a href="#JuMP.@operator"><code>@operator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+)</code></pre><p>Return an <a href="../../moi/reference/nonlinear/#MathOptInterface.AbstractNLPEvaluator"><code>MOI.AbstractNLPEvaluator</code></a> constructed from <code>model</code></p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Before using, you must initialize the evaluator using <a href="../../moi/reference/nonlinear/#MathOptInterface.initialize"><code>MOI.initialize</code></a>.</p></div></div><p><strong>Experimental</strong></p><p>These features may change or be removed in any future version of JuMP.</p><p>Pass <code>_differentiation_backend</code> to specify the differentiation backend used to compute derivatives.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L881-L900">source</a></section></article><h2 id="NoOptimizer"><a class="docs-heading-anchor" href="#NoOptimizer"><code>NoOptimizer</code></a><a id="NoOptimizer-1"></a><a class="docs-heading-anchor-permalink" href="#NoOptimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NoOptimizer" href="#JuMP.NoOptimizer"><code>JuMP.NoOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct NoOptimizer &lt;: Exception end</code></pre><p>No optimizer is set. The optimizer can be provided to the <a href="#JuMP.Model"><code>Model</code></a> constructor or by calling <a href="#JuMP.set_optimizer"><code>set_optimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L640-L645">source</a></section></article><h2 id="NonlinearConstraintIndex"><a class="docs-heading-anchor" href="#NonlinearConstraintIndex"><code>NonlinearConstraintIndex</code></a><a id="NonlinearConstraintIndex-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearConstraintIndex" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearConstraintIndex" href="#JuMP.NonlinearConstraintIndex"><code>JuMP.NonlinearConstraintIndex</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintIndex</code></pre><p>An index to a nonlinear constraint that is returned by <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_constraint"><code>add_constraint</code></a>.</p><p>Given <code>data::Model</code> and <code>c::ConstraintIndex</code>, use <code>data[c]</code> to retrieve the corresponding <a href="../../moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Constraint"><code>Constraint</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Nonlinear/types.jl#L133-L140">source</a></section></article><h2 id="NonlinearConstraintRef"><a class="docs-heading-anchor" href="#NonlinearConstraintRef"><code>NonlinearConstraintRef</code></a><a id="NonlinearConstraintRef-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearConstraintRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearConstraintRef" href="#JuMP.NonlinearConstraintRef"><code>JuMP.NonlinearConstraintRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearConstraintRef</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L402-L404">source</a></section></article><h2 id="NonlinearExpr"><a class="docs-heading-anchor" href="#NonlinearExpr"><code>NonlinearExpr</code></a><a id="NonlinearExpr-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearExpr" href="#JuMP.NonlinearExpr"><code>JuMP.NonlinearExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearExpr</code></pre><p>Alias for <code>GenericNonlinearExpr{VariableRef}</code>, the specific <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L128-L133">source</a></section></article><h2 id="NonlinearExpression"><a class="docs-heading-anchor" href="#NonlinearExpression"><code>NonlinearExpression</code></a><a id="NonlinearExpression-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearExpression" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearExpression" href="#JuMP.NonlinearExpression"><code>JuMP.NonlinearExpression</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearExpression &lt;: AbstractJuMPScalar</code></pre><p>A struct to represent a nonlinear expression.</p><p>Create an expression using <a href="#JuMP.@NLexpression"><code>@NLexpression</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L280-L286">source</a></section></article><h2 id="NonlinearOperator"><a class="docs-heading-anchor" href="#NonlinearOperator"><code>NonlinearOperator</code></a><a id="NonlinearOperator-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearOperator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearOperator" href="#JuMP.NonlinearOperator"><code>JuMP.NonlinearOperator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearOperator(func::Function, head::Symbol)</code></pre><p>A callable struct (functor) representing a function named <code>head</code>.</p><p>When called with <a href="#JuMP.AbstractJuMPScalar"><code>AbstractJuMPScalar</code></a>s, the struct returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p>When called with non-JuMP types, the struct returns the evaluation of <code>func(args...)</code>.</p><p>Unless <code>head</code> is special-cased by the optimizer, the operator must have already been added to the model using <a href="#JuMP.add_nonlinear_operator"><code>add_nonlinear_operator</code></a> or <a href="#JuMP.@operator"><code>@operator</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1826,7 +1826,7 @@
 op_f(x)
 
 julia&gt; bar(2.0)
-4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp_expr.jl#L810-L854">source</a></section></article><h2 id="NonlinearParameter"><a class="docs-heading-anchor" href="#NonlinearParameter"><code>NonlinearParameter</code></a><a id="NonlinearParameter-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearParameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearParameter" href="#JuMP.NonlinearParameter"><code>JuMP.NonlinearParameter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearParameter &lt;: AbstractJuMPScalar</code></pre><p>A struct to represent a nonlinear parameter.</p><p>Create a parameter using <a href="#JuMP.@NLparameter"><code>@NLparameter</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/nlp.jl#L188-L194">source</a></section></article><h2 id="Nonnegatives"><a class="docs-heading-anchor" href="#Nonnegatives"><code>Nonnegatives</code></a><a id="Nonnegatives-1"></a><a class="docs-heading-anchor-permalink" href="#Nonnegatives" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Nonnegatives" href="#JuMP.Nonnegatives"><code>JuMP.Nonnegatives</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Nonnegatives()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Nonnegatives"><code>MOI.Nonnegatives</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+4.0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp_expr.jl#L810-L854">source</a></section></article><h2 id="NonlinearParameter"><a class="docs-heading-anchor" href="#NonlinearParameter"><code>NonlinearParameter</code></a><a id="NonlinearParameter-1"></a><a class="docs-heading-anchor-permalink" href="#NonlinearParameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NonlinearParameter" href="#JuMP.NonlinearParameter"><code>JuMP.NonlinearParameter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NonlinearParameter &lt;: AbstractJuMPScalar</code></pre><p>A struct to represent a nonlinear parameter.</p><p>Create a parameter using <a href="#JuMP.@NLparameter"><code>@NLparameter</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/nlp.jl#L188-L194">source</a></section></article><h2 id="Nonnegatives"><a class="docs-heading-anchor" href="#Nonnegatives"><code>Nonnegatives</code></a><a id="Nonnegatives-1"></a><a class="docs-heading-anchor-permalink" href="#Nonnegatives" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Nonnegatives" href="#JuMP.Nonnegatives"><code>JuMP.Nonnegatives</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Nonnegatives()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Nonnegatives"><code>MOI.Nonnegatives</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x[1:2])
 2-element Vector{VariableRef}:
@@ -1841,7 +1841,7 @@
 julia&gt; b = [5, 6];
 
 julia&gt; @constraint(model, A * x &gt;= b)
-[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L235-L261">source</a></section></article><h2 id="Nonpositives"><a class="docs-heading-anchor" href="#Nonpositives"><code>Nonpositives</code></a><a id="Nonpositives-1"></a><a class="docs-heading-anchor-permalink" href="#Nonpositives" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Nonpositives" href="#JuMP.Nonpositives"><code>JuMP.Nonpositives</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Nonpositives()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Nonpositives"><code>MOI.Nonpositives</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L235-L261">source</a></section></article><h2 id="Nonpositives"><a class="docs-heading-anchor" href="#Nonpositives"><code>Nonpositives</code></a><a id="Nonpositives-1"></a><a class="docs-heading-anchor-permalink" href="#Nonpositives" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Nonpositives" href="#JuMP.Nonpositives"><code>JuMP.Nonpositives</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Nonpositives()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Nonpositives"><code>MOI.Nonpositives</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x[1:2])
 2-element Vector{VariableRef}:
@@ -1856,7 +1856,7 @@
 julia&gt; b = [5, 6];
 
 julia&gt; @constraint(model, A * x &lt;= b)
-[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L264-L290">source</a></section></article><h2 id="OptimizationSense"><a class="docs-heading-anchor" href="#OptimizationSense"><code>OptimizationSense</code></a><a id="OptimizationSense-1"></a><a class="docs-heading-anchor-permalink" href="#OptimizationSense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OptimizationSense" href="#JuMP.OptimizationSense"><code>JuMP.OptimizationSense</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">OptimizationSense</code></pre><p>An enum for the value of the <a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>ObjectiveSense</code></a> attribute.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.MIN_SENSE"><code>MIN_SENSE</code></a>: the goal is to minimize the objective function</li><li><a href="../../moi/reference/models/#MathOptInterface.MAX_SENSE"><code>MAX_SENSE</code></a>: the goal is to maximize the objective function</li><li><a href="../../moi/reference/models/#MathOptInterface.FEASIBILITY_SENSE"><code>FEASIBILITY_SENSE</code></a>: the model does not have an objective function</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1098-L1110">source</a></section></article><h2 id="OptimizeNotCalled"><a class="docs-heading-anchor" href="#OptimizeNotCalled"><code>OptimizeNotCalled</code></a><a id="OptimizeNotCalled-1"></a><a class="docs-heading-anchor-permalink" href="#OptimizeNotCalled" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OptimizeNotCalled" href="#JuMP.OptimizeNotCalled"><code>JuMP.OptimizeNotCalled</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct OptimizeNotCalled &lt;: Exception end</code></pre><p>A result attribute cannot be queried before <a href="#JuMP.optimize!"><code>optimize!</code></a> is called.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L633-L637">source</a></section></article><h2 id="PSDCone"><a class="docs-heading-anchor" href="#PSDCone"><code>PSDCone</code></a><a id="PSDCone-1"></a><a class="docs-heading-anchor-permalink" href="#PSDCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.PSDCone" href="#JuMP.PSDCone"><code>JuMP.PSDCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">PSDCone</code></pre><p>Positive semidefinite cone object that can be used to constrain a square matrix to be positive semidefinite in the <a href="#JuMP.@constraint"><code>@constraint</code></a> macro. If the matrix has type <code>Symmetric</code> then the columns vectorization (the vector obtained by concatenating the columns) of its upper triangular part is constrained to belong to the <code>MOI.PositiveSemidefiniteConeTriangle</code> set, otherwise its column vectorization is constrained to belong to the <code>MOI.PositiveSemidefiniteConeSquare</code> set.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L264-L290">source</a></section></article><h2 id="OptimizationSense"><a class="docs-heading-anchor" href="#OptimizationSense"><code>OptimizationSense</code></a><a id="OptimizationSense-1"></a><a class="docs-heading-anchor-permalink" href="#OptimizationSense" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OptimizationSense" href="#JuMP.OptimizationSense"><code>JuMP.OptimizationSense</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">OptimizationSense</code></pre><p>An enum for the value of the <a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>ObjectiveSense</code></a> attribute.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.MIN_SENSE"><code>MIN_SENSE</code></a>: the goal is to minimize the objective function</li><li><a href="../../moi/reference/models/#MathOptInterface.MAX_SENSE"><code>MAX_SENSE</code></a>: the goal is to maximize the objective function</li><li><a href="../../moi/reference/models/#MathOptInterface.FEASIBILITY_SENSE"><code>FEASIBILITY_SENSE</code></a>: the model does not have an objective function</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1098-L1110">source</a></section></article><h2 id="OptimizeNotCalled"><a class="docs-heading-anchor" href="#OptimizeNotCalled"><code>OptimizeNotCalled</code></a><a id="OptimizeNotCalled-1"></a><a class="docs-heading-anchor-permalink" href="#OptimizeNotCalled" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OptimizeNotCalled" href="#JuMP.OptimizeNotCalled"><code>JuMP.OptimizeNotCalled</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct OptimizeNotCalled &lt;: Exception end</code></pre><p>A result attribute cannot be queried before <a href="#JuMP.optimize!"><code>optimize!</code></a> is called.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L633-L637">source</a></section></article><h2 id="PSDCone"><a class="docs-heading-anchor" href="#PSDCone"><code>PSDCone</code></a><a id="PSDCone-1"></a><a class="docs-heading-anchor-permalink" href="#PSDCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.PSDCone" href="#JuMP.PSDCone"><code>JuMP.PSDCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">PSDCone</code></pre><p>Positive semidefinite cone object that can be used to constrain a square matrix to be positive semidefinite in the <a href="#JuMP.@constraint"><code>@constraint</code></a> macro. If the matrix has type <code>Symmetric</code> then the columns vectorization (the vector obtained by concatenating the columns) of its upper triangular part is constrained to belong to the <code>MOI.PositiveSemidefiniteConeTriangle</code> set, otherwise its column vectorization is constrained to belong to the <code>MOI.PositiveSemidefiniteConeSquare</code> set.</p><p><strong>Example</strong></p><p>Consider the following example:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1892,7 +1892,7 @@
  x - 4
 
 julia&gt; moi_set(constraint_object(cref))
-MathOptInterface.PositiveSemidefiniteConeTriangle(2)</code></pre><p>As we see in the output of the last command, the vectorization of only the upper triangular part of the matrix is constrained to belong to the <code>PositiveSemidefiniteConeSquare</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L75-L137">source</a></section></article><h2 id="Parameter"><a class="docs-heading-anchor" href="#Parameter"><code>Parameter</code></a><a id="Parameter-1"></a><a class="docs-heading-anchor-permalink" href="#Parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Parameter" href="#JuMP.Parameter"><code>JuMP.Parameter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Parameter(value)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Parameter"><code>MOI.Parameter</code></a> set.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+MathOptInterface.PositiveSemidefiniteConeTriangle(2)</code></pre><p>As we see in the output of the last command, the vectorization of only the upper triangular part of the matrix is constrained to belong to the <code>PositiveSemidefiniteConeSquare</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L75-L137">source</a></section></article><h2 id="Parameter"><a class="docs-heading-anchor" href="#Parameter"><code>Parameter</code></a><a id="Parameter-1"></a><a class="docs-heading-anchor-permalink" href="#Parameter" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Parameter" href="#JuMP.Parameter"><code>JuMP.Parameter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Parameter(value)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Parameter"><code>MOI.Parameter</code></a> set.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x in Parameter(2))
 x
@@ -1900,7 +1900,7 @@
 julia&gt; print(model)
 Feasibility
 Subject to
- x ∈ MathOptInterface.Parameter{Float64}(2.0)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L364-L382">source</a></section></article><h2 id="ParameterRef"><a class="docs-heading-anchor" href="#ParameterRef"><code>ParameterRef</code></a><a id="ParameterRef-1"></a><a class="docs-heading-anchor-permalink" href="#ParameterRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ParameterRef" href="#JuMP.ParameterRef"><code>JuMP.ParameterRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">ParameterRef(x::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>x</code> to be a parameter.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ x ∈ MathOptInterface.Parameter{Float64}(2.0)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L364-L382">source</a></section></article><h2 id="ParameterRef"><a class="docs-heading-anchor" href="#ParameterRef"><code>ParameterRef</code></a><a id="ParameterRef-1"></a><a class="docs-heading-anchor-permalink" href="#ParameterRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ParameterRef" href="#JuMP.ParameterRef"><code>JuMP.ParameterRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">ParameterRef(x::GenericVariableRef)</code></pre><p>Return a constraint reference to the constraint constraining <code>x</code> to be a parameter.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.is_parameter"><code>is_parameter</code></a>, <a href="#JuMP.set_parameter_value"><code>set_parameter_value</code></a>, <a href="#JuMP.parameter_value"><code>parameter_value</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, p in Parameter(2))
 p
@@ -1913,7 +1913,7 @@
 julia&gt; ParameterRef(x)
 ERROR: Variable x is not a parameter.
 Stacktrace:
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1493-L1522">source</a></section></article><h2 id="QuadExpr"><a class="docs-heading-anchor" href="#QuadExpr"><code>QuadExpr</code></a><a id="QuadExpr-1"></a><a class="docs-heading-anchor-permalink" href="#QuadExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.QuadExpr" href="#JuMP.QuadExpr"><code>JuMP.QuadExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">QuadExpr</code></pre><p>An alias for <code>GenericQuadExpr{Float64,VariableRef}</code>, the specific <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L542-L547">source</a></section></article><h2 id="QuadTermIterator"><a class="docs-heading-anchor" href="#QuadTermIterator"><code>QuadTermIterator</code></a><a id="QuadTermIterator-1"></a><a class="docs-heading-anchor-permalink" href="#QuadTermIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.QuadTermIterator" href="#JuMP.QuadTermIterator"><code>JuMP.QuadTermIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">QuadTermIterator{GQE&lt;:GenericQuadExpr}</code></pre><p>A struct that implements the <code>iterate</code> protocol in order to iterate over tuples of <code>(coefficient, variable, variable)</code> in the <code>GenericQuadExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L274-L279">source</a></section></article><h2 id="ReferenceMap"><a class="docs-heading-anchor" href="#ReferenceMap"><code>ReferenceMap</code></a><a id="ReferenceMap-1"></a><a class="docs-heading-anchor-permalink" href="#ReferenceMap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ReferenceMap" href="#JuMP.ReferenceMap"><code>JuMP.ReferenceMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericReferenceMap{T}</code></pre><p>Mapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L31-L37">source</a></section></article><h2 id="ResultStatusCode"><a class="docs-heading-anchor" href="#ResultStatusCode"><code>ResultStatusCode</code></a><a id="ResultStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#ResultStatusCode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ResultStatusCode" href="#JuMP.ResultStatusCode"><code>JuMP.ResultStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultStatusCode</code></pre><p>An Enum of possible values for the <a href="../../moi/reference/models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> and <a href="../../moi/reference/models/#MathOptInterface.DualStatus"><code>DualStatus</code></a> attributes.</p><p>The values indicate how to interpret the result vector.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.NO_SOLUTION"><code>NO_SOLUTION</code></a>: the result vector is empty.</li><li><a href="../../moi/reference/models/#MathOptInterface.FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a>: the result vector is a feasible point.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a>: the result vector is feasible if some constraint tolerances are relaxed.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a>: the result vector is an infeasible point.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for a certificate of infeasibility.</li><li><a href="../../moi/reference/models/#MathOptInterface.REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for an ill-posed certificate.</li><li><a href="../../moi/reference/models/#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a>: the result vector contains a solution with an unknown interpretation.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2080">source</a></section></article><h2 id="RotatedSecondOrderCone"><a class="docs-heading-anchor" href="#RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a><a id="RotatedSecondOrderCone-1"></a><a class="docs-heading-anchor-permalink" href="#RotatedSecondOrderCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.RotatedSecondOrderCone" href="#JuMP.RotatedSecondOrderCone"><code>JuMP.RotatedSecondOrderCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RotatedSecondOrderCone</code></pre><p>Rotated second order cone object that can be used to constrain the square of the euclidean norm of a vector <code>x</code> to be less than or equal to <span>$2tu$</span> where <code>t</code> and <code>u</code> are nonnegative scalars. This is a shortcut for the <code>MOI.RotatedSecondOrderCone</code>.</p><p><strong>Example</strong></p><p>The following constrains <span>$\|(x-1, x-2)\|^2_2 \le 2tx$</span> and <span>$t, x \ge 0$</span>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1493-L1522">source</a></section></article><h2 id="QuadExpr"><a class="docs-heading-anchor" href="#QuadExpr"><code>QuadExpr</code></a><a id="QuadExpr-1"></a><a class="docs-heading-anchor-permalink" href="#QuadExpr" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.QuadExpr" href="#JuMP.QuadExpr"><code>JuMP.QuadExpr</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">QuadExpr</code></pre><p>An alias for <code>GenericQuadExpr{Float64,VariableRef}</code>, the specific <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> used by JuMP.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L542-L547">source</a></section></article><h2 id="QuadTermIterator"><a class="docs-heading-anchor" href="#QuadTermIterator"><code>QuadTermIterator</code></a><a id="QuadTermIterator-1"></a><a class="docs-heading-anchor-permalink" href="#QuadTermIterator" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.QuadTermIterator" href="#JuMP.QuadTermIterator"><code>JuMP.QuadTermIterator</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">QuadTermIterator{GQE&lt;:GenericQuadExpr}</code></pre><p>A struct that implements the <code>iterate</code> protocol in order to iterate over tuples of <code>(coefficient, variable, variable)</code> in the <code>GenericQuadExpr</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L274-L279">source</a></section></article><h2 id="ReferenceMap"><a class="docs-heading-anchor" href="#ReferenceMap"><code>ReferenceMap</code></a><a id="ReferenceMap-1"></a><a class="docs-heading-anchor-permalink" href="#ReferenceMap" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ReferenceMap" href="#JuMP.ReferenceMap"><code>JuMP.ReferenceMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericReferenceMap{T}</code></pre><p>Mapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L31-L37">source</a></section></article><h2 id="ResultStatusCode"><a class="docs-heading-anchor" href="#ResultStatusCode"><code>ResultStatusCode</code></a><a id="ResultStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#ResultStatusCode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ResultStatusCode" href="#JuMP.ResultStatusCode"><code>JuMP.ResultStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultStatusCode</code></pre><p>An Enum of possible values for the <a href="../../moi/reference/models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> and <a href="../../moi/reference/models/#MathOptInterface.DualStatus"><code>DualStatus</code></a> attributes.</p><p>The values indicate how to interpret the result vector.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.NO_SOLUTION"><code>NO_SOLUTION</code></a>: the result vector is empty.</li><li><a href="../../moi/reference/models/#MathOptInterface.FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a>: the result vector is a feasible point.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a>: the result vector is feasible if some constraint tolerances are relaxed.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a>: the result vector is an infeasible point.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for a certificate of infeasibility.</li><li><a href="../../moi/reference/models/#MathOptInterface.REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</li><li><a href="../../moi/reference/models/#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for an ill-posed certificate.</li><li><a href="../../moi/reference/models/#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a>: the result vector contains a solution with an unknown interpretation.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2080">source</a></section></article><h2 id="RotatedSecondOrderCone"><a class="docs-heading-anchor" href="#RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a><a id="RotatedSecondOrderCone-1"></a><a class="docs-heading-anchor-permalink" href="#RotatedSecondOrderCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.RotatedSecondOrderCone" href="#JuMP.RotatedSecondOrderCone"><code>JuMP.RotatedSecondOrderCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RotatedSecondOrderCone</code></pre><p>Rotated second order cone object that can be used to constrain the square of the euclidean norm of a vector <code>x</code> to be less than or equal to <span>$2tu$</span> where <code>t</code> and <code>u</code> are nonnegative scalars. This is a shortcut for the <code>MOI.RotatedSecondOrderCone</code>.</p><p><strong>Example</strong></p><p>The following constrains <span>$\|(x-1, x-2)\|^2_2 \le 2tx$</span> and <span>$t, x \ge 0$</span>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1922,7 +1922,7 @@
 t
 
 julia&gt; @constraint(model, [t, x, x-1, x-2] in RotatedSecondOrderCone())
-[t, x, x - 1, x - 2] ∈ MathOptInterface.RotatedSecondOrderCone(4)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L157-L180">source</a></section></article><h2 id="SOS1"><a class="docs-heading-anchor" href="#SOS1"><code>SOS1</code></a><a id="SOS1-1"></a><a class="docs-heading-anchor-permalink" href="#SOS1" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOS1" href="#JuMP.SOS1"><code>JuMP.SOS1</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SOS1</code></pre><p>SOS1 (Special Ordered Sets type 1) object than can be used to constrain a vector <code>x</code> to a set where at most 1 variable can take a non-zero value, all others being at 0. The <code>weights</code>, when specified, induce an ordering of the variables; as such, they should be unique values. The <em>k</em>th element in the set corresponds to the <em>k</em>th weight in <code>weights</code>. See <a href="http://lpsolve.sourceforge.net/5.5/SOS.htm">here</a> for a description of SOS constraints and their potential uses. This is a shortcut for the <code>MathOptInterface.SOS1</code> set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L184-L195">source</a></section></article><h2 id="SOS2"><a class="docs-heading-anchor" href="#SOS2"><code>SOS2</code></a><a id="SOS2-1"></a><a class="docs-heading-anchor-permalink" href="#SOS2" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOS2" href="#JuMP.SOS2"><code>JuMP.SOS2</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SOS2</code></pre><p>SOS1 (Special Ordered Sets type 2) object than can be used to constrain a vector <code>x</code> to a set where at most 2 variables can take a non-zero value, all others being at 0. In addition, if two are non-zero these must be consecutive in their ordering. The <code>weights</code> induce an ordering of the variables; as such, they should be unique values. The <em>k</em>th element in the set corresponds to the <em>k</em>th weight in <code>weights</code>. See <a href="http://lpsolve.sourceforge.net/5.5/SOS.htm">here</a> for a description of SOS constraints and their potential uses. This is a shortcut for the <code>MathOptInterface.SOS2</code> set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L215-L227">source</a></section></article><h2 id="ScalarConstraint"><a class="docs-heading-anchor" href="#ScalarConstraint"><code>ScalarConstraint</code></a><a id="ScalarConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarConstraint" href="#JuMP.ScalarConstraint"><code>JuMP.ScalarConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct ScalarConstraint</code></pre><p>The data for a scalar constraint. The <code>func</code> field contains a JuMP object representing the function and the <code>set</code> field contains the MOI set. See also the <a href="../../moi/manual/constraints/#Constraints">documentation</a> on JuMP&#39;s representation of constraints for more background.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L577-L584">source</a></section></article><h2 id="ScalarShape"><a class="docs-heading-anchor" href="#ScalarShape"><code>ScalarShape</code></a><a id="ScalarShape-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarShape" href="#JuMP.ScalarShape"><code>JuMP.ScalarShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarShape</code></pre><p>Shape of scalar constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L108-L112">source</a></section></article><h2 id="ScalarVariable"><a class="docs-heading-anchor" href="#ScalarVariable"><code>ScalarVariable</code></a><a id="ScalarVariable-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarVariable" href="#JuMP.ScalarVariable"><code>JuMP.ScalarVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarVariable{S,T,U,V} &lt;: AbstractVariable</code></pre><p>A struct used when adding variables.</p><p>See also: <a href="#JuMP.add_variable"><code>add_variable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L199-L205">source</a></section></article><h2 id="SecondOrderCone"><a class="docs-heading-anchor" href="#SecondOrderCone"><code>SecondOrderCone</code></a><a id="SecondOrderCone-1"></a><a class="docs-heading-anchor-permalink" href="#SecondOrderCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SecondOrderCone" href="#JuMP.SecondOrderCone"><code>JuMP.SecondOrderCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SecondOrderCone</code></pre><p>Second order cone object that can be used to constrain the euclidean norm of a vector <code>x</code> to be less than or equal to a nonnegative scalar <code>t</code>. This is a shortcut for the <code>MOI.SecondOrderCone</code>.</p><p><strong>Example</strong></p><p>The following constrains <span>$\|(x-1, x-2)\|_2 \le t$</span> and <span>$t \ge 0$</span>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[t, x, x - 1, x - 2] ∈ MathOptInterface.RotatedSecondOrderCone(4)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L157-L180">source</a></section></article><h2 id="SOS1"><a class="docs-heading-anchor" href="#SOS1"><code>SOS1</code></a><a id="SOS1-1"></a><a class="docs-heading-anchor-permalink" href="#SOS1" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOS1" href="#JuMP.SOS1"><code>JuMP.SOS1</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SOS1</code></pre><p>SOS1 (Special Ordered Sets type 1) object than can be used to constrain a vector <code>x</code> to a set where at most 1 variable can take a non-zero value, all others being at 0. The <code>weights</code>, when specified, induce an ordering of the variables; as such, they should be unique values. The <em>k</em>th element in the set corresponds to the <em>k</em>th weight in <code>weights</code>. See <a href="http://lpsolve.sourceforge.net/5.5/SOS.htm">here</a> for a description of SOS constraints and their potential uses. This is a shortcut for the <code>MathOptInterface.SOS1</code> set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L184-L195">source</a></section></article><h2 id="SOS2"><a class="docs-heading-anchor" href="#SOS2"><code>SOS2</code></a><a id="SOS2-1"></a><a class="docs-heading-anchor-permalink" href="#SOS2" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOS2" href="#JuMP.SOS2"><code>JuMP.SOS2</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SOS2</code></pre><p>SOS1 (Special Ordered Sets type 2) object than can be used to constrain a vector <code>x</code> to a set where at most 2 variables can take a non-zero value, all others being at 0. In addition, if two are non-zero these must be consecutive in their ordering. The <code>weights</code> induce an ordering of the variables; as such, they should be unique values. The <em>k</em>th element in the set corresponds to the <em>k</em>th weight in <code>weights</code>. See <a href="http://lpsolve.sourceforge.net/5.5/SOS.htm">here</a> for a description of SOS constraints and their potential uses. This is a shortcut for the <code>MathOptInterface.SOS2</code> set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L215-L227">source</a></section></article><h2 id="ScalarConstraint"><a class="docs-heading-anchor" href="#ScalarConstraint"><code>ScalarConstraint</code></a><a id="ScalarConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarConstraint" href="#JuMP.ScalarConstraint"><code>JuMP.ScalarConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct ScalarConstraint</code></pre><p>The data for a scalar constraint. The <code>func</code> field contains a JuMP object representing the function and the <code>set</code> field contains the MOI set. See also the <a href="../../moi/manual/constraints/#Constraints">documentation</a> on JuMP&#39;s representation of constraints for more background.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L577-L584">source</a></section></article><h2 id="ScalarShape"><a class="docs-heading-anchor" href="#ScalarShape"><code>ScalarShape</code></a><a id="ScalarShape-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarShape" href="#JuMP.ScalarShape"><code>JuMP.ScalarShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarShape</code></pre><p>Shape of scalar constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L108-L112">source</a></section></article><h2 id="ScalarVariable"><a class="docs-heading-anchor" href="#ScalarVariable"><code>ScalarVariable</code></a><a id="ScalarVariable-1"></a><a class="docs-heading-anchor-permalink" href="#ScalarVariable" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ScalarVariable" href="#JuMP.ScalarVariable"><code>JuMP.ScalarVariable</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarVariable{S,T,U,V} &lt;: AbstractVariable</code></pre><p>A struct used when adding variables.</p><p>See also: <a href="#JuMP.add_variable"><code>add_variable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L199-L205">source</a></section></article><h2 id="SecondOrderCone"><a class="docs-heading-anchor" href="#SecondOrderCone"><code>SecondOrderCone</code></a><a id="SecondOrderCone-1"></a><a class="docs-heading-anchor-permalink" href="#SecondOrderCone" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SecondOrderCone" href="#JuMP.SecondOrderCone"><code>JuMP.SecondOrderCone</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SecondOrderCone</code></pre><p>Second order cone object that can be used to constrain the euclidean norm of a vector <code>x</code> to be less than or equal to a nonnegative scalar <code>t</code>. This is a shortcut for the <code>MOI.SecondOrderCone</code>.</p><p><strong>Example</strong></p><p>The following constrains <span>$\|(x-1, x-2)\|_2 \le t$</span> and <span>$t \ge 0$</span>:</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -1931,7 +1931,7 @@
 t
 
 julia&gt; @constraint(model, [t, x-1, x-2] in SecondOrderCone())
-[t, x - 1, x - 2] ∈ MathOptInterface.SecondOrderCone(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L131-L153">source</a></section></article><h2 id="Semicontinuous"><a class="docs-heading-anchor" href="#Semicontinuous"><code>Semicontinuous</code></a><a id="Semicontinuous-1"></a><a class="docs-heading-anchor-permalink" href="#Semicontinuous" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Semicontinuous" href="#JuMP.Semicontinuous"><code>JuMP.Semicontinuous</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Semicontinuous(lower, upper)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous</code></a> set.</p><p>This short-cut is useful because it automatically promotes <code>lower</code> and <code>upper</code> to the same type, and converts them into the element type supported by the JuMP model.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[t, x - 1, x - 2] ∈ MathOptInterface.SecondOrderCone(3)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L131-L153">source</a></section></article><h2 id="Semicontinuous"><a class="docs-heading-anchor" href="#Semicontinuous"><code>Semicontinuous</code></a><a id="Semicontinuous-1"></a><a class="docs-heading-anchor-permalink" href="#Semicontinuous" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Semicontinuous" href="#JuMP.Semicontinuous"><code>JuMP.Semicontinuous</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Semicontinuous(lower, upper)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous</code></a> set.</p><p>This short-cut is useful because it automatically promotes <code>lower</code> and <code>upper</code> to the same type, and converts them into the element type supported by the JuMP model.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x in Semicontinuous(1, 2))
 x
@@ -1939,7 +1939,7 @@
 julia&gt; print(model)
 Feasibility
 Subject to
- x ∈ MathOptInterface.Semicontinuous{Int64}(1, 2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L292-L314">source</a></section></article><h2 id="Semiinteger"><a class="docs-heading-anchor" href="#Semiinteger"><code>Semiinteger</code></a><a id="Semiinteger-1"></a><a class="docs-heading-anchor-permalink" href="#Semiinteger" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Semiinteger" href="#JuMP.Semiinteger"><code>JuMP.Semiinteger</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Semiinteger(lower, upper)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger</code></a> set.</p><p>This short-cut is useful because it automatically promotes <code>lower</code> and <code>upper</code> to the same type, and converts them into the element type supported by the JuMP model.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ x ∈ MathOptInterface.Semicontinuous{Int64}(1, 2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L292-L314">source</a></section></article><h2 id="Semiinteger"><a class="docs-heading-anchor" href="#Semiinteger"><code>Semiinteger</code></a><a id="Semiinteger-1"></a><a class="docs-heading-anchor-permalink" href="#Semiinteger" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Semiinteger" href="#JuMP.Semiinteger"><code>JuMP.Semiinteger</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Semiinteger(lower, upper)</code></pre><p>A short-cut for the <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger</code></a> set.</p><p>This short-cut is useful because it automatically promotes <code>lower</code> and <code>upper</code> to the same type, and converts them into the element type supported by the JuMP model.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x in Semiinteger(3, 5))
 x
@@ -1947,12 +1947,12 @@
 julia&gt; print(model)
 Feasibility
 Subject to
- x ∈ MathOptInterface.Semiinteger{Int64}(3, 5)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sets.jl#L328-L350">source</a></section></article><h2 id="SensitivityReport"><a class="docs-heading-anchor" href="#SensitivityReport"><code>SensitivityReport</code></a><a id="SensitivityReport-1"></a><a class="docs-heading-anchor-permalink" href="#SensitivityReport" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SensitivityReport" href="#JuMP.SensitivityReport"><code>JuMP.SensitivityReport</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SensitivityReport</code></pre><p>See <a href="#JuMP.lp_sensitivity_report"><code>lp_sensitivity_report</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/lp_sensitivity2.jl#L6-L10">source</a></section></article><h2 id="SkewSymmetricMatrixShape"><a class="docs-heading-anchor" href="#SkewSymmetricMatrixShape"><code>SkewSymmetricMatrixShape</code></a><a id="SkewSymmetricMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SkewSymmetricMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkewSymmetricMatrixShape" href="#JuMP.SkewSymmetricMatrixShape"><code>JuMP.SkewSymmetricMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkewSymmetricMatrixShape</code></pre><p>Shape object for a skew symmetric square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix (without the diagonal) given column by column (or equivalently, the entries of the lower-left triangular part given row by row). The diagonal is zero.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L201-L209">source</a></section></article><h2 id="SkewSymmetricMatrixSpace"><a class="docs-heading-anchor" href="#SkewSymmetricMatrixSpace"><code>SkewSymmetricMatrixSpace</code></a><a id="SkewSymmetricMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#SkewSymmetricMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkewSymmetricMatrixSpace" href="#JuMP.SkewSymmetricMatrixSpace"><code>JuMP.SkewSymmetricMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkewSymmetricMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be skew-symmetric.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ x ∈ MathOptInterface.Semiinteger{Int64}(3, 5)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sets.jl#L328-L350">source</a></section></article><h2 id="SensitivityReport"><a class="docs-heading-anchor" href="#SensitivityReport"><code>SensitivityReport</code></a><a id="SensitivityReport-1"></a><a class="docs-heading-anchor-permalink" href="#SensitivityReport" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SensitivityReport" href="#JuMP.SensitivityReport"><code>JuMP.SensitivityReport</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SensitivityReport</code></pre><p>See <a href="#JuMP.lp_sensitivity_report"><code>lp_sensitivity_report</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/lp_sensitivity2.jl#L6-L10">source</a></section></article><h2 id="SkewSymmetricMatrixShape"><a class="docs-heading-anchor" href="#SkewSymmetricMatrixShape"><code>SkewSymmetricMatrixShape</code></a><a id="SkewSymmetricMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SkewSymmetricMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkewSymmetricMatrixShape" href="#JuMP.SkewSymmetricMatrixShape"><code>JuMP.SkewSymmetricMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkewSymmetricMatrixShape</code></pre><p>Shape object for a skew symmetric square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix (without the diagonal) given column by column (or equivalently, the entries of the lower-left triangular part given row by row). The diagonal is zero.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L201-L209">source</a></section></article><h2 id="SkewSymmetricMatrixSpace"><a class="docs-heading-anchor" href="#SkewSymmetricMatrixSpace"><code>SkewSymmetricMatrixSpace</code></a><a id="SkewSymmetricMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#SkewSymmetricMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkewSymmetricMatrixSpace" href="#JuMP.SkewSymmetricMatrixSpace"><code>JuMP.SkewSymmetricMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkewSymmetricMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be skew-symmetric.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())
 2×2 Matrix{AffExpr}:
  0        Q[1,2]
- -Q[1,2]  0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L37-L53">source</a></section></article><h2 id="SkipModelConvertScalarSetWrapper"><a class="docs-heading-anchor" href="#SkipModelConvertScalarSetWrapper"><code>SkipModelConvertScalarSetWrapper</code></a><a id="SkipModelConvertScalarSetWrapper-1"></a><a class="docs-heading-anchor-permalink" href="#SkipModelConvertScalarSetWrapper" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkipModelConvertScalarSetWrapper" href="#JuMP.SkipModelConvertScalarSetWrapper"><code>JuMP.SkipModelConvertScalarSetWrapper</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkipModelConvertScalarSetWrapper(set::MOI.AbstractScalarSet)</code></pre><p>JuMP uses <a href="#JuMP.model_convert"><code>model_convert</code></a> to automatically promote <a href="../../moi/reference/standard_form/#MathOptInterface.AbstractScalarSet"><code>MOI.AbstractScalarSet</code></a> sets to the same <a href="#JuMP.value_type"><code>value_type</code></a> as the model.</p><p>In cases there this is undesirable, wrap the set in <code>SkipModelConvertScalarSetWrapper</code> to pass the set un-changed to the solver.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ -Q[1,2]  0</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L37-L53">source</a></section></article><h2 id="SkipModelConvertScalarSetWrapper"><a class="docs-heading-anchor" href="#SkipModelConvertScalarSetWrapper"><code>SkipModelConvertScalarSetWrapper</code></a><a id="SkipModelConvertScalarSetWrapper-1"></a><a class="docs-heading-anchor-permalink" href="#SkipModelConvertScalarSetWrapper" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SkipModelConvertScalarSetWrapper" href="#JuMP.SkipModelConvertScalarSetWrapper"><code>JuMP.SkipModelConvertScalarSetWrapper</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SkipModelConvertScalarSetWrapper(set::MOI.AbstractScalarSet)</code></pre><p>JuMP uses <a href="#JuMP.model_convert"><code>model_convert</code></a> to automatically promote <a href="../../moi/reference/standard_form/#MathOptInterface.AbstractScalarSet"><code>MOI.AbstractScalarSet</code></a> sets to the same <a href="#JuMP.value_type"><code>value_type</code></a> as the model.</p><p>In cases there this is undesirable, wrap the set in <code>SkipModelConvertScalarSetWrapper</code> to pass the set un-changed to the solver.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -1960,17 +1960,17 @@
 x = 0.5
 
 julia&gt; @constraint(model, x in SkipModelConvertScalarSetWrapper(MOI.EqualTo(1 // 2)))
-x = 1//2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L1546-L1568">source</a></section></article><h2 id="SquareMatrixShape"><a class="docs-heading-anchor" href="#SquareMatrixShape"><code>SquareMatrixShape</code></a><a id="SquareMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SquareMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SquareMatrixShape" href="#JuMP.SquareMatrixShape"><code>JuMP.SquareMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SquareMatrixShape</code></pre><p>Shape object for a square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L240-L247">source</a></section></article><h2 id="SymmetricMatrixShape"><a class="docs-heading-anchor" href="#SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a><a id="SymmetricMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SymmetricMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SymmetricMatrixShape" href="#JuMP.SymmetricMatrixShape"><code>JuMP.SymmetricMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SymmetricMatrixShape</code></pre><p>Shape object for a symmetric square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L140-L147">source</a></section></article><h2 id="SymmetricMatrixSpace"><a class="docs-heading-anchor" href="#SymmetricMatrixSpace"><code>SymmetricMatrixSpace</code></a><a id="SymmetricMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#SymmetricMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SymmetricMatrixSpace" href="#JuMP.SymmetricMatrixSpace"><code>JuMP.SymmetricMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SymmetricMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be symmetric.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x = 1//2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L1546-L1568">source</a></section></article><h2 id="SquareMatrixShape"><a class="docs-heading-anchor" href="#SquareMatrixShape"><code>SquareMatrixShape</code></a><a id="SquareMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SquareMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SquareMatrixShape" href="#JuMP.SquareMatrixShape"><code>JuMP.SquareMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SquareMatrixShape</code></pre><p>Shape object for a square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L240-L247">source</a></section></article><h2 id="SymmetricMatrixShape"><a class="docs-heading-anchor" href="#SymmetricMatrixShape"><code>SymmetricMatrixShape</code></a><a id="SymmetricMatrixShape-1"></a><a class="docs-heading-anchor-permalink" href="#SymmetricMatrixShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SymmetricMatrixShape" href="#JuMP.SymmetricMatrixShape"><code>JuMP.SymmetricMatrixShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SymmetricMatrixShape</code></pre><p>Shape object for a symmetric square matrix of <code>side_dimension</code> rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L140-L147">source</a></section></article><h2 id="SymmetricMatrixSpace"><a class="docs-heading-anchor" href="#SymmetricMatrixSpace"><code>SymmetricMatrixSpace</code></a><a id="SymmetricMatrixSpace-1"></a><a class="docs-heading-anchor-permalink" href="#SymmetricMatrixSpace" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SymmetricMatrixSpace" href="#JuMP.SymmetricMatrixSpace"><code>JuMP.SymmetricMatrixSpace</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SymmetricMatrixSpace()</code></pre><p>Use in the <a href="#JuMP.@variable"><code>@variable</code></a> macro to constrain a matrix of variables to be symmetric.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, Q[1:2, 1:2] in SymmetricMatrixSpace())
 2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:
  Q[1,1]  Q[1,2]
- Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/sd.jl#L18-L34">source</a></section></article><h2 id="TerminationStatusCode"><a class="docs-heading-anchor" href="#TerminationStatusCode"><code>TerminationStatusCode</code></a><a id="TerminationStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#TerminationStatusCode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.TerminationStatusCode" href="#JuMP.TerminationStatusCode"><code>JuMP.TerminationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatusCode</code></pre><p>An Enum of possible values for the <code>TerminationStatus</code> attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to <a href="../../moi/reference/models/#MathOptInterface.optimize!"><code>optimize!</code></a>.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a>: The algorithm has not started.</li><li><a href="../../moi/reference/models/#MathOptInterface.OPTIMAL"><code>OPTIMAL</code></a>: The algorithm found a globally optimal solution.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE"><code>INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists.</li><li><a href="../../moi/reference/models/#MathOptInterface.DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</li><li><a href="../../moi/reference/models/#MathOptInterface.LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</li><li><a href="../../moi/reference/models/#MathOptInterface.LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a>: The algorithm found a globally optimal solution to relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a>: An iterative algorithm stopped after conducting the maximum number of iterations.</li><li><a href="../../moi/reference/models/#MathOptInterface.TIME_LIMIT"><code>TIME_LIMIT</code></a>: The algorithm stopped after a user-specified computation time.</li><li><a href="../../moi/reference/models/#MathOptInterface.NODE_LIMIT"><code>NODE_LIMIT</code></a>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</li><li><a href="../../moi/reference/models/#MathOptInterface.SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</li><li><a href="../../moi/reference/models/#MathOptInterface.MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a>: The algorithm stopped because it ran out of memory.</li><li><a href="../../moi/reference/models/#MathOptInterface.OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</li><li><a href="../../moi/reference/models/#MathOptInterface.NORM_LIMIT"><code>NORM_LIMIT</code></a>: The algorithm stopped because the norm of an iterate became too large.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_LIMIT"><code>OTHER_LIMIT</code></a>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</li><li><a href="../../moi/reference/models/#MathOptInterface.SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a>: The algorithm stopped because it was unable to continue making progress towards the solution.</li><li><a href="../../moi/reference/models/#MathOptInterface.NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a>: The algorithm stopped because it encountered unrecoverable numerical error.</li><li><a href="../../moi/reference/models/#MathOptInterface.INVALID_MODEL"><code>INVALID_MODEL</code></a>: The algorithm stopped because the model is invalid.</li><li><a href="../../moi/reference/models/#MathOptInterface.INVALID_OPTION"><code>INVALID_OPTION</code></a>: The algorithm stopped because it was provided an invalid option.</li><li><a href="../../moi/reference/models/#MathOptInterface.INTERRUPTED"><code>INTERRUPTED</code></a>: The algorithm stopped because of an interrupt signal.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_ERROR"><code>OTHER_ERROR</code></a>: The algorithm stopped because of an error not covered by one of the statuses defined above.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L2009">source</a></section></article><h2 id="UnorderedPair"><a class="docs-heading-anchor" href="#UnorderedPair"><code>UnorderedPair</code></a><a id="UnorderedPair-1"></a><a class="docs-heading-anchor-permalink" href="#UnorderedPair" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UnorderedPair" href="#JuMP.UnorderedPair"><code>JuMP.UnorderedPair</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">UnorderedPair(a::T, b::T)</code></pre><p>A wrapper type used by <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> with fields <code>.a</code> and <code>.b</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/quad_expr.jl#L18-L22">source</a></section></article><h2 id="UpperBoundRef"><a class="docs-heading-anchor" href="#UpperBoundRef"><code>UpperBoundRef</code></a><a id="UpperBoundRef-1"></a><a class="docs-heading-anchor-permalink" href="#UpperBoundRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UpperBoundRef" href="#JuMP.UpperBoundRef"><code>JuMP.UpperBoundRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">UpperBoundRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the upper bound constraint of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+ Q[1,2]  Q[2,2]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/sd.jl#L18-L34">source</a></section></article><h2 id="TerminationStatusCode"><a class="docs-heading-anchor" href="#TerminationStatusCode"><code>TerminationStatusCode</code></a><a id="TerminationStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#TerminationStatusCode" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.TerminationStatusCode" href="#JuMP.TerminationStatusCode"><code>JuMP.TerminationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatusCode</code></pre><p>An Enum of possible values for the <code>TerminationStatus</code> attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to <a href="../../moi/reference/models/#MathOptInterface.optimize!"><code>optimize!</code></a>.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a>: The algorithm has not started.</li><li><a href="../../moi/reference/models/#MathOptInterface.OPTIMAL"><code>OPTIMAL</code></a>: The algorithm found a globally optimal solution.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE"><code>INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists.</li><li><a href="../../moi/reference/models/#MathOptInterface.DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</li><li><a href="../../moi/reference/models/#MathOptInterface.LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</li><li><a href="../../moi/reference/models/#MathOptInterface.LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</li><li><a href="../../moi/reference/models/#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a>: The algorithm found a globally optimal solution to relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</li><li><a href="../../moi/reference/models/#MathOptInterface.ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a>: An iterative algorithm stopped after conducting the maximum number of iterations.</li><li><a href="../../moi/reference/models/#MathOptInterface.TIME_LIMIT"><code>TIME_LIMIT</code></a>: The algorithm stopped after a user-specified computation time.</li><li><a href="../../moi/reference/models/#MathOptInterface.NODE_LIMIT"><code>NODE_LIMIT</code></a>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</li><li><a href="../../moi/reference/models/#MathOptInterface.SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</li><li><a href="../../moi/reference/models/#MathOptInterface.MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a>: The algorithm stopped because it ran out of memory.</li><li><a href="../../moi/reference/models/#MathOptInterface.OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</li><li><a href="../../moi/reference/models/#MathOptInterface.NORM_LIMIT"><code>NORM_LIMIT</code></a>: The algorithm stopped because the norm of an iterate became too large.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_LIMIT"><code>OTHER_LIMIT</code></a>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</li><li><a href="../../moi/reference/models/#MathOptInterface.SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a>: The algorithm stopped because it was unable to continue making progress towards the solution.</li><li><a href="../../moi/reference/models/#MathOptInterface.NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a>: The algorithm stopped because it encountered unrecoverable numerical error.</li><li><a href="../../moi/reference/models/#MathOptInterface.INVALID_MODEL"><code>INVALID_MODEL</code></a>: The algorithm stopped because the model is invalid.</li><li><a href="../../moi/reference/models/#MathOptInterface.INVALID_OPTION"><code>INVALID_OPTION</code></a>: The algorithm stopped because it was provided an invalid option.</li><li><a href="../../moi/reference/models/#MathOptInterface.INTERRUPTED"><code>INTERRUPTED</code></a>: The algorithm stopped because of an interrupt signal.</li><li><a href="../../moi/reference/models/#MathOptInterface.OTHER_ERROR"><code>OTHER_ERROR</code></a>: The algorithm stopped because of an error not covered by one of the statuses defined above.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L2009">source</a></section></article><h2 id="UnorderedPair"><a class="docs-heading-anchor" href="#UnorderedPair"><code>UnorderedPair</code></a><a id="UnorderedPair-1"></a><a class="docs-heading-anchor-permalink" href="#UnorderedPair" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UnorderedPair" href="#JuMP.UnorderedPair"><code>JuMP.UnorderedPair</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">UnorderedPair(a::T, b::T)</code></pre><p>A wrapper type used by <a href="#JuMP.GenericQuadExpr"><code>GenericQuadExpr</code></a> with fields <code>.a</code> and <code>.b</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/quad_expr.jl#L18-L22">source</a></section></article><h2 id="UpperBoundRef"><a class="docs-heading-anchor" href="#UpperBoundRef"><code>UpperBoundRef</code></a><a id="UpperBoundRef-1"></a><a class="docs-heading-anchor-permalink" href="#UpperBoundRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UpperBoundRef" href="#JuMP.UpperBoundRef"><code>JuMP.UpperBoundRef</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">UpperBoundRef(v::GenericVariableRef)</code></pre><p>Return a constraint reference to the upper bound constraint of <code>v</code>.</p><p>Errors if one does not exist.</p><p>See also <a href="#JuMP.has_upper_bound"><code>has_upper_bound</code></a>, <a href="#JuMP.upper_bound"><code>upper_bound</code></a>, <a href="#JuMP.set_upper_bound"><code>set_upper_bound</code></a>, <a href="#JuMP.delete_upper_bound"><code>delete_upper_bound</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x &lt;= 1.0);
 
 julia&gt; UpperBoundRef(x)
-x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L930-L950">source</a></section></article><h2 id="VariableConstrainedOnCreation"><a class="docs-heading-anchor" href="#VariableConstrainedOnCreation"><code>VariableConstrainedOnCreation</code></a><a id="VariableConstrainedOnCreation-1"></a><a class="docs-heading-anchor-permalink" href="#VariableConstrainedOnCreation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableConstrainedOnCreation" href="#JuMP.VariableConstrainedOnCreation"><code>JuMP.VariableConstrainedOnCreation</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableConstrainedOnCreation &lt;: AbstractVariable</code></pre><p>Variable <code>scalar_variables</code> constrained to belong to <code>set</code>.</p><p>Adding this variable can be understood as doing:</p><pre><code class="language-julia hljs">function JuMP.add_variable(
+x ≤ 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L930-L950">source</a></section></article><h2 id="VariableConstrainedOnCreation"><a class="docs-heading-anchor" href="#VariableConstrainedOnCreation"><code>VariableConstrainedOnCreation</code></a><a id="VariableConstrainedOnCreation-1"></a><a class="docs-heading-anchor-permalink" href="#VariableConstrainedOnCreation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableConstrainedOnCreation" href="#JuMP.VariableConstrainedOnCreation"><code>JuMP.VariableConstrainedOnCreation</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableConstrainedOnCreation &lt;: AbstractVariable</code></pre><p>Variable <code>scalar_variables</code> constrained to belong to <code>set</code>.</p><p>Adding this variable can be understood as doing:</p><pre><code class="language-julia hljs">function JuMP.add_variable(
     model::GenericModel,
     variable::VariableConstrainedOnCreation,
     names,
@@ -1978,9 +1978,9 @@
     var_ref = add_variable(model, variable.scalar_variable, name)
     add_constraint(model, VectorConstraint(var_ref, variable.set))
     return var_ref
-end</code></pre><p>but adds the variables with <code>MOI.add_constrained_variable(model, variable.set)</code> instead. See <a href="https://jump.dev/MathOptInterface.jl/v0.9.3/apireference/#Variables-1">the MOI documentation</a> for the difference between adding the variables with <code>MOI.add_constrained_variable</code> and adding them with <code>MOI.add_variable</code> and adding the constraint separately.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1798-L1819">source</a></section></article><h2 id="VariableInfo"><a class="docs-heading-anchor" href="#VariableInfo"><code>VariableInfo</code></a><a id="VariableInfo-1"></a><a class="docs-heading-anchor-permalink" href="#VariableInfo" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableInfo" href="#JuMP.VariableInfo"><code>JuMP.VariableInfo</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableInfo{S,T,U,V}</code></pre><p>A struct by JuMP internally when creating variables. This may also be used by JuMP extensions to create new types of variables.</p><p>See also: <a href="#JuMP.ScalarVariable"><code>ScalarVariable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L127-L134">source</a></section></article><h2 id="VariableNotOwned"><a class="docs-heading-anchor" href="#VariableNotOwned"><code>VariableNotOwned</code></a><a id="VariableNotOwned-1"></a><a class="docs-heading-anchor-permalink" href="#VariableNotOwned" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableNotOwned" href="#JuMP.VariableNotOwned"><code>JuMP.VariableNotOwned</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VariableNotOwned{V&lt;:AbstractVariableRef} &lt;: Exception
+end</code></pre><p>but adds the variables with <code>MOI.add_constrained_variable(model, variable.set)</code> instead. See <a href="https://jump.dev/MathOptInterface.jl/v0.9.3/apireference/#Variables-1">the MOI documentation</a> for the difference between adding the variables with <code>MOI.add_constrained_variable</code> and adding them with <code>MOI.add_variable</code> and adding the constraint separately.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1798-L1819">source</a></section></article><h2 id="VariableInfo"><a class="docs-heading-anchor" href="#VariableInfo"><code>VariableInfo</code></a><a id="VariableInfo-1"></a><a class="docs-heading-anchor-permalink" href="#VariableInfo" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableInfo" href="#JuMP.VariableInfo"><code>JuMP.VariableInfo</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableInfo{S,T,U,V}</code></pre><p>A struct by JuMP internally when creating variables. This may also be used by JuMP extensions to create new types of variables.</p><p>See also: <a href="#JuMP.ScalarVariable"><code>ScalarVariable</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L127-L134">source</a></section></article><h2 id="VariableNotOwned"><a class="docs-heading-anchor" href="#VariableNotOwned"><code>VariableNotOwned</code></a><a id="VariableNotOwned-1"></a><a class="docs-heading-anchor-permalink" href="#VariableNotOwned" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableNotOwned" href="#JuMP.VariableNotOwned"><code>JuMP.VariableNotOwned</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VariableNotOwned{V&lt;:AbstractVariableRef} &lt;: Exception
     variable::V
-end</code></pre><p>The variable <code>variable</code> was used in a model different to <code>owner_model(variable)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L282-L289">source</a></section></article><h2 id="VariableRef"><a class="docs-heading-anchor" href="#VariableRef"><code>VariableRef</code></a><a id="VariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#VariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableRef" href="#JuMP.VariableRef"><code>JuMP.VariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericVariableRef{T} &lt;: AbstractVariableRef</code></pre><p>Holds a reference to the model and the corresponding MOI.VariableIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L245-L249">source</a></section></article><h2 id="VariablesConstrainedOnCreation"><a class="docs-heading-anchor" href="#VariablesConstrainedOnCreation"><code>VariablesConstrainedOnCreation</code></a><a id="VariablesConstrainedOnCreation-1"></a><a class="docs-heading-anchor-permalink" href="#VariablesConstrainedOnCreation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariablesConstrainedOnCreation" href="#JuMP.VariablesConstrainedOnCreation"><code>JuMP.VariablesConstrainedOnCreation</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablesConstrainedOnCreation &lt;: AbstractVariable</code></pre><p>Vector of variables <code>scalar_variables</code> constrained to belong to <code>set</code>. Adding this variable can be thought as doing:</p><pre><code class="language-julia hljs">function JuMP.add_variable(
+end</code></pre><p>The variable <code>variable</code> was used in a model different to <code>owner_model(variable)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L282-L289">source</a></section></article><h2 id="VariableRef"><a class="docs-heading-anchor" href="#VariableRef"><code>VariableRef</code></a><a id="VariableRef-1"></a><a class="docs-heading-anchor-permalink" href="#VariableRef" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariableRef" href="#JuMP.VariableRef"><code>JuMP.VariableRef</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">GenericVariableRef{T} &lt;: AbstractVariableRef</code></pre><p>Holds a reference to the model and the corresponding MOI.VariableIndex.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L245-L249">source</a></section></article><h2 id="VariablesConstrainedOnCreation"><a class="docs-heading-anchor" href="#VariablesConstrainedOnCreation"><code>VariablesConstrainedOnCreation</code></a><a id="VariablesConstrainedOnCreation-1"></a><a class="docs-heading-anchor-permalink" href="#VariablesConstrainedOnCreation" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VariablesConstrainedOnCreation" href="#JuMP.VariablesConstrainedOnCreation"><code>JuMP.VariablesConstrainedOnCreation</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablesConstrainedOnCreation &lt;: AbstractVariable</code></pre><p>Vector of variables <code>scalar_variables</code> constrained to belong to <code>set</code>. Adding this variable can be thought as doing:</p><pre><code class="language-julia hljs">function JuMP.add_variable(
     model::GenericModel,
     variable::VariablesConstrainedOnCreation,
     names,
@@ -1989,7 +1989,7 @@
     var_refs = add_variable.(model, variable.scalar_variables, v_names)
     add_constraint(model, VectorConstraint(var_refs, variable.set))
     return reshape_vector(var_refs, variable.shape)
-end</code></pre><p>but adds the variables with <code>MOI.add_constrained_variables(model, variable.set)</code> instead. See <a href="https://jump.dev/MathOptInterface.jl/v0.9.3/apireference/#Variables-1">the MOI documentation</a> for the difference between adding the variables with <code>MOI.add_constrained_variables</code> and adding them with <code>MOI.add_variables</code> and adding the constraint separately.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/variables.jl#L1874-L1895">source</a></section></article><h2 id="VectorConstraint"><a class="docs-heading-anchor" href="#VectorConstraint"><code>VectorConstraint</code></a><a id="VectorConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#VectorConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VectorConstraint" href="#JuMP.VectorConstraint"><code>JuMP.VectorConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VectorConstraint</code></pre><p>The data for a vector constraint. The <code>func</code> field contains a JuMP object representing the function and the <code>set</code> field contains the MOI set. The <code>shape</code> field contains an <a href="#JuMP.AbstractShape"><code>AbstractShape</code></a> matching the form in which the constraint was constructed (e.g., by using matrices or flat vectors). See also the <a href="../../moi/manual/constraints/#Constraints">documentation</a> on JuMP&#39;s representation of constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/constraints.jl#L611-L620">source</a></section></article><h2 id="VectorShape"><a class="docs-heading-anchor" href="#VectorShape"><code>VectorShape</code></a><a id="VectorShape-1"></a><a class="docs-heading-anchor-permalink" href="#VectorShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VectorShape" href="#JuMP.VectorShape"><code>JuMP.VectorShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VectorShape</code></pre><p>Vector for which the vectorized form corresponds exactly to the vector given.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/shapes.jl#L116-L120">source</a></section></article><h2 id="Zeros"><a class="docs-heading-anchor" href="#Zeros"><code>Zeros</code></a><a id="Zeros-1"></a><a class="docs-heading-anchor-permalink" href="#Zeros" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Zeros" href="#JuMP.Zeros"><code>JuMP.Zeros</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Zeros()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Zeros"><code>MOI.Zeros</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+end</code></pre><p>but adds the variables with <code>MOI.add_constrained_variables(model, variable.set)</code> instead. See <a href="https://jump.dev/MathOptInterface.jl/v0.9.3/apireference/#Variables-1">the MOI documentation</a> for the difference between adding the variables with <code>MOI.add_constrained_variables</code> and adding them with <code>MOI.add_variables</code> and adding the constraint separately.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/variables.jl#L1874-L1895">source</a></section></article><h2 id="VectorConstraint"><a class="docs-heading-anchor" href="#VectorConstraint"><code>VectorConstraint</code></a><a id="VectorConstraint-1"></a><a class="docs-heading-anchor-permalink" href="#VectorConstraint" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VectorConstraint" href="#JuMP.VectorConstraint"><code>JuMP.VectorConstraint</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">struct VectorConstraint</code></pre><p>The data for a vector constraint. The <code>func</code> field contains a JuMP object representing the function and the <code>set</code> field contains the MOI set. The <code>shape</code> field contains an <a href="#JuMP.AbstractShape"><code>AbstractShape</code></a> matching the form in which the constraint was constructed (e.g., by using matrices or flat vectors). See also the <a href="../../moi/manual/constraints/#Constraints">documentation</a> on JuMP&#39;s representation of constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/constraints.jl#L611-L620">source</a></section></article><h2 id="VectorShape"><a class="docs-heading-anchor" href="#VectorShape"><code>VectorShape</code></a><a id="VectorShape-1"></a><a class="docs-heading-anchor-permalink" href="#VectorShape" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.VectorShape" href="#JuMP.VectorShape"><code>JuMP.VectorShape</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VectorShape</code></pre><p>Vector for which the vectorized form corresponds exactly to the vector given.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/shapes.jl#L116-L120">source</a></section></article><h2 id="Zeros"><a class="docs-heading-anchor" href="#Zeros"><code>Zeros</code></a><a id="Zeros-1"></a><a class="docs-heading-anchor-permalink" href="#Zeros" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.Zeros" href="#JuMP.Zeros"><code>JuMP.Zeros</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">Zeros()</code></pre><p>The JuMP equivalent of the <a href="../../moi/reference/standard_form/#MathOptInterface.Zeros"><code>MOI.Zeros</code></a> set, in which the dimension is inferred from the corresponding function.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x[1:2])
 2-element Vector{VariableRef}:
@@ -2004,7 +2004,7 @@
 julia&gt; b = [5, 6];
 
 julia&gt; @constraint(model, A * x == b)
-[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L293-L319">source</a></section></article><h2 id="ALMOST_DUAL_INFEASIBLE"><a class="docs-heading-anchor" href="#ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a><a id="ALMOST_DUAL_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_DUAL_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_DUAL_INFEASIBLE" href="#JuMP.ALMOST_DUAL_INFEASIBLE"><code>JuMP.ALMOST_DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_INFEASIBLE"><a class="docs-heading-anchor" href="#ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a><a id="ALMOST_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_INFEASIBLE" href="#JuMP.ALMOST_INFEASIBLE"><code>JuMP.ALMOST_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_INFEASIBLE</code>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_LOCALLY_SOLVED"><a class="docs-heading-anchor" href="#ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a><a id="ALMOST_LOCALLY_SOLVED-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_LOCALLY_SOLVED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_LOCALLY_SOLVED" href="#JuMP.ALMOST_LOCALLY_SOLVED"><code>JuMP.ALMOST_LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_OPTIMAL"><a class="docs-heading-anchor" href="#ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a><a id="ALMOST_OPTIMAL-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_OPTIMAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_OPTIMAL" href="#JuMP.ALMOST_OPTIMAL"><code>JuMP.ALMOST_OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_OPTIMAL</code>: The algorithm found a globally optimal solution to relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="AUTOMATIC"><a class="docs-heading-anchor" href="#AUTOMATIC"><code>AUTOMATIC</code></a><a id="AUTOMATIC-1"></a><a class="docs-heading-anchor-permalink" href="#AUTOMATIC" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AUTOMATIC" href="#JuMP.AUTOMATIC"><code>JuMP.AUTOMATIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds a CachingOptimizer in AUTOMATIC mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L57">source</a></section></article><h2 id="DIRECT"><a class="docs-heading-anchor" href="#DIRECT"><code>DIRECT</code></a><a id="DIRECT-1"></a><a class="docs-heading-anchor-permalink" href="#DIRECT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.DIRECT" href="#JuMP.DIRECT"><code>JuMP.DIRECT</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds an AbstractOptimizer. No extra copy of the model is stored. The <code>moi_backend</code> must support <code>add_constraint</code> etc.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L64">source</a></section></article><h2 id="DUAL_INFEASIBLE"><a class="docs-heading-anchor" href="#DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a><a id="DUAL_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#DUAL_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.DUAL_INFEASIBLE" href="#JuMP.DUAL_INFEASIBLE"><code>JuMP.DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="FEASIBILITY_SENSE"><a class="docs-heading-anchor" href="#FEASIBILITY_SENSE"><code>FEASIBILITY_SENSE</code></a><a id="FEASIBILITY_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#FEASIBILITY_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FEASIBILITY_SENSE" href="#JuMP.FEASIBILITY_SENSE"><code>JuMP.FEASIBILITY_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBILITY_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>FEASIBILITY_SENSE</code>: the model does not have an objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="FEASIBLE_POINT"><a class="docs-heading-anchor" href="#FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a><a id="FEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#FEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FEASIBLE_POINT" href="#JuMP.FEASIBLE_POINT"><code>JuMP.FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>FEASIBLE_POINT</code>: the result vector is a feasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBILITY_CERTIFICATE"><a class="docs-heading-anchor" href="#INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a><a id="INFEASIBILITY_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBILITY_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBILITY_CERTIFICATE" href="#JuMP.INFEASIBILITY_CERTIFICATE"><code>JuMP.INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBILITY_CERTIFICATE</code>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE"><a class="docs-heading-anchor" href="#INFEASIBLE"><code>INFEASIBLE</code></a><a id="INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE" href="#JuMP.INFEASIBLE"><code>JuMP.INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE</code>: The algorithm concluded that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE_OR_UNBOUNDED"><a class="docs-heading-anchor" href="#INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a><a id="INFEASIBLE_OR_UNBOUNDED-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE_OR_UNBOUNDED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE_OR_UNBOUNDED" href="#JuMP.INFEASIBLE_OR_UNBOUNDED"><code>JuMP.INFEASIBLE_OR_UNBOUNDED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE_OR_UNBOUNDED</code>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE_POINT"><a class="docs-heading-anchor" href="#INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a><a id="INFEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE_POINT" href="#JuMP.INFEASIBLE_POINT"><code>JuMP.INFEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBLE_POINT</code>: the result vector is an infeasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INTERRUPTED"><a class="docs-heading-anchor" href="#INTERRUPTED"><code>INTERRUPTED</code></a><a id="INTERRUPTED-1"></a><a class="docs-heading-anchor-permalink" href="#INTERRUPTED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INTERRUPTED" href="#JuMP.INTERRUPTED"><code>JuMP.INTERRUPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INTERRUPTED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INTERRUPTED</code>: The algorithm stopped because of an interrupt signal.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INVALID_MODEL"><a class="docs-heading-anchor" href="#INVALID_MODEL"><code>INVALID_MODEL</code></a><a id="INVALID_MODEL-1"></a><a class="docs-heading-anchor-permalink" href="#INVALID_MODEL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INVALID_MODEL" href="#JuMP.INVALID_MODEL"><code>JuMP.INVALID_MODEL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_MODEL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_MODEL</code>: The algorithm stopped because the model is invalid.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INVALID_OPTION"><a class="docs-heading-anchor" href="#INVALID_OPTION"><code>INVALID_OPTION</code></a><a id="INVALID_OPTION-1"></a><a class="docs-heading-anchor-permalink" href="#INVALID_OPTION" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INVALID_OPTION" href="#JuMP.INVALID_OPTION"><code>JuMP.INVALID_OPTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_OPTION::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_OPTION</code>: The algorithm stopped because it was provided an invalid option.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ITERATION_LIMIT"><a class="docs-heading-anchor" href="#ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a><a id="ITERATION_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#ITERATION_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ITERATION_LIMIT" href="#JuMP.ITERATION_LIMIT"><code>JuMP.ITERATION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ITERATION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ITERATION_LIMIT</code>: An iterative algorithm stopped after conducting the maximum number of iterations.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="LOCALLY_INFEASIBLE"><a class="docs-heading-anchor" href="#LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a><a id="LOCALLY_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#LOCALLY_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LOCALLY_INFEASIBLE" href="#JuMP.LOCALLY_INFEASIBLE"><code>JuMP.LOCALLY_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_INFEASIBLE</code>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="LOCALLY_SOLVED"><a class="docs-heading-anchor" href="#LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a><a id="LOCALLY_SOLVED-1"></a><a class="docs-heading-anchor-permalink" href="#LOCALLY_SOLVED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LOCALLY_SOLVED" href="#JuMP.LOCALLY_SOLVED"><code>JuMP.LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MANUAL"><a class="docs-heading-anchor" href="#MANUAL"><code>MANUAL</code></a><a id="MANUAL-1"></a><a class="docs-heading-anchor-permalink" href="#MANUAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MANUAL" href="#JuMP.MANUAL"><code>JuMP.MANUAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds a CachingOptimizer in MANUAL mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L62">source</a></section></article><h2 id="MAX_SENSE"><a class="docs-heading-anchor" href="#MAX_SENSE"><code>MAX_SENSE</code></a><a id="MAX_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#MAX_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MAX_SENSE" href="#JuMP.MAX_SENSE"><code>JuMP.MAX_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MAX_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>MAX_SENSE</code>: the goal is to maximize the objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MEMORY_LIMIT"><a class="docs-heading-anchor" href="#MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a><a id="MEMORY_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#MEMORY_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MEMORY_LIMIT" href="#JuMP.MEMORY_LIMIT"><code>JuMP.MEMORY_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MEMORY_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>MEMORY_LIMIT</code>: The algorithm stopped because it ran out of memory.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MIN_SENSE"><a class="docs-heading-anchor" href="#MIN_SENSE"><code>MIN_SENSE</code></a><a id="MIN_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#MIN_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MIN_SENSE" href="#JuMP.MIN_SENSE"><code>JuMP.MIN_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MIN_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>MIN_SENSE</code>: the goal is to minimize the objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_FEASIBLE_POINT"><a class="docs-heading-anchor" href="#NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a><a id="NEARLY_FEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_FEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_FEASIBLE_POINT" href="#JuMP.NEARLY_FEASIBLE_POINT"><code>JuMP.NEARLY_FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_FEASIBLE_POINT</code>: the result vector is feasible if some constraint tolerances are relaxed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_INFEASIBILITY_CERTIFICATE"><a class="docs-heading-anchor" href="#NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a><a id="NEARLY_INFEASIBILITY_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_INFEASIBILITY_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_INFEASIBILITY_CERTIFICATE" href="#JuMP.NEARLY_INFEASIBILITY_CERTIFICATE"><code>JuMP.NEARLY_INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_INFEASIBILITY_CERTIFICATE</code>: the result satisfies a relaxed criterion for a certificate of infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_REDUCTION_CERTIFICATE"><a class="docs-heading-anchor" href="#NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a><a id="NEARLY_REDUCTION_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_REDUCTION_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_REDUCTION_CERTIFICATE" href="#JuMP.NEARLY_REDUCTION_CERTIFICATE"><code>JuMP.NEARLY_REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_REDUCTION_CERTIFICATE</code>: the result satisfies a relaxed criterion for an ill-posed certificate.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NODE_LIMIT"><a class="docs-heading-anchor" href="#NODE_LIMIT"><code>NODE_LIMIT</code></a><a id="NODE_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#NODE_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NODE_LIMIT" href="#JuMP.NODE_LIMIT"><code>JuMP.NODE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NODE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NODE_LIMIT</code>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NORM_LIMIT"><a class="docs-heading-anchor" href="#NORM_LIMIT"><code>NORM_LIMIT</code></a><a id="NORM_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#NORM_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NORM_LIMIT" href="#JuMP.NORM_LIMIT"><code>JuMP.NORM_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NORM_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NORM_LIMIT</code>: The algorithm stopped because the norm of an iterate became too large.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NO_SOLUTION"><a class="docs-heading-anchor" href="#NO_SOLUTION"><code>NO_SOLUTION</code></a><a id="NO_SOLUTION-1"></a><a class="docs-heading-anchor-permalink" href="#NO_SOLUTION" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NO_SOLUTION" href="#JuMP.NO_SOLUTION"><code>JuMP.NO_SOLUTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NO_SOLUTION::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NO_SOLUTION</code>: the result vector is empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NUMERICAL_ERROR"><a class="docs-heading-anchor" href="#NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a><a id="NUMERICAL_ERROR-1"></a><a class="docs-heading-anchor-permalink" href="#NUMERICAL_ERROR" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NUMERICAL_ERROR" href="#JuMP.NUMERICAL_ERROR"><code>JuMP.NUMERICAL_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NUMERICAL_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NUMERICAL_ERROR</code>: The algorithm stopped because it encountered unrecoverable numerical error.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OBJECTIVE_LIMIT"><a class="docs-heading-anchor" href="#OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a><a id="OBJECTIVE_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#OBJECTIVE_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OBJECTIVE_LIMIT" href="#JuMP.OBJECTIVE_LIMIT"><code>JuMP.OBJECTIVE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OBJECTIVE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OBJECTIVE_LIMIT</code>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OPTIMAL"><a class="docs-heading-anchor" href="#OPTIMAL"><code>OPTIMAL</code></a><a id="OPTIMAL-1"></a><a class="docs-heading-anchor-permalink" href="#OPTIMAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OPTIMAL" href="#JuMP.OPTIMAL"><code>JuMP.OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMAL</code>: The algorithm found a globally optimal solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OPTIMIZE_NOT_CALLED"><a class="docs-heading-anchor" href="#OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a><a id="OPTIMIZE_NOT_CALLED-1"></a><a class="docs-heading-anchor-permalink" href="#OPTIMIZE_NOT_CALLED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OPTIMIZE_NOT_CALLED" href="#JuMP.OPTIMIZE_NOT_CALLED"><code>JuMP.OPTIMIZE_NOT_CALLED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMIZE_NOT_CALLED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMIZE_NOT_CALLED</code>: The algorithm has not started.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_ERROR"><a class="docs-heading-anchor" href="#OTHER_ERROR"><code>OTHER_ERROR</code></a><a id="OTHER_ERROR-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_ERROR" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_ERROR" href="#JuMP.OTHER_ERROR"><code>JuMP.OTHER_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_ERROR</code>: The algorithm stopped because of an error not covered by one of the statuses defined above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_LIMIT"><a class="docs-heading-anchor" href="#OTHER_LIMIT"><code>OTHER_LIMIT</code></a><a id="OTHER_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_LIMIT" href="#JuMP.OTHER_LIMIT"><code>JuMP.OTHER_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_LIMIT</code>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_RESULT_STATUS"><a class="docs-heading-anchor" href="#OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a><a id="OTHER_RESULT_STATUS-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_RESULT_STATUS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_RESULT_STATUS" href="#JuMP.OTHER_RESULT_STATUS"><code>JuMP.OTHER_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>OTHER_RESULT_STATUS</code>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="REDUCTION_CERTIFICATE"><a class="docs-heading-anchor" href="#REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a><a id="REDUCTION_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#REDUCTION_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.REDUCTION_CERTIFICATE" href="#JuMP.REDUCTION_CERTIFICATE"><code>JuMP.REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>REDUCTION_CERTIFICATE</code>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="SLOW_PROGRESS"><a class="docs-heading-anchor" href="#SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a><a id="SLOW_PROGRESS-1"></a><a class="docs-heading-anchor-permalink" href="#SLOW_PROGRESS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SLOW_PROGRESS" href="#JuMP.SLOW_PROGRESS"><code>JuMP.SLOW_PROGRESS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SLOW_PROGRESS::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SLOW_PROGRESS</code>: The algorithm stopped because it was unable to continue making progress towards the solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="SOLUTION_LIMIT"><a class="docs-heading-anchor" href="#SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a><a id="SOLUTION_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#SOLUTION_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOLUTION_LIMIT" href="#JuMP.SOLUTION_LIMIT"><code>JuMP.SOLUTION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SOLUTION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SOLUTION_LIMIT</code>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="TIME_LIMIT"><a class="docs-heading-anchor" href="#TIME_LIMIT"><code>TIME_LIMIT</code></a><a id="TIME_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#TIME_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.TIME_LIMIT" href="#JuMP.TIME_LIMIT"><code>JuMP.TIME_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">TIME_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>TIME_LIMIT</code>: The algorithm stopped after a user-specified computation time.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="UNKNOWN_RESULT_STATUS"><a class="docs-heading-anchor" href="#UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a><a id="UNKNOWN_RESULT_STATUS-1"></a><a class="docs-heading-anchor-permalink" href="#UNKNOWN_RESULT_STATUS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UNKNOWN_RESULT_STATUS" href="#JuMP.UNKNOWN_RESULT_STATUS"><code>JuMP.UNKNOWN_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">UNKNOWN_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>UNKNOWN_RESULT_STATUS</code>: the result vector contains a solution with an unknown interpretation.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="op_and"><a class="docs-heading-anchor" href="#op_and"><code>op_and</code></a><a id="op_and-1"></a><a class="docs-heading-anchor-permalink" href="#op_and" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_and" href="#JuMP.op_and"><code>JuMP.op_and</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_and(x, y)</code></pre><p>A function that falls back to <code>x &amp; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L293-L319">source</a></section></article><h2 id="ALMOST_DUAL_INFEASIBLE"><a class="docs-heading-anchor" href="#ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a><a id="ALMOST_DUAL_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_DUAL_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_DUAL_INFEASIBLE" href="#JuMP.ALMOST_DUAL_INFEASIBLE"><code>JuMP.ALMOST_DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_INFEASIBLE"><a class="docs-heading-anchor" href="#ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a><a id="ALMOST_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_INFEASIBLE" href="#JuMP.ALMOST_INFEASIBLE"><code>JuMP.ALMOST_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_INFEASIBLE</code>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_LOCALLY_SOLVED"><a class="docs-heading-anchor" href="#ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a><a id="ALMOST_LOCALLY_SOLVED-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_LOCALLY_SOLVED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_LOCALLY_SOLVED" href="#JuMP.ALMOST_LOCALLY_SOLVED"><code>JuMP.ALMOST_LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ALMOST_OPTIMAL"><a class="docs-heading-anchor" href="#ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a><a id="ALMOST_OPTIMAL-1"></a><a class="docs-heading-anchor-permalink" href="#ALMOST_OPTIMAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ALMOST_OPTIMAL" href="#JuMP.ALMOST_OPTIMAL"><code>JuMP.ALMOST_OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_OPTIMAL</code>: The algorithm found a globally optimal solution to relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="AUTOMATIC"><a class="docs-heading-anchor" href="#AUTOMATIC"><code>AUTOMATIC</code></a><a id="AUTOMATIC-1"></a><a class="docs-heading-anchor-permalink" href="#AUTOMATIC" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.AUTOMATIC" href="#JuMP.AUTOMATIC"><code>JuMP.AUTOMATIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds a CachingOptimizer in AUTOMATIC mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L57">source</a></section></article><h2 id="DIRECT"><a class="docs-heading-anchor" href="#DIRECT"><code>DIRECT</code></a><a id="DIRECT-1"></a><a class="docs-heading-anchor-permalink" href="#DIRECT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.DIRECT" href="#JuMP.DIRECT"><code>JuMP.DIRECT</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds an AbstractOptimizer. No extra copy of the model is stored. The <code>moi_backend</code> must support <code>add_constraint</code> etc.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L64">source</a></section></article><h2 id="DUAL_INFEASIBLE"><a class="docs-heading-anchor" href="#DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a><a id="DUAL_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#DUAL_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.DUAL_INFEASIBLE" href="#JuMP.DUAL_INFEASIBLE"><code>JuMP.DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="FEASIBILITY_SENSE"><a class="docs-heading-anchor" href="#FEASIBILITY_SENSE"><code>FEASIBILITY_SENSE</code></a><a id="FEASIBILITY_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#FEASIBILITY_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FEASIBILITY_SENSE" href="#JuMP.FEASIBILITY_SENSE"><code>JuMP.FEASIBILITY_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBILITY_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>FEASIBILITY_SENSE</code>: the model does not have an objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="FEASIBLE_POINT"><a class="docs-heading-anchor" href="#FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a><a id="FEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#FEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.FEASIBLE_POINT" href="#JuMP.FEASIBLE_POINT"><code>JuMP.FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>FEASIBLE_POINT</code>: the result vector is a feasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBILITY_CERTIFICATE"><a class="docs-heading-anchor" href="#INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a><a id="INFEASIBILITY_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBILITY_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBILITY_CERTIFICATE" href="#JuMP.INFEASIBILITY_CERTIFICATE"><code>JuMP.INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBILITY_CERTIFICATE</code>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE"><a class="docs-heading-anchor" href="#INFEASIBLE"><code>INFEASIBLE</code></a><a id="INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE" href="#JuMP.INFEASIBLE"><code>JuMP.INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE</code>: The algorithm concluded that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE_OR_UNBOUNDED"><a class="docs-heading-anchor" href="#INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a><a id="INFEASIBLE_OR_UNBOUNDED-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE_OR_UNBOUNDED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE_OR_UNBOUNDED" href="#JuMP.INFEASIBLE_OR_UNBOUNDED"><code>JuMP.INFEASIBLE_OR_UNBOUNDED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE_OR_UNBOUNDED</code>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INFEASIBLE_POINT"><a class="docs-heading-anchor" href="#INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a><a id="INFEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#INFEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INFEASIBLE_POINT" href="#JuMP.INFEASIBLE_POINT"><code>JuMP.INFEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBLE_POINT</code>: the result vector is an infeasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INTERRUPTED"><a class="docs-heading-anchor" href="#INTERRUPTED"><code>INTERRUPTED</code></a><a id="INTERRUPTED-1"></a><a class="docs-heading-anchor-permalink" href="#INTERRUPTED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INTERRUPTED" href="#JuMP.INTERRUPTED"><code>JuMP.INTERRUPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INTERRUPTED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INTERRUPTED</code>: The algorithm stopped because of an interrupt signal.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INVALID_MODEL"><a class="docs-heading-anchor" href="#INVALID_MODEL"><code>INVALID_MODEL</code></a><a id="INVALID_MODEL-1"></a><a class="docs-heading-anchor-permalink" href="#INVALID_MODEL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INVALID_MODEL" href="#JuMP.INVALID_MODEL"><code>JuMP.INVALID_MODEL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_MODEL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_MODEL</code>: The algorithm stopped because the model is invalid.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="INVALID_OPTION"><a class="docs-heading-anchor" href="#INVALID_OPTION"><code>INVALID_OPTION</code></a><a id="INVALID_OPTION-1"></a><a class="docs-heading-anchor-permalink" href="#INVALID_OPTION" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.INVALID_OPTION" href="#JuMP.INVALID_OPTION"><code>JuMP.INVALID_OPTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_OPTION::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_OPTION</code>: The algorithm stopped because it was provided an invalid option.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="ITERATION_LIMIT"><a class="docs-heading-anchor" href="#ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a><a id="ITERATION_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#ITERATION_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.ITERATION_LIMIT" href="#JuMP.ITERATION_LIMIT"><code>JuMP.ITERATION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ITERATION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ITERATION_LIMIT</code>: An iterative algorithm stopped after conducting the maximum number of iterations.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="LOCALLY_INFEASIBLE"><a class="docs-heading-anchor" href="#LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a><a id="LOCALLY_INFEASIBLE-1"></a><a class="docs-heading-anchor-permalink" href="#LOCALLY_INFEASIBLE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LOCALLY_INFEASIBLE" href="#JuMP.LOCALLY_INFEASIBLE"><code>JuMP.LOCALLY_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_INFEASIBLE</code>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="LOCALLY_SOLVED"><a class="docs-heading-anchor" href="#LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a><a id="LOCALLY_SOLVED-1"></a><a class="docs-heading-anchor-permalink" href="#LOCALLY_SOLVED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.LOCALLY_SOLVED" href="#JuMP.LOCALLY_SOLVED"><code>JuMP.LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MANUAL"><a class="docs-heading-anchor" href="#MANUAL"><code>MANUAL</code></a><a id="MANUAL-1"></a><a class="docs-heading-anchor-permalink" href="#MANUAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MANUAL" href="#JuMP.MANUAL"><code>JuMP.MANUAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><p><code>moi_backend</code> field holds a CachingOptimizer in MANUAL mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L62">source</a></section></article><h2 id="MAX_SENSE"><a class="docs-heading-anchor" href="#MAX_SENSE"><code>MAX_SENSE</code></a><a id="MAX_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#MAX_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MAX_SENSE" href="#JuMP.MAX_SENSE"><code>JuMP.MAX_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MAX_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>MAX_SENSE</code>: the goal is to maximize the objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MEMORY_LIMIT"><a class="docs-heading-anchor" href="#MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a><a id="MEMORY_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#MEMORY_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MEMORY_LIMIT" href="#JuMP.MEMORY_LIMIT"><code>JuMP.MEMORY_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MEMORY_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>MEMORY_LIMIT</code>: The algorithm stopped because it ran out of memory.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="MIN_SENSE"><a class="docs-heading-anchor" href="#MIN_SENSE"><code>MIN_SENSE</code></a><a id="MIN_SENSE-1"></a><a class="docs-heading-anchor-permalink" href="#MIN_SENSE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.MIN_SENSE" href="#JuMP.MIN_SENSE"><code>JuMP.MIN_SENSE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MIN_SENSE::OptimizationSense</code></pre><p>An instance of the <a href="#JuMP.OptimizationSense"><code>OptimizationSense</code></a> enum.</p><p><code>MIN_SENSE</code>: the goal is to minimize the objective function</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_FEASIBLE_POINT"><a class="docs-heading-anchor" href="#NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a><a id="NEARLY_FEASIBLE_POINT-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_FEASIBLE_POINT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_FEASIBLE_POINT" href="#JuMP.NEARLY_FEASIBLE_POINT"><code>JuMP.NEARLY_FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_FEASIBLE_POINT</code>: the result vector is feasible if some constraint tolerances are relaxed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_INFEASIBILITY_CERTIFICATE"><a class="docs-heading-anchor" href="#NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a><a id="NEARLY_INFEASIBILITY_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_INFEASIBILITY_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_INFEASIBILITY_CERTIFICATE" href="#JuMP.NEARLY_INFEASIBILITY_CERTIFICATE"><code>JuMP.NEARLY_INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_INFEASIBILITY_CERTIFICATE</code>: the result satisfies a relaxed criterion for a certificate of infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NEARLY_REDUCTION_CERTIFICATE"><a class="docs-heading-anchor" href="#NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a><a id="NEARLY_REDUCTION_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#NEARLY_REDUCTION_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NEARLY_REDUCTION_CERTIFICATE" href="#JuMP.NEARLY_REDUCTION_CERTIFICATE"><code>JuMP.NEARLY_REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_REDUCTION_CERTIFICATE</code>: the result satisfies a relaxed criterion for an ill-posed certificate.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NODE_LIMIT"><a class="docs-heading-anchor" href="#NODE_LIMIT"><code>NODE_LIMIT</code></a><a id="NODE_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#NODE_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NODE_LIMIT" href="#JuMP.NODE_LIMIT"><code>JuMP.NODE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NODE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NODE_LIMIT</code>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NORM_LIMIT"><a class="docs-heading-anchor" href="#NORM_LIMIT"><code>NORM_LIMIT</code></a><a id="NORM_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#NORM_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NORM_LIMIT" href="#JuMP.NORM_LIMIT"><code>JuMP.NORM_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NORM_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NORM_LIMIT</code>: The algorithm stopped because the norm of an iterate became too large.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NO_SOLUTION"><a class="docs-heading-anchor" href="#NO_SOLUTION"><code>NO_SOLUTION</code></a><a id="NO_SOLUTION-1"></a><a class="docs-heading-anchor-permalink" href="#NO_SOLUTION" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NO_SOLUTION" href="#JuMP.NO_SOLUTION"><code>JuMP.NO_SOLUTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NO_SOLUTION::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NO_SOLUTION</code>: the result vector is empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="NUMERICAL_ERROR"><a class="docs-heading-anchor" href="#NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a><a id="NUMERICAL_ERROR-1"></a><a class="docs-heading-anchor-permalink" href="#NUMERICAL_ERROR" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.NUMERICAL_ERROR" href="#JuMP.NUMERICAL_ERROR"><code>JuMP.NUMERICAL_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NUMERICAL_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NUMERICAL_ERROR</code>: The algorithm stopped because it encountered unrecoverable numerical error.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OBJECTIVE_LIMIT"><a class="docs-heading-anchor" href="#OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a><a id="OBJECTIVE_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#OBJECTIVE_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OBJECTIVE_LIMIT" href="#JuMP.OBJECTIVE_LIMIT"><code>JuMP.OBJECTIVE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OBJECTIVE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OBJECTIVE_LIMIT</code>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OPTIMAL"><a class="docs-heading-anchor" href="#OPTIMAL"><code>OPTIMAL</code></a><a id="OPTIMAL-1"></a><a class="docs-heading-anchor-permalink" href="#OPTIMAL" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OPTIMAL" href="#JuMP.OPTIMAL"><code>JuMP.OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMAL</code>: The algorithm found a globally optimal solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OPTIMIZE_NOT_CALLED"><a class="docs-heading-anchor" href="#OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a><a id="OPTIMIZE_NOT_CALLED-1"></a><a class="docs-heading-anchor-permalink" href="#OPTIMIZE_NOT_CALLED" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OPTIMIZE_NOT_CALLED" href="#JuMP.OPTIMIZE_NOT_CALLED"><code>JuMP.OPTIMIZE_NOT_CALLED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMIZE_NOT_CALLED::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMIZE_NOT_CALLED</code>: The algorithm has not started.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_ERROR"><a class="docs-heading-anchor" href="#OTHER_ERROR"><code>OTHER_ERROR</code></a><a id="OTHER_ERROR-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_ERROR" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_ERROR" href="#JuMP.OTHER_ERROR"><code>JuMP.OTHER_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_ERROR</code>: The algorithm stopped because of an error not covered by one of the statuses defined above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_LIMIT"><a class="docs-heading-anchor" href="#OTHER_LIMIT"><code>OTHER_LIMIT</code></a><a id="OTHER_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_LIMIT" href="#JuMP.OTHER_LIMIT"><code>JuMP.OTHER_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_LIMIT</code>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="OTHER_RESULT_STATUS"><a class="docs-heading-anchor" href="#OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a><a id="OTHER_RESULT_STATUS-1"></a><a class="docs-heading-anchor-permalink" href="#OTHER_RESULT_STATUS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.OTHER_RESULT_STATUS" href="#JuMP.OTHER_RESULT_STATUS"><code>JuMP.OTHER_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>OTHER_RESULT_STATUS</code>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="REDUCTION_CERTIFICATE"><a class="docs-heading-anchor" href="#REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a><a id="REDUCTION_CERTIFICATE-1"></a><a class="docs-heading-anchor-permalink" href="#REDUCTION_CERTIFICATE" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.REDUCTION_CERTIFICATE" href="#JuMP.REDUCTION_CERTIFICATE"><code>JuMP.REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>REDUCTION_CERTIFICATE</code>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="SLOW_PROGRESS"><a class="docs-heading-anchor" href="#SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a><a id="SLOW_PROGRESS-1"></a><a class="docs-heading-anchor-permalink" href="#SLOW_PROGRESS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SLOW_PROGRESS" href="#JuMP.SLOW_PROGRESS"><code>JuMP.SLOW_PROGRESS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SLOW_PROGRESS::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SLOW_PROGRESS</code>: The algorithm stopped because it was unable to continue making progress towards the solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="SOLUTION_LIMIT"><a class="docs-heading-anchor" href="#SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a><a id="SOLUTION_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#SOLUTION_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.SOLUTION_LIMIT" href="#JuMP.SOLUTION_LIMIT"><code>JuMP.SOLUTION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SOLUTION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SOLUTION_LIMIT</code>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="TIME_LIMIT"><a class="docs-heading-anchor" href="#TIME_LIMIT"><code>TIME_LIMIT</code></a><a id="TIME_LIMIT-1"></a><a class="docs-heading-anchor-permalink" href="#TIME_LIMIT" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.TIME_LIMIT" href="#JuMP.TIME_LIMIT"><code>JuMP.TIME_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">TIME_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#JuMP.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>TIME_LIMIT</code>: The algorithm stopped after a user-specified computation time.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="UNKNOWN_RESULT_STATUS"><a class="docs-heading-anchor" href="#UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a><a id="UNKNOWN_RESULT_STATUS-1"></a><a class="docs-heading-anchor-permalink" href="#UNKNOWN_RESULT_STATUS" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.UNKNOWN_RESULT_STATUS" href="#JuMP.UNKNOWN_RESULT_STATUS"><code>JuMP.UNKNOWN_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">UNKNOWN_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#JuMP.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>UNKNOWN_RESULT_STATUS</code>: the result vector contains a solution with an unknown interpretation.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L1124-L1130">source</a></section></article><h2 id="op_and"><a class="docs-heading-anchor" href="#op_and"><code>op_and</code></a><a id="op_and-1"></a><a class="docs-heading-anchor-permalink" href="#op_and" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_and" href="#JuMP.op_and"><code>JuMP.op_and</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_and(x, y)</code></pre><p>A function that falls back to <code>x &amp; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2012,7 +2012,7 @@
 false
 
 julia&gt; op_and(true, x)
-true &amp;&amp; x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L515-L534">source</a></section></article><h2 id="op_equal_to"><a class="docs-heading-anchor" href="#op_equal_to"><code>op_equal_to</code></a><a id="op_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_equal_to" href="#JuMP.op_equal_to"><code>JuMP.op_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_equal_to(x, y)</code></pre><p>A function that falls back to <code>x == y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true &amp;&amp; x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L515-L534">source</a></section></article><h2 id="op_equal_to"><a class="docs-heading-anchor" href="#op_equal_to"><code>op_equal_to</code></a><a id="op_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_equal_to" href="#JuMP.op_equal_to"><code>JuMP.op_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_equal_to(x, y)</code></pre><p>A function that falls back to <code>x == y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2020,7 +2020,7 @@
 true
 
 julia&gt; op_equal_to(x, 2)
-x == 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L651-L670">source</a></section></article><h2 id="op_greater_than_or_equal_to"><a class="docs-heading-anchor" href="#op_greater_than_or_equal_to"><code>op_greater_than_or_equal_to</code></a><a id="op_greater_than_or_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_greater_than_or_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_greater_than_or_equal_to" href="#JuMP.op_greater_than_or_equal_to"><code>JuMP.op_greater_than_or_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_greater_than_or_equal_to(x, y)</code></pre><p>A function that falls back to <code>x &gt;= y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x == 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L651-L670">source</a></section></article><h2 id="op_greater_than_or_equal_to"><a class="docs-heading-anchor" href="#op_greater_than_or_equal_to"><code>op_greater_than_or_equal_to</code></a><a id="op_greater_than_or_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_greater_than_or_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_greater_than_or_equal_to" href="#JuMP.op_greater_than_or_equal_to"><code>JuMP.op_greater_than_or_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_greater_than_or_equal_to(x, y)</code></pre><p>A function that falls back to <code>x &gt;= y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2028,7 +2028,7 @@
 true
 
 julia&gt; op_greater_than_or_equal_to(x, 2)
-x &gt;= 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L629-L648">source</a></section></article><h2 id="op_less_than_or_equal_to"><a class="docs-heading-anchor" href="#op_less_than_or_equal_to"><code>op_less_than_or_equal_to</code></a><a id="op_less_than_or_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_less_than_or_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_less_than_or_equal_to" href="#JuMP.op_less_than_or_equal_to"><code>JuMP.op_less_than_or_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_less_than_or_equal_to(x, y)</code></pre><p>A function that falls back to <code>x &lt;= y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x &gt;= 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L629-L648">source</a></section></article><h2 id="op_less_than_or_equal_to"><a class="docs-heading-anchor" href="#op_less_than_or_equal_to"><code>op_less_than_or_equal_to</code></a><a id="op_less_than_or_equal_to-1"></a><a class="docs-heading-anchor-permalink" href="#op_less_than_or_equal_to" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_less_than_or_equal_to" href="#JuMP.op_less_than_or_equal_to"><code>JuMP.op_less_than_or_equal_to</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_less_than_or_equal_to(x, y)</code></pre><p>A function that falls back to <code>x &lt;= y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2036,7 +2036,7 @@
 true
 
 julia&gt; op_less_than_or_equal_to(x, 2)
-x &lt;= 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L607-L626">source</a></section></article><h2 id="op_or"><a class="docs-heading-anchor" href="#op_or"><code>op_or</code></a><a id="op_or-1"></a><a class="docs-heading-anchor-permalink" href="#op_or" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_or" href="#JuMP.op_or"><code>JuMP.op_or</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_or(x, y)</code></pre><p>A function that falls back to <code>x | y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x &lt;= 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L607-L626">source</a></section></article><h2 id="op_or"><a class="docs-heading-anchor" href="#op_or"><code>op_or</code></a><a id="op_or-1"></a><a class="docs-heading-anchor-permalink" href="#op_or" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_or" href="#JuMP.op_or"><code>JuMP.op_or</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_or(x, y)</code></pre><p>A function that falls back to <code>x | y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2044,7 +2044,7 @@
 true
 
 julia&gt; op_or(true, x)
-true || x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L539-L558">source</a></section></article><h2 id="op_strictly_greater_than"><a class="docs-heading-anchor" href="#op_strictly_greater_than"><code>op_strictly_greater_than</code></a><a id="op_strictly_greater_than-1"></a><a class="docs-heading-anchor-permalink" href="#op_strictly_greater_than" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_strictly_greater_than" href="#JuMP.op_strictly_greater_than"><code>JuMP.op_strictly_greater_than</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_strictly_greater_than(x, y)</code></pre><p>A function that falls back to <code>x &gt; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+true || x</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L539-L558">source</a></section></article><h2 id="op_strictly_greater_than"><a class="docs-heading-anchor" href="#op_strictly_greater_than"><code>op_strictly_greater_than</code></a><a id="op_strictly_greater_than-1"></a><a class="docs-heading-anchor-permalink" href="#op_strictly_greater_than" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_strictly_greater_than" href="#JuMP.op_strictly_greater_than"><code>JuMP.op_strictly_greater_than</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_strictly_greater_than(x, y)</code></pre><p>A function that falls back to <code>x &gt; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2052,7 +2052,7 @@
 false
 
 julia&gt; op_strictly_greater_than(x, 2)
-x &gt; 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L585-L604">source</a></section></article><h2 id="op_strictly_less_than"><a class="docs-heading-anchor" href="#op_strictly_less_than"><code>op_strictly_less_than</code></a><a id="op_strictly_less_than-1"></a><a class="docs-heading-anchor-permalink" href="#op_strictly_less_than" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_strictly_less_than" href="#JuMP.op_strictly_less_than"><code>JuMP.op_strictly_less_than</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_strictly_less_than(x, y)</code></pre><p>A function that falls back to <code>x &lt; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x &gt; 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L585-L604">source</a></section></article><h2 id="op_strictly_less_than"><a class="docs-heading-anchor" href="#op_strictly_less_than"><code>op_strictly_less_than</code></a><a id="op_strictly_less_than-1"></a><a class="docs-heading-anchor-permalink" href="#op_strictly_less_than" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="JuMP.op_strictly_less_than" href="#JuMP.op_strictly_less_than"><code>JuMP.op_strictly_less_than</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">op_strictly_less_than(x, y)</code></pre><p>A function that falls back to <code>x &lt; y</code>, but when called with JuMP variables or expressions, returns a <a href="#JuMP.GenericNonlinearExpr"><code>GenericNonlinearExpr</code></a>.</p><p><strong>Example</strong></p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x);
 
@@ -2060,7 +2060,7 @@
 true
 
 julia&gt; op_strictly_less_than(x, 2)
-x &lt; 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/macros.jl#L563-L582">source</a></section></article><h2 id="Base.empty!(::GenericModel)"><a class="docs-heading-anchor" href="#Base.empty!(::GenericModel)"><code>Base.empty!(::GenericModel)</code></a><a id="Base.empty!(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.empty!(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.empty!-Tuple{GenericModel}" href="#Base.empty!-Tuple{GenericModel}"><code>Base.empty!</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">empty!(model::GenericModel)::GenericModel</code></pre><p>Empty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes. Always return the argument.</p><p>Note: removes extensions data.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L800-L807">source</a></section></article><h2 id="Base.isempty(::GenericModel)"><a class="docs-heading-anchor" href="#Base.isempty(::GenericModel)"><code>Base.isempty(::GenericModel)</code></a><a id="Base.isempty(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.isempty(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.isempty-Tuple{GenericModel}" href="#Base.isempty-Tuple{GenericModel}"><code>Base.isempty</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">isempty(model::GenericModel)</code></pre><p>Verifies whether the model is empty, that is, whether the MOI backend is empty and whether the model is in the same state as at its creation apart from optimizer attributes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/JuMP.jl#L827-L833">source</a></section></article><h2 id="Base.copy(::AbstractModel)"><a class="docs-heading-anchor" href="#Base.copy(::AbstractModel)"><code>Base.copy(::AbstractModel)</code></a><a id="Base.copy(::AbstractModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.copy(::AbstractModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.copy-Tuple{AbstractModel}" href="#Base.copy-Tuple{AbstractModel}"><code>Base.copy</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">copy(model::AbstractModel)</code></pre><p>Return a copy of the model <code>model</code>. It is similar to <a href="#JuMP.copy_model"><code>copy_model</code></a> except that it does not return the mapping between the references of <code>model</code> and its copy.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and a constraint <code>cref</code>. It is then copied into a model <code>new_model</code> with the new references assigned to <code>x_new</code> and <code>cref_new</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
+x &lt; 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/macros.jl#L563-L582">source</a></section></article><h2 id="Base.empty!(::GenericModel)"><a class="docs-heading-anchor" href="#Base.empty!(::GenericModel)"><code>Base.empty!(::GenericModel)</code></a><a id="Base.empty!(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.empty!(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.empty!-Tuple{GenericModel}" href="#Base.empty!-Tuple{GenericModel}"><code>Base.empty!</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">empty!(model::GenericModel)::GenericModel</code></pre><p>Empty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes. Always return the argument.</p><p>Note: removes extensions data.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L800-L807">source</a></section></article><h2 id="Base.isempty(::GenericModel)"><a class="docs-heading-anchor" href="#Base.isempty(::GenericModel)"><code>Base.isempty(::GenericModel)</code></a><a id="Base.isempty(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.isempty(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.isempty-Tuple{GenericModel}" href="#Base.isempty-Tuple{GenericModel}"><code>Base.isempty</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">isempty(model::GenericModel)</code></pre><p>Verifies whether the model is empty, that is, whether the MOI backend is empty and whether the model is in the same state as at its creation apart from optimizer attributes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/JuMP.jl#L827-L833">source</a></section></article><h2 id="Base.copy(::AbstractModel)"><a class="docs-heading-anchor" href="#Base.copy(::AbstractModel)"><code>Base.copy(::AbstractModel)</code></a><a id="Base.copy(::AbstractModel)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.copy(::AbstractModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.copy-Tuple{AbstractModel}" href="#Base.copy-Tuple{AbstractModel}"><code>Base.copy</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">copy(model::AbstractModel)</code></pre><p>Return a copy of the model <code>model</code>. It is similar to <a href="#JuMP.copy_model"><code>copy_model</code></a> except that it does not return the mapping between the references of <code>model</code> and its copy.</p><p><strong>Note</strong></p><p>Model copy is not supported in <code>DIRECT</code> mode, i.e. when a model is constructed using the <a href="#JuMP.direct_model"><code>direct_model</code></a> constructor instead of the <a href="#JuMP.Model"><code>Model</code></a> constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the <a href="#JuMP.optimize!"><code>optimize!</code></a> call.</p><p><strong>Example</strong></p><p>In the following example, a model <code>model</code> is constructed with a variable <code>x</code> and a constraint <code>cref</code>. It is then copied into a model <code>new_model</code> with the new references assigned to <code>x_new</code> and <code>cref_new</code>.</p><pre><code class="language-julia-repl hljs">julia&gt; model = Model();
 
 julia&gt; @variable(model, x)
 x
@@ -2074,9 +2074,9 @@
 x
 
 julia&gt; cref_new = model[:cref]
-cref : x = 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/copy.jl#L203-L241">source</a></section></article><h2 id="Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)"><a class="docs-heading-anchor" href="#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)"><code>Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)</code></a><a id="Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.write-Tuple{IO, GenericModel}" href="#Base.write-Tuple{IO, GenericModel}"><code>Base.write</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">Base.write(
+cref : x = 2</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/copy.jl#L203-L241">source</a></section></article><h2 id="Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)"><a class="docs-heading-anchor" href="#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)"><code>Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)</code></a><a id="Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)-1"></a><a class="docs-heading-anchor-permalink" href="#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="Base.write-Tuple{IO, GenericModel}" href="#Base.write-Tuple{IO, GenericModel}"><code>Base.write</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">Base.write(
     io::IO,
     model::GenericModel;
     format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_MOF,
     kwargs...,
-)</code></pre><p>Write the JuMP model <code>model</code> to <code>io</code> in the format <code>format</code>.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/file_formats.jl#L63-L74">source</a></section></article><h2 id="MOI.Utilities.reset_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.reset_optimizer(::GenericModel)"><code>MOI.Utilities.reset_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.reset_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.reset_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.reset_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L285-L291">source</a></section></article><h2 id="MOI.Utilities.drop_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.drop_optimizer(::GenericModel)"><code>MOI.Utilities.drop_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.drop_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.drop_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.drop_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.drop_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.drop_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L300-L306">source</a></section></article><h2 id="MOI.Utilities.attach_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.attach_optimizer(::GenericModel)"><code>MOI.Utilities.attach_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.attach_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.attach_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.attach_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.attach_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.attach_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L313-L319">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../manual/nlp/">« Nonlinear Modeling (Legacy)</a><a class="docs-footer-nextpage" href="../JuMP.Containers/">JuMP.Containers »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><p>Write the JuMP model <code>model</code> to <code>io</code> in the format <code>format</code>.</p><p>Other <code>kwargs</code> are passed to the <code>Model</code> constructor of the chosen format.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/file_formats.jl#L63-L74">source</a></section></article><h2 id="MOI.Utilities.reset_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.reset_optimizer(::GenericModel)"><code>MOI.Utilities.reset_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.reset_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.reset_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.reset_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L285-L291">source</a></section></article><h2 id="MOI.Utilities.drop_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.drop_optimizer(::GenericModel)"><code>MOI.Utilities.drop_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.drop_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.drop_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.drop_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.drop_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.drop_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L300-L306">source</a></section></article><h2 id="MOI.Utilities.attach_optimizer(::GenericModel)"><a class="docs-heading-anchor" href="#MOI.Utilities.attach_optimizer(::GenericModel)"><code>MOI.Utilities.attach_optimizer(::GenericModel)</code></a><a id="MOI.Utilities.attach_optimizer(::GenericModel)-1"></a><a class="docs-heading-anchor-permalink" href="#MOI.Utilities.attach_optimizer(::GenericModel)" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}" href="#MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}"><code>MathOptInterface.Utilities.attach_optimizer</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">MOIU.attach_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.attach_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L313-L319">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../manual/nlp/">« Nonlinear Modeling (Legacy)</a><a class="docs-footer-nextpage" href="../JuMP.Containers/">JuMP.Containers »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/background/algebraic_modeling_languages/index.html b/previews/PR3562/background/algebraic_modeling_languages/index.html
index aa1ed92f324..c5b6b4738f3 100644
--- a/previews/PR3562/background/algebraic_modeling_languages/index.html
+++ b/previews/PR3562/background/algebraic_modeling_languages/index.html
@@ -126,4 +126,4 @@
 julia&gt; highs_knapsack([1.0, 2.0], [0.5, 0.5], 1.25)
 2-element Vector{Float64}:
  0.0
- 2.0</code></pre><p>We&#39;ve now gone from a algebraic model that looked identical to the mathematical model we started with, to a verbose function that uses HiGHS-specific functionality.</p><p>The difference between <code>algebraic_knapsack</code> and <code>highs_knapsack</code> highlights the benefit that algebraic modeling languages provide to users. Moreover, if we used a different solver, the solver-specific function would be entirely different. A key benefit of an algebraic modeling language is that you can change the solver without needing to rewrite the model.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../api/JuMP.Containers/">« JuMP.Containers</a><a class="docs-footer-nextpage" href="../../developers/contributing/">Contributing »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 2.0</code></pre><p>We&#39;ve now gone from a algebraic model that looked identical to the mathematical model we started with, to a verbose function that uses HiGHS-specific functionality.</p><p>The difference between <code>algebraic_knapsack</code> and <code>highs_knapsack</code> highlights the benefit that algebraic modeling languages provide to users. Moreover, if we used a different solver, the solver-specific function would be entirely different. A key benefit of an algebraic modeling language is that you can change the solver without needing to rewrite the model.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../api/JuMP.Containers/">« JuMP.Containers</a><a class="docs-footer-nextpage" href="../../developers/contributing/">Contributing »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/changelog/index.html b/previews/PR3562/changelog/index.html
index b1933765f9e..26a1b2dd6a2 100644
--- a/previews/PR3562/changelog/index.html
+++ b/previews/PR3562/changelog/index.html
@@ -12,4 +12,4 @@
 new_b = backend(model)</code></pre></li><li>All usages of <code>@SDconstraint</code> are deprecated. The new syntax is <code>@constraint(model, X &gt;= Y, PSDCone())</code>.</li><li>Creating a <code>DenseAxisArray</code> with a <code>Number</code> as an axis will now display a warning. This catches a common error in which users write <code>@variable(model, x[length(S)])</code> instead of <code>@variable(model, x[1:length(S)])</code>.</li><li>The <code>caching_mode</code> argument to <code>Model</code>, for example, <code>Model(caching_mode = MOIU.MANUAL)</code> mode has been removed. For more control over the optimizer, use <code>direct_model</code> instead.</li><li>The previously deprecated <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> functions have been removed. Use <code>lp_sensitivity_report</code> instead.</li><li>The <code>.m</code> fields of <code>NonlinearExpression</code> and <code>NonlinearParameter</code> have been renamed to <code>.model</code>.</li><li>Infinite variable bounds are now ignored. Thus, <code>@variable(model, x &lt;= Inf)</code> will show <code>has_upper_bound(x) == false</code>. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.</li><li>The <code>variable_type</code> and <code>constraint_type</code> functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.</li><li>The internal functions <code>moi_mode</code>, <code>moi_bridge_constraints</code>, <code>moi_add_constraint</code>, and <code>moi_add_to_function_constant</code> are no longer exported.</li><li>The un-used method <code>Containers.generate_container</code> has been deleted.</li><li>The <code>Containers</code> API has been refactored, and <code>_build_ref_sets</code> is now public as <code>Containers.build_ref_sets</code>.</li><li>The <code>parse_constraint_</code> methods for extending <code>@constraint</code> at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.</li></ul><h3 id="Added-22"><a class="docs-heading-anchor" href="#Added-22">Added</a><a class="docs-heading-anchor-permalink" href="#Added-22" title="Permalink"></a></h3><ul><li>The <code>TerminationStatusCode</code> and <code>ResultStatusCode</code> enums are now exported by JuMP. Prefer <code>termination_status(model) == OPTIMAL</code> instead of <code>== MOI.OPTIMAL</code>, although the <code>MOI.</code> prefix way still works.</li><li>Copy a <code>x::DenseAxisArray</code> to an <code>Array</code> by calling <code>Array(x)</code>.</li><li><code>NonlinearExpression</code> is now a subtype of <code>AbstractJuMPScalar</code></li><li>Constraints such as <code>@constraint(model, x + 1 in MOI.Integer())</code> are now supported.</li><li><code>primal_feasibility_report</code> now accepts a function as the first argument.</li><li>Scalar variables <code>@variable(model, x[1:2] in MOI.Integer())</code> creates two variables, both of which are constrained to be in the set <code>MOI.Integer</code>.</li><li>Conic constraints can now be specified as inequalities under a different partial ordering. So <code>@constraint(model, x - y in MOI.Nonnegatives())</code> can now be written as <code>@constraint(model, x &gt;= y, MOI.Nonnegatives())</code>.</li><li>Names are now set for vectorized constraints.</li></ul><h3 id="Fixed-30"><a class="docs-heading-anchor" href="#Fixed-30">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-30" title="Permalink"></a></h3><ul><li>Fixed a performance issue when <code>show</code> was called on a <code>SparseAxisArray</code> with a large number of elements.</li><li>Fixed a bug displaying barrier and simplex iterations in <code>solution_summary</code>.</li><li>Fixed a bug by implementing <code>hash</code> for <code>DenseAxisArray</code> and <code>SparseAxisArray</code>.</li><li>Names are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with <code>direct_model</code>.</li><li><code>MutableArithmetics.Zero</code> is converted into a <code>0.0</code> before being returned to the user. Previously, some calls to <code>@expression</code> would return the undocumented <code>MutableArithmetics.Zero()</code> object. One example is summing over an empty set <code>@expression(model, sum(x[i] for i in 1:0))</code>. You will now get <code>0.0</code> instead.</li><li><code>AffExpr</code> and <code>QuadExpr</code> can now be used with <code>== 0</code> instead of <code>iszero</code>. This fixes a number of issues relating to Julia standard libraries such as <code>LinearAlgebra</code> and <code>SparseArrays</code>.</li><li>Fixed a bug when registering a user-defined function with splatting.</li></ul><h3 id="Other-28"><a class="docs-heading-anchor" href="#Other-28">Other</a><a class="docs-heading-anchor-permalink" href="#Other-28" title="Permalink"></a></h3><ul><li>The documentation is now available as a PDF.</li><li>The documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)</li><li>The documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.</li><li>The documentation is now built with Julia 1.6 instead of 1.0.</li><li>Various error messages have been improved to be more readable.</li></ul><h2 id="Version-0.21.10-(September-4,-2021)"><a class="docs-heading-anchor" href="#Version-0.21.10-(September-4,-2021)">Version 0.21.10 (September 4, 2021)</a><a id="Version-0.21.10-(September-4,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.10-(September-4,-2021)" title="Permalink"></a></h2><h3 id="Added-23"><a class="docs-heading-anchor" href="#Added-23">Added</a><a class="docs-heading-anchor-permalink" href="#Added-23" title="Permalink"></a></h3><ul><li>Added <code>add_NL_expression</code></li><li><code>add_NL_xxx</code> functions now support <code>AffExpr</code> and <code>QuadExpr</code> as terms</li></ul><h3 id="Fixed-31"><a class="docs-heading-anchor" href="#Fixed-31">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-31" title="Permalink"></a></h3><ul><li>Fixed a bug in <code>solution_summary</code></li><li>Fixed a bug in <code>relax_integrality</code></li></ul><h3 id="Other-29"><a class="docs-heading-anchor" href="#Other-29">Other</a><a class="docs-heading-anchor-permalink" href="#Other-29" title="Permalink"></a></h3><ul><li>Improved error message in <code>lp_sensitivity_report</code></li></ul><h2 id="Version-0.21.9-(August-1,-2021)"><a class="docs-heading-anchor" href="#Version-0.21.9-(August-1,-2021)">Version 0.21.9 (August 1, 2021)</a><a id="Version-0.21.9-(August-1,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.9-(August-1,-2021)" title="Permalink"></a></h2><h3 id="Added-24"><a class="docs-heading-anchor" href="#Added-24">Added</a><a class="docs-heading-anchor-permalink" href="#Added-24" title="Permalink"></a></h3><ul><li>Containers now support arbitrary container types by passing the type to the <code>container</code> keyword and overloading <code>Containers.container</code>.</li><li><code>is_valid</code> now supports nonlinear constraints</li><li>Added <code>unsafe_backend</code> for querying the inner-most optimizer of a JuMP model.</li><li>Nonlinear parameters now support the plural <code>@NLparameters</code> macro.</li><li>Containers (for example, <code>DenseAxisArray</code>) can now be used in vector-valued constraints.</li></ul><h3 id="Other-30"><a class="docs-heading-anchor" href="#Other-30">Other</a><a class="docs-heading-anchor-permalink" href="#Other-30" title="Permalink"></a></h3><ul><li>Various improvements to the documentation.</li></ul><h2 id="Version-0.21.8-(May-8,-2021)"><a class="docs-heading-anchor" href="#Version-0.21.8-(May-8,-2021)">Version 0.21.8 (May 8, 2021)</a><a id="Version-0.21.8-(May-8,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.8-(May-8,-2021)" title="Permalink"></a></h2><h3 id="Added-25"><a class="docs-heading-anchor" href="#Added-25">Added</a><a class="docs-heading-anchor-permalink" href="#Added-25" title="Permalink"></a></h3><ul><li>The <code>@constraint</code> macro is now extendable in the same way as <code>@variable</code>.</li><li><code>AffExpr</code> and <code>QuadExpr</code> can now be used in nonlinear macros.</li></ul><h3 id="Fixed-32"><a class="docs-heading-anchor" href="#Fixed-32">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-32" title="Permalink"></a></h3><ul><li>Fixed a bug in <code>lp_sensitivity_report</code>.</li><li>Fixed an inference issue when creating empty <code>SparseAxisArray</code>s.</li></ul><h2 id="Version-0.21.7-(April-12,-2021)"><a class="docs-heading-anchor" href="#Version-0.21.7-(April-12,-2021)">Version 0.21.7 (April 12, 2021)</a><a id="Version-0.21.7-(April-12,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.7-(April-12,-2021)" title="Permalink"></a></h2><h3 id="Added-26"><a class="docs-heading-anchor" href="#Added-26">Added</a><a class="docs-heading-anchor-permalink" href="#Added-26" title="Permalink"></a></h3><ul><li>Added <code>primal_feasibility_report</code>, which can be used to check whether a primal point satisfies primal feasibility.</li><li>Added <code>coefficient</code>, which returns the coefficient associated with a variable in affine and quadratic expressions.</li><li>Added <code>copy_conflict</code>, which returns the IIS of an infeasible model.</li><li>Added <code>solution_summary</code>, which returns (and prints) a struct containing a summary of the solution.</li><li>Allow <code>AbstractVector</code> in vector constraints instead of just <code>Vector</code>.</li><li>Added <code>latex_formulation(model)</code> which returns an object representing the latex formulation of a model. Use <code>print(latex_formulation(model))</code> to print the formulation as a string.</li><li>User-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.</li><li>DenseAxisArray&#39;s now support broadcasting over multiple arrays.</li><li>Container indices can now be iterators of <code>Base.SizeUnknown</code>.</li></ul><h3 id="Fixed-33"><a class="docs-heading-anchor" href="#Fixed-33">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-33" title="Permalink"></a></h3><ul><li>Fixed bug in <code>rad2deg</code> and <code>deg2rad</code> in nonlinear expressions.</li><li>Fixed a MethodError bug in <code>Containers</code> when forcing container type.</li><li>Allow partial slicing of a DenseAxisArray, resolving an issue from 2014.</li><li>Fixed a bug printing variable names in IJulia.</li><li>Ending an IJulia cell with <code>model</code> now prints a summary of the model (like in the REPL) not the latex formulation. Use <code>print(model)</code> to print the latex formulation.</li><li>Fixed a bug when copying models containing nested arrays.</li></ul><h3 id="Other-31"><a class="docs-heading-anchor" href="#Other-31">Other</a><a class="docs-heading-anchor-permalink" href="#Other-31" title="Permalink"></a></h3><ul><li>Tutorials are now part of the documentation, and more refactoring has taken place.</li><li>Added JuliaFormatter added as a code formatter.</li><li>Added some precompilation statements to reduce initial latency.</li><li>Various improvements to error messages to make them more helpful.</li><li>Improved performance of <code>value(::NonlinearExpression)</code>.</li><li>Improved performance of <code>fix(::VariableRef)</code>.</li></ul><h2 id="Version-0.21.6-(January-29,-2021)"><a class="docs-heading-anchor" href="#Version-0.21.6-(January-29,-2021)">Version 0.21.6 (January 29, 2021)</a><a id="Version-0.21.6-(January-29,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.6-(January-29,-2021)" title="Permalink"></a></h2><h3 id="Added-27"><a class="docs-heading-anchor" href="#Added-27">Added</a><a class="docs-heading-anchor-permalink" href="#Added-27" title="Permalink"></a></h3><ul><li>Added support for skew symmetric variables via <code>@variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace())</code>.</li><li><code>lp_sensitivity_report</code> has been added which significantly improves the performance of querying the sensitivity summary of an LP. <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> are deprecated.</li><li>Dual warm-starts are now supported with <code>set_dual_start_value</code> and <code>dual_start_value</code>.</li><li><code>∈</code> (<code>\in&lt;tab&gt;</code>) can now be used in macros instead of <code>=</code> or <code>in</code>.</li><li>Use <code>haskey(model::Model, key::Symbol)</code> to check if a name <code>key</code> is registered in a model.</li><li>Added <code>unregister(model::Model, key::Symbol)</code> to unregister a name <code>key</code> from <code>model</code>.</li><li>Added <code>callback_node_status</code> for use in callbacks.</li><li>Added <code>print_bridge_graph</code> to visualize the bridging graph generated by MathOptInterface.</li><li>Improved error message for containers with duplicate indices.</li></ul><h3 id="Fixed-34"><a class="docs-heading-anchor" href="#Fixed-34">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-34" title="Permalink"></a></h3><ul><li>Various fixes to pass tests on Julia 1.6.</li><li>Fixed a bug in the printing of nonlinear expressions in IJulia.</li><li>Fixed a bug when nonlinear expressions are passed to user-defined functions.</li><li>Some internal functions that were previously exported are now no longer exported.</li><li>Fixed a bug when relaxing a fixed binary variable.</li><li>Fixed a <code>StackOverflowError</code> that occurred when <code>SparseAxisArray</code>s had a large number of elements.</li><li>Removed an unnecessary type assertion in <code>list_of_constraint_types</code>.</li><li>Fixed a bug when copying models with registered expressions.</li></ul><h3 id="Other-32"><a class="docs-heading-anchor" href="#Other-32">Other</a><a class="docs-heading-anchor-permalink" href="#Other-32" title="Permalink"></a></h3><ul><li>The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in <code>/examples</code> have now been moved to <code>/docs/src/examples</code> and rewritten using <code>Literate.jl</code>, and they are now included in the documentation.</li><li>JuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.</li><li>The root cause of a large number of method invalidations has been resolved.</li><li>We switched continuous integration from Travis and Appveyor to GitHub Actions.</li></ul><h2 id="Version-0.21.5-(September-18,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.5-(September-18,-2020)">Version 0.21.5 (September 18, 2020)</a><a id="Version-0.21.5-(September-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.5-(September-18,-2020)" title="Permalink"></a></h2><h3 id="Fixed-35"><a class="docs-heading-anchor" href="#Fixed-35">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-35" title="Permalink"></a></h3><ul><li>Fixed deprecation warnings</li><li>Throw <code>DimensionMismatch</code> for incompatibly sized functions and sets</li><li>Unify treatment of <code>keys(x)</code> on JuMP containers</li></ul><h2 id="Version-0.21.4-(September-14,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.4-(September-14,-2020)">Version 0.21.4 (September 14, 2020)</a><a id="Version-0.21.4-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.4-(September-14,-2020)" title="Permalink"></a></h2><h3 id="Added-28"><a class="docs-heading-anchor" href="#Added-28">Added</a><a class="docs-heading-anchor-permalink" href="#Added-28" title="Permalink"></a></h3><ul><li>Add debug info when adding unsupported constraints</li><li>Add <code>relax_integrality</code> for solving continuous relaxation</li><li>Allow querying constraint conflicts</li></ul><h3 id="Fixed-36"><a class="docs-heading-anchor" href="#Fixed-36">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-36" title="Permalink"></a></h3><ul><li>Dispatch on <code>Real</code> for <code>MOI.submit</code></li><li>Implement <code>copy</code> for <code>CustomSet</code> in tests</li><li>Don&#39;t export private macros</li><li>Fix invalid assertion in nonlinear</li><li>Error if constraint has <code>NaN</code> right-hand side</li><li>Improve speed of tests</li><li>Lots of work modularizing files in <code>/test</code></li><li>Improve line numbers in macro error messages</li><li>Print nonlinear subexpressions</li><li>Various documentation updates</li><li>Dependency updates:<ul><li>Datastructures 0.18</li><li>MathOptFormat v0.5</li><li>Prep for MathOptInterface 0.9.15</li></ul></li></ul><h2 id="Version-0.21.3-(June-18,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.3-(June-18,-2020)">Version 0.21.3 (June 18, 2020)</a><a id="Version-0.21.3-(June-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.3-(June-18,-2020)" title="Permalink"></a></h2><ul><li>Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (#2212).</li><li>Added functions <code>simplex_iterations</code>, <code>barrier_iterations</code> and <code>node_count</code> (#2201).</li><li>Added function <code>reduced_cost</code> (#2205).</li><li>Implemented <code>callback_value</code> for affine and quadratic expressions (#2231).</li><li>Support <code>MutableArithmetics.Zero</code> in objective and constraints (#2219).</li><li>Documentation improvements:<ul><li>Mention tutorials in the docs (#2223).</li><li>Update COIN-OR links (#2242).</li><li>Explicit link to the documentation of <code>MOI.FileFormats</code> (#2253).</li><li>Typo fixes (#2261).</li></ul></li><li>Containers improvements:<ul><li>Fix <code>Base.map</code> for <code>DenseAxisArray</code> (#2235).</li><li>Throw <code>BoundsError</code> if number of indices is incorrect for <code>DenseAxisArray</code> and <code>SparseAxisArray</code> (#2240).</li></ul></li><li>Extensibility improvements:<ul><li>Implement a <code>set_objective</code> method fallback that redirects to <code>set_objective_sense</code> and <code>set_objective_function</code> (#2247).</li><li>Add <code>parse_constraint</code> method with arbitrary number of arguments (#2051).</li><li>Add <code>parse_constraint_expr</code> and <code>parse_constraint_head</code> (#2228).</li></ul></li></ul><h2 id="Version-0.21.2-(April-2,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.2-(April-2,-2020)">Version 0.21.2 (April 2, 2020)</a><a id="Version-0.21.2-(April-2,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.2-(April-2,-2020)" title="Permalink"></a></h2><ul><li>Added <code>relative_gap()</code> to access <code>MOI.RelativeGap()</code> attribute (#2199).</li><li>Documentation fixes:<ul><li>Added link to source for docstrings in the documentation (#2207).</li><li>Added docstring for <code>@variables</code> macro (#2216).</li><li>Typo fixes (#2177, #2184, #2182).</li></ul></li><li>Implementation of methods for Base functions:<ul><li>Implemented <code>Base.empty!</code> for <code>JuMP.Model</code> (#2198).</li><li>Implemented <code>Base.conj</code> for JuMP scalar types (#2209).</li></ul></li></ul><h3 id="Fixed-37"><a class="docs-heading-anchor" href="#Fixed-37">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-37" title="Permalink"></a></h3><ul><li>Fixed sum of expression with scalar product in macro (#2178).</li><li>Fixed writing of nonlinear models to MathOptFormat (#2181).</li><li>Fixed construction of empty SparseAxisArray (#2179).</li><li>Fixed constraint with zero function (#2188).</li></ul><h2 id="Version-0.21.1-(Feb-18,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.1-(Feb-18,-2020)">Version 0.21.1 (Feb 18, 2020)</a><a id="Version-0.21.1-(Feb-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.1-(Feb-18,-2020)" title="Permalink"></a></h2><ul><li>Improved the clarity of the <code>with_optimizer</code> deprecation warning.</li></ul><h2 id="Version-0.21.0-(Feb-16,-2020)"><a class="docs-heading-anchor" href="#Version-0.21.0-(Feb-16,-2020)">Version 0.21.0 (Feb 16, 2020)</a><a id="Version-0.21.0-(Feb-16,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.21.0-(Feb-16,-2020)" title="Permalink"></a></h2><h3 id="Breaking-6"><a class="docs-heading-anchor" href="#Breaking-6">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-6" title="Permalink"></a></h3><ul><li><p>Deprecated <code>with_optimizer</code> (#2090, #2084, #2141). You can replace <code>with_optimizer</code> by either nothing, <code>optimizer_with_attributes</code> or a closure:</p><ul><li>replace <code>with_optimizer(Ipopt.Optimizer)</code> by <code>Ipopt.Optimizer</code>.</li><li>replace <code>with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0)</code> by <code>optimizer_with_attributes(Ipopt.Optimizer, &quot;max_cpu_time&quot; =&gt; 60.0)</code>.</li><li>replace <code>with_optimizer(Gurobi.Optimizer, env)</code> by <code>() -&gt; Gurobi.Optimizer(env)</code>.</li><li>replace <code>with_optimizer(Gurobi.Optimizer, env, Presolve=0)</code> by <code>optimizer_with_attributes(() -&gt; Gurobi.Optimizer(env), &quot;Presolve&quot; =&gt; 0)</code>.</li></ul><p>alternatively to <code>optimizer_with_attributes</code>, you can also set the attributes separately with <code>set_optimizer_attribute</code>.</p></li><li><p>Renamed <code>set_parameter</code> and <code>set_parameters</code> to <code>set_optimizer_attribute</code> and <code>set_optimizer_attributes</code> (#2150).</p></li><li><p>Broadcast should now be explicit inside macros. <code>@SDconstraint(model, x &gt;= 1)</code> and <code>@constraint(model, x + 1 in SecondOrderCone())</code> now throw an error instead of broadcasting <code>1</code> along the dimension of <code>x</code> (#2107).</p></li><li><p><code>@SDconstraint(model, x &gt;= 0)</code> is now equivalent to <code>@constraint(model, x in PSDCone())</code> instead of <code>@constraint(model, (x .- 0) in PSDCone())</code> (#2107).</p></li><li><p>The macros now create the containers with <code>map</code> instead of <code>for</code> loops, as a consequence, containers created by <code>@expression</code> can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where <code>@expression</code> could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (#2070).</p></li><li><p>Calling <code>deepcopy(::AbstractModel)</code> now throws an error.</p></li><li><p>The constraint name is now printed in the model string (#2108).</p></li></ul><h3 id="Added-29"><a class="docs-heading-anchor" href="#Added-29">Added</a><a class="docs-heading-anchor-permalink" href="#Added-29" title="Permalink"></a></h3><ul><li>Added support for solver-independent and solver-specific callbacks (#2101).</li><li>Added <code>write_to_file</code> and <code>read_from_file</code>, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (#2114).</li><li>Added support for complementarity constraints (#2132).</li><li>Added support for indicator constraints (#2092).</li><li>Added support for querying multiple solutions with the <code>result</code> keyword (#2100).</li><li>Added support for constraining variables on creation (#2128).</li><li>Added method <code>delete</code> that deletes a vector of variables at once if it is supported by the underlying solver (#2135).</li><li>The arithmetic between JuMP expression has be refactored into the MutableArithmetics package (#2107).</li><li>Improved error on complex values in NLP (#1978).</li><li>Added an example of column generation (#2010).</li></ul><h3 id="Fixed-38"><a class="docs-heading-anchor" href="#Fixed-38">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-38" title="Permalink"></a></h3><ul><li>Incorrect coefficients generated when using Symmetric variables (#2102)</li></ul><h2 id="Version-0.20.1-(Oct-18,-2019)"><a class="docs-heading-anchor" href="#Version-0.20.1-(Oct-18,-2019)">Version 0.20.1 (Oct 18, 2019)</a><a id="Version-0.20.1-(Oct-18,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.20.1-(Oct-18,-2019)" title="Permalink"></a></h2><ul><li>Add sections on <code>@variables</code> and <code>@constraints</code> in the documentation (#2062).</li><li>Fixed product of sparse matrices for Julia v1.3 (#2063).</li><li>Added <code>set_objective_coefficient</code> to modify the coefficient of a linear term of the objective function (#2008).</li><li>Added <code>set_time_limit_sec</code>, <code>unset_time_limit_sec</code> and <code>time_limit_sec</code> to set and query the time limit for the solver in seconds (#2053).</li></ul><h2 id="Version-0.20.0-(Aug-24,-2019)"><a class="docs-heading-anchor" href="#Version-0.20.0-(Aug-24,-2019)">Version 0.20.0 (Aug 24, 2019)</a><a id="Version-0.20.0-(Aug-24,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.20.0-(Aug-24,-2019)" title="Permalink"></a></h2><ul><li>Documentation updates.</li><li>Numerous bug fixes.</li><li>Better error messages (#1977, #1978, #1997, #2017).</li><li>Performance improvements (#1947, #2032).</li><li>Added LP sensitivity summary functions <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> (#1917).</li><li>Added functions <code>dual_objective_value</code>, <code>raw_status</code> and <code>set_parameter</code>.</li><li>Added function <code>set_objective_coefficient</code> to modify the coefficient of a linear term of the objective (#2008).</li><li>Added functions <code>set_normalized_rhs</code>, <code>normalized_rhs</code>, and <code>add_to_function_constant</code> to modify and get the constant part of a constraint (#1935, #1960).</li><li>Added functions <code>set_normalized_coefficient</code> and <code>normalized_coefficient</code> to modify and get the coefficient of a linear term of a constraint (#1935, #1960).</li><li>Numerous other improvements in MOI 0.9, see the <code>NEWS.md</code> file of MOI for more details.</li></ul><h2 id="Version-0.19.2-(June-8,-2019)"><a class="docs-heading-anchor" href="#Version-0.19.2-(June-8,-2019)">Version 0.19.2 (June 8, 2019)</a><a id="Version-0.19.2-(June-8,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.19.2-(June-8,-2019)" title="Permalink"></a></h2><ul><li>Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.</li></ul><h2 id="Version-0.19.1-(May-12,-2019)"><a class="docs-heading-anchor" href="#Version-0.19.1-(May-12,-2019)">Version 0.19.1 (May 12, 2019)</a><a id="Version-0.19.1-(May-12,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.19.1-(May-12,-2019)" title="Permalink"></a></h2><ul><li>Usability and performance improvements.</li><li>Bug fixes.</li></ul><h2 id="Version-0.19.0-(February-15,-2019)"><a class="docs-heading-anchor" href="#Version-0.19.0-(February-15,-2019)">Version 0.19.0 (February 15, 2019)</a><a id="Version-0.19.0-(February-15,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.19.0-(February-15,-2019)" title="Permalink"></a></h2><p><strong>JuMP 0.19 contains significant breaking changes.</strong></p><h3 id="Breaking-7"><a class="docs-heading-anchor" href="#Breaking-7">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-7" title="Permalink"></a></h3><ul><li><p>JuMP&#39;s abstraction layer for communicating with solvers changed from <a href="https://github.com/JuliaOpt/MathProgBase.jl">MathProgBase</a> (MPB) to <a href="https://github.com/JuliaOpt/MathOptInterface.jl">MathOptInterface</a> (MOI). MOI addresses many longstanding design issues. (See @mlubin&#39;s <a href="https://www.juliaopt.org/meetings/bordeaux2018/lubin.pdf">slides</a> from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the <a href="https://www.juliaopt.org/JuMP.jl/dev/installation/">installation guide</a> for a list of solvers that have and have not yet been updated.</p></li><li><p>Most solvers have been renamed to <code>PackageName.Optimizer</code>. For example, <code>GurobiSolver()</code> is now <code>Gurobi.Optimizer</code>.</p></li><li><p>Solvers are no longer added to a model via <code>Model(solver = XXX(kwargs...))</code>. Instead use <code>Model(with_optimizer(XXX, kwargs...))</code>. For example, <code>Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0))</code>.</p></li><li><p>JuMP containers (for example, the objects returned by <code>@variable</code>) have been redesigned. <code>Containers.SparseAxisArray</code> replaces <code>JuMPDict</code>, <code>JuMPArray</code> was rewritten (inspired by <code>AxisArrays</code>) and renamed <code>Containers.DenseAxisArray</code>, and you can now request a container type with the <code>container=</code> keyword to the macros. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/variables/#Variable-containers-1">documentation</a> for more details.</p></li><li><p>The statuses returned by solvers have changed. See the possible status values <a href="https://www.juliaopt.org/MathOptInterface.jl/stable/apireference.html#Termination-Status-1">here</a>. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).</p></li><li><p>Starting values are separate from result values. Use <code>value</code> to query the value of a variable in a solution. Use <code>start_value</code> and <code>set_start_value</code> to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.</p></li><li><p>The data structures for affine and quadratic expressions <code>AffExpr</code> and <code>QuadExpr</code> have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.</p></li><li><p><code>JuMPNLPEvaluator</code> no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.</p></li><li><p>JuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)</p></li><li><p>The <code>norm()</code> syntax is no longer recognized inside macros. Use the <code>SecondOrderCone()</code> set instead.</p></li><li><p>JuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.</p></li><li><p>The symbols <code>:Min</code> and <code>:Max</code> are no longer used as optimization senses. Instead, JuMP uses the <code>OptimizationSense</code> enum from MathOptInterface. <code>@objective(model, Max, ...)</code>, <code>@objective(model, Min, ...)</code>, <code>@NLobjective(model, Max, ...)</code>, and <code>@objective(model, Min, ...)</code> remain valid, but <code>@objective(m, :Max, ...)</code> is no longer accepted.</p></li><li><p>The sign conventions for duals has changed in some cases for consistency with conic duality (see the <a href="https://www.juliaopt.org/MathOptInterface.jl/v0.6.2/apimanual.html#Duals-1">documentation</a>). The <code>shadow_price</code> helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.</p></li><li><p><code>@constraintref</code> is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,</p><pre><code class="language-julia hljs">constraints = Dict() # Optionally, specify types for improved performance.
 for i in 1:N
   constraints[i] = @constraint(model, ...)
-end</code></pre></li><li><p>The <code>lowerbound</code>, <code>upperbound</code>, and <code>basename</code> keyword arguments to the <code>@variable</code> macro have been renamed to <code>lower_bound</code>, <code>upper_bound</code>, and <code>base_name</code>, for consistency with JuMP&#39;s new <a href="https://www.juliaopt.org/JuMP.jl/dev/style/">style recommendations</a>.</p></li><li><p>We rely on broadcasting syntax to apply accessors to collections of variables, for example, <code>value.(x)</code> instead of <code>getvalue(x)</code> for collections. (Use <code>value(x)</code> when <code>x</code> is a scalar object.)</p></li></ul><h3 id="Added-30"><a class="docs-heading-anchor" href="#Added-30">Added</a><a class="docs-heading-anchor-permalink" href="#Added-30" title="Permalink"></a></h3><ul><li><p>Splatting (like <code>f(x...)</code>) is recognized in restricted settings in nonlinear expressions.</p></li><li><p>Support for deleting constraints and variables.</p></li><li><p>The documentation has been completely rewritten using docstrings and Documenter.</p></li><li><p>Support for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).</p></li><li><p>Significantly improved support for modeling new types of constraints and for extending JuMP&#39;s macros.</p></li><li><p>Support for providing dual warm starts.</p></li><li><p>Improved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).</p></li><li><p>Explicit control of whether symmetry-enforcing constraints are added to PSD constraints.</p></li><li><p>Support for modeling exponential cones.</p></li><li><p>Significant improvements in internal code quality and testing.</p></li><li><p>Style and naming guidelines.</p></li><li><p>Direct mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/solvers/">documentation</a>.</p></li></ul><h3 id="Regressions"><a class="docs-heading-anchor" href="#Regressions">Regressions</a><a id="Regressions-1"></a><a class="docs-heading-anchor-permalink" href="#Regressions" title="Permalink"></a></h3><p>There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):</p><ul><li><p>Performance regressions in model generation (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1403">issue</a>). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.</p></li><li><p>Fast incremental NLP solves are not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1185">issue</a>).</p></li><li><p>We do not yet have an implementation of solver-specific callbacks.</p></li><li><p>The column generation syntax in <code>@variable</code> has been removed (that is, the <code>objective</code>, <code>coefficients</code>, and <code>inconstraints</code> keyword arguments). Support for column generation will be re-introduced in a future release.</p></li><li><p>The ability to solve the continuous relaxation (that is, via <code>solve(model; relaxation = true)</code>) is not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1611">issue</a>).</p></li></ul><h2 id="Version-0.18.5-(December-1,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.5-(December-1,-2018)">Version 0.18.5 (December 1, 2018)</a><a id="Version-0.18.5-(December-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.5-(December-1,-2018)" title="Permalink"></a></h2><ul><li>Support views in some derivative evaluation functions.</li><li>Improved compatibility with PackageCompiler.</li></ul><h2 id="Version-0.18.4-(October-8,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.4-(October-8,-2018)">Version 0.18.4 (October 8, 2018)</a><a id="Version-0.18.4-(October-8,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.4-(October-8,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in model printing on Julia 0.7 and 1.0.</li></ul><h2 id="Version-0.18.3-(October-1,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.3-(October-1,-2018)">Version 0.18.3 (October 1, 2018)</a><a id="Version-0.18.3-(October-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.3-(October-1,-2018)" title="Permalink"></a></h2><ul><li>Add support for Julia v1.0 (Thanks @ExpandingMan)</li><li>Fix matrix expressions with quadratic functions (#1508)</li></ul><h2 id="Version-0.18.2-(June-10,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.2-(June-10,-2018)">Version 0.18.2 (June 10, 2018)</a><a id="Version-0.18.2-(June-10,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.2-(June-10,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in second-order derivatives when expressions are present (#1319)</li><li>Fix a bug in <code>@constraintref</code> (#1330)</li></ul><h2 id="Version-0.18.1-(April-9,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.1-(April-9,-2018)">Version 0.18.1 (April 9, 2018)</a><a id="Version-0.18.1-(April-9,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.1-(April-9,-2018)" title="Permalink"></a></h2><ul><li>Fix for nested tuple destructuring (#1193)</li><li>Preserve internal model when relaxation=true (#1209)</li><li>Minor bug fixes and updates for example</li></ul><h2 id="Version-0.18.0-(July-27,-2017)"><a class="docs-heading-anchor" href="#Version-0.18.0-(July-27,-2017)">Version 0.18.0 (July 27, 2017)</a><a id="Version-0.18.0-(July-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.0-(July-27,-2017)" title="Permalink"></a></h2><ul><li>Drop support for Julia 0.5.</li><li>Update for ForwardDiff 0.5.</li><li>Minor bug fixes.</li></ul><h2 id="Version-0.17.1-(June-9,-2017)"><a class="docs-heading-anchor" href="#Version-0.17.1-(June-9,-2017)">Version 0.17.1 (June 9, 2017)</a><a id="Version-0.17.1-(June-9,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.17.1-(June-9,-2017)" title="Permalink"></a></h2><ul><li>Use of <code>constructconstraint!</code> in <code>@SDconstraint</code>.</li><li>Minor bug fixes.</li></ul><h2 id="Version-0.17.0-(May-27,-2017)"><a class="docs-heading-anchor" href="#Version-0.17.0-(May-27,-2017)">Version 0.17.0 (May 27, 2017)</a><a id="Version-0.17.0-(May-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.17.0-(May-27,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: Mixing quadratic and conic constraints is no longer supported.</li><li><strong>Breaking change</strong>: The <code>getvariable</code> and <code>getconstraint</code> functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name <code>x</code>, one should now write <code>m[:x]</code> instead of <code>getvariable(m, :x)</code>. As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.</li><li>Addition of the <code>getobjectivebound</code> function that mirrors the functionality of the MathProgBase <code>getobjbound</code> function except that it takes into account transformations performed by JuMP.</li><li>Minor bug fixes.</li></ul><p>The following changes are primarily of interest to developers of JuMP extensions:</p><ul><li>The new syntax <code>@constraint(model, expr in Cone)</code> creates the constraint ensuring that <code>expr</code> is inside <code>Cone</code>. The <code>Cone</code> argument is passed to <code>constructconstraint!</code> which enables the call to the dispatched to an extension.</li><li>The <code>@variable</code> macro now calls <code>constructvariable!</code> instead of directly calling the <code>Variable</code> constructor. Extra arguments and keyword arguments passed to <code>@variable</code> are passed to <code>constructvariable!</code> which enables the call to be dispatched to an extension.</li><li>Refactor the internal function <code>conicdata</code> (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.</li></ul><h2 id="Version-0.16.2-(March-28,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.2-(March-28,-2017)">Version 0.16.2 (March 28, 2017)</a><a id="Version-0.16.2-(March-28,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.2-(March-28,-2017)" title="Permalink"></a></h2><ul><li>Minor bug fixes and printing tweaks</li><li>Address deprecation warnings for Julia 0.6</li></ul><h2 id="Version-0.16.1-(March-7,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.1-(March-7,-2017)">Version 0.16.1 (March 7, 2017)</a><a id="Version-0.16.1-(March-7,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.1-(March-7,-2017)" title="Permalink"></a></h2><ul><li>Better support for <code>AbstractArray</code> in JuMP (Thanks @tkoolen)</li><li>Minor bug fixes</li></ul><h2 id="Version-0.16.0-(February-23,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.0-(February-23,-2017)">Version 0.16.0 (February 23, 2017)</a><a id="Version-0.16.0-(February-23,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.0-(February-23,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling <code>solve()</code> will result in an error.</li><li><strong>Breaking change</strong>: User-defined functions are no longer global. The first argument to <code>JuMP.register</code> is now a JuMP <code>Model</code> object within whose scope the function will be registered. Calling <code>JuMP.register</code> without a <code>Model</code> now produces an error.</li><li><strong>Breaking change</strong>: Use the new <code>JuMP.fix</code> method to fix a variable to a value or to update the value to which a variable is fixed. Calling <code>setvalue</code> on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)</li><li>Nonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)</li><li>New <code>category</code> keyword to <code>@variable</code>. Used for specifying categories of anonymous variables.</li><li>Compatibility with Julia 0.6-dev.</li><li>Minor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)</li></ul><h2 id="Version-0.15.1-(January-31,-2017)"><a class="docs-heading-anchor" href="#Version-0.15.1-(January-31,-2017)">Version 0.15.1 (January 31, 2017)</a><a id="Version-0.15.1-(January-31,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.15.1-(January-31,-2017)" title="Permalink"></a></h2><ul><li>Bugfix for <code>@LinearConstraints</code> and friends</li></ul><h2 id="Version-0.15.0-(December-22,-2016)"><a class="docs-heading-anchor" href="#Version-0.15.0-(December-22,-2016)">Version 0.15.0 (December 22, 2016)</a><a id="Version-0.15.0-(December-22,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.15.0-(December-22,-2016)" title="Permalink"></a></h2><ul><li>Julia 0.5.0 is the minimum required version for this release.</li><li>Document support for BARON solver</li><li>Enable info callbacks in more states than before, for example, for recording solutions. New <code>when</code> argument to <code>addinfocallback</code> (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/814">#814</a>, thanks @yeesian)</li><li>Improved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:<ul><li>When multiple variables in a model are given the same name</li><li>When non-symbols are used as names, for example, <code>@variable(m, x[1][1:N])</code></li></ul></li><li>Improvements in iterating over JuMP containers (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/836">#836</a>, thanks @IssamT)</li><li>Support for writing variable names in .lp file output (Thanks @leethargo)</li><li>Support for querying duals to SDP problems (Thanks @blegat)</li><li>The comprehension syntax with curly braces <code>sum{}</code>, <code>prod{}</code>, and <code>norm2{}</code> has been deprecated in favor of Julia&#39;s native comprehension syntax <code>sum()</code>, <code>prod()</code> and <code>norm()</code> as previously announced. (For early adopters of the new syntax, <code>norm2()</code> was renamed to <code>norm()</code> without deprecation.)</li><li>Unit tests rewritten to use Base.Test instead of FactCheck</li><li>Improved support for operations with matrices of JuMP types (Thanks @ExpandingMan)</li><li>The syntax to halt a solver from inside a callback has changed from <code>throw(CallbackAbort())</code> to <code>return JuMP.StopTheSolver</code></li><li>Minor bug fixes</li></ul><h2 id="Version-0.14.2-(December-12,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.2-(December-12,-2016)">Version 0.14.2 (December 12, 2016)</a><a id="Version-0.14.2-(December-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.2-(December-12,-2016)" title="Permalink"></a></h2><ul><li>Allow singleton anonymous variables (includes bugfix)</li></ul><h2 id="Version-0.14.1-(September-12,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.1-(September-12,-2016)">Version 0.14.1 (September 12, 2016)</a><a id="Version-0.14.1-(September-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.1-(September-12,-2016)" title="Permalink"></a></h2><ul><li>More consistent handling of states in informational callbacks, includes a new <code>when</code> parameter to <code>addinfocallback</code> for specifying in which state an informational callback should be called.</li></ul><h2 id="Version-0.14.0-(August-7,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.0-(August-7,-2016)">Version 0.14.0 (August 7, 2016)</a><a id="Version-0.14.0-(August-7,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.0-(August-7,-2016)" title="Permalink"></a></h2><ul><li>Compatibility with Julia 0.5 and ForwardDiff 0.2</li><li>Support for &quot;anonymous&quot; variables, constraints, expressions, and parameters, for example, <code>x = @variable(m, [1:N])</code> instead of <code>@variable(m, x[1:N])</code></li><li>Support for retrieving constraints from a model by name via <code>getconstraint</code></li><li><code>@NLconstraint</code> now returns constraint references (as expected).</li><li>Support for vectorized expressions within lazy constraints</li><li>On Julia 0.5, parse new comprehension syntax <code>sum(x[i] for i in 1:N if isodd(i))</code> instead of <code>sum{ x[i], i in 1:N; isodd(i) }</code>. The old syntax with curly braces will be deprecated in JuMP 0.15.</li><li>Now possible to provide nonlinear expressions as &quot;raw&quot; Julia <code>Expr</code> objects instead of using JuMP&#39;s nonlinear macros. This input format is useful for programmatically generated expressions.</li><li><code>s/Mathematical Programming/Mathematical Optimization/</code></li><li>Support for local cuts (Thanks to @madanim, Mehdi Madani)</li><li>Document Xpress interface developed by @joaquimg, Joaquim Dias Garcia</li><li>Minor bug and deprecation fixes (Thanks @odow, @jrevels)</li></ul><h2 id="Version-0.13.2-(May-16,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.2-(May-16,-2016)">Version 0.13.2 (May 16, 2016)</a><a id="Version-0.13.2-(May-16,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.2-(May-16,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for MathProgBase</li></ul><h2 id="Version-0.13.1-(May-3,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.1-(May-3,-2016)">Version 0.13.1 (May 3, 2016)</a><a id="Version-0.13.1-(May-3,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.1-(May-3,-2016)" title="Permalink"></a></h2><ul><li>Fix broken deprecation for <code>registerNLfunction</code>.</li></ul><h2 id="Version-0.13.0-(April-29,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.0-(April-29,-2016)">Version 0.13.0 (April 29, 2016)</a><a id="Version-0.13.0-(April-29,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.0-(April-29,-2016)" title="Permalink"></a></h2><ul><li>Most exported methods and macros have been renamed to avoid camelCase. See the list of changes <a href="https://github.com/JuliaOpt/JuMP.jl/blob/e53d0db67cde2a4b80d0c1281f4b49eb0128a1f5/src/deprecated.jl#L30">here</a>. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.</li><li>Specify variable lower/upper bounds in <code>@variable</code> using the <code>lowerbound</code> and <code>upperbound</code> keyword arguments.</li><li>Change name printed for variable using the <code>basename</code> keyword argument to <code>@variable</code>.</li><li>New <code>@variables</code> macro allows multi-line declaration of groups of variables.</li><li>A number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was <a href="https://youtu.be/qF1lZPJ3a5A">recorded</a> live.</li><li>Compatibility fixes with Julia 0.5.</li><li>The &quot;end&quot; indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see <a href="https://github.com/JuliaOpt/JuMP.jl/issues/730">here</a>.</li></ul><h2 id="Version-0.12.2-(March-9,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.2-(March-9,-2016)">Version 0.12.2 (March 9, 2016)</a><a id="Version-0.12.2-(March-9,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.2-(March-9,-2016)" title="Permalink"></a></h2><ul><li>Small fixes for nonlinear optimization</li></ul><h2 id="Version-0.12.1-(March-1,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.1-(March-1,-2016)">Version 0.12.1 (March 1, 2016)</a><a id="Version-0.12.1-(March-1,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.1-(March-1,-2016)" title="Permalink"></a></h2><ul><li>Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)</li></ul><h2 id="Version-0.12.0-(February-27,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.0-(February-27,-2016)">Version 0.12.0 (February 27, 2016)</a><a id="Version-0.12.0-(February-27,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.0-(February-27,-2016)" title="Permalink"></a></h2><ul><li>The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:<ul><li><code>@defExpr</code> and <code>@defNLExpr</code> now take the model as the first argument. The previous one-argument version of <code>@defExpr</code> is deprecated; all expressions should be named. For example, replace <code>@defExpr(2x+y)</code> with <code>@defExpr(jump_model, my_expr, 2x+y)</code>.</li><li>JuMP no longer uses Julia&#39;s variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.</li><li>Support for user-defined functions integrated within nonlinear JuMP expressions.</li></ul></li><li>Replaced iteration over <code>AffExpr</code> with <code>Number</code>-like scalar iteration; previous iteration behavior is now available via <code>linearterms(::AffExpr)</code>.</li><li>Stopping the solver via <code>throw(CallbackAbort())</code> from a callback no longer triggers an exception. Instead, <code>solve()</code> returns <code>UserLimit</code> status.</li><li><code>getDual()</code> now works for conic problems (Thanks @emreyamangil.)</li></ul><h2 id="Version-0.11.3-(February-4,-2016)"><a class="docs-heading-anchor" href="#Version-0.11.3-(February-4,-2016)">Version 0.11.3 (February 4, 2016)</a><a id="Version-0.11.3-(February-4,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.3-(February-4,-2016)" title="Permalink"></a></h2><ul><li>Bug-fix for problems with quadratic objectives and semidefinite constraints</li></ul><h2 id="Version-0.11.2-(January-14,-2016)"><a class="docs-heading-anchor" href="#Version-0.11.2-(January-14,-2016)">Version 0.11.2 (January 14, 2016)</a><a id="Version-0.11.2-(January-14,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.2-(January-14,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for Mosek</li></ul><h2 id="Version-0.11.1-(December-1,-2015)"><a class="docs-heading-anchor" href="#Version-0.11.1-(December-1,-2015)">Version 0.11.1 (December 1, 2015)</a><a id="Version-0.11.1-(December-1,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.1-(December-1,-2015)" title="Permalink"></a></h2><ul><li>Remove usage of <code>@compat</code> in tests.</li><li>Fix updating quadratic objectives for nonlinear models.</li></ul><h2 id="Version-0.11.0-(November-30,-2015)"><a class="docs-heading-anchor" href="#Version-0.11.0-(November-30,-2015)">Version 0.11.0 (November 30, 2015)</a><a id="Version-0.11.0-(November-30,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.0-(November-30,-2015)" title="Permalink"></a></h2><ul><li>Julia 0.4.0 is the minimum required version for this release.</li><li>Fix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.</li><li>Addition of the <code>solve(m::Model, relaxation=true)</code> keyword argument to solve the standard continuous relaxation of model <code>m</code></li><li>The <code>getConstraintBounds()</code> method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.</li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="Version-0.10.3-(November-20,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.3-(November-20,-2015)">Version 0.10.3 (November 20, 2015)</a><a id="Version-0.10.3-(November-20,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.3-(November-20,-2015)" title="Permalink"></a></h2><ul><li>Fix a rare error when parsing quadratic expressions</li><li>Fix <code>Variable()</code> constructor with default arguments</li><li>Detect unrecognized keywords in <code>solve()</code></li></ul><h2 id="Version-0.10.2-(September-28,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.2-(September-28,-2015)">Version 0.10.2 (September 28, 2015)</a><a id="Version-0.10.2-(September-28,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.2-(September-28,-2015)" title="Permalink"></a></h2><ul><li>Fix for deprecation warnings</li></ul><h2 id="Version-0.10.1-(September-3,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.1-(September-3,-2015)">Version 0.10.1 (September 3, 2015)</a><a id="Version-0.10.1-(September-3,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.1-(September-3,-2015)" title="Permalink"></a></h2><ul><li>Fixes for ambiguity warnings.</li><li>Fix for breaking change in precompilation syntax in Julia 0.4-pre</li></ul><h2 id="Version-0.10.0-(August-31,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.0-(August-31,-2015)">Version 0.10.0 (August 31, 2015)</a><a id="Version-0.10.0-(August-31,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.0-(August-31,-2015)" title="Permalink"></a></h2><ul><li>Support (on Julia 0.4 and later) for conditions in indexing <code>@defVar</code> and <code>@addConstraint</code> constructs, for example, <code>@defVar(m, x[i=1:5,j=1:5; i+j &gt;= 3])</code></li><li>Support for vectorized operations on Variables and expressions. See the documentation for details.</li><li>New <code>getVar()</code> method to access variables in a model by name</li><li>Support for semidefinite programming.</li><li>Dual solutions are now available for general nonlinear problems. You may call <code>getDual</code> on a reference object for a nonlinear constraint, and <code>getDual</code> on a variable object for Lagrange multipliers from active bounds.</li><li>Introduce warnings for two common performance traps: too many calls to <code>getValue()</code> on a collection of variables and use of the <code>+</code> operator in a loop to sum expressions.</li><li>Second-order cone constraints can be written directly with the <code>norm()</code> and <code>norm2{}</code> syntax.</li><li>Implement MathProgBase interface for querying Hessian-vector products.</li><li>Iteration over <code>JuMPContainer</code>s is deprecated; instead, use the <code>keys</code> and <code>values</code> functions, and <code>zip(keys(d),values(d))</code> for the old behavior.</li><li><code>@defVar</code> returns <code>Array{Variable,N}</code> when each of <code>N</code> index sets are of the form <code>1:nᵢ</code>.</li><li>Module precompilation: on Julia 0.4 and later, <code>using JuMP</code> is now much faster.</li></ul><h2 id="Version-0.9.3-(August-11,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.3-(August-11,-2015)">Version 0.9.3 (August 11, 2015)</a><a id="Version-0.9.3-(August-11,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.3-(August-11,-2015)" title="Permalink"></a></h2><ul><li>Fixes for FactCheck testing on julia v0.4.</li></ul><h2 id="Version-0.9.2-(June-27,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.2-(June-27,-2015)">Version 0.9.2 (June 27, 2015)</a><a id="Version-0.9.2-(June-27,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.2-(June-27,-2015)" title="Permalink"></a></h2><ul><li>Fix bug in @addConstraints.</li></ul><h2 id="Version-0.9.1-(April-25,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.1-(April-25,-2015)">Version 0.9.1 (April 25, 2015)</a><a id="Version-0.9.1-(April-25,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.1-(April-25,-2015)" title="Permalink"></a></h2><ul><li>Fix for Julia 0.4-dev.</li><li>Small infrastructure improvements for extensions.</li></ul><h2 id="Version-0.9.0-(April-18,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.0-(April-18,-2015)">Version 0.9.0 (April 18, 2015)</a><a id="Version-0.9.0-(April-18,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.0-(April-18,-2015)" title="Permalink"></a></h2><ul><li>Comparison operators for constructing constraints (for example, <code>2x &gt;= 1</code>) have been deprecated. Instead, construct the constraints explicitly in the <code>@addConstraint</code> macro to add them to the model, or in the <code>@LinearConstraint</code> macro to create a stand-alone linear constraint instance.</li><li><code>getValue()</code> method implemented to compute the value of a nonlinear subexpression</li><li>JuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.</li><li>A number of performance improvements in ReverseDiffSparse for computing derivatives.</li><li><code>MathProgBase.getsolvetime(m)</code> now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)</li><li>Formatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).</li></ul><h2 id="Version-0.8.0-(February-17,-2015)"><a class="docs-heading-anchor" href="#Version-0.8.0-(February-17,-2015)">Version 0.8.0 (February 17, 2015)</a><a id="Version-0.8.0-(February-17,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.8.0-(February-17,-2015)" title="Permalink"></a></h2><ul><li>Nonlinear subexpressions now supported with the <code>@defNLExpr</code> macro.</li><li>SCS supported for solving second-order conic problems.</li><li><code>setXXXCallback</code> family deprecated in favor of <code>addXXXCallback</code>.</li><li>Multiple callbacks of the same type can be registered.</li><li>Added support for informational callbacks via <code>addInfoCallback</code>.</li><li>A <code>CallbackAbort</code> exception can be thrown from callback to safely exit optimization.</li></ul><h2 id="Version-0.7.4-(February-4,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.4-(February-4,-2015)">Version 0.7.4 (February 4, 2015)</a><a id="Version-0.7.4-(February-4,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.4-(February-4,-2015)" title="Permalink"></a></h2><ul><li>Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.</li><li>Two-sided nonlinear constraints are supported.</li><li>Methods for accessing the number of variables and constraints in a model are renamed.</li><li>New default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.</li><li>Small bug fixes.</li></ul><h2 id="Version-0.7.3-(January-14,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.3-(January-14,-2015)">Version 0.7.3 (January 14, 2015)</a><a id="Version-0.7.3-(January-14,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.3-(January-14,-2015)" title="Permalink"></a></h2><ul><li>Fix a method ambiguity conflict with Compose.jl (cosmetic fix)</li></ul><h2 id="Version-0.7.2-(January-9,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.2-(January-9,-2015)">Version 0.7.2 (January 9, 2015)</a><a id="Version-0.7.2-(January-9,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.2-(January-9,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in <code>sum(::JuMPDict)</code></li><li>Added the <code>setCategory</code> function to change a variables category (for example, continuous or binary)</li></ul><p>after construction, and <code>getCategory</code> to retrieve the variable category.</p><h2 id="Version-0.7.1-(January-2,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.1-(January-2,-2015)">Version 0.7.1 (January 2, 2015)</a><a id="Version-0.7.1-(January-2,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.1-(January-2,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.</li></ul><h2 id="Version-0.7.0-(December-29,-2014)"><a class="docs-heading-anchor" href="#Version-0.7.0-(December-29,-2014)">Version 0.7.0 (December 29, 2014)</a><a id="Version-0.7.0-(December-29,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.0-(December-29,-2014)" title="Permalink"></a></h2><h3 id="Linear/quadratic/conic-programming"><a class="docs-heading-anchor" href="#Linear/quadratic/conic-programming">Linear/quadratic/conic programming</a><a id="Linear/quadratic/conic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Linear/quadratic/conic-programming" title="Permalink"></a></h3><ul><li><strong>Breaking change</strong>: The syntax for column-wise model generation has been changed to use keyword arguments in <code>@defVar</code>.</li><li>On Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.</li><li>ECOS supported for solving second-order conic problems.</li></ul><h3 id="_nonlinear_programming_release_notes"><a class="docs-heading-anchor" href="#_nonlinear_programming_release_notes">Nonlinear programming</a><a id="_nonlinear_programming_release_notes-1"></a><a class="docs-heading-anchor-permalink" href="#_nonlinear_programming_release_notes" title="Permalink"></a></h3><ul><li>Support for skipping model generation when solving a sequence of nonlinear models with changing data.</li><li>Fix a memory leak when solving a sequence of nonlinear models.</li><li>The <code>@addNLConstraint</code> macro now supports the three-argument version to define sets of nonlinear constraints.</li><li>KNITRO supported as a nonlinear solver.</li><li>Speed improvements for model generation.</li><li>The <code>@addNLConstraints</code> macro supports adding multiple (groups of) constraints at once. Syntax is similar to <code>@addConstraints</code>.</li><li>Discrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).</li></ul><h3 id="General"><a class="docs-heading-anchor" href="#General">General</a><a id="General-1"></a><a class="docs-heading-anchor-permalink" href="#General" title="Permalink"></a></h3><ul><li>Starting values for variables may now be specified with <code>@defVar(m, x, start=value)</code>.</li><li>The <code>setSolver</code> function allows users to change the solver subsequent to model creation.</li><li>Support for &quot;fixed&quot; variables via the <code>@defVar(m, x == 1)</code> syntax.</li><li>Unit tests rewritten to use FactCheck.jl, improved testing across solvers.</li></ul><h2 id="Version-0.6.3-(October-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.3-(October-19,-2014)">Version 0.6.3 (October 19, 2014)</a><a id="Version-0.6.3-(October-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.3-(October-19,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in multiplying two AffExpr objects.</li></ul><h2 id="Version-0.6.2-(October-11,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.2-(October-11,-2014)">Version 0.6.2 (October 11, 2014)</a><a id="Version-0.6.2-(October-11,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.2-(October-11,-2014)" title="Permalink"></a></h2><ul><li>Further improvements and bug fixes for printing.</li><li>Fixed a bug in <code>@defExpr</code>.</li><li>Support for accessing expression graphs through the MathProgBase NLP interface.</li></ul><h2 id="Version-0.6.1-(September-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.1-(September-19,-2014)">Version 0.6.1 (September 19, 2014)</a><a id="Version-0.6.1-(September-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.1-(September-19,-2014)" title="Permalink"></a></h2><ul><li>Improvements and bug fixes for printing.</li></ul><h2 id="Version-0.6.0-(September-9,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.0-(September-9,-2014)">Version 0.6.0 (September 9, 2014)</a><a id="Version-0.6.0-(September-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.0-(September-9,-2014)" title="Permalink"></a></h2><ul><li>Julia 0.3.0 is the minimum required version for this release.</li><li><code>buildInternalModel(m::Model)</code> added to build solver-level model in memory without optimizing.</li><li>Deprecate <code>load_model_only</code> keyword argument to <code>solve</code>.</li><li>Add groups of constraints with <code>@addConstraints</code> macro.</li><li>Unicode operators now supported, including <code>∑</code> for <code>sum</code>, <code>∏</code> for <code>prod</code>, and <code>≤</code>/<code>≥</code></li><li>Quadratic constraints supported in <code>@addConstraint</code> macro.</li><li>Quadratic objectives supported in <code>@setObjective</code> macro.</li><li>MathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems<ul><li><strong>Breaking change</strong>: <code>IpoptOptions</code> no longer supported to specify solver options, use <code>m = Model(solver=IpoptSolver(options...))</code> instead.</li></ul></li><li>New solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK</li><li>New option to control whether the lazy constraint callback is executed at each node in the B&amp;B tree or just when feasible solutions are found</li><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.</li><li>Add support for index dependencies (for example, triangular indexing) in <code>@defVar</code>, <code>@addConstraint</code>, and <code>@defExpr</code> (for example, <code>@defVar(m, x[i=1:10,j=i:10])</code>).<ul><li>This required some changes to the internal structure of JuMP containers, which may break code that explicitly stored <code>JuMPDict</code> objects.</li></ul></li></ul><h2 id="Version-0.5.8-(September-24,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.8-(September-24,-2014)">Version 0.5.8 (September 24, 2014)</a><a id="Version-0.5.8-(September-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.8-(September-24,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug with specifying solvers (affects Julia 0.2 only)</li></ul><h2 id="Version-0.5.7-(September-5,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.7-(September-5,-2014)">Version 0.5.7 (September 5, 2014)</a><a id="Version-0.5.7-(September-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.7-(September-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in printing models</li></ul><h2 id="Version-0.5.6-(September-2,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.6-(September-2,-2014)">Version 0.5.6 (September 2, 2014)</a><a id="Version-0.5.6-(September-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.6-(September-2,-2014)" title="Permalink"></a></h2><ul><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.<ul><li><strong>Breaking change</strong>: Syntax for <code>Variable()</code> constructor has changed (use of this interface remains discouraged)</li></ul></li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="Version-0.5.5-(July-6,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.5-(July-6,-2014)">Version 0.5.5 (July 6, 2014)</a><a id="Version-0.5.5-(July-6,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.5-(July-6,-2014)" title="Permalink"></a></h2><ul><li>Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.</li></ul><h2 id="Version-0.5.4-(June-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.4-(June-19,-2014)">Version 0.5.4 (June 19, 2014)</a><a id="Version-0.5.4-(June-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.4-(June-19,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in MathProgBase which reduces loading times for <code>using JuMP</code></li><li>Fix error when MIPs not solved to optimality</li></ul><h2 id="Version-0.5.3-(May-21,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.3-(May-21,-2014)">Version 0.5.3 (May 21, 2014)</a><a id="Version-0.5.3-(May-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.3-(May-21,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in ReverseDiffSparse</li></ul><h2 id="Version-0.5.2-(May-9,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.2-(May-9,-2014)">Version 0.5.2 (May 9, 2014)</a><a id="Version-0.5.2-(May-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.2-(May-9,-2014)" title="Permalink"></a></h2><ul><li>Fix compatibility with Julia 0.3 prerelease</li></ul><h2 id="Version-0.5.1-(May-5,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.1-(May-5,-2014)">Version 0.5.1 (May 5, 2014)</a><a id="Version-0.5.1-(May-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.1-(May-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in coefficient handling inside lazy constraints and user cuts</li></ul><h2 id="Version-0.5.0-(May-2,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.0-(May-2,-2014)">Version 0.5.0 (May 2, 2014)</a><a id="Version-0.5.0-(May-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.0-(May-2,-2014)" title="Permalink"></a></h2><ul><li>Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.</li><li><code>getValue</code> for <code>AffExpr</code> and <code>QuadExpr</code></li><li><strong>Breaking change</strong>: <code>getSolverModel</code> replaced by <code>getInternalModel</code>, which returns the internal MathProgBase-level model</li><li>Groups of constraints can be specified with <code>@addConstraint</code> (see documentation for details). This is not a breaking change.</li><li><code>dot(::JuMPDict{Variable},::JuMPDict{Variable})</code> now returns the corresponding quadratic expression.</li></ul><h2 id="Version-0.4.1-(March-24,-2014)"><a class="docs-heading-anchor" href="#Version-0.4.1-(March-24,-2014)">Version 0.4.1 (March 24, 2014)</a><a id="Version-0.4.1-(March-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.4.1-(March-24,-2014)" title="Permalink"></a></h2><ul><li>Fix bug where change in objective sense was ignored when re-solving a model.</li><li>Fix issue with handling zero coefficients in AffExpr.</li></ul><h2 id="Version-0.4.0-(March-10,-2014)"><a class="docs-heading-anchor" href="#Version-0.4.0-(March-10,-2014)">Version 0.4.0 (March 10, 2014)</a><a id="Version-0.4.0-(March-10,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.4.0-(March-10,-2014)" title="Permalink"></a></h2><ul><li>Support for SOS1 and SOS2 constraints.</li><li>Solver-independent callback for user heuristics.</li><li><code>dot</code> and <code>sum</code> implemented for <code>JuMPDict</code> objects. Now you can say <code>@addConstraint(m, dot(a,x) &lt;= b)</code>.</li><li>Developers: support for extensions to JuMP. See definition of Model in <code>src/JuMP.jl</code> for more details.</li><li>Option to construct the low-level model before optimizing.</li></ul><h2 id="Version-0.3.2-(February-17,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.2-(February-17,-2014)">Version 0.3.2 (February 17, 2014)</a><a id="Version-0.3.2-(February-17,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.2-(February-17,-2014)" title="Permalink"></a></h2><ul><li>Improved model printing<ul><li>Preliminary support for IJulia output</li></ul></li></ul><h2 id="Version-0.3.1-(January-30,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.1-(January-30,-2014)">Version 0.3.1 (January 30, 2014)</a><a id="Version-0.3.1-(January-30,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.1-(January-30,-2014)" title="Permalink"></a></h2><ul><li>Documentation updates</li><li>Support for MOSEK</li><li>CPLEXLink renamed to CPLEX</li></ul><h2 id="Version-0.3.0-(January-21,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.0-(January-21,-2014)">Version 0.3.0 (January 21, 2014)</a><a id="Version-0.3.0-(January-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.0-(January-21,-2014)" title="Permalink"></a></h2><ul><li>Unbounded/infeasibility rays: <code>getValue()</code> will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. <code>getDual()</code> will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.</li><li>Solver-independent callbacks for user generated cuts.</li><li>Use new interface for solver-independent QCQP.</li><li><code>setlazycallback</code> renamed to <code>setLazyCallback</code> for consistency.</li></ul><h2 id="Version-0.2.0-(December-15,-2013)"><a class="docs-heading-anchor" href="#Version-0.2.0-(December-15,-2013)">Version 0.2.0 (December 15, 2013)</a><a id="Version-0.2.0-(December-15,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.2.0-(December-15,-2013)" title="Permalink"></a></h2><h3 id="Breaking-8"><a class="docs-heading-anchor" href="#Breaking-8">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-8" title="Permalink"></a></h3><ul><li>Objective sense is specified in <code>setObjective</code> instead of in the <code>Model</code> constructor.</li><li><code>lpsolver</code> and <code>mipsolver</code> merged into single <code>solver</code> option.</li></ul><h3 id="Added-31"><a class="docs-heading-anchor" href="#Added-31">Added</a><a class="docs-heading-anchor-permalink" href="#Added-31" title="Permalink"></a></h3><ul><li>Problem modification with efficient LP restarts and MIP warm-starts.</li><li>Relatedly, column-wise modeling now supported.</li><li>Solver-independent callbacks supported. Currently we support only a &quot;lazy constraint&quot; callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.</li></ul><h2 id="Version-0.1.2-(November-16,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.2-(November-16,-2013)">Version 0.1.2 (November 16, 2013)</a><a id="Version-0.1.2-(November-16,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.2-(November-16,-2013)" title="Permalink"></a></h2><ul><li>Bug fixes for printing, improved error messages.</li><li>Allow <code>AffExpr</code> to be used in macros; for example, <code>ex = y + z; @addConstraint(m, x + 2*ex &lt;= 3)</code></li></ul><h2 id="Version-0.1.1-(October-23,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.1-(October-23,-2013)">Version 0.1.1 (October 23, 2013)</a><a id="Version-0.1.1-(October-23,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.1-(October-23,-2013)" title="Permalink"></a></h2><ul><li>Update for solver specification API changes in MathProgBase.</li></ul><h2 id="Version-0.1.0-(October-3,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.0-(October-3,-2013)">Version 0.1.0 (October 3, 2013)</a><a id="Version-0.1.0-(October-3,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.0-(October-3,-2013)" title="Permalink"></a></h2><ul><li>Initial public release.</li></ul></article><nav class="docs-footer"><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></li><li><p>The <code>lowerbound</code>, <code>upperbound</code>, and <code>basename</code> keyword arguments to the <code>@variable</code> macro have been renamed to <code>lower_bound</code>, <code>upper_bound</code>, and <code>base_name</code>, for consistency with JuMP&#39;s new <a href="https://www.juliaopt.org/JuMP.jl/dev/style/">style recommendations</a>.</p></li><li><p>We rely on broadcasting syntax to apply accessors to collections of variables, for example, <code>value.(x)</code> instead of <code>getvalue(x)</code> for collections. (Use <code>value(x)</code> when <code>x</code> is a scalar object.)</p></li></ul><h3 id="Added-30"><a class="docs-heading-anchor" href="#Added-30">Added</a><a class="docs-heading-anchor-permalink" href="#Added-30" title="Permalink"></a></h3><ul><li><p>Splatting (like <code>f(x...)</code>) is recognized in restricted settings in nonlinear expressions.</p></li><li><p>Support for deleting constraints and variables.</p></li><li><p>The documentation has been completely rewritten using docstrings and Documenter.</p></li><li><p>Support for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).</p></li><li><p>Significantly improved support for modeling new types of constraints and for extending JuMP&#39;s macros.</p></li><li><p>Support for providing dual warm starts.</p></li><li><p>Improved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).</p></li><li><p>Explicit control of whether symmetry-enforcing constraints are added to PSD constraints.</p></li><li><p>Support for modeling exponential cones.</p></li><li><p>Significant improvements in internal code quality and testing.</p></li><li><p>Style and naming guidelines.</p></li><li><p>Direct mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/solvers/">documentation</a>.</p></li></ul><h3 id="Regressions"><a class="docs-heading-anchor" href="#Regressions">Regressions</a><a id="Regressions-1"></a><a class="docs-heading-anchor-permalink" href="#Regressions" title="Permalink"></a></h3><p>There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):</p><ul><li><p>Performance regressions in model generation (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1403">issue</a>). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.</p></li><li><p>Fast incremental NLP solves are not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1185">issue</a>).</p></li><li><p>We do not yet have an implementation of solver-specific callbacks.</p></li><li><p>The column generation syntax in <code>@variable</code> has been removed (that is, the <code>objective</code>, <code>coefficients</code>, and <code>inconstraints</code> keyword arguments). Support for column generation will be re-introduced in a future release.</p></li><li><p>The ability to solve the continuous relaxation (that is, via <code>solve(model; relaxation = true)</code>) is not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1611">issue</a>).</p></li></ul><h2 id="Version-0.18.5-(December-1,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.5-(December-1,-2018)">Version 0.18.5 (December 1, 2018)</a><a id="Version-0.18.5-(December-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.5-(December-1,-2018)" title="Permalink"></a></h2><ul><li>Support views in some derivative evaluation functions.</li><li>Improved compatibility with PackageCompiler.</li></ul><h2 id="Version-0.18.4-(October-8,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.4-(October-8,-2018)">Version 0.18.4 (October 8, 2018)</a><a id="Version-0.18.4-(October-8,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.4-(October-8,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in model printing on Julia 0.7 and 1.0.</li></ul><h2 id="Version-0.18.3-(October-1,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.3-(October-1,-2018)">Version 0.18.3 (October 1, 2018)</a><a id="Version-0.18.3-(October-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.3-(October-1,-2018)" title="Permalink"></a></h2><ul><li>Add support for Julia v1.0 (Thanks @ExpandingMan)</li><li>Fix matrix expressions with quadratic functions (#1508)</li></ul><h2 id="Version-0.18.2-(June-10,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.2-(June-10,-2018)">Version 0.18.2 (June 10, 2018)</a><a id="Version-0.18.2-(June-10,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.2-(June-10,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in second-order derivatives when expressions are present (#1319)</li><li>Fix a bug in <code>@constraintref</code> (#1330)</li></ul><h2 id="Version-0.18.1-(April-9,-2018)"><a class="docs-heading-anchor" href="#Version-0.18.1-(April-9,-2018)">Version 0.18.1 (April 9, 2018)</a><a id="Version-0.18.1-(April-9,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.1-(April-9,-2018)" title="Permalink"></a></h2><ul><li>Fix for nested tuple destructuring (#1193)</li><li>Preserve internal model when relaxation=true (#1209)</li><li>Minor bug fixes and updates for example</li></ul><h2 id="Version-0.18.0-(July-27,-2017)"><a class="docs-heading-anchor" href="#Version-0.18.0-(July-27,-2017)">Version 0.18.0 (July 27, 2017)</a><a id="Version-0.18.0-(July-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.18.0-(July-27,-2017)" title="Permalink"></a></h2><ul><li>Drop support for Julia 0.5.</li><li>Update for ForwardDiff 0.5.</li><li>Minor bug fixes.</li></ul><h2 id="Version-0.17.1-(June-9,-2017)"><a class="docs-heading-anchor" href="#Version-0.17.1-(June-9,-2017)">Version 0.17.1 (June 9, 2017)</a><a id="Version-0.17.1-(June-9,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.17.1-(June-9,-2017)" title="Permalink"></a></h2><ul><li>Use of <code>constructconstraint!</code> in <code>@SDconstraint</code>.</li><li>Minor bug fixes.</li></ul><h2 id="Version-0.17.0-(May-27,-2017)"><a class="docs-heading-anchor" href="#Version-0.17.0-(May-27,-2017)">Version 0.17.0 (May 27, 2017)</a><a id="Version-0.17.0-(May-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.17.0-(May-27,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: Mixing quadratic and conic constraints is no longer supported.</li><li><strong>Breaking change</strong>: The <code>getvariable</code> and <code>getconstraint</code> functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name <code>x</code>, one should now write <code>m[:x]</code> instead of <code>getvariable(m, :x)</code>. As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.</li><li>Addition of the <code>getobjectivebound</code> function that mirrors the functionality of the MathProgBase <code>getobjbound</code> function except that it takes into account transformations performed by JuMP.</li><li>Minor bug fixes.</li></ul><p>The following changes are primarily of interest to developers of JuMP extensions:</p><ul><li>The new syntax <code>@constraint(model, expr in Cone)</code> creates the constraint ensuring that <code>expr</code> is inside <code>Cone</code>. The <code>Cone</code> argument is passed to <code>constructconstraint!</code> which enables the call to the dispatched to an extension.</li><li>The <code>@variable</code> macro now calls <code>constructvariable!</code> instead of directly calling the <code>Variable</code> constructor. Extra arguments and keyword arguments passed to <code>@variable</code> are passed to <code>constructvariable!</code> which enables the call to be dispatched to an extension.</li><li>Refactor the internal function <code>conicdata</code> (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.</li></ul><h2 id="Version-0.16.2-(March-28,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.2-(March-28,-2017)">Version 0.16.2 (March 28, 2017)</a><a id="Version-0.16.2-(March-28,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.2-(March-28,-2017)" title="Permalink"></a></h2><ul><li>Minor bug fixes and printing tweaks</li><li>Address deprecation warnings for Julia 0.6</li></ul><h2 id="Version-0.16.1-(March-7,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.1-(March-7,-2017)">Version 0.16.1 (March 7, 2017)</a><a id="Version-0.16.1-(March-7,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.1-(March-7,-2017)" title="Permalink"></a></h2><ul><li>Better support for <code>AbstractArray</code> in JuMP (Thanks @tkoolen)</li><li>Minor bug fixes</li></ul><h2 id="Version-0.16.0-(February-23,-2017)"><a class="docs-heading-anchor" href="#Version-0.16.0-(February-23,-2017)">Version 0.16.0 (February 23, 2017)</a><a id="Version-0.16.0-(February-23,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.16.0-(February-23,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling <code>solve()</code> will result in an error.</li><li><strong>Breaking change</strong>: User-defined functions are no longer global. The first argument to <code>JuMP.register</code> is now a JuMP <code>Model</code> object within whose scope the function will be registered. Calling <code>JuMP.register</code> without a <code>Model</code> now produces an error.</li><li><strong>Breaking change</strong>: Use the new <code>JuMP.fix</code> method to fix a variable to a value or to update the value to which a variable is fixed. Calling <code>setvalue</code> on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)</li><li>Nonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)</li><li>New <code>category</code> keyword to <code>@variable</code>. Used for specifying categories of anonymous variables.</li><li>Compatibility with Julia 0.6-dev.</li><li>Minor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)</li></ul><h2 id="Version-0.15.1-(January-31,-2017)"><a class="docs-heading-anchor" href="#Version-0.15.1-(January-31,-2017)">Version 0.15.1 (January 31, 2017)</a><a id="Version-0.15.1-(January-31,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.15.1-(January-31,-2017)" title="Permalink"></a></h2><ul><li>Bugfix for <code>@LinearConstraints</code> and friends</li></ul><h2 id="Version-0.15.0-(December-22,-2016)"><a class="docs-heading-anchor" href="#Version-0.15.0-(December-22,-2016)">Version 0.15.0 (December 22, 2016)</a><a id="Version-0.15.0-(December-22,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.15.0-(December-22,-2016)" title="Permalink"></a></h2><ul><li>Julia 0.5.0 is the minimum required version for this release.</li><li>Document support for BARON solver</li><li>Enable info callbacks in more states than before, for example, for recording solutions. New <code>when</code> argument to <code>addinfocallback</code> (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/814">#814</a>, thanks @yeesian)</li><li>Improved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:<ul><li>When multiple variables in a model are given the same name</li><li>When non-symbols are used as names, for example, <code>@variable(m, x[1][1:N])</code></li></ul></li><li>Improvements in iterating over JuMP containers (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/836">#836</a>, thanks @IssamT)</li><li>Support for writing variable names in .lp file output (Thanks @leethargo)</li><li>Support for querying duals to SDP problems (Thanks @blegat)</li><li>The comprehension syntax with curly braces <code>sum{}</code>, <code>prod{}</code>, and <code>norm2{}</code> has been deprecated in favor of Julia&#39;s native comprehension syntax <code>sum()</code>, <code>prod()</code> and <code>norm()</code> as previously announced. (For early adopters of the new syntax, <code>norm2()</code> was renamed to <code>norm()</code> without deprecation.)</li><li>Unit tests rewritten to use Base.Test instead of FactCheck</li><li>Improved support for operations with matrices of JuMP types (Thanks @ExpandingMan)</li><li>The syntax to halt a solver from inside a callback has changed from <code>throw(CallbackAbort())</code> to <code>return JuMP.StopTheSolver</code></li><li>Minor bug fixes</li></ul><h2 id="Version-0.14.2-(December-12,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.2-(December-12,-2016)">Version 0.14.2 (December 12, 2016)</a><a id="Version-0.14.2-(December-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.2-(December-12,-2016)" title="Permalink"></a></h2><ul><li>Allow singleton anonymous variables (includes bugfix)</li></ul><h2 id="Version-0.14.1-(September-12,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.1-(September-12,-2016)">Version 0.14.1 (September 12, 2016)</a><a id="Version-0.14.1-(September-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.1-(September-12,-2016)" title="Permalink"></a></h2><ul><li>More consistent handling of states in informational callbacks, includes a new <code>when</code> parameter to <code>addinfocallback</code> for specifying in which state an informational callback should be called.</li></ul><h2 id="Version-0.14.0-(August-7,-2016)"><a class="docs-heading-anchor" href="#Version-0.14.0-(August-7,-2016)">Version 0.14.0 (August 7, 2016)</a><a id="Version-0.14.0-(August-7,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.14.0-(August-7,-2016)" title="Permalink"></a></h2><ul><li>Compatibility with Julia 0.5 and ForwardDiff 0.2</li><li>Support for &quot;anonymous&quot; variables, constraints, expressions, and parameters, for example, <code>x = @variable(m, [1:N])</code> instead of <code>@variable(m, x[1:N])</code></li><li>Support for retrieving constraints from a model by name via <code>getconstraint</code></li><li><code>@NLconstraint</code> now returns constraint references (as expected).</li><li>Support for vectorized expressions within lazy constraints</li><li>On Julia 0.5, parse new comprehension syntax <code>sum(x[i] for i in 1:N if isodd(i))</code> instead of <code>sum{ x[i], i in 1:N; isodd(i) }</code>. The old syntax with curly braces will be deprecated in JuMP 0.15.</li><li>Now possible to provide nonlinear expressions as &quot;raw&quot; Julia <code>Expr</code> objects instead of using JuMP&#39;s nonlinear macros. This input format is useful for programmatically generated expressions.</li><li><code>s/Mathematical Programming/Mathematical Optimization/</code></li><li>Support for local cuts (Thanks to @madanim, Mehdi Madani)</li><li>Document Xpress interface developed by @joaquimg, Joaquim Dias Garcia</li><li>Minor bug and deprecation fixes (Thanks @odow, @jrevels)</li></ul><h2 id="Version-0.13.2-(May-16,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.2-(May-16,-2016)">Version 0.13.2 (May 16, 2016)</a><a id="Version-0.13.2-(May-16,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.2-(May-16,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for MathProgBase</li></ul><h2 id="Version-0.13.1-(May-3,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.1-(May-3,-2016)">Version 0.13.1 (May 3, 2016)</a><a id="Version-0.13.1-(May-3,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.1-(May-3,-2016)" title="Permalink"></a></h2><ul><li>Fix broken deprecation for <code>registerNLfunction</code>.</li></ul><h2 id="Version-0.13.0-(April-29,-2016)"><a class="docs-heading-anchor" href="#Version-0.13.0-(April-29,-2016)">Version 0.13.0 (April 29, 2016)</a><a id="Version-0.13.0-(April-29,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.13.0-(April-29,-2016)" title="Permalink"></a></h2><ul><li>Most exported methods and macros have been renamed to avoid camelCase. See the list of changes <a href="https://github.com/JuliaOpt/JuMP.jl/blob/e53d0db67cde2a4b80d0c1281f4b49eb0128a1f5/src/deprecated.jl#L30">here</a>. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.</li><li>Specify variable lower/upper bounds in <code>@variable</code> using the <code>lowerbound</code> and <code>upperbound</code> keyword arguments.</li><li>Change name printed for variable using the <code>basename</code> keyword argument to <code>@variable</code>.</li><li>New <code>@variables</code> macro allows multi-line declaration of groups of variables.</li><li>A number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was <a href="https://youtu.be/qF1lZPJ3a5A">recorded</a> live.</li><li>Compatibility fixes with Julia 0.5.</li><li>The &quot;end&quot; indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see <a href="https://github.com/JuliaOpt/JuMP.jl/issues/730">here</a>.</li></ul><h2 id="Version-0.12.2-(March-9,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.2-(March-9,-2016)">Version 0.12.2 (March 9, 2016)</a><a id="Version-0.12.2-(March-9,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.2-(March-9,-2016)" title="Permalink"></a></h2><ul><li>Small fixes for nonlinear optimization</li></ul><h2 id="Version-0.12.1-(March-1,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.1-(March-1,-2016)">Version 0.12.1 (March 1, 2016)</a><a id="Version-0.12.1-(March-1,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.1-(March-1,-2016)" title="Permalink"></a></h2><ul><li>Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)</li></ul><h2 id="Version-0.12.0-(February-27,-2016)"><a class="docs-heading-anchor" href="#Version-0.12.0-(February-27,-2016)">Version 0.12.0 (February 27, 2016)</a><a id="Version-0.12.0-(February-27,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.12.0-(February-27,-2016)" title="Permalink"></a></h2><ul><li>The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:<ul><li><code>@defExpr</code> and <code>@defNLExpr</code> now take the model as the first argument. The previous one-argument version of <code>@defExpr</code> is deprecated; all expressions should be named. For example, replace <code>@defExpr(2x+y)</code> with <code>@defExpr(jump_model, my_expr, 2x+y)</code>.</li><li>JuMP no longer uses Julia&#39;s variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.</li><li>Support for user-defined functions integrated within nonlinear JuMP expressions.</li></ul></li><li>Replaced iteration over <code>AffExpr</code> with <code>Number</code>-like scalar iteration; previous iteration behavior is now available via <code>linearterms(::AffExpr)</code>.</li><li>Stopping the solver via <code>throw(CallbackAbort())</code> from a callback no longer triggers an exception. Instead, <code>solve()</code> returns <code>UserLimit</code> status.</li><li><code>getDual()</code> now works for conic problems (Thanks @emreyamangil.)</li></ul><h2 id="Version-0.11.3-(February-4,-2016)"><a class="docs-heading-anchor" href="#Version-0.11.3-(February-4,-2016)">Version 0.11.3 (February 4, 2016)</a><a id="Version-0.11.3-(February-4,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.3-(February-4,-2016)" title="Permalink"></a></h2><ul><li>Bug-fix for problems with quadratic objectives and semidefinite constraints</li></ul><h2 id="Version-0.11.2-(January-14,-2016)"><a class="docs-heading-anchor" href="#Version-0.11.2-(January-14,-2016)">Version 0.11.2 (January 14, 2016)</a><a id="Version-0.11.2-(January-14,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.2-(January-14,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for Mosek</li></ul><h2 id="Version-0.11.1-(December-1,-2015)"><a class="docs-heading-anchor" href="#Version-0.11.1-(December-1,-2015)">Version 0.11.1 (December 1, 2015)</a><a id="Version-0.11.1-(December-1,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.1-(December-1,-2015)" title="Permalink"></a></h2><ul><li>Remove usage of <code>@compat</code> in tests.</li><li>Fix updating quadratic objectives for nonlinear models.</li></ul><h2 id="Version-0.11.0-(November-30,-2015)"><a class="docs-heading-anchor" href="#Version-0.11.0-(November-30,-2015)">Version 0.11.0 (November 30, 2015)</a><a id="Version-0.11.0-(November-30,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.11.0-(November-30,-2015)" title="Permalink"></a></h2><ul><li>Julia 0.4.0 is the minimum required version for this release.</li><li>Fix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.</li><li>Addition of the <code>solve(m::Model, relaxation=true)</code> keyword argument to solve the standard continuous relaxation of model <code>m</code></li><li>The <code>getConstraintBounds()</code> method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.</li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="Version-0.10.3-(November-20,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.3-(November-20,-2015)">Version 0.10.3 (November 20, 2015)</a><a id="Version-0.10.3-(November-20,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.3-(November-20,-2015)" title="Permalink"></a></h2><ul><li>Fix a rare error when parsing quadratic expressions</li><li>Fix <code>Variable()</code> constructor with default arguments</li><li>Detect unrecognized keywords in <code>solve()</code></li></ul><h2 id="Version-0.10.2-(September-28,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.2-(September-28,-2015)">Version 0.10.2 (September 28, 2015)</a><a id="Version-0.10.2-(September-28,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.2-(September-28,-2015)" title="Permalink"></a></h2><ul><li>Fix for deprecation warnings</li></ul><h2 id="Version-0.10.1-(September-3,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.1-(September-3,-2015)">Version 0.10.1 (September 3, 2015)</a><a id="Version-0.10.1-(September-3,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.1-(September-3,-2015)" title="Permalink"></a></h2><ul><li>Fixes for ambiguity warnings.</li><li>Fix for breaking change in precompilation syntax in Julia 0.4-pre</li></ul><h2 id="Version-0.10.0-(August-31,-2015)"><a class="docs-heading-anchor" href="#Version-0.10.0-(August-31,-2015)">Version 0.10.0 (August 31, 2015)</a><a id="Version-0.10.0-(August-31,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.10.0-(August-31,-2015)" title="Permalink"></a></h2><ul><li>Support (on Julia 0.4 and later) for conditions in indexing <code>@defVar</code> and <code>@addConstraint</code> constructs, for example, <code>@defVar(m, x[i=1:5,j=1:5; i+j &gt;= 3])</code></li><li>Support for vectorized operations on Variables and expressions. See the documentation for details.</li><li>New <code>getVar()</code> method to access variables in a model by name</li><li>Support for semidefinite programming.</li><li>Dual solutions are now available for general nonlinear problems. You may call <code>getDual</code> on a reference object for a nonlinear constraint, and <code>getDual</code> on a variable object for Lagrange multipliers from active bounds.</li><li>Introduce warnings for two common performance traps: too many calls to <code>getValue()</code> on a collection of variables and use of the <code>+</code> operator in a loop to sum expressions.</li><li>Second-order cone constraints can be written directly with the <code>norm()</code> and <code>norm2{}</code> syntax.</li><li>Implement MathProgBase interface for querying Hessian-vector products.</li><li>Iteration over <code>JuMPContainer</code>s is deprecated; instead, use the <code>keys</code> and <code>values</code> functions, and <code>zip(keys(d),values(d))</code> for the old behavior.</li><li><code>@defVar</code> returns <code>Array{Variable,N}</code> when each of <code>N</code> index sets are of the form <code>1:nᵢ</code>.</li><li>Module precompilation: on Julia 0.4 and later, <code>using JuMP</code> is now much faster.</li></ul><h2 id="Version-0.9.3-(August-11,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.3-(August-11,-2015)">Version 0.9.3 (August 11, 2015)</a><a id="Version-0.9.3-(August-11,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.3-(August-11,-2015)" title="Permalink"></a></h2><ul><li>Fixes for FactCheck testing on julia v0.4.</li></ul><h2 id="Version-0.9.2-(June-27,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.2-(June-27,-2015)">Version 0.9.2 (June 27, 2015)</a><a id="Version-0.9.2-(June-27,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.2-(June-27,-2015)" title="Permalink"></a></h2><ul><li>Fix bug in @addConstraints.</li></ul><h2 id="Version-0.9.1-(April-25,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.1-(April-25,-2015)">Version 0.9.1 (April 25, 2015)</a><a id="Version-0.9.1-(April-25,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.1-(April-25,-2015)" title="Permalink"></a></h2><ul><li>Fix for Julia 0.4-dev.</li><li>Small infrastructure improvements for extensions.</li></ul><h2 id="Version-0.9.0-(April-18,-2015)"><a class="docs-heading-anchor" href="#Version-0.9.0-(April-18,-2015)">Version 0.9.0 (April 18, 2015)</a><a id="Version-0.9.0-(April-18,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.9.0-(April-18,-2015)" title="Permalink"></a></h2><ul><li>Comparison operators for constructing constraints (for example, <code>2x &gt;= 1</code>) have been deprecated. Instead, construct the constraints explicitly in the <code>@addConstraint</code> macro to add them to the model, or in the <code>@LinearConstraint</code> macro to create a stand-alone linear constraint instance.</li><li><code>getValue()</code> method implemented to compute the value of a nonlinear subexpression</li><li>JuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.</li><li>A number of performance improvements in ReverseDiffSparse for computing derivatives.</li><li><code>MathProgBase.getsolvetime(m)</code> now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)</li><li>Formatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).</li></ul><h2 id="Version-0.8.0-(February-17,-2015)"><a class="docs-heading-anchor" href="#Version-0.8.0-(February-17,-2015)">Version 0.8.0 (February 17, 2015)</a><a id="Version-0.8.0-(February-17,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.8.0-(February-17,-2015)" title="Permalink"></a></h2><ul><li>Nonlinear subexpressions now supported with the <code>@defNLExpr</code> macro.</li><li>SCS supported for solving second-order conic problems.</li><li><code>setXXXCallback</code> family deprecated in favor of <code>addXXXCallback</code>.</li><li>Multiple callbacks of the same type can be registered.</li><li>Added support for informational callbacks via <code>addInfoCallback</code>.</li><li>A <code>CallbackAbort</code> exception can be thrown from callback to safely exit optimization.</li></ul><h2 id="Version-0.7.4-(February-4,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.4-(February-4,-2015)">Version 0.7.4 (February 4, 2015)</a><a id="Version-0.7.4-(February-4,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.4-(February-4,-2015)" title="Permalink"></a></h2><ul><li>Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.</li><li>Two-sided nonlinear constraints are supported.</li><li>Methods for accessing the number of variables and constraints in a model are renamed.</li><li>New default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.</li><li>Small bug fixes.</li></ul><h2 id="Version-0.7.3-(January-14,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.3-(January-14,-2015)">Version 0.7.3 (January 14, 2015)</a><a id="Version-0.7.3-(January-14,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.3-(January-14,-2015)" title="Permalink"></a></h2><ul><li>Fix a method ambiguity conflict with Compose.jl (cosmetic fix)</li></ul><h2 id="Version-0.7.2-(January-9,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.2-(January-9,-2015)">Version 0.7.2 (January 9, 2015)</a><a id="Version-0.7.2-(January-9,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.2-(January-9,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in <code>sum(::JuMPDict)</code></li><li>Added the <code>setCategory</code> function to change a variables category (for example, continuous or binary)</li></ul><p>after construction, and <code>getCategory</code> to retrieve the variable category.</p><h2 id="Version-0.7.1-(January-2,-2015)"><a class="docs-heading-anchor" href="#Version-0.7.1-(January-2,-2015)">Version 0.7.1 (January 2, 2015)</a><a id="Version-0.7.1-(January-2,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.1-(January-2,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.</li></ul><h2 id="Version-0.7.0-(December-29,-2014)"><a class="docs-heading-anchor" href="#Version-0.7.0-(December-29,-2014)">Version 0.7.0 (December 29, 2014)</a><a id="Version-0.7.0-(December-29,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.7.0-(December-29,-2014)" title="Permalink"></a></h2><h3 id="Linear/quadratic/conic-programming"><a class="docs-heading-anchor" href="#Linear/quadratic/conic-programming">Linear/quadratic/conic programming</a><a id="Linear/quadratic/conic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Linear/quadratic/conic-programming" title="Permalink"></a></h3><ul><li><strong>Breaking change</strong>: The syntax for column-wise model generation has been changed to use keyword arguments in <code>@defVar</code>.</li><li>On Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.</li><li>ECOS supported for solving second-order conic problems.</li></ul><h3 id="_nonlinear_programming_release_notes"><a class="docs-heading-anchor" href="#_nonlinear_programming_release_notes">Nonlinear programming</a><a id="_nonlinear_programming_release_notes-1"></a><a class="docs-heading-anchor-permalink" href="#_nonlinear_programming_release_notes" title="Permalink"></a></h3><ul><li>Support for skipping model generation when solving a sequence of nonlinear models with changing data.</li><li>Fix a memory leak when solving a sequence of nonlinear models.</li><li>The <code>@addNLConstraint</code> macro now supports the three-argument version to define sets of nonlinear constraints.</li><li>KNITRO supported as a nonlinear solver.</li><li>Speed improvements for model generation.</li><li>The <code>@addNLConstraints</code> macro supports adding multiple (groups of) constraints at once. Syntax is similar to <code>@addConstraints</code>.</li><li>Discrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).</li></ul><h3 id="General"><a class="docs-heading-anchor" href="#General">General</a><a id="General-1"></a><a class="docs-heading-anchor-permalink" href="#General" title="Permalink"></a></h3><ul><li>Starting values for variables may now be specified with <code>@defVar(m, x, start=value)</code>.</li><li>The <code>setSolver</code> function allows users to change the solver subsequent to model creation.</li><li>Support for &quot;fixed&quot; variables via the <code>@defVar(m, x == 1)</code> syntax.</li><li>Unit tests rewritten to use FactCheck.jl, improved testing across solvers.</li></ul><h2 id="Version-0.6.3-(October-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.3-(October-19,-2014)">Version 0.6.3 (October 19, 2014)</a><a id="Version-0.6.3-(October-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.3-(October-19,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in multiplying two AffExpr objects.</li></ul><h2 id="Version-0.6.2-(October-11,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.2-(October-11,-2014)">Version 0.6.2 (October 11, 2014)</a><a id="Version-0.6.2-(October-11,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.2-(October-11,-2014)" title="Permalink"></a></h2><ul><li>Further improvements and bug fixes for printing.</li><li>Fixed a bug in <code>@defExpr</code>.</li><li>Support for accessing expression graphs through the MathProgBase NLP interface.</li></ul><h2 id="Version-0.6.1-(September-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.1-(September-19,-2014)">Version 0.6.1 (September 19, 2014)</a><a id="Version-0.6.1-(September-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.1-(September-19,-2014)" title="Permalink"></a></h2><ul><li>Improvements and bug fixes for printing.</li></ul><h2 id="Version-0.6.0-(September-9,-2014)"><a class="docs-heading-anchor" href="#Version-0.6.0-(September-9,-2014)">Version 0.6.0 (September 9, 2014)</a><a id="Version-0.6.0-(September-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.6.0-(September-9,-2014)" title="Permalink"></a></h2><ul><li>Julia 0.3.0 is the minimum required version for this release.</li><li><code>buildInternalModel(m::Model)</code> added to build solver-level model in memory without optimizing.</li><li>Deprecate <code>load_model_only</code> keyword argument to <code>solve</code>.</li><li>Add groups of constraints with <code>@addConstraints</code> macro.</li><li>Unicode operators now supported, including <code>∑</code> for <code>sum</code>, <code>∏</code> for <code>prod</code>, and <code>≤</code>/<code>≥</code></li><li>Quadratic constraints supported in <code>@addConstraint</code> macro.</li><li>Quadratic objectives supported in <code>@setObjective</code> macro.</li><li>MathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems<ul><li><strong>Breaking change</strong>: <code>IpoptOptions</code> no longer supported to specify solver options, use <code>m = Model(solver=IpoptSolver(options...))</code> instead.</li></ul></li><li>New solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK</li><li>New option to control whether the lazy constraint callback is executed at each node in the B&amp;B tree or just when feasible solutions are found</li><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.</li><li>Add support for index dependencies (for example, triangular indexing) in <code>@defVar</code>, <code>@addConstraint</code>, and <code>@defExpr</code> (for example, <code>@defVar(m, x[i=1:10,j=i:10])</code>).<ul><li>This required some changes to the internal structure of JuMP containers, which may break code that explicitly stored <code>JuMPDict</code> objects.</li></ul></li></ul><h2 id="Version-0.5.8-(September-24,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.8-(September-24,-2014)">Version 0.5.8 (September 24, 2014)</a><a id="Version-0.5.8-(September-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.8-(September-24,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug with specifying solvers (affects Julia 0.2 only)</li></ul><h2 id="Version-0.5.7-(September-5,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.7-(September-5,-2014)">Version 0.5.7 (September 5, 2014)</a><a id="Version-0.5.7-(September-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.7-(September-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in printing models</li></ul><h2 id="Version-0.5.6-(September-2,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.6-(September-2,-2014)">Version 0.5.6 (September 2, 2014)</a><a id="Version-0.5.6-(September-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.6-(September-2,-2014)" title="Permalink"></a></h2><ul><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.<ul><li><strong>Breaking change</strong>: Syntax for <code>Variable()</code> constructor has changed (use of this interface remains discouraged)</li></ul></li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="Version-0.5.5-(July-6,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.5-(July-6,-2014)">Version 0.5.5 (July 6, 2014)</a><a id="Version-0.5.5-(July-6,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.5-(July-6,-2014)" title="Permalink"></a></h2><ul><li>Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.</li></ul><h2 id="Version-0.5.4-(June-19,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.4-(June-19,-2014)">Version 0.5.4 (June 19, 2014)</a><a id="Version-0.5.4-(June-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.4-(June-19,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in MathProgBase which reduces loading times for <code>using JuMP</code></li><li>Fix error when MIPs not solved to optimality</li></ul><h2 id="Version-0.5.3-(May-21,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.3-(May-21,-2014)">Version 0.5.3 (May 21, 2014)</a><a id="Version-0.5.3-(May-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.3-(May-21,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in ReverseDiffSparse</li></ul><h2 id="Version-0.5.2-(May-9,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.2-(May-9,-2014)">Version 0.5.2 (May 9, 2014)</a><a id="Version-0.5.2-(May-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.2-(May-9,-2014)" title="Permalink"></a></h2><ul><li>Fix compatibility with Julia 0.3 prerelease</li></ul><h2 id="Version-0.5.1-(May-5,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.1-(May-5,-2014)">Version 0.5.1 (May 5, 2014)</a><a id="Version-0.5.1-(May-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.1-(May-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in coefficient handling inside lazy constraints and user cuts</li></ul><h2 id="Version-0.5.0-(May-2,-2014)"><a class="docs-heading-anchor" href="#Version-0.5.0-(May-2,-2014)">Version 0.5.0 (May 2, 2014)</a><a id="Version-0.5.0-(May-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.5.0-(May-2,-2014)" title="Permalink"></a></h2><ul><li>Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.</li><li><code>getValue</code> for <code>AffExpr</code> and <code>QuadExpr</code></li><li><strong>Breaking change</strong>: <code>getSolverModel</code> replaced by <code>getInternalModel</code>, which returns the internal MathProgBase-level model</li><li>Groups of constraints can be specified with <code>@addConstraint</code> (see documentation for details). This is not a breaking change.</li><li><code>dot(::JuMPDict{Variable},::JuMPDict{Variable})</code> now returns the corresponding quadratic expression.</li></ul><h2 id="Version-0.4.1-(March-24,-2014)"><a class="docs-heading-anchor" href="#Version-0.4.1-(March-24,-2014)">Version 0.4.1 (March 24, 2014)</a><a id="Version-0.4.1-(March-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.4.1-(March-24,-2014)" title="Permalink"></a></h2><ul><li>Fix bug where change in objective sense was ignored when re-solving a model.</li><li>Fix issue with handling zero coefficients in AffExpr.</li></ul><h2 id="Version-0.4.0-(March-10,-2014)"><a class="docs-heading-anchor" href="#Version-0.4.0-(March-10,-2014)">Version 0.4.0 (March 10, 2014)</a><a id="Version-0.4.0-(March-10,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.4.0-(March-10,-2014)" title="Permalink"></a></h2><ul><li>Support for SOS1 and SOS2 constraints.</li><li>Solver-independent callback for user heuristics.</li><li><code>dot</code> and <code>sum</code> implemented for <code>JuMPDict</code> objects. Now you can say <code>@addConstraint(m, dot(a,x) &lt;= b)</code>.</li><li>Developers: support for extensions to JuMP. See definition of Model in <code>src/JuMP.jl</code> for more details.</li><li>Option to construct the low-level model before optimizing.</li></ul><h2 id="Version-0.3.2-(February-17,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.2-(February-17,-2014)">Version 0.3.2 (February 17, 2014)</a><a id="Version-0.3.2-(February-17,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.2-(February-17,-2014)" title="Permalink"></a></h2><ul><li>Improved model printing<ul><li>Preliminary support for IJulia output</li></ul></li></ul><h2 id="Version-0.3.1-(January-30,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.1-(January-30,-2014)">Version 0.3.1 (January 30, 2014)</a><a id="Version-0.3.1-(January-30,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.1-(January-30,-2014)" title="Permalink"></a></h2><ul><li>Documentation updates</li><li>Support for MOSEK</li><li>CPLEXLink renamed to CPLEX</li></ul><h2 id="Version-0.3.0-(January-21,-2014)"><a class="docs-heading-anchor" href="#Version-0.3.0-(January-21,-2014)">Version 0.3.0 (January 21, 2014)</a><a id="Version-0.3.0-(January-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.3.0-(January-21,-2014)" title="Permalink"></a></h2><ul><li>Unbounded/infeasibility rays: <code>getValue()</code> will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. <code>getDual()</code> will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.</li><li>Solver-independent callbacks for user generated cuts.</li><li>Use new interface for solver-independent QCQP.</li><li><code>setlazycallback</code> renamed to <code>setLazyCallback</code> for consistency.</li></ul><h2 id="Version-0.2.0-(December-15,-2013)"><a class="docs-heading-anchor" href="#Version-0.2.0-(December-15,-2013)">Version 0.2.0 (December 15, 2013)</a><a id="Version-0.2.0-(December-15,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.2.0-(December-15,-2013)" title="Permalink"></a></h2><h3 id="Breaking-8"><a class="docs-heading-anchor" href="#Breaking-8">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-8" title="Permalink"></a></h3><ul><li>Objective sense is specified in <code>setObjective</code> instead of in the <code>Model</code> constructor.</li><li><code>lpsolver</code> and <code>mipsolver</code> merged into single <code>solver</code> option.</li></ul><h3 id="Added-31"><a class="docs-heading-anchor" href="#Added-31">Added</a><a class="docs-heading-anchor-permalink" href="#Added-31" title="Permalink"></a></h3><ul><li>Problem modification with efficient LP restarts and MIP warm-starts.</li><li>Relatedly, column-wise modeling now supported.</li><li>Solver-independent callbacks supported. Currently we support only a &quot;lazy constraint&quot; callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.</li></ul><h2 id="Version-0.1.2-(November-16,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.2-(November-16,-2013)">Version 0.1.2 (November 16, 2013)</a><a id="Version-0.1.2-(November-16,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.2-(November-16,-2013)" title="Permalink"></a></h2><ul><li>Bug fixes for printing, improved error messages.</li><li>Allow <code>AffExpr</code> to be used in macros; for example, <code>ex = y + z; @addConstraint(m, x + 2*ex &lt;= 3)</code></li></ul><h2 id="Version-0.1.1-(October-23,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.1-(October-23,-2013)">Version 0.1.1 (October 23, 2013)</a><a id="Version-0.1.1-(October-23,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.1-(October-23,-2013)" title="Permalink"></a></h2><ul><li>Update for solver specification API changes in MathProgBase.</li></ul><h2 id="Version-0.1.0-(October-3,-2013)"><a class="docs-heading-anchor" href="#Version-0.1.0-(October-3,-2013)">Version 0.1.0 (October 3, 2013)</a><a id="Version-0.1.0-(October-3,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#Version-0.1.0-(October-3,-2013)" title="Permalink"></a></h2><ul><li>Initial public release.</li></ul></article><nav class="docs-footer"><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/checklists/index.html b/previews/PR3562/developers/checklists/index.html
index fe85691f406..fa413334966 100644
--- a/previews/PR3562/developers/checklists/index.html
+++ b/previews/PR3562/developers/checklists/index.html
@@ -62,4 +62,4 @@
 
 ## Optional
 
- - [ ] Add package metadata to `docs/packages.toml`</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../roadmap/">« Roadmap</a><a class="docs-footer-nextpage" href="../../packages/solvers/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ - [ ] Add package metadata to `docs/packages.toml`</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../roadmap/">« Roadmap</a><a class="docs-footer-nextpage" href="../../packages/solvers/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/contributing/index.html b/previews/PR3562/developers/contributing/index.html
index cabce589c5d..c5e9f1edd33 100644
--- a/previews/PR3562/developers/contributing/index.html
+++ b/previews/PR3562/developers/contributing/index.html
@@ -25,4 +25,4 @@
 
 $ git checkout master
 
-$ git pull</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>If you have suggestions to improve this guide, please make a pull request. It&#39;s particularly helpful if you do this after your first pull request because you&#39;ll know all the parts that could be explained better.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/algebraic_modeling_languages/">« Algebraic modeling languages</a><a class="docs-footer-nextpage" href="../extensions/">Extensions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+$ git pull</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>If you have suggestions to improve this guide, please make a pull request. It&#39;s particularly helpful if you do this after your first pull request because you&#39;ll know all the parts that could be explained better.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/algebraic_modeling_languages/">« Algebraic modeling languages</a><a class="docs-footer-nextpage" href="../extensions/">Extensions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/custom_solver_binaries/index.html b/previews/PR3562/developers/custom_solver_binaries/index.html
index b5534c2fb37..a93209dd9ed 100644
--- a/previews/PR3562/developers/custom_solver_binaries/index.html
+++ b/previews/PR3562/developers/custom_solver_binaries/index.html
@@ -90,4 +90,4 @@
 libCbc_path = &quot;/usr/local/Cellar/cbc/2.10.5/lib/libCbc.3.10.5&quot;
 libOsiCbc_path = &quot;/usr/local/Cellar/cbc/2.10.5/lib/libOsiCbc.3.10.5&quot;
 libcbcsolver_path = &quot;/usr/local/Cellar/cbc/2.10.5/lib/libCbcSolver.3.10.5&quot;</code></pre><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>Note that capitalization matters, so <code>libcbcsolver_path</code> corresponds to <code>libCbcSolver.3.10.5</code>.</p></div></div><h3 id="Override-entire-artifact"><a class="docs-heading-anchor" href="#Override-entire-artifact">Override entire artifact</a><a id="Override-entire-artifact-1"></a><a class="docs-heading-anchor-permalink" href="#Override-entire-artifact" title="Permalink"></a></h3><p>To use the homebrew install as our custom binary we add the following to <code>~/.julia/artifacts/Overrides.toml</code>:</p><pre><code class="language-julia hljs"># Override for Cbc_jll
-e481bc81db5e229ba1f52b2b4bd57484204b1b06 = &quot;/usr/local/Cellar/cbc/2.10.5&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../extensions/">« Extensions</a><a class="docs-footer-nextpage" href="../style/">Style Guide »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+e481bc81db5e229ba1f52b2b4bd57484204b1b06 = &quot;/usr/local/Cellar/cbc/2.10.5&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../extensions/">« Extensions</a><a class="docs-footer-nextpage" href="../style/">Style Guide »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/extensions/index.html b/previews/PR3562/developers/extensions/index.html
index 207f7ccf4a0..42bf84c580f 100644
--- a/previews/PR3562/developers/extensions/index.html
+++ b/previews/PR3562/developers/extensions/index.html
@@ -301,4 +301,4 @@
         _function_barrier(names, model, F, S)
     end
     return names
-end</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>It is important to explicitly type the <code>F</code> and <code>S</code> arguments. If you leave them untyped, for example, <code>function _function_barrier(names, model, F, S)</code>, Julia will not specialize the function calls and performance will not be improved.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../contributing/">« Contributing</a><a class="docs-footer-nextpage" href="../custom_solver_binaries/">Custom binaries »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>It is important to explicitly type the <code>F</code> and <code>S</code> arguments. If you leave them untyped, for example, <code>function _function_barrier(names, model, F, S)</code>, Julia will not specialize the function calls and performance will not be improved.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../contributing/">« Contributing</a><a class="docs-footer-nextpage" href="../custom_solver_binaries/">Custom binaries »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/roadmap/index.html b/previews/PR3562/developers/roadmap/index.html
index b73ec84e7af..ea3eeb29767 100644
--- a/previews/PR3562/developers/roadmap/index.html
+++ b/previews/PR3562/developers/roadmap/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox" checked/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../contributing/">Contributing</a></li><li><a class="tocitem" href="../extensions/">Extensions</a></li><li><a class="tocitem" href="../custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../style/">Style Guide</a></li><li class="is-active"><a class="tocitem" href>Roadmap</a></li><li><a class="tocitem" href="../checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Developer Docs</a></li><li class="is-active"><a href>Roadmap</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Roadmap</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/developers/roadmap.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Development-roadmap"><a class="docs-heading-anchor" href="#Development-roadmap">Development roadmap</a><a id="Development-roadmap-1"></a><a class="docs-heading-anchor-permalink" href="#Development-roadmap" title="Permalink"></a></h1><p>The JuMP developers have compiled this roadmap document to share their plans and goals with the JuMP community. Contributions to roadmap issues are especially invited.</p><p>Most of these issues will require changes to both JuMP and MathOptInterface, and are non-trivial in their implementation. They are in no particular order, but represent broad themes that we see as areas in which JuMP could be improved.</p><ul><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3106">#3106</a></strong> Make nonlinear programming a first-class citizen. There have been many issues and discussions about this: currently nonlinear constraints are handled through a <code>MOI.NLPBlock</code> and have various limitations and restrictions.<ul><li><a href="https://github.com/jump-dev/JuMP.jl/issues/1185">https://github.com/jump-dev/JuMP.jl/issues/1185</a></li><li><a href="https://github.com/jump-dev/JuMP.jl/issues/1198">https://github.com/jump-dev/JuMP.jl/issues/1198</a></li><li><a href="https://github.com/jump-dev/JuMP.jl/issues/2788">https://github.com/jump-dev/JuMP.jl/issues/2788</a></li><li><a href="https://github.com/jump-dev/MathOptInterface.jl/issues/846">https://github.com/jump-dev/MathOptInterface.jl/issues/846</a></li><li><a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1397">https://github.com/jump-dev/MathOptInterface.jl/issues/1397</a></li></ul></li><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3385">#3385</a></strong> Add support for coefficient types other than <code>Float64</code>: <a href="https://github.com/jump-dev/JuMP.jl/issues/2025">https://github.com/jump-dev/JuMP.jl/issues/2025</a> Since the very beginning, JuMP has hard-coded the coefficient type as <code>Float64</code>. This has made it impossible to support solvers which can use other types such as <code>BigFloat</code> or <code>Rational{BigInt}</code>.</li><li>Add support for constraint programming: <a href="https://github.com/jump-dev/JuMP.jl/issues/2227">https://github.com/jump-dev/JuMP.jl/issues/2227</a> JuMP has a strong focus on linear, conic and nonlinear optimization problems. We want to add better support for constraint programming.</li><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3176">#3176</a></strong> Add support for multiobjective problems: <a href="https://github.com/jump-dev/JuMP.jl/issues/2099">https://github.com/jump-dev/JuMP.jl/issues/2099</a> JuMP is restricted to problems with scalar-valued objectives. We want to extend this to vector-valued problems.</li><li>Refactor the internal code of JuMP&#39;s macros. The code in <code>src/macros.jl</code> is some of the oldest part of JuMP and is difficult to read, modify, and extend. We should overhaul the internals of JuMP&#39;s macros–-without making user-visible breaking changes–-to improve their long-term maintainability.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../style/">« Style Guide</a><a class="docs-footer-nextpage" href="../checklists/">Checklists »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox" checked/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../contributing/">Contributing</a></li><li><a class="tocitem" href="../extensions/">Extensions</a></li><li><a class="tocitem" href="../custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../style/">Style Guide</a></li><li class="is-active"><a class="tocitem" href>Roadmap</a></li><li><a class="tocitem" href="../checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Developer Docs</a></li><li class="is-active"><a href>Roadmap</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Roadmap</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/developers/roadmap.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Development-roadmap"><a class="docs-heading-anchor" href="#Development-roadmap">Development roadmap</a><a id="Development-roadmap-1"></a><a class="docs-heading-anchor-permalink" href="#Development-roadmap" title="Permalink"></a></h1><p>The JuMP developers have compiled this roadmap document to share their plans and goals with the JuMP community. Contributions to roadmap issues are especially invited.</p><p>Most of these issues will require changes to both JuMP and MathOptInterface, and are non-trivial in their implementation. They are in no particular order, but represent broad themes that we see as areas in which JuMP could be improved.</p><ul><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3106">#3106</a></strong> Make nonlinear programming a first-class citizen. There have been many issues and discussions about this: currently nonlinear constraints are handled through a <code>MOI.NLPBlock</code> and have various limitations and restrictions.<ul><li><a href="https://github.com/jump-dev/JuMP.jl/issues/1185">https://github.com/jump-dev/JuMP.jl/issues/1185</a></li><li><a href="https://github.com/jump-dev/JuMP.jl/issues/1198">https://github.com/jump-dev/JuMP.jl/issues/1198</a></li><li><a href="https://github.com/jump-dev/JuMP.jl/issues/2788">https://github.com/jump-dev/JuMP.jl/issues/2788</a></li><li><a href="https://github.com/jump-dev/MathOptInterface.jl/issues/846">https://github.com/jump-dev/MathOptInterface.jl/issues/846</a></li><li><a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1397">https://github.com/jump-dev/MathOptInterface.jl/issues/1397</a></li></ul></li><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3385">#3385</a></strong> Add support for coefficient types other than <code>Float64</code>: <a href="https://github.com/jump-dev/JuMP.jl/issues/2025">https://github.com/jump-dev/JuMP.jl/issues/2025</a> Since the very beginning, JuMP has hard-coded the coefficient type as <code>Float64</code>. This has made it impossible to support solvers which can use other types such as <code>BigFloat</code> or <code>Rational{BigInt}</code>.</li><li>Add support for constraint programming: <a href="https://github.com/jump-dev/JuMP.jl/issues/2227">https://github.com/jump-dev/JuMP.jl/issues/2227</a> JuMP has a strong focus on linear, conic and nonlinear optimization problems. We want to add better support for constraint programming.</li><li><strong>Done <a href="https://github.com/jump-dev/JuMP.jl/pull/3176">#3176</a></strong> Add support for multiobjective problems: <a href="https://github.com/jump-dev/JuMP.jl/issues/2099">https://github.com/jump-dev/JuMP.jl/issues/2099</a> JuMP is restricted to problems with scalar-valued objectives. We want to extend this to vector-valued problems.</li><li>Refactor the internal code of JuMP&#39;s macros. The code in <code>src/macros.jl</code> is some of the oldest part of JuMP and is difficult to read, modify, and extend. We should overhaul the internals of JuMP&#39;s macros–-without making user-visible breaking changes–-to improve their long-term maintainability.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../style/">« Style Guide</a><a class="docs-footer-nextpage" href="../checklists/">Checklists »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/developers/style/index.html b/previews/PR3562/developers/style/index.html
index ec201de9549..ca0cd33f2fb 100644
--- a/previews/PR3562/developers/style/index.html
+++ b/previews/PR3562/developers/style/index.html
@@ -182,4 +182,4 @@
 
 end # module TestPkg
 
-TestPkg.runtests()</code></pre><p>Break the tests into multiple files, with one module per file, so that subsets of the codebase can be tested by calling <code>include</code> with the relevant file.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../custom_solver_binaries/">« Custom binaries</a><a class="docs-footer-nextpage" href="../roadmap/">Roadmap »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+TestPkg.runtests()</code></pre><p>Break the tests into multiple files, with one module per file, so that subsets of the codebase can be tested by calling <code>include</code> with the relevant file.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../custom_solver_binaries/">« Custom binaries</a><a class="docs-footer-nextpage" href="../roadmap/">Roadmap »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/extensions/DimensionalData/index.html b/previews/PR3562/extensions/DimensionalData/index.html
index f92ce1b61a9..e59ecf09eb4 100644
--- a/previews/PR3562/extensions/DimensionalData/index.html
+++ b/previews/PR3562/extensions/DimensionalData/index.html
@@ -39,4 +39,4 @@
 2-element DimArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},1} with dimensions:
   Dim{:j} Categorical{String} String[&quot;a&quot;, &quot;b&quot;] ForwardOrdered
  &quot;a&quot;  x[2,a] + x[3,a] + x[4,a] ≤ 1
- &quot;b&quot;  x[2,b] + x[3,b] + x[4,b] ≤ 1</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>See the <a href="https://rafaqz.github.io/DimensionalData.jl/stable/">DimensionalData.jl documentation</a> for more details on the syntax and features of <code>DimensionalData.DimArray</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../packages/Plasmo/">« plasmo-dev/Plasmo.jl</a><a class="docs-footer-nextpage" href="../../moi/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ &quot;b&quot;  x[2,b] + x[3,b] + x[4,b] ≤ 1</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>See the <a href="https://rafaqz.github.io/DimensionalData.jl/stable/">DimensionalData.jl documentation</a> for more details on the syntax and features of <code>DimensionalData.DimArray</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../packages/Plasmo/">« plasmo-dev/Plasmo.jl</a><a class="docs-footer-nextpage" href="../../moi/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/extensions/introduction/index.html b/previews/PR3562/extensions/introduction/index.html
index f05479379ea..1b1e5be07e4 100644
--- a/previews/PR3562/extensions/introduction/index.html
+++ b/previews/PR3562/extensions/introduction/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox" checked/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#Adding-new-extensions"><span>Adding new extensions</span></a></li><li><a class="tocitem" href="#Weak-dependencies"><span>Weak dependencies</span></a></li></ul></li><li><a class="tocitem" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Extensions</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/extensions/introduction.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>This section of the documentation contains brief documentation for some popular JuMP extensions. The list of extensions is not exhaustive, but instead is intended to help you discover popular JuMP extensions, and to give you an overview of the types of extensions that are possible to write with JuMP.</p><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>Packages beginning with <code>jump-dev/</code> are developed and maintained by the JuMP developers.</p><p>Packages that do not begin with <code>jump-dev/</code> are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.</p><h2 id="Adding-new-extensions"><a class="docs-heading-anchor" href="#Adding-new-extensions">Adding new extensions</a><a id="Adding-new-extensions-1"></a><a class="docs-heading-anchor-permalink" href="#Adding-new-extensions" title="Permalink"></a></h2><p>Written an extension? Add it to this section of the JuMP documentation by making a pull request to the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/packages.toml"><code>docs/packages.toml</code></a> file.</p><h2 id="Weak-dependencies"><a class="docs-heading-anchor" href="#Weak-dependencies">Weak dependencies</a><a id="Weak-dependencies-1"></a><a class="docs-heading-anchor-permalink" href="#Weak-dependencies" title="Permalink"></a></h2><p>Some extensions listed in this section are implemented using the <a href="https://pkgdocs.julialang.org/v1/creating-packages/#Weak-dependencies">weak dependency</a> feature added to Julia in v1.9. These extensions are activated if and only if you have <code>JuMP</code> and the other package loaded into your current scope with <code>using</code> or <code>import</code>.</p><div class="admonition is-compat"><header class="admonition-header">Compat</header><div class="admonition-body"><p>Using a weak dependency requires Julia v1.9 or later.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../packages/SCIP/">« scipopt/SCIP.jl</a><a class="docs-footer-nextpage" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox" checked/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#Adding-new-extensions"><span>Adding new extensions</span></a></li><li><a class="tocitem" href="#Weak-dependencies"><span>Weak dependencies</span></a></li></ul></li><li><a class="tocitem" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Extensions</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/extensions/introduction.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>This section of the documentation contains brief documentation for some popular JuMP extensions. The list of extensions is not exhaustive, but instead is intended to help you discover popular JuMP extensions, and to give you an overview of the types of extensions that are possible to write with JuMP.</p><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>Packages beginning with <code>jump-dev/</code> are developed and maintained by the JuMP developers.</p><p>Packages that do not begin with <code>jump-dev/</code> are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.</p><h2 id="Adding-new-extensions"><a class="docs-heading-anchor" href="#Adding-new-extensions">Adding new extensions</a><a id="Adding-new-extensions-1"></a><a class="docs-heading-anchor-permalink" href="#Adding-new-extensions" title="Permalink"></a></h2><p>Written an extension? Add it to this section of the JuMP documentation by making a pull request to the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/packages.toml"><code>docs/packages.toml</code></a> file.</p><h2 id="Weak-dependencies"><a class="docs-heading-anchor" href="#Weak-dependencies">Weak dependencies</a><a id="Weak-dependencies-1"></a><a class="docs-heading-anchor-permalink" href="#Weak-dependencies" title="Permalink"></a></h2><p>Some extensions listed in this section are implemented using the <a href="https://pkgdocs.julialang.org/v1/creating-packages/#Weak-dependencies">weak dependency</a> feature added to Julia in v1.9. These extensions are activated if and only if you have <code>JuMP</code> and the other package loaded into your current scope with <code>using</code> or <code>import</code>.</p><div class="admonition is-compat"><header class="admonition-header">Compat</header><div class="admonition-body"><p>Using a weak dependency requires Julia v1.9 or later.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../packages/SCIP/">« scipopt/SCIP.jl</a><a class="docs-footer-nextpage" href="../../packages/DiffOpt/">jump-dev/DiffOpt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/index.html b/previews/PR3562/index.html
index ec805bc3703..6cd1286fa55 100644
--- a/previews/PR3562/index.html
+++ b/previews/PR3562/index.html
@@ -10,4 +10,4 @@
     journal = {Mathematical Programming Computation},
     year = {2023},
     doi = {10.1007/s12532-023-00239-3}
-}</code></pre><h2 id="NumFOCUS"><a class="docs-heading-anchor" href="#NumFOCUS">NumFOCUS</a><a id="NumFOCUS-1"></a><a class="docs-heading-anchor-permalink" href="#NumFOCUS" title="Permalink"></a></h2><p><img src="assets/numfocus-logo.png" alt="NumFOCUS logo"/></p><p>JuMP is a Sponsored Project of NumFOCUS, a 501(c)(3) nonprofit charity in the United States. NumFOCUS provides JuMP with fiscal, legal, and administrative support to help ensure the health and sustainability of the project. Visit <a href="https://numfocus.org">numfocus.org</a> for more information.</p><p>You can support JuMP by <a href="https://numfocus.salsalabs.org/donate-to-jump/index.html">donating</a>.</p><p>Donations to JuMP are managed by NumFOCUS. For donors in the United States, your gift is tax-deductible to the extent provided by law. As with any donation, you should consult with your tax adviser about your particular tax situation.</p><p>JuMP&#39;s largest expense is the annual JuMP-dev workshop. Donations will help us provide travel support for JuMP-dev attendees and take advantage of other opportunities that arise to support JuMP development.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>JuMP is licensed under the <a href="https://mozilla.org/MPL/2.0/">MPL-2.0 software license</a>. Consult the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/LICENSE.md">license</a> and the <a href="https://www.mozilla.org/en-US/MPL/2.0/FAQ/">Mozilla FAQ</a> for more information. In addition, JuMP is typically used in conjunction with solver packages and extensions which have their own licences. Consult their package repositories for the specific licenses that apply.</p></article><nav class="docs-footer"><a class="docs-footer-nextpage" href="should_i_use/">Should you use JuMP? »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><h2 id="NumFOCUS"><a class="docs-heading-anchor" href="#NumFOCUS">NumFOCUS</a><a id="NumFOCUS-1"></a><a class="docs-heading-anchor-permalink" href="#NumFOCUS" title="Permalink"></a></h2><p><img src="assets/numfocus-logo.png" alt="NumFOCUS logo"/></p><p>JuMP is a Sponsored Project of NumFOCUS, a 501(c)(3) nonprofit charity in the United States. NumFOCUS provides JuMP with fiscal, legal, and administrative support to help ensure the health and sustainability of the project. Visit <a href="https://numfocus.org">numfocus.org</a> for more information.</p><p>You can support JuMP by <a href="https://numfocus.salsalabs.org/donate-to-jump/index.html">donating</a>.</p><p>Donations to JuMP are managed by NumFOCUS. For donors in the United States, your gift is tax-deductible to the extent provided by law. As with any donation, you should consult with your tax adviser about your particular tax situation.</p><p>JuMP&#39;s largest expense is the annual JuMP-dev workshop. Donations will help us provide travel support for JuMP-dev attendees and take advantage of other opportunities that arise to support JuMP development.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>JuMP is licensed under the <a href="https://mozilla.org/MPL/2.0/">MPL-2.0 software license</a>. Consult the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/LICENSE.md">license</a> and the <a href="https://www.mozilla.org/en-US/MPL/2.0/FAQ/">Mozilla FAQ</a> for more information. In addition, JuMP is typically used in conjunction with solver packages and extensions which have their own licences. Consult their package repositories for the specific licenses that apply.</p></article><nav class="docs-footer"><a class="docs-footer-nextpage" href="should_i_use/">Should you use JuMP? »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/installation/index.html b/previews/PR3562/installation/index.html
index 11690f81c01..a93f4978b8c 100644
--- a/previews/PR3562/installation/index.html
+++ b/previews/PR3562/installation/index.html
@@ -21,4 +21,4 @@
   [4076af6c] ↓ JuMP v0.21.5 ⇒ v0.18.6
   [707a9f91] + JuMPeR v0.6.0
 Updating `~/jump_example/Manifest.toml`
-  ... lines omitted ...</code></pre><p>JuMPeR gets added at version <code>0.6.0</code> (<code>+ JuMPeR v0.6.0</code>), but JuMP gets downgraded from <code>0.21.5</code> to <code>0.18.6</code> (<code>↓ JuMP v0.21.5 ⇒ v0.18.6</code>)! The reason for this is that JuMPeR doesn&#39;t support a version of JuMP newer than <code>0.18.6</code>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Pay careful attention to the output of the package manager when adding new packages, especially when you see a package being downgraded.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../should_i_use/">« Should you use JuMP?</a><a class="docs-footer-nextpage" href="../tutorials/getting_started/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  ... lines omitted ...</code></pre><p>JuMPeR gets added at version <code>0.6.0</code> (<code>+ JuMPeR v0.6.0</code>), but JuMP gets downgraded from <code>0.21.5</code> to <code>0.18.6</code> (<code>↓ JuMP v0.21.5 ⇒ v0.18.6</code>)! The reason for this is that JuMPeR doesn&#39;t support a version of JuMP newer than <code>0.18.6</code>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Pay careful attention to the output of the package manager when adding new packages, especially when you see a package being downgraded.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../should_i_use/">« Should you use JuMP?</a><a class="docs-footer-nextpage" href="../tutorials/getting_started/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/callbacks/index.html b/previews/PR3562/manual/callbacks/index.html
index 1da4e0beb34..93e020ddafb 100644
--- a/previews/PR3562/manual/callbacks/index.html
+++ b/previews/PR3562/manual/callbacks/index.html
@@ -84,4 +84,4 @@
        end
 my_callback_function (generic function with 1 method)
 
-julia&gt; set_attribute(model, MOI.HeuristicCallback(), my_callback_function)</code></pre><p>The third argument to <code>submit</code> is a vector of JuMP variables, and the fourth argument is a vector of values corresponding to each variable.</p><p><code>MOI.submit</code> returns an enum that depends on whether the solver accepted the solution. The possible return codes are:</p><ul><li><code>MOI.HEURISTIC_SOLUTION_ACCEPTED</code></li><li><code>MOI.HEURISTIC_SOLUTION_REJECTED</code></li><li><code>MOI.HEURISTIC_SOLUTION_UNKNOWN</code></li></ul><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Some solvers may accept partial solutions. Others require a feasible integer solution for every variable. If in doubt, provide a complete solution.</p></div></div><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The heuristic solution callback <em>may</em> be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at <em>every</em> fractional primal solution.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solutions/">« Solutions</a><a class="docs-footer-nextpage" href="../complex/">Complex number support »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+julia&gt; set_attribute(model, MOI.HeuristicCallback(), my_callback_function)</code></pre><p>The third argument to <code>submit</code> is a vector of JuMP variables, and the fourth argument is a vector of values corresponding to each variable.</p><p><code>MOI.submit</code> returns an enum that depends on whether the solver accepted the solution. The possible return codes are:</p><ul><li><code>MOI.HEURISTIC_SOLUTION_ACCEPTED</code></li><li><code>MOI.HEURISTIC_SOLUTION_REJECTED</code></li><li><code>MOI.HEURISTIC_SOLUTION_UNKNOWN</code></li></ul><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Some solvers may accept partial solutions. Others require a feasible integer solution for every variable. If in doubt, provide a complete solution.</p></div></div><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The heuristic solution callback <em>may</em> be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at <em>every</em> fractional primal solution.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solutions/">« Solutions</a><a class="docs-footer-nextpage" href="../complex/">Complex number support »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/complex/index.html b/previews/PR3562/manual/complex/index.html
index 43deb166fb0..c151e27e848 100644
--- a/previews/PR3562/manual/complex/index.html
+++ b/previews/PR3562/manual/complex/index.html
@@ -148,4 +148,4 @@
 
 julia&gt; @constraint(model, H in HermitianPSDCone())
 [x[1]  im;
- -im   -x[2]] ∈ HermitianPSDCone()</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>The matrix <code>H</code> in <code>H in HermitianPSDCone()</code> must be a <code>LinearAlgebra.Hermitian</code> matrix type. A <code>build_constraint</code> error will be thrown if the matrix is a different matrix type.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Solver-independent Callbacks</a><a class="docs-footer-nextpage" href="../nonlinear/">Nonlinear Modeling »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ -im   -x[2]] ∈ HermitianPSDCone()</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>The matrix <code>H</code> in <code>H in HermitianPSDCone()</code> must be a <code>LinearAlgebra.Hermitian</code> matrix type. A <code>build_constraint</code> error will be thrown if the matrix is a different matrix type.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Solver-independent Callbacks</a><a class="docs-footer-nextpage" href="../nonlinear/">Nonlinear Modeling »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/constraints/index.html b/previews/PR3562/manual/constraints/index.html
index 093d7b23d28..b0f9260de11 100644
--- a/previews/PR3562/manual/constraints/index.html
+++ b/previews/PR3562/manual/constraints/index.html
@@ -753,4 +753,4 @@
 (x[1] == x[2]) - 0.0 = 0
 
 julia&gt; @constraint(model, x[1] == x[2] := rhs)
-x[1] == x[2] = false</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../expressions/">Expressions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+x[1] == x[2] = false</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../expressions/">Expressions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/containers/index.html b/previews/PR3562/manual/containers/index.html
index af0439e5f3b..4942dc577f6 100644
--- a/previews/PR3562/manual/containers/index.html
+++ b/previews/PR3562/manual/containers/index.html
@@ -224,4 +224,4 @@
 julia&gt; Containers.@container([i = 1:2, j = 1:4; condition(i, j)], i + j)
 JuMP.Containers.SparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 2 entries:
   [1, 2]  =  3
-  [1, 4]  =  5</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../objective/">« Objectives</a><a class="docs-footer-nextpage" href="../solutions/">Solutions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  [1, 4]  =  5</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../objective/">« Objectives</a><a class="docs-footer-nextpage" href="../solutions/">Solutions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/expressions/index.html b/previews/PR3562/manual/expressions/index.html
index b42d14cbfd6..401bdd97f15 100644
--- a/previews/PR3562/manual/expressions/index.html
+++ b/previews/PR3562/manual/expressions/index.html
@@ -227,4 +227,4 @@
 julia&gt; x
 2-element Vector{AffExpr}:
  1.1
- 0</code></pre><p>Note that for large expressions this will be slower due to the allocation of additional temporary objects.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../objective/">Objectives »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 0</code></pre><p>Note that for large expressions this will be slower due to the allocation of additional temporary objects.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../objective/">Objectives »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/models/index.html b/previews/PR3562/manual/models/index.html
index 146003cf7b6..a1488ffa768 100644
--- a/previews/PR3562/manual/models/index.html
+++ b/previews/PR3562/manual/models/index.html
@@ -234,4 +234,4 @@
 If you expected the solver to support your problem, you may have an error in your formulation. Otherwise, consider using a different solver.
 
 The list of available solvers, along with the problem types they support, is available at https://jump.dev/JuMP.jl/stable/installation/#Supported-solvers.
-Stacktrace:</code></pre><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Another downside of direct mode is that the behavior of querying solution information after modifying the problem is solver-specific. This can lead to errors, or the solver silently returning an incorrect value. See <a href="../solutions/#OptimizeNotCalled-errors">OptimizeNotCalled errors</a> for more information.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../tutorials/applications/two_stage_stochastic/">« Two-stage stochastic programs</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Stacktrace:</code></pre><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Another downside of direct mode is that the behavior of querying solution information after modifying the problem is solver-specific. This can lead to errors, or the solver silently returning an incorrect value. See <a href="../solutions/#OptimizeNotCalled-errors">OptimizeNotCalled errors</a> for more information.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../tutorials/applications/two_stage_stochastic/">« Two-stage stochastic programs</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/nlp/index.html b/previews/PR3562/manual/nlp/index.html
index 77194aac9b8..8a98a98a38c 100644
--- a/previews/PR3562/manual/nlp/index.html
+++ b/previews/PR3562/manual/nlp/index.html
@@ -344,4 +344,4 @@
  f1(x[1]) - 1.0 ≤ 0
  f2(x[1], x[2]) - 1.0 ≤ 0
  f3(x[2], x[3], x[4]) - 1.0 ≤ 0
- f4(x[1], x[3], x[4], x[5]) - 1.0 ≤ 0</code></pre><h2 id="Known-performance-issues"><a class="docs-heading-anchor" href="#Known-performance-issues">Known performance issues</a><a id="Known-performance-issues-1"></a><a class="docs-heading-anchor-permalink" href="#Known-performance-issues" title="Permalink"></a></h2><p>The macro-based input to JuMP&#39;s nonlinear interface can cause a performance issue if you:</p><ol><li>write a macro with a large number (hundreds) of terms</li><li>call that macro from within a function instead of from the top-level in global scope.</li></ol><p>The first issue does not depend on the number of resulting terms in the mathematical expression, but rather the number of terms in the Julia <code>Expr</code> representation of that expression. For example, the expression <code>sum(x[i] for i in 1:1_000_000)</code> contains one million mathematical terms, but the <code>Expr</code> representation is just a single sum.</p><p>The most common cause, other than a lot of tedious typing, is if you write a program that automatically writes a JuMP model as a text file, which you later execute. One example is <a href="https://github.com/lanl-ansi/MINLPLib.jl">MINLPlib.jl</a> which automatically transpiled models in the GAMS scalar format into JuMP examples.</p><p>As a rule of thumb, if you are writing programs to automatically generate expressions for the JuMP macros, you should target the <a href="#Raw-expression-input">Raw expression input</a> instead. For more information, read <a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1997">MathOptInterface Issue#1997</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nonlinear/">« Nonlinear Modeling</a><a class="docs-footer-nextpage" href="../../api/JuMP/">JuMP »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ f4(x[1], x[3], x[4], x[5]) - 1.0 ≤ 0</code></pre><h2 id="Known-performance-issues"><a class="docs-heading-anchor" href="#Known-performance-issues">Known performance issues</a><a id="Known-performance-issues-1"></a><a class="docs-heading-anchor-permalink" href="#Known-performance-issues" title="Permalink"></a></h2><p>The macro-based input to JuMP&#39;s nonlinear interface can cause a performance issue if you:</p><ol><li>write a macro with a large number (hundreds) of terms</li><li>call that macro from within a function instead of from the top-level in global scope.</li></ol><p>The first issue does not depend on the number of resulting terms in the mathematical expression, but rather the number of terms in the Julia <code>Expr</code> representation of that expression. For example, the expression <code>sum(x[i] for i in 1:1_000_000)</code> contains one million mathematical terms, but the <code>Expr</code> representation is just a single sum.</p><p>The most common cause, other than a lot of tedious typing, is if you write a program that automatically writes a JuMP model as a text file, which you later execute. One example is <a href="https://github.com/lanl-ansi/MINLPLib.jl">MINLPlib.jl</a> which automatically transpiled models in the GAMS scalar format into JuMP examples.</p><p>As a rule of thumb, if you are writing programs to automatically generate expressions for the JuMP macros, you should target the <a href="#Raw-expression-input">Raw expression input</a> instead. For more information, read <a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1997">MathOptInterface Issue#1997</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nonlinear/">« Nonlinear Modeling</a><a class="docs-footer-nextpage" href="../../api/JuMP/">JuMP »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/nonlinear/index.html b/previews/PR3562/manual/nonlinear/index.html
index 6a5d57655d0..c3e1318bc5e 100644
--- a/previews/PR3562/manual/nonlinear/index.html
+++ b/previews/PR3562/manual/nonlinear/index.html
@@ -196,4 +196,4 @@
         y[i] = x[i]^i
     end
     return sum(y)
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../complex/">« Complex number support</a><a class="docs-footer-nextpage" href="../nlp/">Nonlinear Modeling (Legacy) »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../complex/">« Complex number support</a><a class="docs-footer-nextpage" href="../nlp/">Nonlinear Modeling (Legacy) »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/objective/index.html b/previews/PR3562/manual/objective/index.html
index 5d0c045dc38..4803fb871f0 100644
--- a/previews/PR3562/manual/objective/index.html
+++ b/previews/PR3562/manual/objective/index.html
@@ -165,4 +165,4 @@
  2 x[1]
 
 julia&gt; @constraint(model, obj3 &lt;= 2.0)
-x[1] + x[2] ≤ 2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../expressions/">« Expressions</a><a class="docs-footer-nextpage" href="../containers/">Containers »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+x[1] + x[2] ≤ 2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../expressions/">« Expressions</a><a class="docs-footer-nextpage" href="../containers/">Containers »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/solutions/index.html b/previews/PR3562/manual/solutions/index.html
index 107aa925034..0597b0922b2 100644
--- a/previews/PR3562/manual/solutions/index.html
+++ b/previews/PR3562/manual/solutions/index.html
@@ -371,4 +371,4 @@
   x integer =&gt; 0.1</code></pre><p>You can also use the functional form, where the first argument is a function that maps variables to their primal values:</p><pre><code class="language-julia-repl hljs">julia&gt; optimize!(model)
 
 julia&gt; primal_feasibility_report(v -&gt; value(v), model)
-Dict{Any, Float64}()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../containers/">« Containers</a><a class="docs-footer-nextpage" href="../callbacks/">Solver-independent Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Dict{Any, Float64}()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../containers/">« Containers</a><a class="docs-footer-nextpage" href="../callbacks/">Solver-independent Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/manual/variables/index.html b/previews/PR3562/manual/variables/index.html
index 9d60ee07cb5..25611598d8c 100644
--- a/previews/PR3562/manual/variables/index.html
+++ b/previews/PR3562/manual/variables/index.html
@@ -632,4 +632,4 @@
 p*x
 
 julia&gt; typeof(px)
-QuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})</code></pre><h3 id="When-to-use-a-parameter"><a class="docs-heading-anchor" href="#When-to-use-a-parameter">When to use a parameter</a><a id="When-to-use-a-parameter-1"></a><a class="docs-heading-anchor-permalink" href="#When-to-use-a-parameter" title="Permalink"></a></h3><p>Parameters are most useful when solving nonlinear models in a sequence:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; using JuMP, Ipopt</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model(Ipopt.Optimizer);</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_silent(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x)</code><code class="nohighlight hljs ansi" style="display:block;">x</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, p in Parameter(1.0))</code><code class="nohighlight hljs ansi" style="display:block;">p</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @objective(model, Min, (x - p)^2)</code><code class="nohighlight hljs ansi" style="display:block;">x² - 2 p*x + p²</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; optimize!(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; value(x)</code><code class="nohighlight hljs ansi" style="display:block;">1.0</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_parameter_value(p, 5.0)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; optimize!(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; value(x)</code><code class="nohighlight hljs ansi" style="display:block;">5.0</code></pre><p>Using parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+QuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})</code></pre><h3 id="When-to-use-a-parameter"><a class="docs-heading-anchor" href="#When-to-use-a-parameter">When to use a parameter</a><a id="When-to-use-a-parameter-1"></a><a class="docs-heading-anchor-permalink" href="#When-to-use-a-parameter" title="Permalink"></a></h3><p>Parameters are most useful when solving nonlinear models in a sequence:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; using JuMP, Ipopt</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model(Ipopt.Optimizer);</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_silent(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x)</code><code class="nohighlight hljs ansi" style="display:block;">x</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, p in Parameter(1.0))</code><code class="nohighlight hljs ansi" style="display:block;">p</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @objective(model, Min, (x - p)^2)</code><code class="nohighlight hljs ansi" style="display:block;">x² - 2 p*x + p²</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; optimize!(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; value(x)</code><code class="nohighlight hljs ansi" style="display:block;">1.0</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_parameter_value(p, 5.0)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; optimize!(model)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; value(x)</code><code class="nohighlight hljs ansi" style="display:block;">5.0</code></pre><p>Using parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/background/duality/index.html b/previews/PR3562/moi/background/duality/index.html
index 5fe70bb8ca3..a9f80033162 100644
--- a/previews/PR3562/moi/background/duality/index.html
+++ b/previews/PR3562/moi/background/duality/index.html
@@ -81,4 +81,4 @@
        \max &amp; \sum b_k y_k \\
 \text{s.t.} \;\; &amp; C+C^\top - \sum (A_k+A_k^\top) y_k \in \mathcal{S}_+ \\
             &amp; C-C^\top - \sum(A_k-A_k^\top) y_k = 0
-\end{align}\]</p><p>and we recover <span>$Z = X + X^\top$</span>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../manual/modification/">« Problem modification</a><a class="docs-footer-nextpage" href="../infeasibility_certificates/">Infeasibility certificates »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>and we recover <span>$Z = X + X^\top$</span>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../manual/modification/">« Problem modification</a><a class="docs-footer-nextpage" href="../infeasibility_certificates/">Infeasibility certificates »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/background/infeasibility_certificates/index.html b/previews/PR3562/moi/background/infeasibility_certificates/index.html
index 5dcbaa7ecd0..0d153052efc 100644
--- a/previews/PR3562/moi/background/infeasibility_certificates/index.html
+++ b/previews/PR3562/moi/background/infeasibility_certificates/index.html
@@ -29,4 +29,4 @@
 \end{align}\]</p><p>and:</p><p class="math-container">\[-\sum_{i=1}^m b_i^\top (y_i + \eta d_i) &gt; -\sum_{i=1}^m b_i^\top y_i,\]</p><p>for any feasible dual solution <span>$y$</span>. The latter simplifies to <span>$-\sum_{i=1}^m b_i^\top d_i &gt; 0$</span>. For a maximization problem, the inequality is <span>$\sum_{i=1}^m b_i^\top d_i &lt; 0$</span>. (Note that these are the same inequality, modulo a <code>-</code> sign.)</p><p>If the solver has found a certificate of primal infeasibility:</p><ul><li><a href="../../reference/models/#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a> must be <code>INFEASIBLE</code></li><li><a href="../../reference/models/#MathOptInterface.DualStatus"><code>DualStatus</code></a> must be <code>INFEASIBILITY_CERTIFICATE</code></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintDual"><code>ConstraintDual</code></a> must be the corresponding value of <span>$d$</span></li><li><a href="../../reference/models/#MathOptInterface.DualObjectiveValue"><code>DualObjectiveValue</code></a> must be the value <span>$-\sum_{i=1}^m b_i^\top d_i$</span> for minimization problems and <span>$\sum_{i=1}^m b_i^\top d_i$</span> for maximization problems.</li></ul><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>The choice of whether to scale the ray <span>$d$</span> to have magnitude <code>1</code> is left to the solver.</p></div></div><h3 id="Infeasibility-certificates-of-variable-bounds"><a class="docs-heading-anchor" href="#Infeasibility-certificates-of-variable-bounds">Infeasibility certificates of variable bounds</a><a id="Infeasibility-certificates-of-variable-bounds-1"></a><a class="docs-heading-anchor-permalink" href="#Infeasibility-certificates-of-variable-bounds" title="Permalink"></a></h3><p>Many linear solvers (for example, Gurobi) do not provide explicit access to the primal infeasibility certificate of a variable bound. However, given a set of linear constraints:</p><p class="math-container">\[\begin{align}
 l_A \le A x \le u_A \\
 l_x \le x \le u_x,
-\end{align}\]</p><p>the primal certificate of the variable bounds can be computed using the primal certificate associated with the affine constraints, <span>$d$</span>. (Note that <span>$d$</span> will have one element for each row of the <span>$A$</span> matrix, and that some or all of the elements in the vectors <span>$l_A$</span> and <span>$u_A$</span> may be <span>$\pm \infty$</span>. If both <span>$l_A$</span> and <span>$u_A$</span> are finite for some row, the corresponding element in `<code>d</code>  must be <code>0</code>.)</p><p>Given <span>$d$</span>, compute <span>$\bar{d} = d^\top A$</span>. If the bound is finite, a certificate for the lower variable bound of <span>$x_i$</span> is <span>$\max\{\bar{d}_i, 0\}$</span>, and a certificate for the upper variable bound is <span>$\min\{\bar{d}_i, 0\}$</span>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../duality/">« Duality</a><a class="docs-footer-nextpage" href="../naming_conventions/">Naming conventions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>the primal certificate of the variable bounds can be computed using the primal certificate associated with the affine constraints, <span>$d$</span>. (Note that <span>$d$</span> will have one element for each row of the <span>$A$</span> matrix, and that some or all of the elements in the vectors <span>$l_A$</span> and <span>$u_A$</span> may be <span>$\pm \infty$</span>. If both <span>$l_A$</span> and <span>$u_A$</span> are finite for some row, the corresponding element in `<code>d</code>  must be <code>0</code>.)</p><p>Given <span>$d$</span>, compute <span>$\bar{d} = d^\top A$</span>. If the bound is finite, a certificate for the lower variable bound of <span>$x_i$</span> is <span>$\max\{\bar{d}_i, 0\}$</span>, and a certificate for the upper variable bound is <span>$\min\{\bar{d}_i, 0\}$</span>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../duality/">« Duality</a><a class="docs-footer-nextpage" href="../naming_conventions/">Naming conventions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/background/motivation/index.html b/previews/PR3562/moi/background/motivation/index.html
index 833f0854e70..77676ca12d2 100644
--- a/previews/PR3562/moi/background/motivation/index.html
+++ b/previews/PR3562/moi/background/motivation/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox" checked/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li class="is-active"><a class="tocitem" href>Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../duality/">Duality</a></li><li><a class="tocitem" href="../infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Introduction</a></li><li class="is-active"><a href>Motivation</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Motivation</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/motivation.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Motivation"><a class="docs-heading-anchor" href="#Motivation">Motivation</a><a id="Motivation-1"></a><a class="docs-heading-anchor-permalink" href="#Motivation" title="Permalink"></a></h1><p>MathOptInterface (MOI) is a replacement for <a href="https://github.com/JuliaOpt/MathProgBase.jl">MathProgBase</a>, the first-generation abstraction layer for mathematical optimization previously used by <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a> and <a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a>.</p><p>To address a number of limitations of MathProgBase, MOI is designed to:</p><ul><li>Be simple and extensible<ul><li>unifying linear, quadratic, and conic optimization,</li><li>seamlessly facilitating extensions to essentially arbitrary constraints and functions (for example, indicator constraints, complementarity constraints, and piecewise-linear functions)</li></ul></li><li>Be fast<ul><li>by allowing access to a solver&#39;s in-memory representation of a problem without writing intermediate files (when possible)</li><li>by using multiple dispatch and avoiding requiring containers of non-concrete types</li></ul></li><li>Allow a solver to return multiple results (for example, a pool of solutions)</li><li>Allow a solver to return extra arbitrary information via attributes (for example, variable- and constraint-wise membership in an irreducible inconsistent subset for infeasibility analysis)</li><li>Provide a greatly expanded set of status codes explaining what happened during the optimization procedure</li><li>Enable a solver to more precisely specify which problem classes it supports</li><li>Enable both primal and dual warm starts</li><li>Enable adding and removing both variables and constraints by indices that are not required to be consecutive</li><li>Enable any modification that the solver supports to an existing model</li><li>Avoid requiring the solver wrapper to store an additional copy of the problem data</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../">« Introduction</a><a class="docs-footer-nextpage" href="../../tutorials/example/">Solving a problem using MathOptInterface »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox" checked/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li class="is-active"><a class="tocitem" href>Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../duality/">Duality</a></li><li><a class="tocitem" href="../infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Introduction</a></li><li class="is-active"><a href>Motivation</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Motivation</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/motivation.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Motivation"><a class="docs-heading-anchor" href="#Motivation">Motivation</a><a id="Motivation-1"></a><a class="docs-heading-anchor-permalink" href="#Motivation" title="Permalink"></a></h1><p>MathOptInterface (MOI) is a replacement for <a href="https://github.com/JuliaOpt/MathProgBase.jl">MathProgBase</a>, the first-generation abstraction layer for mathematical optimization previously used by <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a> and <a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a>.</p><p>To address a number of limitations of MathProgBase, MOI is designed to:</p><ul><li>Be simple and extensible<ul><li>unifying linear, quadratic, and conic optimization,</li><li>seamlessly facilitating extensions to essentially arbitrary constraints and functions (for example, indicator constraints, complementarity constraints, and piecewise-linear functions)</li></ul></li><li>Be fast<ul><li>by allowing access to a solver&#39;s in-memory representation of a problem without writing intermediate files (when possible)</li><li>by using multiple dispatch and avoiding requiring containers of non-concrete types</li></ul></li><li>Allow a solver to return multiple results (for example, a pool of solutions)</li><li>Allow a solver to return extra arbitrary information via attributes (for example, variable- and constraint-wise membership in an irreducible inconsistent subset for infeasibility analysis)</li><li>Provide a greatly expanded set of status codes explaining what happened during the optimization procedure</li><li>Enable a solver to more precisely specify which problem classes it supports</li><li>Enable both primal and dual warm starts</li><li>Enable adding and removing both variables and constraints by indices that are not required to be consecutive</li><li>Enable any modification that the solver supports to an existing model</li><li>Avoid requiring the solver wrapper to store an additional copy of the problem data</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../">« Introduction</a><a class="docs-footer-nextpage" href="../../tutorials/example/">Solving a problem using MathOptInterface »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/background/naming_conventions/index.html b/previews/PR3562/moi/background/naming_conventions/index.html
index f9724a4bc32..b56b071df50 100644
--- a/previews/PR3562/moi/background/naming_conventions/index.html
+++ b/previews/PR3562/moi/background/naming_conventions/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox" checked/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../duality/">Duality</a></li><li><a class="tocitem" href="../infeasibility_certificates/">Infeasibility certificates</a></li><li class="is-active"><a class="tocitem" href>Naming conventions</a><ul class="internal"><li><a class="tocitem" href="#Sets"><span>Sets</span></a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Background</a></li><li class="is-active"><a href>Naming conventions</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Naming conventions</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/naming_conventions.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Naming-conventions"><a class="docs-heading-anchor" href="#Naming-conventions">Naming conventions</a><a id="Naming-conventions-1"></a><a class="docs-heading-anchor-permalink" href="#Naming-conventions" title="Permalink"></a></h1><p>MOI follows several conventions for naming functions and structures. These should also be followed by packages extending MOI.</p><h2 id="Sets"><a class="docs-heading-anchor" href="#Sets">Sets</a><a id="Sets-1"></a><a class="docs-heading-anchor-permalink" href="#Sets" title="Permalink"></a></h2><p>Sets encode the structure of constraints. Their names should follow the following conventions:</p><ul><li>Abstract types in the set hierarchy should begin with <code>Abstract</code> and end in <code>Set</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.AbstractScalarSet"><code>AbstractScalarSet</code></a>, <a href="../../reference/standard_form/#MathOptInterface.AbstractVectorSet"><code>AbstractVectorSet</code></a>.</li><li>Vector-valued conic sets should end with <code>Cone</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.NormInfinityCone"><code>NormInfinityCone</code></a>, <a href="../../reference/standard_form/#MathOptInterface.SecondOrderCone"><code>SecondOrderCone</code></a>.</li><li>Vector-valued Cartesian products should be plural and not end in <code>Cone</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.Nonnegatives"><code>Nonnegatives</code></a>, not <code>NonnegativeCone</code>.</li><li>Matrix-valued conic sets should provide two representations: <code>ConeSquare</code> and <code>ConeTriangle</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>RootDetConeTriangle</code></a> and <a href="../../reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>RootDetConeSquare</code></a>. See <a href="../../manual/standard_form/#Matrix-cones">Matrix cones</a> for more details.</li><li>Scalar sets should be singular, not plural, for example, <a href="../../reference/standard_form/#MathOptInterface.Integer"><code>Integer</code></a>, not <code>Integers</code>.</li><li>As much as possible, the names should follow established conventions in the domain where this set is used: for instance, convex sets should have names close to those of <a href="http://web.cvxr.com/cvx/doc/">CVX</a>, and constraint-programming sets should follow <a href="https://www.minizinc.org/doc-latest/en/">MiniZinc</a>&#39;s constraints.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../infeasibility_certificates/">« Infeasibility certificates</a><a class="docs-footer-nextpage" href="../../reference/standard_form/">Standard form »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox" checked/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../duality/">Duality</a></li><li><a class="tocitem" href="../infeasibility_certificates/">Infeasibility certificates</a></li><li class="is-active"><a class="tocitem" href>Naming conventions</a><ul class="internal"><li><a class="tocitem" href="#Sets"><span>Sets</span></a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Background</a></li><li class="is-active"><a href>Naming conventions</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Naming conventions</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/naming_conventions.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Naming-conventions"><a class="docs-heading-anchor" href="#Naming-conventions">Naming conventions</a><a id="Naming-conventions-1"></a><a class="docs-heading-anchor-permalink" href="#Naming-conventions" title="Permalink"></a></h1><p>MOI follows several conventions for naming functions and structures. These should also be followed by packages extending MOI.</p><h2 id="Sets"><a class="docs-heading-anchor" href="#Sets">Sets</a><a id="Sets-1"></a><a class="docs-heading-anchor-permalink" href="#Sets" title="Permalink"></a></h2><p>Sets encode the structure of constraints. Their names should follow the following conventions:</p><ul><li>Abstract types in the set hierarchy should begin with <code>Abstract</code> and end in <code>Set</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.AbstractScalarSet"><code>AbstractScalarSet</code></a>, <a href="../../reference/standard_form/#MathOptInterface.AbstractVectorSet"><code>AbstractVectorSet</code></a>.</li><li>Vector-valued conic sets should end with <code>Cone</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.NormInfinityCone"><code>NormInfinityCone</code></a>, <a href="../../reference/standard_form/#MathOptInterface.SecondOrderCone"><code>SecondOrderCone</code></a>.</li><li>Vector-valued Cartesian products should be plural and not end in <code>Cone</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.Nonnegatives"><code>Nonnegatives</code></a>, not <code>NonnegativeCone</code>.</li><li>Matrix-valued conic sets should provide two representations: <code>ConeSquare</code> and <code>ConeTriangle</code>, for example, <a href="../../reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>RootDetConeTriangle</code></a> and <a href="../../reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>RootDetConeSquare</code></a>. See <a href="../../manual/standard_form/#Matrix-cones">Matrix cones</a> for more details.</li><li>Scalar sets should be singular, not plural, for example, <a href="../../reference/standard_form/#MathOptInterface.Integer"><code>Integer</code></a>, not <code>Integers</code>.</li><li>As much as possible, the names should follow established conventions in the domain where this set is used: for instance, convex sets should have names close to those of <a href="http://web.cvxr.com/cvx/doc/">CVX</a>, and constraint-programming sets should follow <a href="https://www.minizinc.org/doc-latest/en/">MiniZinc</a>&#39;s constraints.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../infeasibility_certificates/">« Infeasibility certificates</a><a class="docs-footer-nextpage" href="../../reference/standard_form/">Standard form »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/changelog/index.html b/previews/PR3562/moi/changelog/index.html
index 53b0ca83607..45c8015a075 100644
--- a/previews/PR3562/moi/changelog/index.html
+++ b/previews/PR3562/moi/changelog/index.html
@@ -31,4 +31,4 @@
         end
         write(path, s)
     end
-end</code></pre><h2 id="v0.9.22-(May-22,-2021)"><a class="docs-heading-anchor" href="#v0.9.22-(May-22,-2021)">v0.9.22 (May 22, 2021)</a><a id="v0.9.22-(May-22,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.22-(May-22,-2021)" title="Permalink"></a></h2><p>This release contains backports from the ongoing development of the v0.10 release.</p><ul><li>Improved type inference in <code>Utilities</code>, <code>Bridges</code> and <code>FileFormats</code> submodules to reduce latency.</li><li>Improved performance of <code>Utilities.is_canonical</code>.</li><li>Fixed <code>Utilities.pass_nonvariable_constraints</code> with bridged variables.</li><li>Fixed performance regression of <code>Utilities.Model</code>.</li><li>Fixed ordering of objective setting in parser.</li></ul><h2 id="v0.9.21-(April-23,-2021)"><a class="docs-heading-anchor" href="#v0.9.21-(April-23,-2021)">v0.9.21 (April 23, 2021)</a><a id="v0.9.21-(April-23,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.21-(April-23,-2021)" title="Permalink"></a></h2><ul><li>Added <code>supports_shift_constant</code>.</li><li>Improve performance of bridging quadratic constraints.</li><li>Add precompilation statements.</li><li>Large improvements to the documentation.</li><li>Fix a variety of inference issues, benefiting precompilation and reducing initial latency.</li><li><code>RawParameter</code>s are now ignored when resetting a <code>CachingOptimizer</code>. Previously, changing the underlying optimizer after <code>RawParameter</code>s were set would throw an error.</li><li><code>Utilities.AbstractModel</code> is being refactored. This may break users interacting with private fields of a model generated using <code>@model</code>.</li></ul><h2 id="v0.9.20-(February-20,-2021)"><a class="docs-heading-anchor" href="#v0.9.20-(February-20,-2021)">v0.9.20 (February 20, 2021)</a><a id="v0.9.20-(February-20,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.20-(February-20,-2021)" title="Permalink"></a></h2><ul><li>Improved performance of <code>Utilities.ScalarFunctionIterator</code></li><li>Added support for <code>compute_conflict</code> to MOI layers</li><li>Added test with zero off-diagonal quadratic term in objective</li><li>Fixed double deletion of nested bridged <code>SingleVariable</code>/<code>VectorOfVariables</code> constraints</li><li>Fixed modification of un-set objective</li><li>Fixed function modification with duplicate terms</li><li>Made unit tests abort without failing if the problem class is not supported</li><li>Formatted code with JuliaFormatter</li><li>Clarified <code>BasisStatusCode</code>&#39;s docstring</li></ul><h2 id="v0.9.19-(December-1,-2020)"><a class="docs-heading-anchor" href="#v0.9.19-(December-1,-2020)">v0.9.19 (December 1, 2020)</a><a id="v0.9.19-(December-1,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.19-(December-1,-2020)" title="Permalink"></a></h2><ul><li>Added <code>CallbackNodeStatus</code> attribute</li><li>Added bridge from <code>GreaterThan</code> or <code>LessThan</code> to <code>Interval</code></li><li>Added tests for infeasibility certificates and double optimize</li><li>Fixed support for Julia v1.6</li><li>Re-organized MOI docs and added documentation for adding a test</li></ul><h2 id="v0.9.18-(November-3,-2020)"><a class="docs-heading-anchor" href="#v0.9.18-(November-3,-2020)">v0.9.18 (November 3, 2020)</a><a id="v0.9.18-(November-3,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.18-(November-3,-2020)" title="Permalink"></a></h2><ul><li>Various improvements for working with complex numbers</li><li>Added <code>GeoMeantoRelEntrBridge</code> to bridge a <code>GeometricMeanCone</code> constraint to a relative entropy constraint</li></ul><h2 id="v0.9.17-(September-21,-2020)"><a class="docs-heading-anchor" href="#v0.9.17-(September-21,-2020)">v0.9.17 (September 21, 2020)</a><a id="v0.9.17-(September-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.17-(September-21,-2020)" title="Permalink"></a></h2><ul><li>Fixed <code>CleverDict</code> with variable of negative index value</li><li>Implement <code>supports_add_constrained_variable</code> for <code>MockOptimizer</code></li></ul><h2 id="v0.9.16-(September-17,-2020)"><a class="docs-heading-anchor" href="#v0.9.16-(September-17,-2020)">v0.9.16 (September 17, 2020)</a><a id="v0.9.16-(September-17,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.16-(September-17,-2020)" title="Permalink"></a></h2><ul><li>Various fixes:<ul><li>32-bit support</li><li><code>CleverDict</code> with abstract value type</li><li>Checks in test suite</li></ul></li></ul><h2 id="v0.9.15-(September-14,-2020)"><a class="docs-heading-anchor" href="#v0.9.15-(September-14,-2020)">v0.9.15 (September 14, 2020)</a><a id="v0.9.15-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.15-(September-14,-2020)" title="Permalink"></a></h2><ul><li>Bridges improvements:<ul><li>(R)SOCtoNonConvexQuad bridge</li><li>ZeroOne bridge</li><li>Use <code>supports_add_constrained_variable</code> in <code>LazyBridgeOptimizer</code></li><li>Exposed <code>VariableBridgeCost</code> and <code>ConstraintBridgeCost</code> attributes</li><li>Prioritize constraining variables on creation according to these costs</li><li>Refactor bridge debugging</li></ul></li><li>Large performance improvements across all submodules</li><li>Lots of documentation improvements</li><li>FileFormats improvements:<ul><li>Update MathOptFormat to v0.5</li><li>Fix supported objectives in <code>FileFormats</code></li></ul></li><li>Testing improvements:<ul><li>Add name option for <code>basic_constraint_test</code></li></ul></li><li>Bug fixes and missing methods<ul><li>Add <code>length</code> for iterators</li><li>Fix bug with duplicate terms</li><li>Fix order of <code>LinearOfConstraintIndices</code></li></ul></li></ul><h2 id="v0.9.14-(May-30,-2020)"><a class="docs-heading-anchor" href="#v0.9.14-(May-30,-2020)">v0.9.14 (May 30, 2020)</a><a id="v0.9.14-(May-30,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.14-(May-30,-2020)" title="Permalink"></a></h2><ul><li>Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).</li><li>Bump JSONSchema dependency from v0.2 to v0.3 (#1090).</li><li>Documentation improvements:<ul><li>Fix typos (#1054, #1060, #1061, #1064, #1069, #1070).</li><li>Remove the outdated recommendation for a package implementing MOI for a solver <code>XXX</code> to be called <code>MathOptInterfaceXXX</code> (#1087).</li></ul></li><li>Utilities improvements:<ul><li>Fix <code>is_canonical</code> for quadratic functions (#1081, #1089).</li><li>Implement <code>add_constrained_variable[s]</code> for <code>CachingOptimizer</code> so that it is added as constrained variables to the underlying optimizer (#1084).</li><li>Add support for custom objective functions for <code>UniversalFallback</code> (#1086).</li><li>Deterministic ordering of constraints in <code>UniversalFallback</code> (#1088).</li></ul></li><li>Testing improvements:<ul><li>Add <code>NormOneCone</code>/<code>NormInfinityCone</code> tests (#1045).</li></ul></li><li>Bridges improvements:<ul><li>Add bridges from <code>Semiinteger</code> and <code>Semicontinuous</code> (#1059).</li><li>Implement getting <code>ConstraintSet</code> for <code>Variable.FlipSignBridge</code> (#1066).</li><li>Fix setting <code>ConstraintFunction</code> for <code>Constraint.ScalarizeBridge</code> (#1093).</li><li>Fix <code>NormOne</code>/<code>NormInf</code> bridges with nonzero constants (#1045).</li><li>Fix StackOverflow in <code>debug</code> (#1063).</li></ul></li><li>FileFormats improvements:<ul><li>[SDPA] Implement the extension for integer variables (#1079).</li><li>[SDPA] Ignore comments after <code>m</code> and <code>nblocks</code> and detect <code>dat-s</code> extension (#1077).</li><li>[SDPA] No scaling of off-diagonal coefficient (#1076).</li><li>[SDPA] Add missing negation of constant (#1075).</li></ul></li></ul><h2 id="v0.9.13-(March-24,-2020)"><a class="docs-heading-anchor" href="#v0.9.13-(March-24,-2020)">v0.9.13 (March 24, 2020)</a><a id="v0.9.13-(March-24,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.13-(March-24,-2020)" title="Permalink"></a></h2><ul><li>Added tests for <code>Semicontinuous</code> and <code>Semiinteger</code> variables (#1033).</li><li>Added tests for using <code>ExprGraph</code>s from NLP evaluators (#1043).</li><li>Update version compatibilities of dependencies (#1034, #1051, #1052).</li><li>Fixed typos in documentation (#1044).</li></ul><h2 id="v0.9.12-(February-28,-2020)"><a class="docs-heading-anchor" href="#v0.9.12-(February-28,-2020)">v0.9.12 (February 28, 2020)</a><a id="v0.9.12-(February-28,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.12-(February-28,-2020)" title="Permalink"></a></h2><ul><li>Fixed writing <code>NLPBlock</code> in MathOptFormat (#1037).</li><li>Fixed <code>MockOptimizer</code> for result attributes with non-one result index (#1039).</li><li>Updated test template with <code>instantiate</code> (#1032).</li></ul><h2 id="v0.9.11-(February-21,-2020)"><a class="docs-heading-anchor" href="#v0.9.11-(February-21,-2020)">v0.9.11 (February 21, 2020)</a><a id="v0.9.11-(February-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.11-(February-21,-2020)" title="Permalink"></a></h2><ul><li>Add an option for the model created by <code>Utilities.@model</code> to be a subtype of <code>AbstractOptimizer</code> (#1031).</li><li>Described dual cone in docstrings of <code>GeoMeanCone</code> and <code>RelativeEntropyCone</code> (#1018, #1028).</li><li>Fixed typos in documentation (#1022, #1024).</li><li>Fixed warning of unsupported attribute (#1027).</li><li>Added more rootdet/logdet conic tests (#1026).</li><li>Implemented <code>ConstraintDual</code> for <code>Constraint.GeoMeanBridge</code>, <code>Constraint.RootDetBridge</code> and <code>Constraint.LogDetBridge</code> and test duals in tests with <code>GeoMeanCone</code> and <code>RootDetConeTriangle</code> and <code>LogDetConeTriangle</code> cones (#1025, #1026).</li></ul><h2 id="v0.9.10-(January-31,-2020)"><a class="docs-heading-anchor" href="#v0.9.10-(January-31,-2020)">v0.9.10 (January 31, 2020)</a><a id="v0.9.10-(January-31,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.10-(January-31,-2020)" title="Permalink"></a></h2><ul><li>Added <code>OptimizerWithAttributes</code> grouping an optimizer constructor and a list of optimizer attributes (#1008).</li><li>Added <code>RelativeEntropyCone</code> with corresponding bridge into exponential cone constraints (#993).</li><li>Added <code>NormSpectralCone</code> and <code>NormNuclearCone</code> with corresponding bridges into positive semidefinite constraints (#976).</li><li>Added <code>supports_constrained_variable(s)</code> (#1004).</li><li>Added <code>dual_set_type</code> (#1002).</li><li>Added tests for vector specialized version of <code>delete</code> (#989, #1011).</li><li>Added PSD3 test (#1007).</li><li>Clarified dual solution of <code>Tests.pow1v</code> and <code>Tests.pow1f</code> (#1013).</li><li>Added support for <code>EqualTo</code> and <code>Zero</code> in <code>Bridges.Constraint.SplitIntervalBridge</code> (#1005).</li><li>Fixed <code>Utilities.vectorize</code> for empty vector (#1003).</li><li>Fixed free variables in LP writer (#1006).</li></ul><h2 id="v0.9.9-(December-29,-2019)"><a class="docs-heading-anchor" href="#v0.9.9-(December-29,-2019)">v0.9.9 (December 29, 2019)</a><a id="v0.9.9-(December-29,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.9-(December-29,-2019)" title="Permalink"></a></h2><ul><li>Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).</li><li>Improved performance of deletion of vector of variables in <code>MOI.Utilities.Model</code> (#983).</li><li>Updated to MutableArithmetics v0.2 (#981).</li><li>Added <code>MutableArithmetics.promote_operation</code> allocation tests (#975).</li><li>Fixed inference issue on Julia v1.1 (#982).</li></ul><h2 id="v0.9.8-(December-19,-2019)"><a class="docs-heading-anchor" href="#v0.9.8-(December-19,-2019)">v0.9.8 (December 19, 2019)</a><a id="v0.9.8-(December-19,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.8-(December-19,-2019)" title="Permalink"></a></h2><ul><li>Implemented MutableArithmetics API (#924).</li><li>Fixed callbacks with <code>CachingOptimizer</code> (#959).</li><li>Fixed <code>MOI.dimension</code> for <code>MOI.Complements</code> (#948).</li><li>Added fallback for <code>add_variables</code> (#972).</li><li>Added <code>is_diagonal_vectorized_index</code> utility (#965).</li><li>Improved linear constraints display in manual (#963, #964).</li><li>Bridges improvements:<ul><li>Added <code>IndicatorSet</code> to <code>SOS1</code> bridge (#877).</li><li>Added support for starting values for <code>Variable.VectorizeBridge</code> (#944).</li><li>Fixed <code>MOI.add_constraints</code> with non-bridged variable constraint on bridged variable (#951).</li><li>Fixed corner cases and docstring of <code>GeoMeanBridge</code> (#961, #962, #966).</li><li>Fixed choice between variable or constraint bridges for constrained variables (#973).</li><li>Improve performance of bridge shortest path (#945, #946, #956).</li><li>Added docstring for <code>test_delete_bridge</code> (#954).</li><li>Added Variable bridge tests (#952).</li></ul></li></ul><h2 id="v0.9.7-(October-30,-2019)"><a class="docs-heading-anchor" href="#v0.9.7-(October-30,-2019)">v0.9.7 (October 30, 2019)</a><a id="v0.9.7-(October-30,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.7-(October-30,-2019)" title="Permalink"></a></h2><ul><li>Implemented <code>_result_index_field</code> for <code>NLPBlockDual</code> (#934).</li><li>Fixed copy of model with starting values for vector constraints (#941).</li><li>Bridges improvements:<ul><li>Improved performance of <code>add_bridge</code> and added <code>has_bridge</code> (#935).</li><li>Added <code>AbstractSetMapBridge</code> for bridges between sets <code>S1</code>, <code>S2</code> such that there is a linear map <code>A</code> such that <code>A*S1 = S2</code> (#933).</li><li>Added support for starting values for <code>FlipSignBridge</code>, <code>VectorizeBridge</code>, <code>ScalarizeBridge</code>, <code>SlackBridge</code>, <code>SplitIntervalBridge</code>, <code>RSOCBridge</code>, <code>SOCRBridge</code> <code>NormInfinityBridge</code>, <code>SOCtoPSDBridge</code> and <code>RSOCtoPSDBridge</code> (#933, #936, #937, #938, #939).</li></ul></li></ul><h2 id="v0.9.6-(October-25,-2019)"><a class="docs-heading-anchor" href="#v0.9.6-(October-25,-2019)">v0.9.6 (October 25, 2019)</a><a id="v0.9.6-(October-25,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.6-(October-25,-2019)" title="Permalink"></a></h2><ul><li>Added complementarity constraints (#913).</li><li>Allowed <code>ModelLike</code> objects as value of attributes (#928).</li><li>Testing improvements:<ul><li>Added <code>dual_objective_value</code> option to <code>MOI.Test.TestConfig</code> (#922).</li><li>Added <code>InvalidIndex</code> tests in <code>basic_constraint_tests</code> (#921).</li><li>Added tests for the constant term in indicator constraint (#929).</li></ul></li><li>Bridges improvements:<ul><li>Added support for starting values for <code>Functionize</code> bridges (#923).</li><li>Added variable indices context to variable bridges (#920).</li><li>Fixed a typo in printing o <code>debug_supports</code> (#927).</li></ul></li></ul><h2 id="v0.9.5-(October-9,-2019)"><a class="docs-heading-anchor" href="#v0.9.5-(October-9,-2019)">v0.9.5 (October 9, 2019)</a><a id="v0.9.5-(October-9,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.5-(October-9,-2019)" title="Permalink"></a></h2><ul><li>Clarified <code>PrimalStatus</code>/<code>DualStatus</code> to be <code>NO_SOLUTION</code> if <code>result_index</code> is out of bounds (#912).</li><li>Added tolerance for checks  and use <code>ResultCount</code> + 1 for the <code>result_index</code> in <code>MOI.Test.solve_result_status</code> (#910, #917).</li><li>Use <code>0.5</code> instead of <code>2.0</code> for power in PowerCone in <code>basic_constraint_test</code> (#916).</li><li>Bridges improvements:<ul><li>Added debug utilities for unsupported variable/constraint/objective (#861).</li><li>Fixed deletion of variables in bridged <code>VectorOfVariables</code> constraints (#909).</li><li>Fixed <code>result_index</code> with objective bridges (#911).</li></ul></li></ul><h2 id="v0.9.4-(October-2,-2019)"><a class="docs-heading-anchor" href="#v0.9.4-(October-2,-2019)">v0.9.4 (October 2, 2019)</a><a id="v0.9.4-(October-2,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.4-(October-2,-2019)" title="Permalink"></a></h2><ul><li>Added solver-independent MIP callbacks (#782).</li><li>Implements submit for <code>Utilities.CachingOptimizer</code> and <code>Bridges.AbstractBridgeOptimizer</code> (#906).</li><li>Added tests for result count of solution attributes (#901, #904).</li><li>Added <code>NumberOfThreads</code> attribute (#892).</li><li>Added <code>Utilities.get_bounds</code> to get the bounds on a variable (#890).</li><li>Added a note on duplicate coefficients in documentation (#581).</li><li>Added result index in <code>ConstraintBasisStatus</code> (#898).</li><li>Added extension dictionary to <code>Utilities.Model</code> (#884, #895).</li><li>Fixed deletion of constrained variables for CachingOptimizer (#905).</li><li>Implemented <code>Utilities.shift_constraint</code> for <code>Test.UnknownScalarSet</code> (#896).</li><li>Bridges improvements:<ul><li>Added Variable.RSOCtoSOCBridge (#907).</li><li>Implemented <code>MOI.get</code> for <code>ConstraintFunction</code>/<code>ConstraintSet</code> for <code>Bridges.Constraint.SquareBridge</code> (#899).</li></ul></li></ul><h2 id="v0.9.3-(September-20,-2019)"><a class="docs-heading-anchor" href="#v0.9.3-(September-20,-2019)">v0.9.3 (September 20, 2019)</a><a id="v0.9.3-(September-20,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.3-(September-20,-2019)" title="Permalink"></a></h2><ul><li>Fixed ambiguity detected in Julia v1.3 (#891, #893).</li><li>Fixed missing sets from <code>ListOfSupportedConstraints</code> (#880).</li><li>Fixed copy of <code>VectorOfVariables</code> constraints with duplicate indices (#886).</li><li>Added extension dictionary to MOIU.Model (#884).</li><li>Implemented <code>MOI.get</code> for function and set for <code>GeoMeanBridge</code> (#888).</li><li>Updated documentation for SingleVariable indices and bridges (#885).</li><li>Testing improvements:<ul><li>Added more comprehensive tests for names (#882).</li><li>Added tests for <code>SingleVariable</code> duals (#883).</li><li>Added tests for <code>DualExponentialCone</code> and <code>DualPowerCone</code> (#873).</li></ul></li><li>Improvements for arbitrary coefficient type:<ul><li>Fixed <code>==</code> for sets with mutable fields (#887).</li><li>Removed some <code>Float64</code> assumptions in bridges (#878).</li><li>Automatic selection of <code>Constraint.[Scalar|Vector]FunctionizeBridge</code> (#889).</li></ul></li></ul><h2 id="v0.9.2-(September-5,-2019)"><a class="docs-heading-anchor" href="#v0.9.2-(September-5,-2019)">v0.9.2 (September 5, 2019)</a><a id="v0.9.2-(September-5,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.2-(September-5,-2019)" title="Permalink"></a></h2><ul><li>Implemented model printing for <code>MOI.ModelLike</code> and specialized it for models defined in MOI (864).</li><li>Generalized <code>contlinear</code> tests for arbitrary coefficient type (#855).</li><li>Fixed <code>supports_constraint</code> for <code>Semiinteger</code> and <code>Semicontinuous</code> and <code>supports</code> for <code>ObjectiveFunction</code> (#859).</li><li>Fixed Allocate-Load copy for single variable constraints (#856).</li><li>Bridges improvements:<ul><li>Add objective bridges (#789).</li><li>Fixed <code>Variable.RSOCtoPSDBridge</code> for dimension 2 (#869).</li><li>Added <code>Variable.SOCtoRSOCBridge</code> (#865).</li><li>Added <code>Constraint.SOCRBridge</code> and disable <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> (#751).</li><li>Fixed <code>added_constraint_types</code> for <code>Contraint.LogDetBridge</code> and <code>Constraint.RootDetBridge</code> (#870).</li></ul></li></ul><h2 id="v0.9.1-(August-22,-2019)"><a class="docs-heading-anchor" href="#v0.9.1-(August-22,-2019)">v0.9.1 (August 22, 2019)</a><a id="v0.9.1-(August-22,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.1-(August-22,-2019)" title="Permalink"></a></h2><ul><li>Fix support for Julia v1.2 (#834).</li><li>L<em>1 and L</em>∞ norm epigraph cones and corresponding bridges to LP were added (#818).</li><li>Added tests to <code>MOI.Test.nametest</code> (#833).</li><li>Fix <code>MOI.Test.soc3test</code> for solvers not supporting infeasibility certificates (#839).</li><li>Implements <code>operate</code> for operators <code>*</code> and <code>/</code> between vector function and constant (#837).</li><li>Implements <code>show</code> for <code>MOI.Utilities.IndexMap</code> (#847).</li><li>Fix corner cases for mapping of variables in <code>MOI.Utilities.CachingOptimizer</code> and substitution of variables in <code>MOI.Bridges.AbstractBridgeOptimizer</code> (#848).</li><li>Fix transformation of constant terms for <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> and <code>MOI.Bridges.Constraint.RSOCtoPSDBridge</code> (#840).</li></ul><h2 id="v0.9.0-(August-13,-2019)"><a class="docs-heading-anchor" href="#v0.9.0-(August-13,-2019)">v0.9.0 (August 13, 2019)</a><a id="v0.9.0-(August-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.0-(August-13,-2019)" title="Permalink"></a></h2><ul><li>Support for Julia v0.6 and v0.7 was dropped (#714, #717).</li><li>A <code>MOI.Utilities.Model</code> implementation of <code>ModelLike</code>, this should replace most use cases of <code>MOI.Utilities.@model</code> (#781).</li><li><code>add_constrained_variable</code> and <code>add_constrained_variables</code> were added (#759).</li><li>Support for indicator constraints was added (#709, #712).</li><li><code>DualObjectiveValue</code> attribute was added (#473).</li><li><code>RawParameter</code> attribute was added (#733).</li><li>A <code>dual_set</code> function was added (#804).</li><li>A <code>Benchmarks</code> submodule was added to facilitate solver benchmarking (#769).</li><li>A <code>submit</code> function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).</li><li>The field of <code>ObjectiveValue</code> was renamed to <code>result_index</code> (#729).</li><li>The <code>_constant</code> and <code>Utilities.getconstant</code> function were renamed to <code>constant</code></li><li><code>REDUCTION_CERTIFICATE</code> result status was added (#734).</li><li>Abstract matrix sets were added (#731).</li><li>Testing improvements:<ul><li>The testing guideline was updated (#728).</li><li>Quadratic tests were added (#697).</li><li>Unit tests for <code>RawStatusString</code>, <code>SolveTime</code>, <code>Silent</code> and <code>SolverName</code> were added (#726, #741).</li><li>A rotated second-order cone test was added (#759).</li><li>A power cone test was added (#768).</li><li>Tests for <code>ZeroOne</code> variables with variable bounds were added (#772).</li><li>An unbounded test was added (#773).</li><li>Existing tests had a few updates (#702, #703, #763).</li></ul></li><li>Documentation improvements:<ul><li>Added a section on <code>CachingOptimizer</code> (#777).</li><li>Added a section on <code>UniversalFallback</code>, <code>Model</code> and <code>@model</code> (#762).</li><li>Transition the knapsack example to a doctest with MockOptimizer (#786).</li></ul></li><li>Utilities improvements:<ul><li>A <code>CleverDict</code> utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).</li><li>The <code>Utilities.constant</code> function was renamed to <code>Utilities.constant_vector</code> (#740).</li><li>Implement optimizer attributes for CachingOptimizer (#745).</li><li>Rename <code>Utilities.add_scalar_constraint</code> to <code>Utilities.normalize_and_add_constraint</code> (#801).</li><li><code>operate</code> with <code>vcat</code>, <code>SingleVariable</code> and <code>VectorOfVariables</code> now returns a <code>VectorOfVariables</code> (#616).</li><li>Fix a type piracy of <code>operate</code> (#784).</li><li>The <code>load_constraint</code> fallback signature was fixed (#760).</li><li>The <code>set_dot</code> function was extended to work with sparse arrays (#805).</li></ul></li><li>Bridges improvements:<ul><li>The bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement <code>ConstraintFunction</code> and <code>ConstraintSet</code> if the user wants to recover them. As a consequence, the <code>@bridge</code> macro was removed (#722).</li><li>Bridge are now instantiated with a <code>bridge_constraint</code> function instead of using a constructor (#730).</li><li>Fix constraint attributes for bridges (#699).</li><li>Constraint bridges were moved to the <code>Bridges/Constraint</code> submodule so they should now inherit from <code>MOI.Bridges.Constraint.Abstract</code> and should implement <code>MOI.Bridges.Constraint.concrete_bridge_type</code> instead of <code>MOI.Bridges.concrete_bridge_type</code> (#756).</li><li>Variable bridges were added in (#759).</li><li>Various improvements (#746, #747).</li></ul></li></ul><h2 id="v0.8.4-(March-13,-2019)"><a class="docs-heading-anchor" href="#v0.8.4-(March-13,-2019)">v0.8.4 (March 13, 2019)</a><a id="v0.8.4-(March-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.4-(March-13,-2019)" title="Permalink"></a></h2><ul><li>Performance improvement in <code>default_copy_to</code> and bridge optimizer (#696).</li><li>Add <code>Silent</code> and implement setting optimizer attributes in caching and mock optimizers (#695).</li><li>Add <code>Functionize</code> bridges (SingleVariable and VectorOfVariables) (#659).</li><li>Minor typo fixes (#694).</li></ul><h2 id="v0.8.3-(March-6,-2019)"><a class="docs-heading-anchor" href="#v0.8.3-(March-6,-2019)">v0.8.3 (March 6, 2019)</a><a id="v0.8.3-(March-6,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.3-(March-6,-2019)" title="Permalink"></a></h2><ul><li>Use zero constant in scalar constraint function of <code>MOI.Test.copytest</code> (#691).</li><li>Fix variable deletion with <code>SingleVariable</code> objective function (#690).</li><li>Fix <code>LazyBridgeOptimizer</code> with bridges that add no constraints (#689).</li><li>Error message improvements (#673, #685, #686, #688).</li><li>Documentation improvements (#682, #683, #687).</li><li>Basis status:<ul><li>Remove <code>VariableBasisStatus</code> (#679).</li><li>Test <code>ConstraintBasisStatus</code> and implement it in bridges (#678).</li></ul></li><li>Fix inference of <code>NumberOfVariables</code> and <code>NumberOfConstraints</code> (#677).</li><li>Implement division between a quadratic function and a number (#675).</li></ul><h2 id="v0.8.2-(February-7,-2019)"><a class="docs-heading-anchor" href="#v0.8.2-(February-7,-2019)">v0.8.2 (February 7, 2019)</a><a id="v0.8.2-(February-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.2-(February-7,-2019)" title="Permalink"></a></h2><ul><li>Add <code>RawStatusString</code> attribute (#629).</li><li>Do not set names to the optimizer but only to the cache in <code>CachingOptimizer</code> (#638).</li><li>Make scalar MOI functions act as scalars in broadcast (#646).</li><li>Add function utilities:<ul><li>Implement <code>Base.zero</code> (#634), <code>Base.iszero</code> (#643), add missing arithmetic operations (#644, #645) and fix division (#648).</li><li>Add a <code>vectorize</code> function that turns a vector of <code>ScalarAffineFunction</code> into a <code>VectorAffineFunction</code> (#642).</li></ul></li><li>Improve support for starting values:<ul><li>Show a warning in copy when starting values are not supported instead of throwing an error (#630).</li><li>Fix <code>UniversalFallback</code> for getting an variable or constraint attribute set to no indices (#623).</li><li>Add a test in <code>contlineartest</code> with partially set <code>VariablePrimalStart</code>.</li></ul></li><li>Bridges improvements:<ul><li>Fix <code>StackOverFlow</code> in <code>LazyBridgeOptimizer</code> when there is a cycle in the graph of bridges.</li><li>Add <code>Slack</code> bridges (#610, #650).</li><li>Add <code>FlipSign</code> bridges (#658).</li></ul></li><li>Add tests with duplicate coefficients in <code>ScalarAffineFunction</code> and <code>VectorAffineFunction</code> (#639).</li><li>Use tolerance to compare <code>VariablePrimal</code> in <code>rotatedsoc1</code> test (#632).</li><li>Use a zero constant in <code>ScalarAffineFunction</code> of constraints in <code>psdt2</code> (#622).</li></ul><h2 id="v0.8.1-(January-7,-2019)"><a class="docs-heading-anchor" href="#v0.8.1-(January-7,-2019)">v0.8.1 (January 7, 2019)</a><a id="v0.8.1-(January-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.1-(January-7,-2019)" title="Permalink"></a></h2><ul><li>Adding an NLP objective now overrides any objective set using the <code>ObjectiveFunction</code> attribute (#619).</li><li>Rename <code>fullbridgeoptimizer</code> into <code>full_bridge_optimizer</code> (#621).</li><li>Allow custom constraint types with <code>full_bridge_optimizer</code> (#617).</li><li>Add <code>Vectorize</code> bridge which transforms scalar linear constraints into vector linear constraints (#615).</li></ul><h2 id="v0.8.0-(December-18,-2018)"><a class="docs-heading-anchor" href="#v0.8.0-(December-18,-2018)">v0.8.0 (December 18, 2018)</a><a id="v0.8.0-(December-18,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.0-(December-18,-2018)" title="Permalink"></a></h2><ul><li>Rename all enum values to follow the JuMP naming guidelines for constants, for example, <code>Optimal</code> becomes <code>OPTIMAL</code>, and <code>DualInfeasible</code> becomes <code>DUAL_INFEASIBLE</code>.</li><li>Rename CachingOptimizer methods for style compliance.</li><li>Add an <code>MOI.TerminationStatusCode</code> called <code>ALMOST_DUAL_INFEASIBLE</code>.</li></ul><h2 id="v0.7.0-(December-13,-2018)"><a class="docs-heading-anchor" href="#v0.7.0-(December-13,-2018)">v0.7.0 (December 13, 2018)</a><a id="v0.7.0-(December-13,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.7.0-(December-13,-2018)" title="Permalink"></a></h2><ul><li>Test that <code>MOI.TerminationStatus</code> is <code>MOI.OptimizeNotCalled</code> before <code>MOI.optimize!</code> is called.</li><li>Check <code>supports_default_copy_to</code> in tests (#594).</li><li>Key pieces of information like optimality, infeasibility, etc., are now reported through <code>TerminationStatusCode</code>. It is typically no longer necessary to check the result statuses in addition to the termination status.</li><li>Add perspective dimension to log-det cone (#593).</li></ul><h2 id="v0.6.4-(November-27,-2018)"><a class="docs-heading-anchor" href="#v0.6.4-(November-27,-2018)">v0.6.4 (November 27, 2018)</a><a id="v0.6.4-(November-27,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.4-(November-27,-2018)" title="Permalink"></a></h2><ul><li>Add <code>OptimizeNotCalled</code> termination status (#577) and improve documentation of other statuses (#575).</li><li>Add a solver naming guideline (#578).</li><li>Make <code>FeasibilitySense</code> the default <code>ObjectiveSense</code> (#579).</li><li>Fix <code>Utilities.@model</code> and <code>Bridges.@bridge</code> macros for functions and sets defined outside MOI (#582).</li><li>Document solver-specific attributes (#580) and implement them in <code>Utilities.CachingOptimizer</code> (#565).</li></ul><h2 id="v0.6.3-(November-16,-2018)"><a class="docs-heading-anchor" href="#v0.6.3-(November-16,-2018)">v0.6.3 (November 16, 2018)</a><a id="v0.6.3-(November-16,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.3-(November-16,-2018)" title="Permalink"></a></h2><ul><li>Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)</li><li>Attributes may now be partially set (some values could be <code>nothing</code>). (#563)</li><li>Performance improvements in Utilities.Model (#549, #567, #568)</li><li>Fix bug in QuadtoSOC (#558).</li><li>New <code>supports_default_copy_to</code> method that optimizers should implement to control caching behavior.</li><li>Documentation improvements.</li></ul><h2 id="v0.6.2-(October-26,-2018)"><a class="docs-heading-anchor" href="#v0.6.2-(October-26,-2018)">v0.6.2 (October 26, 2018)</a><a id="v0.6.2-(October-26,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.2-(October-26,-2018)" title="Permalink"></a></h2><ul><li>Improve hygiene of <code>@model</code> macro (#544).</li><li>Fix bug in copy tests (#543).</li><li>Fix bug in UniversalFallback attribute getter (#540).</li><li>Allow all correct solutions for <code>solve_blank_obj</code> unit test (#537).</li><li>Add errors for Allocate-Load and bad constraints (#534).</li><li>[performance] Add specialized implementation of <code>hash</code> for <code>VariableIndex</code> (#533).</li><li>[performance] Construct the name to object dictionaries lazily in model (#535).</li><li>Add the <code>QuadtoSOC</code> bridge which transforms <code>ScalarQuadraticFunction</code> constraints into <code>RotatedSecondOrderCone</code> (#483).</li></ul><h2 id="v0.6.1-(September-22,-2018)"><a class="docs-heading-anchor" href="#v0.6.1-(September-22,-2018)">v0.6.1 (September 22, 2018)</a><a id="v0.6.1-(September-22,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.1-(September-22,-2018)" title="Permalink"></a></h2><ul><li>Enable <code>PositiveSemidefiniteConeSquare</code> set and quadratic functions in <code>MOIB.fullbridgeoptimizer</code> (#524).</li><li>Add warning in the bridge between <code>PositiveSemidefiniteConeSquare</code> and <code>PositiveSemidefiniteConeTriangle</code> when the matrix is almost symmetric (#522).</li><li>Modify <code>MOIT.copytest</code> to not add multiples constraints on the same variable (#521).</li><li>Add missing keyword argument in one of <code>MOIU.add_scalar_constraint</code> methods (#520).</li></ul><h2 id="v0.6.0-(August-30,-2018)"><a class="docs-heading-anchor" href="#v0.6.0-(August-30,-2018)">v0.6.0 (August 30, 2018)</a><a id="v0.6.0-(August-30,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.0-(August-30,-2018)" title="Permalink"></a></h2><ul><li>The <code>MOIU.@model</code> and <code>MOIB.@bridge</code> macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.</li><li>Rename functions according to the <a href="https://jump.dev/JuMP.jl/stable/developers/style/">JuMP style guide</a>:<ul><li><code>copy!</code> with keyword arguments <code>copynames</code> and <code>warnattributes</code> -&gt; <code>copy_to</code> with keyword arguments <code>copy_names</code> and <code>warn_attributes</code>;</li><li><code>set!</code> -&gt; <code>set</code>;</li><li><code>addvariable[s]!</code> -&gt; <code>add_variable[s]</code>;</li><li><code>supportsconstraint</code> -&gt; <code>supports_constraint</code>;</li><li><code>addconstraint[s]!</code> -&gt; <code>add_constraint[s]</code>;</li><li><code>isvalid</code> -&gt; <code>is_valid</code>;</li><li><code>isempty</code> -&gt; <code>is_empty</code>;</li><li><code>Base.delete!</code> -&gt; <code>delete</code>;</li><li><code>modify!</code> -&gt; <code>modify</code>;</li><li><code>transform!</code> -&gt; <code>transform</code>;</li><li><code>initialize!</code> -&gt; <code>initialize</code>;</li><li><code>write</code> -&gt; <code>write_to_file</code>; and</li><li><code>read!</code> -&gt; <code>read_from_file</code>.</li></ul></li><li>Remove <code>free!</code> (use <code>Base.finalize</code> instead).</li><li>Add the <code>SquarePSD</code> bridge which transforms <code>PositiveSemidefiniteConeTriangle</code> constraints into <code>PositiveSemidefiniteConeTriangle</code>.</li><li>Add result fallback for <code>ConstraintDual</code> of variable-wise constraint, <code>ConstraintPrimal</code> and <code>ObjectiveValue</code>.</li><li>Add tests for <code>ObjectiveBound</code>.</li><li>Add test for empty rows in vector linear constraint.</li><li>Rework errors: <code>CannotError</code> has been renamed <code>NotAllowedError</code> and the distinction between <code>UnsupportedError</code> and <code>NotAllowedError</code> is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)</li><li><code>canget</code> is removed. <code>NoSolution</code> is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).</li></ul><h2 id="v0.5.0-(August-5,-2018)"><a class="docs-heading-anchor" href="#v0.5.0-(August-5,-2018)">v0.5.0 (August 5, 2018)</a><a id="v0.5.0-(August-5,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.5.0-(August-5,-2018)" title="Permalink"></a></h2><ul><li>Fix names with CachingOptimizer.</li><li>Cleanup thanks to @mohamed82008.</li><li>Added a universal fallback for constraints.</li><li>Fast utilities for function canonicalization thanks to @rdeits.</li><li>Renamed <code>dimension</code> field to <code>side_dimension</code> in the context of matrix-like sets.</li><li>New and improved tests for cases like duplicate terms and <code>ObjectiveBound</code>.</li><li>Removed <code>cantransform</code>, <code>canaddconstraint</code>, <code>canaddvariable</code>, <code>canset</code>, <code>canmodify</code>, and <code>candelete</code> functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of <code>UnsupportedError</code> indicate unsupported operations, while subtypes of <code>CannotError</code> indicate operations that cannot be performed in the current state.</li><li>The API for <code>copy!</code> is updated to remove the CopyResult type.</li><li>Updates for the new JuMP style guide.</li></ul><h2 id="v0.4.1-(June-28,-2018)"><a class="docs-heading-anchor" href="#v0.4.1-(June-28,-2018)">v0.4.1 (June 28, 2018)</a><a id="v0.4.1-(June-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.4.1-(June-28,-2018)" title="Permalink"></a></h2><ul><li>Fixes vector function modification on 32 bits.</li><li>Fixes Bellman-Ford algorithm for bridges.</li><li>Added an NLP test with <code>FeasibilitySense</code>.</li><li>Update modification documentation.</li></ul><h2 id="v0.4.0-(June-23,-2018)"><a class="docs-heading-anchor" href="#v0.4.0-(June-23,-2018)">v0.4.0 (June 23, 2018)</a><a id="v0.4.0-(June-23,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.4.0-(June-23,-2018)" title="Permalink"></a></h2><ul><li>Helper constructors for <code>VectorAffineTerm</code> and <code>VectorQuadraticTerm</code>.</li><li>Added <code>modify_lhs</code> to <code>TestConfig</code>.</li><li>Additional unit tests for optimizers.</li><li>Added a type parameter to <code>CachingOptimizer</code> for the <code>optimizer</code> field.</li><li>New API for problem modification (#388)</li><li>Tests pass without deprecation warnings on Julia 0.7.</li><li>Small fixes and documentation updates.</li></ul><h2 id="v0.3.0-(May-25,-2018)"><a class="docs-heading-anchor" href="#v0.3.0-(May-25,-2018)">v0.3.0 (May 25, 2018)</a><a id="v0.3.0-(May-25,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.3.0-(May-25,-2018)" title="Permalink"></a></h2><ul><li>Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.</li><li>Improvements to <code>MockOptimizer</code>.</li><li>Significant changes to <code>Bridges</code>.</li><li>New and improved unit tests.</li><li>Fixes for Julia 0.7.</li></ul><h2 id="v0.2.0-(April-24,-2018)"><a class="docs-heading-anchor" href="#v0.2.0-(April-24,-2018)">v0.2.0 (April 24, 2018)</a><a id="v0.2.0-(April-24,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.2.0-(April-24,-2018)" title="Permalink"></a></h2><ul><li>Improvements to and better coverage of <code>Tests</code>.</li><li>Documentation fixes.</li><li><code>SolverName</code> attribute.</li><li>Changes to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).</li><li>Addition of NLP tests.</li><li>Introduction of <code>UniversalFallback</code>.</li><li><code>copynames</code> keyword argument to <code>MOI.copy!</code>.</li><li>Add Bridges submodule.</li></ul><h2 id="v0.1.0-(February-28,-2018)"><a class="docs-heading-anchor" href="#v0.1.0-(February-28,-2018)">v0.1.0 (February 28, 2018)</a><a id="v0.1.0-(February-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.1.0-(February-28,-2018)" title="Permalink"></a></h2><ul><li>Initial public release.</li><li>The framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.</li></ul></article><nav class="docs-footer"><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><h2 id="v0.9.22-(May-22,-2021)"><a class="docs-heading-anchor" href="#v0.9.22-(May-22,-2021)">v0.9.22 (May 22, 2021)</a><a id="v0.9.22-(May-22,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.22-(May-22,-2021)" title="Permalink"></a></h2><p>This release contains backports from the ongoing development of the v0.10 release.</p><ul><li>Improved type inference in <code>Utilities</code>, <code>Bridges</code> and <code>FileFormats</code> submodules to reduce latency.</li><li>Improved performance of <code>Utilities.is_canonical</code>.</li><li>Fixed <code>Utilities.pass_nonvariable_constraints</code> with bridged variables.</li><li>Fixed performance regression of <code>Utilities.Model</code>.</li><li>Fixed ordering of objective setting in parser.</li></ul><h2 id="v0.9.21-(April-23,-2021)"><a class="docs-heading-anchor" href="#v0.9.21-(April-23,-2021)">v0.9.21 (April 23, 2021)</a><a id="v0.9.21-(April-23,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.21-(April-23,-2021)" title="Permalink"></a></h2><ul><li>Added <code>supports_shift_constant</code>.</li><li>Improve performance of bridging quadratic constraints.</li><li>Add precompilation statements.</li><li>Large improvements to the documentation.</li><li>Fix a variety of inference issues, benefiting precompilation and reducing initial latency.</li><li><code>RawParameter</code>s are now ignored when resetting a <code>CachingOptimizer</code>. Previously, changing the underlying optimizer after <code>RawParameter</code>s were set would throw an error.</li><li><code>Utilities.AbstractModel</code> is being refactored. This may break users interacting with private fields of a model generated using <code>@model</code>.</li></ul><h2 id="v0.9.20-(February-20,-2021)"><a class="docs-heading-anchor" href="#v0.9.20-(February-20,-2021)">v0.9.20 (February 20, 2021)</a><a id="v0.9.20-(February-20,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.20-(February-20,-2021)" title="Permalink"></a></h2><ul><li>Improved performance of <code>Utilities.ScalarFunctionIterator</code></li><li>Added support for <code>compute_conflict</code> to MOI layers</li><li>Added test with zero off-diagonal quadratic term in objective</li><li>Fixed double deletion of nested bridged <code>SingleVariable</code>/<code>VectorOfVariables</code> constraints</li><li>Fixed modification of un-set objective</li><li>Fixed function modification with duplicate terms</li><li>Made unit tests abort without failing if the problem class is not supported</li><li>Formatted code with JuliaFormatter</li><li>Clarified <code>BasisStatusCode</code>&#39;s docstring</li></ul><h2 id="v0.9.19-(December-1,-2020)"><a class="docs-heading-anchor" href="#v0.9.19-(December-1,-2020)">v0.9.19 (December 1, 2020)</a><a id="v0.9.19-(December-1,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.19-(December-1,-2020)" title="Permalink"></a></h2><ul><li>Added <code>CallbackNodeStatus</code> attribute</li><li>Added bridge from <code>GreaterThan</code> or <code>LessThan</code> to <code>Interval</code></li><li>Added tests for infeasibility certificates and double optimize</li><li>Fixed support for Julia v1.6</li><li>Re-organized MOI docs and added documentation for adding a test</li></ul><h2 id="v0.9.18-(November-3,-2020)"><a class="docs-heading-anchor" href="#v0.9.18-(November-3,-2020)">v0.9.18 (November 3, 2020)</a><a id="v0.9.18-(November-3,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.18-(November-3,-2020)" title="Permalink"></a></h2><ul><li>Various improvements for working with complex numbers</li><li>Added <code>GeoMeantoRelEntrBridge</code> to bridge a <code>GeometricMeanCone</code> constraint to a relative entropy constraint</li></ul><h2 id="v0.9.17-(September-21,-2020)"><a class="docs-heading-anchor" href="#v0.9.17-(September-21,-2020)">v0.9.17 (September 21, 2020)</a><a id="v0.9.17-(September-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.17-(September-21,-2020)" title="Permalink"></a></h2><ul><li>Fixed <code>CleverDict</code> with variable of negative index value</li><li>Implement <code>supports_add_constrained_variable</code> for <code>MockOptimizer</code></li></ul><h2 id="v0.9.16-(September-17,-2020)"><a class="docs-heading-anchor" href="#v0.9.16-(September-17,-2020)">v0.9.16 (September 17, 2020)</a><a id="v0.9.16-(September-17,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.16-(September-17,-2020)" title="Permalink"></a></h2><ul><li>Various fixes:<ul><li>32-bit support</li><li><code>CleverDict</code> with abstract value type</li><li>Checks in test suite</li></ul></li></ul><h2 id="v0.9.15-(September-14,-2020)"><a class="docs-heading-anchor" href="#v0.9.15-(September-14,-2020)">v0.9.15 (September 14, 2020)</a><a id="v0.9.15-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.15-(September-14,-2020)" title="Permalink"></a></h2><ul><li>Bridges improvements:<ul><li>(R)SOCtoNonConvexQuad bridge</li><li>ZeroOne bridge</li><li>Use <code>supports_add_constrained_variable</code> in <code>LazyBridgeOptimizer</code></li><li>Exposed <code>VariableBridgeCost</code> and <code>ConstraintBridgeCost</code> attributes</li><li>Prioritize constraining variables on creation according to these costs</li><li>Refactor bridge debugging</li></ul></li><li>Large performance improvements across all submodules</li><li>Lots of documentation improvements</li><li>FileFormats improvements:<ul><li>Update MathOptFormat to v0.5</li><li>Fix supported objectives in <code>FileFormats</code></li></ul></li><li>Testing improvements:<ul><li>Add name option for <code>basic_constraint_test</code></li></ul></li><li>Bug fixes and missing methods<ul><li>Add <code>length</code> for iterators</li><li>Fix bug with duplicate terms</li><li>Fix order of <code>LinearOfConstraintIndices</code></li></ul></li></ul><h2 id="v0.9.14-(May-30,-2020)"><a class="docs-heading-anchor" href="#v0.9.14-(May-30,-2020)">v0.9.14 (May 30, 2020)</a><a id="v0.9.14-(May-30,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.14-(May-30,-2020)" title="Permalink"></a></h2><ul><li>Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).</li><li>Bump JSONSchema dependency from v0.2 to v0.3 (#1090).</li><li>Documentation improvements:<ul><li>Fix typos (#1054, #1060, #1061, #1064, #1069, #1070).</li><li>Remove the outdated recommendation for a package implementing MOI for a solver <code>XXX</code> to be called <code>MathOptInterfaceXXX</code> (#1087).</li></ul></li><li>Utilities improvements:<ul><li>Fix <code>is_canonical</code> for quadratic functions (#1081, #1089).</li><li>Implement <code>add_constrained_variable[s]</code> for <code>CachingOptimizer</code> so that it is added as constrained variables to the underlying optimizer (#1084).</li><li>Add support for custom objective functions for <code>UniversalFallback</code> (#1086).</li><li>Deterministic ordering of constraints in <code>UniversalFallback</code> (#1088).</li></ul></li><li>Testing improvements:<ul><li>Add <code>NormOneCone</code>/<code>NormInfinityCone</code> tests (#1045).</li></ul></li><li>Bridges improvements:<ul><li>Add bridges from <code>Semiinteger</code> and <code>Semicontinuous</code> (#1059).</li><li>Implement getting <code>ConstraintSet</code> for <code>Variable.FlipSignBridge</code> (#1066).</li><li>Fix setting <code>ConstraintFunction</code> for <code>Constraint.ScalarizeBridge</code> (#1093).</li><li>Fix <code>NormOne</code>/<code>NormInf</code> bridges with nonzero constants (#1045).</li><li>Fix StackOverflow in <code>debug</code> (#1063).</li></ul></li><li>FileFormats improvements:<ul><li>[SDPA] Implement the extension for integer variables (#1079).</li><li>[SDPA] Ignore comments after <code>m</code> and <code>nblocks</code> and detect <code>dat-s</code> extension (#1077).</li><li>[SDPA] No scaling of off-diagonal coefficient (#1076).</li><li>[SDPA] Add missing negation of constant (#1075).</li></ul></li></ul><h2 id="v0.9.13-(March-24,-2020)"><a class="docs-heading-anchor" href="#v0.9.13-(March-24,-2020)">v0.9.13 (March 24, 2020)</a><a id="v0.9.13-(March-24,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.13-(March-24,-2020)" title="Permalink"></a></h2><ul><li>Added tests for <code>Semicontinuous</code> and <code>Semiinteger</code> variables (#1033).</li><li>Added tests for using <code>ExprGraph</code>s from NLP evaluators (#1043).</li><li>Update version compatibilities of dependencies (#1034, #1051, #1052).</li><li>Fixed typos in documentation (#1044).</li></ul><h2 id="v0.9.12-(February-28,-2020)"><a class="docs-heading-anchor" href="#v0.9.12-(February-28,-2020)">v0.9.12 (February 28, 2020)</a><a id="v0.9.12-(February-28,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.12-(February-28,-2020)" title="Permalink"></a></h2><ul><li>Fixed writing <code>NLPBlock</code> in MathOptFormat (#1037).</li><li>Fixed <code>MockOptimizer</code> for result attributes with non-one result index (#1039).</li><li>Updated test template with <code>instantiate</code> (#1032).</li></ul><h2 id="v0.9.11-(February-21,-2020)"><a class="docs-heading-anchor" href="#v0.9.11-(February-21,-2020)">v0.9.11 (February 21, 2020)</a><a id="v0.9.11-(February-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.11-(February-21,-2020)" title="Permalink"></a></h2><ul><li>Add an option for the model created by <code>Utilities.@model</code> to be a subtype of <code>AbstractOptimizer</code> (#1031).</li><li>Described dual cone in docstrings of <code>GeoMeanCone</code> and <code>RelativeEntropyCone</code> (#1018, #1028).</li><li>Fixed typos in documentation (#1022, #1024).</li><li>Fixed warning of unsupported attribute (#1027).</li><li>Added more rootdet/logdet conic tests (#1026).</li><li>Implemented <code>ConstraintDual</code> for <code>Constraint.GeoMeanBridge</code>, <code>Constraint.RootDetBridge</code> and <code>Constraint.LogDetBridge</code> and test duals in tests with <code>GeoMeanCone</code> and <code>RootDetConeTriangle</code> and <code>LogDetConeTriangle</code> cones (#1025, #1026).</li></ul><h2 id="v0.9.10-(January-31,-2020)"><a class="docs-heading-anchor" href="#v0.9.10-(January-31,-2020)">v0.9.10 (January 31, 2020)</a><a id="v0.9.10-(January-31,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.10-(January-31,-2020)" title="Permalink"></a></h2><ul><li>Added <code>OptimizerWithAttributes</code> grouping an optimizer constructor and a list of optimizer attributes (#1008).</li><li>Added <code>RelativeEntropyCone</code> with corresponding bridge into exponential cone constraints (#993).</li><li>Added <code>NormSpectralCone</code> and <code>NormNuclearCone</code> with corresponding bridges into positive semidefinite constraints (#976).</li><li>Added <code>supports_constrained_variable(s)</code> (#1004).</li><li>Added <code>dual_set_type</code> (#1002).</li><li>Added tests for vector specialized version of <code>delete</code> (#989, #1011).</li><li>Added PSD3 test (#1007).</li><li>Clarified dual solution of <code>Tests.pow1v</code> and <code>Tests.pow1f</code> (#1013).</li><li>Added support for <code>EqualTo</code> and <code>Zero</code> in <code>Bridges.Constraint.SplitIntervalBridge</code> (#1005).</li><li>Fixed <code>Utilities.vectorize</code> for empty vector (#1003).</li><li>Fixed free variables in LP writer (#1006).</li></ul><h2 id="v0.9.9-(December-29,-2019)"><a class="docs-heading-anchor" href="#v0.9.9-(December-29,-2019)">v0.9.9 (December 29, 2019)</a><a id="v0.9.9-(December-29,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.9-(December-29,-2019)" title="Permalink"></a></h2><ul><li>Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).</li><li>Improved performance of deletion of vector of variables in <code>MOI.Utilities.Model</code> (#983).</li><li>Updated to MutableArithmetics v0.2 (#981).</li><li>Added <code>MutableArithmetics.promote_operation</code> allocation tests (#975).</li><li>Fixed inference issue on Julia v1.1 (#982).</li></ul><h2 id="v0.9.8-(December-19,-2019)"><a class="docs-heading-anchor" href="#v0.9.8-(December-19,-2019)">v0.9.8 (December 19, 2019)</a><a id="v0.9.8-(December-19,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.8-(December-19,-2019)" title="Permalink"></a></h2><ul><li>Implemented MutableArithmetics API (#924).</li><li>Fixed callbacks with <code>CachingOptimizer</code> (#959).</li><li>Fixed <code>MOI.dimension</code> for <code>MOI.Complements</code> (#948).</li><li>Added fallback for <code>add_variables</code> (#972).</li><li>Added <code>is_diagonal_vectorized_index</code> utility (#965).</li><li>Improved linear constraints display in manual (#963, #964).</li><li>Bridges improvements:<ul><li>Added <code>IndicatorSet</code> to <code>SOS1</code> bridge (#877).</li><li>Added support for starting values for <code>Variable.VectorizeBridge</code> (#944).</li><li>Fixed <code>MOI.add_constraints</code> with non-bridged variable constraint on bridged variable (#951).</li><li>Fixed corner cases and docstring of <code>GeoMeanBridge</code> (#961, #962, #966).</li><li>Fixed choice between variable or constraint bridges for constrained variables (#973).</li><li>Improve performance of bridge shortest path (#945, #946, #956).</li><li>Added docstring for <code>test_delete_bridge</code> (#954).</li><li>Added Variable bridge tests (#952).</li></ul></li></ul><h2 id="v0.9.7-(October-30,-2019)"><a class="docs-heading-anchor" href="#v0.9.7-(October-30,-2019)">v0.9.7 (October 30, 2019)</a><a id="v0.9.7-(October-30,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.7-(October-30,-2019)" title="Permalink"></a></h2><ul><li>Implemented <code>_result_index_field</code> for <code>NLPBlockDual</code> (#934).</li><li>Fixed copy of model with starting values for vector constraints (#941).</li><li>Bridges improvements:<ul><li>Improved performance of <code>add_bridge</code> and added <code>has_bridge</code> (#935).</li><li>Added <code>AbstractSetMapBridge</code> for bridges between sets <code>S1</code>, <code>S2</code> such that there is a linear map <code>A</code> such that <code>A*S1 = S2</code> (#933).</li><li>Added support for starting values for <code>FlipSignBridge</code>, <code>VectorizeBridge</code>, <code>ScalarizeBridge</code>, <code>SlackBridge</code>, <code>SplitIntervalBridge</code>, <code>RSOCBridge</code>, <code>SOCRBridge</code> <code>NormInfinityBridge</code>, <code>SOCtoPSDBridge</code> and <code>RSOCtoPSDBridge</code> (#933, #936, #937, #938, #939).</li></ul></li></ul><h2 id="v0.9.6-(October-25,-2019)"><a class="docs-heading-anchor" href="#v0.9.6-(October-25,-2019)">v0.9.6 (October 25, 2019)</a><a id="v0.9.6-(October-25,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.6-(October-25,-2019)" title="Permalink"></a></h2><ul><li>Added complementarity constraints (#913).</li><li>Allowed <code>ModelLike</code> objects as value of attributes (#928).</li><li>Testing improvements:<ul><li>Added <code>dual_objective_value</code> option to <code>MOI.Test.TestConfig</code> (#922).</li><li>Added <code>InvalidIndex</code> tests in <code>basic_constraint_tests</code> (#921).</li><li>Added tests for the constant term in indicator constraint (#929).</li></ul></li><li>Bridges improvements:<ul><li>Added support for starting values for <code>Functionize</code> bridges (#923).</li><li>Added variable indices context to variable bridges (#920).</li><li>Fixed a typo in printing o <code>debug_supports</code> (#927).</li></ul></li></ul><h2 id="v0.9.5-(October-9,-2019)"><a class="docs-heading-anchor" href="#v0.9.5-(October-9,-2019)">v0.9.5 (October 9, 2019)</a><a id="v0.9.5-(October-9,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.5-(October-9,-2019)" title="Permalink"></a></h2><ul><li>Clarified <code>PrimalStatus</code>/<code>DualStatus</code> to be <code>NO_SOLUTION</code> if <code>result_index</code> is out of bounds (#912).</li><li>Added tolerance for checks  and use <code>ResultCount</code> + 1 for the <code>result_index</code> in <code>MOI.Test.solve_result_status</code> (#910, #917).</li><li>Use <code>0.5</code> instead of <code>2.0</code> for power in PowerCone in <code>basic_constraint_test</code> (#916).</li><li>Bridges improvements:<ul><li>Added debug utilities for unsupported variable/constraint/objective (#861).</li><li>Fixed deletion of variables in bridged <code>VectorOfVariables</code> constraints (#909).</li><li>Fixed <code>result_index</code> with objective bridges (#911).</li></ul></li></ul><h2 id="v0.9.4-(October-2,-2019)"><a class="docs-heading-anchor" href="#v0.9.4-(October-2,-2019)">v0.9.4 (October 2, 2019)</a><a id="v0.9.4-(October-2,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.4-(October-2,-2019)" title="Permalink"></a></h2><ul><li>Added solver-independent MIP callbacks (#782).</li><li>Implements submit for <code>Utilities.CachingOptimizer</code> and <code>Bridges.AbstractBridgeOptimizer</code> (#906).</li><li>Added tests for result count of solution attributes (#901, #904).</li><li>Added <code>NumberOfThreads</code> attribute (#892).</li><li>Added <code>Utilities.get_bounds</code> to get the bounds on a variable (#890).</li><li>Added a note on duplicate coefficients in documentation (#581).</li><li>Added result index in <code>ConstraintBasisStatus</code> (#898).</li><li>Added extension dictionary to <code>Utilities.Model</code> (#884, #895).</li><li>Fixed deletion of constrained variables for CachingOptimizer (#905).</li><li>Implemented <code>Utilities.shift_constraint</code> for <code>Test.UnknownScalarSet</code> (#896).</li><li>Bridges improvements:<ul><li>Added Variable.RSOCtoSOCBridge (#907).</li><li>Implemented <code>MOI.get</code> for <code>ConstraintFunction</code>/<code>ConstraintSet</code> for <code>Bridges.Constraint.SquareBridge</code> (#899).</li></ul></li></ul><h2 id="v0.9.3-(September-20,-2019)"><a class="docs-heading-anchor" href="#v0.9.3-(September-20,-2019)">v0.9.3 (September 20, 2019)</a><a id="v0.9.3-(September-20,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.3-(September-20,-2019)" title="Permalink"></a></h2><ul><li>Fixed ambiguity detected in Julia v1.3 (#891, #893).</li><li>Fixed missing sets from <code>ListOfSupportedConstraints</code> (#880).</li><li>Fixed copy of <code>VectorOfVariables</code> constraints with duplicate indices (#886).</li><li>Added extension dictionary to MOIU.Model (#884).</li><li>Implemented <code>MOI.get</code> for function and set for <code>GeoMeanBridge</code> (#888).</li><li>Updated documentation for SingleVariable indices and bridges (#885).</li><li>Testing improvements:<ul><li>Added more comprehensive tests for names (#882).</li><li>Added tests for <code>SingleVariable</code> duals (#883).</li><li>Added tests for <code>DualExponentialCone</code> and <code>DualPowerCone</code> (#873).</li></ul></li><li>Improvements for arbitrary coefficient type:<ul><li>Fixed <code>==</code> for sets with mutable fields (#887).</li><li>Removed some <code>Float64</code> assumptions in bridges (#878).</li><li>Automatic selection of <code>Constraint.[Scalar|Vector]FunctionizeBridge</code> (#889).</li></ul></li></ul><h2 id="v0.9.2-(September-5,-2019)"><a class="docs-heading-anchor" href="#v0.9.2-(September-5,-2019)">v0.9.2 (September 5, 2019)</a><a id="v0.9.2-(September-5,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.2-(September-5,-2019)" title="Permalink"></a></h2><ul><li>Implemented model printing for <code>MOI.ModelLike</code> and specialized it for models defined in MOI (864).</li><li>Generalized <code>contlinear</code> tests for arbitrary coefficient type (#855).</li><li>Fixed <code>supports_constraint</code> for <code>Semiinteger</code> and <code>Semicontinuous</code> and <code>supports</code> for <code>ObjectiveFunction</code> (#859).</li><li>Fixed Allocate-Load copy for single variable constraints (#856).</li><li>Bridges improvements:<ul><li>Add objective bridges (#789).</li><li>Fixed <code>Variable.RSOCtoPSDBridge</code> for dimension 2 (#869).</li><li>Added <code>Variable.SOCtoRSOCBridge</code> (#865).</li><li>Added <code>Constraint.SOCRBridge</code> and disable <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> (#751).</li><li>Fixed <code>added_constraint_types</code> for <code>Contraint.LogDetBridge</code> and <code>Constraint.RootDetBridge</code> (#870).</li></ul></li></ul><h2 id="v0.9.1-(August-22,-2019)"><a class="docs-heading-anchor" href="#v0.9.1-(August-22,-2019)">v0.9.1 (August 22, 2019)</a><a id="v0.9.1-(August-22,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.1-(August-22,-2019)" title="Permalink"></a></h2><ul><li>Fix support for Julia v1.2 (#834).</li><li>L<em>1 and L</em>∞ norm epigraph cones and corresponding bridges to LP were added (#818).</li><li>Added tests to <code>MOI.Test.nametest</code> (#833).</li><li>Fix <code>MOI.Test.soc3test</code> for solvers not supporting infeasibility certificates (#839).</li><li>Implements <code>operate</code> for operators <code>*</code> and <code>/</code> between vector function and constant (#837).</li><li>Implements <code>show</code> for <code>MOI.Utilities.IndexMap</code> (#847).</li><li>Fix corner cases for mapping of variables in <code>MOI.Utilities.CachingOptimizer</code> and substitution of variables in <code>MOI.Bridges.AbstractBridgeOptimizer</code> (#848).</li><li>Fix transformation of constant terms for <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> and <code>MOI.Bridges.Constraint.RSOCtoPSDBridge</code> (#840).</li></ul><h2 id="v0.9.0-(August-13,-2019)"><a class="docs-heading-anchor" href="#v0.9.0-(August-13,-2019)">v0.9.0 (August 13, 2019)</a><a id="v0.9.0-(August-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.9.0-(August-13,-2019)" title="Permalink"></a></h2><ul><li>Support for Julia v0.6 and v0.7 was dropped (#714, #717).</li><li>A <code>MOI.Utilities.Model</code> implementation of <code>ModelLike</code>, this should replace most use cases of <code>MOI.Utilities.@model</code> (#781).</li><li><code>add_constrained_variable</code> and <code>add_constrained_variables</code> were added (#759).</li><li>Support for indicator constraints was added (#709, #712).</li><li><code>DualObjectiveValue</code> attribute was added (#473).</li><li><code>RawParameter</code> attribute was added (#733).</li><li>A <code>dual_set</code> function was added (#804).</li><li>A <code>Benchmarks</code> submodule was added to facilitate solver benchmarking (#769).</li><li>A <code>submit</code> function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).</li><li>The field of <code>ObjectiveValue</code> was renamed to <code>result_index</code> (#729).</li><li>The <code>_constant</code> and <code>Utilities.getconstant</code> function were renamed to <code>constant</code></li><li><code>REDUCTION_CERTIFICATE</code> result status was added (#734).</li><li>Abstract matrix sets were added (#731).</li><li>Testing improvements:<ul><li>The testing guideline was updated (#728).</li><li>Quadratic tests were added (#697).</li><li>Unit tests for <code>RawStatusString</code>, <code>SolveTime</code>, <code>Silent</code> and <code>SolverName</code> were added (#726, #741).</li><li>A rotated second-order cone test was added (#759).</li><li>A power cone test was added (#768).</li><li>Tests for <code>ZeroOne</code> variables with variable bounds were added (#772).</li><li>An unbounded test was added (#773).</li><li>Existing tests had a few updates (#702, #703, #763).</li></ul></li><li>Documentation improvements:<ul><li>Added a section on <code>CachingOptimizer</code> (#777).</li><li>Added a section on <code>UniversalFallback</code>, <code>Model</code> and <code>@model</code> (#762).</li><li>Transition the knapsack example to a doctest with MockOptimizer (#786).</li></ul></li><li>Utilities improvements:<ul><li>A <code>CleverDict</code> utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).</li><li>The <code>Utilities.constant</code> function was renamed to <code>Utilities.constant_vector</code> (#740).</li><li>Implement optimizer attributes for CachingOptimizer (#745).</li><li>Rename <code>Utilities.add_scalar_constraint</code> to <code>Utilities.normalize_and_add_constraint</code> (#801).</li><li><code>operate</code> with <code>vcat</code>, <code>SingleVariable</code> and <code>VectorOfVariables</code> now returns a <code>VectorOfVariables</code> (#616).</li><li>Fix a type piracy of <code>operate</code> (#784).</li><li>The <code>load_constraint</code> fallback signature was fixed (#760).</li><li>The <code>set_dot</code> function was extended to work with sparse arrays (#805).</li></ul></li><li>Bridges improvements:<ul><li>The bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement <code>ConstraintFunction</code> and <code>ConstraintSet</code> if the user wants to recover them. As a consequence, the <code>@bridge</code> macro was removed (#722).</li><li>Bridge are now instantiated with a <code>bridge_constraint</code> function instead of using a constructor (#730).</li><li>Fix constraint attributes for bridges (#699).</li><li>Constraint bridges were moved to the <code>Bridges/Constraint</code> submodule so they should now inherit from <code>MOI.Bridges.Constraint.Abstract</code> and should implement <code>MOI.Bridges.Constraint.concrete_bridge_type</code> instead of <code>MOI.Bridges.concrete_bridge_type</code> (#756).</li><li>Variable bridges were added in (#759).</li><li>Various improvements (#746, #747).</li></ul></li></ul><h2 id="v0.8.4-(March-13,-2019)"><a class="docs-heading-anchor" href="#v0.8.4-(March-13,-2019)">v0.8.4 (March 13, 2019)</a><a id="v0.8.4-(March-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.4-(March-13,-2019)" title="Permalink"></a></h2><ul><li>Performance improvement in <code>default_copy_to</code> and bridge optimizer (#696).</li><li>Add <code>Silent</code> and implement setting optimizer attributes in caching and mock optimizers (#695).</li><li>Add <code>Functionize</code> bridges (SingleVariable and VectorOfVariables) (#659).</li><li>Minor typo fixes (#694).</li></ul><h2 id="v0.8.3-(March-6,-2019)"><a class="docs-heading-anchor" href="#v0.8.3-(March-6,-2019)">v0.8.3 (March 6, 2019)</a><a id="v0.8.3-(March-6,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.3-(March-6,-2019)" title="Permalink"></a></h2><ul><li>Use zero constant in scalar constraint function of <code>MOI.Test.copytest</code> (#691).</li><li>Fix variable deletion with <code>SingleVariable</code> objective function (#690).</li><li>Fix <code>LazyBridgeOptimizer</code> with bridges that add no constraints (#689).</li><li>Error message improvements (#673, #685, #686, #688).</li><li>Documentation improvements (#682, #683, #687).</li><li>Basis status:<ul><li>Remove <code>VariableBasisStatus</code> (#679).</li><li>Test <code>ConstraintBasisStatus</code> and implement it in bridges (#678).</li></ul></li><li>Fix inference of <code>NumberOfVariables</code> and <code>NumberOfConstraints</code> (#677).</li><li>Implement division between a quadratic function and a number (#675).</li></ul><h2 id="v0.8.2-(February-7,-2019)"><a class="docs-heading-anchor" href="#v0.8.2-(February-7,-2019)">v0.8.2 (February 7, 2019)</a><a id="v0.8.2-(February-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.2-(February-7,-2019)" title="Permalink"></a></h2><ul><li>Add <code>RawStatusString</code> attribute (#629).</li><li>Do not set names to the optimizer but only to the cache in <code>CachingOptimizer</code> (#638).</li><li>Make scalar MOI functions act as scalars in broadcast (#646).</li><li>Add function utilities:<ul><li>Implement <code>Base.zero</code> (#634), <code>Base.iszero</code> (#643), add missing arithmetic operations (#644, #645) and fix division (#648).</li><li>Add a <code>vectorize</code> function that turns a vector of <code>ScalarAffineFunction</code> into a <code>VectorAffineFunction</code> (#642).</li></ul></li><li>Improve support for starting values:<ul><li>Show a warning in copy when starting values are not supported instead of throwing an error (#630).</li><li>Fix <code>UniversalFallback</code> for getting an variable or constraint attribute set to no indices (#623).</li><li>Add a test in <code>contlineartest</code> with partially set <code>VariablePrimalStart</code>.</li></ul></li><li>Bridges improvements:<ul><li>Fix <code>StackOverFlow</code> in <code>LazyBridgeOptimizer</code> when there is a cycle in the graph of bridges.</li><li>Add <code>Slack</code> bridges (#610, #650).</li><li>Add <code>FlipSign</code> bridges (#658).</li></ul></li><li>Add tests with duplicate coefficients in <code>ScalarAffineFunction</code> and <code>VectorAffineFunction</code> (#639).</li><li>Use tolerance to compare <code>VariablePrimal</code> in <code>rotatedsoc1</code> test (#632).</li><li>Use a zero constant in <code>ScalarAffineFunction</code> of constraints in <code>psdt2</code> (#622).</li></ul><h2 id="v0.8.1-(January-7,-2019)"><a class="docs-heading-anchor" href="#v0.8.1-(January-7,-2019)">v0.8.1 (January 7, 2019)</a><a id="v0.8.1-(January-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.1-(January-7,-2019)" title="Permalink"></a></h2><ul><li>Adding an NLP objective now overrides any objective set using the <code>ObjectiveFunction</code> attribute (#619).</li><li>Rename <code>fullbridgeoptimizer</code> into <code>full_bridge_optimizer</code> (#621).</li><li>Allow custom constraint types with <code>full_bridge_optimizer</code> (#617).</li><li>Add <code>Vectorize</code> bridge which transforms scalar linear constraints into vector linear constraints (#615).</li></ul><h2 id="v0.8.0-(December-18,-2018)"><a class="docs-heading-anchor" href="#v0.8.0-(December-18,-2018)">v0.8.0 (December 18, 2018)</a><a id="v0.8.0-(December-18,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.8.0-(December-18,-2018)" title="Permalink"></a></h2><ul><li>Rename all enum values to follow the JuMP naming guidelines for constants, for example, <code>Optimal</code> becomes <code>OPTIMAL</code>, and <code>DualInfeasible</code> becomes <code>DUAL_INFEASIBLE</code>.</li><li>Rename CachingOptimizer methods for style compliance.</li><li>Add an <code>MOI.TerminationStatusCode</code> called <code>ALMOST_DUAL_INFEASIBLE</code>.</li></ul><h2 id="v0.7.0-(December-13,-2018)"><a class="docs-heading-anchor" href="#v0.7.0-(December-13,-2018)">v0.7.0 (December 13, 2018)</a><a id="v0.7.0-(December-13,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.7.0-(December-13,-2018)" title="Permalink"></a></h2><ul><li>Test that <code>MOI.TerminationStatus</code> is <code>MOI.OptimizeNotCalled</code> before <code>MOI.optimize!</code> is called.</li><li>Check <code>supports_default_copy_to</code> in tests (#594).</li><li>Key pieces of information like optimality, infeasibility, etc., are now reported through <code>TerminationStatusCode</code>. It is typically no longer necessary to check the result statuses in addition to the termination status.</li><li>Add perspective dimension to log-det cone (#593).</li></ul><h2 id="v0.6.4-(November-27,-2018)"><a class="docs-heading-anchor" href="#v0.6.4-(November-27,-2018)">v0.6.4 (November 27, 2018)</a><a id="v0.6.4-(November-27,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.4-(November-27,-2018)" title="Permalink"></a></h2><ul><li>Add <code>OptimizeNotCalled</code> termination status (#577) and improve documentation of other statuses (#575).</li><li>Add a solver naming guideline (#578).</li><li>Make <code>FeasibilitySense</code> the default <code>ObjectiveSense</code> (#579).</li><li>Fix <code>Utilities.@model</code> and <code>Bridges.@bridge</code> macros for functions and sets defined outside MOI (#582).</li><li>Document solver-specific attributes (#580) and implement them in <code>Utilities.CachingOptimizer</code> (#565).</li></ul><h2 id="v0.6.3-(November-16,-2018)"><a class="docs-heading-anchor" href="#v0.6.3-(November-16,-2018)">v0.6.3 (November 16, 2018)</a><a id="v0.6.3-(November-16,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.3-(November-16,-2018)" title="Permalink"></a></h2><ul><li>Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)</li><li>Attributes may now be partially set (some values could be <code>nothing</code>). (#563)</li><li>Performance improvements in Utilities.Model (#549, #567, #568)</li><li>Fix bug in QuadtoSOC (#558).</li><li>New <code>supports_default_copy_to</code> method that optimizers should implement to control caching behavior.</li><li>Documentation improvements.</li></ul><h2 id="v0.6.2-(October-26,-2018)"><a class="docs-heading-anchor" href="#v0.6.2-(October-26,-2018)">v0.6.2 (October 26, 2018)</a><a id="v0.6.2-(October-26,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.2-(October-26,-2018)" title="Permalink"></a></h2><ul><li>Improve hygiene of <code>@model</code> macro (#544).</li><li>Fix bug in copy tests (#543).</li><li>Fix bug in UniversalFallback attribute getter (#540).</li><li>Allow all correct solutions for <code>solve_blank_obj</code> unit test (#537).</li><li>Add errors for Allocate-Load and bad constraints (#534).</li><li>[performance] Add specialized implementation of <code>hash</code> for <code>VariableIndex</code> (#533).</li><li>[performance] Construct the name to object dictionaries lazily in model (#535).</li><li>Add the <code>QuadtoSOC</code> bridge which transforms <code>ScalarQuadraticFunction</code> constraints into <code>RotatedSecondOrderCone</code> (#483).</li></ul><h2 id="v0.6.1-(September-22,-2018)"><a class="docs-heading-anchor" href="#v0.6.1-(September-22,-2018)">v0.6.1 (September 22, 2018)</a><a id="v0.6.1-(September-22,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.1-(September-22,-2018)" title="Permalink"></a></h2><ul><li>Enable <code>PositiveSemidefiniteConeSquare</code> set and quadratic functions in <code>MOIB.fullbridgeoptimizer</code> (#524).</li><li>Add warning in the bridge between <code>PositiveSemidefiniteConeSquare</code> and <code>PositiveSemidefiniteConeTriangle</code> when the matrix is almost symmetric (#522).</li><li>Modify <code>MOIT.copytest</code> to not add multiples constraints on the same variable (#521).</li><li>Add missing keyword argument in one of <code>MOIU.add_scalar_constraint</code> methods (#520).</li></ul><h2 id="v0.6.0-(August-30,-2018)"><a class="docs-heading-anchor" href="#v0.6.0-(August-30,-2018)">v0.6.0 (August 30, 2018)</a><a id="v0.6.0-(August-30,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.6.0-(August-30,-2018)" title="Permalink"></a></h2><ul><li>The <code>MOIU.@model</code> and <code>MOIB.@bridge</code> macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.</li><li>Rename functions according to the <a href="https://jump.dev/JuMP.jl/stable/developers/style/">JuMP style guide</a>:<ul><li><code>copy!</code> with keyword arguments <code>copynames</code> and <code>warnattributes</code> -&gt; <code>copy_to</code> with keyword arguments <code>copy_names</code> and <code>warn_attributes</code>;</li><li><code>set!</code> -&gt; <code>set</code>;</li><li><code>addvariable[s]!</code> -&gt; <code>add_variable[s]</code>;</li><li><code>supportsconstraint</code> -&gt; <code>supports_constraint</code>;</li><li><code>addconstraint[s]!</code> -&gt; <code>add_constraint[s]</code>;</li><li><code>isvalid</code> -&gt; <code>is_valid</code>;</li><li><code>isempty</code> -&gt; <code>is_empty</code>;</li><li><code>Base.delete!</code> -&gt; <code>delete</code>;</li><li><code>modify!</code> -&gt; <code>modify</code>;</li><li><code>transform!</code> -&gt; <code>transform</code>;</li><li><code>initialize!</code> -&gt; <code>initialize</code>;</li><li><code>write</code> -&gt; <code>write_to_file</code>; and</li><li><code>read!</code> -&gt; <code>read_from_file</code>.</li></ul></li><li>Remove <code>free!</code> (use <code>Base.finalize</code> instead).</li><li>Add the <code>SquarePSD</code> bridge which transforms <code>PositiveSemidefiniteConeTriangle</code> constraints into <code>PositiveSemidefiniteConeTriangle</code>.</li><li>Add result fallback for <code>ConstraintDual</code> of variable-wise constraint, <code>ConstraintPrimal</code> and <code>ObjectiveValue</code>.</li><li>Add tests for <code>ObjectiveBound</code>.</li><li>Add test for empty rows in vector linear constraint.</li><li>Rework errors: <code>CannotError</code> has been renamed <code>NotAllowedError</code> and the distinction between <code>UnsupportedError</code> and <code>NotAllowedError</code> is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)</li><li><code>canget</code> is removed. <code>NoSolution</code> is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).</li></ul><h2 id="v0.5.0-(August-5,-2018)"><a class="docs-heading-anchor" href="#v0.5.0-(August-5,-2018)">v0.5.0 (August 5, 2018)</a><a id="v0.5.0-(August-5,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.5.0-(August-5,-2018)" title="Permalink"></a></h2><ul><li>Fix names with CachingOptimizer.</li><li>Cleanup thanks to @mohamed82008.</li><li>Added a universal fallback for constraints.</li><li>Fast utilities for function canonicalization thanks to @rdeits.</li><li>Renamed <code>dimension</code> field to <code>side_dimension</code> in the context of matrix-like sets.</li><li>New and improved tests for cases like duplicate terms and <code>ObjectiveBound</code>.</li><li>Removed <code>cantransform</code>, <code>canaddconstraint</code>, <code>canaddvariable</code>, <code>canset</code>, <code>canmodify</code>, and <code>candelete</code> functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of <code>UnsupportedError</code> indicate unsupported operations, while subtypes of <code>CannotError</code> indicate operations that cannot be performed in the current state.</li><li>The API for <code>copy!</code> is updated to remove the CopyResult type.</li><li>Updates for the new JuMP style guide.</li></ul><h2 id="v0.4.1-(June-28,-2018)"><a class="docs-heading-anchor" href="#v0.4.1-(June-28,-2018)">v0.4.1 (June 28, 2018)</a><a id="v0.4.1-(June-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.4.1-(June-28,-2018)" title="Permalink"></a></h2><ul><li>Fixes vector function modification on 32 bits.</li><li>Fixes Bellman-Ford algorithm for bridges.</li><li>Added an NLP test with <code>FeasibilitySense</code>.</li><li>Update modification documentation.</li></ul><h2 id="v0.4.0-(June-23,-2018)"><a class="docs-heading-anchor" href="#v0.4.0-(June-23,-2018)">v0.4.0 (June 23, 2018)</a><a id="v0.4.0-(June-23,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.4.0-(June-23,-2018)" title="Permalink"></a></h2><ul><li>Helper constructors for <code>VectorAffineTerm</code> and <code>VectorQuadraticTerm</code>.</li><li>Added <code>modify_lhs</code> to <code>TestConfig</code>.</li><li>Additional unit tests for optimizers.</li><li>Added a type parameter to <code>CachingOptimizer</code> for the <code>optimizer</code> field.</li><li>New API for problem modification (#388)</li><li>Tests pass without deprecation warnings on Julia 0.7.</li><li>Small fixes and documentation updates.</li></ul><h2 id="v0.3.0-(May-25,-2018)"><a class="docs-heading-anchor" href="#v0.3.0-(May-25,-2018)">v0.3.0 (May 25, 2018)</a><a id="v0.3.0-(May-25,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.3.0-(May-25,-2018)" title="Permalink"></a></h2><ul><li>Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.</li><li>Improvements to <code>MockOptimizer</code>.</li><li>Significant changes to <code>Bridges</code>.</li><li>New and improved unit tests.</li><li>Fixes for Julia 0.7.</li></ul><h2 id="v0.2.0-(April-24,-2018)"><a class="docs-heading-anchor" href="#v0.2.0-(April-24,-2018)">v0.2.0 (April 24, 2018)</a><a id="v0.2.0-(April-24,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.2.0-(April-24,-2018)" title="Permalink"></a></h2><ul><li>Improvements to and better coverage of <code>Tests</code>.</li><li>Documentation fixes.</li><li><code>SolverName</code> attribute.</li><li>Changes to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).</li><li>Addition of NLP tests.</li><li>Introduction of <code>UniversalFallback</code>.</li><li><code>copynames</code> keyword argument to <code>MOI.copy!</code>.</li><li>Add Bridges submodule.</li></ul><h2 id="v0.1.0-(February-28,-2018)"><a class="docs-heading-anchor" href="#v0.1.0-(February-28,-2018)">v0.1.0 (February 28, 2018)</a><a id="v0.1.0-(February-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#v0.1.0-(February-28,-2018)" title="Permalink"></a></h2><ul><li>Initial public release.</li><li>The framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.</li></ul></article><nav class="docs-footer"><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/developer/checklists/index.html b/previews/PR3562/moi/developer/checklists/index.html
index 5b2208e3fd6..10c4768102f 100644
--- a/previews/PR3562/moi/developer/checklists/index.html
+++ b/previews/PR3562/moi/developer/checklists/index.html
@@ -113,4 +113,4 @@
 
 ## Documentation
 
- - [ ] The version fields are updated in `docs/src/submodules/FileFormats/overview.md`</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../submodules/Test/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../../release_notes/">Release notes »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ - [ ] The version fields are updated in `docs/src/submodules/FileFormats/overview.md`</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../submodules/Test/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../../release_notes/">Release notes »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/index.html b/previews/PR3562/moi/index.html
index c0c3d0768bf..7e97e7d4916 100644
--- a/previews/PR3562/moi/index.html
+++ b/previews/PR3562/moi/index.html
@@ -10,4 +10,4 @@
     year={2021},
     doi={10.1287/ijoc.2021.1067},
     publisher={INFORMS}
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../extensions/DimensionalData/">« rafaqz/DimensionalData.jl</a><a class="docs-footer-nextpage" href="background/motivation/">Motivation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../extensions/DimensionalData/">« rafaqz/DimensionalData.jl</a><a class="docs-footer-nextpage" href="background/motivation/">Motivation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/constraints/index.html b/previews/PR3562/moi/manual/constraints/index.html
index 2786bccefb3..0df8c8a72d0 100644
--- a/previews/PR3562/moi/manual/constraints/index.html
+++ b/previews/PR3562/moi/manual/constraints/index.html
@@ -23,4 +23,4 @@
 false</code></pre><h2 id="Constraint-attributes"><a class="docs-heading-anchor" href="#Constraint-attributes">Constraint attributes</a><a id="Constraint-attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Constraint-attributes" title="Permalink"></a></h2><p>The following attributes are available for constraints:</p><ul><li><a href="../../reference/constraints/#MathOptInterface.ConstraintName"><code>ConstraintName</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>ConstraintPrimalStart</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintDualStart"><code>ConstraintDualStart</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintPrimal"><code>ConstraintPrimal</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintDual"><code>ConstraintDual</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintBasisStatus"><code>ConstraintBasisStatus</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.CanonicalConstraintFunction"><code>CanonicalConstraintFunction</code></a></li><li><a href="../../reference/constraints/#MathOptInterface.ConstraintSet"><code>ConstraintSet</code></a></li></ul><p>Get and set these attributes using <a href="../../reference/models/#MathOptInterface.get"><code>get</code></a> and <a href="../../reference/models/#MathOptInterface.set"><code>set</code></a>.</p><pre><code class="language-julia-repl hljs">julia&gt; MOI.set(model, MOI.ConstraintName(), c, &quot;con_c&quot;)
 
 julia&gt; MOI.get(model, MOI.ConstraintName(), c)
-&quot;con_c&quot;</code></pre><h2 id="Constraints-by-function-set-pairs"><a class="docs-heading-anchor" href="#Constraints-by-function-set-pairs">Constraints by function-set pairs</a><a id="Constraints-by-function-set-pairs-1"></a><a class="docs-heading-anchor-permalink" href="#Constraints-by-function-set-pairs" title="Permalink"></a></h2><p>Below is a list of common constraint types and how they are represented as function-set pairs in MOI. In the notation below, <span>$x$</span> is a vector of decision variables, <span>$x_i$</span> is a scalar decision variable, <span>$\alpha, \beta$</span> are scalar constants, <span>$a, b$</span> are constant vectors, <code>A</code> is a constant matrix and <span>$\mathbb{R}_+$</span> (resp. <span>$\mathbb{R}_-$</span>) is the set of non-negative (resp. non-positive) real numbers.</p><h3 id="Linear-constraints"><a class="docs-heading-anchor" href="#Linear-constraints">Linear constraints</a><a id="Linear-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Linear-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$a^Tx \le \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$a^Tx \ge \alpha$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$a^Tx = \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right"><span>$\alpha \le a^Tx \le \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>Interval</code></td></tr><tr><td style="text-align: right"><span>$x_i \le \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$x_i \ge \alpha$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$x_i = \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right"><span>$\alpha \le x_i \le \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Interval</code></td></tr><tr><td style="text-align: right"><span>$Ax + b \in \mathbb{R}_+^n$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Nonnegatives</code></td></tr><tr><td style="text-align: right"><span>$Ax + b \in \mathbb{R}_-^n$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Nonpositives</code></td></tr><tr><td style="text-align: right"><span>$Ax + b = 0$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Zeros</code></td></tr></table><p>By convention, solvers are not expected to support nonzero constant terms in the <a href="../../reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a>s the first four rows of the preceding table because they are redundant with the parameters of the sets. For example, encode <span>$2x + 1 \le 2$</span> as <span>$2x \le 1$</span>.</p><p>Constraints with <a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> in <a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan</code></a>, <a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan</code></a>, <a href="../../reference/standard_form/#MathOptInterface.EqualTo"><code>EqualTo</code></a>, or <a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval</code></a> sets have a natural interpretation as variable bounds. As such, it is typically not natural to impose multiple lower- or upper-bounds on the same variable, and the solver interfaces will throw respectively <a href="../../reference/errors/#MathOptInterface.LowerBoundAlreadySet"><code>LowerBoundAlreadySet</code></a> or <a href="../../reference/errors/#MathOptInterface.UpperBoundAlreadySet"><code>UpperBoundAlreadySet</code></a>.</p><p>Moreover, adding two <a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraints on the same variable with the same set is impossible because they share the same index as it is the index of the variable, see <a href="../../reference/constraints/#MathOptInterface.ConstraintIndex"><code>ConstraintIndex</code></a>.</p><p>It is natural, however, to impose upper- and lower-bounds separately as two different constraints on a single variable. The difference between imposing bounds by using a single <a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval</code></a> constraint and by using separate <a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan</code></a> and <a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan</code></a> constraints is that the latter will allow the solver to return separate dual multipliers for the two bounds, while the former will allow the solver to return only a single dual for the interval constraint.</p><h3 id="Conic-constraints"><a class="docs-heading-anchor" href="#Conic-constraints">Conic constraints</a><a id="Conic-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Conic-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$\lVert Ax + b\rVert_2 \le c^Tx + d$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>SecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$y \ge \lVert x \rVert_2$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$2yz \ge \lVert x \rVert_2^2, y,z \ge 0$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>RotatedSecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$(a_1^Tx + b_1,a_2^Tx + b_2,a_3^Tx + b_3) \in \mathcal{E}$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>ExponentialCone</code></td></tr><tr><td style="text-align: right"><span>$A(x) \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeTriangle</code></td></tr><tr><td style="text-align: right"><span>$B(x) \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeSquare</code></td></tr><tr><td style="text-align: right"><span>$x \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeTriangle</code></td></tr><tr><td style="text-align: right"><span>$x \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeSquare</code></td></tr></table><p>where <span>$\mathcal{E}$</span> is the exponential cone (see <a href="../../reference/standard_form/#MathOptInterface.ExponentialCone"><code>ExponentialCone</code></a>), <span>$\mathcal{S}_+$</span> is the set of positive semidefinite symmetric matrices, <span>$A$</span> is an affine map that outputs symmetric matrices and <span>$B$</span> is an affine map that outputs square matrices.</p><h3 id="Quadratic-constraints"><a class="docs-heading-anchor" href="#Quadratic-constraints">Quadratic constraints</a><a id="Quadratic-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Quadratic-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b \ge 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b \le 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b = 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right">Bilinear matrix inequality</td><td style="text-align: right"><code>VectorQuadraticFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteCone...</code></td></tr></table><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>For more details on the internal format of the quadratic functions see <a href="../../reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a> or <a href="../../reference/standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a>.</p></div></div><h3 id="Discrete-and-logical-constraints"><a class="docs-heading-anchor" href="#Discrete-and-logical-constraints">Discrete and logical constraints</a><a id="Discrete-and-logical-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Discrete-and-logical-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$x_i \in \mathbb{Z}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Integer</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0,1\}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>ZeroOne</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0\} \cup [l,u]$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Semicontinuous</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0\} \cup \{l,l+1,\ldots,u-1,u\}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Semiinteger</code></td></tr><tr><td style="text-align: right">At most one component of <span>$x$</span> can be nonzero</td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SOS1</code></td></tr><tr><td style="text-align: right">At most two components of <span>$x$</span> can be nonzero, and if so they must be adjacent components</td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SOS2</code></td></tr><tr><td style="text-align: right"><span>$y = 1 \implies a^T x \in S$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Indicator</code></td></tr></table><h2 id="JuMP-mapping"><a class="docs-heading-anchor" href="#JuMP-mapping">JuMP mapping</a><a id="JuMP-mapping-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-mapping" title="Permalink"></a></h2><p>The following bullet points show examples of how JuMP constraints are translated into MOI function-set pairs:</p><ul><li><code>@constraint(m, 2x + y &lt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>LessThan</code></li><li><code>@constraint(m, 2x + y &gt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>GreaterThan</code></li><li><code>@constraint(m, 2x + y == 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>EqualTo</code></li><li><code>@constraint(m, 0 &lt;= 2x + y &lt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>Interval</code></li><li><code>@constraint(m, 2x + y in ArbitrarySet())</code> becomes <code>ScalarAffineFunction</code>-in-<code>ArbitrarySet</code>.</li></ul><p>Variable bounds are handled in a similar fashion:</p><ul><li><code>@variable(m, x &lt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>LessThan</code></li><li><code>@variable(m, x &gt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>GreaterThan</code></li></ul><p>One notable difference is that a variable with an upper and lower bound is translated into two constraints, rather than an interval, that is:</p><ul><li><code>@variable(m, 0 &lt;= x &lt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>LessThan</code> <em>and</em>  <code>VariableIndex</code>-in-<code>GreaterThan</code>.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../solutions/">Solutions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+&quot;con_c&quot;</code></pre><h2 id="Constraints-by-function-set-pairs"><a class="docs-heading-anchor" href="#Constraints-by-function-set-pairs">Constraints by function-set pairs</a><a id="Constraints-by-function-set-pairs-1"></a><a class="docs-heading-anchor-permalink" href="#Constraints-by-function-set-pairs" title="Permalink"></a></h2><p>Below is a list of common constraint types and how they are represented as function-set pairs in MOI. In the notation below, <span>$x$</span> is a vector of decision variables, <span>$x_i$</span> is a scalar decision variable, <span>$\alpha, \beta$</span> are scalar constants, <span>$a, b$</span> are constant vectors, <code>A</code> is a constant matrix and <span>$\mathbb{R}_+$</span> (resp. <span>$\mathbb{R}_-$</span>) is the set of non-negative (resp. non-positive) real numbers.</p><h3 id="Linear-constraints"><a class="docs-heading-anchor" href="#Linear-constraints">Linear constraints</a><a id="Linear-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Linear-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$a^Tx \le \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$a^Tx \ge \alpha$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$a^Tx = \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right"><span>$\alpha \le a^Tx \le \beta$</span></td><td style="text-align: right"><code>ScalarAffineFunction</code></td><td style="text-align: right"><code>Interval</code></td></tr><tr><td style="text-align: right"><span>$x_i \le \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$x_i \ge \alpha$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$x_i = \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right"><span>$\alpha \le x_i \le \beta$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Interval</code></td></tr><tr><td style="text-align: right"><span>$Ax + b \in \mathbb{R}_+^n$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Nonnegatives</code></td></tr><tr><td style="text-align: right"><span>$Ax + b \in \mathbb{R}_-^n$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Nonpositives</code></td></tr><tr><td style="text-align: right"><span>$Ax + b = 0$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Zeros</code></td></tr></table><p>By convention, solvers are not expected to support nonzero constant terms in the <a href="../../reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a>s the first four rows of the preceding table because they are redundant with the parameters of the sets. For example, encode <span>$2x + 1 \le 2$</span> as <span>$2x \le 1$</span>.</p><p>Constraints with <a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> in <a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan</code></a>, <a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan</code></a>, <a href="../../reference/standard_form/#MathOptInterface.EqualTo"><code>EqualTo</code></a>, or <a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval</code></a> sets have a natural interpretation as variable bounds. As such, it is typically not natural to impose multiple lower- or upper-bounds on the same variable, and the solver interfaces will throw respectively <a href="../../reference/errors/#MathOptInterface.LowerBoundAlreadySet"><code>LowerBoundAlreadySet</code></a> or <a href="../../reference/errors/#MathOptInterface.UpperBoundAlreadySet"><code>UpperBoundAlreadySet</code></a>.</p><p>Moreover, adding two <a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraints on the same variable with the same set is impossible because they share the same index as it is the index of the variable, see <a href="../../reference/constraints/#MathOptInterface.ConstraintIndex"><code>ConstraintIndex</code></a>.</p><p>It is natural, however, to impose upper- and lower-bounds separately as two different constraints on a single variable. The difference between imposing bounds by using a single <a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval</code></a> constraint and by using separate <a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan</code></a> and <a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan</code></a> constraints is that the latter will allow the solver to return separate dual multipliers for the two bounds, while the former will allow the solver to return only a single dual for the interval constraint.</p><h3 id="Conic-constraints"><a class="docs-heading-anchor" href="#Conic-constraints">Conic constraints</a><a id="Conic-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Conic-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$\lVert Ax + b\rVert_2 \le c^Tx + d$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>SecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$y \ge \lVert x \rVert_2$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$2yz \ge \lVert x \rVert_2^2, y,z \ge 0$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>RotatedSecondOrderCone</code></td></tr><tr><td style="text-align: right"><span>$(a_1^Tx + b_1,a_2^Tx + b_2,a_3^Tx + b_3) \in \mathcal{E}$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>ExponentialCone</code></td></tr><tr><td style="text-align: right"><span>$A(x) \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeTriangle</code></td></tr><tr><td style="text-align: right"><span>$B(x) \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeSquare</code></td></tr><tr><td style="text-align: right"><span>$x \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeTriangle</code></td></tr><tr><td style="text-align: right"><span>$x \in \mathcal{S}_+$</span></td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>PositiveSemidefiniteConeSquare</code></td></tr></table><p>where <span>$\mathcal{E}$</span> is the exponential cone (see <a href="../../reference/standard_form/#MathOptInterface.ExponentialCone"><code>ExponentialCone</code></a>), <span>$\mathcal{S}_+$</span> is the set of positive semidefinite symmetric matrices, <span>$A$</span> is an affine map that outputs symmetric matrices and <span>$B$</span> is an affine map that outputs square matrices.</p><h3 id="Quadratic-constraints"><a class="docs-heading-anchor" href="#Quadratic-constraints">Quadratic constraints</a><a id="Quadratic-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Quadratic-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b \ge 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>GreaterThan</code></td></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b \le 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>LessThan</code></td></tr><tr><td style="text-align: right"><span>$\frac{1}{2}x^TQx + a^Tx + b = 0$</span></td><td style="text-align: right"><code>ScalarQuadraticFunction</code></td><td style="text-align: right"><code>EqualTo</code></td></tr><tr><td style="text-align: right">Bilinear matrix inequality</td><td style="text-align: right"><code>VectorQuadraticFunction</code></td><td style="text-align: right"><code>PositiveSemidefiniteCone...</code></td></tr></table><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>For more details on the internal format of the quadratic functions see <a href="../../reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a> or <a href="../../reference/standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a>.</p></div></div><h3 id="Discrete-and-logical-constraints"><a class="docs-heading-anchor" href="#Discrete-and-logical-constraints">Discrete and logical constraints</a><a id="Discrete-and-logical-constraints-1"></a><a class="docs-heading-anchor-permalink" href="#Discrete-and-logical-constraints" title="Permalink"></a></h3><table><tr><th style="text-align: right">Mathematical Constraint</th><th style="text-align: right">MOI Function</th><th style="text-align: right">MOI Set</th></tr><tr><td style="text-align: right"><span>$x_i \in \mathbb{Z}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Integer</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0,1\}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>ZeroOne</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0\} \cup [l,u]$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Semicontinuous</code></td></tr><tr><td style="text-align: right"><span>$x_i \in \{0\} \cup \{l,l+1,\ldots,u-1,u\}$</span></td><td style="text-align: right"><code>VariableIndex</code></td><td style="text-align: right"><code>Semiinteger</code></td></tr><tr><td style="text-align: right">At most one component of <span>$x$</span> can be nonzero</td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SOS1</code></td></tr><tr><td style="text-align: right">At most two components of <span>$x$</span> can be nonzero, and if so they must be adjacent components</td><td style="text-align: right"><code>VectorOfVariables</code></td><td style="text-align: right"><code>SOS2</code></td></tr><tr><td style="text-align: right"><span>$y = 1 \implies a^T x \in S$</span></td><td style="text-align: right"><code>VectorAffineFunction</code></td><td style="text-align: right"><code>Indicator</code></td></tr></table><h2 id="JuMP-mapping"><a class="docs-heading-anchor" href="#JuMP-mapping">JuMP mapping</a><a id="JuMP-mapping-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-mapping" title="Permalink"></a></h2><p>The following bullet points show examples of how JuMP constraints are translated into MOI function-set pairs:</p><ul><li><code>@constraint(m, 2x + y &lt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>LessThan</code></li><li><code>@constraint(m, 2x + y &gt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>GreaterThan</code></li><li><code>@constraint(m, 2x + y == 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>EqualTo</code></li><li><code>@constraint(m, 0 &lt;= 2x + y &lt;= 10)</code> becomes <code>ScalarAffineFunction</code>-in-<code>Interval</code></li><li><code>@constraint(m, 2x + y in ArbitrarySet())</code> becomes <code>ScalarAffineFunction</code>-in-<code>ArbitrarySet</code>.</li></ul><p>Variable bounds are handled in a similar fashion:</p><ul><li><code>@variable(m, x &lt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>LessThan</code></li><li><code>@variable(m, x &gt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>GreaterThan</code></li></ul><p>One notable difference is that a variable with an upper and lower bound is translated into two constraints, rather than an interval, that is:</p><ul><li><code>@variable(m, 0 &lt;= x &lt;= 1)</code> becomes <code>VariableIndex</code>-in-<code>LessThan</code> <em>and</em>  <code>VariableIndex</code>-in-<code>GreaterThan</code>.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../solutions/">Solutions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/models/index.html b/previews/PR3562/moi/manual/models/index.html
index 66c75d69ed9..7517cc848e3 100644
--- a/previews/PR3562/moi/manual/models/index.html
+++ b/previews/PR3562/moi/manual/models/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox" checked/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../standard_form/">Standard form problem</a></li><li class="is-active"><a class="tocitem" href>Models</a><ul class="internal"><li><a class="tocitem" href="#Attributes"><span>Attributes</span></a></li><li><a class="tocitem" href="#ModelLike-API"><span>ModelLike API</span></a></li><li><a class="tocitem" href="#AbstractOptimizer-API"><span>AbstractOptimizer API</span></a></li></ul></li><li><a class="tocitem" href="../variables/">Variables</a></li><li><a class="tocitem" href="../constraints/">Constraints</a></li><li><a class="tocitem" href="../solutions/">Solutions</a></li><li><a class="tocitem" href="../modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/duality/">Duality</a></li><li><a class="tocitem" href="../../background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Manual</a></li><li class="is-active"><a href>Models</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Models</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/models.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Models"><a class="docs-heading-anchor" href="#Models">Models</a><a id="Models-1"></a><a class="docs-heading-anchor-permalink" href="#Models" title="Permalink"></a></h1><p>The most significant part of MOI is the definition of the <strong>model API</strong> that is used to specify an instance of an optimization problem (for example, by adding variables and constraints). Objects that implement the model API must inherit from the <a href="../../reference/models/#MathOptInterface.ModelLike"><code>ModelLike</code></a> abstract type.</p><p>Notably missing from the model API is the method to solve an optimization problem. <code>ModelLike</code> objects may store an instance (for example, in memory or backed by a file format) without being linked to a particular solver. In addition to the model API, MOI defines <a href="../../reference/models/#MathOptInterface.AbstractOptimizer"><code>AbstractOptimizer</code></a> and provides methods to solve the model and interact with solutions. See the <a href="../solutions/#manual_solutions">Solutions</a> section for more details.</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>Throughout the rest of the manual, <code>model</code> is used as a generic <code>ModelLike</code>, and <code>optimizer</code> is used as a generic <code>AbstractOptimizer</code>.</p></div></div><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>MOI does not export functions, but for brevity we often omit qualifying names with the MOI module. Best practice is to have</p><pre><code class="language-julia hljs">import MathOptInterface as MOI</code></pre><p>and prefix all MOI methods with <code>MOI.</code> in user code. If a name is also available in base Julia, we always explicitly use the module prefix, for example, with <code>MOI.get</code>.</p></div></div><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><p>Attributes are properties of the model that can be queried and modified. These include constants such as the number of variables in a model <a href="../../reference/models/#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a>), and properties of variables and constraints such as the name of a variable (<a href="../../reference/variables/#MathOptInterface.VariableName"><code>VariableName</code></a>).</p><p>There are four types of attributes:</p><ul><li>Model attributes (subtypes of <a href="../../reference/models/#MathOptInterface.AbstractModelAttribute"><code>AbstractModelAttribute</code></a>) refer to properties of a model.</li><li>Optimizer attributes (subtypes of <a href="../../reference/models/#MathOptInterface.AbstractOptimizerAttribute"><code>AbstractOptimizerAttribute</code></a>) refer to properties of an optimizer.</li><li>Constraint attributes (subtypes of <a href="../../reference/constraints/#MathOptInterface.AbstractConstraintAttribute"><code>AbstractConstraintAttribute</code></a>) refer to properties of an individual constraint.</li><li>Variable attributes (subtypes of <a href="../../reference/variables/#MathOptInterface.AbstractVariableAttribute"><code>AbstractVariableAttribute</code></a>) refer to properties of an individual variable.</li></ul><p>Some attributes are values that can be queried by the user but not modified, while other attributes can be modified by the user.</p><p>All interactions with attributes occur through the <a href="../../reference/models/#MathOptInterface.get"><code>get</code></a> and <a href="../../reference/models/#MathOptInterface.set"><code>set</code></a> functions.</p><p>Consult the docstrings of each attribute for information on what it represents.</p><h2 id="ModelLike-API"><a class="docs-heading-anchor" href="#ModelLike-API">ModelLike API</a><a id="ModelLike-API-1"></a><a class="docs-heading-anchor-permalink" href="#ModelLike-API" title="Permalink"></a></h2><p>The following attributes are available:</p><ul><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintIndices"><code>ListOfConstraintIndices</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintTypesPresent"><code>ListOfConstraintTypesPresent</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfVariableAttributesSet"><code>ListOfVariableAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfVariableIndices"><code>ListOfVariableIndices</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfConstraints"><code>NumberOfConstraints</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a></li><li><a href="../../reference/models/#MathOptInterface.Name"><code>Name</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveFunction"><code>ObjectiveFunction</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveFunctionType"><code>ObjectiveFunctionType</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveSense"><code>ObjectiveSense</code></a></li></ul><h2 id="AbstractOptimizer-API"><a class="docs-heading-anchor" href="#AbstractOptimizer-API">AbstractOptimizer API</a><a id="AbstractOptimizer-API-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractOptimizer-API" title="Permalink"></a></h2><p>The following attributes are available:</p><ul><li><a href="../../reference/models/#MathOptInterface.DualStatus"><code>DualStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawStatusString"><code>RawStatusString</code></a></li><li><a href="../../reference/models/#MathOptInterface.ResultCount"><code>ResultCount</code></a></li><li><a href="../../reference/models/#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.BarrierIterations"><code>BarrierIterations</code></a></li><li><a href="../../reference/models/#MathOptInterface.DualObjectiveValue"><code>DualObjectiveValue</code></a></li><li><a href="../../reference/models/#MathOptInterface.NodeCount"><code>NodeCount</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfThreads"><code>NumberOfThreads</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveBound"><code>ObjectiveBound</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveValue"><code>ObjectiveValue</code></a></li><li><a href="../../reference/models/#MathOptInterface.RelativeGap"><code>RelativeGap</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawOptimizerAttribute"><code>RawOptimizerAttribute</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawSolver"><code>RawSolver</code></a></li><li><a href="../../reference/models/#MathOptInterface.Silent"><code>Silent</code></a></li><li><a href="../../reference/models/#MathOptInterface.SimplexIterations"><code>SimplexIterations</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolverName"><code>SolverName</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolverVersion"><code>SolverVersion</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolveTimeSec"><code>SolveTimeSec</code></a></li><li><a href="../../reference/models/#MathOptInterface.TimeLimitSec"><code>TimeLimitSec</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveLimit"><code>ObjectiveLimit</code></a></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../standard_form/">« Standard form problem</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox" checked/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../standard_form/">Standard form problem</a></li><li class="is-active"><a class="tocitem" href>Models</a><ul class="internal"><li><a class="tocitem" href="#Attributes"><span>Attributes</span></a></li><li><a class="tocitem" href="#ModelLike-API"><span>ModelLike API</span></a></li><li><a class="tocitem" href="#AbstractOptimizer-API"><span>AbstractOptimizer API</span></a></li></ul></li><li><a class="tocitem" href="../variables/">Variables</a></li><li><a class="tocitem" href="../constraints/">Constraints</a></li><li><a class="tocitem" href="../solutions/">Solutions</a></li><li><a class="tocitem" href="../modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/duality/">Duality</a></li><li><a class="tocitem" href="../../background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../reference/models/">Models</a></li><li><a class="tocitem" href="../../reference/variables/">Variables</a></li><li><a class="tocitem" href="../../reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">Manual</a></li><li class="is-active"><a href>Models</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Models</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/models.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Models"><a class="docs-heading-anchor" href="#Models">Models</a><a id="Models-1"></a><a class="docs-heading-anchor-permalink" href="#Models" title="Permalink"></a></h1><p>The most significant part of MOI is the definition of the <strong>model API</strong> that is used to specify an instance of an optimization problem (for example, by adding variables and constraints). Objects that implement the model API must inherit from the <a href="../../reference/models/#MathOptInterface.ModelLike"><code>ModelLike</code></a> abstract type.</p><p>Notably missing from the model API is the method to solve an optimization problem. <code>ModelLike</code> objects may store an instance (for example, in memory or backed by a file format) without being linked to a particular solver. In addition to the model API, MOI defines <a href="../../reference/models/#MathOptInterface.AbstractOptimizer"><code>AbstractOptimizer</code></a> and provides methods to solve the model and interact with solutions. See the <a href="../solutions/#manual_solutions">Solutions</a> section for more details.</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>Throughout the rest of the manual, <code>model</code> is used as a generic <code>ModelLike</code>, and <code>optimizer</code> is used as a generic <code>AbstractOptimizer</code>.</p></div></div><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>MOI does not export functions, but for brevity we often omit qualifying names with the MOI module. Best practice is to have</p><pre><code class="language-julia hljs">import MathOptInterface as MOI</code></pre><p>and prefix all MOI methods with <code>MOI.</code> in user code. If a name is also available in base Julia, we always explicitly use the module prefix, for example, with <code>MOI.get</code>.</p></div></div><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><p>Attributes are properties of the model that can be queried and modified. These include constants such as the number of variables in a model <a href="../../reference/models/#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a>), and properties of variables and constraints such as the name of a variable (<a href="../../reference/variables/#MathOptInterface.VariableName"><code>VariableName</code></a>).</p><p>There are four types of attributes:</p><ul><li>Model attributes (subtypes of <a href="../../reference/models/#MathOptInterface.AbstractModelAttribute"><code>AbstractModelAttribute</code></a>) refer to properties of a model.</li><li>Optimizer attributes (subtypes of <a href="../../reference/models/#MathOptInterface.AbstractOptimizerAttribute"><code>AbstractOptimizerAttribute</code></a>) refer to properties of an optimizer.</li><li>Constraint attributes (subtypes of <a href="../../reference/constraints/#MathOptInterface.AbstractConstraintAttribute"><code>AbstractConstraintAttribute</code></a>) refer to properties of an individual constraint.</li><li>Variable attributes (subtypes of <a href="../../reference/variables/#MathOptInterface.AbstractVariableAttribute"><code>AbstractVariableAttribute</code></a>) refer to properties of an individual variable.</li></ul><p>Some attributes are values that can be queried by the user but not modified, while other attributes can be modified by the user.</p><p>All interactions with attributes occur through the <a href="../../reference/models/#MathOptInterface.get"><code>get</code></a> and <a href="../../reference/models/#MathOptInterface.set"><code>set</code></a> functions.</p><p>Consult the docstrings of each attribute for information on what it represents.</p><h2 id="ModelLike-API"><a class="docs-heading-anchor" href="#ModelLike-API">ModelLike API</a><a id="ModelLike-API-1"></a><a class="docs-heading-anchor-permalink" href="#ModelLike-API" title="Permalink"></a></h2><p>The following attributes are available:</p><ul><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintIndices"><code>ListOfConstraintIndices</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfConstraintTypesPresent"><code>ListOfConstraintTypesPresent</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfVariableAttributesSet"><code>ListOfVariableAttributesSet</code></a></li><li><a href="../../reference/models/#MathOptInterface.ListOfVariableIndices"><code>ListOfVariableIndices</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfConstraints"><code>NumberOfConstraints</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a></li><li><a href="../../reference/models/#MathOptInterface.Name"><code>Name</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveFunction"><code>ObjectiveFunction</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveFunctionType"><code>ObjectiveFunctionType</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveSense"><code>ObjectiveSense</code></a></li></ul><h2 id="AbstractOptimizer-API"><a class="docs-heading-anchor" href="#AbstractOptimizer-API">AbstractOptimizer API</a><a id="AbstractOptimizer-API-1"></a><a class="docs-heading-anchor-permalink" href="#AbstractOptimizer-API" title="Permalink"></a></h2><p>The following attributes are available:</p><ul><li><a href="../../reference/models/#MathOptInterface.DualStatus"><code>DualStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawStatusString"><code>RawStatusString</code></a></li><li><a href="../../reference/models/#MathOptInterface.ResultCount"><code>ResultCount</code></a></li><li><a href="../../reference/models/#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a></li><li><a href="../../reference/models/#MathOptInterface.BarrierIterations"><code>BarrierIterations</code></a></li><li><a href="../../reference/models/#MathOptInterface.DualObjectiveValue"><code>DualObjectiveValue</code></a></li><li><a href="../../reference/models/#MathOptInterface.NodeCount"><code>NodeCount</code></a></li><li><a href="../../reference/models/#MathOptInterface.NumberOfThreads"><code>NumberOfThreads</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveBound"><code>ObjectiveBound</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveValue"><code>ObjectiveValue</code></a></li><li><a href="../../reference/models/#MathOptInterface.RelativeGap"><code>RelativeGap</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawOptimizerAttribute"><code>RawOptimizerAttribute</code></a></li><li><a href="../../reference/models/#MathOptInterface.RawSolver"><code>RawSolver</code></a></li><li><a href="../../reference/models/#MathOptInterface.Silent"><code>Silent</code></a></li><li><a href="../../reference/models/#MathOptInterface.SimplexIterations"><code>SimplexIterations</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolverName"><code>SolverName</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolverVersion"><code>SolverVersion</code></a></li><li><a href="../../reference/models/#MathOptInterface.SolveTimeSec"><code>SolveTimeSec</code></a></li><li><a href="../../reference/models/#MathOptInterface.TimeLimitSec"><code>TimeLimitSec</code></a></li><li><a href="../../reference/models/#MathOptInterface.ObjectiveLimit"><code>ObjectiveLimit</code></a></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../standard_form/">« Standard form problem</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/modification/index.html b/previews/PR3562/moi/manual/modification/index.html
index 563cb8ca062..35c843aa17e 100644
--- a/previews/PR3562/moi/manual/modification/index.html
+++ b/previews/PR3562/moi/manual/modification/index.html
@@ -126,4 +126,4 @@
        );
 
 julia&gt; MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f
-true</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solutions/">« Solutions</a><a class="docs-footer-nextpage" href="../../background/duality/">Duality »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+true</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solutions/">« Solutions</a><a class="docs-footer-nextpage" href="../../background/duality/">Duality »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/solutions/index.html b/previews/PR3562/moi/manual/solutions/index.html
index 978b5a34775..bf55c2b1742 100644
--- a/previews/PR3562/moi/manual/solutions/index.html
+++ b/previews/PR3562/moi/manual/solutions/index.html
@@ -36,4 +36,4 @@
         end
         rethrow(err)  # Something else went wrong. Rethrow the error
     end
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../modification/">Problem modification »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../modification/">Problem modification »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/standard_form/index.html b/previews/PR3562/moi/manual/standard_form/index.html
index c822abb01fb..db7756922ce 100644
--- a/previews/PR3562/moi/manual/standard_form/index.html
+++ b/previews/PR3562/moi/manual/standard_form/index.html
@@ -7,4 +7,4 @@
     &amp; \min_{x \in \mathbb{R}^n} &amp; f_0(x)
     \\
     &amp; \;\;\text{s.t.} &amp; f_i(x) &amp; \in \mathcal{S}_i &amp; i = 1 \ldots m
-\end{align}\]</p><p>where:</p><ul><li>the functions <span>$f_0, f_1, \ldots, f_m$</span> are specified by <a href="../../reference/standard_form/#MathOptInterface.AbstractFunction"><code>AbstractFunction</code></a> objects</li><li>the sets <span>$\mathcal{S}_1, \ldots, \mathcal{S}_m$</span> are specified by <a href="../../reference/standard_form/#MathOptInterface.AbstractSet"><code>AbstractSet</code></a> objects</li></ul><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>For more information on this standard form, read <a href="https://arxiv.org/pdf/2002.03447.pdf">our paper</a>.</p></div></div><p>MOI defines some commonly used functions and sets, but the interface is extensible to other sets recognized by the solver.</p><h2 id="Functions"><a class="docs-heading-anchor" href="#Functions">Functions</a><a id="Functions-1"></a><a class="docs-heading-anchor-permalink" href="#Functions" title="Permalink"></a></h2><p>The function types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Function</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a></td><td style="text-align: left"><span>$x_j$</span>, the projection onto a single coordinate defined by a variable index <span>$j$</span>.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorOfVariables"><code>VectorOfVariables</code></a></td><td style="text-align: left">The projection onto multiple coordinates (that is, extracting a sub-vector).</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a></td><td style="text-align: left"><span>$a^T x + b$</span>, where <span>$a$</span> is a vector and <span>$b$</span> scalar.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>ScalarNonlinearFunction</code></a></td><td style="text-align: left"><span>$f(x)$</span>, where <span>$f$</span> is a nonlinear function.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>VectorAffineFunction</code></a></td><td style="text-align: left"><span>$A x + b$</span>, where <span>$A$</span> is a matrix and <span>$b$</span> is a vector.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a></td><td style="text-align: left"><span>$\frac{1}{2} x^T Q x + a^T x + b$</span>, where <span>$Q$</span> is a symmetric matrix, <span>$a$</span> is a vector, and <span>$b$</span> is a constant.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a></td><td style="text-align: left">A vector of scalar-valued quadratic functions.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorNonlinearFunction"><code>VectorNonlinearFunction</code></a></td><td style="text-align: left"><span>$f(x)$</span>, where <span>$f$</span> is a vector-valued nonlinear function.</td></tr></table><p>Extensions for nonlinear programming are present but not yet well documented.</p><h2 id="One-dimensional-sets"><a class="docs-heading-anchor" href="#One-dimensional-sets">One-dimensional sets</a><a id="One-dimensional-sets-1"></a><a class="docs-heading-anchor-permalink" href="#One-dimensional-sets" title="Permalink"></a></h2><p>The one-dimensional set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan(u)</code></a></td><td style="text-align: left"><span>$(-\infty, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan(l)</code></a></td><td style="text-align: left"><span>$[l, \infty)$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>EqualTo(v)</code></a></td><td style="text-align: left"><span>$\{v\}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval(l, u)</code></a></td><td style="text-align: left"><span>$[l, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Integer"><code>Integer()</code></a></td><td style="text-align: left"><span>$\mathbb{Z}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ZeroOne"><code>ZeroOne()</code></a></td><td style="text-align: left"><span>$\{ 0, 1 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Semicontinuous"><code>Semicontinuous(l, u)</code></a></td><td style="text-align: left"><span>$\{ 0\} \cup [l, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Semiinteger"><code>Semiinteger(l, u)</code></a></td><td style="text-align: left"><span>$\{ 0\} \cup \{l,l+1,\ldots,u-1,u\}$</span></td></tr></table><h2 id="Vector-cones"><a class="docs-heading-anchor" href="#Vector-cones">Vector cones</a><a id="Vector-cones-1"></a><a class="docs-heading-anchor-permalink" href="#Vector-cones" title="Permalink"></a></h2><p>The vector-valued set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Reals"><code>Reals(d)</code></a></td><td style="text-align: left"><span>$\mathbb{R}^{d}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Zeros"><code>Zeros(d)</code></a></td><td style="text-align: left"><span>$0^{d}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Nonnegatives"><code>Nonnegatives(d)</code></a></td><td style="text-align: left"><span>$\{ x \in \mathbb{R}^{d} : x \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Nonpositives"><code>Nonpositives(d)</code></a></td><td style="text-align: left"><span>$\{ x \in \mathbb{R}^{d} : x \le 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SecondOrderCone"><code>SecondOrderCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \lVert x \rVert_2 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RotatedSecondOrderCone"><code>RotatedSecondOrderCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,x) \in \mathbb{R}^{d} : 2tu \ge \lVert x \rVert_2^2, t \ge 0,u \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ExponentialCone"><code>ExponentialCone()</code></a></td><td style="text-align: left"><span>$\{ (x,y,z) \in \mathbb{R}^3 : y \exp (x/y) \le z, y &gt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.DualExponentialCone"><code>DualExponentialCone()</code></a></td><td style="text-align: left"><span>$\{ (u,v,w) \in \mathbb{R}^3 : -u \exp (v/u) \le \exp(1) w, u &lt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>GeometricMeanCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{1+n} : x \ge 0, t \le \sqrt[n]{x_1 x_2 \cdots x_n} \}$</span> where <span>$n$</span> is <span>$d - 1$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PowerCone"><code>PowerCone(α)</code></a></td><td style="text-align: left"><span>$\{ (x,y,z) \in \mathbb{R}^3 : x^{\alpha} y^{1-\alpha} \ge |z|, x \ge 0,y \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.DualPowerCone"><code>DualPowerCone(α)</code></a></td><td style="text-align: left"><span>$\{ (u,v,w) \in \mathbb{R}^3 : \left(\frac{u}{\alpha}\right(^{\alpha}\left(\frac{v}{1-\alpha}\right)^{1-\alpha} \ge |w|, u,v \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormOneCone"><code>NormOneCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \sum_i \lvert x_i \rvert \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormInfinityCone"><code>NormInfinityCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \max_i \lvert x_i \rvert \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RelativeEntropyCone"><code>RelativeEntropyCone(d)</code></a></td><td style="text-align: left"><span>$\{ (u, v, w) \in \mathbb{R}^{d} : u \ge \sum_i w_i \log (\frac{w_i}{v_i}), v_i \ge 0, w_i \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.HyperRectangle"><code>HyperRectangle(l, u)</code></a></td><td style="text-align: left"><span>$\{x \in \bar{\mathbb{R}}^d: x_i \in [l_i, u_i] \forall i=1,\ldots,d\}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormCone"><code>NormCone(p, d)</code></a></td><td style="text-align: left">``{ (t,x) \in \mathbb{R}^{d} : t \ge \left(\sum\limits_i</td></tr></table><h2 id="Matrix-cones"><a class="docs-heading-anchor" href="#Matrix-cones">Matrix cones</a><a id="Matrix-cones-1"></a><a class="docs-heading-anchor-permalink" href="#Matrix-cones" title="Permalink"></a></h2><p>The matrix-valued set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>RootDetConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ (t,X) \in \mathbb{R}^{1+d(1+d)/2} : t \le \det(X)^{1/d}, X \mbox{ is the upper triangle of a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>RootDetConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ (t,X) \in \mathbb{R}^{1+d^2} : t \le \det(X)^{1/d}, X \mbox{ is a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle"><code>PositiveSemidefiniteConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ X \in \mathbb{R}^{d(d+1)/2} : X \mbox{ is the upper triangle of a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeSquare"><code>PositiveSemidefiniteConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ X \in \mathbb{R}^{d^2} : X \mbox{ is a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LogDetConeTriangle"><code>LogDetConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,X) \in \mathbb{R}^{2+d(1+d)/2} : t \le u\log(\det(X/u)), X \mbox{ is the upper triangle of a PSD matrix}, u &gt; 0  \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LogDetConeSquare"><code>LogDetConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,X) \in \mathbb{R}^{2+d^2} : t \le u \log(\det(X/u)), X \mbox{ is a PSD matrix}, u &gt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormSpectralCone"><code>NormSpectralCone(r, c)</code></a></td><td style="text-align: left"><span>$\{ (t, X) \in \mathbb{R}^{1 + r \times c} : t \ge \sigma_1(X), X \mbox{ is a } r\times c\mbox{ matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormNuclearCone"><code>NormNuclearCone(r, c)</code></a></td><td style="text-align: left"><span>$\{ (t, X) \in \mathbb{R}^{1 + r \times c} : t \ge \sum_i \sigma_i(X), X \mbox{ is a } r\times c\mbox{ matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle"><code>HermitianPositiveSemidefiniteConeTriangle(d)</code></a></td><td style="text-align: left">The cone of Hermitian positive semidefinite matrices, with</td></tr><tr><td style="text-align: left"><code>side_dimension</code> rows and columns.</td><td style="text-align: left"></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Scaled"><code>Scaled(S)</code></a></td><td style="text-align: left">The set <code>S</code> scaled so that <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.set_dot"><code>Utilities.set_dot</code></a> corresponds to <code>LinearAlgebra.dot</code></td></tr></table><p>Some of these cones can take two forms: <code>XXXConeTriangle</code> and <code>XXXConeSquare</code>.</p><p>In <code>XXXConeTriangle</code> sets, the matrix is assumed to be symmetric, and the elements are provided by a vector, in which the entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row).</p><p>In <code>XXXConeSquare</code> sets, the entries of the matrix are given column by column (or equivalently, row by row), and the matrix is constrained to be symmetric. As an example, given a 2-by-2 matrix of variables <code>X</code> and a one-dimensional variable <code>t</code>, we can specify a root-det constraint as <code>[t, X11, X12, X22] ∈ RootDetConeTriangle</code> or <code>[t, X11, X12, X21, X22] ∈ RootDetConeSquare</code>.</p><p>We provide both forms to enable flexibility for solvers who may natively support one or the other. Transformations between <code>XXXConeTriangle</code> and <code>XXXConeSquare</code> are handled by bridges, which removes the chance of conversion mistakes by users or solver developers.</p><h2 id="Multi-dimensional-sets-with-combinatorial-structure"><a class="docs-heading-anchor" href="#Multi-dimensional-sets-with-combinatorial-structure">Multi-dimensional sets with combinatorial structure</a><a id="Multi-dimensional-sets-with-combinatorial-structure-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-dimensional-sets-with-combinatorial-structure" title="Permalink"></a></h2><p>Other sets are vector-valued, with a particular combinatorial structure. Read their docstrings for more information on how to interpret them.</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SOS1"><code>SOS1</code></a></td><td style="text-align: left">A Special Ordered Set (SOS) of Type I</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SOS2"><code>SOS2</code></a></td><td style="text-align: left">A Special Ordered Set (SOS) of Type II</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Indicator"><code>Indicator</code></a></td><td style="text-align: left">A set to specify an indicator constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Complements"><code>Complements</code></a></td><td style="text-align: left">A set to specify a mixed complementarity constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.AllDifferent"><code>AllDifferent</code></a></td><td style="text-align: left">The <code>all_different</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.BinPacking"><code>BinPacking</code></a></td><td style="text-align: left">The <code>bin_packing</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Circuit"><code>Circuit</code></a></td><td style="text-align: left">The <code>circuit</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountAtLeast"><code>CountAtLeast</code></a></td><td style="text-align: left">The <code>at_least</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountBelongs"><code>CountBelongs</code></a></td><td style="text-align: left">The <code>nvalue</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountDistinct"><code>CountDistinct</code></a></td><td style="text-align: left">The <code>distinct</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountGreaterThan"><code>CountGreaterThan</code></a></td><td style="text-align: left">The <code>count_gt</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Cumulative"><code>Cumulative</code></a></td><td style="text-align: left">The <code>cumulative</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Path"><code>Path</code></a></td><td style="text-align: left">The <code>path</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Table"><code>Table</code></a></td><td style="text-align: left">The <code>table</code> global constraint</td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../tutorials/latency/">« Latency</a><a class="docs-footer-nextpage" href="../models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>where:</p><ul><li>the functions <span>$f_0, f_1, \ldots, f_m$</span> are specified by <a href="../../reference/standard_form/#MathOptInterface.AbstractFunction"><code>AbstractFunction</code></a> objects</li><li>the sets <span>$\mathcal{S}_1, \ldots, \mathcal{S}_m$</span> are specified by <a href="../../reference/standard_form/#MathOptInterface.AbstractSet"><code>AbstractSet</code></a> objects</li></ul><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>For more information on this standard form, read <a href="https://arxiv.org/pdf/2002.03447.pdf">our paper</a>.</p></div></div><p>MOI defines some commonly used functions and sets, but the interface is extensible to other sets recognized by the solver.</p><h2 id="Functions"><a class="docs-heading-anchor" href="#Functions">Functions</a><a id="Functions-1"></a><a class="docs-heading-anchor-permalink" href="#Functions" title="Permalink"></a></h2><p>The function types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Function</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a></td><td style="text-align: left"><span>$x_j$</span>, the projection onto a single coordinate defined by a variable index <span>$j$</span>.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorOfVariables"><code>VectorOfVariables</code></a></td><td style="text-align: left">The projection onto multiple coordinates (that is, extracting a sub-vector).</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a></td><td style="text-align: left"><span>$a^T x + b$</span>, where <span>$a$</span> is a vector and <span>$b$</span> scalar.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>ScalarNonlinearFunction</code></a></td><td style="text-align: left"><span>$f(x)$</span>, where <span>$f$</span> is a nonlinear function.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>VectorAffineFunction</code></a></td><td style="text-align: left"><span>$A x + b$</span>, where <span>$A$</span> is a matrix and <span>$b$</span> is a vector.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a></td><td style="text-align: left"><span>$\frac{1}{2} x^T Q x + a^T x + b$</span>, where <span>$Q$</span> is a symmetric matrix, <span>$a$</span> is a vector, and <span>$b$</span> is a constant.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a></td><td style="text-align: left">A vector of scalar-valued quadratic functions.</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.VectorNonlinearFunction"><code>VectorNonlinearFunction</code></a></td><td style="text-align: left"><span>$f(x)$</span>, where <span>$f$</span> is a vector-valued nonlinear function.</td></tr></table><p>Extensions for nonlinear programming are present but not yet well documented.</p><h2 id="One-dimensional-sets"><a class="docs-heading-anchor" href="#One-dimensional-sets">One-dimensional sets</a><a id="One-dimensional-sets-1"></a><a class="docs-heading-anchor-permalink" href="#One-dimensional-sets" title="Permalink"></a></h2><p>The one-dimensional set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LessThan"><code>LessThan(u)</code></a></td><td style="text-align: left"><span>$(-\infty, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan(l)</code></a></td><td style="text-align: left"><span>$[l, \infty)$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GreaterThan"><code>EqualTo(v)</code></a></td><td style="text-align: left"><span>$\{v\}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Interval"><code>Interval(l, u)</code></a></td><td style="text-align: left"><span>$[l, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Integer"><code>Integer()</code></a></td><td style="text-align: left"><span>$\mathbb{Z}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ZeroOne"><code>ZeroOne()</code></a></td><td style="text-align: left"><span>$\{ 0, 1 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Semicontinuous"><code>Semicontinuous(l, u)</code></a></td><td style="text-align: left"><span>$\{ 0\} \cup [l, u]$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Semiinteger"><code>Semiinteger(l, u)</code></a></td><td style="text-align: left"><span>$\{ 0\} \cup \{l,l+1,\ldots,u-1,u\}$</span></td></tr></table><h2 id="Vector-cones"><a class="docs-heading-anchor" href="#Vector-cones">Vector cones</a><a id="Vector-cones-1"></a><a class="docs-heading-anchor-permalink" href="#Vector-cones" title="Permalink"></a></h2><p>The vector-valued set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Reals"><code>Reals(d)</code></a></td><td style="text-align: left"><span>$\mathbb{R}^{d}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Zeros"><code>Zeros(d)</code></a></td><td style="text-align: left"><span>$0^{d}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Nonnegatives"><code>Nonnegatives(d)</code></a></td><td style="text-align: left"><span>$\{ x \in \mathbb{R}^{d} : x \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Nonpositives"><code>Nonpositives(d)</code></a></td><td style="text-align: left"><span>$\{ x \in \mathbb{R}^{d} : x \le 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SecondOrderCone"><code>SecondOrderCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \lVert x \rVert_2 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RotatedSecondOrderCone"><code>RotatedSecondOrderCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,x) \in \mathbb{R}^{d} : 2tu \ge \lVert x \rVert_2^2, t \ge 0,u \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.ExponentialCone"><code>ExponentialCone()</code></a></td><td style="text-align: left"><span>$\{ (x,y,z) \in \mathbb{R}^3 : y \exp (x/y) \le z, y &gt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.DualExponentialCone"><code>DualExponentialCone()</code></a></td><td style="text-align: left"><span>$\{ (u,v,w) \in \mathbb{R}^3 : -u \exp (v/u) \le \exp(1) w, u &lt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>GeometricMeanCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{1+n} : x \ge 0, t \le \sqrt[n]{x_1 x_2 \cdots x_n} \}$</span> where <span>$n$</span> is <span>$d - 1$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PowerCone"><code>PowerCone(α)</code></a></td><td style="text-align: left"><span>$\{ (x,y,z) \in \mathbb{R}^3 : x^{\alpha} y^{1-\alpha} \ge |z|, x \ge 0,y \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.DualPowerCone"><code>DualPowerCone(α)</code></a></td><td style="text-align: left"><span>$\{ (u,v,w) \in \mathbb{R}^3 : \left(\frac{u}{\alpha}\right(^{\alpha}\left(\frac{v}{1-\alpha}\right)^{1-\alpha} \ge |w|, u,v \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormOneCone"><code>NormOneCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \sum_i \lvert x_i \rvert \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormInfinityCone"><code>NormInfinityCone(d)</code></a></td><td style="text-align: left"><span>$\{ (t,x) \in \mathbb{R}^{d} : t \ge \max_i \lvert x_i \rvert \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RelativeEntropyCone"><code>RelativeEntropyCone(d)</code></a></td><td style="text-align: left"><span>$\{ (u, v, w) \in \mathbb{R}^{d} : u \ge \sum_i w_i \log (\frac{w_i}{v_i}), v_i \ge 0, w_i \ge 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.HyperRectangle"><code>HyperRectangle(l, u)</code></a></td><td style="text-align: left"><span>$\{x \in \bar{\mathbb{R}}^d: x_i \in [l_i, u_i] \forall i=1,\ldots,d\}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormCone"><code>NormCone(p, d)</code></a></td><td style="text-align: left">``{ (t,x) \in \mathbb{R}^{d} : t \ge \left(\sum\limits_i</td></tr></table><h2 id="Matrix-cones"><a class="docs-heading-anchor" href="#Matrix-cones">Matrix cones</a><a id="Matrix-cones-1"></a><a class="docs-heading-anchor-permalink" href="#Matrix-cones" title="Permalink"></a></h2><p>The matrix-valued set types implemented in MathOptInterface.jl are:</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>RootDetConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ (t,X) \in \mathbb{R}^{1+d(1+d)/2} : t \le \det(X)^{1/d}, X \mbox{ is the upper triangle of a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>RootDetConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ (t,X) \in \mathbb{R}^{1+d^2} : t \le \det(X)^{1/d}, X \mbox{ is a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle"><code>PositiveSemidefiniteConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ X \in \mathbb{R}^{d(d+1)/2} : X \mbox{ is the upper triangle of a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeSquare"><code>PositiveSemidefiniteConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ X \in \mathbb{R}^{d^2} : X \mbox{ is a PSD matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LogDetConeTriangle"><code>LogDetConeTriangle(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,X) \in \mathbb{R}^{2+d(1+d)/2} : t \le u\log(\det(X/u)), X \mbox{ is the upper triangle of a PSD matrix}, u &gt; 0  \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.LogDetConeSquare"><code>LogDetConeSquare(d)</code></a></td><td style="text-align: left"><span>$\{ (t,u,X) \in \mathbb{R}^{2+d^2} : t \le u \log(\det(X/u)), X \mbox{ is a PSD matrix}, u &gt; 0 \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormSpectralCone"><code>NormSpectralCone(r, c)</code></a></td><td style="text-align: left"><span>$\{ (t, X) \in \mathbb{R}^{1 + r \times c} : t \ge \sigma_1(X), X \mbox{ is a } r\times c\mbox{ matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.NormNuclearCone"><code>NormNuclearCone(r, c)</code></a></td><td style="text-align: left"><span>$\{ (t, X) \in \mathbb{R}^{1 + r \times c} : t \ge \sum_i \sigma_i(X), X \mbox{ is a } r\times c\mbox{ matrix} \}$</span></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle"><code>HermitianPositiveSemidefiniteConeTriangle(d)</code></a></td><td style="text-align: left">The cone of Hermitian positive semidefinite matrices, with</td></tr><tr><td style="text-align: left"><code>side_dimension</code> rows and columns.</td><td style="text-align: left"></td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Scaled"><code>Scaled(S)</code></a></td><td style="text-align: left">The set <code>S</code> scaled so that <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.set_dot"><code>Utilities.set_dot</code></a> corresponds to <code>LinearAlgebra.dot</code></td></tr></table><p>Some of these cones can take two forms: <code>XXXConeTriangle</code> and <code>XXXConeSquare</code>.</p><p>In <code>XXXConeTriangle</code> sets, the matrix is assumed to be symmetric, and the elements are provided by a vector, in which the entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row).</p><p>In <code>XXXConeSquare</code> sets, the entries of the matrix are given column by column (or equivalently, row by row), and the matrix is constrained to be symmetric. As an example, given a 2-by-2 matrix of variables <code>X</code> and a one-dimensional variable <code>t</code>, we can specify a root-det constraint as <code>[t, X11, X12, X22] ∈ RootDetConeTriangle</code> or <code>[t, X11, X12, X21, X22] ∈ RootDetConeSquare</code>.</p><p>We provide both forms to enable flexibility for solvers who may natively support one or the other. Transformations between <code>XXXConeTriangle</code> and <code>XXXConeSquare</code> are handled by bridges, which removes the chance of conversion mistakes by users or solver developers.</p><h2 id="Multi-dimensional-sets-with-combinatorial-structure"><a class="docs-heading-anchor" href="#Multi-dimensional-sets-with-combinatorial-structure">Multi-dimensional sets with combinatorial structure</a><a id="Multi-dimensional-sets-with-combinatorial-structure-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-dimensional-sets-with-combinatorial-structure" title="Permalink"></a></h2><p>Other sets are vector-valued, with a particular combinatorial structure. Read their docstrings for more information on how to interpret them.</p><table><tr><th style="text-align: left">Set</th><th style="text-align: left">Description</th></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SOS1"><code>SOS1</code></a></td><td style="text-align: left">A Special Ordered Set (SOS) of Type I</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.SOS2"><code>SOS2</code></a></td><td style="text-align: left">A Special Ordered Set (SOS) of Type II</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Indicator"><code>Indicator</code></a></td><td style="text-align: left">A set to specify an indicator constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Complements"><code>Complements</code></a></td><td style="text-align: left">A set to specify a mixed complementarity constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.AllDifferent"><code>AllDifferent</code></a></td><td style="text-align: left">The <code>all_different</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.BinPacking"><code>BinPacking</code></a></td><td style="text-align: left">The <code>bin_packing</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Circuit"><code>Circuit</code></a></td><td style="text-align: left">The <code>circuit</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountAtLeast"><code>CountAtLeast</code></a></td><td style="text-align: left">The <code>at_least</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountBelongs"><code>CountBelongs</code></a></td><td style="text-align: left">The <code>nvalue</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountDistinct"><code>CountDistinct</code></a></td><td style="text-align: left">The <code>distinct</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.CountGreaterThan"><code>CountGreaterThan</code></a></td><td style="text-align: left">The <code>count_gt</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Cumulative"><code>Cumulative</code></a></td><td style="text-align: left">The <code>cumulative</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Path"><code>Path</code></a></td><td style="text-align: left">The <code>path</code> global constraint</td></tr><tr><td style="text-align: left"><a href="../../reference/standard_form/#MathOptInterface.Table"><code>Table</code></a></td><td style="text-align: left">The <code>table</code> global constraint</td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../tutorials/latency/">« Latency</a><a class="docs-footer-nextpage" href="../models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/manual/variables/index.html b/previews/PR3562/moi/manual/variables/index.html
index cef9ee415d6..4a8495da9bb 100644
--- a/previews/PR3562/moi/manual/variables/index.html
+++ b/previews/PR3562/moi/manual/variables/index.html
@@ -14,4 +14,4 @@
 false</code></pre><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Not all <code>ModelLike</code> models support deleting variables. A <a href="../../reference/errors/#MathOptInterface.DeleteNotAllowed"><code>DeleteNotAllowed</code></a> error is thrown if this is not supported.</p></div></div><h2 id="Variable-attributes"><a class="docs-heading-anchor" href="#Variable-attributes">Variable attributes</a><a id="Variable-attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Variable-attributes" title="Permalink"></a></h2><p>The following attributes are available for variables:</p><ul><li><a href="../../reference/variables/#MathOptInterface.VariableName"><code>VariableName</code></a></li><li><a href="../../reference/variables/#MathOptInterface.VariablePrimalStart"><code>VariablePrimalStart</code></a></li><li><a href="../../reference/variables/#MathOptInterface.VariablePrimal"><code>VariablePrimal</code></a></li></ul><p>Get and set these attributes using <a href="../../reference/models/#MathOptInterface.get"><code>get</code></a> and <a href="../../reference/models/#MathOptInterface.set"><code>set</code></a>.</p><pre><code class="language-julia-repl hljs">julia&gt; MOI.set(model, MOI.VariableName(), x, &quot;var_x&quot;)
 
 julia&gt; MOI.get(model, MOI.VariableName(), x)
-&quot;var_x&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+&quot;var_x&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/callbacks/index.html b/previews/PR3562/moi/reference/callbacks/index.html
index 30e113f45e3..e056a0387e8 100644
--- a/previews/PR3562/moi/reference/callbacks/index.html
+++ b/previews/PR3562/moi/reference/callbacks/index.html
@@ -33,4 +33,4 @@
         MOI.submit(optimizer, MOI.HeuristicSolution(callback_data), x,
                    values)
     end
-end</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1019-L1045">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HeuristicSolution" href="#MathOptInterface.HeuristicSolution"><code>MathOptInterface.HeuristicSolution</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HeuristicSolution(callback_data)</code></pre><p>Heuristically obtained feasible solution. The solution is submitted as <code>variables, values</code> where <code>values[i]</code> gives the value of <code>variables[i]</code>, similarly to <a href="../models/#MathOptInterface.set"><code>set</code></a>. The <a href="#MathOptInterface.submit"><code>submit</code></a> call returns a <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> indicating whether the provided solution was accepted or rejected.</p><p>This can be submitted only from the <a href="#MathOptInterface.HeuristicCallback"><code>HeuristicCallback</code></a>. The field <code>callback_data</code> is a solver-specific callback type that is passed as the argument to the heuristic callback.</p><p>Some solvers require a complete solution, others only partial solutions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L685-L699">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HeuristicSolutionStatus" href="#MathOptInterface.HeuristicSolutionStatus"><code>MathOptInterface.HeuristicSolutionStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HeuristicSolutionStatus</code></pre><p>An Enum of possible return values for <a href="#MathOptInterface.submit"><code>submit</code></a> with <a href="#MathOptInterface.HeuristicSolution"><code>HeuristicSolution</code></a>. This informs whether the heuristic solution was accepted or rejected.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED"><code>HEURISTIC_SOLUTION_ACCEPTED</code></a>: The heuristic solution was accepted</li><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_REJECTED"><code>HEURISTIC_SOLUTION_REJECTED</code></a>: The heuristic solution was rejected</li><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN"><code>HEURISTIC_SOLUTION_UNKNOWN</code></a>: No information available on the acceptance</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L682">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED" href="#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED"><code>MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_ACCEPTED::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_ACCEPTED</code>: The heuristic solution was accepted</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_REJECTED" href="#MathOptInterface.HEURISTIC_SOLUTION_REJECTED"><code>MathOptInterface.HEURISTIC_SOLUTION_REJECTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_REJECTED::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_REJECTED</code>: The heuristic solution was rejected</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN" href="#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN"><code>MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_UNKNOWN::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_UNKNOWN</code>: No information available on the acceptance</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nonlinear/">« Nonlinear programming</a><a class="docs-footer-nextpage" href="../errors/">Errors »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1019-L1045">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HeuristicSolution" href="#MathOptInterface.HeuristicSolution"><code>MathOptInterface.HeuristicSolution</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HeuristicSolution(callback_data)</code></pre><p>Heuristically obtained feasible solution. The solution is submitted as <code>variables, values</code> where <code>values[i]</code> gives the value of <code>variables[i]</code>, similarly to <a href="../models/#MathOptInterface.set"><code>set</code></a>. The <a href="#MathOptInterface.submit"><code>submit</code></a> call returns a <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> indicating whether the provided solution was accepted or rejected.</p><p>This can be submitted only from the <a href="#MathOptInterface.HeuristicCallback"><code>HeuristicCallback</code></a>. The field <code>callback_data</code> is a solver-specific callback type that is passed as the argument to the heuristic callback.</p><p>Some solvers require a complete solution, others only partial solutions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L685-L699">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HeuristicSolutionStatus" href="#MathOptInterface.HeuristicSolutionStatus"><code>MathOptInterface.HeuristicSolutionStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HeuristicSolutionStatus</code></pre><p>An Enum of possible return values for <a href="#MathOptInterface.submit"><code>submit</code></a> with <a href="#MathOptInterface.HeuristicSolution"><code>HeuristicSolution</code></a>. This informs whether the heuristic solution was accepted or rejected.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED"><code>HEURISTIC_SOLUTION_ACCEPTED</code></a>: The heuristic solution was accepted</li><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_REJECTED"><code>HEURISTIC_SOLUTION_REJECTED</code></a>: The heuristic solution was rejected</li><li><a href="#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN"><code>HEURISTIC_SOLUTION_UNKNOWN</code></a>: No information available on the acceptance</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L682">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED" href="#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED"><code>MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_ACCEPTED::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_ACCEPTED</code>: The heuristic solution was accepted</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_REJECTED" href="#MathOptInterface.HEURISTIC_SOLUTION_REJECTED"><code>MathOptInterface.HEURISTIC_SOLUTION_REJECTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_REJECTED::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_REJECTED</code>: The heuristic solution was rejected</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN" href="#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN"><code>MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">HEURISTIC_SOLUTION_UNKNOWN::HeuristicSolutionStatus</code></pre><p>An instance of the <a href="#MathOptInterface.HeuristicSolutionStatus"><code>HeuristicSolutionStatus</code></a> enum.</p><p><code>HEURISTIC_SOLUTION_UNKNOWN</code>: No information available on the acceptance</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L668-L674">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nonlinear/">« Nonlinear programming</a><a class="docs-footer-nextpage" href="../errors/">Errors »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/constraints/index.html b/previews/PR3562/moi/reference/constraints/index.html
index 2e6986539e4..6d5d39d2259 100644
--- a/previews/PR3562/moi/reference/constraints/index.html
+++ b/previews/PR3562/moi/reference/constraints/index.html
@@ -5,12 +5,12 @@
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
 </script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/duality/">Duality</a></li><li><a class="tocitem" href="../../background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox" checked/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../standard_form/">Standard form</a></li><li><a class="tocitem" href="../models/">Models</a></li><li><a class="tocitem" href="../variables/">Variables</a></li><li class="is-active"><a class="tocitem" href>Constraints</a><ul class="internal"><li><a class="tocitem" href="#Types"><span>Types</span></a></li><li><a class="tocitem" href="#Functions"><span>Functions</span></a></li><li><a class="tocitem" href="#Attributes"><span>Attributes</span></a></li></ul></li><li><a class="tocitem" href="../modification/">Modifications</a></li><li><a class="tocitem" href="../nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../callbacks/">Callbacks</a></li><li><a class="tocitem" href="../errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">API Reference</a></li><li class="is-active"><a href>Constraints</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Constraints</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/constraints.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="constraints_ref"><a class="docs-heading-anchor" href="#constraints_ref">Constraints</a><a id="constraints_ref-1"></a><a class="docs-heading-anchor-permalink" href="#constraints_ref" title="Permalink"></a></h1><h2 id="Types"><a class="docs-heading-anchor" href="#Types">Types</a><a id="Types-1"></a><a class="docs-heading-anchor-permalink" href="#Types" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintIndex" href="#MathOptInterface.ConstraintIndex"><code>MathOptInterface.ConstraintIndex</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintIndex{F, S}</code></pre><p>A type-safe wrapper for <code>Int64</code> for use in referencing <code>F</code>-in-<code>S</code> constraints in a model. The parameter <code>F</code> is the type of the function in the constraint, and the parameter <code>S</code> is the type of set in the constraint. To allow for deletion, indices need not be consecutive. Indices within a constraint type (i.e. <code>F</code>-in-<code>S</code>) must be unique, but non-unique indices across different constraint types are allowed. If <code>F</code> is <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> then the index is equal to the index of the variable. That is for an <code>index::ConstraintIndex{VariableIndex}</code>, we always have</p><pre><code class="language-julia hljs">index.value == MOI.get(model, MOI.ConstraintFunction(), index).value</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L48-L63">source</a></section></article><h2 id="Functions"><a class="docs-heading-anchor" href="#Functions">Functions</a><a id="Functions-1"></a><a class="docs-heading-anchor-permalink" href="#Functions" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.ConstraintIndex}" href="#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.ConstraintIndex}"><code>MathOptInterface.is_valid</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">is_valid(model::ModelLike, index::Index)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether this index refers to a valid object in the model <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L95-L99">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.add_constraint" href="#MathOptInterface.add_constraint"><code>MathOptInterface.add_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_constraint(model::ModelLike, func::F, set::S)::ConstraintIndex{F,S} where {F,S}</code></pre><p>Add the constraint <span>$f(x) \in \mathcal{S}$</span> where <span>$f$</span> is defined by <code>func</code>, and <span>$\mathcal{S}$</span> is defined by <code>set</code>.</p><pre><code class="nohighlight hljs">add_constraint(model::ModelLike, v::VariableIndex, set::S)::ConstraintIndex{VariableIndex,S} where {S}
 add_constraint(model::ModelLike, vec::Vector{VariableIndex}, set::S)::ConstraintIndex{VectorOfVariables,S} where {S}</code></pre><p>Add the constraint <span>$v \in \mathcal{S}$</span> where <span>$v$</span> is the variable (or vector of variables) referenced by <code>v</code> and <span>$\mathcal{S}$</span> is defined by <code>set</code>.</p><ul><li>An <a href="../errors/#MathOptInterface.UnsupportedConstraint"><code>UnsupportedConstraint</code></a> error is thrown if <code>model</code> does not support <code>F</code>-in-<code>S</code> constraints,</li><li>a <a href="../errors/#MathOptInterface.AddConstraintNotAllowed"><code>AddConstraintNotAllowed</code></a> error is thrown if it supports <code>F</code>-in-<code>S</code> constraints but it cannot add the constraint(s) in its current state and</li><li>a <a href="../errors/#MathOptInterface.ScalarFunctionConstantNotZero"><code>ScalarFunctionConstantNotZero</code></a> error may be thrown if <code>func</code> is an <code>AbstractScalarFunction</code> with nonzero constant and <code>set</code> is <a href="../standard_form/#MathOptInterface.EqualTo"><code>EqualTo</code></a>, <a href="../standard_form/#MathOptInterface.GreaterThan"><code>GreaterThan</code></a>, <a href="../standard_form/#MathOptInterface.LessThan"><code>LessThan</code></a> or <a href="../standard_form/#MathOptInterface.Interval"><code>Interval</code></a>.</li><li>a <a href="../errors/#MathOptInterface.LowerBoundAlreadySet"><code>LowerBoundAlreadySet</code></a> error is thrown if <code>F</code> is a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> and a constraint was already added to this variable that sets a lower bound.</li><li>a <a href="../errors/#MathOptInterface.UpperBoundAlreadySet"><code>UpperBoundAlreadySet</code></a> error is thrown if <code>F</code> is a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> and a constraint was already added to this variable that sets an upper bound.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L125-L149">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.add_constraints" href="#MathOptInterface.add_constraints"><code>MathOptInterface.add_constraints</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">add_constraints(model::ModelLike, funcs::Vector{F}, sets::Vector{S})::Vector{ConstraintIndex{F,S}} where {F,S}</code></pre><p>Add the set of constraints specified by each function-set pair in <code>funcs</code> and <code>sets</code>. <code>F</code> and <code>S</code> should be concrete types. This call is equivalent to <code>add_constraint.(model, funcs, sets)</code> but may be more efficient.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L231-L236">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.transform" href="#MathOptInterface.transform"><code>MathOptInterface.transform</code></a> — <span class="docstring-category">Function</span></header><section><div><p><strong>Transform Constraint Set</strong></p><pre><code class="nohighlight hljs">transform(model::ModelLike, c::ConstraintIndex{F,S1}, newset::S2)::ConstraintIndex{F,S2}</code></pre><p>Replace the set in constraint <code>c</code> with <code>newset</code>. The constraint index <code>c</code> will no longer be valid, and the function returns a new constraint index with the correct type.</p><p>Solvers may only support a subset of constraint transforms that they perform efficiently (for example, changing from a <code>LessThan</code> to <code>GreaterThan</code> set). In addition, set modification (where <code>S1 = S2</code>) should be performed via the <code>modify</code> function.</p><p>Typically, the user should delete the constraint and add a new one.</p><p><strong>Examples</strong></p><p>If <code>c</code> is a <code>ConstraintIndex{ScalarAffineFunction{Float64},LessThan{Float64}}</code>,</p><pre><code class="language-julia hljs">c2 = transform(model, c, GreaterThan(0.0))
-transform(model, c, LessThan(0.0)) # errors</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L288-L313">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports_constraint" href="#MathOptInterface.supports_constraint"><code>MathOptInterface.supports_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">supports_constraint(
-    model::ModelLike,
-    ::Type{F},
-    ::Type{S},
-)::Bool where {F&lt;:AbstractFunction,S&lt;:AbstractSet}</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports <code>F</code>-in-<code>S</code> constraints, that is, <code>copy_to(model, src)</code> does not throw <a href="../errors/#MathOptInterface.UnsupportedConstraint"><code>UnsupportedConstraint</code></a> when <code>src</code> contains <code>F</code>-in-<code>S</code> constraints. If <code>F</code>-in-<code>S</code> constraints are only not supported in specific circumstances, e.g. <code>F</code>-in-<code>S</code> constraints cannot be combined with another type of constraint, it should still return <code>true</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L7-L19">source</a></section><section><div><pre><code class="language-julia hljs">MOI.supports_constraint(
+transform(model, c, LessThan(0.0)) # errors</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L288-L313">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports_constraint" href="#MathOptInterface.supports_constraint"><code>MathOptInterface.supports_constraint</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">MOI.supports_constraint(
     BT::Type{&lt;:AbstractBridge},
     F::Type{&lt;:MOI.AbstractFunction},
     S::Type{&lt;:MOI.AbstractSet},
-)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether the bridges of type <code>BT</code> support bridging <code>F</code>-in-<code>S</code> constraints.</p><p><strong>Implementation notes</strong></p><ul><li>This method depends only on the type of the inputs, not the runtime values.</li><li>There is a default fallback, so you need only implement this method for constraint types that the bridge implements.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/Constraint/bridge.jl#L22-L37">source</a></section></article><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.AbstractConstraintAttribute" href="#MathOptInterface.AbstractConstraintAttribute"><code>MathOptInterface.AbstractConstraintAttribute</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractConstraintAttribute</code></pre><p>Abstract supertype for attribute objects that can be used to set or get attributes (properties) of constraints in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L38-L43">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintName" href="#MathOptInterface.ConstraintName"><code>MathOptInterface.ConstraintName</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintName()</code></pre><p>A constraint attribute for a string identifying the constraint.</p><p>It is <em>valid</em> for constraints variables to have the same name; however, constraints with duplicate names cannot be looked up using <a href="../models/#MathOptInterface.get"><code>get</code></a>, regardless of whether they have the same <code>F</code>-in-<code>S</code> type.</p><p><code>ConstraintName</code> has a default value of <code>&quot;&quot;</code> if not set.</p><p><strong>Notes</strong></p><p>You should <em>not</em> implement <code>ConstraintName</code> for <code>VariableIndex</code> constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1555-L1569">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintPrimalStart" href="#MathOptInterface.ConstraintPrimalStart"><code>MathOptInterface.ConstraintPrimalStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintPrimalStart()</code></pre><p>A constraint attribute for the initial assignment to some constraint&#39;s <a href="#MathOptInterface.ConstraintPrimal"><code>ConstraintPrimal</code></a> that the optimizer may use to warm-start the solve.</p><p>May be <code>nothing</code> (unset), a number for <a href="../standard_form/#MathOptInterface.AbstractScalarFunction"><code>AbstractScalarFunction</code></a>, or a vector for <a href="../standard_form/#MathOptInterface.AbstractVectorFunction"><code>AbstractVectorFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1595-L1603">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintDualStart" href="#MathOptInterface.ConstraintDualStart"><code>MathOptInterface.ConstraintDualStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintDualStart()</code></pre><p>A constraint attribute for the initial assignment to some constraint&#39;s <a href="#MathOptInterface.ConstraintDual"><code>ConstraintDual</code></a> that the optimizer may use to warm-start the solve.</p><p>May be <code>nothing</code> (unset), a number for <a href="../standard_form/#MathOptInterface.AbstractScalarFunction"><code>AbstractScalarFunction</code></a>, or a vector for <a href="../standard_form/#MathOptInterface.AbstractVectorFunction"><code>AbstractVectorFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1606-L1614">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintPrimal" href="#MathOptInterface.ConstraintPrimal"><code>MathOptInterface.ConstraintPrimal</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintPrimal(result_index::Int = 1)</code></pre><p>A constraint attribute for the assignment to some constraint&#39;s primal value(s) in result <code>result_index</code>.</p><p>If the constraint is <code>f(x) in S</code>, then in most cases the <code>ConstraintPrimal</code> is the value of <code>f</code>, evaluated at the correspondng <a href="../variables/#MathOptInterface.VariablePrimal"><code>VariablePrimal</code></a> solution.</p><p>However, some conic solvers reformulate <code>b - Ax in S</code> to <code>s = b - Ax, s in S</code>. These solvers may return the value of <code>s</code> for <code>ConstraintPrimal</code>, rather than <code>b - Ax</code>. (Although these are constrained by an equality constraint, due to numerical tolerances they may not be identical.)</p><p>If the solver does not have a primal value for the constraint because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ConstraintPrimal</code> attribute.</p><p>If <code>result_index</code> is omitted, it is 1 by default. See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1617-L1642">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintDual" href="#MathOptInterface.ConstraintDual"><code>MathOptInterface.ConstraintDual</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintDual(result_index::Int = 1)</code></pre><p>A constraint attribute for the assignment to some constraint&#39;s dual value(s) in result <code>result_index</code>. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a dual value for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.DualStatus"><code>DualStatus</code></a> before accessing the <code>ConstraintDual</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1648-L1663">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintBasisStatus" href="#MathOptInterface.ConstraintBasisStatus"><code>MathOptInterface.ConstraintBasisStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintBasisStatus(result_index::Int = 1)</code></pre><p>A constraint attribute for the <code>BasisStatusCode</code> of some constraint in result <code>result_index</code>, with respect to an available optimal solution basis. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a basis statue for the constraint because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ConstraintBasisStatus</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p><strong>Notes</strong></p><p>For the basis status of a variable, query <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>.</p><p><code>ConstraintBasisStatus</code> does not apply to <code>VariableIndex</code> constraints. You can infer the basis status of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint by looking at the result of <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1669-L1693">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintFunction" href="#MathOptInterface.ConstraintFunction"><code>MathOptInterface.ConstraintFunction</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintFunction()</code></pre><p>A constraint attribute for the <code>AbstractFunction</code> object used to define the constraint.</p><p>It is guaranteed to be equivalent but not necessarily identical to the function provided by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1755-L1763">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.CanonicalConstraintFunction" href="#MathOptInterface.CanonicalConstraintFunction"><code>MathOptInterface.CanonicalConstraintFunction</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">CanonicalConstraintFunction()</code></pre><p>A constraint attribute for a canonical representation of the <a href="../standard_form/#MathOptInterface.AbstractFunction"><code>AbstractFunction</code></a> object used to define the constraint.</p><p>Getting this attribute is guaranteed to return a function that is equivalent but not necessarily identical to the function provided by the user.</p><p>By default, <code>MOI.get(model, MOI.CanonicalConstraintFunction(), ci)</code> fallbacks to <code>MOI.Utilities.canonical(MOI.get(model, MOI.ConstraintFunction(), ci))</code>. However, if <code>model</code> knows that the constraint function is canonical then it can implement a specialized method that directly return the function without calling <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.canonical"><code>Utilities.canonical</code></a>. Therefore, the value returned <strong>cannot</strong> be assumed to be a copy of the function stored in <code>model</code>. Moreover, <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.Model"><code>Utilities.Model</code></a> checks with <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.is_canonical"><code>Utilities.is_canonical</code></a> whether the function stored internally is already canonical and if it&#39;s the case, then it returns the function stored internally instead of a copy.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1712-L1730">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintSet" href="#MathOptInterface.ConstraintSet"><code>MathOptInterface.ConstraintSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintSet()</code></pre><p>A constraint attribute for the <code>AbstractSet</code> object used to define the constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1800-L1805">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BasisStatusCode" href="#MathOptInterface.BasisStatusCode"><code>MathOptInterface.BasisStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BasisStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.ConstraintBasisStatus"><code>ConstraintBasisStatus</code></a> and <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a> attributes, explaining the status of a given element with respect to an optimal solution basis.</p><p><strong>Notes</strong></p><ul><li><code>NONBASIC_AT_LOWER</code> and <code>NONBASIC_AT_UPPER</code> should be used only for</li></ul><p>constraints with the <code>Interval</code> set. In this case, they are necessary to distinguish which side of the constraint is active. One-sided constraints (e.g., <code>LessThan</code> and <code>GreaterThan</code>) should use <code>NONBASIC</code> instead of the <code>NONBASIC_AT_*</code> values. This restriction does not apply to <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>, which should return <code>NONBASIC_AT_*</code> regardless of whether the alternative bound exists.</p><ul><li>In linear programs, <code>SUPER_BASIC</code> occurs when a variable with no bounds is not</li></ul><p>in the basis.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.BASIC"><code>BASIC</code></a>: element is in the basis</li><li><a href="#MathOptInterface.NONBASIC"><code>NONBASIC</code></a>: element is not in the basis</li><li><a href="#MathOptInterface.NONBASIC_AT_LOWER"><code>NONBASIC_AT_LOWER</code></a>: element is not in the basis and is at its lower bound</li><li><a href="#MathOptInterface.NONBASIC_AT_UPPER"><code>NONBASIC_AT_UPPER</code></a>: element is not in the basis and is at its upper bound</li><li><a href="#MathOptInterface.SUPER_BASIC"><code>SUPER_BASIC</code></a>: element is not in the basis but is also not at one of its bounds</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1511">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BASIC" href="#MathOptInterface.BASIC"><code>MathOptInterface.BASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">BASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>BASIC</code>: element is in the basis</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC" href="#MathOptInterface.NONBASIC"><code>MathOptInterface.NONBASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC</code>: element is not in the basis</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC_AT_LOWER" href="#MathOptInterface.NONBASIC_AT_LOWER"><code>MathOptInterface.NONBASIC_AT_LOWER</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC_AT_LOWER::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC_AT_LOWER</code>: element is not in the basis and is at its lower bound</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC_AT_UPPER" href="#MathOptInterface.NONBASIC_AT_UPPER"><code>MathOptInterface.NONBASIC_AT_UPPER</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC_AT_UPPER::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC_AT_UPPER</code>: element is not in the basis and is at its upper bound</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SUPER_BASIC" href="#MathOptInterface.SUPER_BASIC"><code>MathOptInterface.SUPER_BASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SUPER_BASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>SUPER_BASIC</code>: element is not in the basis but is also not at one of its bounds</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../modification/">Modifications »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether the bridges of type <code>BT</code> support bridging <code>F</code>-in-<code>S</code> constraints.</p><p><strong>Implementation notes</strong></p><ul><li>This method depends only on the type of the inputs, not the runtime values.</li><li>There is a default fallback, so you need only implement this method for constraint types that the bridge implements.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/Constraint/bridge.jl#L22-L37">source</a></section><section><div><pre><code class="language-julia hljs">supports_constraint(
+    model::ModelLike,
+    ::Type{F},
+    ::Type{S},
+)::Bool where {F&lt;:AbstractFunction,S&lt;:AbstractSet}</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports <code>F</code>-in-<code>S</code> constraints, that is, <code>copy_to(model, src)</code> does not throw <a href="../errors/#MathOptInterface.UnsupportedConstraint"><code>UnsupportedConstraint</code></a> when <code>src</code> contains <code>F</code>-in-<code>S</code> constraints. If <code>F</code>-in-<code>S</code> constraints are only not supported in specific circumstances, e.g. <code>F</code>-in-<code>S</code> constraints cannot be combined with another type of constraint, it should still return <code>true</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/constraints.jl#L7-L19">source</a></section></article><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.AbstractConstraintAttribute" href="#MathOptInterface.AbstractConstraintAttribute"><code>MathOptInterface.AbstractConstraintAttribute</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractConstraintAttribute</code></pre><p>Abstract supertype for attribute objects that can be used to set or get attributes (properties) of constraints in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L38-L43">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintName" href="#MathOptInterface.ConstraintName"><code>MathOptInterface.ConstraintName</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintName()</code></pre><p>A constraint attribute for a string identifying the constraint.</p><p>It is <em>valid</em> for constraints variables to have the same name; however, constraints with duplicate names cannot be looked up using <a href="../models/#MathOptInterface.get"><code>get</code></a>, regardless of whether they have the same <code>F</code>-in-<code>S</code> type.</p><p><code>ConstraintName</code> has a default value of <code>&quot;&quot;</code> if not set.</p><p><strong>Notes</strong></p><p>You should <em>not</em> implement <code>ConstraintName</code> for <code>VariableIndex</code> constraints.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1555-L1569">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintPrimalStart" href="#MathOptInterface.ConstraintPrimalStart"><code>MathOptInterface.ConstraintPrimalStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintPrimalStart()</code></pre><p>A constraint attribute for the initial assignment to some constraint&#39;s <a href="#MathOptInterface.ConstraintPrimal"><code>ConstraintPrimal</code></a> that the optimizer may use to warm-start the solve.</p><p>May be <code>nothing</code> (unset), a number for <a href="../standard_form/#MathOptInterface.AbstractScalarFunction"><code>AbstractScalarFunction</code></a>, or a vector for <a href="../standard_form/#MathOptInterface.AbstractVectorFunction"><code>AbstractVectorFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1595-L1603">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintDualStart" href="#MathOptInterface.ConstraintDualStart"><code>MathOptInterface.ConstraintDualStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintDualStart()</code></pre><p>A constraint attribute for the initial assignment to some constraint&#39;s <a href="#MathOptInterface.ConstraintDual"><code>ConstraintDual</code></a> that the optimizer may use to warm-start the solve.</p><p>May be <code>nothing</code> (unset), a number for <a href="../standard_form/#MathOptInterface.AbstractScalarFunction"><code>AbstractScalarFunction</code></a>, or a vector for <a href="../standard_form/#MathOptInterface.AbstractVectorFunction"><code>AbstractVectorFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1606-L1614">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintPrimal" href="#MathOptInterface.ConstraintPrimal"><code>MathOptInterface.ConstraintPrimal</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintPrimal(result_index::Int = 1)</code></pre><p>A constraint attribute for the assignment to some constraint&#39;s primal value(s) in result <code>result_index</code>.</p><p>If the constraint is <code>f(x) in S</code>, then in most cases the <code>ConstraintPrimal</code> is the value of <code>f</code>, evaluated at the correspondng <a href="../variables/#MathOptInterface.VariablePrimal"><code>VariablePrimal</code></a> solution.</p><p>However, some conic solvers reformulate <code>b - Ax in S</code> to <code>s = b - Ax, s in S</code>. These solvers may return the value of <code>s</code> for <code>ConstraintPrimal</code>, rather than <code>b - Ax</code>. (Although these are constrained by an equality constraint, due to numerical tolerances they may not be identical.)</p><p>If the solver does not have a primal value for the constraint because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ConstraintPrimal</code> attribute.</p><p>If <code>result_index</code> is omitted, it is 1 by default. See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1617-L1642">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintDual" href="#MathOptInterface.ConstraintDual"><code>MathOptInterface.ConstraintDual</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintDual(result_index::Int = 1)</code></pre><p>A constraint attribute for the assignment to some constraint&#39;s dual value(s) in result <code>result_index</code>. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a dual value for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.DualStatus"><code>DualStatus</code></a> before accessing the <code>ConstraintDual</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1648-L1663">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintBasisStatus" href="#MathOptInterface.ConstraintBasisStatus"><code>MathOptInterface.ConstraintBasisStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintBasisStatus(result_index::Int = 1)</code></pre><p>A constraint attribute for the <code>BasisStatusCode</code> of some constraint in result <code>result_index</code>, with respect to an available optimal solution basis. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a basis statue for the constraint because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ConstraintBasisStatus</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p><strong>Notes</strong></p><p>For the basis status of a variable, query <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>.</p><p><code>ConstraintBasisStatus</code> does not apply to <code>VariableIndex</code> constraints. You can infer the basis status of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint by looking at the result of <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1669-L1693">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintFunction" href="#MathOptInterface.ConstraintFunction"><code>MathOptInterface.ConstraintFunction</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintFunction()</code></pre><p>A constraint attribute for the <code>AbstractFunction</code> object used to define the constraint.</p><p>It is guaranteed to be equivalent but not necessarily identical to the function provided by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1755-L1763">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.CanonicalConstraintFunction" href="#MathOptInterface.CanonicalConstraintFunction"><code>MathOptInterface.CanonicalConstraintFunction</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">CanonicalConstraintFunction()</code></pre><p>A constraint attribute for a canonical representation of the <a href="../standard_form/#MathOptInterface.AbstractFunction"><code>AbstractFunction</code></a> object used to define the constraint.</p><p>Getting this attribute is guaranteed to return a function that is equivalent but not necessarily identical to the function provided by the user.</p><p>By default, <code>MOI.get(model, MOI.CanonicalConstraintFunction(), ci)</code> fallbacks to <code>MOI.Utilities.canonical(MOI.get(model, MOI.ConstraintFunction(), ci))</code>. However, if <code>model</code> knows that the constraint function is canonical then it can implement a specialized method that directly return the function without calling <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.canonical"><code>Utilities.canonical</code></a>. Therefore, the value returned <strong>cannot</strong> be assumed to be a copy of the function stored in <code>model</code>. Moreover, <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.Model"><code>Utilities.Model</code></a> checks with <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.is_canonical"><code>Utilities.is_canonical</code></a> whether the function stored internally is already canonical and if it&#39;s the case, then it returns the function stored internally instead of a copy.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1712-L1730">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintSet" href="#MathOptInterface.ConstraintSet"><code>MathOptInterface.ConstraintSet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintSet()</code></pre><p>A constraint attribute for the <code>AbstractSet</code> object used to define the constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1800-L1805">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BasisStatusCode" href="#MathOptInterface.BasisStatusCode"><code>MathOptInterface.BasisStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BasisStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.ConstraintBasisStatus"><code>ConstraintBasisStatus</code></a> and <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a> attributes, explaining the status of a given element with respect to an optimal solution basis.</p><p><strong>Notes</strong></p><ul><li><code>NONBASIC_AT_LOWER</code> and <code>NONBASIC_AT_UPPER</code> should be used only for</li></ul><p>constraints with the <code>Interval</code> set. In this case, they are necessary to distinguish which side of the constraint is active. One-sided constraints (e.g., <code>LessThan</code> and <code>GreaterThan</code>) should use <code>NONBASIC</code> instead of the <code>NONBASIC_AT_*</code> values. This restriction does not apply to <a href="../variables/#MathOptInterface.VariableBasisStatus"><code>VariableBasisStatus</code></a>, which should return <code>NONBASIC_AT_*</code> regardless of whether the alternative bound exists.</p><ul><li>In linear programs, <code>SUPER_BASIC</code> occurs when a variable with no bounds is not</li></ul><p>in the basis.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.BASIC"><code>BASIC</code></a>: element is in the basis</li><li><a href="#MathOptInterface.NONBASIC"><code>NONBASIC</code></a>: element is not in the basis</li><li><a href="#MathOptInterface.NONBASIC_AT_LOWER"><code>NONBASIC_AT_LOWER</code></a>: element is not in the basis and is at its lower bound</li><li><a href="#MathOptInterface.NONBASIC_AT_UPPER"><code>NONBASIC_AT_UPPER</code></a>: element is not in the basis and is at its upper bound</li><li><a href="#MathOptInterface.SUPER_BASIC"><code>SUPER_BASIC</code></a>: element is not in the basis but is also not at one of its bounds</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1511">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BASIC" href="#MathOptInterface.BASIC"><code>MathOptInterface.BASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">BASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>BASIC</code>: element is in the basis</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC" href="#MathOptInterface.NONBASIC"><code>MathOptInterface.NONBASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC</code>: element is not in the basis</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC_AT_LOWER" href="#MathOptInterface.NONBASIC_AT_LOWER"><code>MathOptInterface.NONBASIC_AT_LOWER</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC_AT_LOWER::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC_AT_LOWER</code>: element is not in the basis and is at its lower bound</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NONBASIC_AT_UPPER" href="#MathOptInterface.NONBASIC_AT_UPPER"><code>MathOptInterface.NONBASIC_AT_UPPER</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NONBASIC_AT_UPPER::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>NONBASIC_AT_UPPER</code>: element is not in the basis and is at its upper bound</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SUPER_BASIC" href="#MathOptInterface.SUPER_BASIC"><code>MathOptInterface.SUPER_BASIC</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SUPER_BASIC::BasisStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.BasisStatusCode"><code>BasisStatusCode</code></a> enum.</p><p><code>SUPER_BASIC</code>: element is not in the basis but is also not at one of its bounds</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1482-L1488">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../variables/">« Variables</a><a class="docs-footer-nextpage" href="../modification/">Modifications »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/errors/index.html b/previews/PR3562/moi/reference/errors/index.html
index c2dc3e2571d..fea55a2a48f 100644
--- a/previews/PR3562/moi/reference/errors/index.html
+++ b/previews/PR3562/moi/reference/errors/index.html
@@ -49,4 +49,4 @@
 julia&gt; throw(MOI.UnsupportedNonlinearOperator(:black_box))
 ERROR: MathOptInterface.UnsupportedNonlinearOperator: The nonlinear operator `:black_box` is not supported by the model.
 Stacktrace:
-[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L366-L382">source</a></section></article><p>Note that setting the <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint is not allowed:</p><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SettingVariableIndexNotAllowed" href="#MathOptInterface.SettingVariableIndexNotAllowed"><code>MathOptInterface.SettingVariableIndexNotAllowed</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SettingVariableIndexNotAllowed()</code></pre><p>Error type that should be thrown when the user calls <a href="../models/#MathOptInterface.set"><code>set</code></a> to change the <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L604-L609">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Callbacks</a><a class="docs-footer-nextpage" href="../../submodules/Benchmarks/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+[...]</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L366-L382">source</a></section></article><p>Note that setting the <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint is not allowed:</p><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SettingVariableIndexNotAllowed" href="#MathOptInterface.SettingVariableIndexNotAllowed"><code>MathOptInterface.SettingVariableIndexNotAllowed</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SettingVariableIndexNotAllowed()</code></pre><p>Error type that should be thrown when the user calls <a href="../models/#MathOptInterface.set"><code>set</code></a> to change the <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> of a <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L604-L609">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Callbacks</a><a class="docs-footer-nextpage" href="../../submodules/Benchmarks/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/models/index.html b/previews/PR3562/moi/reference/models/index.html
index 8a22776a96f..7c456de273a 100644
--- a/previews/PR3562/moi/reference/models/index.html
+++ b/previews/PR3562/moi/reference/models/index.html
@@ -3,7 +3,11 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/duality/">Duality</a></li><li><a class="tocitem" href="../../background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox" checked/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../standard_form/">Standard form</a></li><li class="is-active"><a class="tocitem" href>Models</a><ul class="internal"><li><a class="tocitem" href="#Attribute-interface"><span>Attribute interface</span></a></li><li><a class="tocitem" href="#Model-interface"><span>Model interface</span></a></li><li><a class="tocitem" href="#Model-attributes"><span>Model attributes</span></a></li><li><a class="tocitem" href="#Optimizer-interface"><span>Optimizer interface</span></a></li><li><a class="tocitem" href="#Optimizer-attributes"><span>Optimizer attributes</span></a></li></ul></li><li><a class="tocitem" href="../variables/">Variables</a></li><li><a class="tocitem" href="../constraints/">Constraints</a></li><li><a class="tocitem" href="../modification/">Modifications</a></li><li><a class="tocitem" href="../nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../callbacks/">Callbacks</a></li><li><a class="tocitem" href="../errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">API Reference</a></li><li class="is-active"><a href>Models</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Models</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/models.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Models"><a class="docs-heading-anchor" href="#Models">Models</a><a id="Models-1"></a><a class="docs-heading-anchor-permalink" href="#Models" title="Permalink"></a></h1><h2 id="Attribute-interface"><a class="docs-heading-anchor" href="#Attribute-interface">Attribute interface</a><a id="Attribute-interface-1"></a><a class="docs-heading-anchor-permalink" href="#Attribute-interface" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_set_by_optimize" href="#MathOptInterface.is_set_by_optimize"><code>MathOptInterface.is_set_by_optimize</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_set_by_optimize(::AnyAttribute)</code></pre><p>Return a <code>Bool</code> indicating whether the value of the attribute is modified during an <a href="#MathOptInterface.optimize!"><code>optimize!</code></a> call, that is, the attribute is used to query the result of the optimization.</p><p><strong>Important note when defining new attributes</strong></p><p>This function returns <code>false</code> by default so it should be implemented for attributes that are modified by <a href="#MathOptInterface.optimize!"><code>optimize!</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2158-L2169">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_copyable" href="#MathOptInterface.is_copyable"><code>MathOptInterface.is_copyable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_copyable(::AnyAttribute)</code></pre><p>Return a <code>Bool</code> indicating whether the value of the attribute may be copied during <a href="#MathOptInterface.copy_to"><code>copy_to</code></a> using <a href="#MathOptInterface.set"><code>set</code></a>.</p><p><strong>Important note when defining new attributes</strong></p><p>By default <code>is_copyable(attr)</code> returns <code>!is_set_by_optimize(attr)</code>. A specific method should be defined for attributes which are copied indirectly during <a href="#MathOptInterface.copy_to"><code>copy_to</code></a>. For instance, both <code>is_copyable</code> and <a href="#MathOptInterface.is_set_by_optimize"><code>is_set_by_optimize</code></a> return <code>false</code> for the following attributes:</p><ul><li><a href="#MathOptInterface.ListOfOptimizerAttributesSet"><code>ListOfOptimizerAttributesSet</code></a>, <a href="#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a>, <a href="#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a> and <a href="#MathOptInterface.ListOfVariableAttributesSet"><code>ListOfVariableAttributesSet</code></a>.</li><li><a href="#MathOptInterface.SolverName"><code>SolverName</code></a> and <a href="#MathOptInterface.RawSolver"><code>RawSolver</code></a>: these attributes cannot be set.</li><li><a href="#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a> and <a href="#MathOptInterface.ListOfVariableIndices"><code>ListOfVariableIndices</code></a>: these attributes are set indirectly by <a href="../variables/#MathOptInterface.add_variable"><code>add_variable</code></a> and <a href="../variables/#MathOptInterface.add_variables"><code>add_variables</code></a>.</li><li><a href="#MathOptInterface.ObjectiveFunctionType"><code>ObjectiveFunctionType</code></a>: this attribute is set indirectly when setting the <a href="#MathOptInterface.ObjectiveFunction"><code>ObjectiveFunction</code></a> attribute.</li><li><a href="#MathOptInterface.NumberOfConstraints"><code>NumberOfConstraints</code></a>, <a href="#MathOptInterface.ListOfConstraintIndices"><code>ListOfConstraintIndices</code></a>, <a href="#MathOptInterface.ListOfConstraintTypesPresent"><code>ListOfConstraintTypesPresent</code></a>, <a href="../constraints/#MathOptInterface.CanonicalConstraintFunction"><code>CanonicalConstraintFunction</code></a>, <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> and <a href="../constraints/#MathOptInterface.ConstraintSet"><code>ConstraintSet</code></a>: these attributes are set indirectly by <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a> and <a href="../constraints/#MathOptInterface.add_constraints"><code>add_constraints</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2200-L2227">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.get" href="#MathOptInterface.get"><code>MathOptInterface.get</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute)</code></pre><p>Return an attribute <code>attr</code> of the optimizer <code>optimizer</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractModelAttribute)</code></pre><p>Return an attribute <code>attr</code> of the model <code>model</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex)</code></pre><p>If the attribute <code>attr</code> is set for the variable <code>v</code> in the model <code>model</code>, return its value, return <code>nothing</code> otherwise. If the attribute <code>attr</code> is not supported by <code>model</code> then an error should be thrown instead of returning <code>nothing</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractVariableAttribute, v::Vector{VariableIndex})</code></pre><p>Return a vector of attributes corresponding to each variable in the collection <code>v</code> in the model <code>model</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractConstraintAttribute, c::ConstraintIndex)</code></pre><p>If the attribute <code>attr</code> is set for the constraint <code>c</code> in the model <code>model</code>, return its value, return <code>nothing</code> otherwise. If the attribute <code>attr</code> is not supported by <code>model</code> then an error should be thrown instead of returning <code>nothing</code>.</p><pre><code class="language-julia hljs">get(
+</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox" checked/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/duality/">Duality</a></li><li><a class="tocitem" href="../../background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox" checked/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../standard_form/">Standard form</a></li><li class="is-active"><a class="tocitem" href>Models</a><ul class="internal"><li><a class="tocitem" href="#Attribute-interface"><span>Attribute interface</span></a></li><li><a class="tocitem" href="#Model-interface"><span>Model interface</span></a></li><li><a class="tocitem" href="#Model-attributes"><span>Model attributes</span></a></li><li><a class="tocitem" href="#Optimizer-interface"><span>Optimizer interface</span></a></li><li><a class="tocitem" href="#Optimizer-attributes"><span>Optimizer attributes</span></a></li></ul></li><li><a class="tocitem" href="../variables/">Variables</a></li><li><a class="tocitem" href="../constraints/">Constraints</a></li><li><a class="tocitem" href="../modification/">Modifications</a></li><li><a class="tocitem" href="../nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../callbacks/">Callbacks</a></li><li><a class="tocitem" href="../errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">MathOptInterface</a></li><li><a class="is-disabled">API Reference</a></li><li class="is-active"><a href>Models</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Models</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/models.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Models"><a class="docs-heading-anchor" href="#Models">Models</a><a id="Models-1"></a><a class="docs-heading-anchor-permalink" href="#Models" title="Permalink"></a></h1><h2 id="Attribute-interface"><a class="docs-heading-anchor" href="#Attribute-interface">Attribute interface</a><a id="Attribute-interface-1"></a><a class="docs-heading-anchor-permalink" href="#Attribute-interface" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_set_by_optimize" href="#MathOptInterface.is_set_by_optimize"><code>MathOptInterface.is_set_by_optimize</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_set_by_optimize(::AnyAttribute)</code></pre><p>Return a <code>Bool</code> indicating whether the value of the attribute is modified during an <a href="#MathOptInterface.optimize!"><code>optimize!</code></a> call, that is, the attribute is used to query the result of the optimization.</p><p><strong>Important note when defining new attributes</strong></p><p>This function returns <code>false</code> by default so it should be implemented for attributes that are modified by <a href="#MathOptInterface.optimize!"><code>optimize!</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2158-L2169">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_copyable" href="#MathOptInterface.is_copyable"><code>MathOptInterface.is_copyable</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_copyable(::AnyAttribute)</code></pre><p>Return a <code>Bool</code> indicating whether the value of the attribute may be copied during <a href="#MathOptInterface.copy_to"><code>copy_to</code></a> using <a href="#MathOptInterface.set"><code>set</code></a>.</p><p><strong>Important note when defining new attributes</strong></p><p>By default <code>is_copyable(attr)</code> returns <code>!is_set_by_optimize(attr)</code>. A specific method should be defined for attributes which are copied indirectly during <a href="#MathOptInterface.copy_to"><code>copy_to</code></a>. For instance, both <code>is_copyable</code> and <a href="#MathOptInterface.is_set_by_optimize"><code>is_set_by_optimize</code></a> return <code>false</code> for the following attributes:</p><ul><li><a href="#MathOptInterface.ListOfOptimizerAttributesSet"><code>ListOfOptimizerAttributesSet</code></a>, <a href="#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a>, <a href="#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a> and <a href="#MathOptInterface.ListOfVariableAttributesSet"><code>ListOfVariableAttributesSet</code></a>.</li><li><a href="#MathOptInterface.SolverName"><code>SolverName</code></a> and <a href="#MathOptInterface.RawSolver"><code>RawSolver</code></a>: these attributes cannot be set.</li><li><a href="#MathOptInterface.NumberOfVariables"><code>NumberOfVariables</code></a> and <a href="#MathOptInterface.ListOfVariableIndices"><code>ListOfVariableIndices</code></a>: these attributes are set indirectly by <a href="../variables/#MathOptInterface.add_variable"><code>add_variable</code></a> and <a href="../variables/#MathOptInterface.add_variables"><code>add_variables</code></a>.</li><li><a href="#MathOptInterface.ObjectiveFunctionType"><code>ObjectiveFunctionType</code></a>: this attribute is set indirectly when setting the <a href="#MathOptInterface.ObjectiveFunction"><code>ObjectiveFunction</code></a> attribute.</li><li><a href="#MathOptInterface.NumberOfConstraints"><code>NumberOfConstraints</code></a>, <a href="#MathOptInterface.ListOfConstraintIndices"><code>ListOfConstraintIndices</code></a>, <a href="#MathOptInterface.ListOfConstraintTypesPresent"><code>ListOfConstraintTypesPresent</code></a>, <a href="../constraints/#MathOptInterface.CanonicalConstraintFunction"><code>CanonicalConstraintFunction</code></a>, <a href="../constraints/#MathOptInterface.ConstraintFunction"><code>ConstraintFunction</code></a> and <a href="../constraints/#MathOptInterface.ConstraintSet"><code>ConstraintSet</code></a>: these attributes are set indirectly by <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a> and <a href="../constraints/#MathOptInterface.add_constraints"><code>add_constraints</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2200-L2227">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.get" href="#MathOptInterface.get"><code>MathOptInterface.get</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64</code></pre><p>Return the number of variables created by the bridge <code>b</code> in the model.</p><p>See also <a href="#MathOptInterface.NumberOfConstraints"><code>MOI.NumberOfConstraints</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this if the bridge adds new variables.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L42-L53">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)</code></pre><p>Return the list of variables created by the bridge <code>b</code>.</p><p>See also <a href="#MathOptInterface.ListOfVariableIndices"><code>MOI.ListOfVariableIndices</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this if the bridge adds new variables.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L56-L67">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}</code></pre><p>Return the number of constraints of the type <code>F</code>-in-<code>S</code> created by the bridge <code>b</code>.</p><p>See also <a href="#MathOptInterface.NumberOfConstraints"><code>MOI.NumberOfConstraints</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this for the constraint types returned by <a href="../../submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types"><code>added_constraint_types</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L70-L82">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}</code></pre><p>Return a  <code>Vector{ConstraintIndex{F,S}}</code> with indices of all constraints of type <code>F</code>-in-<code>S</code> created by the bride <code>b</code>.</p><p>See also <a href="#MathOptInterface.ListOfConstraintIndices"><code>MOI.ListOfConstraintIndices</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this for the constraint types returned by <a href="../../submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types"><code>added_constraint_types</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L85-L97">source</a></section><section><div><pre><code class="language-julia hljs">function MOI.get(
+    model::MOI.ModelLike,
+    attr::MOI.AbstractConstraintAttribute,
+    bridge::AbstractBridge,
+)</code></pre><p>Return the value of the attribute <code>attr</code> of the model <code>model</code> for the constraint bridged by <code>bridge</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L122-L131">source</a></section><section><div><pre><code class="language-julia hljs">get(model::GenericModel, attr::MathOptInterface.AbstractOptimizerAttribute)</code></pre><p>Return the value of the attribute <code>attr</code> from the model&#39;s MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L624-L628">source</a></section><section><div><pre><code class="language-julia hljs">get(model::GenericModel, attr::MathOptInterface.AbstractModelAttribute)</code></pre><p>Return the value of the attribute <code>attr</code> from the model&#39;s MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L666-L670">source</a></section><section><div><pre><code class="language-julia hljs">get(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute)</code></pre><p>Return an attribute <code>attr</code> of the optimizer <code>optimizer</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractModelAttribute)</code></pre><p>Return an attribute <code>attr</code> of the model <code>model</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex)</code></pre><p>If the attribute <code>attr</code> is set for the variable <code>v</code> in the model <code>model</code>, return its value, return <code>nothing</code> otherwise. If the attribute <code>attr</code> is not supported by <code>model</code> then an error should be thrown instead of returning <code>nothing</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractVariableAttribute, v::Vector{VariableIndex})</code></pre><p>Return a vector of attributes corresponding to each variable in the collection <code>v</code> in the model <code>model</code>.</p><pre><code class="language-julia hljs">get(model::ModelLike, attr::AbstractConstraintAttribute, c::ConstraintIndex)</code></pre><p>If the attribute <code>attr</code> is set for the constraint <code>c</code> in the model <code>model</code>, return its value, return <code>nothing</code> otherwise. If the attribute <code>attr</code> is not supported by <code>model</code> then an error should be thrown instead of returning <code>nothing</code>.</p><pre><code class="language-julia hljs">get(
     model::ModelLike,
     attr::AbstractConstraintAttribute,
     c::Vector{ConstraintIndex{F,S}},
@@ -11,11 +15,12 @@
     model::ModelLike,
     ::Type{ConstraintIndex{F,S}},
     name::String,
-) where {F,S}</code></pre><p>If an <code>F</code>-in-<code>S</code> constraint with name <code>name</code> exists in the model <code>model</code>, return the corresponding index, otherwise return <code>nothing</code>. Errors if two constraints have the same name.</p><pre><code class="language-julia hljs">get(model::ModelLike, ::Type{ConstraintIndex}, name::String)</code></pre><p>If <em>any</em> constraint with name <code>name</code> exists in the model <code>model</code>, return the corresponding index, otherwise return <code>nothing</code>. This version is available for convenience but may incur a performance penalty because it is not type stable. Errors if two constraints have the same name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L314-L388">source</a></section><section><div><pre><code class="language-julia hljs">get(model::GenericModel, attr::MathOptInterface.AbstractOptimizerAttribute)</code></pre><p>Return the value of the attribute <code>attr</code> from the model&#39;s MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L624-L628">source</a></section><section><div><pre><code class="language-julia hljs">get(model::GenericModel, attr::MathOptInterface.AbstractModelAttribute)</code></pre><p>Return the value of the attribute <code>attr</code> from the model&#39;s MOI backend.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L666-L670">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64</code></pre><p>Return the number of variables created by the bridge <code>b</code> in the model.</p><p>See also <a href="#MathOptInterface.NumberOfConstraints"><code>MOI.NumberOfConstraints</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this if the bridge adds new variables.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L42-L53">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)</code></pre><p>Return the list of variables created by the bridge <code>b</code>.</p><p>See also <a href="#MathOptInterface.ListOfVariableIndices"><code>MOI.ListOfVariableIndices</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this if the bridge adds new variables.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L56-L67">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}</code></pre><p>Return the number of constraints of the type <code>F</code>-in-<code>S</code> created by the bridge <code>b</code>.</p><p>See also <a href="#MathOptInterface.NumberOfConstraints"><code>MOI.NumberOfConstraints</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this for the constraint types returned by <a href="../../submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types"><code>added_constraint_types</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L70-L82">source</a></section><section><div><pre><code class="language-julia hljs">MOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}</code></pre><p>Return a  <code>Vector{ConstraintIndex{F,S}}</code> with indices of all constraints of type <code>F</code>-in-<code>S</code> created by the bride <code>b</code>.</p><p>See also <a href="#MathOptInterface.ListOfConstraintIndices"><code>MOI.ListOfConstraintIndices</code></a>.</p><p><strong>Implementation notes</strong></p><ul><li>There is a default fallback, so you need only implement this for the constraint types returned by <a href="../../submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types"><code>added_constraint_types</code></a>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L85-L97">source</a></section><section><div><pre><code class="language-julia hljs">function MOI.get(
+) where {F,S}</code></pre><p>If an <code>F</code>-in-<code>S</code> constraint with name <code>name</code> exists in the model <code>model</code>, return the corresponding index, otherwise return <code>nothing</code>. Errors if two constraints have the same name.</p><pre><code class="language-julia hljs">get(model::ModelLike, ::Type{ConstraintIndex}, name::String)</code></pre><p>If <em>any</em> constraint with name <code>name</code> exists in the model <code>model</code>, return the corresponding index, otherwise return <code>nothing</code>. This version is available for convenience but may incur a performance penalty because it is not type stable. Errors if two constraints have the same name.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L314-L388">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.get!" href="#MathOptInterface.get!"><code>MathOptInterface.get!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get!(output, model::ModelLike, args...)</code></pre><p>An in-place version of <a href="#MathOptInterface.get"><code>get</code></a>.</p><p>The signature matches that of <a href="#MathOptInterface.get"><code>get</code></a> except that the the result is placed in the vector <code>output</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L449-L456">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.set" href="#MathOptInterface.set"><code>MathOptInterface.set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">function MOI.set(
     model::MOI.ModelLike,
     attr::MOI.AbstractConstraintAttribute,
     bridge::AbstractBridge,
-)</code></pre><p>Return the value of the attribute <code>attr</code> of the model <code>model</code> for the constraint bridged by <code>bridge</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L122-L131">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.get!" href="#MathOptInterface.get!"><code>MathOptInterface.get!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">get!(output, model::ModelLike, args...)</code></pre><p>An in-place version of <a href="#MathOptInterface.get"><code>get</code></a>.</p><p>The signature matches that of <a href="#MathOptInterface.get"><code>get</code></a> except that the the result is placed in the vector <code>output</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L449-L456">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.set" href="#MathOptInterface.set"><code>MathOptInterface.set</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">set(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of the optimizer <code>optimizer</code>.</p><pre><code class="language-julia hljs">set(model::ModelLike, attr::AbstractModelAttribute, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of the model <code>model</code>.</p><pre><code class="language-julia hljs">set(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of variable <code>v</code> in model <code>model</code>.</p><pre><code class="language-julia hljs">set(
+    value,
+)</code></pre><p>Set the value of the attribute <code>attr</code> of the model <code>model</code> for the constraint bridged by <code>bridge</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L161-L171">source</a></section><section><div><pre><code class="language-julia hljs">set(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of the optimizer <code>optimizer</code>.</p><pre><code class="language-julia hljs">set(model::ModelLike, attr::AbstractModelAttribute, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of the model <code>model</code>.</p><pre><code class="language-julia hljs">set(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex, value)</code></pre><p>Assign <code>value</code> to the attribute <code>attr</code> of variable <code>v</code> in model <code>model</code>.</p><pre><code class="language-julia hljs">set(
     model::ModelLike,
     attr::AbstractVariableAttribute,
     v::Vector{VariableIndex},
@@ -40,12 +45,11 @@
     ::ConstraintFunction,
     c::ConstraintIndex{F,S},
     func::F,
-) where {F,S}</code></pre><p>Replace the function in constraint <code>c</code> with <code>func</code>. <code>F</code> must match the original function type used to define the constraint.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>Setting the constraint function is not allowed if <code>F</code> is <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a>; a <a href="../errors/#MathOptInterface.SettingVariableIndexNotAllowed"><code>SettingVariableIndexNotAllowed</code></a> error is thrown instead. This is because, it would require changing the index <code>c</code> since the index of <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraints must be the same as the index of the variable.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L462-L548">source</a></section><section><div><pre><code class="language-julia hljs">function MOI.set(
+) where {F,S}</code></pre><p>Replace the function in constraint <code>c</code> with <code>func</code>. <code>F</code> must match the original function type used to define the constraint.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>Setting the constraint function is not allowed if <code>F</code> is <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a>; a <a href="../errors/#MathOptInterface.SettingVariableIndexNotAllowed"><code>SettingVariableIndexNotAllowed</code></a> error is thrown instead. This is because, it would require changing the index <code>c</code> since the index of <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a> constraints must be the same as the index of the variable.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L462-L548">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports" href="#MathOptInterface.supports"><code>MathOptInterface.supports</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">MOI.supports(
     model::MOI.ModelLike,
     attr::MOI.AbstractConstraintAttribute,
-    bridge::AbstractBridge,
-    value,
-)</code></pre><p>Set the value of the attribute <code>attr</code> of the model <code>model</code> for the constraint bridged by <code>bridge</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L161-L171">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports" href="#MathOptInterface.supports"><code>MathOptInterface.supports</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">supports(model::ModelLike, sub::AbstractSubmittable)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the submittable <code>sub</code>.</p><pre><code class="language-julia hljs">supports(model::ModelLike, attr::AbstractOptimizerAttribute)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the optimizer attribute <code>attr</code>. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> shows a warning in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfOptimizerAttributesSet"><code>ListOfOptimizerAttributesSet</code></a> of <code>src</code>; see <a href="#MathOptInterface.copy_to"><code>copy_to</code></a> for more details on how unsupported optimizer attributes are handled in copy.</p><pre><code class="language-julia hljs">supports(model::ModelLike, attr::AbstractModelAttribute)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the model attribute <code>attr</code>. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> cannot be performed in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a> of <code>src</code>.</p><pre><code class="language-julia hljs">supports(
+    BT::Type{&lt;:AbstractBridge},
+)</code></pre><p>Return a <code>Bool</code> indicating whether <code>BT</code> supports setting <code>attr</code> to <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L105-L113">source</a></section><section><div><pre><code class="language-julia hljs">supports(model::ModelLike, sub::AbstractSubmittable)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the submittable <code>sub</code>.</p><pre><code class="language-julia hljs">supports(model::ModelLike, attr::AbstractOptimizerAttribute)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the optimizer attribute <code>attr</code>. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> shows a warning in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfOptimizerAttributesSet"><code>ListOfOptimizerAttributesSet</code></a> of <code>src</code>; see <a href="#MathOptInterface.copy_to"><code>copy_to</code></a> for more details on how unsupported optimizer attributes are handled in copy.</p><pre><code class="language-julia hljs">supports(model::ModelLike, attr::AbstractModelAttribute)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the model attribute <code>attr</code>. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> cannot be performed in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfModelAttributesSet"><code>ListOfModelAttributesSet</code></a> of <code>src</code>.</p><pre><code class="language-julia hljs">supports(
     model::ModelLike,
     attr::AbstractVariableAttribute,
     ::Type{VariableIndex},
@@ -53,11 +57,7 @@
     model::ModelLike,
     attr::AbstractConstraintAttribute,
     ::Type{ConstraintIndex{F,S}},
-)::Bool where {F,S}</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the constraint attribute <code>attr</code> applied to an <code>F</code>-in-<code>S</code> constraint. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> cannot be performed in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a> of <code>src</code>.</p><p>For all five methods, if the attribute is only not supported in specific circumstances, it should still return <code>true</code>.</p><p>Note that <code>supports</code> is only defined for attributes for which <a href="#MathOptInterface.is_copyable"><code>is_copyable</code></a> returns <code>true</code> as other attributes do not appear in the list of attributes set obtained by <code>ListOf...AttributesSet</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L220-L274">source</a></section><section><div><pre><code class="language-julia hljs">MOI.supports(
-    model::MOI.ModelLike,
-    attr::MOI.AbstractConstraintAttribute,
-    BT::Type{&lt;:AbstractBridge},
-)</code></pre><p>Return a <code>Bool</code> indicating whether <code>BT</code> supports setting <code>attr</code> to <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/bridge.jl#L105-L113">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.attribute_value_type" href="#MathOptInterface.attribute_value_type"><code>MathOptInterface.attribute_value_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">attribute_value_type(attr::AnyAttribute)</code></pre><p>Given an attribute <code>attr</code>, return the type of value expected by <a href="#MathOptInterface.get"><code>get</code></a>, or returned by <a href="#MathOptInterface.set"><code>set</code></a>.</p><p><strong>Notes</strong></p><ul><li>Only implement this if it make sense to do so. If un-implemented, the default is <code>Any</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L59-L69">source</a></section></article><h2 id="Model-interface"><a class="docs-heading-anchor" href="#Model-interface">Model interface</a><a id="Model-interface-1"></a><a class="docs-heading-anchor-permalink" href="#Model-interface" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ModelLike" href="#MathOptInterface.ModelLike"><code>MathOptInterface.ModelLike</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelLike</code></pre><p>Abstract supertype for objects that implement the &quot;Model&quot; interface for defining an optimization problem.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L11-L16">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_empty" href="#MathOptInterface.is_empty"><code>MathOptInterface.is_empty</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_empty(model::ModelLike)</code></pre><p>Returns <code>false</code> if the <code>model</code> has any model attribute set or has any variables or constraints.</p><p>Note that an empty model can have optimizer attributes set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L155-L162">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.empty!" href="#MathOptInterface.empty!"><code>MathOptInterface.empty!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">empty!(model::ModelLike)</code></pre><p>Empty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L165-L170">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.write_to_file" href="#MathOptInterface.write_to_file"><code>MathOptInterface.write_to_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">write_to_file(model::ModelLike, filename::String)</code></pre><p>Write the current model to the file at <code>filename</code>.</p><p>Supported file types depend on the model type.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L116-L122">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.read_from_file" href="#MathOptInterface.read_from_file"><code>MathOptInterface.read_from_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">read_from_file(model::ModelLike, filename::String)</code></pre><p>Read the file <code>filename</code> into the model <code>model</code>. If <code>model</code> is non-empty, this may throw an error.</p><p>Supported file types depend on the model type.</p><p><strong>Note</strong></p><p>Once the contents of the file are loaded into the model, users can query the variables via <code>get(model, ListOfVariableIndices())</code>. However, some filetypes, such as LP files, do not maintain an explicit ordering of the variables. Therefore, the returned list may be in an arbitrary order.</p><p>To avoid depending on the order of the indices, look up each variable index by name using  <code>get(model, VariableIndex, &quot;name&quot;)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L130-L147">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports_incremental_interface" href="#MathOptInterface.supports_incremental_interface"><code>MathOptInterface.supports_incremental_interface</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">supports_incremental_interface(model::ModelLike)</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports building incrementally via <a href="../variables/#MathOptInterface.add_variable"><code>add_variable</code></a> and <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a>.</p><p>The main purpose of this function is to determine whether a model can be loaded into <code>model</code> incrementally or whether it should be cached and copied at once instead.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L173-L182">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.copy_to" href="#MathOptInterface.copy_to"><code>MathOptInterface.copy_to</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_to(dest::ModelLike, src::ModelLike)::IndexMap</code></pre><p>Copy the model from <code>src</code> into <code>dest</code>.</p><p>The target <code>dest</code> is emptied, and all previous indices to variables and constraints in <code>dest</code> are invalidated.</p><p>Returns an <a href="#MathOptInterface.IndexMap"><code>IndexMap</code></a> object that translates variable and constraint indices from the <code>src</code> model to the corresponding indices in the <code>dest</code> model.</p><p><strong>Notes</strong></p><ul><li>If a constraint that in <code>src</code> is not supported by <code>dest</code>, then an <a href="../errors/#MathOptInterface.UnsupportedConstraint"><code>UnsupportedConstraint</code></a> error is thrown.</li><li>If an <a href="#MathOptInterface.AbstractModelAttribute"><code>AbstractModelAttribute</code></a>, <a href="../variables/#MathOptInterface.AbstractVariableAttribute"><code>AbstractVariableAttribute</code></a>, or <a href="../constraints/#MathOptInterface.AbstractConstraintAttribute"><code>AbstractConstraintAttribute</code></a> is set in <code>src</code> but not supported by <code>dest</code>, then an <a href="../errors/#MathOptInterface.UnsupportedAttribute"><code>UnsupportedAttribute</code></a> error is thrown.</li></ul><p><a href="#MathOptInterface.AbstractOptimizerAttribute"><code>AbstractOptimizerAttribute</code></a>s are <em>not</em> copied  to the <code>dest</code> model.</p><p><strong>IndexMap</strong></p><p>Implementations of <code>copy_to</code> must return an <a href="#MathOptInterface.IndexMap"><code>IndexMap</code></a>. For technical reasons, this type is defined in the Utilities submodule as <code>MOI.Utilities.IndexMap</code>. However, since it is an integral part of the MOI API, we provide <code>MOI.IndexMap</code> as an alias.</p><p><strong>Example</strong></p><pre><code class="language-julia hljs"># Given empty `ModelLike` objects `src` and `dest`.
+)::Bool where {F,S}</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports the constraint attribute <code>attr</code> applied to an <code>F</code>-in-<code>S</code> constraint. That is, it returns <code>false</code> if <code>copy_to(model, src)</code> cannot be performed in case <code>attr</code> is in the <a href="#MathOptInterface.ListOfConstraintAttributesSet"><code>ListOfConstraintAttributesSet</code></a> of <code>src</code>.</p><p>For all five methods, if the attribute is only not supported in specific circumstances, it should still return <code>true</code>.</p><p>Note that <code>supports</code> is only defined for attributes for which <a href="#MathOptInterface.is_copyable"><code>is_copyable</code></a> returns <code>true</code> as other attributes do not appear in the list of attributes set obtained by <code>ListOf...AttributesSet</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L220-L274">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.attribute_value_type" href="#MathOptInterface.attribute_value_type"><code>MathOptInterface.attribute_value_type</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">attribute_value_type(attr::AnyAttribute)</code></pre><p>Given an attribute <code>attr</code>, return the type of value expected by <a href="#MathOptInterface.get"><code>get</code></a>, or returned by <a href="#MathOptInterface.set"><code>set</code></a>.</p><p><strong>Notes</strong></p><ul><li>Only implement this if it make sense to do so. If un-implemented, the default is <code>Any</code>.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L59-L69">source</a></section></article><h2 id="Model-interface"><a class="docs-heading-anchor" href="#Model-interface">Model interface</a><a id="Model-interface-1"></a><a class="docs-heading-anchor-permalink" href="#Model-interface" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ModelLike" href="#MathOptInterface.ModelLike"><code>MathOptInterface.ModelLike</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelLike</code></pre><p>Abstract supertype for objects that implement the &quot;Model&quot; interface for defining an optimization problem.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L11-L16">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_empty" href="#MathOptInterface.is_empty"><code>MathOptInterface.is_empty</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_empty(model::ModelLike)</code></pre><p>Returns <code>false</code> if the <code>model</code> has any model attribute set or has any variables or constraints.</p><p>Note that an empty model can have optimizer attributes set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L155-L162">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.empty!" href="#MathOptInterface.empty!"><code>MathOptInterface.empty!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">empty!(model::ModelLike)</code></pre><p>Empty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L165-L170">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.write_to_file" href="#MathOptInterface.write_to_file"><code>MathOptInterface.write_to_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">write_to_file(model::ModelLike, filename::String)</code></pre><p>Write the current model to the file at <code>filename</code>.</p><p>Supported file types depend on the model type.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L116-L122">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.read_from_file" href="#MathOptInterface.read_from_file"><code>MathOptInterface.read_from_file</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">read_from_file(model::ModelLike, filename::String)</code></pre><p>Read the file <code>filename</code> into the model <code>model</code>. If <code>model</code> is non-empty, this may throw an error.</p><p>Supported file types depend on the model type.</p><p><strong>Note</strong></p><p>Once the contents of the file are loaded into the model, users can query the variables via <code>get(model, ListOfVariableIndices())</code>. However, some filetypes, such as LP files, do not maintain an explicit ordering of the variables. Therefore, the returned list may be in an arbitrary order.</p><p>To avoid depending on the order of the indices, look up each variable index by name using  <code>get(model, VariableIndex, &quot;name&quot;)</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L130-L147">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports_incremental_interface" href="#MathOptInterface.supports_incremental_interface"><code>MathOptInterface.supports_incremental_interface</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">supports_incremental_interface(model::ModelLike)</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports building incrementally via <a href="../variables/#MathOptInterface.add_variable"><code>add_variable</code></a> and <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a>.</p><p>The main purpose of this function is to determine whether a model can be loaded into <code>model</code> incrementally or whether it should be cached and copied at once instead.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L173-L182">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.copy_to" href="#MathOptInterface.copy_to"><code>MathOptInterface.copy_to</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">copy_to(dest::ModelLike, src::ModelLike)::IndexMap</code></pre><p>Copy the model from <code>src</code> into <code>dest</code>.</p><p>The target <code>dest</code> is emptied, and all previous indices to variables and constraints in <code>dest</code> are invalidated.</p><p>Returns an <a href="#MathOptInterface.IndexMap"><code>IndexMap</code></a> object that translates variable and constraint indices from the <code>src</code> model to the corresponding indices in the <code>dest</code> model.</p><p><strong>Notes</strong></p><ul><li>If a constraint that in <code>src</code> is not supported by <code>dest</code>, then an <a href="../errors/#MathOptInterface.UnsupportedConstraint"><code>UnsupportedConstraint</code></a> error is thrown.</li><li>If an <a href="#MathOptInterface.AbstractModelAttribute"><code>AbstractModelAttribute</code></a>, <a href="../variables/#MathOptInterface.AbstractVariableAttribute"><code>AbstractVariableAttribute</code></a>, or <a href="../constraints/#MathOptInterface.AbstractConstraintAttribute"><code>AbstractConstraintAttribute</code></a> is set in <code>src</code> but not supported by <code>dest</code>, then an <a href="../errors/#MathOptInterface.UnsupportedAttribute"><code>UnsupportedAttribute</code></a> error is thrown.</li></ul><p><a href="#MathOptInterface.AbstractOptimizerAttribute"><code>AbstractOptimizerAttribute</code></a>s are <em>not</em> copied  to the <code>dest</code> model.</p><p><strong>IndexMap</strong></p><p>Implementations of <code>copy_to</code> must return an <a href="#MathOptInterface.IndexMap"><code>IndexMap</code></a>. For technical reasons, this type is defined in the Utilities submodule as <code>MOI.Utilities.IndexMap</code>. However, since it is an integral part of the MOI API, we provide <code>MOI.IndexMap</code> as an alias.</p><p><strong>Example</strong></p><pre><code class="language-julia hljs"># Given empty `ModelLike` objects `src` and `dest`.
 
 x = add_variable(src)
 
@@ -133,4 +133,4 @@
 MOI.get(model, MOI.RelativeGapTolerance())  # returns 1e-3
 # ... and the relative gap of the obtained solution is smaller or equal to the
 # tolerance
-MOI.get(model, MOI.RelativeGap())  # should return something ≤ 1e-3</code></pre><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The mathematical definition of &quot;relative gap&quot;, and its allowed range, are solver-dependent. Typically, solvers expect a value between <code>0.0</code> and <code>1.0</code>.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L890-L919">source</a></section></article><p>List of attributes useful for optimizers</p><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TerminationStatus" href="#MathOptInterface.TerminationStatus"><code>MathOptInterface.TerminationStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatus()</code></pre><p>A model attribute for the <code>TerminationStatusCode</code> explaining why the optimizer stopped.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2038-L2043">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TerminationStatusCode" href="#MathOptInterface.TerminationStatusCode"><code>MathOptInterface.TerminationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatusCode</code></pre><p>An Enum of possible values for the <code>TerminationStatus</code> attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to <a href="#MathOptInterface.optimize!"><code>optimize!</code></a>.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a>: The algorithm has not started.</li><li><a href="#MathOptInterface.OPTIMAL"><code>OPTIMAL</code></a>: The algorithm found a globally optimal solution.</li><li><a href="#MathOptInterface.INFEASIBLE"><code>INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists.</li><li><a href="#MathOptInterface.DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</li><li><a href="#MathOptInterface.LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</li><li><a href="#MathOptInterface.LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</li><li><a href="#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</li><li><a href="#MathOptInterface.ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a>: The algorithm found a globally optimal solution to relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</li><li><a href="#MathOptInterface.ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a>: An iterative algorithm stopped after conducting the maximum number of iterations.</li><li><a href="#MathOptInterface.TIME_LIMIT"><code>TIME_LIMIT</code></a>: The algorithm stopped after a user-specified computation time.</li><li><a href="#MathOptInterface.NODE_LIMIT"><code>NODE_LIMIT</code></a>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</li><li><a href="#MathOptInterface.SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</li><li><a href="#MathOptInterface.MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a>: The algorithm stopped because it ran out of memory.</li><li><a href="#MathOptInterface.OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</li><li><a href="#MathOptInterface.NORM_LIMIT"><code>NORM_LIMIT</code></a>: The algorithm stopped because the norm of an iterate became too large.</li><li><a href="#MathOptInterface.OTHER_LIMIT"><code>OTHER_LIMIT</code></a>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</li><li><a href="#MathOptInterface.SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a>: The algorithm stopped because it was unable to continue making progress towards the solution.</li><li><a href="#MathOptInterface.NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a>: The algorithm stopped because it encountered unrecoverable numerical error.</li><li><a href="#MathOptInterface.INVALID_MODEL"><code>INVALID_MODEL</code></a>: The algorithm stopped because the model is invalid.</li><li><a href="#MathOptInterface.INVALID_OPTION"><code>INVALID_OPTION</code></a>: The algorithm stopped because it was provided an invalid option.</li><li><a href="#MathOptInterface.INTERRUPTED"><code>INTERRUPTED</code></a>: The algorithm stopped because of an interrupt signal.</li><li><a href="#MathOptInterface.OTHER_ERROR"><code>OTHER_ERROR</code></a>: The algorithm stopped because of an error not covered by one of the statuses defined above.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L2009">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OPTIMIZE_NOT_CALLED" href="#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>MathOptInterface.OPTIMIZE_NOT_CALLED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMIZE_NOT_CALLED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMIZE_NOT_CALLED</code>: The algorithm has not started.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OPTIMAL" href="#MathOptInterface.OPTIMAL"><code>MathOptInterface.OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMAL</code>: The algorithm found a globally optimal solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE" href="#MathOptInterface.INFEASIBLE"><code>MathOptInterface.INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE</code>: The algorithm concluded that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DUAL_INFEASIBLE" href="#MathOptInterface.DUAL_INFEASIBLE"><code>MathOptInterface.DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.LOCALLY_SOLVED" href="#MathOptInterface.LOCALLY_SOLVED"><code>MathOptInterface.LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.LOCALLY_INFEASIBLE" href="#MathOptInterface.LOCALLY_INFEASIBLE"><code>MathOptInterface.LOCALLY_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_INFEASIBLE</code>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE_OR_UNBOUNDED" href="#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>MathOptInterface.INFEASIBLE_OR_UNBOUNDED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE_OR_UNBOUNDED</code>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_OPTIMAL" href="#MathOptInterface.ALMOST_OPTIMAL"><code>MathOptInterface.ALMOST_OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_OPTIMAL</code>: The algorithm found a globally optimal solution to relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_INFEASIBLE" href="#MathOptInterface.ALMOST_INFEASIBLE"><code>MathOptInterface.ALMOST_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_INFEASIBLE</code>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_DUAL_INFEASIBLE" href="#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>MathOptInterface.ALMOST_DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_LOCALLY_SOLVED" href="#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>MathOptInterface.ALMOST_LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ITERATION_LIMIT" href="#MathOptInterface.ITERATION_LIMIT"><code>MathOptInterface.ITERATION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ITERATION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ITERATION_LIMIT</code>: An iterative algorithm stopped after conducting the maximum number of iterations.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TIME_LIMIT" href="#MathOptInterface.TIME_LIMIT"><code>MathOptInterface.TIME_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">TIME_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>TIME_LIMIT</code>: The algorithm stopped after a user-specified computation time.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NODE_LIMIT" href="#MathOptInterface.NODE_LIMIT"><code>MathOptInterface.NODE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NODE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NODE_LIMIT</code>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SOLUTION_LIMIT" href="#MathOptInterface.SOLUTION_LIMIT"><code>MathOptInterface.SOLUTION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SOLUTION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SOLUTION_LIMIT</code>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.MEMORY_LIMIT" href="#MathOptInterface.MEMORY_LIMIT"><code>MathOptInterface.MEMORY_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MEMORY_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>MEMORY_LIMIT</code>: The algorithm stopped because it ran out of memory.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OBJECTIVE_LIMIT" href="#MathOptInterface.OBJECTIVE_LIMIT"><code>MathOptInterface.OBJECTIVE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OBJECTIVE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OBJECTIVE_LIMIT</code>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NORM_LIMIT" href="#MathOptInterface.NORM_LIMIT"><code>MathOptInterface.NORM_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NORM_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NORM_LIMIT</code>: The algorithm stopped because the norm of an iterate became too large.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_LIMIT" href="#MathOptInterface.OTHER_LIMIT"><code>MathOptInterface.OTHER_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_LIMIT</code>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SLOW_PROGRESS" href="#MathOptInterface.SLOW_PROGRESS"><code>MathOptInterface.SLOW_PROGRESS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SLOW_PROGRESS::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SLOW_PROGRESS</code>: The algorithm stopped because it was unable to continue making progress towards the solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NUMERICAL_ERROR" href="#MathOptInterface.NUMERICAL_ERROR"><code>MathOptInterface.NUMERICAL_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NUMERICAL_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NUMERICAL_ERROR</code>: The algorithm stopped because it encountered unrecoverable numerical error.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INVALID_MODEL" href="#MathOptInterface.INVALID_MODEL"><code>MathOptInterface.INVALID_MODEL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_MODEL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_MODEL</code>: The algorithm stopped because the model is invalid.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INVALID_OPTION" href="#MathOptInterface.INVALID_OPTION"><code>MathOptInterface.INVALID_OPTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_OPTION::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_OPTION</code>: The algorithm stopped because it was provided an invalid option.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INTERRUPTED" href="#MathOptInterface.INTERRUPTED"><code>MathOptInterface.INTERRUPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INTERRUPTED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INTERRUPTED</code>: The algorithm stopped because of an interrupt signal.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_ERROR" href="#MathOptInterface.OTHER_ERROR"><code>MathOptInterface.OTHER_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_ERROR</code>: The algorithm stopped because of an error not covered by one of the statuses defined above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.PrimalStatus" href="#MathOptInterface.PrimalStatus"><code>MathOptInterface.PrimalStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">PrimalStatus(result_index::Int = 1)</code></pre><p>A model attribute for the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> of the primal result <code>result_index</code>. If <code>result_index</code> is omitted, it defaults to 1.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p>If <code>result_index</code> is larger than the value of <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> then <code>NO_SOLUTION</code> is returned.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2090-L2100">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DualStatus" href="#MathOptInterface.DualStatus"><code>MathOptInterface.DualStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">DualStatus(result_index::Int = 1)</code></pre><p>A model attribute for the <code>ResultStatusCode</code> of the dual result <code>result_index</code>. If <code>result_index</code> is omitted, it defaults to 1.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p>If <code>result_index</code> is larger than the value of <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> then <code>NO_SOLUTION</code> is returned.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2108-L2118">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.RawStatusString" href="#MathOptInterface.RawStatusString"><code>MathOptInterface.RawStatusString</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RawStatusString()</code></pre><p>A model attribute for a solver specific string explaining why the optimizer stopped.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2048-L2053">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ResultCount" href="#MathOptInterface.ResultCount"><code>MathOptInterface.ResultCount</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultCount()</code></pre><p>A model attribute for the number of results available.</p><p><strong>Order of solutions</strong></p><p>A number of attributes contain an index, <code>result_index</code>, which is used to refer to one of the available results. Thus, <code>result_index</code> must be an integer between <code>1</code> and the number of available results.</p><p>As a general rule, the first result (<code>result_index=1</code>) is the most important result (e.g., an optimal solution or an infeasibility certificate). Other results will typically be alternate solutions that the solver found during the search for the first result.</p><p>If a (local) optimal solution is available, i.e., <a href="#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a> is <code>OPTIMAL</code> or <code>LOCALLY_SOLVED</code>, the first result must correspond to the (locally) optimal solution. Other results may be alternative optimal solutions, or they may be other suboptimal solutions; use <a href="#MathOptInterface.ObjectiveValue"><code>ObjectiveValue</code></a> to distingiush between them.</p><p>If a primal or dual infeasibility certificate is available, i.e., <a href="#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a> is <code>INFEASIBLE</code> or <code>DUAL_INFEASIBLE</code> and the corresponding <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> or <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> is <code>INFEASIBILITY_CERTIFICATE</code>, then the first result must be a certificate. Other results may be alternate certificates, or infeasible points.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1357-L1384">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ObjectiveValue" href="#MathOptInterface.ObjectiveValue"><code>MathOptInterface.ObjectiveValue</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ObjectiveValue(result_index::Int = 1)</code></pre><p>A model attribute for the objective value of the primal solution <code>result_index</code>.</p><p>If the solver does not have a primal value for the objective because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ObjectiveValue</code> attribute.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1243-L1257">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DualObjectiveValue" href="#MathOptInterface.DualObjectiveValue"><code>MathOptInterface.DualObjectiveValue</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">DualObjectiveValue(result_index::Int = 1)</code></pre><p>A model attribute for the value of the objective function of the dual problem for the <code>result_index</code>th dual result.</p><p>If the solver does not have a dual value for the objective because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> before accessing the <code>DualObjectiveValue</code> attribute.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1263-L1278">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ObjectiveBound" href="#MathOptInterface.ObjectiveBound"><code>MathOptInterface.ObjectiveBound</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ObjectiveBound()</code></pre><p>A model attribute for the best known bound on the optimal objective value.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1284-L1288">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.RelativeGap" href="#MathOptInterface.RelativeGap"><code>MathOptInterface.RelativeGap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RelativeGap()</code></pre><p>A model attribute for the final relative optimality gap.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The definition of this gap is solver-dependent. However, most solvers implementing this attribute define the relative gap as some variation of <span>$\frac{|b-f|}{|f|}$</span>, where <span>$b$</span> is the best bound and <span>$f$</span> is the best feasible objective value.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1291-L1301">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SolveTimeSec" href="#MathOptInterface.SolveTimeSec"><code>MathOptInterface.SolveTimeSec</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SolveTimeSec()</code></pre><p>A model attribute for the total elapsed solution time (in seconds) as reported by the optimizer.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1306-L1311">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SimplexIterations" href="#MathOptInterface.SimplexIterations"><code>MathOptInterface.SimplexIterations</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SimplexIterations()</code></pre><p>A model attribute for the cumulative number of simplex iterations during the optimization process.</p><p>For a mixed-integer program (MIP), the return value is the total simplex iterations for all nodes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1316-L1324">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BarrierIterations" href="#MathOptInterface.BarrierIterations"><code>MathOptInterface.BarrierIterations</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BarrierIterations()</code></pre><p>A model attribute for the cumulative number of barrier iterations while solving a problem.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1329-L1334">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NodeCount" href="#MathOptInterface.NodeCount"><code>MathOptInterface.NodeCount</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NodeCount()</code></pre><p>A model attribute for the total number of branch-and-bound nodes explored while solving a mixed-integer program (MIP).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1339-L1344">source</a></section></article><h3 id="ResultStatusCode"><a class="docs-heading-anchor" href="#ResultStatusCode">ResultStatusCode</a><a id="ResultStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#ResultStatusCode" title="Permalink"></a></h3><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ResultStatusCode" href="#MathOptInterface.ResultStatusCode"><code>MathOptInterface.ResultStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> and <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> attributes.</p><p>The values indicate how to interpret the result vector.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.NO_SOLUTION"><code>NO_SOLUTION</code></a>: the result vector is empty.</li><li><a href="#MathOptInterface.FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a>: the result vector is a feasible point.</li><li><a href="#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a>: the result vector is feasible if some constraint tolerances are relaxed.</li><li><a href="#MathOptInterface.INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a>: the result vector is an infeasible point.</li><li><a href="#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</li><li><a href="#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for a certificate of infeasibility.</li><li><a href="#MathOptInterface.REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</li><li><a href="#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for an ill-posed certificate.</li><li><a href="#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a>: the result vector contains a solution with an unknown interpretation.</li><li><a href="#MathOptInterface.OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2080">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NO_SOLUTION" href="#MathOptInterface.NO_SOLUTION"><code>MathOptInterface.NO_SOLUTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NO_SOLUTION::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NO_SOLUTION</code>: the result vector is empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.FEASIBLE_POINT" href="#MathOptInterface.FEASIBLE_POINT"><code>MathOptInterface.FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>FEASIBLE_POINT</code>: the result vector is a feasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_FEASIBLE_POINT" href="#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>MathOptInterface.NEARLY_FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_FEASIBLE_POINT</code>: the result vector is feasible if some constraint tolerances are relaxed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE_POINT" href="#MathOptInterface.INFEASIBLE_POINT"><code>MathOptInterface.INFEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBLE_POINT</code>: the result vector is an infeasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBILITY_CERTIFICATE" href="#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>MathOptInterface.INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBILITY_CERTIFICATE</code>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE" href="#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_INFEASIBILITY_CERTIFICATE</code>: the result satisfies a relaxed criterion for a certificate of infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.REDUCTION_CERTIFICATE" href="#MathOptInterface.REDUCTION_CERTIFICATE"><code>MathOptInterface.REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>REDUCTION_CERTIFICATE</code>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_REDUCTION_CERTIFICATE" href="#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>MathOptInterface.NEARLY_REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_REDUCTION_CERTIFICATE</code>: the result satisfies a relaxed criterion for an ill-posed certificate.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.UNKNOWN_RESULT_STATUS" href="#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>MathOptInterface.UNKNOWN_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">UNKNOWN_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>UNKNOWN_RESULT_STATUS</code>: the result vector contains a solution with an unknown interpretation.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_RESULT_STATUS" href="#MathOptInterface.OTHER_RESULT_STATUS"><code>MathOptInterface.OTHER_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>OTHER_RESULT_STATUS</code>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><h3 id="Conflict-Status"><a class="docs-heading-anchor" href="#Conflict-Status">Conflict Status</a><a id="Conflict-Status-1"></a><a class="docs-heading-anchor-permalink" href="#Conflict-Status" title="Permalink"></a></h3><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.compute_conflict!" href="#MathOptInterface.compute_conflict!"><code>MathOptInterface.compute_conflict!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">compute_conflict!(optimizer::AbstractOptimizer)</code></pre><p>Computes a minimal subset of constraints such that the model with the other constraint removed is still infeasible.</p><p>Some solvers call a set of conflicting constraints an Irreducible Inconsistent Subsystem (IIS).</p><p>See also <a href="#MathOptInterface.ConflictStatus"><code>ConflictStatus</code></a> and <a href="#MathOptInterface.ConstraintConflictStatus"><code>ConstraintConflictStatus</code></a>.</p><p><strong>Note</strong></p><p>If the model is modified after a call to <code>compute_conflict!</code>, the implementor is not obliged to purge the conflict. Any calls to the above attributes may return values for the original conflict without a warning. Similarly, when modifying the model, the conflict can be discarded.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L89-L106">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictStatus" href="#MathOptInterface.ConflictStatus"><code>MathOptInterface.ConflictStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictStatus()</code></pre><p>A model attribute for the <a href="#MathOptInterface.ConflictStatusCode"><code>ConflictStatusCode</code></a> explaining why the conflict refiner stopped when computing the conflict.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1410-L1415">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintConflictStatus" href="#MathOptInterface.ConstraintConflictStatus"><code>MathOptInterface.ConstraintConflictStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintConflictStatus()</code></pre><p>A constraint attribute indicating whether the constraint participates in the conflict. Its type is <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1860-L1865">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictStatusCode" href="#MathOptInterface.ConflictStatusCode"><code>MathOptInterface.ConflictStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictStatusCode</code></pre><p>An Enum of possible values for the <code>ConflictStatus</code> attribute. This attribute is meant to explain the reason why the conflict finder stopped executing in the most recent call to <a href="#MathOptInterface.compute_conflict!"><code>compute_conflict!</code></a>.</p><p>Possible values are:</p><ul><li><code>COMPUTE_CONFLICT_NOT_CALLED</code>: the function <a href="#MathOptInterface.compute_conflict!"><code>compute_conflict!</code></a> has not yet been called</li><li><code>NO_CONFLICT_EXISTS</code>: there is no conflict because the problem is feasible</li><li><code>NO_CONFLICT_FOUND</code>: the solver could not find a conflict</li><li><code>CONFLICT_FOUND</code>: at least one conflict could be found</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1389-L1402">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictParticipationStatusCode" href="#MathOptInterface.ConflictParticipationStatusCode"><code>MathOptInterface.ConflictParticipationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictParticipationStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.ConstraintConflictStatus"><code>ConstraintConflictStatus</code></a> attribute. This attribute is meant to indicate whether a given constraint participates or not in the last computed conflict.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.NOT_IN_CONFLICT"><code>NOT_IN_CONFLICT</code></a>: the constraint does not participate in the conflict</li><li><a href="#MathOptInterface.IN_CONFLICT"><code>IN_CONFLICT</code></a>: the constraint participates in the conflict</li><li><a href="#MathOptInterface.MAYBE_IN_CONFLICT"><code>MAYBE_IN_CONFLICT</code></a>: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1857">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NOT_IN_CONFLICT" href="#MathOptInterface.NOT_IN_CONFLICT"><code>MathOptInterface.NOT_IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NOT_IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>NOT_IN_CONFLICT</code>: the constraint does not participate in the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.IN_CONFLICT" href="#MathOptInterface.IN_CONFLICT"><code>MathOptInterface.IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>IN_CONFLICT</code>: the constraint participates in the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.MAYBE_IN_CONFLICT" href="#MathOptInterface.MAYBE_IN_CONFLICT"><code>MathOptInterface.MAYBE_IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MAYBE_IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>MAYBE_IN_CONFLICT</code>: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../standard_form/">« Standard form</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+MOI.get(model, MOI.RelativeGap())  # should return something ≤ 1e-3</code></pre><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The mathematical definition of &quot;relative gap&quot;, and its allowed range, are solver-dependent. Typically, solvers expect a value between <code>0.0</code> and <code>1.0</code>.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L890-L919">source</a></section></article><p>List of attributes useful for optimizers</p><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TerminationStatus" href="#MathOptInterface.TerminationStatus"><code>MathOptInterface.TerminationStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatus()</code></pre><p>A model attribute for the <code>TerminationStatusCode</code> explaining why the optimizer stopped.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2038-L2043">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TerminationStatusCode" href="#MathOptInterface.TerminationStatusCode"><code>MathOptInterface.TerminationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">TerminationStatusCode</code></pre><p>An Enum of possible values for the <code>TerminationStatus</code> attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to <a href="#MathOptInterface.optimize!"><code>optimize!</code></a>.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>OPTIMIZE_NOT_CALLED</code></a>: The algorithm has not started.</li><li><a href="#MathOptInterface.OPTIMAL"><code>OPTIMAL</code></a>: The algorithm found a globally optimal solution.</li><li><a href="#MathOptInterface.INFEASIBLE"><code>INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists.</li><li><a href="#MathOptInterface.DUAL_INFEASIBLE"><code>DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</li><li><a href="#MathOptInterface.LOCALLY_SOLVED"><code>LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</li><li><a href="#MathOptInterface.LOCALLY_INFEASIBLE"><code>LOCALLY_INFEASIBLE</code></a>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</li><li><a href="#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>INFEASIBLE_OR_UNBOUNDED</code></a>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</li><li><a href="#MathOptInterface.ALMOST_OPTIMAL"><code>ALMOST_OPTIMAL</code></a>: The algorithm found a globally optimal solution to relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_INFEASIBLE"><code>ALMOST_INFEASIBLE</code></a>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>ALMOST_DUAL_INFEASIBLE</code></a>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</li><li><a href="#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>ALMOST_LOCALLY_SOLVED</code></a>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</li><li><a href="#MathOptInterface.ITERATION_LIMIT"><code>ITERATION_LIMIT</code></a>: An iterative algorithm stopped after conducting the maximum number of iterations.</li><li><a href="#MathOptInterface.TIME_LIMIT"><code>TIME_LIMIT</code></a>: The algorithm stopped after a user-specified computation time.</li><li><a href="#MathOptInterface.NODE_LIMIT"><code>NODE_LIMIT</code></a>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</li><li><a href="#MathOptInterface.SOLUTION_LIMIT"><code>SOLUTION_LIMIT</code></a>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</li><li><a href="#MathOptInterface.MEMORY_LIMIT"><code>MEMORY_LIMIT</code></a>: The algorithm stopped because it ran out of memory.</li><li><a href="#MathOptInterface.OBJECTIVE_LIMIT"><code>OBJECTIVE_LIMIT</code></a>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</li><li><a href="#MathOptInterface.NORM_LIMIT"><code>NORM_LIMIT</code></a>: The algorithm stopped because the norm of an iterate became too large.</li><li><a href="#MathOptInterface.OTHER_LIMIT"><code>OTHER_LIMIT</code></a>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</li><li><a href="#MathOptInterface.SLOW_PROGRESS"><code>SLOW_PROGRESS</code></a>: The algorithm stopped because it was unable to continue making progress towards the solution.</li><li><a href="#MathOptInterface.NUMERICAL_ERROR"><code>NUMERICAL_ERROR</code></a>: The algorithm stopped because it encountered unrecoverable numerical error.</li><li><a href="#MathOptInterface.INVALID_MODEL"><code>INVALID_MODEL</code></a>: The algorithm stopped because the model is invalid.</li><li><a href="#MathOptInterface.INVALID_OPTION"><code>INVALID_OPTION</code></a>: The algorithm stopped because it was provided an invalid option.</li><li><a href="#MathOptInterface.INTERRUPTED"><code>INTERRUPTED</code></a>: The algorithm stopped because of an interrupt signal.</li><li><a href="#MathOptInterface.OTHER_ERROR"><code>OTHER_ERROR</code></a>: The algorithm stopped because of an error not covered by one of the statuses defined above.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L2009">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OPTIMIZE_NOT_CALLED" href="#MathOptInterface.OPTIMIZE_NOT_CALLED"><code>MathOptInterface.OPTIMIZE_NOT_CALLED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMIZE_NOT_CALLED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMIZE_NOT_CALLED</code>: The algorithm has not started.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OPTIMAL" href="#MathOptInterface.OPTIMAL"><code>MathOptInterface.OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OPTIMAL</code>: The algorithm found a globally optimal solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE" href="#MathOptInterface.INFEASIBLE"><code>MathOptInterface.INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE</code>: The algorithm concluded that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DUAL_INFEASIBLE" href="#MathOptInterface.DUAL_INFEASIBLE"><code>MathOptInterface.DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.LOCALLY_SOLVED" href="#MathOptInterface.LOCALLY_SOLVED"><code>MathOptInterface.LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.LOCALLY_INFEASIBLE" href="#MathOptInterface.LOCALLY_INFEASIBLE"><code>MathOptInterface.LOCALLY_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">LOCALLY_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>LOCALLY_INFEASIBLE</code>: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE_OR_UNBOUNDED" href="#MathOptInterface.INFEASIBLE_OR_UNBOUNDED"><code>MathOptInterface.INFEASIBLE_OR_UNBOUNDED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INFEASIBLE_OR_UNBOUNDED</code>: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_OPTIMAL" href="#MathOptInterface.ALMOST_OPTIMAL"><code>MathOptInterface.ALMOST_OPTIMAL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_OPTIMAL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_OPTIMAL</code>: The algorithm found a globally optimal solution to relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_INFEASIBLE" href="#MathOptInterface.ALMOST_INFEASIBLE"><code>MathOptInterface.ALMOST_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_INFEASIBLE</code>: The algorithm concluded that no feasible solution exists within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_DUAL_INFEASIBLE" href="#MathOptInterface.ALMOST_DUAL_INFEASIBLE"><code>MathOptInterface.ALMOST_DUAL_INFEASIBLE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_DUAL_INFEASIBLE::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_DUAL_INFEASIBLE</code>: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ALMOST_LOCALLY_SOLVED" href="#MathOptInterface.ALMOST_LOCALLY_SOLVED"><code>MathOptInterface.ALMOST_LOCALLY_SOLVED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ALMOST_LOCALLY_SOLVED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ALMOST_LOCALLY_SOLVED</code>: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ITERATION_LIMIT" href="#MathOptInterface.ITERATION_LIMIT"><code>MathOptInterface.ITERATION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">ITERATION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>ITERATION_LIMIT</code>: An iterative algorithm stopped after conducting the maximum number of iterations.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.TIME_LIMIT" href="#MathOptInterface.TIME_LIMIT"><code>MathOptInterface.TIME_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">TIME_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>TIME_LIMIT</code>: The algorithm stopped after a user-specified computation time.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NODE_LIMIT" href="#MathOptInterface.NODE_LIMIT"><code>MathOptInterface.NODE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NODE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NODE_LIMIT</code>: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SOLUTION_LIMIT" href="#MathOptInterface.SOLUTION_LIMIT"><code>MathOptInterface.SOLUTION_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SOLUTION_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SOLUTION_LIMIT</code>: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.MEMORY_LIMIT" href="#MathOptInterface.MEMORY_LIMIT"><code>MathOptInterface.MEMORY_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MEMORY_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>MEMORY_LIMIT</code>: The algorithm stopped because it ran out of memory.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OBJECTIVE_LIMIT" href="#MathOptInterface.OBJECTIVE_LIMIT"><code>MathOptInterface.OBJECTIVE_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OBJECTIVE_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OBJECTIVE_LIMIT</code>: The algorithm stopped because it found a solution better than a minimum limit set by the user.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NORM_LIMIT" href="#MathOptInterface.NORM_LIMIT"><code>MathOptInterface.NORM_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NORM_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NORM_LIMIT</code>: The algorithm stopped because the norm of an iterate became too large.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_LIMIT" href="#MathOptInterface.OTHER_LIMIT"><code>MathOptInterface.OTHER_LIMIT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_LIMIT::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_LIMIT</code>: The algorithm stopped due to a limit not covered by one of the <code>_LIMIT_</code> statuses above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SLOW_PROGRESS" href="#MathOptInterface.SLOW_PROGRESS"><code>MathOptInterface.SLOW_PROGRESS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">SLOW_PROGRESS::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>SLOW_PROGRESS</code>: The algorithm stopped because it was unable to continue making progress towards the solution.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NUMERICAL_ERROR" href="#MathOptInterface.NUMERICAL_ERROR"><code>MathOptInterface.NUMERICAL_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NUMERICAL_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>NUMERICAL_ERROR</code>: The algorithm stopped because it encountered unrecoverable numerical error.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INVALID_MODEL" href="#MathOptInterface.INVALID_MODEL"><code>MathOptInterface.INVALID_MODEL</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_MODEL::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_MODEL</code>: The algorithm stopped because the model is invalid.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INVALID_OPTION" href="#MathOptInterface.INVALID_OPTION"><code>MathOptInterface.INVALID_OPTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INVALID_OPTION::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INVALID_OPTION</code>: The algorithm stopped because it was provided an invalid option.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INTERRUPTED" href="#MathOptInterface.INTERRUPTED"><code>MathOptInterface.INTERRUPTED</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INTERRUPTED::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>INTERRUPTED</code>: The algorithm stopped because of an interrupt signal.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_ERROR" href="#MathOptInterface.OTHER_ERROR"><code>MathOptInterface.OTHER_ERROR</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_ERROR::TerminationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.TerminationStatusCode"><code>TerminationStatusCode</code></a> enum.</p><p><code>OTHER_ERROR</code>: The algorithm stopped because of an error not covered by one of the statuses defined above.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1973-L1979">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.PrimalStatus" href="#MathOptInterface.PrimalStatus"><code>MathOptInterface.PrimalStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">PrimalStatus(result_index::Int = 1)</code></pre><p>A model attribute for the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> of the primal result <code>result_index</code>. If <code>result_index</code> is omitted, it defaults to 1.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p>If <code>result_index</code> is larger than the value of <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> then <code>NO_SOLUTION</code> is returned.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2090-L2100">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DualStatus" href="#MathOptInterface.DualStatus"><code>MathOptInterface.DualStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">DualStatus(result_index::Int = 1)</code></pre><p>A model attribute for the <code>ResultStatusCode</code> of the dual result <code>result_index</code>. If <code>result_index</code> is omitted, it defaults to 1.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p><p>If <code>result_index</code> is larger than the value of <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> then <code>NO_SOLUTION</code> is returned.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2108-L2118">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.RawStatusString" href="#MathOptInterface.RawStatusString"><code>MathOptInterface.RawStatusString</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RawStatusString()</code></pre><p>A model attribute for a solver specific string explaining why the optimizer stopped.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2048-L2053">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ResultCount" href="#MathOptInterface.ResultCount"><code>MathOptInterface.ResultCount</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultCount()</code></pre><p>A model attribute for the number of results available.</p><p><strong>Order of solutions</strong></p><p>A number of attributes contain an index, <code>result_index</code>, which is used to refer to one of the available results. Thus, <code>result_index</code> must be an integer between <code>1</code> and the number of available results.</p><p>As a general rule, the first result (<code>result_index=1</code>) is the most important result (e.g., an optimal solution or an infeasibility certificate). Other results will typically be alternate solutions that the solver found during the search for the first result.</p><p>If a (local) optimal solution is available, i.e., <a href="#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a> is <code>OPTIMAL</code> or <code>LOCALLY_SOLVED</code>, the first result must correspond to the (locally) optimal solution. Other results may be alternative optimal solutions, or they may be other suboptimal solutions; use <a href="#MathOptInterface.ObjectiveValue"><code>ObjectiveValue</code></a> to distingiush between them.</p><p>If a primal or dual infeasibility certificate is available, i.e., <a href="#MathOptInterface.TerminationStatus"><code>TerminationStatus</code></a> is <code>INFEASIBLE</code> or <code>DUAL_INFEASIBLE</code> and the corresponding <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> or <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> is <code>INFEASIBILITY_CERTIFICATE</code>, then the first result must be a certificate. Other results may be alternate certificates, or infeasible points.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1357-L1384">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ObjectiveValue" href="#MathOptInterface.ObjectiveValue"><code>MathOptInterface.ObjectiveValue</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ObjectiveValue(result_index::Int = 1)</code></pre><p>A model attribute for the objective value of the primal solution <code>result_index</code>.</p><p>If the solver does not have a primal value for the objective because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>ObjectiveValue</code> attribute.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1243-L1257">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.DualObjectiveValue" href="#MathOptInterface.DualObjectiveValue"><code>MathOptInterface.DualObjectiveValue</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">DualObjectiveValue(result_index::Int = 1)</code></pre><p>A model attribute for the value of the objective function of the dual problem for the <code>result_index</code>th dual result.</p><p>If the solver does not have a dual value for the objective because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> before accessing the <code>DualObjectiveValue</code> attribute.</p><p>See <a href="#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1263-L1278">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ObjectiveBound" href="#MathOptInterface.ObjectiveBound"><code>MathOptInterface.ObjectiveBound</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ObjectiveBound()</code></pre><p>A model attribute for the best known bound on the optimal objective value.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1284-L1288">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.RelativeGap" href="#MathOptInterface.RelativeGap"><code>MathOptInterface.RelativeGap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RelativeGap()</code></pre><p>A model attribute for the final relative optimality gap.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The definition of this gap is solver-dependent. However, most solvers implementing this attribute define the relative gap as some variation of <span>$\frac{|b-f|}{|f|}$</span>, where <span>$b$</span> is the best bound and <span>$f$</span> is the best feasible objective value.</p></div></div></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1291-L1301">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SolveTimeSec" href="#MathOptInterface.SolveTimeSec"><code>MathOptInterface.SolveTimeSec</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SolveTimeSec()</code></pre><p>A model attribute for the total elapsed solution time (in seconds) as reported by the optimizer.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1306-L1311">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.SimplexIterations" href="#MathOptInterface.SimplexIterations"><code>MathOptInterface.SimplexIterations</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SimplexIterations()</code></pre><p>A model attribute for the cumulative number of simplex iterations during the optimization process.</p><p>For a mixed-integer program (MIP), the return value is the total simplex iterations for all nodes.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1316-L1324">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.BarrierIterations" href="#MathOptInterface.BarrierIterations"><code>MathOptInterface.BarrierIterations</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">BarrierIterations()</code></pre><p>A model attribute for the cumulative number of barrier iterations while solving a problem.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1329-L1334">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NodeCount" href="#MathOptInterface.NodeCount"><code>MathOptInterface.NodeCount</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">NodeCount()</code></pre><p>A model attribute for the total number of branch-and-bound nodes explored while solving a mixed-integer program (MIP).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1339-L1344">source</a></section></article><h3 id="ResultStatusCode"><a class="docs-heading-anchor" href="#ResultStatusCode">ResultStatusCode</a><a id="ResultStatusCode-1"></a><a class="docs-heading-anchor-permalink" href="#ResultStatusCode" title="Permalink"></a></h3><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ResultStatusCode" href="#MathOptInterface.ResultStatusCode"><code>MathOptInterface.ResultStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ResultStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> and <a href="#MathOptInterface.DualStatus"><code>DualStatus</code></a> attributes.</p><p>The values indicate how to interpret the result vector.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.NO_SOLUTION"><code>NO_SOLUTION</code></a>: the result vector is empty.</li><li><a href="#MathOptInterface.FEASIBLE_POINT"><code>FEASIBLE_POINT</code></a>: the result vector is a feasible point.</li><li><a href="#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>NEARLY_FEASIBLE_POINT</code></a>: the result vector is feasible if some constraint tolerances are relaxed.</li><li><a href="#MathOptInterface.INFEASIBLE_POINT"><code>INFEASIBLE_POINT</code></a>: the result vector is an infeasible point.</li><li><a href="#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>INFEASIBILITY_CERTIFICATE</code></a>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</li><li><a href="#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>NEARLY_INFEASIBILITY_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for a certificate of infeasibility.</li><li><a href="#MathOptInterface.REDUCTION_CERTIFICATE"><code>REDUCTION_CERTIFICATE</code></a>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</li><li><a href="#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>NEARLY_REDUCTION_CERTIFICATE</code></a>: the result satisfies a relaxed criterion for an ill-posed certificate.</li><li><a href="#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>UNKNOWN_RESULT_STATUS</code></a>: the result vector contains a solution with an unknown interpretation.</li><li><a href="#MathOptInterface.OTHER_RESULT_STATUS"><code>OTHER_RESULT_STATUS</code></a>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2080">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NO_SOLUTION" href="#MathOptInterface.NO_SOLUTION"><code>MathOptInterface.NO_SOLUTION</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NO_SOLUTION::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NO_SOLUTION</code>: the result vector is empty.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.FEASIBLE_POINT" href="#MathOptInterface.FEASIBLE_POINT"><code>MathOptInterface.FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>FEASIBLE_POINT</code>: the result vector is a feasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_FEASIBLE_POINT" href="#MathOptInterface.NEARLY_FEASIBLE_POINT"><code>MathOptInterface.NEARLY_FEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_FEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_FEASIBLE_POINT</code>: the result vector is feasible if some constraint tolerances are relaxed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBLE_POINT" href="#MathOptInterface.INFEASIBLE_POINT"><code>MathOptInterface.INFEASIBLE_POINT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBLE_POINT::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBLE_POINT</code>: the result vector is an infeasible point.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.INFEASIBILITY_CERTIFICATE" href="#MathOptInterface.INFEASIBILITY_CERTIFICATE"><code>MathOptInterface.INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>INFEASIBILITY_CERTIFICATE</code>: the result vector is an infeasibility certificate. If the <code>PrimalStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the primal result vector is a certificate of dual infeasibility. If the <code>DualStatus</code> is <code>INFEASIBILITY_CERTIFICATE</code>, then the dual result vector is a proof of primal infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE" href="#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE"><code>MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_INFEASIBILITY_CERTIFICATE</code>: the result satisfies a relaxed criterion for a certificate of infeasibility.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.REDUCTION_CERTIFICATE" href="#MathOptInterface.REDUCTION_CERTIFICATE"><code>MathOptInterface.REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>REDUCTION_CERTIFICATE</code>: the result vector is an ill-posed certificate; see <a href="https://arxiv.org/abs/1408.4685">this article</a> for details. If the <code>PrimalStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the primal result vector is a proof that the dual problem is ill-posed. If the <code>DualStatus</code> is <code>REDUCTION_CERTIFICATE</code>, then the dual result vector is a proof that the primal is ill-posed.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NEARLY_REDUCTION_CERTIFICATE" href="#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE"><code>MathOptInterface.NEARLY_REDUCTION_CERTIFICATE</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>NEARLY_REDUCTION_CERTIFICATE</code>: the result satisfies a relaxed criterion for an ill-posed certificate.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.UNKNOWN_RESULT_STATUS" href="#MathOptInterface.UNKNOWN_RESULT_STATUS"><code>MathOptInterface.UNKNOWN_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">UNKNOWN_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>UNKNOWN_RESULT_STATUS</code>: the result vector contains a solution with an unknown interpretation.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.OTHER_RESULT_STATUS" href="#MathOptInterface.OTHER_RESULT_STATUS"><code>MathOptInterface.OTHER_RESULT_STATUS</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">OTHER_RESULT_STATUS::ResultStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ResultStatusCode"><code>ResultStatusCode</code></a> enum.</p><p><code>OTHER_RESULT_STATUS</code>: the result vector contains a solution with an interpretation not covered by one of the statuses defined above</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L2058-L2064">source</a></section></article><h3 id="Conflict-Status"><a class="docs-heading-anchor" href="#Conflict-Status">Conflict Status</a><a id="Conflict-Status-1"></a><a class="docs-heading-anchor-permalink" href="#Conflict-Status" title="Permalink"></a></h3><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.compute_conflict!" href="#MathOptInterface.compute_conflict!"><code>MathOptInterface.compute_conflict!</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">compute_conflict!(optimizer::AbstractOptimizer)</code></pre><p>Computes a minimal subset of constraints such that the model with the other constraint removed is still infeasible.</p><p>Some solvers call a set of conflicting constraints an Irreducible Inconsistent Subsystem (IIS).</p><p>See also <a href="#MathOptInterface.ConflictStatus"><code>ConflictStatus</code></a> and <a href="#MathOptInterface.ConstraintConflictStatus"><code>ConstraintConflictStatus</code></a>.</p><p><strong>Note</strong></p><p>If the model is modified after a call to <code>compute_conflict!</code>, the implementor is not obliged to purge the conflict. Any calls to the above attributes may return values for the original conflict without a warning. Similarly, when modifying the model, the conflict can be discarded.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/MathOptInterface.jl#L89-L106">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictStatus" href="#MathOptInterface.ConflictStatus"><code>MathOptInterface.ConflictStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictStatus()</code></pre><p>A model attribute for the <a href="#MathOptInterface.ConflictStatusCode"><code>ConflictStatusCode</code></a> explaining why the conflict refiner stopped when computing the conflict.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1410-L1415">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConstraintConflictStatus" href="#MathOptInterface.ConstraintConflictStatus"><code>MathOptInterface.ConstraintConflictStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConstraintConflictStatus()</code></pre><p>A constraint attribute indicating whether the constraint participates in the conflict. Its type is <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1860-L1865">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictStatusCode" href="#MathOptInterface.ConflictStatusCode"><code>MathOptInterface.ConflictStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictStatusCode</code></pre><p>An Enum of possible values for the <code>ConflictStatus</code> attribute. This attribute is meant to explain the reason why the conflict finder stopped executing in the most recent call to <a href="#MathOptInterface.compute_conflict!"><code>compute_conflict!</code></a>.</p><p>Possible values are:</p><ul><li><code>COMPUTE_CONFLICT_NOT_CALLED</code>: the function <a href="#MathOptInterface.compute_conflict!"><code>compute_conflict!</code></a> has not yet been called</li><li><code>NO_CONFLICT_EXISTS</code>: there is no conflict because the problem is feasible</li><li><code>NO_CONFLICT_FOUND</code>: the solver could not find a conflict</li><li><code>CONFLICT_FOUND</code>: at least one conflict could be found</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1389-L1402">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ConflictParticipationStatusCode" href="#MathOptInterface.ConflictParticipationStatusCode"><code>MathOptInterface.ConflictParticipationStatusCode</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ConflictParticipationStatusCode</code></pre><p>An Enum of possible values for the <a href="#MathOptInterface.ConstraintConflictStatus"><code>ConstraintConflictStatus</code></a> attribute. This attribute is meant to indicate whether a given constraint participates or not in the last computed conflict.</p><p><strong>Values</strong></p><p>Possible values are:</p><ul><li><a href="#MathOptInterface.NOT_IN_CONFLICT"><code>NOT_IN_CONFLICT</code></a>: the constraint does not participate in the conflict</li><li><a href="#MathOptInterface.IN_CONFLICT"><code>IN_CONFLICT</code></a>: the constraint participates in the conflict</li><li><a href="#MathOptInterface.MAYBE_IN_CONFLICT"><code>MAYBE_IN_CONFLICT</code></a>: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1857">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.NOT_IN_CONFLICT" href="#MathOptInterface.NOT_IN_CONFLICT"><code>MathOptInterface.NOT_IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">NOT_IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>NOT_IN_CONFLICT</code>: the constraint does not participate in the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.IN_CONFLICT" href="#MathOptInterface.IN_CONFLICT"><code>MathOptInterface.IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>IN_CONFLICT</code>: the constraint participates in the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.MAYBE_IN_CONFLICT" href="#MathOptInterface.MAYBE_IN_CONFLICT"><code>MathOptInterface.MAYBE_IN_CONFLICT</code></a> — <span class="docstring-category">Constant</span></header><section><div><pre><code class="language-julia hljs">MAYBE_IN_CONFLICT::ConflictParticipationStatusCode</code></pre><p>An instance of the <a href="#MathOptInterface.ConflictParticipationStatusCode"><code>ConflictParticipationStatusCode</code></a> enum.</p><p><code>MAYBE_IN_CONFLICT</code>: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1843-L1849">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../standard_form/">« Standard form</a><a class="docs-footer-nextpage" href="../variables/">Variables »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/modification/index.html b/previews/PR3562/moi/reference/modification/index.html
index 5305e5dbd2e..707f1530a89 100644
--- a/previews/PR3562/moi/reference/modification/index.html
+++ b/previews/PR3562/moi/reference/modification/index.html
@@ -28,4 +28,4 @@
 )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/modifications.jl#L67-L150">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.AbstractFunctionModification" href="#MathOptInterface.AbstractFunctionModification"><code>MathOptInterface.AbstractFunctionModification</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractFunctionModification</code></pre><p>An abstract supertype for structs which specify partial modifications to functions, to be used for making small modifications instead of replacing the functions entirely.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L739-L745">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ScalarConstantChange" href="#MathOptInterface.ScalarConstantChange"><code>MathOptInterface.ScalarConstantChange</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarConstantChange{T}(new_constant::T)</code></pre><p>A struct used to request a change in the constant term of a scalar-valued function.</p><p>Applicable to <a href="../standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a> and <a href="../standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L748-L755">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VectorConstantChange" href="#MathOptInterface.VectorConstantChange"><code>MathOptInterface.VectorConstantChange</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VectorConstantChange{T}(new_constant::Vector{T})</code></pre><p>A struct used to request a change in the constant vector of a vector-valued function.</p><p>Applicable to <a href="../standard_form/#MathOptInterface.VectorAffineFunction"><code>VectorAffineFunction</code></a> and <a href="../standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L760-L767">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.ScalarCoefficientChange" href="#MathOptInterface.ScalarCoefficientChange"><code>MathOptInterface.ScalarCoefficientChange</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ScalarCoefficientChange{T}(variable::VariableIndex, new_coefficient::T)</code></pre><p>A struct used to request a change in the linear coefficient of a single variable in a scalar-valued function.</p><p>Applicable to <a href="../standard_form/#MathOptInterface.ScalarAffineFunction"><code>ScalarAffineFunction</code></a> and <a href="../standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>ScalarQuadraticFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L772-L779">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.MultirowChange" href="#MathOptInterface.MultirowChange"><code>MathOptInterface.MultirowChange</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">MultirowChange{T}(
     variable::VariableIndex,
     new_coefficients::Vector{Tuple{Int64,T}},
-) where {T}</code></pre><p>A struct used to request a change in the linear coefficients of a single variable in a vector-valued function.</p><p>New coefficients are specified by <code>(output_index, coefficient)</code> tuples.</p><p>Applicable to <a href="../standard_form/#MathOptInterface.VectorAffineFunction"><code>VectorAffineFunction</code></a> and <a href="../standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L793-L805">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../nonlinear/">Nonlinear programming »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+) where {T}</code></pre><p>A struct used to request a change in the linear coefficients of a single variable in a vector-valued function.</p><p>New coefficients are specified by <code>(output_index, coefficient)</code> tuples.</p><p>Applicable to <a href="../standard_form/#MathOptInterface.VectorAffineFunction"><code>VectorAffineFunction</code></a> and <a href="../standard_form/#MathOptInterface.VectorQuadraticFunction"><code>VectorQuadraticFunction</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/functions.jl#L793-L805">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraints/">« Constraints</a><a class="docs-footer-nextpage" href="../nonlinear/">Nonlinear programming »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/nonlinear/index.html b/previews/PR3562/moi/reference/nonlinear/index.html
index e2c16890046..3cc05408c75 100644
--- a/previews/PR3562/moi/reference/nonlinear/index.html
+++ b/previews/PR3562/moi/reference/nonlinear/index.html
@@ -79,4 +79,4 @@
 )</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/nlp.jl#L454-L495">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.constraint_expr" href="#MathOptInterface.constraint_expr"><code>MathOptInterface.constraint_expr</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">constraint_expr(d::AbstractNLPEvaluator, i::Integer)::Expr</code></pre><p>Returns a Julia <code>Expr</code> object representing the expression graph for the <span>$i\text{th}$</span> nonlinear constraint.</p><p><strong>Format</strong></p><p>The format is the same as <a href="#MathOptInterface.objective_expr"><code>objective_expr</code></a>, with an additional comparison operator indicating the sense of and bounds on the constraint.</p><p>For single-sided comparisons, the body of the constraint must be on the left-hand side, and the right-hand side must be a constant.</p><p>For double-sided comparisons (that is, <span>$l \le f(x) \le u$</span>), the body of the constraint must be in the middle, and the left- and right-hand sides must be constants.</p><p>The bounds on the constraints must match the <a href="#MathOptInterface.NLPBoundsPair"><code>NLPBoundsPair</code></a>s passed to <a href="#MathOptInterface.NLPBlockData"><code>NLPBlockData</code></a>.</p><p><strong>Examples</strong></p><pre><code class="language-julia hljs">:(x[MOI.VariableIndex(1)]^2 &lt;= 1.0)
 :(x[MOI.VariableIndex(1)]^2 &gt;= 2.0)
 :(x[MOI.VariableIndex(1)]^2 == 3.0)
-:(4.0 &lt;= x[MOI.VariableIndex(1)]^2 &lt;= 5.0)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/nlp.jl#L498-L527">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../modification/">« Modifications</a><a class="docs-footer-nextpage" href="../callbacks/">Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+:(4.0 &lt;= x[MOI.VariableIndex(1)]^2 &lt;= 5.0)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/nlp.jl#L498-L527">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../modification/">« Modifications</a><a class="docs-footer-nextpage" href="../callbacks/">Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/standard_form/index.html b/previews/PR3562/moi/reference/standard_form/index.html
index 2286105175a..7e2616ed6df 100644
--- a/previews/PR3562/moi/reference/standard_form/index.html
+++ b/previews/PR3562/moi/reference/standard_form/index.html
@@ -990,4 +990,4 @@
            MOI.VectorOfVariables([t; vec(X)]),
            MOI.RootDetConeSquare(2),
        )
-MathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeSquare}(1)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/sets.jl#L1619-L1656">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/naming_conventions/">« Naming conventions</a><a class="docs-footer-nextpage" href="../models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+MathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeSquare}(1)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/sets.jl#L1619-L1656">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/naming_conventions/">« Naming conventions</a><a class="docs-footer-nextpage" href="../models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/reference/variables/index.html b/previews/PR3562/moi/reference/variables/index.html
index 7febbd56b12..7f4aed6ec76 100644
--- a/previews/PR3562/moi/reference/variables/index.html
+++ b/previews/PR3562/moi/reference/variables/index.html
@@ -25,4 +25,4 @@
 )::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports constraining a variable to belong to a set of type <code>S</code> either on creation of the variable with <a href="#MathOptInterface.add_constrained_variable"><code>add_constrained_variable</code></a> or after the variable is created with <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a>.</p><p>By default, this function falls back to <code>supports_add_constrained_variables(model, Reals) &amp;&amp; supports_constraint(model, MOI.VariableIndex, S)</code> which is the correct definition for most models.</p><p><strong>Example</strong></p><p>Suppose that a solver supports only two kind of variables: binary variables and continuous variables with a lower bound. If the solver decides not to support <code>VariableIndex</code>-in-<code>Binary</code> and <code>VariableIndex</code>-in-<code>GreaterThan</code> constraints, it only has to implement <code>add_constrained_variable</code> for these two sets which prevents the user to add both a binary constraint and a lower bound on the same variable. Moreover, if the user adds a <code>VariableIndex</code>-in-<code>GreaterThan</code> constraint, implementing this interface (i.e., <code>supports_add_constrained_variables</code>) enables the constraint to be transparently bridged into a supported constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/variables.jl#L45-L72">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.supports_add_constrained_variables" href="#MathOptInterface.supports_add_constrained_variables"><code>MathOptInterface.supports_add_constrained_variables</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">supports_add_constrained_variables(
     model::ModelLike,
     S::Type{&lt;:AbstractVectorSet}
-)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports constraining a vector of variables to belong to a set of type <code>S</code> either on creation of the vector of variables with <a href="#MathOptInterface.add_constrained_variables"><code>add_constrained_variables</code></a> or after the variable is created with <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a>.</p><p>By default, if <code>S</code> is <code>Reals</code> then this function returns <code>true</code> and otherwise, it falls back to <code>supports_add_constrained_variables(model, Reals) &amp;&amp; supports_constraint(model, MOI.VectorOfVariables, S)</code> which is the correct definition for most models.</p><p><strong>Example</strong></p><p>In the standard conic form (see <a href="../../background/duality/#Duality">Duality</a>), the variables are grouped into several cones and the constraints are affine equality constraints. If <code>Reals</code> is not one of the cones supported by the solvers then it needs to implement <code>supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false</code> as free variables are not supported. The solvers should then implement <code>supports_add_constrained_variables(::Optimizer, ::Type{&lt;:SupportedCones}) = true</code> where <code>SupportedCones</code> is the union of all cone types that are supported; it does not have to implement the method <code>supports_constraint(::Type{VectorOfVariables}, Type{&lt;:SupportedCones})</code> as it should return <code>false</code> and it&#39;s the default. This prevents the user to constrain the same variable in two different cones. When a <code>VectorOfVariables</code>-in-<code>S</code> is added, the variables of the vector have already been created so they already belong to given cones. If bridges are enabled, the constraint will therefore be bridged by adding slack variables in <code>S</code> and equality constraints ensuring that the slack variables are equal to the corresponding variables of the given constraint function.</p><p>Note that there may also be sets for which <code>!supports_add_constrained_variables(model, S)</code> and <code>supports_constraint(model, MOI.VectorOfVariables, S)</code>. For instance, suppose a solver supports positive semidefinite variable constraints and two types of variables: binary variables and nonnegative variables. Then the solver should support adding <code>VectorOfVariables</code>-in-<code>PositiveSemidefiniteConeTriangle</code> constraints, but it should not support creating variables constrained to belong to the <code>PositiveSemidefiniteConeTriangle</code> because the variables in <code>PositiveSemidefiniteConeTriangle</code> should first be created as either binary or non-negative.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/variables.jl#L102-L149">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}" href="#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}"><code>MathOptInterface.is_valid</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">is_valid(model::ModelLike, index::Index)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether this index refers to a valid object in the model <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L95-L99">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}" href="#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}"><code>MathOptInterface.delete</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">delete(model::ModelLike, index::Index)</code></pre><p>Delete the referenced object from the model. Throw <a href="../errors/#MathOptInterface.DeleteNotAllowed"><code>DeleteNotAllowed</code></a> if if <code>index</code> cannot be deleted.</p><p>The following modifications also take effect if <code>Index</code> is <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a>:</p><ul><li>If <code>index</code> used in the objective function, it is removed from the function, i.e., it is substituted for zero.</li><li>For each <code>func</code>-in-<code>set</code> constraint of the model:<ul><li>If <code>func isa VariableIndex</code> and <code>func == index</code> then the constraint is deleted.</li><li>If <code>func isa VectorOfVariables</code> and <code>index in func.variables</code> then<ul><li>if <code>length(func.variables) == 1</code> is one, the constraint is deleted;</li><li>if <code>length(func.variables) &gt; 1</code> and <code>supports_dimension_update(set)</code> then then the variable is removed from <code>func</code> and <code>set</code> is replaced by <code>update_dimension(set, MOI.dimension(set) - 1)</code>.</li><li>Otherwise, a <a href="../errors/#MathOptInterface.DeleteNotAllowed"><code>DeleteNotAllowed</code></a> error is thrown.</li></ul></li><li>Otherwise, the variable is removed from <code>func</code>, i.e., it is substituted for zero.</li></ul></li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L132-L152">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}" href="#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}"><code>MathOptInterface.delete</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">delete(model::ModelLike, indices::Vector{R&lt;:Index}) where {R}</code></pre><p>Delete the referenced objects in the vector <code>indices</code> from the model. It may be assumed that <code>R</code> is a concrete type. The default fallback sequentially deletes the individual items in <code>indices</code>, although specialized implementations may be more efficient.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L155-L162">source</a></section></article><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.AbstractVariableAttribute" href="#MathOptInterface.AbstractVariableAttribute"><code>MathOptInterface.AbstractVariableAttribute</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariableAttribute</code></pre><p>Abstract supertype for attribute objects that can be used to set or get attributes (properties) of variables in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L30-L35">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariableName" href="#MathOptInterface.VariableName"><code>MathOptInterface.VariableName</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableName()</code></pre><p>A variable attribute for a string identifying the variable. It is <em>valid</em> for two variables to have the same name; however, variables with duplicate names cannot be looked up using <a href="../models/#MathOptInterface.get"><code>get</code></a>. It has a default value of <code>&quot;&quot;</code> if not set`.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1429-L1436">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariablePrimalStart" href="#MathOptInterface.VariablePrimalStart"><code>MathOptInterface.VariablePrimalStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablePrimalStart()</code></pre><p>A variable attribute for the initial assignment to some primal variable&#39;s value that the optimizer may use to warm-start the solve. May be a number or <code>nothing</code> (unset).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1441-L1447">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariablePrimal" href="#MathOptInterface.VariablePrimal"><code>MathOptInterface.VariablePrimal</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablePrimal(result_index::Int = 1)</code></pre><p>A variable attribute for the assignment to some primal variable&#39;s value in result <code>result_index</code>. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a primal value for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>VariablePrimal</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1450-L1465">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariableBasisStatus" href="#MathOptInterface.VariableBasisStatus"><code>MathOptInterface.VariableBasisStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableBasisStatus(result_index::Int = 1)</code></pre><p>A variable attribute for the <code>BasisStatusCode</code> of a variable in result <code>result_index</code>, with respect to an available optimal solution basis.</p><p>If the solver does not have a basis statue for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>VariableBasisStatus</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1516-L1531">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>model</code> supports constraining a vector of variables to belong to a set of type <code>S</code> either on creation of the vector of variables with <a href="#MathOptInterface.add_constrained_variables"><code>add_constrained_variables</code></a> or after the variable is created with <a href="../constraints/#MathOptInterface.add_constraint"><code>add_constraint</code></a>.</p><p>By default, if <code>S</code> is <code>Reals</code> then this function returns <code>true</code> and otherwise, it falls back to <code>supports_add_constrained_variables(model, Reals) &amp;&amp; supports_constraint(model, MOI.VectorOfVariables, S)</code> which is the correct definition for most models.</p><p><strong>Example</strong></p><p>In the standard conic form (see <a href="../../background/duality/#Duality">Duality</a>), the variables are grouped into several cones and the constraints are affine equality constraints. If <code>Reals</code> is not one of the cones supported by the solvers then it needs to implement <code>supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false</code> as free variables are not supported. The solvers should then implement <code>supports_add_constrained_variables(::Optimizer, ::Type{&lt;:SupportedCones}) = true</code> where <code>SupportedCones</code> is the union of all cone types that are supported; it does not have to implement the method <code>supports_constraint(::Type{VectorOfVariables}, Type{&lt;:SupportedCones})</code> as it should return <code>false</code> and it&#39;s the default. This prevents the user to constrain the same variable in two different cones. When a <code>VectorOfVariables</code>-in-<code>S</code> is added, the variables of the vector have already been created so they already belong to given cones. If bridges are enabled, the constraint will therefore be bridged by adding slack variables in <code>S</code> and equality constraints ensuring that the slack variables are equal to the corresponding variables of the given constraint function.</p><p>Note that there may also be sets for which <code>!supports_add_constrained_variables(model, S)</code> and <code>supports_constraint(model, MOI.VectorOfVariables, S)</code>. For instance, suppose a solver supports positive semidefinite variable constraints and two types of variables: binary variables and nonnegative variables. Then the solver should support adding <code>VectorOfVariables</code>-in-<code>PositiveSemidefiniteConeTriangle</code> constraints, but it should not support creating variables constrained to belong to the <code>PositiveSemidefiniteConeTriangle</code> because the variables in <code>PositiveSemidefiniteConeTriangle</code> should first be created as either binary or non-negative.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/variables.jl#L102-L149">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}" href="#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}"><code>MathOptInterface.is_valid</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">is_valid(model::ModelLike, index::Index)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether this index refers to a valid object in the model <code>model</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L95-L99">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}" href="#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}"><code>MathOptInterface.delete</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">delete(model::ModelLike, index::Index)</code></pre><p>Delete the referenced object from the model. Throw <a href="../errors/#MathOptInterface.DeleteNotAllowed"><code>DeleteNotAllowed</code></a> if if <code>index</code> cannot be deleted.</p><p>The following modifications also take effect if <code>Index</code> is <a href="../standard_form/#MathOptInterface.VariableIndex"><code>VariableIndex</code></a>:</p><ul><li>If <code>index</code> used in the objective function, it is removed from the function, i.e., it is substituted for zero.</li><li>For each <code>func</code>-in-<code>set</code> constraint of the model:<ul><li>If <code>func isa VariableIndex</code> and <code>func == index</code> then the constraint is deleted.</li><li>If <code>func isa VectorOfVariables</code> and <code>index in func.variables</code> then<ul><li>if <code>length(func.variables) == 1</code> is one, the constraint is deleted;</li><li>if <code>length(func.variables) &gt; 1</code> and <code>supports_dimension_update(set)</code> then then the variable is removed from <code>func</code> and <code>set</code> is replaced by <code>update_dimension(set, MOI.dimension(set) - 1)</code>.</li><li>Otherwise, a <a href="../errors/#MathOptInterface.DeleteNotAllowed"><code>DeleteNotAllowed</code></a> error is thrown.</li></ul></li><li>Otherwise, the variable is removed from <code>func</code>, i.e., it is substituted for zero.</li></ul></li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L132-L152">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}" href="#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}"><code>MathOptInterface.delete</code></a> — <span class="docstring-category">Method</span></header><section><div><pre><code class="language-julia hljs">delete(model::ModelLike, indices::Vector{R&lt;:Index}) where {R}</code></pre><p>Delete the referenced objects in the vector <code>indices</code> from the model. It may be assumed that <code>R</code> is a concrete type. The default fallback sequentially deletes the individual items in <code>indices</code>, although specialized implementations may be more efficient.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/indextypes.jl#L155-L162">source</a></section></article><h2 id="Attributes"><a class="docs-heading-anchor" href="#Attributes">Attributes</a><a id="Attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Attributes" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.AbstractVariableAttribute" href="#MathOptInterface.AbstractVariableAttribute"><code>MathOptInterface.AbstractVariableAttribute</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">AbstractVariableAttribute</code></pre><p>Abstract supertype for attribute objects that can be used to set or get attributes (properties) of variables in the model.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L30-L35">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariableName" href="#MathOptInterface.VariableName"><code>MathOptInterface.VariableName</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableName()</code></pre><p>A variable attribute for a string identifying the variable. It is <em>valid</em> for two variables to have the same name; however, variables with duplicate names cannot be looked up using <a href="../models/#MathOptInterface.get"><code>get</code></a>. It has a default value of <code>&quot;&quot;</code> if not set`.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1429-L1436">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariablePrimalStart" href="#MathOptInterface.VariablePrimalStart"><code>MathOptInterface.VariablePrimalStart</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablePrimalStart()</code></pre><p>A variable attribute for the initial assignment to some primal variable&#39;s value that the optimizer may use to warm-start the solve. May be a number or <code>nothing</code> (unset).</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1441-L1447">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariablePrimal" href="#MathOptInterface.VariablePrimal"><code>MathOptInterface.VariablePrimal</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariablePrimal(result_index::Int = 1)</code></pre><p>A variable attribute for the assignment to some primal variable&#39;s value in result <code>result_index</code>. If <code>result_index</code> is omitted, it is 1 by default.</p><p>If the solver does not have a primal value for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>VariablePrimal</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1450-L1465">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.VariableBasisStatus" href="#MathOptInterface.VariableBasisStatus"><code>MathOptInterface.VariableBasisStatus</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">VariableBasisStatus(result_index::Int = 1)</code></pre><p>A variable attribute for the <code>BasisStatusCode</code> of a variable in result <code>result_index</code>, with respect to an available optimal solution basis.</p><p>If the solver does not have a basis statue for the variable because the <code>result_index</code> is beyond the available solutions (whose number is indicated by the <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> attribute), getting this attribute must throw a <a href="../errors/#MathOptInterface.ResultIndexBoundsError"><code>ResultIndexBoundsError</code></a>. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check <a href="../models/#MathOptInterface.PrimalStatus"><code>PrimalStatus</code></a> before accessing the <code>VariableBasisStatus</code> attribute.</p><p>See <a href="../models/#MathOptInterface.ResultCount"><code>ResultCount</code></a> for information on how the results are ordered.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/attributes.jl#L1516-L1531">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../models/">« Models</a><a class="docs-footer-nextpage" href="../constraints/">Constraints »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/release_notes/index.html b/previews/PR3562/moi/release_notes/index.html
index 5fdbf7e9d74..c094964c6b3 100644
--- a/previews/PR3562/moi/release_notes/index.html
+++ b/previews/PR3562/moi/release_notes/index.html
@@ -31,4 +31,4 @@
         end
         write(path, s)
     end
-end</code></pre><h2 id="[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)"><a class="docs-heading-anchor" href="#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22">v0.9.22</a> (May 22, 2021)</a><a id="[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)" title="Permalink"></a></h2><p>This release contains backports from the ongoing development of the v0.10 release.</p><ul><li>Improved type inference in <code>Utilities</code>, <code>Bridges</code> and <code>FileFormats</code> submodules to reduce latency.</li><li>Improved performance of <code>Utilities.is_canonical</code>.</li><li>Fixed <code>Utilities.pass_nonvariable_constraints</code> with bridged variables.</li><li>Fixed performance regression of <code>Utilities.Model</code>.</li><li>Fixed ordering of objective setting in parser.</li></ul><h2 id="[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)"><a class="docs-heading-anchor" href="#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21">v0.9.21</a> (April 23, 2021)</a><a id="[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)" title="Permalink"></a></h2><ul><li>Added <code>supports_shift_constant</code>.</li><li>Improve performance of bridging quadratic constraints.</li><li>Add precompilation statements.</li><li>Large improvements to the documentation.</li><li>Fix a variety of inference issues, benefiting precompilation and reducing initial latency.</li><li><code>RawParameter</code>s are now ignored when resetting a <code>CachingOptimizer</code>. Previously, changing the underlying optimizer after <code>RawParameter</code>s were set would throw an error.</li><li><code>Utilities.AbstractModel</code> is being refactored. This may break users interacting with private fields of a model generated using <code>@model</code>.</li></ul><h2 id="[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)"><a class="docs-heading-anchor" href="#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20">v0.9.20</a> (February 20, 2021)</a><a id="[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)" title="Permalink"></a></h2><ul><li>Improved performance of <code>Utilities.ScalarFunctionIterator</code></li><li>Added support for <code>compute_conflict</code> to MOI layers</li><li>Added test with zero off-diagonal quadratic term in objective</li><li>Fixed double deletion of nested bridged <code>SingleVariable</code>/<code>VectorOfVariables</code> constraints</li><li>Fixed modification of un-set objective</li><li>Fixed function modification with duplicate terms</li><li>Made unit tests abort without failing if the problem class is not supported</li><li>Formatted code with JuliaFormatter</li><li>Clarified <code>BasisStatusCode</code>&#39;s docstring</li></ul><h2 id="[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)"><a class="docs-heading-anchor" href="#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19">v0.9.19</a> (December 1, 2020)</a><a id="[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)" title="Permalink"></a></h2><ul><li>Added <code>CallbackNodeStatus</code> attribute</li><li>Added bridge from <code>GreaterThan</code> or <code>LessThan</code> to <code>Interval</code></li><li>Added tests for infeasibility certificates and double optimize</li><li>Fixed support for Julia v1.6</li><li>Re-organized MOI docs and added documentation for adding a test</li></ul><h2 id="[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)"><a class="docs-heading-anchor" href="#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18">v0.9.18</a> (November 3, 2020)</a><a id="[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)" title="Permalink"></a></h2><ul><li>Various improvements for working with complex numbers</li><li>Added <code>GeoMeantoRelEntrBridge</code> to bridge a <code>GeometricMeanCone</code> constraint to a relative entropy constraint</li></ul><h2 id="[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)"><a class="docs-heading-anchor" href="#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17">v0.9.17</a> (September 21, 2020)</a><a id="[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)" title="Permalink"></a></h2><ul><li>Fixed <code>CleverDict</code> with variable of negative index value</li><li>Implement <code>supports_add_constrained_variable</code> for <code>MockOptimizer</code></li></ul><h2 id="[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)"><a class="docs-heading-anchor" href="#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16">v0.9.16</a> (September 17, 2020)</a><a id="[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)" title="Permalink"></a></h2><ul><li>Various fixes:<ul><li>32-bit support</li><li><code>CleverDict</code> with abstract value type</li><li>Checks in test suite</li></ul></li></ul><h2 id="[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)"><a class="docs-heading-anchor" href="#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15">v0.9.15</a> (September 14, 2020)</a><a id="[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)" title="Permalink"></a></h2><ul><li>Bridges improvements:<ul><li>(R)SOCtoNonConvexQuad bridge</li><li>ZeroOne bridge</li><li>Use <code>supports_add_constrained_variable</code> in <code>LazyBridgeOptimizer</code></li><li>Exposed <code>VariableBridgeCost</code> and <code>ConstraintBridgeCost</code> attributes</li><li>Prioritize constraining variables on creation according to these costs</li><li>Refactor bridge debugging</li></ul></li><li>Large performance improvements across all submodules</li><li>Lots of documentation improvements</li><li>FileFormats improvements:<ul><li>Update MathOptFormat to v0.5</li><li>Fix supported objectives in <code>FileFormats</code></li></ul></li><li>Testing improvements:<ul><li>Add name option for <code>basic_constraint_test</code></li></ul></li><li>Bug fixes and missing methods<ul><li>Add <code>length</code> for iterators</li><li>Fix bug with duplicate terms</li><li>Fix order of <code>LinearOfConstraintIndices</code></li></ul></li></ul><h2 id="[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)"><a class="docs-heading-anchor" href="#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14">v0.9.14</a> (May 30, 2020)</a><a id="[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)" title="Permalink"></a></h2><ul><li>Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1056">#1056</a>).</li><li>Bump JSONSchema dependency from v0.2 to v0.3 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1090">#1090</a>).</li><li>Documentation improvements:<ul><li>Fix typos (#1054, #1060, #1061, #1064, #1069, #1070).</li><li>Remove the outdated recommendation for a package implementing MOI for a solver <code>XXX</code> to be called <code>MathOptInterfaceXXX</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1087">#1087</a>).</li></ul></li><li>Utilities improvements:<ul><li>Fix <code>is_canonical</code> for quadratic functions (#1081, #1089).</li><li>Implement <code>add_constrained_variable[s]</code> for <code>CachingOptimizer</code> so that it is added as constrained variables to the underlying optimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1084">#1084</a>).</li><li>Add support for custom objective functions for <code>UniversalFallback</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1086">#1086</a>).</li><li>Deterministic ordering of constraints in <code>UniversalFallback</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1088">#1088</a>).</li></ul></li><li>Testing improvements:<ul><li>Add <code>NormOneCone</code>/<code>NormInfinityCone</code> tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1045">#1045</a>).</li></ul></li><li>Bridges improvements:<ul><li>Add bridges from <code>Semiinteger</code> and <code>Semicontinuous</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1059">#1059</a>).</li><li>Implement getting <code>ConstraintSet</code> for <code>Variable.FlipSignBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1066">#1066</a>).</li><li>Fix setting <code>ConstraintFunction</code> for <code>Constraint.ScalarizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1093">#1093</a>).</li><li>Fix <code>NormOne</code>/<code>NormInf</code> bridges with nonzero constants (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1045">#1045</a>).</li><li>Fix StackOverflow in <code>debug</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1063">#1063</a>).</li></ul></li><li>FileFormats improvements:<ul><li>[SDPA] Implement the extension for integer variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1079">#1079</a>).</li><li>[SDPA] Ignore comments after <code>m</code> and <code>nblocks</code> and detect <code>dat-s</code> extension (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1077">#1077</a>).</li><li>[SDPA] No scaling of off-diagonal coefficient (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1076">#1076</a>).</li><li>[SDPA] Add missing negation of constant (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1075">#1075</a>).</li></ul></li></ul><h2 id="[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)"><a class="docs-heading-anchor" href="#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13">v0.9.13</a> (March 24, 2020)</a><a id="[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)" title="Permalink"></a></h2><ul><li>Added tests for <code>Semicontinuous</code> and <code>Semiinteger</code> variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1033">#1033</a>).</li><li>Added tests for using <code>ExprGraph</code>s from NLP evaluators (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1043">#1043</a>).</li><li>Update version compatibilities of dependencies (#1034, #1051, #1052).</li><li>Fixed typos in documentation (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1044">#1044</a>).</li></ul><h2 id="[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)"><a class="docs-heading-anchor" href="#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12">v0.9.12</a> (February 28, 2020)</a><a id="[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)" title="Permalink"></a></h2><ul><li>Fixed writing <code>NLPBlock</code> in MathOptFormat (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1037">#1037</a>).</li><li>Fixed <code>MockOptimizer</code> for result attributes with non-one result index (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1039">#1039</a>).</li><li>Updated test template with <code>instantiate</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1032">#1032</a>).</li></ul><h2 id="[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)"><a class="docs-heading-anchor" href="#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11">v0.9.11</a> (February 21, 2020)</a><a id="[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)" title="Permalink"></a></h2><ul><li>Add an option for the model created by <code>Utilities.@model</code> to be a subtype of <code>AbstractOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1031">#1031</a>).</li><li>Described dual cone in docstrings of <code>GeoMeanCone</code> and <code>RelativeEntropyCone</code> (#1018, #1028).</li><li>Fixed typos in documentation (#1022, #1024).</li><li>Fixed warning of unsupported attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1027">#1027</a>).</li><li>Added more rootdet/logdet conic tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1026">#1026</a>).</li><li>Implemented <code>ConstraintDual</code> for <code>Constraint.GeoMeanBridge</code>, <code>Constraint.RootDetBridge</code> and <code>Constraint.LogDetBridge</code> and test duals in tests with <code>GeoMeanCone</code> and <code>RootDetConeTriangle</code> and <code>LogDetConeTriangle</code> cones (#1025, #1026).</li></ul><h2 id="[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)"><a class="docs-heading-anchor" href="#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10">v0.9.10</a> (January 31, 2020)</a><a id="[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)" title="Permalink"></a></h2><ul><li>Added <code>OptimizerWithAttributes</code> grouping an optimizer constructor and a list of optimizer attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1008">#1008</a>).</li><li>Added <code>RelativeEntropyCone</code> with corresponding bridge into exponential cone constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/993">#993</a>).</li><li>Added <code>NormSpectralCone</code> and <code>NormNuclearCone</code> with corresponding bridges into positive semidefinite constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/976">#976</a>).</li><li>Added <code>supports_constrained_variable(s)</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1004">#1004</a>).</li><li>Added <code>dual_set_type</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1002">#1002</a>).</li><li>Added tests for vector specialized version of <code>delete</code> (#989, #1011).</li><li>Added PSD3 test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1007">#1007</a>).</li><li>Clarified dual solution of <code>Tests.pow1v</code> and <code>Tests.pow1f</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1013">#1013</a>).</li><li>Added support for <code>EqualTo</code> and <code>Zero</code> in <code>Bridges.Constraint.SplitIntervalBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1005">#1005</a>).</li><li>Fixed <code>Utilities.vectorize</code> for empty vector (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1003">#1003</a>).</li><li>Fixed free variables in LP writer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1006">#1006</a>).</li></ul><h2 id="[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)"><a class="docs-heading-anchor" href="#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9">v0.9.9</a> (December 29, 2019)</a><a id="[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)" title="Permalink"></a></h2><ul><li>Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/969">#969</a>).</li><li>Improved performance of deletion of vector of variables in <code>MOI.Utilities.Model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/983">#983</a>).</li><li>Updated to MutableArithmetics v0.2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/981">#981</a>).</li><li>Added <code>MutableArithmetics.promote_operation</code> allocation tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/975">#975</a>).</li><li>Fixed inference issue on Julia v1.1 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/982">#982</a>).</li></ul><h2 id="[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)"><a class="docs-heading-anchor" href="#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8">v0.9.8</a> (December 19, 2019)</a><a id="[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)" title="Permalink"></a></h2><ul><li>Implemented MutableArithmetics API (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/924">#924</a>).</li><li>Fixed callbacks with <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/959">#959</a>).</li><li>Fixed <code>MOI.dimension</code> for <code>MOI.Complements</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/948">#948</a>).</li><li>Added fallback for <code>add_variables</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/972">#972</a>).</li><li>Added <code>is_diagonal_vectorized_index</code> utility (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/965">#965</a>).</li><li>Improved linear constraints display in manual (#963, #964).</li><li>Bridges improvements:<ul><li>Added <code>IndicatorSet</code> to <code>SOS1</code> bridge (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/877">#877</a>).</li><li>Added support for starting values for <code>Variable.VectorizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/944">#944</a>).</li><li>Fixed <code>MOI.add_constraints</code> with non-bridged variable constraint on bridged variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/951">#951</a>).</li><li>Fixed corner cases and docstring of <code>GeoMeanBridge</code> (#961, #962, #966).</li><li>Fixed choice between variable or constraint bridges for constrained variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/973">#973</a>).</li><li>Improve performance of bridge shortest path (#945, #946, #956).</li><li>Added docstring for <code>test_delete_bridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/954">#954</a>).</li><li>Added Variable bridge tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/952">#952</a>).</li></ul></li></ul><h2 id="[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)"><a class="docs-heading-anchor" href="#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7">v0.9.7</a> (October 30, 2019)</a><a id="[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)" title="Permalink"></a></h2><ul><li>Implemented <code>_result_index_field</code> for <code>NLPBlockDual</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/934">#934</a>).</li><li>Fixed copy of model with starting values for vector constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/941">#941</a>).</li><li>Bridges improvements:<ul><li>Improved performance of <code>add_bridge</code> and added <code>has_bridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/935">#935</a>).</li><li>Added <code>AbstractSetMapBridge</code> for bridges between sets <code>S1</code>, <code>S2</code> such that there is a linear map <code>A</code> such that <code>A*S1 = S2</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/933">#933</a>).</li><li>Added support for starting values for <code>FlipSignBridge</code>, <code>VectorizeBridge</code>, <code>ScalarizeBridge</code>, <code>SlackBridge</code>, <code>SplitIntervalBridge</code>, <code>RSOCBridge</code>, <code>SOCRBridge</code> <code>NormInfinityBridge</code>, <code>SOCtoPSDBridge</code> and <code>RSOCtoPSDBridge</code> (#933, #936, #937, #938, #939).</li></ul></li></ul><h2 id="[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)"><a class="docs-heading-anchor" href="#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6">v0.9.6</a> (October 25, 2019)</a><a id="[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)" title="Permalink"></a></h2><ul><li>Added complementarity constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/913">#913</a>).</li><li>Allowed <code>ModelLike</code> objects as value of attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/928">#928</a>).</li><li>Testing improvements:<ul><li>Added <code>dual_objective_value</code> option to <code>MOI.Test.TestConfig</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/922">#922</a>).</li><li>Added <code>InvalidIndex</code> tests in <code>basic_constraint_tests</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/921">#921</a>).</li><li>Added tests for the constant term in indicator constraint (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/929">#929</a>).</li></ul></li><li>Bridges improvements:<ul><li>Added support for starting values for <code>Functionize</code> bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/923">#923</a>).</li><li>Added variable indices context to variable bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/920">#920</a>).</li><li>Fixed a typo in printing o <code>debug_supports</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/927">#927</a>).</li></ul></li></ul><h2 id="[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)"><a class="docs-heading-anchor" href="#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5">v0.9.5</a> (October 9, 2019)</a><a id="[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)" title="Permalink"></a></h2><ul><li>Clarified <code>PrimalStatus</code>/<code>DualStatus</code> to be <code>NO_SOLUTION</code> if <code>result_index</code> is out of bounds (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/912">#912</a>).</li><li>Added tolerance for checks  and use <code>ResultCount</code> + 1 for the <code>result_index</code> in <code>MOI.Test.solve_result_status</code> (#910, #917).</li><li>Use <code>0.5</code> instead of <code>2.0</code> for power in PowerCone in <code>basic_constraint_test</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/916">#916</a>).</li><li>Bridges improvements:<ul><li>Added debug utilities for unsupported variable/constraint/objective (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/861">#861</a>).</li><li>Fixed deletion of variables in bridged <code>VectorOfVariables</code> constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/909">#909</a>).</li><li>Fixed <code>result_index</code> with objective bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/911">#911</a>).</li></ul></li></ul><h2 id="[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)"><a class="docs-heading-anchor" href="#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4">v0.9.4</a> (October 2, 2019)</a><a id="[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)" title="Permalink"></a></h2><ul><li>Added solver-independent MIP callbacks (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/782">#782</a>).</li><li>Implements submit for <code>Utilities.CachingOptimizer</code> and <code>Bridges.AbstractBridgeOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/906">#906</a>).</li><li>Added tests for result count of solution attributes (#901, #904).</li><li>Added <code>NumberOfThreads</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/892">#892</a>).</li><li>Added <code>Utilities.get_bounds</code> to get the bounds on a variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/890">#890</a>).</li><li>Added a note on duplicate coefficients in documentation (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/581">#581</a>).</li><li>Added result index in <code>ConstraintBasisStatus</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/898">#898</a>).</li><li>Added extension dictionary to <code>Utilities.Model</code> (#884, #895).</li><li>Fixed deletion of constrained variables for CachingOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/905">#905</a>).</li><li>Implemented <code>Utilities.shift_constraint</code> for <code>Test.UnknownScalarSet</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/896">#896</a>).</li><li>Bridges improvements:<ul><li>Added Variable.RSOCtoSOCBridge (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/907">#907</a>).</li><li>Implemented <code>MOI.get</code> for <code>ConstraintFunction</code>/<code>ConstraintSet</code> for <code>Bridges.Constraint.SquareBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/899">#899</a>).</li></ul></li></ul><h2 id="[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)"><a class="docs-heading-anchor" href="#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3">v0.9.3</a> (September 20, 2019)</a><a id="[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)" title="Permalink"></a></h2><ul><li>Fixed ambiguity detected in Julia v1.3 (#891, #893).</li><li>Fixed missing sets from <code>ListOfSupportedConstraints</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/880">#880</a>).</li><li>Fixed copy of <code>VectorOfVariables</code> constraints with duplicate indices (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/886">#886</a>).</li><li>Added extension dictionary to MOIU.Model (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/884">#884</a>).</li><li>Implemented <code>MOI.get</code> for function and set for <code>GeoMeanBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/888">#888</a>).</li><li>Updated documentation for SingleVariable indices and bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/885">#885</a>).</li><li>Testing improvements:<ul><li>Added more comprehensive tests for names (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/882">#882</a>).</li><li>Added tests for <code>SingleVariable</code> duals (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/883">#883</a>).</li><li>Added tests for <code>DualExponentialCone</code> and <code>DualPowerCone</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/873">#873</a>).</li></ul></li><li>Improvements for arbitrary coefficient type:<ul><li>Fixed <code>==</code> for sets with mutable fields (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/887">#887</a>).</li><li>Removed some <code>Float64</code> assumptions in bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/878">#878</a>).</li><li>Automatic selection of <code>Constraint.[Scalar|Vector]FunctionizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/889">#889</a>).</li></ul></li></ul><h2 id="[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)"><a class="docs-heading-anchor" href="#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2">v0.9.2</a> (September 5, 2019)</a><a id="[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)" title="Permalink"></a></h2><ul><li>Implemented model printing for <code>MOI.ModelLike</code> and specialized it for models defined in MOI (864).</li><li>Generalized <code>contlinear</code> tests for arbitrary coefficient type (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/855">#855</a>).</li><li>Fixed <code>supports_constraint</code> for <code>Semiinteger</code> and <code>Semicontinuous</code> and <code>supports</code> for <code>ObjectiveFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/859">#859</a>).</li><li>Fixed Allocate-Load copy for single variable constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/856">#856</a>).</li><li>Bridges improvements:<ul><li>Add objective bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/789">#789</a>).</li><li>Fixed <code>Variable.RSOCtoPSDBridge</code> for dimension 2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/869">#869</a>).</li><li>Added <code>Variable.SOCtoRSOCBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/865">#865</a>).</li><li>Added <code>Constraint.SOCRBridge</code> and disable <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/751">#751</a>).</li><li>Fixed <code>added_constraint_types</code> for <code>Contraint.LogDetBridge</code> and <code>Constraint.RootDetBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/870">#870</a>).</li></ul></li></ul><h2 id="[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)"><a class="docs-heading-anchor" href="#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1">v0.9.1</a> (August 22, 2019)</a><a id="[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)" title="Permalink"></a></h2><ul><li>Fix support for Julia v1.2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/834">#834</a>).</li><li>L<em>1 and L</em>∞ norm epigraph cones and corresponding bridges to LP were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/818">#818</a>).</li><li>Added tests to <code>MOI.Test.nametest</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/833">#833</a>).</li><li>Fix <code>MOI.Test.soc3test</code> for solvers not supporting infeasibility certificates (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/839">#839</a>).</li><li>Implements <code>operate</code> for operators <code>*</code> and <code>/</code> between vector function and constant (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/837">#837</a>).</li><li>Implements <code>show</code> for <code>MOI.Utilities.IndexMap</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/847">#847</a>).</li><li>Fix corner cases for mapping of variables in <code>MOI.Utilities.CachingOptimizer</code> and substitution of variables in <code>MOI.Bridges.AbstractBridgeOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/848">#848</a>).</li><li>Fix transformation of constant terms for <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> and <code>MOI.Bridges.Constraint.RSOCtoPSDBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/840">#840</a>).</li></ul><h2 id="[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)"><a class="docs-heading-anchor" href="#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0">v0.9.0</a> (August 13, 2019)</a><a id="[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)" title="Permalink"></a></h2><ul><li>Support for Julia v0.6 and v0.7 was dropped (#714, #717).</li><li>A <code>MOI.Utilities.Model</code> implementation of <code>ModelLike</code>, this should replace most use cases of <code>MOI.Utilities.@model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/781">#781</a>).</li><li><code>add_constrained_variable</code> and <code>add_constrained_variables</code> were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>Support for indicator constraints was added (#709, #712).</li><li><code>DualObjectiveValue</code> attribute was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/473">#473</a>).</li><li><code>RawParameter</code> attribute was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/733">#733</a>).</li><li>A <code>dual_set</code> function was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/804">#804</a>).</li><li>A <code>Benchmarks</code> submodule was added to facilitate solver benchmarking (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/769">#769</a>).</li><li>A <code>submit</code> function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/775">#775</a>).</li><li>The field of <code>ObjectiveValue</code> was renamed to <code>result_index</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/729">#729</a>).</li><li>The <code>_constant</code> and <code>Utilities.getconstant</code> function were renamed to <code>constant</code></li><li><code>REDUCTION_CERTIFICATE</code> result status was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/734">#734</a>).</li><li>Abstract matrix sets were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/731">#731</a>).</li><li>Testing improvements:<ul><li>The testing guideline was updated (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/728">#728</a>).</li><li>Quadratic tests were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/697">#697</a>).</li><li>Unit tests for <code>RawStatusString</code>, <code>SolveTime</code>, <code>Silent</code> and <code>SolverName</code> were added (#726, #741).</li><li>A rotated second-order cone test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>A power cone test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/768">#768</a>).</li><li>Tests for <code>ZeroOne</code> variables with variable bounds were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/772">#772</a>).</li><li>An unbounded test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/773">#773</a>).</li><li>Existing tests had a few updates (#702, #703, #763).</li></ul></li><li>Documentation improvements:<ul><li>Added a section on <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/777">#777</a>).</li><li>Added a section on <code>UniversalFallback</code>, <code>Model</code> and <code>@model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/762">#762</a>).</li><li>Transition the knapsack example to a doctest with MockOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/786">#786</a>).</li></ul></li><li>Utilities improvements:<ul><li>A <code>CleverDict</code> utility was added for a vector that automatically transform into a dictionary once a first index is removed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/767">#767</a>).</li><li>The <code>Utilities.constant</code> function was renamed to <code>Utilities.constant_vector</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/740">#740</a>).</li><li>Implement optimizer attributes for CachingOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/745">#745</a>).</li><li>Rename <code>Utilities.add_scalar_constraint</code> to <code>Utilities.normalize_and_add_constraint</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/801">#801</a>).</li><li><code>operate</code> with <code>vcat</code>, <code>SingleVariable</code> and <code>VectorOfVariables</code> now returns a <code>VectorOfVariables</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/616">#616</a>).</li><li>Fix a type piracy of <code>operate</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/784">#784</a>).</li><li>The <code>load_constraint</code> fallback signature was fixed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/760">#760</a>).</li><li>The <code>set_dot</code> function was extended to work with sparse arrays (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/805">#805</a>).</li></ul></li><li>Bridges improvements:<ul><li>The bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement <code>ConstraintFunction</code> and <code>ConstraintSet</code> if the user wants to recover them. As a consequence, the <code>@bridge</code> macro was removed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/722">#722</a>).</li><li>Bridge are now instantiated with a <code>bridge_constraint</code> function instead of using a constructor (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/730">#730</a>).</li><li>Fix constraint attributes for bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/699">#699</a>).</li><li>Constraint bridges were moved to the <code>Bridges/Constraint</code> submodule so they should now inherit from <code>MOI.Bridges.Constraint.Abstract</code> and should implement <code>MOI.Bridges.Constraint.concrete_bridge_type</code> instead of <code>MOI.Bridges.concrete_bridge_type</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/756">#756</a>).</li><li>Variable bridges were added in (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>Various improvements (#746, #747).</li></ul></li></ul><h2 id="[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)"><a class="docs-heading-anchor" href="#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4">v0.8.4</a> (March 13, 2019)</a><a id="[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)" title="Permalink"></a></h2><ul><li>Performance improvement in <code>default_copy_to</code> and bridge optimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/696">#696</a>).</li><li>Add <code>Silent</code> and implement setting optimizer attributes in caching and mock optimizers (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/695">#695</a>).</li><li>Add <code>Functionize</code> bridges (SingleVariable and VectorOfVariables) (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/659">#659</a>).</li><li>Minor typo fixes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/694">#694</a>).</li></ul><h2 id="[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)"><a class="docs-heading-anchor" href="#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3">v0.8.3</a> (March 6, 2019)</a><a id="[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)" title="Permalink"></a></h2><ul><li>Use zero constant in scalar constraint function of <code>MOI.Test.copytest</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/691">#691</a>).</li><li>Fix variable deletion with <code>SingleVariable</code> objective function (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/690">#690</a>).</li><li>Fix <code>LazyBridgeOptimizer</code> with bridges that add no constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/689">#689</a>).</li><li>Error message improvements (#673, #685, #686, #688).</li><li>Documentation improvements (#682, #683, #687).</li><li>Basis status:<ul><li>Remove <code>VariableBasisStatus</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/679">#679</a>).</li><li>Test <code>ConstraintBasisStatus</code> and implement it in bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/678">#678</a>).</li></ul></li><li>Fix inference of <code>NumberOfVariables</code> and <code>NumberOfConstraints</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/677">#677</a>).</li><li>Implement division between a quadratic function and a number (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/675">#675</a>).</li></ul><h2 id="[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)"><a class="docs-heading-anchor" href="#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2">v0.8.2</a> (February 7, 2019)</a><a id="[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)" title="Permalink"></a></h2><ul><li>Add <code>RawStatusString</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/629">#629</a>).</li><li>Do not set names to the optimizer but only to the cache in <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/638">#638</a>).</li><li>Make scalar MOI functions act as scalars in broadcast (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/646">#646</a>).</li><li>Add function utilities:<ul><li>Implement <code>Base.zero</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/634">#634</a>), <code>Base.iszero</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/643">#643</a>), add missing arithmetic operations (#644, #645) and fix division (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/648">#648</a>).</li><li>Add a <code>vectorize</code> function that turns a vector of <code>ScalarAffineFunction</code> into a <code>VectorAffineFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/642">#642</a>).</li></ul></li><li>Improve support for starting values:<ul><li>Show a warning in copy when starting values are not supported instead of throwing an error (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/630">#630</a>).</li><li>Fix <code>UniversalFallback</code> for getting an variable or constraint attribute set to no indices (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/623">#623</a>).</li><li>Add a test in <code>contlineartest</code> with partially set <code>VariablePrimalStart</code>.</li></ul></li><li>Bridges improvements:<ul><li>Fix <code>StackOverFlow</code> in <code>LazyBridgeOptimizer</code> when there is a cycle in the graph of bridges.</li><li>Add <code>Slack</code> bridges (#610, #650).</li><li>Add <code>FlipSign</code> bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/658">#658</a>).</li></ul></li><li>Add tests with duplicate coefficients in <code>ScalarAffineFunction</code> and <code>VectorAffineFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/639">#639</a>).</li><li>Use tolerance to compare <code>VariablePrimal</code> in <code>rotatedsoc1</code> test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/632">#632</a>).</li><li>Use a zero constant in <code>ScalarAffineFunction</code> of constraints in <code>psdt2</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/622">#622</a>).</li></ul><h2 id="[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)"><a class="docs-heading-anchor" href="#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1">v0.8.1</a> (January 7, 2019)</a><a id="[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)" title="Permalink"></a></h2><ul><li>Adding an NLP objective now overrides any objective set using the <code>ObjectiveFunction</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/619">#619</a>).</li><li>Rename <code>fullbridgeoptimizer</code> into <code>full_bridge_optimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/621">#621</a>).</li><li>Allow custom constraint types with <code>full_bridge_optimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/617">#617</a>).</li><li>Add <code>Vectorize</code> bridge which transforms scalar linear constraints into vector linear constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/615">#615</a>).</li></ul><h2 id="[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)"><a class="docs-heading-anchor" href="#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0">v0.8.0</a> (December 18, 2018)</a><a id="[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)" title="Permalink"></a></h2><ul><li>Rename all enum values to follow the JuMP naming guidelines for constants, for example, <code>Optimal</code> becomes <code>OPTIMAL</code>, and <code>DualInfeasible</code> becomes <code>DUAL_INFEASIBLE</code>.</li><li>Rename CachingOptimizer methods for style compliance.</li><li>Add an <code>MOI.TerminationStatusCode</code> called <code>ALMOST_DUAL_INFEASIBLE</code>.</li></ul><h2 id="[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)"><a class="docs-heading-anchor" href="#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0">v0.7.0</a> (December 13, 2018)</a><a id="[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)" title="Permalink"></a></h2><ul><li>Test that <code>MOI.TerminationStatus</code> is <code>MOI.OptimizeNotCalled</code> before <code>MOI.optimize!</code> is called.</li><li>Check <code>supports_default_copy_to</code> in tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/594">#594</a>).</li><li>Key pieces of information like optimality, infeasibility, etc., are now reported through <code>TerminationStatusCode</code>. It is typically no longer necessary to check the result statuses in addition to the termination status.</li><li>Add perspective dimension to log-det cone (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/593">#593</a>).</li></ul><h2 id="[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)"><a class="docs-heading-anchor" href="#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4">v0.6.4</a> (November 27, 2018)</a><a id="[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)" title="Permalink"></a></h2><ul><li>Add <code>OptimizeNotCalled</code> termination status (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/577">#577</a>) and improve documentation of other statuses (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/575">#575</a>).</li><li>Add a solver naming guideline (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/578">#578</a>).</li><li>Make <code>FeasibilitySense</code> the default <code>ObjectiveSense</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/579">#579</a>).</li><li>Fix <code>Utilities.@model</code> and <code>Bridges.@bridge</code> macros for functions and sets defined outside MOI (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/582">#582</a>).</li><li>Document solver-specific attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/580">#580</a>) and implement them in <code>Utilities.CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/565">#565</a>).</li></ul><h2 id="[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)"><a class="docs-heading-anchor" href="#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3">v0.6.3</a> (November 16, 2018)</a><a id="[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)" title="Permalink"></a></h2><ul><li>Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/549">#549</a>)</li><li>Attributes may now be partially set (some values could be <code>nothing</code>). (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/563">#563</a>)</li><li>Performance improvements in Utilities.Model (#549, #567, #568)</li><li>Fix bug in QuadtoSOC (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/558">#558</a>).</li><li>New <code>supports_default_copy_to</code> method that optimizers should implement to control caching behavior.</li><li>Documentation improvements.</li></ul><h2 id="[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)"><a class="docs-heading-anchor" href="#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2">v0.6.2</a> (October 26, 2018)</a><a id="[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)" title="Permalink"></a></h2><ul><li>Improve hygiene of <code>@model</code> macro (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/544">#544</a>).</li><li>Fix bug in copy tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/543">#543</a>).</li><li>Fix bug in UniversalFallback attribute getter (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/540">#540</a>).</li><li>Allow all correct solutions for <code>solve_blank_obj</code> unit test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/537">#537</a>).</li><li>Add errors for Allocate-Load and bad constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/534">#534</a>).</li><li>[performance] Add specialized implementation of <code>hash</code> for <code>VariableIndex</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/533">#533</a>).</li><li>[performance] Construct the name to object dictionaries lazily in model (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/535">#535</a>).</li><li>Add the <code>QuadtoSOC</code> bridge which transforms <code>ScalarQuadraticFunction</code> constraints into <code>RotatedSecondOrderCone</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/483">#483</a>).</li></ul><h2 id="[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)"><a class="docs-heading-anchor" href="#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1">v0.6.1</a> (September 22, 2018)</a><a id="[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)" title="Permalink"></a></h2><ul><li>Enable <code>PositiveSemidefiniteConeSquare</code> set and quadratic functions in <code>MOIB.fullbridgeoptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/524">#524</a>).</li><li>Add warning in the bridge between <code>PositiveSemidefiniteConeSquare</code> and <code>PositiveSemidefiniteConeTriangle</code> when the matrix is almost symmetric (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/522">#522</a>).</li><li>Modify <code>MOIT.copytest</code> to not add multiples constraints on the same variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/521">#521</a>).</li><li>Add missing keyword argument in one of <code>MOIU.add_scalar_constraint</code> methods (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/520">#520</a>).</li></ul><h2 id="[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)"><a class="docs-heading-anchor" href="#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0">v0.6.0</a> (August 30, 2018)</a><a id="[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)" title="Permalink"></a></h2><ul><li>The <code>MOIU.@model</code> and <code>MOIB.@bridge</code> macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.</li><li>Rename functions according to the <a href="https://jump.dev/JuMP.jl/stable/developers/style/">JuMP style guide</a>:<ul><li><code>copy!</code> with keyword arguments <code>copynames</code> and <code>warnattributes</code> -&gt; <code>copy_to</code> with keyword arguments <code>copy_names</code> and <code>warn_attributes</code>;</li><li><code>set!</code> -&gt; <code>set</code>;</li><li><code>addvariable[s]!</code> -&gt; <code>add_variable[s]</code>;</li><li><code>supportsconstraint</code> -&gt; <code>supports_constraint</code>;</li><li><code>addconstraint[s]!</code> -&gt; <code>add_constraint[s]</code>;</li><li><code>isvalid</code> -&gt; <code>is_valid</code>;</li><li><code>isempty</code> -&gt; <code>is_empty</code>;</li><li><code>Base.delete!</code> -&gt; <code>delete</code>;</li><li><code>modify!</code> -&gt; <code>modify</code>;</li><li><code>transform!</code> -&gt; <code>transform</code>;</li><li><code>initialize!</code> -&gt; <code>initialize</code>;</li><li><code>write</code> -&gt; <code>write_to_file</code>; and</li><li><code>read!</code> -&gt; <code>read_from_file</code>.</li></ul></li><li>Remove <code>free!</code> (use <code>Base.finalize</code> instead).</li><li>Add the <code>SquarePSD</code> bridge which transforms <code>PositiveSemidefiniteConeTriangle</code> constraints into <code>PositiveSemidefiniteConeTriangle</code>.</li><li>Add result fallback for <code>ConstraintDual</code> of variable-wise constraint, <code>ConstraintPrimal</code> and <code>ObjectiveValue</code>.</li><li>Add tests for <code>ObjectiveBound</code>.</li><li>Add test for empty rows in vector linear constraint.</li><li>Rework errors: <code>CannotError</code> has been renamed <code>NotAllowedError</code> and the distinction between <code>UnsupportedError</code> and <code>NotAllowedError</code> is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)</li><li><code>canget</code> is removed. <code>NoSolution</code> is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).</li></ul><h2 id="[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)"><a class="docs-heading-anchor" href="#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0">v0.5.0</a> (August 5, 2018)</a><a id="[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)" title="Permalink"></a></h2><ul><li>Fix names with CachingOptimizer.</li><li>Cleanup thanks to @mohamed82008.</li><li>Added a universal fallback for constraints.</li><li>Fast utilities for function canonicalization thanks to @rdeits.</li><li>Renamed <code>dimension</code> field to <code>side_dimension</code> in the context of matrix-like sets.</li><li>New and improved tests for cases like duplicate terms and <code>ObjectiveBound</code>.</li><li>Removed <code>cantransform</code>, <code>canaddconstraint</code>, <code>canaddvariable</code>, <code>canset</code>, <code>canmodify</code>, and <code>candelete</code> functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of <code>UnsupportedError</code> indicate unsupported operations, while subtypes of <code>CannotError</code> indicate operations that cannot be performed in the current state.</li><li>The API for <code>copy!</code> is updated to remove the CopyResult type.</li><li>Updates for the new JuMP style guide.</li></ul><h2 id="[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)"><a class="docs-heading-anchor" href="#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1">v0.4.1</a> (June 28, 2018)</a><a id="[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)" title="Permalink"></a></h2><ul><li>Fixes vector function modification on 32 bits.</li><li>Fixes Bellman-Ford algorithm for bridges.</li><li>Added an NLP test with <code>FeasibilitySense</code>.</li><li>Update modification documentation.</li></ul><h2 id="[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)"><a class="docs-heading-anchor" href="#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0">v0.4.0</a> (June 23, 2018)</a><a id="[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)" title="Permalink"></a></h2><ul><li>Helper constructors for <code>VectorAffineTerm</code> and <code>VectorQuadraticTerm</code>.</li><li>Added <code>modify_lhs</code> to <code>TestConfig</code>.</li><li>Additional unit tests for optimizers.</li><li>Added a type parameter to <code>CachingOptimizer</code> for the <code>optimizer</code> field.</li><li>New API for problem modification (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/388">#388</a>)</li><li>Tests pass without deprecation warnings on Julia 0.7.</li><li>Small fixes and documentation updates.</li></ul><h2 id="[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)"><a class="docs-heading-anchor" href="#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0">v0.3.0</a> (May 25, 2018)</a><a id="[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)" title="Permalink"></a></h2><ul><li>Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.</li><li>Improvements to <code>MockOptimizer</code>.</li><li>Significant changes to <code>Bridges</code>.</li><li>New and improved unit tests.</li><li>Fixes for Julia 0.7.</li></ul><h2 id="[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)"><a class="docs-heading-anchor" href="#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0">v0.2.0</a> (April 24, 2018)</a><a id="[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)" title="Permalink"></a></h2><ul><li>Improvements to and better coverage of <code>Tests</code>.</li><li>Documentation fixes.</li><li><code>SolverName</code> attribute.</li><li>Changes to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).</li><li>Addition of NLP tests.</li><li>Introduction of <code>UniversalFallback</code>.</li><li><code>copynames</code> keyword argument to <code>MOI.copy!</code>.</li><li>Add Bridges submodule.</li></ul><h2 id="[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)"><a class="docs-heading-anchor" href="#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0">v0.1.0</a> (February 28, 2018)</a><a id="[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)" title="Permalink"></a></h2><ul><li>Initial public release.</li><li>The framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../developer/checklists/">« Checklists</a><a class="docs-footer-nextpage" href="../../release_notes/">Release notes »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><h2 id="[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)"><a class="docs-heading-anchor" href="#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22">v0.9.22</a> (May 22, 2021)</a><a id="[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)" title="Permalink"></a></h2><p>This release contains backports from the ongoing development of the v0.10 release.</p><ul><li>Improved type inference in <code>Utilities</code>, <code>Bridges</code> and <code>FileFormats</code> submodules to reduce latency.</li><li>Improved performance of <code>Utilities.is_canonical</code>.</li><li>Fixed <code>Utilities.pass_nonvariable_constraints</code> with bridged variables.</li><li>Fixed performance regression of <code>Utilities.Model</code>.</li><li>Fixed ordering of objective setting in parser.</li></ul><h2 id="[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)"><a class="docs-heading-anchor" href="#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21">v0.9.21</a> (April 23, 2021)</a><a id="[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)" title="Permalink"></a></h2><ul><li>Added <code>supports_shift_constant</code>.</li><li>Improve performance of bridging quadratic constraints.</li><li>Add precompilation statements.</li><li>Large improvements to the documentation.</li><li>Fix a variety of inference issues, benefiting precompilation and reducing initial latency.</li><li><code>RawParameter</code>s are now ignored when resetting a <code>CachingOptimizer</code>. Previously, changing the underlying optimizer after <code>RawParameter</code>s were set would throw an error.</li><li><code>Utilities.AbstractModel</code> is being refactored. This may break users interacting with private fields of a model generated using <code>@model</code>.</li></ul><h2 id="[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)"><a class="docs-heading-anchor" href="#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20">v0.9.20</a> (February 20, 2021)</a><a id="[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)" title="Permalink"></a></h2><ul><li>Improved performance of <code>Utilities.ScalarFunctionIterator</code></li><li>Added support for <code>compute_conflict</code> to MOI layers</li><li>Added test with zero off-diagonal quadratic term in objective</li><li>Fixed double deletion of nested bridged <code>SingleVariable</code>/<code>VectorOfVariables</code> constraints</li><li>Fixed modification of un-set objective</li><li>Fixed function modification with duplicate terms</li><li>Made unit tests abort without failing if the problem class is not supported</li><li>Formatted code with JuliaFormatter</li><li>Clarified <code>BasisStatusCode</code>&#39;s docstring</li></ul><h2 id="[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)"><a class="docs-heading-anchor" href="#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19">v0.9.19</a> (December 1, 2020)</a><a id="[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)" title="Permalink"></a></h2><ul><li>Added <code>CallbackNodeStatus</code> attribute</li><li>Added bridge from <code>GreaterThan</code> or <code>LessThan</code> to <code>Interval</code></li><li>Added tests for infeasibility certificates and double optimize</li><li>Fixed support for Julia v1.6</li><li>Re-organized MOI docs and added documentation for adding a test</li></ul><h2 id="[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)"><a class="docs-heading-anchor" href="#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18">v0.9.18</a> (November 3, 2020)</a><a id="[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)" title="Permalink"></a></h2><ul><li>Various improvements for working with complex numbers</li><li>Added <code>GeoMeantoRelEntrBridge</code> to bridge a <code>GeometricMeanCone</code> constraint to a relative entropy constraint</li></ul><h2 id="[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)"><a class="docs-heading-anchor" href="#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17">v0.9.17</a> (September 21, 2020)</a><a id="[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)" title="Permalink"></a></h2><ul><li>Fixed <code>CleverDict</code> with variable of negative index value</li><li>Implement <code>supports_add_constrained_variable</code> for <code>MockOptimizer</code></li></ul><h2 id="[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)"><a class="docs-heading-anchor" href="#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16">v0.9.16</a> (September 17, 2020)</a><a id="[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)" title="Permalink"></a></h2><ul><li>Various fixes:<ul><li>32-bit support</li><li><code>CleverDict</code> with abstract value type</li><li>Checks in test suite</li></ul></li></ul><h2 id="[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)"><a class="docs-heading-anchor" href="#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15">v0.9.15</a> (September 14, 2020)</a><a id="[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)" title="Permalink"></a></h2><ul><li>Bridges improvements:<ul><li>(R)SOCtoNonConvexQuad bridge</li><li>ZeroOne bridge</li><li>Use <code>supports_add_constrained_variable</code> in <code>LazyBridgeOptimizer</code></li><li>Exposed <code>VariableBridgeCost</code> and <code>ConstraintBridgeCost</code> attributes</li><li>Prioritize constraining variables on creation according to these costs</li><li>Refactor bridge debugging</li></ul></li><li>Large performance improvements across all submodules</li><li>Lots of documentation improvements</li><li>FileFormats improvements:<ul><li>Update MathOptFormat to v0.5</li><li>Fix supported objectives in <code>FileFormats</code></li></ul></li><li>Testing improvements:<ul><li>Add name option for <code>basic_constraint_test</code></li></ul></li><li>Bug fixes and missing methods<ul><li>Add <code>length</code> for iterators</li><li>Fix bug with duplicate terms</li><li>Fix order of <code>LinearOfConstraintIndices</code></li></ul></li></ul><h2 id="[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)"><a class="docs-heading-anchor" href="#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14">v0.9.14</a> (May 30, 2020)</a><a id="[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)" title="Permalink"></a></h2><ul><li>Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1056">#1056</a>).</li><li>Bump JSONSchema dependency from v0.2 to v0.3 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1090">#1090</a>).</li><li>Documentation improvements:<ul><li>Fix typos (#1054, #1060, #1061, #1064, #1069, #1070).</li><li>Remove the outdated recommendation for a package implementing MOI for a solver <code>XXX</code> to be called <code>MathOptInterfaceXXX</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1087">#1087</a>).</li></ul></li><li>Utilities improvements:<ul><li>Fix <code>is_canonical</code> for quadratic functions (#1081, #1089).</li><li>Implement <code>add_constrained_variable[s]</code> for <code>CachingOptimizer</code> so that it is added as constrained variables to the underlying optimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1084">#1084</a>).</li><li>Add support for custom objective functions for <code>UniversalFallback</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1086">#1086</a>).</li><li>Deterministic ordering of constraints in <code>UniversalFallback</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1088">#1088</a>).</li></ul></li><li>Testing improvements:<ul><li>Add <code>NormOneCone</code>/<code>NormInfinityCone</code> tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1045">#1045</a>).</li></ul></li><li>Bridges improvements:<ul><li>Add bridges from <code>Semiinteger</code> and <code>Semicontinuous</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1059">#1059</a>).</li><li>Implement getting <code>ConstraintSet</code> for <code>Variable.FlipSignBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1066">#1066</a>).</li><li>Fix setting <code>ConstraintFunction</code> for <code>Constraint.ScalarizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1093">#1093</a>).</li><li>Fix <code>NormOne</code>/<code>NormInf</code> bridges with nonzero constants (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1045">#1045</a>).</li><li>Fix StackOverflow in <code>debug</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1063">#1063</a>).</li></ul></li><li>FileFormats improvements:<ul><li>[SDPA] Implement the extension for integer variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1079">#1079</a>).</li><li>[SDPA] Ignore comments after <code>m</code> and <code>nblocks</code> and detect <code>dat-s</code> extension (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1077">#1077</a>).</li><li>[SDPA] No scaling of off-diagonal coefficient (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1076">#1076</a>).</li><li>[SDPA] Add missing negation of constant (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1075">#1075</a>).</li></ul></li></ul><h2 id="[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)"><a class="docs-heading-anchor" href="#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13">v0.9.13</a> (March 24, 2020)</a><a id="[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)" title="Permalink"></a></h2><ul><li>Added tests for <code>Semicontinuous</code> and <code>Semiinteger</code> variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1033">#1033</a>).</li><li>Added tests for using <code>ExprGraph</code>s from NLP evaluators (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1043">#1043</a>).</li><li>Update version compatibilities of dependencies (#1034, #1051, #1052).</li><li>Fixed typos in documentation (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1044">#1044</a>).</li></ul><h2 id="[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)"><a class="docs-heading-anchor" href="#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12">v0.9.12</a> (February 28, 2020)</a><a id="[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)" title="Permalink"></a></h2><ul><li>Fixed writing <code>NLPBlock</code> in MathOptFormat (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1037">#1037</a>).</li><li>Fixed <code>MockOptimizer</code> for result attributes with non-one result index (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1039">#1039</a>).</li><li>Updated test template with <code>instantiate</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1032">#1032</a>).</li></ul><h2 id="[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)"><a class="docs-heading-anchor" href="#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11">v0.9.11</a> (February 21, 2020)</a><a id="[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)" title="Permalink"></a></h2><ul><li>Add an option for the model created by <code>Utilities.@model</code> to be a subtype of <code>AbstractOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1031">#1031</a>).</li><li>Described dual cone in docstrings of <code>GeoMeanCone</code> and <code>RelativeEntropyCone</code> (#1018, #1028).</li><li>Fixed typos in documentation (#1022, #1024).</li><li>Fixed warning of unsupported attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1027">#1027</a>).</li><li>Added more rootdet/logdet conic tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1026">#1026</a>).</li><li>Implemented <code>ConstraintDual</code> for <code>Constraint.GeoMeanBridge</code>, <code>Constraint.RootDetBridge</code> and <code>Constraint.LogDetBridge</code> and test duals in tests with <code>GeoMeanCone</code> and <code>RootDetConeTriangle</code> and <code>LogDetConeTriangle</code> cones (#1025, #1026).</li></ul><h2 id="[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)"><a class="docs-heading-anchor" href="#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10">v0.9.10</a> (January 31, 2020)</a><a id="[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)" title="Permalink"></a></h2><ul><li>Added <code>OptimizerWithAttributes</code> grouping an optimizer constructor and a list of optimizer attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1008">#1008</a>).</li><li>Added <code>RelativeEntropyCone</code> with corresponding bridge into exponential cone constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/993">#993</a>).</li><li>Added <code>NormSpectralCone</code> and <code>NormNuclearCone</code> with corresponding bridges into positive semidefinite constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/976">#976</a>).</li><li>Added <code>supports_constrained_variable(s)</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1004">#1004</a>).</li><li>Added <code>dual_set_type</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1002">#1002</a>).</li><li>Added tests for vector specialized version of <code>delete</code> (#989, #1011).</li><li>Added PSD3 test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1007">#1007</a>).</li><li>Clarified dual solution of <code>Tests.pow1v</code> and <code>Tests.pow1f</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1013">#1013</a>).</li><li>Added support for <code>EqualTo</code> and <code>Zero</code> in <code>Bridges.Constraint.SplitIntervalBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1005">#1005</a>).</li><li>Fixed <code>Utilities.vectorize</code> for empty vector (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1003">#1003</a>).</li><li>Fixed free variables in LP writer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/1006">#1006</a>).</li></ul><h2 id="[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)"><a class="docs-heading-anchor" href="#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9">v0.9.9</a> (December 29, 2019)</a><a id="[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)" title="Permalink"></a></h2><ul><li>Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/969">#969</a>).</li><li>Improved performance of deletion of vector of variables in <code>MOI.Utilities.Model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/983">#983</a>).</li><li>Updated to MutableArithmetics v0.2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/981">#981</a>).</li><li>Added <code>MutableArithmetics.promote_operation</code> allocation tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/975">#975</a>).</li><li>Fixed inference issue on Julia v1.1 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/982">#982</a>).</li></ul><h2 id="[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)"><a class="docs-heading-anchor" href="#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8">v0.9.8</a> (December 19, 2019)</a><a id="[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)" title="Permalink"></a></h2><ul><li>Implemented MutableArithmetics API (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/924">#924</a>).</li><li>Fixed callbacks with <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/959">#959</a>).</li><li>Fixed <code>MOI.dimension</code> for <code>MOI.Complements</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/948">#948</a>).</li><li>Added fallback for <code>add_variables</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/972">#972</a>).</li><li>Added <code>is_diagonal_vectorized_index</code> utility (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/965">#965</a>).</li><li>Improved linear constraints display in manual (#963, #964).</li><li>Bridges improvements:<ul><li>Added <code>IndicatorSet</code> to <code>SOS1</code> bridge (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/877">#877</a>).</li><li>Added support for starting values for <code>Variable.VectorizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/944">#944</a>).</li><li>Fixed <code>MOI.add_constraints</code> with non-bridged variable constraint on bridged variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/951">#951</a>).</li><li>Fixed corner cases and docstring of <code>GeoMeanBridge</code> (#961, #962, #966).</li><li>Fixed choice between variable or constraint bridges for constrained variables (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/973">#973</a>).</li><li>Improve performance of bridge shortest path (#945, #946, #956).</li><li>Added docstring for <code>test_delete_bridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/954">#954</a>).</li><li>Added Variable bridge tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/952">#952</a>).</li></ul></li></ul><h2 id="[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)"><a class="docs-heading-anchor" href="#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7">v0.9.7</a> (October 30, 2019)</a><a id="[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)" title="Permalink"></a></h2><ul><li>Implemented <code>_result_index_field</code> for <code>NLPBlockDual</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/934">#934</a>).</li><li>Fixed copy of model with starting values for vector constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/941">#941</a>).</li><li>Bridges improvements:<ul><li>Improved performance of <code>add_bridge</code> and added <code>has_bridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/935">#935</a>).</li><li>Added <code>AbstractSetMapBridge</code> for bridges between sets <code>S1</code>, <code>S2</code> such that there is a linear map <code>A</code> such that <code>A*S1 = S2</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/933">#933</a>).</li><li>Added support for starting values for <code>FlipSignBridge</code>, <code>VectorizeBridge</code>, <code>ScalarizeBridge</code>, <code>SlackBridge</code>, <code>SplitIntervalBridge</code>, <code>RSOCBridge</code>, <code>SOCRBridge</code> <code>NormInfinityBridge</code>, <code>SOCtoPSDBridge</code> and <code>RSOCtoPSDBridge</code> (#933, #936, #937, #938, #939).</li></ul></li></ul><h2 id="[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)"><a class="docs-heading-anchor" href="#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6">v0.9.6</a> (October 25, 2019)</a><a id="[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)" title="Permalink"></a></h2><ul><li>Added complementarity constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/913">#913</a>).</li><li>Allowed <code>ModelLike</code> objects as value of attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/928">#928</a>).</li><li>Testing improvements:<ul><li>Added <code>dual_objective_value</code> option to <code>MOI.Test.TestConfig</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/922">#922</a>).</li><li>Added <code>InvalidIndex</code> tests in <code>basic_constraint_tests</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/921">#921</a>).</li><li>Added tests for the constant term in indicator constraint (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/929">#929</a>).</li></ul></li><li>Bridges improvements:<ul><li>Added support for starting values for <code>Functionize</code> bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/923">#923</a>).</li><li>Added variable indices context to variable bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/920">#920</a>).</li><li>Fixed a typo in printing o <code>debug_supports</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/927">#927</a>).</li></ul></li></ul><h2 id="[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)"><a class="docs-heading-anchor" href="#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5">v0.9.5</a> (October 9, 2019)</a><a id="[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)" title="Permalink"></a></h2><ul><li>Clarified <code>PrimalStatus</code>/<code>DualStatus</code> to be <code>NO_SOLUTION</code> if <code>result_index</code> is out of bounds (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/912">#912</a>).</li><li>Added tolerance for checks  and use <code>ResultCount</code> + 1 for the <code>result_index</code> in <code>MOI.Test.solve_result_status</code> (#910, #917).</li><li>Use <code>0.5</code> instead of <code>2.0</code> for power in PowerCone in <code>basic_constraint_test</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/916">#916</a>).</li><li>Bridges improvements:<ul><li>Added debug utilities for unsupported variable/constraint/objective (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/861">#861</a>).</li><li>Fixed deletion of variables in bridged <code>VectorOfVariables</code> constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/909">#909</a>).</li><li>Fixed <code>result_index</code> with objective bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/911">#911</a>).</li></ul></li></ul><h2 id="[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)"><a class="docs-heading-anchor" href="#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4">v0.9.4</a> (October 2, 2019)</a><a id="[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)" title="Permalink"></a></h2><ul><li>Added solver-independent MIP callbacks (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/782">#782</a>).</li><li>Implements submit for <code>Utilities.CachingOptimizer</code> and <code>Bridges.AbstractBridgeOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/906">#906</a>).</li><li>Added tests for result count of solution attributes (#901, #904).</li><li>Added <code>NumberOfThreads</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/892">#892</a>).</li><li>Added <code>Utilities.get_bounds</code> to get the bounds on a variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/890">#890</a>).</li><li>Added a note on duplicate coefficients in documentation (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/581">#581</a>).</li><li>Added result index in <code>ConstraintBasisStatus</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/898">#898</a>).</li><li>Added extension dictionary to <code>Utilities.Model</code> (#884, #895).</li><li>Fixed deletion of constrained variables for CachingOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/905">#905</a>).</li><li>Implemented <code>Utilities.shift_constraint</code> for <code>Test.UnknownScalarSet</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/896">#896</a>).</li><li>Bridges improvements:<ul><li>Added Variable.RSOCtoSOCBridge (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/907">#907</a>).</li><li>Implemented <code>MOI.get</code> for <code>ConstraintFunction</code>/<code>ConstraintSet</code> for <code>Bridges.Constraint.SquareBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/899">#899</a>).</li></ul></li></ul><h2 id="[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)"><a class="docs-heading-anchor" href="#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3">v0.9.3</a> (September 20, 2019)</a><a id="[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)" title="Permalink"></a></h2><ul><li>Fixed ambiguity detected in Julia v1.3 (#891, #893).</li><li>Fixed missing sets from <code>ListOfSupportedConstraints</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/880">#880</a>).</li><li>Fixed copy of <code>VectorOfVariables</code> constraints with duplicate indices (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/886">#886</a>).</li><li>Added extension dictionary to MOIU.Model (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/884">#884</a>).</li><li>Implemented <code>MOI.get</code> for function and set for <code>GeoMeanBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/888">#888</a>).</li><li>Updated documentation for SingleVariable indices and bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/885">#885</a>).</li><li>Testing improvements:<ul><li>Added more comprehensive tests for names (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/882">#882</a>).</li><li>Added tests for <code>SingleVariable</code> duals (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/883">#883</a>).</li><li>Added tests for <code>DualExponentialCone</code> and <code>DualPowerCone</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/873">#873</a>).</li></ul></li><li>Improvements for arbitrary coefficient type:<ul><li>Fixed <code>==</code> for sets with mutable fields (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/887">#887</a>).</li><li>Removed some <code>Float64</code> assumptions in bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/878">#878</a>).</li><li>Automatic selection of <code>Constraint.[Scalar|Vector]FunctionizeBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/889">#889</a>).</li></ul></li></ul><h2 id="[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)"><a class="docs-heading-anchor" href="#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2">v0.9.2</a> (September 5, 2019)</a><a id="[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)" title="Permalink"></a></h2><ul><li>Implemented model printing for <code>MOI.ModelLike</code> and specialized it for models defined in MOI (864).</li><li>Generalized <code>contlinear</code> tests for arbitrary coefficient type (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/855">#855</a>).</li><li>Fixed <code>supports_constraint</code> for <code>Semiinteger</code> and <code>Semicontinuous</code> and <code>supports</code> for <code>ObjectiveFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/859">#859</a>).</li><li>Fixed Allocate-Load copy for single variable constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/856">#856</a>).</li><li>Bridges improvements:<ul><li>Add objective bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/789">#789</a>).</li><li>Fixed <code>Variable.RSOCtoPSDBridge</code> for dimension 2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/869">#869</a>).</li><li>Added <code>Variable.SOCtoRSOCBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/865">#865</a>).</li><li>Added <code>Constraint.SOCRBridge</code> and disable <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/751">#751</a>).</li><li>Fixed <code>added_constraint_types</code> for <code>Contraint.LogDetBridge</code> and <code>Constraint.RootDetBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/870">#870</a>).</li></ul></li></ul><h2 id="[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)"><a class="docs-heading-anchor" href="#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1">v0.9.1</a> (August 22, 2019)</a><a id="[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)" title="Permalink"></a></h2><ul><li>Fix support for Julia v1.2 (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/834">#834</a>).</li><li>L<em>1 and L</em>∞ norm epigraph cones and corresponding bridges to LP were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/818">#818</a>).</li><li>Added tests to <code>MOI.Test.nametest</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/833">#833</a>).</li><li>Fix <code>MOI.Test.soc3test</code> for solvers not supporting infeasibility certificates (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/839">#839</a>).</li><li>Implements <code>operate</code> for operators <code>*</code> and <code>/</code> between vector function and constant (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/837">#837</a>).</li><li>Implements <code>show</code> for <code>MOI.Utilities.IndexMap</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/847">#847</a>).</li><li>Fix corner cases for mapping of variables in <code>MOI.Utilities.CachingOptimizer</code> and substitution of variables in <code>MOI.Bridges.AbstractBridgeOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/848">#848</a>).</li><li>Fix transformation of constant terms for <code>MOI.Bridges.Constraint.SOCtoPSDBridge</code> and <code>MOI.Bridges.Constraint.RSOCtoPSDBridge</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/840">#840</a>).</li></ul><h2 id="[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)"><a class="docs-heading-anchor" href="#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0">v0.9.0</a> (August 13, 2019)</a><a id="[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)" title="Permalink"></a></h2><ul><li>Support for Julia v0.6 and v0.7 was dropped (#714, #717).</li><li>A <code>MOI.Utilities.Model</code> implementation of <code>ModelLike</code>, this should replace most use cases of <code>MOI.Utilities.@model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/781">#781</a>).</li><li><code>add_constrained_variable</code> and <code>add_constrained_variables</code> were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>Support for indicator constraints was added (#709, #712).</li><li><code>DualObjectiveValue</code> attribute was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/473">#473</a>).</li><li><code>RawParameter</code> attribute was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/733">#733</a>).</li><li>A <code>dual_set</code> function was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/804">#804</a>).</li><li>A <code>Benchmarks</code> submodule was added to facilitate solver benchmarking (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/769">#769</a>).</li><li>A <code>submit</code> function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/775">#775</a>).</li><li>The field of <code>ObjectiveValue</code> was renamed to <code>result_index</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/729">#729</a>).</li><li>The <code>_constant</code> and <code>Utilities.getconstant</code> function were renamed to <code>constant</code></li><li><code>REDUCTION_CERTIFICATE</code> result status was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/734">#734</a>).</li><li>Abstract matrix sets were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/731">#731</a>).</li><li>Testing improvements:<ul><li>The testing guideline was updated (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/728">#728</a>).</li><li>Quadratic tests were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/697">#697</a>).</li><li>Unit tests for <code>RawStatusString</code>, <code>SolveTime</code>, <code>Silent</code> and <code>SolverName</code> were added (#726, #741).</li><li>A rotated second-order cone test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>A power cone test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/768">#768</a>).</li><li>Tests for <code>ZeroOne</code> variables with variable bounds were added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/772">#772</a>).</li><li>An unbounded test was added (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/773">#773</a>).</li><li>Existing tests had a few updates (#702, #703, #763).</li></ul></li><li>Documentation improvements:<ul><li>Added a section on <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/777">#777</a>).</li><li>Added a section on <code>UniversalFallback</code>, <code>Model</code> and <code>@model</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/762">#762</a>).</li><li>Transition the knapsack example to a doctest with MockOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/786">#786</a>).</li></ul></li><li>Utilities improvements:<ul><li>A <code>CleverDict</code> utility was added for a vector that automatically transform into a dictionary once a first index is removed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/767">#767</a>).</li><li>The <code>Utilities.constant</code> function was renamed to <code>Utilities.constant_vector</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/740">#740</a>).</li><li>Implement optimizer attributes for CachingOptimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/745">#745</a>).</li><li>Rename <code>Utilities.add_scalar_constraint</code> to <code>Utilities.normalize_and_add_constraint</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/801">#801</a>).</li><li><code>operate</code> with <code>vcat</code>, <code>SingleVariable</code> and <code>VectorOfVariables</code> now returns a <code>VectorOfVariables</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/616">#616</a>).</li><li>Fix a type piracy of <code>operate</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/784">#784</a>).</li><li>The <code>load_constraint</code> fallback signature was fixed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/760">#760</a>).</li><li>The <code>set_dot</code> function was extended to work with sparse arrays (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/805">#805</a>).</li></ul></li><li>Bridges improvements:<ul><li>The bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement <code>ConstraintFunction</code> and <code>ConstraintSet</code> if the user wants to recover them. As a consequence, the <code>@bridge</code> macro was removed (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/722">#722</a>).</li><li>Bridge are now instantiated with a <code>bridge_constraint</code> function instead of using a constructor (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/730">#730</a>).</li><li>Fix constraint attributes for bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/699">#699</a>).</li><li>Constraint bridges were moved to the <code>Bridges/Constraint</code> submodule so they should now inherit from <code>MOI.Bridges.Constraint.Abstract</code> and should implement <code>MOI.Bridges.Constraint.concrete_bridge_type</code> instead of <code>MOI.Bridges.concrete_bridge_type</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/756">#756</a>).</li><li>Variable bridges were added in (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/759">#759</a>).</li><li>Various improvements (#746, #747).</li></ul></li></ul><h2 id="[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)"><a class="docs-heading-anchor" href="#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4">v0.8.4</a> (March 13, 2019)</a><a id="[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)" title="Permalink"></a></h2><ul><li>Performance improvement in <code>default_copy_to</code> and bridge optimizer (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/696">#696</a>).</li><li>Add <code>Silent</code> and implement setting optimizer attributes in caching and mock optimizers (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/695">#695</a>).</li><li>Add <code>Functionize</code> bridges (SingleVariable and VectorOfVariables) (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/659">#659</a>).</li><li>Minor typo fixes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/694">#694</a>).</li></ul><h2 id="[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)"><a class="docs-heading-anchor" href="#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3">v0.8.3</a> (March 6, 2019)</a><a id="[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)" title="Permalink"></a></h2><ul><li>Use zero constant in scalar constraint function of <code>MOI.Test.copytest</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/691">#691</a>).</li><li>Fix variable deletion with <code>SingleVariable</code> objective function (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/690">#690</a>).</li><li>Fix <code>LazyBridgeOptimizer</code> with bridges that add no constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/689">#689</a>).</li><li>Error message improvements (#673, #685, #686, #688).</li><li>Documentation improvements (#682, #683, #687).</li><li>Basis status:<ul><li>Remove <code>VariableBasisStatus</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/679">#679</a>).</li><li>Test <code>ConstraintBasisStatus</code> and implement it in bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/678">#678</a>).</li></ul></li><li>Fix inference of <code>NumberOfVariables</code> and <code>NumberOfConstraints</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/677">#677</a>).</li><li>Implement division between a quadratic function and a number (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/675">#675</a>).</li></ul><h2 id="[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)"><a class="docs-heading-anchor" href="#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2">v0.8.2</a> (February 7, 2019)</a><a id="[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)" title="Permalink"></a></h2><ul><li>Add <code>RawStatusString</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/629">#629</a>).</li><li>Do not set names to the optimizer but only to the cache in <code>CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/638">#638</a>).</li><li>Make scalar MOI functions act as scalars in broadcast (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/646">#646</a>).</li><li>Add function utilities:<ul><li>Implement <code>Base.zero</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/634">#634</a>), <code>Base.iszero</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/643">#643</a>), add missing arithmetic operations (#644, #645) and fix division (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/648">#648</a>).</li><li>Add a <code>vectorize</code> function that turns a vector of <code>ScalarAffineFunction</code> into a <code>VectorAffineFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/642">#642</a>).</li></ul></li><li>Improve support for starting values:<ul><li>Show a warning in copy when starting values are not supported instead of throwing an error (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/630">#630</a>).</li><li>Fix <code>UniversalFallback</code> for getting an variable or constraint attribute set to no indices (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/623">#623</a>).</li><li>Add a test in <code>contlineartest</code> with partially set <code>VariablePrimalStart</code>.</li></ul></li><li>Bridges improvements:<ul><li>Fix <code>StackOverFlow</code> in <code>LazyBridgeOptimizer</code> when there is a cycle in the graph of bridges.</li><li>Add <code>Slack</code> bridges (#610, #650).</li><li>Add <code>FlipSign</code> bridges (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/658">#658</a>).</li></ul></li><li>Add tests with duplicate coefficients in <code>ScalarAffineFunction</code> and <code>VectorAffineFunction</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/639">#639</a>).</li><li>Use tolerance to compare <code>VariablePrimal</code> in <code>rotatedsoc1</code> test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/632">#632</a>).</li><li>Use a zero constant in <code>ScalarAffineFunction</code> of constraints in <code>psdt2</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/622">#622</a>).</li></ul><h2 id="[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)"><a class="docs-heading-anchor" href="#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1">v0.8.1</a> (January 7, 2019)</a><a id="[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)" title="Permalink"></a></h2><ul><li>Adding an NLP objective now overrides any objective set using the <code>ObjectiveFunction</code> attribute (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/619">#619</a>).</li><li>Rename <code>fullbridgeoptimizer</code> into <code>full_bridge_optimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/621">#621</a>).</li><li>Allow custom constraint types with <code>full_bridge_optimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/617">#617</a>).</li><li>Add <code>Vectorize</code> bridge which transforms scalar linear constraints into vector linear constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/615">#615</a>).</li></ul><h2 id="[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)"><a class="docs-heading-anchor" href="#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0">v0.8.0</a> (December 18, 2018)</a><a id="[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)" title="Permalink"></a></h2><ul><li>Rename all enum values to follow the JuMP naming guidelines for constants, for example, <code>Optimal</code> becomes <code>OPTIMAL</code>, and <code>DualInfeasible</code> becomes <code>DUAL_INFEASIBLE</code>.</li><li>Rename CachingOptimizer methods for style compliance.</li><li>Add an <code>MOI.TerminationStatusCode</code> called <code>ALMOST_DUAL_INFEASIBLE</code>.</li></ul><h2 id="[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)"><a class="docs-heading-anchor" href="#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0">v0.7.0</a> (December 13, 2018)</a><a id="[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)" title="Permalink"></a></h2><ul><li>Test that <code>MOI.TerminationStatus</code> is <code>MOI.OptimizeNotCalled</code> before <code>MOI.optimize!</code> is called.</li><li>Check <code>supports_default_copy_to</code> in tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/594">#594</a>).</li><li>Key pieces of information like optimality, infeasibility, etc., are now reported through <code>TerminationStatusCode</code>. It is typically no longer necessary to check the result statuses in addition to the termination status.</li><li>Add perspective dimension to log-det cone (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/593">#593</a>).</li></ul><h2 id="[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)"><a class="docs-heading-anchor" href="#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4">v0.6.4</a> (November 27, 2018)</a><a id="[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)" title="Permalink"></a></h2><ul><li>Add <code>OptimizeNotCalled</code> termination status (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/577">#577</a>) and improve documentation of other statuses (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/575">#575</a>).</li><li>Add a solver naming guideline (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/578">#578</a>).</li><li>Make <code>FeasibilitySense</code> the default <code>ObjectiveSense</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/579">#579</a>).</li><li>Fix <code>Utilities.@model</code> and <code>Bridges.@bridge</code> macros for functions and sets defined outside MOI (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/582">#582</a>).</li><li>Document solver-specific attributes (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/580">#580</a>) and implement them in <code>Utilities.CachingOptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/565">#565</a>).</li></ul><h2 id="[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)"><a class="docs-heading-anchor" href="#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3">v0.6.3</a> (November 16, 2018)</a><a id="[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)" title="Permalink"></a></h2><ul><li>Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/549">#549</a>)</li><li>Attributes may now be partially set (some values could be <code>nothing</code>). (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/563">#563</a>)</li><li>Performance improvements in Utilities.Model (#549, #567, #568)</li><li>Fix bug in QuadtoSOC (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/558">#558</a>).</li><li>New <code>supports_default_copy_to</code> method that optimizers should implement to control caching behavior.</li><li>Documentation improvements.</li></ul><h2 id="[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)"><a class="docs-heading-anchor" href="#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2">v0.6.2</a> (October 26, 2018)</a><a id="[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)" title="Permalink"></a></h2><ul><li>Improve hygiene of <code>@model</code> macro (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/544">#544</a>).</li><li>Fix bug in copy tests (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/543">#543</a>).</li><li>Fix bug in UniversalFallback attribute getter (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/540">#540</a>).</li><li>Allow all correct solutions for <code>solve_blank_obj</code> unit test (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/537">#537</a>).</li><li>Add errors for Allocate-Load and bad constraints (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/534">#534</a>).</li><li>[performance] Add specialized implementation of <code>hash</code> for <code>VariableIndex</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/533">#533</a>).</li><li>[performance] Construct the name to object dictionaries lazily in model (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/535">#535</a>).</li><li>Add the <code>QuadtoSOC</code> bridge which transforms <code>ScalarQuadraticFunction</code> constraints into <code>RotatedSecondOrderCone</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/483">#483</a>).</li></ul><h2 id="[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)"><a class="docs-heading-anchor" href="#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1">v0.6.1</a> (September 22, 2018)</a><a id="[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)" title="Permalink"></a></h2><ul><li>Enable <code>PositiveSemidefiniteConeSquare</code> set and quadratic functions in <code>MOIB.fullbridgeoptimizer</code> (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/524">#524</a>).</li><li>Add warning in the bridge between <code>PositiveSemidefiniteConeSquare</code> and <code>PositiveSemidefiniteConeTriangle</code> when the matrix is almost symmetric (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/522">#522</a>).</li><li>Modify <code>MOIT.copytest</code> to not add multiples constraints on the same variable (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/521">#521</a>).</li><li>Add missing keyword argument in one of <code>MOIU.add_scalar_constraint</code> methods (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/520">#520</a>).</li></ul><h2 id="[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)"><a class="docs-heading-anchor" href="#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0">v0.6.0</a> (August 30, 2018)</a><a id="[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)" title="Permalink"></a></h2><ul><li>The <code>MOIU.@model</code> and <code>MOIB.@bridge</code> macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.</li><li>Rename functions according to the <a href="https://jump.dev/JuMP.jl/stable/developers/style/">JuMP style guide</a>:<ul><li><code>copy!</code> with keyword arguments <code>copynames</code> and <code>warnattributes</code> -&gt; <code>copy_to</code> with keyword arguments <code>copy_names</code> and <code>warn_attributes</code>;</li><li><code>set!</code> -&gt; <code>set</code>;</li><li><code>addvariable[s]!</code> -&gt; <code>add_variable[s]</code>;</li><li><code>supportsconstraint</code> -&gt; <code>supports_constraint</code>;</li><li><code>addconstraint[s]!</code> -&gt; <code>add_constraint[s]</code>;</li><li><code>isvalid</code> -&gt; <code>is_valid</code>;</li><li><code>isempty</code> -&gt; <code>is_empty</code>;</li><li><code>Base.delete!</code> -&gt; <code>delete</code>;</li><li><code>modify!</code> -&gt; <code>modify</code>;</li><li><code>transform!</code> -&gt; <code>transform</code>;</li><li><code>initialize!</code> -&gt; <code>initialize</code>;</li><li><code>write</code> -&gt; <code>write_to_file</code>; and</li><li><code>read!</code> -&gt; <code>read_from_file</code>.</li></ul></li><li>Remove <code>free!</code> (use <code>Base.finalize</code> instead).</li><li>Add the <code>SquarePSD</code> bridge which transforms <code>PositiveSemidefiniteConeTriangle</code> constraints into <code>PositiveSemidefiniteConeTriangle</code>.</li><li>Add result fallback for <code>ConstraintDual</code> of variable-wise constraint, <code>ConstraintPrimal</code> and <code>ObjectiveValue</code>.</li><li>Add tests for <code>ObjectiveBound</code>.</li><li>Add test for empty rows in vector linear constraint.</li><li>Rework errors: <code>CannotError</code> has been renamed <code>NotAllowedError</code> and the distinction between <code>UnsupportedError</code> and <code>NotAllowedError</code> is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)</li><li><code>canget</code> is removed. <code>NoSolution</code> is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).</li></ul><h2 id="[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)"><a class="docs-heading-anchor" href="#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0">v0.5.0</a> (August 5, 2018)</a><a id="[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)" title="Permalink"></a></h2><ul><li>Fix names with CachingOptimizer.</li><li>Cleanup thanks to @mohamed82008.</li><li>Added a universal fallback for constraints.</li><li>Fast utilities for function canonicalization thanks to @rdeits.</li><li>Renamed <code>dimension</code> field to <code>side_dimension</code> in the context of matrix-like sets.</li><li>New and improved tests for cases like duplicate terms and <code>ObjectiveBound</code>.</li><li>Removed <code>cantransform</code>, <code>canaddconstraint</code>, <code>canaddvariable</code>, <code>canset</code>, <code>canmodify</code>, and <code>candelete</code> functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of <code>UnsupportedError</code> indicate unsupported operations, while subtypes of <code>CannotError</code> indicate operations that cannot be performed in the current state.</li><li>The API for <code>copy!</code> is updated to remove the CopyResult type.</li><li>Updates for the new JuMP style guide.</li></ul><h2 id="[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)"><a class="docs-heading-anchor" href="#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1">v0.4.1</a> (June 28, 2018)</a><a id="[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)" title="Permalink"></a></h2><ul><li>Fixes vector function modification on 32 bits.</li><li>Fixes Bellman-Ford algorithm for bridges.</li><li>Added an NLP test with <code>FeasibilitySense</code>.</li><li>Update modification documentation.</li></ul><h2 id="[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)"><a class="docs-heading-anchor" href="#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0">v0.4.0</a> (June 23, 2018)</a><a id="[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)" title="Permalink"></a></h2><ul><li>Helper constructors for <code>VectorAffineTerm</code> and <code>VectorQuadraticTerm</code>.</li><li>Added <code>modify_lhs</code> to <code>TestConfig</code>.</li><li>Additional unit tests for optimizers.</li><li>Added a type parameter to <code>CachingOptimizer</code> for the <code>optimizer</code> field.</li><li>New API for problem modification (<a href="https://github.com/jump-dev/MathOptInterface.jl/issues/388">#388</a>)</li><li>Tests pass without deprecation warnings on Julia 0.7.</li><li>Small fixes and documentation updates.</li></ul><h2 id="[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)"><a class="docs-heading-anchor" href="#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0">v0.3.0</a> (May 25, 2018)</a><a id="[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)" title="Permalink"></a></h2><ul><li>Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.</li><li>Improvements to <code>MockOptimizer</code>.</li><li>Significant changes to <code>Bridges</code>.</li><li>New and improved unit tests.</li><li>Fixes for Julia 0.7.</li></ul><h2 id="[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)"><a class="docs-heading-anchor" href="#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0">v0.2.0</a> (April 24, 2018)</a><a id="[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)" title="Permalink"></a></h2><ul><li>Improvements to and better coverage of <code>Tests</code>.</li><li>Documentation fixes.</li><li><code>SolverName</code> attribute.</li><li>Changes to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).</li><li>Addition of NLP tests.</li><li>Introduction of <code>UniversalFallback</code>.</li><li><code>copynames</code> keyword argument to <code>MOI.copy!</code>.</li><li>Add Bridges submodule.</li></ul><h2 id="[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)"><a class="docs-heading-anchor" href="#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)"><a href="https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0">v0.1.0</a> (February 28, 2018)</a><a id="[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)" title="Permalink"></a></h2><ul><li>Initial public release.</li><li>The framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../developer/checklists/">« Checklists</a><a class="docs-footer-nextpage" href="../../release_notes/">Release notes »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Benchmarks/overview/index.html b/previews/PR3562/moi/submodules/Benchmarks/overview/index.html
index fee5a59ee0e..ba90c0ad068 100644
--- a/previews/PR3562/moi/submodules/Benchmarks/overview/index.html
+++ b/previews/PR3562/moi/submodules/Benchmarks/overview/index.html
@@ -21,4 +21,4 @@
 
 MOI.Benchmarks.compare_against_baseline(
     suite, &quot;current&quot;; directory = &quot;/tmp&quot;, verbose = true
-)</code></pre><p>This comparison will create a report detailing improvements and regressions.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../../reference/errors/">« Errors</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><p>This comparison will create a report detailing improvements and regressions.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../../reference/errors/">« Errors</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Benchmarks/reference/index.html b/previews/PR3562/moi/submodules/Benchmarks/reference/index.html
index 66a34e15cb8..b97ade11d4b 100644
--- a/previews/PR3562/moi/submodules/Benchmarks/reference/index.html
+++ b/previews/PR3562/moi/submodules/Benchmarks/reference/index.html
@@ -18,4 +18,4 @@
 )</code></pre><p>Run all benchmarks in <code>suite</code> and compare against files called <code>name</code> in <code>directory</code> that were created by a call to <code>create_baseline</code>.</p><p>A report summarizing the comparison is written to <code>report_filename</code> in <code>directory</code>.</p><p>Extra <code>kwargs</code> are based to <code>BenchmarkTools.run</code>.</p><p><strong>Examples</strong></p><pre><code class="language-julia hljs">my_suite = suite(() -&gt; GLPK.Optimizer())
 compare_against_baseline(
     my_suite, &quot;glpk_master&quot;; directory = &quot;/tmp&quot;, verbose = true
-)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Benchmarks/Benchmarks.jl#L75-L97">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Bridges/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Benchmarks/Benchmarks.jl#L75-L97">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Bridges/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Bridges/list_of_bridges/index.html b/previews/PR3562/moi/submodules/Bridges/list_of_bridges/index.html
index 29cc94ed0df..26a35074eab 100644
--- a/previews/PR3562/moi/submodules/Bridges/list_of_bridges/index.html
+++ b/previews/PR3562/moi/submodules/Bridges/list_of_bridges/index.html
@@ -111,4 +111,4 @@
            &amp;        &amp;        &amp; x_{11}  &amp; x_{12}  &amp; x_{13} \\
            &amp;        &amp;        &amp;         &amp; x_{22}  &amp; x_{23} \\
            &amp;        &amp;        &amp;         &amp;         &amp; x_{33}
-\end{bmatrix}\]</p><p>is positive semidefinite.</p><p>The bridge achieves this reformulation by adding a new set of variables in <code>MOI.PositiveSemidefiniteConeTriangle(6)</code>, and then adding three groups of equality constraints to:</p><ul><li>constrain the two <code>x</code> blocks to be equal</li><li>force the diagonal of the <code>y</code> blocks to be <code>0</code></li><li>force the lower triangular of the <code>y</code> block to be the negative of the upper triangle.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/Variable/bridges/hermitian.jl#L7-L63">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{bmatrix}\]</p><p>is positive semidefinite.</p><p>The bridge achieves this reformulation by adding a new set of variables in <code>MOI.PositiveSemidefiniteConeTriangle(6)</code>, and then adding three groups of equality constraints to:</p><ul><li>constrain the two <code>x</code> blocks to be equal</li><li>force the diagonal of the <code>y</code> blocks to be <code>0</code></li><li>force the lower triangular of the <code>y</code> block to be the negative of the upper triangle.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/Variable/bridges/hermitian.jl#L7-L63">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Bridges/overview/index.html b/previews/PR3562/moi/submodules/Bridges/overview/index.html
index b544821db75..42c60c85f48 100644
--- a/previews/PR3562/moi/submodules/Bridges/overview/index.html
+++ b/previews/PR3562/moi/submodules/Bridges/overview/index.html
@@ -54,4 +54,4 @@
 
 julia&gt; MOI.get(inner_optimizer, MOI.ListOfConstraintTypesPresent())
 1-element Vector{Tuple{Type, Type}}:
- (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Benchmarks/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../list_of_bridges/">List of bridges »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Benchmarks/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../list_of_bridges/">List of bridges »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Bridges/reference/index.html b/previews/PR3562/moi/submodules/Bridges/reference/index.html
index 3cc1aac1154..1250e1981c5 100644
--- a/previews/PR3562/moi/submodules/Bridges/reference/index.html
+++ b/previews/PR3562/moi/submodules/Bridges/reference/index.html
@@ -196,4 +196,4 @@
     cost::Int,
 )</code></pre><p>As an alternative to <code>variable_node</code>, add a virtual edge to <code>graph</code> that represents adding a free variable, followed by a constraint of type <code>constraint_node</code>, with bridging cost <code>cost</code>.</p><p><strong>Why is this needed?</strong></p><p>Variables can either be added as a variable constrained on creation, or as a free variable which then has a constraint added to it.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/graph.jl#L270-L286">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Bridges.bridge_index" href="#MathOptInterface.Bridges.bridge_index"><code>MathOptInterface.Bridges.bridge_index</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">bridge_index(graph::Graph, node::VariableNode)::Int
 bridge_index(graph::Graph, node::ConstraintNode)::Int
-bridge_index(graph::Graph, node::ObjectiveNode)::Int</code></pre><p>Return the optimal index of the bridge to chose from <code>node</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/graph.jl#L303-L309">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Bridges.is_variable_edge_best" href="#MathOptInterface.Bridges.is_variable_edge_best"><code>MathOptInterface.Bridges.is_variable_edge_best</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_variable_edge_best(graph::Graph, node::VariableNode)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>node</code> should be added as a variable constrained on creation, or as a free variable followed by a constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/graph.jl#L325-L330">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../list_of_bridges/">« List of bridges</a><a class="docs-footer-nextpage" href="../../FileFormats/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+bridge_index(graph::Graph, node::ObjectiveNode)::Int</code></pre><p>Return the optimal index of the bridge to chose from <code>node</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/graph.jl#L303-L309">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Bridges.is_variable_edge_best" href="#MathOptInterface.Bridges.is_variable_edge_best"><code>MathOptInterface.Bridges.is_variable_edge_best</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">is_variable_edge_best(graph::Graph, node::VariableNode)::Bool</code></pre><p>Return a <code>Bool</code> indicating whether <code>node</code> should be added as a variable constrained on creation, or as a free variable followed by a constraint.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Bridges/graph.jl#L325-L330">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../list_of_bridges/">« List of bridges</a><a class="docs-footer-nextpage" href="../../FileFormats/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/FileFormats/overview/index.html b/previews/PR3562/moi/submodules/FileFormats/overview/index.html
index 9fe3ed46a8d..6291fc0ba2a 100644
--- a/previews/PR3562/moi/submodules/FileFormats/overview/index.html
+++ b/previews/PR3562/moi/submodules/FileFormats/overview/index.html
@@ -138,4 +138,4 @@
 path:         [variables][1]
 instance:     Dict{String, Any}(&quot;NaMe&quot; =&gt; &quot;x&quot;)
 schema key:   required
-schema value: Any[&quot;name&quot;]</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Bridges/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+schema value: Any[&quot;name&quot;]</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Bridges/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/FileFormats/reference/index.html b/previews/PR3562/moi/submodules/FileFormats/reference/index.html
index 15bab0e7ec5..a72804706cd 100644
--- a/previews/PR3562/moi/submodules/FileFormats/reference/index.html
+++ b/previews/PR3562/moi/submodules/FileFormats/reference/index.html
@@ -19,4 +19,4 @@
 \end{align}\]</p><p>In other words, the standard conic form contains nonnegative and positive semidefinite variables with equality constraints. That is, in the MathOptInterface&#39;s terminology, <a href="../../../reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a>-in-<a href="../../../reference/standard_form/#MathOptInterface.Nonnegatives"><code>MOI.Nonnegatives</code></a>, <a href="../../../reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a>-in-<a href="../../../reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle"><code>MOI.PositiveSemidefiniteConeTriangle</code></a> and <a href="../../../reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction</code></a>-in-<a href="../../../reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo</code></a> constraints.</p><p>If a model is in standard conic form, use <code>Dualization.jl</code> to transform it into the geometric conic form before writting it. Otherwise, the nonnegative (resp. positive semidefinite) variables will be bridged into free variables with affine constraints constraining them to belong to the nonnegative orthant (resp. positive semidefinite cone) by the <a href="../../Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorFunctionizeBridge"><code>MOI.Bridges.Constraint.VectorFunctionizeBridge</code></a>. Moreover, equality constraints will be bridged into pairs of affine constraints in the nonnegative orthant by the <a href="../../Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitIntervalBridge"><code>MOI.Bridges.Constraint.SplitIntervalBridge</code></a> and then the <a href="../../Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorizeBridge"><code>MOI.Bridges.Constraint.VectorizeBridge</code></a>.</p><p>If a solver is in standard conic form, use <code>Dualization.jl</code> to transform the model read into standard conic form before copying it to the solver. Otherwise, the free variables will be bridged into pairs of variables in the nonnegative orthant by the <a href="../../Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.FreeBridge"><code>MOI.Bridges.Variable.FreeBridge</code></a> and affine constraints will be bridged into equality constraints by creating a slack variable by the <a href="../../Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorSlackBridge"><code>MOI.Bridges.Constraint.VectorSlackBridge</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/FileFormats/SDPA/SDPA.jl#L56-L123">source</a></section></article><h2 id="Other-helpers"><a class="docs-heading-anchor" href="#Other-helpers">Other helpers</a><a id="Other-helpers-1"></a><a class="docs-heading-anchor-permalink" href="#Other-helpers" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.FileFormats.NL.SolFileResults" href="#MathOptInterface.FileFormats.NL.SolFileResults"><code>MathOptInterface.FileFormats.NL.SolFileResults</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">SolFileResults(filename::String, model::Model)</code></pre><p>Parse the <code>.sol</code> file <code>filename</code> created by solving <code>model</code> and return a <code>SolFileResults</code> struct.</p><p>The returned struct supports the <code>MOI.get</code> API for querying result attributes such as <a href="../../../reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a>, <a href="../../../reference/variables/#MathOptInterface.VariablePrimal"><code>MOI.VariablePrimal</code></a>, and <a href="../../../reference/constraints/#MathOptInterface.ConstraintDual"><code>MOI.ConstraintDual</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/FileFormats/NL/sol.jl#L21-L30">source</a></section><section><div><pre><code class="language-julia hljs">SolFileResults(
     raw_status::String,
     termination_status::MOI.TerminationStatusCode,
-)</code></pre><p>Return a <code>SolFileResults</code> struct with <a href="../../../reference/models/#MathOptInterface.RawStatusString"><code>MOI.RawStatusString</code></a> set to <code>raw_status</code>, <a href="../../../reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a> set to <code>termination_status</code>, and <a href="../../../reference/models/#MathOptInterface.PrimalStatus"><code>MOI.PrimalStatus</code></a> and <a href="../../../reference/models/#MathOptInterface.DualStatus"><code>MOI.DualStatus</code></a> set to <code>NO_SOLUTION</code>.</p><p>All other attributes are un-set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/FileFormats/NL/sol.jl#L35-L46">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Nonlinear/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><p>Return a <code>SolFileResults</code> struct with <a href="../../../reference/models/#MathOptInterface.RawStatusString"><code>MOI.RawStatusString</code></a> set to <code>raw_status</code>, <a href="../../../reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a> set to <code>termination_status</code>, and <a href="../../../reference/models/#MathOptInterface.PrimalStatus"><code>MOI.PrimalStatus</code></a> and <a href="../../../reference/models/#MathOptInterface.DualStatus"><code>MOI.DualStatus</code></a> set to <code>NO_SOLUTION</code>.</p><p>All other attributes are un-set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/FileFormats/NL/sol.jl#L35-L46">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Nonlinear/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Nonlinear/overview/index.html b/previews/PR3562/moi/submodules/Nonlinear/overview/index.html
index ffa8af45b88..839b273988d 100644
--- a/previews/PR3562/moi/submodules/Nonlinear/overview/index.html
+++ b/previews/PR3562/moi/submodules/Nonlinear/overview/index.html
@@ -184,4 +184,4 @@
                Node(NODE_VARIABLE, 1, 1),
            ],
            [2.0],
-       );</code></pre><p>This is less readable than the other options, but does this data structure meet our design goals?</p><p>Instead of a heap-allocated object for each node, we only have two <code>Vector</code>s for each expression, <code>nodes</code> and <code>values</code>, as well as two constant vectors for the <code>OPERATORS</code>. In addition, all fields are concretely typed, and there are no <code>Union</code> or <code>Any</code> types.</p><p>For our third goal, it is not easy to identify the children of a node, but it is easy to identify the <em>parent</em> of any node. Therefore, we can use <a href="../reference/#MathOptInterface.Nonlinear.adjacency_matrix"><code>Nonlinear.adjacency_matrix</code></a> to compute a sparse matrix that maps parents to their children.</p><p>The tape is also ordered topologically, so that a reverse pass of the nodes evaluates all children nodes before their parent.</p><h3 id="The-design-in-practice"><a class="docs-heading-anchor" href="#The-design-in-practice">The design in practice</a><a id="The-design-in-practice-1"></a><a class="docs-heading-anchor-permalink" href="#The-design-in-practice" title="Permalink"></a></h3><p>In practice, <code>Node</code> and <code>Expression</code> are exactly <a href="../reference/#MathOptInterface.Nonlinear.Node"><code>Nonlinear.Node</code></a> and <a href="../reference/#MathOptInterface.Nonlinear.Expression"><code>Nonlinear.Expression</code></a>. However, <a href="../reference/#MathOptInterface.Nonlinear.NodeType"><code>Nonlinear.NodeType</code></a> has more fields to account for comparison operators such as <code>:&gt;=</code> and <code>:&lt;=</code>, logic operators such as <code>:&amp;&amp;</code> and <code>:||</code>, nonlinear parameters, and nested subexpressions.</p><p>Moreover, instead of storing the operators as global constants, they are stored in <a href="../reference/#MathOptInterface.Nonlinear.OperatorRegistry"><code>Nonlinear.OperatorRegistry</code></a>, and it also stores a vector of logic operators and a vector of comparison operators. In addition to <a href="../reference/#MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS"><code>Nonlinear.DEFAULT_UNIVARIATE_OPERATORS</code></a> and <a href="../reference/#MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS"><code>Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS</code></a>, you can register user-defined functions using <a href="../reference/#MathOptInterface.Nonlinear.register_operator"><code>Nonlinear.register_operator</code></a>.</p><p><a href="../reference/#MathOptInterface.Nonlinear.Model"><code>Nonlinear.Model</code></a> is a struct that stores the <a href="../reference/#MathOptInterface.Nonlinear.OperatorRegistry"><code>Nonlinear.OperatorRegistry</code></a>, as well as a list of parameters and subexpressions in the model.</p><h2 id="ReverseAD"><a class="docs-heading-anchor" href="#ReverseAD">ReverseAD</a><a id="ReverseAD-1"></a><a class="docs-heading-anchor-permalink" href="#ReverseAD" title="Permalink"></a></h2><p><code>Nonlinear.ReverseAD</code> is a submodule for computing derivatives of a nonlinear optimization problem using sparse reverse-mode automatic differentiation (AD).</p><p>This section does not attempt to explain how sparse reverse-mode AD works, but instead explains why MOI contains its own implementation, and highlights notable differences from similar packages.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Don&#39;t use the API in <code>ReverseAD</code> to compute derivatives. Instead, create a <a href="../reference/#MathOptInterface.Nonlinear.Evaluator"><code>Nonlinear.Evaluator</code></a> object with <a href="../reference/#MathOptInterface.Nonlinear.SparseReverseMode"><code>Nonlinear.SparseReverseMode</code></a> as the backend, and then query the MOI API methods.</p></div></div><h3 id="Design-goals"><a class="docs-heading-anchor" href="#Design-goals">Design goals</a><a id="Design-goals-1"></a><a class="docs-heading-anchor-permalink" href="#Design-goals" title="Permalink"></a></h3><p>The JuliaDiff organization maintains a <a href="https://juliadiff.org">list of packages</a> for doing AD in Julia. At last count, there were at least ten packages——not including <code>ReverseAD</code>——for reverse-mode AD in Julia. <code>ReverseAD</code> exists because it has a different set of design goals.</p><ul><li><strong>Goal: handle scale and sparsity.</strong> The types of nonlinear optimization problems that MOI represents can be large scale (10^5 or more functions across 10^5 or more variables) with very sparse derivatives. The ability to compute a sparse Hessian matrix is essential. To the best of our knowledge, <code>ReverseAD</code> is the only reverse-mode AD system in Julia that handles sparsity by default.</li><li><strong>Goal: limit the scope to improve robustness.</strong> Most other AD packages accept arbitrary Julia functions as input and then trace an expression graph using operator overloading. This means they must deal (or detect and ignore) with control flow, I/O, and other vagaries of Julia. In contrast, <code>ReverseAD</code> only accepts functions in the form of <a href="../reference/#MathOptInterface.Nonlinear.Expression"><code>Nonlinear.Expression</code></a>, which greatly limits the range of syntax that it must deal with. By reducing the scope of what we accept as input to functions relevant for mathematical optimization, we can provide a simpler implementation with various performance optimizations.</li><li><strong>Goal: provide outputs which match what solvers expect.</strong> Other AD packages focus on differentiating individual Julia functions. In contrast, <code>ReverseAD</code> has a very specific use-case: to generate outputs needed by the MOI nonlinear API. This means it needs to efficiently compute sparse Hessians, and it needs subexpression handling to avoid recomputing subexpressions that are shared between functions.</li></ul><h3 id="History"><a class="docs-heading-anchor" href="#History">History</a><a id="History-1"></a><a class="docs-heading-anchor-permalink" href="#History" title="Permalink"></a></h3><p><code>ReverseAD</code> started life as <a href="https://github.com/mlubin/ReverseDiffSparse.jl">ReverseDiffSparse.jl</a>, development of which began in early 2014(!). This was well before the other AD packages started development. Because we had a well-tested, working AD in JuMP, there was less motivation to contribute to and explore other AD packages. The lack of historical interaction also meant that other packages were not optimized for the types of problems that JuMP is built for (that is, large-scale sparse problems). When we first created MathOptInterface, we kept the AD in JuMP to simplify the transition, and post-poned the development of a first-class nonlinear interface in MathOptInterface.</p><p>Prior to the introduction of <code>Nonlinear</code>, JuMP&#39;s nonlinear implementation was a confusing mix of functions and types spread across the code base and in the private <code>_Derivatives</code> submodule. This made it hard to swap the AD system for another. The main motivation for refactoring JuMP to create the <code>Nonlinear</code> submodule in MathOptInterface was to abstract the interface between JuMP and the AD system, allowing us to swap-in and test new AD systems in the future.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../FileFormats/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+       );</code></pre><p>This is less readable than the other options, but does this data structure meet our design goals?</p><p>Instead of a heap-allocated object for each node, we only have two <code>Vector</code>s for each expression, <code>nodes</code> and <code>values</code>, as well as two constant vectors for the <code>OPERATORS</code>. In addition, all fields are concretely typed, and there are no <code>Union</code> or <code>Any</code> types.</p><p>For our third goal, it is not easy to identify the children of a node, but it is easy to identify the <em>parent</em> of any node. Therefore, we can use <a href="../reference/#MathOptInterface.Nonlinear.adjacency_matrix"><code>Nonlinear.adjacency_matrix</code></a> to compute a sparse matrix that maps parents to their children.</p><p>The tape is also ordered topologically, so that a reverse pass of the nodes evaluates all children nodes before their parent.</p><h3 id="The-design-in-practice"><a class="docs-heading-anchor" href="#The-design-in-practice">The design in practice</a><a id="The-design-in-practice-1"></a><a class="docs-heading-anchor-permalink" href="#The-design-in-practice" title="Permalink"></a></h3><p>In practice, <code>Node</code> and <code>Expression</code> are exactly <a href="../reference/#MathOptInterface.Nonlinear.Node"><code>Nonlinear.Node</code></a> and <a href="../reference/#MathOptInterface.Nonlinear.Expression"><code>Nonlinear.Expression</code></a>. However, <a href="../reference/#MathOptInterface.Nonlinear.NodeType"><code>Nonlinear.NodeType</code></a> has more fields to account for comparison operators such as <code>:&gt;=</code> and <code>:&lt;=</code>, logic operators such as <code>:&amp;&amp;</code> and <code>:||</code>, nonlinear parameters, and nested subexpressions.</p><p>Moreover, instead of storing the operators as global constants, they are stored in <a href="../reference/#MathOptInterface.Nonlinear.OperatorRegistry"><code>Nonlinear.OperatorRegistry</code></a>, and it also stores a vector of logic operators and a vector of comparison operators. In addition to <a href="../reference/#MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS"><code>Nonlinear.DEFAULT_UNIVARIATE_OPERATORS</code></a> and <a href="../reference/#MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS"><code>Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS</code></a>, you can register user-defined functions using <a href="../reference/#MathOptInterface.Nonlinear.register_operator"><code>Nonlinear.register_operator</code></a>.</p><p><a href="../reference/#MathOptInterface.Nonlinear.Model"><code>Nonlinear.Model</code></a> is a struct that stores the <a href="../reference/#MathOptInterface.Nonlinear.OperatorRegistry"><code>Nonlinear.OperatorRegistry</code></a>, as well as a list of parameters and subexpressions in the model.</p><h2 id="ReverseAD"><a class="docs-heading-anchor" href="#ReverseAD">ReverseAD</a><a id="ReverseAD-1"></a><a class="docs-heading-anchor-permalink" href="#ReverseAD" title="Permalink"></a></h2><p><code>Nonlinear.ReverseAD</code> is a submodule for computing derivatives of a nonlinear optimization problem using sparse reverse-mode automatic differentiation (AD).</p><p>This section does not attempt to explain how sparse reverse-mode AD works, but instead explains why MOI contains its own implementation, and highlights notable differences from similar packages.</p><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>Don&#39;t use the API in <code>ReverseAD</code> to compute derivatives. Instead, create a <a href="../reference/#MathOptInterface.Nonlinear.Evaluator"><code>Nonlinear.Evaluator</code></a> object with <a href="../reference/#MathOptInterface.Nonlinear.SparseReverseMode"><code>Nonlinear.SparseReverseMode</code></a> as the backend, and then query the MOI API methods.</p></div></div><h3 id="Design-goals"><a class="docs-heading-anchor" href="#Design-goals">Design goals</a><a id="Design-goals-1"></a><a class="docs-heading-anchor-permalink" href="#Design-goals" title="Permalink"></a></h3><p>The JuliaDiff organization maintains a <a href="https://juliadiff.org">list of packages</a> for doing AD in Julia. At last count, there were at least ten packages——not including <code>ReverseAD</code>——for reverse-mode AD in Julia. <code>ReverseAD</code> exists because it has a different set of design goals.</p><ul><li><strong>Goal: handle scale and sparsity.</strong> The types of nonlinear optimization problems that MOI represents can be large scale (10^5 or more functions across 10^5 or more variables) with very sparse derivatives. The ability to compute a sparse Hessian matrix is essential. To the best of our knowledge, <code>ReverseAD</code> is the only reverse-mode AD system in Julia that handles sparsity by default.</li><li><strong>Goal: limit the scope to improve robustness.</strong> Most other AD packages accept arbitrary Julia functions as input and then trace an expression graph using operator overloading. This means they must deal (or detect and ignore) with control flow, I/O, and other vagaries of Julia. In contrast, <code>ReverseAD</code> only accepts functions in the form of <a href="../reference/#MathOptInterface.Nonlinear.Expression"><code>Nonlinear.Expression</code></a>, which greatly limits the range of syntax that it must deal with. By reducing the scope of what we accept as input to functions relevant for mathematical optimization, we can provide a simpler implementation with various performance optimizations.</li><li><strong>Goal: provide outputs which match what solvers expect.</strong> Other AD packages focus on differentiating individual Julia functions. In contrast, <code>ReverseAD</code> has a very specific use-case: to generate outputs needed by the MOI nonlinear API. This means it needs to efficiently compute sparse Hessians, and it needs subexpression handling to avoid recomputing subexpressions that are shared between functions.</li></ul><h3 id="History"><a class="docs-heading-anchor" href="#History">History</a><a id="History-1"></a><a class="docs-heading-anchor-permalink" href="#History" title="Permalink"></a></h3><p><code>ReverseAD</code> started life as <a href="https://github.com/mlubin/ReverseDiffSparse.jl">ReverseDiffSparse.jl</a>, development of which began in early 2014(!). This was well before the other AD packages started development. Because we had a well-tested, working AD in JuMP, there was less motivation to contribute to and explore other AD packages. The lack of historical interaction also meant that other packages were not optimized for the types of problems that JuMP is built for (that is, large-scale sparse problems). When we first created MathOptInterface, we kept the AD in JuMP to simplify the transition, and post-poned the development of a first-class nonlinear interface in MathOptInterface.</p><p>Prior to the introduction of <code>Nonlinear</code>, JuMP&#39;s nonlinear implementation was a confusing mix of functions and types spread across the code base and in the private <code>_Derivatives</code> submodule. This made it hard to swap the AD system for another. The main motivation for refactoring JuMP to create the <code>Nonlinear</code> submodule in MathOptInterface was to abstract the interface between JuMP and the AD system, allowing us to swap-in and test new AD systems in the future.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../FileFormats/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Nonlinear/reference/index.html b/previews/PR3562/moi/submodules/Nonlinear/reference/index.html
index b69e2e67b8a..d6b26ece3b7 100644
--- a/previews/PR3562/moi/submodules/Nonlinear/reference/index.html
+++ b/previews/PR3562/moi/submodules/Nonlinear/reference/index.html
@@ -148,4 +148,4 @@
 delete(model, c1)
 evaluator = Evaluator(model)
 MOI.initialize(evaluator, Symbol[])
-ordinal_index(model, c2)  # Returns 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Nonlinear/evaluator.jl#L11-L31">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Utilities/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ordinal_index(model, c2)  # Returns 1</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Nonlinear/evaluator.jl#L11-L31">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Utilities/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Test/overview/index.html b/previews/PR3562/moi/submodules/Test/overview/index.html
index 9e245951fef..7f2fc534456 100644
--- a/previews/PR3562/moi/submodules/Test/overview/index.html
+++ b/previews/PR3562/moi/submodules/Test/overview/index.html
@@ -167,4 +167,4 @@
         ),
     )
     return
-end</code></pre><p>Finally, you also need to implement <a href="../reference/#MathOptInterface.Test.version_added"><code>Test.version_added</code></a>. If we added this test when the latest released version of MOI was <code>v0.10.5</code>, define:</p><pre><code class="language-julia hljs">version_added(::typeof(test_unit_optimize!_twice)) = v&quot;0.10.6&quot;</code></pre><p><strong>Step 6</strong></p><p>Commit the changes to git from <code>~/.julia/dev/MathOptInterface</code> and submit the PR for review.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>If you need help writing a test, <a href="https://github.com/jump-dev/MathOptInterface.jl/issues/new">open an issue on GitHub</a>, or ask the <a href="https://jump.dev/chatroom/">Developer Chatroom</a>.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Utilities/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><p>Finally, you also need to implement <a href="../reference/#MathOptInterface.Test.version_added"><code>Test.version_added</code></a>. If we added this test when the latest released version of MOI was <code>v0.10.5</code>, define:</p><pre><code class="language-julia hljs">version_added(::typeof(test_unit_optimize!_twice)) = v&quot;0.10.6&quot;</code></pre><p><strong>Step 6</strong></p><p>Commit the changes to git from <code>~/.julia/dev/MathOptInterface</code> and submit the PR for review.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>If you need help writing a test, <a href="https://github.com/jump-dev/MathOptInterface.jl/issues/new">open an issue on GitHub</a>, or ask the <a href="https://jump.dev/chatroom/">Developer Chatroom</a>.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Utilities/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Test/reference/index.html b/previews/PR3562/moi/submodules/Test/reference/index.html
index 4598f1da4b3..4f1323f9b56 100644
--- a/previews/PR3562/moi/submodules/Test/reference/index.html
+++ b/previews/PR3562/moi/submodules/Test/reference/index.html
@@ -51,4 +51,4 @@
     end
     return reset_function
 end</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L118-L155">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Test.version_added" href="#MathOptInterface.Test.version_added"><code>MathOptInterface.Test.version_added</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">version_added(::typeof(function_name))</code></pre><p>Returns the version of MOI in which the test <code>function_name</code> was added.</p><p>This method should be implemented for all new tests.</p><p>See the <code>exclude_tests_after</code> keyword of <a href="#MathOptInterface.Test.runtests"><code>runtests</code></a> for more details.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L158-L166">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Test.@requires" href="#MathOptInterface.Test.@requires"><code>MathOptInterface.Test.@requires</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">@requires(x)</code></pre><p>Check that the condition <code>x</code> is <code>true</code>. Otherwise, throw an <a href="#MathOptInterface.Test.RequirementUnmet"><code>RequirementUnmet</code></a> error to indicate that the model does not support something required by the test function.</p><p><strong>Examples</strong></p><pre><code class="language-julia hljs">@requires MOI.supports(model, MOI.Silent())
-@test MOI.get(model, MOI.Silent())</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L288-L301">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Test.RequirementUnmet" href="#MathOptInterface.Test.RequirementUnmet"><code>MathOptInterface.Test.RequirementUnmet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RequirementUnmet(msg::String) &lt;: Exception</code></pre><p>An error for throwing in tests to indicate that the model does not support some requirement expected by the test function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L273-L278">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../../developer/checklists/">Checklists »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@test MOI.get(model, MOI.Silent())</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L288-L301">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Test.RequirementUnmet" href="#MathOptInterface.Test.RequirementUnmet"><code>MathOptInterface.Test.RequirementUnmet</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">RequirementUnmet(msg::String) &lt;: Exception</code></pre><p>An error for throwing in tests to indicate that the model does not support some requirement expected by the test function.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Test/Test.jl#L273-L278">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../../developer/checklists/">Checklists »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Utilities/overview/index.html b/previews/PR3562/moi/submodules/Utilities/overview/index.html
index bbcdac60aae..5f50a36085e 100644
--- a/previews/PR3562/moi/submodules/Utilities/overview/index.html
+++ b/previews/PR3562/moi/submodules/Utilities/overview/index.html
@@ -251,4 +251,4 @@
 index_map = MOI.copy_to(dest, src)
 for (F, S) in MOI.get(src, MOI.ListOfConstraintTypesPresent())
     function_barrier(dest, src, index_map[F, S])
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Nonlinear/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../Nonlinear/reference/">« API Reference</a><a class="docs-footer-nextpage" href="../reference/">API Reference »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/submodules/Utilities/reference/index.html b/previews/PR3562/moi/submodules/Utilities/reference/index.html
index b27983771b0..2b332a4bc0c 100644
--- a/previews/PR3562/moi/submodules/Utilities/reference/index.html
+++ b/previews/PR3562/moi/submodules/Utilities/reference/index.html
@@ -87,7 +87,7 @@
         typeof(CleverDicts.key_to_index),
         typeof(CleverDicts.index_to_key),
     }
-end</code></pre><p>A struct storing <code>F</code>-in-<code>S</code> constraints as a mapping between the constraint indices to the corresponding tuple of function and set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/vector_of_constraints.jl#L24-L39">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.StructOfConstraints" href="#MathOptInterface.Utilities.StructOfConstraints"><code>MathOptInterface.Utilities.StructOfConstraints</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">abstract type StructOfConstraints &lt;: MOI.ModelLike end</code></pre><p>A struct storing a subfields other structs storing constraints of different types.</p><p>See <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>Utilities.@struct_of_constraints_by_function_types</code></a> and <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>Utilities.@struct_of_constraints_by_set_types</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L7-L15">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.@struct_of_constraints_by_function_types" href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>MathOptInterface.Utilities.@struct_of_constraints_by_function_types</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">Utilities.@struct_of_constraints_by_function_types(name, func_types...)</code></pre><p>Given a vector of <code>n</code> function types <code>(F1, F2,..., Fn)</code> in <code>func_types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, C1, C2, ..., Cn}</code>. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> and stores the constraints of function type <code>Fi</code>.</p><p>The expression <code>Fi</code> can also be a union in which case any constraint for which the function type is in the union is stored in the field with type <code>Ci</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L396-L407">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.@struct_of_constraints_by_set_types" href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>MathOptInterface.Utilities.@struct_of_constraints_by_set_types</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">Utilities.@struct_of_constraints_by_set_types(name, func_types...)</code></pre><p>Given a vector of <code>n</code> set types <code>(S1, S2,..., Sn)</code> in <code>func_types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, C1, C2, ..., Cn}</code>. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> and stores the constraints of set type <code>Si</code>. The expression <code>Si</code> can also be a union in which case any constraint for which the set type is in the union is stored in the field with type <code>Ci</code>. This can be useful if <code>Ci</code> is a <a href="#MathOptInterface.Utilities.MatrixOfConstraints"><code>MatrixOfConstraints</code></a> in order to concatenate the coefficients of constraints of several different set types in the same matrix.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L413-L426">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.struct_of_constraint_code" href="#MathOptInterface.Utilities.struct_of_constraint_code"><code>MathOptInterface.Utilities.struct_of_constraint_code</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">struct_of_constraint_code(struct_name, types, field_types = nothing)</code></pre><p>Given a vector of <code>n</code> <code>Union{SymbolFun,_UnionSymbolFS{SymbolFun}}</code> or <code>Union{SymbolSet,_UnionSymbolFS{SymbolSet}}</code> in <code>types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, F1, F2, ..., Fn}</code> if <code>field_types</code> is <code>nothing</code> and a <code>{T}</code> otherwise. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> if <code>field_types</code> is <code>nothing</code> and type <code>field_types[i]</code> otherwise. If <code>types</code> is vector of <code>Union{SymbolFun,_UnionSymbolFS{SymbolFun}}</code> (resp. <code>Union{SymbolSet,_UnionSymbolFS{SymbolSet}}</code>) then the constraints of that function (resp. set) type are stored in the corresponding field.</p><p>This function is used by the macros <a href="#MathOptInterface.Utilities.@model"><code>@model</code></a>, <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>@struct_of_constraints_by_function_types</code></a> and <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>@struct_of_constraints_by_set_types</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L263-L280">source</a></section></article><h2 id="Caching-optimizer"><a class="docs-heading-anchor" href="#Caching-optimizer">Caching optimizer</a><a id="Caching-optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#Caching-optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.CachingOptimizer" href="#MathOptInterface.Utilities.CachingOptimizer"><code>MathOptInterface.Utilities.CachingOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">CachingOptimizer</code></pre><p><code>CachingOptimizer</code> is an intermediate layer that stores a cache of the model and links it with an optimizer. It supports incremental model construction and modification even when the optimizer doesn&#39;t.</p><p><strong>Constructors</strong></p><pre><code class="language-julia hljs">    CachingOptimizer(cache::MOI.ModelLike, optimizer::AbstractOptimizer)</code></pre><p>Creates a <code>CachingOptimizer</code> in <code>AUTOMATIC</code> mode, with the optimizer <code>optimizer</code>.</p><p>The type of the optimizer returned is <code>CachingOptimizer{typeof(optimizer), typeof(cache)}</code> so it does not support the function <code>reset_optimizer(::CachingOptimizer, new_optimizer)</code> if the type of <code>new_optimizer</code> is different from the type of <code>optimizer</code>.</p><pre><code class="language-julia hljs">    CachingOptimizer(cache::MOI.ModelLike, mode::CachingOptimizerMode)</code></pre><p>Creates a <code>CachingOptimizer</code> in the <code>NO_OPTIMIZER</code> state and mode <code>mode</code>.</p><p>The type of the optimizer returned is <code>CachingOptimizer{MOI.AbstractOptimizer,typeof(cache)}</code> so it <em>does</em> support the function <code>reset_optimizer(::CachingOptimizer, new_optimizer)</code> if the type of <code>new_optimizer</code> is different from the type of <code>optimizer</code>.</p><p><strong>About the type</strong></p><p><strong>States</strong></p><p>A <code>CachingOptimizer</code> may be in one of three possible states (<code>CachingOptimizerState</code>):</p><ul><li><code>NO_OPTIMIZER</code>: The CachingOptimizer does not have any optimizer.</li><li><code>EMPTY_OPTIMIZER</code>: The CachingOptimizer has an empty optimizer. The optimizer is not synchronized with the cached model.</li><li><code>ATTACHED_OPTIMIZER</code>: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.</li></ul><p><strong>Modes</strong></p><p>A <code>CachingOptimizer</code> has two modes of operation (<code>CachingOptimizerMode</code>):</p><ul><li><code>MANUAL</code>: The only methods that change the state of the <code>CachingOptimizer</code> are <a href="#MathOptInterface.Utilities.reset_optimizer"><code>Utilities.reset_optimizer</code></a>, <a href="#MathOptInterface.Utilities.drop_optimizer"><code>Utilities.drop_optimizer</code></a>, and <a href="#MathOptInterface.Utilities.attach_optimizer"><code>Utilities.attach_optimizer</code></a>. Attempting to perform an operation in the incorrect state results in an error.</li><li><code>AUTOMATIC</code>: The <code>CachingOptimizer</code> changes its state when necessary. For example, <code>optimize!</code> will automatically call <code>attach_optimizer</code> (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to <code>EMPTY_OPTIMIZER</code> mode.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L18-L76">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.attach_optimizer" href="#MathOptInterface.Utilities.attach_optimizer"><code>MathOptInterface.Utilities.attach_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">attach_optimizer(model::CachingOptimizer)</code></pre><p>Attaches the optimizer to <code>model</code>, copying all model data into it. Can be called only from the <code>EMPTY_OPTIMIZER</code> state. If the copy succeeds, the <code>CachingOptimizer</code> will be in state <code>ATTACHED_OPTIMIZER</code> after the call, otherwise an error is thrown; see <a href="../../../reference/models/#MathOptInterface.copy_to"><code>MOI.copy_to</code></a> for more details on which errors can be thrown.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L208-L216">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.attach_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.attach_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L313-L319">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.reset_optimizer" href="#MathOptInterface.Utilities.reset_optimizer"><code>MathOptInterface.Utilities.reset_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reset_optimizer(m::CachingOptimizer, optimizer::MOI.AbstractOptimizer)</code></pre><p>Sets or resets <code>m</code> to have the given empty optimizer <code>optimizer</code>.</p><p>Can be called from any state. An assertion error will be thrown if <code>optimizer</code> is not empty.</p><p>The <code>CachingOptimizer</code> <code>m</code> will be in state <code>EMPTY_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L163-L172">source</a></section><section><div><pre><code class="language-julia hljs">reset_optimizer(m::CachingOptimizer)</code></pre><p>Detaches and empties the current optimizer. Can be called from <code>ATTACHED_OPTIMIZER</code> or <code>EMPTY_OPTIMIZER</code> state. The <code>CachingOptimizer</code> will be in state <code>EMPTY_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L181-L187">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel, optimizer::MOI.AbstractOptimizer)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L268-L274">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L285-L291">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.drop_optimizer" href="#MathOptInterface.Utilities.drop_optimizer"><code>MathOptInterface.Utilities.drop_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">drop_optimizer(m::CachingOptimizer)</code></pre><p>Drops the optimizer, if one is present. Can be called from any state. The <code>CachingOptimizer</code> will be in state <code>NO_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L196-L201">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.drop_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.drop_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/6ec8617b430058b221acce299ddc0b8255351c10/src/optimizer_interface.jl#L300-L306">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.state" href="#MathOptInterface.Utilities.state"><code>MathOptInterface.Utilities.state</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">state(m::CachingOptimizer)::CachingOptimizerState</code></pre><p>Returns the state of the CachingOptimizer <code>m</code>. See <a href="#MathOptInterface.Utilities.CachingOptimizer"><code>Utilities.CachingOptimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L149-L153">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.mode" href="#MathOptInterface.Utilities.mode"><code>MathOptInterface.Utilities.mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">mode(m::CachingOptimizer)::CachingOptimizerMode</code></pre><p>Returns the operating mode of the CachingOptimizer <code>m</code>. See <a href="#MathOptInterface.Utilities.CachingOptimizer"><code>Utilities.CachingOptimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L156-L160">source</a></section></article><h2 id="Mock-optimizer"><a class="docs-heading-anchor" href="#Mock-optimizer">Mock optimizer</a><a id="Mock-optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#Mock-optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.MockOptimizer" href="#MathOptInterface.Utilities.MockOptimizer"><code>MathOptInterface.Utilities.MockOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">MockOptimizer</code></pre><p><code>MockOptimizer</code> is a fake optimizer especially useful for testing. Its main feature is that it can store the values that should be returned for each attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/mockoptimizer.jl#L17-L23">source</a></section></article><h2 id="Printing"><a class="docs-heading-anchor" href="#Printing">Printing</a><a id="Printing-1"></a><a class="docs-heading-anchor-permalink" href="#Printing" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.latex_formulation" href="#MathOptInterface.Utilities.latex_formulation"><code>MathOptInterface.Utilities.latex_formulation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">latex_formulation(model::MOI.ModelLike; kwargs...)</code></pre><p>Wrap <code>model</code> in a type so that it can be pretty-printed as <code>text/latex</code> in a notebook like IJulia, or in Documenter.</p><p>To render the model, end the cell with <code>latex_formulation(model)</code>, or call <code>display(latex_formulation(model))</code> in to force the display of the model from inside a function.</p><p>Possible keyword arguments are:</p><ul><li><code>simplify_coefficients</code> : Simplify coefficients if possible by omitting them or removing trailing zeros.</li><li><code>default_name</code> : The name given to variables with an empty name.</li><li><code>print_types</code> : Print the MOI type of each function and set for clarity.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/print.jl#L632-L648">source</a></section></article><h2 id="Copy-utilities"><a class="docs-heading-anchor" href="#Copy-utilities">Copy utilities</a><a id="Copy-utilities-1"></a><a class="docs-heading-anchor-permalink" href="#Copy-utilities" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.default_copy_to" href="#MathOptInterface.Utilities.default_copy_to"><code>MathOptInterface.Utilities.default_copy_to</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">default_copy_to(dest::MOI.ModelLike, src::MOI.ModelLike)</code></pre><p>A default implementation of <code>MOI.copy_to(dest, src)</code> for models that implement the incremental interface, i.e., <a href="../../../reference/models/#MathOptInterface.supports_incremental_interface"><code>MOI.supports_incremental_interface</code></a> returns <code>true</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy.jl#L471-L477">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.IndexMap" href="#MathOptInterface.Utilities.IndexMap"><code>MathOptInterface.Utilities.IndexMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">IndexMap()</code></pre><p>The dictionary-like object returned by <a href="../../../reference/models/#MathOptInterface.copy_to"><code>MOI.copy_to</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy/index_map.jl#L25-L29">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.identity_index_map" href="#MathOptInterface.Utilities.identity_index_map"><code>MathOptInterface.Utilities.identity_index_map</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">identity_index_map(model::MOI.ModelLike)</code></pre><p>Return an <a href="#MathOptInterface.Utilities.IndexMap"><code>IndexMap</code></a> that maps all variable and constraint indices of <code>model</code> to themselves.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy/index_map.jl#L46-L51">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.ModelFilter" href="#MathOptInterface.Utilities.ModelFilter"><code>MathOptInterface.Utilities.ModelFilter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelFilter(filter::Function, model::MOI.ModelLike)</code></pre><p>A layer to filter out various components of <code>model</code>.</p><p>The filter function takes a single argument, which is each element from the list returned by the attributes below. It returns <code>true</code> if the element should be visible in the filtered model and <code>false</code> otherwise.</p><p>The components that are filtered are:</p><ul><li>Entire constraint types via:<ul><li><code>MOI.ListOfConstraintTypesPresent</code></li></ul></li><li>Individual constraints via:<ul><li><code>MOI.ListOfConstraintIndices{F,S}</code></li></ul></li><li>Specific attributes via:<ul><li><code>MOI.ListOfModelAttributesSet</code></li><li><code>MOI.ListOfConstraintAttributesSet</code></li><li><code>MOI.ListOfVariableAttributesSet</code></li></ul></li></ul><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The list of attributes filtered may change in a future release. You should write functions that are generic and not limited to the five types listed above. Thus, you should probably define a fallback <code>filter(::Any) = true</code>.</p></div></div><p>See below for examples of how this works.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>This layer has a limited scope. It is intended by be used in conjunction with <code>MOI.copy_to</code>.</p></div></div><p><strong>Example: copy model excluding integer constraints</strong></p><p>Use the <code>do</code> syntax to provide a single function.</p><pre><code class="language-julia hljs">filtered_src = MOI.Utilities.ModelFilter(src) do item
+end</code></pre><p>A struct storing <code>F</code>-in-<code>S</code> constraints as a mapping between the constraint indices to the corresponding tuple of function and set.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/vector_of_constraints.jl#L24-L39">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.StructOfConstraints" href="#MathOptInterface.Utilities.StructOfConstraints"><code>MathOptInterface.Utilities.StructOfConstraints</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">abstract type StructOfConstraints &lt;: MOI.ModelLike end</code></pre><p>A struct storing a subfields other structs storing constraints of different types.</p><p>See <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>Utilities.@struct_of_constraints_by_function_types</code></a> and <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>Utilities.@struct_of_constraints_by_set_types</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L7-L15">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.@struct_of_constraints_by_function_types" href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>MathOptInterface.Utilities.@struct_of_constraints_by_function_types</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">Utilities.@struct_of_constraints_by_function_types(name, func_types...)</code></pre><p>Given a vector of <code>n</code> function types <code>(F1, F2,..., Fn)</code> in <code>func_types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, C1, C2, ..., Cn}</code>. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> and stores the constraints of function type <code>Fi</code>.</p><p>The expression <code>Fi</code> can also be a union in which case any constraint for which the function type is in the union is stored in the field with type <code>Ci</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L396-L407">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.@struct_of_constraints_by_set_types" href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>MathOptInterface.Utilities.@struct_of_constraints_by_set_types</code></a> — <span class="docstring-category">Macro</span></header><section><div><pre><code class="language-julia hljs">Utilities.@struct_of_constraints_by_set_types(name, func_types...)</code></pre><p>Given a vector of <code>n</code> set types <code>(S1, S2,..., Sn)</code> in <code>func_types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, C1, C2, ..., Cn}</code>. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> and stores the constraints of set type <code>Si</code>. The expression <code>Si</code> can also be a union in which case any constraint for which the set type is in the union is stored in the field with type <code>Ci</code>. This can be useful if <code>Ci</code> is a <a href="#MathOptInterface.Utilities.MatrixOfConstraints"><code>MatrixOfConstraints</code></a> in order to concatenate the coefficients of constraints of several different set types in the same matrix.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L413-L426">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.struct_of_constraint_code" href="#MathOptInterface.Utilities.struct_of_constraint_code"><code>MathOptInterface.Utilities.struct_of_constraint_code</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">struct_of_constraint_code(struct_name, types, field_types = nothing)</code></pre><p>Given a vector of <code>n</code> <code>Union{SymbolFun,_UnionSymbolFS{SymbolFun}}</code> or <code>Union{SymbolSet,_UnionSymbolFS{SymbolSet}}</code> in <code>types</code>, defines a subtype of <code>StructOfConstraints</code> of name <code>name</code> and which type parameters <code>{T, F1, F2, ..., Fn}</code> if <code>field_types</code> is <code>nothing</code> and a <code>{T}</code> otherwise. It contains <code>n</code> field where the <code>i</code>th field has type <code>Ci</code> if <code>field_types</code> is <code>nothing</code> and type <code>field_types[i]</code> otherwise. If <code>types</code> is vector of <code>Union{SymbolFun,_UnionSymbolFS{SymbolFun}}</code> (resp. <code>Union{SymbolSet,_UnionSymbolFS{SymbolSet}}</code>) then the constraints of that function (resp. set) type are stored in the corresponding field.</p><p>This function is used by the macros <a href="#MathOptInterface.Utilities.@model"><code>@model</code></a>, <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_function_types"><code>@struct_of_constraints_by_function_types</code></a> and <a href="#MathOptInterface.Utilities.@struct_of_constraints_by_set_types"><code>@struct_of_constraints_by_set_types</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/struct_of_constraints.jl#L263-L280">source</a></section></article><h2 id="Caching-optimizer"><a class="docs-heading-anchor" href="#Caching-optimizer">Caching optimizer</a><a id="Caching-optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#Caching-optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.CachingOptimizer" href="#MathOptInterface.Utilities.CachingOptimizer"><code>MathOptInterface.Utilities.CachingOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">CachingOptimizer</code></pre><p><code>CachingOptimizer</code> is an intermediate layer that stores a cache of the model and links it with an optimizer. It supports incremental model construction and modification even when the optimizer doesn&#39;t.</p><p><strong>Constructors</strong></p><pre><code class="language-julia hljs">    CachingOptimizer(cache::MOI.ModelLike, optimizer::AbstractOptimizer)</code></pre><p>Creates a <code>CachingOptimizer</code> in <code>AUTOMATIC</code> mode, with the optimizer <code>optimizer</code>.</p><p>The type of the optimizer returned is <code>CachingOptimizer{typeof(optimizer), typeof(cache)}</code> so it does not support the function <code>reset_optimizer(::CachingOptimizer, new_optimizer)</code> if the type of <code>new_optimizer</code> is different from the type of <code>optimizer</code>.</p><pre><code class="language-julia hljs">    CachingOptimizer(cache::MOI.ModelLike, mode::CachingOptimizerMode)</code></pre><p>Creates a <code>CachingOptimizer</code> in the <code>NO_OPTIMIZER</code> state and mode <code>mode</code>.</p><p>The type of the optimizer returned is <code>CachingOptimizer{MOI.AbstractOptimizer,typeof(cache)}</code> so it <em>does</em> support the function <code>reset_optimizer(::CachingOptimizer, new_optimizer)</code> if the type of <code>new_optimizer</code> is different from the type of <code>optimizer</code>.</p><p><strong>About the type</strong></p><p><strong>States</strong></p><p>A <code>CachingOptimizer</code> may be in one of three possible states (<code>CachingOptimizerState</code>):</p><ul><li><code>NO_OPTIMIZER</code>: The CachingOptimizer does not have any optimizer.</li><li><code>EMPTY_OPTIMIZER</code>: The CachingOptimizer has an empty optimizer. The optimizer is not synchronized with the cached model.</li><li><code>ATTACHED_OPTIMIZER</code>: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.</li></ul><p><strong>Modes</strong></p><p>A <code>CachingOptimizer</code> has two modes of operation (<code>CachingOptimizerMode</code>):</p><ul><li><code>MANUAL</code>: The only methods that change the state of the <code>CachingOptimizer</code> are <a href="#MathOptInterface.Utilities.reset_optimizer"><code>Utilities.reset_optimizer</code></a>, <a href="#MathOptInterface.Utilities.drop_optimizer"><code>Utilities.drop_optimizer</code></a>, and <a href="#MathOptInterface.Utilities.attach_optimizer"><code>Utilities.attach_optimizer</code></a>. Attempting to perform an operation in the incorrect state results in an error.</li><li><code>AUTOMATIC</code>: The <code>CachingOptimizer</code> changes its state when necessary. For example, <code>optimize!</code> will automatically call <code>attach_optimizer</code> (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to <code>EMPTY_OPTIMIZER</code> mode.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L18-L76">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.attach_optimizer" href="#MathOptInterface.Utilities.attach_optimizer"><code>MathOptInterface.Utilities.attach_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">attach_optimizer(model::CachingOptimizer)</code></pre><p>Attaches the optimizer to <code>model</code>, copying all model data into it. Can be called only from the <code>EMPTY_OPTIMIZER</code> state. If the copy succeeds, the <code>CachingOptimizer</code> will be in state <code>ATTACHED_OPTIMIZER</code> after the call, otherwise an error is thrown; see <a href="../../../reference/models/#MathOptInterface.copy_to"><code>MOI.copy_to</code></a> for more details on which errors can be thrown.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L208-L216">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.attach_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.attach_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L313-L319">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.reset_optimizer" href="#MathOptInterface.Utilities.reset_optimizer"><code>MathOptInterface.Utilities.reset_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">reset_optimizer(m::CachingOptimizer, optimizer::MOI.AbstractOptimizer)</code></pre><p>Sets or resets <code>m</code> to have the given empty optimizer <code>optimizer</code>.</p><p>Can be called from any state. An assertion error will be thrown if <code>optimizer</code> is not empty.</p><p>The <code>CachingOptimizer</code> <code>m</code> will be in state <code>EMPTY_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L163-L172">source</a></section><section><div><pre><code class="language-julia hljs">reset_optimizer(m::CachingOptimizer)</code></pre><p>Detaches and empties the current optimizer. Can be called from <code>ATTACHED_OPTIMIZER</code> or <code>EMPTY_OPTIMIZER</code> state. The <code>CachingOptimizer</code> will be in state <code>EMPTY_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L181-L187">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel, optimizer::MOI.AbstractOptimizer)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L268-L274">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.reset_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.reset_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L285-L291">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.drop_optimizer" href="#MathOptInterface.Utilities.drop_optimizer"><code>MathOptInterface.Utilities.drop_optimizer</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">drop_optimizer(m::CachingOptimizer)</code></pre><p>Drops the optimizer, if one is present. Can be called from any state. The <code>CachingOptimizer</code> will be in state <code>NO_OPTIMIZER</code> after the call.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L196-L201">source</a></section><section><div><pre><code class="language-julia hljs">MOIU.drop_optimizer(model::GenericModel)</code></pre><p>Call <code>MOIU.drop_optimizer</code> on the backend of <code>model</code>.</p><p>Cannot be called in direct mode.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/JuMP.jl/blob/ff869cf288b11f13d43f43e7e7bd7d42f0f49f91/src/optimizer_interface.jl#L300-L306">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.state" href="#MathOptInterface.Utilities.state"><code>MathOptInterface.Utilities.state</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">state(m::CachingOptimizer)::CachingOptimizerState</code></pre><p>Returns the state of the CachingOptimizer <code>m</code>. See <a href="#MathOptInterface.Utilities.CachingOptimizer"><code>Utilities.CachingOptimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L149-L153">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.mode" href="#MathOptInterface.Utilities.mode"><code>MathOptInterface.Utilities.mode</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">mode(m::CachingOptimizer)::CachingOptimizerMode</code></pre><p>Returns the operating mode of the CachingOptimizer <code>m</code>. See <a href="#MathOptInterface.Utilities.CachingOptimizer"><code>Utilities.CachingOptimizer</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/cachingoptimizer.jl#L156-L160">source</a></section></article><h2 id="Mock-optimizer"><a class="docs-heading-anchor" href="#Mock-optimizer">Mock optimizer</a><a id="Mock-optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#Mock-optimizer" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.MockOptimizer" href="#MathOptInterface.Utilities.MockOptimizer"><code>MathOptInterface.Utilities.MockOptimizer</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">MockOptimizer</code></pre><p><code>MockOptimizer</code> is a fake optimizer especially useful for testing. Its main feature is that it can store the values that should be returned for each attribute.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/mockoptimizer.jl#L17-L23">source</a></section></article><h2 id="Printing"><a class="docs-heading-anchor" href="#Printing">Printing</a><a id="Printing-1"></a><a class="docs-heading-anchor-permalink" href="#Printing" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.latex_formulation" href="#MathOptInterface.Utilities.latex_formulation"><code>MathOptInterface.Utilities.latex_formulation</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">latex_formulation(model::MOI.ModelLike; kwargs...)</code></pre><p>Wrap <code>model</code> in a type so that it can be pretty-printed as <code>text/latex</code> in a notebook like IJulia, or in Documenter.</p><p>To render the model, end the cell with <code>latex_formulation(model)</code>, or call <code>display(latex_formulation(model))</code> in to force the display of the model from inside a function.</p><p>Possible keyword arguments are:</p><ul><li><code>simplify_coefficients</code> : Simplify coefficients if possible by omitting them or removing trailing zeros.</li><li><code>default_name</code> : The name given to variables with an empty name.</li><li><code>print_types</code> : Print the MOI type of each function and set for clarity.</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/print.jl#L632-L648">source</a></section></article><h2 id="Copy-utilities"><a class="docs-heading-anchor" href="#Copy-utilities">Copy utilities</a><a id="Copy-utilities-1"></a><a class="docs-heading-anchor-permalink" href="#Copy-utilities" title="Permalink"></a></h2><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.default_copy_to" href="#MathOptInterface.Utilities.default_copy_to"><code>MathOptInterface.Utilities.default_copy_to</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">default_copy_to(dest::MOI.ModelLike, src::MOI.ModelLike)</code></pre><p>A default implementation of <code>MOI.copy_to(dest, src)</code> for models that implement the incremental interface, i.e., <a href="../../../reference/models/#MathOptInterface.supports_incremental_interface"><code>MOI.supports_incremental_interface</code></a> returns <code>true</code>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy.jl#L471-L477">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.IndexMap" href="#MathOptInterface.Utilities.IndexMap"><code>MathOptInterface.Utilities.IndexMap</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">IndexMap()</code></pre><p>The dictionary-like object returned by <a href="../../../reference/models/#MathOptInterface.copy_to"><code>MOI.copy_to</code></a>.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy/index_map.jl#L25-L29">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.identity_index_map" href="#MathOptInterface.Utilities.identity_index_map"><code>MathOptInterface.Utilities.identity_index_map</code></a> — <span class="docstring-category">Function</span></header><section><div><pre><code class="language-julia hljs">identity_index_map(model::MOI.ModelLike)</code></pre><p>Return an <a href="#MathOptInterface.Utilities.IndexMap"><code>IndexMap</code></a> that maps all variable and constraint indices of <code>model</code> to themselves.</p></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/copy/index_map.jl#L46-L51">source</a></section></article><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="MathOptInterface.Utilities.ModelFilter" href="#MathOptInterface.Utilities.ModelFilter"><code>MathOptInterface.Utilities.ModelFilter</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">ModelFilter(filter::Function, model::MOI.ModelLike)</code></pre><p>A layer to filter out various components of <code>model</code>.</p><p>The filter function takes a single argument, which is each element from the list returned by the attributes below. It returns <code>true</code> if the element should be visible in the filtered model and <code>false</code> otherwise.</p><p>The components that are filtered are:</p><ul><li>Entire constraint types via:<ul><li><code>MOI.ListOfConstraintTypesPresent</code></li></ul></li><li>Individual constraints via:<ul><li><code>MOI.ListOfConstraintIndices{F,S}</code></li></ul></li><li>Specific attributes via:<ul><li><code>MOI.ListOfModelAttributesSet</code></li><li><code>MOI.ListOfConstraintAttributesSet</code></li><li><code>MOI.ListOfVariableAttributesSet</code></li></ul></li></ul><div class="admonition is-warning"><header class="admonition-header">Warning</header><div class="admonition-body"><p>The list of attributes filtered may change in a future release. You should write functions that are generic and not limited to the five types listed above. Thus, you should probably define a fallback <code>filter(::Any) = true</code>.</p></div></div><p>See below for examples of how this works.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>This layer has a limited scope. It is intended by be used in conjunction with <code>MOI.copy_to</code>.</p></div></div><p><strong>Example: copy model excluding integer constraints</strong></p><p>Use the <code>do</code> syntax to provide a single function.</p><pre><code class="language-julia hljs">filtered_src = MOI.Utilities.ModelFilter(src) do item
     return item != (MOI.VariableIndex, MOI.Integer)
 end
 MOI.copy_to(dest, filtered_src)</code></pre><p><strong>Example: copy model excluding names</strong></p><p>Use type dispatch to simplify the implementation:</p><pre><code class="language-julia hljs">my_filter(::Any) = true  # Note the generic fallback!
@@ -265,4 +265,4 @@
 For performance, it is recommended that the inner loop lies in a separate
 function to gurantee type-stability.
 
-If you want an iterator of all current outer keys, use [`outer_keys`](@ref).</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/DoubleDicts.jl#L321-L341">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Test/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+If you want an iterator of all current outer keys, use [`outer_keys`](@ref).</code></pre></div><a class="docs-sourcelink" target="_blank" href="https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/src/Utilities/DoubleDicts.jl#L321-L341">source</a></section></article></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../overview/">« Overview</a><a class="docs-footer-nextpage" href="../../Test/overview/">Overview »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/bridging_constraint/index.html b/previews/PR3562/moi/tutorials/bridging_constraint/index.html
index 5bdc85adb90..8be992c5a25 100644
--- a/previews/PR3562/moi/tutorials/bridging_constraint/index.html
+++ b/previews/PR3562/moi/tutorials/bridging_constraint/index.html
@@ -103,4 +103,4 @@
 end</code></pre><h3 id="Bridge-deletion"><a class="docs-heading-anchor" href="#Bridge-deletion">Bridge deletion</a><a id="Bridge-deletion-1"></a><a class="docs-heading-anchor-permalink" href="#Bridge-deletion" title="Permalink"></a></h3><p>When a bridge is deleted, the constraints it added must be deleted too.</p><pre><code class="language-julia hljs">function delete(model::ModelLike, bridge::SignBridge)
     delete(model, bridge.constraint)
     return
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../mathprogbase/">« Transitioning from MathProgBase</a><a class="docs-footer-nextpage" href="../manipulating_expressions/">Manipulating expressions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../mathprogbase/">« Transitioning from MathProgBase</a><a class="docs-footer-nextpage" href="../manipulating_expressions/">Manipulating expressions »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/example/index.html b/previews/PR3562/moi/tutorials/example/index.html
index 3b97b7c6a93..d56197268e8 100644
--- a/previews/PR3562/moi/tutorials/example/index.html
+++ b/previews/PR3562/moi/tutorials/example/index.html
@@ -46,4 +46,4 @@
 3-element Vector{Float64}:
  1.0
  1.0
- 1.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/motivation/">« Motivation</a><a class="docs-footer-nextpage" href="../implementing/">Implementing a solver interface »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 1.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../background/motivation/">« Motivation</a><a class="docs-footer-nextpage" href="../implementing/">Implementing a solver interface »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/implementing/index.html b/previews/PR3562/moi/tutorials/implementing/index.html
index bd3d789ffd4..c9eb19cdf15 100644
--- a/previews/PR3562/moi/tutorials/implementing/index.html
+++ b/previews/PR3562/moi/tutorials/implementing/index.html
@@ -115,4 +115,4 @@
     n = # Code to get NumberOfObjectives
     return n
 end</code></pre><p>Then, the user can write:</p><pre><code class="language-julia hljs">model = Gurobi.Optimizer()
-MOI.set(model, Gurobi.NumberofObjectives(), 3)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../example/">« Solving a problem using MathOptInterface</a><a class="docs-footer-nextpage" href="../mathprogbase/">Transitioning from MathProgBase »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+MOI.set(model, Gurobi.NumberofObjectives(), 3)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../example/">« Solving a problem using MathOptInterface</a><a class="docs-footer-nextpage" href="../mathprogbase/">Transitioning from MathProgBase »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/latency/index.html b/previews/PR3562/moi/tutorials/latency/index.html
index aeeaa991d10..9e90004eb5a 100644
--- a/previews/PR3562/moi/tutorials/latency/index.html
+++ b/previews/PR3562/moi/tutorials/latency/index.html
@@ -130,4 +130,4 @@
 end</code></pre><p>You can create a flame-graph via</p><pre><code class="language-julia hljs">using SnoopComile
 tinf = @snoopi_deep example_diet(GLPK.Optimizer, true)
 using ProfileView
-ProfileView.view(flamegraph(tinf))</code></pre><p>Here&#39;s how things looked in mid-August 2021: <img src="../../assets/latency.png" alt="flamegraph"/></p><p>There are a few opportunities for improvement (non-red flames, particularly on the right). But the main problem is a large red (non-precompilable due to method ownership) flame.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../manipulating_expressions/">« Manipulating expressions</a><a class="docs-footer-nextpage" href="../../manual/standard_form/">Standard form problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ProfileView.view(flamegraph(tinf))</code></pre><p>Here&#39;s how things looked in mid-August 2021: <img src="../../assets/latency.png" alt="flamegraph"/></p><p>There are a few opportunities for improvement (non-red flames, particularly on the right). But the main problem is a large red (non-precompilable due to method ownership) flame.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../manipulating_expressions/">« Manipulating expressions</a><a class="docs-footer-nextpage" href="../../manual/standard_form/">Standard form problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/manipulating_expressions/index.html b/previews/PR3562/moi/tutorials/manipulating_expressions/index.html
index 133bb56d6dc..6d987ff8d08 100644
--- a/previews/PR3562/moi/tutorials/manipulating_expressions/index.html
+++ b/previews/PR3562/moi/tutorials/manipulating_expressions/index.html
@@ -23,4 +23,4 @@
 2-element Vector{MathOptInterface.ScalarAffineFunction{Int64}}:
  (2) + (1) MOI.VariableIndex(1)
  (4) + (2) MOI.VariableIndex(1)</code></pre><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p><a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.eachscalar"><code>Utilities.eachscalar</code></a> returns an iterator on the dimensions, which serves the same purpose as <a href="../../submodules/Utilities/reference/#MathOptInterface.Utilities.scalarize"><code>Utilities.scalarize</code></a>.</p></div></div><p><a href="../../reference/standard_form/#MathOptInterface.output_dimension"><code>output_dimension</code></a> returns the number of dimensions of the output of a function:</p><pre><code class="language-julia-repl hljs">julia&gt; MOI.output_dimension(g)
-2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../bridging_constraint/">« Implementing a constraint bridge</a><a class="docs-footer-nextpage" href="../latency/">Latency »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../bridging_constraint/">« Implementing a constraint bridge</a><a class="docs-footer-nextpage" href="../latency/">Latency »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/moi/tutorials/mathprogbase/index.html b/previews/PR3562/moi/tutorials/mathprogbase/index.html
index 97c6cc5ca4f..d7ab15c1dca 100644
--- a/previews/PR3562/moi/tutorials/mathprogbase/index.html
+++ b/previews/PR3562/moi/tutorials/mathprogbase/index.html
@@ -55,4 +55,4 @@
         objval = objective_value(model),
         sol = value.(x)
     )
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../implementing/">« Implementing a solver interface</a><a class="docs-footer-nextpage" href="../bridging_constraint/">Implementing a constraint bridge »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../implementing/">« Implementing a solver interface</a><a class="docs-footer-nextpage" href="../bridging_constraint/">Implementing a constraint bridge »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Alpine/index.html b/previews/PR3562/packages/Alpine/index.html
index 671aa2ac586..75fb3f2c94e 100644
--- a/previews/PR3562/packages/Alpine/index.html
+++ b/previews/PR3562/packages/Alpine/index.html
@@ -46,4 +46,4 @@
   author={Kim, Jongeun and Richard, Jean-Philippe P. and Tawarmalani, Mohit},
   eprinttype={Optimization Online},
   date={2022}
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Loraine/">« kocvara/Loraine.jl</a><a class="docs-footer-nextpage" href="../Juniper/">lanl-ansi/Juniper.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Loraine/">« kocvara/Loraine.jl</a><a class="docs-footer-nextpage" href="../Juniper/">lanl-ansi/Juniper.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/AmplNLWriter/index.html b/previews/PR3562/packages/AmplNLWriter/index.html
index ef11e4ce652..afa7a0d4317 100644
--- a/previews/PR3562/packages/AmplNLWriter/index.html
+++ b/previews/PR3562/packages/AmplNLWriter/index.html
@@ -12,4 +12,4 @@
 import Bonmin_jll
 model = Model(() -&gt; AmplNLWriter.Optimizer(Bonmin_jll.amplexe))
 set_attribute(model, &quot;bonmin.nlp_log_level&quot;, 0)</code></pre><h3 id="opt-files"><a class="docs-heading-anchor" href="#opt-files">opt files</a><a id="opt-files-1"></a><a class="docs-heading-anchor-permalink" href="#opt-files" title="Permalink"></a></h3><p>Some options need to be specified via an <code>.opt</code> file.</p><p>This file must be located in the current working directory whenever the model is solved.</p><p>The <code>.opt</code> file must be named after the name of the solver, for example, <code>bonmin.opt</code>, and each line must contain an option name and the desired value, separated by a space.</p><p>For example, to set the absolute and relative tolerances in Couenne to <code>1</code> and <code>0.05</code> respectively, the <code>couenne.opt</code> file should contain:</p><pre><code class="language-raw hljs">allowable_gap 1
-allowable_fraction_gap 0.05</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solvers/">« Introduction</a><a class="docs-footer-nextpage" href="../BARON/">jump-dev/BARON.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+allowable_fraction_gap 0.05</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../solvers/">« Introduction</a><a class="docs-footer-nextpage" href="../BARON/">jump-dev/BARON.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/BARON/index.html b/previews/PR3562/packages/BARON/index.html
index 09c8548dc50..2c0a536ab05 100644
--- a/previews/PR3562/packages/BARON/index.html
+++ b/previews/PR3562/packages/BARON/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li class="is-active"><a class="tocitem" href>jump-dev/BARON.jl</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Use-with-JuMP"><span>Use with JuMP</span></a></li><li><a class="tocitem" href="#MathOptInterface-API"><span>MathOptInterface API</span></a></li></ul></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>jump-dev/BARON.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>jump-dev/BARON.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/BARON.jl/blob/v0.8.2/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="BARON.jl"><a class="docs-heading-anchor" href="#BARON.jl">BARON.jl</a><a id="BARON.jl-1"></a><a class="docs-heading-anchor-permalink" href="#BARON.jl" title="Permalink"></a></h1><p><a href="https://github.com/jump-dev/BARON.jl/actions?query=workflow%3ACI"><img src="https://github.com/jump-dev/BARON.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/jump-dev/BARON.jl"><img src="https://codecov.io/gh/jump-dev/BARON.jl/branch/master/graph/badge.svg" alt="codecov"/></a></p><p><a href="https://github.com/jump-dev/BARON.jl">BARON.jl</a> is a wrapper for <a href="http://minlp.com/baron">BARON by The Optimization Firm</a>.</p><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>This wrapper is maintained by the JuMP community and is not officially supported by The Optimization Firm.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>BARON.jl</code> is licensed under the <a href="https://github.com/jump-dev/BARON.jl/blob/master/LICENSE.md">MIT License</a>.</p><p>The underlying solver is a closed-source commercial product for which you must obtain a license from <a href="http://minlp.com">The Optimization Firm</a>, although a small trial version is available for free.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>First, download a copy of <a href="http://minlp.com/">the BARON solver</a> and unpack the executable in a location of your choosing.</p><p>Once installed, set the <code>BARON_EXEC</code> environment variable pointing to the BARON executable (full path, including file name as it differs across platforms), and run <code>Pkg.add(&quot;BARON&quot;)</code>. For example:</p><pre><code class="language-julia hljs">ENV[&quot;BARON_EXEC&quot;] = &quot;/path/to/baron.exe&quot;
 using Pkg
 Pkg.add(&quot;BARON&quot;)</code></pre><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><pre><code class="language-julia hljs">using JuMP, BARON
-model = Model(BARON.Optimizer)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The BARON optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarNonlinearFunction}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../AmplNLWriter/">« jump-dev/AmplNLWriter.jl</a><a class="docs-footer-nextpage" href="../CPLEX/">jump-dev/CPLEX.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+model = Model(BARON.Optimizer)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The BARON optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarNonlinearFunction}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction"><code>MOI.ScalarNonlinearFunction</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../AmplNLWriter/">« jump-dev/AmplNLWriter.jl</a><a class="docs-footer-nextpage" href="../CPLEX/">jump-dev/CPLEX.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/BilevelJuMP/index.html b/previews/PR3562/packages/BilevelJuMP/index.html
index a05768becfc..59f98611e98 100644
--- a/previews/PR3562/packages/BilevelJuMP/index.html
+++ b/previews/PR3562/packages/BilevelJuMP/index.html
@@ -34,4 +34,4 @@
 
 objective_value(model) # = 3 * (3.5 * 8/15) + 8/15 # = 6.13...
 value(x) # = 3.5 * 8/15 # = 1.86...
-value(y) # = 8/15 # = 0.53...</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../InfiniteOpt/">« infiniteopt/InfiniteOpt.jl</a><a class="docs-footer-nextpage" href="../SDDP/">odow/SDDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+value(y) # = 8/15 # = 0.53...</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../InfiniteOpt/">« infiniteopt/InfiniteOpt.jl</a><a class="docs-footer-nextpage" href="../SDDP/">odow/SDDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/CDCS/index.html b/previews/PR3562/packages/CDCS/index.html
index 1b68cce253b..ef97508dff0 100644
--- a/previews/PR3562/packages/CDCS/index.html
+++ b/previews/PR3562/packages/CDCS/index.html
@@ -27,4 +27,4 @@
            mat&quot;cdcsInstall&quot;
        end
 
-julia&gt; mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../OSQP/">« osqp/OSQP.jl</a><a class="docs-footer-nextpage" href="../COSMO/">oxfordcontrol/COSMO.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+julia&gt; mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../OSQP/">« osqp/OSQP.jl</a><a class="docs-footer-nextpage" href="../COSMO/">oxfordcontrol/COSMO.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/CDDLib/index.html b/previews/PR3562/packages/CDDLib/index.html
index 12fa80c6ac0..75b01c93ee2 100644
--- a/previews/PR3562/packages/CDDLib/index.html
+++ b/previews/PR3562/packages/CDDLib/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li class="is-active"><a class="tocitem" href>JuliaPolyhedra/CDDLib.jl</a><ul class="internal"><li><a class="tocitem" href="#Problem-description"><span>Problem description</span></a></li><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Use-with-JuMP"><span>Use with JuMP</span></a></li><li><a class="tocitem" href="#Debugging"><span>Debugging</span></a></li></ul></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>JuliaPolyhedra/CDDLib.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>JuliaPolyhedra/CDDLib.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/JuliaPolyhedra/CDDLib.jl/blob/9700752f849845bc8d8477928baf3df920daadd6/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="CDDLib"><a class="docs-heading-anchor" href="#CDDLib">CDDLib</a><a id="CDDLib-1"></a><a class="docs-heading-anchor-permalink" href="#CDDLib" title="Permalink"></a></h1><p><a href="https://github.com/JuliaPolyhedra/CDDLib.jl/actions?query=workflow%3ACI"><img src="https://github.com/JuliaPolyhedra/CDDLib.jl/workflows/CI/badge.svg?branch=master" alt/></a> <a href="http://codecov.io/github/JuliaPolyhedra/CDDLib.jl?branch=master"><img src="http://codecov.io/github/JuliaPolyhedra/CDDLib.jl/coverage.svg?branch=master" alt/></a> <a href="https://doi.org/10.5281/zenodo.1214581"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.1214581.svg" alt/></a></p><p><a href="https://github.com/JuliaPolyhedra/CDDLib.jl">CDDLib.jl</a> is a wrapper for <a href="https://www.inf.ethz.ch/personal/fukudak/cdd_home/">cddlib</a>.</p><p>CDDLib.jl can be used with C API of cddlib, the higher level interface of <a href="https://github.com/JuliaPolyhedra/Polyhedra.jl">Polyhedra.jl</a>, or as a linear programming solver with <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a> or <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a>.</p><h2 id="Problem-description"><a class="docs-heading-anchor" href="#Problem-description">Problem description</a><a id="Problem-description-1"></a><a class="docs-heading-anchor-permalink" href="#Problem-description" title="Permalink"></a></h2><p>As written in the <a href="https://github.com/cddlib/cddlib">README of cddlib</a>:</p><blockquote><p>The C-library  cddlib is a C implementation of the Double Description Method of Motzkin et al. for generating all vertices (that is, extreme points) and extreme rays of a general convex polyhedron in R^d given by a system of linear inequalities:</p><pre><code class="nohighlight hljs">P = { x=(x1, ..., xd)^T :  b - A  x  &gt;= 0 }</code></pre><p>where  A  is a given m x d real matrix, b is a given m-vector and 0 is the m-vector of all zeros.</p><p>The program can be used for the reverse operation (that is, convex hull computation).  This means that  one can move back and forth between an inequality representation  and a generator (that is, vertex and ray) representation of a polyhedron with cdd.  Also, cdd can solve a linear programming problem, that is, a problem of maximizing and minimizing a linear function over P.</p></blockquote><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>CDDLib.jl is licensed under the <a href="https://github.com/JuliaPolyhedra/CDDLib.jl/blob/master/LICENSE.md">GPL v2 license</a>.</p><p>The underlying solver, <a href="https://github.com/cddlib/cddlib">cddlib/cddlib</a> is also licensed under the <a href="https://github.com/cddlib/cddlib/blob/master/COPYING">GPL v2 license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install CDDLib.jl using the Julia package manager:</p><pre><code class="language-julia hljs">import Pkg
 Pkg.add(&quot;CDDLib&quot;)</code></pre><p>Building the package will download binaries of <a href="https://github.com/cddlib/cddlib">cddlib</a> that are provided by <a href="https://github.com/JuliaBinaryWrappers/cddlib_jll.jl">cddlib_jll.jl</a>.</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>Use <code>CDDLib.Optimizer{Float64}</code> to use CDDLib.jl with <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a>:</p><pre><code class="language-julia hljs">using JuMP, CDDLib
 model = Model(CDDLib.Optimizer{Float64})</code></pre><p>When using CDDLib.jl with <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a>, you can pass a different number type:</p><pre><code class="language-julia hljs">using MathOptInterface, CDDLib
-model = CDDLib.Optimizer{Rational{BigInt}}()</code></pre><h2 id="Debugging"><a class="docs-heading-anchor" href="#Debugging">Debugging</a><a id="Debugging-1"></a><a class="docs-heading-anchor-permalink" href="#Debugging" title="Permalink"></a></h2><p>CDDLib.jl uses two global Boolean variables to enable debugging outputs: <code>debug</code> and <code>log</code>.</p><p>You can query the value of <code>debug</code> and <code>log</code> with <code>get_debug</code> and <code>get_log</code>, and set their values with <code>set_debug</code> and <code>set_log</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../NLopt/">« JuliaOpt/NLopt.jl</a><a class="docs-footer-nextpage" href="../MadNLP/">MadNLP/MadNLP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+model = CDDLib.Optimizer{Rational{BigInt}}()</code></pre><h2 id="Debugging"><a class="docs-heading-anchor" href="#Debugging">Debugging</a><a id="Debugging-1"></a><a class="docs-heading-anchor-permalink" href="#Debugging" title="Permalink"></a></h2><p>CDDLib.jl uses two global Boolean variables to enable debugging outputs: <code>debug</code> and <code>log</code>.</p><p>You can query the value of <code>debug</code> and <code>log</code> with <code>get_debug</code> and <code>get_log</code>, and set their values with <code>set_debug</code> and <code>set_log</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../NLopt/">« JuliaOpt/NLopt.jl</a><a class="docs-footer-nextpage" href="../MadNLP/">MadNLP/MadNLP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/COPT/index.html b/previews/PR3562/packages/COPT/index.html
index f07142966f5..b9c5109e2bf 100644
--- a/previews/PR3562/packages/COPT/index.html
+++ b/previews/PR3562/packages/COPT/index.html
@@ -39,4 +39,4 @@
 @show value.(X)
 @show value.(z)
 @show shadow_price(c1)
-@show shadow_price(c2)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Xpress/">« jump-dev/Xpress.jl</a><a class="docs-footer-nextpage" href="../GAMS/">GAMS-dev/GAMS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@show shadow_price(c2)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Xpress/">« jump-dev/Xpress.jl</a><a class="docs-footer-nextpage" href="../GAMS/">GAMS-dev/GAMS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/COSMO/index.html b/previews/PR3562/packages/COSMO/index.html
index 51d48fbfb0d..45c0c4b2cba 100644
--- a/previews/PR3562/packages/COSMO/index.html
+++ b/previews/PR3562/packages/COSMO/index.html
@@ -34,4 +34,4 @@
   publisher = {Springer},
   doi     = {10.1007/s10957-021-01896-x},
   url     = {https://doi.org/10.1007/s10957-021-01896-x}
-}</code></pre><p>The article is available under Open Access <a href="https://link.springer.com/article/10.1007/s10957-021-01896-x">here</a>.</p><h2 id="Contributing"><a class="docs-heading-anchor" href="#Contributing">Contributing</a><a id="Contributing-1"></a><a class="docs-heading-anchor-permalink" href="#Contributing" title="Permalink"></a></h2><ul><li>Contributions are always welcome. Our style guide can be found <a href="https://github.com/oxfordcontrol/COSMO.jl/wiki/Code-Style-Guide">here</a>.</li><li>Current issues, tasks and future ideas are listed in <a href="https://github.com/oxfordcontrol/ossdp/issues">Issues</a>. Please report any issues or bugs that you encounter.</li><li>As an open source project we are also interested in any projects and applications that use COSMO. Please let us know by opening a GitHub issue.</li></ul><h2 id="Python-Interface"><a class="docs-heading-anchor" href="#Python-Interface">Python - Interface</a><a id="Python-Interface-1"></a><a class="docs-heading-anchor-permalink" href="#Python-Interface" title="Permalink"></a></h2><p>COSMO can also be called from Python. Take a look at: <a href="https://github.com/oxfordcontrol/cosmo-python">cosmo-python</a></p><h2 id="Licence"><a class="docs-heading-anchor" href="#Licence">Licence 🔍</a><a id="Licence-1"></a><a class="docs-heading-anchor-permalink" href="#Licence" title="Permalink"></a></h2><p>This project is licensed under the Apache License - see the <a href="https://github.com/oxfordcontrol/COSMO.jl/blob/master/LICENSE.md">LICENSE.md</a> file for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CDCS/">« oxfordcontrol/CDCS.jl</a><a class="docs-footer-nextpage" href="../Clarabel/">oxfordcontrol/Clarabel.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><p>The article is available under Open Access <a href="https://link.springer.com/article/10.1007/s10957-021-01896-x">here</a>.</p><h2 id="Contributing"><a class="docs-heading-anchor" href="#Contributing">Contributing</a><a id="Contributing-1"></a><a class="docs-heading-anchor-permalink" href="#Contributing" title="Permalink"></a></h2><ul><li>Contributions are always welcome. Our style guide can be found <a href="https://github.com/oxfordcontrol/COSMO.jl/wiki/Code-Style-Guide">here</a>.</li><li>Current issues, tasks and future ideas are listed in <a href="https://github.com/oxfordcontrol/ossdp/issues">Issues</a>. Please report any issues or bugs that you encounter.</li><li>As an open source project we are also interested in any projects and applications that use COSMO. Please let us know by opening a GitHub issue.</li></ul><h2 id="Python-Interface"><a class="docs-heading-anchor" href="#Python-Interface">Python - Interface</a><a id="Python-Interface-1"></a><a class="docs-heading-anchor-permalink" href="#Python-Interface" title="Permalink"></a></h2><p>COSMO can also be called from Python. Take a look at: <a href="https://github.com/oxfordcontrol/cosmo-python">cosmo-python</a></p><h2 id="Licence"><a class="docs-heading-anchor" href="#Licence">Licence 🔍</a><a id="Licence-1"></a><a class="docs-heading-anchor-permalink" href="#Licence" title="Permalink"></a></h2><p>This project is licensed under the Apache License - see the <a href="https://github.com/oxfordcontrol/COSMO.jl/blob/master/LICENSE.md">LICENSE.md</a> file for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CDCS/">« oxfordcontrol/CDCS.jl</a><a class="docs-footer-nextpage" href="../Clarabel/">oxfordcontrol/Clarabel.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/CPLEX/index.html b/previews/PR3562/packages/CPLEX/index.html
index 58e1cd15475..218844399f6 100644
--- a/previews/PR3562/packages/CPLEX/index.html
+++ b/previews/PR3562/packages/CPLEX/index.html
@@ -161,4 +161,4 @@
     x_optimal = value.(x)
     y_optimal = value.(y)
     println(&quot;x: $(x_optimal), y: $(y_optimal)&quot;)
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../BARON/">« jump-dev/BARON.jl</a><a class="docs-footer-nextpage" href="../CSDP/">jump-dev/CSDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../BARON/">« jump-dev/BARON.jl</a><a class="docs-footer-nextpage" href="../CSDP/">jump-dev/CSDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/CSDP/index.html b/previews/PR3562/packages/CSDP/index.html
index d01539bf1fc..a39eb76b296 100644
--- a/previews/PR3562/packages/CSDP/index.html
+++ b/previews/PR3562/packages/CSDP/index.html
@@ -10,4 +10,4 @@
       A(X) = a
          X ⪰ 0</code></pre><p>where <code>A(X) = [⟨A_1, X⟩, ..., ⟨A_m, X⟩]</code>. The corresponding dual is:</p><pre><code class="nohighlight hljs">min ⟨a, y⟩
      A&#39;(y) - C = Z
-             Z ⪰ 0</code></pre><p>where <code>A&#39;(y) = y_1A_1 + ... + y_mA_m</code></p><h2 id="Termination-criteria"><a class="docs-heading-anchor" href="#Termination-criteria">Termination criteria</a><a id="Termination-criteria-1"></a><a class="docs-heading-anchor-permalink" href="#Termination-criteria" title="Permalink"></a></h2><p>CSDP will terminate successfully (or partially) in the following cases:</p><ul><li>If CSDP finds <code>X, Z ⪰ 0</code> such that the following 3 tolerances are satisfied:<ul><li>primal feasibility tolerance: <code>||A(x) - a||_2 / (1 + ||a||_2) &lt; axtol</code></li><li>dual feasibility tolerance: <code>||A&#39;(y) - C - Z||_F / (1 + ||C||_F) &lt; atytol</code></li><li>relative duality gap tolerance: <code>gap / (1 + |⟨a, y⟩| + |⟨C, X⟩|) &lt; objtol</code><ul><li>objective duality gap: if <code>usexygap</code> is <code>0</code>, <code>gap = ⟨a, y⟩ - ⟨C, X⟩</code></li><li>XY duality gap: if <code>usexygap</code> is <code>1</code>, <code>gap = ⟨Z, X⟩</code></li></ul></li></ul>then it returns <code>0</code>.</li><li>If CSDP finds <code>y</code> and <code>Z ⪰ 0</code> such that <code>-⟨a, y⟩ / ||A&#39;(y) - Z||_F &gt; pinftol</code>, it returns <code>1</code> with <code>y</code> such that <code>⟨a, y⟩ = -1</code>.</li><li>If CSDP finds <code>X ⪰ 0</code> such that <code>⟨C, X⟩ / ||A(X)||_2 &gt; dinftol</code>, it returns <code>2</code> with <code>X</code> such that <code>⟨C, X⟩ = 1</code>.</li><li>If CSDP finds <code>X, Z ⪰ 0</code> such that the following 3 tolerances are satisfied with <code>1000*axtol</code>, <code>1000*atytol</code> and <code>1000*objtol</code> but at least one of them is not satisfied with <code>axtol</code>, <code>atytol</code> and <code>objtol</code> and cannot make progress, then it returns <code>3</code>.</li></ul><p>In addition, if the <code>printlevel</code> option is at least <code>1</code>, the following will be printed:</p><ul><li>If the return code is <code>1</code>, CSDP will print <code>⟨a, y⟩</code> and <code>||A&#39;(y) - Z||_F</code></li><li>If the return code is <code>2</code>, CSDP will print <code>⟨C, X⟩</code> and <code>||A(X)||_F</code></li><li>Otherwise, CSDP will print<ul><li>the primal/dual objective value,</li><li>the relative primal/dual infeasibility,</li><li>the objective duality gap <code>⟨a, y⟩ - ⟨C, X⟩</code> and objective relative duality gap <code>(⟨a, y⟩ - ⟨C, X⟩) / (1 + |⟨a, y⟩| + |⟨C, X⟩|)</code>,</li><li>the XY duality gap <code>⟨Z, X⟩</code> and XY relative duality gap <code>⟨Z, X⟩ / (1 + |⟨a, y⟩| + |⟨C, X⟩|)</code></li><li>and the DIMACS error measures.</li></ul></li></ul><p>In theory, for feasible primal and dual solutions, <code>⟨a, y⟩ - ⟨C, X⟩ = ⟨Z, X⟩</code>, so the objective and XY duality gap should be equivalent. However, in practice, there are sometimes solution which satisfy primal and dual feasibility tolerances but have objective duality gap which are not close to XY duality gap. In some cases, the objective duality gap may even become negative (hence the <code>tweakgap</code> option). This is the reason <code>usexygap</code> is <code>1</code> by default.</p><p>CSDP considers that <code>X ⪰ 0</code> (resp. <code>Z ⪰ 0</code>) is satisfied when the Cholesky factorizations can be computed. In practice, this is somewhat more conservative than simply requiring all eigenvalues to be nonnegative.</p><h2 id="Status"><a class="docs-heading-anchor" href="#Status">Status</a><a id="Status-1"></a><a class="docs-heading-anchor-permalink" href="#Status" title="Permalink"></a></h2><p>The table below shows how the different CSDP statuses are converted to the MathOptInterface statuses.</p><table><tr><th style="text-align: right">CSDP code</th><th style="text-align: right">State</th><th style="text-align: right">Description</th><th style="text-align: right">MOI status</th></tr><tr><td style="text-align: right"><code>0</code></td><td style="text-align: right">Success</td><td style="text-align: right">SDP solved</td><td style="text-align: right"><code>MOI.OPTIMAL</code></td></tr><tr><td style="text-align: right"><code>1</code></td><td style="text-align: right">Success</td><td style="text-align: right">The problem is primal infeasible, and we have a certificate</td><td style="text-align: right"><code>MOI.INFEASIBLE</code></td></tr><tr><td style="text-align: right"><code>2</code></td><td style="text-align: right">Success</td><td style="text-align: right">The problem is dual infeasible, and we have a certificate</td><td style="text-align: right"><code>MOI.DUAL_INFEASIBLE</code></td></tr><tr><td style="text-align: right"><code>3</code></td><td style="text-align: right">Partial Success</td><td style="text-align: right">A solution has been found, but full accuracy was not achieved</td><td style="text-align: right"><code>MOI.ALMOST_OPTIMAL</code></td></tr><tr><td style="text-align: right"><code>4</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Maximum iterations reached</td><td style="text-align: right"><code>MOI.ITERATION_LIMIT</code></td></tr><tr><td style="text-align: right"><code>5</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Stuck at edge of primal feasibility</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>6</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Stuck at edge of dual infeasibility</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>7</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Lack of progress</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>8</code></td><td style="text-align: right">Failure</td><td style="text-align: right"><code>X</code>, <code>Z</code>, or <code>O</code> was singular</td><td style="text-align: right"><code>MOI.NUMERICAL_ERROR</code></td></tr><tr><td style="text-align: right"><code>9</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Detected <code>NaN</code> or <code>Inf</code> values</td><td style="text-align: right"><code>MOI.NUMERICAL_ERROR</code></td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CPLEX/">« jump-dev/CPLEX.jl</a><a class="docs-footer-nextpage" href="../Cbc/">jump-dev/Cbc.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+             Z ⪰ 0</code></pre><p>where <code>A&#39;(y) = y_1A_1 + ... + y_mA_m</code></p><h2 id="Termination-criteria"><a class="docs-heading-anchor" href="#Termination-criteria">Termination criteria</a><a id="Termination-criteria-1"></a><a class="docs-heading-anchor-permalink" href="#Termination-criteria" title="Permalink"></a></h2><p>CSDP will terminate successfully (or partially) in the following cases:</p><ul><li>If CSDP finds <code>X, Z ⪰ 0</code> such that the following 3 tolerances are satisfied:<ul><li>primal feasibility tolerance: <code>||A(x) - a||_2 / (1 + ||a||_2) &lt; axtol</code></li><li>dual feasibility tolerance: <code>||A&#39;(y) - C - Z||_F / (1 + ||C||_F) &lt; atytol</code></li><li>relative duality gap tolerance: <code>gap / (1 + |⟨a, y⟩| + |⟨C, X⟩|) &lt; objtol</code><ul><li>objective duality gap: if <code>usexygap</code> is <code>0</code>, <code>gap = ⟨a, y⟩ - ⟨C, X⟩</code></li><li>XY duality gap: if <code>usexygap</code> is <code>1</code>, <code>gap = ⟨Z, X⟩</code></li></ul></li></ul>then it returns <code>0</code>.</li><li>If CSDP finds <code>y</code> and <code>Z ⪰ 0</code> such that <code>-⟨a, y⟩ / ||A&#39;(y) - Z||_F &gt; pinftol</code>, it returns <code>1</code> with <code>y</code> such that <code>⟨a, y⟩ = -1</code>.</li><li>If CSDP finds <code>X ⪰ 0</code> such that <code>⟨C, X⟩ / ||A(X)||_2 &gt; dinftol</code>, it returns <code>2</code> with <code>X</code> such that <code>⟨C, X⟩ = 1</code>.</li><li>If CSDP finds <code>X, Z ⪰ 0</code> such that the following 3 tolerances are satisfied with <code>1000*axtol</code>, <code>1000*atytol</code> and <code>1000*objtol</code> but at least one of them is not satisfied with <code>axtol</code>, <code>atytol</code> and <code>objtol</code> and cannot make progress, then it returns <code>3</code>.</li></ul><p>In addition, if the <code>printlevel</code> option is at least <code>1</code>, the following will be printed:</p><ul><li>If the return code is <code>1</code>, CSDP will print <code>⟨a, y⟩</code> and <code>||A&#39;(y) - Z||_F</code></li><li>If the return code is <code>2</code>, CSDP will print <code>⟨C, X⟩</code> and <code>||A(X)||_F</code></li><li>Otherwise, CSDP will print<ul><li>the primal/dual objective value,</li><li>the relative primal/dual infeasibility,</li><li>the objective duality gap <code>⟨a, y⟩ - ⟨C, X⟩</code> and objective relative duality gap <code>(⟨a, y⟩ - ⟨C, X⟩) / (1 + |⟨a, y⟩| + |⟨C, X⟩|)</code>,</li><li>the XY duality gap <code>⟨Z, X⟩</code> and XY relative duality gap <code>⟨Z, X⟩ / (1 + |⟨a, y⟩| + |⟨C, X⟩|)</code></li><li>and the DIMACS error measures.</li></ul></li></ul><p>In theory, for feasible primal and dual solutions, <code>⟨a, y⟩ - ⟨C, X⟩ = ⟨Z, X⟩</code>, so the objective and XY duality gap should be equivalent. However, in practice, there are sometimes solution which satisfy primal and dual feasibility tolerances but have objective duality gap which are not close to XY duality gap. In some cases, the objective duality gap may even become negative (hence the <code>tweakgap</code> option). This is the reason <code>usexygap</code> is <code>1</code> by default.</p><p>CSDP considers that <code>X ⪰ 0</code> (resp. <code>Z ⪰ 0</code>) is satisfied when the Cholesky factorizations can be computed. In practice, this is somewhat more conservative than simply requiring all eigenvalues to be nonnegative.</p><h2 id="Status"><a class="docs-heading-anchor" href="#Status">Status</a><a id="Status-1"></a><a class="docs-heading-anchor-permalink" href="#Status" title="Permalink"></a></h2><p>The table below shows how the different CSDP statuses are converted to the MathOptInterface statuses.</p><table><tr><th style="text-align: right">CSDP code</th><th style="text-align: right">State</th><th style="text-align: right">Description</th><th style="text-align: right">MOI status</th></tr><tr><td style="text-align: right"><code>0</code></td><td style="text-align: right">Success</td><td style="text-align: right">SDP solved</td><td style="text-align: right"><code>MOI.OPTIMAL</code></td></tr><tr><td style="text-align: right"><code>1</code></td><td style="text-align: right">Success</td><td style="text-align: right">The problem is primal infeasible, and we have a certificate</td><td style="text-align: right"><code>MOI.INFEASIBLE</code></td></tr><tr><td style="text-align: right"><code>2</code></td><td style="text-align: right">Success</td><td style="text-align: right">The problem is dual infeasible, and we have a certificate</td><td style="text-align: right"><code>MOI.DUAL_INFEASIBLE</code></td></tr><tr><td style="text-align: right"><code>3</code></td><td style="text-align: right">Partial Success</td><td style="text-align: right">A solution has been found, but full accuracy was not achieved</td><td style="text-align: right"><code>MOI.ALMOST_OPTIMAL</code></td></tr><tr><td style="text-align: right"><code>4</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Maximum iterations reached</td><td style="text-align: right"><code>MOI.ITERATION_LIMIT</code></td></tr><tr><td style="text-align: right"><code>5</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Stuck at edge of primal feasibility</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>6</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Stuck at edge of dual infeasibility</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>7</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Lack of progress</td><td style="text-align: right"><code>MOI.SLOW_PROGRESS</code></td></tr><tr><td style="text-align: right"><code>8</code></td><td style="text-align: right">Failure</td><td style="text-align: right"><code>X</code>, <code>Z</code>, or <code>O</code> was singular</td><td style="text-align: right"><code>MOI.NUMERICAL_ERROR</code></td></tr><tr><td style="text-align: right"><code>9</code></td><td style="text-align: right">Failure</td><td style="text-align: right">Detected <code>NaN</code> or <code>Inf</code> values</td><td style="text-align: right"><code>MOI.NUMERICAL_ERROR</code></td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CPLEX/">« jump-dev/CPLEX.jl</a><a class="docs-footer-nextpage" href="../Cbc/">jump-dev/Cbc.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Cbc/index.html b/previews/PR3562/packages/Cbc/index.html
index 78372860833..80d67651a38 100644
--- a/previews/PR3562/packages/Cbc/index.html
+++ b/previews/PR3562/packages/Cbc/index.html
@@ -9,4 +9,4 @@
 set_attribute(model, &quot;logLevel&quot;, 1)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The COIN Branch-and-Cut (Cbc) optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS1"><code>MOI.SOS1{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS2"><code>MOI.SOS2{Float64}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><code>Cbc.Status</code></li><li><code>Cbc.SecondaryStatus</code></li><li><a href="../../moi/reference/models/#MathOptInterface.DualStatus"><code>MOI.DualStatus</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.NodeCount"><code>MOI.NodeCount</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.NumberOfVariables"><code>MOI.NumberOfVariables</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveBound"><code>MOI.ObjectiveBound</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveValue"><code>MOI.ObjectiveValue</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.PrimalStatus"><code>MOI.PrimalStatus</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.RelativeGap"><code>MOI.RelativeGap</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ResultCount"><code>MOI.ResultCount</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.SolveTimeSec"><code>MOI.SolveTimeSec</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.TerminationStatus"><code>MOI.TerminationStatus</code></a></li></ul><p>List of supported optimizer attributes:</p><ul><li><code>Cbc.SetVariableNames</code></li><li><a href="../../moi/reference/models/#MathOptInterface.AbsoluteGapTolerance"><code>MOI.AbsoluteGapTolerance</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.NumberOfThreads"><code>MOI.NumberOfThreads</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.RawOptimizerAttribute"><code>MOI.RawOptimizerAttribute</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.RelativeGapTolerance"><code>MOI.RelativeGapTolerance</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.Silent"><code>MOI.Silent</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.SolverName"><code>MOI.SolverName</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.SolverVersion"><code>MOI.SolverVersion</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.TimeLimitSec"><code>MOI.TimeLimitSec</code></a></li></ul><p>List of supported variable attributes:</p><ul><li><a href="../../moi/reference/variables/#MathOptInterface.VariablePrimal"><code>MOI.VariablePrimal</code></a></li><li><a href="../../moi/reference/variables/#MathOptInterface.VariablePrimalStart"><code>MOI.VariablePrimalStart</code></a></li><li><a href="../../moi/reference/variables/#MathOptInterface.VariableName"><code>MOI.VariableName</code></a></li></ul><p>List of supported constraint attributes:</p><ul><li><a href="../../moi/reference/constraints/#MathOptInterface.ConstraintPrimal"><code>MOI.ConstraintPrimal</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>Options are, unfortunately, not well documented.</p><p>The following options are likely to be the most useful:</p><table><tr><th style="text-align: right">Parameter</th><th style="text-align: right">Example</th><th style="text-align: right">Explanation</th></tr><tr><td style="text-align: right"><code>seconds</code></td><td style="text-align: right"><code>60.0</code></td><td style="text-align: right">Solution timeout limit</td></tr><tr><td style="text-align: right"><code>logLevel</code></td><td style="text-align: right"><code>2</code></td><td style="text-align: right">Set to 0 to disable solution output</td></tr><tr><td style="text-align: right"><code>maxSolutions</code></td><td style="text-align: right"><code>1</code></td><td style="text-align: right">Terminate after this many feasible solutions have been found</td></tr><tr><td style="text-align: right"><code>maxNodes</code></td><td style="text-align: right"><code>1</code></td><td style="text-align: right">Terminate after this many branch-and-bound nodes have been evaluated</td></tr><tr><td style="text-align: right"><code>allowableGap</code></td><td style="text-align: right"><code>0.05</code></td><td style="text-align: right">Terminate after optimality gap is less than this value (on an absolute scale)</td></tr><tr><td style="text-align: right"><code>ratioGap</code></td><td style="text-align: right"><code>0.05</code></td><td style="text-align: right">Terminate after optimality gap is smaller than this relative fraction</td></tr><tr><td style="text-align: right"><code>threads</code></td><td style="text-align: right"><code>1</code></td><td style="text-align: right">Set the number of threads to use for parallel branch &amp; bound</td></tr></table><p>The complete list of parameters can be found by running the <code>cbc</code> executable and typing <code>?</code> at the prompt.</p><p>Start the <code>cbc</code> executable from Julia as follows:</p><pre><code class="language-julia hljs">using Cbc_jll
 Cbc_jll.cbc() do exe
     run(`$(exe)`)
-end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CSDP/">« jump-dev/CSDP.jl</a><a class="docs-footer-nextpage" href="../Clp/">jump-dev/Clp.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CSDP/">« jump-dev/CSDP.jl</a><a class="docs-footer-nextpage" href="../Clp/">jump-dev/Clp.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Clarabel/index.html b/previews/PR3562/packages/Clarabel/index.html
index 546be828654..dc7fef3a7ac 100644
--- a/previews/PR3562/packages/Clarabel/index.html
+++ b/previews/PR3562/packages/Clarabel/index.html
@@ -23,4 +23,4 @@
 \text{minimize} &amp; \frac{1}{2}x^T P x + q^T x\\\\[2ex]
 \text{subject to} &amp; Ax + s = b \\\\[1ex]
         &amp; s \in \mathcal{K}
-\end{array}\]</p><p>with decision variables <span>$x \in \mathbb{R}^n$</span>, <span>$s \in \mathbb{R}^m$</span> and data matrices <span>$P=P^\top \succeq 0$</span>, <span>$q \in \mathbb{R}^n$</span>, <span>$A \in \mathbb{R}^{m \times n}$</span>, and <span>$b \in \mathbb{R}^m$</span>. The convex set <span>$\mathcal{K}$</span> is a composition of convex cones.</p><p><strong>For more information see the Clarabel Documentation (<a href="https://oxfordcontrol.github.io/ClarabelDocs/stable">stable</a> |  <a href="https://oxfordcontrol.github.io/ClarabelDocs/dev">dev</a>).</strong></p><p>Clarabel is also available in a Rust implementation with additional language interfaces.  See <a href="https://github.com/oxfordcontrol/Clarabel.rs">here</a>.</p><h2 id="Features"><a class="docs-heading-anchor" href="#Features">Features</a><a id="Features-1"></a><a class="docs-heading-anchor-permalink" href="#Features" title="Permalink"></a></h2><ul><li><strong>Versatile</strong>: Clarabel.jl solves linear programs (LPs), quadratic programs (QPs), second-order cone programs (SOCPs) and semidefinite programs (SDPs). It also solves problems with exponential, power cone and generalized power cone constraints.</li><li><strong>Quadratic objectives</strong>: Unlike interior point solvers based on the standard homogeneous self-dual embedding (HSDE), Clarabel.jl handles quadratic objectives without requiring any epigraphical reformulation of the objective.   It can therefore be significantly faster than other HSDE-based solvers for problems with quadratic objective functions.</li><li><strong>Infeasibility detection</strong>: Infeasible problems are detected using a homogeneous embedding technique.</li><li><strong>JuMP / Convex.jl support</strong>: We provide an interface to <a href="https://jump.dev/JuMP.jl/stable/moi/">MathOptInterface</a> (MOI), which allows you to describe your problem in <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a> and <a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a>.</li><li><strong>Arbitrary precision types</strong>: You can solve problems with any floating point precision, for example, Float32 or Julia&#39;s BigFloat type, using either the native interface, or via MathOptInterface / Convex.jl.</li><li><strong>Open Source</strong>: Our code is available on <a href="https://github.com/oxfordcontrol/Clarabel.jl">GitHub</a> and distributed under the Apache 2.0 License</li></ul><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><ul><li><strong>Clarabel.jl</strong> can be added via the Julia package manager (type <code>]</code>): <code>pkg&gt; add Clarabel</code></li></ul><h2 id="License"><a class="docs-heading-anchor" href="#License">License 🔍</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>This project is licensed under the Apache License  2.0 - see the <a href="https://github.com/oxfordcontrol/Clarabel.jl/blob/main/LICENSE.md">LICENSE.md</a> file for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../COSMO/">« oxfordcontrol/COSMO.jl</a><a class="docs-footer-nextpage" href="../SCIP/">scipopt/SCIP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{array}\]</p><p>with decision variables <span>$x \in \mathbb{R}^n$</span>, <span>$s \in \mathbb{R}^m$</span> and data matrices <span>$P=P^\top \succeq 0$</span>, <span>$q \in \mathbb{R}^n$</span>, <span>$A \in \mathbb{R}^{m \times n}$</span>, and <span>$b \in \mathbb{R}^m$</span>. The convex set <span>$\mathcal{K}$</span> is a composition of convex cones.</p><p><strong>For more information see the Clarabel Documentation (<a href="https://oxfordcontrol.github.io/ClarabelDocs/stable">stable</a> |  <a href="https://oxfordcontrol.github.io/ClarabelDocs/dev">dev</a>).</strong></p><p>Clarabel is also available in a Rust implementation with additional language interfaces.  See <a href="https://github.com/oxfordcontrol/Clarabel.rs">here</a>.</p><h2 id="Features"><a class="docs-heading-anchor" href="#Features">Features</a><a id="Features-1"></a><a class="docs-heading-anchor-permalink" href="#Features" title="Permalink"></a></h2><ul><li><strong>Versatile</strong>: Clarabel.jl solves linear programs (LPs), quadratic programs (QPs), second-order cone programs (SOCPs) and semidefinite programs (SDPs). It also solves problems with exponential, power cone and generalized power cone constraints.</li><li><strong>Quadratic objectives</strong>: Unlike interior point solvers based on the standard homogeneous self-dual embedding (HSDE), Clarabel.jl handles quadratic objectives without requiring any epigraphical reformulation of the objective.   It can therefore be significantly faster than other HSDE-based solvers for problems with quadratic objective functions.</li><li><strong>Infeasibility detection</strong>: Infeasible problems are detected using a homogeneous embedding technique.</li><li><strong>JuMP / Convex.jl support</strong>: We provide an interface to <a href="https://jump.dev/JuMP.jl/stable/moi/">MathOptInterface</a> (MOI), which allows you to describe your problem in <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a> and <a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a>.</li><li><strong>Arbitrary precision types</strong>: You can solve problems with any floating point precision, for example, Float32 or Julia&#39;s BigFloat type, using either the native interface, or via MathOptInterface / Convex.jl.</li><li><strong>Open Source</strong>: Our code is available on <a href="https://github.com/oxfordcontrol/Clarabel.jl">GitHub</a> and distributed under the Apache 2.0 License</li></ul><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><ul><li><strong>Clarabel.jl</strong> can be added via the Julia package manager (type <code>]</code>): <code>pkg&gt; add Clarabel</code></li></ul><h2 id="License"><a class="docs-heading-anchor" href="#License">License 🔍</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>This project is licensed under the Apache License  2.0 - see the <a href="https://github.com/oxfordcontrol/Clarabel.jl/blob/main/LICENSE.md">LICENSE.md</a> file for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../COSMO/">« oxfordcontrol/COSMO.jl</a><a class="docs-footer-nextpage" href="../SCIP/">scipopt/SCIP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Clp/index.html b/previews/PR3562/packages/Clp/index.html
index 466b1a4c826..204c929daa7 100644
--- a/previews/PR3562/packages/Clp/index.html
+++ b/previews/PR3562/packages/Clp/index.html
@@ -7,4 +7,4 @@
 Pkg.add(&quot;Clp&quot;)</code></pre><p>In addition to installing the Clp.jl package, this will also download and install the Clp binaries. You do not need to install Clp separately.</p><p>To use a custom binary, read the <a href="https://jump.dev/JuMP.jl/stable/developers/custom_solver_binaries/">Custom solver binaries</a> section of the JuMP documentation.</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>To use Clp with JuMP, use <code>Clp.Optimizer</code>:</p><pre><code class="language-julia hljs">using JuMP, Clp
 model = Model(Clp.Optimizer)
 set_attribute(model, &quot;LogLevel&quot;, 1)
-set_attribute(model, &quot;Algorithm&quot;, 4)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The Clp optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>Options are, unfortunately, not well documented.</p><p>The following options are likely to be the most useful:</p><table><tr><th style="text-align: right">Parameter</th><th style="text-align: right">Example</th><th style="text-align: right">Explanation</th></tr><tr><td style="text-align: right"><code>PrimalTolerance</code></td><td style="text-align: right"><code>1e-7</code></td><td style="text-align: right">Primal feasibility tolerance</td></tr><tr><td style="text-align: right"><code>DualTolerance</code></td><td style="text-align: right"><code>1e-7</code></td><td style="text-align: right">Dual feasibility tolerance</td></tr><tr><td style="text-align: right"><code>DualObjectiveLimit</code></td><td style="text-align: right"><code>1e308</code></td><td style="text-align: right">When using dual simplex (where the objective is monotonically changing), terminate when the objective exceeds this limit</td></tr><tr><td style="text-align: right"><code>MaximumIterations</code></td><td style="text-align: right"><code>2147483647</code></td><td style="text-align: right">Terminate after performing this number of simplex iterations</td></tr><tr><td style="text-align: right"><code>MaximumSeconds</code></td><td style="text-align: right"><code>-1.0</code></td><td style="text-align: right">Terminate after this many seconds have passed. A negative value means no time limit</td></tr><tr><td style="text-align: right"><code>LogLevel</code></td><td style="text-align: right"><code>1</code></td><td style="text-align: right">Set to 1, 2, 3, or 4 for increasing output. Set to <code>0</code> to disable output</td></tr><tr><td style="text-align: right"><code>PresolveType</code></td><td style="text-align: right"><code>0</code></td><td style="text-align: right">Set to <code>1</code> to disable presolve</td></tr><tr><td style="text-align: right"><code>SolveType</code></td><td style="text-align: right"><code>5</code></td><td style="text-align: right">Solution method: dual simplex (<code>0</code>), primal simplex (<code>1</code>), sprint (<code>2</code>), barrier with crossover (<code>3</code>), barrier without crossover (<code>4</code>), automatic (<code>5</code>)</td></tr><tr><td style="text-align: right"><code>InfeasibleReturn</code></td><td style="text-align: right"><code>0</code></td><td style="text-align: right">Set to 1 to return as soon as the problem is found to be infeasible (by default, an infeasibility proof is computed as well)</td></tr><tr><td style="text-align: right"><code>Scaling</code></td><td style="text-align: right"><code>3</code></td><td style="text-align: right"><code>0</code> -off, <code>1</code> equilibrium, <code>2</code> geometric, <code>3</code> auto, <code>4</code> dynamic(later)</td></tr><tr><td style="text-align: right"><code>Perturbation</code></td><td style="text-align: right"><code>100</code></td><td style="text-align: right">switch on perturbation (<code>50</code>), automatic (<code>100</code>), don&#39;t try perturbing (<code>102</code>)</td></tr></table><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>Clp.Clp_XXX</code> functions, where the names and arguments are identical to the C API.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Cbc/">« jump-dev/Cbc.jl</a><a class="docs-footer-nextpage" href="../Dualization/">jump-dev/Dualization.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;Algorithm&quot;, 4)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The Clp optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>Options are, unfortunately, not well documented.</p><p>The following options are likely to be the most useful:</p><table><tr><th style="text-align: right">Parameter</th><th style="text-align: right">Example</th><th style="text-align: right">Explanation</th></tr><tr><td style="text-align: right"><code>PrimalTolerance</code></td><td style="text-align: right"><code>1e-7</code></td><td style="text-align: right">Primal feasibility tolerance</td></tr><tr><td style="text-align: right"><code>DualTolerance</code></td><td style="text-align: right"><code>1e-7</code></td><td style="text-align: right">Dual feasibility tolerance</td></tr><tr><td style="text-align: right"><code>DualObjectiveLimit</code></td><td style="text-align: right"><code>1e308</code></td><td style="text-align: right">When using dual simplex (where the objective is monotonically changing), terminate when the objective exceeds this limit</td></tr><tr><td style="text-align: right"><code>MaximumIterations</code></td><td style="text-align: right"><code>2147483647</code></td><td style="text-align: right">Terminate after performing this number of simplex iterations</td></tr><tr><td style="text-align: right"><code>MaximumSeconds</code></td><td style="text-align: right"><code>-1.0</code></td><td style="text-align: right">Terminate after this many seconds have passed. A negative value means no time limit</td></tr><tr><td style="text-align: right"><code>LogLevel</code></td><td style="text-align: right"><code>1</code></td><td style="text-align: right">Set to 1, 2, 3, or 4 for increasing output. Set to <code>0</code> to disable output</td></tr><tr><td style="text-align: right"><code>PresolveType</code></td><td style="text-align: right"><code>0</code></td><td style="text-align: right">Set to <code>1</code> to disable presolve</td></tr><tr><td style="text-align: right"><code>SolveType</code></td><td style="text-align: right"><code>5</code></td><td style="text-align: right">Solution method: dual simplex (<code>0</code>), primal simplex (<code>1</code>), sprint (<code>2</code>), barrier with crossover (<code>3</code>), barrier without crossover (<code>4</code>), automatic (<code>5</code>)</td></tr><tr><td style="text-align: right"><code>InfeasibleReturn</code></td><td style="text-align: right"><code>0</code></td><td style="text-align: right">Set to 1 to return as soon as the problem is found to be infeasible (by default, an infeasibility proof is computed as well)</td></tr><tr><td style="text-align: right"><code>Scaling</code></td><td style="text-align: right"><code>3</code></td><td style="text-align: right"><code>0</code> -off, <code>1</code> equilibrium, <code>2</code> geometric, <code>3</code> auto, <code>4</code> dynamic(later)</td></tr><tr><td style="text-align: right"><code>Perturbation</code></td><td style="text-align: right"><code>100</code></td><td style="text-align: right">switch on perturbation (<code>50</code>), automatic (<code>100</code>), don&#39;t try perturbing (<code>102</code>)</td></tr></table><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>Clp.Clp_XXX</code> functions, where the names and arguments are identical to the C API.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Cbc/">« jump-dev/Cbc.jl</a><a class="docs-footer-nextpage" href="../Dualization/">jump-dev/Dualization.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/DAQP/index.html b/previews/PR3562/packages/DAQP/index.html
index b12c947cc05..308d563ecfa 100644
--- a/previews/PR3562/packages/DAQP/index.html
+++ b/previews/PR3562/packages/DAQP/index.html
@@ -5,4 +5,4 @@
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li class="is-active"><a class="tocitem" href>darnstrom/DAQP.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Use-with-JuMP"><span>Use with JuMP</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li></ul></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>darnstrom/DAQP.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>darnstrom/DAQP.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/darnstrom/DAQP.jl/blob/v0.5.1/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="DAQP.jl"><a class="docs-heading-anchor" href="#DAQP.jl">DAQP.jl</a><a id="DAQP.jl-1"></a><a class="docs-heading-anchor-permalink" href="#DAQP.jl" title="Permalink"></a></h1><p><a href="https://github.com/darnstrom/DAQP.jl/actions"><img src="https://github.com/darnstrom/DAQP.jl/workflows/CI/badge.svg" alt/></a> <a href="http://codecov.io/github/darnstrom/DAQP.jl"><img src="http://codecov.io/gh/darnstrom/DAQP.jl/graphs/badge.svg" alt/></a></p><p><a href="https://github.com/darnstrom/DAQP.jl">DAQP.jl</a> is a Julia wrapper for the Quadratic Programming solver <a href="https://github.com/darnstrom/daqp">DAQP</a>.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>DAQP.jl is licensed under the <a href="https://github.com/darnstrom/DAQP.jl/blob/main/LICENSE">MIT license</a>.</p><p>The underlying solver, <a href="https://github.com/darnstrom/daqp">darnstrom/daqp</a> is licensed under the <a href="https://github.com/darnstrom/daqp/blob/master/LICENSE">MIT license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install DAQP.jl using the Julia package manager:</p><pre><code class="language-julia hljs">import Pkg
 Pkg.add(&quot;DAQP&quot;)</code></pre><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>To use DAQP with JuMP, do:</p><pre><code class="language-julia hljs">using JuMP, DAQP
-model = Model(DAQP.Optimizer)</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>General information about the solver is available at <a href="https://darnstrom.github.io/daqp/">https://darnstrom.github.io/daqp/</a>, and specifics for the Julia interface are available at <a href="https://darnstrom.github.io/daqp/start/julia">https://darnstrom.github.io/daqp/start/julia</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Hypatia/">« chriscoey/Hypatia.jl</a><a class="docs-footer-nextpage" href="../Tulip/">ds4dm/Tulip.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+model = Model(DAQP.Optimizer)</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>General information about the solver is available at <a href="https://darnstrom.github.io/daqp/">https://darnstrom.github.io/daqp/</a>, and specifics for the Julia interface are available at <a href="https://darnstrom.github.io/daqp/start/julia">https://darnstrom.github.io/daqp/start/julia</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Hypatia/">« chriscoey/Hypatia.jl</a><a class="docs-footer-nextpage" href="../Tulip/">ds4dm/Tulip.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/DiffOpt/index.html b/previews/PR3562/packages/DiffOpt/index.html
index ec20db201b5..5742bd06588 100644
--- a/previews/PR3562/packages/DiffOpt/index.html
+++ b/previews/PR3562/packages/DiffOpt/index.html
@@ -20,4 +20,4 @@
 # fetch the gradients
 grad_exp = MOI.get(model, DiffOpt.ReverseConstraintFunction(), cons)  # -3 x - 1
 constant(grad_exp)        # -1
-coefficient(grad_exp, x)  # -3</code></pre><h2 id="GSOC2020"><a class="docs-heading-anchor" href="#GSOC2020">GSOC2020</a><a id="GSOC2020-1"></a><a class="docs-heading-anchor-permalink" href="#GSOC2020" title="Permalink"></a></h2><p>DiffOpt began as a <a href="https://summerofcode.withgoogle.com/organizations/4727917315096576/?sp-page=2#5232064888045568">NumFOCUS sponsored Google Summer of Code (2020) project</a></p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../extensions/introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+coefficient(grad_exp, x)  # -3</code></pre><h2 id="GSOC2020"><a class="docs-heading-anchor" href="#GSOC2020">GSOC2020</a><a id="GSOC2020-1"></a><a class="docs-heading-anchor-permalink" href="#GSOC2020" title="Permalink"></a></h2><p>DiffOpt began as a <a href="https://summerofcode.withgoogle.com/organizations/4727917315096576/?sp-page=2#5232064888045568">NumFOCUS sponsored Google Summer of Code (2020) project</a></p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../extensions/introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Dualization/index.html b/previews/PR3562/packages/Dualization/index.html
index 085d4de9218..7d4e8cc65aa 100644
--- a/previews/PR3562/packages/Dualization/index.html
+++ b/previews/PR3562/packages/Dualization/index.html
@@ -10,4 +10,4 @@
 dual_model = dualize(model)</code></pre><p>To solve the dual formulation of a JuMP model, create a <code>dual_optimizer</code>:</p><pre><code class="language-julia hljs">using JuMP, Dualization, SCS
 model = Model(dual_optimizer(SCS.Optimizer))
 # ... build model ...
-optimize!(model)  # Solves the dual instead of the primal</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>The <a href="https://jump.dev/Dualization.jl/stable/">documentation for Dualization.jl</a> includes a detailed description of the dual reformulation, along with examples and an API reference.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Clp/">« jump-dev/Clp.jl</a><a class="docs-footer-nextpage" href="../ECOS/">jump-dev/ECOS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+optimize!(model)  # Solves the dual instead of the primal</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>The <a href="https://jump.dev/Dualization.jl/stable/">documentation for Dualization.jl</a> includes a detailed description of the dual reformulation, along with examples and an API reference.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Clp/">« jump-dev/Clp.jl</a><a class="docs-footer-nextpage" href="../ECOS/">jump-dev/ECOS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/ECOS/index.html b/previews/PR3562/packages/ECOS/index.html
index fe74e83a163..09cea847772 100644
--- a/previews/PR3562/packages/ECOS/index.html
+++ b/previews/PR3562/packages/ECOS/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li class="is-active"><a class="tocitem" href>jump-dev/ECOS.jl</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Use-with-JuMP"><span>Use with JuMP</span></a></li><li><a class="tocitem" href="#MathOptInterface-API"><span>MathOptInterface API</span></a></li><li><a class="tocitem" href="#Options"><span>Options</span></a></li></ul></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>jump-dev/ECOS.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>jump-dev/ECOS.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/ECOS.jl/blob/v1.1.1/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="ECOS.jl"><a class="docs-heading-anchor" href="#ECOS.jl">ECOS.jl</a><a id="ECOS.jl-1"></a><a class="docs-heading-anchor-permalink" href="#ECOS.jl" title="Permalink"></a></h1><p><a href="https://github.com/jump-dev/ECOS.jl/actions?query=workflow%3ACI"><img src="https://github.com/jump-dev/ECOS.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/jump-dev/ECOS.jl"><img src="https://codecov.io/gh/jump-dev/ECOS.jl/branch/master/graph/badge.svg" alt="codecov"/></a></p><p><a href="https://github.com/jump-dev/ECOS.jl">ECOS.jl</a> is a wrapper for the <a href="https://github.com/embotech/ecos">ECOS</a> solver.</p><p>The wrapper has two components:</p><ul><li>a thin wrapper around the complete C API</li><li>an interface to <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a></li></ul><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>This wrapper is maintained by the JuMP community and is not a product of Embotech AG.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>ECOS.jl</code> is licensed under the <a href="https://github.com/jump-dev/ECOS.jl/blob/master/LICENSE.md">MIT License</a>.</p><p>The underlying solver, <a href="https://github.com/embotech/ecos">embotech/ecos</a>, is licensed under the <a href="https://github.com/embotech/ecos/blob/develop/COPYING">GPL v3 license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install ECOS.jl using <code>Pkg.add</code>:</p><pre><code class="language-julia hljs">import Pkg
 Pkg.add(&quot;ECOS&quot;)</code></pre><p>In addition to installing the ECOS.jl package, this will also download and install the ECOS binaries. You do not need to install ECOS separately.</p><p>To use a custom binary, read the <a href="https://jump.dev/JuMP.jl/stable/developers/custom_solver_binaries/">Custom solver binaries</a> section of the JuMP documentation.</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>To use ECOS with <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a>, use <code>ECOS.Optimizer</code>:</p><pre><code class="language-julia hljs">using JuMP, ECOS
 model = Model(ECOS.Optimizer)
-set_attribute(model, &quot;maxit&quot;, 100)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The ECOS optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Nonnegatives"><code>MOI.Nonnegatives</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Zeros"><code>MOI.Zeros</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The following options are supported:</p><table><tr><th style="text-align: right">Parameter</th><th style="text-align: right">Explanation</th></tr><tr><td style="text-align: right"><code>gamma</code></td><td style="text-align: right">scaling the final step length</td></tr><tr><td style="text-align: right"><code>delta</code></td><td style="text-align: right">regularization parameter</td></tr><tr><td style="text-align: right"><code>eps</code></td><td style="text-align: right">regularization threshold</td></tr><tr><td style="text-align: right"><code>feastol</code></td><td style="text-align: right">primal/dual infeasibility tolerance</td></tr><tr><td style="text-align: right"><code>abstol</code></td><td style="text-align: right">absolute tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>reltol</code></td><td style="text-align: right">relative tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>feastol_inacc</code></td><td style="text-align: right">primal/dual infeasibility relaxed tolerance</td></tr><tr><td style="text-align: right"><code>abstol_inacc</code></td><td style="text-align: right">absolute relaxed tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>reltol_inacc</code></td><td style="text-align: right">relative relaxed tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>nitref</code></td><td style="text-align: right">number of iterative refinement steps</td></tr><tr><td style="text-align: right"><code>maxit</code></td><td style="text-align: right">maximum number of iterations</td></tr><tr><td style="text-align: right"><code>verbose</code></td><td style="text-align: right">verbosity <code>bool</code> for <code>PRINTLEVEL &lt; 3</code></td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Dualization/">« jump-dev/Dualization.jl</a><a class="docs-footer-nextpage" href="../GLPK/">jump-dev/GLPK.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;maxit&quot;, 100)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The ECOS optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Nonnegatives"><code>MOI.Nonnegatives</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Zeros"><code>MOI.Zeros</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The following options are supported:</p><table><tr><th style="text-align: right">Parameter</th><th style="text-align: right">Explanation</th></tr><tr><td style="text-align: right"><code>gamma</code></td><td style="text-align: right">scaling the final step length</td></tr><tr><td style="text-align: right"><code>delta</code></td><td style="text-align: right">regularization parameter</td></tr><tr><td style="text-align: right"><code>eps</code></td><td style="text-align: right">regularization threshold</td></tr><tr><td style="text-align: right"><code>feastol</code></td><td style="text-align: right">primal/dual infeasibility tolerance</td></tr><tr><td style="text-align: right"><code>abstol</code></td><td style="text-align: right">absolute tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>reltol</code></td><td style="text-align: right">relative tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>feastol_inacc</code></td><td style="text-align: right">primal/dual infeasibility relaxed tolerance</td></tr><tr><td style="text-align: right"><code>abstol_inacc</code></td><td style="text-align: right">absolute relaxed tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>reltol_inacc</code></td><td style="text-align: right">relative relaxed tolerance on duality gap</td></tr><tr><td style="text-align: right"><code>nitref</code></td><td style="text-align: right">number of iterative refinement steps</td></tr><tr><td style="text-align: right"><code>maxit</code></td><td style="text-align: right">maximum number of iterations</td></tr><tr><td style="text-align: right"><code>verbose</code></td><td style="text-align: right">verbosity <code>bool</code> for <code>PRINTLEVEL &lt; 3</code></td></tr></table></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Dualization/">« jump-dev/Dualization.jl</a><a class="docs-footer-nextpage" href="../GLPK/">jump-dev/GLPK.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/GAMS/index.html b/previews/PR3562/packages/GAMS/index.html
index dba298a8d7e..d1fdca2e1b0 100644
--- a/previews/PR3562/packages/GAMS/index.html
+++ b/previews/PR3562/packages/GAMS/index.html
@@ -22,4 +22,4 @@
 
 MOI.get(model, GAMS.GeneratedConstraintName(), c[2]) # returns eq2
 MOI.get(model, GAMS.OriginalConstraintName(&quot;eq1&quot;))   # returns c[1]
-MOI.get(model, GAMS.OriginalConstraintName(&quot;eq10&quot;))  # returns nothing</code></pre><p>Note that JuMP direct-mode is used.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../COPT/">« COPT-Public/COPT.jl</a><a class="docs-footer-nextpage" href="../NLopt/">JuliaOpt/NLopt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+MOI.get(model, GAMS.OriginalConstraintName(&quot;eq10&quot;))  # returns nothing</code></pre><p>Note that JuMP direct-mode is used.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../COPT/">« COPT-Public/COPT.jl</a><a class="docs-footer-nextpage" href="../NLopt/">JuliaOpt/NLopt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/GLPK/index.html b/previews/PR3562/packages/GLPK/index.html
index 722051b46c9..3ebce9b39a0 100644
--- a/previews/PR3562/packages/GLPK/index.html
+++ b/previews/PR3562/packages/GLPK/index.html
@@ -36,4 +36,4 @@
 @test primal_status(model) == MOI.FEASIBLE_POINT
 @test value(x) == 1
 @test value(y) == 2
-@show reasons</code></pre><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>GLPK.glp_XXX</code> functions, where the names and arguments are identical to the C API. See the <code>/tests</code> folder for inspiration.</p><h2 id="Thread-safety"><a class="docs-heading-anchor" href="#Thread-safety">Thread safety</a><a id="Thread-safety-1"></a><a class="docs-heading-anchor-permalink" href="#Thread-safety" title="Permalink"></a></h2><p>GLPK is not thread-safe and should not be used with multithreading.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ECOS/">« jump-dev/ECOS.jl</a><a class="docs-footer-nextpage" href="../Gurobi/">jump-dev/Gurobi.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@show reasons</code></pre><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>GLPK.glp_XXX</code> functions, where the names and arguments are identical to the C API. See the <code>/tests</code> folder for inspiration.</p><h2 id="Thread-safety"><a class="docs-heading-anchor" href="#Thread-safety">Thread safety</a><a id="Thread-safety-1"></a><a class="docs-heading-anchor-permalink" href="#Thread-safety" title="Permalink"></a></h2><p>GLPK is not thread-safe and should not be used with multithreading.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ECOS/">« jump-dev/ECOS.jl</a><a class="docs-footer-nextpage" href="../Gurobi/">jump-dev/Gurobi.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Gurobi/index.html b/previews/PR3562/packages/Gurobi/index.html
index 0e46e4302fe..f10c6f13e4b 100644
--- a/previews/PR3562/packages/Gurobi/index.html
+++ b/previews/PR3562/packages/Gurobi/index.html
@@ -109,4 +109,4 @@
     println(lower_bound(x[i]))
 end</code></pre><h2 id="Common-errors"><a class="docs-heading-anchor" href="#Common-errors">Common errors</a><a id="Common-errors-1"></a><a class="docs-heading-anchor-permalink" href="#Common-errors" title="Permalink"></a></h2><h3 id="Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?"><a class="docs-heading-anchor" href="#Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?">Using Gurobi v9.0 and you got an error like <code>Q not PSD</code>?</a><a id="Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?-1"></a><a class="docs-heading-anchor-permalink" href="#Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?" title="Permalink"></a></h3><p>You need to set the NonConvex parameter:</p><pre><code class="language-julia hljs">model = Model(Gurobi.Optimizer)
 set_optimizer_attribute(model, &quot;NonConvex&quot;, 2)</code></pre><h3 id="Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X"><a class="docs-heading-anchor" href="#Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X">Gurobi Error 1009: Version number is XX.X, license is for version XX.X</a><a id="Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X-1"></a><a class="docs-heading-anchor-permalink" href="#Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X" title="Permalink"></a></h3><p>Make sure that your license is correct for your Gurobi version. See the <a href="https://support.gurobi.com/hc/en-us/articles/360034784572-How-do-I-check-for-a-valid-license-file-">Gurobi documentation</a> for details.</p><p>Once you are sure that the license and Gurobi versions match, re-install Gurobi.jl by running:</p><pre><code class="language-julia hljs">import Pkg
-Pkg.build(&quot;Gurobi&quot;)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../GLPK/">« jump-dev/GLPK.jl</a><a class="docs-footer-nextpage" href="../HiGHS/">jump-dev/HiGHS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Pkg.build(&quot;Gurobi&quot;)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../GLPK/">« jump-dev/GLPK.jl</a><a class="docs-footer-nextpage" href="../HiGHS/">jump-dev/HiGHS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/HiGHS/index.html b/previews/PR3562/packages/HiGHS/index.html
index 3b7a5aea1c1..31627cedc8d 100644
--- a/previews/PR3562/packages/HiGHS/index.html
+++ b/previews/PR3562/packages/HiGHS/index.html
@@ -7,4 +7,4 @@
 Pkg.add(&quot;HiGHS&quot;)</code></pre><p>In addition to installing the HiGHS.jl package, this will also download and install the HiGHS binaries. You do not need to install HiGHS separately.</p><p>To use a custom binary, read the <a href="https://jump.dev/JuMP.jl/stable/developers/custom_solver_binaries/">Custom solver binaries</a> section of the JuMP documentation.</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>To use HiGHS with JuMP, use <code>HiGHS.Optimizer</code>:</p><pre><code class="language-julia hljs">using JuMP, HiGHS
 model = Model(HiGHS.Optimizer)
 set_attribute(model, &quot;presolve&quot;, &quot;on&quot;)
-set_attribute(model, &quot;time_limit&quot;, 60.0)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The HiGHS optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>See the <a href="https://ergo-code.github.io/HiGHS/dev/options/definitions/">HiGHS documentation</a> for a full list of the available options.</p><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>HiGHS.Highs_xxx</code> functions, where the names and arguments are identical to the C API.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Gurobi/">« jump-dev/Gurobi.jl</a><a class="docs-footer-nextpage" href="../Ipopt/">jump-dev/Ipopt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;time_limit&quot;, 60.0)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The HiGHS optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>See the <a href="https://ergo-code.github.io/HiGHS/dev/options/definitions/">HiGHS documentation</a> for a full list of the available options.</p><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>HiGHS.Highs_xxx</code> functions, where the names and arguments are identical to the C API.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Gurobi/">« jump-dev/Gurobi.jl</a><a class="docs-footer-nextpage" href="../Ipopt/">jump-dev/Ipopt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Hypatia/index.html b/previews/PR3562/packages/Hypatia/index.html
index a14b3e4f961..6c4f6a43257 100644
--- a/previews/PR3562/packages/Hypatia/index.html
+++ b/previews/PR3562/packages/Hypatia/index.html
@@ -42,4 +42,4 @@
     volume={15},
     pages={53--101},
     doi={https://doi.org/10.1007/s12532-022-00226-0}
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../PATHSolver/">« chkwon/PATHSolver.jl</a><a class="docs-footer-nextpage" href="../DAQP/">darnstrom/DAQP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../PATHSolver/">« chkwon/PATHSolver.jl</a><a class="docs-footer-nextpage" href="../DAQP/">darnstrom/DAQP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/InfiniteOpt/index.html b/previews/PR3562/packages/InfiniteOpt/index.html
index 302b6720026..ae2a83ca390 100644
--- a/previews/PR3562/packages/InfiniteOpt/index.html
+++ b/previews/PR3562/packages/InfiniteOpt/index.html
@@ -12,4 +12,4 @@
       doi = {https://doi.org/10.1016/j.compchemeng.2021.107567},
       url = {https://www.sciencedirect.com/science/article/pii/S0098135421003458},
       author = {Joshua L. Pulsipher and Weiqi Zhang and Tyler J. Hongisto and Victor M. Zavala},
-}</code></pre><p>A pre-print version is freely available though <a href="https://arxiv.org/abs/2106.12689">arXiv</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SumOfSquares/">« jump-dev/SumOfSquares.jl</a><a class="docs-footer-nextpage" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><p>A pre-print version is freely available though <a href="https://arxiv.org/abs/2106.12689">arXiv</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SumOfSquares/">« jump-dev/SumOfSquares.jl</a><a class="docs-footer-nextpage" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Ipopt/index.html b/previews/PR3562/packages/Ipopt/index.html
index ed5dd6dfe77..26a49e3bdb9 100644
--- a/previews/PR3562/packages/Ipopt/index.html
+++ b/previews/PR3562/packages/Ipopt/index.html
@@ -118,4 +118,4 @@
 export OMP_PROC_BIND=TRUE</code></pre><h2 id="BLAS-and-LAPACK"><a class="docs-heading-anchor" href="#BLAS-and-LAPACK">BLAS and LAPACK</a><a id="BLAS-and-LAPACK-1"></a><a class="docs-heading-anchor-permalink" href="#BLAS-and-LAPACK" title="Permalink"></a></h2><p>With Julia v1.9 or later, Ipopt and the linear solvers <a href="https://mumps-solver.org/index.php">MUMPS</a> (default), SPRAL, and HSL are compiled with <a href="https://github.com/JuliaLinearAlgebra/libblastrampoline"><code>libblastrampoline</code></a> (LBT), a library that can change between BLAS and LAPACK backends at runtime.</p><p>The default BLAS and LAPACK backend is <a href="https://github.com/OpenMathLib/OpenBLAS">OpenBLAS</a>.</p><p>Using LBT, we can also switch dynamically to other BLAS backends such as Intel MKL and Apple Accelerate. Because Ipopt and the linear solvers heavily rely on BLAS and LAPACK routines, using an optimized backend for a particular platform can improve the performance.</p><h3 id="MKL"><a class="docs-heading-anchor" href="#MKL">MKL</a><a id="MKL-1"></a><a class="docs-heading-anchor-permalink" href="#MKL" title="Permalink"></a></h3><p>If you have <a href="https://github.com/JuliaLinearAlgebra/MKL.jl">MKL.jl</a> installed, switch to MKL by adding <code>using MKL</code> to your code:</p><pre><code class="language-julia hljs">using MKL  # Replace OpenBLAS by Intel MKL
 using Ipopt</code></pre><h3 id="AppleAccelerate"><a class="docs-heading-anchor" href="#AppleAccelerate">AppleAccelerate</a><a id="AppleAccelerate-1"></a><a class="docs-heading-anchor-permalink" href="#AppleAccelerate" title="Permalink"></a></h3><p>If you are using macOS ≥ v13.4 and you have <a href="https://github.com/JuliaLinearAlgebra/AppleAccelerate.jl">AppleAccelerate.jl</a> installed, add <code>using AppleAccelerate</code> to your code:</p><pre><code class="language-julia hljs">using AppleAccelerate  # Replace OpenBLAS by Apple Accelerate
 using Ipopt</code></pre><h3 id="Display-backends"><a class="docs-heading-anchor" href="#Display-backends">Display backends</a><a id="Display-backends-1"></a><a class="docs-heading-anchor-permalink" href="#Display-backends" title="Permalink"></a></h3><p>Check what backends are loaded using:</p><pre><code class="language-julia hljs">import LinearAlgebra
-LinearAlgebra.BLAS.lbt_get_config()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../HiGHS/">« jump-dev/HiGHS.jl</a><a class="docs-footer-nextpage" href="../KNITRO/">jump-dev/KNITRO.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+LinearAlgebra.BLAS.lbt_get_config()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../HiGHS/">« jump-dev/HiGHS.jl</a><a class="docs-footer-nextpage" href="../KNITRO/">jump-dev/KNITRO.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Juniper/index.html b/previews/PR3562/packages/Juniper/index.html
index 9a7f9f35ccf..2f678562aff 100644
--- a/previews/PR3562/packages/Juniper/index.html
+++ b/previews/PR3562/packages/Juniper/index.html
@@ -33,4 +33,4 @@
      year=&quot;2018&quot;,
      publisher=&quot;Springer International Publishing&quot;,
      isbn=&quot;978-3-319-93031-2&quot;
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Alpine/">« lanl-ansi/Alpine.jl</a><a class="docs-footer-nextpage" href="../ProxSDP/">mariohsouto/ProxSDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Alpine/">« lanl-ansi/Alpine.jl</a><a class="docs-footer-nextpage" href="../ProxSDP/">mariohsouto/ProxSDP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/KNITRO/index.html b/previews/PR3562/packages/KNITRO/index.html
index 908d669d94b..e7963314433 100644
--- a/previews/PR3562/packages/KNITRO/index.html
+++ b/previews/PR3562/packages/KNITRO/index.html
@@ -10,4 +10,4 @@
 set_attribute(model, &quot;algorithm&quot;, 4)</code></pre><h2 id="Use-with-AMPL"><a class="docs-heading-anchor" href="#Use-with-AMPL">Use with AMPL</a><a id="Use-with-AMPL-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-AMPL" title="Permalink"></a></h2><p>To use KNITRO with <a href="https://github.com/jump-dev/AmplNLWriter.jl">AmplNLWriter.jl</a>, use <code>KNITRO.amplexe</code>:</p><pre><code class="language-julia hljs">using JuMP
 import AmplNLWriter
 import KNITRO
-model = Model(() -&gt; AmplNLWriter.Optimizer(KNITRO.amplexe, [&quot;outlev=3&quot;]))</code></pre><h2 id="Use-with-other-packages"><a class="docs-heading-anchor" href="#Use-with-other-packages">Use with other packages</a><a id="Use-with-other-packages-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-other-packages" title="Permalink"></a></h2><p>A variety of packages extend KNITRO.jl to support other optimization modeling systems. These include:</p><ul><li><a href="https://github.com/JuliaSmoothOptimizers/NLPModelsKnitro.jl">NLPModelsKnitro</a></li><li><a href="http://optimization.sciml.ai/stable/">Optimization.jl</a></li></ul><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The Knitro optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.VariableIndex}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Complements"><code>MOI.Complements</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart"><code>MOI.NLPBlockDualStart()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>A list of available options is provided in the <a href="https://www.artelys.com/docs/knitro/3_referenceManual/userOptions.html">KNITRO reference manual</a>.</p><h2 id="Low-level-wrapper"><a class="docs-heading-anchor" href="#Low-level-wrapper">Low-level wrapper</a><a id="Low-level-wrapper-1"></a><a class="docs-heading-anchor-permalink" href="#Low-level-wrapper" title="Permalink"></a></h2><p>KNITRO.jl implements most of Knitro&#39;s functionalities. If you aim at using part of Knitro&#39;s API that are not implemented in the MathOptInterface/JuMP ecosystem, you can refer to the low-level API, which wraps Knitro&#39;s C API (whose templates are specified in the file <code>knitro.h</code>).</p><p>Extensive examples using the C wrapper can be found in <code>examples/</code>.</p><h2 id="Multi-threading"><a class="docs-heading-anchor" href="#Multi-threading">Multi-threading</a><a id="Multi-threading-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-threading" title="Permalink"></a></h2><p>Due to limitations in the interaction between Julia and C, KNITRO.jl disables multi-threading if the problem is nonlinear. This will override any options such as <code>par_numthreads</code> that you may have set. Read <a href="https://github.com/jump-dev/KNITRO.jl/issues/93">GitHub issue #93</a> for more details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Ipopt/">« jump-dev/Ipopt.jl</a><a class="docs-footer-nextpage" href="../MiniZinc/">jump-dev/MiniZinc.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+model = Model(() -&gt; AmplNLWriter.Optimizer(KNITRO.amplexe, [&quot;outlev=3&quot;]))</code></pre><h2 id="Use-with-other-packages"><a class="docs-heading-anchor" href="#Use-with-other-packages">Use with other packages</a><a id="Use-with-other-packages-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-other-packages" title="Permalink"></a></h2><p>A variety of packages extend KNITRO.jl to support other optimization modeling systems. These include:</p><ul><li><a href="https://github.com/JuliaSmoothOptimizers/NLPModelsKnitro.jl">NLPModelsKnitro</a></li><li><a href="http://optimization.sciml.ai/stable/">Optimization.jl</a></li></ul><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The Knitro optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.VariableIndex}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorAffineFunction"><code>MOI.VectorAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Complements"><code>MOI.Complements</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart"><code>MOI.NLPBlockDualStart()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>A list of available options is provided in the <a href="https://www.artelys.com/docs/knitro/3_referenceManual/userOptions.html">KNITRO reference manual</a>.</p><h2 id="Low-level-wrapper"><a class="docs-heading-anchor" href="#Low-level-wrapper">Low-level wrapper</a><a id="Low-level-wrapper-1"></a><a class="docs-heading-anchor-permalink" href="#Low-level-wrapper" title="Permalink"></a></h2><p>KNITRO.jl implements most of Knitro&#39;s functionalities. If you aim at using part of Knitro&#39;s API that are not implemented in the MathOptInterface/JuMP ecosystem, you can refer to the low-level API, which wraps Knitro&#39;s C API (whose templates are specified in the file <code>knitro.h</code>).</p><p>Extensive examples using the C wrapper can be found in <code>examples/</code>.</p><h2 id="Multi-threading"><a class="docs-heading-anchor" href="#Multi-threading">Multi-threading</a><a id="Multi-threading-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-threading" title="Permalink"></a></h2><p>Due to limitations in the interaction between Julia and C, KNITRO.jl disables multi-threading if the problem is nonlinear. This will override any options such as <code>par_numthreads</code> that you may have set. Read <a href="https://github.com/jump-dev/KNITRO.jl/issues/93">GitHub issue #93</a> for more details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Ipopt/">« jump-dev/Ipopt.jl</a><a class="docs-footer-nextpage" href="../MiniZinc/">jump-dev/MiniZinc.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Loraine/index.html b/previews/PR3562/packages/Loraine/index.html
index 5434d3c62c9..1f961b5364d 100644
--- a/previews/PR3562/packages/Loraine/index.html
+++ b/previews/PR3562/packages/Loraine/index.html
@@ -29,4 +29,4 @@
   www={https://hal.science/hal-04076509/}
   note={Preprint hal-04076509}
   year={2023}
-}</code></pre><section class="footnotes is-size-7"><ul><li class="footnote" id="footnote-1"><a class="tag is-link" href="#citeref-1">1</a>https://www.youtube.com/watch?v=0D2wNf1lVrI</li></ul></section></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Tulip/">« ds4dm/Tulip.jl</a><a class="docs-footer-nextpage" href="../Alpine/">lanl-ansi/Alpine.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><section class="footnotes is-size-7"><ul><li class="footnote" id="footnote-1"><a class="tag is-link" href="#citeref-1">1</a>https://www.youtube.com/watch?v=0D2wNf1lVrI</li></ul></section></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Tulip/">« ds4dm/Tulip.jl</a><a class="docs-footer-nextpage" href="../Alpine/">lanl-ansi/Alpine.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/MadNLP/index.html b/previews/PR3562/packages/MadNLP/index.html
index bbb879c3e0a..5bfc45a6eae 100644
--- a/previews/PR3562/packages/MadNLP/index.html
+++ b/previews/PR3562/packages/MadNLP/index.html
@@ -45,4 +45,4 @@
   author={Shin, Sungho and Coffrin, Carleton and Sundar, Kaarthik and Zavala, Victor M},
   journal={arXiv preprint arXiv:2010.02404},
   year={2020}
-}</code></pre><h2 id="Bug-reports-and-support"><a class="docs-heading-anchor" href="#Bug-reports-and-support">Bug reports and support</a><a id="Bug-reports-and-support-1"></a><a class="docs-heading-anchor-permalink" href="#Bug-reports-and-support" title="Permalink"></a></h2><p>Please report issues and feature requests via the <a href="https://github.com/MadNLP/MadNLP.jl/issues">GitHub issue tracker</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CDDLib/">« JuliaPolyhedra/CDDLib.jl</a><a class="docs-footer-nextpage" href="../PATHSolver/">chkwon/PATHSolver.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><h2 id="Bug-reports-and-support"><a class="docs-heading-anchor" href="#Bug-reports-and-support">Bug reports and support</a><a id="Bug-reports-and-support-1"></a><a class="docs-heading-anchor-permalink" href="#Bug-reports-and-support" title="Permalink"></a></h2><p>Please report issues and feature requests via the <a href="https://github.com/MadNLP/MadNLP.jl/issues">GitHub issue tracker</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../CDDLib/">« JuliaPolyhedra/CDDLib.jl</a><a class="docs-footer-nextpage" href="../PATHSolver/">chkwon/PATHSolver.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/MiniZinc/index.html b/previews/PR3562/packages/MiniZinc/index.html
index d0c5362b5f0..60eac87bb39 100644
--- a/previews/PR3562/packages/MiniZinc/index.html
+++ b/previews/PR3562/packages/MiniZinc/index.html
@@ -53,4 +53,4 @@
 @constraint(model, x in MOI.AllDifferent(3))
 @objective(model, Max, sum(i * x[i] for i in 1:3))
 optimize!(model)
-@show value.(x)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The MiniZinc <code>Optimizer{T}</code> supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.VariableIndex}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Parameter"><code>MOI.Parameter{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.AllDifferent"><code>MOI.AllDifferent</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.BinPacking"><code>MOI.BinPacking{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Circuit"><code>MOI.Circuit</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountAtLeast"><code>MOI.CountAtLeast</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountBelongs"><code>MOI.CountBelongs</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountDistinct"><code>MOI.CountDistinct</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountGreaterThan"><code>MOI.CountGreaterThan</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Cumulative"><code>MOI.Cumulative</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Path"><code>MOI.Path</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Table"><code>MOI.Table{T}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>Set options using <a href="../../moi/reference/models/#MathOptInterface.RawOptimizerAttribute"><code>MOI.RawOptimizerAttribute</code></a> in MOI or <a href="../../api/JuMP/#JuMP.set_attribute"><code>set_attribute</code></a> in JuMP.</p><p><code>MiniZinc.jl</code> supports the following options:</p><ul><li><code>model_filename::String = &quot;&quot;</code>: the location at which to write out the <code>.mzn</code> file during optimization. This option can be helpful during debugging. If left empty, a temporary file will be used instead.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../KNITRO/">« jump-dev/KNITRO.jl</a><a class="docs-footer-nextpage" href="../MosekTools/">jump-dev/MosekTools.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@show value.(x)</code></pre><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The MiniZinc <code>Optimizer{T}</code> supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.VariableIndex}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{T}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Parameter"><code>MOI.Parameter{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semicontinuous"><code>MOI.Semicontinuous{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Semiinteger"><code>MOI.Semiinteger{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.AllDifferent"><code>MOI.AllDifferent</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.BinPacking"><code>MOI.BinPacking{T}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Circuit"><code>MOI.Circuit</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountAtLeast"><code>MOI.CountAtLeast</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountBelongs"><code>MOI.CountBelongs</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountDistinct"><code>MOI.CountDistinct</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.CountGreaterThan"><code>MOI.CountGreaterThan</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Cumulative"><code>MOI.Cumulative</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Path"><code>MOI.Path</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Table"><code>MOI.Table{T}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.Name"><code>MOI.Name()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li></ul><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>Set options using <a href="../../moi/reference/models/#MathOptInterface.RawOptimizerAttribute"><code>MOI.RawOptimizerAttribute</code></a> in MOI or <a href="../../api/JuMP/#JuMP.set_attribute"><code>set_attribute</code></a> in JuMP.</p><p><code>MiniZinc.jl</code> supports the following options:</p><ul><li><code>model_filename::String = &quot;&quot;</code>: the location at which to write out the <code>.mzn</code> file during optimization. This option can be helpful during debugging. If left empty, a temporary file will be used instead.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../KNITRO/">« jump-dev/KNITRO.jl</a><a class="docs-footer-nextpage" href="../MosekTools/">jump-dev/MosekTools.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/MosekTools/index.html b/previews/PR3562/packages/MosekTools/index.html
index bf8ae86dc9a..1eb6074ebe4 100644
--- a/previews/PR3562/packages/MosekTools/index.html
+++ b/previews/PR3562/packages/MosekTools/index.html
@@ -7,4 +7,4 @@
 using MosekTools
 model = Model(Mosek.Optimizer)
 set_attribute(model, &quot;QUIET&quot;, true)
-set_attribute(model, &quot;INTPNT_CO_TOL_DFEAS&quot;, 1e-7)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The parameter <code>QUIET</code> is a special parameter that when set to <code>true</code> disables all Mosek printing output.</p><p>All other parameters can be found in the <a href="https://docs.mosek.com/8.1/capi/param-groups.html#doc-param-groups">Mosek documentation</a>.</p><p>Note that the prefix <code>MSK_IPAR_</code> (for integer parameters), <code>MSK_DPAR_</code> (for floating point parameters) or <code>MSK_SPAR_</code> (for string parameters) are optional. If they are not given, they are inferred from the type of the value. For example, in the example above, as <code>1e-7</code> is a floating point number, the parameters name used is <code>MSK_DPAR_INTPNT_CO_TOL_DFEAS</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MiniZinc/">« jump-dev/MiniZinc.jl</a><a class="docs-footer-nextpage" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;INTPNT_CO_TOL_DFEAS&quot;, 1e-7)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The parameter <code>QUIET</code> is a special parameter that when set to <code>true</code> disables all Mosek printing output.</p><p>All other parameters can be found in the <a href="https://docs.mosek.com/8.1/capi/param-groups.html#doc-param-groups">Mosek documentation</a>.</p><p>Note that the prefix <code>MSK_IPAR_</code> (for integer parameters), <code>MSK_DPAR_</code> (for floating point parameters) or <code>MSK_SPAR_</code> (for string parameters) are optional. If they are not given, they are inferred from the type of the value. For example, in the example above, as <code>1e-7</code> is a floating point number, the parameters name used is <code>MSK_DPAR_INTPNT_CO_TOL_DFEAS</code>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MiniZinc/">« jump-dev/MiniZinc.jl</a><a class="docs-footer-nextpage" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/MultiObjectiveAlgorithms/index.html b/previews/PR3562/packages/MultiObjectiveAlgorithms/index.html
index 70475bc5357..3d872dd7d4a 100644
--- a/previews/PR3562/packages/MultiObjectiveAlgorithms/index.html
+++ b/previews/PR3562/packages/MultiObjectiveAlgorithms/index.html
@@ -9,4 +9,4 @@
 import MultiObjectiveAlgorithms as MOA
 model = JuMP.Model(() -&gt; MOA.Optimizer(HiGHS.Optimizer))
 set_attribute(model, MOA.Algorithm(), MOA.Dichotomy())
-set_attribute(model, MOA.SolutionLimit(), 4)</code></pre><p>Replace <code>HiGHS.Optimizer</code> with an optimizer capable of solving a single-objective instance of your optimization problem.</p><p>You may set additional optimizer attributes, the supported attributes depend on the choice of solution algorithm.</p><h2 id="Algorithm"><a class="docs-heading-anchor" href="#Algorithm">Algorithm</a><a id="Algorithm-1"></a><a class="docs-heading-anchor-permalink" href="#Algorithm" title="Permalink"></a></h2><p>Set the algorithm using the <code>MOA.Algorithm()</code> attribute.</p><p>The value must be one of the algorithms supported by MOA:</p><ul><li><code>MOA.Chalmet()</code></li><li><code>MOA.Dichotomy()</code></li><li><code>MOA.DominguezRios()</code></li><li><code>MOA.EpsilonConstraint()</code></li><li><code>MOA.Hierarchical()</code></li><li><code>MOA.KirlikSayin()</code></li><li><code>MOA.Lexicographic()</code> [default]</li><li><code>MOA.TambyVanderpooten()</code></li></ul><p>Consult their docstrings for details.</p><h2 id="Other-optimizer-attributes"><a class="docs-heading-anchor" href="#Other-optimizer-attributes">Other optimizer attributes</a><a id="Other-optimizer-attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Other-optimizer-attributes" title="Permalink"></a></h2><p>There are a number of optimizer attributes supported by the algorithms in MOA.</p><p>Each algorithm supports only a subset of the attributes. Consult the algorithm&#39;s docstring for details on which attributes it supports, and how it uses them in the solution process.</p><ul><li><code>MOA.EpsilonConstraintStep()</code></li><li><code>MOA.LexicographicAllPermutations()</code></li><li><code>MOA.ObjectiveAbsoluteTolerance(index::Int)</code></li><li><code>MOA.ObjectivePriority(index::Int)</code></li><li><code>MOA.ObjectiveRelativeTolerance(index::Int)</code></li><li><code>MOA.ObjectiveWeight(index::Int)</code></li><li><code>MOA.SolutionLimit()</code></li><li><code>MOI.TimeLimitSec()</code></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MosekTools/">« jump-dev/MosekTools.jl</a><a class="docs-footer-nextpage" href="../Pajarito/">jump-dev/Pajarito.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, MOA.SolutionLimit(), 4)</code></pre><p>Replace <code>HiGHS.Optimizer</code> with an optimizer capable of solving a single-objective instance of your optimization problem.</p><p>You may set additional optimizer attributes, the supported attributes depend on the choice of solution algorithm.</p><h2 id="Algorithm"><a class="docs-heading-anchor" href="#Algorithm">Algorithm</a><a id="Algorithm-1"></a><a class="docs-heading-anchor-permalink" href="#Algorithm" title="Permalink"></a></h2><p>Set the algorithm using the <code>MOA.Algorithm()</code> attribute.</p><p>The value must be one of the algorithms supported by MOA:</p><ul><li><code>MOA.Chalmet()</code></li><li><code>MOA.Dichotomy()</code></li><li><code>MOA.DominguezRios()</code></li><li><code>MOA.EpsilonConstraint()</code></li><li><code>MOA.Hierarchical()</code></li><li><code>MOA.KirlikSayin()</code></li><li><code>MOA.Lexicographic()</code> [default]</li><li><code>MOA.TambyVanderpooten()</code></li></ul><p>Consult their docstrings for details.</p><h2 id="Other-optimizer-attributes"><a class="docs-heading-anchor" href="#Other-optimizer-attributes">Other optimizer attributes</a><a id="Other-optimizer-attributes-1"></a><a class="docs-heading-anchor-permalink" href="#Other-optimizer-attributes" title="Permalink"></a></h2><p>There are a number of optimizer attributes supported by the algorithms in MOA.</p><p>Each algorithm supports only a subset of the attributes. Consult the algorithm&#39;s docstring for details on which attributes it supports, and how it uses them in the solution process.</p><ul><li><code>MOA.EpsilonConstraintStep()</code></li><li><code>MOA.LexicographicAllPermutations()</code></li><li><code>MOA.ObjectiveAbsoluteTolerance(index::Int)</code></li><li><code>MOA.ObjectivePriority(index::Int)</code></li><li><code>MOA.ObjectiveRelativeTolerance(index::Int)</code></li><li><code>MOA.ObjectiveWeight(index::Int)</code></li><li><code>MOA.SolutionLimit()</code></li><li><code>MOI.TimeLimitSec()</code></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MosekTools/">« jump-dev/MosekTools.jl</a><a class="docs-footer-nextpage" href="../Pajarito/">jump-dev/Pajarito.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/NEOSServer/index.html b/previews/PR3562/packages/NEOSServer/index.html
index 9764c673e8a..03653d5a7de 100644
--- a/previews/PR3562/packages/NEOSServer/index.html
+++ b/previews/PR3562/packages/NEOSServer/index.html
@@ -28,4 +28,4 @@
 
 model = Model() do
     NEOSServer.Optimizer(email=&quot;me@mydomain.com&quot;, solver=&quot;Ipopt&quot;)
-end</code></pre><p><strong>Note: <code>NEOSServer.Optimizer</code> is limited to the following solvers:</strong></p><ul><li><code>&quot;CPLEX&quot;</code></li><li><code>&quot;FICO-Xpress&quot;</code></li><li><code>&quot;Ipopt&quot;</code></li><li><code>&quot;Knitro&quot;</code></li><li><code>&quot;MOSEK&quot;</code></li><li><code>&quot;OCTERACT&quot;</code></li><li><code>&quot;SNOPT&quot;</code></li></ul><h2 id="NEOS-Limits"><a class="docs-heading-anchor" href="#NEOS-Limits">NEOS Limits</a><a id="NEOS-Limits-1"></a><a class="docs-heading-anchor-permalink" href="#NEOS-Limits" title="Permalink"></a></h2><p>NEOS currently limits jobs to an 8 hour time limit, 3 GB of memory, and a 16 MB submission file. If your model exceeds these limits, NEOSServer.jl may be unable to return useful information to the user.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ProxSDP/">« mariohsouto/ProxSDP.jl</a><a class="docs-footer-nextpage" href="../OSQP/">osqp/OSQP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><p><strong>Note: <code>NEOSServer.Optimizer</code> is limited to the following solvers:</strong></p><ul><li><code>&quot;CPLEX&quot;</code></li><li><code>&quot;FICO-Xpress&quot;</code></li><li><code>&quot;Ipopt&quot;</code></li><li><code>&quot;Knitro&quot;</code></li><li><code>&quot;MOSEK&quot;</code></li><li><code>&quot;OCTERACT&quot;</code></li><li><code>&quot;SNOPT&quot;</code></li></ul><h2 id="NEOS-Limits"><a class="docs-heading-anchor" href="#NEOS-Limits">NEOS Limits</a><a id="NEOS-Limits-1"></a><a class="docs-heading-anchor-permalink" href="#NEOS-Limits" title="Permalink"></a></h2><p>NEOS currently limits jobs to an 8 hour time limit, 3 GB of memory, and a 16 MB submission file. If your model exceeds these limits, NEOSServer.jl may be unable to return useful information to the user.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ProxSDP/">« mariohsouto/ProxSDP.jl</a><a class="docs-footer-nextpage" href="../OSQP/">osqp/OSQP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/NLopt/index.html b/previews/PR3562/packages/NLopt/index.html
index f23269a1a79..0e13d0d02c1 100644
--- a/previews/PR3562/packages/NLopt/index.html
+++ b/previews/PR3562/packages/NLopt/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li class="is-active"><a class="tocitem" href>JuliaOpt/NLopt.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Use-with-JuMP"><span>Use with JuMP</span></a></li><li><a class="tocitem" href="#Options"><span>Options</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li></ul></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>JuliaOpt/NLopt.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>JuliaOpt/NLopt.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/packages/NLopt.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="NLopt.jl"><a class="docs-heading-anchor" href="#NLopt.jl">NLopt.jl</a><a id="NLopt.jl-1"></a><a class="docs-heading-anchor-permalink" href="#NLopt.jl" title="Permalink"></a></h1><p><a href="https://github.com/JuliaOpt/NLopt.jl/actions?query=workflow%3ACI"><img src="https://github.com/JuliaOpt/NLopt.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/JuliaOpt/NLopt.jl"><img src="https://codecov.io/gh/JuliaOpt/NLopt.jl/branch/master/graph/badge.svg" alt="codecov"/></a></p><p><a href="https://github.com/JuliaOpt/NLopt.jl">NLopt.jl</a> is a wrapper for the <a href="https://nlopt.readthedocs.io/en/latest/">NLopt</a> library.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>NLopt.jl</code> is licensed under the <a href="https://github.com/JuliaOpt/NLopt.jl/blob/master/LICENSE.md">MIT License</a>.</p><p>The underlying solver, <a href="https://github.com/stevengj/nlopt">stevengj/nlopt</a>, is licensed under the <a href="https://github.com/stevengj/nlopt/blob/master/COPYING">LGPL v3.0 license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install <code>NLopt.jl</code> using the Julia package manager:</p><pre><code class="language-julia hljs">import Pkg
 Pkg.add(&quot;NLopt&quot;)</code></pre><p>In addition to installing the <code>NLopt.jl</code> package, this will also download and install the NLopt binaries. You do not need to install NLopt separately.</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>You can use NLopt with JuMP as follows:</p><pre><code class="language-julia hljs">using JuMP, NLopt
 model = Model(NLopt.Optimizer)
-set_attribute(model, &quot;algorithm&quot;, :LD_MMA)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The <code>algorithm</code> attribute is required. The value must be one of the supported <a href="https://nlopt.readthedocs.io/en/latest/NLopt_Algorithms/">NLopt algorithms</a>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>For more details, see the <a href="https://github.com/JuliaOpt/NLopt.jl/blob/master/README.md">NLopt.jl README</a> or the <a href="https://nlopt.readthedocs.io/en/latest/">NLopt documentation</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../GAMS/">« GAMS-dev/GAMS.jl</a><a class="docs-footer-nextpage" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;algorithm&quot;, :LD_MMA)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The <code>algorithm</code> attribute is required. The value must be one of the supported <a href="https://nlopt.readthedocs.io/en/latest/NLopt_Algorithms/">NLopt algorithms</a>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>For more details, see the <a href="https://github.com/JuliaOpt/NLopt.jl/blob/master/README.md">NLopt.jl README</a> or the <a href="https://nlopt.readthedocs.io/en/latest/">NLopt documentation</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../GAMS/">« GAMS-dev/GAMS.jl</a><a class="docs-footer-nextpage" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/OSQP/index.html b/previews/PR3562/packages/OSQP/index.html
index 34402467aa6..75b7f900a4c 100644
--- a/previews/PR3562/packages/OSQP/index.html
+++ b/previews/PR3562/packages/OSQP/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li class="is-active"><a class="tocitem" href>osqp/OSQP.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Problem-class"><span>Problem class</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li></ul></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>osqp/OSQP.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>osqp/OSQP.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/osqp/OSQP.jl/blob/443706e34c2619acbe65281c60bbe850ca4a8fac/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="OSQP.jl"><a class="docs-heading-anchor" href="#OSQP.jl">OSQP.jl</a><a id="OSQP.jl-1"></a><a class="docs-heading-anchor-permalink" href="#OSQP.jl" title="Permalink"></a></h1><p><a href="https://github.com/osqp/OSQP.jl/actions"><img src="https://github.com/osqp/OSQP.jl/workflows/CI/badge.svg" alt="Build Status"/></a>  <a href="http://codecov.io/github/osqp/OSQP.jl?branch=master"><img src="http://codecov.io/github/osqp/OSQP.jl/coverage.svg?branch=master" alt="codecov.io"/></a></p><p><a href="https://github.com/osqp/OSQP.jl">OSQP.jl</a> is a Julia wrapper for <a href="https://osqp.org/">OSQP</a>: the Operator Splitting QP Solver.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p>OSQP.jl is licensed under the <a href="https://github.com/osqp/OSQP.jl/blob/master/LICENSE.md">Apache-2.0 license</a>.</p><p>The upstream solver, <a href="https://github.com/osqp/osqp">osqp/osqp</a> is also licensed under the <a href="https://github.com/osqp/osqp/blob/master/LICENSE">Apache-2.0 license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install OSQP.jl using the Julia package manager</p><pre><code class="language-julia hljs">import Pkg
 Pkg.add(&quot;OSQP&quot;)</code></pre><h2 id="Problem-class"><a class="docs-heading-anchor" href="#Problem-class">Problem class</a><a id="Problem-class-1"></a><a class="docs-heading-anchor-permalink" href="#Problem-class" title="Permalink"></a></h2><p>The OSQP (Operator Splitting Quadratic Program) solver is a numerical optimization package for solving problems in the form</p><pre><code class="nohighlight hljs">minimize        0.5 x&#39; P x + q&#39; x
 
-subject to      l &lt;= A x &lt;= u</code></pre><p>where <code>x in R^n</code> is the optimization variable. The objective function is defined by a positive semidefinite matrix <code>P in S^n_+</code> and vector <code>q in R^n</code>. The linear constraints are defined by matrix <code>A in R^{m x n}</code> and vectors <code>l in R^m U {-inf}^m</code>, <code>u in R^m U {+inf}^m</code>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>Detailed documentation is available at <a href="https://osqp.org/">https://osqp.org/</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../NEOSServer/">« odow/NEOSServer.jl</a><a class="docs-footer-nextpage" href="../CDCS/">oxfordcontrol/CDCS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+subject to      l &lt;= A x &lt;= u</code></pre><p>where <code>x in R^n</code> is the optimization variable. The objective function is defined by a positive semidefinite matrix <code>P in S^n_+</code> and vector <code>q in R^n</code>. The linear constraints are defined by matrix <code>A in R^{m x n}</code> and vectors <code>l in R^m U {-inf}^m</code>, <code>u in R^m U {+inf}^m</code>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>Detailed documentation is available at <a href="https://osqp.org/">https://osqp.org/</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../NEOSServer/">« odow/NEOSServer.jl</a><a class="docs-footer-nextpage" href="../CDCS/">oxfordcontrol/CDCS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/PATHSolver/index.html b/previews/PR3562/packages/PATHSolver/index.html
index 4df374e7734..d6e81ddc335 100644
--- a/previews/PR3562/packages/PATHSolver/index.html
+++ b/previews/PR3562/packages/PATHSolver/index.html
@@ -165,4 +165,4 @@
  0.8
  1.2</code></pre><h2 id="Thread-safety"><a class="docs-heading-anchor" href="#Thread-safety">Thread safety</a><a id="Thread-safety-1"></a><a class="docs-heading-anchor-permalink" href="#Thread-safety" title="Permalink"></a></h2><p>PATH is not thread-safe and there are no known work-arounds. Do not run it in parallel using <code>Threads.@threads</code>. See <a href="https://github.com/chkwon/PATHSolver.jl/issues/62">issue #62</a> for more details.</p><h2 id="Factorization-methods"><a class="docs-heading-anchor" href="#Factorization-methods">Factorization methods</a><a id="Factorization-methods-1"></a><a class="docs-heading-anchor-permalink" href="#Factorization-methods" title="Permalink"></a></h2><p>By default, <code>PATHSolver.jl</code> will download the <a href="https://web.stanford.edu/group/SOL/software/lusol/">LUSOL</a> shared library. To use LUSOL, set the following options:</p><pre><code class="language-julia hljs">model = Model(PATHSolver.Optimizer)
 set_optimizer_attribute(model, &quot;factorization_method&quot;, &quot;blu_lusol&quot;)
-set_optimizer_attribute(model, &quot;factorization_library_name&quot;, PATHSolver.LUSOL_LIBRARY_PATH)</code></pre><p>To use <code>factorization_method umfpack</code> you will need the <code>umfpack</code> shared library that is available directly from the <a href="http://faculty.cse.tamu.edu/davis/suitesparse.html">developers of that code for academic use</a>.</p><h2 id="Manual-installation"><a class="docs-heading-anchor" href="#Manual-installation">Manual installation</a><a id="Manual-installation-1"></a><a class="docs-heading-anchor-permalink" href="#Manual-installation" title="Permalink"></a></h2><p>By default <code>PATHSolver.jl</code> will download a copy of the <code>libpath</code> library. If you already have one installed and want to use that, set the <code>PATH_JL_LOCATION</code> environment variable to point to the <code>libpath50.xx</code> library.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MadNLP/">« MadNLP/MadNLP.jl</a><a class="docs-footer-nextpage" href="../Hypatia/">chriscoey/Hypatia.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_optimizer_attribute(model, &quot;factorization_library_name&quot;, PATHSolver.LUSOL_LIBRARY_PATH)</code></pre><p>To use <code>factorization_method umfpack</code> you will need the <code>umfpack</code> shared library that is available directly from the <a href="http://faculty.cse.tamu.edu/davis/suitesparse.html">developers of that code for academic use</a>.</p><h2 id="Manual-installation"><a class="docs-heading-anchor" href="#Manual-installation">Manual installation</a><a id="Manual-installation-1"></a><a class="docs-heading-anchor-permalink" href="#Manual-installation" title="Permalink"></a></h2><p>By default <code>PATHSolver.jl</code> will download a copy of the <code>libpath</code> library. If you already have one installed and want to use that, set the <code>PATH_JL_LOCATION</code> environment variable to point to the <code>libpath50.xx</code> library.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MadNLP/">« MadNLP/MadNLP.jl</a><a class="docs-footer-nextpage" href="../Hypatia/">chriscoey/Hypatia.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Pajarito/index.html b/previews/PR3562/packages/Pajarito/index.html
index ef4336c5df2..83349613bdb 100644
--- a/previews/PR3562/packages/Pajarito/index.html
+++ b/previews/PR3562/packages/Pajarito/index.html
@@ -27,4 +27,4 @@
     pages={249--293},
     year={2020},
     publisher={Springer}
-}</code></pre><p>Note this paper describes a legacy MathProgBase version of Pajarito, which is available on the <a href="https://github.com/jump-dev/Pajarito.jl/tree/mathprogbase"><code>mathprogbase</code> branch</a> of this repository. Starting with version v0.8.0, Pajarito supports MathOptInterface instead of MathProgBase.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MultiObjectiveAlgorithms/">« jump-dev/MultiObjectiveAlgorithms.jl</a><a class="docs-footer-nextpage" href="../Pavito/">jump-dev/Pavito.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><p>Note this paper describes a legacy MathProgBase version of Pajarito, which is available on the <a href="https://github.com/jump-dev/Pajarito.jl/tree/mathprogbase"><code>mathprogbase</code> branch</a> of this repository. Starting with version v0.8.0, Pajarito supports MathOptInterface instead of MathProgBase.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../MultiObjectiveAlgorithms/">« jump-dev/MultiObjectiveAlgorithms.jl</a><a class="docs-footer-nextpage" href="../Pavito/">jump-dev/Pavito.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/ParametricOptInterface/index.html b/previews/PR3562/packages/ParametricOptInterface/index.html
index 1f66a732134..600a822aeff 100644
--- a/previews/PR3562/packages/ParametricOptInterface/index.html
+++ b/previews/PR3562/packages/ParametricOptInterface/index.html
@@ -13,4 +13,4 @@
 @objective(model, Min, 2x)
 optimize!(model)
 MOI.set(model, POI.ParameterValue(), p, 2.0)
-optimize!(model)</code></pre><h2 id="GSOC2020"><a class="docs-heading-anchor" href="#GSOC2020">GSOC2020</a><a id="GSOC2020-1"></a><a class="docs-heading-anchor-permalink" href="#GSOC2020" title="Permalink"></a></h2><p>ParametricOptInterface began as a <a href="https://summerofcode.withgoogle.com/archive/2020/projects/4959861055422464">NumFOCUS sponsored Google Summer of Code (2020) project</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../DiffOpt/">« jump-dev/DiffOpt.jl</a><a class="docs-footer-nextpage" href="../PolyJuMP/">jump-dev/PolyJuMP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+optimize!(model)</code></pre><h2 id="GSOC2020"><a class="docs-heading-anchor" href="#GSOC2020">GSOC2020</a><a id="GSOC2020-1"></a><a class="docs-heading-anchor-permalink" href="#GSOC2020" title="Permalink"></a></h2><p>ParametricOptInterface began as a <a href="https://summerofcode.withgoogle.com/archive/2020/projects/4959861055422464">NumFOCUS sponsored Google Summer of Code (2020) project</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../DiffOpt/">« jump-dev/DiffOpt.jl</a><a class="docs-footer-nextpage" href="../PolyJuMP/">jump-dev/PolyJuMP.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Pavito/index.html b/previews/PR3562/packages/Pavito/index.html
index edb316f6902..2b56be385b3 100644
--- a/previews/PR3562/packages/Pavito/index.html
+++ b/previews/PR3562/packages/Pavito/index.html
@@ -13,4 +13,4 @@
         &quot;cont_solver&quot; =&gt;
             optimizer_with_attributes(Ipopt.Optimizer, &quot;print_level&quot; =&gt; 0),
     ),
-)</code></pre><p>The algorithm implemented by Pavito itself is relatively simple; most of the hard work is performed by the MILP solver passed as <code>mip_solver</code> and the NLP solver passed as <code>cont_solver</code>.</p><p><strong>The performance of Pavito depends on these two types of solvers.</strong></p><p>For better performance, you should use a commercial MILP solver such as CPLEX or Gurobi.</p><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The following optimizer attributes can set to a <code>Pavito.Optimizer</code> to modify its behavior:</p><ul><li><code>log_level::Int</code> Verbosity flag: 0 for quiet, higher for basic solve info</li><li><code>timeout::Float64</code> Time limit for algorithm (in seconds)</li><li><code>rel_gap::Float64</code> Relative optimality gap termination condition</li><li><code>mip_solver_drives::Bool</code> Let MILP solver manage convergence (&quot;branch and cut&quot;)</li><li><code>mip_solver::MOI.OptimizerWithAttributes</code> MILP solver</li><li><code>cont_solver::MOI.OptimizerWithAttributes</code> Continuous NLP solver</li></ul><p><strong>Pavito is not yet numerically robust and may require tuning of parameters to improve convergence.</strong></p><p>If the default parameters don&#39;t work for you, please let us know by opening an issue.</p><p>For improved Pavito performance, MILP solver integrality tolerance and feasibility tolerances should typically be tightened, for example to <code>1e-8</code>.</p><h2 id="Bug-reports-and-support"><a class="docs-heading-anchor" href="#Bug-reports-and-support">Bug reports and support</a><a id="Bug-reports-and-support-1"></a><a class="docs-heading-anchor-permalink" href="#Bug-reports-and-support" title="Permalink"></a></h2><p>Please report any issues via the <a href="https://github.com/jump-dev/Pavito.jl/issues">GitHub issue tracker</a>. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. The <a href="https://discourse.julialang.org/c/domain/opt">Optimization (Mathematical)</a> category on Discourse is appropriate for general discussion.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Pajarito/">« jump-dev/Pajarito.jl</a><a class="docs-footer-nextpage" href="../SCS/">jump-dev/SCS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><p>The algorithm implemented by Pavito itself is relatively simple; most of the hard work is performed by the MILP solver passed as <code>mip_solver</code> and the NLP solver passed as <code>cont_solver</code>.</p><p><strong>The performance of Pavito depends on these two types of solvers.</strong></p><p>For better performance, you should use a commercial MILP solver such as CPLEX or Gurobi.</p><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The following optimizer attributes can set to a <code>Pavito.Optimizer</code> to modify its behavior:</p><ul><li><code>log_level::Int</code> Verbosity flag: 0 for quiet, higher for basic solve info</li><li><code>timeout::Float64</code> Time limit for algorithm (in seconds)</li><li><code>rel_gap::Float64</code> Relative optimality gap termination condition</li><li><code>mip_solver_drives::Bool</code> Let MILP solver manage convergence (&quot;branch and cut&quot;)</li><li><code>mip_solver::MOI.OptimizerWithAttributes</code> MILP solver</li><li><code>cont_solver::MOI.OptimizerWithAttributes</code> Continuous NLP solver</li></ul><p><strong>Pavito is not yet numerically robust and may require tuning of parameters to improve convergence.</strong></p><p>If the default parameters don&#39;t work for you, please let us know by opening an issue.</p><p>For improved Pavito performance, MILP solver integrality tolerance and feasibility tolerances should typically be tightened, for example to <code>1e-8</code>.</p><h2 id="Bug-reports-and-support"><a class="docs-heading-anchor" href="#Bug-reports-and-support">Bug reports and support</a><a id="Bug-reports-and-support-1"></a><a class="docs-heading-anchor-permalink" href="#Bug-reports-and-support" title="Permalink"></a></h2><p>Please report any issues via the <a href="https://github.com/jump-dev/Pavito.jl/issues">GitHub issue tracker</a>. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. The <a href="https://discourse.julialang.org/c/domain/opt">Optimization (Mathematical)</a> category on Discourse is appropriate for general discussion.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Pajarito/">« jump-dev/Pajarito.jl</a><a class="docs-footer-nextpage" href="../SCS/">jump-dev/SCS.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Plasmo/index.html b/previews/PR3562/packages/Plasmo/index.html
index c702699e029..1ffef474236 100644
--- a/previews/PR3562/packages/Plasmo/index.html
+++ b/previews/PR3562/packages/Plasmo/index.html
@@ -46,4 +46,4 @@
 eprint = {2006.05378},
 archivePrefix = {arXiv},
 primaryClass = {math.OC}
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDDP/">« odow/SDDP.jl</a><a class="docs-footer-nextpage" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDDP/">« odow/SDDP.jl</a><a class="docs-footer-nextpage" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/PolyJuMP/index.html b/previews/PR3562/packages/PolyJuMP/index.html
index 3fb5ae844ba..58ff57f526c 100644
--- a/previews/PR3562/packages/PolyJuMP/index.html
+++ b/previews/PR3562/packages/PolyJuMP/index.html
@@ -17,4 +17,4 @@
 @constraint(model, a * x * y^2 + y^3 &gt;= a * x)</code></pre><p>you need to specify how to interpret this nonnegativity constraint. To use Sum-of-Arithmetic-Geometric-Exponentials (SAGE), use</p><pre><code class="language-julia hljs">import PolyJuMP
 PolyJuMP.setpolymodule!(model, PolyJuMP.SAGE)</code></pre><p>To use Sum-of-Squares (SOS), use</p><pre><code class="language-julia hljs">import SumOfSquares
 PolyJuMP.setpolymodule!(model, SumOfSquares)</code></pre><p>or replace <code>model = Model()</code> by <code>model = SOSModel()</code>.</p><p>Alternatively, the nonnegativity constraint can be explicit:</p><pre><code class="language-julia hljs">@constraint(model, a * x * y^2 + y^3 - a * x in PolyJuMP.SAGE.Polynomials())
-@constraint(model, a * x * y^2 + y^3 - a * x in SumOfSquares.SOSCone())</code></pre><p>This allows mixing SAGE and SOS constraints in the same model.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>Documentation for <code>PolyJuMP.jl</code> is included in the <a href="https://jump.dev/SumOfSquares.jl/stable">documentation for SumOfSquares.jl</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ParametricOptInterface/">« jump-dev/ParametricOptInterface.jl</a><a class="docs-footer-nextpage" href="../SumOfSquares/">jump-dev/SumOfSquares.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@constraint(model, a * x * y^2 + y^3 - a * x in SumOfSquares.SOSCone())</code></pre><p>This allows mixing SAGE and SOS constraints in the same model.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>Documentation for <code>PolyJuMP.jl</code> is included in the <a href="https://jump.dev/SumOfSquares.jl/stable">documentation for SumOfSquares.jl</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ParametricOptInterface/">« jump-dev/ParametricOptInterface.jl</a><a class="docs-footer-nextpage" href="../SumOfSquares/">jump-dev/SumOfSquares.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/ProxSDP/index.html b/previews/PR3562/packages/ProxSDP/index.html
index ddf5abae723..90f092619fa 100644
--- a/previews/PR3562/packages/ProxSDP/index.html
+++ b/previews/PR3562/packages/ProxSDP/index.html
@@ -56,4 +56,4 @@
   publisher = {Taylor &amp; Francis},
   doi = {10.1080/02331934.2020.1823387},
   URL = {https://doi.org/10.1080/02331934.2020.1823387}
-}</code></pre><p>The preprint version of the paper can be found <a href="https://arxiv.org/abs/1810.05231">here</a>.</p><h2 id="Disclaimer"><a class="docs-heading-anchor" href="#Disclaimer">Disclaimer</a><a id="Disclaimer-1"></a><a class="docs-heading-anchor-permalink" href="#Disclaimer" title="Permalink"></a></h2><ul><li>ProxSDP is a research software, therefore it should not be used in production.</li><li>Please open an issue if you find any problems, developers will try to fix and find alternatives.</li><li>There is no continuous development for 32-bit systems, the package should work, but might reach some issues.</li><li>ProxSDP assumes primal and dual feasibility.</li></ul><h2 id="ROAD-MAP"><a class="docs-heading-anchor" href="#ROAD-MAP">ROAD MAP</a><a id="ROAD-MAP-1"></a><a class="docs-heading-anchor-permalink" href="#ROAD-MAP" title="Permalink"></a></h2><ul><li>Support for exponential and power cones</li><li>Warm start</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Juniper/">« lanl-ansi/Juniper.jl</a><a class="docs-footer-nextpage" href="../NEOSServer/">odow/NEOSServer.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre><p>The preprint version of the paper can be found <a href="https://arxiv.org/abs/1810.05231">here</a>.</p><h2 id="Disclaimer"><a class="docs-heading-anchor" href="#Disclaimer">Disclaimer</a><a id="Disclaimer-1"></a><a class="docs-heading-anchor-permalink" href="#Disclaimer" title="Permalink"></a></h2><ul><li>ProxSDP is a research software, therefore it should not be used in production.</li><li>Please open an issue if you find any problems, developers will try to fix and find alternatives.</li><li>There is no continuous development for 32-bit systems, the package should work, but might reach some issues.</li><li>ProxSDP assumes primal and dual feasibility.</li></ul><h2 id="ROAD-MAP"><a class="docs-heading-anchor" href="#ROAD-MAP">ROAD MAP</a><a id="ROAD-MAP-1"></a><a class="docs-heading-anchor-permalink" href="#ROAD-MAP" title="Permalink"></a></h2><ul><li>Support for exponential and power cones</li><li>Warm start</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Juniper/">« lanl-ansi/Juniper.jl</a><a class="docs-footer-nextpage" href="../NEOSServer/">odow/NEOSServer.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SCIP/index.html b/previews/PR3562/packages/SCIP/index.html
index f4732b25be9..969e1884277 100644
--- a/previews/PR3562/packages/SCIP/index.html
+++ b/previews/PR3562/packages/SCIP/index.html
@@ -10,4 +10,4 @@
 Pkg.build(&quot;SCIP&quot;)</code></pre><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>Use SCIP with JuMP as follows:</p><pre><code class="language-julia hljs">using JuMP, SCIP
 model = Model(SCIP.Optimizer)
 set_attribute(model, &quot;display/verblevel&quot;, 0)
-set_attribute(model, &quot;limits/gap&quot;, 0.05)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>See the <a href="https://scip.zib.de/doc-8.0.0/html/PARAMETERS.php">SCIP documentation</a> for a list of supported options.</p><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The SCIP optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS1"><code>MOI.SOS1{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS2"><code>MOI.SOS2{Float64}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li><li><a href="../../moi/reference/callbacks/#MathOptInterface.UserCutCallback"><code>MOI.UserCutCallback()</code></a></li></ul><h2 id="Design-considerations"><a class="docs-heading-anchor" href="#Design-considerations">Design considerations</a><a id="Design-considerations-1"></a><a class="docs-heading-anchor-permalink" href="#Design-considerations" title="Permalink"></a></h2><h3 id="Wrapping-the-public-API"><a class="docs-heading-anchor" href="#Wrapping-the-public-API">Wrapping the public API</a><a id="Wrapping-the-public-API-1"></a><a class="docs-heading-anchor-permalink" href="#Wrapping-the-public-API" title="Permalink"></a></h3><p>All of the public API methods are wrapped and available within the <code>SCIP</code> package. This includes the <code>scip_*.h</code> and <code>pub_*.h</code> headers that are collected in <code>scip.h</code>, as well as all default constraint handlers (<code>cons_*.h</code>.) </p><p>The wrapped functions do not transform any data structures and work on the <em>raw</em> pointers (for example, <code>SCIP*</code> in C, <code>Ptr{SCIP_}</code> in Julia). Convenience wrapper functions based on Julia types are added as needed.</p><h3 id="Memory-management"><a class="docs-heading-anchor" href="#Memory-management">Memory management</a><a id="Memory-management-1"></a><a class="docs-heading-anchor-permalink" href="#Memory-management" title="Permalink"></a></h3><p>Programming with SCIP requires dealing with variable and constraint objects that use <a href="https://scip.zib.de/doc-8.0.0/html/OBJ.php">reference counting</a> for memory management.</p><p>The <code>SCIP.Optimizer</code> wrapper type collects lists of <code>SCIP_VAR*</code> and <code>SCIP_CONS*</code> under the hood, and it releases all references when it is garbage collected itself (via <code>finalize</code>).</p><p>When adding a variable (<code>add_variable</code>) or a constraint (<code>add_linear_constraint</code>), an integer index is returned. This index can be used to retrieve the <code>SCIP_VAR*</code> or <code>SCIP_CONS*</code> pointer via <code>get_var</code> and <code>get_cons</code> respectively.</p><h3 id="Supported-nonlinear-operators"><a class="docs-heading-anchor" href="#Supported-nonlinear-operators">Supported nonlinear operators</a><a id="Supported-nonlinear-operators-1"></a><a class="docs-heading-anchor-permalink" href="#Supported-nonlinear-operators" title="Permalink"></a></h3><p>Supported operators in nonlinear expressions are as follows:</p><ul><li><code>+</code></li><li><code>-</code></li><li><code>*</code></li><li><code>/</code></li><li><code>^</code></li><li><code>sqrt</code></li><li><code>exp</code></li><li><code>log</code></li><li><code>abs</code></li><li><code>cos</code></li><li><code>sin</code></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Clarabel/">« oxfordcontrol/Clarabel.jl</a><a class="docs-footer-nextpage" href="../../extensions/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;limits/gap&quot;, 0.05)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>See the <a href="https://scip.zib.de/doc-8.0.0/html/PARAMETERS.php">SCIP documentation</a> for a list of supported options.</p><h2 id="MathOptInterface-API"><a class="docs-heading-anchor" href="#MathOptInterface-API">MathOptInterface API</a><a id="MathOptInterface-API-1"></a><a class="docs-heading-anchor-permalink" href="#MathOptInterface-API" title="Permalink"></a></h2><p>The SCIP optimizer supports the following constraints and attributes.</p><p>List of supported objective functions:</p><ul><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveFunction"><code>MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}</code></a></li></ul><p>List of supported variable types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.Reals"><code>MOI.Reals</code></a></li></ul><p>List of supported constraint types:</p><ul><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction"><code>MOI.ScalarAffineFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction"><code>MOI.ScalarQuadraticFunction{Float64}</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.EqualTo"><code>MOI.EqualTo{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.GreaterThan"><code>MOI.GreaterThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Integer"><code>MOI.Integer</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.Interval"><code>MOI.Interval{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.LessThan"><code>MOI.LessThan{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VariableIndex"><code>MOI.VariableIndex</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.ZeroOne"><code>MOI.ZeroOne</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS1"><code>MOI.SOS1{Float64}</code></a></li><li><a href="../../moi/reference/standard_form/#MathOptInterface.VectorOfVariables"><code>MOI.VectorOfVariables</code></a> in <a href="../../moi/reference/standard_form/#MathOptInterface.SOS2"><code>MOI.SOS2{Float64}</code></a></li></ul><p>List of supported model attributes:</p><ul><li><a href="../../moi/reference/nonlinear/#MathOptInterface.NLPBlock"><code>MOI.NLPBlock()</code></a></li><li><a href="../../moi/reference/models/#MathOptInterface.ObjectiveSense"><code>MOI.ObjectiveSense()</code></a></li><li><a href="../../moi/reference/callbacks/#MathOptInterface.UserCutCallback"><code>MOI.UserCutCallback()</code></a></li></ul><h2 id="Design-considerations"><a class="docs-heading-anchor" href="#Design-considerations">Design considerations</a><a id="Design-considerations-1"></a><a class="docs-heading-anchor-permalink" href="#Design-considerations" title="Permalink"></a></h2><h3 id="Wrapping-the-public-API"><a class="docs-heading-anchor" href="#Wrapping-the-public-API">Wrapping the public API</a><a id="Wrapping-the-public-API-1"></a><a class="docs-heading-anchor-permalink" href="#Wrapping-the-public-API" title="Permalink"></a></h3><p>All of the public API methods are wrapped and available within the <code>SCIP</code> package. This includes the <code>scip_*.h</code> and <code>pub_*.h</code> headers that are collected in <code>scip.h</code>, as well as all default constraint handlers (<code>cons_*.h</code>.) </p><p>The wrapped functions do not transform any data structures and work on the <em>raw</em> pointers (for example, <code>SCIP*</code> in C, <code>Ptr{SCIP_}</code> in Julia). Convenience wrapper functions based on Julia types are added as needed.</p><h3 id="Memory-management"><a class="docs-heading-anchor" href="#Memory-management">Memory management</a><a id="Memory-management-1"></a><a class="docs-heading-anchor-permalink" href="#Memory-management" title="Permalink"></a></h3><p>Programming with SCIP requires dealing with variable and constraint objects that use <a href="https://scip.zib.de/doc-8.0.0/html/OBJ.php">reference counting</a> for memory management.</p><p>The <code>SCIP.Optimizer</code> wrapper type collects lists of <code>SCIP_VAR*</code> and <code>SCIP_CONS*</code> under the hood, and it releases all references when it is garbage collected itself (via <code>finalize</code>).</p><p>When adding a variable (<code>add_variable</code>) or a constraint (<code>add_linear_constraint</code>), an integer index is returned. This index can be used to retrieve the <code>SCIP_VAR*</code> or <code>SCIP_CONS*</code> pointer via <code>get_var</code> and <code>get_cons</code> respectively.</p><h3 id="Supported-nonlinear-operators"><a class="docs-heading-anchor" href="#Supported-nonlinear-operators">Supported nonlinear operators</a><a id="Supported-nonlinear-operators-1"></a><a class="docs-heading-anchor-permalink" href="#Supported-nonlinear-operators" title="Permalink"></a></h3><p>Supported operators in nonlinear expressions are as follows:</p><ul><li><code>+</code></li><li><code>-</code></li><li><code>*</code></li><li><code>/</code></li><li><code>^</code></li><li><code>sqrt</code></li><li><code>exp</code></li><li><code>log</code></li><li><code>abs</code></li><li><code>cos</code></li><li><code>sin</code></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Clarabel/">« oxfordcontrol/Clarabel.jl</a><a class="docs-footer-nextpage" href="../../extensions/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SCS/index.html b/previews/PR3562/packages/SCS/index.html
index 10b083dc729..ddc23ac5298 100644
--- a/previews/PR3562/packages/SCS/index.html
+++ b/previews/PR3562/packages/SCS/index.html
@@ -68,4 +68,4 @@
  SCS.IndirectSolver
  SCS.GpuIndirectSolver</code></pre><p>The <code>GpuIndirectSolver</code> is available on <code>Linux x86_64</code> platform only.</p><h2 id="Low-level-wrapper"><a class="docs-heading-anchor" href="#Low-level-wrapper">Low-level wrapper</a><a id="Low-level-wrapper-1"></a><a class="docs-heading-anchor-permalink" href="#Low-level-wrapper" title="Permalink"></a></h2><p>SCS.jl provides a low-level interface to solve a problem directly, without interfacing through MathOptInterface.</p><p><strong>This is an advanced interface with a risk of incorrect usage. For new users, we recommend that you use the JuMP or Convex interfaces instead.</strong></p><p>SCS solves a problem of the form:</p><pre><code class="nohighlight hljs">minimize        1/2 * x&#39; * P * x + c&#39; * x
 subject to      A * x + s = b
-                s in K</code></pre><p>where <code>K</code> is a product cone of:</p><ul><li>zero cone</li><li>positive orthant <code>{ x | x ≥ 0 }</code></li><li>box cone <code>{ (t,x) | t*l ≤ x ≤ t*u}</code></li><li>second-order cone (SOC) <code>{ (t,x) | ||x||_2 ≤ t }</code></li><li>semi-definite cone (SDC) <code>{ X | X is psd }</code></li><li>exponential cone <code>{ (x,y,z) | y e^(x/y) ≤ z, y&gt;0 }</code></li><li>power cone <code>{ (x,y,z) | x^a * y^(1-a) ≥ |z|, x ≥ 0, y ≥ 0 }</code></li><li>dual power cone <code>{ (u,v,w) | (u/a)^a * (v/(1-a))^(1-a) ≥ |w|, u ≥ 0, v ≥ 0 }</code>.</li></ul><p>To solve this problem with SCS, call <code>SCS.scs_solve</code>; see the docstring for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Pavito/">« jump-dev/Pavito.jl</a><a class="docs-footer-nextpage" href="../SDPA/">jump-dev/SDPA.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+                s in K</code></pre><p>where <code>K</code> is a product cone of:</p><ul><li>zero cone</li><li>positive orthant <code>{ x | x ≥ 0 }</code></li><li>box cone <code>{ (t,x) | t*l ≤ x ≤ t*u}</code></li><li>second-order cone (SOC) <code>{ (t,x) | ||x||_2 ≤ t }</code></li><li>semi-definite cone (SDC) <code>{ X | X is psd }</code></li><li>exponential cone <code>{ (x,y,z) | y e^(x/y) ≤ z, y&gt;0 }</code></li><li>power cone <code>{ (x,y,z) | x^a * y^(1-a) ≥ |z|, x ≥ 0, y ≥ 0 }</code></li><li>dual power cone <code>{ (u,v,w) | (u/a)^a * (v/(1-a))^(1-a) ≥ |w|, u ≥ 0, v ≥ 0 }</code>.</li></ul><p>To solve this problem with SCS, call <code>SCS.scs_solve</code>; see the docstring for details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../Pavito/">« jump-dev/Pavito.jl</a><a class="docs-footer-nextpage" href="../SDPA/">jump-dev/SDPA.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SDDP/index.html b/previews/PR3562/packages/SDDP/index.html
index adae3a866ea..0adcfb3f655 100644
--- a/previews/PR3562/packages/SDDP/index.html
+++ b/previews/PR3562/packages/SDDP/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox" checked/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li class="is-active"><a class="tocitem" href>odow/SDDP.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li><li><a class="tocitem" href="#Help"><span>Help</span></a></li></ul></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Extensions</a></li><li class="is-active"><a href>odow/SDDP.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>odow/SDDP.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/odow/SDDP.jl/blob/v1.6.6/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><img src="https://raw.githubusercontent.com/odow/SDDP.jl/e9de84e0a4b57374bd9e0c95148da1501816e4c5/docs/src/assets/logo.png?raw=true" alt="logo" width="100px"/><h1 id="SDDP.jl"><a class="docs-heading-anchor" href="#SDDP.jl">SDDP.jl</a><a id="SDDP.jl-1"></a><a class="docs-heading-anchor-permalink" href="#SDDP.jl" title="Permalink"></a></h1><p><a href="https://github.com/odow/SDDP.jl/actions?query=workflow%3ACI"><img src="https://github.com/odow/SDDP.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/odow/SDDP.jl"><img src="https://codecov.io/gh/odow/SDDP.jl/branch/master/graph/badge.svg" alt="codecov"/></a></p><p><a href="https://github.com/odow/SDDP.jl">SDDP.jl</a> is a JuMP extension for solving large convex multistage stochastic programming problems using stochastic dual dynamic programming.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>SDDP.jl</code> is licensed under the <a href="https://github.com/odow/SDDP.jl/blob/master/LICENSE.md">MPL 2.0 license</a>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>You can find the documentation at <a href="https://sddp.dev">sddp.dev</a>.</p><h2 id="Help"><a class="docs-heading-anchor" href="#Help">Help</a><a id="Help-1"></a><a class="docs-heading-anchor-permalink" href="#Help" title="Permalink"></a></h2><p>If you need help, please <a href="https://github.com/odow/SDDP.jl/issues/new">open a GitHub issue</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../BilevelJuMP/">« joaquimg/BilevelJuMP.jl</a><a class="docs-footer-nextpage" href="../Plasmo/">plasmo-dev/Plasmo.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox" checked/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li class="is-active"><a class="tocitem" href>odow/SDDP.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li><li><a class="tocitem" href="#Help"><span>Help</span></a></li></ul></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Extensions</a></li><li class="is-active"><a href>odow/SDDP.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>odow/SDDP.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/odow/SDDP.jl/blob/v1.6.6/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><img src="https://raw.githubusercontent.com/odow/SDDP.jl/e9de84e0a4b57374bd9e0c95148da1501816e4c5/docs/src/assets/logo.png?raw=true" alt="logo" width="100px"/><h1 id="SDDP.jl"><a class="docs-heading-anchor" href="#SDDP.jl">SDDP.jl</a><a id="SDDP.jl-1"></a><a class="docs-heading-anchor-permalink" href="#SDDP.jl" title="Permalink"></a></h1><p><a href="https://github.com/odow/SDDP.jl/actions?query=workflow%3ACI"><img src="https://github.com/odow/SDDP.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/odow/SDDP.jl"><img src="https://codecov.io/gh/odow/SDDP.jl/branch/master/graph/badge.svg" alt="codecov"/></a></p><p><a href="https://github.com/odow/SDDP.jl">SDDP.jl</a> is a JuMP extension for solving large convex multistage stochastic programming problems using stochastic dual dynamic programming.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>SDDP.jl</code> is licensed under the <a href="https://github.com/odow/SDDP.jl/blob/master/LICENSE.md">MPL 2.0 license</a>.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>You can find the documentation at <a href="https://sddp.dev">sddp.dev</a>.</p><h2 id="Help"><a class="docs-heading-anchor" href="#Help">Help</a><a id="Help-1"></a><a class="docs-heading-anchor-permalink" href="#Help" title="Permalink"></a></h2><p>If you need help, please <a href="https://github.com/odow/SDDP.jl/issues/new">open a GitHub issue</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../BilevelJuMP/">« joaquimg/BilevelJuMP.jl</a><a class="docs-footer-nextpage" href="../Plasmo/">plasmo-dev/Plasmo.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SDPA/index.html b/previews/PR3562/packages/SDPA/index.html
index 5cdec2aa6a7..0f8b3c66447 100644
--- a/previews/PR3562/packages/SDPA/index.html
+++ b/previews/PR3562/packages/SDPA/index.html
@@ -12,4 +12,4 @@
 set_attribute(model, &quot;Mode&quot;, SDPA.PARAMETER_STABLE_BUT_SLOW)</code></pre><p>Note that the parameters are set in the order they are given, so you can set a mode and then modify parameters from this mode.</p><pre><code class="language-julia hljs">using JuMP, SDPA
 model = Model(SDPA.Optimizer)
 set_attribute(model, &quot;Mode&quot;, SDPA.PARAMETER_STABLE_BUT_SLOW)
-set_attribute(model, &quot;MaxIteration&quot;, 100)</code></pre><p>The choice of parameter mode has a large impact on the performance and stability of SDPA, and not necessarily in the way implied by the names of the modes; for example, <code>PARAMETER_UNSTABLE_BUT_FAST</code> can be more stable than the other modes for some problems. You should try each mode to see how it performs on your specific problem. See <a href="https://github.com/jump-dev/SDPA.jl/issues/17">SDPA.jl#17</a> for more details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SCS/">« jump-dev/SCS.jl</a><a class="docs-footer-nextpage" href="../SDPNAL/">jump-dev/SDPNAL.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_attribute(model, &quot;MaxIteration&quot;, 100)</code></pre><p>The choice of parameter mode has a large impact on the performance and stability of SDPA, and not necessarily in the way implied by the names of the modes; for example, <code>PARAMETER_UNSTABLE_BUT_FAST</code> can be more stable than the other modes for some problems. You should try each mode to see how it performs on your specific problem. See <a href="https://github.com/jump-dev/SDPA.jl/issues/17">SDPA.jl#17</a> for more details.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SCS/">« jump-dev/SCS.jl</a><a class="docs-footer-nextpage" href="../SDPNAL/">jump-dev/SDPNAL.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SDPNAL/index.html b/previews/PR3562/packages/SDPNAL/index.html
index d38a25996b3..2966f1b15fc 100644
--- a/previews/PR3562/packages/SDPNAL/index.html
+++ b/previews/PR3562/packages/SDPNAL/index.html
@@ -18,4 +18,4 @@
 &#39;/path/to/SDPNALv1.0/solver:&#39;, ...
 &#39;/path/to/SDPNALv1.0/solver_main_default:&#39;, ...
 &#39;/path/to/SDPNALv1.0/util:&#39;, ...
-% (...)</code></pre><p>If you have <a href="https://github.com/jump-dev/SDPT3.jl">SDPT3</a> in addition to SDPNAL in the MATLAB path (that is, the <code>toolbox/local/pathdef.m</code> file) then you might have issues because both solvers define a <code>validate</code> function, and this might make SDPNAL call SDPT3&#39;s <code>validate</code> function instead of SDPT3&#39;s <code>validate</code> function.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPA/">« jump-dev/SDPA.jl</a><a class="docs-footer-nextpage" href="../SDPT3/">jump-dev/SDPT3.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+% (...)</code></pre><p>If you have <a href="https://github.com/jump-dev/SDPT3.jl">SDPT3</a> in addition to SDPNAL in the MATLAB path (that is, the <code>toolbox/local/pathdef.m</code> file) then you might have issues because both solvers define a <code>validate</code> function, and this might make SDPNAL call SDPT3&#39;s <code>validate</code> function instead of SDPT3&#39;s <code>validate</code> function.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPA/">« jump-dev/SDPA.jl</a><a class="docs-footer-nextpage" href="../SDPT3/">jump-dev/SDPT3.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SDPT3/index.html b/previews/PR3562/packages/SDPT3/index.html
index 43ccc1878db..0a47ec1e3d2 100644
--- a/previews/PR3562/packages/SDPT3/index.html
+++ b/previews/PR3562/packages/SDPT3/index.html
@@ -29,4 +29,4 @@
 
 julia&gt; MATLAB.restoredefaultpath()
 
-julia&gt; MATLAB.mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPNAL/">« jump-dev/SDPNAL.jl</a><a class="docs-footer-nextpage" href="../SeDuMi/">jump-dev/SeDuMi.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+julia&gt; MATLAB.mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPNAL/">« jump-dev/SDPNAL.jl</a><a class="docs-footer-nextpage" href="../SeDuMi/">jump-dev/SeDuMi.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SeDuMi/index.html b/previews/PR3562/packages/SeDuMi/index.html
index 835a11969dd..953c742af3d 100644
--- a/previews/PR3562/packages/SeDuMi/index.html
+++ b/previews/PR3562/packages/SeDuMi/index.html
@@ -17,4 +17,4 @@
            MATLAB.mat&quot;install_sedumi&quot;
        end
 
-julia&gt; MATLAB.mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPT3/">« jump-dev/SDPT3.jl</a><a class="docs-footer-nextpage" href="../Xpress/">jump-dev/Xpress.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+julia&gt; MATLAB.mat&quot;savepath&quot;</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SDPT3/">« jump-dev/SDPT3.jl</a><a class="docs-footer-nextpage" href="../Xpress/">jump-dev/Xpress.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/SumOfSquares/index.html b/previews/PR3562/packages/SumOfSquares/index.html
index 72ab870f5f9..f8921769c4e 100644
--- a/previews/PR3562/packages/SumOfSquares/index.html
+++ b/previews/PR3562/packages/SumOfSquares/index.html
@@ -4,4 +4,4 @@
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
 </script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../solvers/">Introduction</a></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox" checked/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li class="is-active"><a class="tocitem" href>jump-dev/SumOfSquares.jl</a><ul class="internal"><li><a class="tocitem" href="#License"><span>License</span></a></li><li><a class="tocitem" href="#Installation"><span>Installation</span></a></li><li><a class="tocitem" href="#Documentation"><span>Documentation</span></a></li><li><a class="tocitem" href="#Presentations"><span>Presentations</span></a></li><li><a class="tocitem" href="#Citing"><span>Citing</span></a></li></ul></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Extensions</a></li><li class="is-active"><a href>jump-dev/SumOfSquares.jl</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>jump-dev/SumOfSquares.jl</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/SumOfSquares.jl/blob/v0.7.3/README.md" title="View source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="SumOfSquares.jl"><a class="docs-heading-anchor" href="#SumOfSquares.jl">SumOfSquares.jl</a><a id="SumOfSquares.jl-1"></a><a class="docs-heading-anchor-permalink" href="#SumOfSquares.jl" title="Permalink"></a></h1><p><a href="https://github.com/jump-dev/SumOfSquares.jl/actions?query=workflow%3ACI"><img src="https://github.com/jump-dev/SumOfSquares.jl/workflows/CI/badge.svg?branch=master" alt="Build Status"/></a> <a href="https://codecov.io/gh/jump-dev/SumOfSquares.jl"><img src="https://codecov.io/gh/jump-dev/SumOfSquares.jl/branch/master/graph/badge.svg" alt="codecov"/></a> <a href="https://doi.org/10.5281/zenodo.1208672"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.1208672.svg" alt/></a></p><p><a href="https://github.com/jump-dev/SumOfSquares.jl">SumOfSquares.jl</a> is a JuMP extension that, when used in conjunction with <a href="https://github.com/JuliaAlgebra/MultivariatePolynomials.jl">MultivariatePolynomial</a> and <a href="https://github.com/jump-dev/PolyJuMP.jl">PolyJuMP</a>, implements a sum of squares reformulation for polynomial optimization.</p><h2 id="License"><a class="docs-heading-anchor" href="#License">License</a><a id="License-1"></a><a class="docs-heading-anchor-permalink" href="#License" title="Permalink"></a></h2><p><code>SumOfSquares.jl</code> is licensed under the <a href="https://github.com/jump-dev/SumOfSquares.jl/blob/master/LICENSE.md">MIT license</a>.</p><h2 id="Installation"><a class="docs-heading-anchor" href="#Installation">Installation</a><a id="Installation-1"></a><a class="docs-heading-anchor-permalink" href="#Installation" title="Permalink"></a></h2><p>Install <code>SumOfSquares</code> using <code>Pkg.add</code>:</p><pre><code class="language-julia hljs">import Pkg
-Pkg.add(&quot;SumOfSquares&quot;)</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>See <a href="https://jump.dev/SumOfSquares.jl/stable">https://jump.dev/SumOfSquares.jl/stable</a> for the most recently tagged version of the documentation.</p><p>See <a href="https://jump.dev/SumOfSquares.jl/dev">https://jump.dev/SumOfSquares.jl/dev</a> for the in-development version of the documentation.</p><h2 id="Presentations"><a class="docs-heading-anchor" href="#Presentations">Presentations</a><a id="Presentations-1"></a><a class="docs-heading-anchor-permalink" href="#Presentations" title="Permalink"></a></h2><p>Some presentations on, or using, SumOfSquares (see <a href="https://github.com/blegat/SumOfSquaresSlides">blegat/SumOfSquaresSlides</a> for the source code of the presentations):</p><ul><li>Benoît Legat at <a href="https://pretalx.com/juliacon2023/talk/XLT8H3/">JuMP-dev 2023</a> [<a href="https://drive.google.com/file/d/1H-_Ot7tP2g7t95r0K_mTaXvGO1ChUPci/view?usp=drive_link">Slides</a>]</li><li>Benoît Legat, Marek Kaluba and Tillmann Weisser at INFORMS 2022 [<a href="https://drive.google.com/file/d/1rlsIxgcnKWT436k4MNenjHfgH0UYRLAB/view?usp=share_link">Slides</a>]</li><li>Benoît Legat at <a href="http://poema-network.eu/index.php/news-and-events/project-workshops/13-poema-learning-week-2">POEMA Learning Week 2</a></li><li>Benoît Legat, Marek Kaluba and Tillmann Weisser at <a href="https://pretalx.com/juliacon2021/talk/L8DTE3/">JuMP-dev 2021</a> [<a href="https://drive.google.com/file/d/1HtArDFNMQ6IYUqRjSWR3JviJp9xLtSlB/view?usp=sharing">Slides</a>]</li><li>Benoît Legat at <a href="http://meetings2.informs.org/wordpress/annual2020/">INFORMS 2020</a> [<a href="https://drive.google.com/file/d/1lb8NtOWCikTYm6KRUZCSLYgaUjqIsSyV/view?usp=sharing">Slides</a>]</li><li>Tillmann Weisser, Benoît Legat, Chris Coey, Lea Kapelevich and Juan Pablo Vielma at <a href="https://juliacon.org/2019/">JuliaCon 2019</a> [<a href="https://drive.google.com/open?id=1HiA-praFyejE0Z3nVSpFEv938TAcPjA9">Slides</a>] [<a href="https://www.youtube.com/watch?v=cTmqmPcroFo">Video</a>]</li><li>Benoît Legat at <a href="https://cnls.lanl.gov/External/showtalksummary.php?selection=7768">CNLS 2019</a> [<a href="https://drive.google.com/open?id=1kNF18C7RY2zi7jcZBMO1PRXtHuvVTFPn">Slides</a>]</li><li>Benoît Legat at <a href="https://www.euro2019dublin.com/">EURO 2019</a> [<a href="https://drive.google.com/open?id=1Wry56NzzL4QBRSwuhP4AlKOe2i2FL7dk">Slides</a>]</li><li>Benoît Legat at <a href="https://julialang.univ-nantes.fr/programme/">juliaday Nantes 2019</a> [<a href="https://drive.google.com/open?id=1pN3G9Pr8jbzK9EEaJ9a6p_qKwSbxb2bo">Slides</a>]</li><li>Benoît Legat at <a href="https://www.mis.mpg.de/calendar/conferences/2018/nc2018.html">Summer School on Numerical Computing in Algebraic Geometry 2018</a> [<a href="https://drive.google.com/open?id=1pf9rdoVEjAnD164rptLki1AG0AH4i88M">Poster</a>]</li><li>Benoît Legat at <a href="https://jump.dev/meetings/mit2017/">The First Annual JuMP-dev Workshop 2017</a> [<a href="https://drive.google.com/file/d/1ea5eSMvMB3jXPuljzNGmMKied-n50YIo/view?usp=sharing">Slides</a>] [<a href="https://youtu.be/kyo72yWYr54">Video</a>]</li><li><a href="https://docs.google.com/presentation/d/1ASfjB1TdLJmYxT0b6rnyGh9eLbMc-66bTOt3_3yvc90/edit?usp=sharing">Joey Huchette at SIAM Opt 2017</a></li></ul><h2 id="Citing"><a class="docs-heading-anchor" href="#Citing">Citing</a><a id="Citing-1"></a><a class="docs-heading-anchor-permalink" href="#Citing" title="Permalink"></a></h2><p>See <a href="https://github.com/jump-dev/SumOfSquares.jl/blob/master/CITATION.bib">CITATION.bib</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../PolyJuMP/">« jump-dev/PolyJuMP.jl</a><a class="docs-footer-nextpage" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Pkg.add(&quot;SumOfSquares&quot;)</code></pre><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>See <a href="https://jump.dev/SumOfSquares.jl/stable">https://jump.dev/SumOfSquares.jl/stable</a> for the most recently tagged version of the documentation.</p><p>See <a href="https://jump.dev/SumOfSquares.jl/dev">https://jump.dev/SumOfSquares.jl/dev</a> for the in-development version of the documentation.</p><h2 id="Presentations"><a class="docs-heading-anchor" href="#Presentations">Presentations</a><a id="Presentations-1"></a><a class="docs-heading-anchor-permalink" href="#Presentations" title="Permalink"></a></h2><p>Some presentations on, or using, SumOfSquares (see <a href="https://github.com/blegat/SumOfSquaresSlides">blegat/SumOfSquaresSlides</a> for the source code of the presentations):</p><ul><li>Benoît Legat at <a href="https://pretalx.com/juliacon2023/talk/XLT8H3/">JuMP-dev 2023</a> [<a href="https://drive.google.com/file/d/1H-_Ot7tP2g7t95r0K_mTaXvGO1ChUPci/view?usp=drive_link">Slides</a>]</li><li>Benoît Legat, Marek Kaluba and Tillmann Weisser at INFORMS 2022 [<a href="https://drive.google.com/file/d/1rlsIxgcnKWT436k4MNenjHfgH0UYRLAB/view?usp=share_link">Slides</a>]</li><li>Benoît Legat at <a href="http://poema-network.eu/index.php/news-and-events/project-workshops/13-poema-learning-week-2">POEMA Learning Week 2</a></li><li>Benoît Legat, Marek Kaluba and Tillmann Weisser at <a href="https://pretalx.com/juliacon2021/talk/L8DTE3/">JuMP-dev 2021</a> [<a href="https://drive.google.com/file/d/1HtArDFNMQ6IYUqRjSWR3JviJp9xLtSlB/view?usp=sharing">Slides</a>]</li><li>Benoît Legat at <a href="http://meetings2.informs.org/wordpress/annual2020/">INFORMS 2020</a> [<a href="https://drive.google.com/file/d/1lb8NtOWCikTYm6KRUZCSLYgaUjqIsSyV/view?usp=sharing">Slides</a>]</li><li>Tillmann Weisser, Benoît Legat, Chris Coey, Lea Kapelevich and Juan Pablo Vielma at <a href="https://juliacon.org/2019/">JuliaCon 2019</a> [<a href="https://drive.google.com/open?id=1HiA-praFyejE0Z3nVSpFEv938TAcPjA9">Slides</a>] [<a href="https://www.youtube.com/watch?v=cTmqmPcroFo">Video</a>]</li><li>Benoît Legat at <a href="https://cnls.lanl.gov/External/showtalksummary.php?selection=7768">CNLS 2019</a> [<a href="https://drive.google.com/open?id=1kNF18C7RY2zi7jcZBMO1PRXtHuvVTFPn">Slides</a>]</li><li>Benoît Legat at <a href="https://www.euro2019dublin.com/">EURO 2019</a> [<a href="https://drive.google.com/open?id=1Wry56NzzL4QBRSwuhP4AlKOe2i2FL7dk">Slides</a>]</li><li>Benoît Legat at <a href="https://julialang.univ-nantes.fr/programme/">juliaday Nantes 2019</a> [<a href="https://drive.google.com/open?id=1pN3G9Pr8jbzK9EEaJ9a6p_qKwSbxb2bo">Slides</a>]</li><li>Benoît Legat at <a href="https://www.mis.mpg.de/calendar/conferences/2018/nc2018.html">Summer School on Numerical Computing in Algebraic Geometry 2018</a> [<a href="https://drive.google.com/open?id=1pf9rdoVEjAnD164rptLki1AG0AH4i88M">Poster</a>]</li><li>Benoît Legat at <a href="https://jump.dev/meetings/mit2017/">The First Annual JuMP-dev Workshop 2017</a> [<a href="https://drive.google.com/file/d/1ea5eSMvMB3jXPuljzNGmMKied-n50YIo/view?usp=sharing">Slides</a>] [<a href="https://youtu.be/kyo72yWYr54">Video</a>]</li><li><a href="https://docs.google.com/presentation/d/1ASfjB1TdLJmYxT0b6rnyGh9eLbMc-66bTOt3_3yvc90/edit?usp=sharing">Joey Huchette at SIAM Opt 2017</a></li></ul><h2 id="Citing"><a class="docs-heading-anchor" href="#Citing">Citing</a><a id="Citing-1"></a><a class="docs-heading-anchor-permalink" href="#Citing" title="Permalink"></a></h2><p>See <a href="https://github.com/jump-dev/SumOfSquares.jl/blob/master/CITATION.bib">CITATION.bib</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../PolyJuMP/">« jump-dev/PolyJuMP.jl</a><a class="docs-footer-nextpage" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Tulip/index.html b/previews/PR3562/packages/Tulip/index.html
index 3115a8a807e..1b14197da2c 100644
--- a/previews/PR3562/packages/Tulip/index.html
+++ b/previews/PR3562/packages/Tulip/index.html
@@ -28,4 +28,4 @@
   language = {en},
   url      = {https://doi.org/10.1007/s12532-020-00200-8},
   urldate  = {2021-03-07},
-}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../DAQP/">« darnstrom/DAQP.jl</a><a class="docs-footer-nextpage" href="../Loraine/">kocvara/Loraine.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+}</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../DAQP/">« darnstrom/DAQP.jl</a><a class="docs-footer-nextpage" href="../Loraine/">kocvara/Loraine.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/Xpress/index.html b/previews/PR3562/packages/Xpress/index.html
index 4c616181f86..9a8a7b773be 100644
--- a/previews/PR3562/packages/Xpress/index.html
+++ b/previews/PR3562/packages/Xpress/index.html
@@ -8,4 +8,4 @@
 import Pkg
 Pkg.add(&quot;Xpress&quot;)</code></pre><p>By default, building Xpress.jl will fail if the Xpress library is not found. This may not be desirable in certain cases, for example when part of a package&#39;s test suite uses Xpress as an optional test dependency, but Xpress cannot be installed on a CI server running the test suite. To support this use case, the <code>XPRESS_JL_SKIP_LIB_CHECK</code> environment variable may be set (to any value) to make Xpress.jl installable (but not usable).</p><h2 id="Use-with-JuMP"><a class="docs-heading-anchor" href="#Use-with-JuMP">Use with JuMP</a><a id="Use-with-JuMP-1"></a><a class="docs-heading-anchor-permalink" href="#Use-with-JuMP" title="Permalink"></a></h2><p>To use Xpress with JuMP, use:</p><pre><code class="language-julia hljs">using JuMP, Xpress
 model = Model(Xpress.Optimizer)
-set_optimizer(model, &quot;PRESOLVE&quot;, 0)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>For other parameters use <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML/">Xpress Optimizer manual</a> or type <code>julia -e &quot;using Xpress; println(keys(Xpress.XPRS_ATTRIBUTES))&quot;</code>.</p><p>If <code>logfile</code> is set to <code>&quot;&quot;</code>, the log file is disabled and output is printed to the console (<a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML/OUTPUTLOG.html">there might be issues with console output on windows (it is manually implemented with callbacks)</a>). If <code>logfile</code> is set to a file&#39;s path, output is printed to that file. By default, <code>logfile = &quot;&quot;</code> (console).</p><h2 id="Callbacks"><a class="docs-heading-anchor" href="#Callbacks">Callbacks</a><a id="Callbacks-1"></a><a class="docs-heading-anchor-permalink" href="#Callbacks" title="Permalink"></a></h2><p>Solver specific and solver independent callbacks are working in <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a> and, consequently, in <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a>. However, the current implementation should be considered experimental.</p><h2 id="Environment-variables"><a class="docs-heading-anchor" href="#Environment-variables">Environment variables</a><a id="Environment-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Environment-variables" title="Permalink"></a></h2><ul><li><code>XPRESS_JL_SKIP_LIB_CHECK</code>: Used to skip build lib check as previously described.</li><li><code>XPRESS_JL_NO_INFO</code>: Disable license info log.</li><li><code>XPRESS_JL_NO_DEPS_ERROR</code>: Disable error when do deps.jl file is found.</li><li><code>XPRESS_JL_NO_AUTO_INIT</code>: Disable automatic run of <code>Xpress.initialize()</code>.</li></ul><p>Specially useful for explicitly loading the dynamic library.</p><h2 id="Skipping-Xpress.postsolve"><a class="docs-heading-anchor" href="#Skipping-Xpress.postsolve">Skipping Xpress.postsolve</a><a id="Skipping-Xpress.postsolve-1"></a><a class="docs-heading-anchor-permalink" href="#Skipping-Xpress.postsolve" title="Permalink"></a></h2><p>In older versions of Xpress, the command <code>XPRSpostsolve</code> throws an error in infeasible models. In these older versions the post solve should not be executed. To do this, one can use the <code>MOI.RawOptimizerAttribute(&quot;MOI_POST_SOLVE&quot;)</code> to skip this routine.</p><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>Xpress.Lib.XPRSxx</code> functions, where the names and arguments are identical to the C API. </p><p>See the <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML">Xpress documentation</a> for details.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>For more information, consult the <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML">FICO optimizer manual</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SeDuMi/">« jump-dev/SeDuMi.jl</a><a class="docs-footer-nextpage" href="../COPT/">COPT-Public/COPT.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+set_optimizer(model, &quot;PRESOLVE&quot;, 0)</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>For other parameters use <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML/">Xpress Optimizer manual</a> or type <code>julia -e &quot;using Xpress; println(keys(Xpress.XPRS_ATTRIBUTES))&quot;</code>.</p><p>If <code>logfile</code> is set to <code>&quot;&quot;</code>, the log file is disabled and output is printed to the console (<a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML/OUTPUTLOG.html">there might be issues with console output on windows (it is manually implemented with callbacks)</a>). If <code>logfile</code> is set to a file&#39;s path, output is printed to that file. By default, <code>logfile = &quot;&quot;</code> (console).</p><h2 id="Callbacks"><a class="docs-heading-anchor" href="#Callbacks">Callbacks</a><a id="Callbacks-1"></a><a class="docs-heading-anchor-permalink" href="#Callbacks" title="Permalink"></a></h2><p>Solver specific and solver independent callbacks are working in <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a> and, consequently, in <a href="https://github.com/jump-dev/JuMP.jl">JuMP</a>. However, the current implementation should be considered experimental.</p><h2 id="Environment-variables"><a class="docs-heading-anchor" href="#Environment-variables">Environment variables</a><a id="Environment-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Environment-variables" title="Permalink"></a></h2><ul><li><code>XPRESS_JL_SKIP_LIB_CHECK</code>: Used to skip build lib check as previously described.</li><li><code>XPRESS_JL_NO_INFO</code>: Disable license info log.</li><li><code>XPRESS_JL_NO_DEPS_ERROR</code>: Disable error when do deps.jl file is found.</li><li><code>XPRESS_JL_NO_AUTO_INIT</code>: Disable automatic run of <code>Xpress.initialize()</code>.</li></ul><p>Specially useful for explicitly loading the dynamic library.</p><h2 id="Skipping-Xpress.postsolve"><a class="docs-heading-anchor" href="#Skipping-Xpress.postsolve">Skipping Xpress.postsolve</a><a id="Skipping-Xpress.postsolve-1"></a><a class="docs-heading-anchor-permalink" href="#Skipping-Xpress.postsolve" title="Permalink"></a></h2><p>In older versions of Xpress, the command <code>XPRSpostsolve</code> throws an error in infeasible models. In these older versions the post solve should not be executed. To do this, one can use the <code>MOI.RawOptimizerAttribute(&quot;MOI_POST_SOLVE&quot;)</code> to skip this routine.</p><h2 id="C-API"><a class="docs-heading-anchor" href="#C-API">C API</a><a id="C-API-1"></a><a class="docs-heading-anchor-permalink" href="#C-API" title="Permalink"></a></h2><p>The C API can be accessed via <code>Xpress.Lib.XPRSxx</code> functions, where the names and arguments are identical to the C API. </p><p>See the <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML">Xpress documentation</a> for details.</p><h2 id="Documentation"><a class="docs-heading-anchor" href="#Documentation">Documentation</a><a id="Documentation-1"></a><a class="docs-heading-anchor-permalink" href="#Documentation" title="Permalink"></a></h2><p>For more information, consult the <a href="https://www.fico.com/fico-xpress-optimization/docs/latest/solver/optimizer/HTML">FICO optimizer manual</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../SeDuMi/">« jump-dev/SeDuMi.jl</a><a class="docs-footer-nextpage" href="../COPT/">COPT-Public/COPT.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/packages/solvers/index.html b/previews/PR3562/packages/solvers/index.html
index 9f9a6603bfc..1d844bc221a 100644
--- a/previews/PR3562/packages/solvers/index.html
+++ b/previews/PR3562/packages/solvers/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#Adding-new-solvers"><span>Adding new solvers</span></a></li></ul></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/packages/solvers.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>This section of the documentation contains brief documentation for some of the solvers that JuMP supports. The list of solvers is not exhaustive, but instead is intended to help you discover commonly used solvers.</p><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>Packages beginning with <code>jump-dev/</code> are developed and maintained by the JuMP developers. In many cases, these packages wrap external solvers that are not developed by the JuMP developers and, while the Julia packages are all open-source, in some cases the solvers themselves are closed source commercial products.</p><p>Packages that do not begin with <code>jump-dev/</code> are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.</p><h2 id="Adding-new-solvers"><a class="docs-heading-anchor" href="#Adding-new-solvers">Adding new solvers</a><a id="Adding-new-solvers-1"></a><a class="docs-heading-anchor-permalink" href="#Adding-new-solvers" title="Permalink"></a></h2><p>Written a solver? Add it to this section of the JuMP documentation by making a pull request to the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/packages.toml"><code>docs/packages.toml</code></a> file.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../developers/checklists/">« Checklists</a><a class="docs-footer-nextpage" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../assets/documenter.js"></script><script src="../../search_index.js"></script><script src="../../siteinfo.js"></script><script src="../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../assets/themeswap.js"></script><link href="../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../"><img class="docs-light-only" src="../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../">Introduction</a></li><li><a class="tocitem" href="../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../manual/models/">Models</a></li><li><a class="tocitem" href="../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox" checked/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#Affiliation"><span>Affiliation</span></a></li><li><a class="tocitem" href="#Adding-new-solvers"><span>Adding new solvers</span></a></li></ul></li><li><a class="tocitem" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/">Introduction</a></li><li><a class="tocitem" href="../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Solvers</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/packages/solvers.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>This section of the documentation contains brief documentation for some of the solvers that JuMP supports. The list of solvers is not exhaustive, but instead is intended to help you discover commonly used solvers.</p><h2 id="Affiliation"><a class="docs-heading-anchor" href="#Affiliation">Affiliation</a><a id="Affiliation-1"></a><a class="docs-heading-anchor-permalink" href="#Affiliation" title="Permalink"></a></h2><p>Packages beginning with <code>jump-dev/</code> are developed and maintained by the JuMP developers. In many cases, these packages wrap external solvers that are not developed by the JuMP developers and, while the Julia packages are all open-source, in some cases the solvers themselves are closed source commercial products.</p><p>Packages that do not begin with <code>jump-dev/</code> are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.</p><h2 id="Adding-new-solvers"><a class="docs-heading-anchor" href="#Adding-new-solvers">Adding new solvers</a><a id="Adding-new-solvers-1"></a><a class="docs-heading-anchor-permalink" href="#Adding-new-solvers" title="Permalink"></a></h2><p>Written a solver? Add it to this section of the JuMP documentation by making a pull request to the <a href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/packages.toml"><code>docs/packages.toml</code></a> file.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../developers/checklists/">« Checklists</a><a class="docs-footer-nextpage" href="../AmplNLWriter/">jump-dev/AmplNLWriter.jl »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/release_notes/index.html b/previews/PR3562/release_notes/index.html
index 07a50ae6ee8..54e67d57520 100644
--- a/previews/PR3562/release_notes/index.html
+++ b/previews/PR3562/release_notes/index.html
@@ -12,4 +12,4 @@
 new_b = backend(model)</code></pre></li><li>All usages of <code>@SDconstraint</code> are deprecated. The new syntax is <code>@constraint(model, X &gt;= Y, PSDCone())</code>.</li><li>Creating a <code>DenseAxisArray</code> with a <code>Number</code> as an axis will now display a warning. This catches a common error in which users write <code>@variable(model, x[length(S)])</code> instead of <code>@variable(model, x[1:length(S)])</code>.</li><li>The <code>caching_mode</code> argument to <code>Model</code>, for example, <code>Model(caching_mode = MOIU.MANUAL)</code> mode has been removed. For more control over the optimizer, use <code>direct_model</code> instead.</li><li>The previously deprecated <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> functions have been removed. Use <code>lp_sensitivity_report</code> instead.</li><li>The <code>.m</code> fields of <code>NonlinearExpression</code> and <code>NonlinearParameter</code> have been renamed to <code>.model</code>.</li><li>Infinite variable bounds are now ignored. Thus, <code>@variable(model, x &lt;= Inf)</code> will show <code>has_upper_bound(x) == false</code>. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.</li><li>The <code>variable_type</code> and <code>constraint_type</code> functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.</li><li>The internal functions <code>moi_mode</code>, <code>moi_bridge_constraints</code>, <code>moi_add_constraint</code>, and <code>moi_add_to_function_constant</code> are no longer exported.</li><li>The un-used method <code>Containers.generate_container</code> has been deleted.</li><li>The <code>Containers</code> API has been refactored, and <code>_build_ref_sets</code> is now public as <code>Containers.build_ref_sets</code>.</li><li>The <code>parse_constraint_</code> methods for extending <code>@constraint</code> at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.</li></ul><h3 id="Added-22"><a class="docs-heading-anchor" href="#Added-22">Added</a><a class="docs-heading-anchor-permalink" href="#Added-22" title="Permalink"></a></h3><ul><li>The <code>TerminationStatusCode</code> and <code>ResultStatusCode</code> enums are now exported by JuMP. Prefer <code>termination_status(model) == OPTIMAL</code> instead of <code>== MOI.OPTIMAL</code>, although the <code>MOI.</code> prefix way still works.</li><li>Copy a <code>x::DenseAxisArray</code> to an <code>Array</code> by calling <code>Array(x)</code>.</li><li><code>NonlinearExpression</code> is now a subtype of <code>AbstractJuMPScalar</code></li><li>Constraints such as <code>@constraint(model, x + 1 in MOI.Integer())</code> are now supported.</li><li><code>primal_feasibility_report</code> now accepts a function as the first argument.</li><li>Scalar variables <code>@variable(model, x[1:2] in MOI.Integer())</code> creates two variables, both of which are constrained to be in the set <code>MOI.Integer</code>.</li><li>Conic constraints can now be specified as inequalities under a different partial ordering. So <code>@constraint(model, x - y in MOI.Nonnegatives())</code> can now be written as <code>@constraint(model, x &gt;= y, MOI.Nonnegatives())</code>.</li><li>Names are now set for vectorized constraints.</li></ul><h3 id="Fixed-30"><a class="docs-heading-anchor" href="#Fixed-30">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-30" title="Permalink"></a></h3><ul><li>Fixed a performance issue when <code>show</code> was called on a <code>SparseAxisArray</code> with a large number of elements.</li><li>Fixed a bug displaying barrier and simplex iterations in <code>solution_summary</code>.</li><li>Fixed a bug by implementing <code>hash</code> for <code>DenseAxisArray</code> and <code>SparseAxisArray</code>.</li><li>Names are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with <code>direct_model</code>.</li><li><code>MutableArithmetics.Zero</code> is converted into a <code>0.0</code> before being returned to the user. Previously, some calls to <code>@expression</code> would return the undocumented <code>MutableArithmetics.Zero()</code> object. One example is summing over an empty set <code>@expression(model, sum(x[i] for i in 1:0))</code>. You will now get <code>0.0</code> instead.</li><li><code>AffExpr</code> and <code>QuadExpr</code> can now be used with <code>== 0</code> instead of <code>iszero</code>. This fixes a number of issues relating to Julia standard libraries such as <code>LinearAlgebra</code> and <code>SparseArrays</code>.</li><li>Fixed a bug when registering a user-defined function with splatting.</li></ul><h3 id="Other-28"><a class="docs-heading-anchor" href="#Other-28">Other</a><a class="docs-heading-anchor-permalink" href="#Other-28" title="Permalink"></a></h3><ul><li>The documentation is now available as a PDF.</li><li>The documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)</li><li>The documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.</li><li>The documentation is now built with Julia 1.6 instead of 1.0.</li><li>Various error messages have been improved to be more readable.</li></ul><h2 id="[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)"><a class="docs-heading-anchor" href="#[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10">Version 0.21.10</a> (September 4, 2021)</a><a id="[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)" title="Permalink"></a></h2><h3 id="Added-23"><a class="docs-heading-anchor" href="#Added-23">Added</a><a class="docs-heading-anchor-permalink" href="#Added-23" title="Permalink"></a></h3><ul><li>Added <code>add_NL_expression</code></li><li><code>add_NL_xxx</code> functions now support <code>AffExpr</code> and <code>QuadExpr</code> as terms</li></ul><h3 id="Fixed-31"><a class="docs-heading-anchor" href="#Fixed-31">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-31" title="Permalink"></a></h3><ul><li>Fixed a bug in <code>solution_summary</code></li><li>Fixed a bug in <code>relax_integrality</code></li></ul><h3 id="Other-29"><a class="docs-heading-anchor" href="#Other-29">Other</a><a class="docs-heading-anchor-permalink" href="#Other-29" title="Permalink"></a></h3><ul><li>Improved error message in <code>lp_sensitivity_report</code></li></ul><h2 id="[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)"><a class="docs-heading-anchor" href="#[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9">Version 0.21.9</a> (August 1, 2021)</a><a id="[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)" title="Permalink"></a></h2><h3 id="Added-24"><a class="docs-heading-anchor" href="#Added-24">Added</a><a class="docs-heading-anchor-permalink" href="#Added-24" title="Permalink"></a></h3><ul><li>Containers now support arbitrary container types by passing the type to the <code>container</code> keyword and overloading <code>Containers.container</code>.</li><li><code>is_valid</code> now supports nonlinear constraints</li><li>Added <code>unsafe_backend</code> for querying the inner-most optimizer of a JuMP model.</li><li>Nonlinear parameters now support the plural <code>@NLparameters</code> macro.</li><li>Containers (for example, <code>DenseAxisArray</code>) can now be used in vector-valued constraints.</li></ul><h3 id="Other-30"><a class="docs-heading-anchor" href="#Other-30">Other</a><a class="docs-heading-anchor-permalink" href="#Other-30" title="Permalink"></a></h3><ul><li>Various improvements to the documentation.</li></ul><h2 id="[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)"><a class="docs-heading-anchor" href="#[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8">Version 0.21.8</a> (May 8, 2021)</a><a id="[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)" title="Permalink"></a></h2><h3 id="Added-25"><a class="docs-heading-anchor" href="#Added-25">Added</a><a class="docs-heading-anchor-permalink" href="#Added-25" title="Permalink"></a></h3><ul><li>The <code>@constraint</code> macro is now extendable in the same way as <code>@variable</code>.</li><li><code>AffExpr</code> and <code>QuadExpr</code> can now be used in nonlinear macros.</li></ul><h3 id="Fixed-32"><a class="docs-heading-anchor" href="#Fixed-32">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-32" title="Permalink"></a></h3><ul><li>Fixed a bug in <code>lp_sensitivity_report</code>.</li><li>Fixed an inference issue when creating empty <code>SparseAxisArray</code>s.</li></ul><h2 id="[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)"><a class="docs-heading-anchor" href="#[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7">Version 0.21.7</a> (April 12, 2021)</a><a id="[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)" title="Permalink"></a></h2><h3 id="Added-26"><a class="docs-heading-anchor" href="#Added-26">Added</a><a class="docs-heading-anchor-permalink" href="#Added-26" title="Permalink"></a></h3><ul><li>Added <code>primal_feasibility_report</code>, which can be used to check whether a primal point satisfies primal feasibility.</li><li>Added <code>coefficient</code>, which returns the coefficient associated with a variable in affine and quadratic expressions.</li><li>Added <code>copy_conflict</code>, which returns the IIS of an infeasible model.</li><li>Added <code>solution_summary</code>, which returns (and prints) a struct containing a summary of the solution.</li><li>Allow <code>AbstractVector</code> in vector constraints instead of just <code>Vector</code>.</li><li>Added <code>latex_formulation(model)</code> which returns an object representing the latex formulation of a model. Use <code>print(latex_formulation(model))</code> to print the formulation as a string.</li><li>User-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.</li><li>DenseAxisArray&#39;s now support broadcasting over multiple arrays.</li><li>Container indices can now be iterators of <code>Base.SizeUnknown</code>.</li></ul><h3 id="Fixed-33"><a class="docs-heading-anchor" href="#Fixed-33">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-33" title="Permalink"></a></h3><ul><li>Fixed bug in <code>rad2deg</code> and <code>deg2rad</code> in nonlinear expressions.</li><li>Fixed a MethodError bug in <code>Containers</code> when forcing container type.</li><li>Allow partial slicing of a DenseAxisArray, resolving an issue from 2014.</li><li>Fixed a bug printing variable names in IJulia.</li><li>Ending an IJulia cell with <code>model</code> now prints a summary of the model (like in the REPL) not the latex formulation. Use <code>print(model)</code> to print the latex formulation.</li><li>Fixed a bug when copying models containing nested arrays.</li></ul><h3 id="Other-31"><a class="docs-heading-anchor" href="#Other-31">Other</a><a class="docs-heading-anchor-permalink" href="#Other-31" title="Permalink"></a></h3><ul><li>Tutorials are now part of the documentation, and more refactoring has taken place.</li><li>Added JuliaFormatter added as a code formatter.</li><li>Added some precompilation statements to reduce initial latency.</li><li>Various improvements to error messages to make them more helpful.</li><li>Improved performance of <code>value(::NonlinearExpression)</code>.</li><li>Improved performance of <code>fix(::VariableRef)</code>.</li></ul><h2 id="[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)"><a class="docs-heading-anchor" href="#[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6">Version 0.21.6</a> (January 29, 2021)</a><a id="[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)" title="Permalink"></a></h2><h3 id="Added-27"><a class="docs-heading-anchor" href="#Added-27">Added</a><a class="docs-heading-anchor-permalink" href="#Added-27" title="Permalink"></a></h3><ul><li>Added support for skew symmetric variables via <code>@variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace())</code>.</li><li><code>lp_sensitivity_report</code> has been added which significantly improves the performance of querying the sensitivity summary of an LP. <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> are deprecated.</li><li>Dual warm-starts are now supported with <code>set_dual_start_value</code> and <code>dual_start_value</code>.</li><li><code>∈</code> (<code>\in&lt;tab&gt;</code>) can now be used in macros instead of <code>=</code> or <code>in</code>.</li><li>Use <code>haskey(model::Model, key::Symbol)</code> to check if a name <code>key</code> is registered in a model.</li><li>Added <code>unregister(model::Model, key::Symbol)</code> to unregister a name <code>key</code> from <code>model</code>.</li><li>Added <code>callback_node_status</code> for use in callbacks.</li><li>Added <code>print_bridge_graph</code> to visualize the bridging graph generated by MathOptInterface.</li><li>Improved error message for containers with duplicate indices.</li></ul><h3 id="Fixed-34"><a class="docs-heading-anchor" href="#Fixed-34">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-34" title="Permalink"></a></h3><ul><li>Various fixes to pass tests on Julia 1.6.</li><li>Fixed a bug in the printing of nonlinear expressions in IJulia.</li><li>Fixed a bug when nonlinear expressions are passed to user-defined functions.</li><li>Some internal functions that were previously exported are now no longer exported.</li><li>Fixed a bug when relaxing a fixed binary variable.</li><li>Fixed a <code>StackOverflowError</code> that occurred when <code>SparseAxisArray</code>s had a large number of elements.</li><li>Removed an unnecessary type assertion in <code>list_of_constraint_types</code>.</li><li>Fixed a bug when copying models with registered expressions.</li></ul><h3 id="Other-32"><a class="docs-heading-anchor" href="#Other-32">Other</a><a class="docs-heading-anchor-permalink" href="#Other-32" title="Permalink"></a></h3><ul><li>The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in <code>/examples</code> have now been moved to <code>/docs/src/examples</code> and rewritten using <code>Literate.jl</code>, and they are now included in the documentation.</li><li>JuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.</li><li>The root cause of a large number of method invalidations has been resolved.</li><li>We switched continuous integration from Travis and Appveyor to GitHub Actions.</li></ul><h2 id="[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5">Version 0.21.5</a> (September 18, 2020)</a><a id="[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)" title="Permalink"></a></h2><h3 id="Fixed-35"><a class="docs-heading-anchor" href="#Fixed-35">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-35" title="Permalink"></a></h3><ul><li>Fixed deprecation warnings</li><li>Throw <code>DimensionMismatch</code> for incompatibly sized functions and sets</li><li>Unify treatment of <code>keys(x)</code> on JuMP containers</li></ul><h2 id="[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4">Version 0.21.4</a> (September 14, 2020)</a><a id="[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)" title="Permalink"></a></h2><h3 id="Added-28"><a class="docs-heading-anchor" href="#Added-28">Added</a><a class="docs-heading-anchor-permalink" href="#Added-28" title="Permalink"></a></h3><ul><li>Add debug info when adding unsupported constraints</li><li>Add <code>relax_integrality</code> for solving continuous relaxation</li><li>Allow querying constraint conflicts</li></ul><h3 id="Fixed-36"><a class="docs-heading-anchor" href="#Fixed-36">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-36" title="Permalink"></a></h3><ul><li>Dispatch on <code>Real</code> for <code>MOI.submit</code></li><li>Implement <code>copy</code> for <code>CustomSet</code> in tests</li><li>Don&#39;t export private macros</li><li>Fix invalid assertion in nonlinear</li><li>Error if constraint has <code>NaN</code> right-hand side</li><li>Improve speed of tests</li><li>Lots of work modularizing files in <code>/test</code></li><li>Improve line numbers in macro error messages</li><li>Print nonlinear subexpressions</li><li>Various documentation updates</li><li>Dependency updates:<ul><li>Datastructures 0.18</li><li>MathOptFormat v0.5</li><li>Prep for MathOptInterface 0.9.15</li></ul></li></ul><h2 id="[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3">Version 0.21.3</a> (June 18, 2020)</a><a id="[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)" title="Permalink"></a></h2><ul><li>Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (<a href="https://github.com/jump-dev/JuMP.jl/issues/2212">#2212</a>).</li><li>Added functions <code>simplex_iterations</code>, <code>barrier_iterations</code> and <code>node_count</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2201">#2201</a>).</li><li>Added function <code>reduced_cost</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2205">#2205</a>).</li><li>Implemented <code>callback_value</code> for affine and quadratic expressions (<a href="https://github.com/jump-dev/JuMP.jl/issues/2231">#2231</a>).</li><li>Support <code>MutableArithmetics.Zero</code> in objective and constraints (<a href="https://github.com/jump-dev/JuMP.jl/issues/2219">#2219</a>).</li><li>Documentation improvements:<ul><li>Mention tutorials in the docs (<a href="https://github.com/jump-dev/JuMP.jl/issues/2223">#2223</a>).</li><li>Update COIN-OR links (<a href="https://github.com/jump-dev/JuMP.jl/issues/2242">#2242</a>).</li><li>Explicit link to the documentation of <code>MOI.FileFormats</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2253">#2253</a>).</li><li>Typo fixes (<a href="https://github.com/jump-dev/JuMP.jl/issues/2261">#2261</a>).</li></ul></li><li>Containers improvements:<ul><li>Fix <code>Base.map</code> for <code>DenseAxisArray</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2235">#2235</a>).</li><li>Throw <code>BoundsError</code> if number of indices is incorrect for <code>DenseAxisArray</code> and <code>SparseAxisArray</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2240">#2240</a>).</li></ul></li><li>Extensibility improvements:<ul><li>Implement a <code>set_objective</code> method fallback that redirects to <code>set_objective_sense</code> and <code>set_objective_function</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2247">#2247</a>).</li><li>Add <code>parse_constraint</code> method with arbitrary number of arguments (<a href="https://github.com/jump-dev/JuMP.jl/issues/2051">#2051</a>).</li><li>Add <code>parse_constraint_expr</code> and <code>parse_constraint_head</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2228">#2228</a>).</li></ul></li></ul><h2 id="[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2">Version 0.21.2</a> (April 2, 2020)</a><a id="[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)" title="Permalink"></a></h2><ul><li>Added <code>relative_gap()</code> to access <code>MOI.RelativeGap()</code> attribute (<a href="https://github.com/jump-dev/JuMP.jl/issues/2199">#2199</a>).</li><li>Documentation fixes:<ul><li>Added link to source for docstrings in the documentation (<a href="https://github.com/jump-dev/JuMP.jl/issues/2207">#2207</a>).</li><li>Added docstring for <code>@variables</code> macro (<a href="https://github.com/jump-dev/JuMP.jl/issues/2216">#2216</a>).</li><li>Typo fixes (#2177, #2184, #2182).</li></ul></li><li>Implementation of methods for Base functions:<ul><li>Implemented <code>Base.empty!</code> for <code>JuMP.Model</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2198">#2198</a>).</li><li>Implemented <code>Base.conj</code> for JuMP scalar types (<a href="https://github.com/jump-dev/JuMP.jl/issues/2209">#2209</a>).</li></ul></li></ul><h3 id="Fixed-37"><a class="docs-heading-anchor" href="#Fixed-37">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-37" title="Permalink"></a></h3><ul><li>Fixed sum of expression with scalar product in macro (<a href="https://github.com/jump-dev/JuMP.jl/issues/2178">#2178</a>).</li><li>Fixed writing of nonlinear models to MathOptFormat (<a href="https://github.com/jump-dev/JuMP.jl/issues/2181">#2181</a>).</li><li>Fixed construction of empty SparseAxisArray (<a href="https://github.com/jump-dev/JuMP.jl/issues/2179">#2179</a>).</li><li>Fixed constraint with zero function (<a href="https://github.com/jump-dev/JuMP.jl/issues/2188">#2188</a>).</li></ul><h2 id="[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1">Version 0.21.1</a> (Feb 18, 2020)</a><a id="[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)" title="Permalink"></a></h2><ul><li>Improved the clarity of the <code>with_optimizer</code> deprecation warning.</li></ul><h2 id="[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)"><a class="docs-heading-anchor" href="#[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0">Version 0.21.0</a> (Feb 16, 2020)</a><a id="[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)" title="Permalink"></a></h2><h3 id="Breaking-6"><a class="docs-heading-anchor" href="#Breaking-6">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-6" title="Permalink"></a></h3><ul><li><p>Deprecated <code>with_optimizer</code> (#2090, #2084, #2141). You can replace <code>with_optimizer</code> by either nothing, <code>optimizer_with_attributes</code> or a closure:</p><ul><li>replace <code>with_optimizer(Ipopt.Optimizer)</code> by <code>Ipopt.Optimizer</code>.</li><li>replace <code>with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0)</code> by <code>optimizer_with_attributes(Ipopt.Optimizer, &quot;max_cpu_time&quot; =&gt; 60.0)</code>.</li><li>replace <code>with_optimizer(Gurobi.Optimizer, env)</code> by <code>() -&gt; Gurobi.Optimizer(env)</code>.</li><li>replace <code>with_optimizer(Gurobi.Optimizer, env, Presolve=0)</code> by <code>optimizer_with_attributes(() -&gt; Gurobi.Optimizer(env), &quot;Presolve&quot; =&gt; 0)</code>.</li></ul><p>alternatively to <code>optimizer_with_attributes</code>, you can also set the attributes separately with <code>set_optimizer_attribute</code>.</p></li><li><p>Renamed <code>set_parameter</code> and <code>set_parameters</code> to <code>set_optimizer_attribute</code> and <code>set_optimizer_attributes</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2150">#2150</a>).</p></li><li><p>Broadcast should now be explicit inside macros. <code>@SDconstraint(model, x &gt;= 1)</code> and <code>@constraint(model, x + 1 in SecondOrderCone())</code> now throw an error instead of broadcasting <code>1</code> along the dimension of <code>x</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2107">#2107</a>).</p></li><li><p><code>@SDconstraint(model, x &gt;= 0)</code> is now equivalent to <code>@constraint(model, x in PSDCone())</code> instead of <code>@constraint(model, (x .- 0) in PSDCone())</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/2107">#2107</a>).</p></li><li><p>The macros now create the containers with <code>map</code> instead of <code>for</code> loops, as a consequence, containers created by <code>@expression</code> can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where <code>@expression</code> could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (<a href="https://github.com/jump-dev/JuMP.jl/issues/2070">#2070</a>).</p></li><li><p>Calling <code>deepcopy(::AbstractModel)</code> now throws an error.</p></li><li><p>The constraint name is now printed in the model string (<a href="https://github.com/jump-dev/JuMP.jl/issues/2108">#2108</a>).</p></li></ul><h3 id="Added-29"><a class="docs-heading-anchor" href="#Added-29">Added</a><a class="docs-heading-anchor-permalink" href="#Added-29" title="Permalink"></a></h3><ul><li>Added support for solver-independent and solver-specific callbacks (<a href="https://github.com/jump-dev/JuMP.jl/issues/2101">#2101</a>).</li><li>Added <code>write_to_file</code> and <code>read_from_file</code>, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (<a href="https://github.com/jump-dev/JuMP.jl/issues/2114">#2114</a>).</li><li>Added support for complementarity constraints (<a href="https://github.com/jump-dev/JuMP.jl/issues/2132">#2132</a>).</li><li>Added support for indicator constraints (<a href="https://github.com/jump-dev/JuMP.jl/issues/2092">#2092</a>).</li><li>Added support for querying multiple solutions with the <code>result</code> keyword (<a href="https://github.com/jump-dev/JuMP.jl/issues/2100">#2100</a>).</li><li>Added support for constraining variables on creation (<a href="https://github.com/jump-dev/JuMP.jl/issues/2128">#2128</a>).</li><li>Added method <code>delete</code> that deletes a vector of variables at once if it is supported by the underlying solver (<a href="https://github.com/jump-dev/JuMP.jl/issues/2135">#2135</a>).</li><li>The arithmetic between JuMP expression has be refactored into the MutableArithmetics package (<a href="https://github.com/jump-dev/JuMP.jl/issues/2107">#2107</a>).</li><li>Improved error on complex values in NLP (<a href="https://github.com/jump-dev/JuMP.jl/issues/1978">#1978</a>).</li><li>Added an example of column generation (<a href="https://github.com/jump-dev/JuMP.jl/issues/2010">#2010</a>).</li></ul><h3 id="Fixed-38"><a class="docs-heading-anchor" href="#Fixed-38">Fixed</a><a class="docs-heading-anchor-permalink" href="#Fixed-38" title="Permalink"></a></h3><ul><li>Incorrect coefficients generated when using Symmetric variables (<a href="https://github.com/jump-dev/JuMP.jl/issues/2102">#2102</a>)</li></ul><h2 id="[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)"><a class="docs-heading-anchor" href="#[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1">Version 0.20.1</a> (Oct 18, 2019)</a><a id="[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)" title="Permalink"></a></h2><ul><li>Add sections on <code>@variables</code> and <code>@constraints</code> in the documentation (<a href="https://github.com/jump-dev/JuMP.jl/issues/2062">#2062</a>).</li><li>Fixed product of sparse matrices for Julia v1.3 (<a href="https://github.com/jump-dev/JuMP.jl/issues/2063">#2063</a>).</li><li>Added <code>set_objective_coefficient</code> to modify the coefficient of a linear term of the objective function (<a href="https://github.com/jump-dev/JuMP.jl/issues/2008">#2008</a>).</li><li>Added <code>set_time_limit_sec</code>, <code>unset_time_limit_sec</code> and <code>time_limit_sec</code> to set and query the time limit for the solver in seconds (<a href="https://github.com/jump-dev/JuMP.jl/issues/2053">#2053</a>).</li></ul><h2 id="[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)"><a class="docs-heading-anchor" href="#[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0">Version 0.20.0</a> (Aug 24, 2019)</a><a id="[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)" title="Permalink"></a></h2><ul><li>Documentation updates.</li><li>Numerous bug fixes.</li><li>Better error messages (#1977, #1978, #1997, #2017).</li><li>Performance improvements (#1947, #2032).</li><li>Added LP sensitivity summary functions <code>lp_objective_perturbation_range</code> and <code>lp_rhs_perturbation_range</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/1917">#1917</a>).</li><li>Added functions <code>dual_objective_value</code>, <code>raw_status</code> and <code>set_parameter</code>.</li><li>Added function <code>set_objective_coefficient</code> to modify the coefficient of a linear term of the objective (<a href="https://github.com/jump-dev/JuMP.jl/issues/2008">#2008</a>).</li><li>Added functions <code>set_normalized_rhs</code>, <code>normalized_rhs</code>, and <code>add_to_function_constant</code> to modify and get the constant part of a constraint (#1935, #1960).</li><li>Added functions <code>set_normalized_coefficient</code> and <code>normalized_coefficient</code> to modify and get the coefficient of a linear term of a constraint (#1935, #1960).</li><li>Numerous other improvements in MOI 0.9, see the <code>NEWS.md</code> file of MOI for more details.</li></ul><h2 id="[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)"><a class="docs-heading-anchor" href="#[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2">Version 0.19.2</a> (June 8, 2019)</a><a id="[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)" title="Permalink"></a></h2><ul><li>Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.</li></ul><h2 id="[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)"><a class="docs-heading-anchor" href="#[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1">Version 0.19.1</a> (May 12, 2019)</a><a id="[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)" title="Permalink"></a></h2><ul><li>Usability and performance improvements.</li><li>Bug fixes.</li></ul><h2 id="[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)"><a class="docs-heading-anchor" href="#[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0">Version 0.19.0</a> (February 15, 2019)</a><a id="[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)" title="Permalink"></a></h2><p><strong>JuMP 0.19 contains significant breaking changes.</strong></p><h3 id="Breaking-7"><a class="docs-heading-anchor" href="#Breaking-7">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-7" title="Permalink"></a></h3><ul><li><p>JuMP&#39;s abstraction layer for communicating with solvers changed from <a href="https://github.com/JuliaOpt/MathProgBase.jl">MathProgBase</a> (MPB) to <a href="https://github.com/JuliaOpt/MathOptInterface.jl">MathOptInterface</a> (MOI). MOI addresses many longstanding design issues. (See @mlubin&#39;s <a href="https://www.juliaopt.org/meetings/bordeaux2018/lubin.pdf">slides</a> from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the <a href="https://www.juliaopt.org/JuMP.jl/dev/installation/">installation guide</a> for a list of solvers that have and have not yet been updated.</p></li><li><p>Most solvers have been renamed to <code>PackageName.Optimizer</code>. For example, <code>GurobiSolver()</code> is now <code>Gurobi.Optimizer</code>.</p></li><li><p>Solvers are no longer added to a model via <code>Model(solver = XXX(kwargs...))</code>. Instead use <code>Model(with_optimizer(XXX, kwargs...))</code>. For example, <code>Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0))</code>.</p></li><li><p>JuMP containers (for example, the objects returned by <code>@variable</code>) have been redesigned. <code>Containers.SparseAxisArray</code> replaces <code>JuMPDict</code>, <code>JuMPArray</code> was rewritten (inspired by <code>AxisArrays</code>) and renamed <code>Containers.DenseAxisArray</code>, and you can now request a container type with the <code>container=</code> keyword to the macros. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/variables/#Variable-containers-1">documentation</a> for more details.</p></li><li><p>The statuses returned by solvers have changed. See the possible status values <a href="https://www.juliaopt.org/MathOptInterface.jl/stable/apireference.html#Termination-Status-1">here</a>. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).</p></li><li><p>Starting values are separate from result values. Use <code>value</code> to query the value of a variable in a solution. Use <code>start_value</code> and <code>set_start_value</code> to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.</p></li><li><p>The data structures for affine and quadratic expressions <code>AffExpr</code> and <code>QuadExpr</code> have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.</p></li><li><p><code>JuMPNLPEvaluator</code> no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.</p></li><li><p>JuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)</p></li><li><p>The <code>norm()</code> syntax is no longer recognized inside macros. Use the <code>SecondOrderCone()</code> set instead.</p></li><li><p>JuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.</p></li><li><p>The symbols <code>:Min</code> and <code>:Max</code> are no longer used as optimization senses. Instead, JuMP uses the <code>OptimizationSense</code> enum from MathOptInterface. <code>@objective(model, Max, ...)</code>, <code>@objective(model, Min, ...)</code>, <code>@NLobjective(model, Max, ...)</code>, and <code>@objective(model, Min, ...)</code> remain valid, but <code>@objective(m, :Max, ...)</code> is no longer accepted.</p></li><li><p>The sign conventions for duals has changed in some cases for consistency with conic duality (see the <a href="https://www.juliaopt.org/MathOptInterface.jl/v0.6.2/apimanual.html#Duals-1">documentation</a>). The <code>shadow_price</code> helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.</p></li><li><p><code>@constraintref</code> is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,</p><pre><code class="language-julia hljs">constraints = Dict() # Optionally, specify types for improved performance.
 for i in 1:N
   constraints[i] = @constraint(model, ...)
-end</code></pre></li><li><p>The <code>lowerbound</code>, <code>upperbound</code>, and <code>basename</code> keyword arguments to the <code>@variable</code> macro have been renamed to <code>lower_bound</code>, <code>upper_bound</code>, and <code>base_name</code>, for consistency with JuMP&#39;s new <a href="https://www.juliaopt.org/JuMP.jl/dev/style/">style recommendations</a>.</p></li><li><p>We rely on broadcasting syntax to apply accessors to collections of variables, for example, <code>value.(x)</code> instead of <code>getvalue(x)</code> for collections. (Use <code>value(x)</code> when <code>x</code> is a scalar object.)</p></li></ul><h3 id="Added-30"><a class="docs-heading-anchor" href="#Added-30">Added</a><a class="docs-heading-anchor-permalink" href="#Added-30" title="Permalink"></a></h3><ul><li><p>Splatting (like <code>f(x...)</code>) is recognized in restricted settings in nonlinear expressions.</p></li><li><p>Support for deleting constraints and variables.</p></li><li><p>The documentation has been completely rewritten using docstrings and Documenter.</p></li><li><p>Support for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).</p></li><li><p>Significantly improved support for modeling new types of constraints and for extending JuMP&#39;s macros.</p></li><li><p>Support for providing dual warm starts.</p></li><li><p>Improved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).</p></li><li><p>Explicit control of whether symmetry-enforcing constraints are added to PSD constraints.</p></li><li><p>Support for modeling exponential cones.</p></li><li><p>Significant improvements in internal code quality and testing.</p></li><li><p>Style and naming guidelines.</p></li><li><p>Direct mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/solvers/">documentation</a>.</p></li></ul><h3 id="Regressions"><a class="docs-heading-anchor" href="#Regressions">Regressions</a><a id="Regressions-1"></a><a class="docs-heading-anchor-permalink" href="#Regressions" title="Permalink"></a></h3><p>There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):</p><ul><li><p>Performance regressions in model generation (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1403">issue</a>). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.</p></li><li><p>Fast incremental NLP solves are not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1185">issue</a>).</p></li><li><p>We do not yet have an implementation of solver-specific callbacks.</p></li><li><p>The column generation syntax in <code>@variable</code> has been removed (that is, the <code>objective</code>, <code>coefficients</code>, and <code>inconstraints</code> keyword arguments). Support for column generation will be re-introduced in a future release.</p></li><li><p>The ability to solve the continuous relaxation (that is, via <code>solve(model; relaxation = true)</code>) is not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1611">issue</a>).</p></li></ul><h2 id="[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5">Version 0.18.5</a> (December 1, 2018)</a><a id="[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)" title="Permalink"></a></h2><ul><li>Support views in some derivative evaluation functions.</li><li>Improved compatibility with PackageCompiler.</li></ul><h2 id="[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4">Version 0.18.4</a> (October 8, 2018)</a><a id="[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in model printing on Julia 0.7 and 1.0.</li></ul><h2 id="[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3">Version 0.18.3</a> (October 1, 2018)</a><a id="[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)" title="Permalink"></a></h2><ul><li>Add support for Julia v1.0 (Thanks @ExpandingMan)</li><li>Fix matrix expressions with quadratic functions (<a href="https://github.com/jump-dev/JuMP.jl/issues/1508">#1508</a>)</li></ul><h2 id="[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2">Version 0.18.2</a> (June 10, 2018)</a><a id="[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in second-order derivatives when expressions are present (<a href="https://github.com/jump-dev/JuMP.jl/issues/1319">#1319</a>)</li><li>Fix a bug in <code>@constraintref</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/1330">#1330</a>)</li></ul><h2 id="[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1">Version 0.18.1</a> (April 9, 2018)</a><a id="[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)" title="Permalink"></a></h2><ul><li>Fix for nested tuple destructuring (<a href="https://github.com/jump-dev/JuMP.jl/issues/1193">#1193</a>)</li><li>Preserve internal model when relaxation=true (<a href="https://github.com/jump-dev/JuMP.jl/issues/1209">#1209</a>)</li><li>Minor bug fixes and updates for example</li></ul><h2 id="[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)"><a class="docs-heading-anchor" href="#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0">Version 0.18.0</a> (July 27, 2017)</a><a id="[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)" title="Permalink"></a></h2><ul><li>Drop support for Julia 0.5.</li><li>Update for ForwardDiff 0.5.</li><li>Minor bug fixes.</li></ul><h2 id="[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)"><a class="docs-heading-anchor" href="#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1">Version 0.17.1</a> (June 9, 2017)</a><a id="[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)" title="Permalink"></a></h2><ul><li>Use of <code>constructconstraint!</code> in <code>@SDconstraint</code>.</li><li>Minor bug fixes.</li></ul><h2 id="[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)"><a class="docs-heading-anchor" href="#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0">Version 0.17.0</a> (May 27, 2017)</a><a id="[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: Mixing quadratic and conic constraints is no longer supported.</li><li><strong>Breaking change</strong>: The <code>getvariable</code> and <code>getconstraint</code> functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name <code>x</code>, one should now write <code>m[:x]</code> instead of <code>getvariable(m, :x)</code>. As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.</li><li>Addition of the <code>getobjectivebound</code> function that mirrors the functionality of the MathProgBase <code>getobjbound</code> function except that it takes into account transformations performed by JuMP.</li><li>Minor bug fixes.</li></ul><p>The following changes are primarily of interest to developers of JuMP extensions:</p><ul><li>The new syntax <code>@constraint(model, expr in Cone)</code> creates the constraint ensuring that <code>expr</code> is inside <code>Cone</code>. The <code>Cone</code> argument is passed to <code>constructconstraint!</code> which enables the call to the dispatched to an extension.</li><li>The <code>@variable</code> macro now calls <code>constructvariable!</code> instead of directly calling the <code>Variable</code> constructor. Extra arguments and keyword arguments passed to <code>@variable</code> are passed to <code>constructvariable!</code> which enables the call to be dispatched to an extension.</li><li>Refactor the internal function <code>conicdata</code> (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.</li></ul><h2 id="[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2">Version 0.16.2</a> (March 28, 2017)</a><a id="[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)" title="Permalink"></a></h2><ul><li>Minor bug fixes and printing tweaks</li><li>Address deprecation warnings for Julia 0.6</li></ul><h2 id="[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1">Version 0.16.1</a> (March 7, 2017)</a><a id="[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)" title="Permalink"></a></h2><ul><li>Better support for <code>AbstractArray</code> in JuMP (Thanks @tkoolen)</li><li>Minor bug fixes</li></ul><h2 id="[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0">Version 0.16.0</a> (February 23, 2017)</a><a id="[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling <code>solve()</code> will result in an error.</li><li><strong>Breaking change</strong>: User-defined functions are no longer global. The first argument to <code>JuMP.register</code> is now a JuMP <code>Model</code> object within whose scope the function will be registered. Calling <code>JuMP.register</code> without a <code>Model</code> now produces an error.</li><li><strong>Breaking change</strong>: Use the new <code>JuMP.fix</code> method to fix a variable to a value or to update the value to which a variable is fixed. Calling <code>setvalue</code> on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)</li><li>Nonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)</li><li>New <code>category</code> keyword to <code>@variable</code>. Used for specifying categories of anonymous variables.</li><li>Compatibility with Julia 0.6-dev.</li><li>Minor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)</li></ul><h2 id="[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)"><a class="docs-heading-anchor" href="#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1">Version 0.15.1</a> (January 31, 2017)</a><a id="[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)" title="Permalink"></a></h2><ul><li>Bugfix for <code>@LinearConstraints</code> and friends</li></ul><h2 id="[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)"><a class="docs-heading-anchor" href="#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0">Version 0.15.0</a> (December 22, 2016)</a><a id="[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)" title="Permalink"></a></h2><ul><li>Julia 0.5.0 is the minimum required version for this release.</li><li>Document support for BARON solver</li><li>Enable info callbacks in more states than before, for example, for recording solutions. New <code>when</code> argument to <code>addinfocallback</code> (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/814">#814</a>, thanks @yeesian)</li><li>Improved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:<ul><li>When multiple variables in a model are given the same name</li><li>When non-symbols are used as names, for example, <code>@variable(m, x[1][1:N])</code></li></ul></li><li>Improvements in iterating over JuMP containers (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/836">#836</a>, thanks @IssamT)</li><li>Support for writing variable names in .lp file output (Thanks @leethargo)</li><li>Support for querying duals to SDP problems (Thanks @blegat)</li><li>The comprehension syntax with curly braces <code>sum{}</code>, <code>prod{}</code>, and <code>norm2{}</code> has been deprecated in favor of Julia&#39;s native comprehension syntax <code>sum()</code>, <code>prod()</code> and <code>norm()</code> as previously announced. (For early adopters of the new syntax, <code>norm2()</code> was renamed to <code>norm()</code> without deprecation.)</li><li>Unit tests rewritten to use Base.Test instead of FactCheck</li><li>Improved support for operations with matrices of JuMP types (Thanks @ExpandingMan)</li><li>The syntax to halt a solver from inside a callback has changed from <code>throw(CallbackAbort())</code> to <code>return JuMP.StopTheSolver</code></li><li>Minor bug fixes</li></ul><h2 id="[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2">Version 0.14.2</a> (December 12, 2016)</a><a id="[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)" title="Permalink"></a></h2><ul><li>Allow singleton anonymous variables (includes bugfix)</li></ul><h2 id="[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1">Version 0.14.1</a> (September 12, 2016)</a><a id="[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)" title="Permalink"></a></h2><ul><li>More consistent handling of states in informational callbacks, includes a new <code>when</code> parameter to <code>addinfocallback</code> for specifying in which state an informational callback should be called.</li></ul><h2 id="[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0">Version 0.14.0</a> (August 7, 2016)</a><a id="[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)" title="Permalink"></a></h2><ul><li>Compatibility with Julia 0.5 and ForwardDiff 0.2</li><li>Support for &quot;anonymous&quot; variables, constraints, expressions, and parameters, for example, <code>x = @variable(m, [1:N])</code> instead of <code>@variable(m, x[1:N])</code></li><li>Support for retrieving constraints from a model by name via <code>getconstraint</code></li><li><code>@NLconstraint</code> now returns constraint references (as expected).</li><li>Support for vectorized expressions within lazy constraints</li><li>On Julia 0.5, parse new comprehension syntax <code>sum(x[i] for i in 1:N if isodd(i))</code> instead of <code>sum{ x[i], i in 1:N; isodd(i) }</code>. The old syntax with curly braces will be deprecated in JuMP 0.15.</li><li>Now possible to provide nonlinear expressions as &quot;raw&quot; Julia <code>Expr</code> objects instead of using JuMP&#39;s nonlinear macros. This input format is useful for programmatically generated expressions.</li><li><code>s/Mathematical Programming/Mathematical Optimization/</code></li><li>Support for local cuts (Thanks to @madanim, Mehdi Madani)</li><li>Document Xpress interface developed by @joaquimg, Joaquim Dias Garcia</li><li>Minor bug and deprecation fixes (Thanks @odow, @jrevels)</li></ul><h2 id="[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2">Version 0.13.2</a> (May 16, 2016)</a><a id="[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for MathProgBase</li></ul><h2 id="[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1">Version 0.13.1</a> (May 3, 2016)</a><a id="[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)" title="Permalink"></a></h2><ul><li>Fix broken deprecation for <code>registerNLfunction</code>.</li></ul><h2 id="[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0">Version 0.13.0</a> (April 29, 2016)</a><a id="[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)" title="Permalink"></a></h2><ul><li>Most exported methods and macros have been renamed to avoid camelCase. See the list of changes <a href="https://github.com/JuliaOpt/JuMP.jl/blob/e53d0db67cde2a4b80d0c1281f4b49eb0128a1f5/src/deprecated.jl#L30">here</a>. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.</li><li>Specify variable lower/upper bounds in <code>@variable</code> using the <code>lowerbound</code> and <code>upperbound</code> keyword arguments.</li><li>Change name printed for variable using the <code>basename</code> keyword argument to <code>@variable</code>.</li><li>New <code>@variables</code> macro allows multi-line declaration of groups of variables.</li><li>A number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was <a href="https://youtu.be/qF1lZPJ3a5A">recorded</a> live.</li><li>Compatibility fixes with Julia 0.5.</li><li>The &quot;end&quot; indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see <a href="https://github.com/JuliaOpt/JuMP.jl/issues/730">here</a>.</li></ul><h2 id="[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2">Version 0.12.2</a> (March 9, 2016)</a><a id="[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)" title="Permalink"></a></h2><ul><li>Small fixes for nonlinear optimization</li></ul><h2 id="[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1">Version 0.12.1</a> (March 1, 2016)</a><a id="[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)" title="Permalink"></a></h2><ul><li>Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)</li></ul><h2 id="[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0">Version 0.12.0</a> (February 27, 2016)</a><a id="[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)" title="Permalink"></a></h2><ul><li>The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:<ul><li><code>@defExpr</code> and <code>@defNLExpr</code> now take the model as the first argument. The previous one-argument version of <code>@defExpr</code> is deprecated; all expressions should be named. For example, replace <code>@defExpr(2x+y)</code> with <code>@defExpr(jump_model, my_expr, 2x+y)</code>.</li><li>JuMP no longer uses Julia&#39;s variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.</li><li>Support for user-defined functions integrated within nonlinear JuMP expressions.</li></ul></li><li>Replaced iteration over <code>AffExpr</code> with <code>Number</code>-like scalar iteration; previous iteration behavior is now available via <code>linearterms(::AffExpr)</code>.</li><li>Stopping the solver via <code>throw(CallbackAbort())</code> from a callback no longer triggers an exception. Instead, <code>solve()</code> returns <code>UserLimit</code> status.</li><li><code>getDual()</code> now works for conic problems (Thanks @emreyamangil.)</li></ul><h2 id="[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)"><a class="docs-heading-anchor" href="#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3">Version 0.11.3</a> (February 4, 2016)</a><a id="[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)" title="Permalink"></a></h2><ul><li>Bug-fix for problems with quadratic objectives and semidefinite constraints</li></ul><h2 id="[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)"><a class="docs-heading-anchor" href="#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2">Version 0.11.2</a> (January 14, 2016)</a><a id="[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for Mosek</li></ul><h2 id="[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)"><a class="docs-heading-anchor" href="#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1">Version 0.11.1</a> (December 1, 2015)</a><a id="[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)" title="Permalink"></a></h2><ul><li>Remove usage of <code>@compat</code> in tests.</li><li>Fix updating quadratic objectives for nonlinear models.</li></ul><h2 id="[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)"><a class="docs-heading-anchor" href="#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0">Version 0.11.0</a> (November 30, 2015)</a><a id="[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)" title="Permalink"></a></h2><ul><li>Julia 0.4.0 is the minimum required version for this release.</li><li>Fix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.</li><li>Addition of the <code>solve(m::Model, relaxation=true)</code> keyword argument to solve the standard continuous relaxation of model <code>m</code></li><li>The <code>getConstraintBounds()</code> method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.</li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3">Version 0.10.3</a> (November 20, 2015)</a><a id="[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)" title="Permalink"></a></h2><ul><li>Fix a rare error when parsing quadratic expressions</li><li>Fix <code>Variable()</code> constructor with default arguments</li><li>Detect unrecognized keywords in <code>solve()</code></li></ul><h2 id="[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2">Version 0.10.2</a> (September 28, 2015)</a><a id="[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)" title="Permalink"></a></h2><ul><li>Fix for deprecation warnings</li></ul><h2 id="[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1">Version 0.10.1</a> (September 3, 2015)</a><a id="[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)" title="Permalink"></a></h2><ul><li>Fixes for ambiguity warnings.</li><li>Fix for breaking change in precompilation syntax in Julia 0.4-pre</li></ul><h2 id="[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0">Version 0.10.0</a> (August 31, 2015)</a><a id="[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)" title="Permalink"></a></h2><ul><li>Support (on Julia 0.4 and later) for conditions in indexing <code>@defVar</code> and <code>@addConstraint</code> constructs, for example, <code>@defVar(m, x[i=1:5,j=1:5; i+j &gt;= 3])</code></li><li>Support for vectorized operations on Variables and expressions. See the documentation for details.</li><li>New <code>getVar()</code> method to access variables in a model by name</li><li>Support for semidefinite programming.</li><li>Dual solutions are now available for general nonlinear problems. You may call <code>getDual</code> on a reference object for a nonlinear constraint, and <code>getDual</code> on a variable object for Lagrange multipliers from active bounds.</li><li>Introduce warnings for two common performance traps: too many calls to <code>getValue()</code> on a collection of variables and use of the <code>+</code> operator in a loop to sum expressions.</li><li>Second-order cone constraints can be written directly with the <code>norm()</code> and <code>norm2{}</code> syntax.</li><li>Implement MathProgBase interface for querying Hessian-vector products.</li><li>Iteration over <code>JuMPContainer</code>s is deprecated; instead, use the <code>keys</code> and <code>values</code> functions, and <code>zip(keys(d),values(d))</code> for the old behavior.</li><li><code>@defVar</code> returns <code>Array{Variable,N}</code> when each of <code>N</code> index sets are of the form <code>1:nᵢ</code>.</li><li>Module precompilation: on Julia 0.4 and later, <code>using JuMP</code> is now much faster.</li></ul><h2 id="[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3">Version 0.9.3</a> (August 11, 2015)</a><a id="[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)" title="Permalink"></a></h2><ul><li>Fixes for FactCheck testing on julia v0.4.</li></ul><h2 id="[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2">Version 0.9.2</a> (June 27, 2015)</a><a id="[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)" title="Permalink"></a></h2><ul><li>Fix bug in @addConstraints.</li></ul><h2 id="[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1">Version 0.9.1</a> (April 25, 2015)</a><a id="[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)" title="Permalink"></a></h2><ul><li>Fix for Julia 0.4-dev.</li><li>Small infrastructure improvements for extensions.</li></ul><h2 id="[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0">Version 0.9.0</a> (April 18, 2015)</a><a id="[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)" title="Permalink"></a></h2><ul><li>Comparison operators for constructing constraints (for example, <code>2x &gt;= 1</code>) have been deprecated. Instead, construct the constraints explicitly in the <code>@addConstraint</code> macro to add them to the model, or in the <code>@LinearConstraint</code> macro to create a stand-alone linear constraint instance.</li><li><code>getValue()</code> method implemented to compute the value of a nonlinear subexpression</li><li>JuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.</li><li>A number of performance improvements in ReverseDiffSparse for computing derivatives.</li><li><code>MathProgBase.getsolvetime(m)</code> now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)</li><li>Formatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).</li></ul><h2 id="[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)"><a class="docs-heading-anchor" href="#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0">Version 0.8.0</a> (February 17, 2015)</a><a id="[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)" title="Permalink"></a></h2><ul><li>Nonlinear subexpressions now supported with the <code>@defNLExpr</code> macro.</li><li>SCS supported for solving second-order conic problems.</li><li><code>setXXXCallback</code> family deprecated in favor of <code>addXXXCallback</code>.</li><li>Multiple callbacks of the same type can be registered.</li><li>Added support for informational callbacks via <code>addInfoCallback</code>.</li><li>A <code>CallbackAbort</code> exception can be thrown from callback to safely exit optimization.</li></ul><h2 id="[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4">Version 0.7.4</a> (February 4, 2015)</a><a id="[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)" title="Permalink"></a></h2><ul><li>Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.</li><li>Two-sided nonlinear constraints are supported.</li><li>Methods for accessing the number of variables and constraints in a model are renamed.</li><li>New default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.</li><li>Small bug fixes.</li></ul><h2 id="[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3">Version 0.7.3</a> (January 14, 2015)</a><a id="[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)" title="Permalink"></a></h2><ul><li>Fix a method ambiguity conflict with Compose.jl (cosmetic fix)</li></ul><h2 id="[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2">Version 0.7.2</a> (January 9, 2015)</a><a id="[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in <code>sum(::JuMPDict)</code></li><li>Added the <code>setCategory</code> function to change a variables category (for example, continuous or binary)</li></ul><p>after construction, and <code>getCategory</code> to retrieve the variable category.</p><h2 id="[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1">Version 0.7.1</a> (January 2, 2015)</a><a id="[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.</li></ul><h2 id="[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)"><a class="docs-heading-anchor" href="#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0">Version 0.7.0</a> (December 29, 2014)</a><a id="[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)" title="Permalink"></a></h2><h3 id="Linear/quadratic/conic-programming"><a class="docs-heading-anchor" href="#Linear/quadratic/conic-programming">Linear/quadratic/conic programming</a><a id="Linear/quadratic/conic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Linear/quadratic/conic-programming" title="Permalink"></a></h3><ul><li><strong>Breaking change</strong>: The syntax for column-wise model generation has been changed to use keyword arguments in <code>@defVar</code>.</li><li>On Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.</li><li>ECOS supported for solving second-order conic problems.</li></ul><h3 id="_nonlinear_programming_release_notes"><a class="docs-heading-anchor" href="#_nonlinear_programming_release_notes">Nonlinear programming</a><a id="_nonlinear_programming_release_notes-1"></a><a class="docs-heading-anchor-permalink" href="#_nonlinear_programming_release_notes" title="Permalink"></a></h3><ul><li>Support for skipping model generation when solving a sequence of nonlinear models with changing data.</li><li>Fix a memory leak when solving a sequence of nonlinear models.</li><li>The <code>@addNLConstraint</code> macro now supports the three-argument version to define sets of nonlinear constraints.</li><li>KNITRO supported as a nonlinear solver.</li><li>Speed improvements for model generation.</li><li>The <code>@addNLConstraints</code> macro supports adding multiple (groups of) constraints at once. Syntax is similar to <code>@addConstraints</code>.</li><li>Discrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).</li></ul><h3 id="General"><a class="docs-heading-anchor" href="#General">General</a><a id="General-1"></a><a class="docs-heading-anchor-permalink" href="#General" title="Permalink"></a></h3><ul><li>Starting values for variables may now be specified with <code>@defVar(m, x, start=value)</code>.</li><li>The <code>setSolver</code> function allows users to change the solver subsequent to model creation.</li><li>Support for &quot;fixed&quot; variables via the <code>@defVar(m, x == 1)</code> syntax.</li><li>Unit tests rewritten to use FactCheck.jl, improved testing across solvers.</li></ul><h2 id="[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3">Version 0.6.3</a> (October 19, 2014)</a><a id="[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in multiplying two AffExpr objects.</li></ul><h2 id="[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2">Version 0.6.2</a> (October 11, 2014)</a><a id="[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)" title="Permalink"></a></h2><ul><li>Further improvements and bug fixes for printing.</li><li>Fixed a bug in <code>@defExpr</code>.</li><li>Support for accessing expression graphs through the MathProgBase NLP interface.</li></ul><h2 id="[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1">Version 0.6.1</a> (September 19, 2014)</a><a id="[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)" title="Permalink"></a></h2><ul><li>Improvements and bug fixes for printing.</li></ul><h2 id="[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0">Version 0.6.0</a> (September 9, 2014)</a><a id="[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)" title="Permalink"></a></h2><ul><li>Julia 0.3.0 is the minimum required version for this release.</li><li><code>buildInternalModel(m::Model)</code> added to build solver-level model in memory without optimizing.</li><li>Deprecate <code>load_model_only</code> keyword argument to <code>solve</code>.</li><li>Add groups of constraints with <code>@addConstraints</code> macro.</li><li>Unicode operators now supported, including <code>∑</code> for <code>sum</code>, <code>∏</code> for <code>prod</code>, and <code>≤</code>/<code>≥</code></li><li>Quadratic constraints supported in <code>@addConstraint</code> macro.</li><li>Quadratic objectives supported in <code>@setObjective</code> macro.</li><li>MathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems<ul><li><strong>Breaking change</strong>: <code>IpoptOptions</code> no longer supported to specify solver options, use <code>m = Model(solver=IpoptSolver(options...))</code> instead.</li></ul></li><li>New solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK</li><li>New option to control whether the lazy constraint callback is executed at each node in the B&amp;B tree or just when feasible solutions are found</li><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.</li><li>Add support for index dependencies (for example, triangular indexing) in <code>@defVar</code>, <code>@addConstraint</code>, and <code>@defExpr</code> (for example, <code>@defVar(m, x[i=1:10,j=i:10])</code>).<ul><li>This required some changes to the internal structure of JuMP containers, which may break code that explicitly stored <code>JuMPDict</code> objects.</li></ul></li></ul><h2 id="[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8">Version 0.5.8</a> (September 24, 2014)</a><a id="[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug with specifying solvers (affects Julia 0.2 only)</li></ul><h2 id="[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7">Version 0.5.7</a> (September 5, 2014)</a><a id="[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in printing models</li></ul><h2 id="[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6">Version 0.5.6</a> (September 2, 2014)</a><a id="[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)" title="Permalink"></a></h2><ul><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.<ul><li><strong>Breaking change</strong>: Syntax for <code>Variable()</code> constructor has changed (use of this interface remains discouraged)</li></ul></li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5">Version 0.5.5</a> (July 6, 2014)</a><a id="[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)" title="Permalink"></a></h2><ul><li>Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.</li></ul><h2 id="[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4">Version 0.5.4</a> (June 19, 2014)</a><a id="[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in MathProgBase which reduces loading times for <code>using JuMP</code></li><li>Fix error when MIPs not solved to optimality</li></ul><h2 id="[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3">Version 0.5.3</a> (May 21, 2014)</a><a id="[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in ReverseDiffSparse</li></ul><h2 id="[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2">Version 0.5.2</a> (May 9, 2014)</a><a id="[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)" title="Permalink"></a></h2><ul><li>Fix compatibility with Julia 0.3 prerelease</li></ul><h2 id="[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1">Version 0.5.1</a> (May 5, 2014)</a><a id="[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in coefficient handling inside lazy constraints and user cuts</li></ul><h2 id="[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0">Version 0.5.0</a> (May 2, 2014)</a><a id="[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)" title="Permalink"></a></h2><ul><li>Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.</li><li><code>getValue</code> for <code>AffExpr</code> and <code>QuadExpr</code></li><li><strong>Breaking change</strong>: <code>getSolverModel</code> replaced by <code>getInternalModel</code>, which returns the internal MathProgBase-level model</li><li>Groups of constraints can be specified with <code>@addConstraint</code> (see documentation for details). This is not a breaking change.</li><li><code>dot(::JuMPDict{Variable},::JuMPDict{Variable})</code> now returns the corresponding quadratic expression.</li></ul><h2 id="[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)"><a class="docs-heading-anchor" href="#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1">Version 0.4.1</a> (March 24, 2014)</a><a id="[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)" title="Permalink"></a></h2><ul><li>Fix bug where change in objective sense was ignored when re-solving a model.</li><li>Fix issue with handling zero coefficients in AffExpr.</li></ul><h2 id="[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)"><a class="docs-heading-anchor" href="#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0">Version 0.4.0</a> (March 10, 2014)</a><a id="[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)" title="Permalink"></a></h2><ul><li>Support for SOS1 and SOS2 constraints.</li><li>Solver-independent callback for user heuristics.</li><li><code>dot</code> and <code>sum</code> implemented for <code>JuMPDict</code> objects. Now you can say <code>@addConstraint(m, dot(a,x) &lt;= b)</code>.</li><li>Developers: support for extensions to JuMP. See definition of Model in <code>src/JuMP.jl</code> for more details.</li><li>Option to construct the low-level model before optimizing.</li></ul><h2 id="[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2">Version 0.3.2</a> (February 17, 2014)</a><a id="[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)" title="Permalink"></a></h2><ul><li>Improved model printing<ul><li>Preliminary support for IJulia output</li></ul></li></ul><h2 id="[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1">Version 0.3.1</a> (January 30, 2014)</a><a id="[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)" title="Permalink"></a></h2><ul><li>Documentation updates</li><li>Support for MOSEK</li><li>CPLEXLink renamed to CPLEX</li></ul><h2 id="[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0">Version 0.3.0</a> (January 21, 2014)</a><a id="[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)" title="Permalink"></a></h2><ul><li>Unbounded/infeasibility rays: <code>getValue()</code> will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. <code>getDual()</code> will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.</li><li>Solver-independent callbacks for user generated cuts.</li><li>Use new interface for solver-independent QCQP.</li><li><code>setlazycallback</code> renamed to <code>setLazyCallback</code> for consistency.</li></ul><h2 id="[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)"><a class="docs-heading-anchor" href="#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0">Version 0.2.0</a> (December 15, 2013)</a><a id="[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)" title="Permalink"></a></h2><h3 id="Breaking-8"><a class="docs-heading-anchor" href="#Breaking-8">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-8" title="Permalink"></a></h3><ul><li>Objective sense is specified in <code>setObjective</code> instead of in the <code>Model</code> constructor.</li><li><code>lpsolver</code> and <code>mipsolver</code> merged into single <code>solver</code> option.</li></ul><h3 id="Added-31"><a class="docs-heading-anchor" href="#Added-31">Added</a><a class="docs-heading-anchor-permalink" href="#Added-31" title="Permalink"></a></h3><ul><li>Problem modification with efficient LP restarts and MIP warm-starts.</li><li>Relatedly, column-wise modeling now supported.</li><li>Solver-independent callbacks supported. Currently we support only a &quot;lazy constraint&quot; callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.</li></ul><h2 id="[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2">Version 0.1.2</a> (November 16, 2013)</a><a id="[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)" title="Permalink"></a></h2><ul><li>Bug fixes for printing, improved error messages.</li><li>Allow <code>AffExpr</code> to be used in macros; for example, <code>ex = y + z; @addConstraint(m, x + 2*ex &lt;= 3)</code></li></ul><h2 id="[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1">Version 0.1.1</a> (October 23, 2013)</a><a id="[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)" title="Permalink"></a></h2><ul><li>Update for solver specification API changes in MathProgBase.</li></ul><h2 id="[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0">Version 0.1.0</a> (October 3, 2013)</a><a id="[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)" title="Permalink"></a></h2><ul><li>Initial public release.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../moi/release_notes/">« Release notes</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre></li><li><p>The <code>lowerbound</code>, <code>upperbound</code>, and <code>basename</code> keyword arguments to the <code>@variable</code> macro have been renamed to <code>lower_bound</code>, <code>upper_bound</code>, and <code>base_name</code>, for consistency with JuMP&#39;s new <a href="https://www.juliaopt.org/JuMP.jl/dev/style/">style recommendations</a>.</p></li><li><p>We rely on broadcasting syntax to apply accessors to collections of variables, for example, <code>value.(x)</code> instead of <code>getvalue(x)</code> for collections. (Use <code>value(x)</code> when <code>x</code> is a scalar object.)</p></li></ul><h3 id="Added-30"><a class="docs-heading-anchor" href="#Added-30">Added</a><a class="docs-heading-anchor-permalink" href="#Added-30" title="Permalink"></a></h3><ul><li><p>Splatting (like <code>f(x...)</code>) is recognized in restricted settings in nonlinear expressions.</p></li><li><p>Support for deleting constraints and variables.</p></li><li><p>The documentation has been completely rewritten using docstrings and Documenter.</p></li><li><p>Support for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).</p></li><li><p>Significantly improved support for modeling new types of constraints and for extending JuMP&#39;s macros.</p></li><li><p>Support for providing dual warm starts.</p></li><li><p>Improved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).</p></li><li><p>Explicit control of whether symmetry-enforcing constraints are added to PSD constraints.</p></li><li><p>Support for modeling exponential cones.</p></li><li><p>Significant improvements in internal code quality and testing.</p></li><li><p>Style and naming guidelines.</p></li><li><p>Direct mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding <a href="https://www.juliaopt.org/JuMP.jl/dev/solvers/">documentation</a>.</p></li></ul><h3 id="Regressions"><a class="docs-heading-anchor" href="#Regressions">Regressions</a><a id="Regressions-1"></a><a class="docs-heading-anchor-permalink" href="#Regressions" title="Permalink"></a></h3><p>There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):</p><ul><li><p>Performance regressions in model generation (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1403">issue</a>). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.</p></li><li><p>Fast incremental NLP solves are not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1185">issue</a>).</p></li><li><p>We do not yet have an implementation of solver-specific callbacks.</p></li><li><p>The column generation syntax in <code>@variable</code> has been removed (that is, the <code>objective</code>, <code>coefficients</code>, and <code>inconstraints</code> keyword arguments). Support for column generation will be re-introduced in a future release.</p></li><li><p>The ability to solve the continuous relaxation (that is, via <code>solve(model; relaxation = true)</code>) is not yet reimplemented (<a href="https://github.com/JuliaOpt/JuMP.jl/issues/1611">issue</a>).</p></li></ul><h2 id="[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5">Version 0.18.5</a> (December 1, 2018)</a><a id="[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)" title="Permalink"></a></h2><ul><li>Support views in some derivative evaluation functions.</li><li>Improved compatibility with PackageCompiler.</li></ul><h2 id="[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4">Version 0.18.4</a> (October 8, 2018)</a><a id="[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in model printing on Julia 0.7 and 1.0.</li></ul><h2 id="[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3">Version 0.18.3</a> (October 1, 2018)</a><a id="[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)" title="Permalink"></a></h2><ul><li>Add support for Julia v1.0 (Thanks @ExpandingMan)</li><li>Fix matrix expressions with quadratic functions (<a href="https://github.com/jump-dev/JuMP.jl/issues/1508">#1508</a>)</li></ul><h2 id="[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2">Version 0.18.2</a> (June 10, 2018)</a><a id="[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)" title="Permalink"></a></h2><ul><li>Fix a bug in second-order derivatives when expressions are present (<a href="https://github.com/jump-dev/JuMP.jl/issues/1319">#1319</a>)</li><li>Fix a bug in <code>@constraintref</code> (<a href="https://github.com/jump-dev/JuMP.jl/issues/1330">#1330</a>)</li></ul><h2 id="[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)"><a class="docs-heading-anchor" href="#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1">Version 0.18.1</a> (April 9, 2018)</a><a id="[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)" title="Permalink"></a></h2><ul><li>Fix for nested tuple destructuring (<a href="https://github.com/jump-dev/JuMP.jl/issues/1193">#1193</a>)</li><li>Preserve internal model when relaxation=true (<a href="https://github.com/jump-dev/JuMP.jl/issues/1209">#1209</a>)</li><li>Minor bug fixes and updates for example</li></ul><h2 id="[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)"><a class="docs-heading-anchor" href="#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0">Version 0.18.0</a> (July 27, 2017)</a><a id="[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)" title="Permalink"></a></h2><ul><li>Drop support for Julia 0.5.</li><li>Update for ForwardDiff 0.5.</li><li>Minor bug fixes.</li></ul><h2 id="[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)"><a class="docs-heading-anchor" href="#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1">Version 0.17.1</a> (June 9, 2017)</a><a id="[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)" title="Permalink"></a></h2><ul><li>Use of <code>constructconstraint!</code> in <code>@SDconstraint</code>.</li><li>Minor bug fixes.</li></ul><h2 id="[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)"><a class="docs-heading-anchor" href="#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0">Version 0.17.0</a> (May 27, 2017)</a><a id="[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: Mixing quadratic and conic constraints is no longer supported.</li><li><strong>Breaking change</strong>: The <code>getvariable</code> and <code>getconstraint</code> functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name <code>x</code>, one should now write <code>m[:x]</code> instead of <code>getvariable(m, :x)</code>. As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.</li><li>Addition of the <code>getobjectivebound</code> function that mirrors the functionality of the MathProgBase <code>getobjbound</code> function except that it takes into account transformations performed by JuMP.</li><li>Minor bug fixes.</li></ul><p>The following changes are primarily of interest to developers of JuMP extensions:</p><ul><li>The new syntax <code>@constraint(model, expr in Cone)</code> creates the constraint ensuring that <code>expr</code> is inside <code>Cone</code>. The <code>Cone</code> argument is passed to <code>constructconstraint!</code> which enables the call to the dispatched to an extension.</li><li>The <code>@variable</code> macro now calls <code>constructvariable!</code> instead of directly calling the <code>Variable</code> constructor. Extra arguments and keyword arguments passed to <code>@variable</code> are passed to <code>constructvariable!</code> which enables the call to be dispatched to an extension.</li><li>Refactor the internal function <code>conicdata</code> (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.</li></ul><h2 id="[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2">Version 0.16.2</a> (March 28, 2017)</a><a id="[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)" title="Permalink"></a></h2><ul><li>Minor bug fixes and printing tweaks</li><li>Address deprecation warnings for Julia 0.6</li></ul><h2 id="[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1">Version 0.16.1</a> (March 7, 2017)</a><a id="[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)" title="Permalink"></a></h2><ul><li>Better support for <code>AbstractArray</code> in JuMP (Thanks @tkoolen)</li><li>Minor bug fixes</li></ul><h2 id="[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)"><a class="docs-heading-anchor" href="#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0">Version 0.16.0</a> (February 23, 2017)</a><a id="[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)" title="Permalink"></a></h2><ul><li><strong>Breaking change</strong>: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling <code>solve()</code> will result in an error.</li><li><strong>Breaking change</strong>: User-defined functions are no longer global. The first argument to <code>JuMP.register</code> is now a JuMP <code>Model</code> object within whose scope the function will be registered. Calling <code>JuMP.register</code> without a <code>Model</code> now produces an error.</li><li><strong>Breaking change</strong>: Use the new <code>JuMP.fix</code> method to fix a variable to a value or to update the value to which a variable is fixed. Calling <code>setvalue</code> on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)</li><li>Nonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)</li><li>New <code>category</code> keyword to <code>@variable</code>. Used for specifying categories of anonymous variables.</li><li>Compatibility with Julia 0.6-dev.</li><li>Minor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)</li></ul><h2 id="[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)"><a class="docs-heading-anchor" href="#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1">Version 0.15.1</a> (January 31, 2017)</a><a id="[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)" title="Permalink"></a></h2><ul><li>Bugfix for <code>@LinearConstraints</code> and friends</li></ul><h2 id="[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)"><a class="docs-heading-anchor" href="#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0">Version 0.15.0</a> (December 22, 2016)</a><a id="[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)" title="Permalink"></a></h2><ul><li>Julia 0.5.0 is the minimum required version for this release.</li><li>Document support for BARON solver</li><li>Enable info callbacks in more states than before, for example, for recording solutions. New <code>when</code> argument to <code>addinfocallback</code> (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/814">#814</a>, thanks @yeesian)</li><li>Improved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:<ul><li>When multiple variables in a model are given the same name</li><li>When non-symbols are used as names, for example, <code>@variable(m, x[1][1:N])</code></li></ul></li><li>Improvements in iterating over JuMP containers (<a href="https://github.com/JuliaOpt/JuMP.jl/pull/836">#836</a>, thanks @IssamT)</li><li>Support for writing variable names in .lp file output (Thanks @leethargo)</li><li>Support for querying duals to SDP problems (Thanks @blegat)</li><li>The comprehension syntax with curly braces <code>sum{}</code>, <code>prod{}</code>, and <code>norm2{}</code> has been deprecated in favor of Julia&#39;s native comprehension syntax <code>sum()</code>, <code>prod()</code> and <code>norm()</code> as previously announced. (For early adopters of the new syntax, <code>norm2()</code> was renamed to <code>norm()</code> without deprecation.)</li><li>Unit tests rewritten to use Base.Test instead of FactCheck</li><li>Improved support for operations with matrices of JuMP types (Thanks @ExpandingMan)</li><li>The syntax to halt a solver from inside a callback has changed from <code>throw(CallbackAbort())</code> to <code>return JuMP.StopTheSolver</code></li><li>Minor bug fixes</li></ul><h2 id="[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2">Version 0.14.2</a> (December 12, 2016)</a><a id="[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)" title="Permalink"></a></h2><ul><li>Allow singleton anonymous variables (includes bugfix)</li></ul><h2 id="[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1">Version 0.14.1</a> (September 12, 2016)</a><a id="[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)" title="Permalink"></a></h2><ul><li>More consistent handling of states in informational callbacks, includes a new <code>when</code> parameter to <code>addinfocallback</code> for specifying in which state an informational callback should be called.</li></ul><h2 id="[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)"><a class="docs-heading-anchor" href="#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0">Version 0.14.0</a> (August 7, 2016)</a><a id="[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)" title="Permalink"></a></h2><ul><li>Compatibility with Julia 0.5 and ForwardDiff 0.2</li><li>Support for &quot;anonymous&quot; variables, constraints, expressions, and parameters, for example, <code>x = @variable(m, [1:N])</code> instead of <code>@variable(m, x[1:N])</code></li><li>Support for retrieving constraints from a model by name via <code>getconstraint</code></li><li><code>@NLconstraint</code> now returns constraint references (as expected).</li><li>Support for vectorized expressions within lazy constraints</li><li>On Julia 0.5, parse new comprehension syntax <code>sum(x[i] for i in 1:N if isodd(i))</code> instead of <code>sum{ x[i], i in 1:N; isodd(i) }</code>. The old syntax with curly braces will be deprecated in JuMP 0.15.</li><li>Now possible to provide nonlinear expressions as &quot;raw&quot; Julia <code>Expr</code> objects instead of using JuMP&#39;s nonlinear macros. This input format is useful for programmatically generated expressions.</li><li><code>s/Mathematical Programming/Mathematical Optimization/</code></li><li>Support for local cuts (Thanks to @madanim, Mehdi Madani)</li><li>Document Xpress interface developed by @joaquimg, Joaquim Dias Garcia</li><li>Minor bug and deprecation fixes (Thanks @odow, @jrevels)</li></ul><h2 id="[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2">Version 0.13.2</a> (May 16, 2016)</a><a id="[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for MathProgBase</li></ul><h2 id="[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1">Version 0.13.1</a> (May 3, 2016)</a><a id="[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)" title="Permalink"></a></h2><ul><li>Fix broken deprecation for <code>registerNLfunction</code>.</li></ul><h2 id="[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)"><a class="docs-heading-anchor" href="#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0">Version 0.13.0</a> (April 29, 2016)</a><a id="[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)" title="Permalink"></a></h2><ul><li>Most exported methods and macros have been renamed to avoid camelCase. See the list of changes <a href="https://github.com/JuliaOpt/JuMP.jl/blob/e53d0db67cde2a4b80d0c1281f4b49eb0128a1f5/src/deprecated.jl#L30">here</a>. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.</li><li>Specify variable lower/upper bounds in <code>@variable</code> using the <code>lowerbound</code> and <code>upperbound</code> keyword arguments.</li><li>Change name printed for variable using the <code>basename</code> keyword argument to <code>@variable</code>.</li><li>New <code>@variables</code> macro allows multi-line declaration of groups of variables.</li><li>A number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was <a href="https://youtu.be/qF1lZPJ3a5A">recorded</a> live.</li><li>Compatibility fixes with Julia 0.5.</li><li>The &quot;end&quot; indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see <a href="https://github.com/JuliaOpt/JuMP.jl/issues/730">here</a>.</li></ul><h2 id="[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2">Version 0.12.2</a> (March 9, 2016)</a><a id="[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)" title="Permalink"></a></h2><ul><li>Small fixes for nonlinear optimization</li></ul><h2 id="[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1">Version 0.12.1</a> (March 1, 2016)</a><a id="[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)" title="Permalink"></a></h2><ul><li>Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)</li></ul><h2 id="[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)"><a class="docs-heading-anchor" href="#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0">Version 0.12.0</a> (February 27, 2016)</a><a id="[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)" title="Permalink"></a></h2><ul><li>The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:<ul><li><code>@defExpr</code> and <code>@defNLExpr</code> now take the model as the first argument. The previous one-argument version of <code>@defExpr</code> is deprecated; all expressions should be named. For example, replace <code>@defExpr(2x+y)</code> with <code>@defExpr(jump_model, my_expr, 2x+y)</code>.</li><li>JuMP no longer uses Julia&#39;s variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.</li><li>Support for user-defined functions integrated within nonlinear JuMP expressions.</li></ul></li><li>Replaced iteration over <code>AffExpr</code> with <code>Number</code>-like scalar iteration; previous iteration behavior is now available via <code>linearterms(::AffExpr)</code>.</li><li>Stopping the solver via <code>throw(CallbackAbort())</code> from a callback no longer triggers an exception. Instead, <code>solve()</code> returns <code>UserLimit</code> status.</li><li><code>getDual()</code> now works for conic problems (Thanks @emreyamangil.)</li></ul><h2 id="[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)"><a class="docs-heading-anchor" href="#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3">Version 0.11.3</a> (February 4, 2016)</a><a id="[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)" title="Permalink"></a></h2><ul><li>Bug-fix for problems with quadratic objectives and semidefinite constraints</li></ul><h2 id="[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)"><a class="docs-heading-anchor" href="#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2">Version 0.11.2</a> (January 14, 2016)</a><a id="[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)" title="Permalink"></a></h2><ul><li>Compatibility update for Mosek</li></ul><h2 id="[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)"><a class="docs-heading-anchor" href="#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1">Version 0.11.1</a> (December 1, 2015)</a><a id="[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)" title="Permalink"></a></h2><ul><li>Remove usage of <code>@compat</code> in tests.</li><li>Fix updating quadratic objectives for nonlinear models.</li></ul><h2 id="[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)"><a class="docs-heading-anchor" href="#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0">Version 0.11.0</a> (November 30, 2015)</a><a id="[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)" title="Permalink"></a></h2><ul><li>Julia 0.4.0 is the minimum required version for this release.</li><li>Fix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.</li><li>Addition of the <code>solve(m::Model, relaxation=true)</code> keyword argument to solve the standard continuous relaxation of model <code>m</code></li><li>The <code>getConstraintBounds()</code> method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.</li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3">Version 0.10.3</a> (November 20, 2015)</a><a id="[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)" title="Permalink"></a></h2><ul><li>Fix a rare error when parsing quadratic expressions</li><li>Fix <code>Variable()</code> constructor with default arguments</li><li>Detect unrecognized keywords in <code>solve()</code></li></ul><h2 id="[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2">Version 0.10.2</a> (September 28, 2015)</a><a id="[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)" title="Permalink"></a></h2><ul><li>Fix for deprecation warnings</li></ul><h2 id="[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1">Version 0.10.1</a> (September 3, 2015)</a><a id="[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)" title="Permalink"></a></h2><ul><li>Fixes for ambiguity warnings.</li><li>Fix for breaking change in precompilation syntax in Julia 0.4-pre</li></ul><h2 id="[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)"><a class="docs-heading-anchor" href="#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0">Version 0.10.0</a> (August 31, 2015)</a><a id="[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)" title="Permalink"></a></h2><ul><li>Support (on Julia 0.4 and later) for conditions in indexing <code>@defVar</code> and <code>@addConstraint</code> constructs, for example, <code>@defVar(m, x[i=1:5,j=1:5; i+j &gt;= 3])</code></li><li>Support for vectorized operations on Variables and expressions. See the documentation for details.</li><li>New <code>getVar()</code> method to access variables in a model by name</li><li>Support for semidefinite programming.</li><li>Dual solutions are now available for general nonlinear problems. You may call <code>getDual</code> on a reference object for a nonlinear constraint, and <code>getDual</code> on a variable object for Lagrange multipliers from active bounds.</li><li>Introduce warnings for two common performance traps: too many calls to <code>getValue()</code> on a collection of variables and use of the <code>+</code> operator in a loop to sum expressions.</li><li>Second-order cone constraints can be written directly with the <code>norm()</code> and <code>norm2{}</code> syntax.</li><li>Implement MathProgBase interface for querying Hessian-vector products.</li><li>Iteration over <code>JuMPContainer</code>s is deprecated; instead, use the <code>keys</code> and <code>values</code> functions, and <code>zip(keys(d),values(d))</code> for the old behavior.</li><li><code>@defVar</code> returns <code>Array{Variable,N}</code> when each of <code>N</code> index sets are of the form <code>1:nᵢ</code>.</li><li>Module precompilation: on Julia 0.4 and later, <code>using JuMP</code> is now much faster.</li></ul><h2 id="[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3">Version 0.9.3</a> (August 11, 2015)</a><a id="[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)" title="Permalink"></a></h2><ul><li>Fixes for FactCheck testing on julia v0.4.</li></ul><h2 id="[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2">Version 0.9.2</a> (June 27, 2015)</a><a id="[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)" title="Permalink"></a></h2><ul><li>Fix bug in @addConstraints.</li></ul><h2 id="[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1">Version 0.9.1</a> (April 25, 2015)</a><a id="[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)" title="Permalink"></a></h2><ul><li>Fix for Julia 0.4-dev.</li><li>Small infrastructure improvements for extensions.</li></ul><h2 id="[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)"><a class="docs-heading-anchor" href="#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0">Version 0.9.0</a> (April 18, 2015)</a><a id="[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)" title="Permalink"></a></h2><ul><li>Comparison operators for constructing constraints (for example, <code>2x &gt;= 1</code>) have been deprecated. Instead, construct the constraints explicitly in the <code>@addConstraint</code> macro to add them to the model, or in the <code>@LinearConstraint</code> macro to create a stand-alone linear constraint instance.</li><li><code>getValue()</code> method implemented to compute the value of a nonlinear subexpression</li><li>JuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.</li><li>A number of performance improvements in ReverseDiffSparse for computing derivatives.</li><li><code>MathProgBase.getsolvetime(m)</code> now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)</li><li>Formatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).</li></ul><h2 id="[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)"><a class="docs-heading-anchor" href="#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0">Version 0.8.0</a> (February 17, 2015)</a><a id="[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)" title="Permalink"></a></h2><ul><li>Nonlinear subexpressions now supported with the <code>@defNLExpr</code> macro.</li><li>SCS supported for solving second-order conic problems.</li><li><code>setXXXCallback</code> family deprecated in favor of <code>addXXXCallback</code>.</li><li>Multiple callbacks of the same type can be registered.</li><li>Added support for informational callbacks via <code>addInfoCallback</code>.</li><li>A <code>CallbackAbort</code> exception can be thrown from callback to safely exit optimization.</li></ul><h2 id="[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4">Version 0.7.4</a> (February 4, 2015)</a><a id="[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)" title="Permalink"></a></h2><ul><li>Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.</li><li>Two-sided nonlinear constraints are supported.</li><li>Methods for accessing the number of variables and constraints in a model are renamed.</li><li>New default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.</li><li>Small bug fixes.</li></ul><h2 id="[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3">Version 0.7.3</a> (January 14, 2015)</a><a id="[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)" title="Permalink"></a></h2><ul><li>Fix a method ambiguity conflict with Compose.jl (cosmetic fix)</li></ul><h2 id="[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2">Version 0.7.2</a> (January 9, 2015)</a><a id="[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in <code>sum(::JuMPDict)</code></li><li>Added the <code>setCategory</code> function to change a variables category (for example, continuous or binary)</li></ul><p>after construction, and <code>getCategory</code> to retrieve the variable category.</p><h2 id="[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)"><a class="docs-heading-anchor" href="#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1">Version 0.7.1</a> (January 2, 2015)</a><a id="[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)" title="Permalink"></a></h2><ul><li>Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.</li></ul><h2 id="[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)"><a class="docs-heading-anchor" href="#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0">Version 0.7.0</a> (December 29, 2014)</a><a id="[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)" title="Permalink"></a></h2><h3 id="Linear/quadratic/conic-programming"><a class="docs-heading-anchor" href="#Linear/quadratic/conic-programming">Linear/quadratic/conic programming</a><a id="Linear/quadratic/conic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Linear/quadratic/conic-programming" title="Permalink"></a></h3><ul><li><strong>Breaking change</strong>: The syntax for column-wise model generation has been changed to use keyword arguments in <code>@defVar</code>.</li><li>On Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.</li><li>ECOS supported for solving second-order conic problems.</li></ul><h3 id="_nonlinear_programming_release_notes"><a class="docs-heading-anchor" href="#_nonlinear_programming_release_notes">Nonlinear programming</a><a id="_nonlinear_programming_release_notes-1"></a><a class="docs-heading-anchor-permalink" href="#_nonlinear_programming_release_notes" title="Permalink"></a></h3><ul><li>Support for skipping model generation when solving a sequence of nonlinear models with changing data.</li><li>Fix a memory leak when solving a sequence of nonlinear models.</li><li>The <code>@addNLConstraint</code> macro now supports the three-argument version to define sets of nonlinear constraints.</li><li>KNITRO supported as a nonlinear solver.</li><li>Speed improvements for model generation.</li><li>The <code>@addNLConstraints</code> macro supports adding multiple (groups of) constraints at once. Syntax is similar to <code>@addConstraints</code>.</li><li>Discrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).</li></ul><h3 id="General"><a class="docs-heading-anchor" href="#General">General</a><a id="General-1"></a><a class="docs-heading-anchor-permalink" href="#General" title="Permalink"></a></h3><ul><li>Starting values for variables may now be specified with <code>@defVar(m, x, start=value)</code>.</li><li>The <code>setSolver</code> function allows users to change the solver subsequent to model creation.</li><li>Support for &quot;fixed&quot; variables via the <code>@defVar(m, x == 1)</code> syntax.</li><li>Unit tests rewritten to use FactCheck.jl, improved testing across solvers.</li></ul><h2 id="[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3">Version 0.6.3</a> (October 19, 2014)</a><a id="[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in multiplying two AffExpr objects.</li></ul><h2 id="[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2">Version 0.6.2</a> (October 11, 2014)</a><a id="[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)" title="Permalink"></a></h2><ul><li>Further improvements and bug fixes for printing.</li><li>Fixed a bug in <code>@defExpr</code>.</li><li>Support for accessing expression graphs through the MathProgBase NLP interface.</li></ul><h2 id="[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1">Version 0.6.1</a> (September 19, 2014)</a><a id="[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)" title="Permalink"></a></h2><ul><li>Improvements and bug fixes for printing.</li></ul><h2 id="[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)"><a class="docs-heading-anchor" href="#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0">Version 0.6.0</a> (September 9, 2014)</a><a id="[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)" title="Permalink"></a></h2><ul><li>Julia 0.3.0 is the minimum required version for this release.</li><li><code>buildInternalModel(m::Model)</code> added to build solver-level model in memory without optimizing.</li><li>Deprecate <code>load_model_only</code> keyword argument to <code>solve</code>.</li><li>Add groups of constraints with <code>@addConstraints</code> macro.</li><li>Unicode operators now supported, including <code>∑</code> for <code>sum</code>, <code>∏</code> for <code>prod</code>, and <code>≤</code>/<code>≥</code></li><li>Quadratic constraints supported in <code>@addConstraint</code> macro.</li><li>Quadratic objectives supported in <code>@setObjective</code> macro.</li><li>MathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems<ul><li><strong>Breaking change</strong>: <code>IpoptOptions</code> no longer supported to specify solver options, use <code>m = Model(solver=IpoptSolver(options...))</code> instead.</li></ul></li><li>New solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK</li><li>New option to control whether the lazy constraint callback is executed at each node in the B&amp;B tree or just when feasible solutions are found</li><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.</li><li>Add support for index dependencies (for example, triangular indexing) in <code>@defVar</code>, <code>@addConstraint</code>, and <code>@defExpr</code> (for example, <code>@defVar(m, x[i=1:10,j=i:10])</code>).<ul><li>This required some changes to the internal structure of JuMP containers, which may break code that explicitly stored <code>JuMPDict</code> objects.</li></ul></li></ul><h2 id="[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8">Version 0.5.8</a> (September 24, 2014)</a><a id="[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug with specifying solvers (affects Julia 0.2 only)</li></ul><h2 id="[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7">Version 0.5.7</a> (September 5, 2014)</a><a id="[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in printing models</li></ul><h2 id="[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6">Version 0.5.6</a> (September 2, 2014)</a><a id="[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)" title="Permalink"></a></h2><ul><li>Add support for semicontinuous and semi-integer variables for those solvers that support them.<ul><li><strong>Breaking change</strong>: Syntax for <code>Variable()</code> constructor has changed (use of this interface remains discouraged)</li></ul></li><li>Update for breaking changes in MathProgBase</li></ul><h2 id="[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5">Version 0.5.5</a> (July 6, 2014)</a><a id="[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)" title="Permalink"></a></h2><ul><li>Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.</li></ul><h2 id="[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4">Version 0.5.4</a> (June 19, 2014)</a><a id="[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in MathProgBase which reduces loading times for <code>using JuMP</code></li><li>Fix error when MIPs not solved to optimality</li></ul><h2 id="[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3">Version 0.5.3</a> (May 21, 2014)</a><a id="[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)" title="Permalink"></a></h2><ul><li>Update for breaking change in ReverseDiffSparse</li></ul><h2 id="[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2">Version 0.5.2</a> (May 9, 2014)</a><a id="[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)" title="Permalink"></a></h2><ul><li>Fix compatibility with Julia 0.3 prerelease</li></ul><h2 id="[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1">Version 0.5.1</a> (May 5, 2014)</a><a id="[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)" title="Permalink"></a></h2><ul><li>Fix a bug in coefficient handling inside lazy constraints and user cuts</li></ul><h2 id="[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)"><a class="docs-heading-anchor" href="#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0">Version 0.5.0</a> (May 2, 2014)</a><a id="[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)" title="Permalink"></a></h2><ul><li>Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.</li><li><code>getValue</code> for <code>AffExpr</code> and <code>QuadExpr</code></li><li><strong>Breaking change</strong>: <code>getSolverModel</code> replaced by <code>getInternalModel</code>, which returns the internal MathProgBase-level model</li><li>Groups of constraints can be specified with <code>@addConstraint</code> (see documentation for details). This is not a breaking change.</li><li><code>dot(::JuMPDict{Variable},::JuMPDict{Variable})</code> now returns the corresponding quadratic expression.</li></ul><h2 id="[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)"><a class="docs-heading-anchor" href="#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1">Version 0.4.1</a> (March 24, 2014)</a><a id="[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)" title="Permalink"></a></h2><ul><li>Fix bug where change in objective sense was ignored when re-solving a model.</li><li>Fix issue with handling zero coefficients in AffExpr.</li></ul><h2 id="[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)"><a class="docs-heading-anchor" href="#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0">Version 0.4.0</a> (March 10, 2014)</a><a id="[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)" title="Permalink"></a></h2><ul><li>Support for SOS1 and SOS2 constraints.</li><li>Solver-independent callback for user heuristics.</li><li><code>dot</code> and <code>sum</code> implemented for <code>JuMPDict</code> objects. Now you can say <code>@addConstraint(m, dot(a,x) &lt;= b)</code>.</li><li>Developers: support for extensions to JuMP. See definition of Model in <code>src/JuMP.jl</code> for more details.</li><li>Option to construct the low-level model before optimizing.</li></ul><h2 id="[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2">Version 0.3.2</a> (February 17, 2014)</a><a id="[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)" title="Permalink"></a></h2><ul><li>Improved model printing<ul><li>Preliminary support for IJulia output</li></ul></li></ul><h2 id="[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1">Version 0.3.1</a> (January 30, 2014)</a><a id="[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)" title="Permalink"></a></h2><ul><li>Documentation updates</li><li>Support for MOSEK</li><li>CPLEXLink renamed to CPLEX</li></ul><h2 id="[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)"><a class="docs-heading-anchor" href="#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0">Version 0.3.0</a> (January 21, 2014)</a><a id="[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)" title="Permalink"></a></h2><ul><li>Unbounded/infeasibility rays: <code>getValue()</code> will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. <code>getDual()</code> will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.</li><li>Solver-independent callbacks for user generated cuts.</li><li>Use new interface for solver-independent QCQP.</li><li><code>setlazycallback</code> renamed to <code>setLazyCallback</code> for consistency.</li></ul><h2 id="[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)"><a class="docs-heading-anchor" href="#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0">Version 0.2.0</a> (December 15, 2013)</a><a id="[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)" title="Permalink"></a></h2><h3 id="Breaking-8"><a class="docs-heading-anchor" href="#Breaking-8">Breaking</a><a class="docs-heading-anchor-permalink" href="#Breaking-8" title="Permalink"></a></h3><ul><li>Objective sense is specified in <code>setObjective</code> instead of in the <code>Model</code> constructor.</li><li><code>lpsolver</code> and <code>mipsolver</code> merged into single <code>solver</code> option.</li></ul><h3 id="Added-31"><a class="docs-heading-anchor" href="#Added-31">Added</a><a class="docs-heading-anchor-permalink" href="#Added-31" title="Permalink"></a></h3><ul><li>Problem modification with efficient LP restarts and MIP warm-starts.</li><li>Relatedly, column-wise modeling now supported.</li><li>Solver-independent callbacks supported. Currently we support only a &quot;lazy constraint&quot; callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.</li></ul><h2 id="[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2">Version 0.1.2</a> (November 16, 2013)</a><a id="[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)" title="Permalink"></a></h2><ul><li>Bug fixes for printing, improved error messages.</li><li>Allow <code>AffExpr</code> to be used in macros; for example, <code>ex = y + z; @addConstraint(m, x + 2*ex &lt;= 3)</code></li></ul><h2 id="[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1">Version 0.1.1</a> (October 23, 2013)</a><a id="[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)" title="Permalink"></a></h2><ul><li>Update for solver specification API changes in MathProgBase.</li></ul><h2 id="[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)"><a class="docs-heading-anchor" href="#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)"><a href="https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0">Version 0.1.0</a> (October 3, 2013)</a><a id="[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)-1"></a><a class="docs-heading-anchor-permalink" href="#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)" title="Permalink"></a></h2><ul><li>Initial public release.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../moi/release_notes/">« Release notes</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/search_index.js b/previews/PR3562/search_index.js
index c1231e715a7..9b17509c3ae 100644
--- a/previews/PR3562/search_index.js
+++ b/previews/PR3562/search_index.js
@@ -1,3 +1,3 @@
 var documenterSearchIndex = {"docs":
-[{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/nonlinear.md\"","category":"page"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/nonlinear/#Nonlinear-programming","page":"Nonlinear programming","title":"Nonlinear programming","text":"","category":"section"},{"location":"moi/reference/nonlinear/#Types","page":"Nonlinear programming","title":"Types","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"AbstractNLPEvaluator\nNLPBoundsPair\nNLPBlockData","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.AbstractNLPEvaluator","page":"Nonlinear programming","title":"MathOptInterface.AbstractNLPEvaluator","text":"AbstractNLPEvaluator\n\nAbstract supertype for the callback object that is used to query function values, derivatives, and expression graphs.\n\nIt is used in NLPBlockData.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBoundsPair","page":"Nonlinear programming","title":"MathOptInterface.NLPBoundsPair","text":"NLPBoundsPair(lower::Float64, upper::Float64)\n\nA struct holding a pair of lower and upper bounds.\n\n-Inf and Inf can be used to indicate no lower or upper bound, respectively.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockData","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockData","text":"struct NLPBlockData\n    constraint_bounds::Vector{NLPBoundsPair}\n    evaluator::AbstractNLPEvaluator\n    has_objective::Bool\nend\n\nA struct encoding a set of nonlinear constraints of the form lb le g(x) le ub and, if has_objective == true, a nonlinear objective function f(x).\n\nNonlinear objectives override any objective set by using the ObjectiveFunction attribute.\n\nThe evaluator is a callback object that is used to query function values, derivatives, and expression graphs. If has_objective == false, then it is an error to query properties of the objective function, and in Hessian-of-the-Lagrangian queries, σ must be set to zero.\n\nnote: Note\nThroughout the evaluator, all variables are ordered according to ListOfVariableIndices. Hence, MOI copies of nonlinear problems must not re-order variables.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#Attributes","page":"Nonlinear programming","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"NLPBlock\nNLPBlockDual\nNLPBlockDualStart","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlock","page":"Nonlinear programming","title":"MathOptInterface.NLPBlock","text":"NLPBlock()\n\nAn AbstractModelAttribute that stores an NLPBlockData, representing a set of nonlinear constraints, and optionally a nonlinear objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockDual","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockDual","text":"NLPBlockDual(result_index::Int = 1)\n\nAn AbstractModelAttribute for the Lagrange multipliers on the constraints from the NLPBlock in result result_index.\n\nIf result_index is omitted, it is 1 by default.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockDualStart","text":"NLPBlockDualStart()\n\nAn AbstractModelAttribute for the initial assignment of the Lagrange multipliers on the constraints from the NLPBlock that the solver may use to warm-start the solve.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#Functions","page":"Nonlinear programming","title":"Functions","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"initialize\nfeatures_available\neval_objective\neval_constraint\neval_objective_gradient\njacobian_structure\neval_constraint_gradient\nconstraint_gradient_structure\neval_constraint_jacobian\neval_constraint_jacobian_product\neval_constraint_jacobian_transpose_product\nhessian_lagrangian_structure\nhessian_objective_structure\nhessian_constraint_structure\neval_hessian_objective\neval_hessian_constraint\neval_hessian_lagrangian\neval_hessian_lagrangian_product\nobjective_expr\nconstraint_expr","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.initialize","page":"Nonlinear programming","title":"MathOptInterface.initialize","text":"initialize(\n    d::AbstractNLPEvaluator,\n    requested_features::Vector{Symbol},\n)::Nothing\n\nInitialize d with the set of features in requested_features. Check features_available before calling initialize to see what features are supported by d.\n\nwarning: Warning\nThis method must be called before any other methods.\n\nFeatures\n\nThe following features are defined:\n\n:Grad: enables eval_objective_gradient\n:Jac: enables eval_constraint_jacobian\n:JacVec: enables eval_constraint_jacobian_product and eval_constraint_jacobian_transpose_product\n:Hess: enables eval_hessian_lagrangian\n:HessVec: enables eval_hessian_lagrangian_product\n:ExprGraph: enables objective_expr and constraint_expr.\n\nIn all cases, including when requested_features is empty, eval_objective and eval_constraint are supported.\n\nExamples\n\nMOI.initialize(d, Symbol[])\nMOI.initialize(d, [:ExprGraph])\nMOI.initialize(d, MOI.features_available(d))\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.features_available","page":"Nonlinear programming","title":"MathOptInterface.features_available","text":"features_available(d::AbstractNLPEvaluator)::Vector{Symbol}\n\nReturns the subset of features available for this problem instance.\n\nSee initialize for the list of defined features.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_objective","page":"Nonlinear programming","title":"MathOptInterface.eval_objective","text":"eval_objective(d::AbstractNLPEvaluator, x::AbstractVector{T})::T where {T}\n\nEvaluate the objective f(x), returning a scalar value.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint","text":"eval_constraint(d::AbstractNLPEvaluator,\n    g::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nGiven a set of vector-valued constraints l le g(x) le u, evaluate the constraint function g(x), storing the result in the vector g.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that g is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_objective_gradient","page":"Nonlinear programming","title":"MathOptInterface.eval_objective_gradient","text":"eval_objective_gradient(\n    d::AbstractNLPEvaluator,\n    grad::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nEvaluate the gradient of the objective function grad = nabla f(x) as a dense vector, storing the result in the vector grad.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that grad is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.jacobian_structure","page":"Nonlinear programming","title":"MathOptInterface.jacobian_structure","text":"jacobian_structure(d::AbstractNLPEvaluator)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Jacobian matrix: J_g(x) = left beginarrayc nabla g_1(x)  nabla g_2(x)  vdots  nabla g_m(x) endarrayright where g_i is the itextth component of the nonlinear constraints g(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_gradient","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_gradient","text":"eval_constraint_gradient(\n    d::AbstractNLPEvaluator,\n    ∇g::AbstractVector{T},\n    x::AbstractVector{T},\n    i::Int,\n)::Nothing where {T}\n\nEvaluate the gradient of constraint i, nabla g_i(x), and store the non-zero values in ∇g, corresponding to the structure returned by constraint_gradient_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that ∇g is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.constraint_gradient_structure","page":"Nonlinear programming","title":"MathOptInterface.constraint_gradient_structure","text":"constraint_gradient_structure(d::AbstractNLPEvaluator, i::Int)::Vector{Int64}\n\nReturns a vector of indices, where each element indicates the position of a structurally nonzero element in the gradient of constraint nabla g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian","text":"eval_constraint_jacobian(d::AbstractNLPEvaluator,\n    J::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nEvaluates the sparse Jacobian matrix J_g(x) = left beginarrayc nabla g_1(x)  nabla g_2(x)  vdots  nabla g_m(x) endarrayright.\n\nThe result is stored in the vector J in the same order as the indices returned by jacobian_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that J is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian_product","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian_product","text":"eval_constraint_jacobian_product(\n    d::AbstractNLPEvaluator,\n    y::AbstractVector{T},\n    x::AbstractVector{T},\n    w::AbstractVector{T},\n)::Nothing where {T}\n\nComputes the Jacobian-vector product y = J_g(x)w, storing the result in the vector y.\n\nThe vectors have dimensions such that length(w) == length(x), and length(y) is the number of nonlinear constraints.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that y is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian_transpose_product","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian_transpose_product","text":"eval_constraint_jacobian_transpose_product(\n    d::AbstractNLPEvaluator,\n    y::AbstractVector{T},\n    x::AbstractVector{T},\n    w::AbstractVector{T},\n)::Nothing where {T}\n\nComputes the Jacobian-transpose-vector product y = J_g(x)^Tw, storing the result in the vector y.\n\nThe vectors have dimensions such that length(y) == length(x), and length(w) is the number of nonlinear constraints.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that y is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_lagrangian_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_lagrangian_structure","text":"hessian_lagrangian_structure(\n    d::AbstractNLPEvaluator,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian-of-the-Lagrangian matrix: nabla^2 f(x) + sum_i=1^m nabla^2 g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_objective_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_objective_structure","text":"hessian_objective_structure(\n    d::AbstractNLPEvaluator,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian matrix: nabla^2 f(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_constraint_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_constraint_structure","text":"hessian_constraint_structure(\n    d::AbstractNLPEvaluator,\n    i::Int64,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian matrix: nabla^2 g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_objective","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_objective","text":"eval_hessian_objective(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nThis function computes the sparse Hessian matrix: nabla^2 f(x), storing the result in the vector H in the same order as the indices returned by hessian_objective_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_constraint","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_constraint","text":"eval_hessian_constraint(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n    i::Int64,\n)::Nothing where {T}\n\nThis function computes the sparse Hessian matrix: nabla^2 g_i(x), storing the result in the vector H in the same order as the indices returned by hessian_constraint_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_lagrangian","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_lagrangian","text":"eval_hessian_lagrangian(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n    σ::T,\n    μ::AbstractVector{T},\n)::Nothing where {T}\n\nGiven scalar weight σ and vector of constraint weights μ, this function computes the sparse Hessian-of-the-Lagrangian matrix: sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x), storing the result in the vector H in the same order as the indices returned by hessian_lagrangian_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_lagrangian_product","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_lagrangian_product","text":"eval_hessian_lagrangian_product(\n    d::AbstractNLPEvaluator,\n    h::AbstractVector{T},\n    x::AbstractVector{T},\n    v::AbstractVector{T},\n    σ::T,\n    μ::AbstractVector{T},\n)::Nothing where {T}\n\nGiven scalar weight σ and vector of constraint weights μ, computes the Hessian-of-the-Lagrangian-vector product h = left(sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x)right)v, storing the result in the vector h.\n\nThe vectors have dimensions such that length(h) == length(x) == length(v).\n\nImplementation notes\n\nWhen implementing this method, you must not assume that h is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.objective_expr","page":"Nonlinear programming","title":"MathOptInterface.objective_expr","text":"objective_expr(d::AbstractNLPEvaluator)::Expr\n\nReturns a Julia Expr object representing the expression graph of the objective function.\n\nFormat\n\nThe expression has a number of limitations, compared with arbitrary Julia expressions:\n\nAll sums and products are flattened out as simple Expr(:+, ...) and Expr(:*, ...) objects.\nAll decision variables must be of the form Expr(:ref, :x, MOI.VariableIndex(i)), where i is the ith variable in ListOfVariableIndices.\nThere are currently no restrictions on recognized functions; typically these will be built-in Julia functions like ^, exp, log, cos, tan, sqrt, etc., but modeling interfaces may choose to extend these basic functions, or error if they encounter unsupported functions.\n\nExamples\n\nThe expression x_1+sin(x_2exp(x_3)) is represented as\n\n:(x[MOI.VariableIndex(1)] + sin(x[MOI.VariableIndex(2)] / exp(x[MOI.VariableIndex[3]])))\n\nor equivalently\n\nExpr(\n    :call,\n    :+,\n    Expr(:ref, :x, MOI.VariableIndex(1)),\n    Expr(\n        :call,\n        :/,\n        Expr(:call, :sin, Expr(:ref, :x, MOI.VariableIndex(2))),\n        Expr(:call, :exp, Expr(:ref, :x, MOI.VariableIndex(3))),\n    ),\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.constraint_expr","page":"Nonlinear programming","title":"MathOptInterface.constraint_expr","text":"constraint_expr(d::AbstractNLPEvaluator, i::Integer)::Expr\n\nReturns a Julia Expr object representing the expression graph for the itextth nonlinear constraint.\n\nFormat\n\nThe format is the same as objective_expr, with an additional comparison operator indicating the sense of and bounds on the constraint.\n\nFor single-sided comparisons, the body of the constraint must be on the left-hand side, and the right-hand side must be a constant.\n\nFor double-sided comparisons (that is, l le f(x) le u), the body of the constraint must be in the middle, and the left- and right-hand sides must be constants.\n\nThe bounds on the constraints must match the NLPBoundsPairs passed to NLPBlockData.\n\nExamples\n\n:(x[MOI.VariableIndex(1)]^2 <= 1.0)\n:(x[MOI.VariableIndex(1)]^2 >= 2.0)\n:(x[MOI.VariableIndex(1)]^2 == 3.0)\n:(4.0 <= x[MOI.VariableIndex(1)]^2 <= 5.0)\n\n\n\n\n\n","category":"function"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"EditURL = \"https://github.com/jump-dev/Cbc.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"(Image: )","category":"page"},{"location":"packages/Cbc/#Cbc.jl","page":"jump-dev/Cbc.jl","title":"Cbc.jl","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.jl is a wrapper for the COIN-OR Branch and Cut (Cbc) solver.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Cbc/#Affiliation","page":"jump-dev/Cbc.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Cbc/#License","page":"jump-dev/Cbc.jl","title":"License","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The underlying solver, coin-or/Cbc, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Cbc/#Installation","page":"jump-dev/Cbc.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Install Cbc using Pkg.add:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"import Pkg\nPkg.add(\"Cbc\")","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"In addition to installing the Cbc.jl package, this will also download and install the Cbc binaries. You do not need to install Cbc separately.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Cbc/#Use-with-JuMP","page":"jump-dev/Cbc.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"To use Cbc with JuMP, use Cbc.Optimizer:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"using JuMP, Cbc\nmodel = Model(Cbc.Optimizer)\nset_attribute(model, \"logLevel\", 1)","category":"page"},{"location":"packages/Cbc/#MathOptInterface-API","page":"jump-dev/Cbc.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The COIN Branch-and-Cut (Cbc) optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.Status\nCbc.SecondaryStatus\nMOI.DualStatus\nMOI.NodeCount\nMOI.NumberOfVariables\nMOI.ObjectiveBound\nMOI.ObjectiveSense\nMOI.ObjectiveValue\nMOI.PrimalStatus\nMOI.RelativeGap\nMOI.ResultCount\nMOI.SolveTimeSec\nMOI.TerminationStatus","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported optimizer attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.SetVariableNames\nMOI.AbsoluteGapTolerance\nMOI.NumberOfThreads\nMOI.RawOptimizerAttribute\nMOI.RelativeGapTolerance\nMOI.Silent\nMOI.SolverName\nMOI.SolverVersion\nMOI.TimeLimitSec","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported variable attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.VariablePrimal\nMOI.VariablePrimalStart\nMOI.VariableName","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported constraint attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ConstraintPrimal","category":"page"},{"location":"packages/Cbc/#Options","page":"jump-dev/Cbc.jl","title":"Options","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Options are, unfortunately, not well documented.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The following options are likely to be the most useful:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Parameter Example Explanation\nseconds 60.0 Solution timeout limit\nlogLevel 2 Set to 0 to disable solution output\nmaxSolutions 1 Terminate after this many feasible solutions have been found\nmaxNodes 1 Terminate after this many branch-and-bound nodes have been evaluated\nallowableGap 0.05 Terminate after optimality gap is less than this value (on an absolute scale)\nratioGap 0.05 Terminate after optimality gap is smaller than this relative fraction\nthreads 1 Set the number of threads to use for parallel branch & bound","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The complete list of parameters can be found by running the cbc executable and typing ? at the prompt.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Start the cbc executable from Julia as follows:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"using Cbc_jll\nCbc_jll.cbc() do exe\n    run(`$(exe)`)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"EditURL = \"geographic_clustering.jl\"","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Geographical-clustering","page":"Geographical clustering","title":"Geographical clustering","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial was originally contributed by Matthew Helm and Mathieu Tanneau.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"The goal of this exercise is to cluster n cities into k groups, minimizing the total pairwise distance between cities and ensuring that the variance in the total populations of each group is relatively small.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"For this example, we'll use the 20 most populous cities in the United States.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"cities = DataFrames.DataFrame(\n    Union{String,Float64}[\n        \"New York, NY\" 8.405 40.7127 -74.0059\n        \"Los Angeles, CA\" 3.884 34.0522 -118.2436\n        \"Chicago, IL\" 2.718 41.8781 -87.6297\n        \"Houston, TX\" 2.195 29.7604 -95.3698\n        \"Philadelphia, PA\" 1.553 39.9525 -75.1652\n        \"Phoenix, AZ\" 1.513 33.4483 -112.0740\n        \"San Antonio, TX\" 1.409 29.4241 -98.4936\n        \"San Diego, CA\" 1.355 32.7157 -117.1610\n        \"Dallas, TX\" 1.257 32.7766 -96.7969\n        \"San Jose, CA\" 0.998 37.3382 -121.8863\n        \"Austin, TX\" 0.885 30.2671 -97.7430\n        \"Indianapolis, IN\" 0.843 39.7684 -86.1580\n        \"Jacksonville, FL\" 0.842 30.3321 -81.6556\n        \"San Francisco, CA\" 0.837 37.7749 -122.4194\n        \"Columbus, OH\" 0.822 39.9611 -82.9987\n        \"Charlotte, NC\" 0.792 35.2270 -80.8431\n        \"Fort Worth, TX\" 0.792 32.7554 -97.3307\n        \"Detroit, MI\" 0.688 42.3314 -83.0457\n        \"El Paso, TX\" 0.674 31.7775 -106.4424\n        \"Memphis, TN\" 0.653 35.1495 -90.0489\n    ],\n    [\"city\", \"population\", \"lat\", \"lon\"],\n)","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Model-Specifics","page":"Geographical clustering","title":"Model Specifics","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We will cluster these 20 cities into 3 different groups and we will assume that the ideal or target population P for a group is simply the total population of the 20 cities divided by 3:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"n = size(cities, 1)\nk = 3\nP = sum(cities.population) / k","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Obtaining-the-distances-between-each-city","page":"Geographical clustering","title":"Obtaining the distances between each city","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Let's compute the pairwise Haversine distance between each of the cities in our data set and store the result in a variable we'll call dm:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"\"\"\"\n    haversine(lat1, long1, lat2, long2, r = 6372.8)\n\nCompute the haversine distance between two points on a sphere of radius `r`,\nwhere the points are given by the latitude/longitude pairs `lat1/long1` and\n`lat2/long2` (in degrees).\n\"\"\"\nfunction haversine(lat1, long1, lat2, long2, r = 6372.8)\n    lat1, long1 = deg2rad(lat1), deg2rad(long1)\n    lat2, long2 = deg2rad(lat2), deg2rad(long2)\n    hav(a, b) = sin((b - a) / 2)^2\n    inner_term = hav(lat1, lat2) + cos(lat1) * cos(lat2) * hav(long1, long2)\n    d = 2 * r * asin(sqrt(inner_term))\n    # Round distance to nearest kilometer.\n    return round(Int, d)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Our distance matrix is symmetric so we'll convert it to a LowerTriangular matrix so that we can better interpret the objective value of our model:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"dm = LinearAlgebra.LowerTriangular([\n    haversine(cities.lat[i], cities.lon[i], cities.lat[j], cities.lon[j])\n    for i in 1:n, j in 1:n\n])","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Build-the-model","page":"Geographical clustering","title":"Build the model","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now that we have the basics taken  care of, we can set up our model, create decision variables, add constraints, and then solve.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"First, we'll set up a model that leverages the Cbc solver. Next, we'll set up a binary variable x_ik that takes the value 1 if city i is in group k and 0 otherwise. Each city must be in a group, so we'll add the constraint sum_k x_ik = 1 for every i.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:n, 1:k], Bin)\n@constraint(model, [i = 1:n], sum(x[i, :]) == 1);\n# To reduce symmetry, we fix the first city to belong to the first group.\nfix(x[1, 1], 1; force = true)","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"The total population of a group k is Q_k = sum_ix_ikq_i where q_i is simply the i-th value from the population column in our cities DataFrame. Let's add constraints so that alpha leq (Q_k - P) leq beta. We'll set alpha equal to -3 million and beta equal to 3. By adjusting these thresholds you'll find that there is a tradeoff between having relatively even populations between groups and having geographically close cities within each group. In other words, the larger the absolute values of alpha and beta, the closer together the cities in a group will be but the variance between the group populations will be higher.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@variable(model, -3 <= population_diff[1:k] <= 3)\n@constraint(model, population_diff .== x' * cities.population .- P)","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now we need to add one last binary variable z_ij to our model that we'll use to compute the total distance between the cities in our groups, defined as sum_ijd_ijz_ij. Variable z_ij will equal 1 if cities i and j are in the same group, and 0 if they are not in the same group.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"To ensure that z_ij = 1 if and only if cities i and j are in the same group, we add the constraints z_ij geq x_ik + x_jk - 1 for every pair ij and every k:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@variable(model, z[i = 1:n, j = 1:i], Bin)\nfor k in 1:k, i in 1:n, j in 1:i\n    @constraint(model, z[i, j] >= x[i, k] + x[j, k] - 1)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We can now add an objective to our model which will simply be to minimize the dot product of z and our distance matrix, dm.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@objective(model, Min, sum(dm[i, j] * z[i, j] for i in 1:n, j in 1:i));\nnothing #hide","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We can then call optimize! and review the results.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"optimize!(model)","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Reviewing-the-Results","page":"Geographical clustering","title":"Reviewing the Results","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now that we have results, we can add a column to our cities DataFrame for the group and then loop through our x variable to assign each city to its group. Once we have that, we can look at the total population for each group and also look at the cities in each group to verify that they are grouped by geographic proximity.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"cities.group = zeros(n)\n\nfor i in 1:n, j in 1:k\n    if round(Int, value(x[i, j])) == 1\n        cities.group[i] = j\n    end\nend\n\nfor group in DataFrames.groupby(cities, :group)\n    @show group\n    println(\"\")\n    @show sum(group.population)\n    println(\"\")\nend","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/objective/#Objectives","page":"Objectives","title":"Objectives","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"This page describes macros and functions related to linear and quadratic objective functions only, unless otherwise indicated. For nonlinear objective functions, see Nonlinear Modeling.","category":"page"},{"location":"manual/objective/#Set-a-linear-objective","page":"Objectives","title":"Set a linear objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a linear objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use Min to create a minimization objective:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use Max to create a maximization objective:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Max, 2x + 1)\n2 x + 1","category":"page"},{"location":"manual/objective/#Set-a-quadratic-objective","page":"Objectives","title":"Set a quadratic objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a quadratic objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use ^2 to have a variable squared:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, x^2 + 2x + 1)\nx² + 2 x + 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"You can also have bilinear terms between variables:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @objective(model, Max, x * y + x + y)\nx*y + x + y","category":"page"},{"location":"manual/objective/#Set-a-nonlinear-objective","page":"Objectives","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a nonlinear objective function:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x <= 1);\n\njulia> @objective(model, Max, log(x))\nlog(x)","category":"page"},{"location":"manual/objective/#Query-the-objective-function","page":"Objectives","title":"Query the objective function","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use objective_function to return the current objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_function(model)\n2 x + 1","category":"page"},{"location":"manual/objective/#Evaluate-the-objective-function-at-a-point","page":"Objectives","title":"Evaluate the objective function at a point","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use value to evaluate an objective function at a point specifying values for variables.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, 2x[1]^2 + x[1] + 0.5*x[2])\n2 x[1]² + x[1] + 0.5 x[2]\n\njulia> f = objective_function(model)\n2 x[1]² + x[1] + 0.5 x[2]\n\njulia> point = Dict(x[1] => 2.0, x[2] => 1.0);\n\njulia> value(z -> point[z], f)\n10.5","category":"page"},{"location":"manual/objective/#Query-the-objective-sense","page":"Objectives","title":"Query the objective sense","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use objective_sense to return the current objective sense.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_sense(model)\nMIN_SENSE::OptimizationSense = 0","category":"page"},{"location":"manual/objective/#Modify-an-objective","page":"Objectives","title":"Modify an objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"To modify an objective, call @objective with the new objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> @objective(model, Max, -2x)\n-2 x","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Alternatively, use set_objective_function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> new_objective = @expression(model, -2 * x)\n-2 x\n\njulia> set_objective_function(model, new_objective)","category":"page"},{"location":"manual/objective/#Modify-an-objective-coefficient","page":"Objectives","title":"Modify an objective coefficient","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use set_objective_coefficient to modify an objective coefficient.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> set_objective_coefficient(model, x, 3)\n\njulia> objective_function(model)\n3 x","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"info: Info\nThere is no way to modify the coefficient of a quadratic term. Set a new objective instead.","category":"page"},{"location":"manual/objective/#Modify-the-objective-sense","page":"Objectives","title":"Modify the objective sense","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use set_objective_sense to modify the objective sense.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> objective_sense(model)\nMIN_SENSE::OptimizationSense = 0\n\njulia> set_objective_sense(model, MAX_SENSE);\n\njulia> objective_sense(model)\nMAX_SENSE::OptimizationSense = 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Alternatively, call @objective and pass the existing objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> @objective(model, Max, objective_function(model))\n2 x","category":"page"},{"location":"manual/objective/#Set-a-vector-valued-objective","page":"Objectives","title":"Set a vector-valued objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Define a multi-objective optimization problem by passing a vector of objectives:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, [1 + x[1], 2 * x[2]])\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[2]\n\njulia> f = objective_function(model)\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[2]","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"tip: Tip\nThe Multi-objective knapsack tutorial provides an example of solving a multi-objective integer program.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"In most cases, multi-objective optimization solvers will return multiple solutions, corresponding to points on the Pareto frontier. See Multiple solutions for information on how to query and work with multiple solutions.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Note that you must set a single objective sense, that is, you cannot have both minimization and maximization objectives. Work around this limitation by choosing Min and negating any objectives you want to maximize:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @expression(model, obj1, 1 + x[1])\nx[1] + 1\n\njulia> @expression(model, obj2, 2 * x[1])\n2 x[1]\n\njulia> @objective(model, Min, [obj1, -obj2])\n2-element Vector{AffExpr}:\n x[1] + 1\n -2 x[1]","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Defining your objectives as expressions allows flexibility in how you can solve variations of the same problem, with some objectives removed and constrained to be no worse that a fixed value.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @expression(model, obj1, 1 + x[1])\nx[1] + 1\n\njulia> @expression(model, obj2, 2 * x[1])\n2 x[1]\n\njulia> @expression(model, obj3, x[1] + x[2])\nx[1] + x[2]\n\njulia> @objective(model, Min, [obj1, obj2, obj3])  # Three-objective problem\n3-element Vector{AffExpr}:\n x[1] + 1\n 2 x[1]\n x[1] + x[2]\n\njulia> # optimize!(model), look at the solution, talk to stakeholders, then\n       # decide you want to solve a new problem where the third objective is\n       # removed and constrained to be better than 2.0.\n       nothing\n\njulia> @objective(model, Min, [obj1, obj2])   # Two-objective problem\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[1]\n\njulia> @constraint(model, obj3 <= 2.0)\nx[1] + x[2] ≤ 2","category":"page"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/callbacks.md\"","category":"page"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/callbacks/#Callbacks","page":"Callbacks","title":"Callbacks","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"AbstractCallback\nAbstractSubmittable\nsubmit","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.AbstractCallback","page":"Callbacks","title":"MathOptInterface.AbstractCallback","text":"abstract type AbstractCallback <: AbstractModelAttribute end\n\nAbstract type for a model attribute representing a callback function. The value set to subtypes of AbstractCallback is a function that may be called during optimize!. As optimize! is in progress, the result attributes (i.e, the attributes attr such that is_set_by_optimize(attr)) may not be accessible from the callback, hence trying to get result attributes might throw a OptimizeInProgress error.\n\nAt most one callback of each type can be registered. If an optimizer already has a function for a callback type, and the user registers a new function, then the old one is replaced.\n\nThe value of the attribute should be a function taking only one argument, commonly called callback_data, that can be used for instance in LazyConstraintCallback, HeuristicCallback and UserCutCallback.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.AbstractSubmittable","page":"Callbacks","title":"MathOptInterface.AbstractSubmittable","text":"AbstractSubmittable\n\nAbstract supertype for objects that can be submitted to the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.submit","page":"Callbacks","title":"MathOptInterface.submit","text":"submit(\n    optimizer::AbstractOptimizer,\n    sub::AbstractSubmittable,\n    values...,\n)::Nothing\n\nSubmit values to the submittable sub of the optimizer optimizer.\n\nAn UnsupportedSubmittable error is thrown if model does not support the attribute attr (see supports) and a SubmitNotAllowed error is thrown if it supports the submittable sub but it cannot be submitted.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/callbacks/#Attributes","page":"Callbacks","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"CallbackNodeStatus\nCallbackVariablePrimal\nCallbackNodeStatusCode\nCALLBACK_NODE_STATUS_INTEGER\nCALLBACK_NODE_STATUS_FRACTIONAL\nCALLBACK_NODE_STATUS_UNKNOWN","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackNodeStatus","page":"Callbacks","title":"MathOptInterface.CallbackNodeStatus","text":"CallbackNodeStatus(callback_data)\n\nAn optimizer attribute describing the (in)feasibility of the primal solution available from CallbackVariablePrimal during a callback identified by callback_data.\n\nReturns a CallbackNodeStatusCode Enum.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackVariablePrimal","page":"Callbacks","title":"MathOptInterface.CallbackVariablePrimal","text":"CallbackVariablePrimal(callback_data)\n\nA variable attribute for the assignment to some primal variable's value during the callback identified by callback_data.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackNodeStatusCode","page":"Callbacks","title":"MathOptInterface.CallbackNodeStatusCode","text":"CallbackNodeStatusCode\n\nAn Enum of possible return values from calling get with CallbackNodeStatus.\n\nValues\n\nPossible values are:\n\nCALLBACK_NODE_STATUS_INTEGER: the primal solution available from CallbackVariablePrimal is integer feasible.\nCALLBACK_NODE_STATUS_FRACTIONAL: the primal solution available from CallbackVariablePrimal is integer infeasible.\nCALLBACK_NODE_STATUS_UNKNOWN: the primal solution available from CallbackVariablePrimal might be integer feasible or infeasible.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_INTEGER","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_INTEGER","text":"CALLBACK_NODE_STATUS_INTEGER::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_INTEGER: the primal solution available from CallbackVariablePrimal is integer feasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_FRACTIONAL","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_FRACTIONAL","text":"CALLBACK_NODE_STATUS_FRACTIONAL::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_FRACTIONAL: the primal solution available from CallbackVariablePrimal is integer infeasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_UNKNOWN","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_UNKNOWN","text":"CALLBACK_NODE_STATUS_UNKNOWN::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_UNKNOWN: the primal solution available from CallbackVariablePrimal might be integer feasible or infeasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#Lazy-constraints","page":"Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"LazyConstraintCallback\nLazyConstraint","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.LazyConstraintCallback","page":"Callbacks","title":"MathOptInterface.LazyConstraintCallback","text":"LazyConstraintCallback() <: AbstractCallback\n\nThe callback can be used to reduce the feasible set given the current primal solution by submitting a LazyConstraint. For instance, it may be called at an incumbent of a mixed-integer problem. Note that there is no guarantee that the callback is called at every feasible primal solution.\n\nThe current primal solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.LazyConstraintCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # should add a lazy constraint\n        func = # computes function\n        set = # computes set\n        MOI.submit(optimizer, MOI.LazyConstraint(callback_data), func, set)\n    end\nend)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.LazyConstraint","page":"Callbacks","title":"MathOptInterface.LazyConstraint","text":"LazyConstraint(callback_data)\n\nLazy constraint func-in-set submitted as func, set. The optimal solution returned by VariablePrimal will satisfy all lazy constraints that have been submitted.\n\nThis can be submitted only from the LazyConstraintCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the feasible solution callback.\n\nExamples\n\nSuppose x and y are VariableIndexs of optimizer. To add a LazyConstraint for 2x + 3y <= 1, write\n\nfunc = 2.0x + 3.0y\nset = MOI.LessThan(1.0)\nMOI.submit(optimizer, MOI.LazyConstraint(callback_data), func, set)\n\ninside a LazyConstraintCallback of data callback_data.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#User-cuts","page":"Callbacks","title":"User cuts","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"UserCutCallback\nUserCut","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.UserCutCallback","page":"Callbacks","title":"MathOptInterface.UserCutCallback","text":"UserCutCallback() <: AbstractCallback\n\nThe callback can be used to submit UserCut given the current primal solution. For instance, it may be called at fractional (i.e., non-integer) nodes in the branch and bound tree of a mixed-integer problem. Note that there is not guarantee that the callback is called everytime the solver has an infeasible solution.\n\nThe infeasible solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.UserCutCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # can find a user cut\n        func = # computes function\n        set = # computes set\n        MOI.submit(optimizer, MOI.UserCut(callback_data), func, set)\n    end\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.UserCut","page":"Callbacks","title":"MathOptInterface.UserCut","text":"UserCut(callback_data)\n\nConstraint func-to-set suggested to help the solver detect the solution given by CallbackVariablePrimal as infeasible. The cut is submitted as func, set. Typically CallbackVariablePrimal will violate integrality constraints, and a cut would be of the form ScalarAffineFunction-in-LessThan or ScalarAffineFunction-in-GreaterThan. Note that, as opposed to LazyConstraint, the provided constraint cannot modify the feasible set, the constraint should be redundant, e.g., it may be a consequence of affine and integrality constraints.\n\nThis can be submitted only from the UserCutCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the infeasible solution callback.\n\nNote that the solver may silently ignore the provided constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#Heuristic-solutions","page":"Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"HeuristicCallback\nHeuristicSolution\nHeuristicSolutionStatus\nHEURISTIC_SOLUTION_ACCEPTED\nHEURISTIC_SOLUTION_REJECTED\nHEURISTIC_SOLUTION_UNKNOWN","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicCallback","page":"Callbacks","title":"MathOptInterface.HeuristicCallback","text":"HeuristicCallback() <: AbstractCallback\n\nThe callback can be used to submit HeuristicSolution given the current primal solution. For example, it may be called at fractional (i.e., non-integer) nodes in the branch and bound tree of a mixed-integer problem. Note that there is no guarantee that the callback is called every time the solver has an infeasible solution.\n\nThe current primal solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.HeuristicCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # can find a heuristic solution\n        values = # computes heuristic solution\n        MOI.submit(optimizer, MOI.HeuristicSolution(callback_data), x,\n                   values)\n    end\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicSolution","page":"Callbacks","title":"MathOptInterface.HeuristicSolution","text":"HeuristicSolution(callback_data)\n\nHeuristically obtained feasible solution. The solution is submitted as variables, values where values[i] gives the value of variables[i], similarly to set. The submit call returns a HeuristicSolutionStatus indicating whether the provided solution was accepted or rejected.\n\nThis can be submitted only from the HeuristicCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the heuristic callback.\n\nSome solvers require a complete solution, others only partial solutions.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicSolutionStatus","page":"Callbacks","title":"MathOptInterface.HeuristicSolutionStatus","text":"HeuristicSolutionStatus\n\nAn Enum of possible return values for submit with HeuristicSolution. This informs whether the heuristic solution was accepted or rejected.\n\nValues\n\nPossible values are:\n\nHEURISTIC_SOLUTION_ACCEPTED: The heuristic solution was accepted\nHEURISTIC_SOLUTION_REJECTED: The heuristic solution was rejected\nHEURISTIC_SOLUTION_UNKNOWN: No information available on the acceptance\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED","text":"HEURISTIC_SOLUTION_ACCEPTED::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_ACCEPTED: The heuristic solution was accepted\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_REJECTED","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_REJECTED","text":"HEURISTIC_SOLUTION_REJECTED::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_REJECTED: The heuristic solution was rejected\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN","text":"HEURISTIC_SOLUTION_UNKNOWN::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_UNKNOWN: No information available on the acceptance\n\n\n\n\n\n","category":"constant"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"EditURL = \"https://github.com/odow/SDDP.jl/blob/v1.6.6/README.md\"","category":"page"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"<img src=\"https://raw.githubusercontent.com/odow/SDDP.jl/e9de84e0a4b57374bd9e0c95148da1501816e4c5/docs/src/assets/logo.png?raw=true\" alt=\"logo\" width=\"100px\"/>","category":"page"},{"location":"packages/SDDP/#SDDP.jl","page":"odow/SDDP.jl","title":"SDDP.jl","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"SDDP.jl is a JuMP extension for solving large convex multistage stochastic programming problems using stochastic dual dynamic programming.","category":"page"},{"location":"packages/SDDP/#License","page":"odow/SDDP.jl","title":"License","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"SDDP.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/SDDP/#Documentation","page":"odow/SDDP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"You can find the documentation at sddp.dev.","category":"page"},{"location":"packages/SDDP/#Help","page":"odow/SDDP.jl","title":"Help","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"If you need help, please open a GitHub issue.","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"EditURL = \"https://github.com/plasmo-dev/Plasmo.jl/blob/v0.5.4/README.md\"","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"<img src=\"https://github.com/plasmo-dev/Plasmo.jl/blob/main/docs/plasmo_logo.svg?raw=true\"/>","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"(Image: CI) (Image: codecov) (Image: ) (Image: DOI)","category":"page"},{"location":"packages/Plasmo/#Plasmo.jl","page":"plasmo-dev/Plasmo.jl","title":"Plasmo.jl","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Plasmo.jl (Platform for Scalable Modeling and Optimization) is a graph-based algebraic modeling framework that adopts a modular style to create mathematical optimization problems and manage distributed and hierarchical structures. The package has been developed as a JuMP extension and consequently supports  most JuMP syntax and functions. ","category":"page"},{"location":"packages/Plasmo/#Overview","page":"plasmo-dev/Plasmo.jl","title":"Overview","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The core data structure in Plasmo.jl is the OptiGraph. The optigraph contains a set of optinodes which represent self-contained optimization problems and optiedges that represent coupling between optinodes (which produces an underlying hypergraph structure of optinodes and optiedges). Optigraphs can further be embedded within other optigraphs to create nested hierarchical graph structures. The graph structures obtained using Plasmo.jl can be used for simple model and data management, but they can also be used to perform graph partitioning or develop interfaces to structured optimization solvers.","category":"page"},{"location":"packages/Plasmo/#License","page":"plasmo-dev/Plasmo.jl","title":"License","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Plasmo is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Plasmo/#Installation","page":"plasmo-dev/Plasmo.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Install Plasmo using Pkg.add:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"import Pkg\nPkg.add(\"Plasmo\")","category":"page"},{"location":"packages/Plasmo/#Documentation","page":"plasmo-dev/Plasmo.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The latest documentation is available through GitHub Pages. Additional examples can be found in the examples folder.","category":"page"},{"location":"packages/Plasmo/#Simple-Example","page":"plasmo-dev/Plasmo.jl","title":"Simple Example","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"using Plasmo\nusing Ipopt\n\n#create an optigraph\ngraph = OptiGraph()\n\n#add nodes to an optigraph\n@optinode(graph, n1)\n@optinode(graph, n2)\n\n#add variables, constraints, and objective functions to nodes\n@variable(n1, 0 <= x <= 2)\n@variable(n1, 0 <= y <= 3)\n@constraint(n1, x+y <= 4)\n@objective(n1, Min, x)\n\n@variable(n2,x)\n@NLconstraint(n2, exp(x) >= 2)\n\n#add a linkconstraint to couple nodes\n@linkconstraint(graph, n1[:x] == n2[:x])\n\n#optimize with Ipopt\nset_optimizer(graph, Ipopt.Optimizer)\noptimize!(graph)\n\n#Print solution values\nprintln(\"n1[:x] = \", value(n1[:x]))\nprintln(\"n2[:x] = \", value(n2[:x]))","category":"page"},{"location":"packages/Plasmo/#Acknowledgments","page":"plasmo-dev/Plasmo.jl","title":"Acknowledgments","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"This code is based on work supported by the following funding agencies:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"U.S. Department of Energy (DOE), Office of Science, under Contract No. DE-AC02-06CH11357\nDOE Office of Electricity Delivery and Energy Reliability’s Advanced Grid Research and Development program at Argonne National Laboratory\nNational Science Foundation under award NSF-EECS-1609183 and under award CBET-1748516","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The primary developer is Jordan Jalving (@jalving) with support from the following contributors.  ","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Victor Zavala (University of Wisconsin-Madison)\nYankai Cao (University of British Columbia)\nKibaek Kim (Argonne National Laboratory)\nSungho Shin (University of Wisconsin-Madison)","category":"page"},{"location":"packages/Plasmo/#Citing-Plasmo.jl","page":"plasmo-dev/Plasmo.jl","title":"Citing Plasmo.jl","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"If you find Plasmo.jl useful for your work, you may cite the manuscript as:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"@article{JalvingShinZavala2022,\n  title={A Graph-Based Modeling Abstraction for Optimization: Concepts and Implementation in Plasmo.jl},\n  author={Jordan Jalving and Sungho Shin and Victor M. Zavala},\n  journal={Mathematical Programming Computation},\n  year={2022},\n  volume={14},\n  pages={699 - 747}\n}","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"There is also a freely available pre-print:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"@misc{JalvingShinZavala2020,\ntitle = {A Graph-Based Modeling Abstraction for Optimization: Concepts and Implementation in Plasmo.jl},\nauthor = {Jordan Jalving and Sungho Shin and Victor M. Zavala},\nyear = {2020},\neprint = {2006.05378},\narchivePrefix = {arXiv},\nprimaryClass = {math.OC}\n}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"EditURL = \"https://github.com/jump-dev/MiniZinc.jl/blob/v0.3.4/README.md\"","category":"page"},{"location":"packages/MiniZinc/#MiniZinc.jl","page":"jump-dev/MiniZinc.jl","title":"MiniZinc.jl","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl is a wrapper for the MiniZinc constraint modeling language.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"It provides a way to write MathOptInterface models to .mzn files, and a way to interact with libminizinc.","category":"page"},{"location":"packages/MiniZinc/#Affiliation","page":"jump-dev/MiniZinc.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"This wrapper is maintained by the JuMP community and is not part of the MiniZinc project.","category":"page"},{"location":"packages/MiniZinc/#License","page":"jump-dev/MiniZinc.jl","title":"License","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl is licensed under the MIT License.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The underlying project, MiniZinc/libminizinc, is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/MiniZinc/#Install","page":"jump-dev/MiniZinc.jl","title":"Install","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Install MiniZinc.jl using the Julia package manager:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"import Pkg\nPkg.add(\"MiniZinc\")","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Windows","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"On Linux and macOS, this package automatically installs libminizinc. However, we're still working out problems with the install on Windows. To use MiniZinc.jl, you'll need to manually install a copy of libminizinc from minizinc.org or compile one yourself from MiniZinc/libminizinc.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"To teach MiniZinc.jl where to look for libminizinc, set the JULIA_LIBMINIZINC_DIR environment variable. For example:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"ENV[\"JULIA_LIBMINIZINC_DIR\"] = \"C:\\\\Program Files\\\\MiniZinc\"","category":"page"},{"location":"packages/MiniZinc/#Use-with-MathOptInterface","page":"jump-dev/MiniZinc.jl","title":"Use with MathOptInterface","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl supports the constraint programming sets defined in MathOptInterface, as well as (in)equality constraints.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The following example solves the following constraint program:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"xᵢ ∈ {1, 2, 3} ∀i=1,2,3\nzⱼ ∈ {0, 1}    ∀j=1,2\nz₁ <-> x₁ != x₂\nz₂ <-> x₂ != x₃\nz₁ + z₂ = 1","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"julia> import MiniZinc\n\njulia> import MathOptInterface as MOI\n\njulia> function main()\n           model = MOI.Utilities.CachingOptimizer(\n               MiniZinc.Model{Int}(),\n               MiniZinc.Optimizer{Int}(\"chuffed\"),\n           )\n           # xᵢ ∈ {1, 2, 3} ∀i=1,2,3\n           x = MOI.add_variables(model, 3)\n           MOI.add_constraint.(model, x, MOI.Interval(1, 3))\n           MOI.add_constraint.(model, x, MOI.Integer())\n           # zⱼ ∈ {0, 1}    ∀j=1,2\n           z = MOI.add_variables(model, 2)\n           MOI.add_constraint.(model, z, MOI.ZeroOne())\n           # z₁ <-> x₁ != x₂\n           MOI.add_constraint(\n               model,\n               MOI.VectorOfVariables([z[1], x[1], x[2]]),\n               MOI.Reified(MOI.AllDifferent(2)),\n           )\n           # z₂ <-> x₂ != x₃\n           MOI.add_constraint(\n               model,\n               MOI.VectorOfVariables([z[2], x[2], x[3]]),\n               MOI.Reified(MOI.AllDifferent(2)),\n           )\n           # z₁ + z₂ = 1\n           MOI.add_constraint(model, 1 * z[1] + x[2], MOI.EqualTo(1))\n           MOI.optimize!(model)\n           x_star = MOI.get(model, MOI.VariablePrimal(), x)\n           z_star = MOI.get(model, MOI.VariablePrimal(), z)\n           return x_star, z_star\n       end\nmain (generic function with 1 method)\n\njulia> main()\n([1, 1, 3], [0, 1])","category":"page"},{"location":"packages/MiniZinc/#Use-with-JuMP","page":"jump-dev/MiniZinc.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"You can also call MiniZinc from JuMP, using any solver that libminizinc supports. By default, MiniZinc.jl is compiled with \"highs\":","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"using JuMP\nimport MiniZinc\nmodel = Model(() -> MiniZinc.Optimizer{Float64}(\"highs\"))\n@variable(model, 1 <= x[1:3] <= 3, Int)\n@constraint(model, x in MOI.AllDifferent(3))\n@objective(model, Max, sum(i * x[i] for i in 1:3))\noptimize!(model)\n@show value.(x)","category":"page"},{"location":"packages/MiniZinc/#MathOptInterface-API","page":"jump-dev/MiniZinc.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The MiniZinc Optimizer{T} supports the following constraints and attributes.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.Reals","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\nMOI.ScalarAffineFunction{T} in MOI.Integer\nMOI.ScalarAffineFunction{T} in MOI.Interval{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.ZeroOne\nMOI.VariableIndex in MOI.EqualTo{T}\nMOI.VariableIndex in MOI.GreaterThan{T}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{T}\nMOI.VariableIndex in MOI.LessThan{T}\nMOI.VariableIndex in MOI.Parameter{T}\nMOI.VariableIndex in MOI.Semicontinuous{T}\nMOI.VariableIndex in MOI.Semiinteger{T}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.AllDifferent\nMOI.VectorOfVariables in MOI.BinPacking{T}\nMOI.VectorOfVariables in MOI.Circuit\nMOI.VectorOfVariables in MOI.CountAtLeast\nMOI.VectorOfVariables in MOI.CountBelongs\nMOI.VectorOfVariables in MOI.CountDistinct\nMOI.VectorOfVariables in MOI.CountGreaterThan\nMOI.VectorOfVariables in MOI.Cumulative\nMOI.VectorOfVariables in MOI.Path\nMOI.VectorOfVariables in MOI.Table{T}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.NLPBlock()\nMOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/MiniZinc/#Options","page":"jump-dev/MiniZinc.jl","title":"Options","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Set options using MOI.RawOptimizerAttribute in MOI or set_attribute in JuMP.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl supports the following options:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"model_filename::String = \"\": the location at which to write out the .mzn file during optimization. This option can be helpful during debugging. If left empty, a temporary file will be used instead.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"EditURL = \"transp.jl\"","category":"page"},{"location":"tutorials/linear/transp/#The-transportation-problem","page":"The transportation problem","title":"The transportation problem","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial is an adaptation of the transportation problem described in AMPL: A Modeling Language for Mathematical Programming, by R. Fourer, D.M. Gay and B.W. Kernighan.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"The purpose of this tutorial is to demonstrate how to create a JuMP model from an ad-hoc structured text file.","category":"page"},{"location":"tutorials/linear/transp/#Required-packages","page":"The transportation problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"using JuMP\nimport DelimitedFiles\nimport HiGHS","category":"page"},{"location":"tutorials/linear/transp/#Formulation","page":"The transportation problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Suppose that we have a set of factories that produce pogo sticks, and a set of retail stores in which to sell them. Each factory has a maximum number of pogo sticks that it can produce, and each retail store has a demand of pogo sticks that it can sell.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"In the transportation problem, we want to choose the number of pogo sticks to make and ship from each factory to each retail store that minimizes the total shipping cost.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Mathematically, we represent our set of factories by a set of origins i in O and our retail stores by a set of destinations j in D. The maximum supply at each factory is s_i and the demand from each retail store is d_j. The cost of shipping one pogo stick from i to j is c_ij.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"With a little effort, we can model the transportation problem as the following linear program:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"beginaligned\nmin   sum_i in O j in D c_ij x_ij \nst   sum_j in D x_i j le s_i  forall i in O \n       sum_i in O x_i j = d_j  forall j in D \n       x_i j ge 0  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/transp/#Data","page":"The transportation problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"We assume our data is in the form of a text file that has the following form. In practice, we would obtain this text file from the user as input, but for the purpose of this tutorial we're going to create it from Julia.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"open(joinpath(@__DIR__, \"transp.txt\"), \"w\") do io\n    print(\n        io,\n        \"\"\"\n             . FRA  DET LAN WIN  STL  FRE  LAF SUPPLY\n          GARY  39   14  11  14   16   82    8   1400\n          CLEV  27    .  12   .   26   95   17   2600\n          PITT  24   14  17  13   28   99   20   2900\n        DEMAND 900 1200 600 400 1700 1100 1000      0\n        \"\"\",\n    )\n    return\nend","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Here the rows are the origins, the columns are the destinations, and the values are the cost of shipping one pogo stick from the origin to the destination. If pogo stick cannot be transported from a source to a destination, then the value is .. The final row and final column are the demand and supply of each location respectively.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"We didn't account for arcs which do not exist in our formulation, but we can make a small change and fix x_ij = 0 if c_ij = .","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Our first step is to convert this text format into an appropriate Julia datastructure that we can work with. Since our data is tabular with named rows and columns, one option is JuMP's Containers.DenseAxisArray object:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"function read_data(filename::String)\n    data = DelimitedFiles.readdlm(filename)\n    rows, columns = data[2:end, 1], data[1, 2:end]\n    return Containers.DenseAxisArray(data[2:end, 2:end], rows, columns)\nend\n\ndata = read_data(joinpath(@__DIR__, \"transp.txt\"))","category":"page"},{"location":"tutorials/linear/transp/#JuMP-formulation","page":"The transportation problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Following Design patterns for larger models, we code our JuMP model as a function which takes in an input. In this example, we print the output to stdout:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"function solve_transportation_problem(data::Containers.DenseAxisArray)\n    # Get the set of supplies and demands\n    O, D = axes(data)\n    # Drop the SUPPLY and DEMAND nodes from our sets\n    O, D = setdiff(O, [\"DEMAND\"]), setdiff(D, [\"SUPPLY\"])\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x[o in O, d in D] >= 0)\n    # Remove arcs with \".\" cost by fixing them to 0.0.\n    for o in O, d in D\n        if data[o, d] == \".\"\n            fix(x[o, d], 0.0; force = true)\n        end\n    end\n    @objective(\n        model,\n        Min,\n        sum(data[o, d] * x[o, d] for o in O, d in D if data[o, d] != \".\"),\n    )\n    @constraint(model, [o in O], sum(x[o, :]) <= data[o, \"SUPPLY\"])\n    @constraint(model, [d in D], sum(x[:, d]) == data[\"DEMAND\", d])\n    optimize!(model)\n    # Pretty print the solution in the format of the input\n    print(\"    \", join(lpad.(D, 7, ' ')))\n    for o in O\n        print(\"\\n\", o)\n        for d in D\n            if isapprox(value(x[o, d]), 0.0; atol = 1e-6)\n                print(\"      .\")\n            else\n                print(\" \", lpad(value(x[o, d]), 6, ' '))\n            end\n        end\n    end\n    return\nend","category":"page"},{"location":"tutorials/linear/transp/#Solution","page":"The transportation problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Let's solve and view the solution:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"solve_transportation_problem(data)","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/developer/checklists.md\"","category":"page"},{"location":"moi/developer/checklists/#Checklists","page":"Checklists","title":"Checklists","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"The purpose of this page is to collate a series of checklists for commonly performed changes to the source code of MathOptInterface.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"In each case, copy the checklist into the description of the pull request.","category":"page"},{"location":"moi/developer/checklists/#Making-a-release","page":"Checklists","title":"Making a release","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when making a release of the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] `version` field of `Project.toml` has been updated\n       - If a breaking change, increment the MAJOR field and reset others to 0\n       - If adding new features, increment the MINOR field and reset PATCH to 0\n       - If adding bug fixes or documentation changes, increment the PATCH field\n\n## Documentation\n\n - [ ] Add a new entry to `docs/src/changelog.md`, following existing style\n\n## Tests\n\n - [ ] The `solver-tests.yml` GitHub action does not have unexpected failures.\n       To run the action, go to:\n       https://github.com/jump-dev/MathOptInterface.jl/actions/workflows/solver-tests.yml\n       and click \"Run workflow\"","category":"page"},{"location":"moi/developer/checklists/#Adding-a-new-set","page":"Checklists","title":"Adding a new set","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when adding a new set to the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Add a new `AbstractScalarSet` or `AbstractVectorSet` to `src/sets.jl`\n - [ ] If `isbitstype(S) == false`, implement `Base.copy(set::S)`\n - [ ] If `isbitstype(S) == false`, implement `Base.:(==)(x::S, y::S)`\n - [ ] If an `AbstractVectorSet`, implement `dimension(set::S)`, unless the\n       dimension is given by `set.dimension`.\n\n## Utilities\n\n - [ ] If an `AbstractVectorSet`, implement `Utilities.set_dot`,\n       unless the dot product between two vectors in the set is equivalent to\n       `LinearAlgebra.dot`\n - [ ] If an `AbstractVectorSet`, implement `Utilities.set_with_dimension` in\n       `src/Utilities/matrix_of_constraints.jl`\n - [ ] Add the set to the `@model` macro at the bottom of `src/Utilities.model.jl`\n\n## Documentation\n\n - [ ] Add a docstring, which gives the mathematical definition of the set,\n       along with an `## Example` block containing a `jldoctest`\n - [ ] Add the docstring to `docs/src/reference/standard_form.md`\n - [ ] Add the set to the relevant table in `docs/src/manual/standard_form.md`\n\n## Tests\n\n - [ ] Define a new `_set(::Type{S})` method in `src/Test/test_basic_constraint.jl`\n       and add the name of the set to the list at the bottom of that files\n - [ ] If the set has any checks in its constructor, add tests to `test/sets.jl`\n\n## MathOptFormat\n\n - [ ] Open an issue at `https://github.com/jump-dev/MathOptFormat` to add\n       support for the new set {{ replace with link to the issue }}\n\n## Optional\n\n - [ ] Implement `dual_set(::S)` and `dual_set_type(::Type{S})`\n - [ ] Add new tests to the `Test` submodule exercising your new set\n - [ ] Add new bridges to convert your set into more commonly used sets","category":"page"},{"location":"moi/developer/checklists/#Adding-a-new-bridge","page":"Checklists","title":"Adding a new bridge","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when adding a new bridge to the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"The steps are mostly the same, but locations depend on whether the bridge is a Constraint, Objective, or Variable bridge. In each case below, replace XXX with the appropriate type of bridge.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Create a new file in `src/Bridges/XXX/bridges`\n - [ ] Define the bridge, following existing examples. The name of the bridge\n       struct must end in `Bridge`\n - [ ] Check if your bridge can be a subtype of [`MOI.Bridges.Constraint.SetMapBridge`](@ref)\n - [ ] Define a new `const` that is a `SingleBridgeOptimizer` wrapping the\n       new bridge. The name of the const must be the name of the bridge, less\n       the `Bridge` suffix\n - [ ] `include` the file in `src/Bridges/XXX/bridges/XXX.jl`\n - [ ] If the bridge should be enabled by default, add the bridge to\n       `add_all_bridges` at the bottom of `src/Bridges/XXX/XXX.jl`\n\n## Tests\n\n - [ ] Create a new file in the appropriate subdirectory of `tests/Bridges/XXX`\n - [ ] Use `MOI.Bridges.runtests` to test various inputs and outputs of the\n       bridge\n - [ ] If, after opening the pull request to add the bridge, some lines are not\n       covered by the tests, add additional bridge-specific tests to cover the\n       untested lines.\n\n## Documentation\n\n - [ ] Add a docstring which uses the same template as existing bridges.\n - [ ] Add the docstring to `docs/src/submodules/Bridges/list_of_bridges.md`\n\n## Final touch\n\nIf the bridge depends on run-time values of other variables and constraints in\nthe model:\n\n - [ ] Implement `MOI.Utilities.needs_final_touch(::Bridge)`\n - [ ] Implement `MOI.Utilities.final_touch(::Bridge, ::MOI.ModelLike)`\n - [ ] Ensure that `final_touch` can be called multiple times in a row","category":"page"},{"location":"moi/developer/checklists/#Updating-MathOptFormat","page":"Checklists","title":"Updating MathOptFormat","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when updating the version of MathOptFormat.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] The file at `src/FileFormats/MOF/mof.X.Y.schema.json` is updated\n - [ ] The constants `SCHEMA_PATH`, `VERSION`, and `SUPPORTED_VERSIONS` are\n       updated in `src/FileFormats/MOF/MOF.jl`\n\n## New sets\n\n - [ ] New sets are added to the `@model` in `src/FileFormats/MOF/MOF.jl`\n - [ ] New sets are added to the `@enum` in `src/FileFormats/MOF/read.jl`\n - [ ] `set_to_moi` is defined for each set in `src/FileFormats/MOF/read.jl`\n - [ ] `head_name` is defined for each set in `src/FileFormats/MOF/write.jl`\n - [ ] A new unit test calling `_test_model_equality` is aded to\n       `test/FileFormats/MOF/MOF.jl`\n\n## Tests\n\n - [ ] The version field in `test/FileFormats/MOF/nlp.mof.json` is updated\n\n## Documentation\n\n - [ ] The version fields are updated in `docs/src/submodules/FileFormats/overview.md`","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"DocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/expressions/#Expressions","page":"Expressions","title":"Expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"JuMP has three types of expressions: affine, quadratic, and nonlinear. These expressions can be inserted into constraints or into the objective. This is particularly useful if an expression is used in multiple places in the model.","category":"page"},{"location":"manual/expressions/#Affine-expressions","page":"Expressions","title":"Affine expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"There are four ways of constructing an affine expression in JuMP: with the @expression macro, with operator overloading, with the AffExpr constructor, and with add_to_expression!.","category":"page"},{"location":"manual/expressions/#Macros","page":"Expressions","title":"Macros","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The recommended way to create an affine expression is via the @expression macro.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = @expression(model, 2x + y - 1)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This expression can be used in the objective or added to a constraint. For example:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> @objective(model, Min, 2 * ex - 1)\n4 x + 2 y - 3\n\njulia> objective_function(model)\n4 x + 2 y - 3","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Just like variables and constraints, named expressions can also be created. For example","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x[i = 1:3]);\n\njulia> @expression(model, expr[i = 1:3], i * sum(x[j] for j in i:3));\n\njulia> expr\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[2] + 2 x[3]\n 3 x[3]","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/expressions/#Operator-overloading","page":"Expressions","title":"Operator overloading","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Expressions can also be created without macros. However, note that in some cases, this can be much slower that constructing an expression using macros.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = 2x + y - 1\n2 x + y - 1","category":"page"},{"location":"manual/expressions/#Constructors","page":"Expressions","title":"Constructors","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"A third way to create an affine expression is by the AffExpr constructor. The first argument is the constant term, and the remaining arguments are variable-coefficient pairs.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = AffExpr(-1.0, x => 2.0, y => 1.0)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/#add_to_expression!","page":"Expressions","title":"add_to_expression!","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The fourth way to create an affine expression is by using add_to_expression!. Compared to the operator overloading method, this approach is faster because it avoids constructing temporary objects. The @expression macro uses add_to_expression! behind-the-scenes.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = AffExpr(-1.0)\n-1\n\njulia> add_to_expression!(ex, 2.0, x)\n2 x - 1\n\njulia> add_to_expression!(ex, 1.0, y)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"warning: Warning\nRead the section Initializing arrays for some cases to be careful about when using add_to_expression!.","category":"page"},{"location":"manual/expressions/#Removing-zero-terms","page":"Expressions","title":"Removing zero terms","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use drop_zeros! to remove terms from an affine expression with a 0 coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, ex, x + 1 - x)\n0 x + 1\n\njulia> drop_zeros!(ex)\n\njulia> ex\n1","category":"page"},{"location":"manual/expressions/#Coefficients","page":"Expressions","title":"Coefficients","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use coefficient to return the coefficient associated with a variable in an affine expression.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @expression(model, ex, 2x + 1)\n2 x + 1\n\njulia> coefficient(ex, x)\n2.0\n\njulia> coefficient(ex, y)\n0.0","category":"page"},{"location":"manual/expressions/#Quadratic-expressions","page":"Expressions","title":"Quadratic expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Like affine expressions, there are four ways of constructing a quadratic expression in JuMP: macros, operator overloading, constructors, and add_to_expression!.","category":"page"},{"location":"manual/expressions/#Macros-2","page":"Expressions","title":"Macros","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The @expression macro can be used to create quadratic expressions by including quadratic terms.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = @expression(model, x^2 + 2 * x * y + y^2 + x + y - 1)\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#Operator-overloading-2","page":"Expressions","title":"Operator overloading","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Operator overloading can also be used to create quadratic expressions. The same performance warning (discussed in the affine expression section) applies.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = x^2 + 2 * x * y + y^2 + x + y - 1\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#Constructors-2","page":"Expressions","title":"Constructors","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Quadratic expressions can also be created using the QuadExpr constructor. The first argument is an affine expression, and the remaining arguments are pairs, where the first term is a JuMP.UnorderedPair and the second term is the coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> aff_expr = AffExpr(-1.0, x => 1.0, y => 1.0)\nx + y - 1\n\njulia> quad_expr = QuadExpr(\n           aff_expr,\n           UnorderedPair(x, x) => 1.0,\n           UnorderedPair(x, y) => 2.0,\n           UnorderedPair(y, y) => 1.0,\n       )\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#add_to_expression!-2","page":"Expressions","title":"add_to_expression!","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Finally, add_to_expression! can also be used to add quadratic terms.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = QuadExpr(x + y - 1.0)\nx + y - 1\n\njulia> add_to_expression!(ex, 1.0, x, x)\nx² + x + y - 1\n\njulia> add_to_expression!(ex, 2.0, x, y)\nx² + 2 x*y + x + y - 1\n\njulia> add_to_expression!(ex, 1.0, y, y)\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"warning: Warning\nRead the section Initializing arrays for some cases to be careful about when using add_to_expression!.","category":"page"},{"location":"manual/expressions/#Removing-zero-terms-2","page":"Expressions","title":"Removing zero terms","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use drop_zeros! to remove terms from a quadratic expression with a 0 coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, ex, x^2 + x + 1 - x^2)\n0 x² + x + 1\n\njulia> drop_zeros!(ex)\n\njulia> ex\nx + 1","category":"page"},{"location":"manual/expressions/#Coefficients-2","page":"Expressions","title":"Coefficients","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use coefficient to return the coefficient associated with a pair of variables in a quadratic expression.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @expression(model, ex, 2*x*y + 3*x)\n2 x*y + 3 x\n\njulia> coefficient(ex, x, y)\n2.0\n\njulia> coefficient(ex, x, x)\n0.0\n\njulia> coefficient(ex, y, x)\n2.0\n\njulia> coefficient(ex, x)\n3.0","category":"page"},{"location":"manual/expressions/#Nonlinear-expressions","page":"Expressions","title":"Nonlinear expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Nonlinear expressions in JuMP are represented by a NonlinearExpr object. See Nonlinear expressions in detail for more details.","category":"page"},{"location":"manual/expressions/#Initializing-arrays","page":"Expressions","title":"Initializing arrays","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"JuMP implements zero(AffExpr) and one(AffExpr) to support various functions in LinearAlgebra (for example, accessing the off-diagonal of a Diagonal matrix).","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> zero(AffExpr)\n0\n\njulia> one(AffExpr)\n1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"However, this can result in a subtle bug if you call add_to_expression! or the MutableArithmetics API on an element created by zeros or ones:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = zeros(AffExpr, 2)\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 1.1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Notice how we modified x[1], but we also changed x[2]!","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This happened because zeros(AffExpr, 2) calls zero(AffExpr) once to obtain a zero element, and then creates an appropriately sized array filled with the same element.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This also happens with broadcasting calls containing a conversion of 0 or 1:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = Vector{AffExpr}(undef, 2)\n2-element Vector{AffExpr}:\n #undef\n #undef\n\njulia> x .= 0\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 1.1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The recommended way to create an array of empty expressions is as follows:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = Vector{AffExpr}(undef, 2)\n2-element Vector{AffExpr}:\n #undef\n #undef\n\njulia> for i in eachindex(x)\n           x[i] = AffExpr(0.0)\n       end\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 0","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Alternatively, use non-mutating operation to avoid updating x[1] in-place:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = zeros(AffExpr, 2)\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> x[1] += 1.1\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 0","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Note that for large expressions this will be slower due to the allocation of additional temporary objects.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"EditURL = \"factory_schedule.jl\"","category":"page"},{"location":"tutorials/linear/factory_schedule/#The-factory-schedule-example","page":"The factory schedule example","title":"The factory schedule example","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial was originally contributed by @Crghilardi.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial is a Julia translation of Part 5 from Introduction to Linear Programming with Python.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The purpose of this tutorial is to demonstrate how to use DataFrames and delimited files, and to structure your code that is robust to infeasibilities and permits running with different datasets.","category":"page"},{"location":"tutorials/linear/factory_schedule/#Required-packages","page":"The factory schedule example","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"using JuMP\nimport CSV\nimport DataFrames\nimport HiGHS\nimport StatsPlots","category":"page"},{"location":"tutorials/linear/factory_schedule/#Formulation","page":"The factory schedule example","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The Factory Scheduling Problem assumes we are optimizing the production of a good from factories f in F over the course of 12 months m in M.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"If a factory f runs during a month m, a fixed cost of a_f is incurred, the factory must produce x_mf units that is within some minimum and maximum production levels l_f and u_f respectively, and each unit of production incurs a variable cost c_f. Otherwise, the factory can be shut for the month with zero production and no fixed-cost is incurred. We denote the run/not-run decision by z_mf in 0 1, where z_mf is 1 if factory f runs in month m. The factory must produce enough units to satisfy demand d_m.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"With a little effort, we can formulate our problem as the following linear program:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"beginaligned\nmin  sumlimits_f in F m in M a_f z_mf + c_f x_mf \ntextst  x_mf le u_f z_mf  forall f in F m in M \n             x_mf ge l_f z_mf  forall f in F m in M \n             sumlimits_fin F x_mf = d_m  forall f in F m in M \n             z_mf in 0 1  forall f in F m in M\nendaligned","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"However, this formulation has a problem: if demand is too high, we may be unable to satisfy the demand constraint, and the problem will be infeasible.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"tip: Tip\nWhen modeling, consider ways to formulate your model such that it always has a feasible solution. This greatly simplifies debugging data errors that would otherwise result in an infeasible solution. In practice, most practical decisions have a feasible solution. In our case, we could satisfy demand (at a high cost) by buying replacement items for the buyer, or running the factories in overtime to make up the difference.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"We can improve our model by adding a new variable, delta_m, which represents the quantity of unmet demand in each month m. We penalize delta_m by an arbitrarily large value of $10,000/unit in the objective.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"beginaligned\nmin  sumlimits_f in F m in M a_f z_mf + c_f x_mf + sumlimits_m in M10000 delta_m \ntextst  x_mf le u_f z_mf  forall f in F m in M \n             x_mf ge l_f z_mf  forall f in F m in M \n             sumlimits_fin F x_mf - delta_m = d_m  forall f in F m in M \n             z_mf in 0 1  forall f in F m in M \n             delta_m ge 0  forall m in M\nendaligned","category":"page"},{"location":"tutorials/linear/factory_schedule/#Data","page":"The factory schedule example","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The JuMP GitHub repository contains two text files with the data we need for this tutorial.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The first file contains a dataset of our factories, A and B, with their production and cost levels for each month. For the documentation, the file is located at:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factories_filename = joinpath(@__DIR__, \"factory_schedule_factories.txt\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"To run locally, download factory_schedule_factories.txt and update factories_filename appropriately.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The file has the following contents:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"print(read(factories_filename, String))","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"We use the CSV and DataFrames packages to read it into Julia:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factory_df = CSV.read(\n    factories_filename,\n    DataFrames.DataFrame;\n    delim = ' ',\n    ignorerepeated = true,\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The second file contains the demand data by month:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_filename = joinpath(@__DIR__, \"factory_schedule_demand.txt\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"To run locally, download factory_schedule_demand.txt and update demand_filename appropriately.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df = CSV.read(\n    demand_filename,\n    DataFrames.DataFrame;\n    delim = ' ',\n    ignorerepeated = true,\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/#Data-validation","page":"The factory schedule example","title":"Data validation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Before moving on, it's always good practice to validate the data you read from external sources. The more effort you spend here, the fewer issues you will have later. The following function contains a few simple checks, but we could add more. For example, you might want to check that none of the values are too large (or too small), which might indicate a typo or a unit conversion issue (perhaps the variable costs are in $/1000 units instead of $/unit).","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function valiate_data(\n    demand_df::DataFrames.DataFrame,\n    factory_df::DataFrames.DataFrame,\n)\n    # Minimum production must not exceed maximum production.\n    @assert all(factory_df.min_production .<= factory_df.max_production)\n    # Demand, minimum production, fixed costs, and variable costs must all be\n    # non-negative.\n    @assert all(demand_df.demand .>= 0)\n    @assert all(factory_df.min_production .>= 0)\n    @assert all(factory_df.fixed_cost .>= 0)\n    @assert all(factory_df.variable_cost .>= 0)\n    return\nend\n\nvaliate_data(demand_df, factory_df)","category":"page"},{"location":"tutorials/linear/factory_schedule/#JuMP-formulation","page":"The factory schedule example","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Next, we need to code our JuMP formulation. As shown in Design patterns for larger models, it's always good practice to code your model in a function that accepts well-defined input and returns well-defined output.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function solve_factory_scheduling(\n    demand_df::DataFrames.DataFrame,\n    factory_df::DataFrames.DataFrame,\n)\n    # Even though we validated the data above, it's good practice to do it here\n    # too.\n    valiate_data(demand_df, factory_df)\n    months, factories = unique(factory_df.month), unique(factory_df.factory)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, status[months, factories], Bin)\n    @variable(model, production[months, factories], Int)\n    @variable(model, unmet_demand[months] >= 0)\n    # We use `eachrow` to loop through the rows of the dataframe and add the\n    # relevant constraints.\n    for r in eachrow(factory_df)\n        m, f = r.month, r.factory\n        @constraint(model, production[m, f] <= r.max_production * status[m, f])\n        @constraint(model, production[m, f] >= r.min_production * status[m, f])\n    end\n    @constraint(\n        model,\n        [r in eachrow(demand_df)],\n        sum(production[r.month, :]) + unmet_demand[r.month] == r.demand,\n    )\n    @objective(\n        model,\n        Min,\n        10_000 * sum(unmet_demand) + sum(\n            r.fixed_cost * status[r.month, r.factory] +\n            r.variable_cost * production[r.month, r.factory] for\n            r in eachrow(factory_df)\n        )\n    )\n    optimize!(model)\n    schedules = Dict{Symbol,Vector{Float64}}(\n        Symbol(f) => value.(production[:, f]) for f in factories\n    )\n    schedules[:unmet_demand] = value.(unmet_demand)\n    return (\n        termination_status = termination_status(model),\n        cost = objective_value(model),\n        # This `select` statement re-orders the columns in the DataFrame.\n        schedules = DataFrames.select(\n            DataFrames.DataFrame(schedules),\n            [:unmet_demand, :A, :B],\n        ),\n    )\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/#Solution","page":"The factory schedule example","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Now we can call our solve_factory_scheduling function using the data we read in above.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution = solve_factory_scheduling(demand_df, factory_df);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's see what solution contains:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.termination_status","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.cost","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.schedules","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"These schedules will be easier to visualize as a graph:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.groupedbar(\n    Matrix(solution.schedules);\n    bar_position = :stack,\n    labels = [\"unmet demand\" \"A\" \"B\"],\n    xlabel = \"Month\",\n    ylabel = \"Production\",\n    legend = :topleft,\n    color = [\"#20326c\" \"#4063d8\" \"#a0b1ec\"],\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Note that we don't have any unmet demand.","category":"page"},{"location":"tutorials/linear/factory_schedule/#What-happens-if-demand-increases?","page":"The factory schedule example","title":"What happens if demand increases?","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's run an experiment by increasing the demand by 50% in all time periods:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df.demand .*= 1.5","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Now we resolve the problem:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"high_demand_solution = solve_factory_scheduling(demand_df, factory_df);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"and visualize the solution:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.groupedbar(\n    Matrix(high_demand_solution.schedules);\n    bar_position = :stack,\n    labels = [\"unmet demand\" \"A\" \"B\"],\n    xlabel = \"Month\",\n    ylabel = \"Production\",\n    legend = :topleft,\n    color = [\"#20326c\" \"#4063d8\" \"#a0b1ec\"],\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Uh oh, we can't satisfy all of the demand.","category":"page"},{"location":"tutorials/linear/factory_schedule/#How-sensitive-is-the-solution-to-changes-in-variable-cost?","page":"The factory schedule example","title":"How sensitive is the solution to changes in variable cost?","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's run another experiment, this time seeing how the optimal objective value changes as we vary the variable costs of each factory.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"First though, let's reset the demand to it's original level:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df.demand ./= 1.5;\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"For our experiment, we're going to scale the variable costs of both factories by a set of values from 0.0 to 1.5:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"scale_factors = 0:0.1:1.5","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"At a high level, we're going to loop over the scale factors for A, then the scale factors for B, rescale the input data, call our solve_factory_scheduling example, and then store the optimal objective value in the following cost matrix:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"cost = zeros(length(scale_factors), length(scale_factors));\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Because we're modifying factory_df in-place, we need to store the original variable costs in a new column:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factory_df[!, :old_variable_cost] = copy(factory_df.variable_cost);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Then, we need a function to scale the :variable_cost column for a particular factory by a value scale:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function scale_variable_cost(df, factory, scale)\n    rows = df.factory .== factory\n    df[rows, :variable_cost] .=\n        round.(Int, df[rows, :old_variable_cost] .* scale)\n    return\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Our experiment is just a nested for-loop, modifying A and B and storing the cost:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"for (j, a) in enumerate(scale_factors)\n    scale_variable_cost(factory_df, \"A\", a)\n    for (i, b) in enumerate(scale_factors)\n        scale_variable_cost(factory_df, \"B\", b)\n        cost[i, j] = solve_factory_scheduling(demand_df, factory_df).cost\n    end\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's visualize the cost matrix:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.contour(\n    scale_factors,\n    scale_factors,\n    cost;\n    xlabel = \"Scale of factory A\",\n    ylabel = \"Scale of factory B\",\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"What can you infer from the solution?","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"info: Info\nThe Power Systems tutorial explains a number of other ways you can structure a problem to perform a parametric analysis of the solution. In particular, you can use in-place modification to reduce the time it takes to build and solve the resulting models.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"EditURL = \"changelog.md\"","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"CurrentModule = JuMP","category":"page"},{"location":"release_notes/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"release_notes/#[Version-1.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.16.0)-(October-24,-2023)","page":"Release notes","title":"Version 1.16.0 (October 24, 2023)","text":"","category":"section"},{"location":"release_notes/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added := operator for Boolean satisfiability problems (#3530)","category":"page"},{"location":"release_notes/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed text/latex printing of MOI.Interval sets (#3537)\nFixed tests with duplicate function names (#3539)","category":"page"},{"location":"release_notes/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated documentation list of supported solvers (#3527) (#3529) (#3538) (#3542) (#3545) (#3546)\nUpdated to Documenter@1.1 (#3528)\nFixed various tutorials (#3534) (#3532)\nFixed Project.toml compat bounds for standard libraries (#3544)","category":"page"},{"location":"release_notes/#[Version-1.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.15.1)-(September-24,-2023)","page":"Release notes","title":"Version 1.15.1 (September 24, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for single argument min and max operators (#3522)\nFixed error message for add_to_expression! when called with a GenericNonlinearExpr (#3506)\nFixed constraint tags with broadcasted constraints (#3515)\nFixed MethodError in MA.scaling (#3518)\nFixed support for arrays of Parameter variables (#3524)","category":"page"},{"location":"release_notes/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated to Documenter@1 (#3501)\nFixed links to data in tutorials (#3512)\nFixed typo in TSP tutorial (#3516)\nImproved error message for VariableNotOwned errors (#3520)\nFixed various JET errors (#3519)","category":"page"},{"location":"release_notes/#[Version-1.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.15.0)-(September-15,-2023)","page":"Release notes","title":"Version 1.15.0 (September 15, 2023)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it adds an entirely new data structure and API path for working with nonlinear programs. The previous nonlinear interface remains unchanged and is documented at Nonlinear Modeling (Legacy). The new interface is a treated as a non-breaking feature addition and is documented at Nonlinear Modeling.","category":"page"},{"location":"release_notes/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Although the new nonlinear interface is a feature addition, there are two changes which might be breaking for a very small number of users.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The syntax inside JuMP macros is parsed using a different code path, even for linear and quadratic expressions. We made this change to unify how we parse linear, quadratic, and nonlinear expressions. In all cases, the new code returns equivalent expressions, but because of the different order of operations, there are three changes to be aware of when updating:\nThe printed form of the expression may change, for example from x * y to y * x. This can cause tests which test the String representation of a model to fail.\nSome coefficients may change slightly due to floating point round-off error.\nParticularly when working with a JuMP extension, you may encounter a MethodError due to a missing or ambiguous method. These errors are due to previously existing bugs that were not triggered by the previous parsing code. If you encounter such an error, please open a GitHub issue.\nThe methods for Base.:^(x::VariableRef, n::Integer) and Base.:^(x::AffExpr, n::Integer) have changed. Previously, these methods supported only n = 0, 1, 2 and they always returned a QuadExpr, even for the case when n = 0 or n = 1. Now:\nx^0 returns one(T), where T is the value_type of the model (defaults to Float64)\nx^1 returns x\nx^2 returns a QuadExpr\nx^n where !(0 <= n <= 2) returns a NonlinearExpr.\nWe made this change to support nonlinear expressions and to align the mathematical definition of the operation with their return type. (Previously, users were surprised that x^1 returned a QuadExpr.) As a consequence of this change, the methods are now not type-stable. This means that the compiler cannot prove that x^2 returns a QuadExpr. If benchmarking shows that this is a performance problem, you can use the type-stable x * x instead of x^2.","category":"page"},{"location":"release_notes/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added triangle_vec which simplifies adding MOI.LogDetConeTriangle and MOI.RootDetConeTriangle constraints (#3456)\nAdded the new nonlinear interface. This is a very large change. See the documentation at Nonlinear Modeling and the (long) discussion in JuMP.jl#3106. Related PRs are (#3468) (#3472) (#3475) (#3483) (#3487) (#3488) (#3489) (#3504) (#3509)","category":"page"},{"location":"release_notes/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed uses of @nospecialize which cause precompilation failures in Julia v1.6.0 and v1.6.1. (#3464)\nFixed adding a container of Parameter (#3473)\nFixed return type of x^0 and x^1 to no longer return QuadExpr (see note in Breaking section above) (#3474)\nFixed error messages in LowerBoundRef, UpperBoundRef, FixRef, IntegerRef, BinaryRef, ParameterRef and related functions (#3494)\nFixed type inference of empty containers in JuMP macros (#3500)","category":"page"},{"location":"release_notes/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added GAMS to solver documentation (#3357)\nUpdated various tutorials (#3459) (#3460) (#3462) (#3463) (#3465) (#3490) (#3492) (#3503)\nAdded The network multi-commodity flow problem tutorial (#3491)\nAdded Two-stage stochastic programs tutorial (#3466)\nAdded better error messages for unsupported operations in LinearAlgebra (#3476)\nUpdated to the latest version of Documenter (#3484) (#3495) (#3497)\nUpdated GitHub action versions (#3507)","category":"page"},{"location":"release_notes/#[Version-1.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.14.1)-(September-2,-2023)","page":"Release notes","title":"Version 1.14.1 (September 2, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix links in Documentation (#3478)","category":"page"},{"location":"release_notes/#[Version-1.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.14.0)-(August-27,-2023)","page":"Release notes","title":"Version 1.14.0 (August 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added DimensionalData.jl extension (#3413)\nAdded syntactic sugar for the MOI.Parameter set (#3443)\nParameter\nParameterRef\nis_parameter\nparameter_value\nset_parameter_value","category":"page"},{"location":"release_notes/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed model_convert for BridgeableConstraint (#3437)\nFixed printing models with integer coefficients larger than typemax(Int) (#3447)\nFixed support for constant left-hand side functions in a complementarity constraint (#3452)","category":"page"},{"location":"release_notes/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated packages used in documentation (#3444) (#3455)\nFixed docstring tests (#3445)\nFixed printing change for MathOptInterface (#3446)\nFixed typos in documentation (#3448) (#3457)\nAdded SCIP to callback documentation (#3449)","category":"page"},{"location":"release_notes/#[Version-1.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.13.0)-(July-27,-2023)","page":"Release notes","title":"Version 1.13.0 (July 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for generic number types (#3377) (#3385)\nAdded fallback for MOI.AbstractSymmetricMatrixSetTriangle and MOI.AbstractSymmetricMatrixSetSquare (#3424)","category":"page"},{"location":"release_notes/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed set_start_values with MOI.Bridges.Objective.SlackBridge (#3422)\nFixed flakey doctest in variables.md (#3425)\nFixed names on CITATION.bib (#3423)","category":"page"},{"location":"release_notes/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Loraine.jl to the installation table (#3426)\nRemoved Penopt.jl from packages.toml (#3428)\nImproved problem statement in cannery example of tutorial (#3430)\nMinor cleanups in Containers.DenseAxisArray implementation (#3429)\nChanged nested_problems.jl: outer/inner to upper/lower (#3433)\nRemoved second SDP relaxation in OPF tutorial (#3432)","category":"page"},{"location":"release_notes/#[Version-1.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.12.0)-(June-19,-2023)","page":"Release notes","title":"Version 1.12.0 (June 19, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added coefficient_type keyword argument to add_bridge and remove_bridge (#3394)","category":"page"},{"location":"release_notes/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed error message for matrix in HermitianPSDCone (#3369)\nFixed EditURL for custom documentation pages (#3373)\nFixed return type annotations for MOI.ConstraintPrimal and MOI.ConstraintDual (#3381)\nFixed printing change in Julia nightly (#3391)\nFixed printing of Complex coefficients (#3397)\nFixed printing of constraints in text/latex mode (#3405)\nFixed performance issue in Containers.rowtable (#3410)\nFixed bug when variables added to set of wrong dimension (#3411)","category":"page"},{"location":"release_notes/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added more solver READMEs to the documentation (#3358) (#3360) (#3364) (#3365) (#3366) (#3368) (#3372) (#3374) (#3376) (#3379) (#3387) (#3389)\nAdded StatusSwitchingQP.jl to the installation table (#3354)\nUpdated checklist for adding a new solver (#3370)\nUpdated extension-tests.yml action (#3371) (#3375)\nColor logs in GitHub actions (#3392)\nAdded new tutorials\nOptimal power flow (#3395) (#3412)\nLovász numbers (#3399)\nDualization (#3402)\nUpdated JuMP paper citation (#3400)\nChanged GitHub action to upload LaTeX logs when building documentation (#3403)\nFixed printing of SCS log in documentation (#3406)\nUpdated solver versions (#3407)\nUpdated documentation to use Julia v1.9 (#3398)\nReplaced _value_type with MOI.Utilities.value_type (#3414)\nFixed a typo in docstring (#3415)\nRefactored API documentation (#3386)\nUpdated SCIP license (#3420)","category":"page"},{"location":"release_notes/#[Version-1.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.11.1)-(May-19,-2023)","page":"Release notes","title":"Version 1.11.1 (May 19, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a poor error message when sum(::DenseAxisArray; dims) was called (#3338)\nFixed support for dependent sets in the @variable macro (#3344)\nFixed a performance bug in constraints with sparse symmetric matrices (#3349)","category":"page"},{"location":"release_notes/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the printing of complex numbers (#3332)\nWhen printing, sets which contain constants ending in .0 now print as integers. This follows the behavior of constants in functions (#3341)\nAdded InfiniteOpt to the extensions documentation (#3343)\nAdded more documentation for the exponential cone (#3345) (#3347)\nAdded checklists for developers (#3346) (#3355)\nFixed test support upcoming Julia nightly (#3351)\nFixed extension-tests.yml action (#3353)\nAdd more solvers to the documentation (#3359) (#3361) (#3362)","category":"page"},{"location":"release_notes/#[Version-1.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.11.0)-(May-3,-2023)","page":"Release notes","title":"Version 1.11.0 (May 3, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added new methods to print_active_bridges for printing a particular objective, constraint, or variable (#3316)","category":"page"},{"location":"release_notes/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed tests for MOI v1.14.0 release (#3312)\nFixed indexing containers when an axis is Vector{Any} that contains a Vector{Any} element (#3280)\nFixed getindex(::AbstractJuMPScalar) which is called for an expression like x[] (#3314)\nFixed bug in set_string_names_on_creation with a vector of variables (#3322)\nFixed bug in memoize function in nonlinear documentation (#3337)","category":"page"},{"location":"release_notes/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#3317) (#3318) (#3328)\nAdded a test for the order of setting start values (#3315)\nAdded READMEs of solvers and extensions to the docs (#3309) (#3320) (#3327) (#3329) (#3333)\nStyle improvements to src/variables.jl (#3324)\nClarify that column generation does not find global optimum (#3325)\nAdd a GitHub actions workflow for testing extensions prior to release (#3331)\nDocument the release process for JuMP (#3334)\nFix links to discourse and chatroom (#3335)","category":"page"},{"location":"release_notes/#[Version-1.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.10.0)-(April-3,-2023)","page":"Release notes","title":"Version 1.10.0 (April 3, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Nonnegatives, Nonpositives and Zeros, and support vector-valued inequality syntax in the JuMP macros (#3273)\nAdded special support for LinearAlgebra.Symmetric and LinearAlgebra.Hermitian matrices in Zeros constraints (#3281) (#3296)\nAdded HermitianMatrixSpace and the Hermitian tag for generating a matrix of variables that is Hermitian (#3292) (#3293)\nAdded Semicontinuous and Semiinteger (#3302)\nAdded support for keyword indexing of containers (#3237)","category":"page"},{"location":"release_notes/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed [compat] bound for MathOptInterface in Project.toml (#3272)","category":"page"},{"location":"release_notes/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Split out the Nested optimization problems tutorial (#3274)\nUpdated doctests to ensure none have hidden state (#3275) (#3276)\nClarified how lazy constraints may revisit points (#3278)\nAdded P-Norm example (#3282)\nClarified docs that macros create new bindings (#3284)\nFixed threading example (#3283)\nAdded plot to The minimum distortion problem (#3288)\nAdded Google style rules for Vale and fixed warnings (#3285)\nAdded citation for the JuMP 1.0 paper (#3294)\nUpdated package versions in the documentation (#3298)\nAdded comment for the order in which start values must be set (#3303)\nImproved error message for unrecognized constraint operators (#3311)","category":"page"},{"location":"release_notes/#[Version-1.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.9.0)-(March-7,-2023)","page":"Release notes","title":"Version 1.9.0 (March 7, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added get_attribute and set_attribute. These replace get_optimizer_attribute and set_optimizer_attribute, although the _optimizer_ functions remain for backward compatibility. (#3219)\nAdded set_start_values for setting all supported start values in a model (#3238)\nAdd remove_bridge and print_active_bridges (#3259)","category":"page"},{"location":"release_notes/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The matrix returned by a variable in HermitianPSDCone is now a LinearAlgebra.Hermitian matrix. This is potentially breaking if you have written code to assume the return is a Matrix. (#3245) (#3246)\nFixed missing support for Base.isreal of expressions (#3252)","category":"page"},{"location":"release_notes/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a thread safety issue in the Parallelism tutorial (#3240) (#3243)\nImproved the error message when unsupported operators are used in @NL macros (#3236)\nClarified the documentation to say that matrices in HermitianPSDCone must be LinearAlgebra.Hermitian (#3241)\nMinor style fixes to internal macro code (#3247)\nAdd Quantum state discrimination tutorial (#3250)\nImprove error message when begin...end not passed to plural macros (#3255)\nDocument how to register function with varying number of input arguments (#3258)\nTidy tests by removing unneeded JuMP. prefixes (#3260)\nClarified the introduction to the Complex number support tutorial (#3262)\nFixed typos in the Documentation (#3263) (#3266) (#3268) (#3269)","category":"page"},{"location":"release_notes/#[Version-1.8.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.2)-(February-27,-2023)","page":"Release notes","title":"Version 1.8.2 (February 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed dot product between complex JuMP expression and number (#3244)","category":"page"},{"location":"release_notes/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Polish simple SDP examples (#3232)","category":"page"},{"location":"release_notes/#[Version-1.8.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.1)-(February-23,-2023)","page":"Release notes","title":"Version 1.8.1 (February 23, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for init in nonlinear generator expressions (#3226)","category":"page"},{"location":"release_notes/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Use and document import MathOptInterface as MOI (#3222)\nRemoved references in documentation to multiobjective optimization being unsupported (#3223)\nAdded tutorial on multi-objective portfolio optimization (#3227)\nRefactored some of the conic tutorials (#3229)\nFixed typos in the documentation (#3230)\nAdded tutorial on parallelism (#3231)","category":"page"},{"location":"release_notes/#[Version-1.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.0)-(February-16,-2023)","page":"Release notes","title":"Version 1.8.0 (February 16, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added --> syntax support for indicator constraints. The old syntax of => remains supported (#3207)\nAdded <--> syntax for reified constraints. For now, few solvers support reified constraints (#3206)\nAdded fix_discrete_variables. This is most useful for computing the dual of a mixed-integer program (#3208)\nAdded support for vector-valued objectives. For details, see the Multi-objective knapsack tutorial (#3176)","category":"page"},{"location":"release_notes/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report by switching to an explicit LU factorization of the basis matrix (#3182)\nFixed a bug that prevented [; kwarg] arguments in macros (#3220)","category":"page"},{"location":"release_notes/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Minor fixes to the documentation (#3200) (#3201) (#3203) (#3210)\nAdded tutorial Constraint programming (#3202)\nAdded more examples to Tips and Tricks\nRemove _distance_to_set in favor of MOI.Utilities.distance_to_set (#3209)\nImprove The diet problem tutorial by adding the variable as a column in the dataframe (#3213)\nImprove The knapsack problem example tutorial (#3216) (#3217)\nAdded the Ellipsoid approximation tutorial (#3218)","category":"page"},{"location":"release_notes/#[Version-1.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.7.0)-(January-25,-2023)","page":"Release notes","title":"Version 1.7.0 (January 25, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for view of a Containers.DenseAxisArray (#3152) (#3180)\nAdded support for containers of variables in ComplexPlane (#3184)\nAdded support for minimum and maximum generators in nonlinear expressions (#3189)\nAdded SnoopPrecompile statements that reduce the time-to-first-solve in Julia 1.9 (#3193) (#3195) (#3196) (#3197)","category":"page"},{"location":"release_notes/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Large refactoring of the tests (#3166) (#3167) (#3168) (#3169) (#3170) (#3171)\nRemove unreachable code due to VERSION checks (#3172)\nDocument how to test JuMP extensions (#3174)\nFix method ambiguities in Containers (#3173)\nImprove error message that is thrown when = is used instead of == in the @constraint macro (#3178)\nImprove the error message when Bool is used instead of Bin in the @variable macro (#3180)\nUpdate versions of the documentation (#3185)\nTidy the import of packages and remove unnecessary prefixes (#3186) (#3187)\nRefactor src/JuMP.jl by moving methods into more relevant files (#3188)\nFix docstring of Model not appearing in the documentation (#3198)","category":"page"},{"location":"release_notes/#[Version-1.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.6.0)-(January-1,-2023)","page":"Release notes","title":"Version 1.6.0 (January 1, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a result keyword argument to solution_summary to allow summarizing models with multiple solutions (#3138)\nAdded relax_with_penalty!, which is a useful tool when debugging infeasible models (#3140)\nAdded has_start_value (#3157)\nAdded support for HermitianPSDCone in constraints (#3154)","category":"page"},{"location":"release_notes/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed promotion of complex expressions (#3150) (#3164)","category":"page"},{"location":"release_notes/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Benders tutorial with in-place resolves (#3145)\nAdded more Tips and tricks for linear programs (#3144) (#3163)\nClarified documentation that start can depend on the indices of a variable container (#3148)\nReplace instances of length and size by the recommended eachindex and axes (#3149)\nAdded a warning explaining why the model is dirty when accessing solution results from a modified model (#3156)\nClarify documentation that PSD ensures a symmetric matrix (#3159)\nMaintenance of the JuMP test suite (#3146) (#3158) (#3162)","category":"page"},{"location":"release_notes/#[Version-1.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.5.0)-(December-8,-2022)","page":"Release notes","title":"Version 1.5.0 (December 8, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for complex-valued variables:\nHermitianPSDCone (#3109)\nComplexPlane and ComplexVariable (#3134)\nAdd support for MOI.OptimizerWithAttributes in set_optimizer_attribute and get_optimizer_attribute (#3129)","category":"page"},{"location":"release_notes/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed error message for vectorized interval constraints (#3123)\nFixed passing AbstractString to set_optimizer_attribute (#3127)","category":"page"},{"location":"release_notes/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update package versions used in docs (#3119) (#3133) (#3139)\nFixed output of diet tutorial (#3120)\nExplain how to use Dates.period in set_time_limit_sec (#3121)\nUpdate to JuliaFormatter v1.0.15 (#3130)\nFixed HTTP server example in web_app.jl (#3131)\nUpdate docs to build with Documenter#master (#3094)\nAdd tests for LinearAlgebra operations (#3132)\nTidy these release notes (#3135)\nAdded documentation for Complex number support (#3141)\nRemoved the \"workforce scheduling\" and \"steelT3\" tutorials (#3143)","category":"page"},{"location":"release_notes/#[Version-1.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.4.0)-(October-29,-2022)","page":"Release notes","title":"Version 1.4.0 (October 29, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Containers.rowtable which converts a container into a vector of NamedTuples to support the Tables.jl interface. This simplifies converting Containers.DenseAxisArray and Containers.SparseAxisArray objects into tabular forms such as a DataFrame (#3104)\nAdded a new method to Containers.container so that index names are passed to the container (#3088)","category":"page"},{"location":"release_notes/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in copy_to(dest::Model, src::MOI.ModelLike) when src has nonlinear components (#3101)\nFixed the printing of (-1.0 + 0.0im) coefficients in complex expressions (#3112)\nFixed a parsing bug in nonlinear expressions with generator statements that contain multiple for statements (#3116)","category":"page"},{"location":"release_notes/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Converted the multi-commodity flow tutorial to use an SQLite database (#3098)\nFixed a number of typos in the documentation (#3103) (#3107) (#3018)\nImproved various style aspects of the PDF documentation (#3095) (#3098) (#3102)","category":"page"},{"location":"release_notes/#[Version-1.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.3.1)-(September-28,-2022)","page":"Release notes","title":"Version 1.3.1 (September 28, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in relax_integrality (#3087)\nFixed the type stability of operators with Complex arguments (#3072)\nFixed a bug which added additional +() terms to some nonlinear expressions (#3091)\nFixed potential method ambiguities with AffExpr and QuadExpr objects (#3092)","category":"page"},{"location":"release_notes/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added vale as a linter for the documentation (#3080)\nAdded a tutorial on debugging JuMP models (#3043)\nFixed a number of typos in the documentation (#3079) (#3083)\nMany other small tweaks to the documentation (#3068) (#3073) (#3074) (#3075) (#3076) (#3077) (#3078) (#3081) (#3082) (#3084) (#3085) (#3089)","category":"page"},{"location":"release_notes/#[Version-1.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.3.0)-(September-5,-2022)","page":"Release notes","title":"Version 1.3.0 (September 5, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support slicing in SparseAxisArray (#3031)","category":"page"},{"location":"release_notes/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug introduced in v1.2.0 that prevented DenseAxisArrays with Vector keys (#3064)","category":"page"},{"location":"release_notes/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Released the JuMP logos under the CC BY 4.0 license (#3063)\nMinor tweaks to the documentation (#3054) (#3056) (#3057) (#3060) (#3061) (#3065)\nImproved code coverage of a number of files (#3048) (#3049) (#3050) (#3051) (#3052) (#3053) (#3058) (#3059)","category":"page"},{"location":"release_notes/#[Version-1.2.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.2.1)-(August-22,-2022)","page":"Release notes","title":"Version 1.2.1 (August 22, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug when parsing two-sided nonlinear constraints (#3045)","category":"page"},{"location":"release_notes/#[Version-1.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.2.0)-(August-16,-2022)","page":"Release notes","title":"Version 1.2.0 (August 16, 2022)","text":"","category":"section"},{"location":"release_notes/#Breaking-2","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it significantly refactors the internal code for handling nonlinear programs to use the MathOptInterface.Nonlinear submodule that was introduced in MathOptInterface v1.3.0. As a consequence, the internal datastructure in model.nlp_data has been removed, as has the JuMP._Derivatives submodule. Despite the changes, the public API for nonlinear programming has not changed, and any code that uses only the public API and that worked with v1.1.1 will continue to work with v1.2.0.","category":"page"},{"location":"release_notes/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added all_constraints(model; include_variable_in_set_constraints) which simplifies returning a list of all constraint indices in the model.\nAdded the ability to delete nonlinear constraints via delete(::Model, ::NonlinearConstraintRef).\nAdded the ability to provide an explicit Hessian for a multivariate user-defined function.\nAdded support for querying the primal value of a nonlinear constraint via value(::NonlinearConstraintRef)","category":"page"},{"location":"release_notes/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Containers.DenseAxisArray so that it now supports indexing with keys that hash to the same value, even if they are different types, for example, Int32 and Int64.\nFixed a bug printing the model when the solver does not support MOI.Name.","category":"page"},{"location":"release_notes/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a constraint programming formulation to the Sudoku tutorial.\nAdded newly supported solvers Pajarito, Clarabel, and COPT to the installation table.\nFixed a variety of other miscellaneous issues in the documentation.","category":"page"},{"location":"release_notes/#[Version-1.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.1.1)-(June-14,-2022)","page":"Release notes","title":"Version 1.1.1 (June 14, 2022)","text":"","category":"section"},{"location":"release_notes/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed problem displaying LaTeX in the documentation\nMinor updates to the style guide\nUpdated to MOI v1.4.0 in the documentation","category":"page"},{"location":"release_notes/#[Version-1.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.1.0)-(May-25,-2022)","page":"Release notes","title":"Version 1.1.0 (May 25, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added num_constraints(::Model; count_variable_in_set_constraints) to simplify the process of counting the number of constraints in a model\nAdded VariableRef(::ConstraintRef) for querying the variable associated with a bound or integrality constraint.\nAdded set_normalized_coefficients for modifying the variable coefficients of a vector-valued constraint.\nAdded set_string_names_on_creation to disable creating String names for variables and constraints. This can improve performance.","category":"page"},{"location":"release_notes/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug passing nothing to the start keyword of @variable","category":"page"},{"location":"release_notes/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"New tutorials:\nSensitivity analysis of a linear program\nServing web apps\nMinimal ellipse SDP tutorial refactored and improved\nDocs updated to the latest version of each package\nLots of minor fixes and improvements to the documentation","category":"page"},{"location":"release_notes/#[Version-1.0.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.0.0)-(March-24,-2022)","page":"Release notes","title":"Version 1.0.0 (March 24, 2022)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Read more about this release, along with an acknowledgement of all the contributors in our JuMP 1.0.0 is released blog post.","category":"page"},{"location":"release_notes/#Breaking-3","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The previously deprecated functions (v0.23.0, v0.23.1) have been removed. Deprecation was to improve consistency of function names:\nnum_nl_constraints (see num_nonlinear_constraints)\nall_nl_constraints (see all_nonlinear_constraints)\nadd_NL_expression (see add_nonlinear_expression)\nset_NL_objective  (see set_nonlinear_objective)\nadd_NL_constraint (see add_nonlinear_constraint)\nnl_expr_string (see nonlinear_expr_string)\nnl_constraint_string (see nonlinear_constraint_string)\nSymMatrixSpace (see SymmetricMatrixSpace)\nThe unintentionally exported variable JuMP.op_hint has been renamed to the unexported JuMP._OP_HINT","category":"page"},{"location":"release_notes/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .nl files\nFixed a bug broadcasting SparseAxisArrays","category":"page"},{"location":"release_notes/#[Version-0.23.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.2)-(March-14,-2022)","page":"Release notes","title":"Version 0.23.2 (March 14, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added relative_gap to solution_summary\nregister now throws an informative error if the function is not differentiable using ForwardDiff. In some cases, the check in register will encounter a false negative, and the informative error will be thrown at run-time. This usually happens when the function is non-differentiable in a subset of the domain.","category":"page"},{"location":"release_notes/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a scoping issue when extending the container keyword of containers","category":"page"},{"location":"release_notes/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Docs updated to the latest version of each package","category":"page"},{"location":"release_notes/#[Version-0.23.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.1)-(March-2,-2022)","page":"Release notes","title":"Version 0.23.1 (March 2, 2022)","text":"","category":"section"},{"location":"release_notes/#Deprecated","page":"Release notes","title":"Deprecated","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"nl_expr_string and nl_constraint_string have been renamed to nonlinear_expr_string and nonlinear_constraint_string. The old methods still exist with deprecation warnings. This change should impact very few users because to call them you must rely on private internals of the nonlinear API. Users are encouraged to use sprint(show, x) instead, where x is the nonlinear expression or constraint of interest.","category":"page"},{"location":"release_notes/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for Base.abs2(x) where x is a variable or affine expression. This is mainly useful for complex-valued constraints.","category":"page"},{"location":"release_notes/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed addition of complex and real affine expressions\nFixed arithmetic for Complex-valued quadratic expressions\nFixed variable bounds passed as Rational{Int}(Inf)\nFixed printing of the coefficient (0 + 1im)\nFixed a bug when solution_summary is called prior to optimize!","category":"page"},{"location":"release_notes/#[Version-0.23.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.0)-(February-25,-2022)","page":"Release notes","title":"Version 0.23.0 (February 25, 2022)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP v0.23.0 is a breaking release. It is also a release-candidate for JuMP v1.0.0. That is, if no issues are found with the v0.23.0 release, then it will be re-tagged as v1.0.0.","category":"page"},{"location":"release_notes/#Breaking-4","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nMathOptInterface has been updated to v1.0.0\nAll previously deprecated functionality has been removed\nPrintMode, REPLMode and IJuliaMode have been removed in favor of the MIME types MIME\"text/plain\" and MIME\"text/latex\". Replace instances of ::Type{REPLMode} with ::MIME\"text/plain\", REPLMode with MIME(\"text/plain\"), ::Type{IJuliaMode} with ::MIME\"text/latex\", and IJuliaMode with MIME(\"text/latex\").\nFunctions containing the nl_ acronym have been renamed to the more explicit nonlinear_. For example, num_nl_constraints is now num_nonlinear_constraints and set_NL_objective is now set_nonlinear_objective. Calls to the old functions throw an error explaining the new name.\nSymMatrixSpace has been renamed to SymmetricMatrixSpace","category":"page"},{"location":"release_notes/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added nonlinear_dual_start_value and set_nonlinear_dual_start_value\nAdded preliminary support for Complex coefficient types","category":"page"},{"location":"release_notes/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary","category":"page"},{"location":"release_notes/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"MILP examples have been migrated from GLPK to HiGHS\nFixed various typos\nImproved section on setting constraint start values","category":"page"},{"location":"release_notes/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated functionality. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Then install JuMP v0.22.3:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add JuMP@0.22.3\"","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"release_notes/#[Version-0.22.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.3)-(February-10,-2022)","page":"Release notes","title":"Version 0.22.3 (February 10, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a reproducibility issue in the TSP tutorial\nFixed a reproducibility issue in the max_cut_sdp tutorial\nFixed a bug broadcasting an empty SparseAxisArray","category":"page"},{"location":"release_notes/#Other-25","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a warning and improved documentation for the modify-then-query case\nFixed a typo in the docstring of RotatedSecondOrderCone\nAdded Aqua.jl as a check for code health\nAdded introductions to each section of the tutorials\nImproved the column generation and Benders decomposition tutorials\nUpdated documentation to MOI v0.10.8\nUpdated JuliaFormatter to v0.22.2","category":"page"},{"location":"release_notes/#[Version-0.22.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.2)-(January-10,-2022)","page":"Release notes","title":"Version 0.22.2 (January 10, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The function all_nl_constraints now returns all nonlinear constraints in a model\nstart_value and set_start_value can now be used to get and set the primal start for constraint references\nPlural macros now return a tuple containing the elements that were defined instead of nothing\nAnonymous variables are now printed as _[i] where i is the index of the variable instead of noname. Calling name(x) still returns \"\" so this is non-breaking.","category":"page"},{"location":"release_notes/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed handling of min and max in nonlinear expressions\nCartesianIndex is no longer allowed as a key for DenseAxisArrays.","category":"page"},{"location":"release_notes/#Other-26","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the performance of GenericAffExpr\nAdded a tutorial on the Travelling Salesperson Problem\nAdded a tutorial on querying the Hessian of a nonlinear program\nAdded documentation on using custom solver binaries.","category":"page"},{"location":"release_notes/#[Version-0.22.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.1)-(November-29,-2021)","page":"Release notes","title":"Version 0.22.1 (November 29, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Export OptimizationSense enum, with instances: MIN_SENSE, MAX_SENSE, and FEASIBILITY_SENSE\nAdd Base.isempty(::Model) to match Base.empty(::Model)","category":"page"},{"location":"release_notes/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in container with tuples as indices\nFix bug in set_time_limit_sec","category":"page"},{"location":"release_notes/#Other-27","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add tutorial \"Design patterns for larger models\"\nRemove release notes section from PDF\nGeneral edits of the documentation and error messages","category":"page"},{"location":"release_notes/#[Version-0.22.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.0)-(November-10,-2021)","page":"Release notes","title":"Version 0.22.0 (November 10, 2021)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP v0.22 is a breaking release","category":"page"},{"location":"release_notes/#Breaking-5","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP 0.22 contains a number of breaking changes. However, these should be invisible for the majority of users. You will mostly encounter these breaking changes if you: wrote a JuMP extension, accessed backend(model), or called @SDconstraint.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The breaking changes are as follows:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"MathOptInterface has been updated to v0.10.4. For users who have interacted with the MOI backend, this contains a large number of breaking changes. Read the MathOptInterface release notes for more details.\nThe bridge_constraints keyword argument to Model and set_optimizer has been renamed add_bridges to reflect that more thing were bridged than just constraints.\nThe backend(model) field now contains a concrete instance of a MOI.Utilities.CachingOptimizer instead of one with an abstractly typed optimizer field. In most cases, this will lead to improved performance. However, calling set_optimizer after backend invalidates the old backend. For example:\nmodel = Model()\nb = backend(model)\nset_optimizer(model, GLPK.Optimizer)\n@variable(model, x)\n# b is not updated with `x`! Get a new b by calling `backend` again.\nnew_b = backend(model)\nAll usages of @SDconstraint are deprecated. The new syntax is @constraint(model, X >= Y, PSDCone()).\nCreating a DenseAxisArray with a Number as an axis will now display a warning. This catches a common error in which users write @variable(model, x[length(S)]) instead of @variable(model, x[1:length(S)]).\nThe caching_mode argument to Model, for example, Model(caching_mode = MOIU.MANUAL) mode has been removed. For more control over the optimizer, use direct_model instead.\nThe previously deprecated lp_objective_perturbation_range and lp_rhs_perturbation_range functions have been removed. Use lp_sensitivity_report instead.\nThe .m fields of NonlinearExpression and NonlinearParameter have been renamed to .model.\nInfinite variable bounds are now ignored. Thus, @variable(model, x <= Inf) will show has_upper_bound(x) == false. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.\nThe variable_type and constraint_type functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.\nThe internal functions moi_mode, moi_bridge_constraints, moi_add_constraint, and moi_add_to_function_constant are no longer exported.\nThe un-used method Containers.generate_container has been deleted.\nThe Containers API has been refactored, and _build_ref_sets is now public as Containers.build_ref_sets.\nThe parse_constraint_ methods for extending @constraint at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.","category":"page"},{"location":"release_notes/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The TerminationStatusCode and ResultStatusCode enums are now exported by JuMP. Prefer termination_status(model) == OPTIMAL instead of == MOI.OPTIMAL, although the MOI. prefix way still works.\nCopy a x::DenseAxisArray to an Array by calling Array(x).\nNonlinearExpression is now a subtype of AbstractJuMPScalar\nConstraints such as @constraint(model, x + 1 in MOI.Integer()) are now supported.\nprimal_feasibility_report now accepts a function as the first argument.\nScalar variables @variable(model, x[1:2] in MOI.Integer()) creates two variables, both of which are constrained to be in the set MOI.Integer.\nConic constraints can now be specified as inequalities under a different partial ordering. So @constraint(model, x - y in MOI.Nonnegatives()) can now be written as @constraint(model, x >= y, MOI.Nonnegatives()).\nNames are now set for vectorized constraints.","category":"page"},{"location":"release_notes/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue when show was called on a SparseAxisArray with a large number of elements.\nFixed a bug displaying barrier and simplex iterations in solution_summary.\nFixed a bug by implementing hash for DenseAxisArray and SparseAxisArray.\nNames are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with direct_model.\nMutableArithmetics.Zero is converted into a 0.0 before being returned to the user. Previously, some calls to @expression would return the undocumented MutableArithmetics.Zero() object. One example is summing over an empty set @expression(model, sum(x[i] for i in 1:0)). You will now get 0.0 instead.\nAffExpr and QuadExpr can now be used with == 0 instead of iszero. This fixes a number of issues relating to Julia standard libraries such as LinearAlgebra and SparseArrays.\nFixed a bug when registering a user-defined function with splatting.","category":"page"},{"location":"release_notes/#Other-28","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The documentation is now available as a PDF.\nThe documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)\nThe documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.\nThe documentation is now built with Julia 1.6 instead of 1.0.\nVarious error messages have been improved to be more readable.","category":"page"},{"location":"release_notes/#[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)","page":"Release notes","title":"Version 0.21.10 (September 4, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added add_NL_expression\nadd_NL_xxx functions now support AffExpr and QuadExpr as terms","category":"page"},{"location":"release_notes/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary\nFixed a bug in relax_integrality","category":"page"},{"location":"release_notes/#Other-29","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved error message in lp_sensitivity_report","category":"page"},{"location":"release_notes/#[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)","page":"Release notes","title":"Version 0.21.9 (August 1, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Containers now support arbitrary container types by passing the type to the container keyword and overloading Containers.container.\nis_valid now supports nonlinear constraints\nAdded unsafe_backend for querying the inner-most optimizer of a JuMP model.\nNonlinear parameters now support the plural @NLparameters macro.\nContainers (for example, DenseAxisArray) can now be used in vector-valued constraints.","category":"page"},{"location":"release_notes/#Other-30","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements to the documentation.","category":"page"},{"location":"release_notes/#[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)","page":"Release notes","title":"Version 0.21.8 (May 8, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The @constraint macro is now extendable in the same way as @variable.\nAffExpr and QuadExpr can now be used in nonlinear macros.","category":"page"},{"location":"release_notes/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report.\nFixed an inference issue when creating empty SparseAxisArrays.","category":"page"},{"location":"release_notes/#[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)","page":"Release notes","title":"Version 0.21.7 (April 12, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-26","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added primal_feasibility_report, which can be used to check whether a primal point satisfies primal feasibility.\nAdded coefficient, which returns the coefficient associated with a variable in affine and quadratic expressions.\nAdded copy_conflict, which returns the IIS of an infeasible model.\nAdded solution_summary, which returns (and prints) a struct containing a summary of the solution.\nAllow AbstractVector in vector constraints instead of just Vector.\nAdded latex_formulation(model) which returns an object representing the latex formulation of a model. Use print(latex_formulation(model)) to print the formulation as a string.\nUser-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.\nDenseAxisArray's now support broadcasting over multiple arrays.\nContainer indices can now be iterators of Base.SizeUnknown.","category":"page"},{"location":"release_notes/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug in rad2deg and deg2rad in nonlinear expressions.\nFixed a MethodError bug in Containers when forcing container type.\nAllow partial slicing of a DenseAxisArray, resolving an issue from 2014.\nFixed a bug printing variable names in IJulia.\nEnding an IJulia cell with model now prints a summary of the model (like in the REPL) not the latex formulation. Use print(model) to print the latex formulation.\nFixed a bug when copying models containing nested arrays.","category":"page"},{"location":"release_notes/#Other-31","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Tutorials are now part of the documentation, and more refactoring has taken place.\nAdded JuliaFormatter added as a code formatter.\nAdded some precompilation statements to reduce initial latency.\nVarious improvements to error messages to make them more helpful.\nImproved performance of value(::NonlinearExpression).\nImproved performance of fix(::VariableRef).","category":"page"},{"location":"release_notes/#[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)","page":"Release notes","title":"Version 0.21.6 (January 29, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-27","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for skew symmetric variables via @variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace()).\nlp_sensitivity_report has been added which significantly improves the performance of querying the sensitivity summary of an LP. lp_objective_perturbation_range and lp_rhs_perturbation_range are deprecated.\nDual warm-starts are now supported with set_dual_start_value and dual_start_value.\n∈ (\\in<tab>) can now be used in macros instead of = or in.\nUse haskey(model::Model, key::Symbol) to check if a name key is registered in a model.\nAdded unregister(model::Model, key::Symbol) to unregister a name key from model.\nAdded callback_node_status for use in callbacks.\nAdded print_bridge_graph to visualize the bridging graph generated by MathOptInterface.\nImproved error message for containers with duplicate indices.","category":"page"},{"location":"release_notes/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes to pass tests on Julia 1.6.\nFixed a bug in the printing of nonlinear expressions in IJulia.\nFixed a bug when nonlinear expressions are passed to user-defined functions.\nSome internal functions that were previously exported are now no longer exported.\nFixed a bug when relaxing a fixed binary variable.\nFixed a StackOverflowError that occurred when SparseAxisArrays had a large number of elements.\nRemoved an unnecessary type assertion in list_of_constraint_types.\nFixed a bug when copying models with registered expressions.","category":"page"},{"location":"release_notes/#Other-32","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in /examples have now been moved to /docs/src/examples and rewritten using Literate.jl, and they are now included in the documentation.\nJuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.\nThe root cause of a large number of method invalidations has been resolved.\nWe switched continuous integration from Travis and Appveyor to GitHub Actions.","category":"page"},{"location":"release_notes/#[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)","page":"Release notes","title":"Version 0.21.5 (September 18, 2020)","text":"","category":"section"},{"location":"release_notes/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed deprecation warnings\nThrow DimensionMismatch for incompatibly sized functions and sets\nUnify treatment of keys(x) on JuMP containers","category":"page"},{"location":"release_notes/#[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)","page":"Release notes","title":"Version 0.21.4 (September 14, 2020)","text":"","category":"section"},{"location":"release_notes/#Added-28","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add debug info when adding unsupported constraints\nAdd relax_integrality for solving continuous relaxation\nAllow querying constraint conflicts","category":"page"},{"location":"release_notes/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Dispatch on Real for MOI.submit\nImplement copy for CustomSet in tests\nDon't export private macros\nFix invalid assertion in nonlinear\nError if constraint has NaN right-hand side\nImprove speed of tests\nLots of work modularizing files in /test\nImprove line numbers in macro error messages\nPrint nonlinear subexpressions\nVarious documentation updates\nDependency updates:\nDatastructures 0.18\nMathOptFormat v0.5\nPrep for MathOptInterface 0.9.15","category":"page"},{"location":"release_notes/#[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)","page":"Release notes","title":"Version 0.21.3 (June 18, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (#2212).\nAdded functions simplex_iterations, barrier_iterations and node_count (#2201).\nAdded function reduced_cost (#2205).\nImplemented callback_value for affine and quadratic expressions (#2231).\nSupport MutableArithmetics.Zero in objective and constraints (#2219).\nDocumentation improvements:\nMention tutorials in the docs (#2223).\nUpdate COIN-OR links (#2242).\nExplicit link to the documentation of MOI.FileFormats (#2253).\nTypo fixes (#2261).\nContainers improvements:\nFix Base.map for DenseAxisArray (#2235).\nThrow BoundsError if number of indices is incorrect for DenseAxisArray and SparseAxisArray (#2240).\nExtensibility improvements:\nImplement a set_objective method fallback that redirects to set_objective_sense and set_objective_function (#2247).\nAdd parse_constraint method with arbitrary number of arguments (#2051).\nAdd parse_constraint_expr and parse_constraint_head (#2228).","category":"page"},{"location":"release_notes/#[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)","page":"Release notes","title":"Version 0.21.2 (April 2, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added relative_gap() to access MOI.RelativeGap() attribute (#2199).\nDocumentation fixes:\nAdded link to source for docstrings in the documentation (#2207).\nAdded docstring for @variables macro (#2216).\nTypo fixes (#2177, #2184, #2182).\nImplementation of methods for Base functions:\nImplemented Base.empty! for JuMP.Model (#2198).\nImplemented Base.conj for JuMP scalar types (#2209).","category":"page"},{"location":"release_notes/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed sum of expression with scalar product in macro (#2178).\nFixed writing of nonlinear models to MathOptFormat (#2181).\nFixed construction of empty SparseAxisArray (#2179).\nFixed constraint with zero function (#2188).","category":"page"},{"location":"release_notes/#[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)","page":"Release notes","title":"Version 0.21.1 (Feb 18, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the clarity of the with_optimizer deprecation warning.","category":"page"},{"location":"release_notes/#[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)","page":"Release notes","title":"Version 0.21.0 (Feb 16, 2020)","text":"","category":"section"},{"location":"release_notes/#Breaking-6","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Deprecated with_optimizer (#2090, #2084, #2141). You can replace with_optimizer by either nothing, optimizer_with_attributes or a closure:\nreplace with_optimizer(Ipopt.Optimizer) by Ipopt.Optimizer.\nreplace with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0) by optimizer_with_attributes(Ipopt.Optimizer, \"max_cpu_time\" => 60.0).\nreplace with_optimizer(Gurobi.Optimizer, env) by () -> Gurobi.Optimizer(env).\nreplace with_optimizer(Gurobi.Optimizer, env, Presolve=0) by optimizer_with_attributes(() -> Gurobi.Optimizer(env), \"Presolve\" => 0).\nalternatively to optimizer_with_attributes, you can also set the attributes separately with set_optimizer_attribute.\nRenamed set_parameter and set_parameters to set_optimizer_attribute and set_optimizer_attributes (#2150).\nBroadcast should now be explicit inside macros. @SDconstraint(model, x >= 1) and @constraint(model, x + 1 in SecondOrderCone()) now throw an error instead of broadcasting 1 along the dimension of x (#2107).\n@SDconstraint(model, x >= 0) is now equivalent to @constraint(model, x in PSDCone()) instead of @constraint(model, (x .- 0) in PSDCone()) (#2107).\nThe macros now create the containers with map instead of for loops, as a consequence, containers created by @expression can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where @expression could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (#2070).\nCalling deepcopy(::AbstractModel) now throws an error.\nThe constraint name is now printed in the model string (#2108).","category":"page"},{"location":"release_notes/#Added-29","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for solver-independent and solver-specific callbacks (#2101).\nAdded write_to_file and read_from_file, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (#2114).\nAdded support for complementarity constraints (#2132).\nAdded support for indicator constraints (#2092).\nAdded support for querying multiple solutions with the result keyword (#2100).\nAdded support for constraining variables on creation (#2128).\nAdded method delete that deletes a vector of variables at once if it is supported by the underlying solver (#2135).\nThe arithmetic between JuMP expression has be refactored into the MutableArithmetics package (#2107).\nImproved error on complex values in NLP (#1978).\nAdded an example of column generation (#2010).","category":"page"},{"location":"release_notes/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Incorrect coefficients generated when using Symmetric variables (#2102)","category":"page"},{"location":"release_notes/#[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)","page":"Release notes","title":"Version 0.20.1 (Oct 18, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add sections on @variables and @constraints in the documentation (#2062).\nFixed product of sparse matrices for Julia v1.3 (#2063).\nAdded set_objective_coefficient to modify the coefficient of a linear term of the objective function (#2008).\nAdded set_time_limit_sec, unset_time_limit_sec and time_limit_sec to set and query the time limit for the solver in seconds (#2053).","category":"page"},{"location":"release_notes/#[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)","page":"Release notes","title":"Version 0.20.0 (Aug 24, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Documentation updates.\nNumerous bug fixes.\nBetter error messages (#1977, #1978, #1997, #2017).\nPerformance improvements (#1947, #2032).\nAdded LP sensitivity summary functions lp_objective_perturbation_range and lp_rhs_perturbation_range (#1917).\nAdded functions dual_objective_value, raw_status and set_parameter.\nAdded function set_objective_coefficient to modify the coefficient of a linear term of the objective (#2008).\nAdded functions set_normalized_rhs, normalized_rhs, and add_to_function_constant to modify and get the constant part of a constraint (#1935, #1960).\nAdded functions set_normalized_coefficient and normalized_coefficient to modify and get the coefficient of a linear term of a constraint (#1935, #1960).\nNumerous other improvements in MOI 0.9, see the NEWS.md file of MOI for more details.","category":"page"},{"location":"release_notes/#[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)","page":"Release notes","title":"Version 0.19.2 (June 8, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.","category":"page"},{"location":"release_notes/#[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)","page":"Release notes","title":"Version 0.19.1 (May 12, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Usability and performance improvements.\nBug fixes.","category":"page"},{"location":"release_notes/#[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)","page":"Release notes","title":"Version 0.19.0 (February 15, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP 0.19 contains significant breaking changes.","category":"page"},{"location":"release_notes/#Breaking-7","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP's abstraction layer for communicating with solvers changed from MathProgBase (MPB) to MathOptInterface (MOI). MOI addresses many longstanding design issues. (See @mlubin's slides from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the installation guide for a list of solvers that have and have not yet been updated.\nMost solvers have been renamed to PackageName.Optimizer. For example, GurobiSolver() is now Gurobi.Optimizer.\nSolvers are no longer added to a model via Model(solver = XXX(kwargs...)). Instead use Model(with_optimizer(XXX, kwargs...)). For example, Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0)).\nJuMP containers (for example, the objects returned by @variable) have been redesigned. Containers.SparseAxisArray replaces JuMPDict, JuMPArray was rewritten (inspired by AxisArrays) and renamed Containers.DenseAxisArray, and you can now request a container type with the container= keyword to the macros. See the corresponding documentation for more details.\nThe statuses returned by solvers have changed. See the possible status values here. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).\nStarting values are separate from result values. Use value to query the value of a variable in a solution. Use start_value and set_start_value to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.\nThe data structures for affine and quadratic expressions AffExpr and QuadExpr have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.\nJuMPNLPEvaluator no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.\nJuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)\nThe norm() syntax is no longer recognized inside macros. Use the SecondOrderCone() set instead.\nJuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.\nThe symbols :Min and :Max are no longer used as optimization senses. Instead, JuMP uses the OptimizationSense enum from MathOptInterface. @objective(model, Max, ...), @objective(model, Min, ...), @NLobjective(model, Max, ...), and @objective(model, Min, ...) remain valid, but @objective(m, :Max, ...) is no longer accepted.\nThe sign conventions for duals has changed in some cases for consistency with conic duality (see the documentation). The shadow_price helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.\n@constraintref is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,\nconstraints = Dict() # Optionally, specify types for improved performance.\nfor i in 1:N\n  constraints[i] = @constraint(model, ...)\nend\nThe lowerbound, upperbound, and basename keyword arguments to the @variable macro have been renamed to lower_bound, upper_bound, and base_name, for consistency with JuMP's new style recommendations.\nWe rely on broadcasting syntax to apply accessors to collections of variables, for example, value.(x) instead of getvalue(x) for collections. (Use value(x) when x is a scalar object.)","category":"page"},{"location":"release_notes/#Added-30","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Splatting (like f(x...)) is recognized in restricted settings in nonlinear expressions.\nSupport for deleting constraints and variables.\nThe documentation has been completely rewritten using docstrings and Documenter.\nSupport for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).\nSignificantly improved support for modeling new types of constraints and for extending JuMP's macros.\nSupport for providing dual warm starts.\nImproved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).\nExplicit control of whether symmetry-enforcing constraints are added to PSD constraints.\nSupport for modeling exponential cones.\nSignificant improvements in internal code quality and testing.\nStyle and naming guidelines.\nDirect mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding documentation.","category":"page"},{"location":"release_notes/#Regressions","page":"Release notes","title":"Regressions","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Performance regressions in model generation (issue). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.\nFast incremental NLP solves are not yet reimplemented (issue).\nWe do not yet have an implementation of solver-specific callbacks.\nThe column generation syntax in @variable has been removed (that is, the objective, coefficients, and inconstraints keyword arguments). Support for column generation will be re-introduced in a future release.\nThe ability to solve the continuous relaxation (that is, via solve(model; relaxation = true)) is not yet reimplemented (issue).","category":"page"},{"location":"release_notes/#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)","page":"Release notes","title":"Version 0.18.5 (December 1, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support views in some derivative evaluation functions.\nImproved compatibility with PackageCompiler.","category":"page"},{"location":"release_notes/#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)","page":"Release notes","title":"Version 0.18.4 (October 8, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in model printing on Julia 0.7 and 1.0.","category":"page"},{"location":"release_notes/#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)","page":"Release notes","title":"Version 0.18.3 (October 1, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for Julia v1.0 (Thanks @ExpandingMan)\nFix matrix expressions with quadratic functions (#1508)","category":"page"},{"location":"release_notes/#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)","page":"Release notes","title":"Version 0.18.2 (June 10, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in second-order derivatives when expressions are present (#1319)\nFix a bug in @constraintref (#1330)","category":"page"},{"location":"release_notes/#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)","page":"Release notes","title":"Version 0.18.1 (April 9, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for nested tuple destructuring (#1193)\nPreserve internal model when relaxation=true (#1209)\nMinor bug fixes and updates for example","category":"page"},{"location":"release_notes/#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)","page":"Release notes","title":"Version 0.18.0 (July 27, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Drop support for Julia 0.5.\nUpdate for ForwardDiff 0.5.\nMinor bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)","page":"Release notes","title":"Version 0.17.1 (June 9, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Use of constructconstraint! in @SDconstraint.\nMinor bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)","page":"Release notes","title":"Version 0.17.0 (May 27, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: Mixing quadratic and conic constraints is no longer supported.\nBreaking change: The getvariable and getconstraint functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name x, one should now write m[:x] instead of getvariable(m, :x). As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.\nAddition of the getobjectivebound function that mirrors the functionality of the MathProgBase getobjbound function except that it takes into account transformations performed by JuMP.\nMinor bug fixes.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The following changes are primarily of interest to developers of JuMP extensions:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The new syntax @constraint(model, expr in Cone) creates the constraint ensuring that expr is inside Cone. The Cone argument is passed to constructconstraint! which enables the call to the dispatched to an extension.\nThe @variable macro now calls constructvariable! instead of directly calling the Variable constructor. Extra arguments and keyword arguments passed to @variable are passed to constructvariable! which enables the call to be dispatched to an extension.\nRefactor the internal function conicdata (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.","category":"page"},{"location":"release_notes/#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)","page":"Release notes","title":"Version 0.16.2 (March 28, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Minor bug fixes and printing tweaks\nAddress deprecation warnings for Julia 0.6","category":"page"},{"location":"release_notes/#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)","page":"Release notes","title":"Version 0.16.1 (March 7, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Better support for AbstractArray in JuMP (Thanks @tkoolen)\nMinor bug fixes","category":"page"},{"location":"release_notes/#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)","page":"Release notes","title":"Version 0.16.0 (February 23, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling solve() will result in an error.\nBreaking change: User-defined functions are no longer global. The first argument to JuMP.register is now a JuMP Model object within whose scope the function will be registered. Calling JuMP.register without a Model now produces an error.\nBreaking change: Use the new JuMP.fix method to fix a variable to a value or to update the value to which a variable is fixed. Calling setvalue on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)\nNonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)\nNew category keyword to @variable. Used for specifying categories of anonymous variables.\nCompatibility with Julia 0.6-dev.\nMinor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)","category":"page"},{"location":"release_notes/#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)","page":"Release notes","title":"Version 0.15.1 (January 31, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bugfix for @LinearConstraints and friends","category":"page"},{"location":"release_notes/#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)","page":"Release notes","title":"Version 0.15.0 (December 22, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.5.0 is the minimum required version for this release.\nDocument support for BARON solver\nEnable info callbacks in more states than before, for example, for recording solutions. New when argument to addinfocallback (#814, thanks @yeesian)\nImproved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:\nWhen multiple variables in a model are given the same name\nWhen non-symbols are used as names, for example, @variable(m, x[1][1:N])\nImprovements in iterating over JuMP containers (#836, thanks @IssamT)\nSupport for writing variable names in .lp file output (Thanks @leethargo)\nSupport for querying duals to SDP problems (Thanks @blegat)\nThe comprehension syntax with curly braces sum{}, prod{}, and norm2{} has been deprecated in favor of Julia's native comprehension syntax sum(), prod() and norm() as previously announced. (For early adopters of the new syntax, norm2() was renamed to norm() without deprecation.)\nUnit tests rewritten to use Base.Test instead of FactCheck\nImproved support for operations with matrices of JuMP types (Thanks @ExpandingMan)\nThe syntax to halt a solver from inside a callback has changed from throw(CallbackAbort()) to return JuMP.StopTheSolver\nMinor bug fixes","category":"page"},{"location":"release_notes/#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)","page":"Release notes","title":"Version 0.14.2 (December 12, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Allow singleton anonymous variables (includes bugfix)","category":"page"},{"location":"release_notes/#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)","page":"Release notes","title":"Version 0.14.1 (September 12, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"More consistent handling of states in informational callbacks, includes a new when parameter to addinfocallback for specifying in which state an informational callback should be called.","category":"page"},{"location":"release_notes/#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)","page":"Release notes","title":"Version 0.14.0 (August 7, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility with Julia 0.5 and ForwardDiff 0.2\nSupport for \"anonymous\" variables, constraints, expressions, and parameters, for example, x = @variable(m, [1:N]) instead of @variable(m, x[1:N])\nSupport for retrieving constraints from a model by name via getconstraint\n@NLconstraint now returns constraint references (as expected).\nSupport for vectorized expressions within lazy constraints\nOn Julia 0.5, parse new comprehension syntax sum(x[i] for i in 1:N if isodd(i)) instead of sum{ x[i], i in 1:N; isodd(i) }. The old syntax with curly braces will be deprecated in JuMP 0.15.\nNow possible to provide nonlinear expressions as \"raw\" Julia Expr objects instead of using JuMP's nonlinear macros. This input format is useful for programmatically generated expressions.\ns/Mathematical Programming/Mathematical Optimization/\nSupport for local cuts (Thanks to @madanim, Mehdi Madani)\nDocument Xpress interface developed by @joaquimg, Joaquim Dias Garcia\nMinor bug and deprecation fixes (Thanks @odow, @jrevels)","category":"page"},{"location":"release_notes/#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)","page":"Release notes","title":"Version 0.13.2 (May 16, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility update for MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)","page":"Release notes","title":"Version 0.13.1 (May 3, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix broken deprecation for registerNLfunction.","category":"page"},{"location":"release_notes/#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)","page":"Release notes","title":"Version 0.13.0 (April 29, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Most exported methods and macros have been renamed to avoid camelCase. See the list of changes here. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.\nSpecify variable lower/upper bounds in @variable using the lowerbound and upperbound keyword arguments.\nChange name printed for variable using the basename keyword argument to @variable.\nNew @variables macro allows multi-line declaration of groups of variables.\nA number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was recorded live.\nCompatibility fixes with Julia 0.5.\nThe \"end\" indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see here.","category":"page"},{"location":"release_notes/#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)","page":"Release notes","title":"Version 0.12.2 (March 9, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Small fixes for nonlinear optimization","category":"page"},{"location":"release_notes/#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)","page":"Release notes","title":"Version 0.12.1 (March 1, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)","category":"page"},{"location":"release_notes/#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)","page":"Release notes","title":"Version 0.12.0 (February 27, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:\n@defExpr and @defNLExpr now take the model as the first argument. The previous one-argument version of @defExpr is deprecated; all expressions should be named. For example, replace @defExpr(2x+y) with @defExpr(jump_model, my_expr, 2x+y).\nJuMP no longer uses Julia's variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.\nSupport for user-defined functions integrated within nonlinear JuMP expressions.\nReplaced iteration over AffExpr with Number-like scalar iteration; previous iteration behavior is now available via linearterms(::AffExpr).\nStopping the solver via throw(CallbackAbort()) from a callback no longer triggers an exception. Instead, solve() returns UserLimit status.\ngetDual() now works for conic problems (Thanks @emreyamangil.)","category":"page"},{"location":"release_notes/#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)","page":"Release notes","title":"Version 0.11.3 (February 4, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bug-fix for problems with quadratic objectives and semidefinite constraints","category":"page"},{"location":"release_notes/#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)","page":"Release notes","title":"Version 0.11.2 (January 14, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility update for Mosek","category":"page"},{"location":"release_notes/#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)","page":"Release notes","title":"Version 0.11.1 (December 1, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Remove usage of @compat in tests.\nFix updating quadratic objectives for nonlinear models.","category":"page"},{"location":"release_notes/#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)","page":"Release notes","title":"Version 0.11.0 (November 30, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.4.0 is the minimum required version for this release.\nFix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.\nAddition of the solve(m::Model, relaxation=true) keyword argument to solve the standard continuous relaxation of model m\nThe getConstraintBounds() method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)","page":"Release notes","title":"Version 0.10.3 (November 20, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a rare error when parsing quadratic expressions\nFix Variable() constructor with default arguments\nDetect unrecognized keywords in solve()","category":"page"},{"location":"release_notes/#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)","page":"Release notes","title":"Version 0.10.2 (September 28, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for deprecation warnings","category":"page"},{"location":"release_notes/#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)","page":"Release notes","title":"Version 0.10.1 (September 3, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixes for ambiguity warnings.\nFix for breaking change in precompilation syntax in Julia 0.4-pre","category":"page"},{"location":"release_notes/#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)","page":"Release notes","title":"Version 0.10.0 (August 31, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support (on Julia 0.4 and later) for conditions in indexing @defVar and @addConstraint constructs, for example, @defVar(m, x[i=1:5,j=1:5; i+j >= 3])\nSupport for vectorized operations on Variables and expressions. See the documentation for details.\nNew getVar() method to access variables in a model by name\nSupport for semidefinite programming.\nDual solutions are now available for general nonlinear problems. You may call getDual on a reference object for a nonlinear constraint, and getDual on a variable object for Lagrange multipliers from active bounds.\nIntroduce warnings for two common performance traps: too many calls to getValue() on a collection of variables and use of the + operator in a loop to sum expressions.\nSecond-order cone constraints can be written directly with the norm() and norm2{} syntax.\nImplement MathProgBase interface for querying Hessian-vector products.\nIteration over JuMPContainers is deprecated; instead, use the keys and values functions, and zip(keys(d),values(d)) for the old behavior.\n@defVar returns Array{Variable,N} when each of N index sets are of the form 1:nᵢ.\nModule precompilation: on Julia 0.4 and later, using JuMP is now much faster.","category":"page"},{"location":"release_notes/#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)","page":"Release notes","title":"Version 0.9.3 (August 11, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixes for FactCheck testing on julia v0.4.","category":"page"},{"location":"release_notes/#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)","page":"Release notes","title":"Version 0.9.2 (June 27, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in @addConstraints.","category":"page"},{"location":"release_notes/#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)","page":"Release notes","title":"Version 0.9.1 (April 25, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for Julia 0.4-dev.\nSmall infrastructure improvements for extensions.","category":"page"},{"location":"release_notes/#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)","page":"Release notes","title":"Version 0.9.0 (April 18, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Comparison operators for constructing constraints (for example, 2x >= 1) have been deprecated. Instead, construct the constraints explicitly in the @addConstraint macro to add them to the model, or in the @LinearConstraint macro to create a stand-alone linear constraint instance.\ngetValue() method implemented to compute the value of a nonlinear subexpression\nJuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.\nA number of performance improvements in ReverseDiffSparse for computing derivatives.\nMathProgBase.getsolvetime(m) now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)\nFormatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).","category":"page"},{"location":"release_notes/#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)","page":"Release notes","title":"Version 0.8.0 (February 17, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Nonlinear subexpressions now supported with the @defNLExpr macro.\nSCS supported for solving second-order conic problems.\nsetXXXCallback family deprecated in favor of addXXXCallback.\nMultiple callbacks of the same type can be registered.\nAdded support for informational callbacks via addInfoCallback.\nA CallbackAbort exception can be thrown from callback to safely exit optimization.","category":"page"},{"location":"release_notes/#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)","page":"Release notes","title":"Version 0.7.4 (February 4, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.\nTwo-sided nonlinear constraints are supported.\nMethods for accessing the number of variables and constraints in a model are renamed.\nNew default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.\nSmall bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)","page":"Release notes","title":"Version 0.7.3 (January 14, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a method ambiguity conflict with Compose.jl (cosmetic fix)","category":"page"},{"location":"release_notes/#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)","page":"Release notes","title":"Version 0.7.2 (January 9, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in sum(::JuMPDict)\nAdded the setCategory function to change a variables category (for example, continuous or binary)","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"after construction, and getCategory to retrieve the variable category.","category":"page"},{"location":"release_notes/#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)","page":"Release notes","title":"Version 0.7.1 (January 2, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.","category":"page"},{"location":"release_notes/#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)","page":"Release notes","title":"Version 0.7.0 (December 29, 2014)","text":"","category":"section"},{"location":"release_notes/#Linear/quadratic/conic-programming","page":"Release notes","title":"Linear/quadratic/conic programming","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: The syntax for column-wise model generation has been changed to use keyword arguments in @defVar.\nOn Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.\nECOS supported for solving second-order conic problems.","category":"page"},{"location":"release_notes/#_nonlinear_programming_release_notes","page":"Release notes","title":"Nonlinear programming","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for skipping model generation when solving a sequence of nonlinear models with changing data.\nFix a memory leak when solving a sequence of nonlinear models.\nThe @addNLConstraint macro now supports the three-argument version to define sets of nonlinear constraints.\nKNITRO supported as a nonlinear solver.\nSpeed improvements for model generation.\nThe @addNLConstraints macro supports adding multiple (groups of) constraints at once. Syntax is similar to @addConstraints.\nDiscrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).","category":"page"},{"location":"release_notes/#General","page":"Release notes","title":"General","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Starting values for variables may now be specified with @defVar(m, x, start=value).\nThe setSolver function allows users to change the solver subsequent to model creation.\nSupport for \"fixed\" variables via the @defVar(m, x == 1) syntax.\nUnit tests rewritten to use FactCheck.jl, improved testing across solvers.","category":"page"},{"location":"release_notes/#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)","page":"Release notes","title":"Version 0.6.3 (October 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in multiplying two AffExpr objects.","category":"page"},{"location":"release_notes/#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)","page":"Release notes","title":"Version 0.6.2 (October 11, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Further improvements and bug fixes for printing.\nFixed a bug in @defExpr.\nSupport for accessing expression graphs through the MathProgBase NLP interface.","category":"page"},{"location":"release_notes/#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)","page":"Release notes","title":"Version 0.6.1 (September 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improvements and bug fixes for printing.","category":"page"},{"location":"release_notes/#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)","page":"Release notes","title":"Version 0.6.0 (September 9, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.3.0 is the minimum required version for this release.\nbuildInternalModel(m::Model) added to build solver-level model in memory without optimizing.\nDeprecate load_model_only keyword argument to solve.\nAdd groups of constraints with @addConstraints macro.\nUnicode operators now supported, including ∑ for sum, ∏ for prod, and ≤/≥\nQuadratic constraints supported in @addConstraint macro.\nQuadratic objectives supported in @setObjective macro.\nMathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems\nBreaking change: IpoptOptions no longer supported to specify solver options, use m = Model(solver=IpoptSolver(options...)) instead.\nNew solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK\nNew option to control whether the lazy constraint callback is executed at each node in the B&B tree or just when feasible solutions are found\nAdd support for semicontinuous and semi-integer variables for those solvers that support them.\nAdd support for index dependencies (for example, triangular indexing) in @defVar, @addConstraint, and @defExpr (for example, @defVar(m, x[i=1:10,j=i:10])).\nThis required some changes to the internal structure of JuMP containers, which may break code that explicitly stored JuMPDict objects.","category":"page"},{"location":"release_notes/#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)","page":"Release notes","title":"Version 0.5.8 (September 24, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug with specifying solvers (affects Julia 0.2 only)","category":"page"},{"location":"release_notes/#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)","page":"Release notes","title":"Version 0.5.7 (September 5, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in printing models","category":"page"},{"location":"release_notes/#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)","page":"Release notes","title":"Version 0.5.6 (September 2, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for semicontinuous and semi-integer variables for those solvers that support them.\nBreaking change: Syntax for Variable() constructor has changed (use of this interface remains discouraged)\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)","page":"Release notes","title":"Version 0.5.5 (July 6, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.","category":"page"},{"location":"release_notes/#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)","page":"Release notes","title":"Version 0.5.4 (June 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for breaking change in MathProgBase which reduces loading times for using JuMP\nFix error when MIPs not solved to optimality","category":"page"},{"location":"release_notes/#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)","page":"Release notes","title":"Version 0.5.3 (May 21, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for breaking change in ReverseDiffSparse","category":"page"},{"location":"release_notes/#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)","page":"Release notes","title":"Version 0.5.2 (May 9, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix compatibility with Julia 0.3 prerelease","category":"page"},{"location":"release_notes/#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)","page":"Release notes","title":"Version 0.5.1 (May 5, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in coefficient handling inside lazy constraints and user cuts","category":"page"},{"location":"release_notes/#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)","page":"Release notes","title":"Version 0.5.0 (May 2, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.\ngetValue for AffExpr and QuadExpr\nBreaking change: getSolverModel replaced by getInternalModel, which returns the internal MathProgBase-level model\nGroups of constraints can be specified with @addConstraint (see documentation for details). This is not a breaking change.\ndot(::JuMPDict{Variable},::JuMPDict{Variable}) now returns the corresponding quadratic expression.","category":"page"},{"location":"release_notes/#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)","page":"Release notes","title":"Version 0.4.1 (March 24, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug where change in objective sense was ignored when re-solving a model.\nFix issue with handling zero coefficients in AffExpr.","category":"page"},{"location":"release_notes/#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)","page":"Release notes","title":"Version 0.4.0 (March 10, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for SOS1 and SOS2 constraints.\nSolver-independent callback for user heuristics.\ndot and sum implemented for JuMPDict objects. Now you can say @addConstraint(m, dot(a,x) <= b).\nDevelopers: support for extensions to JuMP. See definition of Model in src/JuMP.jl for more details.\nOption to construct the low-level model before optimizing.","category":"page"},{"location":"release_notes/#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)","page":"Release notes","title":"Version 0.3.2 (February 17, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved model printing\nPreliminary support for IJulia output","category":"page"},{"location":"release_notes/#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)","page":"Release notes","title":"Version 0.3.1 (January 30, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Documentation updates\nSupport for MOSEK\nCPLEXLink renamed to CPLEX","category":"page"},{"location":"release_notes/#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)","page":"Release notes","title":"Version 0.3.0 (January 21, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Unbounded/infeasibility rays: getValue() will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. getDual() will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.\nSolver-independent callbacks for user generated cuts.\nUse new interface for solver-independent QCQP.\nsetlazycallback renamed to setLazyCallback for consistency.","category":"page"},{"location":"release_notes/#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)","page":"Release notes","title":"Version 0.2.0 (December 15, 2013)","text":"","category":"section"},{"location":"release_notes/#Breaking-8","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Objective sense is specified in setObjective instead of in the Model constructor.\nlpsolver and mipsolver merged into single solver option.","category":"page"},{"location":"release_notes/#Added-31","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Problem modification with efficient LP restarts and MIP warm-starts.\nRelatedly, column-wise modeling now supported.\nSolver-independent callbacks supported. Currently we support only a \"lazy constraint\" callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.","category":"page"},{"location":"release_notes/#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)","page":"Release notes","title":"Version 0.1.2 (November 16, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bug fixes for printing, improved error messages.\nAllow AffExpr to be used in macros; for example, ex = y + z; @addConstraint(m, x + 2*ex <= 3)","category":"page"},{"location":"release_notes/#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)","page":"Release notes","title":"Version 0.1.1 (October 23, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for solver specification API changes in MathProgBase.","category":"page"},{"location":"release_notes/#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)","page":"Release notes","title":"Version 0.1.0 (October 3, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Initial public release.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\n    import HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/constraints/#jump_constraints","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP is based on the MathOptInterface (MOI) API. Because of this, JuMP uses the following standard form to represent problems:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Each constraint, f_i(x) in mathcalS_i, is composed of a function and a set. For example, instead of calling a^top x le b a less-than-or-equal-to constraint, we say that it is a scalar-affine-in-less-than constraint, where the function a^top x belongs to the less-than set (-infty b. We use the shorthand function-in-set to refer to constraints composed of different types of functions and sets.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This page explains how to write various types of constraints in JuMP. For nonlinear constraints, see Nonlinear Modeling instead.","category":"page"},{"location":"manual/constraints/#Add-a-constraint","page":"Constraints","title":"Add a constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Add a constraint to a JuMP model using the @constraint macro. The syntax to use depends on the type of constraint you wish to add.","category":"page"},{"location":"manual/constraints/#Add-a-linear-constraint","page":"Constraints","title":"Add a linear constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create linear constraints using the @constraint macro:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c1, sum(x) <= 1)\nc1 : x[1] + x[2] + x[3] ≤ 1\n\njulia> @constraint(model, c2, x[1] + 2 * x[3] >= 2)\nc2 : x[1] + 2 x[3] ≥ 2\n\njulia> @constraint(model, c3, sum(i * x[i] for i in 1:3) == 3)\nc3 : x[1] + 2 x[2] + 3 x[3] = 3\n\njulia> @constraint(model, c4, 4 <= 2 * x[2] <= 5)\nc4 : 2 x[2] ∈ [4, 5]","category":"page"},{"location":"manual/constraints/#Normalization","page":"Constraints","title":"Normalization","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP normalizes constraints by moving all of the terms containing variables to the left-hand side and all of the constant terms to the right-hand side. Thus, we get:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, c, 2x + 1 <= 4x + 4)\nc : -2 x ≤ 3","category":"page"},{"location":"manual/constraints/#quad_constraints","page":"Constraints","title":"Add a quadratic constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In addition to affine functions, JuMP also supports constraints with quadratic terms. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @variable(model, t >= 0)\nt\n\njulia> @constraint(model, my_q, x[1]^2 + x[2]^2 <= t^2)\nmy_q : x[1]² + x[2]² - t² ≤ 0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nBecause solvers can take advantage of the knowledge that a constraint is quadratic, prefer adding quadratic constraints using @constraint, rather than @NLconstraint.","category":"page"},{"location":"manual/constraints/#Vectorized-constraints","page":"Constraints","title":"Vectorized constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also add constraints to JuMP using vectorized linear algebra. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> A = [1 2; 3 4]\n2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> b = [5, 6]\n2-element Vector{Int64}:\n 5\n 6\n\njulia> @constraint(model, con_vector, A * x == b)\ncon_vector : [x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)\n\njulia> @constraint(model, con_scalar, A * x .== b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}, ScalarShape}}:\n con_scalar : x[1] + 2 x[2] = 5\n con_scalar : 3 x[1] + 4 x[2] = 6","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The two constraints, == and .== are similar, but subtly different. The first creates a single constraint that is a MOI.VectorAffineFunction in MOI.Zeros constraint. The second creates a vector of MOI.ScalarAffineFunction in MOI.EqualTo constraints.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Which formulation to choose depends on the solver, and what you want to do with the constraint object con_vector or con_scalar.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you are using a conic solver, expect the dual of con_vector to be a Vector{Float64}, and do not intend to delete a row in the constraint, choose the == formulation.\nIf you are using a solver that expects a list of scalar constraints, for example HiGHS, or you wish to delete part of the constraint or access a single row of the constraint, for example, dual(con_scalar[2]), then use the broadcast .==.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP reformulates both constraints into the other form if needed by the solver, but choosing the right format for a particular solver is more efficient.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also use <=, .<= , >=, and .>= as comparison operators in the constraint.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, A * x <= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)\n\njulia> @constraint(model, A * x .<= b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x[1] + 2 x[2] ≤ 5\n 3 x[1] + 4 x[2] ≤ 6\n\njulia> @constraint(model, A * x >= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, A * x .>= b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n x[1] + 2 x[2] ≥ 5\n 3 x[1] + 4 x[2] ≥ 6","category":"page"},{"location":"manual/constraints/#Vectorized-matrix-constraints","page":"Constraints","title":"Vectorized matrix constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In most cases, you cannot use the non-broadcasting syntax for general matrices. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[2,1]  X[2,2]\n\njulia> @constraint(model, X >= 0)\nERROR: At none:1: `@constraint(model, X >= 0)`: Unsupported matrix in vector-valued set. Did you mean to use the broadcasting syntax `.>=` instead? Alternatively, perhaps you are missing a set argument like `@constraint(model, X >= 0, PSDCone())` or `@constraint(model, X >= 0, HermmitianPSDCone())`.\nStacktrace:\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Instead, to represent matrix inequalities you must always use the element-wise broadcasting .==, .>=, or .<=, or use the Set inequality syntax.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"There are two exceptions: if the result of the left-hand side minus the right-hand side is a LinearAlgebra.Symmetric matrix or a LinearAlgebra.Hermitian matrix, you may use the non-broadcasting equality syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> using LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2], Symmetric)\n2×2 Symmetric{VariableRef, Matrix{VariableRef}}:\n X[1,1]  X[1,2]\n X[1,2]  X[2,2]\n\njulia> @constraint(model, X == LinearAlgebra.I)\n[X[1,1] - 1  X[1,2];\n X[1,2]      X[2,2] - 1] ∈ Zeros()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Despite the model showing the matrix in Zeros, this will add only three rows to the constraint matrix because the symmetric constraints are redundant. In contrast, the broadcasting syntax adds four linear constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, X .== LinearAlgebra.I)\n2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}, ScalarShape}}:\n X[1,1] = 1  X[1,2] = 0\n X[1,2] = 0  X[2,2] = 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The same holds for LinearAlgebra.Hermitian matrices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> using LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2] in HermitianPSDCone())\n2×2 Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(X[1,1])                    real(X[1,2]) + imag(X[1,2]) im\n real(X[1,2]) - imag(X[1,2]) im  real(X[2,2])\n\njulia> @constraint(model, X == LinearAlgebra.I)\n[real(X[1,1]) - 1                real(X[1,2]) + imag(X[1,2]) im;\n real(X[1,2]) - imag(X[1,2]) im  real(X[2,2]) - 1] ∈ Zeros()\n\njulia> @constraint(model, X .== LinearAlgebra.I)\n2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{ComplexF64}, MathOptInterface.EqualTo{ComplexF64}}, ScalarShape}}:\n real(X[1,1]) = 1                    real(X[1,2]) + imag(X[1,2]) im = 0\n real(X[1,2]) - imag(X[1,2]) im = 0  real(X[2,2]) = 1","category":"page"},{"location":"manual/constraints/#Containers-of-constraints","page":"Constraints","title":"Containers of constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The @constraint macro supports creating collections of constraints. We'll cover some brief syntax here; read the Constraint containers section for more details:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create arrays of constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=1:3], x[i] <= i^2)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c[1] : x[1] ≤ 1\n c[2] : x[2] ≤ 4\n c[3] : x[3] ≤ 9\n\njulia> c[2]\nc[2] : x[2] ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Sets can be any Julia type that supports iteration:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=2:3, [\"red\", \"blue\"]], x[i] <= i^2)\n2-dimensional DenseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, [\"red\", \"blue\"]\nAnd data, a 2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c[2,red] : x[2] ≤ 4  c[2,blue] : x[2] ≤ 4\n c[3,red] : x[3] ≤ 9  c[3,blue] : x[3] ≤ 9\n\njulia> c[2, \"red\"]\nc[2,red] : x[2] ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Sets can depend upon previous indices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=1:3, j=i:3], x[i] <= j)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  c[1,1] : x[1] ≤ 1\n  [1, 2]  =  c[1,2] : x[1] ≤ 2\n  [1, 3]  =  c[1,3] : x[1] ≤ 3\n  [2, 2]  =  c[2,2] : x[2] ≤ 2\n  [2, 3]  =  c[2,3] : x[2] ≤ 3\n  [3, 3]  =  c[3,3] : x[3] ≤ 3","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and you can filter elements in the sets using the ; syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:9]);\n\njulia> @constraint(model, c[i=1:9; mod(i, 3) == 0], x[i] <= i)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 1, Tuple{Int64}} with 3 entries:\n  [3]  =  c[3] : x[3] ≤ 3\n  [6]  =  c[6] : x[6] ≤ 6\n  [9]  =  c[9] : x[9] ≤ 9","category":"page"},{"location":"manual/constraints/#Registered-constraints","page":"Constraints","title":"Registered constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"When you create constraints, JuMP registers them inside the model using their corresponding symbol. Get a registered name using model[:key]:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, my_c, 2x <= 1)\nmy_c : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: my_c, x\n\njulia> model[:my_c] === my_c\ntrue","category":"page"},{"location":"manual/constraints/#Anonymous-constraints","page":"Constraints","title":"Anonymous constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To reduce the likelihood of accidental bugs, and because JuMP registers constraints inside a model, creating two constraints with the same name is an error:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2x <= 1)\nc : 2 x ≤ 1\n\njulia> @constraint(model, c, 2x <= 1)\nERROR: An object of name c is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :c)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:c]`.\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A common reason for encountering this error is adding constraints in a loop.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"As a work-around, JuMP provides anonymous constraints. Create an anonymous constraint by omitting the name argument:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> c = @constraint(model, 2x <= 1)\n2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create a container of anonymous constraints by dropping the name in front of the [:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> c = @constraint(model, [i = 1:3], x[i] <= i)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x[1] ≤ 1\n x[2] ≤ 2\n x[3] ≤ 3","category":"page"},{"location":"manual/constraints/#Constraint-names","page":"Constraints","title":"Constraint names","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In addition to the symbol that constraints are registered with, constraints have a String name that is used for printing and writing to file formats.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Get and set the name of a constraint using name(::JuMP.ConstraintRef) and set_name(::JuMP.ConstraintRef, ::String):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> @constraint(model, con, x <= 1)\ncon : x ≤ 1\n\njulia> name(con)\n\"con\"\n\njulia> set_name(con, \"my_con_name\")\n\njulia> con\nmy_con_name : x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Override the default choice of name using the base_name keyword:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> con = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Note that names apply to each element of the container, not to the container of constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> name(con[1])\n\"my_con[1]\"\n\njulia> set_name(con[1], \"c\")\n\njulia> con\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nFor some models, setting the string name of each constraint can take a non-trivial portion of the total time required to build the model. Turn off String names by passing set_string_name = false to @constraint:julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, x <= 2, set_string_name = false)\nx ≤ 2See Disable string names for more information.","category":"page"},{"location":"manual/constraints/#Retrieve-a-constraint-by-name","page":"Constraints","title":"Retrieve a constraint by name","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Retrieve a constraint from a model using constraint_by_name:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> constraint_by_name(model, \"c\")\nc : x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the name is not present, nothing will be returned:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> constraint_by_name(model, \"bad_name\")","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can only look up individual constraints using constraint_by_name. Something like this will not work:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> con = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2\n\njulia> constraint_by_name(model, \"my_con\")","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To look up a collection of constraints, do not use constraint_by_name. Instead, register them using the model[:key] = value syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> model[:con] = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2\n\njulia> model[:con]\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/#String-names,-symbolic-names,-and-bindings","page":"Constraints","title":"String names, symbolic names, and bindings","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"It's common for new users to experience confusion relating to constraints. Part of the problem is the difference between the name that a constraint is registered under and the String name used for printing.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Here's a summary of the differences:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Constraints are created using @constraint.\nConstraints can be named or anonymous.\nNamed constraints have the form @constraint(model, c, expr). For named constraints:\nThe String name of the constraint is set to \"c\".\nA Julia variable c is created that binds c to  the JuMP constraint.\nThe name :c is registered as a key in the model with the value c.\nAnonymous constraints have the form c = @constraint(model, expr). For anonymous constraints:\nThe String name of the constraint is set to \"\".\nYou control the name of the Julia variable used as the binding.\nNo name is registered as a key in the model.\nThe base_name keyword can override the String name of the constraint.\nYou can manually register names in the model via model[:key] = value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Here's an example of the differences:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> c_binding = @constraint(model, 2x <= 1, base_name = \"c\")\nc : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x\n\njulia> c\nERROR: UndefVarError: `c` not defined\n\njulia> c_binding\nc : 2 x ≤ 1\n\njulia> name(c_binding)\n\"c\"\n\njulia> model[:c_register] = c_binding\nc : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: c_register, x\n\njulia> model[:c_register]\nc : 2 x ≤ 1\n\njulia> model[:c_register] === c_binding\ntrue\n\njulia> c\nERROR: UndefVarError: `c` not defined","category":"page"},{"location":"manual/constraints/#The-@constraints-macro","page":"Constraints","title":"The @constraints macro","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you have many @constraint calls, use the @constraints macro to improve readability:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraints(model, begin\n           2x <= 1\n           c, x >= -1\n       end)\n(2 x ≤ 1, c : x ≥ -1)\n\njulia> print(model)\nFeasibility\nSubject to\n c : x ≥ -1\n 2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The @constraints macro returns a tuple of the constraints that were defined.","category":"page"},{"location":"manual/constraints/#constraint_duality","page":"Constraints","title":"Duality","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP adopts the notion of conic duality from MathOptInterface. For linear programs, a feasible dual on a >= constraint is nonnegative and a feasible dual on a <= constraint is nonpositive. If the constraint is an equality constraint, it depends on which direction is binding.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nJuMP's definition of duality is independent of the objective sense. That is, the sign of feasible duals associated with a constraint depends on the direction of the constraint and not whether the problem is maximization or minimization. This is a different convention from linear programming duality in some common textbooks. If you have a linear program, and you want the textbook definition, you probably want to use shadow_price and reduced_cost instead.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The dual value associated with a constraint in the most recent solution can be accessed using the dual function. Use has_duals to check if the model has a dual solution available to query. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x <= 1)\ncon : x ≤ 1\n\njulia> @objective(model, Min, -2x)\n-2 x\n\njulia> has_duals(model)\nfalse\n\njulia> optimize!(model)\n\njulia> has_duals(model)\ntrue\n\njulia> dual(con)\n-2.0\n\njulia> @objective(model, Max, 2x)\n2 x\n\njulia> optimize!(model)\n\njulia> dual(con)\n-2.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To help users who may be less familiar with conic duality, JuMP provides shadow_price, which returns a value that can be interpreted as the improvement in the objective in response to an infinitesimal relaxation (on the scale of one unit) in the right-hand side of the constraint. shadow_price can be used only on linear constraints with a <=, >=, or == comparison operator.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In the example above, dual(con) returned -2.0 regardless of the optimization sense. However, in the second case when the optimization sense is Max, shadow_price returns:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> shadow_price(con)\n2.0","category":"page"},{"location":"manual/constraints/#Duals-of-variable-bounds","page":"Constraints","title":"Duals of variable bounds","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To query the dual variables associated with a variable bound, first obtain a constraint reference using one of UpperBoundRef, LowerBoundRef, or FixRef, and then call dual on the returned constraint reference. The reduced_cost function may simplify this process as it returns the shadow price of an active bound of a variable (or zero, if no active bound exists).","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x <= 1)\nx\n\njulia> @objective(model, Min, -2x)\n-2 x\n\njulia> optimize!(model)\n\njulia> dual(UpperBoundRef(x))\n-2.0\n\njulia> reduced_cost(x)\n-2.0","category":"page"},{"location":"manual/constraints/#Modify-a-constant-term","page":"Constraints","title":"Modify a constant term","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This section explains how to modify the constant term in a constraint. There are multiple ways to achieve this goal; we explain three options.","category":"page"},{"location":"manual/constraints/#Option-1:-change-the-right-hand-side","page":"Constraints","title":"Option 1: change the right-hand side","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use set_normalized_rhs to modify the right-hand side (constant) term of a linear or quadratic  constraint. Use normalized_rhs to query the right-hand side term.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> set_normalized_rhs(con, 3)\n\njulia> con\ncon : 2 x ≤ 3\n\njulia> normalized_rhs(con)\n3.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nset_normalized_rhs sets the right-hand side term of the normalized constraint. See Normalization for more details.","category":"page"},{"location":"manual/constraints/#Option-2:-use-fixed-variables","page":"Constraints","title":"Option 2: use fixed variables","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If constraints are complicated, for example, they are composed of a number of components, each of which has a constant term, then it may be difficult to calculate what the right-hand side term is in the standard form.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For this situation, JuMP includes the ability to fix variables to a value using the fix function. Fixing a variable sets its lower and upper bound to the same value. Thus, changes in a constant term can be simulated by adding a new variable and fixing it to different values. Here is an example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, const_term)\nconst_term\n\njulia> @constraint(model, con, 2x <= const_term + 1)\ncon : 2 x - const_term ≤ 1\n\njulia> fix(const_term, 1.0)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The constraint con is now equivalent to 2x <= 2.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nFixed variables are not replaced with constants when communicating the problem to a solver. Therefore, even though const_term is fixed, it is still a decision variable, and so const_term * x is bilinear.","category":"page"},{"location":"manual/constraints/#Option-3:-modify-the-function's-constant-term","page":"Constraints","title":"Option 3: modify the function's constant term","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The third option is to use add_to_function_constant. The constant given is added to the function of a func-in-set constraint. In the following example, adding 2 to the function has the effect of removing 2 to the right-hand side:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> add_to_function_constant(con, 2)\n\njulia> con\ncon : 2 x ≤ -1\n\njulia> normalized_rhs(con)\n-1.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In the case of interval constraints, the constant is removed from each bound:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 0 <= 2x + 1 <= 2)\ncon : 2 x ∈ [-1, 1]\n\njulia> add_to_function_constant(con, 3)\n\njulia> con\ncon : 2 x ∈ [-4, -2]","category":"page"},{"location":"manual/constraints/#Modify-a-variable-coefficient","page":"Constraints","title":"Modify a variable coefficient","text":"","category":"section"},{"location":"manual/constraints/#Scalar-constraints","page":"Constraints","title":"Scalar constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To modify the coefficients for a linear term (modifying the coefficient of a quadratic term is not supported) in a constraint, use set_normalized_coefficient. To query the current coefficient, use normalized_coefficient.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, con, 2x[1] + x[2] <= 1)\ncon : 2 x[1] + x[2] ≤ 1\n\njulia> set_normalized_coefficient(con, x[2], 0)\n\njulia> con\ncon : 2 x[1] ≤ 1\n\njulia> normalized_coefficient(con, x[2])\n0.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nset_normalized_coefficient sets the coefficient of the normalized constraint. See Normalization for more details.","category":"page"},{"location":"manual/constraints/#Vector-constraints","page":"Constraints","title":"Vector constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To modify the coefficients of a vector-valued constraint, use set_normalized_coefficients.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, [2x + 3x, 4x] in MOI.Nonnegatives(2))\ncon : [5 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 3.0)])\n\njulia> con\ncon : [3 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 2.0), (2, 5.0)])\n\njulia> con\ncon : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/#Delete-a-constraint","page":"Constraints","title":"Delete a constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use delete to delete a constraint from a model. Use is_valid to check if a constraint belongs to a model and has not been deleted.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> is_valid(model, con)\ntrue\n\njulia> delete(model, con)\n\njulia> is_valid(model, con)\nfalse","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Deleting a constraint does not unregister the symbolic reference from the model. Therefore, creating a new constraint of the same name will throw an error:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, con, 2x <= 1)\nERROR: An object of name con is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :con)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:con]`.\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"After calling delete, call unregister to remove the symbolic reference:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> unregister(model, :con)\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"info: Info\ndelete does not automatically unregister because we do not distinguish between names that are automatically registered by JuMP macros, and names that are manually registered by the user by setting values in object_dictionary. In addition, deleting a constraint and then adding a new constraint of the same name is an easy way to introduce bugs into your code.","category":"page"},{"location":"manual/constraints/#Start-values","page":"Constraints","title":"Start values","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Provide a starting value (also called warmstart) for a constraint's primal and dual solutions using set_start_value and set_dual_start_value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Query the starting value for a constraint's primal and dual solution using start_value and dual_start_value. If no start value has been set, the methods will return nothing.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x >= 10)\ncon : x ≥ 10\n\njulia> start_value(con)\n\njulia> set_start_value(con, 10.0)\n\njulia> start_value(con)\n10.0\n\njulia> dual_start_value(con)\n\njulia> set_dual_start_value(con, 2)\n\njulia> dual_start_value(con)\n2.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Vector-valued constraints require a vector:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, con, x in SecondOrderCone())\ncon : [x[1], x[2], x[3]] in MathOptInterface.SecondOrderCone(3)\n\njulia> dual_start_value(con)\n\njulia> set_dual_start_value(con, [1.0, 2.0, 3.0])\n\njulia> dual_start_value(con)\n3-element Vector{Float64}:\n 1.0\n 2.0\n 3.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nTo simplify setting start values for all variables and constraints in a model, see set_start_values. The Primal and dual warm-starts tutorial also gives a detailed description of how to iterate over constraints in the model to set custom start values.","category":"page"},{"location":"manual/constraints/#Constraint-containers","page":"Constraints","title":"Constraint containers","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Like Variable containers, JuMP provides a mechanism for building groups of constraints compactly. References to these groups of constraints are returned in containers. Three types of constraint containers are supported: Arrays, DenseAxisArrays, and SparseAxisArrays. We explain each of these in the following.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/constraints/#constraint_arrays","page":"Constraints","title":"Arrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"One way of adding a group of constraints compactly is the following:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:3], i * x <= i + 1)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[1] : x ≤ 2\n con[2] : 2 x ≤ 3\n con[3] : 3 x ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP returns references to the three constraints in an Array that is bound to the Julia variable con. This array can be accessed and sliced as you would with any Julia array:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con[1]\ncon[1] : x ≤ 2\n\njulia> con[2:3]\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[2] : 2 x ≤ 3\n con[3] : 3 x ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Anonymous containers can also be constructed by dropping the name (for example, con) before the square brackets:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con = @constraint(model, [i = 1:2], i * x <= i + 1)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x ≤ 2\n 2 x ≤ 3","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Just like @variable, JuMP will form an Array of constraints when it can determine at parse time that the indices are one-based integer ranges. Therefore con[1:b] will create an Array, but con[a:b] will not. A special case is con[Base.OneTo(n)] which will produce an Array. If JuMP cannot determine that the indices are one-based integer ranges (for example, in the case of con[a:b]), JuMP will create a DenseAxisArray instead.","category":"page"},{"location":"manual/constraints/#DenseAxisArrays","page":"Constraints","title":"DenseAxisArrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The syntax for constructing a DenseAxisArray of constraints is very similar to the syntax for constructing a DenseAxisArray of variables.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:2, j = 2:3], i * x <= j + 1)\n2-dimensional DenseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, 2:3\nAnd data, a 2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[1,2] : x ≤ 3    con[1,3] : x ≤ 4\n con[2,2] : 2 x ≤ 3  con[2,3] : 2 x ≤ 4","category":"page"},{"location":"manual/constraints/#SparseAxisArrays","page":"Constraints","title":"SparseAxisArrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The syntax for constructing a SparseAxisArray of constraints is very similar to the syntax for constructing a SparseAxisArray of variables.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:2, j = 1:2; i != j], i * x <= j + 1)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 2, Tuple{Int64, Int64}} with 2 entries:\n  [1, 2]  =  con[1,2] : x ≤ 3\n  [2, 1]  =  con[2,1] : 2 x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nIf you have many index dimensions and a large amount of sparsity, read Performance considerations.","category":"page"},{"location":"manual/constraints/#Forcing-the-container-type","page":"Constraints","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"When creating a container of constraints, JuMP will attempt to choose the tightest container type that can store the constraints. However, because this happens at parse time, it does not always make the best choice. Just like in @variable, you can force the type of container using the container keyword. For syntax and the reason behind this, take a look at the variable docs.","category":"page"},{"location":"manual/constraints/#Constraints-with-similar-indices","page":"Constraints","title":"Constraints with similar indices","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Containers are often used to create constraints over a set of indices. However, you'll often have cases in which you are repeating the indices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @variable(model, y[1:2]);\n\njulia> @constraints(model, begin\n           [i=1:2, j=1:2, k=1:2], i * x[j] <= k\n           [i=1:2, j=1:2, k=1:2], i * y[j] <= k\n       end);","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This is hard to read and leads to a lot of copy-paste. A more readable way is to use a for-loop:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> for i=1:2, j=1:2, k=1:2\n           @constraints(model, begin\n               i * x[j] <= k\n               i * y[j] <= k\n           end)\n       end","category":"page"},{"location":"manual/constraints/#Accessing-constraints-from-a-model","page":"Constraints","title":"Accessing constraints from a model","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Query the types of function-in-set constraints in a model using list_of_constraint_types:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2] >= i, Int);\n\njulia> @constraint(model, x[1] + x[2] <= 1);\n\njulia> list_of_constraint_types(model)\n3-element Vector{Tuple{Type, Type}}:\n (AffExpr, MathOptInterface.LessThan{Float64})\n (VariableRef, MathOptInterface.GreaterThan{Float64})\n (VariableRef, MathOptInterface.Integer)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For a given combination of function and set type, use num_constraints to access the number of constraints and all_constraints to access a list of their references:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> num_constraints(model, VariableRef, MOI.Integer)\n2\n\njulia> cons = all_constraints(model, VariableRef, MOI.Integer)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}, ScalarShape}}:\n x[1] integer\n x[2] integer","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also count the total number of constraints in the model, but you must explicitly choose whether to count VariableRef constraints such as bound and integrality constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> num_constraints(model; count_variable_in_set_constraints = true)\n5\n\njulia> num_constraints(model; count_variable_in_set_constraints = false)\n1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The same also applies for all_constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> all_constraints(model; include_variable_in_set_constraints = true)\n5-element Vector{ConstraintRef}:\n x[1] + x[2] ≤ 1\n x[1] ≥ 1\n x[2] ≥ 2\n x[1] integer\n x[2] integer\n\njulia> all_constraints(model; include_variable_in_set_constraints = false)\n1-element Vector{ConstraintRef}:\n x[1] + x[2] ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you need finer-grained control on which constraints to include, use a variant of:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> sum(\n           num_constraints(model, F, S) for\n           (F, S) in list_of_constraint_types(model) if F != VariableRef\n       )\n1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use constraint_object to get an instance of an AbstractConstraint object that stores the constraint data:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con = constraint_object(cons[1])\nScalarConstraint{VariableRef, MathOptInterface.Integer}(x[1], MathOptInterface.Integer())\n\njulia> con.func\nx[1]\n\njulia> con.set\nMathOptInterface.Integer()","category":"page"},{"location":"manual/constraints/#MathOptInterface-constraints","page":"Constraints","title":"MathOptInterface constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Because JuMP is based on MathOptInterface, you can add any constraints supported by MathOptInterface using the function-in-set syntax. For a list of supported functions and sets, read Standard form problem.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"note: Note\nWe use MOI as an alias for the MathOptInterface module. This alias is defined by using JuMP. You may also define it in your code as follows:import MathOptInterface as MOI","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, the following two constraints are equivalent:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, 2 * x[1] <= 1)\n2 x[1] ≤ 1\n\njulia> @constraint(model, 2 * x[1] in MOI.LessThan(1.0))\n2 x[1] ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also use any set defined by MathOptInterface:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x - [1; 2; 3] in MOI.Nonnegatives(3))\n[x[1] - 1, x[2] - 2, x[3] - 3] ∈ MathOptInterface.Nonnegatives(3)\n\njulia> @constraint(model, x in MOI.ExponentialCone())\n[x[1], x[2], x[3]] ∈ MathOptInterface.ExponentialCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"info: Info\nSimilar to how JuMP defines the <= and >= syntax as a convenience way to specify MOI.LessThan and MOI.GreaterThan constraints, the remaining sections in this page describe functions and syntax that have been added for the convenience of common modeling situations.","category":"page"},{"location":"manual/constraints/#Set-inequality-syntax","page":"Constraints","title":"Set inequality syntax","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For modeling convenience, the syntax @constraint(model, x >= y, Set()) is short-hand for @constraint(model, x - y in Set()). Therefore, the following calls are equivalent:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> y = [0.5, 0.75];\n\njulia> @constraint(model, x >= y, MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, y <= x, MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, x - y in MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Non-zero constants are not supported in this syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x >= 1, MOI.Nonnegatives(2))\nERROR: Operation `sub_mul` between `Vector{VariableRef}` and `Int64` is not allowed. This most often happens when you write a constraint like `x >= y` where `x` is an array and `y` is a constant. Use the broadcast syntax `x .- y >= 0` instead.\nStacktrace:\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use instead:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x .- 1 >= 0, MOI.Nonnegatives(2))\n[x[1] - 1, x[2] - 1] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/#Second-order-cone-constraints","page":"Constraints","title":"Second-order cone constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A SecondOrderCone constrains the variables t and x to the set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"x_2 le t","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and t ge 0. It can be added as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, t)\nt\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, [t; x] in SecondOrderCone())\n[t, x[1], x[2]] ∈ MathOptInterface.SecondOrderCone(3)","category":"page"},{"location":"manual/constraints/#Rotated-second-order-cone-constraints","page":"Constraints","title":"Rotated second-order cone constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A RotatedSecondOrderCone constrains the variables t, u, and x to the set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"x_2^2 le 2 t cdot u","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and t u ge 0. It can be added as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, t)\nt\n\njulia> @variable(model, u)\nu\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, [t; u; x] in RotatedSecondOrderCone())\n[t, u, x[1], x[2]] ∈ MathOptInterface.RotatedSecondOrderCone(4)","category":"page"},{"location":"manual/constraints/#Special-Ordered-Sets-of-Type-1","page":"Constraints","title":"Special Ordered Sets of Type 1","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In a Special Ordered Set of Type 1 (often denoted SOS-I or SOS1), at most one element can take a non-zero value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Construct SOS-I constraints using the SOS1 set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, x in SOS1())\n[x[1], x[2], x[3]] in MathOptInterface.SOS1{Float64}([1.0, 2.0, 3.0])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Although not required for feasibility, solvers can benefit from an ordering of the variables (for example, the variables represent different factories to build, at most one factory can be built, and the factories can be ordered according to cost). To induce an ordering, a vector of weights can be provided, and the variables are ordered according to their corresponding weight.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, in the constraint:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS1([3.1, 1.2, 2.3]))\n[x[1], x[2], x[3]] in MathOptInterface.SOS1{Float64}([3.1, 1.2, 2.3])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"the variables x have precedence x[2], x[3], x[1].","category":"page"},{"location":"manual/constraints/#Special-Ordered-Sets-of-Type-2","page":"Constraints","title":"Special Ordered Sets of Type 2","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In a Special Ordered Set of Type 2 (SOS-II), at most two elements can be non-zero, and if there are two non-zeros, they must be consecutive according to the ordering induced by a weight vector.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Construct SOS-II constraints using the SOS2 set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS2([3.0, 1.0, 2.0]))\n[x[1], x[2], x[3]] in MathOptInterface.SOS2{Float64}([3.0, 1.0, 2.0])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The possible non-zero pairs are (x[1], x[3]) and (x[2], x[3]):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the weight vector is omitted, JuMP induces an ordering from 1:length(x):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS2())\n[x[1], x[2], x[3]] in MathOptInterface.SOS2{Float64}([1.0, 2.0, 3.0])","category":"page"},{"location":"manual/constraints/#Indicator-constraints","page":"Constraints","title":"Indicator constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Indicator constraints consist of a binary variable and a linear constraint. The constraint holds when the binary variable takes the value 1. The constraint may or may not hold when the binary variable takes the value 0.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To enforce the constraint x + y <= 1 when the binary variable a is 1, use:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @variable(model, a, Bin)\na\n\njulia> @constraint(model, a => {x + y <= 1})\na => {x + y ≤ 1}","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the constraint must hold when a is zero, add ! or ¬ before the binary variable;","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, !a => {x + y <= 1})\n!a => {x + y ≤ 1}","category":"page"},{"location":"manual/constraints/#Semidefinite-constraints","page":"Constraints","title":"Semidefinite constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To constrain a matrix to be positive semidefinite (PSD), use PSDCone:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[2,1]  X[2,2]\n\njulia> @constraint(model, X >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[2,1]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nWhere possible, prefer constructing a matrix of Semidefinite variables using the @variable macro, rather than adding a constraint like @constraint(model, X >= 0, PSDCone()). In some solvers, adding the constraint via @constraint is less efficient, and can result in additional intermediate variables and constraints being added to the model.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The inequality X >= Y between two square matrices X and Y is understood as constraining X - Y to be positive semidefinite.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> Y = [1 2; 2 1]\n2×2 Matrix{Int64}:\n 1  2\n 2  1\n\njulia> @constraint(model, X >= Y, PSDCone())\n[X[1,1] - 1  X[1,2] - 2;\n X[2,1] - 2  X[2,2] - 1] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/#Symmetry","page":"Constraints","title":"Symmetry","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Solvers supporting PSD constraints usually expect to be given a matrix that is symbolically symmetric, that is, for which the expression in corresponding off-diagonal entries are the same. In our example, the expressions of entries (1, 2) and (2, 1) are respectively X[1,2] - 2 and X[2,1] - 2 which are different.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To bridge the gap between the constraint modeled and what the solver expects, solvers may add an equality constraint X[1,2] - 2 == X[2,1] - 2 to force symmetry. Use LinearAlgebra.Symmetric to explicitly tell the solver that the matrix is symmetric:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> import LinearAlgebra\n\njulia> Z = [X[1, 1] X[1, 2]; X[1, 2] X[2, 2]]\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[1,2]  X[2,2]\n\njulia> @constraint(model, LinearAlgebra.Symmetric(Z) >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[1,2]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Note that the lower triangular entries are ignored even if they are different so use it with caution:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, LinearAlgebra.Symmetric(X) >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[1,2]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"(Note the (2, 1) element of the constraint is X[1,2], not X[2,1].)","category":"page"},{"location":"manual/constraints/#Complementarity-constraints","page":"Constraints","title":"Complementarity constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A mixed complementarity constraint F(x) ⟂ x consists of finding x in the interval [lb, ub], such that the following holds:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"F(x) == 0 if lb < x < ub\nF(x) >= 0 if lb == x\nF(x) <= 0 if x == ub","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP supports mixed complementarity constraints via complements(F(x), x) or F(x) ⟂ x in the @constraint macro. The interval set [lb, ub] is obtained from the variable bounds on x.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, to define the problem 2x - 1 ⟂ x with x ∈ [0, ∞), do:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, 2x - 1 ⟂ x)\n[2 x - 1, x] ∈ MathOptInterface.Complements(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This problem has a unique solution at x = 0.5.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The perp operator ⟂ can be entered in most editors (and the Julia REPL) by typing \\perp<tab>.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"An alternative approach that does not require the ⟂ symbol uses the complements function as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, complements(2x - 1, x))\n[2 x - 1, x] ∈ MathOptInterface.Complements(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In both cases, the mapping F(x) is supplied as the first argument, and the matching variable x is supplied as the second.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Vector-valued complementarity constraints are also supported:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @variable(model, -2 <= y[1:2] <= 2)\n2-element Vector{VariableRef}:\n y[1]\n y[2]\n\njulia> M = [1 2; 3 4]\n2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> q = [5, 6]\n2-element Vector{Int64}:\n 5\n 6\n\njulia> @constraint(model, M * y + q ⟂ y)\n[y[1] + 2 y[2] + 5, 3 y[1] + 4 y[2] + 6, y[1], y[2]] ∈ MathOptInterface.Complements(4)","category":"page"},{"location":"manual/constraints/#Boolean-constraints","page":"Constraints","title":"Boolean constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Add a Boolean constraint (a MOI.EqualTo{Bool} set) using the := operator with a Bool right-hand side term:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = GenericModel{Bool}();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x[1] || x[2] := true)\nx[1] || x[2] = true\n\njulia> @constraint(model, x[1] && x[2] := false)\nx[1] && x[2] = false\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariables: 2\n`GenericNonlinearExpr{GenericVariableRef{Bool}}`-in-`MathOptInterface.EqualTo{Bool}`: 2 constraints\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Boolean constraints should not be added using the == operator because JuMP will rewrite the constraint as lhs - rhs = 0, and because constraints like a == b == c require parentheses to disambiguate between (a == b) == c and a == (b == c). In contrast, a == b := c is equivalent to (a == b) := c:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> rhs = false\nfalse\n\njulia> @constraint(model, (x[1] == x[2]) == rhs)\n(x[1] == x[2]) - 0.0 = 0\n\njulia> @constraint(model, x[1] == x[2] := rhs)\nx[1] == x[2] = false","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"EditURL = \"power_systems.jl\"","category":"page"},{"location":"tutorials/applications/power_systems/#Power-Systems","page":"Power Systems","title":"Power Systems","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial was originally contributed by Yury Dvorkin and Miles Lubin.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial demonstrates how to formulate basic power systems engineering models in JuMP.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"We will consider basic \"economic dispatch\" and \"unit commitment\" models without taking into account transmission constraints.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For this tutorial, we use the following packages:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport Plots\nimport StatsPlots","category":"page"},{"location":"tutorials/applications/power_systems/#Economic-dispatch","page":"Power Systems","title":"Economic dispatch","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Economic dispatch (ED) is an optimization problem that minimizes the cost of supplying energy demand subject to operational constraints on power system assets. In its simplest modification, ED is an LP problem solved for an aggregated load and wind forecast and for a single infinitesimal moment.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Mathematically, the ED problem can be written as follows:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"min sum_i in I c^g_i cdot g_i + c^w cdot w","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"where c_i and g_i are the incremental cost ($/MWh) and power output (MW) of the i^th generator, respectively, and c^w and w are the incremental cost ($/MWh) and wind power injection (MW), respectively.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Subject to the constraints:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Minimum (g^min) and maximum (g^max) limits on power outputs of generators: g^min_i leq g_i leq g^max_i\nConstraint on the wind power injection: 0 leq w leq w^f where w and w^f are the wind power injection and wind power forecast, respectively.\nPower balance constraint: sum_i in I g_i + w = d^f where d^f is the demand forecast.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Further reading on ED models can be found in A. J. Wood, B. F. Wollenberg, and G. B. Sheblé, \"Power Generation, Operation and Control,\" Wiley, 2013.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Define some input data about the test system.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"We define some thermal generators:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function ThermalGenerator(\n    min::Float64,\n    max::Float64,\n    fixed_cost::Float64,\n    variable_cost::Float64,\n)\n    return (\n        min = min,\n        max = max,\n        fixed_cost = fixed_cost,\n        variable_cost = variable_cost,\n    )\nend\n\ngenerators = [\n    ThermalGenerator(0.0, 1000.0, 1000.0, 50.0),\n    ThermalGenerator(300.0, 1000.0, 0.0, 100.0),\n]","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"A wind generator","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"WindGenerator(variable_cost::Float64) = (variable_cost = variable_cost,)\n\nwind_generator = WindGenerator(50.0)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"And a scenario","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function Scenario(demand::Float64, wind::Float64)\n    return (demand = demand, wind = wind)\nend\n\nscenario = Scenario(1500.0, 200.0)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Create a function solve_economic_dispatch, which solves the economic dispatch problem for a given set of input parameters.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_economic_dispatch(generators::Vector, wind, scenario)\n    # Define the economic dispatch (ED) model\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    # Define decision variables\n    # power output of generators\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    # wind power injection\n    @variable(model, 0 <= w <= scenario.wind)\n    # Define the objective function\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    # Define the power balance constraint\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    # Solve statement\n    optimize!(model)\n    # return the optimal value of the objective function and its minimizers\n    return (\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        total_cost = objective_value(model),\n    )\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Solve the economic dispatch problem","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"solution = solve_economic_dispatch(generators, wind_generator, scenario);\n\nprintln(\"Dispatch of Generators: \", solution.g, \" MW\")\nprintln(\"Dispatch of Wind: \", solution.w, \" MW\")\nprintln(\"Wind spillage: \", solution.wind_spill, \" MW\")\nprintln(\"Total cost: \\$\", solution.total_cost)","category":"page"},{"location":"tutorials/applications/power_systems/#Economic-dispatch-with-adjustable-incremental-costs","page":"Power Systems","title":"Economic dispatch with adjustable incremental costs","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following exercise we adjust the incremental cost of generator G1 and observe its impact on the total cost.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function scale_generator_cost(g, scale)\n    return ThermalGenerator(g.min, g.max, g.fixed_cost, scale * g.variable_cost)\nend\n\nstart = time()\nc_g_scale_df = DataFrames.DataFrame(;\n    # Scale factor\n    scale = Float64[],\n    # Dispatch of Generator 1 [MW]\n    dispatch_G1 = Float64[],\n    # Dispatch of Generator 2 [MW]\n    dispatch_G2 = Float64[],\n    # Dispatch of Wind [MW]\n    dispatch_wind = Float64[],\n    # Spillage of Wind [MW]\n    spillage_wind = Float64[],\n    # Total cost [$]\n    total_cost = Float64[],\n)\nfor c_g1_scale in 0.5:0.1:3.0\n    # Update the incremental cost of the first generator at every iteration.\n    new_generators = scale_generator_cost.(generators, [c_g1_scale, 1.0])\n    # Solve the economic-dispatch problem with the updated incremental cost\n    sol = solve_economic_dispatch(new_generators, wind_generator, scenario)\n    push!(\n        c_g_scale_df,\n        (c_g1_scale, sol.g[1], sol.g[2], sol.w, sol.wind_spill, sol.total_cost),\n    )\nend\nprint(string(\"elapsed time: \", time() - start, \" seconds\"))","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"c_g_scale_df","category":"page"},{"location":"tutorials/applications/power_systems/#Modifying-the-JuMP-model-in-place","page":"Power Systems","title":"Modifying the JuMP model in-place","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Note that in the previous exercise we entirely rebuilt the optimization model at every iteration of the internal loop, which incurs an additional computational burden. This burden can be alleviated if instead of re-building the entire model, we modify the constraints or objective function, as it shown in the example below.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Compare the computing time in case of the above and below models.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_economic_dispatch_inplace(\n    generators::Vector,\n    wind,\n    scenario,\n    scale::AbstractVector{Float64},\n)\n    obj_out = Float64[]\n    w_out = Float64[]\n    g1_out = Float64[]\n    g2_out = Float64[]\n    # This function only works for two generators\n    @assert length(generators) == 2\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    for c_g1_scale in scale\n        @objective(\n            model,\n            Min,\n            c_g1_scale * generators[1].variable_cost * g[1] +\n            generators[2].variable_cost * g[2] +\n            wind.variable_cost * w,\n        )\n        optimize!(model)\n        push!(obj_out, objective_value(model))\n        push!(w_out, value(w))\n        push!(g1_out, value(g[1]))\n        push!(g2_out, value(g[2]))\n    end\n    df = DataFrames.DataFrame(;\n        scale = scale,\n        dispatch_G1 = g1_out,\n        dispatch_G2 = g2_out,\n        dispatch_wind = w_out,\n        spillage_wind = scenario.wind .- w_out,\n        total_cost = obj_out,\n    )\n    return df\nend\n\nstart = time()\ninplace_df = solve_economic_dispatch_inplace(\n    generators,\n    wind_generator,\n    scenario,\n    0.5:0.1:3.0,\n)\nprint(string(\"elapsed time: \", time() - start, \" seconds\"))","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For small models, adjusting specific constraints or the objective function is sometimes faster and sometimes slower than re-building the entire model. However, as the problem size increases, updating the model in-place is usually faster.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"inplace_df","category":"page"},{"location":"tutorials/applications/power_systems/#Inefficient-usage-of-wind-generators","page":"Power Systems","title":"Inefficient usage of wind generators","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The economic dispatch problem does not perform commitment decisions and, thus, assumes that all generators must be dispatched at least at their minimum power output limit. This approach is not cost efficient and may lead to absurd decisions. For example, if d = sum_i in I g^min_i, the wind power injection must be zero, that is, all available wind generation is spilled, to meet the minimum power output constraints on generators.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following example, we adjust the total demand and observed how it affects wind spillage.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"demand_scale_df = DataFrames.DataFrame(;\n    demand = Float64[],\n    dispatch_G1 = Float64[],\n    dispatch_G2 = Float64[],\n    dispatch_wind = Float64[],\n    spillage_wind = Float64[],\n    total_cost = Float64[],\n)\n\nfunction scale_demand(scenario, scale)\n    return Scenario(scale * scenario.demand, scenario.wind)\nend\n\nfor demand_scale in 0.2:0.1:1.4\n    new_scenario = scale_demand(scenario, demand_scale)\n    sol = solve_economic_dispatch(generators, wind_generator, new_scenario)\n    push!(\n        demand_scale_df,\n        (\n            new_scenario.demand,\n            sol.g[1],\n            sol.g[2],\n            sol.w,\n            sol.wind_spill,\n            sol.total_cost,\n        ),\n    )\nend\n\ndemand_scale_df","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"dispatch_plot = StatsPlots.@df(\n    demand_scale_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_G1, :dispatch_G2],\n        labels = [\"G1\" \"G2\"],\n        title = \"Thermal Dispatch\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand\",\n        ylabel = \"Dispatch [MW]\",\n    ),\n)\n\nwind_plot = StatsPlots.@df(\n    demand_scale_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_wind, :spillage_wind],\n        labels = [\"Dispatch\" \"Spillage\"],\n        title = \"Wind\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand [MW]\",\n        ylabel = \"Energy [MW]\",\n    ),\n)\n\nPlots.plot(dispatch_plot, wind_plot)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This particular drawback can be overcome by introducing binary decisions on the \"on/off\" status of generators. This model is called unit commitment and considered later in these notes.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For further reading on the interplay between wind generation and the minimum power output constraints of generators, we refer interested readers to R. Baldick, \"Wind and energy markets: a case study of Texas,\" IEEE Systems Journal, vol. 6, pp. 27-34, 2012.","category":"page"},{"location":"tutorials/applications/power_systems/#Unit-commitment","page":"Power Systems","title":"Unit commitment","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The Unit Commitment (UC) model can be obtained from ED model by introducing binary variable associated with each generator. This binary variable can attain two values: if it is \"1,\" the generator is synchronized and, thus, can be dispatched, otherwise, that is, if the binary variable is \"0,\" that generator is not synchronized and its power output is set to 0.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"To obtain the mathematical formulation of the UC model, we will modify the constraints of the ED model as follows:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"g^min_i cdot u_ti leq g_i leq g^max_i cdot u_ti","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"where u_i in 01 In this constraint, if u_i = 0, then g_i  = 0. On the other hand, if u_i = 1, then g^min_i leq g_i leq g^max_i.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For further reading on the UC problem we refer interested readers to G. Morales-Espana, J. M. Latorre, and A. Ramos, \"Tight and Compact MILP Formulation for the Thermal Unit Commitment Problem,\" IEEE Transactions on Power Systems, vol. 28, pp. 4897-4908, 2013.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following example we convert the ED model explained above to the UC model.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_unit_commitment(generators::Vector, wind, scenario)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    N = length(generators)\n    @variable(model, 0 <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    # !!! New: add binary on-off variables for each generator\n    @variable(model, u[i = 1:N], Bin)\n    @constraint(model, [i = 1:N], g[i] <= generators[i].max * u[i])\n    @constraint(model, [i = 1:N], g[i] >= generators[i].min * u[i])\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w +\n        # !!! new\n        sum(generators[i].fixed_cost * u[i] for i in 1:N)\n    )\n    optimize!(model)\n    status = termination_status(model)\n    if status != OPTIMAL\n        return (status = status,)\n    end\n    return (\n        status = status,\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        u = value.(u),\n        total_cost = objective_value(model),\n    )\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Solve the unit commitment problem","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"solution = solve_unit_commitment(generators, wind_generator, scenario)\n\nprintln(\"Dispatch of Generators: \", solution.g, \" MW\")\nprintln(\"Commitments of Generators: \", solution.u)\nprintln(\"Dispatch of Wind: \", solution.w, \" MW\")\nprintln(\"Wind spillage: \", solution.wind_spill, \" MW\")\nprintln(\"Total cost: \\$\", solution.total_cost)","category":"page"},{"location":"tutorials/applications/power_systems/#Unit-commitment-as-a-function-of-demand","page":"Power Systems","title":"Unit commitment as a function of demand","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"After implementing the unit commitment model, we can now assess the interplay between the minimum power output constraints on generators and wind generation.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"uc_df = DataFrames.DataFrame(;\n    demand = Float64[],\n    commitment_G1 = Float64[],\n    commitment_G2 = Float64[],\n    dispatch_G1 = Float64[],\n    dispatch_G2 = Float64[],\n    dispatch_wind = Float64[],\n    spillage_wind = Float64[],\n    total_cost = Float64[],\n)\n\nfor demand_scale in 0.2:0.1:1.4\n    new_scenario = scale_demand(scenario, demand_scale)\n    sol = solve_unit_commitment(generators, wind_generator, new_scenario)\n    if sol.status == OPTIMAL\n        push!(\n            uc_df,\n            (\n                new_scenario.demand,\n                sol.u[1],\n                sol.u[2],\n                sol.g[1],\n                sol.g[2],\n                sol.w,\n                sol.wind_spill,\n                sol.total_cost,\n            ),\n        )\n    end\n    println(\"Status: $(sol.status) for demand_scale = $(demand_scale)\")\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"uc_df","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"commitment_plot = StatsPlots.@df(\n    uc_df,\n    Plots.plot(\n        :demand,\n        [:commitment_G1, :commitment_G2],\n        labels = [\"G1\" \"G2\"],\n        title = \"Commitment\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand [MW]\",\n        ylabel = \"Commitment decision {0, 1}\",\n    ),\n)\n\ndispatch_plot = StatsPlots.@df(\n    uc_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_G1, :dispatch_G2, :dispatch_wind],\n        labels = [\"G1\" \"G2\" \"Wind\"],\n        title = \"Dispatch [MW]\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand\",\n        ylabel = \"Dispatch [MW]\",\n    ),\n)\n\nPlots.plot(commitment_plot, dispatch_plot)","category":"page"},{"location":"tutorials/applications/power_systems/#Nonlinear-economic-dispatch","page":"Power Systems","title":"Nonlinear economic dispatch","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"As a final example, we modify our economic dispatch problem in two ways:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The thermal cost function is user-defined\nThe output of the wind is only the square-root of the dispatch","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"import Ipopt\n\n\"\"\"\n    thermal_cost_function(g)\n\nA user-defined thermal cost function in pure-Julia! You can include\nnonlinearities, and even things like control flow.\n\n!!! warning\n    It's still up to you to make sure that the function has a meaningful\n    derivative.\n\"\"\"\nfunction thermal_cost_function(g)\n    if g <= 500\n        return g\n    else\n        return g + 1e-2 * (g - 500)^2\n    end\nend\n\nfunction solve_nonlinear_economic_dispatch(\n    generators::Vector,\n    wind,\n    scenario;\n    silent::Bool = false,\n)\n    model = Model(Ipopt.Optimizer)\n    if silent\n        set_silent(model)\n    end\n    @operator(model, op_tcf, 1, thermal_cost_function)\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * op_tcf(g[i]) for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    @constraint(model, sum(g[i] for i in 1:N) + sqrt(w) == scenario.demand)\n    optimize!(model)\n    return (\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        total_cost = objective_value(model),\n    )\nend\n\nsolution =\n    solve_nonlinear_economic_dispatch(generators, wind_generator, scenario)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Now let's see how the wind is dispatched as a function of the cost:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"wind_cost = 0.0:1:100\nwind_dispatch = Float64[]\nfor c in wind_cost\n    sol = solve_nonlinear_economic_dispatch(\n        generators,\n        WindGenerator(c),\n        scenario;\n        silent = true,\n    )\n    push!(wind_dispatch, sol.w)\nend\n\nPlots.plot(\n    wind_cost,\n    wind_dispatch;\n    xlabel = \"Cost\",\n    ylabel = \"Dispatch [MW]\",\n    label = false,\n)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"EditURL = \"logistic_regression.jl\"","category":"page"},{"location":"tutorials/conic/logistic_regression/#Logistic-regression","page":"Logistic regression","title":"Logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial was originally contributed by François Pacaud.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial shows how to solve a logistic regression problem with JuMP. Logistic regression is a well known method in machine learning, useful when we want to classify binary variables with the help of a given set of features. To this goal, we find the optimal combination of features maximizing the (log)-likelihood onto a training set. From a modern optimization glance, the resulting problem is convex and differentiable. On a modern optimization glance, it is even conic representable.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Formulating-the-logistic-regression-problem","page":"Logistic regression","title":"Formulating the logistic regression problem","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Suppose we have a set of training data-point i = 1 cdots n, where for each i we have a vector of features x_i in mathbbR^p and a categorical observation y_i in -1 1.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"The log-likelihood is given by","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"l(theta) = sum_i=1^n log(dfrac11 + exp(-y_i theta^top x_i))","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"and the optimal theta minimizes the logistic loss function:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"min_theta sum_i=1^n log(1 + exp(-y_i theta^top x_i))","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Most of the time, instead of solving directly the previous optimization problem, we prefer to add a regularization term:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"min_theta sum_i=1^n log(1 + exp(-y_i theta^top x_i)) + lambda  theta ","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"with lambda in mathbbR_+ a penalty and  a norm function. By adding such a regularization term, we avoid overfitting on the training set and usually achieve a greater score in cross-validation.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Reformulation-as-a-conic-optimization-problem","page":"Logistic regression","title":"Reformulation as a conic optimization problem","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"By introducing auxiliary variables t_1 cdots t_n and r, the optimization problem is equivalent to","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_t r theta  sum_i=1^n t_i + lambda r \ntextsubject to   quad t_i geq log(1 + exp(- y_i theta^top x_i)) \n                    quad r geq theta\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Now, the trick is to reformulate the constraints t_i geq log(1 + exp(- y_i theta^top x_i)) with the help of the exponential cone","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"K_exp =  (x y z) in mathbbR^3   y exp(x  y) leq z  ","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Indeed, by passing to the exponential, we see that for all i=1 cdots n, the constraint t_i geq log(1 + exp(- y_i theta^top x_i)) is equivalent to","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"exp(-t_i) + exp(u_i - t_i) leq 1","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"with u_i = -y_i theta^top x_i. Then, by adding two auxiliary variables z_i1 and z_i2 such that z_i1 geq exp(u_i-t_i) and z_i2 geq exp(-t_i), we get the equivalent formulation","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"left\nbeginaligned\n(u_i -t_i  1 z_i1)  in  K_exp  \n(-t_i  1 z_i2)  in  K_exp  \nz_i1 + z_i2  leq  1\nendaligned\nright","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"In this setting, the conic version of the logistic regression problems writes out","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_t z r theta   sum_i=1^n t_i + lambda r \ntextsubject to   quad  (u_i -t_i  1 z_i1)  in  K_exp  \n                    quad  (-t_i  1 z_i2)  in  K_exp  \n                    quad  z_i1 + z_i2  leq  1 \n                    quad u_i = -y_i x_i^top theta \n                    quad r geq theta\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"and thus encompasses 3n + p + 1 variables and 3n + 1 constraints (u_i = -y_i theta^top x_i is only a virtual constraint used to clarify the notation). Thus, if n gg 1, we get a large number of variables and constraints.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Fitting-logistic-regression-with-a-conic-solver","page":"Logistic regression","title":"Fitting logistic regression with a conic solver","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"It is now time to pass to the implementation. We choose SCS as a conic solver.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"using JuMP\nimport Random\nimport SCS","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Random.seed!(2713);\nnothing #hide","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We start by implementing a function to generate a fake dataset, and where we could tune the correlation between the feature variables. The function is a direct transcription of the one used in this blog post.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function generate_dataset(n_samples = 100, n_features = 10; shift = 0.0)\n    X = randn(n_samples, n_features)\n    w = randn(n_features)\n    y = sign.(X * w)\n    X .+= 0.8 * randn(n_samples, n_features) # add noise\n    X .+= shift # shift the points in the feature space\n    X = hcat(X, ones(n_samples, 1))\n    return X, y\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We write a softplus function to formulate each constraint t geq log(1 + exp(u)) with two exponential cones.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function softplus(model, t, u)\n    z = @variable(model, [1:2], lower_bound = 0.0)\n    @constraint(model, sum(z) <= 1.0)\n    @constraint(model, [u - t, 1, z[1]] in MOI.ExponentialCone())\n    @constraint(model, [-t, 1, z[2]] in MOI.ExponentialCone())\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/#\\ell_2-regularized-logistic-regression","page":"Logistic regression","title":"ell_2 regularized logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Then, with the help of the softplus function, we could write our optimization model. In the ell_2 regularization case, the constraint r geq theta_2 rewrites as a second order cone constraint.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function build_logit_model(X, y, λ)\n    n, p = size(X)\n    model = Model()\n    @variable(model, θ[1:p])\n    @variable(model, t[1:n])\n    for i in 1:n\n        u = -(X[i, :]' * θ) * y[i]\n        softplus(model, t[i], u)\n    end\n    # Add ℓ2 regularization\n    @variable(model, 0.0 <= reg)\n    @constraint(model, [reg; θ] in SecondOrderCone())\n    # Define objective\n    @objective(model, Min, sum(t) + λ * reg)\n    return model\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We generate the dataset.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"warning: Warning\nBe careful here, for large n and p SCS could fail to converge.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"n, p = 200, 10\nX, y = generate_dataset(n, p; shift = 10.0);\n\n# We could now solve the logistic regression problem\nλ = 10.0\nmodel = build_logit_model(X, y, λ)\nset_optimizer(model, SCS.Optimizer)\nset_silent(model)\nJuMP.optimize!(model)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"θ♯ = JuMP.value.(model[:θ])","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"It appears that the speed of convergence is not that impacted by the correlation of the dataset, nor by the penalty lambda.","category":"page"},{"location":"tutorials/conic/logistic_regression/#\\ell_1-regularized-logistic-regression","page":"Logistic regression","title":"ell_1 regularized logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We now formulate the logistic problem with a ell_1 regularization term. The ell_1 regularization ensures sparsity in the optimal solution of the resulting optimization problem. Luckily, the ell_1 norm is implemented as a set in MathOptInterface. Thus, we could formulate the sparse logistic regression problem with the help of a MOI.NormOneCone set.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function build_sparse_logit_model(X, y, λ)\n    n, p = size(X)\n    model = Model()\n    @variable(model, θ[1:p])\n    @variable(model, t[1:n])\n    for i in 1:n\n        u = -(X[i, :]' * θ) * y[i]\n        softplus(model, t[i], u)\n    end\n    # Add ℓ1 regularization\n    @variable(model, 0.0 <= reg)\n    @constraint(model, [reg; θ] in MOI.NormOneCone(p + 1))\n    # Define objective\n    @objective(model, Min, sum(t) + λ * reg)\n    return model\nend\n\n# Auxiliary function to count non-null components:\ncount_nonzero(v::Vector; tol = 1e-6) = sum(abs.(v) .>= tol)\n\n# We solve the sparse logistic regression problem on the same dataset as before.\nλ = 10.0\nsparse_model = build_sparse_logit_model(X, y, λ)\nset_optimizer(sparse_model, SCS.Optimizer)\nset_silent(sparse_model)\nJuMP.optimize!(sparse_model)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"θ♯ = JuMP.value.(sparse_model[:θ])\nprintln(\n    \"Number of non-zero components: \",\n    count_nonzero(θ♯),\n    \" (out of \",\n    p,\n    \" features)\",\n)","category":"page"},{"location":"tutorials/conic/logistic_regression/#Extensions","page":"Logistic regression","title":"Extensions","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"A direct extension would be to consider the sparse logistic regression with hard thresholding, which, on contrary to the soft version using a ell_1 regularization, adds an explicit cardinality constraint in its formulation:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_theta   sum_i=1^n log(1 + exp(-y_i theta^top x_i)) + lambda  theta _2^2 \ntextsubject to   quad  theta _0 = k\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"where k is the maximum number of non-zero components in the vector theta, and _0 is the ell_0 pseudo-norm:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":" x_0 = i   x_i neq 0","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"The cardinality constraint theta_0 leq k could be reformulated with binary variables. Thus the hard sparse regression problem could be solved by any solver supporting mixed integer conic problems.","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/motivation.md\"","category":"page"},{"location":"moi/background/motivation/#Motivation","page":"Motivation","title":"Motivation","text":"","category":"section"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"MathOptInterface (MOI) is a replacement for MathProgBase, the first-generation abstraction layer for mathematical optimization previously used by JuMP and Convex.jl.","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"To address a number of limitations of MathProgBase, MOI is designed to:","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"Be simple and extensible\nunifying linear, quadratic, and conic optimization,\nseamlessly facilitating extensions to essentially arbitrary constraints and functions (for example, indicator constraints, complementarity constraints, and piecewise-linear functions)\nBe fast\nby allowing access to a solver's in-memory representation of a problem without writing intermediate files (when possible)\nby using multiple dispatch and avoiding requiring containers of non-concrete types\nAllow a solver to return multiple results (for example, a pool of solutions)\nAllow a solver to return extra arbitrary information via attributes (for example, variable- and constraint-wise membership in an irreducible inconsistent subset for infeasibility analysis)\nProvide a greatly expanded set of status codes explaining what happened during the optimization procedure\nEnable a solver to more precisely specify which problem classes it supports\nEnable both primal and dual warm starts\nEnable adding and removing both variables and constraints by indices that are not required to be consecutive\nEnable any modification that the solver supports to an existing model\nAvoid requiring the solver wrapper to store an additional copy of the problem data","category":"page"},{"location":"tutorials/conic/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Conic programs are a class of convex nonlinear optimization problems which use cones to represent the nonlinearities. They have the form:","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n     min_x in mathbbR^n  f_0(x) \n     textst  f_j(x) in mathcalS_j   j = 1 ldots m\nendalign","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer conic programs (MICPs) are extensions of conic programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/conic/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"JuMP supports a range of conic solvers, although support differs on what types of cones each solver supports. In the list of Supported solvers, \"SOCP\" denotes solvers supporting second-order cones and \"SDP\" denotes solvers supporting semidefinite cones. In addition, solvers such as SCS and Mosek have support for the exponential cone. Moreover, due to the bridging system in MathOptInterface, many of these solvers support a much wider range of exotic cones than they natively support. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"tip: Tip\nDuality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual instead of the primal. If performance is an issue, see the Dualization tutorial for more details.","category":"page"},{"location":"tutorials/conic/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nExperiment design\nLogistic regression\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling conic programs. Look here if you are stuck trying to formulate a problem as a conic program.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/manipulating_expressions.md\"","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Manipulating-expressions","page":"Manipulating expressions","title":"Manipulating expressions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This guide highlights a syntactically appealing way to build expressions at the MOI level, but also to look at their contents. It may be especially useful when writing models or bridge code.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-functions","page":"Manipulating expressions","title":"Creating functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This section details the ways to create functions with MathOptInterface.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-scalar-affine-functions","page":"Manipulating expressions","title":"Creating scalar affine functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"The simplest scalar function is simply a variable:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> x = MOI.add_variable(model) # Create the variable x\nMOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This type of function is extremely simple; to express more complex functions, other types must be used. For instance, a ScalarAffineFunction is a sum of linear terms (a factor times a variable) and a constant. Such an object can be built using the standard constructor:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1, x)], 2) # x + 2\n(2) + (1) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"However, you can also use operators to build the same scalar function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> f = x + 2\n(2) + (1) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-scalar-quadratic-functions","page":"Manipulating expressions","title":"Creating scalar quadratic functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Scalar quadratic functions are stored in ScalarQuadraticFunction objects, in a way that is highly similar to scalar affine functions. You can obtain a quadratic function as a product of affine functions:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> 1 * x * x\n(0) + 1.0 MOI.VariableIndex(1)²\n\njulia> f * f  # (x + 2)²\n(4) + (2) MOI.VariableIndex(1) + (2) MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²\n\njulia> f^2  # (x + 2)² too\n(4) + (2) MOI.VariableIndex(1) + (2) MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-vector-functions","page":"Manipulating expressions","title":"Creating vector functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"A vector function is a function with several values, irrespective of the number of input variables. Similarly to scalar functions, there are three main types of vector functions: VectorOfVariables, VectorAffineFunction, and VectorQuadraticFunction.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"The easiest way to create a vector function is to stack several scalar functions using Utilities.vectorize. It takes a vector as input, and the generated vector function (of the most appropriate type) has each dimension corresponding to a dimension of the vector.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> g = MOI.Utilities.vectorize([f, 2 * f])\n┌                              ┐\n│(2) + (1) MOI.VariableIndex(1)│\n│(4) + (2) MOI.VariableIndex(1)│\n└                              ┘","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"warning: Warning\nUtilities.vectorize only takes a vector of similar scalar functions: you cannot mix VariableIndex and ScalarAffineFunction, for instance. In practice, it means that Utilities.vectorize([x, f]) does not work; you should rather use Utilities.vectorize([1 * x, f]) instead to only have ScalarAffineFunction objects.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Canonicalizing-functions","page":"Manipulating expressions","title":"Canonicalizing functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"In more advanced use cases, you might need to ensure that a function is \"canonical.\" Functions are stored as an array of terms, but there is no check that these terms are redundant: a ScalarAffineFunction object might have two terms with the same variable, like x + x + 1. These terms could be merged without changing the semantics of the function: 2x + 1.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Working with these objects might be cumbersome. Canonicalization helps maintain redundancy to zero.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.is_canonical checks whether a function is already in its canonical form:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.is_canonical(f + f) # (x + 2) + (x + 2) is stored as x + x + 4\nfalse","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.canonical returns the equivalent canonical version of the function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.canonical(f + f) # Returns 2x + 4\n(4) + (2) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Exploring-functions","page":"Manipulating expressions","title":"Exploring functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"At some point, you might need to dig into a function, for instance to map it into solver constructs.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Vector-functions","page":"Manipulating expressions","title":"Vector functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.scalarize returns a vector of scalar functions from a vector function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.scalarize(g) # Returns a vector [f, 2 * f].\n2-element Vector{MathOptInterface.ScalarAffineFunction{Int64}}:\n (2) + (1) MOI.VariableIndex(1)\n (4) + (2) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"note: Note\nUtilities.eachscalar returns an iterator on the dimensions, which serves the same purpose as Utilities.scalarize.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"output_dimension returns the number of dimensions of the output of a function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.output_dimension(g)\n2","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"EditURL = \"https://github.com/jump-dev/MosekTools.jl/blob/v0.15.1/README.md\"","category":"page"},{"location":"packages/MosekTools/#MosekTools.jl","page":"jump-dev/MosekTools.jl","title":"MosekTools.jl","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is the MathOptInterface.jl implementation for the MOSEK solver.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The low-level solver API for MOSEK is found in the package Mosek.jl.","category":"page"},{"location":"packages/MosekTools/#Affiliation","page":"jump-dev/MosekTools.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is maintained by the JuMP community and is not officially supported by MOSEK. However, Mosek.jl is an officially supported product of MOSEK.","category":"page"},{"location":"packages/MosekTools/#License","page":"jump-dev/MosekTools.jl","title":"License","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is licensed under the MIT License.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/MosekTools/#Installation","page":"jump-dev/MosekTools.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The latest release of this package and the master branch are to be used with the latest release of Mosek.jl (which uses MOSEK v10).","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"To use MOSEK v9 (resp. v8), use the v0.12.x (resp. v0.7.x) releases of this package, and the mosekv9 (resp. mosekv8) branch and v1.2.x (resp. v0.9.x) releases of Mosek.jl.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"See the following table for a summary:","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MOSEK Mosek.jl MosekTools.jl release MosekTools.jl branch\nv10 v10 v0.13 master\nv9 v0.12 v0.12 mosekv9\nv8 v0.9 v0.7 mosekv8","category":"page"},{"location":"packages/MosekTools/#Use-with-JuMP","page":"jump-dev/MosekTools.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"using JuMP\nusing MosekTools\nmodel = Model(Mosek.Optimizer)\nset_attribute(model, \"QUIET\", true)\nset_attribute(model, \"INTPNT_CO_TOL_DFEAS\", 1e-7)","category":"page"},{"location":"packages/MosekTools/#Options","page":"jump-dev/MosekTools.jl","title":"Options","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The parameter QUIET is a special parameter that when set to true disables all Mosek printing output.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"All other parameters can be found in the Mosek documentation.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"Note that the prefix MSK_IPAR_ (for integer parameters), MSK_DPAR_ (for floating point parameters) or MSK_SPAR_ (for string parameters) are optional. If they are not given, they are inferred from the type of the value. For example, in the example above, as 1e-7 is a floating point number, the parameters name used is MSK_DPAR_INTPNT_CO_TOL_DFEAS.","category":"page"},{"location":"developers/style/#Style-guide-and-design-principles","page":"Style Guide","title":"Style guide and design principles","text":"","category":"section"},{"location":"developers/style/#Style-guide","page":"Style Guide","title":"Style guide","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section describes the coding style rules that apply to JuMP code and that we recommend for JuMP models and surrounding Julia code. The motivations for a style guide include:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"conveying best practices for writing readable and maintainable code\nreducing the amount of time spent on bike-shedding by establishing basic naming and formatting conventions\nlowering the barrier for new contributors by codifying the existing practices (for example, you can be more confident your code will pass review if you follow the style guide)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In some cases, the JuMP style guide diverges from the Julia style guide. All such cases will be explicitly noted and justified.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The JuMP style guide adopts many recommendations from the Google style guides.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nThe style guide is always a work in progress, and not all JuMP code follows the rules. When modifying JuMP, please fix the style violations of the surrounding code (that is, leave the code tidier than when you started). If large changes are needed, consider separating them into another PR.","category":"page"},{"location":"developers/style/#JuliaFormatter","page":"Style Guide","title":"JuliaFormatter","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"JuMP uses JuliaFormatter.jl as an auto-formatting tool.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"We use the options contained in .JuliaFormatter.toml.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"To format code, cd to the JuMP directory, then run:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"] add JuliaFormatter@1\nusing JuliaFormatter\nformat(\"docs\")\nformat(\"src\")\nformat(\"test\")","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nA continuous integration check verifies that all PRs made to JuMP have passed the formatter.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The following sections outline extra style guide points that are not fixed automatically by JuliaFormatter.","category":"page"},{"location":"developers/style/#Abstract-types-and-composition","page":"Style Guide","title":"Abstract types and composition","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Specifying types for method arguments is mostly optional in Julia. The benefit of abstract method arguments is that it enables functions and types from one package to be used with functions and types from another package via multiple dispatch.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"However, abstractly typed methods have two main drawbacks:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"It's possible to find out that you are working with unexpected types deep in the call chain, potentially leading to hard-to-diagnose MethodErrors.\nUntyped function arguments can lead to correctness problems if the user's choice of input type does not satisfy the assumptions made by the author of the function.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"As a motivating example, consider the following function:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum(x)\n           y = 0.0\n           for i in 1:length(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum (generic function with 1 method)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This function contains a number of implicit assumptions about the type of x:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"x supports 1-based getindex and implements length\nThe element type of x supports addition with 0.0, and then with the result of x + 0.0.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nAs a motivating example for the second point, VariableRef plus Float64 produces an AffExpr. Do not assume that +(::A, ::B) produces an instance of the type A or B.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"my_sum works as expected if the user passes in Vector{Float64}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"but it doesn't respect input types, for example returning a Float64 if the user passes Vector{Int}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum([1, 2, 3])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"but it throws a MethodError if the user passes String:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum(\"abc\")\nERROR: MethodError: no method matching +(::Float64, ::Char)\n[...]","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This particular MethodError is hard to debug, particularly for new users, because it mentions +, Float64, and Char, none of which were called or passed by the user.","category":"page"},{"location":"developers/style/#Dealing-with-MethodErrors","page":"Style Guide","title":"Dealing with MethodErrors","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section diverges from the Julia style guide, as well as other common guides like SciML. The following suggestions are intended to provide a friendlier experience for novice Julia programmers, at the cost of limiting the power and flexibility of advanced Julia programmers.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Code should follow the MethodError principle:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: The MethodError principle\nA user should see a MethodError only for methods that they called directly.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\n# The user sees a MethodError for _internal_function when calling\n# public_function(\"a string\"). This is not very helpful.\npublic_function(x) = _internal_function(x)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\n# The user sees a MethodError for public_function when calling\n# public_function(\"a string\"). This is easy to understand.\npublic_function(x::Integer) = _internal_function(x)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"If it is hard to provide an error message at the top of the call chain, then the following pattern is also ok:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\nfunction _internal_function(x)\n    error(\n        \"Internal error. This probably means that you called \" *\n        \"`public_function()`s with the wrong type.\",\n    )\nend\npublic_function(x) = _internal_function(x)","category":"page"},{"location":"developers/style/#Dealing-with-correctness","page":"Style Guide","title":"Dealing with correctness","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Dealing with correctness is harder, because Julia has no way of formally specifying interfaces that abstract types must implement. Instead, here are two options that you can use when writing and interacting with generic code:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Option 1: use concrete types and let users extend new methods.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In this option, explicitly restrict input arguments to concrete types that are tested and have been validated for correctness. For example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum_option_1(x::Vector{Float64})\n           y = 0.0\n           for i in 1:length(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum_option_1 (generic function with 1 method)\n\njulia> my_sum_option_1([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Using concrete types satisfies the MethodError principle:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_option_1(\"abc\")\nERROR: MethodError: no method matching my_sum_option_1(::String)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"and it allows other types to be supported in future by defining new methods:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum_option_1(x::Array{T,N}) where {T<:Number,N}\n           y = zero(T)\n           for i in eachindex(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum_option_1 (generic function with 2 methods)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Importantly, these methods do not have to be defined in the original package.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nSome usage of abstract types is okay. For example, in my_sum_option_1, we allowed the element type, T, to be a subtype of Number. This is fairly safe, but it still has an implicit assumption that T supports zero(T) and +(::T, ::T).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Option 2: program defensively, and validate all assumptions.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"An alternative is to program defensively, and to rigorously document and validate all assumptions that the code makes. In particular:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"All assumptions on abstract types that aren't guaranteed by the definition of the abstract type (for example, optional methods without a fallback) should be documented.\nIf practical, the assumptions should be checked in code, and informative error messages should be provided to the user if the assumptions are not met. In general, these checks may be expensive, so you should prefer to do this once, at the highest level of the call-chain.\nTests should cover for a range of corner cases and argument types.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"\"\"\"\n    test_my_sum_defensive_assumptions(x::AbstractArray{T}) where {T}\n\nTest the assumptions made by `my_sum_defensive`.\n\"\"\"\nfunction test_my_sum_defensive_assumptions(x::AbstractArray{T}) where {T}\n    try\n        # Some types may not define zero.\n        @assert zero(T) isa T\n        # Check iteration supported\n        @assert iterate(x) isa Union{Nothing,Tuple{T,Int}}\n        # Check that + is defined\n        @assert +(zero(T), zero(T)) isa Any\n    catch err\n        error(\n            \"Unable to call my_sum_defensive(::$(typeof(x))) because \" *\n            \"it failed an internal assumption\",\n        )\n    end\n    return\nend\n\n\"\"\"\n    my_sum_defensive(x::AbstractArray{T}) where {T}\n\nReturn the sum of the elements in the abstract array `x`.\n\n## Assumptions\n\nThis function makes the following assumptions:\n\n * That `zero(T)` is defined\n * That `x` supports the iteration interface\n * That  `+(::T, ::T)` is defined\n\"\"\"\nfunction my_sum_defensive(x::AbstractArray{T}) where {T}\n    test_my_sum_defensive_assumptions(x)\n    y = zero(T)\n    for xi in x\n        y += xi\n    end\n    return y\nend\n\n# output\n\nmy_sum_defensive","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This function works on Vector{Float64}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"as well as Matrix{Rational{Int}}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive([(1//2) + (4//3)im; (6//5) + (7//11)im])\n17//10 + 65//33*im","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"and it throws an error when the assumptions aren't met:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive(['a', 'b', 'c'])\nERROR: Unable to call my_sum_defensive(::Vector{Char}) because it failed an internal assumption\n[...]","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"As an alternative, you may choose not to call test_my_sum_defensive_assumptions within my_sum_defensive, and instead ask users of my_sum_defensive to call it in their tests.","category":"page"},{"location":"developers/style/#Juxtaposed-multiplication","page":"Style Guide","title":"Juxtaposed multiplication","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Only use juxtaposed multiplication when the right-hand side is a symbol.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"2x  # Acceptable if there are space constraints.\n2 * x  # This is preferred if space is not an issue.\n2 * (x + 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"2(x + 1)","category":"page"},{"location":"developers/style/#Empty-vectors","page":"Style Guide","title":"Empty vectors","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For a type T, T[] and Vector{T}() are equivalent ways to create an empty vector with element type T. Prefer T[] because it is more concise.","category":"page"},{"location":"developers/style/#Comments","page":"Style Guide","title":"Comments","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For non-native speakers and for general clarity, comments in code must be proper English sentences with appropriate punctuation.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"# This is a comment demonstrating a good comment.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"# a bad comment","category":"page"},{"location":"developers/style/#JuMP-macro-syntax","page":"Style Guide","title":"JuMP macro syntax","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For consistency, always use parentheses.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable model x >= 0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For consistency, always use constant * variable as opposed to variable * constant. This makes it easier to read models in ambiguous cases like a * x.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"a = 4\n@constraint(model, 3 * x <= 1)\n@constraint(model, a * x <= 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"a = 4\n@constraint(model, x * 3 <= 1)\n@constraint(model, x * a <= 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In order to reduce boilerplate code, prefer the plural form of macros over lots of repeated calls to singular forms.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variables(model, begin\n    x >= 0\n    y >= 1\n    z <= 2\nend)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0)\n@variable(model, y >= 1)\n@variable(model, z <= 2)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"An exception is made for calls with many keyword arguments, since these need to be enclosed in parentheses in order to parse properly.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Acceptable:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0, start = 0.0, base_name = \"my_x\")\n@variable(model, y >= 1, start = 2.0)\n@variable(model, z <= 2, start = -1.0)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Also acceptable:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variables(model, begin\n    x >= 0, (start = 0.0, base_name = \"my_x\")\n    y >= 1, (start = 2.0)\n    z <= 2, (start = -1.0)\nend)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"While we always use in for for-loops, it is acceptable to use = in the container declarations of JuMP macros.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Okay:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x[i=1:3])","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Also okay:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x[i in 1:3])","category":"page"},{"location":"developers/style/#Naming","page":"Style Guide","title":"Naming","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module SomeModule end\nfunction some_function end\nconst SOME_CONSTANT = ...\nstruct SomeStruct\n  some_field::SomeType\nend\n@enum SomeEnum ENUM_VALUE_A ENUM_VALUE_B\nsome_local_variable = ...\nsome_file.jl # Except for ModuleName.jl.","category":"page"},{"location":"developers/style/#Exported-and-non-exported-names","page":"Style Guide","title":"Exported and non-exported names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Begin private module level functions and constants with an underscore. All other objects in the scope of a module should be exported. (See JuMP.jl for an example of how to do this.)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Names beginning with an underscore should only be used for distinguishing between exported (public) and non-exported (private) objects. Therefore, never begin the name of a local variable with an underscore.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module MyModule\n\nexport public_function, PUBLIC_CONSTANT\n\nfunction _private_function()\n    local_variable = 1\n    return\nend\n\nfunction public_function end\n\nconst _PRIVATE_CONSTANT = 3.14159\nconst PUBLIC_CONSTANT = 1.41421\n\nend","category":"page"},{"location":"developers/style/#Use-of-underscores-within-names","page":"Style Guide","title":"Use of underscores within names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The Julia style guide recommends avoiding underscores \"when readable,\" for example, haskey, isequal, remotecall, and remotecall_fetch. This convention creates the potential for unnecessary bikeshedding and also forces the user to recall the presence/absence of an underscore, for example, \"was that argument named basename or base_name?\". For consistency, always use underscores in variable names and function names to separate words.","category":"page"},{"location":"developers/style/#Use-of-!","page":"Style Guide","title":"Use of !","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Julia has a convention of appending ! to a function name if the function modifies its arguments. We recommend to:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Omit ! when the name itself makes it clear that modification is taking place, for example, add_constraint and set_name. We depart from the Julia style guide because ! does not provide a reader with any additional information in this case, and adherence to this convention is not uniform even in base Julia itself (consider Base.println and Base.finalize).\nUse ! in all other cases. In particular it can be used to distinguish between modifying and non-modifying variants of the same function like scale and scale!.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Note that ! is not a self-documenting feature because it is still ambiguous which arguments are modified when multiple arguments are present. Be sure to document which arguments are modified in the method's docstring.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"See also the Julia style guide recommendations for ordering of function arguments.","category":"page"},{"location":"developers/style/#Abbreviations","page":"Style Guide","title":"Abbreviations","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Abbreviate names to make the code more readable, not to save typing. Don't arbitrarily delete letters from a word to abbreviate it (for example, indx). Use abbreviations consistently within a body of code (for example, do not mix con and constr, idx and indx).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Common abbreviations:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"num for number\ncon for constraint","category":"page"},{"location":"developers/style/#No-one-letter-variable-names","page":"Style Guide","title":"No one-letter variable names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Where possible, avoid one-letter variable names.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use model = Model() instead of m = Model()","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Exceptions are made for indices in loops.","category":"page"},{"location":"developers/style/#@enum-vs.-Symbol","page":"Style Guide","title":"@enum vs. Symbol","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The @enum macro lets you define types with a finite number of values that are explicitly enumerated (like enum in C/C++). Symbols are lightweight strings that are used to represent identifiers in Julia (for example, :x).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@enum provides type safety and can have docstrings attached to explain the possible values. Use @enums when applicable, for example, for reporting statuses. Use strings to provide long-form additional information like error messages.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use of Symbol should typically be reserved for identifiers, for example, for lookup in the JuMP model (model[:my_variable]).","category":"page"},{"location":"developers/style/#using-vs.-import","page":"Style Guide","title":"using vs. import","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"using ModuleName brings all symbols exported by the module ModuleName into scope, while import ModuleName brings only the module itself into scope. (See the Julia manual) for examples and more details.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For the same reason that from <module> import * is not recommended in python (PEP 8), avoid using ModuleName except in throw-away scripts or at the REPL. The using statement makes it harder to track where symbols come from and exposes the code to ambiguities when two modules export the same symbol.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Prefer using ModuleName: x, p to import ModuleName.x, ModuleName.p and import MyModule: x, p because the import versions allow method extension without qualifying with the module name.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Similarly, using ModuleName: ModuleName is an acceptable substitute for import ModuleName, because it does not bring all symbols exported by ModuleName into scope. However, we prefer import ModuleName for consistency.","category":"page"},{"location":"developers/style/#Documentation","page":"Style Guide","title":"Documentation","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section describes the writing style that should be used when writing documentation for JuMP (and supporting packages).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"We can recommend the documentation style guides by Divio, Google, and Write the Docs as general reading for those writing documentation. This guide delegates a thorough handling of the topic to those guides and instead elaborates on the points more specific to Julia and documentation that use Documenter.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Be concise\nUse lists instead of long sentences\nUse numbered lists when describing a sequence, for example, (1) do X, (2) then Y\nUse bullet points when the items are not ordered\nExample code should be covered by doctests\nWhen a word is a Julia symbol and not an English word, enclose it with backticks. In addition, if it has a docstring in this doc add a link using @ref. If it is a plural, add the \"s\" after the closing backtick. For example,\n[`VariableRef`](@ref)s\nUse @meta blocks for TODOs and other comments that shouldn't be visible to readers. For example,\n```@meta\n# TODO: Mention also X, Y, and Z.\n```","category":"page"},{"location":"developers/style/#Docstrings","page":"Style Guide","title":"Docstrings","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Every exported object needs a docstring\nAll examples in docstrings should be jldoctests\nAlways use complete English sentences with proper punctuation\nDo not terminate lists with punctuation (for example, as in this doc)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Here is an example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"\"\"\"\n    signature(args; kwargs...)\n\nShort sentence describing the function.\n\nOptional: add a slightly longer paragraph describing the function.\n\n## Notes\n\n - List any notes that the user should be aware of\n\n## Examples\n\n```jldoctest\njulia> 1 + 1\n2\n```\n\"\"\"","category":"page"},{"location":"developers/style/#Testing","page":"Style Guide","title":"Testing","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use a module to encapsulate tests, and structure all tests as functions. This avoids leaking local variables between tests.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Here is a basic skeleton:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module TestPkg\n\nusing Test\n\nfunction runtests()\n    for name in names(@__MODULE__; all = true)\n        if startswith(\"$(name)\", \"test_\")\n            @testset \"$(name)\" begin\n                getfield(@__MODULE__, name)()\n            end\n        end\n    end\n    return\nend\n\n_helper_function() = 2\n\nfunction test_addition()\n    @test 1 + 1 == _helper_function()\n    return\nend\n\nend # module TestPkg\n\nTestPkg.runtests()","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Break the tests into multiple files, with one module per file, so that subsets of the codebase can be tested by calling include with the relevant file.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Test/overview.md\"","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Test/overview/#test_module","page":"Overview","title":"The Test submodule","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"The Test submodule provides tools to help solvers implement unit tests in order to ensure they implement the MathOptInterface API correctly, and to check for solver-correctness.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"We use a centralized repository of tests, so that if we find a bug in one solver, instead of adding a test to that particular repository, we add it here so that all solvers can benefit.","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-test-a-solver","page":"Overview","title":"How to test a solver","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"The skeleton below can be used for the wrapper test file of a solver named FooBar.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"# ============================ /test/MOI_wrapper.jl ============================\nmodule TestFooBar\n\nimport FooBar\nusing Test\n\nimport MathOptInterface as MOI\n\nconst OPTIMIZER = MOI.instantiate(\n    MOI.OptimizerWithAttributes(FooBar.Optimizer, MOI.Silent() => true),\n)\n\nconst BRIDGED = MOI.instantiate(\n    MOI.OptimizerWithAttributes(FooBar.Optimizer, MOI.Silent() => true),\n    with_bridge_type = Float64,\n)\n\n# See the docstring of MOI.Test.Config for other arguments.\nconst CONFIG = MOI.Test.Config(\n    # Modify tolerances as necessary.\n    atol = 1e-6,\n    rtol = 1e-6,\n    # Use MOI.LOCALLY_SOLVED for local solvers.\n    optimal_status = MOI.OPTIMAL,\n    # Pass attributes or MOI functions to `exclude` to skip tests that\n    # rely on this functionality.\n    exclude = Any[MOI.VariableName, MOI.delete],\n)\n\n\"\"\"\n    runtests()\n\nThis function runs all functions in the this Module starting with `test_`.\n\"\"\"\nfunction runtests()\n    for name in names(@__MODULE__; all = true)\n        if startswith(\"$(name)\", \"test_\")\n            @testset \"$(name)\" begin\n                getfield(@__MODULE__, name)()\n            end\n        end\n    end\nend\n\n\"\"\"\n    test_runtests()\n\nThis function runs all the tests in MathOptInterface.Test.\n\nPass arguments to `exclude` to skip tests for functionality that is not\nimplemented or that your solver doesn't support.\n\"\"\"\nfunction test_runtests()\n    MOI.Test.runtests(\n        BRIDGED,\n        CONFIG,\n        exclude = [\n            \"test_attribute_NumberOfThreads\",\n            \"test_quadratic_\",\n        ],\n        # This argument is useful to prevent tests from failing on future\n        # releases of MOI that add new tests. Don't let this number get too far\n        # behind the current MOI release though. You should periodically check\n        # for new tests to fix bugs and implement new features.\n        exclude_tests_after = v\"0.10.5\",\n    )\n    return\nend\n\n\"\"\"\n    test_SolverName()\n\nYou can also write new tests for solver-specific functionality. Write each new\ntest as a function with a name beginning with `test_`.\n\"\"\"\nfunction test_SolverName()\n    @test MOI.get(FooBar.Optimizer(), MOI.SolverName()) == \"FooBar\"\n    return\nend\n\nend # module TestFooBar\n\n# This line at tne end of the file runs all the tests!\nTestFooBar.runtests()","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Then modify your runtests.jl file to include the MOI_wrapper.jl file:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"# ============================ /test/runtests.jl ============================\n\nusing Test\n\n@testset \"MOI\" begin\n    include(\"test/MOI_wrapper.jl\")\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"info: Info\nThe optimizer BRIDGED constructed with instantiate automatically bridges constraints that are not supported by OPTIMIZER using the bridges listed in Bridges. It is recommended for an implementation of MOI to only support constraints that are natively supported by the solver and let bridges transform the constraint to the appropriate form. For this reason it is expected that tests may not pass if OPTIMIZER is used instead of BRIDGED.","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-debug-a-failing-test","page":"Overview","title":"How to debug a failing test","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"When writing a solver, it's likely that you will initially fail many tests. Some failures will be bugs, but other failures you may choose to exclude.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"There are two ways to exclude tests:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Exclude tests whose names contain a string using:\nMOI.Test.runtests(\n    model,\n    config;\n    exclude = String[\"test_to_exclude\", \"test_conic_\"],\n)\nThis will exclude tests whose name contains either of the two strings provided.\nExclude tests which rely on specific functionality using:\nMOI.Test.Config(exclude = Any[MOI.VariableName, MOI.optimize!])\nThis will exclude tests which use the MOI.VariableName attribute, or which call MOI.optimize!.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Each test that fails can be independently called as:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"model = FooBar.Optimizer()\nconfig = MOI.Test.Config()\nMOI.empty!(model)\nMOI.Test.test_category_name_that_failed(model, config)","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"You can look-up the source code of the test that failed by searching for it in the src/Test/test_category.jl file.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nEach test function also has a docstring that explains what the test is for. Use ? MOI.Test.test_category_name_that_failed from the REPL to read it.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Periodically, you should re-run excluded tests to see if they now pass. The easiest way to do this is to swap the exclude keyword argument of runtests to include. For example:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"MOI.Test.runtests(\n    model,\n    config;\n    exclude = String[\"test_to_exclude\", \"test_conic_\"],\n)","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"becomes","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"MOI.Test.runtests(\n    model,\n    config;\n    include = String[\"test_to_exclude\", \"test_conic_\"],\n)","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-add-a-test","page":"Overview","title":"How to add a test","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"To detect bugs in solvers, we add new tests to MOI.Test.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"As an example, ECOS errored calling optimize! twice in a row. (See ECOS.jl PR #72.) We could add a test to ECOS.jl, but that would only stop us from re-introducing the bug to ECOS.jl in the future, but it would not catch other solvers in the ecosystem with the same bug. Instead, if we add a test to MOI.Test, then all solvers will also check that they handle a double optimize call.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"For this test, we care about correctness, rather than performance. therefore, we don't expect solvers to efficiently decide that they have already solved the problem, only that calling optimize! twice doesn't throw an error or give the wrong answer.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 1","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Install the MathOptInterface julia package in dev mode:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"julia> ]\n(@v1.6) pkg> dev MathOptInterface","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 2","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"From here on, proceed with making the following changes in the ~/.julia/dev/MathOptInterface folder (or equivalent dev path on your machine).","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 3","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Since the double-optimize error involves solving an optimization problem, add a new test to src/Test/test_solve.jl:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"\"\"\"\n    test_unit_optimize!_twice(model::MOI.ModelLike, config::Config)\n\nTest that calling `MOI.optimize!` twice does not error.\n\nThis problem was first detected in ECOS.jl PR#72:\nhttps://github.com/jump-dev/ECOS.jl/pull/72\n\"\"\"\nfunction test_unit_optimize!_twice(\n    model::MOI.ModelLike,\n    config::Config{T},\n) where {T}\n    # Use the `@requires` macro to check conditions that the test function\n    # requires to run. Models failing this `@requires` check will silently skip\n    # the test.\n    @requires MOI.supports_constraint(\n        model,\n        MOI.VariableIndex,\n        MOI.GreaterThan{Float64},\n    )\n    @requires _supports(config, MOI.optimize!)\n    # If needed, you can test that the model is empty at the start of the test.\n    # You can assume that this will be the case for tests run via `runtests`.\n    # User's calling tests individually need to call `MOI.empty!` themselves.\n    @test MOI.is_empty(model)\n    # Create a simple model. Try to make this as simple as possible so that the\n    # majority of solvers can run the test.\n    x = MOI.add_variable(model)\n    MOI.add_constraint(model, x, MOI.GreaterThan(one(T)))\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    MOI.set(\n        model,\n        MOI.ObjectiveFunction{MOI.VariableIndex}(),\n        x,\n    )\n    # The main component of the test: does calling `optimize!` twice error?\n    MOI.optimize!(model)\n    MOI.optimize!(model)\n    # Check we have a solution.\n    @test MOI.get(model, MOI.TerminationStatus()) == MOI.OPTIMAL\n    # There is a three-argument version of `Base.isapprox` for checking\n    # approximate equality based on the tolerances defined in `config`:\n    @test isapprox(MOI.get(model, MOI.VariablePrimal(), x), one(T), config)\n    # For code-style, these tests should always `return` `nothing`.\n    return\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"info: Info\nMake sure the function is agnostic to the number type T; don't assume it is a Float64 capable solver.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"We also need to write a test for the test. Place this function immediately below the test you just wrote in the same file:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"function setup_test(\n    ::typeof(test_unit_optimize!_twice),\n    model::MOI.Utilities.MockOptimizer,\n    ::Config,\n)\n    MOI.Utilities.set_mock_optimize!(\n        model,\n        (mock::MOI.Utilities.MockOptimizer) -> MOIU.mock_optimize!(\n            mock,\n            MOI.OPTIMAL,\n            (MOI.FEASIBLE_POINT, [1.0]),\n        ),\n    )\n    return\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Finally, you also need to implement Test.version_added. If we added this test when the latest released version of MOI was v0.10.5, define:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"version_added(::typeof(test_unit_optimize!_twice)) = v\"0.10.6\"","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 6","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Commit the changes to git from ~/.julia/dev/MathOptInterface and submit the PR for review.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nIf you need help writing a test, open an issue on GitHub, or ask the Developer Chatroom.","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Utilities/reference.md\"","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Utilities/reference/#Utilities.Model","page":"API Reference","title":"Utilities.Model","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.Model","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.Model","page":"API Reference","title":"MathOptInterface.Utilities.Model","text":"MOI.Utilities.Model{T}() where {T}\n\nAn implementation of ModelLike that supports all functions and sets defined in MOI. It is parameterized by the coefficient type.\n\nExamples\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Utilities.UniversalFallback","page":"API Reference","title":"Utilities.UniversalFallback","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.UniversalFallback","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.UniversalFallback","page":"API Reference","title":"MathOptInterface.Utilities.UniversalFallback","text":"UniversalFallback\n\nThe UniversalFallback can be applied on a MOI.ModelLike model to create the model UniversalFallback(model) supporting any constraint and attribute. This allows to have a specialized implementation in model for performance critical constraints and attributes while still supporting other attributes with a small performance penalty. Note that model is unaware of constraints and attributes stored by UniversalFallback so this is not appropriate if model is an optimizer (for this reason, MOI.optimize! has not been implemented). In that case, optimizer bridges should be used instead.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Utilities.@model","page":"API Reference","title":"Utilities.@model","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.@model\nUtilities.GenericModel\nUtilities.GenericOptimizer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@model","page":"API Reference","title":"MathOptInterface.Utilities.@model","text":"macro model(\n    model_name,\n    scalar_sets,\n    typed_scalar_sets,\n    vector_sets,\n    typed_vector_sets,\n    scalar_functions,\n    typed_scalar_functions,\n    vector_functions,\n    typed_vector_functions,\n    is_optimizer = false\n)\n\nCreates a type model_name implementing the MOI model interface and supporting all combinations of the provided functions and sets.\n\nEach typed_ scalar/vector sets/functions argument is a tuple of types. A type is \"typed\" if it has a coefficient {T} as the first type parameter.\n\nTuple syntax\n\nTo give no set/function, write (). To give one set or function X, write (X,).\n\nis_optimizer\n\nIf is_optimizer = true, the resulting struct is a of GenericOptimizer, which is a subtype of MOI.AbstractOptimizer, otherwise, it is a GenericModel, which is a subtype of MOI.ModelLike.\n\nVariableIndex\n\nThe function MOI.VariableIndex must not be given in scalar_functions.\nThe model supports MOI.VariableIndex-in-S constraints where S is MOI.EqualTo, MOI.GreaterThan, MOI.LessThan, MOI.Interval, MOI.Integer, MOI.ZeroOne, MOI.Semicontinuous or MOI.Semiinteger.\nThe sets supported with MOI.VariableIndex cannot be controlled from the macro; use UniversalFallback to support more sets.\n\nExamples\n\nThe model describing a linear program would be:\n\n@model(\n    LPModel,                                          # model_name\n    (),                                               # untyped scalar sets\n    (MOI.EqualTo, MOI.GreaterThan, MOI.LessThan, MOI.Interval), #   typed scalar sets\n    (MOI.Zeros, MOI.Nonnegatives, MOI.Nonpositives),  # untyped vector sets\n    (),                                               #   typed vector sets\n    (),                                               # untyped scalar functions\n    (MOI.ScalarAffineFunction,),                      #   typed scalar functions\n    (MOI.VectorOfVariables,),                         # untyped vector functions\n    (MOI.VectorAffineFunction,),                      #   typed vector functions\n    false,                                            # is_optimizer\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.GenericModel","page":"API Reference","title":"MathOptInterface.Utilities.GenericModel","text":"mutable struct GenericModel{T,O,V,C} <: AbstractModelLike{T}\n\nImplements a model supporting coefficients of type T and:\n\nAn objective function stored in .objective::O\nVariables and VariableIndex constraints stored in .variable_bounds::V\nF-in-S constraints (excluding VariableIndex constraints) stored in .constraints::C\n\nAll interactions take place via the MOI interface, so the types O, V, and C must implement the API as needed for their functionality.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.GenericOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.GenericOptimizer","text":"mutable struct GenericOptimizer{T,O,V,C} <: AbstractOptimizer{T}\n\nImplements a model supporting coefficients of type T and:\n\nAn objective function stored in .objective::O\nVariables and VariableIndex constraints stored in .variable_bounds::V\nF-in-S constraints (excluding VariableIndex constraints) stored in .constraints::C\n\nAll interactions take place via the MOI interface, so the types O, V, and C must implement the API as needed for their functionality.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.objective","page":"API Reference","title":".objective","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.ObjectiveContainer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ObjectiveContainer","page":"API Reference","title":"MathOptInterface.Utilities.ObjectiveContainer","text":"ObjectiveContainer{T}\n\nA helper struct to simplify the handling of objective functions in Utilities.Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.variables","page":"API Reference","title":".variables","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.VariablesContainer\nUtilities.FreeVariables","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.VariablesContainer","page":"API Reference","title":"MathOptInterface.Utilities.VariablesContainer","text":"struct VariablesContainer{T} <: AbstractVectorBounds\n    set_mask::Vector{UInt16}\n    lower::Vector{T}\n    upper::Vector{T}\nend\n\nA struct for storing variables and VariableIndex-related constraints. Used in MOI.Utilities.Model by default.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.FreeVariables","page":"API Reference","title":"MathOptInterface.Utilities.FreeVariables","text":"mutable struct FreeVariables <: MOI.ModelLike\n    n::Int64\n    FreeVariables() = new(0)\nend\n\nA struct for storing free variables that can be used as the variables field of GenericModel or GenericModel. It represents a model that does not support any constraint nor objective function.\n\nExample\n\nThe following model type represents a conic model in geometric form. As opposed to VariablesContainer, FreeVariables does not support constraint bounds so they are bridged into an affine constraint in the MOI.Nonnegatives cone as expected for the geometric conic form.\n\njulia> MOI.Utilities.@product_of_sets(\n    Cones,\n    MOI.Zeros,\n    MOI.Nonnegatives,\n    MOI.SecondOrderCone,\n    MOI.PositiveSemidefiniteConeTriangle,\n);\n\njulia> const ConicModel{T} = MOI.Utilities.GenericOptimizer{\n    T,\n    MOI.Utilities.ObjectiveContainer{T},\n    MOI.Utilities.FreeVariables,\n    MOI.Utilities.MatrixOfConstraints{\n        T,\n        MOI.Utilities.MutableSparseMatrixCSC{\n            T,\n            Int,\n            MOI.Utilities.OneBasedIndexing,\n        },\n        Vector{T},\n        Cones{T},\n    },\n};\n\njulia> model = MOI.instantiate(ConicModel{Float64}, with_bridge_type=Float64);\n\njulia> x = MOI.add_variable(model)\nMathOptInterface.VariableIndex(1)\n\njulia> c = MOI.add_constraint(model, x, MOI.GreaterThan(1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n\njulia> MOI.Bridges.is_bridged(model, c)\ntrue\n\njulia> bridge = MOI.Bridges.bridge(model, c)\nMathOptInterface.Bridges.Constraint.VectorizeBridge{Float64, MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives, MathOptInterface.VariableIndex}(MathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1), 1.0)\n\njulia> bridge.vector_constraint\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.Bridges.is_bridged(model, bridge.vector_constraint)\nfalse\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.constraints","page":"API Reference","title":".constraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.VectorOfConstraints\nUtilities.StructOfConstraints\nUtilities.@struct_of_constraints_by_function_types\nUtilities.@struct_of_constraints_by_set_types\nUtilities.struct_of_constraint_code","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.VectorOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.VectorOfConstraints","text":"mutable struct VectorOfConstraints{\n    F<:MOI.AbstractFunction,\n    S<:MOI.AbstractSet,\n} <: MOI.ModelLike\n    constraints::CleverDicts.CleverDict{\n        MOI.ConstraintIndex{F,S},\n        Tuple{F,S},\n        typeof(CleverDicts.key_to_index),\n        typeof(CleverDicts.index_to_key),\n    }\nend\n\nA struct storing F-in-S constraints as a mapping between the constraint indices to the corresponding tuple of function and set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.StructOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.StructOfConstraints","text":"abstract type StructOfConstraints <: MOI.ModelLike end\n\nA struct storing a subfields other structs storing constraints of different types.\n\nSee Utilities.@struct_of_constraints_by_function_types and Utilities.@struct_of_constraints_by_set_types.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@struct_of_constraints_by_function_types","page":"API Reference","title":"MathOptInterface.Utilities.@struct_of_constraints_by_function_types","text":"Utilities.@struct_of_constraints_by_function_types(name, func_types...)\n\nGiven a vector of n function types (F1, F2,..., Fn) in func_types, defines a subtype of StructOfConstraints of name name and which type parameters {T, C1, C2, ..., Cn}. It contains n field where the ith field has type Ci and stores the constraints of function type Fi.\n\nThe expression Fi can also be a union in which case any constraint for which the function type is in the union is stored in the field with type Ci.\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@struct_of_constraints_by_set_types","page":"API Reference","title":"MathOptInterface.Utilities.@struct_of_constraints_by_set_types","text":"Utilities.@struct_of_constraints_by_set_types(name, func_types...)\n\nGiven a vector of n set types (S1, S2,..., Sn) in func_types, defines a subtype of StructOfConstraints of name name and which type parameters {T, C1, C2, ..., Cn}. It contains n field where the ith field has type Ci and stores the constraints of set type Si. The expression Si can also be a union in which case any constraint for which the set type is in the union is stored in the field with type Ci. This can be useful if Ci is a MatrixOfConstraints in order to concatenate the coefficients of constraints of several different set types in the same matrix.\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.struct_of_constraint_code","page":"API Reference","title":"MathOptInterface.Utilities.struct_of_constraint_code","text":"struct_of_constraint_code(struct_name, types, field_types = nothing)\n\nGiven a vector of n Union{SymbolFun,_UnionSymbolFS{SymbolFun}} or Union{SymbolSet,_UnionSymbolFS{SymbolSet}} in types, defines a subtype of StructOfConstraints of name name and which type parameters {T, F1, F2, ..., Fn} if field_types is nothing and a {T} otherwise. It contains n field where the ith field has type Ci if field_types is nothing and type field_types[i] otherwise. If types is vector of Union{SymbolFun,_UnionSymbolFS{SymbolFun}} (resp. Union{SymbolSet,_UnionSymbolFS{SymbolSet}}) then the constraints of that function (resp. set) type are stored in the corresponding field.\n\nThis function is used by the macros @model, @struct_of_constraints_by_function_types and @struct_of_constraints_by_set_types.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Caching-optimizer","page":"API Reference","title":"Caching optimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.CachingOptimizer\nUtilities.attach_optimizer\nUtilities.reset_optimizer\nUtilities.drop_optimizer\nUtilities.state\nUtilities.mode","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.CachingOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.CachingOptimizer","text":"CachingOptimizer\n\nCachingOptimizer is an intermediate layer that stores a cache of the model and links it with an optimizer. It supports incremental model construction and modification even when the optimizer doesn't.\n\nConstructors\n\n    CachingOptimizer(cache::MOI.ModelLike, optimizer::AbstractOptimizer)\n\nCreates a CachingOptimizer in AUTOMATIC mode, with the optimizer optimizer.\n\nThe type of the optimizer returned is CachingOptimizer{typeof(optimizer), typeof(cache)} so it does not support the function reset_optimizer(::CachingOptimizer, new_optimizer) if the type of new_optimizer is different from the type of optimizer.\n\n    CachingOptimizer(cache::MOI.ModelLike, mode::CachingOptimizerMode)\n\nCreates a CachingOptimizer in the NO_OPTIMIZER state and mode mode.\n\nThe type of the optimizer returned is CachingOptimizer{MOI.AbstractOptimizer,typeof(cache)} so it does support the function reset_optimizer(::CachingOptimizer, new_optimizer) if the type of new_optimizer is different from the type of optimizer.\n\nAbout the type\n\nStates\n\nA CachingOptimizer may be in one of three possible states (CachingOptimizerState):\n\nNO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer. The optimizer is not synchronized with the cached model.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.\n\nModes\n\nA CachingOptimizer has two modes of operation (CachingOptimizerMode):\n\nMANUAL: The only methods that change the state of the CachingOptimizer are Utilities.reset_optimizer, Utilities.drop_optimizer, and Utilities.attach_optimizer. Attempting to perform an operation in the incorrect state results in an error.\nAUTOMATIC: The CachingOptimizer changes its state when necessary. For example, optimize! will automatically call attach_optimizer (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.attach_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.attach_optimizer","text":"attach_optimizer(model::CachingOptimizer)\n\nAttaches the optimizer to model, copying all model data into it. Can be called only from the EMPTY_OPTIMIZER state. If the copy succeeds, the CachingOptimizer will be in state ATTACHED_OPTIMIZER after the call, otherwise an error is thrown; see MOI.copy_to for more details on which errors can be thrown.\n\n\n\n\n\nMOIU.attach_optimizer(model::GenericModel)\n\nCall MOIU.attach_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.reset_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.reset_optimizer","text":"reset_optimizer(m::CachingOptimizer, optimizer::MOI.AbstractOptimizer)\n\nSets or resets m to have the given empty optimizer optimizer.\n\nCan be called from any state. An assertion error will be thrown if optimizer is not empty.\n\nThe CachingOptimizer m will be in state EMPTY_OPTIMIZER after the call.\n\n\n\n\n\nreset_optimizer(m::CachingOptimizer)\n\nDetaches and empties the current optimizer. Can be called from ATTACHED_OPTIMIZER or EMPTY_OPTIMIZER state. The CachingOptimizer will be in state EMPTY_OPTIMIZER after the call.\n\n\n\n\n\nMOIU.reset_optimizer(model::GenericModel, optimizer::MOI.AbstractOptimizer)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\nMOIU.reset_optimizer(model::GenericModel)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.drop_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.drop_optimizer","text":"drop_optimizer(m::CachingOptimizer)\n\nDrops the optimizer, if one is present. Can be called from any state. The CachingOptimizer will be in state NO_OPTIMIZER after the call.\n\n\n\n\n\nMOIU.drop_optimizer(model::GenericModel)\n\nCall MOIU.drop_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.state","page":"API Reference","title":"MathOptInterface.Utilities.state","text":"state(m::CachingOptimizer)::CachingOptimizerState\n\nReturns the state of the CachingOptimizer m. See Utilities.CachingOptimizer.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.mode","page":"API Reference","title":"MathOptInterface.Utilities.mode","text":"mode(m::CachingOptimizer)::CachingOptimizerMode\n\nReturns the operating mode of the CachingOptimizer m. See Utilities.CachingOptimizer.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Mock-optimizer","page":"API Reference","title":"Mock optimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MockOptimizer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MockOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.MockOptimizer","text":"MockOptimizer\n\nMockOptimizer is a fake optimizer especially useful for testing. Its main feature is that it can store the values that should be returned for each attribute.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Printing","page":"API Reference","title":"Printing","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.latex_formulation","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.latex_formulation","page":"API Reference","title":"MathOptInterface.Utilities.latex_formulation","text":"latex_formulation(model::MOI.ModelLike; kwargs...)\n\nWrap model in a type so that it can be pretty-printed as text/latex in a notebook like IJulia, or in Documenter.\n\nTo render the model, end the cell with latex_formulation(model), or call display(latex_formulation(model)) in to force the display of the model from inside a function.\n\nPossible keyword arguments are:\n\nsimplify_coefficients : Simplify coefficients if possible by omitting them or removing trailing zeros.\ndefault_name : The name given to variables with an empty name.\nprint_types : Print the MOI type of each function and set for clarity.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Copy-utilities","page":"API Reference","title":"Copy utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.default_copy_to\nUtilities.IndexMap\nUtilities.identity_index_map\nUtilities.ModelFilter","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.default_copy_to","page":"API Reference","title":"MathOptInterface.Utilities.default_copy_to","text":"default_copy_to(dest::MOI.ModelLike, src::MOI.ModelLike)\n\nA default implementation of MOI.copy_to(dest, src) for models that implement the incremental interface, i.e., MOI.supports_incremental_interface returns true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.IndexMap","page":"API Reference","title":"MathOptInterface.Utilities.IndexMap","text":"IndexMap()\n\nThe dictionary-like object returned by MOI.copy_to.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.identity_index_map","page":"API Reference","title":"MathOptInterface.Utilities.identity_index_map","text":"identity_index_map(model::MOI.ModelLike)\n\nReturn an IndexMap that maps all variable and constraint indices of model to themselves.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ModelFilter","page":"API Reference","title":"MathOptInterface.Utilities.ModelFilter","text":"ModelFilter(filter::Function, model::MOI.ModelLike)\n\nA layer to filter out various components of model.\n\nThe filter function takes a single argument, which is each element from the list returned by the attributes below. It returns true if the element should be visible in the filtered model and false otherwise.\n\nThe components that are filtered are:\n\nEntire constraint types via:\nMOI.ListOfConstraintTypesPresent\nIndividual constraints via:\nMOI.ListOfConstraintIndices{F,S}\nSpecific attributes via:\nMOI.ListOfModelAttributesSet\nMOI.ListOfConstraintAttributesSet\nMOI.ListOfVariableAttributesSet\n\nwarning: Warning\nThe list of attributes filtered may change in a future release. You should write functions that are generic and not limited to the five types listed above. Thus, you should probably define a fallback filter(::Any) = true.\n\nSee below for examples of how this works.\n\nnote: Note\nThis layer has a limited scope. It is intended by be used in conjunction with MOI.copy_to.\n\nExample: copy model excluding integer constraints\n\nUse the do syntax to provide a single function.\n\nfiltered_src = MOI.Utilities.ModelFilter(src) do item\n    return item != (MOI.VariableIndex, MOI.Integer)\nend\nMOI.copy_to(dest, filtered_src)\n\nExample: copy model excluding names\n\nUse type dispatch to simplify the implementation:\n\nmy_filter(::Any) = true  # Note the generic fallback!\nmy_filter(::MOI.VariableName) = false\nmy_filter(::MOI.ConstraintName) = false\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nMOI.copy_to(dest, filtered_src)\n\nExample: copy irreducible infeasible subsystem\n\nmy_filter(::Any) = true  # Note the generic fallback!\nfunction my_filter(ci::MOI.ConstraintIndex)\n    status = MOI.get(dest, MOI.ConstraintConflictStatus(), ci)\n    return status != MOI.NOT_IN_CONFLICT\nend\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nMOI.copy_to(dest, filtered_src)\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Penalty-relaxation","page":"API Reference","title":"Penalty relaxation","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.PenaltyRelaxation\nUtilities.ScalarPenaltyRelaxation","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.PenaltyRelaxation","page":"API Reference","title":"MathOptInterface.Utilities.PenaltyRelaxation","text":"PenaltyRelaxation(\n    penalties = Dict{MOI.ConstraintIndex,Float64}();\n    default::Union{Nothing,T} = 1.0,\n)\n\nA problem modifier that, when passed to MOI.modify, destructively modifies the model in-place to create a penalized relaxation of the constraints.\n\nwarning: Warning\nThis is a destructive routine that modifies the model in-place. If you don't want to modify the original model, use JuMP.copy_model to create a copy before calling MOI.modify.\n\nReformulation\n\nSee Utilities.ScalarPenaltyRelaxation for details of the reformulation.\n\nFor each constraint ci, the penalty passed to Utilities.ScalarPenaltyRelaxation is get(penalties, ci, default). If the value is nothing, because ci does not exist in penalties and default = nothing, then the constraint is skipped.\n\nReturn value\n\nMOI.modify(model, PenaltyRelaxation()) returns a Dict{MOI.ConstraintIndex,MOI.ScalarAffineFunction} that maps each constraint index to the corresponding y + z as a MOI.ScalarAffineFunction. In an optimal solution, query the value of these functions to compute the violation of each constraint.\n\nRelax a subset of constraints\n\nTo relax a subset of constraints, pass a penalties dictionary and set default = nothing.\n\nSupported constraint types\n\nThe penalty relaxation is currently limited to modifying MOI.ScalarAffineFunction and MOI.ScalarQuadraticFunction constraints in the linear sets MOI.LessThan, MOI.GreaterThan, MOI.EqualTo and MOI.Interval.\n\nIt does not include variable bound or integrality constraints, because these cannot be modified in-place.\n\nTo modify variable bounds, rewrite them as linear constraints.\n\nExamples\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(default = 2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c] isa MOI.ScalarAffineFunction{Float64}\ntrue\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(Dict(c => 3.0)));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 3.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c] isa MOI.ScalarAffineFunction{Float64}\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ScalarPenaltyRelaxation","page":"API Reference","title":"MathOptInterface.Utilities.ScalarPenaltyRelaxation","text":"ScalarPenaltyRelaxation(penalty::T) where {T}\n\nA problem modifier that, when passed to MOI.modify, destructively modifies the constraint in-place to create a penalized relaxation of the constraint.\n\nwarning: Warning\nThis is a destructive routine that modifies the constraint in-place. If you don't want to modify the original model, use JuMP.copy_model to create a copy before calling MOI.modify.\n\nReformulation\n\nThe penalty relaxation modifies constraints of the form f(x) in S into f(x) + y - z in S, where y z ge 0, and then it introduces a penalty term into the objective of a times (y + z) (if minimizing, else -a), where a is penalty\n\nWhen S is MOI.LessThan or MOI.GreaterThan, we omit y or z respectively as a performance optimization.\n\nReturn value\n\nMOI.modify(model, ci, ScalarPenaltyRelaxation(penalty)) returns y + z as a MOI.ScalarAffineFunction. In an optimal solution, query the value of this function to compute the violation of the constraint.\n\nExamples\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> f = MOI.modify(model, c, MOI.Utilities.ScalarPenaltyRelaxation(2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> f isa MOI.ScalarAffineFunction{Float64}\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MatrixOfConstraints","page":"API Reference","title":"MatrixOfConstraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MatrixOfConstraints","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MatrixOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.MatrixOfConstraints","text":"mutable struct MatrixOfConstraints{T,AT,BT,ST} <: MOI.ModelLike\n    coefficients::AT\n    constants::BT\n    sets::ST\n    caches::Vector{Any}\n    are_indices_mapped::Vector{BitSet}\n    final_touch::Bool\nend\n\nRepresent ScalarAffineFunction and VectorAffinefunction constraints in a matrix form where the linear coefficients of the functions are stored in the coefficients field, the constants of the functions or sets are stored in the constants field. Additional information about the sets are stored in the sets field.\n\nThis model can only be used as the constraints field of a MOI.Utilities.AbstractModel.\n\nWhen the constraints are added, they are stored in the caches field. They are only loaded in the coefficients and constants fields once MOI.Utilities.final_touch is called. For this reason, MatrixOfConstraints should not be used by an incremental interface. Use MOI.copy_to instead.\n\nThe constraints can be added in two different ways:\n\nWith add_constraint, in which case a canonicalized copy of the function is stored in caches.\nWith pass_nonvariable_constraints, in which case the functions and sets are stored themselves in caches without mapping the variable indices. The corresponding index in caches is added in are_indices_mapped. This avoids doing a copy of the function in case the getter of CanonicalConstraintFunction does not make a copy for the source model, e.g., this is the case of VectorOfConstraints.\n\nWe illustrate this with an example. Suppose a model is copied from a src::MOI.Utilities.Model to a bridged model with a MatrixOfConstraints. For all the types that are not bridged, the constraints will be copied with pass_nonvariable_constraints. Hence the functions stored in caches are exactly the same as the ones stored in src. This is ok since this is only during the copy_to operation during which src cannot be modified. On the other hand, for the types that are bridged, the functions added may contain duplicates even if the functions did not contain duplicates in src so duplicates are removed with MOI.Utilities.canonical.\n\nInterface\n\nThe .coefficients::AT type must implement:\n\nAT()\nMOI.empty(::AT)!\nMOI.Utilities.add_column\nMOI.Utilities.set_number_of_rows\nMOI.Utilities.allocate_terms\nMOI.Utilities.load_terms\nMOI.Utilities.final_touch\n\nThe .constants::BT type must implement:\n\nBT()\nBase.empty!(::BT)\nBase.resize(::BT)\nMOI.Utilities.load_constants\nMOI.Utilities.function_constants\nMOI.Utilities.set_from_constants\n\nThe .sets::ST type must implement:\n\nST()\nMOI.is_empty(::ST)\nMOI.empty(::ST)\nMOI.dimension(::ST)\nMOI.is_valid(::ST, ::MOI.ConstraintIndex)\nMOI.get(::ST, ::MOI.ListOfConstraintTypesPresent)\nMOI.get(::ST, ::MOI.NumberOfConstraints)\nMOI.get(::ST, ::MOI.ListOfConstraintIndices)\nMOI.Utilities.set_types\nMOI.Utilities.set_index\nMOI.Utilities.add_set\nMOI.Utilities.rows\nMOI.Utilities.final_touch\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.coefficients","page":"API Reference","title":".coefficients","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.add_column\nUtilities.allocate_terms\nUtilities.set_number_of_rows\nUtilities.load_terms\nUtilities.final_touch\nUtilities.extract_function","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.add_column","page":"API Reference","title":"MathOptInterface.Utilities.add_column","text":"add_column(coefficients)::Nothing\n\nTell coefficients to pre-allocate datastructures as needed to store one column.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.allocate_terms","page":"API Reference","title":"MathOptInterface.Utilities.allocate_terms","text":"allocate_terms(coefficients, index_map, func)::Nothing\n\nTell coefficients that the terms of the function func where the variable indices are mapped with index_map will be loaded with load_terms.\n\nThe function func must be canonicalized before calling allocate_terms. See is_canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_number_of_rows","page":"API Reference","title":"MathOptInterface.Utilities.set_number_of_rows","text":"set_number_of_rows(coefficients, n)::Nothing\n\nTell coefficients to pre-allocate datastructures as needed to store n rows.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.load_terms","page":"API Reference","title":"MathOptInterface.Utilities.load_terms","text":"load_terms(coefficients, index_map, func, offset)::Nothing\n\nLoads the terms of func to coefficients, mapping the variable indices with index_map.\n\nThe ith dimension of func is loaded at the (offset + i)th row of coefficients.\n\nThe function must be allocated first with allocate_terms.\n\nThe function func must be canonicalized, see is_canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.final_touch","page":"API Reference","title":"MathOptInterface.Utilities.final_touch","text":"final_touch(coefficients)::Nothing\n\nInforms the coefficients that all functions have been added with load_terms. No more modification is allowed unless MOI.empty! is called.\n\nfinal_touch(sets)::Nothing\n\nInforms the sets that all functions have been added with add_set. No more modification is allowed unless MOI.empty! is called.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.extract_function","page":"API Reference","title":"MathOptInterface.Utilities.extract_function","text":"extract_function(coefficients, row::Integer, constant::T) where {T}\n\nReturn the MOI.ScalarAffineFunction{T} function corresponding to row row in coefficients.\n\nextract_function(\n    coefficients,\n    rows::UnitRange,\n    constants::Vector{T},\n) where{T}\n\nReturn the MOI.VectorAffineFunction{T} function corresponding to rows rows in coefficients.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MutableSparseMatrixCSC\nUtilities.AbstractIndexing\nUtilities.ZeroBasedIndexing\nUtilities.OneBasedIndexing","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MutableSparseMatrixCSC","page":"API Reference","title":"MathOptInterface.Utilities.MutableSparseMatrixCSC","text":"mutable struct MutableSparseMatrixCSC{Tv,Ti<:Integer,I<:AbstractIndexing}\n    indexing::I\n    m::Int\n    n::Int\n    colptr::Vector{Ti}\n    rowval::Vector{Ti}\n    nzval::Vector{Tv}\n    nz_added::Vector{Ti}\nend\n\nMatrix type loading sparse matrices in the Compressed Sparse Column format. The indexing used is indexing, see AbstractIndexing. The other fields have the same meaning than for SparseArrays.SparseMatrixCSC except that the indexing is different unless indexing is OneBasedIndexing. In addition, nz_added is used to cache the number of non-zero terms that have been added to each column due to the incremental nature of load_terms.\n\nThe matrix is loaded in 5 steps:\n\nMOI.empty! is called.\nMOI.Utilities.add_column and MOI.Utilities.allocate_terms are called in any order.\nMOI.Utilities.set_number_of_rows is called.\nMOI.Utilities.load_terms is called for each affine function.\nMOI.Utilities.final_touch is called.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.AbstractIndexing","page":"API Reference","title":"MathOptInterface.Utilities.AbstractIndexing","text":"abstract type AbstractIndexing end\n\nIndexing to be used for storing the row and column indices of MutableSparseMatrixCSC. See ZeroBasedIndexing and OneBasedIndexing.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ZeroBasedIndexing","page":"API Reference","title":"MathOptInterface.Utilities.ZeroBasedIndexing","text":"struct ZeroBasedIndexing <: AbstractIndexing end\n\nZero-based indexing: the ith row or column has index i - 1. This is useful when the vectors of row and column indices need to be communicated to a library using zero-based indexing such as C libraries.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.OneBasedIndexing","page":"API Reference","title":"MathOptInterface.Utilities.OneBasedIndexing","text":"struct ZeroBasedIndexing <: AbstractIndexing end\n\nOne-based indexing: the ith row or column has index i. This enables an allocation-free conversion of MutableSparseMatrixCSC to SparseArrays.SparseMatrixCSC.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.constants","page":"API Reference","title":".constants","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.load_constants\nUtilities.function_constants\nUtilities.set_from_constants","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.load_constants","page":"API Reference","title":"MathOptInterface.Utilities.load_constants","text":"load_constants(constants, offset, func_or_set)::Nothing\n\nThis function loads the constants of func_or_set in constants at an offset of offset. Where offset is the sum of the dimensions of the constraints already loaded. The storage should be preallocated with resize! before calling this function.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\nThe constants are loaded in three steps:\n\nBase.empty! is called.\nBase.resize! is called with the sum of the dimensions of all constraints.\nMOI.Utilities.load_constants is called for each function for vector constraint or set for scalar constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.function_constants","page":"API Reference","title":"MathOptInterface.Utilities.function_constants","text":"function_constants(constants, rows)\n\nThis function returns the function constants that were loaded with load_constants at the rows rows.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_from_constants","page":"API Reference","title":"MathOptInterface.Utilities.set_from_constants","text":"set_from_constants(constants, S::Type, rows)::S\n\nThis function returns an instance of the set S for which the constants where loaded with load_constants at the rows rows.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.Hyperrectangle","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.Hyperrectangle","page":"API Reference","title":"MathOptInterface.Utilities.Hyperrectangle","text":"struct Hyperrectangle{T} <: AbstractVectorBounds\n    lower::Vector{T}\n    upper::Vector{T}\nend\n\nA struct for the .constants field in MatrixOfConstraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.sets","page":"API Reference","title":".sets","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.set_index\nUtilities.set_types\nUtilities.add_set\nUtilities.rows\nUtilities.num_rows\nUtilities.set_with_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_index","page":"API Reference","title":"MathOptInterface.Utilities.set_index","text":"set_index(sets, ::Type{S})::Union{Int,Nothing} where {S<:MOI.AbstractSet}\n\nReturn an integer corresponding to the index of the set type in the list given by set_types.\n\nIf S is not part of the list, return nothing.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_types","page":"API Reference","title":"MathOptInterface.Utilities.set_types","text":"set_types(sets)::Vector{Type}\n\nReturn the list of the types of the sets allowed in sets.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.add_set","page":"API Reference","title":"MathOptInterface.Utilities.add_set","text":"add_set(sets, i)::Int64\n\nAdd a scalar set of type index i.\n\nadd_set(sets, i, dim)::Int64\n\nAdd a vector set of type index i and dimension dim.\n\nBoth methods return a unique Int64 of the set that can be used to reference this set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.rows","page":"API Reference","title":"MathOptInterface.Utilities.rows","text":"rows(sets, ci::MOI.ConstraintIndex)::Union{Int,UnitRange{Int}}\n\nReturn the rows in 1:MOI.dimension(sets) corresponding to the set of id ci.value.\n\nFor scalar sets, this returns an Int. For vector sets, this returns an UnitRange{Int}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.num_rows","page":"API Reference","title":"MathOptInterface.Utilities.num_rows","text":"num_rows(sets::OrderedProductOfSets, ::Type{S}) where {S}\n\nReturn the number of rows corresponding to a set of type S. That is, it is the sum of the dimensions of the sets of type S.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_with_dimension","page":"API Reference","title":"MathOptInterface.Utilities.set_with_dimension","text":"set_with_dimension(::Type{S}, dim) where {S<:MOI.AbstractVectorSet}\n\nReturns the instance of S of MOI.dimension dim. This needs to be implemented for sets of type S to be useable with MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.ProductOfSets\nUtilities.MixOfScalarSets\nUtilities.@mix_of_scalar_sets\nUtilities.OrderedProductOfSets\nUtilities.@product_of_sets","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ProductOfSets","page":"API Reference","title":"MathOptInterface.Utilities.ProductOfSets","text":"abstract type ProductOfSets{T} end\n\nRepresents a cartesian product of sets of given types.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MixOfScalarSets","page":"API Reference","title":"MathOptInterface.Utilities.MixOfScalarSets","text":"abstract type MixOfScalarSets{T} <: ProductOfSets{T} end\n\nProduct of scalar sets in the order the constraints are added, mixing the constraints of different types.\n\nUse @mix_of_scalar_sets to generate a new subtype.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@mix_of_scalar_sets","page":"API Reference","title":"MathOptInterface.Utilities.@mix_of_scalar_sets","text":"@mix_of_scalar_sets(name, set_types...)\n\nGenerate a new MixOfScalarSets subtype.\n\nExample\n\n@mix_of_scalar_sets(\n    MixedIntegerLinearProgramSets,\n    MOI.GreaterThan{T},\n    MOI.LessThan{T},\n    MOI.EqualTo{T},\n    MOI.Integer,\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.OrderedProductOfSets","page":"API Reference","title":"MathOptInterface.Utilities.OrderedProductOfSets","text":"abstract type OrderedProductOfSets{T} <: ProductOfSets{T} end\n\nProduct of sets in the order the constraints are added, grouping the constraints of the same types contiguously.\n\nUse @product_of_sets to generate new subtypes.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@product_of_sets","page":"API Reference","title":"MathOptInterface.Utilities.@product_of_sets","text":"@product_of_sets(name, set_types...)\n\nGenerate a new OrderedProductOfSets subtype.\n\nExample\n\n@product_of_sets(\n    LinearOrthants,\n    MOI.Zeros,\n    MOI.Nonnegatives,\n    MOI.Nonpositives,\n    MOI.ZeroOne,\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#Fallbacks","page":"API Reference","title":"Fallbacks","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.get_fallback","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.get_fallback","page":"API Reference","title":"MathOptInterface.Utilities.get_fallback","text":"get_fallback(model::MOI.ModelLike, ::MOI.ObjectiveValue)\n\nCompute the objective function value using the VariablePrimal results and the ObjectiveFunction value.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, ::MOI.DualObjectiveValue, T::Type)::T\n\nCompute the dual objective value of type T using the ConstraintDual results and the ConstraintFunction and ConstraintSet values. Note that the nonlinear part of the model is ignored.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, ::MOI.ConstraintPrimal,\n             constraint_index::MOI.ConstraintIndex)\n\nCompute the value of the function of the constraint of index constraint_index using the VariablePrimal results and the ConstraintFunction values.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, attr::MOI.ConstraintDual,\n             ci::MOI.ConstraintIndex{Union{MOI.VariableIndex,\n                                           MOI.VectorOfVariables}})\n\nCompute the dual of the constraint of index ci using the ConstraintDual of other constraints and the ConstraintFunction values. Throws an error if some constraints are quadratic or if there is one another MOI.VariableIndex-in-S or MOI.VectorOfVariables-in-S constraint with one of the variables in the function of the constraint ci.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Function-utilities","page":"API Reference","title":"Function utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for functions:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.eval_variables\nUtilities.map_indices\nUtilities.substitute_variables\nUtilities.filter_variables\nUtilities.remove_variable\nUtilities.all_coefficients\nUtilities.unsafe_add\nUtilities.isapprox_zero\nUtilities.modify_function\nUtilities.zero_with_output_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.eval_variables","page":"API Reference","title":"MathOptInterface.Utilities.eval_variables","text":"eval_variables(value_fn::Function, f::MOI.AbstractFunction)\n\nReturns the value of function f if each variable index vi is evaluated as value_fn(vi).\n\nNote that value_fn must return a Number. See substitute_variables for a similar function where value_fn returns an MOI.AbstractScalarFunction.\n\nwarning: Warning\nThe two-argument version of eval_variables is deprecated and may be removed in MOI v2.0.0. Use the three-argument method eval_variables(::Function, ::MOI.ModelLike, ::MOI.AbstractFunction) instead.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.map_indices","page":"API Reference","title":"MathOptInterface.Utilities.map_indices","text":"map_indices(index_map::Function, attr::MOI.AnyAttribute, x::X)::X where {X}\n\nSubstitute any MOI.VariableIndex (resp. MOI.ConstraintIndex) in x by the MOI.VariableIndex (resp. MOI.ConstraintIndex) of the same type given by index_map(x).\n\nWhen to implement this method for new types X\n\nThis function is used by implementations of MOI.copy_to on constraint functions, attribute values and submittable values. If you define a new attribute whose values x::X contain variable or constraint indices, you must also implement this function.\n\n\n\n\n\nmap_indices(\n    variable_map::AbstractDict{T,T},\n    x::X,\n)::X where {T<:MOI.Index,X}\n\nShortcut for map_indices(vi -> variable_map[vi], x).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.substitute_variables","page":"API Reference","title":"MathOptInterface.Utilities.substitute_variables","text":"substitute_variables(variable_map::Function, x)\n\nSubstitute any MOI.VariableIndex in x by variable_map(x). The variable_map function returns either MOI.VariableIndex or MOI.ScalarAffineFunction, see eval_variables for a similar function where variable_map returns a number.\n\nThis function is used by bridge optimizers on constraint functions, attribute values and submittable values when at least one variable bridge is used hence it needs to be implemented for custom types that are meant to be used as attribute or submittable value.\n\nnote: Note\nWhen implementing a new method, don't use substitute_variables(::Function, because Julia will not specialize on it. Use instead substitute_variables(::F, ...) where {F<:Function}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.filter_variables","page":"API Reference","title":"MathOptInterface.Utilities.filter_variables","text":"filter_variables(keep::Function, f::AbstractFunction)\n\nReturn a new function f with the variable vi such that !keep(vi) removed.\n\nWARNING: Don't define filter_variables(::Function, ...) because Julia will not specialize on this. Define instead filter_variables(::F, ...) where {F<:Function}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.remove_variable","page":"API Reference","title":"MathOptInterface.Utilities.remove_variable","text":"remove_variable(f::AbstractFunction, vi::VariableIndex)\n\nReturn a new function f with the variable vi removed.\n\n\n\n\n\nremove_variable(\n    f::MOI.AbstractFunction,\n    s::MOI.AbstractSet,\n    vi::MOI.VariableIndex,\n)\n\nReturn a tuple (g, t) representing the constraint f-in-s with the variable vi removed. That is, the terms containing the variable vi in the function f are removed and the dimension of the set s is updated if needed (e.g. when f is a VectorOfVariables with vi being one of the variables).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.all_coefficients","page":"API Reference","title":"MathOptInterface.Utilities.all_coefficients","text":"all_coefficients(p::Function, f::MOI.AbstractFunction)\n\nDetermine whether predicate p returns true for all coefficients of f, returning false as soon as the first coefficient of f for which p returns false is encountered (short-circuiting). Similar to all.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.unsafe_add","page":"API Reference","title":"MathOptInterface.Utilities.unsafe_add","text":"unsafe_add(t1::MOI.ScalarAffineTerm, t2::MOI.ScalarAffineTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.ScalarAffineTerm. It is unsafe because it uses the variable of t1 as the variable of the output without checking that it is equal to that of t2.\n\n\n\n\n\nunsafe_add(t1::MOI.ScalarQuadraticTerm, t2::MOI.ScalarQuadraticTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.ScalarQuadraticTerm. It is unsafe because it uses the variable's of t1 as the variable's of the output without checking that they are the same (up to permutation) to those of t2.\n\n\n\n\n\nunsafe_add(t1::MOI.VectorAffineTerm, t2::MOI.VectorAffineTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.VectorAffineTerm. It is unsafe because it uses the output_index and variable of t1 as the output_index and variable of the output term without checking that they are equal to those of t2.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.isapprox_zero","page":"API Reference","title":"MathOptInterface.Utilities.isapprox_zero","text":"isapprox_zero(f::MOI.AbstractFunction, tol)\n\nReturn a Bool indicating whether the function f is approximately zero using tol as a tolerance.\n\nImportant note\n\nThis function assumes that f does not contain any duplicate terms, you might want to first call canonical if that is not guaranteed. For instance, given\n\nf = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.([1, -1], [x, x]), 0)\n\nthen isapprox_zero(f) is false but isapprox_zero(MOIU.canonical(f)) is true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.modify_function","page":"API Reference","title":"MathOptInterface.Utilities.modify_function","text":"modify_function(f::AbstractFunction, change::AbstractFunctionModification)\n\nReturn a copy of the function f, modified according to change.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.zero_with_output_dimension","page":"API Reference","title":"MathOptInterface.Utilities.zero_with_output_dimension","text":"zero_with_output_dimension(::Type{T}, output_dimension::Integer) where {T}\n\nCreate an instance of type T with the output dimension output_dimension.\n\nThis is mostly useful in Bridges, when code needs to be agnostic to the type of vector-valued function that is passed in.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following functions can be used to canonicalize a function:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.is_canonical\nUtilities.canonical\nUtilities.canonicalize!","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.is_canonical","page":"API Reference","title":"MathOptInterface.Utilities.is_canonical","text":"is_canonical(f::Union{ScalarAffineFunction, VectorAffineFunction})\n\nReturns a Bool indicating whether the function is in canonical form. See canonical.\n\n\n\n\n\nis_canonical(f::Union{ScalarQuadraticFunction, VectorQuadraticFunction})\n\nReturns a Bool indicating whether the function is in canonical form. See canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.canonical","page":"API Reference","title":"MathOptInterface.Utilities.canonical","text":"canonical(f::MOI.AbstractFunction)\n\nReturns the function in a canonical form, i.e.\n\nA term appear only once.\nThe coefficients are nonzero.\nThe terms appear in increasing order of variable where there the order of the variables is the order of their value.\nFor a AbstractVectorFunction, the terms are sorted in ascending order of output index.\n\nThe output of canonical can be assumed to be a copy of f, even for VectorOfVariables.\n\nExamples\n\nIf x (resp. y, z) is VariableIndex(1) (resp. 2, 3). The canonical representation of ScalarAffineFunction([y, x, z, x, z], [2, 1, 3, -2, -3], 5) is ScalarAffineFunction([x, y], [-1, 2], 5).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.canonicalize!","page":"API Reference","title":"MathOptInterface.Utilities.canonicalize!","text":"canonicalize!(f::Union{ScalarAffineFunction, VectorAffineFunction})\n\nConvert a function to canonical form in-place, without allocating a copy to hold the result. See canonical.\n\n\n\n\n\ncanonicalize!(f::Union{ScalarQuadraticFunction, VectorQuadraticFunction})\n\nConvert a function to canonical form in-place, without allocating a copy to hold the result. See canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following functions can be used to manipulate functions with basic algebra:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.scalar_type\nUtilities.scalarize\nUtilities.eachscalar\nUtilities.promote_operation\nUtilities.operate\nUtilities.operate!\nUtilities.operate_output_index!\nUtilities.vectorize","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.scalar_type","page":"API Reference","title":"MathOptInterface.Utilities.scalar_type","text":"scalar_type(F::Type{<:MOI.AbstractVectorFunction})\n\nType of functions obtained by indexing objects obtained by calling eachscalar on functions of type F.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.scalarize","page":"API Reference","title":"MathOptInterface.Utilities.scalarize","text":"scalarize(func::MOI.VectorOfVariables, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.SingleVariable}.\n\nSee also eachscalar.\n\n\n\n\n\nscalarize(func::MOI.VectorAffineFunction{T}, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.ScalarAffineFunction{T}}.\n\nSee also eachscalar.\n\n\n\n\n\nscalarize(func::MOI.VectorQuadraticFunction{T}, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.ScalarQuadraticFunction{T}}.\n\nSee also eachscalar.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.eachscalar","page":"API Reference","title":"MathOptInterface.Utilities.eachscalar","text":"eachscalar(f::MOI.AbstractVectorFunction)\n\nReturns an iterator for the scalar components of the vector function.\n\nSee also scalarize.\n\n\n\n\n\neachscalar(f::MOI.AbstractVector)\n\nReturns an iterator for the scalar components of the vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.promote_operation","page":"API Reference","title":"MathOptInterface.Utilities.promote_operation","text":"promote_operation(\n    op::Function,\n    ::Type{T},\n    ArgsTypes::Type{<:Union{T,AbstractVector{T},MOI.AbstractFunction}}...,\n) where {T<:Number}\n\nCompute the return type of the call operate(op, T, args...), where the types of the arguments args are ArgsTypes.\n\nOne assumption is that the element type T is invariant under each operation. That is, op(::T, ::T)::T where op is a +, -, *, and /.\n\nThere are six methods for which we implement Utilities.promote_operation:\n\n+ a. promote_operation(::typeof(+), ::Type{T}, ::Type{F1}, ::Type{F2})\n- a. promote_operation(::typeof(-), ::Type{T}, ::Type{F}) b. promote_operation(::typeof(-), ::Type{T}, ::Type{F1}, ::Type{F2})\n* a. promote_operation(::typeof(*), ::Type{T}, ::Type{T}, ::Type{F}) b. promote_operation(::typeof(*), ::Type{T}, ::Type{F}, ::Type{T}) c. promote_operation(::typeof(*), ::Type{T}, ::Type{F1}, ::Type{F2})    where F1 and F2 are VariableIndex or ScalarAffineFunction d. promote_operation(::typeof(*), ::Type{T}, ::Type{<:Diagonal{T}}, ::Type{F}\n/ a. promote_operation(::typeof(/), ::Type{T}, ::Type{F}, ::Type{T})\nvcat a. promote_operation(::typeof(vcat), ::Type{T}, ::Type{F}...)\nimag a. promote_operation(::typeof(imag), ::Type{T}, ::Type{F})    where F is VariableIndex or VectorOfVariables\n\nIn each case, F (or F1 and F2) is one of the ten supported types, with a restriction that the mathematical operation makes sense, for example, we don't define promote_operation(-, T, F1, F2) where F1 is a scalar-valued function and  F2 is a vector-valued function. The ten supported types are:\n\n::T\n::VariableIndex\n::ScalarAffineFunction{T}\n::ScalarQuadraticFunction{T}\n::ScalarNonlinearFunction\n::AbstractVector{T}\n::VectorOfVariables\n::VectorAffineFunction{T}\n::VectorQuadraticFunction{T}\n::VectorNonlinearFunction\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate","page":"API Reference","title":"MathOptInterface.Utilities.operate","text":"operate(\n    op::Function,\n    ::Type{T},\n    args::Union{T,MOI.AbstractFunction}...,\n)::MOI.AbstractFunction where {T<:Number}\n\nReturns an MOI.AbstractFunction representing the function resulting from the operation op(args...) on functions of coefficient type T.\n\nNo argument can be modified.\n\nMethods\n\n+ a. operate(::typeof(+), ::Type{T}, ::F1) b. operate(::typeof(+), ::Type{T}, ::F1, ::F2) c. operate(::typeof(+), ::Type{T}, ::F1...)\n- a. operate(::typeof(-), ::Type{T}, ::F) b. operate(::typeof(-), ::Type{T}, ::F1, ::F2)\n* a. operate(::typeof(*), ::Type{T}, ::T, ::F) b. operate(::typeof(*), ::Type{T}, ::F, ::T) c. operate(::typeof(*), ::Type{T}, ::F1, ::F2)    where F1 and F2 are VariableIndex or ScalarAffineFunction d. operate(::typeof(*), ::Type{T}, ::Diagonal{T}, ::F)\n/ a. operate(::typeof(/), ::Type{T}, ::F, ::T)\nvcat a. operate(::typeof(vcat), ::Type{T}, ::F...)\nimag a. operate(::typeof(imag), ::Type{T}, ::F)    where F is VariableIndex or VectorOfVariables\n\nOne assumption is that the element type T is invariant under each operation. That is, op(::T, ::T)::T where op is a +, -, *, and /.\n\nIn each case, F (or F1 and F2) is one of the ten supported types, with a restriction that the mathematical operation makes sense, for example, we don't define promote_operation(-, T, F1, F2) where F1 is a scalar-valued function and  F2 is a vector-valued function. The ten supported types are:\n\n::T\n::VariableIndex\n::ScalarAffineFunction{T}\n::ScalarQuadraticFunction{T}\n::ScalarNonlinearFunction\n::AbstractVector{T}\n::VectorOfVariables\n::VectorAffineFunction{T}\n::VectorQuadraticFunction{T}\n::VectorNonlinearFunction\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate!","page":"API Reference","title":"MathOptInterface.Utilities.operate!","text":"operate!(\n    op::Function,\n    ::Type{T},\n    args::Union{T,MOI.AbstractFunction}...,\n)::MOI.AbstractFunction where {T<:Number}\n\nReturns an MOI.AbstractFunction representing the function resulting from the operation op(args...) on functions of coefficient type T.\n\nThe first argument may be modified, in which case the return value is identical to the first argument. For operations which cannot be implemented in-place, this function returns a new object.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate_output_index!","page":"API Reference","title":"MathOptInterface.Utilities.operate_output_index!","text":"operate_output_index!(\n    op::Union{typeof(+),typeof(-)},\n    ::Type{T},\n    output_index::Integer,\n    f::Union{AbstractVector{T},MOI.AbstractVectorFunction}\n    g::Union{T,MOI.AbstractScalarFunction}...\n) where {T<:Number}\n\nReturn an MOI.AbstractVectorFunction in which the scalar function in row output_index is the result of op(f[output_index], g).\n\nThe functions at output index different to output_index are the same as the functions at the same output index in func. The first argument may be modified.\n\nMethods\n\n+ a. operate_output_index!(+, ::Type{T}, ::Int, ::VectorF, ::ScalarF)\n- a. operate_output_index!(-, ::Type{T}, ::Int, ::VectorF, ::ScalarF)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.vectorize","page":"API Reference","title":"MathOptInterface.Utilities.vectorize","text":"vectorize(x::AbstractVector{<:Number})\n\nReturns x.\n\n\n\n\n\nvectorize(x::AbstractVector{MOI.VariableIndex})\n\nReturns the vector of scalar affine functions in the form of a MOI.VectorAffineFunction{T}.\n\n\n\n\n\nvectorize(funcs::AbstractVector{MOI.ScalarAffineFunction{T}}) where T\n\nReturns the vector of scalar affine functions in the form of a MOI.VectorAffineFunction{T}.\n\n\n\n\n\nvectorize(funcs::AbstractVector{MOI.ScalarQuadraticFunction{T}}) where T\n\nReturns the vector of scalar quadratic functions in the form of a MOI.VectorQuadraticFunction{T}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Constraint-utilities","page":"API Reference","title":"Constraint utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for moving the function constant to the set for scalar constraints:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.shift_constant\nUtilities.supports_shift_constant\nUtilities.normalize_and_add_constraint\nUtilities.normalize_constant","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.shift_constant","page":"API Reference","title":"MathOptInterface.Utilities.shift_constant","text":"shift_constant(set::MOI.AbstractScalarSet, offset)\n\nReturns a new scalar set new_set such that func-in-set is equivalent to func + offset-in-new_set.\n\nOnly define this function if it makes sense to!\n\nUse supports_shift_constant to check if the set supports shifting:\n\nif supports_shift_constant(typeof(old_set))\n    new_set = shift_constant(old_set, offset)\n    f.constant = 0\n    add_constraint(model, f, new_set)\nelse\n    add_constraint(model, f, old_set)\nend\n\nSee also supports_shift_constant.\n\nExamples\n\nThe call shift_constant(MOI.Interval(-2, 3), 1) is equal to MOI.Interval(-1, 4).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.supports_shift_constant","page":"API Reference","title":"MathOptInterface.Utilities.supports_shift_constant","text":"supports_shift_constant(::Type{S}) where {S<:MOI.AbstractSet}\n\nReturn true if shift_constant is defined for set S.\n\nSee also shift_constant.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.normalize_and_add_constraint","page":"API Reference","title":"MathOptInterface.Utilities.normalize_and_add_constraint","text":"normalize_and_add_constraint(\n    model::MOI.ModelLike,\n    func::MOI.AbstractScalarFunction,\n    set::MOI.AbstractScalarSet;\n    allow_modify_function::Bool = false,\n)\n\nAdds the scalar constraint obtained by moving the constant term in func to the set in model. If allow_modify_function is true then the function func can be modified.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.normalize_constant","page":"API Reference","title":"MathOptInterface.Utilities.normalize_constant","text":"normalize_constant(\n    func::MOI.AbstractScalarFunction,\n    set::MOI.AbstractScalarSet;\n    allow_modify_function::Bool = false,\n)\n\nReturn the func-in-set constraint in normalized form. That is, if func is MOI.ScalarQuadraticFunction or MOI.ScalarAffineFunction, the constant is moved to the set. If allow_modify_function is true then the function func can be modified.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utility identifies those constraints imposing bounds on a given variable, and returns those bound values:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.get_bounds","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.get_bounds","page":"API Reference","title":"MathOptInterface.Utilities.get_bounds","text":"get_bounds(model::MOI.ModelLike, ::Type{T}, x::MOI.VariableIndex)\n\nReturn a tuple (lb, ub) of type Tuple{T, T}, where lb and ub are lower  and upper bounds, respectively, imposed on x in model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are useful when working with symmetric matrix cones.","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.is_diagonal_vectorized_index\nUtilities.side_dimension_for_vectorized_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.is_diagonal_vectorized_index","page":"API Reference","title":"MathOptInterface.Utilities.is_diagonal_vectorized_index","text":"is_diagonal_vectorized_index(index::Base.Integer)\n\nReturn whether index is the index of a diagonal element in a MOI.AbstractSymmetricMatrixSetTriangle set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.side_dimension_for_vectorized_dimension","page":"API Reference","title":"MathOptInterface.Utilities.side_dimension_for_vectorized_dimension","text":"side_dimension_for_vectorized_dimension(n::Integer)\n\nReturn the dimension d such that MOI.dimension(MOI.PositiveSemidefiniteConeTriangle(d)) is n.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Set-utilities","page":"API Reference","title":"Set utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for sets:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.AbstractDistance\nUtilities.ProjectionUpperBoundDistance\nUtilities.distance_to_set\nUtilities.set_dot","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.AbstractDistance","page":"API Reference","title":"MathOptInterface.Utilities.AbstractDistance","text":"abstract type AbstractDistance end\n\nAn abstract type used to enabble dispatch of Utilities.distance_to_set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ProjectionUpperBoundDistance","page":"API Reference","title":"MathOptInterface.Utilities.ProjectionUpperBoundDistance","text":"ProjectionUpperBoundDistance() <: AbstractDistance\n\nAn upper bound on the minimum distance between point and the closest feasible point in set.\n\nDefinition of distance\n\nThe minimum distance is computed as:\n\nd(x mathcalK) = min_y in mathcalK  x - y \n\nwhere x is point and mathcalK is set. The norm is computed as:\n\nx = sqrtf(x x mathcalK)\n\nwhere f is Utilities.set_dot.\n\nIn the default case, where the set does not have a specialized method for Utilities.set_dot, the norm is equivalent to the Euclidean norm x = sqrtsum x_i^2.\n\nWhy an upper bound?\n\nIn most cases, distance_to_set should return the smallest upper bound, but it may return a larger value if the smallest upper bound is expensive to compute.\n\nFor example, given an epigraph from of a conic set, (t x)  f(x) le t, it may be simpler to return delta such that f(x) le t + delta, rather than computing the nearest projection onto the set.\n\nIf the distance is not the smallest upper bound, the docstring of the appropriate distance_to_set method must describe the way that the distance is computed.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.distance_to_set","page":"API Reference","title":"MathOptInterface.Utilities.distance_to_set","text":"distance_to_set(\n    [d::AbstractDistance = ProjectionUpperBoundDistance()],]\n    point::T,\n    set::MOI.AbstractScalarSet,\n) where {T}\n\ndistance_to_set(\n    [d::AbstractDistance = ProjectionUpperBoundDistance(),]\n    point::AbstractVector{T},\n    set::MOI.AbstractVectorSet,\n) where {T}\n\nCompute the distance between point and set using the distance metric d. If point is in the set set, this function must return zero(T).\n\nIf d is omitted, the default distance is Utilities.ProjectionUpperBoundDistance.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_dot","page":"API Reference","title":"MathOptInterface.Utilities.set_dot","text":"set_dot(x::AbstractVector, y::AbstractVector, set::AbstractVectorSet)\n\nReturn the scalar product between a vector x of the set set and a vector y of the dual of the set s.\n\n\n\n\n\nset_dot(x, y, set::AbstractScalarSet)\n\nReturn the scalar product between a number x of the set set and a number y of the dual of the set s.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#DoubleDicts","page":"API Reference","title":"DoubleDicts","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.DoubleDicts.DoubleDict\nUtilities.DoubleDicts.DoubleDictInner\nUtilities.DoubleDicts.IndexDoubleDict\nUtilities.DoubleDicts.IndexDoubleDictInner\nUtilities.DoubleDicts.outer_keys\nUtilities.DoubleDicts.nonempty_outer_keys","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.DoubleDict","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.DoubleDict","text":"DoubleDict{V}\n\nAn optimized dictionary to map MOI.ConstraintIndex to values of type V.\n\nWorks as a AbstractDict{MOI.ConstraintIndex,V} with minimal differences.\n\nIf V is also a MOI.ConstraintIndex, use IndexDoubleDict.\n\nNote that MOI.ConstraintIndex is not a concrete type, opposed to MOI.ConstraintIndex{MOI.VariableIndex, MOI.Integers}, which is a concrete type.\n\nWhen looping through multiple keys of the same Function-in-Set type, use\n\ninner = dict[F, S]\n\nto return a type-stable DoubleDictInner.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.DoubleDictInner","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.DoubleDictInner","text":"DoubleDictInner{F,S,V}\n\nA type stable inner dictionary of DoubleDict.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.IndexDoubleDict","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.IndexDoubleDict","text":"IndexDoubleDict\n\nA specialized version of [DoubleDict] in which the values are of type MOI.ConstraintIndex\n\nWhen looping through multiple keys of the same Function-in-Set type, use\n\ninner = dict[F, S]\n\nto return a type-stable IndexDoubleDictInner.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.IndexDoubleDictInner","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.IndexDoubleDictInner","text":"IndexDoubleDictInner{F,S}\n\nA type stable inner dictionary of IndexDoubleDict.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.outer_keys","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.outer_keys","text":"outer_keys(d::AbstractDoubleDict)\n\nReturn an iterator over the outer keys of the AbstractDoubleDict d. Each outer key is a Tuple{Type,Type} so that a double loop can be easily used:\n\nfor (F, S) in DoubleDicts.outer_keys(dict)\n    for (k, v) in dict[F, S]\n        # ...\n    end\nend\n\nFor performance, it is recommended that the inner loop lies in a separate function to gurantee type-stability. Some outer keys (F, S) might lead to an empty dict[F, S]. If you want only nonempty dict[F, S], use nonempty_outer_keys.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.nonempty_outer_keys","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.nonempty_outer_keys","text":"nonempty_outer_keys(d::AbstractDoubleDict)\n\nReturn a vector of outer keys of the AbstractDoubleDict d.\n\nOnly outer keys that have a nonempty set of inner keys will be returned.\n\nEach outer key is a Tuple{Type,Type} so that a double loop can be easily used\n\nfor (F, S) in DoubleDicts.nonempty_outer_keys(dict)\n    for (k, v) in dict[F, S]\n        # ...\n    end\nend\nFor performance, it is recommended that the inner loop lies in a separate\nfunction to gurantee type-stability.\n\nIf you want an iterator of all current outer keys, use [`outer_keys`](@ref).\n\n\n\n\n\n","category":"function"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"EditURL = \"https://github.com/jump-dev/Clp.jl/blob/v1.0.3/README.md\"","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"(Image: )","category":"page"},{"location":"packages/Clp/#Clp.jl","page":"jump-dev/Clp.jl","title":"Clp.jl","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Clp.jl is a wrapper for the COIN-OR Linear Programming solver.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Clp/#Affiliation","page":"jump-dev/Clp.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Clp/#License","page":"jump-dev/Clp.jl","title":"License","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Clp.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The underlying solver, coin-or/Clp, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Clp/#Installation","page":"jump-dev/Clp.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Install Clp using Pkg.add:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"import Pkg\nPkg.add(\"Clp\")","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"In addition to installing the Clp.jl package, this will also download and install the Clp binaries. You do not need to install Clp separately.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Clp/#Use-with-JuMP","page":"jump-dev/Clp.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"To use Clp with JuMP, use Clp.Optimizer:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"using JuMP, Clp\nmodel = Model(Clp.Optimizer)\nset_attribute(model, \"LogLevel\", 1)\nset_attribute(model, \"Algorithm\", 4)","category":"page"},{"location":"packages/Clp/#MathOptInterface-API","page":"jump-dev/Clp.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The Clp optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/Clp/#Options","page":"jump-dev/Clp.jl","title":"Options","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Options are, unfortunately, not well documented.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The following options are likely to be the most useful:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Parameter Example Explanation\nPrimalTolerance 1e-7 Primal feasibility tolerance\nDualTolerance 1e-7 Dual feasibility tolerance\nDualObjectiveLimit 1e308 When using dual simplex (where the objective is monotonically changing), terminate when the objective exceeds this limit\nMaximumIterations 2147483647 Terminate after performing this number of simplex iterations\nMaximumSeconds -1.0 Terminate after this many seconds have passed. A negative value means no time limit\nLogLevel 1 Set to 1, 2, 3, or 4 for increasing output. Set to 0 to disable output\nPresolveType 0 Set to 1 to disable presolve\nSolveType 5 Solution method: dual simplex (0), primal simplex (1), sprint (2), barrier with crossover (3), barrier without crossover (4), automatic (5)\nInfeasibleReturn 0 Set to 1 to return as soon as the problem is found to be infeasible (by default, an infeasibility proof is computed as well)\nScaling 3 0 -off, 1 equilibrium, 2 geometric, 3 auto, 4 dynamic(later)\nPerturbation 100 switch on perturbation (50), automatic (100), don't try perturbing (102)","category":"page"},{"location":"packages/Clp/#C-API","page":"jump-dev/Clp.jl","title":"C API","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The C API can be accessed via Clp.Clp_XXX functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/index.md\"","category":"page"},{"location":"moi/#moi_documentation","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"warning: Warning\nThis documentation in this section is a copy of the official MathOptInterface documentation available at https://jump.dev/MathOptInterface.jl/v1.20.1. It is included here to make it easier to link concepts between JuMP and MathOptInterface.","category":"page"},{"location":"moi/#What-is-MathOptInterface?","page":"Introduction","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"MathOptInterface.jl (MOI) is an abstraction layer designed to provide a unified interface to mathematical optimization solvers so that users do not need to understand multiple solver-specific APIs.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"tip: Tip\nThis documentation is aimed at developers writing software interfaces to solvers and modeling languages using the MathOptInterface API. If you are a user interested in solving optimization problems, we encourage you instead to use MOI through a higher-level modeling interface like JuMP or Convex.jl.","category":"page"},{"location":"moi/#How-the-documentation-is-structured","page":"Introduction","title":"How the documentation is structured","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"The Tutorials section contains articles on how to use and implement the MathOptInteraface API. Look here if you want to write a model in MOI, or write an interface to a new solver.\nThe Manual contains short code-snippets that explain how to use the MOI API. Look here for more details on particular areas of MOI.\nThe Background section contains articles on the theory behind MathOptInterface. Look here if you want to understand why, rather than how.\nThe API Reference contains a complete list of functions and types that comprise the MOI API. Look here is you want to know how to use (or implement) a particular function.\nThe Submodules section contains stand-alone documentation for each of the submodules within MOI. These submodules are not required to interface a solver with MOI, but they make the job much easier.","category":"page"},{"location":"moi/#Citing-MathOptInterface","page":"Introduction","title":"Citing MathOptInterface","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"A paper describing the design and features of MathOptInterface is available on arXiv.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"If you find MathOptInterface useful in your work, we kindly request that you cite the following paper:","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"@article{legat2021mathoptinterface,\n    title={{MathOptInterface}: a data structure for mathematical optimization problems},\n    author={Legat, Beno{\\^\\i}t and Dowson, Oscar and Garcia, Joaquim Dias and Lubin, Miles},\n    journal={INFORMS Journal on Computing},\n    year={2021},\n    doi={10.1287/ijoc.2021.1067},\n    publisher={INFORMS}\n}","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"EditURL = \"https://github.com/osqp/OSQP.jl/blob/443706e34c2619acbe65281c60bbe850ca4a8fac/README.md\"","category":"page"},{"location":"packages/OSQP/#OSQP.jl","page":"osqp/OSQP.jl","title":"OSQP.jl","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"(Image: Build Status)  (Image: codecov.io)","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"OSQP.jl is a Julia wrapper for OSQP: the Operator Splitting QP Solver.","category":"page"},{"location":"packages/OSQP/#License","page":"osqp/OSQP.jl","title":"License","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"OSQP.jl is licensed under the Apache-2.0 license.","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"The upstream solver, osqp/osqp is also licensed under the Apache-2.0 license.","category":"page"},{"location":"packages/OSQP/#Installation","page":"osqp/OSQP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"Install OSQP.jl using the Julia package manager","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"import Pkg\nPkg.add(\"OSQP\")","category":"page"},{"location":"packages/OSQP/#Problem-class","page":"osqp/OSQP.jl","title":"Problem class","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"The OSQP (Operator Splitting Quadratic Program) solver is a numerical optimization package for solving problems in the form","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"minimize        0.5 x' P x + q' x\n\nsubject to      l <= A x <= u","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"where x in R^n is the optimization variable. The objective function is defined by a positive semidefinite matrix P in S^n_+ and vector q in R^n. The linear constraints are defined by matrix A in R^{m x n} and vectors l in R^m U {-inf}^m, u in R^m U {+inf}^m.","category":"page"},{"location":"packages/OSQP/#Documentation","page":"osqp/OSQP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"Detailed documentation is available at https://osqp.org/.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS, SCS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/models/#jump_models","page":"Models","title":"Models","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP models are the fundamental building block that we use to construct optimization problems. They hold things like the variables and constraints, as well as which solver to use and even solution information.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nJuMP uses \"optimizer\" as a synonym for \"solver.\" Our convention is to use \"solver\" to refer to the underlying software, and use \"optimizer\" to refer to the Julia object that wraps the solver. For example, HiGHS is a solver, and HiGHS.Optimizer is an optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nSee Supported solvers for a list of available solvers.","category":"page"},{"location":"manual/models/#Create-a-model","page":"Models","title":"Create a model","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Create a model by passing an optimizer to Model:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If you don't know which optimizer you will be using at creation time, create a model without an optimizer, and then call set_optimizer at any time prior to optimize!:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> set_optimizer(model, HiGHS.Optimizer)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nDon't know what the fields Model mode and CachingOptimizer state mean? Read the Backends section.","category":"page"},{"location":"manual/models/#What-is-the-difference?","page":"Models","title":"What is the difference?","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"For most models, there is no difference between passing the optimizer to Model, and calling set_optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if an optimizer does not support a constraint in the model, the timing of when an error will be thrown can differ:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If you pass an optimizer, an error will be thrown when you try to add the constraint.\nIf you call set_optimizer, an error will be thrown when you try to solve the model via optimize!.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Therefore, most users should pass an optimizer to Model because it provides the earliest warning that your solver is not suitable for the model you are trying to build. However, if you are modifying a problem by adding and deleting different constraint types, you may need to use set_optimizer. See Switching optimizer for the relaxed problem for an example of when this is useful.","category":"page"},{"location":"manual/models/#Reducing-time-to-first-solve-latency","page":"Models","title":"Reducing time-to-first-solve latency","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, JuMP uses bridges to reformulate the model you are building into an equivalent model supported by the solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if your model is already supported by the solver, bridges add latency (read The \"time-to-first-solve\" issue). This is particularly noticeable for small models.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"To reduce the \"time-to-first-solve,s\" try passing add_bridges = false.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer; add_bridges = false);","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"or","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> set_optimizer(model, HiGHS.Optimizer; add_bridges = false)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, be wary. If your model and solver combination needs bridges, an error will be thrown:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(SCS.Optimizer; add_bridges = false);\n\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, 2x <= 1)\nERROR: Constraints of type MathOptInterface.ScalarAffineFunction{Float64}-in-MathOptInterface.LessThan{Float64} are not supported by the solver.\n\nIf you expected the solver to support your problem, you may have an error in your formulation. Otherwise, consider using a different solver.\n\nThe list of available solvers, along with the problem types they support, is available at https://jump.dev/JuMP.jl/stable/installation/#Supported-solvers.\n[...]","category":"page"},{"location":"manual/models/#Solvers-which-expect-environments","page":"Models","title":"Solvers which expect environments","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Some solvers accept (or require) positional arguments such as a license environment or a path to a binary executable. For these solvers, you can pass a function to Model which takes zero arguments and returns an instance of the optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common use-case for this is passing an environment or sub-solver to the optimizer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> import HiGHS\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer))\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: MOA[algorithm=MultiObjectiveAlgorithms.Lexicographic, optimizer=HiGHS]","category":"page"},{"location":"manual/models/#solver_options","page":"Models","title":"Solver options","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP uses \"attribute\" as a synonym for \"option.\" Use optimizer_with_attributes to create an optimizer with some attributes initialized:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(\n           optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false),\n       )\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Alternatively, use set_attribute to set an attribute after the model has been created:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_attribute(model, \"output_flag\", false)\n\njulia> get_attribute(model, \"output_flag\")\nfalse","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"You can also modify attributes within an optimizer_with_attributes object:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> solver = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => true);\n\njulia> get_attribute(solver, \"output_flag\")\ntrue\n\njulia> set_attribute(solver, \"output_flag\", false)\n\njulia> get_attribute(solver, \"output_flag\")\nfalse\n\njulia> model = Model(solver);","category":"page"},{"location":"manual/models/#Changing-the-number-types","page":"Models","title":"Changing the number types","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, the coefficients of affine and quadratic expressions are numbers of type either Float64 or Complex{Float64} (see Complex number support).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The type Float64 can be changed using the GenericModel constructor:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = GenericModel{Rational{BigInt}}();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, expr, 1 // 3 * x)\n1//3 x\n\njulia> typeof(expr)\nGenericAffExpr{Rational{BigInt}, GenericVariableRef{Rational{BigInt}}}","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Using a value_type other than Float64 is an advanced operation and should be used only if the underlying solver actually solves the problem using the provided value type.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nNonlinear Modeling is currently restricted to the Float64 number type.","category":"page"},{"location":"manual/models/#Print-the-model","page":"Models","title":"Print the model","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, show(model) will print a summary of the problem:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(); @variable(model, x >= 0); @objective(model, Max, x);\n\njulia> model\nA JuMP Model\nMaximization problem with:\nVariable: 1\nObjective function type: VariableRef\n`VariableRef`-in-`MathOptInterface.GreaterThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use print to print the formulation of the model (in IJulia, this will render as LaTeX.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> print(model)\nMax x\nSubject to\n x ≥ 0","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nThis format is specific to JuMP and may change in any future release. It is not intended to be an instance format. To write the model to a file, use write_to_file instead.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use latex_formulation to display the model in LaTeX form.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> latex_formulation(model)\n$$ \\begin{aligned}\n\\max\\quad & x\\\\\n\\text{Subject to} \\quad & x \\geq 0\\\\\n\\end{aligned} $$","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"In IJulia (and Documenter), ending a cell in with latex_formulation will render the model in LaTeX:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP                # hide\nmodel = Model()           # hide\n@variable(model, x >= 0)  # hide\n@objective(model, Max, x) # hide\nlatex_formulation(model)","category":"page"},{"location":"manual/models/#Turn-off-output","page":"Models","title":"Turn off output","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use set_silent and unset_silent to disable or enable printing output from the solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> unset_silent(model)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nMost solvers will also have a solver-specific option to provide finer-grained control over the output. Consult their README's for details.","category":"page"},{"location":"manual/models/#Set-a-time-limit","page":"Models","title":"Set a time limit","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use set_time_limit_sec, unset_time_limit_sec, and time_limit_sec to manage time limits.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_time_limit_sec(model, 60.0)\n\n\njulia> time_limit_sec(model)\n60.0\n\njulia> unset_time_limit_sec(model)\n\njulia> limit = time_limit_sec(model)\n\njulia> limit === nothing\ntrue","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If your time limit is encoded as a Dates.Period object, use the following code to convert it to Float64 for set_time_limit_sec:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> import Dates\n\njulia> seconds(x::Dates.Period) = 1e-3 * Dates.value(round(x, Dates.Millisecond))\nseconds (generic function with 1 method)\n\njulia> set_time_limit_sec(model, seconds(Dates.Hour(1)))\n\njulia> time_limit_sec(model)\n3600.0","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nSome solvers do not support time limits. In these cases, an error will be thrown.","category":"page"},{"location":"manual/models/#Write-a-model-to-file","page":"Models","title":"Write a model to file","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP can write models to a variety of file-formats using write_to_file and Base.write.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For most common file formats, the file type will be detected from the extension.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For example, here is how to write an MPS file:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> write_to_file(model, \"model.mps\")","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Other supported file formats include:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":".cbf for the Conic Benchmark Format\n.lp for the LP file format\n.mof.json for the MathOptFormat\n.nl for AMPL's NL file format\n.rew for the REW file format\n.sdpa and \".dat-s\" for the SDPA file format","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"To write to a specific io::IO, use Base.write. Specify the file type by passing a MOI.FileFormats.FileFormat enum.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> io = IOBuffer();\n\njulia> write(io, model; format = MOI.FileFormats.FORMAT_MPS)","category":"page"},{"location":"manual/models/#Read-a-model-from-file","page":"Models","title":"Read a model from file","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP models can be created from file formats using read_from_file and Base.read.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = read_from_file(\"model.mps\")\nA JuMP Model\nMinimization problem with:\nVariables: 0\nObjective function type: AffExpr\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> seekstart(io);\n\njulia> model2 = read(io, Model; format = MOI.FileFormats.FORMAT_MPS)\nA JuMP Model\nMinimization problem with:\nVariables: 0\nObjective function type: AffExpr\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"note: Note\nBecause file formats do not serialize the containers of JuMP variables and constraints, the names in the model will not be registered. Therefore, you cannot access named variables and constraints via model[:x]. Instead, use variable_by_name or constraint_by_name to access specific variables or constraints.","category":"page"},{"location":"manual/models/#Relax-integrality","page":"Models","title":"Relax integrality","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use relax_integrality to remove any integrality constraints from the model, such as integer and binary restrictions on variables. relax_integrality returns a function that can be later called with zero arguments to re-add the removed constraints:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> @variable(model, x, Int)\nx\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n1\n\njulia> undo = relax_integrality(model);\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n0\n\njulia> undo()\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n1","category":"page"},{"location":"manual/models/#Switching-optimizer-for-the-relaxed-problem","page":"Models","title":"Switching optimizer for the relaxed problem","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common reason for relaxing integrality is to compute dual variables of the relaxed problem. However, some mixed-integer linear solvers (for example, Cbc) do not return dual solutions, even if the problem does not have integrality restrictions.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Therefore, after relax_integrality you should call set_optimizer with a solver that does support dual solutions, such as Clp.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For example, instead of:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP, Cbc\nmodel = Model(Cbc.Optimizer)\n@variable(model, x, Int)\nundo = relax_integrality(model)\noptimize!(model)\nreduced_cost(x)  # Errors","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"do:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP, Cbc, Clp\nmodel = Model(Cbc.Optimizer)\n@variable(model, x, Int)\nundo = relax_integrality(model)\nset_optimizer(model, Clp.Optimizer)\noptimize!(model)\nreduced_cost(x)  # Works","category":"page"},{"location":"manual/models/#Backends","page":"Models","title":"Backends","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nThis section discusses advanced features of JuMP. For new users, you may want to skip this section. You don't need to know how JuMP manages problems behind the scenes to create and solve JuMP models.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A JuMP Model is a thin layer around a backend of type MOI.ModelLike that stores the optimization problem and acts as the optimization solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if you construct a model like Model(HiGHS.Optimizer), the backend is not a HiGHS.Optimizer, but a more complicated object.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"From JuMP, the MOI backend can be accessed using the backend function. Let's see what the backend of a JuMP Model is:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> b = backend(model)\nMOIU.CachingOptimizer{MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\n  with 0 variable bridges\n  with 0 constraint bridges\n  with 0 objective bridges\n  with inner model A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Uh oh. Even though we passed a HiGHS.Optimizer, the backend is a much more complicated object.","category":"page"},{"location":"manual/models/#CachingOptimizer","page":"Models","title":"CachingOptimizer","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"A MOIU.CachingOptimizer is a layer that abstracts the difference between solvers that support incremental modification (for example, they support adding variables one-by-one), and solvers that require the entire problem in a single API call (for example, they only accept the A, b and c matrices of a linear program).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"It has two parts:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A cache, where the model can be built and modified incrementally\njulia> b.model_cache\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\nAn optimizer, which is used to solve the problem\njulia> b.optimizer\nMOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nThe LazyBridgeOptimizer section explains what a LazyBridgeOptimizer is.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The CachingOptimizer has logic to decide when to copy the problem from the cache to the optimizer, and when it can efficiently update the optimizer in-place.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A CachingOptimizer may be in one of three possible states:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"NO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer, and it is not synchronized with the cached model.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A CachingOptimizer has two modes of operation:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"AUTOMATIC: The CachingOptimizer changes its state when necessary. For example, optimize! will automatically call attach_optimizer (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\nMANUAL: The user must change the state of the CachingOptimizer using MOIU.reset_optimizer(::JuMP.Model), MOIU.drop_optimizer(::JuMP.Model), and MOIU.attach_optimizer(::JuMP.Model). Attempting to perform an operation in the incorrect state results in an error.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default Model will create a CachingOptimizer in AUTOMATIC mode.","category":"page"},{"location":"manual/models/#LazyBridgeOptimizer","page":"Models","title":"LazyBridgeOptimizer","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"The second layer that JuMP applies automatically is a MOI.Bridges.LazyBridgeOptimizer. A MOI.Bridges.LazyBridgeOptimizer is an MOI layer that attempts to transform the problem from the formulation provided by the user into an equivalent problem supported by the solver. This may involve adding new variables and constraints to the optimizer. The transformations are selected from a set of known recipes called bridges.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common example of a bridge is one that splits an interval constraint like @constraint(model, 1 <= x + y <= 2) into two constraints, @constraint(model, x + y >= 1) and @constraint(model, x + y <= 2).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use the add_bridges = false keyword to remove the bridging layer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer; add_bridges = false)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS\n\njulia> backend(model)\nMOIU.CachingOptimizer{HiGHS.Optimizer, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Bridges can be added and removed from a MOI.Bridges.LazyBridgeOptimizer using add_bridge and remove_bridge. Use print_active_bridges to see which bridges are used to reformulate the model. Read the Ellipsoid approximation tutorial for more details.","category":"page"},{"location":"manual/models/#Unsafe-backend","page":"Models","title":"Unsafe backend","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"In some advanced use-cases, it is necessary to work with the inner optimization model directly. To access this model, use unsafe_backend:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> backend(model)\nMOIU.CachingOptimizer{MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\n  with 0 variable bridges\n  with 0 constraint bridges\n  with 0 objective bridges\n  with inner model A HiGHS model with 0 columns and 0 rows.\n\njulia> unsafe_backend(model)\nA HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nbackend and unsafe_backend are advanced routines. Read their docstrings to understand the caveats of their usage, and only call them if you wish to access low-level solver-specific functions.","category":"page"},{"location":"manual/models/#Direct-mode","page":"Models","title":"Direct mode","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Using a CachingOptimizer results in an additional copy of the model being stored by JuMP in the .model_cache field. To avoid this overhead, create a JuMP model using direct_model:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = direct_model(HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nSolvers that do not support incremental modification do not support direct_model. An error will be thrown, telling you to use a CachingOptimizer instead.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The benefit of using direct_model is that there are no extra layers (for example, Cachingoptimizer or LazyBridgeOptimizer) between model and the provided optimizer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> backend(model)\nA HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A downside of direct mode is that there is no bridging layer. Therefore, only constraints which are natively supported by the solver are supported. For example, HiGHS.jl does not implement quadratic constraints:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x[1]^2 + x[2]^2 <= 2)\nERROR: Constraints of type MathOptInterface.ScalarQuadraticFunction{Float64}-in-MathOptInterface.LessThan{Float64} are not supported by the solver.\n\nIf you expected the solver to support your problem, you may have an error in your formulation. Otherwise, consider using a different solver.\n\nThe list of available solvers, along with the problem types they support, is available at https://jump.dev/JuMP.jl/stable/installation/#Supported-solvers.\nStacktrace:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nAnother downside of direct mode is that the behavior of querying solution information after modifying the problem is solver-specific. This can lead to errors, or the solver silently returning an incorrect value. See OptimizeNotCalled errors for more information.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/duality.md\"","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/duality/#Duality","page":"Duality","title":"Duality","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Conic duality is the starting point for MOI's duality conventions. When all functions are affine (or coordinate projections), and all constraint sets are closed convex cones, the model may be called a conic optimization problem.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For a minimization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_x in mathbbR^n  a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and the dual is a maximization problem in standard conic form:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_y_1 ldots y_m  -sum_i=1^m b_i^T y_i + b_0\n\n textst  a_0 - sum_i=1^m A_i^T y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For a maximization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_x in mathbbR^n  a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and the dual is a minimization problem in standard conic form:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_y_1 ldots y_m  sum_i=1^m b_i^T y_i + b_0\n\n textst  a_0 + sum_i=1^m A_i^T y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"A linear inequality constraint a^T x + b ge c is equivalent to a^T x + b - c in mathbbR_+, and a^T x + b le c is equivalent to a^T x + b - c in mathbbR_-. Variable-wise constraints are affine constraints with the appropriate identity mapping in place of A_i.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For the special case of minimization LPs, the MOI primal form can be stated as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_x in mathbbR^n  a_0^T x + b_0\n\n textst\nA_1 x  ge b_1\n A_2 x  le b_2\n A_3 x  = b_3\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"By applying the stated transformations to conic form, taking the dual, and transforming back into linear inequality form, one obtains the following dual:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_y_1y_2y_3  b_1^Ty_1 + b_2^Ty_2 + b_3^Ty_3 + b_0\n\n textst\nA_1^Ty_1 + A_2^Ty_2 + A_3^Ty_3  = a_0\n y_1 ge 0\n y_2 le 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For maximization LPs, the MOI primal form can be stated as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_x in mathbbR^n  a_0^T x + b_0\n\n textst\nA_1 x  ge b_1\n A_2 x  le b_2\n A_3 x  = b_3\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and similarly, the dual is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_y_1y_2y_3  -b_1^Ty_1 - b_2^Ty_2 - b_3^Ty_3 + b_0\n\n textst\nA_1^Ty_1 + A_2^Ty_2 + A_3^Ty_3  = -a_0\n y_1 ge 0\n y_2 le 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"warning: Warning\nFor the LP case, the signs of the feasible dual variables depend only on the sense of the corresponding primal inequality and not on the objective sense.","category":"page"},{"location":"moi/background/duality/#Duality-and-scalar-product","page":"Duality","title":"Duality and scalar product","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The scalar product is different from the canonical one for the sets PositiveSemidefiniteConeTriangle, LogDetConeTriangle, RootDetConeTriangle.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If the set C_i of the section Duality is one of these three cones, then the rows of the matrix A_i corresponding to off-diagonal entries are twice the value of the coefficients field in the VectorAffineFunction for the corresponding rows. See PositiveSemidefiniteConeTriangle for details.","category":"page"},{"location":"moi/background/duality/#Dual-for-problems-with-quadratic-functions","page":"Duality","title":"Dual for problems with quadratic functions","text":"","category":"section"},{"location":"moi/background/duality/#Quadratic-Programs-(QPs)","page":"Duality","title":"Quadratic Programs (QPs)","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For quadratic programs with only affine conic constraints,","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign*\n min_x in mathbbR^n  frac12x^TQ_0x + a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign*","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with cones mathcalC_i subseteq mathbbR^m_i for i = 1 ldots m, consider the Lagrangian function","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x + a_0^T x + b_0 - sum_i = 1^m y_i^T (A_i x + b_i)","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let z(y) denote sum_i = 1^m A_i^T y_i - a_0, the Lagrangian can be rewritten as","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x - z(y)^T x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The condition nabla_x L(x y) = 0 gives","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"0 = nabla_x L(x y) = Q_0x + a_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"which gives Q_0x = z(y). This allows to obtain that","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"min_x in mathbbR^n L(x y) = -frac12x^TQ_0x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so the dual problem is","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"max_y_i in mathcalC_i^* min_x in mathbbR^n -frac12x^TQ_0x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If Q_0 is invertible, we have x = Q_0^-1z(y) hence","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"min_x in mathbbR^n L(x y) = -frac12z(y)^TQ_0^-1z(y) + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so the dual problem is","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"max_y_i in mathcalC_i^* -frac12z(y)^TQ_0^-1z(y) + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/#Quadratically-Constrained-Quadratic-Programs-(QCQPs)","page":"Duality","title":"Quadratically Constrained Quadratic Programs (QCQPs)","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Given a problem with both quadratic function and quadratic objectives:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign*\n min_x in mathbbR^n  frac12x^TQ_0x + a_0^T x + b_0\n\n textst  frac12x^TQ_ix + a_i^T x + b_i  in mathcalC_i  i = 1 ldots m\nendalign*","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with cones mathcalC_i subseteq mathbbR for i = 1 ldots m, consider the Lagrangian function","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x + a_0^T x + b_0 - sum_i = 1^m y_i (frac12x^TQ_ix + a_i^T x + b_i)","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"A pair of primal-dual variables (x^star y^star) is optimal if","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"x^star is a minimizer of\nmin_x in mathbbR^n L(x y^star)\nThat is,\n0 = nabla_x L(x y^star) = Q_0x + a_0 - sum_i = 1^m y_i^star (Q_ix + a_i)\nand y^star is a maximizer of\nmax_y_i in mathcalC_i^* L(x^star y)\nThat is, for all i = 1 ldots m, frac12x^TQ_ix + a_i^T x + b_i is either zero or in the normal cone of mathcalC_i^* at y^star. For instance, if mathcalC_i is  z in mathbbR  z le 0 , this means that if frac12x^TQ_ix + a_i^T x + b_i is nonzero at x^star then y_i^star = 0. This is the classical complementary slackness condition.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If mathcalC_i is a vector set, the discussion remains valid with y_i(frac12x^TQ_ix + a_i^T x + b_i) replaced with the scalar product between y_i and the vector of scalar-valued quadratic functions.","category":"page"},{"location":"moi/background/duality/#Dual-for-square-semidefinite-matrices","page":"Duality","title":"Dual for square semidefinite matrices","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The set PositiveSemidefiniteConeTriangle is a self-dual. That is, querying ConstraintDual of a PositiveSemidefiniteConeTriangle constraint returns a vector that is itself a member of PositiveSemidefiniteConeTriangle.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"However, the dual of PositiveSemidefiniteConeSquare is not so straight forward. This section explains the duality convention we use, and how it is derived.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"info: Info\nIf you have a PositiveSemidefiniteConeSquare constraint, the result matrix A from ConstraintDual is not positive semidefinite. However, A + A^top is positive semidefinite.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let mathcalS_+ be the cone of symmetric semidefinite matrices in the fracn(n+1)2 dimensional space of symmetric mathbbR^n times n matrices. That is, mathcalS_+ is the set PositiveSemidefiniteConeTriangle. It is well known that mathcalS_+ is a self-dual proper cone.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let mathcalP_+ be the cone of symmetric semidefinite matrices in the n^2 dimensional space of mathbbR^n times n matrices. That is mathcalP_+ is the set PositiveSemidefiniteConeSquare.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"In addition, let mathcalD_+ be the cone of matrices A such that A+A^top in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"mathcalP_+ is not proper because it is not solid (it is not n^2 dimensional), so it is not necessarily true that mathcalP_+^** = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"However, this is the case, because we will show that mathcalP_+^* = mathcalD_+ and mathcalD_+^* = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"First, let us see why mathcalP_+^* = mathcalD_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If B is symmetric, then","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"langle AB rangle = langle A^top B^top rangle = langle A^top Brangle","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"2langle A B rangle = langle A B rangle + langle A^top B rangle = langle A + A^top  B rangle","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Therefore, langle ABrangle ge 0 for all B in mathcalP_+ if and only if langle A+A^topBrangle ge 0 for all B in mathcalP_+. Since A+A^top is symmetric, and we know that mathcalS_+ is self-dual, we have shown that mathcalP_+^* is the set of matrices A such that A+A^top in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Second, let us see why mathcalD_+^* = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Since A in mathcalD_+ implies that A^top in mathcalD_+, B in mathcalD_+^* means that langle A+A^topBrangle ge 0 for all A in mathcalD_+, and hence B in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"To see why it should be symmetric, simply notice that if B_ij  B_ji, then langle ABrangle can be made arbitrarily small by setting A_ij = A_ij + s and A_ji = A_ji - s, with s arbitrarily large, and A stays in mathcalD_+ because A+A^top does not change.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Typically, the primal/dual pair for semidefinite programs is presented as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C X rangle \ntextst   langle A_k Xrangle = b_k forall k \n             X                     in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum_k b_k y_k \ntextst   C - sum A_k y_k in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If we allow A_k to be non-symmetric, we should instead use:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C X rangle \ntextst   langle A_k Xrangle = b_k forall k \n             X                     in mathcalD_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum b_k y_k \ntextst   C - sum A_k y_k in mathcalP_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"This is implemented as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C   Z rangle + langle C - C^top S rangle \ntextst   langle A_k Z rangle + langle A_k - A_k^top S rangle = b_k forall k \n             Z in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum b_k y_k \ntextst   C+C^top - sum (A_k+A_k^top) y_k in mathcalS_+ \n             C-C^top - sum(A_k-A_k^top) y_k = 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and we recover Z = X + X^top.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#conic_tips_and_tricks","page":"Tips and Tricks","title":"Tips and Tricks","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial is aimed at providing a simplistic introduction to conic programming using JuMP.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It uses the following packages:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"using JuMP\nimport SCS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"import Random      # hide\nRandom.seed!(1234) # hide\nnothing            # hide","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"tip: Tip\nA good resource for learning more about functions which can be modeled using cones is the MOSEK Modeling Cookbook.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Background-theory","page":"Tips and Tricks","title":"Background theory","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A subset C of a vector space V is a cone if forall x in C and positive scalars lambda  0, the product lambda x in C.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A cone C is a convex cone if lambda x + (1 - lambda) y in C, for any lambda in 0 1, and any x y in C.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Conic programming problems are convex optimization problems in which a convex function is minimized over the intersection of an affine subspace and a convex cone. An example of a conic-form minimization problems, in the primal form is:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n min_x in mathbbR^n  a_0^T x + b_0 \n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The corresponding dual problem is:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max_y_1 ldots y_m  -sum_i=1^m b_i^T y_i + b_0 \n textst  a_0 - sum_i=1^m A_i^T y_i  = 0 \n  y_i  in mathcalC_i^*  i = 1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Second-Order-Cone","page":"Tips and Tricks","title":"Second-Order Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The SecondOrderCone (or Lorentz Cone) of dimension n is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_soc =  (t x) in mathbbR^n  t ge x_2 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It is most commonly used to represent the L2-norm of the vector x:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, sum(x) == 1)\n@constraint(model, [t; x] in SecondOrderCone())\n@objective(model, Min, t)\noptimize!(model)\nvalue(t), value.(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Rotated-Second-Order-Cone","page":"Tips and Tricks","title":"Rotated Second-Order Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A Second-Order Cone rotated by pi4 in the (x_1x_2) plane is called a RotatedSecondOrderCone. It is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_rsoc =  (tux) in mathbbR^n  2tu ge x_2^2 tu ge 0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When u = 0.5, it represents the sum of squares of a vector x:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"data = [1.0, 2.0, 3.0, 4.0]\ntarget = [0.45, 1.04, 1.51, 1.97]\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, θ)\n@variable(model, t)\n@expression(model, residuals, θ * data .- target)\n@constraint(model, [t; 0.5; residuals] in RotatedSecondOrderCone())\n@objective(model, Min, t)\noptimize!(model)\nvalue(θ), value(t)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Exponential-Cone","page":"Tips and Tricks","title":"Exponential Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.ExponentialCone is a set of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_exp =  (xyz) in mathbbR^3  y exp (xy) le z y  0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It can be used to model problems involving log and exp.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Exponential","page":"Tips and Tricks","title":"Exponential","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model exp(x) le z, use (x, 1, z):","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x == 1.5)\n@variable(model, z)\n@objective(model, Min, z)\n@constraint(model, [x, 1, z] in MOI.ExponentialCone())\noptimize!(model)\nvalue(z), exp(1.5)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Logarithm","page":"Tips and Tricks","title":"Logarithm","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model x le log(z), use (x, 1, z):","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x)\n@variable(model, z == 1.5)\n@objective(model, Max, x)\n@constraint(model, [x, 1, z] in MOI.ExponentialCone())\noptimize!(model)\nvalue(x), log(1.5)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Log-sum-exp","page":"Tips and Tricks","title":"Log-sum-exp","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model t ge logleft(sum e^x_iright), use:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"N = 3\nx0 = rand(N)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:N] == x0[i])\n@variable(model, t)\n@objective(model, Min, t)\n@variable(model, u[1:N])\n@constraint(model, sum(u) <= 1)\n@constraint(model, [i = 1:N], [x[i] - t, 1, u[i]] in MOI.ExponentialCone())\noptimize!(model)\nvalue(t), log(sum(exp.(x0)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Entropy","page":"Tips and Tricks","title":"Entropy","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The entropy maximization problem consists of maximizing the entropy function, H(x) = -xlogx subject to linear inequality constraints.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max  - sum_i=1^n x_i log x_i \n textst  mathbf1^top x = 1 \n  Ax leq b\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"We can model this problem using an exponential cone by using the following transformation:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"tleq -xlogx iff tleq xlog(1x)  iff (t x 1) in K_exp","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Thus, our problem becomes,","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max  1^Tt \n textst  Ax leq b \n  1^T x = 1 \n  (t_i x_i 1) in K_exp  forall i = 1 ldots n \nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"m, n = 10, 15\nA, b = randn(m, n), rand(m, 1)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t[1:n])\n@variable(model, x[1:n])\n@objective(model, Max, sum(t))\n@constraint(model, sum(x) == 1)\n@constraint(model, A * x .<= b)\n@constraint(model, [i = 1:n], [t[i], x[i], 1] in MOI.ExponentialCone())\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.ExponentialCone has a dual, the MOI.DualExponentialCone, that offers an alternative formulation that can be more efficient for some formulations.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"There is also the MOI.RelativeEntropyCone for explicitly encoding the relative entropy function","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, x[1:n])\n@objective(model, Max, -t)\n@constraint(model, sum(x) == 1)\n@constraint(model, A * x .<= b)\n@constraint(model, [t; ones(n); x] in MOI.RelativeEntropyCone(2n + 1))\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#PowerCone","page":"Tips and Tricks","title":"PowerCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.PowerCone is a three-dimensional set parameterized by a scalar value α. It has the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_p =  (xyz) in mathbbR^3  x^alpha y^1-alpha ge z x ge 0 y ge 0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The power cone permits a number of reformulations. For example, when p  1, we can model t ge x^p using the power cone (t 1 x) with alpha = 1  p. Thus, to model t ge x^3 with x ge 0","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, x >= 1.5)\n@constraint(model, [t, 1, x] in MOI.PowerCone(1 / 3))\n@objective(model, Min, t)\noptimize!(model)\nvalue(t), value(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.PowerCone has a dual, the MOI.DualPowerCone, that offers an alternative formulation that can be more efficient for some formulations.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#P-Norm","page":"Tips and Tricks","title":"P-Norm","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The p-norm x_p = left(sumlimits_i x_i^pright)^frac1p can be modeled using MOI.PowerCones. See the Mosek Modeling Cookbook for the derivation.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"function p_norm(x::Vector, p)\n    N = length(x)\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, r[1:N])\n    @variable(model, t)\n    @constraint(model, [i = 1:N], [r[i], t, x[i]] in MOI.PowerCone(1 / p))\n    @constraint(model, sum(r) == t)\n    @objective(model, Min, t)\n    optimize!(model)\n    return value(t)\nend\n\nx = rand(5);\nLinearAlgebra.norm(x, 4), p_norm(x, 4)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Positive-Semidefinite-Cone","page":"Tips and Tricks","title":"Positive Semidefinite Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The set of positive semidefinite matrices (PSD) of dimension n form a cone in mathbbR^n. We write this set mathematically as:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"mathcalS_+^n =  X in mathcalS^n mid z^T X z geq 0  forall zin mathbbR^n ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A PSD cone is represented in JuMP using the MOI sets PositiveSemidefiniteConeTriangle (for upper triangle of a PSD matrix) and PositiveSemidefiniteConeSquare (for a complete PSD matrix). However, it is preferable to use the PSDCone shortcut as illustrated below.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Example:-largest-eigenvalue-of-a-symmetric-matrix","page":"Tips and Tricks","title":"Example: largest eigenvalue of a symmetric matrix","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Suppose A has eigenvalues lambda_1 geq lambda_2 ldots geq lambda_n. Then the matrix t I-A has eigenvalues t-lambda_1 t-lambda_2 ldots t-lambda_n. Note that t I-A is PSD exactly when all these eigenvalues are non-negative, and this happens for values t geq lambda_1. Thus, we can model the problem of finding the largest eigenvalue of a symmetric matrix as:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\nlambda_1 = min t \ntext  st  t I-A succeq 0\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A = [3 2 4; 2 0 2; 4 2 3]\nI = Matrix{Float64}(LinearAlgebra.I, 3, 3)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@objective(model, Min, t)\n@constraint(model, t .* I - A in PSDCone())\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#GeometricMeanCone","page":"Tips and Tricks","title":"GeometricMeanCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.GeometricMeanCone is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_geo =  (t x) in mathbbR^n  x ge 0 t le sqrtn-1x_1 x_2 cdots x_n-1 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[1:4])\n@variable(model, t)\n@constraint(model, sum(x) == 1)\n@constraint(model, [t; x] in MOI.GeometricMeanCone(5))\noptimize!(model)\nvalue(t), value.(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#RootDetCone","page":"Tips and Tricks","title":"RootDetCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.RootDetConeSquare is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K =  (t X) in mathbbR^1+d^2  t le det(X)^frac1d ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, X[1:2, 1:2])\n@objective(model, Max, t)\n@constraint(model, [t; vec(X)] in MOI.RootDetConeSquare(2))\n@constraint(model, X .== [2 1; 1 3])\noptimize!(model)\nvalue(t), sqrt(LinearAlgebra.det(value.(X)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"If X is symmetric, then you can use MOI.RootDetConeTriangle instead. This can be more efficient because the solver does not need to add additional constraints to ensure X is symmetric.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When forming the function, use triangle_vec to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, X[1:2, 1:2], Symmetric)\n@objective(model, Max, t)\n@constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(2))\n@constraint(model, X .== [2 1; 1 3])\noptimize!(model)\nvalue(t), sqrt(LinearAlgebra.det(value.(X)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#LogDetCone","page":"Tips and Tricks","title":"LogDetCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.LogDetConeSquare is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K =  (t u X) in mathbbR^2+d^2  t le u log(det(X  u)) ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, u)\n@variable(model, X[1:2, 1:2])\n@objective(model, Max, t)\n@constraint(model, [t; u; vec(X)] in MOI.LogDetConeSquare(2))\n@constraint(model, X .== [2 1; 1 3])\n@constraint(model, u == 0.5)\noptimize!(model)\nvalue(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"If X is symmetric, then you can use MOI.LogDetConeTriangle instead. This can be more efficient because the solver does not need to add additional constraints to ensure X is symmetric.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When forming the function, use triangle_vec to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, u)\n@variable(model, X[1:2, 1:2], Symmetric)\n@objective(model, Max, t)\n@constraint(model, [t; u; triangle_vec(X)] in MOI.LogDetConeTriangle(2))\n@constraint(model, X .== [2 1; 1 3])\n@constraint(model, u == 0.5)\noptimize!(model)\nvalue(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Other-Cones-and-Functions","page":"Tips and Tricks","title":"Other Cones and Functions","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"For other cones supported by JuMP, check out the MathOptInterface Manual.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"EditURL = \"https://github.com/jump-dev/BARON.jl/blob/v0.8.2/README.md\"","category":"page"},{"location":"packages/BARON/#BARON.jl","page":"jump-dev/BARON.jl","title":"BARON.jl","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"BARON.jl is a wrapper for BARON by The Optimization Firm.","category":"page"},{"location":"packages/BARON/#Affiliation","page":"jump-dev/BARON.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"This wrapper is maintained by the JuMP community and is not officially supported by The Optimization Firm.","category":"page"},{"location":"packages/BARON/#License","page":"jump-dev/BARON.jl","title":"License","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"BARON.jl is licensed under the MIT License.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license from The Optimization Firm, although a small trial version is available for free.","category":"page"},{"location":"packages/BARON/#Installation","page":"jump-dev/BARON.jl","title":"Installation","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"First, download a copy of the BARON solver and unpack the executable in a location of your choosing.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"Once installed, set the BARON_EXEC environment variable pointing to the BARON executable (full path, including file name as it differs across platforms), and run Pkg.add(\"BARON\"). For example:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"ENV[\"BARON_EXEC\"] = \"/path/to/baron.exe\"\nusing Pkg\nPkg.add(\"BARON\")","category":"page"},{"location":"packages/BARON/#Use-with-JuMP","page":"jump-dev/BARON.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"using JuMP, BARON\nmodel = Model(BARON.Optimizer)","category":"page"},{"location":"packages/BARON/#MathOptInterface-API","page":"jump-dev/BARON.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"The BARON optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarNonlinearFunction}","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.Reals","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarNonlinearFunction in MOI.EqualTo{Float64}\nMOI.ScalarNonlinearFunction in MOI.GreaterThan{Float64}\nMOI.ScalarNonlinearFunction in MOI.Interval{Float64}\nMOI.ScalarNonlinearFunction in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.NLPBlock()\nMOI.ObjectiveSense()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"EditURL = \"getting_started_with_julia.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Getting-started-with-Julia","page":"Getting started with Julia","title":"Getting started with Julia","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because JuMP is embedded in Julia, knowing some basic Julia is important before you start learning JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nThis tutorial is designed to provide a minimalist crash course in the basics of Julia. You can find resources that provide a more comprehensive introduction to Julia here.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Installing-Julia","page":"Getting started with Julia","title":"Installing Julia","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"To install Julia, download the latest stable release, then follow the platform specific install instructions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nUnless you know otherwise, you probably want the 64-bit version.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Next, you need an IDE to develop in. VS Code is a popular choice, so follow these install instructions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia can also be used with Jupyter notebooks or the reactive notebooks of Pluto.jl.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#The-Julia-REPL","page":"Getting started with Julia","title":"The Julia REPL","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The main way of interacting with Julia is via its REPL (Read Evaluate Print Loop). To access the REPL, start the Julia executable to arrive at the julia> prompt, and then start coding:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"As your programs become larger, write a script as a text file, and then run that file using:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"julia> include(\"path/to/file.jl\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nBecause of Julia's startup latency, running scripts from the command line like the following is slow:$ julia path/to/file.jlUse the REPL or a notebook instead, and read The \"time-to-first-solve\" issue for more information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Code-blocks-in-this-documentation","page":"Getting started with Julia","title":"Code blocks in this documentation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In this documentation you'll see a mix of code examples with and without the julia>.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The Julia prompt is mostly used to demonstrate short code snippets, and the output is exactly what you will see if run from the REPL.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Blocks without the julia> can be copy-pasted into the REPL, but they are used because they enable richer output like plots or LaTeX to be displayed in the online and PDF versions of the documentation. If you run them from the REPL you may see different output.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Where-to-get-help","page":"Getting started with Julia","title":"Where to get help","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Read the documentation\nJuMP https://jump.dev/JuMP.jl/stable/\nJulia https://docs.julialang.org/en/v1/\nAsk (or browse) the Julia community forum: https://discourse.julialang.org\nIf the question is JuMP-related, ask in the Optimization (Mathematical) section, or tag your question with \"jump\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"To access the built-in help at the REPL, type ? to enter help-mode, followed by the name of the function to lookup:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"help?> print\nsearch: print println printstyled sprint isprint prevind parentindices precision escape_string\n\n  print([io::IO], xs...)\n\n  Write to io (or to the default output stream stdout if io is not given) a canonical\n  (un-decorated) text representation. The representation used by print includes minimal formatting\n  and tries to avoid Julia-specific details.\n\n  print falls back to calling show, so most types should just define show. Define print if your\n  type has a separate \"plain\" representation. For example, show displays strings with quotes, and\n  print displays strings without quotes.\n\n  string returns the output of print as a string.\n\n  Examples\n  ≡≡≡≡≡≡≡≡≡≡\n\n  julia> print(\"Hello World!\")\n  Hello World!\n  julia> io = IOBuffer();\n\n  julia> print(io, \"Hello\", ' ', :World!)\n\n  julia> String(take!(io))\n  \"Hello World!\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Numbers-and-arithmetic","page":"Getting started with Julia","title":"Numbers and arithmetic","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Since we want to solve optimization problems, we're going to be using a lot of math. Luckily, Julia is great for math, with all the usual operators:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1\n1 - 2\n2 * 2\n4 / 5\n3^2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Did you notice how Julia didn't print .0 after some of the numbers? Julia is a dynamic language, which means you never have to explicitly declare the type of a variable. However, in the background, Julia is giving each variable a type. Check the type of something using the typeof function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(1)\ntypeof(1.0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Here 1 is an Int64, which is an integer with 64 bits of precision, and 1.0 is a Float64, which is a floating point number with 64-bits of precision.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nIf you aren't familiar with floating point numbers, make sure to read the Floating point numbers section.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We create complex numbers using im:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = 2 + 1im\nreal(x)\nimag(x)\ntypeof(x)\nx * (1 - 2im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nThe curly brackets surround what we call the parameters of a type. You can read Complex{Int64}  as \"a complex number, where the real and imaginary parts are represented by Int64.\" If we call typeof(1.0 + 2.0im) it will be Complex{Float64}, which a complex number with the parts represented by Float64.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"There are also some cool things like an irrational representation of π.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"π","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nTo make π (and most other Greek letters), type \\pi and then press [TAB].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(π)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"However, if we do math with irrational numbers, they get converted to Float64:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(2π / 3)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Floating-point-numbers","page":"Getting started with Julia","title":"Floating point numbers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nIf you aren't familiar with floating point numbers, make sure to read this section carefully.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A Float64 is a floating point approximation of a real number using 64-bits of information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because it is an approximation, things we know hold true in mathematics don't hold true in a computer. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 == 0.3","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A more complicated example is:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"sin(2π / 3) == √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGet √ by typing \\sqrt then press [TAB].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Let's see what the differences are:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 - 0.3\nsin(2π / 3) - √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"They are small, but not zero.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"One way of explaining this difference is to consider how we would write 1 / 3 and 2 / 3 using only four digits after the decimal point. We would write 1 / 3 as 0.3333, and 2 / 3 as 0.6667. So, despite the fact that 2 * (1 / 3) == 2 / 3, 2 * 0.3333 == 0.6666 != 0.6667.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Let's try that again using ≈ (\\approx + [TAB]) instead of ==:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 ≈ 0.3\nsin(2π / 3) ≈ √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"≈ is a clever way of calling the isapprox function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(sin(2π / 3), √3 / 2; atol = 1e-8)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nFloating point is the reason solvers use tolerances when they solve optimization models. A common mistake you're likely to make is checking whether a binary variable is 0 using value(z) == 0. Always remember to use something like isapprox when comparing floating point numbers.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that isapprox will always return false if one of the number being compared is 0 and atol is zero (its default value).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1e-300 ≈ 0.0","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"so always set a nonzero value of atol if one of the arguments can be zero.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(1e-9, 0.0; atol = 1e-8)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGurobi has a good series of articles on the implications of floating point in optimization if you want to read more.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"If you aren't careful, floating point arithmetic can throw up all manner of issues. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1e-16 == 1","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"It even turns out that floating point numbers aren't associative:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"(1 + 1e-16) - 1e-16 == 1 + (1e-16 - 1e-16)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"It's important to note that this issue isn't Julia-specific. It happens in every programming language (try it out in Python).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Vectors,-matrices,-and-arrays","page":"Getting started with Julia","title":"Vectors, matrices, and arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to MATLAB, Julia has native support for vectors, matrices and tensors; all of which are represented by arrays of different dimensions. Vectors are constructed by comma-separated elements surrounded by square brackets:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"b = [5, 6]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Matrices can be constructed with spaces separating the columns, and semicolons separating the rows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A = [1.0 2.0; 3.0 4.0]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We can do linear algebra:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = A \\ b","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nHere is floating point at work again; x is approximately [-4, 4.5].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A * x\nA * x ≈ b","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that when multiplying vectors and matrices, dimensions matter. For example, you can't multiply a vector by a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    b * b\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"But multiplying transposes works:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"b' * b\nb * b'","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Other-common-types","page":"Getting started with Julia","title":"Other common types","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/#Comments","page":"Getting started with Julia","title":"Comments","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Although not technically a type, code comments begin with the # character:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1  # This is a comment","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Multiline comments begin with #= and end with =#:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"#=\nHere is a\nmultiline comment\n=#","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Comments can even be nested inside expressions. This is sometimes helpful when documenting inputs to functions:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(\n    sin(π),\n    0.0;\n    #= We need an explicit atol here because we are comparing with 0 =#\n    atol = 0.001,\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Strings","page":"Getting started with Julia","title":"Strings","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Double quotes are used for strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(\"This is Julia\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Unicode is fine in strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(\"π is about 3.1415\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use println to print a string:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"println(\"Hello, World!\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use $() to interpolate values into a string:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = 123\nprintln(\"The value of x is: $(x)\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use triple-quotes for multiline strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"s = \"\"\"\nHere is\na\nmultiline string\n\"\"\"\n\nprintln(s)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Symbols","page":"Getting started with Julia","title":"Symbols","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia Symbols are a data structure from the compiler that represent Julia identifiers (that is, variable names).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"println(\"The value of x is: $(eval(:x))\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nWe used eval here to demonstrate how Julia links Symbols to variables. However, avoid calling eval in your code. It is usually a sign that your code is doing something that could be more easily achieved a different way. The Community Forum is a good place to ask for advice on alternative approaches.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(:x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can think of a Symbol as a String that takes up less memory, and that can't be modified.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Convert between String and Symbol using their constructors:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"String(:abc)\nSymbol(\"abc\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nSymbols are often (ab)used to stand in for a String or an Enum, when one of the latter is likely a better choice. The JuMP Style guide recommends reserving Symbols for identifiers. See @enum vs. Symbol for more.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Tuples","page":"Getting started with Julia","title":"Tuples","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia makes extensive use of a simple data structure called Tuples. Tuples are immutable collections of values. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t = (\"hello\", 1.2, :foo)\ntypeof(t)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Tuples can be accessed by index, similar to arrays:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t[2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"And they can be \"unpacked\" like so:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"a, b, c = t\nb","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The values can also be given names, which is a convenient way of making light-weight data structures.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t = (word = \"hello\", num = 1.2, sym = :foo)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Values can be accessed using dot syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t.word","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Dictionaries","page":"Getting started with Julia","title":"Dictionaries","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to Python, Julia has native support for dictionaries. Dictionaries provide a very generic way of mapping keys to values.  For example, a map of integers to strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d1 = Dict(1 => \"A\", 2 => \"B\", 4 => \"D\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nType-stuff again: Dict{Int64,String} is a dictionary with Int64 keys and String values.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Looking up a value uses the bracket syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d1[2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dictionaries support non-integer keys and can mix data types:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"A\" => 1, \"B\" => 2.5, \"D\" => 2 - 3im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nJulia types form a hierarchy. Here the value type of the dictionary is Number, which is a generalization of Int64, Float64, and Complex{Int}. Leaf nodes in this hierarchy are called \"concrete\" types, and all others are called \"Abstract.\" In general, having variables with abstract types like Number can lead to slower code, so you should try to make sure every element in a dictionary or vector is the same type. For example, in this case we could represent every element as a Complex{Float64}:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"A\" => 1.0 + 0.0im, \"B\" => 2.5 + 0.0im, \"D\" => 2.0 - 3.0im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dictionaries can be nested:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d2 = Dict(\"A\" => 1, \"B\" => 2, \"D\" => Dict(:foo => 3, :bar => 4))\nd2[\"B\"]\nd2[\"D\"][:foo]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Structs","page":"Getting started with Julia","title":"Structs","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can define custom datastructures with struct:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"struct MyStruct\n    x::Int\n    y::String\n    z::Dict{Int,Int}\nend\n\na = MyStruct(1, \"a\", Dict(2 => 3))\na.x","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"By default, these are not mutable","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    a.x = 2\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"However, you can declare a mutable struct which is mutable:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"mutable struct MyStructMutable\n    x::Int\n    y::String\n    z::Dict{Int,Int}\nend\n\na = MyStructMutable(1, \"a\", Dict(2 => 3))\na.x\na.x = 2\na","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Loops","page":"Getting started with Julia","title":"Loops","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia has native support for for-each style loops with the syntax for <value> in <collection> end:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 1:5\n    println(i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nRanges are constructed as start:stop, or start:step:stop.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 1.2:1.1:5.6\n    println(i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This for-each loop also works with dictionaries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for (key, value) in Dict(\"A\" => 1, \"B\" => 2.5, \"D\" => 2 - 3im)\n    println(\"$(key): $(value)\")\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that in contrast to vector languages like MATLAB and R, loops do not result in a significant performance degradation in Julia.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Control-flow","page":"Getting started with Julia","title":"Control flow","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia control flow is similar to MATLAB, using the keywords if-elseif-else-end, and the logical operators || and && for or and and respectively:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 0:5:15\n    if i < 5\n        println(\"$(i) is less than 5\")\n    elseif i < 10\n        println(\"$(i) is less than 10\")\n    else\n        if i == 10\n            println(\"the value is 10\")\n        else\n            println(\"$(i) is bigger than 10\")\n        end\n    end\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Comprehensions","page":"Getting started with Julia","title":"Comprehensions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to languages like Haskell and Python, Julia supports the use of simple loops in the construction of arrays and dictionaries, called comprehensions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A list of increasing integers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i for i in 1:5]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Matrices can be built by including multiple indices:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i * j for i in 1:5, j in 5:10]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Conditional statements can be used to filter out some values:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i for i in 1:10 if i % 2 == 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A similar syntax can be used for building dictionaries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"$(i)\" => i for i in 1:10 if i % 2 == 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Functions","page":"Getting started with Julia","title":"Functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A simple function is defined as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_hello()\n    return println(\"hello\")\nend\nprint_hello()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Arguments can be added to a function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_it(x)\n    return println(x)\nend\nprint_it(\"hello\")\nprint_it(1.234)\nprint_it(:my_id)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Optional keyword arguments are also possible:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_it(x; prefix = \"value:\")\n    return println(\"$(prefix) $(x)\")\nend\nprint_it(1.234)\nprint_it(1.234; prefix = \"val:\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The keyword return is used to specify the return values of a function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function mult(x; y = 2.0)\n    return x * y\nend\n\nmult(4.0)\nmult(4.0; y = 5.0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Anonymous-functions","page":"Getting started with Julia","title":"Anonymous functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The syntax input -> output creates an anonymous function. These are most useful when passed to other functions. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"f = x -> x^2\nf(2)\nmap(x -> x^2, 1:4)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Type-parameters","page":"Getting started with Julia","title":"Type parameters","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We can constrain the inputs to a function using type parameters, which are :: followed by the type of the input we want. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function foo(x::Int)\n    return x^2\nend\n\nfunction foo(x::Float64)\n    return exp(x)\nend\n\nfunction foo(x::Number)\n    return x + 1\nend\n\nfoo(2)\nfoo(2.0)\nfoo(1 + 1im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"But what happens if we call foo with something we haven't defined it for?","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    foo([1, 2, 3])\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A MethodError means that you passed a function something that didn't match the type that it was expecting. In this case, the error message says that it doesn't know how to handle an Vector{Int64}, but it does know how to handle Float64, Int64, and Number.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nRead the \"Closest candidates\" part of the error message carefully to get a hint as to what was expected.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Broadcasting","page":"Getting started with Julia","title":"Broadcasting","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In the example above, we didn't define what to do if f was passed a Vector. Luckily, Julia provides a convenient syntax for mapping f element-wise over arrays. Just add a . between the name of the function and the opening (. This works for any function, including functions with multiple arguments. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"foo.([1, 2, 3])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGet a MethodError when calling a function that takes a Vector, Matrix, or Array? Try broadcasting.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Mutable-vs-immutable-objects","page":"Getting started with Julia","title":"Mutable vs immutable objects","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Some types in Julia are mutable, which means you can change the values inside them. A good example is an array. You can modify the contents of an array without having to make a new array.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In contrast, types like Float64 are immutable. You cannot modify the contents of a Float64.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This is something to be aware of when passing types into functions. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function mutability_example(mutable_type::Vector{Int}, immutable_type::Int)\n    mutable_type[1] += 1\n    immutable_type += 1\n    return\nend\n\nmutable_type = [1, 2, 3]\nimmutable_type = 1\n\nmutability_example(mutable_type, immutable_type)\n\nprintln(\"mutable_type: $(mutable_type)\")\nprintln(\"immutable_type: $(immutable_type)\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because Vector{Int} is a mutable type, modifying the variable inside the function changed the value outside of the function. In contrast, the change to immutable_type didn't modify the value outside the function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can check mutability with the isimmutable function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isimmutable([1, 2, 3])\nisimmutable(1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#The-package-manager","page":"Getting started with Julia","title":"The package manager","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/#Installing-packages","page":"Getting started with Julia","title":"Installing packages","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"No matter how wonderful Julia's base language is, at some point you will want to use an extension package.  Some of these are built-in, for example random number generation is available in the Random package in the standard library. These packages are loaded with the commands using and import.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Random  # The equivalent of Python's `from Random import *`\nimport Random  # The equivalent of Python's `import Random`\n\nRandom.seed!(33)\n\n[rand() for i in 1:10]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The Package Manager is used to install packages that are not part of Julia's standard library.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"For example the following can be used to install JuMP,","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"For a complete list of registered Julia packages see the package listing at JuliaHub.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"From time to you may wish to use a Julia package that is not registered.  In this case a git repository URL can be used to install the package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Pkg\nPkg.add(\"https://github.com/user-name/MyPackage.jl.git\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Package-environments","page":"Getting started with Julia","title":"Package environments","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"By default, Pkg.add will add packages to Julia's global environment. However, Julia also has built-in support for virtual environments.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Activate a virtual environment with:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"import Pkg; Pkg.activate(\"/path/to/environment\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can see what packages are installed in the current environment with Pkg.status().","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nWe strongly recommend you create a Pkg environment for each project that you create in Julia, and add only the packages that you need, instead of adding lots of packages to the global environment. The Pkg manager documentation has more information on this topic.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#linear_tips_and_tricks","page":"Tips and tricks","title":"Tips and tricks","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"tip: Tip\nA good source of tips is the Mosek Modeling Cookbook.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial collates some tips and tricks you can use when formulating mixed-integer programs. It uses the following packages:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"using JuMP","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Absolute-value","page":"Tips and tricks","title":"Absolute value","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model the absolute value function t ge x, there are a few options. In all cases, these reformulations only work if you are minimizing t \"down\" into x. They do not work if you are trying to maximize x.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-1","page":"Tips and tricks","title":"Option 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option adds two linear inequality constraints:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x)\n@variable(model, t)\n@constraint(model, t >= x)\n@constraint(model, t >= -x)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-2","page":"Tips and tricks","title":"Option 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option uses two non-negative variables and forms expressions for x and t:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, z[1:2] >= 0)\n@expression(model, t, z[1] + z[2])\n@expression(model, x, z[1] - z[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-3","page":"Tips and tricks","title":"Option 3","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option uses MOI.NormOneCone and lets JuMP choose the reformulation:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x)\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormOneCone(2))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#L1-norm","page":"Tips and tricks","title":"L1-norm","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model min x_1, that is, min sumlimits_i x_i, use the MOI.NormOneCone:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormOneCone(1 + length(x)))\n@objective(model, Min, t)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Infinity-norm","page":"Tips and tricks","title":"Infinity-norm","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model min x_infty, that is, min maxlimits_i x_i, use the MOI.NormInfinityCone:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormInfinityCone(1 + length(x)))\n@objective(model, Min, t)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Max","page":"Tips and tricks","title":"Max","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model t ge maxx y, do:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, t)\n@variable(model, x)\n@variable(model, y)\n@constraint(model, t >= x)\n@constraint(model, t >= y)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This reformulation does not work for t ge minx y.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Min","page":"Tips and tricks","title":"Min","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model t le minx y, do:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, t)\n@variable(model, x)\n@variable(model, y)\n@constraint(model, t <= x)\n@constraint(model, t <= y)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This reformulation does not work for t le maxx y.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Boolean-operators","page":"Tips and tricks","title":"Boolean operators","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Binary variables can be used to construct logical operators. Here are some example.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Or","page":"Tips and tricks","title":"Or","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_3 = x_1 lor x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraints(model, begin\n    x[1] <= x[3]\n    x[2] <= x[3]\n    x[3] <= x[1] + x[2]\nend)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#And","page":"Tips and tricks","title":"And","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_3 = x_1 land x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraints(model, begin\n    x[3] <= x[1]\n    x[3] <= x[2]\n    x[3] >= x[1] + x[2] - 1\nend)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Not","page":"Tips and tricks","title":"Not","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_1 neg x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2], Bin)\n@constraint(model, x[1] == 1 - x[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Implies","page":"Tips and tricks","title":"Implies","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_1 implies x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2], Bin)\n@constraint(model, x[1] <= x[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Disjunctions","page":"Tips and tricks","title":"Disjunctions","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/#Problem","page":"Tips and tricks","title":"Problem","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Suppose that we have two constraints a^top x leq b and c^top x leq d, and we want at least one to hold.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick","page":"Tips and tricks","title":"Trick","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Introduce a \"big-M\" multiplied by a binary variable to relax one of the constraints.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example Either x_1 leq 1 or x_2 leq 2.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, y, Bin)\nM = 100\n@constraint(model, x[1] <= 1 + M * y)\n@constraint(model, x[2] <= 2 + M * (1 - y))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"warning: Warning\nIf M is too small, the solution may be suboptimal. If M is too big, the solver may encounter numerical issues. Try to use domain knowledge to choose an M that is just right. Gurobi has a good documentation section on this topic.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Indicator-constraints","page":"Tips and tricks","title":"Indicator constraints","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/#Problem-2","page":"Tips and tricks","title":"Problem","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Suppose we want to model that a certain linear inequality must be satisfied when some other event occurs, that is, for a binary variable z, we want to model the implication:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"z = 1 implies a^top x leq b","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick-1","page":"Tips and tricks","title":"Trick 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Some solvers have native support for indicator constraints.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 1.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\n@constraint(model, z => {sum(x) <= 1})","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\n@constraint(model, !z => {sum(x) <= 1})","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick-2","page":"Tips and tricks","title":"Trick 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"If the solver doesn't support indicator constraints, you an use the big-M trick.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 1.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\nM = 100\n@constraint(model, sum(x) <= 1 + M * (1 - z))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\nM = 100\n@constraint(model, sum(x) <= 1 + M * z)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Semi-continuous-variables","page":"Tips and tricks","title":"Semi-continuous variables","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A semi-continuous variable is a continuous variable between bounds lu that also can assume the value zero, that is: x in 0 cup lu","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x in 0cup 1 2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x in Semicontinuous(1.0, 2.0))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"You can also represent a semi-continuous variable using the reformulation:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x)\n@variable(model, z, Bin)\n@constraint(model, x <= 2 * z)\n@constraint(model, x >= 1 * z)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"When z = 0 the two constraints are equivalent to 0 <= x <= 0. When z = 1, the two constraints are equivalennt to 1 <= x <= 2.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Semi-integer-variables","page":"Tips and tricks","title":"Semi-integer variables","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A semi-integer variable is a variable which assumes integer values between bounds lu and can also assume the value zero: x in 0 cup l u cap mathbbZ","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x in Semiinteger(5.0, 10.0))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"You can also represent a semi-integer variable using the reformulation:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x, Int)\n@variable(model, z, Bin)\n@constraint(model, x <= 10 * z)\n@constraint(model, x >= 5 * z)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"When z = 0 the two constraints are equivalent to 0 <= x <= 0. When z = 1, the two constraints are equivalennt to 5 <= x <= 10.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Special-Ordered-Sets-of-Type-1","page":"Tips and tricks","title":"Special Ordered Sets of Type 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A Special Ordered Set of Type 1 is a set of variables, at most one of which can take a non-zero value, all others being at 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"They most frequently apply where a set of variables are actually binary variables. In other words, we have to choose at most one from a set of possibilities.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraint(model, x in SOS1())","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"You can optionally pass SOS1 a weight vector like","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"@constraint(model, x in SOS1([0.2, 0.5, 0.3]))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"If the decision variables are related and have a physical ordering, then the weight vector, although not used directly in the constraint, can help the solver make a better decision in the solution process.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#tip_sos2","page":"Tips and tricks","title":"Special Ordered Sets of Type 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A Special Ordered Set of type 2 is a set of non-negative variables, of which at most two can be non-zero, and if two are non-zero these must be consecutive in their ordering.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@constraint(model, x in SOS2([3.0, 1.0, 2.0]))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"The ordering provided by the weight vector is more important in this case as the variables need to be consecutive according to the ordering. For example, in the above constraint, the possible pairs are:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Consecutive\n(x[1] and x[3]) as they correspond to 3 and 2 resp. and thus can be non-zero\n(x[2] and x[3]) as they correspond to 1 and 2 resp. and thus can be non-zero\nNon-consecutive\n(x[1] and x[2]) as they correspond to 3 and 1 resp. and thus cannot be non-zero","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Piecewise-linear-approximations","page":"Tips and tricks","title":"Piecewise linear approximations","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"SOSII constraints are most often used to form piecewise linear approximations of a function.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Given a set of points for x:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x̂ = -1:0.5:2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"and a set of corresponding points for y:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"ŷ = x̂ .^ 2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"the piecewise linear approximation is constructed by representing x and y as convex combinations of x̂ and ŷ.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"N = length(x̂)\nmodel = Model();\n@variable(model, -1 <= x <= 2)\n@variable(model, y)\n@variable(model, 0 <= λ[1:N] <= 1)\n@objective(model, Max, y)\n@constraints(model, begin\n    x == sum(x̂[i] * λ[i] for i in 1:N)\n    y == sum(ŷ[i] * λ[i] for i in 1:N)\n    sum(λ) == 1\n    λ in SOS2()\nend)","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"EditURL = \"web_app.jl\"","category":"page"},{"location":"tutorials/applications/web_app/#Serving-web-apps","page":"Serving web apps","title":"Serving web apps","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"This tutorial demonstrates how to setup and serve JuMP models via a REST API.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"In the example app we are building, we solve a trivial mixed-integer program, which is parameterized by the lower bound of a variable. To call the service, users send an HTTP POST request with JSON contents indicating the lower bound. The returned value is the solution of the mixed-integer program as JSON.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"First, we need JuMP and a solver:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"We also need HTTP.jl to act as our REST server, and JSON.jl to marshal data.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"import HTTP\nimport JSON","category":"page"},{"location":"tutorials/applications/web_app/#The-server-side","page":"Serving web apps","title":"The server side","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"The core components of our REST server are endpoints. These are functions which accept a Dict{String,Any} of input parameters, and return a Dict{String,Any} as output. The types are Dict{String,Any} because we're going to read these to and from JSON.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Here's a very simple endpoint: it accepts params as input, formulates and solves a trivial mixed-integer program, and then returns a dictionary with the result.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function endpoint_solve(params::Dict{String,Any})\n    if !haskey(params, \"lower_bound\")\n        return Dict{String,Any}(\n            \"status\" => \"failure\",\n            \"reason\" => \"missing lower_bound param\",\n        )\n    elseif !(params[\"lower_bound\"] isa Real)\n        return Dict{String,Any}(\n            \"status\" => \"failure\",\n            \"reason\" => \"lower_bound is not a number\",\n        )\n    end\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x >= params[\"lower_bound\"], Int)\n    optimize!(model)\n    ret = Dict{String,Any}(\n        \"status\" => \"okay\",\n        \"terminaton_status\" => termination_status(model),\n        \"primal_status\" => primal_status(model),\n    )\n    # Only include the `x` key if it has a value.\n    if has_values(model)\n        ret[\"x\"] = value(x)\n    end\n    return ret\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"When we call this, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"endpoint_solve(Dict{String,Any}(\"lower_bound\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"endpoint_solve(Dict{String,Any}())","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"For a second function, we need a function that accepts an HTTP.Request object and returns an HTTP.Response object.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function serve_solve(request::HTTP.Request)\n    data = JSON.parse(String(request.body))\n    solution = endpoint_solve(data)\n    return HTTP.Response(200, JSON.json(solution))\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Finally, we need an HTTP server. There are a variety of ways you can do this in HTTP.jl. We use an explicit Sockets.listen so we have manual control of when we shutdown the server.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function setup_server(host, port)\n    server = HTTP.Sockets.listen(host, port)\n    HTTP.serve!(host, port; server = server) do request\n        try\n            # Extend the server by adding other endpoints here.\n            if request.target == \"/api/solve\"\n                return serve_solve(request)\n            else\n                return HTTP.Response(404, \"target $(request.target) not found\")\n            end\n        catch err\n            # Log details about the exception server-side\n            @info \"Unhandled exception: $err\"\n            # Return a response to the client\n            return HTTP.Response(500, \"internal error\")\n        end\n    end\n    return server\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"warning: Warning\nHTTP.jl does not serve requests on a separate thread. Therefore, a long-running job will block the main thread, preventing concurrent users from submitting requests. To work-around this, read HTTP.jl issue 798 or watch Building Microservices and Applications in Julia from JuliaCon 2020.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"server = setup_server(HTTP.ip\"127.0.0.1\", 8080)","category":"page"},{"location":"tutorials/applications/web_app/#The-client-side","page":"Serving web apps","title":"The client side","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Now that we have a server, we can send it requests via this function:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function send_request(data::Dict; endpoint::String = \"solve\")\n    ret = HTTP.request(\n        \"POST\",\n        # This should match the URL and endpoint we defined for our server.\n        \"http://127.0.0.1:8080/api/$endpoint\",\n        [\"Content-Type\" => \"application/json\"],\n        JSON.json(data),\n    )\n    if ret.status != 200\n        # This could happen if there are time-outs, network errors, etc.\n        return Dict(\n            \"status\" => \"failure\",\n            \"code\" => ret.status,\n            \"body\" => String(ret.body),\n        )\n    end\n    return JSON.parse(String(ret.body))\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Let's see what happens:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => 0))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"If we don't send a lower_bound, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"invalid_param\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"If we don't send a lower_bound that is a number, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => \"1.2\"))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Finally, we can shutdown our HTTP server:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"close(server)","category":"page"},{"location":"tutorials/applications/web_app/#Next-steps","page":"Serving web apps","title":"Next steps","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"For more complicated examples relating to HTTP servers, consult the HTTP.jl documentation.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"To see how you can integrate this with a larger JuMP model, read Design patterns for larger models.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"EditURL = \"https://github.com/jump-dev/SDPT3.jl/blob/b565aac2a58818090d521f2340e71f597688e4fb/README.md\"","category":"page"},{"location":"packages/SDPT3/#SDPT3.jl","page":"jump-dev/SDPT3.jl","title":"SDPT3.jl","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"SDPT3.jl is wrapper for the SDPT3 solver.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"an exported sdpt3 function that is a thin wrapper on top of the sdpt3 MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SDPT3/#Affiliation","page":"jump-dev/SDPT3.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SDPT3.","category":"page"},{"location":"packages/SDPT3/#License","page":"jump-dev/SDPT3.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"SDPT3.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The underlying solver, SDPT3 is licensed under the GPL v2 License.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"In addition, SDPT3 requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SDPT3/#Use-with-JuMP","page":"jump-dev/SDPT3.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"To use SDPT3 with JuMP, do:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"using JuMP, SDPT3\nmodel = Model(SDPT3.Optimizer)\nset_attribute(model, \"printlevel\", 0)","category":"page"},{"location":"packages/SDPT3/#Installation","page":"jump-dev/SDPT3.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SeDuMi software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Then, install SDPT3.jl using Pkg.add:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"import Pkg\nPkg.add(\"SDPT3\")","category":"page"},{"location":"packages/SDPT3/#SDPT3-not-in-PATH","page":"jump-dev/SDPT3.jl","title":"SDPT3 not in PATH","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If you get the error:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Error using save\nVariable 'jx_sdpt3_arg_out_1' not found.\n\nERROR: LoadError: MATLAB.MEngineError(\"failed to get variable jx_sdpt3_arg_out_1 from MATLAB session\")\nStacktrace:\n[...]","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The error means that we could not find the sdpt3 function with one output argument using the MATLAB C API. This most likely means that you did not add SDPT3 to the MATLAB's path (that is, the toolbox/local/pathdef.m file).","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work, where /path/to/sdpt3/ is the directory where the sdpt3 folder is located:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"julia> using MATLAB\n\njulia> cd(\"/path/to/sdpt3/\") do\n           MATLAB.mat\"install_sdpt3\"\n       end\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"An alternative fix is suggested in the following issue.","category":"page"},{"location":"packages/SDPT3/#Error-in-validate","page":"jump-dev/SDPT3.jl","title":"Error in validate","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If you get the error:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Brace indexing is not supported for variables of this type.\n\nError in validate\n\nError in sdpt3 (line 171)\n   [blk,At,C,b,blkdim,numblk,parbarrier] = validate(blk,At,C,b,par,parbarrier);\n\nError using save\nVariable 'jx_sdpt3_arg_out_1' not found.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"It might mean that you have added SDPNAL in addition to SDPT3 in the MATLAB's path (that is, the toolbox/local/pathdef.m file). Because SDPNAL also defines a validate function, this can make sdpt3 call SDPNAL's validate function instead of SDPT3's validate function, which causes the issue.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"One way to fix this from the Julia REPL is to reset the search path to the factory-installed state using restoredefaultpath:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"julia> using MATLAB\n\njulia> MATLAB.restoredefaultpath()\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"EditURL = \"dualization.jl\"","category":"page"},{"location":"tutorials/conic/dualization/#Dualization","page":"Dualization","title":"Dualization","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The purpose of this tutorial is to explain how to use Dualization.jl to improve the performance of some conic optimization models. There are two important takeaways:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"JuMP reformulates problems to meet the input requirements of the solver, potentially increasing the problem size by adding slack variables and constraints.\nSolving the dual of a conic model can be more efficient than solving the primal.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Dualization.jl is a package which fixes these problems, allowing you to solve the dual instead of the primal with a one-line change to your code.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"using JuMP\nimport Dualization\nimport SCS","category":"page"},{"location":"tutorials/conic/dualization/#Background","page":"Dualization","title":"Background","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Conic optimization solvers typically accept one of two input formulations.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The first is the standard conic form:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   A x = b  \n                               x in mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"in which we have a set of linear equality constraints Ax = b and the variables belong to a cone mathcalK.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The second is the geometric conic form:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   A x - b in mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"in which an affine function Ax - b belongs to a cone mathcalK and the variables are free.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"It is trivial to convert between these two representations, for example, to go from the geometric conic form to the standard conic form we introduce slack variables y:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   beginbmatrixA  -Iendbmatrix beginbmatrixxyendbmatrix = b \n                               beginbmatrixxyendbmatrix in mathbbR^n times mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"and to go from the standard conic form to the geometric conic form, we can rewrite the equality constraint as a function belonging to the {0} cone:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst  beginbmatrixAIendbmatrix x - beginbmatrixb0endbmatrix in 0 times mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"From a theoretical perspective, the two formulations are equivalent, and if you implement a model in the standard conic form and pass it to a geometric conic form solver (or vice versa), then JuMP will automatically reformulate the problem into the correct formulation.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"From a practical perspective though, the reformulations are problematic because the additional slack variables and constraints can make the problem much larger and therefore harder to solve.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"You should also note many problems contain a mix of conic constraints and variables, and so they do not neatly fall into one of the two formulations. In these cases, JuMP reformulates only the variables and constraints as necessary to convert the problem into the desired form.","category":"page"},{"location":"tutorials/conic/dualization/#Primal-and-dual-formulations","page":"Dualization","title":"Primal and dual formulations","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Duality plays a large role in conic optimization. For a detailed description of conic duality, see Duality.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"A useful observation is that if the primal problem is in standard conic form, then the dual problem is in geometric conic form, and vice versa. Moreover, the primal and dual may have a different number of variables and constraints, although which one is smaller depends on the problem. Therefore, instead of reformulating the problem from one form to the other, it can be more efficient to solve the dual instead of the primal.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"To demonstrate, we use a variation of the Maximum cut via SDP example.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The primal formulation (in standard conic form) is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"model_primal = Model()\n@variable(model_primal, X[1:2, 1:2], PSD)\n@objective(model_primal, Max, sum([1 -1; -1 1] .* X))\n@constraint(model_primal, primal_c[i = 1:2], 1 - X[i, i] == 0)\nprint(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This problem has three scalar decision variables (the matrix X is symmetric), two scalar equality constraints, and a constraint that X is positive semidefinite.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The dual of model_primal is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"model_dual = Model()\n@variable(model_dual, y[1:2])\n@objective(model_dual, Min, sum(y))\n@constraint(model_dual, dual_c, [y[1]-1 1; 1 y[2]-1] in PSDCone())\nprint(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This problem has two scalar decision variables, and a 2x2 positive semidefinite matrix constraint.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"tip: Tip\nIf you haven't seen conic duality before, try deriving the dual problem based on the description in Duality. You'll need to know that the dual cone of PSDCone is the PSDCone.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"When we solve model_primal with SCS.Optimizer, SCS reports three variables (variables n: 3), five rows in the constraint matrix (constraints m: 5), and five non-zeros in the matrix (nnz(A): 5):","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_primal, SCS.Optimizer)\noptimize!(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"(There are five rows in the constraint matrix because SCS expects problems in geometric conic form, and so JuMP has reformulated the X, PSD variable constraint into the affine constraint X .+ 0 in PSDCone().)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The solution we obtain is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(X)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(primal_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"objective_value(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"When we solve model_dual with SCS.Optimizer, SCS reports two variables (variables n: 2), three rows in the constraint matrix (constraints m: 3), and two non-zeros in the matrix (nnz(A): 2):","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_dual, SCS.Optimizer)\noptimize!(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"and the solution we obtain is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(dual_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(y)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"objective_value(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This particular problem is small enough that it isn't meaningful to compare the solve times, but in general, we should expect model_dual to solve faster than model_primal because it contains fewer variables and constraints. The difference is particularly noticeable on large-scale optimization problems.","category":"page"},{"location":"tutorials/conic/dualization/#dual_optimizer","page":"Dualization","title":"dual_optimizer","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Manually deriving the conic dual is difficult and error-prone. The package Dualization.jl provides the Dualization.dual_optimizer meta-solver, which wraps any MathOptInterface-compatible solver in an interface that automatically formulates and solves the dual of an input problem.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"To demonstrate, we use Dualization.dual_optimizer to solve model_primal:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_primal, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The performance is the same as if we solved model_dual, and the correct solution is returned to X:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(X)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(primal_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Moreover, if we use dual_optimizer on model_dual, then we get the same performance as if we had solved model_primal:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_dual, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(dual_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(y)","category":"page"},{"location":"tutorials/conic/dualization/#A-mixed-example","page":"Dualization","title":"A mixed example","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The Maximum cut via SDP example is nicely defined because the primal is in standard conic form and the dual is in geometric conic form. However, many practical models contain a mix of the two formulations. One example is The minimum distortion problem:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"D = [0 1 1 1; 1 0 2 2; 1 2 0 2; 1 2 2 0]\nmodel = Model()\n@variable(model, c²)\n@variable(model, Q[1:4, 1:4], PSD)\n@objective(model, Min, c²)\nfor i in 1:4, j in (i+1):4\n    @constraint(model, D[i, j]^2 <= Q[i, i] + Q[j, j] - 2 * Q[i, j])\n    @constraint(model, Q[i, i] + Q[j, j] - 2 * Q[i, j] <= c² * D[i, j]^2)\nend\n@constraint(model, Q[1, 1] == 0)\n@constraint(model, c² >= 1)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"In this formulation, the Q variable is of the form xinmathcalK, but there is also a free variable, c², a linear equality constraint, Q[1, 1] == 0, and some linear inequality constraints. Rather than attempting to derive the formulation that JuMP would pass to SCS and its dual, the simplest solution is to try solving the problem with and without dual_optimizer to see which formulation is most efficient.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model, SCS.Optimizer)\noptimize!(model)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"For this problem, SCS reports that the primal has variables n: 11, constraints m: 24 and that the dual has variables n: 14, constraints m: 24. Therefore, we should probably use the primal formulation because it has fewer variables and the same number of constraints.","category":"page"},{"location":"tutorials/conic/dualization/#When-to-use-dual_optimizer","page":"Dualization","title":"When to use dual_optimizer","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Because it can make the problem larger or smaller, depending on the problem and the choice of solver, there is no definitive rule on when you should use dual_optimizer. However, you should try dual_optimizer if your conic optimization problem takes a long time to solve, or if you need to repeatedly solve similarly structured problems with different data. In some cases solving the dual instead of the primal can make a large difference.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"EditURL = \"ellipse_approx.jl\"","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Ellipsoid-approximation","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial considers the problem of computing extremal ellipsoids: finding ellipsoids that best approximate a given set. As an extension, we show how to use JuMP to inspect the bridges that were used, and how to explore alternative formulations.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The model comes from Section 4.9 \"Applications VII: extremal ellipsoids\" of the book Lectures on Modern Convex Optimization by Ben-Tal and Nemirovski (2001).","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"For a related example, see also the Minimal ellipses tutorial.","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Problem-formulation","page":"Ellipsoid approximation","title":"Problem formulation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Suppose that we are given a set mathcalS consisting of m points in n-dimensional space:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"mathcalS =  x_1 ldots x_m  subset mathbbR^n","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Our goal is to determine an optimal vector c in  mathbbR^n and an optimal n times n real symmetric matrix D such that the ellipse:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"E(D c) =  x  (x - c)^top D ( x - c) leq 1 ","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"contains mathcalS and has the smallest possible volume.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The optimal D and c are given by the optimization problem:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"beginaligned\nmax        quad  t                                                                    \ntextst quad  Z succeq 0 \n                   beginbmatrix s  z^top  z  Z  endbmatrix succeq 0 \n                   x_i^top Z x_i - 2x_i^top z + s leq 1     quad i=1 ldots m \n                   t le sqrtndet(Z)\nendaligned","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"where D = Z_* and c = Z_*^-1 z_*.","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Required-packages","page":"Ellipsoid approximation","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport Random\nimport SCS","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Data","page":"Ellipsoid approximation","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We first need to generate some points to work with.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"function generate_point_cloud(\n    m;            # number of 2-dimensional points\n    a = 10,       # scaling in x direction\n    b = 2,        # scaling in y direction\n    rho = π / 6,  # rotation of points around origin\n    random_seed = 1,\n)\n    rng = Random.MersenneTwister(random_seed)\n    P = randn(rng, Float64, m, 2)\n    Phi = [a*cos(rho) a*sin(rho); -b*sin(rho) b*cos(rho)]\n    S = P * Phi\n    return S\nend","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"For the sake of this example, let's take m = 600:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"S = generate_point_cloud(600);\nnothing #hide","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We will visualise the points (and ellipse) using the Plots package:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"r = 1.1 * maximum(abs.(S))\nplot = Plots.scatter(\n    S[:, 1],\n    S[:, 2];\n    xlim = (-r, r),\n    ylim = (-r, r),\n    label = nothing,\n    c = :green,\n    shape = :x,\n    size = (600, 600),\n)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#JuMP-formulation","page":"Ellipsoid approximation","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Now let's build and the JuMP model. We'll compute D and c after the solve.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"model = Model(SCS.Optimizer)\n# We need to use a tighter tolerance for this example, otherwise the bounding\n# ellipse won't actually be bounding...\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\nm, n = size(S)\n@variable(model, z[1:n])\n@variable(model, Z[1:n, 1:n], PSD)\n@variable(model, s)\n@variable(model, t)\n@constraint(model, [s z'; z Z] >= 0, PSDCone())\n@constraint(\n    model,\n    [i in 1:m],\n    S[i, :]' * Z * S[i, :] - 2 * S[i, :]' * z + s <= 1,\n)\n@constraint(model, [t; vec(Z)] in MOI.RootDetConeSquare(n))\n@objective(model, Max, t)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Results","page":"Ellipsoid approximation","title":"Results","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"After solving the model to optimality we can recover the solution in terms of D and c:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"D = value.(Z)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"c = D \\ value.(z)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Finally, overlaying the solution in the plot we see the minimal volume approximating ellipsoid:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"P = sqrt(D)\nq = -P * c\ndata = [tuple(P \\ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)]\nPlots.plot!(plot, data; c = :crimson, label = nothing)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Alternative-formulations","page":"Ellipsoid approximation","title":"Alternative formulations","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The formulation of model uses MOI.RootDetConeSquare. However, because SCS does not natively support this cone, JuMP automatically reformulates the problem into an equivalent problem that SCS does support. You can see the reformulation that JuMP chose using print_active_bridges:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"There's a lot going on here, but the first bullet is:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"* Unsupported objective: MOI.VariableIndex\n|  bridged by:\n|   MOIB.Objective.FunctionizeBridge{Float64}\n|  introduces:\n|   * Supported objective: MOI.ScalarAffineFunction{Float64}","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This says that SCS does not support a MOI.VariableIndex objective function, and that JuMP used a MOI.Bridges.Objective.FunctionizeBridge to convert it into a MOI.ScalarAffineFunction{Float64} objective function.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We can leave JuMP to do the reformulation, or we can rewrite our model to have an objective function that SCS natively supports:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"@objective(model, Max, 1.0 * t + 0.0);\nnothing #hide","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Re-printing the active bridges:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"we get * Supported objective: MOI.ScalarAffineFunction{Float64}.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We can manually implement some other reformulations to change our model to something that SCS more closely supports by:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Replacing the MOI.VectorOfVariables in MOI.PositiveSemidefiniteConeTriangle constraint @variable(model, Z[1:n, 1:n], PSD) with the MOI.VectorAffineFunction in MOI.PositiveSemidefiniteConeTriangle @constraint(model, Z >= 0, PSDCone()).\nReplacing the MOI.VectorOfVariables in MOI.PositiveSemidefiniteConeSquare constraint [s z'; z Z] >= 0, PSDCone() with the MOI.VectorAffineFunction in MOI.PositiveSemidefiniteConeTriangle @constraint(model, LinearAlgebra.Symmetric([s z'; z Z]) >= 0, PSDCone()).\nReplacing the MOI.ScalarAffineFunction in MOI.GreaterThan constraints with the vectorized equivalent of MOI.VectorAffineFunction in MOI.Nonnegatives\nReplacing the MOI.VectorOfVariables in MOI.RootDetConeSquare constraint with MOI.VectorAffineFunction in MOI.RootDetConeTriangle.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Note that we still need to bridge MOI.PositiveSemidefiniteConeTriangle constraints because SCS uses an internal SCS.ScaledPSDCone set instead.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"model = Model(SCS.Optimizer)\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\n@variable(model, z[1:n])\n@variable(model, s)\n@variable(model, t)\n# The former @variable(model, Z[1:n, 1:n], PSD)\n@variable(model, Z[1:n, 1:n], Symmetric)\n@constraint(model, Z >= 0, PSDCone())\n# The former [s z'; z Z] >= 0, PSDCone()\n@constraint(model, LinearAlgebra.Symmetric([s z'; z Z]) >= 0, PSDCone())\n# The former constraint S[i, :]' * Z * S[i, :] - 2 * S[i, :]' * z + s <= 1\nf = [1 - S[i, :]' * Z * S[i, :] + 2 * S[i, :]' * z - s for i in 1:m]\n@constraint(model, f in MOI.Nonnegatives(m))\n# The former constraint [t; vec(Z)] in MOI.RootDetConeSquare(n)\n@constraint(model, 1 * [t; triangle_vec(Z)] .+ 0 in MOI.RootDetConeTriangle(n))\n# The former @objective(model, Max, t)\n@objective(model, Max, 1 * t + 0)\noptimize!(model)\nsolve_time_1 = solve_time(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This formulation gives the much smaller graph:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The last bullet shows how JuMP reformulated the MOI.RootDetConeTriangle constraint by adding a mix of MOI.PositiveSemidefiniteConeTriangle and MOI.GeometricMeanCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Because SCS doesn't natively support the MOI.GeometricMeanCone, these constraints were further bridged using a MOI.Bridges.Constraint.GeoMeanToPowerBridge to a series of MOI.PowerCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"However, there are many other ways that a MOI.GeometricMeanCone can be reformulated into something that SCS supports. Let's see what happens if we use remove_bridge to remove the MOI.Bridges.Constraint.GeoMeanToPowerBridge:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"remove_bridge(model, MOI.Bridges.Constraint.GeoMeanToPowerBridge)\noptimize!(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This time, the solve took:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"solve_time_2 = solve_time(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"where previously it took","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"solve_time_1","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Why was the solve time different?","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This time, JuMP used a MOI.Bridges.Constraint.GeoMeanBridge to reformulate the constraint into a set of MOI.RotatedSecondOrderCone constraints, which were further reformulated into a set of supported MOI.SecondOrderCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Since the two models are equivalent, we can conclude that for this particular model, the MOI.SecondOrderCone formulation is more efficient.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"In general though, the performance of a particular reformulation is problem- and solver-specific. Therefore, JuMP chooses to minimize the number of bridges in the default reformulation, leaving you to explore alternative formulations using the tools and techniques shown in this tutorial.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"DocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/containers/#Containers","page":"Containers","title":"Containers","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"JuMP provides specialized containers similar to AxisArrays that enable multi-dimensional arrays with non-integer indices.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"These containers are created automatically by JuMP's macros. Each macro has the same basic syntax:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"@macroname(model, name[key1=index1, index2; optional_condition], other stuff)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The containers are generated by the name[key1=index1, index2; optional_condition] syntax. Everything else is specific to the particular macro.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Containers can be named, for example, name[key=index], or unnamed, for example, [key=index]. We call unnamed containers anonymous.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We call the bits inside the square brackets and before the ; the index sets. The index sets can be named, for example, [i = 1:4], or they can be unnamed, for example, [1:4].","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We call the bit inside the square brackets and after the ; the condition. Conditions are optional.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"In addition to the standard JuMP macros like @variable and @constraint, which construct containers of variables and constraints respectively, you can use Containers.@container to construct containers with arbitrary elements.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We will use this macro to explain the three types of containers that are natively supported by JuMP: Array, Containers.DenseAxisArray, and Containers.SparseAxisArray.","category":"page"},{"location":"manual/containers/#Array","page":"Containers","title":"Array","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"An Array is created when the index sets are rectangular and the index sets are of the form 1:n.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container(x[i = 1:2, j = 1:3], (i, j))\n2×3 Matrix{Tuple{Int64, Int64}}:\n (1, 1)  (1, 2)  (1, 3)\n (2, 1)  (2, 2)  (2, 3)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The result is a normal Julia Array, so you can do all the usual things.","category":"page"},{"location":"manual/containers/#Slicing","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Arrays can be sliced","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[:, 1]\n2-element Vector{Tuple{Int64, Int64}}:\n (1, 1)\n (2, 1)\n\njulia> x[2, :]\n3-element Vector{Tuple{Int64, Int64}}:\n (2, 1)\n (2, 2)\n (2, 3)","category":"page"},{"location":"manual/containers/#Looping","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(x)\n           println(x[key])\n       end\n(1, 1)\n(2, 1)\n(1, 2)\n(2, 2)\n(1, 3)\n(2, 3)","category":"page"},{"location":"manual/containers/#Get-the-index-sets","page":"Containers","title":"Get the index sets","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use axes to obtain the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> axes(x)\n(Base.OneTo(2), Base.OneTo(3))","category":"page"},{"location":"manual/containers/#Broadcasting","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over an Array returns an Array","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(x)\n2×3 Matrix{Tuple{Int64, Int64}}:\n (1, 1)  (2, 1)  (3, 1)\n (1, 2)  (2, 2)  (3, 2)","category":"page"},{"location":"manual/containers/#Tables","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the Array into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n6-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Int64, Tuple{Int64, Int64}}}}:\n (I = 1, J = 1, value = (1, 1))\n (I = 2, J = 1, value = (2, 1))\n (I = 1, J = 2, value = (1, 2))\n (I = 2, J = 2, value = (2, 2))\n (I = 1, J = 3, value = (1, 3))\n (I = 2, J = 3, value = (2, 3))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n6×3 DataFrame\n Row │ I      J      value\n     │ Int64  Int64  Tuple…\n─────┼──────────────────────\n   1 │     1      1  (1, 1)\n   2 │     2      1  (2, 1)\n   3 │     1      2  (1, 2)\n   4 │     2      2  (2, 2)\n   5 │     1      3  (1, 3)\n   6 │     2      3  (2, 3)","category":"page"},{"location":"manual/containers/#DenseAxisArray","page":"Containers","title":"DenseAxisArray","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"A Containers.DenseAxisArray is created when the index sets are rectangular, but not of the form 1:n. The index sets can be of any type.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x = Containers.@container([i = 1:2, j = [:A, :B]], (i, j))\n2-dimensional DenseAxisArray{Tuple{Int64, Symbol},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/#Slicing-2","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"DenseAxisArrays can be sliced","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[:, :A]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, Base.OneTo(2)\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :A)\n (2, :A)\n\njulia> x[1, :]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, [:A, :B]\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :A)\n (1, :B)","category":"page"},{"location":"manual/containers/#Looping-2","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(x)\n           println(x[key])\n       end\n(1, :A)\n(2, :A)\n(1, :B)\n(2, :B)","category":"page"},{"location":"manual/containers/#Get-the-index-sets-2","page":"Containers","title":"Get the index sets","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use axes to obtain the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> axes(x)\n(Base.OneTo(2), [:A, :B])","category":"page"},{"location":"manual/containers/#Broadcasting-2","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over a DenseAxisArray returns a DenseAxisArray","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(x)\n2-dimensional DenseAxisArray{Tuple{Symbol, Int64},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{Tuple{Symbol, Int64}}:\n (:A, 1)  (:B, 1)\n (:A, 2)  (:B, 2)","category":"page"},{"location":"manual/containers/#Access-internal-data","page":"Containers","title":"Access internal data","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Array(x) to copy the internal data array into a new Array:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Array(x)\n2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"To access the internal data without a copy, use x.data.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x.data\n2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/#Tables-2","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the DenseAxisArray into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n4-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Symbol, Tuple{Int64, Symbol}}}}:\n (I = 1, J = :A, value = (1, :A))\n (I = 2, J = :A, value = (2, :A))\n (I = 1, J = :B, value = (1, :B))\n (I = 2, J = :B, value = (2, :B))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n4×3 DataFrame\n Row │ I      J       value\n     │ Int64  Symbol  Tuple…\n─────┼────────────────────────\n   1 │     1  A       (1, :A)\n   2 │     2  A       (2, :A)\n   3 │     1  B       (1, :B)\n   4 │     2  B       (2, :B)","category":"page"},{"location":"manual/containers/#Keyword-indexing","page":"Containers","title":"Keyword indexing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If all axes are named, you can use keyword indexing:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[i = 2, j = :A]\n(2, :A)\n\njulia> x[i = :, j = :B]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, Base.OneTo(2)\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :B)\n (2, :B)","category":"page"},{"location":"manual/containers/#SparseAxisArray","page":"Containers","title":"SparseAxisArray","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"A Containers.SparseAxisArray is created when the index sets are non-rectangular. This occurs in two circumstances:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"An index depends on a prior index:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:2, j = i:2], (i, j))\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Int64}, 2, Tuple{Int64, Int64}} with 3 entries:\n  [1, 1]  =  (1, 1)\n  [1, 2]  =  (1, 2)\n  [2, 2]  =  (2, 2)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The [indices; condition] syntax is used:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x = Containers.@container([i = 1:3, j = [:A, :B]; i > 1], (i, j))\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 2, Tuple{Int64, Symbol}} with 4 entries:\n  [2, A]  =  (2, :A)\n  [2, B]  =  (2, :B)\n  [3, A]  =  (3, :A)\n  [3, B]  =  (3, :B)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Here we have the index sets i = 1:3, j = [:A, :B], followed by ;, and then a condition, which evaluates to true or false: i > 1.","category":"page"},{"location":"manual/containers/#Slicing-3","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Slicing is supported:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> y = x[:, :B]\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (2, :B)\n  [3]  =  (3, :B)","category":"page"},{"location":"manual/containers/#Looping-3","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(y)\n           println(y[key])\n       end\n(2, :B)\n(3, :B)","category":"page"},{"location":"manual/containers/#Broadcasting-3","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over a SparseAxisArray returns a SparseAxisArray","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(y)\nJuMP.Containers.SparseAxisArray{Tuple{Symbol, Int64}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (:B, 2)\n  [3]  =  (:B, 3)","category":"page"},{"location":"manual/containers/#Tables-3","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the SparseAxisArray into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n4-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Symbol, Tuple{Int64, Symbol}}}}:\n (I = 2, J = :A, value = (2, :A))\n (I = 2, J = :B, value = (2, :B))\n (I = 3, J = :A, value = (3, :A))\n (I = 3, J = :B, value = (3, :B))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n4×3 DataFrame\n Row │ I      J       value\n     │ Int64  Symbol  Tuple…\n─────┼────────────────────────\n   1 │     2  A       (2, :A)\n   2 │     2  B       (2, :B)\n   3 │     3  A       (3, :A)\n   4 │     3  B       (3, :B)","category":"page"},{"location":"manual/containers/#Keyword-indexing-2","page":"Containers","title":"Keyword indexing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If all axes are named, you can use keyword indexing:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[i = 2, j = :A]\n(2, :A)\n\njulia> x[i = :, j = :B]\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (2, :B)\n  [3]  =  (3, :B)","category":"page"},{"location":"manual/containers/#Forcing-the-container-type","page":"Containers","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Pass container = T to use T as the container. For example:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:2, j = 1:2], i + j, container = Array)\n2×2 Matrix{Int64}:\n 2  3\n 3  4\n\njulia> Containers.@container([i = 1:2, j = 1:2], i + j, container = Dict)\nDict{Tuple{Int64, Int64}, Int64} with 4 entries:\n  (1, 2) => 3\n  (1, 1) => 2\n  (2, 2) => 4\n  (2, 1) => 3","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"You can also pass DenseAxisArray or SparseAxisArray.","category":"page"},{"location":"manual/containers/#How-different-container-types-are-chosen","page":"Containers","title":"How different container types are chosen","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If the compiler can prove at compile time that the index sets are rectangular, and indexed by a compact set of integers that start at 1, Containers.@container will return an array. This is the case if your index sets are visible to the macro as 1:n:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i=1:3, j=1:5], i + j)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"or an instance of Base.OneTo:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = Base.OneTo(3)\nBase.OneTo(3)\n\njulia> Containers.@container([i=set, j=1:5], i + j)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If the compiler can prove that the index set is rectangular, but not necessarily of the form 1:n at compile time, then a Containers.DenseAxisArray will be constructed instead:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = 1:3\n1:3\n\njulia> Containers.@container([i=set, j=1:5], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, Base.OneTo(5)\nAnd data, a 3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"info: Info\nWhat happened here? Although we know that set contains 1:3, at compile time the typeof(set) is a UnitRange{Int}. Therefore, Julia can't prove that the range starts at 1 (it only finds this out at runtime), and it defaults to a  DenseAxisArray. The case where we explicitly wrote i = 1:3 worked because the macro can \"see\" the 1 at compile time.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"However, if you know that the indices do form an Array, you can force the container type with container = Array:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = 1:3\n1:3\n\njulia> Containers.@container([i=set, j=1:5], i + j, container = Array)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Here's another example with something similar:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> a = 1\n1\n\njulia> Containers.@container([i=a:3, j=1:5], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, Base.OneTo(5)\nAnd data, a 3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8\n\njulia> Containers.@container([i=1:a, j=1:5], i + j)\n1×5 Matrix{Int64}:\n 2  3  4  5  6","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Finally, if the compiler cannot prove that the index set is rectangular, a Containers.SparseAxisArray will be created.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"This occurs when some indices depend on a previous one:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i=1:3, j=1:i], i + j)\nJuMP.Containers.SparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  2\n  [2, 1]  =  3\n  [2, 2]  =  4\n  [3, 1]  =  4\n  [3, 2]  =  5\n  [3, 3]  =  6","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"or if there is a condition on the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:5; isodd(i)], i^2)\nJuMP.Containers.SparseAxisArray{Int64, 1, Tuple{Int64}} with 3 entries:\n  [1]  =  1\n  [3]  =  9\n  [5]  =  25","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The condition can depend on multiple indices, the only requirement is that it is an expression that returns true or false:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> condition(i, j) = isodd(i) && iseven(j)\ncondition (generic function with 1 method)\n\njulia> Containers.@container([i = 1:2, j = 1:4; condition(i, j)], i + j)\nJuMP.Containers.SparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 2 entries:\n  [1, 2]  =  3\n  [1, 4]  =  5","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/solutions.md\"","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/solutions/#manual_solutions","page":"Solutions","title":"Solutions","text":"","category":"section"},{"location":"moi/manual/solutions/#Solving-and-retrieving-the-results","page":"Solutions","title":"Solving and retrieving the results","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Once an optimizer is loaded with the objective function and all of the constraints, we can ask the solver to solve the model by calling optimize!.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"MOI.optimize!(optimizer)","category":"page"},{"location":"moi/manual/solutions/#Why-did-the-solver-stop?","page":"Solutions","title":"Why did the solver stop?","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The optimization procedure may stop for a number of reasons. The TerminationStatus attribute of the optimizer returns a TerminationStatusCode object which explains why the solver stopped.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The termination statuses distinguish between proofs of optimality, infeasibility, local convergence, limits, and termination because of something unexpected like invalid problem data or failure to converge.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"A typical usage of the TerminationStatus attribute is as follows:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"status = MOI.get(optimizer, TerminationStatus())\nif status == MOI.OPTIMAL\n    # Ok, we solved the problem!\nelse\n    # Handle other cases.\nend","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"After checking the TerminationStatus, check ResultCount. This attribute returns the number of results that the solver has available to return. A result is defined as a primal-dual pair, but either the primal or the dual may be missing from the result. While the OPTIMAL termination status normally implies that at least one result is available, other statuses do not. For example, in the case of infeasibility, a solver may return no result or a proof of infeasibility. The ResultCount attribute distinguishes between these two cases.","category":"page"},{"location":"moi/manual/solutions/#Primal-solutions","page":"Solutions","title":"Primal solutions","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Use the PrimalStatus optimizer attribute to return a ResultStatusCode describing the status of the primal solution.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Common returns are described below in the Common status situations section.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the primal solution using the VariablePrimal and ConstraintPrimal attributes.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the objective function value using the ObjectiveValue attribute.","category":"page"},{"location":"moi/manual/solutions/#Dual-solutions","page":"Solutions","title":"Dual solutions","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nSee Duality for a discussion of the MOI conventions for primal-dual pairs and certificates.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Use the DualStatus optimizer attribute to return a ResultStatusCode describing the status of the dual solution.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the dual solution using the ConstraintDual attribute.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the dual objective function value using the DualObjectiveValue attribute.","category":"page"},{"location":"moi/manual/solutions/#Common-status-situations","page":"Solutions","title":"Common status situations","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The sections below describe how to interpret typical or interesting status cases for three common classes of solvers. The example cases are illustrative, not comprehensive. Solver wrappers may provide additional information on how the solver's statuses map to MOI statuses.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\n* in the tables indicate that multiple different values are possible.","category":"page"},{"location":"moi/manual/solutions/#Primal-dual-convex-solver","page":"Solutions","title":"Primal-dual convex solver","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Linear programming and conic optimization solvers fall into this category.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nProved optimality OPTIMAL 1 FEASIBLE_POINT FEASIBLE_POINT\nProved infeasible INFEASIBLE 1 NO_SOLUTION INFEASIBILITY_CERTIFICATE\nOptimal within relaxed tolerances ALMOST_OPTIMAL 1 FEASIBLE_POINT FEASIBLE_POINT\nOptimal within relaxed tolerances ALMOST_OPTIMAL 1 ALMOST_FEASIBLE_POINT ALMOST_FEASIBLE_POINT\nDetected an unbounded ray of the primal DUAL_INFEASIBLE 1 INFEASIBILITY_CERTIFICATE NO_SOLUTION\nStall SLOW_PROGRESS 1 * *","category":"page"},{"location":"moi/manual/solutions/#Global-branch-and-bound-solvers","page":"Solutions","title":"Global branch-and-bound solvers","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Mixed-integer programming solvers fall into this category.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nProved optimality OPTIMAL 1 FEASIBLE_POINT NO_SOLUTION\nPresolve detected infeasibility or unboundedness INFEASIBLE_OR_UNBOUNDED 0 NO_SOLUTION NO_SOLUTION\nProved infeasibility INFEASIBLE 0 NO_SOLUTION NO_SOLUTION\nTimed out (no solution) TIME_LIMIT 0 NO_SOLUTION NO_SOLUTION\nTimed out (with a solution) TIME_LIMIT 1 FEASIBLE_POINT NO_SOLUTION\nCPXMIP_OPTIMAL_INFEAS ALMOST_OPTIMAL 1 INFEASIBLE_POINT NO_SOLUTION","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nCPXMIP_OPTIMAL_INFEAS is a CPLEX status that indicates that a preprocessed problem was solved to optimality, but the solver was unable to recover a feasible solution to the original problem. Handling this status was one of the motivating drivers behind the design of MOI.","category":"page"},{"location":"moi/manual/solutions/#Local-search-solvers","page":"Solutions","title":"Local search solvers","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Nonlinear programming solvers fall into this category. It also includes non-global tree search solvers like Juniper.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nConverged to a stationary point LOCALLY_SOLVED 1 FEASIBLE_POINT FEASIBLE_POINT\nCompleted a non-global tree search (with a solution) LOCALLY_SOLVED 1 FEASIBLE_POINT FEASIBLE_POINT\nConverged to an infeasible point LOCALLY_INFEASIBLE 1 INFEASIBLE_POINT *\nCompleted a non-global tree search (no solution found) LOCALLY_INFEASIBLE 0 NO_SOLUTION NO_SOLUTION\nIteration limit ITERATION_LIMIT 1 * *\nDiverging iterates NORM_LIMIT or OBJECTIVE_LIMIT 1 * *","category":"page"},{"location":"moi/manual/solutions/#Querying-solution-attributes","page":"Solutions","title":"Querying solution attributes","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Some solvers will not implement every solution attribute. Therefore, a call like MOI.get(model, MOI.SolveTimeSec()) may throw an UnsupportedAttribute error.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"If you need to write code that is agnostic to the solver (for example, you are writing a library that an end-user passes their choice of solver to), you can work-around this problem using a try-catch:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"function get_solve_time(model)\n    try\n        return MOI.get(model, MOI.SolveTimeSec())\n    catch err\n        if err isa MOI.UnsupportedAttribute\n            return NaN  # Solver doesn't support. Return a placeholder value.\n        end\n        rethrow(err)  # Something else went wrong. Rethrow the error\n    end\nend","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"If, after careful profiling, you find that the try-catch is taking a significant portion of your runtime, you can improve performance by caching the result of the try-catch:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"mutable struct CachedSolveTime{M}\n    model::M\n    supports_solve_time::Bool\n    CachedSolveTime(model::M) where {M} = new(model, true)\nend\n\nfunction get_solve_time(model::CachedSolveTime)\n    if !model.supports_solve_time\n        return NaN\n    end\n    try\n        return MOI.get(model, MOI.SolveTimeSec())\n    catch err\n        if err isa MOI.UnsupportedAttribute\n            model.supports_solve_time = false\n            return NaN\n        end\n        rethrow(err)  # Something else went wrong. Rethrow the error\n    end\nend","category":"page"},{"location":"tutorials/getting_started/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"The purpose of these \"Getting started\" tutorials is to teach new users the basics of Julia and JuMP.","category":"page"},{"location":"tutorials/getting_started/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"The \"Getting started with\" tutorials are basic introductions to different aspects of JuMP and Julia. If you are new to JuMP and Julia, start by reading them in the following order:\nGetting started with Julia\nGetting started with JuMP\nGetting started with sets and indexing\nGetting started with data and plotting\nJulia has a reputation for being \"fast.\" Unfortunately, it is also easy to write slow Julia code. Performance tips contains a number of important tips on how to improve the performance of models you write in JuMP.\nDesign patterns for larger models is a more advanced tutorial that is aimed at users writing large JuMP models. It's in the \"Getting started\" section to give you an early preview of how JuMP makes it easy to structure larger models. If you are new to JuMP you may want to skip or briefly skim this tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"EditURL = \"two_stage_stochastic.jl\"","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Two-stage-stochastic-programs","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The purpose of this tutorial is to demonstrate how to model and solve a two-stage stochastic program.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"using JuMP\nimport Distributions\nimport HiGHS\nimport Plots\nimport StatsPlots\nimport Statistics","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Background","page":"Two-stage stochastic programs","title":"Background","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"During the week, you are a busy practitioner of Operations Research. To escape the drudgery of mathematics, you decide to open a side business selling creamy mushroom pies with puff pastry. After a few weeks, it quickly becomes apparent that operating a food business is not so easy.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The pies must be prepared in the morning, before you open for the day and can gauge the level of demand. If you bake too many, the unsold pies at the end of the day must be discarded and you have wasted time and money on their production. But if you bake too few, then there may be un-served customers and you could have made more money by baking more pies.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"After a few weeks of poor decision making, you decide to put your knowledge of Operations Research to good use, starting with some data collection.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Each pie costs you $2 to make, and you sell them at $5 each. Disposal of an unsold pie costs $0.10. Based on three weeks of data collected, in which you made 200 pies each week, you sold 150, 190, and 200 pies. Thus, as a guess, you assume a triangular distribution of demand with a minimum of 150, a median of 200, and a maximum of 250.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"We can model this problem by a two-stage stochastic program. In the first stage, we decide a quantity of pies to make x. We make this decision before we observe the demand d_omega. In the second stage, we sell y_omega pies, and incur any costs for unsold pies.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"We can formulate this problem as follows:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"beginaligned\nmaxlimits_xy_omega   -2x + mathbbE_omega5y_omega - 01(x - y_omega) \n   y_omega le x               quad forall omega in Omega \n   0 le y_omega le d_omega  quad forall omega in Omega \n   x ge 0\nendaligned","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Sample-Average-approximation","page":"Two-stage stochastic programs","title":"Sample Average approximation","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"If the distribution of demand is continuous, then our problem has an infinite number of variables and constraints. To form a computationally tractable problem, we instead use a finite set of samples drawn from the distribution. This is called sample average approximation (SAA).","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"D = Distributions.TriangularDist(150.0, 250.0, 200.0)\nN = 100\nd = sort!(rand(D, N));\nΩ = 1:N\nP = fill(1 / N, N);\nStatsPlots.histogram(d; bins = 20, label = \"\", xlabel = \"Demand\")","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#JuMP-model","page":"Two-stage stochastic programs","title":"JuMP model","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The implementation of our two-stage stochastic program in JuMP is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y[ω in Ω] <= d[ω])\n@constraint(model, [ω in Ω], y[ω] <= x)\n@expression(model, z[ω in Ω], 5y[ω] - 0.1 * (x - y[ω]))\n@objective(model, Max, -2x + sum(P[ω] * z[ω] for ω in Ω))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The optimal number of pies to make is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"value(x)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The distribution of total profit is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"total_profit = [-2 * value(x) + value(z[ω]) for ω in Ω]","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Let's plot it:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"\"\"\"\n    bin_distribution(x::Vector{Float64}, N::Int)\n\nA helper function that discretizes `x` into bins of width `N`.\n\"\"\"\nbin_distribution(x, N) = N * (floor(minimum(x) / N):ceil(maximum(x) / N))\n\nplot = StatsPlots.histogram(\n    total_profit;\n    bins = bin_distribution(total_profit, 25),\n    label = \"\",\n    xlabel = \"Profit [\\$]\",\n    ylabel = \"Number of outcomes\",\n)\nμ = Statistics.mean(total_profit)\nPlots.vline!(\n    plot,\n    [μ];\n    label = \"Expected profit (\\$$(round(Int, μ)))\",\n    linewidth = 3,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Risk-measures","page":"Two-stage stochastic programs","title":"Risk measures","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"A risk measure is a function which maps a random variable to a real number. Common risk measures include the mean (expectation), median, mode, and maximum. We need a risk measure to convert the distribution of second stage costs into a single number that can be optimized.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Our model currently uses the expectation risk measure, but others are possible too. One popular risk measure is the conditional value at risk (CVaR).","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"CVaR has a parameter gamma, and it computes the expectation of the worst gamma fraction of outcomes.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"If we are maximizing, so that small outcomes are bad, the definition of CVaR is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"CVaR_gammaZ = maxlimits_xi  xi - frac1gammamathbbE_omegaleft(xi - Z)_+right","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"which can be formulated as the linear program:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"beginaligned\nCVaR_gammaZ = maxlimits_xi z_omega   xi - frac1gammasum P_omega z_omega\n  z_omega ge xi - Z_omega  quad forall omega \n  z_omega ge 0  quad forall omega\nendaligned","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"function CVaR(Z::Vector{Float64}, P::Vector{Float64}; γ::Float64)\n    @assert 0 < γ <= 1\n    N = length(Z)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, ξ)\n    @variable(model, z[1:N] >= 0)\n    @constraint(model, [i in 1:N], z[i] >= ξ - Z[i])\n    @objective(model, Max, ξ - 1 / γ * sum(P[i] * z[i] for i in 1:N))\n    optimize!(model)\n    return objective_value(model)\nend","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"When γ is 1.0, we compute the mean of the profit:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_10 = CVaR(total_profit, P; γ = 1.0)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Statistics.mean(total_profit)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"As γ approaches 0.0, we compute the worst-case (minimum) profit:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_00 = CVaR(total_profit, P; γ = 0.0001)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"minimum(total_profit)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"By varying γ between 0 and 1 we can compute some trade-off of these two extremes:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_05 = CVaR(total_profit, P; γ = 0.5)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Let's plot these outcomes on our distribution:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"plot = StatsPlots.histogram(\n    total_profit;\n    bins = bin_distribution(total_profit, 25),\n    label = \"\",\n    xlabel = \"Profit [\\$]\",\n    ylabel = \"Number of outcomes\",\n)\nPlots.vline!(\n    plot,\n    [cvar_10 cvar_05 cvar_00];\n    label = [\"γ = 1.0\" \"γ = 0.5\" \"γ = 0.0\"],\n    linewidth = 3,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Risk-averse-sample-average-approximation","page":"Two-stage stochastic programs","title":"Risk averse sample average approximation","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Because CVaR can be formulated as a linear program, we can form a risk averse sample average approximation model by combining the two formulations:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"γ = 0.4\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y[ω in Ω] <= d[ω])\n@constraint(model, [ω in Ω], y[ω] <= x)\n@expression(model, Z[ω in Ω], 5 * y[ω] - 0.1(x - y[ω]))\n@variable(model, ξ)\n@variable(model, z[ω in Ω] >= 0)\n@constraint(model, [ω in Ω], z[ω] >= ξ - Z[ω])\n@objective(model, Max, -2x + ξ - 1 / γ * sum(P[ω] * z[ω] for ω in Ω))\noptimize!(model)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"When gamma = 04, the optimal number of pies to bake is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"value(x)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The distribution of total profit is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"risk_averse_total_profit = [value(-2x + Z[ω]) for ω in Ω]\nbins = bin_distribution([total_profit; risk_averse_total_profit], 25)\nplot = StatsPlots.histogram(total_profit; label = \"Expectation\", bins = bins)\nStatsPlots.histogram!(\n    plot,\n    risk_averse_total_profit;\n    label = \"CV@R\",\n    bins = bins,\n    alpha = 0.5,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Next-steps","page":"Two-stage stochastic programs","title":"Next steps","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Try solving this problem for different numbers of samples and different distributions.\nRefactor the example to avoid hard-coding the costs. What happens to the solution if the cost of disposing unsold pies increases?\nPlot the optimal number of pies to make for different values of the risk aversion parameter gamma. What is the relationship?","category":"page"},{"location":"packages/solvers/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"This section of the documentation contains brief documentation for some of the solvers that JuMP supports. The list of solvers is not exhaustive, but instead is intended to help you discover commonly used solvers.","category":"page"},{"location":"packages/solvers/#Affiliation","page":"Introduction","title":"Affiliation","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Packages beginning with jump-dev/ are developed and maintained by the JuMP developers. In many cases, these packages wrap external solvers that are not developed by the JuMP developers and, while the Julia packages are all open-source, in some cases the solvers themselves are closed source commercial products.","category":"page"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Packages that do not begin with jump-dev/ are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.","category":"page"},{"location":"packages/solvers/#Adding-new-solvers","page":"Introduction","title":"Adding new solvers","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Written a solver? Add it to this section of the JuMP documentation by making a pull request to the docs/packages.toml file.","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/models.md\"","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/models/#Models","page":"Models","title":"Models","text":"","category":"section"},{"location":"moi/reference/models/#Attribute-interface","page":"Models","title":"Attribute interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"is_set_by_optimize\nis_copyable\nget\nget!\nset\nsupports\nattribute_value_type","category":"page"},{"location":"moi/reference/models/#MathOptInterface.is_set_by_optimize","page":"Models","title":"MathOptInterface.is_set_by_optimize","text":"is_set_by_optimize(::AnyAttribute)\n\nReturn a Bool indicating whether the value of the attribute is modified during an optimize! call, that is, the attribute is used to query the result of the optimization.\n\nImportant note when defining new attributes\n\nThis function returns false by default so it should be implemented for attributes that are modified by optimize!.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.is_copyable","page":"Models","title":"MathOptInterface.is_copyable","text":"is_copyable(::AnyAttribute)\n\nReturn a Bool indicating whether the value of the attribute may be copied during copy_to using set.\n\nImportant note when defining new attributes\n\nBy default is_copyable(attr) returns !is_set_by_optimize(attr). A specific method should be defined for attributes which are copied indirectly during copy_to. For instance, both is_copyable and is_set_by_optimize return false for the following attributes:\n\nListOfOptimizerAttributesSet, ListOfModelAttributesSet, ListOfConstraintAttributesSet and ListOfVariableAttributesSet.\nSolverName and RawSolver: these attributes cannot be set.\nNumberOfVariables and ListOfVariableIndices: these attributes are set indirectly by add_variable and add_variables.\nObjectiveFunctionType: this attribute is set indirectly when setting the ObjectiveFunction attribute.\nNumberOfConstraints, ListOfConstraintIndices, ListOfConstraintTypesPresent, CanonicalConstraintFunction, ConstraintFunction and ConstraintSet: these attributes are set indirectly by add_constraint and add_constraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.get","page":"Models","title":"MathOptInterface.get","text":"get(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute)\n\nReturn an attribute attr of the optimizer optimizer.\n\nget(model::ModelLike, attr::AbstractModelAttribute)\n\nReturn an attribute attr of the model model.\n\nget(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex)\n\nIf the attribute attr is set for the variable v in the model model, return its value, return nothing otherwise. If the attribute attr is not supported by model then an error should be thrown instead of returning nothing.\n\nget(model::ModelLike, attr::AbstractVariableAttribute, v::Vector{VariableIndex})\n\nReturn a vector of attributes corresponding to each variable in the collection v in the model model.\n\nget(model::ModelLike, attr::AbstractConstraintAttribute, c::ConstraintIndex)\n\nIf the attribute attr is set for the constraint c in the model model, return its value, return nothing otherwise. If the attribute attr is not supported by model then an error should be thrown instead of returning nothing.\n\nget(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::Vector{ConstraintIndex{F,S}},\n) where {F,S}\n\nReturn a vector of attributes corresponding to each constraint in the collection c in the model model.\n\nget(model::ModelLike, ::Type{VariableIndex}, name::String)\n\nIf a variable with name name exists in the model model, return the corresponding index, otherwise return nothing. Errors if two variables have the same name.\n\nget(\n    model::ModelLike,\n    ::Type{ConstraintIndex{F,S}},\n    name::String,\n) where {F,S}\n\nIf an F-in-S constraint with name name exists in the model model, return the corresponding index, otherwise return nothing. Errors if two constraints have the same name.\n\nget(model::ModelLike, ::Type{ConstraintIndex}, name::String)\n\nIf any constraint with name name exists in the model model, return the corresponding index, otherwise return nothing. This version is available for convenience but may incur a performance penalty because it is not type stable. Errors if two constraints have the same name.\n\n\n\n\n\nget(model::GenericModel, attr::MathOptInterface.AbstractOptimizerAttribute)\n\nReturn the value of the attribute attr from the model's MOI backend.\n\n\n\n\n\nget(model::GenericModel, attr::MathOptInterface.AbstractModelAttribute)\n\nReturn the value of the attribute attr from the model's MOI backend.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64\n\nReturn the number of variables created by the bridge b in the model.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)\n\nReturn the list of variables created by the bridge b.\n\nSee also MOI.ListOfVariableIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}\n\nReturn the number of constraints of the type F-in-S created by the bridge b.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}\n\nReturn a  Vector{ConstraintIndex{F,S}} with indices of all constraints of type F-in-S created by the bride b.\n\nSee also MOI.ListOfConstraintIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\nfunction MOI.get(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    bridge::AbstractBridge,\n)\n\nReturn the value of the attribute attr of the model model for the constraint bridged by bridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.get!","page":"Models","title":"MathOptInterface.get!","text":"get!(output, model::ModelLike, args...)\n\nAn in-place version of get.\n\nThe signature matches that of get except that the the result is placed in the vector output.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.set","page":"Models","title":"MathOptInterface.set","text":"set(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute, value)\n\nAssign value to the attribute attr of the optimizer optimizer.\n\nset(model::ModelLike, attr::AbstractModelAttribute, value)\n\nAssign value to the attribute attr of the model model.\n\nset(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex, value)\n\nAssign value to the attribute attr of variable v in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractVariableAttribute,\n    v::Vector{VariableIndex},\n    vector_of_values,\n)\n\nAssign a value respectively to the attribute attr of each variable in the collection v in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::ConstraintIndex,\n    value,\n)\n\nAssign a value to the attribute attr of constraint c in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::Vector{ConstraintIndex{F,S}},\n    vector_of_values,\n) where {F,S}\n\nAssign a value respectively to the attribute attr of each constraint in the collection c in model model.\n\nAn UnsupportedAttribute error is thrown if model does not support the attribute attr (see supports) and a SetAttributeNotAllowed error is thrown if it supports the attribute attr but it cannot be set.\n\nset(\n    model::ModelLike,\n    ::ConstraintSet,\n    c::ConstraintIndex{F,S},\n    set::S,\n) where {F,S}\n\nChange the set of constraint c to the new set set which should be of the same type as the original set.\n\nset(\n    model::ModelLike,\n    ::ConstraintFunction,\n    c::ConstraintIndex{F,S},\n    func::F,\n) where {F,S}\n\nReplace the function in constraint c with func. F must match the original function type used to define the constraint.\n\nnote: Note\nSetting the constraint function is not allowed if F is VariableIndex; a SettingVariableIndexNotAllowed error is thrown instead. This is because, it would require changing the index c since the index of VariableIndex constraints must be the same as the index of the variable.\n\n\n\n\n\nfunction MOI.set(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    bridge::AbstractBridge,\n    value,\n)\n\nSet the value of the attribute attr of the model model for the constraint bridged by bridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.supports","page":"Models","title":"MathOptInterface.supports","text":"supports(model::ModelLike, sub::AbstractSubmittable)::Bool\n\nReturn a Bool indicating whether model supports the submittable sub.\n\nsupports(model::ModelLike, attr::AbstractOptimizerAttribute)::Bool\n\nReturn a Bool indicating whether model supports the optimizer attribute attr. That is, it returns false if copy_to(model, src) shows a warning in case attr is in the ListOfOptimizerAttributesSet of src; see copy_to for more details on how unsupported optimizer attributes are handled in copy.\n\nsupports(model::ModelLike, attr::AbstractModelAttribute)::Bool\n\nReturn a Bool indicating whether model supports the model attribute attr. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfModelAttributesSet of src.\n\nsupports(\n    model::ModelLike,\n    attr::AbstractVariableAttribute,\n    ::Type{VariableIndex},\n)::Bool\n\nReturn a Bool indicating whether model supports the variable attribute attr. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfVariableAttributesSet of src.\n\nsupports(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    ::Type{ConstraintIndex{F,S}},\n)::Bool where {F,S}\n\nReturn a Bool indicating whether model supports the constraint attribute attr applied to an F-in-S constraint. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfConstraintAttributesSet of src.\n\nFor all five methods, if the attribute is only not supported in specific circumstances, it should still return true.\n\nNote that supports is only defined for attributes for which is_copyable returns true as other attributes do not appear in the list of attributes set obtained by ListOf...AttributesSet.\n\n\n\n\n\nMOI.supports(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    BT::Type{<:AbstractBridge},\n)\n\nReturn a Bool indicating whether BT supports setting attr to model.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.attribute_value_type","page":"Models","title":"MathOptInterface.attribute_value_type","text":"attribute_value_type(attr::AnyAttribute)\n\nGiven an attribute attr, return the type of value expected by get, or returned by set.\n\nNotes\n\nOnly implement this if it make sense to do so. If un-implemented, the default is Any.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#Model-interface","page":"Models","title":"Model interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"ModelLike\nis_empty\nempty!\nwrite_to_file\nread_from_file\nsupports_incremental_interface\ncopy_to\nIndexMap","category":"page"},{"location":"moi/reference/models/#MathOptInterface.ModelLike","page":"Models","title":"MathOptInterface.ModelLike","text":"ModelLike\n\nAbstract supertype for objects that implement the \"Model\" interface for defining an optimization problem.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.is_empty","page":"Models","title":"MathOptInterface.is_empty","text":"is_empty(model::ModelLike)\n\nReturns false if the model has any model attribute set or has any variables or constraints.\n\nNote that an empty model can have optimizer attributes set.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.empty!","page":"Models","title":"MathOptInterface.empty!","text":"empty!(model::ModelLike)\n\nEmpty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.write_to_file","page":"Models","title":"MathOptInterface.write_to_file","text":"write_to_file(model::ModelLike, filename::String)\n\nWrite the current model to the file at filename.\n\nSupported file types depend on the model type.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.read_from_file","page":"Models","title":"MathOptInterface.read_from_file","text":"read_from_file(model::ModelLike, filename::String)\n\nRead the file filename into the model model. If model is non-empty, this may throw an error.\n\nSupported file types depend on the model type.\n\nNote\n\nOnce the contents of the file are loaded into the model, users can query the variables via get(model, ListOfVariableIndices()). However, some filetypes, such as LP files, do not maintain an explicit ordering of the variables. Therefore, the returned list may be in an arbitrary order.\n\nTo avoid depending on the order of the indices, look up each variable index by name using  get(model, VariableIndex, \"name\").\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.supports_incremental_interface","page":"Models","title":"MathOptInterface.supports_incremental_interface","text":"supports_incremental_interface(model::ModelLike)\n\nReturn a Bool indicating whether model supports building incrementally via add_variable and add_constraint.\n\nThe main purpose of this function is to determine whether a model can be loaded into model incrementally or whether it should be cached and copied at once instead.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.copy_to","page":"Models","title":"MathOptInterface.copy_to","text":"copy_to(dest::ModelLike, src::ModelLike)::IndexMap\n\nCopy the model from src into dest.\n\nThe target dest is emptied, and all previous indices to variables and constraints in dest are invalidated.\n\nReturns an IndexMap object that translates variable and constraint indices from the src model to the corresponding indices in the dest model.\n\nNotes\n\nIf a constraint that in src is not supported by dest, then an UnsupportedConstraint error is thrown.\nIf an AbstractModelAttribute, AbstractVariableAttribute, or AbstractConstraintAttribute is set in src but not supported by dest, then an UnsupportedAttribute error is thrown.\n\nAbstractOptimizerAttributes are not copied  to the dest model.\n\nIndexMap\n\nImplementations of copy_to must return an IndexMap. For technical reasons, this type is defined in the Utilities submodule as MOI.Utilities.IndexMap. However, since it is an integral part of the MOI API, we provide MOI.IndexMap as an alias.\n\nExample\n\n# Given empty `ModelLike` objects `src` and `dest`.\n\nx = add_variable(src)\n\nis_valid(src, x)   # true\nis_valid(dest, x)  # false (`dest` has no variables)\n\nindex_map = copy_to(dest, src)\nis_valid(dest, x) # false (unless index_map[x] == x)\nis_valid(dest, index_map[x]) # true\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.IndexMap","page":"Models","title":"MathOptInterface.IndexMap","text":"IndexMap()\n\nThe dictionary-like object returned by copy_to.\n\nIndexMap\n\nImplementations of copy_to must return an IndexMap. For technical reasons, the IndexMap type is defined in the Utilities submodule as MOI.Utilities.IndexMap. However, since it is an integral part of the MOI API, we provide this MOI.IndexMap as an alias.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#Model-attributes","page":"Models","title":"Model attributes","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractModelAttribute\nName\nObjectiveFunction\nObjectiveFunctionType\nObjectiveSense\nOptimizationSense\nMIN_SENSE\nMAX_SENSE\nFEASIBILITY_SENSE\nNumberOfVariables\nListOfVariableIndices\nListOfConstraintTypesPresent\nNumberOfConstraints\nListOfConstraintIndices\nListOfOptimizerAttributesSet\nListOfModelAttributesSet\nListOfVariableAttributesSet\nListOfConstraintAttributesSet\nUserDefinedFunction\nListOfSupportedNonlinearOperators","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractModelAttribute","page":"Models","title":"MathOptInterface.AbstractModelAttribute","text":"AbstractModelAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.Name","page":"Models","title":"MathOptInterface.Name","text":"Name()\n\nA model attribute for the string identifying the model. It has a default value of \"\" if not set`.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveFunction","page":"Models","title":"MathOptInterface.ObjectiveFunction","text":"ObjectiveFunction{F<:AbstractScalarFunction}()\n\nA model attribute for the objective function which has a type F<:AbstractScalarFunction.\n\nF should be guaranteed to be equivalent but not necessarily identical to the function type provided by the user.\n\nThrows an InexactError if the objective function cannot be converted to F, e.g., the objective function is quadratic and F is ScalarAffineFunction{Float64} or it has non-integer coefficient and F is ScalarAffineFunction{Int}.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveFunctionType","page":"Models","title":"MathOptInterface.ObjectiveFunctionType","text":"ObjectiveFunctionType()\n\nA model attribute for the type F of the objective function set using the ObjectiveFunction{F} attribute.\n\nExamples\n\nIn the following code, attr should be equal to MOI.VariableIndex:\n\nx = MOI.add_variable(model)\nMOI.set(model, MOI.ObjectiveFunction{MOI.VariableIndex}(), x)\nattr = MOI.get(model, MOI.ObjectiveFunctionType())\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveSense","page":"Models","title":"MathOptInterface.ObjectiveSense","text":"ObjectiveSense()\n\nA model attribute for the objective sense of the objective function, which must be an OptimizationSense: MIN_SENSE, MAX_SENSE, or FEASIBILITY_SENSE. The default is FEASIBILITY_SENSE.\n\nInteraction with ObjectiveFunction\n\nSetting the sense to FEASIBILITY_SENSE unsets the ObjectiveFunction attribute. That is, if you first set ObjectiveFunction and then set ObjectiveSense to be FEASIBILITY_SENSE, no objective function will be passed to the solver.\n\nIn addition, some reformulations of ObjectiveFunction via bridges rely on the value of ObjectiveSense. Therefore, you should set ObjectiveSense before setting ObjectiveFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OptimizationSense","page":"Models","title":"MathOptInterface.OptimizationSense","text":"OptimizationSense\n\nAn enum for the value of the ObjectiveSense attribute.\n\nValues\n\nPossible values are:\n\nMIN_SENSE: the goal is to minimize the objective function\nMAX_SENSE: the goal is to maximize the objective function\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.MIN_SENSE","page":"Models","title":"MathOptInterface.MIN_SENSE","text":"MIN_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMIN_SENSE: the goal is to minimize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MAX_SENSE","page":"Models","title":"MathOptInterface.MAX_SENSE","text":"MAX_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMAX_SENSE: the goal is to maximize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.FEASIBILITY_SENSE","page":"Models","title":"MathOptInterface.FEASIBILITY_SENSE","text":"FEASIBILITY_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NumberOfVariables","page":"Models","title":"MathOptInterface.NumberOfVariables","text":"NumberOfVariables()\n\nA model attribute for the number of variables in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfVariableIndices","page":"Models","title":"MathOptInterface.ListOfVariableIndices","text":"ListOfVariableIndices()\n\nA model attribute for the Vector{VariableIndex} of all variable indices present in the model (i.e., of length equal to the value of NumberOfVariables in the order in which they were added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintTypesPresent","page":"Models","title":"MathOptInterface.ListOfConstraintTypesPresent","text":"ListOfConstraintTypesPresent()\n\nA model attribute for the list of tuples of the form (F,S), where F is a function type and S is a set type indicating that the attribute NumberOfConstraints{F,S} has a value greater than zero.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NumberOfConstraints","page":"Models","title":"MathOptInterface.NumberOfConstraints","text":"NumberOfConstraints{F,S}()\n\nA model attribute for the number of constraints of the type F-in-S present in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintIndices","page":"Models","title":"MathOptInterface.ListOfConstraintIndices","text":"ListOfConstraintIndices{F,S}()\n\nA model attribute for the Vector{ConstraintIndex{F,S}} of all constraint indices of type F-in-S in the model (i.e., of length equal to the value of NumberOfConstraints{F,S}) in the order in which they were added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfOptimizerAttributesSet","page":"Models","title":"MathOptInterface.ListOfOptimizerAttributesSet","text":"ListOfOptimizerAttributesSet()\n\nAn optimizer attribute for the Vector{AbstractOptimizerAttribute} of all optimizer attributes that were set.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfModelAttributesSet","page":"Models","title":"MathOptInterface.ListOfModelAttributesSet","text":"ListOfModelAttributesSet()\n\nA model attribute for the Vector{AbstractModelAttribute} of all model attributes attr such that:\n\nis_copyable(attr) returns true, and\nthe attribute was set to the model\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfVariableAttributesSet","page":"Models","title":"MathOptInterface.ListOfVariableAttributesSet","text":"ListOfVariableAttributesSet()\n\nA model attribute for the Vector{AbstractVariableAttribute} of all variable attributes attr such that 1) is_copyable(attr) returns true and 2) the attribute was set to variables.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintAttributesSet","page":"Models","title":"MathOptInterface.ListOfConstraintAttributesSet","text":"ListOfConstraintAttributesSet{F, S}()\n\nA model attribute for the Vector{AbstractConstraintAttribute} of all constraint attributes attr such that:\n\nis_copyable(attr) returns true and\nthe attribute was set to F-in-S constraints.\n\nNote\n\nThe attributes ConstraintFunction and ConstraintSet should not be included in the list even if then have been set with set.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.UserDefinedFunction","page":"Models","title":"MathOptInterface.UserDefinedFunction","text":"UserDefinedFunction(name::Symbol, arity::Int) <: AbstractModelAttribute\n\nSet this attribute to register a user-defined function by the name of name with arity arguments.\n\nOnce registered, name will appear in ListOfSupportedNonlinearOperators.\n\nYou cannot register multiple UserDefinedFunctions with the same name but different arity.\n\nValue type\n\nThe value to be set is a tuple containing one, two, or three functions to evaluate the function, the first-order derivative, and the second-order derivative respectively. Both derivatives are optional, but if you pass the second-order derivative you must also pass the first-order derivative.\n\nFor univariate functions with arity == 1, the functions in the tuple must have the form:\n\nf(x::T)::T: returns the value of the function at x\n∇f(x::T)::T: returns the first-order derivative of f with respect to x\n∇²f(x::T)::T: returns the second-order derivative of f with respect to x.\n\nFor multivariate functions with arity > 1, the functions in the tuple must have the form:\n\nf(x::T...)::T: returns the value of the function at x\n∇f(g::AbstractVector{T}, x::T...)::Nothing: fills the components of g, with g[i] being the first-order partial derivative of f with respect to x[i]\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing: fills the non-zero components  of H, with H[i, j] being the second-order partial derivative of f with  respect to x[i] and then x[j]. H is initialized to the zero matrix,  so you do not need to set any zero elements.\n\nExamples\n\njulia> import MathOptInterface as MOI\n\njulia> f(x, y) = x^2 + y^2\nf (generic function with 1 method)\n\njulia> function ∇f(g, x, y)\n           g .= 2 * x, 2 * y\n           return\n       end\n∇f (generic function with 1 method)\n\njulia> function ∇²f(H, x...)\n           H[1, 1] = H[2, 2] = 2.0\n           return\n       end\n∇²f (generic function with 1 method)\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:f, 2), (f,))\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:g, 2), (f, ∇f))\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:h, 2), (f, ∇f, ∇²f))\n\njulia> x = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n\njulia> obj_f = MOI.ScalarNonlinearFunction(:f, Any[x[1], x[2]])\nf(MOI.VariableIndex(1), MOI.VariableIndex(2))\n\njulia> MOI.set(model, MOI.ObjectiveFunction{typeof(obj_f)}(), obj_f)\n\njulia> print(model)\nMinimize ScalarNonlinearFunction:\n f(v[1], v[2])\n\nSubject to:\n\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfSupportedNonlinearOperators","page":"Models","title":"MathOptInterface.ListOfSupportedNonlinearOperators","text":"ListOfSupportedNonlinearOperators() <: AbstractModelAttribute\n\nWhen queried with get, return a Vector{Symbol} listing the operators supported by the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#Optimizer-interface","page":"Models","title":"Optimizer interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractOptimizer\nOptimizerWithAttributes\noptimize!\noptimize!(::ModelLike, ::ModelLike)\ninstantiate\ndefault_cache","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractOptimizer","page":"Models","title":"MathOptInterface.AbstractOptimizer","text":"AbstractOptimizer <: ModelLike\n\nAbstract supertype for objects representing an instance of an optimization problem tied to a particular solver. This is typically a solver's in-memory representation. In addition to ModelLike, AbstractOptimizer objects let you solve the model and query the solution.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OptimizerWithAttributes","page":"Models","title":"MathOptInterface.OptimizerWithAttributes","text":"struct OptimizerWithAttributes\n    optimizer_constructor\n    params::Vector{Pair{AbstractOptimizerAttribute,<:Any}}\nend\n\nObject grouping an optimizer constructor and a list of optimizer attributes. Instances are created with instantiate.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.optimize!","page":"Models","title":"MathOptInterface.optimize!","text":"optimize!(optimizer::AbstractOptimizer)\n\nOptimize the problem contained in optimizer.\n\nBefore calling optimize!, the problem should first be constructed using the incremental interface (see supports_incremental_interface) or copy_to.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.optimize!-Tuple{MathOptInterface.ModelLike, MathOptInterface.ModelLike}","page":"Models","title":"MathOptInterface.optimize!","text":"optimize!(dest::AbstractOptimizer, src::ModelLike)::Tuple{IndexMap,Bool}\n\nA \"one-shot\" call that copies the problem from src into dest and then uses dest to optimize the problem.\n\nReturns a tuple of an IndexMap and a Bool copied.\n\nThe IndexMap object translates variable and constraint indices from the src model to the corresponding indices in the dest optimizer. See copy_to for details.\nIf copied == true, src was copied to dest and then cached, allowing incremental modification if supported by the solver.\nIf copied == false, a cache of the model was not kept in dest. Therefore, only the solution information (attributes for which is_set_by_optimize is true) is available to query.\n\nnote: Note\nThe main purpose of optimize! method with two arguments is for use in Utilities.CachingOptimizer.\n\nRelationship to the single-argument optimize!\n\nThe default fallback of optimize!(dest::AbstractOptimizer, src::ModelLike) is\n\nfunction optimize!(dest::AbstractOptimizer, src::ModelLike)\n    index_map = copy_to(dest, src)\n    optimize!(dest)\n    return index_map, true\nend\n\nTherefore, subtypes of AbstractOptimizer should either implement this two-argument method, or implement both copy_to(::Optimizer, ::ModelLike) and optimize!(::Optimizer).\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/models/#MathOptInterface.instantiate","page":"Models","title":"MathOptInterface.instantiate","text":"instantiate(\n    optimizer_constructor,\n    with_cache_type::Union{Nothing,Type} = nothing,\n    with_bridge_type::Union{Nothing,Type} = nothing,\n)\n\nCreate an instance of an optimizer by either:\n\ncalling optimizer_constructor.optimizer_constructor() and setting the parameters in optimizer_constructor.params if optimizer_constructor is a OptimizerWithAttributes\ncalling optimizer_constructor() if optimizer_constructor is callable.\n\nwithcachetype\n\nIf with_cache_type is not nothing, then the optimizer is wrapped in a Utilities.CachingOptimizer to store a cache of the model. This is most useful if the optimizer you are constructing does not support the incremental interface (see supports_incremental_interface).\n\nwithbridgetype\n\nIf with_bridge_type is not nothing, the optimizer is wrapped in a Bridges.full_bridge_optimizer, enabling all the bridges defined in the MOI.Bridges submodule with coefficient type with_bridge_type.\n\nIn addition, if the optimizer created by optimizer_constructor does not support the incremental interface (see supports_incremental_interface), then, irrespective of with_cache_type, the optimizer is wrapped in a Utilities.CachingOptimizer to store a cache of the bridged model.\n\nIf with_cache_type and with_bridge_type are both not nothing, then they must be the same type.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.default_cache","page":"Models","title":"MathOptInterface.default_cache","text":"default_cache(optimizer::ModelLike, ::Type{T}) where {T}\n\nReturn a new instance of the default model type to be used as cache for optimizer in a Utilities.CachingOptimizer for holding constraints of coefficient type T. By default, this returns Utilities.UniversalFallback(Utilities.Model{T}()). If copying from a instance of a given model type is faster for optimizer then a new method returning an instance of this model type should be defined.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#Optimizer-attributes","page":"Models","title":"Optimizer attributes","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractOptimizerAttribute\nSolverName\nSolverVersion\nSilent\nTimeLimitSec\nObjectiveLimit\nRawOptimizerAttribute\nNumberOfThreads\nRawSolver\nAbsoluteGapTolerance\nRelativeGapTolerance","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractOptimizerAttribute","page":"Models","title":"MathOptInterface.AbstractOptimizerAttribute","text":"AbstractOptimizerAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of the optimizer.\n\nNotes\n\nThe difference between AbstractOptimizerAttribute and AbstractModelAttribute lies in the behavior of is_empty, empty! and copy_to. Typically optimizer attributes affect only how the model is solved.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolverName","page":"Models","title":"MathOptInterface.SolverName","text":"SolverName()\n\nAn optimizer attribute for the string identifying the solver/optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolverVersion","page":"Models","title":"MathOptInterface.SolverVersion","text":"SolverVersion()\n\nAn optimizer attribute for the string identifying the version of the solver.\n\nnote: Note\nFor solvers supporting semantic versioning, the SolverVersion should be a string of the form \"vMAJOR.MINOR.PATCH\", so that it can be converted to a Julia VersionNumber (e.g., `VersionNumber(\"v1.2.3\")).We do not require Semantic Versioning because some solvers use alternate versioning systems. For example, CPLEX uses Calendar Versioning, so SolverVersion will return a string like \"202001\".\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.Silent","page":"Models","title":"MathOptInterface.Silent","text":"Silent()\n\nAn optimizer attribute for silencing the output of an optimizer. When set to true, it takes precedence over any other attribute controlling verbosity and requires the solver to produce no output. The default value is false which has no effect. In this case the verbosity is controlled by other attributes.\n\nNote\n\nEvery optimizer should have verbosity on by default. For instance, if a solver has a solver-specific log level attribute, the MOI implementation should set it to 1 by default. If the user sets Silent to true, then the log level should be set to 0, even if the user specifically sets a value of log level. If the value of Silent is false then the log level set to the solver is the value given by the user for this solver-specific parameter or 1 if none is given.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.TimeLimitSec","page":"Models","title":"MathOptInterface.TimeLimitSec","text":"TimeLimitSec()\n\nAn optimizer attribute for setting a time limit (in seconnds) for an optimization. When set to nothing, it deactivates the solver time limit. The default value is nothing.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveLimit","page":"Models","title":"MathOptInterface.ObjectiveLimit","text":"ObjectiveLimit()\n\nAn optimizer attribute for setting a limit on the objective value.\n\nThe provided limit must be a Union{Real,Nothing}.\n\nWhen set to nothing, the limit reverts to the solver's default.\n\nThe default value is nothing.\n\nThe solver may stop when the ObjectiveValue is better (lower for minimization, higher for maximization) than the ObjectiveLimit. If stopped, the TerminationStatus should be OBJECTIVE_LIMIT.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawOptimizerAttribute","page":"Models","title":"MathOptInterface.RawOptimizerAttribute","text":"RawOptimizerAttribute(name::String)\n\nAn optimizer attribute for the solver-specific parameter identified by name.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NumberOfThreads","page":"Models","title":"MathOptInterface.NumberOfThreads","text":"NumberOfThreads()\n\nAn optimizer attribute for setting the number of threads used for an optimization. When set to nothing uses solver default. Values are positive integers. The default value is nothing.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawSolver","page":"Models","title":"MathOptInterface.RawSolver","text":"RawSolver()\n\nA model attribute for the object that may be used to access a solver-specific API for this optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.AbsoluteGapTolerance","page":"Models","title":"MathOptInterface.AbsoluteGapTolerance","text":"AbsoluteGapTolerance()\n\nAn optimizer attribute for setting the absolute gap tolerance for an optimization. This is an optimizer attribute, and should be set before calling optimize!. When set to nothing (if supported), uses solver default.\n\nTo set a relative gap tolerance, see RelativeGapTolerance.\n\nwarning: Warning\nThe mathematical definition of \"absolute gap\", and its treatment during the optimization, are solver-dependent. However, assuming no other limit nor issue is encountered during the optimization, most solvers that implement this attribute will stop once f - b  g_abs, where b is the best bound, f is the best feasible objective value, and g_abs is the absolute gap.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RelativeGapTolerance","page":"Models","title":"MathOptInterface.RelativeGapTolerance","text":"RelativeGapTolerance()\n\nAn optimizer attribute for setting the relative gap tolerance for an optimization. This is an optimizer attribute, and should be set before calling optimize!. When set to nothing (if supported), uses solver default.\n\nIf you are looking for the relative gap of the current best solution, see RelativeGap. If no limit nor issue is encountered during the optimization, the value of RelativeGap should be at most as large as RelativeGapTolerance.\n\n# Before optimizing: set relative gap tolerance\n# set 0.1% relative gap tolerance\nMOI.set(model, MOI.RelativeGapTolerance(), 1e-3)\nMOI.optimize!(model)\n\n# After optimizing (assuming all went well)\n# The relative gap tolerance has not changed...\nMOI.get(model, MOI.RelativeGapTolerance())  # returns 1e-3\n# ... and the relative gap of the obtained solution is smaller or equal to the\n# tolerance\nMOI.get(model, MOI.RelativeGap())  # should return something ≤ 1e-3\n\nwarning: Warning\nThe mathematical definition of \"relative gap\", and its allowed range, are solver-dependent. Typically, solvers expect a value between 0.0 and 1.0.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"List of attributes useful for optimizers","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"TerminationStatus\nTerminationStatusCode\nOPTIMIZE_NOT_CALLED\nOPTIMAL\nINFEASIBLE\nDUAL_INFEASIBLE\nLOCALLY_SOLVED\nLOCALLY_INFEASIBLE\nINFEASIBLE_OR_UNBOUNDED\nALMOST_OPTIMAL\nALMOST_INFEASIBLE\nALMOST_DUAL_INFEASIBLE\nALMOST_LOCALLY_SOLVED\nITERATION_LIMIT\nTIME_LIMIT\nNODE_LIMIT\nSOLUTION_LIMIT\nMEMORY_LIMIT\nOBJECTIVE_LIMIT\nNORM_LIMIT\nOTHER_LIMIT\nSLOW_PROGRESS\nNUMERICAL_ERROR\nINVALID_MODEL\nINVALID_OPTION\nINTERRUPTED\nOTHER_ERROR\nPrimalStatus\nDualStatus\nRawStatusString\nResultCount\nObjectiveValue\nDualObjectiveValue\nObjectiveBound\nRelativeGap\nSolveTimeSec\nSimplexIterations\nBarrierIterations\nNodeCount","category":"page"},{"location":"moi/reference/models/#MathOptInterface.TerminationStatus","page":"Models","title":"MathOptInterface.TerminationStatus","text":"TerminationStatus()\n\nA model attribute for the TerminationStatusCode explaining why the optimizer stopped.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.TerminationStatusCode","page":"Models","title":"MathOptInterface.TerminationStatusCode","text":"TerminationStatusCode\n\nAn Enum of possible values for the TerminationStatus attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to optimize!.\n\nValues\n\nPossible values are:\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\nOPTIMAL: The algorithm found a globally optimal solution.\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\nINVALID_MODEL: The algorithm stopped because the model is invalid.\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OPTIMIZE_NOT_CALLED","page":"Models","title":"MathOptInterface.OPTIMIZE_NOT_CALLED","text":"OPTIMIZE_NOT_CALLED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OPTIMAL","page":"Models","title":"MathOptInterface.OPTIMAL","text":"OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMAL: The algorithm found a globally optimal solution.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE","page":"Models","title":"MathOptInterface.INFEASIBLE","text":"INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.DUAL_INFEASIBLE","page":"Models","title":"MathOptInterface.DUAL_INFEASIBLE","text":"DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.LOCALLY_SOLVED","page":"Models","title":"MathOptInterface.LOCALLY_SOLVED","text":"LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.LOCALLY_INFEASIBLE","page":"Models","title":"MathOptInterface.LOCALLY_INFEASIBLE","text":"LOCALLY_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE_OR_UNBOUNDED","page":"Models","title":"MathOptInterface.INFEASIBLE_OR_UNBOUNDED","text":"INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_OPTIMAL","page":"Models","title":"MathOptInterface.ALMOST_OPTIMAL","text":"ALMOST_OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_INFEASIBLE","page":"Models","title":"MathOptInterface.ALMOST_INFEASIBLE","text":"ALMOST_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_DUAL_INFEASIBLE","page":"Models","title":"MathOptInterface.ALMOST_DUAL_INFEASIBLE","text":"ALMOST_DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_LOCALLY_SOLVED","page":"Models","title":"MathOptInterface.ALMOST_LOCALLY_SOLVED","text":"ALMOST_LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ITERATION_LIMIT","page":"Models","title":"MathOptInterface.ITERATION_LIMIT","text":"ITERATION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.TIME_LIMIT","page":"Models","title":"MathOptInterface.TIME_LIMIT","text":"TIME_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NODE_LIMIT","page":"Models","title":"MathOptInterface.NODE_LIMIT","text":"NODE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.SOLUTION_LIMIT","page":"Models","title":"MathOptInterface.SOLUTION_LIMIT","text":"SOLUTION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MEMORY_LIMIT","page":"Models","title":"MathOptInterface.MEMORY_LIMIT","text":"MEMORY_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OBJECTIVE_LIMIT","page":"Models","title":"MathOptInterface.OBJECTIVE_LIMIT","text":"OBJECTIVE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NORM_LIMIT","page":"Models","title":"MathOptInterface.NORM_LIMIT","text":"NORM_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_LIMIT","page":"Models","title":"MathOptInterface.OTHER_LIMIT","text":"OTHER_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.SLOW_PROGRESS","page":"Models","title":"MathOptInterface.SLOW_PROGRESS","text":"SLOW_PROGRESS::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NUMERICAL_ERROR","page":"Models","title":"MathOptInterface.NUMERICAL_ERROR","text":"NUMERICAL_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INVALID_MODEL","page":"Models","title":"MathOptInterface.INVALID_MODEL","text":"INVALID_MODEL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_MODEL: The algorithm stopped because the model is invalid.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INVALID_OPTION","page":"Models","title":"MathOptInterface.INVALID_OPTION","text":"INVALID_OPTION::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INTERRUPTED","page":"Models","title":"MathOptInterface.INTERRUPTED","text":"INTERRUPTED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_ERROR","page":"Models","title":"MathOptInterface.OTHER_ERROR","text":"OTHER_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.PrimalStatus","page":"Models","title":"MathOptInterface.PrimalStatus","text":"PrimalStatus(result_index::Int = 1)\n\nA model attribute for the ResultStatusCode of the primal result result_index. If result_index is omitted, it defaults to 1.\n\nSee ResultCount for information on how the results are ordered.\n\nIf result_index is larger than the value of ResultCount then NO_SOLUTION is returned.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.DualStatus","page":"Models","title":"MathOptInterface.DualStatus","text":"DualStatus(result_index::Int = 1)\n\nA model attribute for the ResultStatusCode of the dual result result_index. If result_index is omitted, it defaults to 1.\n\nSee ResultCount for information on how the results are ordered.\n\nIf result_index is larger than the value of ResultCount then NO_SOLUTION is returned.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawStatusString","page":"Models","title":"MathOptInterface.RawStatusString","text":"RawStatusString()\n\nA model attribute for a solver specific string explaining why the optimizer stopped.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ResultCount","page":"Models","title":"MathOptInterface.ResultCount","text":"ResultCount()\n\nA model attribute for the number of results available.\n\nOrder of solutions\n\nA number of attributes contain an index, result_index, which is used to refer to one of the available results. Thus, result_index must be an integer between 1 and the number of available results.\n\nAs a general rule, the first result (result_index=1) is the most important result (e.g., an optimal solution or an infeasibility certificate). Other results will typically be alternate solutions that the solver found during the search for the first result.\n\nIf a (local) optimal solution is available, i.e., TerminationStatus is OPTIMAL or LOCALLY_SOLVED, the first result must correspond to the (locally) optimal solution. Other results may be alternative optimal solutions, or they may be other suboptimal solutions; use ObjectiveValue to distingiush between them.\n\nIf a primal or dual infeasibility certificate is available, i.e., TerminationStatus is INFEASIBLE or DUAL_INFEASIBLE and the corresponding PrimalStatus or DualStatus is INFEASIBILITY_CERTIFICATE, then the first result must be a certificate. Other results may be alternate certificates, or infeasible points.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveValue","page":"Models","title":"MathOptInterface.ObjectiveValue","text":"ObjectiveValue(result_index::Int = 1)\n\nA model attribute for the objective value of the primal solution result_index.\n\nIf the solver does not have a primal value for the objective because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ObjectiveValue attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.DualObjectiveValue","page":"Models","title":"MathOptInterface.DualObjectiveValue","text":"DualObjectiveValue(result_index::Int = 1)\n\nA model attribute for the value of the objective function of the dual problem for the result_indexth dual result.\n\nIf the solver does not have a dual value for the objective because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check DualStatus before accessing the DualObjectiveValue attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveBound","page":"Models","title":"MathOptInterface.ObjectiveBound","text":"ObjectiveBound()\n\nA model attribute for the best known bound on the optimal objective value.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RelativeGap","page":"Models","title":"MathOptInterface.RelativeGap","text":"RelativeGap()\n\nA model attribute for the final relative optimality gap.\n\nwarning: Warning\nThe definition of this gap is solver-dependent. However, most solvers implementing this attribute define the relative gap as some variation of fracb-ff, where b is the best bound and f is the best feasible objective value.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolveTimeSec","page":"Models","title":"MathOptInterface.SolveTimeSec","text":"SolveTimeSec()\n\nA model attribute for the total elapsed solution time (in seconds) as reported by the optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SimplexIterations","page":"Models","title":"MathOptInterface.SimplexIterations","text":"SimplexIterations()\n\nA model attribute for the cumulative number of simplex iterations during the optimization process.\n\nFor a mixed-integer program (MIP), the return value is the total simplex iterations for all nodes.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.BarrierIterations","page":"Models","title":"MathOptInterface.BarrierIterations","text":"BarrierIterations()\n\nA model attribute for the cumulative number of barrier iterations while solving a problem.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NodeCount","page":"Models","title":"MathOptInterface.NodeCount","text":"NodeCount()\n\nA model attribute for the total number of branch-and-bound nodes explored while solving a mixed-integer program (MIP).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#ResultStatusCode","page":"Models","title":"ResultStatusCode","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"ResultStatusCode\nNO_SOLUTION\nFEASIBLE_POINT\nNEARLY_FEASIBLE_POINT\nINFEASIBLE_POINT\nINFEASIBILITY_CERTIFICATE\nNEARLY_INFEASIBILITY_CERTIFICATE\nREDUCTION_CERTIFICATE\nNEARLY_REDUCTION_CERTIFICATE\nUNKNOWN_RESULT_STATUS\nOTHER_RESULT_STATUS","category":"page"},{"location":"moi/reference/models/#MathOptInterface.ResultStatusCode","page":"Models","title":"MathOptInterface.ResultStatusCode","text":"ResultStatusCode\n\nAn Enum of possible values for the PrimalStatus and DualStatus attributes.\n\nThe values indicate how to interpret the result vector.\n\nValues\n\nPossible values are:\n\nNO_SOLUTION: the result vector is empty.\nFEASIBLE_POINT: the result vector is a feasible point.\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\nINFEASIBLE_POINT: the result vector is an infeasible point.\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NO_SOLUTION","page":"Models","title":"MathOptInterface.NO_SOLUTION","text":"NO_SOLUTION::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNO_SOLUTION: the result vector is empty.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.FEASIBLE_POINT","page":"Models","title":"MathOptInterface.FEASIBLE_POINT","text":"FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nFEASIBLE_POINT: the result vector is a feasible point.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_FEASIBLE_POINT","page":"Models","title":"MathOptInterface.NEARLY_FEASIBLE_POINT","text":"NEARLY_FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE_POINT","page":"Models","title":"MathOptInterface.INFEASIBLE_POINT","text":"INFEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBLE_POINT: the result vector is an infeasible point.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBILITY_CERTIFICATE","page":"Models","title":"MathOptInterface.INFEASIBILITY_CERTIFICATE","text":"INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE","page":"Models","title":"MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE","text":"NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.REDUCTION_CERTIFICATE","page":"Models","title":"MathOptInterface.REDUCTION_CERTIFICATE","text":"REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE","page":"Models","title":"MathOptInterface.NEARLY_REDUCTION_CERTIFICATE","text":"NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.UNKNOWN_RESULT_STATUS","page":"Models","title":"MathOptInterface.UNKNOWN_RESULT_STATUS","text":"UNKNOWN_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_RESULT_STATUS","page":"Models","title":"MathOptInterface.OTHER_RESULT_STATUS","text":"OTHER_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#Conflict-Status","page":"Models","title":"Conflict Status","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"compute_conflict!\nConflictStatus\nConstraintConflictStatus\nConflictStatusCode\nConflictParticipationStatusCode\nNOT_IN_CONFLICT\nIN_CONFLICT\nMAYBE_IN_CONFLICT","category":"page"},{"location":"moi/reference/models/#MathOptInterface.compute_conflict!","page":"Models","title":"MathOptInterface.compute_conflict!","text":"compute_conflict!(optimizer::AbstractOptimizer)\n\nComputes a minimal subset of constraints such that the model with the other constraint removed is still infeasible.\n\nSome solvers call a set of conflicting constraints an Irreducible Inconsistent Subsystem (IIS).\n\nSee also ConflictStatus and ConstraintConflictStatus.\n\nNote\n\nIf the model is modified after a call to compute_conflict!, the implementor is not obliged to purge the conflict. Any calls to the above attributes may return values for the original conflict without a warning. Similarly, when modifying the model, the conflict can be discarded.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.ConflictStatus","page":"Models","title":"MathOptInterface.ConflictStatus","text":"ConflictStatus()\n\nA model attribute for the ConflictStatusCode explaining why the conflict refiner stopped when computing the conflict.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConstraintConflictStatus","page":"Models","title":"MathOptInterface.ConstraintConflictStatus","text":"ConstraintConflictStatus()\n\nA constraint attribute indicating whether the constraint participates in the conflict. Its type is ConflictParticipationStatusCode.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConflictStatusCode","page":"Models","title":"MathOptInterface.ConflictStatusCode","text":"ConflictStatusCode\n\nAn Enum of possible values for the ConflictStatus attribute. This attribute is meant to explain the reason why the conflict finder stopped executing in the most recent call to compute_conflict!.\n\nPossible values are:\n\nCOMPUTE_CONFLICT_NOT_CALLED: the function compute_conflict! has not yet been called\nNO_CONFLICT_EXISTS: there is no conflict because the problem is feasible\nNO_CONFLICT_FOUND: the solver could not find a conflict\nCONFLICT_FOUND: at least one conflict could be found\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConflictParticipationStatusCode","page":"Models","title":"MathOptInterface.ConflictParticipationStatusCode","text":"ConflictParticipationStatusCode\n\nAn Enum of possible values for the ConstraintConflictStatus attribute. This attribute is meant to indicate whether a given constraint participates or not in the last computed conflict.\n\nValues\n\nPossible values are:\n\nNOT_IN_CONFLICT: the constraint does not participate in the conflict\nIN_CONFLICT: the constraint participates in the conflict\nMAYBE_IN_CONFLICT: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NOT_IN_CONFLICT","page":"Models","title":"MathOptInterface.NOT_IN_CONFLICT","text":"NOT_IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nNOT_IN_CONFLICT: the constraint does not participate in the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.IN_CONFLICT","page":"Models","title":"MathOptInterface.IN_CONFLICT","text":"IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nIN_CONFLICT: the constraint participates in the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MAYBE_IN_CONFLICT","page":"Models","title":"MathOptInterface.MAYBE_IN_CONFLICT","text":"MAYBE_IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nMAYBE_IN_CONFLICT: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Nonlinear/reference.md\"","category":"page"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#NonlinearAPI","page":"API Reference","title":"Nonlinear Modeling","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"More information can be found in the Nonlinear section of the manual.","category":"page"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear\nNonlinear.Model","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear","page":"API Reference","title":"MathOptInterface.Nonlinear","text":"Nonlinear\n\nwarning: Warning\nThe Nonlinear submodule is experimental. Until this message is removed, breaking changes may be introduced in any minor or patch release of MathOptInterface.\n\n\n\n\n\n","category":"module"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model","page":"API Reference","title":"MathOptInterface.Nonlinear.Model","text":"Model()\n\nThe core datastructure for representing a nonlinear optimization problem.\n\nIt has the following fields:\n\nobjective::Union{Nothing,Expression} : holds the nonlinear objective function, if one exists, otherwise nothing.\nexpressions::Vector{Expression} : a vector of expressions in the model.\nconstraints::OrderedDict{ConstraintIndex,Constraint} : a map from ConstraintIndex to the corresponding Constraint. An OrderedDict is used instead of a Vector to support constraint deletion.\nparameters::Vector{Float64} : holds the current values of the parameters.\noperators::OperatorRegistry : stores the operators used in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_expressions","page":"API Reference","title":"Expressions","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ExpressionIndex\nNonlinear.add_expression","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ExpressionIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ExpressionIndex","text":"ExpressionIndex\n\nAn index to a nonlinear expression that is returned by add_expression.\n\nGiven data::Model and ex::ExpressionIndex, use data[ex] to retrieve the corresponding Expression.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_expression","page":"API Reference","title":"MathOptInterface.Nonlinear.add_expression","text":"add_expression(model::Model, expr)::ExpressionIndex\n\nParse expr into a Expression and add to model. Returns an ExpressionIndex that can be interpolated into other input expressions.\n\nexpr must be a type that is supported by parse_expression.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nex = add_expression(model, :($x^2 + 1))\nset_objective(model, :(sqrt($ex)))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_parameters","page":"API Reference","title":"Parameters","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ParameterIndex\nNonlinear.add_parameter","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ParameterIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ParameterIndex","text":"ParameterIndex\n\nAn index to a nonlinear parameter that is returned by add_parameter. Given data::Model and p::ParameterIndex, use data[p] to retrieve the current value of the parameter and data[p] = value to set a new value.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_parameter","page":"API Reference","title":"MathOptInterface.Nonlinear.add_parameter","text":"add_parameter(model::Model, value::Float64)::ParameterIndex\n\nAdd a new parameter to model with the default value value. Returns a ParameterIndex that can be interpolated into other input expressions and used to modify the value of the parameter.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\np = add_parameter(model, 1.2)\nc = add_constraint(model, :($x^2 - $p), MOI.LessThan(0.0))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_objectives","page":"API Reference","title":"Objectives","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.set_objective","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.set_objective","page":"API Reference","title":"MathOptInterface.Nonlinear.set_objective","text":"set_objective(model::Model, obj)::Nothing\n\nParse obj into a Expression and set as the objective function of model.\n\nobj must be a type that is supported by parse_expression.\n\nTo remove the objective, pass nothing.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nset_objective(model, :($x^2 + 1))\nset_objective(model, x)\nset_objective(model, nothing)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_constraints","page":"API Reference","title":"Constraints","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ConstraintIndex\nNonlinear.add_constraint\nNonlinear.delete","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ConstraintIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ConstraintIndex","text":"ConstraintIndex\n\nAn index to a nonlinear constraint that is returned by add_constraint.\n\nGiven data::Model and c::ConstraintIndex, use data[c] to retrieve the corresponding Constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_constraint","page":"API Reference","title":"MathOptInterface.Nonlinear.add_constraint","text":"add_constraint(\n    model::Model,\n    func,\n    set::Union{\n        MOI.GreaterThan{Float64},\n        MOI.LessThan{Float64},\n        MOI.Interval{Float64},\n        MOI.EqualTo{Float64},\n    },\n)\n\nParse func and set into a Constraint and add to model. Returns a ConstraintIndex that can be used to delete the constraint or query solution information.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc = add_constraint(model, :($x^2), MOI.LessThan(1.0))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.delete","page":"API Reference","title":"MathOptInterface.Nonlinear.delete","text":"delete(model::Model, c::ConstraintIndex)::Nothing\n\nDelete the constraint index c from model.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc = add_constraint(model, :($x^2), MOI.LessThan(1.0))\ndelete(model, c)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_operators","page":"API Reference","title":"User-defined operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.OperatorRegistry\nNonlinear.DEFAULT_UNIVARIATE_OPERATORS\nNonlinear.DEFAULT_MULTIVARIATE_OPERATORS\nNonlinear.register_operator\nNonlinear.register_operator_if_needed\nNonlinear.assert_registered\nNonlinear.check_return_type\nNonlinear.eval_univariate_function\nNonlinear.eval_univariate_gradient\nNonlinear.eval_univariate_hessian\nNonlinear.eval_multivariate_function\nNonlinear.eval_multivariate_gradient\nNonlinear.eval_multivariate_hessian\nNonlinear.eval_logic_function\nNonlinear.eval_comparison_function","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.OperatorRegistry","page":"API Reference","title":"MathOptInterface.Nonlinear.OperatorRegistry","text":"OperatorRegistry()\n\nCreate a new OperatorRegistry to store and evaluate univariate and multivariate operators.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS","page":"API Reference","title":"MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS","text":"DEFAULT_UNIVARIATE_OPERATORS\n\nThe list of univariate operators that are supported by default.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS\n72-element Vector{Symbol}:\n :+\n :-\n :abs\n :sqrt\n :cbrt\n :abs2\n :inv\n :log\n :log10\n :log2\n ⋮\n :airybi\n :airyaiprime\n :airybiprime\n :besselj0\n :besselj1\n :bessely0\n :bessely1\n :erfcx\n :dawson\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS","page":"API Reference","title":"MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS","text":"DEFAULT_MULTIVARIATE_OPERATORS\n\nThe list of multivariate operators that are supported by default.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n9-element Vector{Symbol}:\n :+\n :-\n :*\n :^\n :/\n :ifelse\n :atan\n :min\n :max\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.register_operator","page":"API Reference","title":"MathOptInterface.Nonlinear.register_operator","text":"register_operator(\n    model::Model,\n    op::Symbol,\n    nargs::Int,\n    f::Function,\n    [∇f::Function],\n    [∇²f::Function],\n)\n\nRegister the user-defined operator op with nargs input arguments in model.\n\nUnivariate functions\n\nf(x::T)::T must be a function that takes a single input argument x and returns the function evaluated at x. If ∇f and ∇²f are not provided, f must support any Real input type T.\n∇f(x::T)::T is a function that takes a single input argument x and returns the first derivative of f with respect to x. If ∇²f is not provided, ∇f must support any Real input type T.\n∇²f(x::T)::T is a function that takes a single input argument x and returns the second derivative of f with respect to x.\n\nMultivariate functions\n\nf(x::T...)::T must be a function that takes a nargs input arguments x and returns the function evaluated at x. If ∇f and ∇²f are not provided, f must support any Real input type T.\n∇f(g::AbstractVector{T}, x::T...)::T is a function that takes a cache vector g of length length(x), and fills each element g[i] with the partial derivative of f with respect to x[i].\n∇²f(H::AbstractMatrix, x::T...)::T is a function that takes a matrix H and fills the lower-triangular components H[i, j] with the Hessian of f with respect to x[i] and x[j] for i >= j.\n\nNotes for multivariate Hessians\n\nH has size(H) == (length(x), length(x)), but you must not access elements H[i, j] for i > j.\nH is dense, but you do not need to fill structural zeros.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.register_operator_if_needed","page":"API Reference","title":"MathOptInterface.Nonlinear.register_operator_if_needed","text":"register_operator_if_needed(\n    registry::OperatorRegistry,\n    op::Symbol,\n    nargs::Int,\n    f::Function;\n)\n\nSimilar to register_operator, but this function warns if the function is not registered, and skips silently if it already is.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.assert_registered","page":"API Reference","title":"MathOptInterface.Nonlinear.assert_registered","text":"assert_registered(registry::OperatorRegistry, op::Symbol, nargs::Int)\n\nThrow an error if op is not registered in registry with nargs arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.check_return_type","page":"API Reference","title":"MathOptInterface.Nonlinear.check_return_type","text":"check_return_type(::Type{T}, ret::S) where {T,S}\n\nOverload this method for new types S to throw an informative error if a user-defined function returns the type S instead of T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_function","text":"eval_univariate_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_gradient","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_gradient","text":"eval_univariate_gradient(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the first-derivative of the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_hessian","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_hessian","text":"eval_univariate_hessian(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the second-derivative of the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_function","text":"eval_multivariate_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the operator op(x)::T, where op is a multivariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_gradient","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_gradient","text":"eval_multivariate_gradient(\n    registry::OperatorRegistry,\n    op::Symbol,\n    g::AbstractVector{T},\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the gradient of operator g .= ∇op(x), where op is a multivariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_hessian","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_hessian","text":"eval_multivariate_hessian(\n    registry::OperatorRegistry,\n    op::Symbol,\n    H::AbstractMatrix,\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the Hessian of operator ∇²op(x), where op is a multivariate function in registry.\n\nThe Hessian is stored in the lower-triangular part of the matrix H.\n\nnote: Note\nImplementations of the Hessian operators will not fill structural zeros. Therefore, before calling this function you should pre-populate the matrix H with 0.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_logic_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_logic_function","text":"eval_logic_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    lhs::T,\n    rhs::T,\n)::Bool where {T}\n\nEvaluate (lhs op rhs)::Bool, where op is a logic operator in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_comparison_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_comparison_function","text":"eval_comparison_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    lhs::T,\n    rhs::T,\n)::Bool where {T}\n\nEvaluate (lhs op rhs)::Bool, where op is a comparison operator in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#Automatic-differentiation-backends","page":"API Reference","title":"Automatic-differentiation backends","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.Evaluator\nNonlinear.AbstractAutomaticDifferentiation\nNonlinear.ExprGraphOnly\nNonlinear.SparseReverseMode","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Evaluator","page":"API Reference","title":"MathOptInterface.Nonlinear.Evaluator","text":"Evaluator(\n    model::Model,\n    backend::AbstractAutomaticDifferentiation,\n    ordered_variables::Vector{MOI.VariableIndex},\n)\n\nCreate Evaluator, a subtype of MOI.AbstractNLPEvaluator, from Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation","page":"API Reference","title":"MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation","text":"AbstractAutomaticDifferentiation\n\nAn abstract type for extending Evaluator.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ExprGraphOnly","page":"API Reference","title":"MathOptInterface.Nonlinear.ExprGraphOnly","text":"ExprGraphOnly() <: AbstractAutomaticDifferentiation\n\nThe default implementation of AbstractAutomaticDifferentiation. The only supported feature is :ExprGraph.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.SparseReverseMode","page":"API Reference","title":"MathOptInterface.Nonlinear.SparseReverseMode","text":"SparseReverseMode() <: AbstractAutomaticDifferentiation\n\nAn implementation of AbstractAutomaticDifferentiation that uses sparse reverse-mode automatic differentiation to compute derivatives. Supports all features in the MOI nonlinear interface.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#Data-structure","page":"API Reference","title":"Data-structure","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.Node\nNonlinear.NodeType\nNonlinear.Expression\nNonlinear.Constraint\nNonlinear.adjacency_matrix\nNonlinear.parse_expression\nNonlinear.convert_to_expr\nNonlinear.ordinal_index","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Node","page":"API Reference","title":"MathOptInterface.Nonlinear.Node","text":"struct Node\n    type::NodeType\n    index::Int\n    parent::Int\nend\n\nA single node in a nonlinear expression tree. Used by Expression.\n\nSee the MathOptInterface documentation for information on how the nodes and values form an expression tree.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.NodeType","page":"API Reference","title":"MathOptInterface.Nonlinear.NodeType","text":"NodeType\n\nAn enum describing the possible node types. Each Node has a .index field, which should be interpreted as follows:\n\nNODE_CALL_MULTIVARIATE: the index into operators.multivariate_operators\nNODE_CALL_UNIVARIATE: the index into operators.univariate_operators\nNODE_LOGIC: the index into operators.logic_operators\nNODE_COMPARISON: the index into operators.comparison_operators\nNODE_MOI_VARIABLE: the value of MOI.VariableIndex(index) in the user's space of the model.\nNODE_VARIABLE: the 1-based index of the internal vector\nNODE_VALUE: the index into the .values field of Expression\nNODE_PARAMETER: the index into data.parameters\nNODE_SUBEXPRESSION:  the index into data.expressions\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Expression","page":"API Reference","title":"MathOptInterface.Nonlinear.Expression","text":"struct Expression\n    nodes::Vector{Node}\n    values::Vector{Float64}\nend\n\nThe core type that represents a nonlinear expression. See the MathOptInterface documentation for information on how the nodes and values form an expression tree.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Constraint","page":"API Reference","title":"MathOptInterface.Nonlinear.Constraint","text":"struct Constraint\n    expression::Expression\n    set::Union{\n        MOI.LessThan{Float64},\n        MOI.GreaterThan{Float64},\n        MOI.EqualTo{Float64},\n        MOI.Interval{Float64},\n    }\nend\n\nA type to hold information relating to the nonlinear constraint f(x) in S, where f(x) is defined by .expression, and S is .set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.adjacency_matrix","page":"API Reference","title":"MathOptInterface.Nonlinear.adjacency_matrix","text":"adjacency_matrix(nodes::Vector{Node})\n\nCompute the sparse adjacency matrix describing the parent-child relationships in nodes.\n\nThe element (i, j) is true if there is an edge from node[j] to node[i]. Since we get a column-oriented matrix, this gives us a fast way to look up the edges leaving any node (i.e., the children).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.parse_expression","page":"API Reference","title":"MathOptInterface.Nonlinear.parse_expression","text":"parse_expression(data::Model, input)::Expression\n\nParse input into a Expression.\n\n\n\n\n\nparse_expression(\n    data::Model,\n    expr::Expression,\n    input::Any,\n    parent_index::Int,\n)::Expression\n\nParse input into a Expression, and add it to expr as a child of expr.nodes[parent_index]. Existing subexpressions and parameters are stored in data.\n\nYou can extend parsing support to new types of objects by overloading this method with a different type on input::Any.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.convert_to_expr","page":"API Reference","title":"MathOptInterface.Nonlinear.convert_to_expr","text":"convert_to_expr(data::Model, expr::Expression)\n\nConvert the Expression expr into a Julia Expr.\n\nsubexpressions are represented by a ExpressionIndex object.\nparameters are represented by a ParameterIndex object.\nvariables are represennted by an MOI.VariableIndex object.\n\n\n\n\n\nconvert_to_expr(\n    evaluator::Evaluator,\n    expr::Expression;\n    moi_output_format::Bool,\n)\n\nConvert the Expression expr into a Julia Expr.\n\nIf moi_output_format = true:\n\nsubexpressions will be converted to Julia Expr and substituted into the output expression.\nthe current value of each parameter will be interpolated into the expression\nvariables will be represented in the form x[MOI.VariableIndex(i)]\n\nIf moi_output_format = false:\n\nsubexpressions will be represented by a ExpressionIndex object.\nparameters will be represented by a ParameterIndex object.\nvariables will be represennted by an MOI.VariableIndex object.\n\nwarning: Warning\nTo use moi_output_format = true, you must have first called MOI.initialize with :ExprGraph as a requested feature.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ordinal_index","page":"API Reference","title":"MathOptInterface.Nonlinear.ordinal_index","text":"ordinal_index(evaluator::Evaluator, c::ConstraintIndex)::Int\n\nReturn the 1-indexed value of the constraint index c in evaluator.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc1 = add_constraint(model, :($x^2), MOI.LessThan(1.0))\nc2 = add_constraint(model, :($x^2), MOI.LessThan(1.0))\nevaluator = Evaluator(model)\nMOI.initialize(evaluator, Symbol[])\nordinal_index(evaluator, c2)  # Returns 2\ndelete(model, c1)\nevaluator = Evaluator(model)\nMOI.initialize(evaluator, Symbol[])\nordinal_index(model, c2)  # Returns 1\n\n\n\n\n\n","category":"function"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"EditURL = \"https://github.com/lanl-ansi/Juniper.jl/blob/62532341586d447f19c7360715333ba62a42bea9/README.md\"","category":"page"},{"location":"packages/Juniper/#Juniper","page":"lanl-ansi/Juniper.jl","title":"Juniper","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"(Image: CI) (Image: codecov) (Image: Documentation)","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Juniper (Jump Nonlinear Integer Program solver) is a solver for mixed-integer nonlinear programs. ","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"It is a heuristic which is not guaranteed to find the global optimum. If you need the global optimum, check out Alpine.","category":"page"},{"location":"packages/Juniper/#Installation","page":"lanl-ansi/Juniper.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Install Juniper using the Julia package manager:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"packages/Juniper/#Use-with-JuMP","page":"lanl-ansi/Juniper.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Use Juniper with JuMP as follows:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"using JuMP, Juniper, Ipopt\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\"=>0)\noptimizer = optimizer_with_attributes(Juniper.Optimizer, \"nl_solver\"=>ipopt)\nmodel = Model(optimizer)\nv = [10, 20, 12, 23, 42]\nw = [12, 45, 12, 22, 21]\n@variable(model, x[1:5], Bin)\n@objective(model, Max, v' * x)\n@constraint(model, sum(w[i]*x[i]^2 for i in 1:5) <= 45)\noptimize!(model)\nprintln(termination_status(model))\nprintln(objective_value(model))\nprintln(value.(x))","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The nl_solver is used by Juniper to solve continuous nonlinear sub-problems while Juniper searches for acceptable assignments to the discrete variables. A common choice is Ipopt, but any optimizer that supports the continuous relaxation of the model may be used.","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"To solve problems with more complex nonlinear functions, use the @NLconstraint and @NLobjective JuMP macros.","category":"page"},{"location":"packages/Juniper/#Documentation","page":"lanl-ansi/Juniper.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The online documentation is available at https://lanl-ansi.github.io/Juniper.jl/stable/.","category":"page"},{"location":"packages/Juniper/#Feasibility-pump","page":"lanl-ansi/Juniper.jl","title":"Feasibility pump","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"If Juniper has difficulty finding feasible solutions on your model, try adding a solver that supports integer variables (for example, HiGHS) to run a feasibility pump:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"using JuMP, Juniper, Ipopt, HiGHS\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0)\nhighs = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false)\nmodel = Model(\n    optimizer_with_attributes(\n        Juniper.Optimizer,\n        \"nl_solver\" => ipopt,\n        \"mip_solver\" => highs,\n    ),\n)","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The feasibility pump is used at the start of Juniper to find a feasible solution before the branch and bound part starts.  For some classes of problems this can be a highly effective pre-processor.","category":"page"},{"location":"packages/Juniper/#Citing-Juniper","page":"lanl-ansi/Juniper.jl","title":"Citing Juniper","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"If you find Juniper useful in your work, we kindly request that you cite the following paper or technical report:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"@inproceedings{juniper,\n     Author = {Ole Kröger and Carleton Coffrin and Hassan Hijazi and Harsha Nagarajan},\n     Title = {Juniper: An Open-Source Nonlinear Branch-and-Bound Solver in Julia},\n     booktitle=\"Integration of Constraint Programming, Artificial Intelligence, and Operations Research\",\n     pages=\"377--386\",\n     year=\"2018\",\n     publisher=\"Springer International Publishing\",\n     isbn=\"978-3-319-93031-2\"\n}","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"EditURL = \"diet.jl\"","category":"page"},{"location":"tutorials/linear/diet/#The-diet-problem","page":"The diet problem","title":"The diet problem","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"The purpose of this tutorial is to demonstrate how to incorporate DataFrames into a JuMP model. As an example, we use classic Stigler diet problem.","category":"page"},{"location":"tutorials/linear/diet/#Required-packages","page":"The diet problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"using JuMP\nimport CSV\nimport DataFrames\nimport HiGHS\nimport Test  #hide","category":"page"},{"location":"tutorials/linear/diet/#Formulation","page":"The diet problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We wish to cook a nutritionally balanced meal by choosing the quantity of each food f to eat from a set of foods F in our kitchen.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Each food f has a cost, c_f, as well as a macro-nutrient profile a_mf for each macro-nutrient m in M.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Because we care about a nutritionally balanced meal, we set some minimum and maximum limits for each nutrient, which we denote l_m and u_m respectively.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Furthermore, because we are optimizers, we seek the minimum cost solution.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"With a little effort, we can formulate our dinner problem as the following linear program:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"beginaligned\nmin  sumlimits_f in F c_f x_f \ntextst   l_m le sumlimits_f in F a_mf x_f le u_m  forall m in M \n x_f ge 0  forall f in F\nendaligned","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"In the rest of this tutorial, we will create and solve this problem in JuMP, and learn what we should cook for dinner.","category":"page"},{"location":"tutorials/linear/diet/#Data","page":"The diet problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"First, we need some data for the problem. For this tutorial, we'll write CSV files to a temporary directory from Julia. If you have existing files, you could change the filenames to point to them instead.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"dir = mktempdir()","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"The first file is a list of foods with their macro-nutrient profile:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"food_csv_filename = joinpath(dir, \"diet_foods.csv\")\nopen(food_csv_filename, \"w\") do io\n    write(\n        io,\n        \"\"\"\n        name,cost,calories,protein,fat,sodium\n        hamburger,2.49,410,24,26,730\n        chicken,2.89,420,32,10,1190\n        hot dog,1.50,560,20,32,1800\n        fries,1.89,380,4,19,270\n        macaroni,2.09,320,12,10,930\n        pizza,1.99,320,15,12,820\n        salad,2.49,320,31,12,1230\n        milk,0.89,100,8,2.5,125\n        ice cream,1.59,330,8,10,180\n        \"\"\",\n    )\n    return\nend\nfoods = CSV.read(food_csv_filename, DataFrames.DataFrame)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Here, F is foods.name and c_f is foods.cost. (We're also playing a bit loose the term \"macro-nutrient\" by including calories and sodium.)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We also need our minimum and maximum limits:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"nutrient_csv_filename = joinpath(dir, \"diet_nutrient.csv\")\nopen(nutrient_csv_filename, \"w\") do io\n    write(\n        io,\n        \"\"\"\n        nutrient,min,max\n        calories,1800,2200\n        protein,91,\n        fat,0,65\n        sodium,0,1779\n        \"\"\",\n    )\n    return\nend\nlimits = CSV.read(nutrient_csv_filename, DataFrames.DataFrame)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Protein is missing data for the maximum. Let's fix that using coalesce:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"limits.max = coalesce.(limits.max, Inf)\nlimits","category":"page"},{"location":"tutorials/linear/diet/#JuMP-formulation","page":"The diet problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Now we're ready to convert our mathematical formulation into a JuMP model.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"First, create a new JuMP model. Since we have a linear program, we'll use HiGHS as our optimizer:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Next, we create a set of decision variables x, with one element for each row in the DataFrame, and each x has a lower bound of 0:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@variable(model, x[foods.name] >= 0)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"To simplify things later on, we store the vector as a new column x in the DataFrame foods. Since x is a DenseAxisArray, we first need to convert it to an Array:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"foods.x = Array(x)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Our objective is to minimize the total cost of purchasing food:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@objective(model, Min, sum(foods.cost .* foods.x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"For the next component, we need to add a constraint that our total intake of each component is within the limits contained in the limits DataFrame:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@constraint(\n    model,\n    [row in eachrow(limits)],\n    row.min <= sum(foods[!, row.nutrient] .* foods.x) <= row.max,\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"What does our model look like?","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"print(model)","category":"page"},{"location":"tutorials/linear/diet/#Solution","page":"The diet problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Let's optimize and take a look at the solution:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"optimize!(model)\nTest.@test primal_status(model) == FEASIBLE_POINT        #hide\nTest.@test objective_value(model) ≈ 11.8288 atol = 1e-4  #hide\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We found an optimal solution. Let's see what the optimal solution is:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"for row in eachrow(foods)\n    println(row.name, \" = \", value(row.x))\nend","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"That's a lot of milk and ice cream, and sadly, we only get 0.6 of a hamburger.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We can also use the function Containers.rowtable to easily convert the result into a DataFrame:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"table = Containers.rowtable(value, x; header = [:food, :quantity])\nsolution = DataFrames.DataFrame(table)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This makes it easy to perform analyses our solution:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"filter!(row -> row.quantity > 0.0, solution)","category":"page"},{"location":"tutorials/linear/diet/#Problem-modification","page":"The diet problem","title":"Problem modification","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"JuMP makes it easy to take an existing model and modify it by adding extra constraints. Let's see what happens if we add a constraint that we can buy at most 6 units of milk or ice cream combined.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"dairy_foods = [\"milk\", \"ice cream\"]\nis_dairy = map(name -> name in dairy_foods, foods.name)\ndairy_constraint = @constraint(model, sum(foods[is_dairy, :x]) <= 6)\noptimize!(model)\nTest.@test termination_status(model) == INFEASIBLE  #hide\nTest.@test primal_status(model) == NO_SOLUTION      #hide\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"There exists no feasible solution to our problem. Looks like we're stuck eating ice cream for dinner.","category":"page"},{"location":"tutorials/linear/diet/#Next-steps","page":"The diet problem","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"You can delete a constraint using delete(model, dairy_constraint). Can you add a different constraint to provide a diet with less dairy?\nSome food items (like hamburgers) are discrete. You can use set_integer to force a variable to take integer values. What happens to the solution if you do?","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"EditURL = \"n-queens.jl\"","category":"page"},{"location":"tutorials/linear/n-queens/#N-Queens","page":"N-Queens","title":"N-Queens","text":"","category":"section"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"This tutorial was originally contributed by Matthew Helm and Mathieu Tanneau.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"The N-Queens problem involves placing N queens on an N x N chessboard such that none of the queens attacks another. In chess, a queen can move vertically, horizontally, and diagonally so there cannot be more than one queen on any given row, column, or diagonal.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"(Image: Four Queens)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Note that none of the queens above are able to attack any other as a result of their careful placement.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"using JuMP\nimport HiGHS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"N-Queens","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"N = 8\n\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Next, let's create an N x N chessboard of binary values. 0 will represent an empty space on the board and 1 will represent a space occupied by one of our queens:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"@variable(model, x[1:N, 1:N], Bin);\nnothing #hide","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Now we can add our constraints:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"There must be exactly one queen in a given row/column","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"for i in 1:N\n    @constraint(model, sum(x[i, :]) == 1)\n    @constraint(model, sum(x[:, i]) == 1)\nend","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"There can only be one queen on any given diagonal","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"for i in -(N - 1):(N-1)\n    @constraint(model, sum(LinearAlgebra.diag(x, i)) <= 1)\n    @constraint(model, sum(LinearAlgebra.diag(reverse(x; dims = 1), i)) <= 1)\nend","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"We are ready to put our model to work and see if it is able to find a feasible solution:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"optimize!(model)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"We can now review the solution that our model found:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"solution = round.(Int, value.(x))","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"EditURL = \"https://github.com/jump-dev/SDPNAL.jl/blob/00a3fa19f4e1235587948113b0b681da17f4dab5/README.md\"","category":"page"},{"location":"packages/SDPNAL/#SDPNAL.jl","page":"jump-dev/SDPNAL.jl","title":"SDPNAL.jl","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"SDPNAL.jl is wrapper for the SDPNALplus solver.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"an exported sdpnalplus function that is a thin wrapper on top of the sdpnalplus MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SDPNAL/#Affiliation","page":"jump-dev/SDPNAL.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SDPNALplus.","category":"page"},{"location":"packages/SDPNAL/#License","page":"jump-dev/SDPNAL.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"SDPNAL.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"The underlying solver, SDPNALplus is licensed under the Creative Commons Attribution-ShareAlike 4.0 International Public License.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"In addition, SDPNAL requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SDPNAL/#Use-with-JuMP","page":"jump-dev/SDPNAL.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"To use SDPNAL with JuMP, do:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"using JuMP, SDPNAL\nmodel = Model(SDPNAL.Optimizer)\nset_attribute(model, \"printlevel\", 0)","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"Note that, contrary to implementation of other solver-independent interfaces, using SDPNAL from JuMP or MOI fully exploits the particular structures of the SDPNAL interface and does not create superfluous slack variables and equality constraints as discussed in the SDPNAL guide:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"A new interface is necessary to facilitate the modeling of an SDP problem for SDPNAL+ because of latter’s flexibility to directly accept inequality constraints of the form “l ≤ B(X) ≤ u”, and bound constraints of the form “L ≤ X ≤ U”. The flexibility can significantly simplify the generation of the data in the SDPNAL+ format as compared to what need to be done in CVX or YALMIP to reformulate them as equality constraints through introducing extra variables. In addition, the final number of equality constraints present in the data input to SDPNAL+ can also be substantially fewer than those present in CVX or YALMIP. It is important to note here that the number of equality constraints present in the generated problem data can greatly affect the computational efficiency of the solvers, especially for interior-point based solvers.","category":"page"},{"location":"packages/SDPNAL/#Installation","page":"jump-dev/SDPNAL.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SDPNALplus software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"Then, install SDPNAL.jl using Pkg.add:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"import Pkg\nPkg.add(\"SDPNAL\")","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"There is a startup.m file at the root of the SDPNAL folder. This adds all subdirectories recursively when MATLAB starts. However, the interface directory contains a .git subdirectory which contains a very large number of files. Because of this, MATLAB crashes if SDPNAL is in its path because the startup.m requests MATLAB to try to parse all the files in the .git folder. To resolve this problem, delete the startup.m file and .git folder, and add the subdirectories manually your toolbox/local/pathdef.m file as follows:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"function p = pathdef\n\n% (...)\n\np = [...\n%%% BEGIN ENTRIES %%%\n'/path/to/SDPNALv1.0:', ...\n'/path/to/SDPNALv1.0/interface:', ...\n'/path/to/SDPNALv1.0/mexfun:', ...\n'/path/to/SDPNALv1.0/solver:', ...\n'/path/to/SDPNALv1.0/solver_main_default:', ...\n'/path/to/SDPNALv1.0/util:', ...\n% (...)","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"If you have SDPT3 in addition to SDPNAL in the MATLAB path (that is, the toolbox/local/pathdef.m file) then you might have issues because both solvers define a validate function, and this might make SDPNAL call SDPT3's validate function instead of SDPT3's validate function.","category":"page"},{"location":"should_i_use/#Should-you-use-JuMP?","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP is an algebraic modeling language for mathematical optimization written in the Julia language.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"This page explains when you should consider using JuMP, and importantly, when you should not use JuMP.","category":"page"},{"location":"should_i_use/#When-should-you-use-JuMP?","page":"Should you use JuMP?","title":"When should you use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"You should use JuMP if you have a constrained optimization problem for which you can formulate using the language of mathematical programming, that is:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"a set of decision variables\na scalar- or vector-valued objective function\na set of constraints.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Key reasons to use JuMP include:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"User friendliness\nJuMP has syntax that mimics natural mathematical expressions. (See the section on algebraic modeling languages.)\nSolver independence\nJuMP uses a generic solver-independent interface provided by the MathOptInterface package, making it easy to change between a number of open-source and commercial optimization software packages (\"solvers\"). The Supported solvers section contains a table of the currently supported solvers.\nEase of embedding\nJuMP itself is written purely in Julia. Solvers are the only binary dependencies.\nJuMP provides automatic installation of many open-source solvers. This is different to modeling languages in Python which require you to download and install a solver yourself.\nBecause it is embedded in a general-purpose programming language, JuMP makes it easy to solve optimization problems as part of a larger workflow, for example, inside a simulation, behind a web server, or as a subproblem in a decomposition algorithm. As a trade-off, JuMP's syntax is constrained by the syntax and functionality available in Julia.\nJuMP is MPL licensed, meaning that it can be embedded in commercial software that complies with the terms of the license.\nSpeed\nBenchmarking has shown that JuMP can create problems at similar speeds to special-purpose modeling languages such as AMPL.\nJuMP communicates with most solvers in memory, avoiding the need to write intermediary files.\nAccess to advanced algorithmic techniques\nJuMP supports efficient in-memory re-solves of linear programs, which previously required using solver-specific or low-level C++ libraries.\nJuMP provides access to solver-independent and solver-dependent Callbacks.","category":"page"},{"location":"should_i_use/#When-should-you-not-use-JuMP?","page":"Should you use JuMP?","title":"When should you not use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP supports a broad range of optimization classes. However, there are still some that it doesn't support, or that are better supported by other software packages.","category":"page"},{"location":"should_i_use/#You-want-to-optimize-a-complicated-Julia-function","page":"Should you use JuMP?","title":"You want to optimize a complicated Julia function","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Packages in Julia compose well. It's common for people to pick two unrelated packages and use them in conjunction to create novel behavior. JuMP isn't one of those packages.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If you want to optimize an ordinary differential equation from DifferentialEquations.jl or tune a neural network from Flux.jl, consider using other packages such as:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Optim.jl\nOptimization.jl\nNLPModels.jl\nNonconvex.jl","category":"page"},{"location":"should_i_use/#Black-box,-derivative-free,-or-unconstrained-optimization","page":"Should you use JuMP?","title":"Black-box, derivative free, or unconstrained optimization","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does support nonlinear programs with constraints and objectives containing user-defined operators. However, the functions must be automatically differentiable, or need to provide explicit derivatives. (See User-defined operators for more information.)","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If your function is a black-box that is non-differentiable (for example, it is the output of a simulation written in C++), JuMP is not the right tool for the job. This also applies if you want to use a derivative free method.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Even if your problem is differentiable, if it is unconstrained there is limited benefit (and downsides in the form of more overhead) to using JuMP over tools which are only concerned with function minimization.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Optim.jl\nOptimization.jl\nNLopt.jl","category":"page"},{"location":"should_i_use/#Optimal-control-problems","page":"Should you use JuMP?","title":"Optimal control problems","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP supports formulating optimal control problems as large nonlinear programs (see, for example, Optimal control for a Space Shuttle reentry trajectory). However, the nonlinear interface has a number of limitations (for example, the need to write out the dynamics in algebraic form) that mean JuMP might not be the right tool for the job.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"CasADi [MATLAB/Python], CasADi.jl\nInfiniteOpt.jl\npyomo.DAE [Python]","category":"page"},{"location":"should_i_use/#Disciplined-convex-programming","page":"Should you use JuMP?","title":"Disciplined convex programming","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does not support disciplined convex programming (DCP).","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Convex.jl\nCVXPY [Python]\nYALMIP [MATLAB]","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"note: Note\nConvex.jl is also built on MathOptInterface, and shares the same set of underlying solvers. However, you input problems differently, and Convex.jl checks that the problem is DCP.","category":"page"},{"location":"should_i_use/#Stochastic-programming","page":"Should you use JuMP?","title":"Stochastic programming","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP requires deterministic input data.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If you have stochastic input data, consider using a JuMP extension such as:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"InfiniteOpt.jl\nStochasticPrograms.jl\nSDDP.jl","category":"page"},{"location":"should_i_use/#Polyhedral-computations","page":"Should you use JuMP?","title":"Polyhedral computations","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does not provide tools for working with the polyhedron formed by the set of linear constraints.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Polyhedra.jl (See the documentation to create a polyhedron from a JuMP model.)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"EditURL = \"design_patterns_for_larger_models.jl\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Design-patterns-for-larger-models","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"JuMP makes it easy to build and solve optimization models. However, once you start to construct larger models, and especially ones that interact with external data sources or have customizable sets of variables and constraints based on client choices, you may find that your scripts become unwieldy. This tutorial demonstrates a variety of ways in which you can structure larger JuMP models to improve their readability and maintainability.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"tip: Tip\nThis tutorial is more advanced than the other \"Getting started\" tutorials. It's in the \"Getting started\" section to give you an early preview of how JuMP makes it easy to structure larger models. However, if you are new to JuMP you may want to briefly skim the tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Overview","page":"Design patterns for larger models","title":"Overview","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"This tutorial uses explanation-by-example. We're going to start with a simple knapsack model, and then expand it to add various features and structure.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#A-simple-script","page":"Design patterns for larger models","title":"A simple script","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Your first prototype of a JuMP model is probably a script that uses a small set of hard-coded data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"using JuMP, HiGHS\nprofit = [5, 3, 2, 7, 4]\nweight = [2, 8, 4, 2, 5]\ncapacity = 10\nN = 5\nmodel = Model(HiGHS.Optimizer)\n@variable(model, x[1:N], Bin)\n@objective(model, Max, sum(profit[i] * x[i] for i in 1:N))\n@constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The benefits of this approach are:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"it is quick to code\nit is quick to make changes.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The downsides include:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"all variables are global (read Performance tips)\nit is easy to introduce errors, for example, having profit and weight be vectors of different lengths, or not match N\nthe solution, x[i], is hard to interpret without knowing the order in which we provided the data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Wrap-the-model-in-a-function","page":"Design patterns for larger models","title":"Wrap the model in a function","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"A good next step is to wrap your model in a function. This is useful for a few reasons:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"it removes global variables\nit encapsulates the JuMP model and forces you to clarify your inputs and outputs\nwe can add some error checking.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_1(profit::Vector, weight::Vector, capacity::Real)\n    if length(profit) != length(weight)\n        throw(DimensionMismatch(\"profit and weight are different sizes\"))\n    end\n    N = length(weight)\n    model = Model(HiGHS.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @objective(model, Max, sum(profit[i] * x[i] for i in 1:N))\n    @constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\n    optimize!(model)\n    return value.(x)\nend\n\nsolve_knapsack_1([5, 3, 2, 7, 4], [2, 8, 4, 2, 5], 10)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Create-better-data-structures","page":"Design patterns for larger models","title":"Create better data structures","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Although we can check for errors like mis-matched vector lengths, if you start to develop models with a lot of data, keeping track of vectors and lengths and indices is fragile and a common source of bugs. A good solution is to use Julia's type system to create an abstraction over your data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For example, we can create a struct that represents a single object, with a constructor that lets us validate assumptions on the input data:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct KnapsackObject\n    profit::Float64\n    weight::Float64\n    function KnapsackObject(profit::Float64, weight::Float64)\n        if weight < 0\n            throw(DomainError(\"Weight of object cannot be negative\"))\n        end\n        return new(profit, weight)\n    end\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"as well as a struct that holds a dictionary of objects and the knapsack's capacity:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct KnapsackData\n    objects::Dict{String,KnapsackObject}\n    capacity::Float64\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Here's what our data might look like now:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"objects = Dict(\n    \"apple\" => KnapsackObject(5.0, 2.0),\n    \"banana\" => KnapsackObject(3.0, 8.0),\n    \"cherry\" => KnapsackObject(2.0, 4.0),\n    \"date\" => KnapsackObject(7.0, 2.0),\n    \"eggplant\" => KnapsackObject(4.0, 5.0),\n)\ndata = KnapsackData(objects, 10.0)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If you want, you can add custom printing to make it easier to visualize:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function Base.show(io::IO, data::KnapsackData)\n    println(io, \"A knapsack with capacity $(data.capacity) and possible items:\")\n    for (k, v) in data.objects\n        println(\n            io,\n            \"  $(rpad(k, 8)) : profit = $(v.profit), weight = $(v.weight)\",\n        )\n    end\n    return\nend\n\ndata","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Then, we can re-write our solve_knapsack function to take our KnapsackData as input:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_2(data::KnapsackData)\n    model = Model(HiGHS.Optimizer)\n    @variable(model, x[keys(data.objects)], Bin)\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend\n\nsolve_knapsack_2(data)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Read-in-data-from-files","page":"Design patterns for larger models","title":"Read in data from files","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Having a data structure is a good step. But it is still annoying that we have to hard-code the data into Julia. A good next step is to separate the data into an external file format; JSON is a common choice.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"json_data = \"\"\"\n{\n    \"objects\": {\n        \"apple\": {\"profit\": 5.0, \"weight\": 2.0},\n        \"banana\": {\"profit\": 3.0, \"weight\": 8.0},\n        \"cherry\": {\"profit\": 2.0, \"weight\": 4.0},\n        \"date\": {\"profit\": 7.0, \"weight\": 2.0},\n        \"eggplant\": {\"profit\": 4.0, \"weight\": 5.0}\n    },\n    \"capacity\": 10.0\n}\n\"\"\"\ntemp_dir = mktempdir()\nknapsack_json_filename = joinpath(temp_dir, \"knapsack.json\")\n# Instead of writing a new file here you could replace `knapsack_json_filename`\n# with the path to a local file.\nwrite(knapsack_json_filename, json_data);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now let's write a function that reads this file and builds a KnapsackData object:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import JSON\n\nfunction read_data(filename)\n    d = JSON.parsefile(filename)\n    return KnapsackData(\n        Dict(\n            k => KnapsackObject(v[\"profit\"], v[\"weight\"]) for\n            (k, v) in d[\"objects\"]\n        ),\n        d[\"capacity\"],\n    )\nend\n\ndata = read_data(knapsack_json_filename)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-options-via-if-else","page":"Design patterns for larger models","title":"Add options via if-else","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"At this point, we have data in a file format which we can load and solve a single problem. For many users, this might be sufficient. However, at some point you may be asked to add features like \"but what if we want to take more than one of a particular item?\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If this is the first time that you've been asked to add a feature, adding options via if-else statements is a good approach. For example, we might write:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_3(data::KnapsackData; binary_knapsack::Bool)\n    model = Model(HiGHS.Optimizer)\n    if binary_knapsack\n        @variable(model, x[keys(data.objects)], Bin)\n    else\n        @variable(model, x[keys(data.objects)] >= 0, Int)\n    end\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we can solve the binary knapsack:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_3(data; binary_knapsack = true)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"And an integer knapsack where we can take more than one copy of each item:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_3(data; binary_knapsack = false)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-configuration-options-via-dispatch","page":"Design patterns for larger models","title":"Add configuration options via dispatch","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If you get repeated requests to add different options, you'll quickly find yourself in a mess of different flags and if-else statements. It's hard to write, hard to read, and hard to ensure you haven't introduced any bugs. A good solution is to use Julia's type dispatch to control the configuration of the model. The easiest way to explain this is by example.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"First, start by defining a new abstract type, as well as new subtypes for each of our options. These types are going to control the configuration of the knapsack model.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"abstract type AbstractConfiguration end\n\nstruct BinaryKnapsackConfig <: AbstractConfiguration end\n\nstruct IntegerKnapsackConfig <: AbstractConfiguration end","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Then, we rewrite our solve_knapsack function to take a config argument, and we introduce an add_knapsack_variables function to abstract the creation of our variables.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_4(data::KnapsackData, config::AbstractConfiguration)\n    model = Model(HiGHS.Optimizer)\n    x = add_knapsack_variables(model, data, config)\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For the binary knapsack problem, add_knapsack_variables looks like this:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    ::BinaryKnapsackConfig,\n)\n    return @variable(model, x[keys(data.objects)], Bin)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For the integer knapsack problem, add_knapsack_variables looks like this:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    ::IntegerKnapsackConfig,\n)\n    return @variable(model, x[keys(data.objects)] >= 0, Int)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we can solve the binary knapsack:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_4(data, BinaryKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"and the integer knapsack problem:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_4(data, IntegerKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The main benefit of the dispatch approach is that you can quickly add new options without needing to modify the existing code. For example:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct UpperBoundedKnapsackConfig <: AbstractConfiguration\n    limit::Int\nend\n\nfunction add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    config::UpperBoundedKnapsackConfig,\n)\n    return @variable(model, 0 <= x[keys(data.objects)] <= config.limit, Int)\nend\n\nsolve_knapsack_4(data, UpperBoundedKnapsackConfig(3))","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Generalize-constraints-and-objectives","page":"Design patterns for larger models","title":"Generalize constraints and objectives","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"It's easy to extend the dispatch approach to constraints and objectives as well. The key points to notice in the next two functions are that:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"we can access registered variables via model[:x]\nwe can define generic functions which accept any AbstractConfiguration as a configuration argument. That means we can implement a single method and have it apply to multiple configuration types.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_constraints(\n    model::Model,\n    data::KnapsackData,\n    ::AbstractConfiguration,\n)\n    x = model[:x]\n    @constraint(\n        model,\n        capacity_constraint,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    return\nend\n\nfunction add_knapsack_objective(\n    model::Model,\n    data::KnapsackData,\n    ::AbstractConfiguration,\n)\n    x = model[:x]\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    return\nend\n\nfunction solve_knapsack_5(data::KnapsackData, config::AbstractConfiguration)\n    model = Model(HiGHS.Optimizer)\n    add_knapsack_variables(model, data, config)\n    add_knapsack_constraints(model, data, config)\n    add_knapsack_objective(model, data, config)\n    optimize!(model)\n    return value.(model[:x])\nend\n\nsolve_knapsack_5(data, BinaryKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Remove-solver-dependence,-add-error-checks","page":"Design patterns for larger models","title":"Remove solver dependence, add error checks","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Compared to where we started, our knapsack model is now significantly different. We've wrapped it in a function, defined some data types, and introduced configuration options to control the variables and constraints that get added. There are a few other steps we can do to further improve things:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"remove the dependence on HiGHS\nadd checks that we found an optimal solution\nadd a helper function to avoid the need to explicitly construct the data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_6(\n    optimizer,\n    data::KnapsackData,\n    config::AbstractConfiguration,\n)\n    model = Model(optimizer)\n    add_knapsack_variables(model, data, config)\n    add_knapsack_constraints(model, data, config)\n    add_knapsack_objective(model, data, config)\n    optimize!(model)\n    if termination_status(model) != OPTIMAL\n        @warn(\"Model not solved to optimality\")\n        return nothing\n    end\n    return value.(model[:x])\nend\n\nfunction solve_knapsack_6(\n    optimizer,\n    data::String,\n    config::AbstractConfiguration,\n)\n    return solve_knapsack_6(optimizer, read_data(data), config)\nend\n\nsolution = solve_knapsack_6(\n    HiGHS.Optimizer,\n    knapsack_json_filename,\n    BinaryKnapsackConfig(),\n)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Create-a-module","page":"Design patterns for larger models","title":"Create a module","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we're ready to expose our model to the wider world. That might be as part of a larger Julia project that we're contributing to, or as a stand-alone script that we can run on-demand. In either case, it's good practice to wrap everything in a module. This further encapsulates our code into a single namespace, and we can add documentation in the form of docstrings.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Some good rules to follow when creating a module are:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"use import in a module instead of using to make it clear which functions are from which packages\nuse _ to start function and type names that are considered private\nadd docstrings to all public variables and functions.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"module KnapsackModel\n\nimport JuMP\nimport JSON\n\nstruct _KnapsackObject\n    profit::Float64\n    weight::Float64\n    function _KnapsackObject(profit::Float64, weight::Float64)\n        if weight < 0\n            throw(DomainError(\"Weight of object cannot be negative\"))\n        end\n        return new(profit, weight)\n    end\nend\n\nstruct _KnapsackData\n    objects::Dict{String,_KnapsackObject}\n    capacity::Float64\nend\n\nfunction _read_data(filename)\n    d = JSON.parsefile(filename)\n    return _KnapsackData(\n        Dict(\n            k => _KnapsackObject(v[\"profit\"], v[\"weight\"]) for\n            (k, v) in d[\"objects\"]\n        ),\n        d[\"capacity\"],\n    )\nend\n\nabstract type _AbstractConfiguration end\n\n\"\"\"\n    BinaryKnapsackConfig()\n\nCreate a binary knapsack problem where each object can be taken 0 or 1 times.\n\"\"\"\nstruct BinaryKnapsackConfig <: _AbstractConfiguration end\n\n\"\"\"\n    IntegerKnapsackConfig()\n\nCreate an integer knapsack problem where each object can be taken any number of\ntimes.\n\"\"\"\nstruct IntegerKnapsackConfig <: _AbstractConfiguration end\n\nfunction _add_knapsack_variables(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::BinaryKnapsackConfig,\n)\n    return JuMP.@variable(model, x[keys(data.objects)], Bin)\nend\n\nfunction _add_knapsack_variables(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::IntegerKnapsackConfig,\n)\n    return JuMP.@variable(model, x[keys(data.objects)] >= 0, Int)\nend\n\nfunction _add_knapsack_constraints(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::_AbstractConfiguration,\n)\n    x = model[:x]\n    JuMP.@constraint(\n        model,\n        capacity_constraint,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    return\nend\n\nfunction _add_knapsack_objective(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::_AbstractConfiguration,\n)\n    x = model[:x]\n    JuMP.@objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    return\nend\n\nfunction _solve_knapsack(\n    optimizer,\n    data::_KnapsackData,\n    config::_AbstractConfiguration,\n)\n    model = JuMP.Model(optimizer)\n    _add_knapsack_variables(model, data, config)\n    _add_knapsack_constraints(model, data, config)\n    _add_knapsack_objective(model, data, config)\n    JuMP.optimize!(model)\n    if JuMP.termination_status(model) != JuMP.OPTIMAL\n        @warn(\"Model not solved to optimality\")\n        return nothing\n    end\n    return JuMP.value.(model[:x])\nend\n\n\"\"\"\n    solve_knapsack(\n        optimizer,\n        knapsack_json_filename::String,\n        config::_AbstractConfiguration,\n    )\n\nSolve the knapsack problem and return the optimal primal solution\n\n# Arguments\n\n * `optimizer` : an object that can be passed to `JuMP.Model` to construct a new\n   JuMP model.\n * `knapsack_json_filename` : the filename of a JSON file containing the data for the\n   problem.\n * `config` : an object to control the type of knapsack model constructed.\n   Valid options are:\n    * `BinaryKnapsackConfig()`\n    * `IntegerKnapsackConfig()`\n\n# Returns\n\n * If an optimal solution exists: a `JuMP.DenseAxisArray` that maps the `String`\n   name of each object to the number of objects to pack into the knapsack.\n * Otherwise, `nothing`, indicating that the problem does not have an optimal\n   solution.\n\n# Examples\n\n```julia\nsolution = solve_knapsack(\n    HiGHS.Optimizer,\n    \"path/to/data.json\",\n    BinaryKnapsackConfig(),\n)\n```\n\n```julia\nsolution = solve_knapsack(\n    MOI.OptimizerWithAttributes(HiGHS.Optimizer, \"output_flag\" => false),\n    \"path/to/data.json\",\n    IntegerKnapsackConfig(),\n)\n```\n\"\"\"\nfunction solve_knapsack(\n    optimizer,\n    knapsack_json_filename::String,\n    config::_AbstractConfiguration,\n)\n    data = _read_data(knapsack_json_filename)\n    return _solve_knapsack(optimizer, data, config)\nend\n\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Finally, you can call your model:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import .KnapsackModel\n\nKnapsackModel.solve_knapsack(\n    HiGHS.Optimizer,\n    knapsack_json_filename,\n    KnapsackModel.BinaryKnapsackConfig(),\n)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"note: Note\nThe . in .KnapsackModel denotes that it is a submodule and not a separate package that we installed with Pkg.add. If you put the KnapsackModel in a separate file, load it with:include(\"path/to/KnapsackModel.jl\")\nimport .KnapsackModel","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-tests","page":"Design patterns for larger models","title":"Add tests","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"As a final step, you should add tests for your model. This often means testing on a small problem for which you can work out the optimal solution by hand. The Julia standard library Test has good unit-testing functionality.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import .KnapsackModel\nusing Test\n\n@testset \"KnapsackModel\" begin\n    @testset \"feasible_binary_knapsack\" begin\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            knapsack_json_filename,\n            KnapsackModel.BinaryKnapsackConfig(),\n        )\n        @test isapprox(x[\"apple\"], 1, atol = 1e-5)\n        @test isapprox(x[\"banana\"], 0, atol = 1e-5)\n        @test isapprox(x[\"cherry\"], 0, atol = 1e-5)\n        @test isapprox(x[\"date\"], 1, atol = 1e-5)\n        @test isapprox(x[\"eggplant\"], 1, atol = 1e-5)\n    end\n    @testset \"feasible_integer_knapsack\" begin\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            knapsack_json_filename,\n            KnapsackModel.IntegerKnapsackConfig(),\n        )\n        @test isapprox(x[\"apple\"], 0, atol = 1e-5)\n        @test isapprox(x[\"banana\"], 0, atol = 1e-5)\n        @test isapprox(x[\"cherry\"], 0, atol = 1e-5)\n        @test isapprox(x[\"date\"], 5, atol = 1e-5)\n        @test isapprox(x[\"eggplant\"], 0, atol = 1e-5)\n    end\n    @testset \"infeasible_binary_knapsack\" begin\n        dir = mktempdir()\n        infeasible_filename = joinpath(dir, \"infeasible.json\")\n        write(\n            infeasible_filename,\n            \"\"\"{\n                \"objects\": {\n                    \"apple\": {\"profit\": 5.0, \"weight\": 2.0},\n                    \"banana\": {\"profit\": 3.0, \"weight\": 8.0},\n                    \"cherry\": {\"profit\": 2.0, \"weight\": 4.0},\n                    \"date\": {\"profit\": 7.0, \"weight\": 2.0},\n                    \"eggplant\": {\"profit\": 4.0, \"weight\": 5.0}\n                },\n                \"capacity\": -10.0\n            }\"\"\",\n        )\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            infeasible_filename,\n            KnapsackModel.BinaryKnapsackConfig(),\n        )\n        @test x === nothing\n    end\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"tip: Tip\nPlace these tests in a separate file test_knapsack_model.jl so that you can run the tests by adding include(\"test_knapsack_model.jl\") to any file where needed.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Next-steps","page":"Design patterns for larger models","title":"Next steps","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"We've only briefly scratched the surface of ways to create and structure large JuMP models, so consider this tutorial a starting point, rather than a comprehensive list of all the possible ways to structure JuMP models.  If you are embarking on a large project that uses JuMP, a good next step is to look at ways people have written large JuMP projects \"in the wild.\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Here are some good examples (all co-incidentally related to energy):","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"AnyMOD.jl\nJuMP-dev 2021 talk\nsource code\nPowerModels.jl\nJuMP-dev 2021 talk\nsource code\nPowerSimulations.jl\nJuliaCon 2021 talk\nsource code\nUnitCommitment.jl\nJuMP-dev 2021 talk\nsource code","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"EditURL = \"multi.jl\"","category":"page"},{"location":"tutorials/linear/multi/#The-multi-commodity-flow-problem","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial is a JuMP implementation of the multi-commodity transportation model described in AMPL: A Modeling Language for Mathematical Programming, by R. Fourer, D.M. Gay and B.W. Kernighan.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate creating a JuMP model from an SQLite database.","category":"page"},{"location":"tutorials/linear/multi/#Required-packages","page":"The multi-commodity flow problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport SQLite\nimport Tables\n\nconst DBInterface = SQLite.DBInterface","category":"page"},{"location":"tutorials/linear/multi/#Formulation","page":"The multi-commodity flow problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The multi-commondity flow problem is a simple extension of The transportation problem to multiple types of products. Briefly, we start with the formulation of the transportation problem:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"beginaligned\nmin   sum_i in O j in D c_ij x_ij \nst   sum_j in D x_i j le s_i  forall i in O \n       sum_i in O x_i j = d_j  forall j in D \n       x_i j ge 0  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"but introduce a set of products P, resulting in:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"beginaligned\nmin   sum_i in O j in D k in P c_ijk x_ijk \nst   sum_j in D x_i j k le s_ik  forall i in O k in P \n       sum_i in O x_i j k = d_jk  forall j in D k in P \n       x_i jk ge 0  forall i in O j in D k in P \n       sum_k in P x_i j k le u_ij  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Note that the last constraint is new; it says that there is a maximum quantity of goods (of any type) that can be transported from origin i to destination j.","category":"page"},{"location":"tutorials/linear/multi/#Data","page":"The multi-commodity flow problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"For the purpose of this tutorial, the JuMP repository contains an example database called multi.sqlite.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"filename = joinpath(@__DIR__, \"multi.sqlite\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"To run locally, download multi.sqlite and update filename appropriately.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Load the database using SQLite.DB:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"db = SQLite.DB(filename)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"A quick way to see the schema of the database is via SQLite.tables:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"SQLite.tables(db)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We interact with the database by executing queries, and then piping the results to an appropriate table. One example is a DataFrame:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"DBInterface.execute(db, \"SELECT * FROM locations\") |> DataFrames.DataFrame","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"But other table types are supported, such as Tables.rowtable:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"DBInterface.execute(db, \"SELECT * FROM locations\") |> Tables.rowtable","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"A rowtable is a Vector of NamedTuples.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"You can construct more complicated SQL queries:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"origins =\n    DBInterface.execute(\n        db,\n        \"SELECT location FROM locations WHERE type = \\\"origin\\\"\",\n    ) |> Tables.rowtable","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"But for our purpose, we just want the list of strings:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"origins = map(y -> y.location, origins)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We can compose these two operations to get a list of destinations:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"destinations =\n    DBInterface.execute(\n        db,\n        \"SELECT location FROM locations WHERE type = \\\"destination\\\"\",\n    ) |>\n    Tables.rowtable |>\n    x -> map(y -> y.location, x)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"And a list of products from our products table:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"products =\n    DBInterface.execute(db, \"SELECT product FROM products\") |>\n    Tables.rowtable |>\n    x -> map(y -> y.product, x)","category":"page"},{"location":"tutorials/linear/multi/#JuMP-formulation","page":"The multi-commodity flow problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We start by creating a model and our decision variables:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[origins, destinations, products] >= 0)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"One approach when working with databases is to extract all of the data into a Julia datastructure. For example, let's pull the cost table into a DataFrame and then construct our objective by iterating over the rows of the DataFrame:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"cost = DBInterface.execute(db, \"SELECT * FROM cost\") |> DataFrames.DataFrame\n@objective(\n    model,\n    Max,\n    sum(r.cost * x[r.origin, r.destination, r.product] for r in eachrow(cost)),\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"If we don't want to use a DataFrame, we can use a Tables.rowtable instead:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"supply = DBInterface.execute(db, \"SELECT * FROM supply\") |> Tables.rowtable\nfor r in supply\n    @constraint(model, sum(x[r.origin, :, r.product]) <= r.supply)\nend","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Another approach is to execute the query, and then to iterate through the rows of the query using Tables.rows:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"demand = DBInterface.execute(db, \"SELECT * FROM demand\")\nfor r in Tables.rows(demand)\n    @constraint(model, sum(x[:, r.destination, r.product]) == r.demand)\nend","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"warning: Warning\nIterating through the rows of a query result works by incrementing a cursor inside the database. As a consequence, you cannot call Tables.rows twice on the same query result.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The SQLite queries can be arbitrarily complex. For example, here's a query which builds every possible origin-destination pair:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"od_pairs = DBInterface.execute(\n    db,\n    \"\"\"\n    SELECT a.location as 'origin',\n           b.location as 'destination'\n    FROM locations a\n    INNER JOIN locations b\n    ON a.type = 'origin' AND b.type = 'destination'\n    \"\"\",\n)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"With a constraint that we cannot send more than 625 units between each pair:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"for r in Tables.rows(od_pairs)\n    @constraint(model, sum(x[r.origin, r.destination, :]) <= 625)\nend","category":"page"},{"location":"tutorials/linear/multi/#Solution","page":"The multi-commodity flow problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Finally, we can optimize the model:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"and print the solution:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"begin\n    println(\"         \", join(products, ' '))\n    for o in origins, d in destinations\n        v = lpad.([round(Int, value(x[o, d, p])) for p in products], 5)\n        println(o, \" \", d, \" \", join(replace.(v, \"   0\" => \"  . \"), \" \"))\n    end\nend","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"EditURL = \"finance.jl\"","category":"page"},{"location":"tutorials/linear/finance/#Financial-modeling-problems","page":"Financial modeling problems","title":"Financial modeling problems","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Optimization models play an increasingly important role in financial decisions. Many computational finance problems can be solved efficiently using modern optimization techniques.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"In this tutorial we will discuss two such examples taken from the book Optimization Methods in Finance.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/finance/#Short-term-financing","page":"Financial modeling problems","title":"Short-term financing","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Corporations routinely face the problem of financing short term cash commitments such as the following:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Month Jan Feb Mar Apr May Jun\nNet Cash Flow -150 -100 200 -200 50 300","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Net cash flow requirements are given in thousands of dollars. The company has the following sources of funds:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"A line of credit of up to $100K at an interest rate of 1% per month,\nIn any one of the first three months, it can issue 90-day commercial paper bearing a total interest of 2% for the 3-month period,\nExcess funds can be invested at an interest rate of 0.3% per month.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Our task is to find out the most economical way to use these 3 sources such that we end up with the most amount of money at the end of June.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We model this problem in the following manner:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We will use the following decision variables:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"the amount u_i drawn from the line of credit in month i\nthe amount v_i of commercial paper issued in month i\nthe excess funds w_i in month i","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Here we have three types of constraints:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"for every month, cash inflow = cash outflow for each month\nupper bounds on u_i\nnonnegativity of the decision variables u_i, v_i and w_i.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Our objective will be to simply maximize the company's wealth in June, which say we represent with the variable m.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"financing = Model(HiGHS.Optimizer)\n\n@variables(financing, begin\n    0 <= u[1:5] <= 100\n    0 <= v[1:3]\n    0 <= w[1:5]\n    m\nend)\n\n@objective(financing, Max, m)\n\n@constraints(\n    financing,\n    begin\n        u[1] + v[1] - w[1] == 150 # January\n        u[2] + v[2] - w[2] - 1.01u[1] + 1.003w[1] == 100 # February\n        u[3] + v[3] - w[3] - 1.01u[2] + 1.003w[2] == -200 # March\n        u[4] - w[4] - 1.02v[1] - 1.01u[3] + 1.003w[3] == 200 # April\n        u[5] - w[5] - 1.02v[2] - 1.01u[4] + 1.003w[4] == -50 # May\n        -m - 1.02v[3] - 1.01u[5] + 1.003w[5] == -300 # June\n    end\n)\n\noptimize!(financing)\n\nobjective_value(financing)","category":"page"},{"location":"tutorials/linear/finance/#Combinatorial-auctions","page":"Financial modeling problems","title":"Combinatorial auctions","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"In many auctions, the value that a bidder has for a set of items may not be the sum of the values that he has for individual items.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Examples are equity trading, electricity markets, pollution right auctions and auctions for airport landing slots.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"To take this into account, combinatorial auctions allow the bidders to submit bids on combinations of items.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Let M=12 ldots m be the set of items that the auctioneer has to sell. A bid is a pair B_j=left(S_j p_jright) where S_j subseteq M is a nonempty set of items and p_j is the price offer for this set.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Suppose that the auctioneer has received n bids B_1 B_2 ldots B_n The goal of this problem is to help an auctioneer determine the winners in order to maximize his revenue.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We model this problem by taking a decision variable y_j for every bid. We add a constraint that each item i is sold at most once. This gives us the following model:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"beginaligned\nmax  sum_i=1^n p_j y_j \ntext  st    sum_j  i in S_j y_j leq 1  forall i=12 ldots m \n y_j in01  forall j in12 ldots n\nendaligned","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"bid_values = [6 3 12 12 8 16]\nbid_items = [[1], [2], [3 4], [1 3], [2 4], [1 3 4]]\n\nauction = Model(HiGHS.Optimizer)\n@variable(auction, y[1:6], Bin)\n@objective(auction, Max, sum(y' .* bid_values))\nfor i in 1:6\n    @constraint(auction, sum(y[j] for j in 1:6 if i in bid_items[j]) <= 1)\nend\n\noptimize!(auction)\n\nobjective_value(auction)","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"value.(y)","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"EditURL = \"https://github.com/GAMS-dev/GAMS.jl/blob/c5dee9f929e9d2f4433ae09fa92b8d872c9c43e0/README.md\"","category":"page"},{"location":"packages/GAMS/#GAMS.jl","page":"GAMS-dev/GAMS.jl","title":"GAMS.jl","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl provides a MathOptInterface Optimizer to solve JuMP models using GAMS.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS comes with dozens of supported solvers. Among them are: ALPHAECP, ANTIGONE, BARON, CBC, CONOPT, CPLEX, DICOPT, GUROBI, IPOPT, KNITRO, LINDO, LINDOGLOBAL, MINOS, MOSEK, NLPEC, PATH, QUADMINOS, SBB, SHOT, SCIP, SNOPT, SOPLEX, XPRESS. Find a complete list here.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl supports the following JuMP features:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"linear, quadratic and nonlinear (convex and non-convex) objective and constraints\ncontinuous, binary, integer, semi-continuous and semi-integer variables\nSOS1 and SOS2 sets\ncomplementarity constraints","category":"page"},{"location":"packages/GAMS/#Installation","page":"GAMS-dev/GAMS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Download GAMS and obtain a GAMS license. Please note that GAMS also offers a free community license.\n(optional) Add the GAMS system directory to the PATH variable in order to find GAMS automatically.\nInstall GAMS.jl using the Julia package manager:\nusing Pkg\nPkg.add(\"GAMS\")","category":"page"},{"location":"packages/GAMS/#Usage","page":"GAMS-dev/GAMS.jl","title":"Usage","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Using GAMS as optimizer for your JuMP model:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"using GAMS, JuMP\nmodel = Model(GAMS.Optimizer)","category":"page"},{"location":"packages/GAMS/#GAMS-System","page":"GAMS-dev/GAMS.jl","title":"GAMS System","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"If the GAMS system directory has been added to the PATH variable (you can check this with print(ENV[\"PATH\"])), GAMS.jl will find it automatically. Otherwise, or if you like to switch between systems, the system directory can be specified by (one of the following):","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"SysDir\", \"<gams_system_dir>\")\nset_optimizer_attribute(model, GAMS.SysDir(), \"<gams_system_dir>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Analogously, you can specify a working directory with \"WorkDir\" or GAMS.WorkDir(). If no working directory has been set, GAMS.jl will create a temporary one.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"If you want to use the same GAMS workspace (same system and working directory) for multiple models, you can create a GAMSWorkspace first with either of the following","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"ws = GAMS.GAMSWorkspace()\nws = GAMS.GAMSWorkspace(\"<gams_system_dir>\")\nws = GAMS.GAMSWorkspace(\"<gams_system_dir>\", \"<gams_working_dir>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"and then pass it to your models:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"model = Model(() -> GAMS.Optimizer(ws))","category":"page"},{"location":"packages/GAMS/#GAMS-Options","page":"GAMS-dev/GAMS.jl","title":"GAMS Options","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS command line options can be specified by","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"<option>\", \"<solver_name>\")\nset_optimizer_attribute(model, GAMS.<option>(), \"<solver_name>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"where <option> is either HoldFixed, IterLim, License, LogOption, NodLim, OptCA, OptCR, ResLim, Solver, Threads, Trace, TraceOpt as well as LP, MIP, RMIP, NLP, DNLP, CNS, MINLP, RMINLP, QCP, MIQCP, RMIQCP, MCP or MPEC. Note that GAMS.ResLim() is equivalent to MOI.TimeLimitSec() and GAMS.Threads() to MOI.NumberOfThreads(). Options LimCol, LimRow, SolPrint and SolveLink cannot be changed and are set to 0, 0, 0 and 5, respectively.","category":"page"},{"location":"packages/GAMS/#Model-Type","page":"GAMS-dev/GAMS.jl","title":"Model Type","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl will automatically choose a GAMS model type for you. Choosing a different model type:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, GAMS.ModelType(), \"<model_type>\")","category":"page"},{"location":"packages/GAMS/#GAMS-Solver-Options","page":"GAMS-dev/GAMS.jl","title":"GAMS Solver Options","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Specifying GAMS solver options:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"<solver_option_name>\", <option_value>)","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Note that passing a solver option is only valid when explicitly choosing a GAMS solver and not using the default.","category":"page"},{"location":"packages/GAMS/#GAMS-Names-vs.-JuMP-Names","page":"GAMS-dev/GAMS.jl","title":"GAMS Names vs. JuMP Names","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS uses generated variable and constraint names although it is possible to pass the JuMP names to the GAMS optimizer, because GAMS is more restrictive when it comes to variable and constraint naming. Use the attributes GeneratedVariableName, GeneratedConstraintName, OriginalVariableName, OriginalConstraintName to query a GAMS symbol name from a JuMP symbol and vice versa. This can help for debugging, for example in case of GAMS compilation errors. For example:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"using GAMS\n\nmodel = direct_model(GAMS.Optimizer())\n\n@variable(model, x[1:2,1:3] >= 0)\n@constraint(model, c[i = 1:2], sum(x[i,j] for j = 1:3) <= 10)\n\nMOI.get(model, GAMS.GeneratedVariableName(), x[2,2]) # returns x4\nMOI.get(model, GAMS.OriginalVariableName(\"x6\"))      # returns x[2,3]\nMOI.get(model, GAMS.OriginalVariableName(\"x10\"))     # returns nothing\n\nMOI.get(model, GAMS.GeneratedConstraintName(), c[2]) # returns eq2\nMOI.get(model, GAMS.OriginalConstraintName(\"eq1\"))   # returns c[1]\nMOI.get(model, GAMS.OriginalConstraintName(\"eq10\"))  # returns nothing","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Note that JuMP direct-mode is used.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"EditURL = \"https://github.com/jump-dev/Pavito.jl/blob/v0.3.7/README.md\"","category":"page"},{"location":"packages/Pavito/#Pavito.jl","page":"jump-dev/Pavito.jl","title":"Pavito.jl","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"(Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito.jl is a mixed-integer convex programming (MICP) solver package written in Julia.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"MICP problems are convex, except for restrictions that some variables take binary or integer values.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito solves MICP problems by constructing sequential polyhedral outer-approximations of the convex feasible set, similar to Bonmin.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito accesses state-of-the-art MILP solvers and continuous, derivative-based nonlinear programming (NLP) solvers through MathOptInterface.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For algorithms that use a conic solver instead of an NLP solver, use Pajarito. Pajarito is a robust mixed-integer conic solver that can handle such established problem classes as mixed-integer second-order cone programming (MISOCP) and mixed-integer semidefinite programming (MISDP).","category":"page"},{"location":"packages/Pavito/#License","page":"jump-dev/Pavito.jl","title":"License","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Pavito/#Installation","page":"jump-dev/Pavito.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Install Pavito using Pkg.add:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"import Pkg\nPkg.add(\"Pavito\")","category":"page"},{"location":"packages/Pavito/#Use-with-JuMP","page":"jump-dev/Pavito.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"To use Pavito with JuMP, use Pavito.Optimizer:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"using JuMP, Pavito\nimport GLPK, Ipopt\nmodel = Model(\n    optimizer_with_attributes(\n        Pavito.Optimizer,\n        \"mip_solver\" => optimizer_with_attributes(GLPK.Optimizer),\n        \"cont_solver\" =>\n            optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0),\n    ),\n)","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The algorithm implemented by Pavito itself is relatively simple; most of the hard work is performed by the MILP solver passed as mip_solver and the NLP solver passed as cont_solver.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The performance of Pavito depends on these two types of solvers.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For better performance, you should use a commercial MILP solver such as CPLEX or Gurobi.","category":"page"},{"location":"packages/Pavito/#Options","page":"jump-dev/Pavito.jl","title":"Options","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The following optimizer attributes can set to a Pavito.Optimizer to modify its behavior:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"log_level::Int Verbosity flag: 0 for quiet, higher for basic solve info\ntimeout::Float64 Time limit for algorithm (in seconds)\nrel_gap::Float64 Relative optimality gap termination condition\nmip_solver_drives::Bool Let MILP solver manage convergence (\"branch and cut\")\nmip_solver::MOI.OptimizerWithAttributes MILP solver\ncont_solver::MOI.OptimizerWithAttributes Continuous NLP solver","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito is not yet numerically robust and may require tuning of parameters to improve convergence.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"If the default parameters don't work for you, please let us know by opening an issue.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For improved Pavito performance, MILP solver integrality tolerance and feasibility tolerances should typically be tightened, for example to 1e-8.","category":"page"},{"location":"packages/Pavito/#Bug-reports-and-support","page":"jump-dev/Pavito.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Please report any issues via the GitHub issue tracker. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. The Optimization (Mathematical) category on Discourse is appropriate for general discussion.","category":"page"},{"location":"developers/contributing/#How-to-contribute-to-JuMP","page":"Contributing","title":"How to contribute to JuMP","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Welcome, this document explains some ways you can contribute to JuMP.","category":"page"},{"location":"developers/contributing/#Code-of-Conduct","page":"Contributing","title":"Code of Conduct","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"This project and everyone participating in it is governed by the JuMP Code of Conduct. By participating, you are expected to uphold this code.","category":"page"},{"location":"developers/contributing/#Join-the-community-forum","page":"Contributing","title":"Join the community forum","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"First up, join the community forum.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The forum is a good place to ask questions about how to use JuMP. You can also use the forum to discuss possible feature requests and bugs before raising a GitHub issue (more on this below).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Aside from asking questions, the easiest way you can contribute to JuMP is to help answer questions on the forum.","category":"page"},{"location":"developers/contributing/#Join-the-developer-chatroom","page":"Contributing","title":"Join the developer chatroom","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If you're interested in contributing code to JuMP, the next place to join is the developer chatroom. Let us know what you have in mind, and we can point you in the right direction.","category":"page"},{"location":"developers/contributing/#Improve-the-documentation","page":"Contributing","title":"Improve the documentation","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Chances are, if you asked (or answered) a question on the community forum, then it is a sign that the documentation could be improved. Moreover, since it is your question, you are probably the best-placed person to improve it.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The docs are written in Markdown and are built using Documenter.jl. You can find the source of all the docs here.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If your change is small (like fixing typos, or one or two sentence corrections), the easiest way to do this is via GitHub's online editor. (GitHub has help on how to do this.)","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If your change is larger, or touches multiple files, you will need to make the change locally and then use Git to submit a pull request. (See Contribute code to JuMP below for more on this.)","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you need any help, come join the developer chatroom and we will walk you through the process.","category":"page"},{"location":"developers/contributing/#File-a-bug-report","page":"Contributing","title":"File a bug report","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Another way to contribute to JuMP is to file bug reports.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Make sure you read the info in the box where you write the body of the issue before posting. You can also find a copy of that info here.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you're unsure whether you have a real bug, post on the community forum first. Someone will either help you fix the problem, or let you know the most appropriate place to open a bug report.","category":"page"},{"location":"developers/contributing/#Contribute-code-to-JuMP","page":"Contributing","title":"Contribute code to JuMP","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Finally, you can also contribute code to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nIf you do not have experience with Git, GitHub, and Julia development, the first steps can be a little daunting. However, there are lots of tutorials available online, including these for:GitHub\nGit and GitHub\nGit\nJulia package developmentIf you need any help, come join the developer chatroom and we will walk you through the process.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once you are familiar with Git and GitHub, the workflow for contributing code to JuMP is similar to the following:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 1: decide what to work on","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The first step is to find an open issue (or open a new one) for the problem you want to solve. Then, before spending too much time on it, discuss what you are planning to do in the issue to see if other contributors are fine with your proposed changes. Getting feedback early can improve code quality, and avoid time spent writing code that does not get merged into JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nAt this point, remember to be patient and polite; you may get a lot of comments on your issue. However, do not be afraid. Comments mean that people are willing to help you improve the code that you are contributing to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 2: fork JuMP","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Go to https://github.com/jump-dev/JuMP.jl and click the \"Fork\" button in the top-right corner. This will create a copy of JuMP under your GitHub account.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 3: install JuMP locally","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Open Julia and run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"] dev JuMP","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"This will download the JuMP Git repository to ~/.julia/dev/JuMP. If you're on Windows, this will be C:\\\\Users\\\\<my_name>\\\\.julia\\\\dev\\\\JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\n] command means \"first type ] to enter the Julia pkg mode, then type the rest. Don't copy-paste the code directly.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 4: checkout a new branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"note: Note\nIn the following, replace any instance of GITHUB_ACCOUNT with your GitHub user name.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The next step is to checkout a development branch. In a terminal (or command prompt on Windows), run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git remote add GITHUB_ACCOUNT https://github.com/GITHUB_ACCOUNT/JuMP.jl.git\n\n$ git checkout master\n\n$ git pull\n\n$ git checkout -b my_new_branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nLines starting with $ mean \"run these in a terminal (command prompt on Windows).\"","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 5: make changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Now make any changes to the source code inside the ~/.julia/dev/JuMP directory.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Make sure you:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Follow the Style guide and run JuliaFormatter\nAdd tests and documentation for any changes or new features","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nWhen you change the source code, you'll need to restart Julia for the changes to take effect. This is a pain, so install Revise.jl.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 6a: test your code changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"To test that your changes work, run the JuMP test-suite by opening Julia and running:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"cd(\"~/.julia/dev/JuMP\")\n] activate .\n] test","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nRunning the tests might take a long time (~10–15 minutes).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you're using Revise.jl, you can also run the tests by calling include:include(\"test/runtests.jl\")This can be faster if you want to re-run the tests multiple times.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 6b: test your documentation changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Open Julia, then run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"cd(\"~/.julia/dev/JuMP/docs\")\n] activate .\ninclude(\"src/make.jl\")","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nBuilding the documentation might take a long time (~10 minutes).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf there's a problem with the tests that you don't know how to fix, don't worry. Continue to step 5, and one of the JuMP contributors will comment on your pull request telling you how to fix things.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 7: make a pull request","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once you've made changes, you're ready to push the changes to GitHub. Run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git add .\n\n$ git commit -m \"A descriptive message of the changes\"\n\n$ git push -u GITHUB_ACCOUNT my_new_branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Then go to https://github.com/jump-dev/JuMP.jl and follow the instructions that pop up to open a pull request.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 8: respond to comments","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"At this point, remember to be patient and polite; you may get a lot of comments on your pull request. However, do not be afraid. A lot of comments means that people are willing to help you improve the code that you are contributing to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"To respond to the comments, go back to step 5, make any changes, test the changes in step 6, and then make a new commit in step 7. Your PR will automatically update.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 9: cleaning up","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once the PR is merged, clean-up your Git repository ready for the next contribution.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git checkout master\n\n$ git pull","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"note: Note\nIf you have suggestions to improve this guide, please make a pull request. It's particularly helpful if you do this after your first pull request because you'll know all the parts that could be explained better.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"EditURL = \"knapsack.jl\"","category":"page"},{"location":"tutorials/linear/knapsack/#The-knapsack-problem-example","page":"The knapsack problem example","title":"The knapsack problem example","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The purpose of this tutorial is to demonstrate how to formulate and solve a simple optimization problem.","category":"page"},{"location":"tutorials/linear/knapsack/#Required-packages","page":"The knapsack problem example","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/knapsack/#Formulation","page":"The knapsack problem example","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The knapsack problem is a classical optimization problem: given a set of items and a container with a fixed capacity, choose a subset of items having the greatest combined value that will fit within the container without exceeding the capacity.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The name of the problem suggests its analogy to packing for a trip, where the baggage weight limit is the capacity and the goal is to pack the most profitable combination of belongings.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We can formulate the knapsack problem as the integer linear program:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"beginaligned\nmax   sum_i=1^n c_i x_i      \nst   sum_i=1^n w_i x_i le C \n         x_i in 01quad forall i=1ldotsn\nendaligned","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"where C is the capacity, and there is a choice between n items, with item i having weight w_i, profit c_i. Decision variable x_i is equal to 1 if the item is chosen and 0 if not.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This formulation can be written more compactly as:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"beginaligned\nmax   c^top x       \nst   w^top x le C \n         x text binary \nendaligned","category":"page"},{"location":"tutorials/linear/knapsack/#Data","page":"The knapsack problem example","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The data for the problem consists of two vectors (one for the profits and one for the weights) along with a capacity.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"There are five objects:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"n = 5;\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"For our example, we use a capacity of 10 units:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"capacity = 10.0;\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"and the profit and cost data:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"profit = [5.0, 3.0, 2.0, 7.0, 4.0];\nweight = [2.0, 8.0, 4.0, 2.0, 5.0];\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/#JuMP-formulation","page":"The knapsack problem example","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Let's begin constructing the JuMP model for our knapsack problem.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"First, we'll create a Model object for holding model elements as we construct each part. We'll also set the solver that will ultimately be called to solve the model, once it's constructed.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Next we need the decision variables representing which items are chosen:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@variable(model, x[1:n], Bin)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We now want to constrain those variables so that their combined weight is less than or equal to the given capacity:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@constraint(model, sum(weight[i] * x[i] for i in 1:n) <= capacity)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Finally, our objective is to maximize the combined profit of the chosen items:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@objective(model, Max, sum(profit[i] * x[i] for i in 1:n))","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Let's print a human-readable description of the model and check that the model looks as expected:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"print(model)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We can now solve the optimization problem and inspect the results.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The items chosen are","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"items_chosen = [i for i in 1:n if value(x[i]) > 0.5]","category":"page"},{"location":"tutorials/linear/knapsack/#Writing-a-function","page":"The knapsack problem example","title":"Writing a function","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"After working interactively, it is good practice to implement your model in a function.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The function can be used to ensure that the model is given well-defined input data with validation checks, and that the solution process went as expected.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"function solve_knapsack_problem(;\n    profit::Vector{Float64},\n    weight::Vector{Float64},\n    capacity::Float64,\n)\n    n = length(weight)\n    # The profit and weight vectors must be of equal length.\n    @assert length(profit) == n\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x[1:n], Bin)\n    @objective(model, Max, profit' * x)\n    @constraint(model, weight' * x <= capacity)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    @assert primal_status(model) == FEASIBLE_POINT\n    println(\"Objective is: \", objective_value(model))\n    println(\"Solution is:\")\n    for i in 1:n\n        print(\"x[$i] = \", round(Int, value(x[i])))\n        println(\", c[$i] / w[$i] = \", profit[i] / weight[i])\n    end\n    chosen_items = [i for i in 1:n if value(x[i]) > 0.5]\n    return return chosen_items\nend\n\nsolve_knapsack_problem(; profit = profit, weight = weight, capacity = capacity)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We observe that the chosen items (1, 4, and 5) have the best profit to weight ratio in this particular example.","category":"page"},{"location":"tutorials/linear/knapsack/#Next-steps","page":"The knapsack problem example","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Here are some things to try next:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Call the function with different data. What happens as the capacity increases?\nWhat happens if the profit and weight vectors are different lengths?\nInstead of creating a binary variable with Bin, we could have written @variable(model, 0 <= x[1:n] <= 1, Int). Verify that this formulation finds the same solution. What happens if we are allowed to take more than one of each item?","category":"page"},{"location":"developers/checklists/#Checklists","page":"Checklists","title":"Checklists","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"The purpose of this page is to collate a series of checklists for commonly performed changes to the source code of JuMP.","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"In each case, copy the checklist into the description of the pull request.","category":"page"},{"location":"developers/checklists/#Making-a-release","page":"Checklists","title":"Making a release","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"In preparation for a release, use the following checklist. These steps can be done in the same commit, or separately. The last commit should have the message \"Prep for vX.Y.Z.\"","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"## Pre-release\n\n - [ ] Check that the pinned packages in `docs/Project.toml` are updated. We pin\n       the versions so that changes in the solvers (changes in printing, small\n       numeric changes) do not break the printing of the JuMP docs in arbitrary\n       commits.\n - [ ] Check that the `rev` fields in `docs/packages.toml` are updated. We pin\n       the versions of solvers and extensions to ensure that changes to their\n       READMEs do not break the JuMP docs in arbitrary commits, and to ensure\n       that the versions are compatible with the latest JuMP and\n       MathOptInterface releases.\n - [ ] Update `docs/src/changelog.md`\n - [ ] Run https://github.com/jump-dev/JuMP.jl/actions/workflows/extension-tests.yml\n       using a `workflow_dispatch` trigger to check for any changes in JuMP that\n       broke extensions.\n - [ ] Change the version number in `Project.toml`\n - [ ] Update the links in README.md\n - [ ] The commit messages in this PR do not contain `[ci skip]`\n\n## The release\n\n - [ ] After merging this pull request, comment `[at]JuliaRegistrator register` in\n       the GitHub commit. This should automatically publish a new version to the\n       Julia registry, as well as create a tag, and rebuild the documentation\n       for this tag.\n\n       These steps can take quite a bit of time (1 hour or more), so don't be\n       surprised if the new documentation takes a while to appear. In addition,\n       the links in the README will be broken until JuliaHub fetches the new\n       version on their servers.\n\n## Post-release\n\n - [ ] Once the tag is created, update the relevant `release-` branch. The latest\n       release branch at the time of writing is `release-1.0` (we haven't\n       back-ported any patches that needed to create a `release-1.Y` branch). To\n       to update the release branch with the v1.10.0 tag, do:\n       ```\n       git checkout release-1.0\n       git pull\n       git merge v1.10.0\n       git push\n       ```","category":"page"},{"location":"developers/checklists/#Adding-a-new-solver-to-the-documentation","page":"Checklists","title":"Adding a new solver to the documentation","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"Use the following checklist when adding a new solver to the JuMP documentation.","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Check that the solver is a registered Julia package\n - [ ] Check that the solver supports the long-term support release of Julia\n - [ ] Check that the solver has a MathOptInterface wrapper\n - [ ] Check that the tests call `MOI.Test.runtests`. Some test excludes are\n       permissible, but the reason for skipping a particular test should be\n       documented.\n - [ ] Check that the README and/or documentation provides an example of how to\n       use the solver with JuMP\n\n## Documentation\n\n - [ ] Add a new row to the table in `docs/src/installation.md`\n\n## Optional\n\n - [ ] Add package metadata to `docs/packages.toml`","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"EditURL = \"simple_examples.jl\"","category":"page"},{"location":"tutorials/conic/simple_examples/#Simple-semidefinite-programming-examples","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial is a collection of examples of small conic programs from the field of semidefinite programming (SDP).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial makes use of the following packages:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport Random\nimport SCS\nimport Test","category":"page"},{"location":"tutorials/conic/simple_examples/#Maximum-cut-via-SDP","page":"Simple semidefinite programming examples","title":"Maximum cut via SDP","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The maximum cut problem is a classical example in graph theory, where we seek to partition a graph into two complementary sets, such that the weight of edges between the two sets is maximized. This problem is NP-hard, but it is possible to obtain an approximate solution using the semidefinite programming relaxation:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    textmax  quad   025 LX \n    text    st quad   mathrmdiag(X) = e \n                 qquad X succeq 0","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"where L is the weighted graph Laplacian and e is a vector of ones. For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Goemans, M. X., & Williamson, D. P. (1995). Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. Journal of the ACM (JACM), 42(6), 1115-1145.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"\"\"\"\n    svd_cholesky(X::AbstractMatrix, rtol)\n\nReturn the matrix `U` of the Cholesky decomposition of `X` as `U' * U`.\nNote that we do not use the `LinearAlgebra.cholesky` function because it\nrequires the matrix to be positive definite while `X` may be only\npositive *semi*definite.\n\nWe use the convention `U' * U` instead of `U * U'` to be consistent with\n`LinearAlgebra.cholesky`.\n\"\"\"\nfunction svd_cholesky(X::AbstractMatrix)\n    F = LinearAlgebra.svd(X)\n    # We now have `X ≈ `F.U * D² * F.U'` where:\n    D = LinearAlgebra.Diagonal(sqrt.(F.S))\n    # So `X ≈ U' * U` where `U` is:\n    return (F.U * D)'\nend\n\nfunction solve_max_cut_sdp(weights)\n    N = size(weights, 1)\n    # Calculate the (weighted) Laplacian of the graph: L = D - W.\n    L = LinearAlgebra.diagm(0 => weights * ones(N)) - weights\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, X[1:N, 1:N], PSD)\n    for i in 1:N\n        set_start_value(X[i, i], 1.0)\n    end\n    @objective(model, Max, 0.25 * LinearAlgebra.dot(L, X))\n    @constraint(model, LinearAlgebra.diag(X) .== 1)\n    optimize!(model)\n    V = svd_cholesky(value(X))\n    Random.seed!(N)\n    r = rand(N)\n    r /= LinearAlgebra.norm(r)\n    cut = [LinearAlgebra.dot(r, V[:, i]) > 0 for i in 1:N]\n    S = findall(cut)\n    T = findall(.!cut)\n    println(\"Solution:\")\n    println(\" (S, T) = ({\", join(S, \", \"), \"}, {\", join(T, \", \"), \"})\")\n    return S, T\nend","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 5 --- [2]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1}, {2})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 5; 5 0])","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 5 --- [2]\n |  \\          |\n |    \\        |\n 7      6      1\n |        \\    |\n |          \\  |\n[3] --- 1 --- [4]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1}, {2, 3, 4})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 5 7 6; 5 0 0 1; 7 0 0 1; 6 1 1 0])","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 1 --- [2]\n |             |\n |             |\n 5             9\n |             |\n |             |\n[3] --- 2 --- [4]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1, 4}, {2, 3})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 1 5 0; 1 0 0 9; 5 0 0 2; 0 9 2 0])","category":"page"},{"location":"tutorials/conic/simple_examples/#K-means-clustering-via-SDP","page":"Simple semidefinite programming examples","title":"K-means clustering via SDP","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given a set of points a_1 ldots a_m  in mathbbR^n, allocate them to k clusters.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Peng, J., & Wei, Y. (2007). Approximating k-means-type clustering via semidefinite programming. SIAM Journal on Optimization, 18(1), 186-205.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_k_means_clustering()\n    a = [[2.0, 2.0], [2.5, 2.1], [7.0, 7.0], [2.2, 2.3], [6.8, 7.0], [7.2, 7.5]]\n    m = length(a)\n    num_clusters = 2\n    W = zeros(m, m)\n    for i in 1:m, j in i+1:m\n        W[i, j] = W[j, i] = exp(-LinearAlgebra.norm(a[i] - a[j]) / 1.0)\n    end\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, Z[1:m, 1:m] >= 0, PSD)\n    @objective(model, Min, LinearAlgebra.tr(W * (LinearAlgebra.I - Z)))\n    @constraint(model, [i = 1:m], sum(Z[i, :]) .== 1)\n    @constraint(model, LinearAlgebra.tr(Z) == num_clusters)\n    optimize!(model)\n    Z_val = value.(Z)\n    current_cluster, visited = 0, Set{Int}()\n    for i in 1:m\n        if !(i in visited)\n            current_cluster += 1\n            println(\"Cluster $current_cluster\")\n            for j in i:m\n                if isapprox(Z_val[i, i], Z_val[i, j]; atol = 1e-3)\n                    println(a[j])\n                    push!(visited, j)\n                end\n            end\n        end\n    end\n    return\nend\n\nexample_k_means_clustering()","category":"page"},{"location":"tutorials/conic/simple_examples/#The-correlation-problem","page":"Simple semidefinite programming examples","title":"The correlation problem","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given three random variables A, B, and C, and given bounds on two of the three correlation coefficients:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    -02 leq ρ_AB leq -01 \n    04  leq ρ_BC leq  05","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"our problem is to determine upper and lower bounds on other correlation coefficient ρ_AC.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"We solve an SDP to make use of the following positive semidefinite property of the correlation matrix:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"beginbmatrix\n      1     ρ_AB   ρ_AC \n     ρ_AB   1      ρ_BC  \n     ρ_AC   ρ_BC   1\nendbmatrix succeq 0","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_correlation_problem()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, X[1:3, 1:3], PSD)\n    S = [\"A\", \"B\", \"C\"]\n    ρ = Containers.DenseAxisArray(X, S, S)\n    @constraint(model, [i in S], ρ[i, i] == 1)\n    @constraint(model, -0.2 <= ρ[\"A\", \"B\"] <= -0.1)\n    @constraint(model, 0.4 <= ρ[\"B\", \"C\"] <= 0.5)\n    @objective(model, Max, ρ[\"A\", \"C\"])\n    optimize!(model)\n    println(\"An upper bound for ρ_AC is $(value(ρ[\"A\", \"C\"]))\")\n    @objective(model, Min, ρ[\"A\", \"C\"])\n    optimize!(model)\n    println(\"A lower bound for ρ_AC is $(value(ρ[\"A\", \"C\"]))\")\n    return\nend\n\nexample_correlation_problem()","category":"page"},{"location":"tutorials/conic/simple_examples/#The-minimum-distortion-problem","page":"Simple semidefinite programming examples","title":"The minimum distortion problem","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This example arises from computational geometry, in particular the problem of embedding a general finite metric space into a Euclidean space.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"It is known that the 4-point metric space defined by the star graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"  [1]\n    \\\n     1\n      \\\n      [0] —- 1 -- [2]\n      /\n     1\n    /\n  [3]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"cannot be exactly embedded into a Euclidean space of any dimension, where distances are computed by length of the shortest path between vertices. A distance-preserving embedding would require the three leaf nodes to form an equilateral triangle of side length 2, with the centre node (0) mapped to an equidistant point at distance 1; this is impossible since the triangle inequality in Euclidean space implies all points would need to be simultaneously collinear.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Here we will formulate and solve an SDP to compute the best possible embedding, that is, the embedding f assigning each vertex v to a vector f(v) that minimizes the distortion c such that","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    Da b leq f(a) - f(b) leq c  Da b","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"for all edges (a b) in the graph, where Da b is the distance in the graph metric space.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Any embedding f can be characterized by a Gram matrix Q, which is PSD and such that","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    f(a) - f(b)^2 = Qa a + Qb b - 2 Qa b","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The matrix entry Qab represents the inner product of f(a) with f(b).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"We therefore impose the constraint","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    Da b^2 leq Qa a + Qb b - 2 Qa b leq c^2  Da b^2","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"for all edges (a b) in the graph and minimize c^2, which gives us the SDP formulation below. Since we may choose any point to be the origin, we fix the first vertex at 0.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"J. Matoušek (2002), Lectures on discrete geometry, Springer, pp. 378-379","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"N. Linial (2002), Finite metric spaces–combinatorics, geometry and algorithms, Proceedings of the ICM, Vol. 3, 573-586","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_minimum_distortion()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    D = [\n        0.0 1.0 1.0 1.0\n        1.0 0.0 2.0 2.0\n        1.0 2.0 0.0 2.0\n        1.0 2.0 2.0 0.0\n    ]\n    @variable(model, c² >= 1.0)\n    @variable(model, Q[1:4, 1:4], PSD)\n    for i in 1:4, j in (i+1):4\n        @constraint(model, D[i, j]^2 <= Q[i, i] + Q[j, j] - 2 * Q[i, j])\n        @constraint(model, Q[i, i] + Q[j, j] - 2 * Q[i, j] <= c² * D[i, j]^2)\n    end\n    fix(Q[1, 1], 0)\n    @objective(model, Min, c²)\n    optimize!(model)\n    Test.@test termination_status(model) == OPTIMAL\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 4 / 3 atol = 1e-4\n    # Recover the minimal distorted embedding:\n    X = [zeros(3) sqrt(value.(Q)[2:end, 2:end])]\n    return Plots.plot(\n        X[1, :],\n        X[2, :],\n        X[3, :];\n        seriestype = :mesh3d,\n        connections = ([0, 0, 0, 1], [1, 2, 3, 2], [2, 3, 1, 3]),\n        legend = false,\n        fillalpha = 0.1,\n        lw = 3,\n        ratio = :equal,\n        xlim = (-1.1, 1.1),\n        ylim = (-1.1, 1.1),\n        zlim = (-1.5, 1.0),\n        zticks = -1:1,\n        camera = (60, 30),\n    )\nend\n\nexample_minimum_distortion()","category":"page"},{"location":"tutorials/conic/simple_examples/#Lovász-numbers","page":"Simple semidefinite programming examples","title":"Lovász numbers","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The Lovász number of a graph, also known as Lovász's theta-function, is a number that lies between two important and related numbers that are computationally hard to determine, namely the chromatic and clique numbers of the graph. It is possible however to efficient compute the Lovász number as the optimal value of a semidefinite program.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Consider the pentagon graph:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"     [5]\n    /   \\\n   /     \\\n [1]     [4]\n  |       |\n  |       |\n [2] --- [3]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"with five vertices and edges. Its Lovász number is known to be precisely sqrt5 approx 2236, lying between 2 (the largest clique size) and 3 (the smallest number needed for a vertex coloring).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Let i  j be integers such that 1 leq i  j leq 5. We define A^ij to be the 5 times 5 symmetric matrix with entries (ij) and (ji) equal to 1, with all other entries 0. Let  E be the graph's edge set; in this example, E contains (1,2), (2,3), (3,4), (4,5), (5,1) and their transposes. The Lovász number can be computed from the program","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"beginalign\n    textmax       quad    J  X  \n    text    st  quad A^ij  X = 0 text for all  (ij) notin E   \n                     quad I  X = 1 \n                     quad X succeq 0\nendalign","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"where J is the matrix filled with ones, and I is the identity matrix.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Barvinok, A. (2002). A course in convexity, American Mathematical Society (Vol. 54), pp. 182-185.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Knuth, D. E. (1994), The sandwich theorem, Electronic Journal of Combinatorics, Volume 1, Issue 1, A1.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_theta_problem()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    E = [(1, 2), (2, 3), (3, 4), (4, 5), (5, 1)]\n    @variable(model, X[1:5, 1:5], PSD)\n    for i in 1:5\n        for j in (i+1):5\n            if !((i, j) in E || (j, i) in E)\n                A = zeros(Int, 5, 5)\n                A[i, j] = 1\n                A[j, i] = 1\n                @constraint(model, LinearAlgebra.dot(A, X) == 0)\n            end\n        end\n    end\n    @constraint(model, LinearAlgebra.tr(LinearAlgebra.I * X) == 1)\n    J = ones(Int, 5, 5)\n    @objective(model, Max, LinearAlgebra.dot(J, X))\n    optimize!(model)\n    Test.@test termination_status(model) == OPTIMAL\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ sqrt(5) rtol = 1e-4\n    println(\"The Lovász number is: $(objective_value(model))\")\n    return\nend\n\nexample_theta_problem()","category":"page"},{"location":"tutorials/conic/simple_examples/#Robust-uncertainty-sets","page":"Simple semidefinite programming examples","title":"Robust uncertainty sets","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This example computes the Value at Risk for a data-driven uncertainty set. Closed-form expressions for the optimal value are available. For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Bertsimas, D., Gupta, V., & Kallus, N. (2018). Data-driven robust optimization. Mathematical Programming, 167, 235-292.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_robust_uncertainty_sets()\n    R, d, 𝛿, ɛ = 1, 3, 0.05, 0.05\n    N = ceil((2 + 2 * log(2 / 𝛿))^2) + 1\n    c, μhat, M = randn(d), rand(d), rand(d, d)\n    Σhat = 1 / (d - 1) * (M - ones(d) * μhat')' * (M - ones(d) * μhat')\n    Γ1(𝛿, N) = R / sqrt(N) * (2 + sqrt(2 * log(1 / 𝛿)))\n    Γ2(𝛿, N) = 2 * R^2 / sqrt(N) * (2 + sqrt(2 * log(2 / 𝛿)))\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, Σ[1:d, 1:d], PSD)\n    @variable(model, u[1:d])\n    @variable(model, μ[1:d])\n    @constraint(model, [Γ1(𝛿 / 2, N); μ - μhat] in SecondOrderCone())\n    @constraint(model, [Γ2(𝛿 / 2, N); vec(Σ - Σhat)] in SecondOrderCone())\n    @constraint(model, [((1-ɛ)/ɛ) (u - μ)'; (u-μ) Σ] >= 0, PSDCone())\n    @objective(model, Max, c' * u)\n    optimize!(model)\n    exact =\n        μhat' * c +\n        Γ1(𝛿 / 2, N) * LinearAlgebra.norm(c) +\n        sqrt((1 - ɛ) / ɛ) *\n        sqrt(c' * (Σhat + Γ2(𝛿 / 2, N) * LinearAlgebra.I) * c)\n    Test.@test objective_value(model) ≈ exact atol = 1e-2\n    return\nend\n\nexample_robust_uncertainty_sets()","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/callbacks/#callbacks_manual","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Many mixed-integer (linear, conic, and nonlinear) programming solvers offer the ability to modify the solve process. Examples include changing branching decisions in branch-and-bound, adding custom cutting planes, providing custom heuristics to find feasible solutions, or implementing on-demand separators to add new constraints only when they are violated by the current solution (also known as lazy constraints).","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"While historically this functionality has been limited to solver-specific interfaces, JuMP provides solver-independent support for three types of callbacks:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"lazy constraints\nuser-cuts\nheuristic solutions","category":"page"},{"location":"manual/callbacks/#Available-solvers","page":"Solver-independent Callbacks","title":"Available solvers","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Solver-independent callback support is limited to a few solvers. This includes CPLEX, GLPK, Gurobi, Xpress, and SCIP.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nWhile JuMP provides a solver-independent way of accessing callbacks, you should not assume that you will see identical behavior when running the same code on different solvers. For example, some solvers may ignore user-cuts for various reasons, while other solvers may add every user-cut. Read the underlying solver's callback documentation to understand details specific to each solver.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"tip: Tip\nThis page discusses solver-independent callbacks. However, each solver listed above also provides a solver-dependent callback to provide access to the full range of solver-specific features. Consult the solver's README for an example of how to use the solver-dependent callback. This will require you to understand the C interface of the solver.","category":"page"},{"location":"manual/callbacks/#Things-you-can-and-cannot-do-during-solver-independent-callbacks","page":"Solver-independent Callbacks","title":"Things you can and cannot do during solver-independent callbacks","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"There is a limited range of things you can do during a callback. Only use the functions and macros explicitly stated in this page of the documentation, or in the Callbacks tutorial.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Using any other part of the JuMP API (for example, adding a constraint with @constraint or modifying a variable bound with set_lower_bound) is undefined behavior, and your solver may throw an error, return an incorrect solution, or result in a segfault that aborts Julia.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"In each of the three solver-independent callbacks, there are two things you may query:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"callback_node_status returns an MOI.CallbackNodeStatusCode enum indicating if the current primal solution is integer feasible.\ncallback_value returns the current primal solution of a variable.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"If you need to query any other information, use a solver-dependent callback instead. Each solver supporting a solver-dependent callback has information on how to use it in the README of their GitHub repository.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"If you want to modify the problem in a callback, you must use a lazy constraint.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nYou can only set each callback once. Calling set twice will over-write the earlier callback. In addition, if you use a solver-independent callback, you cannot set a solver-dependent callback.","category":"page"},{"location":"manual/callbacks/#Lazy-constraints","page":"Solver-independent Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Lazy constraints are useful when the full set of constraints is too large to explicitly include in the initial formulation. When a MIP solver reaches a new solution, for example with a heuristic or by solving a problem at a node in the branch-and-bound tree, it will give the user the chance to provide constraints that would make the current solution infeasible. For some more information about lazy constraints, see this blog post by Paul Rubin.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A lazy constraint callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           status = callback_node_status(cb_data, model)\n           if status == MOI.CALLBACK_NODE_STATUS_FRACTIONAL\n               # `callback_value(cb_data, x)` is not integer (to some tolerance).\n               # If, for example, your lazy constraint generator requires an\n               # integer-feasible primal solution, you can add a `return` here.\n               return\n           elseif status == MOI.CALLBACK_NODE_STATUS_INTEGER\n               # `callback_value(cb_data, x)` is integer (to some tolerance).\n           else\n               @assert status == MOI.CALLBACK_NODE_STATUS_UNKNOWN\n               # `callback_value(cb_data, x)` might be fractional or integer.\n           end\n           x_val = callback_value(cb_data, x)\n           if x_val > 2 + 1e-6\n               con = @build_constraint(x <= 2)\n               MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n           end\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.LazyConstraintCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe lazy constraint callback may be called at fractional or integer nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every primal solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nOnly add a lazy constraint if your primal solution violates the constraint. Adding the lazy constraint irrespective of feasibility may result in the solver returning an incorrect solution, or lead to many constraints being added, slowing down the solution process.model = Model(GLPK.Optimizer)\n@variable(model, x <= 10, Int)\n@objective(model, Max, x)\nfunction bad_callback_function(cb_data)\n    # Don't do this!\n    con = @build_constraint(x <= 2)\n    MOI.submit(model, MOI.LazyConstraint(cb_data), con)\nend\nfunction good_callback_function(cb_data)\n    if callback_value(x) > 2\n        con = @build_constraint(x <= 2)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nset_attribute(model, MOI.LazyConstraintCallback(), good_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nDuring the solve, a solver may visit a point that was cut off by a previous lazy constraint, for example, because the earlier lazy constraint was removed during presolve. If this happens, you must re-add the lazy constraint.","category":"page"},{"location":"manual/callbacks/#User-cuts","page":"Solver-independent Callbacks","title":"User cuts","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"User cuts, or simply cuts, provide a way for the user to tighten the LP relaxation using problem-specific knowledge that the solver cannot or is unable to infer from the model. Just like with lazy constraints, when a MIP solver reaches a new node in the branch-and-bound tree, it will give the user the chance to provide cuts to make the current relaxed (fractional) solution infeasible in the hopes of obtaining an integer solution. For more details about the difference between user cuts and lazy constraints see the aforementioned blog post.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A user-cut callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10.5, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           x_val = callback_value(cb_data, x)\n           con = @build_constraint(x <= floor(x_val))\n           MOI.submit(model, MOI.UserCut(cb_data), con)\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.UserCutCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nUser cuts must not change the set of integer feasible solutions. Equivalently, user cuts can only remove fractional solutions. If you add a cut that removes an integer solution (even one that is not optimal), the solver may return an incorrect solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe user-cut callback may be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every fractional primal solution.","category":"page"},{"location":"manual/callbacks/#Heuristic-solutions","page":"Solver-independent Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Integer programming solvers frequently include heuristics that run at the nodes of the branch-and-bound tree. They aim to find integer solutions quicker than plain branch-and-bound would to tighten the bound, allowing us to fathom nodes quicker and to tighten the integrality gap.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Some heuristics take integer solutions and explore their \"local neighborhood\" (for example, flipping binary variables, fix some variables and solve a smaller MILP) and others take fractional solutions and attempt to round them in an intelligent way.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"You may want to add a heuristic of your own if you have some special insight into the problem structure that the solver is not aware of, for example, you can consistently take fractional solutions and intelligently guess integer solutions from them.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A heuristic solution callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10.5, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           x_val = callback_value(cb_data, x)\n           status = MOI.submit(\n               model, MOI.HeuristicSolution(cb_data), [x], [floor(Int, x_val)]\n           )\n           println(\"I submitted a heuristic solution, and the status was: \", status)\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.HeuristicCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"The third argument to submit is a vector of JuMP variables, and the fourth argument is a vector of values corresponding to each variable.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"MOI.submit returns an enum that depends on whether the solver accepted the solution. The possible return codes are:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"MOI.HEURISTIC_SOLUTION_ACCEPTED\nMOI.HEURISTIC_SOLUTION_REJECTED\nMOI.HEURISTIC_SOLUTION_UNKNOWN","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nSome solvers may accept partial solutions. Others require a feasible integer solution for every variable. If in doubt, provide a complete solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe heuristic solution callback may be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every fractional primal solution.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/infeasibility_certificates.md\"","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasibility-certificates","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"When given a conic problem that is infeasible or unbounded, some solvers can produce a certificate of infeasibility. This page explains what a certificate of infeasibility is, and the related conventions that MathOptInterface adopts.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Conic-duality","page":"Infeasibility certificates","title":"Conic duality","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"MathOptInterface uses conic duality to define infeasibility certificates. A full explanation is given in the section Duality, but here is a brief overview.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Minimization-problems","page":"Infeasibility certificates","title":"Minimization problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a minimization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n min_x in mathbbR^n  a_0^top x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and the dual is a maximization problem in standard conic form:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n max_y_1 ldots y_m  -sum_i=1^m b_i^top y_i + b_0\n\n textst  a_0 - sum_i=1^m A_i^top y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Maximization-problems","page":"Infeasibility certificates","title":"Maximization problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a maximization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n max_x in mathbbR^n  a_0^top x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and the dual is a minimization problem in standard conic form:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n min_y_1 ldots y_m  sum_i=1^m b_i^top y_i + b_0\n\n textst  a_0 + sum_i=1^m A_i^top y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/#Unbounded-problems","page":"Infeasibility certificates","title":"Unbounded problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A problem is unbounded if and only if:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"there exists a feasible primal solution\nthe dual is infeasible.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A feasible primal solution—if one exists—can be obtained by setting ObjectiveSense to FEASIBILITY_SENSE before optimizing. Therefore, most solvers stop after they prove the dual is infeasible via a certificate of dual infeasibility, but before they have found a feasible primal solution. This is also the reason that MathOptInterface defines the DUAL_INFEASIBLE status instead of UNBOUNDED.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A certificate of dual infeasibility is an improving ray of the primal problem. That is, there exists some vector d such that for all eta  0:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A_i (x + eta d) + b_i in mathcalC_i  i = 1 ldots m","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and (for minimization problems):","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"a_0^top (x + eta d) + b_0  a_0^top x + b_0","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"for any feasible point x. The latter simplifies to a_0^top d  0. For maximization problems, the inequality is reversed, so that a_0^top d  0.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"If the solver has found a certificate of dual infeasibility:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"TerminationStatus must be DUAL_INFEASIBLE\nPrimalStatus must be INFEASIBILITY_CERTIFICATE\nVariablePrimal must be the corresponding value of d\nConstraintPrimal must be the corresponding value of A_i d\nObjectiveValue must be the value a_0^top d. Note that this is the value of the objective function at d, ignoring the constant b_0.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"note: Note\nThe choice of whether to scale the ray d to have magnitude 1 is left to the solver.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasible-problems","page":"Infeasibility certificates","title":"Infeasible problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A certificate of primal infeasibility is an improving ray of the dual problem. However, because infeasibility is independent of the objective function, we first homogenize the primal problem by removing its objective.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a minimization problem, a dual improving ray is some vector d such that for all eta  0:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n-sum_i=1^m A_i^top (y_i + eta d_i)  = 0 \n(y_i + eta d_i)  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"-sum_i=1^m b_i^top (y_i + eta d_i)  -sum_i=1^m b_i^top y_i","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"for any feasible dual solution y. The latter simplifies to -sum_i=1^m b_i^top d_i  0. For a maximization problem, the inequality is sum_i=1^m b_i^top d_i  0. (Note that these are the same inequality, modulo a - sign.)","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"If the solver has found a certificate of primal infeasibility:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"TerminationStatus must be INFEASIBLE\nDualStatus must be INFEASIBILITY_CERTIFICATE\nConstraintDual must be the corresponding value of d\nDualObjectiveValue must be the value -sum_i=1^m b_i^top d_i for minimization problems and sum_i=1^m b_i^top d_i for maximization problems.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"note: Note\nThe choice of whether to scale the ray d to have magnitude 1 is left to the solver.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasibility-certificates-of-variable-bounds","page":"Infeasibility certificates","title":"Infeasibility certificates of variable bounds","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"Many linear solvers (for example, Gurobi) do not provide explicit access to the primal infeasibility certificate of a variable bound. However, given a set of linear constraints:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\nl_A le A x le u_A \nl_x le x le u_x\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"the primal certificate of the variable bounds can be computed using the primal certificate associated with the affine constraints, d. (Note that d will have one element for each row of the A matrix, and that some or all of the elements in the vectors l_A and u_A may be pm infty. If both l_A and u_A are finite for some row, the corresponding element in `d  must be 0.)","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"Given d, compute bard = d^top A. If the bound is finite, a certificate for the lower variable bound of x_i is maxbard_i 0, and a certificate for the upper variable bound is minbard_i 0.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"EditURL = \"debugging.jl\"","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging","page":"Debugging","title":"Debugging","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Dealing with bugs is an unavoidable part of coding optimization models in any framework, including JuMP. Sources of bugs include not only generic coding errors (method errors, typos, off-by-one issues), but also semantic mistakes in the formulation of an optimization problem and the incorrect use of a solver.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This tutorial explains some common sources of bugs and modeling issues that you might encounter when writing models in JuMP, and it suggests a variety of strategies to deal with them.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nThis tutorial is more advanced than the other \"Getting started\" tutorials. It's in the \"Getting started\" section to give you an early preview of how to debug JuMP models. However, if you are new to JuMP, you may want to briefly skim the tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/debugging/#Getting-help","page":"Debugging","title":"Getting help","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Debugging can be a frustrating part of modeling, particularly if you're new to optimization and programming. If you're stuck, join the community forum to search for answers to commonly asked questions.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Before asking a new question, make sure to read the post Make it easier to help you, which contains a number of tips on how to ask a good question.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Above all else, take time to simplify your code as much as possible. The fewer lines of code you can post that reproduces the same issue, the faster someone can answer your question.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-Julia-code","page":"Debugging","title":"Debugging Julia code","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Read the Debugging chapter in the book ThinkJulia.jl. It has a number of great tips and tricks for debugging Julia code.","category":"page"},{"location":"tutorials/getting_started/debugging/#Solve-failures","page":"Debugging","title":"Solve failures","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When a solver experiences an issue that prevents it from finding an optimal solution (or proving that one does not exist), JuMP may return one of a number of termination_statuses.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"For example, if the solver found a solution, but experienced numerical imprecision, it may return a status such as ALMOST_OPTIMAL or ALMOST_LOCALLY_SOLVED indicating that the problem was solved to a relaxed set of tolerances. Alternatively, the solver may return a problematic status such as NUMERICAL_ERROR, SLOW_PROGRESS, or OTHER_ERROR, indicating that it could not find a solution to the problem.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Most solvers can experience numerical imprecision because they use floating-point arithmetic to perform operations such as addition, subtraction, and multiplication. These operations aren't exact, and small errors can accrue between the theoretical value and the value that the computer computes. For example:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"0.1 * 3 == 0.3","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nRead the Guidelines for numerical issues section of the Gurobi documentation, along with the Debugging numerical problems section of the YALMIP documentation.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of solve failures are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Very large numbers and very small numbers as problem coefficients. Exactly what \"large\" is depends on the solver and the problem, but in general, values above 1e6 or smaller than 1e-6 cause problems.\nNonlinear problems with functions that are not defined in parts of their domain. For example, minimizing log(x) where x >= 0 is undefined when x = 0 (a common starting value).","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of solve failures include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Rescale variables in the problem and their associated coefficients to make the magnitudes of all coefficients in the 1e-4 to 1e4 range. For example, that might mean rescaling a variable from measuring distance in centimeters to kilometers.\nTry a different solver. Some solvers might be more robust than others for a particular problem.\nRead the documentation of your solver, and try settings that encourage numerical robustness.\nSet bounds or add constraints so that all nonlinear functions are defined across all of the feasible region. This particularly applies for functions like 1 / x and log(x) which are not defined for x = 0.","category":"page"},{"location":"tutorials/getting_started/debugging/#Incorrect-results","page":"Debugging","title":"Incorrect results","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Sometimes, you might find that the solver returns an \"optimal\" solution that is incorrect according to the model you are trying to solve (perhaps the solution is suboptimal, or it doesn't satisfy some of the constraints).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Incorrect results can be hard to detect and debug, because the solver gives no hints that there is a problem. Indeed, the termination_status will likely be OPTIMAL and a solution will be available.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-2","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of incorrect results are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A modeling error, so that your JuMP model does not match the formulation you have on paper\nNot accounting for the tolerances that solvers use (for example, if x is binary, a value like x = 1.0000001 may still be considered feasible)\nA bug in JuMP or the solver.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"The probability of the issue being a bug in JuMP or the solver is much smaller than a modeling error. When in doubt, first assume there is a bug in your code before assuming that there is a bug in JuMP.","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-2","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of incorrect results include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Print your JuMP model to see if it matches the formulation you have on paper. Look out for incorrect signs + instead of -, and off-by-one errors such as x[t] instead of x[t-1].\nCheck that you are not using exact comparisons like value(x) == 1.0; always use isapprox(value(x), 1.0; atol = 1e-6) where you manually specify the comparison tolerance.\nTry a different solver. If one solver succeeds where another doesn't this is a sign that the problem is a numerical issue or a bug in the solver.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-an-infeasible-model","page":"Debugging","title":"Debugging an infeasible model","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A model is infeasible if there is no primal solution that satisfies all of the constraints. In general, an infeasible model means one of two things:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Your problem really has no feasible solution\nThere is a mistake in your model.","category":"page"},{"location":"tutorials/getting_started/debugging/#Example","page":"Debugging","title":"Example","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A simple example of an infeasible model is:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n@objective(model, Max, 2x + 1)\n@constraint(model, con, 2x - 1 <= -2)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"because the bound says that x >= 0, but we can rewrite the constraint to be x <= -1/2. When the problem is infeasible, JuMP may return one of a number of statuses. The most common is INFEASIBLE:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Depending on the solver, you may also receive INFEASIBLE_OR_UNBOUNDED or LOCALLY_INFEASIBLE.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A termination status of INFEASIBLE_OR_UNBOUNDED means that the solver could not prove if the solver was infeasible or unbounded, only that the model does not have a finite feasible optimal solution.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Nonlinear optimizers such as Ipopt may return the status LOCALLY_INFEASIBLE. This does not mean that the solver proved no feasible solution exists, only that it could not find one. If you know a primal feasible point, try providing it as a starting point using set_start_value and re-optimize.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-3","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of infeasibility are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Incorrect units, for example, using a lower bound of megawatts and an upper bound of kilowatts\nUsing + instead of - in a constraint\nOff-by-one and related errors, for example, using x[t] instead of x[t-1] in part of a constraint\nOtherwise invalid mathematical formulations","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-3","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of infeasibility include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Iteratively comment out a constraint (or block of constraints) and re-solve the problem. When you find a constraint that makes the problem infeasible when added, check the constraint carefully for errors.\nIf the problem is still infeasible with all constraints commented out, check all variable bounds. Do they use the right data?\nIf you have a known feasible solution, use primal_feasibility_report to evaluate the constraints and check for violations. You'll probably find that you have a typo in one of the constraints.\nTry a different solver. Sometimes, solvers have bugs, and they can incorrectly report a problem as infeasible when it isn't. If you find such a case where one solver reports the problem is infeasible and another can find an optimal solution, please report it by opening an issue on the GitHub repository of the solver that reports infeasibility.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nSome solvers also have specialized support for debugging sources of infeasibility via an irreducible infeasible subsystem. To see if your solver has support, try calling compute_conflict!:julia> compute_conflict!(model)\nERROR: ArgumentError: The optimizer HiGHS.Optimizer does not support `compute_conflict!`In this case, HiGHS does not support computing conflicts, but other solvers such as Gurobi and CPLEX do. If the solver does support computing conflicts, read Conflicts for more details.","category":"page"},{"location":"tutorials/getting_started/debugging/#Penalty-relaxation","page":"Debugging","title":"Penalty relaxation","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Another strategy to debug sources of infeasibility is the relax_with_penalty! function.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"The penalty relaxation modifies constraints of the form f(x) in S into f(x) + y - z in S, where y z ge 0, and then it introduces a penalty term into the objective of a times (y + z) (if minimizing, else -a), where a is a penalty.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"map = relax_with_penalty!(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Here map is a dictionary which maps constraint indices to an affine expression representing (y + z).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"If we optimize the relaxed model, this time we get a feasible solution:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Iterate over the contents of map to see which constraints are violated:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"for (con, penalty) in map\n    violation = value(penalty)\n    if violation > 0\n        println(\"Constraint `$(name(con))` is violated by $violation\")\n    end\nend","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Once you find a violated constraint in the relaxed problem, take a look to see if there is a typo or other common mistake in that particular constraint.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Consult the docstring relax_with_penalty! for information on how to modify the penalty cost term a, either for every constraint in the model or a particular subset of the constraints.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When using relax_with_penalty!, you should be aware that:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Variable bounds and integrality restrictions are not relaxed. If the problem is still infeasible after calling relax_with_penalty!, check the variable bounds.\nYou cannot undo the penalty relaxation. If you need an unmodified model, rebuild the problem, or call copy_model before calling relax_with_penalty!.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-an-unbounded-model","page":"Debugging","title":"Debugging an unbounded model","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A model is unbounded if there is no limit on how good the objective value can get. Most often, an unbounded model means that you have an error in your modeling, because all physical systems have limits. (You cannot make an infinite amount of profit.)","category":"page"},{"location":"tutorials/getting_started/debugging/#Example-2","page":"Debugging","title":"Example","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A simple example of an unbounded model is:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n@objective(model, Max, 2x + 1)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"because we can increase x without limit, and the objective value 2x + 1 gets better as x increases.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When the problem is unbounded, JuMP may return one of a number of statuses. The most common is DUAL_INFEASIBLE:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Depending on the solver, you may also receive INFEASIBLE_OR_UNBOUNDED or an error code like NORM_LIMIT.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-4","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of unboundedness are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Using Max instead of Min\nOmitting variable bounds, such as 0 <= x <= 1\nUsing + instead of - in a term of the objective function.","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-4","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of unboundedness include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Double check whether you intended Min or Max in the @objective line.\nPrint the objective function with print(objective_function(model)) and verify that the value and sign of each coefficient is as you expect.\nAdd large bounds to all variables that are free or have one-sided bounds, then re-solve the problem. Because all variables are now bounded, the problem will have a finite optimal solution. Look at the value of each variable in the optimal solution to see if it is at one of the new bounds. If it is, you either need to specify a better bound for that variable, or there might be a mistake in the objective function associated with that variable (for example, a + instead of a -).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"If there are too many variables to add bounds to, or there are too many terms to examine by hand, another strategy is to create a new variable with a large upper bound (if maximizing, lower bound if minimizing) and a constraint that the variable must be less-than or equal to the expression of the objective function. For example:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n# @objective(model, Max, 2x + 1)\n@variable(model, objective <= 10_000)\n@constraint(model, objective <= 2x + 1)\n@objective(model, Max, objective)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This new model has a finite optimal solution, so we can solve it and then look for variables with large positive or negative values in the optimal solution.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\nfor var in all_variables(model)\n    if var == objective\n        continue\n    end\n    if abs(value(var)) > 1e3\n        println(\"Variable `$(name(var))` may be unbounded\")\n    end\nend","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"EditURL = \"https://github.com/jump-dev/ECOS.jl/blob/v1.1.1/README.md\"","category":"page"},{"location":"packages/ECOS/#ECOS.jl","page":"jump-dev/ECOS.jl","title":"ECOS.jl","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"ECOS.jl is a wrapper for the ECOS solver.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/ECOS/#Affiliation","page":"jump-dev/ECOS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"This wrapper is maintained by the JuMP community and is not a product of Embotech AG.","category":"page"},{"location":"packages/ECOS/#License","page":"jump-dev/ECOS.jl","title":"License","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"ECOS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The underlying solver, embotech/ecos, is licensed under the GPL v3 license.","category":"page"},{"location":"packages/ECOS/#Installation","page":"jump-dev/ECOS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"Install ECOS.jl using Pkg.add:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"import Pkg\nPkg.add(\"ECOS\")","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"In addition to installing the ECOS.jl package, this will also download and install the ECOS binaries. You do not need to install ECOS separately.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/ECOS/#Use-with-JuMP","page":"jump-dev/ECOS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"To use ECOS with JuMP, use ECOS.Optimizer:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"using JuMP, ECOS\nmodel = Model(ECOS.Optimizer)\nset_attribute(model, \"maxit\", 100)","category":"page"},{"location":"packages/ECOS/#MathOptInterface-API","page":"jump-dev/ECOS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The ECOS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.VectorAffineFunction{Float64} in MOI.Nonnegatives\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorAffineFunction{Float64} in MOI.Zeros","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/ECOS/#Options","page":"jump-dev/ECOS.jl","title":"Options","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The following options are supported:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"Parameter Explanation\ngamma scaling the final step length\ndelta regularization parameter\neps regularization threshold\nfeastol primal/dual infeasibility tolerance\nabstol absolute tolerance on duality gap\nreltol relative tolerance on duality gap\nfeastol_inacc primal/dual infeasibility relaxed tolerance\nabstol_inacc absolute relaxed tolerance on duality gap\nreltol_inacc relative relaxed tolerance on duality gap\nnitref number of iterative refinement steps\nmaxit maximum number of iterations\nverbose verbosity bool for PRINTLEVEL < 3","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Nonlinear/overview.md\"","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#nonlinear_developers","page":"Overview","title":"Nonlinear","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nThe Nonlinear submodule is experimental. Until this message is removed, breaking changes may be introduced in any minor or patch release of MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The Nonlinear submodule contains data structures and functions for working with a nonlinear optimization problem in the form of an expression graph. This page explains the API and describes the rationale behind its design.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Standard-form","page":"Overview","title":"Standard form","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear programs (NLPs) are a class of optimization problems in which some of the constraints or the objective function are nonlinear:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"beginalign\n    min_x in mathbbR^n  f_0(x) \n    textst  l_j le f_j(x) le u_j  j = 1 ldots m\nendalign","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There may be additional constraints, as well as things like variable bounds and integrality restrictions, but we do not consider them here because they are best dealt with by other components of MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#API-overview","page":"Overview","title":"API overview","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The core element of the Nonlinear submodule is Nonlinear.Model:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> const Nonlinear = MOI.Nonlinear;\n\njulia> model = Nonlinear.Model()\nA Nonlinear.Model with:\n 0 objectives\n 0 parameters\n 0 expressions\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model is a mutable struct that stores all of the nonlinear information added to the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Decision-variables","page":"Overview","title":"Decision variables","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Decision variables are represented by VariableIndexes. The user is responsible for creating these using MOI.VariableIndex(i), where i is the column associated with the variable.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Expressions","page":"Overview","title":"Expressions","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The input data structure is a Julia Expr. The input expressions can incorporate VariableIndexes, but these must be interpolated into the expression with $:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> input = :(1 + sin($x)^2)\n:(1 + sin(MathOptInterface.VariableIndex(1)) ^ 2)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There are a number of restrictions on the input Expr:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"It cannot contain macros\nIt cannot contain broadcasting\nIt cannot contain splatting (except in limited situations)\nIt cannot contain linear algebra, such as matrix-vector products\nIt cannot contain generator expressions, including sum(i for i in S)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Given an input expression, add an expression using Nonlinear.add_expression:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Nonlinear.add_expression(model, input)\nMathOptInterface.Nonlinear.ExpressionIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, expr, is a Nonlinear.ExpressionIndex that can then be interpolated into other input expressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Looking  again at model, we see:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model\nA Nonlinear.Model with:\n 0 objectives\n 0 parameters\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Parameters","page":"Overview","title":"Parameters","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In addition to constant literals like 1 or 1.23, you can create parameters. Parameters are placeholders whose values can change before passing the expression to the solver. Create a parameter using Nonlinear.add_parameter, which accepts a default value:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> p = Nonlinear.add_parameter(model, 1.23)\nMathOptInterface.Nonlinear.ParameterIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, p, is a Nonlinear.ParameterIndex that can then be interpolated into other input expressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Looking  again at model, we see:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model\nA Nonlinear.Model with:\n 0 objectives\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Update a parameter as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model[p]\n1.23\n\njulia> model[p] = 4.56\n4.56\n\njulia> model[p]\n4.56","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Objectives","page":"Overview","title":"Objectives","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Set a nonlinear objective using Nonlinear.set_objective:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, :($p + $expr + $x))\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Clear a nonlinear objective by passing nothing:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, nothing)\n\njulia> model\nA Nonlinear.Model with:\n 0 objectives\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"But we'll re-add the objective for later:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, :($p + $expr + $x));","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Constraints","page":"Overview","title":"Constraints","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Add a constraint using Nonlinear.add_constraint:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> c = Nonlinear.add_constraint(model, :(1 + sqrt($x)), MOI.LessThan(2.0))\nMathOptInterface.Nonlinear.ConstraintIndex(1)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 1 constraint","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, c, is a Nonlinear.ConstraintIndex that is a unique identifier for the constraint. Interval constraints are also supported:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> c2 = Nonlinear.add_constraint(model, :(1 + sqrt($x)), MOI.Interval(-1.0, 2.0))\nMathOptInterface.Nonlinear.ConstraintIndex(2)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 2 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Delete a constraint using Nonlinear.delete:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.delete(model, c2)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 1 constraint","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#User-defined-operators","page":"Overview","title":"User-defined operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear supports a wide range of univariate and multivariate operators. However, you can also define your own operators by registering them.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Univariate-operators","page":"Overview","title":"Univariate operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Register a univariate user-defined operator using Nonlinear.register_operator:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f(x) = 1 + sin(x)^2\nf (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f, 1, f)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now, you can use :my_f in expressions:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> new_expr = Nonlinear.add_expression(model, :(my_f($x + 1)))\nMathOptInterface.Nonlinear.ExpressionIndex(2)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear will compute first- and second-derivatives of the registered operator using ForwardDiff.jl. Override this by passing functions which compute the respective derivative:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f′(x) = 2 * sin(x) * cos(x)\nf′ (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f2, 1, f, f′)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"or","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f′′(x) = 2 * (cos(x)^2 - sin(x)^2)\nf′′ (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f3, 1, f, f′, f′′)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Multivariate-operators","page":"Overview","title":"Multivariate operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Register a multivariate user-defined operator using Nonlinear.register_operator:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> g(x...) = x[1]^2 + x[1] * x[2] + x[2]^2\ng (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_g, 2, g)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now, you can use :my_g in expressions:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> new_expr = Nonlinear.add_expression(model, :(my_g($x + 1, $x)))\nMathOptInterface.Nonlinear.ExpressionIndex(3)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear will compute the gradient of the registered operator using ForwardDiff.jl. (Hessian information is not supported.) Override this by passing a function to compute the gradient:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> function ∇g(ret, x...)\n           ret[1] = 2 * x[1] + x[2]\n           ret[2] = x[1] + 2 * x[2]\n           return\n       end\n∇g (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_g2, 2, g, ∇g)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_MOI_interface","page":"Overview","title":"MathOptInterface","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"MathOptInterface communicates the nonlinear portion of an optimization problem to solvers using concrete subtypes of AbstractNLPEvaluator, which implement the Nonlinear programming API.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Create an AbstractNLPEvaluator from Nonlinear.Model using Nonlinear.Evaluator.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Evaluator requires an Nonlinear.AbstractAutomaticDifferentiation backend and an ordered list of the variables that are included in the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There following backends are available to choose from within MOI, although other packages may add more options by sub-typing Nonlinear.AbstractAutomaticDifferentiation:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.ExprGraphOnly\nNonlinear.SparseReverseMode.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> evaluator = Nonlinear.Evaluator(model, Nonlinear.ExprGraphOnly(), [x])\nNonlinear.Evaluator with available features:\n  * :ExprGraph","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The functions of the Nonlinear programming API implemented by Nonlinear.Evaluator depends upon the chosen Nonlinear.AbstractAutomaticDifferentiation backend.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The :ExprGraph feature means we can call objective_expr and constraint_expr to retrieve the expression graph of the problem. However, we cannot call gradient terms such as eval_objective_gradient because Nonlinear.ExprGraphOnly does not have the capability to differentiate a nonlinear expression.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"If, instead, we pass Nonlinear.SparseReverseMode, then we get access to :Grad, the gradient of the objective function, :Jac, the Jacobian matrix of the constraints, :JacVec, the ability to compute Jacobian-vector products, and :ExprGraph.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> evaluator = Nonlinear.Evaluator(\n           model,\n           Nonlinear.SparseReverseMode(),\n           [x],\n       )\nNonlinear.Evaluator with available features:\n  * :Grad\n  * :Jac\n  * :JacVec\n  * :ExprGraph","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"However, before using the evaluator, we need to call initialize:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> MOI.initialize(evaluator, [:Grad, :Jac, :JacVec, :ExprGraph])","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now we can call methods like eval_objective:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = [1.0]\n1-element Vector{Float64}:\n 1.0\n\njulia> MOI.eval_objective(evaluator, x)\n7.268073418273571","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"and eval_objective_gradient:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> grad = [0.0]\n1-element Vector{Float64}:\n 0.0\n\njulia> MOI.eval_objective_gradient(evaluator, grad, x)\n\njulia> grad\n1-element Vector{Float64}:\n 1.909297426825682","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Instead of passing Nonlinear.Evaluator directly to solvers, solvers query the NLPBlock attribute, which returns an NLPBlockData. This object wraps an Nonlinear.Evaluator and includes other information such as constraint bounds and whether the evaluator has a nonlinear objective. Create and set NLPBlockData as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> block = MOI.NLPBlockData(evaluator);\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}());\n\njulia> MOI.set(model, MOI.NLPBlock(), block);","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nOnly call NLPBlockData once you have finished modifying the problem in model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Putting everything together, you can create a nonlinear optimization problem in MathOptInterface as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"import MathOptInterface as MOI\n\nfunction build_model(\n    model::MOI.ModelLike;\n    backend::MOI.Nonlinear.AbstractAutomaticDifferentiation,\n)\n    x = MOI.add_variable(model)\n    y = MOI.add_variable(model)\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    nl_model = MOI.Nonlinear.Model()\n    MOI.Nonlinear.set_objective(nl_model, :($x^2 + $y^2))\n    evaluator = MOI.Nonlinear.Evaluator(nl_model, backend, [x, y])\n    MOI.set(model, MOI.NLPBlock(), MOI.NLPBlockData(evaluator))\n    return\nend\n\n# Replace `model` and `backend` with your optimizer and backend of choice.\nmodel = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nbuild_model(model; backend = MOI.Nonlinear.SparseReverseMode())","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Expression-graph-representation","page":"Overview","title":"Expression-graph representation","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model stores nonlinear expressions in Nonlinear.Expressions. This section explains the design of the expression graph data structure in Nonlinear.Expression.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Given a nonlinear function like f(x) = sin(x)^2 + x, a conceptual aid for thinking about the graph representation of the expression is to convert it into Polish prefix notation:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"f(x, y) = (+ (^ (sin x) 2) x)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This format identifies each operator (function), as well as a list of arguments. Operators can be univariate, like sin, or multivariate, like +.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"A common way of representing Polish prefix notation in code is as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.VariableIndex(1);\n\njulia> struct ExprNode\n           op::Symbol\n           children::Vector{Union{ExprNode,Float64,MOI.VariableIndex}}\n       end\n\njulia> expr = ExprNode(:+, [ExprNode(:^, [ExprNode(:sin, [x]), 2.0]), x]);","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This data structure follows our Polish prefix notation very closely, and we can easily identify the arguments to an operator. However, it has a significant draw-back: each node in the graph requires a Vector, which is heap-allocated and tracked by Julia's garbage collector (GC). For large models, we can expect to have millions of nodes in the expression graph, so this overhead quickly becomes prohibitive for computation.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"An alternative is to record the expression as a linear tape:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Any[:+, 2, :^, 2, :sin, 1, x, 2.0, x]\n9-element Vector{Any}:\n  :+\n 2\n  :^\n 2\n  :sin\n 1\n  MOI.VariableIndex(1)\n 2.0\n  MOI.VariableIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The Int after each operator Symbol specifies the number of arguments.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This data-structure is a single vector, which resolves our problem with the GC, but each element is the abstract type, Any, and so any operations on it will lead to slower dynamic dispatch. It's also hard to identify the children of each operation without reading the entire tape.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"To summarize, representing expression graphs in Julia has the following challenges:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nodes in the expression graph should not contain a heap-allocated object\nAll data-structures should be concretely typed\nIt should be easy to identify the children of a node","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Sketch-of-the-design-in-Nonlinear","page":"Overview","title":"Sketch of the design in Nonlinear","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear overcomes these problems by decomposing the data structure into a number of different concrete-typed vectors.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"First, we create vectors of the supported uni- and multivariate operators.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> const UNIVARIATE_OPERATORS = [:sin];\n\njulia> const MULTIVARIATE_OPERATORS = [:+, :^];","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, there are many more supported operations than the ones listed here.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Second, we create an enum to represent the different types of nodes present in the expression graph:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> @enum(\n           NodeType,\n           NODE_CALL_MULTIVARIATE,\n           NODE_CALL_UNIVARIATE,\n           NODE_VARIABLE,\n           NODE_VALUE,\n       )","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, there are node types other than the ones listed here.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Third, we create two concretely typed structs as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> struct Node\n           type::NodeType\n           parent::Int\n           index::Int\n       end\n\njulia> struct Expression\n           nodes::Vector{Node}\n           values::Vector{Float64}\n       end","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"For each node node in the .nodes field, if node.type is:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"NODE_CALL_MULTIVARIATE, we look up MULTIVARIATE_OPERATORS[node.index] to retrieve the operator\nNODE_CALL_UNIVARIATE, we look up UNIVARIATE_OPERATORS[node.index] to retrieve the operator\nNODE_VARIABLE, we create MOI.VariableIndex(node.index)\nNODE_VALUE, we look up values[node.index]","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The .parent field of each node is the integer index of the parent node in .nodes. For the first node, the parent is -1 by convention.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Therefore, we can represent our function as:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Expression(\n           [\n               Node(NODE_CALL_MULTIVARIATE, -1, 1),\n               Node(NODE_CALL_MULTIVARIATE, 1, 2),\n               Node(NODE_CALL_UNIVARIATE, 2, 1),\n               Node(NODE_VARIABLE, 3, 1),\n               Node(NODE_VALUE, 2, 1),\n               Node(NODE_VARIABLE, 1, 1),\n           ],\n           [2.0],\n       );","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This is less readable than the other options, but does this data structure meet our design goals?","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Instead of a heap-allocated object for each node, we only have two Vectors for each expression, nodes and values, as well as two constant vectors for the OPERATORS. In addition, all fields are concretely typed, and there are no Union or Any types.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"For our third goal, it is not easy to identify the children of a node, but it is easy to identify the parent of any node. Therefore, we can use Nonlinear.adjacency_matrix to compute a sparse matrix that maps parents to their children.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The tape is also ordered topologically, so that a reverse pass of the nodes evaluates all children nodes before their parent.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#The-design-in-practice","page":"Overview","title":"The design in practice","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, Node and Expression are exactly Nonlinear.Node and Nonlinear.Expression. However, Nonlinear.NodeType has more fields to account for comparison operators such as :>= and :<=, logic operators such as :&& and :||, nonlinear parameters, and nested subexpressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Moreover, instead of storing the operators as global constants, they are stored in Nonlinear.OperatorRegistry, and it also stores a vector of logic operators and a vector of comparison operators. In addition to Nonlinear.DEFAULT_UNIVARIATE_OPERATORS and Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS, you can register user-defined functions using Nonlinear.register_operator.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model is a struct that stores the Nonlinear.OperatorRegistry, as well as a list of parameters and subexpressions in the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#ReverseAD","page":"Overview","title":"ReverseAD","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.ReverseAD is a submodule for computing derivatives of a nonlinear optimization problem using sparse reverse-mode automatic differentiation (AD).","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This section does not attempt to explain how sparse reverse-mode AD works, but instead explains why MOI contains its own implementation, and highlights notable differences from similar packages.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nDon't use the API in ReverseAD to compute derivatives. Instead, create a Nonlinear.Evaluator object with Nonlinear.SparseReverseMode as the backend, and then query the MOI API methods.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Design-goals","page":"Overview","title":"Design goals","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The JuliaDiff organization maintains a list of packages for doing AD in Julia. At last count, there were at least ten packages——not including ReverseAD——for reverse-mode AD in Julia. ReverseAD exists because it has a different set of design goals.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Goal: handle scale and sparsity. The types of nonlinear optimization problems that MOI represents can be large scale (10^5 or more functions across 10^5 or more variables) with very sparse derivatives. The ability to compute a sparse Hessian matrix is essential. To the best of our knowledge, ReverseAD is the only reverse-mode AD system in Julia that handles sparsity by default.\nGoal: limit the scope to improve robustness. Most other AD packages accept arbitrary Julia functions as input and then trace an expression graph using operator overloading. This means they must deal (or detect and ignore) with control flow, I/O, and other vagaries of Julia. In contrast, ReverseAD only accepts functions in the form of Nonlinear.Expression, which greatly limits the range of syntax that it must deal with. By reducing the scope of what we accept as input to functions relevant for mathematical optimization, we can provide a simpler implementation with various performance optimizations.\nGoal: provide outputs which match what solvers expect. Other AD packages focus on differentiating individual Julia functions. In contrast, ReverseAD has a very specific use-case: to generate outputs needed by the MOI nonlinear API. This means it needs to efficiently compute sparse Hessians, and it needs subexpression handling to avoid recomputing subexpressions that are shared between functions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#History","page":"Overview","title":"History","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"ReverseAD started life as ReverseDiffSparse.jl, development of which began in early 2014(!). This was well before the other AD packages started development. Because we had a well-tested, working AD in JuMP, there was less motivation to contribute to and explore other AD packages. The lack of historical interaction also meant that other packages were not optimized for the types of problems that JuMP is built for (that is, large-scale sparse problems). When we first created MathOptInterface, we kept the AD in JuMP to simplify the transition, and post-poned the development of a first-class nonlinear interface in MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Prior to the introduction of Nonlinear, JuMP's nonlinear implementation was a confusing mix of functions and types spread across the code base and in the private _Derivatives submodule. This made it hard to swap the AD system for another. The main motivation for refactoring JuMP to create the Nonlinear submodule in MathOptInterface was to abstract the interface between JuMP and the AD system, allowing us to swap-in and test new AD systems in the future.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"EditURL = \"https://github.com/ds4dm/Tulip.jl/blob/069cdda082dfde31ef7ee935c96e833a3b96b33c/README.md\"","category":"page"},{"location":"packages/Tulip/#Tulip","page":"ds4dm/Tulip.jl","title":"Tulip","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"(Image: DOI) (Image: ) (Image: )","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip is an open-source interior-point solver for linear optimization, written in pure Julia. It implements the homogeneous primal-dual interior-point algorithm with multiple centrality corrections, and therefore handles unbounded and infeasible problems. Tulip’s main feature is that its algorithmic framework is disentangled from linear algebra implementations. This allows to seamlessly integrate specialized routines for structured problems.","category":"page"},{"location":"packages/Tulip/#License","page":"ds4dm/Tulip.jl","title":"License","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Tulip/#Installation","page":"ds4dm/Tulip.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Install Tulip using the Julia package manager:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"import Pkg\nPkg.add(\"Tulip\")","category":"page"},{"location":"packages/Tulip/#Usage","page":"ds4dm/Tulip.jl","title":"Usage","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The recommended way of using Tulip is through JuMP or MathOptInterface (MOI).","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The low-level interface is still under development and is likely change in the future. The MOI interface is more stable.","category":"page"},{"location":"packages/Tulip/#Using-with-JuMP","page":"ds4dm/Tulip.jl","title":"Using with JuMP","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip follows the syntax convention PackageName.Optimizer:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using JuMP\nimport Tulip\nmodel = Model(Tulip.Optimizer)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Linear objectives, linear constraints and lower/upper bounds on variables are supported.","category":"page"},{"location":"packages/Tulip/#Using-with-MOI","page":"ds4dm/Tulip.jl","title":"Using with MOI","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The type Tulip.Optimizer is parametrized by the model's arithmetic, for example, Float64 or BigFloat. This allows to solve problem in higher numerical precision. See the documentation for more details.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"import MathOptInterface as MOI\nimport Tulip\nmodel = Tulip.Optimizer{Float64}()   # Create a model in Float64 precision\nmodel = Tulip.Optimizer()            # Defaults to the above call\nmodel = Tulip.Optimizer{BigFloat}()  # Create a model in BigFloat precision","category":"page"},{"location":"packages/Tulip/#Solver-parameters","page":"ds4dm/Tulip.jl","title":"Solver parameters","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"See the documentation for a full list of parameters.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in JuMP, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using JuMP, Tulip\nmodel = Model(Tulip.Optimizer)\nset_attribute(model, \"IPM_IterationsLimit\", 200)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in MathOptInterface, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using Tulip\nimport MathOptInterface as MOI\nmodel = Tulip.Optimizer{Float64}()\nMOI.set(model, MOI.RawOptimizerAttribute(\"IPM_IterationsLimit\"), 200)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in the Tulip API, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using Tulip\nmodel = Tulip.Model{Float64}()\nTulip.set_parameter(model, \"IPM_IterationsLimit\", 200)","category":"page"},{"location":"packages/Tulip/#Command-line-executable","page":"ds4dm/Tulip.jl","title":"Command-line executable","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"See app building instructions.","category":"page"},{"location":"packages/Tulip/#Citing-Tulip.jl","page":"ds4dm/Tulip.jl","title":"Citing Tulip.jl","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"If you use Tulip in your work, we kindly ask that you cite the following reference (preprint available here).","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"@Article{Tulip.jl,\n  author   = {Tanneau, Mathieu and Anjos, Miguel F. and Lodi, Andrea},\n  journal  = {Mathematical Programming Computation},\n  title    = {Design and implementation of a modular interior-point solver for linear optimization},\n  year     = {2021},\n  issn     = {1867-2957},\n  month    = feb,\n  doi      = {10.1007/s12532-020-00200-8},\n  language = {en},\n  url      = {https://doi.org/10.1007/s12532-020-00200-8},\n  urldate  = {2021-03-07},\n}","category":"page"},{"location":"tutorials/algorithms/parallelism/#Parallelism","page":"Parallelism","title":"Parallelism","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The purpose of this tutorial is to give a brief overview of parallelism in Julia as it pertains to JuMP, and to explain some of the things to be aware of when writing parallel algorithms involving JuMP models.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Multi-threading-and-distributed-computing","page":"Parallelism","title":"Multi-threading and distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"There are two main types of parallelism in Julia:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Multi-threading\nDistributed computing","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"In multi-threading, multiple tasks are run in a single Julia process and share the same memory. In distributed computing, tasks are run in multiple Julia processes with independent memory spaces. This can include processes across multiple physical machines, such as in a high-performance computing cluster.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Choosing and understanding the type of parallelism you are using is important because the code you write for each type is different, and there are different limitations and benefits to each approach. However, the best choice is highly problem dependent, so you may want to experiment with both approaches to determine what works for your situation.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Multi-threading","page":"Parallelism","title":"Multi-threading","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use multi-threading with Julia, you must either start Julia with the command line flag --threads=N, or you must set the JULIA_NUM_THREADS environment variable before launching Julia. For this documentation, we set the environment variable to:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> ENV[\"JULIA_NUM_THREADS\"]\n\"4\"","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"You can check how many threads are available using:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Threads.nthreads()\n4","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The easiest way to use multi-threading in Julia is by placing the Threads.@threads macro in front of a for-loop:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> @time begin\n           ids = Int[]\n           my_lock = Threads.ReentrantLock()\n           Threads.@threads for i in 1:Threads.nthreads()\n               global ids, my_lock\n               Threads.lock(my_lock) do\n                   push!(ids, Threads.threadid())\n               end\n               sleep(1.0)\n           end\n       end\n  1.037087 seconds (31.32 k allocations: 1.836 MiB, 2.02% compilation time)","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"This for-loop sleeps for 1 second on each iteration. Thus, if it had executed sequentially, it should have taken the same number of seconds as there are threads available. Instead, it took only 1 second, showing that the iterations were executed simultaneously. We can verify this by checking the Threads.threadid() of the thread that executed each iteration:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> ids\n4-element Vector{Int64}:\n 2\n 4\n 1\n 3","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nWhen working with threads, you need to avoid race conditions, in which two threads attempt to write to the same variable at the same time. In the above example we avoided a race condition by using ReentrantLock. See the Multi-threading section of the Julia documentation for more details.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Distributed-computing","page":"Parallelism","title":"Distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use distributed computing with Julia, use the Distributed package:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> import Distributed","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Like multi-threading, we need to tell Julia how many processes to add. We can do this either by starting Julia with the -p N command line argument, or by using Distributed.addprocs:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> import Pkg\n\njulia> project = Pkg.project();\n\njulia> workers = Distributed.addprocs(4; exeflags = \"--project=$(project.path)\")\n4-element Vector{Int64}:\n 2\n 3\n 4\n 5","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nNot loading the parent environment with --project is a common mistake.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The added processes are \"worker\" processes that we can use to do computation with. They are orchestrated by the process with the id 1. You can check what process the code is currently running on using Distributed.myid()","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.myid()\n1","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"As a general rule, to get maximum performance you should add as many processes as you have logical cores available.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unlike the for-loop approach of multi-threading, distributed computing extends the Julia map function to a \"parallel-map\" function Distributed.pmap. For each element in the list of arguments to map over, Julia will copy the element to an idle worker process and evaluate the function, passing the element as an input argument.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> function hard_work(i::Int)\n           sleep(1.0)\n           return Distributed.myid()\n       end\nhard_work (generic function with 1 method)\n\njulia> Distributed.pmap(hard_work, 1:4)\nERROR: On worker 2:\nUndefVarError: #hard_work not defined\nStacktrace:\n[...]","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unfortunately, if you try this code directly, you will get an error message that says On worker 2: UndefVarError: hard_work not defined. The error is thrown because, although process 1 knows what the hard_work function is, the worker processes do not.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To fix the error, we need to use Distributed.@everywhere, which evaluates the code on every process:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.@everywhere begin\n           function hard_work(i::Int)\n               sleep(1.0)\n               return Distributed.myid()\n           end\n       end","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Now if we run pmap, we see that it took only 1 second instead of 4, and that it executed on each of the worker processes:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> @time ids = Distributed.pmap(hard_work, 1:4)\n  1.202006 seconds (216.39 k allocations: 13.301 MiB, 4.07% compilation time)\n4-element Vector{Int64}:\n 2\n 3\n 5\n 4","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"tip: Tip\nFor more information, read the Julia documentation Distributed Computing.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Using-parallelism-the-wrong-way","page":"Parallelism","title":"Using parallelism the wrong way","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/#When-building-a-JuMP-model","page":"Parallelism","title":"When building a JuMP model","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"For large problems, building the model in JuMP can be a bottleneck, and you may consider trying to write code that builds the model in parallel, for example, by wrapping a for-loop that adds constraints with Threads.@threads.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unfortunately, you cannot build a JuMP model in parallel, and attempting to do so may error or produce incorrect results.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"In most cases, we find that the reason for the bottleneck is not JuMP, but in how you are constructing the problem data, and that with changes, it is possible to build a model in a way that is not the bottleneck in the solution process.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"tip: Tip\nLooking for help to make your code run faster? Ask for help on the community forum. Make sure to include a reproducible example of your code.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-a-single-JuMP-model","page":"Parallelism","title":"With a single JuMP model","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"A common approach people try is to use parallelism with a single JuMP model. For example, to optimize a model over multiple right-hand side vectors, you may try:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"using JuMP\nimport HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x)\n@objective(model, Min, x)\nsolutions = Pair{Int,Float64}[]\nmy_lock = Threads.ReentrantLock()\nThreads.@threads for i in 1:10\n    set_lower_bound(x, i)\n    optimize!(model)\n    Threads.lock(my_lock) do\n        push!(solutions, i => objective_value(model))\n    end\nend","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"This will not work, and attempting to do so may error, crash Julia or produce incorrect results.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Using-parallelism-the-right-way","page":"Parallelism","title":"Using parallelism the right way","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use parallelism with JuMP, the simplest rule to remember is that each worker must have its own instance of a JuMP model.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-multi-threading","page":"Parallelism","title":"With multi-threading","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"With multi-threading, create a new instance of model in each iteration of the for-loop:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> using JuMP\n\njulia> import HiGHS\n\njulia> solutions = Pair{Int,Float64}[]\n\njulia> my_lock = Threads.ReentrantLock();\n\njulia> Threads.@threads for i in 1:10\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           set_attribute(model, MOI.NumberOfThreads(), 1)\n           @variable(model, x)\n           @objective(model, Min, x)\n           set_lower_bound(x, i)\n           optimize!(model)\n           Threads.lock(my_lock) do\n               push!(solutions, i => objective_value(model))\n           end\n       end\n\njulia> solutions\n10-element Vector{Pair{Int64, Float64}}:\n  7 => 7.0\n  4 => 4.0\n  1 => 1.0\n  9 => 9.0\n  5 => 5.0\n  8 => 8.0\n 10 => 10.0\n  2 => 2.0\n  6 => 6.0\n  3 => 3.0","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nFor some solvers, it may be necessary to limit the number of threads used internally by the solver to 1 by setting the MOI.NumberOfThreads attribute.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-distributed-computing","page":"Parallelism","title":"With distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"With distributed computing, remember to evaluate all of the code on all of the processes using Distributed.@everywhere, and then write a function which creates a new instance of the model on every evaluation:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.@everywhere begin\n           using JuMP\n           import HiGHS\n       end\n\njulia> Distributed.@everywhere begin\n           function solve_model_with_right_hand_side(i)\n               model = Model(HiGHS.Optimizer)\n               set_silent(model)\n               @variable(model, x)\n               @objective(model, Min, x)\n               set_lower_bound(x, i)\n               optimize!(model)\n               return objective_value(model)\n           end\n       end\n\njulia> solutions = Distributed.pmap(solve_model_with_right_hand_side, 1:10)\n10-element Vector{Float64}:\n  1.0\n  2.0\n  3.0\n  4.0\n  5.0\n  6.0\n  7.0\n  8.0\n  9.0\n 10.0","category":"page"},{"location":"tutorials/algorithms/parallelism/#Other-types-of-parallelism","page":"Parallelism","title":"Other types of parallelism","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/#GPU","page":"Parallelism","title":"GPU","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"JuMP does not support GPU programming, and few solvers support execution on a GPU.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Parallelism-within-the-solver","page":"Parallelism","title":"Parallelism within the solver","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Many solvers use parallelism internally. For example, commercial solvers like Gurobi and CPLEX both parallelize the search in branch-and-bound. Solvers supporting internal parallelism will typically support the MOI.NumberOfThreads attribute, which you can set using set_attribute.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/variables.md\"","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/variables/#Variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"moi/manual/variables/#Add-a-variable","page":"Variables","title":"Add a variable","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Use add_variable to add a single variable.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"add_variable returns a VariableIndex type, which is used to refer to the added variable in other calls.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Check if a VariableIndex is valid using is_valid.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.is_valid(model, x)\ntrue","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Use add_variables to add a number of variables.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> y = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nThe integer does not necessarily correspond to the column inside an optimizer.","category":"page"},{"location":"moi/manual/variables/#Delete-a-variable","page":"Variables","title":"Delete a variable","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Delete a variable using delete.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.delete(model, x)\n\njulia> MOI.is_valid(model, x)\nfalse","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nNot all ModelLike models support deleting variables. A DeleteNotAllowed error is thrown if this is not supported.","category":"page"},{"location":"moi/manual/variables/#Variable-attributes","page":"Variables","title":"Variable attributes","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"The following attributes are available for variables:","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"VariableName\nVariablePrimalStart\nVariablePrimal","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Get and set these attributes using get and set.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.set(model, MOI.VariableName(), x, \"var_x\")\n\njulia> MOI.get(model, MOI.VariableName(), x)\n\"var_x\"","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"EditURL = \"complementarity.jl\"","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Mixed-complementarity-problems","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial is a collection of mixed complementarity programs.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"using JuMP\nimport PATHSolver","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Background","page":"Mixed complementarity problems","title":"Background","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"A mixed complementarity problem has the form:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"beginalign\n    F_i(x) perp x_i  i = 1 ldots n \n    l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"where the perp constraint enforces the following relations:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"If l_i  x_i  u_i, then F_i(x) = 0\nIf l_i = x_i, then F_i(x) ge 0\nIf x_i = u_i, then F_i(x) le 0","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"You may have seen a complementarity problem written as 0 le F(x) perp x ge 0. This is a special case of a mixed complementarity problem in which l_i = 0 and u_i = infty.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"Importantly, a mixed complementarity problem does not have an objective, and no other constraint types are present.","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Linear-complementarity","page":"Mixed complementarity problems","title":"Linear complementarity","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"Form a mixed complementarity problem using the perp symbol ⟂ (type \\perp<tab> in the REPL).","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"M = [0 0 -1 -1; 0 0 1 -2; 1 -1 2 -2; 1 2 -2 4]\nq = [2, 2, -2, -6]\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[1:4] <= 10, start = 0)\n@constraint(model, M * x + q ⟂ x)\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Other-ways-of-writing-linear-complementarity-problems","page":"Mixed complementarity problems","title":"Other ways of writing linear complementarity problems","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"You do not need to use a single vector of variables, and the complementarity constraints can be given in any order. In addition, you can use the perp symbol, the complements(F, x) syntax, or the MOI.Complements set.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"model = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, 0 <= w <= 10, start = 0)\n@variable(model, 0 <= x <= 10, start = 0)\n@variable(model, 0 <= y <= 10, start = 0)\n@variable(model, 0 <= z <= 10, start = 0)\n@constraint(model, complements(y - 2z + 2, x))\n@constraint(model, [-y - z + 2, w] in MOI.Complements(2))\n@constraint(model, w + 2x - 2y + 4z - 6 ⟂ z)\n@constraint(model, w - x + 2y - 2z - 2 ⟂ y)\noptimize!(model)\nvalue.([w, x, y, z])","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Transportation","page":"Mixed complementarity problems","title":"Transportation","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is a reformulation of the transportation problem from Chapter 3.3 of Dantzig, G.B. (1963). Linear Programming and Extensions. Princeton University Press, Princeton, New Jersey. It is based on the GAMS model gamslib_transmcp.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"capacity = Dict(\"seattle\" => 350, \"san-diego\" => 600)\ndemand = Dict(\"new-york\" => 325, \"chicago\" => 300, \"topeka\" => 275)\ncost = Dict(\n    (\"seattle\" => \"new-york\") => 90 * 2.5 / 1_000,\n    (\"seattle\" => \"chicago\") => 90 * 1.7 / 1_000,\n    (\"seattle\" => \"topeka\") => 90 * 1.8 / 1_000,\n    (\"san-diego\" => \"new-york\") => 90 * 2.5 / 1_000,\n    (\"san-diego\" => \"chicago\") => 90 * 1.8 / 1_000,\n    (\"san-diego\" => \"topeka\") => 90 * 1.4 / 1_000,\n)\nplants, markets = keys(capacity), keys(demand)\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, w[i in plants] >= 0)\n@variable(model, p[j in markets] >= 0)\n@variable(model, x[i in plants, j in markets] >= 0)\n@constraints(\n    model,\n    begin\n        [i in plants, j in markets], w[i] + cost[i=>j] - p[j] ⟂ x[i, j]\n        [i in plants], capacity[i] - sum(x[i, :]) ⟂ w[i]\n        [j in markets], sum(x[:, j]) - demand[j] ⟂ p[j]\n    end\n)\noptimize!(model)\nvalue.(p)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Expected-utility-of-insurance","page":"Mixed complementarity problems","title":"Expected utility of insurance","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is taken from a lecture of the course AAE706, given by Thomas F. Rutherford at the University of Wisconsin, Madison. It models the expected coverage of insurance K that a rational actor would obtain to insure a risk that occurs with probability pi and results in a loss of L.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"pi = 0.01  # Probability of a bad outcome\nL = 0.5    # Loss with a bad outcome\nγ = 0.02   # Premium for coverage\nσ = 0.5    # Elasticity\nρ = -1     # Risk exponent\nU(C) = C^ρ / ρ\nMU(C) = C^(ρ - 1)\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, EU, start = 1)   # Expected utilitiy\n@variable(model, EV, start = 1)   # Equivalent variation in income\n@variable(model, C_G, start = 1)  # Consumption on a good day\n@variable(model, C_B, start = 1)  # Consumption on a bad day\n@variable(model, K, start = 1)    # Coverage\n@constraints(\n    model,\n    begin\n        (1 - pi) * U(C_G) + pi * U(C_B) - EU ⟂ EU\n        100 * (((1 - pi) * C_G^ρ + pi * C_B^ρ)^(1 / ρ) - 1) - EV ⟂ EV\n        1 - γ * K - C_G ⟂ C_G\n        1 - L + (1 - γ) * K - C_B ⟂ C_B\n        γ * ((1 - pi) * MU(C_G) + pi * MU(C_B)) - pi * MU(C_B) ⟂ K\n    end\n)\noptimize!(model)\nvalue(K)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Electricity-consumption","page":"Mixed complementarity problems","title":"Electricity consumption","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is a mixed complementarity formulation of example 3.3.1 from D’Aertrycke, G., Ehrenmann, A., Ralph, D., & Smeers, Y. (2017). Risk trading in capacity equilibrium models.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example models a risk neutral competitive equilibrium between a producer and a consumer of electricity.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"In our example, we assume a producer is looking to invest in a new power plant with capacity x [MW]. This plant has an annualized capital cost of I [€/MW] and an operating cost of C [€/MWh]. There are 8760 hours in a year.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"After making the capital investment, there are five possible consumption scenarios, omega, which occur with probability theta_omega. In each scenario , the producer makes Y_ω MW of electricity.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"There is one consumer in the model, who has a quadratic utility function, U(Q_ω) = A_ω Q_ω + fracB_ω Q_ω^22.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"We now build and solve the mixed complementarity problem with a few brief comments. The economic justification for the model would require a larger tutorial than the space available here. Consult the original text for details.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"I = 90_000                     # Annualized capital cost\nC = 60                         # Operation cost per MWh\nτ = 8_760                      # Hours per year\nθ = [0.2, 0.2, 0.2, 0.2, 0.2]  # Scenario probabilities\nA = [300, 350, 400, 450, 500]  # Utility function coefficients\nB = 1                          # Utility function coefficients\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, x >= 0, start = 1)           # Installed capacity\n@variable(model, Q[ω = 1:5] >= 0, start = 1)  # Consumption\n@variable(model, Y[ω = 1:5] >= 0, start = 1)  # Production\n@variable(model, P[ω = 1:5], start = 1)       # Electricity price\n@variable(model, μ[ω = 1:5] >= 0, start = 1)  # Capital scarcity margin\n# Unit investment cost equals annualized scarcity profit or investment is 0\n@constraint(model, I - τ * θ' * μ ⟂ x)\n# Difference between price and scarcity margin is equal to operation cost\n@constraint(model, [ω = 1:5], C - (P[ω] - μ[ω]) ⟂ Y[ω])\n# Price is equal to consumer's marginal utility\n@constraint(model, [ω = 1:5], P[ω] - (A[ω] - B * Q[ω]) ⟂ Q[ω])\n# Production is equal to consumption\n@constraint(model, [ω = 1:5], Y[ω] - Q[ω] ⟂ P[ω])\n# Production does not exceed capacity\n@constraint(model, [ω = 1:5], x - Y[ω] ⟂ μ[ω])\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"An equilibrium solution is to build 389 MW:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value(x)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"The production in each scenario is:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value.(Q)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"The price in each scenario is:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value.(P)","category":"page"},{"location":"developers/custom_solver_binaries/#How-to-use-a-custom-binary","page":"Custom binaries","title":"How to use a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Many solvers are not written in Julia, but instead in languages like C or C++. JuMP interacts with these solvers through binary dependencies.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"For many open-source solvers, we automatically install the appropriate binary when you run Pkg.add(\"Solver\"). For example, Pkg.add(\"ECOS\") will also install the ECOS binary.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"This page explains how this installation works, and how you can use a custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"compat: Compat\nThese instructions require Julia 1.6 or later.","category":"page"},{"location":"developers/custom_solver_binaries/#Background","page":"Custom binaries","title":"Background","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Each solver that JuMP supports is structured as a Julia package. For example, the interface for the ECOS solver is provided by the ECOS.jl package.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"tip: Tip\nThis page uses the example of ECOS.jl because it is simple to compile. Other solvers follow similar conventions. For example, the interface to the Clp solver is provided by Clp.jl.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The ECOS.jl package provides an interface between the C API of ECOS and MathOptInterface. However, it does not handle the installation of the solver binary; that is the job for a JLL package.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"A JLL is a Julia package that wraps a pre-compiled binary. Binaries are built using Yggdrasil (for example, ECOS) and hosted in the JuliaBinaryWrappers GitHub repository (for example, ECOS_jll.jl).","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"JLL packages contain little code. Their only job is to dlopen a dynamic library, along with any dependencies.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"JLL packages manage their binary dependencies using Julia's artifact system. Each JLL package has an Artifacts.toml file which describes where to find each binary artifact for each different platform that it might be installed on. Here is the Artifacts.toml file for ECOS_jll.jl.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The binaries installed by the JLL package should be sufficient for most users. In rare cases, however, you may require a custom binary. The two main reasons to use a custom binary are:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"You want a binary with custom compilation settings (for example, debugging)\nYou want a binary with a set of dependencies that are not available on Yggdrasil (for example, a commercial solver like Gurobi or CPLEX).","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The following sections explain how to replace the binaries provided by a JLL package with the custom ones you have compiled. As a reminder, we use ECOS as an example for simplicity, but the steps are the same for other solvers.","category":"page"},{"location":"developers/custom_solver_binaries/#jll_structure","page":"Custom binaries","title":"Explore the JLL you want to override","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The first step is to explore the structure and filenames of the JLL package we want to override.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Find the location of the files using .artifact_dir:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.artifact_dir\n\"/Users/oscar/.julia/artifacts/2addb75332eff5a1657b46bb6bf30d2410bc7ecf\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"tip: Tip\nThis path may be different on other machines.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Here is what it contains:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> readdir(ECOS_jll.artifact_dir)\n4-element Vector{String}:\n \"include\"\n \"lib\"\n \"logs\"\n \"share\"\n\njulia> readdir(joinpath(ECOS_jll.artifact_dir, \"lib\"))\n1-element Vector{String}:\n \"libecos.dylib\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Other solvers may have a bin directory containing executables. To use a custom binary of ECOS, we need to replace /lib/libecos.dylib with our custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#compile_ecos","page":"Custom binaries","title":"Compile a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The next step is to compile a custom binary. Because ECOS is written in C with no dependencies, this is easy to do if you have a C compiler:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"oscar@Oscars-MBP jll_example % git clone https://github.com/embotech/ecos.git\n[... lines omitted ...]\noscar@Oscars-MBP jll_example % cd ecos\noscar@Oscars-MBP ecos % make shared\n[... many lines omitted...]\noscar@Oscars-MBP ecos % mkdir lib\noscar@Oscars-MBP ecos % cp libecos.dylib lib","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"warning: Warning\nCompiling custom solver binaries is an advanced operation. Due to the complexities of compiling various solvers, the JuMP community is unable to help you diagnose and fix compilation issues.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"After this compilation step, we now have a folder /tmp/jll_example/ecos that contains lib and include directories with the same files as ECOS_jll:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> readdir(joinpath(\"ecos\", \"lib\"))\n1-element Vector{String}:\n \"libecos.dylib\"","category":"page"},{"location":"developers/custom_solver_binaries/#Overriding-a-single-library","page":"Custom binaries","title":"Overriding a single library","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To override the libecos library, we need to know what ECOS_jll calls it. (In most cases, it will also be libecos, but not always.)","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"There are two ways you can check.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Check the bottom of the JLL's GitHub README. For example, ECOS_jll has a single LibraryProduct called libecos.\nType ECOS_jll. and the press the [TAB] key twice to auto-complete available options:\njulia> ECOS_jll.\nLIBPATH           PATH_list          best_wrapper       get_libecos_path   libecos_handle\nLIBPATH_list      __init__           dev_jll            is_available       libecos_path\nPATH              artifact_dir       find_artifact_dir  libecos\nHere you can see there is libecos, and more usefully for us, libecos_path.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Once you know the name of the variable to override (the one that ends in _path), use  Preferences.jl to specify a new path:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"using Preferences\nset_preferences!(\n    \"LocalPreferences.toml\",\n    \"ECOS_jll\",\n    \"libecos_path\" => \"/tmp/jll_example/ecos/lib/libecos\"\n)","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"This will create a file in your current directory called LocalPreferences.toml with the contents:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"[ECOS_jll]\nlibecos_path = \"/tmp/jll_example/ecos/lib/libecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Now if you restart Julia, you will see:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.libecos\n\"/tmp/jll_example/ecos/lib/libecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To go back to using the default library, just delete the LocalPreferences.toml file.","category":"page"},{"location":"developers/custom_solver_binaries/#Overriding-an-entire-artifact","page":"Custom binaries","title":"Overriding an entire artifact","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Sometimes a solver may provide a number of libraries and executables, and specifying the path for each of the becomes tedious. In this case, we can use Julia's Override.toml to replace an entire artifact.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Overriding an entire artifact requires you to replicate the structure and contents of the JLL package that we explored above.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"In most cases you need only reproduce the include, lib, and bin directories (if they exist). You can safely ignore any logs or share directories. Take careful note of what files each directory contains and what they are called.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"For our ECOS example, we already reproduced the structure when we compiled ECOS.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"So, now we need to tell Julia to use our custom installation instead of the default. We can do this by making an override file at ~/.julia/artifacts/Overrides.toml.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Overrides.toml has the following content:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"# Override for ECOS_jll\n2addb75332eff5a1657b46bb6bf30d2410bc7ecf = \"/tmp/jll_example/ecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"where 2addb75332eff5a1657b46bb6bf30d2410bc7ecf is the folder from the original ECOS_jll.artifact_dir and \"/tmp/jll_example/ecos\" is the location of our new installation. Replace these as appropriate for your system.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"If you restart Julia after creating the override file, you will see:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.artifact_dir\n\"/tmp/jll_example/ecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Now when we use ECOS it will use our custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#Using-Cbc-with-a-custom-binary","page":"Custom binaries","title":"Using Cbc with a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"As a second example, we demonstrate how to use Cbc.jl with a custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#Explore-the-JLL-you-want-to-override","page":"Custom binaries","title":"Explore the JLL you want to override","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"First, let's check where Cbc_jll is installed:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using Cbc_jll\n\njulia> Cbc_jll.artifact_dir\n\"/Users/oscar/.julia/artifacts/e481bc81db5e229ba1f52b2b4bd57484204b1b06\"\n\njulia> readdir(Cbc_jll.artifact_dir)\n5-element Vector{String}:\n \"bin\"\n \"include\"\n \"lib\"\n \"logs\"\n \"share\"\n\njulia> readdir(joinpath(Cbc_jll.artifact_dir, \"bin\"))\n1-element Vector{String}:\n \"cbc\"\n\njulia> readdir(joinpath(Cbc_jll.artifact_dir, \"lib\"))\n10-element Vector{String}:\n \"libCbc.3.10.5.dylib\"\n \"libCbc.3.dylib\"\n \"libCbc.dylib\"\n \"libCbcSolver.3.10.5.dylib\"\n \"libCbcSolver.3.dylib\"\n \"libCbcSolver.dylib\"\n \"libOsiCbc.3.10.5.dylib\"\n \"libOsiCbc.3.dylib\"\n \"libOsiCbc.dylib\"\n \"pkgconfig\"","category":"page"},{"location":"developers/custom_solver_binaries/#Compile-a-custom-binary","page":"Custom binaries","title":"Compile a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Next, we need to compile Cbc. Cbc can be difficult to compile (it has a lot of dependencies), but for macOS users there is a homebrew recipe:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"(base) oscar@Oscars-MBP jll_example % brew install cbc\n[ ... lines omitted ... ]\n(base) oscar@Oscars-MBP jll_example % brew list cbc\n/usr/local/Cellar/cbc/2.10.5/bin/cbc\n/usr/local/Cellar/cbc/2.10.5/include/cbc/ (76 files)\n/usr/local/Cellar/cbc/2.10.5/lib/libCbc.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/libCbcSolver.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/libOsiCbc.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/pkgconfig/ (2 files)\n/usr/local/Cellar/cbc/2.10.5/lib/ (6 other files)\n/usr/local/Cellar/cbc/2.10.5/share/cbc/ (59 files)\n/usr/local/Cellar/cbc/2.10.5/share/coin/ (4 files)","category":"page"},{"location":"developers/custom_solver_binaries/#Override-single-libraries","page":"Custom binaries","title":"Override single libraries","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To use Preferences.jl to override specific libraries we first check the names of each library in Cbc_jll:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> Cbc_jll.\nLIBPATH               cbc                    get_libcbcsolver_path  libOsiCbc_path\nLIBPATH_list          cbc_path               is_available           libcbcsolver\nPATH                  dev_jll                libCbc                 libcbcsolver_handle\nPATH_list             find_artifact_dir      libCbc_handle          libcbcsolver_path\n__init__              get_cbc_path           libCbc_path\nartifact_dir          get_libCbc_path        libOsiCbc\nbest_wrapper          get_libOsiCbc_path     libOsiCbc_handle","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Then we add the following to LocalPreferences.toml:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"[Cbc_jll]\ncbc_path = \"/usr/local/Cellar/cbc/2.10.5/bin/cbc\"\nlibCbc_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libCbc.3.10.5\"\nlibOsiCbc_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libOsiCbc.3.10.5\"\nlibcbcsolver_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libCbcSolver.3.10.5\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"info: Info\nNote that capitalization matters, so libcbcsolver_path corresponds to libCbcSolver.3.10.5.","category":"page"},{"location":"developers/custom_solver_binaries/#Override-entire-artifact","page":"Custom binaries","title":"Override entire artifact","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To use the homebrew install as our custom binary we add the following to ~/.julia/artifacts/Overrides.toml:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"# Override for Cbc_jll\ne481bc81db5e229ba1f52b2b4bd57484204b1b06 = \"/usr/local/Cellar/cbc/2.10.5\"","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"EditURL = \"benders_decomposition.jl\"","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#benders_decomposition_classical","page":"Benders decomposition","title":"Benders decomposition","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial was originally contributed by Shuvomoy Das Gupta.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial describes how to implement Benders decomposition in JuMP. It uses the following packages:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"using JuMP\nimport GLPK\nimport Printf","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Theory","page":"Benders decomposition","title":"Theory","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition is a useful algorithm for solving convex optimization problems with a large number of variables. It works best when a larger problem can be decomposed into two (or more) smaller problems that are individually much easier to solve. This tutorial demonstrates Benders decomposition on the following mixed-integer linear program:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\ntextmin          c_1^top x+c_2^top y   \ntextsubject to   A_1 x+ A_2 y le b      \n                     x ge 0                 \n                     y ge 0                 \n                     x in mathbbZ^n\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"where b in mathbbR^m, A_1 in mathbbR^m times n, A_2 in mathbbR^m times p and mathbbZ is the set of integers.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Any mixed integer programming problem can be written in the form above.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"If there are relatively few integer variables, and many more continuous variables, then it may be beneficial to decompose the problem into a small problem containing only integer variables and a linear program containing only continuous variables. Hopefully, the linear program will be much easier to solve in isolation than in the full mixed-integer linear program.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For example, if we knew a feasible solution for x, we could obtain a solution for y by solving:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\nV_2(x) =  textmin            c_2^top y                        \n          textsubject to     A_2 y le b - A_1 x  quad pi \n                                 y ge 0\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"where pi is the dual variable associated with the constraints. Because this is a linear program, it is easy to solve.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Replacing the c_2^top y component of the objective in our original problem with V_2 yields:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\ntextmin          c_1^top x + V_2(x) \ntextsubject to   x ge 0             \n                     x in mathbbZ^n\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This problem looks a lot simpler to solve, but we need to do something else with V_2 first.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Because x is a constant that appears on the right-hand side of the constraints, V_2 is a convex function with respect to x, and the dual variable pi can be multiplied by -A_1 to obtain a subgradient of V_2(x) with respect to x. Therefore, if we have a candidate solution x_k, then we can solve V_2(x_k) and obtain a feasible dual vector pi_k. Using these values, we can construct a first-order Taylor-series approximation of V_2 about the point x_k:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"V_2(x) ge V_2(x_k) + -pi_k^top A_1 (x - x_k)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"By convexity, we know that this inequality holds for all x, and we call these inequalities cuts.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition is an iterative technique that replaces V_2(x) with a new decision variable theta, and approximates it from below using cuts:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\nV_1^K =  textmin           c_1^top x + theta  \n          textsubject to   x ge 0              \n                              x in mathbbZ^n   \n                              theta ge M         \n                              theta ge V_2(x_k) + pi_k^top(x - x_k)  quad forall k = 1ldotsK\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This integer program is called the first-stage subproblem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"To generate cuts, we solve V_1^K to obtain a candidate first-stage solution x_k, then we use that solution to solve V_2(x_k). Then, using the optimal objective value and dual solution from V_2, we add a new cut to form V_1^K+1 and repeat.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Bounds","page":"Benders decomposition","title":"Bounds","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Due to convexity, we know that V_2(x) ge theta for all x. Therefore, the optimal objective value of V_1^K provides a valid lower bound on the objective value of the full problem. In addition, if we take a feasible solution for x from the first-stage problem, then c_1^top x + V_2(x) is a valid upper bound on the objective value of the full problem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition uses the lower and upper bounds to determine when it has found the global optimal solution.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Input-data","page":"Benders decomposition","title":"Input data","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As an example for this tutorial, we use the input data is from page 139 of Garfinkel, R. & Nemhauser, G. L. Integer programming. (Wiley, 1972).","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"c_1 = [1, 4]\nc_2 = [2, 3]\ndim_x = length(c_1)\ndim_y = length(c_2)\nb = [-2; -3]\nA_1 = [1 -3; -1 -3]\nA_2 = [1 -2; -1 -1]\nM = -1000;\nnothing #hide","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#benders_iterative","page":"Benders decomposition","title":"Iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"warning: Warning\nThis is a basic implementation for pedagogical purposes. We haven't discussed Benders feasibility cuts, or any of the computational tricks that are required to build a performant implementation for large-scale problems. See In-place iterative method for one improvement that helps computation time.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We start by formulating the first-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"model = Model(GLPK.Optimizer)\n@variable(model, x[1:dim_x] >= 0, Int)\n@variable(model, θ >= M)\n@objective(model, Min, c_1' * x + θ)\nprint(model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For the next step, we need a function that takes a first-stage candidate solution x and returns the optimal solution from the second-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function solve_subproblem(x)\n    model = Model(GLPK.Optimizer)\n    @variable(model, y[1:dim_y] >= 0)\n    con = @constraint(model, A_2 * y .<= b - A_1 * x)\n    @objective(model, Min, c_2' * y)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    return (obj = objective_value(model), y = value.(y), π = dual.(con))\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Note that solve_subproblem returns a NamedTuple of the objective value, the optimal primal solution for y, and the optimal dual solution for π.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We're almost ready for our optimization loop, but first, here's a helpful function for logging:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function print_iteration(k, args...)\n    f(x) = Printf.@sprintf(\"%12.4e\", x)\n    println(lpad(k, 9), \" \", join(f.(args), \" \"))\n    return\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We also need to put a limit on the number of iterations before termination:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"MAXIMUM_ITERATIONS = 100","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"And a way to check if the lower and upper bounds are close-enough to terminate:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"ABSOLUTE_OPTIMALITY_GAP = 1e-6","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now we're ready to iterate Benders decomposition:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"println(\"Iteration  Lower Bound  Upper Bound          Gap\")\nfor k in 1:MAXIMUM_ITERATIONS\n    optimize!(model)\n    lower_bound = objective_value(model)\n    x_k = value.(x)\n    ret = solve_subproblem(x_k)\n    upper_bound = c_1' * x_k + ret.obj\n    gap = (upper_bound - lower_bound) / upper_bound\n    print_iteration(k, lower_bound, upper_bound, gap)\n    if gap < ABSOLUTE_OPTIMALITY_GAP\n        println(\"Terminating with the optimal solution\")\n        break\n    end\n    cut = @constraint(model, θ >= ret.obj + -ret.π' * A_1 * (x .- x_k))\n    @info \"Adding the cut $(cut)\"\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(model)\nx_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Callback-method","page":"Benders decomposition","title":"Callback method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"The Iterative method section implemented Benders decomposition using a loop. In each iteration, we re-solved the first-stage subproblem to generate a candidate solution. However, modern MILP solvers such as CPLEX, Gurobi, and GLPK provide lazy constraint callbacks which allow us to add new cuts while the solver is running. This can be more efficient than an iterative method because we can avoid repeating work such as solving the root node of the first-stage MILP at each iteration.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"tip: Tip\nFor more information on callbacks, read the page Solver-independent callbacks.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As before, we construct the same first-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"lazy_model = Model(GLPK.Optimizer)\n@variable(lazy_model, x[1:dim_x] >= 0, Int)\n@variable(lazy_model, θ >= M)\n@objective(lazy_model, Min, c_1' * x + θ)\nprint(lazy_model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"What differs is that we write a callback function instead of a loop:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"number_of_subproblem_solves = 0\nfunction my_callback(cb_data)\n    status = callback_node_status(cb_data, lazy_model)\n    if status != MOI.CALLBACK_NODE_STATUS_INTEGER\n        # Only add the constraint if `x` is an integer feasible solution\n        return\n    end\n    x_k = callback_value.(cb_data, x)\n    θ_k = callback_value(cb_data, θ)\n    global number_of_subproblem_solves += 1\n    ret = solve_subproblem(x_k)\n    if θ_k < (ret.obj - 1e-6)\n        # Only add the constraint if θ_k violates the constraint\n        cut = @build_constraint(θ >= ret.obj + -ret.π' * A_1 * (x .- x_k))\n        MOI.submit(lazy_model, MOI.LazyConstraint(cb_data), cut)\n    end\n    return\nend\n\nset_attribute(lazy_model, MOI.LazyConstraintCallback(), my_callback)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now when we optimize!, our callback is run:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(lazy_model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For this model, the callback algorithm required more solves of the subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"number_of_subproblem_solves","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"But for larger problems, you can expect the callback algorithm to be more efficient than the iterative algorithm.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"x_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#In-place-iterative-method","page":"Benders decomposition","title":"In-place iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Our implementation of the iterative method has a problem: every time we need to solve the subproblem, we must rebuild it from scratch. This is expensive, and it can be the bottleneck in the solution process. We can improve our implementation by using re-using the subproblem between solves.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"First, we create our first-stage problem as usual:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"model = Model(GLPK.Optimizer)\n@variable(model, x[1:dim_x] >= 0, Int)\n@variable(model, θ >= M)\n@objective(model, Min, c_1' * x + θ)\nprint(model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Then, instead of building the subproblem in a function, we build it once here:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"subproblem = Model(GLPK.Optimizer)\n@variable(subproblem, x_copy[1:dim_x])\n@variable(subproblem, y[1:dim_y] >= 0)\n@constraint(subproblem, A_1 * x_copy + A_2 * y .<= b)\n@objective(subproblem, Min, c_2' * y)\nprint(subproblem)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This formulation is slightly different. We have included a copy of the x variables, x_copy, and used x_copy in the left-hand side of the constraints.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Our function to solve the subproblem is also slightly different. First, we need to fix the value of the x_copy variables to the value of x from the first-stage problem, and second, we compute the dual using the reduced_cost of x_copy, not the dual of the linear constraints:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function solve_subproblem(model, x)\n    fix.(model[:x_copy], x)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    return (\n        obj = objective_value(model),\n        y = value.(model[:y]),\n        π = reduced_cost.(model[:x_copy]),\n    )\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now we're ready to iterate our in-place Benders decomposition, but this time the cut computation is slightly different. Because we used reduced_cost on the x_copy variables, the value of π is a valid subgradient on the objective of subproblem with respect to x. Therefore, we don't need to multiply the duals by -A_1, and so our cut uses ret.π' instead of -ret.π' * A_1:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"println(\"Iteration  Lower Bound  Upper Bound          Gap\")\nfor k in 1:MAXIMUM_ITERATIONS\n    optimize!(model)\n    lower_bound = objective_value(model)\n    x_k = value.(x)\n    ret = solve_subproblem(subproblem, x_k)\n    upper_bound = c_1' * x_k + ret.obj\n    gap = (upper_bound - lower_bound) / upper_bound\n    print_iteration(k, lower_bound, upper_bound, gap)\n    if gap < ABSOLUTE_OPTIMALITY_GAP\n        println(\"Terminating with the optimal solution\")\n        break\n    end\n    cut = @constraint(model, θ >= ret.obj + ret.π' * (x .- x_k))\n    @info \"Adding the cut $(cut)\"\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(model)\nx_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(subproblem, x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For larger problems, the benefit of re-using the same subproblem and not needing to multiply the duals by A_1 in the cut coefficient usually outweights the cost of needing a copy of the x variables in the subproblem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As a secondary benefit, because we no longer need an explicit representation of A_1 in the cut, we can build the model and subproblem formulations using arbitrary JuMP syntax; they do not need to be in matrix form.","category":"page"},{"location":"tutorials/nonlinear/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Nonlinear programs (NLPs) are a class of optimization problems in which some of the constraints or the objective function are nonlinear:","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n    min_x in mathbbR^n  f_0(x) \n    textst  l_j le f_j(x) le u_j  j = 1 ldots m \n     l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer nonlinear linear programs (MINLPs) are extensions of nonlinear programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/nonlinear/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"JuMP supports a range of nonlinear solvers; look for \"NLP\" in the list of Supported solvers. However, very few solvers support mixed-integer nonlinear linear programs. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"If the only nonlinearities in your model are quadratic terms (that is, multiplication between two decision variables), you can also use second-order cone solvers, which are indicated by \"SOCP.\" In most cases, these solvers are restricted to convex quadratic problems and will error if you pass a nonconvex quadratic function; however, Gurobi has the ability to solve nonconvex quadratic terms.","category":"page"},{"location":"tutorials/nonlinear/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nRocket Control\nOptimal control for a Space Shuttle reentry trajectory\nPortfolio optimization\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling nonlinear programs. Look here if you are stuck trying to formulate a problem as a nonlinear program.\nThe Computing Hessians is an advanced tutorial which explains how to compute the Hessian of the Lagrangian of a nonlinear program. This is useful only in particular cases.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"EditURL = \"https://github.com/jump-dev/Pajarito.jl/blob/4c7efa915bdc900b51d1c8290dc771403b11230a/README.md\"","category":"page"},{"location":"packages/Pajarito/#Pajarito.jl","page":"jump-dev/Pajarito.jl","title":"Pajarito.jl","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito is a mixed-integer convex programming (MICP) solver package written in Julia.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"MICP problems are convex except for restrictions that some variables take binary or integer values.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito solves MICP problems in conic form, by constructing sequential polyhedral outer approximations of the conic feasible set.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The underlying algorithm has theoretical finite-time convergence under reasonable assumptions.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito accesses state-of-the-art MILP solvers and continuous conic solvers through MathOptInterface.","category":"page"},{"location":"packages/Pajarito/#Pavito","page":"jump-dev/Pajarito.jl","title":"Pavito","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"For algorithms that use a derivative-based nonlinear programming (NLP) solver (for example, Ipopt) instead of a conic solver, use Pavito.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pavito is a convex mixed-integer nonlinear programming (convex MINLP) solver. Because Pavito relies on gradient cuts, it can fail near points of non-differentiability. Pajarito may be more robust than Pavito on non-smooth problems.","category":"page"},{"location":"packages/Pajarito/#License","page":"jump-dev/Pajarito.jl","title":"License","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Pajarito/#Installation","page":"jump-dev/Pajarito.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Install Pajarito using Pkg.add:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"import Pkg\nPkg.add(\"Pajarito\")","category":"page"},{"location":"packages/Pajarito/#MIP-and-continuous-solvers","page":"jump-dev/Pajarito.jl","title":"MIP and continuous solvers","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The algorithm implemented by Pajarito itself is relatively simple, and most of the hard work is performed by the MIP outer approximation (OA) solver and the continuous conic solver.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Therefore, in addition to installing Pajarito, you must also install a mixed-integer linear programming solver and a continuous conic solver.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The performance of Pajarito depends on these two types of solvers.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The OA solver (typically a mixed-integer linear solver) is specified by the oa_solver option. You must first load the Julia package that provides this solver, for example, using Gurobi. The continuous conic solver is specified by the conic_solver option.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"See JuMP's list of supported solvers.","category":"page"},{"location":"packages/Pajarito/#Use-with-JuMP","page":"jump-dev/Pajarito.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"To use Pajarito with JuMP, use:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"using JuMP, Pajarito, HiGHS, Hypatia\nmodel = Model(\n    optimizer_with_attributes(\n        Pajarito.Optimizer,\n        \"oa_solver\" => optimizer_with_attributes(\n            HiGHS.Optimizer,\n            MOI.Silent() => true,\n            \"mip_feasibility_tolerance\" => 1e-8,\n            \"mip_rel_gap\" => 1e-6,\n        ),\n        \"conic_solver\" =>\n            optimizer_with_attributes(Hypatia.Optimizer, MOI.Silent() => true),\n    )\n)\nset_attribute(model, \"time_limit\", 60)","category":"page"},{"location":"packages/Pajarito/#Options","page":"jump-dev/Pajarito.jl","title":"Options","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"We list Pajarito's options below.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"verbose::Bool toggles printing\ntol_feas::Float64 is the feasibility tolerance for conic constraints\ntol_rel_gap::Float64 is the relative optimality gap tolerance\ntol_abs_gap::Float64 is the absolute optimality gap tolerance\ntime_limit::Float64 sets the time limit (in seconds)\niteration_limit::Int sets the iteration limit (for the iterative method)\nuse_iterative_method::Union{Nothing,Bool} toggles the iterative algorithm; if nothing is specified, Pajarito defaults to the OA-solver-driven (single tree) algorithm if lazy callbacks are supported by the OA solver\nuse_extended_form::Bool toggles the use of extended formulations for the second-order cone\nsolve_relaxation::Bool toggles solution of the continuous conic relaxation\nsolve_subproblems::Bool toggles solution of the continuous conic subproblems\nuse_init_fixed_oa::Bool toggles initial fixed OA cuts\noa_solver::Union{Nothing,MOI.OptimizerWithAttributes} is the OA solver\nconic_solver::Union{Nothing,MOI.OptimizerWithAttributes} is the conic solver","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito may require tuning of parameters to improve convergence.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"For example, it often helps to tighten the OA solver's integrality tolerance. OA solver and conic solver options must be specified directly to those solvers.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Note: if solve_subproblems is true, Pajarito usually returns a solution constructed from one of the conic solver's feasible solutions; since the conic solver is not subject to the same feasibility tolerances as the OA solver, Pajarito's solution will not necessarily satisfy tol_feas.","category":"page"},{"location":"packages/Pajarito/#Cone-interface","page":"jump-dev/Pajarito.jl","title":"Cone interface","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito has a generic cone interface (see the cones folder that allows the user to add support for new convex cones.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"To illustrate, in the experimental package PajaritoExtras we have extended Pajarito by adding support for several cones recognized by Hypatia.jl (a continuous conic solver with its own generic cone interface).","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The examples folder of PajaritoExtras also contains many applied mixed-integer convex problems that are solved using Pajarito.","category":"page"},{"location":"packages/Pajarito/#Bug-reports-and-support","page":"jump-dev/Pajarito.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Please report any issues via the GitHub issue tracker.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"All types of issues are welcome and encouraged; this includes bug reports, documentation typos, and feature requests. The Optimization (Mathematical) category on Discourse is appropriate for general discussion.","category":"page"},{"location":"packages/Pajarito/#References","page":"jump-dev/Pajarito.jl","title":"References","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"If you find Pajarito useful in your work, we kindly request that you cite the following paper (arXiv preprint), which is recommended reading for advanced users:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"@article{CoeyLubinVielma2020,\n    title={Outer approximation with conic certificates for mixed-integer convex problems},\n    author={Coey, Chris and Lubin, Miles and Vielma, Juan Pablo},\n    journal={Mathematical Programming Computation},\n    volume={12},\n    number={2},\n    pages={249--293},\n    year={2020},\n    publisher={Springer}\n}","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Note this paper describes a legacy MathProgBase version of Pajarito, which is available on the mathprogbase branch of this repository. Starting with version v0.8.0, Pajarito supports MathOptInterface instead of MathProgBase.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/models.md\"","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/models/#Models","page":"Models","title":"Models","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The most significant part of MOI is the definition of the model API that is used to specify an instance of an optimization problem (for example, by adding variables and constraints). Objects that implement the model API must inherit from the ModelLike abstract type.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Notably missing from the model API is the method to solve an optimization problem. ModelLike objects may store an instance (for example, in memory or backed by a file format) without being linked to a particular solver. In addition to the model API, MOI defines AbstractOptimizer and provides methods to solve the model and interact with solutions. See the Solutions section for more details.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"info: Info\nThroughout the rest of the manual, model is used as a generic ModelLike, and optimizer is used as a generic AbstractOptimizer.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"tip: Tip\nMOI does not export functions, but for brevity we often omit qualifying names with the MOI module. Best practice is to haveimport MathOptInterface as MOIand prefix all MOI methods with MOI. in user code. If a name is also available in base Julia, we always explicitly use the module prefix, for example, with MOI.get.","category":"page"},{"location":"moi/manual/models/#Attributes","page":"Models","title":"Attributes","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Attributes are properties of the model that can be queried and modified. These include constants such as the number of variables in a model NumberOfVariables), and properties of variables and constraints such as the name of a variable (VariableName).","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"There are four types of attributes:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Model attributes (subtypes of AbstractModelAttribute) refer to properties of a model.\nOptimizer attributes (subtypes of AbstractOptimizerAttribute) refer to properties of an optimizer.\nConstraint attributes (subtypes of AbstractConstraintAttribute) refer to properties of an individual constraint.\nVariable attributes (subtypes of AbstractVariableAttribute) refer to properties of an individual variable.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Some attributes are values that can be queried by the user but not modified, while other attributes can be modified by the user.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"All interactions with attributes occur through the get and set functions.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Consult the docstrings of each attribute for information on what it represents.","category":"page"},{"location":"moi/manual/models/#ModelLike-API","page":"Models","title":"ModelLike API","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The following attributes are available:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"ListOfConstraintAttributesSet\nListOfConstraintIndices\nListOfConstraintTypesPresent\nListOfModelAttributesSet\nListOfVariableAttributesSet\nListOfVariableIndices\nNumberOfConstraints\nNumberOfVariables\nName\nObjectiveFunction\nObjectiveFunctionType\nObjectiveSense","category":"page"},{"location":"moi/manual/models/#AbstractOptimizer-API","page":"Models","title":"AbstractOptimizer API","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The following attributes are available:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"DualStatus\nPrimalStatus\nRawStatusString\nResultCount\nTerminationStatus\nBarrierIterations\nDualObjectiveValue\nNodeCount\nNumberOfThreads\nObjectiveBound\nObjectiveValue\nRelativeGap\nRawOptimizerAttribute\nRawSolver\nSilent\nSimplexIterations\nSolverName\nSolverVersion\nSolveTimeSec\nTimeLimitSec\nObjectiveLimit","category":"page"},{"location":"packages/NLopt/#NLopt.jl","page":"JuliaOpt/NLopt.jl","title":"NLopt.jl","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"NLopt.jl is a wrapper for the NLopt library.","category":"page"},{"location":"packages/NLopt/#License","page":"JuliaOpt/NLopt.jl","title":"License","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"NLopt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"The underlying solver, stevengj/nlopt, is licensed under the LGPL v3.0 license.","category":"page"},{"location":"packages/NLopt/#Installation","page":"JuliaOpt/NLopt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"Install NLopt.jl using the Julia package manager:","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"import Pkg\nPkg.add(\"NLopt\")","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"In addition to installing the NLopt.jl package, this will also download and install the NLopt binaries. You do not need to install NLopt separately.","category":"page"},{"location":"packages/NLopt/#Use-with-JuMP","page":"JuliaOpt/NLopt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"You can use NLopt with JuMP as follows:","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"using JuMP, NLopt\nmodel = Model(NLopt.Optimizer)\nset_attribute(model, \"algorithm\", :LD_MMA)","category":"page"},{"location":"packages/NLopt/#Options","page":"JuliaOpt/NLopt.jl","title":"Options","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"The algorithm attribute is required. The value must be one of the supported NLopt algorithms.","category":"page"},{"location":"packages/NLopt/#Documentation","page":"JuliaOpt/NLopt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"For more details, see the NLopt.jl README or the NLopt documentation.","category":"page"},{"location":"tutorials/linear/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Linear programs (LPs) are a fundamental class of optimization problems of the form:","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n    min_x in mathbbR^n  sumlimits_i=1^n c_i x_i \n    textst  l_j le sumlimits_i=1^n a_ij x_i le u_j  j = 1 ldots m \n     l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"The most important thing to note is that all terms are of the form coefficient * variable, and that there are no nonlinear terms or multiplications between variables.","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer linear programs (MILPs) are extensions of linear programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/linear/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Almost all solvers support linear programs; look for \"LP\" in the list of Supported solvers. However, fewer solvers support mixed-integer linear programs. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/linear/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nThe diet problem\nThe cannery problem\nThe facility location problem\nFinancial modeling problems\nNetwork flow problems\nN-Queens\nSudoku\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling linear programs. Look here if you are stuck trying to formulate a problem as a linear program.\nThe Sensitivity analysis of a linear program tutorial explains how to create sensitivity reports like those produced by the Excel Solver.\nThe Callbacks tutorial explains how to write a variety of solver-independent callbacks. Look here if you want to write a callback.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/#nonlinear_tips_and_tricks","page":"Tips and tricks","title":"Tips and tricks","text":"","category":"section"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This example collates some tips and tricks you can use when formulating nonlinear programs. It uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"using JuMP\nimport Ipopt\nimport Test","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/#User-defined-operators-with-vector-outputs","page":"Tips and tricks","title":"User-defined operators with vector outputs","text":"","category":"section"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A common situation is to have a user-defined operator like the following that returns multiple outputs (we define function_calls to keep track of how many times we call this method):","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"function_calls = 0\nfunction foo(x, y)\n    global function_calls += 1\n    common_term = x^2 + y^2\n    term_1 = sqrt(1 + common_term)\n    term_2 = common_term\n    return term_1, term_2\nend","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"For example, the first term might be used in the objective, and the second term might be used in a constraint, and often they share work that is expensive to evaluate.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This is a problem for JuMP, because it requires user-defined operators to return a single number. One option is to define two separate functions, the first returning the first argument, and the second returning the second argument.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"foo_1(x, y) = foo(x, y)[1]\nfoo_2(x, y) = foo(x, y)[2]","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"However, if the common term is expensive to compute, this approach is wasteful because it will evaluate the expensive term twice. Let's have a look at how many times we evaluate x^2 + y^2 during a solve:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:2] >= 0, start = 0.1)\n@operator(model, op_foo_1, 2, foo_1)\n@operator(model, op_foo_2, 2, foo_2)\n@objective(model, Max, op_foo_1(x[1], x[2]))\n@constraint(model, op_foo_2(x[1], x[2]) <= 2)\nfunction_calls = 0\noptimize!(model)\nTest.@test objective_value(model) ≈ √3 atol = 1e-4\nTest.@test value.(x) ≈ [1.0, 1.0] atol = 1e-4\nprintln(\"Naive approach: function calls = $(function_calls)\")","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"An alternative approach is to use memoization, which uses a cache to store the result of function evaluations. We can write a memoization function as follows:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"\"\"\"\n    memoize(foo::Function, n_outputs::Int)\n\nTake a function `foo` and return a vector of length `n_outputs`, where element\n`i` is a function that returns the equivalent of `foo(x...)[i]`.\n\nTo avoid duplication of work, cache the most-recent evaluations of `foo`.\nBecause `foo_i` is auto-differentiated with ForwardDiff, our cache needs to\nwork when `x` is a `Float64` and a `ForwardDiff.Dual`.\n\"\"\"\nfunction memoize(foo::Function, n_outputs::Int)\n    last_x, last_f = nothing, nothing\n    last_dx, last_dfdx = nothing, nothing\n    function foo_i(i, x::T...) where {T<:Real}\n        if T == Float64\n            if x !== last_x\n                last_x, last_f = x, foo(x...)\n            end\n            return last_f[i]::T\n        else\n            if x !== last_dx\n                last_dx, last_dfdx = x, foo(x...)\n            end\n            return last_dfdx[i]::T\n        end\n    end\n    return [(x...) -> foo_i(i, x...) for i in 1:n_outputs]\nend","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Let's see how it works. First, construct the memoized versions of foo:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"memoized_foo = memoize(foo, 2)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Now try evaluating the first element of memoized_foo.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"function_calls = 0\nmemoized_foo[1](1.0, 1.0)\nprintln(\"function_calls = \", function_calls)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"As expected, this evaluated the function once. However, if we call the function again, we hit the cache instead of needing to re-compute foo and function_calls is still 1!","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"memoized_foo[1](1.0, 1.0)\nprintln(\"function_calls = \", function_calls)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Now let's see how this works during a real solve:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:2] >= 0, start = 0.1)\n@operator(model, op_foo_1, 2, memoized_foo[1])\n@operator(model, op_foo_2, 2, memoized_foo[2])\n@objective(model, Max, op_foo_1(x[1], x[2]))\n@constraint(model, op_foo_2(x[1], x[2]) <= 2)\nfunction_calls = 0\noptimize!(model)\nTest.@test objective_value(model) ≈ √3 atol = 1e-4\nTest.@test value.(x) ≈ [1.0, 1.0] atol = 1e-4\nprintln(\"Memoized approach: function_calls = $(function_calls)\")","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Compared to the naive approach, the memoized approach requires half as many function evaluations.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"EditURL = \"rocket_control.jl\"","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Rocket-Control","page":"Rocket Control","title":"Rocket Control","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial was originally contributed by Iain Dunning.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The purpose of this tutorial is to demonstrate how to setup and solve a nonlinear optimization problem.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The example is an optimal control problem of a nonlinear rocket.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"using JuMP\nimport Ipopt\nimport Plots","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Overview","page":"Rocket Control","title":"Overview","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Our goal is to maximize the final altitude of a vertically launched rocket.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We can control the thrust of the rocket, and must take account of the rocket mass, fuel consumption rate, gravity, and aerodynamic drag.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Let us consider the basic description of the model (for the full description, including parameters for the rocket, see COPS3).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"There are three state variables in our model:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Velocity: x_v(t)\nAltitude: x_h(t)\nMass of rocket and remaining fuel, x_m(t)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"and a single control variable:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Thrust: u_t(t).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"There are three equations that control the dynamics of the rocket:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Rate of ascent: fracd x_hdt = x_v\nAcceleration: fracd x_vdt = fracu_t - D(x_h x_v)x_m - g(x_h)\nRate of mass loss: fracd x_mdt = -fracu_tc","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"where drag D(x_h x_v) is a function of altitude and velocity, gravity g(x_h) is a function of altitude, and c is a constant.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"These forces are defined as:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"D(x_h x_v) = D_c cdot x_v^2 cdot e^-h_c left( fracx_h-x_h(0)x_h(0) right)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"and g(x_h) = g_0 cdot left( fracx_h(0)x_h right)^2","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We use a discretized model of time, with a fixed number of time steps, T.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Our goal is thus to maximize x_h(T).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Data","page":"Rocket Control","title":"Data","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"All parameters in this model have been normalized to be dimensionless, and they are taken from COPS3.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"h_0 = 1                      # Initial height\nv_0 = 0                      # Initial velocity\nm_0 = 1.0                    # Initial mass\nm_T = 0.6                    # Final mass\ng_0 = 1                      # Gravity at the surface\nh_c = 500                    # Used for drag\nc = 0.5 * sqrt(g_0 * h_0)    # Thrust-to-fuel mass\nD_c = 0.5 * 620 * m_0 / g_0  # Drag scaling\nu_t_max = 3.5 * g_0 * m_0    # Maximum thrust\nT_max = 0.2                  # Number of seconds\nT = 1_000                    # Number of time steps\nΔt = 0.2 / T;                # Time per discretized step\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#JuMP-formulation","page":"Rocket Control","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"First, we create a model and choose an optimizer. Since this is a nonlinear program, we need to use a nonlinear solver like Ipopt. We cannot use a linear solver like HiGHS.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Next, we create our state and control variables, which are each indexed by t. It is good practice for nonlinear programs to always provide a starting solution for each variable.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"@variable(model, x_v[1:T] >= 0, start = v_0)           # Velocity\n@variable(model, x_h[1:T] >= 0, start = h_0)           # Height\n@variable(model, x_m[1:T] >= m_T, start = m_0)         # Mass\n@variable(model, 0 <= u_t[1:T] <= u_t_max, start = 0); # Thrust\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We implement boundary conditions by fixing variables to values.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"fix(x_v[1], v_0; force = true)\nfix(x_h[1], h_0; force = true)\nfix(x_m[1], m_0; force = true)\nfix(u_t[T], 0.0; force = true)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The objective is to maximize altitude at end of time of flight.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"@objective(model, Max, x_h[T])","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Forces are defined as functions:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"D(x_h, x_v) = D_c * x_v^2 * exp(-h_c * (x_h - h_0) / h_0)\ng(x_h) = g_0 * (h_0 / x_h)^2","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The dynamical equations are implemented as constraints.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"ddt(x::Vector, t::Int) = (x[t] - x[t-1]) / Δt\n@constraint(model, [t in 2:T], ddt(x_h, t) == x_v[t-1])\n@constraint(\n    model,\n    [t in 2:T],\n    ddt(x_v, t) == (u_t[t-1] - D(x_h[t-1], x_v[t-1])) / x_m[t-1] - g(x_h[t-1]),\n)\n@constraint(model, [t in 2:T], ddt(x_m, t) == -u_t[t-1] / c);\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Now we optimize the model and check that we found a solution:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Finally, we plot the solution:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"function plot_trajectory(y; kwargs...)\n    return Plots.plot(\n        (1:T) * Δt,\n        value.(y);\n        xlabel = \"Time (s)\",\n        legend = false,\n        kwargs...,\n    )\nend\n\nPlots.plot(\n    plot_trajectory(x_h; ylabel = \"Altitude\"),\n    plot_trajectory(x_m; ylabel = \"Mass\"),\n    plot_trajectory(x_v; ylabel = \"Velocity\"),\n    plot_trajectory(u_t; ylabel = \"Thrust\");\n    layout = (2, 2),\n)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Next-steps","page":"Rocket Control","title":"Next steps","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Experiment with different values for the constants. How does the solution change? In particular, what happens if you change T_max?\nThe dynamical equations use rectangular integration for the right-hand side terms. Modify the equations to use the Trapezoidal rule instead. (As an example, x_v[t-1] would become 0.5 * (x_v[t-1] + x_v[t]).) Is there a difference?","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"EditURL = \"querying_hessians.jl\"","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#Computing-Hessians","page":"Computing Hessians","title":"Computing Hessians","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"The purpose of this tutorial is to demonstrate how to compute the Hessian of the Lagrangian of a nonlinear program.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"warning: Warning\nThis is an advanced tutorial that interacts with the low-level nonlinear interface of MathOptInterface.By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Given a nonlinear program:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"beginalign\n min_x in mathbbR^n  f(x) \n textst  l le g_i(x) le u\nendalign","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"the Hessian of the Lagrangian is computed as:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"H(x sigma mu) = sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"where x is a primal point, sigma is a scalar (typically 1), and mu is a vector of weights corresponding to the Lagrangian dual of the constraints.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"using JuMP\nimport Ipopt\nimport LinearAlgebra\nimport Random\nimport SparseArrays","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#The-basic-model","page":"Computing Hessians","title":"The basic model","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"To demonstrate how to interact with the lower-level nonlinear interface, we need an example model. The exact model isn't important; we use the model from The Rosenbrock function tutorial, with some additional constraints to demonstrate various features of the lower-level interface.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:2], start = -i)\n@constraint(model, g_1, x[1]^2 <= 1)\n@constraint(model, g_2, (x[1] + x[2])^2 <= 2)\n@objective(model, Min, (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2)\noptimize!(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#The-analytic-solution","page":"Computing Hessians","title":"The analytic solution","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"With a little work, it is possible to analytically derive the correct hessian:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function analytic_hessian(x, σ, μ)\n    g_1_H = [2.0 0.0; 0.0 0.0]\n    g_2_H = [2.0 2.0; 2.0 2.0]\n    f_H = zeros(2, 2)\n    f_H[1, 1] = 2.0 + 1200.0 * x[1]^2 - 400.0 * x[2]\n    f_H[1, 2] = f_H[2, 1] = -400.0 * x[1]\n    f_H[2, 2] = 200.0\n    return σ * f_H + μ' * [g_1_H, g_2_H]\nend","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Here are various points:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [0, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [1, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [0, 1])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 1, [0, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#Create-a-nonlinear-model","page":"Computing Hessians","title":"Create a nonlinear model","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"JuMP delegates automatic differentiation to the MOI.Nonlinear submodule. Therefore, to compute the Hessian of the Lagrangian, we need to create a MOI.Nonlinear.Model object:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"rows = Any[]\nnlp = MOI.Nonlinear.Model()\nfor (F, S) in list_of_constraint_types(model)\n    if F <: VariableRef\n        continue  # Skip variable bounds\n    end\n    for ci in all_constraints(model, F, S)\n        push!(rows, ci)\n        object = constraint_object(ci)\n        MOI.Nonlinear.add_constraint(nlp, object.func, object.set)\n    end\nend\nMOI.Nonlinear.set_objective(nlp, objective_function(model))\nnlp","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"It is important that we save the constraint indices in a vector rows, so that we know the order of the constraints in the nonlinear model.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Next, we need to convert our model into an MOI.Nonlinear.Evaluator, specifying an automatic differentiation backend. In this case, we use MOI.Nonlinear.SparseReverseMode:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"evaluator = MOI.Nonlinear.Evaluator(\n    nlp,\n    MOI.Nonlinear.SparseReverseMode(),\n    index.(all_variables(model)),\n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Before computing anything with the evaluator, we need to initialize it. Use MOI.features_available to see what we can query:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.features_available(evaluator)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Consult the MOI documentation for specifics, but to obtain the Hessian matrix, we need to initialize :Hess:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.initialize(evaluator, [:Hess])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI represents the Hessian as a sparse matrix. Get the sparsity pattern as follows:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"hessian_sparsity = MOI.hessian_lagrangian_structure(evaluator)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"The sparsity pattern has a few properties of interest:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Each element (i, j) indicates a structural non-zero in row i and column j\nThe list may contain duplicates, in which case we should add the values together\nThe list does not need to be sorted\nThe list may contain any mix of lower- or upper-triangular indices","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This format matches Julia's sparse-triplet form of a SparseArray, so we can convert from the sparse Hessian representation to a Julia SparseArray as follows:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"I = [i for (i, _) in hessian_sparsity]\nJ = [j for (_, j) in hessian_sparsity]\nV = zeros(length(hessian_sparsity))\nn = num_variables(model)\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Of course, knowing where the zeros are isn't very interesting. We really want to compute the value of the Hessian matrix at a point.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.eval_hessian_lagrangian(evaluator, V, ones(n), 1.0, ones(length(rows)))\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"In practice, we often want to compute the value of the hessian at the optimal solution.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"First, we compute the primal solution. To do so, we need a vector of the variables in the order that they were passed to the solver:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"x = all_variables(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Here x[1] is the variable that corresponds to column 1, and so on. Here's the optimal primal solution:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"x_optimal = value.(x)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Next, we need the optimal dual solution associated with the nonlinear constraints (this is where it is important to record the order of the constraints as we added them to nlp):","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"y_optimal = dual.(rows)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Now we can compute the Hessian at the optimal primal-dual point:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.eval_hessian_lagrangian(evaluator, V, x_optimal, 1.0, y_optimal)\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"However, this Hessian isn't quite right because it isn't symmetric. We can fix this by filling in the appropriate off-diagonal terms:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function fill_off_diagonal(H)\n    ret = H + H'\n    row_vals = SparseArrays.rowvals(ret)\n    non_zeros = SparseArrays.nonzeros(ret)\n    for col in 1:size(ret, 2)\n        for i in SparseArrays.nzrange(ret, col)\n            if col == row_vals[i]\n                non_zeros[i] /= 2\n            end\n        end\n    end\n    return ret\nend\n\nfill_off_diagonal(H)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Putting everything together:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function compute_optimal_hessian(model::Model)\n    rows = Any[]\n    nlp = MOI.Nonlinear.Model()\n    for (F, S) in list_of_constraint_types(model)\n        for ci in all_constraints(model, F, S)\n            push!(rows, ci)\n            object = constraint_object(ci)\n            MOI.Nonlinear.add_constraint(nlp, object.func, object.set)\n        end\n    end\n    MOI.Nonlinear.set_objective(nlp, objective_function(model))\n    x = all_variables(model)\n    backend = MOI.Nonlinear.SparseReverseMode()\n    evaluator = MOI.Nonlinear.Evaluator(nlp, backend, index.(x))\n    MOI.initialize(evaluator, [:Hess])\n    hessian_sparsity = MOI.hessian_lagrangian_structure(evaluator)\n    I = [i for (i, _) in hessian_sparsity]\n    J = [j for (_, j) in hessian_sparsity]\n    V = zeros(length(hessian_sparsity))\n    MOI.eval_hessian_lagrangian(evaluator, V, value.(x), 1.0, dual.(rows))\n    H = SparseArrays.sparse(I, J, V, length(x), length(x))\n    return Matrix(fill_off_diagonal(H))\nend\n\nH_star = compute_optimal_hessian(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"If we compare our solution against the analytical solution:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian(value.(x), 1.0, dual.([g_1, g_2]))","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"If we look at the eigenvalues of the Hessian:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"LinearAlgebra.eigvals(H_star)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"we see that they are all positive. Therefore, the Hessian is positive definite, and so the solution found by Ipopt is a local minimizer.","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"EditURL = \"https://github.com/jump-dev/SumOfSquares.jl/blob/v0.7.3/README.md\"","category":"page"},{"location":"packages/SumOfSquares/#SumOfSquares.jl","page":"jump-dev/SumOfSquares.jl","title":"SumOfSquares.jl","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"(Image: Build Status) (Image: codecov) (Image: )","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"SumOfSquares.jl is a JuMP extension that, when used in conjunction with MultivariatePolynomial and PolyJuMP, implements a sum of squares reformulation for polynomial optimization.","category":"page"},{"location":"packages/SumOfSquares/#License","page":"jump-dev/SumOfSquares.jl","title":"License","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"SumOfSquares.jl is licensed under the MIT license.","category":"page"},{"location":"packages/SumOfSquares/#Installation","page":"jump-dev/SumOfSquares.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Install SumOfSquares using Pkg.add:","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"import Pkg\nPkg.add(\"SumOfSquares\")","category":"page"},{"location":"packages/SumOfSquares/#Documentation","page":"jump-dev/SumOfSquares.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See https://jump.dev/SumOfSquares.jl/stable for the most recently tagged version of the documentation.","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See https://jump.dev/SumOfSquares.jl/dev for the in-development version of the documentation.","category":"page"},{"location":"packages/SumOfSquares/#Presentations","page":"jump-dev/SumOfSquares.jl","title":"Presentations","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Some presentations on, or using, SumOfSquares (see blegat/SumOfSquaresSlides for the source code of the presentations):","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Benoît Legat at JuMP-dev 2023 [Slides]\nBenoît Legat, Marek Kaluba and Tillmann Weisser at INFORMS 2022 [Slides]\nBenoît Legat at POEMA Learning Week 2\nBenoît Legat, Marek Kaluba and Tillmann Weisser at JuMP-dev 2021 [Slides]\nBenoît Legat at INFORMS 2020 [Slides]\nTillmann Weisser, Benoît Legat, Chris Coey, Lea Kapelevich and Juan Pablo Vielma at JuliaCon 2019 [Slides] [Video]\nBenoît Legat at CNLS 2019 [Slides]\nBenoît Legat at EURO 2019 [Slides]\nBenoît Legat at juliaday Nantes 2019 [Slides]\nBenoît Legat at Summer School on Numerical Computing in Algebraic Geometry 2018 [Poster]\nBenoît Legat at The First Annual JuMP-dev Workshop 2017 [Slides] [Video]\nJoey Huchette at SIAM Opt 2017","category":"page"},{"location":"packages/SumOfSquares/#Citing","page":"jump-dev/SumOfSquares.jl","title":"Citing","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See CITATION.bib.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"EditURL = \"cutting_stock_column_generation.jl\"","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Column-generation","page":"Column generation","title":"Column generation","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The purpose of this tutorial is to demonstrate the column generation algorithm. As an example, it solves the Cutting stock problem.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport Plots\nimport SparseArrays","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Background","page":"Column generation","title":"Background","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The cutting stock problem is about cutting large rolls of paper into smaller pieces.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We denote the set of possible sized pieces that a roll can be cut into by iin 1ldotsI. Each piece i has a width, w_i, and a demand, d_i. The width of the large roll is W.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our objective is to minimize the number of rolls needed to meet all demand.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Here's the data that we are going to use in this tutorial:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"struct Piece\n    w::Float64\n    d::Int\nend\n\nstruct Data\n    pieces::Vector{Piece}\n    W::Float64\nend\n\nfunction Base.show(io::IO, d::Data)\n    println(io, \"Data for the cutting stock problem:\")\n    println(io, \"  W = $(d.W)\")\n    println(io, \"with pieces:\")\n    println(io, \"   i   w_i d_i\")\n    println(io, \"  ------------\")\n    for (i, p) in enumerate(d.pieces)\n        println(io, lpad(i, 4), \" \", lpad(p.w, 5), \" \", lpad(p.d, 3))\n    end\n    return\nend\n\nfunction get_data()\n    data = [\n        75.0 38\n        75.0 44\n        75.0 30\n        75.0 41\n        75.0 36\n        53.8 33\n        53.0 36\n        51.0 41\n        50.2 35\n        32.2 37\n        30.8 44\n        29.8 49\n        20.1 37\n        16.2 36\n        14.5 42\n        11.0 33\n        8.6 47\n        8.2 35\n        6.6 49\n        5.1 42\n    ]\n    return Data([Piece(data[i, 1], data[i, 2]) for i in axes(data, 1)], 100.0)\nend\n\ndata = get_data()","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Mathematical-formulation","page":"Column generation","title":"Mathematical formulation","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"To formulate the cutting stock problem as a mixed-integer linear program, we assume that there is a set of large rolls j=1ldotsJ to use. Then, we introduce two classes of decision variables:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_ij ge 0 textinteger  forall i=1ldotsI j=1ldotsJ\ny_j in 0 1 forall j=1ldotsJ","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"y_j is a binary variable that indicates if we use roll j, and x_ij counts how many pieces of size i that we cut from roll j.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our mixed-integer linear program is therefore:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    min  sumlimits_j=1^J y_j \n    textst  sumlimits_i=1^N w_i x_ij le W y_j  forall j=1ldotsJ \n          sumlimits_j=1^J x_ij ge d_i  forall i=1ldotsI \n          x_ij ge 0  forall i=1ldotsN j=1ldotsJ \n          x_ij in mathbbZ  forall i=1ldotsI j=1ldotsJ \n          y_j in 0 1  forall j=1ldotsJ \nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The objective is to minimize the number of rolls that we use, and the two constraints ensure that we respect the total width of each large roll and that we satisfy demand exactly.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The JuMP formulation of this model is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"I = length(data.pieces)\nJ = 1_000  # Some large number\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:I, 1:J] >= 0, Int)\n@variable(model, y[1:J], Bin)\n@objective(model, Min, sum(y))\n@constraint(model, [i in 1:I], sum(x[i, :]) >= data.pieces[i].d)\n@constraint(\n    model,\n    [j in 1:J],\n    sum(data.pieces[i].w * x[i, j] for i in 1:I) <= data.W * y[j],\n);\nnothing #hide","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Unfortunately, we can't solve this formulation for realistic instances because it takes a very long time to solve. (Try removing the time limit.)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"set_time_limit_sec(model, 5.0)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"However, there is a formulation that solves much faster, and that is to use a column generation scheme.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Column-generation-theory","page":"Column generation","title":"Column generation theory","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The key insight for column generation is to recognize that feasible columns in the x matrix of variables encode cutting patterns.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For example, if we look only at the roll j=1, then a feasible solution is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_11 = 1 (1 unit of piece #1)\nx_131 = 1 (1 unit of piece #13)\nAll other x_i1 = 0","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Another solution is","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_201 = 19 (19 unit of piece #20)\nAll other x_i1 = 0","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Cutting patterns like x_11 = 1 and x_21 = 1 are infeasible because the combined length is greater than W.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Since there are a finite number of ways that we could cut a roll into a valid cutting pattern, we could create a set of all possible cutting patterns p = 1ldotsP, with data a_ip indicating how many units of piece i we cut in pattern p. Then, we can formulate our mixed-integer linear program as:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    min  sumlimits_p=1^P x_p \n    textst  sumlimits_p=1^P a_ip x_p ge d_i  forall i=1ldotsI \n          x_p ge 0  forall p=1ldotsP \n          x_p in mathbbZ  forall p=1ldotsP\nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Unfortunately, there will be a very large number of these patterns, so it is often intractable to enumerate all columns p=1ldotsP.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Column generation is an iterative algorithm that starts with a small set of initial patterns, and then cleverly chooses new columns to add to the main MILP so that we find the optimal solution without having to enumerate every column.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Choosing-the-initial-set-of-patterns","page":"Column generation","title":"Choosing the initial set of patterns","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For the initial set of patterns, we create a trivial cutting pattern which cuts as many units of piece i as will fit.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"patterns = map(1:I) do i\n    n_pieces = floor(Int, data.W / data.pieces[i].w)\n    return SparseArrays.sparsevec([i], [n_pieces], I)\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We can visualize the patterns as follows:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"\"\"\"\n    cutting_locations(data::Data, pattern::SparseArrays.SparseVector)\n\nA function which returns a vector of the locations along the roll at which to\ncut in order to produce pattern `pattern`.\n\"\"\"\nfunction cutting_locations(data::Data, pattern::SparseArrays.SparseVector)\n    locations = Float64[]\n    offset = 0.0\n    for (i, c) in zip(SparseArrays.findnz(pattern)...)\n        for _ in 1:c\n            offset += data.pieces[i].w\n            push!(locations, offset)\n        end\n    end\n    return locations\nend\n\nfunction plot_patterns(data::Data, patterns)\n    plot = Plots.bar(;\n        xlims = (0, length(patterns) + 1),\n        ylims = (0, data.W),\n        xlabel = \"Pattern\",\n        ylabel = \"Roll length\",\n    )\n    for (i, p) in enumerate(patterns)\n        locations = cutting_locations(data, p)\n        Plots.bar!(\n            plot,\n            fill(i, length(locations)),\n            reverse(locations);\n            bar_width = 0.6,\n            label = false,\n            color = \"#90caf9\",\n        )\n    end\n    return plot\nend\n\nplot_patterns(data, patterns)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#The-base-problem","page":"Column generation","title":"The base problem","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Using the initial set of patterns, we can create and optimize our base model:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:length(patterns)] >= 0, Int)\n@objective(model, Min, sum(x))\n@constraint(model, demand[i in 1:I], patterns[i]' * x >= data.pieces[i].d)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This solution requires 421 rolls. This solution is sub-optimal because the model does not contain the full set of possible patterns.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"How do we find a new column that leads to an improved solution?","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Choosing-new-columns","page":"Column generation","title":"Choosing new columns","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Column generation chooses a new column by relaxing the integrality constraint on x and looking at the dual variable pi_i associated with demand constraint i.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For example, the dual of demand[13] is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"unset_integer.(x)\noptimize!(model)\nπ_13 = dual(demand[13])","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Using the economic interpretation of the dual variable, we can say that a one unit increase in demand for piece i will cost an extra pi_i rolls. Alternatively, we can say that a one unit increase in the left-hand side (for example, due to a new cutting pattern) will save us pi_i rolls. Therefore, we want a new column that maximizes the savings associated with the dual variables, while respecting the total width of the roll:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    max  sumlimits_i=1^I pi_i y_i \n    textst  sumlimits_i=1^I w_i y_i le W \n          y_i ge 0  forall i=1ldotsI \n          y_i in mathbbZ  forall i=1ldotsI \nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If this problem, called the pricing problem, has an objective value greater than 1, then we estimate than adding y as the coefficients of a new column will decrease the objective by more than the cost of an extra roll.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Here is code to solve the pricing problem:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"function solve_pricing(data::Data, π::Vector{Float64})\n    I = length(π)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, y[1:I] >= 0, Int)\n    @constraint(model, sum(data.pieces[i].w * y[i] for i in 1:I) <= data.W)\n    @objective(model, Max, sum(π[i] * y[i] for i in 1:I))\n    optimize!(model)\n    number_of_rolls_saved = objective_value(model)\n    if number_of_rolls_saved > 1 + 1e-8\n        # Benefit of pattern is more than the cost of a new roll plus some\n        # tolerance\n        return SparseArrays.sparse(round.(Int, value.(y)))\n    end\n    return nothing\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If we solve the pricing problem with an artificial dual vector:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solve_pricing(data, [1.0 / i for i in 1:I])","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"the solution is a roll with 1 unit of piece #1, 1 unit of piece #17, and 3 units of piece #20.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If we solve the pricing problem with a dual vector of zeros, then the benefit of the new pattern is less than the cost of a roll, and so the function returns nothing:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solve_pricing(data, zeros(I))","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Iterative-algorithm","page":"Column generation","title":"Iterative algorithm","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Now we can combine our base model with the pricing subproblem in an iterative column generation scheme:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"while true\n    # Solve the linear relaxation\n    optimize!(model)\n    # Obtain a new dual vector\n    π = dual.(demand)\n    # Solve the pricing problem\n    new_pattern = solve_pricing(data, π)\n    # Stop iterating if there is no new pattern\n    if new_pattern === nothing\n        @info \"No new patterns, terminating the algorithm.\"\n        break\n    end\n    push!(patterns, new_pattern)\n    # Create a new column\n    push!(x, @variable(model, lower_bound = 0))\n    # Update the objective coefficient of the new column\n    set_objective_coefficient(model, x[end], 1.0)\n    # Update the non-zeros in the coefficient matrix\n    for (i, count) in zip(SparseArrays.findnz(new_pattern)...)\n        set_normalized_coefficient(demand[i], x[end], count)\n    end\n    println(\"Found new pattern. Total patterns = $(length(patterns))\")\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We found lots of new patterns. Here's pattern 21:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"patterns[21]","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Let's have a look at the patterns now:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"plot_patterns(data, patterns)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Looking-at-the-solution","page":"Column generation","title":"Looking at the solution","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Let's see how many of each column we need:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solution = DataFrames.DataFrame([\n    (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)\n])\nfilter!(row -> row.rolls > 0, solution)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Since we solved a linear program, some of our columns have fractional solutions. We can create a integer feasible solution by rounding up the orders. This requires 341 rolls:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"sum(ceil.(Int, solution.rolls))","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Alternatively, we can re-introduce the integrality constraints and resolve the problem:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"set_integer.(x)\noptimize!(model)\nsolution = DataFrames.DataFrame([\n    (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)\n])\nfilter!(row -> row.rolls > 0, solution)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This now requires 334 rolls:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"sum(solution.rolls)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Note that this may not be the global minimum because we are not adding new columns during the solution of the mixed-integer problem model (an algorithm known as branch and price). Nevertheless, the column generation algorithm typically finds good integer feasible solutions to an otherwise intractable optimization problem.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Next-steps","page":"Column generation","title":"Next steps","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our objective function is to minimize the total number of rolls. What is the total length of waste? How does that compare to the total demand?\nWriting the optimization algorithm is only part of the challenge. Can you develop a better way to communicate the solution to stakeholders?","category":"page"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/constraints.md\"","category":"page"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/constraints/#constraints_ref","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"moi/reference/constraints/#Types","page":"Constraints","title":"Types","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"ConstraintIndex","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintIndex","page":"Constraints","title":"MathOptInterface.ConstraintIndex","text":"ConstraintIndex{F, S}\n\nA type-safe wrapper for Int64 for use in referencing F-in-S constraints in a model. The parameter F is the type of the function in the constraint, and the parameter S is the type of set in the constraint. To allow for deletion, indices need not be consecutive. Indices within a constraint type (i.e. F-in-S) must be unique, but non-unique indices across different constraint types are allowed. If F is VariableIndex then the index is equal to the index of the variable. That is for an index::ConstraintIndex{VariableIndex}, we always have\n\nindex.value == MOI.get(model, MOI.ConstraintFunction(), index).value\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#Functions","page":"Constraints","title":"Functions","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"is_valid(::ModelLike,::ConstraintIndex)\nadd_constraint\nadd_constraints\ntransform\nsupports_constraint","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.ConstraintIndex}","page":"Constraints","title":"MathOptInterface.is_valid","text":"is_valid(model::ModelLike, index::Index)::Bool\n\nReturn a Bool indicating whether this index refers to a valid object in the model model.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/constraints/#MathOptInterface.add_constraint","page":"Constraints","title":"MathOptInterface.add_constraint","text":"add_constraint(model::ModelLike, func::F, set::S)::ConstraintIndex{F,S} where {F,S}\n\nAdd the constraint f(x) in mathcalS where f is defined by func, and mathcalS is defined by set.\n\nadd_constraint(model::ModelLike, v::VariableIndex, set::S)::ConstraintIndex{VariableIndex,S} where {S}\nadd_constraint(model::ModelLike, vec::Vector{VariableIndex}, set::S)::ConstraintIndex{VectorOfVariables,S} where {S}\n\nAdd the constraint v in mathcalS where v is the variable (or vector of variables) referenced by v and mathcalS is defined by set.\n\nAn UnsupportedConstraint error is thrown if model does not support F-in-S constraints,\na AddConstraintNotAllowed error is thrown if it supports F-in-S constraints but it cannot add the constraint(s) in its current state and\na ScalarFunctionConstantNotZero error may be thrown if func is an AbstractScalarFunction with nonzero constant and set is EqualTo, GreaterThan, LessThan or Interval.\na LowerBoundAlreadySet error is thrown if F is a VariableIndex and a constraint was already added to this variable that sets a lower bound.\na UpperBoundAlreadySet error is thrown if F is a VariableIndex and a constraint was already added to this variable that sets an upper bound.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.add_constraints","page":"Constraints","title":"MathOptInterface.add_constraints","text":"add_constraints(model::ModelLike, funcs::Vector{F}, sets::Vector{S})::Vector{ConstraintIndex{F,S}} where {F,S}\n\nAdd the set of constraints specified by each function-set pair in funcs and sets. F and S should be concrete types. This call is equivalent to add_constraint.(model, funcs, sets) but may be more efficient.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.transform","page":"Constraints","title":"MathOptInterface.transform","text":"Transform Constraint Set\n\ntransform(model::ModelLike, c::ConstraintIndex{F,S1}, newset::S2)::ConstraintIndex{F,S2}\n\nReplace the set in constraint c with newset. The constraint index c will no longer be valid, and the function returns a new constraint index with the correct type.\n\nSolvers may only support a subset of constraint transforms that they perform efficiently (for example, changing from a LessThan to GreaterThan set). In addition, set modification (where S1 = S2) should be performed via the modify function.\n\nTypically, the user should delete the constraint and add a new one.\n\nExamples\n\nIf c is a ConstraintIndex{ScalarAffineFunction{Float64},LessThan{Float64}},\n\nc2 = transform(model, c, GreaterThan(0.0))\ntransform(model, c, LessThan(0.0)) # errors\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.supports_constraint","page":"Constraints","title":"MathOptInterface.supports_constraint","text":"supports_constraint(\n    model::ModelLike,\n    ::Type{F},\n    ::Type{S},\n)::Bool where {F<:AbstractFunction,S<:AbstractSet}\n\nReturn a Bool indicating whether model supports F-in-S constraints, that is, copy_to(model, src) does not throw UnsupportedConstraint when src contains F-in-S constraints. If F-in-S constraints are only not supported in specific circumstances, e.g. F-in-S constraints cannot be combined with another type of constraint, it should still return true.\n\n\n\n\n\nMOI.supports_constraint(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging F-in-S constraints.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method for constraint types that the bridge implements.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#Attributes","page":"Constraints","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"AbstractConstraintAttribute\nConstraintName\nConstraintPrimalStart\nConstraintDualStart\nConstraintPrimal\nConstraintDual\nConstraintBasisStatus\nConstraintFunction\nCanonicalConstraintFunction\nConstraintSet\nBasisStatusCode\nBASIC\nNONBASIC\nNONBASIC_AT_LOWER\nNONBASIC_AT_UPPER\nSUPER_BASIC","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.AbstractConstraintAttribute","page":"Constraints","title":"MathOptInterface.AbstractConstraintAttribute","text":"AbstractConstraintAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of constraints in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintName","page":"Constraints","title":"MathOptInterface.ConstraintName","text":"ConstraintName()\n\nA constraint attribute for a string identifying the constraint.\n\nIt is valid for constraints variables to have the same name; however, constraints with duplicate names cannot be looked up using get, regardless of whether they have the same F-in-S type.\n\nConstraintName has a default value of \"\" if not set.\n\nNotes\n\nYou should not implement ConstraintName for VariableIndex constraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart","page":"Constraints","title":"MathOptInterface.ConstraintPrimalStart","text":"ConstraintPrimalStart()\n\nA constraint attribute for the initial assignment to some constraint's ConstraintPrimal that the optimizer may use to warm-start the solve.\n\nMay be nothing (unset), a number for AbstractScalarFunction, or a vector for AbstractVectorFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintDualStart","page":"Constraints","title":"MathOptInterface.ConstraintDualStart","text":"ConstraintDualStart()\n\nA constraint attribute for the initial assignment to some constraint's ConstraintDual that the optimizer may use to warm-start the solve.\n\nMay be nothing (unset), a number for AbstractScalarFunction, or a vector for AbstractVectorFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintPrimal","page":"Constraints","title":"MathOptInterface.ConstraintPrimal","text":"ConstraintPrimal(result_index::Int = 1)\n\nA constraint attribute for the assignment to some constraint's primal value(s) in result result_index.\n\nIf the constraint is f(x) in S, then in most cases the ConstraintPrimal is the value of f, evaluated at the correspondng VariablePrimal solution.\n\nHowever, some conic solvers reformulate b - Ax in S to s = b - Ax, s in S. These solvers may return the value of s for ConstraintPrimal, rather than b - Ax. (Although these are constrained by an equality constraint, due to numerical tolerances they may not be identical.)\n\nIf the solver does not have a primal value for the constraint because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ConstraintPrimal attribute.\n\nIf result_index is omitted, it is 1 by default. See ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintDual","page":"Constraints","title":"MathOptInterface.ConstraintDual","text":"ConstraintDual(result_index::Int = 1)\n\nA constraint attribute for the assignment to some constraint's dual value(s) in result result_index. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a dual value for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check DualStatus before accessing the ConstraintDual attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintBasisStatus","page":"Constraints","title":"MathOptInterface.ConstraintBasisStatus","text":"ConstraintBasisStatus(result_index::Int = 1)\n\nA constraint attribute for the BasisStatusCode of some constraint in result result_index, with respect to an available optimal solution basis. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a basis statue for the constraint because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ConstraintBasisStatus attribute.\n\nSee ResultCount for information on how the results are ordered.\n\nNotes\n\nFor the basis status of a variable, query VariableBasisStatus.\n\nConstraintBasisStatus does not apply to VariableIndex constraints. You can infer the basis status of a VariableIndex constraint by looking at the result of VariableBasisStatus.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintFunction","page":"Constraints","title":"MathOptInterface.ConstraintFunction","text":"ConstraintFunction()\n\nA constraint attribute for the AbstractFunction object used to define the constraint.\n\nIt is guaranteed to be equivalent but not necessarily identical to the function provided by the user.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.CanonicalConstraintFunction","page":"Constraints","title":"MathOptInterface.CanonicalConstraintFunction","text":"CanonicalConstraintFunction()\n\nA constraint attribute for a canonical representation of the AbstractFunction object used to define the constraint.\n\nGetting this attribute is guaranteed to return a function that is equivalent but not necessarily identical to the function provided by the user.\n\nBy default, MOI.get(model, MOI.CanonicalConstraintFunction(), ci) fallbacks to MOI.Utilities.canonical(MOI.get(model, MOI.ConstraintFunction(), ci)). However, if model knows that the constraint function is canonical then it can implement a specialized method that directly return the function without calling Utilities.canonical. Therefore, the value returned cannot be assumed to be a copy of the function stored in model. Moreover, Utilities.Model checks with Utilities.is_canonical whether the function stored internally is already canonical and if it's the case, then it returns the function stored internally instead of a copy.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintSet","page":"Constraints","title":"MathOptInterface.ConstraintSet","text":"ConstraintSet()\n\nA constraint attribute for the AbstractSet object used to define the constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.BasisStatusCode","page":"Constraints","title":"MathOptInterface.BasisStatusCode","text":"BasisStatusCode\n\nAn Enum of possible values for the ConstraintBasisStatus and VariableBasisStatus attributes, explaining the status of a given element with respect to an optimal solution basis.\n\nNotes\n\nNONBASIC_AT_LOWER and NONBASIC_AT_UPPER should be used only for\n\nconstraints with the Interval set. In this case, they are necessary to distinguish which side of the constraint is active. One-sided constraints (e.g., LessThan and GreaterThan) should use NONBASIC instead of the NONBASIC_AT_* values. This restriction does not apply to VariableBasisStatus, which should return NONBASIC_AT_* regardless of whether the alternative bound exists.\n\nIn linear programs, SUPER_BASIC occurs when a variable with no bounds is not\n\nin the basis.\n\nValues\n\nPossible values are:\n\nBASIC: element is in the basis\nNONBASIC: element is not in the basis\nNONBASIC_AT_LOWER: element is not in the basis and is at its lower bound\nNONBASIC_AT_UPPER: element is not in the basis and is at its upper bound\nSUPER_BASIC: element is not in the basis but is also not at one of its bounds\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.BASIC","page":"Constraints","title":"MathOptInterface.BASIC","text":"BASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nBASIC: element is in the basis\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC","page":"Constraints","title":"MathOptInterface.NONBASIC","text":"NONBASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC: element is not in the basis\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC_AT_LOWER","page":"Constraints","title":"MathOptInterface.NONBASIC_AT_LOWER","text":"NONBASIC_AT_LOWER::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC_AT_LOWER: element is not in the basis and is at its lower bound\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC_AT_UPPER","page":"Constraints","title":"MathOptInterface.NONBASIC_AT_UPPER","text":"NONBASIC_AT_UPPER::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC_AT_UPPER: element is not in the basis and is at its upper bound\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.SUPER_BASIC","page":"Constraints","title":"MathOptInterface.SUPER_BASIC","text":"SUPER_BASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nSUPER_BASIC: element is not in the basis but is also not at one of its bounds\n\n\n\n\n\n","category":"constant"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"EditURL = \"tsp_lazy_constraints.jl\"","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_lazy","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial was originally contributed by Daniel Schermer.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial describes how to implement the Traveling Salesperson Problem in JuMP using solver-independent lazy constraints that dynamically separate subtours. To be more precise, we use lazy constraints to cut off infeasible subtours only when necessary and not before needed.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"It uses the following packages:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"using JuMP\nimport GLPK\nimport Random\nimport Plots","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_model","page":"Traveling Salesperson Problem","title":"Mathematical Formulation","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Assume that we are given a complete graph mathcalG(VE) where V is the set of vertices (or cities) and E is the set of edges (or roads). For each pair of vertices i j in V i neq j the edge (ij) in E is associated with a weight (or distance) d_ij in mathbbR^+.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"For this tutorial, we assume the problem to be symmetric, that is, d_ij = d_ji  forall ij in V.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"In the Traveling Salesperson Problem, we are tasked with finding a tour with minimal length that visits every vertex exactly once and then returns to the point of origin, that is, a Hamiltonian cycle with minimal weight.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"To model the problem, we introduce a binary variable, x_ij in 01  forall i j in V, that indicates if edge (ij) is part of the tour or not. Using these variables, the Traveling Salesperson Problem can be modeled as the following integer linear program.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_objective","page":"Traveling Salesperson Problem","title":"Objective Function","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The objective is to minimize the length of the tour (due to the assumed symmetry, the second sum only contains ij):","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"textmin  sum_i in V  sum_j in V i  j d_ij x_ij","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Note that it is also possible to use the following objective function instead:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"textmin  sum_i in V  sum_j in V dfracd_ij x_ij2","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_constraints","page":"Traveling Salesperson Problem","title":"Constraints","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"There are four classes of constraints in our formulation.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"First, due to the presumed symmetry, the following constraints must hold:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"x_ij = x_ji quad forall ij in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Second, for each vertex i, exactly two edges must be selected that connect it to other vertices j in the graph G:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"sum_j in V x_ij = 2 quad forall i in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Third, we do not permit loops to occur:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"x_ii = 0 quad forall i in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The fourth constraint is more complicated. A major difficulty of the Traveling Salesperson Problem arises from the fact that we need to prevent subtours, that is, several distinct Hamiltonian cycles existing on subgraphs of G.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Note that the previous constraints do not guarantee that the solution will be free of subtours. To this end, by S we label a subset of vertices. Then, for each proper subset S subset V, the following constraints guarantee that no subtour may occur:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"sum_i in S sum_j in S i  j x_ij leq vert S vert - 1 quad forall S subset V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Problematically, we require exponentially many of these constraints as vert V vert increases. Therefore, we will add these constraints only when necessary.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_implementation","page":"Traveling Salesperson Problem","title":"Implementation","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"There are two ways we can eliminate subtours in JuMP, both of which will be shown in what follows:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"iteratively solving a new model that incorporates previously identified subtours,\nor adding violated subtours as lazy constraints.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Data","page":"Traveling Salesperson Problem","title":"Data","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The vertices are assumed to be randomly distributed in the Euclidean space; thus, the weight (distance) of each edge is defined as follows.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function generate_distance_matrix(n; random_seed = 1)\n    rng = Random.MersenneTwister(random_seed)\n    X = 100 * rand(rng, n)\n    Y = 100 * rand(rng, n)\n    d = [sqrt((X[i] - X[j])^2 + (Y[i] - Y[j])^2) for i in 1:n, j in 1:n]\n    return X, Y, d\nend\n\nn = 40\nX, Y, d = generate_distance_matrix(n)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"For the JuMP model, we first initialize the model object. Then, we create the binary decision variables and add the objective function and constraints. By defining the x matrix as Symmetric, we do not need to add explicit constraints that x[i, j] == x[j, i].","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function build_tsp_model(d, n)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:n, 1:n], Bin, Symmetric)\n    @objective(model, Min, sum(d .* x) / 2)\n    @constraint(model, [i in 1:n], sum(x[i, :]) == 2)\n    @constraint(model, [i in 1:n], x[i, i] == 0)\n    return model\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"To search for violated constraints, based on the edges that are currently in the solution (that is, those that have value x_ij = 1), we identify the shortest cycle through the function subtour(). Whenever a subtour has been identified, a constraint corresponding to the form above can be added to the model.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function subtour(edges::Vector{Tuple{Int,Int}}, n)\n    shortest_subtour, unvisited = collect(1:n), Set(collect(1:n))\n    while !isempty(unvisited)\n        this_cycle, neighbors = Int[], unvisited\n        while !isempty(neighbors)\n            current = pop!(neighbors)\n            push!(this_cycle, current)\n            if length(this_cycle) > 1\n                pop!(unvisited, current)\n            end\n            neighbors =\n                [j for (i, j) in edges if i == current && j in unvisited]\n        end\n        if length(this_cycle) < length(shortest_subtour)\n            shortest_subtour = this_cycle\n        end\n    end\n    return shortest_subtour\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Let us declare a helper function selected_edges() that will be repeatedly used in what follows.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function selected_edges(x::Matrix{Float64}, n)\n    return Tuple{Int,Int}[(i, j) for i in 1:n, j in 1:n if x[i, j] > 0.5]\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Other helper functions for computing subtours:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"subtour(x::Matrix{Float64}) = subtour(selected_edges(x, size(x, 1)), size(x, 1))\nsubtour(x::AbstractMatrix{VariableRef}) = subtour(value.(x))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Iterative-method","page":"Traveling Salesperson Problem","title":"Iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"An iterative way of eliminating subtours is the following.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"However, it is reasonable to assume that this is not the most efficient way: whenever a new subtour elimination constraint is added to the model, the optimization has to start from the very beginning.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"That way, the solver will repeatedly discard useful information encountered during previous solves (for example, all cuts, the incumbent solution, or lower bounds).","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"info: Info\nNote that, in principle, it would also be feasible to add all subtours (instead of just the shortest one) to the model. However, preventing just the shortest cycle is often sufficient for breaking other subtours and will keep the model size smaller.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"iterative_model = build_tsp_model(d, n)\noptimize!(iterative_model)\ntime_iterated = solve_time(iterative_model)\ncycle = subtour(iterative_model[:x])\nwhile 1 < length(cycle) < n\n    println(\"Found cycle of length $(length(cycle))\")\n    S = [(i, j) for (i, j) in Iterators.product(cycle, cycle) if i < j]\n    @constraint(\n        iterative_model,\n        sum(iterative_model[:x][i, j] for (i, j) in S) <= length(cycle) - 1,\n    )\n    optimize!(iterative_model)\n    global time_iterated += solve_time(iterative_model)\n    global cycle = subtour(iterative_model[:x])\nend\n\nobjective_value(iterative_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"time_iterated","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"As a quick sanity check, we visualize the optimal tour to verify that no subtour is present:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function plot_tour(X, Y, x)\n    plot = Plots.plot()\n    for (i, j) in selected_edges(x, size(x, 1))\n        Plots.plot!([X[i], X[j]], [Y[i], Y[j]]; legend = false)\n    end\n    return plot\nend\n\nplot_tour(X, Y, value.(iterative_model[:x]))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Lazy-constraint-method","page":"Traveling Salesperson Problem","title":"Lazy constraint method","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"A more sophisticated approach makes use of lazy constraints. To be more precise, we do this through the subtour_elimination_callback() below, which is only run whenever we encounter a new integer-feasible solution.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"lazy_model = build_tsp_model(d, n)\nfunction subtour_elimination_callback(cb_data)\n    status = callback_node_status(cb_data, lazy_model)\n    if status != MOI.CALLBACK_NODE_STATUS_INTEGER\n        return  # Only run at integer solutions\n    end\n    cycle = subtour(callback_value.(cb_data, lazy_model[:x]))\n    if !(1 < length(cycle) < n)\n        return  # Only add a constraint if there is a cycle\n    end\n    println(\"Found cycle of length $(length(cycle))\")\n    S = [(i, j) for (i, j) in Iterators.product(cycle, cycle) if i < j]\n    con = @build_constraint(\n        sum(lazy_model[:x][i, j] for (i, j) in S) <= length(cycle) - 1,\n    )\n    MOI.submit(lazy_model, MOI.LazyConstraint(cb_data), con)\n    return\nend\nset_attribute(\n    lazy_model,\n    MOI.LazyConstraintCallback(),\n    subtour_elimination_callback,\n)\noptimize!(lazy_model)\nobjective_value(lazy_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This finds the same optimal tour:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"plot_tour(X, Y, value.(lazy_model[:x]))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Surprisingly, for this particular model with GLPK, the solution time is worse than the iterative method:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"time_lazy = solve_time(lazy_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"In most other cases and solvers, however, the lazy time should be faster than the iterative method.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"CurrentModule = JuMP","category":"page"},{"location":"changelog/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"changelog/#Version-1.16.0-(October-24,-2023)","page":"Release notes","title":"Version 1.16.0 (October 24, 2023)","text":"","category":"section"},{"location":"changelog/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added := operator for Boolean satisfiability problems (#3530)","category":"page"},{"location":"changelog/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed text/latex printing of MOI.Interval sets (#3537)\nFixed tests with duplicate function names (#3539)","category":"page"},{"location":"changelog/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated documentation list of supported solvers (#3527) (#3529) (#3538) (#3542) (#3545) (#3546)\nUpdated to Documenter@1.1 (#3528)\nFixed various tutorials (#3534) (#3532)\nFixed Project.toml compat bounds for standard libraries (#3544)","category":"page"},{"location":"changelog/#Version-1.15.1-(September-24,-2023)","page":"Release notes","title":"Version 1.15.1 (September 24, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for single argument min and max operators (#3522)\nFixed error message for add_to_expression! when called with a GenericNonlinearExpr (#3506)\nFixed constraint tags with broadcasted constraints (#3515)\nFixed MethodError in MA.scaling (#3518)\nFixed support for arrays of Parameter variables (#3524)","category":"page"},{"location":"changelog/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated to Documenter@1 (#3501)\nFixed links to data in tutorials (#3512)\nFixed typo in TSP tutorial (#3516)\nImproved error message for VariableNotOwned errors (#3520)\nFixed various JET errors (#3519)","category":"page"},{"location":"changelog/#Version-1.15.0-(September-15,-2023)","page":"Release notes","title":"Version 1.15.0 (September 15, 2023)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it adds an entirely new data structure and API path for working with nonlinear programs. The previous nonlinear interface remains unchanged and is documented at Nonlinear Modeling (Legacy). The new interface is a treated as a non-breaking feature addition and is documented at Nonlinear Modeling.","category":"page"},{"location":"changelog/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Although the new nonlinear interface is a feature addition, there are two changes which might be breaking for a very small number of users.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The syntax inside JuMP macros is parsed using a different code path, even for linear and quadratic expressions. We made this change to unify how we parse linear, quadratic, and nonlinear expressions. In all cases, the new code returns equivalent expressions, but because of the different order of operations, there are three changes to be aware of when updating:\nThe printed form of the expression may change, for example from x * y to y * x. This can cause tests which test the String representation of a model to fail.\nSome coefficients may change slightly due to floating point round-off error.\nParticularly when working with a JuMP extension, you may encounter a MethodError due to a missing or ambiguous method. These errors are due to previously existing bugs that were not triggered by the previous parsing code. If you encounter such an error, please open a GitHub issue.\nThe methods for Base.:^(x::VariableRef, n::Integer) and Base.:^(x::AffExpr, n::Integer) have changed. Previously, these methods supported only n = 0, 1, 2 and they always returned a QuadExpr, even for the case when n = 0 or n = 1. Now:\nx^0 returns one(T), where T is the value_type of the model (defaults to Float64)\nx^1 returns x\nx^2 returns a QuadExpr\nx^n where !(0 <= n <= 2) returns a NonlinearExpr.\nWe made this change to support nonlinear expressions and to align the mathematical definition of the operation with their return type. (Previously, users were surprised that x^1 returned a QuadExpr.) As a consequence of this change, the methods are now not type-stable. This means that the compiler cannot prove that x^2 returns a QuadExpr. If benchmarking shows that this is a performance problem, you can use the type-stable x * x instead of x^2.","category":"page"},{"location":"changelog/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added triangle_vec which simplifies adding MOI.LogDetConeTriangle and MOI.RootDetConeTriangle constraints (#3456)\nAdded the new nonlinear interface. This is a very large change. See the documentation at Nonlinear Modeling and the (long) discussion in JuMP.jl#3106. Related PRs are (#3468) (#3472) (#3475) (#3483) (#3487) (#3488) (#3489) (#3504) (#3509)","category":"page"},{"location":"changelog/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed uses of @nospecialize which cause precompilation failures in Julia v1.6.0 and v1.6.1. (#3464)\nFixed adding a container of Parameter (#3473)\nFixed return type of x^0 and x^1 to no longer return QuadExpr (see note in Breaking section above) (#3474)\nFixed error messages in LowerBoundRef, UpperBoundRef, FixRef, IntegerRef, BinaryRef, ParameterRef and related functions (#3494)\nFixed type inference of empty containers in JuMP macros (#3500)","category":"page"},{"location":"changelog/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added GAMS to solver documentation (#3357)\nUpdated various tutorials (#3459) (#3460) (#3462) (#3463) (#3465) (#3490) (#3492) (#3503)\nAdded The network multi-commodity flow problem tutorial (#3491)\nAdded Two-stage stochastic programs tutorial (#3466)\nAdded better error messages for unsupported operations in LinearAlgebra (#3476)\nUpdated to the latest version of Documenter (#3484) (#3495) (#3497)\nUpdated GitHub action versions (#3507)","category":"page"},{"location":"changelog/#Version-1.14.1-(September-2,-2023)","page":"Release notes","title":"Version 1.14.1 (September 2, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix links in Documentation (#3478)","category":"page"},{"location":"changelog/#Version-1.14.0-(August-27,-2023)","page":"Release notes","title":"Version 1.14.0 (August 27, 2023)","text":"","category":"section"},{"location":"changelog/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added DimensionalData.jl extension (#3413)\nAdded syntactic sugar for the MOI.Parameter set (#3443)\nParameter\nParameterRef\nis_parameter\nparameter_value\nset_parameter_value","category":"page"},{"location":"changelog/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed model_convert for BridgeableConstraint (#3437)\nFixed printing models with integer coefficients larger than typemax(Int) (#3447)\nFixed support for constant left-hand side functions in a complementarity constraint (#3452)","category":"page"},{"location":"changelog/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated packages used in documentation (#3444) (#3455)\nFixed docstring tests (#3445)\nFixed printing change for MathOptInterface (#3446)\nFixed typos in documentation (#3448) (#3457)\nAdded SCIP to callback documentation (#3449)","category":"page"},{"location":"changelog/#Version-1.13.0-(July-27,-2023)","page":"Release notes","title":"Version 1.13.0 (July 27, 2023)","text":"","category":"section"},{"location":"changelog/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for generic number types (#3377) (#3385)\nAdded fallback for MOI.AbstractSymmetricMatrixSetTriangle and MOI.AbstractSymmetricMatrixSetSquare (#3424)","category":"page"},{"location":"changelog/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed set_start_values with MOI.Bridges.Objective.SlackBridge (#3422)\nFixed flakey doctest in variables.md (#3425)\nFixed names on CITATION.bib (#3423)","category":"page"},{"location":"changelog/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Loraine.jl to the installation table (#3426)\nRemoved Penopt.jl from packages.toml (#3428)\nImproved problem statement in cannery example of tutorial (#3430)\nMinor cleanups in Containers.DenseAxisArray implementation (#3429)\nChanged nested_problems.jl: outer/inner to upper/lower (#3433)\nRemoved second SDP relaxation in OPF tutorial (#3432)","category":"page"},{"location":"changelog/#Version-1.12.0-(June-19,-2023)","page":"Release notes","title":"Version 1.12.0 (June 19, 2023)","text":"","category":"section"},{"location":"changelog/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added coefficient_type keyword argument to add_bridge and remove_bridge (#3394)","category":"page"},{"location":"changelog/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed error message for matrix in HermitianPSDCone (#3369)\nFixed EditURL for custom documentation pages (#3373)\nFixed return type annotations for MOI.ConstraintPrimal and MOI.ConstraintDual (#3381)\nFixed printing change in Julia nightly (#3391)\nFixed printing of Complex coefficients (#3397)\nFixed printing of constraints in text/latex mode (#3405)\nFixed performance issue in Containers.rowtable (#3410)\nFixed bug when variables added to set of wrong dimension (#3411)","category":"page"},{"location":"changelog/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added more solver READMEs to the documentation (#3358) (#3360) (#3364) (#3365) (#3366) (#3368) (#3372) (#3374) (#3376) (#3379) (#3387) (#3389)\nAdded StatusSwitchingQP.jl to the installation table (#3354)\nUpdated checklist for adding a new solver (#3370)\nUpdated extension-tests.yml action (#3371) (#3375)\nColor logs in GitHub actions (#3392)\nAdded new tutorials\nOptimal power flow (#3395) (#3412)\nLovász numbers (#3399)\nDualization (#3402)\nUpdated JuMP paper citation (#3400)\nChanged GitHub action to upload LaTeX logs when building documentation (#3403)\nFixed printing of SCS log in documentation (#3406)\nUpdated solver versions (#3407)\nUpdated documentation to use Julia v1.9 (#3398)\nReplaced _value_type with MOI.Utilities.value_type (#3414)\nFixed a typo in docstring (#3415)\nRefactored API documentation (#3386)\nUpdated SCIP license (#3420)","category":"page"},{"location":"changelog/#Version-1.11.1-(May-19,-2023)","page":"Release notes","title":"Version 1.11.1 (May 19, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a poor error message when sum(::DenseAxisArray; dims) was called (#3338)\nFixed support for dependent sets in the @variable macro (#3344)\nFixed a performance bug in constraints with sparse symmetric matrices (#3349)","category":"page"},{"location":"changelog/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the printing of complex numbers (#3332)\nWhen printing, sets which contain constants ending in .0 now print as integers. This follows the behavior of constants in functions (#3341)\nAdded InfiniteOpt to the extensions documentation (#3343)\nAdded more documentation for the exponential cone (#3345) (#3347)\nAdded checklists for developers (#3346) (#3355)\nFixed test support upcoming Julia nightly (#3351)\nFixed extension-tests.yml action (#3353)\nAdd more solvers to the documentation (#3359) (#3361) (#3362)","category":"page"},{"location":"changelog/#Version-1.11.0-(May-3,-2023)","page":"Release notes","title":"Version 1.11.0 (May 3, 2023)","text":"","category":"section"},{"location":"changelog/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added new methods to print_active_bridges for printing a particular objective, constraint, or variable (#3316)","category":"page"},{"location":"changelog/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed tests for MOI v1.14.0 release (#3312)\nFixed indexing containers when an axis is Vector{Any} that contains a Vector{Any} element (#3280)\nFixed getindex(::AbstractJuMPScalar) which is called for an expression like x[] (#3314)\nFixed bug in set_string_names_on_creation with a vector of variables (#3322)\nFixed bug in memoize function in nonlinear documentation (#3337)","category":"page"},{"location":"changelog/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#3317) (#3318) (#3328)\nAdded a test for the order of setting start values (#3315)\nAdded READMEs of solvers and extensions to the docs (#3309) (#3320) (#3327) (#3329) (#3333)\nStyle improvements to src/variables.jl (#3324)\nClarify that column generation does not find global optimum (#3325)\nAdd a GitHub actions workflow for testing extensions prior to release (#3331)\nDocument the release process for JuMP (#3334)\nFix links to discourse and chatroom (#3335)","category":"page"},{"location":"changelog/#Version-1.10.0-(April-3,-2023)","page":"Release notes","title":"Version 1.10.0 (April 3, 2023)","text":"","category":"section"},{"location":"changelog/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Nonnegatives, Nonpositives and Zeros, and support vector-valued inequality syntax in the JuMP macros (#3273)\nAdded special support for LinearAlgebra.Symmetric and LinearAlgebra.Hermitian matrices in Zeros constraints (#3281) (#3296)\nAdded HermitianMatrixSpace and the Hermitian tag for generating a matrix of variables that is Hermitian (#3292) (#3293)\nAdded Semicontinuous and Semiinteger (#3302)\nAdded support for keyword indexing of containers (#3237)","category":"page"},{"location":"changelog/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed [compat] bound for MathOptInterface in Project.toml (#3272)","category":"page"},{"location":"changelog/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Split out the Nested optimization problems tutorial (#3274)\nUpdated doctests to ensure none have hidden state (#3275) (#3276)\nClarified how lazy constraints may revisit points (#3278)\nAdded P-Norm example (#3282)\nClarified docs that macros create new bindings (#3284)\nFixed threading example (#3283)\nAdded plot to The minimum distortion problem (#3288)\nAdded Google style rules for Vale and fixed warnings (#3285)\nAdded citation for the JuMP 1.0 paper (#3294)\nUpdated package versions in the documentation (#3298)\nAdded comment for the order in which start values must be set (#3303)\nImproved error message for unrecognized constraint operators (#3311)","category":"page"},{"location":"changelog/#Version-1.9.0-(March-7,-2023)","page":"Release notes","title":"Version 1.9.0 (March 7, 2023)","text":"","category":"section"},{"location":"changelog/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added get_attribute and set_attribute. These replace get_optimizer_attribute and set_optimizer_attribute, although the _optimizer_ functions remain for backward compatibility. (#3219)\nAdded set_start_values for setting all supported start values in a model (#3238)\nAdd remove_bridge and print_active_bridges (#3259)","category":"page"},{"location":"changelog/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The matrix returned by a variable in HermitianPSDCone is now a LinearAlgebra.Hermitian matrix. This is potentially breaking if you have written code to assume the return is a Matrix. (#3245) (#3246)\nFixed missing support for Base.isreal of expressions (#3252)","category":"page"},{"location":"changelog/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a thread safety issue in the Parallelism tutorial (#3240) (#3243)\nImproved the error message when unsupported operators are used in @NL macros (#3236)\nClarified the documentation to say that matrices in HermitianPSDCone must be LinearAlgebra.Hermitian (#3241)\nMinor style fixes to internal macro code (#3247)\nAdd Quantum state discrimination tutorial (#3250)\nImprove error message when begin...end not passed to plural macros (#3255)\nDocument how to register function with varying number of input arguments (#3258)\nTidy tests by removing unneeded JuMP. prefixes (#3260)\nClarified the introduction to the Complex number support tutorial (#3262)\nFixed typos in the Documentation (#3263) (#3266) (#3268) (#3269)","category":"page"},{"location":"changelog/#Version-1.8.2-(February-27,-2023)","page":"Release notes","title":"Version 1.8.2 (February 27, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed dot product between complex JuMP expression and number (#3244)","category":"page"},{"location":"changelog/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Polish simple SDP examples (#3232)","category":"page"},{"location":"changelog/#Version-1.8.1-(February-23,-2023)","page":"Release notes","title":"Version 1.8.1 (February 23, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for init in nonlinear generator expressions (#3226)","category":"page"},{"location":"changelog/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Use and document import MathOptInterface as MOI (#3222)\nRemoved references in documentation to multiobjective optimization being unsupported (#3223)\nAdded tutorial on multi-objective portfolio optimization (#3227)\nRefactored some of the conic tutorials (#3229)\nFixed typos in the documentation (#3230)\nAdded tutorial on parallelism (#3231)","category":"page"},{"location":"changelog/#Version-1.8.0-(February-16,-2023)","page":"Release notes","title":"Version 1.8.0 (February 16, 2023)","text":"","category":"section"},{"location":"changelog/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added --> syntax support for indicator constraints. The old syntax of => remains supported (#3207)\nAdded <--> syntax for reified constraints. For now, few solvers support reified constraints (#3206)\nAdded fix_discrete_variables. This is most useful for computing the dual of a mixed-integer program (#3208)\nAdded support for vector-valued objectives. For details, see the Multi-objective knapsack tutorial (#3176)","category":"page"},{"location":"changelog/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report by switching to an explicit LU factorization of the basis matrix (#3182)\nFixed a bug that prevented [; kwarg] arguments in macros (#3220)","category":"page"},{"location":"changelog/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Minor fixes to the documentation (#3200) (#3201) (#3203) (#3210)\nAdded tutorial Constraint programming (#3202)\nAdded more examples to Tips and Tricks\nRemove _distance_to_set in favor of MOI.Utilities.distance_to_set (#3209)\nImprove The diet problem tutorial by adding the variable as a column in the dataframe (#3213)\nImprove The knapsack problem example tutorial (#3216) (#3217)\nAdded the Ellipsoid approximation tutorial (#3218)","category":"page"},{"location":"changelog/#Version-1.7.0-(January-25,-2023)","page":"Release notes","title":"Version 1.7.0 (January 25, 2023)","text":"","category":"section"},{"location":"changelog/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for view of a Containers.DenseAxisArray (#3152) (#3180)\nAdded support for containers of variables in ComplexPlane (#3184)\nAdded support for minimum and maximum generators in nonlinear expressions (#3189)\nAdded SnoopPrecompile statements that reduce the time-to-first-solve in Julia 1.9 (#3193) (#3195) (#3196) (#3197)","category":"page"},{"location":"changelog/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Large refactoring of the tests (#3166) (#3167) (#3168) (#3169) (#3170) (#3171)\nRemove unreachable code due to VERSION checks (#3172)\nDocument how to test JuMP extensions (#3174)\nFix method ambiguities in Containers (#3173)\nImprove error message that is thrown when = is used instead of == in the @constraint macro (#3178)\nImprove the error message when Bool is used instead of Bin in the @variable macro (#3180)\nUpdate versions of the documentation (#3185)\nTidy the import of packages and remove unnecessary prefixes (#3186) (#3187)\nRefactor src/JuMP.jl by moving methods into more relevant files (#3188)\nFix docstring of Model not appearing in the documentation (#3198)","category":"page"},{"location":"changelog/#Version-1.6.0-(January-1,-2023)","page":"Release notes","title":"Version 1.6.0 (January 1, 2023)","text":"","category":"section"},{"location":"changelog/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a result keyword argument to solution_summary to allow summarizing models with multiple solutions (#3138)\nAdded relax_with_penalty!, which is a useful tool when debugging infeasible models (#3140)\nAdded has_start_value (#3157)\nAdded support for HermitianPSDCone in constraints (#3154)","category":"page"},{"location":"changelog/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed promotion of complex expressions (#3150) (#3164)","category":"page"},{"location":"changelog/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Benders tutorial with in-place resolves (#3145)\nAdded more Tips and tricks for linear programs (#3144) (#3163)\nClarified documentation that start can depend on the indices of a variable container (#3148)\nReplace instances of length and size by the recommended eachindex and axes (#3149)\nAdded a warning explaining why the model is dirty when accessing solution results from a modified model (#3156)\nClarify documentation that PSD ensures a symmetric matrix (#3159)\nMaintenance of the JuMP test suite (#3146) (#3158) (#3162)","category":"page"},{"location":"changelog/#Version-1.5.0-(December-8,-2022)","page":"Release notes","title":"Version 1.5.0 (December 8, 2022)","text":"","category":"section"},{"location":"changelog/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for complex-valued variables:\nHermitianPSDCone (#3109)\nComplexPlane and ComplexVariable (#3134)\nAdd support for MOI.OptimizerWithAttributes in set_optimizer_attribute and get_optimizer_attribute (#3129)","category":"page"},{"location":"changelog/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed error message for vectorized interval constraints (#3123)\nFixed passing AbstractString to set_optimizer_attribute (#3127)","category":"page"},{"location":"changelog/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update package versions used in docs (#3119) (#3133) (#3139)\nFixed output of diet tutorial (#3120)\nExplain how to use Dates.period in set_time_limit_sec (#3121)\nUpdate to JuliaFormatter v1.0.15 (#3130)\nFixed HTTP server example in web_app.jl (#3131)\nUpdate docs to build with Documenter#master (#3094)\nAdd tests for LinearAlgebra operations (#3132)\nTidy these release notes (#3135)\nAdded documentation for Complex number support (#3141)\nRemoved the \"workforce scheduling\" and \"steelT3\" tutorials (#3143)","category":"page"},{"location":"changelog/#Version-1.4.0-(October-29,-2022)","page":"Release notes","title":"Version 1.4.0 (October 29, 2022)","text":"","category":"section"},{"location":"changelog/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Containers.rowtable which converts a container into a vector of NamedTuples to support the Tables.jl interface. This simplifies converting Containers.DenseAxisArray and Containers.SparseAxisArray objects into tabular forms such as a DataFrame (#3104)\nAdded a new method to Containers.container so that index names are passed to the container (#3088)","category":"page"},{"location":"changelog/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in copy_to(dest::Model, src::MOI.ModelLike) when src has nonlinear components (#3101)\nFixed the printing of (-1.0 + 0.0im) coefficients in complex expressions (#3112)\nFixed a parsing bug in nonlinear expressions with generator statements that contain multiple for statements (#3116)","category":"page"},{"location":"changelog/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Converted the multi-commodity flow tutorial to use an SQLite database (#3098)\nFixed a number of typos in the documentation (#3103) (#3107) (#3018)\nImproved various style aspects of the PDF documentation (#3095) (#3098) (#3102)","category":"page"},{"location":"changelog/#Version-1.3.1-(September-28,-2022)","page":"Release notes","title":"Version 1.3.1 (September 28, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in relax_integrality (#3087)\nFixed the type stability of operators with Complex arguments (#3072)\nFixed a bug which added additional +() terms to some nonlinear expressions (#3091)\nFixed potential method ambiguities with AffExpr and QuadExpr objects (#3092)","category":"page"},{"location":"changelog/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added vale as a linter for the documentation (#3080)\nAdded a tutorial on debugging JuMP models (#3043)\nFixed a number of typos in the documentation (#3079) (#3083)\nMany other small tweaks to the documentation (#3068) (#3073) (#3074) (#3075) (#3076) (#3077) (#3078) (#3081) (#3082) (#3084) (#3085) (#3089)","category":"page"},{"location":"changelog/#Version-1.3.0-(September-5,-2022)","page":"Release notes","title":"Version 1.3.0 (September 5, 2022)","text":"","category":"section"},{"location":"changelog/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support slicing in SparseAxisArray (#3031)","category":"page"},{"location":"changelog/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug introduced in v1.2.0 that prevented DenseAxisArrays with Vector keys (#3064)","category":"page"},{"location":"changelog/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Released the JuMP logos under the CC BY 4.0 license (#3063)\nMinor tweaks to the documentation (#3054) (#3056) (#3057) (#3060) (#3061) (#3065)\nImproved code coverage of a number of files (#3048) (#3049) (#3050) (#3051) (#3052) (#3053) (#3058) (#3059)","category":"page"},{"location":"changelog/#Version-1.2.1-(August-22,-2022)","page":"Release notes","title":"Version 1.2.1 (August 22, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug when parsing two-sided nonlinear constraints (#3045)","category":"page"},{"location":"changelog/#Version-1.2.0-(August-16,-2022)","page":"Release notes","title":"Version 1.2.0 (August 16, 2022)","text":"","category":"section"},{"location":"changelog/#Breaking-2","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it significantly refactors the internal code for handling nonlinear programs to use the MathOptInterface.Nonlinear submodule that was introduced in MathOptInterface v1.3.0. As a consequence, the internal datastructure in model.nlp_data has been removed, as has the JuMP._Derivatives submodule. Despite the changes, the public API for nonlinear programming has not changed, and any code that uses only the public API and that worked with v1.1.1 will continue to work with v1.2.0.","category":"page"},{"location":"changelog/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added all_constraints(model; include_variable_in_set_constraints) which simplifies returning a list of all constraint indices in the model.\nAdded the ability to delete nonlinear constraints via delete(::Model, ::NonlinearConstraintRef).\nAdded the ability to provide an explicit Hessian for a multivariate user-defined function.\nAdded support for querying the primal value of a nonlinear constraint via value(::NonlinearConstraintRef)","category":"page"},{"location":"changelog/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Containers.DenseAxisArray so that it now supports indexing with keys that hash to the same value, even if they are different types, for example, Int32 and Int64.\nFixed a bug printing the model when the solver does not support MOI.Name.","category":"page"},{"location":"changelog/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a constraint programming formulation to the Sudoku tutorial.\nAdded newly supported solvers Pajarito, Clarabel, and COPT to the installation table.\nFixed a variety of other miscellaneous issues in the documentation.","category":"page"},{"location":"changelog/#Version-1.1.1-(June-14,-2022)","page":"Release notes","title":"Version 1.1.1 (June 14, 2022)","text":"","category":"section"},{"location":"changelog/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed problem displaying LaTeX in the documentation\nMinor updates to the style guide\nUpdated to MOI v1.4.0 in the documentation","category":"page"},{"location":"changelog/#Version-1.1.0-(May-25,-2022)","page":"Release notes","title":"Version 1.1.0 (May 25, 2022)","text":"","category":"section"},{"location":"changelog/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added num_constraints(::Model; count_variable_in_set_constraints) to simplify the process of counting the number of constraints in a model\nAdded VariableRef(::ConstraintRef) for querying the variable associated with a bound or integrality constraint.\nAdded set_normalized_coefficients for modifying the variable coefficients of a vector-valued constraint.\nAdded set_string_names_on_creation to disable creating String names for variables and constraints. This can improve performance.","category":"page"},{"location":"changelog/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug passing nothing to the start keyword of @variable","category":"page"},{"location":"changelog/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"New tutorials:\nSensitivity analysis of a linear program\nServing web apps\nMinimal ellipse SDP tutorial refactored and improved\nDocs updated to the latest version of each package\nLots of minor fixes and improvements to the documentation","category":"page"},{"location":"changelog/#Version-1.0.0-(March-24,-2022)","page":"Release notes","title":"Version 1.0.0 (March 24, 2022)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Read more about this release, along with an acknowledgement of all the contributors in our JuMP 1.0.0 is released blog post.","category":"page"},{"location":"changelog/#Breaking-3","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The previously deprecated functions (v0.23.0, v0.23.1) have been removed. Deprecation was to improve consistency of function names:\nnum_nl_constraints (see num_nonlinear_constraints)\nall_nl_constraints (see all_nonlinear_constraints)\nadd_NL_expression (see add_nonlinear_expression)\nset_NL_objective  (see set_nonlinear_objective)\nadd_NL_constraint (see add_nonlinear_constraint)\nnl_expr_string (see nonlinear_expr_string)\nnl_constraint_string (see nonlinear_constraint_string)\nSymMatrixSpace (see SymmetricMatrixSpace)\nThe unintentionally exported variable JuMP.op_hint has been renamed to the unexported JuMP._OP_HINT","category":"page"},{"location":"changelog/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .nl files\nFixed a bug broadcasting SparseAxisArrays","category":"page"},{"location":"changelog/#Version-0.23.2-(March-14,-2022)","page":"Release notes","title":"Version 0.23.2 (March 14, 2022)","text":"","category":"section"},{"location":"changelog/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added relative_gap to solution_summary\nregister now throws an informative error if the function is not differentiable using ForwardDiff. In some cases, the check in register will encounter a false negative, and the informative error will be thrown at run-time. This usually happens when the function is non-differentiable in a subset of the domain.","category":"page"},{"location":"changelog/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a scoping issue when extending the container keyword of containers","category":"page"},{"location":"changelog/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Docs updated to the latest version of each package","category":"page"},{"location":"changelog/#Version-0.23.1-(March-2,-2022)","page":"Release notes","title":"Version 0.23.1 (March 2, 2022)","text":"","category":"section"},{"location":"changelog/#Deprecated","page":"Release notes","title":"Deprecated","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"nl_expr_string and nl_constraint_string have been renamed to nonlinear_expr_string and nonlinear_constraint_string. The old methods still exist with deprecation warnings. This change should impact very few users because to call them you must rely on private internals of the nonlinear API. Users are encouraged to use sprint(show, x) instead, where x is the nonlinear expression or constraint of interest.","category":"page"},{"location":"changelog/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for Base.abs2(x) where x is a variable or affine expression. This is mainly useful for complex-valued constraints.","category":"page"},{"location":"changelog/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed addition of complex and real affine expressions\nFixed arithmetic for Complex-valued quadratic expressions\nFixed variable bounds passed as Rational{Int}(Inf)\nFixed printing of the coefficient (0 + 1im)\nFixed a bug when solution_summary is called prior to optimize!","category":"page"},{"location":"changelog/#Version-0.23.0-(February-25,-2022)","page":"Release notes","title":"Version 0.23.0 (February 25, 2022)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP v0.23.0 is a breaking release. It is also a release-candidate for JuMP v1.0.0. That is, if no issues are found with the v0.23.0 release, then it will be re-tagged as v1.0.0.","category":"page"},{"location":"changelog/#Breaking-4","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nMathOptInterface has been updated to v1.0.0\nAll previously deprecated functionality has been removed\nPrintMode, REPLMode and IJuliaMode have been removed in favor of the MIME types MIME\"text/plain\" and MIME\"text/latex\". Replace instances of ::Type{REPLMode} with ::MIME\"text/plain\", REPLMode with MIME(\"text/plain\"), ::Type{IJuliaMode} with ::MIME\"text/latex\", and IJuliaMode with MIME(\"text/latex\").\nFunctions containing the nl_ acronym have been renamed to the more explicit nonlinear_. For example, num_nl_constraints is now num_nonlinear_constraints and set_NL_objective is now set_nonlinear_objective. Calls to the old functions throw an error explaining the new name.\nSymMatrixSpace has been renamed to SymmetricMatrixSpace","category":"page"},{"location":"changelog/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added nonlinear_dual_start_value and set_nonlinear_dual_start_value\nAdded preliminary support for Complex coefficient types","category":"page"},{"location":"changelog/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary","category":"page"},{"location":"changelog/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"MILP examples have been migrated from GLPK to HiGHS\nFixed various typos\nImproved section on setting constraint start values","category":"page"},{"location":"changelog/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated functionality. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Then install JuMP v0.22.3:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add JuMP@0.22.3\"","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"changelog/#Version-0.22.3-(February-10,-2022)","page":"Release notes","title":"Version 0.22.3 (February 10, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a reproducibility issue in the TSP tutorial\nFixed a reproducibility issue in the max_cut_sdp tutorial\nFixed a bug broadcasting an empty SparseAxisArray","category":"page"},{"location":"changelog/#Other-25","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a warning and improved documentation for the modify-then-query case\nFixed a typo in the docstring of RotatedSecondOrderCone\nAdded Aqua.jl as a check for code health\nAdded introductions to each section of the tutorials\nImproved the column generation and Benders decomposition tutorials\nUpdated documentation to MOI v0.10.8\nUpdated JuliaFormatter to v0.22.2","category":"page"},{"location":"changelog/#Version-0.22.2-(January-10,-2022)","page":"Release notes","title":"Version 0.22.2 (January 10, 2022)","text":"","category":"section"},{"location":"changelog/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The function all_nl_constraints now returns all nonlinear constraints in a model\nstart_value and set_start_value can now be used to get and set the primal start for constraint references\nPlural macros now return a tuple containing the elements that were defined instead of nothing\nAnonymous variables are now printed as _[i] where i is the index of the variable instead of noname. Calling name(x) still returns \"\" so this is non-breaking.","category":"page"},{"location":"changelog/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed handling of min and max in nonlinear expressions\nCartesianIndex is no longer allowed as a key for DenseAxisArrays.","category":"page"},{"location":"changelog/#Other-26","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the performance of GenericAffExpr\nAdded a tutorial on the Travelling Salesperson Problem\nAdded a tutorial on querying the Hessian of a nonlinear program\nAdded documentation on using custom solver binaries.","category":"page"},{"location":"changelog/#Version-0.22.1-(November-29,-2021)","page":"Release notes","title":"Version 0.22.1 (November 29, 2021)","text":"","category":"section"},{"location":"changelog/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Export OptimizationSense enum, with instances: MIN_SENSE, MAX_SENSE, and FEASIBILITY_SENSE\nAdd Base.isempty(::Model) to match Base.empty(::Model)","category":"page"},{"location":"changelog/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in container with tuples as indices\nFix bug in set_time_limit_sec","category":"page"},{"location":"changelog/#Other-27","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add tutorial \"Design patterns for larger models\"\nRemove release notes section from PDF\nGeneral edits of the documentation and error messages","category":"page"},{"location":"changelog/#Version-0.22.0-(November-10,-2021)","page":"Release notes","title":"Version 0.22.0 (November 10, 2021)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP v0.22 is a breaking release","category":"page"},{"location":"changelog/#Breaking-5","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP 0.22 contains a number of breaking changes. However, these should be invisible for the majority of users. You will mostly encounter these breaking changes if you: wrote a JuMP extension, accessed backend(model), or called @SDconstraint.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The breaking changes are as follows:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"MathOptInterface has been updated to v0.10.4. For users who have interacted with the MOI backend, this contains a large number of breaking changes. Read the MathOptInterface release notes for more details.\nThe bridge_constraints keyword argument to Model and set_optimizer has been renamed add_bridges to reflect that more thing were bridged than just constraints.\nThe backend(model) field now contains a concrete instance of a MOI.Utilities.CachingOptimizer instead of one with an abstractly typed optimizer field. In most cases, this will lead to improved performance. However, calling set_optimizer after backend invalidates the old backend. For example:\nmodel = Model()\nb = backend(model)\nset_optimizer(model, GLPK.Optimizer)\n@variable(model, x)\n# b is not updated with `x`! Get a new b by calling `backend` again.\nnew_b = backend(model)\nAll usages of @SDconstraint are deprecated. The new syntax is @constraint(model, X >= Y, PSDCone()).\nCreating a DenseAxisArray with a Number as an axis will now display a warning. This catches a common error in which users write @variable(model, x[length(S)]) instead of @variable(model, x[1:length(S)]).\nThe caching_mode argument to Model, for example, Model(caching_mode = MOIU.MANUAL) mode has been removed. For more control over the optimizer, use direct_model instead.\nThe previously deprecated lp_objective_perturbation_range and lp_rhs_perturbation_range functions have been removed. Use lp_sensitivity_report instead.\nThe .m fields of NonlinearExpression and NonlinearParameter have been renamed to .model.\nInfinite variable bounds are now ignored. Thus, @variable(model, x <= Inf) will show has_upper_bound(x) == false. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.\nThe variable_type and constraint_type functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.\nThe internal functions moi_mode, moi_bridge_constraints, moi_add_constraint, and moi_add_to_function_constant are no longer exported.\nThe un-used method Containers.generate_container has been deleted.\nThe Containers API has been refactored, and _build_ref_sets is now public as Containers.build_ref_sets.\nThe parse_constraint_ methods for extending @constraint at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.","category":"page"},{"location":"changelog/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The TerminationStatusCode and ResultStatusCode enums are now exported by JuMP. Prefer termination_status(model) == OPTIMAL instead of == MOI.OPTIMAL, although the MOI. prefix way still works.\nCopy a x::DenseAxisArray to an Array by calling Array(x).\nNonlinearExpression is now a subtype of AbstractJuMPScalar\nConstraints such as @constraint(model, x + 1 in MOI.Integer()) are now supported.\nprimal_feasibility_report now accepts a function as the first argument.\nScalar variables @variable(model, x[1:2] in MOI.Integer()) creates two variables, both of which are constrained to be in the set MOI.Integer.\nConic constraints can now be specified as inequalities under a different partial ordering. So @constraint(model, x - y in MOI.Nonnegatives()) can now be written as @constraint(model, x >= y, MOI.Nonnegatives()).\nNames are now set for vectorized constraints.","category":"page"},{"location":"changelog/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue when show was called on a SparseAxisArray with a large number of elements.\nFixed a bug displaying barrier and simplex iterations in solution_summary.\nFixed a bug by implementing hash for DenseAxisArray and SparseAxisArray.\nNames are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with direct_model.\nMutableArithmetics.Zero is converted into a 0.0 before being returned to the user. Previously, some calls to @expression would return the undocumented MutableArithmetics.Zero() object. One example is summing over an empty set @expression(model, sum(x[i] for i in 1:0)). You will now get 0.0 instead.\nAffExpr and QuadExpr can now be used with == 0 instead of iszero. This fixes a number of issues relating to Julia standard libraries such as LinearAlgebra and SparseArrays.\nFixed a bug when registering a user-defined function with splatting.","category":"page"},{"location":"changelog/#Other-28","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The documentation is now available as a PDF.\nThe documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)\nThe documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.\nThe documentation is now built with Julia 1.6 instead of 1.0.\nVarious error messages have been improved to be more readable.","category":"page"},{"location":"changelog/#Version-0.21.10-(September-4,-2021)","page":"Release notes","title":"Version 0.21.10 (September 4, 2021)","text":"","category":"section"},{"location":"changelog/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added add_NL_expression\nadd_NL_xxx functions now support AffExpr and QuadExpr as terms","category":"page"},{"location":"changelog/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary\nFixed a bug in relax_integrality","category":"page"},{"location":"changelog/#Other-29","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved error message in lp_sensitivity_report","category":"page"},{"location":"changelog/#Version-0.21.9-(August-1,-2021)","page":"Release notes","title":"Version 0.21.9 (August 1, 2021)","text":"","category":"section"},{"location":"changelog/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Containers now support arbitrary container types by passing the type to the container keyword and overloading Containers.container.\nis_valid now supports nonlinear constraints\nAdded unsafe_backend for querying the inner-most optimizer of a JuMP model.\nNonlinear parameters now support the plural @NLparameters macro.\nContainers (for example, DenseAxisArray) can now be used in vector-valued constraints.","category":"page"},{"location":"changelog/#Other-30","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Various improvements to the documentation.","category":"page"},{"location":"changelog/#Version-0.21.8-(May-8,-2021)","page":"Release notes","title":"Version 0.21.8 (May 8, 2021)","text":"","category":"section"},{"location":"changelog/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The @constraint macro is now extendable in the same way as @variable.\nAffExpr and QuadExpr can now be used in nonlinear macros.","category":"page"},{"location":"changelog/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report.\nFixed an inference issue when creating empty SparseAxisArrays.","category":"page"},{"location":"changelog/#Version-0.21.7-(April-12,-2021)","page":"Release notes","title":"Version 0.21.7 (April 12, 2021)","text":"","category":"section"},{"location":"changelog/#Added-26","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added primal_feasibility_report, which can be used to check whether a primal point satisfies primal feasibility.\nAdded coefficient, which returns the coefficient associated with a variable in affine and quadratic expressions.\nAdded copy_conflict, which returns the IIS of an infeasible model.\nAdded solution_summary, which returns (and prints) a struct containing a summary of the solution.\nAllow AbstractVector in vector constraints instead of just Vector.\nAdded latex_formulation(model) which returns an object representing the latex formulation of a model. Use print(latex_formulation(model)) to print the formulation as a string.\nUser-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.\nDenseAxisArray's now support broadcasting over multiple arrays.\nContainer indices can now be iterators of Base.SizeUnknown.","category":"page"},{"location":"changelog/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug in rad2deg and deg2rad in nonlinear expressions.\nFixed a MethodError bug in Containers when forcing container type.\nAllow partial slicing of a DenseAxisArray, resolving an issue from 2014.\nFixed a bug printing variable names in IJulia.\nEnding an IJulia cell with model now prints a summary of the model (like in the REPL) not the latex formulation. Use print(model) to print the latex formulation.\nFixed a bug when copying models containing nested arrays.","category":"page"},{"location":"changelog/#Other-31","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Tutorials are now part of the documentation, and more refactoring has taken place.\nAdded JuliaFormatter added as a code formatter.\nAdded some precompilation statements to reduce initial latency.\nVarious improvements to error messages to make them more helpful.\nImproved performance of value(::NonlinearExpression).\nImproved performance of fix(::VariableRef).","category":"page"},{"location":"changelog/#Version-0.21.6-(January-29,-2021)","page":"Release notes","title":"Version 0.21.6 (January 29, 2021)","text":"","category":"section"},{"location":"changelog/#Added-27","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for skew symmetric variables via @variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace()).\nlp_sensitivity_report has been added which significantly improves the performance of querying the sensitivity summary of an LP. lp_objective_perturbation_range and lp_rhs_perturbation_range are deprecated.\nDual warm-starts are now supported with set_dual_start_value and dual_start_value.\n∈ (\\in<tab>) can now be used in macros instead of = or in.\nUse haskey(model::Model, key::Symbol) to check if a name key is registered in a model.\nAdded unregister(model::Model, key::Symbol) to unregister a name key from model.\nAdded callback_node_status for use in callbacks.\nAdded print_bridge_graph to visualize the bridging graph generated by MathOptInterface.\nImproved error message for containers with duplicate indices.","category":"page"},{"location":"changelog/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Various fixes to pass tests on Julia 1.6.\nFixed a bug in the printing of nonlinear expressions in IJulia.\nFixed a bug when nonlinear expressions are passed to user-defined functions.\nSome internal functions that were previously exported are now no longer exported.\nFixed a bug when relaxing a fixed binary variable.\nFixed a StackOverflowError that occurred when SparseAxisArrays had a large number of elements.\nRemoved an unnecessary type assertion in list_of_constraint_types.\nFixed a bug when copying models with registered expressions.","category":"page"},{"location":"changelog/#Other-32","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in /examples have now been moved to /docs/src/examples and rewritten using Literate.jl, and they are now included in the documentation.\nJuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.\nThe root cause of a large number of method invalidations has been resolved.\nWe switched continuous integration from Travis and Appveyor to GitHub Actions.","category":"page"},{"location":"changelog/#Version-0.21.5-(September-18,-2020)","page":"Release notes","title":"Version 0.21.5 (September 18, 2020)","text":"","category":"section"},{"location":"changelog/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed deprecation warnings\nThrow DimensionMismatch for incompatibly sized functions and sets\nUnify treatment of keys(x) on JuMP containers","category":"page"},{"location":"changelog/#Version-0.21.4-(September-14,-2020)","page":"Release notes","title":"Version 0.21.4 (September 14, 2020)","text":"","category":"section"},{"location":"changelog/#Added-28","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add debug info when adding unsupported constraints\nAdd relax_integrality for solving continuous relaxation\nAllow querying constraint conflicts","category":"page"},{"location":"changelog/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Dispatch on Real for MOI.submit\nImplement copy for CustomSet in tests\nDon't export private macros\nFix invalid assertion in nonlinear\nError if constraint has NaN right-hand side\nImprove speed of tests\nLots of work modularizing files in /test\nImprove line numbers in macro error messages\nPrint nonlinear subexpressions\nVarious documentation updates\nDependency updates:\nDatastructures 0.18\nMathOptFormat v0.5\nPrep for MathOptInterface 0.9.15","category":"page"},{"location":"changelog/#Version-0.21.3-(June-18,-2020)","page":"Release notes","title":"Version 0.21.3 (June 18, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (#2212).\nAdded functions simplex_iterations, barrier_iterations and node_count (#2201).\nAdded function reduced_cost (#2205).\nImplemented callback_value for affine and quadratic expressions (#2231).\nSupport MutableArithmetics.Zero in objective and constraints (#2219).\nDocumentation improvements:\nMention tutorials in the docs (#2223).\nUpdate COIN-OR links (#2242).\nExplicit link to the documentation of MOI.FileFormats (#2253).\nTypo fixes (#2261).\nContainers improvements:\nFix Base.map for DenseAxisArray (#2235).\nThrow BoundsError if number of indices is incorrect for DenseAxisArray and SparseAxisArray (#2240).\nExtensibility improvements:\nImplement a set_objective method fallback that redirects to set_objective_sense and set_objective_function (#2247).\nAdd parse_constraint method with arbitrary number of arguments (#2051).\nAdd parse_constraint_expr and parse_constraint_head (#2228).","category":"page"},{"location":"changelog/#Version-0.21.2-(April-2,-2020)","page":"Release notes","title":"Version 0.21.2 (April 2, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added relative_gap() to access MOI.RelativeGap() attribute (#2199).\nDocumentation fixes:\nAdded link to source for docstrings in the documentation (#2207).\nAdded docstring for @variables macro (#2216).\nTypo fixes (#2177, #2184, #2182).\nImplementation of methods for Base functions:\nImplemented Base.empty! for JuMP.Model (#2198).\nImplemented Base.conj for JuMP scalar types (#2209).","category":"page"},{"location":"changelog/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed sum of expression with scalar product in macro (#2178).\nFixed writing of nonlinear models to MathOptFormat (#2181).\nFixed construction of empty SparseAxisArray (#2179).\nFixed constraint with zero function (#2188).","category":"page"},{"location":"changelog/#Version-0.21.1-(Feb-18,-2020)","page":"Release notes","title":"Version 0.21.1 (Feb 18, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the clarity of the with_optimizer deprecation warning.","category":"page"},{"location":"changelog/#Version-0.21.0-(Feb-16,-2020)","page":"Release notes","title":"Version 0.21.0 (Feb 16, 2020)","text":"","category":"section"},{"location":"changelog/#Breaking-6","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Deprecated with_optimizer (#2090, #2084, #2141). You can replace with_optimizer by either nothing, optimizer_with_attributes or a closure:\nreplace with_optimizer(Ipopt.Optimizer) by Ipopt.Optimizer.\nreplace with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0) by optimizer_with_attributes(Ipopt.Optimizer, \"max_cpu_time\" => 60.0).\nreplace with_optimizer(Gurobi.Optimizer, env) by () -> Gurobi.Optimizer(env).\nreplace with_optimizer(Gurobi.Optimizer, env, Presolve=0) by optimizer_with_attributes(() -> Gurobi.Optimizer(env), \"Presolve\" => 0).\nalternatively to optimizer_with_attributes, you can also set the attributes separately with set_optimizer_attribute.\nRenamed set_parameter and set_parameters to set_optimizer_attribute and set_optimizer_attributes (#2150).\nBroadcast should now be explicit inside macros. @SDconstraint(model, x >= 1) and @constraint(model, x + 1 in SecondOrderCone()) now throw an error instead of broadcasting 1 along the dimension of x (#2107).\n@SDconstraint(model, x >= 0) is now equivalent to @constraint(model, x in PSDCone()) instead of @constraint(model, (x .- 0) in PSDCone()) (#2107).\nThe macros now create the containers with map instead of for loops, as a consequence, containers created by @expression can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where @expression could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (#2070).\nCalling deepcopy(::AbstractModel) now throws an error.\nThe constraint name is now printed in the model string (#2108).","category":"page"},{"location":"changelog/#Added-29","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for solver-independent and solver-specific callbacks (#2101).\nAdded write_to_file and read_from_file, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (#2114).\nAdded support for complementarity constraints (#2132).\nAdded support for indicator constraints (#2092).\nAdded support for querying multiple solutions with the result keyword (#2100).\nAdded support for constraining variables on creation (#2128).\nAdded method delete that deletes a vector of variables at once if it is supported by the underlying solver (#2135).\nThe arithmetic between JuMP expression has be refactored into the MutableArithmetics package (#2107).\nImproved error on complex values in NLP (#1978).\nAdded an example of column generation (#2010).","category":"page"},{"location":"changelog/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Incorrect coefficients generated when using Symmetric variables (#2102)","category":"page"},{"location":"changelog/#Version-0.20.1-(Oct-18,-2019)","page":"Release notes","title":"Version 0.20.1 (Oct 18, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add sections on @variables and @constraints in the documentation (#2062).\nFixed product of sparse matrices for Julia v1.3 (#2063).\nAdded set_objective_coefficient to modify the coefficient of a linear term of the objective function (#2008).\nAdded set_time_limit_sec, unset_time_limit_sec and time_limit_sec to set and query the time limit for the solver in seconds (#2053).","category":"page"},{"location":"changelog/#Version-0.20.0-(Aug-24,-2019)","page":"Release notes","title":"Version 0.20.0 (Aug 24, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Documentation updates.\nNumerous bug fixes.\nBetter error messages (#1977, #1978, #1997, #2017).\nPerformance improvements (#1947, #2032).\nAdded LP sensitivity summary functions lp_objective_perturbation_range and lp_rhs_perturbation_range (#1917).\nAdded functions dual_objective_value, raw_status and set_parameter.\nAdded function set_objective_coefficient to modify the coefficient of a linear term of the objective (#2008).\nAdded functions set_normalized_rhs, normalized_rhs, and add_to_function_constant to modify and get the constant part of a constraint (#1935, #1960).\nAdded functions set_normalized_coefficient and normalized_coefficient to modify and get the coefficient of a linear term of a constraint (#1935, #1960).\nNumerous other improvements in MOI 0.9, see the NEWS.md file of MOI for more details.","category":"page"},{"location":"changelog/#Version-0.19.2-(June-8,-2019)","page":"Release notes","title":"Version 0.19.2 (June 8, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.","category":"page"},{"location":"changelog/#Version-0.19.1-(May-12,-2019)","page":"Release notes","title":"Version 0.19.1 (May 12, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Usability and performance improvements.\nBug fixes.","category":"page"},{"location":"changelog/#Version-0.19.0-(February-15,-2019)","page":"Release notes","title":"Version 0.19.0 (February 15, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP 0.19 contains significant breaking changes.","category":"page"},{"location":"changelog/#Breaking-7","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP's abstraction layer for communicating with solvers changed from MathProgBase (MPB) to MathOptInterface (MOI). MOI addresses many longstanding design issues. (See @mlubin's slides from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the installation guide for a list of solvers that have and have not yet been updated.\nMost solvers have been renamed to PackageName.Optimizer. For example, GurobiSolver() is now Gurobi.Optimizer.\nSolvers are no longer added to a model via Model(solver = XXX(kwargs...)). Instead use Model(with_optimizer(XXX, kwargs...)). For example, Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0)).\nJuMP containers (for example, the objects returned by @variable) have been redesigned. Containers.SparseAxisArray replaces JuMPDict, JuMPArray was rewritten (inspired by AxisArrays) and renamed Containers.DenseAxisArray, and you can now request a container type with the container= keyword to the macros. See the corresponding documentation for more details.\nThe statuses returned by solvers have changed. See the possible status values here. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).\nStarting values are separate from result values. Use value to query the value of a variable in a solution. Use start_value and set_start_value to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.\nThe data structures for affine and quadratic expressions AffExpr and QuadExpr have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.\nJuMPNLPEvaluator no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.\nJuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)\nThe norm() syntax is no longer recognized inside macros. Use the SecondOrderCone() set instead.\nJuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.\nThe symbols :Min and :Max are no longer used as optimization senses. Instead, JuMP uses the OptimizationSense enum from MathOptInterface. @objective(model, Max, ...), @objective(model, Min, ...), @NLobjective(model, Max, ...), and @objective(model, Min, ...) remain valid, but @objective(m, :Max, ...) is no longer accepted.\nThe sign conventions for duals has changed in some cases for consistency with conic duality (see the documentation). The shadow_price helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.\n@constraintref is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,\nconstraints = Dict() # Optionally, specify types for improved performance.\nfor i in 1:N\n  constraints[i] = @constraint(model, ...)\nend\nThe lowerbound, upperbound, and basename keyword arguments to the @variable macro have been renamed to lower_bound, upper_bound, and base_name, for consistency with JuMP's new style recommendations.\nWe rely on broadcasting syntax to apply accessors to collections of variables, for example, value.(x) instead of getvalue(x) for collections. (Use value(x) when x is a scalar object.)","category":"page"},{"location":"changelog/#Added-30","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Splatting (like f(x...)) is recognized in restricted settings in nonlinear expressions.\nSupport for deleting constraints and variables.\nThe documentation has been completely rewritten using docstrings and Documenter.\nSupport for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).\nSignificantly improved support for modeling new types of constraints and for extending JuMP's macros.\nSupport for providing dual warm starts.\nImproved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).\nExplicit control of whether symmetry-enforcing constraints are added to PSD constraints.\nSupport for modeling exponential cones.\nSignificant improvements in internal code quality and testing.\nStyle and naming guidelines.\nDirect mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding documentation.","category":"page"},{"location":"changelog/#Regressions","page":"Release notes","title":"Regressions","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Performance regressions in model generation (issue). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.\nFast incremental NLP solves are not yet reimplemented (issue).\nWe do not yet have an implementation of solver-specific callbacks.\nThe column generation syntax in @variable has been removed (that is, the objective, coefficients, and inconstraints keyword arguments). Support for column generation will be re-introduced in a future release.\nThe ability to solve the continuous relaxation (that is, via solve(model; relaxation = true)) is not yet reimplemented (issue).","category":"page"},{"location":"changelog/#Version-0.18.5-(December-1,-2018)","page":"Release notes","title":"Version 0.18.5 (December 1, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support views in some derivative evaluation functions.\nImproved compatibility with PackageCompiler.","category":"page"},{"location":"changelog/#Version-0.18.4-(October-8,-2018)","page":"Release notes","title":"Version 0.18.4 (October 8, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in model printing on Julia 0.7 and 1.0.","category":"page"},{"location":"changelog/#Version-0.18.3-(October-1,-2018)","page":"Release notes","title":"Version 0.18.3 (October 1, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for Julia v1.0 (Thanks @ExpandingMan)\nFix matrix expressions with quadratic functions (#1508)","category":"page"},{"location":"changelog/#Version-0.18.2-(June-10,-2018)","page":"Release notes","title":"Version 0.18.2 (June 10, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in second-order derivatives when expressions are present (#1319)\nFix a bug in @constraintref (#1330)","category":"page"},{"location":"changelog/#Version-0.18.1-(April-9,-2018)","page":"Release notes","title":"Version 0.18.1 (April 9, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for nested tuple destructuring (#1193)\nPreserve internal model when relaxation=true (#1209)\nMinor bug fixes and updates for example","category":"page"},{"location":"changelog/#Version-0.18.0-(July-27,-2017)","page":"Release notes","title":"Version 0.18.0 (July 27, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Drop support for Julia 0.5.\nUpdate for ForwardDiff 0.5.\nMinor bug fixes.","category":"page"},{"location":"changelog/#Version-0.17.1-(June-9,-2017)","page":"Release notes","title":"Version 0.17.1 (June 9, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Use of constructconstraint! in @SDconstraint.\nMinor bug fixes.","category":"page"},{"location":"changelog/#Version-0.17.0-(May-27,-2017)","page":"Release notes","title":"Version 0.17.0 (May 27, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: Mixing quadratic and conic constraints is no longer supported.\nBreaking change: The getvariable and getconstraint functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name x, one should now write m[:x] instead of getvariable(m, :x). As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.\nAddition of the getobjectivebound function that mirrors the functionality of the MathProgBase getobjbound function except that it takes into account transformations performed by JuMP.\nMinor bug fixes.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The following changes are primarily of interest to developers of JuMP extensions:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The new syntax @constraint(model, expr in Cone) creates the constraint ensuring that expr is inside Cone. The Cone argument is passed to constructconstraint! which enables the call to the dispatched to an extension.\nThe @variable macro now calls constructvariable! instead of directly calling the Variable constructor. Extra arguments and keyword arguments passed to @variable are passed to constructvariable! which enables the call to be dispatched to an extension.\nRefactor the internal function conicdata (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.","category":"page"},{"location":"changelog/#Version-0.16.2-(March-28,-2017)","page":"Release notes","title":"Version 0.16.2 (March 28, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Minor bug fixes and printing tweaks\nAddress deprecation warnings for Julia 0.6","category":"page"},{"location":"changelog/#Version-0.16.1-(March-7,-2017)","page":"Release notes","title":"Version 0.16.1 (March 7, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Better support for AbstractArray in JuMP (Thanks @tkoolen)\nMinor bug fixes","category":"page"},{"location":"changelog/#Version-0.16.0-(February-23,-2017)","page":"Release notes","title":"Version 0.16.0 (February 23, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling solve() will result in an error.\nBreaking change: User-defined functions are no longer global. The first argument to JuMP.register is now a JuMP Model object within whose scope the function will be registered. Calling JuMP.register without a Model now produces an error.\nBreaking change: Use the new JuMP.fix method to fix a variable to a value or to update the value to which a variable is fixed. Calling setvalue on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)\nNonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)\nNew category keyword to @variable. Used for specifying categories of anonymous variables.\nCompatibility with Julia 0.6-dev.\nMinor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)","category":"page"},{"location":"changelog/#Version-0.15.1-(January-31,-2017)","page":"Release notes","title":"Version 0.15.1 (January 31, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bugfix for @LinearConstraints and friends","category":"page"},{"location":"changelog/#Version-0.15.0-(December-22,-2016)","page":"Release notes","title":"Version 0.15.0 (December 22, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.5.0 is the minimum required version for this release.\nDocument support for BARON solver\nEnable info callbacks in more states than before, for example, for recording solutions. New when argument to addinfocallback (#814, thanks @yeesian)\nImproved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:\nWhen multiple variables in a model are given the same name\nWhen non-symbols are used as names, for example, @variable(m, x[1][1:N])\nImprovements in iterating over JuMP containers (#836, thanks @IssamT)\nSupport for writing variable names in .lp file output (Thanks @leethargo)\nSupport for querying duals to SDP problems (Thanks @blegat)\nThe comprehension syntax with curly braces sum{}, prod{}, and norm2{} has been deprecated in favor of Julia's native comprehension syntax sum(), prod() and norm() as previously announced. (For early adopters of the new syntax, norm2() was renamed to norm() without deprecation.)\nUnit tests rewritten to use Base.Test instead of FactCheck\nImproved support for operations with matrices of JuMP types (Thanks @ExpandingMan)\nThe syntax to halt a solver from inside a callback has changed from throw(CallbackAbort()) to return JuMP.StopTheSolver\nMinor bug fixes","category":"page"},{"location":"changelog/#Version-0.14.2-(December-12,-2016)","page":"Release notes","title":"Version 0.14.2 (December 12, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Allow singleton anonymous variables (includes bugfix)","category":"page"},{"location":"changelog/#Version-0.14.1-(September-12,-2016)","page":"Release notes","title":"Version 0.14.1 (September 12, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"More consistent handling of states in informational callbacks, includes a new when parameter to addinfocallback for specifying in which state an informational callback should be called.","category":"page"},{"location":"changelog/#Version-0.14.0-(August-7,-2016)","page":"Release notes","title":"Version 0.14.0 (August 7, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility with Julia 0.5 and ForwardDiff 0.2\nSupport for \"anonymous\" variables, constraints, expressions, and parameters, for example, x = @variable(m, [1:N]) instead of @variable(m, x[1:N])\nSupport for retrieving constraints from a model by name via getconstraint\n@NLconstraint now returns constraint references (as expected).\nSupport for vectorized expressions within lazy constraints\nOn Julia 0.5, parse new comprehension syntax sum(x[i] for i in 1:N if isodd(i)) instead of sum{ x[i], i in 1:N; isodd(i) }. The old syntax with curly braces will be deprecated in JuMP 0.15.\nNow possible to provide nonlinear expressions as \"raw\" Julia Expr objects instead of using JuMP's nonlinear macros. This input format is useful for programmatically generated expressions.\ns/Mathematical Programming/Mathematical Optimization/\nSupport for local cuts (Thanks to @madanim, Mehdi Madani)\nDocument Xpress interface developed by @joaquimg, Joaquim Dias Garcia\nMinor bug and deprecation fixes (Thanks @odow, @jrevels)","category":"page"},{"location":"changelog/#Version-0.13.2-(May-16,-2016)","page":"Release notes","title":"Version 0.13.2 (May 16, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility update for MathProgBase","category":"page"},{"location":"changelog/#Version-0.13.1-(May-3,-2016)","page":"Release notes","title":"Version 0.13.1 (May 3, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix broken deprecation for registerNLfunction.","category":"page"},{"location":"changelog/#Version-0.13.0-(April-29,-2016)","page":"Release notes","title":"Version 0.13.0 (April 29, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Most exported methods and macros have been renamed to avoid camelCase. See the list of changes here. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.\nSpecify variable lower/upper bounds in @variable using the lowerbound and upperbound keyword arguments.\nChange name printed for variable using the basename keyword argument to @variable.\nNew @variables macro allows multi-line declaration of groups of variables.\nA number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was recorded live.\nCompatibility fixes with Julia 0.5.\nThe \"end\" indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see here.","category":"page"},{"location":"changelog/#Version-0.12.2-(March-9,-2016)","page":"Release notes","title":"Version 0.12.2 (March 9, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Small fixes for nonlinear optimization","category":"page"},{"location":"changelog/#Version-0.12.1-(March-1,-2016)","page":"Release notes","title":"Version 0.12.1 (March 1, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)","category":"page"},{"location":"changelog/#Version-0.12.0-(February-27,-2016)","page":"Release notes","title":"Version 0.12.0 (February 27, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:\n@defExpr and @defNLExpr now take the model as the first argument. The previous one-argument version of @defExpr is deprecated; all expressions should be named. For example, replace @defExpr(2x+y) with @defExpr(jump_model, my_expr, 2x+y).\nJuMP no longer uses Julia's variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.\nSupport for user-defined functions integrated within nonlinear JuMP expressions.\nReplaced iteration over AffExpr with Number-like scalar iteration; previous iteration behavior is now available via linearterms(::AffExpr).\nStopping the solver via throw(CallbackAbort()) from a callback no longer triggers an exception. Instead, solve() returns UserLimit status.\ngetDual() now works for conic problems (Thanks @emreyamangil.)","category":"page"},{"location":"changelog/#Version-0.11.3-(February-4,-2016)","page":"Release notes","title":"Version 0.11.3 (February 4, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bug-fix for problems with quadratic objectives and semidefinite constraints","category":"page"},{"location":"changelog/#Version-0.11.2-(January-14,-2016)","page":"Release notes","title":"Version 0.11.2 (January 14, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility update for Mosek","category":"page"},{"location":"changelog/#Version-0.11.1-(December-1,-2015)","page":"Release notes","title":"Version 0.11.1 (December 1, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Remove usage of @compat in tests.\nFix updating quadratic objectives for nonlinear models.","category":"page"},{"location":"changelog/#Version-0.11.0-(November-30,-2015)","page":"Release notes","title":"Version 0.11.0 (November 30, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.4.0 is the minimum required version for this release.\nFix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.\nAddition of the solve(m::Model, relaxation=true) keyword argument to solve the standard continuous relaxation of model m\nThe getConstraintBounds() method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"changelog/#Version-0.10.3-(November-20,-2015)","page":"Release notes","title":"Version 0.10.3 (November 20, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a rare error when parsing quadratic expressions\nFix Variable() constructor with default arguments\nDetect unrecognized keywords in solve()","category":"page"},{"location":"changelog/#Version-0.10.2-(September-28,-2015)","page":"Release notes","title":"Version 0.10.2 (September 28, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for deprecation warnings","category":"page"},{"location":"changelog/#Version-0.10.1-(September-3,-2015)","page":"Release notes","title":"Version 0.10.1 (September 3, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixes for ambiguity warnings.\nFix for breaking change in precompilation syntax in Julia 0.4-pre","category":"page"},{"location":"changelog/#Version-0.10.0-(August-31,-2015)","page":"Release notes","title":"Version 0.10.0 (August 31, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support (on Julia 0.4 and later) for conditions in indexing @defVar and @addConstraint constructs, for example, @defVar(m, x[i=1:5,j=1:5; i+j >= 3])\nSupport for vectorized operations on Variables and expressions. See the documentation for details.\nNew getVar() method to access variables in a model by name\nSupport for semidefinite programming.\nDual solutions are now available for general nonlinear problems. You may call getDual on a reference object for a nonlinear constraint, and getDual on a variable object for Lagrange multipliers from active bounds.\nIntroduce warnings for two common performance traps: too many calls to getValue() on a collection of variables and use of the + operator in a loop to sum expressions.\nSecond-order cone constraints can be written directly with the norm() and norm2{} syntax.\nImplement MathProgBase interface for querying Hessian-vector products.\nIteration over JuMPContainers is deprecated; instead, use the keys and values functions, and zip(keys(d),values(d)) for the old behavior.\n@defVar returns Array{Variable,N} when each of N index sets are of the form 1:nᵢ.\nModule precompilation: on Julia 0.4 and later, using JuMP is now much faster.","category":"page"},{"location":"changelog/#Version-0.9.3-(August-11,-2015)","page":"Release notes","title":"Version 0.9.3 (August 11, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixes for FactCheck testing on julia v0.4.","category":"page"},{"location":"changelog/#Version-0.9.2-(June-27,-2015)","page":"Release notes","title":"Version 0.9.2 (June 27, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in @addConstraints.","category":"page"},{"location":"changelog/#Version-0.9.1-(April-25,-2015)","page":"Release notes","title":"Version 0.9.1 (April 25, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for Julia 0.4-dev.\nSmall infrastructure improvements for extensions.","category":"page"},{"location":"changelog/#Version-0.9.0-(April-18,-2015)","page":"Release notes","title":"Version 0.9.0 (April 18, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Comparison operators for constructing constraints (for example, 2x >= 1) have been deprecated. Instead, construct the constraints explicitly in the @addConstraint macro to add them to the model, or in the @LinearConstraint macro to create a stand-alone linear constraint instance.\ngetValue() method implemented to compute the value of a nonlinear subexpression\nJuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.\nA number of performance improvements in ReverseDiffSparse for computing derivatives.\nMathProgBase.getsolvetime(m) now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)\nFormatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).","category":"page"},{"location":"changelog/#Version-0.8.0-(February-17,-2015)","page":"Release notes","title":"Version 0.8.0 (February 17, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Nonlinear subexpressions now supported with the @defNLExpr macro.\nSCS supported for solving second-order conic problems.\nsetXXXCallback family deprecated in favor of addXXXCallback.\nMultiple callbacks of the same type can be registered.\nAdded support for informational callbacks via addInfoCallback.\nA CallbackAbort exception can be thrown from callback to safely exit optimization.","category":"page"},{"location":"changelog/#Version-0.7.4-(February-4,-2015)","page":"Release notes","title":"Version 0.7.4 (February 4, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.\nTwo-sided nonlinear constraints are supported.\nMethods for accessing the number of variables and constraints in a model are renamed.\nNew default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.\nSmall bug fixes.","category":"page"},{"location":"changelog/#Version-0.7.3-(January-14,-2015)","page":"Release notes","title":"Version 0.7.3 (January 14, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a method ambiguity conflict with Compose.jl (cosmetic fix)","category":"page"},{"location":"changelog/#Version-0.7.2-(January-9,-2015)","page":"Release notes","title":"Version 0.7.2 (January 9, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in sum(::JuMPDict)\nAdded the setCategory function to change a variables category (for example, continuous or binary)","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"after construction, and getCategory to retrieve the variable category.","category":"page"},{"location":"changelog/#Version-0.7.1-(January-2,-2015)","page":"Release notes","title":"Version 0.7.1 (January 2, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.","category":"page"},{"location":"changelog/#Version-0.7.0-(December-29,-2014)","page":"Release notes","title":"Version 0.7.0 (December 29, 2014)","text":"","category":"section"},{"location":"changelog/#Linear/quadratic/conic-programming","page":"Release notes","title":"Linear/quadratic/conic programming","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: The syntax for column-wise model generation has been changed to use keyword arguments in @defVar.\nOn Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.\nECOS supported for solving second-order conic problems.","category":"page"},{"location":"changelog/#_nonlinear_programming_release_notes","page":"Release notes","title":"Nonlinear programming","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for skipping model generation when solving a sequence of nonlinear models with changing data.\nFix a memory leak when solving a sequence of nonlinear models.\nThe @addNLConstraint macro now supports the three-argument version to define sets of nonlinear constraints.\nKNITRO supported as a nonlinear solver.\nSpeed improvements for model generation.\nThe @addNLConstraints macro supports adding multiple (groups of) constraints at once. Syntax is similar to @addConstraints.\nDiscrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).","category":"page"},{"location":"changelog/#General","page":"Release notes","title":"General","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Starting values for variables may now be specified with @defVar(m, x, start=value).\nThe setSolver function allows users to change the solver subsequent to model creation.\nSupport for \"fixed\" variables via the @defVar(m, x == 1) syntax.\nUnit tests rewritten to use FactCheck.jl, improved testing across solvers.","category":"page"},{"location":"changelog/#Version-0.6.3-(October-19,-2014)","page":"Release notes","title":"Version 0.6.3 (October 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in multiplying two AffExpr objects.","category":"page"},{"location":"changelog/#Version-0.6.2-(October-11,-2014)","page":"Release notes","title":"Version 0.6.2 (October 11, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Further improvements and bug fixes for printing.\nFixed a bug in @defExpr.\nSupport for accessing expression graphs through the MathProgBase NLP interface.","category":"page"},{"location":"changelog/#Version-0.6.1-(September-19,-2014)","page":"Release notes","title":"Version 0.6.1 (September 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improvements and bug fixes for printing.","category":"page"},{"location":"changelog/#Version-0.6.0-(September-9,-2014)","page":"Release notes","title":"Version 0.6.0 (September 9, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.3.0 is the minimum required version for this release.\nbuildInternalModel(m::Model) added to build solver-level model in memory without optimizing.\nDeprecate load_model_only keyword argument to solve.\nAdd groups of constraints with @addConstraints macro.\nUnicode operators now supported, including ∑ for sum, ∏ for prod, and ≤/≥\nQuadratic constraints supported in @addConstraint macro.\nQuadratic objectives supported in @setObjective macro.\nMathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems\nBreaking change: IpoptOptions no longer supported to specify solver options, use m = Model(solver=IpoptSolver(options...)) instead.\nNew solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK\nNew option to control whether the lazy constraint callback is executed at each node in the B&B tree or just when feasible solutions are found\nAdd support for semicontinuous and semi-integer variables for those solvers that support them.\nAdd support for index dependencies (for example, triangular indexing) in @defVar, @addConstraint, and @defExpr (for example, @defVar(m, x[i=1:10,j=i:10])).\nThis required some changes to the internal structure of JuMP containers, which may break code that explicitly stored JuMPDict objects.","category":"page"},{"location":"changelog/#Version-0.5.8-(September-24,-2014)","page":"Release notes","title":"Version 0.5.8 (September 24, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug with specifying solvers (affects Julia 0.2 only)","category":"page"},{"location":"changelog/#Version-0.5.7-(September-5,-2014)","page":"Release notes","title":"Version 0.5.7 (September 5, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in printing models","category":"page"},{"location":"changelog/#Version-0.5.6-(September-2,-2014)","page":"Release notes","title":"Version 0.5.6 (September 2, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for semicontinuous and semi-integer variables for those solvers that support them.\nBreaking change: Syntax for Variable() constructor has changed (use of this interface remains discouraged)\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"changelog/#Version-0.5.5-(July-6,-2014)","page":"Release notes","title":"Version 0.5.5 (July 6, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.","category":"page"},{"location":"changelog/#Version-0.5.4-(June-19,-2014)","page":"Release notes","title":"Version 0.5.4 (June 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for breaking change in MathProgBase which reduces loading times for using JuMP\nFix error when MIPs not solved to optimality","category":"page"},{"location":"changelog/#Version-0.5.3-(May-21,-2014)","page":"Release notes","title":"Version 0.5.3 (May 21, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for breaking change in ReverseDiffSparse","category":"page"},{"location":"changelog/#Version-0.5.2-(May-9,-2014)","page":"Release notes","title":"Version 0.5.2 (May 9, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix compatibility with Julia 0.3 prerelease","category":"page"},{"location":"changelog/#Version-0.5.1-(May-5,-2014)","page":"Release notes","title":"Version 0.5.1 (May 5, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in coefficient handling inside lazy constraints and user cuts","category":"page"},{"location":"changelog/#Version-0.5.0-(May-2,-2014)","page":"Release notes","title":"Version 0.5.0 (May 2, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.\ngetValue for AffExpr and QuadExpr\nBreaking change: getSolverModel replaced by getInternalModel, which returns the internal MathProgBase-level model\nGroups of constraints can be specified with @addConstraint (see documentation for details). This is not a breaking change.\ndot(::JuMPDict{Variable},::JuMPDict{Variable}) now returns the corresponding quadratic expression.","category":"page"},{"location":"changelog/#Version-0.4.1-(March-24,-2014)","page":"Release notes","title":"Version 0.4.1 (March 24, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug where change in objective sense was ignored when re-solving a model.\nFix issue with handling zero coefficients in AffExpr.","category":"page"},{"location":"changelog/#Version-0.4.0-(March-10,-2014)","page":"Release notes","title":"Version 0.4.0 (March 10, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for SOS1 and SOS2 constraints.\nSolver-independent callback for user heuristics.\ndot and sum implemented for JuMPDict objects. Now you can say @addConstraint(m, dot(a,x) <= b).\nDevelopers: support for extensions to JuMP. See definition of Model in src/JuMP.jl for more details.\nOption to construct the low-level model before optimizing.","category":"page"},{"location":"changelog/#Version-0.3.2-(February-17,-2014)","page":"Release notes","title":"Version 0.3.2 (February 17, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved model printing\nPreliminary support for IJulia output","category":"page"},{"location":"changelog/#Version-0.3.1-(January-30,-2014)","page":"Release notes","title":"Version 0.3.1 (January 30, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Documentation updates\nSupport for MOSEK\nCPLEXLink renamed to CPLEX","category":"page"},{"location":"changelog/#Version-0.3.0-(January-21,-2014)","page":"Release notes","title":"Version 0.3.0 (January 21, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Unbounded/infeasibility rays: getValue() will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. getDual() will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.\nSolver-independent callbacks for user generated cuts.\nUse new interface for solver-independent QCQP.\nsetlazycallback renamed to setLazyCallback for consistency.","category":"page"},{"location":"changelog/#Version-0.2.0-(December-15,-2013)","page":"Release notes","title":"Version 0.2.0 (December 15, 2013)","text":"","category":"section"},{"location":"changelog/#Breaking-8","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Objective sense is specified in setObjective instead of in the Model constructor.\nlpsolver and mipsolver merged into single solver option.","category":"page"},{"location":"changelog/#Added-31","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Problem modification with efficient LP restarts and MIP warm-starts.\nRelatedly, column-wise modeling now supported.\nSolver-independent callbacks supported. Currently we support only a \"lazy constraint\" callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.","category":"page"},{"location":"changelog/#Version-0.1.2-(November-16,-2013)","page":"Release notes","title":"Version 0.1.2 (November 16, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bug fixes for printing, improved error messages.\nAllow AffExpr to be used in macros; for example, ex = y + z; @addConstraint(m, x + 2*ex <= 3)","category":"page"},{"location":"changelog/#Version-0.1.1-(October-23,-2013)","page":"Release notes","title":"Version 0.1.1 (October 23, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for solver specification API changes in MathProgBase.","category":"page"},{"location":"changelog/#Version-0.1.0-(October-3,-2013)","page":"Release notes","title":"Version 0.1.0 (October 3, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Initial public release.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"EditURL = \"https://github.com/JuliaPolyhedra/CDDLib.jl/blob/9700752f849845bc8d8477928baf3df920daadd6/README.md\"","category":"page"},{"location":"packages/CDDLib/#CDDLib","page":"JuliaPolyhedra/CDDLib.jl","title":"CDDLib","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"(Image: ) (Image: ) (Image: )","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl is a wrapper for cddlib.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl can be used with C API of cddlib, the higher level interface of Polyhedra.jl, or as a linear programming solver with JuMP or MathOptInterface.","category":"page"},{"location":"packages/CDDLib/#Problem-description","page":"JuliaPolyhedra/CDDLib.jl","title":"Problem description","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"As written in the README of cddlib:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"The C-library  cddlib is a C implementation of the Double Description Method of Motzkin et al. for generating all vertices (that is, extreme points) and extreme rays of a general convex polyhedron in R^d given by a system of linear inequalities:P = { x=(x1, ..., xd)^T :  b - A  x  >= 0 }where  A  is a given m x d real matrix, b is a given m-vector and 0 is the m-vector of all zeros.The program can be used for the reverse operation (that is, convex hull computation).  This means that  one can move back and forth between an inequality representation  and a generator (that is, vertex and ray) representation of a polyhedron with cdd.  Also, cdd can solve a linear programming problem, that is, a problem of maximizing and minimizing a linear function over P.","category":"page"},{"location":"packages/CDDLib/#License","page":"JuliaPolyhedra/CDDLib.jl","title":"License","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl is licensed under the GPL v2 license.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"The underlying solver, cddlib/cddlib is also licensed under the GPL v2 license.","category":"page"},{"location":"packages/CDDLib/#Installation","page":"JuliaPolyhedra/CDDLib.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Install CDDLib.jl using the Julia package manager:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"import Pkg\nPkg.add(\"CDDLib\")","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Building the package will download binaries of cddlib that are provided by cddlib_jll.jl.","category":"page"},{"location":"packages/CDDLib/#Use-with-JuMP","page":"JuliaPolyhedra/CDDLib.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Use CDDLib.Optimizer{Float64} to use CDDLib.jl with JuMP:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"using JuMP, CDDLib\nmodel = Model(CDDLib.Optimizer{Float64})","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"When using CDDLib.jl with MathOptInterface, you can pass a different number type:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"using MathOptInterface, CDDLib\nmodel = CDDLib.Optimizer{Rational{BigInt}}()","category":"page"},{"location":"packages/CDDLib/#Debugging","page":"JuliaPolyhedra/CDDLib.jl","title":"Debugging","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl uses two global Boolean variables to enable debugging outputs: debug and log.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"You can query the value of debug and log with get_debug and get_log, and set their values with set_debug and set_log.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"EditURL = \"https://github.com/scipopt/SCIP.jl/blob/v0.11.14/README.md\"","category":"page"},{"location":"packages/SCIP/#SCIP.jl","page":"scipopt/SCIP.jl","title":"SCIP.jl","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"(Image: Build Status) (Image: codecov) (Image: Genie Downloads)","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"SCIP.jl is a Julia interface to the SCIP solver.","category":"page"},{"location":"packages/SCIP/#Affiliation","page":"scipopt/SCIP.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"This wrapper is maintained by the SCIP project with the help of the JuMP community.","category":"page"},{"location":"packages/SCIP/#License","page":"scipopt/SCIP.jl","title":"License","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"SCIP.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The underlying solver, scipopt/scip, is licensed under the Apache 2.0 license.","category":"page"},{"location":"packages/SCIP/#Installation","page":"scipopt/SCIP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Install SCIP.jl using Pkg.add:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"import Pkg\nPkg.add(\"SCIP\")","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"On MacOS and Linux, installing the SCIP Julia package will work out of the box and install the SCIP_jll.jl and SCIP_PaPILO_jll.jl dependencies.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"On Windows, a separate installation of SCIP is still mandatory, as detailed in the \"Custom installation\" section below.","category":"page"},{"location":"packages/SCIP/#Custom-installations","page":"scipopt/SCIP.jl","title":"Custom installations","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"If you use an older Julia version, Windows, or you want a custom SCIP installation, you must manually install the SCIP binaries.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Once installed, set the SCIPOPTDIR environment variable to point to the installation path, that is, depending on your operating system, $SCIPOPTDIR/lib/libscip.so, $SCIPOPTDIR/lib/libscip.dylib, or $SCIPOPTDIR/bin/scip.dll must exist.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Then, install SCIP.jl using Pkg.add and Pkg.build:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"ENV[\"SCIPOPTDIR\"] = \"/Users/Oscar/code/SCIP\"\nimport Pkg\nPkg.add(\"SCIP\")\nPkg.build(\"SCIP\")","category":"page"},{"location":"packages/SCIP/#Use-with-JuMP","page":"scipopt/SCIP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Use SCIP with JuMP as follows:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"using JuMP, SCIP\nmodel = Model(SCIP.Optimizer)\nset_attribute(model, \"display/verblevel\", 0)\nset_attribute(model, \"limits/gap\", 0.05)","category":"page"},{"location":"packages/SCIP/#Options","page":"scipopt/SCIP.jl","title":"Options","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"See the SCIP documentation for a list of supported options.","category":"page"},{"location":"packages/SCIP/#MathOptInterface-API","page":"scipopt/SCIP.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The SCIP optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.Reals","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.NLPBlock()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/SCIP/#Design-considerations","page":"scipopt/SCIP.jl","title":"Design considerations","text":"","category":"section"},{"location":"packages/SCIP/#Wrapping-the-public-API","page":"scipopt/SCIP.jl","title":"Wrapping the public API","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"All of the public API methods are wrapped and available within the SCIP package. This includes the scip_*.h and pub_*.h headers that are collected in scip.h, as well as all default constraint handlers (cons_*.h.) ","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The wrapped functions do not transform any data structures and work on the raw pointers (for example, SCIP* in C, Ptr{SCIP_} in Julia). Convenience wrapper functions based on Julia types are added as needed.","category":"page"},{"location":"packages/SCIP/#Memory-management","page":"scipopt/SCIP.jl","title":"Memory management","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Programming with SCIP requires dealing with variable and constraint objects that use reference counting for memory management.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The SCIP.Optimizer wrapper type collects lists of SCIP_VAR* and SCIP_CONS* under the hood, and it releases all references when it is garbage collected itself (via finalize).","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"When adding a variable (add_variable) or a constraint (add_linear_constraint), an integer index is returned. This index can be used to retrieve the SCIP_VAR* or SCIP_CONS* pointer via get_var and get_cons respectively.","category":"page"},{"location":"packages/SCIP/#Supported-nonlinear-operators","page":"scipopt/SCIP.jl","title":"Supported nonlinear operators","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Supported operators in nonlinear expressions are as follows:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"+\n-\n*\n/\n^\nsqrt\nexp\nlog\nabs\ncos\nsin","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"EditURL = \"nested_problems.jl\"","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Nested-optimization-problems","page":"Nested optimization problems","title":"Nested optimization problems","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"In this tutorial we explain how to write a nested optimization problem, where an upper problem uses the results from the optimization of a lower subproblem.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"To model the problem, we define a user-defined operator to handle the decomposition of the lower problem inside the upper one. Finally, we show how to improve the performance by using a cache that avoids resolving the lower problem.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"For a simpler example of writing a user-defined operator, see the User-defined Hessians tutorial.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"using JuMP\nimport Ipopt","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"In the rest of this tutorial, our goal is to solve the bilevel optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\nminlimits_xz  x_1^2 + x_2^2 + z \nst             beginarrayr l\n                      z = maxlimits_y  x_1^2 y_1 + x_2^2 y_2  - x_1 y_1^4 - 2 x_2 y_2^4 \n                      st                   (y_1 - 10)^2 + (y_2 - 10)^2 le 25\n                  endarray \n                 x ge 0\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This bilevel optimization problem is composed of two nested optimization problems. An upper level, involving variables x, and a lower level, involving variables y. From the perspective of the lower-level problem, the values of x are fixed parameters, and so the model optimizes y given those fixed parameters. Simultaneously, the upper-level problem optimizes x and z given the response of y.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Decomposition","page":"Nested optimization problems","title":"Decomposition","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"There are a few ways to solve this problem, but we are going to use a nonlinear decomposition method. The first step is to write a function to compute the lower-level problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\n  V(x_1 x_2) = maxlimits_y  x_1^2 y_1 + x_2^2 y_2  - x_1 y_1^4 - 2 x_2 y_2^4 \n                           st  (y_1 - 10)^2 + (y_2 - 10)^2 le 25\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function solve_lower_level(x...)\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, y[1:2])\n    @objective(\n        model,\n        Max,\n        x[1]^2 * y[1] + x[2]^2 * y[2] - x[1] * y[1]^4 - 2 * x[2] * y[2]^4,\n    )\n    @constraint(model, (y[1] - 10)^2 + (y[2] - 10)^2 <= 25)\n    optimize!(model)\n    @assert termination_status(model) == LOCALLY_SOLVED\n    return objective_value(model), value.(y)\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"The next function takes a value of x and returns the optimal lower-level objective-value and the optimal response y. The reason why we need both the objective and the optimal y will be made clear shortly, but for now let us define:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function V(x...)\n    f, _ = solve_lower_level(x...)\n    return f\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Then, we can substitute V into our full problem to create:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\nminlimits_x  x_1^2 + x_2^2 + V(x_1 x_2) \nst             x ge 0\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This looks like a nonlinear optimization problem with a user-defined operator V! However, because V solves an optimization problem internally, we can't use automatic differentiation to compute the first and second derivatives. Instead, we can use JuMP's ability to pass callback functions for the gradient and Hessian instead.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"First up, we need to define the gradient of V with respect to x. In general, this may be difficult to compute, but because x appears only in the objective, we can just differentiate the objective function with respect to x, giving:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function ∇V(g::AbstractVector, x...)\n    _, y = solve_lower_level(x...)\n    g[1] = 2 * x[1] * y[1] - y[1]^4\n    g[2] = 2 * x[2] * y[2] - 2 * y[2]^4\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Second, we need to define the Hessian of V with respect to x. This is a symmetric matrix, but in our example only the diagonal elements are non-zero:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function ∇²V(H::AbstractMatrix, x...)\n    _, y = solve_lower_level(x...)\n    H[1, 1] = 2 * y[1]\n    H[2, 2] = 2 * y[2]\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"info: Info\nProviding an explicit Hessian function is optional if first derivatives are already available.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"We now have enough to define our bilevel optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2] >= 0)\n@operator(model, op_V, 2, V, ∇V, ∇²V)\n@objective(model, Min, x[1]^2 + x[2]^2 + op_V(x[1], x[2]))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"The optimal objective value is:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"objective_value(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"and the optimal upper-level decision variables x are:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"value.(x)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"To compute the optimal lower-level decision variables, we need to call solve_lower_level with the optimal upper-level decision variables:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"_, y = solve_lower_level(value.(x)...)\ny","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Improving-performance","page":"Nested optimization problems","title":"Improving performance","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Our solution approach works, but it has a performance problem: every time we need to compute the value, gradient, or Hessian of V, we have to re-solve the lower-level optimization problem. This is wasteful, because we will often call the gradient and Hessian at the same point, and so solving the problem twice with the same input repeats work unnecessarily.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"We can work around this by using a cache:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"mutable struct Cache\n    x::Any\n    f::Float64\n    y::Vector{Float64}\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"with a function to update the cache if needed:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function _update_if_needed(cache::Cache, x...)\n    if cache.x !== x\n        cache.f, cache.y = solve_lower_level(x...)\n        cache.x = x\n    end\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Then, we define cached versions of out three functions which call _updated_if_needed and return values from the cache.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function cached_f(cache::Cache, x...)\n    _update_if_needed(cache, x...)\n    return cache.f\nend\n\nfunction cached_∇f(cache::Cache, g::AbstractVector, x...)\n    _update_if_needed(cache, x...)\n    g[1] = 2 * x[1] * cache.y[1] - cache.y[1]^4\n    g[2] = 2 * x[2] * cache.y[2] - 2 * cache.y[2]^4\n    return\nend\n\nfunction cached_∇²f(cache::Cache, H::AbstractMatrix, x...)\n    _update_if_needed(cache, x...)\n    H[1, 1] = 2 * cache.y[1]\n    H[2, 2] = 2 * cache.y[2]\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Now we're ready to setup and solve the upper level optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2] >= 0)\ncache = Cache(Float64[], NaN, Float64[])\n@operator(\n    model,\n    op_cached_f,\n    2,\n    (x...) -> cached_f(cache, x...),\n    (g, x...) -> cached_∇f(cache, g, x...),\n    (H, x...) -> cached_∇²f(cache, H, x...),\n)\n@objective(model, Min, x[1]^2 + x[2]^2 + op_cached_f(x[1], x[2]))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"an we can check we get the same objective value:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"objective_value(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"and upper-level decision variable x:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"value.(x)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"EditURL = \"https://github.com/jump-dev/PolyJuMP.jl/blob/6b280429865850cd278f0d74d0a176d53da2980e/README.md\"","category":"page"},{"location":"packages/PolyJuMP/#PolyJuMP.jl","page":"jump-dev/PolyJuMP.jl","title":"PolyJuMP.jl","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"PolyJuMP.jl is a JuMP extension for formulating and solving polynomial optimization problems. This extension includes the following:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Polynomial functions on JuMP decisions variables. These can be solved with the PolyJuMP.QCQP.Optimizer or PolyJuMP.KKT.Optimizer.\nConstraints that a polynomial is nonnegative where the coefficients of the polynomials depend on JuMP decision variables. These nonnegativity constraints can be reformulated using sufficient conditions using PolyJuMP.RelativeEntropy submodule or SumOfSquares.jl.","category":"page"},{"location":"packages/PolyJuMP/#License","page":"jump-dev/PolyJuMP.jl","title":"License","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"PolyJuMP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/PolyJuMP/#Installation","page":"jump-dev/PolyJuMP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Install PolyJuMP using Pkg.add:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import Pkg\nPkg.add(\"PolyJuMP\")","category":"page"},{"location":"packages/PolyJuMP/#Use-with-JuMP","page":"jump-dev/PolyJuMP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use QCQP solver with JuMP, use a nonconvex QCQP solver, for example Gurobi.Optimizer, and PolyJuMP.QCQP.Optimizer:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using JuMP, PolyJuMP, Gurobi\nmodel = Model(() -> PolyJuMP.QCQP.Optimizer(Gurobi.Optimizer))","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use KKT solver with JuMP, use solver of algebraic systems of equations implementing the SemialgebraicSets interface, for example HomotopyContinuation.SemialgebraicSetsHCSolver, and PolyJuMP.KKT.Optimizer:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using JuMP, PolyJuMP, HomotopyContinuation\nmodel = Model(optimizer_with_attributes(\n    PolyJuMP.KKT.Optimizer,\n    \"solver\" => HomotopyContinuation.SemialgebraicSetsHCSolver(),\n))","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"For a nonnegativity constraint on a polynomial, for example","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using DynamicPolynomials\n@polyvar x y\nusing JuMP\nmodel = Model()\n@variable(model, a)\n@constraint(model, a * x * y^2 + y^3 >= a * x)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"you need to specify how to interpret this nonnegativity constraint. To use Sum-of-Arithmetic-Geometric-Exponentials (SAGE), use","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import PolyJuMP\nPolyJuMP.setpolymodule!(model, PolyJuMP.SAGE)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use Sum-of-Squares (SOS), use","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import SumOfSquares\nPolyJuMP.setpolymodule!(model, SumOfSquares)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"or replace model = Model() by model = SOSModel().","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Alternatively, the nonnegativity constraint can be explicit:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"@constraint(model, a * x * y^2 + y^3 - a * x in PolyJuMP.SAGE.Polynomials())\n@constraint(model, a * x * y^2 + y^3 - a * x in SumOfSquares.SOSCone())","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"This allows mixing SAGE and SOS constraints in the same model.","category":"page"},{"location":"packages/PolyJuMP/#Documentation","page":"jump-dev/PolyJuMP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Documentation for PolyJuMP.jl is included in the documentation for SumOfSquares.jl.","category":"page"},{"location":"extensions/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"This section of the documentation contains brief documentation for some popular JuMP extensions. The list of extensions is not exhaustive, but instead is intended to help you discover popular JuMP extensions, and to give you an overview of the types of extensions that are possible to write with JuMP.","category":"page"},{"location":"extensions/introduction/#Affiliation","page":"Introduction","title":"Affiliation","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Packages beginning with jump-dev/ are developed and maintained by the JuMP developers.","category":"page"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Packages that do not begin with jump-dev/ are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.","category":"page"},{"location":"extensions/introduction/#Adding-new-extensions","page":"Introduction","title":"Adding new extensions","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Written an extension? Add it to this section of the JuMP documentation by making a pull request to the docs/packages.toml file.","category":"page"},{"location":"extensions/introduction/#Weak-dependencies","page":"Introduction","title":"Weak dependencies","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Some extensions listed in this section are implemented using the weak dependency feature added to Julia in v1.9. These extensions are activated if and only if you have JuMP and the other package loaded into your current scope with using or import.","category":"page"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"compat: Compat\nUsing a weak dependency requires Julia v1.9 or later.","category":"page"},{"location":"installation/#Installation-Guide","page":"Installation Guide","title":"Installation Guide","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This guide explains how to install Julia and  JuMP. If you have installation troubles, read the Common installation issues section below.","category":"page"},{"location":"installation/#Install-Julia","page":"Installation Guide","title":"Install Julia","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMP is a package for Julia. To use JuMP, first download and install Julia.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nIf you are new to Julia, read our Getting started with Julia tutorial.","category":"page"},{"location":"installation/#Choosing-a-version","page":"Installation Guide","title":"Choosing a version","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"You can install the \"Current stable release\" or the \"Long-term support (LTS) release.\"","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"The \"Current stable release\" is the latest release of Julia. It has access to newer features, and is likely faster.\nThe \"Long-term support release\" is an older version of Julia that has continued to receive bug and security fixes. However, it may not have the latest features or performance improvements.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"For most users, you should install the \"Current stable release,\" and whenever Julia releases a new version of the current stable release, you should update your version of Julia. Note that any code you write on one version of the current stable release will continue to work on all subsequent releases.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"For users in restricted software environments (for example, your enterprise IT controls what software you can install), you may be better off installing the long-term support release because you will not have to update Julia as frequently.","category":"page"},{"location":"installation/#Install-JuMP","page":"Installation Guide","title":"Install JuMP","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"From Julia, JuMP is installed using the built-in package manager:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nWe recommend you create a Pkg environment for each project you use JuMP for, instead of adding lots of packages to the global environment. The Pkg manager documentation has more information on this topic.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"When we release a new version of JuMP, you can update with:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.update(\"JuMP\")","category":"page"},{"location":"installation/#Install-a-solver","page":"Installation Guide","title":"Install a solver","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMP depends on solvers to solve optimization problems. Therefore, you will need to install one before you can solve problems with JuMP.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Install a solver using the Julia package manager, replacing \"Clp\" by the Julia package name as appropriate.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.add(\"Clp\")","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Once installed, you can use Clp as a solver with JuMP as follows, using set_attribute to set solver-specific options:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"using JuMP\nusing Clp\nmodel = Model(Clp.Optimizer)\nset_attribute(model, \"LogLevel\" => 1)\nset_attribute(model, \"PrimalTolerance\" => 1e-7)","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nMost packages follow the ModuleName.Optimizer naming convention, but exceptions may exist. See the README of the Julia package's GitHub repository for more details on how to use a particular solver, including any solver-specific options.","category":"page"},{"location":"installation/#Supported-solvers","page":"Installation Guide","title":"Supported solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Most solvers are not written in Julia, and some require commercial licenses to use, so installation is often more complex.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"If a solver has Manual in the Installation column, the solver requires a manual installation step, such as downloading and installing a binary, or obtaining a commercial license. Consult the README of the relevant Julia package for more information.\nIf the solver has Manualᴹ in the Installation column, the solver requires an installation of MATLAB.\nIf the Installation column is missing an entry, installing the Julia package will download and install any relevant solver binaries automatically, and you shouldn't need to do anything other than Pkg.add.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Solvers with a missing entry in the Julia Package column are written in Julia. The link in the Solver column is the corresponding Julia package.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Solver Julia Package Installation License Supports\nAlpine.jl   Triad NS (MI)NLP\nArtelys Knitro KNITRO.jl Manual Comm. (MI)LP, (MI)SOCP, (MI)NLP\nBARON BARON.jl Manual Comm. (MI)NLP\nBonmin AmplNLWriter.jl  EPL (MI)NLP\nCbc Cbc.jl  EPL (MI)LP\nCDCS CDCS.jl Manualᴹ GPL LP, SOCP, SDP\nCDD CDDLib.jl  GPL LP\nClarabel.jl   Apache LP, QP, SOCP, SDP\nClp Clp.jl  EPL LP\nCOPT COPT.jl  Comm. (MI)LP, SOCP, SDP\nCOSMO.jl   Apache LP, QP, SOCP, SDP\nCouenne AmplNLWriter.jl  EPL (MI)NLP\nCPLEX CPLEX.jl Manual Comm. (MI)LP, (MI)SOCP\nCSDP CSDP.jl  EPL LP, SDP\nDAQP DAQP.jl  MIT (Mixed-binary) QP\nEAGO.jl   MIT (MI)NLP\nECOS ECOS.jl  GPL LP, SOCP\nFICO Xpress Xpress.jl Manual Comm. (MI)LP, (MI)SOCP\nGLPK GLPK.jl  GPL (MI)LP\nGurobi Gurobi.jl Manual Comm. (MI)LP, (MI)SOCP\nHiGHS HiGHS.jl  MIT (MI)LP, QP\nHypatia.jl   MIT LP, SOCP, SDP\nIpopt Ipopt.jl  EPL LP, QP, NLP\nJuniper.jl   MIT (MI)SOCP, (MI)NLP\nLoraine.jl   MIT LP, SDP\nMadNLP.jl   MIT LP, QP, NLP\nMiniZinc MiniZinc.jl Manual MPL-2 CP-SAT\nMinotaur AmplNLWriter.jl Manual BSD-like (MI)NLP\nMOSEK MosekTools.jl Manual Comm. (MI)LP, (MI)SOCP, SDP\nNLopt NLopt.jl  GPL LP, QP, NLP\nOcteract AmplNLWriter.jl  Comm. (MI)NLP\nOSQP OSQP.jl  Apache LP, QP\nPATH PATHSolver.jl  MIT MCP\nPajarito.jl   MPL-2 (MI)NLP, (MI)SOCP, (MI)SDP\nPavito.jl   MPL-2 (MI)NLP\nPenbmi Penopt.jl  Comm. Bilinear SDP\nPolyJuMP.KKT PolyJuMP.jl  MIT NLP\nPolyJuMP.QCQP PolyJuMP.jl  MIT NLP\nProxSDP.jl   MIT LP, SOCP, SDP\nRAPOSa AmplNLWriter.jl Manual RAPOSa (MI)NLP\nSCIP SCIP.jl  Apache (MI)LP, (MI)NLP\nSCS SCS.jl  MIT LP, QP, SOCP, SDP\nSDPA SDPA.jl, SDPAFamily.jl  GPL LP, SDP\nSDPNAL SDPNAL.jl Manualᴹ CC BY-SA LP, SDP\nSDPT3 SDPT3.jl Manualᴹ GPL LP, SOCP, SDP\nSeDuMi SeDuMi.jl Manualᴹ GPL LP, SOCP, SDP\nStatusSwitchingQP.jl   MIT LP, QP\nTulip.jl   MPL-2 LP","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Where:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"LP = Linear programming\nQP = Quadratic programming\nSOCP = Second-order conic programming (including problems with convex quadratic constraints or objective)\nMCP = Mixed-complementarity programming\nNLP = Nonlinear programming\nSDP = Semidefinite programming\n(MI)XXX = Mixed-integer equivalent of problem type XXX\nCP-SAT = Constraint programming and Boolean satisfiability","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nDeveloped a solver or solver wrapper? This table is open for new contributions. Edit the installation.md file, and use the checklist Adding a new solver to the documentation when opening the pull request.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nDeveloping a solver or solver wrapper? See Models and the MathOptInterface docs for more details on how JuMP interacts with solvers. Please get in touch via the Developer Chatroom with any questions about connecting new solvers with JuMP.","category":"page"},{"location":"installation/#AMPL-based-solvers","page":"Installation Guide","title":"AMPL-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use AmplNLWriter to access solvers that support the NL format.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Some solvers, such as Bonmin and Couenne can be installed via the Julia package manager. Others need to be manually installed.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Consult the AMPL documentation for a complete list of supported solvers.","category":"page"},{"location":"installation/#GAMS-based-solvers","page":"Installation Guide","title":"GAMS-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use GAMS.jl to access solvers available through GAMS. Such solvers include: AlphaECP, Antigone, BARON, CONOPT, Couenne, LocalSolver, PATHNLP, SHOT, SNOPT, SoPlex. See a complete list here.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nGAMS.jl requires an installation of the commercial software GAMS for which a free community license exists.","category":"page"},{"location":"installation/#NEOS-based-solvers","page":"Installation Guide","title":"NEOS-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use NEOSServer.jl to access solvers available through the NEOS Server.","category":"page"},{"location":"installation/#Common-installation-issues","page":"Installation Guide","title":"Common installation issues","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nWhen in doubt, run import Pkg; Pkg.update() to see if updating your packages fixes the issue. Remember you will need to exit Julia and start a new session for the changes to take effect.","category":"page"},{"location":"installation/#Check-the-version-of-your-packages","page":"Installation Guide","title":"Check the version of your packages","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Each package is versioned with a three-part number of the form vX.Y.Z. You can check which versions you have installed with import Pkg; Pkg.status().","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This should almost always be the most-recent release. You can check the releases of a package by going to the relevant GitHub page, and navigating to the \"releases\" page. For example, the list of JuMP releases is available at: https://github.com/jump-dev/JuMP.jl/releases.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"If you post on the community forum, please include the output of Pkg.status().","category":"page"},{"location":"installation/#Unsatisfiable-requirements-detected","page":"Installation Guide","title":"Unsatisfiable requirements detected","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Did you get an error like Unsatisfiable requirements detected for package JuMP? The Pkg documentation has a section on how to understand and manage these conflicts.","category":"page"},{"location":"installation/#Installing-new-packages-can-make-JuMP-downgrade-to-an-earlier-version","page":"Installation Guide","title":"Installing new packages can make JuMP downgrade to an earlier version","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Another common complaint is that after adding a new package, code that previously worked no longer works.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This usually happens because the new package is not compatible with the latest version of JuMP. Therefore, the package manager rolls-back JuMP to an earlier version. Here's an example.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"First, we add JuMP:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"(jump_example) pkg> add JuMP\n  Resolving package versions...\nUpdating `~/jump_example/Project.toml`\n  [4076af6c] + JuMP v0.21.5\nUpdating `~/jump_example/Manifest.toml`\n  ... lines omitted ...","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"The + JuMP v0.21.5 line indicates that JuMP has been added at version 0.21.5. However, watch what happens when we add JuMPeR:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"(jump_example) pkg> add JuMPeR\n  Resolving package versions...\nUpdating `~/jump_example/Project.toml`\n  [4076af6c] ↓ JuMP v0.21.5 ⇒ v0.18.6\n  [707a9f91] + JuMPeR v0.6.0\nUpdating `~/jump_example/Manifest.toml`\n  ... lines omitted ...","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMPeR gets added at version 0.6.0 (+ JuMPeR v0.6.0), but JuMP gets downgraded from 0.21.5 to 0.18.6 (↓ JuMP v0.21.5 ⇒ v0.18.6)! The reason for this is that JuMPeR doesn't support a version of JuMP newer than 0.18.6.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nPay careful attention to the output of the package manager when adding new packages, especially when you see a package being downgraded.","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/errors.md\"","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/errors/#Errors","page":"Errors","title":"Errors","text":"","category":"section"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an MOI call fails on a model, precise errors should be thrown when possible instead of simply calling error with a message. The docstrings for the respective methods describe the errors that the implementation should throw in certain situations. This error-reporting system allows code to distinguish between internal errors (that should be shown to the user) and unsupported operations which may have automatic workarounds.","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an invalid index is used in an MOI call, an InvalidIndex is thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"InvalidIndex","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.InvalidIndex","page":"Errors","title":"MathOptInterface.InvalidIndex","text":"struct InvalidIndex{IndexType<:Index} <: Exception\n    index::IndexType\nend\n\nAn error indicating that the index index is invalid.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an invalid result index is used to retrieve an attribute, a ResultIndexBoundsError is thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"ResultIndexBoundsError\ncheck_result_index_bounds","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.ResultIndexBoundsError","page":"Errors","title":"MathOptInterface.ResultIndexBoundsError","text":"struct ResultIndexBoundsError{AttrType} <: Exception\n    attr::AttrType\n    result_count::Int\nend\n\nAn error indicating that the requested attribute attr could not be retrieved, because the solver returned too few results compared to what was requested. For instance, the user tries to retrieve VariablePrimal(2) when only one solution is available, or when the model is infeasible and has no solution.\n\nSee also: check_result_index_bounds.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.check_result_index_bounds","page":"Errors","title":"MathOptInterface.check_result_index_bounds","text":"check_result_index_bounds(model::ModelLike, attr)\n\nThis function checks whether enough results are available in the model for the requested attr, using its result_index field. If the model does not have sufficient results to answer the query, it throws a ResultIndexBoundsError.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"As discussed in JuMP mapping, for scalar constraint with a nonzero function constant, a ScalarFunctionConstantNotZero exception may be thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"ScalarFunctionConstantNotZero","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.ScalarFunctionConstantNotZero","page":"Errors","title":"MathOptInterface.ScalarFunctionConstantNotZero","text":"struct ScalarFunctionConstantNotZero{T, F, S} <: Exception\n    constant::T\nend\n\nAn error indicating that the constant part of the function in the constraint F-in-S is nonzero.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Some VariableIndex constraints cannot be combined on the same variable:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"LowerBoundAlreadySet\nUpperBoundAlreadySet","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.LowerBoundAlreadySet","page":"Errors","title":"MathOptInterface.LowerBoundAlreadySet","text":"LowerBoundAlreadySet{S1, S2}\n\nError thrown when setting a VariableIndex-in-S2 when a VariableIndex-in-S1 has already been added and the sets S1, S2 both set a lower bound, i.e. they are EqualTo, GreaterThan, Interval, Semicontinuous or Semiinteger.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UpperBoundAlreadySet","page":"Errors","title":"MathOptInterface.UpperBoundAlreadySet","text":"UpperBoundAlreadySet{S1, S2}\n\nError thrown when setting a VariableIndex-in-S2 when a VariableIndex-in-S1 has already been added and the sets S1, S2 both set an upper bound, i.e. they are EqualTo, LessThan, Interval, Semicontinuous or Semiinteger.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"As discussed in AbstractCallback, trying to get attributes inside a callback may throw:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"OptimizeInProgress","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.OptimizeInProgress","page":"Errors","title":"MathOptInterface.OptimizeInProgress","text":"struct OptimizeInProgress{AttrType<:AnyAttribute} <: Exception\n    attr::AttrType\nend\n\nError thrown from optimizer when MOI.get(optimizer, attr) is called inside an AbstractCallback while it is only defined once optimize! has completed. This can only happen when is_set_by_optimize(attr) is true.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Trying to submit the wrong type of AbstractSubmittable inside an AbstractCallback (for example, a UserCut inside a LazyConstraintCallback) will throw:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"InvalidCallbackUsage","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.InvalidCallbackUsage","page":"Errors","title":"MathOptInterface.InvalidCallbackUsage","text":"struct InvalidCallbackUsage{C, S} <: Exception\n    callback::C\n    submittable::S\nend\n\nAn error indicating that submittable cannot be submitted inside callback.\n\nFor example, UserCut cannot be submitted inside LazyConstraintCallback.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"The rest of the errors defined in MOI fall in two categories represented by the following two abstract types:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"UnsupportedError\nNotAllowedError","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedError","page":"Errors","title":"MathOptInterface.UnsupportedError","text":"UnsupportedError <: Exception\n\nAbstract type for error thrown when an element is not supported by the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.NotAllowedError","page":"Errors","title":"MathOptInterface.NotAllowedError","text":"NotAllowedError <: Exception\n\nAbstract type for error thrown when an operation is supported but cannot be applied in the current state of the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"The different UnsupportedError and NotAllowedError are the following errors:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"UnsupportedAttribute\nSetAttributeNotAllowed\nAddVariableNotAllowed\nUnsupportedConstraint\nAddConstraintNotAllowed\nModifyConstraintNotAllowed\nModifyObjectiveNotAllowed\nDeleteNotAllowed\nUnsupportedSubmittable\nSubmitNotAllowed\nUnsupportedNonlinearOperator","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedAttribute","page":"Errors","title":"MathOptInterface.UnsupportedAttribute","text":"struct UnsupportedAttribute{AttrType} <: UnsupportedError\n    attr::AttrType\n    message::String\nend\n\nAn error indicating that the attribute attr is not supported by the model, i.e. that supports returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.SetAttributeNotAllowed","page":"Errors","title":"MathOptInterface.SetAttributeNotAllowed","text":"struct SetAttributeNotAllowed{AttrType} <: NotAllowedError\n    attr::AttrType\n    message::String\nend\n\nAn error indicating that the attribute attr is supported (see supports) but cannot be set for some reason (see the error string).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.AddVariableNotAllowed","page":"Errors","title":"MathOptInterface.AddVariableNotAllowed","text":"struct AddVariableNotAllowed <: NotAllowedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that variables cannot be added to the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedConstraint","page":"Errors","title":"MathOptInterface.UnsupportedConstraint","text":"struct UnsupportedConstraint{F<:AbstractFunction, S<:AbstractSet} <: UnsupportedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that constraints of type F-in-S are not supported by the model, i.e. that supports_constraint returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.AddConstraintNotAllowed","page":"Errors","title":"MathOptInterface.AddConstraintNotAllowed","text":"struct AddConstraintNotAllowed{F<:AbstractFunction, S<:AbstractSet} <: NotAllowedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that constraints of type F-in-S are supported (see supports_constraint) but cannot be added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.ModifyConstraintNotAllowed","page":"Errors","title":"MathOptInterface.ModifyConstraintNotAllowed","text":"struct ModifyConstraintNotAllowed{F<:AbstractFunction, S<:AbstractSet,\n                                  C<:AbstractFunctionModification} <: NotAllowedError\n    constraint_index::ConstraintIndex{F, S}\n    change::C\n    message::String\nend\n\nAn error indicating that the constraint modification change cannot be applied to the constraint of index ci.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.ModifyObjectiveNotAllowed","page":"Errors","title":"MathOptInterface.ModifyObjectiveNotAllowed","text":"struct ModifyObjectiveNotAllowed{C<:AbstractFunctionModification} <: NotAllowedError\n    change::C\n    message::String\nend\n\nAn error indicating that the objective modification change cannot be applied to the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.DeleteNotAllowed","page":"Errors","title":"MathOptInterface.DeleteNotAllowed","text":"struct DeleteNotAllowed{IndexType <: Index} <: NotAllowedError\n    index::IndexType\n    message::String\nend\n\nAn error indicating that the index index cannot be deleted.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedSubmittable","page":"Errors","title":"MathOptInterface.UnsupportedSubmittable","text":"struct UnsupportedSubmittable{SubmitType} <: UnsupportedError\n    sub::SubmitType\n    message::String\nend\n\nAn error indicating that the submittable sub is not supported by the model, i.e. that supports returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.SubmitNotAllowed","page":"Errors","title":"MathOptInterface.SubmitNotAllowed","text":"struct SubmitNotAllowed{SubmitTyp<:AbstractSubmittable} <: NotAllowedError\n    sub::SubmitType\n    message::String\nend\n\nAn error indicating that the submittable sub is supported (see supports) but cannot be added for some reason (see the error string).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedNonlinearOperator","page":"Errors","title":"MathOptInterface.UnsupportedNonlinearOperator","text":"UnsupportedNonlinearOperator(head::Symbol[, message::String]) <: UnsupportedError\n\nAn error thrown by optimizers if they do not support the operator head in a ScalarNonlinearFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> throw(MOI.UnsupportedNonlinearOperator(:black_box))\nERROR: MathOptInterface.UnsupportedNonlinearOperator: The nonlinear operator `:black_box` is not supported by the model.\nStacktrace:\n[...]\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Note that setting the ConstraintFunction of a VariableIndex constraint is not allowed:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"SettingVariableIndexNotAllowed","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.SettingVariableIndexNotAllowed","page":"Errors","title":"MathOptInterface.SettingVariableIndexNotAllowed","text":"SettingVariableIndexNotAllowed()\n\nError type that should be thrown when the user calls set to change the ConstraintFunction of a VariableIndex constraint.\n\n\n\n\n\n","category":"type"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"EditURL = \"https://github.com/jump-dev/CPLEX.jl/blob/v1.0.0/README.md\"","category":"page"},{"location":"packages/CPLEX/#CPLEX.jl","page":"jump-dev/CPLEX.jl","title":"CPLEX.jl","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl is a wrapper for the IBM® ILOG® CPLEX® Optimization Studio.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl has two components:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The C API can be accessed via CPLEX.CPXxx functions, where the names and arguments are identical to the C API. See the CPLEX documentation for details.","category":"page"},{"location":"packages/CPLEX/#Affiliation","page":"jump-dev/CPLEX.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"This wrapper is maintained by the JuMP community and is not officially supported by IBM. However, we thank IBM for providing us with a CPLEX license to test CPLEX.jl on GitHub. If you are a commercial customer interested in official support for CPLEX in Julia, let them know.","category":"page"},{"location":"packages/CPLEX/#License","page":"jump-dev/CPLEX.jl","title":"License","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl is licensed under the MIT License.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Free CPLEX licenses are available for academics and students.","category":"page"},{"location":"packages/CPLEX/#Installation","page":"jump-dev/CPLEX.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl requires CPLEX version 12.10, 20.1, or 22.1.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"First, obtain a license of CPLEX and install CPLEX solver, following the instructions on IBM's website.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Once installed, set the CPLEX_STUDIO_BINARIES environment variable as appropriate and run Pkg.add(\"CPLEX\"). For example:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"# On Windows, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"C:\\\\Program Files\\\\CPLEX_Studio1210\\\\cplex\\\\bin\\\\x86-64_win\\\\\"\n# On OSX, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"/Applications/CPLEX_Studio1210/cplex/bin/x86-64_osx/\"\n# On Unix, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"/opt/CPLEX_Studio1210/cplex/bin/x86-64_linux/\"\n\nimport Pkg\nPkg.add(\"CPLEX\")","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"note: Note\nThe exact path may differ. Check which folder you installed CPLEX in, and update the path accordingly.","category":"page"},{"location":"packages/CPLEX/#Use-with-JuMP","page":"jump-dev/CPLEX.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Use CPLEX.jl with JuMP as follows:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX\nmodel = Model(CPLEX.Optimizer)\nset_attribute(model, \"CPX_PARAM_EPINT\", 1e-8)","category":"page"},{"location":"packages/CPLEX/#MathOptInterface-API","page":"jump-dev/CPLEX.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The CPLEX optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}\nMOI.ObjectiveFunction{MOI.VectorAffineFunction{Float64}}","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.Reals","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ConflictStatus()\nMOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/CPLEX/#Options","page":"jump-dev/CPLEX.jl","title":"Options","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Options match those of the C API in the CPLEX documentation.","category":"page"},{"location":"packages/CPLEX/#Callbacks","page":"jump-dev/CPLEX.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl provides a solver-specific callback to CPLEX:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX, Test\n\nmodel = direct_model(CPLEX.Optimizer())\nset_silent(model)\n\n# This is very, very important!!! Only use callbacks in single-threaded mode.\nMOI.set(model, MOI.NumberOfThreads(), 1)\n\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\ncb_calls = Clong[]\nfunction my_callback_function(cb_data::CPLEX.CallbackContext, context_id::Clong)\n    # You can reference variables outside the function as normal\n    push!(cb_calls, context_id)\n    # You can select where the callback is run\n    if context_id != CPX_CALLBACKCONTEXT_CANDIDATE\n        return\n    end\n    ispoint_p = Ref{Cint}()\n    ret = CPXcallbackcandidateispoint(cb_data, ispoint_p)\n    if ret != 0 || ispoint_p[] == 0\n        return  # No candidate point available or error\n    end\n    # You can query CALLBACKINFO items\n    valueP = Ref{Cdouble}()\n    ret = CPXcallbackgetinfodbl(cb_data, CPXCALLBACKINFO_BEST_BND, valueP)\n    @info \"Best bound is currently: $(valueP[])\"\n    # As well as any other C API\n    x_p = Vector{Cdouble}(undef, 2)\n    obj_p = Ref{Cdouble}()\n    ret = CPXcallbackgetincumbent(cb_data, x_p, 0, 1, obj_p)\n    if ret == 0\n        @info \"Objective incumbent is: $(obj_p[])\"\n        @info \"Incumbent solution is: $(x_p)\"\n        # Use CPLEX.column to map between variable references and the 1-based\n        # column.\n        x_col = CPLEX.column(cb_data, index(x))\n        @info \"x = $(x_p[x_col])\"\n    else\n        # Unable to query incumbent.\n    end\n\n    # Before querying `callback_value`, you must call:\n    CPLEX.load_callback_variable_primal(cb_data, context_id)\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    # You can submit solver-independent MathOptInterface attributes such as\n    # lazy constraints, user-cuts, and heuristic solutions.\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y + x <= 3)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nMOI.set(model, CPLEX.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2","category":"page"},{"location":"packages/CPLEX/#Annotations-for-automatic-Benders'-decomposition","page":"jump-dev/CPLEX.jl","title":"Annotations for automatic Benders' decomposition","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Here is an example of using the annotation feature for automatic Benders' decomposition:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX\n\nfunction add_annotation(\n    model::JuMP.Model,\n    variable_classification::Dict;\n    all_variables::Bool = true,\n)\n    num_variables = sum(length(it) for it in values(variable_classification))\n    if all_variables\n        @assert num_variables == JuMP.num_variables(model)\n    end\n    indices, annotations = CPXINT[], CPXLONG[]\n    for (key, value) in variable_classification\n        for variable_ref in value\n            push!(indices, variable_ref.index.value - 1)\n            push!(annotations, CPX_BENDERS_MASTERVALUE + key)\n        end\n    end\n    cplex = backend(model)\n    index_p = Ref{CPXINT}()\n    CPXnewlongannotation(\n        cplex.env,\n        cplex.lp,\n        CPX_BENDERS_ANNOTATION,\n        CPX_BENDERS_MASTERVALUE,\n    )\n    CPXgetlongannotationindex(\n        cplex.env,\n        cplex.lp,\n        CPX_BENDERS_ANNOTATION,\n        index_p,\n    )\n    CPXsetlongannotations(\n        cplex.env,\n        cplex.lp,\n        index_p[],\n        CPX_ANNOTATIONOBJ_COL,\n        length(indices),\n        indices,\n        annotations,\n    )\n    return\nend\n\n# Problem\n\nfunction illustrate_full_annotation()\n    c_1, c_2 = [1, 4], [2, 3]\n    dim_x, dim_y = length(c_1), length(c_2)\n    b = [-2; -3]\n    A_1, A_2 = [1 -3; -1 -3], [1 -2; -1 -1]\n    model = JuMP.direct_model(CPLEX.Optimizer())\n    set_optimizer_attribute(model, \"CPXPARAM_Benders_Strategy\", 1)\n    @variable(model, x[1:dim_x] >= 0, Bin)\n    @variable(model, y[1:dim_y] >= 0)\n    variable_classification = Dict(0 => [x[1], x[2]], 1 => [y[1], y[2]])\n    @constraint(model, A_2 * y + A_1 * x .<= b)\n    @objective(model, Min, c_1' * x + c_2' * y)\n    add_annotation(model, variable_classification)\n    optimize!(model)\n    x_optimal = value.(x)\n    y_optimal = value.(y)\n    println(\"x: $(x_optimal), y: $(y_optimal)\")\nend\n\nfunction illustrate_partial_annotation()\n    c_1, c_2 = [1, 4], [2, 3]\n    dim_x, dim_y = length(c_1), length(c_2)\n    b = [-2; -3]\n    A_1, A_2 = [1 -3; -1 -3], [1 -2; -1 -1]\n    model = JuMP.direct_model(CPLEX.Optimizer())\n    # Note that the \"CPXPARAM_Benders_Strategy\" has to be set to 2 if partial\n    # annotation is provided. If \"CPXPARAM_Benders_Strategy\" is set to 1, then\n    # the following error will be thrown:\n    # `CPLEX Error  2002: Invalid Benders decomposition.`\n    set_optimizer_attribute(model, \"CPXPARAM_Benders_Strategy\", 2)\n    @variable(model, x[1:dim_x] >= 0, Bin)\n    @variable(model, y[1:dim_y] >= 0)\n    variable_classification = Dict(0 => [x[1]], 1 => [y[1], y[2]])\n    @constraint(model, A_2 * y + A_1 * x .<= b)\n    @objective(model, Min, c_1' * x + c_2' * y)\n    add_annotation(model, variable_classification; all_variables = false)\n    optimize!(model)\n    x_optimal = value.(x)\n    y_optimal = value.(y)\n    println(\"x: $(x_optimal), y: $(y_optimal)\")\nend","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"EditURL = nothing","category":"page"},{"location":"api/JuMP/#DocumenterReference_JuMP","page":"JuMP","title":"JuMP","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"This page lists the public API of JuMP.","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"info: Info\nThis page is an unstructured list of the JuMP API. For a more structured overview, read the Manual or Tutorial parts of this documentation.","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"Load all of the public the API into the current scope with:","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"using JuMP","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"Alternatively, load only the module with:","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"import JuMP","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"and then prefix all calls with JuMP. to create JuMP.<NAME>.","category":"page"},{"location":"api/JuMP/#@NLconstraint","page":"JuMP","title":"@NLconstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLconstraint","category":"page"},{"location":"api/JuMP/#JuMP.@NLconstraint","page":"JuMP","title":"JuMP.@NLconstraint","text":"@NLconstraint(model::GenericModel, expr)\n\nAdd a constraint described by the nonlinear expression expr. See also @constraint. For example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @NLconstraint(model, sin(x) <= 1)\nsin(x) - 1.0 ≤ 0\n\njulia> @NLconstraint(model, [i = 1:3], sin(i * x) <= 1 / i)\n3-element Vector{NonlinearConstraintRef{ScalarShape}}:\n (sin(1.0 * x) - 1.0 / 1.0) - 0.0 ≤ 0\n (sin(2.0 * x) - 1.0 / 2.0) - 0.0 ≤ 0\n (sin(3.0 * x) - 1.0 / 3.0) - 0.0 ≤ 0\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLconstraints","page":"JuMP","title":"@NLconstraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLconstraints","category":"page"},{"location":"api/JuMP/#JuMP.@NLconstraints","page":"JuMP","title":"JuMP.@NLconstraints","text":"@NLconstraints(model, args...)\n\nAdds multiple nonlinear constraints to model at once, in the same fashion as the @NLconstraint macro.\n\nThe model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the constraints that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, t);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @NLconstraints(model, begin\n           t >= sqrt(x^2 + y^2)\n           [i = 1:2], z[i] <= log(a[i])\n       end)\n((t - sqrt(x ^ 2.0 + y ^ 2.0)) - 0.0 ≥ 0, NonlinearConstraintRef{ScalarShape}[(z[1] - log(4.0)) - 0.0 ≤ 0, (z[2] - log(5.0)) - 0.0 ≤ 0])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLexpression","page":"JuMP","title":"@NLexpression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLexpression","category":"page"},{"location":"api/JuMP/#JuMP.@NLexpression","page":"JuMP","title":"JuMP.@NLexpression","text":"@NLexpression(args...)\n\nEfficiently build a nonlinear expression which can then be inserted in other nonlinear constraints and the objective. See also [@expression]. For example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @NLexpression(model, my_expr, sin(x)^2 + cos(x^2))\nsubexpression[1]: sin(x) ^ 2.0 + cos(x ^ 2.0)\n\njulia> @NLconstraint(model, my_expr + y >= 5)\n(subexpression[1] + y) - 5.0 ≥ 0\n\njulia> @NLobjective(model, Min, my_expr)\n\nIndexing over sets and anonymous expressions are also supported:\n\njulia> @NLexpression(model, my_expr_1[i=1:3], sin(i * x))\n3-element Vector{NonlinearExpression}:\n subexpression[2]: sin(1.0 * x)\n subexpression[3]: sin(2.0 * x)\n subexpression[4]: sin(3.0 * x)\n\njulia> my_expr_2 = @NLexpression(model, log(1 + sum(exp(my_expr_1[i]) for i in 1:2)))\nsubexpression[5]: log(1.0 + (exp(subexpression[2]) + exp(subexpression[3])))\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLexpressions","page":"JuMP","title":"@NLexpressions","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLexpressions","category":"page"},{"location":"api/JuMP/#JuMP.@NLexpressions","page":"JuMP","title":"JuMP.@NLexpressions","text":"@NLexpressions(model, args...)\n\nAdds multiple nonlinear expressions to model at once, in the same fashion as the @NLexpression macro.\n\nThe model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the expressions that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @NLexpressions(model, begin\n           my_expr, sqrt(x^2 + y^2)\n           my_expr_1[i = 1:2], log(a[i]) - z[i]\n       end)\n(subexpression[1]: sqrt(x ^ 2.0 + y ^ 2.0), NonlinearExpression[subexpression[2]: log(4.0) - z[1], subexpression[3]: log(5.0) - z[2]])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLobjective","page":"JuMP","title":"@NLobjective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLobjective","category":"page"},{"location":"api/JuMP/#JuMP.@NLobjective","page":"JuMP","title":"JuMP.@NLobjective","text":"@NLobjective(model, sense, expression)\n\nAdd a nonlinear objective to model with optimization sense sense. sense must be Max or Min.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @NLobjective(model, Max, 2x + 1 + sin(x))\n\njulia> print(model)\nMax 2.0 * x + 1.0 + sin(x)\nSubject to\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLparameter","page":"JuMP","title":"@NLparameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLparameter","category":"page"},{"location":"api/JuMP/#JuMP.@NLparameter","page":"JuMP","title":"JuMP.@NLparameter","text":"@NLparameter(model, param == value)\n\nCreate and return a nonlinear parameter param attached to the model model with initial value set to value. Nonlinear parameters may be used only in nonlinear expressions.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, x == 10)\nx == 10.0\n\njulia> value(x)\n10.0\n\n@NLparameter(model, value = param_value)\n\nCreate and return an anonymous nonlinear parameter param attached to the model model with initial value set to param_value. Nonlinear parameters may be used only in nonlinear expressions.\n\nExample\n\njulia> model = Model();\n\njulia> x = @NLparameter(model, value = 10)\nparameter[1] == 10.0\n\njulia> value(x)\n10.0\n\n@NLparameter(model, param_collection[...] == value_expr)\n\nCreate and return a collection of nonlinear parameters param_collection attached to the model model with initial value set to value_expr (may depend on index sets). Uses the same syntax for specifying index sets as @variable.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, y[i = 1:3] == 2 * i)\n3-element Vector{NonlinearParameter}:\n parameter[1] == 2.0\n parameter[2] == 4.0\n parameter[3] == 6.0\n\njulia> value(y[2])\n4.0\n\n@NLparameter(model, [...] == value_expr)\n\nCreate and return an anonymous collection of nonlinear parameters attached to the model model with initial value set to value_expr (may depend on index sets). Uses the same syntax for specifying index sets as @variable.\n\nExample\n\njulia> model = Model();\n\njulia> y = @NLparameter(model, [i = 1:3] == 2 * i)\n3-element Vector{NonlinearParameter}:\n parameter[1] == 2.0\n parameter[2] == 4.0\n parameter[3] == 6.0\n\njulia> value(y[2])\n4.0\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLparameters","page":"JuMP","title":"@NLparameters","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLparameters","category":"page"},{"location":"api/JuMP/#JuMP.@NLparameters","page":"JuMP","title":"JuMP.@NLparameters","text":" @NLparameters(model, args...)\n\nCreate and return multiple nonlinear parameters attached to model model, in the same fashion as @NLparameter macro.\n\nThe model must be the first argument, and multiple parameters can be added on multiple lines wrapped in a begin ... end block. Distinct parameters need to be placed on separate lines as in the following example.\n\nThe macro returns a tuple containing the parameters that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameters(model, begin\n           x == 10\n           b == 156\n       end);\n\njulia> value(x)\n10.0\n\n\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@build_constraint","page":"JuMP","title":"@build_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@build_constraint","category":"page"},{"location":"api/JuMP/#JuMP.@build_constraint","page":"JuMP","title":"JuMP.@build_constraint","text":"@build_constraint(constraint_expr)\n\nConstructs a ScalarConstraint or VectorConstraint using the same machinery as @constraint but without adding the constraint to a model.\n\nConstraints using broadcast operators like x .<= 1 are also supported and will create arrays of ScalarConstraint or VectorConstraint.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @build_constraint(2x >= 1)\nScalarConstraint{AffExpr, MathOptInterface.GreaterThan{Float64}}(2 x, MathOptInterface.GreaterThan{Float64}(1.0))\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@constraint","page":"JuMP","title":"@constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@constraint","category":"page"},{"location":"api/JuMP/#JuMP.@constraint","page":"JuMP","title":"JuMP.@constraint","text":"@constraint(m::GenericModel, expr, kw_args...)\n\nAdd a constraint described by the expression expr.\n\n@constraint(m::GenericModel, ref[i=..., j=..., ...], expr, kw_args...)\n\nAdd a group of constraints described by the expression expr parametrized by i, j, ...\n\nThe expression expr can either be\n\nof the form func in set constraining the function func to belong to the set set which is either a MOI.AbstractSet or one of the JuMP shortcuts SecondOrderCone, RotatedSecondOrderCone and PSDCone, e.g. @constraint(model, [1, x-1, y-2] in SecondOrderCone()) constrains the norm of [x-1, y-2] be less than 1;\nof the form a sign b, where sign is one of ==, ≥, >=, ≤ and <= building the single constraint enforcing the comparison to hold for the expression a and b, e.g. @constraint(m, x^2 + y^2 == 1) constrains x and y to lie on the unit circle;\nof the form a ≤ b ≤ c or a ≥ b ≥ c (where ≤ and <= (resp. ≥ and >=) can be used interchangeably) constraining the paired the expression b to lie between a and c;\nof the forms @constraint(m, a .sign b) or @constraint(m, a .sign b .sign c) which broadcast the constraint creation to each element of the vectors.\n\nThe recognized keyword arguments in kw_args are the following:\n\nbase_name: Sets the name prefix used to generate constraint names. It corresponds to the constraint name for scalar constraints, otherwise, the constraint names are set to base_name[...] for each index ... of the axes axes.\ncontainer: Specify the container type.\nset_string_name::Bool = true: control whether to set the MOI.ConstraintName attribute. Passing set_string_name = false can improve performance.\n\nNote for extending the constraint macro\n\nEach constraint will be created using add_constraint(m, build_constraint(_error, func, set)) where\n\n_error is an error function showing the constraint call in addition to the error message given as argument,\nfunc is the expression that is constrained\nand set is the set in which it is constrained to belong.\n\nFor expr of the first type (i.e. @constraint(m, func in set)), func and set are passed unchanged to build_constraint but for the other types, they are determined from the expressions and signs. For instance, @constraint(m, x^2 + y^2 == 1) is transformed into add_constraint(m, build_constraint(_error, x^2 + y^2, MOI.EqualTo(1.0))).\n\nTo extend JuMP to accept new constraints of this form, it is necessary to add the corresponding methods to build_constraint. Note that this will likely mean that either func or set will be some custom type, rather than e.g. a Symbol, since we will likely want to dispatch on the type of the function or set appearing in the constraint.\n\nFor extensions that need to create constraints with more information than just func and set, an additional positional argument can be specified to @constraint that will then be passed on build_constraint. Hence, we can enable this syntax by defining extensions of build_constraint(_error, func, set, my_arg; kw_args...). This produces the user syntax: @constraint(model, ref[...], expr, my_arg, kw_args...).\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@constraints","page":"JuMP","title":"@constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@constraints","category":"page"},{"location":"api/JuMP/#JuMP.@constraints","page":"JuMP","title":"JuMP.@constraints","text":"@constraints(model, args...)\n\nAdds groups of constraints at once, in the same fashion as the @constraint macro.\n\nThe model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the constraints that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, w);\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:3]);\n\njulia> @constraints(model, begin\n           x >= 1\n           y - w <= 2\n           sum_to_one[i=1:3], z[i] + y == 1\n       end);\n\njulia> print(model)\nFeasibility\nSubject to\n sum_to_one[1] : y + z[1] = 1\n sum_to_one[2] : y + z[2] = 1\n sum_to_one[3] : y + z[3] = 1\n x ≥ 1\n -w + y ≤ 2\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@expression","page":"JuMP","title":"@expression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@expression","category":"page"},{"location":"api/JuMP/#JuMP.@expression","page":"JuMP","title":"JuMP.@expression","text":"@expression(args...)\n\nEfficiently builds a linear or quadratic expression but does not add to model immediately. Instead, returns the expression which can then be inserted in other constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:5]);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z);\n\njulia> @expression(model, shared, sum(i * x[i] for i in 1:5))\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5]\n\njulia> @constraint(model, shared + y >= 5)\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5] + y ≥ 5\n\njulia> @constraint(model, shared + z <= 10)\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5] + z ≤ 10\n\nThe ref accepts index sets in the same way as @variable, and those indices can be used in the construction of the expressions:\n\njulia> @expression(model, expr[i = 1:3], i * sum(x[j] for j in 1:3))\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[1] + 2 x[2] + 2 x[3]\n 3 x[1] + 3 x[2] + 3 x[3]\n\nAnonymous syntax is also supported:\n\njulia> expr = @expression(model, [i in 1:3], i * sum(x[j] for j in 1:3))\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[1] + 2 x[2] + 2 x[3]\n 3 x[1] + 3 x[2] + 3 x[3]\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@expressions","page":"JuMP","title":"@expressions","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@expressions","category":"page"},{"location":"api/JuMP/#JuMP.@expressions","page":"JuMP","title":"JuMP.@expressions","text":"@expressions(model, args...)\n\nAdds multiple expressions to model at once, in the same fashion as the @expression macro.\n\nThe model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the expressions that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @expressions(model, begin\n           my_expr, x^2 + y^2\n           my_expr_1[i = 1:2], a[i] - z[i]\n       end)\n(x² + y², AffExpr[-z[1] + 4, -z[2] + 5])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@objective","page":"JuMP","title":"@objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@objective","category":"page"},{"location":"api/JuMP/#JuMP.@objective","page":"JuMP","title":"JuMP.@objective","text":"@objective(model::GenericModel, sense, func)\n\nSet the objective sense to sense and objective function to func. The objective sense can be either Min, Max, MOI.MIN_SENSE, MOI.MAX_SENSE or MOI.FEASIBILITY_SENSE; see MOI.ObjectiveSense.\n\nIn order to set the sense programmatically, i.e., when sense is a Julia variable whose value is the sense, one of the three MOI.ObjectiveSense values should be used.\n\nExample\n\nTo minimize the value of the variable x, do as follows:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @objective(model, Min, x)\nx\n\nTo maximize the value of the affine expression 2x - 1, do as follows:\n\njulia> @objective(model, Max, 2x - 1)\n2 x - 1\n\nTo set a quadratic objective and set the objective sense programmatically, do as follows:\n\njulia> sense = MIN_SENSE\nMIN_SENSE::OptimizationSense = 0\n\njulia> @objective(model, sense, x^2 - 2x + 1)\nx² - 2 x + 1\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@operator","page":"JuMP","title":"@operator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@operator","category":"page"},{"location":"api/JuMP/#JuMP.@operator","page":"JuMP","title":"JuMP.@operator","text":"@operator(model, operator, dim, f[, ∇f[, ∇²f]])\n\nAdd the nonlinear operator operator in model with dim arguments, and create a new NonlinearOperator object called operator in the current scope.\n\nThe function f evaluates the operator and must return a scalar.\n\nThe optional function ∇f evaluates the first derivative, and the optional function ∇²f evaluates the second derivative.\n\n∇²f may be provided only if ∇f is also provided.\n\nUnivariate syntax\n\nIf dim == 1, then the method signatures of each function must be:\n\nf(::T)::T where {T<:Real}\n∇f(::T)::T where {T<:Real}\n∇²f(::T)::T where {T<:Real}\n\nMultivariate syntax\n\nIf dim > 1, then the method signatures of each function must be:\n\nf(x::T...)::T where {T<:Real}\n∇f(g::AbstractVector{T}, x::T...)::Nothing where {T<:Real}\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T<:Real}\n\nWhere the gradient vector g and Hessian matrix H are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :op_f)\n\njulia> @objective(model, Min, op_f(x))\nop_f(x)\n\njulia> op_f(2.0)\n4.0\n\njulia> model[:op_f]\nNonlinearOperator(f, :op_f)\n\njulia> model[:op_f](x)\nop_f(x)\n\nNon-macro version\n\nThis macro is provided as helpful syntax that matches the style of the rest of the JuMP macros. However, you may also add operators outside the macro using add_nonlinear_operator. For example:\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f)\nNonlinearOperator(f, :op_f)\n\nis equivalent to\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> op_f = model[:op_f] = add_nonlinear_operator(model, 1, f; name = :op_f)\nNonlinearOperator(f, :op_f)\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@variable","page":"JuMP","title":"@variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@variable","category":"page"},{"location":"api/JuMP/#JuMP.@variable","page":"JuMP","title":"JuMP.@variable","text":"@variable(model, expr, args..., kw_args...)\n\nAdd a variable to the model model described by the expression expr, the positional arguments args and the keyword arguments kw_args.\n\nAnonymous and named variables\n\nexpr must be one of the forms:\n\nOmitted, like @variable(model), which creates an anonymous variable\nA single symbol like @variable(model, x)\nA container expression like @variable(model, x[i=1:3])\nAn anoymous container expression like @variable(model, [i=1:3])\n\nBounds\n\nIn addition, the expression can have bounds, such as:\n\n@variable(model, x >= 0)\n@variable(model, x <= 0)\n@variable(model, x == 0)\n@variable(model, 0 <= x <= 1)\n\nand bounds can depend on the indices of the container expressions:\n\n@variable(model, -i <= x[i=1:3] <= i)\n\nSets\n\nYou can explicitly specify the set to which the variable belongs:\n\n@variable(model, x in MOI.Interval(0.0, 1.0))\n\nFor more information on this syntax, read Variables constrained on creation.\n\nPositional arguments\n\nThe recognized positional arguments in args are the following:\n\nBin: restricts the variable to the MOI.ZeroOne set, that is, {0, 1}. For example, @variable(model, x, Bin). Note: you cannot use @variable(model, Bin), use the binary keyword instead.\nInt: restricts the variable to the set of integers, that is, ..., -2, -1,  0, 1, 2, ... For example, @variable(model, x, Int). Note: you cannot use  @variable(model, Int), use the integer keyword instead.\nSymmetric: Only available when creating a square matrix of variables, i.e., when expr is of the form varname[1:n,1:n] or varname[i=1:n,j=1:n], it creates a symmetric matrix of variables.\nPSD: A restrictive extension to Symmetric which constraints a square matrix of variables to Symmetric and constrains to be positive semidefinite.\n\nKeyword arguments\n\nFour keyword arguments are useful in all cases:\n\nbase_name: Sets the name prefix used to generate variable names. It corresponds to the variable name for scalar variable, otherwise, the variable names are set to base_name[...] for each index ... of the axes axes.\nstart::Float64: specify the value passed to set_start_value for each variable\ncontainer: specify the container type. See Forcing the container type for more information.\nset_string_name::Bool = true: control whether to set the MOI.VariableName attribute. Passing set_string_name = false can improve performance.\n\nOther keyword arguments are needed to disambiguate sitations with anonymous variables:\n\nlower_bound::Float64: an alternative to x >= lb, sets the value of the variable lower bound.\nupper_bound::Float64: an alternative to x <= ub, sets the value of the variable upper bound.\nbinary::Bool: an alternative to passing Bin, sets whether the variable is binary or not.\ninteger::Bool: an alternative to passing Int, sets whether the variable is integer or not.\nset::MOI.AbstractSet: an alternative to using x in set\nvariable_type: used by JuMP extensions. See Extend @variable for more information.\n\nExample\n\nThe following are equivalent ways of creating a variable x of name x with lower bound 0:\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0)\nx\n\njulia> model = Model();\n\njulia> @variable(model, x, lower_bound = 0)\nx\n\njulia> model = Model();\n\njulia> x = @variable(model, base_name = \"x\", lower_bound = 0)\nx\n\nOther examples:\n\njulia> model = Model();\n\njulia> @variable(model, x[i=1:3] <= i, Int, start = sqrt(i), lower_bound = -i)\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @variable(model, y[i=1:3], container = DenseAxisArray, set = MOI.ZeroOne())\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, Base.OneTo(3)\nAnd data, a 3-element Vector{VariableRef}:\n y[1]\n y[2]\n y[3]\n\njulia> @variable(model, z[i=1:3], set_string_name = false)\n3-element Vector{VariableRef}:\n _[7]\n _[8]\n _[9]\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@variables","page":"JuMP","title":"@variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@variables","category":"page"},{"location":"api/JuMP/#JuMP.@variables","page":"JuMP","title":"JuMP.@variables","text":"@variables(model, args...)\n\nAdds multiple variables to model at once, in the same fashion as the @variable macro.\n\nThe model must be the first argument, and multiple variables can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the variables that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variables(model, begin\n           x\n           y[i = 1:2] >= 0, (start = i)\n           z, Bin, (start = 0, base_name = \"Z\")\n       end)\n(x, VariableRef[y[1], y[2]], Z)\n\nnote: Note\nKeyword arguments must be contained within parentheses (refer to the example above).\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#add_bridge","page":"JuMP","title":"add_bridge","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_bridge","category":"page"},{"location":"api/JuMP/#JuMP.add_bridge","page":"JuMP","title":"JuMP.add_bridge","text":"add_bridge(\n    model::GenericModel{T},\n    BT::Type{<:MOI.Bridges.AbstractBridge};\n    coefficient_type::Type{S} = T,\n) where {T,S}\n\nAdd BT{T} to the list of bridges that can be used to transform unsupported constraints into an equivalent formulation using only constraints supported by the optimizer.\n\nSee also: remove_bridge.\n\nExample\n\njulia> model = Model();\n\njulia> add_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> add_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64}\n       )\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_constraint","page":"JuMP","title":"add_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_constraint","category":"page"},{"location":"api/JuMP/#JuMP.add_constraint","page":"JuMP","title":"JuMP.add_constraint","text":"add_constraint(model::GenericModel, con::AbstractConstraint, name::String=\"\")\n\nAdd a constraint con to Model model and sets its name.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_constraint","page":"JuMP","title":"add_nonlinear_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_constraint","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_constraint","page":"JuMP","title":"JuMP.add_nonlinear_constraint","text":"add_nonlinear_constraint(model::Model, expr::Expr)\n\nAdd a nonlinear constraint described by the Julia expression ex to model.\n\nThis function is most useful if the expression ex is generated programmatically, and you cannot use @NLconstraint.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> add_nonlinear_constraint(model, :($(x) + $(x)^2 <= 1))\n(x + x ^ 2.0) - 1.0 ≤ 0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_expression","page":"JuMP","title":"add_nonlinear_expression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_expression","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_expression","page":"JuMP","title":"JuMP.add_nonlinear_expression","text":"add_nonlinear_expression(model::Model, expr::Expr)\n\nAdd a nonlinear expression expr to model.\n\nThis function is most useful if the expression expr is generated programmatically, and you cannot use @NLexpression.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> add_nonlinear_expression(model, :($(x) + $(x)^2))\nsubexpression[1]: x + x ^ 2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_operator","page":"JuMP","title":"add_nonlinear_operator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_operator","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_operator","page":"JuMP","title":"JuMP.add_nonlinear_operator","text":"add_nonlinear_operator(\n    model::Model,\n    dim::Int,\n    f::Function,\n    [∇f::Function,]\n    [∇²f::Function];\n    [name::Symbol = Symbol(f),]\n)\n\nAdd a new nonlinear operator with dim input arguments to model and associate it with the name name.\n\nThe function f evaluates the operator and must return a scalar.\n\nThe optional function ∇f evaluates the first derivative, and the optional function ∇²f evaluates the second derivative.\n\n∇²f may be provided only if ∇f is also provided.\n\nUnivariate syntax\n\nIf dim == 1, then the method signatures of each function must be:\n\nf(::T)::T where {T<:Real}\n∇f(::T)::T where {T<:Real}\n∇²f(::T)::T where {T<:Real}\n\nMultivariate syntax\n\nIf dim > 1, then the method signatures of each function must be:\n\nf(x::T...)::T where {T<:Real}\n∇f(g::AbstractVector{T}, x::T...)::Nothing where {T<:Real}\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T<:Real}\n\nWhere the gradient vector g and Hessian matrix H are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> op_f = add_nonlinear_operator(model, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :f)\n\njulia> @objective(model, Min, op_f(x))\nf(x)\n\njulia> op_f(2.0)\n4.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_parameter","page":"JuMP","title":"add_nonlinear_parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_parameter","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_parameter","page":"JuMP","title":"JuMP.add_nonlinear_parameter","text":"add_nonlinear_parameter(model::Model, value::Real)\n\nAdd an anonymous parameter to the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_to_expression!","page":"JuMP","title":"add_to_expression!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_to_expression!","category":"page"},{"location":"api/JuMP/#JuMP.add_to_expression!","page":"JuMP","title":"JuMP.add_to_expression!","text":"add_to_expression!(expression, terms...)\n\nUpdates expression in place to expression + (*)(terms...). This is typically much more efficient than expression += (*)(terms...). For example, add_to_expression!(expression, a, b) produces the same result as expression += a*b, and add_to_expression!(expression, a) produces the same result as expression += a.\n\nOnly a few methods are defined, mostly for internal use, and only for the cases when (1) they can be implemented efficiently and (2) expression is capable of storing the result. For example, add_to_expression!(::AffExpr, ::GenericVariableRef, ::GenericVariableRef) is not defined because a GenericAffExpr cannot store the product of two variables.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_to_function_constant","page":"JuMP","title":"add_to_function_constant","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_to_function_constant","category":"page"},{"location":"api/JuMP/#JuMP.add_to_function_constant","page":"JuMP","title":"JuMP.add_to_function_constant","text":"add_to_function_constant(constraint::ConstraintRef, value)\n\nAdd value to the function constant term.\n\nNote that for scalar constraints, JuMP will aggregate all constant terms onto the right-hand side of the constraint so instead of modifying the function, the set will be translated by -value. For example, given a constraint 2x <= 3, add_to_function_constant(c, 4) will modify it to 2x <= -1.\n\nExample\n\nFor scalar constraints, the set is translated by -value:\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 0 <= 2x - 1 <= 2)\ncon : 2 x ∈ [1, 3]\n\njulia> add_to_function_constant(con, 4)\n\njulia> con\ncon : 2 x ∈ [-3, -1]\n\nFor vector constraints, the constant is added to the function:\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, con, [x + y, x, y] in SecondOrderCone())\ncon : [x + y, x, y] ∈ MathOptInterface.SecondOrderCone(3)\n\njulia> add_to_function_constant(con, [1, 2, 2])\n\njulia> con\ncon : [x + y + 1, x + 2, y + 2] ∈ MathOptInterface.SecondOrderCone(3)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_variable","page":"JuMP","title":"add_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_variable","category":"page"},{"location":"api/JuMP/#JuMP.add_variable","page":"JuMP","title":"JuMP.add_variable","text":"add_variable(m::GenericModel, v::AbstractVariable, name::String=\"\")\n\nAdd a variable v to Model m and sets its name.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_constraints","page":"JuMP","title":"all_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_constraints","category":"page"},{"location":"api/JuMP/#JuMP.all_constraints","page":"JuMP","title":"JuMP.all_constraints","text":"all_constraints(model::GenericModel, function_type, set_type)::Vector{<:ConstraintRef}\n\nReturn a list of all constraints currently in the model where the function has type function_type and the set has type set_type. The constraints are ordered by creation time.\n\nSee also list_of_constraint_types and num_constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> all_constraints(model, VariableRef, MOI.GreaterThan{Float64})\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n x ≥ 0\n\njulia> all_constraints(model, VariableRef, MOI.ZeroOne)\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}, ScalarShape}}:\n x binary\n\njulia> all_constraints(model, AffExpr, MOI.LessThan{Float64})\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n 2 x ≤ 1\n\n\n\n\n\nall_constraints(\n    model::GenericModel;\n    include_variable_in_set_constraints::Bool,\n)::Vector{ConstraintRef}\n\nReturn a list of all constraints in model.\n\nIf include_variable_in_set_constraints == true, then VariableRef constraints such as VariableRef-in-Integer are included. To return only the structural constraints (e.g., the rows in the constraint matrix of a linear program), pass include_variable_in_set_constraints = false.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Int);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> @NLconstraint(model, x^2 <= 1);\n\njulia> all_constraints(model; include_variable_in_set_constraints = true)\n4-element Vector{ConstraintRef}:\n 2 x ≤ 1\n x ≥ 0\n x integer\n x ^ 2.0 - 1.0 ≤ 0\n\njulia> all_constraints(model; include_variable_in_set_constraints = false)\n2-element Vector{ConstraintRef}:\n 2 x ≤ 1\n x ^ 2.0 - 1.0 ≤ 0\n\nPerformance considerations\n\nNote that this function is type-unstable because it returns an abstractly typed vector. If performance is a problem, consider using list_of_constraint_types and a function barrier. See the Performance tips for extensions section of the documentation for more details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_nonlinear_constraints","page":"JuMP","title":"all_nonlinear_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_nonlinear_constraints","category":"page"},{"location":"api/JuMP/#JuMP.all_nonlinear_constraints","page":"JuMP","title":"JuMP.all_nonlinear_constraints","text":"all_nonlinear_constraints(model::GenericModel)\n\nReturn a vector of all nonlinear constraint references in the model in the order they were added to the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_variables","page":"JuMP","title":"all_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_variables","category":"page"},{"location":"api/JuMP/#JuMP.all_variables","page":"JuMP","title":"JuMP.all_variables","text":"all_variables(model::GenericModel{T})::Vector{GenericVariableRef{T}} where {T}\n\nReturns a list of all variables currently in the model. The variables are ordered by creation time.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> all_variables(model)\n2-element Vector{VariableRef}:\n x\n y\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#anonymous_name","page":"JuMP","title":"anonymous_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.anonymous_name","category":"page"},{"location":"api/JuMP/#JuMP.anonymous_name","page":"JuMP","title":"JuMP.anonymous_name","text":"anonymous_name(::MIME, x::AbstractVariableRef)\n\nThe name to use for an anonymous variable x when printing.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#backend","page":"JuMP","title":"backend","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.backend","category":"page"},{"location":"api/JuMP/#JuMP.backend","page":"JuMP","title":"JuMP.backend","text":"backend(model::GenericModel)\n\nReturn the lower-level MathOptInterface model that sits underneath JuMP. This model depends on which operating mode JuMP is in (see mode).\n\nIf JuMP is in DIRECT mode (i.e., the model was created using direct_model), the backend will be the optimizer passed to direct_model.\nIf JuMP is in MANUAL or AUTOMATIC mode, the backend is a MOI.Utilities.CachingOptimizer.\n\nThis function should only be used by advanced users looking to access low-level MathOptInterface or solver-specific functionality.\n\nNotes\n\nIf JuMP is not in DIRECT mode, the type returned by backend may change between any JuMP releases. Therefore, only use the public API exposed by MathOptInterface, and do not access internal fields. If you require access to the innermost optimizer, see unsafe_backend. Alternatively, use direct_model to create a JuMP model in DIRECT mode.\n\nSee also: unsafe_backend.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#barrier_iterations","page":"JuMP","title":"barrier_iterations","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.barrier_iterations","category":"page"},{"location":"api/JuMP/#JuMP.barrier_iterations","page":"JuMP","title":"JuMP.barrier_iterations","text":"barrier_iterations(model::GenericModel)\n\nGets the cumulative number of barrier iterations during the most recent optimization.\n\nSolvers must implement MOI.BarrierIterations() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#bridge_constraints","page":"JuMP","title":"bridge_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.bridge_constraints","category":"page"},{"location":"api/JuMP/#JuMP.bridge_constraints","page":"JuMP","title":"JuMP.bridge_constraints","text":"bridge_constraints(model::GenericModel)\n\nWhen in direct mode, return false.\n\nWhen in manual or automatic mode, return a Bool indicating whether the optimizer is set and unsupported constraints are automatically bridged to equivalent supported constraints when an appropriate transformation is available.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#build_constraint","page":"JuMP","title":"build_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.build_constraint","category":"page"},{"location":"api/JuMP/#JuMP.build_constraint","page":"JuMP","title":"JuMP.build_constraint","text":"build_constraint(\n    _error::Function,\n    Q::LinearAlgebra.Symmetric{V, M},\n    ::PSDCone,\n) where {V<:AbstractJuMPScalar,M<:AbstractMatrix{V}}\n\nReturn a VectorConstraint of shape SymmetricMatrixShape constraining the matrix Q to be positive semidefinite.\n\nThis function is used by the @constraint macros as follows:\n\njulia> import LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, LinearAlgebra.Symmetric(Q) in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ PSDCone()\n\nThe form above is usually used when the entries of Q are affine or quadratic expressions, but it can also be used when the entries are variables to get the reference of the semidefinite constraint, e.g.,\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\njulia> @constraint(model, Q in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ PSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    Q::AbstractMatrix{<:AbstractJuMPScalar},\n    ::PSDCone,\n)\n\nReturn a VectorConstraint of shape SquareMatrixShape constraining the matrix Q to be symmetric and positive semidefinite.\n\nThis function is used by the @constraint macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, Q in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[2,1]  Q[2,2]] ∈ PSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    Q::LinearAlgebra.Hermitian{V,M},\n    ::HermitianPSDCone,\n) where {V<:AbstractJuMPScalar,M<:AbstractMatrix{V}}\n\nReturn a VectorConstraint of shape HermitianMatrixShape constraining the matrix Q to be Hermitian positive semidefinite.\n\nThis function is used by the @constraint macros as follows:\n\njulia> import LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, LinearAlgebra.Hermitian(Q) in HermitianPSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ HermitianPSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    f::AbstractVector{<:AbstractJuMPScalar},\n    ::Nonnegatives,\n    extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},\n)\n\nA helper method that re-writes\n\n@constraint(model, X >= Y, extra)\n\ninto\n\n@constraint(model, X - Y in extra)\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    f::AbstractVector{<:AbstractJuMPScalar},\n    ::Nonpositives,\n    extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},\n)\n\nA helper method that re-writes\n\n@constraint(model, Y <= X, extra)\n\ninto\n\n@constraint(model, X - Y in extra)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#build_variable","page":"JuMP","title":"build_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.build_variable","category":"page"},{"location":"api/JuMP/#JuMP.build_variable","page":"JuMP","title":"JuMP.build_variable","text":"build_variable(\n    _error::Function,\n    info::VariableInfo,\n    args...;\n    kwargs...,\n)\n\nReturn a new AbstractVariable object.\n\nThis method should only be implemented by developers creating JuMP extensions. It should never be called by users of JuMP.\n\nArguments\n\n_error: a function to call instead of error. _error annotates the error message with additional information for the user.\ninfo: an instance of VariableInfo. This has a variety of fields relating to the variable such as info.lower_bound and info.binary.\nargs: optional additional positional arguments for extending the @variable macro.\nkwargs: optional keyword arguments for extending the @variable macro.\n\nSee also: @variable\n\nwarning: Warning\nExtensions should define a method with ONE positional argument to dispatch the call to a different method. Creating an extension that relies on multiple positional arguments leads to MethodErrors if the user passes the arguments in the wrong order.\n\nExample\n\n@variable(model, x, Foo)\n\nwill call\n\nbuild_variable(_error::Function, info::VariableInfo, ::Type{Foo})\n\nPassing special-case positional arguments such as Bin, Int, and PSD is okay, along with keyword arguments:\n\n@variable(model, x, Int, Foo(), mykwarg = true)\n# or\n@variable(model, x, Foo(), Int, mykwarg = true)\n\nwill call\n\nbuild_variable(_error::Function, info::VariableInfo, ::Foo; mykwarg)\n\nand info.integer will be true.\n\nNote that the order of the positional arguments does not matter.\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::SymmetricMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape SymmetricMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::SkewSymmetricMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape SkewSymmetricMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        Q[1,2]\n -Q[1,2]  0\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::HermitianMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape HermitianMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im\n real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::PSDCone)\n\nReturn a VariablesConstrainedOnCreation of shape SymmetricMatrixShape constraining the variables to be positive semidefinite.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], PSD)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#callback_node_status","page":"JuMP","title":"callback_node_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.callback_node_status","category":"page"},{"location":"api/JuMP/#JuMP.callback_node_status","page":"JuMP","title":"JuMP.callback_node_status","text":"callback_node_status(cb_data, model::GenericModel)\n\nReturn an MOI.CallbackNodeStatusCode enum, indicating if the current primal solution available from callback_value is integer feasible.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#callback_value","page":"JuMP","title":"callback_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.callback_value","category":"page"},{"location":"api/JuMP/#JuMP.callback_value","page":"JuMP","title":"JuMP.callback_value","text":"callback_value(cb_data, x::GenericVariableRef)\n\nReturn the primal solution of a variable inside a callback.\n\ncb_data is the argument to the callback function, and the type is dependent on the solver.\n\n\n\n\n\ncallback_value(cb_data, expr::Union{GenericAffExpr, GenericQuadExpr})\n\nReturn the primal solution of an affine or quadratic expression inside a callback by getting the value for each variable appearing in the expression.\n\ncb_data is the argument to the callback function, and the type is dependent on the solver.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#check_belongs_to_model","page":"JuMP","title":"check_belongs_to_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.check_belongs_to_model","category":"page"},{"location":"api/JuMP/#JuMP.check_belongs_to_model","page":"JuMP","title":"JuMP.check_belongs_to_model","text":"check_belongs_to_model(func::AbstractJuMPScalar, model::AbstractModel)\n\nThrow VariableNotOwned if the owner_model of one of the variables of the function func is not model.\n\ncheck_belongs_to_model(constraint::AbstractConstraint, model::AbstractModel)\n\nThrow VariableNotOwned if the owner_model of one of the variables of the constraint constraint is not model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#coefficient","page":"JuMP","title":"coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.coefficient","category":"page"},{"location":"api/JuMP/#JuMP.coefficient","page":"JuMP","title":"JuMP.coefficient","text":"coefficient(v1::GenericVariableRef{T}, v2::GenericVariableRef{T}) where {T}\n\nReturn one(T) if v1 == v2, and zero(T) otherwise.\n\nThis is a fallback for other coefficient methods to simplify code in which the expression may be a single variable.\n\n\n\n\n\ncoefficient(a::GenericAffExpr{C,V}, v::V) where {C,V}\n\nReturn the coefficient associated with variable v in the affine expression a.\n\n\n\n\n\ncoefficient(a::GenericAffExpr{C,V}, v1::V, v2::V) where {C,V}\n\nReturn the coefficient associated with the term v1 * v2 in the quadratic expression a.\n\nNote that coefficient(a, v1, v2) is the same as coefficient(a, v2, v1).\n\n\n\n\n\ncoefficient(a::GenericQuadExpr{C,V}, v::V) where {C,V}\n\nReturn the coefficient associated with variable v in the affine component of a.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#compute_conflict!","page":"JuMP","title":"compute_conflict!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.compute_conflict!","category":"page"},{"location":"api/JuMP/#JuMP.compute_conflict!","page":"JuMP","title":"JuMP.compute_conflict!","text":"compute_conflict!(model::GenericModel)\n\nCompute a conflict if the model is infeasible. If an optimizer has not been set yet (see set_optimizer), a NoOptimizer error is thrown.\n\nThe status of the conflict can be checked with the MOI.ConflictStatus model attribute. Then, the status for each constraint can be queried with the MOI.ConstraintConflictStatus attribute.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constant","page":"JuMP","title":"constant","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constant","category":"page"},{"location":"api/JuMP/#JuMP.constant","page":"JuMP","title":"JuMP.constant","text":"constant(aff::GenericAffExpr{C, V})::C\n\nReturn the constant of the affine expression.\n\n\n\n\n\nconstant(aff::GenericQuadExpr{C, V})::C\n\nReturn the constant of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_by_name","page":"JuMP","title":"constraint_by_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_by_name","category":"page"},{"location":"api/JuMP/#JuMP.constraint_by_name","page":"JuMP","title":"JuMP.constraint_by_name","text":"constraint_by_name(model::AbstractModel,\n                   name::String)::Union{ConstraintRef, Nothing}\n\nReturn the reference of the constraint with name attribute name or Nothing if no constraint has this name attribute. Throws an error if several constraints have name as their name attribute.\n\nconstraint_by_name(model::AbstractModel,\n                   name::String,\n                   F::Type{<:Union{AbstractJuMPScalar,\n                                   Vector{<:AbstractJuMPScalar},\n                                   MOI.AbstactFunction}},\n                   S::Type{<:MOI.AbstractSet})::Union{ConstraintRef, Nothing}\n\nSimilar to the method above, except that it throws an error if the constraint is not an F-in-S contraint where F is either the JuMP or MOI type of the function, and S is the MOI type of the set. This method is recommended if you know the type of the function and set since its returned type can be inferred while for the method above (i.e. without F and S), the exact return type of the constraint index cannot be inferred.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x^2 == 1)\ncon : x² = 1\n\njulia> constraint_by_name(model, \"kon\")\n\njulia> constraint_by_name(model, \"con\")\ncon : x² = 1\n\njulia> constraint_by_name(model, \"con\", AffExpr, MOI.EqualTo{Float64})\n\njulia> constraint_by_name(model, \"con\", QuadExpr, MOI.EqualTo{Float64})\ncon : x² = 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_object","page":"JuMP","title":"constraint_object","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_object","category":"page"},{"location":"api/JuMP/#JuMP.constraint_object","page":"JuMP","title":"JuMP.constraint_object","text":"constraint_object(con_ref::ConstraintRef)\n\nReturn the underlying constraint data for the constraint referenced by ref.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_ref_with_index","page":"JuMP","title":"constraint_ref_with_index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_ref_with_index","category":"page"},{"location":"api/JuMP/#JuMP.constraint_ref_with_index","page":"JuMP","title":"JuMP.constraint_ref_with_index","text":"constraint_ref_with_index(model::AbstractModel, index::MOI.ConstraintIndex)\n\nReturn a ConstraintRef of model corresponding to index.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_string","page":"JuMP","title":"constraint_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_string","category":"page"},{"location":"api/JuMP/#JuMP.constraint_string","page":"JuMP","title":"JuMP.constraint_string","text":"constraint_string(\n    mode::MIME,\n    ref::ConstraintRef;\n    in_math_mode::Bool = false)\n\nReturn a string representation of the constraint ref, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraints_string","page":"JuMP","title":"constraints_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraints_string","category":"page"},{"location":"api/JuMP/#JuMP.constraints_string","page":"JuMP","title":"JuMP.constraints_string","text":"constraints_string(mode, model::AbstractModel)::Vector{String}\n\nReturn a list of Strings describing each constraint of the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_conflict","page":"JuMP","title":"copy_conflict","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_conflict","category":"page"},{"location":"api/JuMP/#JuMP.copy_conflict","page":"JuMP","title":"JuMP.copy_conflict","text":"copy_conflict(model::GenericModel)\n\nReturn a copy of the current conflict for the model model and a GenericReferenceMap that can be used to obtain the variable and constraint reference of the new model corresponding to a given model's reference.\n\nThis is a convenience function that provides a filtering function for copy_model.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and two constraints c1 and c2. This model has no solution, as the two constraints are mutually exclusive. The solver is asked to compute a conflict with compute_conflict!. The parts of model participating in the conflict are then copied into a model iis_model.\n\njulia> using JuMP\n\njulia> import Gurobi\n\njulia> model = Model(Gurobi.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, c1, x >= 2)\nc1 : x ≥ 2\n\njulia> @constraint(model, c2, x <= 1)\nc2 : x ≤ 1\n\njulia> optimize!(model)\n\njulia> compute_conflict!(model)\n\njulia> if get_attribute(model, MOI.ConflictStatus()) == MOI.CONFLICT_FOUND\n           iis_model, reference_map = copy_conflict(model)\n           print(iis_model)\n       end\nFeasibility\nSubject to\n c1 : x ≥ 2\n c2 : x ≤ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_extension_data","page":"JuMP","title":"copy_extension_data","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_extension_data","category":"page"},{"location":"api/JuMP/#JuMP.copy_extension_data","page":"JuMP","title":"JuMP.copy_extension_data","text":"copy_extension_data(data, new_model::AbstractModel, model::AbstractModel)\n\nReturn a copy of the extension data data of the model model to the extension data of the new model new_model.\n\nA method should be added for any JuMP extension storing data in the ext field.\n\nwarning: Warning\nDo not engage in type piracy by implementing this method for types of data that you did not define! JuMP extensions should store types that they define in model.ext, rather than regular Julia types.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_model","page":"JuMP","title":"copy_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_model","category":"page"},{"location":"api/JuMP/#JuMP.copy_model","page":"JuMP","title":"JuMP.copy_model","text":"copy_model(model::GenericModel; filter_constraints::Union{Nothing, Function}=nothing)\n\nReturn a copy of the model model and a GenericReferenceMap that can be used to obtain the variable and constraint reference of the new model corresponding to a given model's reference. A Base.copy(::AbstractModel) method has also been implemented, it is similar to copy_model but does not return the reference map.\n\nIf the filter_constraints argument is given, only the constraints for which this function returns true will be copied. This function is given a constraint reference as argument.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and a constraint cref. It is then copied into a model new_model with the new references assigned to x_new and cref_new.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, cref, x == 2)\ncref : x = 2\n\njulia> new_model, reference_map = copy_model(model);\n\njulia> x_new = reference_map[x]\nx\n\njulia> cref_new = reference_map[cref]\ncref : x = 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete","page":"JuMP","title":"delete","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete","category":"page"},{"location":"api/JuMP/#JuMP.delete","page":"JuMP","title":"JuMP.delete","text":"delete(model::GenericModel, con_ref::ConstraintRef)\n\nDelete the constraint associated with constraint_ref from the model model.\n\nNote that delete does not unregister the name from the model, so adding a new constraint of the same name will throw an error. Use unregister to unregister the name after deletion.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, c, 2x <= 1)\nc : 2 x ≤ 1\n\njulia> delete(model, c)\n\njulia> unregister(model, :c)\n\njulia> print(model)\nFeasibility\nSubject to\n\njulia> model[:c]\nERROR: KeyError: key :c not found\nStacktrace:\n[...]\n\n\n\n\n\ndelete(model::GenericModel, con_refs::Vector{<:ConstraintRef})\n\nDelete the constraints associated with con_refs from the model model. Solvers may implement specialized methods for deleting multiple constraints of the same concrete type, i.e., when isconcretetype(eltype(con_refs)). These may be more efficient than repeatedly calling the single constraint delete method.\n\nSee also: unregister\n\n\n\n\n\ndelete(model::GenericModel, variable_ref::GenericVariableRef)\n\nDelete the variable associated with variable_ref from the model model.\n\nNote that delete does not unregister the name from the model, so adding a new variable of the same name will throw an error. Use unregister to unregister the name after deletion.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> delete(model, x)\n\njulia> unregister(model, :x)\n\njulia> print(model)\nFeasibility\nSubject to\n\njulia> model[:x]\nERROR: KeyError: key :x not found\nStacktrace:\n[...]\n\n\n\n\n\ndelete(model::GenericModel, variable_refs::Vector{<:GenericVariableRef})\n\nDelete the variables associated with variable_refs from the model model. Solvers may implement methods for deleting multiple variables that are more efficient than repeatedly calling the single variable delete method.\n\nSee also: unregister\n\n\n\n\n\ndelete(model::Model, c::NonlinearConstraintRef)\n\nDelete the nonlinear constraint c from model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete_lower_bound","page":"JuMP","title":"delete_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.delete_lower_bound","page":"JuMP","title":"JuMP.delete_lower_bound","text":"delete_lower_bound(v::GenericVariableRef)\n\nDelete the lower bound constraint of a variable.\n\nSee also LowerBoundRef, has_lower_bound, lower_bound, set_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> has_lower_bound(x)\ntrue\n\njulia> delete_lower_bound(x)\n\njulia> has_lower_bound(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete_upper_bound","page":"JuMP","title":"delete_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.delete_upper_bound","page":"JuMP","title":"JuMP.delete_upper_bound","text":"delete_upper_bound(v::GenericVariableRef)\n\nDelete the upper bound constraint of a variable.\n\nErrors if one does not exist.\n\nSee also UpperBoundRef, has_upper_bound, upper_bound, set_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> has_upper_bound(x)\ntrue\n\njulia> delete_upper_bound(x)\n\njulia> has_upper_bound(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#direct_generic_model","page":"JuMP","title":"direct_generic_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.direct_generic_model","category":"page"},{"location":"api/JuMP/#JuMP.direct_generic_model","page":"JuMP","title":"JuMP.direct_generic_model","text":"direct_generic_model(\n    value_type::Type{T},\n    backend::MOI.ModelLike;\n) where {T<:Real}\n\nReturn a new JuMP model using backend to store the model and solve it.\n\nAs opposed to the Model constructor, no cache of the model is stored outside of backend and no bridges are automatically applied to backend.\n\nNotes\n\nThe absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this direct mode:\n\nWhen backend does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the Model constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when optimize! is called.\nNo constraint bridging is supported by default.\nThe optimizer used cannot be changed the model is constructed.\nThe model created cannot be copied.\n\n\n\n\n\ndirect_generic_model(::Type{T}, factory::MOI.OptimizerWithAttributes)\n\nCreate a direct_generic_model using factory, a MOI.OptimizerWithAttributes object created by optimizer_with_attributes.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer,\n           \"presolve\" => \"off\",\n           MOI.Silent() => true,\n       );\n\njulia> model = direct_generic_model(Float64, optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = direct_generic_model(Float64, HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#direct_model","page":"JuMP","title":"direct_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.direct_model","category":"page"},{"location":"api/JuMP/#JuMP.direct_model","page":"JuMP","title":"JuMP.direct_model","text":"direct_model(backend::MOI.ModelLike)\n\nReturn a new JuMP model using backend to store the model and solve it.\n\nAs opposed to the Model constructor, no cache of the model is stored outside of backend and no bridges are automatically applied to backend.\n\nNotes\n\nThe absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this direct mode:\n\nWhen backend does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the Model constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when optimize! is called.\nNo constraint bridging is supported by default.\nThe optimizer used cannot be changed the model is constructed.\nThe model created cannot be copied.\n\n\n\n\n\ndirect_model(factory::MOI.OptimizerWithAttributes)\n\nCreate a direct_model using factory, a MOI.OptimizerWithAttributes object created by optimizer_with_attributes.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer,\n           \"presolve\" => \"off\",\n           MOI.Silent() => true,\n       );\n\njulia> model = direct_model(optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = direct_model(HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#drop_zeros!","page":"JuMP","title":"drop_zeros!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.drop_zeros!","category":"page"},{"location":"api/JuMP/#JuMP.drop_zeros!","page":"JuMP","title":"JuMP.drop_zeros!","text":"drop_zeros!(expr::GenericAffExpr)\n\nRemove terms in the affine expression with 0 coefficients.\n\n\n\n\n\ndrop_zeros!(expr::GenericQuadExpr)\n\nRemove terms in the quadratic expression with 0 coefficients.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual","page":"JuMP","title":"dual","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual","category":"page"},{"location":"api/JuMP/#JuMP.dual","page":"JuMP","title":"JuMP.dual","text":"dual(con_ref::ConstraintRef; result::Int = 1)\n\nReturn the dual value of constraint con_ref associated with result index result of the most-recent solution returned by the solver.\n\nUse has_dual to check if a result exists before asking for values.\n\nSee also: result_count, shadow_price.\n\n\n\n\n\ndual(c::NonlinearConstraintRef)\n\nReturn the dual of the nonlinear constraint c.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_objective_value","page":"JuMP","title":"dual_objective_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_objective_value","category":"page"},{"location":"api/JuMP/#JuMP.dual_objective_value","page":"JuMP","title":"JuMP.dual_objective_value","text":"dual_objective_value(model::GenericModel; result::Int = 1)\n\nReturn the value of the objective of the dual problem associated with result index result of the most-recent solution returned by the solver.\n\nThrows MOI.UnsupportedAttribute{MOI.DualObjectiveValue} if the solver does not support this attribute.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_shape","page":"JuMP","title":"dual_shape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_shape","category":"page"},{"location":"api/JuMP/#JuMP.dual_shape","page":"JuMP","title":"JuMP.dual_shape","text":"dual_shape(shape::AbstractShape)::AbstractShape\n\nReturns the shape of the dual space of the space of objects of shape shape. By default, the dual_shape of a shape is itself. See the examples section below for an example for which this is not the case.\n\nExample\n\nConsider polynomial constraints for which the dual is moment constraints and moment constraints for which the dual is polynomial constraints. Shapes for polynomials can be defined as follows:\n\nstruct Polynomial\n    coefficients::Vector{Float64}\n    monomials::Vector{Monomial}\nend\nstruct PolynomialShape <: AbstractShape\n    monomials::Vector{Monomial}\nend\nJuMP.reshape_vector(x::Vector, shape::PolynomialShape) = Polynomial(x, shape.monomials)\n\nand a shape for moments can be defined as follows:\n\nstruct Moments\n    coefficients::Vector{Float64}\n    monomials::Vector{Monomial}\nend\nstruct MomentsShape <: AbstractShape\n    monomials::Vector{Monomial}\nend\nJuMP.reshape_vector(x::Vector, shape::MomentsShape) = Moments(x, shape.monomials)\n\nThen dual_shape allows the definition of the shape of the dual of polynomial and moment constraints:\n\ndual_shape(shape::PolynomialShape) = MomentsShape(shape.monomials)\ndual_shape(shape::MomentsShape) = PolynomialShape(shape.monomials)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_start_value","page":"JuMP","title":"dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.dual_start_value","page":"JuMP","title":"JuMP.dual_start_value","text":"dual_start_value(con_ref::ConstraintRef)\n\nReturn the dual start value (MOI attribute ConstraintDualStart) of the constraint con_ref.\n\nNote: If no dual start value has been set, dual_start_value will return nothing.\n\nSee also set_dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_status","page":"JuMP","title":"dual_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_status","category":"page"},{"location":"api/JuMP/#JuMP.dual_status","page":"JuMP","title":"JuMP.dual_status","text":"dual_status(model::GenericModel; result::Int = 1)\n\nReturn a MOI.ResultStatusCode describing the status of the most recent dual solution of the solver (i.e., the MOI.DualStatus attribute) associated with the result index result.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#error_if_direct_mode","page":"JuMP","title":"error_if_direct_mode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.error_if_direct_mode","category":"page"},{"location":"api/JuMP/#JuMP.error_if_direct_mode","page":"JuMP","title":"JuMP.error_if_direct_mode","text":"error_if_direct_mode(model::GenericModel, func::Symbol)\n\nErrors if model is in direct mode during a call from the function named func.\n\nUsed internally within JuMP, or by JuMP extensions who do not want to support models in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix","page":"JuMP","title":"fix","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix","category":"page"},{"location":"api/JuMP/#JuMP.fix","page":"JuMP","title":"JuMP.fix","text":"fix(v::GenericVariableRef, value::Number; force::Bool = false)\n\nFix a variable to a value. Update the fixing constraint if one exists, otherwise create a new one.\n\nIf the variable already has variable bounds and force=false, calling fix will throw an error. If force=true, existing variable bounds will be deleted, and the fixing constraint will be added. Note a variable will have no bounds after a call to unfix.\n\nSee also FixRef, is_fixed, fix_value, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0)\n\njulia> is_fixed(x)\ntrue\n\njulia> model = Model();\n\njulia> @variable(model, 0 <= x <= 1);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0; force = true)\n\njulia> is_fixed(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix_discrete_variables","page":"JuMP","title":"fix_discrete_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix_discrete_variables","category":"page"},{"location":"api/JuMP/#JuMP.fix_discrete_variables","page":"JuMP","title":"JuMP.fix_discrete_variables","text":"fix_discrete_variables([var_value::Function = value,] model::GenericModel)\n\nModifies model to convert all binary and integer variables to continuous variables with fixed bounds of var_value(x).\n\nReturn\n\nReturns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.\n\nNotes\n\nAn error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).\nAll other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin, start = 1);\n\njulia> @variable(model, 1 <= y <= 10, Int, start = 2);\n\njulia> @objective(model, Min, x + y);\n\njulia> undo_relax = fix_discrete_variables(start_value, model);\n\njulia> print(model)\nMin x + y\nSubject to\n x = 1\n y = 2\n\njulia> undo_relax()\n\njulia> print(model)\nMin x + y\nSubject to\n y ≥ 1\n y ≤ 10\n y integer\n x binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix_value","page":"JuMP","title":"fix_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix_value","category":"page"},{"location":"api/JuMP/#JuMP.fix_value","page":"JuMP","title":"JuMP.fix_value","text":"fix_value(v::GenericVariableRef)\n\nReturn the value to which a variable is fixed.\n\nError if one does not exist.\n\nSee also FixRef, is_fixed, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> fix_value(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#flatten!","page":"JuMP","title":"flatten!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.flatten!","category":"page"},{"location":"api/JuMP/#JuMP.flatten!","page":"JuMP","title":"JuMP.flatten!","text":"flatten!(expr::GenericNonlinearExpr)\n\nFlatten a nonlinear expression in-place by lifting nested + and * nodes into a single n-ary operation.\n\nMotivation\n\nNonlinear expressions created using operator overloading can be deeply nested and unbalanced. For example, prod(x for i in 1:4) creates *(x, *(x, *(x, x))) instead of the more preferable *(x, x, x, x).\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> y = prod(x for i in 1:4)\n((x²) * x) * x\n\njulia> flatten!(y)\n(x²) * x * x\n\njulia> flatten!(sin(prod(x for i in 1:4)))\nsin((x²) * x * x)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#function_string","page":"JuMP","title":"function_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.function_string","category":"page"},{"location":"api/JuMP/#JuMP.function_string","page":"JuMP","title":"JuMP.function_string","text":"function_string(\n    mode::MIME,\n    func::Union{JuMP.AbstractJuMPScalar,Vector{<:JuMP.AbstractJuMPScalar}},\n)\n\nReturn a String representing the function func using print mode mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#get_attribute","page":"JuMP","title":"get_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.get_attribute","category":"page"},{"location":"api/JuMP/#JuMP.get_attribute","page":"JuMP","title":"JuMP.get_attribute","text":"get_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute)\nget_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute)\nget_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute)\n\nGet the value of a solver-specifc attribute attr.\n\nThis is equivalent to calling MOI.get with the associated MOI model and, for variables and constraints, with the associated MOI.VariableIndex or MOI.ConstraintIndex.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2 * x <= 1)\nc : 2 x ≤ 1\n\njulia> get_attribute(model, MOI.Name())\n\"\"\n\njulia> get_attribute(x, MOI.VariableName())\n\"x\"\n\njulia> get_attribute(c, MOI.ConstraintName())\n\"c\"\n\n\n\n\n\nget_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n)\n\nGet the value of a solver-specifc attribute attr.\n\nThis is equivalent to calling MOI.get with the associated MOI model.\n\nIf attr is an AbstractString, it is converted to MOI.RawOptimizerAttribute.\n\nExample\n\njulia> import HiGHS\n\njulia> opt = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => true);\n\njulia> model = Model(opt);\n\njulia> get_attribute(model, \"output_flag\")\ntrue\n\njulia> get_attribute(model, MOI.RawOptimizerAttribute(\"output_flag\"))\ntrue\n\njulia> get_attribute(opt, \"output_flag\")\ntrue\n\njulia> get_attribute(opt, MOI.RawOptimizerAttribute(\"output_flag\"))\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#get_optimizer_attribute","page":"JuMP","title":"get_optimizer_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.get_optimizer_attribute","category":"page"},{"location":"api/JuMP/#JuMP.get_optimizer_attribute","page":"JuMP","title":"JuMP.get_optimizer_attribute","text":"get_optimizer_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n)\n\nReturn the value associated with the solver-specific attribute attr.\n\nIf attr is an AbstractString, this is equivalent to get_optimizer_attribute(model, MOI.RawOptimizerAttribute(name)).\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using get_attribute instead.\n\nSee also: set_optimizer_attribute, set_optimizer_attributes.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> get_optimizer_attribute(model, MOI.Silent())\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_duals","page":"JuMP","title":"has_duals","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_duals","category":"page"},{"location":"api/JuMP/#JuMP.has_duals","page":"JuMP","title":"JuMP.has_duals","text":"has_duals(model::GenericModel; result::Int = 1)\n\nReturn true if the solver has a dual solution in result index result available to query, otherwise return false.\n\nSee also dual, shadow_price, and result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_lower_bound","page":"JuMP","title":"has_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.has_lower_bound","page":"JuMP","title":"JuMP.has_lower_bound","text":"has_lower_bound(v::GenericVariableRef)\n\nReturn true if v has a lower bound. If true, the lower bound can be queried with lower_bound.\n\nSee also LowerBoundRef, lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> has_lower_bound(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_start_value","page":"JuMP","title":"has_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_start_value","category":"page"},{"location":"api/JuMP/#JuMP.has_start_value","page":"JuMP","title":"JuMP.has_start_value","text":"has_start_value(variable::AbstractVariableRef)\n\nReturn true if the variable has a start value set otherwise return false.\n\nSee also set_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_upper_bound","page":"JuMP","title":"has_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.has_upper_bound","page":"JuMP","title":"JuMP.has_upper_bound","text":"has_upper_bound(v::GenericVariableRef)\n\nReturn true if v has a upper bound. If true, the upper bound can be queried with upper_bound.\n\nSee also UpperBoundRef, upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> has_upper_bound(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_values","page":"JuMP","title":"has_values","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_values","category":"page"},{"location":"api/JuMP/#JuMP.has_values","page":"JuMP","title":"JuMP.has_values","text":"has_values(model::GenericModel; result::Int = 1)\n\nReturn true if the solver has a primal solution in result index result available to query, otherwise return false.\n\nSee also value and result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#in_set_string","page":"JuMP","title":"in_set_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.in_set_string","category":"page"},{"location":"api/JuMP/#JuMP.in_set_string","page":"JuMP","title":"JuMP.in_set_string","text":"in_set_string(mode::MIME, set)\n\nReturn a String representing the membership to the set set using print mode mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#index","page":"JuMP","title":"index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.index","category":"page"},{"location":"api/JuMP/#JuMP.index","page":"JuMP","title":"JuMP.index","text":"index(cr::ConstraintRef)::MOI.ConstraintIndex\n\nReturn the index of the constraint that corresponds to cr in the MOI backend.\n\n\n\n\n\nindex(v::GenericVariableRef)::MOI.VariableIndex\n\nReturn the index of the variable that corresponds to v in the MOI backend.\n\n\n\n\n\nindex(p::NonlinearParameter)::MOI.Nonlinear.ParameterIndex\n\nReturn the index of the nonlinear parameter associated with p.\n\n\n\n\n\nindex(ex::NonlinearExpression)::MOI.Nonlinear.ExpressionIndex\n\nReturn the index of the nonlinear expression associated with ex.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_binary","page":"JuMP","title":"is_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_binary","category":"page"},{"location":"api/JuMP/#JuMP.is_binary","page":"JuMP","title":"JuMP.is_binary","text":"is_binary(v::GenericVariableRef)\n\nReturn true if v is constrained to be binary.\n\nSee also BinaryRef, set_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> is_binary(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_fixed","page":"JuMP","title":"is_fixed","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_fixed","category":"page"},{"location":"api/JuMP/#JuMP.is_fixed","page":"JuMP","title":"JuMP.is_fixed","text":"is_fixed(v::GenericVariableRef)\n\nReturn true if v is a fixed variable. If true, the fixed value can be queried with fix_value.\n\nSee also FixRef, fix_value, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0)\n\njulia> is_fixed(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_integer","page":"JuMP","title":"is_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_integer","category":"page"},{"location":"api/JuMP/#JuMP.is_integer","page":"JuMP","title":"JuMP.is_integer","text":"is_integer(v::GenericVariableRef)\n\nReturn true if v is constrained to be integer.\n\nSee also IntegerRef, set_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_integer(x)\nfalse\n\njulia> set_integer(x)\n\njulia> is_integer(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_parameter","page":"JuMP","title":"is_parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_parameter","category":"page"},{"location":"api/JuMP/#JuMP.is_parameter","page":"JuMP","title":"JuMP.is_parameter","text":"is_parameter(x::GenericVariableRef)::Bool\n\nReturn true if x is constrained to be a parameter.\n\nSee also ParameterRef, set_parameter_value, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> is_parameter(p)\ntrue\n\njulia> @variable(model, x)\nx\n\njulia> is_parameter(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_valid","page":"JuMP","title":"is_valid","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_valid","category":"page"},{"location":"api/JuMP/#JuMP.is_valid","page":"JuMP","title":"JuMP.is_valid","text":"is_valid(model::GenericModel, con_ref::ConstraintRef{<:AbstractModel})\n\nReturn true if constraint_ref refers to a valid constraint in model.\n\n\n\n\n\nis_valid(model::GenericModel, variable_ref::GenericVariableRef)\n\nReturn true if variable refers to a valid variable in model.\n\n\n\n\n\nis_valid(model::Model, c::NonlinearConstraintRef)\n\nReturn true if c refers to a valid nonlinear constraint in model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#isequal_canonical","page":"JuMP","title":"isequal_canonical","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.isequal_canonical","category":"page"},{"location":"api/JuMP/#JuMP.isequal_canonical","page":"JuMP","title":"JuMP.isequal_canonical","text":"isequal_canonical(\n    aff::GenericAffExpr{C,V},\n    other::GenericAffExpr{C,V}\n) where {C,V}\n\nReturn true if aff is equal to other after dropping zeros and disregarding the order. Mainly useful for testing.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#jump_function","page":"JuMP","title":"jump_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.jump_function","category":"page"},{"location":"api/JuMP/#JuMP.jump_function","page":"JuMP","title":"JuMP.jump_function","text":"jump_function(x)\n\nGiven an MathOptInterface object x, return the JuMP equivalent.\n\nSee also: moi_function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#jump_function_type","page":"JuMP","title":"jump_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.jump_function_type","category":"page"},{"location":"api/JuMP/#JuMP.jump_function_type","page":"JuMP","title":"JuMP.jump_function_type","text":"jump_function_type(::Type{T}) where {T}\n\nGiven an MathOptInterface object type T, return the JuMP equivalent.\n\nSee also: moi_function_type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#latex_formulation","page":"JuMP","title":"latex_formulation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.latex_formulation","category":"page"},{"location":"api/JuMP/#JuMP.latex_formulation","page":"JuMP","title":"JuMP.latex_formulation","text":"latex_formulation(model::AbstractModel)\n\nWrap model in a type so that it can be pretty-printed as text/latex in a notebook like IJulia, or in Documenter.\n\nTo render the model, end the cell with latex_formulation(model), or call display(latex_formulation(model)) in to force the display of the model from inside a function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#linear_terms","page":"JuMP","title":"linear_terms","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.linear_terms","category":"page"},{"location":"api/JuMP/#JuMP.linear_terms","page":"JuMP","title":"JuMP.linear_terms","text":"linear_terms(aff::GenericAffExpr{C, V})\n\nProvides an iterator over coefficient-variable tuples (a_i::C, x_i::V) in the linear part of the affine expression.\n\n\n\n\n\nlinear_terms(quad::GenericQuadExpr{C, V})\n\nProvides an iterator over tuples (coefficient::C, variable::V) in the linear part of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#list_of_constraint_types","page":"JuMP","title":"list_of_constraint_types","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.list_of_constraint_types","category":"page"},{"location":"api/JuMP/#JuMP.list_of_constraint_types","page":"JuMP","title":"JuMP.list_of_constraint_types","text":"list_of_constraint_types(model::GenericModel)::Vector{Tuple{Type,Type}}\n\nReturn a list of tuples of the form (F, S) where F is a JuMP function type and S is an MOI set type such that all_constraints(model, F, S) returns a nonempty list.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> list_of_constraint_types(model)\n3-element Vector{Tuple{Type, Type}}:\n (AffExpr, MathOptInterface.LessThan{Float64})\n (VariableRef, MathOptInterface.GreaterThan{Float64})\n (VariableRef, MathOptInterface.ZeroOne)\n\nPerformance considerations\n\nIterating over the list of function and set types is a type-unstable operation. Consider using a function barrier. See the Performance tips for extensions section of the documentation for more details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#lower_bound","page":"JuMP","title":"lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.lower_bound","page":"JuMP","title":"JuMP.lower_bound","text":"lower_bound(v::GenericVariableRef)\n\nReturn the lower bound of a variable. Error if one does not exist.\n\nSee also LowerBoundRef, has_lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> lower_bound(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#lp_sensitivity_report","page":"JuMP","title":"lp_sensitivity_report","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.lp_sensitivity_report","category":"page"},{"location":"api/JuMP/#JuMP.lp_sensitivity_report","page":"JuMP","title":"JuMP.lp_sensitivity_report","text":"lp_sensitivity_report(model::GenericModel{T}; atol::T = Base.rtoldefault(T))::SensitivityReport{T} where {T}\n\nGiven a linear program model with a current optimal basis, return a SensitivityReport object, which maps:\n\nEvery variable reference to a tuple (d_lo, d_hi)::Tuple{T,T}, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.\nEvery constraint reference to a tuple (d_lo, d_hi)::Tuple{T,T}, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.\n\nBoth tuples are relative, rather than absolute. So given a objective coefficient of 1.0 and a tuple (-0.5, 0.5), the objective coefficient can range between 1.0 - 0.5 an 1.0 + 0.5.\n\natol is the primal/dual optimality tolerance, and should match the tolerance of the solver used to compute the basis.\n\nNote: interval constraints are NOT supported.\n\nExample\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, -1 <= x <= 2)\nx\n\njulia> @objective(model, Min, x)\nx\n\njulia> optimize!(model)\n\njulia> report = lp_sensitivity_report(model; atol = 1e-7);\n\njulia> dx_lo, dx_hi = report[x]\n(-1.0, Inf)\n\njulia> println(\n           \"The objective coefficient of `x` can decrease by $dx_lo or \" *\n           \"increase by $dx_hi.\"\n       )\nThe objective coefficient of `x` can decrease by -1.0 or increase by Inf.\n\njulia> dRHS_lo, dRHS_hi = report[LowerBoundRef(x)]\n(-Inf, 3.0)\n\njulia> println(\n           \"The lower bound of `x` can decrease by $dRHS_lo or increase \" *\n           \"by $dRHS_hi.\"\n       )\nThe lower bound of `x` can decrease by -Inf or increase by 3.0.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#map_coefficients","page":"JuMP","title":"map_coefficients","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.map_coefficients","category":"page"},{"location":"api/JuMP/#JuMP.map_coefficients","page":"JuMP","title":"JuMP.map_coefficients","text":"map_coefficients(f::Function, a::GenericAffExpr)\n\nApply f to the coefficients and constant term of an GenericAffExpr a and return a new expression.\n\nSee also: map_coefficients_inplace!\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = GenericAffExpr(1.0, x => 1.0)\nx + 1\n\njulia> map_coefficients(c -> 2 * c, a)\n2 x + 2\n\njulia> a\nx + 1\n\n\n\n\n\nmap_coefficients(f::Function, a::GenericQuadExpr)\n\nApply f to the coefficients and constant term of an GenericQuadExpr a and return a new expression.\n\nSee also: map_coefficients_inplace!\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = @expression(model, x^2 + x + 1)\nx² + x + 1\n\njulia> map_coefficients(c -> 2 * c, a)\n2 x² + 2 x + 2\n\njulia> a\nx² + x + 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#map_coefficients_inplace!","page":"JuMP","title":"map_coefficients_inplace!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.map_coefficients_inplace!","category":"page"},{"location":"api/JuMP/#JuMP.map_coefficients_inplace!","page":"JuMP","title":"JuMP.map_coefficients_inplace!","text":"map_coefficients_inplace!(f::Function, a::GenericAffExpr)\n\nApply f to the coefficients and constant term of an GenericAffExpr a and update them in-place.\n\nSee also: map_coefficients\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = GenericAffExpr(1.0, x => 1.0)\nx + 1\n\njulia> map_coefficients_inplace!(c -> 2 * c, a)\n2 x + 2\n\njulia> a\n2 x + 2\n\n\n\n\n\nmap_coefficients_inplace!(f::Function, a::GenericQuadExpr)\n\nApply f to the coefficients and constant term of an GenericQuadExpr a and update them in-place.\n\nSee also: map_coefficients\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = @expression(model, x^2 + x + 1)\nx² + x + 1\n\njulia> map_coefficients_inplace!(c -> 2 * c, a)\n2 x² + 2 x + 2\n\njulia> a\n2 x² + 2 x + 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#mode","page":"JuMP","title":"mode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.mode","category":"page"},{"location":"api/JuMP/#JuMP.mode","page":"JuMP","title":"JuMP.mode","text":"mode(model::GenericModel)\n\nReturn the ModelMode (DIRECT, AUTOMATIC, or MANUAL) of model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#model_convert","page":"JuMP","title":"model_convert","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.model_convert","category":"page"},{"location":"api/JuMP/#JuMP.model_convert","page":"JuMP","title":"JuMP.model_convert","text":"model_convert(\n    model::AbstractModel,\n    rhs::Union{\n        AbstractConstraint,\n        Number,\n        AbstractJuMPScalar,\n        MOI.AbstractSet,\n    },\n)\n\nConvert the coefficients and constants of functions and sets in the rhs to the coefficient type value_type(typeof(model)).\n\nPurpose\n\nCreating and adding a constraint is a two-step process. The first step calls build_constraint, and the result of that is passed to add_constraint.\n\nHowever, because build_constraint does not take the model as an argument, the coefficients and constants of the function or set might be different than value_type(typeof(model)).\n\nTherefore, the result of build_constraint is converted in a call to model_convert before the result is passed to add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#model_string","page":"JuMP","title":"model_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.model_string","category":"page"},{"location":"api/JuMP/#JuMP.model_string","page":"JuMP","title":"JuMP.model_string","text":"model_string(mode::MIME, model::AbstractModel)\n\nReturn a String representation of model given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_function","page":"JuMP","title":"moi_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_function","category":"page"},{"location":"api/JuMP/#JuMP.moi_function","page":"JuMP","title":"JuMP.moi_function","text":"moi_function(x)\n\nGiven a JuMP object x, return the MathOptInterface equivalent.\n\nSee also: jump_function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_function_type","page":"JuMP","title":"moi_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_function_type","category":"page"},{"location":"api/JuMP/#JuMP.moi_function_type","page":"JuMP","title":"JuMP.moi_function_type","text":"moi_function_type(::Type{T}) where {T}\n\nGiven a JuMP object type T, return the MathOptInterface equivalent.\n\nSee also: jump_function_type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_set","page":"JuMP","title":"moi_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_set","category":"page"},{"location":"api/JuMP/#JuMP.moi_set","page":"JuMP","title":"JuMP.moi_set","text":"moi_set(constraint::AbstractConstraint)\n\nReturn the set of the constraint constraint in the function-in-set form as a MathOptInterface.AbstractSet.\n\nmoi_set(s::AbstractVectorSet, dim::Int)\n\nReturns the MOI set of dimension dim corresponding to the JuMP set s.\n\nmoi_set(s::AbstractScalarSet)\n\nReturns the MOI set corresponding to the JuMP set s.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#name","page":"JuMP","title":"name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.name","category":"page"},{"location":"api/JuMP/#JuMP.name","page":"JuMP","title":"JuMP.name","text":"name(con_ref::ConstraintRef)\n\nGet a constraint's name attribute.\n\n\n\n\n\nname(v::GenericVariableRef)::String\n\nGet a variable's name attribute.\n\n\n\n\n\nname(model::AbstractModel)\n\nReturn the MOI.Name attribute of model's backend, or a default if empty.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#node_count","page":"JuMP","title":"node_count","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.node_count","category":"page"},{"location":"api/JuMP/#JuMP.node_count","page":"JuMP","title":"JuMP.node_count","text":"node_count(model::GenericModel)\n\nGets the total number of branch-and-bound nodes explored during the most recent optimization in a Mixed Integer Program.\n\nSolvers must implement MOI.NodeCount() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_constraint_string","page":"JuMP","title":"nonlinear_constraint_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_constraint_string","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_constraint_string","page":"JuMP","title":"JuMP.nonlinear_constraint_string","text":"nonlinear_constraint_string(\n    model::GenericModel,\n    mode::MIME,\n    c::_NonlinearConstraint,\n)\n\nReturn a string representation of the nonlinear constraint c belonging to model, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_dual_start_value","page":"JuMP","title":"nonlinear_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_dual_start_value","page":"JuMP","title":"JuMP.nonlinear_dual_start_value","text":"nonlinear_dual_start_value(model::Model)\n\nReturn the current value of the MOI attribute MOI.NLPBlockDualStart.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_expr_string","page":"JuMP","title":"nonlinear_expr_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_expr_string","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_expr_string","page":"JuMP","title":"JuMP.nonlinear_expr_string","text":"nonlinear_expr_string(\n    model::GenericModel,\n    mode::MIME,\n    c::MOI.Nonlinear.Expression,\n)\n\nReturn a string representation of the nonlinear expression c belonging to model, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_model","page":"JuMP","title":"nonlinear_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_model","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_model","page":"JuMP","title":"JuMP.nonlinear_model","text":"nonlinear_model(\n    model::GenericModel;\n    force::Bool = false,\n)::Union{MOI.Nonlinear.Model,Nothing}\n\nIf model has nonlinear components, return a MOI.Nonlinear.Model, otherwise return nothing.\n\nIf force, always return a MOI.Nonlinear.Model, and if one does not exist for the model, create an empty one.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#normalized_coefficient","page":"JuMP","title":"normalized_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.normalized_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.normalized_coefficient","page":"JuMP","title":"JuMP.normalized_coefficient","text":"normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef)\n\nReturn the coefficient associated with variable in constraint after JuMP has normalized the constraint into its standard form. See also set_normalized_coefficient.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#normalized_rhs","page":"JuMP","title":"normalized_rhs","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.normalized_rhs","category":"page"},{"location":"api/JuMP/#JuMP.normalized_rhs","page":"JuMP","title":"JuMP.normalized_rhs","text":"normalized_rhs(con_ref::ConstraintRef)\n\nReturn the right-hand side term of con_ref after JuMP has converted the constraint into its normalized form. See also set_normalized_rhs.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_constraints","page":"JuMP","title":"num_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_constraints","category":"page"},{"location":"api/JuMP/#JuMP.num_constraints","page":"JuMP","title":"JuMP.num_constraints","text":"num_constraints(model::GenericModel, function_type, set_type)::Int64\n\nReturn the number of constraints currently in the model where the function has type function_type and the set has type set_type.\n\nSee also list_of_constraint_types and all_constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, y in MOI.GreaterThan(1.0));\n\njulia> @constraint(model, y <= 1.0);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> num_constraints(model, VariableRef, MOI.GreaterThan{Float64})\n2\n\njulia> num_constraints(model, VariableRef, MOI.ZeroOne)\n1\n\njulia> num_constraints(model, AffExpr, MOI.LessThan{Float64})\n2\n\n\n\n\n\nnum_constraints(model::GenericModel; count_variable_in_set_constraints::Bool)\n\nReturn the number of constraints in model.\n\nIf count_variable_in_set_constraints == true, then VariableRef constraints such as VariableRef-in-Integer are included. To count only the number of structural constraints (e.g., the rows in the constraint matrix of a linear program), pass count_variable_in_set_constraints = false.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Int);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> num_constraints(model; count_variable_in_set_constraints = true)\n3\n\njulia> num_constraints(model; count_variable_in_set_constraints = false)\n1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_nonlinear_constraints","page":"JuMP","title":"num_nonlinear_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_nonlinear_constraints","category":"page"},{"location":"api/JuMP/#JuMP.num_nonlinear_constraints","page":"JuMP","title":"JuMP.num_nonlinear_constraints","text":"num_nonlinear_constraints(model::GenericModel)\n\nReturns the number of nonlinear constraints associated with the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_variables","page":"JuMP","title":"num_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_variables","category":"page"},{"location":"api/JuMP/#JuMP.num_variables","page":"JuMP","title":"JuMP.num_variables","text":"num_variables(model::GenericModel)::Int64\n\nReturns number of variables in model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#object_dictionary","page":"JuMP","title":"object_dictionary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.object_dictionary","category":"page"},{"location":"api/JuMP/#JuMP.object_dictionary","page":"JuMP","title":"JuMP.object_dictionary","text":"object_dictionary(model::GenericModel)\n\nReturn the dictionary that maps the symbol name of a variable, constraint, or expression to the corresponding object.\n\nObjects are registered to a specific symbol in the macros. For example, @variable(model, x[1:2, 1:2]) registers the array of variables x to the symbol :x.\n\nThis method should be defined for any subtype of AbstractModel.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_bound","page":"JuMP","title":"objective_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_bound","category":"page"},{"location":"api/JuMP/#JuMP.objective_bound","page":"JuMP","title":"JuMP.objective_bound","text":"objective_bound(model::GenericModel)\n\nReturn the best known bound on the optimal objective value after a call to optimize!(model).\n\nFor scalar-valued objectives, this function returns a Float64. For vector-valued objectives, it returns a Vector{Float64}.\n\nIn the case of a vector-valued objective, this returns the ideal point, that is, the point obtained if each objective was optimized independently.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function","page":"JuMP","title":"objective_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function","category":"page"},{"location":"api/JuMP/#JuMP.objective_function","page":"JuMP","title":"JuMP.objective_function","text":"objective_function(\n    model::GenericModel,\n    T::Type = objective_function_type(model),\n)\n\nReturn an object of type T representing the objective function.\n\nError if the objective is not convertible to type T.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_function(model, AffExpr)\n2 x + 1\n\njulia> objective_function(model, QuadExpr)\n2 x + 1\n\njulia> typeof(objective_function(model, QuadExpr))\nQuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})\n\nWe see with the last two commands that even if the objective function is affine, as it is convertible to a quadratic function, it can be queried as a quadratic function and the result is quadratic.\n\nHowever, it is not convertible to a variable.\n\njulia> objective_function(model, VariableRef)\nERROR: InexactError: convert(MathOptInterface.VariableIndex, 1.0 + 2.0 MOI.VariableIndex(1))\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function_string","page":"JuMP","title":"objective_function_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function_string","category":"page"},{"location":"api/JuMP/#JuMP.objective_function_string","page":"JuMP","title":"JuMP.objective_function_string","text":"objective_function_string(mode, model::AbstractModel)::String\n\nReturn a String describing the objective function of the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function_type","page":"JuMP","title":"objective_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function_type","category":"page"},{"location":"api/JuMP/#JuMP.objective_function_type","page":"JuMP","title":"JuMP.objective_function_type","text":"objective_function_type(model::GenericModel)::AbstractJuMPScalar\n\nReturn the type of the objective function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_sense","page":"JuMP","title":"objective_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_sense","category":"page"},{"location":"api/JuMP/#JuMP.objective_sense","page":"JuMP","title":"JuMP.objective_sense","text":"objective_sense(model::GenericModel)::MOI.OptimizationSense\n\nReturn the objective sense.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_value","page":"JuMP","title":"objective_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_value","category":"page"},{"location":"api/JuMP/#JuMP.objective_value","page":"JuMP","title":"JuMP.objective_value","text":"objective_value(model::GenericModel; result::Int = 1)\n\nReturn the objective value associated with result index result of the most-recent solution returned by the solver.\n\nFor scalar-valued objectives, this function returns a Float64. For vector-valued objectives, it returns a Vector{Float64}.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#op_ifelse","page":"JuMP","title":"op_ifelse","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_ifelse","category":"page"},{"location":"api/JuMP/#JuMP.op_ifelse","page":"JuMP","title":"JuMP.op_ifelse","text":"op_ifelse(a, x, y)\n\nA function that falls back to ifelse(a, x, y), but when called with a JuMP variables or expression in the first argument, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_ifelse(true, 1.0, 2.0)\n1.0\n\njulia> op_ifelse(x, 1.0, 2.0)\nifelse(x, 1.0, 2.0)\n\njulia> op_ifelse(true, x, 2.0)\nx\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#operator_to_set","page":"JuMP","title":"operator_to_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.operator_to_set","category":"page"},{"location":"api/JuMP/#JuMP.operator_to_set","page":"JuMP","title":"JuMP.operator_to_set","text":"operator_to_set(_error::Function, ::Val{sense_symbol})\n\nConverts a sense symbol to a set set such that @constraint(model, func sense_symbol 0) is equivalent to @constraint(model, func in set) for any func::AbstractJuMPScalar.\n\nExample\n\nOnce a custom set is defined you can directly create a JuMP constraint with it:\n\njulia> struct CustomSet{T} <: MOI.AbstractScalarSet\n           value::T\n       end\n\njulia> Base.copy(x::CustomSet) = CustomSet(x.value)\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> cref = @constraint(model, x in CustomSet(1.0))\nx ∈ CustomSet{Float64}(1.0)\n\nHowever, there might be an appropriate sign that could be used in order to provide a more convenient syntax:\n\njulia> JuMP.operator_to_set(::Function, ::Val{:⊰}) = CustomSet(0.0)\n\njulia> MOIU.supports_shift_constant(::Type{<:CustomSet}) = true\n\njulia> MOIU.shift_constant(set::CustomSet, value) = CustomSet(set.value + value)\n\njulia> cref = @constraint(model, x ⊰ 1)\nx ∈ CustomSet{Float64}(1.0)\n\nNote that the whole function is first moved to the right-hand side, then the sign is transformed into a set with zero constant and finally the constant is moved to the set with MOIU.shift_constant.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#operator_warn","page":"JuMP","title":"operator_warn","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.operator_warn","category":"page"},{"location":"api/JuMP/#JuMP.operator_warn","page":"JuMP","title":"JuMP.operator_warn","text":"operator_warn(model::AbstractModel)\noperator_warn(model::GenericModel)\n\nThis function is called on the model whenever two affine expressions are added together without using destructive_add!, and at least one of the two expressions has more than 50 terms.\n\nFor the case of Model, if this function is called more than 20,000 times then a warning is generated once.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimize!","page":"JuMP","title":"optimize!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimize!","category":"page"},{"location":"api/JuMP/#JuMP.optimize!","page":"JuMP","title":"JuMP.optimize!","text":"optimize!(\n    model::GenericModel;\n    ignore_optimize_hook = (model.optimize_hook === nothing),\n    _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =\n        MOI.Nonlinear.SparseReverseMode(),\n    kwargs...,\n)\n\nOptimize the model.\n\nIf an optimizer has not been set yet (see set_optimizer), a NoOptimizer error is thrown.\n\nIf ignore_optimize_hook == true, the optimize hook is ignored and the model is solved as if the hook was not set. Keyword arguments kwargs are passed to the optimize_hook. An error is thrown if optimize_hook is nothing and keyword arguments are provided.\n\nExperimental features\n\nThese features may change or be removed in any future version of JuMP.\n\nPass _differentiation_backend to set the MOI.Nonlinear.AbstractAutomaticDifferentiation backend used to compute derivatives of nonlinear programs.\n\nIf you require only :ExprGraph, it is more efficient to pass _differentiation_backend = MOI.Nonlinear.ExprGraphOnly().\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimizer_index","page":"JuMP","title":"optimizer_index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimizer_index","category":"page"},{"location":"api/JuMP/#JuMP.optimizer_index","page":"JuMP","title":"JuMP.optimizer_index","text":"optimizer_index(x::GenericVariableRef)::MOI.VariableIndex\noptimizer_index(x::ConstraintRef{<:GenericModel})::MOI.ConstraintIndex\n\nReturn the index that corresponds to x in the optimizer model.\n\nThrows NoOptimizer if no optimizer is set, and throws an ErrorException if the optimizer is set but is not attached.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimizer_with_attributes","page":"JuMP","title":"optimizer_with_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimizer_with_attributes","category":"page"},{"location":"api/JuMP/#JuMP.optimizer_with_attributes","page":"JuMP","title":"JuMP.optimizer_with_attributes","text":"optimizer_with_attributes(optimizer_constructor, attrs::Pair...)\n\nGroups an optimizer constructor with the list of attributes attrs. Note that it is equivalent to MOI.OptimizerWithAttributes.\n\nWhen provided to the Model constructor or to set_optimizer, it creates an optimizer by calling optimizer_constructor(), and then sets the attributes using set_attribute.\n\nSee also: set_attribute, get_attribute.\n\nNote\n\nThe string names of the attributes are specific to each solver. One should consult the solver's documentation to find the attributes of interest.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer, \"presolve\" => \"off\", MOI.Silent() => true,\n       );\n\njulia> model = Model(optimizer);\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#owner_model","page":"JuMP","title":"owner_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.owner_model","category":"page"},{"location":"api/JuMP/#JuMP.owner_model","page":"JuMP","title":"JuMP.owner_model","text":"owner_model(s::AbstractJuMPScalar)\n\nReturn the model owning the scalar s.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parameter_value","page":"JuMP","title":"parameter_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parameter_value","category":"page"},{"location":"api/JuMP/#JuMP.parameter_value","page":"JuMP","title":"JuMP.parameter_value","text":"parameter_value(x::GenericVariableRef)\n\nReturn the value of the parameter x.\n\nErrors if x is not a parameter.\n\nSee also ParameterRef, is_parameter, set_parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> parameter_value(p)\n2.0\n\njulia> set_parameter_value(p, 2.5)\n\njulia> parameter_value(p)\n2.5\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint","page":"JuMP","title":"parse_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint","page":"JuMP","title":"JuMP.parse_constraint","text":"parse_constraint(_error::Function, expr::Expr)\n\nThe entry-point for all constraint-related parsing.\n\nArguments\n\nThe _error function is passed everywhere to provide better error messages\nexpr comes from the @constraint macro. There are two possibilities:\n@constraint(model, expr)\n@constraint(model, name[args], expr)\nIn both cases, expr is the main component of the constraint.\n\nSupported syntax\n\nJuMP currently supports the following expr objects:\n\nlhs <= rhs\nlhs == rhs\nlhs >= rhs\nl <= body <= u\nu >= body >= l\nlhs ⟂ rhs\nlhs in rhs\nlhs ∈ rhs\nz => {constraint}\n!z => {constraint}\n\nas well as all broadcasted variants.\n\nExtensions\n\nThe infrastructure behind parse_constraint is extendable. See parse_constraint_head and parse_constraint_call for details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint_call","page":"JuMP","title":"parse_constraint_call","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint_call","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint_call","page":"JuMP","title":"JuMP.parse_constraint_call","text":"parse_constraint_call(\n    _error::Function,\n    is_vectorized::Bool,\n    ::Val{op},\n    args...,\n)\n\nImplement this method to intercept the parsing of a :call expression with operator op.\n\nwarning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.\n\nArguments\n\n_error: a function that accepts a String and throws the string as an error, along with some descriptive information of the macro from which it was thrown.\nis_vectorized: a boolean to indicate if op should be broadcast or not\nop: the first element of the .args field of the Expr to intercept\nargs...: the .args field of the Expr.\n\nReturns\n\nThis function must return:\n\nparse_code::Expr: an expression containing any setup or rewriting code that needs to be called before build_constraint\nbuild_code::Expr: an expression that calls build_constraint( or build_constraint.( depending on is_vectorized.\n\nSee also: parse_constraint_head, build_constraint\n\n\n\n\n\nparse_constraint_call(\n    _error::Function,\n    vectorized::Bool,\n    ::Val{op},\n    lhs,\n    rhs,\n) where {op}\n\nFallback handler for binary operators. These might be infix operators like @constraint(model, lhs op rhs), or normal operators like @constraint(model, op(lhs, rhs)).\n\nIn both cases, we rewrite as lhs - rhs in operator_to_set(_error, op).\n\nSee operator_to_set for details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint_head","page":"JuMP","title":"parse_constraint_head","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint_head","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint_head","page":"JuMP","title":"JuMP.parse_constraint_head","text":"parse_constraint_head(_error::Function, ::Val{head}, args...)\n\nImplement this method to intercept the parsing of an expression with head head.\n\nwarning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.\n\nArguments\n\n_error: a function that accepts a String and throws the string as an error, along with some descriptive information of the macro from which it was thrown.\nhead: the .head field of the Expr to intercept\nargs...: the .args field of the Expr.\n\nReturns\n\nThis function must return:\n\nis_vectorized::Bool: whether the expression represents a broadcasted expression like x .<= 1\nparse_code::Expr: an expression containing any setup or rewriting code that needs to be called before build_constraint\nbuild_code::Expr: an expression that calls build_constraint( or build_constraint.( depending on is_vectorized.\n\nExisting implementations\n\nJuMP currently implements:\n\n::Val{:call}, which forwards calls to parse_constraint_call\n::Val{:comparison}, which handles the special case of l <= body <= u.\n\nSee also: parse_constraint_call, build_constraint\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_one_operator_variable","page":"JuMP","title":"parse_one_operator_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_one_operator_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_one_operator_variable","page":"JuMP","title":"JuMP.parse_one_operator_variable","text":"parse_one_operator_variable(_error::Function, infoexpr::_VariableInfoExpr, sense::Val{S}, value) where S\n\nUpdate infoexr for a variable expression in the @variable macro of the form variable name S value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_ternary_variable","page":"JuMP","title":"parse_ternary_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_ternary_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_ternary_variable","page":"JuMP","title":"JuMP.parse_ternary_variable","text":"parse_ternary_variable(_error, variable_info, lhs_sense, lhs, rhs_sense, rhs)\n\nA hook for JuMP extensiosn to intercept the parsing of a :comparison expression, which has the form lhs lhs_sense variable rhs_sense rhs.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_variable","page":"JuMP","title":"parse_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_variable","page":"JuMP","title":"JuMP.parse_variable","text":"parse_variable(_error::Function, ::_VariableInfoExpr, args...)\n\nA hook for extensions to intercept the parsing of inequality constraints in the @variable macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#primal_feasibility_report","page":"JuMP","title":"primal_feasibility_report","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.primal_feasibility_report","category":"page"},{"location":"api/JuMP/#JuMP.primal_feasibility_report","page":"JuMP","title":"JuMP.primal_feasibility_report","text":"primal_feasibility_report(\n    model::GenericModel{T},\n    point::AbstractDict{GenericVariableRef{T},T} = _last_primal_solution(model),\n    atol::T = zero(T),\n    skip_missing::Bool = false,\n)::Dict{Any,T}\n\nGiven a dictionary point, which maps variables to primal values, return a dictionary whose keys are the constraints with an infeasibility greater than the supplied tolerance atol. The value corresponding to each key is the respective infeasibility. Infeasibility is defined as the distance between the primal value of the constraint (see MOI.ConstraintPrimal) and the nearest point by Euclidean distance in the corresponding set.\n\nNotes\n\nIf skip_missing = true, constraints containing variables that are not in point will be ignored.\nIf skip_missing = false and a partial primal solution is provided, an error will be thrown.\nIf no point is provided, the primal solution from the last time the model was solved is used.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, 0.5 <= x <= 1);\n\njulia> primal_feasibility_report(model, Dict(x => 0.2))\nDict{Any, Float64} with 1 entry:\n  x ≥ 0.5 => 0.3\n\n\n\n\n\nprimal_feasibility_report(\n    point::Function,\n    model::GenericModel{T};\n    atol::T = zero(T),\n    skip_missing::Bool = false,\n) where {T}\n\nA form of primal_feasibility_report where a function is passed as the first argument instead of a dictionary as the second argument.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, 0.5 <= x <= 1, start = 1.3);\n\njulia> primal_feasibility_report(model) do v\n           return start_value(v)\n       end\nDict{Any, Float64} with 1 entry:\n  x ≤ 1 => 0.3\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#primal_status","page":"JuMP","title":"primal_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.primal_status","category":"page"},{"location":"api/JuMP/#JuMP.primal_status","page":"JuMP","title":"JuMP.primal_status","text":"primal_status(model::GenericModel; result::Int = 1)\n\nReturn a MOI.ResultStatusCode describing the status of the most recent primal solution of the solver (i.e., the MOI.PrimalStatus attribute) associated with the result index result.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#print_active_bridges","page":"JuMP","title":"print_active_bridges","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.print_active_bridges","category":"page"},{"location":"api/JuMP/#JuMP.print_active_bridges","page":"JuMP","title":"JuMP.print_active_bridges","text":"print_active_bridges([io::IO = stdout,] model::GenericModel)\n\nPrint a list of the variable, constraint, and objective bridges that are currently used in the model.\n\n\n\n\n\nprint_active_bridges([io::IO = stdout,] model::GenericModel, ::Type{F}) where {F}\n\nPrint a list of bridges required for an objective function of type F.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO = stdout,]\n    model::GenericModel,\n    F::Type,\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint a list of bridges required for a constraint of type F-in-S.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO = stdout,]\n    model::GenericModel,\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint a list of bridges required to add a variable constrained to the set S.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#print_bridge_graph","page":"JuMP","title":"print_bridge_graph","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.print_bridge_graph","category":"page"},{"location":"api/JuMP/#JuMP.print_bridge_graph","page":"JuMP","title":"JuMP.print_bridge_graph","text":" print_bridge_graph([io::IO,] model::GenericModel)\n\nPrint the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model.\n\nwarning: Warning\nThis function is intended for advanced users. If you want to see only the bridges that are currently used, use print_active_bridges instead.\n\nExplanation of output\n\nEach node in the hyper-graph corresponds to a variable, constraint, or objective type.\n\nVariable nodes are indicated by [ ]\nConstraint nodes are indicated by ( )\nObjective nodes are indicated by | |\n\nThe number inside each pair of brackets is an index of the node in the hyper-graph.\n\nNote that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.\n\nFor more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  \"MathOptInterface: a data structure for mathematical optimization problems.\" URL: https://arxiv.org/abs/2002.03447\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#quad_terms","page":"JuMP","title":"quad_terms","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.quad_terms","category":"page"},{"location":"api/JuMP/#JuMP.quad_terms","page":"JuMP","title":"JuMP.quad_terms","text":"quad_terms(quad::GenericQuadExpr{C, V})\n\nProvides an iterator over tuples (coefficient::C, var_1::V, var_2::V) in the quadratic part of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#raw_status","page":"JuMP","title":"raw_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.raw_status","category":"page"},{"location":"api/JuMP/#JuMP.raw_status","page":"JuMP","title":"JuMP.raw_status","text":"raw_status(model::GenericModel)\n\nReturn the reason why the solver stopped in its own words (i.e., the MathOptInterface model attribute RawStatusString).\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#read_from_file","page":"JuMP","title":"read_from_file","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.read_from_file","category":"page"},{"location":"api/JuMP/#JuMP.read_from_file","page":"JuMP","title":"JuMP.read_from_file","text":"read_from_file(\n    filename::String;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,\n    kwargs...,\n)\n\nReturn a JuMP model read from filename in the format format.\n\nIf the filename ends in .gz, it will be uncompressed using Gzip. If the filename ends in .bz2, it will be uncompressed using BZip2.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reduced_cost","page":"JuMP","title":"reduced_cost","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reduced_cost","category":"page"},{"location":"api/JuMP/#JuMP.reduced_cost","page":"JuMP","title":"JuMP.reduced_cost","text":"reduced_cost(x::GenericVariableRef{T})::T where {T}\n\nReturn the reduced cost associated with variable x.\n\nEquivalent to querying the shadow price of the active variable bound (if one exists and is active).\n\nSee also: shadow_price.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#register","page":"JuMP","title":"register","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.register","category":"page"},{"location":"api/JuMP/#JuMP.register","page":"JuMP","title":"JuMP.register","text":"register(\n    model::Model,\n    op::Symbol,\n    dimension::Integer,\n    f::Function;\n    autodiff:Bool = false,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol op.\n\nThe function f must support all subtypes of Real as arguments. Do not assume that the inputs are Float64.\n\nNotes\n\nFor this method, you must explicitly set autodiff = true, because no user-provided gradient function ∇f is given.\nSecond-derivative information is only computed if dimension == 1.\nop does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::T) where {T<:Real} = x^2\nf (generic function with 1 method)\n\njulia> register(model, :foo, 1, f; autodiff = true)\n\njulia> @NLobjective(model, Min, foo(x))\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> g(x::T, y::T) where {T<:Real} = x * y\ng (generic function with 1 method)\n\njulia> register(model, :g, 2, g; autodiff = true)\n\njulia> @NLobjective(model, Min, g(x[1], x[2]))\n\n\n\n\n\nregister(\n    model::Model,\n    s::Symbol,\n    dimension::Integer,\n    f::Function,\n    ∇f::Function;\n    autodiff:Bool = false,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol s. In addition, provide a gradient function ∇f.\n\nThe functions fand ∇f must support all subtypes of Real as arguments. Do not assume that the inputs are Float64.\n\nNotes\n\nIf the function f is univariate (i.e., dimension == 1), ∇f must return a number which represents the first-order derivative of the function f.\nIf the function f is multi-variate, ∇f must have a signature matching ∇f(g::AbstractVector{T}, args::T...) where {T<:Real}, where the first argument is a vector g that is modified in-place with the gradient.\nIf autodiff = true and dimension == 1, use automatic differentiation to compute the second-order derivative information. If autodiff = false, only first-order derivative information will be used.\ns does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::T) where {T<:Real} = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::T) where {T<:Real} = 2 * x\n∇f (generic function with 1 method)\n\njulia> register(model, :foo, 1, f, ∇f; autodiff = true)\n\njulia> @NLobjective(model, Min, foo(x))\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> g(x::T, y::T) where {T<:Real} = x * y\ng (generic function with 1 method)\n\njulia> function ∇g(g::AbstractVector{T}, x::T, y::T) where {T<:Real}\n           g[1] = y\n           g[2] = x\n           return\n       end\n∇g (generic function with 1 method)\n\njulia> register(model, :g, 2, g, ∇g)\n\njulia> @NLobjective(model, Min, g(x[1], x[2]))\n\n\n\n\n\nregister(\n    model::Model,\n    s::Symbol,\n    dimension::Integer,\n    f::Function,\n    ∇f::Function,\n    ∇²f::Function,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol s. In addition, provide a gradient function ∇f and a hessian function ∇²f.\n\n∇f and ∇²f must return numbers corresponding to the first- and second-order derivatives of the function f respectively.\n\nNotes\n\nBecause automatic differentiation is not used, you can assume the inputs are all Float64.\nThis method will throw an error if dimension > 1.\ns does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> register(model, :foo, 1, f, ∇f, ∇²f)\n\njulia> @NLobjective(model, Min, foo(x))\n\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relative_gap","page":"JuMP","title":"relative_gap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relative_gap","category":"page"},{"location":"api/JuMP/#JuMP.relative_gap","page":"JuMP","title":"JuMP.relative_gap","text":"relative_gap(model::GenericModel)\n\nReturn the final relative optimality gap after a call to optimize!(model). Exact value depends upon implementation of MathOptInterface.RelativeGap() by the particular solver used for optimization.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relax_integrality","page":"JuMP","title":"relax_integrality","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relax_integrality","category":"page"},{"location":"api/JuMP/#JuMP.relax_integrality","page":"JuMP","title":"JuMP.relax_integrality","text":"relax_integrality(model::GenericModel)\n\nModifies model to \"relax\" all binary and integrality constraints on variables. Specifically,\n\nBinary constraints are deleted, and variable bounds are tightened if necessary to ensure the variable is constrained to the interval 0 1.\nIntegrality constraints are deleted without modifying variable bounds.\nAn error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).\nAll other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.\n\nReturns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> @variable(model, 1 <= y <= 10, Int);\n\njulia> @objective(model, Min, x + y);\n\njulia> undo_relax = relax_integrality(model);\n\njulia> print(model)\nMin x + y\nSubject to\n x ≥ 0\n y ≥ 1\n x ≤ 1\n y ≤ 10\n\njulia> undo_relax()\n\njulia> print(model)\nMin x + y\nSubject to\n y ≥ 1\n y ≤ 10\n y integer\n x binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relax_with_penalty!","page":"JuMP","title":"relax_with_penalty!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relax_with_penalty!","category":"page"},{"location":"api/JuMP/#JuMP.relax_with_penalty!","page":"JuMP","title":"JuMP.relax_with_penalty!","text":"relax_with_penalty!(\n    model::GenericModel{T},\n    [penalties::Dict{ConstraintRef,T}];\n    [default::Union{Nothing,Real} = nothing,]\n) where {T}\n\nDestructively modify the model in-place to create a penalized relaxation of the constraints.\n\nwarning: Warning\nThis is a destructive routine that modifies the model in-place. If you don't want to modify the original model, use copy_model to create a copy before calling relax_with_penalty!.\n\nReformulation\n\nSee MOI.Utilities.ScalarPenaltyRelaxation for details of the reformulation.\n\nFor each constraint ci, the penalty passed to MOI.Utilities.ScalarPenaltyRelaxation is get(penalties, ci, default). If the value is nothing, because ci does not exist in penalties and default = nothing, then the constraint is skipped.\n\nReturn value\n\nThis function returns a Dict{ConstraintRef,AffExpr} that maps each constraint index to the corresponding y + z as an AffExpr. In an optimal solution, query the value of these functions to compute the violation of each constraint.\n\nRelax a subset of constraints\n\nTo relax a subset of constraints, pass a penalties dictionary and set default = nothing.\n\nExample\n\njulia> function new_model()\n           model = Model()\n           @variable(model, x)\n           @objective(model, Max, 2x + 1)\n           @constraint(model, c1, 2x - 1 <= -2)\n           @constraint(model, c2, 3x >= 0)\n           return model\n       end\nnew_model (generic function with 1 method)\n\njulia> model_1 = new_model();\n\njulia> penalty_map = relax_with_penalty!(model_1; default = 2.0);\n\njulia> penalty_map[model_1[:c1]]\n_[3]\n\njulia> penalty_map[model_1[:c2]]\n_[2]\n\njulia> print(model_1)\nMax 2 x - 2 _[2] - 2 _[3] + 1\nSubject to\n c2 : 3 x + _[2] ≥ 0\n c1 : 2 x - _[3] ≤ -1\n _[2] ≥ 0\n _[3] ≥ 0\n\njulia> model_2 = new_model();\n\njulia> relax_with_penalty!(model_2, Dict(model_2[:c2] => 3.0))\nDict{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.GreaterThan{Float64}}, ScalarShape}, AffExpr} with 1 entry:\n  c2 : 3 x + _[2] ≥ 0 => _[2]\n\njulia> print(model_2)\nMax 2 x - 3 _[2] + 1\nSubject to\n c2 : 3 x + _[2] ≥ 0\n c1 : 2 x ≤ -1\n _[2] ≥ 0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#remove_bridge","page":"JuMP","title":"remove_bridge","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.remove_bridge","category":"page"},{"location":"api/JuMP/#JuMP.remove_bridge","page":"JuMP","title":"JuMP.remove_bridge","text":"remove_bridge(\n    model::GenericModel{S},\n    BT::Type{<:MOI.Bridges.AbstractBridge};\n    coefficient_type::Type{T} = S,\n) where {S,T}\n\nRemove BT{T} from the list of bridges that can be used to transform unsupported constraints into an equivalent formulation using only constraints supported by the optimizer.\n\nSee also: add_bridge.\n\nExample\n\njulia> model = Model();\n\njulia> add_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> remove_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> add_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64},\n       )\n\njulia> remove_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64},\n       )\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reshape_set","page":"JuMP","title":"reshape_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reshape_set","category":"page"},{"location":"api/JuMP/#JuMP.reshape_set","page":"JuMP","title":"JuMP.reshape_set","text":"reshape_set(vectorized_set::MOI.AbstractSet, shape::AbstractShape)\n\nReturn a set in its original shape shape given its vectorized form vectorized_form.\n\nExample\n\nGiven a SymmetricMatrixShape of vectorized form [1, 2, 3] in MOI.PositiveSemidefinieConeTriangle(2), the following code returns the set of the original constraint Symmetric(Matrix[1 2; 2 3]) in PSDCone():\n\njulia> reshape_set(MOI.PositiveSemidefiniteConeTriangle(2), SymmetricMatrixShape(2))\nPSDCone()\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reshape_vector","page":"JuMP","title":"reshape_vector","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reshape_vector","category":"page"},{"location":"api/JuMP/#JuMP.reshape_vector","page":"JuMP","title":"JuMP.reshape_vector","text":"reshape_vector(vectorized_form::Vector, shape::AbstractShape)\n\nReturn an object in its original shape shape given its vectorized form vectorized_form.\n\nExample\n\nGiven a SymmetricMatrixShape of vectorized form [1, 2, 3], the following code returns the matrix Symmetric(Matrix[1 2; 2 3]):\n\njulia> reshape_vector([1, 2, 3], SymmetricMatrixShape(2))\n2×2 LinearAlgebra.Symmetric{Int64, Matrix{Int64}}:\n 1  2\n 2  3\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#result_count","page":"JuMP","title":"result_count","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.result_count","category":"page"},{"location":"api/JuMP/#JuMP.result_count","page":"JuMP","title":"JuMP.result_count","text":"result_count(model::GenericModel)\n\nReturn the number of results available to query after a call to optimize!.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reverse_sense","page":"JuMP","title":"reverse_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reverse_sense","category":"page"},{"location":"api/JuMP/#JuMP.reverse_sense","page":"JuMP","title":"JuMP.reverse_sense","text":"reverse_sense(::Val{T}) where {T}\n\nGiven an (in)equality symbol T, return a new Val object with the opposite (in)equality symbol.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_attribute","page":"JuMP","title":"set_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_attribute","category":"page"},{"location":"api/JuMP/#JuMP.set_attribute","page":"JuMP","title":"JuMP.set_attribute","text":"set_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute, value)\nset_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute, value)\nset_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute, value)\n\nSet the value of a solver-specifc attribute attr to value.\n\nThis is equivalent to calling MOI.set with the associated MOI model and, for variables and constraints, with the associated MOI.VariableIndex or MOI.ConstraintIndex.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2 * x <= 1)\nc : 2 x ≤ 1\n\njulia> set_attribute(model, MOI.Name(), \"model_new\")\n\njulia> set_attribute(x, MOI.VariableName(), \"x_new\")\n\njulia> set_attribute(c, MOI.ConstraintName(), \"c_new\")\n\n\n\n\n\nset_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n    value,\n)\n\nSet the value of a solver-specifc attribute attr to value.\n\nThis is equivalent to calling MOI.set with the associated MOI model.\n\nIf attr is an AbstractString, it is converted to MOI.RawOptimizerAttribute.\n\nExample\n\njulia> import HiGHS\n\njulia> opt = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false);\n\njulia> model = Model(opt);\n\njulia> set_attribute(model, \"output_flag\", false)\n\njulia> set_attribute(model, MOI.RawOptimizerAttribute(\"output_flag\"), true)\n\njulia> set_attribute(opt, \"output_flag\", true)\n\njulia> set_attribute(opt, MOI.RawOptimizerAttribute(\"output_flag\"), false)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_attributes","page":"JuMP","title":"set_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_attributes","category":"page"},{"location":"api/JuMP/#JuMP.set_attributes","page":"JuMP","title":"JuMP.set_attributes","text":"set_attributes(\n    destination::Union{\n        GenericModel,\n        MOI.OptimizerWithAttributes,\n        GenericVariableRef,\n        ConstraintRef,\n    },\n    pairs::Pair...,\n)\n\nGiven a list of attribute => value pairs, calls set_attribute(destination, attribute, value) for each pair.\n\nSee also: set_attribute, get_attribute.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_attributes(model, \"tol\" => 1e-4, \"max_iter\" => 100)\n\nis equivalent to:\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_attribute(model, \"tol\", 1e-4)\n\njulia> set_attribute(model, \"max_iter\", 100)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_binary","page":"JuMP","title":"set_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_binary","category":"page"},{"location":"api/JuMP/#JuMP.set_binary","page":"JuMP","title":"JuMP.set_binary","text":"set_binary(v::GenericVariableRef)\n\nAdd a constraint on the variable v that it must take values in the set 01.\n\nSee also BinaryRef, is_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_binary(x)\nfalse\n\njulia> set_binary(x)\n\njulia> is_binary(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_dual_start_value","page":"JuMP","title":"set_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_dual_start_value","page":"JuMP","title":"JuMP.set_dual_start_value","text":"set_dual_start_value(con_ref::ConstraintRef, value)\n\nSet the dual start value (MOI attribute ConstraintDualStart) of the constraint con_ref to value. To remove a dual start value set it to nothing.\n\nSee also dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_integer","page":"JuMP","title":"set_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_integer","category":"page"},{"location":"api/JuMP/#JuMP.set_integer","page":"JuMP","title":"JuMP.set_integer","text":"set_integer(variable_ref::GenericVariableRef)\n\nAdd an integrality constraint on the variable variable_ref.\n\nSee also IntegerRef, is_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_integer(x)\nfalse\n\njulia> set_integer(x)\n\njulia> is_integer(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_lower_bound","page":"JuMP","title":"set_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.set_lower_bound","page":"JuMP","title":"JuMP.set_lower_bound","text":"set_lower_bound(v::GenericVariableRef, lower::Number)\n\nSet the lower bound of a variable. If one does not exist, create a new lower bound constraint.\n\nSee also LowerBoundRef, has_lower_bound, lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> lower_bound(x)\n1.0\n\njulia> set_lower_bound(x, 2.0)\n\njulia> lower_bound(x)\n2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_name","page":"JuMP","title":"set_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_name","category":"page"},{"location":"api/JuMP/#JuMP.set_name","page":"JuMP","title":"JuMP.set_name","text":"set_name(con_ref::ConstraintRef, s::AbstractString)\n\nSet a constraint's name attribute.\n\n\n\n\n\nset_name(v::GenericVariableRef, s::AbstractString)\n\nSet a variable's name attribute.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_nonlinear_dual_start_value","page":"JuMP","title":"set_nonlinear_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_nonlinear_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_nonlinear_dual_start_value","page":"JuMP","title":"JuMP.set_nonlinear_dual_start_value","text":"set_nonlinear_dual_start_value(\n    model::Model,\n    start::Union{Nothing,Vector{Float64}},\n)\n\nSet the value of the MOI attribute MOI.NLPBlockDualStart.\n\nThe start vector corresponds to the Lagrangian duals of the nonlinear constraints, in the order given by all_nonlinear_constraints. That is, you must pass a single start vector corresponding to all of the nonlinear constraints in a single function call; you cannot set the dual start value of nonlinear constraints one-by-one. The example below demonstrates how to use all_nonlinear_constraints to create a mapping between the nonlinear constraint references and the start vector.\n\nPass nothing to unset a previous start.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> nl1 = @NLconstraint(model, x[1] <= sqrt(x[2]));\n\njulia> nl2 = @NLconstraint(model, x[1] >= exp(x[2]));\n\njulia> start = Dict(nl1 => -1.0, nl2 => 1.0);\n\njulia> start_vector = [start[con] for con in all_nonlinear_constraints(model)]\n2-element Vector{Float64}:\n -1.0\n  1.0\n\njulia> set_nonlinear_dual_start_value(model, start_vector)\n\njulia> nonlinear_dual_start_value(model)\n2-element Vector{Float64}:\n -1.0\n  1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_nonlinear_objective","page":"JuMP","title":"set_nonlinear_objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_nonlinear_objective","category":"page"},{"location":"api/JuMP/#JuMP.set_nonlinear_objective","page":"JuMP","title":"JuMP.set_nonlinear_objective","text":"set_nonlinear_objective(\n    model::Model,\n    sense::MOI.OptimizationSense,\n    expr::Expr,\n)\n\nSet the nonlinear objective of model to the expression expr, with the optimization sense sense.\n\nThis function is most useful if the expression expr is generated programmatically, and you cannot use @NLobjective.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\nYou must use MIN_SENSE or MAX_SENSE instead of Min and Max.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> set_nonlinear_objective(model, MIN_SENSE, :($(x) + $(x)^2))\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_coefficient","page":"JuMP","title":"set_normalized_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_coefficient","page":"JuMP","title":"JuMP.set_normalized_coefficient","text":"set_normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef, value)\n\nSet the coefficient of variable in the constraint constraint to value.\n\nNote that prior to this step, JuMP will aggregate multiple terms containing the same variable. For example, given a constraint 2x + 3x <= 2, set_normalized_coefficient(con, x, 4) will create the constraint 4x <= 2.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, 2x + 3x <= 2)\ncon : 5 x ≤ 2\n\njulia> set_normalized_coefficient(con, x, 4)\n\njulia> con\ncon : 4 x ≤ 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_coefficients","page":"JuMP","title":"set_normalized_coefficients","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_coefficients","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_coefficients","page":"JuMP","title":"JuMP.set_normalized_coefficients","text":"set_normalized_coefficients(\n    con_ref::ConstraintRef,\n    variable,\n    new_coefficients::Vector{Tuple{Int64,T}},\n)\n\nSet the coefficients of variable in the constraint con_ref to new_coefficients, where each element in new_coefficients is a tuple which maps the row to a new coefficient.\n\nNote that prior to this step, during constraint creation, JuMP will aggregate multiple terms containing the same variable.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, [2x + 3x, 4x] in MOI.Nonnegatives(2))\ncon : [5 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 2.0), (2, 5.0)])\n\njulia> con\ncon : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_rhs","page":"JuMP","title":"set_normalized_rhs","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_rhs","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_rhs","page":"JuMP","title":"JuMP.set_normalized_rhs","text":"set_normalized_rhs(con_ref::ConstraintRef, value)\n\nSet the right-hand side term of constraint to value.\n\nNote that prior to this step, JuMP will aggregate all constant terms onto the right-hand side of the constraint. For example, given a constraint 2x + 1 <= 2, set_normalized_rhs(con, 4) will create the constraint 2x <= 4, not 2x + 1 <= 4.\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x + 1 <= 2)\ncon : 2 x ≤ 1\n\njulia> set_normalized_rhs(con, 4)\n\njulia> con\ncon : 2 x ≤ 4\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective","page":"JuMP","title":"set_objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective","category":"page"},{"location":"api/JuMP/#JuMP.set_objective","page":"JuMP","title":"JuMP.set_objective","text":"set_objective(model::AbstractModel, sense::MOI.OptimizationSense, func)\n\nThe functional equivalent of the @objective macro.\n\nSets the objective sense and objective function simultaneously, and is equivalent to calling set_objective_sense and set_objective_function separately.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_objective(model, MIN_SENSE, x)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_coefficient","page":"JuMP","title":"set_objective_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_coefficient","page":"JuMP","title":"JuMP.set_objective_coefficient","text":"set_objective_coefficient(model::GenericModel, variable::GenericVariableRef, coefficient::Real)\n\nSet the linear objective coefficient associated with Variable to coefficient.\n\nNote: this function will throw an error if a nonlinear objective is set.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_function","page":"JuMP","title":"set_objective_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_function","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_function","page":"JuMP","title":"JuMP.set_objective_function","text":"set_objective_function(model::GenericModel, func::MOI.AbstractFunction)\nset_objective_function(model::GenericModel, func::AbstractJuMPScalar)\nset_objective_function(model::GenericModel, func::Real)\nset_objective_function(model::GenericModel, func::Vector{<:AbstractJuMPScalar})\n\nSets the objective function of the model to the given function. See set_objective_sense to set the objective sense. These are low-level functions; the recommended way to set the objective is with the @objective macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_sense","page":"JuMP","title":"set_objective_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_sense","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_sense","page":"JuMP","title":"JuMP.set_objective_sense","text":"set_objective_sense(model::GenericModel, sense::MOI.OptimizationSense)\n\nSets the objective sense of the model to the given sense. See set_objective_function to set the objective function. These are low-level functions; the recommended way to set the objective is with the @objective macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimize_hook","page":"JuMP","title":"set_optimize_hook","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimize_hook","category":"page"},{"location":"api/JuMP/#JuMP.set_optimize_hook","page":"JuMP","title":"JuMP.set_optimize_hook","text":"set_optimize_hook(model::GenericModel, f::Union{Function,Nothing})\n\nSet the function f as the optimize hook for model.\n\nf should have a signature f(model::GenericModel; kwargs...), where the kwargs are those passed to optimize!.\n\nNotes\n\nThe optimize hook should generally modify the model, or some external state in some way, and then call optimize!(model; ignore_optimize_hook = true) to optimize the problem, bypassing the hook.\nUse set_optimize_hook(model, nothing) to unset an optimize hook.\n\nExample\n\njulia> model = Model();\n\njulia> function my_hook(model::Model; kwargs...)\n           println(kwargs)\n           println(\"Calling with `ignore_optimize_hook = true`\")\n           optimize!(model; ignore_optimize_hook = true)\n           return\n       end\nmy_hook (generic function with 1 method)\n\njulia> set_optimize_hook(model, my_hook)\nmy_hook (generic function with 1 method)\n\njulia> optimize!(model; test_arg = true)\nBase.Pairs{Symbol, Bool, Tuple{Symbol}, NamedTuple{(:test_arg,), Tuple{Bool}}}(:test_arg => 1)\nCalling with `ignore_optimize_hook = true`\nERROR: NoOptimizer()\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer","page":"JuMP","title":"set_optimizer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer","page":"JuMP","title":"JuMP.set_optimizer","text":"set_optimizer(\n    model::GenericModel,\n    optimizer_factory;\n    add_bridges::Bool = true,\n)\n\nCreates an empty MathOptInterface.AbstractOptimizer instance by calling optimizer_factory() and sets it as the optimizer of model. Specifically, optimizer_factory must be callable with zero arguments and return an empty MathOptInterface.AbstractOptimizer.\n\nIf add_bridges is true, constraints and objectives that are not supported by the optimizer are automatically bridged to equivalent supported formulation. Passing add_bridges = false can improve performance if the solver natively supports all of the elements in model.\n\nSee set_attribute for setting solver-specific parameters of the optimizer.\n\nExample\n\njulia> import HiGHS\n\njulia> model = Model();\n\njulia> set_optimizer(model, () -> HiGHS.Optimizer())\n\njulia> set_optimizer(model, HiGHS.Optimizer; add_bridges = false)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer_attribute","page":"JuMP","title":"set_optimizer_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer_attribute","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer_attribute","page":"JuMP","title":"JuMP.set_optimizer_attribute","text":"set_optimizer_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n    value,\n)\n\nSet the solver-specific attribute attr in model to value.\n\nIf attr is an AbstractString, this is equivalent to set_optimizer_attribute(model, MOI.RawOptimizerAttribute(name), value).\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using set_attribute instead.\n\nSee also: set_optimizer_attributes, get_optimizer_attribute.\n\nExample\n\njulia> model = Model();\n\njulia> set_optimizer_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer_attributes","page":"JuMP","title":"set_optimizer_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer_attributes","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer_attributes","page":"JuMP","title":"JuMP.set_optimizer_attributes","text":"set_optimizer_attributes(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    pairs::Pair...,\n)\n\nGiven a list of attribute => value pairs, calls set_optimizer_attribute(model, attribute, value) for each pair.\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using set_attributes instead.\n\nSee also: set_optimizer_attribute, get_optimizer_attribute.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_optimizer_attributes(model, \"tol\" => 1e-4, \"max_iter\" => 100)\n\nis equivalent to:\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_optimizer_attribute(model, \"tol\", 1e-4)\n\njulia> set_optimizer_attribute(model, \"max_iter\", 100)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_parameter_value","page":"JuMP","title":"set_parameter_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_parameter_value","category":"page"},{"location":"api/JuMP/#JuMP.set_parameter_value","page":"JuMP","title":"JuMP.set_parameter_value","text":"set_parameter_value(x::GenericVariableRef, value)\n\nUpdate the parameter constraint on the variable x to value.\n\nErrors if x is not a parameter.\n\nSee also ParameterRef, is_parameter, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> parameter_value(p)\n2.0\n\njulia> set_parameter_value(p, 2.5)\n\njulia> parameter_value(p)\n2.5\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_silent","page":"JuMP","title":"set_silent","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_silent","category":"page"},{"location":"api/JuMP/#JuMP.set_silent","page":"JuMP","title":"JuMP.set_silent","text":"set_silent(model::GenericModel)\n\nTakes precedence over any other attribute controlling verbosity and requires the solver to produce no output.\n\nSee also: unset_silent.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_start_value","page":"JuMP","title":"set_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_start_value","page":"JuMP","title":"JuMP.set_start_value","text":"set_start_value(con_ref::ConstraintRef, value)\n\nSet the primal start value (MOI.ConstraintPrimalStart) of the constraint con_ref to value. To remove a primal start value set it to nothing.\n\nSee also start_value.\n\n\n\n\n\nset_start_value(variable::GenericVariableRef, value::Union{Real,Nothing})\n\nSet the start value (MOI attribute VariablePrimalStart) of the variable to value.\n\nPass nothing to unset the start value.\n\nNote: VariablePrimalStarts are sometimes called \"MIP-starts\" or \"warmstarts\".\n\nSee also start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_start_values","page":"JuMP","title":"set_start_values","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_start_values","category":"page"},{"location":"api/JuMP/#JuMP.set_start_values","page":"JuMP","title":"JuMP.set_start_values","text":"set_start_values(\n    model::GenericModel;\n    variable_primal_start::Union{Nothing,Function} = value,\n    constraint_primal_start::Union{Nothing,Function} = value,\n    constraint_dual_start::Union{Nothing,Function} = dual,\n    nonlinear_dual_start::Union{Nothing,Function} = nonlinear_dual_start_value,\n)\n\nSet the primal and dual starting values in model using the functions provided.\n\nIf any keyword argument is nothing, the corresponding start value is skipped.\n\nIf the optimizer does not support setting the starting value, the value will be skipped.\n\nvariable_primal_start\n\nThis function controls the primal starting solution for the variables. It is equivalent to calling set_start_value for each variable, or setting the MOI.VariablePrimalStart attribute.\n\nIf it is a function, it must have the form variable_primal_start(x::VariableRef) that maps each variable x to the starting primal value.\n\nThe default is value.\n\nconstraint_primal_start\n\nThis function controls the primal starting solution for the constraints. It is equivalent to calling set_start_value for each constraint, or setting the MOI.ConstraintPrimalStart attribute.\n\nIf it is a function, it must have the form constraint_primal_start(ci::ConstraintRef) that maps each constraint ci to the starting primal value.\n\nThe default is value.\n\nconstraint_dual_start\n\nThis function controls the dual starting solution for the constraints. It is equivalent to calling set_dual_start_value for each constraint, or setting the MOI.ConstraintDualStart attribute.\n\nIf it is a function, it must have the form constraint_dual_start(ci::ConstraintRef) that maps each constraint ci to the starting dual value.\n\nThe default is dual.\n\nnonlinear_dual_start\n\nThis function controls the dual starting solution for the nonlinear constraints It is equivalent to calling set_nonlinear_dual_start_value.\n\nIf it is a function, it must have the form nonlinear_dual_start(model::GenericModel) that returns a vector corresponding to the dual start of the constraints.\n\nThe default is nonlinear_dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_string_names_on_creation","page":"JuMP","title":"set_string_names_on_creation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_string_names_on_creation","category":"page"},{"location":"api/JuMP/#JuMP.set_string_names_on_creation","page":"JuMP","title":"JuMP.set_string_names_on_creation","text":"set_string_names_on_creation(model::GenericModel, value::Bool)\n\nSet the default argument of the set_string_name keyword in the @variable and @constraint macros to value. This is used to determine whether to assign String names to all variables and constraints in model.\n\nBy default, value is true. However, for larger models calling set_string_names_on_creation(model, false) can improve performance at the cost of reducing the readability of printing and solver log messages.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_time_limit_sec","page":"JuMP","title":"set_time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.set_time_limit_sec","page":"JuMP","title":"JuMP.set_time_limit_sec","text":"set_time_limit_sec(model::GenericModel, limit::Float64)\n\nSet the time limit (in seconds) of the solver.\n\nCan be unset using unset_time_limit_sec or with limit set to nothing.\n\nSee also: unset_time_limit_sec, time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_upper_bound","page":"JuMP","title":"set_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.set_upper_bound","page":"JuMP","title":"JuMP.set_upper_bound","text":"set_upper_bound(v::GenericVariableRef, upper::Number)\n\nSet the upper bound of a variable. If one does not exist, create an upper bound constraint.\n\nSee also UpperBoundRef, has_upper_bound, upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> upper_bound(x)\n1.0\n\njulia> set_upper_bound(x, 2.0)\n\njulia> upper_bound(x)\n2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_value","page":"JuMP","title":"set_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_value","category":"page"},{"location":"api/JuMP/#JuMP.set_value","page":"JuMP","title":"JuMP.set_value","text":"set_value(p::NonlinearParameter, v::Number)\n\nStore the value v in the nonlinear parameter p.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, p == 0)\np == 0.0\n\njulia> set_value(p, 5)\n5\n\njulia> value(p)\n5.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#shadow_price","page":"JuMP","title":"shadow_price","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.shadow_price","category":"page"},{"location":"api/JuMP/#JuMP.shadow_price","page":"JuMP","title":"JuMP.shadow_price","text":"shadow_price(con_ref::ConstraintRef)\n\nReturn the change in the objective from an infinitesimal relaxation of the constraint.\n\nThis value is computed from dual and can be queried only when has_duals is true and the objective sense is MIN_SENSE or MAX_SENSE (not FEASIBILITY_SENSE). For linear constraints, the shadow prices differ at most in sign from the dual value depending on the objective sense.\n\nSee also reduced_cost.\n\nNotes\n\nThe function simply translates signs from dual and does not validate the conditions needed to guarantee the sensitivity interpretation of the shadow price. The caller is responsible, e.g., for checking whether the solver converged to an optimal primal-dual pair or a proof of infeasibility.\nThe computation is based on the current objective sense of the model. If this has changed since the last solve, the results will be incorrect.\nRelaxation of equality constraints (and hence the shadow price) is defined based on which sense of the equality constraint is active.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#shape","page":"JuMP","title":"shape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.shape","category":"page"},{"location":"api/JuMP/#JuMP.shape","page":"JuMP","title":"JuMP.shape","text":"shape(c::AbstractConstraint)::AbstractShape\n\nReturn the shape of the constraint c.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_backend_summary","page":"JuMP","title":"show_backend_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_backend_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_backend_summary","page":"JuMP","title":"JuMP.show_backend_summary","text":"show_backend_summary(io::IO, model::GenericModel)\n\nPrint a summary of the optimizer backing model.\n\nAbstractModels should implement this method.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_constraints_summary","page":"JuMP","title":"show_constraints_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_constraints_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_constraints_summary","page":"JuMP","title":"JuMP.show_constraints_summary","text":"show_constraints_summary(io::IO, model::AbstractModel)\n\nWrite to io a summary of the number of constraints.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_objective_function_summary","page":"JuMP","title":"show_objective_function_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_objective_function_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_objective_function_summary","page":"JuMP","title":"JuMP.show_objective_function_summary","text":"show_objective_function_summary(io::IO, model::AbstractModel)\n\nWrite to io a summary of the objective function type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#simplex_iterations","page":"JuMP","title":"simplex_iterations","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.simplex_iterations","category":"page"},{"location":"api/JuMP/#JuMP.simplex_iterations","page":"JuMP","title":"JuMP.simplex_iterations","text":"simplex_iterations(model::GenericModel)\n\nGets the cumulative number of simplex iterations during the most-recent optimization.\n\nSolvers must implement MOI.SimplexIterations() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solution_summary","page":"JuMP","title":"solution_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solution_summary","category":"page"},{"location":"api/JuMP/#JuMP.solution_summary","page":"JuMP","title":"JuMP.solution_summary","text":"solution_summary(model::GenericModel; result::Int = 1, verbose::Bool = false)\n\nReturn a struct that can be used print a summary of the solution in result result.\n\nIf verbose=true, write out the primal solution for every variable and the dual solution for every constraint, excluding those with empty names.\n\nExample\n\nWhen called at the REPL, the summary is automatically printed:\n\njulia> model = Model();\n\njulia> solution_summary(model)\n* Solver : No optimizer attached.\n\n* Status\n  Result count       : 0\n  Termination status : OPTIMIZE_NOT_CALLED\n  Message from the solver:\n  \"optimize not called\"\n\n* Candidate solution (result #1)\n  Primal status      : NO_SOLUTION\n  Dual status        : NO_SOLUTION\n\n* Work counters\n\nUse print to force the printing of the summary from inside a function:\n\njulia> model = Model();\n\njulia> function foo(model)\n           print(solution_summary(model))\n           return\n       end\nfoo (generic function with 1 method)\n\njulia> foo(model)\n* Solver : No optimizer attached.\n\n* Status\n  Result count       : 0\n  Termination status : OPTIMIZE_NOT_CALLED\n  Message from the solver:\n  \"optimize not called\"\n\n* Candidate solution (result #1)\n  Primal status      : NO_SOLUTION\n  Dual status        : NO_SOLUTION\n\n* Work counters\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solve_time","page":"JuMP","title":"solve_time","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solve_time","category":"page"},{"location":"api/JuMP/#JuMP.solve_time","page":"JuMP","title":"JuMP.solve_time","text":"solve_time(model::GenericModel)\n\nIf available, returns the solve time reported by the solver. Returns \"ArgumentError: ModelLike of type Solver.Optimizer does not support accessing the attribute MathOptInterface.SolveTimeSec()\" if the attribute is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solver_name","page":"JuMP","title":"solver_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solver_name","category":"page"},{"location":"api/JuMP/#JuMP.solver_name","page":"JuMP","title":"JuMP.solver_name","text":"solver_name(model::GenericModel)\n\nIf available, returns the SolverName property of the underlying optimizer.\n\nReturns \"No optimizer attached\" in AUTOMATIC or MANUAL modes when no optimizer is attached.\n\nReturns \"SolverName() attribute not implemented by the optimizer.\" if the attribute is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#start_value","page":"JuMP","title":"start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.start_value","category":"page"},{"location":"api/JuMP/#JuMP.start_value","page":"JuMP","title":"JuMP.start_value","text":"start_value(con_ref::ConstraintRef)\n\nReturn the primal start value (MOI.ConstraintPrimalStart) of the constraint con_ref.\n\nNote: If no primal start value has been set, start_value will return nothing.\n\nSee also set_start_value.\n\n\n\n\n\nstart_value(v::GenericVariableRef)\n\nReturn the start value (MOI attribute VariablePrimalStart) of the variable v.\n\nNote: VariablePrimalStarts are sometimes called \"MIP-starts\" or \"warmstarts\".\n\nSee also set_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#termination_status","page":"JuMP","title":"termination_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.termination_status","category":"page"},{"location":"api/JuMP/#JuMP.termination_status","page":"JuMP","title":"JuMP.termination_status","text":"termination_status(model::GenericModel)\n\nReturn a MOI.TerminationStatusCode describing why the solver stopped (i.e., the MOI.TerminationStatus attribute).\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#time_limit_sec","page":"JuMP","title":"time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.time_limit_sec","page":"JuMP","title":"JuMP.time_limit_sec","text":"time_limit_sec(model::GenericModel)\n\nReturn the time limit (in seconds) of the model.\n\nReturns nothing if unset.\n\nSee also: set_time_limit_sec, unset_time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#triangle_vec","page":"JuMP","title":"triangle_vec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.triangle_vec","category":"page"},{"location":"api/JuMP/#JuMP.triangle_vec","page":"JuMP","title":"JuMP.triangle_vec","text":"triangle_vec(matrix::Matrix)\n\nReturn the upper triangle of a matrix concatenated into a vector in the order required by JuMP and MathOptInterface for Triangle sets.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, X[1:3, 1:3], Symmetric);\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(3))\n[t, X[1,1], X[1,2], X[2,2], X[1,3], X[2,3], X[3,3]] ∈ MathOptInterface.RootDetConeTriangle(3)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unfix","page":"JuMP","title":"unfix","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unfix","category":"page"},{"location":"api/JuMP/#JuMP.unfix","page":"JuMP","title":"JuMP.unfix","text":"unfix(v::GenericVariableRef)\n\nDelete the fixing constraint of a variable.\n\nError if one does not exist.\n\nSee also FixRef, is_fixed, fix_value, fix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> is_fixed(x)\ntrue\n\njulia> unfix(x)\n\njulia> is_fixed(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unregister","page":"JuMP","title":"unregister","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unregister","category":"page"},{"location":"api/JuMP/#JuMP.unregister","page":"JuMP","title":"JuMP.unregister","text":"unregister(model::GenericModel, key::Symbol)\n\nUnregister the name key from model so that a new variable, constraint, or expression can be created with the same key.\n\nNote that this will not delete the object model[key]; it will just remove the reference at model[key]. To delete the object, use delete as well.\n\nSee also: delete, object_dictionary.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n\nStacktrace:\n[...]\n\njulia> num_variables(model)\n1\n\njulia> unregister(model, :x)\n\njulia> @variable(model, x)\nx\n\njulia> num_variables(model)\n2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unsafe_backend","page":"JuMP","title":"unsafe_backend","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unsafe_backend","category":"page"},{"location":"api/JuMP/#JuMP.unsafe_backend","page":"JuMP","title":"JuMP.unsafe_backend","text":"unsafe_backend(model::GenericModel)\n\nReturn the innermost optimizer associated with the JuMP model model.\n\nThis function should only be used by advanced users looking to access low-level solver-specific functionality. It has a high-risk of incorrect usage. We strongly suggest you use the alternative suggested below.\n\nSee also: backend.\n\nUnsafe behavior\n\nThis function is unsafe for two main reasons.\n\nFirst, the formulation and order of variables and constraints in the unsafe backend may be different to the variables and constraints in model. This can happen because of bridges, or because the solver requires the variables or constraints in a specific order. In addition, the variable or constraint index returned by index at the JuMP level may be different to the index of the corresponding variable or constraint in the unsafe_backend. There is no solution to this. Use the alternative suggested below instead.\n\nSecond, the unsafe_backend may be empty, or lack some modifications made to the JuMP model. Thus, before calling unsafe_backend you should first call MOI.Utilities.attach_optimizer to ensure that the backend is synchronized with the JuMP model.\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS\n\njulia> MOI.Utilities.attach_optimizer(model)\n\njulia> inner = unsafe_backend(model)\nA HiGHS model with 0 columns and 0 rows.\n\nMoreover, if you modify the JuMP model, the reference you have to the backend (i.e., inner in the example above) may be out-dated, and you should call MOI.Utilities.attach_optimizer again.\n\nThis function is also unsafe in the reverse direction: if you modify the unsafe backend, e.g., by adding a new constraint to inner, the changes may be silently discarded by JuMP when the JuMP model is modified or solved.\n\nAlternative\n\nInstead of unsafe_backend, create a model using direct_model and call backend instead.\n\nFor example, instead of:\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> MOI.Utilities.attach_optimizer(model)\n\njulia> highs = unsafe_backend(model)\nA HiGHS model with 1 columns and 0 rows.\n\nUse:\n\njulia> import HiGHS\n\njulia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> highs = backend(model)  # No need to call `attach_optimizer`.\nA HiGHS model with 1 columns and 0 rows.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_binary","page":"JuMP","title":"unset_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_binary","category":"page"},{"location":"api/JuMP/#JuMP.unset_binary","page":"JuMP","title":"JuMP.unset_binary","text":"unset_binary(variable_ref::GenericVariableRef)\n\nRemove the binary constraint on the variable variable_ref.\n\nSee also BinaryRef, is_binary, set_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> is_binary(x)\ntrue\n\njulia> unset_binary(x)\n\njulia> is_binary(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_integer","page":"JuMP","title":"unset_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_integer","category":"page"},{"location":"api/JuMP/#JuMP.unset_integer","page":"JuMP","title":"JuMP.unset_integer","text":"unset_integer(variable_ref::GenericVariableRef)\n\nRemove the integrality constraint on the variable variable_ref.\n\nErrors if one does not exist.\n\nSee also IntegerRef, is_integer, set_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Int);\n\njulia> is_integer(x)\ntrue\n\njulia> unset_integer(x)\n\njulia> is_integer(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_silent","page":"JuMP","title":"unset_silent","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_silent","category":"page"},{"location":"api/JuMP/#JuMP.unset_silent","page":"JuMP","title":"JuMP.unset_silent","text":"unset_silent(model::GenericModel)\n\nNeutralize the effect of the set_silent function and let the solver attributes control the verbosity.\n\nSee also: set_silent.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_time_limit_sec","page":"JuMP","title":"unset_time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.unset_time_limit_sec","page":"JuMP","title":"JuMP.unset_time_limit_sec","text":"unset_time_limit_sec(model::GenericModel)\n\nUnset the time limit of the solver.\n\nSee also: set_time_limit_sec, time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#upper_bound","page":"JuMP","title":"upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.upper_bound","page":"JuMP","title":"JuMP.upper_bound","text":"upper_bound(v::GenericVariableRef)\n\nReturn the upper bound of a variable.\n\nError if one does not exist.\n\nSee also UpperBoundRef, has_upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> upper_bound(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#value","page":"JuMP","title":"value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.value","category":"page"},{"location":"api/JuMP/#JuMP.value","page":"JuMP","title":"JuMP.value","text":"value(con_ref::ConstraintRef; result::Int = 1)\n\nReturn the primal value of constraint con_ref associated with result index result of the most-recent solution returned by the solver.\n\nThat is, if con_ref is the reference of a constraint func-in-set, it returns the value of func evaluated at the value of the variables (given by value(::GenericVariableRef)).\n\nUse has_values to check if a result exists before asking for values.\n\nSee also: result_count.\n\nNote\n\nFor scalar constraints, the constant is moved to the set so it is not taken into account in the primal value of the constraint. For instance, the constraint @constraint(model, 2x + 3y + 1 == 5) is transformed into 2x + 3y-in-MOI.EqualTo(4) so the value returned by this function is the evaluation of 2x + 3y.\n\n\n\n\n\nvalue(var_value::Function, con_ref::ConstraintRef)\n\nEvaluate the primal value of the constraint con_ref using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericVariableRef; result = 1)\n\nReturn the value of variable v associated with result index result of the most-recent returned by the solver.\n\nUse has_values to check if a result exists before asking for values.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, v::GenericVariableRef)\n\nEvaluate the value of the variable v as var_value(v).\n\n\n\n\n\nvalue(var_value::Function, ex::GenericAffExpr)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericAffExpr; result::Int = 1)\n\nReturn the value of the GenericAffExpr v associated with result index result of the most-recent solution returned by the solver.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, ex::GenericQuadExpr)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericQuadExpr; result::Int = 1)\n\nReturn the value of the GenericQuadExpr v associated with result index result of the most-recent solution returned by the solver.\n\nReplaces getvalue for most use cases.\n\nSee also: result_count.\n\n\n\n\n\nvalue(p::NonlinearParameter)\n\nReturn the current value stored in the nonlinear parameter p.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, p == 10)\np == 10.0\n\njulia> value(p)\n10.0\n\n\n\n\n\nvalue(ex::NonlinearExpression; result::Int = 1)\n\nReturn the value of the NonlinearExpression ex associated with result index result of the most-recent solution returned by the solver.\n\nReplaces getvalue for most use cases.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, ex::NonlinearExpression)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(c::NonlinearConstraintRef; result::Int = 1)\n\nReturn the value of the NonlinearConstraintRef c associated with result index result of the most-recent solution returned by the solver.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, c::NonlinearConstraintRef)\n\nEvaluate c using var_value(v) as the value for each variable v.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#value_type","page":"JuMP","title":"value_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.value_type","category":"page"},{"location":"api/JuMP/#JuMP.value_type","page":"JuMP","title":"JuMP.value_type","text":"value_type(::Type{<:Union{AbstractModel,AbstractVariableRef}})\n\nReturn the return type of value for variables of that model. It defaults to Float64 if it is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#variable_by_name","page":"JuMP","title":"variable_by_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.variable_by_name","category":"page"},{"location":"api/JuMP/#JuMP.variable_by_name","page":"JuMP","title":"JuMP.variable_by_name","text":"variable_by_name(\n    model::AbstractModel,\n    name::String,\n)::Union{AbstractVariableRef,Nothing}\n\nReturns the reference of the variable with name attribute name or Nothing if no variable has this name attribute. Throws an error if several variables have name as their name attribute.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> variable_by_name(model, \"x\")\nx\n\njulia> @variable(model, base_name=\"x\")\nx\n\njulia> variable_by_name(model, \"x\")\nERROR: Multiple variables have the name x.\nStacktrace:\n [1] error(::String) at ./error.jl:33\n [2] get(::MOIU.Model{Float64}, ::Type{MathOptInterface.VariableIndex}, ::String) at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/model.jl:222\n [3] get at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/universalfallback.jl:201 [inlined]\n [4] get(::MathOptInterface.Utilities.CachingOptimizer{MathOptInterface.AbstractOptimizer,MathOptInterface.Utilities.UniversalFallback{MOIU.Model{Float64}}}, ::Type{MathOptInterface.VariableIndex}, ::String) at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/cachingoptimizer.jl:490\n [5] variable_by_name(::GenericModel, ::String) at /home/blegat/.julia/dev/JuMP/src/variables.jl:268\n [6] top-level scope at none:0\n\njulia> var = @variable(model, base_name=\"y\")\ny\n\njulia> variable_by_name(model, \"y\")\ny\n\njulia> set_name(var, \"z\")\n\njulia> variable_by_name(model, \"y\")\n\njulia> variable_by_name(model, \"z\")\nz\n\njulia> @variable(model, u[1:2])\n2-element Vector{VariableRef}:\n u[1]\n u[2]\n\njulia> variable_by_name(model, \"u[2]\")\nu[2]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#variable_ref_type","page":"JuMP","title":"variable_ref_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.variable_ref_type","category":"page"},{"location":"api/JuMP/#JuMP.variable_ref_type","page":"JuMP","title":"JuMP.variable_ref_type","text":"variable_ref_type(::Union{F,Type{F}}) where {F}\n\nA helper function used internally by JuMP and some JuMP extensions. Returns the variable type associated with the model or expression type F.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#vectorize","page":"JuMP","title":"vectorize","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.vectorize","category":"page"},{"location":"api/JuMP/#JuMP.vectorize","page":"JuMP","title":"JuMP.vectorize","text":"vectorize(matrix::AbstractMatrix, ::Shape)\n\nConvert the matrix into a vector according to Shape.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#write_to_file","page":"JuMP","title":"write_to_file","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.write_to_file","category":"page"},{"location":"api/JuMP/#JuMP.write_to_file","page":"JuMP","title":"JuMP.write_to_file","text":"write_to_file(\n    model::GenericModel,\n    filename::String;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,\n    kwargs...,\n)\n\nWrite the JuMP model model to filename in the format format.\n\nIf the filename ends in .gz, it will be compressed using Gzip. If the filename ends in .bz2, it will be compressed using BZip2.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#AbstractConstraint","page":"JuMP","title":"AbstractConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractConstraint","category":"page"},{"location":"api/JuMP/#JuMP.AbstractConstraint","page":"JuMP","title":"JuMP.AbstractConstraint","text":"abstract type AbstractConstraint\n\nAn abstract base type for all constraint types. AbstractConstraints store the function and set directly, unlike ConstraintRefs that are merely references to constraints stored in a model. AbstractConstraints do not need to be attached to a model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractJuMPScalar","page":"JuMP","title":"AbstractJuMPScalar","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractJuMPScalar","category":"page"},{"location":"api/JuMP/#JuMP.AbstractJuMPScalar","page":"JuMP","title":"JuMP.AbstractJuMPScalar","text":"AbstractJuMPScalar <: MutableArithmetics.AbstractMutable\n\nAbstract base type for all scalar types\n\nThe subtyping of AbstractMutable will allow calls of some Base functions to be redirected to a method in MA that handles type promotion more carefully (e.g. the promotion in sparse matrix products in SparseArrays usually does not work for JuMP types) and exploits the mutability of AffExpr and QuadExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractModel","page":"JuMP","title":"AbstractModel","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractModel","category":"page"},{"location":"api/JuMP/#JuMP.AbstractModel","page":"JuMP","title":"JuMP.AbstractModel","text":"AbstractModel\n\nAn abstract type that should be subtyped for users creating JuMP extensions.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractScalarSet","page":"JuMP","title":"AbstractScalarSet","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractScalarSet","category":"page"},{"location":"api/JuMP/#JuMP.AbstractScalarSet","page":"JuMP","title":"JuMP.AbstractScalarSet","text":"AbstractScalarSet\n\nAn abstract type for defining new scalar sets in JuMP.\n\nImplement moi_set(::AbstractScalarSet) to convert the type into an MOI set.\n\nSee also: moi_set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractShape","page":"JuMP","title":"AbstractShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractShape","category":"page"},{"location":"api/JuMP/#JuMP.AbstractShape","page":"JuMP","title":"JuMP.AbstractShape","text":"AbstractShape\n\nAbstract vectorizable shape. Given a flat vector form of an object of shape shape, the original object can be obtained by reshape_vector.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVariable","page":"JuMP","title":"AbstractVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVariable","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVariable","page":"JuMP","title":"JuMP.AbstractVariable","text":"AbstractVariable\n\nVariable returned by build_variable. It represents a variable that has not been added yet to any model. It can be added to a given model with add_variable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVariableRef","page":"JuMP","title":"AbstractVariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVariableRef","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVariableRef","page":"JuMP","title":"JuMP.AbstractVariableRef","text":"AbstractVariableRef\n\nVariable returned by add_variable. Affine (resp. quadratic) operations with variables of type V<:AbstractVariableRef and coefficients of type T     create a GenericAffExpr{T,V} (resp. GenericQuadExpr{T,V}).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVectorSet","page":"JuMP","title":"AbstractVectorSet","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVectorSet","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVectorSet","page":"JuMP","title":"JuMP.AbstractVectorSet","text":"AbstractVectorSet\n\nAn abstract type for defining new sets in JuMP.\n\nImplement moi_set(::AbstractVectorSet, dim::Int) to convert the type into an MOI set.\n\nSee also: moi_set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AffExpr","page":"JuMP","title":"AffExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AffExpr","category":"page"},{"location":"api/JuMP/#JuMP.AffExpr","page":"JuMP","title":"JuMP.AffExpr","text":"AffExpr\n\nAlias for GenericAffExpr{Float64,VariableRef}, the specific GenericAffExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#BinaryRef","page":"JuMP","title":"BinaryRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.BinaryRef","category":"page"},{"location":"api/JuMP/#JuMP.BinaryRef","page":"JuMP","title":"JuMP.BinaryRef","text":"BinaryRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining v to be binary. Errors if one does not exist.\n\nSee also is_binary, set_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> BinaryRef(x)\nx binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#BridgeableConstraint","page":"JuMP","title":"BridgeableConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.BridgeableConstraint","category":"page"},{"location":"api/JuMP/#JuMP.BridgeableConstraint","page":"JuMP","title":"JuMP.BridgeableConstraint","text":"BridgeableConstraint(\n    constraint::C,\n    bridge_type::B;\n    coefficient_type::Type{T} = Float64,\n) where {C<:AbstractConstraint,B<:Type{<:MOI.Bridges.AbstractBridge},T}\n\nAn AbstractConstraint representinng that constraint that can be bridged by the bridge of type bridge_type{coefficient_type}.\n\nAdding a BridgeableConstraint to a model is equivalent to:\n\nadd_bridge(model, bridge_type; coefficient_type = coefficient_type)\nadd_constraint(model, constraint)\n\nExample\n\nGiven a new scalar set type CustomSet with a bridge CustomBridge that can bridge F-in-CustomSet constraints, when the user does:\n\nmodel = Model()\n@variable(model, x)\n@constraint(model, x + 1 in CustomSet())\noptimize!(model)\n\nwith an optimizer that does not support F-in-CustomSet constraints, the constraint will not be bridged unless they first call add_bridge(model, CustomBridge).\n\nIn order to automatically add the CustomBridge to any model to which an F-in-CustomSet is added, add the following method:\n\nfunction JuMP.build_constraint(\n    _error::Function,\n    func::AbstractJuMPScalar,\n    set::CustomSet,\n)\n    constraint = ScalarConstraint(func, set)\n    return BridgeableConstraint(constraint, CustomBridge)\nend\n\nNote\n\nJuMP extensions should extend JuMP.build_constraint only if they also defined CustomSet, for three reasons:\n\nIt is problematic if multiple extensions overload the same JuMP method.\nA missing method will not inform the users that they forgot to load the extension module defining the build_constraint method.\nDefining a method where neither the function nor any of the argument types are defined in the package is called type piracy and is discouraged in the Julia style guide.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ComplexPlane","page":"JuMP","title":"ComplexPlane","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ComplexPlane","category":"page"},{"location":"api/JuMP/#JuMP.ComplexPlane","page":"JuMP","title":"JuMP.ComplexPlane","text":"ComplexPlane\n\nComplex plane object that can be used to create a complex variable in the @variable macro.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, x in ComplexPlane())\nreal(x) + imag(x) im\n\njulia> all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\nWe see in the output of the last command that two real variables were created. The Julia variable x binds to an affine expression in terms of these two variables that parametrize the complex plane.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ComplexVariable","page":"JuMP","title":"ComplexVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ComplexVariable","category":"page"},{"location":"api/JuMP/#JuMP.ComplexVariable","page":"JuMP","title":"JuMP.ComplexVariable","text":"ComplexVariable{S,T,U,V} <: AbstractVariable\n\nA struct used when adding complex variables.\n\nSee also: ComplexPlane.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ConstraintNotOwned","page":"JuMP","title":"ConstraintNotOwned","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ConstraintNotOwned","category":"page"},{"location":"api/JuMP/#JuMP.ConstraintNotOwned","page":"JuMP","title":"JuMP.ConstraintNotOwned","text":"struct ConstraintNotOwned{C <: ConstraintRef} <: Exception\n    constraint_ref::C\nend\n\nThe constraint constraint_ref was used in a model different to owner_model(constraint_ref).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ConstraintRef","page":"JuMP","title":"ConstraintRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ConstraintRef","category":"page"},{"location":"api/JuMP/#JuMP.ConstraintRef","page":"JuMP","title":"JuMP.ConstraintRef","text":"ConstraintRef\n\nHolds a reference to the model and the corresponding MOI.ConstraintIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#FixRef","page":"JuMP","title":"FixRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FixRef","category":"page"},{"location":"api/JuMP/#JuMP.FixRef","page":"JuMP","title":"JuMP.FixRef","text":"FixRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint fixing the value of v.\n\nErrors if one does not exist.\n\nSee also is_fixed, fix_value, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> FixRef(x)\nx = 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#GenericAffExpr","page":"JuMP","title":"GenericAffExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericAffExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericAffExpr","page":"JuMP","title":"JuMP.GenericAffExpr","text":"mutable struct GenericAffExpr{CoefType,VarType} <: AbstractJuMPScalar\n    constant::CoefType\n    terms::OrderedDict{VarType,CoefType}\nend\n\nAn expression type representing an affine expression of the form: sum a_i x_i + c.\n\nFields\n\n.constant: the constant c in the expression.\n.terms: an OrderedDict, with keys of VarType and values of CoefType describing the sparse vector a.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericModel","page":"JuMP","title":"GenericModel","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericModel","category":"page"},{"location":"api/JuMP/#JuMP.GenericModel","page":"JuMP","title":"JuMP.GenericModel","text":"GenericModel{T}(\n    [optimizer_factory;]\n    add_bridges::Bool = true,\n) where {T<:Real}\n\nCreate a new instance of a JuMP model.\n\nIf optimizer_factory is provided, the model is initialized with the optimizer returned by MOI.instantiate(optimizer_factory).\n\nIf optimizer_factory is not provided, use set_optimizer to set the optimizer before calling optimize!.\n\nIf add_bridges, JuMP adds a MOI.Bridges.LazyBridgeOptimizer to automatically reformulate the problem into a form supported by the optimizer.\n\nValue type T\n\nPassing a type other than Float64 as the value type T is an advanced operation. The value type must match that expected by the chosen optimizer. Consult the optimizers documentation for details.\n\nIf not documented, assume that the optimizer supports only Float64.\n\nChoosing an unsupported value type will throw an MOI.UnsupportedConstraint or an MOI.UnsupportedAttribute error, the timing of which (during the model construction or during a call to optimize!) depends on how the solver is interfaced to JuMP.\n\nExample\n\njulia> model = GenericModel{BigFloat}();\n\njulia> typeof(model)\nGenericModel{BigFloat}\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericNonlinearExpr","page":"JuMP","title":"GenericNonlinearExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericNonlinearExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericNonlinearExpr","page":"JuMP","title":"JuMP.GenericNonlinearExpr","text":"GenericNonlinearExpr{V}(head::Symbol, args::Vector{Any})\nGenericNonlinearExpr{V}(head::Symbol, args::Any...)\n\nThe scalar-valued nonlinear function head(args...), represented as a symbolic expression tree, with the call operator head and ordered arguments in args.\n\nV is the type of AbstractVariableRef present in the expression, and is used to help dispatch JuMP extensions.\n\nhead\n\nThe head::Symbol must be an operator supported by the model.\n\nThe default list of supported univariate operators is given by:\n\nMOI.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS\n\nand the default list of supported multivariate operators is given by:\n\nMOI.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n\nAdditional operators can be add using @operator.\n\nSee the full list of operators supported by a MOI.ModelLike by querying the MOI.ListOfSupportedNonlinearOperators attribute.\n\nargs\n\nThe vector args contains the arguments to the nonlinear function. If the operator is univariate, it must contain one element. Otherwise, it may contain multiple elements.\n\nGiven a subtype of AbstractVariableRef, V, for GenericNonlinearExpr{V}, each element must be one of the following:\n\nA constant value of type <:Real\nA V\nA GenericAffExpr{T,V}\nA GenericQuadExpr{T,V}\nA GenericNonlinearExpr{V}\n\nwhere T<:Real and T == value_type(V).\n\nUnsupported operators\n\nIf the optimizer does not support head, an MOI.UnsupportedNonlinearOperator error will be thrown.\n\nThere is no guarantee about when this error will be thrown; it may be thrown when the function is first added to the model, or it may be thrown when optimize! is called.\n\nExample\n\nTo represent the function f(x) = sin(x)^2, do:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f = sin(x)^2\nsin(x) ^ 2.0\n\njulia> f = GenericNonlinearExpr{VariableRef}(\n           :^,\n           GenericNonlinearExpr{VariableRef}(:sin, x),\n           2.0,\n       )\nsin(x) ^ 2.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericQuadExpr","page":"JuMP","title":"GenericQuadExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericQuadExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericQuadExpr","page":"JuMP","title":"JuMP.GenericQuadExpr","text":"mutable struct GenericQuadExpr{CoefType,VarType} <: AbstractJuMPScalar\n    aff::GenericAffExpr{CoefType,VarType}\n    terms::OrderedDict{UnorderedPair{VarType}, CoefType}\nend\n\nAn expression type representing an quadratic expression of the form: sum q_ij x_i x_j + sum a_i x_i + c.\n\nFields\n\n.aff: an GenericAffExpr representing the affine portion of the expression.\n.terms: an OrderedDict, with keys of UnorderedPair{VarType} and values of CoefType, describing the sparse list of terms q.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericReferenceMap","page":"JuMP","title":"GenericReferenceMap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericReferenceMap","category":"page"},{"location":"api/JuMP/#JuMP.GenericReferenceMap","page":"JuMP","title":"JuMP.GenericReferenceMap","text":"GenericReferenceMap{T}\n\nMapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericVariableRef","page":"JuMP","title":"GenericVariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericVariableRef","category":"page"},{"location":"api/JuMP/#JuMP.GenericVariableRef","page":"JuMP","title":"JuMP.GenericVariableRef","text":"GenericVariableRef{T} <: AbstractVariableRef\n\nHolds a reference to the model and the corresponding MOI.VariableIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianMatrixShape","page":"JuMP","title":"HermitianMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.HermitianMatrixShape","page":"JuMP","title":"JuMP.HermitianMatrixShape","text":"HermitianMatrixShape\n\nShape object for a Hermitian square matrix of side_dimension rows and columns. The vectorized form corresponds to MOI.HermitianPositiveSemidefiniteConeTriangle.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianMatrixSpace","page":"JuMP","title":"HermitianMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.HermitianMatrixSpace","page":"JuMP","title":"JuMP.HermitianMatrixSpace","text":"HermitianMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be hermitian.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im\n real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianPSDCone","page":"JuMP","title":"HermitianPSDCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianPSDCone","category":"page"},{"location":"api/JuMP/#JuMP.HermitianPSDCone","page":"JuMP","title":"JuMP.HermitianPSDCone","text":"HermitianPSDCone\n\nHermitian positive semidefinite cone object that can be used to create a Hermitian positive semidefinite square matrix in the @variable and @constraint macros.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, H[1:3, 1:3] in HermitianPSDCone())\n3×3 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    …  real(H[1,3]) + imag(H[1,3]) im\n real(H[1,2]) - imag(H[1,2]) im     real(H[2,3]) + imag(H[2,3]) im\n real(H[1,3]) - imag(H[1,3]) im     real(H[3,3])\n\njulia> all_variables(model)\n9-element Vector{VariableRef}:\n real(H[1,1])\n real(H[1,2])\n real(H[2,2])\n real(H[1,3])\n real(H[2,3])\n real(H[3,3])\n imag(H[1,2])\n imag(H[1,3])\n imag(H[2,3])\n\njulia> all_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle)\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.HermitianPositiveSemidefiniteConeTriangle}}}:\n [real(H[1,1]), real(H[1,2]), real(H[2,2]), real(H[1,3]), real(H[2,3]), real(H[3,3]), imag(H[1,2]), imag(H[1,3]), imag(H[2,3])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(3)\n\nWe see in the output of the last commands that 9 real variables were created. The matrix H contrains affine expressions in terms of these 9 variables that parametrize a Hermitian matrix.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#IntegerRef","page":"JuMP","title":"IntegerRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.IntegerRef","category":"page"},{"location":"api/JuMP/#JuMP.IntegerRef","page":"JuMP","title":"JuMP.IntegerRef","text":"IntegerRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining v to be integer.\n\nErrors if one does not exist.\n\nSee also is_integer, set_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Int);\n\njulia> IntegerRef(x)\nx integer\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#LinearTermIterator","page":"JuMP","title":"LinearTermIterator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LinearTermIterator","category":"page"},{"location":"api/JuMP/#JuMP.LinearTermIterator","page":"JuMP","title":"JuMP.LinearTermIterator","text":"LinearTermIterator{GAE<:GenericAffExpr}\n\nA struct that implements the iterate protocol in order to iterate over tuples of (coefficient, variable) in the GenericAffExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#LowerBoundRef","page":"JuMP","title":"LowerBoundRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LowerBoundRef","category":"page"},{"location":"api/JuMP/#JuMP.LowerBoundRef","page":"JuMP","title":"JuMP.LowerBoundRef","text":"LowerBoundRef(v::GenericVariableRef)\n\nReturn a constraint reference to the lower bound constraint of v.\n\nErrors if one does not exist.\n\nSee also has_lower_bound, lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> LowerBoundRef(x)\nx ≥ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#Model","page":"JuMP","title":"Model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Model","category":"page"},{"location":"api/JuMP/#JuMP.Model","page":"JuMP","title":"JuMP.Model","text":"Model([optimizer_factory;] add_bridges::Bool = true)\n\nCreate a new instance of a JuMP model.\n\nIf optimizer_factory is provided, the model is initialized with thhe optimizer returned by MOI.instantiate(optimizer_factory).\n\nIf optimizer_factory is not provided, use set_optimizer to set the optimizer before calling optimize!.\n\nIf add_bridges, JuMP adds a MOI.Bridges.LazyBridgeOptimizer to automatically reformulate the problem into a form supported by the optimizer.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> solver_name(model)\n\"Ipopt\"\n\njulia> import HiGHS\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer); add_bridges = false);\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ModelMode","page":"JuMP","title":"ModelMode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ModelMode","category":"page"},{"location":"api/JuMP/#JuMP.ModelMode","page":"JuMP","title":"JuMP.ModelMode","text":"ModelMode\n\nAn enum to describe the state of the CachingOptimizer inside a JuMP model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NLPEvaluator","page":"JuMP","title":"NLPEvaluator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NLPEvaluator","category":"page"},{"location":"api/JuMP/#JuMP.NLPEvaluator","page":"JuMP","title":"JuMP.NLPEvaluator","text":"NLPEvaluator(\n    model::Model,\n    _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =\n        MOI.Nonlinear.SparseReverseMode(),\n)\n\nReturn an MOI.AbstractNLPEvaluator constructed from model\n\nwarning: Warning\nBefore using, you must initialize the evaluator using MOI.initialize.\n\nExperimental\n\nThese features may change or be removed in any future version of JuMP.\n\nPass _differentiation_backend to specify the differentiation backend used to compute derivatives.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#NoOptimizer","page":"JuMP","title":"NoOptimizer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NoOptimizer","category":"page"},{"location":"api/JuMP/#JuMP.NoOptimizer","page":"JuMP","title":"JuMP.NoOptimizer","text":"struct NoOptimizer <: Exception end\n\nNo optimizer is set. The optimizer can be provided to the Model constructor or by calling set_optimizer.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearConstraintIndex","page":"JuMP","title":"NonlinearConstraintIndex","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearConstraintIndex","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearConstraintIndex","page":"JuMP","title":"JuMP.NonlinearConstraintIndex","text":"ConstraintIndex\n\nAn index to a nonlinear constraint that is returned by add_constraint.\n\nGiven data::Model and c::ConstraintIndex, use data[c] to retrieve the corresponding Constraint.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearConstraintRef","page":"JuMP","title":"NonlinearConstraintRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearConstraintRef","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearConstraintRef","page":"JuMP","title":"JuMP.NonlinearConstraintRef","text":"NonlinearConstraintRef\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearExpr","page":"JuMP","title":"NonlinearExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearExpr","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearExpr","page":"JuMP","title":"JuMP.NonlinearExpr","text":"NonlinearExpr\n\nAlias for GenericNonlinearExpr{VariableRef}, the specific GenericNonlinearExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearExpression","page":"JuMP","title":"NonlinearExpression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearExpression","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearExpression","page":"JuMP","title":"JuMP.NonlinearExpression","text":"NonlinearExpression <: AbstractJuMPScalar\n\nA struct to represent a nonlinear expression.\n\nCreate an expression using @NLexpression.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearOperator","page":"JuMP","title":"NonlinearOperator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearOperator","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearOperator","page":"JuMP","title":"JuMP.NonlinearOperator","text":"NonlinearOperator(func::Function, head::Symbol)\n\nA callable struct (functor) representing a function named head.\n\nWhen called with AbstractJuMPScalars, the struct returns a GenericNonlinearExpr.\n\nWhen called with non-JuMP types, the struct returns the evaluation of func(args...).\n\nUnless head is special-cased by the optimizer, the operator must have already been added to the model using add_nonlinear_operator or @operator.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :op_f)\n\njulia> bar = NonlinearOperator(f, :op_f)\nNonlinearOperator(f, :op_f)\n\njulia> @objective(model, Min, bar(x))\nop_f(x)\n\njulia> bar(2.0)\n4.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearParameter","page":"JuMP","title":"NonlinearParameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearParameter","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearParameter","page":"JuMP","title":"JuMP.NonlinearParameter","text":"NonlinearParameter <: AbstractJuMPScalar\n\nA struct to represent a nonlinear parameter.\n\nCreate a parameter using @NLparameter.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Nonnegatives","page":"JuMP","title":"Nonnegatives","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Nonnegatives","category":"page"},{"location":"api/JuMP/#JuMP.Nonnegatives","page":"JuMP","title":"JuMP.Nonnegatives","text":"Nonnegatives()\n\nThe JuMP equivalent of the MOI.Nonnegatives set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Nonnegatives())\n[x[1], x[2]] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x >= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Nonpositives","page":"JuMP","title":"Nonpositives","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Nonpositives","category":"page"},{"location":"api/JuMP/#JuMP.Nonpositives","page":"JuMP","title":"JuMP.Nonpositives","text":"Nonpositives()\n\nThe JuMP equivalent of the MOI.Nonpositives set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Nonpositives())\n[x[1], x[2]] ∈ MathOptInterface.Nonpositives(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x <= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#OptimizationSense","page":"JuMP","title":"OptimizationSense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OptimizationSense","category":"page"},{"location":"api/JuMP/#JuMP.OptimizationSense","page":"JuMP","title":"JuMP.OptimizationSense","text":"OptimizationSense\n\nAn enum for the value of the ObjectiveSense attribute.\n\nValues\n\nPossible values are:\n\nMIN_SENSE: the goal is to minimize the objective function\nMAX_SENSE: the goal is to maximize the objective function\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#OptimizeNotCalled","page":"JuMP","title":"OptimizeNotCalled","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OptimizeNotCalled","category":"page"},{"location":"api/JuMP/#JuMP.OptimizeNotCalled","page":"JuMP","title":"JuMP.OptimizeNotCalled","text":"struct OptimizeNotCalled <: Exception end\n\nA result attribute cannot be queried before optimize! is called.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#PSDCone","page":"JuMP","title":"PSDCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.PSDCone","category":"page"},{"location":"api/JuMP/#JuMP.PSDCone","page":"JuMP","title":"JuMP.PSDCone","text":"PSDCone\n\nPositive semidefinite cone object that can be used to constrain a square matrix to be positive semidefinite in the @constraint macro. If the matrix has type Symmetric then the columns vectorization (the vector obtained by concatenating the columns) of its upper triangular part is constrained to belong to the MOI.PositiveSemidefiniteConeTriangle set, otherwise its column vectorization is constrained to belong to the MOI.PositiveSemidefiniteConeSquare set.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> a = [ x 2x\n            2x  x];\n\njulia> b = [1 2\n            2 4];\n\njulia> cref = @constraint(model, a >= b, PSDCone())\n[x - 1    2 x - 2;\n 2 x - 2  x - 4] ∈ PSDCone()\n\njulia> jump_function(constraint_object(cref))\n4-element Vector{AffExpr}:\n x - 1\n 2 x - 2\n 2 x - 2\n x - 4\n\njulia> moi_set(constraint_object(cref))\nMathOptInterface.PositiveSemidefiniteConeSquare(2)\n\nWe see in the output of the last command that the vectorization of the matrix is constrained to belong to the PositiveSemidefiniteConeSquare.\n\njulia> using LinearAlgebra # For Symmetric\n\njulia> cref = @constraint(model, Symmetric(a - b) in PSDCone())\n[x - 1    2 x - 2;\n 2 x - 2  x - 4] ∈ PSDCone()\n\njulia> jump_function(constraint_object(cref))\n3-element Vector{AffExpr}:\n x - 1\n 2 x - 2\n x - 4\n\njulia> moi_set(constraint_object(cref))\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\nAs we see in the output of the last command, the vectorization of only the upper triangular part of the matrix is constrained to belong to the PositiveSemidefiniteConeSquare.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Parameter","page":"JuMP","title":"Parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Parameter","category":"page"},{"location":"api/JuMP/#JuMP.Parameter","page":"JuMP","title":"JuMP.Parameter","text":"Parameter(value)\n\nA short-cut for the MOI.Parameter set.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Parameter(2))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Parameter{Float64}(2.0)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ParameterRef","page":"JuMP","title":"ParameterRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ParameterRef","category":"page"},{"location":"api/JuMP/#JuMP.ParameterRef","page":"JuMP","title":"JuMP.ParameterRef","text":"ParameterRef(x::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining x to be a parameter.\n\nErrors if one does not exist.\n\nSee also is_parameter, set_parameter_value, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> ParameterRef(p)\np ∈ MathOptInterface.Parameter{Float64}(2.0)\n\njulia> @variable(model, x);\n\njulia> ParameterRef(x)\nERROR: Variable x is not a parameter.\nStacktrace:\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#QuadExpr","page":"JuMP","title":"QuadExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.QuadExpr","category":"page"},{"location":"api/JuMP/#JuMP.QuadExpr","page":"JuMP","title":"JuMP.QuadExpr","text":"QuadExpr\n\nAn alias for GenericQuadExpr{Float64,VariableRef}, the specific GenericQuadExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#QuadTermIterator","page":"JuMP","title":"QuadTermIterator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.QuadTermIterator","category":"page"},{"location":"api/JuMP/#JuMP.QuadTermIterator","page":"JuMP","title":"JuMP.QuadTermIterator","text":"QuadTermIterator{GQE<:GenericQuadExpr}\n\nA struct that implements the iterate protocol in order to iterate over tuples of (coefficient, variable, variable) in the GenericQuadExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ReferenceMap","page":"JuMP","title":"ReferenceMap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ReferenceMap","category":"page"},{"location":"api/JuMP/#JuMP.ReferenceMap","page":"JuMP","title":"JuMP.ReferenceMap","text":"GenericReferenceMap{T}\n\nMapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ResultStatusCode","page":"JuMP","title":"ResultStatusCode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ResultStatusCode","category":"page"},{"location":"api/JuMP/#JuMP.ResultStatusCode","page":"JuMP","title":"JuMP.ResultStatusCode","text":"ResultStatusCode\n\nAn Enum of possible values for the PrimalStatus and DualStatus attributes.\n\nThe values indicate how to interpret the result vector.\n\nValues\n\nPossible values are:\n\nNO_SOLUTION: the result vector is empty.\nFEASIBLE_POINT: the result vector is a feasible point.\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\nINFEASIBLE_POINT: the result vector is an infeasible point.\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#RotatedSecondOrderCone","page":"JuMP","title":"RotatedSecondOrderCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.RotatedSecondOrderCone","category":"page"},{"location":"api/JuMP/#JuMP.RotatedSecondOrderCone","page":"JuMP","title":"JuMP.RotatedSecondOrderCone","text":"RotatedSecondOrderCone\n\nRotated second order cone object that can be used to constrain the square of the euclidean norm of a vector x to be less than or equal to 2tu where t and u are nonnegative scalars. This is a shortcut for the MOI.RotatedSecondOrderCone.\n\nExample\n\nThe following constrains (x-1 x-2)^2_2 le 2tx and t x ge 0:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t, x, x-1, x-2] in RotatedSecondOrderCone())\n[t, x, x - 1, x - 2] ∈ MathOptInterface.RotatedSecondOrderCone(4)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SOS1","page":"JuMP","title":"SOS1","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOS1","category":"page"},{"location":"api/JuMP/#JuMP.SOS1","page":"JuMP","title":"JuMP.SOS1","text":"SOS1\n\nSOS1 (Special Ordered Sets type 1) object than can be used to constrain a vector x to a set where at most 1 variable can take a non-zero value, all others being at 0. The weights, when specified, induce an ordering of the variables; as such, they should be unique values. The kth element in the set corresponds to the kth weight in weights. See here for a description of SOS constraints and their potential uses. This is a shortcut for the MathOptInterface.SOS1 set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SOS2","page":"JuMP","title":"SOS2","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOS2","category":"page"},{"location":"api/JuMP/#JuMP.SOS2","page":"JuMP","title":"JuMP.SOS2","text":"SOS2\n\nSOS1 (Special Ordered Sets type 2) object than can be used to constrain a vector x to a set where at most 2 variables can take a non-zero value, all others being at 0. In addition, if two are non-zero these must be consecutive in their ordering. The weights induce an ordering of the variables; as such, they should be unique values. The kth element in the set corresponds to the kth weight in weights. See here for a description of SOS constraints and their potential uses. This is a shortcut for the MathOptInterface.SOS2 set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarConstraint","page":"JuMP","title":"ScalarConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarConstraint","category":"page"},{"location":"api/JuMP/#JuMP.ScalarConstraint","page":"JuMP","title":"JuMP.ScalarConstraint","text":"struct ScalarConstraint\n\nThe data for a scalar constraint. The func field contains a JuMP object representing the function and the set field contains the MOI set. See also the documentation on JuMP's representation of constraints for more background.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarShape","page":"JuMP","title":"ScalarShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarShape","category":"page"},{"location":"api/JuMP/#JuMP.ScalarShape","page":"JuMP","title":"JuMP.ScalarShape","text":"ScalarShape\n\nShape of scalar constraints.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarVariable","page":"JuMP","title":"ScalarVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarVariable","category":"page"},{"location":"api/JuMP/#JuMP.ScalarVariable","page":"JuMP","title":"JuMP.ScalarVariable","text":"ScalarVariable{S,T,U,V} <: AbstractVariable\n\nA struct used when adding variables.\n\nSee also: add_variable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SecondOrderCone","page":"JuMP","title":"SecondOrderCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SecondOrderCone","category":"page"},{"location":"api/JuMP/#JuMP.SecondOrderCone","page":"JuMP","title":"JuMP.SecondOrderCone","text":"SecondOrderCone\n\nSecond order cone object that can be used to constrain the euclidean norm of a vector x to be less than or equal to a nonnegative scalar t. This is a shortcut for the MOI.SecondOrderCone.\n\nExample\n\nThe following constrains (x-1 x-2)_2 le t and t ge 0:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t, x-1, x-2] in SecondOrderCone())\n[t, x - 1, x - 2] ∈ MathOptInterface.SecondOrderCone(3)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Semicontinuous","page":"JuMP","title":"Semicontinuous","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Semicontinuous","category":"page"},{"location":"api/JuMP/#JuMP.Semicontinuous","page":"JuMP","title":"JuMP.Semicontinuous","text":"Semicontinuous(lower, upper)\n\nA short-cut for the MOI.Semicontinuous set.\n\nThis short-cut is useful because it automatically promotes lower and upper to the same type, and converts them into the element type supported by the JuMP model.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Semicontinuous(1, 2))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Semicontinuous{Int64}(1, 2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Semiinteger","page":"JuMP","title":"Semiinteger","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Semiinteger","category":"page"},{"location":"api/JuMP/#JuMP.Semiinteger","page":"JuMP","title":"JuMP.Semiinteger","text":"Semiinteger(lower, upper)\n\nA short-cut for the MOI.Semiinteger set.\n\nThis short-cut is useful because it automatically promotes lower and upper to the same type, and converts them into the element type supported by the JuMP model.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Semiinteger(3, 5))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Semiinteger{Int64}(3, 5)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SensitivityReport","page":"JuMP","title":"SensitivityReport","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SensitivityReport","category":"page"},{"location":"api/JuMP/#JuMP.SensitivityReport","page":"JuMP","title":"JuMP.SensitivityReport","text":"SensitivityReport\n\nSee lp_sensitivity_report.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkewSymmetricMatrixShape","page":"JuMP","title":"SkewSymmetricMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkewSymmetricMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SkewSymmetricMatrixShape","page":"JuMP","title":"JuMP.SkewSymmetricMatrixShape","text":"SkewSymmetricMatrixShape\n\nShape object for a skew symmetric square matrix of side_dimension rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix (without the diagonal) given column by column (or equivalently, the entries of the lower-left triangular part given row by row). The diagonal is zero.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkewSymmetricMatrixSpace","page":"JuMP","title":"SkewSymmetricMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkewSymmetricMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.SkewSymmetricMatrixSpace","page":"JuMP","title":"JuMP.SkewSymmetricMatrixSpace","text":"SkewSymmetricMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be skew-symmetric.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        Q[1,2]\n -Q[1,2]  0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkipModelConvertScalarSetWrapper","page":"JuMP","title":"SkipModelConvertScalarSetWrapper","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkipModelConvertScalarSetWrapper","category":"page"},{"location":"api/JuMP/#JuMP.SkipModelConvertScalarSetWrapper","page":"JuMP","title":"JuMP.SkipModelConvertScalarSetWrapper","text":"SkipModelConvertScalarSetWrapper(set::MOI.AbstractScalarSet)\n\nJuMP uses model_convert to automatically promote MOI.AbstractScalarSet sets to the same value_type as the model.\n\nIn cases there this is undesirable, wrap the set in SkipModelConvertScalarSetWrapper to pass the set un-changed to the solver.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, x in MOI.EqualTo(1 // 2))\nx = 0.5\n\njulia> @constraint(model, x in SkipModelConvertScalarSetWrapper(MOI.EqualTo(1 // 2)))\nx = 1//2\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SquareMatrixShape","page":"JuMP","title":"SquareMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SquareMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SquareMatrixShape","page":"JuMP","title":"JuMP.SquareMatrixShape","text":"SquareMatrixShape\n\nShape object for a square matrix of side_dimension rows and columns. The vectorized form contains the entries of the the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SymmetricMatrixShape","page":"JuMP","title":"SymmetricMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SymmetricMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SymmetricMatrixShape","page":"JuMP","title":"JuMP.SymmetricMatrixShape","text":"SymmetricMatrixShape\n\nShape object for a symmetric square matrix of side_dimension rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SymmetricMatrixSpace","page":"JuMP","title":"SymmetricMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SymmetricMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.SymmetricMatrixSpace","page":"JuMP","title":"JuMP.SymmetricMatrixSpace","text":"SymmetricMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be symmetric.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SymmetricMatrixSpace())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#TerminationStatusCode","page":"JuMP","title":"TerminationStatusCode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.TerminationStatusCode","category":"page"},{"location":"api/JuMP/#JuMP.TerminationStatusCode","page":"JuMP","title":"JuMP.TerminationStatusCode","text":"TerminationStatusCode\n\nAn Enum of possible values for the TerminationStatus attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to optimize!.\n\nValues\n\nPossible values are:\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\nOPTIMAL: The algorithm found a globally optimal solution.\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\nINVALID_MODEL: The algorithm stopped because the model is invalid.\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#UnorderedPair","page":"JuMP","title":"UnorderedPair","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UnorderedPair","category":"page"},{"location":"api/JuMP/#JuMP.UnorderedPair","page":"JuMP","title":"JuMP.UnorderedPair","text":"UnorderedPair(a::T, b::T)\n\nA wrapper type used by GenericQuadExpr with fields .a and .b.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#UpperBoundRef","page":"JuMP","title":"UpperBoundRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UpperBoundRef","category":"page"},{"location":"api/JuMP/#JuMP.UpperBoundRef","page":"JuMP","title":"JuMP.UpperBoundRef","text":"UpperBoundRef(v::GenericVariableRef)\n\nReturn a constraint reference to the upper bound constraint of v.\n\nErrors if one does not exist.\n\nSee also has_upper_bound, upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> UpperBoundRef(x)\nx ≤ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#VariableConstrainedOnCreation","page":"JuMP","title":"VariableConstrainedOnCreation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableConstrainedOnCreation","category":"page"},{"location":"api/JuMP/#JuMP.VariableConstrainedOnCreation","page":"JuMP","title":"JuMP.VariableConstrainedOnCreation","text":"VariableConstrainedOnCreation <: AbstractVariable\n\nVariable scalar_variables constrained to belong to set.\n\nAdding this variable can be understood as doing:\n\nfunction JuMP.add_variable(\n    model::GenericModel,\n    variable::VariableConstrainedOnCreation,\n    names,\n)\n    var_ref = add_variable(model, variable.scalar_variable, name)\n    add_constraint(model, VectorConstraint(var_ref, variable.set))\n    return var_ref\nend\n\nbut adds the variables with MOI.add_constrained_variable(model, variable.set) instead. See the MOI documentation for the difference between adding the variables with MOI.add_constrained_variable and adding them with MOI.add_variable and adding the constraint separately.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableInfo","page":"JuMP","title":"VariableInfo","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableInfo","category":"page"},{"location":"api/JuMP/#JuMP.VariableInfo","page":"JuMP","title":"JuMP.VariableInfo","text":"VariableInfo{S,T,U,V}\n\nA struct by JuMP internally when creating variables. This may also be used by JuMP extensions to create new types of variables.\n\nSee also: ScalarVariable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableNotOwned","page":"JuMP","title":"VariableNotOwned","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableNotOwned","category":"page"},{"location":"api/JuMP/#JuMP.VariableNotOwned","page":"JuMP","title":"JuMP.VariableNotOwned","text":"struct VariableNotOwned{V<:AbstractVariableRef} <: Exception\n    variable::V\nend\n\nThe variable variable was used in a model different to owner_model(variable).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableRef","page":"JuMP","title":"VariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableRef","category":"page"},{"location":"api/JuMP/#JuMP.VariableRef","page":"JuMP","title":"JuMP.VariableRef","text":"GenericVariableRef{T} <: AbstractVariableRef\n\nHolds a reference to the model and the corresponding MOI.VariableIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariablesConstrainedOnCreation","page":"JuMP","title":"VariablesConstrainedOnCreation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariablesConstrainedOnCreation","category":"page"},{"location":"api/JuMP/#JuMP.VariablesConstrainedOnCreation","page":"JuMP","title":"JuMP.VariablesConstrainedOnCreation","text":"VariablesConstrainedOnCreation <: AbstractVariable\n\nVector of variables scalar_variables constrained to belong to set. Adding this variable can be thought as doing:\n\nfunction JuMP.add_variable(\n    model::GenericModel,\n    variable::VariablesConstrainedOnCreation,\n    names,\n)\n    v_names = vectorize(names, variable.shape)\n    var_refs = add_variable.(model, variable.scalar_variables, v_names)\n    add_constraint(model, VectorConstraint(var_refs, variable.set))\n    return reshape_vector(var_refs, variable.shape)\nend\n\nbut adds the variables with MOI.add_constrained_variables(model, variable.set) instead. See the MOI documentation for the difference between adding the variables with MOI.add_constrained_variables and adding them with MOI.add_variables and adding the constraint separately.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VectorConstraint","page":"JuMP","title":"VectorConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VectorConstraint","category":"page"},{"location":"api/JuMP/#JuMP.VectorConstraint","page":"JuMP","title":"JuMP.VectorConstraint","text":"struct VectorConstraint\n\nThe data for a vector constraint. The func field contains a JuMP object representing the function and the set field contains the MOI set. The shape field contains an AbstractShape matching the form in which the constraint was constructed (e.g., by using matrices or flat vectors). See also the documentation on JuMP's representation of constraints.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VectorShape","page":"JuMP","title":"VectorShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VectorShape","category":"page"},{"location":"api/JuMP/#JuMP.VectorShape","page":"JuMP","title":"JuMP.VectorShape","text":"VectorShape\n\nVector for which the vectorized form corresponds exactly to the vector given.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Zeros","page":"JuMP","title":"Zeros","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Zeros","category":"page"},{"location":"api/JuMP/#JuMP.Zeros","page":"JuMP","title":"JuMP.Zeros","text":"Zeros()\n\nThe JuMP equivalent of the MOI.Zeros set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Zeros())\n[x[1], x[2]] ∈ MathOptInterface.Zeros(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x == b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ALMOST_DUAL_INFEASIBLE","page":"JuMP","title":"ALMOST_DUAL_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_DUAL_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_DUAL_INFEASIBLE","page":"JuMP","title":"JuMP.ALMOST_DUAL_INFEASIBLE","text":"ALMOST_DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_INFEASIBLE","page":"JuMP","title":"ALMOST_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_INFEASIBLE","page":"JuMP","title":"JuMP.ALMOST_INFEASIBLE","text":"ALMOST_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_LOCALLY_SOLVED","page":"JuMP","title":"ALMOST_LOCALLY_SOLVED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_LOCALLY_SOLVED","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_LOCALLY_SOLVED","page":"JuMP","title":"JuMP.ALMOST_LOCALLY_SOLVED","text":"ALMOST_LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_OPTIMAL","page":"JuMP","title":"ALMOST_OPTIMAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_OPTIMAL","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_OPTIMAL","page":"JuMP","title":"JuMP.ALMOST_OPTIMAL","text":"ALMOST_OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#AUTOMATIC","page":"JuMP","title":"AUTOMATIC","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AUTOMATIC","category":"page"},{"location":"api/JuMP/#JuMP.AUTOMATIC","page":"JuMP","title":"JuMP.AUTOMATIC","text":"moi_backend field holds a CachingOptimizer in AUTOMATIC mode.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#DIRECT","page":"JuMP","title":"DIRECT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.DIRECT","category":"page"},{"location":"api/JuMP/#JuMP.DIRECT","page":"JuMP","title":"JuMP.DIRECT","text":"moi_backend field holds an AbstractOptimizer. No extra copy of the model is stored. The moi_backend must support add_constraint etc.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#DUAL_INFEASIBLE","page":"JuMP","title":"DUAL_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.DUAL_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.DUAL_INFEASIBLE","page":"JuMP","title":"JuMP.DUAL_INFEASIBLE","text":"DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#FEASIBILITY_SENSE","page":"JuMP","title":"FEASIBILITY_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FEASIBILITY_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.FEASIBILITY_SENSE","page":"JuMP","title":"JuMP.FEASIBILITY_SENSE","text":"FEASIBILITY_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#FEASIBLE_POINT","page":"JuMP","title":"FEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.FEASIBLE_POINT","page":"JuMP","title":"JuMP.FEASIBLE_POINT","text":"FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nFEASIBLE_POINT: the result vector is a feasible point.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"INFEASIBILITY_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBILITY_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"JuMP.INFEASIBILITY_CERTIFICATE","text":"INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE","page":"JuMP","title":"INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE","page":"JuMP","title":"JuMP.INFEASIBLE","text":"INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE_OR_UNBOUNDED","page":"JuMP","title":"INFEASIBLE_OR_UNBOUNDED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE_OR_UNBOUNDED","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE_OR_UNBOUNDED","page":"JuMP","title":"JuMP.INFEASIBLE_OR_UNBOUNDED","text":"INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE_POINT","page":"JuMP","title":"INFEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE_POINT","page":"JuMP","title":"JuMP.INFEASIBLE_POINT","text":"INFEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBLE_POINT: the result vector is an infeasible point.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INTERRUPTED","page":"JuMP","title":"INTERRUPTED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INTERRUPTED","category":"page"},{"location":"api/JuMP/#JuMP.INTERRUPTED","page":"JuMP","title":"JuMP.INTERRUPTED","text":"INTERRUPTED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INVALID_MODEL","page":"JuMP","title":"INVALID_MODEL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INVALID_MODEL","category":"page"},{"location":"api/JuMP/#JuMP.INVALID_MODEL","page":"JuMP","title":"JuMP.INVALID_MODEL","text":"INVALID_MODEL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_MODEL: The algorithm stopped because the model is invalid.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INVALID_OPTION","page":"JuMP","title":"INVALID_OPTION","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INVALID_OPTION","category":"page"},{"location":"api/JuMP/#JuMP.INVALID_OPTION","page":"JuMP","title":"JuMP.INVALID_OPTION","text":"INVALID_OPTION::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ITERATION_LIMIT","page":"JuMP","title":"ITERATION_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ITERATION_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.ITERATION_LIMIT","page":"JuMP","title":"JuMP.ITERATION_LIMIT","text":"ITERATION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#LOCALLY_INFEASIBLE","page":"JuMP","title":"LOCALLY_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LOCALLY_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.LOCALLY_INFEASIBLE","page":"JuMP","title":"JuMP.LOCALLY_INFEASIBLE","text":"LOCALLY_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#LOCALLY_SOLVED","page":"JuMP","title":"LOCALLY_SOLVED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LOCALLY_SOLVED","category":"page"},{"location":"api/JuMP/#JuMP.LOCALLY_SOLVED","page":"JuMP","title":"JuMP.LOCALLY_SOLVED","text":"LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MANUAL","page":"JuMP","title":"MANUAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MANUAL","category":"page"},{"location":"api/JuMP/#JuMP.MANUAL","page":"JuMP","title":"JuMP.MANUAL","text":"moi_backend field holds a CachingOptimizer in MANUAL mode.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MAX_SENSE","page":"JuMP","title":"MAX_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MAX_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.MAX_SENSE","page":"JuMP","title":"JuMP.MAX_SENSE","text":"MAX_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMAX_SENSE: the goal is to maximize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MEMORY_LIMIT","page":"JuMP","title":"MEMORY_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MEMORY_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.MEMORY_LIMIT","page":"JuMP","title":"JuMP.MEMORY_LIMIT","text":"MEMORY_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MIN_SENSE","page":"JuMP","title":"MIN_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MIN_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.MIN_SENSE","page":"JuMP","title":"JuMP.MIN_SENSE","text":"MIN_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMIN_SENSE: the goal is to minimize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_FEASIBLE_POINT","page":"JuMP","title":"NEARLY_FEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_FEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_FEASIBLE_POINT","page":"JuMP","title":"JuMP.NEARLY_FEASIBLE_POINT","text":"NEARLY_FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"NEARLY_INFEASIBILITY_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","text":"NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_REDUCTION_CERTIFICATE","page":"JuMP","title":"NEARLY_REDUCTION_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_REDUCTION_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_REDUCTION_CERTIFICATE","page":"JuMP","title":"JuMP.NEARLY_REDUCTION_CERTIFICATE","text":"NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NODE_LIMIT","page":"JuMP","title":"NODE_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NODE_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.NODE_LIMIT","page":"JuMP","title":"JuMP.NODE_LIMIT","text":"NODE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NORM_LIMIT","page":"JuMP","title":"NORM_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NORM_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.NORM_LIMIT","page":"JuMP","title":"JuMP.NORM_LIMIT","text":"NORM_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NO_SOLUTION","page":"JuMP","title":"NO_SOLUTION","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NO_SOLUTION","category":"page"},{"location":"api/JuMP/#JuMP.NO_SOLUTION","page":"JuMP","title":"JuMP.NO_SOLUTION","text":"NO_SOLUTION::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNO_SOLUTION: the result vector is empty.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NUMERICAL_ERROR","page":"JuMP","title":"NUMERICAL_ERROR","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NUMERICAL_ERROR","category":"page"},{"location":"api/JuMP/#JuMP.NUMERICAL_ERROR","page":"JuMP","title":"JuMP.NUMERICAL_ERROR","text":"NUMERICAL_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OBJECTIVE_LIMIT","page":"JuMP","title":"OBJECTIVE_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OBJECTIVE_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.OBJECTIVE_LIMIT","page":"JuMP","title":"JuMP.OBJECTIVE_LIMIT","text":"OBJECTIVE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OPTIMAL","page":"JuMP","title":"OPTIMAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OPTIMAL","category":"page"},{"location":"api/JuMP/#JuMP.OPTIMAL","page":"JuMP","title":"JuMP.OPTIMAL","text":"OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMAL: The algorithm found a globally optimal solution.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OPTIMIZE_NOT_CALLED","page":"JuMP","title":"OPTIMIZE_NOT_CALLED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OPTIMIZE_NOT_CALLED","category":"page"},{"location":"api/JuMP/#JuMP.OPTIMIZE_NOT_CALLED","page":"JuMP","title":"JuMP.OPTIMIZE_NOT_CALLED","text":"OPTIMIZE_NOT_CALLED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_ERROR","page":"JuMP","title":"OTHER_ERROR","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_ERROR","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_ERROR","page":"JuMP","title":"JuMP.OTHER_ERROR","text":"OTHER_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_LIMIT","page":"JuMP","title":"OTHER_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_LIMIT","page":"JuMP","title":"JuMP.OTHER_LIMIT","text":"OTHER_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_RESULT_STATUS","page":"JuMP","title":"OTHER_RESULT_STATUS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_RESULT_STATUS","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_RESULT_STATUS","page":"JuMP","title":"JuMP.OTHER_RESULT_STATUS","text":"OTHER_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#REDUCTION_CERTIFICATE","page":"JuMP","title":"REDUCTION_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.REDUCTION_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.REDUCTION_CERTIFICATE","page":"JuMP","title":"JuMP.REDUCTION_CERTIFICATE","text":"REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#SLOW_PROGRESS","page":"JuMP","title":"SLOW_PROGRESS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SLOW_PROGRESS","category":"page"},{"location":"api/JuMP/#JuMP.SLOW_PROGRESS","page":"JuMP","title":"JuMP.SLOW_PROGRESS","text":"SLOW_PROGRESS::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#SOLUTION_LIMIT","page":"JuMP","title":"SOLUTION_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOLUTION_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.SOLUTION_LIMIT","page":"JuMP","title":"JuMP.SOLUTION_LIMIT","text":"SOLUTION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#TIME_LIMIT","page":"JuMP","title":"TIME_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.TIME_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.TIME_LIMIT","page":"JuMP","title":"JuMP.TIME_LIMIT","text":"TIME_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#UNKNOWN_RESULT_STATUS","page":"JuMP","title":"UNKNOWN_RESULT_STATUS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UNKNOWN_RESULT_STATUS","category":"page"},{"location":"api/JuMP/#JuMP.UNKNOWN_RESULT_STATUS","page":"JuMP","title":"JuMP.UNKNOWN_RESULT_STATUS","text":"UNKNOWN_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_and","page":"JuMP","title":"op_and","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_and","category":"page"},{"location":"api/JuMP/#JuMP.op_and","page":"JuMP","title":"JuMP.op_and","text":"op_and(x, y)\n\nA function that falls back to x & y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_and(true, false)\nfalse\n\njulia> op_and(true, x)\ntrue && x\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_equal_to","page":"JuMP","title":"op_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_equal_to","page":"JuMP","title":"JuMP.op_equal_to","text":"op_equal_to(x, y)\n\nA function that falls back to x == y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_equal_to(2, 2)\ntrue\n\njulia> op_equal_to(x, 2)\nx == 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_greater_than_or_equal_to","page":"JuMP","title":"op_greater_than_or_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_greater_than_or_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_greater_than_or_equal_to","page":"JuMP","title":"JuMP.op_greater_than_or_equal_to","text":"op_greater_than_or_equal_to(x, y)\n\nA function that falls back to x >= y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_greater_than_or_equal_to(2, 2)\ntrue\n\njulia> op_greater_than_or_equal_to(x, 2)\nx >= 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_less_than_or_equal_to","page":"JuMP","title":"op_less_than_or_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_less_than_or_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_less_than_or_equal_to","page":"JuMP","title":"JuMP.op_less_than_or_equal_to","text":"op_less_than_or_equal_to(x, y)\n\nA function that falls back to x <= y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_less_than_or_equal_to(2, 2)\ntrue\n\njulia> op_less_than_or_equal_to(x, 2)\nx <= 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_or","page":"JuMP","title":"op_or","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_or","category":"page"},{"location":"api/JuMP/#JuMP.op_or","page":"JuMP","title":"JuMP.op_or","text":"op_or(x, y)\n\nA function that falls back to x | y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_or(true, false)\ntrue\n\njulia> op_or(true, x)\ntrue || x\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_strictly_greater_than","page":"JuMP","title":"op_strictly_greater_than","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_strictly_greater_than","category":"page"},{"location":"api/JuMP/#JuMP.op_strictly_greater_than","page":"JuMP","title":"JuMP.op_strictly_greater_than","text":"op_strictly_greater_than(x, y)\n\nA function that falls back to x > y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_strictly_greater_than(1, 2)\nfalse\n\njulia> op_strictly_greater_than(x, 2)\nx > 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_strictly_less_than","page":"JuMP","title":"op_strictly_less_than","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_strictly_less_than","category":"page"},{"location":"api/JuMP/#JuMP.op_strictly_less_than","page":"JuMP","title":"JuMP.op_strictly_less_than","text":"op_strictly_less_than(x, y)\n\nA function that falls back to x < y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_strictly_less_than(1, 2)\ntrue\n\njulia> op_strictly_less_than(x, 2)\nx < 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#Base.empty!(::GenericModel)","page":"JuMP","title":"Base.empty!(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.empty!(::GenericModel)","category":"page"},{"location":"api/JuMP/#Base.empty!-Tuple{GenericModel}","page":"JuMP","title":"Base.empty!","text":"empty!(model::GenericModel)::GenericModel\n\nEmpty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes. Always return the argument.\n\nNote: removes extensions data.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.isempty(::GenericModel)","page":"JuMP","title":"Base.isempty(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.isempty(::GenericModel)","category":"page"},{"location":"api/JuMP/#Base.isempty-Tuple{GenericModel}","page":"JuMP","title":"Base.isempty","text":"isempty(model::GenericModel)\n\nVerifies whether the model is empty, that is, whether the MOI backend is empty and whether the model is in the same state as at its creation apart from optimizer attributes.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.copy(::AbstractModel)","page":"JuMP","title":"Base.copy(::AbstractModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.copy(::AbstractModel)","category":"page"},{"location":"api/JuMP/#Base.copy-Tuple{AbstractModel}","page":"JuMP","title":"Base.copy","text":"copy(model::AbstractModel)\n\nReturn a copy of the model model. It is similar to copy_model except that it does not return the mapping between the references of model and its copy.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and a constraint cref. It is then copied into a model new_model with the new references assigned to x_new and cref_new.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, cref, x == 2)\ncref : x = 2\n\njulia> new_model = copy(model);\n\njulia> x_new = model[:x]\nx\n\njulia> cref_new = model[:cref]\ncref : x = 2\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)","page":"JuMP","title":"Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)","category":"page"},{"location":"api/JuMP/#Base.write-Tuple{IO, GenericModel}","page":"JuMP","title":"Base.write","text":"Base.write(\n    io::IO,\n    model::GenericModel;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_MOF,\n    kwargs...,\n)\n\nWrite the JuMP model model to io in the format format.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.reset_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.reset_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.reset_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.reset_optimizer","text":"MOIU.reset_optimizer(model::GenericModel)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.drop_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.drop_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.drop_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.drop_optimizer","text":"MOIU.drop_optimizer(model::GenericModel)\n\nCall MOIU.drop_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.attach_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.attach_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.attach_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.attach_optimizer","text":"MOIU.attach_optimizer(model::GenericModel)\n\nCall MOIU.attach_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"EditURL = \"constraint_programming.jl\"","category":"page"},{"location":"tutorials/linear/constraint_programming/#Constraint-programming","page":"Constraint programming","title":"Constraint programming","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"JuMP supports a range of constraint-programming type constraints via the corresponding sets in MathOptInterface. For most constraints, there are reformulations built-in that convert the constraint programming constraint into a mixed-integer programming equivalent.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Because of this reformulation, all variables must be integer, and they must typically have finite bounds. An error will be thrown if the reformulation requires finiteness and you have a variable with non-finite bounds.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/constraint_programming/#AllDifferent","page":"Constraint programming","title":"AllDifferent","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.AllDifferent set ensures that every element in a list takes a different integer value.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 1 <= x[1:4] <= 4, Int)\n@constraint(model, x in MOI.AllDifferent(4))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#BinPacking","page":"Constraint programming","title":"BinPacking","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.BinPacking set can be used to divide up a set of items into different groups, such that the sum of their weights does not exceed the capacity of a bin.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"weights, capacity = Float64[1, 1, 2, 2, 3], 3.0;\nnumber_of_bins = 3\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 1 <= x[1:length(weights)] <= number_of_bins, Int)\n@constraint(model, x in MOI.BinPacking(capacity, weights))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Here, the value of x[i] is the bin that item i was placed into.","category":"page"},{"location":"tutorials/linear/constraint_programming/#Circuit","page":"Constraint programming","title":"Circuit","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.Circuit set is used to construct a tour of a list of N variables. They will each be assigned an integer from 1 to N, that describes the successor to each variable in the list:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:4], Int)\n@constraint(model, x in MOI.Circuit(4))\noptimize!(model)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Let's see what tour was found, starting at node number 1:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"y = round.(Int, value.(x))\ntour = Int[1]\nwhile length(tour) < length(y)\n    push!(tour, y[tour[end]])\nend\ntour","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountAtLeast","page":"Constraint programming","title":"CountAtLeast","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountAtLeast set is used to ensure that at least n elements in a set of variables belong to a set of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, here is a model with three variables, constrained between 0 and 5:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[1:3] <= 5, Int)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"If we want to ensure that at least one element of each set {x[1], x[2]} and {x[2], x[3]} is in the set {3}, then we create a list of variables by concatenating the sets together:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"variables = [x[1], x[2], x[2], x[3]]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Then we need a partition list that contains the number of elements in each set of variables:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"partitions = [2, 2]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Finally, we need a set of values that the elements must be a part of:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"values = Set([3])","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"And the number of elements that must be part of the set values:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"n = 1","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The constraint is:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"@constraint(model, variables in MOI.CountAtLeast(n, partitions, values))","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"To ensure the uniqueness of the solution, we'll add a constraint that x[2] must be <= 2. This ensures that the only feasible solution is for x[1] and x[3] to be 3:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"@constraint(model, x[2] <= 2)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Let's check that we found a valid solution:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"optimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountBelongs","page":"Constraint programming","title":"CountBelongs","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountBelongs set is used to count how many elements in a set of variables belong to a set of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, to count how many elements in a set of 4 variables belong to the set {2, 3}, do:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@objective(model, Max, sum(x))\nset = Set([2, 3])\n@constraint(model, [n; x] in MOI.CountBelongs(1 + length(x), set))\noptimize!(model)\nvalue(n), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountDistinct","page":"Constraint programming","title":"CountDistinct","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountDistinct set is used to count the number of distinct elements in a set of variables.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@objective(model, Max, sum(x))\n@constraint(model, [n; x] in MOI.CountDistinct(1 + length(x)))\noptimize!(model)\nvalue(n), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountGreaterThan","page":"Constraint programming","title":"CountGreaterThan","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountGreaterThan set is used to strictly upper-bound the number of distinct elements in a set of variables that have a value equal to another variable.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, to count the number n of times that y appears in the vector x, use:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@variable(model, 3 <= y <= 4, Int)\n@objective(model, Max, sum(x))\n@constraint(model, [n; y; x] in MOI.CountGreaterThan(1 + 1 + length(x)))\noptimize!(model)\nvalue(n), value(y), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Here n is strictly greater than the count, and there is no limit on how large n could be. For example, n = 100 is also a feasible solution. The only constraint is that n cannot be equal to or smaller than the number of times that y appears.","category":"page"},{"location":"tutorials/linear/constraint_programming/#Table","page":"Constraint programming","title":"Table","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.Table set is used to select a single row from a matrix of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, given a matrix:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"table = Float64[1 1 0; 0 1 1; 1 0 1; 1 1 1]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"we can constraint a 3-element vector x to equal one of the rows in table via:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:3], Int)\n@constraint(model, x in MOI.Table(table))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"EditURL = \"https://github.com/oxfordcontrol/COSMO.jl/blob/v0.8.8/README.md\"","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<div style=\"display: flex; justify-content: flex-end\" margin=0px>\n    <img src=\"https://github.com/migarstka/COSMO_assets/blob/master/star_badge_3.png?raw=true\" align=\"right\" width=6%>\n</div>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<h1 align=\"center\" margin=0px>\n  <img src=\"https://github.com/migarstka/COSMO_assets/blob/master/cosmo_rocket_with_convex_set.png?raw=true\" width=70%>\n</h1>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n   <a href=\"https://github.com/oxfordcontrol/COSMO.jl/actions\"><img src=\"https://github.com/oxfordcontrol/COSMO.jl/workflows/ci/badge.svg?branch=master\"></a>\n  <a href=\"https://codecov.io/gh/oxfordcontrol/COSMO.jl\"><img src=\"https://codecov.io/gh/oxfordcontrol/COSMO.jl/branch/master/graph/badge.svg?raw=true\"></a>\n  <a href=\"https://oxfordcontrol.github.io/COSMO.jl/stable\"><img src=\"https://img.shields.io/badge/Documentation-stable-purple.svg?raw=true\"></a>\n  <a href=\"https://opensource.org/licenses/Apache-2.0\"><img src=\"https://img.shields.io/badge/License-Apache%202.0-blue.svg?raw=true\"></a>\n  <a href=\"https://github.com/oxfordcontrol/COSMO.jl/releases\"><img src=\"https://img.shields.io/badge/Release-v0.8.8-blue.svg?raw=true\"></a>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n  <a href=\"#features\">Features</a> •\n  <a href=\"#installation\">Installation</a> •\n  <a href=\"NEWS.md\">News</a> •\n  <a href=\"#citing-\">Citing</a> •\n  <a href=\"#contributing\">Contributing</a>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"This is a Julia implementation of the Conic operator splitting method (COSMO) solver. It can solve large convex conic optimization problems of the following form:","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n<img src=\"https://github.com/migarstka/COSMO_assets/blob/master/cosmo_format.png?raw=true\" width=220px>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"with decision variables x ϵ R^n, s ϵ R^m and data matrices P=P'>=0, q ϵ R^n, A ϵ R^(m×n), and b ϵ R^m. The convex set K is a composition of convex sets and cones.","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"For more information take a look at the COSMO.jl Documentation (stable |  dev).","category":"page"},{"location":"packages/COSMO/#Features","page":"oxfordcontrol/COSMO.jl","title":"Features","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"Versatile: COSMO solves linear programs, quadratic programs, second-order cone programs, semidefinite programs and problems involving exponential and power cones\nQuad SDPs: Positive semidefinite programs with quadratic objective functions are natively supported\nSafeguarded acceleration: robust and faster convergence to higher precision using COSMOAccelerators\nInfeasibility detection: Infeasible problems are detected without a homogeneous self-dual embedding of the problem\nJuMP / Convex.jl support: We provide an interface to MathOptInterface (MOI), which allows you to describe your problem in JuMP and Convex.jl.\nWarm starting: COSMO supports warm starting of the decision variables\nCustom sets and linear solver: Customize COSMO's components by defining your own convex constraint sets and by choosing from a number of direct and indirect linear system solvers, for example, QDLDL, Pardiso, Conjugate Gradient and MINRES\nArbitrary precision types: You can solve problems with any floating point precision.\nOpen Source: Our code is free to use and distributed under the Apache 2.0 Licence\nChordal decomposition: COSMO tries to decompose large structured PSD constraints using chordal decomposition techniques. This often results in a significant speedup compared to the original problem.\nSmart clique merging: After an initial decomposition of a structured SDP, COSMO recombines overlapping cliques/blocks to speed up the algorithm.","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<div align=\"center\" margin=0px>\n  <img src=\"https://github.com/oxfordcontrol/COSMO.jl/blob/master/docs/src/assets/example_decomposition.gif?raw=true\" width=45%>\n</div>","category":"page"},{"location":"packages/COSMO/#Installation","page":"oxfordcontrol/COSMO.jl","title":"Installation","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"COSMO can be added via the Julia package manager (type ]): pkg> add COSMO","category":"page"},{"location":"packages/COSMO/#Citing","page":"oxfordcontrol/COSMO.jl","title":"Citing","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"If you find COSMO useful in your project, we kindly request that you cite the following paper:","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"@Article{Garstka_2021,\n  author  = {Michael Garstka and Mark Cannon and Paul Goulart},\n  journal = {Journal of Optimization Theory and Applications},\n  title   = {{COSMO}: A Conic Operator Splitting Method for Convex Conic Problems},\n  volume  = {190},\n  number  = {3},\n  pages   = {779--810},\n  year    = {2021},\n  publisher = {Springer},\n  doi     = {10.1007/s10957-021-01896-x},\n  url     = {https://doi.org/10.1007/s10957-021-01896-x}\n}","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"The article is available under Open Access here.","category":"page"},{"location":"packages/COSMO/#Contributing","page":"oxfordcontrol/COSMO.jl","title":"Contributing","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"Contributions are always welcome. Our style guide can be found here.\nCurrent issues, tasks and future ideas are listed in Issues. Please report any issues or bugs that you encounter.\nAs an open source project we are also interested in any projects and applications that use COSMO. Please let us know by opening a GitHub issue.","category":"page"},{"location":"packages/COSMO/#Python-Interface","page":"oxfordcontrol/COSMO.jl","title":"Python - Interface","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"COSMO can also be called from Python. Take a look at: cosmo-python","category":"page"},{"location":"packages/COSMO/#Licence","page":"oxfordcontrol/COSMO.jl","title":"Licence 🔍","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"This project is licensed under the Apache License - see the LICENSE.md file for details.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"EditURL = \"https://github.com/chriscoey/Hypatia.jl/blob/v0.7.3/README.md\"","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"<img src=\"https://github.com/chriscoey/Hypatia.jl/wiki/hypatia_logo.png?raw=true\" alt=\"Hypatia logo\" width=\"358\"/>","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"(Image: Build Status) (Image: codecov) (Image: ) (Image: )","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is a highly customizable open source interior point solver for generic conic optimization problems, written in Julia.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"For more information on Hypatia, please see:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"documentation for Hypatia's conic form, predefined cones, and interfaces\ncones reference for cone definitions and oracles\nexamples folder for applied examples and instances\nbenchmarks folder for scripts used to run and analyze various computational benchmarks","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"and preprints of our papers:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Solving natural conic formulations with Hypatia.jl for computational arguments for expanding the class of cones recognized by conic solvers\nPerformance enhancements for a generic conic interior point algorithm for a description of Hypatia's algorithm and our enhanced stepping procedures\nSum of squares generalizations for conic sets for barriers and computational techniques for our generalized polynomial sum of squares cones\nConic optimization with spectral functions on Euclidean Jordan algebras for barriers and computational techniques for many of our epigraph/hypograph cones","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"and corresponding raw results CSV files generated by our run scripts in the benchmarks folder.","category":"page"},{"location":"packages/Hypatia/#License","page":"chriscoey/Hypatia.jl","title":"License","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is licensed under the MIT License (see LICENSE).","category":"page"},{"location":"packages/Hypatia/#Installation","page":"chriscoey/Hypatia.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"To use Hypatia, install Julia, then at the Julia REPL, type:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"using Hypatia\nusing Pkg\nPkg.add(\"Hypatia\")","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is an experimental solver and a work in progress, and may not run with older releases of Julia. Default options/parameters are not well-tuned, so we encourage you to experiment with these.","category":"page"},{"location":"packages/Hypatia/#Usage","page":"chriscoey/Hypatia.jl","title":"Usage","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia can be accessed through a low-level native Julia interface or through open-source modeling tools such as JuMP and Convex.jl. The native interface is more expressive, allowing Hypatia to solve conic models expressed with generic real floating point types and structured matrices or linear operators, for example. However, it is typically sufficient and more convenient to use JuMP.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Using JuMP, we can model a simple D-optimal experiment design problem and call Hypatia:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"using LinearAlgebra\nusing JuMP\nusing Hypatia\n\nmodel = Model(() -> Hypatia.Optimizer(verbose = false))\n@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) == 5)\n@variable(model, hypo)\n@objective(model, Max, hypo)\nV = rand(2, 3)\nQ = V * diagm(x) * V'\naff = vcat(hypo, [Q[i, j] for i in 1:2 for j in 1:i]...)\n@constraint(model, aff in MOI.RootDetConeTriangle(2))\n\n# solve and query solution\noptimize!(model)\ntermination_status(model)\nobjective_value(model)\nvalue.(x)","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"See our D-optimal design example for more information and references.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Many more examples using the native interface or JuMP can be found in the examples folder.","category":"page"},{"location":"packages/Hypatia/#Contributing","page":"chriscoey/Hypatia.jl","title":"Contributing","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Comments, questions, suggestions, and improvements/extensions to the code or documentation are welcomed. Please reach out on Discourse, or submit an issue or contribute a PR on our GitHub. If contributing code, try to maintain consistent style and add docstrings or comments for clarity. New examples are welcomed and should be implemented similarly to the existing examples.","category":"page"},{"location":"packages/Hypatia/#Acknowledgements","page":"chriscoey/Hypatia.jl","title":"Acknowledgements","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"This work has been partially funded by the National Science Foundation under grant OAC-1835443 and the Office of Naval Research under grant N00014-18-1-2079.","category":"page"},{"location":"packages/Hypatia/#Citing-Hypatia","page":"chriscoey/Hypatia.jl","title":"Citing Hypatia","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"If you find Hypatia solver useful, please cite our solver paper:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"@article{coey2022solving,\n    title={Solving natural conic formulations with {H}ypatia.jl},\n    author={Chris Coey and Lea Kapelevich and Juan Pablo Vielma},\n    year={2022},\n    journal={INFORMS Journal on Computing},\n    publisher={INFORMS},\n    volume={34},\n    number={5},\n    pages={2686--2699},\n    doi={https://doi.org/10.1287/ijoc.2022.1202}\n}","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"If you find aspects of Hypatia's IPM implementation useful, please cite our algorithm paper:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"@article{coey2022performance,\n    title={Performance enhancements for a generic conic interior point algorithm},\n    author={Chris Coey and Lea Kapelevich and Juan Pablo Vielma},\n    year={2023},\n    journal={Mathematical Programming Computation},\n    publisher={Springer},\n    volume={15},\n    pages={53--101},\n    doi={https://doi.org/10.1007/s12532-022-00226-0}\n}","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"EditURL = \"arbitrary_precision.jl\"","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Arbitrary-precision-arithmetic","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The purpose of this tutorial is to explain how to use a solver which supports arithmetic using a number type other than Float64.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"using JuMP\nimport CDDLib\nimport Clarabel","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Higher-precision-arithmetic","page":"Arbitrary precision arithmetic","title":"Higher-precision arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"To create a model with a number type other than Float64, use GenericModel with an optimizer which supports the same number type:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"model = GenericModel{BigFloat}(Clarabel.Optimizer{BigFloat})","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding decision variables is the same as a normal JuMP model, except that values are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@variable(model, -1 <= x[1:2] <= sqrt(big\"2\"))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Note that each x is now a GenericVariableRef{BigFloat}, which means that the value of x in a solution will be a BigFloat.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The lower and upper bounds of the decision variables are also BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"lower_bound(x[1])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(lower_bound(x[1]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"upper_bound(x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(upper_bound(x[2]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding constraints is the same as a normal JuMP model, except that coefficients are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c, x[1] == big\"2\" * x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The function is a GenericAffExpr with BigFloat for the coefficient and variable types;","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"constraint = constraint_object(c)\ntypeof(constraint.func)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and the set is a MOI.EqualTo{BigFloat}:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(constraint.set)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding and objective is the same as a normal JuMP model, except that coefficients are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@objective(model, Min, 3x[1]^2 + 2x[2]^2 - x[1] - big\"4\" * x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(objective_function(model))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Here's the model we have built:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"print(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Let's solve and inspect the solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The value of each decision variable is a BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"as well as other solution attributes like the objective value:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and dual solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"dual(c)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This problem has an analytic solution of x = [3//7, 3//14]. Currently, our solution has an error of approximately 1e-9:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x) .- [3 // 7, 3 // 14]","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"But by reducing the tolerances, we can obtain a more accurate solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"set_attribute(model, \"tol_gap_abs\", 1e-32)\nset_attribute(model, \"tol_gap_rel\", 1e-32)\noptimize!(model)\nvalue.(x) .- [3 // 7, 3 // 14]","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Rational-arithmetic","page":"Arbitrary precision arithmetic","title":"Rational arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"In addition to higher-precision floating point number types like BigFloat, JuMP also supports solvers with exact rational arithmetic. One example is CDDLib.jl, which supports the Rational{BigInt} number type:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"model = GenericModel{Rational{BigInt}}(CDDLib.Optimizer{Rational{BigInt}})","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"As before, we can create variables using rational bounds:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@variable(model, 1 // 7 <= x[1:2] <= 2 // 3)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"lower_bound(x[1])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(lower_bound(x[1]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"As well as constraints:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c1, (2 // 1) * x[1] + x[2] <= 1)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c2, x[1] + 3x[2] <= 9 // 4)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and objective functions:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@objective(model, Max, sum(x))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Here's the model we have built:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"print(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Let's solve and inspect the solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The optimal values are given in exact rational arithmetic:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value(c2)","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/standard_form.md\"","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/standard_form/#Standard-form-problem","page":"Standard form problem","title":"Standard form problem","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"MathOptInterface represents optimization problems in the standard form:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"where:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"the functions f_0 f_1 ldots f_m are specified by AbstractFunction objects\nthe sets mathcalS_1 ldots mathcalS_m are specified by AbstractSet objects","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"tip: Tip\nFor more information on this standard form, read our paper.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"MOI defines some commonly used functions and sets, but the interface is extensible to other sets recognized by the solver.","category":"page"},{"location":"moi/manual/standard_form/#Functions","page":"Standard form problem","title":"Functions","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The function types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Function Description\nVariableIndex x_j, the projection onto a single coordinate defined by a variable index j.\nVectorOfVariables The projection onto multiple coordinates (that is, extracting a sub-vector).\nScalarAffineFunction a^T x + b, where a is a vector and b scalar.\nScalarNonlinearFunction f(x), where f is a nonlinear function.\nVectorAffineFunction A x + b, where A is a matrix and b is a vector.\nScalarQuadraticFunction frac12 x^T Q x + a^T x + b, where Q is a symmetric matrix, a is a vector, and b is a constant.\nVectorQuadraticFunction A vector of scalar-valued quadratic functions.\nVectorNonlinearFunction f(x), where f is a vector-valued nonlinear function.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Extensions for nonlinear programming are present but not yet well documented.","category":"page"},{"location":"moi/manual/standard_form/#One-dimensional-sets","page":"Standard form problem","title":"One-dimensional sets","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The one-dimensional set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nLessThan(u) (-infty u\nGreaterThan(l) l infty)\nEqualTo(v) v\nInterval(l, u) l u\nInteger() mathbbZ\nZeroOne()  0 1 \nSemicontinuous(l, u)  0 cup l u\nSemiinteger(l, u)  0 cup ll+1ldotsu-1u","category":"page"},{"location":"moi/manual/standard_form/#Vector-cones","page":"Standard form problem","title":"Vector cones","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The vector-valued set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nReals(d) mathbbR^d\nZeros(d) 0^d\nNonnegatives(d)  x in mathbbR^d  x ge 0 \nNonpositives(d)  x in mathbbR^d  x le 0 \nSecondOrderCone(d)  (tx) in mathbbR^d  t ge lVert x rVert_2 \nRotatedSecondOrderCone(d)  (tux) in mathbbR^d  2tu ge lVert x rVert_2^2 t ge 0u ge 0 \nExponentialCone()  (xyz) in mathbbR^3  y exp (xy) le z y  0 \nDualExponentialCone()  (uvw) in mathbbR^3  -u exp (vu) le exp(1) w u  0 \nGeometricMeanCone(d)  (tx) in mathbbR^1+n  x ge 0 t le sqrtnx_1 x_2 cdots x_n  where n is d - 1\nPowerCone(α)  (xyz) in mathbbR^3  x^alpha y^1-alpha ge z x ge 0y ge 0 \nDualPowerCone(α)  (uvw) in mathbbR^3  left(fracualpharight(^alphaleft(fracv1-alpharight)^1-alpha ge w uv ge 0 \nNormOneCone(d)  (tx) in mathbbR^d  t ge sum_i lvert x_i rvert \nNormInfinityCone(d)  (tx) in mathbbR^d  t ge max_i lvert x_i rvert \nRelativeEntropyCone(d)  (u v w) in mathbbR^d  u ge sum_i w_i log (fracw_iv_i) v_i ge 0 w_i ge 0 \nHyperRectangle(l, u) x in barmathbbR^d x_i in l_i u_i forall i=1ldotsd\nNormCone(p, d) ``{ (t,x) \\in \\mathbb{R}^{d} : t \\ge \\left(\\sum\\limits_i","category":"page"},{"location":"moi/manual/standard_form/#Matrix-cones","page":"Standard form problem","title":"Matrix cones","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The matrix-valued set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nRootDetConeTriangle(d)  (tX) in mathbbR^1+d(1+d)2  t le det(X)^1d X mbox is the upper triangle of a PSD matrix \nRootDetConeSquare(d)  (tX) in mathbbR^1+d^2  t le det(X)^1d X mbox is a PSD matrix \nPositiveSemidefiniteConeTriangle(d)  X in mathbbR^d(d+1)2  X mbox is the upper triangle of a PSD matrix \nPositiveSemidefiniteConeSquare(d)  X in mathbbR^d^2  X mbox is a PSD matrix \nLogDetConeTriangle(d)  (tuX) in mathbbR^2+d(1+d)2  t le ulog(det(Xu)) X mbox is the upper triangle of a PSD matrix u  0  \nLogDetConeSquare(d)  (tuX) in mathbbR^2+d^2  t le u log(det(Xu)) X mbox is a PSD matrix u  0 \nNormSpectralCone(r, c)  (t X) in mathbbR^1 + r times c  t ge sigma_1(X) X mbox is a  rtimes cmbox matrix \nNormNuclearCone(r, c)  (t X) in mathbbR^1 + r times c  t ge sum_i sigma_i(X) X mbox is a  rtimes cmbox matrix \nHermitianPositiveSemidefiniteConeTriangle(d) The cone of Hermitian positive semidefinite matrices, with\nside_dimension rows and columns. \nScaled(S) The set S scaled so that Utilities.set_dot corresponds to LinearAlgebra.dot","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Some of these cones can take two forms: XXXConeTriangle and XXXConeSquare.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"In XXXConeTriangle sets, the matrix is assumed to be symmetric, and the elements are provided by a vector, in which the entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row).","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"In XXXConeSquare sets, the entries of the matrix are given column by column (or equivalently, row by row), and the matrix is constrained to be symmetric. As an example, given a 2-by-2 matrix of variables X and a one-dimensional variable t, we can specify a root-det constraint as [t, X11, X12, X22] ∈ RootDetConeTriangle or [t, X11, X12, X21, X22] ∈ RootDetConeSquare.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"We provide both forms to enable flexibility for solvers who may natively support one or the other. Transformations between XXXConeTriangle and XXXConeSquare are handled by bridges, which removes the chance of conversion mistakes by users or solver developers.","category":"page"},{"location":"moi/manual/standard_form/#Multi-dimensional-sets-with-combinatorial-structure","page":"Standard form problem","title":"Multi-dimensional sets with combinatorial structure","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Other sets are vector-valued, with a particular combinatorial structure. Read their docstrings for more information on how to interpret them.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nSOS1 A Special Ordered Set (SOS) of Type I\nSOS2 A Special Ordered Set (SOS) of Type II\nIndicator A set to specify an indicator constraint\nComplements A set to specify a mixed complementarity constraint\nAllDifferent The all_different global constraint\nBinPacking The bin_packing global constraint\nCircuit The circuit global constraint\nCountAtLeast The at_least global constraint\nCountBelongs The nvalue global constraint\nCountDistinct The distinct global constraint\nCountGreaterThan The count_gt global constraint\nCumulative The cumulative global constraint\nPath The path global constraint\nTable The table global constraint","category":"page"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/variables.md\"","category":"page"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/variables/#Variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"moi/reference/variables/#Functions","page":"Variables","title":"Functions","text":"","category":"section"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"add_variable\nadd_variables\nadd_constrained_variable\nadd_constrained_variables\nsupports_add_constrained_variable\nsupports_add_constrained_variables\nis_valid(::ModelLike,::VariableIndex)\ndelete(::ModelLike, ::VariableIndex)\ndelete(::ModelLike, ::Vector{VariableIndex})","category":"page"},{"location":"moi/reference/variables/#MathOptInterface.add_variable","page":"Variables","title":"MathOptInterface.add_variable","text":"add_variable(model::ModelLike)::VariableIndex\n\nAdd a scalar variable to the model, returning a variable index.\n\nA AddVariableNotAllowed error is thrown if adding variables cannot be done in the current state of the model model.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_variables","page":"Variables","title":"MathOptInterface.add_variables","text":"add_variables(model::ModelLike, n::Int)::Vector{VariableIndex}\n\nAdd n scalar variables to the model, returning a vector of variable indices.\n\nA AddVariableNotAllowed error is thrown if adding variables cannot be done in the current state of the model model.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_constrained_variable","page":"Variables","title":"MathOptInterface.add_constrained_variable","text":"add_constrained_variable(\n    model::ModelLike,\n    set::AbstractScalarSet\n)::Tuple{MOI.VariableIndex,\n         MOI.ConstraintIndex{MOI.VariableIndex, typeof(set)}}\n\nAdd to model a scalar variable constrained to belong to set, returning the index of the variable created and the index of the constraint constraining the variable to belong to set.\n\nBy default, this function falls back to creating a free variable with add_variable and then constraining it to belong to set with add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_constrained_variables","page":"Variables","title":"MathOptInterface.add_constrained_variables","text":"add_constrained_variables(\n    model::ModelLike,\n    sets::AbstractVector{<:AbstractScalarSet}\n)::Tuple{\n    Vector{MOI.VariableIndex},\n    Vector{MOI.ConstraintIndex{MOI.VariableIndex,eltype(sets)}},\n}\n\nAdd to model scalar variables constrained to belong to sets, returning the indices of the variables created and the indices of the constraints constraining the variables to belong to each set in sets. That is, if it returns variables and constraints, constraints[i] is the index of the constraint constraining variable[i] to belong to sets[i].\n\nBy default, this function falls back to calling add_constrained_variable on each set.\n\n\n\n\n\nadd_constrained_variables(\n    model::ModelLike,\n    set::AbstractVectorSet,\n)::Tuple{\n    Vector{MOI.VariableIndex},\n    MOI.ConstraintIndex{MOI.VectorOfVariables,typeof(set)},\n}\n\nAdd to model a vector of variables constrained to belong to set, returning the indices of the variables created and the index of the constraint constraining the vector of variables to belong to set.\n\nBy default, this function falls back to creating free variables with add_variables and then constraining it to belong to set with add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.supports_add_constrained_variable","page":"Variables","title":"MathOptInterface.supports_add_constrained_variable","text":"supports_add_constrained_variable(\n    model::ModelLike,\n    S::Type{<:AbstractScalarSet}\n)::Bool\n\nReturn a Bool indicating whether model supports constraining a variable to belong to a set of type S either on creation of the variable with add_constrained_variable or after the variable is created with add_constraint.\n\nBy default, this function falls back to supports_add_constrained_variables(model, Reals) && supports_constraint(model, MOI.VariableIndex, S) which is the correct definition for most models.\n\nExample\n\nSuppose that a solver supports only two kind of variables: binary variables and continuous variables with a lower bound. If the solver decides not to support VariableIndex-in-Binary and VariableIndex-in-GreaterThan constraints, it only has to implement add_constrained_variable for these two sets which prevents the user to add both a binary constraint and a lower bound on the same variable. Moreover, if the user adds a VariableIndex-in-GreaterThan constraint, implementing this interface (i.e., supports_add_constrained_variables) enables the constraint to be transparently bridged into a supported constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.supports_add_constrained_variables","page":"Variables","title":"MathOptInterface.supports_add_constrained_variables","text":"supports_add_constrained_variables(\n    model::ModelLike,\n    S::Type{<:AbstractVectorSet}\n)::Bool\n\nReturn a Bool indicating whether model supports constraining a vector of variables to belong to a set of type S either on creation of the vector of variables with add_constrained_variables or after the variable is created with add_constraint.\n\nBy default, if S is Reals then this function returns true and otherwise, it falls back to supports_add_constrained_variables(model, Reals) && supports_constraint(model, MOI.VectorOfVariables, S) which is the correct definition for most models.\n\nExample\n\nIn the standard conic form (see Duality), the variables are grouped into several cones and the constraints are affine equality constraints. If Reals is not one of the cones supported by the solvers then it needs to implement supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false as free variables are not supported. The solvers should then implement supports_add_constrained_variables(::Optimizer, ::Type{<:SupportedCones}) = true where SupportedCones is the union of all cone types that are supported; it does not have to implement the method supports_constraint(::Type{VectorOfVariables}, Type{<:SupportedCones}) as it should return false and it's the default. This prevents the user to constrain the same variable in two different cones. When a VectorOfVariables-in-S is added, the variables of the vector have already been created so they already belong to given cones. If bridges are enabled, the constraint will therefore be bridged by adding slack variables in S and equality constraints ensuring that the slack variables are equal to the corresponding variables of the given constraint function.\n\nNote that there may also be sets for which !supports_add_constrained_variables(model, S) and supports_constraint(model, MOI.VectorOfVariables, S). For instance, suppose a solver supports positive semidefinite variable constraints and two types of variables: binary variables and nonnegative variables. Then the solver should support adding VectorOfVariables-in-PositiveSemidefiniteConeTriangle constraints, but it should not support creating variables constrained to belong to the PositiveSemidefiniteConeTriangle because the variables in PositiveSemidefiniteConeTriangle should first be created as either binary or non-negative.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}","page":"Variables","title":"MathOptInterface.is_valid","text":"is_valid(model::ModelLike, index::Index)::Bool\n\nReturn a Bool indicating whether this index refers to a valid object in the model model.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}","page":"Variables","title":"MathOptInterface.delete","text":"delete(model::ModelLike, index::Index)\n\nDelete the referenced object from the model. Throw DeleteNotAllowed if if index cannot be deleted.\n\nThe following modifications also take effect if Index is VariableIndex:\n\nIf index used in the objective function, it is removed from the function, i.e., it is substituted for zero.\nFor each func-in-set constraint of the model:\nIf func isa VariableIndex and func == index then the constraint is deleted.\nIf func isa VectorOfVariables and index in func.variables then\nif length(func.variables) == 1 is one, the constraint is deleted;\nif length(func.variables) > 1 and supports_dimension_update(set) then then the variable is removed from func and set is replaced by update_dimension(set, MOI.dimension(set) - 1).\nOtherwise, a DeleteNotAllowed error is thrown.\nOtherwise, the variable is removed from func, i.e., it is substituted for zero.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}","page":"Variables","title":"MathOptInterface.delete","text":"delete(model::ModelLike, indices::Vector{R<:Index}) where {R}\n\nDelete the referenced objects in the vector indices from the model. It may be assumed that R is a concrete type. The default fallback sequentially deletes the individual items in indices, although specialized implementations may be more efficient.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#Attributes","page":"Variables","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"AbstractVariableAttribute\nVariableName\nVariablePrimalStart\nVariablePrimal\nVariableBasisStatus","category":"page"},{"location":"moi/reference/variables/#MathOptInterface.AbstractVariableAttribute","page":"Variables","title":"MathOptInterface.AbstractVariableAttribute","text":"AbstractVariableAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of variables in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariableName","page":"Variables","title":"MathOptInterface.VariableName","text":"VariableName()\n\nA variable attribute for a string identifying the variable. It is valid for two variables to have the same name; however, variables with duplicate names cannot be looked up using get. It has a default value of \"\" if not set`.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariablePrimalStart","page":"Variables","title":"MathOptInterface.VariablePrimalStart","text":"VariablePrimalStart()\n\nA variable attribute for the initial assignment to some primal variable's value that the optimizer may use to warm-start the solve. May be a number or nothing (unset).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariablePrimal","page":"Variables","title":"MathOptInterface.VariablePrimal","text":"VariablePrimal(result_index::Int = 1)\n\nA variable attribute for the assignment to some primal variable's value in result result_index. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a primal value for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the VariablePrimal attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariableBasisStatus","page":"Variables","title":"MathOptInterface.VariableBasisStatus","text":"VariableBasisStatus(result_index::Int = 1)\n\nA variable attribute for the BasisStatusCode of a variable in result result_index, with respect to an available optimal solution basis.\n\nIf the solver does not have a basis statue for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the VariableBasisStatus attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"EditURL = \"https://github.com/jump-dev/Ipopt.jl/blob/v1.5.0/README.md\"","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"(Image: )","category":"page"},{"location":"packages/Ipopt/#Ipopt.jl","page":"jump-dev/Ipopt.jl","title":"Ipopt.jl","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl is a wrapper for the Ipopt solver.","category":"page"},{"location":"packages/Ipopt/#Affiliation","page":"jump-dev/Ipopt.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Ipopt/#License","page":"jump-dev/Ipopt.jl","title":"License","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The underlying solver, coin-or/Ipopt, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Ipopt/#Installation","page":"jump-dev/Ipopt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Install Ipopt.jl using the Julia package manager:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import Pkg\nPkg.add(\"Ipopt\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"In addition to installing the Ipopt.jl package, this will also download and install the Ipopt binaries. You do not need to install Ipopt separately.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"For details on using a different linear solver, see the Linear Solvers section below. You do not need a custom binary to change the linear solver.","category":"page"},{"location":"packages/Ipopt/#Use-with-JuMP","page":"jump-dev/Ipopt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"You can use Ipopt with JuMP as follows:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"max_cpu_time\", 60.0)\nset_attribute(model, \"print_level\", 0)","category":"page"},{"location":"packages/Ipopt/#MathOptInterface-API","page":"jump-dev/Ipopt.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The Ipopt optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.BarrierIterations\nMOI.NLPBlock\nMOI.NLPBlockDualStart\nMOI.Name\nMOI.ObjectiveSense\nMOI.SolveTimeSec","category":"page"},{"location":"packages/Ipopt/#Options","page":"jump-dev/Ipopt.jl","title":"Options","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Supported options are listed in the Ipopt documentation.","category":"page"},{"location":"packages/Ipopt/#Solver-specific-callbacks","page":"jump-dev/Ipopt.jl","title":"Solver-specific callbacks","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt provides a callback that can be used to log the status of the optimization during a solve. It can also be used to terminate the optimization by returning false. Here is an example:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt, Test\nmodel = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x >= 1)\n@objective(model, Min, x + 0.5)\nx_vals = Float64[]\nfunction my_callback(\n   alg_mod::Cint,\n   iter_count::Cint,\n   obj_value::Float64,\n   inf_pr::Float64,\n   inf_du::Float64,\n   mu::Float64,\n   d_norm::Float64,\n   regularization_size::Float64,\n   alpha_du::Float64,\n   alpha_pr::Float64,\n   ls_trials::Cint,\n)\n   push!(x_vals, callback_value(model, x))\n   @test isapprox(obj_value, 1.0 * x_vals[end] + 0.5, atol = 1e-1)\n   # return `true` to keep going, or `false` to terminate the optimization.\n   return iter_count < 1\nend\nMOI.set(model, Ipopt.CallbackFunction(), my_callback)\noptimize!(model)\n@test MOI.get(model, MOI.TerminationStatus()) == MOI.INTERRUPTED\n@test length(x_vals) == 2","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"See the Ipopt documentation for an explanation of the arguments to the callback. They are identical to the output contained in the logging table printed to the screen.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To access the current solution and primal, dual, and complementarity violations of each iteration, use Ipopt.GetIpoptCurrentViolations and Ipopt.GetIpoptCurrentIterate. The two functions are identical to the ones in the Ipopt C interface.","category":"page"},{"location":"packages/Ipopt/#C-API","page":"jump-dev/Ipopt.jl","title":"C API","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl wraps the Ipopt C interface with minimal modifications.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"A complete example is available in the test/C_wrapper.jl file.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"For simplicity, the five callbacks required by Ipopt are slightly different to the C interface. They are as follows:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"\"\"\"\n   eval_f(x::Vector{Float64})::Float64\n\nReturns the objective value `f(x)`.\n\"\"\"\nfunction eval_f end\n\n\"\"\"\n   eval_grad_f(x::Vector{Float64}, grad_f::Vector{Float64})::Nothing\n\nFills `grad_f` in-place with the gradient of the objective function evaluated at\n`x`.\n\"\"\"\nfunction eval_grad_f end\n\n\"\"\"\n   eval_g(x::Vector{Float64}, g::Vector{Float64})::Nothing\n\nFills `g` in-place with the value of the constraints evaluated at `x`.\n\"\"\"\nfunction eval_g end\n\n\"\"\"\n   eval_jac_g(\n      x::Vector{Float64},\n      rows::Vector{Cint},\n      cols::Vector{Cint},\n      values::Union{Nothing,Vector{Float64}},\n   )::Nothing\n\nCompute the Jacobian matrix.\n\n* If `values === nothing`\n   - Fill `rows` and `cols` with the 1-indexed sparsity structure\n* Otherwise:\n   - Fill `values` with the elements of the Jacobian matrix according to the\n     sparsity structure.\n\n!!! warning\n    If `values === nothing`, `x` is an undefined object. Accessing any elements\n    in it will cause Julia to segfault.\n\"\"\"\nfunction eval_jac_g end\n\n\"\"\"\n   eval_h(\n      x::Vector{Float64},\n      rows::Vector{Cint},\n      cols::Vector{Cint},\n      obj_factor::Float64,\n      lambda::Float64,\n      values::Union{Nothing,Vector{Float64}},\n   )::Nothing\n\nCompute the Hessian-of-the-Lagrangian matrix.\n\n* If `values === nothing`\n   - Fill `rows` and `cols` with the 1-indexed sparsity structure\n* Otherwise:\n   - Fill `values` with the Hessian matrix according to the sparsity structure.\n\n!!! warning\n    If `values === nothing`, `x` is an undefined object. Accessing any elements\n    in it will cause Julia to segfault.\n\"\"\"\nfunction eval_h end","category":"page"},{"location":"packages/Ipopt/#INVALID_MODEL-error","page":"jump-dev/Ipopt.jl","title":"INVALID_MODEL error","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you get a termination status MOI.INVALID_MODEL, it is probably because you have some undefined value in your model, for example, a division by zero. Fix this by removing the division, or by imposing variable bounds so that you cut off the undefined region.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Instead of","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x)\n@NLobjective(model, 1 / x)","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"do","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x >= 0.0001)\n@NLobjective(model, 1 / x)","category":"page"},{"location":"packages/Ipopt/#Linear-Solvers","page":"jump-dev/Ipopt.jl","title":"Linear Solvers","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To improve performance, Ipopt supports a number of linear solvers.","category":"page"},{"location":"packages/Ipopt/#HSL","page":"jump-dev/Ipopt.jl","title":"HSL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Obtain a license and download HSL_jll.jl from https://licences.stfc.ac.uk/product/julia-hsl.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"There are two versions available: LBT and OpenBLAS. LBT is the recommended option for Julia ≥ v1.9.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Install this download into your current environment using:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import Pkg\nPkg.develop(path = \"/full/path/to/HSL_jll.jl\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Then, use a linear solver in HSL by setting the hsllib and linear_solver attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nimport HSL_jll\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"hsllib\", HSL_jll.libhsl_path)\nset_attribute(model, \"linear_solver\", \"ma86\")","category":"page"},{"location":"packages/Ipopt/#macOS-users","page":"jump-dev/Ipopt.jl","title":"macOS users","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Due to the security policy of macOS, Mac users may need to delete the quarantine attribute of the ZIP archive before extracting. For example:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"xattr -d com.apple.quarantine lbt_HSL_jll.jl-2023.5.26.zip\nxattr -d com.apple.quarantine openblas_HSL_jll.jl-2023.5.26.zip","category":"page"},{"location":"packages/Ipopt/#Pardiso","page":"jump-dev/Ipopt.jl","title":"Pardiso","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Download Pardiso from https://www.pardiso-project.org. Save the shared library somewhere, and record the filename.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Then, use Pardiso by setting the pardisolib and linear_solver attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"pardisolib\", \"/full/path/to/libpardiso\")\nset_attribute(model, \"linear_solver\", \"pardiso\")","category":"page"},{"location":"packages/Ipopt/#SPRAL","page":"jump-dev/Ipopt.jl","title":"SPRAL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you use Ipopt.jl with Julia ≥ v1.9, the linear solver SPRAL is available. You can use it by setting the linear_solver attribute:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"linear_solver\", \"spral\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Note that the following environment variables must be set before starting Julia:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"export OMP_CANCELLATION=TRUE\nexport OMP_PROC_BIND=TRUE","category":"page"},{"location":"packages/Ipopt/#BLAS-and-LAPACK","page":"jump-dev/Ipopt.jl","title":"BLAS and LAPACK","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"With Julia v1.9 or later, Ipopt and the linear solvers MUMPS (default), SPRAL, and HSL are compiled with libblastrampoline (LBT), a library that can change between BLAS and LAPACK backends at runtime.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The default BLAS and LAPACK backend is OpenBLAS.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Using LBT, we can also switch dynamically to other BLAS backends such as Intel MKL and Apple Accelerate. Because Ipopt and the linear solvers heavily rely on BLAS and LAPACK routines, using an optimized backend for a particular platform can improve the performance.","category":"page"},{"location":"packages/Ipopt/#MKL","page":"jump-dev/Ipopt.jl","title":"MKL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you have MKL.jl installed, switch to MKL by adding using MKL to your code:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using MKL  # Replace OpenBLAS by Intel MKL\nusing Ipopt","category":"page"},{"location":"packages/Ipopt/#AppleAccelerate","page":"jump-dev/Ipopt.jl","title":"AppleAccelerate","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you are using macOS ≥ v13.4 and you have AppleAccelerate.jl installed, add using AppleAccelerate to your code:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using AppleAccelerate  # Replace OpenBLAS by Apple Accelerate\nusing Ipopt","category":"page"},{"location":"packages/Ipopt/#Display-backends","page":"jump-dev/Ipopt.jl","title":"Display backends","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Check what backends are loaded using:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import LinearAlgebra\nLinearAlgebra.BLAS.lbt_get_config()","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/release_notes.md\"","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"CurrentModule = MathOptInterface","category":"page"},{"location":"moi/release_notes/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"moi/release_notes/#[v1.20.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.20.1)-(September-24,-2023)","page":"Release notes","title":"v1.20.1 (September 24, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Removed incorrect kwargs in some copy_to methods (#2272)\nFixed ConstraintDualStart for Bridges.Constraint.SplitIntervalBridge (#2275)\nFixed MethodError when trying to modify a variable objective (#2278)\nFixed stack-overflow in Utilities.operate(+,...) with many arguments (#2285) (#2290)","category":"page"},{"location":"moi/release_notes/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MathOptSetDistances to solver-tests.yml (#2265)\nUpdated Documenter (#2266)\nFixed various JET errors (#2267) (#2269) (#2270) (#2271) (#2276) (#2277) (#2289)\nVarious style improvements\nReplaced using Package with import Package where possible (#2274)\nRemoved Utilities.EMPTYSTRING (#2283)\nRemoved unnecessary const acronyms in Utilities (#2280) (#2281)\nRemoved invalid and unused method (#2286)\nRefactored src/Utilities/model.jl (#2287)","category":"page"},{"location":"moi/release_notes/#[v1.20.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.20.0)-(September-7,-2023)","page":"Release notes","title":"v1.20.0 (September 7, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Scaled set (#2237) (#2263) (#2264)\nAdded ObjectiveLimit attribute (#2257)","category":"page"},{"location":"moi/release_notes/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Updated dependencies (#2258)\nImproved performance of ScalarNonlinearFunction utilities (#2259)\nFixed docstrings (#2261)","category":"page"},{"location":"moi/release_notes/#[v1.19.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.19.0)-(August-15,-2023)","page":"Release notes","title":"v1.19.0 (August 15, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added VectorNonlinearFunction (#2201)\nAdded new bridges\nBridges.Constraint.IntegerToZeroOneBridge (#2205)\nBridges.Constraint.ToScalarQuadraticBridge (#2235)\nBridges.Constraint.ToVectorQuadraticBridge (#2235)\nBridges.Constraint.ToScalarNonlinearBridge (#2233) (#2235)\nBridges.Constraint.FunctionConversionBridge (#2235)\nAdded Bridges.bridging_cost for setting a bridge-specific cost in the bridging graph (#2235)\nAdded Utilities.eval_variables support for ScalarNonlinearFunction (#2218) (#2219)\nAdded support for ScalarNonlinearFunction in FileFormats.NL (#2228) (#2231)\nAdded support for writing non-Float64 functions in FileFormats.MOF\nAdded Utilities.lazy_map (#2254)","category":"page"},{"location":"moi/release_notes/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed method ambiguities in operate (#2224)\nFixed reading LP file with a -infinity <= x <= +infinity variable (#2225)\nFixed missing require in Test.test_nonlinear_duals (#2230)\nFixed bug in ConstraintPrimal of Bridges.Constraint.QuadtoSOCBridge (#2240)","category":"page"},{"location":"moi/release_notes/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added extensions to solver-tests.yml (#2229)\nRefactored test/Benchmarks (#2234)\nFixed warnings in tests (#2241) (#2243)\nSmall refactoring of bridges for upcoming VectorNonlinearFunction (#2244) (#2245)\nFixed various typos (#2251) (#2255)\nPartitioned how we run the tests on GitHub actions (#2252) (#2253)","category":"page"},{"location":"moi/release_notes/#[v1.18.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.18.0)-(June-23,-2023)","page":"Release notes","title":"v1.18.0 (June 23, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.SlackBridgePrimalDualStart (#2194)\nAdded constraint_gradient_structure and eval_constraint_gradient (#2200)","category":"page"},{"location":"moi/release_notes/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a missing @require in MOI.Test (#2195) (#2196)\nFixed incorrect usage of Utilities.operate! in bridges (#2207) (#2216)\nFixed splatting nonlinear expression with univariate operator (#2221)","category":"page"},{"location":"moi/release_notes/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Removed unused argument names (#2199)\nReduced memory requirement for tests (#2204)\nRefactored Utilities.promote_operation (#2206)\nImproved code style in Utilities/mutable_arithmetics.jl (#2209)\nRefactored various methods in Utilities/functions.jl (#2208) (#2212) (#2213) (#2214) (#2215)","category":"page"},{"location":"moi/release_notes/#[v1.17.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.17.1)-(June-6,-2023)","page":"Release notes","title":"v1.17.1 (June 6, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed comparison of ScalarFunctionConstantNotZero (#2190)","category":"page"},{"location":"moi/release_notes/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added documentation for enum instances (#2186)\nUpdated chatroom links in documentation (#2188)\nChanged the documentation to build on Julia v1.9 (#2191)","category":"page"},{"location":"moi/release_notes/#[v1.17.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.17.0)-(June-1,-2023)","page":"Release notes","title":"v1.17.0 (June 1, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added ScalarNonlinearFunction (#2059)\nAdded a variety of tests for NormSpectralCone, NormNuclearCone, and HermitianPositiveSemidefiniteConeTriangle (#2174)\nAdded Utilities.value_type for computing the value type of a function (#2176)","category":"page"},{"location":"moi/release_notes/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for external sets in Utilities.loadfromstring! (#2177)\nFixed promote_operation for ScalarNonlinearFunction (#2179)\nFixed two issues in FileFormats.LP when reading files with quadratic functions (#2182) (#2184)","category":"page"},{"location":"moi/release_notes/#[v1.16.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.16.0)-(May-16,-2023)","page":"Release notes","title":"v1.16.0 (May 16, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for MathOptFormat v1.3 and v1.4 (#2158) (#2169)\nAdded new method to the nonlinear API (#2162) (#2164)\neval_hessian_constraint\neval_hessian_objective\nhessian_constraint_structure\nhessian_objective_structure\nAdded new sets\nNormCone (#2119)\nScaledPositiveSemidefiniteConeTriangle (#2154)","category":"page"},{"location":"moi/release_notes/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for Julia v1.9 to work around a bug in the upstream Julia compiler (#2161) (#2163)\nFixed a correctness bug in Bridges.Constraint.HermitianToSymmetricPSDBridge (#2171)\nFixed convert(::VariableIndex, ::ScalarAffineFunction) when the function has terms with 0coefficients (#2173)","category":"page"},{"location":"moi/release_notes/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed solver-tests.yml (#2157)\nUpdated documentation links to developer chatroom (#2160)\nAdded various tests for bridges (#2156)\nAdded checklists to the developer documentation (#2167) (#2168)","category":"page"},{"location":"moi/release_notes/#[v1.15.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.15.1)-(April-25,-2023)","page":"Release notes","title":"v1.15.1 (April 25, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed deleting a variable in a bridged objective (#2150)","category":"page"},{"location":"moi/release_notes/#[v1.15.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.15.0)-(April-19,-2023)","page":"Release notes","title":"v1.15.0 (April 19, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.VectorFunctionizeBridge (#2139)","category":"page"},{"location":"moi/release_notes/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for Rational in Bridges.Constraint.SplitIntervalBridge (#2137)\nFixed printing of LaTeX models (#2141)\nFixed modify in Bridges.Objective.VectorSlackBridge (#2144)\nFixed NAME record with spaces in FileFormats.MPS (#2146)\nFixed deleting a variable in a bridged objective (#2147)","category":"page"},{"location":"moi/release_notes/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add a test for variables in one-sided open Interval sets (#2133)\nMinor style fixes in the source code (#2148)","category":"page"},{"location":"moi/release_notes/#[v1.14.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.14.1)-(April-6,-2023)","page":"Release notes","title":"v1.14.1 (April 6, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Bridges.print_active_bridges (#2135)","category":"page"},{"location":"moi/release_notes/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a warning when an ambiguous string is passed to exclude in Test.runtests (#2136)","category":"page"},{"location":"moi/release_notes/#[v1.14.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.14.0)-(April-4,-2023)","page":"Release notes","title":"v1.14.0 (April 4, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for starting values in Bridges.Constraint.QuadtoSOCBridge (#2115)\nAdded support for Regex in the include and exclude arguments to Test.runtests (#2129)\nAdded Bridges.print_active_bridges methods for individual objectives and constraints (#2128)","category":"page"},{"location":"moi/release_notes/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ResultCount when parsing .sol files in FileFormats.NL (#2130)","category":"page"},{"location":"moi/release_notes/#[v1.13.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.2)-(March-21,-2023)","page":"Release notes","title":"v1.13.2 (March 21, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed splatting of containers in MOI.Nonlinear (#2120)\nFixed a bug reading LP files with default bounds (#2121)\nFixed a bug in which Bridges.Constraint.HermitianToSymmetricPSDBridge was not enabled by default (#2123)","category":"page"},{"location":"moi/release_notes/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#2114)\nFunctions now print to the REPL in algebraic form. This is potentially breaking if you have tests which rely on a specific String form of MOI functions. (#2112) (#2126)","category":"page"},{"location":"moi/release_notes/#[v1.13.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.1)-(March-3,-2023)","page":"Release notes","title":"v1.13.1 (March 3, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added the Google style guide to the documentation linter Vale, and fixed the resulting warnings (#2110)\nImproved the docstrings in src/functions.jl (#2108)","category":"page"},{"location":"moi/release_notes/#[v1.13.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.0)-(February-28,-2023)","page":"Release notes","title":"v1.13.0 (February 28, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Constraint.NumberConversionBridge (#2091)\nAdded Parameter set (#2095) (#2105) (#2106) (#2109)\nAdded with_cache_type argument to instantiate (#2097)\nAdded support for HermitianPositiveSemidefiniteConeTriangle in Utilities.Model (#2100)","category":"page"},{"location":"moi/release_notes/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug when Utilities.@product_of_sets is empty (#2101)\nFixed Bridges.print_active_bridges when variable bridge is an AbstractScalarSet (#2107)","category":"page"},{"location":"moi/release_notes/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added tests for vector-valued objective functions in FileFormats.MOF (#2093)\nUsed and documented preference for import MathOptInterface as MOI (#2096)\nFix and test links in the documentation with linkcheck = true (#2098)\nImproved docstrings of sets in src/sets.jl (#2099)\nSkip checking flakey links in documentation with linkcheck_ignore  (#2103)","category":"page"},{"location":"moi/release_notes/#[v1.12.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.12.0)-(February-10,-2023)","page":"Release notes","title":"v1.12.0 (February 10, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for vector-valued objective functions (#2070)\nAdded a Utilities.distance_to_set method for SecondOrderCone (#2060)","category":"page"},{"location":"moi/release_notes/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a number of constraint bridges so that Bridges.final_touch can be called multiple times without forcing a rebuild of the reformulation (#2089)","category":"page"},{"location":"moi/release_notes/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new tests that a set ObjectiveFunction appears in ListOfModelAttributesSet (#2085)\nImproved the docstrings of a number of constraint-programming related sets (#2087)","category":"page"},{"location":"moi/release_notes/#[v1.11.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.5)-(January-24,-2023)","page":"Release notes","title":"v1.11.5 (January 24, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .lp files with an off-diagonal quadratic objective (#2082)","category":"page"},{"location":"moi/release_notes/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added SnoopPrecompile directives for reduced time-to-first-X in Julia v1.9 (#2080)","category":"page"},{"location":"moi/release_notes/#[v1.11.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.4)-(January-12,-2023)","page":"Release notes","title":"v1.11.4 (January 12, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .lp files with an Integer section (#2078)","category":"page"},{"location":"moi/release_notes/#[v1.11.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.3)-(January-12,-2023)","page":"Release notes","title":"v1.11.3 (January 12, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance bug when deleting a vector of constraints (#2072)\nFixed a bug reading .lp files with terms like x -1 y (#2076)","category":"page"},{"location":"moi/release_notes/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Documented the two-argument method of optimize! (#2074)","category":"page"},{"location":"moi/release_notes/#[v1.11.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.2)-(January-2,-2023)","page":"Release notes","title":"v1.11.2 (January 2, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with ConstraintName set for VariableIndex constraints (#2066)\nFixed a bug reading .mof.json files with nonlinear objectives and no constraints (#2068)","category":"page"},{"location":"moi/release_notes/#[v1.11.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.1)-(December-22,-2022)","page":"Release notes","title":"v1.11.1 (December 22, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with integer coefficients for affine and quadratic functions (#2063)","category":"page"},{"location":"moi/release_notes/#[v1.11.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.0)-(December-2,-2022)","page":"Release notes","title":"v1.11.0 (December 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Utilities.PenaltyRelaxation and Utilities.ScalarPenaltyRelaxation (#1995)\nAdded Utilities.distance_to_set (#2048)\nAdded support for ConstraintPrimalStart and ConstraintDualStart in FileFormats.MOF (#2056)","category":"page"},{"location":"moi/release_notes/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Tidied these release notes (#2055)","category":"page"},{"location":"moi/release_notes/#[v1.10.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.10.0)-(November-22,-2022)","page":"Release notes","title":"v1.10.0 (November 22, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new methods set(::OptimizerWithAttributes, ::RawOptimizerAttribute, value) and get(::OptimizerWithAttributes, ::RawOptimizerAttribute) (#2049)\nAdded new methods Utilities.DoubleDicts.outer_keys and Utilities.DoubleDicts.nonempty_outer_keys (#2052)","category":"page"},{"location":"moi/release_notes/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed Bridges.Objective.SlackBridge when the objective function is complex-valued (#2036) (#2038)\nFixed docstring of Test.runtests to clarify the warn_unsupported argument (#2037)\nFixed reading of free variables in FileFormats.LP (#2044)\nFixed numerous edge cases reading files from QPLIB using FileFormats.LP (#2042) (#2044)\nFixed situations in which x^y returns a complex value in Nonlinear (#2050)","category":"page"},{"location":"moi/release_notes/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved the error message thrown when a user-defined nonlinear function does not accept splatted input (#2032)\nRemoved specialized iterators for keys and values in Utilities.CleverDicts (#2051)","category":"page"},{"location":"moi/release_notes/#[v1.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.9.0)-(October-29,-2022)","page":"Release notes","title":"v1.9.0 (October 29, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added default fallback for getting ListOfConstraintIndices and NumberOfConstraints when the constraint type is unsupported by the model (#2021)\nAdded support for min and max in nonlinear expressions (#2023)\nAdded support for Indicator{EqualTo{T}} constraints in FileFormats.MPS (#2022)\nAdded default fallback for write_to_file and read_from_file (#2029)","category":"page"},{"location":"moi/release_notes/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed Constraint.ZeroOneBridge by adding new bounds as affine constraints instead of variable bounds (#1879)\nFixed reading free rows in FileFormats.MPS files (#2009)\nFixed parsing of OBJSENSE blocks in FileFormats.MPS files (#2016) (#2019)\nFixed the parsing of deeply nested nonlinear expressions by removing the use of recursion (#2020)\nFixed the requirements check in Test.test_constrainnt_get_ConstraintIndex (#2024)","category":"page"},{"location":"moi/release_notes/#[v1.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.2)-(September-20,-2022)","page":"Release notes","title":"v1.8.2 (September 20, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Documentation","page":"Release notes","title":"Documentation","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added vale as a documentation linter (#2002)\nImproved styling of code blocks in the PDF (#1999) (#2000)\nFixed a number of typos in the documentation (#2001) (#2003)","category":"page"},{"location":"moi/release_notes/#[v1.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.1)-(September-12,-2022)","page":"Release notes","title":"v1.8.1 (September 12, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in supports(::AbstractBridgeOptimizer for constraint attributes (#1991) (#1992)","category":"page"},{"location":"moi/release_notes/#[v1.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.0)-(September-1,-2022)","page":"Release notes","title":"v1.8.0 (September 1, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new sets\nHyperRectangle (#1961)\nReified (#1955)\nAdded new bridges (#1955)\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.SplitHyperRectangleBridge\nAdded support for atan(y, x) in Nonlinear (#1987)","category":"page"},{"location":"moi/release_notes/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Lazily construct expressions in Nonlinear so that expressions are updated when Nonlinear.Parameter values are updated (#1984)\nAllow NORM_LIMIT as a TerminationStatus for unbounded problems in Test (#1990)","category":"page"},{"location":"moi/release_notes/#[v1.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.7.0)-(August-16,-2022)","page":"Release notes","title":"v1.7.0 (August 16, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new sets\nHermitianPositiveSemidefiniteConeTriangle\nAdded new optimizer-independent options\nRelativeGapTolerance\nAbsoluteGapTolerance\nAdded new bridges\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Objective.QuadratizeBridge\nAdded support for generic number types in Utilities.loadfromstring!\nUpdated FileFormats.MOF to MathOptFormat v1.1, enabling support for constraint programming sets in the MOF file format\nAdded support in various FileFormats for\nindicator constraints in FileFormats.MPS\nquadratic constraints and an objective in FileFormats.LP\nquadratic constraints and an objective in FileFormats.MPS","category":"page"},{"location":"moi/release_notes/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed some missing promotion rules","category":"page"},{"location":"moi/release_notes/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved the performance of Jacobian products in Nonlinear\nRemoved an un-needed copy in Utilities.modify_function\nVarious clean-ups in Bridges/bridge_optimizer.jl","category":"page"},{"location":"moi/release_notes/#[v1.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.6.1)-(July-23,-2022)","page":"Release notes","title":"v1.6.1 (July 23, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for ExponentialCone in MatrixOfConstraints\nFix PSDSquare_3 test to reflect a previously fixed bug getting the ConstraintDual of a PositiveSemidefiniteConeSquare constraint","category":"page"},{"location":"moi/release_notes/#[v1.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.6.0)-(July-2,-2022)","page":"Release notes","title":"v1.6.0 (July 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.needs_final_touch and Bridges.final_touch\nAdded new bridges from constraint programming sets to mixed-integer linear programs:\nAllDifferentToCountDistinctBridge\nCountAtLeastToCountBelongsBridge\nCountBelongsToMILPBridge\nCountDistinctToMILPBridge\nCountGreaterThanToMILPBridge\nCircuitToMILPBridge","category":"page"},{"location":"moi/release_notes/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Relax an instance of ::Vector to ::AbstractVector in MOI.Nonlinear\nFix BinPackingToMILPBridge to respect variable bounds\nFix SemiToBinaryBridge to throw error if other bounds are set","category":"page"},{"location":"moi/release_notes/#[v1.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.5.0)-(June-27,-2022)","page":"Release notes","title":"v1.5.0 (June 27, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added GetAttributeNotAllowed for solvers to indicate when getting an attribute encounters an error\nAdded Utilities.get_fallback support for ObjectiveValue and DualObjectiveValue\nAdded new bridges:\nRootDetConeSquare to RootDetConeTriangle\nLogDetConeSquare to LogDetConeTriangle\nBinPacking to a mixed-integer linear program\nTable to a mixed-integer linear program\nAdded Bridges.print_active_bridges to display the current optimal hyper-path in a Bridges.LazyBridgeOptimizer","category":"page"},{"location":"moi/release_notes/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ZeroOne tests with lower and upper bounds\nFixed error in FileFormats.LP when reading a malformed file\nFixed reading of nonlinear programs in FileFormats.MOF\nFixed bug in ConstraintDual when using SquareBridge","category":"page"},{"location":"moi/release_notes/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved documentation of nonlinear API\nDocumented duality convention for PositiveSemidefiniteConeSquare sets\nFixed typo in Bridges.Constraint.QuadToSOCBridge docstring","category":"page"},{"location":"moi/release_notes/#[v1.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.4.0)-(June-9,-2022)","page":"Release notes","title":"v1.4.0 (June 9, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a number of sets for constraint programming:\nAllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nTable\nAdded support for user-defined hessians in Nonlinear\nAdded Bridges.runtests to simplify the testing of bridge implementations","category":"page"},{"location":"moi/release_notes/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in FileFormats.NL when writing univariate *","category":"page"},{"location":"moi/release_notes/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Began a large refactoring of the Bridges submodule, with greatly improved documentation.","category":"page"},{"location":"moi/release_notes/#[v1.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.3.0)-(May-27,-2022)","page":"Release notes","title":"v1.3.0 (May 27, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add MOI.Nonlinear submodule. This is a large new submodule that has been refactored from code that was in JuMP. For now, it should be considered experimental.\nAdd FileFormats.NL.SolFileResults(::IO, ::Model)\nAdd FileFormats.NL.read!(::IO, ::Model)\nAdd MOI.modify that accepts a vector of modifications","category":"page"},{"location":"moi/release_notes/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Test which attempted to include non-.jl files\nFixed a bug in FileFormats for models with open interval constraints","category":"page"},{"location":"moi/release_notes/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in Utilities.DoubleDict\nVarious minor improvements to the documentation","category":"page"},{"location":"moi/release_notes/#[v1.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.2.0)-(April-25,-2022)","page":"Release notes","title":"v1.2.0 (April 25, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add support for the FORMAT_REW/.rew file format in FileFormats.","category":"page"},{"location":"moi/release_notes/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug handling of default variable bounds in FileFormats.LP\nFix FileFormats.MPS to not write OBJSENSE by default since this is only supported by some readers.","category":"page"},{"location":"moi/release_notes/#[v1.1.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.2)-(March-31,-2022)","page":"Release notes","title":"v1.1.2 (March 31, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix a range of bugs in FileFormats.LP\nFix reading of problem dimensions in FileFormats.SDPA","category":"page"},{"location":"moi/release_notes/#[v1.1.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.1)-(March-23,-2022)","page":"Release notes","title":"v1.1.1 (March 23, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in test_model_UpperBoundAlreadySet\nFix bug in test_infeasible_ tests\nFix bug in test_objective_ObjectiveFunction_blank\nRelax restriction of MOI.AbstractOptimizer to MOI.ModelLike in Utilities.CachingOptimizer and instantiate.","category":"page"},{"location":"moi/release_notes/#New-tests","page":"Release notes","title":"New tests","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add test_conic_empty_matrix that checks conic solvers support problems with no variables.","category":"page"},{"location":"moi/release_notes/#[v1.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.0)-(March-2,-2022)","page":"Release notes","title":"v1.1.0 (March 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.throw_unsupported(::UniversalFallback) for simplifying solver wrappers which copy from a UniversalFallback.","category":"page"},{"location":"moi/release_notes/#[v1.0.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.2)-(March-1,-2022)","page":"Release notes","title":"v1.0.2 (March 1, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in the test_model_ScalarFunctionConstantNotZero test\nFixed the error type when an AbstractFunctionConversionBridge cannot get or set an attribute\nIdentified a correctness bug in RSOCtoPSDBridge. We now thrown an error instead of returning an incorrect result.","category":"page"},{"location":"moi/release_notes/#[v1.0.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.1)-(February-25,-2022)","page":"Release notes","title":"v1.0.1 (February 25, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in which OptimizerAttributes were not copied in CachingOptimizer\nFixed a bug in which shift_constant did not promote mixed types of coefficients\nFixed a bug in which deleting a constraint of a bridged variable threw ErrorException instead of MOI.DeleteNotAllowed\nFixed a bug in which add_constraint in MatrixOfConstraints did not canonicalize the function\nFixed a bug when modifying scalar constants of a function containing a bridged variable\nFixed a bug in which final_touch was not always called with a CachingOptimizer","category":"page"},{"location":"moi/release_notes/#[v1.0.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.0)-(February-17,-2022)","page":"Release notes","title":"v1.0.0 (February 17, 2022)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Although tagged as a breaking release, v1.0.0 is v0.10.9 with deprecations removed, similar to how Julia 1.0 was Julia 0.7 with deprecations removed.","category":"page"},{"location":"moi/release_notes/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nAll deprecations have been removed","category":"page"},{"location":"moi/release_notes/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated features. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Then install MathOptInterface v0.10.9:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add MathOptInterface@0.10\"","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"moi/release_notes/#[v0.10.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.9)-(February-16,-2022)","page":"Release notes","title":"v0.10.9 (February 16, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.FreeVariables as a new VariablesConstrainer for conic solvers\nAdded MOI.default_cache for specifying the model used in CachingOptimizer","category":"page"},{"location":"moi/release_notes/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed LaTeX printing of MOI.Interval sets","category":"page"},{"location":"moi/release_notes/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Aqua.jl as a CI check, and fixed suggested issues\nThe constructors of GeoMeanBridge, StructOfConstraints, and CachingOptimizer were changed from outer to inner constructors. This change is technically breaking, but does not impact users who followed the documented API.","category":"page"},{"location":"moi/release_notes/#[v0.10.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.8)-(February-3,-2022)","page":"Release notes","title":"v0.10.8 (February 3, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a Base.read! for FileFormats.LP.","category":"page"},{"location":"moi/release_notes/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in MutableSparseMatrix\nFixed a bug when calling operate!(vcat, ...) with Number arguments\nRemoved unintended export of deprecated symbols\nFixed a bug with PowerCone and DualPowerCone in MatrixOfConstraints.","category":"page"},{"location":"moi/release_notes/#[v0.10.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.7)-(January-5,-2022)","page":"Release notes","title":"v0.10.7 (January 5, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added test for modifying the constant vector in a VectorAffineFunction-in-Zeros constraint.","category":"page"},{"location":"moi/release_notes/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed the order in which sets are added to a LazyBridgeOptimizer. Compared to v0.10.6, this may result in bridged models being created with a different number (and order) of variables and constraints. However, it was necessary to fix cases which were previously rejected as unsupported, even though there was a valid bridge transformation.\nFixed an error message in FileFormats.CBF\nFixed comparison in test_linear_integration_Interval\nFixed errors for ConstraintPrimal in a CachingOptimizer\nFixed printing of models with non-Float64 coefficients.","category":"page"},{"location":"moi/release_notes/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements to reduce time-to-first-solve latency\nImproved error message when an optimizer does not support compute_conflict!","category":"page"},{"location":"moi/release_notes/#[v0.10.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.6)-(November-30,-2021)","page":"Release notes","title":"v0.10.6 (November 30, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new documentation and tests for infeasibility certificates\nAdded a version control system for the tests in MOI.Test.runtests. Pass exclude_tests_after = v\"0.10.5\" to run tests added in v0.10.5 and earlier.\nMOI.Test.runtests now supports generic number types. To specify the number type T, pass MOI.Test.Config(T).\nAdded infeasible_status to MOI.Test.Config for solvers which return LOCALLY_INFEASIBLE\nCachingOptimizers now use a fallback for ConstraintPrimal. This should enable solvers using a CachingOptimizer to pass tests requiring ConstraintPrimal.","category":"page"},{"location":"moi/release_notes/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a StackOverflow bug in copy_to\nFixed error thrown when nonconvex quadratic constraints cannot be bridged\nFixed a bug in copy_to for FileFormats.NL.Model\nFixed a bug in FileFormats.NL when printing large integers\nRemove a common test failure for LowerBoundAlreadySet tests\nUtilities.num_rows is now exported\nRemove parts of failing test_model_copy_to_xxx tests due to bridges","category":"page"},{"location":"moi/release_notes/#[v0.10.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.5)-(November-7,-2021)","page":"Release notes","title":"v0.10.5 (November 7, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed getter in UniversalFallback\nFixed test_solve_conflict_zeroone_ii","category":"page"},{"location":"moi/release_notes/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Make normalize_and_add_constraint more flexible\nUpdate paper BibTeX","category":"page"},{"location":"moi/release_notes/#[v0.10.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.4)-(October-26,-2021)","page":"Release notes","title":"v0.10.4 (October 26, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add SolverVersion attribute\nAdd new tests:\ntest_solve_conflict_zeroone_ii\ntest_nonlinear_objective\nUtilities.VariablesContainer now supports ConstraintFunction and ConstraintSet\nThe documentation is now available as a PDF","category":"page"},{"location":"moi/release_notes/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Update to MutableArithmetics 0.3\nVarious improvements to the documentation","category":"page"},{"location":"moi/release_notes/#[v0.10.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.3)-(September-18,-2021)","page":"Release notes","title":"v0.10.3 (September 18, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-39","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug which prevented callbacks from working through a CachingOptimizer\nFixed bug in Test submodule","category":"page"},{"location":"moi/release_notes/#[v0.10.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.2)-(September-16,-2021)","page":"Release notes","title":"v0.10.2 (September 16, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Updated MathOptFormat to v1.0\nUpdated JSONSchema to v1.0\nAdded Utilities.set_with_dimension\nAdded two-argument optimize!(::AbstractOptimizer, ::ModelLike)\nThe experimental feature copy_to_and_optimize! has been removed\nDet bridges now support getting ConstraintFunction and ConstraintSet\nVarious minor bug fixes identified by improved testing","category":"page"},{"location":"moi/release_notes/#[v0.10.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.1)-(September-8,-2021)","page":"Release notes","title":"v0.10.1 (September 8, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes to MOI.Test","category":"page"},{"location":"moi/release_notes/#[v0.10.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.0)-(September-6,-2021)","page":"Release notes","title":"v0.10.0 (September 6, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"MOI v0.10 is a significant breaking release. There are a large number of user-visible breaking changes and code refactors, as well as a substantial number of new features.","category":"page"},{"location":"moi/release_notes/#Breaking-in-MOI","page":"Release notes","title":"Breaking in MOI","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"SingleVariable has been removed; use VariableIndex instead\nSingleVariableConstraintNameError has been renamed to VariableIndexConstraintNameError\nSettingSingleVariableFunctionNotAllowed has been renamed to SettingVariableIndexFunctionNotAllowed\nVariableIndex constraints should not support ConstraintName\nVariableIndex constraints should not support ConstraintBasisStatus; implement VariableBasisStatus instead\nListOfConstraints has been renamed to ListOfConstraintTypesPresent\nListOfConstraintTypesPresent should now return Tuple{Type,Type} instead of Tuple{DataType,DataType}\nSolveTime has been renamed to SolveTimeSec\nIndicatorSet has been renamed to Indicator\nRawParameter has been renamed to RawOptimizerAttribute and now takes String instead of Any as the only argument\nThe .N field in result attributes has been renamed to .result_index\nThe .variable_index field in ScalarAffineTerm has been renamed to .variable\nThe .variable_index_1 field in ScalarQuadraticTerm has been renamed to .variable_1\nThe .variable_index_2 field in ScalarQuadraticTerm has been renamed to .variable_2\nThe order of affine_terms and quadratic_terms in ScalarQuadraticFunction and VectorQuadraticFunction have been reversed. Both functions now accept quadratic, affine, and constant terms in that order.\nThe index_value function has been removed. Use .value instead.\nisapprox has been removed for SOS1 and SOS2.\nThe dimension argument to Complements(dimension::Int) should now be the length of the corresponding function, instead of half the length. An ArgumentError is thrown if dimension is not even.\ncopy_to no longer takes keyword arguments:\ncopy_names: now copy names if they are supported by the destination solver\nfilter_constraints: use Utilities.ModelFilter instead\nwarn_attributes: never warn about optimizer attributes","category":"page"},{"location":"moi/release_notes/#Breaking-in-Bridges","page":"Release notes","title":"Breaking in Bridges","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Constraint.RSOCBridge has been renamed to Constraint.RSOCtoSOCBridge\nConstraint.SOCRBridge has been renamed to Constraint.SOCtoRSOCBridge\nBridges now return vectors that can be modified by the user. Previously, some bridges returned views instead of copies.\nBridges.IndexInVector has been unified into a single type. Previously, there was a different type for each submodule within Bridges\nThe signature of indicator bridges has been fixed. Use MOI.Bridges.Constraint.IndicatortoSOS1{Float64}(model).","category":"page"},{"location":"moi/release_notes/#Breaking-in-FileFormats","page":"Release notes","title":"Breaking in FileFormats","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"FileFormats.MOF.Model no longer accepts validate argument. Use the JSONSchema package to validate the MOF file. See the documentation for more information.","category":"page"},{"location":"moi/release_notes/#Breaking-in-Utilities","page":"Release notes","title":"Breaking in Utilities","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The datastructure of Utilities.Model (and models created with Utilities.@model) has been significantly refactored in a breaking way. This includes the way that objective functions and variable-related information is stored.\nUtilities.supports_default_copy has been renamed to supports_incremental_interface\nUtilities.automatic_copy_to has been renamed to Utilities.default_copy_to\nThe allocate-load API has been removed\nCachingOptimizers are now initialized as EMPTY_OPTIMIZER instead of ATTACHED_OPTIMIZER. If your code relies on the optimizer being attached, call MOIU.attach_optimizer(model) after creation.\nThe field names of Utilities.IndexMap have been renamed to var_map and con_map. Accessing these fields directly is considered a private detail that may change. Use the public getindex and setindex! API instead.\nThe size argument to Utilities.CleverDicts.CleverDict(::Integer) has been removed.\nThe size argument to Utilities.IndexMap(::Integer) has been removed.\nUtilities.DoubleDicts have been significantly refactored. Consult the source code for details.\nUtilities.test_models_equal has been moved to MOI.Test","category":"page"},{"location":"moi/release_notes/#Breaking-in-Test","page":"Release notes","title":"Breaking in Test","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"MOI.Test has been renamed to MOI.DeprecatedTest\nAn entirely new MOI.Test submodule has been written. See the documentation for details. The new MOI.Test submodule may find many bugs in the implementations of existing solvers that were previously untested.","category":"page"},{"location":"moi/release_notes/#Other-changes:","page":"Release notes","title":"Other changes:","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"attribute_value_type has been added\ncopy_to_and_optimize! has been added\nVariableBasisStatus has been added\nprint(model) now prints a human-readable description of the model\nVarious improvements to the FileFormats submodule\nFileFormats.CBF was refactored and received bugfixes\nSupport for MathOptFormat v0.6 was added in FileFormats.MOF\nFileFormats.MPS has had bugfixes and support for more features such as OBJSENSE and objective constants.\nFileFormats.NL has been added to support nonlinear files\nImproved type inference throughout to reduce latency","category":"page"},{"location":"moi/release_notes/#Updating","page":"Release notes","title":"Updating","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"A helpful script when updating is:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"for (root, dirs, files) in walkdir(\".\")\n    for file in files\n        if !endswith(file, \".jl\")\n            continue\n        end\n        path = joinpath(root, file)\n        s = read(path, String)\n        for pair in [\n            \".variable_index\" => \".variable\",\n            \"RawParameter\" => \"RawOptimizerAttribute\",\n            \"ListOfConstraints\" => \"ListOfConstraintTypesPresent\",\n            \"TestConfig\" => \"Config\",\n            \"attr.N\" => \"attr.result_index\",\n            \"SolveTime\" => \"SolveTimeSec\",\n            \"DataType\" => \"Type\",\n            \"Utilities.supports_default_copy_to\" =>\n                \"supports_incremental_interface\",\n            \"SingleVariableConstraintNameError\" =>\n                \"VariableIndexConstraintNameError\",\n            \"SettingSingleVariableFunctionNotAllowed\" =>\n                \"SettingVariableIndexFunctionNotAllowed\",\n            \"automatic_copy_to\" => \"default_copy_to\",\n        ]\n            s = replace(s, pair)\n        end\n        write(path, s)\n    end\nend","category":"page"},{"location":"moi/release_notes/#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)","page":"Release notes","title":"v0.9.22 (May 22, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"This release contains backports from the ongoing development of the v0.10 release.","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved type inference in Utilities, Bridges and FileFormats submodules to reduce latency.\nImproved performance of Utilities.is_canonical.\nFixed Utilities.pass_nonvariable_constraints with bridged variables.\nFixed performance regression of Utilities.Model.\nFixed ordering of objective setting in parser.","category":"page"},{"location":"moi/release_notes/#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)","page":"Release notes","title":"v0.9.21 (April 23, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added supports_shift_constant.\nImprove performance of bridging quadratic constraints.\nAdd precompilation statements.\nLarge improvements to the documentation.\nFix a variety of inference issues, benefiting precompilation and reducing initial latency.\nRawParameters are now ignored when resetting a CachingOptimizer. Previously, changing the underlying optimizer after RawParameters were set would throw an error.\nUtilities.AbstractModel is being refactored. This may break users interacting with private fields of a model generated using @model.","category":"page"},{"location":"moi/release_notes/#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)","page":"Release notes","title":"v0.9.20 (February 20, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved performance of Utilities.ScalarFunctionIterator\nAdded support for compute_conflict to MOI layers\nAdded test with zero off-diagonal quadratic term in objective\nFixed double deletion of nested bridged SingleVariable/VectorOfVariables constraints\nFixed modification of un-set objective\nFixed function modification with duplicate terms\nMade unit tests abort without failing if the problem class is not supported\nFormatted code with JuliaFormatter\nClarified BasisStatusCode's docstring","category":"page"},{"location":"moi/release_notes/#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)","page":"Release notes","title":"v0.9.19 (December 1, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added CallbackNodeStatus attribute\nAdded bridge from GreaterThan or LessThan to Interval\nAdded tests for infeasibility certificates and double optimize\nFixed support for Julia v1.6\nRe-organized MOI docs and added documentation for adding a test","category":"page"},{"location":"moi/release_notes/#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)","page":"Release notes","title":"v0.9.18 (November 3, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements for working with complex numbers\nAdded GeoMeantoRelEntrBridge to bridge a GeometricMeanCone constraint to a relative entropy constraint","category":"page"},{"location":"moi/release_notes/#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)","page":"Release notes","title":"v0.9.17 (September 21, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed CleverDict with variable of negative index value\nImplement supports_add_constrained_variable for MockOptimizer","category":"page"},{"location":"moi/release_notes/#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)","page":"Release notes","title":"v0.9.16 (September 17, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes:\n32-bit support\nCleverDict with abstract value type\nChecks in test suite","category":"page"},{"location":"moi/release_notes/#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)","page":"Release notes","title":"v0.9.15 (September 14, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Bridges improvements:\n(R)SOCtoNonConvexQuad bridge\nZeroOne bridge\nUse supports_add_constrained_variable in LazyBridgeOptimizer\nExposed VariableBridgeCost and ConstraintBridgeCost attributes\nPrioritize constraining variables on creation according to these costs\nRefactor bridge debugging\nLarge performance improvements across all submodules\nLots of documentation improvements\nFileFormats improvements:\nUpdate MathOptFormat to v0.5\nFix supported objectives in FileFormats\nTesting improvements:\nAdd name option for basic_constraint_test\nBug fixes and missing methods\nAdd length for iterators\nFix bug with duplicate terms\nFix order of LinearOfConstraintIndices","category":"page"},{"location":"moi/release_notes/#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)","page":"Release notes","title":"v0.9.14 (May 30, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).\nBump JSONSchema dependency from v0.2 to v0.3 (#1090).\nDocumentation improvements:\nFix typos (#1054, #1060, #1061, #1064, #1069, #1070).\nRemove the outdated recommendation for a package implementing MOI for a solver XXX to be called MathOptInterfaceXXX (#1087).\nUtilities improvements:\nFix is_canonical for quadratic functions (#1081, #1089).\nImplement add_constrained_variable[s] for CachingOptimizer so that it is added as constrained variables to the underlying optimizer (#1084).\nAdd support for custom objective functions for UniversalFallback (#1086).\nDeterministic ordering of constraints in UniversalFallback (#1088).\nTesting improvements:\nAdd NormOneCone/NormInfinityCone tests (#1045).\nBridges improvements:\nAdd bridges from Semiinteger and Semicontinuous (#1059).\nImplement getting ConstraintSet for Variable.FlipSignBridge (#1066).\nFix setting ConstraintFunction for Constraint.ScalarizeBridge (#1093).\nFix NormOne/NormInf bridges with nonzero constants (#1045).\nFix StackOverflow in debug (#1063).\nFileFormats improvements:\n[SDPA] Implement the extension for integer variables (#1079).\n[SDPA] Ignore comments after m and nblocks and detect dat-s extension (#1077).\n[SDPA] No scaling of off-diagonal coefficient (#1076).\n[SDPA] Add missing negation of constant (#1075).","category":"page"},{"location":"moi/release_notes/#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)","page":"Release notes","title":"v0.9.13 (March 24, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added tests for Semicontinuous and Semiinteger variables (#1033).\nAdded tests for using ExprGraphs from NLP evaluators (#1043).\nUpdate version compatibilities of dependencies (#1034, #1051, #1052).\nFixed typos in documentation (#1044).","category":"page"},{"location":"moi/release_notes/#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)","page":"Release notes","title":"v0.9.12 (February 28, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed writing NLPBlock in MathOptFormat (#1037).\nFixed MockOptimizer for result attributes with non-one result index (#1039).\nUpdated test template with instantiate (#1032).","category":"page"},{"location":"moi/release_notes/#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)","page":"Release notes","title":"v0.9.11 (February 21, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add an option for the model created by Utilities.@model to be a subtype of AbstractOptimizer (#1031).\nDescribed dual cone in docstrings of GeoMeanCone and RelativeEntropyCone (#1018, #1028).\nFixed typos in documentation (#1022, #1024).\nFixed warning of unsupported attribute (#1027).\nAdded more rootdet/logdet conic tests (#1026).\nImplemented ConstraintDual for Constraint.GeoMeanBridge, Constraint.RootDetBridge and Constraint.LogDetBridge and test duals in tests with GeoMeanCone and RootDetConeTriangle and LogDetConeTriangle cones (#1025, #1026).","category":"page"},{"location":"moi/release_notes/#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)","page":"Release notes","title":"v0.9.10 (January 31, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added OptimizerWithAttributes grouping an optimizer constructor and a list of optimizer attributes (#1008).\nAdded RelativeEntropyCone with corresponding bridge into exponential cone constraints (#993).\nAdded NormSpectralCone and NormNuclearCone with corresponding bridges into positive semidefinite constraints (#976).\nAdded supports_constrained_variable(s) (#1004).\nAdded dual_set_type (#1002).\nAdded tests for vector specialized version of delete (#989, #1011).\nAdded PSD3 test (#1007).\nClarified dual solution of Tests.pow1v and Tests.pow1f (#1013).\nAdded support for EqualTo and Zero in Bridges.Constraint.SplitIntervalBridge (#1005).\nFixed Utilities.vectorize for empty vector (#1003).\nFixed free variables in LP writer (#1006).","category":"page"},{"location":"moi/release_notes/#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)","page":"Release notes","title":"v0.9.9 (December 29, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).\nImproved performance of deletion of vector of variables in MOI.Utilities.Model (#983).\nUpdated to MutableArithmetics v0.2 (#981).\nAdded MutableArithmetics.promote_operation allocation tests (#975).\nFixed inference issue on Julia v1.1 (#982).","category":"page"},{"location":"moi/release_notes/#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)","page":"Release notes","title":"v0.9.8 (December 19, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented MutableArithmetics API (#924).\nFixed callbacks with CachingOptimizer (#959).\nFixed MOI.dimension for MOI.Complements (#948).\nAdded fallback for add_variables (#972).\nAdded is_diagonal_vectorized_index utility (#965).\nImproved linear constraints display in manual (#963, #964).\nBridges improvements:\nAdded IndicatorSet to SOS1 bridge (#877).\nAdded support for starting values for Variable.VectorizeBridge (#944).\nFixed MOI.add_constraints with non-bridged variable constraint on bridged variable (#951).\nFixed corner cases and docstring of GeoMeanBridge (#961, #962, #966).\nFixed choice between variable or constraint bridges for constrained variables (#973).\nImprove performance of bridge shortest path (#945, #946, #956).\nAdded docstring for test_delete_bridge (#954).\nAdded Variable bridge tests (#952).","category":"page"},{"location":"moi/release_notes/#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)","page":"Release notes","title":"v0.9.7 (October 30, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented _result_index_field for NLPBlockDual (#934).\nFixed copy of model with starting values for vector constraints (#941).\nBridges improvements:\nImproved performance of add_bridge and added has_bridge (#935).\nAdded AbstractSetMapBridge for bridges between sets S1, S2 such that there is a linear map A such that A*S1 = S2 (#933).\nAdded support for starting values for FlipSignBridge, VectorizeBridge, ScalarizeBridge, SlackBridge, SplitIntervalBridge, RSOCBridge, SOCRBridge NormInfinityBridge, SOCtoPSDBridge and RSOCtoPSDBridge (#933, #936, #937, #938, #939).","category":"page"},{"location":"moi/release_notes/#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)","page":"Release notes","title":"v0.9.6 (October 25, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added complementarity constraints (#913).\nAllowed ModelLike objects as value of attributes (#928).\nTesting improvements:\nAdded dual_objective_value option to MOI.Test.TestConfig (#922).\nAdded InvalidIndex tests in basic_constraint_tests (#921).\nAdded tests for the constant term in indicator constraint (#929).\nBridges improvements:\nAdded support for starting values for Functionize bridges (#923).\nAdded variable indices context to variable bridges (#920).\nFixed a typo in printing o debug_supports (#927).","category":"page"},{"location":"moi/release_notes/#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)","page":"Release notes","title":"v0.9.5 (October 9, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Clarified PrimalStatus/DualStatus to be NO_SOLUTION if result_index is out of bounds (#912).\nAdded tolerance for checks  and use ResultCount + 1 for the result_index in MOI.Test.solve_result_status (#910, #917).\nUse 0.5 instead of 2.0 for power in PowerCone in basic_constraint_test (#916).\nBridges improvements:\nAdded debug utilities for unsupported variable/constraint/objective (#861).\nFixed deletion of variables in bridged VectorOfVariables constraints (#909).\nFixed result_index with objective bridges (#911).","category":"page"},{"location":"moi/release_notes/#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)","page":"Release notes","title":"v0.9.4 (October 2, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added solver-independent MIP callbacks (#782).\nImplements submit for Utilities.CachingOptimizer and Bridges.AbstractBridgeOptimizer (#906).\nAdded tests for result count of solution attributes (#901, #904).\nAdded NumberOfThreads attribute (#892).\nAdded Utilities.get_bounds to get the bounds on a variable (#890).\nAdded a note on duplicate coefficients in documentation (#581).\nAdded result index in ConstraintBasisStatus (#898).\nAdded extension dictionary to Utilities.Model (#884, #895).\nFixed deletion of constrained variables for CachingOptimizer (#905).\nImplemented Utilities.shift_constraint for Test.UnknownScalarSet (#896).\nBridges improvements:\nAdded Variable.RSOCtoSOCBridge (#907).\nImplemented MOI.get for ConstraintFunction/ConstraintSet for Bridges.Constraint.SquareBridge (#899).","category":"page"},{"location":"moi/release_notes/#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)","page":"Release notes","title":"v0.9.3 (September 20, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ambiguity detected in Julia v1.3 (#891, #893).\nFixed missing sets from ListOfSupportedConstraints (#880).\nFixed copy of VectorOfVariables constraints with duplicate indices (#886).\nAdded extension dictionary to MOIU.Model (#884).\nImplemented MOI.get for function and set for GeoMeanBridge (#888).\nUpdated documentation for SingleVariable indices and bridges (#885).\nTesting improvements:\nAdded more comprehensive tests for names (#882).\nAdded tests for SingleVariable duals (#883).\nAdded tests for DualExponentialCone and DualPowerCone (#873).\nImprovements for arbitrary coefficient type:\nFixed == for sets with mutable fields (#887).\nRemoved some Float64 assumptions in bridges (#878).\nAutomatic selection of Constraint.[Scalar|Vector]FunctionizeBridge (#889).","category":"page"},{"location":"moi/release_notes/#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)","page":"Release notes","title":"v0.9.2 (September 5, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented model printing for MOI.ModelLike and specialized it for models defined in MOI (864).\nGeneralized contlinear tests for arbitrary coefficient type (#855).\nFixed supports_constraint for Semiinteger and Semicontinuous and supports for ObjectiveFunction (#859).\nFixed Allocate-Load copy for single variable constraints (#856).\nBridges improvements:\nAdd objective bridges (#789).\nFixed Variable.RSOCtoPSDBridge for dimension 2 (#869).\nAdded Variable.SOCtoRSOCBridge (#865).\nAdded Constraint.SOCRBridge and disable MOI.Bridges.Constraint.SOCtoPSDBridge (#751).\nFixed added_constraint_types for Contraint.LogDetBridge and Constraint.RootDetBridge (#870).","category":"page"},{"location":"moi/release_notes/#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)","page":"Release notes","title":"v0.9.1 (August 22, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix support for Julia v1.2 (#834).\nL1 and L∞ norm epigraph cones and corresponding bridges to LP were added (#818).\nAdded tests to MOI.Test.nametest (#833).\nFix MOI.Test.soc3test for solvers not supporting infeasibility certificates (#839).\nImplements operate for operators * and / between vector function and constant (#837).\nImplements show for MOI.Utilities.IndexMap (#847).\nFix corner cases for mapping of variables in MOI.Utilities.CachingOptimizer and substitution of variables in MOI.Bridges.AbstractBridgeOptimizer (#848).\nFix transformation of constant terms for MOI.Bridges.Constraint.SOCtoPSDBridge and MOI.Bridges.Constraint.RSOCtoPSDBridge (#840).","category":"page"},{"location":"moi/release_notes/#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)","page":"Release notes","title":"v0.9.0 (August 13, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Support for Julia v0.6 and v0.7 was dropped (#714, #717).\nA MOI.Utilities.Model implementation of ModelLike, this should replace most use cases of MOI.Utilities.@model (#781).\nadd_constrained_variable and add_constrained_variables were added (#759).\nSupport for indicator constraints was added (#709, #712).\nDualObjectiveValue attribute was added (#473).\nRawParameter attribute was added (#733).\nA dual_set function was added (#804).\nA Benchmarks submodule was added to facilitate solver benchmarking (#769).\nA submit function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).\nThe field of ObjectiveValue was renamed to result_index (#729).\nThe _constant and Utilities.getconstant function were renamed to constant\nREDUCTION_CERTIFICATE result status was added (#734).\nAbstract matrix sets were added (#731).\nTesting improvements:\nThe testing guideline was updated (#728).\nQuadratic tests were added (#697).\nUnit tests for RawStatusString, SolveTime, Silent and SolverName were added (#726, #741).\nA rotated second-order cone test was added (#759).\nA power cone test was added (#768).\nTests for ZeroOne variables with variable bounds were added (#772).\nAn unbounded test was added (#773).\nExisting tests had a few updates (#702, #703, #763).\nDocumentation improvements:\nAdded a section on CachingOptimizer (#777).\nAdded a section on UniversalFallback, Model and @model (#762).\nTransition the knapsack example to a doctest with MockOptimizer (#786).\nUtilities improvements:\nA CleverDict utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).\nThe Utilities.constant function was renamed to Utilities.constant_vector (#740).\nImplement optimizer attributes for CachingOptimizer (#745).\nRename Utilities.add_scalar_constraint to Utilities.normalize_and_add_constraint (#801).\noperate with vcat, SingleVariable and VectorOfVariables now returns a VectorOfVariables (#616).\nFix a type piracy of operate (#784).\nThe load_constraint fallback signature was fixed (#760).\nThe set_dot function was extended to work with sparse arrays (#805).\nBridges improvements:\nThe bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement ConstraintFunction and ConstraintSet if the user wants to recover them. As a consequence, the @bridge macro was removed (#722).\nBridge are now instantiated with a bridge_constraint function instead of using a constructor (#730).\nFix constraint attributes for bridges (#699).\nConstraint bridges were moved to the Bridges/Constraint submodule so they should now inherit from MOI.Bridges.Constraint.Abstract and should implement MOI.Bridges.Constraint.concrete_bridge_type instead of MOI.Bridges.concrete_bridge_type (#756).\nVariable bridges were added in (#759).\nVarious improvements (#746, #747).","category":"page"},{"location":"moi/release_notes/#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)","page":"Release notes","title":"v0.8.4 (March 13, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Performance improvement in default_copy_to and bridge optimizer (#696).\nAdd Silent and implement setting optimizer attributes in caching and mock optimizers (#695).\nAdd Functionize bridges (SingleVariable and VectorOfVariables) (#659).\nMinor typo fixes (#694).","category":"page"},{"location":"moi/release_notes/#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)","page":"Release notes","title":"v0.8.3 (March 6, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Use zero constant in scalar constraint function of MOI.Test.copytest (#691).\nFix variable deletion with SingleVariable objective function (#690).\nFix LazyBridgeOptimizer with bridges that add no constraints (#689).\nError message improvements (#673, #685, #686, #688).\nDocumentation improvements (#682, #683, #687).\nBasis status:\nRemove VariableBasisStatus (#679).\nTest ConstraintBasisStatus and implement it in bridges (#678).\nFix inference of NumberOfVariables and NumberOfConstraints (#677).\nImplement division between a quadratic function and a number (#675).","category":"page"},{"location":"moi/release_notes/#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)","page":"Release notes","title":"v0.8.2 (February 7, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add RawStatusString attribute (#629).\nDo not set names to the optimizer but only to the cache in CachingOptimizer (#638).\nMake scalar MOI functions act as scalars in broadcast (#646).\nAdd function utilities:\nImplement Base.zero (#634), Base.iszero (#643), add missing arithmetic operations (#644, #645) and fix division (#648).\nAdd a vectorize function that turns a vector of ScalarAffineFunction into a VectorAffineFunction (#642).\nImprove support for starting values:\nShow a warning in copy when starting values are not supported instead of throwing an error (#630).\nFix UniversalFallback for getting an variable or constraint attribute set to no indices (#623).\nAdd a test in contlineartest with partially set VariablePrimalStart.\nBridges improvements:\nFix StackOverFlow in LazyBridgeOptimizer when there is a cycle in the graph of bridges.\nAdd Slack bridges (#610, #650).\nAdd FlipSign bridges (#658).\nAdd tests with duplicate coefficients in ScalarAffineFunction and VectorAffineFunction (#639).\nUse tolerance to compare VariablePrimal in rotatedsoc1 test (#632).\nUse a zero constant in ScalarAffineFunction of constraints in psdt2 (#622).","category":"page"},{"location":"moi/release_notes/#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)","page":"Release notes","title":"v0.8.1 (January 7, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Adding an NLP objective now overrides any objective set using the ObjectiveFunction attribute (#619).\nRename fullbridgeoptimizer into full_bridge_optimizer (#621).\nAllow custom constraint types with full_bridge_optimizer (#617).\nAdd Vectorize bridge which transforms scalar linear constraints into vector linear constraints (#615).","category":"page"},{"location":"moi/release_notes/#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)","page":"Release notes","title":"v0.8.0 (December 18, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Rename all enum values to follow the JuMP naming guidelines for constants, for example, Optimal becomes OPTIMAL, and DualInfeasible becomes DUAL_INFEASIBLE.\nRename CachingOptimizer methods for style compliance.\nAdd an MOI.TerminationStatusCode called ALMOST_DUAL_INFEASIBLE.","category":"page"},{"location":"moi/release_notes/#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)","page":"Release notes","title":"v0.7.0 (December 13, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Test that MOI.TerminationStatus is MOI.OptimizeNotCalled before MOI.optimize! is called.\nCheck supports_default_copy_to in tests (#594).\nKey pieces of information like optimality, infeasibility, etc., are now reported through TerminationStatusCode. It is typically no longer necessary to check the result statuses in addition to the termination status.\nAdd perspective dimension to log-det cone (#593).","category":"page"},{"location":"moi/release_notes/#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)","page":"Release notes","title":"v0.6.4 (November 27, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add OptimizeNotCalled termination status (#577) and improve documentation of other statuses (#575).\nAdd a solver naming guideline (#578).\nMake FeasibilitySense the default ObjectiveSense (#579).\nFix Utilities.@model and Bridges.@bridge macros for functions and sets defined outside MOI (#582).\nDocument solver-specific attributes (#580) and implement them in Utilities.CachingOptimizer (#565).","category":"page"},{"location":"moi/release_notes/#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)","page":"Release notes","title":"v0.6.3 (November 16, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)\nAttributes may now be partially set (some values could be nothing). (#563)\nPerformance improvements in Utilities.Model (#549, #567, #568)\nFix bug in QuadtoSOC (#558).\nNew supports_default_copy_to method that optimizers should implement to control caching behavior.\nDocumentation improvements.","category":"page"},{"location":"moi/release_notes/#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)","page":"Release notes","title":"v0.6.2 (October 26, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improve hygiene of @model macro (#544).\nFix bug in copy tests (#543).\nFix bug in UniversalFallback attribute getter (#540).\nAllow all correct solutions for solve_blank_obj unit test (#537).\nAdd errors for Allocate-Load and bad constraints (#534).\n[performance] Add specialized implementation of hash for VariableIndex (#533).\n[performance] Construct the name to object dictionaries lazily in model (#535).\nAdd the QuadtoSOC bridge which transforms ScalarQuadraticFunction constraints into RotatedSecondOrderCone (#483).","category":"page"},{"location":"moi/release_notes/#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)","page":"Release notes","title":"v0.6.1 (September 22, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Enable PositiveSemidefiniteConeSquare set and quadratic functions in MOIB.fullbridgeoptimizer (#524).\nAdd warning in the bridge between PositiveSemidefiniteConeSquare and PositiveSemidefiniteConeTriangle when the matrix is almost symmetric (#522).\nModify MOIT.copytest to not add multiples constraints on the same variable (#521).\nAdd missing keyword argument in one of MOIU.add_scalar_constraint methods (#520).","category":"page"},{"location":"moi/release_notes/#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)","page":"Release notes","title":"v0.6.0 (August 30, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The MOIU.@model and MOIB.@bridge macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.\nRename functions according to the JuMP style guide:\ncopy! with keyword arguments copynames and warnattributes -> copy_to with keyword arguments copy_names and warn_attributes;\nset! -> set;\naddvariable[s]! -> add_variable[s];\nsupportsconstraint -> supports_constraint;\naddconstraint[s]! -> add_constraint[s];\nisvalid -> is_valid;\nisempty -> is_empty;\nBase.delete! -> delete;\nmodify! -> modify;\ntransform! -> transform;\ninitialize! -> initialize;\nwrite -> write_to_file; and\nread! -> read_from_file.\nRemove free! (use Base.finalize instead).\nAdd the SquarePSD bridge which transforms PositiveSemidefiniteConeTriangle constraints into PositiveSemidefiniteConeTriangle.\nAdd result fallback for ConstraintDual of variable-wise constraint, ConstraintPrimal and ObjectiveValue.\nAdd tests for ObjectiveBound.\nAdd test for empty rows in vector linear constraint.\nRework errors: CannotError has been renamed NotAllowedError and the distinction between UnsupportedError and NotAllowedError is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)\ncanget is removed. NoSolution is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).","category":"page"},{"location":"moi/release_notes/#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)","page":"Release notes","title":"v0.5.0 (August 5, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix names with CachingOptimizer.\nCleanup thanks to @mohamed82008.\nAdded a universal fallback for constraints.\nFast utilities for function canonicalization thanks to @rdeits.\nRenamed dimension field to side_dimension in the context of matrix-like sets.\nNew and improved tests for cases like duplicate terms and ObjectiveBound.\nRemoved cantransform, canaddconstraint, canaddvariable, canset, canmodify, and candelete functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of UnsupportedError indicate unsupported operations, while subtypes of CannotError indicate operations that cannot be performed in the current state.\nThe API for copy! is updated to remove the CopyResult type.\nUpdates for the new JuMP style guide.","category":"page"},{"location":"moi/release_notes/#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)","page":"Release notes","title":"v0.4.1 (June 28, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixes vector function modification on 32 bits.\nFixes Bellman-Ford algorithm for bridges.\nAdded an NLP test with FeasibilitySense.\nUpdate modification documentation.","category":"page"},{"location":"moi/release_notes/#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)","page":"Release notes","title":"v0.4.0 (June 23, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Helper constructors for VectorAffineTerm and VectorQuadraticTerm.\nAdded modify_lhs to TestConfig.\nAdditional unit tests for optimizers.\nAdded a type parameter to CachingOptimizer for the optimizer field.\nNew API for problem modification (#388)\nTests pass without deprecation warnings on Julia 0.7.\nSmall fixes and documentation updates.","category":"page"},{"location":"moi/release_notes/#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)","page":"Release notes","title":"v0.3.0 (May 25, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.\nImprovements to MockOptimizer.\nSignificant changes to Bridges.\nNew and improved unit tests.\nFixes for Julia 0.7.","category":"page"},{"location":"moi/release_notes/#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)","page":"Release notes","title":"v0.2.0 (April 24, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improvements to and better coverage of Tests.\nDocumentation fixes.\nSolverName attribute.\nChanges to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).\nAddition of NLP tests.\nIntroduction of UniversalFallback.\ncopynames keyword argument to MOI.copy!.\nAdd Bridges submodule.","category":"page"},{"location":"moi/release_notes/#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)","page":"Release notes","title":"v0.1.0 (February 28, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Initial public release.\nThe framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"developers/extensions/#extensions_manual","page":"Extensions","title":"Extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP provides a variety of ways to extend the basic modeling functionality.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nThis documentation in this section is still a work-in-progress. The best place to look for ideas and help when writing a new JuMP extension are existing JuMP extensions. Examples include:BilevelJuMP.jl\nColuna.jl\nInfiniteOpt.jl\nPlasmo.jl\nPolyJuMP.jl\nSDDP.jl\nStochasticPrograms.jl\nSumOfSquares.jl\nvOptGeneric.jl","category":"page"},{"location":"developers/extensions/#Compatibility","page":"Extensions","title":"Compatibility","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"When writing JuMP extensions, you should carefully consider the compatibility guarantees that JuMP makes. In particular:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"All functions, structs, and constants which do not begin with an underscore (_) are public. These are always safe to use, and they should all have corresponding documentation.\nAll identifiers beginning with an underscore (_) are private. These are not safe to use, because they may break in any JuMP release, including patch releases.\nUnless explicitly mentioned in the documentation, all fields of a struct are private. These are not safe to use, because they may break in any JuMP release, including patch releases. An example of a field which is safe to use is the model.ext extension dictionary, which is documented in The extension dictionary.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"In general, we strongly encourage you to use only the public API of JuMP. If you are missing a feature, please open a GitHub issue.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"However, if you do use the private API (for example, because your feature request has not been implemented yet), then you must carefully restrict the versions of JuMP that your package is compatible with in the Project.toml file. The easiest way to do this is via the hyphen specifiers. For example, if your package supports all JuMP versions between v1.0.0 and v1.1.1, do:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP = \"1.0.0 - 1.1.1\"","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Then, whenever JuMP releases a new version, you should check if your package is still compatible and update the bound accordingly.","category":"page"},{"location":"developers/extensions/#Define-a-new-set","page":"Extensions","title":"Define a new set","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To define a new set for JuMP, subtype MOI.AbstractScalarSet or MOI.AbstractVectorSet and implement Base.copy for the set.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct NewMOIVectorSet <: MOI.AbstractVectorSet\n           dimension::Int\n       end\n\njulia> Base.copy(x::NewMOIVectorSet) = x\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x in NewMOIVectorSet(2))\n[x[1], x[2]] ∈ NewMOIVectorSet(2)","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"However, for vector-sets, this requires the user to specify the dimension argument to their set, even though we could infer it from the length of x!","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"You can make a more user-friendly set by subtyping AbstractVectorSet and implementing moi_set.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct NewVectorSet <: JuMP.AbstractVectorSet end\n\njulia> JuMP.moi_set(::NewVectorSet, dim::Int) = NewMOIVectorSet(dim)\n\njulia> @constraint(model, x in NewVectorSet())\n[x[1], x[2]] ∈ NewMOIVectorSet(2)","category":"page"},{"location":"developers/extensions/#extend_variable_macro","page":"Extensions","title":"Extend @variable","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Just as Bin and Int create binary and integer variables, you can extend the @variable macro to create new types of variables. Here is an explanation by example, where we create a AddTwice type, that creates a tuple of two JuMP variables instead of a single variable.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"First, create a new struct. This can be anything. Our struct holds a VariableInfo object that stores bound information, and whether the variable is binary or integer.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct AddTwice\n           info::JuMP.VariableInfo\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Second, implement build_variable, which takes ::Type{AddTwice} as an argument, and returns an instance of AddTwice. Note that you can also receive keyword arguments.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function JuMP.build_variable(\n           _err::Function,\n           info::JuMP.VariableInfo,\n           ::Type{AddTwice};\n           kwargs...\n       )\n           println(\"Can also use $kwargs here.\")\n           return AddTwice(info)\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Third, implement add_variable, which takes the instance of AddTwice from the previous step, and returns something. Typically, you will want to call add_variable here. For example, our AddTwice call is going to add two JuMP variables.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function JuMP.add_variable(\n           model::JuMP.Model,\n           duplicate::AddTwice,\n           name::String,\n       )\n           a = JuMP.add_variable(\n               model,\n               JuMP.ScalarVariable(duplicate.info),\n               \"$(name)_a\",\n            )\n           b = JuMP.add_variable(\n               model,\n               JuMP.ScalarVariable(duplicate.info),\n               \"$(name)_b\",\n            )\n           return (a, b)\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Now AddTwice can be passed to @variable similar to Bin or Int, or through the variable_type keyword. However, now it adds two variables instead of one.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2], variable_type = AddTwice, kw = i)\nCan also use Base.Pairs(:kw => 1) here.\nCan also use Base.Pairs(:kw => 2) here.\n2-element Vector{Tuple{VariableRef, VariableRef}}:\n (x[1]_a, x[1]_b)\n (x[2]_a, x[2]_b)\n\njulia> num_variables(model)\n4\n\njulia> first(x[1])\nx[1]_a\n\njulia> last(x[2])\nx[2]_b","category":"page"},{"location":"developers/extensions/#Extend-[@constraint](@ref)","page":"Extensions","title":"Extend @constraint","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The @constraint macro has three steps that can be intercepted and extended: parse time, build time, and add time.","category":"page"},{"location":"developers/extensions/#Parse","page":"Extensions","title":"Parse","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To extend the @constraint macro at parse time, implement one of the following methods:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_head\nparse_constraint_call","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_head should be implemented to intercept an expression based on the .head field of Base.Expr. For example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP\n\njulia> const MutableArithmetics = JuMP._MA;\n\njulia> model = Model(); @variable(model, x);\n\njulia> function JuMP.parse_constraint_head(\n           _error::Function,\n           ::Val{:≔},\n           lhs,\n           rhs,\n       )\n           println(\"Rewriting ≔ as ==\")\n           new_lhs, parse_code = MutableArithmetics.rewrite(lhs)\n           build_code = :(\n               build_constraint($(_error), $(new_lhs), MOI.EqualTo($(rhs)))\n           )\n           return false, parse_code, build_code\n       end\n\njulia> @constraint(model, x + x ≔ 1.0)\nRewriting ≔ as ==\n2 x = 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_call should be implemented to intercept an expression of the form Expr(:call, op, args...). For example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP\n\njulia> const MutableArithmetics = JuMP._MA;\n\njulia> model = Model(); @variable(model, x);\n\njulia> function JuMP.parse_constraint_call(\n           _error::Function,\n           is_vectorized::Bool,\n           ::Val{:my_equal_to},\n           lhs,\n           rhs,\n       )\n           println(\"Rewriting my_equal_to to ==\")\n           new_lhs, parse_code = MutableArithmetics.rewrite(lhs)\n           build_code = if is_vectorized\n               :(build_constraint($(_error), $(new_lhs), MOI.EqualTo($(rhs)))\n           )\n           else\n               :(build_constraint.($(_error), $(new_lhs), MOI.EqualTo($(rhs))))\n           end\n           return parse_code, build_code\n       end\n\njulia> @constraint(model, my_equal_to(x + x, 1.0))\nRewriting my_equal_to to ==\n2 x = 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nWhen parsing a constraint you can recurse into sub-constraint (for example, the {expr} in z => {x <= 1}) by calling parse_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To prevent JuMP from promoting the set to the same value type as the model, use SkipModelConvertScalarSetWrapper.","category":"page"},{"location":"developers/extensions/#Build","page":"Extensions","title":"Build","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To extend the @constraint macro at build time, implement a new build_constraint method.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"This may mean implementing a method for a specific function or set created at parse time, or it may mean implementing a method which handles additional positional arguments.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"build_constraint must return an AbstractConstraint, which can either be an AbstractConstraint already supported by JuMP, for example, ScalarConstraint or VectorConstraint, or a custom AbstractConstraint with a corresponding add_constraint method (see Add).","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nThe easiest way to extend @constraint is via an additional positional argument to build_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here is an example of adding extra arguments to build_constraint:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model(); @variable(model, x);\n\njulia> struct MyConstrType end\n\njulia> function JuMP.build_constraint(\n            _error::Function,\n            f::JuMP.GenericAffExpr,\n            set::MOI.EqualTo,\n            extra::Type{MyConstrType};\n            d = 0,\n       )\n            new_set = MOI.LessThan(set.value + d)\n            return JuMP.build_constraint(_error, f, new_set)\n       end\n\njulia> @constraint(model, my_con, x == 0, MyConstrType, d = 2)\nmy_con : x ≤ 2","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"note: Note\nOnly a single positional argument can be given to a particular constraint. Extensions that seek to pass multiple arguments (for example, Foo and Bar) should combine them into one argument type (for example, FooBar).","category":"page"},{"location":"developers/extensions/#extension_add_constraint","page":"Extensions","title":"Add","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"build_constraint returns an AbstractConstraint object. To extend @constraint at add time, define a subtype of AbstractConstraint, implement build_constraint to return an instance of the new type, and then implement add_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here is an example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model(); @variable(model, x);\n\njulia> struct MyTag\n           name::String\n       end\n\njulia> struct MyConstraint{S} <: AbstractConstraint\n           name::String\n           f::AffExpr\n           s::S\n       end\n\njulia> function JuMP.build_constraint(\n            _error::Function,\n            f::AffExpr,\n            set::MOI.AbstractScalarSet,\n            extra::MyTag,\n       )\n            return MyConstraint(extra.name, f, set)\n       end\n\njulia> function JuMP.add_constraint(\n            model::Model,\n            con::MyConstraint,\n            name::String,\n       )\n            return add_constraint(\n                model,\n                ScalarConstraint(con.f, con.s),\n                \"$(con.name)[$(name)]\",\n            )\n       end\n\njulia> @constraint(model, my_con, 2x <= 1, MyTag(\"my_prefix\"))\nmy_prefix[my_con] : 2 x - 1 ≤ 0","category":"page"},{"location":"developers/extensions/#The-extension-dictionary","page":"Extensions","title":"The extension dictionary","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Every JuMP model has a field .ext::Dict{Symbol,Any} that can be used by extensions. This is useful if your extensions to @variable and @constraint need to store information between calls.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The most common way to initialize a model with information in the .ext dictionary is to provide a new constructor:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function MyModel()\n           model = Model()\n           model.ext[:MyModel] = 1\n           return model\n       end\nMyModel (generic function with 1 method)\n\njulia> model = MyModel()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> model.ext\nDict{Symbol, Any} with 1 entry:\n  :MyModel => 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If you define extension data, implement copy_extension_data to support copy_model.","category":"page"},{"location":"developers/extensions/#Defining-new-JuMP-models","page":"Extensions","title":"Defining new JuMP models","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If extending individual calls to @variable and @constraint is not sufficient, it is possible to implement a new model via a subtype of AbstractModel. You can also define new AbstractVariableRefs to create different types of JuMP variables.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nExtending JuMP in this manner is an advanced operation. We strongly encourage you to consider how you can use the methods mentioned in the previous sections to achieve your aims instead of defining new model and variable types. Consult the developer chatroom before starting work on this.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If you define new types, you will need to implement a considerable number of methods, and doing so will require a detailed understanding of the JuMP internals. Therefore, the list of methods to implement is currently undocumented.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The easiest way to extend JuMP by defining a new model type is to follow an existing example. A simple example to follow is the JuMPExtension module in the JuMP test suite. The best example of an external JuMP extension that implements an AbstractModel is InfiniteOpt.jl.","category":"page"},{"location":"developers/extensions/#Testing-JuMP-extensions","page":"Extensions","title":"Testing JuMP extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The JuMP test suite contains a large number of tests for JuMP extensions. You can run these tests by copying the MIT-licensed Kokako.jl file in the JuMP tests into your /test folder, and then adding this snippet to your /test/runtests.jl file:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"using MyJuMPExtension\nimport JuMP\ninclude(\"Kokako.jl\")\nconst MODULES_TO_TEST = Kokako.include_modules_to_test(JuMP)\nKokako.run_tests(\n    MODULES_TO_TEST,\n    MyJuMPExtension.MyModel,\n    MyJuMPExtension.MyVariableRef;\n    test_prefix = \"test_extension_\",\n)","category":"page"},{"location":"developers/extensions/#Set-an-optimize!-hook","page":"Extensions","title":"Set an optimize! hook","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Some extensions require modification to the problem after the user has finished constructing the problem, but before optimize! is called. For these situations, JuMP provides set_optimize_hook, which lets you intercept the optimize! call.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here's a simple example of adding an optimize hook that extends optimize! to take a keyword argument silent:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP, HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> @variable(model, x >= 1.5, Int);\n\njulia> @objective(model, Min, x);\n\njulia> function silent_hook(model; silent::Bool)\n           if silent\n               set_silent(model)\n           else\n               unset_silent(model)\n           end\n           ## Make sure you set ignore_optimize_hook = true, or we'll\n           ## recursively enter the optimize hook!\n           return optimize!(model; ignore_optimize_hook = true)\n       end\nsilent_hook (generic function with 1 method)\n\njulia> set_optimize_hook(model, silent_hook)\nsilent_hook (generic function with 1 method)\n\njulia> optimize!(model; silent = true)\n\njulia> optimize!(model; silent = false)\nSolution has               num          max          sum\nCol     infeasibilities      0            0            0\nInteger infeasibilities      0            0            0\nRow     infeasibilities      0            0            0\nRow     residuals            0            0            0\nPresolving model\n0 rows, 0 cols, 0 nonzeros\n0 rows, 0 cols, 0 nonzeros\nPresolve: Optimal\n\nSolving report\n  Status            Optimal\n  Primal bound      2\n  Dual bound        2\n  Gap               0% (tolerance: 0.01%)\n  Solution status   feasible\n                    2 (objective)\n                    0 (bound viol.)\n                    0 (int. viol.)\n                    0 (row viol.)\n  Timing            0.00 (total)\n                    0.00 (presolve)\n                    0.00 (postsolve)\n  Nodes             0\n  LP iterations     0 (total)\n                    0 (strong br.)\n                    0 (separation)\n                    0 (heuristics)","category":"page"},{"location":"developers/extensions/#Creating-new-container-types","page":"Extensions","title":"Creating new container types","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP macros (for example, @variable) accept a container keyword argument to force the type of container that is chosen. By default, JuMP supports container = Array, container = DenseAxisArray, container = SparseAxisArray and container = Auto. You can extend support to user-defined types by implementing Containers.container.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"For example, here is a container that reverses the order of the indices:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct Foo end\n\njulia> function Containers.container(f::Function, indices, ::Type{Foo})\n           return reverse([f(i...) for i in indices])\n       end\n\njulia> model = Model();\n\njulia> @variable(model, x[1:3], container = Foo)\n3-element Vector{VariableRef}:\n x[3]\n x[2]\n x[1]\n\njulia> x[1]\nx[3]\n\njulia> @variable(model, y[1:3, 1:2], container = Foo)\n3×2 Matrix{VariableRef}:\n y[3,2]  y[3,1]\n y[2,2]  y[2,1]\n y[1,2]  y[1,1]\n\njulia> y[1, 1]\ny[3,2]\n\njulia> @variable(model, z[i=1:3; isodd(i)], container = Foo)\n2-element Vector{VariableRef}:\n z[3]\n z[1]\n\njulia> z[2]\nz[1]","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nIf you are a general user, you should not need to create a new container type. Instead, consider following User-defined containers and create a new container using standard Julia syntax. For example:julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> y = reverse(x)\n3-element Vector{VariableRef}:\n x[3]\n x[2]\n x[1]","category":"page"},{"location":"developers/extensions/#Performance-tips-for-extensions","page":"Extensions","title":"Performance tips for extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The function-in-set design of MathOptInterface causes type stability issues in Julia if you try to iterate over all of the constraints in a model. The easiest way to fix this is to use a function barrier.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"For example, instead of:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"function all_names_slow(model)\n    names = Set{String}()\n    for ci in all_constraints(model)\n        push!(names, name(ci))\n    end\n    return names\nend","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"use:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"function _function_barrier(names, model, ::Type{F}, ::Type{S}) where {F,S}\n    for ci in all_constraints(model, F, S)\n        push!(names, name(ci))\n    end\n    return\nend\n\nfunction all_names_fast(model)\n    names = Set{String}()\n    for (F, S) in list_of_constraint_types(model)\n        _function_barrier(names, model, F, S)\n    end\n    return names\nend","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"note: Note\nIt is important to explicitly type the F and S arguments. If you leave them untyped, for example, function _function_barrier(names, model, F, S), Julia will not specialize the function calls and performance will not be improved.","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/changelog.md\"","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"CurrentModule = MathOptInterface","category":"page"},{"location":"moi/changelog/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"moi/changelog/#v1.20.1-(September-24,-2023)","page":"Release notes","title":"v1.20.1 (September 24, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Removed incorrect kwargs in some copy_to methods (#2272)\nFixed ConstraintDualStart for Bridges.Constraint.SplitIntervalBridge (#2275)\nFixed MethodError when trying to modify a variable objective (#2278)\nFixed stack-overflow in Utilities.operate(+,...) with many arguments (#2285) (#2290)","category":"page"},{"location":"moi/changelog/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MathOptSetDistances to solver-tests.yml (#2265)\nUpdated Documenter (#2266)\nFixed various JET errors (#2267) (#2269) (#2270) (#2271) (#2276) (#2277) (#2289)\nVarious style improvements\nReplaced using Package with import Package where possible (#2274)\nRemoved Utilities.EMPTYSTRING (#2283)\nRemoved unnecessary const acronyms in Utilities (#2280) (#2281)\nRemoved invalid and unused method (#2286)\nRefactored src/Utilities/model.jl (#2287)","category":"page"},{"location":"moi/changelog/#v1.20.0-(September-7,-2023)","page":"Release notes","title":"v1.20.0 (September 7, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Scaled set (#2237) (#2263) (#2264)\nAdded ObjectiveLimit attribute (#2257)","category":"page"},{"location":"moi/changelog/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Updated dependencies (#2258)\nImproved performance of ScalarNonlinearFunction utilities (#2259)\nFixed docstrings (#2261)","category":"page"},{"location":"moi/changelog/#v1.19.0-(August-15,-2023)","page":"Release notes","title":"v1.19.0 (August 15, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added VectorNonlinearFunction (#2201)\nAdded new bridges\nBridges.Constraint.IntegerToZeroOneBridge (#2205)\nBridges.Constraint.ToScalarQuadraticBridge (#2235)\nBridges.Constraint.ToVectorQuadraticBridge (#2235)\nBridges.Constraint.ToScalarNonlinearBridge (#2233) (#2235)\nBridges.Constraint.FunctionConversionBridge (#2235)\nAdded Bridges.bridging_cost for setting a bridge-specific cost in the bridging graph (#2235)\nAdded Utilities.eval_variables support for ScalarNonlinearFunction (#2218) (#2219)\nAdded support for ScalarNonlinearFunction in FileFormats.NL (#2228) (#2231)\nAdded support for writing non-Float64 functions in FileFormats.MOF\nAdded Utilities.lazy_map (#2254)","category":"page"},{"location":"moi/changelog/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed method ambiguities in operate (#2224)\nFixed reading LP file with a -infinity <= x <= +infinity variable (#2225)\nFixed missing require in Test.test_nonlinear_duals (#2230)\nFixed bug in ConstraintPrimal of Bridges.Constraint.QuadtoSOCBridge (#2240)","category":"page"},{"location":"moi/changelog/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added extensions to solver-tests.yml (#2229)\nRefactored test/Benchmarks (#2234)\nFixed warnings in tests (#2241) (#2243)\nSmall refactoring of bridges for upcoming VectorNonlinearFunction (#2244) (#2245)\nFixed various typos (#2251) (#2255)\nPartitioned how we run the tests on GitHub actions (#2252) (#2253)","category":"page"},{"location":"moi/changelog/#v1.18.0-(June-23,-2023)","page":"Release notes","title":"v1.18.0 (June 23, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.SlackBridgePrimalDualStart (#2194)\nAdded constraint_gradient_structure and eval_constraint_gradient (#2200)","category":"page"},{"location":"moi/changelog/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a missing @require in MOI.Test (#2195) (#2196)\nFixed incorrect usage of Utilities.operate! in bridges (#2207) (#2216)\nFixed splatting nonlinear expression with univariate operator (#2221)","category":"page"},{"location":"moi/changelog/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Removed unused argument names (#2199)\nReduced memory requirement for tests (#2204)\nRefactored Utilities.promote_operation (#2206)\nImproved code style in Utilities/mutable_arithmetics.jl (#2209)\nRefactored various methods in Utilities/functions.jl (#2208) (#2212) (#2213) (#2214) (#2215)","category":"page"},{"location":"moi/changelog/#v1.17.1-(June-6,-2023)","page":"Release notes","title":"v1.17.1 (June 6, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed comparison of ScalarFunctionConstantNotZero (#2190)","category":"page"},{"location":"moi/changelog/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added documentation for enum instances (#2186)\nUpdated chatroom links in documentation (#2188)\nChanged the documentation to build on Julia v1.9 (#2191)","category":"page"},{"location":"moi/changelog/#v1.17.0-(June-1,-2023)","page":"Release notes","title":"v1.17.0 (June 1, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added ScalarNonlinearFunction (#2059)\nAdded a variety of tests for NormSpectralCone, NormNuclearCone, and HermitianPositiveSemidefiniteConeTriangle (#2174)\nAdded Utilities.value_type for computing the value type of a function (#2176)","category":"page"},{"location":"moi/changelog/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for external sets in Utilities.loadfromstring! (#2177)\nFixed promote_operation for ScalarNonlinearFunction (#2179)\nFixed two issues in FileFormats.LP when reading files with quadratic functions (#2182) (#2184)","category":"page"},{"location":"moi/changelog/#v1.16.0-(May-16,-2023)","page":"Release notes","title":"v1.16.0 (May 16, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for MathOptFormat v1.3 and v1.4 (#2158) (#2169)\nAdded new method to the nonlinear API (#2162) (#2164)\neval_hessian_constraint\neval_hessian_objective\nhessian_constraint_structure\nhessian_objective_structure\nAdded new sets\nNormCone (#2119)\nScaledPositiveSemidefiniteConeTriangle (#2154)","category":"page"},{"location":"moi/changelog/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for Julia v1.9 to work around a bug in the upstream Julia compiler (#2161) (#2163)\nFixed a correctness bug in Bridges.Constraint.HermitianToSymmetricPSDBridge (#2171)\nFixed convert(::VariableIndex, ::ScalarAffineFunction) when the function has terms with 0coefficients (#2173)","category":"page"},{"location":"moi/changelog/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed solver-tests.yml (#2157)\nUpdated documentation links to developer chatroom (#2160)\nAdded various tests for bridges (#2156)\nAdded checklists to the developer documentation (#2167) (#2168)","category":"page"},{"location":"moi/changelog/#v1.15.1-(April-25,-2023)","page":"Release notes","title":"v1.15.1 (April 25, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed deleting a variable in a bridged objective (#2150)","category":"page"},{"location":"moi/changelog/#v1.15.0-(April-19,-2023)","page":"Release notes","title":"v1.15.0 (April 19, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.VectorFunctionizeBridge (#2139)","category":"page"},{"location":"moi/changelog/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for Rational in Bridges.Constraint.SplitIntervalBridge (#2137)\nFixed printing of LaTeX models (#2141)\nFixed modify in Bridges.Objective.VectorSlackBridge (#2144)\nFixed NAME record with spaces in FileFormats.MPS (#2146)\nFixed deleting a variable in a bridged objective (#2147)","category":"page"},{"location":"moi/changelog/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add a test for variables in one-sided open Interval sets (#2133)\nMinor style fixes in the source code (#2148)","category":"page"},{"location":"moi/changelog/#v1.14.1-(April-6,-2023)","page":"Release notes","title":"v1.14.1 (April 6, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Bridges.print_active_bridges (#2135)","category":"page"},{"location":"moi/changelog/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a warning when an ambiguous string is passed to exclude in Test.runtests (#2136)","category":"page"},{"location":"moi/changelog/#v1.14.0-(April-4,-2023)","page":"Release notes","title":"v1.14.0 (April 4, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for starting values in Bridges.Constraint.QuadtoSOCBridge (#2115)\nAdded support for Regex in the include and exclude arguments to Test.runtests (#2129)\nAdded Bridges.print_active_bridges methods for individual objectives and constraints (#2128)","category":"page"},{"location":"moi/changelog/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ResultCount when parsing .sol files in FileFormats.NL (#2130)","category":"page"},{"location":"moi/changelog/#v1.13.2-(March-21,-2023)","page":"Release notes","title":"v1.13.2 (March 21, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed splatting of containers in MOI.Nonlinear (#2120)\nFixed a bug reading LP files with default bounds (#2121)\nFixed a bug in which Bridges.Constraint.HermitianToSymmetricPSDBridge was not enabled by default (#2123)","category":"page"},{"location":"moi/changelog/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#2114)\nFunctions now print to the REPL in algebraic form. This is potentially breaking if you have tests which rely on a specific String form of MOI functions. (#2112) (#2126)","category":"page"},{"location":"moi/changelog/#v1.13.1-(March-3,-2023)","page":"Release notes","title":"v1.13.1 (March 3, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added the Google style guide to the documentation linter Vale, and fixed the resulting warnings (#2110)\nImproved the docstrings in src/functions.jl (#2108)","category":"page"},{"location":"moi/changelog/#v1.13.0-(February-28,-2023)","page":"Release notes","title":"v1.13.0 (February 28, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Constraint.NumberConversionBridge (#2091)\nAdded Parameter set (#2095) (#2105) (#2106) (#2109)\nAdded with_cache_type argument to instantiate (#2097)\nAdded support for HermitianPositiveSemidefiniteConeTriangle in Utilities.Model (#2100)","category":"page"},{"location":"moi/changelog/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug when Utilities.@product_of_sets is empty (#2101)\nFixed Bridges.print_active_bridges when variable bridge is an AbstractScalarSet (#2107)","category":"page"},{"location":"moi/changelog/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added tests for vector-valued objective functions in FileFormats.MOF (#2093)\nUsed and documented preference for import MathOptInterface as MOI (#2096)\nFix and test links in the documentation with linkcheck = true (#2098)\nImproved docstrings of sets in src/sets.jl (#2099)\nSkip checking flakey links in documentation with linkcheck_ignore  (#2103)","category":"page"},{"location":"moi/changelog/#v1.12.0-(February-10,-2023)","page":"Release notes","title":"v1.12.0 (February 10, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for vector-valued objective functions (#2070)\nAdded a Utilities.distance_to_set method for SecondOrderCone (#2060)","category":"page"},{"location":"moi/changelog/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a number of constraint bridges so that Bridges.final_touch can be called multiple times without forcing a rebuild of the reformulation (#2089)","category":"page"},{"location":"moi/changelog/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new tests that a set ObjectiveFunction appears in ListOfModelAttributesSet (#2085)\nImproved the docstrings of a number of constraint-programming related sets (#2087)","category":"page"},{"location":"moi/changelog/#v1.11.5-(January-24,-2023)","page":"Release notes","title":"v1.11.5 (January 24, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .lp files with an off-diagonal quadratic objective (#2082)","category":"page"},{"location":"moi/changelog/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added SnoopPrecompile directives for reduced time-to-first-X in Julia v1.9 (#2080)","category":"page"},{"location":"moi/changelog/#v1.11.4-(January-12,-2023)","page":"Release notes","title":"v1.11.4 (January 12, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .lp files with an Integer section (#2078)","category":"page"},{"location":"moi/changelog/#v1.11.3-(January-12,-2023)","page":"Release notes","title":"v1.11.3 (January 12, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance bug when deleting a vector of constraints (#2072)\nFixed a bug reading .lp files with terms like x -1 y (#2076)","category":"page"},{"location":"moi/changelog/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Documented the two-argument method of optimize! (#2074)","category":"page"},{"location":"moi/changelog/#v1.11.2-(January-2,-2023)","page":"Release notes","title":"v1.11.2 (January 2, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with ConstraintName set for VariableIndex constraints (#2066)\nFixed a bug reading .mof.json files with nonlinear objectives and no constraints (#2068)","category":"page"},{"location":"moi/changelog/#v1.11.1-(December-22,-2022)","page":"Release notes","title":"v1.11.1 (December 22, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with integer coefficients for affine and quadratic functions (#2063)","category":"page"},{"location":"moi/changelog/#v1.11.0-(December-2,-2022)","page":"Release notes","title":"v1.11.0 (December 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Utilities.PenaltyRelaxation and Utilities.ScalarPenaltyRelaxation (#1995)\nAdded Utilities.distance_to_set (#2048)\nAdded support for ConstraintPrimalStart and ConstraintDualStart in FileFormats.MOF (#2056)","category":"page"},{"location":"moi/changelog/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Tidied these release notes (#2055)","category":"page"},{"location":"moi/changelog/#v1.10.0-(November-22,-2022)","page":"Release notes","title":"v1.10.0 (November 22, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new methods set(::OptimizerWithAttributes, ::RawOptimizerAttribute, value) and get(::OptimizerWithAttributes, ::RawOptimizerAttribute) (#2049)\nAdded new methods Utilities.DoubleDicts.outer_keys and Utilities.DoubleDicts.nonempty_outer_keys (#2052)","category":"page"},{"location":"moi/changelog/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed Bridges.Objective.SlackBridge when the objective function is complex-valued (#2036) (#2038)\nFixed docstring of Test.runtests to clarify the warn_unsupported argument (#2037)\nFixed reading of free variables in FileFormats.LP (#2044)\nFixed numerous edge cases reading files from QPLIB using FileFormats.LP (#2042) (#2044)\nFixed situations in which x^y returns a complex value in Nonlinear (#2050)","category":"page"},{"location":"moi/changelog/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved the error message thrown when a user-defined nonlinear function does not accept splatted input (#2032)\nRemoved specialized iterators for keys and values in Utilities.CleverDicts (#2051)","category":"page"},{"location":"moi/changelog/#v1.9.0-(October-29,-2022)","page":"Release notes","title":"v1.9.0 (October 29, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added default fallback for getting ListOfConstraintIndices and NumberOfConstraints when the constraint type is unsupported by the model (#2021)\nAdded support for min and max in nonlinear expressions (#2023)\nAdded support for Indicator{EqualTo{T}} constraints in FileFormats.MPS (#2022)\nAdded default fallback for write_to_file and read_from_file (#2029)","category":"page"},{"location":"moi/changelog/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed Constraint.ZeroOneBridge by adding new bounds as affine constraints instead of variable bounds (#1879)\nFixed reading free rows in FileFormats.MPS files (#2009)\nFixed parsing of OBJSENSE blocks in FileFormats.MPS files (#2016) (#2019)\nFixed the parsing of deeply nested nonlinear expressions by removing the use of recursion (#2020)\nFixed the requirements check in Test.test_constrainnt_get_ConstraintIndex (#2024)","category":"page"},{"location":"moi/changelog/#v1.8.2-(September-20,-2022)","page":"Release notes","title":"v1.8.2 (September 20, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Documentation","page":"Release notes","title":"Documentation","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added vale as a documentation linter (#2002)\nImproved styling of code blocks in the PDF (#1999) (#2000)\nFixed a number of typos in the documentation (#2001) (#2003)","category":"page"},{"location":"moi/changelog/#v1.8.1-(September-12,-2022)","page":"Release notes","title":"v1.8.1 (September 12, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in supports(::AbstractBridgeOptimizer for constraint attributes (#1991) (#1992)","category":"page"},{"location":"moi/changelog/#v1.8.0-(September-1,-2022)","page":"Release notes","title":"v1.8.0 (September 1, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new sets\nHyperRectangle (#1961)\nReified (#1955)\nAdded new bridges (#1955)\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.SplitHyperRectangleBridge\nAdded support for atan(y, x) in Nonlinear (#1987)","category":"page"},{"location":"moi/changelog/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Lazily construct expressions in Nonlinear so that expressions are updated when Nonlinear.Parameter values are updated (#1984)\nAllow NORM_LIMIT as a TerminationStatus for unbounded problems in Test (#1990)","category":"page"},{"location":"moi/changelog/#v1.7.0-(August-16,-2022)","page":"Release notes","title":"v1.7.0 (August 16, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new sets\nHermitianPositiveSemidefiniteConeTriangle\nAdded new optimizer-independent options\nRelativeGapTolerance\nAbsoluteGapTolerance\nAdded new bridges\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Objective.QuadratizeBridge\nAdded support for generic number types in Utilities.loadfromstring!\nUpdated FileFormats.MOF to MathOptFormat v1.1, enabling support for constraint programming sets in the MOF file format\nAdded support in various FileFormats for\nindicator constraints in FileFormats.MPS\nquadratic constraints and an objective in FileFormats.LP\nquadratic constraints and an objective in FileFormats.MPS","category":"page"},{"location":"moi/changelog/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed some missing promotion rules","category":"page"},{"location":"moi/changelog/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved the performance of Jacobian products in Nonlinear\nRemoved an un-needed copy in Utilities.modify_function\nVarious clean-ups in Bridges/bridge_optimizer.jl","category":"page"},{"location":"moi/changelog/#v1.6.1-(July-23,-2022)","page":"Release notes","title":"v1.6.1 (July 23, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for ExponentialCone in MatrixOfConstraints\nFix PSDSquare_3 test to reflect a previously fixed bug getting the ConstraintDual of a PositiveSemidefiniteConeSquare constraint","category":"page"},{"location":"moi/changelog/#v1.6.0-(July-2,-2022)","page":"Release notes","title":"v1.6.0 (July 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.needs_final_touch and Bridges.final_touch\nAdded new bridges from constraint programming sets to mixed-integer linear programs:\nAllDifferentToCountDistinctBridge\nCountAtLeastToCountBelongsBridge\nCountBelongsToMILPBridge\nCountDistinctToMILPBridge\nCountGreaterThanToMILPBridge\nCircuitToMILPBridge","category":"page"},{"location":"moi/changelog/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Relax an instance of ::Vector to ::AbstractVector in MOI.Nonlinear\nFix BinPackingToMILPBridge to respect variable bounds\nFix SemiToBinaryBridge to throw error if other bounds are set","category":"page"},{"location":"moi/changelog/#v1.5.0-(June-27,-2022)","page":"Release notes","title":"v1.5.0 (June 27, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added GetAttributeNotAllowed for solvers to indicate when getting an attribute encounters an error\nAdded Utilities.get_fallback support for ObjectiveValue and DualObjectiveValue\nAdded new bridges:\nRootDetConeSquare to RootDetConeTriangle\nLogDetConeSquare to LogDetConeTriangle\nBinPacking to a mixed-integer linear program\nTable to a mixed-integer linear program\nAdded Bridges.print_active_bridges to display the current optimal hyper-path in a Bridges.LazyBridgeOptimizer","category":"page"},{"location":"moi/changelog/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ZeroOne tests with lower and upper bounds\nFixed error in FileFormats.LP when reading a malformed file\nFixed reading of nonlinear programs in FileFormats.MOF\nFixed bug in ConstraintDual when using SquareBridge","category":"page"},{"location":"moi/changelog/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved documentation of nonlinear API\nDocumented duality convention for PositiveSemidefiniteConeSquare sets\nFixed typo in Bridges.Constraint.QuadToSOCBridge docstring","category":"page"},{"location":"moi/changelog/#v1.4.0-(June-9,-2022)","page":"Release notes","title":"v1.4.0 (June 9, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a number of sets for constraint programming:\nAllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nTable\nAdded support for user-defined hessians in Nonlinear\nAdded Bridges.runtests to simplify the testing of bridge implementations","category":"page"},{"location":"moi/changelog/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in FileFormats.NL when writing univariate *","category":"page"},{"location":"moi/changelog/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Began a large refactoring of the Bridges submodule, with greatly improved documentation.","category":"page"},{"location":"moi/changelog/#v1.3.0-(May-27,-2022)","page":"Release notes","title":"v1.3.0 (May 27, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add MOI.Nonlinear submodule. This is a large new submodule that has been refactored from code that was in JuMP. For now, it should be considered experimental.\nAdd FileFormats.NL.SolFileResults(::IO, ::Model)\nAdd FileFormats.NL.read!(::IO, ::Model)\nAdd MOI.modify that accepts a vector of modifications","category":"page"},{"location":"moi/changelog/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Test which attempted to include non-.jl files\nFixed a bug in FileFormats for models with open interval constraints","category":"page"},{"location":"moi/changelog/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in Utilities.DoubleDict\nVarious minor improvements to the documentation","category":"page"},{"location":"moi/changelog/#v1.2.0-(April-25,-2022)","page":"Release notes","title":"v1.2.0 (April 25, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add support for the FORMAT_REW/.rew file format in FileFormats.","category":"page"},{"location":"moi/changelog/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix bug handling of default variable bounds in FileFormats.LP\nFix FileFormats.MPS to not write OBJSENSE by default since this is only supported by some readers.","category":"page"},{"location":"moi/changelog/#v1.1.2-(March-31,-2022)","page":"Release notes","title":"v1.1.2 (March 31, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix a range of bugs in FileFormats.LP\nFix reading of problem dimensions in FileFormats.SDPA","category":"page"},{"location":"moi/changelog/#v1.1.1-(March-23,-2022)","page":"Release notes","title":"v1.1.1 (March 23, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in test_model_UpperBoundAlreadySet\nFix bug in test_infeasible_ tests\nFix bug in test_objective_ObjectiveFunction_blank\nRelax restriction of MOI.AbstractOptimizer to MOI.ModelLike in Utilities.CachingOptimizer and instantiate.","category":"page"},{"location":"moi/changelog/#New-tests","page":"Release notes","title":"New tests","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add test_conic_empty_matrix that checks conic solvers support problems with no variables.","category":"page"},{"location":"moi/changelog/#v1.1.0-(March-2,-2022)","page":"Release notes","title":"v1.1.0 (March 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.throw_unsupported(::UniversalFallback) for simplifying solver wrappers which copy from a UniversalFallback.","category":"page"},{"location":"moi/changelog/#v1.0.2-(March-1,-2022)","page":"Release notes","title":"v1.0.2 (March 1, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in the test_model_ScalarFunctionConstantNotZero test\nFixed the error type when an AbstractFunctionConversionBridge cannot get or set an attribute\nIdentified a correctness bug in RSOCtoPSDBridge. We now thrown an error instead of returning an incorrect result.","category":"page"},{"location":"moi/changelog/#v1.0.1-(February-25,-2022)","page":"Release notes","title":"v1.0.1 (February 25, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in which OptimizerAttributes were not copied in CachingOptimizer\nFixed a bug in which shift_constant did not promote mixed types of coefficients\nFixed a bug in which deleting a constraint of a bridged variable threw ErrorException instead of MOI.DeleteNotAllowed\nFixed a bug in which add_constraint in MatrixOfConstraints did not canonicalize the function\nFixed a bug when modifying scalar constants of a function containing a bridged variable\nFixed a bug in which final_touch was not always called with a CachingOptimizer","category":"page"},{"location":"moi/changelog/#v1.0.0-(February-17,-2022)","page":"Release notes","title":"v1.0.0 (February 17, 2022)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Although tagged as a breaking release, v1.0.0 is v0.10.9 with deprecations removed, similar to how Julia 1.0 was Julia 0.7 with deprecations removed.","category":"page"},{"location":"moi/changelog/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nAll deprecations have been removed","category":"page"},{"location":"moi/changelog/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated features. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Then install MathOptInterface v0.10.9:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add MathOptInterface@0.10\"","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"moi/changelog/#v0.10.9-(February-16,-2022)","page":"Release notes","title":"v0.10.9 (February 16, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.FreeVariables as a new VariablesConstrainer for conic solvers\nAdded MOI.default_cache for specifying the model used in CachingOptimizer","category":"page"},{"location":"moi/changelog/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed LaTeX printing of MOI.Interval sets","category":"page"},{"location":"moi/changelog/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Aqua.jl as a CI check, and fixed suggested issues\nThe constructors of GeoMeanBridge, StructOfConstraints, and CachingOptimizer were changed from outer to inner constructors. This change is technically breaking, but does not impact users who followed the documented API.","category":"page"},{"location":"moi/changelog/#v0.10.8-(February-3,-2022)","page":"Release notes","title":"v0.10.8 (February 3, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a Base.read! for FileFormats.LP.","category":"page"},{"location":"moi/changelog/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in MutableSparseMatrix\nFixed a bug when calling operate!(vcat, ...) with Number arguments\nRemoved unintended export of deprecated symbols\nFixed a bug with PowerCone and DualPowerCone in MatrixOfConstraints.","category":"page"},{"location":"moi/changelog/#v0.10.7-(January-5,-2022)","page":"Release notes","title":"v0.10.7 (January 5, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added test for modifying the constant vector in a VectorAffineFunction-in-Zeros constraint.","category":"page"},{"location":"moi/changelog/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed the order in which sets are added to a LazyBridgeOptimizer. Compared to v0.10.6, this may result in bridged models being created with a different number (and order) of variables and constraints. However, it was necessary to fix cases which were previously rejected as unsupported, even though there was a valid bridge transformation.\nFixed an error message in FileFormats.CBF\nFixed comparison in test_linear_integration_Interval\nFixed errors for ConstraintPrimal in a CachingOptimizer\nFixed printing of models with non-Float64 coefficients.","category":"page"},{"location":"moi/changelog/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various improvements to reduce time-to-first-solve latency\nImproved error message when an optimizer does not support compute_conflict!","category":"page"},{"location":"moi/changelog/#v0.10.6-(November-30,-2021)","page":"Release notes","title":"v0.10.6 (November 30, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new documentation and tests for infeasibility certificates\nAdded a version control system for the tests in MOI.Test.runtests. Pass exclude_tests_after = v\"0.10.5\" to run tests added in v0.10.5 and earlier.\nMOI.Test.runtests now supports generic number types. To specify the number type T, pass MOI.Test.Config(T).\nAdded infeasible_status to MOI.Test.Config for solvers which return LOCALLY_INFEASIBLE\nCachingOptimizers now use a fallback for ConstraintPrimal. This should enable solvers using a CachingOptimizer to pass tests requiring ConstraintPrimal.","category":"page"},{"location":"moi/changelog/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a StackOverflow bug in copy_to\nFixed error thrown when nonconvex quadratic constraints cannot be bridged\nFixed a bug in copy_to for FileFormats.NL.Model\nFixed a bug in FileFormats.NL when printing large integers\nRemove a common test failure for LowerBoundAlreadySet tests\nUtilities.num_rows is now exported\nRemove parts of failing test_model_copy_to_xxx tests due to bridges","category":"page"},{"location":"moi/changelog/#v0.10.5-(November-7,-2021)","page":"Release notes","title":"v0.10.5 (November 7, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed getter in UniversalFallback\nFixed test_solve_conflict_zeroone_ii","category":"page"},{"location":"moi/changelog/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Make normalize_and_add_constraint more flexible\nUpdate paper BibTeX","category":"page"},{"location":"moi/changelog/#v0.10.4-(October-26,-2021)","page":"Release notes","title":"v0.10.4 (October 26, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add SolverVersion attribute\nAdd new tests:\ntest_solve_conflict_zeroone_ii\ntest_nonlinear_objective\nUtilities.VariablesContainer now supports ConstraintFunction and ConstraintSet\nThe documentation is now available as a PDF","category":"page"},{"location":"moi/changelog/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Update to MutableArithmetics 0.3\nVarious improvements to the documentation","category":"page"},{"location":"moi/changelog/#v0.10.3-(September-18,-2021)","page":"Release notes","title":"v0.10.3 (September 18, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-39","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug which prevented callbacks from working through a CachingOptimizer\nFixed bug in Test submodule","category":"page"},{"location":"moi/changelog/#v0.10.2-(September-16,-2021)","page":"Release notes","title":"v0.10.2 (September 16, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Updated MathOptFormat to v1.0\nUpdated JSONSchema to v1.0\nAdded Utilities.set_with_dimension\nAdded two-argument optimize!(::AbstractOptimizer, ::ModelLike)\nThe experimental feature copy_to_and_optimize! has been removed\nDet bridges now support getting ConstraintFunction and ConstraintSet\nVarious minor bug fixes identified by improved testing","category":"page"},{"location":"moi/changelog/#v0.10.1-(September-8,-2021)","page":"Release notes","title":"v0.10.1 (September 8, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various fixes to MOI.Test","category":"page"},{"location":"moi/changelog/#v0.10.0-(September-6,-2021)","page":"Release notes","title":"v0.10.0 (September 6, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"MOI v0.10 is a significant breaking release. There are a large number of user-visible breaking changes and code refactors, as well as a substantial number of new features.","category":"page"},{"location":"moi/changelog/#Breaking-in-MOI","page":"Release notes","title":"Breaking in MOI","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"SingleVariable has been removed; use VariableIndex instead\nSingleVariableConstraintNameError has been renamed to VariableIndexConstraintNameError\nSettingSingleVariableFunctionNotAllowed has been renamed to SettingVariableIndexFunctionNotAllowed\nVariableIndex constraints should not support ConstraintName\nVariableIndex constraints should not support ConstraintBasisStatus; implement VariableBasisStatus instead\nListOfConstraints has been renamed to ListOfConstraintTypesPresent\nListOfConstraintTypesPresent should now return Tuple{Type,Type} instead of Tuple{DataType,DataType}\nSolveTime has been renamed to SolveTimeSec\nIndicatorSet has been renamed to Indicator\nRawParameter has been renamed to RawOptimizerAttribute and now takes String instead of Any as the only argument\nThe .N field in result attributes has been renamed to .result_index\nThe .variable_index field in ScalarAffineTerm has been renamed to .variable\nThe .variable_index_1 field in ScalarQuadraticTerm has been renamed to .variable_1\nThe .variable_index_2 field in ScalarQuadraticTerm has been renamed to .variable_2\nThe order of affine_terms and quadratic_terms in ScalarQuadraticFunction and VectorQuadraticFunction have been reversed. Both functions now accept quadratic, affine, and constant terms in that order.\nThe index_value function has been removed. Use .value instead.\nisapprox has been removed for SOS1 and SOS2.\nThe dimension argument to Complements(dimension::Int) should now be the length of the corresponding function, instead of half the length. An ArgumentError is thrown if dimension is not even.\ncopy_to no longer takes keyword arguments:\ncopy_names: now copy names if they are supported by the destination solver\nfilter_constraints: use Utilities.ModelFilter instead\nwarn_attributes: never warn about optimizer attributes","category":"page"},{"location":"moi/changelog/#Breaking-in-Bridges","page":"Release notes","title":"Breaking in Bridges","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Constraint.RSOCBridge has been renamed to Constraint.RSOCtoSOCBridge\nConstraint.SOCRBridge has been renamed to Constraint.SOCtoRSOCBridge\nBridges now return vectors that can be modified by the user. Previously, some bridges returned views instead of copies.\nBridges.IndexInVector has been unified into a single type. Previously, there was a different type for each submodule within Bridges\nThe signature of indicator bridges has been fixed. Use MOI.Bridges.Constraint.IndicatortoSOS1{Float64}(model).","category":"page"},{"location":"moi/changelog/#Breaking-in-FileFormats","page":"Release notes","title":"Breaking in FileFormats","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"FileFormats.MOF.Model no longer accepts validate argument. Use the JSONSchema package to validate the MOF file. See the documentation for more information.","category":"page"},{"location":"moi/changelog/#Breaking-in-Utilities","page":"Release notes","title":"Breaking in Utilities","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The datastructure of Utilities.Model (and models created with Utilities.@model) has been significantly refactored in a breaking way. This includes the way that objective functions and variable-related information is stored.\nUtilities.supports_default_copy has been renamed to supports_incremental_interface\nUtilities.automatic_copy_to has been renamed to Utilities.default_copy_to\nThe allocate-load API has been removed\nCachingOptimizers are now initialized as EMPTY_OPTIMIZER instead of ATTACHED_OPTIMIZER. If your code relies on the optimizer being attached, call MOIU.attach_optimizer(model) after creation.\nThe field names of Utilities.IndexMap have been renamed to var_map and con_map. Accessing these fields directly is considered a private detail that may change. Use the public getindex and setindex! API instead.\nThe size argument to Utilities.CleverDicts.CleverDict(::Integer) has been removed.\nThe size argument to Utilities.IndexMap(::Integer) has been removed.\nUtilities.DoubleDicts have been significantly refactored. Consult the source code for details.\nUtilities.test_models_equal has been moved to MOI.Test","category":"page"},{"location":"moi/changelog/#Breaking-in-Test","page":"Release notes","title":"Breaking in Test","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"MOI.Test has been renamed to MOI.DeprecatedTest\nAn entirely new MOI.Test submodule has been written. See the documentation for details. The new MOI.Test submodule may find many bugs in the implementations of existing solvers that were previously untested.","category":"page"},{"location":"moi/changelog/#Other-changes:","page":"Release notes","title":"Other changes:","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"attribute_value_type has been added\ncopy_to_and_optimize! has been added\nVariableBasisStatus has been added\nprint(model) now prints a human-readable description of the model\nVarious improvements to the FileFormats submodule\nFileFormats.CBF was refactored and received bugfixes\nSupport for MathOptFormat v0.6 was added in FileFormats.MOF\nFileFormats.MPS has had bugfixes and support for more features such as OBJSENSE and objective constants.\nFileFormats.NL has been added to support nonlinear files\nImproved type inference throughout to reduce latency","category":"page"},{"location":"moi/changelog/#Updating","page":"Release notes","title":"Updating","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"A helpful script when updating is:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"for (root, dirs, files) in walkdir(\".\")\n    for file in files\n        if !endswith(file, \".jl\")\n            continue\n        end\n        path = joinpath(root, file)\n        s = read(path, String)\n        for pair in [\n            \".variable_index\" => \".variable\",\n            \"RawParameter\" => \"RawOptimizerAttribute\",\n            \"ListOfConstraints\" => \"ListOfConstraintTypesPresent\",\n            \"TestConfig\" => \"Config\",\n            \"attr.N\" => \"attr.result_index\",\n            \"SolveTime\" => \"SolveTimeSec\",\n            \"DataType\" => \"Type\",\n            \"Utilities.supports_default_copy_to\" =>\n                \"supports_incremental_interface\",\n            \"SingleVariableConstraintNameError\" =>\n                \"VariableIndexConstraintNameError\",\n            \"SettingSingleVariableFunctionNotAllowed\" =>\n                \"SettingVariableIndexFunctionNotAllowed\",\n            \"automatic_copy_to\" => \"default_copy_to\",\n        ]\n            s = replace(s, pair)\n        end\n        write(path, s)\n    end\nend","category":"page"},{"location":"moi/changelog/#v0.9.22-(May-22,-2021)","page":"Release notes","title":"v0.9.22 (May 22, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"This release contains backports from the ongoing development of the v0.10 release.","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved type inference in Utilities, Bridges and FileFormats submodules to reduce latency.\nImproved performance of Utilities.is_canonical.\nFixed Utilities.pass_nonvariable_constraints with bridged variables.\nFixed performance regression of Utilities.Model.\nFixed ordering of objective setting in parser.","category":"page"},{"location":"moi/changelog/#v0.9.21-(April-23,-2021)","page":"Release notes","title":"v0.9.21 (April 23, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added supports_shift_constant.\nImprove performance of bridging quadratic constraints.\nAdd precompilation statements.\nLarge improvements to the documentation.\nFix a variety of inference issues, benefiting precompilation and reducing initial latency.\nRawParameters are now ignored when resetting a CachingOptimizer. Previously, changing the underlying optimizer after RawParameters were set would throw an error.\nUtilities.AbstractModel is being refactored. This may break users interacting with private fields of a model generated using @model.","category":"page"},{"location":"moi/changelog/#v0.9.20-(February-20,-2021)","page":"Release notes","title":"v0.9.20 (February 20, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved performance of Utilities.ScalarFunctionIterator\nAdded support for compute_conflict to MOI layers\nAdded test with zero off-diagonal quadratic term in objective\nFixed double deletion of nested bridged SingleVariable/VectorOfVariables constraints\nFixed modification of un-set objective\nFixed function modification with duplicate terms\nMade unit tests abort without failing if the problem class is not supported\nFormatted code with JuliaFormatter\nClarified BasisStatusCode's docstring","category":"page"},{"location":"moi/changelog/#v0.9.19-(December-1,-2020)","page":"Release notes","title":"v0.9.19 (December 1, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added CallbackNodeStatus attribute\nAdded bridge from GreaterThan or LessThan to Interval\nAdded tests for infeasibility certificates and double optimize\nFixed support for Julia v1.6\nRe-organized MOI docs and added documentation for adding a test","category":"page"},{"location":"moi/changelog/#v0.9.18-(November-3,-2020)","page":"Release notes","title":"v0.9.18 (November 3, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various improvements for working with complex numbers\nAdded GeoMeantoRelEntrBridge to bridge a GeometricMeanCone constraint to a relative entropy constraint","category":"page"},{"location":"moi/changelog/#v0.9.17-(September-21,-2020)","page":"Release notes","title":"v0.9.17 (September 21, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed CleverDict with variable of negative index value\nImplement supports_add_constrained_variable for MockOptimizer","category":"page"},{"location":"moi/changelog/#v0.9.16-(September-17,-2020)","page":"Release notes","title":"v0.9.16 (September 17, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various fixes:\n32-bit support\nCleverDict with abstract value type\nChecks in test suite","category":"page"},{"location":"moi/changelog/#v0.9.15-(September-14,-2020)","page":"Release notes","title":"v0.9.15 (September 14, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Bridges improvements:\n(R)SOCtoNonConvexQuad bridge\nZeroOne bridge\nUse supports_add_constrained_variable in LazyBridgeOptimizer\nExposed VariableBridgeCost and ConstraintBridgeCost attributes\nPrioritize constraining variables on creation according to these costs\nRefactor bridge debugging\nLarge performance improvements across all submodules\nLots of documentation improvements\nFileFormats improvements:\nUpdate MathOptFormat to v0.5\nFix supported objectives in FileFormats\nTesting improvements:\nAdd name option for basic_constraint_test\nBug fixes and missing methods\nAdd length for iterators\nFix bug with duplicate terms\nFix order of LinearOfConstraintIndices","category":"page"},{"location":"moi/changelog/#v0.9.14-(May-30,-2020)","page":"Release notes","title":"v0.9.14 (May 30, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).\nBump JSONSchema dependency from v0.2 to v0.3 (#1090).\nDocumentation improvements:\nFix typos (#1054, #1060, #1061, #1064, #1069, #1070).\nRemove the outdated recommendation for a package implementing MOI for a solver XXX to be called MathOptInterfaceXXX (#1087).\nUtilities improvements:\nFix is_canonical for quadratic functions (#1081, #1089).\nImplement add_constrained_variable[s] for CachingOptimizer so that it is added as constrained variables to the underlying optimizer (#1084).\nAdd support for custom objective functions for UniversalFallback (#1086).\nDeterministic ordering of constraints in UniversalFallback (#1088).\nTesting improvements:\nAdd NormOneCone/NormInfinityCone tests (#1045).\nBridges improvements:\nAdd bridges from Semiinteger and Semicontinuous (#1059).\nImplement getting ConstraintSet for Variable.FlipSignBridge (#1066).\nFix setting ConstraintFunction for Constraint.ScalarizeBridge (#1093).\nFix NormOne/NormInf bridges with nonzero constants (#1045).\nFix StackOverflow in debug (#1063).\nFileFormats improvements:\n[SDPA] Implement the extension for integer variables (#1079).\n[SDPA] Ignore comments after m and nblocks and detect dat-s extension (#1077).\n[SDPA] No scaling of off-diagonal coefficient (#1076).\n[SDPA] Add missing negation of constant (#1075).","category":"page"},{"location":"moi/changelog/#v0.9.13-(March-24,-2020)","page":"Release notes","title":"v0.9.13 (March 24, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added tests for Semicontinuous and Semiinteger variables (#1033).\nAdded tests for using ExprGraphs from NLP evaluators (#1043).\nUpdate version compatibilities of dependencies (#1034, #1051, #1052).\nFixed typos in documentation (#1044).","category":"page"},{"location":"moi/changelog/#v0.9.12-(February-28,-2020)","page":"Release notes","title":"v0.9.12 (February 28, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed writing NLPBlock in MathOptFormat (#1037).\nFixed MockOptimizer for result attributes with non-one result index (#1039).\nUpdated test template with instantiate (#1032).","category":"page"},{"location":"moi/changelog/#v0.9.11-(February-21,-2020)","page":"Release notes","title":"v0.9.11 (February 21, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add an option for the model created by Utilities.@model to be a subtype of AbstractOptimizer (#1031).\nDescribed dual cone in docstrings of GeoMeanCone and RelativeEntropyCone (#1018, #1028).\nFixed typos in documentation (#1022, #1024).\nFixed warning of unsupported attribute (#1027).\nAdded more rootdet/logdet conic tests (#1026).\nImplemented ConstraintDual for Constraint.GeoMeanBridge, Constraint.RootDetBridge and Constraint.LogDetBridge and test duals in tests with GeoMeanCone and RootDetConeTriangle and LogDetConeTriangle cones (#1025, #1026).","category":"page"},{"location":"moi/changelog/#v0.9.10-(January-31,-2020)","page":"Release notes","title":"v0.9.10 (January 31, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added OptimizerWithAttributes grouping an optimizer constructor and a list of optimizer attributes (#1008).\nAdded RelativeEntropyCone with corresponding bridge into exponential cone constraints (#993).\nAdded NormSpectralCone and NormNuclearCone with corresponding bridges into positive semidefinite constraints (#976).\nAdded supports_constrained_variable(s) (#1004).\nAdded dual_set_type (#1002).\nAdded tests for vector specialized version of delete (#989, #1011).\nAdded PSD3 test (#1007).\nClarified dual solution of Tests.pow1v and Tests.pow1f (#1013).\nAdded support for EqualTo and Zero in Bridges.Constraint.SplitIntervalBridge (#1005).\nFixed Utilities.vectorize for empty vector (#1003).\nFixed free variables in LP writer (#1006).","category":"page"},{"location":"moi/changelog/#v0.9.9-(December-29,-2019)","page":"Release notes","title":"v0.9.9 (December 29, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).\nImproved performance of deletion of vector of variables in MOI.Utilities.Model (#983).\nUpdated to MutableArithmetics v0.2 (#981).\nAdded MutableArithmetics.promote_operation allocation tests (#975).\nFixed inference issue on Julia v1.1 (#982).","category":"page"},{"location":"moi/changelog/#v0.9.8-(December-19,-2019)","page":"Release notes","title":"v0.9.8 (December 19, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented MutableArithmetics API (#924).\nFixed callbacks with CachingOptimizer (#959).\nFixed MOI.dimension for MOI.Complements (#948).\nAdded fallback for add_variables (#972).\nAdded is_diagonal_vectorized_index utility (#965).\nImproved linear constraints display in manual (#963, #964).\nBridges improvements:\nAdded IndicatorSet to SOS1 bridge (#877).\nAdded support for starting values for Variable.VectorizeBridge (#944).\nFixed MOI.add_constraints with non-bridged variable constraint on bridged variable (#951).\nFixed corner cases and docstring of GeoMeanBridge (#961, #962, #966).\nFixed choice between variable or constraint bridges for constrained variables (#973).\nImprove performance of bridge shortest path (#945, #946, #956).\nAdded docstring for test_delete_bridge (#954).\nAdded Variable bridge tests (#952).","category":"page"},{"location":"moi/changelog/#v0.9.7-(October-30,-2019)","page":"Release notes","title":"v0.9.7 (October 30, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented _result_index_field for NLPBlockDual (#934).\nFixed copy of model with starting values for vector constraints (#941).\nBridges improvements:\nImproved performance of add_bridge and added has_bridge (#935).\nAdded AbstractSetMapBridge for bridges between sets S1, S2 such that there is a linear map A such that A*S1 = S2 (#933).\nAdded support for starting values for FlipSignBridge, VectorizeBridge, ScalarizeBridge, SlackBridge, SplitIntervalBridge, RSOCBridge, SOCRBridge NormInfinityBridge, SOCtoPSDBridge and RSOCtoPSDBridge (#933, #936, #937, #938, #939).","category":"page"},{"location":"moi/changelog/#v0.9.6-(October-25,-2019)","page":"Release notes","title":"v0.9.6 (October 25, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added complementarity constraints (#913).\nAllowed ModelLike objects as value of attributes (#928).\nTesting improvements:\nAdded dual_objective_value option to MOI.Test.TestConfig (#922).\nAdded InvalidIndex tests in basic_constraint_tests (#921).\nAdded tests for the constant term in indicator constraint (#929).\nBridges improvements:\nAdded support for starting values for Functionize bridges (#923).\nAdded variable indices context to variable bridges (#920).\nFixed a typo in printing o debug_supports (#927).","category":"page"},{"location":"moi/changelog/#v0.9.5-(October-9,-2019)","page":"Release notes","title":"v0.9.5 (October 9, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Clarified PrimalStatus/DualStatus to be NO_SOLUTION if result_index is out of bounds (#912).\nAdded tolerance for checks  and use ResultCount + 1 for the result_index in MOI.Test.solve_result_status (#910, #917).\nUse 0.5 instead of 2.0 for power in PowerCone in basic_constraint_test (#916).\nBridges improvements:\nAdded debug utilities for unsupported variable/constraint/objective (#861).\nFixed deletion of variables in bridged VectorOfVariables constraints (#909).\nFixed result_index with objective bridges (#911).","category":"page"},{"location":"moi/changelog/#v0.9.4-(October-2,-2019)","page":"Release notes","title":"v0.9.4 (October 2, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added solver-independent MIP callbacks (#782).\nImplements submit for Utilities.CachingOptimizer and Bridges.AbstractBridgeOptimizer (#906).\nAdded tests for result count of solution attributes (#901, #904).\nAdded NumberOfThreads attribute (#892).\nAdded Utilities.get_bounds to get the bounds on a variable (#890).\nAdded a note on duplicate coefficients in documentation (#581).\nAdded result index in ConstraintBasisStatus (#898).\nAdded extension dictionary to Utilities.Model (#884, #895).\nFixed deletion of constrained variables for CachingOptimizer (#905).\nImplemented Utilities.shift_constraint for Test.UnknownScalarSet (#896).\nBridges improvements:\nAdded Variable.RSOCtoSOCBridge (#907).\nImplemented MOI.get for ConstraintFunction/ConstraintSet for Bridges.Constraint.SquareBridge (#899).","category":"page"},{"location":"moi/changelog/#v0.9.3-(September-20,-2019)","page":"Release notes","title":"v0.9.3 (September 20, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ambiguity detected in Julia v1.3 (#891, #893).\nFixed missing sets from ListOfSupportedConstraints (#880).\nFixed copy of VectorOfVariables constraints with duplicate indices (#886).\nAdded extension dictionary to MOIU.Model (#884).\nImplemented MOI.get for function and set for GeoMeanBridge (#888).\nUpdated documentation for SingleVariable indices and bridges (#885).\nTesting improvements:\nAdded more comprehensive tests for names (#882).\nAdded tests for SingleVariable duals (#883).\nAdded tests for DualExponentialCone and DualPowerCone (#873).\nImprovements for arbitrary coefficient type:\nFixed == for sets with mutable fields (#887).\nRemoved some Float64 assumptions in bridges (#878).\nAutomatic selection of Constraint.[Scalar|Vector]FunctionizeBridge (#889).","category":"page"},{"location":"moi/changelog/#v0.9.2-(September-5,-2019)","page":"Release notes","title":"v0.9.2 (September 5, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented model printing for MOI.ModelLike and specialized it for models defined in MOI (864).\nGeneralized contlinear tests for arbitrary coefficient type (#855).\nFixed supports_constraint for Semiinteger and Semicontinuous and supports for ObjectiveFunction (#859).\nFixed Allocate-Load copy for single variable constraints (#856).\nBridges improvements:\nAdd objective bridges (#789).\nFixed Variable.RSOCtoPSDBridge for dimension 2 (#869).\nAdded Variable.SOCtoRSOCBridge (#865).\nAdded Constraint.SOCRBridge and disable MOI.Bridges.Constraint.SOCtoPSDBridge (#751).\nFixed added_constraint_types for Contraint.LogDetBridge and Constraint.RootDetBridge (#870).","category":"page"},{"location":"moi/changelog/#v0.9.1-(August-22,-2019)","page":"Release notes","title":"v0.9.1 (August 22, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix support for Julia v1.2 (#834).\nL1 and L∞ norm epigraph cones and corresponding bridges to LP were added (#818).\nAdded tests to MOI.Test.nametest (#833).\nFix MOI.Test.soc3test for solvers not supporting infeasibility certificates (#839).\nImplements operate for operators * and / between vector function and constant (#837).\nImplements show for MOI.Utilities.IndexMap (#847).\nFix corner cases for mapping of variables in MOI.Utilities.CachingOptimizer and substitution of variables in MOI.Bridges.AbstractBridgeOptimizer (#848).\nFix transformation of constant terms for MOI.Bridges.Constraint.SOCtoPSDBridge and MOI.Bridges.Constraint.RSOCtoPSDBridge (#840).","category":"page"},{"location":"moi/changelog/#v0.9.0-(August-13,-2019)","page":"Release notes","title":"v0.9.0 (August 13, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Support for Julia v0.6 and v0.7 was dropped (#714, #717).\nA MOI.Utilities.Model implementation of ModelLike, this should replace most use cases of MOI.Utilities.@model (#781).\nadd_constrained_variable and add_constrained_variables were added (#759).\nSupport for indicator constraints was added (#709, #712).\nDualObjectiveValue attribute was added (#473).\nRawParameter attribute was added (#733).\nA dual_set function was added (#804).\nA Benchmarks submodule was added to facilitate solver benchmarking (#769).\nA submit function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).\nThe field of ObjectiveValue was renamed to result_index (#729).\nThe _constant and Utilities.getconstant function were renamed to constant\nREDUCTION_CERTIFICATE result status was added (#734).\nAbstract matrix sets were added (#731).\nTesting improvements:\nThe testing guideline was updated (#728).\nQuadratic tests were added (#697).\nUnit tests for RawStatusString, SolveTime, Silent and SolverName were added (#726, #741).\nA rotated second-order cone test was added (#759).\nA power cone test was added (#768).\nTests for ZeroOne variables with variable bounds were added (#772).\nAn unbounded test was added (#773).\nExisting tests had a few updates (#702, #703, #763).\nDocumentation improvements:\nAdded a section on CachingOptimizer (#777).\nAdded a section on UniversalFallback, Model and @model (#762).\nTransition the knapsack example to a doctest with MockOptimizer (#786).\nUtilities improvements:\nA CleverDict utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).\nThe Utilities.constant function was renamed to Utilities.constant_vector (#740).\nImplement optimizer attributes for CachingOptimizer (#745).\nRename Utilities.add_scalar_constraint to Utilities.normalize_and_add_constraint (#801).\noperate with vcat, SingleVariable and VectorOfVariables now returns a VectorOfVariables (#616).\nFix a type piracy of operate (#784).\nThe load_constraint fallback signature was fixed (#760).\nThe set_dot function was extended to work with sparse arrays (#805).\nBridges improvements:\nThe bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement ConstraintFunction and ConstraintSet if the user wants to recover them. As a consequence, the @bridge macro was removed (#722).\nBridge are now instantiated with a bridge_constraint function instead of using a constructor (#730).\nFix constraint attributes for bridges (#699).\nConstraint bridges were moved to the Bridges/Constraint submodule so they should now inherit from MOI.Bridges.Constraint.Abstract and should implement MOI.Bridges.Constraint.concrete_bridge_type instead of MOI.Bridges.concrete_bridge_type (#756).\nVariable bridges were added in (#759).\nVarious improvements (#746, #747).","category":"page"},{"location":"moi/changelog/#v0.8.4-(March-13,-2019)","page":"Release notes","title":"v0.8.4 (March 13, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Performance improvement in default_copy_to and bridge optimizer (#696).\nAdd Silent and implement setting optimizer attributes in caching and mock optimizers (#695).\nAdd Functionize bridges (SingleVariable and VectorOfVariables) (#659).\nMinor typo fixes (#694).","category":"page"},{"location":"moi/changelog/#v0.8.3-(March-6,-2019)","page":"Release notes","title":"v0.8.3 (March 6, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Use zero constant in scalar constraint function of MOI.Test.copytest (#691).\nFix variable deletion with SingleVariable objective function (#690).\nFix LazyBridgeOptimizer with bridges that add no constraints (#689).\nError message improvements (#673, #685, #686, #688).\nDocumentation improvements (#682, #683, #687).\nBasis status:\nRemove VariableBasisStatus (#679).\nTest ConstraintBasisStatus and implement it in bridges (#678).\nFix inference of NumberOfVariables and NumberOfConstraints (#677).\nImplement division between a quadratic function and a number (#675).","category":"page"},{"location":"moi/changelog/#v0.8.2-(February-7,-2019)","page":"Release notes","title":"v0.8.2 (February 7, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add RawStatusString attribute (#629).\nDo not set names to the optimizer but only to the cache in CachingOptimizer (#638).\nMake scalar MOI functions act as scalars in broadcast (#646).\nAdd function utilities:\nImplement Base.zero (#634), Base.iszero (#643), add missing arithmetic operations (#644, #645) and fix division (#648).\nAdd a vectorize function that turns a vector of ScalarAffineFunction into a VectorAffineFunction (#642).\nImprove support for starting values:\nShow a warning in copy when starting values are not supported instead of throwing an error (#630).\nFix UniversalFallback for getting an variable or constraint attribute set to no indices (#623).\nAdd a test in contlineartest with partially set VariablePrimalStart.\nBridges improvements:\nFix StackOverFlow in LazyBridgeOptimizer when there is a cycle in the graph of bridges.\nAdd Slack bridges (#610, #650).\nAdd FlipSign bridges (#658).\nAdd tests with duplicate coefficients in ScalarAffineFunction and VectorAffineFunction (#639).\nUse tolerance to compare VariablePrimal in rotatedsoc1 test (#632).\nUse a zero constant in ScalarAffineFunction of constraints in psdt2 (#622).","category":"page"},{"location":"moi/changelog/#v0.8.1-(January-7,-2019)","page":"Release notes","title":"v0.8.1 (January 7, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Adding an NLP objective now overrides any objective set using the ObjectiveFunction attribute (#619).\nRename fullbridgeoptimizer into full_bridge_optimizer (#621).\nAllow custom constraint types with full_bridge_optimizer (#617).\nAdd Vectorize bridge which transforms scalar linear constraints into vector linear constraints (#615).","category":"page"},{"location":"moi/changelog/#v0.8.0-(December-18,-2018)","page":"Release notes","title":"v0.8.0 (December 18, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Rename all enum values to follow the JuMP naming guidelines for constants, for example, Optimal becomes OPTIMAL, and DualInfeasible becomes DUAL_INFEASIBLE.\nRename CachingOptimizer methods for style compliance.\nAdd an MOI.TerminationStatusCode called ALMOST_DUAL_INFEASIBLE.","category":"page"},{"location":"moi/changelog/#v0.7.0-(December-13,-2018)","page":"Release notes","title":"v0.7.0 (December 13, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Test that MOI.TerminationStatus is MOI.OptimizeNotCalled before MOI.optimize! is called.\nCheck supports_default_copy_to in tests (#594).\nKey pieces of information like optimality, infeasibility, etc., are now reported through TerminationStatusCode. It is typically no longer necessary to check the result statuses in addition to the termination status.\nAdd perspective dimension to log-det cone (#593).","category":"page"},{"location":"moi/changelog/#v0.6.4-(November-27,-2018)","page":"Release notes","title":"v0.6.4 (November 27, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add OptimizeNotCalled termination status (#577) and improve documentation of other statuses (#575).\nAdd a solver naming guideline (#578).\nMake FeasibilitySense the default ObjectiveSense (#579).\nFix Utilities.@model and Bridges.@bridge macros for functions and sets defined outside MOI (#582).\nDocument solver-specific attributes (#580) and implement them in Utilities.CachingOptimizer (#565).","category":"page"},{"location":"moi/changelog/#v0.6.3-(November-16,-2018)","page":"Release notes","title":"v0.6.3 (November 16, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)\nAttributes may now be partially set (some values could be nothing). (#563)\nPerformance improvements in Utilities.Model (#549, #567, #568)\nFix bug in QuadtoSOC (#558).\nNew supports_default_copy_to method that optimizers should implement to control caching behavior.\nDocumentation improvements.","category":"page"},{"location":"moi/changelog/#v0.6.2-(October-26,-2018)","page":"Release notes","title":"v0.6.2 (October 26, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improve hygiene of @model macro (#544).\nFix bug in copy tests (#543).\nFix bug in UniversalFallback attribute getter (#540).\nAllow all correct solutions for solve_blank_obj unit test (#537).\nAdd errors for Allocate-Load and bad constraints (#534).\n[performance] Add specialized implementation of hash for VariableIndex (#533).\n[performance] Construct the name to object dictionaries lazily in model (#535).\nAdd the QuadtoSOC bridge which transforms ScalarQuadraticFunction constraints into RotatedSecondOrderCone (#483).","category":"page"},{"location":"moi/changelog/#v0.6.1-(September-22,-2018)","page":"Release notes","title":"v0.6.1 (September 22, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Enable PositiveSemidefiniteConeSquare set and quadratic functions in MOIB.fullbridgeoptimizer (#524).\nAdd warning in the bridge between PositiveSemidefiniteConeSquare and PositiveSemidefiniteConeTriangle when the matrix is almost symmetric (#522).\nModify MOIT.copytest to not add multiples constraints on the same variable (#521).\nAdd missing keyword argument in one of MOIU.add_scalar_constraint methods (#520).","category":"page"},{"location":"moi/changelog/#v0.6.0-(August-30,-2018)","page":"Release notes","title":"v0.6.0 (August 30, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The MOIU.@model and MOIB.@bridge macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.\nRename functions according to the JuMP style guide:\ncopy! with keyword arguments copynames and warnattributes -> copy_to with keyword arguments copy_names and warn_attributes;\nset! -> set;\naddvariable[s]! -> add_variable[s];\nsupportsconstraint -> supports_constraint;\naddconstraint[s]! -> add_constraint[s];\nisvalid -> is_valid;\nisempty -> is_empty;\nBase.delete! -> delete;\nmodify! -> modify;\ntransform! -> transform;\ninitialize! -> initialize;\nwrite -> write_to_file; and\nread! -> read_from_file.\nRemove free! (use Base.finalize instead).\nAdd the SquarePSD bridge which transforms PositiveSemidefiniteConeTriangle constraints into PositiveSemidefiniteConeTriangle.\nAdd result fallback for ConstraintDual of variable-wise constraint, ConstraintPrimal and ObjectiveValue.\nAdd tests for ObjectiveBound.\nAdd test for empty rows in vector linear constraint.\nRework errors: CannotError has been renamed NotAllowedError and the distinction between UnsupportedError and NotAllowedError is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)\ncanget is removed. NoSolution is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).","category":"page"},{"location":"moi/changelog/#v0.5.0-(August-5,-2018)","page":"Release notes","title":"v0.5.0 (August 5, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix names with CachingOptimizer.\nCleanup thanks to @mohamed82008.\nAdded a universal fallback for constraints.\nFast utilities for function canonicalization thanks to @rdeits.\nRenamed dimension field to side_dimension in the context of matrix-like sets.\nNew and improved tests for cases like duplicate terms and ObjectiveBound.\nRemoved cantransform, canaddconstraint, canaddvariable, canset, canmodify, and candelete functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of UnsupportedError indicate unsupported operations, while subtypes of CannotError indicate operations that cannot be performed in the current state.\nThe API for copy! is updated to remove the CopyResult type.\nUpdates for the new JuMP style guide.","category":"page"},{"location":"moi/changelog/#v0.4.1-(June-28,-2018)","page":"Release notes","title":"v0.4.1 (June 28, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixes vector function modification on 32 bits.\nFixes Bellman-Ford algorithm for bridges.\nAdded an NLP test with FeasibilitySense.\nUpdate modification documentation.","category":"page"},{"location":"moi/changelog/#v0.4.0-(June-23,-2018)","page":"Release notes","title":"v0.4.0 (June 23, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Helper constructors for VectorAffineTerm and VectorQuadraticTerm.\nAdded modify_lhs to TestConfig.\nAdditional unit tests for optimizers.\nAdded a type parameter to CachingOptimizer for the optimizer field.\nNew API for problem modification (#388)\nTests pass without deprecation warnings on Julia 0.7.\nSmall fixes and documentation updates.","category":"page"},{"location":"moi/changelog/#v0.3.0-(May-25,-2018)","page":"Release notes","title":"v0.3.0 (May 25, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.\nImprovements to MockOptimizer.\nSignificant changes to Bridges.\nNew and improved unit tests.\nFixes for Julia 0.7.","category":"page"},{"location":"moi/changelog/#v0.2.0-(April-24,-2018)","page":"Release notes","title":"v0.2.0 (April 24, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improvements to and better coverage of Tests.\nDocumentation fixes.\nSolverName attribute.\nChanges to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).\nAddition of NLP tests.\nIntroduction of UniversalFallback.\ncopynames keyword argument to MOI.copy!.\nAdd Bridges submodule.","category":"page"},{"location":"moi/changelog/#v0.1.0-(February-28,-2018)","page":"Release notes","title":"v0.1.0 (February 28, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Initial public release.\nThe framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/overview.md\"","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/overview/#The-Bridges-submodule","page":"Overview","title":"The Bridges submodule","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The Bridges module simplifies the process of converting models between equivalent formulations.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nRead our paper for more details on how bridges are implemented.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Why-bridges?","page":"Overview","title":"Why bridges?","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"A constraint can often be written in a number of equivalent formulations. For example, the constraint l le a^top x le u (ScalarAffineFunction-in-Interval) could be re-formulated as two constraints: a^top x ge l (ScalarAffineFunction-in-GreaterThan) and a^top x le u (ScalarAffineFunction-in-LessThan). An alternative re-formulation is to add a dummy variable y with the constraints l le y le u (VariableIndex-in-Interval) and a^top x - y = 0 (ScalarAffineFunction-in-EqualTo).","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"To avoid each solver having to code these transformations manually, MathOptInterface provides bridges.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"A bridge is a small transformation from one constraint type to another (potentially collection of) constraint type.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Because these bridges are included in MathOptInterface, they can be re-used by any optimizer. Some bridges also implement constraint modifications and constraint primal and dual translations.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Several bridges can be used in combination to transform a single constraint into a form that the solver may understand. Choosing the bridges to use takes the form of finding a shortest path in the hyper-graph of bridges. The methodology is detailed in the MOI paper.","category":"page"},{"location":"moi/submodules/Bridges/overview/#The-three-types-of-bridges","page":"Overview","title":"The three types of bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"There are three types of bridges in MathOptInterface:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Constraint bridges\nVariable bridges\nObjective bridges","category":"page"},{"location":"moi/submodules/Bridges/overview/#Constraint-bridges","page":"Overview","title":"Constraint bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Constraint bridges convert constraints formulated by the user into an equivalent form supported by the solver. Constraint bridges are subtypes of Bridges.Constraint.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"In particular, constraint bridges can focus on rewriting the function of a constraint, and do not change the set. Function bridges are subtypes of Bridges.Constraint.AbstractFunctionConversionBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented constraint bridges for more details on the types of transformations that are available. Function bridges are Bridges.Constraint.ScalarFunctionizeBridge and Bridges.Constraint.VectorFunctionizeBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/#variable_bridges","page":"Overview","title":"Variable bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Variable bridges convert variables added by the user, either free with add_variable/add_variables, or constrained with add_constrained_variable/add_constrained_variables, into an equivalent form supported by the solver. Variable bridges are subtypes of Bridges.Variable.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented variable bridges for more details on the types of transformations that are available.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Objective-bridges","page":"Overview","title":"Objective bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Objective bridges convert the ObjectiveFunction set by the user into an equivalent form supported by the solver. Objective bridges are subtypes of Bridges.Objective.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented objective bridges for more details on the types of transformations that are available.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Bridges.full_bridge_optimizer","page":"Overview","title":"Bridges.full_bridge_optimizer","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nUnless you have an advanced use-case, this is probably the only function you need to care about.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"To enable the full power of MathOptInterface's bridges, wrap an optimizer in a Bridges.full_bridge_optimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.full_bridge_optimizer(inner_optimizer, Float64)\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Now, use optimizer as normal, and bridging will happen lazily behind the scenes. By lazily, we mean that bridging will happen if and only if the constraint is not supported by the inner_optimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"info: Info\nMost bridges are added by default in Bridges.full_bridge_optimizer. However, for technical reasons, some bridges are not added by default. Three examples include Bridges.Constraint.SOCtoPSDBridge, Bridges.Constraint.SOCtoNonConvexQuadBridge and Bridges.Constraint.RSOCtoNonConvexQuadBridge. See the docs of those bridges for more information.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Add-a-single-bridge","page":"Overview","title":"Add a single bridge","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"If you don't want to use Bridges.full_bridge_optimizer, you can wrap an optimizer in a single bridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"However, this will force the constraint to be bridged, even if the inner_optimizer supports it.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.Constraint.SplitInterval{Float64}(inner_optimizer)\nMOIB.Constraint.SingleBridgeOptimizer{MOIB.Constraint.SplitIntervalBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\njulia> x = MOI.add_variable(optimizer)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(optimizer, x, MOI.Interval(0.0, 1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\njulia> MOI.get(optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})\n\njulia> MOI.get(inner_optimizer, MOI.ListOfConstraintTypesPresent())\n2-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64})\n (MathOptInterface.VariableIndex, MathOptInterface.LessThan{Float64})","category":"page"},{"location":"moi/submodules/Bridges/overview/#Bridges.LazyBridgeOptimizer","page":"Overview","title":"Bridges.LazyBridgeOptimizer","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"If you don't want to use Bridges.full_bridge_optimizer, but you need more than a single bridge (or you want the bridging to happen lazily), you can manually construct a Bridges.LazyBridgeOptimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"First, wrap an inner optimizer:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.LazyBridgeOptimizer(inner_optimizer)\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Then use Bridges.add_bridge to add individual bridges:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Bridges.add_bridge(optimizer, MOI.Bridges.Constraint.SplitIntervalBridge{Float64})\n\njulia> MOI.Bridges.add_bridge(optimizer, MOI.Bridges.Objective.FunctionizeBridge{Float64})","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Now the constraints will be bridged only if needed:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.add_variable(optimizer)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(optimizer, x, MOI.Interval(0.0, 1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\njulia> MOI.get(optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})\n\njulia> MOI.get(inner_optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"EditURL = \"getting_started_with_data_and_plotting.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Getting-started-with-data-and-plotting","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"In this tutorial we will learn how to read tabular data into Julia, and some of the basics of plotting.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"If you're new to Julia, start by reading Getting started with Julia and Getting started with JuMP first.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"note: Note\nThere are multiple ways to read the same kind of data into Julia. This tutorial focuses on DataFrames.jl because it provides the ecosystem to work with most of the required file types in a straightforward manner.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Before we get started, we need this constant to point to where the data files are.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import JuMP\nconst DATA_DIR = joinpath(\n    dirname(pathof(JuMP)),\n    joinpath(\"..\", \"docs\", \"src\", \"tutorials\", \"getting_started\", \"data\"),\n);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Where-to-get-help","page":"Getting started with data and plotting","title":"Where to get help","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Read the documentation","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.jl: http://docs.juliaplots.org/latest/\nCSV.jl: http://csv.juliadata.org/stable\nDataFrames.jl: https://dataframes.juliadata.org/stable/","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Preliminaries","page":"Getting started with data and plotting","title":"Preliminaries","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To get started, we need to install some packages.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#DataFrames.jl","page":"Getting started with data and plotting","title":"DataFrames.jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The DataFrames package provides a set of tools for working with tabular data. It is available through the Julia package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using Pkg\nPkg.add(\"DataFrames\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import DataFrames","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"info: What is a DataFrame?\nA DataFrame is a data structure like a table or spreadsheet. You can use it for storing and exploring a set of related data values. Think of it as a smarter array for holding tabular data.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Plots.jl","page":"Getting started with data and plotting","title":"Plots.jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The Plots package provides a set of tools for plotting. It is available through the Julia package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using Pkg\nPkg.add(\"Plots\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import Plots","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#CSV-.jl","page":"Getting started with data and plotting","title":"CSV .jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"CSV and other delimited text files can be read by the CSV.jl package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Pkg.add(\"CSV\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import CSV","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#DataFrame-basics","page":"Getting started with data and plotting","title":"DataFrame basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To read a CSV file into a DataFrame, we use the CSV.read function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df = CSV.read(joinpath(DATA_DIR, \"StarWars.csv\"), DataFrames.DataFrame)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's try plotting some of this data","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.scatter(\n    csv_df.Weight,\n    csv_df.Height;\n    xlabel = \"Weight\",\n    ylabel = \"Height\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"That doesn't look right. What happened? If you look at the dataframe above, it read Weight in as a String column because there are \"NA\" fields. Let's correct that, by telling CSV to consider \"NA\" as missing.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df = CSV.read(\n    joinpath(DATA_DIR, \"StarWars.csv\"),\n    DataFrames.DataFrame;\n    missingstring = \"NA\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then let's re-plot our data","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.scatter(\n    csv_df.Weight,\n    csv_df.Height;\n    title = \"Height vs Weight of StarWars characters\",\n    xlabel = \"Weight\",\n    ylabel = \"Height\",\n    label = false,\n    ylims = (0, 3),\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"That looks better.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"tip: Tip\nRead the CSV documentation for other parsing options.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.jl supports manipulation using functions similar to pandas. For example, split the dataframe into groups based on eye-color:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"by_eyecolor = DataFrames.groupby(csv_df, :Eyecolor)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then recombine into a single dataframe based on a function operating over the split dataframes:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"eyecolor_count = DataFrames.combine(by_eyecolor) do df\n    return DataFrames.nrow(df)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can rename columns:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.rename!(eyecolor_count, :x1 => :count)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Drop some missing rows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.dropmissing!(eyecolor_count, :Eyecolor)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then we can visualize the data:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"sort!(eyecolor_count, :count; rev = true)\nPlots.bar(\n    eyecolor_count.Eyecolor,\n    eyecolor_count.count;\n    xlabel = \"Eye color\",\n    ylabel = \"Number of characters\",\n    label = false,\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Other-Delimited-Files","page":"Getting started with data and plotting","title":"Other Delimited Files","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also use the CSV.jl package to read any other delimited text file format.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"By default, CSV.File will try to detect a file's delimiter from the first 10 lines of the file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Candidate delimiters include ',', '\\t', ' ', '|', ';', and ':'. If it can't auto-detect the delimiter, it will assume ','.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's take the example of space separated data.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"ss_df = CSV.read(joinpath(DATA_DIR, \"Cereal.txt\"), DataFrames.DataFrame)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also specify the delimiter as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"delim_df = CSV.read(\n    joinpath(DATA_DIR, \"Soccer.txt\"),\n    DataFrames.DataFrame;\n    delim = \"::\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Working-with-DataFrames","page":"Getting started with data and plotting","title":"Working with DataFrames","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Now that we have read the required data into a DataFrame, let us look at some basic operations we can perform on it.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Querying-Basic-Information","page":"Getting started with data and plotting","title":"Querying Basic Information","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The size function gets us the dimensions of the DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.size(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also use the nrow and ncol functions to get the number of rows and columns respectively:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.nrow(ss_df), DataFrames.ncol(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The describe function gives basic summary statistics of data in a DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.describe(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Names of every column can be obtained by the names function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.names(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Corresponding data types are obtained using the broadcasted eltype function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"eltype.(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Accessing-the-Data","page":"Getting started with data and plotting","title":"Accessing the Data","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Similar to regular arrays, we use numerical indexing to access elements of a DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The following are different ways to access a column:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[!, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[!, :Name]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df.Name","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[:, 1] # Note that this creates a copy.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The following are different ways to access a row:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1:1, :]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1, :] # This produces a DataFrameRow.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can change the values just as we normally assign values.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Assign a range to scalar:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1:3, :Height] .= 1.83","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Assign a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[4:6, :Height] = [1.8, 1.6, 1.8]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"tip: Tip\nThere are a lot more things which can be done with a DataFrame. Read the docs for more information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"For information on dplyr-type syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Read the DataFrames.jl documentation\nCheck out DataFramesMeta.jl","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Example:-the-passport-problem","page":"Getting started with data and plotting","title":"Example: the passport problem","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's now apply what we have learned to solve a real problem.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Data-manipulation","page":"Getting started with data and plotting","title":"Data manipulation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The Passport Index Dataset lists travel visa requirements for 199 countries, in .csv format. Our task is to find the minimum number of passports required to visit all countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"passport_data = CSV.read(\n    joinpath(DATA_DIR, \"passport-index-matrix.csv\"),\n    DataFrames.DataFrame,\n);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"In this dataset, the first column represents a passport (=from) and each remaining column represents a foreign country (=to).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values in each cell are as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"3 = visa-free travel\n2 = eTA is required\n1 = visa can be obtained on arrival\n0 = visa is required\n-1 is for all instances where passport and destination are the same","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Our task is to find out the minimum number of passports needed to visit every country without requiring a visa.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values we are interested in are -1 and 3. Let's modify the dataframe so that the -1 and 3 are 1 (true), and all others are 0 (false):","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"function modifier(x)\n    if x == -1 || x == 3\n        return 1\n    else\n        return 0\n    end\nend\n\nfor country in passport_data.Passport\n    passport_data[!, country] = modifier.(passport_data[!, country])\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values in the cells now represent:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"1 = no visa required for travel\n0 = visa required for travel","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#JuMP-Modeling","page":"Getting started with data and plotting","title":"JuMP Modeling","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To model the problem as a mixed-integer linear program, we need a binary decision variable x_c for each country c. x_c is 1 if we select passport c and 0 otherwise. Our objective is to minimize the sum sum x_c over all countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Since we wish to visit all the countries, for every country, we must own at least one passport that lets us travel to that country visa free. For one destination, this can be mathematically represented as sum_c in C a_cd cdot x_d geq 1, where a is the passport_data dataframe.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Thus, we can represent this problem using the following model:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"beginaligned\nmin  sum_c in C x_c \ntextst  sum_c in C a_cd x_c geq 1  forall d in C \n x_c in 01  forall c in C\nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We'll now solve the problem using JuMP:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"First, create the set of countries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"C = passport_data.Passport","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then, create the model and initialize the decision variables:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[C], Bin)\n@objective(model, Min, sum(x))\n@constraint(model, [d in C], passport_data[!, d]' * x >= 1)\nmodel","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Now optimize:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"optimize!(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can use the solution_summary function to get an overview of the solution:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"solution_summary(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Solution","page":"Getting started with data and plotting","title":"Solution","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's  have a look at the solution in more detail:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"println(\"Minimum number of passports needed: \", objective_value(model))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"println(\"Optimal passports:\")\nfor c in C\n    if value(x[c]) > 0.5\n        println(\" * \", c)\n    end\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We need some passports, like New Zealand and the United States, which have widespread access to a large number of countries. However, we also need passports like North Korea which only have visa-free access to a very limited number of countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"note: Note\nWe use value(x[c]) > 0.5 rather than value(x[c]) == 1 to avoid excluding solutions like x[c] = 0.99999 that are \"1\" to some tolerance.","category":"page"},{"location":"developers/roadmap/#Development-roadmap","page":"Roadmap","title":"Development roadmap","text":"","category":"section"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"The JuMP developers have compiled this roadmap document to share their plans and goals with the JuMP community. Contributions to roadmap issues are especially invited.","category":"page"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"Most of these issues will require changes to both JuMP and MathOptInterface, and are non-trivial in their implementation. They are in no particular order, but represent broad themes that we see as areas in which JuMP could be improved.","category":"page"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"Done #3106 Make nonlinear programming a first-class citizen. There have been many issues and discussions about this: currently nonlinear constraints are handled through a MOI.NLPBlock and have various limitations and restrictions.\nhttps://github.com/jump-dev/JuMP.jl/issues/1185\nhttps://github.com/jump-dev/JuMP.jl/issues/1198\nhttps://github.com/jump-dev/JuMP.jl/issues/2788\nhttps://github.com/jump-dev/MathOptInterface.jl/issues/846\nhttps://github.com/jump-dev/MathOptInterface.jl/issues/1397\nDone #3385 Add support for coefficient types other than Float64: https://github.com/jump-dev/JuMP.jl/issues/2025 Since the very beginning, JuMP has hard-coded the coefficient type as Float64. This has made it impossible to support solvers which can use other types such as BigFloat or Rational{BigInt}.\nAdd support for constraint programming: https://github.com/jump-dev/JuMP.jl/issues/2227 JuMP has a strong focus on linear, conic and nonlinear optimization problems. We want to add better support for constraint programming.\nDone #3176 Add support for multiobjective problems: https://github.com/jump-dev/JuMP.jl/issues/2099 JuMP is restricted to problems with scalar-valued objectives. We want to extend this to vector-valued problems.\nRefactor the internal code of JuMP's macros. The code in src/macros.jl is some of the oldest part of JuMP and is difficult to read, modify, and extend. We should overhaul the internals of JuMP's macros–-without making user-visible breaking changes–-to improve their long-term maintainability.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"EditURL = \"simple_examples.jl\"","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Simple-examples","page":"Simple examples","title":"Simple examples","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This tutorial is a collection of examples of small nonlinear programs. It uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"using JuMP\nimport Ipopt\nimport Random\nimport Statistics\nimport Test","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#The-Rosenbrock-function","page":"Simple examples","title":"The Rosenbrock function","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"A nonlinear example of the classical Rosenbrock function.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_rosenbrock()\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, x)\n    @variable(model, y)\n    @objective(model, Min, (1 - x)^2 + 100 * (y - x^2)^2)\n    optimize!(model)\n    Test.@test termination_status(model) == LOCALLY_SOLVED\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 0.0 atol = 1e-10\n    Test.@test value(x) ≈ 1.0\n    Test.@test value(y) ≈ 1.0\n    return\nend\n\nexample_rosenbrock()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#The-clnlbeam-problem","page":"Simple examples","title":"The clnlbeam problem","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Based on an AMPL model by Hande Y. Benson","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Copyright (C) 2001 Princeton University All Rights Reserved","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Permission to use, copy, modify, and distribute this software and its documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appear in all copies and that the copyright notice and this permission notice appear in all supporting documentation.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Source:","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"H. Maurer and H.D. Mittelman, \"The non-linear beam via optimal control with bound state variables,\" Optimal Control Applications and Methods 12, pp. 19-31, 1991.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_clnlbeam()\n    N = 1000\n    h = 1 / N\n    alpha = 350\n    model = Model(Ipopt.Optimizer)\n    @variables(model, begin\n        -1 <= t[1:(N+1)] <= 1\n        -0.05 <= x[1:(N+1)] <= 0.05\n        u[1:(N+1)]\n    end)\n    @objective(\n        model,\n        Min,\n        sum(\n            0.5 * h * (u[i+1]^2 + u[i]^2) +\n            0.5 * alpha * h * (cos(t[i+1]) + cos(t[i])) for i in 1:N\n        ),\n    )\n    @constraint(\n        model,\n        [i = 1:N],\n        x[i+1] - x[i] - 0.5 * h * (sin(t[i+1]) + sin(t[i])) == 0,\n    )\n    @constraint(\n        model,\n        [i = 1:N],\n        t[i+1] - t[i] - 0.5 * h * u[i+1] - 0.5 * h * u[i] == 0,\n    )\n    optimize!(model)\n    println(\"\"\"\n    termination_status = $(termination_status(model))\n    primal_status      = $(primal_status(model))\n    objective_value    = $(objective_value(model))\n    \"\"\")\n    return\nend\n\nexample_clnlbeam()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Maximum-likelihood-estimation","page":"Simple examples","title":"Maximum likelihood estimation","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This example uses nonlinear optimization to compute the maximum likelihood estimate (MLE) of the parameters of a normal distribution, a.k.a., the sample mean and variance.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_mle()\n    n = 1_000\n    Random.seed!(1234)\n    data = randn(n)\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, μ, start = 0.0)\n    @variable(model, σ >= 0.0, start = 1.0)\n    @objective(\n        model,\n        Max,\n        n / 2 * log(1 / (2 * π * σ^2)) -\n        sum((data[i] - μ)^2 for i in 1:n) / (2 * σ^2)\n    )\n    optimize!(model)\n    println(\"μ             = \", value(μ))\n    println(\"mean(data)    = \", Statistics.mean(data))\n    println(\"σ^2           = \", value(σ)^2)\n    println(\"var(data)     = \", Statistics.var(data))\n    println(\"MLE objective = \", objective_value(model))\n    Test.@test value(μ) ≈ Statistics.mean(data) atol = 1e-3\n    Test.@test value(σ)^2 ≈ Statistics.var(data) atol = 1e-2\n    # You can even do constrained MLE!\n    @constraint(model, μ == σ^2)\n    optimize!(model)\n    Test.@test value(μ) ≈ value(σ)^2\n    println()\n    println(\"With constraint μ == σ^2:\")\n    println(\"μ                         = \", value(μ))\n    println(\"σ^2                       = \", value(σ)^2)\n    println(\"Constrained MLE objective = \", objective_value(model))\n    return\nend\n\nexample_mle()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Quadratically-constrained-programs","page":"Simple examples","title":"Quadratically constrained programs","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"A simple quadratically constrained program based on an example from Gurobi.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_qcp()\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, x)\n    @variable(model, y >= 0)\n    @variable(model, z >= 0)\n    @objective(model, Max, x)\n    @constraint(model, x + y + z == 1)\n    @constraint(model, x * x + y * y - z * z <= 0)\n    @constraint(model, x * x - y * z <= 0)\n    optimize!(model)\n    print(model)\n    println(\"Objective value: \", objective_value(model))\n    println(\"x = \", value(x))\n    println(\"y = \", value(y))\n    Test.@test termination_status(model) == LOCALLY_SOLVED\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 0.32699 atol = 1e-5\n    Test.@test value(x) ≈ 0.32699 atol = 1e-5\n    Test.@test value(y) ≈ 0.25707 atol = 1e-5\n    return\nend\n\nexample_qcp()","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"EditURL = \"https://github.com/jump-dev/GLPK.jl/blob/v1.1.3/README.md\"","category":"page"},{"location":"packages/GLPK/#GLPK.jl","page":"jump-dev/GLPK.jl","title":"GLPK.jl","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK.jl is a wrapper for the GNU Linear Programming Kit library.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/GLPK/#Affiliation","page":"jump-dev/GLPK.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"This wrapper is maintained by the JuMP community and is not an GNU project.","category":"page"},{"location":"packages/GLPK/#License","page":"jump-dev/GLPK.jl","title":"License","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK.jl is licensed under the GPL v3 license.","category":"page"},{"location":"packages/GLPK/#Installation","page":"jump-dev/GLPK.jl","title":"Installation","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Install GLPK using Pkg.add:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"import Pkg\nPkg.add(\"GLPK\")","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"In addition to installing the GLPK.jl package, this will also download and install the GLPK binaries. You do not need to install GLPK separately.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/GLPK/#Use-with-JuMP","page":"jump-dev/GLPK.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"To use GLPK with JuMP, use GLPK.Optimizer:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"using JuMP, GLPK\nmodel = Model(GLPK.Optimizer)\nset_attribute(model, \"tm_lim\", 60 * 1_000)\nset_attribute(model, \"msg_lev\", GLPK.GLP_MSG_OFF)","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"If the model is primal or dual infeasible, GLPK will attempt to find a certificate of infeasibility. This can be expensive, particularly if you do not intend to use the certificate. If this is the case, use:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"model = Model(() -> GLPK.Optimizer(; want_infeasibility_certificates = false))","category":"page"},{"location":"packages/GLPK/#MathOptInterface-API","page":"jump-dev/GLPK.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The GLPK optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.Reals","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/GLPK/#Options","page":"jump-dev/GLPK.jl","title":"Options","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Options for GLPK are comprehensively documented in the PDF documentation, but they are hard to find.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Options when solving a linear program are defined in Section 2.8.1\nOptions when solving a mixed-integer program are defined in Section 2.10.5","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"However, the following options are likely to be the most useful:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Parameter Example Explanation\nmsg_lev GLPK.GLP_MSG_ALL Message level for terminal output\npresolve GLPK.GLP_ON Turn presolve on or off\ntol_int 1e-5 Absolute tolerance for integer feasibility\ntol_obj 1e-7 Relative objective tolerance for mixed-integer programs","category":"page"},{"location":"packages/GLPK/#Callbacks","page":"jump-dev/GLPK.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Here is an example using GLPK's solver-specific callbacks.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"using JuMP, GLPK, Test\n\nmodel = Model(GLPK.Optimizer)\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\nreasons = UInt8[]\nfunction my_callback_function(cb_data)\n    reason = GLPK.glp_ios_reason(cb_data.tree)\n    push!(reasons, reason)\n    if reason != GLPK.GLP_IROWGEN\n        return\n    end\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nMOI.set(model, GLPK.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2\n@show reasons","category":"page"},{"location":"packages/GLPK/#C-API","page":"jump-dev/GLPK.jl","title":"C API","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The C API can be accessed via GLPK.glp_XXX functions, where the names and arguments are identical to the C API. See the /tests folder for inspiration.","category":"page"},{"location":"packages/GLPK/#Thread-safety","page":"jump-dev/GLPK.jl","title":"Thread safety","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK is not thread-safe and should not be used with multithreading.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"EditURL = \"callbacks.jl\"","category":"page"},{"location":"tutorials/linear/callbacks/#callbacks_tutorial","page":"Callbacks","title":"Callbacks","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"The purpose of the tutorial is to demonstrate the various solver-independent and solver-dependent callbacks that are supported by JuMP.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"The tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"using JuMP\nimport GLPK\nimport Random","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"info: Info\nThis tutorial uses the MathOptInterface API. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/linear/callbacks/#Lazy-constraints","page":"Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a lazy constraint callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_lazy_constraint()\n    model = Model(GLPK.Optimizer)\n    @variable(model, 0 <= x <= 2.5, Int)\n    @variable(model, 0 <= y <= 2.5, Int)\n    @objective(model, Max, y)\n    lazy_called = false\n    function my_callback_function(cb_data)\n        lazy_called = true\n        x_val = callback_value(cb_data, x)\n        y_val = callback_value(cb_data, y)\n        println(\"Called from (x, y) = ($x_val, $y_val)\")\n        status = callback_node_status(cb_data, model)\n        if status == MOI.CALLBACK_NODE_STATUS_FRACTIONAL\n            println(\" - Solution is integer infeasible!\")\n        elseif status == MOI.CALLBACK_NODE_STATUS_INTEGER\n            println(\" - Solution is integer feasible!\")\n        else\n            @assert status == MOI.CALLBACK_NODE_STATUS_UNKNOWN\n            println(\" - I don't know if the solution is integer feasible :(\")\n        end\n        if y_val - x_val > 1 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        elseif y_val + x_val > 3 + 1e-6\n            con = @build_constraint(y + x <= 3)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        end\n    end\n    set_attribute(model, MOI.LazyConstraintCallback(), my_callback_function)\n    optimize!(model)\n    println(\"Optimal solution (x, y) = ($(value(x)), $(value(y)))\")\n    return\nend\n\nexample_lazy_constraint()","category":"page"},{"location":"tutorials/linear/callbacks/#User-cuts","page":"Callbacks","title":"User-cuts","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a user-cut callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_user_cut_constraint()\n    Random.seed!(1)\n    N = 30\n    item_weights, item_values = rand(N), rand(N)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @constraint(model, sum(item_weights[i] * x[i] for i in 1:N) <= 10)\n    @objective(model, Max, sum(item_values[i] * x[i] for i in 1:N))\n    callback_called = false\n    function my_callback_function(cb_data)\n        callback_called = true\n        x_vals = callback_value.(Ref(cb_data), x)\n        accumulated = sum(item_weights[i] for i in 1:N if x_vals[i] > 1e-4)\n        println(\"Called with accumulated = $(accumulated)\")\n        n_terms = sum(1 for i in 1:N if x_vals[i] > 1e-4)\n        if accumulated > 10\n            con = @build_constraint(\n                sum(x[i] for i in 1:N if x_vals[i] > 0.5) <= n_terms - 1\n            )\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.UserCut(cb_data), con)\n        end\n    end\n    set_attribute(model, MOI.UserCutCallback(), my_callback_function)\n    optimize!(model)\n    @show callback_called\n    return\nend\n\nexample_user_cut_constraint()","category":"page"},{"location":"tutorials/linear/callbacks/#Heuristic-solutions","page":"Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a heuristic solution callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_heuristic_solution()\n    Random.seed!(1)\n    N = 30\n    item_weights, item_values = rand(N), rand(N)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @constraint(model, sum(item_weights[i] * x[i] for i in 1:N) <= 10)\n    @objective(model, Max, sum(item_values[i] * x[i] for i in 1:N))\n    callback_called = false\n    function my_callback_function(cb_data)\n        callback_called = true\n        x_vals = callback_value.(Ref(cb_data), x)\n        ret =\n            MOI.submit(model, MOI.HeuristicSolution(cb_data), x, floor.(x_vals))\n        println(\"Heuristic solution status = $(ret)\")\n    end\n    set_attribute(model, MOI.HeuristicCallback(), my_callback_function)\n    optimize!(model)\n    return\nend\n\nexample_heuristic_solution()","category":"page"},{"location":"tutorials/linear/callbacks/#GLPK-solver-dependent-callback","page":"Callbacks","title":"GLPK solver-dependent callback","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using GLPK's solver-dependent callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_solver_dependent_callback()\n    model = Model(GLPK.Optimizer)\n    @variable(model, 0 <= x <= 2.5, Int)\n    @variable(model, 0 <= y <= 2.5, Int)\n    @objective(model, Max, y)\n    lazy_called = false\n    function my_callback_function(cb_data)\n        lazy_called = true\n        reason = GLPK.glp_ios_reason(cb_data.tree)\n        println(\"Called from reason = $(reason)\")\n        if reason != GLPK.GLP_IROWGEN\n            return\n        end\n        x_val = callback_value(cb_data, x)\n        y_val = callback_value(cb_data, y)\n        if y_val - x_val > 1 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        elseif y_val + x_val > 3 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        end\n    end\n    set_attribute(model, GLPK.CallbackFunction(), my_callback_function)\n    optimize!(model)\n    return\nend\n\nexample_solver_dependent_callback()","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/nlp/#Nonlinear-Modeling-(Legacy)","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nThis page describes the legacy nonlinear interface to JuMP. It has a number of quirks and limitations that prompted the development of a new nonlinear interface. The new interface is documented at Nonlinear Modeling. This legacy interface will remain for all future v1.X releases of JuMP. The two nonlinear interfaces cannot be combined.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP has support for general smooth nonlinear (convex and nonconvex) optimization problems. JuMP is able to provide exact, sparse second-order derivatives to solvers. This information can improve solver accuracy and performance.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"There are three main changes to solve nonlinear programs in JuMP.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLobjective instead of @objective\nUse @NLconstraint instead of @constraint\nUse @NLexpression instead of @expression","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nThere are some restrictions on what syntax you can use in the @NLxxx macros. Make sure to read the Syntax notes.","category":"page"},{"location":"manual/nlp/#Set-a-nonlinear-objective","page":"Nonlinear Modeling (Legacy)","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLobjective to set a nonlinear objective.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @NLobjective(model, Min, exp(x[1]) - sqrt(x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"To modify a nonlinear objective, call @NLobjective again.","category":"page"},{"location":"manual/nlp/#Add-a-nonlinear-constraint","page":"Nonlinear Modeling (Legacy)","title":"Add a nonlinear constraint","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLconstraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @NLconstraint(model, exp(x[1]) <= 1)\nexp(x[1]) - 1.0 ≤ 0\n\njulia> @NLconstraint(model, [i = 1:2], x[i]^i >= i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n x[1] ^ 1.0 - 1.0 ≥ 0\n x[2] ^ 2.0 - 2.0 ≥ 0\n\njulia> @NLconstraint(model, con[i = 1:2], prod(x[j] for j = 1:i) == i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n (*)(x[1]) - 1.0 = 0\n x[1] * x[2] - 2.0 = 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nYou can only create nonlinear constraints with <=, >=, and ==. More general Nonlinear-in-Set constraints are not supported.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Delete a nonlinear constraint using delete:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> delete(model, con[1])","category":"page"},{"location":"manual/nlp/#Create-a-nonlinear-expression","page":"Nonlinear Modeling (Legacy)","title":"Create a nonlinear expression","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLexpression to create nonlinear expression objects. The syntax is identical to @expression, except that the expression can contain nonlinear terms.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> expr = @NLexpression(model, exp(x[1]) + sqrt(x[2]))\nsubexpression[1]: exp(x[1]) + sqrt(x[2])\n\njulia> my_anon_expr = @NLexpression(model, [i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[2]: sin(x[1])\n subexpression[3]: sin(x[2])\n\njulia> @NLexpression(model, my_expr[i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[4]: sin(x[1])\n subexpression[5]: sin(x[2])","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear expression can be used in @NLobjective, @NLconstraint, and even nested in other @NLexpressions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @NLobjective(model, Min, expr^2 + 1)\n\njulia> @NLconstraint(model, [i = 1:2], my_expr[i] <= i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n subexpression[4] - 1.0 ≤ 0\n subexpression[5] - 2.0 ≤ 0\n\njulia> @NLexpression(model, nested[i = 1:2], sin(my_expr[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[6]: sin(subexpression[4])\n subexpression[7]: sin(subexpression[5])","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use value to query the value of a nonlinear expression:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> set_start_value(x[1], 1.0)\n\njulia> value(start_value, nested[1])\n0.7456241416655579\n\njulia> sin(sin(1.0))\n0.7456241416655579","category":"page"},{"location":"manual/nlp/#Create-a-nonlinear-parameter","page":"Nonlinear Modeling (Legacy)","title":"Create a nonlinear parameter","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For nonlinear models only, JuMP offers a syntax for explicit \"parameter\" objects, which are constants in the model that can be efficiently updated between solves.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters are declared by using the @NLparameter macro and may be indexed by arbitrary sets analogously to JuMP variables and expressions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The initial value of the parameter must be provided on the right-hand side of the == sign.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLparameter(model, p[i = 1:2] == i)\n2-element Vector{NonlinearParameter}:\n parameter[1] == 1.0\n parameter[2] == 2.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Create anonymous parameters using the value keyword:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> anon_parameter = @NLparameter(model, value = 1)\nparameter[3] == 1.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nA parameter is not an optimization variable. It must be fixed to a value with ==. If you want a parameter that is <= or >=, create a variable instead using @variable.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use value and set_value to query or update the value of a parameter.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> value.(p)\n2-element Vector{Float64}:\n 1.0\n 2.0\n\njulia> set_value(p[2], 3.0)\n3.0\n\njulia> value.(p)\n2-element Vector{Float64}:\n 1.0\n 3.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters must be used within nonlinear macros only.","category":"page"},{"location":"manual/nlp/#When-to-use-a-parameter","page":"Nonlinear Modeling (Legacy)","title":"When to use a parameter","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters are useful when solving nonlinear models in a sequence:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, z)\n@NLparameter(model, x == 1.0)\n@NLobjective(model, Min, (z - x)^2)\noptimize!(model)\n@show value(z) # Equals 1.0.\n\n# Now, update the value of x to solve a different problem.\nset_value(x, 5.0)\noptimize!(model)\n@show value(z) # Equals 5.0\nnothing #hide","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nUsing nonlinear parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.","category":"page"},{"location":"manual/nlp/#Syntax-notes","page":"Nonlinear Modeling (Legacy)","title":"Syntax notes","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The syntax accepted in nonlinear macros is more restricted than the syntax for linear and quadratic macros. We note some important points below.","category":"page"},{"location":"manual/nlp/#Scalar-operations-only","page":"Nonlinear Modeling (Legacy)","title":"Scalar operations only","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Except for the splatting syntax discussed below, all expressions must be simple scalar operations. You cannot use dot, matrix-vector products, vector slices, etc.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @variable(model, y);\n\njulia> c = [1, 2];\n\njulia> @NLobjective(model, Min, c' * x + 3y)\nERROR: Unexpected array [1 2] in nonlinear expression. Nonlinear expressions may contain only scalar expressions.\n[...]","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Translate vector operations into explicit sum() operations:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @NLobjective(model, Min, sum(c[i] * x[i] for i = 1:2) + 3y)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Or use an @expression:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @expression(model, expr, c' * x)\nx[1] + 2 x[2]\n\njulia> @NLobjective(model, Min, expr + 3y)\n","category":"page"},{"location":"manual/nlp/#Splatting","page":"Nonlinear Modeling (Legacy)","title":"Splatting","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The splatting operator   ... is recognized in a very restricted setting for expanding function   arguments. The expression splatted can be only a symbol. More complex   expressions are not recognized.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @NLconstraint(model, *(x...) <= 1.0)\nx[1] * x[2] * x[3] - 1.0 ≤ 0\n\njulia> @NLconstraint(model, *((x / 2)...) <= 0.0)\nERROR: Unsupported use of the splatting operator. JuMP supports splatting only symbols. For example, `x...` is ok, but `(x + 1)...`, `[x; y]...` and `g(f(y)...)` are not.","category":"page"},{"location":"manual/nlp/#User-defined-Functions","page":"Nonlinear Modeling (Legacy)","title":"User-defined Functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP natively supports the set of univariate and multivariate functions recognized by the MOI.Nonlinear submodule. In addition to this list of functions, it is possible to register custom user-defined nonlinear functions. User-defined functions can be used anywhere in @NLobjective, @NLconstraint, and @NLexpression.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP will attempt to automatically register functions it detects in your nonlinear expressions, which usually means manually registering a function is not needed. Two exceptions are if you want to provide custom derivatives, or if the function is not available in the scope of the nonlinear expression.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nUser-defined functions must return a scalar output. For a work-around, see User-defined operators with vector outputs.","category":"page"},{"location":"manual/nlp/#Automatic-differentiation","page":"Nonlinear Modeling (Legacy)","title":"Automatic differentiation","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP does not support black-box optimization, so all user-defined functions must provide derivatives in some form. Fortunately, JuMP supports automatic differentiation of user-defined functions, a feature to our knowledge not available in any comparable modeling systems.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nAutomatic differentiation is not finite differencing. JuMP's automatically computed derivatives are not subject to approximation error.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP uses ForwardDiff.jl to perform automatic differentiation; see the ForwardDiff.jl documentation for a description of how to write a function suitable for automatic differentiation.","category":"page"},{"location":"manual/nlp/#Common-mistakes-when-writing-a-user-defined-function","page":"Nonlinear Modeling (Legacy)","title":"Common mistakes when writing a user-defined function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nGet an error like No method matching Float64(::ForwardDiff.Dual)? Read this section, and see the guidelines at ForwardDiff.jl.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The most common error is that your user-defined function is not generic with respect to the number type, that is, don't assume that the input to the function is Float64.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x::Float64) = 2 * x  # This will not work.\nf(x::Real)    = 2 * x  # This is good.\nf(x)          = 2 * x  # This is also good.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Another reason you may encounter this error is if you create arrays inside your function which are Float64.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"function bad_f(x...)\n    y = zeros(length(x))  # This constructs an array of `Float64`!\n    for i = 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend\n\nfunction good_f(x::T...) where {T<:Real}\n    y = zeros(T, length(x))  # Construct an array of type `T` instead!\n    for i = 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend","category":"page"},{"location":"manual/nlp/#Register-a-function","page":"Nonlinear Modeling (Legacy)","title":"Register a function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"To register a user-defined function with derivatives computed by automatic differentiation, use the register method as in the following example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\n\nmodel = Model()\n\nregister(model, :square, 1, square; autodiff = true)\nregister(model, :my_f, 2, f; autodiff = true)\n\n@variable(model, x[1:2] >= 0.5)\n@NLobjective(model, Min, my_f(x[1], square(x[2])))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The above code creates a JuMP model with the objective function (x[1] - 1)^2 + (x[2]^2 - 2)^2. The arguments to register are:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The model for which the functions are registered.\nA Julia symbol object which serves as the name of the user-defined function in JuMP expressions.\nThe number of input arguments that the function takes.\nThe Julia method which computes the function\nA flag to instruct JuMP to compute exact gradients automatically.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"tip: Tip\nThe symbol :my_f doesn't have to match the name of the function f. However, it's more readable if it does. Make sure you use my_f and not f in the macros.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nUser-defined functions cannot be re-registered and will not update if you modify the underlying Julia function. If you want to change a user-defined function between solves, rebuild the model or use a different name. To use a different name programmatically, see Raw expression input.","category":"page"},{"location":"manual/nlp/#Register-a-function-and-gradient","page":"Nonlinear Modeling (Legacy)","title":"Register a function and gradient","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Forward-mode automatic differentiation as implemented by ForwardDiff.jl has a computational cost that scales linearly with the number of input dimensions. As such, it is not the most efficient way to compute gradients of user-defined functions if the number of input arguments is large. In this case, users may want to provide their own routines for evaluating gradients.","category":"page"},{"location":"manual/nlp/#Univariate-functions","page":"Nonlinear Modeling (Legacy)","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For univariate functions, the gradient function ∇f returns a number that represents the first-order derivative:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x) = x^2\n∇f(x) = 2x\nmodel = Model()\nregister(model, :my_square, 1, f, ∇f; autodiff = true)\n@variable(model, x >= 0)\n@NLobjective(model, Min, my_square(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"If autodiff = true, JuMP will use automatic differentiation to compute the hessian.","category":"page"},{"location":"manual/nlp/#Multivariate-functions","page":"Nonlinear Modeling (Legacy)","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For multivariate functions, the gradient function ∇f must take a gradient vector as the first argument that is filled in-place:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x, y) = (x - 1)^2 + (y - 2)^2\nfunction ∇f(g::AbstractVector{T}, x::T, y::T) where {T}\n    g[1] = 2 * (x - 1)\n    g[2] = 2 * (y - 2)\n    return\nend\n\nmodel = Model()\nregister(model, :my_square, 2, f, ∇f)\n@variable(model, x[1:2] >= 0)\n@NLobjective(model, Min, my_square(x[1], x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nMake sure the first argument to ∇f supports an AbstractVector, and do not assume the input is Float64.","category":"page"},{"location":"manual/nlp/#Register-a-function,-gradient,-and-hessian","page":"Nonlinear Modeling (Legacy)","title":"Register a function, gradient, and hessian","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"You can also register a function with the second-order derivative information, which is a scalar for univariate functions, and a symmetric matrix for multivariate functions.","category":"page"},{"location":"manual/nlp/#Univariate-functions-2","page":"Nonlinear Modeling (Legacy)","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Pass a function which returns a number representing the second-order derivative:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x) = x^2\n∇f(x) = 2x\n∇²f(x) = 2\nmodel = Model()\nregister(model, :my_square, 1, f, ∇f, ∇²f)\n@variable(model, x >= 0)\n@NLobjective(model, Min, my_square(x))","category":"page"},{"location":"manual/nlp/#Multivariate-functions-2","page":"Nonlinear Modeling (Legacy)","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For multivariate functions, the hessian function ∇²f must take an AbstractMatrix as the first argument, the lower-triangular of which is filled in-place:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2\nfunction ∇f(g, x...)\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend\nfunction ∇²f(H, x...)\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1]  <-- Not needed. Fill the lower-triangular only.\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend\n\nmodel = Model()\nregister(model, :rosenbrock, 2, f, ∇f, ∇²f)\n@variable(model, x[1:2])\n@NLobjective(model, Min, rosenbrock(x[1], x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nYou may assume the Hessian matrix H is initialized with zeros, and because H is symmetric, you need only to fill in the non-zero of the lower-triangular terms. The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). However, you may assume only that H supports size(H) and setindex!. Finally, the matrix is treated as dense, so the performance will be poor on functions with high-dimensional input.","category":"page"},{"location":"manual/nlp/#User-defined-functions-with-vector-inputs","page":"Nonlinear Modeling (Legacy)","title":"User-defined functions with vector inputs","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"User-defined functions which take vectors as input arguments (for example, f(x::Vector)) are not supported. Instead, use Julia's splatting syntax to create a function with scalar arguments. For example, instead of","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x::Vector) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"define:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x...) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This function f can be used in a JuMP model as follows:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nmodel = Model()\n@variable(model, x[1:5] >= 0)\nf(x...) = sum(x[i]^i for i in 1:length(x))\nregister(model, :f, 5, f; autodiff = true)\n@NLobjective(model, Min, f(x...))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"tip: Tip\nMake sure to read the syntax restrictions of Splatting.","category":"page"},{"location":"manual/nlp/#Factors-affecting-solution-time","page":"Nonlinear Modeling (Legacy)","title":"Factors affecting solution time","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The execution time when solving a nonlinear programming problem can be divided into two parts, the time spent in the optimization algorithm (the solver) and the time spent evaluating the nonlinear functions and corresponding derivatives. Ipopt explicitly displays these two timings in its output, for example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Total CPU secs in IPOPT (w/o function evaluations)   =      7.412\nTotal CPU secs in NLP function evaluations           =      2.083","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For Ipopt in particular, one can improve the performance by installing advanced sparse linear algebra packages, see Installation Guide. For other solvers, see their respective documentation for performance tips.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The function evaluation time, on the other hand, is the responsibility of the modeling language. JuMP computes derivatives by using reverse-mode automatic differentiation with graph coloring methods for exploiting sparsity of the Hessian matrix. As a conservative bound, JuMP's performance here currently may be expected to be within a factor of 5 of AMPL's. Our paper in SIAM Review has more details.","category":"page"},{"location":"manual/nlp/#Querying-derivatives-from-a-JuMP-model","page":"Nonlinear Modeling (Legacy)","title":"Querying derivatives from a JuMP model","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For some advanced use cases, one may want to directly query the derivatives of a JuMP model instead of handing the problem off to a solver. Internally, JuMP implements the MOI.AbstractNLPEvaluator interface. To obtain an NLP evaluator object from a JuMP model, use NLPEvaluator. index returns the MOI.VariableIndex corresponding to a JuMP variable. MOI.VariableIndex itself is a type-safe wrapper for Int64 (stored in the .value field.)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> raw_index(v::MOI.VariableIndex) = v.value\nraw_index (generic function with 1 method)\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @NLobjective(model, Min, sin(x) + sin(y))\n\njulia> values = zeros(2)\n2-element Vector{Float64}:\n 0.0\n 0.0\n\njulia> x_index = raw_index(JuMP.index(x))\n1\n\njulia> y_index = raw_index(JuMP.index(y))\n2\n\njulia> values[x_index] = 2.0\n2.0\n\njulia> values[y_index] = 3.0\n3.0\n\njulia> d = NLPEvaluator(model)\nNonlinear.Evaluator with available features:\n  * :Grad\n  * :Jac\n  * :JacVec\n  * :Hess\n  * :HessVec\n  * :ExprGraph\n\njulia> MOI.initialize(d, [:Grad])\n\njulia> MOI.eval_objective(d, values)\n1.0504174348855488\n\njulia> sin(2.0) + sin(3.0)\n1.0504174348855488\n\njulia> ∇f = zeros(2)\n2-element Vector{Float64}:\n 0.0\n 0.0\n\njulia> MOI.eval_objective_gradient(d, ∇f, values)\n\njulia> ∇f[x_index], ∇f[y_index]\n(-0.4161468365471424, -0.9899924966004454)\n\njulia> cos(2.0), cos(3.0)\n(-0.4161468365471424, -0.9899924966004454)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Only nonlinear constraints (those added with @NLconstraint), and nonlinear objectives (added with @NLobjective) exist in the scope of the NLPEvaluator.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The NLPEvaluator does not evaluate derivatives of linear or quadratic constraints or objectives.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The index method applied to a nonlinear constraint reference object returns its index as a MOI.Nonlinear.ConstraintIndex. For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLconstraint(model, cons1, sin(x) <= 1);\n\njulia> @NLconstraint(model, cons2, x + 5 == 10);\n\njulia> typeof(cons1)\nNonlinearConstraintRef{ScalarShape} (alias for ConstraintRef{GenericModel{Float64}, MathOptInterface.Nonlinear.ConstraintIndex, ScalarShape})\n\njulia> index(cons1)\nMathOptInterface.Nonlinear.ConstraintIndex(1)\n\njulia> index(cons2)\nMathOptInterface.Nonlinear.ConstraintIndex(2)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"# TODO: Provide a link for how to access the linear and quadratic parts of the\n# model.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Note that for one-sided nonlinear constraints, JuMP subtracts any values on the right-hand side when computing expressions. In other words, one-sided nonlinear constraints are always transformed to have a right-hand side of zero.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This method of querying derivatives directly from a JuMP model is convenient for interacting with the model in a structured way, for example, for accessing derivatives of specific variables. For example, in statistical maximum likelihood estimation problems, one is often interested in the Hessian matrix at the optimal solution, which can be queried using the NLPEvaluator.","category":"page"},{"location":"manual/nlp/#Raw-expression-input","page":"Nonlinear Modeling (Legacy)","title":"Raw expression input","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nThis section requires advanced knowledge of Julia's Expr. You should read the Expressions and evaluation section of the Julia documentation first.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"In addition to the @NLexpression, @NLobjective and @NLconstraint macros, it is also possible to provide Julia Expr objects directly by using add_nonlinear_expression, set_nonlinear_objective and add_nonlinear_constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This input form may be useful if the expressions are generated programmatically, or if you experience compilation issues with the macro input (see Known performance issues for more information).","category":"page"},{"location":"manual/nlp/#Add-a-nonlinear-expression","page":"Nonlinear Modeling (Legacy)","title":"Add a nonlinear expression","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use add_nonlinear_expression to add a nonlinear expression to the model.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> expr = :($(x) + sin($(x)^2))\n:(x + sin(x ^ 2))\n\njulia> expr_ref = add_nonlinear_expression(model, expr)\nsubexpression[1]: x + sin(x ^ 2.0)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr_ref = @NLexpression(model, x + sin(x^2))\nsubexpression[1]: x + sin(x ^ 2.0)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"note: Note\nYou must interpolate the variables directly into the expression expr.","category":"page"},{"location":"manual/nlp/#Set-the-objective-function","page":"Nonlinear Modeling (Legacy)","title":"Set the objective function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use set_nonlinear_objective to set a nonlinear objective.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = :($(x) + $(x)^2)\n:(x + x ^ 2)\n\njulia> set_nonlinear_objective(model, MIN_SENSE, expr)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLobjective(model, Min, x + x^2)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"note: Note\nYou must use MIN_SENSE or MAX_SENSE instead of Min and Max.","category":"page"},{"location":"manual/nlp/#Add-a-constraint","page":"Nonlinear Modeling (Legacy)","title":"Add a constraint","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use add_nonlinear_constraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = :($(x) + $(x)^2)\n:(x + x ^ 2)\n\njulia> add_nonlinear_constraint(model, :($(expr) <= 1))\n(x + x ^ 2.0) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLconstraint(model, Min, x + x^2 <= 1)\n(x + x ^ 2.0) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#More-complicated-examples","page":"Nonlinear Modeling (Legacy)","title":"More complicated examples","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Raw expression input is most useful when the expressions are generated programmatically, often in conjunction with user-defined functions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As an example, we construct a model with the nonlinear constraints f(x) <= 1, where f(x) = x^2 and f(x) = sin(x)^2:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> function main(functions::Vector{Function})\n           model = Model()\n           @variable(model, x)\n           for (i, f) in enumerate(functions)\n               f_sym = Symbol(\"f_$(i)\")\n               register(model, f_sym, 1, f; autodiff = true)\n               add_nonlinear_constraint(model, :($(f_sym)($(x)) <= 1))\n           end\n           print(model)\n           return\n       end\nmain (generic function with 1 method)\n\njulia> main([x -> x^2, x -> sin(x)^2])\nFeasibility\nSubject to\n f_1(x) - 1.0 ≤ 0\n f_2(x) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As another example, we construct a model with the constraint x^2 + sin(x)^2 <= 1:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> function main(functions::Vector{Function})\n           model = Model()\n           @variable(model, x)\n           expr = Expr(:call, :+)\n           for (i, f) in enumerate(functions)\n               f_sym = Symbol(\"f_$(i)\")\n               register(model, f_sym, 1, f; autodiff = true)\n               push!(expr.args, :($(f_sym)($(x))))\n           end\n           add_nonlinear_constraint(model, :($(expr) <= 1))\n           print(model)\n           return\n       end\nmain (generic function with 1 method)\n\njulia> main([x -> x^2, x -> sin(x)^2])\nFeasibility\nSubject to\n (f_1(x) + f_2(x)) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#Registered-functions-with-a-variable-number-of-arguments","page":"Nonlinear Modeling (Legacy)","title":"Registered functions with a variable number of arguments","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"User defined functions require a fixed number of input arguments. However, sometimes you will want to use a registered function like:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> f(x...) = sum(exp(x[i]^2) for i in 1:length(x));","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"with different numbers of arguments.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The solution is to register the same function f for each unique number of input arguments, making sure to use a unique name each time. For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> A = [[1], [1, 2], [2, 3, 4], [1, 3, 4, 5]];\n\njulia> model = Model();\n\njulia> @variable(model, x[1:5]);\n\njulia> funcs = Set{Symbol}();\n\njulia> for a in A\n           key = Symbol(\"f$(length(a))\")\n           if !(key in funcs)\n               push!(funcs, key)\n               register(model, key, length(a), f; autodiff = true)\n           end\n           add_nonlinear_constraint(model, :($key($(x[a]...)) <= 1))\n       end\n\njulia> print(model)\nFeasibility\nSubject to\n f1(x[1]) - 1.0 ≤ 0\n f2(x[1], x[2]) - 1.0 ≤ 0\n f3(x[2], x[3], x[4]) - 1.0 ≤ 0\n f4(x[1], x[3], x[4], x[5]) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#Known-performance-issues","page":"Nonlinear Modeling (Legacy)","title":"Known performance issues","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The macro-based input to JuMP's nonlinear interface can cause a performance issue if you:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"write a macro with a large number (hundreds) of terms\ncall that macro from within a function instead of from the top-level in global scope.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The first issue does not depend on the number of resulting terms in the mathematical expression, but rather the number of terms in the Julia Expr representation of that expression. For example, the expression sum(x[i] for i in 1:1_000_000) contains one million mathematical terms, but the Expr representation is just a single sum.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The most common cause, other than a lot of tedious typing, is if you write a program that automatically writes a JuMP model as a text file, which you later execute. One example is MINLPlib.jl which automatically transpiled models in the GAMS scalar format into JuMP examples.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As a rule of thumb, if you are writing programs to automatically generate expressions for the JuMP macros, you should target the Raw expression input instead. For more information, read MathOptInterface Issue#1997.","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"EditURL = \"https://github.com/jump-dev/Dualization.jl/blob/v0.5.8/README.md\"","category":"page"},{"location":"packages/Dualization/#Dualization.jl","page":"jump-dev/Dualization.jl","title":"Dualization.jl","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"(Image: Build Status) (Image: codecov) (Image: DOI)","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl is an extension package for MathOptInterface.jl that formulates the dual of conic optimization problems.","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl has two main features:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"The Dualization.dualize function that computes the dual formulation of either a MathOptInterface.jl or a JuMP model.\nThe Dualization.dual_optimizer function that creates a MathOptInterface-compatible optimizer that solves the dual of the problem instead of the primal.","category":"page"},{"location":"packages/Dualization/#License","page":"jump-dev/Dualization.jl","title":"License","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Dualization/#Installation","page":"jump-dev/Dualization.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Install Dualization using Pkg.add:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"import Pkg\nPkg.add(\"Dualization\")","category":"page"},{"location":"packages/Dualization/#Use-with-JuMP","page":"jump-dev/Dualization.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"To compute the dual formulation of a JuMP model, use dualize:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"using JuMP, Dualization\nmodel = Model()\n# ... build model ...\ndual_model = dualize(model)","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"To solve the dual formulation of a JuMP model, create a dual_optimizer:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"using JuMP, Dualization, SCS\nmodel = Model(dual_optimizer(SCS.Optimizer))\n# ... build model ...\noptimize!(model)  # Solves the dual instead of the primal","category":"page"},{"location":"packages/Dualization/#Documentation","page":"jump-dev/Dualization.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"The documentation for Dualization.jl includes a detailed description of the dual reformulation, along with examples and an API reference.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/constraints.md\"","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/constraints/#Constraints","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"moi/manual/constraints/#Add-a-constraint","page":"Constraints","title":"Add a constraint","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use add_constraint to add a single constraint.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> c = MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonnegatives(2))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"add_constraint returns a ConstraintIndex type, which is used to refer to the added constraint in other calls.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Check if a ConstraintIndex is valid using is_valid.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.is_valid(model, c)\ntrue","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use add_constraints to add a number of constraints of the same type.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> c = MOI.add_constraints(\n           model,\n           [x[1], x[2]],\n           [MOI.GreaterThan(0.0), MOI.GreaterThan(1.0)]\n       )\n2-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(2)","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"This time, a vector of ConstraintIndex are returned.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use supports_constraint to check if the model supports adding a constraint type.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.supports_constraint(\n           model,\n           MOI.VariableIndex,\n           MOI.GreaterThan{Float64},\n       )\ntrue","category":"page"},{"location":"moi/manual/constraints/#Delete-a-constraint","page":"Constraints","title":"Delete a constraint","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use delete to delete a constraint.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.delete(model, c)\n\njulia> MOI.is_valid(model, c)\nfalse","category":"page"},{"location":"moi/manual/constraints/#Constraint-attributes","page":"Constraints","title":"Constraint attributes","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"The following attributes are available for constraints:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"ConstraintName\nConstraintPrimalStart\nConstraintDualStart\nConstraintPrimal\nConstraintDual\nConstraintBasisStatus\nConstraintFunction\nCanonicalConstraintFunction\nConstraintSet","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Get and set these attributes using get and set.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.set(model, MOI.ConstraintName(), c, \"con_c\")\n\njulia> MOI.get(model, MOI.ConstraintName(), c)\n\"con_c\"","category":"page"},{"location":"moi/manual/constraints/#Constraints-by-function-set-pairs","page":"Constraints","title":"Constraints by function-set pairs","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Below is a list of common constraint types and how they are represented as function-set pairs in MOI. In the notation below, x is a vector of decision variables, x_i is a scalar decision variable, alpha beta are scalar constants, a b are constant vectors, A is a constant matrix and mathbbR_+ (resp. mathbbR_-) is the set of non-negative (resp. non-positive) real numbers.","category":"page"},{"location":"moi/manual/constraints/#Linear-constraints","page":"Constraints","title":"Linear constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\na^Tx le beta ScalarAffineFunction LessThan\na^Tx ge alpha ScalarAffineFunction GreaterThan\na^Tx = beta ScalarAffineFunction EqualTo\nalpha le a^Tx le beta ScalarAffineFunction Interval\nx_i le beta VariableIndex LessThan\nx_i ge alpha VariableIndex GreaterThan\nx_i = beta VariableIndex EqualTo\nalpha le x_i le beta VariableIndex Interval\nAx + b in mathbbR_+^n VectorAffineFunction Nonnegatives\nAx + b in mathbbR_-^n VectorAffineFunction Nonpositives\nAx + b = 0 VectorAffineFunction Zeros","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"By convention, solvers are not expected to support nonzero constant terms in the ScalarAffineFunctions the first four rows of the preceding table because they are redundant with the parameters of the sets. For example, encode 2x + 1 le 2 as 2x le 1.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Constraints with VariableIndex in LessThan, GreaterThan, EqualTo, or Interval sets have a natural interpretation as variable bounds. As such, it is typically not natural to impose multiple lower- or upper-bounds on the same variable, and the solver interfaces will throw respectively LowerBoundAlreadySet or UpperBoundAlreadySet.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Moreover, adding two VariableIndex constraints on the same variable with the same set is impossible because they share the same index as it is the index of the variable, see ConstraintIndex.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"It is natural, however, to impose upper- and lower-bounds separately as two different constraints on a single variable. The difference between imposing bounds by using a single Interval constraint and by using separate LessThan and GreaterThan constraints is that the latter will allow the solver to return separate dual multipliers for the two bounds, while the former will allow the solver to return only a single dual for the interval constraint.","category":"page"},{"location":"moi/manual/constraints/#Conic-constraints","page":"Constraints","title":"Conic constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nlVert Ax + brVert_2 le c^Tx + d VectorAffineFunction SecondOrderCone\ny ge lVert x rVert_2 VectorOfVariables SecondOrderCone\n2yz ge lVert x rVert_2^2 yz ge 0 VectorOfVariables RotatedSecondOrderCone\n(a_1^Tx + b_1a_2^Tx + b_2a_3^Tx + b_3) in mathcalE VectorAffineFunction ExponentialCone\nA(x) in mathcalS_+ VectorAffineFunction PositiveSemidefiniteConeTriangle\nB(x) in mathcalS_+ VectorAffineFunction PositiveSemidefiniteConeSquare\nx in mathcalS_+ VectorOfVariables PositiveSemidefiniteConeTriangle\nx in mathcalS_+ VectorOfVariables PositiveSemidefiniteConeSquare","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"where mathcalE is the exponential cone (see ExponentialCone), mathcalS_+ is the set of positive semidefinite symmetric matrices, A is an affine map that outputs symmetric matrices and B is an affine map that outputs square matrices.","category":"page"},{"location":"moi/manual/constraints/#Quadratic-constraints","page":"Constraints","title":"Quadratic constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nfrac12x^TQx + a^Tx + b ge 0 ScalarQuadraticFunction GreaterThan\nfrac12x^TQx + a^Tx + b le 0 ScalarQuadraticFunction LessThan\nfrac12x^TQx + a^Tx + b = 0 ScalarQuadraticFunction EqualTo\nBilinear matrix inequality VectorQuadraticFunction PositiveSemidefiniteCone...","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"note: Note\nFor more details on the internal format of the quadratic functions see ScalarQuadraticFunction or VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/constraints/#Discrete-and-logical-constraints","page":"Constraints","title":"Discrete and logical constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nx_i in mathbbZ VariableIndex Integer\nx_i in 01 VariableIndex ZeroOne\nx_i in 0 cup lu VariableIndex Semicontinuous\nx_i in 0 cup ll+1ldotsu-1u VariableIndex Semiinteger\nAt most one component of x can be nonzero VectorOfVariables SOS1\nAt most two components of x can be nonzero, and if so they must be adjacent components VectorOfVariables SOS2\ny = 1 implies a^T x in S VectorAffineFunction Indicator","category":"page"},{"location":"moi/manual/constraints/#JuMP-mapping","page":"Constraints","title":"JuMP mapping","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"The following bullet points show examples of how JuMP constraints are translated into MOI function-set pairs:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@constraint(m, 2x + y <= 10) becomes ScalarAffineFunction-in-LessThan\n@constraint(m, 2x + y >= 10) becomes ScalarAffineFunction-in-GreaterThan\n@constraint(m, 2x + y == 10) becomes ScalarAffineFunction-in-EqualTo\n@constraint(m, 0 <= 2x + y <= 10) becomes ScalarAffineFunction-in-Interval\n@constraint(m, 2x + y in ArbitrarySet()) becomes ScalarAffineFunction-in-ArbitrarySet.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Variable bounds are handled in a similar fashion:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@variable(m, x <= 1) becomes VariableIndex-in-LessThan\n@variable(m, x >= 1) becomes VariableIndex-in-GreaterThan","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"One notable difference is that a variable with an upper and lower bound is translated into two constraints, rather than an interval, that is:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@variable(m, 0 <= x <= 1) becomes VariableIndex-in-LessThan and  VariableIndex-in-GreaterThan.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"EditURL = \"sudoku.jl\"","category":"page"},{"location":"tutorials/linear/sudoku/#Sudoku","page":"Sudoku","title":"Sudoku","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"This tutorial was originally contributed by Iain Dunning.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Sudoku is a popular number puzzle. The goal is to place the digits 1 to 9 on a nine-by-nine grid, with some of the digits already filled in. Your solution must satisfy the following rules:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The numbers 1 to 9 must appear in each 3x3 square\nThe numbers 1 to 9 must appear in each row\nThe numbers 1 to 9 must appear in each column","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Here is a partially solved Sudoku problem:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"(Image: Partially solved Sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Solving a Sudoku isn't an optimization problem with an objective; its actually a feasibility problem: we wish to find a feasible solution that satisfies these rules. You can think of it as an optimization problem with an objective of 0.","category":"page"},{"location":"tutorials/linear/sudoku/#Mixed-integer-linear-programming-formulation","page":"Sudoku","title":"Mixed-integer linear programming formulation","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We can model this problem using 0-1 integer programming: a problem where all the decision variables are binary. We'll use JuMP to create the model, and then we can solve it with any integer programming solver.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"using JuMP\nusing HiGHS","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We will define a binary variable (a variable that is either 0 or 1) for each possible number in each possible cell. The meaning of each variable is as follows: x[i,j,k] = 1  if and only if cell (i,j) has number k, where i is the row and j is the column.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create a model","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sudoku = Model(HiGHS.Optimizer)\nset_silent(sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create our variables","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@variable(sudoku, x[i = 1:9, j = 1:9, k = 1:9], Bin);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Now we can begin to add our constraints. We'll actually start with something obvious to us as humans, but what we need to enforce: that there can be only one number per cell.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:9  # For each row\n    for j in 1:9  # and each column\n        # Sum across all the possible digits. One and only one of the digits\n        # can be in this cell, so the sum must be equal to one.\n        @constraint(sudoku, sum(x[i, j, k] for k in 1:9) == 1)\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Next we'll add the constraints for the rows and the columns. These constraints are all very similar, so much so that we can actually add them at the same time.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for ind in 1:9  # Each row, OR each column\n    for k in 1:9  # Each digit\n        # Sum across columns (j) - row constraint\n        @constraint(sudoku, sum(x[ind, j, k] for j in 1:9) == 1)\n        # Sum across rows (i) - column constraint\n        @constraint(sudoku, sum(x[i, ind, k] for i in 1:9) == 1)\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Finally, we have the to enforce the constraint that each digit appears once in each of the nine 3x3 sub-grids. Our strategy will be to index over the top-left corners of each 3x3 square with for loops, then sum over the squares.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:3:7\n    for j in 1:3:7\n        for k in 1:9\n            # i is the top left row, j is the top left column.\n            # We'll sum from i to i+2, for example, i=4, r=4, 5, 6.\n            @constraint(\n                sudoku,\n                sum(x[r, c, k] for r in i:(i+2), c in j:(j+2)) == 1\n            )\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The final step is to add the initial solution as a set of constraints. We'll solve the problem that is in the picture at the start of the tutorial. We'll put a 0 if there is no digit in that location.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The given digits","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"init_sol = [\n    5 3 0 0 7 0 0 0 0\n    6 0 0 1 9 5 0 0 0\n    0 9 8 0 0 0 0 6 0\n    8 0 0 0 6 0 0 0 3\n    4 0 0 8 0 3 0 0 1\n    7 0 0 0 2 0 0 0 6\n    0 6 0 0 0 0 2 8 0\n    0 0 0 4 1 9 0 0 5\n    0 0 0 0 8 0 0 7 9\n]\nfor i in 1:9\n    for j in 1:9\n        # If the space isn't empty\n        if init_sol[i, j] != 0\n            # Then the corresponding variable for that digit and location must\n            # be 1.\n            fix(x[i, j, init_sol[i, j]], 1; force = true)\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"solve problem","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"optimize!(sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Extract the values of x","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"x_val = value.(x);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create a matrix to store the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol = zeros(Int, 9, 9)  # 9x9 matrix of integers\nfor i in 1:9\n    for j in 1:9\n        for k in 1:9\n            # Integer programs are solved as a series of linear programs so the\n            # values might not be precisely 0 and 1. We can round them to\n            # the nearest integer to make it easier.\n            if round(Int, x_val[i, j, k]) == 1\n                sol[i, j] = k\n            end\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Display the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Which is the correct solution:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"(Image: Solved Sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/#Constraint-programming-formulation","page":"Sudoku","title":"Constraint programming formulation","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We can also model this problem using constraint programming and the all-different constraint, which says that no two elements of a vector can take the same value.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Because of the reformulation system in MathOptInterface, we can still solve this problem using HiGHS.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n# HiGHS v1.2 has a bug in presolve which causes the problem to be classified as\n# infeasible.\nset_attribute(model, \"presolve\", \"off\")","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Instead of the binary variables, we directly define a 9x9 grid of integer values between 1 and 9:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@variable(model, 1 <= x[1:9, 1:9] <= 9, Int);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Then, we enforce that the values in each row must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@constraint(model, [i = 1:9], x[i, :] in MOI.AllDifferent(9));\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"That the values in each column must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@constraint(model, [j = 1:9], x[:, j] in MOI.AllDifferent(9));\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"And that the values in each 3x3 sub-grid must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in (0, 3, 6), j in (0, 3, 6)\n    @constraint(model, vec(x[i.+(1:3), j.+(1:3)]) in MOI.AllDifferent(9))\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Finally, as before we set the initial solution and optimize:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:9, j in 1:9\n    if init_sol[i, j] != 0\n        fix(x[i, j], init_sol[i, j]; force = true)\n    end\nend\n\noptimize!(model)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Display the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"csp_sol = round.(Int, value.(x))","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Which is the same as we found before:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol == csp_sol","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"EditURL = \"performance_tips.jl\"","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Performance-tips","page":"Performance tips","title":"Performance tips","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"By now you should have read the other \"getting started\" tutorials. You're almost ready to write your own models, but before you do so there are some important things to be aware of.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Read-the-Julia-performance-tips","page":"Performance tips","title":"Read the Julia performance tips","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"The first thing to do is read the Performance tips section of the Julia manual. The most important rule is to avoid global variables. This is particularly important if you're learning JuMP after using a language like MATLAB.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#The-\"time-to-first-solve\"-issue","page":"Performance tips","title":"The \"time-to-first-solve\" issue","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Similar to the infamous time-to-first-plot plotting problem, JuMP suffers from time-to-first-solve latency. This latency occurs because the first time you call JuMP code in each session, Julia needs to compile a lot of code specific to your problem. This issue is actively being worked on, but there are a few things you can do to improve things.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-1:-don't-call-JuMP-from-the-command-line","page":"Performance tips","title":"Suggestion 1: don't call JuMP from the command line","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"In other languages, you might be used to a workflow like:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ julia my_script.jl","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"This doesn't work for JuMP, because we have to pay the compilation latency every time you run the script. Instead, use one of the suggested workflows from the Julia documentation.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-2:-disable-bridges-if-none-are-being-used","page":"Performance tips","title":"Suggestion 2: disable bridges if none are being used","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"At present, the majority of the latency problems are caused by JuMP's bridging mechanism. If you only use constraints that are natively supported by the solver, you can disable bridges by passing add_bridges = false to Model.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model(HiGHS.Optimizer; add_bridges = false)","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-3:-use-PackageCompiler","page":"Performance tips","title":"Suggestion 3: use PackageCompiler","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"As an example of compilation latency, consider the following linear program with two variables and two constraints:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using JuMP, HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\noptimize!(model)\nopen(\"model.log\", \"w\") do io\n    print(io, solution_summary(model; verbose = true))\n    return\nend","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Saving the problem in model.jl and calling from the command line results in:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ time julia model.jl\n15.78s user 0.48s system 100% cpu 16.173 total","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Clearly, 16 seconds is a large overhead to pay for solving this trivial model. However, the compilation latency is independent on the problem size, and so 16 seconds of additional overhead may be tolerable for larger models that take minutes or hours to solve.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"In cases where the compilation latency is intolerable, JuMP is compatible with the PackageCompiler.jl package, which makes it easy to generate a custom sysimage (a binary extension to Julia that caches compiled code) that dramatically reduces the compilation latency. A custom image for our problem can be created as follows:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using PackageCompiler, Libdl\nPackageCompiler.create_sysimage(\n    [\"JuMP\", \"HiGHS\"],\n    sysimage_path = \"customimage.\" * Libdl.dlext,\n    precompile_execution_file = \"model.jl\",\n)","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"When Julia is run with the custom image, the run time is now 0.7 seconds instead of 16:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ time julia --sysimage customimage model.jl\n0.68s user 0.22s system 153% cpu 0.587 total","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Other performance tweaks, such as disabling bridges or using direct mode can reduce this time further.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"note: Note\ncreate_sysimage only needs to be run once, and the same sysimage can be used–to a slight detriment of performance–even if we modify model.jl or run a different file.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Use-macros-to-build-expressions","page":"Performance tips","title":"Use macros to build expressions","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Use JuMP's macros (or add_to_expression!) to build expressions. Avoid constructing expressions outside the macros.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Constructing an expression outside the macro results in intermediate copies of the expression. For example,","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"x[1] + x[2] + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"is equivalent to","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"a = x[1]\nb = a + x[2]\nc = b + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Since we only care about c, the a and b expressions are not needed and constructing them slows the program down.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"JuMP's macros rewrite the expressions to operate in-place and avoid these extra copies. Because they allocate less memory, they are faster, particularly for large expressions.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Here's an example.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model()\n@variable(model, x[1:3])","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Here's what happens if we construct the expression outside the macro:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"@allocated x[1] + x[2] + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"info: Info\nThe @allocated measures how many bytes were allocated during the evaluation of an expression. Fewer is better.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"If we use the @expression macro, we get many fewer allocations:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"@allocated @expression(model, x[1] + x[2] + x[3])","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Disable-string-names","page":"Performance tips","title":"Disable string names","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"By default, JuMP creates String names for variables and constraints and passes these to the solver. The benefit of passing names is that it improves the readability of log messages from the solver (for example, \"variable x has invalid bounds\" instead of \"variable v1203 has invalid bounds\"), but for larger models the overhead of passing names can be non-trivial.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Disable the creation of String names by setting set_string_name = false in the @variable and @constraint macros, or by calling set_string_names_on_creation to disable all names for a particular model:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model();\nset_string_names_on_creation(model, false)\n@variable(model, x)\n@constraint(model, c, 2x <= 1)","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Note that this doesn't change how symbolic names and bindings are stored:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"x\nmodel[:x]\nx === model[:x]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"But you can no longer look up the variable by the string name:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"variable_by_name(model, \"x\") === nothing","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"info: Info\nFor more information on the difference between string names, symbolic names, and bindings, see String names, symbolic names, and bindings.","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"EditURL = \"https://github.com/jump-dev/ParametricOptInterface.jl/blob/v0.5.1/README.md\"","category":"page"},{"location":"packages/ParametricOptInterface/#ParametricOptInterface.jl","page":"jump-dev/ParametricOptInterface.jl","title":"ParametricOptInterface.jl","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"(Image: stable docs) (Image: development docs) (Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface.jl is a package that adds parameters to models in JuMP and MathOptInterface.","category":"page"},{"location":"packages/ParametricOptInterface/#License","page":"jump-dev/ParametricOptInterface.jl","title":"License","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface.jl is licensed under the MIT License.","category":"page"},{"location":"packages/ParametricOptInterface/#Installation","page":"jump-dev/ParametricOptInterface.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"Install ParametricOptInterface using Pkg.add:","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"import Pkg\nPkg.add(\"ParametricOptInterface\")","category":"page"},{"location":"packages/ParametricOptInterface/#Documentation","page":"jump-dev/ParametricOptInterface.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"The documentation for ParametricOptInterface.jl includes a detailed description of the theory behind the package, along with examples, tutorials, and an API reference.","category":"page"},{"location":"packages/ParametricOptInterface/#Use-with-JuMP","page":"jump-dev/ParametricOptInterface.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"Use ParametricOptInterface with JuMP by following this brief example:","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"using JuMP, HiGHS\nimport ParametricOptInterface as POI\nmodel = Model(() -> POI.Optimizer(HiGHS.Optimizer()))\n@variable(model, x)\n@variable(model, p in POI.Parameter(1.0))\n@constraint(model, cons, x + p >= 3)\n@objective(model, Min, 2x)\noptimize!(model)\nMOI.set(model, POI.ParameterValue(), p, 2.0)\noptimize!(model)","category":"page"},{"location":"packages/ParametricOptInterface/#GSOC2020","page":"jump-dev/ParametricOptInterface.jl","title":"GSOC2020","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface began as a NumFOCUS sponsored Google Summer of Code (2020) project.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"EditURL = \"quantum_discrimination.jl\"","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Quantum-state-discrimination","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial solves the problem of quantum state discrimination.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The purpose of this tutorial to demonstrate how to solve problems involving complex-valued decision variables and the HermitianPSDCone. See Complex number support for more details.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Required-packages","page":"Quantum state discrimination","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial makes use of the following packages:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"using JuMP\nimport LinearAlgebra\nimport SCS","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Formulation","page":"Quantum state discrimination","title":"Formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"A d-dimensional quantum state, rho, can be defined by a complex-valued Hermitian matrix with a trace of 1. Assume we have N d-dimensional quantum states, rho_i_i=1^N, each of which is equally likely.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The goal of the quantum state discrimination problem is to choose a positive operator-valued measure (POVM),  E_i _i=1^N, such that if we observe E_i then the most probable state that we are in is rho_i.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Each POVM element, E_i, is a complex-valued Hermitian matrix, and there is a requirement that sumlimits_i=1^N E_i = mathbfI.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To choose a POVM, we want to maximize the probability that we guess the quantum state correctly. This can be formulated as the following optimization problem:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"beginaligned\nmaxlimits_E   frac1N sumlimits_i=1^N operatornametr(rho_i  E_i) \ntextst       sumlimits_i=1^N E_i = mathbfI \n                      E_i succeq 0  forall i = 1ldotsN\nendaligned","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Data","page":"Quantum state discrimination","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To setup our problem, we need N d-dimensional quantum states. To keep the problem simple, we use N = 2 and d = 2.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"N, d = 2, 2","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"We then generated N random d-dimensional quantum states:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"function random_state(d)\n    x = randn(ComplexF64, (d, d))\n    y = x * x'\n    return LinearAlgebra.Hermitian(y / LinearAlgebra.tr(y))\nend\n\nρ = [random_state(d) for i in 1:N]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#JuMP-formulation","page":"Quantum state discrimination","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To model the problem in JuMP, we need a solver that supports positive semidefinite matrices:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"model = Model(SCS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Then, we construct our set of E variables:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"E = [@variable(model, [1:d, 1:d] in HermitianPSDCone()) for i in 1:N]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Here we have created a vector of matrices. This is different to other modeling languages such as YALMIP, which allow you to create a multi-dimensional array in which 2-dimensional slices of the array are Hermitian matrices.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"We also need to enforce the constraint that sumlimits_i=1^N E_i = mathbfI:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@constraint(model, sum(E) == LinearAlgebra.I)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This constraint is a complex-valued equality constraint. In the solver, it will be decomposed onto two types of equality constraints: one to enforce equality of the real components, and one to enforce equality of the imaginary components.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Our objective is to maximize the expected probability of guessing correctly:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@objective(\n    model,\n    Max,\n    sum(real(LinearAlgebra.tr(ρ[i] * E[i])) for i in 1:N) / N,\n)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Now we optimize:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The probability of guessing correctly is:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"When N = 2, there is a known analytical solution of:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"0.5 + 0.25 * sum(LinearAlgebra.svdvals(ρ[1] - ρ[2]))","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"proving that we found the optimal solution.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Finally, the optimal POVM is:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"solution = [value.(e) for e in E]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"tip: Tip\nDuality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual of this problem instead of the primal. If performance is an issue, see the Dualization tutorial for more details.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Alternative-formulation","page":"Quantum state discrimination","title":"Alternative formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The formulation above includes N Hermitian matrices and a set of linear equality constraints. We can simplify the problem by replacing E_N with E_N = I - sumlimits_i=1^N-1 E_i. This results in:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"model = Model(SCS.Optimizer)\nset_silent(model)\nE = [@variable(model, [1:d, 1:d] in HermitianPSDCone()) for i in 1:N-1]\nE_N = LinearAlgebra.Hermitian(LinearAlgebra.I - sum(E))\n@constraint(model, E_N in HermitianPSDCone())\npush!(E, E_N)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The objective can also be simplified, by observing that it is equivalent to:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@objective(model, Max, real(LinearAlgebra.dot(ρ, E)) / N)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Then we can check that we get the same solution:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"objective_value(model)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/implementing.md\"","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/implementing/#Implementing-a-solver-interface","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"This guide outlines the basic steps to implement an interface to MathOptInterface for a new solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"danger: Danger\nImplementing an interface to MathOptInterface for a new solver is a lot of work. Before starting, we recommend that you join the Developer chatroom and explain a little bit about the solver you are wrapping. If you have questions that are not answered by this guide, please ask them in the Developer chatroom so we can improve this guide.","category":"page"},{"location":"moi/tutorials/implementing/#A-note-on-the-API","page":"Implementing a solver interface","title":"A note on the API","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The API of MathOptInterface is large and varied. In order to support the diversity of solvers and use-cases, we make heavy use of duck-typing. That is, solvers are not expected to implement the full API, nor is there a well-defined minimal subset of what must be implemented. Instead, you should implement the API as necessary to make the solver function as you require.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The main reason for using duck-typing is that solvers work in different ways and target different use-cases.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Some solvers support incremental problem construction, support modification after a solve, and have native support for things like variable names.\nOther solvers are \"one-shot\" solvers that require all of the problem data to construct and solve the problem in a single function call. They do not support modification or things like variable names.\nOther \"solvers\" are not solvers at all, but things like file readers. These may only support functions like read_from_file, and may not even support the ability to add variables or constraints directly.\nFinally, some \"solvers\" are layers which take a problem as input, transform it according to some rules, and pass the transformed problem to an inner solver.","category":"page"},{"location":"moi/tutorials/implementing/#Preliminaries","page":"Implementing a solver interface","title":"Preliminaries","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before starting on your wrapper, you should do some background research and make the solver accessible via Julia.","category":"page"},{"location":"moi/tutorials/implementing/#Decide-if-MathOptInterface-is-right-for-you","page":"Implementing a solver interface","title":"Decide if MathOptInterface is right for you","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The first step in writing a wrapper is to decide whether implementing an interface is the right thing to do.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface is an abstraction layer for unifying constrained mathematical optimization solvers. If your solver doesn't fit in the category, for example, it implements a derivative-free algorithm for unconstrained objective functions, MathOptInterface may not be the right tool for the job.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf you're not sure whether you should write an interface, ask in the Developer chatroom.","category":"page"},{"location":"moi/tutorials/implementing/#Find-a-similar-solver-already-wrapped","page":"Implementing a solver interface","title":"Find a similar solver already wrapped","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to find (if possible) a similar solver that is already wrapped. Although not strictly necessary, this will be a good place to look for inspiration when implementing your wrapper.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The JuMP documentation has a good list of solvers, along with the problem classes they support.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf you're not sure which solver is most similar, ask in the Developer chatroom.","category":"page"},{"location":"moi/tutorials/implementing/#Create-a-low-level-interface","page":"Implementing a solver interface","title":"Create a low-level interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before writing a MathOptInterface wrapper, you first need to be able to call the solver from Julia.","category":"page"},{"location":"moi/tutorials/implementing/#Wrapping-solvers-written-in-Julia","page":"Implementing a solver interface","title":"Wrapping solvers written in Julia","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver is written in Julia, there's nothing to do here. Go to the next section.","category":"page"},{"location":"moi/tutorials/implementing/#Wrapping-solvers-written-in-C","page":"Implementing a solver interface","title":"Wrapping solvers written in C","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Julia is well suited to wrapping solvers written in C.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nThis is not true for C++. If you have a solver written in C++, first write a C interface, then wrap the C interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before writing a MathOptInterface wrapper, there are a few extra steps.","category":"page"},{"location":"moi/tutorials/implementing/#Create-a-JLL","page":"Implementing a solver interface","title":"Create a JLL","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If the C code is publicly available under an open source license, create a JLL package via Yggdrasil. The easiest way to do this is to copy an existing solver. Good examples to follow are the COIN-OR solvers.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nBuilding the solver via Yggdrasil is non-trivial. Please ask the Developer chatroom for help.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If the code is commercial or not publicly available, the user will need to manually install the solver. See Gurobi.jl or CPLEX.jl for examples of how to structure this.","category":"page"},{"location":"moi/tutorials/implementing/#implement_clang","page":"Implementing a solver interface","title":"Use Clang.jl to wrap the C API","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to use Clang.jl to automatically wrap the C API. The easiest way to do this is to follow an example. Good examples to follow are Cbc.jl and HiGHS.jl.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Sometimes, you will need to make manual modifications to the resulting files.","category":"page"},{"location":"moi/tutorials/implementing/#Solvers-written-in-other-languages","page":"Implementing a solver interface","title":"Solvers written in other languages","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Ask the Developer chatroom for advice. You may be able to use one of the JuliaInterop packages to call out to the solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, SeDuMi.jl uses MATLAB.jl to call the SeDuMi solver written in MATLAB.","category":"page"},{"location":"moi/tutorials/implementing/#Structuring-the-package","page":"Implementing a solver interface","title":"Structuring the package","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Structure your wrapper as a Julia package. Consult the Julia documentation if you haven't done this before.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI solver interfaces may be in the same package as the solver itself (either the C wrapper if the solver is accessible through C, or the Julia code if the solver is written in Julia, for example), or in a separate package which depends on the solver package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"note: Note\nThe JuMP core contributors request that you do not use \"JuMP\" in the name of your package without prior consent.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Your package should have the following structure:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"/.github\n    /workflows\n        ci.yml\n        format_check.yml\n        TagBot.yml\n/gen\n    gen.jl  # Code to wrap the C API\n/src\n    NewSolver.jl\n    /gen\n        libnewsolver_api.jl\n        libnewsolver_common.jl\n    /MOI_wrapper\n        MOI_wrapper.jl\n        other_files.jl\n/test\n    runtests.jl\n    /MOI_wrapper\n        MOI_wrapper.jl\n.gitignore\n.JuliaFormatter.toml\nREADME.md\nLICENSE.md\nProject.toml","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The /.github folder contains the scripts for GitHub actions. The easiest way to write these is to copy the ones from an existing solver.\nThe /gen and /src/gen folders are only needed if you are wrapping a solver written in C.\nThe /src/MOI_wrapper folder contains the Julia code for the MOI wrapper.\nThe /test folder contains code for testing your package. See Setup tests for more information.\nThe .JuliaFormatter.toml and .github/workflows/format_check.yml enforce code formatting using JuliaFormatter.jl. Check existing solvers or JuMP.jl for details.","category":"page"},{"location":"moi/tutorials/implementing/#Documentation","page":"Implementing a solver interface","title":"Documentation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Your package must include documentation explaining how to use the package. The easiest approach is to include documentation in your README.md. A more involved option is to use Documenter.jl.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Examples of packages with README-based documentation include:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Cbc.jl\nHiGHS.jl\nSCS.jl","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Examples of packages with Documenter-based documentation include:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Alpine.jl\nCOSMO.jl\nJuniper.jl","category":"page"},{"location":"moi/tutorials/implementing/#Setup-tests","page":"Implementing a solver interface","title":"Setup tests","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The best way to implement an interface to MathOptInterface is via test-driven development.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The MOI.Test submodule contains a large test suite to help check that you have implemented things correctly.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Follow the guide How to test a solver to set up the tests for your package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nRun the tests frequently when developing. However, at the start there is going to be a lot of errors. Start by excluding large classes of tests (for example, exclude = [\"test_basic_\", \"test_model_\"], implement any missing methods until the tests pass, then remove an exclusion and repeat.","category":"page"},{"location":"moi/tutorials/implementing/#Initial-code","page":"Implementing a solver interface","title":"Initial code","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By this point, you should have a package setup with tests, formatting, and access to the underlying solver. Now it's time to start writing the wrapper.","category":"page"},{"location":"moi/tutorials/implementing/#The-Optimizer-object","page":"Implementing a solver interface","title":"The Optimizer object","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The first object to create is a subtype of AbstractOptimizer. This type is going to store everything related to the problem.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By convention, these optimizers should not be exported and should be named PackageName.Optimizer.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"import MathOptInterface as MOI\n\nstruct Optimizer <: MOI.AbstractOptimizer\n    # Fields go here\nend","category":"page"},{"location":"moi/tutorials/implementing/#Optimizer-objects-for-C-solvers","page":"Implementing a solver interface","title":"Optimizer objects for C solvers","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nThis section is important if you wrap a solver written in C.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Wrapping a solver written in C will require the use of pointers, and for you to manually free the solver's memory when the Optimizer is garbage collected by Julia.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Never pass a pointer directly to a Julia ccall function.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Instead, store the pointer as a field in your Optimizer, and implement Base.cconvert and Base.unsafe_convert. Then you can pass Optimizer to any ccall function that expects the pointer.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In addition, make sure you implement a finalizer for each model you create.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If newsolver_createProblem() is the low-level function that creates the problem pointer in C, and newsolver_freeProblem(::Ptr{Cvoid}) is the low-level function that frees memory associated with the pointer, your Optimizer() function should look like this:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"struct Optimizer <: MOI.AbstractOptimizer\n    ptr::Ptr{Cvoid}\n\n    function Optimizer()\n        ptr = newsolver_createProblem()\n        model = Optimizer(ptr)\n        finalizer(model) do m\n            newsolver_freeProblem(m)\n            return\n        end\n        return model\n    end\nend\n\nBase.cconvert(::Type{Ptr{Cvoid}}, model::Optimizer) = model\nBase.unsafe_convert(::Type{Ptr{Cvoid}}, model::Optimizer) = model.ptr","category":"page"},{"location":"moi/tutorials/implementing/#Implement-methods-for-Optimizer","page":"Implementing a solver interface","title":"Implement methods for Optimizer","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All Optimizers must implement the following methods:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"empty!\nis_empty","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Other methods, detailed below, are optional or depend on how you implement the interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nFor this and all future methods, read the docstrings to understand what each method does, what it expects as input, and what it produces as output. If it isn't clear, let us know and we will improve the docstrings. It is also very helpful to look at an existing wrapper for a similar solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You should also implement Base.show(::IO, ::Optimizer) to print a nice string when someone prints your model. For example","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function Base.show(io::IO, model::Optimizer)\n    return print(io, \"NewSolver with the pointer $(model.ptr)\")\nend","category":"page"},{"location":"moi/tutorials/implementing/#Implement-attributes","page":"Implementing a solver interface","title":"Implement attributes","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface uses attributes to manage different aspects of the problem.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For each attribute","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"get gets the current value of the attribute\nset sets a new value of the attribute. Not all attributes can be set. For example, the user can't modify the SolverName.\nsupports returns a Bool indicating whether the solver supports the attribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nUse attribute_value_type to check the value expected by a given attribute. You should make sure that your get function correctly infers to this type (or a subtype of it).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Each column in the table indicates whether you need to implement the particular method for each attribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nSolverName Yes No No\nSolverVersion Yes No No\nRawSolver Yes No No\nName Yes Yes Yes\nSilent Yes Yes Yes\nTimeLimitSec Yes Yes Yes\nObjectiveLimit Yes Yes Yes\nRawOptimizerAttribute Yes Yes Yes\nNumberOfThreads Yes Yes Yes\nAbsoluteGapTolerance Yes Yes Yes\nRelativeGapTolerance Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.get(model::Optimizer, ::MOI.Silent)\n    return # true if MOI.Silent is set\nend\n\nfunction MOI.set(model::Optimizer, ::MOI.Silent, v::Bool)\n    if v\n        # Set a parameter to turn off printing\n    else\n        # Restore the default printing\n    end\n    return\nend\n\nMOI.supports(::Optimizer, ::MOI.Silent) = true","category":"page"},{"location":"moi/tutorials/implementing/#Define-supports_constraint","page":"Implementing a solver interface","title":"Define supports_constraint","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to define which constraints and objective functions you plan to support.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For each function-set constraint pair, define supports_constraint:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.supports_constraint(\n    ::Optimizer,\n    ::Type{MOI.VariableIndex},\n    ::Type{MOI.ZeroOne},\n)\n    return true\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"To make this easier, you may want to use Unions:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.supports_constraint(\n    ::Optimizer,\n    ::Type{MOI.VariableIndex},\n    ::Type{<:Union{MOI.LessThan,MOI.GreaterThan,MOI.EqualTo}},\n)\n    return true\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nOnly support a constraint if your solver has native support for it.","category":"page"},{"location":"moi/tutorials/implementing/#The-big-decision:-incremental-modification?","page":"Implementing a solver interface","title":"The big decision: incremental modification?","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Now you need to decide whether to support incremental modification or not.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Incremental modification means that the user can add variables and constraints one-by-one without needing to rebuild the entire problem, and they can modify the problem data after an optimize! call. Supporting incremental modification means implementing functions like add_variable and add_constraint.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The alternative is to accept the problem data in a single optimize! or copy_to function call. Because these functions see all of the data at once, it can typically call a more efficient function to load data into the underlying solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Good examples of solvers supporting incremental modification are MILP solvers like GLPK.jl and Gurobi.jl. Examples of non-incremental solvers are AmplNLWriter.jl and SCS.jl","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"It is possible for a solver to implement both approaches, but you should probably start with one for simplicity.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nOnly support incremental modification if your solver has native support for it.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In general, supporting incremental modification is more work, and it usually requires some extra book-keeping. However, it provides a more efficient interface to the solver if the problem is going to be resolved multiple times with small modifications. Moreover, once you've implemented incremental modification, it's usually not much extra work to add a copy_to interface. The converse is not true.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf this is your first time writing an interface, start with the one-shot optimize!.","category":"page"},{"location":"moi/tutorials/implementing/#The-non-incremental-interface","page":"Implementing a solver interface","title":"The non-incremental interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"There are two ways to implement the non-incremental interface. The first uses a two-argument version of optimize!, the second implements copy_to followed by the one-argument version of optimize!.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver does not support modification, and requires all data to solve the problem in a single function call, you should implement the \"one-shot\" optimize!.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"optimize!(::ModelLike, ::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver separates data loading and the actual optimization into separate steps, implement the copy_to interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"copy_to(::ModelLike, ::ModelLike)\noptimize!(::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/#The-incremental-interface","page":"Implementing a solver interface","title":"The incremental interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nWriting this interface is a lot of work. The easiest way is to consult the source code of a similar solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"To implement the incremental interface, implement the following functions:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"add_variable\nadd_variables\nadd_constraint\nadd_constraints\nis_valid\ndelete\noptimize!(::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nSolvers do not have to support AbstractScalarFunction in GreaterThan, LessThan, EqualTo, or Interval with a nonzero constant in the function. Throw ScalarFunctionConstantNotZero if the function constant is not zero.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In addition, you should implement the following model attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfModelAttributesSet Yes No No\nObjectiveFunctionType Yes No No\nObjectiveFunction Yes Yes Yes\nObjectiveSense Yes Yes Yes\nName Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Variable-related attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfVariableAttributesSet Yes No No\nNumberOfVariables Yes No No\nListOfVariableIndices Yes No No","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Constraint-related attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfConstraintAttributesSet Yes No No\nNumberOfConstraints Yes No No\nListOfConstraintTypesPresent Yes No No\nConstraintFunction Yes Yes No\nConstraintSet Yes Yes No","category":"page"},{"location":"moi/tutorials/implementing/#Modifications","page":"Implementing a solver interface","title":"Modifications","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver supports modifying data in-place, implement modify for the following AbstractModifications:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ScalarConstantChange\nScalarCoefficientChange\nVectorConstantChange\nMultirowChange","category":"page"},{"location":"moi/tutorials/implementing/#Variables-constrained-on-creation","page":"Implementing a solver interface","title":"Variables constrained on creation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Some solvers require variables be associated with a set when they are created. This conflicts with the incremental modification approach, since you cannot first add a free variable and then constrain it to the set.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If this is the case, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"add_constrained_variable\nadd_constrained_variables\nsupports_add_constrained_variables","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By default, MathOptInterface assumes solvers support free variables. If your solver does not support free variables, define:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI.supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false","category":"page"},{"location":"moi/tutorials/implementing/#Incremental-and-copy_to","page":"Implementing a solver interface","title":"Incremental and copy_to","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you implement the incremental interface, you have the option of also implementing copy_to.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you don't want to implement copy_to, for example, because the solver has no API for building the problem in a single function call, define the following fallback:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI.supports_incremental_interface(::Optimizer) = true\n\nfunction MOI.copy_to(dest::Optimizer, src::MOI.ModelLike)\n    return MOI.Utilities.default_copy_to(dest, src)\nend","category":"page"},{"location":"moi/tutorials/implementing/#implement_names","page":"Implementing a solver interface","title":"Names","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Regardless of which interface you implement, you have the option of implementing the Name attribute for variables and constraints:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nVariableName Yes Yes Yes\nConstraintName Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you implement names, you must also implement the following three methods:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.get(model::Optimizer, ::Type{MOI.VariableIndex}, name::String)\n    return # The variable named `name`.\nend\n\nfunction MOI.get(model::Optimizer, ::Type{MOI.ConstraintIndex}, name::String)\n    return # The constraint any type named `name`.\nend\n\nfunction MOI.get(\n    model::Optimizer,\n    ::Type{MOI.ConstraintIndex{F,S}},\n    name::String,\n) where {F,S}\n    return # The constraint of type F-in-S named `name`.\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"These methods have the following rules:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If there is no variable or constraint with the name, return nothing\nIf there is a single variable or constraint with that name, return the variable or constraint\nIf there are multiple variables or constraints with the name, throw an error.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nYou should not implement ConstraintName for VariableIndex constraints. If you implement ConstraintName for other constraints, you can add the following two methods to disable ConstraintName for VariableIndex constraints.function MOI.supports(\n    ::Optimizer,\n    ::MOI.ConstraintName,\n    ::Type{<:MOI.ConstraintIndex{MOI.VariableIndex,<:MOI.AbstractScalarSet}},\n)\n    return throw(MOI.VariableIndexConstraintNameError())\nend\nfunction MOI.set(\n    ::Optimizer,\n    ::MOI.ConstraintName,\n    ::MOI.ConstraintIndex{MOI.VariableIndex,<:MOI.AbstractScalarSet},\n    ::String,\n)\n    return throw(MOI.VariableIndexConstraintNameError())\nend","category":"page"},{"location":"moi/tutorials/implementing/#Solutions","page":"Implementing a solver interface","title":"Solutions","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Implement optimize! to solve the model:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"optimize!","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All Optimizers must implement the following attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"DualStatus\nPrimalStatus\nRawStatusString\nResultCount\nTerminationStatus","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nYou only need to implement get for solution attributes. Don't implement set or supports.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"note: Note\nSolver wrappers should document how the low-level statuses map to the MOI statuses. Statuses like NEARLY_FEASIBLE_POINT and INFEASIBLE_POINT, are designed to be used when the solver explicitly indicates that relaxed tolerances are satisfied or the returned point is infeasible, respectively.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You should also implement the following attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ObjectiveValue\nSolveTimeSec\nVariablePrimal","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nAttributes like VariablePrimal and ObjectiveValue are indexed by the result count. Use MOI.check_result_index_bounds(model, attr) to throw an error if the attribute is not available.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver returns dual solutions, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ConstraintDual\nDualObjectiveValue","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For integer solvers, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ObjectiveBound\nRelativeGap","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If applicable, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"SimplexIterations\nBarrierIterations\nNodeCount","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver uses the Simplex method, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ConstraintBasisStatus","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver accepts primal or dual warm-starts, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"VariablePrimalStart\nConstraintDualStart","category":"page"},{"location":"moi/tutorials/implementing/#Other-tips","page":"Implementing a solver interface","title":"Other tips","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Here are some other points to be aware of when writing your wrapper.","category":"page"},{"location":"moi/tutorials/implementing/#Unsupported-constraints-at-runtime","page":"Implementing a solver interface","title":"Unsupported constraints at runtime","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In some cases, your solver may support a particular type of constraint (for example, quadratic constraints), but only if the data meets some condition (for example, it is convex).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In this case, declare that you support the constraint, and throw AddConstraintNotAllowed.","category":"page"},{"location":"moi/tutorials/implementing/#Dealing-with-multiple-variable-bounds","page":"Implementing a solver interface","title":"Dealing with multiple variable bounds","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface uses VariableIndex constraints to represent variable bounds. Defining multiple variable bounds on a single variable is not allowed.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Throw LowerBoundAlreadySet or UpperBoundAlreadySet if the user adds a constraint that results in multiple bounds.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Only throw if the constraints conflict. It is okay to add VariableIndex-in-GreaterThan and then VariableIndex-in-LessThan, but not VariableIndex-in-Interval and then VariableIndex-in-LessThan,","category":"page"},{"location":"moi/tutorials/implementing/#Expect-duplicate-coefficients","page":"Implementing a solver interface","title":"Expect duplicate coefficients","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Solvers must expect that functions such as ScalarAffineFunction and VectorQuadraticFunction may contain duplicate coefficients.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, ScalarAffineFunction([ScalarAffineTerm(x, 1), ScalarAffineTerm(x, 1)], 0.0).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Use Utilities.canonical to return a new function with the duplicate coefficients aggregated together.","category":"page"},{"location":"moi/tutorials/implementing/#Don't-modify-user-data","page":"Implementing a solver interface","title":"Don't modify user-data","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All data passed to the solver must be copied immediately to internal data structures. Solvers may not modify any input vectors and must assume that input vectors will not be modified by users in the future.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"This applies, for example, to the terms vector in ScalarAffineFunction. Vectors returned to the user, for example, via ObjectiveFunction or ConstraintFunction attributes, must not be modified by the solver afterwards. The in-place version of get! can be used by users to avoid extra copies in this case.","category":"page"},{"location":"moi/tutorials/implementing/#Column-Generation","page":"Implementing a solver interface","title":"Column Generation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"There is no special interface for column generation. If the solver has a special API for setting coefficients in existing constraints when adding a new variable, it is possible to queue modifications and new variables and then call the solver's API once all of the new coefficients are known.","category":"page"},{"location":"moi/tutorials/implementing/#Solver-specific-attributes","page":"Implementing a solver interface","title":"Solver-specific attributes","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You don't need to restrict yourself to the attributes defined in the MathOptInterface.jl package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Solver-specific attributes should be specified by creating an appropriate subtype of AbstractModelAttribute, AbstractOptimizerAttribute, AbstractVariableAttribute, or AbstractConstraintAttribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, Gurobi.jl adds attributes for multiobjective optimization by defining:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"struct NumberOfObjectives <: MOI.AbstractModelAttribute end\n\nfunction MOI.set(model::Optimizer, ::NumberOfObjectives, n::Integer)\n    # Code to set NumberOfObjectives\n    return\nend\n\nfunction MOI.get(model::Optimizer, ::NumberOfObjectives)\n    n = # Code to get NumberOfObjectives\n    return n\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Then, the user can write:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"model = Gurobi.Optimizer()\nMOI.set(model, Gurobi.NumberofObjectives(), 3)","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Benchmarks/reference.md\"","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Benchmarks/reference/#Benchmarks","page":"API Reference","title":"Benchmarks","text":"","category":"section"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"Functions to help benchmark the performance of solver wrappers. See The Benchmarks submodule for more details.","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"Benchmarks.suite\nBenchmarks.create_baseline\nBenchmarks.compare_against_baseline","category":"page"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.suite","page":"API Reference","title":"MathOptInterface.Benchmarks.suite","text":"suite(\n    new_model::Function;\n    exclude::Vector{Regex} = Regex[]\n)\n\nCreate a suite of benchmarks. new_model should be a function that takes no arguments, and returns a new instance of the optimizer you wish to benchmark.\n\nUse exclude to exclude a subset of benchmarks.\n\nExamples\n\nsuite() do\n    GLPK.Optimizer()\nend\nsuite(exclude = [r\"delete\"]) do\n    Gurobi.Optimizer(OutputFlag=0)\nend\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.create_baseline","page":"API Reference","title":"MathOptInterface.Benchmarks.create_baseline","text":"create_baseline(suite, name::String; directory::String = \"\"; kwargs...)\n\nRun all benchmarks in suite and save to files called name in directory.\n\nExtra kwargs are based to BenchmarkTools.run.\n\nExamples\n\nmy_suite = suite(() -> GLPK.Optimizer())\ncreate_baseline(my_suite, \"glpk_master\"; directory = \"/tmp\", verbose = true)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.compare_against_baseline","page":"API Reference","title":"MathOptInterface.Benchmarks.compare_against_baseline","text":"compare_against_baseline(\n    suite, name::String; directory::String = \"\",\n    report_filename::String = \"report.txt\"\n)\n\nRun all benchmarks in suite and compare against files called name in directory that were created by a call to create_baseline.\n\nA report summarizing the comparison is written to report_filename in directory.\n\nExtra kwargs are based to BenchmarkTools.run.\n\nExamples\n\nmy_suite = suite(() -> GLPK.Optimizer())\ncompare_against_baseline(\n    my_suite, \"glpk_master\"; directory = \"/tmp\", verbose = true\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/bridging_constraint.md\"","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Implementing-a-constraint-bridge","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This guide outlines the basic steps to create a new bridge from a constraint expressed in the formalism Function-in-Set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Preliminaries","page":"Implementing a constraint bridge","title":"Preliminaries","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"First, decide on the set you want to bridge. Then, study its properties: the most important one is whether the set is scalar or vector, which impacts the dimensionality of the functions that can be used with the set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A scalar function only has one dimension. MOI defines three types of scalar functions: a variable (VariableIndex), an affine function (ScalarAffineFunction), or a quadratic function (ScalarQuadraticFunction).\nA vector function has several dimensions (at least one). MOI defines three types of vector functions: several variables (VectorOfVariables), an affine function (VectorAffineFunction), or a quadratic function (VectorQuadraticFunction). The main difference with scalar functions is that the order of dimensions can be very important: for instance, in an indicator constraint (Indicator), the first dimension indicates whether the constraint about the second dimension is active.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To explain how to implement a bridge, we present the example of Bridges.Constraint.FlipSignBridge. This bridge maps <= (LessThan) constraints to >= (GreaterThan) constraints. This corresponds to reversing the sign of the inequality. We focus on scalar affine functions (we disregard the cases of a single variable or of quadratic functions). This example is a simplified version of the code included in MOI.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Four-mandatory-parts-in-a-constraint-bridge","page":"Implementing a constraint bridge","title":"Four mandatory parts in a constraint bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The first part of a constraint bridge is a new concrete subtype of Bridges.Constraint.AbstractBridge. This type must have fields to store all the new variables and constraints that the bridge will add. Typically, these types are parametrized by the type of the coefficients in the model.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Then, three sets of functions must be defined:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Bridges.Constraint.bridge_constraint: this function implements the bridge and creates the required variables and constraints.\nsupports_constraint: these functions must return true when the combination of function and set is supported by the bridge. By default, the base implementation always returns false and the bridge does not have to provide this implementation.\nBridges.added_constrained_variable_types and Bridges.added_constraint_types: these functions return the types of variables and constraints that this bridge adds. They are used to compute the set of other bridges that are required to use the one you are defining, if need be.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"More functions can be implemented, for instance to retrieve properties from the bridge or deleting a bridged constraint.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Structure-for-the-bridge","page":"Implementing a constraint bridge","title":"1. Structure for the bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A typical struct behind a bridge depends on the type of the coefficients that are used for the model (typically Float64, but coefficients might also be integers or complex numbers).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This structure must hold a reference to all the variables and the constraints that are created as part of the bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The type of this structure is used throughout MOI as an identifier for the bridge. It is passed as argument to most functions related to bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The best practice is to have the name of this type end with Bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, the bridge maps any ScalarAffineFunction{T}-in-LessThan{T} constraint to a single ScalarAffineFunction{T}-in-GreaterThan{T} constraint. The affine function has coefficients of type T. The bridge is parametrized with T, so that the constraint that the bridge creates also has coefficients of type T.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"struct SignBridge{T<:Number} <: Bridges.Constraint.AbstractBridge\n    constraint::ConstraintIndex{ScalarAffineFunction{T}, GreaterThan{T}}\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Bridge-creation","page":"Implementing a constraint bridge","title":"2. Bridge creation","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The function Bridges.Constraint.bridge_constraint is called whenever the bridge is instantiated for a specific model, with the given function and set. The arguments to bridge_constraint are similar to add_constraint, with the exception of the first argument: it is the Type of the struct defined in the first step (for our example, Type{SignBridge{T}}).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"bridge_constraint returns an instance of the struct defined in the first step. the first step.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, the bridge constraint could be defined as:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.Constraint.bridge_constraint(\n    ::Type{SignBridge{T}}, # Bridge to use.\n    model::ModelLike, # Model to which the constraint is being added.\n    f::ScalarAffineFunction{T}, # Function to rewrite.\n    s::LessThan{T}, # Set to rewrite.\n) where {T}\n    # Create the variables and constraints required for the bridge.\n    con = add_constraint(model, -f, GreaterThan(-s.upper))\n\n    # Return an instance of the bridge type with a reference to all the\n    # variables and constraints that were created in this function.\n    return SignBridge(con)\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Supported-constraint-types","page":"Implementing a constraint bridge","title":"3. Supported constraint types","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The function supports_constraint determines whether the bridge type supports a given combination of function and set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This function must closely match bridge_constraint, because it will not be called if supports_constraint returns false.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function supports_constraint(\n    ::Type{SignBridge{T}}, # Bridge to use.\n    ::Type{ScalarAffineFunction{T}}, # Function to rewrite.\n    ::Type{LessThan{T}}, # Set to rewrite.\n) where {T}\n    # Do some computation to ensure that the constraint is supported.\n    # Typically, you can directly return true.\n    return true\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Metadata-about-the-bridge","page":"Implementing a constraint bridge","title":"4. Metadata about the bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To determine whether a bridge can be used, MOI uses a shortest-path algorithm that uses the variable types and the constraints that the bridge can create. This information is communicated from the bridge to MOI using the functions Bridges.added_constrained_variable_types and Bridges.added_constraint_types. Both return lists of tuples: either a list of 1-tuples containing the variable types (typically, ZeroOne or Integer) or a list of 2-tuples contained the functions and sets (like ScalarAffineFunction{T}-GreaterThan).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For our example, the bridge does not create any constrained variables, and only ScalarAffineFunction{T}-in-GreaterThan{T} constraints:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.added_constrained_variable_types(::Type{SignBridge{T}}) where {T}\n    # The bridge does not create variables, return an empty list of tuples:\n    return Tuple{Type}[]\nend\n\nfunction Bridges.added_constraint_types(::Type{SignBridge{T}}) where {T}\n    return Tuple{Type,Type}[\n        # One element per F-in-S the bridge creates.\n        (ScalarAffineFunction{T}, GreaterThan{T}),\n    ]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A bridge that creates binary variables would rather have this definition of added_constrained_variable_types:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.added_constrained_variable_types(::Type{SomeBridge{T}}) where {T}\n    # The bridge only creates binary variables:\n    return Tuple{Type}[(ZeroOne,)]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nIf you declare the creation of constrained variables in added_constrained_variable_types, the corresponding constraint type VariableIndex must not be indicated in added_constraint_types. This would restrict the use of the bridge to solvers that can add such a constraint after the variable is created.More concretely, if you declare in added_constrained_variable_types that your bridge creates binary variables (ZeroOne), and if you never add such a constraint afterward (you do not call add_constraint(model, var, ZeroOne())), then you must not list (VariableIndex, ZeroOne) in added_constraint_types.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Typically, the function Bridges.Constraint.concrete_bridge_type does not have to be defined for most bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-registration","page":"Implementing a constraint bridge","title":"Bridge registration","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For a bridge to be used by MOI, it must be known by MOI.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#SingleBridgeOptimizer","page":"Implementing a constraint bridge","title":"SingleBridgeOptimizer","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The first way to do so is to create a single-bridge optimizer. This type of optimizer wraps another optimizer and adds the possibility to use only one bridge. It is especially useful when unit testing bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"It is common practice to use the same name as the type defined for the bridge (SignBridge, in our example) without the suffix Bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"const Sign{T,OT<: ModelLike} =\n    SingleBridgeOptimizer{SignBridge{T}, OT}","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In the context of unit tests, this bridge is used in conjunction with a Utilities.MockOptimizer:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"mock = Utilities.MockOptimizer(\n    Utilities.UniversalFallback(Utilities.Model{Float64}()),\n)\nbridged_mock = Sign{Float64}(mock)","category":"page"},{"location":"moi/tutorials/bridging_constraint/#New-bridge-for-a-LazyBridgeOptimizer","page":"Implementing a constraint bridge","title":"New bridge for a LazyBridgeOptimizer","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Typical user-facing models for MOI are based on Bridges.LazyBridgeOptimizer. For instance, this type of model is returned by Bridges.full_bridge_optimizer. These models can be added more bridges by using Bridges.add_bridge:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"inner_optimizer = Utilities.Model{Float64}()\noptimizer = Bridges.full_bridge_optimizer(inner_optimizer, Float64)\nBridges.add_bridge(optimizer, SignBridge{Float64})","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-improvements","page":"Implementing a constraint bridge","title":"Bridge improvements","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/#Attribute-retrieval","page":"Implementing a constraint bridge","title":"Attribute retrieval","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Like models, bridges have attributes that can be retrieved using get and set. The most important ones are the number of variables and constraints, but also the lists of variables and constraints.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, we only have one constraint and only have to implement the NumberOfConstraints and ListOfConstraintIndices attributes:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    ::SignBridge{T},\n    ::NumberOfConstraints{\n        ScalarAffineFunction{T},\n        GreaterThan{T},\n    },\n) where {T}\n    return 1\nend\n\nfunction get(\n    bridge::SignBridge{T},\n    ::ListOfConstraintIndices{\n        ScalarAffineFunction{T},\n        GreaterThan{T},\n    },\n) where {T}\n    return [bridge.constraint]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"You must implement one such pair of functions for each type of constraint the bridge adds to the model.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nAvoid returning a list from the bridge object without copying it. Users must be able to change the contents of the returned list without altering the bridge object.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For variables, the situation is simpler. If your bridge creates new variables, you must implement the NumberOfVariables and ListOfVariableIndices attributes. However, these attributes do not have parameters, unlike their constraint counterparts. Only two functions suffice:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    ::SignBridge{T},\n    ::NumberOfVariables,\n) where {T}\n    return 0\nend\n\nfunction get(\n    ::SignBridge{T},\n    ::ListOfVariableIndices,\n) where {T}\n    return VariableIndex[]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In order for the user to be able to access the function and set of the original constraint, the bridge needs to implement getters for the ConstraintFunction and ConstraintSet attributes:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    model::MOI.ModelLike,\n    attr::MOI.ConstraintFunction,\n    bridge::SignBridge,\n)\n    return -MOI.get(model, attr, bridge.constraint)\nend\n\nfunction get(\n    model::MOI.ModelLike,\n    attr::MOI.ConstraintSet,\n    bridge::SignBridge,\n)\n    set = MOI.get(model, attr, bridge.constraint)\n    return MOI.LessThan(-set.lower)\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nAlternatively, one could store the original function and set in SignBridge during Bridges.Constraint.bridge_constraint to make these getters simpler and more efficient. On the other hand, this will increase the memory footprint of the bridges as the garbage collector won't be able to delete that object. The convention is to not store the function in the bridge and not care too much about the efficiency of these getters. If the user needs efficient getters for ConstraintFunction then they should use a Utilities.CachingOptimizer.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Model-modifications","page":"Implementing a constraint bridge","title":"Model modifications","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To avoid copying the model when the user request to change a constraint, MOI provides modify. Bridges can also implement this API to allow certain changes, such as coefficient changes.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our case, a modification of a coefficient in the original constraint (for example, replacing the value of the coefficient of a variable in the affine function) must be transmitted to the constraint created by the bridge, but with a sign change.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function modify(\n    model::ModelLike,\n    bridge::SignBridge,\n    change::ScalarCoefficientChange,\n)\n    modify(\n        model,\n        bridge.constraint,\n        ScalarCoefficientChange(change.variable, -change.new_coefficient),\n    )\n    return\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-deletion","page":"Implementing a constraint bridge","title":"Bridge deletion","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"When a bridge is deleted, the constraints it added must be deleted too.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function delete(model::ModelLike, bridge::SignBridge)\n    delete(model, bridge.constraint)\n    return\nend","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"EditURL = \"experiment_design.jl\"","category":"page"},{"location":"tutorials/conic/experiment_design/#Experiment-design","page":"Experiment design","title":"Experiment design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial was originally contributed by Arpit Bhatia and Chris Coey.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial covers experiment design examples (D-optimal, A-optimal, and E-optimal) from section 7.5 of the book Convex Optimization by Boyd and Vandenberghe.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The tutorial uses the following packages","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"using JuMP\nimport SCS\nimport LinearAlgebra\nimport Random","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"We set a seed so the random numbers are repeatable:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Random.seed!(1234)","category":"page"},{"location":"tutorials/conic/experiment_design/#The-relaxed-experiment-design-problem","page":"Experiment design","title":"The relaxed experiment design problem","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The basic experiment design problem is as follows.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Given the menu of possible choices for experiments, v_1 ldots v_p, and the total number m of experiments to be carried out, choose the numbers of each type of experiment, that is, m_1 ldots m_p to make the error covariance E small (in  some sense).","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The variables m_1 ldots m_p must, of course, be integers and sum to m the given total number of experiments. This leads to the optimization problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nminleft(mathrmwrt mathbfS_+^nright)  E=left(sum_j=1^p m_j v_j v_j^Tright)^-1 \ntextsubject to  m_i geq 0 \n sumlimits_i=1^p m_i = m \n m_i in mathbbZquad i=1ldotsp\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The basic experiment design problem can be a hard combinatorial problem when m, the total number of experiments, is comparable to n, since in this case the m_i are all small integers.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In the case when m is large compared to n, however, a good approximate solution can be found by ignoring, or relaxing, the constraint that the m_i are integers.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Let lambda_i=m_i  m which is the fraction of the total number of experiments for which a_j=v_i or the relative frequency of experiment i. We can express the error covariance in terms of lambda_i as:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"E=frac1mleft(sum_i=1^p lambda_i v_i v_i^Tright)^-1","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The vector lambda in mathbfR^p satisfies lambda succeq 0 mathbf1^T lambda=1 and also, each lambda_i is an integer multiple of 1  m. By ignoring this last constraint, we arrive at the problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nminleft(mathrmwrt mathbfS_+^nright)  E=(1  m)left(sum_i=1^p lambda_i v_i v_i^Tright)^-1 \ntextsubject toquad  lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Several scalarizations have been proposed for the experiment design problem, which is a vector optimization problem over the positive semidefinite cone.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"q = 4 # dimension of estimate space\np = 8 # number of experimental vectors\nn_max = 3 # upper bound on lambda\nn = 12\n\nV = randn(q, p)\n\neye = Matrix{Float64}(LinearAlgebra.I, q, q);\nnothing #hide","category":"page"},{"location":"tutorials/conic/experiment_design/#A-optimal-design","page":"Experiment design","title":"A-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In A-optimal experiment design, we minimize tr E, the trace of the covariance matrix. This objective is simply the mean of the norm of the error squared:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"mathbfEe_2^2=mathbfE operatornametrleft(e e^Tright)=operatornametr E","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The A-optimal experiment design problem in SDP form is","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  mathbf1^T u \ntextsubject to  leftbeginalignedsum_i=1^p lambda_i v_i v_i^T  e_k  e_k^T  u_kendalignedright succeq 0 quad k=1 ldots n \n lambda succeq 0 \n  mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"aOpt = Model(SCS.Optimizer)\nset_silent(aOpt)\n@variable(aOpt, np[1:p], lower_bound = 0, upper_bound = n_max)\n@variable(aOpt, u[1:q], lower_bound = 0)\n@constraint(aOpt, sum(np) <= n)\nfor i in 1:q\n    matrix = [\n        V*LinearAlgebra.diagm(0 => np ./ n)*V' eye[:, i]\n        eye[i, :]' u[i]\n    ]\n    @constraint(aOpt, matrix >= 0, PSDCone())\nend\n@objective(aOpt, Min, sum(u))\noptimize!(aOpt)\nobjective_value(aOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"tutorials/conic/experiment_design/#E-optimal-design","page":"Experiment design","title":"E-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In E -optimal design, we minimize the norm of the error covariance matrix, that is, the maximum eigenvalue of E.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Since the diameter (twice the longest semi-axis) of the confidence ellipsoid mathcalE is proportional to E_2^1  2, minimizing E_2 can be interpreted geometrically as minimizing the diameter of the confidence ellipsoid.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"E-optimal design can also be interpreted as minimizing the maximum variance of q^T e, over all q with q_2=1. The E-optimal experiment design problem in SDP form is:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  t \ntextsubject to  sum_i=1^p lambda_i v_i v_i^T succeq t I \n lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"eOpt = Model(SCS.Optimizer)\nset_silent(eOpt)\n@variable(eOpt, 0 <= np[1:p] <= n_max)\n@variable(eOpt, t)\n@constraint(\n    eOpt,\n    V * LinearAlgebra.diagm(0 => np ./ n) * V' - (t .* eye) >= 0,\n    PSDCone(),\n)\n@constraint(eOpt, sum(np) <= n)\n@objective(eOpt, Max, t)\noptimize!(eOpt)\nobjective_value(eOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"tutorials/conic/experiment_design/#D-optimal-design","page":"Experiment design","title":"D-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The most widely used scalarization is called D -optimal design, in which we minimize the determinant of the error covariance matrix E. This corresponds to designing the experiment to minimize the volume of the resulting confidence ellipsoid (for a fixed confidence level). Ignoring the constant factor 1  m in E, and taking the logarithm of the objective, we can pose this problem as convex optimization problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  log operatornamedetleft(sum_i=1^p lambda_i v_i v_i^Tright)^-1 \ntextsubject to  lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"dOpt = Model(SCS.Optimizer)\nset_silent(dOpt)\n@variable(dOpt, np[1:p], lower_bound = 0, upper_bound = n_max)\n@variable(dOpt, t)\n@objective(dOpt, Max, t)\n@constraint(dOpt, sum(np) <= n)\nE = V * LinearAlgebra.diagm(0 => np ./ n) * V'\n@constraint(dOpt, [t; 1; triangle_vec(E)] in MOI.LogDetConeTriangle(q))\noptimize!(dOpt)\nobjective_value(dOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/nonlinear/#Nonlinear-Modeling","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nThis page describes a new nonlinear interface to JuMP. It replaces the legacy @NL interface, which is documented at Nonlinear Modeling (Legacy). The API described below is stable, and it will not break with future 1.X releases of JuMP. However, solver support may be limited, and there may be gaps in functionality compared with the legacy interface. To report a bug, or request a missing feature, please open an issue.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP has support for nonlinear (convex and nonconvex) optimization problems. JuMP is able to automatically provide exact, sparse second-order derivatives to solvers. This information can improve solver accuracy and performance.","category":"page"},{"location":"manual/nonlinear/#Set-a-nonlinear-objective","page":"Nonlinear Modeling","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @objective to set a nonlinear objective.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, exp(x[1]) - sqrt(x[2]))\nexp(x[1]) - sqrt(x[2])","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"To modify a nonlinear objective, call @objective again.","category":"page"},{"location":"manual/nonlinear/#Add-a-nonlinear-constraint","page":"Nonlinear Modeling","title":"Add a nonlinear constraint","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @constraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, exp(x[1]) <= 1)\nexp(x[1]) - 1.0 ≤ 0\n\njulia> @constraint(model, con[i = 1:2], 2^x[i] >= i)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarNonlinearFunction, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n con[1] : (2.0 ^ x[1]) - 1.0 ≥ 0\n con[2] : (2.0 ^ x[2]) - 2.0 ≥ 0","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Delete a nonlinear constraint using delete:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> delete(model, con[1])","category":"page"},{"location":"manual/nonlinear/#Create-a-nonlinear-expression","page":"Nonlinear Modeling","title":"Create a nonlinear expression","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @expression to create nonlinear expression objects:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> expr = @expression(model, exp(x[1]) + sqrt(x[2]))\nexp(x[1]) + sqrt(x[2])\n\njulia> my_anon_expr = @expression(model, [i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpr}:\n sin(x[1])\n sin(x[2])\n\njulia> @expression(model, my_expr[i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpr}:\n sin(x[1])\n sin(x[2])","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"A NonlinearExpr can be used in @objective, @constraint, and even nested in other @expressions.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> @objective(model, Min, expr^2 + 1)\n((exp(x[1]) + sqrt(x[2])) ^ 2.0) + 1.0\n\njulia> @constraint(model, [i = 1:2], my_expr[i] <= i)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarNonlinearFunction, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n sin(x[1]) - 1.0 ≤ 0\n sin(x[2]) - 2.0 ≤ 0\n\njulia> @expression(model, nested[i = 1:2], sin(my_expr[i]))\n2-element Vector{NonlinearExpr}:\n sin(sin(x[1]))\n sin(sin(x[2]))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use value to query the value of a nonlinear expression:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> set_start_value(x[1], 1.0)\n\njulia> value(start_value, nested[1])\n0.7456241416655579\n\njulia> sin(sin(1.0))\n0.7456241416655579","category":"page"},{"location":"manual/nonlinear/#Nonlinear-expressions-in-detail","page":"Nonlinear Modeling","title":"Nonlinear expressions in detail","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions in JuMP are represented by a NonlinearExpr object.","category":"page"},{"location":"manual/nonlinear/#Constructors","page":"Nonlinear Modeling","title":"Constructors","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions can be created using the NonlinearExpr constructors:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = NonlinearExpr(:sin, Any[x])\nsin(x)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"or via operator overloading:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = sin(x)\nsin(x)","category":"page"},{"location":"manual/nonlinear/#Supported-arguments","page":"Nonlinear Modeling","title":"Supported arguments","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions can contain a mix of numbers, AffExpr, QuadExpr, and other NonlinearExpr:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> aff = x + 1;\n\njulia> quad = x^2 + x;\n\njulia> expr = cos(x) * sin(quad) + aff\n(cos(x) * sin(x² + x)) + (x + 1)","category":"page"},{"location":"manual/nonlinear/#Supported-operators","page":"Nonlinear Modeling","title":"Supported operators","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The list of supported operators may vary between solvers. Given an optimizer, query the list of supported operators using MOI.ListOfSupportedNonlinearOperators:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> import Ipopt\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.get(Ipopt.Optimizer(), MOI.ListOfSupportedNonlinearOperators())\n85-element Vector{Symbol}:\n :+\n :-\n :abs\n :sqrt\n :cbrt\n :abs2\n :inv\n :log\n :log10\n :log2\n ⋮\n :min\n :max\n :&&\n :||\n :<=\n :(==)\n :>=\n :<\n :>","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"In some univariate cases, the operator is defined in SpecialFunctions.jl. To use these functions, you must explicitly import SpecialFunctions.jl","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> import Ipopt\n\njulia> op = MOI.get(Ipopt.Optimizer(), MOI.ListOfSupportedNonlinearOperators());\n\njulia> :erfcx in op\ntrue\n\njulia> :dawson in op\ntrue\n\njulia> import SpecialFunctions\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, SpecialFunctions.erfcx(x))\nerfcx(x)\n\njulia> @expression(model, SpecialFunctions.dawson(x))\ndawson(x)","category":"page"},{"location":"manual/nonlinear/#Limitations","page":"Nonlinear Modeling","title":"Limitations","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Some nonlinear expressions cannot be created via operator overloading. For example, to minimize the likelihood of bugs in user-code, we have not overloaded comparisons such as < and >= between JuMP objects:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> x < 1\nERROR: Cannot evaluate `<` between a variable and a number.\n[...]","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Instead, wrap the expression in the @expression macro:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = @expression(model, x < 1)\nx < 1","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For technical reasons, other operators that are not overloaded include ||, &&, and ifelse.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = @expression(model, ifelse(x < -1 || x >= 1, x^2, 0.0))\nifelse((x < -1) || (x >= 1), x², 0.0)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"As an alternative, use the JuMP.op_ functions, which fallback to the various comparison and logical operators:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = op_ifelse(\n           op_or(op_strictly_less_than(x, -1), op_greater_than_or_equal_to(x, 1)),\n           x^2,\n           0.0,\n       )\nifelse((x < -1) || (x >= 1), x², 0.0)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The available functions are:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP function Julia function\nop_ifelse ifelse\nop_and &&\nop_or ||\nop_greater_than_or_equal_to >=\nop_less_than_or_equal_to <=\nop_equal_to ==\nop_strictly_greater_than >\nop_strictly_less_than <","category":"page"},{"location":"manual/nonlinear/#Fields","page":"Nonlinear Modeling","title":"Fields","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Each NonlinearExpr has two fields.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The .head field is a Symbol that represents the operator being called:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> expr.head\n:sin","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The .args field is a Vector{Any} containing the arguments to the operator:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> expr.args\n1-element Vector{Any}:\n x","category":"page"},{"location":"manual/nonlinear/#jump_user_defined_operators","page":"Nonlinear Modeling","title":"User-defined operators","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"In addition to a standard list of univariate and multivariate operators recognized by the MOI.Nonlinear submodule, JuMP supports user-defined Julia operators.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nUser-defined operators must return a scalar output. For a work-around, see User-defined operators with vector outputs.","category":"page"},{"location":"manual/nonlinear/#Add-an-operator","page":"Nonlinear Modeling","title":"Add an operator","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Add a user-defined operator using the @operator macro:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\nmodel = Model();\n@operator(model, op_square, 1, square)\n@operator(model, op_f, 2, f)\n@variable(model, x[1:2]);\n@objective(model, Min, op_f(x[1], op_square(x[2])))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The arguments to @operator are:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The model to which the operator is added.\nA Julia symbol object which serves as the name of the user-defined operator in JuMP expressions. This name must not be the same as that of the function.\nThe number of scalar input arguments that the function takes.\nA Julia method which computes the function.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nUser-defined operators cannot be deleted.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"You can obtain a reference to the operator using the model[:key] syntax:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nmodel = Model();\n@operator(model, op_square, 1, square)\nop_square_2 = model[:op_square]","category":"page"},{"location":"manual/nonlinear/#Add-an-operator-without-macros","page":"Nonlinear Modeling","title":"Add an operator without macros","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The @operator macro is syntactic sugar for add_nonlinear_operator. Thus, the non-macro version of the preceding example is:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\nmodel = Model();\nop_square = add_nonlinear_operator(model, 1, square; name = :op_square)\nmodel[:op_square] = op_square\nop_f = add_nonlinear_operator(model, 2, f; name = :op_f)\nmodel[:op_f] = op_f\n@variable(model, x[1:2]);\n@objective(model, Min, op_f(x[1], op_square(x[2])))","category":"page"},{"location":"manual/nonlinear/#Operators-with-the-same-name-as-an-existing-function","page":"Nonlinear Modeling","title":"Operators with the same name as an existing function","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"A common error encountered is the following:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> using JuMP\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> @operator(model, f, 1, f)\nERROR: Unable to add the nonlinear operator `:f` with the same name as\nan existing function.\n[...]","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"This error occurs because @operator(model, f, 1, f) is equivalent to:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> f = add_nonlinear_operator(model, 1, f; name = :f)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"but f already exists as a Julia function.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"If you evaluate the function without adding it as an operator, JuMP will trace the function using operator overloading:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> @variable(model, x);\n\njulia> f(x)\nx²","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"To force JuMP to treat f as a user-defined operator and not trace it, add the operator using add_nonlinear_operator and define a new method which manually creates a NonlinearExpr:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> _ = add_nonlinear_operator(model, 1, f; name = :f)\nNonlinearOperator(f, :f)\n\njulia> f(x::AbstractJuMPScalar) = NonlinearExpr(:f, Any[x])\nf (generic function with 2 methods)\n\njulia> @expression(model, log(f(x)))\nlog(f(x))","category":"page"},{"location":"manual/nonlinear/#Gradients-and-Hessians","page":"Nonlinear Modeling","title":"Gradients and Hessians","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"By default, JuMP will use automatic differentiation to compute the gradient and Hessian of user-defined operators. If your function is not amenable to automatic differentiation, or you can compute analytic derivatives, you may pass additional arguments to @operator to compute the first- and second-derivatives.","category":"page"},{"location":"manual/nonlinear/#Univariate-functions","page":"Nonlinear Modeling","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For univariate functions, a gradient function ∇f returns a number that represents the first-order derivative. You may, in addition, pass a third function which returns a number representing the second-order derivative:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nf(x) = x^2\n∇f(x) = 2x\n∇²f(x) = 2\nmodel = Model();\n@operator(model, op_f, 1, f, ∇f, ∇²f)  # Providing ∇²f is optional\n@variable(model, x)\n@objective(model, Min, op_f(x))","category":"page"},{"location":"manual/nonlinear/#Multivariate-functions","page":"Nonlinear Modeling","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For multivariate functions, the gradient function ∇f must take an AbstractVector as the first argument that is filled in-place. The Hessian function, ∇²f, must take an AbstractMatrix as the first argument, the lower-triangular of which is filled in-place:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nf(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2\nfunction ∇f(g::AbstractVector{T}, x::T...) where {T}\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend\nfunction ∇²f(H::AbstractMatrix{T}, x::T...) where {T}\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1]  <-- Not needed. Fill the lower-triangular only.\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend\nmodel = Model();\n@operator(model, rosenbrock, 2, f, ∇f, ∇²f)  # Providing ∇²f is optional\n@variable(model, x[1:2])\n@objective(model, Min, rosenbrock(x[1], x[2]))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"You may assume the Hessian matrix H is initialized with zeros, and because H is symmetric, you need only to fill in the non-zero lower-triangular terms. The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). Moreover, you may assume only that H supports size(H) and setindex!. Finally, the matrix is treated as dense, so the performance will be poor on functions with high-dimensional input.","category":"page"},{"location":"manual/nonlinear/#User-defined-operators-with-vector-inputs","page":"Nonlinear Modeling","title":"User-defined operators with vector inputs","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"User-defined operators which take vectors as input arguments (for example, f(x::Vector)) are not supported. Instead, use Julia's splatting syntax to create a function with scalar arguments. For example, instead of:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x::Vector) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"define:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x...) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Another approach is to define the splatted function as an anonymous function:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nmodel = Model();\n@variable(model, x[1:5])\nf(x::Vector) = sum(x[i]^i for i in 1:length(x))\n@operator(model, op_f, 5, (x...) -> f(collect(x)))\n@objective(model, Min, op_f(x...))","category":"page"},{"location":"manual/nonlinear/#Automatic-differentiation","page":"Nonlinear Modeling","title":"Automatic differentiation","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP does not support black-box optimization, so all user-defined operators must provide derivatives in some form. Fortunately, JuMP supports automatic differentiation of user-defined operators.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"info: Info\nAutomatic differentiation is not finite differencing. JuMP's automatically computed derivatives are not subject to approximation error.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP uses ForwardDiff.jl to perform automatic differentiation of user-defined operators; see the ForwardDiff.jl documentation for a description of how to write a function suitable for automatic differentiation.","category":"page"},{"location":"manual/nonlinear/#Common-mistakes-when-writing-a-user-defined-operator","page":"Nonlinear Modeling","title":"Common mistakes when writing a user-defined operator","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nGet an error like No method matching Float64(::ForwardDiff.Dual)? Read this section, and see the guidelines at ForwardDiff.jl.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The most common error is that your user-defined operator is not generic with respect to the number type, that is, don't assume that the input to the function is Float64.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x::Float64) = 2 * x  # This will not work.\nf(x::Real)    = 2 * x  # This is good.\nf(x)          = 2 * x  # This is also good.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Another reason you may encounter this error is if you create arrays inside your function which are Float64.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"function bad_f(x...)\n    y = zeros(length(x))  # This constructs an array of `Float64`!\n    for i in 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend\n\nfunction good_f(x::T...) where {T<:Real}\n    y = zeros(T, length(x))  # Construct an array of type `T` instead!\n    for i in 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/naming_conventions.md\"","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/naming_conventions/#Naming-conventions","page":"Naming conventions","title":"Naming conventions","text":"","category":"section"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"MOI follows several conventions for naming functions and structures. These should also be followed by packages extending MOI.","category":"page"},{"location":"moi/background/naming_conventions/#Sets","page":"Naming conventions","title":"Sets","text":"","category":"section"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"Sets encode the structure of constraints. Their names should follow the following conventions:","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"Abstract types in the set hierarchy should begin with Abstract and end in Set, for example, AbstractScalarSet, AbstractVectorSet.\nVector-valued conic sets should end with Cone, for example, NormInfinityCone, SecondOrderCone.\nVector-valued Cartesian products should be plural and not end in Cone, for example, Nonnegatives, not NonnegativeCone.\nMatrix-valued conic sets should provide two representations: ConeSquare and ConeTriangle, for example, RootDetConeTriangle and RootDetConeSquare. See Matrix cones for more details.\nScalar sets should be singular, not plural, for example, Integer, not Integers.\nAs much as possible, the names should follow established conventions in the domain where this set is used: for instance, convex sets should have names close to those of CVX, and constraint-programming sets should follow MiniZinc's constraints.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"EditURL = \"https://github.com/oxfordcontrol/CDCS.jl/blob/6d165e8e8f59b14e36e99c8efb19f0e775bbd292/README.md\"","category":"page"},{"location":"packages/CDCS/#CDCS.jl","page":"oxfordcontrol/CDCS.jl","title":"CDCS.jl","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"CDCS.jl is a wrapper for the CDCS solver. ","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"an exported cdcs function that is a thin wrapper on top of the cdcs MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/CDCS/#License","page":"oxfordcontrol/CDCS.jl","title":"License","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"CDCS.jl is licensed under the MIT license.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The underlying solver oxfordcontrol/CDCS is licensed under the LGPL v3 license.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"In addition, CDCS requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/CDCS/#Installation","page":"oxfordcontrol/CDCS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package and that the CDCS software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"Then, install CDCS.jl using Pkg.add:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"import Pkg\nPkg.add(\"CDCS\")","category":"page"},{"location":"packages/CDCS/#Use-with-JuMP","page":"oxfordcontrol/CDCS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"To use CDCS with JuMP, do:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"using JuMP, CDCS\nmodel = Model(CDCS.Optimizer)\nset_attribute(model, \"verbose\", 0)","category":"page"},{"location":"packages/CDCS/#Troubleshooting","page":"oxfordcontrol/CDCS.jl","title":"Troubleshooting","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"If you get the error:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"Undefined function or variable 'cdcs'.\n\nError using save\nVariable 'jx_cdcs_arg_out_1' not found.\n\nLinear Programming example: Error During Test at /home/blegat/.julia/dev/CDCS/test/lp.jl:5\n  Got exception outside of a @test\n  MATLAB.MEngineError(\"failed to get variable jx_cdcs_arg_out_1 from MATLAB session\")\n  Stacktrace:\n    [1] get_mvariable(session::MATLAB.MSession, name::Symbol)\n      @ MATLAB ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:164\n    [2] mxcall(::MATLAB.MSession, ::Symbol, ::Int64, ::Matrix{Float64}, ::Vararg{Any})\n      @ MATLAB ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:297\n    [3] mxcall\n      @ ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:317 [inlined]\n    [4] cdcs(A::Matrix{Float64}, b::Vector{Float64}, c::Vector{Float64}, K::CDCS.Cone; kws::Base.Pairs{Symbol, Int64, Tuple{Symbol}, NamedTuple{(:verbose,), Tuple{Int64}}})","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The error means that we could not find the cdcs function with one output argument using the MATLAB C API.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"This most likely means that you did not add CDCS to the MATLAB's path (that is, the toolbox/local/pathdef.m file).","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work where /path/to/CDCS/ is the directory where the CDCS folder is located:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"julia> using MATLAB\n\njulia> cd(\"/path/to/CDCS/\") do\n           mat\"cdcsInstall\"\n       end\n\njulia> mat\"savepath\"","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"EditURL = \"https://github.com/jump-dev/MultiObjectiveAlgorithms.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"<img src=\"https://raw.githubusercontent.com/jump-dev/MultiObjectiveAlgorithms.jl/master/moa.png?raw=true\" alt=\"An image of the Moa bird. Licensed into the Public Domain by https://freesvg.org/moa\" width=\"100px\"/>","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#MultiObjectiveAlgorithms.jl","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"MultiObjectiveAlgorithms.jl","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MultiObjectiveAlgorithms.jl (MOA) is a collection of algorithms for multi-objective optimization.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#License","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"License","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MultiObjectiveAlgorithms.jl is licensed under the MPL 2.0 License.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Installation","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Install MOA using Pkg.add:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"import Pkg\nPkg.add(\"MultiObjectiveAlgorithms\")","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Use-with-JuMP","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Use MultiObjectiveAlgorithms with JuMP as follows:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA\nmodel = JuMP.Model(() -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.Dichotomy())\nset_attribute(model, MOA.SolutionLimit(), 4)","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Replace HiGHS.Optimizer with an optimizer capable of solving a single-objective instance of your optimization problem.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"You may set additional optimizer attributes, the supported attributes depend on the choice of solution algorithm.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Algorithm","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Algorithm","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Set the algorithm using the MOA.Algorithm() attribute.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"The value must be one of the algorithms supported by MOA:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MOA.Chalmet()\nMOA.Dichotomy()\nMOA.DominguezRios()\nMOA.EpsilonConstraint()\nMOA.Hierarchical()\nMOA.KirlikSayin()\nMOA.Lexicographic() [default]\nMOA.TambyVanderpooten()","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Consult their docstrings for details.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Other-optimizer-attributes","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Other optimizer attributes","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"There are a number of optimizer attributes supported by the algorithms in MOA.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Each algorithm supports only a subset of the attributes. Consult the algorithm's docstring for details on which attributes it supports, and how it uses them in the solution process.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MOA.EpsilonConstraintStep()\nMOA.LexicographicAllPermutations()\nMOA.ObjectiveAbsoluteTolerance(index::Int)\nMOA.ObjectivePriority(index::Int)\nMOA.ObjectiveRelativeTolerance(index::Int)\nMOA.ObjectiveWeight(index::Int)\nMOA.SolutionLimit()\nMOI.TimeLimitSec()","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"EditURL = \"multi_commodity_network.jl\"","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#The-network-multi-commodity-flow-problem","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial is a variation of The multi-commodity flow problem where the graph is a network instead of a bipartite graph.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate a style of modeling that uses relational algebra.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Required-packages","page":"The network multi-commodity flow problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport SQLite\nimport SQLite.DBInterface","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Formulation","page":"The network multi-commodity flow problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The network multi-commondity flow problem is an extension of the The multi-commodity flow problem, where instead of having a bipartite graph of supply and demand nodes, the graph can contains a set of nodes, i in mathcalN , which each have a (potentially zero) supply capacity, u^s_ip, and (potentially zero) a demand, d_ip for each commodity p in P. The nodes are connected by a set of edges (i j) in mathcalE, which have a shipment cost c^x_ijp and a total flow capacity of u^x_ij.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Our take is to choose an optimal supply for each node s_ip, as well as the optimal transshipment x_ijp that minimizes the total cost.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The mathematical formulation is:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"beginaligned\nmin   sum_(ij)inmathcalE p in P c^x_ijp x_ijp + sum_iinmathcalN p in P c^s_ip s_ip \nst   s_ip + sum_(j i) in mathcalE x_jip - sum_(ij) in mathcalE x_ijp = d_ip  forall i in mathcalN p in P \n           x_ijp ge 0            forall (i j) in mathcalE p in P \n           sum_p in P x_ijp le u^x_ij            forall (i j) in mathcalE \n           0 le s_ip le u^s_ip  forall i in mathcalN p in P\nendaligned","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate how this model can be built using relational algebra instead of a direct math-to-code translation of the summations.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Data","page":"The network multi-commodity flow problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the purpose of this tutorial, the JuMP repository contains an example database called commodity_nz.db:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"filename = joinpath(@__DIR__, \"commodity_nz.db\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"To run locally, download commodity_nz.db and update filename appropriately.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Load the database using SQLite.DB:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"db = SQLite.DB(filename)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"A quick way to see the schema of the database is via SQLite.tables:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"SQLite.tables(db)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We interact with the database by executing queries and then loading the results into a DataFrame:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"function get_table(db, table)\n    query = DBInterface.execute(db, \"SELECT * FROM $table\")\n    return DataFrames.DataFrame(query)\nend","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The shipping table contains the set of arcs and their distances:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping = get_table(db, \"shipping\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The products table contains the shipping cost per kilometer of each product:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_products = get_table(db, \"products\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The supply table contains the supply capacity of each node, as well as the cost:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_supply = get_table(db, \"supply\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The demand table contains the demand of each node:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_demand = get_table(db, \"demand\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#JuMP-formulation","page":"The network multi-commodity flow problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We start by creating a model and our decision variables:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the shipping decisions, we create a new column in df_shipping called x_flow, which has one non-negative decision variable for each row:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping.x_flow = @variable(model, x[1:size(df_shipping, 1)] >= 0)\ndf_shipping","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the supply, we add a variable to each row, and then set the upper bound to the capacity of each supply node:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_supply.x_supply = @variable(model, s[1:size(df_supply, 1)] >= 0)\nset_upper_bound.(df_supply.x_supply, df_supply.capacity)\ndf_supply","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Our objective is to minimize the shipping cost plus the supply cost. To compute the flow cost, we need to join the shipping table, which contains distance_km with the products table, which contains cost_per_km:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_cost = DataFrames.leftjoin(df_shipping, df_products; on = [:product])\ndf_cost.flow_cost = df_cost.cost_per_km .* df_cost.distance_km\ndf_cost","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Then we can use linear algebra to compute the inner product between two columns:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"@objective(\n    model,\n    Min,\n    df_cost.flow_cost' * df_shipping.x_flow +\n    df_supply.cost' * df_supply.x_supply\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the flow capacities on each arc, we use DataFrames.groupby to partition the flow variables based on :origin and :destination, and then we constrain their sum to be less than a fixed capacity.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"capacity = 30\nfor df in DataFrames.groupby(df_shipping, [:origin, :destination])\n    @constraint(model, sum(df.x_flow) <= capacity)\nend","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For each node in the graph, we need to compute a mass balance constraint which says that for each product, the supply, plus the flow into the node, and less the flow out of the node is equal to the demand.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can compute an expression for the flow out of each node using DataFrames.groupby on the origin and product columns of the df_shipping table:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_flow_out = DataFrames.DataFrame(\n    (node = i.origin, product = i.product, x_flow_out = sum(df.x_flow)) for\n    (i, df) in pairs(DataFrames.groupby(df_shipping, [:origin, :product]))\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can compute an expression for the flow into each node using DataFrames.groupby on the destination and product columns of the df_shipping table:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_flow_in = DataFrames.DataFrame(\n    (node = i.destination, product = i.product, x_flow_in = sum(df.x_flow))\n    for (i, df) in\n    pairs(DataFrames.groupby(df_shipping, [:destination, :product]))\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can join the two tables together using DataFrames.outerjoin. We need to use outerjoin here because there might be missing rows.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.outerjoin(df_flow_in, df_flow_out; on = [:node, :product])","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Next, we need to join the supply column:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.leftjoin(\n    df,\n    DataFrames.select(df_supply, [:origin, :product, :x_supply]);\n    on = [:node => :origin, :product],\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"and then the demand column","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.leftjoin(\n    df,\n    DataFrames.select(df_demand, [:destination, :product, :demand]);\n    on = [:node => :destination, :product],\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Now we're ready to add our mass balance constraint. Because some rows contain missing values, we need to use coalesce to convert any missing into a numeric value:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"@constraint(\n    model,\n    [r in eachrow(df)],\n    coalesce(r.x_supply, 0.0) + coalesce(r.x_flow_in, 0.0) -\n    coalesce(r.x_flow_out, 0.0) == coalesce(r.demand, 0.0),\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Solution","page":"The network multi-commodity flow problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Finally, we can optimize the model:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"update the solution in the DataFrames:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping.x_flow = value.(df_shipping.x_flow)\ndf_supply.x_supply = value.(df_supply.x_supply);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"and display the optimal solution for flows:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"DataFrames.select(\n    filter!(row -> row.x_flow > 0.0, df_shipping),\n    [:origin, :destination, :product, :x_flow],\n)","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/standard_form.md\"","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/standard_form/#Standard-form","page":"Standard form","title":"Standard form","text":"","category":"section"},{"location":"moi/reference/standard_form/#Functions","page":"Standard form","title":"Functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractFunction\noutput_dimension\nconstant","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractFunction","page":"Standard form","title":"MathOptInterface.AbstractFunction","text":"AbstractFunction\n\nAbstract supertype for function objects.\n\nRequired methods\n\nAll functions must implement:\n\nBase.copy\nBase.isapprox\nconstant\n\nAbstract subtypes of AbstractFunction may require additional methods to be implemented.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.output_dimension","page":"Standard form","title":"MathOptInterface.output_dimension","text":"output_dimension(f::AbstractFunction)\n\nReturn 1 if f is an AbstractScalarFunction, or the number of output components if f is an AbstractVectorFunction.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.constant","page":"Standard form","title":"MathOptInterface.constant","text":"constant(f::AbstractFunction[, ::Type{T}]) where {T}\n\nReturns the constant term of a scalar-valued function, or the constant vector of a vector-valued function.\n\nIf f is untyped and T is provided, returns zero(T).\n\n\n\n\n\nconstant(set::Union{EqualTo,GreaterThan,LessThan,Parameter})\n\nReturns the constant term of the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.constant(MOI.GreaterThan(1.0))\n1.0\n\njulia> MOI.constant(MOI.LessThan(2.5))\n2.5\n\njulia> MOI.constant(MOI.EqualTo(3))\n3\n\njulia> MOI.constant(MOI.Parameter(4.5))\n4.5\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#Scalar-functions","page":"Standard form","title":"Scalar functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractScalarFunction\nVariableIndex\nScalarAffineTerm\nScalarAffineFunction\nScalarQuadraticTerm\nScalarQuadraticFunction\nScalarNonlinearFunction","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractScalarFunction","page":"Standard form","title":"MathOptInterface.AbstractScalarFunction","text":"abstract type AbstractScalarFunction <: AbstractFunction\n\nAbstract supertype for scalar-valued AbstractFunctions.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VariableIndex","page":"Standard form","title":"MathOptInterface.VariableIndex","text":"VariableIndex\n\nA type-safe wrapper for Int64 for use in referencing variables in a model. To allow for deletion, indices need not be consecutive.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarAffineTerm","page":"Standard form","title":"MathOptInterface.ScalarAffineTerm","text":"ScalarAffineTerm{T}(coefficient::T, variable::VariableIndex) where {T}\n\nRepresents the scalar-valued term coefficient * variable.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarAffineTerm(2.0, x)\nMathOptInterface.ScalarAffineTerm{Float64}(2.0, MOI.VariableIndex(1))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction","page":"Standard form","title":"MathOptInterface.ScalarAffineFunction","text":"ScalarAffineFunction{T}(terms::ScalarAffineTerm{T}, constant::T) where {T}\n\nRepresents the scalar-valued affine function a^top x + b, where:\n\na^top x is represented by the vector of ScalarAffineTerms\nb is a scalar constant::T\n\nDuplicates\n\nDuplicate variable indices in terms are accepted, and the corresponding coefficients are summed together.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> terms = [MOI.ScalarAffineTerm(2.0, x), MOI.ScalarAffineTerm(3.0, x)]\n2-element Vector{MathOptInterface.ScalarAffineTerm{Float64}}:\n MathOptInterface.ScalarAffineTerm{Float64}(2.0, MOI.VariableIndex(1))\n MathOptInterface.ScalarAffineTerm{Float64}(3.0, MOI.VariableIndex(1))\n\njulia> f = MOI.ScalarAffineFunction(terms, 4.0)\n4.0 + 2.0 MOI.VariableIndex(1) + 3.0 MOI.VariableIndex(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarQuadraticTerm","page":"Standard form","title":"MathOptInterface.ScalarQuadraticTerm","text":"ScalarQuadraticTerm{T}(\n    coefficient::T,\n    variable_1::VariableIndex,\n    variable_2::VariableIndex,\n) where {T}\n\nRepresents the scalar-valued term c x_i x_j where c is coefficient, x_i is variable_1 and x_j is variable_2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarQuadraticTerm(2.0, x, x)\nMathOptInterface.ScalarQuadraticTerm{Float64}(2.0, MOI.VariableIndex(1), MOI.VariableIndex(1))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction","page":"Standard form","title":"MathOptInterface.ScalarQuadraticFunction","text":"ScalarQuadraticFunction{T}(\n    quadratic_terms::Vector{ScalarQuadraticTerm{T}},\n    affine_terms::Vector{ScalarAffineTerm{T}},\n    constant::T,\n) wher {T}\n\nThe scalar-valued quadratic function frac12x^top Q x + a^top x + b, where:\n\nQ is the symmetric matrix given by the vector of ScalarQuadraticTerms\na^top x is a sparse vector given by the vector of ScalarAffineTerms\nb is the scalar constant::T.\n\nDuplicates\n\nDuplicate indices in quadratic_terms or affine_terms are accepted, and the corresponding coefficients are summed together.\n\nIn quadratic_terms, \"mirrored\" indices, (q, r) and (r, q) where r and q are VariableIndexes, are considered duplicates; only one needs to be specified.\n\nThe 0.5 factor\n\nCoupled with the interpretation of mirrored indices, the 0.5 factor in front of the Q matrix is a common source of bugs.\n\nAs a rule, to represent a * x^2 + b * x * y:\n\nThe coefficient a in front of squared variables (diagonal elements in Q) must be doubled when creating a ScalarQuadraticTerm\nThe coefficient b in front of off-diagonal elements in Q should be left as b, be cause the mirrored index b * y * x will be implicitly added.\n\nExample\n\nTo represent the function f(x y) = 2 * x^2 + 3 * x * y + 4 * x + 5, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> y = MOI.VariableIndex(2);\n\njulia> constant = 5.0;\n\njulia> affine_terms = [MOI.ScalarAffineTerm(4.0, x)];\n\njulia> quadratic_terms = [\n           MOI.ScalarQuadraticTerm(4.0, x, x),  # Note the changed coefficient\n           MOI.ScalarQuadraticTerm(3.0, x, y),\n       ]\n2-element Vector{MathOptInterface.ScalarQuadraticTerm{Float64}}:\n MathOptInterface.ScalarQuadraticTerm{Float64}(4.0, MOI.VariableIndex(1), MOI.VariableIndex(1))\n MathOptInterface.ScalarQuadraticTerm{Float64}(3.0, MOI.VariableIndex(1), MOI.VariableIndex(2))\n\njulia> f = MOI.ScalarQuadraticFunction(quadratic_terms, affine_terms, constant)\n5.0 + 4.0 MOI.VariableIndex(1) + 2.0 MOI.VariableIndex(1)² + 3.0 MOI.VariableIndex(1)*MOI.VariableIndex(2)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction","page":"Standard form","title":"MathOptInterface.ScalarNonlinearFunction","text":"ScalarNonlinearFunction(head::Symbol, args::Vector{Any})\n\nThe scalar-valued nonlinear function head(args...), represented as a symbolic expression tree, with the call operator head and ordered arguments in args.\n\nhead\n\nThe head::Symbol must be an operator supported by the model.\n\nThe default list of supported univariate operators is given by:\n\nNonlinear.DEFAULT_UNIVARIATE_OPERATORS\n\nand the default list of supported multivariate operators is given by:\n\nNonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n\nAdditional operators can be registered by setting a UserDefinedFunction attribute.\n\nSee the full list of operators supported by a ModelLike by querying ListOfSupportedNonlinearOperators.\n\nargs\n\nThe vector args contains the arguments to the nonlinear function. If the operator is univariate, it must contain one element. Otherwise, it may contain multiple elements.\n\nEach element must be one of the following:\n\nA constant value of type T<:Real\nA VariableIndex\nA ScalarAffineFunction\nA ScalarQuadraticFunction\nA ScalarNonlinearFunction\n\nUnsupported operators\n\nIf the optimizer does not support head, an UnsupportedNonlinearOperator error will be thrown.\n\nThere is no guarantee about when this error will be thrown; it may be thrown when the function is first added to the model, or it may be thrown when optimize! is called.\n\nExample\n\nTo represent the function f(x) = sin(x)^2, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarNonlinearFunction(\n           :^,\n           Any[MOI.ScalarNonlinearFunction(:sin, Any[x]), 2],\n       )\n^(sin(MOI.VariableIndex(1)), (2))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Vector-functions","page":"Standard form","title":"Vector functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractVectorFunction\nVectorOfVariables\nVectorAffineTerm\nVectorAffineFunction\nVectorQuadraticTerm\nVectorQuadraticFunction\nVectorNonlinearFunction","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractVectorFunction","page":"Standard form","title":"MathOptInterface.AbstractVectorFunction","text":"abstract type AbstractVectorFunction <: AbstractFunction\n\nAbstract supertype for vector-valued AbstractFunctions.\n\nRequired methods\n\nAll subtypes of AbstractVectorFunction must implement:\n\noutput_dimension\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorOfVariables","page":"Standard form","title":"MathOptInterface.VectorOfVariables","text":"VectorOfVariables(variables::Vector{VariableIndex}) <: AbstractVectorFunction\n\nThe vector-valued function f(x) = variables, where variables is a subset of VariableIndexes in the model.\n\nThe list of variables may contain duplicates.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex.(1:2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> f = MOI.VectorOfVariables([x[1], x[2], x[1]])\n┌                    ┐\n│MOI.VariableIndex(1)│\n│MOI.VariableIndex(2)│\n│MOI.VariableIndex(1)│\n└                    ┘\n\njulia> MOI.output_dimension(f)\n3\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorAffineTerm","page":"Standard form","title":"MathOptInterface.VectorAffineTerm","text":"VectorAffineTerm{T}(\n    output_index::Int64,\n    scalar_term::ScalarAffineTerm{T},\n) where {T}\n\nA VectorAffineTerm is a scalar_term that appears in the output_index row of the vector-valued VectorAffineFunction or VectorQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x))\nMathOptInterface.VectorAffineTerm{Float64}(2, MathOptInterface.ScalarAffineTerm{Float64}(3.0, MOI.VariableIndex(1)))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorAffineFunction","page":"Standard form","title":"MathOptInterface.VectorAffineFunction","text":"VectorAffineFunction{T}(\n    terms::Vector{VectorAffineTerm{T}},\n    constants::Vector{T},\n) where {T}\n\nThe vector-valued affine function A x + b, where:\n\nA x is the sparse matrix given by the vector of VectorAffineTerms\nb is the vector constants\n\nDuplicates\n\nDuplicate indices in the A are accepted, and the corresponding coefficients are summed together.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> terms = [\n           MOI.VectorAffineTerm(Int64(1), MOI.ScalarAffineTerm(2.0, x)),\n           MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x)),\n       ];\n\njulia> f = MOI.VectorAffineFunction(terms, [4.0, 5.0])\n┌                              ┐\n│4.0 + 2.0 MOI.VariableIndex(1)│\n│5.0 + 3.0 MOI.VariableIndex(1)│\n└                              ┘\n\njulia> MOI.output_dimension(f)\n2\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorQuadraticTerm","page":"Standard form","title":"MathOptInterface.VectorQuadraticTerm","text":"VectorQuadraticTerm{T}(\n    output_index::Int64,\n    scalar_term::ScalarQuadraticTerm{T},\n) where {T}\n\nA VectorQuadraticTerm is a ScalarQuadraticTerm scalar_term that appears in the output_index row of the vector-valued VectorQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> MOI.VectorQuadraticTerm(Int64(2), MOI.ScalarQuadraticTerm(3.0, x, x))\nMathOptInterface.VectorQuadraticTerm{Float64}(2, MathOptInterface.ScalarQuadraticTerm{Float64}(3.0, MOI.VariableIndex(1), MOI.VariableIndex(1)))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorQuadraticFunction","page":"Standard form","title":"MathOptInterface.VectorQuadraticFunction","text":"VectorQuadraticFunction{T}(\n    quadratic_terms::Vector{VectorQuadraticTerm{T}},\n    affine_terms::Vector{VectorAffineTerm{T}},\n    constants::Vector{T},\n) where {T}\n\nThe vector-valued quadratic function with ith component (\"output index\") defined as frac12x^top Q_i x + a_i^top x + b_i, where:\n\nfrac12x^top Q_i x is the symmetric matrix given by the VectorQuadraticTerm elements in quadratic_terms with output_index == i\na_i^top x is the sparse vector given by the VectorAffineTerm elements in affine_terms with output_index == i\nb_i is a scalar given by constants[i]\n\nDuplicates\n\nDuplicate indices in quadratic_terms and affine_terms with the same output_index are handled in the same manner as duplicates in ScalarQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> y = MOI.VariableIndex(2);\n\njulia> constants = [4.0, 5.0];\n\njulia> affine_terms = [\n           MOI.VectorAffineTerm(Int64(1), MOI.ScalarAffineTerm(2.0, x)),\n           MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x)),\n       ];\n\njulia> quad_terms = [\n        MOI.VectorQuadraticTerm(Int64(1), MOI.ScalarQuadraticTerm(2.0, x, x)),\n        MOI.VectorQuadraticTerm(Int64(2), MOI.ScalarQuadraticTerm(3.0, x, y)),\n           ];\n\njulia> f = MOI.VectorQuadraticFunction(quad_terms, affine_terms, constants)\n┌                                                                              ┐\n│4.0 + 2.0 MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²                    │\n│5.0 + 3.0 MOI.VariableIndex(1) + 3.0 MOI.VariableIndex(1)*MOI.VariableIndex(2)│\n└                                                                              ┘\n\njulia> MOI.output_dimension(f)\n2\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorNonlinearFunction","page":"Standard form","title":"MathOptInterface.VectorNonlinearFunction","text":"VectorNonlinearFunction(args::Vector{ScalarNonlinearFunction})\n\nThe vector-valued nonlinear function composed of a vector of ScalarNonlinearFunction.\n\nargs\n\nThe vector args contains the scalar elements of the nonlinear function. Each element must be a ScalarNonlinearFunction, but if you pass a Vector{Any}, the elements can be automatically converted from one of the following:\n\nA constant value of type T<:Real\nA VariableIndex\nA ScalarAffineFunction\nA ScalarQuadraticFunction\nA ScalarNonlinearFunction\n\nExample\n\nTo represent the function f(x) = sin(x)^2 x, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> g = MOI.ScalarNonlinearFunction(\n           :^,\n           Any[MOI.ScalarNonlinearFunction(:sin, Any[x]), 2.0],\n       )\n^(sin(MOI.VariableIndex(1)), 2.0)\n\njulia> MOI.VectorNonlinearFunction([g, x])\n┌                                 ┐\n│^(sin(MOI.VariableIndex(1)), 2.0)│\n│+(MOI.VariableIndex(1))          │\n└                                 ┘\n\nNote the automatic conversion from x to +(x).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Sets","page":"Standard form","title":"Sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractSet\nAbstractScalarSet\nAbstractVectorSet","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSet","page":"Standard form","title":"MathOptInterface.AbstractSet","text":"AbstractSet\n\nAbstract supertype for set objects used to encode constraints.\n\nRequired methods\n\nFor sets of type S with isbitstype(S) == false, you must implement:\n\nBase.copy(set::S)\nBase.:(==)(x::S, y::S)\n\nSubtypes of AbstractSet such as AbstractScalarSet and AbstractVectorSet may prescribe additional required methods.\n\nOptional methods\n\nYou may optionally implement:\n\ndual_set\ndual_set_type\n\nNote for developers\n\nWhen creating a new set, the set struct must not contain any VariableIndex or ConstraintIndex objects.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractScalarSet","page":"Standard form","title":"MathOptInterface.AbstractScalarSet","text":"AbstractScalarSet\n\nAbstract supertype for subsets of mathbbR.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractVectorSet","page":"Standard form","title":"MathOptInterface.AbstractVectorSet","text":"AbstractVectorSet\n\nAbstract supertype for subsets of mathbbR^n for some n.\n\nRequired methods\n\nAll AbstractVectorSets of type S must implement:\n\ndimension, unless the dimension is stored in the set.dimension field\nUtilities.set_dot, unless the dot product between two vectors in the set is equivalent to LinearAlgebra.dot.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Utilities","page":"Standard form","title":"Utilities","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"dimension\ndual_set\ndual_set_type\nconstant(s::EqualTo)\nsupports_dimension_update\nupdate_dimension","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.dimension","page":"Standard form","title":"MathOptInterface.dimension","text":"dimension(set::AbstractSet)\n\nReturn the output_dimension that an AbstractFunction should have to be used with the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dimension(MOI.Reals(4))\n4\n\njulia> MOI.dimension(MOI.LessThan(3.0))\n1\n\njulia> MOI.dimension(MOI.PositiveSemidefiniteConeTriangle(2))\n3\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.dual_set","page":"Standard form","title":"MathOptInterface.dual_set","text":"dual_set(set::AbstractSet)\n\nReturn the dual set of set, that is the dual cone of the set. This follows the definition of duality discussed in Duality.\n\nSee Dual cone for more information.\n\nIf the dual cone is not defined it returns an error.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dual_set(MOI.Reals(4))\nMathOptInterface.Zeros(4)\n\njulia> MOI.dual_set(MOI.SecondOrderCone(5))\nMathOptInterface.SecondOrderCone(5)\n\njulia> MOI.dual_set(MOI.ExponentialCone())\nMathOptInterface.DualExponentialCone()\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.dual_set_type","page":"Standard form","title":"MathOptInterface.dual_set_type","text":"dual_set_type(S::Type{<:AbstractSet})\n\nReturn the type of dual set of sets of type S, as returned by dual_set. If the dual cone is not defined it returns an error.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dual_set_type(MOI.Reals)\nMathOptInterface.Zeros\n\njulia> MOI.dual_set_type(MOI.SecondOrderCone)\nMathOptInterface.SecondOrderCone\n\njulia> MOI.dual_set_type(MOI.ExponentialCone)\nMathOptInterface.DualExponentialCone\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.constant-Tuple{MathOptInterface.EqualTo}","page":"Standard form","title":"MathOptInterface.constant","text":"constant(set::Union{EqualTo,GreaterThan,LessThan,Parameter})\n\nReturns the constant term of the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.constant(MOI.GreaterThan(1.0))\n1.0\n\njulia> MOI.constant(MOI.LessThan(2.5))\n2.5\n\njulia> MOI.constant(MOI.EqualTo(3))\n3\n\njulia> MOI.constant(MOI.Parameter(4.5))\n4.5\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/standard_form/#MathOptInterface.supports_dimension_update","page":"Standard form","title":"MathOptInterface.supports_dimension_update","text":"supports_dimension_update(S::Type{<:MOI.AbstractVectorSet})\n\nReturn a Bool indicating whether the elimination of any dimension of n-dimensional sets of type S give an n-1-dimensional set S. By default, this function returns false so it should only be implemented for sets that supports dimension update.\n\nFor instance, supports_dimension_update(MOI.Nonnegatives) is true because the elimination of any dimension of the n-dimensional nonnegative orthant gives the n-1-dimensional nonnegative orthant. However supports_dimension_update(MOI.ExponentialCone) is false.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.update_dimension","page":"Standard form","title":"MathOptInterface.update_dimension","text":"update_dimension(s::AbstractVectorSet, new_dim::Int)\n\nReturns a set with the dimension modified to new_dim.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#Scalar-sets","page":"Standard form","title":"Scalar sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized scalar sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"GreaterThan\nLessThan\nEqualTo\nInterval\nInteger\nZeroOne\nSemicontinuous\nSemiinteger\nParameter","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.GreaterThan","page":"Standard form","title":"MathOptInterface.GreaterThan","text":"GreaterThan{T<:Real}(lower::T)\n\nThe set lower infty) subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.GreaterThan(0.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LessThan","page":"Standard form","title":"MathOptInterface.LessThan","text":"LessThan{T<:Real}(upper::T)\n\nThe set (-infty upper subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.LessThan(2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.LessThan{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.EqualTo","page":"Standard form","title":"MathOptInterface.EqualTo","text":"EqualTo{T<:Number}(value::T)\n\nThe set containing the single point value subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.EqualTo(2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.EqualTo{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Interval","page":"Standard form","title":"MathOptInterface.Interval","text":"Interval{T<:Real}(lower::T, upper::T)\n\nThe interval lower upper subseteq mathbbR cup -infty +infty.\n\nIf lower or upper is -Inf or Inf, respectively, the set is interpreted as a one-sided interval.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Interval(1.0, 2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Integer","page":"Standard form","title":"MathOptInterface.Integer","text":"Integer()\n\nThe set of integers, mathbbZ.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Integer())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ZeroOne","page":"Standard form","title":"MathOptInterface.ZeroOne","text":"ZeroOne()\n\nThe set 0 1.\n\nVariables belonging to the ZeroOne set are also known as \"binary\" variables.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.ZeroOne())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Semicontinuous","page":"Standard form","title":"MathOptInterface.Semicontinuous","text":"Semicontinuous{T<:Real}(lower::T, upper::T)\n\nThe set 0 cup lower upper.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Semicontinuous(2.0, 3.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Semicontinuous{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Semiinteger","page":"Standard form","title":"MathOptInterface.Semiinteger","text":"Semiinteger{T<:Real}(lower::T, upper::T)\n\nThe set 0 cup lower lower+1 ldots upper-1 upper.\n\nNote that if lower and upper are not equivalent to an integer, then the solver may throw an error, or it may round up lower and round down upper to the nearest integers.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Semiinteger(2.0, 3.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Semiinteger{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Parameter","page":"Standard form","title":"MathOptInterface.Parameter","text":"Parameter{T<:Number}(value::T)\n\nThe set containing the single point value subseteq mathbbR.\n\nThe Parameter set is conceptually similar to the EqualTo set, except that a variable constrained to the Parameter set cannot have other constraints added to it, and the Parameter set can never be deleted. Thus, solvers are free to treat the variable as a constant, and they need not add it as a decision variable to the model.\n\nBecause of this behavior, you must add parameters using add_constrained_variable, and solvers should declare supports_add_constrained_variable and not supports_constraint for the Parameter set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> p, ci = MOI.add_constrained_variable(model, MOI.Parameter(2.5))\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Parameter{Float64}}(1))\n\njulia> MOI.set(model, MOI.ConstraintSet(), ci, MOI.Parameter(3.0))\n\njulia> MOI.get(model, MOI.ConstraintSet(), ci)\nMathOptInterface.Parameter{Float64}(3.0)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Vector-sets","page":"Standard form","title":"Vector sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized vector sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"Reals\nZeros\nNonnegatives\nNonpositives\nNormInfinityCone\nNormOneCone\nNormCone\nSecondOrderCone\nRotatedSecondOrderCone\nGeometricMeanCone\nExponentialCone\nDualExponentialCone\nPowerCone\nDualPowerCone\nRelativeEntropyCone\nNormSpectralCone\nNormNuclearCone\nSOS1\nSOS2\nIndicator\nComplements\nHyperRectangle\nScaled","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.Reals","page":"Standard form","title":"MathOptInterface.Reals","text":"Reals(dimension::Int)\n\nThe set mathbbR^dimension (containing all points) of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Reals(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Reals}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Zeros","page":"Standard form","title":"MathOptInterface.Zeros","text":"Zeros(dimension::Int)\n\nThe set  0 ^dimension (containing only the origin) of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Zeros(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Zeros}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Nonnegatives","page":"Standard form","title":"MathOptInterface.Nonnegatives","text":"Nonnegatives(dimension::Int)\n\nThe nonnegative orthant  x in mathbbR^dimension  x ge 0  of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonnegatives(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Nonpositives","page":"Standard form","title":"MathOptInterface.Nonpositives","text":"Nonpositives(dimension::Int)\n\nThe nonpositive orthant  x in mathbbR^dimension  x le 0  of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonpositives(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonpositives}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormInfinityCone","page":"Standard form","title":"MathOptInterface.NormInfinityCone","text":"NormInfinityCone(dimension::Int)\n\nThe ell_infty-norm cone  (tx) in mathbbR^dimension  t ge lVert x rVert_infty = max_i lvert x_i rvert  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormInfinityCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormInfinityCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormOneCone","page":"Standard form","title":"MathOptInterface.NormOneCone","text":"NormOneCone(dimension::Int)\n\nThe ell_1-norm cone  (tx) in mathbbR^dimension  t ge lVert x rVert_1 = sum_i lvert x_i rvert  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormOneCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormOneCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormCone","page":"Standard form","title":"MathOptInterface.NormCone","text":"NormCone(p::Float64, dimension::Int)\n\nThe ell_p-norm cone  (tx) in mathbbR^dimension  t ge left(sumlimits_i x_i^pright)^frac1p  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormCone(3, 4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SecondOrderCone","page":"Standard form","title":"MathOptInterface.SecondOrderCone","text":"SecondOrderCone(dimension::Int)\n\nThe second-order cone (or Lorenz cone or ell_2-norm cone)  (tx) in mathbbR^dimension  t ge lVert x rVert_2  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.SecondOrderCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SecondOrderCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RotatedSecondOrderCone","page":"Standard form","title":"MathOptInterface.RotatedSecondOrderCone","text":"RotatedSecondOrderCone(dimension::Int)\n\nThe rotated second-order cone  (tux) in mathbbR^dimension  2tu ge lVert x rVert_2^2 tu ge 0  of dimension dimension.\n\nThe dimension must be at least 2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> u = MOI.add_variable(model)\nMOI.VariableIndex(2)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; u; x]),\n           MOI.RotatedSecondOrderCone(5),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RotatedSecondOrderCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.GeometricMeanCone","page":"Standard form","title":"MathOptInterface.GeometricMeanCone","text":"GeometricMeanCone(dimension::Int)\n\nThe geometric mean cone  (tx) in mathbbR^n+1  x ge 0 t le sqrtnx_1 x_2 cdots x_n , where dimension = n + 1 >= 2.\n\nDuality note\n\nThe dual of the geometric mean cone is  (u v) in mathbbR^n+1  u le 0 v ge 0 -u le n sqrtnprod_i v_i , where dimension = n + 1 >= 2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; x]),\n           MOI.GeometricMeanCone(4),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.GeometricMeanCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ExponentialCone","page":"Standard form","title":"MathOptInterface.ExponentialCone","text":"ExponentialCone()\n\nThe 3-dimensional exponential cone  (xyz) in mathbbR^3  y exp (xy) le z y  0 .\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.ExponentialCone())\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.ExponentialCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.DualExponentialCone","page":"Standard form","title":"MathOptInterface.DualExponentialCone","text":"DualExponentialCone()\n\nThe 3-dimensional dual exponential cone  (uvw) in mathbbR^3  -u exp (vu) le exp(1) w u  0 .\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.DualExponentialCone())\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.DualExponentialCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.PowerCone","page":"Standard form","title":"MathOptInterface.PowerCone","text":"PowerCone{T<:Real}(exponent::T)\n\nThe 3-dimensional power cone  (xyz) in mathbbR^3  x^exponent y^1-exponent ge z x ge 0 y ge 0  with parameter exponent.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.PowerCone(0.5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.PowerCone{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.DualPowerCone","page":"Standard form","title":"MathOptInterface.DualPowerCone","text":"DualPowerCone{T<:Real}(exponent::T)\n\nThe 3-dimensional power cone  (uvw) in mathbbR^3  (fracuexponent)^exponent (fracv1-exponent)^1-exponent ge w u ge 0 v ge 0  with parameter exponent.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.DualPowerCone(0.5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.DualPowerCone{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RelativeEntropyCone","page":"Standard form","title":"MathOptInterface.RelativeEntropyCone","text":"RelativeEntropyCone(dimension::Int)\n\nThe relative entropy cone  (u v w) in mathbbR^1+2n  u ge sum_i=1^n w_i log(fracw_iv_i) v_i ge 0 w_i ge 0 , where dimension = 2n + 1 >= 1.\n\nDuality note\n\nThe dual of the relative entropy cone is  (u v w) in mathbbR^1+2n  forall i w_i ge u (log (fracuv_i) - 1) v_i ge 0 u  0  of dimension dimension=2n+1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> u = MOI.add_variable(model);\n\njulia> v = MOI.add_variables(model, 3);\n\njulia> w = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([u; v; w]),\n           MOI.RelativeEntropyCone(7),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RelativeEntropyCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormSpectralCone","page":"Standard form","title":"MathOptInterface.NormSpectralCone","text":"NormSpectralCone(row_dim::Int, column_dim::Int)\n\nThe epigraph of the matrix spectral norm (maximum singular value function)  (t X) in mathbbR^1 + row_dim times column_dim  t ge sigma_1(X) , where sigma_i is the ith singular value of the matrix X of non-negative row dimension row_dim and column dimension column_dim.\n\nThe matrix X is vectorized by stacking the columns, matching the behavior of Julia's vec function.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 6), 2, 3)\n2×3 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)  MOI.VariableIndex(6)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)  MOI.VariableIndex(7)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.NormSpectralCone(2, 3),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormSpectralCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormNuclearCone","page":"Standard form","title":"MathOptInterface.NormNuclearCone","text":"NormNuclearCone(row_dim::Int, column_dim::Int)\n\nThe epigraph of the matrix nuclear norm (sum of singular values function)  (t X) in mathbbR^1 + row_dim times column_dim  t ge sum_i sigma_i(X) , where sigma_i is the ith singular value of the matrix X of non-negative row dimension row_dim and column dimension column_dim.\n\nThe matrix X is vectorized by stacking the columns, matching the behavior of Julia's vec function.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 6), 2, 3)\n2×3 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)  MOI.VariableIndex(6)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)  MOI.VariableIndex(7)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.NormNuclearCone(2, 3),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormNuclearCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SOS1","page":"Standard form","title":"MathOptInterface.SOS1","text":"SOS1{T<:Real}(weights::Vector{T})\n\nThe set corresponding to the Special Ordered Set (SOS) constraint of Type I.\n\nOf the variables in the set, at most one can be nonzero.\n\nThe weights induce an ordering of the variables such that the kth element in the set corresponds to the kth weight in weights. Solvers may use these weights to improve the efficiency of the solution process, but the ordering does not change the set of feasible solutions.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.SOS1([1.0, 3.0, 2.5]),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SOS1{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SOS2","page":"Standard form","title":"MathOptInterface.SOS2","text":"SOS2{T<:Real}(weights::Vector{T})\n\nThe set corresponding to the Special Ordered Set (SOS) constraint of Type II.\n\nThe weights induce an ordering of the variables such that the kth element in the set corresponds to the kth weight in weights. Therefore, the weights must be unique.\n\nOf the variables in the set, at most two can be nonzero, and if two are nonzero, they must be adjacent in the ordering of the set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.SOS2([1.0, 3.0, 2.5]),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SOS2{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Indicator","page":"Standard form","title":"MathOptInterface.Indicator","text":"Indicator{A<:ActivationCondition,S<:AbstractScalarSet}(set::S)\n\nThe set corresponding to an indicator constraint.\n\nWhen A is ACTIVATE_ON_ZERO, this means: (y x) in 0 1 times mathbbR^n  y = 0 implies x in set\n\nWhen A is ACTIVATE_ON_ONE, this means: (y x) in 0 1 times mathbbR^n  y = 1 implies x in set\n\nNotes\n\nMost solvers expect that the first row of the function is interpretable as a variable index x_i (e.g., 1.0 * x + 0.0). An error will be thrown if this is not the case.\n\nExample\n\nThe constraint (y x) in 0 1 times mathbbR^2  y = 1 implies x_1 + x_2 leq 9  is defined as\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> y, _ = MOI.add_constrained_variable(model, MOI.ZeroOne())\n(MOI.VariableIndex(3), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(3))\n\njulia> f = MOI.VectorAffineFunction(\n           [\n               MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, y)),\n               MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(1.0, x[1])),\n               MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(1.0, x[2])),\n           ],\n           [0.0, 0.0],\n       )\n┌                                                         ┐\n│0.0 + 1.0 MOI.VariableIndex(3)                           │\n│0.0 + 1.0 MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(2)│\n└                                                         ┘\n\njulia> s = MOI.Indicator{MOI.ACTIVATE_ON_ONE}(MOI.LessThan(9.0))\nMathOptInterface.Indicator{MathOptInterface.ACTIVATE_ON_ONE, MathOptInterface.LessThan{Float64}}(MathOptInterface.LessThan{Float64}(9.0))\n\njulia> MOI.add_constraint(model, f, s)\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Indicator{MathOptInterface.ACTIVATE_ON_ONE, MathOptInterface.LessThan{Float64}}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Complements","page":"Standard form","title":"MathOptInterface.Complements","text":"Complements(dimension::Base.Integer)\n\nThe set corresponding to a mixed complementarity constraint.\n\nComplementarity constraints should be specified with an AbstractVectorFunction-in-Complements(dimension) constraint.\n\nThe dimension of the vector-valued function F must be dimension. This defines a complementarity constraint between the scalar function F[i] and the variable in F[i + dimension/2]. Thus, F[i + dimension/2] must be interpretable as a single variable x_i (e.g., 1.0 * x + 0.0), and dimension must be even.\n\nThe mixed complementarity problem consists of finding x_i in the interval [lb, ub] (i.e., in the set Interval(lb, ub)), such that the following holds:\n\nF_i(x) == 0 if lb_i < x_i < ub_i\nF_i(x) >= 0 if lb_i == x_i\nF_i(x) <= 0 if x_i == ub_i\n\nClassically, the bounding set for x_i is Interval(0, Inf), which recovers: 0 <= F_i(x) ⟂ x_i >= 0, where the ⟂ operator implies F_i(x) * x_i = 0.\n\nExample\n\nThe problem:\n\nx -in- Interval(-1, 1)\n[-4 * x - 3, x] -in- Complements(2)\n\ndefines the mixed complementarity problem where the following holds:\n\n-4 * x - 3 == 0 if -1 < x < 1\n-4 * x - 3 >= 0 if x == -1\n-4 * x - 3 <= 0 if x == 1\n\nThere are three solutions:\n\nx = -3/4 with F(x) = 0\nx = -1 with F(x) = 1\nx = 1 with F(x) = -7\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x, _ = MOI.add_constrained_variable(model, MOI.Interval(-1.0, 1.0));\n\njulia> MOI.add_constraint(\n            model,\n            MOI.Utilities.vectorize([-4.0 * x - 3.0, x]),\n            MOI.Complements(2),\n        )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Complements}(1)\n\nThe function F can also be defined in terms of single variables. For example, the problem:\n\n[x_3, x_4] -in- Nonnegatives(2)\n[x_1, x_2, x_3, x_4] -in- Complements(4)\n\ndefines the complementarity problem where 0 <= x_1 ⟂ x_3 >= 0 and 0 <= x_2 ⟂ x_4 >= 0.\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 4);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x[3:4]), MOI.Nonnegatives(2))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.add_constraint(\n            model,\n            MOI.VectorOfVariables(x),\n            MOI.Complements(4),\n        )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Complements}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.HyperRectangle","page":"Standard form","title":"MathOptInterface.HyperRectangle","text":"HyperRectangle(lower::Vector{T}, upper::Vector{T}) where {T}\n\nThe set x in barmathbbR^d x_i in lower_i upper_i forall i=1ldotsd.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3)\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.HyperRectangle(zeros(3), ones(3)),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.HyperRectangle{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Scaled","page":"Standard form","title":"MathOptInterface.Scaled","text":"struct Scaled{S<:AbstractVectorSet} <: AbstractVectorSet\n    set::S\nend\n\nGiven a vector a in mathbbR^d and a set representing the set mathcalS in mathbbR^d such that Utilities.set_dot for x in mathcalS and y in mathcalS^* is\n\nsum_i=1^d a_i x_i y_i\n\nthe set Scaled(set) is defined as\n\n (sqrta_1 x_1 sqrta_2 x_2 ldots sqrta_d x_d)  x in S \n\nExample\n\nThis can be used to scale a vector of numbers\n\njulia> import MathOptInterface as MOI\n\njulia> set = MOI.PositiveSemidefiniteConeTriangle(2)\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\njulia> a = MOI.Utilities.SetDotScalingVector{Float64}(set)\n3-element MathOptInterface.Utilities.SetDotScalingVector{Float64, MathOptInterface.PositiveSemidefiniteConeTriangle}:\n 1.0\n 1.4142135623730951\n 1.0\n\njulia> using LinearAlgebra\n\njulia> MOI.Utilities.operate(*, Float64, Diagonal(a), ones(3))\n3-element Vector{Float64}:\n 1.0\n 1.4142135623730951\n 1.0\n\nIt can be also used to scale a vector of function\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> func = MOI.VectorOfVariables(x)\n┌                    ┐\n│MOI.VariableIndex(1)│\n│MOI.VariableIndex(2)│\n│MOI.VariableIndex(3)│\n└                    ┘\n\njulia> set = MOI.PositiveSemidefiniteConeTriangle(2)\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\njulia> MOI.Utilities.operate(*, Float64, Diagonal(a), func)\n┌                                             ┐\n│0.0 + 1.0 MOI.VariableIndex(1)               │\n│0.0 + 1.4142135623730951 MOI.VariableIndex(2)│\n│0.0 + 1.0 MOI.VariableIndex(3)               │\n└                                             ┘\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Constraint-programming-sets","page":"Standard form","title":"Constraint programming sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nReified\nTable","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AllDifferent","page":"Standard form","title":"MathOptInterface.AllDifferent","text":"AllDifferent(dimension::Int)\n\nThe set x in mathbbZ^d such that no two elements in x take the same value and dimension = d.\n\nAlso known as\n\nThis constraint is called all_different in MiniZinc, and is sometimes also called distinct.\n\nExample\n\nTo enforce x[1] != x[2] AND x[1] != x[3] AND x[2] != x[3]:\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.AllDifferent(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.AllDifferent}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.BinPacking","page":"Standard form","title":"MathOptInterface.BinPacking","text":"BinPacking(c::T, w::Vector{T}) where {T}\n\nThe set x in mathbbZ^d where d = length(w), such that each item i in 1:d of weight w[i] is put into bin x[i], and the total weight of each bin does not exceed c.\n\nThere are additional assumptions that the capacity, c, and the weights, w, must all be non-negative.\n\nThe bin numbers depend on the bounds of x, so they may be something other than the integers 1:d.\n\nAlso known as\n\nThis constraint is called bin_packing in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> bins = MOI.add_variables(model, 5)\n5-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n\njulia> weights = Float64[1, 1, 2, 2, 3]\n5-element Vector{Float64}:\n 1.0\n 1.0\n 2.0\n 2.0\n 3.0\n\njulia> MOI.add_constraint.(model, bins, MOI.Integer())\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(5)\n\njulia> MOI.add_constraint.(model, bins, MOI.Interval(4.0, 6.0))\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(2)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(5)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(bins), MOI.BinPacking(3.0, weights))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.BinPacking{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Circuit","page":"Standard form","title":"MathOptInterface.Circuit","text":"Circuit(dimension::Int)\n\nThe set x in 1d^d that constraints x to be a circuit, such that x_i = j means that j is the successor of i, and dimension = d.\n\nGraphs with multiple independent circuits, such as [2, 1, 3] and [2, 1, 4, 3], are not valid.\n\nAlso known as\n\nThis constraint is called circuit in MiniZinc, and it is equivalent to forming a (potentially sub-optimal) tour in the travelling salesperson problem.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Circuit(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Circuit}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountAtLeast","page":"Standard form","title":"MathOptInterface.CountAtLeast","text":"CountAtLeast(n::Int, d::Vector{Int}, set::Set{Int})\n\nThe set x in mathbbZ^d_1 + d_2 + ldots d_N, where x is partitioned into N subsets (x_1  ldots x_d_1, x_d_1 + 1  ldots x_d_1 + d_2 and so on), and at least n elements of each subset take one of the values in set.\n\nAlso known as\n\nThis constraint is called at_least in MiniZinc.\n\nExample\n\nTo ensure that 3 appears at least once in each of the subsets {a, b} and {b, c}:\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> a, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> b, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> c, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(3), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(3))\n\njulia> x, d, set = [a, b, b, c], [2, 2], [3]\n(MathOptInterface.VariableIndex[MOI.VariableIndex(1), MOI.VariableIndex(2), MOI.VariableIndex(2), MOI.VariableIndex(3)], [2, 2], [3])\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.CountAtLeast(1, d, Set(set)))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountAtLeast}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountBelongs","page":"Standard form","title":"MathOptInterface.CountBelongs","text":"CountBelongs(dimenson::Int, set::Set{Int})\n\nThe set (n x) in mathbbZ^1+d, such that n elements of the vector x take on of the values in set and dimension = 1 + d.\n\nAlso known as\n\nThis constraint is called among by MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> n, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n\njulia> set = Set([3, 4, 5])\nSet{Int64} with 3 elements:\n  5\n  4\n  3\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([n; x]), MOI.CountBelongs(4, set))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountBelongs}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountDistinct","page":"Standard form","title":"MathOptInterface.CountDistinct","text":"CountDistinct(dimension::Int)\n\nThe set (n x) in mathbbZ^1+d, such that the number of distinct values in x is n and dimension = 1 + d.\n\nAlso known as\n\nThis constraint is called nvalues in MiniZinc.\n\nExample\n\nTo model:\n\nif n == 1`, thenx[1] == x[2] == x[3]`\nif n == 2, then\nx[1] == x[2] != x[3] or\nx[1] != x[2] == x[3] or\nx[1] == x[3] != x[2]\nif n == 3, then x[1] != x[2], x[2] != x[3] and x[3] != x[1]\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> n, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(vcat(n, x)), MOI.CountDistinct(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountDistinct}(1)\n\nRelationship to AllDifferent\n\nWhen the first element is d, CountDistinct is equivalent to an AllDifferent constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountGreaterThan","page":"Standard form","title":"MathOptInterface.CountGreaterThan","text":"CountGreaterThan(dimension::Int)\n\nThe set (c y x) in mathbbZ^1+1+d, such that c is strictly greater than the number of occurances of y in x and dimension = 1 + 1 + d.\n\nAlso known as\n\nThis constraint is called count_gt in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> c, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> y, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([c; y; x]), MOI.CountGreaterThan(5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountGreaterThan}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Cumulative","page":"Standard form","title":"MathOptInterface.Cumulative","text":"Cumulative(dimension::Int)\n\nThe set (s d r b) in mathbbZ^3n+1, representing the cumulative global constraint, where n == length(s) == length(r) == length(b) and dimension = 3n + 1.\n\nCumulative requires that a set of tasks given by start times s, durations d, and resource requirements r, never requires more than the global resource bound b at any one time.\n\nAlso known as\n\nThis constraint is called cumulative in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> s = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> d = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n MOI.VariableIndex(6)\n\njulia> r = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(7)\n MOI.VariableIndex(8)\n MOI.VariableIndex(9)\n\njulia> b, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(10), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(10))\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([s; d; r; b]), MOI.Cumulative(10))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Cumulative}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Path","page":"Standard form","title":"MathOptInterface.Path","text":"Path(from::Vector{Int}, to::Vector{Int})\n\nGiven a graph comprised of a set of nodes 1..N and a set of arcs 1..E represented by an edge from node from[i] to node to[i], Path constrains the set (s t ns es) in (1N)times(1E)times01^Ntimes01^E, to form subgraph that is a path from node s to node t, where node n is in the path if ns[n] is 1, and edge e is in the path if es[e] is 1.\n\nThe path must be acyclic, and it must traverse all nodes n for which ns[n] is 1, and all edges e for which es[e] is 1.\n\nAlso known as\n\nThis constraint is called path in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> N, E = 4, 5\n(4, 5)\n\njulia> from = [1, 1, 2, 2, 3]\n5-element Vector{Int64}:\n 1\n 1\n 2\n 2\n 3\n\njulia> to = [2, 3, 3, 4, 4]\n5-element Vector{Int64}:\n 2\n 3\n 3\n 4\n 4\n\njulia> s, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> t, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> ns = MOI.add_variables(model, N)\n4-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n MOI.VariableIndex(6)\n\njulia> MOI.add_constraint.(model, ns, MOI.ZeroOne())\n4-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(5)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(6)\n\njulia> es = MOI.add_variables(model, E)\n5-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(7)\n MOI.VariableIndex(8)\n MOI.VariableIndex(9)\n MOI.VariableIndex(10)\n MOI.VariableIndex(11)\n\njulia> MOI.add_constraint.(model, es, MOI.ZeroOne())\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(7)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(8)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(9)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(10)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(11)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([s; t; ns; es]), MOI.Path(from, to))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Path}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Reified","page":"Standard form","title":"MathOptInterface.Reified","text":"Reified(set::AbstractSet)\n\nThe constraint z f(x) in Reified(S) ensures that f(x) in S if and only if z == 1, where z in 0 1.\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\n\njulia> z, _ = MOI.add_constrained_variable(model, MOI.ZeroOne())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1))\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(2)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.Utilities.vectorize([z, 2.0 * x]),\n           MOI.Reified(MOI.GreaterThan(1.0)),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Reified{MathOptInterface.GreaterThan{Float64}}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Table","page":"Standard form","title":"MathOptInterface.Table","text":"Table(table::Matrix{T}) where {T}\n\nThe set x in mathbbR^d where d = size(table, 2), such that x belongs to one row of table. That is, there exists some j in 1:size(table, 1), such that x[i] = table[j, i] for all i=1:size(table, 2).\n\nAlso known as\n\nThis constraint is called table in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3)\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> table = Float64[1 1 0; 0 1 1; 1 0 1; 1 1 1]\n4×3 Matrix{Float64}:\n 1.0  1.0  0.0\n 0.0  1.0  1.0\n 1.0  0.0  1.0\n 1.0  1.0  1.0\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Table(table))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Table{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Matrix-sets","page":"Standard form","title":"Matrix sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"Matrix sets are vectorized to be subtypes of AbstractVectorSet.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"For sets of symmetric matrices, storing both the (i, j) and (j, i) elements is redundant. Use the AbstractSymmetricMatrixSetTriangle set to represent only the vectorization of the upper triangular part of the matrix.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"When the matrix of expressions constrained to be in the set is not symmetric, and hence additional constraints are needed to force the equality of the (i, j) and (j, i) elements, use the AbstractSymmetricMatrixSetSquare set.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"The Bridges.Constraint.SquareBridge can transform a set from the square form to the triangular_form by adding appropriate constraints if the (i, j) and (j, i) expressions are different.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractSymmetricMatrixSetTriangle\nAbstractSymmetricMatrixSetSquare\nside_dimension\ntriangular_form","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSymmetricMatrixSetTriangle","page":"Standard form","title":"MathOptInterface.AbstractSymmetricMatrixSetTriangle","text":"abstract type AbstractSymmetricMatrixSetTriangle <: AbstractVectorSet end\n\nAbstract supertype for subsets of the (vectorized) cone of symmetric matrices, with side_dimension rows and columns. The entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row). A vectorized cone of dimension n corresponds to a square matrix with side dimension sqrt14 + 2 n - 12. (Because a d times d matrix has d(d + 1)  2 elements in the upper or lower triangle.)\n\nExample\n\nThe matrix\n\nbeginbmatrix\n  1  2  4\n  2  3  5\n  4  5  6\nendbmatrix\n\nhas side_dimension 3 and vectorization (1 2 3 4 5 6).\n\nNote\n\nTwo packed storage formats exist for symmetric matrices, the respective orders of the entries are:\n\nupper triangular column by column (or lower triangular row by row);\nlower triangular column by column (or upper triangular row by row).\n\nThe advantage of the first format is the mapping between the (i, j) matrix indices and the k index of the vectorized form. It is simpler and does not depend on the side dimension of the matrix. Indeed,\n\nthe entry of matrix indices (i, j) has vectorized index k = div((j - 1) * j, 2) + i if i leq j and k = div((i - 1) * i, 2) + j if j leq i;\nand the entry with vectorized index k has matrix indices i = div(1 + isqrt(8k - 7), 2) and j = k - div((i - 1) * i, 2) or j = div(1 + isqrt(8k - 7), 2) and i = k - div((j - 1) * j, 2).\n\nDuality note\n\nThe scalar product for the symmetric matrix in its vectorized form is the sum of the pairwise product of the diagonal entries plus twice the sum of the pairwise product of the upper diagonal entries; see [p. 634, 1]. This has important consequence for duality.\n\nConsider for example the following problem (PositiveSemidefiniteConeTriangle is a subtype of AbstractSymmetricMatrixSetTriangle)\n\nbeginalign*\n     max_x in mathbbR  x\n    \n     textst \n    (1 -x 1)  in textPositiveSemidefiniteConeTriangle(2)\nendalign*\n\nThe dual is the following problem\n\nbeginalign*\n     min_x in mathbbR^3  y_1 + y_3\n    \n     textst  2y_2  = 1\n      y  in textPositiveSemidefiniteConeTriangle(2)\nendalign*\n\nWhy do we use 2y_2 in the dual constraint instead of y_2 ? The reason is that 2y_2 is the scalar product between y and the symmetric matrix whose vectorized form is (0 1 0). Indeed, with our modified scalar products we have\n\nlangle\n(0 1 0)\n(y_1 y_2 y_3)\nrangle\n=\nmathrmtrace\nbeginpmatrix\n  0  1\n  1  0\nendpmatrix\nbeginpmatrix\n  y_1  y_2\n  y_2  y_3\nendpmatrix\n= 2y_2\n\nReferences\n\n[1] Boyd, S. and Vandenberghe, L.. Convex optimization. Cambridge university     press, 2004.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSymmetricMatrixSetSquare","page":"Standard form","title":"MathOptInterface.AbstractSymmetricMatrixSetSquare","text":"abstract type AbstractSymmetricMatrixSetSquare <: AbstractVectorSet end\n\nAbstract supertype for subsets of the (vectorized) cone of symmetric matrices, with side_dimension rows and columns. The entries of the matrix are given column by column (or equivalently, row by row). The matrix is both constrained to be symmetric and to have its triangular_form belong to the corresponding set. That is, if the functions in entries (i j) and (j i) are different, then a constraint will be added to make sure that the entries are equal.\n\nExample\n\nPositiveSemidefiniteConeSquare is a subtype of AbstractSymmetricMatrixSetSquare and constraining the matrix\n\nbeginbmatrix\n  1  -y\n  -z  0\nendbmatrix\n\nto be symmetric positive semidefinite can be achieved by constraining the vector (1 -z -y 0) (or (1 -y -z 0)) to belong to the PositiveSemidefiniteConeSquare(2). It both constrains y = z and (1 -y 0) (or (1 -z 0)) to be in PositiveSemidefiniteConeTriangle(2), since triangular_form(PositiveSemidefiniteConeSquare) is PositiveSemidefiniteConeTriangle.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.side_dimension","page":"Standard form","title":"MathOptInterface.side_dimension","text":"side_dimension(\n    set::Union{\n        AbstractSymmetricMatrixSetTriangle,\n        AbstractSymmetricMatrixSetSquare,\n        HermitianPositiveSemidefiniteConeTriangle,\n    },\n)\n\nSide dimension of the matrices in set.\n\nConvention\n\nBy convention, the side dimension should be stored in the side_dimension field. If this is not the case for a subtype of AbstractSymmetricMatrixSetTriangle, or AbstractSymmetricMatrixSetSquare you must implement this method.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.triangular_form","page":"Standard form","title":"MathOptInterface.triangular_form","text":"triangular_form(S::Type{<:AbstractSymmetricMatrixSetSquare})\ntriangular_form(set::AbstractSymmetricMatrixSetSquare)\n\nReturn the AbstractSymmetricMatrixSetTriangle corresponding to the vectorization of the upper triangular part of matrices in the AbstractSymmetricMatrixSetSquare set.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized matrix sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"PositiveSemidefiniteConeTriangle\nPositiveSemidefiniteConeSquare\nHermitianPositiveSemidefiniteConeTriangle\nLogDetConeTriangle\nLogDetConeSquare\nRootDetConeTriangle\nRootDetConeSquare","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle","page":"Standard form","title":"MathOptInterface.PositiveSemidefiniteConeTriangle","text":"PositiveSemidefiniteConeTriangle(side_dimension::Int) <: AbstractSymmetricMatrixSetTriangle\n\nThe (vectorized) cone of symmetric positive semidefinite matrices, with non-negative side_dimension rows and columns.\n\nSee AbstractSymmetricMatrixSetTriangle for more details on the vectorized form.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeSquare","page":"Standard form","title":"MathOptInterface.PositiveSemidefiniteConeSquare","text":"PositiveSemidefiniteConeSquare(side_dimension::Int) <: AbstractSymmetricMatrixSetSquare\n\nThe cone of symmetric positive semidefinite matrices, with non-negative side length side_dimension.\n\nSee AbstractSymmetricMatrixSetSquare for more details on the vectorized form.\n\nThe entries of the matrix are given column by column (or equivalently, row by row).\n\nThe matrix is both constrained to be symmetric and to be positive semidefinite. That is, if the functions in entries (i j) and (j i) are different, then a constraint will be added to make sure that the entries are equal.\n\nExample\n\nConstraining the matrix\n\nbeginbmatrix\n  1  -y\n  -z  0\nendbmatrix\n\nto be symmetric positive semidefinite can be achieved by constraining the vector (1 -z -y 0) (or (1 -y -z 0)) to belong to the PositiveSemidefiniteConeSquare(2).\n\nIt both constrains y = z and (1 -y 0) (or (1 -z 0)) to be in PositiveSemidefiniteConeTriangle(2).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle","page":"Standard form","title":"MathOptInterface.HermitianPositiveSemidefiniteConeTriangle","text":"HermitianPositiveSemidefiniteConeTriangle(side_dimension::Int) <: AbstractVectorSet\n\nThe (vectorized) cone of Hermitian positive semidefinite matrices, with non-negative side_dimension rows and columns.\n\nBecaue the matrix is Hermitian, the diagonal elements are real, and the complex-valued lower triangular entries are obtained as the conjugate of corresponding upper triangular entries.\n\nVectorization format\n\nThe vectorized form starts with real part of the entries of the upper triangular part of the matrix, given column by column as explained in AbstractSymmetricMatrixSetSquare.\n\nIt is then followed by the imaginary part of the off-diagonal entries of the upper triangular part, also given column by column.\n\nFor example, the matrix\n\nbeginbmatrix\n  1  2 + 7im  4 + 8im\n  2 - 7im  3  5 + 9im\n  4 - 8im  5 - 9im  6\nendbmatrix\n\nhas side_dimension 3 and is represented as the vector 1 2 3 4 5 6 7 8 9.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LogDetConeTriangle","page":"Standard form","title":"MathOptInterface.LogDetConeTriangle","text":"LogDetConeTriangle(side_dimension::Int)\n\nThe log-determinant cone  (t u X) in mathbbR^2 + d(d+1)2  t le u log(det(Xu)) u  0 , where the matrix X is represented in the same symmetric packed format as in the PositiveSemidefiniteConeTriangle.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; X]),\n           MOI.LogDetConeTriangle(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.LogDetConeTriangle}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LogDetConeSquare","page":"Standard form","title":"MathOptInterface.LogDetConeSquare","text":"LogDetConeSquare(side_dimension::Int)\n\nThe log-determinant cone  (t u X) in mathbbR^2 + d^2  t le u log(det(Xu)) X text symmetric u  0 , where the matrix X is represented in the same format as in the PositiveSemidefiniteConeSquare.\n\nSimilarly to PositiveSemidefiniteConeSquare, constraints are added to ensure that X is symmetric.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 4), 2, 2)\n2×2 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.LogDetConeSquare(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.LogDetConeSquare}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RootDetConeTriangle","page":"Standard form","title":"MathOptInterface.RootDetConeTriangle","text":"RootDetConeTriangle(side_dimension::Int)\n\nThe root-determinant cone  (t X) in mathbbR^1 + d(d+1)2  t le det(X)^1d , where the matrix X is represented in the same symmetric packed format as in the PositiveSemidefiniteConeTriangle.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; X]),\n           MOI.RootDetConeTriangle(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeTriangle}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RootDetConeSquare","page":"Standard form","title":"MathOptInterface.RootDetConeSquare","text":"RootDetConeSquare(side_dimension::Int)\n\nThe root-determinant cone  (t X) in mathbbR^1 + d^2  t le det(X)^1d X text symmetric , where the matrix X is represented in the same format as PositiveSemidefiniteConeSquare.\n\nSimilarly to PositiveSemidefiniteConeSquare, constraints are added to ensure that X is symmetric.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 4), 2, 2)\n2×2 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.RootDetConeSquare(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeSquare}(1)\n\n\n\n\n\n","category":"type"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"EditURL = \"https://github.com/darnstrom/DAQP.jl/blob/v0.5.1/README.md\"","category":"page"},{"location":"packages/DAQP/#DAQP.jl","page":"darnstrom/DAQP.jl","title":"DAQP.jl","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"(Image: ) (Image: )","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"DAQP.jl is a Julia wrapper for the Quadratic Programming solver DAQP.","category":"page"},{"location":"packages/DAQP/#License","page":"darnstrom/DAQP.jl","title":"License","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"DAQP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"The underlying solver, darnstrom/daqp is licensed under the MIT license.","category":"page"},{"location":"packages/DAQP/#Installation","page":"darnstrom/DAQP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"Install DAQP.jl using the Julia package manager:","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"import Pkg\nPkg.add(\"DAQP\")","category":"page"},{"location":"packages/DAQP/#Use-with-JuMP","page":"darnstrom/DAQP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"To use DAQP with JuMP, do:","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"using JuMP, DAQP\nmodel = Model(DAQP.Optimizer)","category":"page"},{"location":"packages/DAQP/#Documentation","page":"darnstrom/DAQP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"General information about the solver is available at https://darnstrom.github.io/daqp/, and specifics for the Julia interface are available at https://darnstrom.github.io/daqp/start/julia.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"EditURL = \"portfolio.jl\"","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Portfolio-optimization","page":"Portfolio optimization","title":"Portfolio optimization","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Optimization models play an increasingly important role in financial decisions. Many computational finance problems can be solved efficiently using modern optimization techniques.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial solves the famous Markowitz Portfolio Optimization problem with data from lecture notes from a course taught at Georgia Tech by Shabbir Ahmed.","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Required-packages","page":"Portfolio optimization","title":"Required packages","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"using JuMP\nimport DataFrames\nimport Ipopt\nimport MultiObjectiveAlgorithms as MOA\nimport Plots\nimport Statistics\nimport StatsPlots","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Formulation","page":"Portfolio optimization","title":"Formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Suppose we are considering investing 1000 dollars in three non-dividend paying stocks, IBM (IBM), Walmart (WMT), and Southern Electric (SEHI), for a one-month period.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We will use the initial money to buy shares of the three stocks at the current market prices, hold these for one month, and sell the shares off at the prevailing market prices at the end of the month.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"As a rational investor, we hope to make some profit out of this endeavor, that is, the return on our investment should be positive.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Suppose we bought a stock at p dollars per share in the beginning of the month, and sold it off at s dollars per share at the end of the month. Then the one-month return on a share of the stock is fracs-pp.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Since the stock prices are quite uncertain, so is the end-of-month return on our investment. Our goal is to invest in such a way that the expected end-of-month return is at least $50 or 5%. Furthermore, we want to make sure that the “risk” of not achieving our desired return is minimum.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Note that we are solving the problem under the following assumptions:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We can trade any continuum of shares.\nNo short-selling is allowed.\nThere are no transaction costs.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We model this problem by taking decision variables x_i i=123 denoting the dollars invested in each of the 3 stocks.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Let us denote by tilder_i the random variable corresponding to the monthly return (increase in the stock price) per dollar for stock i.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Then, the return (or profit) on x_i dollars invested in stock i is tilder_i x_i and the total (random) return on our investment is sum_i=1^3 tilder_i x_i The expected return on our investment is then mathbbEleftsum_i=1^3 tilder_i x_iright=sum_i=1^3 overliner_i x_i where overliner_i is the expected value of the tilder_i","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Now we need to quantify the notion of “risk” in our investment.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Markowitz, in his Nobel prize winning work, showed that a rational investor’s notion of minimizing risk can be closely approximated by minimizing the variance of the return of the investment portfolio. This variance is given by:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"operatornameVarleftsum_i=1^3 tilder_i x_iright = sum_i=1^3 sum_j=1^3 x_i x_j sigma_i j","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"where sigma_i j is the covariance of the return of stock i with stock j.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Note that the right hand side of the equation is the most reduced form of the expression and we have not shown the intermediate steps involved in getting to this form. We can also write this equation as:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"operatornameVarleftsum_i=1^3 tilder_i x_iright =x^top Q x","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Where Q is the covariance matrix for the random vector tilder.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Finally, we can write the model as:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"beginaligned\nmin x^top Q x \ntext  st  sum_i=1^3 x_i leq 1000 \noverliner^top x geq 50 \nx geq 0\nendaligned","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Data","page":"Portfolio optimization","title":"Data","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"For the data in our problem, we use the stock prices given below, in monthly values from November 2000, through November 2001.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"df = DataFrames.DataFrame(\n    [\n        93.043 51.826 1.063\n        84.585 52.823 0.938\n        111.453 56.477 1.000\n        99.525 49.805 0.938\n        95.819 50.287 1.438\n        114.708 51.521 1.700\n        111.515 51.531 2.540\n        113.211 48.664 2.390\n        104.942 55.744 3.120\n        99.827 47.916 2.980\n        91.607 49.438 1.900\n        107.937 51.336 1.750\n        115.590 55.081 1.800\n    ],\n    [:IBM, :WMT, :SEHI],\n)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Next, we compute the percentage return for the stock in each month:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"returns = diff(Matrix(df); dims = 1) ./ Matrix(df[1:end-1, :])","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The expected monthly return is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"r = vec(Statistics.mean(returns; dims = 1))","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"and the covariance matrix is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Q = Statistics.cov(returns)","category":"page"},{"location":"tutorials/nonlinear/portfolio/#JuMP-formulation","page":"Portfolio optimization","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:3] >= 0)\n@objective(model, Min, x' * Q * x)\n@constraint(model, sum(x) <= 1000)\n@constraint(model, r' * x >= 50)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The optimal allocation of our assets is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"value.(x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"So we spend $497 on IBM, and $503 on SEHI. This results in a variance of:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"scalar_variance = value(x' * Q * x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"and an expected return of:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"scalar_return = value(r' * x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Multi-objective-portfolio-optimization","page":"Portfolio optimization","title":"Multi-objective portfolio optimization","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The previous model returned a single solution that minimized the variance, ensuring that our expected return was at least $50. In practice, we might be willing to accept a slightly higher variance if it meant a much increased expected return. To explore this problem space, we can instead formulate our portfolio optimization problem with two objectives:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"to minimize the variance\nto maximize the expected return","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The solution to this bi-objective problem is the efficient frontier of modern portfolio theory, and each point in the solution is a point with the best return for a fixed level of risk.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"model = Model(() -> MOA.Optimizer(Ipopt.Optimizer))\nset_silent(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We also need to choose a solution algorithm for MOA. For our problem, the efficient frontier will have an infinite number of solutions. Since we cannot find all of the solutions, we choose an approximation algorithm and limit the number of solution points that are returned:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"set_optimizer_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\nset_optimizer_attribute(model, MOA.SolutionLimit(), 50)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Now we can define the rest of the model:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) <= 1000)\n@expression(model, variance, x' * Q * x)\n@expression(model, expected_return, r' * x)\n# We want to minimize variance and maximize expected return, but we must pick\n# a single objective sense `Min`, and negate any `Max` objectives:\n@objective(model, Min, [variance, -expected_return])\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The algorithm found 50 different solutions. Let's plot them to see how they differ:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"objective_space = Plots.hline(\n    [scalar_return];\n    label = \"Single-objective solution\",\n    linecolor = :red,\n)\nPlots.vline!(objective_space, [scalar_variance]; label = \"\", linecolor = :red)\nPlots.scatter!(\n    objective_space,\n    [value(variance; result = i) for i in 1:result_count(model)],\n    [value(expected_return; result = i) for i in 1:result_count(model)];\n    xlabel = \"Variance\",\n    ylabel = \"Expected Return\",\n    label = \"\",\n    title = \"Objective space\",\n    markercolor = \"white\",\n    markersize = 5,\n    legend = :bottomright,\n)\nfor i in 1:result_count(model)\n    y = objective_value(model; result = i)\n    Plots.annotate!(objective_space, y[1], -y[2], (i, 3))\nend\n\ndecision_space = StatsPlots.groupedbar(\n    vcat([value.(x; result = i)' for i in 1:result_count(model)]...);\n    bar_position = :stack,\n    label = [\"IBM\" \"WMT\" \"SEHI\"],\n    xlabel = \"Solution #\",\n    ylabel = \"Investment (\\$)\",\n    title = \"Decision space\",\n)\nPlots.plot(objective_space, decision_space; layout = (2, 1), size = (600, 600))","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Perhaps our trade-off wasn't so bad after all. Our original solution corresponded to picking a solution #17. If we buy more SEHI, we can increase the return, but the variance also increases. If we buy less SEHI, such as a solution like #5 or #6, then we can achieve the corresponding return without deploying all of our capital. We should also note that at no point should we buy WMT.","category":"page"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"EditURL = \"https://github.com/joaquimg/BilevelJuMP.jl/blob/v0.6.2/README.md\"","category":"page"},{"location":"packages/BilevelJuMP/#BilevelJuMP.jl","page":"joaquimg/BilevelJuMP.jl","title":"BilevelJuMP.jl","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"(Image: Build Status) (Image: Codecov branch) (Image: Docs dev) (Image: Docs stable)","category":"page"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"BilevelJuMP.jl is a JuMP extension for modeling and solving bilevel optimization problems.","category":"page"},{"location":"packages/BilevelJuMP/#License","page":"joaquimg/BilevelJuMP.jl","title":"License","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"BilevelJuMP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/BilevelJuMP/#Documentation","page":"joaquimg/BilevelJuMP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"You can find the documentation at https://joaquimg.github.io/BilevelJuMP.jl/stable/.","category":"page"},{"location":"packages/BilevelJuMP/#Help","page":"joaquimg/BilevelJuMP.jl","title":"Help","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"If you need help, please open a GitHub issue.","category":"page"},{"location":"packages/BilevelJuMP/#Example","page":"joaquimg/BilevelJuMP.jl","title":"Example","text":"","category":"section"},{"location":"packages/BilevelJuMP/#Install","page":"joaquimg/BilevelJuMP.jl","title":"Install","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"import Pkg\nPkg.add(\"BilevelJuMP\")\nPkg.add(\"HiGHS\")","category":"page"},{"location":"packages/BilevelJuMP/#Run","page":"joaquimg/BilevelJuMP.jl","title":"Run","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"using BilevelJuMP, HiGHS\n\nmodel = BilevelModel(\n    HiGHS.Optimizer,\n    mode = BilevelJuMP.FortunyAmatMcCarlMode(primal_big_M = 100, dual_big_M = 100)\n)\n\n@variable(Lower(model), x)\n@variable(Upper(model), y)\n\n@objective(Upper(model), Min, 3x + y)\n@constraints(Upper(model), begin\n    x <= 5\n    y <= 8\n    y >= 0\nend)\n\n@objective(Lower(model), Min, -x)\n@constraints(Lower(model), begin\n     x +  y <= 8\n    4x +  y >= 8\n    2x +  y <= 13\n    2x - 7y <= 0\nend)\n\noptimize!(model)\n\nobjective_value(model) # = 3 * (3.5 * 8/15) + 8/15 # = 6.13...\nvalue(x) # = 3.5 * 8/15 # = 1.86...\nvalue(y) # = 8/15 # = 0.53...","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"EditURL = \"https://github.com/oxfordcontrol/Clarabel.jl/blob/v0.6.0/README.md\"","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<h1 align=\"center\" margin=0px>\n  <img src=\"https://github.com/oxfordcontrol/ClarabelDocs/blob/main/docs/src/assets/logo-banner-light-jl.png?raw=true\" class=\"display-light-only\" width=60%>\n  <img src=\"https://github.com/oxfordcontrol/ClarabelDocs/blob/main/docs/src/assets/logo-banner-dark-jl.png?raw=true\" class=\"display-dark-only\"   width=60%>\n  <br>\nInterior Point Conic Optimization for Julia\n</h1>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<p align=\"center\">\n   <a href=\"https://github.com/oxfordcontrol/Clarabel.jl/actions\"><img src=\"https://github.com/oxfordcontrol/Clarabel.jl/workflows/ci/badge.svg?branch=main\"></a>\n  <a href=\"https://codecov.io/gh/oxfordcontrol/Clarabel.jl\"><img src=\"https://codecov.io/gh/oxfordcontrol/Clarabel.jl/branch/main/graph/badge.svg?raw=true\"></a>\n  <a href=\"https://oxfordcontrol.github.io/ClarabelDocs/stable\"><img src=\"https://img.shields.io/badge/Documentation-stable-purple.svg?raw=true\"></a>\n  <a href=\"https://opensource.org/licenses/Apache-2.0\"><img src=\"https://img.shields.io/badge/License-Apache%202.0-blue.svg?raw=true\"></a>\n  <a href=\"https://github.com/oxfordcontrol/Clarabel.jl/releases\"><img src=\"https://img.shields.io/badge/Release-v0.6.0-blue.svg?raw=true\"></a>\n</p>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<p align=\"center\">\n  <a href=\"#features\">Features</a> •\n  <a href=\"#installation\">Installation</a> •\n  <a href=\"#license-\">License</a> •\n  <a href=\"https://oxfordcontrol.github.io/ClarabelDocs/stable\">Documentation</a>\n</p>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel.jl is a Julia implementation of an interior point numerical solver for convex optimization problems using a novel homogeneous embedding.  Clarabel.jl solves the following problem:","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"beginarrayr\ntextminimize  frac12x^T P x + q^T x2ex\ntextsubject to  Ax + s = b 1ex\n         s in mathcalK\nendarray","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"with decision variables x in mathbbR^n, s in mathbbR^m and data matrices P=P^top succeq 0, q in mathbbR^n, A in mathbbR^m times n, and b in mathbbR^m. The convex set mathcalK is a composition of convex cones.","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"For more information see the Clarabel Documentation (stable |  dev).","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel is also available in a Rust implementation with additional language interfaces.  See here.","category":"page"},{"location":"packages/Clarabel/#Features","page":"oxfordcontrol/Clarabel.jl","title":"Features","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Versatile: Clarabel.jl solves linear programs (LPs), quadratic programs (QPs), second-order cone programs (SOCPs) and semidefinite programs (SDPs). It also solves problems with exponential, power cone and generalized power cone constraints.\nQuadratic objectives: Unlike interior point solvers based on the standard homogeneous self-dual embedding (HSDE), Clarabel.jl handles quadratic objectives without requiring any epigraphical reformulation of the objective.   It can therefore be significantly faster than other HSDE-based solvers for problems with quadratic objective functions.\nInfeasibility detection: Infeasible problems are detected using a homogeneous embedding technique.\nJuMP / Convex.jl support: We provide an interface to MathOptInterface (MOI), which allows you to describe your problem in JuMP and Convex.jl.\nArbitrary precision types: You can solve problems with any floating point precision, for example, Float32 or Julia's BigFloat type, using either the native interface, or via MathOptInterface / Convex.jl.\nOpen Source: Our code is available on GitHub and distributed under the Apache 2.0 License","category":"page"},{"location":"packages/Clarabel/#Installation","page":"oxfordcontrol/Clarabel.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel.jl can be added via the Julia package manager (type ]): pkg> add Clarabel","category":"page"},{"location":"packages/Clarabel/#License","page":"oxfordcontrol/Clarabel.jl","title":"License 🔍","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"This project is licensed under the Apache License  2.0 - see the LICENSE.md file for details.","category":"page"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/modification.md\"","category":"page"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/modification/#Modifications","page":"Modifications","title":"Modifications","text":"","category":"section"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"modify\nAbstractFunctionModification\nScalarConstantChange\nVectorConstantChange\nScalarCoefficientChange\nMultirowChange","category":"page"},{"location":"moi/reference/modification/#MathOptInterface.modify","page":"Modifications","title":"MathOptInterface.modify","text":"Constraint Function\n\nmodify(model::ModelLike, ci::ConstraintIndex, change::AbstractFunctionModification)\n\nApply the modification specified by change to the function of constraint ci.\n\nAn ModifyConstraintNotAllowed error is thrown if modifying constraints is not supported by the model model.\n\nExamples\n\nmodify(model, ci, ScalarConstantChange(10.0))\n\nObjective Function\n\nmodify(model::ModelLike, ::ObjectiveFunction, change::AbstractFunctionModification)\n\nApply the modification specified by change to the objective function of model. To change the function completely, call set instead.\n\nAn ModifyObjectiveNotAllowed error is thrown if modifying objectives is not supported by the model model.\n\nExamples\n\nmodify(model, ObjectiveFunction{ScalarAffineFunction{Float64}}(), ScalarConstantChange(10.0))\n\nMultiple modifications in Constraint Functions\n\nmodify(\n    model::ModelLike,\n    cis::AbstractVector{<:ConstraintIndex},\n    changes::AbstractVector{<:AbstractFunctionModification},\n)\n\nApply multiple modifications specified by changes to the functions of constraints cis.\n\nA ModifyConstraintNotAllowed error is thrown if modifying constraints is not supported by model.\n\nExamples\n\nmodify(\n    model,\n    [ci, ci],\n    [\n        ScalarCoefficientChange{Float64}(VariableIndex(1), 1.0),\n        ScalarCoefficientChange{Float64}(VariableIndex(2), 0.5),\n    ],\n)\n\nMultiple modifications in the Objective Function\n\nmodify(\n    model::ModelLike,\n    attr::ObjectiveFunction,\n    changes::AbstractVector{<:AbstractFunctionModification},\n)\n\nApply multiple modifications specified by changes to the functions of constraints cis.\n\nA ModifyObjectiveNotAllowed error is thrown if modifying objective coefficients is not supported by model.\n\nExamples\n\nmodify(\n    model,\n    ObjectiveFunction{ScalarAffineFunction{Float64}}(),\n    [\n        ScalarCoefficientChange{Float64}(VariableIndex(1), 1.0),\n        ScalarCoefficientChange{Float64}(VariableIndex(2), 0.5),\n    ],\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/modification/#MathOptInterface.AbstractFunctionModification","page":"Modifications","title":"MathOptInterface.AbstractFunctionModification","text":"AbstractFunctionModification\n\nAn abstract supertype for structs which specify partial modifications to functions, to be used for making small modifications instead of replacing the functions entirely.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.ScalarConstantChange","page":"Modifications","title":"MathOptInterface.ScalarConstantChange","text":"ScalarConstantChange{T}(new_constant::T)\n\nA struct used to request a change in the constant term of a scalar-valued function.\n\nApplicable to ScalarAffineFunction and ScalarQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.VectorConstantChange","page":"Modifications","title":"MathOptInterface.VectorConstantChange","text":"VectorConstantChange{T}(new_constant::Vector{T})\n\nA struct used to request a change in the constant vector of a vector-valued function.\n\nApplicable to VectorAffineFunction and VectorQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.ScalarCoefficientChange","page":"Modifications","title":"MathOptInterface.ScalarCoefficientChange","text":"ScalarCoefficientChange{T}(variable::VariableIndex, new_coefficient::T)\n\nA struct used to request a change in the linear coefficient of a single variable in a scalar-valued function.\n\nApplicable to ScalarAffineFunction and ScalarQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.MultirowChange","page":"Modifications","title":"MathOptInterface.MultirowChange","text":"MultirowChange{T}(\n    variable::VariableIndex,\n    new_coefficients::Vector{Tuple{Int64,T}},\n) where {T}\n\nA struct used to request a change in the linear coefficients of a single variable in a vector-valued function.\n\nNew coefficients are specified by (output_index, coefficient) tuples.\n\nApplicable to VectorAffineFunction and VectorQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"EditURL = \"https://github.com/mariohsouto/ProxSDP.jl/blob/de648be90e63a69b8d8e4f82371760065d242fe9/README.md\"","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"<img src=\"https://raw.githubusercontent.com/mariohsouto/ProxSDP_aux/19f261a912e29c77cb68ee2957a5b41128824f22/logo_proxSDP.png?raw=true\" alt=\"logo\"/>","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"(Image: ) (Image: ) (Image: )","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is an open-source semidefinite programming (SDP) solver based on the paper \"Exploiting Low-Rank Structure in Semidefinite Programming by Approximate Operator Splitting\".","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The main advantage of ProxSDP over other state-of-the-art solvers is the ability to exploit the low-rank structure inherent to several SDP problems.","category":"page"},{"location":"packages/ProxSDP/#Overview-of-problems-ProxSDP-can-solve","page":"mariohsouto/ProxSDP.jl","title":"Overview of problems ProxSDP can solve","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"General conic convex optimization problems with the presence of the positive semidefinite cone, second-order cone and positive orthant;\nSemidefinite relaxation of nonconvex problems, for example, max-cut, binary MIMO, optimal power flow, sensor localization, sum-of-squares;\nControl theory problems with LMI constraints;\nNuclear norm minimization problems, for example, matrix completion;","category":"page"},{"location":"packages/ProxSDP/#License","page":"mariohsouto/ProxSDP.jl","title":"License","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is licensed under the MIT License.","category":"page"},{"location":"packages/ProxSDP/#Installation","page":"mariohsouto/ProxSDP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Install ProxSDP using Julia package manager:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"import Pkg\nPkg.add(\"ProxSDP\")","category":"page"},{"location":"packages/ProxSDP/#Documentation","page":"mariohsouto/ProxSDP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The online documentation is available at https://mariohsouto.github.io/ProxSDP.jl/latest/.","category":"page"},{"location":"packages/ProxSDP/#Usage","page":"mariohsouto/ProxSDP.jl","title":"Usage","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Consider the semidefinite programming relaxation of the max-cut problem, as in:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"max   0.25 * W•X\ns.t.  diag(X) = 1,\n      X ≽ 0,","category":"page"},{"location":"packages/ProxSDP/#JuMP","page":"mariohsouto/ProxSDP.jl","title":"JuMP","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"This problem can be solved by the following code using ProxSDP and JuMP.","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"# Load packages\nusing ProxSDP, JuMP, LinearAlgebra\n# Number of vertices\nn = 4\n# Graph weights\nW = [\n    18.0  -5.0  -7.0  -6.0\n    -5.0   6.0   0.0  -1.0\n    -7.0   0.0   8.0  -1.0\n    -6.0  -1.0  -1.0   8.0\n]\n# Build Max-Cut SDP relaxation via JuMP\nmodel = Model(ProxSDP.Optimizer)\nset_optimizer_attribute(model, \"log_verbose\", true)\nset_optimizer_attribute(model, \"tol_gap\", 1e-4)\nset_optimizer_attribute(model, \"tol_feasibility\", 1e-4)\n@variable(model, X[1:n, 1:n], PSD)\n@objective(model, Max, 0.25 * LinearAlgebra.dot(W, X))\n@constraint(model, LinearAlgebra.diag(X) .== 1)\n# Solve optimization problem with ProxSDP\noptimize!(model)\n# Retrieve solution\nXsol = value.(X)","category":"page"},{"location":"packages/ProxSDP/#Convex.jl","page":"mariohsouto/ProxSDP.jl","title":"Convex.jl","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Another alternative is to use ProxSDP via Convex.jl:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"# Load packages\nusing Convex, ProxSDP\n# Number of vertices\nn = 4\n# Graph weights\nW = [\n    18.0  -5.0  -7.0  -6.0\n    -5.0   6.0   0.0  -1.0\n    -7.0   0.0   8.0  -1.0\n    -6.0  -1.0  -1.0   8.0\n]\n# Define optimization problem\nX = Semidefinite(n)\nproblem = maximize(0.25 * dot(W, X), diag(X) == 1)\n# Solve optimization problem with ProxSDP\nsolve!(problem, ProxSDP.Optimizer(log_verbose=true, tol_gap=1e-4, tol_feasibility=1e-4))\n# Get the objective value\nproblem.optval\n# Retrieve solution\nevaluate(X)","category":"page"},{"location":"packages/ProxSDP/#Citing-this-package","page":"mariohsouto/ProxSDP.jl","title":"Citing this package","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The published version of the paper can be found here and the arXiv version here.","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"We kindly request that you cite the paper as:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"@article{souto2020exploiting,\n  author = {Mario Souto and Joaquim D. Garcia and \\'Alvaro Veiga},\n  title = {Exploiting low-rank structure in semidefinite programming by approximate operator splitting},\n  journal = {Optimization},\n  pages = {1-28},\n  year  = {2020},\n  publisher = {Taylor & Francis},\n  doi = {10.1080/02331934.2020.1823387},\n  URL = {https://doi.org/10.1080/02331934.2020.1823387}\n}","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The preprint version of the paper can be found here.","category":"page"},{"location":"packages/ProxSDP/#Disclaimer","page":"mariohsouto/ProxSDP.jl","title":"Disclaimer","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is a research software, therefore it should not be used in production.\nPlease open an issue if you find any problems, developers will try to fix and find alternatives.\nThere is no continuous development for 32-bit systems, the package should work, but might reach some issues.\nProxSDP assumes primal and dual feasibility.","category":"page"},{"location":"packages/ProxSDP/#ROAD-MAP","page":"mariohsouto/ProxSDP.jl","title":"ROAD MAP","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Support for exponential and power cones\nWarm start","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"EditURL = \"https://github.com/jump-dev/HiGHS.jl/blob/v1.7.4/README.md\"","category":"page"},{"location":"packages/HiGHS/#HiGHS.jl","page":"jump-dev/HiGHS.jl","title":"HiGHS.jl","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"HiGHS.jl is a wrapper for the HiGHS solver.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"It has two components:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/HiGHS/#Affiliation","page":"jump-dev/HiGHS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"This wrapper is maintained by the JuMP community and is not an official project of the HiGHS developers.","category":"page"},{"location":"packages/HiGHS/#License","page":"jump-dev/HiGHS.jl","title":"License","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"HiGHS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The underlying solver, ERGO-Code/HiGHS, is licensed under the MIT license.","category":"page"},{"location":"packages/HiGHS/#Installation","page":"jump-dev/HiGHS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"Install HiGHS as follows:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"import Pkg\nPkg.add(\"HiGHS\")","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"In addition to installing the HiGHS.jl package, this will also download and install the HiGHS binaries. You do not need to install HiGHS separately.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/HiGHS/#Use-with-JuMP","page":"jump-dev/HiGHS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"To use HiGHS with JuMP, use HiGHS.Optimizer:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"using JuMP, HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_attribute(model, \"presolve\", \"on\")\nset_attribute(model, \"time_limit\", 60.0)","category":"page"},{"location":"packages/HiGHS/#MathOptInterface-API","page":"jump-dev/HiGHS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The HiGHS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/HiGHS/#Options","page":"jump-dev/HiGHS.jl","title":"Options","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"See the HiGHS documentation for a full list of the available options.","category":"page"},{"location":"packages/HiGHS/#C-API","page":"jump-dev/HiGHS.jl","title":"C API","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The C API can be accessed via HiGHS.Highs_xxx functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/FileFormats/reference.md\"","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/FileFormats/reference/#File-Formats","page":"API Reference","title":"File Formats","text":"","category":"section"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"Functions to help read and write MOI models to/from various file formats. See The FileFormats submodule for more details.","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"FileFormats.Model\nFileFormats.FileFormat\nFileFormats.CBF.Model\nFileFormats.LP.Model\nFileFormats.MOF.Model\nFileFormats.MPS.Model\nFileFormats.NL.Model\nFileFormats.SDPA.Model","category":"page"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.Model","page":"API Reference","title":"MathOptInterface.FileFormats.Model","text":"Model(\n    ;\n    format::FileFormat = FORMAT_AUTOMATIC,\n    filename::Union{Nothing, String} = nothing,\n    kwargs...\n)\n\nReturn model corresponding to the FileFormat format, or, if format == FORMAT_AUTOMATIC, guess the format from filename.\n\nThe filename argument is only needed if format == FORMAT_AUTOMATIC.\n\nkwargs are passed to the underlying model constructor.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.FileFormat","page":"API Reference","title":"MathOptInterface.FileFormats.FileFormat","text":"FileFormat\n\nList of accepted export formats.\n\nFORMAT_AUTOMATIC: try to detect the file format based on the file name\nFORMAT_CBF: the Conic Benchmark format\nFORMAT_LP: the LP file format\nFORMAT_MOF: the MathOptFormat file format\nFORMAT_MPS: the MPS file format\nFORMAT_NL: the AMPL .nl file format\nFORMAT_REW: the .rew file format, which is MPS with generic names\nFORMAT_SDPA: the SemiDefinite Programming Algorithm format\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.CBF.Model","page":"API Reference","title":"MathOptInterface.FileFormats.CBF.Model","text":"Model()\n\nCreate an empty instance of FileFormats.CBF.Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.LP.Model","page":"API Reference","title":"MathOptInterface.FileFormats.LP.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.LP.Model.\n\nKeyword arguments are:\n\nmaximum_length::Int=255: the maximum length for the name of a variable. lp_solve 5.0 allows only 16 characters, while CPLEX 12.5+ allow 255.\nwarn::Bool=false: print a warning when variables or constraints are renamed.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.MOF.Model","page":"API Reference","title":"MathOptInterface.FileFormats.MOF.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.MOF.Model.\n\nKeyword arguments are:\n\nprint_compact::Bool=false: print the JSON file in a compact format without spaces or newlines.\nwarn::Bool=false: print a warning when variables or constraints are renamed\ndifferentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation = MOI.Nonlinear.SparseReverseMode(): automatic differentiation backend to use when reading models with nonlinear constraints and objectives.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.MPS.Model","page":"API Reference","title":"MathOptInterface.FileFormats.MPS.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.MPS.Model.\n\nKeyword arguments are:\n\nwarn::Bool=false: print a warning when variables or constraints are renamed.\nprint_objsense::Bool=false: print the OBJSENSE section when writing\ngeneric_names::Bool=false: strip all names in the model and replace them with the generic names C$i and R$i for the i'th column and row respectively.\nquadratic_format::QuadraticFormat = kQuadraticFormatGurobi: specify the solver-specific extension used when writing the quadratic components of the model. Options are kQuadraticFormatGurobi, kQuadraticFormatCPLEX, and kQuadraticFormatMosek.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.NL.Model","page":"API Reference","title":"MathOptInterface.FileFormats.NL.Model","text":"Model(; use_nlp_block::Bool = true)\n\nCreate a new Optimizer object.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.SDPA.Model","page":"API Reference","title":"MathOptInterface.FileFormats.SDPA.Model","text":"Model(; number_type::Type = Float64)\n\nCreate an empty instance of FileFormats.SDPA.Model{number_type}.\n\nIt is important to be aware that the SDPA file format is interpreted in geometric form and not standard conic form. The standard conic form and geometric conic form are two dual standard forms for semidefinite programs (SDPs). The geometric conic form of an SDP is as follows:\n\nbeginalign\n min_y in mathbbR^m  b^T y\n\n textst  sum_i=1^m A_i y_i - C  in mathbbK\nendalign\n\nwhere mathcalK is a cartesian product of nonnegative orthant and positive semidefinite matrices that align with a block diagonal structure shared with the matrices A_i and C.\n\nIn other words, the geometric conic form contains free variables and affine constraints in either the nonnegative orthant or the positive semidefinite cone. That is, in the MathOptInterface's terminology, MOI.VectorAffineFunction-in-MOI.Nonnegatives and MOI.VectorAffineFunction-in-MOI.PositiveSemidefiniteConeTriangle constraints.\n\nThe corresponding standard conic form of the dual SDP is as follows:\n\nbeginalign\n max_X in mathbbK  texttr(CX)\n\n textst  texttr(A_iX)  = b_i  i = 1 ldots m\nendalign\n\nIn other words, the standard conic form contains nonnegative and positive semidefinite variables with equality constraints. That is, in the MathOptInterface's terminology, MOI.VectorOfVariables-in-MOI.Nonnegatives, MOI.VectorOfVariables-in-MOI.PositiveSemidefiniteConeTriangle and MOI.ScalarAffineFunction-in-MOI.EqualTo constraints.\n\nIf a model is in standard conic form, use Dualization.jl to transform it into the geometric conic form before writting it. Otherwise, the nonnegative (resp. positive semidefinite) variables will be bridged into free variables with affine constraints constraining them to belong to the nonnegative orthant (resp. positive semidefinite cone) by the MOI.Bridges.Constraint.VectorFunctionizeBridge. Moreover, equality constraints will be bridged into pairs of affine constraints in the nonnegative orthant by the MOI.Bridges.Constraint.SplitIntervalBridge and then the MOI.Bridges.Constraint.VectorizeBridge.\n\nIf a solver is in standard conic form, use Dualization.jl to transform the model read into standard conic form before copying it to the solver. Otherwise, the free variables will be bridged into pairs of variables in the nonnegative orthant by the MOI.Bridges.Variable.FreeBridge and affine constraints will be bridged into equality constraints by creating a slack variable by the MOI.Bridges.Constraint.VectorSlackBridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#Other-helpers","page":"API Reference","title":"Other helpers","text":"","category":"section"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"FileFormats.NL.SolFileResults","category":"page"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.NL.SolFileResults","page":"API Reference","title":"MathOptInterface.FileFormats.NL.SolFileResults","text":"SolFileResults(filename::String, model::Model)\n\nParse the .sol file filename created by solving model and return a SolFileResults struct.\n\nThe returned struct supports the MOI.get API for querying result attributes such as MOI.TerminationStatus, MOI.VariablePrimal, and MOI.ConstraintDual.\n\n\n\n\n\nSolFileResults(\n    raw_status::String,\n    termination_status::MOI.TerminationStatusCode,\n)\n\nReturn a SolFileResults struct with MOI.RawStatusString set to raw_status, MOI.TerminationStatus set to termination_status, and MOI.PrimalStatus and MOI.DualStatus set to NO_SOLUTION.\n\nAll other attributes are un-set.\n\n\n\n\n\n","category":"type"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"EditURL = \"https://github.com/COPT-Public/COPT.jl/blob/v1.1.11/README.md\"","category":"page"},{"location":"packages/COPT/#COPT.jl","page":"COPT-Public/COPT.jl","title":"COPT.jl","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT.jl is a wrapper for the COPT (Cardinal Optimizer), a mathematical optimization solver for large-scale optimization problems.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT includes high-performance solvers for LP, MIP, SOCP, convex QP/QCP and SDP.","category":"page"},{"location":"packages/COPT/#License","page":"COPT-Public/COPT.jl","title":"License","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT.jl is licensed under the MIT license.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/COPT/#Note","page":"COPT-Public/COPT.jl","title":"Note","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"When COPT is upgraded to a newer version, you may see an error message such as ERROR: COPT error 4: Unable to create COPT environment, which indicates that you will need to reapply and upgrade your COPT license files as well.","category":"page"},{"location":"packages/COPT/#Installation","page":"COPT-Public/COPT.jl","title":"Installation","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"Install COPT using the Julia package manager","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"import Pkg\nPkg.add(\"COPT\")","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"When there is no local version of COPT installed, installing COPT.jl will automatically download the necessary solver binaries.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"Without a license, you can solve small models for non-commercial purpose. We strongly recommend that you apply for a license by following the link above.","category":"page"},{"location":"packages/COPT/#Note-2","page":"COPT-Public/COPT.jl","title":"Note","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"MacOS Apple M1/ARM: on MacOS with Apple M1 chips, Intel based programs can run via Rosetta. When you installed the COPT binaries manually, then please make sure that the COPT build matches the Julia build. We recommend the Intel based COPT and Julia build, as the Apple M1/ARM build of Julia is experimental.","category":"page"},{"location":"packages/COPT/#Use-with-JuMP","page":"COPT-Public/COPT.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"To use COPT with JuMP, use COPT.Optimizer:","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"using JuMP\nusing COPT\nmodel = Model(COPT.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\nprint(model)\noptimize!(model)\n@show termination_status(model)\n@show primal_status(model)\n@show dual_status(model)\n@show objective_value(model)\n@show value(x)\n@show value(y)\n@show shadow_price(c1)\n@show shadow_price(c2)","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"To use the semidefinite programming solver in COPT with JuMP, use COPT.ConeOptimizer:","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"using JuMP\nusing COPT\nusing LinearAlgebra\nmodel = Model(COPT.ConeOptimizer)\nC = [1.0 -1.0; -1.0 2.0]\n@variable(model, X[1:2, 1:2], PSD)\n@variable(model, z[1:2] >= 0)\n@objective(model, Min, C ⋅ X)\n@constraint(model, c1, X[1, 1] - z[1] == 1)\n@constraint(model, c2, X[2, 2] - z[2] == 1)\noptimize!(model)\n@show termination_status(model)\n@show primal_status(model)\n@show dual_status(model)\n@show objective_value(model)\n@show value.(X)\n@show value.(z)\n@show shadow_price(c1)\n@show shadow_price(c2)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"EditURL = \"cannery.jl\"","category":"page"},{"location":"tutorials/linear/cannery/#The-cannery-problem","page":"The cannery problem","title":"The cannery problem","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial solves the cannery problem from Dantzig, Linear Programming and Extensions, Princeton University Press, Princeton, NJ, 1963. This class of problem is known as a transshipment problem.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"The purpose of this tutorial is to demonstrate how to use JSON data in the formulation of a JuMP model.","category":"page"},{"location":"tutorials/linear/cannery/#Required-packages","page":"The cannery problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"using JuMP\nimport HiGHS\nimport JSON","category":"page"},{"location":"tutorials/linear/cannery/#Formulation","page":"The cannery problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"The cannery problem assumes we are optimizing the shipment of cases of cans from production plants p in P to markets m in M.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Each production plant p has a capacity c_p, and each market m has a demand d_m. The shipping cost per case of cans from plant p to market m is d_pm.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We wish to find the distribution plan x_pm, the number of cases of cans to ship from plant p to market m, for p in P and m in M that minimizes the shipping costs. We can formulate our problem as the following linear program:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"beginaligned\nmin  sumlimits_p in Psumlimits_m in M d_pm x_pm \ntextst  sumlimits_m in M x_pm le c_p  forall p in P \n             sumlimits_p in P x_pm ge d_m  forall m in M \n             x_pm ge 0  forall p in P m in M\nendaligned","category":"page"},{"location":"tutorials/linear/cannery/#Data","page":"The cannery problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"A key feature of the tutorial is to demonstrate how to load data from JSON.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"For simplicity, we've hard-coded it below. But if the data was available as a .json file, we could use data = JSON.parsefile(filename) to read in the data.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"data = JSON.parse(\"\"\"\n{\n    \"plants\": {\n        \"Seattle\": {\"capacity\": 350},\n        \"San-Diego\": {\"capacity\": 600}\n    },\n    \"markets\": {\n        \"New-York\": {\"demand\": 300},\n        \"Chicago\": {\"demand\": 300},\n        \"Topeka\": {\"demand\": 300}\n    },\n    \"distances\": {\n        \"Seattle => New-York\": 2.5,\n        \"Seattle => Chicago\": 1.7,\n        \"Seattle => Topeka\": 1.8,\n        \"San-Diego => New-York\": 2.5,\n        \"San-Diego => Chicago\": 1.8,\n        \"San-Diego => Topeka\": 1.4\n    }\n}\n\"\"\")","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Create the set of plants:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"P = keys(data[\"plants\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Create the set of markets:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"M = keys(data[\"markets\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We also need a function to compute the distance from plant to market:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"distance(p::String, m::String) = data[\"distances\"][\"$(p) => $(m)\"]","category":"page"},{"location":"tutorials/linear/cannery/#JuMP-formulation","page":"The cannery problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Now we're ready to convert our mathematical formulation into a JuMP model.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"First, create a new JuMP model. Since we have a linear program, we'll use HiGHS as our optimizer:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Our decision variables are indexed over the set of plants and markets:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@variable(model, x[P, M] >= 0)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We need a constraint that each plant can ship no more than its capacity:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@constraint(model, [p in P], sum(x[p, :]) <= data[\"plants\"][p][\"capacity\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"and each market must receive at least its demand:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@constraint(model, [m in M], sum(x[:, m]) >= data[\"markets\"][m][\"demand\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Finally, our objective is to minimize the transportation distance:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@objective(model, Min, sum(distance(p, m) * x[p, m] for p in P, m in M));\nnothing #hide","category":"page"},{"location":"tutorials/linear/cannery/#Solution","page":"The cannery problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Let's optimize and look at the solution:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"What's the optimal shipment?","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"for p in P, m in M\n    println(p, \" => \", m, \": \", value(x[p, m]))\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Utilities/overview.md\"","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Utilities/overview/#The-Utilities-submodule","page":"Overview","title":"The Utilities submodule","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The Utilities submodule provides a variety of functions and datastructures for managing MOI.ModelLike objects.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.Model","page":"Overview","title":"Utilities.Model","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities.Model provides an implementation of a ModelLike that efficiently supports all functions and sets defined within MOI. However, given the extensibility of MOI, this might not cover all use cases.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Create a model as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.UniversalFallback","page":"Overview","title":"Utilities.UniversalFallback","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities.UniversalFallback is a layer that sits on top of any ModelLike and provides non-specialized (slower) fallbacks for constraints and attributes that the underlying ModelLike does not support.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For example, Utilities.Model doesn't support some variable attributes like VariablePrimalStart, so JuMP uses a combination of Universal fallback and Utilities.Model as a generic problem cache:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nAdding a UniversalFallback means that your model will now support all constraints, even if the inner-model does not. This can lead to unexpected behavior.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.@model","page":"Overview","title":"Utilities.@model","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For advanced use cases that need efficient support for functions and sets defined outside of MOI (but still known at compile time), we provide the Utilities.@model macro.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The @model macro takes a name (for a new type, which must not exist yet), eight tuples specifying the types of constraints that are supported, and then a Bool indicating the type is a subtype of MOI.AbstractOptimizer (if true) or MOI.ModelLike (if false).","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The eight tuples are in the following order:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Un-typed scalar sets, for example, Integer\nTyped scalar sets, for example, LessThan\nUn-typed vector sets, for example, Nonnegatives\nTyped vector sets, for example, PowerCone\nUn-typed scalar functions, for example, VariableIndex\nTyped scalar functions, for example, ScalarAffineFunction\nUn-typed vector functions, for example, VectorOfVariables\nTyped vector functions, for example, VectorAffineFunction","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The tuples can contain more than one element. Typed-sets must be specified without their type parameter, for example, MOI.LessThan, not MOI.LessThan{Float64}.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Here is an example:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.@model(\n           MyNewModel,\n           (MOI.Integer,),                  # Un-typed scalar sets\n           (MOI.GreaterThan,),              # Typed scalar sets\n           (MOI.Nonnegatives,),             # Un-typed vector sets\n           (MOI.PowerCone,),                # Typed vector sets\n           (MOI.VariableIndex,),            # Un-typed scalar functions\n           (MOI.ScalarAffineFunction,),     # Typed scalar functions\n           (MOI.VectorOfVariables,),        # Un-typed vector functions\n           (MOI.VectorAffineFunction,),     # Typed vector functions\n           true,                            # <:MOI.AbstractOptimizer?\n       )\nMathOptInterface.Utilities.GenericOptimizer{T, MathOptInterface.Utilities.ObjectiveContainer{T}, MathOptInterface.Utilities.VariablesContainer{T}, MyNewModelFunctionConstraints{T}} where T\n\njulia> model = MyNewModel{Float64}()\nMOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MyNewModelFunctionConstraints{Float64}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nMyNewModel supports every VariableIndex-in-Set constraint, as well as VariableIndex, ScalarAffineFunction, and ScalarQuadraticFunction objective functions. Implement MOI.supports as needed to forbid constraint and objective function combinations.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"As another example, PATHSolver, which only supports VectorAffineFunction-in-Complements defines its optimizer as:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.@model(\n           PathOptimizer,\n           (),  # Scalar sets\n           (),  # Typed scalar sets\n           (MOI.Complements,),  # Vector sets\n           (),  # Typed vector sets\n           (),  # Scalar functions\n           (),  # Typed scalar functions\n           (),  # Vector functions\n           (MOI.VectorAffineFunction,),  # Typed vector functions\n           true,  # is_optimizer\n       )\nMathOptInterface.Utilities.GenericOptimizer{T, MathOptInterface.Utilities.ObjectiveContainer{T}, MathOptInterface.Utilities.VariablesContainer{T}, MathOptInterface.Utilities.VectorOfConstraints{MathOptInterface.VectorAffineFunction{T}, MathOptInterface.Complements}} where T","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"However, PathOptimizer does not support some VariableIndex-in-Set constraints, so we must explicitly define:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> function MOI.supports_constraint(\n           ::PathOptimizer,\n           ::Type{MOI.VariableIndex},\n           ::Type{Union{<:MOI.Semiinteger,MOI.Semicontinuous,MOI.ZeroOne,MOI.Integer}}\n       )\n           return false\n       end","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Finally, PATH doesn't support an objective function, so we need to add:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.supports(::PathOptimizer, ::MOI.ObjectiveFunction) = false","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nThis macro creates a new type, so it must be called from the top-level of a module, for example, it cannot be called from inside a function.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.CachingOptimizer","page":"Overview","title":"Utilities.CachingOptimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A [Utilities.CachingOptimizer] is an MOI layer that abstracts the difference between solvers that support incremental modification (for example, they support adding variables one-by-one), and solvers that require the entire problem in a single API call (for example, they only accept the A, b and c matrices of a linear program).","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"It has two parts:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A cache, where the model can be built and modified incrementally\nAn optimizer, which is used to solve the problem","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.CachingOptimizer(\n           MOI.Utilities.Model{Float64}(),\n           PathOptimizer{Float64}(),\n       )\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A Utilities.CachingOptimizer may be in one of three possible states:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"NO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer, and it is not synchronized with the cached model. Modifications are forwarded to the cache, but not to the optimizer.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model. Modifications are forwarded to the optimizer. If the optimizer does not support modifications, and error will be thrown.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.attach_optimizer to go from EMPTY_OPTIMIZER to ATTACHED_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.attach_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state ATTACHED_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"info: Info\nYou must be in ATTACHED_OPTIMIZER to use optimize!.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.reset_optimizer to go from ATTACHED_OPTIMIZER to EMPTY_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.reset_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"info: Info\nCalling MOI.empty!(model) also resets the state to EMPTY_OPTIMIZER. So after emptying a model, the modification will only be applied to the cache.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.drop_optimizer to go from any state to NO_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.drop_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state NO_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer nothing","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Pass an empty optimizer to Utilities.reset_optimizer to go from NO_OPTIMIZER to EMPTY_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.reset_optimizer(model, PathOptimizer{Float64}())\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Deciding when to attach and reset the optimizer is tedious, and you will often write code like this:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"try\n    # modification\ncatch\n    MOI.Utilities.reset_optimizer(model)\n    # Re-try modification\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"To make this easier, Utilities.CachingOptimizer has two modes of operation:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"AUTOMATIC: The CachingOptimizer changes its state when necessary. Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\nMANUAL: The user must change the state of the CachingOptimizer. Attempting to perform an operation in the incorrect state results in an error.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"By default, AUTOMATIC mode is chosen. However, you can create a CachingOptimizer in MANUAL mode as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.CachingOptimizer(\n           MOI.Utilities.Model{Float64}(),\n           MOI.Utilities.MANUAL,\n       )\nMOIU.CachingOptimizer{MOI.AbstractOptimizer, MOIU.Model{Float64}}\nin state NO_OPTIMIZER\nin mode MANUAL\nwith model cache MOIU.Model{Float64}\nwith optimizer nothing\n\njulia> MOI.Utilities.reset_optimizer(model, PathOptimizer{Float64}())\n\njulia> model\nMOIU.CachingOptimizer{MOI.AbstractOptimizer, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode MANUAL\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/#Printing","page":"Overview","title":"Printing","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use print to print the formulation of the model.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x_var\")\n\njulia> MOI.add_constraint(model, x, MOI.ZeroOne())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1)\n\njulia> MOI.set(model, MOI.ObjectiveFunction{typeof(x)}(), x)\n\njulia> MOI.set(model, MOI.ObjectiveSense(), MOI.MAX_SENSE)\n\njulia> print(model)\nMaximize VariableIndex:\n x_var\n\nSubject to:\n\nVariableIndex-in-ZeroOne\n x_var ∈ {0, 1}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.latex_formulation to display the model in LaTeX form:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.latex_formulation(model)\n$$ \\begin{aligned}\n\\max\\quad & x\\_var \\\\\n\\text{Subject to}\\\\\n & \\text{VariableIndex-in-ZeroOne} \\\\\n & x\\_var \\in \\{0, 1\\} \\\\\n\\end{aligned} $$","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nIn IJulia, calling print or ending a cell with Utilities.latex_formulation will render the model in LaTeX.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.PenaltyRelaxation","page":"Overview","title":"Utilities.PenaltyRelaxation","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Pass Utilities.PenaltyRelaxation to modify to relax the problem by adding penalized slack variables to the constraints. This is helpful when debugging sources of infeasible models.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x\")\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(Dict(c => 2.0)));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 x - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c]\n0.0 + 1.0 MOI.VariableIndex(2)","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"You can also modify a single constraint using Utilities.ScalarPenaltyRelaxation:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x\")\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> f = MOI.modify(model, c, MOI.Utilities.ScalarPenaltyRelaxation(2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 x - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> f\n0.0 + 1.0 MOI.VariableIndex(2)","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.MatrixOfConstraints","page":"Overview","title":"Utilities.MatrixOfConstraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The constraints of Utilities.Model are stored as a vector of tuples of function and set in a Utilities.VectorOfConstraints. Other representations can be used by parameterizing the type Utilities.GenericModel (resp. Utilities.GenericOptimizer). For instance, if all non-VariableIndex constraints are affine, the coefficients of all the constraints can be stored in a single sparse matrix using Utilities.MatrixOfConstraints. The constraints storage can even be customized up to a point where it exactly matches the storage of the solver of interest, in which case copy_to can be implemented for the solver by calling copy_to to this custom model.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For instance, Clp defines the following model:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"MOI.Utilities.@product_of_scalar_sets(LP, MOI.EqualTo{T}, MOI.LessThan{T}, MOI.GreaterThan{T})\nconst Model = MOI.Utilities.GenericModel{\n    Float64,\n    MOI.Utilities.MatrixOfConstraints{\n        Float64,\n        MOI.Utilities.MutableSparseMatrixCSC{Float64,Cint,MOI.Utilities.ZeroBasedIndexing},\n        MOI.Utilities.Hyperrectangle{Float64},\n        LP{Float64},\n    },\n}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The copy_to operation can now be implemented as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function _copy_to(dest::Optimizer, src::Model)\n    @assert MOI.is_empty(dest)\n    A = src.constraints.coefficients\n    row_bounds = src.constraints.constants\n    Clp_loadProblem(\n        dest,\n        A.n,\n        A.m,\n        A.colptr,\n        A.rowval,\n        A.nzval,\n        src.lower_bound,\n        src.upper_bound,\n        # (...) objective vector (omitted),\n        row_bounds.lower,\n        row_bounds.upper,\n    )\n    # Set objective sense and constant (omitted)\n    return\nend\n\nfunction MOI.copy_to(dest::Optimizer, src::Model)\n    _copy_to(dest, src)\n    return MOI.Utilities.identity_index_map(src)\nend\n\nfunction MOI.copy_to(\n    dest::Optimizer,\n    src::MOI.Utilities.UniversalFallback{Model},\n)\n    # Copy attributes from `src` to `dest` and error in case any unsupported\n    # constraints or attributes are set in `UniversalFallback`.\n    return MOI.copy_to(dest, src.model)\nend\n\nfunction MOI.copy_to(\n    dest::Optimizer,\n    src::MOI.ModelLike,\n)\n    model = Model()\n    index_map = MOI.copy_to(model, src)\n    _copy_to(dest, model)\n    return index_map\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/#ModelFilter","page":"Overview","title":"ModelFilter","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities provides Utilities.ModelFilter as a useful tool to copy a subset of a model. For example, given an infeasible model, we can copy the irreducible infeasible subsystem (for models implementing ConstraintConflictStatus) as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"my_filter(::Any) = true\nfunction my_filter(ci::MOI.ConstraintIndex)\n    status = MOI.get(dest, MOI.ConstraintConflictStatus(), ci)\n    return status != MOI.NOT_IN_CONFLICT\nend\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nindex_map = MOI.copy_to(dest, filtered_src)","category":"page"},{"location":"moi/submodules/Utilities/overview/#Fallbacks","page":"Overview","title":"Fallbacks","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The value of some attributes can be inferred from the value of other attributes.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For example, the value of ObjectiveValue can be computed using ObjectiveFunction and VariablePrimal.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"When a solver gives direct access to an attribute, it is better to return this value. However, if this is not the case, Utilities.get_fallback can be used instead. For example:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function MOI.get(model::Optimizer, attr::MOI.ObjectiveFunction)\n    return MOI.Utilities.get_fallback(model, attr)\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/#DoubleDicts","page":"Overview","title":"DoubleDicts","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"When writing MOI interfaces, we often need to handle situations in which we map ConstraintIndexs to different values. For example, to a string for ConstraintName.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"One option is to use a dictionary like Dict{MOI.ConstraintIndex,String}. However, this incurs a performance cost because the key is not a concrete type.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The DoubleDicts submodule helps this situation by providing two types main types Utilities.DoubleDicts.DoubleDict and Utilities.DoubleDicts.IndexDoubleDict. These types act like normal dictionaries, but internally they use more efficient dictionaries specialized to the type of the function-set pair.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The most common usage of a DoubleDict is in the index_map returned by copy_to. Performance can be improved, by using a function barrier. That is, instead of code like:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"index_map = MOI.copy_to(dest, src)\nfor (F, S) in MOI.get(src, MOI.ListOfConstraintTypesPresent())\n    for ci in MOI.get(src, MOI.ListOfConstraintIndices{F,S}())\n        dest_ci = index_map[ci]\n        # ...\n    end\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"use instead:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function function_barrier(\n    dest,\n    src,\n    index_map::MOI.Utilities.DoubleDicts.IndexDoubleDictInner{F,S},\n) where {F,S}\n    for ci in MOI.get(src, MOI.ListOfConstraintIndices{F,S}())\n        dest_ci = index_map[ci]\n        # ...\n    end\n    return\nend\n\nindex_map = MOI.copy_to(dest, src)\nfor (F, S) in MOI.get(src, MOI.ListOfConstraintTypesPresent())\n    function_barrier(dest, src, index_map[F, S])\nend","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Test/reference.md\"","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Test/reference/#test_reference","page":"API Reference","title":"The Test submodule","text":"","category":"section"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"Functions to help test implementations of MOI. See The Test submodule for more details.","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"Test.Config\nTest.runtests\nTest.setup_test\nTest.version_added\nTest.@requires\nTest.RequirementUnmet","category":"page"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.Config","page":"API Reference","title":"MathOptInterface.Test.Config","text":"Config(\n    ::Type{T} = Float64;\n    atol::Real = Base.rtoldefault(T),\n    rtol::Real = Base.rtoldefault(T),\n    optimal_status::MOI.TerminationStatusCode = MOI.OPTIMAL,\n    infeasible_status::MOI.TerminationStatusCode = MOI.INFEASIBLE,\n    exclude::Vector{Any} = Any[],\n) where {T}\n\nReturn an object that is used to configure various tests.\n\nConfiguration arguments\n\natol::Real = Base.rtoldefault(T): Control the absolute tolerance used when comparing solutions.\nrtol::Real = Base.rtoldefault(T): Control the relative tolerance used when comparing solutions.\noptimal_status = MOI.OPTIMAL: Set to MOI.LOCALLY_SOLVED if the solver cannot prove global optimality.\ninfeasible_status = MOI.INFEASIBLE: Set to MOI.LOCALLY_INFEASIBLE if the solver cannot prove global infeasibility.\nexclude = Vector{Any}: Pass attributes or functions to exclude to skip parts of tests that require certain functionality. Common arguments include:\nMOI.delete to skip deletion-related tests\nMOI.optimize! to skip optimize-related tests\nMOI.ConstraintDual to skip dual-related tests\nMOI.VariableName to skip setting variable names\nMOI.ConstraintName to skip setting constraint names\n\nExamples\n\nFor a nonlinear solver that finds local optima and does not support finding dual variables or constraint names:\n\nConfig(\n    Float64;\n    optimal_status = MOI.LOCALLY_SOLVED,\n    exclude = Any[\n        MOI.ConstraintDual,\n        MOI.VariableName,\n        MOI.ConstraintName,\n        MOI.delete,\n    ],\n)\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.runtests","page":"API Reference","title":"MathOptInterface.Test.runtests","text":"runtests(\n    model::MOI.ModelLike,\n    config::Config;\n    include::Vector{Union{String,Regex}} = String[],\n    exclude::Vector{Union{String,Regex}} = String[],\n    warn_unsupported::Bool = false,\n    exclude_tests_after::VersionNumber = v\"999.0.0\",\n)\n\nRun all tests in MathOptInterface.Test on model.\n\nConfiguration arguments\n\nconfig is a Test.Config object that can be used to modify the behavior of tests.\nIf include is not empty, only run tests if an element from include occursin the name of the test.\nIf exclude is not empty, skip tests if an element from exclude occursin the name of the test.\nexclude takes priority over include.\nIf warn_unsupported is false, runtests will silently skip tests that fail with a MOI.NotAllowedError, MOI.UnsupportedError, or RequirementUnmet error. (The latter is thrown when an @requires statement returns false.) When warn_unsupported is true, a warning will be printed. For most cases the default behavior, false, is what you want, since these tests likely test functionality that is not supported by model. However, it can be useful to run  warn_unsupported = true to check you are not skipping tests due to a missing supports_constraint method or equivalent.\nexclude_tests_after is a version number that excludes any tests to MOI added after that version number. This is useful for solvers who can declare a fixed set of tests, and not cause their tests to break if a new patch of MOI is released with a new test.\n\nSee also: setup_test.\n\nExample\n\nconfig = MathOptInterface.Test.Config()\nMathOptInterface.Test.runtests(\n    model,\n    config;\n    include = [\"test_linear_\", r\"^test_model_Name$\"],\n    exclude = [\"VariablePrimalStart\"],\n    warn_unsupported = true,\n    exclude_tests_after = v\"0.10.5\",\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.setup_test","page":"API Reference","title":"MathOptInterface.Test.setup_test","text":"setup_test(::typeof(f), model::MOI.ModelLike, config::Config)\n\nOverload this method to modify model before running the test function f on model with config. You can also modify the fields in config (e.g., to loosen the default tolerances).\n\nThis function should either return nothing, or return a function which, when called with zero arguments, undoes the setup to return the model to its previous state. You do not need to undo any modifications to config.\n\nThis function is most useful when writing new tests of the tests for MOI, but it can also be used to set test-specific tolerances, etc.\n\nSee also: runtests\n\nExample\n\nfunction MOI.Test.setup_test(\n    ::typeof(MOI.Test.test_linear_VariablePrimalStart_partial),\n    mock::MOIU.MockOptimizer,\n    ::MOI.Test.Config,\n)\n    MOIU.set_mock_optimize!(\n        mock,\n        (mock::MOIU.MockOptimizer) -> MOIU.mock_optimize!(mock, [1.0, 0.0]),\n    )\n    mock.eval_variable_constraint_dual = false\n\n    function reset_function()\n        mock.eval_variable_constraint_dual = true\n        return\n    end\n    return reset_function\nend\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.version_added","page":"API Reference","title":"MathOptInterface.Test.version_added","text":"version_added(::typeof(function_name))\n\nReturns the version of MOI in which the test function_name was added.\n\nThis method should be implemented for all new tests.\n\nSee the exclude_tests_after keyword of runtests for more details.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.@requires","page":"API Reference","title":"MathOptInterface.Test.@requires","text":"@requires(x)\n\nCheck that the condition x is true. Otherwise, throw an RequirementUnmet error to indicate that the model does not support something required by the test function.\n\nExamples\n\n@requires MOI.supports(model, MOI.Silent())\n@test MOI.get(model, MOI.Silent())\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.RequirementUnmet","page":"API Reference","title":"MathOptInterface.Test.RequirementUnmet","text":"RequirementUnmet(msg::String) <: Exception\n\nAn error for throwing in tests to indicate that the model does not support some requirement expected by the test function.\n\n\n\n\n\n","category":"type"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"EditURL = \"https://github.com/MadNLP/MadNLP.jl/blob/389561354a774441cd5e6b3aa5fffee102ed222e/README.md\"","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"<img src=\"https://github.com/MadNLP/MadNLP.jl/blob/master/logo-full.svg?raw=true\"/>","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Documentation Build Status Coverage DOI\n(Image: doc) (Image: build) (Image: codecov) (Image: DOI)","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is a nonlinear programming (NLP) solver, purely implemented in Julia. MadNLP implements a filter line-search algorithm, as that used in Ipopt. MadNLP seeks to streamline the development of modeling and algorithmic paradigms in order to exploit structures and to make efficient use of high-performance computers.","category":"page"},{"location":"packages/MadNLP/#License","page":"MadNLP/MadNLP.jl","title":"License","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is available under the MIT license.","category":"page"},{"location":"packages/MadNLP/#Installation","page":"MadNLP/MadNLP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"pkg> add MadNLP","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Optionally, various extension packages can be installed together:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"pkg> add MadNLPHSL, MadNLPPardiso, MadNLPMumps, MadNLPGPU, MadNLPGraph, MadNLPKrylov","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"These packages are stored in the lib subdirectory within the main MadNLP repository. Some extension packages may require additional dependencies or specific hardware. For the instructions for the build procedure, see the following links:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLPHSL\nMadNLPPardiso\nMadNLPGPU","category":"page"},{"location":"packages/MadNLP/#Usage","page":"MadNLP/MadNLP.jl","title":"Usage","text":"","category":"section"},{"location":"packages/MadNLP/#Interfaces","page":"MadNLP/MadNLP.jl","title":"Interfaces","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is interfaced with modeling packages:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"JuMP\nPlasmo\nNLPModels.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Users can pass various options to MadNLP also through the modeling packages. The interface-specific syntax are shown below. To see the list of MadNLP solver options, check the OPTIONS.md file.","category":"page"},{"location":"packages/MadNLP/#JuMP-interface","page":"MadNLP/MadNLP.jl","title":"JuMP interface","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, JuMP\nmodel = Model(()->MadNLP.Optimizer(print_level=MadNLP.INFO, max_iter=100))\n@variable(model, x, start = 0.0)\n@variable(model, y, start = 0.0)\n@NLobjective(model, Min, (1 - x)^2 + 100 * (y - x^2)^2)\noptimize!(model)","category":"page"},{"location":"packages/MadNLP/#NLPModels-interface","page":"MadNLP/MadNLP.jl","title":"NLPModels interface","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, CUTEst\nmodel = CUTEstModel(\"PRIMALC1\")\nmadnlp(model, print_level=MadNLP.WARN, max_wall_time=3600)","category":"page"},{"location":"packages/MadNLP/#Plasmo-interface-(requires-extension-MadNLPGraph)","page":"MadNLP/MadNLP.jl","title":"Plasmo interface (requires extension MadNLPGraph)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGraph, Plasmo\ngraph = OptiGraph()\n@optinode(graph,n1)\n@optinode(graph,n2)\n@variable(n1,0 <= x <= 2)\n@variable(n1,0 <= y <= 3)\n@constraint(n1,x+y <= 4)\n@objective(n1,Min,x)\n@variable(n2,x)\n@NLnodeconstraint(n2,exp(x) >= 2)\n@linkconstraint(graph,n1[:x] == n2[:x])\nMadNLP.optimize!(graph; print_level=MadNLP.DEBUG, max_iter=100)","category":"page"},{"location":"packages/MadNLP/#Linear-Solvers","page":"MadNLP/MadNLP.jl","title":"Linear Solvers","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is interfaced with non-Julia sparse/dense linear solvers:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Umfpack\nMKL-Pardiso\nMKL-Lapack\nHSL solvers (requires extension)\nPardiso (requires extension)\nMumps  (requires extension)\ncuSOLVER (requires extension)","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Each linear solver in MadNLP is a Julia type, and the linear_solver option should be specified by the actual type. Note that the linear solvers are always exported to Main.","category":"page"},{"location":"packages/MadNLP/#Built-in-Solvers:-Umfpack,-PardisoMKL,-LapackCPU","page":"MadNLP/MadNLP.jl","title":"Built-in Solvers: Umfpack, PardisoMKL, LapackCPU","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=UmfpackSolver)) # default\nmodel = Model(()->MadNLP.Optimizer(linear_solver=LapackCPUSolver))","category":"page"},{"location":"packages/MadNLP/#HSL-(requires-extension-MadNLPHSL)","page":"MadNLP/MadNLP.jl","title":"HSL (requires extension MadNLPHSL)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPHSL, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma27Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma57Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma77Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma86Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma97Solver))","category":"page"},{"location":"packages/MadNLP/#Mumps-(requires-extension-MadNLPMumps)","page":"MadNLP/MadNLP.jl","title":"Mumps (requires extension MadNLPMumps)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPMumps, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MumpsSolver))","category":"page"},{"location":"packages/MadNLP/#Pardiso-(requires-extension-MadNLPPardiso)","page":"MadNLP/MadNLP.jl","title":"Pardiso (requires extension MadNLPPardiso)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPPardiso, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=PardisoSolver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=PardisoMKLSolver))","category":"page"},{"location":"packages/MadNLP/#LapackGPU-(requires-extension-MadNLPGPU)","page":"MadNLP/MadNLP.jl","title":"LapackGPU (requires extension MadNLPGPU)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGPU, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=LapackGPUSolver))","category":"page"},{"location":"packages/MadNLP/#Schur-and-Schwarz-(requires-extension-MadNLPGraph)","page":"MadNLP/MadNLP.jl","title":"Schur and Schwarz (requires extension MadNLPGraph)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGraph, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MadNLPSchwarz))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MadNLPSchur))","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"The solvers in MadNLPGraph (Schur and Schwarz) use multi-thread parallelism; thus, Julia session should be started with -t flag.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"julia -t 16 # to use 16 threads","category":"page"},{"location":"packages/MadNLP/#Citing-MadNLP.jl","page":"MadNLP/MadNLP.jl","title":"Citing MadNLP.jl","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"If you use MadNLP.jl in your research, we would greatly appreciate your citing it.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"@article{shin2020graph,\n  title={Graph-Based Modeling and Decomposition of Energy Infrastructures},\n  author={Shin, Sungho and Coffrin, Carleton and Sundar, Kaarthik and Zavala, Victor M},\n  journal={arXiv preprint arXiv:2010.02404},\n  year={2020}\n}","category":"page"},{"location":"packages/MadNLP/#Bug-reports-and-support","page":"MadNLP/MadNLP.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Please report issues and feature requests via the GitHub issue tracker.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/variables/#jump_variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The term variable in mathematical optimization has many meanings. For example, optimization variables (also called decision variables) are the unknowns x that we are solving for in the problem:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To complicate things, Julia uses variable to mean a binding between a name and a value. For example, in the statement:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x = 1\n1","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"x is a variable that stores the value 1.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP uses variable in a third way, to mean an instance of the VariableRef struct. JuMP variables are the link between Julia and the optimization variables inside a JuMP model.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This page explains how to create and manage JuMP variables in a variety of contexts.","category":"page"},{"location":"manual/variables/#Create-a-variable","page":"Variables","title":"Create a variable","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create variables using the @variable macro:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> typeof(x)\nVariableRef (alias for GenericVariableRef{Float64})\n\njulia> num_variables(model)\n1","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here x is a Julia variable that is bound to a VariableRef object, and we have added 1 decision variable to our model.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To make the binding more explicit, we could have written:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model, x)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"but there is no need to in general; the macro does it for us.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When creating a variable, you can also specify variable bounds:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x_free)\nx_free\n\njulia> @variable(model, x_lower >= 0)\nx_lower\n\njulia> @variable(model, x_upper <= 1)\nx_upper\n\njulia> @variable(model, 2 <= x_interval <= 3)\nx_interval\n\njulia> @variable(model, x_fixed == 4)\nx_fixed\n\njulia> print(model)\nFeasibility\nSubject to\n x_fixed = 4\n x_lower ≥ 0\n x_interval ≥ 2\n x_upper ≤ 1\n x_interval ≤ 3","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nWhen creating a variable with a single lower- or upper-bound, and the value of the bound is not a numeric literal (for example, 1 or 1.0), the name of the variable must appear on the left-hand side. Putting the name on the right-hand side is an error. For example, to create a variable x:a = 1\n@variable(model, x >= 1)      # ✓ Okay\n@variable(model, 1.0 <= x)    # ✓ Okay\n@variable(model, x >= a)      # ✓ Okay\n@variable(model, a <= x)      # × Not okay\n@variable(model, x >= 1 / 2)  # ✓ Okay\n@variable(model, 1 / 2 <= x)  # × Not okay","category":"page"},{"location":"manual/variables/#Containers-of-variables","page":"Variables","title":"Containers of variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The @variable macro also supports creating collections of JuMP variables. We'll cover some brief syntax here; read the Variable containers section for more details.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"You can create arrays of JuMP variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]\n\njulia> x[1, 2]\nx[1,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Index sets can be named, and bounds can depend on those names:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, sqrt(i) <= x[i = 1:3] <= i^2)\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> x[2]\nx[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Sets can be any Julia type that supports iteration:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i = 2:3, j = 1:2:3, [\"red\", \"blue\"]] >= 0)\n3-dimensional DenseAxisArray{VariableRef,3,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\n    Dimension 3, [\"red\", \"blue\"]\nAnd data, a 2×2×2 Array{VariableRef, 3}:\n[:, :, \"red\"] =\n x[2,1,red]  x[2,3,red]\n x[3,1,red]  x[3,3,red]\n\n[:, :, \"blue\"] =\n x[2,1,blue]  x[2,3,blue]\n x[3,1,blue]  x[3,3,blue]\n\njulia> x[2, 1, \"red\"]\nx[2,1,red]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Sets can depend upon previous indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, u[i = 1:2, j = i:3])\nJuMP.Containers.SparseAxisArray{VariableRef, 2, Tuple{Int64, Int64}} with 5 entries:\n  [1, 1]  =  u[1,1]\n  [1, 2]  =  u[1,2]\n  [1, 3]  =  u[1,3]\n  [2, 2]  =  u[2,2]\n  [2, 3]  =  u[2,3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"and we can filter elements in the sets using the ; syntax:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, v[i = 1:9; mod(i, 3) == 0])\nJuMP.Containers.SparseAxisArray{VariableRef, 1, Tuple{Int64}} with 3 entries:\n  [3]  =  v[3]\n  [6]  =  v[6]\n  [9]  =  v[9]","category":"page"},{"location":"manual/variables/#Registered-variables","page":"Variables","title":"Registered variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When you create variables, JuMP registers them inside the model using their corresponding symbol. Get a registered name using model[:key]:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> @variable(model, x)\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x\n\njulia> model[:x] === x\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Registered names are most useful when you start to write larger models and want to break up the model construction into functions:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> function set_objective(model::Model)\n           @objective(model, Min, 2 * model[:my_x] + 1)\n           return\n       end\nset_objective (generic function with 1 method)\n\njulia> model = Model();\n\njulia> @variable(model, my_x);\n\njulia> set_objective(model)\n\njulia> print(model)\nMin 2 my_x + 1\nSubject to","category":"page"},{"location":"manual/variables/#anonymous_variables","page":"Variables","title":"Anonymous variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To reduce the likelihood of accidental bugs, and because JuMP registers variables inside a model, creating two variables with the same name is an error:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n[...]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"A common reason for encountering this error is adding variables in a loop.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"As a work-around, JuMP provides anonymous variables. Create a scalar valued anonymous variable by omitting the name argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model)\n_[1]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Anonymous variables get printed as an underscore followed by a unique index of the variable.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nThe index of the variable may not correspond to the column of the variable in the solver.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a container of anonymous JuMP variables by dropping the name in front of the [:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> y = @variable(model, [1:2])\n2-element Vector{VariableRef}:\n _[1]\n _[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The <= and >= short-hand cannot be used to set bounds on scalar-valued anonymous JuMP variables. Instead, use the lower_bound and upper_bound keywords:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x_lower = @variable(model, lower_bound = 1.0)\n_[1]\n\njulia> x_upper = @variable(model, upper_bound = 2.0)\n_[2]\n\njulia> x_interval = @variable(model, lower_bound = 3.0, upper_bound = 4.0)\n_[3]","category":"page"},{"location":"manual/variables/#Variable-names","page":"Variables","title":"Variable names","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"In addition to the symbol that variables are registered with, JuMP variables have a String name that is used for printing and writing to file formats.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Get and set the name of a variable using name and set_name:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> name(x)\n\"x\"\n\njulia> set_name(x, \"my_x_name\")\n\njulia> x\nmy_x_name","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Override the default choice of name using the base_name keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Note that names apply to each element of the container, not to the container of variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> name(x[1])\n\"my_var[1]\"\n\njulia> set_name(x[1], \"my_x\")\n\njulia> x\n2-element Vector{VariableRef}:\n my_x\n my_var[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nFor some models, setting the string name of each variable can take a non-trivial portion of the total time required to build the model. Turn off String names by passing set_string_name = false to @variable:julia> model = Model();\n\njulia> @variable(model, x, set_string_name = false)\n_[1]See Disable string names for more information.","category":"page"},{"location":"manual/variables/#Retrieve-a-variable-by-name","page":"Variables","title":"Retrieve a variable by name","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Retrieve a variable from a model using variable_by_name:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> variable_by_name(model, \"my_x\")\nmy_x","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If the name is not present, nothing will be returned:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> variable_by_name(model, \"bad_name\")","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"You can only look up individual variables using variable_by_name. Something like this will not work:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, [i = 1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]\n\njulia> variable_by_name(model, \"my_var\")","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To look up a collection of variables, do not use variable_by_name. Instead, register them using the model[:key] = value syntax:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> model[:x] = @variable(model, [i = 1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]\n\njulia> model[:x]\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]","category":"page"},{"location":"manual/variables/#variable_names_and_bindings","page":"Variables","title":"String names, symbolic names, and bindings","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"It's common for new users to experience confusion relating to JuMP variables. Part of the problem is the overloaded use of \"variable\" in mathematical optimization, along with the difference between the name that a variable is registered under and the String name used for printing.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here's a summary of the differences:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP variables are created using @variable.\nJuMP variables can be named or anonymous.\nNamed JuMP variables have the form @variable(model, x). For named variables:\nThe String name of the variable is set to \"x\".\nA Julia variable x is created that binds x to  the JuMP variable.\nThe name :x is registered as a key in the model with the value x.\nAnonymous JuMP variables have the form x = @variable(model). For anonymous variables:\nThe String name of the variable is set to \"\". When printed, this is replaced with \"_[i]\" where i is the index of the variable.\nYou control the name of the Julia variable used as the binding.\nNo name is registered as a key in the model.\nThe base_name keyword can override the String name of the variable.\nYou can manually register names in the model via model[:key] = value","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here's an example that should make things clearer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x_binding = @variable(model, base_name = \"x\")\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> x\nERROR: UndefVarError: `x` not defined\n\njulia> x_binding\nx\n\njulia> name(x_binding)\n\"x\"\n\njulia> model[:x_register] = x_binding\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x_register\n\njulia> model[:x_register]\nx\n\njulia> model[:x_register] === x_binding\ntrue\n\njulia> x\nERROR: UndefVarError: `x` not defined","category":"page"},{"location":"manual/variables/#Create,-delete,-and-modify-variable-bounds","page":"Variables","title":"Create, delete, and modify variable bounds","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Query whether a variable has a bound using has_lower_bound, has_upper_bound, and is_fixed:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> has_lower_bound(x_free)\nfalse\n\njulia> has_upper_bound(x_upper)\ntrue\n\njulia> is_fixed(x_fixed)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If a variable has a particular bound, query the value of it using lower_bound, upper_bound, and fix_value:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> lower_bound(x_interval)\n2.0\n\njulia> upper_bound(x_interval)\n3.0\n\njulia> fix_value(x_fixed)\n4.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Querying the value of a bound that does not exist will result in an error.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete variable bounds using delete_lower_bound, delete_upper_bound, and unfix:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> delete_lower_bound(x_lower)\n\njulia> has_lower_bound(x_lower)\nfalse\n\njulia> delete_upper_bound(x_upper)\n\njulia> has_upper_bound(x_upper)\nfalse\n\njulia> unfix(x_fixed)\n\njulia> is_fixed(x_fixed)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Set or update variable bounds using set_lower_bound, set_upper_bound, and fix:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> set_lower_bound(x_lower, 1.1)\n\njulia> set_upper_bound(x_upper, 2.1)\n\njulia> fix(x_fixed, 4.1)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Fixing a variable with existing bounds will throw an error. To delete the bounds prior to fixing, use fix(variable, value; force = true).","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x >= 1)\nx\n\njulia> fix(x, 2)\nERROR: Unable to fix x to 2 because it has existing variable bounds. Consider calling `JuMP.fix(variable, value; force=true)` which will delete existing bounds before fixing the variable.\n\njulia> fix(x, 2; force = true)\n\n\njulia> fix_value(x)\n2.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nUse fix instead of @constraint(model, x == 2). The former modifies variable bounds, while the latter adds a new linear constraint to the problem.","category":"page"},{"location":"manual/variables/#Binary-variables","page":"Variables","title":"Binary variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Binary variables are constrained to the set x in 0 1.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a binary variable by passing Bin as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, Bin)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Check if a variable is binary using is_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> is_binary(x)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete a binary constraint using unset_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unset_binary(x)\n\njulia> is_binary(x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Binary variables can also be created by setting the binary keyword to true:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, binary=true)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"or by using set_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_binary(x)","category":"page"},{"location":"manual/variables/#Integer-variables","page":"Variables","title":"Integer variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Integer variables are constrained to the set x in mathbbZ.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create an integer variable by passing Int as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, Int)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Check if a variable is integer using is_integer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> is_integer(x)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete an integer constraint using unset_integer.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unset_integer(x)\n\njulia> is_integer(x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Integer variables can also be created by setting the integer keyword to true:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, integer=true)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"or by using set_integer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_integer(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nThe relax_integrality function relaxes all integrality constraints in the model, returning a function that can be called to undo the operation later on.","category":"page"},{"location":"manual/variables/#Semi-integer-and-semi-continuous-variables","page":"Variables","title":"Semi-integer and semi-continuous variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Semi-continuous variables are constrained to the set x in 0 cup l u.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a semi-continuous variable using the Semicontinuous set:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x in Semicontinuous(1.5, 3.5))\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Semi-integer variables  are constrained to the set x in 0 cup l l+1 dots u.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a semi-integer variable using the Semiinteger set:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x in Semiinteger(1.0, 3.0))\nx","category":"page"},{"location":"manual/variables/#Start-values","page":"Variables","title":"Start values","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"There are two ways to provide a primal starting solution (also called MIP-start or a warmstart) for each variable:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using the start keyword in the @variable macro\nusing set_start_value","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The starting value of a variable can be queried using start_value. If no start value has been set, start_value will return nothing.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> start_value(x)\n\njulia> @variable(model, y, start = 1)\ny\n\njulia> start_value(y)\n1.0\n\njulia> set_start_value(y, 2)\n\njulia> start_value(y)\n2.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The start keyword argument can depend on the indices of a variable container:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, z[i = 1:2], start = i^2)\n2-element Vector{VariableRef}:\n z[1]\n z[2]\n\njulia> start_value.(z)\n2-element Vector{Float64}:\n 1.0\n 4.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nSome solvers do not support start values. If a solver does not support start values, an MathOptInterface.UnsupportedAttribute{MathOptInterface.VariablePrimalStart} error will be thrown.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nTo set the optimal solution from a previous solve as a new starting value, use all_variables to get a vector of all the variables in the model, then run:x = all_variables(model)\nx_solution = value.(x)\nset_start_value.(x, x_solution)Alternatively, use set_start_values.","category":"page"},{"location":"manual/variables/#delete_a_variable","page":"Variables","title":"Delete a variable","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use delete to delete a variable from a model. Use is_valid to check if a variable belongs to a model and has not been deleted.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> is_valid(model, x)\ntrue\n\njulia> delete(model, x)\n\njulia> is_valid(model, x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Deleting a variable does not unregister the corresponding name from the model. Therefore, creating a new variable of the same name will throw an error:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n[...]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"After calling delete, call unregister to remove the symbolic reference:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unregister(model, :x)\n\njulia> @variable(model, x)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"info: Info\ndelete does not automatically unregister because we do not distinguish between names that are automatically registered by JuMP macros and names that are manually registered by the user by setting values in object_dictionary. In addition, deleting a variable and then adding a new variable of the same name is an easy way to introduce bugs into your code.","category":"page"},{"location":"manual/variables/#Variable-containers","page":"Variables","title":"Variable containers","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP provides a mechanism for creating collections of variables in three types of data structures, which we refer to as containers.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The three types are Arrays, DenseAxisArrays, and SparseAxisArrays. We explain each of these in the following.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/variables/#Arrays","page":"Variables","title":"Arrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We have already seen the creation of an array of JuMP variables with the x[1:2] syntax. This can be extended to create multi-dimensional arrays of JuMP variables. For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Arrays of JuMP variables can be indexed and sliced as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x[1, 2]\nx[1,2]\n\njulia> x[2, :]\n2-element Vector{VariableRef}:\n x[2,1]\n x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Variable bounds can depend upon the indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=1:2] >= 2i + j)\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]\n\njulia> lower_bound.(x)\n2×2 Matrix{Float64}:\n 3.0  4.0\n 5.0  6.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP will form an Array of JuMP variables when it can determine at compile time that the indices are one-based integer ranges. Therefore x[1:b] will create an Array of JuMP variables, but x[a:b] will not. If JuMP cannot determine that the indices are one-based integer ranges (for example, in the case of x[a:b]), JuMP will create a DenseAxisArray instead.","category":"page"},{"location":"manual/variables/#variable_jump_arrays","page":"Variables","title":"DenseAxisArrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We often want to create arrays where the indices are not one-based integer ranges. For example, we may want to create a variable indexed by the name of a product or a location. The syntax is the same as that above, except with an arbitrary vector as an index as opposed to a one-based range. The biggest difference is that instead of returning an Array of JuMP variables, JuMP will return a DenseAxisArray. For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, [:A,:B]])\n2-dimensional DenseAxisArray{VariableRef,2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{VariableRef}:\n x[1,A]  x[1,B]\n x[2,A]  x[2,B]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"DenseAxisArrays can be indexed and sliced as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x[1, :A]\nx[1,A]\n\njulia> x[2, :]\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, [:A, :B]\nAnd data, a 2-element Vector{VariableRef}:\n x[2,A]\n x[2,B]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Bounds can depend upon indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=2:3, j=1:2:3] >= 0.5i + j)\n2-dimensional DenseAxisArray{VariableRef,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\nAnd data, a 2×2 Matrix{VariableRef}:\n x[2,1]  x[2,3]\n x[3,1]  x[3,3]\n\njulia> lower_bound.(x)\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\nAnd data, a 2×2 Matrix{Float64}:\n 2.0  4.0\n 2.5  4.5","category":"page"},{"location":"manual/variables/#variable_sparseaxisarrays","page":"Variables","title":"SparseAxisArrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The third container type that JuMP natively supports is SparseAxisArray. These arrays are created when the indices do not form a rectangular set. For example, this applies when indices have a dependence upon previous indices (called triangular indexing). JuMP supports this as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=i:2])\nJuMP.Containers.SparseAxisArray{VariableRef, 2, Tuple{Int64, Int64}} with 3 entries:\n  [1, 1]  =  x[1,1]\n  [1, 2]  =  x[1,2]\n  [2, 2]  =  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We can also conditionally create variables via a JuMP-specific syntax. This syntax appends a comparison check that depends upon the named indices and is separated from the indices by a semi-colon (;). For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:4; mod(i, 2)==0])\nJuMP.Containers.SparseAxisArray{VariableRef, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  x[2]\n  [4]  =  x[4]","category":"page"},{"location":"manual/variables/#Performance-considerations","page":"Variables","title":"Performance considerations","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When using the semi-colon as a filter, JuMP iterates over all indices and evaluates the conditional for each combination. If there are many index dimensions and a large amount of sparsity, this can be inefficient.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> N = 10\n10\n\njulia> S = [(1, 1, 1), (N, N, N)]\n2-element Vector{Tuple{Int64, Int64, Int64}}:\n (1, 1, 1)\n (10, 10, 10)\n\njulia> @time @variable(model, x1[i=1:N, j=1:N, k=1:N; (i, j, k) in S])\n  0.203861 seconds (392.22 k allocations: 23.977 MiB, 99.10% compilation time)\nJuMP.Containers.SparseAxisArray{VariableRef, 3, Tuple{Int64, Int64, Int64}} with 2 entries:\n  [1, 1, 1   ]  =  x1[1,1,1]\n  [10, 10, 10]  =  x1[10,10,10]\n\njulia> @time @variable(model, x2[S])\n  0.045407 seconds (65.24 k allocations: 3.771 MiB, 99.15% compilation time)\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, [(1, 1, 1), (10, 10, 10)]\nAnd data, a 2-element Vector{VariableRef}:\n x2[(1, 1, 1)]\n x2[(10, 10, 10)]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The first option is slower because it is equivalent to:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x1 = Dict{NTuple{3,Int},VariableRef}()\nDict{Tuple{Int64, Int64, Int64}, VariableRef}()\n\njulia> for i in 1:N\n           for j in 1:N\n               for k in 1:N\n                   if (i, j, k) in S\n                       x1[i, j, k] = @variable(model, base_name = \"x1[$i,$j,$k]\")\n                   end\n               end\n           end\n       end\n\njulia> x1\nDict{Tuple{Int64, Int64, Int64}, VariableRef} with 2 entries:\n  (1, 1, 1)    => x1[1,1,1]\n  (10, 10, 10) => x1[10,10,10]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If performance is a concern, explicitly construct the set of indices instead of using the filtering syntax.","category":"page"},{"location":"manual/variables/#variable_forcing","page":"Variables","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When creating a container of JuMP variables, JuMP will attempt to choose the tightest container type that can store the JuMP variables. Thus, it will prefer to create an Array before a DenseAxisArray and a DenseAxisArray before a SparseAxisArray. However, because this happens at compile time, JuMP does not always make the best choice. To illustrate this, consider the following example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> A = 1:2\n1:2\n\njulia> @variable(model, x[A])\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, 1:2\nAnd data, a 2-element Vector{VariableRef}:\n x[1]\n x[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Since the value (and type) of A is unknown at parsing time, JuMP is unable to infer that A is a one-based integer range. Therefore, JuMP creates a DenseAxisArray, even though it could store these two variables in a standard one-dimensional Array.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We can share our knowledge that it is possible to store these JuMP variables as an array by setting the container keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, y[A], container=Array)\n2-element Vector{VariableRef}:\n y[1]\n y[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP now creates a vector of JuMP variables instead of a DenseAxisArray. Choosing an invalid container type will throw an error.","category":"page"},{"location":"manual/variables/#User-defined-containers","page":"Variables","title":"User-defined containers","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"In addition to the built-in container types, you can create your own collections of JuMP variables.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nThis is a point that users often overlook: you are not restricted to the built-in container types in JuMP.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, the following code creates a dictionary with symmetric matrices as the values:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> variables = Dict{Symbol,Array{VariableRef,2}}(\n           key => @variable(model, [1:2, 1:2], Symmetric, base_name = \"$(key)\")\n           for key in [:A, :B]\n       )\nDict{Symbol, Matrix{VariableRef}} with 2 entries:\n  :A => [A[1,1] A[1,2]; A[1,2] A[2,2]]\n  :B => [B[1,1] B[1,2]; B[1,2] B[2,2]]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Another common scenario is a request to add variables to existing containers, for example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using JuMP\nmodel = Model()\n@variable(model, x[1:2] >= 0)\n# Later I want to add\n@variable(model, x[3:4] >= 0)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This is not possible with the built-in JuMP container types. However, you can use regular Julia types instead:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = model[:x] = @variable(model, [1:2], lower_bound = 0, base_name = \"x\")\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> append!(x, @variable(model, [1:2], lower_bound = 0, base_name = \"y\"));\n\njulia> model[:x]\n4-element Vector{VariableRef}:\n x[1]\n x[2]\n y[1]\n y[2]","category":"page"},{"location":"manual/variables/#Semidefinite-variables","page":"Variables","title":"Semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a square matrix of JuMP variables to be positive semidefinite by passing PSD as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], PSD)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This will ensure that x is symmetric, and that all of its eigenvalues are nonnegative.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nx must be a square 2-dimensional Array of JuMP variables; it cannot be a DenseAxisArray or a SparseAxisArray.","category":"page"},{"location":"manual/variables/#Symmetric-variables","page":"Variables","title":"Symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a square matrix of JuMP variables to be symmetric (but not necessarily positive semidefinite) by passing Symmetric  as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#variables","page":"Variables","title":"The @variables macro","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If you have many @variable calls, JuMP provides the macro @variables that can improve readability:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variables(model, begin\n           x\n           y[i=1:2] >= i, (start = i, base_name = \"Y_$i\")\n           z, Bin\n       end)\n(x, VariableRef[Y_1[1], Y_2[2]], z)\n\njulia> print(model)\nFeasibility\nSubject to\n Y_1[1] ≥ 1\n Y_2[2] ≥ 2\n z binary","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The @variables macro returns a tuple of the variables that were defined.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nKeyword arguments must be contained within parentheses.","category":"page"},{"location":"manual/variables/#jump_variables_on_creation","page":"Variables","title":"Variables constrained on creation","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"All uses of the @variable macro documented so far translate into separate calls for variable creation and the adding of any bound or integrality constraints.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, @variable(model, x >= 0, Int), is equivalent to:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_lower_bound(x, 0.0)\n\njulia> set_integer(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Importantly, the bound and integrality constraints are added after the variable has been created.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"However, some solvers require a set specifying the variable domain to be given when the variable is first created. We say that these variables are constrained on creation.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use in within @variable to access the special syntax for constraining variables on creation.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, the following creates a vector of variables that belong to the SecondOrderCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, y[1:3] in SecondOrderCone())\n3-element Vector{VariableRef}:\n y[1]\n y[2]\n y[3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For contrast, the standard syntax is as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, x in SecondOrderCone())\n[x[1], x[2], x[3]] ∈ MathOptInterface.SecondOrderCone(3)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"An alternate syntax to x in Set is to use the set keyword of @variable. This is most useful when creating anonymous variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model, [1:3], set = SecondOrderCone())\n3-element Vector{VariableRef}:\n _[1]\n _[2]\n _[3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nYou cannot delete the constraint associated with a variable constrained on creation.","category":"page"},{"location":"manual/variables/#Example:-positive-semidefinite-variables","page":"Variables","title":"Example: positive semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"An alternative to the syntax in Semidefinite variables, declare a matrix of JuMP variables to be positive semidefinite using PSDCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in PSDCone())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#Example:-symmetric-variables","page":"Variables","title":"Example: symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"As an alternative to the syntax in Symmetric variables, declare a matrix of JuMP variables to be symmetric using SymmetricMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in SymmetricMatrixSpace())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#Example:-skew-symmetric-variables","page":"Variables","title":"Example: skew-symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be skew-symmetric using SkewSymmetricMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        x[1,2]\n -x[1,2]  0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nEven though x is a 2 by 2 matrix, only one decision variable is added to model; the remaining elements in x are linear transformations of the single variable.","category":"page"},{"location":"manual/variables/#Example:-Hermitian-positive-semidefinite-variables","page":"Variables","title":"Example: Hermitian positive semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be Hermitian positive semidefinite using HermitianPSDCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, H[1:2, 1:2] in HermitianPSDCone())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    real(H[1,2]) + imag(H[1,2]) im\n real(H[1,2]) - imag(H[1,2]) im  real(H[2,2])","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This adds 4 real variables in the MOI.HermitianPositiveSemidefiniteConeTriangle:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> first(all_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle))\n[real(H[1,1]), real(H[1,2]), real(H[2,2]), imag(H[1,2])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(2)","category":"page"},{"location":"manual/variables/#Example:-Hermitian-variables","page":"Variables","title":"Example: Hermitian variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be Hermitian using the Hermitian tag:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], Hermitian)\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(x[1,1])                    real(x[1,2]) + imag(x[1,2]) im\n real(x[1,2]) - imag(x[1,2]) im  real(x[2,2])","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This is equivalent to declaring the variable in HermitianMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(x[1,1])                    real(x[1,2]) + imag(x[1,2]) im\n real(x[1,2]) - imag(x[1,2]) im  real(x[2,2])","category":"page"},{"location":"manual/variables/#Why-use-variables-constrained-on-creation?","page":"Variables","title":"Why use variables constrained on creation?","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For most users, it does not matter if you use the constrained on creation syntax. Therefore, use whatever syntax you find most convenient.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"However, if you use direct_model, you may be forced to use the constrained on creation syntax.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The technical difference between variables constrained on creation and the standard JuMP syntax is that variables constrained on creation calls MOI.add_constrained_variables, while the standard JuMP syntax calls MOI.add_variables and then MOI.add_constraint.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Consult the implementation of solver package you are using to see if your solver requires MOI.add_constrained_variables.","category":"page"},{"location":"manual/variables/#Parameters","page":"Variables","title":"Parameters","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Some solvers have explicit support for parameters, which are constants in the model that can be efficiently updated between solves.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP implements parameters by a decision variable constrained on creation to the Parameter set.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, p[i = 1:2] in Parameter(i))\n2-element Vector{VariableRef}:\n p[1]\n p[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create anonymous parameters using the set keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> anon_parameter = @variable(model, set = Parameter(1.0))\n_[4]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use parameter_value and set_parameter_value to query or update the value of a parameter.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> parameter_value.(p)\n2-element Vector{Float64}:\n 1.0\n 2.0\n\njulia> set_parameter_value(p[2], 3.0)\n\njulia> parameter_value.(p)\n2-element Vector{Float64}:\n 1.0\n 3.0","category":"page"},{"location":"manual/variables/#Limitations","page":"Variables","title":"Limitations","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Parameters are implemented as decision variables belonging to the Parameter set. If the solver supports the MOI.Parameter set, it may decide to replace all instances of the parameter variable by the associated constant. If the solver does not support parameters, it will add the parameter as a decision variable with fixed bounds.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The most important implication of this design is that JuMP treats a parameter multiplied by a decision variable as a quadratic expression, even though it is equivalent to a linear expression.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, p in Parameter(2));\n\njulia> px = @expression(model, p * x)\np*x\n\njulia> typeof(px)\nQuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})","category":"page"},{"location":"manual/variables/#When-to-use-a-parameter","page":"Variables","title":"When to use a parameter","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Parameters are most useful when solving nonlinear models in a sequence:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer);\nset_silent(model)\n@variable(model, x)\n@variable(model, p in Parameter(1.0))\n@objective(model, Min, (x - p)^2)\noptimize!(model)\nvalue(x)\nset_parameter_value(p, 5.0)\noptimize!(model)\nvalue(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Using parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/FileFormats/overview.md\"","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/FileFormats/overview/#The-FileFormats-submodule","page":"Overview","title":"The FileFormats submodule","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The FileFormats module provides functions for reading and writing MOI models using write_to_file and read_from_file.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Supported-file-types","page":"Overview","title":"Supported file types","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"You must read and write files to a FileFormats.Model object. Specific the file-type by passing a FileFormats.FileFormat enum. For example:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The Conic Benchmark Format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_CBF)\nA Conic Benchmark Format (CBF) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The LP file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_LP)\nA .LP-file model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The MathOptFormat file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The MPS file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The NL file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_NL)\nAn AMPL (.nl) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The REW file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_REW)\nA Mathematical Programming System (MPS) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Note that the REW format is identical to the MPS file format, except that all names are replaced with generic identifiers.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The SDPA file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_SDPA)\nA SemiDefinite Programming Algorithm Format (SDPA) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Write-to-file","page":"Overview","title":"Write to file","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"To write a model src to a MathOptFormat file, use:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> MOI.add_variable(src)\nMOI.VariableIndex(1)\n\njulia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap with 1 entry:\n  MOI.VariableIndex(1) => MOI.VariableIndex(1)\n\njulia> MOI.write_to_file(dest, \"file.mof.json\")\n\njulia> print(read(\"file.mof.json\", String))\n{\n  \"name\": \"MathOptFormat Model\",\n  \"version\": {\n    \"major\": 1,\n    \"minor\": 5\n  },\n  \"variables\": [\n    {\n      \"name\": \"x1\"\n    }\n  ],\n  \"objective\": {\n    \"sense\": \"feasibility\"\n  },\n  \"constraints\": []\n}","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Read-from-file","page":"Overview","title":"Read from file","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"To read a MathOptFormat file, use:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model\n\njulia> MOI.read_from_file(dest, \"file.mof.json\")\n\njulia> MOI.get(dest, MOI.ListOfVariableIndices())\n1-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n\njulia> rm(\"file.mof.json\")  # Clean up after ourselves.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Detecting-the-file-type-automatically","page":"Overview","title":"Detecting the file-type automatically","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Instead of the format keyword, you can also use the filename keyword argument to FileFormats.Model. This will attempt to automatically guess the format from the file extension. For example:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> MOI.write_to_file(dest, \"file.cbf.gz\")\n\njulia> src_2 = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> MOI.write_to_file(dest, \"file.cbf.gz\")\n\njulia> src_2 = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.read_from_file(src_2, \"file.cbf.gz\")\n\njulia> rm(\"file.cbf.gz\")  # Clean up after ourselves.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Note how the compression format (GZip) is also automatically detected from the filename.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Unsupported-constraints","page":"Overview","title":"Unsupported constraints","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"In some cases src may contain constraints that are not supported by the file format (for example, the CBF format supports integer variables but not binary). If so, copy src to a bridged model using Bridges.full_bridge_optimizer:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"src = MOI.Utilities.Model{Float64}()\nx = MOI.add_variable(model)\nMOI.add_constraint(model, x, MOI.ZeroOne())\ndest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_CBF)\nbridged = MOI.Bridges.full_bridge_optimizer(dest, Float64)\nMOI.copy_to(bridged, src)\nMOI.write_to_file(dest, \"my_model.cbf\")","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"note: Note\nEven after bridging, it may still not be possible to write the model to file because of unsupported constraints (for example, PSD variables in the LP file format).","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Read-and-write-to-io","page":"Overview","title":"Read and write to io","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"In addition to write_to_file and read_from_file, you can read and write directly from IO streams using Base.write and Base.read!:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> io = IOBuffer();\n\njulia> write(io, dest)\n\njulia> seekstart(io);\n\njulia> src_2 = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model\n\njulia> read!(io, src_2);","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Validating-MOF-files","page":"Overview","title":"Validating MOF files","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"MathOptFormat files are governed by a schema. Use JSONSchema.jl to check if a .mof.json file satisfies the schema.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"First, construct the schema object as follows:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> import JSON, JSONSchema\n\njulia> schema = JSONSchema.Schema(JSON.parsefile(MOI.FileFormats.MOF.SCHEMA_PATH))\nA JSONSchema","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Then, check if a model file is valid using isvalid:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> good_model = JSON.parse(\"\"\"\n       {\n         \"version\": {\n           \"major\": 1,\n           \"minor\": 5\n         },\n         \"variables\": [{\"name\": \"x\"}],\n         \"objective\": {\"sense\": \"feasibility\"},\n         \"constraints\": []\n       }\n       \"\"\");\n\njulia> isvalid(schema, good_model)\ntrue","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"If we construct an invalid file, for example by mis-typing name as NaMe, the validation fails:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> bad_model = JSON.parse(\"\"\"\n       {\n         \"version\": {\n           \"major\": 1,\n           \"minor\": 5\n         },\n         \"variables\": [{\"NaMe\": \"x\"}],\n         \"objective\": {\"sense\": \"feasibility\"},\n         \"constraints\": []\n       }\n       \"\"\");\n\njulia> isvalid(schema, bad_model)\nfalse","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Use JSONSchema.validate to obtain more insight into why the validation failed:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> JSONSchema.validate(schema, bad_model)\nValidation failed:\npath:         [variables][1]\ninstance:     Dict{String, Any}(\"NaMe\" => \"x\")\nschema key:   required\nschema value: Any[\"name\"]","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"EditURL = \"https://github.com/lanl-ansi/Alpine.jl/blob/v0.5.5/README.md\"","category":"page"},{"location":"packages/Alpine/#Alpine,-a-global-solver-for-non-convex-MINLPs","page":"lanl-ansi/Alpine.jl","title":"Alpine, a global solver for non-convex MINLPs","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"(Image: CI) (Image: codecov) (Image: Documentation)  (Image: version)","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"ALPINE (glob(AL) o(P)timization for mixed-(I)nteger programs with (N)onlinear (E)quations), is a novel global optimization solver that uses an adaptive, piecewise convexification scheme and constraint programming methods to solve non-convex Mixed-Integer Non-Linear Programs (MINLPs) efficiently. MINLPs are typically \"hard\" optimization problems which appear in numerous applications (see MINLPLib.jl). ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine is entirely built upon JuMP and MathOptInterface in Julia, which provides incredible flexibility for usage and further development.","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine globally solves a given MINLP by:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Analyzing the problem's expressions (objective & constraints) and applies appropriate convex relaxations and polyhedral outer-approximations\nPerforming sequential optimization-based bound tightening (OBBT) and an iterative MIP-based adaptive partitioning scheme via piecewise polyhedral relaxations with a guarantee of global convergence","category":"page"},{"location":"packages/Alpine/#Installation","page":"lanl-ansi/Alpine.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Install Alpine using the Julia package manager:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"import Pkg\nPkg.add(\"Alpine\")","category":"page"},{"location":"packages/Alpine/#Usage-with-JuMP","page":"lanl-ansi/Alpine.jl","title":"Usage with JuMP","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Use Alpine with JuMP as follows:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"using JuMP, Alpine, Ipopt, HiGHS\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0)\nhighs = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false)\nmodel = Model(\n    optimizer_with_attributes(\n        Juniper.Optimizer,\n        \"nlp_solver\" => ipopt,\n        \"mip_solver\" => highs,\n    ),\n)","category":"page"},{"location":"packages/Alpine/#Documentation","page":"lanl-ansi/Alpine.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"For more details, see the online documentation.","category":"page"},{"location":"packages/Alpine/#Support-problem-types","page":"lanl-ansi/Alpine.jl","title":"Support problem types","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine can currently handle MINLPs with polynomials in constraints and/or in the objective. Currently, there is no support for exponential cones and Positive Semi-Definite (PSD) cones in MINLPs. Alpine is also a good fit for subsets of the MINLP family, for example, Mixed-Integer Quadratically Constrained Quadratic Programs (MIQCQPs), Non-Linear Programs (NLPs), etc.","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"For more details, check out this video on Alpine.jl at JuMP-dev 2018.","category":"page"},{"location":"packages/Alpine/#Underlying-solvers","page":"lanl-ansi/Alpine.jl","title":"Underlying solvers","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Though the MIP-based bounding algorithm implemented in Alpine is quite involved, most of the computational bottleneck arises in the underlying MIP solvers. Since every iteration of Alpine solves an MIP sub-problem, which is typically a convex MILP/MIQCQP, Alpine's run time heavily depends on the run-time of these solvers. For the best performance of Alpine, we recommend using the commercial solver Gurobi, which is available free for academic purposes. However, due to the flexibility offered by JuMP, the following MIP and NLP solvers are supported in Alpine: ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Solver Julia Package\nCPLEX CPLEX.jl\nCbc Cbc.jl\nGurobi Gurobi.jl\nIpopt Ipopt.jl\nBonmin Bonmin.jl\nArtelys KNITRO KNITRO.jl\nXpress Xpress.jl\nHiGHS HiGHS.jl","category":"page"},{"location":"packages/Alpine/#Bug-reports-and-support","page":"lanl-ansi/Alpine.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Please report any issues via the GitHub issue tracker. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. ","category":"page"},{"location":"packages/Alpine/#Challenging-Problems","page":"lanl-ansi/Alpine.jl","title":"Challenging Problems","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"We are seeking out hard benchmark instances for MINLPs. Please get in touch either by opening an issue or privately if you would like to share any hard instances.","category":"page"},{"location":"packages/Alpine/#Citing-Alpine","page":"lanl-ansi/Alpine.jl","title":"Citing Alpine","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"If you find Alpine useful in your work, we kindly request that you cite the following papers (PDF, PDF)","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"@article{alpine_JOGO2019,\n  title = {An adaptive, multivariate partitioning algorithm for global optimization of nonconvex programs},\n  author = {Nagarajan, Harsha and Lu, Mowen and Wang, Site and Bent, Russell and Sundar, Kaarthik},\n  journal = {Journal of Global Optimization},\n  year = {2019},\n  issn = {1573-2916},\n  doi = {10.1007/s10898-018-00734-1},\n}\n\n@inproceedings{alpine_CP2016,\n  title = {Tightening {McCormick} relaxations for nonlinear programs via dynamic multivariate partitioning},\n  author = {Nagarajan, Harsha and Lu, Mowen and Yamangil, Emre and Bent, Russell},\n  booktitle = {International Conference on Principles and Practice of Constraint Programming},\n  pages = {369--387},\n  year = {2016},\n  organization = {Springer},\n  doi = {10.1007/978-3-319-44953-1_24},\n}","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"If you find the underlying piecewise polyhedral formulations implemented in Alpine useful in your work, we kindly request that you cite the following papers (link-1, link-2): ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"@article{alpine_ORL2021,\n  title = {Piecewise polyhedral formulations for a multilinear term},\n  author = {Sundar, Kaarthik and Nagarajan, Harsha and Linderoth, Jeff and Wang, Site and Bent, Russell},\n  journal = {Operations Research Letters},\n  volume = {49},\n  number = {1},\n  pages = {144--149},\n  year = {2021},\n  publisher = {Elsevier}\n}\n\n@article{alpine_OptOnline2022,\n  title={Piecewise Polyhedral Relaxations of Multilinear Optimization},\n  author={Kim, Jongeun and Richard, Jean-Philippe P. and Tawarmalani, Mohit},\n  eprinttype={Optimization Online},\n  date={2022}\n}","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"EditURL = \"https://github.com/odow/NEOSServer.jl/blob/v1.1.0/README.md\"","category":"page"},{"location":"packages/NEOSServer/#NEOSServer.jl","page":"odow/NEOSServer.jl","title":"NEOSServer.jl","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOSServer.jl is a wrapper for the NEOS Server, a free internet-based service for solving numerical optimization problems.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"See here for the full list of solvers and input formats that NEOS supports.","category":"page"},{"location":"packages/NEOSServer/#License","page":"odow/NEOSServer.jl","title":"License","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOSServer.jl is licensed under the MIT License.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"However, use of the NEOS Server requires you to comply with NEOS Server terms of use. In particular, the commercial solvers are to be used solely for academic, non-commercial research purposes.","category":"page"},{"location":"packages/NEOSServer/#Installation","page":"odow/NEOSServer.jl","title":"Installation","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Install NEOSServer.jl using the package manager:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"import Pkg\nPkg.add(\"NEOSServer\")","category":"page"},{"location":"packages/NEOSServer/#The-NEOS-API","page":"odow/NEOSServer.jl","title":"The NEOS API","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"This package contains an interface for the NEOS XML-RPC API.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"The following example shows how you can interact with the API. Wrapped XML-RPC functions begin with neos_ and are exported.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"using NEOSServer\n\n# Create a server. You must supply a valid email:\nserver = NEOSServer.Server(\"me@mydomain.com\")\n\n# Print the NEOS welcome message:\nprintln(neos_welcome(server))\n\n# Get an XML template:\nxml_string = neos_getSolverTemplate(server, \"milp\", \"Cbc\", \"AMPL\")\n\n# Modify template with problem data...\n\n# Submit the XML job to NEOS:\njob = neos_submitJob(server, xml_string)\n\n# Get the status of the Job from NEOS:\nstatus = neos_getJobStatus(server, job)\n\n# Get the final results:\nresults = neos_getFinalResults(server, job)","category":"page"},{"location":"packages/NEOSServer/#Use-with-JuMP","page":"odow/NEOSServer.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Use NEOSServer.jl with JuMP as follows:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"using JuMP, NEOSServer\n\nmodel = Model() do\n    NEOSServer.Optimizer(email=\"me@mydomain.com\", solver=\"Ipopt\")\nend","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Note: NEOSServer.Optimizer is limited to the following solvers:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"\"CPLEX\"\n\"FICO-Xpress\"\n\"Ipopt\"\n\"Knitro\"\n\"MOSEK\"\n\"OCTERACT\"\n\"SNOPT\"","category":"page"},{"location":"packages/NEOSServer/#NEOS-Limits","page":"odow/NEOSServer.jl","title":"NEOS Limits","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOS currently limits jobs to an 8 hour time limit, 3 GB of memory, and a 16 MB submission file. If your model exceeds these limits, NEOSServer.jl may be unable to return useful information to the user.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"EditURL = \"https://github.com/jump-dev/SDPA.jl/blob/v0.5.0/README.md\"","category":"page"},{"location":"packages/SDPA/#SDPA.jl","page":"jump-dev/SDPA.jl","title":"SDPA.jl","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA.jl is a wrapper for the SDPA semidefinite programming solver in double precision floating point arithmetic.","category":"page"},{"location":"packages/SDPA/#Affiliation","page":"jump-dev/SDPA.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"This wrapper is maintained by the JuMP community and is not a product of the SDPA developers.","category":"page"},{"location":"packages/SDPA/#License","page":"jump-dev/SDPA.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The underlying solver, SDPA is licensed under the GPL v2 license.","category":"page"},{"location":"packages/SDPA/#Installation","page":"jump-dev/SDPA.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Install SDPA using Pkg.add:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"import Pkg\nPkg.add(\"SDPA\")","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"In addition to installing the SDPA.jl package, this will also download and install the SDPA binaries. (You do not need to install SDPA separately.)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"If you see an error similar to:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"INFO: Precompiling module GZip.\nERROR: LoadError: LoadError: error compiling anonymous: could not load library \"libz\"","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"please see GZip.jl#54 or Flux.jl#343. In particular, in Ubuntu this issue may be resolved by running","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"sudo apt-get install zlib1g-dev","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"See SDPAFamily for the other solvers, SDPA-GMP, SDPA-DD, and SDPA-QD of the family.","category":"page"},{"location":"packages/SDPA/#Use-with-JuMP","page":"jump-dev/SDPA.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_DEFAULT)","category":"page"},{"location":"packages/SDPA/#MathOptInterface-API","page":"jump-dev/SDPA.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The SDPA optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.Nonnegatives\nMOI.PositiveSemidefiniteConeTriangle","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/SDPA/#Options","page":"jump-dev/SDPA.jl","title":"Options","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA has three modes that give default value to all ten parameters.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The following table gives the default values for each parameter and mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Parameter PARAMETER_DEFAULT PARAMETER_UNSTABLE_BUT_FAST PARAMETER_STABLE_BUT_SLOW\nMaxIteration 100 100 1000\nEpsilonStar 1.0e-7 1.0e-7 1.0e-7\nLambdaStar 1.0e+2 1.0e+2 1.0e+4\nOmegaStar 2.0 2.0 2.0\nLowerBound 1.0e+5 1.0e+5 1.0e+5\nUpperBound 1.0e+5 1.0e+5 1.0e+5\nBetaStar 0.1 0.01 0.1\nBetaBar 0.2 0.02 0.3\nGammaStar 0.9 0.95 0.8\nEpsilonDash 1.0e-7 1.0e-7 1.0e-7","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"By default, we put SDPA in the SDPA.PARAMETER_DEFAULT mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Change the mode using the \"Mode\" option:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_STABLE_BUT_SLOW)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Note that the parameters are set in the order they are given, so you can set a mode and then modify parameters from this mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_STABLE_BUT_SLOW)\nset_attribute(model, \"MaxIteration\", 100)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The choice of parameter mode has a large impact on the performance and stability of SDPA, and not necessarily in the way implied by the names of the modes; for example, PARAMETER_UNSTABLE_BUT_FAST can be more stable than the other modes for some problems. You should try each mode to see how it performs on your specific problem. See SDPA.jl#17 for more details.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/mathprogbase.md\"","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-from-MathProgBase","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface is a replacement for MathProgBase.jl. However, it is not a direct replacement.","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-a-solver-interface","page":"Transitioning from MathProgBase","title":"Transitioning a solver interface","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface is more extensive than MathProgBase which may make its implementation seem daunting at first. There are however numerous utilities in MathOptInterface that the simplify implementation process.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"For more information, read Implementing a solver interface.","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-the-high-level-functions","page":"Transitioning from MathProgBase","title":"Transitioning the high-level functions","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface doesn't provide replacements for the high-level interfaces in MathProgBase. We recommend you use JuMP as a modeling interface instead.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"tip: Tip\nIf you haven't used JuMP before, start with the tutorial Getting started with JuMP","category":"page"},{"location":"moi/tutorials/mathprogbase/#linprog","page":"Transitioning from MathProgBase","title":"linprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from linprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction linprog(c, A, sense, b, l, u, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, l[i] <= x[i=1:N] <= u[i])\n    @objective(model, Min, c' * x)\n    eq_rows, ge_rows, le_rows = sense .== '=', sense .== '>', sense .== '<'\n    @constraint(model, A[eq_rows, :] * x .== b[eq_rows])\n    @constraint(model, A[ge_rows, :] * x .>= b[ge_rows])\n    @constraint(model, A[le_rows, :] * x .<= b[le_rows])\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/tutorials/mathprogbase/#mixintprog","page":"Transitioning from MathProgBase","title":"mixintprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from mixintprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction mixintprog(c, A, rowlb, rowub, vartypes, lb, ub, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, lb[i] <= x[i=1:N] <= ub[i])\n    for i in 1:N\n        if vartypes[i] == :Bin\n            set_binary(x[i])\n        elseif vartypes[i] == :Int\n            set_integer(x[i])\n        end\n    end\n    @objective(model, Min, c' * x)\n    @constraint(model, rowlb .<= A * x .<= rowub)\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/tutorials/mathprogbase/#quadprog","page":"Transitioning from MathProgBase","title":"quadprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from quadprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction quadprog(c, Q, A, rowlb, rowub, lb, ub, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, lb[i] <= x[i=1:N] <= ub[i])\n    @objective(model, Min, c' * x + 0.5 * x' * Q * x)\n    @constraint(model, rowlb .<= A * x .<= rowub)\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/reference.md\"","category":"page"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/reference/#Bridges","page":"API Reference","title":"Bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/#AbstractBridge-API","page":"API Reference","title":"AbstractBridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.AbstractBridge\nBridges.added_constrained_variable_types\nBridges.added_constraint_types\nget(::Bridges.AbstractBridge, ::NumberOfVariables)\nget(::Bridges.AbstractBridge, ::ListOfVariableIndices)\nget(::Bridges.AbstractBridge, ::NumberOfConstraints)\nget(::Bridges.AbstractBridge, ::ListOfConstraintIndices)\nBridges.needs_final_touch\nBridges.final_touch\nBridges.bridging_cost","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.AbstractBridge","text":"abstract type AbstractBridge end\n\nAn abstract type representing a bridged constraint or variable in a MOI.Bridges.AbstractBridgeOptimizer.\n\nAll bridges must implement:\n\nadded_constrained_variable_types\nadded_constraint_types\nMOI.get(::AbstractBridge, ::MOI.NumberOfVariables)\nMOI.get(::AbstractBridge, ::MOI.ListOfVariableIndices)\nMOI.get(::AbstractBridge, ::MOI.NumberOfConstraints)\nMOI.get(::AbstractBridge, ::MOI.ListOfConstraintIndices)\n\nSubtypes of AbstractBridge may have additional requirements. Consult their docstrings for details.\n\nIn addition, all subtypes may optionally implement the following constraint attributes with the bridge in place of the constraint index:\n\nMOI.ConstraintDual\nMOI.ConstraintPrimal\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.added_constrained_variable_types","page":"API Reference","title":"MathOptInterface.Bridges.added_constrained_variable_types","text":"added_constrained_variable_types(\n    BT::Type{<:AbstractBridge},\n)::Vector{Tuple{Type}}\n\nReturn a list of the types of constrained variables that bridges of concrete type BT add.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value. If the bridge may add a constrained variable, the type must be included in the return vector.\nIf the bridge adds a free variable via MOI.add_variable or MOI.add_variables, the return vector must include (MOI.Reals,).\n\nExample\n\njulia> MOI.Bridges.added_constrained_variable_types(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n       )\n1-element Vector{Tuple{Type}}:\n (MathOptInterface.Nonnegatives,)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types","page":"API Reference","title":"MathOptInterface.Bridges.added_constraint_types","text":"added_constraint_types(\n    BT::Type{<:AbstractBridge},\n)::Vector{Tuple{Type,Type}}\n\nReturn a list of the types of constraints that bridges of concrete type BT add.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value. If the bridge may add a constraint, the type must be included in the return vector.\n\nExample\n\njulia> MOI.Bridges.added_constraint_types(\n           MOI.Bridges.Constraint.ZeroOneBridge{Float64},\n       )\n2-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.Interval{Float64})\n (MathOptInterface.VariableIndex, MathOptInterface.Integer)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.NumberOfVariables}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64\n\nReturn the number of variables created by the bridge b in the model.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.ListOfVariableIndices}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)\n\nReturn the list of variables created by the bridge b.\n\nSee also MOI.ListOfVariableIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.NumberOfConstraints}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}\n\nReturn the number of constraints of the type F-in-S created by the bridge b.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.ListOfConstraintIndices}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}\n\nReturn a  Vector{ConstraintIndex{F,S}} with indices of all constraints of type F-in-S created by the bride b.\n\nSee also MOI.ListOfConstraintIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.needs_final_touch","page":"API Reference","title":"MathOptInterface.Bridges.needs_final_touch","text":"needs_final_touch(bridge::AbstractBridge)::Bool\n\nReturn whether final_touch is implemented by bridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.final_touch","page":"API Reference","title":"MathOptInterface.Bridges.final_touch","text":"final_touch(bridge::AbstractBridge, model::MOI.ModelLike)::Nothing\n\nA function that is called immediately prior to MOI.optimize! to allow bridges to modify their reformulations with repsect to other variables and constraints in model.\n\nFor example, if the correctness of bridge depends on the bounds of a variable or the fact that variables are integer, then the bridge can implement final_touch to check assumptions immediately before a call to MOI.optimize!.\n\nIf you implement this method, you must also implement needs_final_touch.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridging_cost","page":"API Reference","title":"MathOptInterface.Bridges.bridging_cost","text":"bridging_cost(b::AbstractBridgeOptimizer, S::Type{<:MOI.AbstractSet}})\n\nReturn the cost of bridging variables constrained in S on creation, is_bridged(b, S) is assumed to be true.\n\nbridging_cost(\n    b::AbstractBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)\n\nReturn the cost of bridging F-in-S constraints.\n\nis_bridged(b, S) is assumed to be true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Constraint-bridge-API","page":"API Reference","title":"Constraint bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Constraint.AbstractBridge\nsupports_constraint(::Type{<:Bridges.Constraint.AbstractBridge}, ::Type{<:AbstractFunction}, ::Type{<:AbstractSet})\nBridges.Constraint.concrete_bridge_type\nBridges.Constraint.bridge_constraint\nBridges.Constraint.AbstractFunctionConversionBridge\nBridges.Constraint.SingleBridgeOptimizer\nBridges.Constraint.add_all_bridges\nBridges.Constraint.FlipSignBridge\nBridges.Constraint.AbstractToIntervalBridge\nBridges.Constraint.SetMapBridge\nBridges.Constraint.conversion_cost","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractType\n\nSubtype of MOI.Bridges.AbstractBridge for constraint bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nMOI.supports_constraint(::Type{<:AbstractBridge}, ::Type{<:MOI.AbstractFunction}, ::Type{<:MOI.AbstractSet})\nconcrete_bridge_type\nbridge_constraint\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.supports_constraint-Tuple{Type{<:MathOptInterface.Bridges.Constraint.AbstractBridge}, Type{<:MathOptInterface.AbstractFunction}, Type{<:MathOptInterface.AbstractSet}}","page":"API Reference","title":"MathOptInterface.supports_constraint","text":"MOI.supports_constraint(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging F-in-S constraints.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method for constraint types that the bridge implements.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet}\n)::Type\n\nReturn the concrete type of the bridge supporting F-in-S constraints.\n\nThis function can only be called if MOI.supports_constraint(BT, F, S) is true.\n\nExample\n\nThe SplitIntervalBridge bridges a MOI.VariableIndex-in-MOI.Interval constraint into a MOI.VariableIndex-in-MOI.GreaterThan and a MOI.VariableIndex-in-MOI.LessThan constraint.\n\njulia> MOI.Bridges.Constraint.concrete_bridge_type(\n           MOI.Bridges.Constraint.SplitIntervalBridge{Float64},\n           MOI.VariableIndex,\n           MOI.Interval{Float64},\n       )\nMathOptInterface.Bridges.Constraint.SplitIntervalBridge{Float64, MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}, MathOptInterface.GreaterThan{Float64}, MathOptInterface.LessThan{Float64}}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.bridge_constraint","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.bridge_constraint","text":"bridge_constraint(\n    BT::Type{<:AbstractBridge},\n    model::MOI.ModelLike,\n    func::AbstractFunction,\n    set::MOI.AbstractSet,\n)::BT\n\nBridge the constraint func-in-set using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT should be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractFunctionConversionBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractFunctionConversionBridge","text":"abstract type AbstractFunctionConversionBridge{F,S} <: AbstractBridge end\n\nAbstract type to support writing bridges in which the function changes but the set does not.\n\nBy convention, the transformed function is stored in the .constraint field.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn AbstractBridgeOptimizer that always bridges any objective function supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the objective function if it is supported by the bridge BT and unsupported by model.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Constraint.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Constraint.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.add_all_bridges","text":"add_all_bridges(bridged_model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Constraint submodule to bridged_model. The coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.FlipSignBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.FlipSignBridge","text":"FlipSignBridge{T,S1,S2,F,G}\n\nAn abstract type that simplifies the creation of other bridges.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractToIntervalBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractToIntervalBridge","text":"AbstractToIntervalBridge{T<:AbstractFloat,S,F}\n\nAn abstract type that simplifies the creation of other bridges.\n\nwarning: Warning\nT must be a AbstractFloat type because otherwise typemin and typemax would either be not implemented (e.g. BigInt), or would not give infinite value (e.g. Int). For this reason, this bridge is only added to MOI.Bridges.full_bridge_optimizer when T is a subtype of AbstractFloat.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.SetMapBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.SetMapBridge","text":"abstract type SetMapBridge{T,S2,S1,F,G} <: AbstractBridge end\n\nConsider two type of sets, S1 and S2, and a linear mapping A such that the image of a set of type S1 under A is a set of type S2.\n\nA SetMapBridge{T,S2,S1,F,G} is a bridge that maps G-in-S1 constraints into F-in-S2 by mapping the function through A.\n\nThe linear map A is described by;\n\nMOI.Bridges.map_set\nMOI.Bridges.map_function.\n\nImplementing a method for these two functions is sufficient to bridge constraints. However, in order for the getters and setters of attributes such as dual solutions and starting values to work as well, a method for the following functions must be implemented:\n\nMOI.Bridges.inverse_map_set\nMOI.Bridges.inverse_map_function\nMOI.Bridges.adjoint_map_function\nMOI.Bridges.inverse_adjoint_map_function\n\nSee the docstrings of each function to see which feature would be missing if it was not implemented for a given bridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.conversion_cost","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.conversion_cost","text":"conversion_cost(\n    F::Type{<:MOI.AbstractFunction},\n    G::Type{<:MOI.AbstractFunction},\n)::Float64\n\nReturn a Float64 returning the cost of converting any function of type G to a function of type F with convert.\n\nThis cost is used to compute MOI.Bridges.bridging_cost.\n\nThe default cost is Inf, which means that MOI.Bridges.Constraint.FunctionConversionBridge should not attempt the conversion.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Objective-bridge-API","page":"API Reference","title":"Objective bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Objective.AbstractBridge\nBridges.Objective.supports_objective_function\nBridges.set_objective_function_type\nBridges.Objective.concrete_bridge_type\nBridges.Objective.bridge_objective\nBridges.Objective.SingleBridgeOptimizer\nBridges.Objective.add_all_bridges","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Objective.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractBridge end\n\nSubtype of MOI.Bridges.AbstractBridge for objective bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nsupports_objective_function\nconcrete_bridge_type\nbridge_objective\nMOI.Bridges.set_objective_function_type\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.supports_objective_function","page":"API Reference","title":"MathOptInterface.Bridges.Objective.supports_objective_function","text":"supports_objective_function(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging objective functions of type F.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method For objective functions that the bridge implements.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.set_objective_function_type","page":"API Reference","title":"MathOptInterface.Bridges.set_objective_function_type","text":"set_objective_function_type(\n    BT::Type{<:Objective.AbstractBridge},\n)::Type{<:MOI.AbstractFunction}\n\nReturn the type of objective function that bridges of concrete type BT set.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value.\n\nExample\n\njulia> MOI.Bridges.set_objective_function_type(\n           MOI.Bridges.Objective.FunctionizeBridge{Float64},\n       )\nMathOptInterface.ScalarAffineFunction{Float64}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Objective.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n)::Type\n\nReturn the concrete type of the bridge supporting objective functions of type F.\n\nThis function can only be called if MOI.supports_objective_function(BT, F) is true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.bridge_objective","page":"API Reference","title":"MathOptInterface.Bridges.Objective.bridge_objective","text":"bridge_objective(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    model::MOI.ModelLike,\n    func::MOI.AbstractFunction,\n)::BT\n\nBridge the objective function func using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT must be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Objective.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn AbstractBridgeOptimizer that always bridges any objective function supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the objective function if it is supported by the bridge BT and unsupported by model.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Objective.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Objective.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Objective.add_all_bridges","text":"add_all_bridges(model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Objective submodule to model.\n\nThe coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#ref_variable_bridges","page":"API Reference","title":"Variable bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Variable.AbstractBridge\nBridges.Variable.supports_constrained_variable\nBridges.Variable.concrete_bridge_type\nBridges.Variable.bridge_constrained_variable\nBridges.Variable.SingleBridgeOptimizer\nBridges.Variable.add_all_bridges\nBridges.Variable.FlipSignBridge\nBridges.Variable.SetMapBridge\nBridges.Variable.unbridged_map","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractBridge end\n\nSubtype of MOI.Bridges.AbstractBridge for variable bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nsupports_constrained_variable\nconcrete_bridge_type\nbridge_constrained_variable\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.supports_constrained_variable","page":"API Reference","title":"MathOptInterface.Bridges.Variable.supports_constrained_variable","text":"supports_constrained_variable(\n    BT::Type{<:AbstractBridge},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging constrained variables in S. That is, it returns true if the bridge of type BT converts constrained variables of type S into a form supported by the solver.\n\nImplementation notes\n\nThis method depends only on the type of the bridge and set, not the runtime values.\nThere is a default fallback, so you need only implement this method for sets that the bridge implements.\n\nExample\n\njulia> MOI.Bridges.Variable.supports_constrained_variable(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n           MOI.Nonpositives,\n       )\ntrue\n\njulia> MOI.Bridges.Variable.supports_constrained_variable(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n           MOI.Nonnegatives,\n       )\nfalse\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Variable.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:AbstractBridge},\n    S::Type{<:MOI.AbstractSet},\n)::Type\n\nReturn the concrete type of the bridge supporting variables in S constraints.\n\nThis function can only be called if MOI.supports_constrained_variable(BT, S) is true.\n\nExamples\n\nAs a variable in MOI.GreaterThan is bridged into variables in MOI.Nonnegatives by the VectorizeBridge:\n\njulia> MOI.Bridges.Variable.concrete_bridge_type(\n           MOI.Bridges.Variable.VectorizeBridge{Float64},\n           MOI.GreaterThan{Float64},\n       )\nMathOptInterface.Bridges.Variable.VectorizeBridge{Float64, MathOptInterface.Nonnegatives}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.bridge_constrained_variable","page":"API Reference","title":"MathOptInterface.Bridges.Variable.bridge_constrained_variable","text":"bridge_constrained_variable(\n    BT::Type{<:AbstractBridge},\n    model::MOI.ModelLike,\n    set::MOI.AbstractSet,\n)::BT\n\nBridge the constrained variable in set using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT must be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Variable.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn MOI.Bridges.AbstractBridgeOptimizer that always bridges any variables constrained on creation supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the variables constrained on creation if they are supported by the bridge BT and unsupported by model.\n\nwarning: Warning\nTwo SingleBridgeOptimizers cannot be used together as both of them assume that the underlying model only returns variable indices with nonnegative values. Use MOI.Bridges.LazyBridgeOptimizer instead.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Variable.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 variable bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Variable.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 variable bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Variable.add_all_bridges","text":"add_all_bridges(model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Variable submodule to model.\n\nThe coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.FlipSignBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.FlipSignBridge","text":"abstract type FlipSignBridge{T,S1,S2} <: SetMapBridge{T,S2,S1} end\n\nAn abstract type that simplifies the creation of other bridges.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.SetMapBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.SetMapBridge","text":"abstract type SetMapBridge{T,S1,S2} <: AbstractBridge end\n\nConsider two type of sets, S1 and S2, and a linear mapping A such that the image of a set of type S1 under A is a set of type S2.\n\nA SetMapBridge{T,S1,S2} is a bridge that substitutes constrained variables in S2 into the image through A of constrained variables in S1.\n\nThe linear map A is described by:\n\nMOI.Bridges.map_set\nMOI.Bridges.map_function\n\nImplementing a method for these two functions is sufficient to bridge constrained variables. However, in order for the getters and setters of attributes such as dual solutions and starting values to work as well, a method for the following functions must be implemented:\n\nMOI.Bridges.inverse_map_set\nMOI.Bridges.inverse_map_function\nMOI.Bridges.adjoint_map_function\nMOI.Bridges.inverse_adjoint_map_function.\n\nSee the docstrings of each function to see which feature would be missing if it was not implemented for a given bridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.unbridged_map","page":"API Reference","title":"MathOptInterface.Bridges.Variable.unbridged_map","text":"unbridged_map(\n   bridge::MOI.Bridges.Variable.AbstractBridge,\n    vi::MOI.VariableIndex,\n)\n\nFor a bridged variable in a scalar set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vi.\n\nunbridged_map(\n    bridge::MOI.Bridges.Variable.AbstractBridge,\n    vis::Vector{MOI.VariableIndex},\n)\n\nFor a bridged variable in a vector set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vis. If this method is not implemented, it falls back to calling the following method for every variable of vis.\n\nunbridged_map(\n    bridge::MOI.Bridges.Variable.AbstractBridge,\n    vi::MOI.VariableIndex,\n    i::MOI.Bridges.IndexInVector,\n)\n\nFor a bridged variable in a vector set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vi corresponding to the ith variable of the vector.\n\nIf there is no way to recover the expression in terms of the bridged variable(s) vi(s), return nothing. See ZerosBridge for an example of bridge returning nothing.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#AbstractBridgeOptimizer-API","page":"API Reference","title":"AbstractBridgeOptimizer API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.AbstractBridgeOptimizer\nBridges.bridged_variable_function\nBridges.unbridged_variable_function\nBridges.bridged_function\nBridges.supports_constraint_bridges\nBridges.recursive_model","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.AbstractBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.AbstractBridgeOptimizer","text":"abstract type AbstractBridgeOptimizer <: MOI.AbstractOptimizer end\n\nAn abstract type that implements generic functions for bridges.\n\nImplementation notes\n\nBy convention, the inner optimizer should be stored in a model field. If not, the optimizer must implement MOI.optimize!.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridged_variable_function","page":"API Reference","title":"MathOptInterface.Bridges.bridged_variable_function","text":"bridged_variable_function(\n    b::AbstractBridgeOptimizer,\n    vi::MOI.VariableIndex,\n)\n\nReturn a MOI.AbstractScalarFunction of variables of b.model that equals vi. That is, if the variable vi is bridged, it returns its expression in terms of the variables of b.model. Otherwise, it returns vi.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.unbridged_variable_function","page":"API Reference","title":"MathOptInterface.Bridges.unbridged_variable_function","text":"unbridged_variable_function(\n    b::AbstractBridgeOptimizer,\n    vi::MOI.VariableIndex,\n)\n\nReturn a MOI.AbstractScalarFunction of variables of b that equals vi. That is, if the variable vi is an internal variable of b.model created by a bridge but not visible to the user, it returns its expression in terms of the variables of bridged variables. Otherwise, it returns vi.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridged_function","page":"API Reference","title":"MathOptInterface.Bridges.bridged_function","text":"bridged_function(b::AbstractBridgeOptimizer, value)::typeof(value)\n\nSubstitute any bridged MOI.VariableIndex in value by an equivalent expression in terms of variables of b.model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.supports_constraint_bridges","page":"API Reference","title":"MathOptInterface.Bridges.supports_constraint_bridges","text":"supports_constraint_bridges(b::AbstractBridgeOptimizer)::Bool\n\nReturn a Bool indicating if b supports MOI.Bridges.Constraint.AbstractBridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.recursive_model","page":"API Reference","title":"MathOptInterface.Bridges.recursive_model","text":"recursive_model(b::AbstractBridgeOptimizer)\n\nIf a variable, constraint, or objective is bridged, return the context of the inner variables. For most optimizers, this should be b.model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#LazyBridgeOptimizer-API","page":"API Reference","title":"LazyBridgeOptimizer API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.LazyBridgeOptimizer\nBridges.full_bridge_optimizer\nBridges.ListOfNonstandardBridges\nBridges.add_bridge\nBridges.remove_bridge\nBridges.has_bridge\nBridges.print_active_bridges\nBridges.print_graph\nBridges.debug_supports_constraint\nBridges.debug_supports","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.LazyBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.LazyBridgeOptimizer","text":"LazyBridgeOptimizer(model::MOI.ModelLike)\n\nThe LazyBridgeOptimizer is a bridge optimizer that supports multiple bridges, and only bridges things which are not supported by the internal model.\n\nInternally, the LazyBridgeOptimizer solves a shortest hyper-path problem to determine which bridges to use.\n\nIn general, you should use full_bridge_optimizer instead of this constructor because full_bridge_optimizer automatically adds a large number of supported bridges.\n\nSee also: add_bridge, remove_bridge, has_bridge and full_bridge_optimizer.\n\nExample\n\njulia> model = MOI.Bridges.LazyBridgeOptimizer(MOI.Utilities.Model{Float64}())\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\njulia> MOI.Bridges.add_bridge(model, MOI.Bridges.Variable.FreeBridge{Float64})\n\njulia> MOI.Bridges.has_bridge(model, MOI.Bridges.Variable.FreeBridge{Float64})\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.full_bridge_optimizer","page":"API Reference","title":"MathOptInterface.Bridges.full_bridge_optimizer","text":"full_bridge_optimizer(model::MOI.ModelLike, ::Type{T}) where {T}\n\nReturns a LazyBridgeOptimizer bridging model for every bridge defined in this package (see below for the few exceptions) and for the coefficient type T, as well as the bridges in the list returned by the ListOfNonstandardBridges attribute.\n\nExample\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> bridged_model = MOI.Bridges.full_bridge_optimizer(model, Float64);\n\nExceptions\n\nThe following bridges are not added by full_bridge_optimizer, except if they are in the list returned by the ListOfNonstandardBridges attribute:\n\nConstraint.SOCtoNonConvexQuadBridge\nConstraint.RSOCtoNonConvexQuadBridge](@ref)\nConstraint.SOCtoPSDBridge\nIf T is not a subtype of AbstractFloat, subtypes of Constraint.AbstractToIntervalBridge\nConstraint.GreaterToIntervalBridge\nConstraint.LessToIntervalBridge)\n\nSee the docstring of the each bridge for the reason they are not added.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ListOfNonstandardBridges","page":"API Reference","title":"MathOptInterface.Bridges.ListOfNonstandardBridges","text":"ListOfNonstandardBridges{T}() <: MOI.AbstractOptimizerAttribute\n\nAny optimizer can be wrapped in a LazyBridgeOptimizer using full_bridge_optimizer. However, by default LazyBridgeOptimizer uses a limited set of bridges that are:\n\nimplemented in MOI.Bridges\ngenerally applicable for all optimizers.\n\nFor some optimizers however, it is useful to add additional bridges, such as those that are implemented in external packages (e.g., within the solver package itself) or only apply in certain circumstances (e.g., Constraint.SOCtoNonConvexQuadBridge).\n\nSuch optimizers should implement the ListOfNonstandardBridges attribute to return a vector of bridge types that are added by full_bridge_optimizer in addition to the list of default bridges.\n\nNote that optimizers implementing ListOfNonstandardBridges may require package-specific functions or sets to be used if the non-standard bridges are not added. Therefore, you are recommended to use model = MOI.instantiate(Package.Optimizer; with_bridge_type = T) instead of model = MOI.instantiate(Package.Optimizer). See MOI.instantiate.\n\nExamples\n\nAn optimizer using a non-default bridge in MOI.Bridges\n\nSolvers supporting MOI.ScalarQuadraticFunction can support MOI.SecondOrderCone and MOI.RotatedSecondOrderCone by defining:\n\nfunction MOI.get(::MyQuadraticOptimizer, ::ListOfNonstandardBridges{Float64})\n    return Type[\n        MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge{Float64},\n        MOI.Bridges.Constraint.RSOCtoNonConvexQuadBridge{Float64},\n    ]\nend\n\nAn optimizer defining an internal bridge\n\nSuppose an optimizer can exploit specific structure of a constraint, e.g., it can exploit the structure of the matrix A in the linear system of equations A * x = b.\n\nThe optimizer can define the function:\n\nstruct MatrixAffineFunction{T} <: MOI.AbstractVectorFunction\n    A::SomeStructuredMatrixType{T}\n    b::Vector{T}\nend\n\nand then a bridge\n\nstruct MatrixAffineFunctionBridge{T} <: MOI.Constraint.AbstractBridge\n    # ...\nend\n# ...\n\nfrom VectorAffineFunction{T} to the MatrixAffineFunction. Finally, it defines:\n\nfunction MOI.get(::Optimizer{T}, ::ListOfNonstandardBridges{T}) where {T}\n    return Type[MatrixAffineFunctionBridge{T}]\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_bridge","page":"API Reference","title":"MathOptInterface.Bridges.add_bridge","text":"add_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nEnable the use of the bridges of type BT by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.remove_bridge","page":"API Reference","title":"MathOptInterface.Bridges.remove_bridge","text":"remove_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nDisable the use of the bridges of type BT by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.has_bridge","page":"API Reference","title":"MathOptInterface.Bridges.has_bridge","text":"has_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nReturn a Bool indicating whether the bridges of type BT are used by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.print_active_bridges","page":"API Reference","title":"MathOptInterface.Bridges.print_active_bridges","text":"print_active_bridges([io::IO=stdout,] b::MOI.Bridges.LazyBridgeOptimizer)\n\nPrint the set of bridges that are active in the model b.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction}\n)\n\nPrint the set of bridges required for an objective function of type F.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint the set of bridges required for a constraint of type F-in-S.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    S::Type{<:MOI.AbstractSet}\n)\n\nPrint the set of bridges required for a variable constrained to set S.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.print_graph","page":"API Reference","title":"MathOptInterface.Bridges.print_graph","text":"print_graph([io::IO = stdout,] b::LazyBridgeOptimizer)\n\nPrint the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model by all the bridges added to b.\n\nEach node in the hyper-graph corresponds to a variable, constraint, or objective type.\n\nVariable nodes are indicated by [ ]\nConstraint nodes are indicated by ( )\nObjective nodes are indicated by | |\n\nThe number inside each pair of brackets is an index of the node in the hyper-graph.\n\nNote that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.\n\nTo see which nodes are used, call print_active_bridges.\n\nFor more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  \"MathOptInterface: a data structure for mathematical optimization problems.\" URL: https://arxiv.org/abs/2002.03447\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.debug_supports_constraint","page":"API Reference","title":"MathOptInterface.Bridges.debug_supports_constraint","text":"debug_supports_constraint(\n    b::LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet};\n    io::IO = Base.stdout,\n)\n\nPrints to io explanations for the value of MOI.supports_constraint with the same arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.debug_supports","page":"API Reference","title":"MathOptInterface.Bridges.debug_supports","text":"debug_supports(\n    b::LazyBridgeOptimizer,\n    ::MOI.ObjectiveFunction{F};\n    io::IO = Base.stdout,\n) where F\n\nPrints to io explanations for the value of MOI.supports with the same arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#constraint_set_map","page":"API Reference","title":"SetMap API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.map_set\nBridges.inverse_map_set\nBridges.map_function\nBridges.inverse_map_function\nBridges.adjoint_map_function\nBridges.inverse_adjoint_map_function","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.map_set","page":"API Reference","title":"MathOptInterface.Bridges.map_set","text":"map_set(::Type{BT}, set) where {BT}\n\nReturn the image of set through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for bridging the constraint and setting the MOI.ConstraintSet.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_map_set","page":"API Reference","title":"MathOptInterface.Bridges.inverse_map_set","text":"inverse_map_set(::Type{BT}, set) where {BT}\n\nReturn the preimage of set through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintSet.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.map_function","page":"API Reference","title":"MathOptInterface.Bridges.map_function","text":"map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintPrimal of variable bridges. For constraint bridges, this is used for bridging the constraint, setting the MOI.ConstraintFunction and MOI.ConstraintPrimalStart and modifying the function with MOI.modify.\n\nmap_function(::Type{BT}, func, i::IndexInVector) where {BT}\n\nReturn the scalar function at the ith index of the vector function that would be returned by map_function(BT, func) except that it may compute the ith element. This is used by bridged_function and for getting the MOI.VariablePrimal and MOI.VariablePrimalStart of variable bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_map_function","page":"API Reference","title":"MathOptInterface.Bridges.inverse_map_function","text":"inverse_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the inverse of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used by Variable.unbridged_map and for setting the MOI.VariablePrimalStart of variable bridges and for getting the MOI.ConstraintFunction, the MOI.ConstraintPrimal and the MOI.ConstraintPrimalStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.adjoint_map_function","page":"API Reference","title":"MathOptInterface.Bridges.adjoint_map_function","text":"adjoint_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the adjoint of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintDual and MOI.ConstraintDualStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_adjoint_map_function","page":"API Reference","title":"MathOptInterface.Bridges.inverse_adjoint_map_function","text":"inverse_adjoint_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the inverse of the adjoint of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintDual of variable bridges and setting the MOI.ConstraintDualStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Bridging-graph-API","page":"API Reference","title":"Bridging graph API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Graph\nBridges.VariableNode\nBridges.ConstraintNode\nBridges.ObjectiveNode\nBridges.Edge\nBridges.ObjectiveEdge\nBridges.add_node\nBridges.add_edge\nBridges.set_variable_constraint_node\nBridges.bridge_index\nBridges.is_variable_edge_best","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Graph","page":"API Reference","title":"MathOptInterface.Bridges.Graph","text":"Graph()\n\nA type-stable datastructure for computing the shortest hyperpath problem.\n\nNodes\n\nThere are three types of nodes in the graph:\n\nVariableNode\nConstraintNode\nObjectiveNode\n\nAdd nodes to the graph using add_node.\n\nEdges\n\nThere are two types of edges in the graph:\n\nEdge\nObjectiveEdge\n\nAdd edges to the graph using add_edge.\n\nFor the ability to add a variable constrained  on creation as a free variable followed by a constraint, use set_variable_constraint_node.\n\nOptimal hyper-edges\n\nUse bridge_index to compute the minimum-cost bridge leaving a node.\n\nNote that bridge_index lazy runs a Bellman-Ford algorithm to compute the set of minimum cost edges. Thus, the first call to bridge_index after adding new nodes or edges will take longer than subsequent calls.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.VariableNode","page":"API Reference","title":"MathOptInterface.Bridges.VariableNode","text":"VariableNode(index::Int)\n\nA node in Graph representing a variable constrained on creation.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ConstraintNode","page":"API Reference","title":"MathOptInterface.Bridges.ConstraintNode","text":"ConstraintNode(index::Int)\n\nA node in Graph representing a constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ObjectiveNode","page":"API Reference","title":"MathOptInterface.Bridges.ObjectiveNode","text":"ObjectiveNode(index::Int)\n\nA node in Graph representing an objective function.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Edge","page":"API Reference","title":"MathOptInterface.Bridges.Edge","text":"Edge(\n    bridge_index::Int,\n    added_variables::Vector{VariableNode},\n    added_constraints::Vector{ConstraintNode},\n    cost::Float64 = 1.0,\n)\n\nReturn a new datastructure representing an edge in Graph that starts at a VariableNode or a ConstraintNode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ObjectiveEdge","page":"API Reference","title":"MathOptInterface.Bridges.ObjectiveEdge","text":"ObjectiveEdge(\n    bridge_index::Int,\n    added_variables::Vector{VariableNode},\n    added_constraints::Vector{ConstraintNode},\n)\n\nReturn a new datastructure representing an edge in Graph that starts at an ObjectiveNode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_node","page":"API Reference","title":"MathOptInterface.Bridges.add_node","text":"add_node(graph::Graph, ::Type{VariableNode})::VariableNode\nadd_node(graph::Graph, ::Type{ConstraintNode})::ConstraintNode\nadd_node(graph::Graph, ::Type{ObjectiveNode})::ObjectiveNode\n\nAdd a new node to graph.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_edge","page":"API Reference","title":"MathOptInterface.Bridges.add_edge","text":"add_edge(graph::Graph, node::VariableNode, edge::Edge)::Nothing\nadd_edge(graph::Graph, node::ConstraintNode, edge::Edge)::Nothing\nadd_edge(graph::Graph, node::ObjectiveNode, edge::ObjectiveEdge)::Nothing\n\nAdd edge to graph, where edge starts at node and connects to the nodes defined in edge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.set_variable_constraint_node","page":"API Reference","title":"MathOptInterface.Bridges.set_variable_constraint_node","text":"set_variable_constraint_node(\n    graph::Graph,\n    variable_node::VariableNode,\n    constraint_node::ConstraintNode,\n    cost::Int,\n)\n\nAs an alternative to variable_node, add a virtual edge to graph that represents adding a free variable, followed by a constraint of type constraint_node, with bridging cost cost.\n\nWhy is this needed?\n\nVariables can either be added as a variable constrained on creation, or as a free variable which then has a constraint added to it.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridge_index","page":"API Reference","title":"MathOptInterface.Bridges.bridge_index","text":"bridge_index(graph::Graph, node::VariableNode)::Int\nbridge_index(graph::Graph, node::ConstraintNode)::Int\nbridge_index(graph::Graph, node::ObjectiveNode)::Int\n\nReturn the optimal index of the bridge to chose from node.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.is_variable_edge_best","page":"API Reference","title":"MathOptInterface.Bridges.is_variable_edge_best","text":"is_variable_edge_best(graph::Graph, node::VariableNode)::Bool\n\nReturn a Bool indicating whether node should be added as a variable constrained on creation, or as a free variable followed by a constraint.\n\n\n\n\n\n","category":"function"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"EditURL = \"https://github.com/kocvara/Loraine.jl/blob/v0.1.2/README.md\"","category":"page"},{"location":"packages/Loraine/#Loraine.jl","page":"kocvara/Loraine.jl","title":"Loraine.jl","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Sweet Lor(r)aine, let the party carry on[1]...","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"[1]: https://www.youtube.com/watch?v=0D2wNf1lVrI","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine.jl is a Julia implementation of an interior point method algorithm for linear semidefinite optimization problems. ","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The special feature of Loraine is the iterative solver for linear systems. This is to be used for problems with (very) low rank solution matrix.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Standard (non-low-rank) problems and linear programs can be solved using the direct solver; then the user gets a standard IP method akin SDPT3.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"There is also a MATLAB version of the code at kocvara/Loraine.m.","category":"page"},{"location":"packages/Loraine/#License-and-Original-Contributors","page":"kocvara/Loraine.jl","title":"License and Original Contributors","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine is licensed under the MIT License.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine was developed by Soodeh Habibi and Michal Kočvara, University of Birmingham, and Michael Stingl, University of Erlangen, for H2020 ITN POEMA. ","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The JuMP interface was provided by Benoît Legat. His help is greatly acknowledged.","category":"page"},{"location":"packages/Loraine/#Installation","page":"kocvara/Loraine.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Install Loraine using Pkg.add:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"import Pkg\nPkg.add(\"Loraine\")","category":"page"},{"location":"packages/Loraine/#Use-with-JuMP","page":"kocvara/Loraine.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"To use Loraine with JuMP, use Loraine.Optimizer:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"using JuMP, Loraine\nmodel = Model(Loraine.Optimizer)\nset_attribute(model, \"maxit\", 100)","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"To solve an SDP problem stored in SDPA format, do","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"using JuMP, Loraine\nmodel = read_from_file(\"examples/data/theta1.dat-s\")\nset_optimizer(model, Loraine.Optimizer)\noptimize!(model)","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"For more examples, the folder examples includes a few examples of how to use Loraine via JuMP; in particular, solve_sdpa.jl reads an SDP in the SDPA input format and solves it by Loraine. A few sample problems can be found in folder examples/data.","category":"page"},{"location":"packages/Loraine/#Rank-one-data","page":"kocvara/Loraine.jl","title":"Rank-one data","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"If the solution does not have low rank, it is recommended to use a direct  solver kit = 0. However, if you know that your data matrices are all rank-one,  use the option datarank = -1 to get a significant reduction in the complexity  (and CPU time). Examples of such problems are maxG11 and thetaG11 from the  SDPLIB collection.","category":"page"},{"location":"packages/Loraine/#Options","page":"kocvara/Loraine.jl","title":"Options","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The list of options:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"kit             # kit = 0 for direct solver; kit = 1 for CG [0]\ntol_cg          # tolerance for CG solver [1.0e-2]\ntol_cg_up       # tolerance update [0.5]\ntol_cg_min      # minimal tolerance for CG solver [1.0e-6]\neDIMACS         # epsilon for DIMACS error stopping criterion [1.0e-5]\npreconditioner  # 0...no; 1...H_alpha; 2...H_beta; 4...hybrid [1]\nerank           # estimated rank [1]\naamat           # 0..A^TA; 1..diag(A^TA); 2..identity [2]\nverb            # 2..full output; 1..short output; 0..no output [1]\ndatarank        # 0..full rank matrices expected [0]\n                # -1..rank-1 matrices expected, converted to vectors, if possible\n                # (TBD) 1..vectors expected for low-rank data matrices\ninitpoint       # 0..Loraine heuristics, 1..SDPT3-like heuristics [0]\ntiming          # 1..yes, 0..no\nmaxit           # maximal number of global iterations [200]","category":"page"},{"location":"packages/Loraine/#Citing","page":"kocvara/Loraine.jl","title":"Citing","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"If you find Loraine useful, please cite the following paper:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"@article{loraine2023,\n  title={Loraine-An interior-point solver for low-rank semidefinite programming},\n  author={Habibi, Soodeh and Ko{\\v{c}}vara, Michal and Stingl, Michael},\n  www={https://hal.science/hal-04076509/}\n  note={Preprint hal-04076509}\n  year={2023}\n}","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"EditURL = nothing","category":"page"},{"location":"api/JuMP.Containers/#DocumenterReference_JuMP.Containers","page":"JuMP.Containers","title":"JuMP.Containers","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"This page lists the public API of JuMP.Containers.","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"info: Info\nThis page is an unstructured list of the JuMP.Containers API. For a more structured overview, read the Manual or Tutorial parts of this documentation.","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"Load all of the public the API into the current scope with:","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"using JuMP.Containers","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"Alternatively, load only the module with:","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"import JuMP.Containers","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"and then prefix all calls with JuMP.Containers. to create JuMP.Containers.<NAME>.","category":"page"},{"location":"api/JuMP.Containers/#DenseAxisArray","page":"JuMP.Containers","title":"DenseAxisArray","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.DenseAxisArray","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.DenseAxisArray","page":"JuMP.Containers","title":"JuMP.Containers.DenseAxisArray","text":"DenseAxisArray(data::Array{T, N}, axes...) where {T, N}\n\nConstruct a JuMP array with the underlying data specified by the data array and the given axes. Exactly N axes must be provided, and their lengths must match size(data) in the corresponding dimensions.\n\nExample\n\njulia> array = Containers.DenseAxisArray([1 2; 3 4], [:a, :b], 2:3)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 2:3\nAnd data, a 2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> array[:b, 3]\n4\n\n\n\n\n\nDenseAxisArray{T}(undef, axes...) where T\n\nConstruct an uninitialized DenseAxisArray with element-type T indexed over the given axes.\n\nExample\n\njulia> array = Containers.DenseAxisArray{Float64}(undef, [:a, :b], 1:2);\n\njulia> fill!(array, 1.0)\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 1:2\nAnd data, a 2×2 Matrix{Float64}:\n 1.0  1.0\n 1.0  1.0\n\njulia> array[:a, 2] = 5.0\n5.0\n\njulia> array[:a, 2]\n5.0\n\njulia> array\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 1:2\nAnd data, a 2×2 Matrix{Float64}:\n 1.0  5.0\n 1.0  1.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#SparseAxisArray","page":"JuMP.Containers","title":"SparseAxisArray","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.SparseAxisArray","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.SparseAxisArray","page":"JuMP.Containers","title":"JuMP.Containers.SparseAxisArray","text":"struct SparseAxisArray{T,N,K<:NTuple{N, Any}} <: AbstractArray{T,N}\n    data::Dict{K,T}\nend\n\nN-dimensional array with elements of type T where only a subset of the entries are defined. The entries with indices idx = (i1, i2, ..., iN) in keys(data) has value data[idx]. Note that as opposed to SparseArrays.AbstractSparseArray, the missing entries are not assumed to be zero(T), they are simply not part of the array. This means that the result of map(f, sa::SparseAxisArray) or f.(sa::SparseAxisArray) has the same sparsity structure than sa even if f(zero(T)) is not zero.\n\nExample\n\njulia> dict = Dict((:a, 2) => 1.0, (:a, 3) => 2.0, (:b, 3) => 3.0)\nDict{Tuple{Symbol, Int64}, Float64} with 3 entries:\n  (:a, 3) => 2.0\n  (:b, 3) => 3.0\n  (:a, 2) => 1.0\n\njulia> array = Containers.SparseAxisArray(dict)\nSparseAxisArray{Float64, 2, Tuple{Symbol, Int64}} with 3 entries:\n  [a, 2]  =  1.0\n  [a, 3]  =  2.0\n  [b, 3]  =  3.0\n\njulia> array[:b, 3]\n3.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.@container","page":"JuMP.Containers","title":"Containers.@container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.@container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.@container","page":"JuMP.Containers","title":"JuMP.Containers.@container","text":"@container([i=..., j=..., ...], expr[, container = :Auto])\n\nCreate a container with indices i, j, ... and values given by expr that may depend on the value of the indices.\n\n@container(ref[i=..., j=..., ...], expr[, container = :Auto])\n\nSame as above but the container is assigned to the variable of name ref.\n\nThe type of container can be controlled by the container keyword.\n\nnote: Note\nWhen the index set is explicitly given as 1:n for any expression n, it is transformed to Base.OneTo(n) before being given to container.\n\nExample\n\njulia> Containers.@container([i = 1:3, j = 1:3], i + j)\n3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> I = 1:3\n1:3\n\njulia> Containers.@container(x[i = I, j = I], i + j);\n\njulia> x\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, 1:3\nAnd data, a 3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.@container([i = 2:3, j = 1:3], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, Base.OneTo(3)\nAnd data, a 2×3 Matrix{Int64}:\n 3  4  5\n 4  5  6\n\njulia> Containers.@container([i = 1:3, j = 1:3; i <= j], i + j)\nSparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  2\n  [1, 2]  =  3\n  [1, 3]  =  4\n  [2, 2]  =  4\n  [2, 3]  =  5\n  [3, 3]  =  6\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP.Containers/#Containers.container","page":"JuMP.Containers","title":"Containers.container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.container","page":"JuMP.Containers","title":"JuMP.Containers.container","text":"container(f::Function, indices[[, ::Type{C} = AutoContainerType], names])\n\nCreate a container of type C with index names names, indices indices and values at given indices given by f.\n\nIf the method with names is not specialized on Type{C}, it falls back to calling  container(f, indices, c) for backwards compatibility with containers not supporting index names.\n\nExample\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(Base.OneTo(3), Base.OneTo(3)))\n3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(1:3, 1:3))\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, 1:3\nAnd data, a 3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(2:3, Base.OneTo(3)))\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, Base.OneTo(3)\nAnd data, a 2×3 Matrix{Int64}:\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.nested(() -> 1:3, i -> i:3, condition = (i, j) -> isodd(i) || isodd(j)))\nSparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 5 entries:\n  [1, 1]  =  2\n  [1, 2]  =  3\n  [1, 3]  =  4\n  [2, 3]  =  5\n  [3, 3]  =  6\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.rowtable","page":"JuMP.Containers","title":"Containers.rowtable","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.rowtable","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.rowtable","page":"JuMP.Containers","title":"JuMP.Containers.rowtable","text":"rowtable([f::Function=identity,] x; [header::Vector{Symbol} = Symbol[]])\n\nApplies the function f to all elements of the variable container x, returning the result as a Vector of NamedTuples, where header is a vector containing the corresponding axis names.\n\nIf x is an N-dimensional array, there must be N+1 names, so that the last name corresponds to the result of f(x[i]).\n\nIf header is left empty, then the default header is [:x1, :x2, ..., :xN, :y].\n\ninfo: Info\nA Vector of NamedTuples implements the Tables.jl interface, and so the result can be used as input for any function that consumes a 'Tables.jl' compatible source.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=i:2] >= 0, start = i+j);\n\njulia> Containers.rowtable(start_value, x; header = [:i, :j, :start])\n3-element Vector{NamedTuple{(:i, :j, :start), Tuple{Int64, Int64, Float64}}}:\n (i = 1, j = 2, start = 3.0)\n (i = 1, j = 1, start = 2.0)\n (i = 2, j = 2, start = 4.0)\n\njulia> Containers.rowtable(x)\n3-element Vector{NamedTuple{(:x1, :x2, :y), Tuple{Int64, Int64, VariableRef}}}:\n (x1 = 1, x2 = 2, y = x[1,2])\n (x1 = 1, x2 = 1, y = x[1,1])\n (x1 = 2, x2 = 2, y = x[2,2])\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.default_container","page":"JuMP.Containers","title":"Containers.default_container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.default_container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.default_container","page":"JuMP.Containers","title":"JuMP.Containers.default_container","text":"default_container(indices)\n\nIf indices is a NestedIterator, return a SparseAxisArray. Otherwise, indices should be a VectorizedProductIterator and the function returns Array if all iterators of the product are Base.OneTo and retunrs DenseAxisArray otherwise.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.nested","page":"JuMP.Containers","title":"Containers.nested","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.nested","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.nested","page":"JuMP.Containers","title":"JuMP.Containers.nested","text":"nested(iterators...; condition = (args...) -> true)\n\nCreate a NestedIterator.\n\nExample\n\njulia> iterator = Containers.nested(\n           () -> 1:2,\n           (i,) -> [\"A\", \"B\"];\n           condition = (i, j) -> isodd(i) || j == \"B\",\n       );\n\njulia> collect(iterator)\n3-element Vector{Tuple{Int64, String}}:\n (1, \"A\")\n (1, \"B\")\n (2, \"B\")\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.vectorized_product","page":"JuMP.Containers","title":"Containers.vectorized_product","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.vectorized_product","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.vectorized_product","page":"JuMP.Containers","title":"JuMP.Containers.vectorized_product","text":"vectorized_product(iterators...)\n\nCreated a VectorizedProductIterator.\n\nExample\n\njulia> iterator = Containers.vectorized_product(1:2, [\"A\", \"B\"]);\n\njulia> collect(iterator)\n2×2 Matrix{Tuple{Int64, String}}:\n (1, \"A\")  (1, \"B\")\n (2, \"A\")  (2, \"B\")\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.build_ref_sets","page":"JuMP.Containers","title":"Containers.build_ref_sets","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.build_ref_sets","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.build_ref_sets","page":"JuMP.Containers","title":"JuMP.Containers.build_ref_sets","text":"build_ref_sets(_error::Function, expr)\n\nHelper function for macros to construct container objects.\n\nwarning: Warning\nThis function is for advanced users implementing JuMP extensions. See container_code for more details.\n\nArguments\n\n_error: a function that takes a String and throws an error, potentially annotating the input string with extra information such as from which macro it was thrown from. Use error if you do not want a modified error message.\nexpr: an Expr that specifies the container, e.g., :(x[i = 1:3, [:red, :blue], k = S; i + k <= 6])\n\nReturns\n\nindex_vars: a Vector{Any} of names for the index variables, e.g., [:i, gensym(), :k]. These may also be expressions, like :((i, j)) from a call like :(x[(i, j) in S]).\nindices: an iterator over the indices, for example, Containers.NestedIterator\n\nExample\n\nSee container_code for a worked example.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.container_code","page":"JuMP.Containers","title":"Containers.container_code","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.container_code","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.container_code","page":"JuMP.Containers","title":"JuMP.Containers.container_code","text":"container_code(\n    index_vars::Vector{Any},\n    indices::Expr,\n    code,\n    requested_container::Union{Symbol,Expr},\n)\n\nUsed in macros to construct a call to container. This should be used in conjunction with build_ref_sets.\n\nArguments\n\nindex_vars::Vector{Any}: a vector of names for the indices of the container. These may also be expressions, like :((i, j)) from a call like :(x[(i, j) in S]).\nindices::Expr: an expression that evaluates to an iterator of the indices.\ncode: an expression or literal constant for the value to be stored in the container as a function of the named index_vars.\nrequested_container: passed to the third argument of container. For built-in JuMP types, choose one of :Array, :DenseAxisArray, :SparseAxisArray, or :Auto. For a user-defined container, this expression must evaluate to the correct type.\n\nwarning: Warning\nIn most cases, you should esc(code) before passing it to container_code.\n\nExample\n\njulia> macro foo(ref_sets, code)\n           index_vars, indices = Containers.build_ref_sets(error, ref_sets)\n           return Containers.container_code(\n               index_vars,\n               indices,\n               esc(code),\n               :Auto,\n            )\n       end\n@foo (macro with 1 method)\n\njulia> @foo(x[i=1:2, j=[\"A\", \"B\"]], j^i)\n2-dimensional DenseAxisArray{String,2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [\"A\", \"B\"]\nAnd data, a 2×2 Matrix{String}:\n \"A\"   \"B\"\n \"AA\"  \"BB\"\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.AutoContainerType","page":"JuMP.Containers","title":"Containers.AutoContainerType","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.AutoContainerType","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.AutoContainerType","page":"JuMP.Containers","title":"JuMP.Containers.AutoContainerType","text":"AutoContainerType\n\nPass AutoContainerType to container to let the container type be chosen based on the type of the indices using default_container.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.NestedIterator","page":"JuMP.Containers","title":"Containers.NestedIterator","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.NestedIterator","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.NestedIterator","page":"JuMP.Containers","title":"JuMP.Containers.NestedIterator","text":"struct NestedIterator{T}\n    iterators::T # Tuple of functions\n    condition::Function\nend\n\nIterators over the tuples that are produced by a nested for loop.\n\nConstruct a NestedIterator using nested.\n\nExample\n\njulia> iterators = (() -> 1:2, (i,) -> [\"A\", \"B\"]);\n\njulia> condition = (i, j) -> isodd(i) || j == \"B\";\n\njulia> x = Containers.NestedIterator(iterators, condition);\n\njulia> for (i, j) in x\n           println((i, j))\n       end\n(1, \"A\")\n(1, \"B\")\n(2, \"B\")\n\nis the same as\n\njulia> for i in iterators[1]()\n           for j in iterators[2](i)\n               if condition(i, j)\n                   println((i, j))\n               end\n           end\n       end\n(1, \"A\")\n(1, \"B\")\n(2, \"B\")\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.VectorizedProductIterator","page":"JuMP.Containers","title":"Containers.VectorizedProductIterator","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.VectorizedProductIterator","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.VectorizedProductIterator","page":"JuMP.Containers","title":"JuMP.Containers.VectorizedProductIterator","text":"struct VectorizedProductIterator{T}\n    prod::Iterators.ProductIterator{T}\nend\n\nA wrapper type for Iterators.ProuctIterator that discards shape information and returns a Vector.\n\nConstruct a VectorizedProductIterator using vectorized_product.\n\n\n\n\n\n","category":"type"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"EditURL = \"multi_objective_knapsack.jl\"","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Multi-objective-knapsack","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The purpose of this tutorial is to demonstrate how to create and solve a multi-objective linear program. In addition, it demonstrates how to work with solvers which return multiple solutions.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Required-packages","page":"Multi-objective knapsack","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA\nimport Plots\nimport Test  #hide","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl is a package which implements a variety of algorithms for solving multi-objective optimization problems. Because it is a long package name, we import it instead as MOA.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Formulation","page":"Multi-objective knapsack","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The knapsack problem is a classic problem in mixed-integer programming. Given a collection of items i in I, each of which has an associated weight, w_i, and profit, p_i, the knapsack problem determines which profit-maximizing subset of items to pack into a knapsack such that the total weight is less than a capacity c. The mathematical formulation is:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"beginaligned\nmax  sumlimits_i in I p_i x_i \ntextst   sumlimits_i in I w_i x_i le c\n x_i in 0 1  forall i in I\nendaligned","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"where x_i is 1 if we pack item i into the knapsack and 0 otherwise.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"For this tutorial, we extend the single-objective knapsack problem by adding another objective: given a desirability rating, r_i, we wish to maximize the total desirability of the items in our knapsack. Thus, our mathematical formulation is now:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"beginaligned\nmax  sumlimits_i in I p_i x_i \n      sumlimits_i in I r_i x_i \ntextst   sumlimits_i in I w_i x_i le c\n x_i in 0 1  forall i in I\nendaligned","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Data","page":"Multi-objective knapsack","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The data for this example was taken from vOptGeneric, and the original author was @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"profit = [77, 94, 71, 63, 96, 82, 85, 75, 72, 91, 99, 63, 84, 87, 79, 94, 90]\ndesire = [65, 90, 90, 77, 95, 84, 70, 94, 66, 92, 74, 97, 60, 60, 65, 97, 93]\nweight = [80, 87, 68, 72, 66, 77, 99, 85, 70, 93, 98, 72, 100, 89, 67, 86, 91]\ncapacity = 900\nN = length(profit)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Comparing the capacity to the total weight of all the items:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"capacity / sum(weight)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"shows that we can take approximately 64% of the items.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Plotting the items, we see that there are a range of items with different profits and desirability. Some items have a high profit and a high desirability, others have a low profit and a high desirability (and vice versa).","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Plots.scatter(\n    profit,\n    desire;\n    xlabel = \"Profit\",\n    ylabel = \"Desire\",\n    legend = false,\n)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The goal of the bi-objective knapsack problem is to choose a subset which maximizes both objectives.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#JuMP-formulation","page":"Multi-objective knapsack","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Our JuMP formulation is a direct translation of the mathematical formulation:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"model = Model()\n@variable(model, x[1:N], Bin)\n@constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\n@expression(model, profit_expr, sum(profit[i] * x[i] for i in 1:N))\n@expression(model, desire_expr, sum(desire[i] * x[i] for i in 1:N))\n@objective(model, Max, [profit_expr, desire_expr])","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Note how we form a multi-objective program by passing a vector of scalar objective functions.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Solution","page":"Multi-objective knapsack","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"To solve our model, we need an optimizer which supports multi-objective linear programs. One option is to use the MultiObjectiveAlgorithms.jl package.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"set_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_silent(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl supports many different algorithms for solving multiobjective optimization problems. One option is the epsilon-constraint method:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"set_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Let's solve the problem and see the solution","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"There are 9 solutions available. We can also use result_count to see how many solutions are available:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"result_count(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Accessing-multiple-solutions","page":"Multi-objective knapsack","title":"Accessing multiple solutions","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Access the nine different solutions in the model using the result keyword to solution_summary, value, and objective_value:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"solution_summary(model; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"objective_value(model; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Note that because we set a vector of two objective functions, the objective value is a vector with two elements. We can also query the value of each objective separately:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"value(profit_expr; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Visualizing-objective-space","page":"Multi-objective knapsack","title":"Visualizing objective space","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Unlike single-objective optimization problems, multi-objective optimization problems do not have a single optimal solution. Instead, the solutions returned represent possible trade-offs that the decision maker can choose between the two objectives. A common way to visualize this is by plotting the objective values of each of the solutions:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"plot = Plots.scatter(\n    [value(profit_expr; result = i) for i in 1:result_count(model)],\n    [value(desire_expr; result = i) for i in 1:result_count(model)];\n    xlabel = \"Profit\",\n    ylabel = \"Desire\",\n    title = \"Objective space\",\n    label = \"\",\n    xlims = (915, 960),\n)\nfor i in 1:result_count(model)\n    y = objective_value(model; result = i)\n    Plots.annotate!(y[1] - 1, y[2], (i, 10))\nend\nideal_point = objective_bound(model)\nPlots.scatter!([ideal_point[1]], [ideal_point[2]]; label = \"Ideal point\")","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Visualizing the objective space lets the decision maker choose a solution that suits their personal preferences. For example, result #7 is close to the maximum value of profit, but offers significantly higher desirability compared with solutions #8 and #9.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The set of items that are chosen in solution #7 are:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"items_chosen = [i for i in 1:N if value(x[i]; result = 7) > 0.9]","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Next-steps","page":"Multi-objective knapsack","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl implements a number of different algorithms. Try solving the same problem using MOA.Dichotomy(). Does it find the same solution?","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"EditURL = \"https://github.com/chkwon/PATHSolver.jl/blob/v1.6.1/README.md\"","category":"page"},{"location":"packages/PATHSolver/#PATHSolver.jl","page":"chkwon/PATHSolver.jl","title":"PATHSolver.jl","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl is a wrapper for the PATH solver.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"a thin wrapper around the C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"You can solve any complementarity problem using the wrapper around the C API, although you must manually provide the callback functions, including the Jacobian.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The MathOptInterface wrapper is more limited, supporting only linear complementarity problems, but it enables PATHSolver to be used with JuMP.","category":"page"},{"location":"packages/PATHSolver/#Affiliation","page":"chkwon/PATHSolver.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of PATH. However, we are in close contact with the PATH developers, and they have given us permission to re-distribute the PATH binaries for automatic installation.","category":"page"},{"location":"packages/PATHSolver/#License","page":"chkwon/PATHSolver.jl","title":"License","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl is licensed under the MIT License.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The underlying solver, path is closed source and requires a license.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Without a license, the PATH Solver can solve problem instances up to with up to 300 variables and 2000 non-zeros. For larger problems, this web page provides a temporary license that is valid for a year.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"You can either store the license in the PATH_LICENSE_STRING environment variable, or you can use the PATHSolver.c_api_License_SetString function immediately after importing the PATHSolver package:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"import PATHSolver\nPATHSolver.c_api_License_SetString(\"<LICENSE STRING>\")","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"where <LICENSE STRING> is replaced by the current license string.","category":"page"},{"location":"packages/PATHSolver/#Installation","page":"chkwon/PATHSolver.jl","title":"Installation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Install PATHSolver.jl as follows:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"import Pkg\nPkg.add(\"PATHSolver\")","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default, PATHSolver.jl will download a copy of the underlying PATH solver. To use a different version of PATH, see the Manual Installation section below.","category":"page"},{"location":"packages/PATHSolver/#Use-with-JuMP","page":"chkwon/PATHSolver.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"julia> using JuMP, PATHSolver\n\njulia> M = [\n           0  0 -1 -1\n           0  0  1 -2\n           1 -1  2 -2\n           1  2 -2  4\n       ]\n4×4 Array{Int64,2}:\n 0   0  -1  -1\n 0   0   1  -2\n 1  -1   2  -2\n 1   2  -2   4\n\njulia> q = [2, 2, -2, -6]\n4-element Array{Int64,1}:\n  2\n  2\n -2\n -6\n\njulia> model = Model(PATHSolver.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: Path 5.0.00\n\njulia> set_optimizer_attribute(model, \"output\", \"no\")\n\njulia> @variable(model, x[1:4] >= 0)\n4-element Array{VariableRef,1}:\n x[1]\n x[2]\n x[3]\n x[4]\n\njulia> @constraint(model, M * x .+ q ⟂ x)\n[-x[3] - x[4] + 2, x[3] - 2 x[4] + 2, x[1] - x[2] + 2 x[3] - 2 x[4] - 2, x[1] + 2 x[2] - 2 x[3] + 4 x[4] - 6, x[1], x[2], x[3], x[4]] ∈ MOI.Complements(4)\n\njulia> optimize!(model)\nReading options file /var/folders/bg/dzq_hhvx1dxgy6gb5510pxj80000gn/T/tmpiSsCRO\nRead of options file complete.\n\nPath 5.0.00 (Mon Aug 19 10:57:18 2019)\nWritten by Todd Munson, Steven Dirkse, Youngdae Kim, and Michael Ferris\n\njulia> value.(x)\n4-element Array{Float64,1}:\n 2.8\n 0.0\n 0.7999999999999998\n 1.2\n\njulia> termination_status(model)\nLOCALLY_SOLVED::TerminationStatusCode = 4","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Note that options are set using JuMP.set_optimizer_attribute.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The list of options supported by PATH can be found here: https://pages.cs.wisc.edu/~ferris/path/options.pdf","category":"page"},{"location":"packages/PATHSolver/#MathOptInterface-API","page":"chkwon/PATHSolver.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The Path 5.0.03 optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.Reals","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VectorOfVariables in MOI.Complements\nMOI.VectorAffineFunction{Float64} in MOI.Complements\nMOI.VectorQuadraticFunction{Float64} in MOI.Complements\nMOI.VectorNonlinearFunction in MOI.Complements","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.Name()","category":"page"},{"location":"packages/PATHSolver/#Use-with-the-C-API","page":"chkwon/PATHSolver.jl","title":"Use with the C API","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl wraps the PATH C API using PATHSolver.c_api_XXX for the C method XXX. However, using the C API directly from Julia can be challenging, particularly with respect to avoiding issues with Julia's garbage collector.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Instead, we recommend that you use the PATHSolver.solve_mcp function, which wrappers the C API into a single call. See the docstring of PATHSolver.solve_mcp for a detailed description of the arguments.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Here is the same example using PATHSolver.solve_mcp. Note that you must manually construct the sparse Jacobian callback.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"julia> import PATHSolver\n\njulia> M = [\n           0  0 -1 -1\n           0  0  1 -2\n           1 -1  2 -2\n           1  2 -2  4\n       ]\n4×4 Matrix{Int64}:\n 0   0  -1  -1\n 0   0   1  -2\n 1  -1   2  -2\n 1   2  -2   4\n\njulia> q = [2, 2, -2, -6]\n4-element Vector{Int64}:\n  2\n  2\n -2\n -6\n\njulia> function F(n::Cint, x::Vector{Cdouble}, f::Vector{Cdouble})\n           @assert n == length(x) == length(f)\n           f .= M * x .+ q\n           return Cint(0)\n       end\nF (generic function with 1 method)\n\njulia> function J(\n           n::Cint,\n           nnz::Cint,\n           x::Vector{Cdouble},\n           col::Vector{Cint},\n           len::Vector{Cint},\n           row::Vector{Cint},\n           data::Vector{Cdouble},\n       )\n           @assert n == length(x) == length(col) == length(len) == 4\n           @assert nnz == length(row) == length(data)\n           i = 1\n           for c in 1:n\n               col[c], len[c] = i, 0\n               for r in 1:n\n                   if !iszero(M[r, c])\n                       row[i], data[i] = r, M[r, c]\n                       len[c] += 1\n                       i += 1\n                   end\n               end\n           end\n           return Cint(0)\n       end\nJ (generic function with 1 method)\n\njulia> status, z, info = PATHSolver.solve_mcp(\n           F,\n           J,\n           fill(0.0, 4),  # Lower bounds\n           fill(Inf, 4),  # Upper bounds\n           fill(0.0, 4);  # Starting point\n           nnz = 12,      # Number of nonzeros in the Jacobian\n           output = \"yes\",\n       )\nReading options file /var/folders/bg/dzq_hhvx1dxgy6gb5510pxj80000gn/T/jl_iftYBS\n > output yes\nRead of options file complete.\n\nPath 5.0.03 (Fri Jun 26 09:58:07 2020)\nWritten by Todd Munson, Steven Dirkse, Youngdae Kim, and Michael Ferris\n\nCrash Log\nmajor  func  diff  size  residual    step       prox   (label)\n    0     0             1.2649e+01             0.0e+00 (f[    4])\n    1     2     4     2 1.0535e+01  8.0e-01    0.0e+00 (f[    1])\n    2     3     2     4 8.4815e-01  1.0e+00    0.0e+00 (f[    4])\n    3     4     0     3 4.4409e-16  1.0e+00    0.0e+00 (f[    3])\npn_search terminated: no basis change.\n\nMajor Iteration Log\nmajor minor  func  grad  residual    step  type prox    inorm  (label)\n    0     0     5     4 4.4409e-16           I 0.0e+00 4.4e-16 (f[    3])\n\nMajor Iterations. . . . 0\nMinor Iterations. . . . 0\nRestarts. . . . . . . . 0\nCrash Iterations. . . . 3\nGradient Steps. . . . . 0\nFunction Evaluations. . 5\nGradient Evaluations. . 4\nBasis Time. . . . . . . 0.000016\nTotal Time. . . . . . . 0.044383\nResidual. . . . . . . . 4.440892e-16\n(PATHSolver.MCP_Solved, [2.8, 0.0, 0.8, 1.2], PATHSolver.Information(4.4408920985006247e-16, 0.0, 0.0, 0.044383, 1.6e-5, 0.0, 0, 0, 3, 5, 4, 0, 0, 0, 0, false, false, false, true, false, false, false))\n\njulia> status\nMCP_Solved::MCP_Termination = 1\n\njulia> z\n4-element Vector{Float64}:\n 2.8\n 0.0\n 0.8\n 1.2","category":"page"},{"location":"packages/PATHSolver/#Thread-safety","page":"chkwon/PATHSolver.jl","title":"Thread safety","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATH is not thread-safe and there are no known work-arounds. Do not run it in parallel using Threads.@threads. See issue #62 for more details.","category":"page"},{"location":"packages/PATHSolver/#Factorization-methods","page":"chkwon/PATHSolver.jl","title":"Factorization methods","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default, PATHSolver.jl will download the LUSOL shared library. To use LUSOL, set the following options:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"model = Model(PATHSolver.Optimizer)\nset_optimizer_attribute(model, \"factorization_method\", \"blu_lusol\")\nset_optimizer_attribute(model, \"factorization_library_name\", PATHSolver.LUSOL_LIBRARY_PATH)","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"To use factorization_method umfpack you will need the umfpack shared library that is available directly from the developers of that code for academic use.","category":"page"},{"location":"packages/PATHSolver/#Manual-installation","page":"chkwon/PATHSolver.jl","title":"Manual installation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default PATHSolver.jl will download a copy of the libpath library. If you already have one installed and want to use that, set the PATH_JL_LOCATION environment variable to point to the libpath50.xx library.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"EditURL = \"multi_objective_examples.jl\"","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Simple-multi-objective-examples","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial contains a number of examples of multi-objective programs from the literature.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Required-packages","page":"Simple multi-objective examples","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-linear-problem","page":"Simple multi-objective examples","title":"Bi-objective linear problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Example 6.3 (from Steuer, 1985), page 154 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"model = Model()\nset_silent(model)\n@variable(model, x1 >= 0)\n@variable(model, 0 <= x2 <= 3)\n@objective(model, Min, [3x1 + x2, -x1 - 2x2])\n@constraint(model, 3x1 - x2 <= 6)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.Lexicographic())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    println(\"x = \", value.([x1, x2]; result = i))\nend","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-linear-assignment-problem","page":"Simple multi-objective examples","title":"Bi-objective linear assignment problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Example 9.38 (from Ulungu and Teghem, 1994), page 255 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"C1 = [5 1 4 7; 6 2 2 6; 2 8 4 4; 3 5 7 1]\nC2 = [3 6 4 2; 1 3 8 3; 5 2 2 3; 4 2 3 5]\nn = size(C2, 1)\nmodel = Model()\nset_silent(model)\n@variable(model, x[1:n, 1:n], Bin)\n@objective(model, Min, [sum(C1 .* x), sum(C2 .* x)])\n@constraint(model, [i = 1:n], sum(x[i, :]) == 1)\n@constraint(model, [j = 1:n], sum(x[:, j]) == 1)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    println(\"x = \", round.(Int, value.(x; result = i)))\nend","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-shortest-path-problem","page":"Simple multi-objective examples","title":"Bi-objective shortest path problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Exercise 9.5 page 269 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"M = 50\nC1 = [\n    M 4 5 M M M\n    M M 2 1 2 7\n    M M M 5 2 M\n    M M 5 M M 3\n    M M M M M 4\n    M M M M M M\n]\nC2 = [\n    M 3 1 M M M\n    M M 1 4 2 2\n    M M M 1 7 M\n    M M 1 M M 2\n    M M M M M 2\n    M M M M M M\n]\nn = size(C2, 1)\nmodel = Model()\nset_silent(model)\n@variable(model, x[1:n, 1:n], Bin)\n@objective(model, Min, [sum(C1 .* x), sum(C2 .* x)])\n@constraint(model, sum(x[1, :]) == 1)\n@constraint(model, sum(x[:, n]) == 1)\n@constraint(model, [i = 2:n-1], sum(x[i, :]) - sum(x[:, i]) == 0)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    X = round.(Int, value.(x; result = i))\n    print(\"Path:\")\n    for ind in findall(val -> val ≈ 1, X)\n        i, j = ind.I\n        print(\" $i->$j\")\n    end\n    println()\nend","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"EditURL = \"network_flows.jl\"","category":"page"},{"location":"tutorials/linear/network_flows/#Network-flow-problems","page":"Network flow problems","title":"Network flow problems","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"In graph theory, a flow network (also known as a transportation network) is a directed graph where each edge has a capacity and each edge receives a flow. The amount of flow on an edge cannot exceed the capacity of the edge.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Often in operations research, a directed graph is called a network, the vertices are called nodes and the edges are called arcs.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"A flow must satisfy the restriction that the amount of flow into a node equals the amount of flow out of it,  unless it is a source, which has only outgoing flow, or sink, which has only incoming flow.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"A network can be used to model traffic in a computer network, circulation with demands, fluids in pipes,  currents in an electrical circuit, or anything similar in which something travels through a network of nodes.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/network_flows/#The-shortest-path-problem","page":"Network flow problems","title":"The shortest path problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Suppose that each arc (i j) of a graph is assigned a scalar cost a_ij, and suppose that we define the cost of a forward path to be the sum of the costs of its arcs.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Given a pair of nodes, the shortest path problem is to find a forward path that connects these nodes and has minimum cost.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmin  sum_forall e(ij) in E a_ij times x_ij \nst  sum_j x_ij - sum_k x_ki = b_i  forall i\n x_e in 01  forall e in E\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"where b_i is 1 if i is the starting node, -1 if i is the ending node, and 0 otherwise.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    0 100 30 0 0\n    0 0 20 0 0\n    0 0 0 10 60\n    0 15 0 0 50\n    0 0 0 0 0\n]\nn = size(G)[1]\nb = [1, -1, 0, 0, 0]\nshortest_path = Model(HiGHS.Optimizer)\nset_silent(shortest_path)\n@variable(shortest_path, x[1:n, 1:n], Bin)\n# Arcs with zero cost are not a part of the path as they do no exist\n@constraint(shortest_path, [i = 1:n, j = 1:n; G[i, j] == 0], x[i, j] == 0)\n# Flow conservation constraint\n@constraint(shortest_path, [i = 1:n], sum(x[i, :]) - sum(x[:, i]) == b[i],)\n@objective(shortest_path, Min, sum(G .* x))\noptimize!(shortest_path)\nobjective_value(shortest_path)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(x)","category":"page"},{"location":"tutorials/linear/network_flows/#The-assignment-problem","page":"Network flow problems","title":"The assignment problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Suppose that there are n persons and n objects that we have to match on a one-to-one basis. There is a benefit or value a_ij for matching person i with object j, and we want to assign persons to objects so as to maximize the total benefit.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"There is also a restriction that person i can be assigned to object j only if (i j) belongs to a given set of pairs A.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Mathematically, we want to find a set of person-object pairs (1 j_1) (n j_n) from A such that the objects j_1j_n are all distinct, and the total benefit sum_i=1^y a_ij_i is maximized.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmax  sum_(ij) in A a_ij times y_ij \nst  sum_j(ij) in A y_ij = 1  forall i = 12n \n sum_i(ij) in A y_ij = 1  forall j = 12n \n y_ij in 01  forall (ij) in 12k\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    6 4 5 0\n    0 3 6 0\n    5 0 4 3\n    7 5 5 5\n]\nn = size(G)[1]\nassignment = Model(HiGHS.Optimizer)\nset_silent(assignment)\n@variable(assignment, y[1:n, 1:n], Bin)\n# One person can only be assigned to one object\n@constraint(assignment, [i = 1:n], sum(y[:, i]) == 1)\n# One object can only be assigned to one person\n@constraint(assignment, [j = 1:n], sum(y[j, :]) == 1)\n@objective(assignment, Max, sum(G .* y))\noptimize!(assignment)\nobjective_value(assignment)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(y)","category":"page"},{"location":"tutorials/linear/network_flows/#The-max-flow-problem","page":"Network flow problems","title":"The max-flow problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"In the max-flow problem, we have a graph with two special nodes: the source, denoted by s, and the sink, denoted by t.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"The objective is to move as much flow as possible from s into t while observing the capacity constraints.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmax  sum_v(sv) in E f(sv) \nst  sum_u(uv) in E f(uv)  = sum_w(vw) in E f(vw)  forall v in V - st \n f(uv) leq c(uv)  forall (uv) in E \n f(uv) geq 0  forall (uv) in E\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    0 3 2 2 0 0 0 0\n    0 0 0 0 5 1 0 0\n    0 0 0 0 1 3 1 0\n    0 0 0 0 0 1 0 0\n    0 0 0 0 0 0 0 4\n    0 0 0 0 0 0 0 2\n    0 0 0 0 0 0 0 4\n    0 0 0 0 0 0 0 0\n]\nn = size(G)[1]\nmax_flow = Model(HiGHS.Optimizer)\n@variable(max_flow, f[1:n, 1:n] >= 0)\n# Capacity constraints\n@constraint(max_flow, [i = 1:n, j = 1:n], f[i, j] <= G[i, j])\n# Flow conservation constraints\n@constraint(max_flow, [i = 1:n; i != 1 && i != 8], sum(f[i, :]) == sum(f[:, i]))\n@objective(max_flow, Max, sum(f[1, :]))\noptimize!(max_flow)\nobjective_value(max_flow)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(f)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"EditURL = \"optimal_power_flow.jl\"","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Optimal-power-flow","page":"Optimal power flow","title":"Optimal power flow","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial was originally contributed by James Foster (@jd-foster).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial formulates and solves an alternating current optimal power flow (AC-OPF) problem, a much-studied nonlinear problem from the field of electrical engineering.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Once we've formulated and solved the nonlinear problem, we will turn our focus to obtaining a good estimate of the objective value at the global optimum through the use of semidefinite programming.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"One main purpose of this tutorial is to highlight JuMP's ability to directly formulate problems involving complex-valued decision variables and complex matrix cones such as the  HermitianPSDCone object.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For another example of modeling with complex decision variables, see the Quantum state discrimination tutorial, and see the Complex number support section of the manual for more details.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial takes a matrix-oriented approach focused on network nodes that simplifies the construction of semidefinite programs. Another approach is to formulate the problem focusing on network lines where it is easier to work with flow constraints. A general approach is provided by the Julia/JuMP package PowerModels.jl, an open-source framework to a broad range of power flow model formulations along with utilities for working with detailed network data.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Required-packages","page":"Optimal power flow","title":"Required packages","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"using JuMP\nimport Clarabel\nimport DataFrames\nimport Ipopt\nimport LinearAlgebra\nimport SparseArrays","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Initial-formulation","page":"Optimal power flow","title":"Initial formulation","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Optimal power flow problems for electrical transmission typically pose the following question: what is the most cost-effective operation of electricity generators while meeting constraints on the safe limits of network components?","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We'll use the 9-node network test case case9mod to explore this problem. The graph of the network, shown here, has three nodes (or buses) each for the different purposes of generation G (nodes 1, 2, and 3), trans-shipment (nodes 4, 6, and 8), and demand D (nodes 5, 7, and 9).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(Image: Nine Nodes)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This example is a modified version of the MATPOWER test case case9 (archive) created by Bukhsh et al. (2013) for their test case archive of optimal power flow problems with local optima. This test case is also extensively evaluated in Krasko and Rebennack (2017).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Here bus and network node are taken as analogous terms, as are branch and transmission line.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For future reference, let's name the number of nodes in the network:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"N = 9;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The network data can be summarised using a small number of arrays. Using the sparsevec function from the SparseArrays standard library package, we can give the indices and values of the non-zero data points:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"# Real generation: lower (`lb`) and upper (`ub`) bounds\nP_Gen_lb = SparseArrays.sparsevec([1, 2, 3], [10, 10, 10], N)\nP_Gen_ub = SparseArrays.sparsevec([1, 2, 3], [250, 300, 270], N)\n# Reactive generation: lower (`lb`) and upper (`ub`) bounds\nQ_Gen_lb = SparseArrays.sparsevec([1, 2, 3], [-5, -5, -5], N)\nQ_Gen_ub = SparseArrays.sparsevec([1, 2, 3], [300, 300, 300], N)\n# Power demand levels (real, reactive, and complex form)\nP_Demand = SparseArrays.sparsevec([5, 7, 9], [54, 60, 75], N)\nQ_Demand = SparseArrays.sparsevec([5, 7, 9], [18, 21, 30], N)\nS_Demand = P_Demand + im * Q_Demand","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The key decision variables are the real power injections P^G and reactive power injections Q^Gover the allowed range of the generators. All other buses must restrict their generation variables to 0. On the other hand, these non-generator nodes have a fixed  real and reactive power demand, denoted P^D and Q^D respectively (these are fixed at 0 in the case of trans-shipment and generator nodes).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The cost of operating each generator is modeled as a quadratic function of its real power output; in our specific test case, the objective function to minimize is:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginalign\n    min          011  (P^G_1)^2 +   5 P^G_1 + 150  \n               +  0085  (P^G_2)^2 + 12 P^G_2 + 600  \n               + 01225   (P^G_3)^2 +     P^G_3 + 335 \nendalign","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's create an initial JuMP model with some of this data:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, P_Gen_lb[i] <= P_G[i in 1:N] <= P_Gen_ub[i])\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Even before solving an optimization problem, we can estimate a lower bound on the best objective value by substituting the lower bound on each generator's real power range (all 10, as it turns out in this case):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"basic_lower_bound = value(lower_bound, objective_function(model));\nprintln(\"Objective value (basic lower bound) : $basic_lower_bound\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"to see that we can do no better than an objective cost of 1188.75.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(Direct substitution works because a quadratic function of a single variable x with positive coefficients is strictly increasing for all x geq 0.)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"In fact, we can get a quick but even better estimate from the direct observation that the real power generated must meet or exceed the real power demand:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, sum(P_G) >= sum(P_Demand))\noptimize!(model)\nbetter_lower_bound = round(objective_value(model); digits = 2)\nprintln(\"Objective value (better lower bound): $better_lower_bound\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"However, there are additional power flow constraints that must be satisfied.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Power must flow from one or more generation nodes through the transmission lines and end up at a demand node. The state variables of our steady-state alternating current (AC) electrical network are complex-valued voltage variables V_1 ldots V_N. Voltages capture both a magnitude and phase of the node's electrical state in relation to the rest of the system. An AC power system also extends the notion of resistance in wires found in a direct current (DC) circuit to a  complex quantity, known as the impedance, of each transmission line. The reciprocal of impedance is known as admittance. Together, these complex quantities are used to express a complex version of Ohm's law: current flow through a line is proportional to the difference in voltages on each end of the line, multiplied by the admittance.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Network-data","page":"Optimal power flow","title":"Network data","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's assemble the data we need for writing the complex power flow constraints. The data for the problem consists of a list of the real and imaginary parts of the line impedance. We obtain the following data table from the branch data section of the case9mod MATPOWER format file:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"branch_data = DataFrames.DataFrame([\n    (1, 4, 0.0, 0.0576, 0.0),\n    (4, 5, 0.017, 0.092, 0.158),\n    (6, 5, 0.039, 0.17, 0.358),\n    (3, 6, 0.0, 0.0586, 0.0),\n    (6, 7, 0.0119, 0.1008, 0.209),\n    (8, 7, 0.0085, 0.072, 0.149),\n    (2, 8, 0.0, 0.0625, 0.0),\n    (8, 9, 0.032, 0.161, 0.306),\n    (4, 9, 0.01, 0.085, 0.176),\n]);\nDataFrames.rename!(branch_data, [:F_BUS, :T_BUS, :BR_R, :BR_X, :BR_Bc])","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The first two columns describe the network, supplying the from and to connection points of the lines. The last three columns give the branch resistance, branch reactance and line-charging susceptance.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We will also need to reference the base_MVA number (used for re-scaling):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"base_MVA = 100;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and the number of lines:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"M = size(branch_data, 1)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"From the first two columns of the branch data table, we can create a sparse incidence matrix that simplifies handling of the network layout:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"A =\n    SparseArrays.sparse(branch_data.F_BUS, 1:M, 1, N, M) +\n    SparseArrays.sparse(branch_data.T_BUS, 1:M, -1, N, M)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We form the network impedance vector from the next two columns","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"z = (branch_data.BR_R .+ im * branch_data.BR_X) / base_MVA;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and calculate it's corresponding bus admittance matrix as","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y_0 = A * SparseArrays.spdiagm(1 ./ z) * A';\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"while the last column gives the branch line-charging susceptance","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"y_sh = 1 / 2 * (im * branch_data.BR_Bc) * base_MVA;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and leads to the shunt admittance matrix","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y_sh = SparseArrays.spdiagm(\n    LinearAlgebra.diag(A * SparseArrays.spdiagm(y_sh) * A'),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(The construction of the shunt admittance matrix Y_sh looks somewhat more complicated than Y_0 because we only want to add the diagonal elements in the calculation; the line-charging is used only in the nodal voltage terms and not the line voltage terms.)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The full bus admittance matrix Y is then defined as","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y = Y_0 + Y_sh;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#JuMP-model","page":"Optimal power flow","title":"JuMP model","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Now we're ready to write the complex power flow constraints we need to more accurately model the electricity system.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We'll introduce a number of constraints that model both the physics and operational requirements.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's start by initializing a new model:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Then we'll create the nodal power generation variables:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@variable(\n    model,\n    S_G[i in 1:N] in ComplexPlane(),\n    lower_bound = P_Gen_lb[i] + Q_Gen_lb[i] * im,\n    upper_bound = P_Gen_ub[i] + Q_Gen_ub[i] * im,\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We need complex nodal voltages (the system state variables):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@variable(model, V[1:N] in ComplexPlane(), start = 1.0 + 0.0im)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and operational constraints for maintaining voltage magnitude levels:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, [i in 1:N], 0.9^2 <= real(V[i])^2 + imag(V[i])^2 <= 1.1^2)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We also need to fix an origin or reference angle from which all other complex voltage angles (arguments) are determined. Here we will use node 1 as the nominated reference bus. Fixing the imaginary component of a reference bus to zero sets its complex voltage angle to 0, while constraining the real part to be non-negative disallows equivalent solutions that are just a reflection by 180 degrees:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, imag(V[1]) == 0);\n@constraint(model, real(V[1]) >= 0);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The power flow equations express a conservation of energy (power) principle, where power generated less the power consumed must balance the power exchanged with the network:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, S_G - S_Demand .== V .* conj(Y * V))","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"As above, the objective function is a quadratic cost of real power:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"P_G = real(S_G)\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We're finally ready to solve our nonlinear AC-OPF problem:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"objval_solution = round(objective_value(model); digits = 2)\nprintln(\"Objective value (feasible solution) : $(objval_solution)\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The solution's power generation (in rectangular form) and complex voltage values (in polar form using degrees) are:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"DataFrames.DataFrame(;\n    Bus = 1:N,\n    ComplexPowerGen = round.(value.(S_G); digits = 2),\n    VoltageMagnitude = round.(abs.(value.(V)); digits = 2),\n    VoltageAngle_Deg = round.(rad2deg.(angle.(value.(V))); digits = 2),\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Relaxations-and-better-objective-bounds","page":"Optimal power flow","title":"Relaxations and better objective bounds","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The Ipopt solver uses an interior-point algorithm. It has local optimality guarantees, but is unable to certify whether the solution is globally optimal. The solution we found is indeed globally optimal. The work to verify this has been done in Bukhsh et al. (2013) and Krasko and Rebennack (2017), and different solvers (such as Gurobi, SCIP and GLOMIQO) are also able to verify this.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The techniques of convex relaxations can also be used to improve on our current best lower bound:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"better_lower_bound","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"To this end, observe that the nonlinear constraints in the AC-OPF formulation are quadratic equalities for power flow along with quadratic voltage inequalities.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's linearize these constraints by first making the substitution W = V V^*, where:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W = V V^* quad iff  quad W_ii =  V_i ^2 quad W_ik = V_i  overlineV_k quad forall i  k in  1 ldots N ","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and where V^* is the conjugate transpose of V.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"On the face of it, this turns a quadratic voltage bound constraint like:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"v_L leq V_i ^2  leq v_U quad i  in  1 ldots N ","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"for some real v_L and v_U into a simple two-sided bound:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"v_L leq W_ii  leq v_U","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"while each quadratic expression for the nodal power term:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"    S^Node_i = V_i overline(YV)_i","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"becomes the linear combination:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"    S^Node_i = (E_ii Y^T) bullet W","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Here A bullet B = operatornametr(A^* B) is the Frobenius inner product of two complex matrices, while E_kn denotes the matrix unit with a single nonzero entry of 1 in row k and column n.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"E(k, n) = SparseArrays.sparse([k], [n], 1, N, N);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Of course, we've shifted the nonlinearity into the equality constraint W = V V^*: it is this constraint we will now relax using a semidefinite programming approach.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We will make use of complex voltages and relax W = V V^* to;","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W succeq V V^*","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"where the relation succeq is the ordering in the Hermitian positive semidefinite cone.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The above constraint is equivalent to:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginbmatrix 1  V^*  V  W  endbmatrix succeq 0","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"by the theory of the Schur complement. This matrix inequality implies a number of second-order cone constraints by taking certain 2 times 2 minors of the matrix for each i  in  1 ldots N :","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginbmatrix 1  V_i^*  V_i  W_ii  endbmatrix succeq 0","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"which is equivalent to the real second-order cone inequality:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"operatornamereal(W_ii) geq operatornamereal(V_i)^2 + operatornameimag(V_i)^2","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We include these implied constraints as well for demonstration purposes.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Putting it all together we get the following semidefinite relaxation of the AC-OPF problem:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Clarabel.Optimizer)\nset_attribute(model, \"tol_gap_rel\", 1e-3)\nset_attribute(model, \"tol_feas\", 1e-3)\nset_attribute(model, \"tol_ktratio\", 1e-3)\n@variable(\n    model,\n    S_G[i in 1:N] in ComplexPlane(),\n    lower_bound = P_Gen_lb[i] + Q_Gen_lb[i] * im,\n    upper_bound = P_Gen_ub[i] + Q_Gen_ub[i] * im,\n)\n@variable(model, W[1:N, 1:N] in HermitianPSDCone())\n@variable(model, V[1:N] in ComplexPlane(), start = 1.0 + 0.0im)\n@constraint(model, [i in 1:N], 0.9^2 <= real(W[i, i]) <= 1.1^2)\n@constraint(model, real(V[1]) >= 0)\n@constraint(model, imag(V[1]) == 0)\n@constraint(model, 0.9 <= real(V[1]) <= 1.1)\n@constraint(model, LinearAlgebra.Hermitian([1 V'; V W]) in HermitianPSDCone())\n# 2 x 2 minor inequalities:\n@constraint(\n    model,\n    [i in 1:N],\n    [0.5, real(W[i, i]), real(V[i]), imag(V[i])] in RotatedSecondOrderCone()\n)\n@constraint(\n    model,\n    [i in 1:N],\n    S_G[i] - S_Demand[i] == LinearAlgebra.tr((conj(Y) * E(i, i)) * W),\n)\nP_G = real(S_G)\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n)\noptimize!(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"sdp_relaxation_lower_bound = round(objective_value(model); digits = 2)\nprintln(\n    \"Objective value (W & V relax. lower bound): $sdp_relaxation_lower_bound\",\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We can more easily see solution values by rounding out noisy data:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W_1 = SparseArrays.sparse(round.(value.(W); digits = 2))","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and recover an approximation to the voltage variables as:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"DataFrames.DataFrame(;\n    Bus = 1:N,\n    Magnitude = round.(abs.(value.(V)); digits = 2),\n    AngleDeg = round.(rad2deg.(angle.(value.(V))); digits = 2),\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For further information on exploiting sparsity see Jabr (2012).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This relaxation has the advantage that we can work directly with complex voltages to extend the formulation, strengthen the relaxation and gain additional approximate information about the voltage variables.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#References-and-further-resources","page":"Optimal power flow","title":"References and further resources","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Bukhsh, W. A., Grothey, A., McKinnon, K. I., & Trodden, P. A. Local solutions of the optimal power flow problem. IEEE Transactions on Power Systems, 28(4), 4780-4788 (2013).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Jabr, R. A. Exploiting sparsity in SDP relaxations of the OPF problem. IEEE Transactions on Power Systems, 27(2), 1138-1139 (2011).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Krasko, V., & S. Rebennack. Chapter 15: Global Optimization: Optimal Power Flow Problem. In Advances and Trends in Optimization with Engineering Applications, 187—205. MOS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics, 2017.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Test case case9mod: from the Test Case Archive of Optimal Power Flow (OPF) Problems with Local Optima.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"R. D. Zimmerman, C. E. Murillo-Sanchez, & R. J. Thomas, MATPOWER: Steady-State Operations, Planning and Analysis Tools for Power Systems Research and Education, IEEE Transactions on Power Systems, 26(1), 12-19 (2011).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"MATPOWER data format: MATPOWER manual; see especially Appendix B \"Data File Format\" and Table B-3.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/list_of_bridges.md\"","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#List-of-bridges","page":"List of bridges","title":"List of bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"This section describes the Bridges.AbstractBridges that are implemented in MathOptInterface.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#constraint_bridges_ref","page":"List of bridges","title":"Constraint bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Constraint.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Constraint.GreaterToIntervalBridge\nBridges.Constraint.LessToIntervalBridge\nBridges.Constraint.GreaterToLessBridge\nBridges.Constraint.LessToGreaterBridge\nBridges.Constraint.NonnegToNonposBridge\nBridges.Constraint.NonposToNonnegBridge\nBridges.Constraint.VectorizeBridge\nBridges.Constraint.ScalarizeBridge\nBridges.Constraint.ScalarSlackBridge\nBridges.Constraint.VectorSlackBridge\nBridges.Constraint.ScalarFunctionizeBridge\nBridges.Constraint.VectorFunctionizeBridge\nBridges.Constraint.ToScalarQuadraticBridge\nBridges.Constraint.ToVectorQuadraticBridge\nBridges.Constraint.ToScalarNonlinearBridge\nBridges.Constraint.FunctionConversionBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitHyperRectangleBridge\nBridges.Constraint.SplitIntervalBridge\nBridges.Constraint.SOCtoRSOCBridge\nBridges.Constraint.RSOCtoSOCBridge\nBridges.Constraint.SOCtoNonConvexQuadBridge\nBridges.Constraint.RSOCtoNonConvexQuadBridge\nBridges.Constraint.QuadtoSOCBridge\nBridges.Constraint.SOCtoPSDBridge\nBridges.Constraint.RSOCtoPSDBridge\nBridges.Constraint.NormInfinityBridge\nBridges.Constraint.NormOneBridge\nBridges.Constraint.NormToPowerBridge\nBridges.Constraint.NormOneConeToNormConeBridge\nBridges.Constraint.SecondOrderConeToNormConeBridge\nBridges.Constraint.NormInfinityConeToNormConeBridge\nBridges.Constraint.GeoMeantoRelEntrBridge\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.GeoMeanBridge\nBridges.Constraint.RelativeEntropyBridge\nBridges.Constraint.NormSpectralBridge\nBridges.Constraint.NormNuclearBridge\nBridges.Constraint.SquareBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.SetDotScalingBridge\nBridges.Constraint.SetDotInverseScalingBridge\nBridges.Constraint.RootDetBridge\nBridges.Constraint.LogDetBridge\nBridges.Constraint.IndicatorActiveOnFalseBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.IndicatorSOS1Bridge\nBridges.Constraint.SemiToBinaryBridge\nBridges.Constraint.ZeroOneBridge\nBridges.Constraint.IntegerToZeroOneBridge\nBridges.Constraint.NumberConversionBridge\nBridges.Constraint.AllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.BinPackingToMILPBridge\nBridges.Constraint.CircuitToMILPBridge\nBridges.Constraint.CountAtLeastToCountBelongsBridge\nBridges.Constraint.CountBelongsToMILPBridge\nBridges.Constraint.CountDistinctToMILPBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.CountGreaterThanToMILPBridge\nBridges.Constraint.TableToMILPBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GreaterToIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GreaterToIntervalBridge","text":"GreaterToIntervalBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nGreaterToIntervalBridge implements the following reformulations:\n\nf(x) ge l into f(x) in l infty)\n\nSource node\n\nGreaterToIntervalBridge supports:\n\nF in MOI.GreaterThan{T}\n\nTarget nodes\n\nGreaterToIntervalBridge creates:\n\nF in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LessToIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LessToIntervalBridge","text":"LessToIntervalBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nLessToIntervalBridge implements the following reformulations:\n\nf(x) le u into f(x) in (-infty u\n\nSource node\n\nLessToIntervalBridge supports:\n\nF in MOI.LessThan{T}\n\nTarget nodes\n\nLessToIntervalBridge creates:\n\nF in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GreaterToLessBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GreaterToLessBridge","text":"GreaterToLessBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nGreaterToLessBridge implements the following reformulation:\n\nf(x) ge l into -f(x) le -l\n\nSource node\n\nGreaterToLessBridge supports:\n\nG in MOI.GreaterThan{T}\n\nTarget nodes\n\nGreaterToLessBridge creates:\n\nF in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LessToGreaterBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LessToGreaterBridge","text":"LessToGreaterBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nLessToGreaterBridge implements the following reformulation:\n\nf(x) le u into -f(x) ge -u\n\nSource node\n\nLessToGreaterBridge supports:\n\nG in MOI.LessThan{T}\n\nTarget nodes\n\nLessToGreaterBridge creates:\n\nF in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NonnegToNonposBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NonnegToNonposBridge","text":"NonnegToNonposBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNonnegToNonposBridge implements the following reformulation:\n\nf(x) in mathbbR_+ into -f(x) in mathbbR_-\n\nSource node\n\nNonnegToNonposBridge supports:\n\nG in MOI.Nonnegatives\n\nTarget nodes\n\nNonnegToNonposBridge creates:\n\nF in MOI.Nonpositives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NonposToNonnegBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NonposToNonnegBridge","text":"NonposToNonnegBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNonposToNonnegBridge implements the following reformulation:\n\nf(x) in mathbbR_- into -f(x) in mathbbR_+\n\nSource node\n\nNonposToNonnegBridge supports:\n\nG in MOI.Nonpositives\n\nTarget nodes\n\nNonposToNonnegBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorizeBridge","text":"VectorizeBridge{T,F,S,G} <: Bridges.Constraint.AbstractBridge\n\nVectorizeBridge implements the following reformulations:\n\ng(x) ge a into g(x) - a in mathbbR_+\ng(x) le a into g(x) - a in mathbbR_-\ng(x) == a into g(x) - a in 0\n\nwhere T is the coefficient type of g(x) - a.\n\nSource node\n\nVectorizeBridge supports:\n\nG in MOI.GreaterThan{T}\nG in MOI.LessThan{T}\nG in MOI.EqualTo{T}\n\nTarget nodes\n\nVectorizeBridge creates:\n\nF in S, where S is one of MOI.Nonnegatives, MOI.Nonpositives, MOI.Zeros depending on the type of the input set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarizeBridge","text":"ScalarizeBridge{T,F,S}\n\nScalarizeBridge implements the following reformulations:\n\nf(x) - a in mathbbR_+ into f_i(x) ge a_i for all i\nf(x) - a in mathbbR_- into f_i(x) le a_i for all i\nf(x) - a in 0 into f_i(x) == a_i for all i\n\nSource node\n\nScalarizeBridge supports:\n\nG in MOI.Nonnegatives{T}\nG in MOI.Nonpositives{T}\nG in MOI.Zeros{T}\n\nTarget nodes\n\nScalarizeBridge creates:\n\nF in S, where S is one of MOI.GreaterThan{T}, MOI.LessThan{T}, and MOI.EqualTo{T}, depending on the type of the input set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarSlackBridge","text":"ScalarSlackBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nScalarSlackBridge implements the following reformulation:\n\nf(x) in S into f(x) - y == 0 and y in S\n\nSource node\n\nScalarSlackBridge supports:\n\nG in S, where G is not MOI.VariableIndex and S is not MOI.EqualTo\n\nTarget nodes\n\nScalarSlackBridge creates:\n\nF in MOI.EqualTo{T}\nMOI.VariableIndex in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorSlackBridge","text":"VectorSlackBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nVectorSlackBridge implements the following reformulation:\n\nf(x) in S into f(x) - y in 0 and y in S\n\nSource node\n\nVectorSlackBridge supports:\n\nG in S, where G is not MOI.VectorOfVariables and S is not MOI.Zeros\n\nTarget nodes\n\nVectorSlackBridge creates:\n\nF in MOI.Zeros\nMOI.VectorOfVariables in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarFunctionizeBridge","text":"ScalarFunctionizeBridge{T,S} = FunctionConversionBridge{T,MOI.ScalarAffineFunction{T},MOI.VariableIndex,S}\n\nScalarFunctionizeBridge implements the following reformulations:\n\nx in S into 1x + 0 in S\n\nSource node\n\nScalarFunctionizeBridge supports:\n\nMOI.VariableIndex in S\n\nTarget nodes\n\nScalarFunctionizeBridge creates:\n\nMOI.ScalarAffineFunction{T} in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorFunctionizeBridge","text":"VectorFunctionizeBridge{T,S} = FunctionConversionBridge{T,MOI.VectorAffineFunction{T},S}\n\nVectorFunctionizeBridge implements the following reformulations:\n\nx in S into Ix + 0 in S\n\nSource node\n\nVectorFunctionizeBridge supports:\n\nMOI.VectorOfVariables in S\n\nTarget nodes\n\nVectorFunctionizeBridge creates:\n\nMOI.VectorAffineFunction{T} in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToScalarQuadraticBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToScalarQuadraticBridge","text":"ToScalarQuadraticBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToScalarQuadraticBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract scalar function and f is a MOI.ScalarQuadraticFunction.\n\nSource node\n\nToScalarQuadraticBridge supports:\n\nG<:AbstractScalarFunction in S\n\nTarget nodes\n\nToScalarQuadraticBridge creates:\n\nMOI.ScalarQuadraticFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToVectorQuadraticBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToVectorQuadraticBridge","text":"ToVectorQuadraticBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToVectorQuadraticBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract vector function and f is a MOI.VectorQuadraticFunction.\n\nSource node\n\nToVectorQuadraticBridge supports:\n\nG<:AbstractVectorFunction in S\n\nTarget nodes\n\nToVectorQuadraticBridge creates:\n\nMOI.VectorQuadraticFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToScalarNonlinearBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToScalarNonlinearBridge","text":"ToScalarNonlinearBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToScalarNonlinearBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract scalar function and f is a MOI.ScalarNonlinearFunction.\n\nSource node\n\nToScalarNonlinearBridge supports:\n\nG<:AbstractScalarFunction in S\n\nTarget nodes\n\nToScalarNonlinearBridge creates:\n\nMOI.ScalarNonlinearFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.FunctionConversionBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.FunctionConversionBridge","text":"FunctionConversionBridge{T,F,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nFunctionConversionBridge implements the following reformulations:\n\ng(x) in S into f(x) in S\n\nfor these pairs of functions:\n\nMOI.ScalarAffineFunctionto [MOI.ScalarQuadraticFunction`](@ref)\nMOI.ScalarQuadraticFunction  to MOI.ScalarNonlinearFunction\nMOI.VectorAffineFunction to MOI.VectorQuadraticFunction\n\nSource node\n\nFunctionConversionBridge supports:\n\nG in S\n\nTarget nodes\n\nFunctionConversionBridge creates:\n\nF in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitComplexEqualToBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitComplexEqualToBridge","text":"SplitComplexEqualToBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSplitComplexEqualToBridge implements the following reformulation:\n\nf(x) + g(x) * im = a + b * im into f(x) = a and g(x) = b\n\nSource node\n\nSplitComplexEqualToBridge supports:\n\nG in MOI.EqualTo{Complex{T}}\n\nwhere G is a function with Complex coefficients.\n\nTarget nodes\n\nSplitComplexEqualToBridge creates:\n\nF in MOI.EqualTo{T}\n\nwhere F is the type of the real/imaginary part of G.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitComplexZerosBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitComplexZerosBridge","text":"SplitComplexZerosBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSplitComplexZerosBridge implements the following reformulation:\n\nf(x) in 0^n into textRe(f(x)) in 0^n and textIm(f(x)) in 0^n\n\nSource node\n\nSplitComplexZerosBridge supports:\n\nG in MOI.Zeros\n\nwhere G is a function with Complex coefficients.\n\nTarget nodes\n\nSplitComplexZerosBridge creates:\n\nF in MOI.Zeros\n\nwhere F is the type of the real/imaginary part of G.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitHyperRectangleBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitHyperRectangleBridge","text":"SplitHyperRectangleBridge{T,G,F} <: Bridges.Constraint.AbstractBridge\n\nSplitHyperRectangleBridge implements the following reformulation:\n\nf(x) in textsfHyperRectangle(l u) to f(x) - l u - f(x) in mathbbR_+.\n\nSource node\n\nSplitHyperRectangleBridge supports:\n\nF in MOI.HyperRectangle\n\nTarget nodes\n\nSplitHyperRectangleBridge creates:\n\nG in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitIntervalBridge","text":"SplitIntervalBridge{T,F,S,LS,US} <: Bridges.Constraint.AbstractBridge\n\nSplitIntervalBridge implements the following reformulations:\n\nl le f(x) le u into f(x) ge l and f(x) le u\nf(x) = b into f(x) ge b and f(x) le b\nf(x) in 0 into f(x) in mathbbR_+ and f(x) in mathbbR_-\n\nSource node\n\nSplitIntervalBridge supports:\n\nF in MOI.Interval{T}\nF in MOI.EqualTo{T}\nF in MOI.Zeros\n\nTarget nodes\n\nSplitIntervalBridge creates:\n\nF in MOI.LessThan{T}\nF in MOI.GreaterThan{T}\n\nor\n\nF in MOI.Nonnegatives\nF in MOI.Nonpositives\n\nnote: Note\nIf T<:AbstractFloat and S is MOI.Interval{T} then no lower (resp. upper) bound constraint is created if the lower (resp. upper) bound is typemin(T) (resp. typemax(T)). Similarly, when MOI.ConstraintSet is set, a lower or upper bound constraint may be deleted or created accordingly.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoRSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoRSOCBridge","text":"SOCtoRSOCBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSOCtoRSOCBridge implements the following reformulation:\n\nx_2 le t into (t+x_1)(t-x_1)ge (x_2ldotsx_N)_2^2\n\nAssumptions\n\nSOCtoRSOCBridge assumes that the length of x is at least one.\n\nSource node\n\nSOCtoRSOCBridge supports:\n\nG in MOI.SecondOrderCone\n\nTarget node\n\nSOCtoRSOCBridge creates:\n\nF in MOI.RotatedSecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoSOCBridge","text":"RSOCtoSOCBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoSOCBridge implements the following reformulation:\n\nx_2^2 le 2tu into fract - usqrt 2 x_2 le  fract + usqrt 2\n\nSource node\n\nRSOCtoSOCBridge supports:\n\nG in MOI.RotatedSecondOrderCone\n\nTarget node\n\nRSOCtoSOCBridge creates:\n\nF in MOI.SecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoNonConvexQuadBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoNonConvexQuadBridge","text":"SOCtoNonConvexQuadBridge{T} <: Bridges.Constraint.AbstractBridge\n\nSOCtoNonConvexQuadBridge implements the following reformulations:\n\nx_2 le t into sum x^2 - t^2 le 0 and 1t + 0 ge 0\n\nThe MOI.ScalarAffineFunction 1t + 0 is used in case the variable has other bound constraints.\n\nwarning: Warning\nThis transformation starts from a convex constraint and creates a non-convex constraint. Unless the solver has explicit support for detecting second-order cones in quadratic form, this may (wrongly) be interpreted by the solver as being non-convex. Therefore, this bridge is not added automatically by MOI.Bridges.full_bridge_optimizer. Care is recommended when adding this bridge to a optimizer.\n\nSource node\n\nSOCtoNonConvexQuadBridge supports:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\nTarget nodes\n\nSOCtoNonConvexQuadBridge creates:\n\nMOI.ScalarQuadraticFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoNonConvexQuadBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoNonConvexQuadBridge","text":"RSOCtoNonConvexQuadBridge{T} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoNonConvexQuadBridge implements the following reformulations:\n\nx_2^2 le 2tu into sum x^2 - 2tu le 0, 1t + 0 ge 0, and 1u + 0 ge 0.\n\nThe MOI.ScalarAffineFunctions 1t + 0 and 1u + 0 are used in case the variables have other bound constraints.\n\nwarning: Warning\nThis transformation starts from a convex constraint and creates a non-convex constraint. Unless the solver has explicit support for detecting rotated second-order cones in quadratic form, this may (wrongly) be interpreted by the solver as being non-convex. Therefore, this bridge is not added automatically by MOI.Bridges.full_bridge_optimizer. Care is recommended when adding this bridge to a optimizer.\n\nSource node\n\nRSOCtoNonConvexQuadBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoNonConvexQuadBridge creates:\n\nMOI.ScalarQuadraticFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.QuadtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.QuadtoSOCBridge","text":"QuadtoSOCBridge{T} <: Bridges.Constraint.AbstractBridge\n\nQuadtoSOCBridge converts quadratic inequalities\n\nfrac12x^T Q x + a^T x le ub\n\ninto MOI.RotatedSecondOrderCone constraints, but it only applies when Q is positive definite.\n\nThis is because, if Q is positive definite, there exists U such that Q = U^T U, and so the inequality can then be rewritten as;\n\nU x_2^2 le 2 (-a^T x + ub)\n\nTherefore, QuadtoSOCBridge implements the following reformulations:\n\nfrac12x^T Q x + a^T x le ub into (1 -a^T x + ub Ux) in RotatedSecondOrderCone where Q = U^T U\nfrac12x^T Q x + a^T x ge lb into (1 a^T x - lb Ux) in RotatedSecondOrderCone where -Q = U^T U\n\nSource node\n\nQuadtoSOCBridge supports:\n\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\nTarget nodes\n\nRelativeEntropyBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.RotatedSecondOrderCone\n\nErrors\n\nThis bridge errors if Q is not positive definite.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoPSDBridge","text":"SOCtoPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSOCtoPSDBridge implements the following reformulation:\n\nx_2 le t into leftbeginarrayc ct  x^top  x  t mathbfIendarrayrightsucceq 0\n\nwarning: Warning\nThis bridge is not added by default by MOI.Bridges.full_bridge_optimizer because bridging second order cone constraints to semidefinite constraints can be achieved by the SOCtoRSOCBridge followed by the RSOCtoPSDBridge, while creating a smaller semidefinite constraint.\n\nSource node\n\nSOCtoPSDBridge supports:\n\nG in MOI.SecondOrderCone\n\nTarget nodes\n\nSOCtoPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoPSDBridge","text":"RSOCtoPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoPSDBridge implements the following reformulation:\n\nx_2^2 le 2tcdot u into leftbeginarrayc ct  x^top  x  2tu mathbfIendarrayrightsucceq 0\n\nSource node\n\nRSOCtoPSDBridge supports:\n\nG in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormInfinityBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormInfinityBridge","text":"NormInfinityBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormInfinityBridge implements the following reformulation:\n\nx_infty le t into t - x_i t + x_i in mathbbR_+.\n\nSource node\n\nNormInfinityBridge supports:\n\nG in MOI.NormInfinityCone{T}\n\nTarget nodes\n\nNormInfinityBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormOneBridge","text":"NormOneBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormOneBridge implements the following reformulation:\n\nsum x_i le t into t - sum y_i y_i - x_i y_i + x_i in mathbbR_+.\n\nSource node\n\nNormOneBridge supports:\n\nG in MOI.NormOneCone{T}\n\nTarget nodes\n\nNormOneBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormToPowerBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormToPowerBridge","text":"NormToPowerBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormToPowerBridge implements the following reformulation:\n\n(t x) in NormCone(p 1+d) into (r_i t x_i) in PowerCone(1  p) for all i, and sumlimits_i r_i == t.\n\nFor details, see Alizadeh, F., and Goldfarb, D. (2001). \"Second-order cone programming.\" Mathematical Programming, Series B, 95:3-51.\n\nSource node\n\nNormToPowerBridge supports:\n\nF in MOI.NormCone\n\nTarget nodes\n\nNormToPowerBridge creates:\n\nF in MOI.PowerCone{T}\nMOI.ScalarAffineFunction in MOI.EqualTo\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormOneConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormOneConeToNormConeBridge","text":"NormOneConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormOneConeToNormConeBridge implements the following reformulations:\n\n(t x) in NormOneCone(d) into (t x) in NormCone(1 d)\n\nSource node\n\nNormOneConeToNormConeBridge supports:\n\nF in MOI.NormOneCone\n\nTarget nodes\n\nNormOneConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SecondOrderConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SecondOrderConeToNormConeBridge","text":"SecondOrderConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nSecondOrderConeToNormConeBridge implements the following reformulations:\n\n(t x) in SecondOrderCone(d) into (t x) in NormCone(2 d)\n\nSource node\n\nSecondOrderConeToNormConeBridge supports:\n\nF in MOI.SecondOrderCone\n\nTarget nodes\n\nSecondOrderConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormInfinityConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormInfinityConeToNormConeBridge","text":"NormInfinityConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormInfinityConeToNormConeBridge implements the following reformulations:\n\n(t x) in NormInfinityCone(d) into (t x) in NormCone(Inf d)\n\nSource node\n\nNormInfinityConeToNormConeBridge supports:\n\nF in MOI.NormInfinityCone\n\nTarget nodes\n\nNormInfinityConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeantoRelEntrBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeantoRelEntrBridge","text":"GeoMeantoRelEntrBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nGeoMeantoRelEntrBridge implements the following reformulation:\n\n(u w) in GeometricMeanCone into (0 w (u + y)mathbf1)in RelativeEntropyCone and y ge 0\n\nSource node\n\nGeoMeantoRelEntrBridge supports:\n\nH in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeantoRelEntrBridge creates:\n\nG in MOI.RelativeEntropyCone\nF in MOI.Nonnegatives\n\nDerivation\n\nThe derivation of the bridge is as follows:\n\nbeginaligned\n(u w) in GeometricMeanCone iff  u le left(prod_i=1^n w_iright)^1n \niff  0 le u + y le left(prod_i=1^n w_iright)^1n y ge 0 \niff  1 le fracleft(prod_i=1^n w_iright)^1nu + y y ge 0 \niff  1 le left(prod_i=1^n fracw_iu + yright)^1n y ge 0 \niff  0 le sum_i=1^n logleft(fracw_iu + yright) y ge 0 \niff  0 ge sum_i=1^n logleft(fracu + yw_iright) y ge 0 \niff  0 ge sum_i=1^n (u + y) logleft(fracu + yw_iright) y ge 0 \niff  (0 w (u + y)mathbf1) in RelativeEntropyCone y ge 0 \nendaligned\n\nThis derivation assumes that u + y  0, which is enforced by the relative entropy cone.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge","text":"GeoMeanToPowerBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nGeoMeanToPowerBridge implements the following reformulation:\n\n(y x) in GeometricMeanCone(1+d) into (x_1 t y) in PowerCone(1d) and (t x_2  x_d) in GeometricMeanCone(d), which is then recursively expanded into more PowerCone constraints.\n\nSource node\n\nGeoMeanToPowerBridge supports:\n\nF in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeanToPowerBridge creates:\n\nF in MOI.PowerCone{T}\nMOI.VectorOfVariables in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeanBridge","text":"GeoMeanBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nGeoMeanBridge implements a reformulation from MOI.GeometricMeanCone into MOI.RotatedSecondOrderCone.\n\nThe reformulation is best described in an example.\n\nConsider the cone of dimension 4:\n\nt le sqrt3x_1 x_2 x_3\n\nThis can be rewritten as exists y ge 0 such that:\n\nbeginalign*\n  t  le y\n  y^4  le x_1 x_2 x_3 y\nendalign*\n\nNote that we need to create y and not use t^4 directly because t is allowed to be negative.\n\nThis is equivalent to:\n\nbeginalign*\n    t  le fracy_1sqrt4\n    y_1^2  le 2y_2 y_3\n    y_2^2  le 2x_1 x_2 \n    y_3^2  le 2x_3(y_1sqrt4) \n    y      ge 0\nendalign*\n\nMore generally, you can show how the geometric mean code is recursively expanded into a set of new variables y in MOI.Nonnegatives, a set of MOI.RotatedSecondOrderCone constraints, and a MOI.LessThan constraint between t and y_1.\n\nSource node\n\nGeoMeanBridge supports:\n\nH in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeanBridge creates:\n\nF in MOI.LessThan{T}\nG in MOI.RotatedSecondOrderCone\nG in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RelativeEntropyBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RelativeEntropyBridge","text":"RelativeEntropyBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nRelativeEntropyBridge implements the following reformulation that converts a MOI.RelativeEntropyCone into an MOI.ExponentialCone:\n\nu ge sum_i=1^n w_i log left(fracw_iv_iright) into y_i ge 0, u ge sum_i=1^n y_i, and (-y_i w_i v_i) in ExponentialCone.\n\nSource node\n\nRelativeEntropyBridge supports:\n\nH in MOI.RelativeEntropyCone\n\nTarget nodes\n\nRelativeEntropyBridge creates:\n\nF in MOI.GreaterThan{T}\nG in MOI.ExponentialCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormSpectralBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormSpectralBridge","text":"NormSpectralBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormSpectralBridge implements the following reformulation:\n\nt ge sigma_1(X) into leftbeginarrayc ctmathbfI  X^top  X  t mathbfIendarrayright succeq 0\n\nSource node\n\nNormSpectralBridge supports:\n\nG in MOI.NormSpectralCone\n\nTarget nodes\n\nNormSpectralBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormNuclearBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormNuclearBridge","text":"NormNuclearBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nNormNuclearBridge implements the following reformulation:\n\nt ge sum_i sigma_i (X) into leftbeginarrayc cU  X^top  X  Vendarrayright succeq 0 and 2t ge tr(U) + tr(V).\n\nSource node\n\nNormNuclearBridge supports:\n\nH in MOI.NormNuclearCone\n\nTarget nodes\n\nNormNuclearBridge creates:\n\nF in MOI.GreaterThan{T}\nG in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SquareBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SquareBridge","text":"SquareBridge{T,F,G,TT,ST} <: Bridges.Constraint.AbstractBridge\n\nSquareBridge implements the following reformulations:\n\n(t u X) in LogDetConeSquare into (t u Y) in LogDetConeTriangle\n(t X) in RootDetConeSquare into (t Y) in RootDetConeTriangle\nX in AbstractSymmetricMatrixSetSquare into Y in AbstractSymmetricMatrixSetTriangle\n\nwhere Y is the upper triangluar component of X.\n\nIn addition, constraints are added as necessary to constrain the matrix X to be symmetric. For example, the constraint for the matrix:\n\nbeginpmatrix\n  1       1 + x  2 - 3x\n  1 +  x  2 + x  3 -  x\n  2 - 3x  2 + x      2x\nendpmatrix\n\ncan be broken down to the constraint of the symmetric matrix\n\nbeginpmatrix\n  1       1 + x  2 - 3x\n  cdot  2 + x  3 -  x\n  cdot  cdot     2x\nendpmatrix\n\nand the equality constraint between the off-diagonal entries (2, 3) and (3, 2) 3 - x == 2 + x. Note that no symmetrization constraint needs to be added between the off-diagonal entries (1, 2) and (2, 1) or between (1, 3) and (3, 1) because the expressions are the same.\n\nSource node\n\nSquareBridge supports:\n\nF in ST\n\nTarget nodes\n\nSquareBridge creates:\n\nG in TT\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.HermitianToSymmetricPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.HermitianToSymmetricPSDBridge","text":"HermitianToSymmetricPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nHermitianToSymmetricPSDBridge implements the following reformulation:\n\nHermitian positive semidefinite n x n complex matrix to a symmetric positive semidefinite 2n x 2n real matrix.\n\nSee also MOI.Bridges.Variable.HermitianToSymmetricPSDBridge.\n\nSource node\n\nHermitianToSymmetricPSDBridge supports:\n\nG in MOI.HermitianPositiveSemidefiniteConeTriangle\n\nTarget node\n\nHermitianToSymmetricPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\nReformulation\n\nThe reformulation is best described by example.\n\nThe Hermitian matrix:\n\nbeginbmatrix\n  x_11             x_12 + y_12im  x_13 + y_13im\n  x_12 - y_12im  x_22             x_23 + y_23im\n  x_13 - y_13im  x_23 - y_23im  x_33\nendbmatrix\n\nis positive semidefinite if and only if the symmetric matrix:\n\nbeginbmatrix\n    x_11  x_12  x_13  0        y_12   y_13 \n            x_22  x_23  -y_12  0        y_23 \n                    x_33  -y_13  -y_23  0      \n                            x_11   x_12   x_13 \n                                     x_22   x_23 \n                                              x_33\nendbmatrix\n\nis positive semidefinite.\n\nThe bridge achieves this reformulation by constraining the above matrix to belong to the MOI.PositiveSemidefiniteConeTriangle(6).\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SetDotScalingBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SetDotScalingBridge","text":"SetDotScalingBridge{T,S,F,G} <: Bridges.Constraint.AbstractBridge\n\nSetDotScalingBridge implements the reformulation from constraints in S to constraints in MOI.Scaled{S}.\n\nSource node\n\nSetDotScalingBridge supports:\n\nG in S\n\nTarget node\n\nSetDotScalingBridge creates:\n\nF in MOI.Scaled{S}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SetDotInverseScalingBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SetDotInverseScalingBridge","text":"SetDotInverseScalingBridge{T,S,F,G} <: Bridges.Constraint.AbstractBridge\n\nSetDotInverseScalingBridge implements the reformulation from constraints in the MOI.Scaled{S} to constraints in the S.\n\nSource node\n\nSetDotInverseScalingBridge supports:\n\nG in MOI.Scaled{S}\n\nTarget node\n\nSetDotInverseScalingBridge creates:\n\nF in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RootDetBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RootDetBridge","text":"RootDetBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nThe MOI.RootDetConeTriangle is representable by MOI.PositiveSemidefiniteConeTriangle and MOI.GeometricMeanCone constraints, see [1, p. 149].\n\nIndeed, t le det(X)^1n if and only if there exists a lower triangular matrix Δ such that:\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  (t mathrmDiag(Δ))  in GeometricMeanCone\nendalign*\n\nSource node\n\nRootDetBridge supports:\n\nI in MOI.RootDetConeTriangle\n\nTarget nodes\n\nRootDetBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\nG in MOI.GeometricMeanCone\n\n[1] Ben-Tal, Aharon, and Arkadi Nemirovski. Lectures on modern convex     optimization: analysis, algorithms, and engineering applications. Society     for Industrial and Applied Mathematics, 2001.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LogDetBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LogDetBridge","text":"LogDetBridge{T,F,G,H,I} <: Bridges.Constraint.AbstractBridge\n\nThe MOI.LogDetConeTriangle is representable by MOI.PositiveSemidefiniteConeTriangle and MOI.ExponentialCone constraints.\n\nIndeed, logdet(X) = sumlimits_i=1^n log(delta_i) where delta_i are the eigenvalues of X.\n\nAdapting the method from [1, p. 149], we see that t le u log(det(Xu)) for u  0 if and only if there exists a lower triangular matrix Δ such that\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  t - sum_i=1^n u logleft(fracΔ_iiuright)  le 0\nendalign*\n\nWhich we reformulate further into\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  (l_i u  Δ_ii)  in ExponentialConequad forall i \n  t - sum_i=1^n l_i  le 0\nendalign*\n\nSource node\n\nLogDetBridge supports:\n\nI in MOI.LogDetConeTriangle\n\nTarget nodes\n\nLogDetBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\nG in MOI.ExponentialCone\nH in MOI.LessThan{T}\n\n[1] Ben-Tal, Aharon, and Arkadi Nemirovski. Lectures on modern convex     optimization: analysis, algorithms, and engineering applications. Society     for Industrial and Applied Mathematics, 2001.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorActiveOnFalseBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorActiveOnFalseBridge","text":"IndicatorActiveOnFalseBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nIndicatorActiveOnFalseBridge implements the following reformulation:\n\nneg z implies f(x) in S into y implies f(x) in S, z + y = 1, and y in 0 1\n\nSource node\n\nIndicatorActiveOnFalseBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ZERO,S}\n\nTarget nodes\n\nIndicatorActiveOnFalseBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ONE,S}\nMOI.ScalarAffineFunction{T} in MOI.EqualTo\nMOI.VariableIndex in MOI.ZeroOne\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorGreaterToLessThanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorGreaterToLessThanBridge","text":"IndicatorGreaterToLessThanBridge{T,A} <: Bridges.Constraint.AbstractBridge\n\nIndicatorGreaterToLessThanBridge implements the following reformulation:\n\nz implies f(x) ge l into z implies -f(x) le -l\n\nSource node\n\nIndicatorGreaterToLessThanBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.GreaterThan{T}}\n\nTarget nodes\n\nIndicatorGreaterToLessThanBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.LessThan{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorLessToGreaterThanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorLessToGreaterThanBridge","text":"IndicatorLessToGreaterThanBridge{T,A} <: Bridges.Constraint.AbstractBridge\n\nIndicatorLessToGreaterThanBridge implements the following reformulations:\n\nz implies f(x) le u into z implies -f(x) ge -u\n\nSource node\n\nIndicatorLessToGreaterThanBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.LessThan{T}}\n\nTarget nodes\n\nIndicatorLessToGreaterThanBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.GreaterThan{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorSOS1Bridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorSOS1Bridge","text":"IndicatorSOS1Bridge{T,S} <: Bridges.Constraint.AbstractBridge\n\nIndicatorSOS1Bridge implements the following reformulation:\n\nz implies f(x) in S into f(x) + y in S, SOS1(y z)\n\nwarning: Warning\nThis bridge assumes that the solver supports MOI.SOS1{T} constraints in which one of the variables (y) is continuous.\n\nSource node\n\nIndicatorSOS1Bridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ONE,S}\n\nTarget nodes\n\nIndicatorSOS1Bridge creates:\n\nMOI.ScalarAffineFunction{T} in S\nMOI.VectorOfVariables in MOI.SOS1{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SemiToBinaryBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SemiToBinaryBridge","text":"SemiToBinaryBridge{T,S} <: Bridges.Constraint.AbstractBridge\n\nSemiToBinaryBridge implements the following reformulations:\n\nx in 0 cup l u into\nbeginaligned\nx leq z u \nx geq z l \nz in 0 1\nendaligned\nx in 0 cup l ldots u into\nbeginaligned\nx leq z u \nx geq z l \nz in 0 1 \nx in mathbbZ\nendaligned\n\nSource node\n\nSemiToBinaryBridge supports:\n\nMOI.VariableIndex in MOI.Semicontinuous{T}\nMOI.VariableIndex in MOI.Semiinteger{T}\n\nTarget nodes\n\nSemiToBinaryBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\nMOI.VariableIndex{T} in MOI.Integer (if S is MOI.Semiinteger{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ZeroOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ZeroOneBridge","text":"ZeroOneBridge{T} <: Bridges.Constraint.AbstractBridge\n\nZeroOneBridge implements the following reformulation:\n\nx in 0 1 into x in mathbbZ, 1x in 0 1.\n\nnote: Note\nZeroOneBridge adds a linear constraint instead of adding variable bounds to avoid conflicting with bounds set by the user.\n\nSource node\n\nZeroOneBridge supports:\n\nMOI.VariableIndex in MOI.ZeroOne\n\nTarget nodes\n\nZeroOneBridge creates:\n\nMOI.VariableIndex in MOI.Integer\nMOI.ScalarAffineFunction{T} in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IntegerToZeroOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IntegerToZeroOneBridge","text":"IntegerToZeroOneBridge{T} <: Bridges.Constraint.AbstractBridge\n\nIntegerToZeroOneBridge implements the following reformulation:\n\nx in mathbfZ into y_i in 0 1, x == lb + sum 2^i-1 y_i.\n\nSource node\n\nIntegerToZeroOneBridge supports:\n\nVariableIndex in MOI.Integer\n\nTarget nodes\n\nIntegerToZeroOneBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\nDeveloper note\n\nThis bridge is implemented as a constraint bridge instead of a variable bridge because we don't want to substitute the linear combination of y for every instance of x. Doing so would be expensive and greatly reduce the sparsity of the constraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NumberConversionBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NumberConversionBridge","text":"NumberConversionBridge{T,F1,S1,F2,S2} <: Bridges.Constraint.AbstractBridge\n\nNumberConversionBridge implements the following reformulation:\n\nf1(x) in S1 to f2(x) in S2\n\nwhere f and S are the same functional form, but differ in their coefficient type.\n\nSource node\n\nNumberConversionBridge supports:\n\nF1 in S1\n\nTarget node\n\nNumberConversionBridge creates:\n\nF2 in S2\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.AllDifferentToCountDistinctBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.AllDifferentToCountDistinctBridge","text":"AllDifferentToCountDistinctBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nAllDifferentToCountDistinctBridge implements the following reformulations:\n\nx in textsfAllDifferent(d) to (n x) in textsfCountDistinct(1+d) and n = d\nf(x) in textsfAllDifferent(d) to (d f(x)) in textsfCountDistinct(1+d)\n\nSource node\n\nAllDifferentToCountDistinctBridge supports:\n\nF in MOI.AllDifferent\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nAllDifferentToCountDistinctBridge creates:\n\nF in MOI.CountDistinct\nMOI.VariableIndex in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ReifiedAllDifferentToCountDistinctBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ReifiedAllDifferentToCountDistinctBridge","text":"ReifiedAllDifferentToCountDistinctBridge{T,F} <:\nBridges.Constraint.AbstractBridge\n\nReifiedAllDifferentToCountDistinctBridge implements the following reformulations:\n\nr iff x in textsfAllDifferent(d) to r iff (n x) in textsfCountDistinct(1+d) and n = d\nr iff f(x) in textsfAllDifferent(d) to r iff (d f(x)) in textsfCountDistinct(1+d)\n\nSource node\n\nReifiedAllDifferentToCountDistinctBridge supports:\n\nF in MOI.Reified{MOI.AllDifferent}\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nReifiedAllDifferentToCountDistinctBridge creates:\n\nF in MOI.Reified{MOI.CountDistinct}\nMOI.VariableIndex in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.BinPackingToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.BinPackingToMILPBridge","text":"BinPackingToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nBinPackingToMILPBridge implements the following reformulation:\n\nx in BinPacking(c w) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we add the capacity constraint for all possible bins j:\n\nsumlimits_i w_i z_ij le c forall j in bigcup_i=1ldotsd S_i\n\nSource node\n\nBinPackingToMILPBridge supports:\n\nF in MOI.BinPacking{T}\n\nTarget nodes\n\nBinPackingToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CircuitToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CircuitToMILPBridge","text":"CircuitToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCircuitToMILPBridge implements the following reformulation:\n\nx in textsfCircuit(d) to the Miller-Tucker-Zemlin formulation of the Traveling Salesperson Problem.\n\nSource node\n\nCircuitToMILPBridge supports:\n\nF in MOI.Circuit\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCircuitToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{T}\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountAtLeastToCountBelongsBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountAtLeastToCountBelongsBridge","text":"CountAtLeastToCountBelongsBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountAtLeastToCountBelongsBridge implements the following reformulation:\n\nx in textsfCountAtLeast(n d mathcalS) to (n_i x_d_i) in textsfCountBelongs(1+d mathcalS) and n_i ge n for all i.\n\nSource node\n\nCountAtLeastToCountBelongsBridge supports:\n\nF in MOI.CountAtLeast\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountAtLeastToCountBelongsBridge creates:\n\nF in MOI.CountBelongs\nMOI.VariableIndex in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountBelongsToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountBelongsToMILPBridge","text":"CountBelongsToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountBelongsToMILPBridge implements the following reformulation:\n\n(n x) in textsfCountBelongs(1+d mathcalS) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nFinally, n is constrained to be the number of z_ij elements that are in mathcalS:\n\nn - sumlimits_iin 1ldots d j in mathcalS z_ij = 0\n\nSource node\n\nCountBelongsToMILPBridge supports:\n\nF in MOI.CountBelongs\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountBelongsToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountDistinctToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountDistinctToMILPBridge","text":"CountDistinctToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountDistinctToMILPBridge implements the following reformulation:\n\n(n x) in textsfCountDistinct(1+d) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we introduce new binary variables y_j, which are 1 if a variable takes the value j in the optimal solution and 0 otherwise.\n\nbeginaligned\ny_j in 0 1                          forall j in bigcup_i=1ldotsd S_i \ny_j le sumlimits_i in 1ldots d j in S_i z_ij le M y_j   forall j in bigcup_i=1ldotsd S_i\nendaligned\n\nFinally, n is constrained to be the number of y_j elements that are non-zero:\n\nn - sumlimits_j in bigcup_i=1ldotsd S_i y_j = 0\n\nSource node\n\nCountDistinctToMILPBridge supports:\n\nF in MOI.CountDistinct\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountDistinctToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ReifiedCountDistinctToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ReifiedCountDistinctToMILPBridge","text":"ReifiedCountDistinctToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nReifiedCountDistinctToMILPBridge implements the following reformulation:\n\nr iff (n x) in textsfCountDistinct(1+d) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we introduce new binary variables y_j, which are 1 if a variable takes the value j in the optimal solution and 0 otherwise.\n\nbeginaligned\ny_j in 0 1                          forall j in bigcup_i=1ldotsd S_i \ny_j le sumlimits_i in 1ldots d j in S_i z_ij le M y_j   forall j in bigcup_i=1ldotsd S_i\nendaligned\n\nFinally, n is constrained to be the number of y_j elements that are non-zero, with some slack:\n\nn - sumlimits_j in bigcup_i=1ldotsd S_i y_j = delta^+ - delta^-\n\nAnd then the slack is constrained to respect the reif variable r:\n\nbeginaligned\nd_1 le delta^+ le M d_1 \nd_2 le delta^- le M d_s \nd_1 + d_2 + r = 1             \nd_1 d_2 in 0 1\nendaligned\n\nSource node\n\nReifiedCountDistinctToMILPBridge supports:\n\nF in MOI.Reified{MOI.CountDistinct}\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nReifiedCountDistinctToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountGreaterThanToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountGreaterThanToMILPBridge","text":"CountGreaterThanToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountGreaterThanToMILPBridge implements the following reformulation:\n\n(c y x) in CountGreaterThan() into a mixed-integer linear program.\n\nSource node\n\nCountGreaterThanToMILPBridge supports:\n\nF in MOI.CountGreaterThan\n\nTarget nodes\n\nCountGreaterThanToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.TableToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.TableToMILPBridge","text":"TableToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nTableToMILPBridge implements the following reformulation:\n\nx in Table(t) into\nbeginaligned\nz_j in 0 1                      quad forall i j \nsumlimits_j=1^n z_j = 1                           \nsumlimits_j=1^n t_ij z_j = x_i  quad forall i\nendaligned\n\nSource node\n\nTableToMILPBridge supports:\n\nF in MOI.Table{T}\n\nTarget nodes\n\nTableToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#objective_bridges_ref","page":"List of bridges","title":"Objective bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Objective.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Objective.FunctionizeBridge\nBridges.Objective.QuadratizeBridge\nBridges.Objective.SlackBridge\nBridges.Objective.VectorFunctionizeBridge\nBridges.Objective.VectorSlackBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.FunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.FunctionizeBridge","text":"FunctionizeBridge{T}\n\nFunctionizeBridge implements the following reformulations:\n\nmin x into min1x + 0\nmax x into max1x + 0\n\nwhere T is the coefficient type of 1 and 0.\n\nSource node\n\nFunctionizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.VariableIndex}\n\nTarget nodes\n\nFunctionizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.QuadratizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.QuadratizeBridge","text":"QuadratizeBridge{T}\n\nQuadratizeBridge implements the following reformulations:\n\nmin a^top x + b into minx^top mathbf0 x + a^top x + b\nmax a^top x + b into maxx^top mathbf0 x + a^top x + b\n\nwhere T is the coefficient type of 0.\n\nSource node\n\nQuadratizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\n\nTarget nodes\n\nQuadratizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.SlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.SlackBridge","text":"SlackBridge{T,F,G}\n\nSlackBridge implements the following reformulations:\n\nminf(x) into miny f(x) - y le 0\nmaxf(x) into maxy f(x) - y ge 0\n\nwhere F is the type of f(x) - y, G is the type of f(x), and T is the coefficient type of f(x).\n\nSource node\n\nSlackBridge supports:\n\nMOI.ObjectiveFunction{G}\n\nTarget nodes\n\nSlackBridge creates:\n\nOne variable node: MOI.VariableIndex in MOI.Reals\nOne objective node: MOI.ObjectiveFunction{MOI.VariableIndex}\nOne constraint node, that depends on the MOI.ObjectiveSense:\nF-in-MOI.LessThan if MIN_SENSE\nF-in-MOI.GreaterThan if MAX_SENSE\n\nwarning: Warning\nWhen using this bridge, changing the optimization sense is not supported. Set the sense to MOI.FEASIBILITY_SENSE first to delete the bridge, then set MOI.ObjectiveSense and re-add the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.VectorFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.VectorFunctionizeBridge","text":"VectorFunctionizeBridge{T}\n\nVectorFunctionizeBridge implements the following reformulations:\n\nmin x into min1x + 0\nmax x into max1x + 0\n\nwhere T is the coefficient type of 1 and 0.\n\nSource node\n\nVectorFunctionizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.VectorOfVariables}\n\nTarget nodes\n\nVectorFunctionizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.VectorAffineFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.VectorSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.VectorSlackBridge","text":"VectorSlackBridge{T,F,G}\n\nVectorSlackBridge implements the following reformulations:\n\nminf(x) into miny y - f(x) in mathbbR_+ \nmaxf(x) into maxy f(x) - y in mathbbR_+ \n\nwhere F is the type of f(x) - y, G is the type of f(x), and T is the coefficient type of f(x).\n\nSource node\n\nVectorSlackBridge supports:\n\nMOI.ObjectiveFunction{G}\n\nTarget nodes\n\nVectorSlackBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Reals\nOne objective node: MOI.ObjectiveFunction{MOI.VectorOfVariables}\nOne constraint node: F-in-MOI.Nonnegatives\n\nwarning: Warning\nWhen using this bridge, changing the optimization sense is not supported. Set the sense to MOI.FEASIBILITY_SENSE first to delete the bridge, then set MOI.ObjectiveSense and re-add the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#variable_bridges_ref","page":"List of bridges","title":"Variable bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Variable.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Variable.FreeBridge\nBridges.Variable.NonposToNonnegBridge\nBridges.Variable.RSOCtoPSDBridge\nBridges.Variable.RSOCtoSOCBridge\nBridges.Variable.SOCtoRSOCBridge\nBridges.Variable.VectorizeBridge\nBridges.Variable.ZerosBridge\nBridges.Variable.HermitianToSymmetricPSDBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.FreeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.FreeBridge","text":"FreeBridge{T} <: Bridges.Variable.AbstractBridge\n\nFreeBridge implements the following reformulation:\n\nx in mathbbR into y z ge 0 with the substitution rule x = y - z,\n\nwhere T is the coefficient type of y - z.\n\nSource node\n\nFreeBridge supports:\n\nMOI.VectorOfVariables in MOI.Reals\n\nTarget nodes\n\nFreeBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.NonposToNonnegBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.NonposToNonnegBridge","text":"NonposToNonnegBridge{T} <: Bridges.Variable.AbstractBridge\n\nNonposToNonnegBridge implements the following reformulation:\n\nx in mathbbR_- into y in mathbbR_+ with the substitution rule x = -y,\n\nwhere T is the coefficient type of -y.\n\nSource node\n\nNonposToNonnegBridge supports:\n\nMOI.VectorOfVariables in MOI.Nonpositives\n\nTarget nodes\n\nNonposToNonnegBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Nonnegatives,\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.RSOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.RSOCtoPSDBridge","text":"RSOCtoPSDBridge{T} <: Bridges.Variable.AbstractBridge\n\nRSOCtoPSDBridge implements the following reformulation:\n\nx_2^2 le 2tu where t u ge 0 into Y succeq 0, with the substitution rule: Y = leftbeginarrayc ct  x^top  x  2u mathbfIendarrayright\n\nAdditional bounds are added to ensure the off-diagonals of the 2uI submatrix are 0, and linear constraints are added to ennsure the diagonal of 2uI takes the same values.\n\nAs a special case, if x = 0, then RSOCtoPSDBridge reformulates into (t u) in mathbbR_+.\n\nSource node\n\nRSOCtoPSDBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoPSDBridge creates:\n\nOne variable node that depends on the input dimension:\nMOI.VectorOfVariables in MOI.Nonnegatives if dimension is 1 or 2\nMOI.VectorOfVariables in\nMOI.PositiveSemidefiniteConeTriangle otherwise\nThe constraint node MOI.VariableIndex in MOI.EqualTo\nThe constrant node MOI.ScalarAffineFunction in MOI.EqualTo\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.RSOCtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.RSOCtoSOCBridge","text":"RSOCtoSOCBridge{T} <: Bridges.Variable.AbstractBridge\n\nRSOCtoSOCBridge implements the following reformulation:\n\nx_2^2 le 2tu into v_2 le w, with the substitution rules t = fracwsqrt 2 + fracv_1sqrt 2, u = fracwsqrt 2 - fracv_1sqrt 2, and x = (v_2ldotsv_N).\n\nSource node\n\nRSOCtoSOCBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget node\n\nRSOCtoSOCBridge creates:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.SOCtoRSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.SOCtoRSOCBridge","text":"SOCtoRSOCBridge{T} <: Bridges.Variable.AbstractBridge\n\nSOCtoRSOCBridge implements the following reformulation:\n\nx_2 le t into 2uv ge w_2^2, with the substitution rules t = fracusqrt 2 + fracvsqrt 2, x = (fracusqrt 2 - fracvsqrt 2 w).\n\nAssumptions\n\nSOCtoRSOCBridge assumes that x ge 1.\n\nSource node\n\nSOCtoRSOCBridge supports:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\nTarget node\n\nSOCtoRSOCBridge creates:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.VectorizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.VectorizeBridge","text":"VectorizeBridge{T,S} <: Bridges.Variable.AbstractBridge\n\nVectorizeBridge implements the following reformulations:\n\nx ge a into y in mathbbR_+ with the substitution rule x = a + y\nx le a into y in mathbbR_- with the substitution rule x = a + y\nx == a into y in 0 with the substitution rule x = a + y\n\nwhere T is the coefficient type of a + y.\n\nSource node\n\nVectorizeBridge supports:\n\nMOI.VariableIndex in MOI.GreaterThan{T}\nMOI.VariableIndex in MOI.LessThan{T}\nMOI.VariableIndex in MOI.EqualTo{T}\n\nTarget nodes\n\nVectorizeBridge creates:\n\nOne variable node: MOI.VectorOfVariables in S, where S is one of MOI.Nonnegatives, MOI.Nonpositives, MOI.Zeros depending on the type of S.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.ZerosBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.ZerosBridge","text":"ZerosBridge{T} <: Bridges.Variable.AbstractBridge\n\nZerosBridge implements the following reformulation:\n\nx in 0 into the substitution rule x = 0,\n\nwhere T is the coefficient type of 0.\n\nSource node\n\nZerosBridge supports:\n\nMOI.VectorOfVariables in MOI.Zeros\n\nTarget nodes\n\nZerosBridge does not create target nodes. It replaces all instances of x with 0 via substitution. This means that no variables are created in the underlying model.\n\nCaveats\n\nThe bridged variables are similar to parameters with zero values. Parameters with non-zero values can be created with constrained variables in MOI.EqualTo by combining a VectorizeBridge and this bridge.\n\nHowever, functions modified by ZerosBridge cannot be unbridged. That is, for a given function, we cannot determine if the bridged variables were used.\n\nA related implication is that this bridge does not support MOI.ConstraintDual. However, if a MOI.Utilities.CachingOptimizer is used, the dual can be determined by the bridged optimizer using MOI.Utilities.get_fallback because the caching optimizer records the unbridged function.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.HermitianToSymmetricPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.HermitianToSymmetricPSDBridge","text":"HermitianToSymmetricPSDBridge{T} <: Bridges.Variable.AbstractBridge\n\nHermitianToSymmetricPSDBridge implements the following reformulation:\n\nHermitian positive semidefinite n x n complex matrix to a symmetric positive semidefinite 2n x 2n real matrix satisfying equality constraints described below.\n\nSource node\n\nHermitianToSymmetricPSDBridge supports:\n\nMOI.VectorOfVariables in MOI.HermitianPositiveSemidefiniteConeTriangle\n\nTarget node\n\nHermitianToSymmetricPSDBridge creates:\n\nMOI.VectorOfVariables in MOI.PositiveSemidefiniteConeTriangle\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\nReformulation\n\nThe reformulation is best described by example.\n\nThe Hermitian matrix:\n\nbeginbmatrix\n  x_11             x_12 + y_12im  x_13 + y_13im\n  x_12 - y_12im  x_22             x_23 + y_23im\n  x_13 - y_13im  x_23 - y_23im  x_33\nendbmatrix\n\nis positive semidefinite if and only if the symmetric matrix:\n\nbeginbmatrix\n    x_11  x_12  x_13  0        y_12   y_13 \n            x_22  x_23  -y_12  0        y_23 \n                    x_33  -y_13  -y_23  0      \n                            x_11   x_12   x_13 \n                                     x_22   x_23 \n                                              x_33\nendbmatrix\n\nis positive semidefinite.\n\nThe bridge achieves this reformulation by adding a new set of variables in MOI.PositiveSemidefiniteConeTriangle(6), and then adding three groups of equality constraints to:\n\nconstrain the two x blocks to be equal\nforce the diagonal of the y blocks to be 0\nforce the lower triangular of the y block to be the negative of the upper triangle.\n\n\n\n\n\n","category":"type"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"EditURL = \"https://github.com/jump-dev/SeDuMi.jl/blob/v0.4.2/README.md\"","category":"page"},{"location":"packages/SeDuMi/#SeDuMi.jl","page":"jump-dev/SeDuMi.jl","title":"SeDuMi.jl","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"SeDuMi.jl is wrapper for the SeDuMi solver.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"an exported sedumi function that is a thin wrapper on top of the sedumi MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SeDuMi/#Affiliation","page":"jump-dev/SeDuMi.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SeDuMi.","category":"page"},{"location":"packages/SeDuMi/#License","page":"jump-dev/SeDuMi.jl","title":"License","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"SeDuMi.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The underlying solver, sqlp/sedumi is licensed under the GPL v2 license.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"In addition, SeDuMi requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SeDuMi/#Use-with-JuMP","page":"jump-dev/SeDuMi.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"To use SeDuMi with JuMP, do:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"using JuMP, SeDuMi\nmodel = Model(SeDuMi.Optimizer)\nset_attribute(model, \"fid\", 0)","category":"page"},{"location":"packages/SeDuMi/#Installation","page":"jump-dev/SeDuMi.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SeDuMi software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"Then, install SeDuMi.jl using Pkg.add:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"import Pkg\nPkg.add(\"SeDuMi\")","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"If you get the error:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"Undefined function or variable 'sedumi'.\n\nError using save\nVariable 'jx_sedumi_arg_out_1' not found.\n\nERROR: LoadError: MATLAB.MEngineError(\"failed to get variable jx_sedumi_arg_out_1 from MATLAB session\")","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The error means that we couldn't find the sedumi function with one output argument using the MATLAB C API.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"This most likely means that you did not add SeDuMi to the MATLAB's path, that is, the toolbox/local/pathdef.m file.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work, where /path/to/sedumi/ is the directory where the sedumi folder is located:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"julia> import MATLAB\n\njulia> cd(\"/path/to/sedumi/\") do\n           MATLAB.mat\"install_sedumi\"\n       end\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"EditURL = \"space_shuttle_reentry_trajectory.jl\"","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial was originally contributed by Henrique Ferrolho.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial demonstrates how to compute a reentry trajectory for the Space Shuttle, by formulating and solving a nonlinear programming problem. The problem was drawn from Chapter 6 of \"Practical Methods for Optimal Control and Estimation Using Nonlinear Programming\", by John T. Betts.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"tip: Tip\nThis tutorial is a more-complicated version of the Rocket Control example. If you are new to solving nonlinear programs in JuMP, you may want to start there instead.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The motion of the vehicle is defined by the following set of DAEs:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\ndoth  = v sin gamma  \ndotphi  = fracvr cos gamma sin psi  cos theta  \ndottheta  = fracvr cos gamma cos psi  \ndotv  = -fracDm - g sin gamma  \ndotgamma  = fracLm v cos(beta) + cos gamma left ( fracvr - fracgv right )  \ndotpsi  = frac1m v cos gamma L sin(beta) + fracvr cos theta cos gamma sin psi sin theta  \nq  le q_U  \nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"where the aerodynamic heating on the vehicle wing leading edge is q = q_a q_r and the dynamic variables are","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n     h  quad textaltitude (ft)     qquad   gamma quad textflight path angle (rad) \n  phi  quad textlongitude (rad)   qquad   psi   quad textazimuth (rad)           \ntheta  quad textlatitude (rad)    qquad   alpha quad textangle of attack (rad)   \n     v  quad textvelocity (ftsec) qquad   beta  quad textbank angle (rad)\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The aerodynamic and atmospheric forces on the vehicle are specified by the following quantities (English units):","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n           D  = frac12 c_D S rho v^2                  qquad  a_0  = -020704 \n           L  = frac12 c_L S rho v^2                  qquad  a_1  =  0029244 \n           g  = mu  r^2                                   qquad  mu  =  014076539 times 10^17 \n           r  = R_e + h                                     qquad  b_0  =  007854 \n        rho  = rho_0 exp-hh_r                         qquad  b_1  = -061592  times 10^-2 \n      rho_0  = 0002378                                    qquad  b_2  =  0621408 times 10^-3 \n         h_r  = 23800                                       qquad  q_r  =  17700 sqrtrho (00001 v)^307 \n         c_L  = a_0 + a_1 hatalpha                      qquad  q_a  =  c_0 + c_1 hatalpha + c_2 hatalpha^2 + c_3 hatalpha^3 \n         c_D  = b_0 + b_1 hatalpha + b_2 hatalpha^2 qquad  c_0  =  10672181 \nhatalpha  = 180 alpha  pi                            qquad  c_1  = -019213774 times 10^-1 \n         R_e  = 20902900                                    qquad  c_2  =  021286289 times 10^-3 \n           S  = 2690                                        qquad  c_3  = -010117249 times 10^-5\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The reentry trajectory begins at an altitude where the aerodynamic forces are quite small with a weight of w = 203000 (lb) and mass m = w  g_0 (slug), where g_0 = 32174 (ft/sec^2). The initial conditions are as follows:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n     h  = 260000 text ft  qquad  v       = 25600 text ftsec \n  phi  = 0      text deg qquad  gamma  = -1    text deg    \ntheta  = 0      text deg qquad  psi    = 90    text deg\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The final point on the reentry trajectory occurs at the unknown (free) time t_F, at the so-called terminal area energy management (TAEM) interface, which is defined by the conditions","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"h = 80000 text ft qquad v = 2500 text ftsec qquad gamma = -5 text deg","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"As explained in the book, our goal is to maximize the final cross-range, which is equivalent to maximizing the final latitude of the vehicle, that is, J = theta(t_F).","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"(Image: Max cross-range shuttle reentry)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Approach","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Approach","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"We will use a discretized model of time, with a fixed number of discretized points, n. The decision variables at each point are going to be the state of the vehicle and the controls commanded to it. In addition, we will also make each time step size Delta t a decision variable; that way, we can either fix the time step size easily, or allow the solver to fine-tune the duration between each adjacent pair of points. Finally, in order to approximate the derivatives of the problem dynamics, we will use either rectangular or trapezoidal integration.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"warning: Warning\nDo not try to actually land a Space Shuttle using this notebook. There's no mesh refinement going on, which can lead to unrealistic trajectories having position and velocity errors with orders of magnitude 10^4 ft and 10^2 ft/sec, respectively.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"using JuMP\nimport Interpolations\nimport Ipopt\n\n# Global variables\nconst w = 203000.0  # weight (lb)\nconst g₀ = 32.174    # acceleration (ft/sec^2)\nconst m = w / g₀    # mass (slug)\n\n# Aerodynamic and atmospheric forces on the vehicle\nconst ρ₀ = 0.002378\nconst hᵣ = 23800.0\nconst Rₑ = 20902900.0\nconst μ = 0.14076539e17\nconst S = 2690.0\nconst a₀ = -0.20704\nconst a₁ = 0.029244\nconst b₀ = 0.07854\nconst b₁ = -0.61592e-2\nconst b₂ = 0.621408e-3\nconst c₀ = 1.0672181\nconst c₁ = -0.19213774e-1\nconst c₂ = 0.21286289e-3\nconst c₃ = -0.10117249e-5\n\n# Initial conditions\nconst h_s = 2.6          # altitude (ft) / 1e5\nconst ϕ_s = deg2rad(0)   # longitude (rad)\nconst θ_s = deg2rad(0)   # latitude (rad)\nconst v_s = 2.56         # velocity (ft/sec) / 1e4\nconst γ_s = deg2rad(-1)  # flight path angle (rad)\nconst ψ_s = deg2rad(90)  # azimuth (rad)\nconst α_s = deg2rad(0)   # angle of attack (rad)\nconst β_s = deg2rad(0)   # bank angle (rad)\nconst t_s = 1.00         # time step (sec)\n\n# Final conditions, the so-called Terminal Area Energy Management (TAEM)\nconst h_t = 0.8          # altitude (ft) / 1e5\nconst v_t = 0.25         # velocity (ft/sec) / 1e4\nconst γ_t = deg2rad(-5)  # flight path angle (rad)\n\n# Number of mesh points (knots) to be used\nconst n = 503\n\n# Integration scheme to be used for the dynamics\nconst integration_rule = \"rectangular\"\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"tip: Choose a good linear solver\nPicking a good linear solver is extremely important to maximize the performance of nonlinear solvers. For the best results, it is advised to experiment different linear solvers.For example, the linear solver MA27 is outdated and can be quite slow. MA57 is a much better alternative, especially for highly-sparse problems (such as trajectory optimization problems).","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"# Uncomment the lines below to pass user options to the solver\nuser_options = (\n# \"mu_strategy\" => \"monotone\",\n# \"linear_solver\" => \"ma27\",\n)\n\n# Create JuMP model, using Ipopt as the solver\nmodel = Model(optimizer_with_attributes(Ipopt.Optimizer, user_options...))\n\n@variables(model, begin\n    0 ≤ scaled_h[1:n]                # altitude (ft) / 1e5\n    ϕ[1:n]                # longitude (rad)\n    deg2rad(-89) ≤ θ[1:n] ≤ deg2rad(89)  # latitude (rad)\n    1e-4 ≤ scaled_v[1:n]                # velocity (ft/sec) / 1e4\n    deg2rad(-89) ≤ γ[1:n] ≤ deg2rad(89)  # flight path angle (rad)\n    ψ[1:n]                # azimuth (rad)\n    deg2rad(-90) ≤ α[1:n] ≤ deg2rad(90)  # angle of attack (rad)\n    deg2rad(-89) ≤ β[1:n] ≤ deg2rad(1)  # bank angle (rad)\n    #        3.5 ≤       Δt[1:n] ≤ 4.5          # time step (sec)\n    Δt[1:n] == 4.0         # time step (sec)\nend);\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"info: Info\nAbove you can find two alternatives for the Δt variables.The first one, 3.5 ≤ Δt[1:n] ≤ 4.5 (currently commented), allows some wiggle room for the solver to adjust the time step size between pairs of mesh points. This is neat because it allows the solver to figure out which parts of the flight require more dense discretization than others. (Remember, the number of discretized points is fixed, and this example does not implement mesh refinement.) However, this makes the problem more complex to solve, and therefore leads to a longer computation time.The second line, Δt[1:n] == 4.0, fixes the duration of every time step to exactly 4.0 seconds. This allows the problem to be solved faster. However, to do this we need to know beforehand that the close-to-optimal total duration of the flight is ~2009 seconds. Therefore, if we split the total duration in slices of 4.0 seconds, we know that we require n = 503 knots to discretize the whole trajectory.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"# Fix initial conditions\nfix(scaled_h[1], h_s; force = true)\nfix(ϕ[1], ϕ_s; force = true)\nfix(θ[1], θ_s; force = true)\nfix(scaled_v[1], v_s; force = true)\nfix(γ[1], γ_s; force = true)\nfix(ψ[1], ψ_s; force = true)\n\n# Fix final conditions\nfix(scaled_h[n], h_t; force = true)\nfix(scaled_v[n], v_t; force = true)\nfix(γ[n], γ_t; force = true)\n\n# Initial guess: linear interpolation between boundary conditions\nx_s = [h_s, ϕ_s, θ_s, v_s, γ_s, ψ_s, α_s, β_s, t_s]\nx_t = [h_t, ϕ_s, θ_s, v_t, γ_t, ψ_s, α_s, β_s, t_s]\ninterp_linear = Interpolations.LinearInterpolation([1, n], [x_s, x_t])\ninitial_guess = mapreduce(transpose, vcat, interp_linear.(1:n))\nset_start_value.(all_variables(model), vec(initial_guess))\n\n# Functions to restore `h` and `v` to their true scale\n@expression(model, h[j = 1:n], scaled_h[j] * 1e5)\n@expression(model, v[j = 1:n], scaled_v[j] * 1e4)\n\n# Helper functions\n@expression(model, c_L[j = 1:n], a₀ + a₁ * rad2deg(α[j]))\n@expression(model, c_D[j = 1:n], b₀ + b₁ * rad2deg(α[j]) + b₂ * rad2deg(α[j])^2)\n@expression(model, ρ[j = 1:n], ρ₀ * exp(-h[j] / hᵣ))\n@expression(model, D[j = 1:n], 0.5 * c_D[j] * S * ρ[j] * v[j]^2)\n@expression(model, L[j = 1:n], 0.5 * c_L[j] * S * ρ[j] * v[j]^2)\n@expression(model, r[j = 1:n], Rₑ + h[j])\n@expression(model, g[j = 1:n], μ / r[j]^2)\n\n# Motion of the vehicle as a differential-algebraic system of equations (DAEs)\n@expression(model, δh[j = 1:n], v[j] * sin(γ[j]))\n@expression(\n    model,\n    δϕ[j = 1:n],\n    (v[j] / r[j]) * cos(γ[j]) * sin(ψ[j]) / cos(θ[j])\n)\n@expression(model, δθ[j = 1:n], (v[j] / r[j]) * cos(γ[j]) * cos(ψ[j]))\n@expression(model, δv[j = 1:n], -(D[j] / m) - g[j] * sin(γ[j]))\n@expression(\n    model,\n    δγ[j = 1:n],\n    (L[j] / (m * v[j])) * cos(β[j]) +\n    cos(γ[j]) * ((v[j] / r[j]) - (g[j] / v[j]))\n)\n@expression(\n    model,\n    δψ[j = 1:n],\n    (1 / (m * v[j] * cos(γ[j]))) * L[j] * sin(β[j]) +\n    (v[j] / (r[j] * cos(θ[j]))) * cos(γ[j]) * sin(ψ[j]) * sin(θ[j])\n)\n\n# System dynamics\nfor j in 2:n\n    i = j - 1  # index of previous knot\n\n    if integration_rule == \"rectangular\"\n        # Rectangular integration\n        @constraint(model, h[j] == h[i] + Δt[i] * δh[i])\n        @constraint(model, ϕ[j] == ϕ[i] + Δt[i] * δϕ[i])\n        @constraint(model, θ[j] == θ[i] + Δt[i] * δθ[i])\n        @constraint(model, v[j] == v[i] + Δt[i] * δv[i])\n        @constraint(model, γ[j] == γ[i] + Δt[i] * δγ[i])\n        @constraint(model, ψ[j] == ψ[i] + Δt[i] * δψ[i])\n    elseif integration_rule == \"trapezoidal\"\n        # Trapezoidal integration\n        @constraint(model, h[j] == h[i] + 0.5 * Δt[i] * (δh[j] + δh[i]))\n        @constraint(model, ϕ[j] == ϕ[i] + 0.5 * Δt[i] * (δϕ[j] + δϕ[i]))\n        @constraint(model, θ[j] == θ[i] + 0.5 * Δt[i] * (δθ[j] + δθ[i]))\n        @constraint(model, v[j] == v[i] + 0.5 * Δt[i] * (δv[j] + δv[i]))\n        @constraint(model, γ[j] == γ[i] + 0.5 * Δt[i] * (δγ[j] + δγ[i]))\n        @constraint(model, ψ[j] == ψ[i] + 0.5 * Δt[i] * (δψ[j] + δψ[i]))\n    else\n        @error \"Unexpected integration rule '$(integration_rule)'\"\n    end\nend\n\n# Objective: Maximize cross-range\n@objective(model, Max, θ[n])\n\nset_silent(model)  # Hide solver's verbose output\noptimize!(model)  # Solve for the control and state\n@assert termination_status(model) == LOCALLY_SOLVED\n\n# Show final cross-range of the solution\nprintln(\n    \"Final latitude θ = \",\n    round(objective_value(model) |> rad2deg; digits = 2),\n    \"°\",\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Plotting-the-results","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Plotting the results","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"Let's plot the results to visualize the optimal trajectory:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"using Plots\nts = cumsum([0; value.(Δt)])[1:end-1]\nplt_altitude = plot(\n    ts,\n    value.(scaled_h);\n    legend = nothing,\n    title = \"Altitude (100,000 ft)\",\n)\nplt_longitude =\n    plot(ts, rad2deg.(value.(ϕ)); legend = nothing, title = \"Longitude (deg)\")\nplt_latitude =\n    plot(ts, rad2deg.(value.(θ)); legend = nothing, title = \"Latitude (deg)\")\nplt_velocity = plot(\n    ts,\n    value.(scaled_v);\n    legend = nothing,\n    title = \"Velocity (1000 ft/sec)\",\n)\nplt_flight_path =\n    plot(ts, rad2deg.(value.(γ)); legend = nothing, title = \"Flight Path (deg)\")\nplt_azimuth =\n    plot(ts, rad2deg.(value.(ψ)); legend = nothing, title = \"Azimuth (deg)\")\n\nplot(\n    plt_altitude,\n    plt_velocity,\n    plt_longitude,\n    plt_flight_path,\n    plt_latitude,\n    plt_azimuth;\n    layout = grid(3, 2),\n    linewidth = 2,\n    size = (700, 700),\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"function q(h, v, a)\n    ρ(h) = ρ₀ * exp(-h / hᵣ)\n    qᵣ(h, v) = 17700 * √ρ(h) * (0.0001 * v)^3.07\n    qₐ(a) = c₀ + c₁ * rad2deg(a) + c₂ * rad2deg(a)^2 + c₃ * rad2deg(a)^3\n    # Aerodynamic heating on the vehicle wing leading edge\n    return qₐ(a) * qᵣ(h, v)\nend\n\nplt_attack_angle = plot(\n    ts[1:end-1],\n    rad2deg.(value.(α)[1:end-1]);\n    legend = nothing,\n    title = \"Angle of Attack (deg)\",\n)\nplt_bank_angle = plot(\n    ts[1:end-1],\n    rad2deg.(value.(β)[1:end-1]);\n    legend = nothing,\n    title = \"Bank Angle (deg)\",\n)\nplt_heating = plot(\n    ts,\n    q.(value.(scaled_h) * 1e5, value.(scaled_v) * 1e4, value.(α));\n    legend = nothing,\n    title = \"Heating (BTU/ft/ft/sec)\",\n)\n\nplot(\n    plt_attack_angle,\n    plt_bank_angle,\n    plt_heating;\n    layout = grid(3, 1),\n    linewidth = 2,\n    size = (700, 700),\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"plot(\n    rad2deg.(value.(ϕ)),\n    rad2deg.(value.(θ)),\n    value.(scaled_h);\n    linewidth = 2,\n    legend = nothing,\n    title = \"Space Shuttle Reentry Trajectory\",\n    xlabel = \"Longitude (deg)\",\n    ylabel = \"Latitude (deg)\",\n    zlabel = \"Altitude (100,000 ft)\",\n)","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"background/algebraic_modeling_languages/#algebraic-modeling-language","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP is an algebraic modeling language for mathematical optimization written in the Julia language. In this page, we explain what an algebraic modeling language actually is.","category":"page"},{"location":"background/algebraic_modeling_languages/#What-is-an-algebraic-modeling-language?","page":"Algebraic modeling languages","title":"What is an algebraic modeling language?","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"If you have taken a class in mixed-integer linear programming, you will have seen a formulation like:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmin       c^top x \ntextst  A x = b  \n             x ge 0  \n             x_i in mathbbZ quad forall i in mathcalI\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"where c, A, and b are appropriately sized vectors and matrices of data, and mathcalI denotes the set of variables that are integer.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Solvers expect problems in a standard form like this because it limits the types of constraints that they need to consider. This makes writing a solver much easier.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"info: What is a solver?\nA solver is a software package that computes solutions to one or more classes of problems.For example, HiGHS is a solver for linear programming (LP) and mixed integer programming (MIP) problems. It incorporates algorithms such as the simplex method and the interior-point method.JuMP currently supports a number of open-source and commercial solvers, which can be viewed in the Supported-solvers table.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Despite the textbook view of a linear program, you probably formulated problems algebraically like so:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmax       sumlimits_i = 1^n c_i x_i                   \ntextst  sumlimits_i = 1^n w_i x_i le b             \n             x_i ge 0 quad forall i = 1ldotsn          \n             x_i in mathbbZ quad forall i = 1ldotsn\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"info: Info\nDo you recognize this formulation? It's the knapsack problem.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Users prefer to write problems in algebraic form because it is more convenient. For example, we used le b, even though the standard form only supported constraints of the form Ax = b.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"We could convert our knapsack problem into the standard form by adding a new slack variable x_0:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmax       sumlimits_i = 1^n c_i x_i            \ntextst  x_0 + sumlimits_i = 1^n w_i x_i = b  \n             x_i ge 0 quad forall i = 0ldotsn   \n             x_i in mathbbZ quad forall i = 1ldotsn\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"However, as models get more complicated, this manual conversion becomes more and more error-prone.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"An algebraic modeling language is a tool that simplifies the translation between the algebraic form of the modeler, and the standard form of the solver.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Each algebraic modeling language has two main parts:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"A domain specific language for the user to write down problems in algebraic form.\nA converter from the algebraic form into a standard form supported by the solver (and back again).","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Part 2 is less trivial than it might seem, because each solver has a unique application programming interface (API) and data structure for representing optimization models and obtaining results.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP uses the MathOptInterface.jl package to abstract these differences between solvers.","category":"page"},{"location":"background/algebraic_modeling_languages/#What-is-MathOptInterface?","page":"Algebraic modeling languages","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"MathOptInterface (MOI) is an abstraction layer designed to provide an interface to mathematical optimization solvers so that users do not need to understand multiple solver-specific APIs. MOI can be used directly, or through a higher-level modeling interface like JuMP.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"There are three main parts to MathOptInterface:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"A solver-independent API that abstracts concepts such as adding and deleting variables and constraints, setting and getting parameters, and querying results. For more information on the MathOptInterface API, read the documentation.\nAn automatic rewriting system based on equivalent formulations of a constraint. For more information on this rewriting system, read the LazyBridgeOptimizer section of the manual, and our paper on arXiv.\nUtilities for managing how and when models are copied to solvers. For more information on this, read the CachingOptimizer section of the manual.","category":"page"},{"location":"background/algebraic_modeling_languages/#From-user-to-solver","page":"Algebraic modeling languages","title":"From user to solver","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"This section provides a brief summary of the steps that happen in order to translate the model that the user writes into a model that the solver understands.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-I:-writing-in-algebraic-form","page":"Algebraic modeling languages","title":"Step I: writing in algebraic form","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP provides the first part of an algebraic modeling language using the @variable, @objective, and @constraint macros.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"For example, here's how we write the knapsack problem in JuMP:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using JuMP, HiGHS\n\njulia> function algebraic_knapsack(c, w, b)\n           n = length(c)\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x[1:n] >= 0, Int)\n           @objective(model, Max, sum(c[i] * x[i] for i = 1:n))\n           @constraint(model, sum(w[i] * x[i] for i = 1:n) <= b)\n           optimize!(model)\n           return value.(x)\n       end\nalgebraic_knapsack (generic function with 1 method)\n\njulia> algebraic_knapsack([1, 2], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"This formulation is compact, and it closely matches the algebraic formulation of the model we wrote out above.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-II:-algebraic-to-functional","page":"Algebraic modeling languages","title":"Step II: algebraic to functional","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"For the next step, JuMP's macros re-write the variables and constraints into a functional form. Here's what the JuMP code looks like after this step:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using JuMP, HiGHS\n\njulia> function nonalgebraic_knapsack(c, w, b)\n           n = length(c)\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           x = [VariableRef(model) for i = 1:n]\n           for i = 1:n\n               set_lower_bound(x[i], 0)\n               set_integer(x[i])\n               set_name(x[i], \"x[$i]\")\n           end\n           obj = AffExpr(0.0)\n           for i = 1:n\n               add_to_expression!(obj, c[i], x[i])\n           end\n           set_objective(model, MAX_SENSE, obj)\n           lhs = AffExpr(0.0)\n           for i = 1:n\n               add_to_expression!(lhs, w[i], x[i])\n           end\n           con = build_constraint(error, lhs, MOI.LessThan(b))\n           add_constraint(model, con)\n           optimize!(model)\n           return value.(x)\n       end\nnonalgebraic_knapsack (generic function with 1 method)\n\njulia> nonalgebraic_knapsack([1, 2], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Hopefully you agree that the macro version is much easier to read.","category":"page"},{"location":"background/algebraic_modeling_languages/#Part-III:-JuMP-to-MathOptInterface","page":"Algebraic modeling languages","title":"Part III: JuMP to MathOptInterface","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"In the third step, JuMP converts the functional form of the problem, that is, nonalgebraic_knapsack, into the MathOptInterface API:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> import MathOptInterface as MOI\n\njulia> import HiGHS\n\njulia> function mathoptinterface_knapsack(optimizer, c, w, b)\n           n = length(c)\n           model = MOI.instantiate(optimizer)\n           MOI.set(model, MOI.Silent(), true)\n           x = MOI.add_variables(model, n)\n           for i in 1:n\n               MOI.add_constraint(model, x[i], MOI.GreaterThan(0.0))\n               MOI.add_constraint(model, x[i], MOI.Integer())\n               MOI.set(model, MOI.VariableName(), x[i], \"x[$i]\")\n           end\n           MOI.set(model, MOI.ObjectiveSense(), MOI.MAX_SENSE)\n           obj = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(c, x), 0.0)\n           MOI.set(model, MOI.ObjectiveFunction{typeof(obj)}(), obj)\n           MOI.add_constraint(\n               model,\n               MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(w, x), 0.0),\n               MOI.LessThan(b),\n           )\n           MOI.optimize!(model)\n           return MOI.get.(model, MOI.VariablePrimal(), x)\n       end\nmathoptinterface_knapsack (generic function with 1 method)\n\njulia> mathoptinterface_knapsack(HiGHS.Optimizer, [1.0, 2.0], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"The code is becoming more verbose and looking less like the mathematical formulation that we started with.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-IV:-MathOptInterface-to-HiGHS","page":"Algebraic modeling languages","title":"Step IV: MathOptInterface to HiGHS","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"As a final step, the HiGHS.jl package converts the MathOptInterface form, that is, mathoptinterface_knapsack, into a HiGHS-specific API:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using HiGHS\n\njulia> function highs_knapsack(c, w, b)\n           n = length(c)\n           model = Highs_create()\n           Highs_setBoolOptionValue(model, \"output_flag\", false)\n           for i in 1:n\n               Highs_addCol(model, c[i], 0.0, Inf, 0, C_NULL, C_NULL)\n               Highs_changeColIntegrality(model, i-1, 1)\n           end\n           Highs_changeObjectiveSense(model, -1)\n           Highs_addRow(\n               model,\n               -Inf,\n               b,\n               Cint(length(w)),\n               collect(Cint(0):Cint(n-1)),\n               w,\n           )\n           Highs_run(model)\n           x = fill(NaN, 2)\n           Highs_getSolution(model, x, C_NULL, C_NULL, C_NULL)\n           Highs_destroy(model)\n           return x\n       end\nhighs_knapsack (generic function with 1 method)\n\njulia> highs_knapsack([1.0, 2.0], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"We've now gone from a algebraic model that looked identical to the mathematical model we started with, to a verbose function that uses HiGHS-specific functionality.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"The difference between algebraic_knapsack and highs_knapsack highlights the benefit that algebraic modeling languages provide to users. Moreover, if we used a different solver, the solver-specific function would be entirely different. A key benefit of an algebraic modeling language is that you can change the solver without needing to rewrite the model.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"EditURL = \"https://github.com/jump-dev/AmplNLWriter.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/AmplNLWriter/#AmplNLWriter.jl","page":"jump-dev/AmplNLWriter.jl","title":"AmplNLWriter.jl","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"(Image: Build Status) (Image: MINLPTests) (Image: codecov)","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"AmplNLWriter.jl is an interface between MathOptInterface.jl and AMPL-enabled solvers.","category":"page"},{"location":"packages/AmplNLWriter/#Affiliation","page":"jump-dev/AmplNLWriter.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"This wrapper is maintained by the JuMP community and has no official connection with the AMPL modeling language or AMPL Optimization Inc.","category":"page"},{"location":"packages/AmplNLWriter/#Installation","page":"jump-dev/AmplNLWriter.jl","title":"Installation","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Install AmplNLWriter using Pkg.add:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"import Pkg\nPkg.add(\"AmplNLWriter\")","category":"page"},{"location":"packages/AmplNLWriter/#Use-with-JuMP","page":"jump-dev/AmplNLWriter.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"AmplNLWriter requires an AMPL compatible solver binary to function.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Pass a string pointing to any AMPL-compatible solver binary as the first positional argument to AmplNLWriter.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"For example, if the bonmin executable is on the system path, use:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter\nmodel = Model(() -> AmplNLWriter.Optimizer(\"bonmin\"))","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"If the solver is not on the system path, pass the full path to the solver:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter\nmodel = Model(() -> AmplNLWriter.Optimizer(\"/Users/Oscar/ampl.macos64/bonmin\"))","category":"page"},{"location":"packages/AmplNLWriter/#Precompiled-binaries","page":"jump-dev/AmplNLWriter.jl","title":"Precompiled binaries","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"To simplify the process of installing solver binaries, a number of Julia packages provide precompiled binaries that are compatible with AmplNLWriter. These are generally the name of the solver, followed by _jll. For example, bomin is provided by the Bonmin_jll package.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"To call Bonmin via AmplNLWriter.jl, install the Bonmin_jll package, then run:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter, Bonmin_jll\nmodel = Model(() -> AmplNLWriter.Optimizer(Bonmin_jll.amplexe))","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Supported packages include:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Solver Julia Package Executable\nBonmin Bonmin_jll.jl Bomin_jll.amplexe\nCouenne Couenne_jll.jl Couenne_jll.amplexe\nIpopt Ipopt_jll.jl Ipopt_jll.amplexe\nSHOT SHOT_jll.jl SHOT_jll.amplexe\nKNITRO KNITRO.jl KNITRO.amplexe","category":"page"},{"location":"packages/AmplNLWriter/#MathOptInterface-API","page":"jump-dev/AmplNLWriter.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"The AmplNLWriter optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.Reals","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.NLPBlock()\nMOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Note that some solver executables may not support the full list of constraint types. For example, Ipopt_jll does not support MOI.Integer or MOI.ZeroOne constraints.","category":"page"},{"location":"packages/AmplNLWriter/#Options","page":"jump-dev/AmplNLWriter.jl","title":"Options","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"A list of available options for each solver can be found here:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Bonmin (plus Ipopt options)\nCouenne (plus Ipopt and Bonmin options)\nIpopt\nSHOT","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Set an option using set_attribute. For example, to set the \"bonmin.nlp_log_level\" option to 0 in Bonmin, use:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP\nimport AmplNLWriter\nimport Bonmin_jll\nmodel = Model(() -> AmplNLWriter.Optimizer(Bonmin_jll.amplexe))\nset_attribute(model, \"bonmin.nlp_log_level\", 0)","category":"page"},{"location":"packages/AmplNLWriter/#opt-files","page":"jump-dev/AmplNLWriter.jl","title":"opt files","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Some options need to be specified via an .opt file.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"This file must be located in the current working directory whenever the model is solved.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"The .opt file must be named after the name of the solver, for example, bonmin.opt, and each line must contain an option name and the desired value, separated by a space.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"For example, to set the absolute and relative tolerances in Couenne to 1 and 0.05 respectively, the couenne.opt file should contain:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"allowable_gap 1\nallowable_fraction_gap 0.05","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"<img class=\"display-light-only\" src=\"assets/logo-with-text.svg\" alt=\"JuMP logo\"/>\n<img class=\"display-dark-only\" src=\"assets/logo-dark-with-text.svg\" alt=\"JuMP logo\"/>","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"# These comments do not display in the HTML output.\n# See https://github.com/JuliaDocs/Documenter.jl/issues/674.","category":"page"},{"location":"#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"\\begin{figure}[h]\n  \\includegraphics[width=0.5\\textwidth]{assets/logo-with-text.pdf}\n\\end{figure}","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Welcome to the documentation for JuMP.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"note: Note\nThis documentation is also available in PDF format: JuMP.pdf.","category":"page"},{"location":"#What-is-JuMP?","page":"Introduction","title":"What is JuMP?","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is a domain-specific modeling language for mathematical optimization embedded in Julia. It currently supports a number of open-source and commercial solvers for a variety of problem classes, including linear, mixed-integer, second-order conic, semidefinite, and nonlinear programming.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"tip: Tip\nIf you aren't sure if you should use JuMP, read Should you use JuMP?.","category":"page"},{"location":"#Resources-for-getting-started","page":"Introduction","title":"Resources for getting started","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"There are few ways to get started with JuMP:","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Read the Installation Guide.\nRead the introductory tutorials Getting started with Julia and Getting started with JuMP.\nBrowse some of our modeling tutorials, including classics such as The diet problem, or the Maximum likelihood estimation problem using nonlinear programming.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"tip: Tip\nNeed help? Join the community forum to search for answers to commonly asked questions.Before asking a question, make sure to read the post make it easier to help you, which contains a number of tips on how to ask a good question.","category":"page"},{"location":"#How-the-documentation-is-structured","page":"Introduction","title":"How the documentation is structured","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Tutorials contain worked examples of solving problems with JuMP. Start here if you are new to JuMP, or you have a particular problem class you want to model.\nThe Manual contains short code-snippets that explain how to achieve specific tasks in JuMP. Look here if you want to know how to achieve a particular task, such as how to Delete a variable or how to Modify an objective coefficient.\nThe API Reference contains a complete list of the functions you can use in JuMP. Look here if you want to know how to use a particular function.\nThe Background information section contains background reading material to provide context to JuMP. Look here if you want an understanding of what JuMP is and why we created it, rather than how to use it.\nThe Developer docs section contains information for people contributing to JuMP development or writing JuMP extensions. Don't worry about this section if you are using JuMP to formulate and solve problems as a user.\nThe MathOptInterface section is a self-contained copy of the documentation for MathOptInterface. Look here for functions and constants beginning with MOI., as well as for general information on how MathOptInterface works.","category":"page"},{"location":"#Citing-JuMP","page":"Introduction","title":"Citing JuMP","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"If you find JuMP useful in your work, we kindly request that you cite the following paper (preprint):","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"@article{Lubin2023,\n    author = {Miles Lubin and Oscar Dowson and Joaquim {Dias Garcia} and Joey Huchette and Beno{\\^i}t Legat and Juan Pablo Vielma},\n    title = {{JuMP} 1.0: {R}ecent improvements to a modeling language for mathematical optimization},\n    journal = {Mathematical Programming Computation},\n    year = {2023},\n    doi = {10.1007/s12532-023-00239-3}\n}","category":"page"},{"location":"#NumFOCUS","page":"Introduction","title":"NumFOCUS","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"(Image: NumFOCUS logo)","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is a Sponsored Project of NumFOCUS, a 501(c)(3) nonprofit charity in the United States. NumFOCUS provides JuMP with fiscal, legal, and administrative support to help ensure the health and sustainability of the project. Visit numfocus.org for more information.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"You can support JuMP by donating.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Donations to JuMP are managed by NumFOCUS. For donors in the United States, your gift is tax-deductible to the extent provided by law. As with any donation, you should consult with your tax adviser about your particular tax situation.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP's largest expense is the annual JuMP-dev workshop. Donations will help us provide travel support for JuMP-dev attendees and take advantage of other opportunities that arise to support JuMP development.","category":"page"},{"location":"#License","page":"Introduction","title":"License","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is licensed under the MPL-2.0 software license. Consult the license and the Mozilla FAQ for more information. In addition, JuMP is typically used in conjunction with solver packages and extensions which have their own licences. Consult their package repositories for the specific licenses that apply.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"EditURL = \"https://github.com/infiniteopt/InfiniteOpt.jl/blob/v0.5.8/docs/jump/README.md\"","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"(Image: Logo)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt.jl is a JuMP extension for expressing and solving infinite-dimensional optimization problems. Such areas include stochastic programming, dynamic programming, space-time optimization, and more. InfiniteOpt serves as an easy-to-use modeling interface for these advanced problem types that can be used by those with little background in these areas. It also it contains a wealth of capabilities making it a powerful and convenient tool for advanced users.  ","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Current Version Documentation Build Status Citation\n(Image: ) (Image: ) (Image: Build Status) (Image: codecov.io) (Image: DOI)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt builds upon JuMP to add support for many complex modeling objects which  include:","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Infinite parameters (for example, time, space, and/or uncertainty)\nFinite parameters (similar to ParameterJuMP)\nInfinite variables (decision functions) (for example, y(t x))\nDerivatives (for example, fracpartial y(t x)partial t)\nMeasures (for example, int_t in Ty(t x)dt and mathbbEy(xi))\nFirst class nonlinear modeling","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"The unifying modeling abstraction behind InfiniteOpt captures a wide spectrum  of disciplines which include dynamic, PDE, stochastic, and semi-infinite  optimization. Moreover, we facilitate transferring techniques between these  to synthesize new optimization paradigms.","category":"page"},{"location":"packages/InfiniteOpt/#License","page":"infiniteopt/InfiniteOpt.jl","title":"License","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt is licensed under the MIT \"Expat\" license.","category":"page"},{"location":"packages/InfiniteOpt/#Installation","page":"infiniteopt/InfiniteOpt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt.jl is a registered Julia package and  can be installed by entering the following in the REPL.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"julia> import Pkg; Pkg.add(\"InfiniteOpt\")","category":"page"},{"location":"packages/InfiniteOpt/#Documentation","page":"infiniteopt/InfiniteOpt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Please visit our documentation pages  to learn more. These pages are quite extensive and feature overviews, guides, manuals, tutorials, examples, and more.","category":"page"},{"location":"packages/InfiniteOpt/#Questions","page":"infiniteopt/InfiniteOpt.jl","title":"Questions","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"For additional help please visit and post in our  discussion forum.","category":"page"},{"location":"packages/InfiniteOpt/#Citing","page":"infiniteopt/InfiniteOpt.jl","title":"Citing","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"(Image: DOI)  (Image: DOI)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"If you use InfiniteOpt.jl in your research, we would greatly appreciate your  citing it.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"@article{pulsipher2022unifying,\n      title = {A unifying modeling abstraction for infinite-dimensional optimization},\n      journal = {Computers & Chemical Engineering},\n      volume = {156},\n      year = {2022},\n      issn = {0098-1354},\n      doi = {https://doi.org/10.1016/j.compchemeng.2021.107567},\n      url = {https://www.sciencedirect.com/science/article/pii/S0098135421003458},\n      author = {Joshua L. Pulsipher and Weiqi Zhang and Tyler J. Hongisto and Victor M. Zavala},\n}","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"A pre-print version is freely available though arXiv.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"EditURL = \"getting_started_with_sets_and_indexing.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Getting-started-with-sets-and-indexing","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Most introductory courses to linear programming will teach you to identify sets over which the decision variables and constraints are indexed. Therefore, it is common to write variables such as x_i for all i in I.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"A common stumbling block for new users to JuMP is that JuMP does not provide specialized syntax for constructing and manipulating these sets.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We made this decision because Julia already provides a wealth of data structures for working with sets.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"In contrast, because tools like AMPL are stand-alone software packages, they had to define their own syntax for set construction and manipulation. Indeed, the AMPL Book has two entire chapters devoted to sets and indexing (Chapter 5, \"Simple Sets and Indexing,\" and Chapter 6, \"Compound Sets and Indexing\").","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The purpose of this tutorial is to demonstrate a variety of ways in which you can construct and manipulate sets for optimization models.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"If you haven't already, you should first read Getting started with JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"using JuMP","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Unordered-sets","page":"Getting started with sets and indexing","title":"Unordered sets","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Unordered sets are useful to describe non-numeric indices, such as the names of cities or types of products.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The most common way to construct a set is by creating a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"animals = [\"dog\", \"cat\", \"chicken\", \"cow\", \"pig\"]\nmodel = Model()\n@variable(model, x[animals])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We can also use things like the keys of a dictionary:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"weight_of_animals = Dict(\n    \"dog\" => 20.0,\n    \"cat\" => 5.0,\n    \"chicken\" => 2.0,\n    \"cow\" => 720.0,\n    \"pig\" => 150.0,\n)\nanimal_keys = keys(weight_of_animals)\nmodel = Model()\n@variable(model, x[animal_keys])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"A third option is to use Julia's Set object.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"animal_set = Set()\nfor animal in keys(weight_of_animals)\n    push!(animal_set, animal)\nend\nanimal_set","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The nice thing about Sets is that they automatically remove duplicates:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"push!(animal_set, \"dog\")\nanimal_set","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Note how dog does not appear twice.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[animal_set])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-of-numbers","page":"Getting started with sets and indexing","title":"Sets of numbers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Sets of numbers are useful to describe sets that are ordered, such as years or elements in a vector. The easiest way to create sets of numbers is to use Julia's range syntax.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"These can start at 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[1:4])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"but they don't have to:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[2012:2021])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Ranges also have a start:step:stop syntax. So the Olympic years are:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[1896:4:2020])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-of-other-things","page":"Getting started with sets and indexing","title":"Sets of other things","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"An important observation is that you can have any Julia type as the element of a set. It doesn't have to be a String or a Number. For example, you can have tuples:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"sources = [\"A\", \"B\", \"C\"]\nsinks = [\"D\", \"E\"]\nS = [(source, sink) for source in sources, sink in sinks]\nmodel = Model()\n@variable(model, x[S])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"x[(\"A\", \"D\")]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"For multi-dimensional sets, you can use JuMP's syntax for constructing Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[sources, sinks])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"x[\"A\", \"D\"]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"info: Info\nNote how we indexed x[\"A\", \"D\"] instead of x[(\"A\", \"D\")] as above.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-to-watch-out-for","page":"Getting started with sets and indexing","title":"Sets to watch out for","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"JuMP supports any sets which are iterable, that is, the set set supports a for-loop like: [i for i in set]. This causes a few common errors.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"First, if T = 3, you may pass the integer T by mistake instead of a range like 1:T:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\nT = 3\n@variable(model, x[T])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This results in a single variable being created, instead of three as desired. Because this is a common error, a warning is printed, advising you to pass a Vector{Int} instead:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, x_fixed[[T]])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Second, because Strings are iterable, passing a \"index\" as a singleton index is the same as passing ['i', 'n', 'd', 'e', 'x']:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, y[\"index\"])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This time, a warning is not printed, but the work-around is similar, pass a Vector{String} instead:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, y_fixed[[\"index\"]])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"tip: Tip\nAs a rule of thumb, if you want an index with one element, avoid confusion by passing [index] instead of index.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Set-operations","page":"Getting started with sets and indexing","title":"Set operations","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Julia has built-in support for set operations such as union, intersect, and setdiff.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Therefore, to create a set of all years in which the summer Olympics were held, we can use:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"baseline = 1896:4:2020\ncancelled = [1916, 1940, 1944, 2020]\noff_year = [2021]\nolympic_years = union(setdiff(baseline, cancelled), off_year)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"You can also find the number of elements (that is, the cardinality) in a set using length:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"length(olympic_years)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Set-membership-operations","page":"Getting started with sets and indexing","title":"Set membership operations","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"To compute membership of sets, use the in function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"2000 in olympic_years","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"2001 in olympic_years","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Indexing-expressions","page":"Getting started with sets and indexing","title":"Indexing expressions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Use Julia's generator syntax to compute new sets, such as the list of Olympic years that are divisible by 3:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"olympic_3_years = [year for year in olympic_years if mod(year, 3) == 0]\nmodel = Model()\n@variable(model, x[olympic_3_years])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Alternatively, use JuMP's syntax for constructing Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[year in olympic_years; mod(year, 3) == 0])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Compound-sets","page":"Getting started with sets and indexing","title":"Compound sets","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Consider a transportation problem in which we need to ship goods between cities. We have been provided a list of cities:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"cities = [\"Auckland\", \"Wellington\", \"Christchurch\", \"Dunedin\"]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"and a distance matrix which records the shipping distance between pairs of cities. If we can't ship between two cities, the distance is 0.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"distances = [0 643 1071 1426; 0 0 436 790; 0 0 0 360; 1426 0 0 0]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Let's have a look at ways we could write a model with an objective function to minimize the total shipping cost. For simplicity, we'll ignore all constraints.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Fix-unused-variables","page":"Getting started with sets and indexing","title":"Fix unused variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"One approach is to fix all variables that we can't use to zero. Most solvers are smart-enough to remove these during a presolve phase, so it has a very small impact on performance:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[1:N, 1:N] >= 0)\nfor i in 1:N, j in 1:N\n    if distances[i, j] == 0\n        fix(x[i, j], 0.0; force = true)\n    end\nend\n@objective(model, Min, sum(distances[i, j] * x[i, j] for i in 1:N, j in 1:N))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Filtered-summation","page":"Getting started with sets and indexing","title":"Filtered summation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Another approach is to define filters whenever we want to sum over our decision variables:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[1:N, 1:N] >= 0)\n@objective(\n    model,\n    Min,\n    sum(\n        distances[i, j] * x[i, j] for i in 1:N, j in 1:N if distances[i, j] > 0\n    ),\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Filtered-indexing","page":"Getting started with sets and indexing","title":"Filtered indexing","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We could also use JuMP's support for Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[i = 1:N, j = 1:N; distances[i, j] > 0])\n@objective(model, Min, sum(distances[i...] * x[i] for i in eachindex(x)))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"note: Note\nThe i... is called  a \"splat.\" It converts a tuple like (1, 2) into two indices like distances[1, 2].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Converting-to-a-different-data-structure","page":"Getting started with sets and indexing","title":"Converting to a different data structure","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Another approach, and one that is often the most readable, is to convert the data you have into something that is easier to work with. Originally, we had a vector of strings and a matrix of distances. What we really need is something that maps usable origin-destination pairs to distances. A dictionary is an obvious choice:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"routes = Dict(\n    (a, b) => distances[i, j] for\n    (i, a) in enumerate(cities), (j, b) in enumerate(cities) if\n    distances[i, j] > 0\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Then, we can create our model like so:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[keys(routes)])\n@objective(model, Min, sum(v * x[k] for (k, v) in routes))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This has a number of benefits over the other approaches, including a compacter algebraic model and variables that are named in a more meaningful way.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"tip: Tip\nIf you're struggling to formulate a problem using the available syntax in JuMP, it's probably a sign that you should convert your data into a different form.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Next-steps","page":"Getting started with sets and indexing","title":"Next steps","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The purpose of this tutorial was to show how JuMP does not have specialized syntax for set creation and manipulation. Instead, you should use the tools provided by Julia itself.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This is both an opportunity and a challenge, because you are free to pick the syntax and data structures that best suit your problem, but for new users it can be daunting to decide which structure to use.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Read through some of the other JuMP tutorials to get inspiration and ideas for how you can use Julia's syntax and data structures to your advantage.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"EditURL = \"https://github.com/jump-dev/Gurobi.jl/blob/v1.0.4/README.md\"","category":"page"},{"location":"packages/Gurobi/#Gurobi.jl","page":"jump-dev/Gurobi.jl","title":"Gurobi.jl","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi.jl is a wrapper for the Gurobi Optimizer.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"It has two components:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Gurobi/#Affiliation","page":"jump-dev/Gurobi.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This wrapper is maintained by the JuMP community with help from Gurobi. ","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you encounter a problem with this interface, please either open an issue in this repository directly or create a topic in the Julia Discourse with the gurobi tag. ","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you encounter a problem with the Gurobi solver, please post in Gurobi’s Community Forum, or if you are a commercial customer, please contact Gurobi directly through the Gurobi Help Center.","category":"page"},{"location":"packages/Gurobi/#License","page":"jump-dev/Gurobi.jl","title":"License","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Free Gurobi licenses are available for academics and students.","category":"page"},{"location":"packages/Gurobi/#Installation","page":"jump-dev/Gurobi.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"First, obtain a license of Gurobi and install Gurobi solver.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Then, set the GUROBI_HOME environment variable as appropriate and run Pkg.add(\"Gurobi\"):","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"# On Windows, this might be\nENV[\"GUROBI_HOME\"] = \"C:\\\\Program Files\\\\gurobi1000\\\\win64\"\n# ... or perhaps ...\nENV[\"GUROBI_HOME\"] = \"C:\\\\gurobi1000\\\\win64\"\n# On Mac, this might be\nENV[\"GUROBI_HOME\"] = \"/Library/gurobi1000/mac64\"\n\nimport Pkg\nPkg.add(\"Gurobi\")","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Note: your path may differ. Check which folder you installed Gurobi in, and update the path accordingly.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"By default, building Gurobi.jl will fail if the Gurobi library is not found. This may not be desirable in certain cases, for example when part of a package's test suite uses Gurobi as an optional test dependency, but Gurobi cannot be installed on a CI server running the test suite. To support this use case, the GUROBI_JL_SKIP_LIB_CHECK environment variable may be set (to any value) to make Gurobi.jl installable (but not usable).","category":"page"},{"location":"packages/Gurobi/#Use-with-JuMP","page":"jump-dev/Gurobi.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"To use Gurobi with JuMP, use Gurobi.Optimizer:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nmodel = Model(Gurobi.Optimizer)\nset_attribute(model, \"TimeLimit\", 100)\nset_attribute(model, \"Presolve\", 0)","category":"page"},{"location":"packages/Gurobi/#MathOptInterface-API","page":"jump-dev/Gurobi.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The Gurobi optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}\nMOI.ObjectiveFunction{MOI.VectorAffineFunction{Float64}}","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/Gurobi/#Options","page":"jump-dev/Gurobi.jl","title":"Options","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi Documentation for a list and description of allowable parameters.","category":"page"},{"location":"packages/Gurobi/#C-API","page":"jump-dev/Gurobi.jl","title":"C API","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The C API can be accessed via Gurobi.GRBxx functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi documentation for details.","category":"page"},{"location":"packages/Gurobi/#Reusing-the-same-Gurobi-environment-for-multiple-solves","page":"jump-dev/Gurobi.jl","title":"Reusing the same Gurobi environment for multiple solves","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"When using this package via other packages such as JuMP.jl, the default behavior is to obtain a new Gurobi license token every time a model is created. If you are using Gurobi in a setting where the number of concurrent Gurobi uses is limited (for example, \"Single-Use\" or \"Floating-Use\" licenses), you might instead prefer to obtain a single license token that is shared by all models that your program solves.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"You can do this by passing a Gurobi.Env() object as the first parameter to Gurobi.Optimizer. For example:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nconst GRB_ENV = Gurobi.Env()\n\nmodel_1 = Model(() -> Gurobi.Optimizer(GRB_ENV))\n\n# The solvers can have different options too\nmodel_2 = direct_model(Gurobi.Optimizer(GRB_ENV))\nset_attribute(model_2, \"OutputFlag\", 0)","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you create a module with a Gurobi.Env as a module-level constant, use an __init__ function to ensure that a new environment is created each time the module is loaded:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"module MyModule\n\nimport Gurobi\n\nconst GRB_ENV_REF = Ref{Gurobi.Env}()\n\nfunction __init__()\n    global GRB_ENV_REF\n    GRB_ENV_REF[] = Gurobi.Env()\n    return\nend\n\n# Note the need for GRB_ENV_REF[] not GRB_ENV_REF\ncreate_optimizer() = Gurobi.Optimizer(GRB_ENV_REF[])\n\nend","category":"page"},{"location":"packages/Gurobi/#Accessing-Gurobi-specific-attributes","page":"jump-dev/Gurobi.jl","title":"Accessing Gurobi-specific attributes","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Get and set Gurobi-specific variable, constraint, and model attributes as follows:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nmodel = direct_model(Gurobi.Optimizer())\n@variable(model, x >= 0)\n@constraint(model, c, 2x >= 1)\n@objective(model, Min, x)\nMOI.set(model, Gurobi.ConstraintAttribute(\"Lazy\"), c, 2)\noptimize!(model)\nMOI.get(model, Gurobi.VariableAttribute(\"LB\"), x)  # Returns 0.0\nMOI.get(model, Gurobi.ModelAttribute(\"NumConstrs\")) # Returns 1","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"A complete list of supported Gurobi attributes can be found in their online documentation.","category":"page"},{"location":"packages/Gurobi/#Callbacks","page":"jump-dev/Gurobi.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Here is an example using Gurobi's solver-specific callbacks.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi, Test\n\nmodel = direct_model(Gurobi.Optimizer())\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\ncb_calls = Cint[]\nfunction my_callback_function(cb_data, cb_where::Cint)\n    # You can reference variables outside the function as normal\n    push!(cb_calls, cb_where)\n    # You can select where the callback is run\n    if cb_where != GRB_CB_MIPSOL && cb_where != GRB_CB_MIPNODE\n        return\n    end\n    # You can query a callback attribute using GRBcbget\n    if cb_where == GRB_CB_MIPNODE\n        resultP = Ref{Cint}()\n        GRBcbget(cb_data, cb_where, GRB_CB_MIPNODE_STATUS, resultP)\n        if resultP[] != GRB_OPTIMAL\n            return  # Solution is something other than optimal.\n        end\n    end\n    # Before querying `callback_value`, you must call:\n    Gurobi.load_callback_variable_primal(cb_data, cb_where)\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    # You can submit solver-independent MathOptInterface attributes such as\n    # lazy constraints, user-cuts, and heuristic solutions.\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y + x <= 3)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\n    if rand() < 0.1\n        # You can terminate the callback as follows:\n        GRBterminate(backend(model))\n    end\n    return\nend\n# You _must_ set this parameter if using lazy constraints.\nMOI.set(model, MOI.RawOptimizerAttribute(\"LazyConstraints\"), 1)\nMOI.set(model, Gurobi.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi documentation for other information that can be queried with GRBcbget.","category":"page"},{"location":"packages/Gurobi/#Common-Performance-Pitfall-with-JuMP","page":"jump-dev/Gurobi.jl","title":"Common Performance Pitfall with JuMP","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi's API works differently than most solvers. Any changes to the model are not applied immediately, but instead go sit in a internal buffer (making any modifications appear to be instantaneous) waiting for a call to GRBupdatemodel (where the work is  done).","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This leads to a common performance pitfall that has the following message as its main symptom:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Warning: excessive time spent in model updates. Consider calling update less frequently.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This often means the JuMP program was structured in such a way that Gurobi.jl ends up calling GRBupdatemodel in each iteration of a loop.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Usually, it is possible (and easy) to restructure the JuMP program in a way it stays ssolver-agnostic and has a close-to-ideal performance with Gurobi.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"To guide such restructuring it is good to keep in mind the following bits of information:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"GRBupdatemodel is only called if changes were done since last GRBupdatemodel (that is, if the internal buffer is not empty).\nGRBupdatemodel is called when JuMP.optimize! is called, but this often is not the source of the problem.\nGRBupdatemodel may be called when any model attribute is queried, even if that specific attribute was not changed. This often the source of the problem.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The worst-case scenario is, therefore, a loop of modify-query-modify-query, even if what is being modified and what is being queried are two completely distinct things.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"As an example, instead of:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\n@variable(model, x[1:100] >= 0)\nfor i in 1:100\n    set_upper_bound(x[i], i)\n    # `GRBupdatemodel` called on each iteration of this loop.\n    println(lower_bound(x[i]))\nend","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"do","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\n@variable(model, x[1:100] >= 0)\n# All modifications are done before any queries.\nfor i in 1:100\n    set_upper_bound(x[i], i)\nend\nfor i in 1:100\n    # Only the first `lower_bound` query may trigger an `GRBupdatemodel`.\n    println(lower_bound(x[i]))\nend","category":"page"},{"location":"packages/Gurobi/#Common-errors","page":"jump-dev/Gurobi.jl","title":"Common errors","text":"","category":"section"},{"location":"packages/Gurobi/#Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?","page":"jump-dev/Gurobi.jl","title":"Using Gurobi v9.0 and you got an error like Q not PSD?","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"You need to set the NonConvex parameter:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\nset_optimizer_attribute(model, \"NonConvex\", 2)","category":"page"},{"location":"packages/Gurobi/#Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X","page":"jump-dev/Gurobi.jl","title":"Gurobi Error 1009: Version number is XX.X, license is for version XX.X","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Make sure that your license is correct for your Gurobi version. See the Gurobi documentation for details.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Once you are sure that the license and Gurobi versions match, re-install Gurobi.jl by running:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"import Pkg\nPkg.build(\"Gurobi\")","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Benchmarks/overview.md\"","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Benchmarks/overview/#The-Benchmarks-submodule","page":"Overview","title":"The Benchmarks submodule","text":"","category":"section"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"To aid the development of efficient solver wrappers, MathOptInterface provides benchmarking capability. Benchmarking a wrapper follows a two-step process.","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"First, prior to making changes, create a baseline for the benchmark results on a given benchmark suite as follows:","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"using SolverPackage  # Replace with your choice of solver.\nimport MathOptInterface as MOI\n\nsuite = MOI.Benchmarks.suite() do\n    SolverPackage.Optimizer()\nend\n\nMOI.Benchmarks.create_baseline(\n    suite, \"current\"; directory = \"/tmp\", verbose = true\n)","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"Use the exclude argument to Benchmarks.suite to exclude benchmarks that the solver doesn't support.","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"Second, after making changes to the package, re-run the benchmark suite and compare to the prior saved results:","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"using SolverPackage\nimport MathOptInterface as MOI\n\nsuite = MOI.Benchmarks.suite() do\n    SolverPackage.Optimizer()\nend\n\nMOI.Benchmarks.compare_against_baseline(\n    suite, \"current\"; directory = \"/tmp\", verbose = true\n)","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"This comparison will create a report detailing improvements and regressions.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/latency.md\"","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/latency/#Latency","page":"Latency","title":"Latency","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"MathOptInterface suffers the \"time-to-first-solve\" problem of start-up latency.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This hurts both the user- and developer-experience of MathOptInterface. In the first case, because simple models have a multi-second delay before solving, and in the latter, because our tests take so long to run.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This page contains some advice on profiling and fixing latency-related problems in the MathOptInterface.jl repository.","category":"page"},{"location":"moi/tutorials/latency/#Background","page":"Latency","title":"Background","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Before reading this part of the documentation, you should familiarize yourself with the reasons for latency in Julia and how to fix them.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Read the blogposts on julialang.org on precompilation and SnoopCompile\nRead the SnoopCompile documentation.\nWatch Tim Holy's talk at JuliaCon 2021\nWatch the package development workshop at JuliaCon 2021","category":"page"},{"location":"moi/tutorials/latency/#Causes","page":"Latency","title":"Causes","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are three main causes of latency in MathOptInterface:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"A large number of types\nLack of method ownership\nType-instability in the bridge layer","category":"page"},{"location":"moi/tutorials/latency/#A-large-number-of-types","page":"Latency","title":"A large number of types","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Julia is very good at specializing method calls based on the input type. Each specialization has a compilation cost, but the benefit of faster run-time performance.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"The best-case scenario is for a method to be called a large number of times with a single set of argument types. The worst-case scenario is for a method to be called a single time for a large set of argument types.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Because of MathOptInterface's function-in-set formulation, we fall into the worst-case situation.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This is a fundamental limitation of Julia, so there isn't much we can do about it. However, if we can precompile MathOptInterface, much of the cost can be shifted from start-up latency to the time it takes to precompile a package on installation.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"However, there are two things which make MathOptInterface hard to precompile.","category":"page"},{"location":"moi/tutorials/latency/#Lack-of-method-ownership","page":"Latency","title":"Lack of method ownership","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Lack of method ownership happens when a call is made using a mix of structs and methods from different modules. Because of this, no single module \"owns\" the method that is being dispatched, and so it cannot be precompiled.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"tip: Tip\nThis is a slightly simplified explanation. Read the precompilation tutorial for a more in-depth discussion on back-edges.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Unfortunately, the design of MOI means that this is a frequent occurrence: we have a bunch of types in MOI.Utilities that wrap types defined in external packages (for example, the Optimizers), which implement methods of functions defined in MOI (for example, add_variable, add_constraint).","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Here's a simple example of method-ownership in practice:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"module MyMOI\nstruct Wrapper{T}\n    inner::T\nend\noptimize!(x::Wrapper) = optimize!(x.inner)\nend  # MyMOI\n\nmodule MyOptimizer\nusing ..MyMOI\nstruct Optimizer end\nMyMOI.optimize!(x::Optimizer) = 1\nend  # MyOptimizer\n\nusing SnoopCompile\nmodel = MyMOI.Wrapper(MyOptimizer.Optimizer())\n\njulia> tinf = @snoopi_deep MyMOI.optimize!(model)\nInferenceTimingNode: 0.008256/0.008543 on InferenceFrameInfo for Core.Compiler.Timings.ROOT() with 1 direct children","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"The result is that there was one method that required type inference. If we visualize tinf:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"using ProfileView\nProfileView.view(flamegraph(tinf))","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"we see a flamegraph with a large red-bar indicating that the method MyMOI.optimize(MyMOI.Wrapper{MyOptimizer.Optimizer}) cannot be precompiled.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"To fix this, we need to designate a module to \"own\" that method (that is, create a back-edge). The easiest way to do this is for MyOptimizer to call MyMOI.optimize(MyMOI.Wrapper{MyOptimizer.Optimizer}) during using MyOptimizer. Let's see that in practice:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"module MyMOI\nstruct Wrapper{T}\n    inner::T\nend\noptimize(x::Wrapper) = optimize(x.inner)\nend  # MyMOI\n\nmodule MyOptimizer\nusing ..MyMOI\nstruct Optimizer end\nMyMOI.optimize(x::Optimizer) = 1\n# The syntax of this let-while loop is very particular:\n#  * `let ... end` keeps everything local to avoid polluting the MyOptimizer\n#    namespace\n#  * `while true ... break end` runs the code once, and forces Julia to compile\n#    the inner loop, rather than interpret it.\nlet\n    while true\n        model = MyMOI.Wrapper(Optimizer())\n        MyMOI.optimize(model)\n        break\n    end\nend\nend  # MyOptimizer\n\nusing SnoopCompile\nmodel = MyMOI.Wrapper(MyOptimizer.Optimizer())\n\njulia> tinf = @snoopi_deep MyMOI.optimize(model)\nInferenceTimingNode: 0.006822/0.006822 on InferenceFrameInfo for Core.Compiler.Timings.ROOT() with 0 direct children","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are now 0 direct children that required type inference because the method was already stored in MyOptimizer!","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Unfortunately, this trick only works if the call-chain is fully inferrable. If there are breaks (due to type instability), then the benefit of doing this is reduced. And unfortunately for us, the design of MathOptInterface has a lot of type instabilities.","category":"page"},{"location":"moi/tutorials/latency/#Type-instability-in-the-bridge-layer","page":"Latency","title":"Type instability in the bridge layer","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Most of MathOptInterface is pretty good at ensuring type-stability. However, a key component is not type stable, and that is the bridging layer.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"In particular, the bridging layer defines Bridges.LazyBridgeOptimizer, which has fields like:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"struct LazyBridgeOptimizer\n    constraint_bridge_types::Vector{Any}\n    constraint_node::Dict{Tuple{Type,Type},ConstraintNode}\n    constraint_types::Vector{Tuple{Type,Type}}\nend","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This is because the LazyBridgeOptimizer needs to be able to deal with any function-in-set type passed to it, and we also allow users to pass additional bridges that they defined in external packages.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"So to recap, MathOptInterface suffers package latency because:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"there are a large number of types and functions\nand these are split between multiple modules, including external packages\nand there are type-instabilities like those in the bridging layer.","category":"page"},{"location":"moi/tutorials/latency/#Resolutions","page":"Latency","title":"Resolutions","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are no magic solutions to reduce latency. Issue #1313 tracks progress on reducing latency in MathOptInterface.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"A useful script is the following (replace GLPK as needed):","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"import GLPK\nimport MathOptInterface as MOI\n\nfunction example_diet(optimizer, bridge)\n    category_data = [\n        1800.0 2200.0;\n          91.0    Inf;\n           0.0   65.0;\n           0.0 1779.0\n    ]\n    cost = [2.49, 2.89, 1.50, 1.89, 2.09, 1.99, 2.49, 0.89, 1.59]\n    food_data = [\n        410 24 26 730;\n        420 32 10 1190;\n        560 20 32 1800;\n        380  4 19 270;\n        320 12 10 930;\n        320 15 12 820;\n        320 31 12 1230;\n        100  8 2.5 125;\n        330  8 10 180\n    ]\n    bridge_model = if bridge\n        MOI.instantiate(optimizer; with_bridge_type=Float64)\n    else\n        MOI.instantiate(optimizer)\n    end\n    model = MOI.Utilities.CachingOptimizer(\n        MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}()),\n        MOI.Utilities.AUTOMATIC,\n    )\n    MOI.Utilities.reset_optimizer(model, bridge_model)\n    MOI.set(model, MOI.Silent(), true)\n    nutrition = MOI.add_variables(model, size(category_data, 1))\n    for (i, v) in enumerate(nutrition)\n        MOI.add_constraint(model, v, MOI.GreaterThan(category_data[i, 1]))\n        MOI.add_constraint(model, v, MOI.LessThan(category_data[i, 2]))\n    end\n    buy = MOI.add_variables(model, size(food_data, 1))\n    MOI.add_constraint.(model, buy, MOI.GreaterThan(0.0))\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    f = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(cost, buy), 0.0)\n    MOI.set(model, MOI.ObjectiveFunction{typeof(f)}(), f)\n    for (j, n) in enumerate(nutrition)\n        f = MOI.ScalarAffineFunction(\n            MOI.ScalarAffineTerm.(food_data[:, j], buy),\n            0.0,\n        )\n        push!(f.terms, MOI.ScalarAffineTerm(-1.0, n))\n        MOI.add_constraint(model, f, MOI.EqualTo(0.0))\n    end\n    MOI.optimize!(model)\n    term_status = MOI.get(model, MOI.TerminationStatus())\n    @assert term_status == MOI.OPTIMAL\n    MOI.add_constraint(\n        model,\n        MOI.ScalarAffineFunction(\n            MOI.ScalarAffineTerm.(1.0, [buy[end-1], buy[end]]),\n            0.0,\n        ),\n        MOI.LessThan(6.0),\n    )\n    MOI.optimize!(model)\n    @assert MOI.get(model, MOI.TerminationStatus()) == MOI.INFEASIBLE\n    return\nend\n\nif length(ARGS) > 0\n    bridge = get(ARGS, 2, \"\") != \"--no-bridge\"\n    println(\"Running: $(ARGS[1]) $(get(ARGS, 2, \"\"))\")\n    @time example_diet(GLPK.Optimizer, bridge)\n    @time example_diet(GLPK.Optimizer, bridge)\n    exit(0)\nend","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"You can create a flame-graph via","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"using SnoopComile\ntinf = @snoopi_deep example_diet(GLPK.Optimizer, true)\nusing ProfileView\nProfileView.view(flamegraph(tinf))","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Here's how things looked in mid-August 2021: (Image: flamegraph)","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are a few opportunities for improvement (non-red flames, particularly on the right). But the main problem is a large red (non-precompilable due to method ownership) flame.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"EditURL = \"https://github.com/jump-dev/SCS.jl/blob/v1.3.1/README.md\"","category":"page"},{"location":"packages/SCS/#SCS.jl","page":"jump-dev/SCS.jl","title":"SCS.jl","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl is a wrapper for the SCS splitting cone solver.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS can solve linear programs, second-order cone programs, semidefinite programs, exponential cone programs, and power cone programs.","category":"page"},{"location":"packages/SCS/#Affiliation","page":"jump-dev/SCS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This wrapper is maintained by the JuMP community and is not a project of the SCS developers.","category":"page"},{"location":"packages/SCS/#License","page":"jump-dev/SCS.jl","title":"License","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The underlying solver, cvxgrp/scs, is licensed under the MIT License.","category":"page"},{"location":"packages/SCS/#Installation","page":"jump-dev/SCS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"Install SCS.jl using the Julia package manager:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"import Pkg\nPkg.add(\"SCS\")","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"In addition to installing the SCS.jl package, this will also download and install the SCS binaries. (You do not need to install SCS separately.)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/SCS/#Use-with-Convex.jl","page":"jump-dev/SCS.jl","title":"Use with Convex.jl","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This example shows how we can model a simple knapsack problem with Convex and use SCS to solve it.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"using Convex, SCS\nitems  = [:Gold, :Silver, :Bronze]\nvalues = [5.0, 3.0, 1.0]\nweights = [2.0, 1.5, 0.3]\n\n# Define a variable of size 3, each index representing an item\nx = Variable(3)\np = maximize(x' * values, 0 <= x, x <= 1, x' * weights <= 3)\nsolve!(p, SCS.Optimizer)\nprintln([items x.value])\n# [:Gold 0.9999971880377178\n#  :Silver 0.46667637765641057\n#  :Bronze 0.9999998036351865]","category":"page"},{"location":"packages/SCS/#Use-with-JuMP","page":"jump-dev/SCS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This example shows how we can model a simple knapsack problem with JuMP and use SCS to solve it.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"using JuMP, SCS\nitems  = [:Gold, :Silver, :Bronze]\nvalues = Dict(:Gold => 5.0,  :Silver => 3.0,  :Bronze => 1.0)\nweight = Dict(:Gold => 2.0,  :Silver => 1.5,  :Bronze => 0.3)\n\nmodel = Model(SCS.Optimizer)\n@variable(model, 0 <= take[items] <= 1)  # Define a variable for each item\n@objective(model, Max, sum(values[item] * take[item] for item in items))\n@constraint(model, sum(weight[item] * take[item] for item in items) <= 3)\noptimize!(model)\nprintln(value.(take))\n# 1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n#     Dimension 1, Symbol[:Gold, :Silver, :Bronze]\n# And data, a 3-element Array{Float64,1}:\n#  1.0000002002226671\n#  0.4666659513182934\n#  1.0000007732744878","category":"page"},{"location":"packages/SCS/#MathOptInterface-API","page":"jump-dev/SCS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The SCS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.VectorAffineFunction{Float64} in MOI.DualExponentialCone\nMOI.VectorAffineFunction{Float64} in MOI.DualPowerCone{Float64}\nMOI.VectorAffineFunction{Float64} in MOI.ExponentialCone\nMOI.VectorAffineFunction{Float64} in MOI.Nonnegatives\nMOI.VectorAffineFunction{Float64} in MOI.PowerCone{Float64}\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorAffineFunction{Float64} in MOI.Zeros\nMOI.VectorAffineFunction{Float64} in SCS.ScaledPSDCone","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/SCS/#Options","page":"jump-dev/SCS.jl","title":"Options","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"All SCS solver options can be set through Convex.jl or MathOptInterface.jl.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"For example:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"model = Model(optimizer_with_attributes(SCS.Optimizer, \"max_iters\" => 10))\n\n# via MathOptInterface:\noptimizer = SCS.Optimizer()\nMOI.set(optimizer, MOI.RawOptimizerAttribute(\"max_iters\"), 10)\nMOI.set(optimizer, MOI.RawOptimizerAttribute(\"verbose\"), 0)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"Common options are:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"max_iters: the maximum number of iterations to take\nverbose: turn printing on (1) or off (0)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"See the glbopts.h header for other options.","category":"page"},{"location":"packages/SCS/#Linear-solvers","page":"jump-dev/SCS.jl","title":"Linear solvers","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS uses a linear solver internally, see this section of SCS documentation. SCS.jl ships with","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.DirectSolver (sparse direct, the default) and\nSCS.LinearSolver (sparse indirect, by conjugate gradient)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"enabled.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The find currently available linear solvers one can inspect SCS.available_solvers:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> using SCS\n\njulia> SCS.available_solvers\n2-element Vector{DataType}:\n SCS.DirectSolver\n SCS.IndirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To select the linear solver of choice:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"pass the linear_solver option to optimizer_with_attributes, or to MOI.OptimizerWithAttributes;\nspecify the solver as the first argument when using scs_solve directly (see the low-level wrapper section below).","category":"page"},{"location":"packages/SCS/#SCS-with-MKL-Pardiso-linear-solver","page":"jump-dev/SCS.jl","title":"SCS with MKL Pardiso linear solver","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To enable the MKL Pardiso (direct sparse) solver one needs to load MKL_jll before SCS:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> import Pkg\n\njulia> Pkg.add(Pkg.PackageSpec(name = \"MKL_jll\", version = \"2022.2\"))\n\njulia> using MKL_jll    # This must be called before `using SCS`.\n\njulia> using SCS\n\njulia> SCS.available_solvers\n3-element Vector{DataType}:\n SCS.DirectSolver\n SCS.IndirectSolver\n SCS.MKLDirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The MKLDirectSolver is available on Linux x86_64 platform only.","category":"page"},{"location":"packages/SCS/#SCS-with-Sparse-GPU-indirect-solver-(CUDA-only)","page":"jump-dev/SCS.jl","title":"SCS with Sparse GPU indirect solver (CUDA only)","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To enable the indirect linear solver on GPU one needs to load CUDA_jll before SCS:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> import Pkg\n\njulia> Pkg.add(Pkg.PackageSpec(name = \"CUDA_jll\", version = \"10.1\"))\n\njulia> using CUDA_jll  # This must be called before `using SCS`.\n\njulia> using SCS\n\njulia> SCS.available_solvers\n3-element Array{DataType,1}:\n SCS.DirectSolver\n SCS.IndirectSolver\n SCS.GpuIndirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The GpuIndirectSolver is available on Linux x86_64 platform only.","category":"page"},{"location":"packages/SCS/#Low-level-wrapper","page":"jump-dev/SCS.jl","title":"Low-level wrapper","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl provides a low-level interface to solve a problem directly, without interfacing through MathOptInterface.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This is an advanced interface with a risk of incorrect usage. For new users, we recommend that you use the JuMP or Convex interfaces instead.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS solves a problem of the form:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"minimize        1/2 * x' * P * x + c' * x\nsubject to      A * x + s = b\n                s in K","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"where K is a product cone of:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"zero cone\npositive orthant { x | x ≥ 0 }\nbox cone { (t,x) | t*l ≤ x ≤ t*u}\nsecond-order cone (SOC) { (t,x) | ||x||_2 ≤ t }\nsemi-definite cone (SDC) { X | X is psd }\nexponential cone { (x,y,z) | y e^(x/y) ≤ z, y>0 }\npower cone { (x,y,z) | x^a * y^(1-a) ≥ |z|, x ≥ 0, y ≥ 0 }\ndual power cone { (u,v,w) | (u/a)^a * (v/(1-a))^(1-a) ≥ |w|, u ≥ 0, v ≥ 0 }.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To solve this problem with SCS, call SCS.scs_solve; see the docstring for details.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"EditURL = \"lp_sensitivity.jl\"","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Sensitivity-analysis-of-a-linear-program","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial explains how to use the lp_sensitivity_report function to create sensitivity reports like those that are produced by the Excel Solver. This is most often used in introductory classes to linear programming.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"In brief, sensitivity analysis of a linear program is about asking two questions:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Given an optimal solution, how much can the objective coefficients change by before a different solution becomes optimal?\nGiven an optimal solution, how much can the right-hand side of a linear constraint change by before a different solution becomes optimal?","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"JuMP provides a function, lp_sensitivity_report, to help us compute these values, but this tutorial extends that to create more informative tables in the form of a DataFrame.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Setup","page":"Sensitivity analysis of a linear program","title":"Setup","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"using JuMP\nimport HiGHS\nimport DataFrames","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"as well as this small linear program:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"model = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@variable(model, z <= 1)\n@objective(model, Min, 12x + 20y - z)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\n@constraint(model, c3, x + y <= 20)\noptimize!(model)\nsolution_summary(model; verbose = true)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Can you identify:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"The objective coefficient of each variable?\nThe right-hand side of each constraint?\nThe optimal primal and dual solutions?","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Sensitivity-reports","page":"Sensitivity analysis of a linear program","title":"Sensitivity reports","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Now let's call lp_sensitivity_report:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report = lp_sensitivity_report(model)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"It returns a SensitivityReport object, which maps:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Every variable reference to a tuple (d_lo, d_hi)::Tuple{Float64,Float64}, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.\nEvery constraint reference to a tuple (d_lo, d_hi)::Tuple{Float64,Float64}, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Both tuples are relative, rather than absolute. So, given an objective coefficient of 1.0 and a tuple (-0.5, 0.5), the objective coefficient can range between 1.0 - 0.5 an 1.0 + 0.5.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"For example:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report[x]","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"indicates that the objective coefficient on x, that is, 12, can decrease by -0.333 or increase by 3.0 and the primal solution (15, 1.25) will remain optimal. In addition:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report[c1]","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"means that the right-hand side of the c1 constraint (100), can decrease by 4 units, or increase by 2.85 units, and the primal solution (15, 1.25) will remain optimal.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Variable-sensitivity","page":"Sensitivity analysis of a linear program","title":"Variable sensitivity","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"By themselves, the tuples aren't informative. Let's put them in context by collating a range of other information about a variable:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"function variable_report(xi)\n    return (\n        name = name(xi),\n        lower_bound = has_lower_bound(xi) ? lower_bound(xi) : -Inf,\n        value = value(xi),\n        upper_bound = has_upper_bound(xi) ? upper_bound(xi) : Inf,\n        reduced_cost = reduced_cost(xi),\n        obj_coefficient = coefficient(objective_function(model), xi),\n        allowed_decrease = report[xi][1],\n        allowed_increase = report[xi][2],\n    )\nend","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Calling our function on x:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"x_report = variable_report(x)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"That's a bit hard to read, so let's call this on every variable in the model and put things into a DataFrame:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"variable_df =\n    DataFrames.DataFrame(variable_report(xi) for xi in all_variables(model))","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Constraint-sensitivity","page":"Sensitivity analysis of a linear program","title":"Constraint sensitivity","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"We can do something similar with constraints:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"function constraint_report(c::ConstraintRef)\n    return (\n        name = name(c),\n        value = value(c),\n        rhs = normalized_rhs(c),\n        slack = normalized_rhs(c) - value(c),\n        shadow_price = shadow_price(c),\n        allowed_decrease = report[c][1],\n        allowed_increase = report[c][2],\n    )\nend\n\nc1_report = constraint_report(c1)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"That's a bit hard to read, so let's call this on every variable in the model and put things into a DataFrame:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"constraint_df = DataFrames.DataFrame(\n    constraint_report(ci) for (F, S) in list_of_constraint_types(model) for\n    ci in all_constraints(model, F, S) if F == AffExpr\n)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Analysis-questions","page":"Sensitivity analysis of a linear program","title":"Analysis questions","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Now we can use these dataframes to ask questions of the solution.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"For example, we can find basic variables by looking for variables with a reduced cost of 0:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"basic = filter(row -> iszero(row.reduced_cost), variable_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"and non-basic variables by looking for non-zero reduced costs:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"non_basic = filter(row -> !iszero(row.reduced_cost), variable_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"we can also find constraints that are binding by looking for zero slacks:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"binding = filter(row -> iszero(row.slack), constraint_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"or non-zero shadow prices:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"binding2 = filter(row -> !iszero(row.shadow_price), constraint_df)","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/example.md\"","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/example/#Solving-a-problem-using-MathOptInterface","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"In this tutorial we demonstrate how to use MathOptInterface to solve the binary-constrained knapsack problem:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"beginaligned\nmax   c^top x       \nst   w^top x le C \n         x_i in 01quad forall i=1ldotsn\nendaligned","category":"page"},{"location":"moi/tutorials/example/#Required-packages","page":"Solving a problem using MathOptInterface","title":"Required packages","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Load the MathOptInterface module and define the shorthand MOI:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"import MathOptInterface as MOI","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"As an optimizer, we choose GLPK:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"using GLPK\noptimizer = GLPK.Optimizer()","category":"page"},{"location":"moi/tutorials/example/#Define-the-data","page":"Solving a problem using MathOptInterface","title":"Define the data","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"We first define the constants of the problem:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> c = [1.0, 2.0, 3.0]\n3-element Vector{Float64}:\n 1.0\n 2.0\n 3.0\n\njulia> w = [0.3, 0.5, 1.0]\n3-element Vector{Float64}:\n 0.3\n 0.5\n 1.0\n\njulia> C = 3.2\n3.2","category":"page"},{"location":"moi/tutorials/example/#Add-the-variables","page":"Solving a problem using MathOptInterface","title":"Add the variables","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> x = MOI.add_variables(optimizer, length(c));","category":"page"},{"location":"moi/tutorials/example/#set_objective_example","page":"Solving a problem using MathOptInterface","title":"Set the objective","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.set(\n           optimizer,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(c, x), 0.0),\n       );\n\njulia> MOI.set(optimizer, MOI.ObjectiveSense(), MOI.MAX_SENSE)","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"tip: Tip\nMOI.ScalarAffineTerm.(c, x) is a shortcut for [MOI.ScalarAffineTerm(c[i], x[i]) for i = 1:3]. This is Julia's broadcast syntax in action, and is used quite often throughout MOI.","category":"page"},{"location":"moi/tutorials/example/#Add-the-constraints","page":"Solving a problem using MathOptInterface","title":"Add the constraints","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"We add the knapsack constraint and integrality constraints:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.add_constraint(\n           optimizer,\n           MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(w, x), 0.0),\n           MOI.LessThan(C),\n       );","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Add integrality constraints:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> for x_i in x\n           MOI.add_constraint(optimizer, x_i, MOI.ZeroOne())\n       end","category":"page"},{"location":"moi/tutorials/example/#Optimize-the-model","page":"Solving a problem using MathOptInterface","title":"Optimize the model","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.optimize!(optimizer)","category":"page"},{"location":"moi/tutorials/example/#Understand-why-the-solver-stopped","page":"Solving a problem using MathOptInterface","title":"Understand why the solver stopped","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"The first thing to check after optimization is why the solver stopped, for example, did it stop because of a time limit or did it stop because it found the optimal solution?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.TerminationStatus())\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Looks like we found an optimal solution.","category":"page"},{"location":"moi/tutorials/example/#Understand-what-solution-was-returned","page":"Solving a problem using MathOptInterface","title":"Understand what solution was returned","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.ResultCount())\n1\n\njulia> MOI.get(optimizer, MOI.PrimalStatus())\nFEASIBLE_POINT::ResultStatusCode = 1\n\njulia> MOI.get(optimizer, MOI.DualStatus())\nNO_SOLUTION::ResultStatusCode = 0","category":"page"},{"location":"moi/tutorials/example/#Query-the-objective","page":"Solving a problem using MathOptInterface","title":"Query the objective","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"What is its objective value?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.ObjectiveValue())\n6.0","category":"page"},{"location":"moi/tutorials/example/#Query-the-primal-solution","page":"Solving a problem using MathOptInterface","title":"Query the primal solution","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"And what is the value of the variables x?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.VariablePrimal(), x)\n3-element Vector{Float64}:\n 1.0\n 1.0\n 1.0","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"EditURL = \"https://github.com/jump-dev/DiffOpt.jl/blob/v0.4.2/README.md\"","category":"page"},{"location":"packages/DiffOpt/#DiffOpt.jl","page":"jump-dev/DiffOpt.jl","title":"DiffOpt.jl","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"(Image: stable docs) (Image: development docs) (Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt.jl is a package for differentiating convex optimization programs with respect to the program parameters. DiffOpt currently supports linear, quadratic, and conic programs.","category":"page"},{"location":"packages/DiffOpt/#License","page":"jump-dev/DiffOpt.jl","title":"License","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/DiffOpt/#Installation","page":"jump-dev/DiffOpt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"Install DiffOpt using Pkg.add:","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"import Pkg\nPkg.add(\"DiffOpt\")","category":"page"},{"location":"packages/DiffOpt/#Documentation","page":"jump-dev/DiffOpt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"The documentation for DiffOpt.jl includes a detailed description of the theory behind the package, along with examples, tutorials, and an API reference.","category":"page"},{"location":"packages/DiffOpt/#Use-with-JuMP","page":"jump-dev/DiffOpt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"Use DiffOpt with JuMP by following this brief example:","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"using JuMP, DiffOpt, HiGHS\n# Create a model using the wrapper\nmodel = Model(() -> DiffOpt.diff_optimizer(HiGHS.Optimizer))\n# Define your model and solve it\n@variable(model, x)\n@constraint(model, cons, x >= 3)\n@objective(model, Min, 2x)\noptimize!(model)\n# Choose the problem parameters to differentiate with respect to, and set their\n# perturbations.\nMOI.set(model, DiffOpt.ReverseVariablePrimal(), x, 1.0)\n# Differentiate the model\nDiffOpt.reverse_differentiate!(model)\n# fetch the gradients\ngrad_exp = MOI.get(model, DiffOpt.ReverseConstraintFunction(), cons)  # -3 x - 1\nconstant(grad_exp)        # -1\ncoefficient(grad_exp, x)  # -3","category":"page"},{"location":"packages/DiffOpt/#GSOC2020","page":"jump-dev/DiffOpt.jl","title":"GSOC2020","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt began as a NumFOCUS sponsored Google Summer of Code (2020) project","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"EditURL = \"https://github.com/jump-dev/KNITRO.jl/blob/4c56de7684c42dd3c83c5fbc515ae1f424eed524/README.md\"","category":"page"},{"location":"packages/KNITRO/#KNITRO.jl","page":"jump-dev/KNITRO.jl","title":"KNITRO.jl","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl is a wrapper for the Artelys Knitro solver.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"It has two components:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"a thin wrapper around the C API\nan interface to MathOptInterface.","category":"page"},{"location":"packages/KNITRO/#Affiliation","page":"jump-dev/KNITRO.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"This wrapper is maintained by the JuMP community with help from Artelys.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Contact Artelys support if you encounter any problem with this interface or the solver.","category":"page"},{"location":"packages/KNITRO/#License","page":"jump-dev/KNITRO.jl","title":"License","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl is licensed under the MIT License.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/KNITRO/#Installation","page":"jump-dev/KNITRO.jl","title":"Installation","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"First, obtain a license and install a copy of KNITRO from Artelys.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Then, install KNITRO.jl using the Julia package manager:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"import Pkg\nPkg.add(\"KNITRO\")","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"If you are having trouble installing KNITRO.jl, here are several things to try:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Make sure that you have defined your global variables correctly, for example with export KNITRODIR=\"/path/to/knitro-vXXX-$OS-64\" and export LD_LIBRARY_PATH=\"$LD_LIBRARY_PATH:$KNITRODIR/lib\". You can check that KNITRO.jl sees your library with using KNITRO; KNITRO.has_knitro().\nIf KNITRO.has_knitro() returns false but you are confident that your paths are correct, try running Pkg.build(\"KNITRO\") and restarting Julia. In at least one user's experience, installing and using KNITRO in a temporary Julia environment (activated with ] activate --temp) does not work and the need to manually build is likely the reason why.","category":"page"},{"location":"packages/KNITRO/#Use-with-JuMP","page":"jump-dev/KNITRO.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"To use KNITRO with JuMP, use KNITRO.Optimizer:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"using JuMP, KNITRO\nmodel = Model(KNITRO.Optimizer)\nset_attribute(model, \"outlev\", 1)\nset_attribute(model, \"algorithm\", 4)","category":"page"},{"location":"packages/KNITRO/#Use-with-AMPL","page":"jump-dev/KNITRO.jl","title":"Use with AMPL","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"To use KNITRO with AmplNLWriter.jl, use KNITRO.amplexe:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"using JuMP\nimport AmplNLWriter\nimport KNITRO\nmodel = Model(() -> AmplNLWriter.Optimizer(KNITRO.amplexe, [\"outlev=3\"]))","category":"page"},{"location":"packages/KNITRO/#Use-with-other-packages","page":"jump-dev/KNITRO.jl","title":"Use with other packages","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"A variety of packages extend KNITRO.jl to support other optimization modeling systems. These include:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"NLPModelsKnitro\nOptimization.jl","category":"page"},{"location":"packages/KNITRO/#MathOptInterface-API","page":"jump-dev/KNITRO.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"The Knitro optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.Reals","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorOfVariables in MOI.Complements\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.NLPBlock()\nMOI.NLPBlockDualStart()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/KNITRO/#Options","page":"jump-dev/KNITRO.jl","title":"Options","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"A list of available options is provided in the KNITRO reference manual.","category":"page"},{"location":"packages/KNITRO/#Low-level-wrapper","page":"jump-dev/KNITRO.jl","title":"Low-level wrapper","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl implements most of Knitro's functionalities. If you aim at using part of Knitro's API that are not implemented in the MathOptInterface/JuMP ecosystem, you can refer to the low-level API, which wraps Knitro's C API (whose templates are specified in the file knitro.h).","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Extensive examples using the C wrapper can be found in examples/.","category":"page"},{"location":"packages/KNITRO/#Multi-threading","page":"jump-dev/KNITRO.jl","title":"Multi-threading","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Due to limitations in the interaction between Julia and C, KNITRO.jl disables multi-threading if the problem is nonlinear. This will override any options such as par_numthreads that you may have set. Read GitHub issue #93 for more details.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/modification.md\"","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/modification/#Problem-modification","page":"Problem modification","title":"Problem modification","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"In addition to adding and deleting constraints and variables, MathOptInterface supports modifying, in-place, coefficients in the constraints and the objective function of a model.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"These modifications can be grouped into two categories:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"modifications which replace the set of function of a constraint with a new set or function\nmodifications which change, in-place, a component of a function","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"warning: Warning\nSome ModelLike objects do not support problem modification.","category":"page"},{"location":"moi/manual/modification/#Modify-the-set-of-a-constraint","page":"Problem modification","title":"Modify the set of a constraint","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use set and ConstraintSet to modify the set of a constraint by replacing it with a new instance of the same type.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.set(model, MOI.ConstraintSet(), c, MOI.EqualTo(2.0));\n\njulia> MOI.get(model, MOI.ConstraintSet(), c) == MOI.EqualTo(2.0)\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"However, the following will fail as the new set is of a different type to the original set:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(model, MOI.ConstraintSet(), c, MOI.GreaterThan(2.0))\nERROR: [...]","category":"page"},{"location":"moi/manual/modification/#Special-cases:-set-transforms","page":"Problem modification","title":"Special cases: set transforms","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"If our constraint is an affine inequality, then this corresponds to modifying the right-hand side of a constraint in linear programming.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"In some special cases, solvers may support efficiently changing the set of a constraint (for example, from LessThan to GreaterThan). For these cases, MathOptInterface provides the transform method.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"The transform function returns a new constraint index, and the old constraint index (that is, c) is no longer valid.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.LessThan(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}(1)\n\njulia> new_c = MOI.transform(model, c, MOI.GreaterThan(2.0));\n\njulia> MOI.is_valid(model, c)\nfalse\n\njulia> MOI.is_valid(model, new_c)\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"note: Note\ntransform cannot be called with a set of the same type. Use set instead.","category":"page"},{"location":"moi/manual/modification/#Modify-the-function-of-a-constraint","page":"Problem modification","title":"Modify the function of a constraint","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use set and ConstraintFunction to modify the function of a constraint by replacing it with a new instance of the same type.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(2.0, x)], 1.0);\n\njulia> MOI.set(model, MOI.ConstraintFunction(), c, new_f);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"However, the following will fail as the new function is of a different type to the original function:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(model, MOI.ConstraintFunction(), c, x)\nERROR: [...]","category":"page"},{"location":"moi/manual/modification/#Modify-constant-term-in-a-scalar-function","page":"Problem modification","title":"Modify constant term in a scalar function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and ScalarConstantChange to modify the constant term in a ScalarAffineFunction or ScalarQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.modify(model, c, MOI.ScalarConstantChange(1.0));\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 1.0);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"ScalarConstantChange can also be used to modify the objective function by passing an instance of ObjectiveFunction:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           new_f,\n       );\n\njulia> MOI.modify(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           MOI.ScalarConstantChange(-1.0)\n       );\n\njulia> MOI.get(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n       ) ≈ MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], -1.0)\ntrue","category":"page"},{"location":"moi/manual/modification/#Modify-constant-terms-in-a-vector-function","page":"Problem modification","title":"Modify constant terms in a vector function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and VectorConstantChange to modify the constant vector in a VectorAffineFunction or VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.VectorAffineFunction([\n                   MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n                   MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n               ],\n               [0.0, 0.0],\n           ),\n           MOI.Nonnegatives(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.modify(model, c, MOI.VectorConstantChange([3.0, 4.0]));\n\njulia> new_f = MOI.VectorAffineFunction(\n           [\n        MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n        MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n           ],\n           [3.0, 4.0],\n       );\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/#Modify-affine-coefficients-in-a-scalar-function","page":"Problem modification","title":"Modify affine coefficients in a scalar function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and ScalarCoefficientChange to modify the affine coefficient of a ScalarAffineFunction or ScalarQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.modify(model, c, MOI.ScalarCoefficientChange(x, 2.0));\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(2.0, x)], 0.0);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"ScalarCoefficientChange can also be used to modify the objective function by passing an instance of ObjectiveFunction.","category":"page"},{"location":"moi/manual/modification/#Modify-affine-coefficients-in-a-vector-function","page":"Problem modification","title":"Modify affine coefficients in a vector function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and MultirowChange to modify a vector of affine coefficients in a VectorAffineFunction or a VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.VectorAffineFunction([\n                   MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n                   MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n               ],\n               [0.0, 0.0],\n           ),\n           MOI.Nonnegatives(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.modify(model, c, MOI.MultirowChange(x, [(1, 3.0), (2, 4.0)]));\n\njulia> new_f = MOI.VectorAffineFunction(\n           [\n        MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(3.0, x)),\n        MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(4.0, x)),\n           ],\n           [0.0, 0.0],\n       );\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"EditURL = \"facility_location.jl\"","category":"page"},{"location":"tutorials/linear/facility_location/#The-facility-location-problem","page":"The facility location problem","title":"The facility location problem","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial was originally contributed by Mathieu Tanneau and Alexis Montoison.","category":"page"},{"location":"tutorials/linear/facility_location/#Required-packages","page":"The facility location problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"using JuMP\nimport HiGHS\nimport LinearAlgebra\nimport Plots\nimport Random","category":"page"},{"location":"tutorials/linear/facility_location/#Uncapacitated-facility-location","page":"The facility location problem","title":"Uncapacitated facility location","text":"","category":"section"},{"location":"tutorials/linear/facility_location/#Problem-description","page":"The facility location problem","title":"Problem description","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"We are given","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"A set M=1 dots m of clients\nA set N= 1 dots n of sites where a facility can be built","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Decision variables Decision variables are split into two categories:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Binary variable y_j indicates whether facility j is built or not\nBinary variable x_i j indicates whether client i is assigned to facility j","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Objective The objective is to minimize the total cost of serving all clients. This costs breaks down into two components:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Fixed cost of building a facility.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"In this example, this cost is f_j = 1  forall j.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Cost of serving clients from the assigned facility.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"In this example, the cost c_i j of serving client i from facility j is the Euclidean distance between the two.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Constraints","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Each customer must be served by exactly one facility\nA facility cannot serve any client unless it is open","category":"page"},{"location":"tutorials/linear/facility_location/#MILP-formulation","page":"The facility location problem","title":"MILP formulation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The problem can be formulated as the following MILP:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nmin_x y    \nsum_i j c_i j x_i j +\nsum_j f_j y_j \nst \nsum_j x_i j = 1  forall i in M \n x_i j leq y_j  forall i in M j in N \n x_i j y_j in 0 1  forall i in M j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"where the first set of constraints ensures that each client is served exactly once, and the second set of constraints ensures that no client is served from an unopened facility.","category":"page"},{"location":"tutorials/linear/facility_location/#Problem-data","page":"The facility location problem","title":"Problem data","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"To ensure reproducibility, we set the random number seed:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Random.seed!(314)","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Here's the data we need:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"# Number of clients\nm = 12\n# Number of facility locations\nn = 5\n\n# Clients' locations\nx_c, y_c = rand(m), rand(m)\n\n# Facilities' potential locations\nx_f, y_f = rand(n), rand(n)\n\n# Fixed costs\nf = ones(n);\n\n# Distance\nc = zeros(m, n)\nfor i in 1:m\n    for j in 1:n\n        c[i, j] = LinearAlgebra.norm([x_c[i] - x_f[j], y_c[i] - y_f[j]], 2)\n    end\nend","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the data","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Plots.scatter(\n    x_c,\n    y_c;\n    label = \"Clients\",\n    markershape = :circle,\n    markercolor = :blue,\n)\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = \"Facility\",\n    markershape = :square,\n    markercolor = :white,\n    markersize = 6,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)","category":"page"},{"location":"tutorials/linear/facility_location/#JuMP-implementation","page":"The facility location problem","title":"JuMP implementation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Create a JuMP model","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, y[1:n], Bin);\n@variable(model, x[1:m, 1:n], Bin);\n# Each client is served exactly once\n@constraint(model, client_service[i in 1:m], sum(x[i, j] for j in 1:n) == 1);\n# A facility must be open to serve a client\n@constraint(model, open_facility[i in 1:m, j in 1:n], x[i, j] <= y[j]);\n@objective(model, Min, f' * y + sum(c .* x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Solve the uncapacitated facility location problem with HiGHS","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"optimize!(model)\nprintln(\"Optimal value: \", objective_value(model))","category":"page"},{"location":"tutorials/linear/facility_location/#Visualizing-the-solution","page":"The facility location problem","title":"Visualizing the solution","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The threshold 1e-5 ensure that edges between clients and facilities are drawn when x[i, j] ≈ 1.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"x_is_selected = isapprox.(value.(x), 1; atol = 1e-5);\ny_is_selected = isapprox.(value.(y), 1; atol = 1e-5);\n\np = Plots.scatter(\n    x_c,\n    y_c;\n    markershape = :circle,\n    markercolor = :blue,\n    label = nothing,\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    markershape = :square,\n    markercolor = [(y_is_selected[j] ? :red : :white) for j in 1:n],\n    markersize = 6,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n    label = nothing,\n)\n\nfor i in 1:m, j in 1:n\n    if x_is_selected[i, j]\n        Plots.plot!(\n            [x_c[i], x_f[j]],\n            [y_c[i], y_f[j]];\n            color = :black,\n            label = nothing,\n        )\n    end\nend\n\np","category":"page"},{"location":"tutorials/linear/facility_location/#Capacitated-facility-location","page":"The facility location problem","title":"Capacitated facility location","text":"","category":"section"},{"location":"tutorials/linear/facility_location/#Problem-formulation","page":"The facility location problem","title":"Problem formulation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The capacitated variant introduces a capacity constraint on each facility, that is, clients have a certain level of demand to be served, while each facility only has finite capacity which cannot be exceeded.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Specifically,","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The demand of client i is denoted by a_i geq 0\nThe capacity of facility j is denoted by q_j geq 0","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The capacity constraints then write","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nsum_i a_i x_i j leq q_j y_j  forall j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Note that, if y_j is set to 0, the capacity constraint above automatically forces x_i j to 0.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Thus, the capacitated facility location can be formulated as follows","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nmin_x y    \nsum_i j c_i j x_i j +\nsum_j f_j y_j \nst \nsum_j x_i j = 1  forall i in M \n sum_i a_i x_i j leq q_j y_j  forall j in N \n x_i j y_j in 0 1  forall i in M j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"For simplicity, we will assume that there is enough capacity to serve the demand, that is, there exists at least one feasible solution.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"We need some new data:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"# Demands\na = rand(1:3, m);\n\n# Capacities\nq = rand(5:10, n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the data","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Plots.scatter(\n    x_c,\n    y_c;\n    label = nothing,\n    markershape = :circle,\n    markercolor = :blue,\n    markersize = 2 .* (2 .+ a),\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = nothing,\n    markershape = :rect,\n    markercolor = :white,\n    markersize = q,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)","category":"page"},{"location":"tutorials/linear/facility_location/#JuMP-implementation-2","page":"The facility location problem","title":"JuMP implementation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Create a JuMP model","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, y[1:n], Bin);\n@variable(model, x[1:m, 1:n], Bin);\n# Each client is served exactly once\n@constraint(model, client_service[i in 1:m], sum(x[i, :]) == 1);\n# Capacity constraint\n@constraint(model, capacity, x' * a .<= (q .* y));\n# Objective\n@objective(model, Min, f' * y + sum(c .* x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Solve the problem","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"optimize!(model)\nprintln(\"Optimal value: \", objective_value(model))","category":"page"},{"location":"tutorials/linear/facility_location/#Visualizing-the-solution-2","page":"The facility location problem","title":"Visualizing the solution","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The threshold 1e-5 ensure that edges between clients and facilities are drawn when x[i, j] ≈ 1.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"x_is_selected = isapprox.(value.(x), 1; atol = 1e-5);\ny_is_selected = isapprox.(value.(y), 1; atol = 1e-5);\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the solution","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"p = Plots.scatter(\n    x_c,\n    y_c;\n    label = nothing,\n    markershape = :circle,\n    markercolor = :blue,\n    markersize = 2 .* (2 .+ a),\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = nothing,\n    markershape = :rect,\n    markercolor = [(y_is_selected[j] ? :red : :white) for j in 1:n],\n    markersize = q,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)\n\nfor i in 1:m, j in 1:n\n    if x_is_selected[i, j]\n        Plots.plot!(\n            [x_c[i], x_f[j]],\n            [y_c[i], y_f[j]];\n            color = :black,\n            label = nothing,\n        )\n    end\nend\n\np","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"EditURL = \"https://github.com/jump-dev/Xpress.jl/blob/v0.16.1/README.md\"","category":"page"},{"location":"packages/Xpress/#Xpress.jl","page":"jump-dev/Xpress.jl","title":"Xpress.jl","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Xpress.jl is a wrapper for the FICO Xpress Solver.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"It has two components:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Xpress/#Affiliation","page":"jump-dev/Xpress.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The Xpress wrapper for Julia is community driven and not officially supported by FICO Xpress. If you are a commercial customer interested in official support for Julia from FICO Xpress, let them know.","category":"page"},{"location":"packages/Xpress/#License","page":"jump-dev/Xpress.jl","title":"License","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Xpress.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/Xpress/#Installation","page":"jump-dev/Xpress.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"First, obtain a license of Xpress and install Xpress solver, following the instructions on the FICO website.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Then, install this package using:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"import Pkg\nPkg.add(\"Xpress\")","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"If you encounter an error, make sure that the XPRESSDIR environmental variable is set to the path of the Xpress directory. This should be part of a standard installation. The Xpress library will be searched for in XPRESSDIR/lib on Unix platforms and XPRESSDIR/bin on Windows.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For example, on macOS, you may need:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"ENV[\"XPRESSDIR\"] = \"/Applications/FICO Xpress/xpressmp/\"\nimport Pkg\nPkg.add(\"Xpress\")","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"By default, building Xpress.jl will fail if the Xpress library is not found. This may not be desirable in certain cases, for example when part of a package's test suite uses Xpress as an optional test dependency, but Xpress cannot be installed on a CI server running the test suite. To support this use case, the XPRESS_JL_SKIP_LIB_CHECK environment variable may be set (to any value) to make Xpress.jl installable (but not usable).","category":"page"},{"location":"packages/Xpress/#Use-with-JuMP","page":"jump-dev/Xpress.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"To use Xpress with JuMP, use:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"using JuMP, Xpress\nmodel = Model(Xpress.Optimizer)\nset_optimizer(model, \"PRESOLVE\", 0)","category":"page"},{"location":"packages/Xpress/#Options","page":"jump-dev/Xpress.jl","title":"Options","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For other parameters use Xpress Optimizer manual or type julia -e \"using Xpress; println(keys(Xpress.XPRS_ATTRIBUTES))\".","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"If logfile is set to \"\", the log file is disabled and output is printed to the console (there might be issues with console output on windows (it is manually implemented with callbacks)). If logfile is set to a file's path, output is printed to that file. By default, logfile = \"\" (console).","category":"page"},{"location":"packages/Xpress/#Callbacks","page":"jump-dev/Xpress.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Solver specific and solver independent callbacks are working in MathOptInterface and, consequently, in JuMP. However, the current implementation should be considered experimental.","category":"page"},{"location":"packages/Xpress/#Environment-variables","page":"jump-dev/Xpress.jl","title":"Environment variables","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"XPRESS_JL_SKIP_LIB_CHECK: Used to skip build lib check as previously described.\nXPRESS_JL_NO_INFO: Disable license info log.\nXPRESS_JL_NO_DEPS_ERROR: Disable error when do deps.jl file is found.\nXPRESS_JL_NO_AUTO_INIT: Disable automatic run of Xpress.initialize().","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Specially useful for explicitly loading the dynamic library.","category":"page"},{"location":"packages/Xpress/#Skipping-Xpress.postsolve","page":"jump-dev/Xpress.jl","title":"Skipping Xpress.postsolve","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"In older versions of Xpress, the command XPRSpostsolve throws an error in infeasible models. In these older versions the post solve should not be executed. To do this, one can use the MOI.RawOptimizerAttribute(\"MOI_POST_SOLVE\") to skip this routine.","category":"page"},{"location":"packages/Xpress/#C-API","page":"jump-dev/Xpress.jl","title":"C API","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The C API can be accessed via Xpress.Lib.XPRSxx functions, where the names and arguments are identical to the C API. ","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"See the Xpress documentation for details.","category":"page"},{"location":"packages/Xpress/#Documentation","page":"jump-dev/Xpress.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For more information, consult the FICO optimizer manual.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"EditURL = \"mip_duality.jl\"","category":"page"},{"location":"tutorials/linear/mip_duality/#Computing-the-duals-of-a-mixed-integer-program","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial explains how to compute the duals of a mixed-integer linear program by fixing the discrete variables to their optimal solution and resolving as a linear program.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/mip_duality/#The-model","page":"Computing the duals of a mixed-integer program","title":"The model","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Our example model is the unit commitment example from Unit commitment. The details are unimportant, other than to note that there are two types of continuous variables, g and w, representing the quantity of generation from thermal and wind plants, and a discrete variable dispatch, which is 1 if plant i is operating, and 0 if not.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"We are interested in the \"dual\" of the power_balance constraint, because it represents the marginal price of electricity that consumers should pay for their consumption.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"generators = [\n    (min = 0.0, max = 1000.0, fixed_cost = 1000.0, variable_cost = 50.0),\n    (min = 300.0, max = 1000.0, fixed_cost = 0.0, variable_cost = 100.0),\n]\nN = length(generators)\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variables(model, begin\n    generators[i].min <= g[i = 1:N] <= generators[i].max\n    0 <= w <= 200\n    dispatch[i = 1:N], Bin\nend)\n@constraints(model, begin\n    power_balance, sum(g[i] for i in 1:N) + w == 1500\n    [i = 1:N], g[i] <= generators[i].max * dispatch[i]\n    [i = 1:N], g[i] >= generators[i].min * dispatch[i]\nend)\n@objective(\n    model,\n    Min,\n    sum(\n        generators[i].fixed_cost * dispatch[i] +\n        generators[i].variable_cost * g[i] for i in 1:N\n    )\n)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/#Manually-fix-the-variables","page":"Computing the duals of a mixed-integer program","title":"Manually fix the variables","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"If we optimize this model, we obtain a dual_status of NO_SOLUTION:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This is because HiGHS cannot compute the duals of a mixed-integer program. We can work around this problem by fixing the integer variables to their optimal solution, relaxing integrality, and re-solving as a linear program.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"discrete_values = value.(dispatch)\nfix.(dispatch, discrete_values; force = true)\nunset_binary.(dispatch)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Now if we re-solve the problem, we obtain a FEASIBLE_POINT for the dual:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"and a marginal price of electricity of $100/MWh:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"dual(power_balance)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"To reset the problem back to a mixed-integer linear program, we need to unfix and call set_binary:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"unfix.(dispatch)\nset_binary.(dispatch)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/#Use-fix_discrete_variables","page":"Computing the duals of a mixed-integer program","title":"Use fix_discrete_variables","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Manually choosing the variables to relax and fix works for our small example, but it becomes more difficult in problems with a larger number of binary and integer variables. To automate the process we just did manually, JuMP provides the fix_discrete_variables function:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"undo = fix_discrete_variables(model);\nnothing #hide","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Here undo is a function that, when called with no arguments, returns the model to the original mixed-integer formulation.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"tip: Tip\nAfter calling fix_discrete_variables, you can set a new solver with set_optimizer if your mixed-integer solver does not support computing a dual solution.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"print(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"dual(power_balance)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Finally, call undo to revert the reformulation","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"undo()\nprint(model)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"EditURL = \"min_ellipse.jl\"","category":"page"},{"location":"tutorials/conic/min_ellipse/#Minimal-ellipses","page":"Minimal ellipses","title":"Minimal ellipses","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This example comes from section 8.4.1 of the book Convex Optimization by Boyd and Vandenberghe (2004).","category":"page"},{"location":"tutorials/conic/min_ellipse/#Formulation","page":"Minimal ellipses","title":"Formulation","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Given a set of m ellipses of the form:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"E(A b c) =  x  x^top A x + 2 b^top x + c leq 0 ","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"the minimal ellipse problem finds an ellipse with the minimum area that encloses the given ellipses.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"It is convenient to parameterize the minimal enclosing ellipse as","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":" x   Px + q  leq 1 ","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Then the optimal P and q are given by the convex semidefinite program;","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"beginaligned\ntextmaximize   quad log(det(P))  \ntextsubject to   quad tau_i geq 0  i = 1 ldots m \n quadbeginbmatrix\n    P^2 - tau_i A_i          P q - tau_i b_i       0        \n    (P q - tau_i b_i)^top    -1 - tau_i c_i     (P q)^top \n                0                     (P q)       - P^2       \nendbmatrix\npreceq 0 text (PSD)    i=1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"with helper variables tau.","category":"page"},{"location":"tutorials/conic/min_ellipse/#Required-packages","page":"Minimal ellipses","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport SCS","category":"page"},{"location":"tutorials/conic/min_ellipse/#Data","page":"Minimal ellipses","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"First, define the m input ellipses (here m = 6), parameterized as x^T A_i x + 2 b_i^T x + c leq 0:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"struct Ellipse\n    A::Matrix{Float64}\n    b::Vector{Float64}\n    c::Float64\n    function Ellipse(A::Matrix{Float64}, b::Vector{Float64}, c::Float64)\n        @assert isreal(A) && LinearAlgebra.issymmetric(A)\n        return new(A, b, c)\n    end\nend\n\nellipses = [\n    Ellipse([1.2576 -0.3873; -0.3873 0.3467], [0.2722, 0.1969], 0.1831),\n    Ellipse([1.4125 -2.1777; -2.1777 6.7775], [-1.228, -0.0521], 0.3295),\n    Ellipse([1.7018 0.8141; 0.8141 1.7538], [-0.4049, 1.5713], 0.2077),\n    Ellipse([0.9742 -0.7202; -0.7202 1.5444], [0.0265, 0.5623], 0.2362),\n    Ellipse([0.6798 -0.1424; -0.1424 0.6871], [-0.4301, -1.0157], 0.3284),\n    Ellipse([0.1796 -0.1423; -0.1423 2.6181], [-0.3286, 0.557], 0.4931),\n];\nnothing #hide","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"We visualise the ellipses using the Plots package:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"function plot_ellipse(plot, ellipse::Ellipse)\n    A, b, c = ellipse.A, ellipse.b, ellipse.c\n    θ = range(0, 2pi + 0.05; step = 0.05)\n    # Some linear algebra to convert θ into (x,y) coordinates.\n    x_y = √A \\ (√(b' * (A \\ b) - c) .* hcat(cos.(θ), sin.(θ)) .- (√A \\ b)')'\n    Plots.plot!(plot, x_y[1, :], x_y[2, :]; label = nothing, c = :navy)\n    return\nend\n\nplot = Plots.plot(; size = (600, 600))\nfor ellipse in ellipses\n    plot_ellipse(plot, ellipse)\nend\nplot","category":"page"},{"location":"tutorials/conic/min_ellipse/#Build-the-model","page":"Minimal ellipses","title":"Build the model","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Now let's build the model, using the change-of-variables P² = P^2 and P_q = P q. We'll recover the true value of P and q after the solve.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"model = Model(SCS.Optimizer)\n# We need to use a tighter tolerance for this example, otherwise the bounding\n# ellipse won't actually be bounding...\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\nm, n = length(ellipses), size(first(ellipses).A, 1)\n@variable(model, τ[1:m] >= 0)\n@variable(model, P²[1:n, 1:n], PSD)\n@variable(model, P_q[1:n])\n\nfor (i, ellipse) in enumerate(ellipses)\n    A, b, c = ellipse.A, ellipse.b, ellipse.c\n    X = [\n        #! format: off\n        (P² - τ[i] * A)   (P_q - τ[i] * b) zeros(n, n)\n        (P_q - τ[i] * b)' (-1 - τ[i] * c)  P_q'\n        zeros(n, n)       P_q              -P²\n        #! format: on\n    ]\n    @constraint(model, LinearAlgebra.Symmetric(X) <= 0, PSDCone())\nend","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"We cannot directly represent the objective log(det(P)), so we introduce the conic reformulation:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"@variable(model, log_det_P)\n@constraint(model, [log_det_P; 1; vec(P²)] in MOI.LogDetConeSquare(n))\n@objective(model, Max, log_det_P)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Now, solve the program:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/min_ellipse/#Results","page":"Minimal ellipses","title":"Results","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"After solving the model to optimality we can recover the solution in terms of P and q:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"P = sqrt(value.(P²))\nq = P \\ value.(P_q)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Finally, overlaying the solution in the plot we see the minimal area enclosing ellipsoid:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Plots.plot!(\n    plot,\n    [tuple(P \\ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)];\n    c = :crimson,\n    label = nothing,\n)","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"EditURL = \"https://github.com/jump-dev/CSDP.jl/blob/3a2e4c7c48af7660a5ba258a4cb6549ab0970366/README.md\"","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"(Image: )","category":"page"},{"location":"packages/CSDP/#CSDP.jl","page":"jump-dev/CSDP.jl","title":"CSDP.jl","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP.jl is a wrapper for the COIN-OR SemiDefinite Programming solver.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"a thin wrapper around the low-level C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/CSDP/#Affiliation","page":"jump-dev/CSDP.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The original algorithm is described by B. Borchers (1999). CSDP, A C Library for Semidefinite Programming. Optimization Methods and Software. 11(1), 613-623. [preprint]","category":"page"},{"location":"packages/CSDP/#License","page":"jump-dev/CSDP.jl","title":"License","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP.jl is licensed under the MIT License.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The underlying solver, coin-or/Csdp, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/CSDP/#Installation","page":"jump-dev/CSDP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"Install CSDP using Pkg.add:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"import Pkg\nPkg.add(\"CSDP\")","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In addition to installing the CSDP.jl package, this will also download and install the CSDP binaries. You do not need to install CSDP separately.","category":"page"},{"location":"packages/CSDP/#Use-with-JuMP","page":"jump-dev/CSDP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"To use CSDP with JuMP, use CSDP.Optimizer:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"using JuMP, CSDP\nmodel = Model(CSDP.Optimizer)\nset_attribute(model, \"maxiter\", 1000)","category":"page"},{"location":"packages/CSDP/#MathOptInterface-API","page":"jump-dev/CSDP.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The CSDP optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.Nonnegatives\nMOI.PositiveSemidefiniteConeTriangle","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/CSDP/#Options","page":"jump-dev/CSDP.jl","title":"Options","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The CSDP options are listed in the table below.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"Name Default Value Explanation\naxtol 1.0e-8 Tolerance for primal feasibility\natytol 1.0e-8 Tolerance for dual feasibility\nobjtol 1.0e-8 Tolerance for relative duality gap\npinftol 1.0e8 Tolerance for determining primal infeasibility\ndinftol 1.0e8 Tolerance for determining dual infeasibility\nmaxiter 100 Limit for the total number of iterations\nminstepfrac 0.90 The minstepfrac and maxstepfrac parameters determine how close to the edge of the feasible region CSDP will step\nmaxstepfrac 0.97 The minstepfrac and maxstepfrac parameters determine how close to the edge of the feasible region CSDP will step\nminstepp 1.0e-8 If the primal step is shorter than minstepp then CSDP declares a line search failure\nminstepd 1.0e-8 If the dual step is shorter than minstepd then CSDP declares a line search failure\nusexzgap 1 If usexzgap is 0 then CSDP will use the objective duality gap d - p instead of the XY duality gap ⟨Z, X⟩\ntweakgap 0 If tweakgap is set to 1, and usexzgap is set to 0, then CSDP will attempt to \"fix\" negative duality gaps\naffine 0 If affine is set to 1, then CSDP will take only primal-dual affine steps and not make use of the barrier term. This can be useful for some problems that do not have feasible solutions that are strictly in the interior of the cone of semidefinite matrices\nperturbobj 1 The perturbobj parameter determines whether the objective function will be perturbed to help deal with problems that have unbounded optimal solution sets. If perturbobj is 0, then the objective will not be perturbed. If perturbobj is 1, then the objective function will be perturbed by a default amount. Larger values of perturbobj (for example, 100) increase the size of the perturbation. This can be helpful in solving some difficult problems.\nfastmode 0 The fastmode parameter determines whether or not CSDP will skip certain time consuming operations that slightly improve the accuracy of the solutions. If fastmode is set to 1, then CSDP may be somewhat faster, but also somewhat less accurate\nprintlevel 1 The printlevel parameter determines how much debugging information is output. Use a printlevel of 0 for no output and a printlevel of 1 for normal output. Higher values of printlevel will generate more debugging output","category":"page"},{"location":"packages/CSDP/#Problem-representation","page":"jump-dev/CSDP.jl","title":"Problem representation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The primal is represented internally by CSDP as follows:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"max ⟨C, X⟩\n      A(X) = a\n         X ⪰ 0","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"where A(X) = [⟨A_1, X⟩, ..., ⟨A_m, X⟩]. The corresponding dual is:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"min ⟨a, y⟩\n     A'(y) - C = Z\n             Z ⪰ 0","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"where A'(y) = y_1A_1 + ... + y_mA_m","category":"page"},{"location":"packages/CSDP/#Termination-criteria","page":"jump-dev/CSDP.jl","title":"Termination criteria","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP will terminate successfully (or partially) in the following cases:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"If CSDP finds X, Z ⪰ 0 such that the following 3 tolerances are satisfied:\nprimal feasibility tolerance: ||A(x) - a||_2 / (1 + ||a||_2) < axtol\ndual feasibility tolerance: ||A'(y) - C - Z||_F / (1 + ||C||_F) < atytol\nrelative duality gap tolerance: gap / (1 + |⟨a, y⟩| + |⟨C, X⟩|) < objtol\nobjective duality gap: if usexygap is 0, gap = ⟨a, y⟩ - ⟨C, X⟩\nXY duality gap: if usexygap is 1, gap = ⟨Z, X⟩\nthen it returns 0.\nIf CSDP finds y and Z ⪰ 0 such that -⟨a, y⟩ / ||A'(y) - Z||_F > pinftol, it returns 1 with y such that ⟨a, y⟩ = -1.\nIf CSDP finds X ⪰ 0 such that ⟨C, X⟩ / ||A(X)||_2 > dinftol, it returns 2 with X such that ⟨C, X⟩ = 1.\nIf CSDP finds X, Z ⪰ 0 such that the following 3 tolerances are satisfied with 1000*axtol, 1000*atytol and 1000*objtol but at least one of them is not satisfied with axtol, atytol and objtol and cannot make progress, then it returns 3.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In addition, if the printlevel option is at least 1, the following will be printed:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"If the return code is 1, CSDP will print ⟨a, y⟩ and ||A'(y) - Z||_F\nIf the return code is 2, CSDP will print ⟨C, X⟩ and ||A(X)||_F\nOtherwise, CSDP will print\nthe primal/dual objective value,\nthe relative primal/dual infeasibility,\nthe objective duality gap ⟨a, y⟩ - ⟨C, X⟩ and objective relative duality gap (⟨a, y⟩ - ⟨C, X⟩) / (1 + |⟨a, y⟩| + |⟨C, X⟩|),\nthe XY duality gap ⟨Z, X⟩ and XY relative duality gap ⟨Z, X⟩ / (1 + |⟨a, y⟩| + |⟨C, X⟩|)\nand the DIMACS error measures.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In theory, for feasible primal and dual solutions, ⟨a, y⟩ - ⟨C, X⟩ = ⟨Z, X⟩, so the objective and XY duality gap should be equivalent. However, in practice, there are sometimes solution which satisfy primal and dual feasibility tolerances but have objective duality gap which are not close to XY duality gap. In some cases, the objective duality gap may even become negative (hence the tweakgap option). This is the reason usexygap is 1 by default.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP considers that X ⪰ 0 (resp. Z ⪰ 0) is satisfied when the Cholesky factorizations can be computed. In practice, this is somewhat more conservative than simply requiring all eigenvalues to be nonnegative.","category":"page"},{"location":"packages/CSDP/#Status","page":"jump-dev/CSDP.jl","title":"Status","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The table below shows how the different CSDP statuses are converted to the MathOptInterface statuses.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP code State Description MOI status\n0 Success SDP solved MOI.OPTIMAL\n1 Success The problem is primal infeasible, and we have a certificate MOI.INFEASIBLE\n2 Success The problem is dual infeasible, and we have a certificate MOI.DUAL_INFEASIBLE\n3 Partial Success A solution has been found, but full accuracy was not achieved MOI.ALMOST_OPTIMAL\n4 Failure Maximum iterations reached MOI.ITERATION_LIMIT\n5 Failure Stuck at edge of primal feasibility MOI.SLOW_PROGRESS\n6 Failure Stuck at edge of dual infeasibility MOI.SLOW_PROGRESS\n7 Failure Lack of progress MOI.SLOW_PROGRESS\n8 Failure X, Z, or O was singular MOI.NUMERICAL_ERROR\n9 Failure Detected NaN or Inf values MOI.NUMERICAL_ERROR","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"EditURL = \"start_values.jl\"","category":"page"},{"location":"tutorials/conic/start_values/#Primal-and-dual-warm-starts","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"","category":"section"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Some conic solvers have the ability to set warm-starts for the primal and dual solution. This can improve performance, particularly if you are repeatedly solving a sequence of related problems.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"tip: Tip\nSee set_start_values for a generic implementation of this function that was added to JuMP after this tutorial was written.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"In this tutorial, we demonstrate how to write a function that sets the primal and dual starts as the optimal solution stored in a model. It is intended to be a starting point for which you can modify if you want to do something similar in your own code.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"warning: Warning\nThis tutorial does not set start values for nonlinear models.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"using JuMP\nimport SCS","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"The main component of this tutorial is the following function. The most important observation is that we cache all of the solution values first, and then we modify the model second. (Alternating between querying a value and modifying the model is not allowed in JuMP.)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"function set_optimal_start_values(model::Model)\n    # Store a mapping of the variable primal solution\n    variable_primal = Dict(x => value(x) for x in all_variables(model))\n    # In the following, we loop through every constraint and store a mapping\n    # from the constraint index to a tuple containing the primal and dual\n    # solutions.\n    constraint_solution = Dict()\n    for (F, S) in list_of_constraint_types(model)\n        # We add a try-catch here because some constraint types might not\n        # support getting the primal or dual solution.\n        try\n            for ci in all_constraints(model, F, S)\n                constraint_solution[ci] = (value(ci), dual(ci))\n            end\n        catch\n            @info(\"Something went wrong getting $F-in-$S. Skipping\")\n        end\n    end\n    # Now we can loop through our cached solutions and set the starting values.\n    for (x, primal_start) in variable_primal\n        set_start_value(x, primal_start)\n    end\n    for (ci, (primal_start, dual_start)) in constraint_solution\n        set_start_value(ci, primal_start)\n        set_dual_start_value(ci, dual_start)\n    end\n    return\nend","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"To test our function, we use the following linear program:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"model = Model(SCS.Optimizer)\n@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) <= 1)\n@objective(model, Max, sum(i * x[i] for i in 1:3))\noptimize!(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"By looking at the log, we can see that SCS took 75 iterations to find the optimal solution. Now we set the optimal solution as our starting point:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"set_optimal_start_values(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"and we re-optimize:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"optimize!(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Now the optimization terminates after 0 iterations because our starting point is already optimal.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Note that some solvers do not support setting some parts of the starting solution, for example, they may support only set_start_value for variables. If you encounter an UnsupportedSupported attribute error for MOI.VariablePrimalStart, MOI.ConstraintPrimalStart, or MOI.ConstraintDualStart, comment out the corresponding part of the set_optimal_start_values function.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/solutions/#jump_solutions","page":"Solutions","title":"Solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"This section of the manual describes how to access a solved solution to a problem. It uses the following model as an example:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x >= 0)\n           @variable(model, y[[:a, :b]] <= 1)\n           @objective(model, Max, -12x - 20y[:a])\n           @expression(model, my_expr, 6x + 8y[:a])\n           @constraint(model, my_expr >= 100)\n           @constraint(model, c1, 7x + 12y[:a] >= 120)\n           optimize!(model)\n           print(model)\n       end\nMax -12 x - 20 y[a]\nSubject to\n 6 x + 8 y[a] ≥ 100\n c1 : 7 x + 12 y[a] ≥ 120\n x ≥ 0\n y[a] ≤ 1\n y[b] ≤ 1","category":"page"},{"location":"manual/solutions/#Solutions-summary","page":"Solutions","title":"Solutions summary","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"solution_summary can be used for checking the summary of the optimization solutions.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> solution_summary(model)\n* Solver : HiGHS\n\n* Status\n  Result count       : 1\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"kHighsModelStatusOptimal\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : FEASIBLE_POINT\n  Objective value    : -2.05143e+02\n  Objective bound    : -0.00000e+00\n  Relative gap       : Inf\n  Dual objective value : -2.05143e+02\n\n* Work counters\n  Solve time (sec)   : 6.70987e-04\n  Simplex iterations : 2\n  Barrier iterations : 0\n  Node count         : -1\n\njulia> solution_summary(model; verbose = true)\n* Solver : HiGHS\n\n* Status\n  Result count       : 1\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"kHighsModelStatusOptimal\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : FEASIBLE_POINT\n  Objective value    : -2.05143e+02\n  Objective bound    : -0.00000e+00\n  Relative gap       : Inf\n  Dual objective value : -2.05143e+02\n  Primal solution :\n    x : 1.54286e+01\n    y[a] : 1.00000e+00\n    y[b] : 1.00000e+00\n  Dual solution :\n    c1 : 1.71429e+00\n\n* Work counters\n  Solve time (sec)   : 6.70987e-04\n  Simplex iterations : 2\n  Barrier iterations : 0\n  Node count         : -1","category":"page"},{"location":"manual/solutions/#Why-did-the-solver-stop?","page":"Solutions","title":"Why did the solver stop?","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Usetermination_status to understand why the solver stopped.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The MOI.TerminationStatusCode enum describes the full list of statuses that could be returned.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Common return values include OPTIMAL, LOCALLY_SOLVED, INFEASIBLE, DUAL_INFEASIBLE, and TIME_LIMIT.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nA return status of OPTIMAL means the solver found (and proved) a globally optimal solution. A return status of LOCALLY_SOLVED means the solver found a locally optimal solution (which may also be globally optimal, but it could not prove so).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nA return status of DUAL_INFEASIBLE does not guarantee that the primal is unbounded. When the dual is infeasible, the primal is unbounded if there exists a feasible primal solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use raw_status to get a solver-specific string explaining why the optimization stopped:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> raw_status(model)\n\"kHighsModelStatusOptimal\"","category":"page"},{"location":"manual/solutions/#Primal-solutions","page":"Solutions","title":"Primal solutions","text":"","category":"section"},{"location":"manual/solutions/#Primal-solution-status","page":"Solutions","title":"Primal solution status","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use primal_status to return an MOI.ResultStatusCode enum describing the status of the primal solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_status(model)\nFEASIBLE_POINT::ResultStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Other common returns are NO_SOLUTION, and INFEASIBILITY_CERTIFICATE. The first means that the solver doesn't have a solution to return, and the second means that the primal solution is a certificate of dual infeasibility (a primal unbounded ray).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use has_values, which returns true if there is a solution that can be queried, and false otherwise.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> has_values(model)\ntrue","category":"page"},{"location":"manual/solutions/#Objective-values","page":"Solutions","title":"Objective values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The objective value of a solved problem can be obtained via objective_value. The best known bound on the optimal objective value can be obtained via objective_bound. If the solver supports it, the value of the dual objective can be obtained via dual_objective_value.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> objective_value(model)\n-205.14285714285714\n\njulia> objective_bound(model)  # HiGHS only implements objective bound for MIPs\n-0.0\n\njulia> dual_objective_value(model)\n-205.1428571428571","category":"page"},{"location":"manual/solutions/#Primal-solution-values","page":"Solutions","title":"Primal solution values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the solver has a primal solution to return, use value to access it:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(x)\n15.428571428571429","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Broadcast value over containers:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value.(y)\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n 1.0\n 1.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"value also works on expressions:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(my_expr)\n100.57142857142857","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"and constraints:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(c1)\n120.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nCalling value on a constraint returns the constraint function evaluated at the solution.","category":"page"},{"location":"manual/solutions/#Dual-solutions","page":"Solutions","title":"Dual solutions","text":"","category":"section"},{"location":"manual/solutions/#Dual-solution-status","page":"Solutions","title":"Dual solution status","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use dual_status to return an MOI.ResultStatusCode enum describing the status of the dual solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual_status(model)\nFEASIBLE_POINT::ResultStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Other common returns are NO_SOLUTION, and INFEASIBILITY_CERTIFICATE. The first means that the solver doesn't have a solution to return, and the second means that the dual solution is a certificate of primal infeasibility (a dual unbounded ray).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use has_duals, which returns true if there is a solution that can be queried, and false otherwise.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> has_duals(model)\ntrue","category":"page"},{"location":"manual/solutions/#Dual-solution-values","page":"Solutions","title":"Dual solution values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the solver has a dual solution to return, use dual to access it:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual(c1)\n1.7142857142857142","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the duals of variable bounds using LowerBoundRef, UpperBoundRef, and FixRef:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual(LowerBoundRef(x))\n0.0\n\njulia> dual.(UpperBoundRef.(y))\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n -0.5714285714285694\n  0.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nJuMP's definition of duality is independent of the objective sense. That is, the sign of feasible duals associated with a constraint depends on the direction of the constraint and not whether the problem is maximization or minimization. This is a different convention from linear programming duality in some common textbooks. If you have a linear program, and you want the textbook definition, you probably want to use shadow_price and reduced_cost instead.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> shadow_price(c1)\n1.7142857142857142\n\njulia> reduced_cost(x)\n-0.0\n\njulia> reduced_cost.(y)\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n  0.5714285714285694\n -0.0","category":"page"},{"location":"manual/solutions/#Recommended-workflow","page":"Solutions","title":"Recommended workflow","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The recommended workflow for solving a model and querying the solution is something like the following:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           if termination_status(model) == OPTIMAL\n               println(\"Solution is optimal\")\n           elseif termination_status(model) == TIME_LIMIT && has_values(model)\n               println(\"Solution is suboptimal due to a time limit, but a primal solution is available\")\n           else\n               error(\"The model was not solved correctly.\")\n           end\n           println(\"  objective value = \", objective_value(model))\n           if primal_status(model) == FEASIBLE_POINT\n               println(\"  primal solution: x = \", value(x))\n           end\n           if dual_status(model) == FEASIBLE_POINT\n               println(\"  dual solution: c1 = \", dual(c1))\n           end\n       end\nSolution is optimal\n  objective value = -205.14285714285714\n  primal solution: x = 15.428571428571429\n  dual solution: c1 = 1.7142857142857142","category":"page"},{"location":"manual/solutions/#OptimizeNotCalled-errors","page":"Solutions","title":"OptimizeNotCalled errors","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Due to differences in how solvers cache solutions internally, modifying a model after calling optimize! will reset the model into the MOI.OPTIMIZE_NOT_CALLED state. If you then attempt to query solution information, an OptimizeNotCalled error will be thrown.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you are iteratively querying solution information and modifying a model, query all the results first, then modify the problem.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"For example, instead of:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> x_val = value(x)\n┌ Warning: The model has been modified since the last call to `optimize!` (or `optimize!` has not been called yet). If you are iteratively querying solution information and modifying a model, query all the results first, then modify the model.\n└ @ JuMP ~/work/JuMP.jl/JuMP.jl/src/optimizer_interface.jl:712\nERROR: OptimizeNotCalled()\nStacktrace:\n[...]\n\njulia> termination_status(model)\nOPTIMIZE_NOT_CALLED::TerminationStatusCode = 0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"do","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model);\n\njulia> x_val = value(x)\n0.0\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> set_lower_bound(x, x_val)\n\njulia> termination_status(model)\nOPTIMIZE_NOT_CALLED::TerminationStatusCode = 0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you know that your particular solver supports querying solution information after modifications, you can use direct_model to bypass the MOI.OPTIMIZE_NOT_CALLED state:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> x_val = value(x)\n0.0\n\njulia> set_lower_bound(x, x_val)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nBe careful doing this. If your particular solver does not support querying solution information after modification, it may silently return incorrect solutions or throw an error.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"# TODO: How to accurately measure the solve time.","category":"page"},{"location":"manual/solutions/#Accessing-attributes","page":"Solutions","title":"Accessing attributes","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"MathOptInterface defines many model attributes that can be queried. Some attributes can be directly accessed by getter functions. These include:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"solve_time\nrelative_gap\nsimplex_iterations\nbarrier_iterations\nnode_count","category":"page"},{"location":"manual/solutions/#Sensitivity-analysis-for-LP","page":"Solutions","title":"Sensitivity analysis for LP","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Given an LP problem and an optimal solution corresponding to a basis, we can question how much an objective coefficient or standard form right-hand side coefficient (c.f., normalized_rhs) can change without violating primal or dual feasibility of the basic solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Note that not all solvers compute the basis, and for sensitivity analysis, the solver interface must implement MOI.ConstraintBasisStatus.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"tip: Tip\nRead the Sensitivity analysis of a linear program for more information on sensitivity analysis.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To give a simple example, we could analyze the sensitivity of the optimal solution to the following (non-degenerate) LP problem:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x[1:2])\n           set_lower_bound(x[2], -0.5)\n           set_upper_bound(x[2], 0.5)\n           @constraint(model, c1, x[1] + x[2] <= 1)\n           @constraint(model, c2, x[1] - x[2] <= 1)\n           @objective(model, Max, x[1])\n           print(model)\n       end\nMax x[1]\nSubject to\n c1 : x[1] + x[2] ≤ 1\n c2 : x[1] - x[2] ≤ 1\n x[2] ≥ -0.5\n x[2] ≤ 0.5","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To analyze the sensitivity of the problem we could check the allowed perturbation ranges of, for example, the cost coefficients and the right-hand side coefficient of the constraint c1 as follows:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n  1.0\n -0.0\n\njulia> report = lp_sensitivity_report(model);\n\njulia> x1_lo, x1_hi = report[x[1]]\n(-1.0, Inf)\n\njulia> println(\"The objective coefficient of x[1] could decrease by $(x1_lo) or increase by $(x1_hi).\")\nThe objective coefficient of x[1] could decrease by -1.0 or increase by Inf.\n\njulia> x2_lo, x2_hi = report[x[2]]\n(-1.0, 1.0)\n\njulia> println(\"The objective coefficient of x[2] could decrease by $(x2_lo) or increase by $(x2_hi).\")\nThe objective coefficient of x[2] could decrease by -1.0 or increase by 1.0.\n\njulia> c_lo, c_hi = report[c1]\n(-1.0, 1.0)\n\njulia> println(\"The RHS of c1 could decrease by $(c_lo) or increase by $(c_hi).\")\nThe RHS of c1 could decrease by -1.0 or increase by 1.0.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The range associated with a variable is the range of the allowed perturbation of the corresponding objective coefficient. Note that the current primal solution remains optimal within this range; however the corresponding dual solution might change since a cost coefficient is perturbed. Similarly, the range associated with a constraint is the range of the allowed perturbation of the corresponding right-hand side coefficient. In this range the current dual solution remains optimal, but the optimal primal solution might change.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the problem is degenerate, there are multiple optimal bases and hence these ranges might not be as intuitive and seem too narrow, for example, a larger cost coefficient perturbation might not invalidate the optimality of the current primal solution. Moreover, if a problem is degenerate, due to finite precision, it can happen that, for example, a perturbation seems to invalidate a basis even though it doesn't (again providing too narrow ranges). To prevent this, increase the atol keyword argument to lp_sensitivity_report. Note that this might make the ranges too wide for numerically challenging instances. Thus, do not blindly trust these ranges, especially not for highly degenerate or numerically unstable instances.","category":"page"},{"location":"manual/solutions/#Conflicts","page":"Solutions","title":"Conflicts","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"When the model you input is infeasible, some solvers can help you find the cause of this infeasibility by offering a conflict, that is, a subset of the constraints that create this infeasibility. Depending on the solver, this can also be called an IIS (irreducible inconsistent subsystem).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If supported by the solver, use compute_conflict! to trigger the computation of a conflict. Once this process is finished, query the MOI.ConflictStatus attribute to check if a conflict was found.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If found, copy the IIS to a new model using copy_conflict, which you can then print or write to a file for easier debugging:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> using JuMP\n\njulia> import Gurobi\n\njulia> model = Model(Gurobi.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: Gurobi\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, c1, x >= 2)\nc1 : x ≥ 2.0\n\njulia> @constraint(model, c2, x <= 1)\nc2 : x ≤ 1.0\n\njulia> optimize!(model)\n\njulia> compute_conflict!(model)\n\njulia> if get_attribute(model, MOI.ConflictStatus()) == MOI.CONFLICT_FOUND\n           iis_model, _ = copy_conflict(model)\n           print(iis_model)\n       end\nFeasibility\nSubject to\n c1 : x ≥ 2.0\n c2 : x ≤ 1.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you need more control over the list of constraints that appear in the conflict, iterate over the list of constraints and query the MOI.ConstraintConflictStatus attribute:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> list_of_conflicting_constraints = ConstraintRef[]\nConstraintRef[]\n\njulia> for (F, S) in list_of_constraint_types(model)\n           for con in all_constraints(model, F, S)\n               if get_attribute(con, MOI.ConstraintConflictStatus()) == MOI.IN_CONFLICT\n                   push!(list_of_conflicting_constraints, con)\n               end\n           end\n       end\n\njulia> list_of_conflicting_constraints\n2-element Vector{ConstraintRef}:\n c1 : x ≥ 2.0\n c2 : x ≤ 1.0","category":"page"},{"location":"manual/solutions/#Multiple-solutions","page":"Solutions","title":"Multiple solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Some solvers support returning multiple solutions. You can check how many solutions are available to query using result_count.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Functions for querying the solutions, for example, primal_status, dual_status, value, dual, and solution_summary all take an additional keyword argument result which can be used to specify which result to return.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nEven if termination_status is OPTIMAL, some of the returned solutions may be suboptimal. However, if the solver found at least one optimal solution, then result = 1 will always return an optimal solution. Use objective_value to assess the quality of the remaining solutions.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> using JuMP\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> import HiGHS\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer));\n\njulia> set_attribute(model, MOA.Algorithm(), MOA.Dichotomy())\n\njulia> set_silent(model)\n\njulia> @variable(model, x1 >= 0)\nx1\n\njulia> @variable(model, 0 <= x2 <= 3)\nx2\n\njulia> @objective(model, Min, [3x1 + x2, -x1 - 2x2])\n2-element Vector{AffExpr}:\n 3 x1 + x2\n -x1 - 2 x2\n\njulia> @constraint(model, 3x1 - x2 <= 6)\n3 x1 - x2 ≤ 6\n\njulia> optimize!(model)\n\njulia> solution_summary(model; result = 1)\n* Solver : MOA[algorithm=MultiObjectiveAlgorithms.Dichotomy, optimizer=HiGHS]\n\n* Status\n  Result count       : 3\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"Solve complete. Found 3 solution(s)\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : NO_SOLUTION\n  Objective value    : [0.00000e+00,0.00000e+00]\n  Objective bound    : [0.00000e+00,-9.00000e+00]\n  Relative gap       : Inf\n  Dual objective value : -9.00000e+00\n\n* Work counters\n  Solve time (sec)   : 1.82720e-03\n  Simplex iterations : 1\n  Barrier iterations : 0\n  Node count         : -1\n\njulia> for i in 1:result_count(model)\n           println(\"Solution $i\")\n           println(\"   x = \", value.([x1, x2]; result = i))\n           println(\" obj = \", objective_value(model; result = i))\n       end\nSolution 1\n   x = [0.0, 0.0]\n obj = [0.0, 0.0]\nSolution 2\n   x = [0.0, 3.0]\n obj = [3.0, -6.0]\nSolution 3\n   x = [3.0, 3.0]\n obj = [12.0, -9.0]","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"tip: Tip\nThe Multi-objective knapsack tutorial provides more examples of querying multiple solutions.","category":"page"},{"location":"manual/solutions/#Checking-feasibility-of-solutions","page":"Solutions","title":"Checking feasibility of solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To check the feasibility of a primal solution, use primal_feasibility_report, which takes a model, a dictionary mapping each variable to a primal solution value (defaults to the last solved solution), and a tolerance atol (defaults to 0.0).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The function returns a dictionary which maps the infeasible constraint references to the distance between the primal value of the constraint and the nearest point in the corresponding set. A point is classed as infeasible if the distance is greater than the supplied tolerance atol.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"# Add a filter here because the output of the dictionary is not ordered, and\n# changes in printing order will cause the doctest to fail.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 1, Int);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, c1, x + y <= 1.95);\n\njulia> point = Dict(x => 1.9, y => 0.06);\n\njulia> primal_feasibility_report(model, point)\nDict{Any, Float64} with 2 entries:\n  x integer         => 0.1\n  c1 : x + y ≤ 1.95 => 0.01\n\njulia> primal_feasibility_report(model, point; atol = 0.02)\nDict{Any, Float64} with 1 entry:\n  x integer => 0.1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the point is feasible, an empty dictionary is returned:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_feasibility_report(model, Dict(x => 1.0, y => 0.0))\nDict{Any, Float64}()","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To use the primal solution from a solve, omit the point argument:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> primal_feasibility_report(model; atol = 0.0)\nDict{Any, Float64}()","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Calling primal_feasibility_report without the point argument is useful when primal_status is FEASIBLE_POINT or NEARLY_FEASIBLE_POINT, and you want to assess the solution quality.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nTo apply primal_feasibility_report to infeasible models, you must also provide a candidate point (solvers generally do not provide one). To diagnose the source of infeasibility, see Conflicts.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Pass skip_mising = true to skip constraints which contain variables that are not in point:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_feasibility_report(model, Dict(x => 2.1); skip_missing = true)\nDict{Any, Float64} with 1 entry:\n  x integer => 0.1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use the functional form, where the first argument is a function that maps variables to their primal values:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> primal_feasibility_report(v -> value(v), model)\nDict{Any, Float64}()","category":"page"},{"location":"extensions/DimensionalData/#DimensionalData.jl","page":"rafaqz/DimensionalData.jl","title":"DimensionalData.jl","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"DimensionalData.jl provides tools and abstractions for working with rectangular arrays that have named dimensions.","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"compat: Compat\nUsing the DimensionalData extension with JuMP requires Julia v1.9 or later.","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"The DimensionalData extension in JuMP lets you construct a DimensionalData.DimArray as an alternative to Containers.DenseAxisArray in the JuMP macros.","category":"page"},{"location":"extensions/DimensionalData/#License","page":"rafaqz/DimensionalData.jl","title":"License","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"DimensionalData.jl is licensed under the MIT license.","category":"page"},{"location":"extensions/DimensionalData/#Installation","page":"rafaqz/DimensionalData.jl","title":"Installation","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Install DimensionalData using Pkg.add:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"import Pkg\nPkg.add(\"DimensionalData\")","category":"page"},{"location":"extensions/DimensionalData/#Use-with-JuMP","page":"rafaqz/DimensionalData.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Activate the extension by loading both JuMP and DimensionalData:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> using JuMP, DimensionalData","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Then, pass container = DimensionalData.DimArray in the @variable, @constraint, or @expression macros:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> model = Model();\n\njulia> @variable(\n           model,\n           x[i = 2:4, j = [\"a\", \"b\"]] >= i,\n           container = DimensionalData.DimArray,\n       )\n3×2 DimArray{VariableRef,2} with dimensions:\n  Dim{:i} Sampled{Int64} 2:4 ForwardOrdered Regular Points,\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n    \"a\"     \"b\"\n 2  x[2,a]  x[2,b]\n 3  x[3,a]  x[3,b]\n 4  x[4,a]  x[4,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Here x is a DimensionalData.Dim array object, so indexing uses the DimensionalData syntax:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> x[At(2), At(\"a\")]\nx[2,a]\n\njulia> x[2, 2]\nx[3,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"You can use container = DimensionalData.DimArray in the @expression macro:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> @expression(\n           model,\n           expr[j = [\"a\", \"b\"]],\n           sum(x[At(i), At(j)] for i in 2:4),\n           container = DimensionalData.DimArray,\n       )\n2-element DimArray{AffExpr,1} with dimensions:\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n \"a\"  x[2,a] + x[3,a] + x[4,a]\n \"b\"  x[2,b] + x[3,b] + x[4,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"and in @constraint:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> @constraint(\n           model,\n           [j = [\"a\", \"b\"]],\n           expr[At(j)] <= 1,\n           container = DimensionalData.DimArray,\n       )\n2-element DimArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},1} with dimensions:\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n \"a\"  x[2,a] + x[3,a] + x[4,a] ≤ 1\n \"b\"  x[2,b] + x[3,b] + x[4,b] ≤ 1","category":"page"},{"location":"extensions/DimensionalData/#Documentation","page":"rafaqz/DimensionalData.jl","title":"Documentation","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"See the DimensionalData.jl documentation for more details on the syntax and features of DimensionalData.DimArray.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"EditURL = \"getting_started_with_JuMP.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Getting-started-with-JuMP","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"This tutorial is aimed at providing a quick introduction to writing and solving optimization models with JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"If you're new to Julia, start by reading Getting started with Julia.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-JuMP?","page":"Getting started with JuMP","title":"What is JuMP?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP (\"Julia for Mathematical Programming\") is an open-source modeling language that is embedded in Julia. It allows users to formulate various classes of optimization problems (linear, mixed-integer, quadratic, conic quadratic, semidefinite, and nonlinear) with easy-to-read code. These problems can then be solved using state-of-the-art open-source and commercial solvers.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP also makes advanced optimization techniques easily accessible from a high-level language.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-a-solver?","page":"Getting started with JuMP","title":"What is a solver?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"A solver is a software package that incorporates algorithms for finding solutions to one or more classes of problem.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"For example, HiGHS is a solver for linear programming (LP) and mixed integer programming (MIP) problems. It incorporates algorithms such as the simplex method and the interior-point method.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"The Supported-solvers table lists the open-source and commercial solvers that JuMP currently supports.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-MathOptInterface?","page":"Getting started with JuMP","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Each solver has its own concepts and data structures for representing optimization models and obtaining results.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"MathOptInterface (MOI) is an abstraction layer that JuMP uses to convert from the problem written in JuMP to the solver-specific data structures for each solver.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"MOI can be used directly, or through a higher-level modeling interface like JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Because JuMP is built on top of MOI, you'll often see the MathOptInterface. prefix displayed when JuMP types are printed. However, you'll only need to understand and interact with MOI to accomplish advanced tasks such as creating solver-independent callbacks.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Installation","page":"Getting started with JuMP","title":"Installation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP is a package for Julia. From Julia, JuMP is installed by using the built-in package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"You also need to include a Julia package which provides an appropriate solver. One such solver is HiGHS.Optimizer, which is provided by the HiGHS.jl package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"import Pkg\nPkg.add(\"HiGHS\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"See Installation Guide for a list of other solvers you can use.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#An-example","page":"Getting started with JuMP","title":"An example","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Let's solve the following linear programming problem using JuMP and HiGHS. We will first look at the complete code to solve the problem and then go through it step by step.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Here's the problem:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"beginaligned\n min  12x + 20y \n textst  6x + 8y geq 100 \n  7x + 12y geq 120 \n  x geq 0 \n  y in 0 3 \nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"And here's the code to solve this problem:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using JuMP\nusing HiGHS\nmodel = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\nprint(model)\noptimize!(model)\ntermination_status(model)\nprimal_status(model)\ndual_status(model)\nobjective_value(model)\nvalue(x)\nvalue(y)\nshadow_price(c1)\nshadow_price(c2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Step-by-step","page":"Getting started with JuMP","title":"Step-by-step","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Once JuMP is installed, to use JuMP in your programs write:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using JuMP","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We also need to include a Julia package which provides an appropriate solver. We want to use HiGHS.Optimizer here which is provided by the HiGHS.jl package:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using HiGHS","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP builds problems incrementally in a Model object. Create a model by passing an optimizer to the Model function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Variables are modeled using @variable:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, x >= 0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"info: Info\nThe macro creates a new Julia object, x, in the current scope. We could have made this more explicit by writing:x = @variable(model, x >= 0)but the macro does this automatically for us to save writing x twice.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Variables can have lower and upper bounds:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, 0 <= y <= 30)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"The objective is set using @objective:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@objective(model, Min, 12x + 20y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Constraints are modeled using @constraint. Here, c1 and c2 are the names of our constraint:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Call print to display the model:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"print(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"To solve the optimization problem, call the optimize! function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"optimize!(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"info: Info\nThe ! after optimize is part of the name. It's nothing special. Julia has a convention that functions which mutate their arguments should end in !. A common example is push!.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Now let's see what information we can query about the solution.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"termination_status tells us why the solver stopped:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"termination_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"In this case, the solver found an optimal solution.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Check primal_status to see if the solver found a primal feasible point:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"primal_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"and dual_status to see if the solver found a dual feasible point:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"dual_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Now we know that our solver found an optimal solution, and that it has a primal and a dual solution to query.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Query the objective value using objective_value:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"objective_value(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"the primal solution using value:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"value(x)\nvalue(y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"and the dual solution using shadow_price:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"shadow_price(c1)\nshadow_price(c2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"That's it for our simple model. In the rest of this tutorial, we expand on some of the basic JuMP operations.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Model-basics","page":"Getting started with JuMP","title":"Model basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a model by passing an optimizer:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Alternatively, call set_optimizer at any point before calling optimize!:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()\nset_optimizer(model, HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"For some solvers, you can also use direct_model, which offers a more efficient connection to the underlying solver:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = direct_model(HiGHS.Optimizer())","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"warning: Warning\nSome solvers do not support direct_model!","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Solver-Options","page":"Getting started with JuMP","title":"Solver Options","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Pass options to solvers with optimizer_with_attributes:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model =\n    Model(optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"note: Note\nThese options are solver-specific. To find out the various options available, see the GitHub README of the individual solver packages. The link to each solver's GitHub page is in the Supported solvers table.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"You can also pass options with set_attribute:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)\nset_attribute(model, \"output_flag\", false)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Solution-basics","page":"Getting started with JuMP","title":"Solution basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We saw above how to use termination_status and primal_status to understand the solution returned by the solver.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"However, only query solution attributes like value and objective_value if there is an available solution. Here's a recommended way to check:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"function solve_infeasible()\n    model = Model(HiGHS.Optimizer)\n    @variable(model, 0 <= x <= 1)\n    @variable(model, 0 <= y <= 1)\n    @constraint(model, x + y >= 3)\n    @objective(model, Max, x + 2y)\n    optimize!(model)\n    if termination_status(model) != OPTIMAL\n        @warn(\"The model was not solved correctly.\")\n        return\n    end\n    return value(x), value(y)\nend\n\nsolve_infeasible()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Variable-basics","page":"Getting started with JuMP","title":"Variable basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Let's create a new empty model to explain some of the variable syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Variable-bounds","page":"Getting started with JuMP","title":"Variable bounds","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"All of the variables we have created till now have had a bound. We can also create a free variable.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, free_x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"While creating a variable, instead of using the <= and >= syntax, we can also use the lower_bound and upper_bound keyword arguments.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, keyword_x, lower_bound = 1, upper_bound = 2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can query whether a variable has a bound using the has_lower_bound and has_upper_bound functions. The values of the bound can be obtained using the lower_bound and upper_bound functions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"has_upper_bound(keyword_x)\nupper_bound(keyword_x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Note querying the value of a bound that does not exist will result in an error.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"try                         #hide\n    lower_bound(free_x)\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#tutorial_variable_container","page":"Getting started with JuMP","title":"Containers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We have already seen how to add a single variable to a model using the @variable macro. Now let's look at ways to add multiple variables to a model.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP provides data structures for adding collections of variables to a model. These data structures are referred to as containers and are of three types: Arrays, DenseAxisArrays, and SparseAxisArrays.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Arrays","page":"Getting started with JuMP","title":"Arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP arrays are created when you have integer indices that start at 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, a[1:2, 1:2])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"a[1, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"a[2, :]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an n-dimensional variable x in R^n with bounds l le x le u (l u in R^n) as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"n = 10\nl = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10];\nu = [10, 11, 12, 13, 14, 15, 16, 17, 18, 19];\n\n@variable(model, l[i] <= x[i = 1:n] <= u[i])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also create variable bounds that depend upon the indices:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, y[i = 1:2, j = 1:2] >= 2i + j)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#DenseAxisArrays","page":"Getting started with JuMP","title":"DenseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"DenseAxisArrays are used when the indices are not one-based integer ranges. The syntax is similar except with an arbitrary vector as an index as opposed to a one-based range:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, z[i = 2:3, j = 1:2:3] >= 0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Indices do not have to be integers. They can be any Julia type:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, w[1:5, [\"red\", \"blue\"]] <= 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a DenseAxisArray as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"z[2, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"w[2:3, [\"red\", \"blue\"]]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"See Forcing the container type for more details.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#SparseAxisArrays","page":"Getting started with JuMP","title":"SparseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"SparseAxisArrays are created when the indices do not form a Cartesian product. For example, this applies when indices have a dependence upon previous indices (called triangular indexing):","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, u[i = 1:2, j = i:3])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also conditionally create variables by adding a comparison check that depends upon the named indices and is separated from the indices by a semi-colon ;:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, v[i = 1:9; mod(i, 3) == 0])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a DenseAxisArray as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"u[1, 2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"v[[3, 6]]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Integrality","page":"Getting started with JuMP","title":"Integrality","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP can create binary and integer variables. Binary variables are constrained to the set  0 1, and integer variables are constrained to the set mathbbZ.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Integer-variables","page":"Getting started with JuMP","title":"Integer variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an integer variable by passing  Int:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, integer_x, Int)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or setting the integer keyword to true:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, integer_z, integer = true)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Binary-variables","page":"Getting started with JuMP","title":"Binary variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a binary variable by passing Bin:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, binary_x, Bin)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or setting the binary keyword to true:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, binary_z, binary = true)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Constraint-basics","page":"Getting started with JuMP","title":"Constraint basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We'll need a new model to explain some of the constraint  basics:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()\n@variable(model, x)\n@variable(model, y)\n@variable(model, z[1:10]);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#tutorial_constraint_container","page":"Getting started with JuMP","title":"Containers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Just as we had containers for variables, JuMP also provides Arrays, DenseAxisArrays, and SparseAxisArrays for storing collections of constraints. Examples for each container type are given below.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Arrays-2","page":"Getting started with JuMP","title":"Arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an Array of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:3], i * x <= i + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#DenseAxisArrays-2","page":"Getting started with JuMP","title":"DenseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an DenseAxisArray of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:2, j = 2:3], i * x <= j + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#SparseAxisArrays-2","page":"Getting started with JuMP","title":"SparseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an SparseAxisArray of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:2, j = 1:2; i != j], i * x <= j + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Constraints-in-a-loop","page":"Getting started with JuMP","title":"Constraints in a loop","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can add constraints using regular Julia loops:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"for i in 1:3\n    @constraint(model, 6x + 4y >= 5i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or use for each loops inside the @constraint macro:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i in 1:3], 6x + 4y >= 5i)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also create constraints such as sum _i = 1^10 z_i leq 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, sum(z[i] for i in 1:10) <= 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Objective-functions","page":"Getting started with JuMP","title":"Objective functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Set an objective function with @objective:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, y >= 0)\n@objective(model, Min, 2x + y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a maximization objective using Max:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@objective(model, Max, 2x + y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"tip: Tip\nCalling @objective multiple times will over-write the previous objective. This can be useful when you want to solve the same problem with different objectives.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Vectorized-syntax","page":"Getting started with JuMP","title":"Vectorized syntax","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also add constraints and an objective to JuMP using vectorized linear algebra. We'll illustrate this by solving an LP in standard form that is,","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"beginaligned\n min  c^T x \n textst  A x = b \n  x ge 0\nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"vector_model = Model(HiGHS.Optimizer)\nA = [1 1 9 5; 3 5 0 8; 2 0 6 13]\nb = [7, 3, 5]\nc = [1, 3, 5, 2]\n@variable(vector_model, x[1:4] >= 0)\n@constraint(vector_model, A * x .== b)\n@objective(vector_model, Min, c' * x)\noptimize!(vector_model)\nobjective_value(vector_model)","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\n    import HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/complex/#Complex-number-support","page":"Complex number support","title":"Complex number support","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"This page explains the complex-valued variables and constraints that JuMP supports. For a worked-example using these features, read the Quantum state discrimination tutorial.","category":"page"},{"location":"manual/complex/#Complex-valued-variables","page":"Complex number support","title":"Complex-valued variables","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Create a complex-valued variable using ComplexPlane:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, x in ComplexPlane())\nreal(x) + imag(x) im","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Note that x is not a VariableRef; instead, it is an affine expression with Complex{Float64}-valued coefficients:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(x)\nGenericAffExpr{ComplexF64, VariableRef}","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Behind the scenes, JuMP has created two real-valued variables, with names \"real(x)\" and \"imag(x)\":","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\njulia> name.(all_variables(model))\n2-element Vector{String}:\n \"real(x)\"\n \"imag(x)\"","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Use the real and imag functions on x to return a real-valued affine expression representing each variable:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(real(x))\nAffExpr (alias for GenericAffExpr{Float64, GenericVariableRef{Float64}})\n\njulia> typeof(imag(x))\nAffExpr (alias for GenericAffExpr{Float64, GenericVariableRef{Float64}})","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"To create an anonymous variable, use the set keyword argument:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> x = @variable(model, set = ComplexPlane())\n_[1] + _[2] im","category":"page"},{"location":"manual/complex/#Complex-valued-variable-bounds","page":"Complex number support","title":"Complex-valued variable bounds","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Because complex-valued variables lack a total ordering, the definition of a variable bound for a complex-valued variable is ambiguous. If you pass a real- or complex-valued argument to keywords such as lower_bound, upper_bound, and start_value, JuMP will apply the real and imaginary parts to the associated real-valued variables.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(\n           model,\n           x in ComplexPlane(),\n           lower_bound = 1.0,\n           upper_bound = 2.0 + 3.0im,\n           start = 4im,\n       )\nreal(x) + imag(x) im\n\njulia> vars = all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\njulia> lower_bound.(vars)\n2-element Vector{Float64}:\n 1.0\n 0.0\n\njulia> upper_bound.(vars)\n2-element Vector{Float64}:\n 2.0\n 3.0\n\njulia> start_value.(vars)\n2-element Vector{Float64}:\n 0.0\n 4.0","category":"page"},{"location":"manual/complex/#Complex-valued-equality-constraints","page":"Complex number support","title":"Complex-valued equality constraints","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"JuMP reformulates complex-valued equality constraints into two real-valued constraints: one representing the real part, and one representing the imaginary part. Thus, complex-valued equality constraints can be solved any solver that supports the real-valued constraint type.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"For example:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, (1 + 2im) * x[1] + 3 * x[2] == 4 + 5im)\n(1 + 2im) x[1] + 3 x[2] = (4 + 5im)\n\njulia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n 2.5\n 0.5","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"is equivalent to","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, 1 * x[1] + 3 * x[2] == 4)  # real component\nx[1] + 3 x[2] = 4\n\njulia> @constraint(model, 2 * x[1] == 5)  # imag component\n2 x[1] = 5\n\njulia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n 2.5\n 0.5","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"This also applies if the variables are complex-valued:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x in ComplexPlane());\n\njulia> @constraint(model, (1 + 2im) * x + 3 * x == 4 + 5im)\n(4 + 2im) real(x) + (-2 + 4im) imag(x) = (4 + 5im)\n\njulia> optimize!(model)\n\njulia> value(x)\n1.3 + 0.6000000000000001im","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"which is equivalent to","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x_real);\n\njulia> @variable(model, x_imag);\n\njulia> @constraint(model, x_real - 2 * x_imag + 3 * x_real == 4)\n4 x_real - 2 x_imag = 4\n\njulia> @constraint(model, x_imag + 2 * x_real + 3 * x_imag == 5)\n2 x_real + 4 x_imag = 5\n\njulia> optimize!(model)\n\njulia> value(x_real) + value(x_imag) * im\n1.3 + 0.6000000000000001im","category":"page"},{"location":"manual/complex/#Hermitian-PSD-Cones","page":"Complex number support","title":"Hermitian PSD Cones","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"JuMP supports creating matrices where are Hermitian.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, H[1:3, 1:3] in HermitianPSDCone())\n3×3 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    …  real(H[1,3]) + imag(H[1,3]) im\n real(H[1,2]) - imag(H[1,2]) im     real(H[2,3]) + imag(H[2,3]) im\n real(H[1,3]) - imag(H[1,3]) im     real(H[3,3])","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Behind the scenes, JuMP has created nine real-valued decision variables:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> all_variables(model)\n9-element Vector{VariableRef}:\n real(H[1,1])\n real(H[1,2])\n real(H[2,2])\n real(H[1,3])\n real(H[2,3])\n real(H[3,3])\n imag(H[1,2])\n imag(H[1,3])\n imag(H[2,3])","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"and a Vector{VariableRef}-in-MOI.HermitianPositiveSemidefiniteConeTriangle constraint:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> num_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle)\n1","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"The MOI.HermitianPositiveSemidefiniteConeTriangle set can be efficiently bridged to MOI.PositiveSemidefiniteConeTriangle, so it can be solved by any solver that supports PSD constraints.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Each element of H is an affine expression with Complex{Float64}-valued coefficients:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(H[1, 1])\nGenericAffExpr{ComplexF64, VariableRef}\n\njulia> typeof(H[2, 1])\nGenericAffExpr{ComplexF64, VariableRef}","category":"page"},{"location":"manual/complex/#Hermitian-PSD-constraints","page":"Complex number support","title":"Hermitian PSD constraints","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"The HermitianPSDCone can also be used in the @constraint macro:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> import LinearAlgebra\n\njulia> H = LinearAlgebra.Hermitian([x[1] 1im; -1im -x[2]])\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n x[1]  im\n -im   -x[2]\n\njulia> @constraint(model, H in HermitianPSDCone())\n[x[1]  im;\n -im   -x[2]] ∈ HermitianPSDCone()","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"note: Note\nThe matrix H in H in HermitianPSDCone() must be a LinearAlgebra.Hermitian matrix type. A build_constraint error will be thrown if the matrix is a different matrix type.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"EditURL = \"user_defined_hessians.jl\"","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/#User-defined-Hessians","page":"User-defined Hessians","title":"User-defined Hessians","text":"","category":"section"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"In this tutorial, we explain how to write a user-defined operator (see User-defined operators) with a Hessian matrix explicitly provided by the user.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"For a more advanced example, see Nested optimization problems.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"using JuMP\nimport Ipopt","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/#Rosenbrock-example","page":"User-defined Hessians","title":"Rosenbrock example","text":"","category":"section"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"As a simple example, we consider the Rosenbrock function:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"rosenbrock(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"which has the gradient vector:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"function ∇rosenbrock(g::AbstractVector, x...)\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"and the Hessian matrix:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"function ∇²rosenbrock(H::AbstractMatrix, x...)\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1] <-- not needed because Hessian is symmetric\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"You may assume the Hessian matrix H is initialized with zeros, and because it is symmetric you need only to fill in the non-zero of the lower-triangular terms.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). However, you may assume only that H supports size(H) and setindex!.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"Now that we have the function, its gradient, and its Hessian, we can construct a JuMP model, add the operator, and use it in a macro:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2])\n@operator(model, op_rosenbrock, 2, rosenbrock, ∇rosenbrock, ∇²rosenbrock)\n@objective(model, Min, op_rosenbrock(x[1], x[2]))\noptimize!(model)\nsolution_summary(model; verbose = true)","category":"page"}]
+[{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/nonlinear.md\"","category":"page"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/nonlinear/#Nonlinear-programming","page":"Nonlinear programming","title":"Nonlinear programming","text":"","category":"section"},{"location":"moi/reference/nonlinear/#Types","page":"Nonlinear programming","title":"Types","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"AbstractNLPEvaluator\nNLPBoundsPair\nNLPBlockData","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.AbstractNLPEvaluator","page":"Nonlinear programming","title":"MathOptInterface.AbstractNLPEvaluator","text":"AbstractNLPEvaluator\n\nAbstract supertype for the callback object that is used to query function values, derivatives, and expression graphs.\n\nIt is used in NLPBlockData.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBoundsPair","page":"Nonlinear programming","title":"MathOptInterface.NLPBoundsPair","text":"NLPBoundsPair(lower::Float64, upper::Float64)\n\nA struct holding a pair of lower and upper bounds.\n\n-Inf and Inf can be used to indicate no lower or upper bound, respectively.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockData","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockData","text":"struct NLPBlockData\n    constraint_bounds::Vector{NLPBoundsPair}\n    evaluator::AbstractNLPEvaluator\n    has_objective::Bool\nend\n\nA struct encoding a set of nonlinear constraints of the form lb le g(x) le ub and, if has_objective == true, a nonlinear objective function f(x).\n\nNonlinear objectives override any objective set by using the ObjectiveFunction attribute.\n\nThe evaluator is a callback object that is used to query function values, derivatives, and expression graphs. If has_objective == false, then it is an error to query properties of the objective function, and in Hessian-of-the-Lagrangian queries, σ must be set to zero.\n\nnote: Note\nThroughout the evaluator, all variables are ordered according to ListOfVariableIndices. Hence, MOI copies of nonlinear problems must not re-order variables.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#Attributes","page":"Nonlinear programming","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"NLPBlock\nNLPBlockDual\nNLPBlockDualStart","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlock","page":"Nonlinear programming","title":"MathOptInterface.NLPBlock","text":"NLPBlock()\n\nAn AbstractModelAttribute that stores an NLPBlockData, representing a set of nonlinear constraints, and optionally a nonlinear objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockDual","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockDual","text":"NLPBlockDual(result_index::Int = 1)\n\nAn AbstractModelAttribute for the Lagrange multipliers on the constraints from the NLPBlock in result result_index.\n\nIf result_index is omitted, it is 1 by default.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#MathOptInterface.NLPBlockDualStart","page":"Nonlinear programming","title":"MathOptInterface.NLPBlockDualStart","text":"NLPBlockDualStart()\n\nAn AbstractModelAttribute for the initial assignment of the Lagrange multipliers on the constraints from the NLPBlock that the solver may use to warm-start the solve.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/nonlinear/#Functions","page":"Nonlinear programming","title":"Functions","text":"","category":"section"},{"location":"moi/reference/nonlinear/","page":"Nonlinear programming","title":"Nonlinear programming","text":"initialize\nfeatures_available\neval_objective\neval_constraint\neval_objective_gradient\njacobian_structure\neval_constraint_gradient\nconstraint_gradient_structure\neval_constraint_jacobian\neval_constraint_jacobian_product\neval_constraint_jacobian_transpose_product\nhessian_lagrangian_structure\nhessian_objective_structure\nhessian_constraint_structure\neval_hessian_objective\neval_hessian_constraint\neval_hessian_lagrangian\neval_hessian_lagrangian_product\nobjective_expr\nconstraint_expr","category":"page"},{"location":"moi/reference/nonlinear/#MathOptInterface.initialize","page":"Nonlinear programming","title":"MathOptInterface.initialize","text":"initialize(\n    d::AbstractNLPEvaluator,\n    requested_features::Vector{Symbol},\n)::Nothing\n\nInitialize d with the set of features in requested_features. Check features_available before calling initialize to see what features are supported by d.\n\nwarning: Warning\nThis method must be called before any other methods.\n\nFeatures\n\nThe following features are defined:\n\n:Grad: enables eval_objective_gradient\n:Jac: enables eval_constraint_jacobian\n:JacVec: enables eval_constraint_jacobian_product and eval_constraint_jacobian_transpose_product\n:Hess: enables eval_hessian_lagrangian\n:HessVec: enables eval_hessian_lagrangian_product\n:ExprGraph: enables objective_expr and constraint_expr.\n\nIn all cases, including when requested_features is empty, eval_objective and eval_constraint are supported.\n\nExamples\n\nMOI.initialize(d, Symbol[])\nMOI.initialize(d, [:ExprGraph])\nMOI.initialize(d, MOI.features_available(d))\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.features_available","page":"Nonlinear programming","title":"MathOptInterface.features_available","text":"features_available(d::AbstractNLPEvaluator)::Vector{Symbol}\n\nReturns the subset of features available for this problem instance.\n\nSee initialize for the list of defined features.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_objective","page":"Nonlinear programming","title":"MathOptInterface.eval_objective","text":"eval_objective(d::AbstractNLPEvaluator, x::AbstractVector{T})::T where {T}\n\nEvaluate the objective f(x), returning a scalar value.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint","text":"eval_constraint(d::AbstractNLPEvaluator,\n    g::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nGiven a set of vector-valued constraints l le g(x) le u, evaluate the constraint function g(x), storing the result in the vector g.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that g is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_objective_gradient","page":"Nonlinear programming","title":"MathOptInterface.eval_objective_gradient","text":"eval_objective_gradient(\n    d::AbstractNLPEvaluator,\n    grad::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nEvaluate the gradient of the objective function grad = nabla f(x) as a dense vector, storing the result in the vector grad.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that grad is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.jacobian_structure","page":"Nonlinear programming","title":"MathOptInterface.jacobian_structure","text":"jacobian_structure(d::AbstractNLPEvaluator)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Jacobian matrix: J_g(x) = left beginarrayc nabla g_1(x)  nabla g_2(x)  vdots  nabla g_m(x) endarrayright where g_i is the itextth component of the nonlinear constraints g(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_gradient","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_gradient","text":"eval_constraint_gradient(\n    d::AbstractNLPEvaluator,\n    ∇g::AbstractVector{T},\n    x::AbstractVector{T},\n    i::Int,\n)::Nothing where {T}\n\nEvaluate the gradient of constraint i, nabla g_i(x), and store the non-zero values in ∇g, corresponding to the structure returned by constraint_gradient_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that ∇g is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.constraint_gradient_structure","page":"Nonlinear programming","title":"MathOptInterface.constraint_gradient_structure","text":"constraint_gradient_structure(d::AbstractNLPEvaluator, i::Int)::Vector{Int64}\n\nReturns a vector of indices, where each element indicates the position of a structurally nonzero element in the gradient of constraint nabla g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian","text":"eval_constraint_jacobian(d::AbstractNLPEvaluator,\n    J::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nEvaluates the sparse Jacobian matrix J_g(x) = left beginarrayc nabla g_1(x)  nabla g_2(x)  vdots  nabla g_m(x) endarrayright.\n\nThe result is stored in the vector J in the same order as the indices returned by jacobian_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that J is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian_product","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian_product","text":"eval_constraint_jacobian_product(\n    d::AbstractNLPEvaluator,\n    y::AbstractVector{T},\n    x::AbstractVector{T},\n    w::AbstractVector{T},\n)::Nothing where {T}\n\nComputes the Jacobian-vector product y = J_g(x)w, storing the result in the vector y.\n\nThe vectors have dimensions such that length(w) == length(x), and length(y) is the number of nonlinear constraints.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that y is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_constraint_jacobian_transpose_product","page":"Nonlinear programming","title":"MathOptInterface.eval_constraint_jacobian_transpose_product","text":"eval_constraint_jacobian_transpose_product(\n    d::AbstractNLPEvaluator,\n    y::AbstractVector{T},\n    x::AbstractVector{T},\n    w::AbstractVector{T},\n)::Nothing where {T}\n\nComputes the Jacobian-transpose-vector product y = J_g(x)^Tw, storing the result in the vector y.\n\nThe vectors have dimensions such that length(y) == length(x), and length(w) is the number of nonlinear constraints.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that y is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_lagrangian_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_lagrangian_structure","text":"hessian_lagrangian_structure(\n    d::AbstractNLPEvaluator,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian-of-the-Lagrangian matrix: nabla^2 f(x) + sum_i=1^m nabla^2 g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_objective_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_objective_structure","text":"hessian_objective_structure(\n    d::AbstractNLPEvaluator,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian matrix: nabla^2 f(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.hessian_constraint_structure","page":"Nonlinear programming","title":"MathOptInterface.hessian_constraint_structure","text":"hessian_constraint_structure(\n    d::AbstractNLPEvaluator,\n    i::Int64,\n)::Vector{Tuple{Int64,Int64}}\n\nReturns a vector of tuples, (row, column), where each indicates the position of a structurally nonzero element in the Hessian matrix: nabla^2 g_i(x).\n\nThe indices are not required to be sorted and can contain duplicates, in which case the solver should combine the corresponding elements by adding them together.\n\nAny mix of lower and upper-triangular indices is valid. Elements (i,j) and (j,i), if both present, should be treated as duplicates.\n\nThe sparsity structure is assumed to be independent of the point x.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_objective","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_objective","text":"eval_hessian_objective(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n)::Nothing where {T}\n\nThis function computes the sparse Hessian matrix: nabla^2 f(x), storing the result in the vector H in the same order as the indices returned by hessian_objective_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_constraint","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_constraint","text":"eval_hessian_constraint(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n    i::Int64,\n)::Nothing where {T}\n\nThis function computes the sparse Hessian matrix: nabla^2 g_i(x), storing the result in the vector H in the same order as the indices returned by hessian_constraint_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_lagrangian","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_lagrangian","text":"eval_hessian_lagrangian(\n    d::AbstractNLPEvaluator,\n    H::AbstractVector{T},\n    x::AbstractVector{T},\n    σ::T,\n    μ::AbstractVector{T},\n)::Nothing where {T}\n\nGiven scalar weight σ and vector of constraint weights μ, this function computes the sparse Hessian-of-the-Lagrangian matrix: sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x), storing the result in the vector H in the same order as the indices returned by hessian_lagrangian_structure.\n\nImplementation notes\n\nWhen implementing this method, you must not assume that H is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.eval_hessian_lagrangian_product","page":"Nonlinear programming","title":"MathOptInterface.eval_hessian_lagrangian_product","text":"eval_hessian_lagrangian_product(\n    d::AbstractNLPEvaluator,\n    h::AbstractVector{T},\n    x::AbstractVector{T},\n    v::AbstractVector{T},\n    σ::T,\n    μ::AbstractVector{T},\n)::Nothing where {T}\n\nGiven scalar weight σ and vector of constraint weights μ, computes the Hessian-of-the-Lagrangian-vector product h = left(sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x)right)v, storing the result in the vector h.\n\nThe vectors have dimensions such that length(h) == length(x) == length(v).\n\nImplementation notes\n\nWhen implementing this method, you must not assume that h is Vector{Float64}, but you may assume that it supports setindex! and length. For example, it may be the view of a vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.objective_expr","page":"Nonlinear programming","title":"MathOptInterface.objective_expr","text":"objective_expr(d::AbstractNLPEvaluator)::Expr\n\nReturns a Julia Expr object representing the expression graph of the objective function.\n\nFormat\n\nThe expression has a number of limitations, compared with arbitrary Julia expressions:\n\nAll sums and products are flattened out as simple Expr(:+, ...) and Expr(:*, ...) objects.\nAll decision variables must be of the form Expr(:ref, :x, MOI.VariableIndex(i)), where i is the ith variable in ListOfVariableIndices.\nThere are currently no restrictions on recognized functions; typically these will be built-in Julia functions like ^, exp, log, cos, tan, sqrt, etc., but modeling interfaces may choose to extend these basic functions, or error if they encounter unsupported functions.\n\nExamples\n\nThe expression x_1+sin(x_2exp(x_3)) is represented as\n\n:(x[MOI.VariableIndex(1)] + sin(x[MOI.VariableIndex(2)] / exp(x[MOI.VariableIndex[3]])))\n\nor equivalently\n\nExpr(\n    :call,\n    :+,\n    Expr(:ref, :x, MOI.VariableIndex(1)),\n    Expr(\n        :call,\n        :/,\n        Expr(:call, :sin, Expr(:ref, :x, MOI.VariableIndex(2))),\n        Expr(:call, :exp, Expr(:ref, :x, MOI.VariableIndex(3))),\n    ),\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/nonlinear/#MathOptInterface.constraint_expr","page":"Nonlinear programming","title":"MathOptInterface.constraint_expr","text":"constraint_expr(d::AbstractNLPEvaluator, i::Integer)::Expr\n\nReturns a Julia Expr object representing the expression graph for the itextth nonlinear constraint.\n\nFormat\n\nThe format is the same as objective_expr, with an additional comparison operator indicating the sense of and bounds on the constraint.\n\nFor single-sided comparisons, the body of the constraint must be on the left-hand side, and the right-hand side must be a constant.\n\nFor double-sided comparisons (that is, l le f(x) le u), the body of the constraint must be in the middle, and the left- and right-hand sides must be constants.\n\nThe bounds on the constraints must match the NLPBoundsPairs passed to NLPBlockData.\n\nExamples\n\n:(x[MOI.VariableIndex(1)]^2 <= 1.0)\n:(x[MOI.VariableIndex(1)]^2 >= 2.0)\n:(x[MOI.VariableIndex(1)]^2 == 3.0)\n:(4.0 <= x[MOI.VariableIndex(1)]^2 <= 5.0)\n\n\n\n\n\n","category":"function"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"EditURL = \"https://github.com/jump-dev/Cbc.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"(Image: )","category":"page"},{"location":"packages/Cbc/#Cbc.jl","page":"jump-dev/Cbc.jl","title":"Cbc.jl","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.jl is a wrapper for the COIN-OR Branch and Cut (Cbc) solver.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Cbc/#Affiliation","page":"jump-dev/Cbc.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Cbc/#License","page":"jump-dev/Cbc.jl","title":"License","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The underlying solver, coin-or/Cbc, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Cbc/#Installation","page":"jump-dev/Cbc.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Install Cbc using Pkg.add:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"import Pkg\nPkg.add(\"Cbc\")","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"In addition to installing the Cbc.jl package, this will also download and install the Cbc binaries. You do not need to install Cbc separately.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Cbc/#Use-with-JuMP","page":"jump-dev/Cbc.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"To use Cbc with JuMP, use Cbc.Optimizer:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"using JuMP, Cbc\nmodel = Model(Cbc.Optimizer)\nset_attribute(model, \"logLevel\", 1)","category":"page"},{"location":"packages/Cbc/#MathOptInterface-API","page":"jump-dev/Cbc.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The COIN Branch-and-Cut (Cbc) optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.Status\nCbc.SecondaryStatus\nMOI.DualStatus\nMOI.NodeCount\nMOI.NumberOfVariables\nMOI.ObjectiveBound\nMOI.ObjectiveSense\nMOI.ObjectiveValue\nMOI.PrimalStatus\nMOI.RelativeGap\nMOI.ResultCount\nMOI.SolveTimeSec\nMOI.TerminationStatus","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported optimizer attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Cbc.SetVariableNames\nMOI.AbsoluteGapTolerance\nMOI.NumberOfThreads\nMOI.RawOptimizerAttribute\nMOI.RelativeGapTolerance\nMOI.Silent\nMOI.SolverName\nMOI.SolverVersion\nMOI.TimeLimitSec","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported variable attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.VariablePrimal\nMOI.VariablePrimalStart\nMOI.VariableName","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"List of supported constraint attributes:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"MOI.ConstraintPrimal","category":"page"},{"location":"packages/Cbc/#Options","page":"jump-dev/Cbc.jl","title":"Options","text":"","category":"section"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Options are, unfortunately, not well documented.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The following options are likely to be the most useful:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Parameter Example Explanation\nseconds 60.0 Solution timeout limit\nlogLevel 2 Set to 0 to disable solution output\nmaxSolutions 1 Terminate after this many feasible solutions have been found\nmaxNodes 1 Terminate after this many branch-and-bound nodes have been evaluated\nallowableGap 0.05 Terminate after optimality gap is less than this value (on an absolute scale)\nratioGap 0.05 Terminate after optimality gap is smaller than this relative fraction\nthreads 1 Set the number of threads to use for parallel branch & bound","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"The complete list of parameters can be found by running the cbc executable and typing ? at the prompt.","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"Start the cbc executable from Julia as follows:","category":"page"},{"location":"packages/Cbc/","page":"jump-dev/Cbc.jl","title":"jump-dev/Cbc.jl","text":"using Cbc_jll\nCbc_jll.cbc() do exe\n    run(`$(exe)`)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"EditURL = \"geographic_clustering.jl\"","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Geographical-clustering","page":"Geographical clustering","title":"Geographical clustering","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial was originally contributed by Matthew Helm and Mathieu Tanneau.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"The goal of this exercise is to cluster n cities into k groups, minimizing the total pairwise distance between cities and ensuring that the variance in the total populations of each group is relatively small.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"For this example, we'll use the 20 most populous cities in the United States.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"cities = DataFrames.DataFrame(\n    Union{String,Float64}[\n        \"New York, NY\" 8.405 40.7127 -74.0059\n        \"Los Angeles, CA\" 3.884 34.0522 -118.2436\n        \"Chicago, IL\" 2.718 41.8781 -87.6297\n        \"Houston, TX\" 2.195 29.7604 -95.3698\n        \"Philadelphia, PA\" 1.553 39.9525 -75.1652\n        \"Phoenix, AZ\" 1.513 33.4483 -112.0740\n        \"San Antonio, TX\" 1.409 29.4241 -98.4936\n        \"San Diego, CA\" 1.355 32.7157 -117.1610\n        \"Dallas, TX\" 1.257 32.7766 -96.7969\n        \"San Jose, CA\" 0.998 37.3382 -121.8863\n        \"Austin, TX\" 0.885 30.2671 -97.7430\n        \"Indianapolis, IN\" 0.843 39.7684 -86.1580\n        \"Jacksonville, FL\" 0.842 30.3321 -81.6556\n        \"San Francisco, CA\" 0.837 37.7749 -122.4194\n        \"Columbus, OH\" 0.822 39.9611 -82.9987\n        \"Charlotte, NC\" 0.792 35.2270 -80.8431\n        \"Fort Worth, TX\" 0.792 32.7554 -97.3307\n        \"Detroit, MI\" 0.688 42.3314 -83.0457\n        \"El Paso, TX\" 0.674 31.7775 -106.4424\n        \"Memphis, TN\" 0.653 35.1495 -90.0489\n    ],\n    [\"city\", \"population\", \"lat\", \"lon\"],\n)","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Model-Specifics","page":"Geographical clustering","title":"Model Specifics","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We will cluster these 20 cities into 3 different groups and we will assume that the ideal or target population P for a group is simply the total population of the 20 cities divided by 3:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"n = size(cities, 1)\nk = 3\nP = sum(cities.population) / k","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Obtaining-the-distances-between-each-city","page":"Geographical clustering","title":"Obtaining the distances between each city","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Let's compute the pairwise Haversine distance between each of the cities in our data set and store the result in a variable we'll call dm:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"\"\"\"\n    haversine(lat1, long1, lat2, long2, r = 6372.8)\n\nCompute the haversine distance between two points on a sphere of radius `r`,\nwhere the points are given by the latitude/longitude pairs `lat1/long1` and\n`lat2/long2` (in degrees).\n\"\"\"\nfunction haversine(lat1, long1, lat2, long2, r = 6372.8)\n    lat1, long1 = deg2rad(lat1), deg2rad(long1)\n    lat2, long2 = deg2rad(lat2), deg2rad(long2)\n    hav(a, b) = sin((b - a) / 2)^2\n    inner_term = hav(lat1, lat2) + cos(lat1) * cos(lat2) * hav(long1, long2)\n    d = 2 * r * asin(sqrt(inner_term))\n    # Round distance to nearest kilometer.\n    return round(Int, d)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Our distance matrix is symmetric so we'll convert it to a LowerTriangular matrix so that we can better interpret the objective value of our model:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"dm = LinearAlgebra.LowerTriangular([\n    haversine(cities.lat[i], cities.lon[i], cities.lat[j], cities.lon[j])\n    for i in 1:n, j in 1:n\n])","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Build-the-model","page":"Geographical clustering","title":"Build the model","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now that we have the basics taken  care of, we can set up our model, create decision variables, add constraints, and then solve.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"First, we'll set up a model that leverages the Cbc solver. Next, we'll set up a binary variable x_ik that takes the value 1 if city i is in group k and 0 otherwise. Each city must be in a group, so we'll add the constraint sum_k x_ik = 1 for every i.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:n, 1:k], Bin)\n@constraint(model, [i = 1:n], sum(x[i, :]) == 1);\n# To reduce symmetry, we fix the first city to belong to the first group.\nfix(x[1, 1], 1; force = true)","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"The total population of a group k is Q_k = sum_ix_ikq_i where q_i is simply the i-th value from the population column in our cities DataFrame. Let's add constraints so that alpha leq (Q_k - P) leq beta. We'll set alpha equal to -3 million and beta equal to 3. By adjusting these thresholds you'll find that there is a tradeoff between having relatively even populations between groups and having geographically close cities within each group. In other words, the larger the absolute values of alpha and beta, the closer together the cities in a group will be but the variance between the group populations will be higher.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@variable(model, -3 <= population_diff[1:k] <= 3)\n@constraint(model, population_diff .== x' * cities.population .- P)","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now we need to add one last binary variable z_ij to our model that we'll use to compute the total distance between the cities in our groups, defined as sum_ijd_ijz_ij. Variable z_ij will equal 1 if cities i and j are in the same group, and 0 if they are not in the same group.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"To ensure that z_ij = 1 if and only if cities i and j are in the same group, we add the constraints z_ij geq x_ik + x_jk - 1 for every pair ij and every k:","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@variable(model, z[i = 1:n, j = 1:i], Bin)\nfor k in 1:k, i in 1:n, j in 1:i\n    @constraint(model, z[i, j] >= x[i, k] + x[j, k] - 1)\nend","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We can now add an objective to our model which will simply be to minimize the dot product of z and our distance matrix, dm.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"@objective(model, Min, sum(dm[i, j] * z[i, j] for i in 1:n, j in 1:i));\nnothing #hide","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"We can then call optimize! and review the results.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"optimize!(model)","category":"page"},{"location":"tutorials/linear/geographic_clustering/#Reviewing-the-Results","page":"Geographical clustering","title":"Reviewing the Results","text":"","category":"section"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"Now that we have results, we can add a column to our cities DataFrame for the group and then loop through our x variable to assign each city to its group. Once we have that, we can look at the total population for each group and also look at the cities in each group to verify that they are grouped by geographic proximity.","category":"page"},{"location":"tutorials/linear/geographic_clustering/","page":"Geographical clustering","title":"Geographical clustering","text":"cities.group = zeros(n)\n\nfor i in 1:n, j in 1:k\n    if round(Int, value(x[i, j])) == 1\n        cities.group[i] = j\n    end\nend\n\nfor group in DataFrames.groupby(cities, :group)\n    @show group\n    println(\"\")\n    @show sum(group.population)\n    println(\"\")\nend","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/objective/#Objectives","page":"Objectives","title":"Objectives","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"This page describes macros and functions related to linear and quadratic objective functions only, unless otherwise indicated. For nonlinear objective functions, see Nonlinear Modeling.","category":"page"},{"location":"manual/objective/#Set-a-linear-objective","page":"Objectives","title":"Set a linear objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a linear objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use Min to create a minimization objective:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use Max to create a maximization objective:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Max, 2x + 1)\n2 x + 1","category":"page"},{"location":"manual/objective/#Set-a-quadratic-objective","page":"Objectives","title":"Set a quadratic objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a quadratic objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use ^2 to have a variable squared:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, x^2 + 2x + 1)\nx² + 2 x + 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"You can also have bilinear terms between variables:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @objective(model, Max, x * y + x + y)\nx*y + x + y","category":"page"},{"location":"manual/objective/#Set-a-nonlinear-objective","page":"Objectives","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use the @objective macro to set a nonlinear objective function:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x <= 1);\n\njulia> @objective(model, Max, log(x))\nlog(x)","category":"page"},{"location":"manual/objective/#Query-the-objective-function","page":"Objectives","title":"Query the objective function","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use objective_function to return the current objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_function(model)\n2 x + 1","category":"page"},{"location":"manual/objective/#Evaluate-the-objective-function-at-a-point","page":"Objectives","title":"Evaluate the objective function at a point","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use value to evaluate an objective function at a point specifying values for variables.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, 2x[1]^2 + x[1] + 0.5*x[2])\n2 x[1]² + x[1] + 0.5 x[2]\n\njulia> f = objective_function(model)\n2 x[1]² + x[1] + 0.5 x[2]\n\njulia> point = Dict(x[1] => 2.0, x[2] => 1.0);\n\njulia> value(z -> point[z], f)\n10.5","category":"page"},{"location":"manual/objective/#Query-the-objective-sense","page":"Objectives","title":"Query the objective sense","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use objective_sense to return the current objective sense.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_sense(model)\nMIN_SENSE::OptimizationSense = 0","category":"page"},{"location":"manual/objective/#Modify-an-objective","page":"Objectives","title":"Modify an objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"To modify an objective, call @objective with the new objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> @objective(model, Max, -2x)\n-2 x","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Alternatively, use set_objective_function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> new_objective = @expression(model, -2 * x)\n-2 x\n\njulia> set_objective_function(model, new_objective)","category":"page"},{"location":"manual/objective/#Modify-an-objective-coefficient","page":"Objectives","title":"Modify an objective coefficient","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use set_objective_coefficient to modify an objective coefficient.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> set_objective_coefficient(model, x, 3)\n\njulia> objective_function(model)\n3 x","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"info: Info\nThere is no way to modify the coefficient of a quadratic term. Set a new objective instead.","category":"page"},{"location":"manual/objective/#Modify-the-objective-sense","page":"Objectives","title":"Modify the objective sense","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Use set_objective_sense to modify the objective sense.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> objective_sense(model)\nMIN_SENSE::OptimizationSense = 0\n\njulia> set_objective_sense(model, MAX_SENSE);\n\njulia> objective_sense(model)\nMAX_SENSE::OptimizationSense = 1","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Alternatively, call @objective and pass the existing objective function.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @objective(model, Min, 2x)\n2 x\n\njulia> @objective(model, Max, objective_function(model))\n2 x","category":"page"},{"location":"manual/objective/#Set-a-vector-valued-objective","page":"Objectives","title":"Set a vector-valued objective","text":"","category":"section"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Define a multi-objective optimization problem by passing a vector of objectives:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, [1 + x[1], 2 * x[2]])\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[2]\n\njulia> f = objective_function(model)\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[2]","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"tip: Tip\nThe Multi-objective knapsack tutorial provides an example of solving a multi-objective integer program.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"In most cases, multi-objective optimization solvers will return multiple solutions, corresponding to points on the Pareto frontier. See Multiple solutions for information on how to query and work with multiple solutions.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Note that you must set a single objective sense, that is, you cannot have both minimization and maximization objectives. Work around this limitation by choosing Min and negating any objectives you want to maximize:","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @expression(model, obj1, 1 + x[1])\nx[1] + 1\n\njulia> @expression(model, obj2, 2 * x[1])\n2 x[1]\n\njulia> @objective(model, Min, [obj1, -obj2])\n2-element Vector{AffExpr}:\n x[1] + 1\n -2 x[1]","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"Defining your objectives as expressions allows flexibility in how you can solve variations of the same problem, with some objectives removed and constrained to be no worse that a fixed value.","category":"page"},{"location":"manual/objective/","page":"Objectives","title":"Objectives","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @expression(model, obj1, 1 + x[1])\nx[1] + 1\n\njulia> @expression(model, obj2, 2 * x[1])\n2 x[1]\n\njulia> @expression(model, obj3, x[1] + x[2])\nx[1] + x[2]\n\njulia> @objective(model, Min, [obj1, obj2, obj3])  # Three-objective problem\n3-element Vector{AffExpr}:\n x[1] + 1\n 2 x[1]\n x[1] + x[2]\n\njulia> # optimize!(model), look at the solution, talk to stakeholders, then\n       # decide you want to solve a new problem where the third objective is\n       # removed and constrained to be better than 2.0.\n       nothing\n\njulia> @objective(model, Min, [obj1, obj2])   # Two-objective problem\n2-element Vector{AffExpr}:\n x[1] + 1\n 2 x[1]\n\njulia> @constraint(model, obj3 <= 2.0)\nx[1] + x[2] ≤ 2","category":"page"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/callbacks.md\"","category":"page"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/callbacks/#Callbacks","page":"Callbacks","title":"Callbacks","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"AbstractCallback\nAbstractSubmittable\nsubmit","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.AbstractCallback","page":"Callbacks","title":"MathOptInterface.AbstractCallback","text":"abstract type AbstractCallback <: AbstractModelAttribute end\n\nAbstract type for a model attribute representing a callback function. The value set to subtypes of AbstractCallback is a function that may be called during optimize!. As optimize! is in progress, the result attributes (i.e, the attributes attr such that is_set_by_optimize(attr)) may not be accessible from the callback, hence trying to get result attributes might throw a OptimizeInProgress error.\n\nAt most one callback of each type can be registered. If an optimizer already has a function for a callback type, and the user registers a new function, then the old one is replaced.\n\nThe value of the attribute should be a function taking only one argument, commonly called callback_data, that can be used for instance in LazyConstraintCallback, HeuristicCallback and UserCutCallback.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.AbstractSubmittable","page":"Callbacks","title":"MathOptInterface.AbstractSubmittable","text":"AbstractSubmittable\n\nAbstract supertype for objects that can be submitted to the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.submit","page":"Callbacks","title":"MathOptInterface.submit","text":"submit(\n    optimizer::AbstractOptimizer,\n    sub::AbstractSubmittable,\n    values...,\n)::Nothing\n\nSubmit values to the submittable sub of the optimizer optimizer.\n\nAn UnsupportedSubmittable error is thrown if model does not support the attribute attr (see supports) and a SubmitNotAllowed error is thrown if it supports the submittable sub but it cannot be submitted.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/callbacks/#Attributes","page":"Callbacks","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"CallbackNodeStatus\nCallbackVariablePrimal\nCallbackNodeStatusCode\nCALLBACK_NODE_STATUS_INTEGER\nCALLBACK_NODE_STATUS_FRACTIONAL\nCALLBACK_NODE_STATUS_UNKNOWN","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackNodeStatus","page":"Callbacks","title":"MathOptInterface.CallbackNodeStatus","text":"CallbackNodeStatus(callback_data)\n\nAn optimizer attribute describing the (in)feasibility of the primal solution available from CallbackVariablePrimal during a callback identified by callback_data.\n\nReturns a CallbackNodeStatusCode Enum.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackVariablePrimal","page":"Callbacks","title":"MathOptInterface.CallbackVariablePrimal","text":"CallbackVariablePrimal(callback_data)\n\nA variable attribute for the assignment to some primal variable's value during the callback identified by callback_data.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CallbackNodeStatusCode","page":"Callbacks","title":"MathOptInterface.CallbackNodeStatusCode","text":"CallbackNodeStatusCode\n\nAn Enum of possible return values from calling get with CallbackNodeStatus.\n\nValues\n\nPossible values are:\n\nCALLBACK_NODE_STATUS_INTEGER: the primal solution available from CallbackVariablePrimal is integer feasible.\nCALLBACK_NODE_STATUS_FRACTIONAL: the primal solution available from CallbackVariablePrimal is integer infeasible.\nCALLBACK_NODE_STATUS_UNKNOWN: the primal solution available from CallbackVariablePrimal might be integer feasible or infeasible.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_INTEGER","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_INTEGER","text":"CALLBACK_NODE_STATUS_INTEGER::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_INTEGER: the primal solution available from CallbackVariablePrimal is integer feasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_FRACTIONAL","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_FRACTIONAL","text":"CALLBACK_NODE_STATUS_FRACTIONAL::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_FRACTIONAL: the primal solution available from CallbackVariablePrimal is integer infeasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.CALLBACK_NODE_STATUS_UNKNOWN","page":"Callbacks","title":"MathOptInterface.CALLBACK_NODE_STATUS_UNKNOWN","text":"CALLBACK_NODE_STATUS_UNKNOWN::CallbackNodeStatusCode\n\nAn instance of the CallbackNodeStatusCode enum.\n\nCALLBACK_NODE_STATUS_UNKNOWN: the primal solution available from CallbackVariablePrimal might be integer feasible or infeasible.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#Lazy-constraints","page":"Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"LazyConstraintCallback\nLazyConstraint","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.LazyConstraintCallback","page":"Callbacks","title":"MathOptInterface.LazyConstraintCallback","text":"LazyConstraintCallback() <: AbstractCallback\n\nThe callback can be used to reduce the feasible set given the current primal solution by submitting a LazyConstraint. For instance, it may be called at an incumbent of a mixed-integer problem. Note that there is no guarantee that the callback is called at every feasible primal solution.\n\nThe current primal solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.LazyConstraintCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # should add a lazy constraint\n        func = # computes function\n        set = # computes set\n        MOI.submit(optimizer, MOI.LazyConstraint(callback_data), func, set)\n    end\nend)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.LazyConstraint","page":"Callbacks","title":"MathOptInterface.LazyConstraint","text":"LazyConstraint(callback_data)\n\nLazy constraint func-in-set submitted as func, set. The optimal solution returned by VariablePrimal will satisfy all lazy constraints that have been submitted.\n\nThis can be submitted only from the LazyConstraintCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the feasible solution callback.\n\nExamples\n\nSuppose x and y are VariableIndexs of optimizer. To add a LazyConstraint for 2x + 3y <= 1, write\n\nfunc = 2.0x + 3.0y\nset = MOI.LessThan(1.0)\nMOI.submit(optimizer, MOI.LazyConstraint(callback_data), func, set)\n\ninside a LazyConstraintCallback of data callback_data.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#User-cuts","page":"Callbacks","title":"User cuts","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"UserCutCallback\nUserCut","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.UserCutCallback","page":"Callbacks","title":"MathOptInterface.UserCutCallback","text":"UserCutCallback() <: AbstractCallback\n\nThe callback can be used to submit UserCut given the current primal solution. For instance, it may be called at fractional (i.e., non-integer) nodes in the branch and bound tree of a mixed-integer problem. Note that there is not guarantee that the callback is called everytime the solver has an infeasible solution.\n\nThe infeasible solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.UserCutCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # can find a user cut\n        func = # computes function\n        set = # computes set\n        MOI.submit(optimizer, MOI.UserCut(callback_data), func, set)\n    end\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.UserCut","page":"Callbacks","title":"MathOptInterface.UserCut","text":"UserCut(callback_data)\n\nConstraint func-to-set suggested to help the solver detect the solution given by CallbackVariablePrimal as infeasible. The cut is submitted as func, set. Typically CallbackVariablePrimal will violate integrality constraints, and a cut would be of the form ScalarAffineFunction-in-LessThan or ScalarAffineFunction-in-GreaterThan. Note that, as opposed to LazyConstraint, the provided constraint cannot modify the feasible set, the constraint should be redundant, e.g., it may be a consequence of affine and integrality constraints.\n\nThis can be submitted only from the UserCutCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the infeasible solution callback.\n\nNote that the solver may silently ignore the provided constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#Heuristic-solutions","page":"Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"moi/reference/callbacks/","page":"Callbacks","title":"Callbacks","text":"HeuristicCallback\nHeuristicSolution\nHeuristicSolutionStatus\nHEURISTIC_SOLUTION_ACCEPTED\nHEURISTIC_SOLUTION_REJECTED\nHEURISTIC_SOLUTION_UNKNOWN","category":"page"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicCallback","page":"Callbacks","title":"MathOptInterface.HeuristicCallback","text":"HeuristicCallback() <: AbstractCallback\n\nThe callback can be used to submit HeuristicSolution given the current primal solution. For example, it may be called at fractional (i.e., non-integer) nodes in the branch and bound tree of a mixed-integer problem. Note that there is no guarantee that the callback is called every time the solver has an infeasible solution.\n\nThe current primal solution is accessed through CallbackVariablePrimal. Trying to access other result attributes will throw OptimizeInProgress as discussed in AbstractCallback.\n\nExamples\n\nx = MOI.add_variables(optimizer, 8)\nMOI.set(optimizer, MOI.HeuristicCallback(), callback_data -> begin\n    sol = MOI.get(optimizer, MOI.CallbackVariablePrimal(callback_data), x)\n    if # can find a heuristic solution\n        values = # computes heuristic solution\n        MOI.submit(optimizer, MOI.HeuristicSolution(callback_data), x,\n                   values)\n    end\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicSolution","page":"Callbacks","title":"MathOptInterface.HeuristicSolution","text":"HeuristicSolution(callback_data)\n\nHeuristically obtained feasible solution. The solution is submitted as variables, values where values[i] gives the value of variables[i], similarly to set. The submit call returns a HeuristicSolutionStatus indicating whether the provided solution was accepted or rejected.\n\nThis can be submitted only from the HeuristicCallback. The field callback_data is a solver-specific callback type that is passed as the argument to the heuristic callback.\n\nSome solvers require a complete solution, others only partial solutions.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HeuristicSolutionStatus","page":"Callbacks","title":"MathOptInterface.HeuristicSolutionStatus","text":"HeuristicSolutionStatus\n\nAn Enum of possible return values for submit with HeuristicSolution. This informs whether the heuristic solution was accepted or rejected.\n\nValues\n\nPossible values are:\n\nHEURISTIC_SOLUTION_ACCEPTED: The heuristic solution was accepted\nHEURISTIC_SOLUTION_REJECTED: The heuristic solution was rejected\nHEURISTIC_SOLUTION_UNKNOWN: No information available on the acceptance\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_ACCEPTED","text":"HEURISTIC_SOLUTION_ACCEPTED::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_ACCEPTED: The heuristic solution was accepted\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_REJECTED","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_REJECTED","text":"HEURISTIC_SOLUTION_REJECTED::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_REJECTED: The heuristic solution was rejected\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/callbacks/#MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN","page":"Callbacks","title":"MathOptInterface.HEURISTIC_SOLUTION_UNKNOWN","text":"HEURISTIC_SOLUTION_UNKNOWN::HeuristicSolutionStatus\n\nAn instance of the HeuristicSolutionStatus enum.\n\nHEURISTIC_SOLUTION_UNKNOWN: No information available on the acceptance\n\n\n\n\n\n","category":"constant"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"EditURL = \"https://github.com/odow/SDDP.jl/blob/v1.6.6/README.md\"","category":"page"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"<img src=\"https://raw.githubusercontent.com/odow/SDDP.jl/e9de84e0a4b57374bd9e0c95148da1501816e4c5/docs/src/assets/logo.png?raw=true\" alt=\"logo\" width=\"100px\"/>","category":"page"},{"location":"packages/SDDP/#SDDP.jl","page":"odow/SDDP.jl","title":"SDDP.jl","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"SDDP.jl is a JuMP extension for solving large convex multistage stochastic programming problems using stochastic dual dynamic programming.","category":"page"},{"location":"packages/SDDP/#License","page":"odow/SDDP.jl","title":"License","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"SDDP.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/SDDP/#Documentation","page":"odow/SDDP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"You can find the documentation at sddp.dev.","category":"page"},{"location":"packages/SDDP/#Help","page":"odow/SDDP.jl","title":"Help","text":"","category":"section"},{"location":"packages/SDDP/","page":"odow/SDDP.jl","title":"odow/SDDP.jl","text":"If you need help, please open a GitHub issue.","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"EditURL = \"https://github.com/plasmo-dev/Plasmo.jl/blob/v0.5.4/README.md\"","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"<img src=\"https://github.com/plasmo-dev/Plasmo.jl/blob/main/docs/plasmo_logo.svg?raw=true\"/>","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"(Image: CI) (Image: codecov) (Image: ) (Image: DOI)","category":"page"},{"location":"packages/Plasmo/#Plasmo.jl","page":"plasmo-dev/Plasmo.jl","title":"Plasmo.jl","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Plasmo.jl (Platform for Scalable Modeling and Optimization) is a graph-based algebraic modeling framework that adopts a modular style to create mathematical optimization problems and manage distributed and hierarchical structures. The package has been developed as a JuMP extension and consequently supports  most JuMP syntax and functions. ","category":"page"},{"location":"packages/Plasmo/#Overview","page":"plasmo-dev/Plasmo.jl","title":"Overview","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The core data structure in Plasmo.jl is the OptiGraph. The optigraph contains a set of optinodes which represent self-contained optimization problems and optiedges that represent coupling between optinodes (which produces an underlying hypergraph structure of optinodes and optiedges). Optigraphs can further be embedded within other optigraphs to create nested hierarchical graph structures. The graph structures obtained using Plasmo.jl can be used for simple model and data management, but they can also be used to perform graph partitioning or develop interfaces to structured optimization solvers.","category":"page"},{"location":"packages/Plasmo/#License","page":"plasmo-dev/Plasmo.jl","title":"License","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Plasmo is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Plasmo/#Installation","page":"plasmo-dev/Plasmo.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Install Plasmo using Pkg.add:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"import Pkg\nPkg.add(\"Plasmo\")","category":"page"},{"location":"packages/Plasmo/#Documentation","page":"plasmo-dev/Plasmo.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The latest documentation is available through GitHub Pages. Additional examples can be found in the examples folder.","category":"page"},{"location":"packages/Plasmo/#Simple-Example","page":"plasmo-dev/Plasmo.jl","title":"Simple Example","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"using Plasmo\nusing Ipopt\n\n#create an optigraph\ngraph = OptiGraph()\n\n#add nodes to an optigraph\n@optinode(graph, n1)\n@optinode(graph, n2)\n\n#add variables, constraints, and objective functions to nodes\n@variable(n1, 0 <= x <= 2)\n@variable(n1, 0 <= y <= 3)\n@constraint(n1, x+y <= 4)\n@objective(n1, Min, x)\n\n@variable(n2,x)\n@NLconstraint(n2, exp(x) >= 2)\n\n#add a linkconstraint to couple nodes\n@linkconstraint(graph, n1[:x] == n2[:x])\n\n#optimize with Ipopt\nset_optimizer(graph, Ipopt.Optimizer)\noptimize!(graph)\n\n#Print solution values\nprintln(\"n1[:x] = \", value(n1[:x]))\nprintln(\"n2[:x] = \", value(n2[:x]))","category":"page"},{"location":"packages/Plasmo/#Acknowledgments","page":"plasmo-dev/Plasmo.jl","title":"Acknowledgments","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"This code is based on work supported by the following funding agencies:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"U.S. Department of Energy (DOE), Office of Science, under Contract No. DE-AC02-06CH11357\nDOE Office of Electricity Delivery and Energy Reliability’s Advanced Grid Research and Development program at Argonne National Laboratory\nNational Science Foundation under award NSF-EECS-1609183 and under award CBET-1748516","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"The primary developer is Jordan Jalving (@jalving) with support from the following contributors.  ","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"Victor Zavala (University of Wisconsin-Madison)\nYankai Cao (University of British Columbia)\nKibaek Kim (Argonne National Laboratory)\nSungho Shin (University of Wisconsin-Madison)","category":"page"},{"location":"packages/Plasmo/#Citing-Plasmo.jl","page":"plasmo-dev/Plasmo.jl","title":"Citing Plasmo.jl","text":"","category":"section"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"If you find Plasmo.jl useful for your work, you may cite the manuscript as:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"@article{JalvingShinZavala2022,\n  title={A Graph-Based Modeling Abstraction for Optimization: Concepts and Implementation in Plasmo.jl},\n  author={Jordan Jalving and Sungho Shin and Victor M. Zavala},\n  journal={Mathematical Programming Computation},\n  year={2022},\n  volume={14},\n  pages={699 - 747}\n}","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"There is also a freely available pre-print:","category":"page"},{"location":"packages/Plasmo/","page":"plasmo-dev/Plasmo.jl","title":"plasmo-dev/Plasmo.jl","text":"@misc{JalvingShinZavala2020,\ntitle = {A Graph-Based Modeling Abstraction for Optimization: Concepts and Implementation in Plasmo.jl},\nauthor = {Jordan Jalving and Sungho Shin and Victor M. Zavala},\nyear = {2020},\neprint = {2006.05378},\narchivePrefix = {arXiv},\nprimaryClass = {math.OC}\n}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"EditURL = \"https://github.com/jump-dev/MiniZinc.jl/blob/v0.3.4/README.md\"","category":"page"},{"location":"packages/MiniZinc/#MiniZinc.jl","page":"jump-dev/MiniZinc.jl","title":"MiniZinc.jl","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl is a wrapper for the MiniZinc constraint modeling language.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"It provides a way to write MathOptInterface models to .mzn files, and a way to interact with libminizinc.","category":"page"},{"location":"packages/MiniZinc/#Affiliation","page":"jump-dev/MiniZinc.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"This wrapper is maintained by the JuMP community and is not part of the MiniZinc project.","category":"page"},{"location":"packages/MiniZinc/#License","page":"jump-dev/MiniZinc.jl","title":"License","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl is licensed under the MIT License.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The underlying project, MiniZinc/libminizinc, is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/MiniZinc/#Install","page":"jump-dev/MiniZinc.jl","title":"Install","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Install MiniZinc.jl using the Julia package manager:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"import Pkg\nPkg.add(\"MiniZinc\")","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Windows","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"On Linux and macOS, this package automatically installs libminizinc. However, we're still working out problems with the install on Windows. To use MiniZinc.jl, you'll need to manually install a copy of libminizinc from minizinc.org or compile one yourself from MiniZinc/libminizinc.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"To teach MiniZinc.jl where to look for libminizinc, set the JULIA_LIBMINIZINC_DIR environment variable. For example:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"ENV[\"JULIA_LIBMINIZINC_DIR\"] = \"C:\\\\Program Files\\\\MiniZinc\"","category":"page"},{"location":"packages/MiniZinc/#Use-with-MathOptInterface","page":"jump-dev/MiniZinc.jl","title":"Use with MathOptInterface","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl supports the constraint programming sets defined in MathOptInterface, as well as (in)equality constraints.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The following example solves the following constraint program:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"xᵢ ∈ {1, 2, 3} ∀i=1,2,3\nzⱼ ∈ {0, 1}    ∀j=1,2\nz₁ <-> x₁ != x₂\nz₂ <-> x₂ != x₃\nz₁ + z₂ = 1","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"julia> import MiniZinc\n\njulia> import MathOptInterface as MOI\n\njulia> function main()\n           model = MOI.Utilities.CachingOptimizer(\n               MiniZinc.Model{Int}(),\n               MiniZinc.Optimizer{Int}(\"chuffed\"),\n           )\n           # xᵢ ∈ {1, 2, 3} ∀i=1,2,3\n           x = MOI.add_variables(model, 3)\n           MOI.add_constraint.(model, x, MOI.Interval(1, 3))\n           MOI.add_constraint.(model, x, MOI.Integer())\n           # zⱼ ∈ {0, 1}    ∀j=1,2\n           z = MOI.add_variables(model, 2)\n           MOI.add_constraint.(model, z, MOI.ZeroOne())\n           # z₁ <-> x₁ != x₂\n           MOI.add_constraint(\n               model,\n               MOI.VectorOfVariables([z[1], x[1], x[2]]),\n               MOI.Reified(MOI.AllDifferent(2)),\n           )\n           # z₂ <-> x₂ != x₃\n           MOI.add_constraint(\n               model,\n               MOI.VectorOfVariables([z[2], x[2], x[3]]),\n               MOI.Reified(MOI.AllDifferent(2)),\n           )\n           # z₁ + z₂ = 1\n           MOI.add_constraint(model, 1 * z[1] + x[2], MOI.EqualTo(1))\n           MOI.optimize!(model)\n           x_star = MOI.get(model, MOI.VariablePrimal(), x)\n           z_star = MOI.get(model, MOI.VariablePrimal(), z)\n           return x_star, z_star\n       end\nmain (generic function with 1 method)\n\njulia> main()\n([1, 1, 3], [0, 1])","category":"page"},{"location":"packages/MiniZinc/#Use-with-JuMP","page":"jump-dev/MiniZinc.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"You can also call MiniZinc from JuMP, using any solver that libminizinc supports. By default, MiniZinc.jl is compiled with \"highs\":","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"using JuMP\nimport MiniZinc\nmodel = Model(() -> MiniZinc.Optimizer{Float64}(\"highs\"))\n@variable(model, 1 <= x[1:3] <= 3, Int)\n@constraint(model, x in MOI.AllDifferent(3))\n@objective(model, Max, sum(i * x[i] for i in 1:3))\noptimize!(model)\n@show value.(x)","category":"page"},{"location":"packages/MiniZinc/#MathOptInterface-API","page":"jump-dev/MiniZinc.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"The MiniZinc Optimizer{T} supports the following constraints and attributes.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.Reals","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\nMOI.ScalarAffineFunction{T} in MOI.Integer\nMOI.ScalarAffineFunction{T} in MOI.Interval{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.ZeroOne\nMOI.VariableIndex in MOI.EqualTo{T}\nMOI.VariableIndex in MOI.GreaterThan{T}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{T}\nMOI.VariableIndex in MOI.LessThan{T}\nMOI.VariableIndex in MOI.Parameter{T}\nMOI.VariableIndex in MOI.Semicontinuous{T}\nMOI.VariableIndex in MOI.Semiinteger{T}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.AllDifferent\nMOI.VectorOfVariables in MOI.BinPacking{T}\nMOI.VectorOfVariables in MOI.Circuit\nMOI.VectorOfVariables in MOI.CountAtLeast\nMOI.VectorOfVariables in MOI.CountBelongs\nMOI.VectorOfVariables in MOI.CountDistinct\nMOI.VectorOfVariables in MOI.CountGreaterThan\nMOI.VectorOfVariables in MOI.Cumulative\nMOI.VectorOfVariables in MOI.Path\nMOI.VectorOfVariables in MOI.Table{T}","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MOI.NLPBlock()\nMOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/MiniZinc/#Options","page":"jump-dev/MiniZinc.jl","title":"Options","text":"","category":"section"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"Set options using MOI.RawOptimizerAttribute in MOI or set_attribute in JuMP.","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"MiniZinc.jl supports the following options:","category":"page"},{"location":"packages/MiniZinc/","page":"jump-dev/MiniZinc.jl","title":"jump-dev/MiniZinc.jl","text":"model_filename::String = \"\": the location at which to write out the .mzn file during optimization. This option can be helpful during debugging. If left empty, a temporary file will be used instead.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"EditURL = \"transp.jl\"","category":"page"},{"location":"tutorials/linear/transp/#The-transportation-problem","page":"The transportation problem","title":"The transportation problem","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial is an adaptation of the transportation problem described in AMPL: A Modeling Language for Mathematical Programming, by R. Fourer, D.M. Gay and B.W. Kernighan.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"The purpose of this tutorial is to demonstrate how to create a JuMP model from an ad-hoc structured text file.","category":"page"},{"location":"tutorials/linear/transp/#Required-packages","page":"The transportation problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"using JuMP\nimport DelimitedFiles\nimport HiGHS","category":"page"},{"location":"tutorials/linear/transp/#Formulation","page":"The transportation problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Suppose that we have a set of factories that produce pogo sticks, and a set of retail stores in which to sell them. Each factory has a maximum number of pogo sticks that it can produce, and each retail store has a demand of pogo sticks that it can sell.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"In the transportation problem, we want to choose the number of pogo sticks to make and ship from each factory to each retail store that minimizes the total shipping cost.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Mathematically, we represent our set of factories by a set of origins i in O and our retail stores by a set of destinations j in D. The maximum supply at each factory is s_i and the demand from each retail store is d_j. The cost of shipping one pogo stick from i to j is c_ij.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"With a little effort, we can model the transportation problem as the following linear program:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"beginaligned\nmin   sum_i in O j in D c_ij x_ij \nst   sum_j in D x_i j le s_i  forall i in O \n       sum_i in O x_i j = d_j  forall j in D \n       x_i j ge 0  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/transp/#Data","page":"The transportation problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"We assume our data is in the form of a text file that has the following form. In practice, we would obtain this text file from the user as input, but for the purpose of this tutorial we're going to create it from Julia.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"open(joinpath(@__DIR__, \"transp.txt\"), \"w\") do io\n    print(\n        io,\n        \"\"\"\n             . FRA  DET LAN WIN  STL  FRE  LAF SUPPLY\n          GARY  39   14  11  14   16   82    8   1400\n          CLEV  27    .  12   .   26   95   17   2600\n          PITT  24   14  17  13   28   99   20   2900\n        DEMAND 900 1200 600 400 1700 1100 1000      0\n        \"\"\",\n    )\n    return\nend","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Here the rows are the origins, the columns are the destinations, and the values are the cost of shipping one pogo stick from the origin to the destination. If pogo stick cannot be transported from a source to a destination, then the value is .. The final row and final column are the demand and supply of each location respectively.","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"We didn't account for arcs which do not exist in our formulation, but we can make a small change and fix x_ij = 0 if c_ij = .","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Our first step is to convert this text format into an appropriate Julia datastructure that we can work with. Since our data is tabular with named rows and columns, one option is JuMP's Containers.DenseAxisArray object:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"function read_data(filename::String)\n    data = DelimitedFiles.readdlm(filename)\n    rows, columns = data[2:end, 1], data[1, 2:end]\n    return Containers.DenseAxisArray(data[2:end, 2:end], rows, columns)\nend\n\ndata = read_data(joinpath(@__DIR__, \"transp.txt\"))","category":"page"},{"location":"tutorials/linear/transp/#JuMP-formulation","page":"The transportation problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Following Design patterns for larger models, we code our JuMP model as a function which takes in an input. In this example, we print the output to stdout:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"function solve_transportation_problem(data::Containers.DenseAxisArray)\n    # Get the set of supplies and demands\n    O, D = axes(data)\n    # Drop the SUPPLY and DEMAND nodes from our sets\n    O, D = setdiff(O, [\"DEMAND\"]), setdiff(D, [\"SUPPLY\"])\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x[o in O, d in D] >= 0)\n    # Remove arcs with \".\" cost by fixing them to 0.0.\n    for o in O, d in D\n        if data[o, d] == \".\"\n            fix(x[o, d], 0.0; force = true)\n        end\n    end\n    @objective(\n        model,\n        Min,\n        sum(data[o, d] * x[o, d] for o in O, d in D if data[o, d] != \".\"),\n    )\n    @constraint(model, [o in O], sum(x[o, :]) <= data[o, \"SUPPLY\"])\n    @constraint(model, [d in D], sum(x[:, d]) == data[\"DEMAND\", d])\n    optimize!(model)\n    # Pretty print the solution in the format of the input\n    print(\"    \", join(lpad.(D, 7, ' ')))\n    for o in O\n        print(\"\\n\", o)\n        for d in D\n            if isapprox(value(x[o, d]), 0.0; atol = 1e-6)\n                print(\"      .\")\n            else\n                print(\" \", lpad(value(x[o, d]), 6, ' '))\n            end\n        end\n    end\n    return\nend","category":"page"},{"location":"tutorials/linear/transp/#Solution","page":"The transportation problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"Let's solve and view the solution:","category":"page"},{"location":"tutorials/linear/transp/","page":"The transportation problem","title":"The transportation problem","text":"solve_transportation_problem(data)","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/developer/checklists.md\"","category":"page"},{"location":"moi/developer/checklists/#Checklists","page":"Checklists","title":"Checklists","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"The purpose of this page is to collate a series of checklists for commonly performed changes to the source code of MathOptInterface.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"In each case, copy the checklist into the description of the pull request.","category":"page"},{"location":"moi/developer/checklists/#Making-a-release","page":"Checklists","title":"Making a release","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when making a release of the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] `version` field of `Project.toml` has been updated\n       - If a breaking change, increment the MAJOR field and reset others to 0\n       - If adding new features, increment the MINOR field and reset PATCH to 0\n       - If adding bug fixes or documentation changes, increment the PATCH field\n\n## Documentation\n\n - [ ] Add a new entry to `docs/src/changelog.md`, following existing style\n\n## Tests\n\n - [ ] The `solver-tests.yml` GitHub action does not have unexpected failures.\n       To run the action, go to:\n       https://github.com/jump-dev/MathOptInterface.jl/actions/workflows/solver-tests.yml\n       and click \"Run workflow\"","category":"page"},{"location":"moi/developer/checklists/#Adding-a-new-set","page":"Checklists","title":"Adding a new set","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when adding a new set to the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Add a new `AbstractScalarSet` or `AbstractVectorSet` to `src/sets.jl`\n - [ ] If `isbitstype(S) == false`, implement `Base.copy(set::S)`\n - [ ] If `isbitstype(S) == false`, implement `Base.:(==)(x::S, y::S)`\n - [ ] If an `AbstractVectorSet`, implement `dimension(set::S)`, unless the\n       dimension is given by `set.dimension`.\n\n## Utilities\n\n - [ ] If an `AbstractVectorSet`, implement `Utilities.set_dot`,\n       unless the dot product between two vectors in the set is equivalent to\n       `LinearAlgebra.dot`\n - [ ] If an `AbstractVectorSet`, implement `Utilities.set_with_dimension` in\n       `src/Utilities/matrix_of_constraints.jl`\n - [ ] Add the set to the `@model` macro at the bottom of `src/Utilities.model.jl`\n\n## Documentation\n\n - [ ] Add a docstring, which gives the mathematical definition of the set,\n       along with an `## Example` block containing a `jldoctest`\n - [ ] Add the docstring to `docs/src/reference/standard_form.md`\n - [ ] Add the set to the relevant table in `docs/src/manual/standard_form.md`\n\n## Tests\n\n - [ ] Define a new `_set(::Type{S})` method in `src/Test/test_basic_constraint.jl`\n       and add the name of the set to the list at the bottom of that files\n - [ ] If the set has any checks in its constructor, add tests to `test/sets.jl`\n\n## MathOptFormat\n\n - [ ] Open an issue at `https://github.com/jump-dev/MathOptFormat` to add\n       support for the new set {{ replace with link to the issue }}\n\n## Optional\n\n - [ ] Implement `dual_set(::S)` and `dual_set_type(::Type{S})`\n - [ ] Add new tests to the `Test` submodule exercising your new set\n - [ ] Add new bridges to convert your set into more commonly used sets","category":"page"},{"location":"moi/developer/checklists/#Adding-a-new-bridge","page":"Checklists","title":"Adding a new bridge","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when adding a new bridge to the MathOptInterface repository.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"The steps are mostly the same, but locations depend on whether the bridge is a Constraint, Objective, or Variable bridge. In each case below, replace XXX with the appropriate type of bridge.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Create a new file in `src/Bridges/XXX/bridges`\n - [ ] Define the bridge, following existing examples. The name of the bridge\n       struct must end in `Bridge`\n - [ ] Check if your bridge can be a subtype of [`MOI.Bridges.Constraint.SetMapBridge`](@ref)\n - [ ] Define a new `const` that is a `SingleBridgeOptimizer` wrapping the\n       new bridge. The name of the const must be the name of the bridge, less\n       the `Bridge` suffix\n - [ ] `include` the file in `src/Bridges/XXX/bridges/XXX.jl`\n - [ ] If the bridge should be enabled by default, add the bridge to\n       `add_all_bridges` at the bottom of `src/Bridges/XXX/XXX.jl`\n\n## Tests\n\n - [ ] Create a new file in the appropriate subdirectory of `tests/Bridges/XXX`\n - [ ] Use `MOI.Bridges.runtests` to test various inputs and outputs of the\n       bridge\n - [ ] If, after opening the pull request to add the bridge, some lines are not\n       covered by the tests, add additional bridge-specific tests to cover the\n       untested lines.\n\n## Documentation\n\n - [ ] Add a docstring which uses the same template as existing bridges.\n - [ ] Add the docstring to `docs/src/submodules/Bridges/list_of_bridges.md`\n\n## Final touch\n\nIf the bridge depends on run-time values of other variables and constraints in\nthe model:\n\n - [ ] Implement `MOI.Utilities.needs_final_touch(::Bridge)`\n - [ ] Implement `MOI.Utilities.final_touch(::Bridge, ::MOI.ModelLike)`\n - [ ] Ensure that `final_touch` can be called multiple times in a row","category":"page"},{"location":"moi/developer/checklists/#Updating-MathOptFormat","page":"Checklists","title":"Updating MathOptFormat","text":"","category":"section"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"Use this checklist when updating the version of MathOptFormat.","category":"page"},{"location":"moi/developer/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] The file at `src/FileFormats/MOF/mof.X.Y.schema.json` is updated\n - [ ] The constants `SCHEMA_PATH`, `VERSION`, and `SUPPORTED_VERSIONS` are\n       updated in `src/FileFormats/MOF/MOF.jl`\n\n## New sets\n\n - [ ] New sets are added to the `@model` in `src/FileFormats/MOF/MOF.jl`\n - [ ] New sets are added to the `@enum` in `src/FileFormats/MOF/read.jl`\n - [ ] `set_to_moi` is defined for each set in `src/FileFormats/MOF/read.jl`\n - [ ] `head_name` is defined for each set in `src/FileFormats/MOF/write.jl`\n - [ ] A new unit test calling `_test_model_equality` is aded to\n       `test/FileFormats/MOF/MOF.jl`\n\n## Tests\n\n - [ ] The version field in `test/FileFormats/MOF/nlp.mof.json` is updated\n\n## Documentation\n\n - [ ] The version fields are updated in `docs/src/submodules/FileFormats/overview.md`","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"DocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/expressions/#Expressions","page":"Expressions","title":"Expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"JuMP has three types of expressions: affine, quadratic, and nonlinear. These expressions can be inserted into constraints or into the objective. This is particularly useful if an expression is used in multiple places in the model.","category":"page"},{"location":"manual/expressions/#Affine-expressions","page":"Expressions","title":"Affine expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"There are four ways of constructing an affine expression in JuMP: with the @expression macro, with operator overloading, with the AffExpr constructor, and with add_to_expression!.","category":"page"},{"location":"manual/expressions/#Macros","page":"Expressions","title":"Macros","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The recommended way to create an affine expression is via the @expression macro.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = @expression(model, 2x + y - 1)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This expression can be used in the objective or added to a constraint. For example:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> @objective(model, Min, 2 * ex - 1)\n4 x + 2 y - 3\n\njulia> objective_function(model)\n4 x + 2 y - 3","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Just like variables and constraints, named expressions can also be created. For example","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x[i = 1:3]);\n\njulia> @expression(model, expr[i = 1:3], i * sum(x[j] for j in i:3));\n\njulia> expr\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[2] + 2 x[3]\n 3 x[3]","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/expressions/#Operator-overloading","page":"Expressions","title":"Operator overloading","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Expressions can also be created without macros. However, note that in some cases, this can be much slower that constructing an expression using macros.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = 2x + y - 1\n2 x + y - 1","category":"page"},{"location":"manual/expressions/#Constructors","page":"Expressions","title":"Constructors","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"A third way to create an affine expression is by the AffExpr constructor. The first argument is the constant term, and the remaining arguments are variable-coefficient pairs.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = AffExpr(-1.0, x => 2.0, y => 1.0)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/#add_to_expression!","page":"Expressions","title":"add_to_expression!","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The fourth way to create an affine expression is by using add_to_expression!. Compared to the operator overloading method, this approach is faster because it avoids constructing temporary objects. The @expression macro uses add_to_expression! behind-the-scenes.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = AffExpr(-1.0)\n-1\n\njulia> add_to_expression!(ex, 2.0, x)\n2 x - 1\n\njulia> add_to_expression!(ex, 1.0, y)\n2 x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"warning: Warning\nRead the section Initializing arrays for some cases to be careful about when using add_to_expression!.","category":"page"},{"location":"manual/expressions/#Removing-zero-terms","page":"Expressions","title":"Removing zero terms","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use drop_zeros! to remove terms from an affine expression with a 0 coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, ex, x + 1 - x)\n0 x + 1\n\njulia> drop_zeros!(ex)\n\njulia> ex\n1","category":"page"},{"location":"manual/expressions/#Coefficients","page":"Expressions","title":"Coefficients","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use coefficient to return the coefficient associated with a variable in an affine expression.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @expression(model, ex, 2x + 1)\n2 x + 1\n\njulia> coefficient(ex, x)\n2.0\n\njulia> coefficient(ex, y)\n0.0","category":"page"},{"location":"manual/expressions/#Quadratic-expressions","page":"Expressions","title":"Quadratic expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Like affine expressions, there are four ways of constructing a quadratic expression in JuMP: macros, operator overloading, constructors, and add_to_expression!.","category":"page"},{"location":"manual/expressions/#Macros-2","page":"Expressions","title":"Macros","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The @expression macro can be used to create quadratic expressions by including quadratic terms.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = @expression(model, x^2 + 2 * x * y + y^2 + x + y - 1)\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#Operator-overloading-2","page":"Expressions","title":"Operator overloading","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Operator overloading can also be used to create quadratic expressions. The same performance warning (discussed in the affine expression section) applies.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = x^2 + 2 * x * y + y^2 + x + y - 1\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#Constructors-2","page":"Expressions","title":"Constructors","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Quadratic expressions can also be created using the QuadExpr constructor. The first argument is an affine expression, and the remaining arguments are pairs, where the first term is a JuMP.UnorderedPair and the second term is the coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> aff_expr = AffExpr(-1.0, x => 1.0, y => 1.0)\nx + y - 1\n\njulia> quad_expr = QuadExpr(\n           aff_expr,\n           UnorderedPair(x, x) => 1.0,\n           UnorderedPair(x, y) => 2.0,\n           UnorderedPair(y, y) => 1.0,\n       )\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/#add_to_expression!-2","page":"Expressions","title":"add_to_expression!","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Finally, add_to_expression! can also be used to add quadratic terms.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> ex = QuadExpr(x + y - 1.0)\nx + y - 1\n\njulia> add_to_expression!(ex, 1.0, x, x)\nx² + x + y - 1\n\njulia> add_to_expression!(ex, 2.0, x, y)\nx² + 2 x*y + x + y - 1\n\njulia> add_to_expression!(ex, 1.0, y, y)\nx² + 2 x*y + y² + x + y - 1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"warning: Warning\nRead the section Initializing arrays for some cases to be careful about when using add_to_expression!.","category":"page"},{"location":"manual/expressions/#Removing-zero-terms-2","page":"Expressions","title":"Removing zero terms","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use drop_zeros! to remove terms from a quadratic expression with a 0 coefficient.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, ex, x^2 + x + 1 - x^2)\n0 x² + x + 1\n\njulia> drop_zeros!(ex)\n\njulia> ex\nx + 1","category":"page"},{"location":"manual/expressions/#Coefficients-2","page":"Expressions","title":"Coefficients","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Use coefficient to return the coefficient associated with a pair of variables in a quadratic expression.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @expression(model, ex, 2*x*y + 3*x)\n2 x*y + 3 x\n\njulia> coefficient(ex, x, y)\n2.0\n\njulia> coefficient(ex, x, x)\n0.0\n\njulia> coefficient(ex, y, x)\n2.0\n\njulia> coefficient(ex, x)\n3.0","category":"page"},{"location":"manual/expressions/#Nonlinear-expressions","page":"Expressions","title":"Nonlinear expressions","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Nonlinear expressions in JuMP are represented by a NonlinearExpr object. See Nonlinear expressions in detail for more details.","category":"page"},{"location":"manual/expressions/#Initializing-arrays","page":"Expressions","title":"Initializing arrays","text":"","category":"section"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"JuMP implements zero(AffExpr) and one(AffExpr) to support various functions in LinearAlgebra (for example, accessing the off-diagonal of a Diagonal matrix).","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> zero(AffExpr)\n0\n\njulia> one(AffExpr)\n1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"However, this can result in a subtle bug if you call add_to_expression! or the MutableArithmetics API on an element created by zeros or ones:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = zeros(AffExpr, 2)\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 1.1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Notice how we modified x[1], but we also changed x[2]!","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This happened because zeros(AffExpr, 2) calls zero(AffExpr) once to obtain a zero element, and then creates an appropriately sized array filled with the same element.","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"This also happens with broadcasting calls containing a conversion of 0 or 1:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = Vector{AffExpr}(undef, 2)\n2-element Vector{AffExpr}:\n #undef\n #undef\n\njulia> x .= 0\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 1.1","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"The recommended way to create an array of empty expressions is as follows:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = Vector{AffExpr}(undef, 2)\n2-element Vector{AffExpr}:\n #undef\n #undef\n\njulia> for i in eachindex(x)\n           x[i] = AffExpr(0.0)\n       end\n\njulia> add_to_expression!(x[1], 1.1)\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 0","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Alternatively, use non-mutating operation to avoid updating x[1] in-place:","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"julia> x = zeros(AffExpr, 2)\n2-element Vector{AffExpr}:\n 0\n 0\n\njulia> x[1] += 1.1\n1.1\n\njulia> x\n2-element Vector{AffExpr}:\n 1.1\n 0","category":"page"},{"location":"manual/expressions/","page":"Expressions","title":"Expressions","text":"Note that for large expressions this will be slower due to the allocation of additional temporary objects.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"EditURL = \"factory_schedule.jl\"","category":"page"},{"location":"tutorials/linear/factory_schedule/#The-factory-schedule-example","page":"The factory schedule example","title":"The factory schedule example","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial was originally contributed by @Crghilardi.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial is a Julia translation of Part 5 from Introduction to Linear Programming with Python.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The purpose of this tutorial is to demonstrate how to use DataFrames and delimited files, and to structure your code that is robust to infeasibilities and permits running with different datasets.","category":"page"},{"location":"tutorials/linear/factory_schedule/#Required-packages","page":"The factory schedule example","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"using JuMP\nimport CSV\nimport DataFrames\nimport HiGHS\nimport StatsPlots","category":"page"},{"location":"tutorials/linear/factory_schedule/#Formulation","page":"The factory schedule example","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The Factory Scheduling Problem assumes we are optimizing the production of a good from factories f in F over the course of 12 months m in M.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"If a factory f runs during a month m, a fixed cost of a_f is incurred, the factory must produce x_mf units that is within some minimum and maximum production levels l_f and u_f respectively, and each unit of production incurs a variable cost c_f. Otherwise, the factory can be shut for the month with zero production and no fixed-cost is incurred. We denote the run/not-run decision by z_mf in 0 1, where z_mf is 1 if factory f runs in month m. The factory must produce enough units to satisfy demand d_m.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"With a little effort, we can formulate our problem as the following linear program:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"beginaligned\nmin  sumlimits_f in F m in M a_f z_mf + c_f x_mf \ntextst  x_mf le u_f z_mf  forall f in F m in M \n             x_mf ge l_f z_mf  forall f in F m in M \n             sumlimits_fin F x_mf = d_m  forall f in F m in M \n             z_mf in 0 1  forall f in F m in M\nendaligned","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"However, this formulation has a problem: if demand is too high, we may be unable to satisfy the demand constraint, and the problem will be infeasible.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"tip: Tip\nWhen modeling, consider ways to formulate your model such that it always has a feasible solution. This greatly simplifies debugging data errors that would otherwise result in an infeasible solution. In practice, most practical decisions have a feasible solution. In our case, we could satisfy demand (at a high cost) by buying replacement items for the buyer, or running the factories in overtime to make up the difference.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"We can improve our model by adding a new variable, delta_m, which represents the quantity of unmet demand in each month m. We penalize delta_m by an arbitrarily large value of $10,000/unit in the objective.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"beginaligned\nmin  sumlimits_f in F m in M a_f z_mf + c_f x_mf + sumlimits_m in M10000 delta_m \ntextst  x_mf le u_f z_mf  forall f in F m in M \n             x_mf ge l_f z_mf  forall f in F m in M \n             sumlimits_fin F x_mf - delta_m = d_m  forall f in F m in M \n             z_mf in 0 1  forall f in F m in M \n             delta_m ge 0  forall m in M\nendaligned","category":"page"},{"location":"tutorials/linear/factory_schedule/#Data","page":"The factory schedule example","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The JuMP GitHub repository contains two text files with the data we need for this tutorial.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The first file contains a dataset of our factories, A and B, with their production and cost levels for each month. For the documentation, the file is located at:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factories_filename = joinpath(@__DIR__, \"factory_schedule_factories.txt\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"To run locally, download factory_schedule_factories.txt and update factories_filename appropriately.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The file has the following contents:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"print(read(factories_filename, String))","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"We use the CSV and DataFrames packages to read it into Julia:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factory_df = CSV.read(\n    factories_filename,\n    DataFrames.DataFrame;\n    delim = ' ',\n    ignorerepeated = true,\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"The second file contains the demand data by month:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_filename = joinpath(@__DIR__, \"factory_schedule_demand.txt\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"To run locally, download factory_schedule_demand.txt and update demand_filename appropriately.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df = CSV.read(\n    demand_filename,\n    DataFrames.DataFrame;\n    delim = ' ',\n    ignorerepeated = true,\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/#Data-validation","page":"The factory schedule example","title":"Data validation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Before moving on, it's always good practice to validate the data you read from external sources. The more effort you spend here, the fewer issues you will have later. The following function contains a few simple checks, but we could add more. For example, you might want to check that none of the values are too large (or too small), which might indicate a typo or a unit conversion issue (perhaps the variable costs are in $/1000 units instead of $/unit).","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function valiate_data(\n    demand_df::DataFrames.DataFrame,\n    factory_df::DataFrames.DataFrame,\n)\n    # Minimum production must not exceed maximum production.\n    @assert all(factory_df.min_production .<= factory_df.max_production)\n    # Demand, minimum production, fixed costs, and variable costs must all be\n    # non-negative.\n    @assert all(demand_df.demand .>= 0)\n    @assert all(factory_df.min_production .>= 0)\n    @assert all(factory_df.fixed_cost .>= 0)\n    @assert all(factory_df.variable_cost .>= 0)\n    return\nend\n\nvaliate_data(demand_df, factory_df)","category":"page"},{"location":"tutorials/linear/factory_schedule/#JuMP-formulation","page":"The factory schedule example","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Next, we need to code our JuMP formulation. As shown in Design patterns for larger models, it's always good practice to code your model in a function that accepts well-defined input and returns well-defined output.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function solve_factory_scheduling(\n    demand_df::DataFrames.DataFrame,\n    factory_df::DataFrames.DataFrame,\n)\n    # Even though we validated the data above, it's good practice to do it here\n    # too.\n    valiate_data(demand_df, factory_df)\n    months, factories = unique(factory_df.month), unique(factory_df.factory)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, status[months, factories], Bin)\n    @variable(model, production[months, factories], Int)\n    @variable(model, unmet_demand[months] >= 0)\n    # We use `eachrow` to loop through the rows of the dataframe and add the\n    # relevant constraints.\n    for r in eachrow(factory_df)\n        m, f = r.month, r.factory\n        @constraint(model, production[m, f] <= r.max_production * status[m, f])\n        @constraint(model, production[m, f] >= r.min_production * status[m, f])\n    end\n    @constraint(\n        model,\n        [r in eachrow(demand_df)],\n        sum(production[r.month, :]) + unmet_demand[r.month] == r.demand,\n    )\n    @objective(\n        model,\n        Min,\n        10_000 * sum(unmet_demand) + sum(\n            r.fixed_cost * status[r.month, r.factory] +\n            r.variable_cost * production[r.month, r.factory] for\n            r in eachrow(factory_df)\n        )\n    )\n    optimize!(model)\n    schedules = Dict{Symbol,Vector{Float64}}(\n        Symbol(f) => value.(production[:, f]) for f in factories\n    )\n    schedules[:unmet_demand] = value.(unmet_demand)\n    return (\n        termination_status = termination_status(model),\n        cost = objective_value(model),\n        # This `select` statement re-orders the columns in the DataFrame.\n        schedules = DataFrames.select(\n            DataFrames.DataFrame(schedules),\n            [:unmet_demand, :A, :B],\n        ),\n    )\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/#Solution","page":"The factory schedule example","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Now we can call our solve_factory_scheduling function using the data we read in above.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution = solve_factory_scheduling(demand_df, factory_df);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's see what solution contains:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.termination_status","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.cost","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"solution.schedules","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"These schedules will be easier to visualize as a graph:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.groupedbar(\n    Matrix(solution.schedules);\n    bar_position = :stack,\n    labels = [\"unmet demand\" \"A\" \"B\"],\n    xlabel = \"Month\",\n    ylabel = \"Production\",\n    legend = :topleft,\n    color = [\"#20326c\" \"#4063d8\" \"#a0b1ec\"],\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Note that we don't have any unmet demand.","category":"page"},{"location":"tutorials/linear/factory_schedule/#What-happens-if-demand-increases?","page":"The factory schedule example","title":"What happens if demand increases?","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's run an experiment by increasing the demand by 50% in all time periods:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df.demand .*= 1.5","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Now we resolve the problem:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"high_demand_solution = solve_factory_scheduling(demand_df, factory_df);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"and visualize the solution:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.groupedbar(\n    Matrix(high_demand_solution.schedules);\n    bar_position = :stack,\n    labels = [\"unmet demand\" \"A\" \"B\"],\n    xlabel = \"Month\",\n    ylabel = \"Production\",\n    legend = :topleft,\n    color = [\"#20326c\" \"#4063d8\" \"#a0b1ec\"],\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Uh oh, we can't satisfy all of the demand.","category":"page"},{"location":"tutorials/linear/factory_schedule/#How-sensitive-is-the-solution-to-changes-in-variable-cost?","page":"The factory schedule example","title":"How sensitive is the solution to changes in variable cost?","text":"","category":"section"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's run another experiment, this time seeing how the optimal objective value changes as we vary the variable costs of each factory.","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"First though, let's reset the demand to it's original level:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"demand_df.demand ./= 1.5;\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"For our experiment, we're going to scale the variable costs of both factories by a set of values from 0.0 to 1.5:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"scale_factors = 0:0.1:1.5","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"At a high level, we're going to loop over the scale factors for A, then the scale factors for B, rescale the input data, call our solve_factory_scheduling example, and then store the optimal objective value in the following cost matrix:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"cost = zeros(length(scale_factors), length(scale_factors));\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Because we're modifying factory_df in-place, we need to store the original variable costs in a new column:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"factory_df[!, :old_variable_cost] = copy(factory_df.variable_cost);\nnothing #hide","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Then, we need a function to scale the :variable_cost column for a particular factory by a value scale:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"function scale_variable_cost(df, factory, scale)\n    rows = df.factory .== factory\n    df[rows, :variable_cost] .=\n        round.(Int, df[rows, :old_variable_cost] .* scale)\n    return\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Our experiment is just a nested for-loop, modifying A and B and storing the cost:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"for (j, a) in enumerate(scale_factors)\n    scale_variable_cost(factory_df, \"A\", a)\n    for (i, b) in enumerate(scale_factors)\n        scale_variable_cost(factory_df, \"B\", b)\n        cost[i, j] = solve_factory_scheduling(demand_df, factory_df).cost\n    end\nend","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"Let's visualize the cost matrix:","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"StatsPlots.contour(\n    scale_factors,\n    scale_factors,\n    cost;\n    xlabel = \"Scale of factory A\",\n    ylabel = \"Scale of factory B\",\n)","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"What can you infer from the solution?","category":"page"},{"location":"tutorials/linear/factory_schedule/","page":"The factory schedule example","title":"The factory schedule example","text":"info: Info\nThe Power Systems tutorial explains a number of other ways you can structure a problem to perform a parametric analysis of the solution. In particular, you can use in-place modification to reduce the time it takes to build and solve the resulting models.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"EditURL = \"changelog.md\"","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"CurrentModule = JuMP","category":"page"},{"location":"release_notes/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"release_notes/#[Version-1.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.16.0)-(October-24,-2023)","page":"Release notes","title":"Version 1.16.0 (October 24, 2023)","text":"","category":"section"},{"location":"release_notes/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added := operator for Boolean satisfiability problems (#3530)","category":"page"},{"location":"release_notes/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed text/latex printing of MOI.Interval sets (#3537)\nFixed tests with duplicate function names (#3539)","category":"page"},{"location":"release_notes/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated documentation list of supported solvers (#3527) (#3529) (#3538) (#3542) (#3545) (#3546)\nUpdated to Documenter@1.1 (#3528)\nFixed various tutorials (#3534) (#3532)\nFixed Project.toml compat bounds for standard libraries (#3544)","category":"page"},{"location":"release_notes/#[Version-1.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.15.1)-(September-24,-2023)","page":"Release notes","title":"Version 1.15.1 (September 24, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for single argument min and max operators (#3522)\nFixed error message for add_to_expression! when called with a GenericNonlinearExpr (#3506)\nFixed constraint tags with broadcasted constraints (#3515)\nFixed MethodError in MA.scaling (#3518)\nFixed support for arrays of Parameter variables (#3524)","category":"page"},{"location":"release_notes/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated to Documenter@1 (#3501)\nFixed links to data in tutorials (#3512)\nFixed typo in TSP tutorial (#3516)\nImproved error message for VariableNotOwned errors (#3520)\nFixed various JET errors (#3519)","category":"page"},{"location":"release_notes/#[Version-1.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.15.0)-(September-15,-2023)","page":"Release notes","title":"Version 1.15.0 (September 15, 2023)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it adds an entirely new data structure and API path for working with nonlinear programs. The previous nonlinear interface remains unchanged and is documented at Nonlinear Modeling (Legacy). The new interface is a treated as a non-breaking feature addition and is documented at Nonlinear Modeling.","category":"page"},{"location":"release_notes/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Although the new nonlinear interface is a feature addition, there are two changes which might be breaking for a very small number of users.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The syntax inside JuMP macros is parsed using a different code path, even for linear and quadratic expressions. We made this change to unify how we parse linear, quadratic, and nonlinear expressions. In all cases, the new code returns equivalent expressions, but because of the different order of operations, there are three changes to be aware of when updating:\nThe printed form of the expression may change, for example from x * y to y * x. This can cause tests which test the String representation of a model to fail.\nSome coefficients may change slightly due to floating point round-off error.\nParticularly when working with a JuMP extension, you may encounter a MethodError due to a missing or ambiguous method. These errors are due to previously existing bugs that were not triggered by the previous parsing code. If you encounter such an error, please open a GitHub issue.\nThe methods for Base.:^(x::VariableRef, n::Integer) and Base.:^(x::AffExpr, n::Integer) have changed. Previously, these methods supported only n = 0, 1, 2 and they always returned a QuadExpr, even for the case when n = 0 or n = 1. Now:\nx^0 returns one(T), where T is the value_type of the model (defaults to Float64)\nx^1 returns x\nx^2 returns a QuadExpr\nx^n where !(0 <= n <= 2) returns a NonlinearExpr.\nWe made this change to support nonlinear expressions and to align the mathematical definition of the operation with their return type. (Previously, users were surprised that x^1 returned a QuadExpr.) As a consequence of this change, the methods are now not type-stable. This means that the compiler cannot prove that x^2 returns a QuadExpr. If benchmarking shows that this is a performance problem, you can use the type-stable x * x instead of x^2.","category":"page"},{"location":"release_notes/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added triangle_vec which simplifies adding MOI.LogDetConeTriangle and MOI.RootDetConeTriangle constraints (#3456)\nAdded the new nonlinear interface. This is a very large change. See the documentation at Nonlinear Modeling and the (long) discussion in JuMP.jl#3106. Related PRs are (#3468) (#3472) (#3475) (#3483) (#3487) (#3488) (#3489) (#3504) (#3509)","category":"page"},{"location":"release_notes/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed uses of @nospecialize which cause precompilation failures in Julia v1.6.0 and v1.6.1. (#3464)\nFixed adding a container of Parameter (#3473)\nFixed return type of x^0 and x^1 to no longer return QuadExpr (see note in Breaking section above) (#3474)\nFixed error messages in LowerBoundRef, UpperBoundRef, FixRef, IntegerRef, BinaryRef, ParameterRef and related functions (#3494)\nFixed type inference of empty containers in JuMP macros (#3500)","category":"page"},{"location":"release_notes/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added GAMS to solver documentation (#3357)\nUpdated various tutorials (#3459) (#3460) (#3462) (#3463) (#3465) (#3490) (#3492) (#3503)\nAdded The network multi-commodity flow problem tutorial (#3491)\nAdded Two-stage stochastic programs tutorial (#3466)\nAdded better error messages for unsupported operations in LinearAlgebra (#3476)\nUpdated to the latest version of Documenter (#3484) (#3495) (#3497)\nUpdated GitHub action versions (#3507)","category":"page"},{"location":"release_notes/#[Version-1.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.14.1)-(September-2,-2023)","page":"Release notes","title":"Version 1.14.1 (September 2, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix links in Documentation (#3478)","category":"page"},{"location":"release_notes/#[Version-1.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.14.0)-(August-27,-2023)","page":"Release notes","title":"Version 1.14.0 (August 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added DimensionalData.jl extension (#3413)\nAdded syntactic sugar for the MOI.Parameter set (#3443)\nParameter\nParameterRef\nis_parameter\nparameter_value\nset_parameter_value","category":"page"},{"location":"release_notes/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed model_convert for BridgeableConstraint (#3437)\nFixed printing models with integer coefficients larger than typemax(Int) (#3447)\nFixed support for constant left-hand side functions in a complementarity constraint (#3452)","category":"page"},{"location":"release_notes/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Updated packages used in documentation (#3444) (#3455)\nFixed docstring tests (#3445)\nFixed printing change for MathOptInterface (#3446)\nFixed typos in documentation (#3448) (#3457)\nAdded SCIP to callback documentation (#3449)","category":"page"},{"location":"release_notes/#[Version-1.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.13.0)-(July-27,-2023)","page":"Release notes","title":"Version 1.13.0 (July 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for generic number types (#3377) (#3385)\nAdded fallback for MOI.AbstractSymmetricMatrixSetTriangle and MOI.AbstractSymmetricMatrixSetSquare (#3424)","category":"page"},{"location":"release_notes/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed set_start_values with MOI.Bridges.Objective.SlackBridge (#3422)\nFixed flakey doctest in variables.md (#3425)\nFixed names on CITATION.bib (#3423)","category":"page"},{"location":"release_notes/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Loraine.jl to the installation table (#3426)\nRemoved Penopt.jl from packages.toml (#3428)\nImproved problem statement in cannery example of tutorial (#3430)\nMinor cleanups in Containers.DenseAxisArray implementation (#3429)\nChanged nested_problems.jl: outer/inner to upper/lower (#3433)\nRemoved second SDP relaxation in OPF tutorial (#3432)","category":"page"},{"location":"release_notes/#[Version-1.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.12.0)-(June-19,-2023)","page":"Release notes","title":"Version 1.12.0 (June 19, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added coefficient_type keyword argument to add_bridge and remove_bridge (#3394)","category":"page"},{"location":"release_notes/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed error message for matrix in HermitianPSDCone (#3369)\nFixed EditURL for custom documentation pages (#3373)\nFixed return type annotations for MOI.ConstraintPrimal and MOI.ConstraintDual (#3381)\nFixed printing change in Julia nightly (#3391)\nFixed printing of Complex coefficients (#3397)\nFixed printing of constraints in text/latex mode (#3405)\nFixed performance issue in Containers.rowtable (#3410)\nFixed bug when variables added to set of wrong dimension (#3411)","category":"page"},{"location":"release_notes/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added more solver READMEs to the documentation (#3358) (#3360) (#3364) (#3365) (#3366) (#3368) (#3372) (#3374) (#3376) (#3379) (#3387) (#3389)\nAdded StatusSwitchingQP.jl to the installation table (#3354)\nUpdated checklist for adding a new solver (#3370)\nUpdated extension-tests.yml action (#3371) (#3375)\nColor logs in GitHub actions (#3392)\nAdded new tutorials\nOptimal power flow (#3395) (#3412)\nLovász numbers (#3399)\nDualization (#3402)\nUpdated JuMP paper citation (#3400)\nChanged GitHub action to upload LaTeX logs when building documentation (#3403)\nFixed printing of SCS log in documentation (#3406)\nUpdated solver versions (#3407)\nUpdated documentation to use Julia v1.9 (#3398)\nReplaced _value_type with MOI.Utilities.value_type (#3414)\nFixed a typo in docstring (#3415)\nRefactored API documentation (#3386)\nUpdated SCIP license (#3420)","category":"page"},{"location":"release_notes/#[Version-1.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.11.1)-(May-19,-2023)","page":"Release notes","title":"Version 1.11.1 (May 19, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a poor error message when sum(::DenseAxisArray; dims) was called (#3338)\nFixed support for dependent sets in the @variable macro (#3344)\nFixed a performance bug in constraints with sparse symmetric matrices (#3349)","category":"page"},{"location":"release_notes/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the printing of complex numbers (#3332)\nWhen printing, sets which contain constants ending in .0 now print as integers. This follows the behavior of constants in functions (#3341)\nAdded InfiniteOpt to the extensions documentation (#3343)\nAdded more documentation for the exponential cone (#3345) (#3347)\nAdded checklists for developers (#3346) (#3355)\nFixed test support upcoming Julia nightly (#3351)\nFixed extension-tests.yml action (#3353)\nAdd more solvers to the documentation (#3359) (#3361) (#3362)","category":"page"},{"location":"release_notes/#[Version-1.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.11.0)-(May-3,-2023)","page":"Release notes","title":"Version 1.11.0 (May 3, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added new methods to print_active_bridges for printing a particular objective, constraint, or variable (#3316)","category":"page"},{"location":"release_notes/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed tests for MOI v1.14.0 release (#3312)\nFixed indexing containers when an axis is Vector{Any} that contains a Vector{Any} element (#3280)\nFixed getindex(::AbstractJuMPScalar) which is called for an expression like x[] (#3314)\nFixed bug in set_string_names_on_creation with a vector of variables (#3322)\nFixed bug in memoize function in nonlinear documentation (#3337)","category":"page"},{"location":"release_notes/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#3317) (#3318) (#3328)\nAdded a test for the order of setting start values (#3315)\nAdded READMEs of solvers and extensions to the docs (#3309) (#3320) (#3327) (#3329) (#3333)\nStyle improvements to src/variables.jl (#3324)\nClarify that column generation does not find global optimum (#3325)\nAdd a GitHub actions workflow for testing extensions prior to release (#3331)\nDocument the release process for JuMP (#3334)\nFix links to discourse and chatroom (#3335)","category":"page"},{"location":"release_notes/#[Version-1.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.10.0)-(April-3,-2023)","page":"Release notes","title":"Version 1.10.0 (April 3, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Nonnegatives, Nonpositives and Zeros, and support vector-valued inequality syntax in the JuMP macros (#3273)\nAdded special support for LinearAlgebra.Symmetric and LinearAlgebra.Hermitian matrices in Zeros constraints (#3281) (#3296)\nAdded HermitianMatrixSpace and the Hermitian tag for generating a matrix of variables that is Hermitian (#3292) (#3293)\nAdded Semicontinuous and Semiinteger (#3302)\nAdded support for keyword indexing of containers (#3237)","category":"page"},{"location":"release_notes/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed [compat] bound for MathOptInterface in Project.toml (#3272)","category":"page"},{"location":"release_notes/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Split out the Nested optimization problems tutorial (#3274)\nUpdated doctests to ensure none have hidden state (#3275) (#3276)\nClarified how lazy constraints may revisit points (#3278)\nAdded P-Norm example (#3282)\nClarified docs that macros create new bindings (#3284)\nFixed threading example (#3283)\nAdded plot to The minimum distortion problem (#3288)\nAdded Google style rules for Vale and fixed warnings (#3285)\nAdded citation for the JuMP 1.0 paper (#3294)\nUpdated package versions in the documentation (#3298)\nAdded comment for the order in which start values must be set (#3303)\nImproved error message for unrecognized constraint operators (#3311)","category":"page"},{"location":"release_notes/#[Version-1.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.9.0)-(March-7,-2023)","page":"Release notes","title":"Version 1.9.0 (March 7, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added get_attribute and set_attribute. These replace get_optimizer_attribute and set_optimizer_attribute, although the _optimizer_ functions remain for backward compatibility. (#3219)\nAdded set_start_values for setting all supported start values in a model (#3238)\nAdd remove_bridge and print_active_bridges (#3259)","category":"page"},{"location":"release_notes/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The matrix returned by a variable in HermitianPSDCone is now a LinearAlgebra.Hermitian matrix. This is potentially breaking if you have written code to assume the return is a Matrix. (#3245) (#3246)\nFixed missing support for Base.isreal of expressions (#3252)","category":"page"},{"location":"release_notes/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a thread safety issue in the Parallelism tutorial (#3240) (#3243)\nImproved the error message when unsupported operators are used in @NL macros (#3236)\nClarified the documentation to say that matrices in HermitianPSDCone must be LinearAlgebra.Hermitian (#3241)\nMinor style fixes to internal macro code (#3247)\nAdd Quantum state discrimination tutorial (#3250)\nImprove error message when begin...end not passed to plural macros (#3255)\nDocument how to register function with varying number of input arguments (#3258)\nTidy tests by removing unneeded JuMP. prefixes (#3260)\nClarified the introduction to the Complex number support tutorial (#3262)\nFixed typos in the Documentation (#3263) (#3266) (#3268) (#3269)","category":"page"},{"location":"release_notes/#[Version-1.8.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.2)-(February-27,-2023)","page":"Release notes","title":"Version 1.8.2 (February 27, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed dot product between complex JuMP expression and number (#3244)","category":"page"},{"location":"release_notes/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Polish simple SDP examples (#3232)","category":"page"},{"location":"release_notes/#[Version-1.8.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.1)-(February-23,-2023)","page":"Release notes","title":"Version 1.8.1 (February 23, 2023)","text":"","category":"section"},{"location":"release_notes/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for init in nonlinear generator expressions (#3226)","category":"page"},{"location":"release_notes/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Use and document import MathOptInterface as MOI (#3222)\nRemoved references in documentation to multiobjective optimization being unsupported (#3223)\nAdded tutorial on multi-objective portfolio optimization (#3227)\nRefactored some of the conic tutorials (#3229)\nFixed typos in the documentation (#3230)\nAdded tutorial on parallelism (#3231)","category":"page"},{"location":"release_notes/#[Version-1.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.8.0)-(February-16,-2023)","page":"Release notes","title":"Version 1.8.0 (February 16, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added --> syntax support for indicator constraints. The old syntax of => remains supported (#3207)\nAdded <--> syntax for reified constraints. For now, few solvers support reified constraints (#3206)\nAdded fix_discrete_variables. This is most useful for computing the dual of a mixed-integer program (#3208)\nAdded support for vector-valued objectives. For details, see the Multi-objective knapsack tutorial (#3176)","category":"page"},{"location":"release_notes/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report by switching to an explicit LU factorization of the basis matrix (#3182)\nFixed a bug that prevented [; kwarg] arguments in macros (#3220)","category":"page"},{"location":"release_notes/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Minor fixes to the documentation (#3200) (#3201) (#3203) (#3210)\nAdded tutorial Constraint programming (#3202)\nAdded more examples to Tips and Tricks\nRemove _distance_to_set in favor of MOI.Utilities.distance_to_set (#3209)\nImprove The diet problem tutorial by adding the variable as a column in the dataframe (#3213)\nImprove The knapsack problem example tutorial (#3216) (#3217)\nAdded the Ellipsoid approximation tutorial (#3218)","category":"page"},{"location":"release_notes/#[Version-1.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.7.0)-(January-25,-2023)","page":"Release notes","title":"Version 1.7.0 (January 25, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for view of a Containers.DenseAxisArray (#3152) (#3180)\nAdded support for containers of variables in ComplexPlane (#3184)\nAdded support for minimum and maximum generators in nonlinear expressions (#3189)\nAdded SnoopPrecompile statements that reduce the time-to-first-solve in Julia 1.9 (#3193) (#3195) (#3196) (#3197)","category":"page"},{"location":"release_notes/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Large refactoring of the tests (#3166) (#3167) (#3168) (#3169) (#3170) (#3171)\nRemove unreachable code due to VERSION checks (#3172)\nDocument how to test JuMP extensions (#3174)\nFix method ambiguities in Containers (#3173)\nImprove error message that is thrown when = is used instead of == in the @constraint macro (#3178)\nImprove the error message when Bool is used instead of Bin in the @variable macro (#3180)\nUpdate versions of the documentation (#3185)\nTidy the import of packages and remove unnecessary prefixes (#3186) (#3187)\nRefactor src/JuMP.jl by moving methods into more relevant files (#3188)\nFix docstring of Model not appearing in the documentation (#3198)","category":"page"},{"location":"release_notes/#[Version-1.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.6.0)-(January-1,-2023)","page":"Release notes","title":"Version 1.6.0 (January 1, 2023)","text":"","category":"section"},{"location":"release_notes/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a result keyword argument to solution_summary to allow summarizing models with multiple solutions (#3138)\nAdded relax_with_penalty!, which is a useful tool when debugging infeasible models (#3140)\nAdded has_start_value (#3157)\nAdded support for HermitianPSDCone in constraints (#3154)","category":"page"},{"location":"release_notes/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed promotion of complex expressions (#3150) (#3164)","category":"page"},{"location":"release_notes/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Benders tutorial with in-place resolves (#3145)\nAdded more Tips and tricks for linear programs (#3144) (#3163)\nClarified documentation that start can depend on the indices of a variable container (#3148)\nReplace instances of length and size by the recommended eachindex and axes (#3149)\nAdded a warning explaining why the model is dirty when accessing solution results from a modified model (#3156)\nClarify documentation that PSD ensures a symmetric matrix (#3159)\nMaintenance of the JuMP test suite (#3146) (#3158) (#3162)","category":"page"},{"location":"release_notes/#[Version-1.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.5.0)-(December-8,-2022)","page":"Release notes","title":"Version 1.5.0 (December 8, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for complex-valued variables:\nHermitianPSDCone (#3109)\nComplexPlane and ComplexVariable (#3134)\nAdd support for MOI.OptimizerWithAttributes in set_optimizer_attribute and get_optimizer_attribute (#3129)","category":"page"},{"location":"release_notes/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed error message for vectorized interval constraints (#3123)\nFixed passing AbstractString to set_optimizer_attribute (#3127)","category":"page"},{"location":"release_notes/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update package versions used in docs (#3119) (#3133) (#3139)\nFixed output of diet tutorial (#3120)\nExplain how to use Dates.period in set_time_limit_sec (#3121)\nUpdate to JuliaFormatter v1.0.15 (#3130)\nFixed HTTP server example in web_app.jl (#3131)\nUpdate docs to build with Documenter#master (#3094)\nAdd tests for LinearAlgebra operations (#3132)\nTidy these release notes (#3135)\nAdded documentation for Complex number support (#3141)\nRemoved the \"workforce scheduling\" and \"steelT3\" tutorials (#3143)","category":"page"},{"location":"release_notes/#[Version-1.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.4.0)-(October-29,-2022)","page":"Release notes","title":"Version 1.4.0 (October 29, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Containers.rowtable which converts a container into a vector of NamedTuples to support the Tables.jl interface. This simplifies converting Containers.DenseAxisArray and Containers.SparseAxisArray objects into tabular forms such as a DataFrame (#3104)\nAdded a new method to Containers.container so that index names are passed to the container (#3088)","category":"page"},{"location":"release_notes/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in copy_to(dest::Model, src::MOI.ModelLike) when src has nonlinear components (#3101)\nFixed the printing of (-1.0 + 0.0im) coefficients in complex expressions (#3112)\nFixed a parsing bug in nonlinear expressions with generator statements that contain multiple for statements (#3116)","category":"page"},{"location":"release_notes/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Converted the multi-commodity flow tutorial to use an SQLite database (#3098)\nFixed a number of typos in the documentation (#3103) (#3107) (#3018)\nImproved various style aspects of the PDF documentation (#3095) (#3098) (#3102)","category":"page"},{"location":"release_notes/#[Version-1.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.3.1)-(September-28,-2022)","page":"Release notes","title":"Version 1.3.1 (September 28, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in relax_integrality (#3087)\nFixed the type stability of operators with Complex arguments (#3072)\nFixed a bug which added additional +() terms to some nonlinear expressions (#3091)\nFixed potential method ambiguities with AffExpr and QuadExpr objects (#3092)","category":"page"},{"location":"release_notes/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added vale as a linter for the documentation (#3080)\nAdded a tutorial on debugging JuMP models (#3043)\nFixed a number of typos in the documentation (#3079) (#3083)\nMany other small tweaks to the documentation (#3068) (#3073) (#3074) (#3075) (#3076) (#3077) (#3078) (#3081) (#3082) (#3084) (#3085) (#3089)","category":"page"},{"location":"release_notes/#[Version-1.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.3.0)-(September-5,-2022)","page":"Release notes","title":"Version 1.3.0 (September 5, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support slicing in SparseAxisArray (#3031)","category":"page"},{"location":"release_notes/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug introduced in v1.2.0 that prevented DenseAxisArrays with Vector keys (#3064)","category":"page"},{"location":"release_notes/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Released the JuMP logos under the CC BY 4.0 license (#3063)\nMinor tweaks to the documentation (#3054) (#3056) (#3057) (#3060) (#3061) (#3065)\nImproved code coverage of a number of files (#3048) (#3049) (#3050) (#3051) (#3052) (#3053) (#3058) (#3059)","category":"page"},{"location":"release_notes/#[Version-1.2.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.2.1)-(August-22,-2022)","page":"Release notes","title":"Version 1.2.1 (August 22, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug when parsing two-sided nonlinear constraints (#3045)","category":"page"},{"location":"release_notes/#[Version-1.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.2.0)-(August-16,-2022)","page":"Release notes","title":"Version 1.2.0 (August 16, 2022)","text":"","category":"section"},{"location":"release_notes/#Breaking-2","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it significantly refactors the internal code for handling nonlinear programs to use the MathOptInterface.Nonlinear submodule that was introduced in MathOptInterface v1.3.0. As a consequence, the internal datastructure in model.nlp_data has been removed, as has the JuMP._Derivatives submodule. Despite the changes, the public API for nonlinear programming has not changed, and any code that uses only the public API and that worked with v1.1.1 will continue to work with v1.2.0.","category":"page"},{"location":"release_notes/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added all_constraints(model; include_variable_in_set_constraints) which simplifies returning a list of all constraint indices in the model.\nAdded the ability to delete nonlinear constraints via delete(::Model, ::NonlinearConstraintRef).\nAdded the ability to provide an explicit Hessian for a multivariate user-defined function.\nAdded support for querying the primal value of a nonlinear constraint via value(::NonlinearConstraintRef)","category":"page"},{"location":"release_notes/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Containers.DenseAxisArray so that it now supports indexing with keys that hash to the same value, even if they are different types, for example, Int32 and Int64.\nFixed a bug printing the model when the solver does not support MOI.Name.","category":"page"},{"location":"release_notes/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a constraint programming formulation to the Sudoku tutorial.\nAdded newly supported solvers Pajarito, Clarabel, and COPT to the installation table.\nFixed a variety of other miscellaneous issues in the documentation.","category":"page"},{"location":"release_notes/#[Version-1.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.1.1)-(June-14,-2022)","page":"Release notes","title":"Version 1.1.1 (June 14, 2022)","text":"","category":"section"},{"location":"release_notes/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed problem displaying LaTeX in the documentation\nMinor updates to the style guide\nUpdated to MOI v1.4.0 in the documentation","category":"page"},{"location":"release_notes/#[Version-1.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.1.0)-(May-25,-2022)","page":"Release notes","title":"Version 1.1.0 (May 25, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added num_constraints(::Model; count_variable_in_set_constraints) to simplify the process of counting the number of constraints in a model\nAdded VariableRef(::ConstraintRef) for querying the variable associated with a bound or integrality constraint.\nAdded set_normalized_coefficients for modifying the variable coefficients of a vector-valued constraint.\nAdded set_string_names_on_creation to disable creating String names for variables and constraints. This can improve performance.","category":"page"},{"location":"release_notes/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug passing nothing to the start keyword of @variable","category":"page"},{"location":"release_notes/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"New tutorials:\nSensitivity analysis of a linear program\nServing web apps\nMinimal ellipse SDP tutorial refactored and improved\nDocs updated to the latest version of each package\nLots of minor fixes and improvements to the documentation","category":"page"},{"location":"release_notes/#[Version-1.0.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v1.0.0)-(March-24,-2022)","page":"Release notes","title":"Version 1.0.0 (March 24, 2022)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Read more about this release, along with an acknowledgement of all the contributors in our JuMP 1.0.0 is released blog post.","category":"page"},{"location":"release_notes/#Breaking-3","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The previously deprecated functions (v0.23.0, v0.23.1) have been removed. Deprecation was to improve consistency of function names:\nnum_nl_constraints (see num_nonlinear_constraints)\nall_nl_constraints (see all_nonlinear_constraints)\nadd_NL_expression (see add_nonlinear_expression)\nset_NL_objective  (see set_nonlinear_objective)\nadd_NL_constraint (see add_nonlinear_constraint)\nnl_expr_string (see nonlinear_expr_string)\nnl_constraint_string (see nonlinear_constraint_string)\nSymMatrixSpace (see SymmetricMatrixSpace)\nThe unintentionally exported variable JuMP.op_hint has been renamed to the unexported JuMP._OP_HINT","category":"page"},{"location":"release_notes/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .nl files\nFixed a bug broadcasting SparseAxisArrays","category":"page"},{"location":"release_notes/#[Version-0.23.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.2)-(March-14,-2022)","page":"Release notes","title":"Version 0.23.2 (March 14, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added relative_gap to solution_summary\nregister now throws an informative error if the function is not differentiable using ForwardDiff. In some cases, the check in register will encounter a false negative, and the informative error will be thrown at run-time. This usually happens when the function is non-differentiable in a subset of the domain.","category":"page"},{"location":"release_notes/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a scoping issue when extending the container keyword of containers","category":"page"},{"location":"release_notes/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Docs updated to the latest version of each package","category":"page"},{"location":"release_notes/#[Version-0.23.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.1)-(March-2,-2022)","page":"Release notes","title":"Version 0.23.1 (March 2, 2022)","text":"","category":"section"},{"location":"release_notes/#Deprecated","page":"Release notes","title":"Deprecated","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"nl_expr_string and nl_constraint_string have been renamed to nonlinear_expr_string and nonlinear_constraint_string. The old methods still exist with deprecation warnings. This change should impact very few users because to call them you must rely on private internals of the nonlinear API. Users are encouraged to use sprint(show, x) instead, where x is the nonlinear expression or constraint of interest.","category":"page"},{"location":"release_notes/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for Base.abs2(x) where x is a variable or affine expression. This is mainly useful for complex-valued constraints.","category":"page"},{"location":"release_notes/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed addition of complex and real affine expressions\nFixed arithmetic for Complex-valued quadratic expressions\nFixed variable bounds passed as Rational{Int}(Inf)\nFixed printing of the coefficient (0 + 1im)\nFixed a bug when solution_summary is called prior to optimize!","category":"page"},{"location":"release_notes/#[Version-0.23.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.23.0)-(February-25,-2022)","page":"Release notes","title":"Version 0.23.0 (February 25, 2022)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP v0.23.0 is a breaking release. It is also a release-candidate for JuMP v1.0.0. That is, if no issues are found with the v0.23.0 release, then it will be re-tagged as v1.0.0.","category":"page"},{"location":"release_notes/#Breaking-4","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nMathOptInterface has been updated to v1.0.0\nAll previously deprecated functionality has been removed\nPrintMode, REPLMode and IJuliaMode have been removed in favor of the MIME types MIME\"text/plain\" and MIME\"text/latex\". Replace instances of ::Type{REPLMode} with ::MIME\"text/plain\", REPLMode with MIME(\"text/plain\"), ::Type{IJuliaMode} with ::MIME\"text/latex\", and IJuliaMode with MIME(\"text/latex\").\nFunctions containing the nl_ acronym have been renamed to the more explicit nonlinear_. For example, num_nl_constraints is now num_nonlinear_constraints and set_NL_objective is now set_nonlinear_objective. Calls to the old functions throw an error explaining the new name.\nSymMatrixSpace has been renamed to SymmetricMatrixSpace","category":"page"},{"location":"release_notes/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added nonlinear_dual_start_value and set_nonlinear_dual_start_value\nAdded preliminary support for Complex coefficient types","category":"page"},{"location":"release_notes/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary","category":"page"},{"location":"release_notes/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"MILP examples have been migrated from GLPK to HiGHS\nFixed various typos\nImproved section on setting constraint start values","category":"page"},{"location":"release_notes/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated functionality. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Then install JuMP v0.22.3:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add JuMP@0.22.3\"","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"release_notes/#[Version-0.22.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.3)-(February-10,-2022)","page":"Release notes","title":"Version 0.22.3 (February 10, 2022)","text":"","category":"section"},{"location":"release_notes/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a reproducibility issue in the TSP tutorial\nFixed a reproducibility issue in the max_cut_sdp tutorial\nFixed a bug broadcasting an empty SparseAxisArray","category":"page"},{"location":"release_notes/#Other-25","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added a warning and improved documentation for the modify-then-query case\nFixed a typo in the docstring of RotatedSecondOrderCone\nAdded Aqua.jl as a check for code health\nAdded introductions to each section of the tutorials\nImproved the column generation and Benders decomposition tutorials\nUpdated documentation to MOI v0.10.8\nUpdated JuliaFormatter to v0.22.2","category":"page"},{"location":"release_notes/#[Version-0.22.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.2)-(January-10,-2022)","page":"Release notes","title":"Version 0.22.2 (January 10, 2022)","text":"","category":"section"},{"location":"release_notes/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The function all_nl_constraints now returns all nonlinear constraints in a model\nstart_value and set_start_value can now be used to get and set the primal start for constraint references\nPlural macros now return a tuple containing the elements that were defined instead of nothing\nAnonymous variables are now printed as _[i] where i is the index of the variable instead of noname. Calling name(x) still returns \"\" so this is non-breaking.","category":"page"},{"location":"release_notes/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed handling of min and max in nonlinear expressions\nCartesianIndex is no longer allowed as a key for DenseAxisArrays.","category":"page"},{"location":"release_notes/#Other-26","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the performance of GenericAffExpr\nAdded a tutorial on the Travelling Salesperson Problem\nAdded a tutorial on querying the Hessian of a nonlinear program\nAdded documentation on using custom solver binaries.","category":"page"},{"location":"release_notes/#[Version-0.22.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.1)-(November-29,-2021)","page":"Release notes","title":"Version 0.22.1 (November 29, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Export OptimizationSense enum, with instances: MIN_SENSE, MAX_SENSE, and FEASIBILITY_SENSE\nAdd Base.isempty(::Model) to match Base.empty(::Model)","category":"page"},{"location":"release_notes/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in container with tuples as indices\nFix bug in set_time_limit_sec","category":"page"},{"location":"release_notes/#Other-27","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add tutorial \"Design patterns for larger models\"\nRemove release notes section from PDF\nGeneral edits of the documentation and error messages","category":"page"},{"location":"release_notes/#[Version-0.22.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.22.0)-(November-10,-2021)","page":"Release notes","title":"Version 0.22.0 (November 10, 2021)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP v0.22 is a breaking release","category":"page"},{"location":"release_notes/#Breaking-5","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP 0.22 contains a number of breaking changes. However, these should be invisible for the majority of users. You will mostly encounter these breaking changes if you: wrote a JuMP extension, accessed backend(model), or called @SDconstraint.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The breaking changes are as follows:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"MathOptInterface has been updated to v0.10.4. For users who have interacted with the MOI backend, this contains a large number of breaking changes. Read the MathOptInterface release notes for more details.\nThe bridge_constraints keyword argument to Model and set_optimizer has been renamed add_bridges to reflect that more thing were bridged than just constraints.\nThe backend(model) field now contains a concrete instance of a MOI.Utilities.CachingOptimizer instead of one with an abstractly typed optimizer field. In most cases, this will lead to improved performance. However, calling set_optimizer after backend invalidates the old backend. For example:\nmodel = Model()\nb = backend(model)\nset_optimizer(model, GLPK.Optimizer)\n@variable(model, x)\n# b is not updated with `x`! Get a new b by calling `backend` again.\nnew_b = backend(model)\nAll usages of @SDconstraint are deprecated. The new syntax is @constraint(model, X >= Y, PSDCone()).\nCreating a DenseAxisArray with a Number as an axis will now display a warning. This catches a common error in which users write @variable(model, x[length(S)]) instead of @variable(model, x[1:length(S)]).\nThe caching_mode argument to Model, for example, Model(caching_mode = MOIU.MANUAL) mode has been removed. For more control over the optimizer, use direct_model instead.\nThe previously deprecated lp_objective_perturbation_range and lp_rhs_perturbation_range functions have been removed. Use lp_sensitivity_report instead.\nThe .m fields of NonlinearExpression and NonlinearParameter have been renamed to .model.\nInfinite variable bounds are now ignored. Thus, @variable(model, x <= Inf) will show has_upper_bound(x) == false. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.\nThe variable_type and constraint_type functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.\nThe internal functions moi_mode, moi_bridge_constraints, moi_add_constraint, and moi_add_to_function_constant are no longer exported.\nThe un-used method Containers.generate_container has been deleted.\nThe Containers API has been refactored, and _build_ref_sets is now public as Containers.build_ref_sets.\nThe parse_constraint_ methods for extending @constraint at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.","category":"page"},{"location":"release_notes/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The TerminationStatusCode and ResultStatusCode enums are now exported by JuMP. Prefer termination_status(model) == OPTIMAL instead of == MOI.OPTIMAL, although the MOI. prefix way still works.\nCopy a x::DenseAxisArray to an Array by calling Array(x).\nNonlinearExpression is now a subtype of AbstractJuMPScalar\nConstraints such as @constraint(model, x + 1 in MOI.Integer()) are now supported.\nprimal_feasibility_report now accepts a function as the first argument.\nScalar variables @variable(model, x[1:2] in MOI.Integer()) creates two variables, both of which are constrained to be in the set MOI.Integer.\nConic constraints can now be specified as inequalities under a different partial ordering. So @constraint(model, x - y in MOI.Nonnegatives()) can now be written as @constraint(model, x >= y, MOI.Nonnegatives()).\nNames are now set for vectorized constraints.","category":"page"},{"location":"release_notes/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue when show was called on a SparseAxisArray with a large number of elements.\nFixed a bug displaying barrier and simplex iterations in solution_summary.\nFixed a bug by implementing hash for DenseAxisArray and SparseAxisArray.\nNames are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with direct_model.\nMutableArithmetics.Zero is converted into a 0.0 before being returned to the user. Previously, some calls to @expression would return the undocumented MutableArithmetics.Zero() object. One example is summing over an empty set @expression(model, sum(x[i] for i in 1:0)). You will now get 0.0 instead.\nAffExpr and QuadExpr can now be used with == 0 instead of iszero. This fixes a number of issues relating to Julia standard libraries such as LinearAlgebra and SparseArrays.\nFixed a bug when registering a user-defined function with splatting.","category":"page"},{"location":"release_notes/#Other-28","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The documentation is now available as a PDF.\nThe documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)\nThe documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.\nThe documentation is now built with Julia 1.6 instead of 1.0.\nVarious error messages have been improved to be more readable.","category":"page"},{"location":"release_notes/#[Version-0.21.10](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.10)-(September-4,-2021)","page":"Release notes","title":"Version 0.21.10 (September 4, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added add_NL_expression\nadd_NL_xxx functions now support AffExpr and QuadExpr as terms","category":"page"},{"location":"release_notes/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary\nFixed a bug in relax_integrality","category":"page"},{"location":"release_notes/#Other-29","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved error message in lp_sensitivity_report","category":"page"},{"location":"release_notes/#[Version-0.21.9](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.9)-(August-1,-2021)","page":"Release notes","title":"Version 0.21.9 (August 1, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Containers now support arbitrary container types by passing the type to the container keyword and overloading Containers.container.\nis_valid now supports nonlinear constraints\nAdded unsafe_backend for querying the inner-most optimizer of a JuMP model.\nNonlinear parameters now support the plural @NLparameters macro.\nContainers (for example, DenseAxisArray) can now be used in vector-valued constraints.","category":"page"},{"location":"release_notes/#Other-30","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements to the documentation.","category":"page"},{"location":"release_notes/#[Version-0.21.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.8)-(May-8,-2021)","page":"Release notes","title":"Version 0.21.8 (May 8, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The @constraint macro is now extendable in the same way as @variable.\nAffExpr and QuadExpr can now be used in nonlinear macros.","category":"page"},{"location":"release_notes/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report.\nFixed an inference issue when creating empty SparseAxisArrays.","category":"page"},{"location":"release_notes/#[Version-0.21.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.7)-(April-12,-2021)","page":"Release notes","title":"Version 0.21.7 (April 12, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-26","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added primal_feasibility_report, which can be used to check whether a primal point satisfies primal feasibility.\nAdded coefficient, which returns the coefficient associated with a variable in affine and quadratic expressions.\nAdded copy_conflict, which returns the IIS of an infeasible model.\nAdded solution_summary, which returns (and prints) a struct containing a summary of the solution.\nAllow AbstractVector in vector constraints instead of just Vector.\nAdded latex_formulation(model) which returns an object representing the latex formulation of a model. Use print(latex_formulation(model)) to print the formulation as a string.\nUser-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.\nDenseAxisArray's now support broadcasting over multiple arrays.\nContainer indices can now be iterators of Base.SizeUnknown.","category":"page"},{"location":"release_notes/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug in rad2deg and deg2rad in nonlinear expressions.\nFixed a MethodError bug in Containers when forcing container type.\nAllow partial slicing of a DenseAxisArray, resolving an issue from 2014.\nFixed a bug printing variable names in IJulia.\nEnding an IJulia cell with model now prints a summary of the model (like in the REPL) not the latex formulation. Use print(model) to print the latex formulation.\nFixed a bug when copying models containing nested arrays.","category":"page"},{"location":"release_notes/#Other-31","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Tutorials are now part of the documentation, and more refactoring has taken place.\nAdded JuliaFormatter added as a code formatter.\nAdded some precompilation statements to reduce initial latency.\nVarious improvements to error messages to make them more helpful.\nImproved performance of value(::NonlinearExpression).\nImproved performance of fix(::VariableRef).","category":"page"},{"location":"release_notes/#[Version-0.21.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.6)-(January-29,-2021)","page":"Release notes","title":"Version 0.21.6 (January 29, 2021)","text":"","category":"section"},{"location":"release_notes/#Added-27","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for skew symmetric variables via @variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace()).\nlp_sensitivity_report has been added which significantly improves the performance of querying the sensitivity summary of an LP. lp_objective_perturbation_range and lp_rhs_perturbation_range are deprecated.\nDual warm-starts are now supported with set_dual_start_value and dual_start_value.\n∈ (\\in<tab>) can now be used in macros instead of = or in.\nUse haskey(model::Model, key::Symbol) to check if a name key is registered in a model.\nAdded unregister(model::Model, key::Symbol) to unregister a name key from model.\nAdded callback_node_status for use in callbacks.\nAdded print_bridge_graph to visualize the bridging graph generated by MathOptInterface.\nImproved error message for containers with duplicate indices.","category":"page"},{"location":"release_notes/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes to pass tests on Julia 1.6.\nFixed a bug in the printing of nonlinear expressions in IJulia.\nFixed a bug when nonlinear expressions are passed to user-defined functions.\nSome internal functions that were previously exported are now no longer exported.\nFixed a bug when relaxing a fixed binary variable.\nFixed a StackOverflowError that occurred when SparseAxisArrays had a large number of elements.\nRemoved an unnecessary type assertion in list_of_constraint_types.\nFixed a bug when copying models with registered expressions.","category":"page"},{"location":"release_notes/#Other-32","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in /examples have now been moved to /docs/src/examples and rewritten using Literate.jl, and they are now included in the documentation.\nJuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.\nThe root cause of a large number of method invalidations has been resolved.\nWe switched continuous integration from Travis and Appveyor to GitHub Actions.","category":"page"},{"location":"release_notes/#[Version-0.21.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.5)-(September-18,-2020)","page":"Release notes","title":"Version 0.21.5 (September 18, 2020)","text":"","category":"section"},{"location":"release_notes/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed deprecation warnings\nThrow DimensionMismatch for incompatibly sized functions and sets\nUnify treatment of keys(x) on JuMP containers","category":"page"},{"location":"release_notes/#[Version-0.21.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.4)-(September-14,-2020)","page":"Release notes","title":"Version 0.21.4 (September 14, 2020)","text":"","category":"section"},{"location":"release_notes/#Added-28","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add debug info when adding unsupported constraints\nAdd relax_integrality for solving continuous relaxation\nAllow querying constraint conflicts","category":"page"},{"location":"release_notes/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Dispatch on Real for MOI.submit\nImplement copy for CustomSet in tests\nDon't export private macros\nFix invalid assertion in nonlinear\nError if constraint has NaN right-hand side\nImprove speed of tests\nLots of work modularizing files in /test\nImprove line numbers in macro error messages\nPrint nonlinear subexpressions\nVarious documentation updates\nDependency updates:\nDatastructures 0.18\nMathOptFormat v0.5\nPrep for MathOptInterface 0.9.15","category":"page"},{"location":"release_notes/#[Version-0.21.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.3)-(June-18,-2020)","page":"Release notes","title":"Version 0.21.3 (June 18, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (#2212).\nAdded functions simplex_iterations, barrier_iterations and node_count (#2201).\nAdded function reduced_cost (#2205).\nImplemented callback_value for affine and quadratic expressions (#2231).\nSupport MutableArithmetics.Zero in objective and constraints (#2219).\nDocumentation improvements:\nMention tutorials in the docs (#2223).\nUpdate COIN-OR links (#2242).\nExplicit link to the documentation of MOI.FileFormats (#2253).\nTypo fixes (#2261).\nContainers improvements:\nFix Base.map for DenseAxisArray (#2235).\nThrow BoundsError if number of indices is incorrect for DenseAxisArray and SparseAxisArray (#2240).\nExtensibility improvements:\nImplement a set_objective method fallback that redirects to set_objective_sense and set_objective_function (#2247).\nAdd parse_constraint method with arbitrary number of arguments (#2051).\nAdd parse_constraint_expr and parse_constraint_head (#2228).","category":"page"},{"location":"release_notes/#[Version-0.21.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.2)-(April-2,-2020)","page":"Release notes","title":"Version 0.21.2 (April 2, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added relative_gap() to access MOI.RelativeGap() attribute (#2199).\nDocumentation fixes:\nAdded link to source for docstrings in the documentation (#2207).\nAdded docstring for @variables macro (#2216).\nTypo fixes (#2177, #2184, #2182).\nImplementation of methods for Base functions:\nImplemented Base.empty! for JuMP.Model (#2198).\nImplemented Base.conj for JuMP scalar types (#2209).","category":"page"},{"location":"release_notes/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixed sum of expression with scalar product in macro (#2178).\nFixed writing of nonlinear models to MathOptFormat (#2181).\nFixed construction of empty SparseAxisArray (#2179).\nFixed constraint with zero function (#2188).","category":"page"},{"location":"release_notes/#[Version-0.21.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.1)-(Feb-18,-2020)","page":"Release notes","title":"Version 0.21.1 (Feb 18, 2020)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved the clarity of the with_optimizer deprecation warning.","category":"page"},{"location":"release_notes/#[Version-0.21.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.21.0)-(Feb-16,-2020)","page":"Release notes","title":"Version 0.21.0 (Feb 16, 2020)","text":"","category":"section"},{"location":"release_notes/#Breaking-6","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Deprecated with_optimizer (#2090, #2084, #2141). You can replace with_optimizer by either nothing, optimizer_with_attributes or a closure:\nreplace with_optimizer(Ipopt.Optimizer) by Ipopt.Optimizer.\nreplace with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0) by optimizer_with_attributes(Ipopt.Optimizer, \"max_cpu_time\" => 60.0).\nreplace with_optimizer(Gurobi.Optimizer, env) by () -> Gurobi.Optimizer(env).\nreplace with_optimizer(Gurobi.Optimizer, env, Presolve=0) by optimizer_with_attributes(() -> Gurobi.Optimizer(env), \"Presolve\" => 0).\nalternatively to optimizer_with_attributes, you can also set the attributes separately with set_optimizer_attribute.\nRenamed set_parameter and set_parameters to set_optimizer_attribute and set_optimizer_attributes (#2150).\nBroadcast should now be explicit inside macros. @SDconstraint(model, x >= 1) and @constraint(model, x + 1 in SecondOrderCone()) now throw an error instead of broadcasting 1 along the dimension of x (#2107).\n@SDconstraint(model, x >= 0) is now equivalent to @constraint(model, x in PSDCone()) instead of @constraint(model, (x .- 0) in PSDCone()) (#2107).\nThe macros now create the containers with map instead of for loops, as a consequence, containers created by @expression can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where @expression could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (#2070).\nCalling deepcopy(::AbstractModel) now throws an error.\nThe constraint name is now printed in the model string (#2108).","category":"page"},{"location":"release_notes/#Added-29","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Added support for solver-independent and solver-specific callbacks (#2101).\nAdded write_to_file and read_from_file, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (#2114).\nAdded support for complementarity constraints (#2132).\nAdded support for indicator constraints (#2092).\nAdded support for querying multiple solutions with the result keyword (#2100).\nAdded support for constraining variables on creation (#2128).\nAdded method delete that deletes a vector of variables at once if it is supported by the underlying solver (#2135).\nThe arithmetic between JuMP expression has be refactored into the MutableArithmetics package (#2107).\nImproved error on complex values in NLP (#1978).\nAdded an example of column generation (#2010).","category":"page"},{"location":"release_notes/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Incorrect coefficients generated when using Symmetric variables (#2102)","category":"page"},{"location":"release_notes/#[Version-0.20.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.1)-(Oct-18,-2019)","page":"Release notes","title":"Version 0.20.1 (Oct 18, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add sections on @variables and @constraints in the documentation (#2062).\nFixed product of sparse matrices for Julia v1.3 (#2063).\nAdded set_objective_coefficient to modify the coefficient of a linear term of the objective function (#2008).\nAdded set_time_limit_sec, unset_time_limit_sec and time_limit_sec to set and query the time limit for the solver in seconds (#2053).","category":"page"},{"location":"release_notes/#[Version-0.20.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.20.0)-(Aug-24,-2019)","page":"Release notes","title":"Version 0.20.0 (Aug 24, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Documentation updates.\nNumerous bug fixes.\nBetter error messages (#1977, #1978, #1997, #2017).\nPerformance improvements (#1947, #2032).\nAdded LP sensitivity summary functions lp_objective_perturbation_range and lp_rhs_perturbation_range (#1917).\nAdded functions dual_objective_value, raw_status and set_parameter.\nAdded function set_objective_coefficient to modify the coefficient of a linear term of the objective (#2008).\nAdded functions set_normalized_rhs, normalized_rhs, and add_to_function_constant to modify and get the constant part of a constraint (#1935, #1960).\nAdded functions set_normalized_coefficient and normalized_coefficient to modify and get the coefficient of a linear term of a constraint (#1935, #1960).\nNumerous other improvements in MOI 0.9, see the NEWS.md file of MOI for more details.","category":"page"},{"location":"release_notes/#[Version-0.19.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.2)-(June-8,-2019)","page":"Release notes","title":"Version 0.19.2 (June 8, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.","category":"page"},{"location":"release_notes/#[Version-0.19.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.1)-(May-12,-2019)","page":"Release notes","title":"Version 0.19.1 (May 12, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Usability and performance improvements.\nBug fixes.","category":"page"},{"location":"release_notes/#[Version-0.19.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.19.0)-(February-15,-2019)","page":"Release notes","title":"Version 0.19.0 (February 15, 2019)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP 0.19 contains significant breaking changes.","category":"page"},{"location":"release_notes/#Breaking-7","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"JuMP's abstraction layer for communicating with solvers changed from MathProgBase (MPB) to MathOptInterface (MOI). MOI addresses many longstanding design issues. (See @mlubin's slides from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the installation guide for a list of solvers that have and have not yet been updated.\nMost solvers have been renamed to PackageName.Optimizer. For example, GurobiSolver() is now Gurobi.Optimizer.\nSolvers are no longer added to a model via Model(solver = XXX(kwargs...)). Instead use Model(with_optimizer(XXX, kwargs...)). For example, Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0)).\nJuMP containers (for example, the objects returned by @variable) have been redesigned. Containers.SparseAxisArray replaces JuMPDict, JuMPArray was rewritten (inspired by AxisArrays) and renamed Containers.DenseAxisArray, and you can now request a container type with the container= keyword to the macros. See the corresponding documentation for more details.\nThe statuses returned by solvers have changed. See the possible status values here. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).\nStarting values are separate from result values. Use value to query the value of a variable in a solution. Use start_value and set_start_value to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.\nThe data structures for affine and quadratic expressions AffExpr and QuadExpr have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.\nJuMPNLPEvaluator no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.\nJuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)\nThe norm() syntax is no longer recognized inside macros. Use the SecondOrderCone() set instead.\nJuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.\nThe symbols :Min and :Max are no longer used as optimization senses. Instead, JuMP uses the OptimizationSense enum from MathOptInterface. @objective(model, Max, ...), @objective(model, Min, ...), @NLobjective(model, Max, ...), and @objective(model, Min, ...) remain valid, but @objective(m, :Max, ...) is no longer accepted.\nThe sign conventions for duals has changed in some cases for consistency with conic duality (see the documentation). The shadow_price helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.\n@constraintref is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,\nconstraints = Dict() # Optionally, specify types for improved performance.\nfor i in 1:N\n  constraints[i] = @constraint(model, ...)\nend\nThe lowerbound, upperbound, and basename keyword arguments to the @variable macro have been renamed to lower_bound, upper_bound, and base_name, for consistency with JuMP's new style recommendations.\nWe rely on broadcasting syntax to apply accessors to collections of variables, for example, value.(x) instead of getvalue(x) for collections. (Use value(x) when x is a scalar object.)","category":"page"},{"location":"release_notes/#Added-30","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Splatting (like f(x...)) is recognized in restricted settings in nonlinear expressions.\nSupport for deleting constraints and variables.\nThe documentation has been completely rewritten using docstrings and Documenter.\nSupport for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).\nSignificantly improved support for modeling new types of constraints and for extending JuMP's macros.\nSupport for providing dual warm starts.\nImproved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).\nExplicit control of whether symmetry-enforcing constraints are added to PSD constraints.\nSupport for modeling exponential cones.\nSignificant improvements in internal code quality and testing.\nStyle and naming guidelines.\nDirect mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding documentation.","category":"page"},{"location":"release_notes/#Regressions","page":"Release notes","title":"Regressions","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Performance regressions in model generation (issue). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.\nFast incremental NLP solves are not yet reimplemented (issue).\nWe do not yet have an implementation of solver-specific callbacks.\nThe column generation syntax in @variable has been removed (that is, the objective, coefficients, and inconstraints keyword arguments). Support for column generation will be re-introduced in a future release.\nThe ability to solve the continuous relaxation (that is, via solve(model; relaxation = true)) is not yet reimplemented (issue).","category":"page"},{"location":"release_notes/#[Version-0.18.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.5)-(December-1,-2018)","page":"Release notes","title":"Version 0.18.5 (December 1, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support views in some derivative evaluation functions.\nImproved compatibility with PackageCompiler.","category":"page"},{"location":"release_notes/#[Version-0.18.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.4)-(October-8,-2018)","page":"Release notes","title":"Version 0.18.4 (October 8, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in model printing on Julia 0.7 and 1.0.","category":"page"},{"location":"release_notes/#[Version-0.18.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.3)-(October-1,-2018)","page":"Release notes","title":"Version 0.18.3 (October 1, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for Julia v1.0 (Thanks @ExpandingMan)\nFix matrix expressions with quadratic functions (#1508)","category":"page"},{"location":"release_notes/#[Version-0.18.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.2)-(June-10,-2018)","page":"Release notes","title":"Version 0.18.2 (June 10, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in second-order derivatives when expressions are present (#1319)\nFix a bug in @constraintref (#1330)","category":"page"},{"location":"release_notes/#[Version-0.18.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.1)-(April-9,-2018)","page":"Release notes","title":"Version 0.18.1 (April 9, 2018)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for nested tuple destructuring (#1193)\nPreserve internal model when relaxation=true (#1209)\nMinor bug fixes and updates for example","category":"page"},{"location":"release_notes/#[Version-0.18.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.18.0)-(July-27,-2017)","page":"Release notes","title":"Version 0.18.0 (July 27, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Drop support for Julia 0.5.\nUpdate for ForwardDiff 0.5.\nMinor bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.17.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.1)-(June-9,-2017)","page":"Release notes","title":"Version 0.17.1 (June 9, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Use of constructconstraint! in @SDconstraint.\nMinor bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.17.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.17.0)-(May-27,-2017)","page":"Release notes","title":"Version 0.17.0 (May 27, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: Mixing quadratic and conic constraints is no longer supported.\nBreaking change: The getvariable and getconstraint functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name x, one should now write m[:x] instead of getvariable(m, :x). As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.\nAddition of the getobjectivebound function that mirrors the functionality of the MathProgBase getobjbound function except that it takes into account transformations performed by JuMP.\nMinor bug fixes.","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The following changes are primarily of interest to developers of JuMP extensions:","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The new syntax @constraint(model, expr in Cone) creates the constraint ensuring that expr is inside Cone. The Cone argument is passed to constructconstraint! which enables the call to the dispatched to an extension.\nThe @variable macro now calls constructvariable! instead of directly calling the Variable constructor. Extra arguments and keyword arguments passed to @variable are passed to constructvariable! which enables the call to be dispatched to an extension.\nRefactor the internal function conicdata (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.","category":"page"},{"location":"release_notes/#[Version-0.16.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.2)-(March-28,-2017)","page":"Release notes","title":"Version 0.16.2 (March 28, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Minor bug fixes and printing tweaks\nAddress deprecation warnings for Julia 0.6","category":"page"},{"location":"release_notes/#[Version-0.16.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.1)-(March-7,-2017)","page":"Release notes","title":"Version 0.16.1 (March 7, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Better support for AbstractArray in JuMP (Thanks @tkoolen)\nMinor bug fixes","category":"page"},{"location":"release_notes/#[Version-0.16.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.16.0)-(February-23,-2017)","page":"Release notes","title":"Version 0.16.0 (February 23, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling solve() will result in an error.\nBreaking change: User-defined functions are no longer global. The first argument to JuMP.register is now a JuMP Model object within whose scope the function will be registered. Calling JuMP.register without a Model now produces an error.\nBreaking change: Use the new JuMP.fix method to fix a variable to a value or to update the value to which a variable is fixed. Calling setvalue on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)\nNonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)\nNew category keyword to @variable. Used for specifying categories of anonymous variables.\nCompatibility with Julia 0.6-dev.\nMinor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)","category":"page"},{"location":"release_notes/#[Version-0.15.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.1)-(January-31,-2017)","page":"Release notes","title":"Version 0.15.1 (January 31, 2017)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bugfix for @LinearConstraints and friends","category":"page"},{"location":"release_notes/#[Version-0.15.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.15.0)-(December-22,-2016)","page":"Release notes","title":"Version 0.15.0 (December 22, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.5.0 is the minimum required version for this release.\nDocument support for BARON solver\nEnable info callbacks in more states than before, for example, for recording solutions. New when argument to addinfocallback (#814, thanks @yeesian)\nImproved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:\nWhen multiple variables in a model are given the same name\nWhen non-symbols are used as names, for example, @variable(m, x[1][1:N])\nImprovements in iterating over JuMP containers (#836, thanks @IssamT)\nSupport for writing variable names in .lp file output (Thanks @leethargo)\nSupport for querying duals to SDP problems (Thanks @blegat)\nThe comprehension syntax with curly braces sum{}, prod{}, and norm2{} has been deprecated in favor of Julia's native comprehension syntax sum(), prod() and norm() as previously announced. (For early adopters of the new syntax, norm2() was renamed to norm() without deprecation.)\nUnit tests rewritten to use Base.Test instead of FactCheck\nImproved support for operations with matrices of JuMP types (Thanks @ExpandingMan)\nThe syntax to halt a solver from inside a callback has changed from throw(CallbackAbort()) to return JuMP.StopTheSolver\nMinor bug fixes","category":"page"},{"location":"release_notes/#[Version-0.14.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.2)-(December-12,-2016)","page":"Release notes","title":"Version 0.14.2 (December 12, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Allow singleton anonymous variables (includes bugfix)","category":"page"},{"location":"release_notes/#[Version-0.14.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.1)-(September-12,-2016)","page":"Release notes","title":"Version 0.14.1 (September 12, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"More consistent handling of states in informational callbacks, includes a new when parameter to addinfocallback for specifying in which state an informational callback should be called.","category":"page"},{"location":"release_notes/#[Version-0.14.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.14.0)-(August-7,-2016)","page":"Release notes","title":"Version 0.14.0 (August 7, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility with Julia 0.5 and ForwardDiff 0.2\nSupport for \"anonymous\" variables, constraints, expressions, and parameters, for example, x = @variable(m, [1:N]) instead of @variable(m, x[1:N])\nSupport for retrieving constraints from a model by name via getconstraint\n@NLconstraint now returns constraint references (as expected).\nSupport for vectorized expressions within lazy constraints\nOn Julia 0.5, parse new comprehension syntax sum(x[i] for i in 1:N if isodd(i)) instead of sum{ x[i], i in 1:N; isodd(i) }. The old syntax with curly braces will be deprecated in JuMP 0.15.\nNow possible to provide nonlinear expressions as \"raw\" Julia Expr objects instead of using JuMP's nonlinear macros. This input format is useful for programmatically generated expressions.\ns/Mathematical Programming/Mathematical Optimization/\nSupport for local cuts (Thanks to @madanim, Mehdi Madani)\nDocument Xpress interface developed by @joaquimg, Joaquim Dias Garcia\nMinor bug and deprecation fixes (Thanks @odow, @jrevels)","category":"page"},{"location":"release_notes/#[Version-0.13.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.2)-(May-16,-2016)","page":"Release notes","title":"Version 0.13.2 (May 16, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility update for MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.13.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.1)-(May-3,-2016)","page":"Release notes","title":"Version 0.13.1 (May 3, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix broken deprecation for registerNLfunction.","category":"page"},{"location":"release_notes/#[Version-0.13.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.13.0)-(April-29,-2016)","page":"Release notes","title":"Version 0.13.0 (April 29, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Most exported methods and macros have been renamed to avoid camelCase. See the list of changes here. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.\nSpecify variable lower/upper bounds in @variable using the lowerbound and upperbound keyword arguments.\nChange name printed for variable using the basename keyword argument to @variable.\nNew @variables macro allows multi-line declaration of groups of variables.\nA number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was recorded live.\nCompatibility fixes with Julia 0.5.\nThe \"end\" indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see here.","category":"page"},{"location":"release_notes/#[Version-0.12.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.2)-(March-9,-2016)","page":"Release notes","title":"Version 0.12.2 (March 9, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Small fixes for nonlinear optimization","category":"page"},{"location":"release_notes/#[Version-0.12.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.1)-(March-1,-2016)","page":"Release notes","title":"Version 0.12.1 (March 1, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)","category":"page"},{"location":"release_notes/#[Version-0.12.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.12.0)-(February-27,-2016)","page":"Release notes","title":"Version 0.12.0 (February 27, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:\n@defExpr and @defNLExpr now take the model as the first argument. The previous one-argument version of @defExpr is deprecated; all expressions should be named. For example, replace @defExpr(2x+y) with @defExpr(jump_model, my_expr, 2x+y).\nJuMP no longer uses Julia's variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.\nSupport for user-defined functions integrated within nonlinear JuMP expressions.\nReplaced iteration over AffExpr with Number-like scalar iteration; previous iteration behavior is now available via linearterms(::AffExpr).\nStopping the solver via throw(CallbackAbort()) from a callback no longer triggers an exception. Instead, solve() returns UserLimit status.\ngetDual() now works for conic problems (Thanks @emreyamangil.)","category":"page"},{"location":"release_notes/#[Version-0.11.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.3)-(February-4,-2016)","page":"Release notes","title":"Version 0.11.3 (February 4, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bug-fix for problems with quadratic objectives and semidefinite constraints","category":"page"},{"location":"release_notes/#[Version-0.11.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.2)-(January-14,-2016)","page":"Release notes","title":"Version 0.11.2 (January 14, 2016)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Compatibility update for Mosek","category":"page"},{"location":"release_notes/#[Version-0.11.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.1)-(December-1,-2015)","page":"Release notes","title":"Version 0.11.1 (December 1, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Remove usage of @compat in tests.\nFix updating quadratic objectives for nonlinear models.","category":"page"},{"location":"release_notes/#[Version-0.11.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.11.0)-(November-30,-2015)","page":"Release notes","title":"Version 0.11.0 (November 30, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.4.0 is the minimum required version for this release.\nFix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.\nAddition of the solve(m::Model, relaxation=true) keyword argument to solve the standard continuous relaxation of model m\nThe getConstraintBounds() method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.10.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.3)-(November-20,-2015)","page":"Release notes","title":"Version 0.10.3 (November 20, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a rare error when parsing quadratic expressions\nFix Variable() constructor with default arguments\nDetect unrecognized keywords in solve()","category":"page"},{"location":"release_notes/#[Version-0.10.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.2)-(September-28,-2015)","page":"Release notes","title":"Version 0.10.2 (September 28, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for deprecation warnings","category":"page"},{"location":"release_notes/#[Version-0.10.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.1)-(September-3,-2015)","page":"Release notes","title":"Version 0.10.1 (September 3, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixes for ambiguity warnings.\nFix for breaking change in precompilation syntax in Julia 0.4-pre","category":"page"},{"location":"release_notes/#[Version-0.10.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.10.0)-(August-31,-2015)","page":"Release notes","title":"Version 0.10.0 (August 31, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support (on Julia 0.4 and later) for conditions in indexing @defVar and @addConstraint constructs, for example, @defVar(m, x[i=1:5,j=1:5; i+j >= 3])\nSupport for vectorized operations on Variables and expressions. See the documentation for details.\nNew getVar() method to access variables in a model by name\nSupport for semidefinite programming.\nDual solutions are now available for general nonlinear problems. You may call getDual on a reference object for a nonlinear constraint, and getDual on a variable object for Lagrange multipliers from active bounds.\nIntroduce warnings for two common performance traps: too many calls to getValue() on a collection of variables and use of the + operator in a loop to sum expressions.\nSecond-order cone constraints can be written directly with the norm() and norm2{} syntax.\nImplement MathProgBase interface for querying Hessian-vector products.\nIteration over JuMPContainers is deprecated; instead, use the keys and values functions, and zip(keys(d),values(d)) for the old behavior.\n@defVar returns Array{Variable,N} when each of N index sets are of the form 1:nᵢ.\nModule precompilation: on Julia 0.4 and later, using JuMP is now much faster.","category":"page"},{"location":"release_notes/#[Version-0.9.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.3)-(August-11,-2015)","page":"Release notes","title":"Version 0.9.3 (August 11, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fixes for FactCheck testing on julia v0.4.","category":"page"},{"location":"release_notes/#[Version-0.9.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.2)-(June-27,-2015)","page":"Release notes","title":"Version 0.9.2 (June 27, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in @addConstraints.","category":"page"},{"location":"release_notes/#[Version-0.9.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.1)-(April-25,-2015)","page":"Release notes","title":"Version 0.9.1 (April 25, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix for Julia 0.4-dev.\nSmall infrastructure improvements for extensions.","category":"page"},{"location":"release_notes/#[Version-0.9.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.9.0)-(April-18,-2015)","page":"Release notes","title":"Version 0.9.0 (April 18, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Comparison operators for constructing constraints (for example, 2x >= 1) have been deprecated. Instead, construct the constraints explicitly in the @addConstraint macro to add them to the model, or in the @LinearConstraint macro to create a stand-alone linear constraint instance.\ngetValue() method implemented to compute the value of a nonlinear subexpression\nJuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.\nA number of performance improvements in ReverseDiffSparse for computing derivatives.\nMathProgBase.getsolvetime(m) now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)\nFormatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).","category":"page"},{"location":"release_notes/#[Version-0.8.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.8.0)-(February-17,-2015)","page":"Release notes","title":"Version 0.8.0 (February 17, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Nonlinear subexpressions now supported with the @defNLExpr macro.\nSCS supported for solving second-order conic problems.\nsetXXXCallback family deprecated in favor of addXXXCallback.\nMultiple callbacks of the same type can be registered.\nAdded support for informational callbacks via addInfoCallback.\nA CallbackAbort exception can be thrown from callback to safely exit optimization.","category":"page"},{"location":"release_notes/#[Version-0.7.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.4)-(February-4,-2015)","page":"Release notes","title":"Version 0.7.4 (February 4, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.\nTwo-sided nonlinear constraints are supported.\nMethods for accessing the number of variables and constraints in a model are renamed.\nNew default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.\nSmall bug fixes.","category":"page"},{"location":"release_notes/#[Version-0.7.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.3)-(January-14,-2015)","page":"Release notes","title":"Version 0.7.3 (January 14, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a method ambiguity conflict with Compose.jl (cosmetic fix)","category":"page"},{"location":"release_notes/#[Version-0.7.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.2)-(January-9,-2015)","page":"Release notes","title":"Version 0.7.2 (January 9, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in sum(::JuMPDict)\nAdded the setCategory function to change a variables category (for example, continuous or binary)","category":"page"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"after construction, and getCategory to retrieve the variable category.","category":"page"},{"location":"release_notes/#[Version-0.7.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.1)-(January-2,-2015)","page":"Release notes","title":"Version 0.7.1 (January 2, 2015)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.","category":"page"},{"location":"release_notes/#[Version-0.7.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.7.0)-(December-29,-2014)","page":"Release notes","title":"Version 0.7.0 (December 29, 2014)","text":"","category":"section"},{"location":"release_notes/#Linear/quadratic/conic-programming","page":"Release notes","title":"Linear/quadratic/conic programming","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Breaking change: The syntax for column-wise model generation has been changed to use keyword arguments in @defVar.\nOn Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.\nECOS supported for solving second-order conic problems.","category":"page"},{"location":"release_notes/#_nonlinear_programming_release_notes","page":"Release notes","title":"Nonlinear programming","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for skipping model generation when solving a sequence of nonlinear models with changing data.\nFix a memory leak when solving a sequence of nonlinear models.\nThe @addNLConstraint macro now supports the three-argument version to define sets of nonlinear constraints.\nKNITRO supported as a nonlinear solver.\nSpeed improvements for model generation.\nThe @addNLConstraints macro supports adding multiple (groups of) constraints at once. Syntax is similar to @addConstraints.\nDiscrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).","category":"page"},{"location":"release_notes/#General","page":"Release notes","title":"General","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Starting values for variables may now be specified with @defVar(m, x, start=value).\nThe setSolver function allows users to change the solver subsequent to model creation.\nSupport for \"fixed\" variables via the @defVar(m, x == 1) syntax.\nUnit tests rewritten to use FactCheck.jl, improved testing across solvers.","category":"page"},{"location":"release_notes/#[Version-0.6.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.3)-(October-19,-2014)","page":"Release notes","title":"Version 0.6.3 (October 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in multiplying two AffExpr objects.","category":"page"},{"location":"release_notes/#[Version-0.6.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.2)-(October-11,-2014)","page":"Release notes","title":"Version 0.6.2 (October 11, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Further improvements and bug fixes for printing.\nFixed a bug in @defExpr.\nSupport for accessing expression graphs through the MathProgBase NLP interface.","category":"page"},{"location":"release_notes/#[Version-0.6.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.1)-(September-19,-2014)","page":"Release notes","title":"Version 0.6.1 (September 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improvements and bug fixes for printing.","category":"page"},{"location":"release_notes/#[Version-0.6.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.6.0)-(September-9,-2014)","page":"Release notes","title":"Version 0.6.0 (September 9, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Julia 0.3.0 is the minimum required version for this release.\nbuildInternalModel(m::Model) added to build solver-level model in memory without optimizing.\nDeprecate load_model_only keyword argument to solve.\nAdd groups of constraints with @addConstraints macro.\nUnicode operators now supported, including ∑ for sum, ∏ for prod, and ≤/≥\nQuadratic constraints supported in @addConstraint macro.\nQuadratic objectives supported in @setObjective macro.\nMathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems\nBreaking change: IpoptOptions no longer supported to specify solver options, use m = Model(solver=IpoptSolver(options...)) instead.\nNew solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK\nNew option to control whether the lazy constraint callback is executed at each node in the B&B tree or just when feasible solutions are found\nAdd support for semicontinuous and semi-integer variables for those solvers that support them.\nAdd support for index dependencies (for example, triangular indexing) in @defVar, @addConstraint, and @defExpr (for example, @defVar(m, x[i=1:10,j=i:10])).\nThis required some changes to the internal structure of JuMP containers, which may break code that explicitly stored JuMPDict objects.","category":"page"},{"location":"release_notes/#[Version-0.5.8](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.8)-(September-24,-2014)","page":"Release notes","title":"Version 0.5.8 (September 24, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug with specifying solvers (affects Julia 0.2 only)","category":"page"},{"location":"release_notes/#[Version-0.5.7](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.7)-(September-5,-2014)","page":"Release notes","title":"Version 0.5.7 (September 5, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in printing models","category":"page"},{"location":"release_notes/#[Version-0.5.6](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.6)-(September-2,-2014)","page":"Release notes","title":"Version 0.5.6 (September 2, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Add support for semicontinuous and semi-integer variables for those solvers that support them.\nBreaking change: Syntax for Variable() constructor has changed (use of this interface remains discouraged)\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"release_notes/#[Version-0.5.5](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.5)-(July-6,-2014)","page":"Release notes","title":"Version 0.5.5 (July 6, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.","category":"page"},{"location":"release_notes/#[Version-0.5.4](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.4)-(June-19,-2014)","page":"Release notes","title":"Version 0.5.4 (June 19, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for breaking change in MathProgBase which reduces loading times for using JuMP\nFix error when MIPs not solved to optimality","category":"page"},{"location":"release_notes/#[Version-0.5.3](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.3)-(May-21,-2014)","page":"Release notes","title":"Version 0.5.3 (May 21, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for breaking change in ReverseDiffSparse","category":"page"},{"location":"release_notes/#[Version-0.5.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.2)-(May-9,-2014)","page":"Release notes","title":"Version 0.5.2 (May 9, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix compatibility with Julia 0.3 prerelease","category":"page"},{"location":"release_notes/#[Version-0.5.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.1)-(May-5,-2014)","page":"Release notes","title":"Version 0.5.1 (May 5, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix a bug in coefficient handling inside lazy constraints and user cuts","category":"page"},{"location":"release_notes/#[Version-0.5.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.5.0)-(May-2,-2014)","page":"Release notes","title":"Version 0.5.0 (May 2, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.\ngetValue for AffExpr and QuadExpr\nBreaking change: getSolverModel replaced by getInternalModel, which returns the internal MathProgBase-level model\nGroups of constraints can be specified with @addConstraint (see documentation for details). This is not a breaking change.\ndot(::JuMPDict{Variable},::JuMPDict{Variable}) now returns the corresponding quadratic expression.","category":"page"},{"location":"release_notes/#[Version-0.4.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.1)-(March-24,-2014)","page":"Release notes","title":"Version 0.4.1 (March 24, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug where change in objective sense was ignored when re-solving a model.\nFix issue with handling zero coefficients in AffExpr.","category":"page"},{"location":"release_notes/#[Version-0.4.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.4.0)-(March-10,-2014)","page":"Release notes","title":"Version 0.4.0 (March 10, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Support for SOS1 and SOS2 constraints.\nSolver-independent callback for user heuristics.\ndot and sum implemented for JuMPDict objects. Now you can say @addConstraint(m, dot(a,x) <= b).\nDevelopers: support for extensions to JuMP. See definition of Model in src/JuMP.jl for more details.\nOption to construct the low-level model before optimizing.","category":"page"},{"location":"release_notes/#[Version-0.3.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.2)-(February-17,-2014)","page":"Release notes","title":"Version 0.3.2 (February 17, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Improved model printing\nPreliminary support for IJulia output","category":"page"},{"location":"release_notes/#[Version-0.3.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.1)-(January-30,-2014)","page":"Release notes","title":"Version 0.3.1 (January 30, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Documentation updates\nSupport for MOSEK\nCPLEXLink renamed to CPLEX","category":"page"},{"location":"release_notes/#[Version-0.3.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.3.0)-(January-21,-2014)","page":"Release notes","title":"Version 0.3.0 (January 21, 2014)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Unbounded/infeasibility rays: getValue() will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. getDual() will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.\nSolver-independent callbacks for user generated cuts.\nUse new interface for solver-independent QCQP.\nsetlazycallback renamed to setLazyCallback for consistency.","category":"page"},{"location":"release_notes/#[Version-0.2.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.2.0)-(December-15,-2013)","page":"Release notes","title":"Version 0.2.0 (December 15, 2013)","text":"","category":"section"},{"location":"release_notes/#Breaking-8","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Objective sense is specified in setObjective instead of in the Model constructor.\nlpsolver and mipsolver merged into single solver option.","category":"page"},{"location":"release_notes/#Added-31","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Problem modification with efficient LP restarts and MIP warm-starts.\nRelatedly, column-wise modeling now supported.\nSolver-independent callbacks supported. Currently we support only a \"lazy constraint\" callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.","category":"page"},{"location":"release_notes/#[Version-0.1.2](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.2)-(November-16,-2013)","page":"Release notes","title":"Version 0.1.2 (November 16, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Bug fixes for printing, improved error messages.\nAllow AffExpr to be used in macros; for example, ex = y + z; @addConstraint(m, x + 2*ex <= 3)","category":"page"},{"location":"release_notes/#[Version-0.1.1](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.1)-(October-23,-2013)","page":"Release notes","title":"Version 0.1.1 (October 23, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Update for solver specification API changes in MathProgBase.","category":"page"},{"location":"release_notes/#[Version-0.1.0](https://github.com/jump-dev/JuMP.jl/releases/tag/v0.1.0)-(October-3,-2013)","page":"Release notes","title":"Version 0.1.0 (October 3, 2013)","text":"","category":"section"},{"location":"release_notes/","page":"Release notes","title":"Release notes","text":"Initial public release.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\n    import HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/constraints/#jump_constraints","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP is based on the MathOptInterface (MOI) API. Because of this, JuMP uses the following standard form to represent problems:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Each constraint, f_i(x) in mathcalS_i, is composed of a function and a set. For example, instead of calling a^top x le b a less-than-or-equal-to constraint, we say that it is a scalar-affine-in-less-than constraint, where the function a^top x belongs to the less-than set (-infty b. We use the shorthand function-in-set to refer to constraints composed of different types of functions and sets.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This page explains how to write various types of constraints in JuMP. For nonlinear constraints, see Nonlinear Modeling instead.","category":"page"},{"location":"manual/constraints/#Add-a-constraint","page":"Constraints","title":"Add a constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Add a constraint to a JuMP model using the @constraint macro. The syntax to use depends on the type of constraint you wish to add.","category":"page"},{"location":"manual/constraints/#Add-a-linear-constraint","page":"Constraints","title":"Add a linear constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create linear constraints using the @constraint macro:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c1, sum(x) <= 1)\nc1 : x[1] + x[2] + x[3] ≤ 1\n\njulia> @constraint(model, c2, x[1] + 2 * x[3] >= 2)\nc2 : x[1] + 2 x[3] ≥ 2\n\njulia> @constraint(model, c3, sum(i * x[i] for i in 1:3) == 3)\nc3 : x[1] + 2 x[2] + 3 x[3] = 3\n\njulia> @constraint(model, c4, 4 <= 2 * x[2] <= 5)\nc4 : 2 x[2] ∈ [4, 5]","category":"page"},{"location":"manual/constraints/#Normalization","page":"Constraints","title":"Normalization","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP normalizes constraints by moving all of the terms containing variables to the left-hand side and all of the constant terms to the right-hand side. Thus, we get:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, c, 2x + 1 <= 4x + 4)\nc : -2 x ≤ 3","category":"page"},{"location":"manual/constraints/#quad_constraints","page":"Constraints","title":"Add a quadratic constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In addition to affine functions, JuMP also supports constraints with quadratic terms. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @variable(model, t >= 0)\nt\n\njulia> @constraint(model, my_q, x[1]^2 + x[2]^2 <= t^2)\nmy_q : x[1]² + x[2]² - t² ≤ 0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nBecause solvers can take advantage of the knowledge that a constraint is quadratic, prefer adding quadratic constraints using @constraint, rather than @NLconstraint.","category":"page"},{"location":"manual/constraints/#Vectorized-constraints","page":"Constraints","title":"Vectorized constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also add constraints to JuMP using vectorized linear algebra. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> A = [1 2; 3 4]\n2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> b = [5, 6]\n2-element Vector{Int64}:\n 5\n 6\n\njulia> @constraint(model, con_vector, A * x == b)\ncon_vector : [x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)\n\njulia> @constraint(model, con_scalar, A * x .== b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}, ScalarShape}}:\n con_scalar : x[1] + 2 x[2] = 5\n con_scalar : 3 x[1] + 4 x[2] = 6","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The two constraints, == and .== are similar, but subtly different. The first creates a single constraint that is a MOI.VectorAffineFunction in MOI.Zeros constraint. The second creates a vector of MOI.ScalarAffineFunction in MOI.EqualTo constraints.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Which formulation to choose depends on the solver, and what you want to do with the constraint object con_vector or con_scalar.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you are using a conic solver, expect the dual of con_vector to be a Vector{Float64}, and do not intend to delete a row in the constraint, choose the == formulation.\nIf you are using a solver that expects a list of scalar constraints, for example HiGHS, or you wish to delete part of the constraint or access a single row of the constraint, for example, dual(con_scalar[2]), then use the broadcast .==.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP reformulates both constraints into the other form if needed by the solver, but choosing the right format for a particular solver is more efficient.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also use <=, .<= , >=, and .>= as comparison operators in the constraint.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, A * x <= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)\n\njulia> @constraint(model, A * x .<= b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x[1] + 2 x[2] ≤ 5\n 3 x[1] + 4 x[2] ≤ 6\n\njulia> @constraint(model, A * x >= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, A * x .>= b)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n x[1] + 2 x[2] ≥ 5\n 3 x[1] + 4 x[2] ≥ 6","category":"page"},{"location":"manual/constraints/#Vectorized-matrix-constraints","page":"Constraints","title":"Vectorized matrix constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In most cases, you cannot use the non-broadcasting syntax for general matrices. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[2,1]  X[2,2]\n\njulia> @constraint(model, X >= 0)\nERROR: At none:1: `@constraint(model, X >= 0)`: Unsupported matrix in vector-valued set. Did you mean to use the broadcasting syntax `.>=` instead? Alternatively, perhaps you are missing a set argument like `@constraint(model, X >= 0, PSDCone())` or `@constraint(model, X >= 0, HermmitianPSDCone())`.\nStacktrace:\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Instead, to represent matrix inequalities you must always use the element-wise broadcasting .==, .>=, or .<=, or use the Set inequality syntax.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"There are two exceptions: if the result of the left-hand side minus the right-hand side is a LinearAlgebra.Symmetric matrix or a LinearAlgebra.Hermitian matrix, you may use the non-broadcasting equality syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> using LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2], Symmetric)\n2×2 Symmetric{VariableRef, Matrix{VariableRef}}:\n X[1,1]  X[1,2]\n X[1,2]  X[2,2]\n\njulia> @constraint(model, X == LinearAlgebra.I)\n[X[1,1] - 1  X[1,2];\n X[1,2]      X[2,2] - 1] ∈ Zeros()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Despite the model showing the matrix in Zeros, this will add only three rows to the constraint matrix because the symmetric constraints are redundant. In contrast, the broadcasting syntax adds four linear constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, X .== LinearAlgebra.I)\n2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}, ScalarShape}}:\n X[1,1] = 1  X[1,2] = 0\n X[1,2] = 0  X[2,2] = 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The same holds for LinearAlgebra.Hermitian matrices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> using LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2] in HermitianPSDCone())\n2×2 Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(X[1,1])                    real(X[1,2]) + imag(X[1,2]) im\n real(X[1,2]) - imag(X[1,2]) im  real(X[2,2])\n\njulia> @constraint(model, X == LinearAlgebra.I)\n[real(X[1,1]) - 1                real(X[1,2]) + imag(X[1,2]) im;\n real(X[1,2]) - imag(X[1,2]) im  real(X[2,2]) - 1] ∈ Zeros()\n\njulia> @constraint(model, X .== LinearAlgebra.I)\n2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{ComplexF64}, MathOptInterface.EqualTo{ComplexF64}}, ScalarShape}}:\n real(X[1,1]) = 1                    real(X[1,2]) + imag(X[1,2]) im = 0\n real(X[1,2]) - imag(X[1,2]) im = 0  real(X[2,2]) = 1","category":"page"},{"location":"manual/constraints/#Containers-of-constraints","page":"Constraints","title":"Containers of constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The @constraint macro supports creating collections of constraints. We'll cover some brief syntax here; read the Constraint containers section for more details:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create arrays of constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=1:3], x[i] <= i^2)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c[1] : x[1] ≤ 1\n c[2] : x[2] ≤ 4\n c[3] : x[3] ≤ 9\n\njulia> c[2]\nc[2] : x[2] ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Sets can be any Julia type that supports iteration:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=2:3, [\"red\", \"blue\"]], x[i] <= i^2)\n2-dimensional DenseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, [\"red\", \"blue\"]\nAnd data, a 2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c[2,red] : x[2] ≤ 4  c[2,blue] : x[2] ≤ 4\n c[3,red] : x[3] ≤ 9  c[3,blue] : x[3] ≤ 9\n\njulia> c[2, \"red\"]\nc[2,red] : x[2] ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Sets can depend upon previous indices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, c[i=1:3, j=i:3], x[i] <= j)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  c[1,1] : x[1] ≤ 1\n  [1, 2]  =  c[1,2] : x[1] ≤ 2\n  [1, 3]  =  c[1,3] : x[1] ≤ 3\n  [2, 2]  =  c[2,2] : x[2] ≤ 2\n  [2, 3]  =  c[2,3] : x[2] ≤ 3\n  [3, 3]  =  c[3,3] : x[3] ≤ 3","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and you can filter elements in the sets using the ; syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:9]);\n\njulia> @constraint(model, c[i=1:9; mod(i, 3) == 0], x[i] <= i)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 1, Tuple{Int64}} with 3 entries:\n  [3]  =  c[3] : x[3] ≤ 3\n  [6]  =  c[6] : x[6] ≤ 6\n  [9]  =  c[9] : x[9] ≤ 9","category":"page"},{"location":"manual/constraints/#Registered-constraints","page":"Constraints","title":"Registered constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"When you create constraints, JuMP registers them inside the model using their corresponding symbol. Get a registered name using model[:key]:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, my_c, 2x <= 1)\nmy_c : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: my_c, x\n\njulia> model[:my_c] === my_c\ntrue","category":"page"},{"location":"manual/constraints/#Anonymous-constraints","page":"Constraints","title":"Anonymous constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To reduce the likelihood of accidental bugs, and because JuMP registers constraints inside a model, creating two constraints with the same name is an error:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2x <= 1)\nc : 2 x ≤ 1\n\njulia> @constraint(model, c, 2x <= 1)\nERROR: An object of name c is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :c)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:c]`.\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A common reason for encountering this error is adding constraints in a loop.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"As a work-around, JuMP provides anonymous constraints. Create an anonymous constraint by omitting the name argument:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> c = @constraint(model, 2x <= 1)\n2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Create a container of anonymous constraints by dropping the name in front of the [:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> c = @constraint(model, [i = 1:3], x[i] <= i)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x[1] ≤ 1\n x[2] ≤ 2\n x[3] ≤ 3","category":"page"},{"location":"manual/constraints/#Constraint-names","page":"Constraints","title":"Constraint names","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In addition to the symbol that constraints are registered with, constraints have a String name that is used for printing and writing to file formats.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Get and set the name of a constraint using name(::JuMP.ConstraintRef) and set_name(::JuMP.ConstraintRef, ::String):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> @constraint(model, con, x <= 1)\ncon : x ≤ 1\n\njulia> name(con)\n\"con\"\n\njulia> set_name(con, \"my_con_name\")\n\njulia> con\nmy_con_name : x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Override the default choice of name using the base_name keyword:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> con = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Note that names apply to each element of the container, not to the container of constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> name(con[1])\n\"my_con[1]\"\n\njulia> set_name(con[1], \"c\")\n\njulia> con\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n c : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nFor some models, setting the string name of each constraint can take a non-trivial portion of the total time required to build the model. Turn off String names by passing set_string_name = false to @constraint:julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, x <= 2, set_string_name = false)\nx ≤ 2See Disable string names for more information.","category":"page"},{"location":"manual/constraints/#Retrieve-a-constraint-by-name","page":"Constraints","title":"Retrieve a constraint by name","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Retrieve a constraint from a model using constraint_by_name:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> constraint_by_name(model, \"c\")\nc : x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the name is not present, nothing will be returned:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> constraint_by_name(model, \"bad_name\")","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can only look up individual constraints using constraint_by_name. Something like this will not work:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> con = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2\n\njulia> constraint_by_name(model, \"my_con\")","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To look up a collection of constraints, do not use constraint_by_name. Instead, register them using the model[:key] = value syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(); @variable(model, x);\n\njulia> model[:con] = @constraint(model, [i=1:2], x <= i, base_name = \"my_con\")\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2\n\njulia> model[:con]\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n my_con[1] : x ≤ 1\n my_con[2] : x ≤ 2","category":"page"},{"location":"manual/constraints/#String-names,-symbolic-names,-and-bindings","page":"Constraints","title":"String names, symbolic names, and bindings","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"It's common for new users to experience confusion relating to constraints. Part of the problem is the difference between the name that a constraint is registered under and the String name used for printing.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Here's a summary of the differences:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Constraints are created using @constraint.\nConstraints can be named or anonymous.\nNamed constraints have the form @constraint(model, c, expr). For named constraints:\nThe String name of the constraint is set to \"c\".\nA Julia variable c is created that binds c to  the JuMP constraint.\nThe name :c is registered as a key in the model with the value c.\nAnonymous constraints have the form c = @constraint(model, expr). For anonymous constraints:\nThe String name of the constraint is set to \"\".\nYou control the name of the Julia variable used as the binding.\nNo name is registered as a key in the model.\nThe base_name keyword can override the String name of the constraint.\nYou can manually register names in the model via model[:key] = value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Here's an example of the differences:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> c_binding = @constraint(model, 2x <= 1, base_name = \"c\")\nc : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x\n\njulia> c\nERROR: UndefVarError: `c` not defined\n\njulia> c_binding\nc : 2 x ≤ 1\n\njulia> name(c_binding)\n\"c\"\n\njulia> model[:c_register] = c_binding\nc : 2 x ≤ 1\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\n`AffExpr`-in-`MathOptInterface.LessThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: c_register, x\n\njulia> model[:c_register]\nc : 2 x ≤ 1\n\njulia> model[:c_register] === c_binding\ntrue\n\njulia> c\nERROR: UndefVarError: `c` not defined","category":"page"},{"location":"manual/constraints/#The-@constraints-macro","page":"Constraints","title":"The @constraints macro","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you have many @constraint calls, use the @constraints macro to improve readability:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraints(model, begin\n           2x <= 1\n           c, x >= -1\n       end)\n(2 x ≤ 1, c : x ≥ -1)\n\njulia> print(model)\nFeasibility\nSubject to\n c : x ≥ -1\n 2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The @constraints macro returns a tuple of the constraints that were defined.","category":"page"},{"location":"manual/constraints/#constraint_duality","page":"Constraints","title":"Duality","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP adopts the notion of conic duality from MathOptInterface. For linear programs, a feasible dual on a >= constraint is nonnegative and a feasible dual on a <= constraint is nonpositive. If the constraint is an equality constraint, it depends on which direction is binding.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nJuMP's definition of duality is independent of the objective sense. That is, the sign of feasible duals associated with a constraint depends on the direction of the constraint and not whether the problem is maximization or minimization. This is a different convention from linear programming duality in some common textbooks. If you have a linear program, and you want the textbook definition, you probably want to use shadow_price and reduced_cost instead.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The dual value associated with a constraint in the most recent solution can be accessed using the dual function. Use has_duals to check if the model has a dual solution available to query. For example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x <= 1)\ncon : x ≤ 1\n\njulia> @objective(model, Min, -2x)\n-2 x\n\njulia> has_duals(model)\nfalse\n\njulia> optimize!(model)\n\njulia> has_duals(model)\ntrue\n\njulia> dual(con)\n-2.0\n\njulia> @objective(model, Max, 2x)\n2 x\n\njulia> optimize!(model)\n\njulia> dual(con)\n-2.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To help users who may be less familiar with conic duality, JuMP provides shadow_price, which returns a value that can be interpreted as the improvement in the objective in response to an infinitesimal relaxation (on the scale of one unit) in the right-hand side of the constraint. shadow_price can be used only on linear constraints with a <=, >=, or == comparison operator.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In the example above, dual(con) returned -2.0 regardless of the optimization sense. However, in the second case when the optimization sense is Max, shadow_price returns:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> shadow_price(con)\n2.0","category":"page"},{"location":"manual/constraints/#Duals-of-variable-bounds","page":"Constraints","title":"Duals of variable bounds","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To query the dual variables associated with a variable bound, first obtain a constraint reference using one of UpperBoundRef, LowerBoundRef, or FixRef, and then call dual on the returned constraint reference. The reduced_cost function may simplify this process as it returns the shadow price of an active bound of a variable (or zero, if no active bound exists).","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x <= 1)\nx\n\njulia> @objective(model, Min, -2x)\n-2 x\n\njulia> optimize!(model)\n\njulia> dual(UpperBoundRef(x))\n-2.0\n\njulia> reduced_cost(x)\n-2.0","category":"page"},{"location":"manual/constraints/#Modify-a-constant-term","page":"Constraints","title":"Modify a constant term","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This section explains how to modify the constant term in a constraint. There are multiple ways to achieve this goal; we explain three options.","category":"page"},{"location":"manual/constraints/#Option-1:-change-the-right-hand-side","page":"Constraints","title":"Option 1: change the right-hand side","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use set_normalized_rhs to modify the right-hand side (constant) term of a linear or quadratic  constraint. Use normalized_rhs to query the right-hand side term.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> set_normalized_rhs(con, 3)\n\njulia> con\ncon : 2 x ≤ 3\n\njulia> normalized_rhs(con)\n3.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nset_normalized_rhs sets the right-hand side term of the normalized constraint. See Normalization for more details.","category":"page"},{"location":"manual/constraints/#Option-2:-use-fixed-variables","page":"Constraints","title":"Option 2: use fixed variables","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If constraints are complicated, for example, they are composed of a number of components, each of which has a constant term, then it may be difficult to calculate what the right-hand side term is in the standard form.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For this situation, JuMP includes the ability to fix variables to a value using the fix function. Fixing a variable sets its lower and upper bound to the same value. Thus, changes in a constant term can be simulated by adding a new variable and fixing it to different values. Here is an example:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, const_term)\nconst_term\n\njulia> @constraint(model, con, 2x <= const_term + 1)\ncon : 2 x - const_term ≤ 1\n\njulia> fix(const_term, 1.0)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The constraint con is now equivalent to 2x <= 2.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nFixed variables are not replaced with constants when communicating the problem to a solver. Therefore, even though const_term is fixed, it is still a decision variable, and so const_term * x is bilinear.","category":"page"},{"location":"manual/constraints/#Option-3:-modify-the-function's-constant-term","page":"Constraints","title":"Option 3: modify the function's constant term","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The third option is to use add_to_function_constant. The constant given is added to the function of a func-in-set constraint. In the following example, adding 2 to the function has the effect of removing 2 to the right-hand side:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> add_to_function_constant(con, 2)\n\njulia> con\ncon : 2 x ≤ -1\n\njulia> normalized_rhs(con)\n-1.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In the case of interval constraints, the constant is removed from each bound:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 0 <= 2x + 1 <= 2)\ncon : 2 x ∈ [-1, 1]\n\njulia> add_to_function_constant(con, 3)\n\njulia> con\ncon : 2 x ∈ [-4, -2]","category":"page"},{"location":"manual/constraints/#Modify-a-variable-coefficient","page":"Constraints","title":"Modify a variable coefficient","text":"","category":"section"},{"location":"manual/constraints/#Scalar-constraints","page":"Constraints","title":"Scalar constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To modify the coefficients for a linear term (modifying the coefficient of a quadratic term is not supported) in a constraint, use set_normalized_coefficient. To query the current coefficient, use normalized_coefficient.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, con, 2x[1] + x[2] <= 1)\ncon : 2 x[1] + x[2] ≤ 1\n\njulia> set_normalized_coefficient(con, x[2], 0)\n\njulia> con\ncon : 2 x[1] ≤ 1\n\njulia> normalized_coefficient(con, x[2])\n0.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nset_normalized_coefficient sets the coefficient of the normalized constraint. See Normalization for more details.","category":"page"},{"location":"manual/constraints/#Vector-constraints","page":"Constraints","title":"Vector constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To modify the coefficients of a vector-valued constraint, use set_normalized_coefficients.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, [2x + 3x, 4x] in MOI.Nonnegatives(2))\ncon : [5 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 3.0)])\n\njulia> con\ncon : [3 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 2.0), (2, 5.0)])\n\njulia> con\ncon : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/#Delete-a-constraint","page":"Constraints","title":"Delete a constraint","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use delete to delete a constraint from a model. Use is_valid to check if a constraint belongs to a model and has not been deleted.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1\n\njulia> is_valid(model, con)\ntrue\n\njulia> delete(model, con)\n\njulia> is_valid(model, con)\nfalse","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Deleting a constraint does not unregister the symbolic reference from the model. Therefore, creating a new constraint of the same name will throw an error:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, con, 2x <= 1)\nERROR: An object of name con is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :con)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:con]`.\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"After calling delete, call unregister to remove the symbolic reference:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> unregister(model, :con)\n\njulia> @constraint(model, con, 2x <= 1)\ncon : 2 x ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"info: Info\ndelete does not automatically unregister because we do not distinguish between names that are automatically registered by JuMP macros, and names that are manually registered by the user by setting values in object_dictionary. In addition, deleting a constraint and then adding a new constraint of the same name is an easy way to introduce bugs into your code.","category":"page"},{"location":"manual/constraints/#Start-values","page":"Constraints","title":"Start values","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Provide a starting value (also called warmstart) for a constraint's primal and dual solutions using set_start_value and set_dual_start_value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Query the starting value for a constraint's primal and dual solution using start_value and dual_start_value. If no start value has been set, the methods will return nothing.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x >= 10)\ncon : x ≥ 10\n\njulia> start_value(con)\n\njulia> set_start_value(con, 10.0)\n\njulia> start_value(con)\n10.0\n\njulia> dual_start_value(con)\n\njulia> set_dual_start_value(con, 2)\n\njulia> dual_start_value(con)\n2.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Vector-valued constraints require a vector:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, con, x in SecondOrderCone())\ncon : [x[1], x[2], x[3]] in MathOptInterface.SecondOrderCone(3)\n\njulia> dual_start_value(con)\n\njulia> set_dual_start_value(con, [1.0, 2.0, 3.0])\n\njulia> dual_start_value(con)\n3-element Vector{Float64}:\n 1.0\n 2.0\n 3.0","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nTo simplify setting start values for all variables and constraints in a model, see set_start_values. The Primal and dual warm-starts tutorial also gives a detailed description of how to iterate over constraints in the model to set custom start values.","category":"page"},{"location":"manual/constraints/#Constraint-containers","page":"Constraints","title":"Constraint containers","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Like Variable containers, JuMP provides a mechanism for building groups of constraints compactly. References to these groups of constraints are returned in containers. Three types of constraint containers are supported: Arrays, DenseAxisArrays, and SparseAxisArrays. We explain each of these in the following.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/constraints/#constraint_arrays","page":"Constraints","title":"Arrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"One way of adding a group of constraints compactly is the following:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:3], i * x <= i + 1)\n3-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[1] : x ≤ 2\n con[2] : 2 x ≤ 3\n con[3] : 3 x ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP returns references to the three constraints in an Array that is bound to the Julia variable con. This array can be accessed and sliced as you would with any Julia array:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con[1]\ncon[1] : x ≤ 2\n\njulia> con[2:3]\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[2] : 2 x ≤ 3\n con[3] : 3 x ≤ 4","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Anonymous containers can also be constructed by dropping the name (for example, con) before the square brackets:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con = @constraint(model, [i = 1:2], i * x <= i + 1)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n x ≤ 2\n 2 x ≤ 3","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Just like @variable, JuMP will form an Array of constraints when it can determine at parse time that the indices are one-based integer ranges. Therefore con[1:b] will create an Array, but con[a:b] will not. A special case is con[Base.OneTo(n)] which will produce an Array. If JuMP cannot determine that the indices are one-based integer ranges (for example, in the case of con[a:b]), JuMP will create a DenseAxisArray instead.","category":"page"},{"location":"manual/constraints/#DenseAxisArrays","page":"Constraints","title":"DenseAxisArrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The syntax for constructing a DenseAxisArray of constraints is very similar to the syntax for constructing a DenseAxisArray of variables.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:2, j = 2:3], i * x <= j + 1)\n2-dimensional DenseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, 2:3\nAnd data, a 2×2 Matrix{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n con[1,2] : x ≤ 3    con[1,3] : x ≤ 4\n con[2,2] : 2 x ≤ 3  con[2,3] : 2 x ≤ 4","category":"page"},{"location":"manual/constraints/#SparseAxisArrays","page":"Constraints","title":"SparseAxisArrays","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The syntax for constructing a SparseAxisArray of constraints is very similar to the syntax for constructing a SparseAxisArray of variables.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con[i = 1:2, j = 1:2; i != j], i * x <= j + 1)\nJuMP.Containers.SparseAxisArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}, 2, Tuple{Int64, Int64}} with 2 entries:\n  [1, 2]  =  con[1,2] : x ≤ 3\n  [2, 1]  =  con[2,1] : 2 x ≤ 2","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"warning: Warning\nIf you have many index dimensions and a large amount of sparsity, read Performance considerations.","category":"page"},{"location":"manual/constraints/#Forcing-the-container-type","page":"Constraints","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"When creating a container of constraints, JuMP will attempt to choose the tightest container type that can store the constraints. However, because this happens at parse time, it does not always make the best choice. Just like in @variable, you can force the type of container using the container keyword. For syntax and the reason behind this, take a look at the variable docs.","category":"page"},{"location":"manual/constraints/#Constraints-with-similar-indices","page":"Constraints","title":"Constraints with similar indices","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Containers are often used to create constraints over a set of indices. However, you'll often have cases in which you are repeating the indices:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @variable(model, y[1:2]);\n\njulia> @constraints(model, begin\n           [i=1:2, j=1:2, k=1:2], i * x[j] <= k\n           [i=1:2, j=1:2, k=1:2], i * y[j] <= k\n       end);","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This is hard to read and leads to a lot of copy-paste. A more readable way is to use a for-loop:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> for i=1:2, j=1:2, k=1:2\n           @constraints(model, begin\n               i * x[j] <= k\n               i * y[j] <= k\n           end)\n       end","category":"page"},{"location":"manual/constraints/#Accessing-constraints-from-a-model","page":"Constraints","title":"Accessing constraints from a model","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Query the types of function-in-set constraints in a model using list_of_constraint_types:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2] >= i, Int);\n\njulia> @constraint(model, x[1] + x[2] <= 1);\n\njulia> list_of_constraint_types(model)\n3-element Vector{Tuple{Type, Type}}:\n (AffExpr, MathOptInterface.LessThan{Float64})\n (VariableRef, MathOptInterface.GreaterThan{Float64})\n (VariableRef, MathOptInterface.Integer)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For a given combination of function and set type, use num_constraints to access the number of constraints and all_constraints to access a list of their references:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> num_constraints(model, VariableRef, MOI.Integer)\n2\n\njulia> cons = all_constraints(model, VariableRef, MOI.Integer)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}, ScalarShape}}:\n x[1] integer\n x[2] integer","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also count the total number of constraints in the model, but you must explicitly choose whether to count VariableRef constraints such as bound and integrality constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> num_constraints(model; count_variable_in_set_constraints = true)\n5\n\njulia> num_constraints(model; count_variable_in_set_constraints = false)\n1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The same also applies for all_constraints:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> all_constraints(model; include_variable_in_set_constraints = true)\n5-element Vector{ConstraintRef}:\n x[1] + x[2] ≤ 1\n x[1] ≥ 1\n x[2] ≥ 2\n x[1] integer\n x[2] integer\n\njulia> all_constraints(model; include_variable_in_set_constraints = false)\n1-element Vector{ConstraintRef}:\n x[1] + x[2] ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If you need finer-grained control on which constraints to include, use a variant of:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> sum(\n           num_constraints(model, F, S) for\n           (F, S) in list_of_constraint_types(model) if F != VariableRef\n       )\n1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use constraint_object to get an instance of an AbstractConstraint object that stores the constraint data:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> con = constraint_object(cons[1])\nScalarConstraint{VariableRef, MathOptInterface.Integer}(x[1], MathOptInterface.Integer())\n\njulia> con.func\nx[1]\n\njulia> con.set\nMathOptInterface.Integer()","category":"page"},{"location":"manual/constraints/#MathOptInterface-constraints","page":"Constraints","title":"MathOptInterface constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Because JuMP is based on MathOptInterface, you can add any constraints supported by MathOptInterface using the function-in-set syntax. For a list of supported functions and sets, read Standard form problem.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"note: Note\nWe use MOI as an alias for the MathOptInterface module. This alias is defined by using JuMP. You may also define it in your code as follows:import MathOptInterface as MOI","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, the following two constraints are equivalent:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @constraint(model, 2 * x[1] <= 1)\n2 x[1] ≤ 1\n\njulia> @constraint(model, 2 * x[1] in MOI.LessThan(1.0))\n2 x[1] ≤ 1","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"You can also use any set defined by MathOptInterface:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x - [1; 2; 3] in MOI.Nonnegatives(3))\n[x[1] - 1, x[2] - 2, x[3] - 3] ∈ MathOptInterface.Nonnegatives(3)\n\njulia> @constraint(model, x in MOI.ExponentialCone())\n[x[1], x[2], x[3]] ∈ MathOptInterface.ExponentialCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"info: Info\nSimilar to how JuMP defines the <= and >= syntax as a convenience way to specify MOI.LessThan and MOI.GreaterThan constraints, the remaining sections in this page describe functions and syntax that have been added for the convenience of common modeling situations.","category":"page"},{"location":"manual/constraints/#Set-inequality-syntax","page":"Constraints","title":"Set inequality syntax","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For modeling convenience, the syntax @constraint(model, x >= y, Set()) is short-hand for @constraint(model, x - y in Set()). Therefore, the following calls are equivalent:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> y = [0.5, 0.75];\n\njulia> @constraint(model, x >= y, MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, y <= x, MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> @constraint(model, x - y in MOI.Nonnegatives(2))\n[x[1] - 0.5, x[2] - 0.75] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Non-zero constants are not supported in this syntax:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x >= 1, MOI.Nonnegatives(2))\nERROR: Operation `sub_mul` between `Vector{VariableRef}` and `Int64` is not allowed. This most often happens when you write a constraint like `x >= y` where `x` is an array and `y` is a constant. Use the broadcast syntax `x .- y >= 0` instead.\nStacktrace:\n[...]","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Use instead:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x .- 1 >= 0, MOI.Nonnegatives(2))\n[x[1] - 1, x[2] - 1] ∈ MathOptInterface.Nonnegatives(2)","category":"page"},{"location":"manual/constraints/#Second-order-cone-constraints","page":"Constraints","title":"Second-order cone constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A SecondOrderCone constrains the variables t and x to the set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"x_2 le t","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and t ge 0. It can be added as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, t)\nt\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, [t; x] in SecondOrderCone())\n[t, x[1], x[2]] ∈ MathOptInterface.SecondOrderCone(3)","category":"page"},{"location":"manual/constraints/#Rotated-second-order-cone-constraints","page":"Constraints","title":"Rotated second-order cone constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A RotatedSecondOrderCone constrains the variables t, u, and x to the set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"x_2^2 le 2 t cdot u","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"and t u ge 0. It can be added as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, t)\nt\n\njulia> @variable(model, u)\nu\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, [t; u; x] in RotatedSecondOrderCone())\n[t, u, x[1], x[2]] ∈ MathOptInterface.RotatedSecondOrderCone(4)","category":"page"},{"location":"manual/constraints/#Special-Ordered-Sets-of-Type-1","page":"Constraints","title":"Special Ordered Sets of Type 1","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In a Special Ordered Set of Type 1 (often denoted SOS-I or SOS1), at most one element can take a non-zero value.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Construct SOS-I constraints using the SOS1 set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, x in SOS1())\n[x[1], x[2], x[3]] in MathOptInterface.SOS1{Float64}([1.0, 2.0, 3.0])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Although not required for feasibility, solvers can benefit from an ordering of the variables (for example, the variables represent different factories to build, at most one factory can be built, and the factories can be ordered according to cost). To induce an ordering, a vector of weights can be provided, and the variables are ordered according to their corresponding weight.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, in the constraint:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS1([3.1, 1.2, 2.3]))\n[x[1], x[2], x[3]] in MathOptInterface.SOS1{Float64}([3.1, 1.2, 2.3])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"the variables x have precedence x[2], x[3], x[1].","category":"page"},{"location":"manual/constraints/#Special-Ordered-Sets-of-Type-2","page":"Constraints","title":"Special Ordered Sets of Type 2","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In a Special Ordered Set of Type 2 (SOS-II), at most two elements can be non-zero, and if there are two non-zeros, they must be consecutive according to the ordering induced by a weight vector.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Construct SOS-II constraints using the SOS2 set:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS2([3.0, 1.0, 2.0]))\n[x[1], x[2], x[3]] in MathOptInterface.SOS2{Float64}([3.0, 1.0, 2.0])","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The possible non-zero pairs are (x[1], x[3]) and (x[2], x[3]):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the weight vector is omitted, JuMP induces an ordering from 1:length(x):","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, x in SOS2())\n[x[1], x[2], x[3]] in MathOptInterface.SOS2{Float64}([1.0, 2.0, 3.0])","category":"page"},{"location":"manual/constraints/#Indicator-constraints","page":"Constraints","title":"Indicator constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Indicator constraints consist of a binary variable and a linear constraint. The constraint holds when the binary variable takes the value 1. The constraint may or may not hold when the binary variable takes the value 0.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To enforce the constraint x + y <= 1 when the binary variable a is 1, use:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @variable(model, a, Bin)\na\n\njulia> @constraint(model, a => {x + y <= 1})\na => {x + y ≤ 1}","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"If the constraint must hold when a is zero, add ! or ¬ before the binary variable;","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, !a => {x + y <= 1})\n!a => {x + y ≤ 1}","category":"page"},{"location":"manual/constraints/#Semidefinite-constraints","page":"Constraints","title":"Semidefinite constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To constrain a matrix to be positive semidefinite (PSD), use PSDCone:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, X[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[2,1]  X[2,2]\n\njulia> @constraint(model, X >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[2,1]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"tip: Tip\nWhere possible, prefer constructing a matrix of Semidefinite variables using the @variable macro, rather than adding a constraint like @constraint(model, X >= 0, PSDCone()). In some solvers, adding the constraint via @constraint is less efficient, and can result in additional intermediate variables and constraints being added to the model.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The inequality X >= Y between two square matrices X and Y is understood as constraining X - Y to be positive semidefinite.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> Y = [1 2; 2 1]\n2×2 Matrix{Int64}:\n 1  2\n 2  1\n\njulia> @constraint(model, X >= Y, PSDCone())\n[X[1,1] - 1  X[1,2] - 2;\n X[2,1] - 2  X[2,2] - 1] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/#Symmetry","page":"Constraints","title":"Symmetry","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Solvers supporting PSD constraints usually expect to be given a matrix that is symbolically symmetric, that is, for which the expression in corresponding off-diagonal entries are the same. In our example, the expressions of entries (1, 2) and (2, 1) are respectively X[1,2] - 2 and X[2,1] - 2 which are different.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"To bridge the gap between the constraint modeled and what the solver expects, solvers may add an equality constraint X[1,2] - 2 == X[2,1] - 2 to force symmetry. Use LinearAlgebra.Symmetric to explicitly tell the solver that the matrix is symmetric:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> import LinearAlgebra\n\njulia> Z = [X[1, 1] X[1, 2]; X[1, 2] X[2, 2]]\n2×2 Matrix{VariableRef}:\n X[1,1]  X[1,2]\n X[1,2]  X[2,2]\n\njulia> @constraint(model, LinearAlgebra.Symmetric(Z) >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[1,2]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Note that the lower triangular entries are ignored even if they are different so use it with caution:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, LinearAlgebra.Symmetric(X) >= 0, PSDCone())\n[X[1,1]  X[1,2];\n X[1,2]  X[2,2]] ∈ PSDCone()","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"(Note the (2, 1) element of the constraint is X[1,2], not X[2,1].)","category":"page"},{"location":"manual/constraints/#Complementarity-constraints","page":"Constraints","title":"Complementarity constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"A mixed complementarity constraint F(x) ⟂ x consists of finding x in the interval [lb, ub], such that the following holds:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"F(x) == 0 if lb < x < ub\nF(x) >= 0 if lb == x\nF(x) <= 0 if x == ub","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"JuMP supports mixed complementarity constraints via complements(F(x), x) or F(x) ⟂ x in the @constraint macro. The interval set [lb, ub] is obtained from the variable bounds on x.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"For example, to define the problem 2x - 1 ⟂ x with x ∈ [0, ∞), do:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, 2x - 1 ⟂ x)\n[2 x - 1, x] ∈ MathOptInterface.Complements(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"This problem has a unique solution at x = 0.5.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"The perp operator ⟂ can be entered in most editors (and the Julia REPL) by typing \\perp<tab>.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"An alternative approach that does not require the ⟂ symbol uses the complements function as follows:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @constraint(model, complements(2x - 1, x))\n[2 x - 1, x] ∈ MathOptInterface.Complements(2)","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"In both cases, the mapping F(x) is supplied as the first argument, and the matching variable x is supplied as the second.","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Vector-valued complementarity constraints are also supported:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> @variable(model, -2 <= y[1:2] <= 2)\n2-element Vector{VariableRef}:\n y[1]\n y[2]\n\njulia> M = [1 2; 3 4]\n2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> q = [5, 6]\n2-element Vector{Int64}:\n 5\n 6\n\njulia> @constraint(model, M * y + q ⟂ y)\n[y[1] + 2 y[2] + 5, 3 y[1] + 4 y[2] + 6, y[1], y[2]] ∈ MathOptInterface.Complements(4)","category":"page"},{"location":"manual/constraints/#Boolean-constraints","page":"Constraints","title":"Boolean constraints","text":"","category":"section"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Add a Boolean constraint (a MOI.EqualTo{Bool} set) using the := operator with a Bool right-hand side term:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = GenericModel{Bool}();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x[1] || x[2] := true)\nx[1] || x[2] = true\n\njulia> @constraint(model, x[1] && x[2] := false)\nx[1] && x[2] = false\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariables: 2\n`GenericNonlinearExpr{GenericVariableRef{Bool}}`-in-`MathOptInterface.EqualTo{Bool}`: 2 constraints\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"Boolean constraints should not be added using the == operator because JuMP will rewrite the constraint as lhs - rhs = 0, and because constraints like a == b == c require parentheses to disambiguate between (a == b) == c and a == (b == c). In contrast, a == b := c is equivalent to (a == b) := c:","category":"page"},{"location":"manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> rhs = false\nfalse\n\njulia> @constraint(model, (x[1] == x[2]) == rhs)\n(x[1] == x[2]) - 0.0 = 0\n\njulia> @constraint(model, x[1] == x[2] := rhs)\nx[1] == x[2] = false","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"EditURL = \"power_systems.jl\"","category":"page"},{"location":"tutorials/applications/power_systems/#Power-Systems","page":"Power Systems","title":"Power Systems","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial was originally contributed by Yury Dvorkin and Miles Lubin.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This tutorial demonstrates how to formulate basic power systems engineering models in JuMP.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"We will consider basic \"economic dispatch\" and \"unit commitment\" models without taking into account transmission constraints.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For this tutorial, we use the following packages:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport Plots\nimport StatsPlots","category":"page"},{"location":"tutorials/applications/power_systems/#Economic-dispatch","page":"Power Systems","title":"Economic dispatch","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Economic dispatch (ED) is an optimization problem that minimizes the cost of supplying energy demand subject to operational constraints on power system assets. In its simplest modification, ED is an LP problem solved for an aggregated load and wind forecast and for a single infinitesimal moment.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Mathematically, the ED problem can be written as follows:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"min sum_i in I c^g_i cdot g_i + c^w cdot w","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"where c_i and g_i are the incremental cost ($/MWh) and power output (MW) of the i^th generator, respectively, and c^w and w are the incremental cost ($/MWh) and wind power injection (MW), respectively.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Subject to the constraints:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Minimum (g^min) and maximum (g^max) limits on power outputs of generators: g^min_i leq g_i leq g^max_i\nConstraint on the wind power injection: 0 leq w leq w^f where w and w^f are the wind power injection and wind power forecast, respectively.\nPower balance constraint: sum_i in I g_i + w = d^f where d^f is the demand forecast.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Further reading on ED models can be found in A. J. Wood, B. F. Wollenberg, and G. B. Sheblé, \"Power Generation, Operation and Control,\" Wiley, 2013.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Define some input data about the test system.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"We define some thermal generators:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function ThermalGenerator(\n    min::Float64,\n    max::Float64,\n    fixed_cost::Float64,\n    variable_cost::Float64,\n)\n    return (\n        min = min,\n        max = max,\n        fixed_cost = fixed_cost,\n        variable_cost = variable_cost,\n    )\nend\n\ngenerators = [\n    ThermalGenerator(0.0, 1000.0, 1000.0, 50.0),\n    ThermalGenerator(300.0, 1000.0, 0.0, 100.0),\n]","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"A wind generator","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"WindGenerator(variable_cost::Float64) = (variable_cost = variable_cost,)\n\nwind_generator = WindGenerator(50.0)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"And a scenario","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function Scenario(demand::Float64, wind::Float64)\n    return (demand = demand, wind = wind)\nend\n\nscenario = Scenario(1500.0, 200.0)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Create a function solve_economic_dispatch, which solves the economic dispatch problem for a given set of input parameters.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_economic_dispatch(generators::Vector, wind, scenario)\n    # Define the economic dispatch (ED) model\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    # Define decision variables\n    # power output of generators\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    # wind power injection\n    @variable(model, 0 <= w <= scenario.wind)\n    # Define the objective function\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    # Define the power balance constraint\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    # Solve statement\n    optimize!(model)\n    # return the optimal value of the objective function and its minimizers\n    return (\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        total_cost = objective_value(model),\n    )\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Solve the economic dispatch problem","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"solution = solve_economic_dispatch(generators, wind_generator, scenario);\n\nprintln(\"Dispatch of Generators: \", solution.g, \" MW\")\nprintln(\"Dispatch of Wind: \", solution.w, \" MW\")\nprintln(\"Wind spillage: \", solution.wind_spill, \" MW\")\nprintln(\"Total cost: \\$\", solution.total_cost)","category":"page"},{"location":"tutorials/applications/power_systems/#Economic-dispatch-with-adjustable-incremental-costs","page":"Power Systems","title":"Economic dispatch with adjustable incremental costs","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following exercise we adjust the incremental cost of generator G1 and observe its impact on the total cost.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function scale_generator_cost(g, scale)\n    return ThermalGenerator(g.min, g.max, g.fixed_cost, scale * g.variable_cost)\nend\n\nstart = time()\nc_g_scale_df = DataFrames.DataFrame(;\n    # Scale factor\n    scale = Float64[],\n    # Dispatch of Generator 1 [MW]\n    dispatch_G1 = Float64[],\n    # Dispatch of Generator 2 [MW]\n    dispatch_G2 = Float64[],\n    # Dispatch of Wind [MW]\n    dispatch_wind = Float64[],\n    # Spillage of Wind [MW]\n    spillage_wind = Float64[],\n    # Total cost [$]\n    total_cost = Float64[],\n)\nfor c_g1_scale in 0.5:0.1:3.0\n    # Update the incremental cost of the first generator at every iteration.\n    new_generators = scale_generator_cost.(generators, [c_g1_scale, 1.0])\n    # Solve the economic-dispatch problem with the updated incremental cost\n    sol = solve_economic_dispatch(new_generators, wind_generator, scenario)\n    push!(\n        c_g_scale_df,\n        (c_g1_scale, sol.g[1], sol.g[2], sol.w, sol.wind_spill, sol.total_cost),\n    )\nend\nprint(string(\"elapsed time: \", time() - start, \" seconds\"))","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"c_g_scale_df","category":"page"},{"location":"tutorials/applications/power_systems/#Modifying-the-JuMP-model-in-place","page":"Power Systems","title":"Modifying the JuMP model in-place","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Note that in the previous exercise we entirely rebuilt the optimization model at every iteration of the internal loop, which incurs an additional computational burden. This burden can be alleviated if instead of re-building the entire model, we modify the constraints or objective function, as it shown in the example below.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Compare the computing time in case of the above and below models.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_economic_dispatch_inplace(\n    generators::Vector,\n    wind,\n    scenario,\n    scale::AbstractVector{Float64},\n)\n    obj_out = Float64[]\n    w_out = Float64[]\n    g1_out = Float64[]\n    g2_out = Float64[]\n    # This function only works for two generators\n    @assert length(generators) == 2\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    for c_g1_scale in scale\n        @objective(\n            model,\n            Min,\n            c_g1_scale * generators[1].variable_cost * g[1] +\n            generators[2].variable_cost * g[2] +\n            wind.variable_cost * w,\n        )\n        optimize!(model)\n        push!(obj_out, objective_value(model))\n        push!(w_out, value(w))\n        push!(g1_out, value(g[1]))\n        push!(g2_out, value(g[2]))\n    end\n    df = DataFrames.DataFrame(;\n        scale = scale,\n        dispatch_G1 = g1_out,\n        dispatch_G2 = g2_out,\n        dispatch_wind = w_out,\n        spillage_wind = scenario.wind .- w_out,\n        total_cost = obj_out,\n    )\n    return df\nend\n\nstart = time()\ninplace_df = solve_economic_dispatch_inplace(\n    generators,\n    wind_generator,\n    scenario,\n    0.5:0.1:3.0,\n)\nprint(string(\"elapsed time: \", time() - start, \" seconds\"))","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For small models, adjusting specific constraints or the objective function is sometimes faster and sometimes slower than re-building the entire model. However, as the problem size increases, updating the model in-place is usually faster.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"inplace_df","category":"page"},{"location":"tutorials/applications/power_systems/#Inefficient-usage-of-wind-generators","page":"Power Systems","title":"Inefficient usage of wind generators","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The economic dispatch problem does not perform commitment decisions and, thus, assumes that all generators must be dispatched at least at their minimum power output limit. This approach is not cost efficient and may lead to absurd decisions. For example, if d = sum_i in I g^min_i, the wind power injection must be zero, that is, all available wind generation is spilled, to meet the minimum power output constraints on generators.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following example, we adjust the total demand and observed how it affects wind spillage.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"demand_scale_df = DataFrames.DataFrame(;\n    demand = Float64[],\n    dispatch_G1 = Float64[],\n    dispatch_G2 = Float64[],\n    dispatch_wind = Float64[],\n    spillage_wind = Float64[],\n    total_cost = Float64[],\n)\n\nfunction scale_demand(scenario, scale)\n    return Scenario(scale * scenario.demand, scenario.wind)\nend\n\nfor demand_scale in 0.2:0.1:1.4\n    new_scenario = scale_demand(scenario, demand_scale)\n    sol = solve_economic_dispatch(generators, wind_generator, new_scenario)\n    push!(\n        demand_scale_df,\n        (\n            new_scenario.demand,\n            sol.g[1],\n            sol.g[2],\n            sol.w,\n            sol.wind_spill,\n            sol.total_cost,\n        ),\n    )\nend\n\ndemand_scale_df","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"dispatch_plot = StatsPlots.@df(\n    demand_scale_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_G1, :dispatch_G2],\n        labels = [\"G1\" \"G2\"],\n        title = \"Thermal Dispatch\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand\",\n        ylabel = \"Dispatch [MW]\",\n    ),\n)\n\nwind_plot = StatsPlots.@df(\n    demand_scale_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_wind, :spillage_wind],\n        labels = [\"Dispatch\" \"Spillage\"],\n        title = \"Wind\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand [MW]\",\n        ylabel = \"Energy [MW]\",\n    ),\n)\n\nPlots.plot(dispatch_plot, wind_plot)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"This particular drawback can be overcome by introducing binary decisions on the \"on/off\" status of generators. This model is called unit commitment and considered later in these notes.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For further reading on the interplay between wind generation and the minimum power output constraints of generators, we refer interested readers to R. Baldick, \"Wind and energy markets: a case study of Texas,\" IEEE Systems Journal, vol. 6, pp. 27-34, 2012.","category":"page"},{"location":"tutorials/applications/power_systems/#Unit-commitment","page":"Power Systems","title":"Unit commitment","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The Unit Commitment (UC) model can be obtained from ED model by introducing binary variable associated with each generator. This binary variable can attain two values: if it is \"1,\" the generator is synchronized and, thus, can be dispatched, otherwise, that is, if the binary variable is \"0,\" that generator is not synchronized and its power output is set to 0.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"To obtain the mathematical formulation of the UC model, we will modify the constraints of the ED model as follows:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"g^min_i cdot u_ti leq g_i leq g^max_i cdot u_ti","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"where u_i in 01 In this constraint, if u_i = 0, then g_i  = 0. On the other hand, if u_i = 1, then g^min_i leq g_i leq g^max_i.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"For further reading on the UC problem we refer interested readers to G. Morales-Espana, J. M. Latorre, and A. Ramos, \"Tight and Compact MILP Formulation for the Thermal Unit Commitment Problem,\" IEEE Transactions on Power Systems, vol. 28, pp. 4897-4908, 2013.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"In the following example we convert the ED model explained above to the UC model.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"function solve_unit_commitment(generators::Vector, wind, scenario)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    N = length(generators)\n    @variable(model, 0 <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @constraint(model, sum(g[i] for i in 1:N) + w == scenario.demand)\n    # !!! New: add binary on-off variables for each generator\n    @variable(model, u[i = 1:N], Bin)\n    @constraint(model, [i = 1:N], g[i] <= generators[i].max * u[i])\n    @constraint(model, [i = 1:N], g[i] >= generators[i].min * u[i])\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * g[i] for i in 1:N) +\n        wind.variable_cost * w +\n        # !!! new\n        sum(generators[i].fixed_cost * u[i] for i in 1:N)\n    )\n    optimize!(model)\n    status = termination_status(model)\n    if status != OPTIMAL\n        return (status = status,)\n    end\n    return (\n        status = status,\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        u = value.(u),\n        total_cost = objective_value(model),\n    )\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Solve the unit commitment problem","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"solution = solve_unit_commitment(generators, wind_generator, scenario)\n\nprintln(\"Dispatch of Generators: \", solution.g, \" MW\")\nprintln(\"Commitments of Generators: \", solution.u)\nprintln(\"Dispatch of Wind: \", solution.w, \" MW\")\nprintln(\"Wind spillage: \", solution.wind_spill, \" MW\")\nprintln(\"Total cost: \\$\", solution.total_cost)","category":"page"},{"location":"tutorials/applications/power_systems/#Unit-commitment-as-a-function-of-demand","page":"Power Systems","title":"Unit commitment as a function of demand","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"After implementing the unit commitment model, we can now assess the interplay between the minimum power output constraints on generators and wind generation.","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"uc_df = DataFrames.DataFrame(;\n    demand = Float64[],\n    commitment_G1 = Float64[],\n    commitment_G2 = Float64[],\n    dispatch_G1 = Float64[],\n    dispatch_G2 = Float64[],\n    dispatch_wind = Float64[],\n    spillage_wind = Float64[],\n    total_cost = Float64[],\n)\n\nfor demand_scale in 0.2:0.1:1.4\n    new_scenario = scale_demand(scenario, demand_scale)\n    sol = solve_unit_commitment(generators, wind_generator, new_scenario)\n    if sol.status == OPTIMAL\n        push!(\n            uc_df,\n            (\n                new_scenario.demand,\n                sol.u[1],\n                sol.u[2],\n                sol.g[1],\n                sol.g[2],\n                sol.w,\n                sol.wind_spill,\n                sol.total_cost,\n            ),\n        )\n    end\n    println(\"Status: $(sol.status) for demand_scale = $(demand_scale)\")\nend","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"uc_df","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"commitment_plot = StatsPlots.@df(\n    uc_df,\n    Plots.plot(\n        :demand,\n        [:commitment_G1, :commitment_G2],\n        labels = [\"G1\" \"G2\"],\n        title = \"Commitment\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand [MW]\",\n        ylabel = \"Commitment decision {0, 1}\",\n    ),\n)\n\ndispatch_plot = StatsPlots.@df(\n    uc_df,\n    Plots.plot(\n        :demand,\n        [:dispatch_G1, :dispatch_G2, :dispatch_wind],\n        labels = [\"G1\" \"G2\" \"Wind\"],\n        title = \"Dispatch [MW]\",\n        legend = :bottomright,\n        linewidth = 3,\n        xlabel = \"Demand\",\n        ylabel = \"Dispatch [MW]\",\n    ),\n)\n\nPlots.plot(commitment_plot, dispatch_plot)","category":"page"},{"location":"tutorials/applications/power_systems/#Nonlinear-economic-dispatch","page":"Power Systems","title":"Nonlinear economic dispatch","text":"","category":"section"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"As a final example, we modify our economic dispatch problem in two ways:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"The thermal cost function is user-defined\nThe output of the wind is only the square-root of the dispatch","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"import Ipopt\n\n\"\"\"\n    thermal_cost_function(g)\n\nA user-defined thermal cost function in pure-Julia! You can include\nnonlinearities, and even things like control flow.\n\n!!! warning\n    It's still up to you to make sure that the function has a meaningful\n    derivative.\n\"\"\"\nfunction thermal_cost_function(g)\n    if g <= 500\n        return g\n    else\n        return g + 1e-2 * (g - 500)^2\n    end\nend\n\nfunction solve_nonlinear_economic_dispatch(\n    generators::Vector,\n    wind,\n    scenario;\n    silent::Bool = false,\n)\n    model = Model(Ipopt.Optimizer)\n    if silent\n        set_silent(model)\n    end\n    @operator(model, op_tcf, 1, thermal_cost_function)\n    N = length(generators)\n    @variable(model, generators[i].min <= g[i = 1:N] <= generators[i].max)\n    @variable(model, 0 <= w <= scenario.wind)\n    @objective(\n        model,\n        Min,\n        sum(generators[i].variable_cost * op_tcf(g[i]) for i in 1:N) +\n        wind.variable_cost * w,\n    )\n    @constraint(model, sum(g[i] for i in 1:N) + sqrt(w) == scenario.demand)\n    optimize!(model)\n    return (\n        g = value.(g),\n        w = value(w),\n        wind_spill = scenario.wind - value(w),\n        total_cost = objective_value(model),\n    )\nend\n\nsolution =\n    solve_nonlinear_economic_dispatch(generators, wind_generator, scenario)","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"Now let's see how the wind is dispatched as a function of the cost:","category":"page"},{"location":"tutorials/applications/power_systems/","page":"Power Systems","title":"Power Systems","text":"wind_cost = 0.0:1:100\nwind_dispatch = Float64[]\nfor c in wind_cost\n    sol = solve_nonlinear_economic_dispatch(\n        generators,\n        WindGenerator(c),\n        scenario;\n        silent = true,\n    )\n    push!(wind_dispatch, sol.w)\nend\n\nPlots.plot(\n    wind_cost,\n    wind_dispatch;\n    xlabel = \"Cost\",\n    ylabel = \"Dispatch [MW]\",\n    label = false,\n)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"EditURL = \"logistic_regression.jl\"","category":"page"},{"location":"tutorials/conic/logistic_regression/#Logistic-regression","page":"Logistic regression","title":"Logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial was originally contributed by François Pacaud.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"This tutorial shows how to solve a logistic regression problem with JuMP. Logistic regression is a well known method in machine learning, useful when we want to classify binary variables with the help of a given set of features. To this goal, we find the optimal combination of features maximizing the (log)-likelihood onto a training set. From a modern optimization glance, the resulting problem is convex and differentiable. On a modern optimization glance, it is even conic representable.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Formulating-the-logistic-regression-problem","page":"Logistic regression","title":"Formulating the logistic regression problem","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Suppose we have a set of training data-point i = 1 cdots n, where for each i we have a vector of features x_i in mathbbR^p and a categorical observation y_i in -1 1.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"The log-likelihood is given by","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"l(theta) = sum_i=1^n log(dfrac11 + exp(-y_i theta^top x_i))","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"and the optimal theta minimizes the logistic loss function:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"min_theta sum_i=1^n log(1 + exp(-y_i theta^top x_i))","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Most of the time, instead of solving directly the previous optimization problem, we prefer to add a regularization term:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"min_theta sum_i=1^n log(1 + exp(-y_i theta^top x_i)) + lambda  theta ","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"with lambda in mathbbR_+ a penalty and  a norm function. By adding such a regularization term, we avoid overfitting on the training set and usually achieve a greater score in cross-validation.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Reformulation-as-a-conic-optimization-problem","page":"Logistic regression","title":"Reformulation as a conic optimization problem","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"By introducing auxiliary variables t_1 cdots t_n and r, the optimization problem is equivalent to","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_t r theta  sum_i=1^n t_i + lambda r \ntextsubject to   quad t_i geq log(1 + exp(- y_i theta^top x_i)) \n                    quad r geq theta\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Now, the trick is to reformulate the constraints t_i geq log(1 + exp(- y_i theta^top x_i)) with the help of the exponential cone","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"K_exp =  (x y z) in mathbbR^3   y exp(x  y) leq z  ","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Indeed, by passing to the exponential, we see that for all i=1 cdots n, the constraint t_i geq log(1 + exp(- y_i theta^top x_i)) is equivalent to","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"exp(-t_i) + exp(u_i - t_i) leq 1","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"with u_i = -y_i theta^top x_i. Then, by adding two auxiliary variables z_i1 and z_i2 such that z_i1 geq exp(u_i-t_i) and z_i2 geq exp(-t_i), we get the equivalent formulation","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"left\nbeginaligned\n(u_i -t_i  1 z_i1)  in  K_exp  \n(-t_i  1 z_i2)  in  K_exp  \nz_i1 + z_i2  leq  1\nendaligned\nright","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"In this setting, the conic version of the logistic regression problems writes out","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_t z r theta   sum_i=1^n t_i + lambda r \ntextsubject to   quad  (u_i -t_i  1 z_i1)  in  K_exp  \n                    quad  (-t_i  1 z_i2)  in  K_exp  \n                    quad  z_i1 + z_i2  leq  1 \n                    quad u_i = -y_i x_i^top theta \n                    quad r geq theta\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"and thus encompasses 3n + p + 1 variables and 3n + 1 constraints (u_i = -y_i theta^top x_i is only a virtual constraint used to clarify the notation). Thus, if n gg 1, we get a large number of variables and constraints.","category":"page"},{"location":"tutorials/conic/logistic_regression/#Fitting-logistic-regression-with-a-conic-solver","page":"Logistic regression","title":"Fitting logistic regression with a conic solver","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"It is now time to pass to the implementation. We choose SCS as a conic solver.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"using JuMP\nimport Random\nimport SCS","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Random.seed!(2713);\nnothing #hide","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We start by implementing a function to generate a fake dataset, and where we could tune the correlation between the feature variables. The function is a direct transcription of the one used in this blog post.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function generate_dataset(n_samples = 100, n_features = 10; shift = 0.0)\n    X = randn(n_samples, n_features)\n    w = randn(n_features)\n    y = sign.(X * w)\n    X .+= 0.8 * randn(n_samples, n_features) # add noise\n    X .+= shift # shift the points in the feature space\n    X = hcat(X, ones(n_samples, 1))\n    return X, y\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We write a softplus function to formulate each constraint t geq log(1 + exp(u)) with two exponential cones.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function softplus(model, t, u)\n    z = @variable(model, [1:2], lower_bound = 0.0)\n    @constraint(model, sum(z) <= 1.0)\n    @constraint(model, [u - t, 1, z[1]] in MOI.ExponentialCone())\n    @constraint(model, [-t, 1, z[2]] in MOI.ExponentialCone())\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/#\\ell_2-regularized-logistic-regression","page":"Logistic regression","title":"ell_2 regularized logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"Then, with the help of the softplus function, we could write our optimization model. In the ell_2 regularization case, the constraint r geq theta_2 rewrites as a second order cone constraint.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function build_logit_model(X, y, λ)\n    n, p = size(X)\n    model = Model()\n    @variable(model, θ[1:p])\n    @variable(model, t[1:n])\n    for i in 1:n\n        u = -(X[i, :]' * θ) * y[i]\n        softplus(model, t[i], u)\n    end\n    # Add ℓ2 regularization\n    @variable(model, 0.0 <= reg)\n    @constraint(model, [reg; θ] in SecondOrderCone())\n    # Define objective\n    @objective(model, Min, sum(t) + λ * reg)\n    return model\nend","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We generate the dataset.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"warning: Warning\nBe careful here, for large n and p SCS could fail to converge.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"n, p = 200, 10\nX, y = generate_dataset(n, p; shift = 10.0);\n\n# We could now solve the logistic regression problem\nλ = 10.0\nmodel = build_logit_model(X, y, λ)\nset_optimizer(model, SCS.Optimizer)\nset_silent(model)\nJuMP.optimize!(model)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"θ♯ = JuMP.value.(model[:θ])","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"It appears that the speed of convergence is not that impacted by the correlation of the dataset, nor by the penalty lambda.","category":"page"},{"location":"tutorials/conic/logistic_regression/#\\ell_1-regularized-logistic-regression","page":"Logistic regression","title":"ell_1 regularized logistic regression","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"We now formulate the logistic problem with a ell_1 regularization term. The ell_1 regularization ensures sparsity in the optimal solution of the resulting optimization problem. Luckily, the ell_1 norm is implemented as a set in MathOptInterface. Thus, we could formulate the sparse logistic regression problem with the help of a MOI.NormOneCone set.","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"function build_sparse_logit_model(X, y, λ)\n    n, p = size(X)\n    model = Model()\n    @variable(model, θ[1:p])\n    @variable(model, t[1:n])\n    for i in 1:n\n        u = -(X[i, :]' * θ) * y[i]\n        softplus(model, t[i], u)\n    end\n    # Add ℓ1 regularization\n    @variable(model, 0.0 <= reg)\n    @constraint(model, [reg; θ] in MOI.NormOneCone(p + 1))\n    # Define objective\n    @objective(model, Min, sum(t) + λ * reg)\n    return model\nend\n\n# Auxiliary function to count non-null components:\ncount_nonzero(v::Vector; tol = 1e-6) = sum(abs.(v) .>= tol)\n\n# We solve the sparse logistic regression problem on the same dataset as before.\nλ = 10.0\nsparse_model = build_sparse_logit_model(X, y, λ)\nset_optimizer(sparse_model, SCS.Optimizer)\nset_silent(sparse_model)\nJuMP.optimize!(sparse_model)","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"θ♯ = JuMP.value.(sparse_model[:θ])\nprintln(\n    \"Number of non-zero components: \",\n    count_nonzero(θ♯),\n    \" (out of \",\n    p,\n    \" features)\",\n)","category":"page"},{"location":"tutorials/conic/logistic_regression/#Extensions","page":"Logistic regression","title":"Extensions","text":"","category":"section"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"A direct extension would be to consider the sparse logistic regression with hard thresholding, which, on contrary to the soft version using a ell_1 regularization, adds an explicit cardinality constraint in its formulation:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"beginaligned\nmin_theta   sum_i=1^n log(1 + exp(-y_i theta^top x_i)) + lambda  theta _2^2 \ntextsubject to   quad  theta _0 = k\nendaligned","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"where k is the maximum number of non-zero components in the vector theta, and _0 is the ell_0 pseudo-norm:","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":" x_0 = i   x_i neq 0","category":"page"},{"location":"tutorials/conic/logistic_regression/","page":"Logistic regression","title":"Logistic regression","text":"The cardinality constraint theta_0 leq k could be reformulated with binary variables. Thus the hard sparse regression problem could be solved by any solver supporting mixed integer conic problems.","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/motivation.md\"","category":"page"},{"location":"moi/background/motivation/#Motivation","page":"Motivation","title":"Motivation","text":"","category":"section"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"MathOptInterface (MOI) is a replacement for MathProgBase, the first-generation abstraction layer for mathematical optimization previously used by JuMP and Convex.jl.","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"To address a number of limitations of MathProgBase, MOI is designed to:","category":"page"},{"location":"moi/background/motivation/","page":"Motivation","title":"Motivation","text":"Be simple and extensible\nunifying linear, quadratic, and conic optimization,\nseamlessly facilitating extensions to essentially arbitrary constraints and functions (for example, indicator constraints, complementarity constraints, and piecewise-linear functions)\nBe fast\nby allowing access to a solver's in-memory representation of a problem without writing intermediate files (when possible)\nby using multiple dispatch and avoiding requiring containers of non-concrete types\nAllow a solver to return multiple results (for example, a pool of solutions)\nAllow a solver to return extra arbitrary information via attributes (for example, variable- and constraint-wise membership in an irreducible inconsistent subset for infeasibility analysis)\nProvide a greatly expanded set of status codes explaining what happened during the optimization procedure\nEnable a solver to more precisely specify which problem classes it supports\nEnable both primal and dual warm starts\nEnable adding and removing both variables and constraints by indices that are not required to be consecutive\nEnable any modification that the solver supports to an existing model\nAvoid requiring the solver wrapper to store an additional copy of the problem data","category":"page"},{"location":"tutorials/conic/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Conic programs are a class of convex nonlinear optimization problems which use cones to represent the nonlinearities. They have the form:","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n     min_x in mathbbR^n  f_0(x) \n     textst  f_j(x) in mathcalS_j   j = 1 ldots m\nendalign","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer conic programs (MICPs) are extensions of conic programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/conic/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"JuMP supports a range of conic solvers, although support differs on what types of cones each solver supports. In the list of Supported solvers, \"SOCP\" denotes solvers supporting second-order cones and \"SDP\" denotes solvers supporting semidefinite cones. In addition, solvers such as SCS and Mosek have support for the exponential cone. Moreover, due to the bridging system in MathOptInterface, many of these solvers support a much wider range of exotic cones than they natively support. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"tip: Tip\nDuality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual instead of the primal. If performance is an issue, see the Dualization tutorial for more details.","category":"page"},{"location":"tutorials/conic/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/conic/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nExperiment design\nLogistic regression\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling conic programs. Look here if you are stuck trying to formulate a problem as a conic program.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/manipulating_expressions.md\"","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Manipulating-expressions","page":"Manipulating expressions","title":"Manipulating expressions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This guide highlights a syntactically appealing way to build expressions at the MOI level, but also to look at their contents. It may be especially useful when writing models or bridge code.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-functions","page":"Manipulating expressions","title":"Creating functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This section details the ways to create functions with MathOptInterface.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-scalar-affine-functions","page":"Manipulating expressions","title":"Creating scalar affine functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"The simplest scalar function is simply a variable:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> x = MOI.add_variable(model) # Create the variable x\nMOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"This type of function is extremely simple; to express more complex functions, other types must be used. For instance, a ScalarAffineFunction is a sum of linear terms (a factor times a variable) and a constant. Such an object can be built using the standard constructor:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1, x)], 2) # x + 2\n(2) + (1) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"However, you can also use operators to build the same scalar function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> f = x + 2\n(2) + (1) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-scalar-quadratic-functions","page":"Manipulating expressions","title":"Creating scalar quadratic functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Scalar quadratic functions are stored in ScalarQuadraticFunction objects, in a way that is highly similar to scalar affine functions. You can obtain a quadratic function as a product of affine functions:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> 1 * x * x\n(0) + 1.0 MOI.VariableIndex(1)²\n\njulia> f * f  # (x + 2)²\n(4) + (2) MOI.VariableIndex(1) + (2) MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²\n\njulia> f^2  # (x + 2)² too\n(4) + (2) MOI.VariableIndex(1) + (2) MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Creating-vector-functions","page":"Manipulating expressions","title":"Creating vector functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"A vector function is a function with several values, irrespective of the number of input variables. Similarly to scalar functions, there are three main types of vector functions: VectorOfVariables, VectorAffineFunction, and VectorQuadraticFunction.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"The easiest way to create a vector function is to stack several scalar functions using Utilities.vectorize. It takes a vector as input, and the generated vector function (of the most appropriate type) has each dimension corresponding to a dimension of the vector.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> g = MOI.Utilities.vectorize([f, 2 * f])\n┌                              ┐\n│(2) + (1) MOI.VariableIndex(1)│\n│(4) + (2) MOI.VariableIndex(1)│\n└                              ┘","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"warning: Warning\nUtilities.vectorize only takes a vector of similar scalar functions: you cannot mix VariableIndex and ScalarAffineFunction, for instance. In practice, it means that Utilities.vectorize([x, f]) does not work; you should rather use Utilities.vectorize([1 * x, f]) instead to only have ScalarAffineFunction objects.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Canonicalizing-functions","page":"Manipulating expressions","title":"Canonicalizing functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"In more advanced use cases, you might need to ensure that a function is \"canonical.\" Functions are stored as an array of terms, but there is no check that these terms are redundant: a ScalarAffineFunction object might have two terms with the same variable, like x + x + 1. These terms could be merged without changing the semantics of the function: 2x + 1.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Working with these objects might be cumbersome. Canonicalization helps maintain redundancy to zero.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.is_canonical checks whether a function is already in its canonical form:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.is_canonical(f + f) # (x + 2) + (x + 2) is stored as x + x + 4\nfalse","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.canonical returns the equivalent canonical version of the function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.canonical(f + f) # Returns 2x + 4\n(4) + (2) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Exploring-functions","page":"Manipulating expressions","title":"Exploring functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"At some point, you might need to dig into a function, for instance to map it into solver constructs.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/#Vector-functions","page":"Manipulating expressions","title":"Vector functions","text":"","category":"section"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"Utilities.scalarize returns a vector of scalar functions from a vector function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.Utilities.scalarize(g) # Returns a vector [f, 2 * f].\n2-element Vector{MathOptInterface.ScalarAffineFunction{Int64}}:\n (2) + (1) MOI.VariableIndex(1)\n (4) + (2) MOI.VariableIndex(1)","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"note: Note\nUtilities.eachscalar returns an iterator on the dimensions, which serves the same purpose as Utilities.scalarize.","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"output_dimension returns the number of dimensions of the output of a function:","category":"page"},{"location":"moi/tutorials/manipulating_expressions/","page":"Manipulating expressions","title":"Manipulating expressions","text":"julia> MOI.output_dimension(g)\n2","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"EditURL = \"https://github.com/jump-dev/MosekTools.jl/blob/v0.15.1/README.md\"","category":"page"},{"location":"packages/MosekTools/#MosekTools.jl","page":"jump-dev/MosekTools.jl","title":"MosekTools.jl","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is the MathOptInterface.jl implementation for the MOSEK solver.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The low-level solver API for MOSEK is found in the package Mosek.jl.","category":"page"},{"location":"packages/MosekTools/#Affiliation","page":"jump-dev/MosekTools.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is maintained by the JuMP community and is not officially supported by MOSEK. However, Mosek.jl is an officially supported product of MOSEK.","category":"page"},{"location":"packages/MosekTools/#License","page":"jump-dev/MosekTools.jl","title":"License","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MosekTools.jl is licensed under the MIT License.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/MosekTools/#Installation","page":"jump-dev/MosekTools.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The latest release of this package and the master branch are to be used with the latest release of Mosek.jl (which uses MOSEK v10).","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"To use MOSEK v9 (resp. v8), use the v0.12.x (resp. v0.7.x) releases of this package, and the mosekv9 (resp. mosekv8) branch and v1.2.x (resp. v0.9.x) releases of Mosek.jl.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"See the following table for a summary:","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"MOSEK Mosek.jl MosekTools.jl release MosekTools.jl branch\nv10 v10 v0.13 master\nv9 v0.12 v0.12 mosekv9\nv8 v0.9 v0.7 mosekv8","category":"page"},{"location":"packages/MosekTools/#Use-with-JuMP","page":"jump-dev/MosekTools.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"using JuMP\nusing MosekTools\nmodel = Model(Mosek.Optimizer)\nset_attribute(model, \"QUIET\", true)\nset_attribute(model, \"INTPNT_CO_TOL_DFEAS\", 1e-7)","category":"page"},{"location":"packages/MosekTools/#Options","page":"jump-dev/MosekTools.jl","title":"Options","text":"","category":"section"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"The parameter QUIET is a special parameter that when set to true disables all Mosek printing output.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"All other parameters can be found in the Mosek documentation.","category":"page"},{"location":"packages/MosekTools/","page":"jump-dev/MosekTools.jl","title":"jump-dev/MosekTools.jl","text":"Note that the prefix MSK_IPAR_ (for integer parameters), MSK_DPAR_ (for floating point parameters) or MSK_SPAR_ (for string parameters) are optional. If they are not given, they are inferred from the type of the value. For example, in the example above, as 1e-7 is a floating point number, the parameters name used is MSK_DPAR_INTPNT_CO_TOL_DFEAS.","category":"page"},{"location":"developers/style/#Style-guide-and-design-principles","page":"Style Guide","title":"Style guide and design principles","text":"","category":"section"},{"location":"developers/style/#Style-guide","page":"Style Guide","title":"Style guide","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section describes the coding style rules that apply to JuMP code and that we recommend for JuMP models and surrounding Julia code. The motivations for a style guide include:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"conveying best practices for writing readable and maintainable code\nreducing the amount of time spent on bike-shedding by establishing basic naming and formatting conventions\nlowering the barrier for new contributors by codifying the existing practices (for example, you can be more confident your code will pass review if you follow the style guide)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In some cases, the JuMP style guide diverges from the Julia style guide. All such cases will be explicitly noted and justified.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The JuMP style guide adopts many recommendations from the Google style guides.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nThe style guide is always a work in progress, and not all JuMP code follows the rules. When modifying JuMP, please fix the style violations of the surrounding code (that is, leave the code tidier than when you started). If large changes are needed, consider separating them into another PR.","category":"page"},{"location":"developers/style/#JuliaFormatter","page":"Style Guide","title":"JuliaFormatter","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"JuMP uses JuliaFormatter.jl as an auto-formatting tool.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"We use the options contained in .JuliaFormatter.toml.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"To format code, cd to the JuMP directory, then run:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"] add JuliaFormatter@1\nusing JuliaFormatter\nformat(\"docs\")\nformat(\"src\")\nformat(\"test\")","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nA continuous integration check verifies that all PRs made to JuMP have passed the formatter.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The following sections outline extra style guide points that are not fixed automatically by JuliaFormatter.","category":"page"},{"location":"developers/style/#Abstract-types-and-composition","page":"Style Guide","title":"Abstract types and composition","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Specifying types for method arguments is mostly optional in Julia. The benefit of abstract method arguments is that it enables functions and types from one package to be used with functions and types from another package via multiple dispatch.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"However, abstractly typed methods have two main drawbacks:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"It's possible to find out that you are working with unexpected types deep in the call chain, potentially leading to hard-to-diagnose MethodErrors.\nUntyped function arguments can lead to correctness problems if the user's choice of input type does not satisfy the assumptions made by the author of the function.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"As a motivating example, consider the following function:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum(x)\n           y = 0.0\n           for i in 1:length(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum (generic function with 1 method)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This function contains a number of implicit assumptions about the type of x:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"x supports 1-based getindex and implements length\nThe element type of x supports addition with 0.0, and then with the result of x + 0.0.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nAs a motivating example for the second point, VariableRef plus Float64 produces an AffExpr. Do not assume that +(::A, ::B) produces an instance of the type A or B.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"my_sum works as expected if the user passes in Vector{Float64}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"but it doesn't respect input types, for example returning a Float64 if the user passes Vector{Int}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum([1, 2, 3])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"but it throws a MethodError if the user passes String:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum(\"abc\")\nERROR: MethodError: no method matching +(::Float64, ::Char)\n[...]","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This particular MethodError is hard to debug, particularly for new users, because it mentions +, Float64, and Char, none of which were called or passed by the user.","category":"page"},{"location":"developers/style/#Dealing-with-MethodErrors","page":"Style Guide","title":"Dealing with MethodErrors","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section diverges from the Julia style guide, as well as other common guides like SciML. The following suggestions are intended to provide a friendlier experience for novice Julia programmers, at the cost of limiting the power and flexibility of advanced Julia programmers.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Code should follow the MethodError principle:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: The MethodError principle\nA user should see a MethodError only for methods that they called directly.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\n# The user sees a MethodError for _internal_function when calling\n# public_function(\"a string\"). This is not very helpful.\npublic_function(x) = _internal_function(x)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\n# The user sees a MethodError for public_function when calling\n# public_function(\"a string\"). This is easy to understand.\npublic_function(x::Integer) = _internal_function(x)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"If it is hard to provide an error message at the top of the call chain, then the following pattern is also ok:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"_internal_function(x::Integer) = x + 1\nfunction _internal_function(x)\n    error(\n        \"Internal error. This probably means that you called \" *\n        \"`public_function()`s with the wrong type.\",\n    )\nend\npublic_function(x) = _internal_function(x)","category":"page"},{"location":"developers/style/#Dealing-with-correctness","page":"Style Guide","title":"Dealing with correctness","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Dealing with correctness is harder, because Julia has no way of formally specifying interfaces that abstract types must implement. Instead, here are two options that you can use when writing and interacting with generic code:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Option 1: use concrete types and let users extend new methods.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In this option, explicitly restrict input arguments to concrete types that are tested and have been validated for correctness. For example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum_option_1(x::Vector{Float64})\n           y = 0.0\n           for i in 1:length(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum_option_1 (generic function with 1 method)\n\njulia> my_sum_option_1([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Using concrete types satisfies the MethodError principle:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_option_1(\"abc\")\nERROR: MethodError: no method matching my_sum_option_1(::String)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"and it allows other types to be supported in future by defining new methods:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> function my_sum_option_1(x::Array{T,N}) where {T<:Number,N}\n           y = zero(T)\n           for i in eachindex(x)\n               y += x[i]\n           end\n           return y\n       end\nmy_sum_option_1 (generic function with 2 methods)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Importantly, these methods do not have to be defined in the original package.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"info: Info\nSome usage of abstract types is okay. For example, in my_sum_option_1, we allowed the element type, T, to be a subtype of Number. This is fairly safe, but it still has an implicit assumption that T supports zero(T) and +(::T, ::T).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Option 2: program defensively, and validate all assumptions.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"An alternative is to program defensively, and to rigorously document and validate all assumptions that the code makes. In particular:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"All assumptions on abstract types that aren't guaranteed by the definition of the abstract type (for example, optional methods without a fallback) should be documented.\nIf practical, the assumptions should be checked in code, and informative error messages should be provided to the user if the assumptions are not met. In general, these checks may be expensive, so you should prefer to do this once, at the highest level of the call-chain.\nTests should cover for a range of corner cases and argument types.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"\"\"\"\n    test_my_sum_defensive_assumptions(x::AbstractArray{T}) where {T}\n\nTest the assumptions made by `my_sum_defensive`.\n\"\"\"\nfunction test_my_sum_defensive_assumptions(x::AbstractArray{T}) where {T}\n    try\n        # Some types may not define zero.\n        @assert zero(T) isa T\n        # Check iteration supported\n        @assert iterate(x) isa Union{Nothing,Tuple{T,Int}}\n        # Check that + is defined\n        @assert +(zero(T), zero(T)) isa Any\n    catch err\n        error(\n            \"Unable to call my_sum_defensive(::$(typeof(x))) because \" *\n            \"it failed an internal assumption\",\n        )\n    end\n    return\nend\n\n\"\"\"\n    my_sum_defensive(x::AbstractArray{T}) where {T}\n\nReturn the sum of the elements in the abstract array `x`.\n\n## Assumptions\n\nThis function makes the following assumptions:\n\n * That `zero(T)` is defined\n * That `x` supports the iteration interface\n * That  `+(::T, ::T)` is defined\n\"\"\"\nfunction my_sum_defensive(x::AbstractArray{T}) where {T}\n    test_my_sum_defensive_assumptions(x)\n    y = zero(T)\n    for xi in x\n        y += xi\n    end\n    return y\nend\n\n# output\n\nmy_sum_defensive","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This function works on Vector{Float64}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive([1.0, 2.0, 3.0])\n6.0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"as well as Matrix{Rational{Int}}:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive([(1//2) + (4//3)im; (6//5) + (7//11)im])\n17//10 + 65//33*im","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"and it throws an error when the assumptions aren't met:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"julia> my_sum_defensive(['a', 'b', 'c'])\nERROR: Unable to call my_sum_defensive(::Vector{Char}) because it failed an internal assumption\n[...]","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"As an alternative, you may choose not to call test_my_sum_defensive_assumptions within my_sum_defensive, and instead ask users of my_sum_defensive to call it in their tests.","category":"page"},{"location":"developers/style/#Juxtaposed-multiplication","page":"Style Guide","title":"Juxtaposed multiplication","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Only use juxtaposed multiplication when the right-hand side is a symbol.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"2x  # Acceptable if there are space constraints.\n2 * x  # This is preferred if space is not an issue.\n2 * (x + 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"2(x + 1)","category":"page"},{"location":"developers/style/#Empty-vectors","page":"Style Guide","title":"Empty vectors","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For a type T, T[] and Vector{T}() are equivalent ways to create an empty vector with element type T. Prefer T[] because it is more concise.","category":"page"},{"location":"developers/style/#Comments","page":"Style Guide","title":"Comments","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For non-native speakers and for general clarity, comments in code must be proper English sentences with appropriate punctuation.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"# This is a comment demonstrating a good comment.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"# a bad comment","category":"page"},{"location":"developers/style/#JuMP-macro-syntax","page":"Style Guide","title":"JuMP macro syntax","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For consistency, always use parentheses.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable model x >= 0","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For consistency, always use constant * variable as opposed to variable * constant. This makes it easier to read models in ambiguous cases like a * x.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"a = 4\n@constraint(model, 3 * x <= 1)\n@constraint(model, a * x <= 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"a = 4\n@constraint(model, x * 3 <= 1)\n@constraint(model, x * a <= 1)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"In order to reduce boilerplate code, prefer the plural form of macros over lots of repeated calls to singular forms.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Good:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variables(model, begin\n    x >= 0\n    y >= 1\n    z <= 2\nend)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Bad:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0)\n@variable(model, y >= 1)\n@variable(model, z <= 2)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"An exception is made for calls with many keyword arguments, since these need to be enclosed in parentheses in order to parse properly.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Acceptable:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x >= 0, start = 0.0, base_name = \"my_x\")\n@variable(model, y >= 1, start = 2.0)\n@variable(model, z <= 2, start = -1.0)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Also acceptable:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variables(model, begin\n    x >= 0, (start = 0.0, base_name = \"my_x\")\n    y >= 1, (start = 2.0)\n    z <= 2, (start = -1.0)\nend)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"While we always use in for for-loops, it is acceptable to use = in the container declarations of JuMP macros.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Okay:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x[i=1:3])","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Also okay:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@variable(model, x[i in 1:3])","category":"page"},{"location":"developers/style/#Naming","page":"Style Guide","title":"Naming","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module SomeModule end\nfunction some_function end\nconst SOME_CONSTANT = ...\nstruct SomeStruct\n  some_field::SomeType\nend\n@enum SomeEnum ENUM_VALUE_A ENUM_VALUE_B\nsome_local_variable = ...\nsome_file.jl # Except for ModuleName.jl.","category":"page"},{"location":"developers/style/#Exported-and-non-exported-names","page":"Style Guide","title":"Exported and non-exported names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Begin private module level functions and constants with an underscore. All other objects in the scope of a module should be exported. (See JuMP.jl for an example of how to do this.)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Names beginning with an underscore should only be used for distinguishing between exported (public) and non-exported (private) objects. Therefore, never begin the name of a local variable with an underscore.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module MyModule\n\nexport public_function, PUBLIC_CONSTANT\n\nfunction _private_function()\n    local_variable = 1\n    return\nend\n\nfunction public_function end\n\nconst _PRIVATE_CONSTANT = 3.14159\nconst PUBLIC_CONSTANT = 1.41421\n\nend","category":"page"},{"location":"developers/style/#Use-of-underscores-within-names","page":"Style Guide","title":"Use of underscores within names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The Julia style guide recommends avoiding underscores \"when readable,\" for example, haskey, isequal, remotecall, and remotecall_fetch. This convention creates the potential for unnecessary bikeshedding and also forces the user to recall the presence/absence of an underscore, for example, \"was that argument named basename or base_name?\". For consistency, always use underscores in variable names and function names to separate words.","category":"page"},{"location":"developers/style/#Use-of-!","page":"Style Guide","title":"Use of !","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Julia has a convention of appending ! to a function name if the function modifies its arguments. We recommend to:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Omit ! when the name itself makes it clear that modification is taking place, for example, add_constraint and set_name. We depart from the Julia style guide because ! does not provide a reader with any additional information in this case, and adherence to this convention is not uniform even in base Julia itself (consider Base.println and Base.finalize).\nUse ! in all other cases. In particular it can be used to distinguish between modifying and non-modifying variants of the same function like scale and scale!.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Note that ! is not a self-documenting feature because it is still ambiguous which arguments are modified when multiple arguments are present. Be sure to document which arguments are modified in the method's docstring.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"See also the Julia style guide recommendations for ordering of function arguments.","category":"page"},{"location":"developers/style/#Abbreviations","page":"Style Guide","title":"Abbreviations","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Abbreviate names to make the code more readable, not to save typing. Don't arbitrarily delete letters from a word to abbreviate it (for example, indx). Use abbreviations consistently within a body of code (for example, do not mix con and constr, idx and indx).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Common abbreviations:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"num for number\ncon for constraint","category":"page"},{"location":"developers/style/#No-one-letter-variable-names","page":"Style Guide","title":"No one-letter variable names","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Where possible, avoid one-letter variable names.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use model = Model() instead of m = Model()","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Exceptions are made for indices in loops.","category":"page"},{"location":"developers/style/#@enum-vs.-Symbol","page":"Style Guide","title":"@enum vs. Symbol","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"The @enum macro lets you define types with a finite number of values that are explicitly enumerated (like enum in C/C++). Symbols are lightweight strings that are used to represent identifiers in Julia (for example, :x).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"@enum provides type safety and can have docstrings attached to explain the possible values. Use @enums when applicable, for example, for reporting statuses. Use strings to provide long-form additional information like error messages.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use of Symbol should typically be reserved for identifiers, for example, for lookup in the JuMP model (model[:my_variable]).","category":"page"},{"location":"developers/style/#using-vs.-import","page":"Style Guide","title":"using vs. import","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"using ModuleName brings all symbols exported by the module ModuleName into scope, while import ModuleName brings only the module itself into scope. (See the Julia manual) for examples and more details.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"For the same reason that from <module> import * is not recommended in python (PEP 8), avoid using ModuleName except in throw-away scripts or at the REPL. The using statement makes it harder to track where symbols come from and exposes the code to ambiguities when two modules export the same symbol.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Prefer using ModuleName: x, p to import ModuleName.x, ModuleName.p and import MyModule: x, p because the import versions allow method extension without qualifying with the module name.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Similarly, using ModuleName: ModuleName is an acceptable substitute for import ModuleName, because it does not bring all symbols exported by ModuleName into scope. However, we prefer import ModuleName for consistency.","category":"page"},{"location":"developers/style/#Documentation","page":"Style Guide","title":"Documentation","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"This section describes the writing style that should be used when writing documentation for JuMP (and supporting packages).","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"We can recommend the documentation style guides by Divio, Google, and Write the Docs as general reading for those writing documentation. This guide delegates a thorough handling of the topic to those guides and instead elaborates on the points more specific to Julia and documentation that use Documenter.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Be concise\nUse lists instead of long sentences\nUse numbered lists when describing a sequence, for example, (1) do X, (2) then Y\nUse bullet points when the items are not ordered\nExample code should be covered by doctests\nWhen a word is a Julia symbol and not an English word, enclose it with backticks. In addition, if it has a docstring in this doc add a link using @ref. If it is a plural, add the \"s\" after the closing backtick. For example,\n[`VariableRef`](@ref)s\nUse @meta blocks for TODOs and other comments that shouldn't be visible to readers. For example,\n```@meta\n# TODO: Mention also X, Y, and Z.\n```","category":"page"},{"location":"developers/style/#Docstrings","page":"Style Guide","title":"Docstrings","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Every exported object needs a docstring\nAll examples in docstrings should be jldoctests\nAlways use complete English sentences with proper punctuation\nDo not terminate lists with punctuation (for example, as in this doc)","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Here is an example:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"\"\"\"\n    signature(args; kwargs...)\n\nShort sentence describing the function.\n\nOptional: add a slightly longer paragraph describing the function.\n\n## Notes\n\n - List any notes that the user should be aware of\n\n## Examples\n\n```jldoctest\njulia> 1 + 1\n2\n```\n\"\"\"","category":"page"},{"location":"developers/style/#Testing","page":"Style Guide","title":"Testing","text":"","category":"section"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Use a module to encapsulate tests, and structure all tests as functions. This avoids leaking local variables between tests.","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Here is a basic skeleton:","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"module TestPkg\n\nusing Test\n\nfunction runtests()\n    for name in names(@__MODULE__; all = true)\n        if startswith(\"$(name)\", \"test_\")\n            @testset \"$(name)\" begin\n                getfield(@__MODULE__, name)()\n            end\n        end\n    end\n    return\nend\n\n_helper_function() = 2\n\nfunction test_addition()\n    @test 1 + 1 == _helper_function()\n    return\nend\n\nend # module TestPkg\n\nTestPkg.runtests()","category":"page"},{"location":"developers/style/","page":"Style Guide","title":"Style Guide","text":"Break the tests into multiple files, with one module per file, so that subsets of the codebase can be tested by calling include with the relevant file.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Test/overview.md\"","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Test/overview/#test_module","page":"Overview","title":"The Test submodule","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"The Test submodule provides tools to help solvers implement unit tests in order to ensure they implement the MathOptInterface API correctly, and to check for solver-correctness.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"We use a centralized repository of tests, so that if we find a bug in one solver, instead of adding a test to that particular repository, we add it here so that all solvers can benefit.","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-test-a-solver","page":"Overview","title":"How to test a solver","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"The skeleton below can be used for the wrapper test file of a solver named FooBar.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"# ============================ /test/MOI_wrapper.jl ============================\nmodule TestFooBar\n\nimport FooBar\nusing Test\n\nimport MathOptInterface as MOI\n\nconst OPTIMIZER = MOI.instantiate(\n    MOI.OptimizerWithAttributes(FooBar.Optimizer, MOI.Silent() => true),\n)\n\nconst BRIDGED = MOI.instantiate(\n    MOI.OptimizerWithAttributes(FooBar.Optimizer, MOI.Silent() => true),\n    with_bridge_type = Float64,\n)\n\n# See the docstring of MOI.Test.Config for other arguments.\nconst CONFIG = MOI.Test.Config(\n    # Modify tolerances as necessary.\n    atol = 1e-6,\n    rtol = 1e-6,\n    # Use MOI.LOCALLY_SOLVED for local solvers.\n    optimal_status = MOI.OPTIMAL,\n    # Pass attributes or MOI functions to `exclude` to skip tests that\n    # rely on this functionality.\n    exclude = Any[MOI.VariableName, MOI.delete],\n)\n\n\"\"\"\n    runtests()\n\nThis function runs all functions in the this Module starting with `test_`.\n\"\"\"\nfunction runtests()\n    for name in names(@__MODULE__; all = true)\n        if startswith(\"$(name)\", \"test_\")\n            @testset \"$(name)\" begin\n                getfield(@__MODULE__, name)()\n            end\n        end\n    end\nend\n\n\"\"\"\n    test_runtests()\n\nThis function runs all the tests in MathOptInterface.Test.\n\nPass arguments to `exclude` to skip tests for functionality that is not\nimplemented or that your solver doesn't support.\n\"\"\"\nfunction test_runtests()\n    MOI.Test.runtests(\n        BRIDGED,\n        CONFIG,\n        exclude = [\n            \"test_attribute_NumberOfThreads\",\n            \"test_quadratic_\",\n        ],\n        # This argument is useful to prevent tests from failing on future\n        # releases of MOI that add new tests. Don't let this number get too far\n        # behind the current MOI release though. You should periodically check\n        # for new tests to fix bugs and implement new features.\n        exclude_tests_after = v\"0.10.5\",\n    )\n    return\nend\n\n\"\"\"\n    test_SolverName()\n\nYou can also write new tests for solver-specific functionality. Write each new\ntest as a function with a name beginning with `test_`.\n\"\"\"\nfunction test_SolverName()\n    @test MOI.get(FooBar.Optimizer(), MOI.SolverName()) == \"FooBar\"\n    return\nend\n\nend # module TestFooBar\n\n# This line at tne end of the file runs all the tests!\nTestFooBar.runtests()","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Then modify your runtests.jl file to include the MOI_wrapper.jl file:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"# ============================ /test/runtests.jl ============================\n\nusing Test\n\n@testset \"MOI\" begin\n    include(\"test/MOI_wrapper.jl\")\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"info: Info\nThe optimizer BRIDGED constructed with instantiate automatically bridges constraints that are not supported by OPTIMIZER using the bridges listed in Bridges. It is recommended for an implementation of MOI to only support constraints that are natively supported by the solver and let bridges transform the constraint to the appropriate form. For this reason it is expected that tests may not pass if OPTIMIZER is used instead of BRIDGED.","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-debug-a-failing-test","page":"Overview","title":"How to debug a failing test","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"When writing a solver, it's likely that you will initially fail many tests. Some failures will be bugs, but other failures you may choose to exclude.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"There are two ways to exclude tests:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Exclude tests whose names contain a string using:\nMOI.Test.runtests(\n    model,\n    config;\n    exclude = String[\"test_to_exclude\", \"test_conic_\"],\n)\nThis will exclude tests whose name contains either of the two strings provided.\nExclude tests which rely on specific functionality using:\nMOI.Test.Config(exclude = Any[MOI.VariableName, MOI.optimize!])\nThis will exclude tests which use the MOI.VariableName attribute, or which call MOI.optimize!.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Each test that fails can be independently called as:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"model = FooBar.Optimizer()\nconfig = MOI.Test.Config()\nMOI.empty!(model)\nMOI.Test.test_category_name_that_failed(model, config)","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"You can look-up the source code of the test that failed by searching for it in the src/Test/test_category.jl file.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nEach test function also has a docstring that explains what the test is for. Use ? MOI.Test.test_category_name_that_failed from the REPL to read it.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Periodically, you should re-run excluded tests to see if they now pass. The easiest way to do this is to swap the exclude keyword argument of runtests to include. For example:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"MOI.Test.runtests(\n    model,\n    config;\n    exclude = String[\"test_to_exclude\", \"test_conic_\"],\n)","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"becomes","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"MOI.Test.runtests(\n    model,\n    config;\n    include = String[\"test_to_exclude\", \"test_conic_\"],\n)","category":"page"},{"location":"moi/submodules/Test/overview/#How-to-add-a-test","page":"Overview","title":"How to add a test","text":"","category":"section"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"To detect bugs in solvers, we add new tests to MOI.Test.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"As an example, ECOS errored calling optimize! twice in a row. (See ECOS.jl PR #72.) We could add a test to ECOS.jl, but that would only stop us from re-introducing the bug to ECOS.jl in the future, but it would not catch other solvers in the ecosystem with the same bug. Instead, if we add a test to MOI.Test, then all solvers will also check that they handle a double optimize call.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"For this test, we care about correctness, rather than performance. therefore, we don't expect solvers to efficiently decide that they have already solved the problem, only that calling optimize! twice doesn't throw an error or give the wrong answer.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 1","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Install the MathOptInterface julia package in dev mode:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"julia> ]\n(@v1.6) pkg> dev MathOptInterface","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 2","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"From here on, proceed with making the following changes in the ~/.julia/dev/MathOptInterface folder (or equivalent dev path on your machine).","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 3","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Since the double-optimize error involves solving an optimization problem, add a new test to src/Test/test_solve.jl:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"\"\"\"\n    test_unit_optimize!_twice(model::MOI.ModelLike, config::Config)\n\nTest that calling `MOI.optimize!` twice does not error.\n\nThis problem was first detected in ECOS.jl PR#72:\nhttps://github.com/jump-dev/ECOS.jl/pull/72\n\"\"\"\nfunction test_unit_optimize!_twice(\n    model::MOI.ModelLike,\n    config::Config{T},\n) where {T}\n    # Use the `@requires` macro to check conditions that the test function\n    # requires to run. Models failing this `@requires` check will silently skip\n    # the test.\n    @requires MOI.supports_constraint(\n        model,\n        MOI.VariableIndex,\n        MOI.GreaterThan{Float64},\n    )\n    @requires _supports(config, MOI.optimize!)\n    # If needed, you can test that the model is empty at the start of the test.\n    # You can assume that this will be the case for tests run via `runtests`.\n    # User's calling tests individually need to call `MOI.empty!` themselves.\n    @test MOI.is_empty(model)\n    # Create a simple model. Try to make this as simple as possible so that the\n    # majority of solvers can run the test.\n    x = MOI.add_variable(model)\n    MOI.add_constraint(model, x, MOI.GreaterThan(one(T)))\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    MOI.set(\n        model,\n        MOI.ObjectiveFunction{MOI.VariableIndex}(),\n        x,\n    )\n    # The main component of the test: does calling `optimize!` twice error?\n    MOI.optimize!(model)\n    MOI.optimize!(model)\n    # Check we have a solution.\n    @test MOI.get(model, MOI.TerminationStatus()) == MOI.OPTIMAL\n    # There is a three-argument version of `Base.isapprox` for checking\n    # approximate equality based on the tolerances defined in `config`:\n    @test isapprox(MOI.get(model, MOI.VariablePrimal(), x), one(T), config)\n    # For code-style, these tests should always `return` `nothing`.\n    return\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"info: Info\nMake sure the function is agnostic to the number type T; don't assume it is a Float64 capable solver.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"We also need to write a test for the test. Place this function immediately below the test you just wrote in the same file:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"function setup_test(\n    ::typeof(test_unit_optimize!_twice),\n    model::MOI.Utilities.MockOptimizer,\n    ::Config,\n)\n    MOI.Utilities.set_mock_optimize!(\n        model,\n        (mock::MOI.Utilities.MockOptimizer) -> MOIU.mock_optimize!(\n            mock,\n            MOI.OPTIMAL,\n            (MOI.FEASIBLE_POINT, [1.0]),\n        ),\n    )\n    return\nend","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Finally, you also need to implement Test.version_added. If we added this test when the latest released version of MOI was v0.10.5, define:","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"version_added(::typeof(test_unit_optimize!_twice)) = v\"0.10.6\"","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Step 6","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"Commit the changes to git from ~/.julia/dev/MathOptInterface and submit the PR for review.","category":"page"},{"location":"moi/submodules/Test/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nIf you need help writing a test, open an issue on GitHub, or ask the Developer Chatroom.","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Utilities/reference.md\"","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Utilities/reference/#Utilities.Model","page":"API Reference","title":"Utilities.Model","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.Model","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.Model","page":"API Reference","title":"MathOptInterface.Utilities.Model","text":"MOI.Utilities.Model{T}() where {T}\n\nAn implementation of ModelLike that supports all functions and sets defined in MOI. It is parameterized by the coefficient type.\n\nExamples\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Utilities.UniversalFallback","page":"API Reference","title":"Utilities.UniversalFallback","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.UniversalFallback","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.UniversalFallback","page":"API Reference","title":"MathOptInterface.Utilities.UniversalFallback","text":"UniversalFallback\n\nThe UniversalFallback can be applied on a MOI.ModelLike model to create the model UniversalFallback(model) supporting any constraint and attribute. This allows to have a specialized implementation in model for performance critical constraints and attributes while still supporting other attributes with a small performance penalty. Note that model is unaware of constraints and attributes stored by UniversalFallback so this is not appropriate if model is an optimizer (for this reason, MOI.optimize! has not been implemented). In that case, optimizer bridges should be used instead.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Utilities.@model","page":"API Reference","title":"Utilities.@model","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.@model\nUtilities.GenericModel\nUtilities.GenericOptimizer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@model","page":"API Reference","title":"MathOptInterface.Utilities.@model","text":"macro model(\n    model_name,\n    scalar_sets,\n    typed_scalar_sets,\n    vector_sets,\n    typed_vector_sets,\n    scalar_functions,\n    typed_scalar_functions,\n    vector_functions,\n    typed_vector_functions,\n    is_optimizer = false\n)\n\nCreates a type model_name implementing the MOI model interface and supporting all combinations of the provided functions and sets.\n\nEach typed_ scalar/vector sets/functions argument is a tuple of types. A type is \"typed\" if it has a coefficient {T} as the first type parameter.\n\nTuple syntax\n\nTo give no set/function, write (). To give one set or function X, write (X,).\n\nis_optimizer\n\nIf is_optimizer = true, the resulting struct is a of GenericOptimizer, which is a subtype of MOI.AbstractOptimizer, otherwise, it is a GenericModel, which is a subtype of MOI.ModelLike.\n\nVariableIndex\n\nThe function MOI.VariableIndex must not be given in scalar_functions.\nThe model supports MOI.VariableIndex-in-S constraints where S is MOI.EqualTo, MOI.GreaterThan, MOI.LessThan, MOI.Interval, MOI.Integer, MOI.ZeroOne, MOI.Semicontinuous or MOI.Semiinteger.\nThe sets supported with MOI.VariableIndex cannot be controlled from the macro; use UniversalFallback to support more sets.\n\nExamples\n\nThe model describing a linear program would be:\n\n@model(\n    LPModel,                                          # model_name\n    (),                                               # untyped scalar sets\n    (MOI.EqualTo, MOI.GreaterThan, MOI.LessThan, MOI.Interval), #   typed scalar sets\n    (MOI.Zeros, MOI.Nonnegatives, MOI.Nonpositives),  # untyped vector sets\n    (),                                               #   typed vector sets\n    (),                                               # untyped scalar functions\n    (MOI.ScalarAffineFunction,),                      #   typed scalar functions\n    (MOI.VectorOfVariables,),                         # untyped vector functions\n    (MOI.VectorAffineFunction,),                      #   typed vector functions\n    false,                                            # is_optimizer\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.GenericModel","page":"API Reference","title":"MathOptInterface.Utilities.GenericModel","text":"mutable struct GenericModel{T,O,V,C} <: AbstractModelLike{T}\n\nImplements a model supporting coefficients of type T and:\n\nAn objective function stored in .objective::O\nVariables and VariableIndex constraints stored in .variable_bounds::V\nF-in-S constraints (excluding VariableIndex constraints) stored in .constraints::C\n\nAll interactions take place via the MOI interface, so the types O, V, and C must implement the API as needed for their functionality.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.GenericOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.GenericOptimizer","text":"mutable struct GenericOptimizer{T,O,V,C} <: AbstractOptimizer{T}\n\nImplements a model supporting coefficients of type T and:\n\nAn objective function stored in .objective::O\nVariables and VariableIndex constraints stored in .variable_bounds::V\nF-in-S constraints (excluding VariableIndex constraints) stored in .constraints::C\n\nAll interactions take place via the MOI interface, so the types O, V, and C must implement the API as needed for their functionality.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.objective","page":"API Reference","title":".objective","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.ObjectiveContainer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ObjectiveContainer","page":"API Reference","title":"MathOptInterface.Utilities.ObjectiveContainer","text":"ObjectiveContainer{T}\n\nA helper struct to simplify the handling of objective functions in Utilities.Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.variables","page":"API Reference","title":".variables","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.VariablesContainer\nUtilities.FreeVariables","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.VariablesContainer","page":"API Reference","title":"MathOptInterface.Utilities.VariablesContainer","text":"struct VariablesContainer{T} <: AbstractVectorBounds\n    set_mask::Vector{UInt16}\n    lower::Vector{T}\n    upper::Vector{T}\nend\n\nA struct for storing variables and VariableIndex-related constraints. Used in MOI.Utilities.Model by default.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.FreeVariables","page":"API Reference","title":"MathOptInterface.Utilities.FreeVariables","text":"mutable struct FreeVariables <: MOI.ModelLike\n    n::Int64\n    FreeVariables() = new(0)\nend\n\nA struct for storing free variables that can be used as the variables field of GenericModel or GenericModel. It represents a model that does not support any constraint nor objective function.\n\nExample\n\nThe following model type represents a conic model in geometric form. As opposed to VariablesContainer, FreeVariables does not support constraint bounds so they are bridged into an affine constraint in the MOI.Nonnegatives cone as expected for the geometric conic form.\n\njulia> MOI.Utilities.@product_of_sets(\n    Cones,\n    MOI.Zeros,\n    MOI.Nonnegatives,\n    MOI.SecondOrderCone,\n    MOI.PositiveSemidefiniteConeTriangle,\n);\n\njulia> const ConicModel{T} = MOI.Utilities.GenericOptimizer{\n    T,\n    MOI.Utilities.ObjectiveContainer{T},\n    MOI.Utilities.FreeVariables,\n    MOI.Utilities.MatrixOfConstraints{\n        T,\n        MOI.Utilities.MutableSparseMatrixCSC{\n            T,\n            Int,\n            MOI.Utilities.OneBasedIndexing,\n        },\n        Vector{T},\n        Cones{T},\n    },\n};\n\njulia> model = MOI.instantiate(ConicModel{Float64}, with_bridge_type=Float64);\n\njulia> x = MOI.add_variable(model)\nMathOptInterface.VariableIndex(1)\n\njulia> c = MOI.add_constraint(model, x, MOI.GreaterThan(1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n\njulia> MOI.Bridges.is_bridged(model, c)\ntrue\n\njulia> bridge = MOI.Bridges.bridge(model, c)\nMathOptInterface.Bridges.Constraint.VectorizeBridge{Float64, MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives, MathOptInterface.VariableIndex}(MathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1), 1.0)\n\njulia> bridge.vector_constraint\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.Bridges.is_bridged(model, bridge.vector_constraint)\nfalse\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.constraints","page":"API Reference","title":".constraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.VectorOfConstraints\nUtilities.StructOfConstraints\nUtilities.@struct_of_constraints_by_function_types\nUtilities.@struct_of_constraints_by_set_types\nUtilities.struct_of_constraint_code","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.VectorOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.VectorOfConstraints","text":"mutable struct VectorOfConstraints{\n    F<:MOI.AbstractFunction,\n    S<:MOI.AbstractSet,\n} <: MOI.ModelLike\n    constraints::CleverDicts.CleverDict{\n        MOI.ConstraintIndex{F,S},\n        Tuple{F,S},\n        typeof(CleverDicts.key_to_index),\n        typeof(CleverDicts.index_to_key),\n    }\nend\n\nA struct storing F-in-S constraints as a mapping between the constraint indices to the corresponding tuple of function and set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.StructOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.StructOfConstraints","text":"abstract type StructOfConstraints <: MOI.ModelLike end\n\nA struct storing a subfields other structs storing constraints of different types.\n\nSee Utilities.@struct_of_constraints_by_function_types and Utilities.@struct_of_constraints_by_set_types.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@struct_of_constraints_by_function_types","page":"API Reference","title":"MathOptInterface.Utilities.@struct_of_constraints_by_function_types","text":"Utilities.@struct_of_constraints_by_function_types(name, func_types...)\n\nGiven a vector of n function types (F1, F2,..., Fn) in func_types, defines a subtype of StructOfConstraints of name name and which type parameters {T, C1, C2, ..., Cn}. It contains n field where the ith field has type Ci and stores the constraints of function type Fi.\n\nThe expression Fi can also be a union in which case any constraint for which the function type is in the union is stored in the field with type Ci.\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@struct_of_constraints_by_set_types","page":"API Reference","title":"MathOptInterface.Utilities.@struct_of_constraints_by_set_types","text":"Utilities.@struct_of_constraints_by_set_types(name, func_types...)\n\nGiven a vector of n set types (S1, S2,..., Sn) in func_types, defines a subtype of StructOfConstraints of name name and which type parameters {T, C1, C2, ..., Cn}. It contains n field where the ith field has type Ci and stores the constraints of set type Si. The expression Si can also be a union in which case any constraint for which the set type is in the union is stored in the field with type Ci. This can be useful if Ci is a MatrixOfConstraints in order to concatenate the coefficients of constraints of several different set types in the same matrix.\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.struct_of_constraint_code","page":"API Reference","title":"MathOptInterface.Utilities.struct_of_constraint_code","text":"struct_of_constraint_code(struct_name, types, field_types = nothing)\n\nGiven a vector of n Union{SymbolFun,_UnionSymbolFS{SymbolFun}} or Union{SymbolSet,_UnionSymbolFS{SymbolSet}} in types, defines a subtype of StructOfConstraints of name name and which type parameters {T, F1, F2, ..., Fn} if field_types is nothing and a {T} otherwise. It contains n field where the ith field has type Ci if field_types is nothing and type field_types[i] otherwise. If types is vector of Union{SymbolFun,_UnionSymbolFS{SymbolFun}} (resp. Union{SymbolSet,_UnionSymbolFS{SymbolSet}}) then the constraints of that function (resp. set) type are stored in the corresponding field.\n\nThis function is used by the macros @model, @struct_of_constraints_by_function_types and @struct_of_constraints_by_set_types.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Caching-optimizer","page":"API Reference","title":"Caching optimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.CachingOptimizer\nUtilities.attach_optimizer\nUtilities.reset_optimizer\nUtilities.drop_optimizer\nUtilities.state\nUtilities.mode","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.CachingOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.CachingOptimizer","text":"CachingOptimizer\n\nCachingOptimizer is an intermediate layer that stores a cache of the model and links it with an optimizer. It supports incremental model construction and modification even when the optimizer doesn't.\n\nConstructors\n\n    CachingOptimizer(cache::MOI.ModelLike, optimizer::AbstractOptimizer)\n\nCreates a CachingOptimizer in AUTOMATIC mode, with the optimizer optimizer.\n\nThe type of the optimizer returned is CachingOptimizer{typeof(optimizer), typeof(cache)} so it does not support the function reset_optimizer(::CachingOptimizer, new_optimizer) if the type of new_optimizer is different from the type of optimizer.\n\n    CachingOptimizer(cache::MOI.ModelLike, mode::CachingOptimizerMode)\n\nCreates a CachingOptimizer in the NO_OPTIMIZER state and mode mode.\n\nThe type of the optimizer returned is CachingOptimizer{MOI.AbstractOptimizer,typeof(cache)} so it does support the function reset_optimizer(::CachingOptimizer, new_optimizer) if the type of new_optimizer is different from the type of optimizer.\n\nAbout the type\n\nStates\n\nA CachingOptimizer may be in one of three possible states (CachingOptimizerState):\n\nNO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer. The optimizer is not synchronized with the cached model.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.\n\nModes\n\nA CachingOptimizer has two modes of operation (CachingOptimizerMode):\n\nMANUAL: The only methods that change the state of the CachingOptimizer are Utilities.reset_optimizer, Utilities.drop_optimizer, and Utilities.attach_optimizer. Attempting to perform an operation in the incorrect state results in an error.\nAUTOMATIC: The CachingOptimizer changes its state when necessary. For example, optimize! will automatically call attach_optimizer (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.attach_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.attach_optimizer","text":"attach_optimizer(model::CachingOptimizer)\n\nAttaches the optimizer to model, copying all model data into it. Can be called only from the EMPTY_OPTIMIZER state. If the copy succeeds, the CachingOptimizer will be in state ATTACHED_OPTIMIZER after the call, otherwise an error is thrown; see MOI.copy_to for more details on which errors can be thrown.\n\n\n\n\n\nMOIU.attach_optimizer(model::GenericModel)\n\nCall MOIU.attach_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.reset_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.reset_optimizer","text":"reset_optimizer(m::CachingOptimizer, optimizer::MOI.AbstractOptimizer)\n\nSets or resets m to have the given empty optimizer optimizer.\n\nCan be called from any state. An assertion error will be thrown if optimizer is not empty.\n\nThe CachingOptimizer m will be in state EMPTY_OPTIMIZER after the call.\n\n\n\n\n\nreset_optimizer(m::CachingOptimizer)\n\nDetaches and empties the current optimizer. Can be called from ATTACHED_OPTIMIZER or EMPTY_OPTIMIZER state. The CachingOptimizer will be in state EMPTY_OPTIMIZER after the call.\n\n\n\n\n\nMOIU.reset_optimizer(model::GenericModel, optimizer::MOI.AbstractOptimizer)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\nMOIU.reset_optimizer(model::GenericModel)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.drop_optimizer","page":"API Reference","title":"MathOptInterface.Utilities.drop_optimizer","text":"drop_optimizer(m::CachingOptimizer)\n\nDrops the optimizer, if one is present. Can be called from any state. The CachingOptimizer will be in state NO_OPTIMIZER after the call.\n\n\n\n\n\nMOIU.drop_optimizer(model::GenericModel)\n\nCall MOIU.drop_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.state","page":"API Reference","title":"MathOptInterface.Utilities.state","text":"state(m::CachingOptimizer)::CachingOptimizerState\n\nReturns the state of the CachingOptimizer m. See Utilities.CachingOptimizer.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.mode","page":"API Reference","title":"MathOptInterface.Utilities.mode","text":"mode(m::CachingOptimizer)::CachingOptimizerMode\n\nReturns the operating mode of the CachingOptimizer m. See Utilities.CachingOptimizer.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Mock-optimizer","page":"API Reference","title":"Mock optimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MockOptimizer","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MockOptimizer","page":"API Reference","title":"MathOptInterface.Utilities.MockOptimizer","text":"MockOptimizer\n\nMockOptimizer is a fake optimizer especially useful for testing. Its main feature is that it can store the values that should be returned for each attribute.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Printing","page":"API Reference","title":"Printing","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.latex_formulation","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.latex_formulation","page":"API Reference","title":"MathOptInterface.Utilities.latex_formulation","text":"latex_formulation(model::MOI.ModelLike; kwargs...)\n\nWrap model in a type so that it can be pretty-printed as text/latex in a notebook like IJulia, or in Documenter.\n\nTo render the model, end the cell with latex_formulation(model), or call display(latex_formulation(model)) in to force the display of the model from inside a function.\n\nPossible keyword arguments are:\n\nsimplify_coefficients : Simplify coefficients if possible by omitting them or removing trailing zeros.\ndefault_name : The name given to variables with an empty name.\nprint_types : Print the MOI type of each function and set for clarity.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Copy-utilities","page":"API Reference","title":"Copy utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.default_copy_to\nUtilities.IndexMap\nUtilities.identity_index_map\nUtilities.ModelFilter","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.default_copy_to","page":"API Reference","title":"MathOptInterface.Utilities.default_copy_to","text":"default_copy_to(dest::MOI.ModelLike, src::MOI.ModelLike)\n\nA default implementation of MOI.copy_to(dest, src) for models that implement the incremental interface, i.e., MOI.supports_incremental_interface returns true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.IndexMap","page":"API Reference","title":"MathOptInterface.Utilities.IndexMap","text":"IndexMap()\n\nThe dictionary-like object returned by MOI.copy_to.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.identity_index_map","page":"API Reference","title":"MathOptInterface.Utilities.identity_index_map","text":"identity_index_map(model::MOI.ModelLike)\n\nReturn an IndexMap that maps all variable and constraint indices of model to themselves.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ModelFilter","page":"API Reference","title":"MathOptInterface.Utilities.ModelFilter","text":"ModelFilter(filter::Function, model::MOI.ModelLike)\n\nA layer to filter out various components of model.\n\nThe filter function takes a single argument, which is each element from the list returned by the attributes below. It returns true if the element should be visible in the filtered model and false otherwise.\n\nThe components that are filtered are:\n\nEntire constraint types via:\nMOI.ListOfConstraintTypesPresent\nIndividual constraints via:\nMOI.ListOfConstraintIndices{F,S}\nSpecific attributes via:\nMOI.ListOfModelAttributesSet\nMOI.ListOfConstraintAttributesSet\nMOI.ListOfVariableAttributesSet\n\nwarning: Warning\nThe list of attributes filtered may change in a future release. You should write functions that are generic and not limited to the five types listed above. Thus, you should probably define a fallback filter(::Any) = true.\n\nSee below for examples of how this works.\n\nnote: Note\nThis layer has a limited scope. It is intended by be used in conjunction with MOI.copy_to.\n\nExample: copy model excluding integer constraints\n\nUse the do syntax to provide a single function.\n\nfiltered_src = MOI.Utilities.ModelFilter(src) do item\n    return item != (MOI.VariableIndex, MOI.Integer)\nend\nMOI.copy_to(dest, filtered_src)\n\nExample: copy model excluding names\n\nUse type dispatch to simplify the implementation:\n\nmy_filter(::Any) = true  # Note the generic fallback!\nmy_filter(::MOI.VariableName) = false\nmy_filter(::MOI.ConstraintName) = false\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nMOI.copy_to(dest, filtered_src)\n\nExample: copy irreducible infeasible subsystem\n\nmy_filter(::Any) = true  # Note the generic fallback!\nfunction my_filter(ci::MOI.ConstraintIndex)\n    status = MOI.get(dest, MOI.ConstraintConflictStatus(), ci)\n    return status != MOI.NOT_IN_CONFLICT\nend\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nMOI.copy_to(dest, filtered_src)\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#Penalty-relaxation","page":"API Reference","title":"Penalty relaxation","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.PenaltyRelaxation\nUtilities.ScalarPenaltyRelaxation","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.PenaltyRelaxation","page":"API Reference","title":"MathOptInterface.Utilities.PenaltyRelaxation","text":"PenaltyRelaxation(\n    penalties = Dict{MOI.ConstraintIndex,Float64}();\n    default::Union{Nothing,T} = 1.0,\n)\n\nA problem modifier that, when passed to MOI.modify, destructively modifies the model in-place to create a penalized relaxation of the constraints.\n\nwarning: Warning\nThis is a destructive routine that modifies the model in-place. If you don't want to modify the original model, use JuMP.copy_model to create a copy before calling MOI.modify.\n\nReformulation\n\nSee Utilities.ScalarPenaltyRelaxation for details of the reformulation.\n\nFor each constraint ci, the penalty passed to Utilities.ScalarPenaltyRelaxation is get(penalties, ci, default). If the value is nothing, because ci does not exist in penalties and default = nothing, then the constraint is skipped.\n\nReturn value\n\nMOI.modify(model, PenaltyRelaxation()) returns a Dict{MOI.ConstraintIndex,MOI.ScalarAffineFunction} that maps each constraint index to the corresponding y + z as a MOI.ScalarAffineFunction. In an optimal solution, query the value of these functions to compute the violation of each constraint.\n\nRelax a subset of constraints\n\nTo relax a subset of constraints, pass a penalties dictionary and set default = nothing.\n\nSupported constraint types\n\nThe penalty relaxation is currently limited to modifying MOI.ScalarAffineFunction and MOI.ScalarQuadraticFunction constraints in the linear sets MOI.LessThan, MOI.GreaterThan, MOI.EqualTo and MOI.Interval.\n\nIt does not include variable bound or integrality constraints, because these cannot be modified in-place.\n\nTo modify variable bounds, rewrite them as linear constraints.\n\nExamples\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(default = 2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c] isa MOI.ScalarAffineFunction{Float64}\ntrue\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(Dict(c => 3.0)));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 3.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c] isa MOI.ScalarAffineFunction{Float64}\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ScalarPenaltyRelaxation","page":"API Reference","title":"MathOptInterface.Utilities.ScalarPenaltyRelaxation","text":"ScalarPenaltyRelaxation(penalty::T) where {T}\n\nA problem modifier that, when passed to MOI.modify, destructively modifies the constraint in-place to create a penalized relaxation of the constraint.\n\nwarning: Warning\nThis is a destructive routine that modifies the constraint in-place. If you don't want to modify the original model, use JuMP.copy_model to create a copy before calling MOI.modify.\n\nReformulation\n\nThe penalty relaxation modifies constraints of the form f(x) in S into f(x) + y - z in S, where y z ge 0, and then it introduces a penalty term into the objective of a times (y + z) (if minimizing, else -a), where a is penalty\n\nWhen S is MOI.LessThan or MOI.GreaterThan, we omit y or z respectively as a performance optimization.\n\nReturn value\n\nMOI.modify(model, ci, ScalarPenaltyRelaxation(penalty)) returns y + z as a MOI.ScalarAffineFunction. In an optimal solution, query the value of this function to compute the violation of the constraint.\n\nExamples\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> f = MOI.modify(model, c, MOI.Utilities.ScalarPenaltyRelaxation(2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 v[1] - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> f isa MOI.ScalarAffineFunction{Float64}\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MatrixOfConstraints","page":"API Reference","title":"MatrixOfConstraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MatrixOfConstraints","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MatrixOfConstraints","page":"API Reference","title":"MathOptInterface.Utilities.MatrixOfConstraints","text":"mutable struct MatrixOfConstraints{T,AT,BT,ST} <: MOI.ModelLike\n    coefficients::AT\n    constants::BT\n    sets::ST\n    caches::Vector{Any}\n    are_indices_mapped::Vector{BitSet}\n    final_touch::Bool\nend\n\nRepresent ScalarAffineFunction and VectorAffinefunction constraints in a matrix form where the linear coefficients of the functions are stored in the coefficients field, the constants of the functions or sets are stored in the constants field. Additional information about the sets are stored in the sets field.\n\nThis model can only be used as the constraints field of a MOI.Utilities.AbstractModel.\n\nWhen the constraints are added, they are stored in the caches field. They are only loaded in the coefficients and constants fields once MOI.Utilities.final_touch is called. For this reason, MatrixOfConstraints should not be used by an incremental interface. Use MOI.copy_to instead.\n\nThe constraints can be added in two different ways:\n\nWith add_constraint, in which case a canonicalized copy of the function is stored in caches.\nWith pass_nonvariable_constraints, in which case the functions and sets are stored themselves in caches without mapping the variable indices. The corresponding index in caches is added in are_indices_mapped. This avoids doing a copy of the function in case the getter of CanonicalConstraintFunction does not make a copy for the source model, e.g., this is the case of VectorOfConstraints.\n\nWe illustrate this with an example. Suppose a model is copied from a src::MOI.Utilities.Model to a bridged model with a MatrixOfConstraints. For all the types that are not bridged, the constraints will be copied with pass_nonvariable_constraints. Hence the functions stored in caches are exactly the same as the ones stored in src. This is ok since this is only during the copy_to operation during which src cannot be modified. On the other hand, for the types that are bridged, the functions added may contain duplicates even if the functions did not contain duplicates in src so duplicates are removed with MOI.Utilities.canonical.\n\nInterface\n\nThe .coefficients::AT type must implement:\n\nAT()\nMOI.empty(::AT)!\nMOI.Utilities.add_column\nMOI.Utilities.set_number_of_rows\nMOI.Utilities.allocate_terms\nMOI.Utilities.load_terms\nMOI.Utilities.final_touch\n\nThe .constants::BT type must implement:\n\nBT()\nBase.empty!(::BT)\nBase.resize(::BT)\nMOI.Utilities.load_constants\nMOI.Utilities.function_constants\nMOI.Utilities.set_from_constants\n\nThe .sets::ST type must implement:\n\nST()\nMOI.is_empty(::ST)\nMOI.empty(::ST)\nMOI.dimension(::ST)\nMOI.is_valid(::ST, ::MOI.ConstraintIndex)\nMOI.get(::ST, ::MOI.ListOfConstraintTypesPresent)\nMOI.get(::ST, ::MOI.NumberOfConstraints)\nMOI.get(::ST, ::MOI.ListOfConstraintIndices)\nMOI.Utilities.set_types\nMOI.Utilities.set_index\nMOI.Utilities.add_set\nMOI.Utilities.rows\nMOI.Utilities.final_touch\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.coefficients","page":"API Reference","title":".coefficients","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.add_column\nUtilities.allocate_terms\nUtilities.set_number_of_rows\nUtilities.load_terms\nUtilities.final_touch\nUtilities.extract_function","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.add_column","page":"API Reference","title":"MathOptInterface.Utilities.add_column","text":"add_column(coefficients)::Nothing\n\nTell coefficients to pre-allocate datastructures as needed to store one column.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.allocate_terms","page":"API Reference","title":"MathOptInterface.Utilities.allocate_terms","text":"allocate_terms(coefficients, index_map, func)::Nothing\n\nTell coefficients that the terms of the function func where the variable indices are mapped with index_map will be loaded with load_terms.\n\nThe function func must be canonicalized before calling allocate_terms. See is_canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_number_of_rows","page":"API Reference","title":"MathOptInterface.Utilities.set_number_of_rows","text":"set_number_of_rows(coefficients, n)::Nothing\n\nTell coefficients to pre-allocate datastructures as needed to store n rows.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.load_terms","page":"API Reference","title":"MathOptInterface.Utilities.load_terms","text":"load_terms(coefficients, index_map, func, offset)::Nothing\n\nLoads the terms of func to coefficients, mapping the variable indices with index_map.\n\nThe ith dimension of func is loaded at the (offset + i)th row of coefficients.\n\nThe function must be allocated first with allocate_terms.\n\nThe function func must be canonicalized, see is_canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.final_touch","page":"API Reference","title":"MathOptInterface.Utilities.final_touch","text":"final_touch(coefficients)::Nothing\n\nInforms the coefficients that all functions have been added with load_terms. No more modification is allowed unless MOI.empty! is called.\n\nfinal_touch(sets)::Nothing\n\nInforms the sets that all functions have been added with add_set. No more modification is allowed unless MOI.empty! is called.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.extract_function","page":"API Reference","title":"MathOptInterface.Utilities.extract_function","text":"extract_function(coefficients, row::Integer, constant::T) where {T}\n\nReturn the MOI.ScalarAffineFunction{T} function corresponding to row row in coefficients.\n\nextract_function(\n    coefficients,\n    rows::UnitRange,\n    constants::Vector{T},\n) where{T}\n\nReturn the MOI.VectorAffineFunction{T} function corresponding to rows rows in coefficients.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.MutableSparseMatrixCSC\nUtilities.AbstractIndexing\nUtilities.ZeroBasedIndexing\nUtilities.OneBasedIndexing","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MutableSparseMatrixCSC","page":"API Reference","title":"MathOptInterface.Utilities.MutableSparseMatrixCSC","text":"mutable struct MutableSparseMatrixCSC{Tv,Ti<:Integer,I<:AbstractIndexing}\n    indexing::I\n    m::Int\n    n::Int\n    colptr::Vector{Ti}\n    rowval::Vector{Ti}\n    nzval::Vector{Tv}\n    nz_added::Vector{Ti}\nend\n\nMatrix type loading sparse matrices in the Compressed Sparse Column format. The indexing used is indexing, see AbstractIndexing. The other fields have the same meaning than for SparseArrays.SparseMatrixCSC except that the indexing is different unless indexing is OneBasedIndexing. In addition, nz_added is used to cache the number of non-zero terms that have been added to each column due to the incremental nature of load_terms.\n\nThe matrix is loaded in 5 steps:\n\nMOI.empty! is called.\nMOI.Utilities.add_column and MOI.Utilities.allocate_terms are called in any order.\nMOI.Utilities.set_number_of_rows is called.\nMOI.Utilities.load_terms is called for each affine function.\nMOI.Utilities.final_touch is called.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.AbstractIndexing","page":"API Reference","title":"MathOptInterface.Utilities.AbstractIndexing","text":"abstract type AbstractIndexing end\n\nIndexing to be used for storing the row and column indices of MutableSparseMatrixCSC. See ZeroBasedIndexing and OneBasedIndexing.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ZeroBasedIndexing","page":"API Reference","title":"MathOptInterface.Utilities.ZeroBasedIndexing","text":"struct ZeroBasedIndexing <: AbstractIndexing end\n\nZero-based indexing: the ith row or column has index i - 1. This is useful when the vectors of row and column indices need to be communicated to a library using zero-based indexing such as C libraries.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.OneBasedIndexing","page":"API Reference","title":"MathOptInterface.Utilities.OneBasedIndexing","text":"struct ZeroBasedIndexing <: AbstractIndexing end\n\nOne-based indexing: the ith row or column has index i. This enables an allocation-free conversion of MutableSparseMatrixCSC to SparseArrays.SparseMatrixCSC.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.constants","page":"API Reference","title":".constants","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.load_constants\nUtilities.function_constants\nUtilities.set_from_constants","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.load_constants","page":"API Reference","title":"MathOptInterface.Utilities.load_constants","text":"load_constants(constants, offset, func_or_set)::Nothing\n\nThis function loads the constants of func_or_set in constants at an offset of offset. Where offset is the sum of the dimensions of the constraints already loaded. The storage should be preallocated with resize! before calling this function.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\nThe constants are loaded in three steps:\n\nBase.empty! is called.\nBase.resize! is called with the sum of the dimensions of all constraints.\nMOI.Utilities.load_constants is called for each function for vector constraint or set for scalar constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.function_constants","page":"API Reference","title":"MathOptInterface.Utilities.function_constants","text":"function_constants(constants, rows)\n\nThis function returns the function constants that were loaded with load_constants at the rows rows.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_from_constants","page":"API Reference","title":"MathOptInterface.Utilities.set_from_constants","text":"set_from_constants(constants, S::Type, rows)::S\n\nThis function returns an instance of the set S for which the constants where loaded with load_constants at the rows rows.\n\nThis function should be implemented to be usable as storage of constants for MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.Hyperrectangle","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.Hyperrectangle","page":"API Reference","title":"MathOptInterface.Utilities.Hyperrectangle","text":"struct Hyperrectangle{T} <: AbstractVectorBounds\n    lower::Vector{T}\n    upper::Vector{T}\nend\n\nA struct for the .constants field in MatrixOfConstraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#.sets","page":"API Reference","title":".sets","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.set_index\nUtilities.set_types\nUtilities.add_set\nUtilities.rows\nUtilities.num_rows\nUtilities.set_with_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_index","page":"API Reference","title":"MathOptInterface.Utilities.set_index","text":"set_index(sets, ::Type{S})::Union{Int,Nothing} where {S<:MOI.AbstractSet}\n\nReturn an integer corresponding to the index of the set type in the list given by set_types.\n\nIf S is not part of the list, return nothing.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_types","page":"API Reference","title":"MathOptInterface.Utilities.set_types","text":"set_types(sets)::Vector{Type}\n\nReturn the list of the types of the sets allowed in sets.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.add_set","page":"API Reference","title":"MathOptInterface.Utilities.add_set","text":"add_set(sets, i)::Int64\n\nAdd a scalar set of type index i.\n\nadd_set(sets, i, dim)::Int64\n\nAdd a vector set of type index i and dimension dim.\n\nBoth methods return a unique Int64 of the set that can be used to reference this set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.rows","page":"API Reference","title":"MathOptInterface.Utilities.rows","text":"rows(sets, ci::MOI.ConstraintIndex)::Union{Int,UnitRange{Int}}\n\nReturn the rows in 1:MOI.dimension(sets) corresponding to the set of id ci.value.\n\nFor scalar sets, this returns an Int. For vector sets, this returns an UnitRange{Int}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.num_rows","page":"API Reference","title":"MathOptInterface.Utilities.num_rows","text":"num_rows(sets::OrderedProductOfSets, ::Type{S}) where {S}\n\nReturn the number of rows corresponding to a set of type S. That is, it is the sum of the dimensions of the sets of type S.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_with_dimension","page":"API Reference","title":"MathOptInterface.Utilities.set_with_dimension","text":"set_with_dimension(::Type{S}, dim) where {S<:MOI.AbstractVectorSet}\n\nReturns the instance of S of MOI.dimension dim. This needs to be implemented for sets of type S to be useable with MatrixOfConstraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.ProductOfSets\nUtilities.MixOfScalarSets\nUtilities.@mix_of_scalar_sets\nUtilities.OrderedProductOfSets\nUtilities.@product_of_sets","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ProductOfSets","page":"API Reference","title":"MathOptInterface.Utilities.ProductOfSets","text":"abstract type ProductOfSets{T} end\n\nRepresents a cartesian product of sets of given types.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.MixOfScalarSets","page":"API Reference","title":"MathOptInterface.Utilities.MixOfScalarSets","text":"abstract type MixOfScalarSets{T} <: ProductOfSets{T} end\n\nProduct of scalar sets in the order the constraints are added, mixing the constraints of different types.\n\nUse @mix_of_scalar_sets to generate a new subtype.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@mix_of_scalar_sets","page":"API Reference","title":"MathOptInterface.Utilities.@mix_of_scalar_sets","text":"@mix_of_scalar_sets(name, set_types...)\n\nGenerate a new MixOfScalarSets subtype.\n\nExample\n\n@mix_of_scalar_sets(\n    MixedIntegerLinearProgramSets,\n    MOI.GreaterThan{T},\n    MOI.LessThan{T},\n    MOI.EqualTo{T},\n    MOI.Integer,\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.OrderedProductOfSets","page":"API Reference","title":"MathOptInterface.Utilities.OrderedProductOfSets","text":"abstract type OrderedProductOfSets{T} <: ProductOfSets{T} end\n\nProduct of sets in the order the constraints are added, grouping the constraints of the same types contiguously.\n\nUse @product_of_sets to generate new subtypes.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.@product_of_sets","page":"API Reference","title":"MathOptInterface.Utilities.@product_of_sets","text":"@product_of_sets(name, set_types...)\n\nGenerate a new OrderedProductOfSets subtype.\n\nExample\n\n@product_of_sets(\n    LinearOrthants,\n    MOI.Zeros,\n    MOI.Nonnegatives,\n    MOI.Nonpositives,\n    MOI.ZeroOne,\n)\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Utilities/reference/#Fallbacks","page":"API Reference","title":"Fallbacks","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.get_fallback","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.get_fallback","page":"API Reference","title":"MathOptInterface.Utilities.get_fallback","text":"get_fallback(model::MOI.ModelLike, ::MOI.ObjectiveValue)\n\nCompute the objective function value using the VariablePrimal results and the ObjectiveFunction value.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, ::MOI.DualObjectiveValue, T::Type)::T\n\nCompute the dual objective value of type T using the ConstraintDual results and the ConstraintFunction and ConstraintSet values. Note that the nonlinear part of the model is ignored.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, ::MOI.ConstraintPrimal,\n             constraint_index::MOI.ConstraintIndex)\n\nCompute the value of the function of the constraint of index constraint_index using the VariablePrimal results and the ConstraintFunction values.\n\n\n\n\n\nget_fallback(model::MOI.ModelLike, attr::MOI.ConstraintDual,\n             ci::MOI.ConstraintIndex{Union{MOI.VariableIndex,\n                                           MOI.VectorOfVariables}})\n\nCompute the dual of the constraint of index ci using the ConstraintDual of other constraints and the ConstraintFunction values. Throws an error if some constraints are quadratic or if there is one another MOI.VariableIndex-in-S or MOI.VectorOfVariables-in-S constraint with one of the variables in the function of the constraint ci.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Function-utilities","page":"API Reference","title":"Function utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for functions:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.eval_variables\nUtilities.map_indices\nUtilities.substitute_variables\nUtilities.filter_variables\nUtilities.remove_variable\nUtilities.all_coefficients\nUtilities.unsafe_add\nUtilities.isapprox_zero\nUtilities.modify_function\nUtilities.zero_with_output_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.eval_variables","page":"API Reference","title":"MathOptInterface.Utilities.eval_variables","text":"eval_variables(value_fn::Function, f::MOI.AbstractFunction)\n\nReturns the value of function f if each variable index vi is evaluated as value_fn(vi).\n\nNote that value_fn must return a Number. See substitute_variables for a similar function where value_fn returns an MOI.AbstractScalarFunction.\n\nwarning: Warning\nThe two-argument version of eval_variables is deprecated and may be removed in MOI v2.0.0. Use the three-argument method eval_variables(::Function, ::MOI.ModelLike, ::MOI.AbstractFunction) instead.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.map_indices","page":"API Reference","title":"MathOptInterface.Utilities.map_indices","text":"map_indices(index_map::Function, attr::MOI.AnyAttribute, x::X)::X where {X}\n\nSubstitute any MOI.VariableIndex (resp. MOI.ConstraintIndex) in x by the MOI.VariableIndex (resp. MOI.ConstraintIndex) of the same type given by index_map(x).\n\nWhen to implement this method for new types X\n\nThis function is used by implementations of MOI.copy_to on constraint functions, attribute values and submittable values. If you define a new attribute whose values x::X contain variable or constraint indices, you must also implement this function.\n\n\n\n\n\nmap_indices(\n    variable_map::AbstractDict{T,T},\n    x::X,\n)::X where {T<:MOI.Index,X}\n\nShortcut for map_indices(vi -> variable_map[vi], x).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.substitute_variables","page":"API Reference","title":"MathOptInterface.Utilities.substitute_variables","text":"substitute_variables(variable_map::Function, x)\n\nSubstitute any MOI.VariableIndex in x by variable_map(x). The variable_map function returns either MOI.VariableIndex or MOI.ScalarAffineFunction, see eval_variables for a similar function where variable_map returns a number.\n\nThis function is used by bridge optimizers on constraint functions, attribute values and submittable values when at least one variable bridge is used hence it needs to be implemented for custom types that are meant to be used as attribute or submittable value.\n\nnote: Note\nWhen implementing a new method, don't use substitute_variables(::Function, because Julia will not specialize on it. Use instead substitute_variables(::F, ...) where {F<:Function}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.filter_variables","page":"API Reference","title":"MathOptInterface.Utilities.filter_variables","text":"filter_variables(keep::Function, f::AbstractFunction)\n\nReturn a new function f with the variable vi such that !keep(vi) removed.\n\nWARNING: Don't define filter_variables(::Function, ...) because Julia will not specialize on this. Define instead filter_variables(::F, ...) where {F<:Function}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.remove_variable","page":"API Reference","title":"MathOptInterface.Utilities.remove_variable","text":"remove_variable(f::AbstractFunction, vi::VariableIndex)\n\nReturn a new function f with the variable vi removed.\n\n\n\n\n\nremove_variable(\n    f::MOI.AbstractFunction,\n    s::MOI.AbstractSet,\n    vi::MOI.VariableIndex,\n)\n\nReturn a tuple (g, t) representing the constraint f-in-s with the variable vi removed. That is, the terms containing the variable vi in the function f are removed and the dimension of the set s is updated if needed (e.g. when f is a VectorOfVariables with vi being one of the variables).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.all_coefficients","page":"API Reference","title":"MathOptInterface.Utilities.all_coefficients","text":"all_coefficients(p::Function, f::MOI.AbstractFunction)\n\nDetermine whether predicate p returns true for all coefficients of f, returning false as soon as the first coefficient of f for which p returns false is encountered (short-circuiting). Similar to all.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.unsafe_add","page":"API Reference","title":"MathOptInterface.Utilities.unsafe_add","text":"unsafe_add(t1::MOI.ScalarAffineTerm, t2::MOI.ScalarAffineTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.ScalarAffineTerm. It is unsafe because it uses the variable of t1 as the variable of the output without checking that it is equal to that of t2.\n\n\n\n\n\nunsafe_add(t1::MOI.ScalarQuadraticTerm, t2::MOI.ScalarQuadraticTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.ScalarQuadraticTerm. It is unsafe because it uses the variable's of t1 as the variable's of the output without checking that they are the same (up to permutation) to those of t2.\n\n\n\n\n\nunsafe_add(t1::MOI.VectorAffineTerm, t2::MOI.VectorAffineTerm)\n\nSums the coefficients of t1 and t2 and returns an output MOI.VectorAffineTerm. It is unsafe because it uses the output_index and variable of t1 as the output_index and variable of the output term without checking that they are equal to those of t2.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.isapprox_zero","page":"API Reference","title":"MathOptInterface.Utilities.isapprox_zero","text":"isapprox_zero(f::MOI.AbstractFunction, tol)\n\nReturn a Bool indicating whether the function f is approximately zero using tol as a tolerance.\n\nImportant note\n\nThis function assumes that f does not contain any duplicate terms, you might want to first call canonical if that is not guaranteed. For instance, given\n\nf = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.([1, -1], [x, x]), 0)\n\nthen isapprox_zero(f) is false but isapprox_zero(MOIU.canonical(f)) is true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.modify_function","page":"API Reference","title":"MathOptInterface.Utilities.modify_function","text":"modify_function(f::AbstractFunction, change::AbstractFunctionModification)\n\nReturn a copy of the function f, modified according to change.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.zero_with_output_dimension","page":"API Reference","title":"MathOptInterface.Utilities.zero_with_output_dimension","text":"zero_with_output_dimension(::Type{T}, output_dimension::Integer) where {T}\n\nCreate an instance of type T with the output dimension output_dimension.\n\nThis is mostly useful in Bridges, when code needs to be agnostic to the type of vector-valued function that is passed in.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following functions can be used to canonicalize a function:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.is_canonical\nUtilities.canonical\nUtilities.canonicalize!","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.is_canonical","page":"API Reference","title":"MathOptInterface.Utilities.is_canonical","text":"is_canonical(f::Union{ScalarAffineFunction, VectorAffineFunction})\n\nReturns a Bool indicating whether the function is in canonical form. See canonical.\n\n\n\n\n\nis_canonical(f::Union{ScalarQuadraticFunction, VectorQuadraticFunction})\n\nReturns a Bool indicating whether the function is in canonical form. See canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.canonical","page":"API Reference","title":"MathOptInterface.Utilities.canonical","text":"canonical(f::MOI.AbstractFunction)\n\nReturns the function in a canonical form, i.e.\n\nA term appear only once.\nThe coefficients are nonzero.\nThe terms appear in increasing order of variable where there the order of the variables is the order of their value.\nFor a AbstractVectorFunction, the terms are sorted in ascending order of output index.\n\nThe output of canonical can be assumed to be a copy of f, even for VectorOfVariables.\n\nExamples\n\nIf x (resp. y, z) is VariableIndex(1) (resp. 2, 3). The canonical representation of ScalarAffineFunction([y, x, z, x, z], [2, 1, 3, -2, -3], 5) is ScalarAffineFunction([x, y], [-1, 2], 5).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.canonicalize!","page":"API Reference","title":"MathOptInterface.Utilities.canonicalize!","text":"canonicalize!(f::Union{ScalarAffineFunction, VectorAffineFunction})\n\nConvert a function to canonical form in-place, without allocating a copy to hold the result. See canonical.\n\n\n\n\n\ncanonicalize!(f::Union{ScalarQuadraticFunction, VectorQuadraticFunction})\n\nConvert a function to canonical form in-place, without allocating a copy to hold the result. See canonical.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following functions can be used to manipulate functions with basic algebra:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.scalar_type\nUtilities.scalarize\nUtilities.eachscalar\nUtilities.promote_operation\nUtilities.operate\nUtilities.operate!\nUtilities.operate_output_index!\nUtilities.vectorize","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.scalar_type","page":"API Reference","title":"MathOptInterface.Utilities.scalar_type","text":"scalar_type(F::Type{<:MOI.AbstractVectorFunction})\n\nType of functions obtained by indexing objects obtained by calling eachscalar on functions of type F.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.scalarize","page":"API Reference","title":"MathOptInterface.Utilities.scalarize","text":"scalarize(func::MOI.VectorOfVariables, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.SingleVariable}.\n\nSee also eachscalar.\n\n\n\n\n\nscalarize(func::MOI.VectorAffineFunction{T}, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.ScalarAffineFunction{T}}.\n\nSee also eachscalar.\n\n\n\n\n\nscalarize(func::MOI.VectorQuadraticFunction{T}, ignore_constants::Bool = false)\n\nReturns a vector of scalar functions making up the vector function in the form of a Vector{MOI.ScalarQuadraticFunction{T}}.\n\nSee also eachscalar.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.eachscalar","page":"API Reference","title":"MathOptInterface.Utilities.eachscalar","text":"eachscalar(f::MOI.AbstractVectorFunction)\n\nReturns an iterator for the scalar components of the vector function.\n\nSee also scalarize.\n\n\n\n\n\neachscalar(f::MOI.AbstractVector)\n\nReturns an iterator for the scalar components of the vector.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.promote_operation","page":"API Reference","title":"MathOptInterface.Utilities.promote_operation","text":"promote_operation(\n    op::Function,\n    ::Type{T},\n    ArgsTypes::Type{<:Union{T,AbstractVector{T},MOI.AbstractFunction}}...,\n) where {T<:Number}\n\nCompute the return type of the call operate(op, T, args...), where the types of the arguments args are ArgsTypes.\n\nOne assumption is that the element type T is invariant under each operation. That is, op(::T, ::T)::T where op is a +, -, *, and /.\n\nThere are six methods for which we implement Utilities.promote_operation:\n\n+ a. promote_operation(::typeof(+), ::Type{T}, ::Type{F1}, ::Type{F2})\n- a. promote_operation(::typeof(-), ::Type{T}, ::Type{F}) b. promote_operation(::typeof(-), ::Type{T}, ::Type{F1}, ::Type{F2})\n* a. promote_operation(::typeof(*), ::Type{T}, ::Type{T}, ::Type{F}) b. promote_operation(::typeof(*), ::Type{T}, ::Type{F}, ::Type{T}) c. promote_operation(::typeof(*), ::Type{T}, ::Type{F1}, ::Type{F2})    where F1 and F2 are VariableIndex or ScalarAffineFunction d. promote_operation(::typeof(*), ::Type{T}, ::Type{<:Diagonal{T}}, ::Type{F}\n/ a. promote_operation(::typeof(/), ::Type{T}, ::Type{F}, ::Type{T})\nvcat a. promote_operation(::typeof(vcat), ::Type{T}, ::Type{F}...)\nimag a. promote_operation(::typeof(imag), ::Type{T}, ::Type{F})    where F is VariableIndex or VectorOfVariables\n\nIn each case, F (or F1 and F2) is one of the ten supported types, with a restriction that the mathematical operation makes sense, for example, we don't define promote_operation(-, T, F1, F2) where F1 is a scalar-valued function and  F2 is a vector-valued function. The ten supported types are:\n\n::T\n::VariableIndex\n::ScalarAffineFunction{T}\n::ScalarQuadraticFunction{T}\n::ScalarNonlinearFunction\n::AbstractVector{T}\n::VectorOfVariables\n::VectorAffineFunction{T}\n::VectorQuadraticFunction{T}\n::VectorNonlinearFunction\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate","page":"API Reference","title":"MathOptInterface.Utilities.operate","text":"operate(\n    op::Function,\n    ::Type{T},\n    args::Union{T,MOI.AbstractFunction}...,\n)::MOI.AbstractFunction where {T<:Number}\n\nReturns an MOI.AbstractFunction representing the function resulting from the operation op(args...) on functions of coefficient type T.\n\nNo argument can be modified.\n\nMethods\n\n+ a. operate(::typeof(+), ::Type{T}, ::F1) b. operate(::typeof(+), ::Type{T}, ::F1, ::F2) c. operate(::typeof(+), ::Type{T}, ::F1...)\n- a. operate(::typeof(-), ::Type{T}, ::F) b. operate(::typeof(-), ::Type{T}, ::F1, ::F2)\n* a. operate(::typeof(*), ::Type{T}, ::T, ::F) b. operate(::typeof(*), ::Type{T}, ::F, ::T) c. operate(::typeof(*), ::Type{T}, ::F1, ::F2)    where F1 and F2 are VariableIndex or ScalarAffineFunction d. operate(::typeof(*), ::Type{T}, ::Diagonal{T}, ::F)\n/ a. operate(::typeof(/), ::Type{T}, ::F, ::T)\nvcat a. operate(::typeof(vcat), ::Type{T}, ::F...)\nimag a. operate(::typeof(imag), ::Type{T}, ::F)    where F is VariableIndex or VectorOfVariables\n\nOne assumption is that the element type T is invariant under each operation. That is, op(::T, ::T)::T where op is a +, -, *, and /.\n\nIn each case, F (or F1 and F2) is one of the ten supported types, with a restriction that the mathematical operation makes sense, for example, we don't define promote_operation(-, T, F1, F2) where F1 is a scalar-valued function and  F2 is a vector-valued function. The ten supported types are:\n\n::T\n::VariableIndex\n::ScalarAffineFunction{T}\n::ScalarQuadraticFunction{T}\n::ScalarNonlinearFunction\n::AbstractVector{T}\n::VectorOfVariables\n::VectorAffineFunction{T}\n::VectorQuadraticFunction{T}\n::VectorNonlinearFunction\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate!","page":"API Reference","title":"MathOptInterface.Utilities.operate!","text":"operate!(\n    op::Function,\n    ::Type{T},\n    args::Union{T,MOI.AbstractFunction}...,\n)::MOI.AbstractFunction where {T<:Number}\n\nReturns an MOI.AbstractFunction representing the function resulting from the operation op(args...) on functions of coefficient type T.\n\nThe first argument may be modified, in which case the return value is identical to the first argument. For operations which cannot be implemented in-place, this function returns a new object.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.operate_output_index!","page":"API Reference","title":"MathOptInterface.Utilities.operate_output_index!","text":"operate_output_index!(\n    op::Union{typeof(+),typeof(-)},\n    ::Type{T},\n    output_index::Integer,\n    f::Union{AbstractVector{T},MOI.AbstractVectorFunction}\n    g::Union{T,MOI.AbstractScalarFunction}...\n) where {T<:Number}\n\nReturn an MOI.AbstractVectorFunction in which the scalar function in row output_index is the result of op(f[output_index], g).\n\nThe functions at output index different to output_index are the same as the functions at the same output index in func. The first argument may be modified.\n\nMethods\n\n+ a. operate_output_index!(+, ::Type{T}, ::Int, ::VectorF, ::ScalarF)\n- a. operate_output_index!(-, ::Type{T}, ::Int, ::VectorF, ::ScalarF)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.vectorize","page":"API Reference","title":"MathOptInterface.Utilities.vectorize","text":"vectorize(x::AbstractVector{<:Number})\n\nReturns x.\n\n\n\n\n\nvectorize(x::AbstractVector{MOI.VariableIndex})\n\nReturns the vector of scalar affine functions in the form of a MOI.VectorAffineFunction{T}.\n\n\n\n\n\nvectorize(funcs::AbstractVector{MOI.ScalarAffineFunction{T}}) where T\n\nReturns the vector of scalar affine functions in the form of a MOI.VectorAffineFunction{T}.\n\n\n\n\n\nvectorize(funcs::AbstractVector{MOI.ScalarQuadraticFunction{T}}) where T\n\nReturns the vector of scalar quadratic functions in the form of a MOI.VectorQuadraticFunction{T}.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Constraint-utilities","page":"API Reference","title":"Constraint utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for moving the function constant to the set for scalar constraints:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.shift_constant\nUtilities.supports_shift_constant\nUtilities.normalize_and_add_constraint\nUtilities.normalize_constant","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.shift_constant","page":"API Reference","title":"MathOptInterface.Utilities.shift_constant","text":"shift_constant(set::MOI.AbstractScalarSet, offset)\n\nReturns a new scalar set new_set such that func-in-set is equivalent to func + offset-in-new_set.\n\nOnly define this function if it makes sense to!\n\nUse supports_shift_constant to check if the set supports shifting:\n\nif supports_shift_constant(typeof(old_set))\n    new_set = shift_constant(old_set, offset)\n    f.constant = 0\n    add_constraint(model, f, new_set)\nelse\n    add_constraint(model, f, old_set)\nend\n\nSee also supports_shift_constant.\n\nExamples\n\nThe call shift_constant(MOI.Interval(-2, 3), 1) is equal to MOI.Interval(-1, 4).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.supports_shift_constant","page":"API Reference","title":"MathOptInterface.Utilities.supports_shift_constant","text":"supports_shift_constant(::Type{S}) where {S<:MOI.AbstractSet}\n\nReturn true if shift_constant is defined for set S.\n\nSee also shift_constant.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.normalize_and_add_constraint","page":"API Reference","title":"MathOptInterface.Utilities.normalize_and_add_constraint","text":"normalize_and_add_constraint(\n    model::MOI.ModelLike,\n    func::MOI.AbstractScalarFunction,\n    set::MOI.AbstractScalarSet;\n    allow_modify_function::Bool = false,\n)\n\nAdds the scalar constraint obtained by moving the constant term in func to the set in model. If allow_modify_function is true then the function func can be modified.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.normalize_constant","page":"API Reference","title":"MathOptInterface.Utilities.normalize_constant","text":"normalize_constant(\n    func::MOI.AbstractScalarFunction,\n    set::MOI.AbstractScalarSet;\n    allow_modify_function::Bool = false,\n)\n\nReturn the func-in-set constraint in normalized form. That is, if func is MOI.ScalarQuadraticFunction or MOI.ScalarAffineFunction, the constant is moved to the set. If allow_modify_function is true then the function func can be modified.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utility identifies those constraints imposing bounds on a given variable, and returns those bound values:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.get_bounds","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.get_bounds","page":"API Reference","title":"MathOptInterface.Utilities.get_bounds","text":"get_bounds(model::MOI.ModelLike, ::Type{T}, x::MOI.VariableIndex)\n\nReturn a tuple (lb, ub) of type Tuple{T, T}, where lb and ub are lower  and upper bounds, respectively, imposed on x in model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are useful when working with symmetric matrix cones.","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.is_diagonal_vectorized_index\nUtilities.side_dimension_for_vectorized_dimension","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.is_diagonal_vectorized_index","page":"API Reference","title":"MathOptInterface.Utilities.is_diagonal_vectorized_index","text":"is_diagonal_vectorized_index(index::Base.Integer)\n\nReturn whether index is the index of a diagonal element in a MOI.AbstractSymmetricMatrixSetTriangle set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.side_dimension_for_vectorized_dimension","page":"API Reference","title":"MathOptInterface.Utilities.side_dimension_for_vectorized_dimension","text":"side_dimension_for_vectorized_dimension(n::Integer)\n\nReturn the dimension d such that MOI.dimension(MOI.PositiveSemidefiniteConeTriangle(d)) is n.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#Set-utilities","page":"API Reference","title":"Set utilities","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"The following utilities are available for sets:","category":"page"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.AbstractDistance\nUtilities.ProjectionUpperBoundDistance\nUtilities.distance_to_set\nUtilities.set_dot","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.AbstractDistance","page":"API Reference","title":"MathOptInterface.Utilities.AbstractDistance","text":"abstract type AbstractDistance end\n\nAn abstract type used to enabble dispatch of Utilities.distance_to_set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.ProjectionUpperBoundDistance","page":"API Reference","title":"MathOptInterface.Utilities.ProjectionUpperBoundDistance","text":"ProjectionUpperBoundDistance() <: AbstractDistance\n\nAn upper bound on the minimum distance between point and the closest feasible point in set.\n\nDefinition of distance\n\nThe minimum distance is computed as:\n\nd(x mathcalK) = min_y in mathcalK  x - y \n\nwhere x is point and mathcalK is set. The norm is computed as:\n\nx = sqrtf(x x mathcalK)\n\nwhere f is Utilities.set_dot.\n\nIn the default case, where the set does not have a specialized method for Utilities.set_dot, the norm is equivalent to the Euclidean norm x = sqrtsum x_i^2.\n\nWhy an upper bound?\n\nIn most cases, distance_to_set should return the smallest upper bound, but it may return a larger value if the smallest upper bound is expensive to compute.\n\nFor example, given an epigraph from of a conic set, (t x)  f(x) le t, it may be simpler to return delta such that f(x) le t + delta, rather than computing the nearest projection onto the set.\n\nIf the distance is not the smallest upper bound, the docstring of the appropriate distance_to_set method must describe the way that the distance is computed.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.distance_to_set","page":"API Reference","title":"MathOptInterface.Utilities.distance_to_set","text":"distance_to_set(\n    [d::AbstractDistance = ProjectionUpperBoundDistance()],]\n    point::T,\n    set::MOI.AbstractScalarSet,\n) where {T}\n\ndistance_to_set(\n    [d::AbstractDistance = ProjectionUpperBoundDistance(),]\n    point::AbstractVector{T},\n    set::MOI.AbstractVectorSet,\n) where {T}\n\nCompute the distance between point and set using the distance metric d. If point is in the set set, this function must return zero(T).\n\nIf d is omitted, the default distance is Utilities.ProjectionUpperBoundDistance.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.set_dot","page":"API Reference","title":"MathOptInterface.Utilities.set_dot","text":"set_dot(x::AbstractVector, y::AbstractVector, set::AbstractVectorSet)\n\nReturn the scalar product between a vector x of the set set and a vector y of the dual of the set s.\n\n\n\n\n\nset_dot(x, y, set::AbstractScalarSet)\n\nReturn the scalar product between a number x of the set set and a number y of the dual of the set s.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#DoubleDicts","page":"API Reference","title":"DoubleDicts","text":"","category":"section"},{"location":"moi/submodules/Utilities/reference/","page":"API Reference","title":"API Reference","text":"Utilities.DoubleDicts.DoubleDict\nUtilities.DoubleDicts.DoubleDictInner\nUtilities.DoubleDicts.IndexDoubleDict\nUtilities.DoubleDicts.IndexDoubleDictInner\nUtilities.DoubleDicts.outer_keys\nUtilities.DoubleDicts.nonempty_outer_keys","category":"page"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.DoubleDict","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.DoubleDict","text":"DoubleDict{V}\n\nAn optimized dictionary to map MOI.ConstraintIndex to values of type V.\n\nWorks as a AbstractDict{MOI.ConstraintIndex,V} with minimal differences.\n\nIf V is also a MOI.ConstraintIndex, use IndexDoubleDict.\n\nNote that MOI.ConstraintIndex is not a concrete type, opposed to MOI.ConstraintIndex{MOI.VariableIndex, MOI.Integers}, which is a concrete type.\n\nWhen looping through multiple keys of the same Function-in-Set type, use\n\ninner = dict[F, S]\n\nto return a type-stable DoubleDictInner.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.DoubleDictInner","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.DoubleDictInner","text":"DoubleDictInner{F,S,V}\n\nA type stable inner dictionary of DoubleDict.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.IndexDoubleDict","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.IndexDoubleDict","text":"IndexDoubleDict\n\nA specialized version of [DoubleDict] in which the values are of type MOI.ConstraintIndex\n\nWhen looping through multiple keys of the same Function-in-Set type, use\n\ninner = dict[F, S]\n\nto return a type-stable IndexDoubleDictInner.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.IndexDoubleDictInner","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.IndexDoubleDictInner","text":"IndexDoubleDictInner{F,S}\n\nA type stable inner dictionary of IndexDoubleDict.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.outer_keys","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.outer_keys","text":"outer_keys(d::AbstractDoubleDict)\n\nReturn an iterator over the outer keys of the AbstractDoubleDict d. Each outer key is a Tuple{Type,Type} so that a double loop can be easily used:\n\nfor (F, S) in DoubleDicts.outer_keys(dict)\n    for (k, v) in dict[F, S]\n        # ...\n    end\nend\n\nFor performance, it is recommended that the inner loop lies in a separate function to gurantee type-stability. Some outer keys (F, S) might lead to an empty dict[F, S]. If you want only nonempty dict[F, S], use nonempty_outer_keys.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Utilities/reference/#MathOptInterface.Utilities.DoubleDicts.nonempty_outer_keys","page":"API Reference","title":"MathOptInterface.Utilities.DoubleDicts.nonempty_outer_keys","text":"nonempty_outer_keys(d::AbstractDoubleDict)\n\nReturn a vector of outer keys of the AbstractDoubleDict d.\n\nOnly outer keys that have a nonempty set of inner keys will be returned.\n\nEach outer key is a Tuple{Type,Type} so that a double loop can be easily used\n\nfor (F, S) in DoubleDicts.nonempty_outer_keys(dict)\n    for (k, v) in dict[F, S]\n        # ...\n    end\nend\nFor performance, it is recommended that the inner loop lies in a separate\nfunction to gurantee type-stability.\n\nIf you want an iterator of all current outer keys, use [`outer_keys`](@ref).\n\n\n\n\n\n","category":"function"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"EditURL = \"https://github.com/jump-dev/Clp.jl/blob/v1.0.3/README.md\"","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"(Image: )","category":"page"},{"location":"packages/Clp/#Clp.jl","page":"jump-dev/Clp.jl","title":"Clp.jl","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Clp.jl is a wrapper for the COIN-OR Linear Programming solver.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Clp/#Affiliation","page":"jump-dev/Clp.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Clp/#License","page":"jump-dev/Clp.jl","title":"License","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Clp.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The underlying solver, coin-or/Clp, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Clp/#Installation","page":"jump-dev/Clp.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Install Clp using Pkg.add:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"import Pkg\nPkg.add(\"Clp\")","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"In addition to installing the Clp.jl package, this will also download and install the Clp binaries. You do not need to install Clp separately.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Clp/#Use-with-JuMP","page":"jump-dev/Clp.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"To use Clp with JuMP, use Clp.Optimizer:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"using JuMP, Clp\nmodel = Model(Clp.Optimizer)\nset_attribute(model, \"LogLevel\", 1)\nset_attribute(model, \"Algorithm\", 4)","category":"page"},{"location":"packages/Clp/#MathOptInterface-API","page":"jump-dev/Clp.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The Clp optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/Clp/#Options","page":"jump-dev/Clp.jl","title":"Options","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Options are, unfortunately, not well documented.","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The following options are likely to be the most useful:","category":"page"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"Parameter Example Explanation\nPrimalTolerance 1e-7 Primal feasibility tolerance\nDualTolerance 1e-7 Dual feasibility tolerance\nDualObjectiveLimit 1e308 When using dual simplex (where the objective is monotonically changing), terminate when the objective exceeds this limit\nMaximumIterations 2147483647 Terminate after performing this number of simplex iterations\nMaximumSeconds -1.0 Terminate after this many seconds have passed. A negative value means no time limit\nLogLevel 1 Set to 1, 2, 3, or 4 for increasing output. Set to 0 to disable output\nPresolveType 0 Set to 1 to disable presolve\nSolveType 5 Solution method: dual simplex (0), primal simplex (1), sprint (2), barrier with crossover (3), barrier without crossover (4), automatic (5)\nInfeasibleReturn 0 Set to 1 to return as soon as the problem is found to be infeasible (by default, an infeasibility proof is computed as well)\nScaling 3 0 -off, 1 equilibrium, 2 geometric, 3 auto, 4 dynamic(later)\nPerturbation 100 switch on perturbation (50), automatic (100), don't try perturbing (102)","category":"page"},{"location":"packages/Clp/#C-API","page":"jump-dev/Clp.jl","title":"C API","text":"","category":"section"},{"location":"packages/Clp/","page":"jump-dev/Clp.jl","title":"jump-dev/Clp.jl","text":"The C API can be accessed via Clp.Clp_XXX functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/index.md\"","category":"page"},{"location":"moi/#moi_documentation","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"warning: Warning\nThis documentation in this section is a copy of the official MathOptInterface documentation available at https://jump.dev/MathOptInterface.jl/v1.20.1. It is included here to make it easier to link concepts between JuMP and MathOptInterface.","category":"page"},{"location":"moi/#What-is-MathOptInterface?","page":"Introduction","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"MathOptInterface.jl (MOI) is an abstraction layer designed to provide a unified interface to mathematical optimization solvers so that users do not need to understand multiple solver-specific APIs.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"tip: Tip\nThis documentation is aimed at developers writing software interfaces to solvers and modeling languages using the MathOptInterface API. If you are a user interested in solving optimization problems, we encourage you instead to use MOI through a higher-level modeling interface like JuMP or Convex.jl.","category":"page"},{"location":"moi/#How-the-documentation-is-structured","page":"Introduction","title":"How the documentation is structured","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"The Tutorials section contains articles on how to use and implement the MathOptInteraface API. Look here if you want to write a model in MOI, or write an interface to a new solver.\nThe Manual contains short code-snippets that explain how to use the MOI API. Look here for more details on particular areas of MOI.\nThe Background section contains articles on the theory behind MathOptInterface. Look here if you want to understand why, rather than how.\nThe API Reference contains a complete list of functions and types that comprise the MOI API. Look here is you want to know how to use (or implement) a particular function.\nThe Submodules section contains stand-alone documentation for each of the submodules within MOI. These submodules are not required to interface a solver with MOI, but they make the job much easier.","category":"page"},{"location":"moi/#Citing-MathOptInterface","page":"Introduction","title":"Citing MathOptInterface","text":"","category":"section"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"A paper describing the design and features of MathOptInterface is available on arXiv.","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"If you find MathOptInterface useful in your work, we kindly request that you cite the following paper:","category":"page"},{"location":"moi/","page":"Introduction","title":"Introduction","text":"@article{legat2021mathoptinterface,\n    title={{MathOptInterface}: a data structure for mathematical optimization problems},\n    author={Legat, Beno{\\^\\i}t and Dowson, Oscar and Garcia, Joaquim Dias and Lubin, Miles},\n    journal={INFORMS Journal on Computing},\n    year={2021},\n    doi={10.1287/ijoc.2021.1067},\n    publisher={INFORMS}\n}","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"EditURL = \"https://github.com/osqp/OSQP.jl/blob/443706e34c2619acbe65281c60bbe850ca4a8fac/README.md\"","category":"page"},{"location":"packages/OSQP/#OSQP.jl","page":"osqp/OSQP.jl","title":"OSQP.jl","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"(Image: Build Status)  (Image: codecov.io)","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"OSQP.jl is a Julia wrapper for OSQP: the Operator Splitting QP Solver.","category":"page"},{"location":"packages/OSQP/#License","page":"osqp/OSQP.jl","title":"License","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"OSQP.jl is licensed under the Apache-2.0 license.","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"The upstream solver, osqp/osqp is also licensed under the Apache-2.0 license.","category":"page"},{"location":"packages/OSQP/#Installation","page":"osqp/OSQP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"Install OSQP.jl using the Julia package manager","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"import Pkg\nPkg.add(\"OSQP\")","category":"page"},{"location":"packages/OSQP/#Problem-class","page":"osqp/OSQP.jl","title":"Problem class","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"The OSQP (Operator Splitting Quadratic Program) solver is a numerical optimization package for solving problems in the form","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"minimize        0.5 x' P x + q' x\n\nsubject to      l <= A x <= u","category":"page"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"where x in R^n is the optimization variable. The objective function is defined by a positive semidefinite matrix P in S^n_+ and vector q in R^n. The linear constraints are defined by matrix A in R^{m x n} and vectors l in R^m U {-inf}^m, u in R^m U {+inf}^m.","category":"page"},{"location":"packages/OSQP/#Documentation","page":"osqp/OSQP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/OSQP/","page":"osqp/OSQP.jl","title":"osqp/OSQP.jl","text":"Detailed documentation is available at https://osqp.org/.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS, SCS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/models/#jump_models","page":"Models","title":"Models","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP models are the fundamental building block that we use to construct optimization problems. They hold things like the variables and constraints, as well as which solver to use and even solution information.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nJuMP uses \"optimizer\" as a synonym for \"solver.\" Our convention is to use \"solver\" to refer to the underlying software, and use \"optimizer\" to refer to the Julia object that wraps the solver. For example, HiGHS is a solver, and HiGHS.Optimizer is an optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nSee Supported solvers for a list of available solvers.","category":"page"},{"location":"manual/models/#Create-a-model","page":"Models","title":"Create a model","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Create a model by passing an optimizer to Model:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If you don't know which optimizer you will be using at creation time, create a model without an optimizer, and then call set_optimizer at any time prior to optimize!:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> set_optimizer(model, HiGHS.Optimizer)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nDon't know what the fields Model mode and CachingOptimizer state mean? Read the Backends section.","category":"page"},{"location":"manual/models/#What-is-the-difference?","page":"Models","title":"What is the difference?","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"For most models, there is no difference between passing the optimizer to Model, and calling set_optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if an optimizer does not support a constraint in the model, the timing of when an error will be thrown can differ:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If you pass an optimizer, an error will be thrown when you try to add the constraint.\nIf you call set_optimizer, an error will be thrown when you try to solve the model via optimize!.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Therefore, most users should pass an optimizer to Model because it provides the earliest warning that your solver is not suitable for the model you are trying to build. However, if you are modifying a problem by adding and deleting different constraint types, you may need to use set_optimizer. See Switching optimizer for the relaxed problem for an example of when this is useful.","category":"page"},{"location":"manual/models/#Reducing-time-to-first-solve-latency","page":"Models","title":"Reducing time-to-first-solve latency","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, JuMP uses bridges to reformulate the model you are building into an equivalent model supported by the solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if your model is already supported by the solver, bridges add latency (read The \"time-to-first-solve\" issue). This is particularly noticeable for small models.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"To reduce the \"time-to-first-solve,s\" try passing add_bridges = false.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer; add_bridges = false);","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"or","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> set_optimizer(model, HiGHS.Optimizer; add_bridges = false)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, be wary. If your model and solver combination needs bridges, an error will be thrown:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(SCS.Optimizer; add_bridges = false);\n\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, 2x <= 1)\nERROR: Constraints of type MathOptInterface.ScalarAffineFunction{Float64}-in-MathOptInterface.LessThan{Float64} are not supported by the solver.\n\nIf you expected the solver to support your problem, you may have an error in your formulation. Otherwise, consider using a different solver.\n\nThe list of available solvers, along with the problem types they support, is available at https://jump.dev/JuMP.jl/stable/installation/#Supported-solvers.\n[...]","category":"page"},{"location":"manual/models/#Solvers-which-expect-environments","page":"Models","title":"Solvers which expect environments","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Some solvers accept (or require) positional arguments such as a license environment or a path to a binary executable. For these solvers, you can pass a function to Model which takes zero arguments and returns an instance of the optimizer.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common use-case for this is passing an environment or sub-solver to the optimizer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> import HiGHS\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer))\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: MOA[algorithm=MultiObjectiveAlgorithms.Lexicographic, optimizer=HiGHS]","category":"page"},{"location":"manual/models/#solver_options","page":"Models","title":"Solver options","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP uses \"attribute\" as a synonym for \"option.\" Use optimizer_with_attributes to create an optimizer with some attributes initialized:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(\n           optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false),\n       )\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Alternatively, use set_attribute to set an attribute after the model has been created:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_attribute(model, \"output_flag\", false)\n\njulia> get_attribute(model, \"output_flag\")\nfalse","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"You can also modify attributes within an optimizer_with_attributes object:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> solver = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => true);\n\njulia> get_attribute(solver, \"output_flag\")\ntrue\n\njulia> set_attribute(solver, \"output_flag\", false)\n\njulia> get_attribute(solver, \"output_flag\")\nfalse\n\njulia> model = Model(solver);","category":"page"},{"location":"manual/models/#Changing-the-number-types","page":"Models","title":"Changing the number types","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, the coefficients of affine and quadratic expressions are numbers of type either Float64 or Complex{Float64} (see Complex number support).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The type Float64 can be changed using the GenericModel constructor:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = GenericModel{Rational{BigInt}}();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, expr, 1 // 3 * x)\n1//3 x\n\njulia> typeof(expr)\nGenericAffExpr{Rational{BigInt}, GenericVariableRef{Rational{BigInt}}}","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Using a value_type other than Float64 is an advanced operation and should be used only if the underlying solver actually solves the problem using the provided value type.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nNonlinear Modeling is currently restricted to the Float64 number type.","category":"page"},{"location":"manual/models/#Print-the-model","page":"Models","title":"Print the model","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default, show(model) will print a summary of the problem:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(); @variable(model, x >= 0); @objective(model, Max, x);\n\njulia> model\nA JuMP Model\nMaximization problem with:\nVariable: 1\nObjective function type: VariableRef\n`VariableRef`-in-`MathOptInterface.GreaterThan{Float64}`: 1 constraint\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use print to print the formulation of the model (in IJulia, this will render as LaTeX.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> print(model)\nMax x\nSubject to\n x ≥ 0","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nThis format is specific to JuMP and may change in any future release. It is not intended to be an instance format. To write the model to a file, use write_to_file instead.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use latex_formulation to display the model in LaTeX form.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> latex_formulation(model)\n$$ \\begin{aligned}\n\\max\\quad & x\\\\\n\\text{Subject to} \\quad & x \\geq 0\\\\\n\\end{aligned} $$","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"In IJulia (and Documenter), ending a cell in with latex_formulation will render the model in LaTeX:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP                # hide\nmodel = Model()           # hide\n@variable(model, x >= 0)  # hide\n@objective(model, Max, x) # hide\nlatex_formulation(model)","category":"page"},{"location":"manual/models/#Turn-off-output","page":"Models","title":"Turn off output","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use set_silent and unset_silent to disable or enable printing output from the solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> unset_silent(model)","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"tip: Tip\nMost solvers will also have a solver-specific option to provide finer-grained control over the output. Consult their README's for details.","category":"page"},{"location":"manual/models/#Set-a-time-limit","page":"Models","title":"Set a time limit","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use set_time_limit_sec, unset_time_limit_sec, and time_limit_sec to manage time limits.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_time_limit_sec(model, 60.0)\n\n\njulia> time_limit_sec(model)\n60.0\n\njulia> unset_time_limit_sec(model)\n\njulia> limit = time_limit_sec(model)\n\njulia> limit === nothing\ntrue","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"If your time limit is encoded as a Dates.Period object, use the following code to convert it to Float64 for set_time_limit_sec:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> import Dates\n\njulia> seconds(x::Dates.Period) = 1e-3 * Dates.value(round(x, Dates.Millisecond))\nseconds (generic function with 1 method)\n\njulia> set_time_limit_sec(model, seconds(Dates.Hour(1)))\n\njulia> time_limit_sec(model)\n3600.0","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nSome solvers do not support time limits. In these cases, an error will be thrown.","category":"page"},{"location":"manual/models/#Write-a-model-to-file","page":"Models","title":"Write a model to file","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP can write models to a variety of file-formats using write_to_file and Base.write.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For most common file formats, the file type will be detected from the extension.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For example, here is how to write an MPS file:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> write_to_file(model, \"model.mps\")","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Other supported file formats include:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":".cbf for the Conic Benchmark Format\n.lp for the LP file format\n.mof.json for the MathOptFormat\n.nl for AMPL's NL file format\n.rew for the REW file format\n.sdpa and \".dat-s\" for the SDPA file format","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"To write to a specific io::IO, use Base.write. Specify the file type by passing a MOI.FileFormats.FileFormat enum.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> io = IOBuffer();\n\njulia> write(io, model; format = MOI.FileFormats.FORMAT_MPS)","category":"page"},{"location":"manual/models/#Read-a-model-from-file","page":"Models","title":"Read a model from file","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"JuMP models can be created from file formats using read_from_file and Base.read.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = read_from_file(\"model.mps\")\nA JuMP Model\nMinimization problem with:\nVariables: 0\nObjective function type: AffExpr\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> seekstart(io);\n\njulia> model2 = read(io, Model; format = MOI.FileFormats.FORMAT_MPS)\nA JuMP Model\nMinimization problem with:\nVariables: 0\nObjective function type: AffExpr\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"note: Note\nBecause file formats do not serialize the containers of JuMP variables and constraints, the names in the model will not be registered. Therefore, you cannot access named variables and constraints via model[:x]. Instead, use variable_by_name or constraint_by_name to access specific variables or constraints.","category":"page"},{"location":"manual/models/#Relax-integrality","page":"Models","title":"Relax integrality","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use relax_integrality to remove any integrality constraints from the model, such as integer and binary restrictions on variables. relax_integrality returns a function that can be later called with zero arguments to re-add the removed constraints:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model();\n\njulia> @variable(model, x, Int)\nx\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n1\n\njulia> undo = relax_integrality(model);\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n0\n\njulia> undo()\n\njulia> num_constraints(model, VariableRef, MOI.Integer)\n1","category":"page"},{"location":"manual/models/#Switching-optimizer-for-the-relaxed-problem","page":"Models","title":"Switching optimizer for the relaxed problem","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common reason for relaxing integrality is to compute dual variables of the relaxed problem. However, some mixed-integer linear solvers (for example, Cbc) do not return dual solutions, even if the problem does not have integrality restrictions.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Therefore, after relax_integrality you should call set_optimizer with a solver that does support dual solutions, such as Clp.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"For example, instead of:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP, Cbc\nmodel = Model(Cbc.Optimizer)\n@variable(model, x, Int)\nundo = relax_integrality(model)\noptimize!(model)\nreduced_cost(x)  # Errors","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"do:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"using JuMP, Cbc, Clp\nmodel = Model(Cbc.Optimizer)\n@variable(model, x, Int)\nundo = relax_integrality(model)\nset_optimizer(model, Clp.Optimizer)\noptimize!(model)\nreduced_cost(x)  # Works","category":"page"},{"location":"manual/models/#Backends","page":"Models","title":"Backends","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nThis section discusses advanced features of JuMP. For new users, you may want to skip this section. You don't need to know how JuMP manages problems behind the scenes to create and solve JuMP models.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A JuMP Model is a thin layer around a backend of type MOI.ModelLike that stores the optimization problem and acts as the optimization solver.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"However, if you construct a model like Model(HiGHS.Optimizer), the backend is not a HiGHS.Optimizer, but a more complicated object.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"From JuMP, the MOI backend can be accessed using the backend function. Let's see what the backend of a JuMP Model is:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> b = backend(model)\nMOIU.CachingOptimizer{MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\n  with 0 variable bridges\n  with 0 constraint bridges\n  with 0 objective bridges\n  with inner model A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Uh oh. Even though we passed a HiGHS.Optimizer, the backend is a much more complicated object.","category":"page"},{"location":"manual/models/#CachingOptimizer","page":"Models","title":"CachingOptimizer","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"A MOIU.CachingOptimizer is a layer that abstracts the difference between solvers that support incremental modification (for example, they support adding variables one-by-one), and solvers that require the entire problem in a single API call (for example, they only accept the A, b and c matrices of a linear program).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"It has two parts:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A cache, where the model can be built and modified incrementally\njulia> b.model_cache\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\nAn optimizer, which is used to solve the problem\njulia> b.optimizer\nMOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"info: Info\nThe LazyBridgeOptimizer section explains what a LazyBridgeOptimizer is.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The CachingOptimizer has logic to decide when to copy the problem from the cache to the optimizer, and when it can efficiently update the optimizer in-place.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A CachingOptimizer may be in one of three possible states:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"NO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer, and it is not synchronized with the cached model.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A CachingOptimizer has two modes of operation:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"AUTOMATIC: The CachingOptimizer changes its state when necessary. For example, optimize! will automatically call attach_optimizer (an optimizer must have been previously set). Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\nMANUAL: The user must change the state of the CachingOptimizer using MOIU.reset_optimizer(::JuMP.Model), MOIU.drop_optimizer(::JuMP.Model), and MOIU.attach_optimizer(::JuMP.Model). Attempting to perform an operation in the incorrect state results in an error.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"By default Model will create a CachingOptimizer in AUTOMATIC mode.","category":"page"},{"location":"manual/models/#LazyBridgeOptimizer","page":"Models","title":"LazyBridgeOptimizer","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"The second layer that JuMP applies automatically is a MOI.Bridges.LazyBridgeOptimizer. A MOI.Bridges.LazyBridgeOptimizer is an MOI layer that attempts to transform the problem from the formulation provided by the user into an equivalent problem supported by the solver. This may involve adding new variables and constraints to the optimizer. The transformations are selected from a set of known recipes called bridges.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A common example of a bridge is one that splits an interval constraint like @constraint(model, 1 <= x + y <= 2) into two constraints, @constraint(model, x + y >= 1) and @constraint(model, x + y <= 2).","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Use the add_bridges = false keyword to remove the bridging layer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = Model(HiGHS.Optimizer; add_bridges = false)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS\n\njulia> backend(model)\nMOIU.CachingOptimizer{HiGHS.Optimizer, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer A HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"Bridges can be added and removed from a MOI.Bridges.LazyBridgeOptimizer using add_bridge and remove_bridge. Use print_active_bridges to see which bridges are used to reformulate the model. Read the Ellipsoid approximation tutorial for more details.","category":"page"},{"location":"manual/models/#Unsafe-backend","page":"Models","title":"Unsafe backend","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"In some advanced use-cases, it is necessary to work with the inner optimization model directly. To access this model, use unsafe_backend:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> backend(model)\nMOIU.CachingOptimizer{MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}, MOIU.UniversalFallback{MOIU.Model{Float64}}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.UniversalFallback{MOIU.Model{Float64}}\n  fallback for MOIU.Model{Float64}\nwith optimizer MOIB.LazyBridgeOptimizer{HiGHS.Optimizer}\n  with 0 variable bridges\n  with 0 constraint bridges\n  with 0 objective bridges\n  with inner model A HiGHS model with 0 columns and 0 rows.\n\njulia> unsafe_backend(model)\nA HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nbackend and unsafe_backend are advanced routines. Read their docstrings to understand the caveats of their usage, and only call them if you wish to access low-level solver-specific functions.","category":"page"},{"location":"manual/models/#Direct-mode","page":"Models","title":"Direct mode","text":"","category":"section"},{"location":"manual/models/","page":"Models","title":"Models","text":"Using a CachingOptimizer results in an additional copy of the model being stored by JuMP in the .model_cache field. To avoid this overhead, create a JuMP model using direct_model:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = direct_model(HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nSolvers that do not support incremental modification do not support direct_model. An error will be thrown, telling you to use a CachingOptimizer instead.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"The benefit of using direct_model is that there are no extra layers (for example, Cachingoptimizer or LazyBridgeOptimizer) between model and the provided optimizer:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> backend(model)\nA HiGHS model with 0 columns and 0 rows.","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"A downside of direct mode is that there is no bridging layer. Therefore, only constraints which are natively supported by the solver are supported. For example, HiGHS.jl does not implement quadratic constraints:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"julia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x[1]^2 + x[2]^2 <= 2)\nERROR: Constraints of type MathOptInterface.ScalarQuadraticFunction{Float64}-in-MathOptInterface.LessThan{Float64} are not supported by the solver.\n\nIf you expected the solver to support your problem, you may have an error in your formulation. Otherwise, consider using a different solver.\n\nThe list of available solvers, along with the problem types they support, is available at https://jump.dev/JuMP.jl/stable/installation/#Supported-solvers.\nStacktrace:","category":"page"},{"location":"manual/models/","page":"Models","title":"Models","text":"warning: Warning\nAnother downside of direct mode is that the behavior of querying solution information after modifying the problem is solver-specific. This can lead to errors, or the solver silently returning an incorrect value. See OptimizeNotCalled errors for more information.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/duality.md\"","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/duality/#Duality","page":"Duality","title":"Duality","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Conic duality is the starting point for MOI's duality conventions. When all functions are affine (or coordinate projections), and all constraint sets are closed convex cones, the model may be called a conic optimization problem.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For a minimization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_x in mathbbR^n  a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and the dual is a maximization problem in standard conic form:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_y_1 ldots y_m  -sum_i=1^m b_i^T y_i + b_0\n\n textst  a_0 - sum_i=1^m A_i^T y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For a maximization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_x in mathbbR^n  a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and the dual is a minimization problem in standard conic form:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_y_1 ldots y_m  sum_i=1^m b_i^T y_i + b_0\n\n textst  a_0 + sum_i=1^m A_i^T y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"A linear inequality constraint a^T x + b ge c is equivalent to a^T x + b - c in mathbbR_+, and a^T x + b le c is equivalent to a^T x + b - c in mathbbR_-. Variable-wise constraints are affine constraints with the appropriate identity mapping in place of A_i.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For the special case of minimization LPs, the MOI primal form can be stated as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_x in mathbbR^n  a_0^T x + b_0\n\n textst\nA_1 x  ge b_1\n A_2 x  le b_2\n A_3 x  = b_3\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"By applying the stated transformations to conic form, taking the dual, and transforming back into linear inequality form, one obtains the following dual:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_y_1y_2y_3  b_1^Ty_1 + b_2^Ty_2 + b_3^Ty_3 + b_0\n\n textst\nA_1^Ty_1 + A_2^Ty_2 + A_3^Ty_3  = a_0\n y_1 ge 0\n y_2 le 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For maximization LPs, the MOI primal form can be stated as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n max_x in mathbbR^n  a_0^T x + b_0\n\n textst\nA_1 x  ge b_1\n A_2 x  le b_2\n A_3 x  = b_3\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and similarly, the dual is:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n min_y_1y_2y_3  -b_1^Ty_1 - b_2^Ty_2 - b_3^Ty_3 + b_0\n\n textst\nA_1^Ty_1 + A_2^Ty_2 + A_3^Ty_3  = -a_0\n y_1 ge 0\n y_2 le 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"warning: Warning\nFor the LP case, the signs of the feasible dual variables depend only on the sense of the corresponding primal inequality and not on the objective sense.","category":"page"},{"location":"moi/background/duality/#Duality-and-scalar-product","page":"Duality","title":"Duality and scalar product","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The scalar product is different from the canonical one for the sets PositiveSemidefiniteConeTriangle, LogDetConeTriangle, RootDetConeTriangle.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If the set C_i of the section Duality is one of these three cones, then the rows of the matrix A_i corresponding to off-diagonal entries are twice the value of the coefficients field in the VectorAffineFunction for the corresponding rows. See PositiveSemidefiniteConeTriangle for details.","category":"page"},{"location":"moi/background/duality/#Dual-for-problems-with-quadratic-functions","page":"Duality","title":"Dual for problems with quadratic functions","text":"","category":"section"},{"location":"moi/background/duality/#Quadratic-Programs-(QPs)","page":"Duality","title":"Quadratic Programs (QPs)","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"For quadratic programs with only affine conic constraints,","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign*\n min_x in mathbbR^n  frac12x^TQ_0x + a_0^T x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign*","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with cones mathcalC_i subseteq mathbbR^m_i for i = 1 ldots m, consider the Lagrangian function","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x + a_0^T x + b_0 - sum_i = 1^m y_i^T (A_i x + b_i)","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let z(y) denote sum_i = 1^m A_i^T y_i - a_0, the Lagrangian can be rewritten as","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x - z(y)^T x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The condition nabla_x L(x y) = 0 gives","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"0 = nabla_x L(x y) = Q_0x + a_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"which gives Q_0x = z(y). This allows to obtain that","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"min_x in mathbbR^n L(x y) = -frac12x^TQ_0x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so the dual problem is","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"max_y_i in mathcalC_i^* min_x in mathbbR^n -frac12x^TQ_0x + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If Q_0 is invertible, we have x = Q_0^-1z(y) hence","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"min_x in mathbbR^n L(x y) = -frac12z(y)^TQ_0^-1z(y) + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so the dual problem is","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"max_y_i in mathcalC_i^* -frac12z(y)^TQ_0^-1z(y) + b_0 - sum_i = 1^m y_i^T b_i","category":"page"},{"location":"moi/background/duality/#Quadratically-Constrained-Quadratic-Programs-(QCQPs)","page":"Duality","title":"Quadratically Constrained Quadratic Programs (QCQPs)","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Given a problem with both quadratic function and quadratic objectives:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign*\n min_x in mathbbR^n  frac12x^TQ_0x + a_0^T x + b_0\n\n textst  frac12x^TQ_ix + a_i^T x + b_i  in mathcalC_i  i = 1 ldots m\nendalign*","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with cones mathcalC_i subseteq mathbbR for i = 1 ldots m, consider the Lagrangian function","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"L(x y) = frac12x^TQ_0x + a_0^T x + b_0 - sum_i = 1^m y_i (frac12x^TQ_ix + a_i^T x + b_i)","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"A pair of primal-dual variables (x^star y^star) is optimal if","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"x^star is a minimizer of\nmin_x in mathbbR^n L(x y^star)\nThat is,\n0 = nabla_x L(x y^star) = Q_0x + a_0 - sum_i = 1^m y_i^star (Q_ix + a_i)\nand y^star is a maximizer of\nmax_y_i in mathcalC_i^* L(x^star y)\nThat is, for all i = 1 ldots m, frac12x^TQ_ix + a_i^T x + b_i is either zero or in the normal cone of mathcalC_i^* at y^star. For instance, if mathcalC_i is  z in mathbbR  z le 0 , this means that if frac12x^TQ_ix + a_i^T x + b_i is nonzero at x^star then y_i^star = 0. This is the classical complementary slackness condition.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If mathcalC_i is a vector set, the discussion remains valid with y_i(frac12x^TQ_ix + a_i^T x + b_i) replaced with the scalar product between y_i and the vector of scalar-valued quadratic functions.","category":"page"},{"location":"moi/background/duality/#Dual-for-square-semidefinite-matrices","page":"Duality","title":"Dual for square semidefinite matrices","text":"","category":"section"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"The set PositiveSemidefiniteConeTriangle is a self-dual. That is, querying ConstraintDual of a PositiveSemidefiniteConeTriangle constraint returns a vector that is itself a member of PositiveSemidefiniteConeTriangle.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"However, the dual of PositiveSemidefiniteConeSquare is not so straight forward. This section explains the duality convention we use, and how it is derived.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"info: Info\nIf you have a PositiveSemidefiniteConeSquare constraint, the result matrix A from ConstraintDual is not positive semidefinite. However, A + A^top is positive semidefinite.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let mathcalS_+ be the cone of symmetric semidefinite matrices in the fracn(n+1)2 dimensional space of symmetric mathbbR^n times n matrices. That is, mathcalS_+ is the set PositiveSemidefiniteConeTriangle. It is well known that mathcalS_+ is a self-dual proper cone.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Let mathcalP_+ be the cone of symmetric semidefinite matrices in the n^2 dimensional space of mathbbR^n times n matrices. That is mathcalP_+ is the set PositiveSemidefiniteConeSquare.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"In addition, let mathcalD_+ be the cone of matrices A such that A+A^top in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"mathcalP_+ is not proper because it is not solid (it is not n^2 dimensional), so it is not necessarily true that mathcalP_+^** = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"However, this is the case, because we will show that mathcalP_+^* = mathcalD_+ and mathcalD_+^* = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"First, let us see why mathcalP_+^* = mathcalD_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If B is symmetric, then","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"langle AB rangle = langle A^top B^top rangle = langle A^top Brangle","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"so","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"2langle A B rangle = langle A B rangle + langle A^top B rangle = langle A + A^top  B rangle","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Therefore, langle ABrangle ge 0 for all B in mathcalP_+ if and only if langle A+A^topBrangle ge 0 for all B in mathcalP_+. Since A+A^top is symmetric, and we know that mathcalS_+ is self-dual, we have shown that mathcalP_+^* is the set of matrices A such that A+A^top in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Second, let us see why mathcalD_+^* = mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Since A in mathcalD_+ implies that A^top in mathcalD_+, B in mathcalD_+^* means that langle A+A^topBrangle ge 0 for all A in mathcalD_+, and hence B in mathcalP_+.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"To see why it should be symmetric, simply notice that if B_ij  B_ji, then langle ABrangle can be made arbitrarily small by setting A_ij = A_ij + s and A_ji = A_ji - s, with s arbitrarily large, and A stays in mathcalD_+ because A+A^top does not change.","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"Typically, the primal/dual pair for semidefinite programs is presented as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C X rangle \ntextst   langle A_k Xrangle = b_k forall k \n             X                     in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum_k b_k y_k \ntextst   C - sum A_k y_k in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"If we allow A_k to be non-symmetric, we should instead use:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C X rangle \ntextst   langle A_k Xrangle = b_k forall k \n             X                     in mathcalD_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum b_k y_k \ntextst   C - sum A_k y_k in mathcalP_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"This is implemented as:","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       min  langle C   Z rangle + langle C - C^top S rangle \ntextst   langle A_k Z rangle + langle A_k - A_k^top S rangle = b_k forall k \n             Z in mathcalS_+\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"with the dual","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"beginalign\n       max  sum b_k y_k \ntextst   C+C^top - sum (A_k+A_k^top) y_k in mathcalS_+ \n             C-C^top - sum(A_k-A_k^top) y_k = 0\nendalign","category":"page"},{"location":"moi/background/duality/","page":"Duality","title":"Duality","text":"and we recover Z = X + X^top.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#conic_tips_and_tricks","page":"Tips and Tricks","title":"Tips and Tricks","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"This tutorial is aimed at providing a simplistic introduction to conic programming using JuMP.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It uses the following packages:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"using JuMP\nimport SCS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"import Random      # hide\nRandom.seed!(1234) # hide\nnothing            # hide","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"tip: Tip\nA good resource for learning more about functions which can be modeled using cones is the MOSEK Modeling Cookbook.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Background-theory","page":"Tips and Tricks","title":"Background theory","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A subset C of a vector space V is a cone if forall x in C and positive scalars lambda  0, the product lambda x in C.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A cone C is a convex cone if lambda x + (1 - lambda) y in C, for any lambda in 0 1, and any x y in C.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Conic programming problems are convex optimization problems in which a convex function is minimized over the intersection of an affine subspace and a convex cone. An example of a conic-form minimization problems, in the primal form is:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n min_x in mathbbR^n  a_0^T x + b_0 \n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The corresponding dual problem is:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max_y_1 ldots y_m  -sum_i=1^m b_i^T y_i + b_0 \n textst  a_0 - sum_i=1^m A_i^T y_i  = 0 \n  y_i  in mathcalC_i^*  i = 1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Second-Order-Cone","page":"Tips and Tricks","title":"Second-Order Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The SecondOrderCone (or Lorentz Cone) of dimension n is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_soc =  (t x) in mathbbR^n  t ge x_2 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It is most commonly used to represent the L2-norm of the vector x:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, sum(x) == 1)\n@constraint(model, [t; x] in SecondOrderCone())\n@objective(model, Min, t)\noptimize!(model)\nvalue(t), value.(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Rotated-Second-Order-Cone","page":"Tips and Tricks","title":"Rotated Second-Order Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A Second-Order Cone rotated by pi4 in the (x_1x_2) plane is called a RotatedSecondOrderCone. It is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_rsoc =  (tux) in mathbbR^n  2tu ge x_2^2 tu ge 0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When u = 0.5, it represents the sum of squares of a vector x:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"data = [1.0, 2.0, 3.0, 4.0]\ntarget = [0.45, 1.04, 1.51, 1.97]\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, θ)\n@variable(model, t)\n@expression(model, residuals, θ * data .- target)\n@constraint(model, [t; 0.5; residuals] in RotatedSecondOrderCone())\n@objective(model, Min, t)\noptimize!(model)\nvalue(θ), value(t)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Exponential-Cone","page":"Tips and Tricks","title":"Exponential Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.ExponentialCone is a set of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_exp =  (xyz) in mathbbR^3  y exp (xy) le z y  0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"It can be used to model problems involving log and exp.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Exponential","page":"Tips and Tricks","title":"Exponential","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model exp(x) le z, use (x, 1, z):","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x == 1.5)\n@variable(model, z)\n@objective(model, Min, z)\n@constraint(model, [x, 1, z] in MOI.ExponentialCone())\noptimize!(model)\nvalue(z), exp(1.5)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Logarithm","page":"Tips and Tricks","title":"Logarithm","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model x le log(z), use (x, 1, z):","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x)\n@variable(model, z == 1.5)\n@objective(model, Max, x)\n@constraint(model, [x, 1, z] in MOI.ExponentialCone())\noptimize!(model)\nvalue(x), log(1.5)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Log-sum-exp","page":"Tips and Tricks","title":"Log-sum-exp","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"To model t ge logleft(sum e^x_iright), use:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"N = 3\nx0 = rand(N)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:N] == x0[i])\n@variable(model, t)\n@objective(model, Min, t)\n@variable(model, u[1:N])\n@constraint(model, sum(u) <= 1)\n@constraint(model, [i = 1:N], [x[i] - t, 1, u[i]] in MOI.ExponentialCone())\noptimize!(model)\nvalue(t), log(sum(exp.(x0)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Entropy","page":"Tips and Tricks","title":"Entropy","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The entropy maximization problem consists of maximizing the entropy function, H(x) = -xlogx subject to linear inequality constraints.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max  - sum_i=1^n x_i log x_i \n textst  mathbf1^top x = 1 \n  Ax leq b\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"We can model this problem using an exponential cone by using the following transformation:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"tleq -xlogx iff tleq xlog(1x)  iff (t x 1) in K_exp","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Thus, our problem becomes,","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\n max  1^Tt \n textst  Ax leq b \n  1^T x = 1 \n  (t_i x_i 1) in K_exp  forall i = 1 ldots n \nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"m, n = 10, 15\nA, b = randn(m, n), rand(m, 1)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t[1:n])\n@variable(model, x[1:n])\n@objective(model, Max, sum(t))\n@constraint(model, sum(x) == 1)\n@constraint(model, A * x .<= b)\n@constraint(model, [i = 1:n], [t[i], x[i], 1] in MOI.ExponentialCone())\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.ExponentialCone has a dual, the MOI.DualExponentialCone, that offers an alternative formulation that can be more efficient for some formulations.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"There is also the MOI.RelativeEntropyCone for explicitly encoding the relative entropy function","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, x[1:n])\n@objective(model, Max, -t)\n@constraint(model, sum(x) == 1)\n@constraint(model, A * x .<= b)\n@constraint(model, [t; ones(n); x] in MOI.RelativeEntropyCone(2n + 1))\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#PowerCone","page":"Tips and Tricks","title":"PowerCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.PowerCone is a three-dimensional set parameterized by a scalar value α. It has the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_p =  (xyz) in mathbbR^3  x^alpha y^1-alpha ge z x ge 0 y ge 0 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The power cone permits a number of reformulations. For example, when p  1, we can model t ge x^p using the power cone (t 1 x) with alpha = 1  p. Thus, to model t ge x^3 with x ge 0","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, x >= 1.5)\n@constraint(model, [t, 1, x] in MOI.PowerCone(1 / 3))\n@objective(model, Min, t)\noptimize!(model)\nvalue(t), value(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.PowerCone has a dual, the MOI.DualPowerCone, that offers an alternative formulation that can be more efficient for some formulations.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#P-Norm","page":"Tips and Tricks","title":"P-Norm","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The p-norm x_p = left(sumlimits_i x_i^pright)^frac1p can be modeled using MOI.PowerCones. See the Mosek Modeling Cookbook for the derivation.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"function p_norm(x::Vector, p)\n    N = length(x)\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, r[1:N])\n    @variable(model, t)\n    @constraint(model, [i = 1:N], [r[i], t, x[i]] in MOI.PowerCone(1 / p))\n    @constraint(model, sum(r) == t)\n    @objective(model, Min, t)\n    optimize!(model)\n    return value(t)\nend\n\nx = rand(5);\nLinearAlgebra.norm(x, 4), p_norm(x, 4)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Positive-Semidefinite-Cone","page":"Tips and Tricks","title":"Positive Semidefinite Cone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The set of positive semidefinite matrices (PSD) of dimension n form a cone in mathbbR^n. We write this set mathematically as:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"mathcalS_+^n =  X in mathcalS^n mid z^T X z geq 0  forall zin mathbbR^n ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A PSD cone is represented in JuMP using the MOI sets PositiveSemidefiniteConeTriangle (for upper triangle of a PSD matrix) and PositiveSemidefiniteConeSquare (for a complete PSD matrix). However, it is preferable to use the PSDCone shortcut as illustrated below.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Example:-largest-eigenvalue-of-a-symmetric-matrix","page":"Tips and Tricks","title":"Example: largest eigenvalue of a symmetric matrix","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"Suppose A has eigenvalues lambda_1 geq lambda_2 ldots geq lambda_n. Then the matrix t I-A has eigenvalues t-lambda_1 t-lambda_2 ldots t-lambda_n. Note that t I-A is PSD exactly when all these eigenvalues are non-negative, and this happens for values t geq lambda_1. Thus, we can model the problem of finding the largest eigenvalue of a symmetric matrix as:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"beginaligned\nlambda_1 = min t \ntext  st  t I-A succeq 0\nendaligned","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"A = [3 2 4; 2 0 2; 4 2 3]\nI = Matrix{Float64}(LinearAlgebra.I, 3, 3)\nmodel = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@objective(model, Min, t)\n@constraint(model, t .* I - A in PSDCone())\noptimize!(model)\nobjective_value(model)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#GeometricMeanCone","page":"Tips and Tricks","title":"GeometricMeanCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.GeometricMeanCone is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K_geo =  (t x) in mathbbR^n  x ge 0 t le sqrtn-1x_1 x_2 cdots x_n-1 ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, x[1:4])\n@variable(model, t)\n@constraint(model, sum(x) == 1)\n@constraint(model, [t; x] in MOI.GeometricMeanCone(5))\noptimize!(model)\nvalue(t), value.(x)","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#RootDetCone","page":"Tips and Tricks","title":"RootDetCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.RootDetConeSquare is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K =  (t X) in mathbbR^1+d^2  t le det(X)^frac1d ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, X[1:2, 1:2])\n@objective(model, Max, t)\n@constraint(model, [t; vec(X)] in MOI.RootDetConeSquare(2))\n@constraint(model, X .== [2 1; 1 3])\noptimize!(model)\nvalue(t), sqrt(LinearAlgebra.det(value.(X)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"If X is symmetric, then you can use MOI.RootDetConeTriangle instead. This can be more efficient because the solver does not need to add additional constraints to ensure X is symmetric.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When forming the function, use triangle_vec to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, X[1:2, 1:2], Symmetric)\n@objective(model, Max, t)\n@constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(2))\n@constraint(model, X .== [2 1; 1 3])\noptimize!(model)\nvalue(t), sqrt(LinearAlgebra.det(value.(X)))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#LogDetCone","page":"Tips and Tricks","title":"LogDetCone","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"The MOI.LogDetConeSquare is a cone of the form:","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"K =  (t u X) in mathbbR^2+d^2  t le u log(det(X  u)) ","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, u)\n@variable(model, X[1:2, 1:2])\n@objective(model, Max, t)\n@constraint(model, [t; u; vec(X)] in MOI.LogDetConeSquare(2))\n@constraint(model, X .== [2 1; 1 3])\n@constraint(model, u == 0.5)\noptimize!(model)\nvalue(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"If X is symmetric, then you can use MOI.LogDetConeTriangle instead. This can be more efficient because the solver does not need to add additional constraints to ensure X is symmetric.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"When forming the function, use triangle_vec to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.","category":"page"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"model = Model(SCS.Optimizer)\nset_silent(model)\n@variable(model, t)\n@variable(model, u)\n@variable(model, X[1:2, 1:2], Symmetric)\n@objective(model, Max, t)\n@constraint(model, [t; u; triangle_vec(X)] in MOI.LogDetConeTriangle(2))\n@constraint(model, X .== [2 1; 1 3])\n@constraint(model, u == 0.5)\noptimize!(model)\nvalue(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))","category":"page"},{"location":"tutorials/conic/tips_and_tricks/#Other-Cones-and-Functions","page":"Tips and Tricks","title":"Other Cones and Functions","text":"","category":"section"},{"location":"tutorials/conic/tips_and_tricks/","page":"Tips and Tricks","title":"Tips and Tricks","text":"For other cones supported by JuMP, check out the MathOptInterface Manual.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"EditURL = \"https://github.com/jump-dev/BARON.jl/blob/v0.8.2/README.md\"","category":"page"},{"location":"packages/BARON/#BARON.jl","page":"jump-dev/BARON.jl","title":"BARON.jl","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"BARON.jl is a wrapper for BARON by The Optimization Firm.","category":"page"},{"location":"packages/BARON/#Affiliation","page":"jump-dev/BARON.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"This wrapper is maintained by the JuMP community and is not officially supported by The Optimization Firm.","category":"page"},{"location":"packages/BARON/#License","page":"jump-dev/BARON.jl","title":"License","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"BARON.jl is licensed under the MIT License.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license from The Optimization Firm, although a small trial version is available for free.","category":"page"},{"location":"packages/BARON/#Installation","page":"jump-dev/BARON.jl","title":"Installation","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"First, download a copy of the BARON solver and unpack the executable in a location of your choosing.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"Once installed, set the BARON_EXEC environment variable pointing to the BARON executable (full path, including file name as it differs across platforms), and run Pkg.add(\"BARON\"). For example:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"ENV[\"BARON_EXEC\"] = \"/path/to/baron.exe\"\nusing Pkg\nPkg.add(\"BARON\")","category":"page"},{"location":"packages/BARON/#Use-with-JuMP","page":"jump-dev/BARON.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"using JuMP, BARON\nmodel = Model(BARON.Optimizer)","category":"page"},{"location":"packages/BARON/#MathOptInterface-API","page":"jump-dev/BARON.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"The BARON optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarNonlinearFunction}","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.Reals","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarNonlinearFunction in MOI.EqualTo{Float64}\nMOI.ScalarNonlinearFunction in MOI.GreaterThan{Float64}\nMOI.ScalarNonlinearFunction in MOI.Interval{Float64}\nMOI.ScalarNonlinearFunction in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/BARON/","page":"jump-dev/BARON.jl","title":"jump-dev/BARON.jl","text":"MOI.NLPBlock()\nMOI.ObjectiveSense()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"EditURL = \"getting_started_with_julia.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Getting-started-with-Julia","page":"Getting started with Julia","title":"Getting started with Julia","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because JuMP is embedded in Julia, knowing some basic Julia is important before you start learning JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nThis tutorial is designed to provide a minimalist crash course in the basics of Julia. You can find resources that provide a more comprehensive introduction to Julia here.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Installing-Julia","page":"Getting started with Julia","title":"Installing Julia","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"To install Julia, download the latest stable release, then follow the platform specific install instructions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nUnless you know otherwise, you probably want the 64-bit version.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Next, you need an IDE to develop in. VS Code is a popular choice, so follow these install instructions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia can also be used with Jupyter notebooks or the reactive notebooks of Pluto.jl.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#The-Julia-REPL","page":"Getting started with Julia","title":"The Julia REPL","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The main way of interacting with Julia is via its REPL (Read Evaluate Print Loop). To access the REPL, start the Julia executable to arrive at the julia> prompt, and then start coding:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"As your programs become larger, write a script as a text file, and then run that file using:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"julia> include(\"path/to/file.jl\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nBecause of Julia's startup latency, running scripts from the command line like the following is slow:$ julia path/to/file.jlUse the REPL or a notebook instead, and read The \"time-to-first-solve\" issue for more information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Code-blocks-in-this-documentation","page":"Getting started with Julia","title":"Code blocks in this documentation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In this documentation you'll see a mix of code examples with and without the julia>.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The Julia prompt is mostly used to demonstrate short code snippets, and the output is exactly what you will see if run from the REPL.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Blocks without the julia> can be copy-pasted into the REPL, but they are used because they enable richer output like plots or LaTeX to be displayed in the online and PDF versions of the documentation. If you run them from the REPL you may see different output.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Where-to-get-help","page":"Getting started with Julia","title":"Where to get help","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Read the documentation\nJuMP https://jump.dev/JuMP.jl/stable/\nJulia https://docs.julialang.org/en/v1/\nAsk (or browse) the Julia community forum: https://discourse.julialang.org\nIf the question is JuMP-related, ask in the Optimization (Mathematical) section, or tag your question with \"jump\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"To access the built-in help at the REPL, type ? to enter help-mode, followed by the name of the function to lookup:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"help?> print\nsearch: print println printstyled sprint isprint prevind parentindices precision escape_string\n\n  print([io::IO], xs...)\n\n  Write to io (or to the default output stream stdout if io is not given) a canonical\n  (un-decorated) text representation. The representation used by print includes minimal formatting\n  and tries to avoid Julia-specific details.\n\n  print falls back to calling show, so most types should just define show. Define print if your\n  type has a separate \"plain\" representation. For example, show displays strings with quotes, and\n  print displays strings without quotes.\n\n  string returns the output of print as a string.\n\n  Examples\n  ≡≡≡≡≡≡≡≡≡≡\n\n  julia> print(\"Hello World!\")\n  Hello World!\n  julia> io = IOBuffer();\n\n  julia> print(io, \"Hello\", ' ', :World!)\n\n  julia> String(take!(io))\n  \"Hello World!\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Numbers-and-arithmetic","page":"Getting started with Julia","title":"Numbers and arithmetic","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Since we want to solve optimization problems, we're going to be using a lot of math. Luckily, Julia is great for math, with all the usual operators:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1\n1 - 2\n2 * 2\n4 / 5\n3^2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Did you notice how Julia didn't print .0 after some of the numbers? Julia is a dynamic language, which means you never have to explicitly declare the type of a variable. However, in the background, Julia is giving each variable a type. Check the type of something using the typeof function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(1)\ntypeof(1.0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Here 1 is an Int64, which is an integer with 64 bits of precision, and 1.0 is a Float64, which is a floating point number with 64-bits of precision.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nIf you aren't familiar with floating point numbers, make sure to read the Floating point numbers section.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We create complex numbers using im:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = 2 + 1im\nreal(x)\nimag(x)\ntypeof(x)\nx * (1 - 2im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nThe curly brackets surround what we call the parameters of a type. You can read Complex{Int64}  as \"a complex number, where the real and imaginary parts are represented by Int64.\" If we call typeof(1.0 + 2.0im) it will be Complex{Float64}, which a complex number with the parts represented by Float64.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"There are also some cool things like an irrational representation of π.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"π","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nTo make π (and most other Greek letters), type \\pi and then press [TAB].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(π)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"However, if we do math with irrational numbers, they get converted to Float64:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(2π / 3)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Floating-point-numbers","page":"Getting started with Julia","title":"Floating point numbers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nIf you aren't familiar with floating point numbers, make sure to read this section carefully.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A Float64 is a floating point approximation of a real number using 64-bits of information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because it is an approximation, things we know hold true in mathematics don't hold true in a computer. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 == 0.3","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A more complicated example is:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"sin(2π / 3) == √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGet √ by typing \\sqrt then press [TAB].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Let's see what the differences are:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 - 0.3\nsin(2π / 3) - √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"They are small, but not zero.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"One way of explaining this difference is to consider how we would write 1 / 3 and 2 / 3 using only four digits after the decimal point. We would write 1 / 3 as 0.3333, and 2 / 3 as 0.6667. So, despite the fact that 2 * (1 / 3) == 2 / 3, 2 * 0.3333 == 0.6666 != 0.6667.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Let's try that again using ≈ (\\approx + [TAB]) instead of ==:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"0.1 * 3 ≈ 0.3\nsin(2π / 3) ≈ √3 / 2","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"≈ is a clever way of calling the isapprox function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(sin(2π / 3), √3 / 2; atol = 1e-8)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nFloating point is the reason solvers use tolerances when they solve optimization models. A common mistake you're likely to make is checking whether a binary variable is 0 using value(z) == 0. Always remember to use something like isapprox when comparing floating point numbers.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that isapprox will always return false if one of the number being compared is 0 and atol is zero (its default value).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1e-300 ≈ 0.0","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"so always set a nonzero value of atol if one of the arguments can be zero.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(1e-9, 0.0; atol = 1e-8)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGurobi has a good series of articles on the implications of floating point in optimization if you want to read more.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"If you aren't careful, floating point arithmetic can throw up all manner of issues. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1e-16 == 1","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"It even turns out that floating point numbers aren't associative:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"(1 + 1e-16) - 1e-16 == 1 + (1e-16 - 1e-16)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"It's important to note that this issue isn't Julia-specific. It happens in every programming language (try it out in Python).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Vectors,-matrices,-and-arrays","page":"Getting started with Julia","title":"Vectors, matrices, and arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to MATLAB, Julia has native support for vectors, matrices and tensors; all of which are represented by arrays of different dimensions. Vectors are constructed by comma-separated elements surrounded by square brackets:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"b = [5, 6]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Matrices can be constructed with spaces separating the columns, and semicolons separating the rows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A = [1.0 2.0; 3.0 4.0]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We can do linear algebra:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = A \\ b","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nHere is floating point at work again; x is approximately [-4, 4.5].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A * x\nA * x ≈ b","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that when multiplying vectors and matrices, dimensions matter. For example, you can't multiply a vector by a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    b * b\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"But multiplying transposes works:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"b' * b\nb * b'","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Other-common-types","page":"Getting started with Julia","title":"Other common types","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/#Comments","page":"Getting started with Julia","title":"Comments","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Although not technically a type, code comments begin with the # character:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"1 + 1  # This is a comment","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Multiline comments begin with #= and end with =#:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"#=\nHere is a\nmultiline comment\n=#","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Comments can even be nested inside expressions. This is sometimes helpful when documenting inputs to functions:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isapprox(\n    sin(π),\n    0.0;\n    #= We need an explicit atol here because we are comparing with 0 =#\n    atol = 0.001,\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Strings","page":"Getting started with Julia","title":"Strings","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Double quotes are used for strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(\"This is Julia\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Unicode is fine in strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(\"π is about 3.1415\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use println to print a string:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"println(\"Hello, World!\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use $() to interpolate values into a string:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"x = 123\nprintln(\"The value of x is: $(x)\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Use triple-quotes for multiline strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"s = \"\"\"\nHere is\na\nmultiline string\n\"\"\"\n\nprintln(s)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Symbols","page":"Getting started with Julia","title":"Symbols","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia Symbols are a data structure from the compiler that represent Julia identifiers (that is, variable names).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"println(\"The value of x is: $(eval(:x))\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"warning: Warning\nWe used eval here to demonstrate how Julia links Symbols to variables. However, avoid calling eval in your code. It is usually a sign that your code is doing something that could be more easily achieved a different way. The Community Forum is a good place to ask for advice on alternative approaches.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"typeof(:x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can think of a Symbol as a String that takes up less memory, and that can't be modified.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Convert between String and Symbol using their constructors:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"String(:abc)\nSymbol(\"abc\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nSymbols are often (ab)used to stand in for a String or an Enum, when one of the latter is likely a better choice. The JuMP Style guide recommends reserving Symbols for identifiers. See @enum vs. Symbol for more.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Tuples","page":"Getting started with Julia","title":"Tuples","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia makes extensive use of a simple data structure called Tuples. Tuples are immutable collections of values. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t = (\"hello\", 1.2, :foo)\ntypeof(t)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Tuples can be accessed by index, similar to arrays:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t[2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"And they can be \"unpacked\" like so:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"a, b, c = t\nb","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The values can also be given names, which is a convenient way of making light-weight data structures.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t = (word = \"hello\", num = 1.2, sym = :foo)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Values can be accessed using dot syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"t.word","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Dictionaries","page":"Getting started with Julia","title":"Dictionaries","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to Python, Julia has native support for dictionaries. Dictionaries provide a very generic way of mapping keys to values.  For example, a map of integers to strings:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d1 = Dict(1 => \"A\", 2 => \"B\", 4 => \"D\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nType-stuff again: Dict{Int64,String} is a dictionary with Int64 keys and String values.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Looking up a value uses the bracket syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d1[2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dictionaries support non-integer keys and can mix data types:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"A\" => 1, \"B\" => 2.5, \"D\" => 2 - 3im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nJulia types form a hierarchy. Here the value type of the dictionary is Number, which is a generalization of Int64, Float64, and Complex{Int}. Leaf nodes in this hierarchy are called \"concrete\" types, and all others are called \"Abstract.\" In general, having variables with abstract types like Number can lead to slower code, so you should try to make sure every element in a dictionary or vector is the same type. For example, in this case we could represent every element as a Complex{Float64}:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"A\" => 1.0 + 0.0im, \"B\" => 2.5 + 0.0im, \"D\" => 2.0 - 3.0im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dictionaries can be nested:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"d2 = Dict(\"A\" => 1, \"B\" => 2, \"D\" => Dict(:foo => 3, :bar => 4))\nd2[\"B\"]\nd2[\"D\"][:foo]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Structs","page":"Getting started with Julia","title":"Structs","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can define custom datastructures with struct:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"struct MyStruct\n    x::Int\n    y::String\n    z::Dict{Int,Int}\nend\n\na = MyStruct(1, \"a\", Dict(2 => 3))\na.x","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"By default, these are not mutable","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    a.x = 2\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"However, you can declare a mutable struct which is mutable:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"mutable struct MyStructMutable\n    x::Int\n    y::String\n    z::Dict{Int,Int}\nend\n\na = MyStructMutable(1, \"a\", Dict(2 => 3))\na.x\na.x = 2\na","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Loops","page":"Getting started with Julia","title":"Loops","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia has native support for for-each style loops with the syntax for <value> in <collection> end:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 1:5\n    println(i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"info: Info\nRanges are constructed as start:stop, or start:step:stop.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 1.2:1.1:5.6\n    println(i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This for-each loop also works with dictionaries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for (key, value) in Dict(\"A\" => 1, \"B\" => 2.5, \"D\" => 2 - 3im)\n    println(\"$(key): $(value)\")\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Note that in contrast to vector languages like MATLAB and R, loops do not result in a significant performance degradation in Julia.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Control-flow","page":"Getting started with Julia","title":"Control flow","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Julia control flow is similar to MATLAB, using the keywords if-elseif-else-end, and the logical operators || and && for or and and respectively:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"for i in 0:5:15\n    if i < 5\n        println(\"$(i) is less than 5\")\n    elseif i < 10\n        println(\"$(i) is less than 10\")\n    else\n        if i == 10\n            println(\"the value is 10\")\n        else\n            println(\"$(i) is bigger than 10\")\n        end\n    end\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Comprehensions","page":"Getting started with Julia","title":"Comprehensions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Similar to languages like Haskell and Python, Julia supports the use of simple loops in the construction of arrays and dictionaries, called comprehensions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A list of increasing integers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i for i in 1:5]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Matrices can be built by including multiple indices:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i * j for i in 1:5, j in 5:10]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Conditional statements can be used to filter out some values:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"[i for i in 1:10 if i % 2 == 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A similar syntax can be used for building dictionaries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Dict(\"$(i)\" => i for i in 1:10 if i % 2 == 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Functions","page":"Getting started with Julia","title":"Functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A simple function is defined as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_hello()\n    return println(\"hello\")\nend\nprint_hello()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Arguments can be added to a function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_it(x)\n    return println(x)\nend\nprint_it(\"hello\")\nprint_it(1.234)\nprint_it(:my_id)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Optional keyword arguments are also possible:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function print_it(x; prefix = \"value:\")\n    return println(\"$(prefix) $(x)\")\nend\nprint_it(1.234)\nprint_it(1.234; prefix = \"val:\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The keyword return is used to specify the return values of a function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function mult(x; y = 2.0)\n    return x * y\nend\n\nmult(4.0)\nmult(4.0; y = 5.0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Anonymous-functions","page":"Getting started with Julia","title":"Anonymous functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The syntax input -> output creates an anonymous function. These are most useful when passed to other functions. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"f = x -> x^2\nf(2)\nmap(x -> x^2, 1:4)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Type-parameters","page":"Getting started with Julia","title":"Type parameters","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"We can constrain the inputs to a function using type parameters, which are :: followed by the type of the input we want. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function foo(x::Int)\n    return x^2\nend\n\nfunction foo(x::Float64)\n    return exp(x)\nend\n\nfunction foo(x::Number)\n    return x + 1\nend\n\nfoo(2)\nfoo(2.0)\nfoo(1 + 1im)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"But what happens if we call foo with something we haven't defined it for?","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"try                         #hide\n    foo([1, 2, 3])\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"A MethodError means that you passed a function something that didn't match the type that it was expecting. In this case, the error message says that it doesn't know how to handle an Vector{Int64}, but it does know how to handle Float64, Int64, and Number.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nRead the \"Closest candidates\" part of the error message carefully to get a hint as to what was expected.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Broadcasting","page":"Getting started with Julia","title":"Broadcasting","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In the example above, we didn't define what to do if f was passed a Vector. Luckily, Julia provides a convenient syntax for mapping f element-wise over arrays. Just add a . between the name of the function and the opening (. This works for any function, including functions with multiple arguments. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"foo.([1, 2, 3])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nGet a MethodError when calling a function that takes a Vector, Matrix, or Array? Try broadcasting.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Mutable-vs-immutable-objects","page":"Getting started with Julia","title":"Mutable vs immutable objects","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Some types in Julia are mutable, which means you can change the values inside them. A good example is an array. You can modify the contents of an array without having to make a new array.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"In contrast, types like Float64 are immutable. You cannot modify the contents of a Float64.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"This is something to be aware of when passing types into functions. For example:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"function mutability_example(mutable_type::Vector{Int}, immutable_type::Int)\n    mutable_type[1] += 1\n    immutable_type += 1\n    return\nend\n\nmutable_type = [1, 2, 3]\nimmutable_type = 1\n\nmutability_example(mutable_type, immutable_type)\n\nprintln(\"mutable_type: $(mutable_type)\")\nprintln(\"immutable_type: $(immutable_type)\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Because Vector{Int} is a mutable type, modifying the variable inside the function changed the value outside of the function. In contrast, the change to immutable_type didn't modify the value outside the function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can check mutability with the isimmutable function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"isimmutable([1, 2, 3])\nisimmutable(1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#The-package-manager","page":"Getting started with Julia","title":"The package manager","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/#Installing-packages","page":"Getting started with Julia","title":"Installing packages","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"No matter how wonderful Julia's base language is, at some point you will want to use an extension package.  Some of these are built-in, for example random number generation is available in the Random package in the standard library. These packages are loaded with the commands using and import.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Random  # The equivalent of Python's `from Random import *`\nimport Random  # The equivalent of Python's `import Random`\n\nRandom.seed!(33)\n\n[rand() for i in 1:10]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"The Package Manager is used to install packages that are not part of Julia's standard library.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"For example the following can be used to install JuMP,","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"For a complete list of registered Julia packages see the package listing at JuliaHub.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"From time to you may wish to use a Julia package that is not registered.  In this case a git repository URL can be used to install the package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"using Pkg\nPkg.add(\"https://github.com/user-name/MyPackage.jl.git\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/#Package-environments","page":"Getting started with Julia","title":"Package environments","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"By default, Pkg.add will add packages to Julia's global environment. However, Julia also has built-in support for virtual environments.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"Activate a virtual environment with:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"import Pkg; Pkg.activate(\"/path/to/environment\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"You can see what packages are installed in the current environment with Pkg.status().","category":"page"},{"location":"tutorials/getting_started/getting_started_with_julia/","page":"Getting started with Julia","title":"Getting started with Julia","text":"tip: Tip\nWe strongly recommend you create a Pkg environment for each project that you create in Julia, and add only the packages that you need, instead of adding lots of packages to the global environment. The Pkg manager documentation has more information on this topic.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#linear_tips_and_tricks","page":"Tips and tricks","title":"Tips and tricks","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"tip: Tip\nA good source of tips is the Mosek Modeling Cookbook.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial collates some tips and tricks you can use when formulating mixed-integer programs. It uses the following packages:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"using JuMP","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Absolute-value","page":"Tips and tricks","title":"Absolute value","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model the absolute value function t ge x, there are a few options. In all cases, these reformulations only work if you are minimizing t \"down\" into x. They do not work if you are trying to maximize x.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-1","page":"Tips and tricks","title":"Option 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option adds two linear inequality constraints:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x)\n@variable(model, t)\n@constraint(model, t >= x)\n@constraint(model, t >= -x)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-2","page":"Tips and tricks","title":"Option 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option uses two non-negative variables and forms expressions for x and t:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, z[1:2] >= 0)\n@expression(model, t, z[1] + z[2])\n@expression(model, x, z[1] - z[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Option-3","page":"Tips and tricks","title":"Option 3","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This option uses MOI.NormOneCone and lets JuMP choose the reformulation:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x)\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormOneCone(2))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#L1-norm","page":"Tips and tricks","title":"L1-norm","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model min x_1, that is, min sumlimits_i x_i, use the MOI.NormOneCone:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormOneCone(1 + length(x)))\n@objective(model, Min, t)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Infinity-norm","page":"Tips and tricks","title":"Infinity-norm","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model min x_infty, that is, min maxlimits_i x_i, use the MOI.NormInfinityCone:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@variable(model, t)\n@constraint(model, [t; x] in MOI.NormInfinityCone(1 + length(x)))\n@objective(model, Min, t)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Max","page":"Tips and tricks","title":"Max","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model t ge maxx y, do:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, t)\n@variable(model, x)\n@variable(model, y)\n@constraint(model, t >= x)\n@constraint(model, t >= y)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This reformulation does not work for t ge minx y.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Min","page":"Tips and tricks","title":"Min","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"To model t le minx y, do:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, t)\n@variable(model, x)\n@variable(model, y)\n@constraint(model, t <= x)\n@constraint(model, t <= y)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This reformulation does not work for t le maxx y.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Boolean-operators","page":"Tips and tricks","title":"Boolean operators","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Binary variables can be used to construct logical operators. Here are some example.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Or","page":"Tips and tricks","title":"Or","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_3 = x_1 lor x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraints(model, begin\n    x[1] <= x[3]\n    x[2] <= x[3]\n    x[3] <= x[1] + x[2]\nend)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#And","page":"Tips and tricks","title":"And","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_3 = x_1 land x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraints(model, begin\n    x[3] <= x[1]\n    x[3] <= x[2]\n    x[3] >= x[1] + x[2] - 1\nend)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Not","page":"Tips and tricks","title":"Not","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_1 neg x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2], Bin)\n@constraint(model, x[1] == 1 - x[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Implies","page":"Tips and tricks","title":"Implies","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x_1 implies x_2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2], Bin)\n@constraint(model, x[1] <= x[2])","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Disjunctions","page":"Tips and tricks","title":"Disjunctions","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/#Problem","page":"Tips and tricks","title":"Problem","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Suppose that we have two constraints a^top x leq b and c^top x leq d, and we want at least one to hold.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick","page":"Tips and tricks","title":"Trick","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Introduce a \"big-M\" multiplied by a binary variable to relax one of the constraints.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example Either x_1 leq 1 or x_2 leq 2.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, y, Bin)\nM = 100\n@constraint(model, x[1] <= 1 + M * y)\n@constraint(model, x[2] <= 2 + M * (1 - y))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"warning: Warning\nIf M is too small, the solution may be suboptimal. If M is too big, the solver may encounter numerical issues. Try to use domain knowledge to choose an M that is just right. Gurobi has a good documentation section on this topic.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Indicator-constraints","page":"Tips and tricks","title":"Indicator constraints","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/#Problem-2","page":"Tips and tricks","title":"Problem","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Suppose we want to model that a certain linear inequality must be satisfied when some other event occurs, that is, for a binary variable z, we want to model the implication:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"z = 1 implies a^top x leq b","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick-1","page":"Tips and tricks","title":"Trick 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Some solvers have native support for indicator constraints.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 1.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\n@constraint(model, z => {sum(x) <= 1})","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\n@constraint(model, !z => {sum(x) <= 1})","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Trick-2","page":"Tips and tricks","title":"Trick 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"If the solver doesn't support indicator constraints, you an use the big-M trick.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 1.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\nM = 100\n@constraint(model, sum(x) <= 1 + M * (1 - z))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x_1 + x_2 leq 1 if z = 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:2])\n@variable(model, z, Bin)\nM = 100\n@constraint(model, sum(x) <= 1 + M * z)","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Semi-continuous-variables","page":"Tips and tricks","title":"Semi-continuous variables","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"info: Info\nThis section uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A semi-continuous variable is a continuous variable between bounds lu that also can assume the value zero, that is: x in 0 cup lu","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Example x in 0cup 1 2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x in Semicontinuous(1.0, 2.0))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Semi-integer-variables","page":"Tips and tricks","title":"Semi-integer variables","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A semi-integer variable is a variable which assumes integer values between bounds lu and can also assume the value zero: x in 0 cup l u cap mathbbZ","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x in Semiinteger(5.0, 10.0))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Special-Ordered-Sets-of-Type-1","page":"Tips and tricks","title":"Special Ordered Sets of Type 1","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A Special Ordered Set of Type 1 is a set of variables, at most one of which can take a non-zero value, all others being at 0.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"They most frequently apply where a set of variables are actually binary variables. In other words, we have to choose at most one from a set of possibilities.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3], Bin)\n@constraint(model, x in SOS1())","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"You can optionally pass SOS1 a weight vector like","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"@constraint(model, x in SOS1([0.2, 0.5, 0.3]))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"If the decision variables are related and have a physical ordering, then the weight vector, although not used directly in the constraint, can help the solver make a better decision in the solution process.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#tip_sos2","page":"Tips and tricks","title":"Special Ordered Sets of Type 2","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A Special Ordered Set of type 2 is a set of non-negative variables, of which at most two can be non-zero, and if two are non-zero these must be consecutive in their ordering.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model();\n@variable(model, x[1:3])\n@constraint(model, x in SOS2([3.0, 1.0, 2.0]))","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"The ordering provided by the weight vector is more important in this case as the variables need to be consecutive according to the ordering. For example, in the above constraint, the possible pairs are:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Consecutive\n(x[1] and x[3]) as they correspond to 3 and 2 resp. and thus can be non-zero\n(x[2] and x[3]) as they correspond to 1 and 2 resp. and thus can be non-zero\nNon-consecutive\n(x[1] and x[2]) as they correspond to 3 and 1 resp. and thus cannot be non-zero","category":"page"},{"location":"tutorials/linear/tips_and_tricks/#Piecewise-linear-approximations","page":"Tips and tricks","title":"Piecewise linear approximations","text":"","category":"section"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"SOSII constraints are most often used to form piecewise linear approximations of a function.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Given a set of points for x:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"x̂ = -1:0.5:2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"and a set of corresponding points for y:","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"ŷ = x̂ .^ 2","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"the piecewise linear approximation is constructed by representing x and y as convex combinations of x̂ and ŷ.","category":"page"},{"location":"tutorials/linear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"N = length(x̂)\nmodel = Model();\n@variable(model, -1 <= x <= 2)\n@variable(model, y)\n@variable(model, 0 <= λ[1:N] <= 1)\n@objective(model, Max, y)\n@constraints(model, begin\n    x == sum(x̂[i] * λ[i] for i in 1:N)\n    y == sum(ŷ[i] * λ[i] for i in 1:N)\n    sum(λ) == 1\n    λ in SOS2()\nend)","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"EditURL = \"web_app.jl\"","category":"page"},{"location":"tutorials/applications/web_app/#Serving-web-apps","page":"Serving web apps","title":"Serving web apps","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"This tutorial demonstrates how to setup and serve JuMP models via a REST API.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"In the example app we are building, we solve a trivial mixed-integer program, which is parameterized by the lower bound of a variable. To call the service, users send an HTTP POST request with JSON contents indicating the lower bound. The returned value is the solution of the mixed-integer program as JSON.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"First, we need JuMP and a solver:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"We also need HTTP.jl to act as our REST server, and JSON.jl to marshal data.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"import HTTP\nimport JSON","category":"page"},{"location":"tutorials/applications/web_app/#The-server-side","page":"Serving web apps","title":"The server side","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"The core components of our REST server are endpoints. These are functions which accept a Dict{String,Any} of input parameters, and return a Dict{String,Any} as output. The types are Dict{String,Any} because we're going to read these to and from JSON.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Here's a very simple endpoint: it accepts params as input, formulates and solves a trivial mixed-integer program, and then returns a dictionary with the result.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function endpoint_solve(params::Dict{String,Any})\n    if !haskey(params, \"lower_bound\")\n        return Dict{String,Any}(\n            \"status\" => \"failure\",\n            \"reason\" => \"missing lower_bound param\",\n        )\n    elseif !(params[\"lower_bound\"] isa Real)\n        return Dict{String,Any}(\n            \"status\" => \"failure\",\n            \"reason\" => \"lower_bound is not a number\",\n        )\n    end\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x >= params[\"lower_bound\"], Int)\n    optimize!(model)\n    ret = Dict{String,Any}(\n        \"status\" => \"okay\",\n        \"terminaton_status\" => termination_status(model),\n        \"primal_status\" => primal_status(model),\n    )\n    # Only include the `x` key if it has a value.\n    if has_values(model)\n        ret[\"x\"] = value(x)\n    end\n    return ret\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"When we call this, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"endpoint_solve(Dict{String,Any}(\"lower_bound\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"endpoint_solve(Dict{String,Any}())","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"For a second function, we need a function that accepts an HTTP.Request object and returns an HTTP.Response object.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function serve_solve(request::HTTP.Request)\n    data = JSON.parse(String(request.body))\n    solution = endpoint_solve(data)\n    return HTTP.Response(200, JSON.json(solution))\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Finally, we need an HTTP server. There are a variety of ways you can do this in HTTP.jl. We use an explicit Sockets.listen so we have manual control of when we shutdown the server.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function setup_server(host, port)\n    server = HTTP.Sockets.listen(host, port)\n    HTTP.serve!(host, port; server = server) do request\n        try\n            # Extend the server by adding other endpoints here.\n            if request.target == \"/api/solve\"\n                return serve_solve(request)\n            else\n                return HTTP.Response(404, \"target $(request.target) not found\")\n            end\n        catch err\n            # Log details about the exception server-side\n            @info \"Unhandled exception: $err\"\n            # Return a response to the client\n            return HTTP.Response(500, \"internal error\")\n        end\n    end\n    return server\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"warning: Warning\nHTTP.jl does not serve requests on a separate thread. Therefore, a long-running job will block the main thread, preventing concurrent users from submitting requests. To work-around this, read HTTP.jl issue 798 or watch Building Microservices and Applications in Julia from JuliaCon 2020.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"server = setup_server(HTTP.ip\"127.0.0.1\", 8080)","category":"page"},{"location":"tutorials/applications/web_app/#The-client-side","page":"Serving web apps","title":"The client side","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Now that we have a server, we can send it requests via this function:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"function send_request(data::Dict; endpoint::String = \"solve\")\n    ret = HTTP.request(\n        \"POST\",\n        # This should match the URL and endpoint we defined for our server.\n        \"http://127.0.0.1:8080/api/$endpoint\",\n        [\"Content-Type\" => \"application/json\"],\n        JSON.json(data),\n    )\n    if ret.status != 200\n        # This could happen if there are time-outs, network errors, etc.\n        return Dict(\n            \"status\" => \"failure\",\n            \"code\" => ret.status,\n            \"body\" => String(ret.body),\n        )\n    end\n    return JSON.parse(String(ret.body))\nend","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Let's see what happens:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => 0))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"If we don't send a lower_bound, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"invalid_param\" => 1.2))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"If we don't send a lower_bound that is a number, we get:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"send_request(Dict(\"lower_bound\" => \"1.2\"))","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"Finally, we can shutdown our HTTP server:","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"close(server)","category":"page"},{"location":"tutorials/applications/web_app/#Next-steps","page":"Serving web apps","title":"Next steps","text":"","category":"section"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"For more complicated examples relating to HTTP servers, consult the HTTP.jl documentation.","category":"page"},{"location":"tutorials/applications/web_app/","page":"Serving web apps","title":"Serving web apps","text":"To see how you can integrate this with a larger JuMP model, read Design patterns for larger models.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"EditURL = \"https://github.com/jump-dev/SDPT3.jl/blob/b565aac2a58818090d521f2340e71f597688e4fb/README.md\"","category":"page"},{"location":"packages/SDPT3/#SDPT3.jl","page":"jump-dev/SDPT3.jl","title":"SDPT3.jl","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"SDPT3.jl is wrapper for the SDPT3 solver.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"an exported sdpt3 function that is a thin wrapper on top of the sdpt3 MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SDPT3/#Affiliation","page":"jump-dev/SDPT3.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SDPT3.","category":"page"},{"location":"packages/SDPT3/#License","page":"jump-dev/SDPT3.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"SDPT3.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The underlying solver, SDPT3 is licensed under the GPL v2 License.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"In addition, SDPT3 requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SDPT3/#Use-with-JuMP","page":"jump-dev/SDPT3.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"To use SDPT3 with JuMP, do:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"using JuMP, SDPT3\nmodel = Model(SDPT3.Optimizer)\nset_attribute(model, \"printlevel\", 0)","category":"page"},{"location":"packages/SDPT3/#Installation","page":"jump-dev/SDPT3.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SeDuMi software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Then, install SDPT3.jl using Pkg.add:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"import Pkg\nPkg.add(\"SDPT3\")","category":"page"},{"location":"packages/SDPT3/#SDPT3-not-in-PATH","page":"jump-dev/SDPT3.jl","title":"SDPT3 not in PATH","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If you get the error:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Error using save\nVariable 'jx_sdpt3_arg_out_1' not found.\n\nERROR: LoadError: MATLAB.MEngineError(\"failed to get variable jx_sdpt3_arg_out_1 from MATLAB session\")\nStacktrace:\n[...]","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"The error means that we could not find the sdpt3 function with one output argument using the MATLAB C API. This most likely means that you did not add SDPT3 to the MATLAB's path (that is, the toolbox/local/pathdef.m file).","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work, where /path/to/sdpt3/ is the directory where the sdpt3 folder is located:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"julia> using MATLAB\n\njulia> cd(\"/path/to/sdpt3/\") do\n           MATLAB.mat\"install_sdpt3\"\n       end\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"An alternative fix is suggested in the following issue.","category":"page"},{"location":"packages/SDPT3/#Error-in-validate","page":"jump-dev/SDPT3.jl","title":"Error in validate","text":"","category":"section"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"If you get the error:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"Brace indexing is not supported for variables of this type.\n\nError in validate\n\nError in sdpt3 (line 171)\n   [blk,At,C,b,blkdim,numblk,parbarrier] = validate(blk,At,C,b,par,parbarrier);\n\nError using save\nVariable 'jx_sdpt3_arg_out_1' not found.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"It might mean that you have added SDPNAL in addition to SDPT3 in the MATLAB's path (that is, the toolbox/local/pathdef.m file). Because SDPNAL also defines a validate function, this can make sdpt3 call SDPNAL's validate function instead of SDPT3's validate function, which causes the issue.","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"One way to fix this from the Julia REPL is to reset the search path to the factory-installed state using restoredefaultpath:","category":"page"},{"location":"packages/SDPT3/","page":"jump-dev/SDPT3.jl","title":"jump-dev/SDPT3.jl","text":"julia> using MATLAB\n\njulia> MATLAB.restoredefaultpath()\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"EditURL = \"dualization.jl\"","category":"page"},{"location":"tutorials/conic/dualization/#Dualization","page":"Dualization","title":"Dualization","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The purpose of this tutorial is to explain how to use Dualization.jl to improve the performance of some conic optimization models. There are two important takeaways:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"JuMP reformulates problems to meet the input requirements of the solver, potentially increasing the problem size by adding slack variables and constraints.\nSolving the dual of a conic model can be more efficient than solving the primal.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Dualization.jl is a package which fixes these problems, allowing you to solve the dual instead of the primal with a one-line change to your code.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"using JuMP\nimport Dualization\nimport SCS","category":"page"},{"location":"tutorials/conic/dualization/#Background","page":"Dualization","title":"Background","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Conic optimization solvers typically accept one of two input formulations.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The first is the standard conic form:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   A x = b  \n                               x in mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"in which we have a set of linear equality constraints Ax = b and the variables belong to a cone mathcalK.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The second is the geometric conic form:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   A x - b in mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"in which an affine function Ax - b belongs to a cone mathcalK and the variables are free.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"It is trivial to convert between these two representations, for example, to go from the geometric conic form to the standard conic form we introduce slack variables y:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst   beginbmatrixA  -Iendbmatrix beginbmatrixxyendbmatrix = b \n                               beginbmatrixxyendbmatrix in mathbbR^n times mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"and to go from the standard conic form to the geometric conic form, we can rewrite the equality constraint as a function belonging to the {0} cone:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"beginalign\n    min_x in mathbbR^n   c^top x \n              textst  beginbmatrixAIendbmatrix x - beginbmatrixb0endbmatrix in 0 times mathcalK\nendalign","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"From a theoretical perspective, the two formulations are equivalent, and if you implement a model in the standard conic form and pass it to a geometric conic form solver (or vice versa), then JuMP will automatically reformulate the problem into the correct formulation.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"From a practical perspective though, the reformulations are problematic because the additional slack variables and constraints can make the problem much larger and therefore harder to solve.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"You should also note many problems contain a mix of conic constraints and variables, and so they do not neatly fall into one of the two formulations. In these cases, JuMP reformulates only the variables and constraints as necessary to convert the problem into the desired form.","category":"page"},{"location":"tutorials/conic/dualization/#Primal-and-dual-formulations","page":"Dualization","title":"Primal and dual formulations","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Duality plays a large role in conic optimization. For a detailed description of conic duality, see Duality.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"A useful observation is that if the primal problem is in standard conic form, then the dual problem is in geometric conic form, and vice versa. Moreover, the primal and dual may have a different number of variables and constraints, although which one is smaller depends on the problem. Therefore, instead of reformulating the problem from one form to the other, it can be more efficient to solve the dual instead of the primal.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"To demonstrate, we use a variation of the Maximum cut via SDP example.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The primal formulation (in standard conic form) is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"model_primal = Model()\n@variable(model_primal, X[1:2, 1:2], PSD)\n@objective(model_primal, Max, sum([1 -1; -1 1] .* X))\n@constraint(model_primal, primal_c[i = 1:2], 1 - X[i, i] == 0)\nprint(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This problem has three scalar decision variables (the matrix X is symmetric), two scalar equality constraints, and a constraint that X is positive semidefinite.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The dual of model_primal is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"model_dual = Model()\n@variable(model_dual, y[1:2])\n@objective(model_dual, Min, sum(y))\n@constraint(model_dual, dual_c, [y[1]-1 1; 1 y[2]-1] in PSDCone())\nprint(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This problem has two scalar decision variables, and a 2x2 positive semidefinite matrix constraint.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"tip: Tip\nIf you haven't seen conic duality before, try deriving the dual problem based on the description in Duality. You'll need to know that the dual cone of PSDCone is the PSDCone.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"When we solve model_primal with SCS.Optimizer, SCS reports three variables (variables n: 3), five rows in the constraint matrix (constraints m: 5), and five non-zeros in the matrix (nnz(A): 5):","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_primal, SCS.Optimizer)\noptimize!(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"(There are five rows in the constraint matrix because SCS expects problems in geometric conic form, and so JuMP has reformulated the X, PSD variable constraint into the affine constraint X .+ 0 in PSDCone().)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The solution we obtain is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(X)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(primal_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"objective_value(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"When we solve model_dual with SCS.Optimizer, SCS reports two variables (variables n: 2), three rows in the constraint matrix (constraints m: 3), and two non-zeros in the matrix (nnz(A): 2):","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_dual, SCS.Optimizer)\noptimize!(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"and the solution we obtain is:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(dual_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(y)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"objective_value(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"This particular problem is small enough that it isn't meaningful to compare the solve times, but in general, we should expect model_dual to solve faster than model_primal because it contains fewer variables and constraints. The difference is particularly noticeable on large-scale optimization problems.","category":"page"},{"location":"tutorials/conic/dualization/#dual_optimizer","page":"Dualization","title":"dual_optimizer","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Manually deriving the conic dual is difficult and error-prone. The package Dualization.jl provides the Dualization.dual_optimizer meta-solver, which wraps any MathOptInterface-compatible solver in an interface that automatically formulates and solves the dual of an input problem.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"To demonstrate, we use Dualization.dual_optimizer to solve model_primal:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_primal, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model_primal)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The performance is the same as if we solved model_dual, and the correct solution is returned to X:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(X)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(primal_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Moreover, if we use dual_optimizer on model_dual, then we get the same performance as if we had solved model_primal:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model_dual, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model_dual)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"dual.(dual_c)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"value.(y)","category":"page"},{"location":"tutorials/conic/dualization/#A-mixed-example","page":"Dualization","title":"A mixed example","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"The Maximum cut via SDP example is nicely defined because the primal is in standard conic form and the dual is in geometric conic form. However, many practical models contain a mix of the two formulations. One example is The minimum distortion problem:","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"D = [0 1 1 1; 1 0 2 2; 1 2 0 2; 1 2 2 0]\nmodel = Model()\n@variable(model, c²)\n@variable(model, Q[1:4, 1:4], PSD)\n@objective(model, Min, c²)\nfor i in 1:4, j in (i+1):4\n    @constraint(model, D[i, j]^2 <= Q[i, i] + Q[j, j] - 2 * Q[i, j])\n    @constraint(model, Q[i, i] + Q[j, j] - 2 * Q[i, j] <= c² * D[i, j]^2)\nend\n@constraint(model, Q[1, 1] == 0)\n@constraint(model, c² >= 1)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"In this formulation, the Q variable is of the form xinmathcalK, but there is also a free variable, c², a linear equality constraint, Q[1, 1] == 0, and some linear inequality constraints. Rather than attempting to derive the formulation that JuMP would pass to SCS and its dual, the simplest solution is to try solving the problem with and without dual_optimizer to see which formulation is most efficient.","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model, SCS.Optimizer)\noptimize!(model)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"set_optimizer(model, Dualization.dual_optimizer(SCS.Optimizer))\noptimize!(model)","category":"page"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"For this problem, SCS reports that the primal has variables n: 11, constraints m: 24 and that the dual has variables n: 14, constraints m: 24. Therefore, we should probably use the primal formulation because it has fewer variables and the same number of constraints.","category":"page"},{"location":"tutorials/conic/dualization/#When-to-use-dual_optimizer","page":"Dualization","title":"When to use dual_optimizer","text":"","category":"section"},{"location":"tutorials/conic/dualization/","page":"Dualization","title":"Dualization","text":"Because it can make the problem larger or smaller, depending on the problem and the choice of solver, there is no definitive rule on when you should use dual_optimizer. However, you should try dual_optimizer if your conic optimization problem takes a long time to solve, or if you need to repeatedly solve similarly structured problems with different data. In some cases solving the dual instead of the primal can make a large difference.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"EditURL = \"ellipse_approx.jl\"","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Ellipsoid-approximation","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial considers the problem of computing extremal ellipsoids: finding ellipsoids that best approximate a given set. As an extension, we show how to use JuMP to inspect the bridges that were used, and how to explore alternative formulations.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The model comes from Section 4.9 \"Applications VII: extremal ellipsoids\" of the book Lectures on Modern Convex Optimization by Ben-Tal and Nemirovski (2001).","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"For a related example, see also the Minimal ellipses tutorial.","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Problem-formulation","page":"Ellipsoid approximation","title":"Problem formulation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Suppose that we are given a set mathcalS consisting of m points in n-dimensional space:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"mathcalS =  x_1 ldots x_m  subset mathbbR^n","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Our goal is to determine an optimal vector c in  mathbbR^n and an optimal n times n real symmetric matrix D such that the ellipse:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"E(D c) =  x  (x - c)^top D ( x - c) leq 1 ","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"contains mathcalS and has the smallest possible volume.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The optimal D and c are given by the optimization problem:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"beginaligned\nmax        quad  t                                                                    \ntextst quad  Z succeq 0 \n                   beginbmatrix s  z^top  z  Z  endbmatrix succeq 0 \n                   x_i^top Z x_i - 2x_i^top z + s leq 1     quad i=1 ldots m \n                   t le sqrtndet(Z)\nendaligned","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"where D = Z_* and c = Z_*^-1 z_*.","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Required-packages","page":"Ellipsoid approximation","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport Random\nimport SCS","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Data","page":"Ellipsoid approximation","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We first need to generate some points to work with.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"function generate_point_cloud(\n    m;            # number of 2-dimensional points\n    a = 10,       # scaling in x direction\n    b = 2,        # scaling in y direction\n    rho = π / 6,  # rotation of points around origin\n    random_seed = 1,\n)\n    rng = Random.MersenneTwister(random_seed)\n    P = randn(rng, Float64, m, 2)\n    Phi = [a*cos(rho) a*sin(rho); -b*sin(rho) b*cos(rho)]\n    S = P * Phi\n    return S\nend","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"For the sake of this example, let's take m = 600:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"S = generate_point_cloud(600);\nnothing #hide","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We will visualise the points (and ellipse) using the Plots package:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"r = 1.1 * maximum(abs.(S))\nplot = Plots.scatter(\n    S[:, 1],\n    S[:, 2];\n    xlim = (-r, r),\n    ylim = (-r, r),\n    label = nothing,\n    c = :green,\n    shape = :x,\n    size = (600, 600),\n)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#JuMP-formulation","page":"Ellipsoid approximation","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Now let's build and the JuMP model. We'll compute D and c after the solve.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"model = Model(SCS.Optimizer)\n# We need to use a tighter tolerance for this example, otherwise the bounding\n# ellipse won't actually be bounding...\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\nm, n = size(S)\n@variable(model, z[1:n])\n@variable(model, Z[1:n, 1:n], PSD)\n@variable(model, s)\n@variable(model, t)\n@constraint(model, [s z'; z Z] >= 0, PSDCone())\n@constraint(\n    model,\n    [i in 1:m],\n    S[i, :]' * Z * S[i, :] - 2 * S[i, :]' * z + s <= 1,\n)\n@constraint(model, [t; vec(Z)] in MOI.RootDetConeSquare(n))\n@objective(model, Max, t)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Results","page":"Ellipsoid approximation","title":"Results","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"After solving the model to optimality we can recover the solution in terms of D and c:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"D = value.(Z)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"c = D \\ value.(z)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Finally, overlaying the solution in the plot we see the minimal volume approximating ellipsoid:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"P = sqrt(D)\nq = -P * c\ndata = [tuple(P \\ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)]\nPlots.plot!(plot, data; c = :crimson, label = nothing)","category":"page"},{"location":"tutorials/conic/ellipse_approx/#Alternative-formulations","page":"Ellipsoid approximation","title":"Alternative formulations","text":"","category":"section"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The formulation of model uses MOI.RootDetConeSquare. However, because SCS does not natively support this cone, JuMP automatically reformulates the problem into an equivalent problem that SCS does support. You can see the reformulation that JuMP chose using print_active_bridges:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"There's a lot going on here, but the first bullet is:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"* Unsupported objective: MOI.VariableIndex\n|  bridged by:\n|   MOIB.Objective.FunctionizeBridge{Float64}\n|  introduces:\n|   * Supported objective: MOI.ScalarAffineFunction{Float64}","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This says that SCS does not support a MOI.VariableIndex objective function, and that JuMP used a MOI.Bridges.Objective.FunctionizeBridge to convert it into a MOI.ScalarAffineFunction{Float64} objective function.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We can leave JuMP to do the reformulation, or we can rewrite our model to have an objective function that SCS natively supports:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"@objective(model, Max, 1.0 * t + 0.0);\nnothing #hide","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Re-printing the active bridges:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"we get * Supported objective: MOI.ScalarAffineFunction{Float64}.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"We can manually implement some other reformulations to change our model to something that SCS more closely supports by:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Replacing the MOI.VectorOfVariables in MOI.PositiveSemidefiniteConeTriangle constraint @variable(model, Z[1:n, 1:n], PSD) with the MOI.VectorAffineFunction in MOI.PositiveSemidefiniteConeTriangle @constraint(model, Z >= 0, PSDCone()).\nReplacing the MOI.VectorOfVariables in MOI.PositiveSemidefiniteConeSquare constraint [s z'; z Z] >= 0, PSDCone() with the MOI.VectorAffineFunction in MOI.PositiveSemidefiniteConeTriangle @constraint(model, LinearAlgebra.Symmetric([s z'; z Z]) >= 0, PSDCone()).\nReplacing the MOI.ScalarAffineFunction in MOI.GreaterThan constraints with the vectorized equivalent of MOI.VectorAffineFunction in MOI.Nonnegatives\nReplacing the MOI.VectorOfVariables in MOI.RootDetConeSquare constraint with MOI.VectorAffineFunction in MOI.RootDetConeTriangle.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Note that we still need to bridge MOI.PositiveSemidefiniteConeTriangle constraints because SCS uses an internal SCS.ScaledPSDCone set instead.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"model = Model(SCS.Optimizer)\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\n@variable(model, z[1:n])\n@variable(model, s)\n@variable(model, t)\n# The former @variable(model, Z[1:n, 1:n], PSD)\n@variable(model, Z[1:n, 1:n], Symmetric)\n@constraint(model, Z >= 0, PSDCone())\n# The former [s z'; z Z] >= 0, PSDCone()\n@constraint(model, LinearAlgebra.Symmetric([s z'; z Z]) >= 0, PSDCone())\n# The former constraint S[i, :]' * Z * S[i, :] - 2 * S[i, :]' * z + s <= 1\nf = [1 - S[i, :]' * Z * S[i, :] + 2 * S[i, :]' * z - s for i in 1:m]\n@constraint(model, f in MOI.Nonnegatives(m))\n# The former constraint [t; vec(Z)] in MOI.RootDetConeSquare(n)\n@constraint(model, 1 * [t; triangle_vec(Z)] .+ 0 in MOI.RootDetConeTriangle(n))\n# The former @objective(model, Max, t)\n@objective(model, Max, 1 * t + 0)\noptimize!(model)\nsolve_time_1 = solve_time(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This formulation gives the much smaller graph:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"The last bullet shows how JuMP reformulated the MOI.RootDetConeTriangle constraint by adding a mix of MOI.PositiveSemidefiniteConeTriangle and MOI.GeometricMeanCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Because SCS doesn't natively support the MOI.GeometricMeanCone, these constraints were further bridged using a MOI.Bridges.Constraint.GeoMeanToPowerBridge to a series of MOI.PowerCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"However, there are many other ways that a MOI.GeometricMeanCone can be reformulated into something that SCS supports. Let's see what happens if we use remove_bridge to remove the MOI.Bridges.Constraint.GeoMeanToPowerBridge:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"remove_bridge(model, MOI.Bridges.Constraint.GeoMeanToPowerBridge)\noptimize!(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This time, the solve took:","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"solve_time_2 = solve_time(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"where previously it took","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"solve_time_1","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Why was the solve time different?","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"print_active_bridges(model)","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"This time, JuMP used a MOI.Bridges.Constraint.GeoMeanBridge to reformulate the constraint into a set of MOI.RotatedSecondOrderCone constraints, which were further reformulated into a set of supported MOI.SecondOrderCone constraints.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"Since the two models are equivalent, we can conclude that for this particular model, the MOI.SecondOrderCone formulation is more efficient.","category":"page"},{"location":"tutorials/conic/ellipse_approx/","page":"Ellipsoid approximation","title":"Ellipsoid approximation","text":"In general though, the performance of a particular reformulation is problem- and solver-specific. Therefore, JuMP chooses to minimize the number of bridges in the default reformulation, leaving you to explore alternative formulations using the tools and techniques shown in this tutorial.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"DocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/containers/#Containers","page":"Containers","title":"Containers","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"JuMP provides specialized containers similar to AxisArrays that enable multi-dimensional arrays with non-integer indices.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"These containers are created automatically by JuMP's macros. Each macro has the same basic syntax:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"@macroname(model, name[key1=index1, index2; optional_condition], other stuff)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The containers are generated by the name[key1=index1, index2; optional_condition] syntax. Everything else is specific to the particular macro.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Containers can be named, for example, name[key=index], or unnamed, for example, [key=index]. We call unnamed containers anonymous.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We call the bits inside the square brackets and before the ; the index sets. The index sets can be named, for example, [i = 1:4], or they can be unnamed, for example, [1:4].","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We call the bit inside the square brackets and after the ; the condition. Conditions are optional.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"In addition to the standard JuMP macros like @variable and @constraint, which construct containers of variables and constraints respectively, you can use Containers.@container to construct containers with arbitrary elements.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"We will use this macro to explain the three types of containers that are natively supported by JuMP: Array, Containers.DenseAxisArray, and Containers.SparseAxisArray.","category":"page"},{"location":"manual/containers/#Array","page":"Containers","title":"Array","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"An Array is created when the index sets are rectangular and the index sets are of the form 1:n.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container(x[i = 1:2, j = 1:3], (i, j))\n2×3 Matrix{Tuple{Int64, Int64}}:\n (1, 1)  (1, 2)  (1, 3)\n (2, 1)  (2, 2)  (2, 3)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The result is a normal Julia Array, so you can do all the usual things.","category":"page"},{"location":"manual/containers/#Slicing","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Arrays can be sliced","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[:, 1]\n2-element Vector{Tuple{Int64, Int64}}:\n (1, 1)\n (2, 1)\n\njulia> x[2, :]\n3-element Vector{Tuple{Int64, Int64}}:\n (2, 1)\n (2, 2)\n (2, 3)","category":"page"},{"location":"manual/containers/#Looping","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(x)\n           println(x[key])\n       end\n(1, 1)\n(2, 1)\n(1, 2)\n(2, 2)\n(1, 3)\n(2, 3)","category":"page"},{"location":"manual/containers/#Get-the-index-sets","page":"Containers","title":"Get the index sets","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use axes to obtain the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> axes(x)\n(Base.OneTo(2), Base.OneTo(3))","category":"page"},{"location":"manual/containers/#Broadcasting","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over an Array returns an Array","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(x)\n2×3 Matrix{Tuple{Int64, Int64}}:\n (1, 1)  (2, 1)  (3, 1)\n (1, 2)  (2, 2)  (3, 2)","category":"page"},{"location":"manual/containers/#Tables","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the Array into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n6-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Int64, Tuple{Int64, Int64}}}}:\n (I = 1, J = 1, value = (1, 1))\n (I = 2, J = 1, value = (2, 1))\n (I = 1, J = 2, value = (1, 2))\n (I = 2, J = 2, value = (2, 2))\n (I = 1, J = 3, value = (1, 3))\n (I = 2, J = 3, value = (2, 3))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n6×3 DataFrame\n Row │ I      J      value\n     │ Int64  Int64  Tuple…\n─────┼──────────────────────\n   1 │     1      1  (1, 1)\n   2 │     2      1  (2, 1)\n   3 │     1      2  (1, 2)\n   4 │     2      2  (2, 2)\n   5 │     1      3  (1, 3)\n   6 │     2      3  (2, 3)","category":"page"},{"location":"manual/containers/#DenseAxisArray","page":"Containers","title":"DenseAxisArray","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"A Containers.DenseAxisArray is created when the index sets are rectangular, but not of the form 1:n. The index sets can be of any type.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x = Containers.@container([i = 1:2, j = [:A, :B]], (i, j))\n2-dimensional DenseAxisArray{Tuple{Int64, Symbol},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/#Slicing-2","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"DenseAxisArrays can be sliced","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[:, :A]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, Base.OneTo(2)\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :A)\n (2, :A)\n\njulia> x[1, :]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, [:A, :B]\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :A)\n (1, :B)","category":"page"},{"location":"manual/containers/#Looping-2","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(x)\n           println(x[key])\n       end\n(1, :A)\n(2, :A)\n(1, :B)\n(2, :B)","category":"page"},{"location":"manual/containers/#Get-the-index-sets-2","page":"Containers","title":"Get the index sets","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use axes to obtain the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> axes(x)\n(Base.OneTo(2), [:A, :B])","category":"page"},{"location":"manual/containers/#Broadcasting-2","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over a DenseAxisArray returns a DenseAxisArray","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(x)\n2-dimensional DenseAxisArray{Tuple{Symbol, Int64},2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{Tuple{Symbol, Int64}}:\n (:A, 1)  (:B, 1)\n (:A, 2)  (:B, 2)","category":"page"},{"location":"manual/containers/#Access-internal-data","page":"Containers","title":"Access internal data","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Array(x) to copy the internal data array into a new Array:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Array(x)\n2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"To access the internal data without a copy, use x.data.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x.data\n2×2 Matrix{Tuple{Int64, Symbol}}:\n (1, :A)  (1, :B)\n (2, :A)  (2, :B)","category":"page"},{"location":"manual/containers/#Tables-2","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the DenseAxisArray into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n4-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Symbol, Tuple{Int64, Symbol}}}}:\n (I = 1, J = :A, value = (1, :A))\n (I = 2, J = :A, value = (2, :A))\n (I = 1, J = :B, value = (1, :B))\n (I = 2, J = :B, value = (2, :B))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n4×3 DataFrame\n Row │ I      J       value\n     │ Int64  Symbol  Tuple…\n─────┼────────────────────────\n   1 │     1  A       (1, :A)\n   2 │     2  A       (2, :A)\n   3 │     1  B       (1, :B)\n   4 │     2  B       (2, :B)","category":"page"},{"location":"manual/containers/#Keyword-indexing","page":"Containers","title":"Keyword indexing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If all axes are named, you can use keyword indexing:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[i = 2, j = :A]\n(2, :A)\n\njulia> x[i = :, j = :B]\n1-dimensional DenseAxisArray{Tuple{Int64, Symbol},1,...} with index sets:\n    Dimension 1, Base.OneTo(2)\nAnd data, a 2-element Vector{Tuple{Int64, Symbol}}:\n (1, :B)\n (2, :B)","category":"page"},{"location":"manual/containers/#SparseAxisArray","page":"Containers","title":"SparseAxisArray","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"A Containers.SparseAxisArray is created when the index sets are non-rectangular. This occurs in two circumstances:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"An index depends on a prior index:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:2, j = i:2], (i, j))\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Int64}, 2, Tuple{Int64, Int64}} with 3 entries:\n  [1, 1]  =  (1, 1)\n  [1, 2]  =  (1, 2)\n  [2, 2]  =  (2, 2)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The [indices; condition] syntax is used:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x = Containers.@container([i = 1:3, j = [:A, :B]; i > 1], (i, j))\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 2, Tuple{Int64, Symbol}} with 4 entries:\n  [2, A]  =  (2, :A)\n  [2, B]  =  (2, :B)\n  [3, A]  =  (3, :A)\n  [3, B]  =  (3, :B)","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Here we have the index sets i = 1:3, j = [:A, :B], followed by ;, and then a condition, which evaluates to true or false: i > 1.","category":"page"},{"location":"manual/containers/#Slicing-3","page":"Containers","title":"Slicing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Slicing is supported:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> y = x[:, :B]\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (2, :B)\n  [3]  =  (3, :B)","category":"page"},{"location":"manual/containers/#Looping-3","page":"Containers","title":"Looping","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use eachindex to loop over the elements:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> for key in eachindex(y)\n           println(y[key])\n       end\n(2, :B)\n(3, :B)","category":"page"},{"location":"manual/containers/#Broadcasting-3","page":"Containers","title":"Broadcasting","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Broadcasting over a SparseAxisArray returns a SparseAxisArray","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> swap(x::Tuple) = (last(x), first(x))\nswap (generic function with 1 method)\n\njulia> swap.(y)\nJuMP.Containers.SparseAxisArray{Tuple{Symbol, Int64}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (:B, 2)\n  [3]  =  (:B, 3)","category":"page"},{"location":"manual/containers/#Tables-3","page":"Containers","title":"Tables","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Use Containers.rowtable to convert the SparseAxisArray into a Tables.jl compatible Vector{<:NamedTuple}:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> table = Containers.rowtable(x; header = [:I, :J, :value])\n4-element Vector{NamedTuple{(:I, :J, :value), Tuple{Int64, Symbol, Tuple{Int64, Symbol}}}}:\n (I = 2, J = :A, value = (2, :A))\n (I = 2, J = :B, value = (2, :B))\n (I = 3, J = :A, value = (3, :A))\n (I = 3, J = :B, value = (3, :B))","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Because it supports the Tables.jl interface, you can pass it to any function which accepts a table as input:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> import DataFrames;\n\njulia> DataFrames.DataFrame(table)\n4×3 DataFrame\n Row │ I      J       value\n     │ Int64  Symbol  Tuple…\n─────┼────────────────────────\n   1 │     2  A       (2, :A)\n   2 │     2  B       (2, :B)\n   3 │     3  A       (3, :A)\n   4 │     3  B       (3, :B)","category":"page"},{"location":"manual/containers/#Keyword-indexing-2","page":"Containers","title":"Keyword indexing","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If all axes are named, you can use keyword indexing:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> x[i = 2, j = :A]\n(2, :A)\n\njulia> x[i = :, j = :B]\nJuMP.Containers.SparseAxisArray{Tuple{Int64, Symbol}, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  (2, :B)\n  [3]  =  (3, :B)","category":"page"},{"location":"manual/containers/#Forcing-the-container-type","page":"Containers","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Pass container = T to use T as the container. For example:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:2, j = 1:2], i + j, container = Array)\n2×2 Matrix{Int64}:\n 2  3\n 3  4\n\njulia> Containers.@container([i = 1:2, j = 1:2], i + j, container = Dict)\nDict{Tuple{Int64, Int64}, Int64} with 4 entries:\n  (1, 2) => 3\n  (1, 1) => 2\n  (2, 2) => 4\n  (2, 1) => 3","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"You can also pass DenseAxisArray or SparseAxisArray.","category":"page"},{"location":"manual/containers/#How-different-container-types-are-chosen","page":"Containers","title":"How different container types are chosen","text":"","category":"section"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If the compiler can prove at compile time that the index sets are rectangular, and indexed by a compact set of integers that start at 1, Containers.@container will return an array. This is the case if your index sets are visible to the macro as 1:n:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i=1:3, j=1:5], i + j)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"or an instance of Base.OneTo:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = Base.OneTo(3)\nBase.OneTo(3)\n\njulia> Containers.@container([i=set, j=1:5], i + j)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"If the compiler can prove that the index set is rectangular, but not necessarily of the form 1:n at compile time, then a Containers.DenseAxisArray will be constructed instead:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = 1:3\n1:3\n\njulia> Containers.@container([i=set, j=1:5], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, Base.OneTo(5)\nAnd data, a 3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"info: Info\nWhat happened here? Although we know that set contains 1:3, at compile time the typeof(set) is a UnitRange{Int}. Therefore, Julia can't prove that the range starts at 1 (it only finds this out at runtime), and it defaults to a  DenseAxisArray. The case where we explicitly wrote i = 1:3 worked because the macro can \"see\" the 1 at compile time.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"However, if you know that the indices do form an Array, you can force the container type with container = Array:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> set = 1:3\n1:3\n\njulia> Containers.@container([i=set, j=1:5], i + j, container = Array)\n3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Here's another example with something similar:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> a = 1\n1\n\njulia> Containers.@container([i=a:3, j=1:5], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, Base.OneTo(5)\nAnd data, a 3×5 Matrix{Int64}:\n 2  3  4  5  6\n 3  4  5  6  7\n 4  5  6  7  8\n\njulia> Containers.@container([i=1:a, j=1:5], i + j)\n1×5 Matrix{Int64}:\n 2  3  4  5  6","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"Finally, if the compiler cannot prove that the index set is rectangular, a Containers.SparseAxisArray will be created.","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"This occurs when some indices depend on a previous one:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i=1:3, j=1:i], i + j)\nJuMP.Containers.SparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  2\n  [2, 1]  =  3\n  [2, 2]  =  4\n  [3, 1]  =  4\n  [3, 2]  =  5\n  [3, 3]  =  6","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"or if there is a condition on the index sets:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> Containers.@container([i = 1:5; isodd(i)], i^2)\nJuMP.Containers.SparseAxisArray{Int64, 1, Tuple{Int64}} with 3 entries:\n  [1]  =  1\n  [3]  =  9\n  [5]  =  25","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"The condition can depend on multiple indices, the only requirement is that it is an expression that returns true or false:","category":"page"},{"location":"manual/containers/","page":"Containers","title":"Containers","text":"julia> condition(i, j) = isodd(i) && iseven(j)\ncondition (generic function with 1 method)\n\njulia> Containers.@container([i = 1:2, j = 1:4; condition(i, j)], i + j)\nJuMP.Containers.SparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 2 entries:\n  [1, 2]  =  3\n  [1, 4]  =  5","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/solutions.md\"","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/solutions/#manual_solutions","page":"Solutions","title":"Solutions","text":"","category":"section"},{"location":"moi/manual/solutions/#Solving-and-retrieving-the-results","page":"Solutions","title":"Solving and retrieving the results","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Once an optimizer is loaded with the objective function and all of the constraints, we can ask the solver to solve the model by calling optimize!.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"MOI.optimize!(optimizer)","category":"page"},{"location":"moi/manual/solutions/#Why-did-the-solver-stop?","page":"Solutions","title":"Why did the solver stop?","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The optimization procedure may stop for a number of reasons. The TerminationStatus attribute of the optimizer returns a TerminationStatusCode object which explains why the solver stopped.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The termination statuses distinguish between proofs of optimality, infeasibility, local convergence, limits, and termination because of something unexpected like invalid problem data or failure to converge.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"A typical usage of the TerminationStatus attribute is as follows:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"status = MOI.get(optimizer, TerminationStatus())\nif status == MOI.OPTIMAL\n    # Ok, we solved the problem!\nelse\n    # Handle other cases.\nend","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"After checking the TerminationStatus, check ResultCount. This attribute returns the number of results that the solver has available to return. A result is defined as a primal-dual pair, but either the primal or the dual may be missing from the result. While the OPTIMAL termination status normally implies that at least one result is available, other statuses do not. For example, in the case of infeasibility, a solver may return no result or a proof of infeasibility. The ResultCount attribute distinguishes between these two cases.","category":"page"},{"location":"moi/manual/solutions/#Primal-solutions","page":"Solutions","title":"Primal solutions","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Use the PrimalStatus optimizer attribute to return a ResultStatusCode describing the status of the primal solution.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Common returns are described below in the Common status situations section.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the primal solution using the VariablePrimal and ConstraintPrimal attributes.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the objective function value using the ObjectiveValue attribute.","category":"page"},{"location":"moi/manual/solutions/#Dual-solutions","page":"Solutions","title":"Dual solutions","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nSee Duality for a discussion of the MOI conventions for primal-dual pairs and certificates.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Use the DualStatus optimizer attribute to return a ResultStatusCode describing the status of the dual solution.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the dual solution using the ConstraintDual attribute.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the dual objective function value using the DualObjectiveValue attribute.","category":"page"},{"location":"moi/manual/solutions/#Common-status-situations","page":"Solutions","title":"Common status situations","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"The sections below describe how to interpret typical or interesting status cases for three common classes of solvers. The example cases are illustrative, not comprehensive. Solver wrappers may provide additional information on how the solver's statuses map to MOI statuses.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\n* in the tables indicate that multiple different values are possible.","category":"page"},{"location":"moi/manual/solutions/#Primal-dual-convex-solver","page":"Solutions","title":"Primal-dual convex solver","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Linear programming and conic optimization solvers fall into this category.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nProved optimality OPTIMAL 1 FEASIBLE_POINT FEASIBLE_POINT\nProved infeasible INFEASIBLE 1 NO_SOLUTION INFEASIBILITY_CERTIFICATE\nOptimal within relaxed tolerances ALMOST_OPTIMAL 1 FEASIBLE_POINT FEASIBLE_POINT\nOptimal within relaxed tolerances ALMOST_OPTIMAL 1 ALMOST_FEASIBLE_POINT ALMOST_FEASIBLE_POINT\nDetected an unbounded ray of the primal DUAL_INFEASIBLE 1 INFEASIBILITY_CERTIFICATE NO_SOLUTION\nStall SLOW_PROGRESS 1 * *","category":"page"},{"location":"moi/manual/solutions/#Global-branch-and-bound-solvers","page":"Solutions","title":"Global branch-and-bound solvers","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Mixed-integer programming solvers fall into this category.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nProved optimality OPTIMAL 1 FEASIBLE_POINT NO_SOLUTION\nPresolve detected infeasibility or unboundedness INFEASIBLE_OR_UNBOUNDED 0 NO_SOLUTION NO_SOLUTION\nProved infeasibility INFEASIBLE 0 NO_SOLUTION NO_SOLUTION\nTimed out (no solution) TIME_LIMIT 0 NO_SOLUTION NO_SOLUTION\nTimed out (with a solution) TIME_LIMIT 1 FEASIBLE_POINT NO_SOLUTION\nCPXMIP_OPTIMAL_INFEAS ALMOST_OPTIMAL 1 INFEASIBLE_POINT NO_SOLUTION","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nCPXMIP_OPTIMAL_INFEAS is a CPLEX status that indicates that a preprocessed problem was solved to optimality, but the solver was unable to recover a feasible solution to the original problem. Handling this status was one of the motivating drivers behind the design of MOI.","category":"page"},{"location":"moi/manual/solutions/#Local-search-solvers","page":"Solutions","title":"Local search solvers","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Nonlinear programming solvers fall into this category. It also includes non-global tree search solvers like Juniper.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"What happened? TerminationStatus ResultCount PrimalStatus DualStatus\nConverged to a stationary point LOCALLY_SOLVED 1 FEASIBLE_POINT FEASIBLE_POINT\nCompleted a non-global tree search (with a solution) LOCALLY_SOLVED 1 FEASIBLE_POINT FEASIBLE_POINT\nConverged to an infeasible point LOCALLY_INFEASIBLE 1 INFEASIBLE_POINT *\nCompleted a non-global tree search (no solution found) LOCALLY_INFEASIBLE 0 NO_SOLUTION NO_SOLUTION\nIteration limit ITERATION_LIMIT 1 * *\nDiverging iterates NORM_LIMIT or OBJECTIVE_LIMIT 1 * *","category":"page"},{"location":"moi/manual/solutions/#Querying-solution-attributes","page":"Solutions","title":"Querying solution attributes","text":"","category":"section"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"Some solvers will not implement every solution attribute. Therefore, a call like MOI.get(model, MOI.SolveTimeSec()) may throw an UnsupportedAttribute error.","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"If you need to write code that is agnostic to the solver (for example, you are writing a library that an end-user passes their choice of solver to), you can work-around this problem using a try-catch:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"function get_solve_time(model)\n    try\n        return MOI.get(model, MOI.SolveTimeSec())\n    catch err\n        if err isa MOI.UnsupportedAttribute\n            return NaN  # Solver doesn't support. Return a placeholder value.\n        end\n        rethrow(err)  # Something else went wrong. Rethrow the error\n    end\nend","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"If, after careful profiling, you find that the try-catch is taking a significant portion of your runtime, you can improve performance by caching the result of the try-catch:","category":"page"},{"location":"moi/manual/solutions/","page":"Solutions","title":"Solutions","text":"mutable struct CachedSolveTime{M}\n    model::M\n    supports_solve_time::Bool\n    CachedSolveTime(model::M) where {M} = new(model, true)\nend\n\nfunction get_solve_time(model::CachedSolveTime)\n    if !model.supports_solve_time\n        return NaN\n    end\n    try\n        return MOI.get(model, MOI.SolveTimeSec())\n    catch err\n        if err isa MOI.UnsupportedAttribute\n            model.supports_solve_time = false\n            return NaN\n        end\n        rethrow(err)  # Something else went wrong. Rethrow the error\n    end\nend","category":"page"},{"location":"tutorials/getting_started/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"The purpose of these \"Getting started\" tutorials is to teach new users the basics of Julia and JuMP.","category":"page"},{"location":"tutorials/getting_started/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/getting_started/introduction/","page":"Introduction","title":"Introduction","text":"The \"Getting started with\" tutorials are basic introductions to different aspects of JuMP and Julia. If you are new to JuMP and Julia, start by reading them in the following order:\nGetting started with Julia\nGetting started with JuMP\nGetting started with sets and indexing\nGetting started with data and plotting\nJulia has a reputation for being \"fast.\" Unfortunately, it is also easy to write slow Julia code. Performance tips contains a number of important tips on how to improve the performance of models you write in JuMP.\nDesign patterns for larger models is a more advanced tutorial that is aimed at users writing large JuMP models. It's in the \"Getting started\" section to give you an early preview of how JuMP makes it easy to structure larger models. If you are new to JuMP you may want to skip or briefly skim this tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"EditURL = \"two_stage_stochastic.jl\"","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Two-stage-stochastic-programs","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The purpose of this tutorial is to demonstrate how to model and solve a two-stage stochastic program.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"using JuMP\nimport Distributions\nimport HiGHS\nimport Plots\nimport StatsPlots\nimport Statistics","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Background","page":"Two-stage stochastic programs","title":"Background","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"During the week, you are a busy practitioner of Operations Research. To escape the drudgery of mathematics, you decide to open a side business selling creamy mushroom pies with puff pastry. After a few weeks, it quickly becomes apparent that operating a food business is not so easy.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The pies must be prepared in the morning, before you open for the day and can gauge the level of demand. If you bake too many, the unsold pies at the end of the day must be discarded and you have wasted time and money on their production. But if you bake too few, then there may be un-served customers and you could have made more money by baking more pies.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"After a few weeks of poor decision making, you decide to put your knowledge of Operations Research to good use, starting with some data collection.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Each pie costs you $2 to make, and you sell them at $5 each. Disposal of an unsold pie costs $0.10. Based on three weeks of data collected, in which you made 200 pies each week, you sold 150, 190, and 200 pies. Thus, as a guess, you assume a triangular distribution of demand with a minimum of 150, a median of 200, and a maximum of 250.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"We can model this problem by a two-stage stochastic program. In the first stage, we decide a quantity of pies to make x. We make this decision before we observe the demand d_omega. In the second stage, we sell y_omega pies, and incur any costs for unsold pies.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"We can formulate this problem as follows:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"beginaligned\nmaxlimits_xy_omega   -2x + mathbbE_omega5y_omega - 01(x - y_omega) \n   y_omega le x               quad forall omega in Omega \n   0 le y_omega le d_omega  quad forall omega in Omega \n   x ge 0\nendaligned","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Sample-Average-approximation","page":"Two-stage stochastic programs","title":"Sample Average approximation","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"If the distribution of demand is continuous, then our problem has an infinite number of variables and constraints. To form a computationally tractable problem, we instead use a finite set of samples drawn from the distribution. This is called sample average approximation (SAA).","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"D = Distributions.TriangularDist(150.0, 250.0, 200.0)\nN = 100\nd = sort!(rand(D, N));\nΩ = 1:N\nP = fill(1 / N, N);\nStatsPlots.histogram(d; bins = 20, label = \"\", xlabel = \"Demand\")","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#JuMP-model","page":"Two-stage stochastic programs","title":"JuMP model","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The implementation of our two-stage stochastic program in JuMP is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y[ω in Ω] <= d[ω])\n@constraint(model, [ω in Ω], y[ω] <= x)\n@expression(model, z[ω in Ω], 5y[ω] - 0.1 * (x - y[ω]))\n@objective(model, Max, -2x + sum(P[ω] * z[ω] for ω in Ω))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The optimal number of pies to make is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"value(x)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The distribution of total profit is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"total_profit = [-2 * value(x) + value(z[ω]) for ω in Ω]","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Let's plot it:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"\"\"\"\n    bin_distribution(x::Vector{Float64}, N::Int)\n\nA helper function that discretizes `x` into bins of width `N`.\n\"\"\"\nbin_distribution(x, N) = N * (floor(minimum(x) / N):ceil(maximum(x) / N))\n\nplot = StatsPlots.histogram(\n    total_profit;\n    bins = bin_distribution(total_profit, 25),\n    label = \"\",\n    xlabel = \"Profit [\\$]\",\n    ylabel = \"Number of outcomes\",\n)\nμ = Statistics.mean(total_profit)\nPlots.vline!(\n    plot,\n    [μ];\n    label = \"Expected profit (\\$$(round(Int, μ)))\",\n    linewidth = 3,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Risk-measures","page":"Two-stage stochastic programs","title":"Risk measures","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"A risk measure is a function which maps a random variable to a real number. Common risk measures include the mean (expectation), median, mode, and maximum. We need a risk measure to convert the distribution of second stage costs into a single number that can be optimized.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Our model currently uses the expectation risk measure, but others are possible too. One popular risk measure is the conditional value at risk (CVaR).","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"CVaR has a parameter gamma, and it computes the expectation of the worst gamma fraction of outcomes.","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"If we are maximizing, so that small outcomes are bad, the definition of CVaR is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"CVaR_gammaZ = maxlimits_xi  xi - frac1gammamathbbE_omegaleft(xi - Z)_+right","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"which can be formulated as the linear program:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"beginaligned\nCVaR_gammaZ = maxlimits_xi z_omega   xi - frac1gammasum P_omega z_omega\n  z_omega ge xi - Z_omega  quad forall omega \n  z_omega ge 0  quad forall omega\nendaligned","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"function CVaR(Z::Vector{Float64}, P::Vector{Float64}; γ::Float64)\n    @assert 0 < γ <= 1\n    N = length(Z)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, ξ)\n    @variable(model, z[1:N] >= 0)\n    @constraint(model, [i in 1:N], z[i] >= ξ - Z[i])\n    @objective(model, Max, ξ - 1 / γ * sum(P[i] * z[i] for i in 1:N))\n    optimize!(model)\n    return objective_value(model)\nend","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"When γ is 1.0, we compute the mean of the profit:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_10 = CVaR(total_profit, P; γ = 1.0)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Statistics.mean(total_profit)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"As γ approaches 0.0, we compute the worst-case (minimum) profit:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_00 = CVaR(total_profit, P; γ = 0.0001)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"minimum(total_profit)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"By varying γ between 0 and 1 we can compute some trade-off of these two extremes:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"cvar_05 = CVaR(total_profit, P; γ = 0.5)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Let's plot these outcomes on our distribution:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"plot = StatsPlots.histogram(\n    total_profit;\n    bins = bin_distribution(total_profit, 25),\n    label = \"\",\n    xlabel = \"Profit [\\$]\",\n    ylabel = \"Number of outcomes\",\n)\nPlots.vline!(\n    plot,\n    [cvar_10 cvar_05 cvar_00];\n    label = [\"γ = 1.0\" \"γ = 0.5\" \"γ = 0.0\"],\n    linewidth = 3,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Risk-averse-sample-average-approximation","page":"Two-stage stochastic programs","title":"Risk averse sample average approximation","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Because CVaR can be formulated as a linear program, we can form a risk averse sample average approximation model by combining the two formulations:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"γ = 0.4\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y[ω in Ω] <= d[ω])\n@constraint(model, [ω in Ω], y[ω] <= x)\n@expression(model, Z[ω in Ω], 5 * y[ω] - 0.1(x - y[ω]))\n@variable(model, ξ)\n@variable(model, z[ω in Ω] >= 0)\n@constraint(model, [ω in Ω], z[ω] >= ξ - Z[ω])\n@objective(model, Max, -2x + ξ - 1 / γ * sum(P[ω] * z[ω] for ω in Ω))\noptimize!(model)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"When gamma = 04, the optimal number of pies to bake is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"value(x)","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"The distribution of total profit is:","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"risk_averse_total_profit = [value(-2x + Z[ω]) for ω in Ω]\nbins = bin_distribution([total_profit; risk_averse_total_profit], 25)\nplot = StatsPlots.histogram(total_profit; label = \"Expectation\", bins = bins)\nStatsPlots.histogram!(\n    plot,\n    risk_averse_total_profit;\n    label = \"CV@R\",\n    bins = bins,\n    alpha = 0.5,\n)\nplot","category":"page"},{"location":"tutorials/applications/two_stage_stochastic/#Next-steps","page":"Two-stage stochastic programs","title":"Next steps","text":"","category":"section"},{"location":"tutorials/applications/two_stage_stochastic/","page":"Two-stage stochastic programs","title":"Two-stage stochastic programs","text":"Try solving this problem for different numbers of samples and different distributions.\nRefactor the example to avoid hard-coding the costs. What happens to the solution if the cost of disposing unsold pies increases?\nPlot the optimal number of pies to make for different values of the risk aversion parameter gamma. What is the relationship?","category":"page"},{"location":"packages/solvers/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"This section of the documentation contains brief documentation for some of the solvers that JuMP supports. The list of solvers is not exhaustive, but instead is intended to help you discover commonly used solvers.","category":"page"},{"location":"packages/solvers/#Affiliation","page":"Introduction","title":"Affiliation","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Packages beginning with jump-dev/ are developed and maintained by the JuMP developers. In many cases, these packages wrap external solvers that are not developed by the JuMP developers and, while the Julia packages are all open-source, in some cases the solvers themselves are closed source commercial products.","category":"page"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Packages that do not begin with jump-dev/ are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.","category":"page"},{"location":"packages/solvers/#Adding-new-solvers","page":"Introduction","title":"Adding new solvers","text":"","category":"section"},{"location":"packages/solvers/","page":"Introduction","title":"Introduction","text":"Written a solver? Add it to this section of the JuMP documentation by making a pull request to the docs/packages.toml file.","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/models.md\"","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/models/#Models","page":"Models","title":"Models","text":"","category":"section"},{"location":"moi/reference/models/#Attribute-interface","page":"Models","title":"Attribute interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"is_set_by_optimize\nis_copyable\nget\nget!\nset\nsupports\nattribute_value_type","category":"page"},{"location":"moi/reference/models/#MathOptInterface.is_set_by_optimize","page":"Models","title":"MathOptInterface.is_set_by_optimize","text":"is_set_by_optimize(::AnyAttribute)\n\nReturn a Bool indicating whether the value of the attribute is modified during an optimize! call, that is, the attribute is used to query the result of the optimization.\n\nImportant note when defining new attributes\n\nThis function returns false by default so it should be implemented for attributes that are modified by optimize!.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.is_copyable","page":"Models","title":"MathOptInterface.is_copyable","text":"is_copyable(::AnyAttribute)\n\nReturn a Bool indicating whether the value of the attribute may be copied during copy_to using set.\n\nImportant note when defining new attributes\n\nBy default is_copyable(attr) returns !is_set_by_optimize(attr). A specific method should be defined for attributes which are copied indirectly during copy_to. For instance, both is_copyable and is_set_by_optimize return false for the following attributes:\n\nListOfOptimizerAttributesSet, ListOfModelAttributesSet, ListOfConstraintAttributesSet and ListOfVariableAttributesSet.\nSolverName and RawSolver: these attributes cannot be set.\nNumberOfVariables and ListOfVariableIndices: these attributes are set indirectly by add_variable and add_variables.\nObjectiveFunctionType: this attribute is set indirectly when setting the ObjectiveFunction attribute.\nNumberOfConstraints, ListOfConstraintIndices, ListOfConstraintTypesPresent, CanonicalConstraintFunction, ConstraintFunction and ConstraintSet: these attributes are set indirectly by add_constraint and add_constraints.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.get","page":"Models","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64\n\nReturn the number of variables created by the bridge b in the model.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)\n\nReturn the list of variables created by the bridge b.\n\nSee also MOI.ListOfVariableIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}\n\nReturn the number of constraints of the type F-in-S created by the bridge b.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\nMOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}\n\nReturn a  Vector{ConstraintIndex{F,S}} with indices of all constraints of type F-in-S created by the bride b.\n\nSee also MOI.ListOfConstraintIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\nfunction MOI.get(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    bridge::AbstractBridge,\n)\n\nReturn the value of the attribute attr of the model model for the constraint bridged by bridge.\n\n\n\n\n\nget(model::GenericModel, attr::MathOptInterface.AbstractOptimizerAttribute)\n\nReturn the value of the attribute attr from the model's MOI backend.\n\n\n\n\n\nget(model::GenericModel, attr::MathOptInterface.AbstractModelAttribute)\n\nReturn the value of the attribute attr from the model's MOI backend.\n\n\n\n\n\nget(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute)\n\nReturn an attribute attr of the optimizer optimizer.\n\nget(model::ModelLike, attr::AbstractModelAttribute)\n\nReturn an attribute attr of the model model.\n\nget(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex)\n\nIf the attribute attr is set for the variable v in the model model, return its value, return nothing otherwise. If the attribute attr is not supported by model then an error should be thrown instead of returning nothing.\n\nget(model::ModelLike, attr::AbstractVariableAttribute, v::Vector{VariableIndex})\n\nReturn a vector of attributes corresponding to each variable in the collection v in the model model.\n\nget(model::ModelLike, attr::AbstractConstraintAttribute, c::ConstraintIndex)\n\nIf the attribute attr is set for the constraint c in the model model, return its value, return nothing otherwise. If the attribute attr is not supported by model then an error should be thrown instead of returning nothing.\n\nget(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::Vector{ConstraintIndex{F,S}},\n) where {F,S}\n\nReturn a vector of attributes corresponding to each constraint in the collection c in the model model.\n\nget(model::ModelLike, ::Type{VariableIndex}, name::String)\n\nIf a variable with name name exists in the model model, return the corresponding index, otherwise return nothing. Errors if two variables have the same name.\n\nget(\n    model::ModelLike,\n    ::Type{ConstraintIndex{F,S}},\n    name::String,\n) where {F,S}\n\nIf an F-in-S constraint with name name exists in the model model, return the corresponding index, otherwise return nothing. Errors if two constraints have the same name.\n\nget(model::ModelLike, ::Type{ConstraintIndex}, name::String)\n\nIf any constraint with name name exists in the model model, return the corresponding index, otherwise return nothing. This version is available for convenience but may incur a performance penalty because it is not type stable. Errors if two constraints have the same name.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.get!","page":"Models","title":"MathOptInterface.get!","text":"get!(output, model::ModelLike, args...)\n\nAn in-place version of get.\n\nThe signature matches that of get except that the the result is placed in the vector output.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.set","page":"Models","title":"MathOptInterface.set","text":"function MOI.set(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    bridge::AbstractBridge,\n    value,\n)\n\nSet the value of the attribute attr of the model model for the constraint bridged by bridge.\n\n\n\n\n\nset(optimizer::AbstractOptimizer, attr::AbstractOptimizerAttribute, value)\n\nAssign value to the attribute attr of the optimizer optimizer.\n\nset(model::ModelLike, attr::AbstractModelAttribute, value)\n\nAssign value to the attribute attr of the model model.\n\nset(model::ModelLike, attr::AbstractVariableAttribute, v::VariableIndex, value)\n\nAssign value to the attribute attr of variable v in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractVariableAttribute,\n    v::Vector{VariableIndex},\n    vector_of_values,\n)\n\nAssign a value respectively to the attribute attr of each variable in the collection v in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::ConstraintIndex,\n    value,\n)\n\nAssign a value to the attribute attr of constraint c in model model.\n\nset(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    c::Vector{ConstraintIndex{F,S}},\n    vector_of_values,\n) where {F,S}\n\nAssign a value respectively to the attribute attr of each constraint in the collection c in model model.\n\nAn UnsupportedAttribute error is thrown if model does not support the attribute attr (see supports) and a SetAttributeNotAllowed error is thrown if it supports the attribute attr but it cannot be set.\n\nset(\n    model::ModelLike,\n    ::ConstraintSet,\n    c::ConstraintIndex{F,S},\n    set::S,\n) where {F,S}\n\nChange the set of constraint c to the new set set which should be of the same type as the original set.\n\nset(\n    model::ModelLike,\n    ::ConstraintFunction,\n    c::ConstraintIndex{F,S},\n    func::F,\n) where {F,S}\n\nReplace the function in constraint c with func. F must match the original function type used to define the constraint.\n\nnote: Note\nSetting the constraint function is not allowed if F is VariableIndex; a SettingVariableIndexNotAllowed error is thrown instead. This is because, it would require changing the index c since the index of VariableIndex constraints must be the same as the index of the variable.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.supports","page":"Models","title":"MathOptInterface.supports","text":"MOI.supports(\n    model::MOI.ModelLike,\n    attr::MOI.AbstractConstraintAttribute,\n    BT::Type{<:AbstractBridge},\n)\n\nReturn a Bool indicating whether BT supports setting attr to model.\n\n\n\n\n\nsupports(model::ModelLike, sub::AbstractSubmittable)::Bool\n\nReturn a Bool indicating whether model supports the submittable sub.\n\nsupports(model::ModelLike, attr::AbstractOptimizerAttribute)::Bool\n\nReturn a Bool indicating whether model supports the optimizer attribute attr. That is, it returns false if copy_to(model, src) shows a warning in case attr is in the ListOfOptimizerAttributesSet of src; see copy_to for more details on how unsupported optimizer attributes are handled in copy.\n\nsupports(model::ModelLike, attr::AbstractModelAttribute)::Bool\n\nReturn a Bool indicating whether model supports the model attribute attr. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfModelAttributesSet of src.\n\nsupports(\n    model::ModelLike,\n    attr::AbstractVariableAttribute,\n    ::Type{VariableIndex},\n)::Bool\n\nReturn a Bool indicating whether model supports the variable attribute attr. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfVariableAttributesSet of src.\n\nsupports(\n    model::ModelLike,\n    attr::AbstractConstraintAttribute,\n    ::Type{ConstraintIndex{F,S}},\n)::Bool where {F,S}\n\nReturn a Bool indicating whether model supports the constraint attribute attr applied to an F-in-S constraint. That is, it returns false if copy_to(model, src) cannot be performed in case attr is in the ListOfConstraintAttributesSet of src.\n\nFor all five methods, if the attribute is only not supported in specific circumstances, it should still return true.\n\nNote that supports is only defined for attributes for which is_copyable returns true as other attributes do not appear in the list of attributes set obtained by ListOf...AttributesSet.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.attribute_value_type","page":"Models","title":"MathOptInterface.attribute_value_type","text":"attribute_value_type(attr::AnyAttribute)\n\nGiven an attribute attr, return the type of value expected by get, or returned by set.\n\nNotes\n\nOnly implement this if it make sense to do so. If un-implemented, the default is Any.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#Model-interface","page":"Models","title":"Model interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"ModelLike\nis_empty\nempty!\nwrite_to_file\nread_from_file\nsupports_incremental_interface\ncopy_to\nIndexMap","category":"page"},{"location":"moi/reference/models/#MathOptInterface.ModelLike","page":"Models","title":"MathOptInterface.ModelLike","text":"ModelLike\n\nAbstract supertype for objects that implement the \"Model\" interface for defining an optimization problem.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.is_empty","page":"Models","title":"MathOptInterface.is_empty","text":"is_empty(model::ModelLike)\n\nReturns false if the model has any model attribute set or has any variables or constraints.\n\nNote that an empty model can have optimizer attributes set.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.empty!","page":"Models","title":"MathOptInterface.empty!","text":"empty!(model::ModelLike)\n\nEmpty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.write_to_file","page":"Models","title":"MathOptInterface.write_to_file","text":"write_to_file(model::ModelLike, filename::String)\n\nWrite the current model to the file at filename.\n\nSupported file types depend on the model type.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.read_from_file","page":"Models","title":"MathOptInterface.read_from_file","text":"read_from_file(model::ModelLike, filename::String)\n\nRead the file filename into the model model. If model is non-empty, this may throw an error.\n\nSupported file types depend on the model type.\n\nNote\n\nOnce the contents of the file are loaded into the model, users can query the variables via get(model, ListOfVariableIndices()). However, some filetypes, such as LP files, do not maintain an explicit ordering of the variables. Therefore, the returned list may be in an arbitrary order.\n\nTo avoid depending on the order of the indices, look up each variable index by name using  get(model, VariableIndex, \"name\").\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.supports_incremental_interface","page":"Models","title":"MathOptInterface.supports_incremental_interface","text":"supports_incremental_interface(model::ModelLike)\n\nReturn a Bool indicating whether model supports building incrementally via add_variable and add_constraint.\n\nThe main purpose of this function is to determine whether a model can be loaded into model incrementally or whether it should be cached and copied at once instead.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.copy_to","page":"Models","title":"MathOptInterface.copy_to","text":"copy_to(dest::ModelLike, src::ModelLike)::IndexMap\n\nCopy the model from src into dest.\n\nThe target dest is emptied, and all previous indices to variables and constraints in dest are invalidated.\n\nReturns an IndexMap object that translates variable and constraint indices from the src model to the corresponding indices in the dest model.\n\nNotes\n\nIf a constraint that in src is not supported by dest, then an UnsupportedConstraint error is thrown.\nIf an AbstractModelAttribute, AbstractVariableAttribute, or AbstractConstraintAttribute is set in src but not supported by dest, then an UnsupportedAttribute error is thrown.\n\nAbstractOptimizerAttributes are not copied  to the dest model.\n\nIndexMap\n\nImplementations of copy_to must return an IndexMap. For technical reasons, this type is defined in the Utilities submodule as MOI.Utilities.IndexMap. However, since it is an integral part of the MOI API, we provide MOI.IndexMap as an alias.\n\nExample\n\n# Given empty `ModelLike` objects `src` and `dest`.\n\nx = add_variable(src)\n\nis_valid(src, x)   # true\nis_valid(dest, x)  # false (`dest` has no variables)\n\nindex_map = copy_to(dest, src)\nis_valid(dest, x) # false (unless index_map[x] == x)\nis_valid(dest, index_map[x]) # true\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.IndexMap","page":"Models","title":"MathOptInterface.IndexMap","text":"IndexMap()\n\nThe dictionary-like object returned by copy_to.\n\nIndexMap\n\nImplementations of copy_to must return an IndexMap. For technical reasons, the IndexMap type is defined in the Utilities submodule as MOI.Utilities.IndexMap. However, since it is an integral part of the MOI API, we provide this MOI.IndexMap as an alias.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#Model-attributes","page":"Models","title":"Model attributes","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractModelAttribute\nName\nObjectiveFunction\nObjectiveFunctionType\nObjectiveSense\nOptimizationSense\nMIN_SENSE\nMAX_SENSE\nFEASIBILITY_SENSE\nNumberOfVariables\nListOfVariableIndices\nListOfConstraintTypesPresent\nNumberOfConstraints\nListOfConstraintIndices\nListOfOptimizerAttributesSet\nListOfModelAttributesSet\nListOfVariableAttributesSet\nListOfConstraintAttributesSet\nUserDefinedFunction\nListOfSupportedNonlinearOperators","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractModelAttribute","page":"Models","title":"MathOptInterface.AbstractModelAttribute","text":"AbstractModelAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.Name","page":"Models","title":"MathOptInterface.Name","text":"Name()\n\nA model attribute for the string identifying the model. It has a default value of \"\" if not set`.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveFunction","page":"Models","title":"MathOptInterface.ObjectiveFunction","text":"ObjectiveFunction{F<:AbstractScalarFunction}()\n\nA model attribute for the objective function which has a type F<:AbstractScalarFunction.\n\nF should be guaranteed to be equivalent but not necessarily identical to the function type provided by the user.\n\nThrows an InexactError if the objective function cannot be converted to F, e.g., the objective function is quadratic and F is ScalarAffineFunction{Float64} or it has non-integer coefficient and F is ScalarAffineFunction{Int}.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveFunctionType","page":"Models","title":"MathOptInterface.ObjectiveFunctionType","text":"ObjectiveFunctionType()\n\nA model attribute for the type F of the objective function set using the ObjectiveFunction{F} attribute.\n\nExamples\n\nIn the following code, attr should be equal to MOI.VariableIndex:\n\nx = MOI.add_variable(model)\nMOI.set(model, MOI.ObjectiveFunction{MOI.VariableIndex}(), x)\nattr = MOI.get(model, MOI.ObjectiveFunctionType())\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveSense","page":"Models","title":"MathOptInterface.ObjectiveSense","text":"ObjectiveSense()\n\nA model attribute for the objective sense of the objective function, which must be an OptimizationSense: MIN_SENSE, MAX_SENSE, or FEASIBILITY_SENSE. The default is FEASIBILITY_SENSE.\n\nInteraction with ObjectiveFunction\n\nSetting the sense to FEASIBILITY_SENSE unsets the ObjectiveFunction attribute. That is, if you first set ObjectiveFunction and then set ObjectiveSense to be FEASIBILITY_SENSE, no objective function will be passed to the solver.\n\nIn addition, some reformulations of ObjectiveFunction via bridges rely on the value of ObjectiveSense. Therefore, you should set ObjectiveSense before setting ObjectiveFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OptimizationSense","page":"Models","title":"MathOptInterface.OptimizationSense","text":"OptimizationSense\n\nAn enum for the value of the ObjectiveSense attribute.\n\nValues\n\nPossible values are:\n\nMIN_SENSE: the goal is to minimize the objective function\nMAX_SENSE: the goal is to maximize the objective function\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.MIN_SENSE","page":"Models","title":"MathOptInterface.MIN_SENSE","text":"MIN_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMIN_SENSE: the goal is to minimize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MAX_SENSE","page":"Models","title":"MathOptInterface.MAX_SENSE","text":"MAX_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMAX_SENSE: the goal is to maximize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.FEASIBILITY_SENSE","page":"Models","title":"MathOptInterface.FEASIBILITY_SENSE","text":"FEASIBILITY_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NumberOfVariables","page":"Models","title":"MathOptInterface.NumberOfVariables","text":"NumberOfVariables()\n\nA model attribute for the number of variables in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfVariableIndices","page":"Models","title":"MathOptInterface.ListOfVariableIndices","text":"ListOfVariableIndices()\n\nA model attribute for the Vector{VariableIndex} of all variable indices present in the model (i.e., of length equal to the value of NumberOfVariables in the order in which they were added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintTypesPresent","page":"Models","title":"MathOptInterface.ListOfConstraintTypesPresent","text":"ListOfConstraintTypesPresent()\n\nA model attribute for the list of tuples of the form (F,S), where F is a function type and S is a set type indicating that the attribute NumberOfConstraints{F,S} has a value greater than zero.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NumberOfConstraints","page":"Models","title":"MathOptInterface.NumberOfConstraints","text":"NumberOfConstraints{F,S}()\n\nA model attribute for the number of constraints of the type F-in-S present in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintIndices","page":"Models","title":"MathOptInterface.ListOfConstraintIndices","text":"ListOfConstraintIndices{F,S}()\n\nA model attribute for the Vector{ConstraintIndex{F,S}} of all constraint indices of type F-in-S in the model (i.e., of length equal to the value of NumberOfConstraints{F,S}) in the order in which they were added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfOptimizerAttributesSet","page":"Models","title":"MathOptInterface.ListOfOptimizerAttributesSet","text":"ListOfOptimizerAttributesSet()\n\nAn optimizer attribute for the Vector{AbstractOptimizerAttribute} of all optimizer attributes that were set.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfModelAttributesSet","page":"Models","title":"MathOptInterface.ListOfModelAttributesSet","text":"ListOfModelAttributesSet()\n\nA model attribute for the Vector{AbstractModelAttribute} of all model attributes attr such that:\n\nis_copyable(attr) returns true, and\nthe attribute was set to the model\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfVariableAttributesSet","page":"Models","title":"MathOptInterface.ListOfVariableAttributesSet","text":"ListOfVariableAttributesSet()\n\nA model attribute for the Vector{AbstractVariableAttribute} of all variable attributes attr such that 1) is_copyable(attr) returns true and 2) the attribute was set to variables.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfConstraintAttributesSet","page":"Models","title":"MathOptInterface.ListOfConstraintAttributesSet","text":"ListOfConstraintAttributesSet{F, S}()\n\nA model attribute for the Vector{AbstractConstraintAttribute} of all constraint attributes attr such that:\n\nis_copyable(attr) returns true and\nthe attribute was set to F-in-S constraints.\n\nNote\n\nThe attributes ConstraintFunction and ConstraintSet should not be included in the list even if then have been set with set.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.UserDefinedFunction","page":"Models","title":"MathOptInterface.UserDefinedFunction","text":"UserDefinedFunction(name::Symbol, arity::Int) <: AbstractModelAttribute\n\nSet this attribute to register a user-defined function by the name of name with arity arguments.\n\nOnce registered, name will appear in ListOfSupportedNonlinearOperators.\n\nYou cannot register multiple UserDefinedFunctions with the same name but different arity.\n\nValue type\n\nThe value to be set is a tuple containing one, two, or three functions to evaluate the function, the first-order derivative, and the second-order derivative respectively. Both derivatives are optional, but if you pass the second-order derivative you must also pass the first-order derivative.\n\nFor univariate functions with arity == 1, the functions in the tuple must have the form:\n\nf(x::T)::T: returns the value of the function at x\n∇f(x::T)::T: returns the first-order derivative of f with respect to x\n∇²f(x::T)::T: returns the second-order derivative of f with respect to x.\n\nFor multivariate functions with arity > 1, the functions in the tuple must have the form:\n\nf(x::T...)::T: returns the value of the function at x\n∇f(g::AbstractVector{T}, x::T...)::Nothing: fills the components of g, with g[i] being the first-order partial derivative of f with respect to x[i]\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing: fills the non-zero components  of H, with H[i, j] being the second-order partial derivative of f with  respect to x[i] and then x[j]. H is initialized to the zero matrix,  so you do not need to set any zero elements.\n\nExamples\n\njulia> import MathOptInterface as MOI\n\njulia> f(x, y) = x^2 + y^2\nf (generic function with 1 method)\n\njulia> function ∇f(g, x, y)\n           g .= 2 * x, 2 * y\n           return\n       end\n∇f (generic function with 1 method)\n\njulia> function ∇²f(H, x...)\n           H[1, 1] = H[2, 2] = 2.0\n           return\n       end\n∇²f (generic function with 1 method)\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:f, 2), (f,))\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:g, 2), (f, ∇f))\n\njulia> MOI.set(model, MOI.UserDefinedFunction(:h, 2), (f, ∇f, ∇²f))\n\njulia> x = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n\njulia> obj_f = MOI.ScalarNonlinearFunction(:f, Any[x[1], x[2]])\nf(MOI.VariableIndex(1), MOI.VariableIndex(2))\n\njulia> MOI.set(model, MOI.ObjectiveFunction{typeof(obj_f)}(), obj_f)\n\njulia> print(model)\nMinimize ScalarNonlinearFunction:\n f(v[1], v[2])\n\nSubject to:\n\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ListOfSupportedNonlinearOperators","page":"Models","title":"MathOptInterface.ListOfSupportedNonlinearOperators","text":"ListOfSupportedNonlinearOperators() <: AbstractModelAttribute\n\nWhen queried with get, return a Vector{Symbol} listing the operators supported by the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#Optimizer-interface","page":"Models","title":"Optimizer interface","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractOptimizer\nOptimizerWithAttributes\noptimize!\noptimize!(::ModelLike, ::ModelLike)\ninstantiate\ndefault_cache","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractOptimizer","page":"Models","title":"MathOptInterface.AbstractOptimizer","text":"AbstractOptimizer <: ModelLike\n\nAbstract supertype for objects representing an instance of an optimization problem tied to a particular solver. This is typically a solver's in-memory representation. In addition to ModelLike, AbstractOptimizer objects let you solve the model and query the solution.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OptimizerWithAttributes","page":"Models","title":"MathOptInterface.OptimizerWithAttributes","text":"struct OptimizerWithAttributes\n    optimizer_constructor\n    params::Vector{Pair{AbstractOptimizerAttribute,<:Any}}\nend\n\nObject grouping an optimizer constructor and a list of optimizer attributes. Instances are created with instantiate.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.optimize!","page":"Models","title":"MathOptInterface.optimize!","text":"optimize!(optimizer::AbstractOptimizer)\n\nOptimize the problem contained in optimizer.\n\nBefore calling optimize!, the problem should first be constructed using the incremental interface (see supports_incremental_interface) or copy_to.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.optimize!-Tuple{MathOptInterface.ModelLike, MathOptInterface.ModelLike}","page":"Models","title":"MathOptInterface.optimize!","text":"optimize!(dest::AbstractOptimizer, src::ModelLike)::Tuple{IndexMap,Bool}\n\nA \"one-shot\" call that copies the problem from src into dest and then uses dest to optimize the problem.\n\nReturns a tuple of an IndexMap and a Bool copied.\n\nThe IndexMap object translates variable and constraint indices from the src model to the corresponding indices in the dest optimizer. See copy_to for details.\nIf copied == true, src was copied to dest and then cached, allowing incremental modification if supported by the solver.\nIf copied == false, a cache of the model was not kept in dest. Therefore, only the solution information (attributes for which is_set_by_optimize is true) is available to query.\n\nnote: Note\nThe main purpose of optimize! method with two arguments is for use in Utilities.CachingOptimizer.\n\nRelationship to the single-argument optimize!\n\nThe default fallback of optimize!(dest::AbstractOptimizer, src::ModelLike) is\n\nfunction optimize!(dest::AbstractOptimizer, src::ModelLike)\n    index_map = copy_to(dest, src)\n    optimize!(dest)\n    return index_map, true\nend\n\nTherefore, subtypes of AbstractOptimizer should either implement this two-argument method, or implement both copy_to(::Optimizer, ::ModelLike) and optimize!(::Optimizer).\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/models/#MathOptInterface.instantiate","page":"Models","title":"MathOptInterface.instantiate","text":"instantiate(\n    optimizer_constructor,\n    with_cache_type::Union{Nothing,Type} = nothing,\n    with_bridge_type::Union{Nothing,Type} = nothing,\n)\n\nCreate an instance of an optimizer by either:\n\ncalling optimizer_constructor.optimizer_constructor() and setting the parameters in optimizer_constructor.params if optimizer_constructor is a OptimizerWithAttributes\ncalling optimizer_constructor() if optimizer_constructor is callable.\n\nwithcachetype\n\nIf with_cache_type is not nothing, then the optimizer is wrapped in a Utilities.CachingOptimizer to store a cache of the model. This is most useful if the optimizer you are constructing does not support the incremental interface (see supports_incremental_interface).\n\nwithbridgetype\n\nIf with_bridge_type is not nothing, the optimizer is wrapped in a Bridges.full_bridge_optimizer, enabling all the bridges defined in the MOI.Bridges submodule with coefficient type with_bridge_type.\n\nIn addition, if the optimizer created by optimizer_constructor does not support the incremental interface (see supports_incremental_interface), then, irrespective of with_cache_type, the optimizer is wrapped in a Utilities.CachingOptimizer to store a cache of the bridged model.\n\nIf with_cache_type and with_bridge_type are both not nothing, then they must be the same type.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.default_cache","page":"Models","title":"MathOptInterface.default_cache","text":"default_cache(optimizer::ModelLike, ::Type{T}) where {T}\n\nReturn a new instance of the default model type to be used as cache for optimizer in a Utilities.CachingOptimizer for holding constraints of coefficient type T. By default, this returns Utilities.UniversalFallback(Utilities.Model{T}()). If copying from a instance of a given model type is faster for optimizer then a new method returning an instance of this model type should be defined.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#Optimizer-attributes","page":"Models","title":"Optimizer attributes","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"AbstractOptimizerAttribute\nSolverName\nSolverVersion\nSilent\nTimeLimitSec\nObjectiveLimit\nRawOptimizerAttribute\nNumberOfThreads\nRawSolver\nAbsoluteGapTolerance\nRelativeGapTolerance","category":"page"},{"location":"moi/reference/models/#MathOptInterface.AbstractOptimizerAttribute","page":"Models","title":"MathOptInterface.AbstractOptimizerAttribute","text":"AbstractOptimizerAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of the optimizer.\n\nNotes\n\nThe difference between AbstractOptimizerAttribute and AbstractModelAttribute lies in the behavior of is_empty, empty! and copy_to. Typically optimizer attributes affect only how the model is solved.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolverName","page":"Models","title":"MathOptInterface.SolverName","text":"SolverName()\n\nAn optimizer attribute for the string identifying the solver/optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolverVersion","page":"Models","title":"MathOptInterface.SolverVersion","text":"SolverVersion()\n\nAn optimizer attribute for the string identifying the version of the solver.\n\nnote: Note\nFor solvers supporting semantic versioning, the SolverVersion should be a string of the form \"vMAJOR.MINOR.PATCH\", so that it can be converted to a Julia VersionNumber (e.g., `VersionNumber(\"v1.2.3\")).We do not require Semantic Versioning because some solvers use alternate versioning systems. For example, CPLEX uses Calendar Versioning, so SolverVersion will return a string like \"202001\".\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.Silent","page":"Models","title":"MathOptInterface.Silent","text":"Silent()\n\nAn optimizer attribute for silencing the output of an optimizer. When set to true, it takes precedence over any other attribute controlling verbosity and requires the solver to produce no output. The default value is false which has no effect. In this case the verbosity is controlled by other attributes.\n\nNote\n\nEvery optimizer should have verbosity on by default. For instance, if a solver has a solver-specific log level attribute, the MOI implementation should set it to 1 by default. If the user sets Silent to true, then the log level should be set to 0, even if the user specifically sets a value of log level. If the value of Silent is false then the log level set to the solver is the value given by the user for this solver-specific parameter or 1 if none is given.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.TimeLimitSec","page":"Models","title":"MathOptInterface.TimeLimitSec","text":"TimeLimitSec()\n\nAn optimizer attribute for setting a time limit (in seconnds) for an optimization. When set to nothing, it deactivates the solver time limit. The default value is nothing.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveLimit","page":"Models","title":"MathOptInterface.ObjectiveLimit","text":"ObjectiveLimit()\n\nAn optimizer attribute for setting a limit on the objective value.\n\nThe provided limit must be a Union{Real,Nothing}.\n\nWhen set to nothing, the limit reverts to the solver's default.\n\nThe default value is nothing.\n\nThe solver may stop when the ObjectiveValue is better (lower for minimization, higher for maximization) than the ObjectiveLimit. If stopped, the TerminationStatus should be OBJECTIVE_LIMIT.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawOptimizerAttribute","page":"Models","title":"MathOptInterface.RawOptimizerAttribute","text":"RawOptimizerAttribute(name::String)\n\nAn optimizer attribute for the solver-specific parameter identified by name.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NumberOfThreads","page":"Models","title":"MathOptInterface.NumberOfThreads","text":"NumberOfThreads()\n\nAn optimizer attribute for setting the number of threads used for an optimization. When set to nothing uses solver default. Values are positive integers. The default value is nothing.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawSolver","page":"Models","title":"MathOptInterface.RawSolver","text":"RawSolver()\n\nA model attribute for the object that may be used to access a solver-specific API for this optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.AbsoluteGapTolerance","page":"Models","title":"MathOptInterface.AbsoluteGapTolerance","text":"AbsoluteGapTolerance()\n\nAn optimizer attribute for setting the absolute gap tolerance for an optimization. This is an optimizer attribute, and should be set before calling optimize!. When set to nothing (if supported), uses solver default.\n\nTo set a relative gap tolerance, see RelativeGapTolerance.\n\nwarning: Warning\nThe mathematical definition of \"absolute gap\", and its treatment during the optimization, are solver-dependent. However, assuming no other limit nor issue is encountered during the optimization, most solvers that implement this attribute will stop once f - b  g_abs, where b is the best bound, f is the best feasible objective value, and g_abs is the absolute gap.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RelativeGapTolerance","page":"Models","title":"MathOptInterface.RelativeGapTolerance","text":"RelativeGapTolerance()\n\nAn optimizer attribute for setting the relative gap tolerance for an optimization. This is an optimizer attribute, and should be set before calling optimize!. When set to nothing (if supported), uses solver default.\n\nIf you are looking for the relative gap of the current best solution, see RelativeGap. If no limit nor issue is encountered during the optimization, the value of RelativeGap should be at most as large as RelativeGapTolerance.\n\n# Before optimizing: set relative gap tolerance\n# set 0.1% relative gap tolerance\nMOI.set(model, MOI.RelativeGapTolerance(), 1e-3)\nMOI.optimize!(model)\n\n# After optimizing (assuming all went well)\n# The relative gap tolerance has not changed...\nMOI.get(model, MOI.RelativeGapTolerance())  # returns 1e-3\n# ... and the relative gap of the obtained solution is smaller or equal to the\n# tolerance\nMOI.get(model, MOI.RelativeGap())  # should return something ≤ 1e-3\n\nwarning: Warning\nThe mathematical definition of \"relative gap\", and its allowed range, are solver-dependent. Typically, solvers expect a value between 0.0 and 1.0.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"List of attributes useful for optimizers","category":"page"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"TerminationStatus\nTerminationStatusCode\nOPTIMIZE_NOT_CALLED\nOPTIMAL\nINFEASIBLE\nDUAL_INFEASIBLE\nLOCALLY_SOLVED\nLOCALLY_INFEASIBLE\nINFEASIBLE_OR_UNBOUNDED\nALMOST_OPTIMAL\nALMOST_INFEASIBLE\nALMOST_DUAL_INFEASIBLE\nALMOST_LOCALLY_SOLVED\nITERATION_LIMIT\nTIME_LIMIT\nNODE_LIMIT\nSOLUTION_LIMIT\nMEMORY_LIMIT\nOBJECTIVE_LIMIT\nNORM_LIMIT\nOTHER_LIMIT\nSLOW_PROGRESS\nNUMERICAL_ERROR\nINVALID_MODEL\nINVALID_OPTION\nINTERRUPTED\nOTHER_ERROR\nPrimalStatus\nDualStatus\nRawStatusString\nResultCount\nObjectiveValue\nDualObjectiveValue\nObjectiveBound\nRelativeGap\nSolveTimeSec\nSimplexIterations\nBarrierIterations\nNodeCount","category":"page"},{"location":"moi/reference/models/#MathOptInterface.TerminationStatus","page":"Models","title":"MathOptInterface.TerminationStatus","text":"TerminationStatus()\n\nA model attribute for the TerminationStatusCode explaining why the optimizer stopped.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.TerminationStatusCode","page":"Models","title":"MathOptInterface.TerminationStatusCode","text":"TerminationStatusCode\n\nAn Enum of possible values for the TerminationStatus attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to optimize!.\n\nValues\n\nPossible values are:\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\nOPTIMAL: The algorithm found a globally optimal solution.\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\nINVALID_MODEL: The algorithm stopped because the model is invalid.\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.OPTIMIZE_NOT_CALLED","page":"Models","title":"MathOptInterface.OPTIMIZE_NOT_CALLED","text":"OPTIMIZE_NOT_CALLED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OPTIMAL","page":"Models","title":"MathOptInterface.OPTIMAL","text":"OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMAL: The algorithm found a globally optimal solution.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE","page":"Models","title":"MathOptInterface.INFEASIBLE","text":"INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.DUAL_INFEASIBLE","page":"Models","title":"MathOptInterface.DUAL_INFEASIBLE","text":"DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.LOCALLY_SOLVED","page":"Models","title":"MathOptInterface.LOCALLY_SOLVED","text":"LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.LOCALLY_INFEASIBLE","page":"Models","title":"MathOptInterface.LOCALLY_INFEASIBLE","text":"LOCALLY_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE_OR_UNBOUNDED","page":"Models","title":"MathOptInterface.INFEASIBLE_OR_UNBOUNDED","text":"INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_OPTIMAL","page":"Models","title":"MathOptInterface.ALMOST_OPTIMAL","text":"ALMOST_OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_INFEASIBLE","page":"Models","title":"MathOptInterface.ALMOST_INFEASIBLE","text":"ALMOST_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_DUAL_INFEASIBLE","page":"Models","title":"MathOptInterface.ALMOST_DUAL_INFEASIBLE","text":"ALMOST_DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ALMOST_LOCALLY_SOLVED","page":"Models","title":"MathOptInterface.ALMOST_LOCALLY_SOLVED","text":"ALMOST_LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.ITERATION_LIMIT","page":"Models","title":"MathOptInterface.ITERATION_LIMIT","text":"ITERATION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.TIME_LIMIT","page":"Models","title":"MathOptInterface.TIME_LIMIT","text":"TIME_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NODE_LIMIT","page":"Models","title":"MathOptInterface.NODE_LIMIT","text":"NODE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.SOLUTION_LIMIT","page":"Models","title":"MathOptInterface.SOLUTION_LIMIT","text":"SOLUTION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MEMORY_LIMIT","page":"Models","title":"MathOptInterface.MEMORY_LIMIT","text":"MEMORY_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OBJECTIVE_LIMIT","page":"Models","title":"MathOptInterface.OBJECTIVE_LIMIT","text":"OBJECTIVE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NORM_LIMIT","page":"Models","title":"MathOptInterface.NORM_LIMIT","text":"NORM_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_LIMIT","page":"Models","title":"MathOptInterface.OTHER_LIMIT","text":"OTHER_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.SLOW_PROGRESS","page":"Models","title":"MathOptInterface.SLOW_PROGRESS","text":"SLOW_PROGRESS::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NUMERICAL_ERROR","page":"Models","title":"MathOptInterface.NUMERICAL_ERROR","text":"NUMERICAL_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INVALID_MODEL","page":"Models","title":"MathOptInterface.INVALID_MODEL","text":"INVALID_MODEL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_MODEL: The algorithm stopped because the model is invalid.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INVALID_OPTION","page":"Models","title":"MathOptInterface.INVALID_OPTION","text":"INVALID_OPTION::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INTERRUPTED","page":"Models","title":"MathOptInterface.INTERRUPTED","text":"INTERRUPTED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_ERROR","page":"Models","title":"MathOptInterface.OTHER_ERROR","text":"OTHER_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.PrimalStatus","page":"Models","title":"MathOptInterface.PrimalStatus","text":"PrimalStatus(result_index::Int = 1)\n\nA model attribute for the ResultStatusCode of the primal result result_index. If result_index is omitted, it defaults to 1.\n\nSee ResultCount for information on how the results are ordered.\n\nIf result_index is larger than the value of ResultCount then NO_SOLUTION is returned.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.DualStatus","page":"Models","title":"MathOptInterface.DualStatus","text":"DualStatus(result_index::Int = 1)\n\nA model attribute for the ResultStatusCode of the dual result result_index. If result_index is omitted, it defaults to 1.\n\nSee ResultCount for information on how the results are ordered.\n\nIf result_index is larger than the value of ResultCount then NO_SOLUTION is returned.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RawStatusString","page":"Models","title":"MathOptInterface.RawStatusString","text":"RawStatusString()\n\nA model attribute for a solver specific string explaining why the optimizer stopped.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ResultCount","page":"Models","title":"MathOptInterface.ResultCount","text":"ResultCount()\n\nA model attribute for the number of results available.\n\nOrder of solutions\n\nA number of attributes contain an index, result_index, which is used to refer to one of the available results. Thus, result_index must be an integer between 1 and the number of available results.\n\nAs a general rule, the first result (result_index=1) is the most important result (e.g., an optimal solution or an infeasibility certificate). Other results will typically be alternate solutions that the solver found during the search for the first result.\n\nIf a (local) optimal solution is available, i.e., TerminationStatus is OPTIMAL or LOCALLY_SOLVED, the first result must correspond to the (locally) optimal solution. Other results may be alternative optimal solutions, or they may be other suboptimal solutions; use ObjectiveValue to distingiush between them.\n\nIf a primal or dual infeasibility certificate is available, i.e., TerminationStatus is INFEASIBLE or DUAL_INFEASIBLE and the corresponding PrimalStatus or DualStatus is INFEASIBILITY_CERTIFICATE, then the first result must be a certificate. Other results may be alternate certificates, or infeasible points.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveValue","page":"Models","title":"MathOptInterface.ObjectiveValue","text":"ObjectiveValue(result_index::Int = 1)\n\nA model attribute for the objective value of the primal solution result_index.\n\nIf the solver does not have a primal value for the objective because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ObjectiveValue attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.DualObjectiveValue","page":"Models","title":"MathOptInterface.DualObjectiveValue","text":"DualObjectiveValue(result_index::Int = 1)\n\nA model attribute for the value of the objective function of the dual problem for the result_indexth dual result.\n\nIf the solver does not have a dual value for the objective because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check DualStatus before accessing the DualObjectiveValue attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ObjectiveBound","page":"Models","title":"MathOptInterface.ObjectiveBound","text":"ObjectiveBound()\n\nA model attribute for the best known bound on the optimal objective value.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.RelativeGap","page":"Models","title":"MathOptInterface.RelativeGap","text":"RelativeGap()\n\nA model attribute for the final relative optimality gap.\n\nwarning: Warning\nThe definition of this gap is solver-dependent. However, most solvers implementing this attribute define the relative gap as some variation of fracb-ff, where b is the best bound and f is the best feasible objective value.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SolveTimeSec","page":"Models","title":"MathOptInterface.SolveTimeSec","text":"SolveTimeSec()\n\nA model attribute for the total elapsed solution time (in seconds) as reported by the optimizer.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.SimplexIterations","page":"Models","title":"MathOptInterface.SimplexIterations","text":"SimplexIterations()\n\nA model attribute for the cumulative number of simplex iterations during the optimization process.\n\nFor a mixed-integer program (MIP), the return value is the total simplex iterations for all nodes.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.BarrierIterations","page":"Models","title":"MathOptInterface.BarrierIterations","text":"BarrierIterations()\n\nA model attribute for the cumulative number of barrier iterations while solving a problem.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NodeCount","page":"Models","title":"MathOptInterface.NodeCount","text":"NodeCount()\n\nA model attribute for the total number of branch-and-bound nodes explored while solving a mixed-integer program (MIP).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#ResultStatusCode","page":"Models","title":"ResultStatusCode","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"ResultStatusCode\nNO_SOLUTION\nFEASIBLE_POINT\nNEARLY_FEASIBLE_POINT\nINFEASIBLE_POINT\nINFEASIBILITY_CERTIFICATE\nNEARLY_INFEASIBILITY_CERTIFICATE\nREDUCTION_CERTIFICATE\nNEARLY_REDUCTION_CERTIFICATE\nUNKNOWN_RESULT_STATUS\nOTHER_RESULT_STATUS","category":"page"},{"location":"moi/reference/models/#MathOptInterface.ResultStatusCode","page":"Models","title":"MathOptInterface.ResultStatusCode","text":"ResultStatusCode\n\nAn Enum of possible values for the PrimalStatus and DualStatus attributes.\n\nThe values indicate how to interpret the result vector.\n\nValues\n\nPossible values are:\n\nNO_SOLUTION: the result vector is empty.\nFEASIBLE_POINT: the result vector is a feasible point.\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\nINFEASIBLE_POINT: the result vector is an infeasible point.\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NO_SOLUTION","page":"Models","title":"MathOptInterface.NO_SOLUTION","text":"NO_SOLUTION::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNO_SOLUTION: the result vector is empty.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.FEASIBLE_POINT","page":"Models","title":"MathOptInterface.FEASIBLE_POINT","text":"FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nFEASIBLE_POINT: the result vector is a feasible point.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_FEASIBLE_POINT","page":"Models","title":"MathOptInterface.NEARLY_FEASIBLE_POINT","text":"NEARLY_FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBLE_POINT","page":"Models","title":"MathOptInterface.INFEASIBLE_POINT","text":"INFEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBLE_POINT: the result vector is an infeasible point.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.INFEASIBILITY_CERTIFICATE","page":"Models","title":"MathOptInterface.INFEASIBILITY_CERTIFICATE","text":"INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE","page":"Models","title":"MathOptInterface.NEARLY_INFEASIBILITY_CERTIFICATE","text":"NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.REDUCTION_CERTIFICATE","page":"Models","title":"MathOptInterface.REDUCTION_CERTIFICATE","text":"REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.NEARLY_REDUCTION_CERTIFICATE","page":"Models","title":"MathOptInterface.NEARLY_REDUCTION_CERTIFICATE","text":"NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.UNKNOWN_RESULT_STATUS","page":"Models","title":"MathOptInterface.UNKNOWN_RESULT_STATUS","text":"UNKNOWN_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.OTHER_RESULT_STATUS","page":"Models","title":"MathOptInterface.OTHER_RESULT_STATUS","text":"OTHER_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#Conflict-Status","page":"Models","title":"Conflict Status","text":"","category":"section"},{"location":"moi/reference/models/","page":"Models","title":"Models","text":"compute_conflict!\nConflictStatus\nConstraintConflictStatus\nConflictStatusCode\nConflictParticipationStatusCode\nNOT_IN_CONFLICT\nIN_CONFLICT\nMAYBE_IN_CONFLICT","category":"page"},{"location":"moi/reference/models/#MathOptInterface.compute_conflict!","page":"Models","title":"MathOptInterface.compute_conflict!","text":"compute_conflict!(optimizer::AbstractOptimizer)\n\nComputes a minimal subset of constraints such that the model with the other constraint removed is still infeasible.\n\nSome solvers call a set of conflicting constraints an Irreducible Inconsistent Subsystem (IIS).\n\nSee also ConflictStatus and ConstraintConflictStatus.\n\nNote\n\nIf the model is modified after a call to compute_conflict!, the implementor is not obliged to purge the conflict. Any calls to the above attributes may return values for the original conflict without a warning. Similarly, when modifying the model, the conflict can be discarded.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/models/#MathOptInterface.ConflictStatus","page":"Models","title":"MathOptInterface.ConflictStatus","text":"ConflictStatus()\n\nA model attribute for the ConflictStatusCode explaining why the conflict refiner stopped when computing the conflict.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConstraintConflictStatus","page":"Models","title":"MathOptInterface.ConstraintConflictStatus","text":"ConstraintConflictStatus()\n\nA constraint attribute indicating whether the constraint participates in the conflict. Its type is ConflictParticipationStatusCode.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConflictStatusCode","page":"Models","title":"MathOptInterface.ConflictStatusCode","text":"ConflictStatusCode\n\nAn Enum of possible values for the ConflictStatus attribute. This attribute is meant to explain the reason why the conflict finder stopped executing in the most recent call to compute_conflict!.\n\nPossible values are:\n\nCOMPUTE_CONFLICT_NOT_CALLED: the function compute_conflict! has not yet been called\nNO_CONFLICT_EXISTS: there is no conflict because the problem is feasible\nNO_CONFLICT_FOUND: the solver could not find a conflict\nCONFLICT_FOUND: at least one conflict could be found\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.ConflictParticipationStatusCode","page":"Models","title":"MathOptInterface.ConflictParticipationStatusCode","text":"ConflictParticipationStatusCode\n\nAn Enum of possible values for the ConstraintConflictStatus attribute. This attribute is meant to indicate whether a given constraint participates or not in the last computed conflict.\n\nValues\n\nPossible values are:\n\nNOT_IN_CONFLICT: the constraint does not participate in the conflict\nIN_CONFLICT: the constraint participates in the conflict\nMAYBE_IN_CONFLICT: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/models/#MathOptInterface.NOT_IN_CONFLICT","page":"Models","title":"MathOptInterface.NOT_IN_CONFLICT","text":"NOT_IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nNOT_IN_CONFLICT: the constraint does not participate in the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.IN_CONFLICT","page":"Models","title":"MathOptInterface.IN_CONFLICT","text":"IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nIN_CONFLICT: the constraint participates in the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/models/#MathOptInterface.MAYBE_IN_CONFLICT","page":"Models","title":"MathOptInterface.MAYBE_IN_CONFLICT","text":"MAYBE_IN_CONFLICT::ConflictParticipationStatusCode\n\nAn instance of the ConflictParticipationStatusCode enum.\n\nMAYBE_IN_CONFLICT: the constraint may participate in the conflict, the solver was not able to prove that the constraint can be excluded from the conflict\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Nonlinear/reference.md\"","category":"page"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#NonlinearAPI","page":"API Reference","title":"Nonlinear Modeling","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"More information can be found in the Nonlinear section of the manual.","category":"page"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear\nNonlinear.Model","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear","page":"API Reference","title":"MathOptInterface.Nonlinear","text":"Nonlinear\n\nwarning: Warning\nThe Nonlinear submodule is experimental. Until this message is removed, breaking changes may be introduced in any minor or patch release of MathOptInterface.\n\n\n\n\n\n","category":"module"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Model","page":"API Reference","title":"MathOptInterface.Nonlinear.Model","text":"Model()\n\nThe core datastructure for representing a nonlinear optimization problem.\n\nIt has the following fields:\n\nobjective::Union{Nothing,Expression} : holds the nonlinear objective function, if one exists, otherwise nothing.\nexpressions::Vector{Expression} : a vector of expressions in the model.\nconstraints::OrderedDict{ConstraintIndex,Constraint} : a map from ConstraintIndex to the corresponding Constraint. An OrderedDict is used instead of a Vector to support constraint deletion.\nparameters::Vector{Float64} : holds the current values of the parameters.\noperators::OperatorRegistry : stores the operators used in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_expressions","page":"API Reference","title":"Expressions","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ExpressionIndex\nNonlinear.add_expression","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ExpressionIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ExpressionIndex","text":"ExpressionIndex\n\nAn index to a nonlinear expression that is returned by add_expression.\n\nGiven data::Model and ex::ExpressionIndex, use data[ex] to retrieve the corresponding Expression.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_expression","page":"API Reference","title":"MathOptInterface.Nonlinear.add_expression","text":"add_expression(model::Model, expr)::ExpressionIndex\n\nParse expr into a Expression and add to model. Returns an ExpressionIndex that can be interpolated into other input expressions.\n\nexpr must be a type that is supported by parse_expression.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nex = add_expression(model, :($x^2 + 1))\nset_objective(model, :(sqrt($ex)))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_parameters","page":"API Reference","title":"Parameters","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ParameterIndex\nNonlinear.add_parameter","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ParameterIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ParameterIndex","text":"ParameterIndex\n\nAn index to a nonlinear parameter that is returned by add_parameter. Given data::Model and p::ParameterIndex, use data[p] to retrieve the current value of the parameter and data[p] = value to set a new value.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_parameter","page":"API Reference","title":"MathOptInterface.Nonlinear.add_parameter","text":"add_parameter(model::Model, value::Float64)::ParameterIndex\n\nAdd a new parameter to model with the default value value. Returns a ParameterIndex that can be interpolated into other input expressions and used to modify the value of the parameter.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\np = add_parameter(model, 1.2)\nc = add_constraint(model, :($x^2 - $p), MOI.LessThan(0.0))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_objectives","page":"API Reference","title":"Objectives","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.set_objective","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.set_objective","page":"API Reference","title":"MathOptInterface.Nonlinear.set_objective","text":"set_objective(model::Model, obj)::Nothing\n\nParse obj into a Expression and set as the objective function of model.\n\nobj must be a type that is supported by parse_expression.\n\nTo remove the objective, pass nothing.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nset_objective(model, :($x^2 + 1))\nset_objective(model, x)\nset_objective(model, nothing)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_constraints","page":"API Reference","title":"Constraints","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.ConstraintIndex\nNonlinear.add_constraint\nNonlinear.delete","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ConstraintIndex","page":"API Reference","title":"MathOptInterface.Nonlinear.ConstraintIndex","text":"ConstraintIndex\n\nAn index to a nonlinear constraint that is returned by add_constraint.\n\nGiven data::Model and c::ConstraintIndex, use data[c] to retrieve the corresponding Constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.add_constraint","page":"API Reference","title":"MathOptInterface.Nonlinear.add_constraint","text":"add_constraint(\n    model::Model,\n    func,\n    set::Union{\n        MOI.GreaterThan{Float64},\n        MOI.LessThan{Float64},\n        MOI.Interval{Float64},\n        MOI.EqualTo{Float64},\n    },\n)\n\nParse func and set into a Constraint and add to model. Returns a ConstraintIndex that can be used to delete the constraint or query solution information.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc = add_constraint(model, :($x^2), MOI.LessThan(1.0))\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.delete","page":"API Reference","title":"MathOptInterface.Nonlinear.delete","text":"delete(model::Model, c::ConstraintIndex)::Nothing\n\nDelete the constraint index c from model.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc = add_constraint(model, :($x^2), MOI.LessThan(1.0))\ndelete(model, c)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#nonlinear_api_operators","page":"API Reference","title":"User-defined operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.OperatorRegistry\nNonlinear.DEFAULT_UNIVARIATE_OPERATORS\nNonlinear.DEFAULT_MULTIVARIATE_OPERATORS\nNonlinear.register_operator\nNonlinear.register_operator_if_needed\nNonlinear.assert_registered\nNonlinear.check_return_type\nNonlinear.eval_univariate_function\nNonlinear.eval_univariate_gradient\nNonlinear.eval_univariate_hessian\nNonlinear.eval_multivariate_function\nNonlinear.eval_multivariate_gradient\nNonlinear.eval_multivariate_hessian\nNonlinear.eval_logic_function\nNonlinear.eval_comparison_function","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.OperatorRegistry","page":"API Reference","title":"MathOptInterface.Nonlinear.OperatorRegistry","text":"OperatorRegistry()\n\nCreate a new OperatorRegistry to store and evaluate univariate and multivariate operators.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS","page":"API Reference","title":"MathOptInterface.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS","text":"DEFAULT_UNIVARIATE_OPERATORS\n\nThe list of univariate operators that are supported by default.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS\n72-element Vector{Symbol}:\n :+\n :-\n :abs\n :sqrt\n :cbrt\n :abs2\n :inv\n :log\n :log10\n :log2\n ⋮\n :airybi\n :airyaiprime\n :airybiprime\n :besselj0\n :besselj1\n :bessely0\n :bessely1\n :erfcx\n :dawson\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS","page":"API Reference","title":"MathOptInterface.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS","text":"DEFAULT_MULTIVARIATE_OPERATORS\n\nThe list of multivariate operators that are supported by default.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n9-element Vector{Symbol}:\n :+\n :-\n :*\n :^\n :/\n :ifelse\n :atan\n :min\n :max\n\n\n\n\n\n","category":"constant"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.register_operator","page":"API Reference","title":"MathOptInterface.Nonlinear.register_operator","text":"register_operator(\n    model::Model,\n    op::Symbol,\n    nargs::Int,\n    f::Function,\n    [∇f::Function],\n    [∇²f::Function],\n)\n\nRegister the user-defined operator op with nargs input arguments in model.\n\nUnivariate functions\n\nf(x::T)::T must be a function that takes a single input argument x and returns the function evaluated at x. If ∇f and ∇²f are not provided, f must support any Real input type T.\n∇f(x::T)::T is a function that takes a single input argument x and returns the first derivative of f with respect to x. If ∇²f is not provided, ∇f must support any Real input type T.\n∇²f(x::T)::T is a function that takes a single input argument x and returns the second derivative of f with respect to x.\n\nMultivariate functions\n\nf(x::T...)::T must be a function that takes a nargs input arguments x and returns the function evaluated at x. If ∇f and ∇²f are not provided, f must support any Real input type T.\n∇f(g::AbstractVector{T}, x::T...)::T is a function that takes a cache vector g of length length(x), and fills each element g[i] with the partial derivative of f with respect to x[i].\n∇²f(H::AbstractMatrix, x::T...)::T is a function that takes a matrix H and fills the lower-triangular components H[i, j] with the Hessian of f with respect to x[i] and x[j] for i >= j.\n\nNotes for multivariate Hessians\n\nH has size(H) == (length(x), length(x)), but you must not access elements H[i, j] for i > j.\nH is dense, but you do not need to fill structural zeros.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.register_operator_if_needed","page":"API Reference","title":"MathOptInterface.Nonlinear.register_operator_if_needed","text":"register_operator_if_needed(\n    registry::OperatorRegistry,\n    op::Symbol,\n    nargs::Int,\n    f::Function;\n)\n\nSimilar to register_operator, but this function warns if the function is not registered, and skips silently if it already is.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.assert_registered","page":"API Reference","title":"MathOptInterface.Nonlinear.assert_registered","text":"assert_registered(registry::OperatorRegistry, op::Symbol, nargs::Int)\n\nThrow an error if op is not registered in registry with nargs arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.check_return_type","page":"API Reference","title":"MathOptInterface.Nonlinear.check_return_type","text":"check_return_type(::Type{T}, ret::S) where {T,S}\n\nOverload this method for new types S to throw an informative error if a user-defined function returns the type S instead of T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_function","text":"eval_univariate_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_gradient","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_gradient","text":"eval_univariate_gradient(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the first-derivative of the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_univariate_hessian","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_univariate_hessian","text":"eval_univariate_hessian(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::T,\n) where {T}\n\nEvaluate the second-derivative of the operator op(x)::T, where op is a univariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_function","text":"eval_multivariate_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the operator op(x)::T, where op is a multivariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_gradient","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_gradient","text":"eval_multivariate_gradient(\n    registry::OperatorRegistry,\n    op::Symbol,\n    g::AbstractVector{T},\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the gradient of operator g .= ∇op(x), where op is a multivariate function in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_multivariate_hessian","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_multivariate_hessian","text":"eval_multivariate_hessian(\n    registry::OperatorRegistry,\n    op::Symbol,\n    H::AbstractMatrix,\n    x::AbstractVector{T},\n) where {T}\n\nEvaluate the Hessian of operator ∇²op(x), where op is a multivariate function in registry.\n\nThe Hessian is stored in the lower-triangular part of the matrix H.\n\nnote: Note\nImplementations of the Hessian operators will not fill structural zeros. Therefore, before calling this function you should pre-populate the matrix H with 0.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_logic_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_logic_function","text":"eval_logic_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    lhs::T,\n    rhs::T,\n)::Bool where {T}\n\nEvaluate (lhs op rhs)::Bool, where op is a logic operator in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.eval_comparison_function","page":"API Reference","title":"MathOptInterface.Nonlinear.eval_comparison_function","text":"eval_comparison_function(\n    registry::OperatorRegistry,\n    op::Symbol,\n    lhs::T,\n    rhs::T,\n)::Bool where {T}\n\nEvaluate (lhs op rhs)::Bool, where op is a comparison operator in registry.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#Automatic-differentiation-backends","page":"API Reference","title":"Automatic-differentiation backends","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.Evaluator\nNonlinear.AbstractAutomaticDifferentiation\nNonlinear.ExprGraphOnly\nNonlinear.SparseReverseMode","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Evaluator","page":"API Reference","title":"MathOptInterface.Nonlinear.Evaluator","text":"Evaluator(\n    model::Model,\n    backend::AbstractAutomaticDifferentiation,\n    ordered_variables::Vector{MOI.VariableIndex},\n)\n\nCreate Evaluator, a subtype of MOI.AbstractNLPEvaluator, from Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation","page":"API Reference","title":"MathOptInterface.Nonlinear.AbstractAutomaticDifferentiation","text":"AbstractAutomaticDifferentiation\n\nAn abstract type for extending Evaluator.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ExprGraphOnly","page":"API Reference","title":"MathOptInterface.Nonlinear.ExprGraphOnly","text":"ExprGraphOnly() <: AbstractAutomaticDifferentiation\n\nThe default implementation of AbstractAutomaticDifferentiation. The only supported feature is :ExprGraph.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.SparseReverseMode","page":"API Reference","title":"MathOptInterface.Nonlinear.SparseReverseMode","text":"SparseReverseMode() <: AbstractAutomaticDifferentiation\n\nAn implementation of AbstractAutomaticDifferentiation that uses sparse reverse-mode automatic differentiation to compute derivatives. Supports all features in the MOI nonlinear interface.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#Data-structure","page":"API Reference","title":"Data-structure","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/reference/","page":"API Reference","title":"API Reference","text":"Nonlinear.Node\nNonlinear.NodeType\nNonlinear.Expression\nNonlinear.Constraint\nNonlinear.adjacency_matrix\nNonlinear.parse_expression\nNonlinear.convert_to_expr\nNonlinear.ordinal_index","category":"page"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Node","page":"API Reference","title":"MathOptInterface.Nonlinear.Node","text":"struct Node\n    type::NodeType\n    index::Int\n    parent::Int\nend\n\nA single node in a nonlinear expression tree. Used by Expression.\n\nSee the MathOptInterface documentation for information on how the nodes and values form an expression tree.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.NodeType","page":"API Reference","title":"MathOptInterface.Nonlinear.NodeType","text":"NodeType\n\nAn enum describing the possible node types. Each Node has a .index field, which should be interpreted as follows:\n\nNODE_CALL_MULTIVARIATE: the index into operators.multivariate_operators\nNODE_CALL_UNIVARIATE: the index into operators.univariate_operators\nNODE_LOGIC: the index into operators.logic_operators\nNODE_COMPARISON: the index into operators.comparison_operators\nNODE_MOI_VARIABLE: the value of MOI.VariableIndex(index) in the user's space of the model.\nNODE_VARIABLE: the 1-based index of the internal vector\nNODE_VALUE: the index into the .values field of Expression\nNODE_PARAMETER: the index into data.parameters\nNODE_SUBEXPRESSION:  the index into data.expressions\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Expression","page":"API Reference","title":"MathOptInterface.Nonlinear.Expression","text":"struct Expression\n    nodes::Vector{Node}\n    values::Vector{Float64}\nend\n\nThe core type that represents a nonlinear expression. See the MathOptInterface documentation for information on how the nodes and values form an expression tree.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.Constraint","page":"API Reference","title":"MathOptInterface.Nonlinear.Constraint","text":"struct Constraint\n    expression::Expression\n    set::Union{\n        MOI.LessThan{Float64},\n        MOI.GreaterThan{Float64},\n        MOI.EqualTo{Float64},\n        MOI.Interval{Float64},\n    }\nend\n\nA type to hold information relating to the nonlinear constraint f(x) in S, where f(x) is defined by .expression, and S is .set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.adjacency_matrix","page":"API Reference","title":"MathOptInterface.Nonlinear.adjacency_matrix","text":"adjacency_matrix(nodes::Vector{Node})\n\nCompute the sparse adjacency matrix describing the parent-child relationships in nodes.\n\nThe element (i, j) is true if there is an edge from node[j] to node[i]. Since we get a column-oriented matrix, this gives us a fast way to look up the edges leaving any node (i.e., the children).\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.parse_expression","page":"API Reference","title":"MathOptInterface.Nonlinear.parse_expression","text":"parse_expression(data::Model, input)::Expression\n\nParse input into a Expression.\n\n\n\n\n\nparse_expression(\n    data::Model,\n    expr::Expression,\n    input::Any,\n    parent_index::Int,\n)::Expression\n\nParse input into a Expression, and add it to expr as a child of expr.nodes[parent_index]. Existing subexpressions and parameters are stored in data.\n\nYou can extend parsing support to new types of objects by overloading this method with a different type on input::Any.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.convert_to_expr","page":"API Reference","title":"MathOptInterface.Nonlinear.convert_to_expr","text":"convert_to_expr(data::Model, expr::Expression)\n\nConvert the Expression expr into a Julia Expr.\n\nsubexpressions are represented by a ExpressionIndex object.\nparameters are represented by a ParameterIndex object.\nvariables are represennted by an MOI.VariableIndex object.\n\n\n\n\n\nconvert_to_expr(\n    evaluator::Evaluator,\n    expr::Expression;\n    moi_output_format::Bool,\n)\n\nConvert the Expression expr into a Julia Expr.\n\nIf moi_output_format = true:\n\nsubexpressions will be converted to Julia Expr and substituted into the output expression.\nthe current value of each parameter will be interpolated into the expression\nvariables will be represented in the form x[MOI.VariableIndex(i)]\n\nIf moi_output_format = false:\n\nsubexpressions will be represented by a ExpressionIndex object.\nparameters will be represented by a ParameterIndex object.\nvariables will be represennted by an MOI.VariableIndex object.\n\nwarning: Warning\nTo use moi_output_format = true, you must have first called MOI.initialize with :ExprGraph as a requested feature.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Nonlinear/reference/#MathOptInterface.Nonlinear.ordinal_index","page":"API Reference","title":"MathOptInterface.Nonlinear.ordinal_index","text":"ordinal_index(evaluator::Evaluator, c::ConstraintIndex)::Int\n\nReturn the 1-indexed value of the constraint index c in evaluator.\n\nExamples\n\nmodel = Model()\nx = MOI.VariableIndex(1)\nc1 = add_constraint(model, :($x^2), MOI.LessThan(1.0))\nc2 = add_constraint(model, :($x^2), MOI.LessThan(1.0))\nevaluator = Evaluator(model)\nMOI.initialize(evaluator, Symbol[])\nordinal_index(evaluator, c2)  # Returns 2\ndelete(model, c1)\nevaluator = Evaluator(model)\nMOI.initialize(evaluator, Symbol[])\nordinal_index(model, c2)  # Returns 1\n\n\n\n\n\n","category":"function"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"EditURL = \"https://github.com/lanl-ansi/Juniper.jl/blob/62532341586d447f19c7360715333ba62a42bea9/README.md\"","category":"page"},{"location":"packages/Juniper/#Juniper","page":"lanl-ansi/Juniper.jl","title":"Juniper","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"(Image: CI) (Image: codecov) (Image: Documentation)","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Juniper (Jump Nonlinear Integer Program solver) is a solver for mixed-integer nonlinear programs. ","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"It is a heuristic which is not guaranteed to find the global optimum. If you need the global optimum, check out Alpine.","category":"page"},{"location":"packages/Juniper/#Installation","page":"lanl-ansi/Juniper.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Install Juniper using the Julia package manager:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"packages/Juniper/#Use-with-JuMP","page":"lanl-ansi/Juniper.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"Use Juniper with JuMP as follows:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"using JuMP, Juniper, Ipopt\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\"=>0)\noptimizer = optimizer_with_attributes(Juniper.Optimizer, \"nl_solver\"=>ipopt)\nmodel = Model(optimizer)\nv = [10, 20, 12, 23, 42]\nw = [12, 45, 12, 22, 21]\n@variable(model, x[1:5], Bin)\n@objective(model, Max, v' * x)\n@constraint(model, sum(w[i]*x[i]^2 for i in 1:5) <= 45)\noptimize!(model)\nprintln(termination_status(model))\nprintln(objective_value(model))\nprintln(value.(x))","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The nl_solver is used by Juniper to solve continuous nonlinear sub-problems while Juniper searches for acceptable assignments to the discrete variables. A common choice is Ipopt, but any optimizer that supports the continuous relaxation of the model may be used.","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"To solve problems with more complex nonlinear functions, use the @NLconstraint and @NLobjective JuMP macros.","category":"page"},{"location":"packages/Juniper/#Documentation","page":"lanl-ansi/Juniper.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The online documentation is available at https://lanl-ansi.github.io/Juniper.jl/stable/.","category":"page"},{"location":"packages/Juniper/#Feasibility-pump","page":"lanl-ansi/Juniper.jl","title":"Feasibility pump","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"If Juniper has difficulty finding feasible solutions on your model, try adding a solver that supports integer variables (for example, HiGHS) to run a feasibility pump:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"using JuMP, Juniper, Ipopt, HiGHS\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0)\nhighs = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false)\nmodel = Model(\n    optimizer_with_attributes(\n        Juniper.Optimizer,\n        \"nl_solver\" => ipopt,\n        \"mip_solver\" => highs,\n    ),\n)","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"The feasibility pump is used at the start of Juniper to find a feasible solution before the branch and bound part starts.  For some classes of problems this can be a highly effective pre-processor.","category":"page"},{"location":"packages/Juniper/#Citing-Juniper","page":"lanl-ansi/Juniper.jl","title":"Citing Juniper","text":"","category":"section"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"If you find Juniper useful in your work, we kindly request that you cite the following paper or technical report:","category":"page"},{"location":"packages/Juniper/","page":"lanl-ansi/Juniper.jl","title":"lanl-ansi/Juniper.jl","text":"@inproceedings{juniper,\n     Author = {Ole Kröger and Carleton Coffrin and Hassan Hijazi and Harsha Nagarajan},\n     Title = {Juniper: An Open-Source Nonlinear Branch-and-Bound Solver in Julia},\n     booktitle=\"Integration of Constraint Programming, Artificial Intelligence, and Operations Research\",\n     pages=\"377--386\",\n     year=\"2018\",\n     publisher=\"Springer International Publishing\",\n     isbn=\"978-3-319-93031-2\"\n}","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"EditURL = \"diet.jl\"","category":"page"},{"location":"tutorials/linear/diet/#The-diet-problem","page":"The diet problem","title":"The diet problem","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"The purpose of this tutorial is to demonstrate how to incorporate DataFrames into a JuMP model. As an example, we use classic Stigler diet problem.","category":"page"},{"location":"tutorials/linear/diet/#Required-packages","page":"The diet problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"using JuMP\nimport CSV\nimport DataFrames\nimport HiGHS\nimport Test  #hide","category":"page"},{"location":"tutorials/linear/diet/#Formulation","page":"The diet problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We wish to cook a nutritionally balanced meal by choosing the quantity of each food f to eat from a set of foods F in our kitchen.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Each food f has a cost, c_f, as well as a macro-nutrient profile a_mf for each macro-nutrient m in M.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Because we care about a nutritionally balanced meal, we set some minimum and maximum limits for each nutrient, which we denote l_m and u_m respectively.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Furthermore, because we are optimizers, we seek the minimum cost solution.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"With a little effort, we can formulate our dinner problem as the following linear program:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"beginaligned\nmin  sumlimits_f in F c_f x_f \ntextst   l_m le sumlimits_f in F a_mf x_f le u_m  forall m in M \n x_f ge 0  forall f in F\nendaligned","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"In the rest of this tutorial, we will create and solve this problem in JuMP, and learn what we should cook for dinner.","category":"page"},{"location":"tutorials/linear/diet/#Data","page":"The diet problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"First, we need some data for the problem. For this tutorial, we'll write CSV files to a temporary directory from Julia. If you have existing files, you could change the filenames to point to them instead.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"dir = mktempdir()","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"The first file is a list of foods with their macro-nutrient profile:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"food_csv_filename = joinpath(dir, \"diet_foods.csv\")\nopen(food_csv_filename, \"w\") do io\n    write(\n        io,\n        \"\"\"\n        name,cost,calories,protein,fat,sodium\n        hamburger,2.49,410,24,26,730\n        chicken,2.89,420,32,10,1190\n        hot dog,1.50,560,20,32,1800\n        fries,1.89,380,4,19,270\n        macaroni,2.09,320,12,10,930\n        pizza,1.99,320,15,12,820\n        salad,2.49,320,31,12,1230\n        milk,0.89,100,8,2.5,125\n        ice cream,1.59,330,8,10,180\n        \"\"\",\n    )\n    return\nend\nfoods = CSV.read(food_csv_filename, DataFrames.DataFrame)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Here, F is foods.name and c_f is foods.cost. (We're also playing a bit loose the term \"macro-nutrient\" by including calories and sodium.)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We also need our minimum and maximum limits:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"nutrient_csv_filename = joinpath(dir, \"diet_nutrient.csv\")\nopen(nutrient_csv_filename, \"w\") do io\n    write(\n        io,\n        \"\"\"\n        nutrient,min,max\n        calories,1800,2200\n        protein,91,\n        fat,0,65\n        sodium,0,1779\n        \"\"\",\n    )\n    return\nend\nlimits = CSV.read(nutrient_csv_filename, DataFrames.DataFrame)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Protein is missing data for the maximum. Let's fix that using coalesce:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"limits.max = coalesce.(limits.max, Inf)\nlimits","category":"page"},{"location":"tutorials/linear/diet/#JuMP-formulation","page":"The diet problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Now we're ready to convert our mathematical formulation into a JuMP model.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"First, create a new JuMP model. Since we have a linear program, we'll use HiGHS as our optimizer:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Next, we create a set of decision variables x, with one element for each row in the DataFrame, and each x has a lower bound of 0:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@variable(model, x[foods.name] >= 0)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"To simplify things later on, we store the vector as a new column x in the DataFrame foods. Since x is a DenseAxisArray, we first need to convert it to an Array:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"foods.x = Array(x)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Our objective is to minimize the total cost of purchasing food:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@objective(model, Min, sum(foods.cost .* foods.x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"For the next component, we need to add a constraint that our total intake of each component is within the limits contained in the limits DataFrame:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"@constraint(\n    model,\n    [row in eachrow(limits)],\n    row.min <= sum(foods[!, row.nutrient] .* foods.x) <= row.max,\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"What does our model look like?","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"print(model)","category":"page"},{"location":"tutorials/linear/diet/#Solution","page":"The diet problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"Let's optimize and take a look at the solution:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"optimize!(model)\nTest.@test primal_status(model) == FEASIBLE_POINT        #hide\nTest.@test objective_value(model) ≈ 11.8288 atol = 1e-4  #hide\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We found an optimal solution. Let's see what the optimal solution is:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"for row in eachrow(foods)\n    println(row.name, \" = \", value(row.x))\nend","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"That's a lot of milk and ice cream, and sadly, we only get 0.6 of a hamburger.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"We can also use the function Containers.rowtable to easily convert the result into a DataFrame:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"table = Containers.rowtable(value, x; header = [:food, :quantity])\nsolution = DataFrames.DataFrame(table)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"This makes it easy to perform analyses our solution:","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"filter!(row -> row.quantity > 0.0, solution)","category":"page"},{"location":"tutorials/linear/diet/#Problem-modification","page":"The diet problem","title":"Problem modification","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"JuMP makes it easy to take an existing model and modify it by adding extra constraints. Let's see what happens if we add a constraint that we can buy at most 6 units of milk or ice cream combined.","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"dairy_foods = [\"milk\", \"ice cream\"]\nis_dairy = map(name -> name in dairy_foods, foods.name)\ndairy_constraint = @constraint(model, sum(foods[is_dairy, :x]) <= 6)\noptimize!(model)\nTest.@test termination_status(model) == INFEASIBLE  #hide\nTest.@test primal_status(model) == NO_SOLUTION      #hide\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"There exists no feasible solution to our problem. Looks like we're stuck eating ice cream for dinner.","category":"page"},{"location":"tutorials/linear/diet/#Next-steps","page":"The diet problem","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/diet/","page":"The diet problem","title":"The diet problem","text":"You can delete a constraint using delete(model, dairy_constraint). Can you add a different constraint to provide a diet with less dairy?\nSome food items (like hamburgers) are discrete. You can use set_integer to force a variable to take integer values. What happens to the solution if you do?","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"EditURL = \"n-queens.jl\"","category":"page"},{"location":"tutorials/linear/n-queens/#N-Queens","page":"N-Queens","title":"N-Queens","text":"","category":"section"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"This tutorial was originally contributed by Matthew Helm and Mathieu Tanneau.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"The N-Queens problem involves placing N queens on an N x N chessboard such that none of the queens attacks another. In chess, a queen can move vertically, horizontally, and diagonally so there cannot be more than one queen on any given row, column, or diagonal.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"(Image: Four Queens)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Note that none of the queens above are able to attack any other as a result of their careful placement.","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"using JuMP\nimport HiGHS\nimport LinearAlgebra","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"N-Queens","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"N = 8\n\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Next, let's create an N x N chessboard of binary values. 0 will represent an empty space on the board and 1 will represent a space occupied by one of our queens:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"@variable(model, x[1:N, 1:N], Bin);\nnothing #hide","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"Now we can add our constraints:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"There must be exactly one queen in a given row/column","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"for i in 1:N\n    @constraint(model, sum(x[i, :]) == 1)\n    @constraint(model, sum(x[:, i]) == 1)\nend","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"There can only be one queen on any given diagonal","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"for i in -(N - 1):(N-1)\n    @constraint(model, sum(LinearAlgebra.diag(x, i)) <= 1)\n    @constraint(model, sum(LinearAlgebra.diag(reverse(x; dims = 1), i)) <= 1)\nend","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"We are ready to put our model to work and see if it is able to find a feasible solution:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"optimize!(model)","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"We can now review the solution that our model found:","category":"page"},{"location":"tutorials/linear/n-queens/","page":"N-Queens","title":"N-Queens","text":"solution = round.(Int, value.(x))","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"EditURL = \"https://github.com/jump-dev/SDPNAL.jl/blob/00a3fa19f4e1235587948113b0b681da17f4dab5/README.md\"","category":"page"},{"location":"packages/SDPNAL/#SDPNAL.jl","page":"jump-dev/SDPNAL.jl","title":"SDPNAL.jl","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"SDPNAL.jl is wrapper for the SDPNALplus solver.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"an exported sdpnalplus function that is a thin wrapper on top of the sdpnalplus MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SDPNAL/#Affiliation","page":"jump-dev/SDPNAL.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SDPNALplus.","category":"page"},{"location":"packages/SDPNAL/#License","page":"jump-dev/SDPNAL.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"SDPNAL.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"The underlying solver, SDPNALplus is licensed under the Creative Commons Attribution-ShareAlike 4.0 International Public License.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"In addition, SDPNAL requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SDPNAL/#Use-with-JuMP","page":"jump-dev/SDPNAL.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"To use SDPNAL with JuMP, do:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"using JuMP, SDPNAL\nmodel = Model(SDPNAL.Optimizer)\nset_attribute(model, \"printlevel\", 0)","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"Note that, contrary to implementation of other solver-independent interfaces, using SDPNAL from JuMP or MOI fully exploits the particular structures of the SDPNAL interface and does not create superfluous slack variables and equality constraints as discussed in the SDPNAL guide:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"A new interface is necessary to facilitate the modeling of an SDP problem for SDPNAL+ because of latter’s flexibility to directly accept inequality constraints of the form “l ≤ B(X) ≤ u”, and bound constraints of the form “L ≤ X ≤ U”. The flexibility can significantly simplify the generation of the data in the SDPNAL+ format as compared to what need to be done in CVX or YALMIP to reformulate them as equality constraints through introducing extra variables. In addition, the final number of equality constraints present in the data input to SDPNAL+ can also be substantially fewer than those present in CVX or YALMIP. It is important to note here that the number of equality constraints present in the generated problem data can greatly affect the computational efficiency of the solvers, especially for interior-point based solvers.","category":"page"},{"location":"packages/SDPNAL/#Installation","page":"jump-dev/SDPNAL.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SDPNALplus software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"Then, install SDPNAL.jl using Pkg.add:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"import Pkg\nPkg.add(\"SDPNAL\")","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"There is a startup.m file at the root of the SDPNAL folder. This adds all subdirectories recursively when MATLAB starts. However, the interface directory contains a .git subdirectory which contains a very large number of files. Because of this, MATLAB crashes if SDPNAL is in its path because the startup.m requests MATLAB to try to parse all the files in the .git folder. To resolve this problem, delete the startup.m file and .git folder, and add the subdirectories manually your toolbox/local/pathdef.m file as follows:","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"function p = pathdef\n\n% (...)\n\np = [...\n%%% BEGIN ENTRIES %%%\n'/path/to/SDPNALv1.0:', ...\n'/path/to/SDPNALv1.0/interface:', ...\n'/path/to/SDPNALv1.0/mexfun:', ...\n'/path/to/SDPNALv1.0/solver:', ...\n'/path/to/SDPNALv1.0/solver_main_default:', ...\n'/path/to/SDPNALv1.0/util:', ...\n% (...)","category":"page"},{"location":"packages/SDPNAL/","page":"jump-dev/SDPNAL.jl","title":"jump-dev/SDPNAL.jl","text":"If you have SDPT3 in addition to SDPNAL in the MATLAB path (that is, the toolbox/local/pathdef.m file) then you might have issues because both solvers define a validate function, and this might make SDPNAL call SDPT3's validate function instead of SDPT3's validate function.","category":"page"},{"location":"should_i_use/#Should-you-use-JuMP?","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP is an algebraic modeling language for mathematical optimization written in the Julia language.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"This page explains when you should consider using JuMP, and importantly, when you should not use JuMP.","category":"page"},{"location":"should_i_use/#When-should-you-use-JuMP?","page":"Should you use JuMP?","title":"When should you use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"You should use JuMP if you have a constrained optimization problem for which you can formulate using the language of mathematical programming, that is:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"a set of decision variables\na scalar- or vector-valued objective function\na set of constraints.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Key reasons to use JuMP include:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"User friendliness\nJuMP has syntax that mimics natural mathematical expressions. (See the section on algebraic modeling languages.)\nSolver independence\nJuMP uses a generic solver-independent interface provided by the MathOptInterface package, making it easy to change between a number of open-source and commercial optimization software packages (\"solvers\"). The Supported solvers section contains a table of the currently supported solvers.\nEase of embedding\nJuMP itself is written purely in Julia. Solvers are the only binary dependencies.\nJuMP provides automatic installation of many open-source solvers. This is different to modeling languages in Python which require you to download and install a solver yourself.\nBecause it is embedded in a general-purpose programming language, JuMP makes it easy to solve optimization problems as part of a larger workflow, for example, inside a simulation, behind a web server, or as a subproblem in a decomposition algorithm. As a trade-off, JuMP's syntax is constrained by the syntax and functionality available in Julia.\nJuMP is MPL licensed, meaning that it can be embedded in commercial software that complies with the terms of the license.\nSpeed\nBenchmarking has shown that JuMP can create problems at similar speeds to special-purpose modeling languages such as AMPL.\nJuMP communicates with most solvers in memory, avoiding the need to write intermediary files.\nAccess to advanced algorithmic techniques\nJuMP supports efficient in-memory re-solves of linear programs, which previously required using solver-specific or low-level C++ libraries.\nJuMP provides access to solver-independent and solver-dependent Callbacks.","category":"page"},{"location":"should_i_use/#When-should-you-not-use-JuMP?","page":"Should you use JuMP?","title":"When should you not use JuMP?","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP supports a broad range of optimization classes. However, there are still some that it doesn't support, or that are better supported by other software packages.","category":"page"},{"location":"should_i_use/#You-want-to-optimize-a-complicated-Julia-function","page":"Should you use JuMP?","title":"You want to optimize a complicated Julia function","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Packages in Julia compose well. It's common for people to pick two unrelated packages and use them in conjunction to create novel behavior. JuMP isn't one of those packages.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If you want to optimize an ordinary differential equation from DifferentialEquations.jl or tune a neural network from Flux.jl, consider using other packages such as:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Optim.jl\nOptimization.jl\nNLPModels.jl\nNonconvex.jl","category":"page"},{"location":"should_i_use/#Black-box,-derivative-free,-or-unconstrained-optimization","page":"Should you use JuMP?","title":"Black-box, derivative free, or unconstrained optimization","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does support nonlinear programs with constraints and objectives containing user-defined operators. However, the functions must be automatically differentiable, or need to provide explicit derivatives. (See User-defined operators for more information.)","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If your function is a black-box that is non-differentiable (for example, it is the output of a simulation written in C++), JuMP is not the right tool for the job. This also applies if you want to use a derivative free method.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Even if your problem is differentiable, if it is unconstrained there is limited benefit (and downsides in the form of more overhead) to using JuMP over tools which are only concerned with function minimization.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Optim.jl\nOptimization.jl\nNLopt.jl","category":"page"},{"location":"should_i_use/#Optimal-control-problems","page":"Should you use JuMP?","title":"Optimal control problems","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP supports formulating optimal control problems as large nonlinear programs (see, for example, Optimal control for a Space Shuttle reentry trajectory). However, the nonlinear interface has a number of limitations (for example, the need to write out the dynamics in algebraic form) that mean JuMP might not be the right tool for the job.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"CasADi [MATLAB/Python], CasADi.jl\nInfiniteOpt.jl\npyomo.DAE [Python]","category":"page"},{"location":"should_i_use/#Disciplined-convex-programming","page":"Should you use JuMP?","title":"Disciplined convex programming","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does not support disciplined convex programming (DCP).","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Convex.jl\nCVXPY [Python]\nYALMIP [MATLAB]","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"note: Note\nConvex.jl is also built on MathOptInterface, and shares the same set of underlying solvers. However, you input problems differently, and Convex.jl checks that the problem is DCP.","category":"page"},{"location":"should_i_use/#Stochastic-programming","page":"Should you use JuMP?","title":"Stochastic programming","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP requires deterministic input data.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"If you have stochastic input data, consider using a JuMP extension such as:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"InfiniteOpt.jl\nStochasticPrograms.jl\nSDDP.jl","category":"page"},{"location":"should_i_use/#Polyhedral-computations","page":"Should you use JuMP?","title":"Polyhedral computations","text":"","category":"section"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"JuMP does not provide tools for working with the polyhedron formed by the set of linear constraints.","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Alternatives to consider are:","category":"page"},{"location":"should_i_use/","page":"Should you use JuMP?","title":"Should you use JuMP?","text":"Polyhedra.jl (See the documentation to create a polyhedron from a JuMP model.)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"EditURL = \"design_patterns_for_larger_models.jl\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Design-patterns-for-larger-models","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"JuMP makes it easy to build and solve optimization models. However, once you start to construct larger models, and especially ones that interact with external data sources or have customizable sets of variables and constraints based on client choices, you may find that your scripts become unwieldy. This tutorial demonstrates a variety of ways in which you can structure larger JuMP models to improve their readability and maintainability.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"tip: Tip\nThis tutorial is more advanced than the other \"Getting started\" tutorials. It's in the \"Getting started\" section to give you an early preview of how JuMP makes it easy to structure larger models. However, if you are new to JuMP you may want to briefly skim the tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Overview","page":"Design patterns for larger models","title":"Overview","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"This tutorial uses explanation-by-example. We're going to start with a simple knapsack model, and then expand it to add various features and structure.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#A-simple-script","page":"Design patterns for larger models","title":"A simple script","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Your first prototype of a JuMP model is probably a script that uses a small set of hard-coded data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"using JuMP, HiGHS\nprofit = [5, 3, 2, 7, 4]\nweight = [2, 8, 4, 2, 5]\ncapacity = 10\nN = 5\nmodel = Model(HiGHS.Optimizer)\n@variable(model, x[1:N], Bin)\n@objective(model, Max, sum(profit[i] * x[i] for i in 1:N))\n@constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The benefits of this approach are:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"it is quick to code\nit is quick to make changes.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The downsides include:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"all variables are global (read Performance tips)\nit is easy to introduce errors, for example, having profit and weight be vectors of different lengths, or not match N\nthe solution, x[i], is hard to interpret without knowing the order in which we provided the data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Wrap-the-model-in-a-function","page":"Design patterns for larger models","title":"Wrap the model in a function","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"A good next step is to wrap your model in a function. This is useful for a few reasons:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"it removes global variables\nit encapsulates the JuMP model and forces you to clarify your inputs and outputs\nwe can add some error checking.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_1(profit::Vector, weight::Vector, capacity::Real)\n    if length(profit) != length(weight)\n        throw(DimensionMismatch(\"profit and weight are different sizes\"))\n    end\n    N = length(weight)\n    model = Model(HiGHS.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @objective(model, Max, sum(profit[i] * x[i] for i in 1:N))\n    @constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\n    optimize!(model)\n    return value.(x)\nend\n\nsolve_knapsack_1([5, 3, 2, 7, 4], [2, 8, 4, 2, 5], 10)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Create-better-data-structures","page":"Design patterns for larger models","title":"Create better data structures","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Although we can check for errors like mis-matched vector lengths, if you start to develop models with a lot of data, keeping track of vectors and lengths and indices is fragile and a common source of bugs. A good solution is to use Julia's type system to create an abstraction over your data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For example, we can create a struct that represents a single object, with a constructor that lets us validate assumptions on the input data:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct KnapsackObject\n    profit::Float64\n    weight::Float64\n    function KnapsackObject(profit::Float64, weight::Float64)\n        if weight < 0\n            throw(DomainError(\"Weight of object cannot be negative\"))\n        end\n        return new(profit, weight)\n    end\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"as well as a struct that holds a dictionary of objects and the knapsack's capacity:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct KnapsackData\n    objects::Dict{String,KnapsackObject}\n    capacity::Float64\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Here's what our data might look like now:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"objects = Dict(\n    \"apple\" => KnapsackObject(5.0, 2.0),\n    \"banana\" => KnapsackObject(3.0, 8.0),\n    \"cherry\" => KnapsackObject(2.0, 4.0),\n    \"date\" => KnapsackObject(7.0, 2.0),\n    \"eggplant\" => KnapsackObject(4.0, 5.0),\n)\ndata = KnapsackData(objects, 10.0)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If you want, you can add custom printing to make it easier to visualize:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function Base.show(io::IO, data::KnapsackData)\n    println(io, \"A knapsack with capacity $(data.capacity) and possible items:\")\n    for (k, v) in data.objects\n        println(\n            io,\n            \"  $(rpad(k, 8)) : profit = $(v.profit), weight = $(v.weight)\",\n        )\n    end\n    return\nend\n\ndata","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Then, we can re-write our solve_knapsack function to take our KnapsackData as input:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_2(data::KnapsackData)\n    model = Model(HiGHS.Optimizer)\n    @variable(model, x[keys(data.objects)], Bin)\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend\n\nsolve_knapsack_2(data)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Read-in-data-from-files","page":"Design patterns for larger models","title":"Read in data from files","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Having a data structure is a good step. But it is still annoying that we have to hard-code the data into Julia. A good next step is to separate the data into an external file format; JSON is a common choice.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"json_data = \"\"\"\n{\n    \"objects\": {\n        \"apple\": {\"profit\": 5.0, \"weight\": 2.0},\n        \"banana\": {\"profit\": 3.0, \"weight\": 8.0},\n        \"cherry\": {\"profit\": 2.0, \"weight\": 4.0},\n        \"date\": {\"profit\": 7.0, \"weight\": 2.0},\n        \"eggplant\": {\"profit\": 4.0, \"weight\": 5.0}\n    },\n    \"capacity\": 10.0\n}\n\"\"\"\ntemp_dir = mktempdir()\nknapsack_json_filename = joinpath(temp_dir, \"knapsack.json\")\n# Instead of writing a new file here you could replace `knapsack_json_filename`\n# with the path to a local file.\nwrite(knapsack_json_filename, json_data);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now let's write a function that reads this file and builds a KnapsackData object:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import JSON\n\nfunction read_data(filename)\n    d = JSON.parsefile(filename)\n    return KnapsackData(\n        Dict(\n            k => KnapsackObject(v[\"profit\"], v[\"weight\"]) for\n            (k, v) in d[\"objects\"]\n        ),\n        d[\"capacity\"],\n    )\nend\n\ndata = read_data(knapsack_json_filename)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-options-via-if-else","page":"Design patterns for larger models","title":"Add options via if-else","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"At this point, we have data in a file format which we can load and solve a single problem. For many users, this might be sufficient. However, at some point you may be asked to add features like \"but what if we want to take more than one of a particular item?\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If this is the first time that you've been asked to add a feature, adding options via if-else statements is a good approach. For example, we might write:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_3(data::KnapsackData; binary_knapsack::Bool)\n    model = Model(HiGHS.Optimizer)\n    if binary_knapsack\n        @variable(model, x[keys(data.objects)], Bin)\n    else\n        @variable(model, x[keys(data.objects)] >= 0, Int)\n    end\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we can solve the binary knapsack:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_3(data; binary_knapsack = true)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"And an integer knapsack where we can take more than one copy of each item:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_3(data; binary_knapsack = false)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-configuration-options-via-dispatch","page":"Design patterns for larger models","title":"Add configuration options via dispatch","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"If you get repeated requests to add different options, you'll quickly find yourself in a mess of different flags and if-else statements. It's hard to write, hard to read, and hard to ensure you haven't introduced any bugs. A good solution is to use Julia's type dispatch to control the configuration of the model. The easiest way to explain this is by example.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"First, start by defining a new abstract type, as well as new subtypes for each of our options. These types are going to control the configuration of the knapsack model.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"abstract type AbstractConfiguration end\n\nstruct BinaryKnapsackConfig <: AbstractConfiguration end\n\nstruct IntegerKnapsackConfig <: AbstractConfiguration end","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Then, we rewrite our solve_knapsack function to take a config argument, and we introduce an add_knapsack_variables function to abstract the creation of our variables.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_4(data::KnapsackData, config::AbstractConfiguration)\n    model = Model(HiGHS.Optimizer)\n    x = add_knapsack_variables(model, data, config)\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    @constraint(\n        model,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    optimize!(model)\n    return value.(x)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For the binary knapsack problem, add_knapsack_variables looks like this:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    ::BinaryKnapsackConfig,\n)\n    return @variable(model, x[keys(data.objects)], Bin)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"For the integer knapsack problem, add_knapsack_variables looks like this:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    ::IntegerKnapsackConfig,\n)\n    return @variable(model, x[keys(data.objects)] >= 0, Int)\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we can solve the binary knapsack:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_4(data, BinaryKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"and the integer knapsack problem:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"solve_knapsack_4(data, IntegerKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"The main benefit of the dispatch approach is that you can quickly add new options without needing to modify the existing code. For example:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"struct UpperBoundedKnapsackConfig <: AbstractConfiguration\n    limit::Int\nend\n\nfunction add_knapsack_variables(\n    model::Model,\n    data::KnapsackData,\n    config::UpperBoundedKnapsackConfig,\n)\n    return @variable(model, 0 <= x[keys(data.objects)] <= config.limit, Int)\nend\n\nsolve_knapsack_4(data, UpperBoundedKnapsackConfig(3))","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Generalize-constraints-and-objectives","page":"Design patterns for larger models","title":"Generalize constraints and objectives","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"It's easy to extend the dispatch approach to constraints and objectives as well. The key points to notice in the next two functions are that:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"we can access registered variables via model[:x]\nwe can define generic functions which accept any AbstractConfiguration as a configuration argument. That means we can implement a single method and have it apply to multiple configuration types.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function add_knapsack_constraints(\n    model::Model,\n    data::KnapsackData,\n    ::AbstractConfiguration,\n)\n    x = model[:x]\n    @constraint(\n        model,\n        capacity_constraint,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    return\nend\n\nfunction add_knapsack_objective(\n    model::Model,\n    data::KnapsackData,\n    ::AbstractConfiguration,\n)\n    x = model[:x]\n    @objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    return\nend\n\nfunction solve_knapsack_5(data::KnapsackData, config::AbstractConfiguration)\n    model = Model(HiGHS.Optimizer)\n    add_knapsack_variables(model, data, config)\n    add_knapsack_constraints(model, data, config)\n    add_knapsack_objective(model, data, config)\n    optimize!(model)\n    return value.(model[:x])\nend\n\nsolve_knapsack_5(data, BinaryKnapsackConfig())","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Remove-solver-dependence,-add-error-checks","page":"Design patterns for larger models","title":"Remove solver dependence, add error checks","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Compared to where we started, our knapsack model is now significantly different. We've wrapped it in a function, defined some data types, and introduced configuration options to control the variables and constraints that get added. There are a few other steps we can do to further improve things:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"remove the dependence on HiGHS\nadd checks that we found an optimal solution\nadd a helper function to avoid the need to explicitly construct the data.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"function solve_knapsack_6(\n    optimizer,\n    data::KnapsackData,\n    config::AbstractConfiguration,\n)\n    model = Model(optimizer)\n    add_knapsack_variables(model, data, config)\n    add_knapsack_constraints(model, data, config)\n    add_knapsack_objective(model, data, config)\n    optimize!(model)\n    if termination_status(model) != OPTIMAL\n        @warn(\"Model not solved to optimality\")\n        return nothing\n    end\n    return value.(model[:x])\nend\n\nfunction solve_knapsack_6(\n    optimizer,\n    data::String,\n    config::AbstractConfiguration,\n)\n    return solve_knapsack_6(optimizer, read_data(data), config)\nend\n\nsolution = solve_knapsack_6(\n    HiGHS.Optimizer,\n    knapsack_json_filename,\n    BinaryKnapsackConfig(),\n)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Create-a-module","page":"Design patterns for larger models","title":"Create a module","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Now we're ready to expose our model to the wider world. That might be as part of a larger Julia project that we're contributing to, or as a stand-alone script that we can run on-demand. In either case, it's good practice to wrap everything in a module. This further encapsulates our code into a single namespace, and we can add documentation in the form of docstrings.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Some good rules to follow when creating a module are:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"use import in a module instead of using to make it clear which functions are from which packages\nuse _ to start function and type names that are considered private\nadd docstrings to all public variables and functions.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"module KnapsackModel\n\nimport JuMP\nimport JSON\n\nstruct _KnapsackObject\n    profit::Float64\n    weight::Float64\n    function _KnapsackObject(profit::Float64, weight::Float64)\n        if weight < 0\n            throw(DomainError(\"Weight of object cannot be negative\"))\n        end\n        return new(profit, weight)\n    end\nend\n\nstruct _KnapsackData\n    objects::Dict{String,_KnapsackObject}\n    capacity::Float64\nend\n\nfunction _read_data(filename)\n    d = JSON.parsefile(filename)\n    return _KnapsackData(\n        Dict(\n            k => _KnapsackObject(v[\"profit\"], v[\"weight\"]) for\n            (k, v) in d[\"objects\"]\n        ),\n        d[\"capacity\"],\n    )\nend\n\nabstract type _AbstractConfiguration end\n\n\"\"\"\n    BinaryKnapsackConfig()\n\nCreate a binary knapsack problem where each object can be taken 0 or 1 times.\n\"\"\"\nstruct BinaryKnapsackConfig <: _AbstractConfiguration end\n\n\"\"\"\n    IntegerKnapsackConfig()\n\nCreate an integer knapsack problem where each object can be taken any number of\ntimes.\n\"\"\"\nstruct IntegerKnapsackConfig <: _AbstractConfiguration end\n\nfunction _add_knapsack_variables(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::BinaryKnapsackConfig,\n)\n    return JuMP.@variable(model, x[keys(data.objects)], Bin)\nend\n\nfunction _add_knapsack_variables(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::IntegerKnapsackConfig,\n)\n    return JuMP.@variable(model, x[keys(data.objects)] >= 0, Int)\nend\n\nfunction _add_knapsack_constraints(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::_AbstractConfiguration,\n)\n    x = model[:x]\n    JuMP.@constraint(\n        model,\n        capacity_constraint,\n        sum(v.weight * x[k] for (k, v) in data.objects) <= data.capacity,\n    )\n    return\nend\n\nfunction _add_knapsack_objective(\n    model::JuMP.Model,\n    data::_KnapsackData,\n    ::_AbstractConfiguration,\n)\n    x = model[:x]\n    JuMP.@objective(model, Max, sum(v.profit * x[k] for (k, v) in data.objects))\n    return\nend\n\nfunction _solve_knapsack(\n    optimizer,\n    data::_KnapsackData,\n    config::_AbstractConfiguration,\n)\n    model = JuMP.Model(optimizer)\n    _add_knapsack_variables(model, data, config)\n    _add_knapsack_constraints(model, data, config)\n    _add_knapsack_objective(model, data, config)\n    JuMP.optimize!(model)\n    if JuMP.termination_status(model) != JuMP.OPTIMAL\n        @warn(\"Model not solved to optimality\")\n        return nothing\n    end\n    return JuMP.value.(model[:x])\nend\n\n\"\"\"\n    solve_knapsack(\n        optimizer,\n        knapsack_json_filename::String,\n        config::_AbstractConfiguration,\n    )\n\nSolve the knapsack problem and return the optimal primal solution\n\n# Arguments\n\n * `optimizer` : an object that can be passed to `JuMP.Model` to construct a new\n   JuMP model.\n * `knapsack_json_filename` : the filename of a JSON file containing the data for the\n   problem.\n * `config` : an object to control the type of knapsack model constructed.\n   Valid options are:\n    * `BinaryKnapsackConfig()`\n    * `IntegerKnapsackConfig()`\n\n# Returns\n\n * If an optimal solution exists: a `JuMP.DenseAxisArray` that maps the `String`\n   name of each object to the number of objects to pack into the knapsack.\n * Otherwise, `nothing`, indicating that the problem does not have an optimal\n   solution.\n\n# Examples\n\n```julia\nsolution = solve_knapsack(\n    HiGHS.Optimizer,\n    \"path/to/data.json\",\n    BinaryKnapsackConfig(),\n)\n```\n\n```julia\nsolution = solve_knapsack(\n    MOI.OptimizerWithAttributes(HiGHS.Optimizer, \"output_flag\" => false),\n    \"path/to/data.json\",\n    IntegerKnapsackConfig(),\n)\n```\n\"\"\"\nfunction solve_knapsack(\n    optimizer,\n    knapsack_json_filename::String,\n    config::_AbstractConfiguration,\n)\n    data = _read_data(knapsack_json_filename)\n    return _solve_knapsack(optimizer, data, config)\nend\n\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Finally, you can call your model:","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import .KnapsackModel\n\nKnapsackModel.solve_knapsack(\n    HiGHS.Optimizer,\n    knapsack_json_filename,\n    KnapsackModel.BinaryKnapsackConfig(),\n)","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"note: Note\nThe . in .KnapsackModel denotes that it is a submodule and not a separate package that we installed with Pkg.add. If you put the KnapsackModel in a separate file, load it with:include(\"path/to/KnapsackModel.jl\")\nimport .KnapsackModel","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Add-tests","page":"Design patterns for larger models","title":"Add tests","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"As a final step, you should add tests for your model. This often means testing on a small problem for which you can work out the optimal solution by hand. The Julia standard library Test has good unit-testing functionality.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"import .KnapsackModel\nusing Test\n\n@testset \"KnapsackModel\" begin\n    @testset \"feasible_binary_knapsack\" begin\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            knapsack_json_filename,\n            KnapsackModel.BinaryKnapsackConfig(),\n        )\n        @test isapprox(x[\"apple\"], 1, atol = 1e-5)\n        @test isapprox(x[\"banana\"], 0, atol = 1e-5)\n        @test isapprox(x[\"cherry\"], 0, atol = 1e-5)\n        @test isapprox(x[\"date\"], 1, atol = 1e-5)\n        @test isapprox(x[\"eggplant\"], 1, atol = 1e-5)\n    end\n    @testset \"feasible_integer_knapsack\" begin\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            knapsack_json_filename,\n            KnapsackModel.IntegerKnapsackConfig(),\n        )\n        @test isapprox(x[\"apple\"], 0, atol = 1e-5)\n        @test isapprox(x[\"banana\"], 0, atol = 1e-5)\n        @test isapprox(x[\"cherry\"], 0, atol = 1e-5)\n        @test isapprox(x[\"date\"], 5, atol = 1e-5)\n        @test isapprox(x[\"eggplant\"], 0, atol = 1e-5)\n    end\n    @testset \"infeasible_binary_knapsack\" begin\n        dir = mktempdir()\n        infeasible_filename = joinpath(dir, \"infeasible.json\")\n        write(\n            infeasible_filename,\n            \"\"\"{\n                \"objects\": {\n                    \"apple\": {\"profit\": 5.0, \"weight\": 2.0},\n                    \"banana\": {\"profit\": 3.0, \"weight\": 8.0},\n                    \"cherry\": {\"profit\": 2.0, \"weight\": 4.0},\n                    \"date\": {\"profit\": 7.0, \"weight\": 2.0},\n                    \"eggplant\": {\"profit\": 4.0, \"weight\": 5.0}\n                },\n                \"capacity\": -10.0\n            }\"\"\",\n        )\n        x = KnapsackModel.solve_knapsack(\n            HiGHS.Optimizer,\n            infeasible_filename,\n            KnapsackModel.BinaryKnapsackConfig(),\n        )\n        @test x === nothing\n    end\nend","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"tip: Tip\nPlace these tests in a separate file test_knapsack_model.jl so that you can run the tests by adding include(\"test_knapsack_model.jl\") to any file where needed.","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/#Next-steps","page":"Design patterns for larger models","title":"Next steps","text":"","category":"section"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"We've only briefly scratched the surface of ways to create and structure large JuMP models, so consider this tutorial a starting point, rather than a comprehensive list of all the possible ways to structure JuMP models.  If you are embarking on a large project that uses JuMP, a good next step is to look at ways people have written large JuMP projects \"in the wild.\"","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"Here are some good examples (all co-incidentally related to energy):","category":"page"},{"location":"tutorials/getting_started/design_patterns_for_larger_models/","page":"Design patterns for larger models","title":"Design patterns for larger models","text":"AnyMOD.jl\nJuMP-dev 2021 talk\nsource code\nPowerModels.jl\nJuMP-dev 2021 talk\nsource code\nPowerSimulations.jl\nJuliaCon 2021 talk\nsource code\nUnitCommitment.jl\nJuMP-dev 2021 talk\nsource code","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"EditURL = \"multi.jl\"","category":"page"},{"location":"tutorials/linear/multi/#The-multi-commodity-flow-problem","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial is a JuMP implementation of the multi-commodity transportation model described in AMPL: A Modeling Language for Mathematical Programming, by R. Fourer, D.M. Gay and B.W. Kernighan.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate creating a JuMP model from an SQLite database.","category":"page"},{"location":"tutorials/linear/multi/#Required-packages","page":"The multi-commodity flow problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport SQLite\nimport Tables\n\nconst DBInterface = SQLite.DBInterface","category":"page"},{"location":"tutorials/linear/multi/#Formulation","page":"The multi-commodity flow problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The multi-commondity flow problem is a simple extension of The transportation problem to multiple types of products. Briefly, we start with the formulation of the transportation problem:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"beginaligned\nmin   sum_i in O j in D c_ij x_ij \nst   sum_j in D x_i j le s_i  forall i in O \n       sum_i in O x_i j = d_j  forall j in D \n       x_i j ge 0  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"but introduce a set of products P, resulting in:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"beginaligned\nmin   sum_i in O j in D k in P c_ijk x_ijk \nst   sum_j in D x_i j k le s_ik  forall i in O k in P \n       sum_i in O x_i j k = d_jk  forall j in D k in P \n       x_i jk ge 0  forall i in O j in D k in P \n       sum_k in P x_i j k le u_ij  forall i in O j in D\nendaligned","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Note that the last constraint is new; it says that there is a maximum quantity of goods (of any type) that can be transported from origin i to destination j.","category":"page"},{"location":"tutorials/linear/multi/#Data","page":"The multi-commodity flow problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"For the purpose of this tutorial, the JuMP repository contains an example database called multi.sqlite.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"filename = joinpath(@__DIR__, \"multi.sqlite\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"To run locally, download multi.sqlite and update filename appropriately.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Load the database using SQLite.DB:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"db = SQLite.DB(filename)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"A quick way to see the schema of the database is via SQLite.tables:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"SQLite.tables(db)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We interact with the database by executing queries, and then piping the results to an appropriate table. One example is a DataFrame:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"DBInterface.execute(db, \"SELECT * FROM locations\") |> DataFrames.DataFrame","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"But other table types are supported, such as Tables.rowtable:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"DBInterface.execute(db, \"SELECT * FROM locations\") |> Tables.rowtable","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"A rowtable is a Vector of NamedTuples.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"You can construct more complicated SQL queries:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"origins =\n    DBInterface.execute(\n        db,\n        \"SELECT location FROM locations WHERE type = \\\"origin\\\"\",\n    ) |> Tables.rowtable","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"But for our purpose, we just want the list of strings:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"origins = map(y -> y.location, origins)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We can compose these two operations to get a list of destinations:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"destinations =\n    DBInterface.execute(\n        db,\n        \"SELECT location FROM locations WHERE type = \\\"destination\\\"\",\n    ) |>\n    Tables.rowtable |>\n    x -> map(y -> y.location, x)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"And a list of products from our products table:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"products =\n    DBInterface.execute(db, \"SELECT product FROM products\") |>\n    Tables.rowtable |>\n    x -> map(y -> y.product, x)","category":"page"},{"location":"tutorials/linear/multi/#JuMP-formulation","page":"The multi-commodity flow problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"We start by creating a model and our decision variables:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[origins, destinations, products] >= 0)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"One approach when working with databases is to extract all of the data into a Julia datastructure. For example, let's pull the cost table into a DataFrame and then construct our objective by iterating over the rows of the DataFrame:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"cost = DBInterface.execute(db, \"SELECT * FROM cost\") |> DataFrames.DataFrame\n@objective(\n    model,\n    Max,\n    sum(r.cost * x[r.origin, r.destination, r.product] for r in eachrow(cost)),\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"If we don't want to use a DataFrame, we can use a Tables.rowtable instead:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"supply = DBInterface.execute(db, \"SELECT * FROM supply\") |> Tables.rowtable\nfor r in supply\n    @constraint(model, sum(x[r.origin, :, r.product]) <= r.supply)\nend","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Another approach is to execute the query, and then to iterate through the rows of the query using Tables.rows:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"demand = DBInterface.execute(db, \"SELECT * FROM demand\")\nfor r in Tables.rows(demand)\n    @constraint(model, sum(x[:, r.destination, r.product]) == r.demand)\nend","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"warning: Warning\nIterating through the rows of a query result works by incrementing a cursor inside the database. As a consequence, you cannot call Tables.rows twice on the same query result.","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"The SQLite queries can be arbitrarily complex. For example, here's a query which builds every possible origin-destination pair:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"od_pairs = DBInterface.execute(\n    db,\n    \"\"\"\n    SELECT a.location as 'origin',\n           b.location as 'destination'\n    FROM locations a\n    INNER JOIN locations b\n    ON a.type = 'origin' AND b.type = 'destination'\n    \"\"\",\n)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"With a constraint that we cannot send more than 625 units between each pair:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"for r in Tables.rows(od_pairs)\n    @constraint(model, sum(x[r.origin, r.destination, :]) <= 625)\nend","category":"page"},{"location":"tutorials/linear/multi/#Solution","page":"The multi-commodity flow problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"Finally, we can optimize the model:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"and print the solution:","category":"page"},{"location":"tutorials/linear/multi/","page":"The multi-commodity flow problem","title":"The multi-commodity flow problem","text":"begin\n    println(\"         \", join(products, ' '))\n    for o in origins, d in destinations\n        v = lpad.([round(Int, value(x[o, d, p])) for p in products], 5)\n        println(o, \" \", d, \" \", join(replace.(v, \"   0\" => \"  . \"), \" \"))\n    end\nend","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"EditURL = \"finance.jl\"","category":"page"},{"location":"tutorials/linear/finance/#Financial-modeling-problems","page":"Financial modeling problems","title":"Financial modeling problems","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Optimization models play an increasingly important role in financial decisions. Many computational finance problems can be solved efficiently using modern optimization techniques.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"In this tutorial we will discuss two such examples taken from the book Optimization Methods in Finance.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/finance/#Short-term-financing","page":"Financial modeling problems","title":"Short-term financing","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Corporations routinely face the problem of financing short term cash commitments such as the following:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Month Jan Feb Mar Apr May Jun\nNet Cash Flow -150 -100 200 -200 50 300","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Net cash flow requirements are given in thousands of dollars. The company has the following sources of funds:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"A line of credit of up to $100K at an interest rate of 1% per month,\nIn any one of the first three months, it can issue 90-day commercial paper bearing a total interest of 2% for the 3-month period,\nExcess funds can be invested at an interest rate of 0.3% per month.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Our task is to find out the most economical way to use these 3 sources such that we end up with the most amount of money at the end of June.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We model this problem in the following manner:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We will use the following decision variables:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"the amount u_i drawn from the line of credit in month i\nthe amount v_i of commercial paper issued in month i\nthe excess funds w_i in month i","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Here we have three types of constraints:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"for every month, cash inflow = cash outflow for each month\nupper bounds on u_i\nnonnegativity of the decision variables u_i, v_i and w_i.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Our objective will be to simply maximize the company's wealth in June, which say we represent with the variable m.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"financing = Model(HiGHS.Optimizer)\n\n@variables(financing, begin\n    0 <= u[1:5] <= 100\n    0 <= v[1:3]\n    0 <= w[1:5]\n    m\nend)\n\n@objective(financing, Max, m)\n\n@constraints(\n    financing,\n    begin\n        u[1] + v[1] - w[1] == 150 # January\n        u[2] + v[2] - w[2] - 1.01u[1] + 1.003w[1] == 100 # February\n        u[3] + v[3] - w[3] - 1.01u[2] + 1.003w[2] == -200 # March\n        u[4] - w[4] - 1.02v[1] - 1.01u[3] + 1.003w[3] == 200 # April\n        u[5] - w[5] - 1.02v[2] - 1.01u[4] + 1.003w[4] == -50 # May\n        -m - 1.02v[3] - 1.01u[5] + 1.003w[5] == -300 # June\n    end\n)\n\noptimize!(financing)\n\nobjective_value(financing)","category":"page"},{"location":"tutorials/linear/finance/#Combinatorial-auctions","page":"Financial modeling problems","title":"Combinatorial auctions","text":"","category":"section"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"In many auctions, the value that a bidder has for a set of items may not be the sum of the values that he has for individual items.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Examples are equity trading, electricity markets, pollution right auctions and auctions for airport landing slots.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"To take this into account, combinatorial auctions allow the bidders to submit bids on combinations of items.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Let M=12 ldots m be the set of items that the auctioneer has to sell. A bid is a pair B_j=left(S_j p_jright) where S_j subseteq M is a nonempty set of items and p_j is the price offer for this set.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"Suppose that the auctioneer has received n bids B_1 B_2 ldots B_n The goal of this problem is to help an auctioneer determine the winners in order to maximize his revenue.","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"We model this problem by taking a decision variable y_j for every bid. We add a constraint that each item i is sold at most once. This gives us the following model:","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"beginaligned\nmax  sum_i=1^n p_j y_j \ntext  st    sum_j  i in S_j y_j leq 1  forall i=12 ldots m \n y_j in01  forall j in12 ldots n\nendaligned","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"bid_values = [6 3 12 12 8 16]\nbid_items = [[1], [2], [3 4], [1 3], [2 4], [1 3 4]]\n\nauction = Model(HiGHS.Optimizer)\n@variable(auction, y[1:6], Bin)\n@objective(auction, Max, sum(y' .* bid_values))\nfor i in 1:6\n    @constraint(auction, sum(y[j] for j in 1:6 if i in bid_items[j]) <= 1)\nend\n\noptimize!(auction)\n\nobjective_value(auction)","category":"page"},{"location":"tutorials/linear/finance/","page":"Financial modeling problems","title":"Financial modeling problems","text":"value.(y)","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"EditURL = \"https://github.com/GAMS-dev/GAMS.jl/blob/c5dee9f929e9d2f4433ae09fa92b8d872c9c43e0/README.md\"","category":"page"},{"location":"packages/GAMS/#GAMS.jl","page":"GAMS-dev/GAMS.jl","title":"GAMS.jl","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl provides a MathOptInterface Optimizer to solve JuMP models using GAMS.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS comes with dozens of supported solvers. Among them are: ALPHAECP, ANTIGONE, BARON, CBC, CONOPT, CPLEX, DICOPT, GUROBI, IPOPT, KNITRO, LINDO, LINDOGLOBAL, MINOS, MOSEK, NLPEC, PATH, QUADMINOS, SBB, SHOT, SCIP, SNOPT, SOPLEX, XPRESS. Find a complete list here.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl supports the following JuMP features:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"linear, quadratic and nonlinear (convex and non-convex) objective and constraints\ncontinuous, binary, integer, semi-continuous and semi-integer variables\nSOS1 and SOS2 sets\ncomplementarity constraints","category":"page"},{"location":"packages/GAMS/#Installation","page":"GAMS-dev/GAMS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Download GAMS and obtain a GAMS license. Please note that GAMS also offers a free community license.\n(optional) Add the GAMS system directory to the PATH variable in order to find GAMS automatically.\nInstall GAMS.jl using the Julia package manager:\nusing Pkg\nPkg.add(\"GAMS\")","category":"page"},{"location":"packages/GAMS/#Usage","page":"GAMS-dev/GAMS.jl","title":"Usage","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Using GAMS as optimizer for your JuMP model:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"using GAMS, JuMP\nmodel = Model(GAMS.Optimizer)","category":"page"},{"location":"packages/GAMS/#GAMS-System","page":"GAMS-dev/GAMS.jl","title":"GAMS System","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"If the GAMS system directory has been added to the PATH variable (you can check this with print(ENV[\"PATH\"])), GAMS.jl will find it automatically. Otherwise, or if you like to switch between systems, the system directory can be specified by (one of the following):","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"SysDir\", \"<gams_system_dir>\")\nset_optimizer_attribute(model, GAMS.SysDir(), \"<gams_system_dir>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Analogously, you can specify a working directory with \"WorkDir\" or GAMS.WorkDir(). If no working directory has been set, GAMS.jl will create a temporary one.","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"If you want to use the same GAMS workspace (same system and working directory) for multiple models, you can create a GAMSWorkspace first with either of the following","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"ws = GAMS.GAMSWorkspace()\nws = GAMS.GAMSWorkspace(\"<gams_system_dir>\")\nws = GAMS.GAMSWorkspace(\"<gams_system_dir>\", \"<gams_working_dir>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"and then pass it to your models:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"model = Model(() -> GAMS.Optimizer(ws))","category":"page"},{"location":"packages/GAMS/#GAMS-Options","page":"GAMS-dev/GAMS.jl","title":"GAMS Options","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS command line options can be specified by","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"<option>\", \"<solver_name>\")\nset_optimizer_attribute(model, GAMS.<option>(), \"<solver_name>\")","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"where <option> is either HoldFixed, IterLim, License, LogOption, NodLim, OptCA, OptCR, ResLim, Solver, Threads, Trace, TraceOpt as well as LP, MIP, RMIP, NLP, DNLP, CNS, MINLP, RMINLP, QCP, MIQCP, RMIQCP, MCP or MPEC. Note that GAMS.ResLim() is equivalent to MOI.TimeLimitSec() and GAMS.Threads() to MOI.NumberOfThreads(). Options LimCol, LimRow, SolPrint and SolveLink cannot be changed and are set to 0, 0, 0 and 5, respectively.","category":"page"},{"location":"packages/GAMS/#Model-Type","page":"GAMS-dev/GAMS.jl","title":"Model Type","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS.jl will automatically choose a GAMS model type for you. Choosing a different model type:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, GAMS.ModelType(), \"<model_type>\")","category":"page"},{"location":"packages/GAMS/#GAMS-Solver-Options","page":"GAMS-dev/GAMS.jl","title":"GAMS Solver Options","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Specifying GAMS solver options:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"set_optimizer_attribute(model, \"<solver_option_name>\", <option_value>)","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Note that passing a solver option is only valid when explicitly choosing a GAMS solver and not using the default.","category":"page"},{"location":"packages/GAMS/#GAMS-Names-vs.-JuMP-Names","page":"GAMS-dev/GAMS.jl","title":"GAMS Names vs. JuMP Names","text":"","category":"section"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"GAMS uses generated variable and constraint names although it is possible to pass the JuMP names to the GAMS optimizer, because GAMS is more restrictive when it comes to variable and constraint naming. Use the attributes GeneratedVariableName, GeneratedConstraintName, OriginalVariableName, OriginalConstraintName to query a GAMS symbol name from a JuMP symbol and vice versa. This can help for debugging, for example in case of GAMS compilation errors. For example:","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"using GAMS\n\nmodel = direct_model(GAMS.Optimizer())\n\n@variable(model, x[1:2,1:3] >= 0)\n@constraint(model, c[i = 1:2], sum(x[i,j] for j = 1:3) <= 10)\n\nMOI.get(model, GAMS.GeneratedVariableName(), x[2,2]) # returns x4\nMOI.get(model, GAMS.OriginalVariableName(\"x6\"))      # returns x[2,3]\nMOI.get(model, GAMS.OriginalVariableName(\"x10\"))     # returns nothing\n\nMOI.get(model, GAMS.GeneratedConstraintName(), c[2]) # returns eq2\nMOI.get(model, GAMS.OriginalConstraintName(\"eq1\"))   # returns c[1]\nMOI.get(model, GAMS.OriginalConstraintName(\"eq10\"))  # returns nothing","category":"page"},{"location":"packages/GAMS/","page":"GAMS-dev/GAMS.jl","title":"GAMS-dev/GAMS.jl","text":"Note that JuMP direct-mode is used.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"EditURL = \"https://github.com/jump-dev/Pavito.jl/blob/v0.3.7/README.md\"","category":"page"},{"location":"packages/Pavito/#Pavito.jl","page":"jump-dev/Pavito.jl","title":"Pavito.jl","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"(Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito.jl is a mixed-integer convex programming (MICP) solver package written in Julia.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"MICP problems are convex, except for restrictions that some variables take binary or integer values.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito solves MICP problems by constructing sequential polyhedral outer-approximations of the convex feasible set, similar to Bonmin.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito accesses state-of-the-art MILP solvers and continuous, derivative-based nonlinear programming (NLP) solvers through MathOptInterface.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For algorithms that use a conic solver instead of an NLP solver, use Pajarito. Pajarito is a robust mixed-integer conic solver that can handle such established problem classes as mixed-integer second-order cone programming (MISOCP) and mixed-integer semidefinite programming (MISDP).","category":"page"},{"location":"packages/Pavito/#License","page":"jump-dev/Pavito.jl","title":"License","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Pavito/#Installation","page":"jump-dev/Pavito.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Install Pavito using Pkg.add:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"import Pkg\nPkg.add(\"Pavito\")","category":"page"},{"location":"packages/Pavito/#Use-with-JuMP","page":"jump-dev/Pavito.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"To use Pavito with JuMP, use Pavito.Optimizer:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"using JuMP, Pavito\nimport GLPK, Ipopt\nmodel = Model(\n    optimizer_with_attributes(\n        Pavito.Optimizer,\n        \"mip_solver\" => optimizer_with_attributes(GLPK.Optimizer),\n        \"cont_solver\" =>\n            optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0),\n    ),\n)","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The algorithm implemented by Pavito itself is relatively simple; most of the hard work is performed by the MILP solver passed as mip_solver and the NLP solver passed as cont_solver.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The performance of Pavito depends on these two types of solvers.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For better performance, you should use a commercial MILP solver such as CPLEX or Gurobi.","category":"page"},{"location":"packages/Pavito/#Options","page":"jump-dev/Pavito.jl","title":"Options","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"The following optimizer attributes can set to a Pavito.Optimizer to modify its behavior:","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"log_level::Int Verbosity flag: 0 for quiet, higher for basic solve info\ntimeout::Float64 Time limit for algorithm (in seconds)\nrel_gap::Float64 Relative optimality gap termination condition\nmip_solver_drives::Bool Let MILP solver manage convergence (\"branch and cut\")\nmip_solver::MOI.OptimizerWithAttributes MILP solver\ncont_solver::MOI.OptimizerWithAttributes Continuous NLP solver","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Pavito is not yet numerically robust and may require tuning of parameters to improve convergence.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"If the default parameters don't work for you, please let us know by opening an issue.","category":"page"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"For improved Pavito performance, MILP solver integrality tolerance and feasibility tolerances should typically be tightened, for example to 1e-8.","category":"page"},{"location":"packages/Pavito/#Bug-reports-and-support","page":"jump-dev/Pavito.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Pavito/","page":"jump-dev/Pavito.jl","title":"jump-dev/Pavito.jl","text":"Please report any issues via the GitHub issue tracker. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. The Optimization (Mathematical) category on Discourse is appropriate for general discussion.","category":"page"},{"location":"developers/contributing/#How-to-contribute-to-JuMP","page":"Contributing","title":"How to contribute to JuMP","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Welcome, this document explains some ways you can contribute to JuMP.","category":"page"},{"location":"developers/contributing/#Code-of-Conduct","page":"Contributing","title":"Code of Conduct","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"This project and everyone participating in it is governed by the JuMP Code of Conduct. By participating, you are expected to uphold this code.","category":"page"},{"location":"developers/contributing/#Join-the-community-forum","page":"Contributing","title":"Join the community forum","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"First up, join the community forum.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The forum is a good place to ask questions about how to use JuMP. You can also use the forum to discuss possible feature requests and bugs before raising a GitHub issue (more on this below).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Aside from asking questions, the easiest way you can contribute to JuMP is to help answer questions on the forum.","category":"page"},{"location":"developers/contributing/#Join-the-developer-chatroom","page":"Contributing","title":"Join the developer chatroom","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If you're interested in contributing code to JuMP, the next place to join is the developer chatroom. Let us know what you have in mind, and we can point you in the right direction.","category":"page"},{"location":"developers/contributing/#Improve-the-documentation","page":"Contributing","title":"Improve the documentation","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Chances are, if you asked (or answered) a question on the community forum, then it is a sign that the documentation could be improved. Moreover, since it is your question, you are probably the best-placed person to improve it.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The docs are written in Markdown and are built using Documenter.jl. You can find the source of all the docs here.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If your change is small (like fixing typos, or one or two sentence corrections), the easiest way to do this is via GitHub's online editor. (GitHub has help on how to do this.)","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"If your change is larger, or touches multiple files, you will need to make the change locally and then use Git to submit a pull request. (See Contribute code to JuMP below for more on this.)","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you need any help, come join the developer chatroom and we will walk you through the process.","category":"page"},{"location":"developers/contributing/#File-a-bug-report","page":"Contributing","title":"File a bug report","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Another way to contribute to JuMP is to file bug reports.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Make sure you read the info in the box where you write the body of the issue before posting. You can also find a copy of that info here.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you're unsure whether you have a real bug, post on the community forum first. Someone will either help you fix the problem, or let you know the most appropriate place to open a bug report.","category":"page"},{"location":"developers/contributing/#Contribute-code-to-JuMP","page":"Contributing","title":"Contribute code to JuMP","text":"","category":"section"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Finally, you can also contribute code to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nIf you do not have experience with Git, GitHub, and Julia development, the first steps can be a little daunting. However, there are lots of tutorials available online, including these for:GitHub\nGit and GitHub\nGit\nJulia package developmentIf you need any help, come join the developer chatroom and we will walk you through the process.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once you are familiar with Git and GitHub, the workflow for contributing code to JuMP is similar to the following:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 1: decide what to work on","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The first step is to find an open issue (or open a new one) for the problem you want to solve. Then, before spending too much time on it, discuss what you are planning to do in the issue to see if other contributors are fine with your proposed changes. Getting feedback early can improve code quality, and avoid time spent writing code that does not get merged into JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nAt this point, remember to be patient and polite; you may get a lot of comments on your issue. However, do not be afraid. Comments mean that people are willing to help you improve the code that you are contributing to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 2: fork JuMP","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Go to https://github.com/jump-dev/JuMP.jl and click the \"Fork\" button in the top-right corner. This will create a copy of JuMP under your GitHub account.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 3: install JuMP locally","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Open Julia and run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"] dev JuMP","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"This will download the JuMP Git repository to ~/.julia/dev/JuMP. If you're on Windows, this will be C:\\\\Users\\\\<my_name>\\\\.julia\\\\dev\\\\JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\n] command means \"first type ] to enter the Julia pkg mode, then type the rest. Don't copy-paste the code directly.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 4: checkout a new branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"note: Note\nIn the following, replace any instance of GITHUB_ACCOUNT with your GitHub user name.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"The next step is to checkout a development branch. In a terminal (or command prompt on Windows), run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git remote add GITHUB_ACCOUNT https://github.com/GITHUB_ACCOUNT/JuMP.jl.git\n\n$ git checkout master\n\n$ git pull\n\n$ git checkout -b my_new_branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nLines starting with $ mean \"run these in a terminal (command prompt on Windows).\"","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 5: make changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Now make any changes to the source code inside the ~/.julia/dev/JuMP directory.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Make sure you:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Follow the Style guide and run JuliaFormatter\nAdd tests and documentation for any changes or new features","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nWhen you change the source code, you'll need to restart Julia for the changes to take effect. This is a pain, so install Revise.jl.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 6a: test your code changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"To test that your changes work, run the JuMP test-suite by opening Julia and running:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"cd(\"~/.julia/dev/JuMP\")\n] activate .\n] test","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nRunning the tests might take a long time (~10–15 minutes).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf you're using Revise.jl, you can also run the tests by calling include:include(\"test/runtests.jl\")This can be faster if you want to re-run the tests multiple times.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 6b: test your documentation changes","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Open Julia, then run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"cd(\"~/.julia/dev/JuMP/docs\")\n] activate .\ninclude(\"src/make.jl\")","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"warning: Warning\nBuilding the documentation might take a long time (~10 minutes).","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"tip: Tip\nIf there's a problem with the tests that you don't know how to fix, don't worry. Continue to step 5, and one of the JuMP contributors will comment on your pull request telling you how to fix things.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 7: make a pull request","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once you've made changes, you're ready to push the changes to GitHub. Run:","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git add .\n\n$ git commit -m \"A descriptive message of the changes\"\n\n$ git push -u GITHUB_ACCOUNT my_new_branch","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Then go to https://github.com/jump-dev/JuMP.jl and follow the instructions that pop up to open a pull request.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 8: respond to comments","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"At this point, remember to be patient and polite; you may get a lot of comments on your pull request. However, do not be afraid. A lot of comments means that people are willing to help you improve the code that you are contributing to JuMP.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"To respond to the comments, go back to step 5, make any changes, test the changes in step 6, and then make a new commit in step 7. Your PR will automatically update.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Step 9: cleaning up","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"Once the PR is merged, clean-up your Git repository ready for the next contribution.","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"$ cd ~/.julia/dev/JuMP\n\n$ git checkout master\n\n$ git pull","category":"page"},{"location":"developers/contributing/","page":"Contributing","title":"Contributing","text":"note: Note\nIf you have suggestions to improve this guide, please make a pull request. It's particularly helpful if you do this after your first pull request because you'll know all the parts that could be explained better.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"EditURL = \"knapsack.jl\"","category":"page"},{"location":"tutorials/linear/knapsack/#The-knapsack-problem-example","page":"The knapsack problem example","title":"The knapsack problem example","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The purpose of this tutorial is to demonstrate how to formulate and solve a simple optimization problem.","category":"page"},{"location":"tutorials/linear/knapsack/#Required-packages","page":"The knapsack problem example","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/knapsack/#Formulation","page":"The knapsack problem example","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The knapsack problem is a classical optimization problem: given a set of items and a container with a fixed capacity, choose a subset of items having the greatest combined value that will fit within the container without exceeding the capacity.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The name of the problem suggests its analogy to packing for a trip, where the baggage weight limit is the capacity and the goal is to pack the most profitable combination of belongings.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We can formulate the knapsack problem as the integer linear program:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"beginaligned\nmax   sum_i=1^n c_i x_i      \nst   sum_i=1^n w_i x_i le C \n         x_i in 01quad forall i=1ldotsn\nendaligned","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"where C is the capacity, and there is a choice between n items, with item i having weight w_i, profit c_i. Decision variable x_i is equal to 1 if the item is chosen and 0 if not.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"This formulation can be written more compactly as:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"beginaligned\nmax   c^top x       \nst   w^top x le C \n         x text binary \nendaligned","category":"page"},{"location":"tutorials/linear/knapsack/#Data","page":"The knapsack problem example","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The data for the problem consists of two vectors (one for the profits and one for the weights) along with a capacity.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"There are five objects:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"n = 5;\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"For our example, we use a capacity of 10 units:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"capacity = 10.0;\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"and the profit and cost data:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"profit = [5.0, 3.0, 2.0, 7.0, 4.0];\nweight = [2.0, 8.0, 4.0, 2.0, 5.0];\nnothing #hide","category":"page"},{"location":"tutorials/linear/knapsack/#JuMP-formulation","page":"The knapsack problem example","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Let's begin constructing the JuMP model for our knapsack problem.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"First, we'll create a Model object for holding model elements as we construct each part. We'll also set the solver that will ultimately be called to solve the model, once it's constructed.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Next we need the decision variables representing which items are chosen:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@variable(model, x[1:n], Bin)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We now want to constrain those variables so that their combined weight is less than or equal to the given capacity:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@constraint(model, sum(weight[i] * x[i] for i in 1:n) <= capacity)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Finally, our objective is to maximize the combined profit of the chosen items:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"@objective(model, Max, sum(profit[i] * x[i] for i in 1:n))","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Let's print a human-readable description of the model and check that the model looks as expected:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"print(model)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We can now solve the optimization problem and inspect the results.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The items chosen are","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"items_chosen = [i for i in 1:n if value(x[i]) > 0.5]","category":"page"},{"location":"tutorials/linear/knapsack/#Writing-a-function","page":"The knapsack problem example","title":"Writing a function","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"After working interactively, it is good practice to implement your model in a function.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"The function can be used to ensure that the model is given well-defined input data with validation checks, and that the solution process went as expected.","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"function solve_knapsack_problem(;\n    profit::Vector{Float64},\n    weight::Vector{Float64},\n    capacity::Float64,\n)\n    n = length(weight)\n    # The profit and weight vectors must be of equal length.\n    @assert length(profit) == n\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, x[1:n], Bin)\n    @objective(model, Max, profit' * x)\n    @constraint(model, weight' * x <= capacity)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    @assert primal_status(model) == FEASIBLE_POINT\n    println(\"Objective is: \", objective_value(model))\n    println(\"Solution is:\")\n    for i in 1:n\n        print(\"x[$i] = \", round(Int, value(x[i])))\n        println(\", c[$i] / w[$i] = \", profit[i] / weight[i])\n    end\n    chosen_items = [i for i in 1:n if value(x[i]) > 0.5]\n    return return chosen_items\nend\n\nsolve_knapsack_problem(; profit = profit, weight = weight, capacity = capacity)","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"We observe that the chosen items (1, 4, and 5) have the best profit to weight ratio in this particular example.","category":"page"},{"location":"tutorials/linear/knapsack/#Next-steps","page":"The knapsack problem example","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Here are some things to try next:","category":"page"},{"location":"tutorials/linear/knapsack/","page":"The knapsack problem example","title":"The knapsack problem example","text":"Call the function with different data. What happens as the capacity increases?\nWhat happens if the profit and weight vectors are different lengths?\nInstead of creating a binary variable with Bin, we could have written @variable(model, 0 <= x[1:n] <= 1, Int). Verify that this formulation finds the same solution. What happens if we are allowed to take more than one of each item?","category":"page"},{"location":"developers/checklists/#Checklists","page":"Checklists","title":"Checklists","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"The purpose of this page is to collate a series of checklists for commonly performed changes to the source code of JuMP.","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"In each case, copy the checklist into the description of the pull request.","category":"page"},{"location":"developers/checklists/#Making-a-release","page":"Checklists","title":"Making a release","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"In preparation for a release, use the following checklist. These steps can be done in the same commit, or separately. The last commit should have the message \"Prep for vX.Y.Z.\"","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"## Pre-release\n\n - [ ] Check that the pinned packages in `docs/Project.toml` are updated. We pin\n       the versions so that changes in the solvers (changes in printing, small\n       numeric changes) do not break the printing of the JuMP docs in arbitrary\n       commits.\n - [ ] Check that the `rev` fields in `docs/packages.toml` are updated. We pin\n       the versions of solvers and extensions to ensure that changes to their\n       READMEs do not break the JuMP docs in arbitrary commits, and to ensure\n       that the versions are compatible with the latest JuMP and\n       MathOptInterface releases.\n - [ ] Update `docs/src/changelog.md`\n - [ ] Run https://github.com/jump-dev/JuMP.jl/actions/workflows/extension-tests.yml\n       using a `workflow_dispatch` trigger to check for any changes in JuMP that\n       broke extensions.\n - [ ] Change the version number in `Project.toml`\n - [ ] Update the links in README.md\n - [ ] The commit messages in this PR do not contain `[ci skip]`\n\n## The release\n\n - [ ] After merging this pull request, comment `[at]JuliaRegistrator register` in\n       the GitHub commit. This should automatically publish a new version to the\n       Julia registry, as well as create a tag, and rebuild the documentation\n       for this tag.\n\n       These steps can take quite a bit of time (1 hour or more), so don't be\n       surprised if the new documentation takes a while to appear. In addition,\n       the links in the README will be broken until JuliaHub fetches the new\n       version on their servers.\n\n## Post-release\n\n - [ ] Once the tag is created, update the relevant `release-` branch. The latest\n       release branch at the time of writing is `release-1.0` (we haven't\n       back-ported any patches that needed to create a `release-1.Y` branch). To\n       to update the release branch with the v1.10.0 tag, do:\n       ```\n       git checkout release-1.0\n       git pull\n       git merge v1.10.0\n       git push\n       ```","category":"page"},{"location":"developers/checklists/#Adding-a-new-solver-to-the-documentation","page":"Checklists","title":"Adding a new solver to the documentation","text":"","category":"section"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"Use the following checklist when adding a new solver to the JuMP documentation.","category":"page"},{"location":"developers/checklists/","page":"Checklists","title":"Checklists","text":"## Basic\n\n - [ ] Check that the solver is a registered Julia package\n - [ ] Check that the solver supports the long-term support release of Julia\n - [ ] Check that the solver has a MathOptInterface wrapper\n - [ ] Check that the tests call `MOI.Test.runtests`. Some test excludes are\n       permissible, but the reason for skipping a particular test should be\n       documented.\n - [ ] Check that the README and/or documentation provides an example of how to\n       use the solver with JuMP\n\n## Documentation\n\n - [ ] Add a new row to the table in `docs/src/installation.md`\n\n## Optional\n\n - [ ] Add package metadata to `docs/packages.toml`","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"EditURL = \"simple_examples.jl\"","category":"page"},{"location":"tutorials/conic/simple_examples/#Simple-semidefinite-programming-examples","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial is a collection of examples of small conic programs from the field of semidefinite programming (SDP).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This tutorial makes use of the following packages:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport Random\nimport SCS\nimport Test","category":"page"},{"location":"tutorials/conic/simple_examples/#Maximum-cut-via-SDP","page":"Simple semidefinite programming examples","title":"Maximum cut via SDP","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The maximum cut problem is a classical example in graph theory, where we seek to partition a graph into two complementary sets, such that the weight of edges between the two sets is maximized. This problem is NP-hard, but it is possible to obtain an approximate solution using the semidefinite programming relaxation:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    textmax  quad   025 LX \n    text    st quad   mathrmdiag(X) = e \n                 qquad X succeq 0","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"where L is the weighted graph Laplacian and e is a vector of ones. For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Goemans, M. X., & Williamson, D. P. (1995). Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. Journal of the ACM (JACM), 42(6), 1115-1145.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"\"\"\"\n    svd_cholesky(X::AbstractMatrix, rtol)\n\nReturn the matrix `U` of the Cholesky decomposition of `X` as `U' * U`.\nNote that we do not use the `LinearAlgebra.cholesky` function because it\nrequires the matrix to be positive definite while `X` may be only\npositive *semi*definite.\n\nWe use the convention `U' * U` instead of `U * U'` to be consistent with\n`LinearAlgebra.cholesky`.\n\"\"\"\nfunction svd_cholesky(X::AbstractMatrix)\n    F = LinearAlgebra.svd(X)\n    # We now have `X ≈ `F.U * D² * F.U'` where:\n    D = LinearAlgebra.Diagonal(sqrt.(F.S))\n    # So `X ≈ U' * U` where `U` is:\n    return (F.U * D)'\nend\n\nfunction solve_max_cut_sdp(weights)\n    N = size(weights, 1)\n    # Calculate the (weighted) Laplacian of the graph: L = D - W.\n    L = LinearAlgebra.diagm(0 => weights * ones(N)) - weights\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, X[1:N, 1:N], PSD)\n    for i in 1:N\n        set_start_value(X[i, i], 1.0)\n    end\n    @objective(model, Max, 0.25 * LinearAlgebra.dot(L, X))\n    @constraint(model, LinearAlgebra.diag(X) .== 1)\n    optimize!(model)\n    V = svd_cholesky(value(X))\n    Random.seed!(N)\n    r = rand(N)\n    r /= LinearAlgebra.norm(r)\n    cut = [LinearAlgebra.dot(r, V[:, i]) > 0 for i in 1:N]\n    S = findall(cut)\n    T = findall(.!cut)\n    println(\"Solution:\")\n    println(\" (S, T) = ({\", join(S, \", \"), \"}, {\", join(T, \", \"), \"})\")\n    return S, T\nend","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 5 --- [2]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1}, {2})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 5; 5 0])","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 5 --- [2]\n |  \\          |\n |    \\        |\n 7      6      1\n |        \\    |\n |          \\  |\n[3] --- 1 --- [4]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1}, {2, 3, 4})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 5 7 6; 5 0 0 1; 7 0 0 1; 6 1 1 0])","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given the graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"[1] --- 1 --- [2]\n |             |\n |             |\n 5             9\n |             |\n |             |\n[3] --- 2 --- [4]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The solution is (S, T)  = ({1, 4}, {2, 3})","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"S, T = solve_max_cut_sdp([0 1 5 0; 1 0 0 9; 5 0 0 2; 0 9 2 0])","category":"page"},{"location":"tutorials/conic/simple_examples/#K-means-clustering-via-SDP","page":"Simple semidefinite programming examples","title":"K-means clustering via SDP","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given a set of points a_1 ldots a_m  in mathbbR^n, allocate them to k clusters.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Peng, J., & Wei, Y. (2007). Approximating k-means-type clustering via semidefinite programming. SIAM Journal on Optimization, 18(1), 186-205.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_k_means_clustering()\n    a = [[2.0, 2.0], [2.5, 2.1], [7.0, 7.0], [2.2, 2.3], [6.8, 7.0], [7.2, 7.5]]\n    m = length(a)\n    num_clusters = 2\n    W = zeros(m, m)\n    for i in 1:m, j in i+1:m\n        W[i, j] = W[j, i] = exp(-LinearAlgebra.norm(a[i] - a[j]) / 1.0)\n    end\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, Z[1:m, 1:m] >= 0, PSD)\n    @objective(model, Min, LinearAlgebra.tr(W * (LinearAlgebra.I - Z)))\n    @constraint(model, [i = 1:m], sum(Z[i, :]) .== 1)\n    @constraint(model, LinearAlgebra.tr(Z) == num_clusters)\n    optimize!(model)\n    Z_val = value.(Z)\n    current_cluster, visited = 0, Set{Int}()\n    for i in 1:m\n        if !(i in visited)\n            current_cluster += 1\n            println(\"Cluster $current_cluster\")\n            for j in i:m\n                if isapprox(Z_val[i, i], Z_val[i, j]; atol = 1e-3)\n                    println(a[j])\n                    push!(visited, j)\n                end\n            end\n        end\n    end\n    return\nend\n\nexample_k_means_clustering()","category":"page"},{"location":"tutorials/conic/simple_examples/#The-correlation-problem","page":"Simple semidefinite programming examples","title":"The correlation problem","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Given three random variables A, B, and C, and given bounds on two of the three correlation coefficients:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    -02 leq ρ_AB leq -01 \n    04  leq ρ_BC leq  05","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"our problem is to determine upper and lower bounds on other correlation coefficient ρ_AC.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"We solve an SDP to make use of the following positive semidefinite property of the correlation matrix:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"beginbmatrix\n      1     ρ_AB   ρ_AC \n     ρ_AB   1      ρ_BC  \n     ρ_AC   ρ_BC   1\nendbmatrix succeq 0","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_correlation_problem()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, X[1:3, 1:3], PSD)\n    S = [\"A\", \"B\", \"C\"]\n    ρ = Containers.DenseAxisArray(X, S, S)\n    @constraint(model, [i in S], ρ[i, i] == 1)\n    @constraint(model, -0.2 <= ρ[\"A\", \"B\"] <= -0.1)\n    @constraint(model, 0.4 <= ρ[\"B\", \"C\"] <= 0.5)\n    @objective(model, Max, ρ[\"A\", \"C\"])\n    optimize!(model)\n    println(\"An upper bound for ρ_AC is $(value(ρ[\"A\", \"C\"]))\")\n    @objective(model, Min, ρ[\"A\", \"C\"])\n    optimize!(model)\n    println(\"A lower bound for ρ_AC is $(value(ρ[\"A\", \"C\"]))\")\n    return\nend\n\nexample_correlation_problem()","category":"page"},{"location":"tutorials/conic/simple_examples/#The-minimum-distortion-problem","page":"Simple semidefinite programming examples","title":"The minimum distortion problem","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This example arises from computational geometry, in particular the problem of embedding a general finite metric space into a Euclidean space.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"It is known that the 4-point metric space defined by the star graph","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"  [1]\n    \\\n     1\n      \\\n      [0] —- 1 -- [2]\n      /\n     1\n    /\n  [3]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"cannot be exactly embedded into a Euclidean space of any dimension, where distances are computed by length of the shortest path between vertices. A distance-preserving embedding would require the three leaf nodes to form an equilateral triangle of side length 2, with the centre node (0) mapped to an equidistant point at distance 1; this is impossible since the triangle inequality in Euclidean space implies all points would need to be simultaneously collinear.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Here we will formulate and solve an SDP to compute the best possible embedding, that is, the embedding f assigning each vertex v to a vector f(v) that minimizes the distortion c such that","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    Da b leq f(a) - f(b) leq c  Da b","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"for all edges (a b) in the graph, where Da b is the distance in the graph metric space.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Any embedding f can be characterized by a Gram matrix Q, which is PSD and such that","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    f(a) - f(b)^2 = Qa a + Qb b - 2 Qa b","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The matrix entry Qab represents the inner product of f(a) with f(b).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"We therefore impose the constraint","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"    Da b^2 leq Qa a + Qb b - 2 Qa b leq c^2  Da b^2","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"for all edges (a b) in the graph and minimize c^2, which gives us the SDP formulation below. Since we may choose any point to be the origin, we fix the first vertex at 0.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"J. Matoušek (2002), Lectures on discrete geometry, Springer, pp. 378-379","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"N. Linial (2002), Finite metric spaces–combinatorics, geometry and algorithms, Proceedings of the ICM, Vol. 3, 573-586","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_minimum_distortion()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    D = [\n        0.0 1.0 1.0 1.0\n        1.0 0.0 2.0 2.0\n        1.0 2.0 0.0 2.0\n        1.0 2.0 2.0 0.0\n    ]\n    @variable(model, c² >= 1.0)\n    @variable(model, Q[1:4, 1:4], PSD)\n    for i in 1:4, j in (i+1):4\n        @constraint(model, D[i, j]^2 <= Q[i, i] + Q[j, j] - 2 * Q[i, j])\n        @constraint(model, Q[i, i] + Q[j, j] - 2 * Q[i, j] <= c² * D[i, j]^2)\n    end\n    fix(Q[1, 1], 0)\n    @objective(model, Min, c²)\n    optimize!(model)\n    Test.@test termination_status(model) == OPTIMAL\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 4 / 3 atol = 1e-4\n    # Recover the minimal distorted embedding:\n    X = [zeros(3) sqrt(value.(Q)[2:end, 2:end])]\n    return Plots.plot(\n        X[1, :],\n        X[2, :],\n        X[3, :];\n        seriestype = :mesh3d,\n        connections = ([0, 0, 0, 1], [1, 2, 3, 2], [2, 3, 1, 3]),\n        legend = false,\n        fillalpha = 0.1,\n        lw = 3,\n        ratio = :equal,\n        xlim = (-1.1, 1.1),\n        ylim = (-1.1, 1.1),\n        zlim = (-1.5, 1.0),\n        zticks = -1:1,\n        camera = (60, 30),\n    )\nend\n\nexample_minimum_distortion()","category":"page"},{"location":"tutorials/conic/simple_examples/#Lovász-numbers","page":"Simple semidefinite programming examples","title":"Lovász numbers","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"The Lovász number of a graph, also known as Lovász's theta-function, is a number that lies between two important and related numbers that are computationally hard to determine, namely the chromatic and clique numbers of the graph. It is possible however to efficient compute the Lovász number as the optimal value of a semidefinite program.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Consider the pentagon graph:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"     [5]\n    /   \\\n   /     \\\n [1]     [4]\n  |       |\n  |       |\n [2] --- [3]","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"with five vertices and edges. Its Lovász number is known to be precisely sqrt5 approx 2236, lying between 2 (the largest clique size) and 3 (the smallest number needed for a vertex coloring).","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Let i  j be integers such that 1 leq i  j leq 5. We define A^ij to be the 5 times 5 symmetric matrix with entries (ij) and (ji) equal to 1, with all other entries 0. Let  E be the graph's edge set; in this example, E contains (1,2), (2,3), (3,4), (4,5), (5,1) and their transposes. The Lovász number can be computed from the program","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"beginalign\n    textmax       quad    J  X  \n    text    st  quad A^ij  X = 0 text for all  (ij) notin E   \n                     quad I  X = 1 \n                     quad X succeq 0\nendalign","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"where J is the matrix filled with ones, and I is the identity matrix.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Barvinok, A. (2002). A course in convexity, American Mathematical Society (Vol. 54), pp. 182-185.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Knuth, D. E. (1994), The sandwich theorem, Electronic Journal of Combinatorics, Volume 1, Issue 1, A1.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_theta_problem()\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    E = [(1, 2), (2, 3), (3, 4), (4, 5), (5, 1)]\n    @variable(model, X[1:5, 1:5], PSD)\n    for i in 1:5\n        for j in (i+1):5\n            if !((i, j) in E || (j, i) in E)\n                A = zeros(Int, 5, 5)\n                A[i, j] = 1\n                A[j, i] = 1\n                @constraint(model, LinearAlgebra.dot(A, X) == 0)\n            end\n        end\n    end\n    @constraint(model, LinearAlgebra.tr(LinearAlgebra.I * X) == 1)\n    J = ones(Int, 5, 5)\n    @objective(model, Max, LinearAlgebra.dot(J, X))\n    optimize!(model)\n    Test.@test termination_status(model) == OPTIMAL\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ sqrt(5) rtol = 1e-4\n    println(\"The Lovász number is: $(objective_value(model))\")\n    return\nend\n\nexample_theta_problem()","category":"page"},{"location":"tutorials/conic/simple_examples/#Robust-uncertainty-sets","page":"Simple semidefinite programming examples","title":"Robust uncertainty sets","text":"","category":"section"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"This example computes the Value at Risk for a data-driven uncertainty set. Closed-form expressions for the optimal value are available. For more details, see:","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"Bertsimas, D., Gupta, V., & Kallus, N. (2018). Data-driven robust optimization. Mathematical Programming, 167, 235-292.","category":"page"},{"location":"tutorials/conic/simple_examples/","page":"Simple semidefinite programming examples","title":"Simple semidefinite programming examples","text":"function example_robust_uncertainty_sets()\n    R, d, 𝛿, ɛ = 1, 3, 0.05, 0.05\n    N = ceil((2 + 2 * log(2 / 𝛿))^2) + 1\n    c, μhat, M = randn(d), rand(d), rand(d, d)\n    Σhat = 1 / (d - 1) * (M - ones(d) * μhat')' * (M - ones(d) * μhat')\n    Γ1(𝛿, N) = R / sqrt(N) * (2 + sqrt(2 * log(1 / 𝛿)))\n    Γ2(𝛿, N) = 2 * R^2 / sqrt(N) * (2 + sqrt(2 * log(2 / 𝛿)))\n    model = Model(SCS.Optimizer)\n    set_silent(model)\n    @variable(model, Σ[1:d, 1:d], PSD)\n    @variable(model, u[1:d])\n    @variable(model, μ[1:d])\n    @constraint(model, [Γ1(𝛿 / 2, N); μ - μhat] in SecondOrderCone())\n    @constraint(model, [Γ2(𝛿 / 2, N); vec(Σ - Σhat)] in SecondOrderCone())\n    @constraint(model, [((1-ɛ)/ɛ) (u - μ)'; (u-μ) Σ] >= 0, PSDCone())\n    @objective(model, Max, c' * u)\n    optimize!(model)\n    exact =\n        μhat' * c +\n        Γ1(𝛿 / 2, N) * LinearAlgebra.norm(c) +\n        sqrt((1 - ɛ) / ɛ) *\n        sqrt(c' * (Σhat + Γ2(𝛿 / 2, N) * LinearAlgebra.I) * c)\n    Test.@test objective_value(model) ≈ exact atol = 1e-2\n    return\nend\n\nexample_robust_uncertainty_sets()","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/callbacks/#callbacks_manual","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Many mixed-integer (linear, conic, and nonlinear) programming solvers offer the ability to modify the solve process. Examples include changing branching decisions in branch-and-bound, adding custom cutting planes, providing custom heuristics to find feasible solutions, or implementing on-demand separators to add new constraints only when they are violated by the current solution (also known as lazy constraints).","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"While historically this functionality has been limited to solver-specific interfaces, JuMP provides solver-independent support for three types of callbacks:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"lazy constraints\nuser-cuts\nheuristic solutions","category":"page"},{"location":"manual/callbacks/#Available-solvers","page":"Solver-independent Callbacks","title":"Available solvers","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Solver-independent callback support is limited to a few solvers. This includes CPLEX, GLPK, Gurobi, Xpress, and SCIP.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nWhile JuMP provides a solver-independent way of accessing callbacks, you should not assume that you will see identical behavior when running the same code on different solvers. For example, some solvers may ignore user-cuts for various reasons, while other solvers may add every user-cut. Read the underlying solver's callback documentation to understand details specific to each solver.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"tip: Tip\nThis page discusses solver-independent callbacks. However, each solver listed above also provides a solver-dependent callback to provide access to the full range of solver-specific features. Consult the solver's README for an example of how to use the solver-dependent callback. This will require you to understand the C interface of the solver.","category":"page"},{"location":"manual/callbacks/#Things-you-can-and-cannot-do-during-solver-independent-callbacks","page":"Solver-independent Callbacks","title":"Things you can and cannot do during solver-independent callbacks","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"There is a limited range of things you can do during a callback. Only use the functions and macros explicitly stated in this page of the documentation, or in the Callbacks tutorial.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Using any other part of the JuMP API (for example, adding a constraint with @constraint or modifying a variable bound with set_lower_bound) is undefined behavior, and your solver may throw an error, return an incorrect solution, or result in a segfault that aborts Julia.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"In each of the three solver-independent callbacks, there are two things you may query:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"callback_node_status returns an MOI.CallbackNodeStatusCode enum indicating if the current primal solution is integer feasible.\ncallback_value returns the current primal solution of a variable.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"If you need to query any other information, use a solver-dependent callback instead. Each solver supporting a solver-dependent callback has information on how to use it in the README of their GitHub repository.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"If you want to modify the problem in a callback, you must use a lazy constraint.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nYou can only set each callback once. Calling set twice will over-write the earlier callback. In addition, if you use a solver-independent callback, you cannot set a solver-dependent callback.","category":"page"},{"location":"manual/callbacks/#Lazy-constraints","page":"Solver-independent Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Lazy constraints are useful when the full set of constraints is too large to explicitly include in the initial formulation. When a MIP solver reaches a new solution, for example with a heuristic or by solving a problem at a node in the branch-and-bound tree, it will give the user the chance to provide constraints that would make the current solution infeasible. For some more information about lazy constraints, see this blog post by Paul Rubin.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A lazy constraint callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           status = callback_node_status(cb_data, model)\n           if status == MOI.CALLBACK_NODE_STATUS_FRACTIONAL\n               # `callback_value(cb_data, x)` is not integer (to some tolerance).\n               # If, for example, your lazy constraint generator requires an\n               # integer-feasible primal solution, you can add a `return` here.\n               return\n           elseif status == MOI.CALLBACK_NODE_STATUS_INTEGER\n               # `callback_value(cb_data, x)` is integer (to some tolerance).\n           else\n               @assert status == MOI.CALLBACK_NODE_STATUS_UNKNOWN\n               # `callback_value(cb_data, x)` might be fractional or integer.\n           end\n           x_val = callback_value(cb_data, x)\n           if x_val > 2 + 1e-6\n               con = @build_constraint(x <= 2)\n               MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n           end\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.LazyConstraintCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe lazy constraint callback may be called at fractional or integer nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every primal solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nOnly add a lazy constraint if your primal solution violates the constraint. Adding the lazy constraint irrespective of feasibility may result in the solver returning an incorrect solution, or lead to many constraints being added, slowing down the solution process.model = Model(GLPK.Optimizer)\n@variable(model, x <= 10, Int)\n@objective(model, Max, x)\nfunction bad_callback_function(cb_data)\n    # Don't do this!\n    con = @build_constraint(x <= 2)\n    MOI.submit(model, MOI.LazyConstraint(cb_data), con)\nend\nfunction good_callback_function(cb_data)\n    if callback_value(x) > 2\n        con = @build_constraint(x <= 2)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nset_attribute(model, MOI.LazyConstraintCallback(), good_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nDuring the solve, a solver may visit a point that was cut off by a previous lazy constraint, for example, because the earlier lazy constraint was removed during presolve. If this happens, you must re-add the lazy constraint.","category":"page"},{"location":"manual/callbacks/#User-cuts","page":"Solver-independent Callbacks","title":"User cuts","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"User cuts, or simply cuts, provide a way for the user to tighten the LP relaxation using problem-specific knowledge that the solver cannot or is unable to infer from the model. Just like with lazy constraints, when a MIP solver reaches a new node in the branch-and-bound tree, it will give the user the chance to provide cuts to make the current relaxed (fractional) solution infeasible in the hopes of obtaining an integer solution. For more details about the difference between user cuts and lazy constraints see the aforementioned blog post.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A user-cut callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10.5, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           x_val = callback_value(cb_data, x)\n           con = @build_constraint(x <= floor(x_val))\n           MOI.submit(model, MOI.UserCut(cb_data), con)\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.UserCutCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nUser cuts must not change the set of integer feasible solutions. Equivalently, user cuts can only remove fractional solutions. If you add a cut that removes an integer solution (even one that is not optimal), the solver may return an incorrect solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe user-cut callback may be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every fractional primal solution.","category":"page"},{"location":"manual/callbacks/#Heuristic-solutions","page":"Solver-independent Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Integer programming solvers frequently include heuristics that run at the nodes of the branch-and-bound tree. They aim to find integer solutions quicker than plain branch-and-bound would to tighten the bound, allowing us to fathom nodes quicker and to tighten the integrality gap.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"Some heuristics take integer solutions and explore their \"local neighborhood\" (for example, flipping binary variables, fix some variables and solve a smaller MILP) and others take fractional solutions and attempt to round them in an intelligent way.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"You may want to add a heuristic of your own if you have some special insight into the problem structure that the solver is not aware of, for example, you can consistently take fractional solutions and intelligently guess integer solutions from them.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"A heuristic solution callback can be set using the following syntax:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"julia> import GLPK\n\njulia> model = Model(GLPK.Optimizer);\n\njulia> @variable(model, x <= 10.5, Int)\nx\n\njulia> @objective(model, Max, x)\nx\n\njulia> function my_callback_function(cb_data)\n           x_val = callback_value(cb_data, x)\n           status = MOI.submit(\n               model, MOI.HeuristicSolution(cb_data), [x], [floor(Int, x_val)]\n           )\n           println(\"I submitted a heuristic solution, and the status was: \", status)\n       end\nmy_callback_function (generic function with 1 method)\n\njulia> set_attribute(model, MOI.HeuristicCallback(), my_callback_function)","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"The third argument to submit is a vector of JuMP variables, and the fourth argument is a vector of values corresponding to each variable.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"MOI.submit returns an enum that depends on whether the solver accepted the solution. The possible return codes are:","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"MOI.HEURISTIC_SOLUTION_ACCEPTED\nMOI.HEURISTIC_SOLUTION_REJECTED\nMOI.HEURISTIC_SOLUTION_UNKNOWN","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"warning: Warning\nSome solvers may accept partial solutions. Others require a feasible integer solution for every variable. If in doubt, provide a complete solution.","category":"page"},{"location":"manual/callbacks/","page":"Solver-independent Callbacks","title":"Solver-independent Callbacks","text":"info: Info\nThe heuristic solution callback may be called at fractional nodes in the branch-and-bound tree. There is no guarantee that the callback is called at every fractional primal solution.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/infeasibility_certificates.md\"","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasibility-certificates","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"When given a conic problem that is infeasible or unbounded, some solvers can produce a certificate of infeasibility. This page explains what a certificate of infeasibility is, and the related conventions that MathOptInterface adopts.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Conic-duality","page":"Infeasibility certificates","title":"Conic duality","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"MathOptInterface uses conic duality to define infeasibility certificates. A full explanation is given in the section Duality, but here is a brief overview.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Minimization-problems","page":"Infeasibility certificates","title":"Minimization problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a minimization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n min_x in mathbbR^n  a_0^top x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and the dual is a maximization problem in standard conic form:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n max_y_1 ldots y_m  -sum_i=1^m b_i^top y_i + b_0\n\n textst  a_0 - sum_i=1^m A_i^top y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"where each mathcalC_i is a closed convex cone and mathcalC_i^* is its dual cone.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Maximization-problems","page":"Infeasibility certificates","title":"Maximization problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a maximization problem in geometric conic form, the primal is:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n max_x in mathbbR^n  a_0^top x + b_0\n\n textst  A_i x + b_i  in mathcalC_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and the dual is a minimization problem in standard conic form:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n min_y_1 ldots y_m  sum_i=1^m b_i^top y_i + b_0\n\n textst  a_0 + sum_i=1^m A_i^top y_i  = 0\n\n  y_i  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/#Unbounded-problems","page":"Infeasibility certificates","title":"Unbounded problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A problem is unbounded if and only if:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"there exists a feasible primal solution\nthe dual is infeasible.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A feasible primal solution—if one exists—can be obtained by setting ObjectiveSense to FEASIBILITY_SENSE before optimizing. Therefore, most solvers stop after they prove the dual is infeasible via a certificate of dual infeasibility, but before they have found a feasible primal solution. This is also the reason that MathOptInterface defines the DUAL_INFEASIBLE status instead of UNBOUNDED.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A certificate of dual infeasibility is an improving ray of the primal problem. That is, there exists some vector d such that for all eta  0:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A_i (x + eta d) + b_i in mathcalC_i  i = 1 ldots m","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and (for minimization problems):","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"a_0^top (x + eta d) + b_0  a_0^top x + b_0","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"for any feasible point x. The latter simplifies to a_0^top d  0. For maximization problems, the inequality is reversed, so that a_0^top d  0.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"If the solver has found a certificate of dual infeasibility:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"TerminationStatus must be DUAL_INFEASIBLE\nPrimalStatus must be INFEASIBILITY_CERTIFICATE\nVariablePrimal must be the corresponding value of d\nConstraintPrimal must be the corresponding value of A_i d\nObjectiveValue must be the value a_0^top d. Note that this is the value of the objective function at d, ignoring the constant b_0.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"note: Note\nThe choice of whether to scale the ray d to have magnitude 1 is left to the solver.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasible-problems","page":"Infeasibility certificates","title":"Infeasible problems","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"A certificate of primal infeasibility is an improving ray of the dual problem. However, because infeasibility is independent of the objective function, we first homogenize the primal problem by removing its objective.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"For a minimization problem, a dual improving ray is some vector d such that for all eta  0:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\n-sum_i=1^m A_i^top (y_i + eta d_i)  = 0 \n(y_i + eta d_i)  in mathcalC_i^*  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"and:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"-sum_i=1^m b_i^top (y_i + eta d_i)  -sum_i=1^m b_i^top y_i","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"for any feasible dual solution y. The latter simplifies to -sum_i=1^m b_i^top d_i  0. For a maximization problem, the inequality is sum_i=1^m b_i^top d_i  0. (Note that these are the same inequality, modulo a - sign.)","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"If the solver has found a certificate of primal infeasibility:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"TerminationStatus must be INFEASIBLE\nDualStatus must be INFEASIBILITY_CERTIFICATE\nConstraintDual must be the corresponding value of d\nDualObjectiveValue must be the value -sum_i=1^m b_i^top d_i for minimization problems and sum_i=1^m b_i^top d_i for maximization problems.","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"note: Note\nThe choice of whether to scale the ray d to have magnitude 1 is left to the solver.","category":"page"},{"location":"moi/background/infeasibility_certificates/#Infeasibility-certificates-of-variable-bounds","page":"Infeasibility certificates","title":"Infeasibility certificates of variable bounds","text":"","category":"section"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"Many linear solvers (for example, Gurobi) do not provide explicit access to the primal infeasibility certificate of a variable bound. However, given a set of linear constraints:","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"beginalign\nl_A le A x le u_A \nl_x le x le u_x\nendalign","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"the primal certificate of the variable bounds can be computed using the primal certificate associated with the affine constraints, d. (Note that d will have one element for each row of the A matrix, and that some or all of the elements in the vectors l_A and u_A may be pm infty. If both l_A and u_A are finite for some row, the corresponding element in `d  must be 0.)","category":"page"},{"location":"moi/background/infeasibility_certificates/","page":"Infeasibility certificates","title":"Infeasibility certificates","text":"Given d, compute bard = d^top A. If the bound is finite, a certificate for the lower variable bound of x_i is maxbard_i 0, and a certificate for the upper variable bound is minbard_i 0.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"EditURL = \"debugging.jl\"","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging","page":"Debugging","title":"Debugging","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Dealing with bugs is an unavoidable part of coding optimization models in any framework, including JuMP. Sources of bugs include not only generic coding errors (method errors, typos, off-by-one issues), but also semantic mistakes in the formulation of an optimization problem and the incorrect use of a solver.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This tutorial explains some common sources of bugs and modeling issues that you might encounter when writing models in JuMP, and it suggests a variety of strategies to deal with them.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nThis tutorial is more advanced than the other \"Getting started\" tutorials. It's in the \"Getting started\" section to give you an early preview of how to debug JuMP models. However, if you are new to JuMP, you may want to briefly skim the tutorial, and come back to it once you have written a few JuMP models.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/debugging/#Getting-help","page":"Debugging","title":"Getting help","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Debugging can be a frustrating part of modeling, particularly if you're new to optimization and programming. If you're stuck, join the community forum to search for answers to commonly asked questions.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Before asking a new question, make sure to read the post Make it easier to help you, which contains a number of tips on how to ask a good question.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Above all else, take time to simplify your code as much as possible. The fewer lines of code you can post that reproduces the same issue, the faster someone can answer your question.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-Julia-code","page":"Debugging","title":"Debugging Julia code","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Read the Debugging chapter in the book ThinkJulia.jl. It has a number of great tips and tricks for debugging Julia code.","category":"page"},{"location":"tutorials/getting_started/debugging/#Solve-failures","page":"Debugging","title":"Solve failures","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When a solver experiences an issue that prevents it from finding an optimal solution (or proving that one does not exist), JuMP may return one of a number of termination_statuses.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"For example, if the solver found a solution, but experienced numerical imprecision, it may return a status such as ALMOST_OPTIMAL or ALMOST_LOCALLY_SOLVED indicating that the problem was solved to a relaxed set of tolerances. Alternatively, the solver may return a problematic status such as NUMERICAL_ERROR, SLOW_PROGRESS, or OTHER_ERROR, indicating that it could not find a solution to the problem.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Most solvers can experience numerical imprecision because they use floating-point arithmetic to perform operations such as addition, subtraction, and multiplication. These operations aren't exact, and small errors can accrue between the theoretical value and the value that the computer computes. For example:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"0.1 * 3 == 0.3","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nRead the Guidelines for numerical issues section of the Gurobi documentation, along with the Debugging numerical problems section of the YALMIP documentation.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of solve failures are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Very large numbers and very small numbers as problem coefficients. Exactly what \"large\" is depends on the solver and the problem, but in general, values above 1e6 or smaller than 1e-6 cause problems.\nNonlinear problems with functions that are not defined in parts of their domain. For example, minimizing log(x) where x >= 0 is undefined when x = 0 (a common starting value).","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of solve failures include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Rescale variables in the problem and their associated coefficients to make the magnitudes of all coefficients in the 1e-4 to 1e4 range. For example, that might mean rescaling a variable from measuring distance in centimeters to kilometers.\nTry a different solver. Some solvers might be more robust than others for a particular problem.\nRead the documentation of your solver, and try settings that encourage numerical robustness.\nSet bounds or add constraints so that all nonlinear functions are defined across all of the feasible region. This particularly applies for functions like 1 / x and log(x) which are not defined for x = 0.","category":"page"},{"location":"tutorials/getting_started/debugging/#Incorrect-results","page":"Debugging","title":"Incorrect results","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Sometimes, you might find that the solver returns an \"optimal\" solution that is incorrect according to the model you are trying to solve (perhaps the solution is suboptimal, or it doesn't satisfy some of the constraints).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Incorrect results can be hard to detect and debug, because the solver gives no hints that there is a problem. Indeed, the termination_status will likely be OPTIMAL and a solution will be available.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-2","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of incorrect results are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A modeling error, so that your JuMP model does not match the formulation you have on paper\nNot accounting for the tolerances that solvers use (for example, if x is binary, a value like x = 1.0000001 may still be considered feasible)\nA bug in JuMP or the solver.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"The probability of the issue being a bug in JuMP or the solver is much smaller than a modeling error. When in doubt, first assume there is a bug in your code before assuming that there is a bug in JuMP.","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-2","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of incorrect results include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Print your JuMP model to see if it matches the formulation you have on paper. Look out for incorrect signs + instead of -, and off-by-one errors such as x[t] instead of x[t-1].\nCheck that you are not using exact comparisons like value(x) == 1.0; always use isapprox(value(x), 1.0; atol = 1e-6) where you manually specify the comparison tolerance.\nTry a different solver. If one solver succeeds where another doesn't this is a sign that the problem is a numerical issue or a bug in the solver.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-an-infeasible-model","page":"Debugging","title":"Debugging an infeasible model","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A model is infeasible if there is no primal solution that satisfies all of the constraints. In general, an infeasible model means one of two things:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Your problem really has no feasible solution\nThere is a mistake in your model.","category":"page"},{"location":"tutorials/getting_started/debugging/#Example","page":"Debugging","title":"Example","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A simple example of an infeasible model is:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n@objective(model, Max, 2x + 1)\n@constraint(model, con, 2x - 1 <= -2)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"because the bound says that x >= 0, but we can rewrite the constraint to be x <= -1/2. When the problem is infeasible, JuMP may return one of a number of statuses. The most common is INFEASIBLE:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Depending on the solver, you may also receive INFEASIBLE_OR_UNBOUNDED or LOCALLY_INFEASIBLE.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A termination status of INFEASIBLE_OR_UNBOUNDED means that the solver could not prove if the solver was infeasible or unbounded, only that the model does not have a finite feasible optimal solution.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Nonlinear optimizers such as Ipopt may return the status LOCALLY_INFEASIBLE. This does not mean that the solver proved no feasible solution exists, only that it could not find one. If you know a primal feasible point, try providing it as a starting point using set_start_value and re-optimize.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-3","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of infeasibility are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Incorrect units, for example, using a lower bound of megawatts and an upper bound of kilowatts\nUsing + instead of - in a constraint\nOff-by-one and related errors, for example, using x[t] instead of x[t-1] in part of a constraint\nOtherwise invalid mathematical formulations","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-3","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of infeasibility include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Iteratively comment out a constraint (or block of constraints) and re-solve the problem. When you find a constraint that makes the problem infeasible when added, check the constraint carefully for errors.\nIf the problem is still infeasible with all constraints commented out, check all variable bounds. Do they use the right data?\nIf you have a known feasible solution, use primal_feasibility_report to evaluate the constraints and check for violations. You'll probably find that you have a typo in one of the constraints.\nTry a different solver. Sometimes, solvers have bugs, and they can incorrectly report a problem as infeasible when it isn't. If you find such a case where one solver reports the problem is infeasible and another can find an optimal solution, please report it by opening an issue on the GitHub repository of the solver that reports infeasibility.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"tip: Tip\nSome solvers also have specialized support for debugging sources of infeasibility via an irreducible infeasible subsystem. To see if your solver has support, try calling compute_conflict!:julia> compute_conflict!(model)\nERROR: ArgumentError: The optimizer HiGHS.Optimizer does not support `compute_conflict!`In this case, HiGHS does not support computing conflicts, but other solvers such as Gurobi and CPLEX do. If the solver does support computing conflicts, read Conflicts for more details.","category":"page"},{"location":"tutorials/getting_started/debugging/#Penalty-relaxation","page":"Debugging","title":"Penalty relaxation","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Another strategy to debug sources of infeasibility is the relax_with_penalty! function.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"The penalty relaxation modifies constraints of the form f(x) in S into f(x) + y - z in S, where y z ge 0, and then it introduces a penalty term into the objective of a times (y + z) (if minimizing, else -a), where a is a penalty.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"map = relax_with_penalty!(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Here map is a dictionary which maps constraint indices to an affine expression representing (y + z).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"If we optimize the relaxed model, this time we get a feasible solution:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Iterate over the contents of map to see which constraints are violated:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"for (con, penalty) in map\n    violation = value(penalty)\n    if violation > 0\n        println(\"Constraint `$(name(con))` is violated by $violation\")\n    end\nend","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Once you find a violated constraint in the relaxed problem, take a look to see if there is a typo or other common mistake in that particular constraint.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Consult the docstring relax_with_penalty! for information on how to modify the penalty cost term a, either for every constraint in the model or a particular subset of the constraints.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When using relax_with_penalty!, you should be aware that:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Variable bounds and integrality restrictions are not relaxed. If the problem is still infeasible after calling relax_with_penalty!, check the variable bounds.\nYou cannot undo the penalty relaxation. If you need an unmodified model, rebuild the problem, or call copy_model before calling relax_with_penalty!.","category":"page"},{"location":"tutorials/getting_started/debugging/#Debugging-an-unbounded-model","page":"Debugging","title":"Debugging an unbounded model","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A model is unbounded if there is no limit on how good the objective value can get. Most often, an unbounded model means that you have an error in your modeling, because all physical systems have limits. (You cannot make an infinite amount of profit.)","category":"page"},{"location":"tutorials/getting_started/debugging/#Example-2","page":"Debugging","title":"Example","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"A simple example of an unbounded model is:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n@objective(model, Max, 2x + 1)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"because we can increase x without limit, and the objective value 2x + 1 gets better as x increases.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"When the problem is unbounded, JuMP may return one of a number of statuses. The most common is DUAL_INFEASIBLE:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\ntermination_status(model)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Depending on the solver, you may also receive INFEASIBLE_OR_UNBOUNDED or an error code like NORM_LIMIT.","category":"page"},{"location":"tutorials/getting_started/debugging/#Common-sources-4","page":"Debugging","title":"Common sources","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Common sources of unboundedness are:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Using Max instead of Min\nOmitting variable bounds, such as 0 <= x <= 1\nUsing + instead of - in a term of the objective function.","category":"page"},{"location":"tutorials/getting_started/debugging/#Strategies-4","page":"Debugging","title":"Strategies","text":"","category":"section"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Strategies to debug sources of unboundedness include:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"Double check whether you intended Min or Max in the @objective line.\nPrint the objective function with print(objective_function(model)) and verify that the value and sign of each coefficient is as you expect.\nAdd large bounds to all variables that are free or have one-sided bounds, then re-solve the problem. Because all variables are now bounded, the problem will have a finite optimal solution. Look at the value of each variable in the optimal solution to see if it is at one of the new bounds. If it is, you either need to specify a better bound for that variable, or there might be a mistake in the objective function associated with that variable (for example, a + instead of a -).","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"If there are too many variables to add bounds to, or there are too many terms to examine by hand, another strategy is to create a new variable with a large upper bound (if maximizing, lower bound if minimizing) and a constraint that the variable must be less-than or equal to the expression of the objective function. For example:","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"model = Model(HiGHS.Optimizer);\nset_silent(model)\n@variable(model, x >= 0)\n# @objective(model, Max, 2x + 1)\n@variable(model, objective <= 10_000)\n@constraint(model, objective <= 2x + 1)\n@objective(model, Max, objective)","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"This new model has a finite optimal solution, so we can solve it and then look for variables with large positive or negative values in the optimal solution.","category":"page"},{"location":"tutorials/getting_started/debugging/","page":"Debugging","title":"Debugging","text":"optimize!(model)\nfor var in all_variables(model)\n    if var == objective\n        continue\n    end\n    if abs(value(var)) > 1e3\n        println(\"Variable `$(name(var))` may be unbounded\")\n    end\nend","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"EditURL = \"https://github.com/jump-dev/ECOS.jl/blob/v1.1.1/README.md\"","category":"page"},{"location":"packages/ECOS/#ECOS.jl","page":"jump-dev/ECOS.jl","title":"ECOS.jl","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"ECOS.jl is a wrapper for the ECOS solver.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/ECOS/#Affiliation","page":"jump-dev/ECOS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"This wrapper is maintained by the JuMP community and is not a product of Embotech AG.","category":"page"},{"location":"packages/ECOS/#License","page":"jump-dev/ECOS.jl","title":"License","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"ECOS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The underlying solver, embotech/ecos, is licensed under the GPL v3 license.","category":"page"},{"location":"packages/ECOS/#Installation","page":"jump-dev/ECOS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"Install ECOS.jl using Pkg.add:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"import Pkg\nPkg.add(\"ECOS\")","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"In addition to installing the ECOS.jl package, this will also download and install the ECOS binaries. You do not need to install ECOS separately.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/ECOS/#Use-with-JuMP","page":"jump-dev/ECOS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"To use ECOS with JuMP, use ECOS.Optimizer:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"using JuMP, ECOS\nmodel = Model(ECOS.Optimizer)\nset_attribute(model, \"maxit\", 100)","category":"page"},{"location":"packages/ECOS/#MathOptInterface-API","page":"jump-dev/ECOS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The ECOS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.VectorAffineFunction{Float64} in MOI.Nonnegatives\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorAffineFunction{Float64} in MOI.Zeros","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/ECOS/#Options","page":"jump-dev/ECOS.jl","title":"Options","text":"","category":"section"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"The following options are supported:","category":"page"},{"location":"packages/ECOS/","page":"jump-dev/ECOS.jl","title":"jump-dev/ECOS.jl","text":"Parameter Explanation\ngamma scaling the final step length\ndelta regularization parameter\neps regularization threshold\nfeastol primal/dual infeasibility tolerance\nabstol absolute tolerance on duality gap\nreltol relative tolerance on duality gap\nfeastol_inacc primal/dual infeasibility relaxed tolerance\nabstol_inacc absolute relaxed tolerance on duality gap\nreltol_inacc relative relaxed tolerance on duality gap\nnitref number of iterative refinement steps\nmaxit maximum number of iterations\nverbose verbosity bool for PRINTLEVEL < 3","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Nonlinear/overview.md\"","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#nonlinear_developers","page":"Overview","title":"Nonlinear","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nThe Nonlinear submodule is experimental. Until this message is removed, breaking changes may be introduced in any minor or patch release of MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The Nonlinear submodule contains data structures and functions for working with a nonlinear optimization problem in the form of an expression graph. This page explains the API and describes the rationale behind its design.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Standard-form","page":"Overview","title":"Standard form","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear programs (NLPs) are a class of optimization problems in which some of the constraints or the objective function are nonlinear:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"beginalign\n    min_x in mathbbR^n  f_0(x) \n    textst  l_j le f_j(x) le u_j  j = 1 ldots m\nendalign","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There may be additional constraints, as well as things like variable bounds and integrality restrictions, but we do not consider them here because they are best dealt with by other components of MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#API-overview","page":"Overview","title":"API overview","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The core element of the Nonlinear submodule is Nonlinear.Model:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> const Nonlinear = MOI.Nonlinear;\n\njulia> model = Nonlinear.Model()\nA Nonlinear.Model with:\n 0 objectives\n 0 parameters\n 0 expressions\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model is a mutable struct that stores all of the nonlinear information added to the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Decision-variables","page":"Overview","title":"Decision variables","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Decision variables are represented by VariableIndexes. The user is responsible for creating these using MOI.VariableIndex(i), where i is the column associated with the variable.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Expressions","page":"Overview","title":"Expressions","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The input data structure is a Julia Expr. The input expressions can incorporate VariableIndexes, but these must be interpolated into the expression with $:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> input = :(1 + sin($x)^2)\n:(1 + sin(MathOptInterface.VariableIndex(1)) ^ 2)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There are a number of restrictions on the input Expr:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"It cannot contain macros\nIt cannot contain broadcasting\nIt cannot contain splatting (except in limited situations)\nIt cannot contain linear algebra, such as matrix-vector products\nIt cannot contain generator expressions, including sum(i for i in S)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Given an input expression, add an expression using Nonlinear.add_expression:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Nonlinear.add_expression(model, input)\nMathOptInterface.Nonlinear.ExpressionIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, expr, is a Nonlinear.ExpressionIndex that can then be interpolated into other input expressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Looking  again at model, we see:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model\nA Nonlinear.Model with:\n 0 objectives\n 0 parameters\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Parameters","page":"Overview","title":"Parameters","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In addition to constant literals like 1 or 1.23, you can create parameters. Parameters are placeholders whose values can change before passing the expression to the solver. Create a parameter using Nonlinear.add_parameter, which accepts a default value:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> p = Nonlinear.add_parameter(model, 1.23)\nMathOptInterface.Nonlinear.ParameterIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, p, is a Nonlinear.ParameterIndex that can then be interpolated into other input expressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Looking  again at model, we see:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model\nA Nonlinear.Model with:\n 0 objectives\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Update a parameter as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> model[p]\n1.23\n\njulia> model[p] = 4.56\n4.56\n\njulia> model[p]\n4.56","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Objectives","page":"Overview","title":"Objectives","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Set a nonlinear objective using Nonlinear.set_objective:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, :($p + $expr + $x))\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Clear a nonlinear objective by passing nothing:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, nothing)\n\njulia> model\nA Nonlinear.Model with:\n 0 objectives\n 1 parameter\n 1 expression\n 0 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"But we'll re-add the objective for later:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.set_objective(model, :($p + $expr + $x));","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_Constraints","page":"Overview","title":"Constraints","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Add a constraint using Nonlinear.add_constraint:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> c = Nonlinear.add_constraint(model, :(1 + sqrt($x)), MOI.LessThan(2.0))\nMathOptInterface.Nonlinear.ConstraintIndex(1)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 1 constraint","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The return value, c, is a Nonlinear.ConstraintIndex that is a unique identifier for the constraint. Interval constraints are also supported:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> c2 = Nonlinear.add_constraint(model, :(1 + sqrt($x)), MOI.Interval(-1.0, 2.0))\nMathOptInterface.Nonlinear.ConstraintIndex(2)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 2 constraints","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Delete a constraint using Nonlinear.delete:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> Nonlinear.delete(model, c2)\n\njulia> model\nA Nonlinear.Model with:\n 1 objective\n 1 parameter\n 1 expression\n 1 constraint","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#User-defined-operators","page":"Overview","title":"User-defined operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear supports a wide range of univariate and multivariate operators. However, you can also define your own operators by registering them.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Univariate-operators","page":"Overview","title":"Univariate operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Register a univariate user-defined operator using Nonlinear.register_operator:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f(x) = 1 + sin(x)^2\nf (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f, 1, f)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now, you can use :my_f in expressions:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> new_expr = Nonlinear.add_expression(model, :(my_f($x + 1)))\nMathOptInterface.Nonlinear.ExpressionIndex(2)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear will compute first- and second-derivatives of the registered operator using ForwardDiff.jl. Override this by passing functions which compute the respective derivative:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f′(x) = 2 * sin(x) * cos(x)\nf′ (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f2, 1, f, f′)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"or","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> f′′(x) = 2 * (cos(x)^2 - sin(x)^2)\nf′′ (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_f3, 1, f, f′, f′′)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Multivariate-operators","page":"Overview","title":"Multivariate operators","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Register a multivariate user-defined operator using Nonlinear.register_operator:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> g(x...) = x[1]^2 + x[1] * x[2] + x[2]^2\ng (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_g, 2, g)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now, you can use :my_g in expressions:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> new_expr = Nonlinear.add_expression(model, :(my_g($x + 1, $x)))\nMathOptInterface.Nonlinear.ExpressionIndex(3)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"By default, Nonlinear will compute the gradient of the registered operator using ForwardDiff.jl. (Hessian information is not supported.) Override this by passing a function to compute the gradient:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> function ∇g(ret, x...)\n           ret[1] = 2 * x[1] + x[2]\n           ret[2] = x[1] + 2 * x[2]\n           return\n       end\n∇g (generic function with 1 method)\n\njulia> Nonlinear.register_operator(model, :my_g2, 2, g, ∇g)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Nonlinear_MOI_interface","page":"Overview","title":"MathOptInterface","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"MathOptInterface communicates the nonlinear portion of an optimization problem to solvers using concrete subtypes of AbstractNLPEvaluator, which implement the Nonlinear programming API.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Create an AbstractNLPEvaluator from Nonlinear.Model using Nonlinear.Evaluator.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Evaluator requires an Nonlinear.AbstractAutomaticDifferentiation backend and an ordered list of the variables that are included in the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"There following backends are available to choose from within MOI, although other packages may add more options by sub-typing Nonlinear.AbstractAutomaticDifferentiation:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.ExprGraphOnly\nNonlinear.SparseReverseMode.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> evaluator = Nonlinear.Evaluator(model, Nonlinear.ExprGraphOnly(), [x])\nNonlinear.Evaluator with available features:\n  * :ExprGraph","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The functions of the Nonlinear programming API implemented by Nonlinear.Evaluator depends upon the chosen Nonlinear.AbstractAutomaticDifferentiation backend.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The :ExprGraph feature means we can call objective_expr and constraint_expr to retrieve the expression graph of the problem. However, we cannot call gradient terms such as eval_objective_gradient because Nonlinear.ExprGraphOnly does not have the capability to differentiate a nonlinear expression.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"If, instead, we pass Nonlinear.SparseReverseMode, then we get access to :Grad, the gradient of the objective function, :Jac, the Jacobian matrix of the constraints, :JacVec, the ability to compute Jacobian-vector products, and :ExprGraph.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> evaluator = Nonlinear.Evaluator(\n           model,\n           Nonlinear.SparseReverseMode(),\n           [x],\n       )\nNonlinear.Evaluator with available features:\n  * :Grad\n  * :Jac\n  * :JacVec\n  * :ExprGraph","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"However, before using the evaluator, we need to call initialize:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> MOI.initialize(evaluator, [:Grad, :Jac, :JacVec, :ExprGraph])","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Now we can call methods like eval_objective:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = [1.0]\n1-element Vector{Float64}:\n 1.0\n\njulia> MOI.eval_objective(evaluator, x)\n7.268073418273571","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"and eval_objective_gradient:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> grad = [0.0]\n1-element Vector{Float64}:\n 0.0\n\njulia> MOI.eval_objective_gradient(evaluator, grad, x)\n\njulia> grad\n1-element Vector{Float64}:\n 1.909297426825682","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Instead of passing Nonlinear.Evaluator directly to solvers, solvers query the NLPBlock attribute, which returns an NLPBlockData. This object wraps an Nonlinear.Evaluator and includes other information such as constraint bounds and whether the evaluator has a nonlinear objective. Create and set NLPBlockData as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> block = MOI.NLPBlockData(evaluator);\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}());\n\njulia> MOI.set(model, MOI.NLPBlock(), block);","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nOnly call NLPBlockData once you have finished modifying the problem in model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Putting everything together, you can create a nonlinear optimization problem in MathOptInterface as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"import MathOptInterface as MOI\n\nfunction build_model(\n    model::MOI.ModelLike;\n    backend::MOI.Nonlinear.AbstractAutomaticDifferentiation,\n)\n    x = MOI.add_variable(model)\n    y = MOI.add_variable(model)\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    nl_model = MOI.Nonlinear.Model()\n    MOI.Nonlinear.set_objective(nl_model, :($x^2 + $y^2))\n    evaluator = MOI.Nonlinear.Evaluator(nl_model, backend, [x, y])\n    MOI.set(model, MOI.NLPBlock(), MOI.NLPBlockData(evaluator))\n    return\nend\n\n# Replace `model` and `backend` with your optimizer and backend of choice.\nmodel = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nbuild_model(model; backend = MOI.Nonlinear.SparseReverseMode())","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Expression-graph-representation","page":"Overview","title":"Expression-graph representation","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model stores nonlinear expressions in Nonlinear.Expressions. This section explains the design of the expression graph data structure in Nonlinear.Expression.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Given a nonlinear function like f(x) = sin(x)^2 + x, a conceptual aid for thinking about the graph representation of the expression is to convert it into Polish prefix notation:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"f(x, y) = (+ (^ (sin x) 2) x)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This format identifies each operator (function), as well as a list of arguments. Operators can be univariate, like sin, or multivariate, like +.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"A common way of representing Polish prefix notation in code is as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.VariableIndex(1);\n\njulia> struct ExprNode\n           op::Symbol\n           children::Vector{Union{ExprNode,Float64,MOI.VariableIndex}}\n       end\n\njulia> expr = ExprNode(:+, [ExprNode(:^, [ExprNode(:sin, [x]), 2.0]), x]);","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This data structure follows our Polish prefix notation very closely, and we can easily identify the arguments to an operator. However, it has a significant draw-back: each node in the graph requires a Vector, which is heap-allocated and tracked by Julia's garbage collector (GC). For large models, we can expect to have millions of nodes in the expression graph, so this overhead quickly becomes prohibitive for computation.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"An alternative is to record the expression as a linear tape:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Any[:+, 2, :^, 2, :sin, 1, x, 2.0, x]\n9-element Vector{Any}:\n  :+\n 2\n  :^\n 2\n  :sin\n 1\n  MOI.VariableIndex(1)\n 2.0\n  MOI.VariableIndex(1)","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The Int after each operator Symbol specifies the number of arguments.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This data-structure is a single vector, which resolves our problem with the GC, but each element is the abstract type, Any, and so any operations on it will lead to slower dynamic dispatch. It's also hard to identify the children of each operation without reading the entire tape.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"To summarize, representing expression graphs in Julia has the following challenges:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nodes in the expression graph should not contain a heap-allocated object\nAll data-structures should be concretely typed\nIt should be easy to identify the children of a node","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Sketch-of-the-design-in-Nonlinear","page":"Overview","title":"Sketch of the design in Nonlinear","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear overcomes these problems by decomposing the data structure into a number of different concrete-typed vectors.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"First, we create vectors of the supported uni- and multivariate operators.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> const UNIVARIATE_OPERATORS = [:sin];\n\njulia> const MULTIVARIATE_OPERATORS = [:+, :^];","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, there are many more supported operations than the ones listed here.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Second, we create an enum to represent the different types of nodes present in the expression graph:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> @enum(\n           NodeType,\n           NODE_CALL_MULTIVARIATE,\n           NODE_CALL_UNIVARIATE,\n           NODE_VARIABLE,\n           NODE_VALUE,\n       )","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, there are node types other than the ones listed here.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Third, we create two concretely typed structs as follows:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> struct Node\n           type::NodeType\n           parent::Int\n           index::Int\n       end\n\njulia> struct Expression\n           nodes::Vector{Node}\n           values::Vector{Float64}\n       end","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"For each node node in the .nodes field, if node.type is:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"NODE_CALL_MULTIVARIATE, we look up MULTIVARIATE_OPERATORS[node.index] to retrieve the operator\nNODE_CALL_UNIVARIATE, we look up UNIVARIATE_OPERATORS[node.index] to retrieve the operator\nNODE_VARIABLE, we create MOI.VariableIndex(node.index)\nNODE_VALUE, we look up values[node.index]","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The .parent field of each node is the integer index of the parent node in .nodes. For the first node, the parent is -1 by convention.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Therefore, we can represent our function as:","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"julia> expr = Expression(\n           [\n               Node(NODE_CALL_MULTIVARIATE, -1, 1),\n               Node(NODE_CALL_MULTIVARIATE, 1, 2),\n               Node(NODE_CALL_UNIVARIATE, 2, 1),\n               Node(NODE_VARIABLE, 3, 1),\n               Node(NODE_VALUE, 2, 1),\n               Node(NODE_VARIABLE, 1, 1),\n           ],\n           [2.0],\n       );","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This is less readable than the other options, but does this data structure meet our design goals?","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Instead of a heap-allocated object for each node, we only have two Vectors for each expression, nodes and values, as well as two constant vectors for the OPERATORS. In addition, all fields are concretely typed, and there are no Union or Any types.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"For our third goal, it is not easy to identify the children of a node, but it is easy to identify the parent of any node. Therefore, we can use Nonlinear.adjacency_matrix to compute a sparse matrix that maps parents to their children.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The tape is also ordered topologically, so that a reverse pass of the nodes evaluates all children nodes before their parent.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#The-design-in-practice","page":"Overview","title":"The design in practice","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"In practice, Node and Expression are exactly Nonlinear.Node and Nonlinear.Expression. However, Nonlinear.NodeType has more fields to account for comparison operators such as :>= and :<=, logic operators such as :&& and :||, nonlinear parameters, and nested subexpressions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Moreover, instead of storing the operators as global constants, they are stored in Nonlinear.OperatorRegistry, and it also stores a vector of logic operators and a vector of comparison operators. In addition to Nonlinear.DEFAULT_UNIVARIATE_OPERATORS and Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS, you can register user-defined functions using Nonlinear.register_operator.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.Model is a struct that stores the Nonlinear.OperatorRegistry, as well as a list of parameters and subexpressions in the model.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#ReverseAD","page":"Overview","title":"ReverseAD","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Nonlinear.ReverseAD is a submodule for computing derivatives of a nonlinear optimization problem using sparse reverse-mode automatic differentiation (AD).","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"This section does not attempt to explain how sparse reverse-mode AD works, but instead explains why MOI contains its own implementation, and highlights notable differences from similar packages.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nDon't use the API in ReverseAD to compute derivatives. Instead, create a Nonlinear.Evaluator object with Nonlinear.SparseReverseMode as the backend, and then query the MOI API methods.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#Design-goals","page":"Overview","title":"Design goals","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"The JuliaDiff organization maintains a list of packages for doing AD in Julia. At last count, there were at least ten packages——not including ReverseAD——for reverse-mode AD in Julia. ReverseAD exists because it has a different set of design goals.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Goal: handle scale and sparsity. The types of nonlinear optimization problems that MOI represents can be large scale (10^5 or more functions across 10^5 or more variables) with very sparse derivatives. The ability to compute a sparse Hessian matrix is essential. To the best of our knowledge, ReverseAD is the only reverse-mode AD system in Julia that handles sparsity by default.\nGoal: limit the scope to improve robustness. Most other AD packages accept arbitrary Julia functions as input and then trace an expression graph using operator overloading. This means they must deal (or detect and ignore) with control flow, I/O, and other vagaries of Julia. In contrast, ReverseAD only accepts functions in the form of Nonlinear.Expression, which greatly limits the range of syntax that it must deal with. By reducing the scope of what we accept as input to functions relevant for mathematical optimization, we can provide a simpler implementation with various performance optimizations.\nGoal: provide outputs which match what solvers expect. Other AD packages focus on differentiating individual Julia functions. In contrast, ReverseAD has a very specific use-case: to generate outputs needed by the MOI nonlinear API. This means it needs to efficiently compute sparse Hessians, and it needs subexpression handling to avoid recomputing subexpressions that are shared between functions.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/#History","page":"Overview","title":"History","text":"","category":"section"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"ReverseAD started life as ReverseDiffSparse.jl, development of which began in early 2014(!). This was well before the other AD packages started development. Because we had a well-tested, working AD in JuMP, there was less motivation to contribute to and explore other AD packages. The lack of historical interaction also meant that other packages were not optimized for the types of problems that JuMP is built for (that is, large-scale sparse problems). When we first created MathOptInterface, we kept the AD in JuMP to simplify the transition, and post-poned the development of a first-class nonlinear interface in MathOptInterface.","category":"page"},{"location":"moi/submodules/Nonlinear/overview/","page":"Overview","title":"Overview","text":"Prior to the introduction of Nonlinear, JuMP's nonlinear implementation was a confusing mix of functions and types spread across the code base and in the private _Derivatives submodule. This made it hard to swap the AD system for another. The main motivation for refactoring JuMP to create the Nonlinear submodule in MathOptInterface was to abstract the interface between JuMP and the AD system, allowing us to swap-in and test new AD systems in the future.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"EditURL = \"https://github.com/ds4dm/Tulip.jl/blob/069cdda082dfde31ef7ee935c96e833a3b96b33c/README.md\"","category":"page"},{"location":"packages/Tulip/#Tulip","page":"ds4dm/Tulip.jl","title":"Tulip","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"(Image: DOI) (Image: ) (Image: )","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip is an open-source interior-point solver for linear optimization, written in pure Julia. It implements the homogeneous primal-dual interior-point algorithm with multiple centrality corrections, and therefore handles unbounded and infeasible problems. Tulip’s main feature is that its algorithmic framework is disentangled from linear algebra implementations. This allows to seamlessly integrate specialized routines for structured problems.","category":"page"},{"location":"packages/Tulip/#License","page":"ds4dm/Tulip.jl","title":"License","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Tulip/#Installation","page":"ds4dm/Tulip.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Install Tulip using the Julia package manager:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"import Pkg\nPkg.add(\"Tulip\")","category":"page"},{"location":"packages/Tulip/#Usage","page":"ds4dm/Tulip.jl","title":"Usage","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The recommended way of using Tulip is through JuMP or MathOptInterface (MOI).","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The low-level interface is still under development and is likely change in the future. The MOI interface is more stable.","category":"page"},{"location":"packages/Tulip/#Using-with-JuMP","page":"ds4dm/Tulip.jl","title":"Using with JuMP","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Tulip follows the syntax convention PackageName.Optimizer:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using JuMP\nimport Tulip\nmodel = Model(Tulip.Optimizer)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"Linear objectives, linear constraints and lower/upper bounds on variables are supported.","category":"page"},{"location":"packages/Tulip/#Using-with-MOI","page":"ds4dm/Tulip.jl","title":"Using with MOI","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"The type Tulip.Optimizer is parametrized by the model's arithmetic, for example, Float64 or BigFloat. This allows to solve problem in higher numerical precision. See the documentation for more details.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"import MathOptInterface as MOI\nimport Tulip\nmodel = Tulip.Optimizer{Float64}()   # Create a model in Float64 precision\nmodel = Tulip.Optimizer()            # Defaults to the above call\nmodel = Tulip.Optimizer{BigFloat}()  # Create a model in BigFloat precision","category":"page"},{"location":"packages/Tulip/#Solver-parameters","page":"ds4dm/Tulip.jl","title":"Solver parameters","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"See the documentation for a full list of parameters.","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in JuMP, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using JuMP, Tulip\nmodel = Model(Tulip.Optimizer)\nset_attribute(model, \"IPM_IterationsLimit\", 200)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in MathOptInterface, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using Tulip\nimport MathOptInterface as MOI\nmodel = Tulip.Optimizer{Float64}()\nMOI.set(model, MOI.RawOptimizerAttribute(\"IPM_IterationsLimit\"), 200)","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"To set parameters in the Tulip API, use:","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"using Tulip\nmodel = Tulip.Model{Float64}()\nTulip.set_parameter(model, \"IPM_IterationsLimit\", 200)","category":"page"},{"location":"packages/Tulip/#Command-line-executable","page":"ds4dm/Tulip.jl","title":"Command-line executable","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"See app building instructions.","category":"page"},{"location":"packages/Tulip/#Citing-Tulip.jl","page":"ds4dm/Tulip.jl","title":"Citing Tulip.jl","text":"","category":"section"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"If you use Tulip in your work, we kindly ask that you cite the following reference (preprint available here).","category":"page"},{"location":"packages/Tulip/","page":"ds4dm/Tulip.jl","title":"ds4dm/Tulip.jl","text":"@Article{Tulip.jl,\n  author   = {Tanneau, Mathieu and Anjos, Miguel F. and Lodi, Andrea},\n  journal  = {Mathematical Programming Computation},\n  title    = {Design and implementation of a modular interior-point solver for linear optimization},\n  year     = {2021},\n  issn     = {1867-2957},\n  month    = feb,\n  doi      = {10.1007/s12532-020-00200-8},\n  language = {en},\n  url      = {https://doi.org/10.1007/s12532-020-00200-8},\n  urldate  = {2021-03-07},\n}","category":"page"},{"location":"tutorials/algorithms/parallelism/#Parallelism","page":"Parallelism","title":"Parallelism","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The purpose of this tutorial is to give a brief overview of parallelism in Julia as it pertains to JuMP, and to explain some of the things to be aware of when writing parallel algorithms involving JuMP models.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Multi-threading-and-distributed-computing","page":"Parallelism","title":"Multi-threading and distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"There are two main types of parallelism in Julia:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Multi-threading\nDistributed computing","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"In multi-threading, multiple tasks are run in a single Julia process and share the same memory. In distributed computing, tasks are run in multiple Julia processes with independent memory spaces. This can include processes across multiple physical machines, such as in a high-performance computing cluster.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Choosing and understanding the type of parallelism you are using is important because the code you write for each type is different, and there are different limitations and benefits to each approach. However, the best choice is highly problem dependent, so you may want to experiment with both approaches to determine what works for your situation.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Multi-threading","page":"Parallelism","title":"Multi-threading","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use multi-threading with Julia, you must either start Julia with the command line flag --threads=N, or you must set the JULIA_NUM_THREADS environment variable before launching Julia. For this documentation, we set the environment variable to:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> ENV[\"JULIA_NUM_THREADS\"]\n\"4\"","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"You can check how many threads are available using:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Threads.nthreads()\n4","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The easiest way to use multi-threading in Julia is by placing the Threads.@threads macro in front of a for-loop:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> @time begin\n           ids = Int[]\n           my_lock = Threads.ReentrantLock()\n           Threads.@threads for i in 1:Threads.nthreads()\n               global ids, my_lock\n               Threads.lock(my_lock) do\n                   push!(ids, Threads.threadid())\n               end\n               sleep(1.0)\n           end\n       end\n  1.037087 seconds (31.32 k allocations: 1.836 MiB, 2.02% compilation time)","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"This for-loop sleeps for 1 second on each iteration. Thus, if it had executed sequentially, it should have taken the same number of seconds as there are threads available. Instead, it took only 1 second, showing that the iterations were executed simultaneously. We can verify this by checking the Threads.threadid() of the thread that executed each iteration:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> ids\n4-element Vector{Int64}:\n 2\n 4\n 1\n 3","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nWhen working with threads, you need to avoid race conditions, in which two threads attempt to write to the same variable at the same time. In the above example we avoided a race condition by using ReentrantLock. See the Multi-threading section of the Julia documentation for more details.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Distributed-computing","page":"Parallelism","title":"Distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use distributed computing with Julia, use the Distributed package:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> import Distributed","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Like multi-threading, we need to tell Julia how many processes to add. We can do this either by starting Julia with the -p N command line argument, or by using Distributed.addprocs:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> import Pkg\n\njulia> project = Pkg.project();\n\njulia> workers = Distributed.addprocs(4; exeflags = \"--project=$(project.path)\")\n4-element Vector{Int64}:\n 2\n 3\n 4\n 5","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nNot loading the parent environment with --project is a common mistake.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"The added processes are \"worker\" processes that we can use to do computation with. They are orchestrated by the process with the id 1. You can check what process the code is currently running on using Distributed.myid()","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.myid()\n1","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"As a general rule, to get maximum performance you should add as many processes as you have logical cores available.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unlike the for-loop approach of multi-threading, distributed computing extends the Julia map function to a \"parallel-map\" function Distributed.pmap. For each element in the list of arguments to map over, Julia will copy the element to an idle worker process and evaluate the function, passing the element as an input argument.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> function hard_work(i::Int)\n           sleep(1.0)\n           return Distributed.myid()\n       end\nhard_work (generic function with 1 method)\n\njulia> Distributed.pmap(hard_work, 1:4)\nERROR: On worker 2:\nUndefVarError: #hard_work not defined\nStacktrace:\n[...]","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unfortunately, if you try this code directly, you will get an error message that says On worker 2: UndefVarError: hard_work not defined. The error is thrown because, although process 1 knows what the hard_work function is, the worker processes do not.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To fix the error, we need to use Distributed.@everywhere, which evaluates the code on every process:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.@everywhere begin\n           function hard_work(i::Int)\n               sleep(1.0)\n               return Distributed.myid()\n           end\n       end","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Now if we run pmap, we see that it took only 1 second instead of 4, and that it executed on each of the worker processes:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> @time ids = Distributed.pmap(hard_work, 1:4)\n  1.202006 seconds (216.39 k allocations: 13.301 MiB, 4.07% compilation time)\n4-element Vector{Int64}:\n 2\n 3\n 5\n 4","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"tip: Tip\nFor more information, read the Julia documentation Distributed Computing.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Using-parallelism-the-wrong-way","page":"Parallelism","title":"Using parallelism the wrong way","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/#When-building-a-JuMP-model","page":"Parallelism","title":"When building a JuMP model","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"For large problems, building the model in JuMP can be a bottleneck, and you may consider trying to write code that builds the model in parallel, for example, by wrapping a for-loop that adds constraints with Threads.@threads.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Unfortunately, you cannot build a JuMP model in parallel, and attempting to do so may error or produce incorrect results.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"In most cases, we find that the reason for the bottleneck is not JuMP, but in how you are constructing the problem data, and that with changes, it is possible to build a model in a way that is not the bottleneck in the solution process.","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"tip: Tip\nLooking for help to make your code run faster? Ask for help on the community forum. Make sure to include a reproducible example of your code.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-a-single-JuMP-model","page":"Parallelism","title":"With a single JuMP model","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"A common approach people try is to use parallelism with a single JuMP model. For example, to optimize a model over multiple right-hand side vectors, you may try:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"using JuMP\nimport HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x)\n@objective(model, Min, x)\nsolutions = Pair{Int,Float64}[]\nmy_lock = Threads.ReentrantLock()\nThreads.@threads for i in 1:10\n    set_lower_bound(x, i)\n    optimize!(model)\n    Threads.lock(my_lock) do\n        push!(solutions, i => objective_value(model))\n    end\nend","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"This will not work, and attempting to do so may error, crash Julia or produce incorrect results.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Using-parallelism-the-right-way","page":"Parallelism","title":"Using parallelism the right way","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"To use parallelism with JuMP, the simplest rule to remember is that each worker must have its own instance of a JuMP model.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-multi-threading","page":"Parallelism","title":"With multi-threading","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"With multi-threading, create a new instance of model in each iteration of the for-loop:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> using JuMP\n\njulia> import HiGHS\n\njulia> solutions = Pair{Int,Float64}[]\n\njulia> my_lock = Threads.ReentrantLock();\n\njulia> Threads.@threads for i in 1:10\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           set_attribute(model, MOI.NumberOfThreads(), 1)\n           @variable(model, x)\n           @objective(model, Min, x)\n           set_lower_bound(x, i)\n           optimize!(model)\n           Threads.lock(my_lock) do\n               push!(solutions, i => objective_value(model))\n           end\n       end\n\njulia> solutions\n10-element Vector{Pair{Int64, Float64}}:\n  7 => 7.0\n  4 => 4.0\n  1 => 1.0\n  9 => 9.0\n  5 => 5.0\n  8 => 8.0\n 10 => 10.0\n  2 => 2.0\n  6 => 6.0\n  3 => 3.0","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"warning: Warning\nFor some solvers, it may be necessary to limit the number of threads used internally by the solver to 1 by setting the MOI.NumberOfThreads attribute.","category":"page"},{"location":"tutorials/algorithms/parallelism/#With-distributed-computing","page":"Parallelism","title":"With distributed computing","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"With distributed computing, remember to evaluate all of the code on all of the processes using Distributed.@everywhere, and then write a function which creates a new instance of the model on every evaluation:","category":"page"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"julia> Distributed.@everywhere begin\n           using JuMP\n           import HiGHS\n       end\n\njulia> Distributed.@everywhere begin\n           function solve_model_with_right_hand_side(i)\n               model = Model(HiGHS.Optimizer)\n               set_silent(model)\n               @variable(model, x)\n               @objective(model, Min, x)\n               set_lower_bound(x, i)\n               optimize!(model)\n               return objective_value(model)\n           end\n       end\n\njulia> solutions = Distributed.pmap(solve_model_with_right_hand_side, 1:10)\n10-element Vector{Float64}:\n  1.0\n  2.0\n  3.0\n  4.0\n  5.0\n  6.0\n  7.0\n  8.0\n  9.0\n 10.0","category":"page"},{"location":"tutorials/algorithms/parallelism/#Other-types-of-parallelism","page":"Parallelism","title":"Other types of parallelism","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/#GPU","page":"Parallelism","title":"GPU","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"JuMP does not support GPU programming, and few solvers support execution on a GPU.","category":"page"},{"location":"tutorials/algorithms/parallelism/#Parallelism-within-the-solver","page":"Parallelism","title":"Parallelism within the solver","text":"","category":"section"},{"location":"tutorials/algorithms/parallelism/","page":"Parallelism","title":"Parallelism","text":"Many solvers use parallelism internally. For example, commercial solvers like Gurobi and CPLEX both parallelize the search in branch-and-bound. Solvers supporting internal parallelism will typically support the MOI.NumberOfThreads attribute, which you can set using set_attribute.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/variables.md\"","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/variables/#Variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"moi/manual/variables/#Add-a-variable","page":"Variables","title":"Add a variable","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Use add_variable to add a single variable.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"add_variable returns a VariableIndex type, which is used to refer to the added variable in other calls.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Check if a VariableIndex is valid using is_valid.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.is_valid(model, x)\ntrue","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Use add_variables to add a number of variables.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> y = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nThe integer does not necessarily correspond to the column inside an optimizer.","category":"page"},{"location":"moi/manual/variables/#Delete-a-variable","page":"Variables","title":"Delete a variable","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Delete a variable using delete.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.delete(model, x)\n\njulia> MOI.is_valid(model, x)\nfalse","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nNot all ModelLike models support deleting variables. A DeleteNotAllowed error is thrown if this is not supported.","category":"page"},{"location":"moi/manual/variables/#Variable-attributes","page":"Variables","title":"Variable attributes","text":"","category":"section"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"The following attributes are available for variables:","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"VariableName\nVariablePrimalStart\nVariablePrimal","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"Get and set these attributes using get and set.","category":"page"},{"location":"moi/manual/variables/","page":"Variables","title":"Variables","text":"julia> MOI.set(model, MOI.VariableName(), x, \"var_x\")\n\njulia> MOI.get(model, MOI.VariableName(), x)\n\"var_x\"","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"EditURL = \"complementarity.jl\"","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Mixed-complementarity-problems","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial is a collection of mixed complementarity programs.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"using JuMP\nimport PATHSolver","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Background","page":"Mixed complementarity problems","title":"Background","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"A mixed complementarity problem has the form:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"beginalign\n    F_i(x) perp x_i  i = 1 ldots n \n    l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"where the perp constraint enforces the following relations:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"If l_i  x_i  u_i, then F_i(x) = 0\nIf l_i = x_i, then F_i(x) ge 0\nIf x_i = u_i, then F_i(x) le 0","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"You may have seen a complementarity problem written as 0 le F(x) perp x ge 0. This is a special case of a mixed complementarity problem in which l_i = 0 and u_i = infty.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"Importantly, a mixed complementarity problem does not have an objective, and no other constraint types are present.","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Linear-complementarity","page":"Mixed complementarity problems","title":"Linear complementarity","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"Form a mixed complementarity problem using the perp symbol ⟂ (type \\perp<tab> in the REPL).","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"M = [0 0 -1 -1; 0 0 1 -2; 1 -1 2 -2; 1 2 -2 4]\nq = [2, 2, -2, -6]\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[1:4] <= 10, start = 0)\n@constraint(model, M * x + q ⟂ x)\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Other-ways-of-writing-linear-complementarity-problems","page":"Mixed complementarity problems","title":"Other ways of writing linear complementarity problems","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"You do not need to use a single vector of variables, and the complementarity constraints can be given in any order. In addition, you can use the perp symbol, the complements(F, x) syntax, or the MOI.Complements set.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"model = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, 0 <= w <= 10, start = 0)\n@variable(model, 0 <= x <= 10, start = 0)\n@variable(model, 0 <= y <= 10, start = 0)\n@variable(model, 0 <= z <= 10, start = 0)\n@constraint(model, complements(y - 2z + 2, x))\n@constraint(model, [-y - z + 2, w] in MOI.Complements(2))\n@constraint(model, w + 2x - 2y + 4z - 6 ⟂ z)\n@constraint(model, w - x + 2y - 2z - 2 ⟂ y)\noptimize!(model)\nvalue.([w, x, y, z])","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Transportation","page":"Mixed complementarity problems","title":"Transportation","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is a reformulation of the transportation problem from Chapter 3.3 of Dantzig, G.B. (1963). Linear Programming and Extensions. Princeton University Press, Princeton, New Jersey. It is based on the GAMS model gamslib_transmcp.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"capacity = Dict(\"seattle\" => 350, \"san-diego\" => 600)\ndemand = Dict(\"new-york\" => 325, \"chicago\" => 300, \"topeka\" => 275)\ncost = Dict(\n    (\"seattle\" => \"new-york\") => 90 * 2.5 / 1_000,\n    (\"seattle\" => \"chicago\") => 90 * 1.7 / 1_000,\n    (\"seattle\" => \"topeka\") => 90 * 1.8 / 1_000,\n    (\"san-diego\" => \"new-york\") => 90 * 2.5 / 1_000,\n    (\"san-diego\" => \"chicago\") => 90 * 1.8 / 1_000,\n    (\"san-diego\" => \"topeka\") => 90 * 1.4 / 1_000,\n)\nplants, markets = keys(capacity), keys(demand)\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, w[i in plants] >= 0)\n@variable(model, p[j in markets] >= 0)\n@variable(model, x[i in plants, j in markets] >= 0)\n@constraints(\n    model,\n    begin\n        [i in plants, j in markets], w[i] + cost[i=>j] - p[j] ⟂ x[i, j]\n        [i in plants], capacity[i] - sum(x[i, :]) ⟂ w[i]\n        [j in markets], sum(x[:, j]) - demand[j] ⟂ p[j]\n    end\n)\noptimize!(model)\nvalue.(p)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Expected-utility-of-insurance","page":"Mixed complementarity problems","title":"Expected utility of insurance","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is taken from a lecture of the course AAE706, given by Thomas F. Rutherford at the University of Wisconsin, Madison. It models the expected coverage of insurance K that a rational actor would obtain to insure a risk that occurs with probability pi and results in a loss of L.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"pi = 0.01  # Probability of a bad outcome\nL = 0.5    # Loss with a bad outcome\nγ = 0.02   # Premium for coverage\nσ = 0.5    # Elasticity\nρ = -1     # Risk exponent\nU(C) = C^ρ / ρ\nMU(C) = C^(ρ - 1)\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, EU, start = 1)   # Expected utilitiy\n@variable(model, EV, start = 1)   # Equivalent variation in income\n@variable(model, C_G, start = 1)  # Consumption on a good day\n@variable(model, C_B, start = 1)  # Consumption on a bad day\n@variable(model, K, start = 1)    # Coverage\n@constraints(\n    model,\n    begin\n        (1 - pi) * U(C_G) + pi * U(C_B) - EU ⟂ EU\n        100 * (((1 - pi) * C_G^ρ + pi * C_B^ρ)^(1 / ρ) - 1) - EV ⟂ EV\n        1 - γ * K - C_G ⟂ C_G\n        1 - L + (1 - γ) * K - C_B ⟂ C_B\n        γ * ((1 - pi) * MU(C_G) + pi * MU(C_B)) - pi * MU(C_B) ⟂ K\n    end\n)\noptimize!(model)\nvalue(K)","category":"page"},{"location":"tutorials/nonlinear/complementarity/#Electricity-consumption","page":"Mixed complementarity problems","title":"Electricity consumption","text":"","category":"section"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example is a mixed complementarity formulation of example 3.3.1 from D’Aertrycke, G., Ehrenmann, A., Ralph, D., & Smeers, Y. (2017). Risk trading in capacity equilibrium models.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"This example models a risk neutral competitive equilibrium between a producer and a consumer of electricity.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"In our example, we assume a producer is looking to invest in a new power plant with capacity x [MW]. This plant has an annualized capital cost of I [€/MW] and an operating cost of C [€/MWh]. There are 8760 hours in a year.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"After making the capital investment, there are five possible consumption scenarios, omega, which occur with probability theta_omega. In each scenario , the producer makes Y_ω MW of electricity.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"There is one consumer in the model, who has a quadratic utility function, U(Q_ω) = A_ω Q_ω + fracB_ω Q_ω^22.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"We now build and solve the mixed complementarity problem with a few brief comments. The economic justification for the model would require a larger tutorial than the space available here. Consult the original text for details.","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"I = 90_000                     # Annualized capital cost\nC = 60                         # Operation cost per MWh\nτ = 8_760                      # Hours per year\nθ = [0.2, 0.2, 0.2, 0.2, 0.2]  # Scenario probabilities\nA = [300, 350, 400, 450, 500]  # Utility function coefficients\nB = 1                          # Utility function coefficients\nmodel = Model(PATHSolver.Optimizer)\nset_silent(model)\n@variable(model, x >= 0, start = 1)           # Installed capacity\n@variable(model, Q[ω = 1:5] >= 0, start = 1)  # Consumption\n@variable(model, Y[ω = 1:5] >= 0, start = 1)  # Production\n@variable(model, P[ω = 1:5], start = 1)       # Electricity price\n@variable(model, μ[ω = 1:5] >= 0, start = 1)  # Capital scarcity margin\n# Unit investment cost equals annualized scarcity profit or investment is 0\n@constraint(model, I - τ * θ' * μ ⟂ x)\n# Difference between price and scarcity margin is equal to operation cost\n@constraint(model, [ω = 1:5], C - (P[ω] - μ[ω]) ⟂ Y[ω])\n# Price is equal to consumer's marginal utility\n@constraint(model, [ω = 1:5], P[ω] - (A[ω] - B * Q[ω]) ⟂ Q[ω])\n# Production is equal to consumption\n@constraint(model, [ω = 1:5], Y[ω] - Q[ω] ⟂ P[ω])\n# Production does not exceed capacity\n@constraint(model, [ω = 1:5], x - Y[ω] ⟂ μ[ω])\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"An equilibrium solution is to build 389 MW:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value(x)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"The production in each scenario is:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value.(Q)","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"The price in each scenario is:","category":"page"},{"location":"tutorials/nonlinear/complementarity/","page":"Mixed complementarity problems","title":"Mixed complementarity problems","text":"value.(P)","category":"page"},{"location":"developers/custom_solver_binaries/#How-to-use-a-custom-binary","page":"Custom binaries","title":"How to use a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Many solvers are not written in Julia, but instead in languages like C or C++. JuMP interacts with these solvers through binary dependencies.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"For many open-source solvers, we automatically install the appropriate binary when you run Pkg.add(\"Solver\"). For example, Pkg.add(\"ECOS\") will also install the ECOS binary.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"This page explains how this installation works, and how you can use a custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"compat: Compat\nThese instructions require Julia 1.6 or later.","category":"page"},{"location":"developers/custom_solver_binaries/#Background","page":"Custom binaries","title":"Background","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Each solver that JuMP supports is structured as a Julia package. For example, the interface for the ECOS solver is provided by the ECOS.jl package.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"tip: Tip\nThis page uses the example of ECOS.jl because it is simple to compile. Other solvers follow similar conventions. For example, the interface to the Clp solver is provided by Clp.jl.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The ECOS.jl package provides an interface between the C API of ECOS and MathOptInterface. However, it does not handle the installation of the solver binary; that is the job for a JLL package.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"A JLL is a Julia package that wraps a pre-compiled binary. Binaries are built using Yggdrasil (for example, ECOS) and hosted in the JuliaBinaryWrappers GitHub repository (for example, ECOS_jll.jl).","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"JLL packages contain little code. Their only job is to dlopen a dynamic library, along with any dependencies.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"JLL packages manage their binary dependencies using Julia's artifact system. Each JLL package has an Artifacts.toml file which describes where to find each binary artifact for each different platform that it might be installed on. Here is the Artifacts.toml file for ECOS_jll.jl.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The binaries installed by the JLL package should be sufficient for most users. In rare cases, however, you may require a custom binary. The two main reasons to use a custom binary are:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"You want a binary with custom compilation settings (for example, debugging)\nYou want a binary with a set of dependencies that are not available on Yggdrasil (for example, a commercial solver like Gurobi or CPLEX).","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The following sections explain how to replace the binaries provided by a JLL package with the custom ones you have compiled. As a reminder, we use ECOS as an example for simplicity, but the steps are the same for other solvers.","category":"page"},{"location":"developers/custom_solver_binaries/#jll_structure","page":"Custom binaries","title":"Explore the JLL you want to override","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The first step is to explore the structure and filenames of the JLL package we want to override.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Find the location of the files using .artifact_dir:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.artifact_dir\n\"/Users/oscar/.julia/artifacts/2addb75332eff5a1657b46bb6bf30d2410bc7ecf\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"tip: Tip\nThis path may be different on other machines.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Here is what it contains:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> readdir(ECOS_jll.artifact_dir)\n4-element Vector{String}:\n \"include\"\n \"lib\"\n \"logs\"\n \"share\"\n\njulia> readdir(joinpath(ECOS_jll.artifact_dir, \"lib\"))\n1-element Vector{String}:\n \"libecos.dylib\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Other solvers may have a bin directory containing executables. To use a custom binary of ECOS, we need to replace /lib/libecos.dylib with our custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#compile_ecos","page":"Custom binaries","title":"Compile a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"The next step is to compile a custom binary. Because ECOS is written in C with no dependencies, this is easy to do if you have a C compiler:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"oscar@Oscars-MBP jll_example % git clone https://github.com/embotech/ecos.git\n[... lines omitted ...]\noscar@Oscars-MBP jll_example % cd ecos\noscar@Oscars-MBP ecos % make shared\n[... many lines omitted...]\noscar@Oscars-MBP ecos % mkdir lib\noscar@Oscars-MBP ecos % cp libecos.dylib lib","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"warning: Warning\nCompiling custom solver binaries is an advanced operation. Due to the complexities of compiling various solvers, the JuMP community is unable to help you diagnose and fix compilation issues.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"After this compilation step, we now have a folder /tmp/jll_example/ecos that contains lib and include directories with the same files as ECOS_jll:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> readdir(joinpath(\"ecos\", \"lib\"))\n1-element Vector{String}:\n \"libecos.dylib\"","category":"page"},{"location":"developers/custom_solver_binaries/#Overriding-a-single-library","page":"Custom binaries","title":"Overriding a single library","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To override the libecos library, we need to know what ECOS_jll calls it. (In most cases, it will also be libecos, but not always.)","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"There are two ways you can check.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Check the bottom of the JLL's GitHub README. For example, ECOS_jll has a single LibraryProduct called libecos.\nType ECOS_jll. and the press the [TAB] key twice to auto-complete available options:\njulia> ECOS_jll.\nLIBPATH           PATH_list          best_wrapper       get_libecos_path   libecos_handle\nLIBPATH_list      __init__           dev_jll            is_available       libecos_path\nPATH              artifact_dir       find_artifact_dir  libecos\nHere you can see there is libecos, and more usefully for us, libecos_path.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Once you know the name of the variable to override (the one that ends in _path), use  Preferences.jl to specify a new path:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"using Preferences\nset_preferences!(\n    \"LocalPreferences.toml\",\n    \"ECOS_jll\",\n    \"libecos_path\" => \"/tmp/jll_example/ecos/lib/libecos\"\n)","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"This will create a file in your current directory called LocalPreferences.toml with the contents:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"[ECOS_jll]\nlibecos_path = \"/tmp/jll_example/ecos/lib/libecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Now if you restart Julia, you will see:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.libecos\n\"/tmp/jll_example/ecos/lib/libecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To go back to using the default library, just delete the LocalPreferences.toml file.","category":"page"},{"location":"developers/custom_solver_binaries/#Overriding-an-entire-artifact","page":"Custom binaries","title":"Overriding an entire artifact","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Sometimes a solver may provide a number of libraries and executables, and specifying the path for each of the becomes tedious. In this case, we can use Julia's Override.toml to replace an entire artifact.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Overriding an entire artifact requires you to replicate the structure and contents of the JLL package that we explored above.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"In most cases you need only reproduce the include, lib, and bin directories (if they exist). You can safely ignore any logs or share directories. Take careful note of what files each directory contains and what they are called.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"For our ECOS example, we already reproduced the structure when we compiled ECOS.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"So, now we need to tell Julia to use our custom installation instead of the default. We can do this by making an override file at ~/.julia/artifacts/Overrides.toml.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Overrides.toml has the following content:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"# Override for ECOS_jll\n2addb75332eff5a1657b46bb6bf30d2410bc7ecf = \"/tmp/jll_example/ecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"where 2addb75332eff5a1657b46bb6bf30d2410bc7ecf is the folder from the original ECOS_jll.artifact_dir and \"/tmp/jll_example/ecos\" is the location of our new installation. Replace these as appropriate for your system.","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"If you restart Julia after creating the override file, you will see:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using ECOS_jll\n\njulia> ECOS_jll.artifact_dir\n\"/tmp/jll_example/ecos\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Now when we use ECOS it will use our custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#Using-Cbc-with-a-custom-binary","page":"Custom binaries","title":"Using Cbc with a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"As a second example, we demonstrate how to use Cbc.jl with a custom binary.","category":"page"},{"location":"developers/custom_solver_binaries/#Explore-the-JLL-you-want-to-override","page":"Custom binaries","title":"Explore the JLL you want to override","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"First, let's check where Cbc_jll is installed:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> using Cbc_jll\n\njulia> Cbc_jll.artifact_dir\n\"/Users/oscar/.julia/artifacts/e481bc81db5e229ba1f52b2b4bd57484204b1b06\"\n\njulia> readdir(Cbc_jll.artifact_dir)\n5-element Vector{String}:\n \"bin\"\n \"include\"\n \"lib\"\n \"logs\"\n \"share\"\n\njulia> readdir(joinpath(Cbc_jll.artifact_dir, \"bin\"))\n1-element Vector{String}:\n \"cbc\"\n\njulia> readdir(joinpath(Cbc_jll.artifact_dir, \"lib\"))\n10-element Vector{String}:\n \"libCbc.3.10.5.dylib\"\n \"libCbc.3.dylib\"\n \"libCbc.dylib\"\n \"libCbcSolver.3.10.5.dylib\"\n \"libCbcSolver.3.dylib\"\n \"libCbcSolver.dylib\"\n \"libOsiCbc.3.10.5.dylib\"\n \"libOsiCbc.3.dylib\"\n \"libOsiCbc.dylib\"\n \"pkgconfig\"","category":"page"},{"location":"developers/custom_solver_binaries/#Compile-a-custom-binary","page":"Custom binaries","title":"Compile a custom binary","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Next, we need to compile Cbc. Cbc can be difficult to compile (it has a lot of dependencies), but for macOS users there is a homebrew recipe:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"(base) oscar@Oscars-MBP jll_example % brew install cbc\n[ ... lines omitted ... ]\n(base) oscar@Oscars-MBP jll_example % brew list cbc\n/usr/local/Cellar/cbc/2.10.5/bin/cbc\n/usr/local/Cellar/cbc/2.10.5/include/cbc/ (76 files)\n/usr/local/Cellar/cbc/2.10.5/lib/libCbc.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/libCbcSolver.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/libOsiCbc.3.10.5.dylib\n/usr/local/Cellar/cbc/2.10.5/lib/pkgconfig/ (2 files)\n/usr/local/Cellar/cbc/2.10.5/lib/ (6 other files)\n/usr/local/Cellar/cbc/2.10.5/share/cbc/ (59 files)\n/usr/local/Cellar/cbc/2.10.5/share/coin/ (4 files)","category":"page"},{"location":"developers/custom_solver_binaries/#Override-single-libraries","page":"Custom binaries","title":"Override single libraries","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To use Preferences.jl to override specific libraries we first check the names of each library in Cbc_jll:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"julia> Cbc_jll.\nLIBPATH               cbc                    get_libcbcsolver_path  libOsiCbc_path\nLIBPATH_list          cbc_path               is_available           libcbcsolver\nPATH                  dev_jll                libCbc                 libcbcsolver_handle\nPATH_list             find_artifact_dir      libCbc_handle          libcbcsolver_path\n__init__              get_cbc_path           libCbc_path\nartifact_dir          get_libCbc_path        libOsiCbc\nbest_wrapper          get_libOsiCbc_path     libOsiCbc_handle","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"Then we add the following to LocalPreferences.toml:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"[Cbc_jll]\ncbc_path = \"/usr/local/Cellar/cbc/2.10.5/bin/cbc\"\nlibCbc_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libCbc.3.10.5\"\nlibOsiCbc_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libOsiCbc.3.10.5\"\nlibcbcsolver_path = \"/usr/local/Cellar/cbc/2.10.5/lib/libCbcSolver.3.10.5\"","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"info: Info\nNote that capitalization matters, so libcbcsolver_path corresponds to libCbcSolver.3.10.5.","category":"page"},{"location":"developers/custom_solver_binaries/#Override-entire-artifact","page":"Custom binaries","title":"Override entire artifact","text":"","category":"section"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"To use the homebrew install as our custom binary we add the following to ~/.julia/artifacts/Overrides.toml:","category":"page"},{"location":"developers/custom_solver_binaries/","page":"Custom binaries","title":"Custom binaries","text":"# Override for Cbc_jll\ne481bc81db5e229ba1f52b2b4bd57484204b1b06 = \"/usr/local/Cellar/cbc/2.10.5\"","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"EditURL = \"benders_decomposition.jl\"","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#benders_decomposition_classical","page":"Benders decomposition","title":"Benders decomposition","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial was originally contributed by Shuvomoy Das Gupta.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This tutorial describes how to implement Benders decomposition in JuMP. It uses the following packages:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"using JuMP\nimport GLPK\nimport Printf","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Theory","page":"Benders decomposition","title":"Theory","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition is a useful algorithm for solving convex optimization problems with a large number of variables. It works best when a larger problem can be decomposed into two (or more) smaller problems that are individually much easier to solve. This tutorial demonstrates Benders decomposition on the following mixed-integer linear program:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\ntextmin          c_1^top x+c_2^top y   \ntextsubject to   A_1 x+ A_2 y le b      \n                     x ge 0                 \n                     y ge 0                 \n                     x in mathbbZ^n\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"where b in mathbbR^m, A_1 in mathbbR^m times n, A_2 in mathbbR^m times p and mathbbZ is the set of integers.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Any mixed integer programming problem can be written in the form above.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"If there are relatively few integer variables, and many more continuous variables, then it may be beneficial to decompose the problem into a small problem containing only integer variables and a linear program containing only continuous variables. Hopefully, the linear program will be much easier to solve in isolation than in the full mixed-integer linear program.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For example, if we knew a feasible solution for x, we could obtain a solution for y by solving:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\nV_2(x) =  textmin            c_2^top y                        \n          textsubject to     A_2 y le b - A_1 x  quad pi \n                                 y ge 0\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"where pi is the dual variable associated with the constraints. Because this is a linear program, it is easy to solve.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Replacing the c_2^top y component of the objective in our original problem with V_2 yields:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\ntextmin          c_1^top x + V_2(x) \ntextsubject to   x ge 0             \n                     x in mathbbZ^n\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This problem looks a lot simpler to solve, but we need to do something else with V_2 first.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Because x is a constant that appears on the right-hand side of the constraints, V_2 is a convex function with respect to x, and the dual variable pi can be multiplied by -A_1 to obtain a subgradient of V_2(x) with respect to x. Therefore, if we have a candidate solution x_k, then we can solve V_2(x_k) and obtain a feasible dual vector pi_k. Using these values, we can construct a first-order Taylor-series approximation of V_2 about the point x_k:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"V_2(x) ge V_2(x_k) + -pi_k^top A_1 (x - x_k)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"By convexity, we know that this inequality holds for all x, and we call these inequalities cuts.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition is an iterative technique that replaces V_2(x) with a new decision variable theta, and approximates it from below using cuts:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"beginaligned\nV_1^K =  textmin           c_1^top x + theta  \n          textsubject to   x ge 0              \n                              x in mathbbZ^n   \n                              theta ge M         \n                              theta ge V_2(x_k) + pi_k^top(x - x_k)  quad forall k = 1ldotsK\nendaligned","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This integer program is called the first-stage subproblem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"To generate cuts, we solve V_1^K to obtain a candidate first-stage solution x_k, then we use that solution to solve V_2(x_k). Then, using the optimal objective value and dual solution from V_2, we add a new cut to form V_1^K+1 and repeat.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Bounds","page":"Benders decomposition","title":"Bounds","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Due to convexity, we know that V_2(x) ge theta for all x. Therefore, the optimal objective value of V_1^K provides a valid lower bound on the objective value of the full problem. In addition, if we take a feasible solution for x from the first-stage problem, then c_1^top x + V_2(x) is a valid upper bound on the objective value of the full problem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Benders decomposition uses the lower and upper bounds to determine when it has found the global optimal solution.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Input-data","page":"Benders decomposition","title":"Input data","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As an example for this tutorial, we use the input data is from page 139 of Garfinkel, R. & Nemhauser, G. L. Integer programming. (Wiley, 1972).","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"c_1 = [1, 4]\nc_2 = [2, 3]\ndim_x = length(c_1)\ndim_y = length(c_2)\nb = [-2; -3]\nA_1 = [1 -3; -1 -3]\nA_2 = [1 -2; -1 -1]\nM = -1000;\nnothing #hide","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#benders_iterative","page":"Benders decomposition","title":"Iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"warning: Warning\nThis is a basic implementation for pedagogical purposes. We haven't discussed Benders feasibility cuts, or any of the computational tricks that are required to build a performant implementation for large-scale problems. See In-place iterative method for one improvement that helps computation time.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We start by formulating the first-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"model = Model(GLPK.Optimizer)\n@variable(model, x[1:dim_x] >= 0, Int)\n@variable(model, θ >= M)\n@objective(model, Min, c_1' * x + θ)\nprint(model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For the next step, we need a function that takes a first-stage candidate solution x and returns the optimal solution from the second-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function solve_subproblem(x)\n    model = Model(GLPK.Optimizer)\n    @variable(model, y[1:dim_y] >= 0)\n    con = @constraint(model, A_2 * y .<= b - A_1 * x)\n    @objective(model, Min, c_2' * y)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    return (obj = objective_value(model), y = value.(y), π = dual.(con))\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Note that solve_subproblem returns a NamedTuple of the objective value, the optimal primal solution for y, and the optimal dual solution for π.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We're almost ready for our optimization loop, but first, here's a helpful function for logging:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function print_iteration(k, args...)\n    f(x) = Printf.@sprintf(\"%12.4e\", x)\n    println(lpad(k, 9), \" \", join(f.(args), \" \"))\n    return\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"We also need to put a limit on the number of iterations before termination:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"MAXIMUM_ITERATIONS = 100","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"And a way to check if the lower and upper bounds are close-enough to terminate:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"ABSOLUTE_OPTIMALITY_GAP = 1e-6","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now we're ready to iterate Benders decomposition:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"println(\"Iteration  Lower Bound  Upper Bound          Gap\")\nfor k in 1:MAXIMUM_ITERATIONS\n    optimize!(model)\n    lower_bound = objective_value(model)\n    x_k = value.(x)\n    ret = solve_subproblem(x_k)\n    upper_bound = c_1' * x_k + ret.obj\n    gap = (upper_bound - lower_bound) / upper_bound\n    print_iteration(k, lower_bound, upper_bound, gap)\n    if gap < ABSOLUTE_OPTIMALITY_GAP\n        println(\"Terminating with the optimal solution\")\n        break\n    end\n    cut = @constraint(model, θ >= ret.obj + -ret.π' * A_1 * (x .- x_k))\n    @info \"Adding the cut $(cut)\"\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(model)\nx_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#Callback-method","page":"Benders decomposition","title":"Callback method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"The Iterative method section implemented Benders decomposition using a loop. In each iteration, we re-solved the first-stage subproblem to generate a candidate solution. However, modern MILP solvers such as CPLEX, Gurobi, and GLPK provide lazy constraint callbacks which allow us to add new cuts while the solver is running. This can be more efficient than an iterative method because we can avoid repeating work such as solving the root node of the first-stage MILP at each iteration.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"tip: Tip\nFor more information on callbacks, read the page Solver-independent callbacks.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As before, we construct the same first-stage subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"lazy_model = Model(GLPK.Optimizer)\n@variable(lazy_model, x[1:dim_x] >= 0, Int)\n@variable(lazy_model, θ >= M)\n@objective(lazy_model, Min, c_1' * x + θ)\nprint(lazy_model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"What differs is that we write a callback function instead of a loop:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"number_of_subproblem_solves = 0\nfunction my_callback(cb_data)\n    status = callback_node_status(cb_data, lazy_model)\n    if status != MOI.CALLBACK_NODE_STATUS_INTEGER\n        # Only add the constraint if `x` is an integer feasible solution\n        return\n    end\n    x_k = callback_value.(cb_data, x)\n    θ_k = callback_value(cb_data, θ)\n    global number_of_subproblem_solves += 1\n    ret = solve_subproblem(x_k)\n    if θ_k < (ret.obj - 1e-6)\n        # Only add the constraint if θ_k violates the constraint\n        cut = @build_constraint(θ >= ret.obj + -ret.π' * A_1 * (x .- x_k))\n        MOI.submit(lazy_model, MOI.LazyConstraint(cb_data), cut)\n    end\n    return\nend\n\nset_attribute(lazy_model, MOI.LazyConstraintCallback(), my_callback)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now when we optimize!, our callback is run:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(lazy_model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For this model, the callback algorithm required more solves of the subproblem:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"number_of_subproblem_solves","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"But for larger problems, you can expect the callback algorithm to be more efficient than the iterative algorithm.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"x_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/#In-place-iterative-method","page":"Benders decomposition","title":"In-place iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Our implementation of the iterative method has a problem: every time we need to solve the subproblem, we must rebuild it from scratch. This is expensive, and it can be the bottleneck in the solution process. We can improve our implementation by using re-using the subproblem between solves.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"First, we create our first-stage problem as usual:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"model = Model(GLPK.Optimizer)\n@variable(model, x[1:dim_x] >= 0, Int)\n@variable(model, θ >= M)\n@objective(model, Min, c_1' * x + θ)\nprint(model)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Then, instead of building the subproblem in a function, we build it once here:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"subproblem = Model(GLPK.Optimizer)\n@variable(subproblem, x_copy[1:dim_x])\n@variable(subproblem, y[1:dim_y] >= 0)\n@constraint(subproblem, A_1 * x_copy + A_2 * y .<= b)\n@objective(subproblem, Min, c_2' * y)\nprint(subproblem)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"This formulation is slightly different. We have included a copy of the x variables, x_copy, and used x_copy in the left-hand side of the constraints.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Our function to solve the subproblem is also slightly different. First, we need to fix the value of the x_copy variables to the value of x from the first-stage problem, and second, we compute the dual using the reduced_cost of x_copy, not the dual of the linear constraints:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"function solve_subproblem(model, x)\n    fix.(model[:x_copy], x)\n    optimize!(model)\n    @assert termination_status(model) == OPTIMAL\n    return (\n        obj = objective_value(model),\n        y = value.(model[:y]),\n        π = reduced_cost.(model[:x_copy]),\n    )\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Now we're ready to iterate our in-place Benders decomposition, but this time the cut computation is slightly different. Because we used reduced_cost on the x_copy variables, the value of π is a valid subgradient on the objective of subproblem with respect to x. Therefore, we don't need to multiply the duals by -A_1, and so our cut uses ret.π' instead of -ret.π' * A_1:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"println(\"Iteration  Lower Bound  Upper Bound          Gap\")\nfor k in 1:MAXIMUM_ITERATIONS\n    optimize!(model)\n    lower_bound = objective_value(model)\n    x_k = value.(x)\n    ret = solve_subproblem(subproblem, x_k)\n    upper_bound = c_1' * x_k + ret.obj\n    gap = (upper_bound - lower_bound) / upper_bound\n    print_iteration(k, lower_bound, upper_bound, gap)\n    if gap < ABSOLUTE_OPTIMALITY_GAP\n        println(\"Terminating with the optimal solution\")\n        break\n    end\n    cut = @constraint(model, θ >= ret.obj + ret.π' * (x .- x_k))\n    @info \"Adding the cut $(cut)\"\nend","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"Finally, we can obtain the optimal solution:","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimize!(model)\nx_optimal = value.(x)","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"optimal_ret = solve_subproblem(subproblem, x_optimal)\ny_optimal = optimal_ret.y","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"For larger problems, the benefit of re-using the same subproblem and not needing to multiply the duals by A_1 in the cut coefficient usually outweights the cost of needing a copy of the x variables in the subproblem.","category":"page"},{"location":"tutorials/algorithms/benders_decomposition/","page":"Benders decomposition","title":"Benders decomposition","text":"As a secondary benefit, because we no longer need an explicit representation of A_1 in the cut, we can build the model and subproblem formulations using arbitrary JuMP syntax; they do not need to be in matrix form.","category":"page"},{"location":"tutorials/nonlinear/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Nonlinear programs (NLPs) are a class of optimization problems in which some of the constraints or the objective function are nonlinear:","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n    min_x in mathbbR^n  f_0(x) \n    textst  l_j le f_j(x) le u_j  j = 1 ldots m \n     l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer nonlinear linear programs (MINLPs) are extensions of nonlinear programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/nonlinear/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"JuMP supports a range of nonlinear solvers; look for \"NLP\" in the list of Supported solvers. However, very few solvers support mixed-integer nonlinear linear programs. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"If the only nonlinearities in your model are quadratic terms (that is, multiplication between two decision variables), you can also use second-order cone solvers, which are indicated by \"SOCP.\" In most cases, these solvers are restricted to convex quadratic problems and will error if you pass a nonconvex quadratic function; however, Gurobi has the ability to solve nonconvex quadratic terms.","category":"page"},{"location":"tutorials/nonlinear/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/nonlinear/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nRocket Control\nOptimal control for a Space Shuttle reentry trajectory\nPortfolio optimization\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling nonlinear programs. Look here if you are stuck trying to formulate a problem as a nonlinear program.\nThe Computing Hessians is an advanced tutorial which explains how to compute the Hessian of the Lagrangian of a nonlinear program. This is useful only in particular cases.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"EditURL = \"https://github.com/jump-dev/Pajarito.jl/blob/4c7efa915bdc900b51d1c8290dc771403b11230a/README.md\"","category":"page"},{"location":"packages/Pajarito/#Pajarito.jl","page":"jump-dev/Pajarito.jl","title":"Pajarito.jl","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito is a mixed-integer convex programming (MICP) solver package written in Julia.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"MICP problems are convex except for restrictions that some variables take binary or integer values.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito solves MICP problems in conic form, by constructing sequential polyhedral outer approximations of the conic feasible set.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The underlying algorithm has theoretical finite-time convergence under reasonable assumptions.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito accesses state-of-the-art MILP solvers and continuous conic solvers through MathOptInterface.","category":"page"},{"location":"packages/Pajarito/#Pavito","page":"jump-dev/Pajarito.jl","title":"Pavito","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"For algorithms that use a derivative-based nonlinear programming (NLP) solver (for example, Ipopt) instead of a conic solver, use Pavito.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pavito is a convex mixed-integer nonlinear programming (convex MINLP) solver. Because Pavito relies on gradient cuts, it can fail near points of non-differentiability. Pajarito may be more robust than Pavito on non-smooth problems.","category":"page"},{"location":"packages/Pajarito/#License","page":"jump-dev/Pajarito.jl","title":"License","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito.jl is licensed under the MPL 2.0 license.","category":"page"},{"location":"packages/Pajarito/#Installation","page":"jump-dev/Pajarito.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Install Pajarito using Pkg.add:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"import Pkg\nPkg.add(\"Pajarito\")","category":"page"},{"location":"packages/Pajarito/#MIP-and-continuous-solvers","page":"jump-dev/Pajarito.jl","title":"MIP and continuous solvers","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The algorithm implemented by Pajarito itself is relatively simple, and most of the hard work is performed by the MIP outer approximation (OA) solver and the continuous conic solver.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Therefore, in addition to installing Pajarito, you must also install a mixed-integer linear programming solver and a continuous conic solver.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The performance of Pajarito depends on these two types of solvers.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The OA solver (typically a mixed-integer linear solver) is specified by the oa_solver option. You must first load the Julia package that provides this solver, for example, using Gurobi. The continuous conic solver is specified by the conic_solver option.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"See JuMP's list of supported solvers.","category":"page"},{"location":"packages/Pajarito/#Use-with-JuMP","page":"jump-dev/Pajarito.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"To use Pajarito with JuMP, use:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"using JuMP, Pajarito, HiGHS, Hypatia\nmodel = Model(\n    optimizer_with_attributes(\n        Pajarito.Optimizer,\n        \"oa_solver\" => optimizer_with_attributes(\n            HiGHS.Optimizer,\n            MOI.Silent() => true,\n            \"mip_feasibility_tolerance\" => 1e-8,\n            \"mip_rel_gap\" => 1e-6,\n        ),\n        \"conic_solver\" =>\n            optimizer_with_attributes(Hypatia.Optimizer, MOI.Silent() => true),\n    )\n)\nset_attribute(model, \"time_limit\", 60)","category":"page"},{"location":"packages/Pajarito/#Options","page":"jump-dev/Pajarito.jl","title":"Options","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"We list Pajarito's options below.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"verbose::Bool toggles printing\ntol_feas::Float64 is the feasibility tolerance for conic constraints\ntol_rel_gap::Float64 is the relative optimality gap tolerance\ntol_abs_gap::Float64 is the absolute optimality gap tolerance\ntime_limit::Float64 sets the time limit (in seconds)\niteration_limit::Int sets the iteration limit (for the iterative method)\nuse_iterative_method::Union{Nothing,Bool} toggles the iterative algorithm; if nothing is specified, Pajarito defaults to the OA-solver-driven (single tree) algorithm if lazy callbacks are supported by the OA solver\nuse_extended_form::Bool toggles the use of extended formulations for the second-order cone\nsolve_relaxation::Bool toggles solution of the continuous conic relaxation\nsolve_subproblems::Bool toggles solution of the continuous conic subproblems\nuse_init_fixed_oa::Bool toggles initial fixed OA cuts\noa_solver::Union{Nothing,MOI.OptimizerWithAttributes} is the OA solver\nconic_solver::Union{Nothing,MOI.OptimizerWithAttributes} is the conic solver","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito may require tuning of parameters to improve convergence.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"For example, it often helps to tighten the OA solver's integrality tolerance. OA solver and conic solver options must be specified directly to those solvers.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Note: if solve_subproblems is true, Pajarito usually returns a solution constructed from one of the conic solver's feasible solutions; since the conic solver is not subject to the same feasibility tolerances as the OA solver, Pajarito's solution will not necessarily satisfy tol_feas.","category":"page"},{"location":"packages/Pajarito/#Cone-interface","page":"jump-dev/Pajarito.jl","title":"Cone interface","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Pajarito has a generic cone interface (see the cones folder that allows the user to add support for new convex cones.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"To illustrate, in the experimental package PajaritoExtras we have extended Pajarito by adding support for several cones recognized by Hypatia.jl (a continuous conic solver with its own generic cone interface).","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"The examples folder of PajaritoExtras also contains many applied mixed-integer convex problems that are solved using Pajarito.","category":"page"},{"location":"packages/Pajarito/#Bug-reports-and-support","page":"jump-dev/Pajarito.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Please report any issues via the GitHub issue tracker.","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"All types of issues are welcome and encouraged; this includes bug reports, documentation typos, and feature requests. The Optimization (Mathematical) category on Discourse is appropriate for general discussion.","category":"page"},{"location":"packages/Pajarito/#References","page":"jump-dev/Pajarito.jl","title":"References","text":"","category":"section"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"If you find Pajarito useful in your work, we kindly request that you cite the following paper (arXiv preprint), which is recommended reading for advanced users:","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"@article{CoeyLubinVielma2020,\n    title={Outer approximation with conic certificates for mixed-integer convex problems},\n    author={Coey, Chris and Lubin, Miles and Vielma, Juan Pablo},\n    journal={Mathematical Programming Computation},\n    volume={12},\n    number={2},\n    pages={249--293},\n    year={2020},\n    publisher={Springer}\n}","category":"page"},{"location":"packages/Pajarito/","page":"jump-dev/Pajarito.jl","title":"jump-dev/Pajarito.jl","text":"Note this paper describes a legacy MathProgBase version of Pajarito, which is available on the mathprogbase branch of this repository. Starting with version v0.8.0, Pajarito supports MathOptInterface instead of MathProgBase.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/models.md\"","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/models/#Models","page":"Models","title":"Models","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The most significant part of MOI is the definition of the model API that is used to specify an instance of an optimization problem (for example, by adding variables and constraints). Objects that implement the model API must inherit from the ModelLike abstract type.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Notably missing from the model API is the method to solve an optimization problem. ModelLike objects may store an instance (for example, in memory or backed by a file format) without being linked to a particular solver. In addition to the model API, MOI defines AbstractOptimizer and provides methods to solve the model and interact with solutions. See the Solutions section for more details.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"info: Info\nThroughout the rest of the manual, model is used as a generic ModelLike, and optimizer is used as a generic AbstractOptimizer.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"tip: Tip\nMOI does not export functions, but for brevity we often omit qualifying names with the MOI module. Best practice is to haveimport MathOptInterface as MOIand prefix all MOI methods with MOI. in user code. If a name is also available in base Julia, we always explicitly use the module prefix, for example, with MOI.get.","category":"page"},{"location":"moi/manual/models/#Attributes","page":"Models","title":"Attributes","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Attributes are properties of the model that can be queried and modified. These include constants such as the number of variables in a model NumberOfVariables), and properties of variables and constraints such as the name of a variable (VariableName).","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"There are four types of attributes:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Model attributes (subtypes of AbstractModelAttribute) refer to properties of a model.\nOptimizer attributes (subtypes of AbstractOptimizerAttribute) refer to properties of an optimizer.\nConstraint attributes (subtypes of AbstractConstraintAttribute) refer to properties of an individual constraint.\nVariable attributes (subtypes of AbstractVariableAttribute) refer to properties of an individual variable.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Some attributes are values that can be queried by the user but not modified, while other attributes can be modified by the user.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"All interactions with attributes occur through the get and set functions.","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"Consult the docstrings of each attribute for information on what it represents.","category":"page"},{"location":"moi/manual/models/#ModelLike-API","page":"Models","title":"ModelLike API","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The following attributes are available:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"ListOfConstraintAttributesSet\nListOfConstraintIndices\nListOfConstraintTypesPresent\nListOfModelAttributesSet\nListOfVariableAttributesSet\nListOfVariableIndices\nNumberOfConstraints\nNumberOfVariables\nName\nObjectiveFunction\nObjectiveFunctionType\nObjectiveSense","category":"page"},{"location":"moi/manual/models/#AbstractOptimizer-API","page":"Models","title":"AbstractOptimizer API","text":"","category":"section"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"The following attributes are available:","category":"page"},{"location":"moi/manual/models/","page":"Models","title":"Models","text":"DualStatus\nPrimalStatus\nRawStatusString\nResultCount\nTerminationStatus\nBarrierIterations\nDualObjectiveValue\nNodeCount\nNumberOfThreads\nObjectiveBound\nObjectiveValue\nRelativeGap\nRawOptimizerAttribute\nRawSolver\nSilent\nSimplexIterations\nSolverName\nSolverVersion\nSolveTimeSec\nTimeLimitSec\nObjectiveLimit","category":"page"},{"location":"packages/NLopt/#NLopt.jl","page":"JuliaOpt/NLopt.jl","title":"NLopt.jl","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"NLopt.jl is a wrapper for the NLopt library.","category":"page"},{"location":"packages/NLopt/#License","page":"JuliaOpt/NLopt.jl","title":"License","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"NLopt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"The underlying solver, stevengj/nlopt, is licensed under the LGPL v3.0 license.","category":"page"},{"location":"packages/NLopt/#Installation","page":"JuliaOpt/NLopt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"Install NLopt.jl using the Julia package manager:","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"import Pkg\nPkg.add(\"NLopt\")","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"In addition to installing the NLopt.jl package, this will also download and install the NLopt binaries. You do not need to install NLopt separately.","category":"page"},{"location":"packages/NLopt/#Use-with-JuMP","page":"JuliaOpt/NLopt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"You can use NLopt with JuMP as follows:","category":"page"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"using JuMP, NLopt\nmodel = Model(NLopt.Optimizer)\nset_attribute(model, \"algorithm\", :LD_MMA)","category":"page"},{"location":"packages/NLopt/#Options","page":"JuliaOpt/NLopt.jl","title":"Options","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"The algorithm attribute is required. The value must be one of the supported NLopt algorithms.","category":"page"},{"location":"packages/NLopt/#Documentation","page":"JuliaOpt/NLopt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/NLopt/","page":"JuliaOpt/NLopt.jl","title":"JuliaOpt/NLopt.jl","text":"For more details, see the NLopt.jl README or the NLopt documentation.","category":"page"},{"location":"tutorials/linear/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Linear programs (LPs) are a fundamental class of optimization problems of the form:","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"beginalign\n    min_x in mathbbR^n  sumlimits_i=1^n c_i x_i \n    textst  l_j le sumlimits_i=1^n a_ij x_i le u_j  j = 1 ldots m \n     l_i le x_i le u_i  i = 1 ldots n\nendalign","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"The most important thing to note is that all terms are of the form coefficient * variable, and that there are no nonlinear terms or multiplications between variables.","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Mixed-integer linear programs (MILPs) are extensions of linear programs in which some (or all) of the decision variables take discrete values.","category":"page"},{"location":"tutorials/linear/introduction/#How-to-choose-a-solver","page":"Introduction","title":"How to choose a solver","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Almost all solvers support linear programs; look for \"LP\" in the list of Supported solvers. However, fewer solvers support mixed-integer linear programs. Solvers supporting discrete variables start with \"(MI)\" in the list of Supported solvers.","category":"page"},{"location":"tutorials/linear/introduction/#How-these-tutorials-are-structured","page":"Introduction","title":"How these tutorials are structured","text":"","category":"section"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"tutorials/linear/introduction/","page":"Introduction","title":"Introduction","text":"The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.\nThe diet problem\nThe cannery problem\nThe facility location problem\nFinancial modeling problems\nNetwork flow problems\nN-Queens\nSudoku\nThe Tips and tricks tutorial contains a number of helpful reformulations and tricks you can use when modeling linear programs. Look here if you are stuck trying to formulate a problem as a linear program.\nThe Sensitivity analysis of a linear program tutorial explains how to create sensitivity reports like those produced by the Excel Solver.\nThe Callbacks tutorial explains how to write a variety of solver-independent callbacks. Look here if you want to write a callback.\nThe remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"EditURL = \"tips_and_tricks.jl\"","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/#nonlinear_tips_and_tricks","page":"Tips and tricks","title":"Tips and tricks","text":"","category":"section"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This example collates some tips and tricks you can use when formulating nonlinear programs. It uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"using JuMP\nimport Ipopt\nimport Test","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/#User-defined-operators-with-vector-outputs","page":"Tips and tricks","title":"User-defined operators with vector outputs","text":"","category":"section"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"A common situation is to have a user-defined operator like the following that returns multiple outputs (we define function_calls to keep track of how many times we call this method):","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"function_calls = 0\nfunction foo(x, y)\n    global function_calls += 1\n    common_term = x^2 + y^2\n    term_1 = sqrt(1 + common_term)\n    term_2 = common_term\n    return term_1, term_2\nend","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"For example, the first term might be used in the objective, and the second term might be used in a constraint, and often they share work that is expensive to evaluate.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"This is a problem for JuMP, because it requires user-defined operators to return a single number. One option is to define two separate functions, the first returning the first argument, and the second returning the second argument.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"foo_1(x, y) = foo(x, y)[1]\nfoo_2(x, y) = foo(x, y)[2]","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"However, if the common term is expensive to compute, this approach is wasteful because it will evaluate the expensive term twice. Let's have a look at how many times we evaluate x^2 + y^2 during a solve:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:2] >= 0, start = 0.1)\n@operator(model, op_foo_1, 2, foo_1)\n@operator(model, op_foo_2, 2, foo_2)\n@objective(model, Max, op_foo_1(x[1], x[2]))\n@constraint(model, op_foo_2(x[1], x[2]) <= 2)\nfunction_calls = 0\noptimize!(model)\nTest.@test objective_value(model) ≈ √3 atol = 1e-4\nTest.@test value.(x) ≈ [1.0, 1.0] atol = 1e-4\nprintln(\"Naive approach: function calls = $(function_calls)\")","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"An alternative approach is to use memoization, which uses a cache to store the result of function evaluations. We can write a memoization function as follows:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"\"\"\"\n    memoize(foo::Function, n_outputs::Int)\n\nTake a function `foo` and return a vector of length `n_outputs`, where element\n`i` is a function that returns the equivalent of `foo(x...)[i]`.\n\nTo avoid duplication of work, cache the most-recent evaluations of `foo`.\nBecause `foo_i` is auto-differentiated with ForwardDiff, our cache needs to\nwork when `x` is a `Float64` and a `ForwardDiff.Dual`.\n\"\"\"\nfunction memoize(foo::Function, n_outputs::Int)\n    last_x, last_f = nothing, nothing\n    last_dx, last_dfdx = nothing, nothing\n    function foo_i(i, x::T...) where {T<:Real}\n        if T == Float64\n            if x !== last_x\n                last_x, last_f = x, foo(x...)\n            end\n            return last_f[i]::T\n        else\n            if x !== last_dx\n                last_dx, last_dfdx = x, foo(x...)\n            end\n            return last_dfdx[i]::T\n        end\n    end\n    return [(x...) -> foo_i(i, x...) for i in 1:n_outputs]\nend","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Let's see how it works. First, construct the memoized versions of foo:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"memoized_foo = memoize(foo, 2)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Now try evaluating the first element of memoized_foo.","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"function_calls = 0\nmemoized_foo[1](1.0, 1.0)\nprintln(\"function_calls = \", function_calls)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"As expected, this evaluated the function once. However, if we call the function again, we hit the cache instead of needing to re-compute foo and function_calls is still 1!","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"memoized_foo[1](1.0, 1.0)\nprintln(\"function_calls = \", function_calls)","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Now let's see how this works during a real solve:","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:2] >= 0, start = 0.1)\n@operator(model, op_foo_1, 2, memoized_foo[1])\n@operator(model, op_foo_2, 2, memoized_foo[2])\n@objective(model, Max, op_foo_1(x[1], x[2]))\n@constraint(model, op_foo_2(x[1], x[2]) <= 2)\nfunction_calls = 0\noptimize!(model)\nTest.@test objective_value(model) ≈ √3 atol = 1e-4\nTest.@test value.(x) ≈ [1.0, 1.0] atol = 1e-4\nprintln(\"Memoized approach: function_calls = $(function_calls)\")","category":"page"},{"location":"tutorials/nonlinear/tips_and_tricks/","page":"Tips and tricks","title":"Tips and tricks","text":"Compared to the naive approach, the memoized approach requires half as many function evaluations.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"EditURL = \"rocket_control.jl\"","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Rocket-Control","page":"Rocket Control","title":"Rocket Control","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial was originally contributed by Iain Dunning.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The purpose of this tutorial is to demonstrate how to setup and solve a nonlinear optimization problem.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The example is an optimal control problem of a nonlinear rocket.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"using JuMP\nimport Ipopt\nimport Plots","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Overview","page":"Rocket Control","title":"Overview","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Our goal is to maximize the final altitude of a vertically launched rocket.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We can control the thrust of the rocket, and must take account of the rocket mass, fuel consumption rate, gravity, and aerodynamic drag.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Let us consider the basic description of the model (for the full description, including parameters for the rocket, see COPS3).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"There are three state variables in our model:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Velocity: x_v(t)\nAltitude: x_h(t)\nMass of rocket and remaining fuel, x_m(t)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"and a single control variable:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Thrust: u_t(t).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"There are three equations that control the dynamics of the rocket:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Rate of ascent: fracd x_hdt = x_v\nAcceleration: fracd x_vdt = fracu_t - D(x_h x_v)x_m - g(x_h)\nRate of mass loss: fracd x_mdt = -fracu_tc","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"where drag D(x_h x_v) is a function of altitude and velocity, gravity g(x_h) is a function of altitude, and c is a constant.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"These forces are defined as:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"D(x_h x_v) = D_c cdot x_v^2 cdot e^-h_c left( fracx_h-x_h(0)x_h(0) right)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"and g(x_h) = g_0 cdot left( fracx_h(0)x_h right)^2","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We use a discretized model of time, with a fixed number of time steps, T.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Our goal is thus to maximize x_h(T).","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Data","page":"Rocket Control","title":"Data","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"All parameters in this model have been normalized to be dimensionless, and they are taken from COPS3.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"h_0 = 1                      # Initial height\nv_0 = 0                      # Initial velocity\nm_0 = 1.0                    # Initial mass\nm_T = 0.6                    # Final mass\ng_0 = 1                      # Gravity at the surface\nh_c = 500                    # Used for drag\nc = 0.5 * sqrt(g_0 * h_0)    # Thrust-to-fuel mass\nD_c = 0.5 * 620 * m_0 / g_0  # Drag scaling\nu_t_max = 3.5 * g_0 * m_0    # Maximum thrust\nT_max = 0.2                  # Number of seconds\nT = 1_000                    # Number of time steps\nΔt = 0.2 / T;                # Time per discretized step\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#JuMP-formulation","page":"Rocket Control","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"First, we create a model and choose an optimizer. Since this is a nonlinear program, we need to use a nonlinear solver like Ipopt. We cannot use a linear solver like HiGHS.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Next, we create our state and control variables, which are each indexed by t. It is good practice for nonlinear programs to always provide a starting solution for each variable.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"@variable(model, x_v[1:T] >= 0, start = v_0)           # Velocity\n@variable(model, x_h[1:T] >= 0, start = h_0)           # Height\n@variable(model, x_m[1:T] >= m_T, start = m_0)         # Mass\n@variable(model, 0 <= u_t[1:T] <= u_t_max, start = 0); # Thrust\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"We implement boundary conditions by fixing variables to values.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"fix(x_v[1], v_0; force = true)\nfix(x_h[1], h_0; force = true)\nfix(x_m[1], m_0; force = true)\nfix(u_t[T], 0.0; force = true)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The objective is to maximize altitude at end of time of flight.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"@objective(model, Max, x_h[T])","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Forces are defined as functions:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"D(x_h, x_v) = D_c * x_v^2 * exp(-h_c * (x_h - h_0) / h_0)\ng(x_h) = g_0 * (h_0 / x_h)^2","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"The dynamical equations are implemented as constraints.","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"ddt(x::Vector, t::Int) = (x[t] - x[t-1]) / Δt\n@constraint(model, [t in 2:T], ddt(x_h, t) == x_v[t-1])\n@constraint(\n    model,\n    [t in 2:T],\n    ddt(x_v, t) == (u_t[t-1] - D(x_h[t-1], x_v[t-1])) / x_m[t-1] - g(x_h[t-1]),\n)\n@constraint(model, [t in 2:T], ddt(x_m, t) == -u_t[t-1] / c);\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Now we optimize the model and check that we found a solution:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Finally, we plot the solution:","category":"page"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"function plot_trajectory(y; kwargs...)\n    return Plots.plot(\n        (1:T) * Δt,\n        value.(y);\n        xlabel = \"Time (s)\",\n        legend = false,\n        kwargs...,\n    )\nend\n\nPlots.plot(\n    plot_trajectory(x_h; ylabel = \"Altitude\"),\n    plot_trajectory(x_m; ylabel = \"Mass\"),\n    plot_trajectory(x_v; ylabel = \"Velocity\"),\n    plot_trajectory(u_t; ylabel = \"Thrust\");\n    layout = (2, 2),\n)","category":"page"},{"location":"tutorials/nonlinear/rocket_control/#Next-steps","page":"Rocket Control","title":"Next steps","text":"","category":"section"},{"location":"tutorials/nonlinear/rocket_control/","page":"Rocket Control","title":"Rocket Control","text":"Experiment with different values for the constants. How does the solution change? In particular, what happens if you change T_max?\nThe dynamical equations use rectangular integration for the right-hand side terms. Modify the equations to use the Trapezoidal rule instead. (As an example, x_v[t-1] would become 0.5 * (x_v[t-1] + x_v[t]).) Is there a difference?","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"EditURL = \"querying_hessians.jl\"","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#Computing-Hessians","page":"Computing Hessians","title":"Computing Hessians","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"The purpose of this tutorial is to demonstrate how to compute the Hessian of the Lagrangian of a nonlinear program.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"warning: Warning\nThis is an advanced tutorial that interacts with the low-level nonlinear interface of MathOptInterface.By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Given a nonlinear program:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"beginalign\n min_x in mathbbR^n  f(x) \n textst  l le g_i(x) le u\nendalign","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"the Hessian of the Lagrangian is computed as:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"H(x sigma mu) = sigmanabla^2 f(x) + sum_i=1^m mu_i nabla^2 g_i(x)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"where x is a primal point, sigma is a scalar (typically 1), and mu is a vector of weights corresponding to the Lagrangian dual of the constraints.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"using JuMP\nimport Ipopt\nimport LinearAlgebra\nimport Random\nimport SparseArrays","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#The-basic-model","page":"Computing Hessians","title":"The basic model","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"To demonstrate how to interact with the lower-level nonlinear interface, we need an example model. The exact model isn't important; we use the model from The Rosenbrock function tutorial, with some additional constraints to demonstrate various features of the lower-level interface.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:2], start = -i)\n@constraint(model, g_1, x[1]^2 <= 1)\n@constraint(model, g_2, (x[1] + x[2])^2 <= 2)\n@objective(model, Min, (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2)\noptimize!(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#The-analytic-solution","page":"Computing Hessians","title":"The analytic solution","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"With a little work, it is possible to analytically derive the correct hessian:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function analytic_hessian(x, σ, μ)\n    g_1_H = [2.0 0.0; 0.0 0.0]\n    g_2_H = [2.0 2.0; 2.0 2.0]\n    f_H = zeros(2, 2)\n    f_H[1, 1] = 2.0 + 1200.0 * x[1]^2 - 400.0 * x[2]\n    f_H[1, 2] = f_H[2, 1] = -400.0 * x[1]\n    f_H[2, 2] = 200.0\n    return σ * f_H + μ' * [g_1_H, g_2_H]\nend","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Here are various points:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [0, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [1, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 0, [0, 1])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian([1, 1], 1, [0, 0])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/#Create-a-nonlinear-model","page":"Computing Hessians","title":"Create a nonlinear model","text":"","category":"section"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"JuMP delegates automatic differentiation to the MOI.Nonlinear submodule. Therefore, to compute the Hessian of the Lagrangian, we need to create a MOI.Nonlinear.Model object:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"rows = Any[]\nnlp = MOI.Nonlinear.Model()\nfor (F, S) in list_of_constraint_types(model)\n    if F <: VariableRef\n        continue  # Skip variable bounds\n    end\n    for ci in all_constraints(model, F, S)\n        push!(rows, ci)\n        object = constraint_object(ci)\n        MOI.Nonlinear.add_constraint(nlp, object.func, object.set)\n    end\nend\nMOI.Nonlinear.set_objective(nlp, objective_function(model))\nnlp","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"It is important that we save the constraint indices in a vector rows, so that we know the order of the constraints in the nonlinear model.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Next, we need to convert our model into an MOI.Nonlinear.Evaluator, specifying an automatic differentiation backend. In this case, we use MOI.Nonlinear.SparseReverseMode:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"evaluator = MOI.Nonlinear.Evaluator(\n    nlp,\n    MOI.Nonlinear.SparseReverseMode(),\n    index.(all_variables(model)),\n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Before computing anything with the evaluator, we need to initialize it. Use MOI.features_available to see what we can query:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.features_available(evaluator)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Consult the MOI documentation for specifics, but to obtain the Hessian matrix, we need to initialize :Hess:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.initialize(evaluator, [:Hess])","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI represents the Hessian as a sparse matrix. Get the sparsity pattern as follows:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"hessian_sparsity = MOI.hessian_lagrangian_structure(evaluator)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"The sparsity pattern has a few properties of interest:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Each element (i, j) indicates a structural non-zero in row i and column j\nThe list may contain duplicates, in which case we should add the values together\nThe list does not need to be sorted\nThe list may contain any mix of lower- or upper-triangular indices","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"This format matches Julia's sparse-triplet form of a SparseArray, so we can convert from the sparse Hessian representation to a Julia SparseArray as follows:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"I = [i for (i, _) in hessian_sparsity]\nJ = [j for (_, j) in hessian_sparsity]\nV = zeros(length(hessian_sparsity))\nn = num_variables(model)\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Of course, knowing where the zeros are isn't very interesting. We really want to compute the value of the Hessian matrix at a point.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.eval_hessian_lagrangian(evaluator, V, ones(n), 1.0, ones(length(rows)))\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"In practice, we often want to compute the value of the hessian at the optimal solution.","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"First, we compute the primal solution. To do so, we need a vector of the variables in the order that they were passed to the solver:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"x = all_variables(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Here x[1] is the variable that corresponds to column 1, and so on. Here's the optimal primal solution:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"x_optimal = value.(x)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Next, we need the optimal dual solution associated with the nonlinear constraints (this is where it is important to record the order of the constraints as we added them to nlp):","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"y_optimal = dual.(rows)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Now we can compute the Hessian at the optimal primal-dual point:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"MOI.eval_hessian_lagrangian(evaluator, V, x_optimal, 1.0, y_optimal)\nH = SparseArrays.sparse(I, J, V, n, n)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"However, this Hessian isn't quite right because it isn't symmetric. We can fix this by filling in the appropriate off-diagonal terms:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function fill_off_diagonal(H)\n    ret = H + H'\n    row_vals = SparseArrays.rowvals(ret)\n    non_zeros = SparseArrays.nonzeros(ret)\n    for col in 1:size(ret, 2)\n        for i in SparseArrays.nzrange(ret, col)\n            if col == row_vals[i]\n                non_zeros[i] /= 2\n            end\n        end\n    end\n    return ret\nend\n\nfill_off_diagonal(H)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"Putting everything together:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"function compute_optimal_hessian(model::Model)\n    rows = Any[]\n    nlp = MOI.Nonlinear.Model()\n    for (F, S) in list_of_constraint_types(model)\n        for ci in all_constraints(model, F, S)\n            push!(rows, ci)\n            object = constraint_object(ci)\n            MOI.Nonlinear.add_constraint(nlp, object.func, object.set)\n        end\n    end\n    MOI.Nonlinear.set_objective(nlp, objective_function(model))\n    x = all_variables(model)\n    backend = MOI.Nonlinear.SparseReverseMode()\n    evaluator = MOI.Nonlinear.Evaluator(nlp, backend, index.(x))\n    MOI.initialize(evaluator, [:Hess])\n    hessian_sparsity = MOI.hessian_lagrangian_structure(evaluator)\n    I = [i for (i, _) in hessian_sparsity]\n    J = [j for (_, j) in hessian_sparsity]\n    V = zeros(length(hessian_sparsity))\n    MOI.eval_hessian_lagrangian(evaluator, V, value.(x), 1.0, dual.(rows))\n    H = SparseArrays.sparse(I, J, V, length(x), length(x))\n    return Matrix(fill_off_diagonal(H))\nend\n\nH_star = compute_optimal_hessian(model)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"If we compare our solution against the analytical solution:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"analytic_hessian(value.(x), 1.0, dual.([g_1, g_2]))","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"If we look at the eigenvalues of the Hessian:","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"LinearAlgebra.eigvals(H_star)","category":"page"},{"location":"tutorials/nonlinear/querying_hessians/","page":"Computing Hessians","title":"Computing Hessians","text":"we see that they are all positive. Therefore, the Hessian is positive definite, and so the solution found by Ipopt is a local minimizer.","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"EditURL = \"https://github.com/jump-dev/SumOfSquares.jl/blob/v0.7.3/README.md\"","category":"page"},{"location":"packages/SumOfSquares/#SumOfSquares.jl","page":"jump-dev/SumOfSquares.jl","title":"SumOfSquares.jl","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"(Image: Build Status) (Image: codecov) (Image: )","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"SumOfSquares.jl is a JuMP extension that, when used in conjunction with MultivariatePolynomial and PolyJuMP, implements a sum of squares reformulation for polynomial optimization.","category":"page"},{"location":"packages/SumOfSquares/#License","page":"jump-dev/SumOfSquares.jl","title":"License","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"SumOfSquares.jl is licensed under the MIT license.","category":"page"},{"location":"packages/SumOfSquares/#Installation","page":"jump-dev/SumOfSquares.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Install SumOfSquares using Pkg.add:","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"import Pkg\nPkg.add(\"SumOfSquares\")","category":"page"},{"location":"packages/SumOfSquares/#Documentation","page":"jump-dev/SumOfSquares.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See https://jump.dev/SumOfSquares.jl/stable for the most recently tagged version of the documentation.","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See https://jump.dev/SumOfSquares.jl/dev for the in-development version of the documentation.","category":"page"},{"location":"packages/SumOfSquares/#Presentations","page":"jump-dev/SumOfSquares.jl","title":"Presentations","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Some presentations on, or using, SumOfSquares (see blegat/SumOfSquaresSlides for the source code of the presentations):","category":"page"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"Benoît Legat at JuMP-dev 2023 [Slides]\nBenoît Legat, Marek Kaluba and Tillmann Weisser at INFORMS 2022 [Slides]\nBenoît Legat at POEMA Learning Week 2\nBenoît Legat, Marek Kaluba and Tillmann Weisser at JuMP-dev 2021 [Slides]\nBenoît Legat at INFORMS 2020 [Slides]\nTillmann Weisser, Benoît Legat, Chris Coey, Lea Kapelevich and Juan Pablo Vielma at JuliaCon 2019 [Slides] [Video]\nBenoît Legat at CNLS 2019 [Slides]\nBenoît Legat at EURO 2019 [Slides]\nBenoît Legat at juliaday Nantes 2019 [Slides]\nBenoît Legat at Summer School on Numerical Computing in Algebraic Geometry 2018 [Poster]\nBenoît Legat at The First Annual JuMP-dev Workshop 2017 [Slides] [Video]\nJoey Huchette at SIAM Opt 2017","category":"page"},{"location":"packages/SumOfSquares/#Citing","page":"jump-dev/SumOfSquares.jl","title":"Citing","text":"","category":"section"},{"location":"packages/SumOfSquares/","page":"jump-dev/SumOfSquares.jl","title":"jump-dev/SumOfSquares.jl","text":"See CITATION.bib.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"EditURL = \"cutting_stock_column_generation.jl\"","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Column-generation","page":"Column generation","title":"Column generation","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The purpose of this tutorial is to demonstrate the column generation algorithm. As an example, it solves the Cutting stock problem.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport Plots\nimport SparseArrays","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Background","page":"Column generation","title":"Background","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The cutting stock problem is about cutting large rolls of paper into smaller pieces.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We denote the set of possible sized pieces that a roll can be cut into by iin 1ldotsI. Each piece i has a width, w_i, and a demand, d_i. The width of the large roll is W.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our objective is to minimize the number of rolls needed to meet all demand.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Here's the data that we are going to use in this tutorial:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"struct Piece\n    w::Float64\n    d::Int\nend\n\nstruct Data\n    pieces::Vector{Piece}\n    W::Float64\nend\n\nfunction Base.show(io::IO, d::Data)\n    println(io, \"Data for the cutting stock problem:\")\n    println(io, \"  W = $(d.W)\")\n    println(io, \"with pieces:\")\n    println(io, \"   i   w_i d_i\")\n    println(io, \"  ------------\")\n    for (i, p) in enumerate(d.pieces)\n        println(io, lpad(i, 4), \" \", lpad(p.w, 5), \" \", lpad(p.d, 3))\n    end\n    return\nend\n\nfunction get_data()\n    data = [\n        75.0 38\n        75.0 44\n        75.0 30\n        75.0 41\n        75.0 36\n        53.8 33\n        53.0 36\n        51.0 41\n        50.2 35\n        32.2 37\n        30.8 44\n        29.8 49\n        20.1 37\n        16.2 36\n        14.5 42\n        11.0 33\n        8.6 47\n        8.2 35\n        6.6 49\n        5.1 42\n    ]\n    return Data([Piece(data[i, 1], data[i, 2]) for i in axes(data, 1)], 100.0)\nend\n\ndata = get_data()","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Mathematical-formulation","page":"Column generation","title":"Mathematical formulation","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"To formulate the cutting stock problem as a mixed-integer linear program, we assume that there is a set of large rolls j=1ldotsJ to use. Then, we introduce two classes of decision variables:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_ij ge 0 textinteger  forall i=1ldotsI j=1ldotsJ\ny_j in 0 1 forall j=1ldotsJ","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"y_j is a binary variable that indicates if we use roll j, and x_ij counts how many pieces of size i that we cut from roll j.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our mixed-integer linear program is therefore:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    min  sumlimits_j=1^J y_j \n    textst  sumlimits_i=1^N w_i x_ij le W y_j  forall j=1ldotsJ \n          sumlimits_j=1^J x_ij ge d_i  forall i=1ldotsI \n          x_ij ge 0  forall i=1ldotsN j=1ldotsJ \n          x_ij in mathbbZ  forall i=1ldotsI j=1ldotsJ \n          y_j in 0 1  forall j=1ldotsJ \nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The objective is to minimize the number of rolls that we use, and the two constraints ensure that we respect the total width of each large roll and that we satisfy demand exactly.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The JuMP formulation of this model is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"I = length(data.pieces)\nJ = 1_000  # Some large number\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:I, 1:J] >= 0, Int)\n@variable(model, y[1:J], Bin)\n@objective(model, Min, sum(y))\n@constraint(model, [i in 1:I], sum(x[i, :]) >= data.pieces[i].d)\n@constraint(\n    model,\n    [j in 1:J],\n    sum(data.pieces[i].w * x[i, j] for i in 1:I) <= data.W * y[j],\n);\nnothing #hide","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Unfortunately, we can't solve this formulation for realistic instances because it takes a very long time to solve. (Try removing the time limit.)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"set_time_limit_sec(model, 5.0)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"However, there is a formulation that solves much faster, and that is to use a column generation scheme.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Column-generation-theory","page":"Column generation","title":"Column generation theory","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"The key insight for column generation is to recognize that feasible columns in the x matrix of variables encode cutting patterns.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For example, if we look only at the roll j=1, then a feasible solution is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_11 = 1 (1 unit of piece #1)\nx_131 = 1 (1 unit of piece #13)\nAll other x_i1 = 0","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Another solution is","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"x_201 = 19 (19 unit of piece #20)\nAll other x_i1 = 0","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Cutting patterns like x_11 = 1 and x_21 = 1 are infeasible because the combined length is greater than W.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Since there are a finite number of ways that we could cut a roll into a valid cutting pattern, we could create a set of all possible cutting patterns p = 1ldotsP, with data a_ip indicating how many units of piece i we cut in pattern p. Then, we can formulate our mixed-integer linear program as:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    min  sumlimits_p=1^P x_p \n    textst  sumlimits_p=1^P a_ip x_p ge d_i  forall i=1ldotsI \n          x_p ge 0  forall p=1ldotsP \n          x_p in mathbbZ  forall p=1ldotsP\nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Unfortunately, there will be a very large number of these patterns, so it is often intractable to enumerate all columns p=1ldotsP.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Column generation is an iterative algorithm that starts with a small set of initial patterns, and then cleverly chooses new columns to add to the main MILP so that we find the optimal solution without having to enumerate every column.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Choosing-the-initial-set-of-patterns","page":"Column generation","title":"Choosing the initial set of patterns","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For the initial set of patterns, we create a trivial cutting pattern which cuts as many units of piece i as will fit.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"patterns = map(1:I) do i\n    n_pieces = floor(Int, data.W / data.pieces[i].w)\n    return SparseArrays.sparsevec([i], [n_pieces], I)\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We can visualize the patterns as follows:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"\"\"\"\n    cutting_locations(data::Data, pattern::SparseArrays.SparseVector)\n\nA function which returns a vector of the locations along the roll at which to\ncut in order to produce pattern `pattern`.\n\"\"\"\nfunction cutting_locations(data::Data, pattern::SparseArrays.SparseVector)\n    locations = Float64[]\n    offset = 0.0\n    for (i, c) in zip(SparseArrays.findnz(pattern)...)\n        for _ in 1:c\n            offset += data.pieces[i].w\n            push!(locations, offset)\n        end\n    end\n    return locations\nend\n\nfunction plot_patterns(data::Data, patterns)\n    plot = Plots.bar(;\n        xlims = (0, length(patterns) + 1),\n        ylims = (0, data.W),\n        xlabel = \"Pattern\",\n        ylabel = \"Roll length\",\n    )\n    for (i, p) in enumerate(patterns)\n        locations = cutting_locations(data, p)\n        Plots.bar!(\n            plot,\n            fill(i, length(locations)),\n            reverse(locations);\n            bar_width = 0.6,\n            label = false,\n            color = \"#90caf9\",\n        )\n    end\n    return plot\nend\n\nplot_patterns(data, patterns)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#The-base-problem","page":"Column generation","title":"The base problem","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Using the initial set of patterns, we can create and optimize our base model:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:length(patterns)] >= 0, Int)\n@objective(model, Min, sum(x))\n@constraint(model, demand[i in 1:I], patterns[i]' * x >= data.pieces[i].d)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This solution requires 421 rolls. This solution is sub-optimal because the model does not contain the full set of possible patterns.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"How do we find a new column that leads to an improved solution?","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Choosing-new-columns","page":"Column generation","title":"Choosing new columns","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Column generation chooses a new column by relaxing the integrality constraint on x and looking at the dual variable pi_i associated with demand constraint i.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"For example, the dual of demand[13] is:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"unset_integer.(x)\noptimize!(model)\nπ_13 = dual(demand[13])","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Using the economic interpretation of the dual variable, we can say that a one unit increase in demand for piece i will cost an extra pi_i rolls. Alternatively, we can say that a one unit increase in the left-hand side (for example, due to a new cutting pattern) will save us pi_i rolls. Therefore, we want a new column that maximizes the savings associated with the dual variables, while respecting the total width of the roll:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"beginalign\n    max  sumlimits_i=1^I pi_i y_i \n    textst  sumlimits_i=1^I w_i y_i le W \n          y_i ge 0  forall i=1ldotsI \n          y_i in mathbbZ  forall i=1ldotsI \nendalign","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If this problem, called the pricing problem, has an objective value greater than 1, then we estimate than adding y as the coefficients of a new column will decrease the objective by more than the cost of an extra roll.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Here is code to solve the pricing problem:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"function solve_pricing(data::Data, π::Vector{Float64})\n    I = length(π)\n    model = Model(HiGHS.Optimizer)\n    set_silent(model)\n    @variable(model, y[1:I] >= 0, Int)\n    @constraint(model, sum(data.pieces[i].w * y[i] for i in 1:I) <= data.W)\n    @objective(model, Max, sum(π[i] * y[i] for i in 1:I))\n    optimize!(model)\n    number_of_rolls_saved = objective_value(model)\n    if number_of_rolls_saved > 1 + 1e-8\n        # Benefit of pattern is more than the cost of a new roll plus some\n        # tolerance\n        return SparseArrays.sparse(round.(Int, value.(y)))\n    end\n    return nothing\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If we solve the pricing problem with an artificial dual vector:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solve_pricing(data, [1.0 / i for i in 1:I])","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"the solution is a roll with 1 unit of piece #1, 1 unit of piece #17, and 3 units of piece #20.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"If we solve the pricing problem with a dual vector of zeros, then the benefit of the new pattern is less than the cost of a roll, and so the function returns nothing:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solve_pricing(data, zeros(I))","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Iterative-algorithm","page":"Column generation","title":"Iterative algorithm","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Now we can combine our base model with the pricing subproblem in an iterative column generation scheme:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"while true\n    # Solve the linear relaxation\n    optimize!(model)\n    # Obtain a new dual vector\n    π = dual.(demand)\n    # Solve the pricing problem\n    new_pattern = solve_pricing(data, π)\n    # Stop iterating if there is no new pattern\n    if new_pattern === nothing\n        @info \"No new patterns, terminating the algorithm.\"\n        break\n    end\n    push!(patterns, new_pattern)\n    # Create a new column\n    push!(x, @variable(model, lower_bound = 0))\n    # Update the objective coefficient of the new column\n    set_objective_coefficient(model, x[end], 1.0)\n    # Update the non-zeros in the coefficient matrix\n    for (i, count) in zip(SparseArrays.findnz(new_pattern)...)\n        set_normalized_coefficient(demand[i], x[end], count)\n    end\n    println(\"Found new pattern. Total patterns = $(length(patterns))\")\nend","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"We found lots of new patterns. Here's pattern 21:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"patterns[21]","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Let's have a look at the patterns now:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"plot_patterns(data, patterns)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Looking-at-the-solution","page":"Column generation","title":"Looking at the solution","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Let's see how many of each column we need:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"solution = DataFrames.DataFrame([\n    (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)\n])\nfilter!(row -> row.rolls > 0, solution)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Since we solved a linear program, some of our columns have fractional solutions. We can create a integer feasible solution by rounding up the orders. This requires 341 rolls:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"sum(ceil.(Int, solution.rolls))","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Alternatively, we can re-introduce the integrality constraints and resolve the problem:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"set_integer.(x)\noptimize!(model)\nsolution = DataFrames.DataFrame([\n    (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)\n])\nfilter!(row -> row.rolls > 0, solution)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"This now requires 334 rolls:","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"sum(solution.rolls)","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Note that this may not be the global minimum because we are not adding new columns during the solution of the mixed-integer problem model (an algorithm known as branch and price). Nevertheless, the column generation algorithm typically finds good integer feasible solutions to an otherwise intractable optimization problem.","category":"page"},{"location":"tutorials/algorithms/cutting_stock_column_generation/#Next-steps","page":"Column generation","title":"Next steps","text":"","category":"section"},{"location":"tutorials/algorithms/cutting_stock_column_generation/","page":"Column generation","title":"Column generation","text":"Our objective function is to minimize the total number of rolls. What is the total length of waste? How does that compare to the total demand?\nWriting the optimization algorithm is only part of the challenge. Can you develop a better way to communicate the solution to stakeholders?","category":"page"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/constraints.md\"","category":"page"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/constraints/#constraints_ref","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"moi/reference/constraints/#Types","page":"Constraints","title":"Types","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"ConstraintIndex","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintIndex","page":"Constraints","title":"MathOptInterface.ConstraintIndex","text":"ConstraintIndex{F, S}\n\nA type-safe wrapper for Int64 for use in referencing F-in-S constraints in a model. The parameter F is the type of the function in the constraint, and the parameter S is the type of set in the constraint. To allow for deletion, indices need not be consecutive. Indices within a constraint type (i.e. F-in-S) must be unique, but non-unique indices across different constraint types are allowed. If F is VariableIndex then the index is equal to the index of the variable. That is for an index::ConstraintIndex{VariableIndex}, we always have\n\nindex.value == MOI.get(model, MOI.ConstraintFunction(), index).value\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#Functions","page":"Constraints","title":"Functions","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"is_valid(::ModelLike,::ConstraintIndex)\nadd_constraint\nadd_constraints\ntransform\nsupports_constraint","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.ConstraintIndex}","page":"Constraints","title":"MathOptInterface.is_valid","text":"is_valid(model::ModelLike, index::Index)::Bool\n\nReturn a Bool indicating whether this index refers to a valid object in the model model.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/constraints/#MathOptInterface.add_constraint","page":"Constraints","title":"MathOptInterface.add_constraint","text":"add_constraint(model::ModelLike, func::F, set::S)::ConstraintIndex{F,S} where {F,S}\n\nAdd the constraint f(x) in mathcalS where f is defined by func, and mathcalS is defined by set.\n\nadd_constraint(model::ModelLike, v::VariableIndex, set::S)::ConstraintIndex{VariableIndex,S} where {S}\nadd_constraint(model::ModelLike, vec::Vector{VariableIndex}, set::S)::ConstraintIndex{VectorOfVariables,S} where {S}\n\nAdd the constraint v in mathcalS where v is the variable (or vector of variables) referenced by v and mathcalS is defined by set.\n\nAn UnsupportedConstraint error is thrown if model does not support F-in-S constraints,\na AddConstraintNotAllowed error is thrown if it supports F-in-S constraints but it cannot add the constraint(s) in its current state and\na ScalarFunctionConstantNotZero error may be thrown if func is an AbstractScalarFunction with nonzero constant and set is EqualTo, GreaterThan, LessThan or Interval.\na LowerBoundAlreadySet error is thrown if F is a VariableIndex and a constraint was already added to this variable that sets a lower bound.\na UpperBoundAlreadySet error is thrown if F is a VariableIndex and a constraint was already added to this variable that sets an upper bound.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.add_constraints","page":"Constraints","title":"MathOptInterface.add_constraints","text":"add_constraints(model::ModelLike, funcs::Vector{F}, sets::Vector{S})::Vector{ConstraintIndex{F,S}} where {F,S}\n\nAdd the set of constraints specified by each function-set pair in funcs and sets. F and S should be concrete types. This call is equivalent to add_constraint.(model, funcs, sets) but may be more efficient.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.transform","page":"Constraints","title":"MathOptInterface.transform","text":"Transform Constraint Set\n\ntransform(model::ModelLike, c::ConstraintIndex{F,S1}, newset::S2)::ConstraintIndex{F,S2}\n\nReplace the set in constraint c with newset. The constraint index c will no longer be valid, and the function returns a new constraint index with the correct type.\n\nSolvers may only support a subset of constraint transforms that they perform efficiently (for example, changing from a LessThan to GreaterThan set). In addition, set modification (where S1 = S2) should be performed via the modify function.\n\nTypically, the user should delete the constraint and add a new one.\n\nExamples\n\nIf c is a ConstraintIndex{ScalarAffineFunction{Float64},LessThan{Float64}},\n\nc2 = transform(model, c, GreaterThan(0.0))\ntransform(model, c, LessThan(0.0)) # errors\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#MathOptInterface.supports_constraint","page":"Constraints","title":"MathOptInterface.supports_constraint","text":"MOI.supports_constraint(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging F-in-S constraints.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method for constraint types that the bridge implements.\n\n\n\n\n\nsupports_constraint(\n    model::ModelLike,\n    ::Type{F},\n    ::Type{S},\n)::Bool where {F<:AbstractFunction,S<:AbstractSet}\n\nReturn a Bool indicating whether model supports F-in-S constraints, that is, copy_to(model, src) does not throw UnsupportedConstraint when src contains F-in-S constraints. If F-in-S constraints are only not supported in specific circumstances, e.g. F-in-S constraints cannot be combined with another type of constraint, it should still return true.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/constraints/#Attributes","page":"Constraints","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/constraints/","page":"Constraints","title":"Constraints","text":"AbstractConstraintAttribute\nConstraintName\nConstraintPrimalStart\nConstraintDualStart\nConstraintPrimal\nConstraintDual\nConstraintBasisStatus\nConstraintFunction\nCanonicalConstraintFunction\nConstraintSet\nBasisStatusCode\nBASIC\nNONBASIC\nNONBASIC_AT_LOWER\nNONBASIC_AT_UPPER\nSUPER_BASIC","category":"page"},{"location":"moi/reference/constraints/#MathOptInterface.AbstractConstraintAttribute","page":"Constraints","title":"MathOptInterface.AbstractConstraintAttribute","text":"AbstractConstraintAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of constraints in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintName","page":"Constraints","title":"MathOptInterface.ConstraintName","text":"ConstraintName()\n\nA constraint attribute for a string identifying the constraint.\n\nIt is valid for constraints variables to have the same name; however, constraints with duplicate names cannot be looked up using get, regardless of whether they have the same F-in-S type.\n\nConstraintName has a default value of \"\" if not set.\n\nNotes\n\nYou should not implement ConstraintName for VariableIndex constraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart","page":"Constraints","title":"MathOptInterface.ConstraintPrimalStart","text":"ConstraintPrimalStart()\n\nA constraint attribute for the initial assignment to some constraint's ConstraintPrimal that the optimizer may use to warm-start the solve.\n\nMay be nothing (unset), a number for AbstractScalarFunction, or a vector for AbstractVectorFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintDualStart","page":"Constraints","title":"MathOptInterface.ConstraintDualStart","text":"ConstraintDualStart()\n\nA constraint attribute for the initial assignment to some constraint's ConstraintDual that the optimizer may use to warm-start the solve.\n\nMay be nothing (unset), a number for AbstractScalarFunction, or a vector for AbstractVectorFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintPrimal","page":"Constraints","title":"MathOptInterface.ConstraintPrimal","text":"ConstraintPrimal(result_index::Int = 1)\n\nA constraint attribute for the assignment to some constraint's primal value(s) in result result_index.\n\nIf the constraint is f(x) in S, then in most cases the ConstraintPrimal is the value of f, evaluated at the correspondng VariablePrimal solution.\n\nHowever, some conic solvers reformulate b - Ax in S to s = b - Ax, s in S. These solvers may return the value of s for ConstraintPrimal, rather than b - Ax. (Although these are constrained by an equality constraint, due to numerical tolerances they may not be identical.)\n\nIf the solver does not have a primal value for the constraint because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ConstraintPrimal attribute.\n\nIf result_index is omitted, it is 1 by default. See ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintDual","page":"Constraints","title":"MathOptInterface.ConstraintDual","text":"ConstraintDual(result_index::Int = 1)\n\nA constraint attribute for the assignment to some constraint's dual value(s) in result result_index. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a dual value for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a primal solution is available), the result is undefined. Users should first check DualStatus before accessing the ConstraintDual attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintBasisStatus","page":"Constraints","title":"MathOptInterface.ConstraintBasisStatus","text":"ConstraintBasisStatus(result_index::Int = 1)\n\nA constraint attribute for the BasisStatusCode of some constraint in result result_index, with respect to an available optimal solution basis. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a basis statue for the constraint because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the ConstraintBasisStatus attribute.\n\nSee ResultCount for information on how the results are ordered.\n\nNotes\n\nFor the basis status of a variable, query VariableBasisStatus.\n\nConstraintBasisStatus does not apply to VariableIndex constraints. You can infer the basis status of a VariableIndex constraint by looking at the result of VariableBasisStatus.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintFunction","page":"Constraints","title":"MathOptInterface.ConstraintFunction","text":"ConstraintFunction()\n\nA constraint attribute for the AbstractFunction object used to define the constraint.\n\nIt is guaranteed to be equivalent but not necessarily identical to the function provided by the user.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.CanonicalConstraintFunction","page":"Constraints","title":"MathOptInterface.CanonicalConstraintFunction","text":"CanonicalConstraintFunction()\n\nA constraint attribute for a canonical representation of the AbstractFunction object used to define the constraint.\n\nGetting this attribute is guaranteed to return a function that is equivalent but not necessarily identical to the function provided by the user.\n\nBy default, MOI.get(model, MOI.CanonicalConstraintFunction(), ci) fallbacks to MOI.Utilities.canonical(MOI.get(model, MOI.ConstraintFunction(), ci)). However, if model knows that the constraint function is canonical then it can implement a specialized method that directly return the function without calling Utilities.canonical. Therefore, the value returned cannot be assumed to be a copy of the function stored in model. Moreover, Utilities.Model checks with Utilities.is_canonical whether the function stored internally is already canonical and if it's the case, then it returns the function stored internally instead of a copy.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.ConstraintSet","page":"Constraints","title":"MathOptInterface.ConstraintSet","text":"ConstraintSet()\n\nA constraint attribute for the AbstractSet object used to define the constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.BasisStatusCode","page":"Constraints","title":"MathOptInterface.BasisStatusCode","text":"BasisStatusCode\n\nAn Enum of possible values for the ConstraintBasisStatus and VariableBasisStatus attributes, explaining the status of a given element with respect to an optimal solution basis.\n\nNotes\n\nNONBASIC_AT_LOWER and NONBASIC_AT_UPPER should be used only for\n\nconstraints with the Interval set. In this case, they are necessary to distinguish which side of the constraint is active. One-sided constraints (e.g., LessThan and GreaterThan) should use NONBASIC instead of the NONBASIC_AT_* values. This restriction does not apply to VariableBasisStatus, which should return NONBASIC_AT_* regardless of whether the alternative bound exists.\n\nIn linear programs, SUPER_BASIC occurs when a variable with no bounds is not\n\nin the basis.\n\nValues\n\nPossible values are:\n\nBASIC: element is in the basis\nNONBASIC: element is not in the basis\nNONBASIC_AT_LOWER: element is not in the basis and is at its lower bound\nNONBASIC_AT_UPPER: element is not in the basis and is at its upper bound\nSUPER_BASIC: element is not in the basis but is also not at one of its bounds\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/constraints/#MathOptInterface.BASIC","page":"Constraints","title":"MathOptInterface.BASIC","text":"BASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nBASIC: element is in the basis\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC","page":"Constraints","title":"MathOptInterface.NONBASIC","text":"NONBASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC: element is not in the basis\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC_AT_LOWER","page":"Constraints","title":"MathOptInterface.NONBASIC_AT_LOWER","text":"NONBASIC_AT_LOWER::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC_AT_LOWER: element is not in the basis and is at its lower bound\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.NONBASIC_AT_UPPER","page":"Constraints","title":"MathOptInterface.NONBASIC_AT_UPPER","text":"NONBASIC_AT_UPPER::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nNONBASIC_AT_UPPER: element is not in the basis and is at its upper bound\n\n\n\n\n\n","category":"constant"},{"location":"moi/reference/constraints/#MathOptInterface.SUPER_BASIC","page":"Constraints","title":"MathOptInterface.SUPER_BASIC","text":"SUPER_BASIC::BasisStatusCode\n\nAn instance of the BasisStatusCode enum.\n\nSUPER_BASIC: element is not in the basis but is also not at one of its bounds\n\n\n\n\n\n","category":"constant"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"EditURL = \"tsp_lazy_constraints.jl\"","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_lazy","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial was originally contributed by Daniel Schermer.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This tutorial describes how to implement the Traveling Salesperson Problem in JuMP using solver-independent lazy constraints that dynamically separate subtours. To be more precise, we use lazy constraints to cut off infeasible subtours only when necessary and not before needed.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"It uses the following packages:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"using JuMP\nimport GLPK\nimport Random\nimport Plots","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_model","page":"Traveling Salesperson Problem","title":"Mathematical Formulation","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Assume that we are given a complete graph mathcalG(VE) where V is the set of vertices (or cities) and E is the set of edges (or roads). For each pair of vertices i j in V i neq j the edge (ij) in E is associated with a weight (or distance) d_ij in mathbbR^+.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"For this tutorial, we assume the problem to be symmetric, that is, d_ij = d_ji  forall ij in V.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"In the Traveling Salesperson Problem, we are tasked with finding a tour with minimal length that visits every vertex exactly once and then returns to the point of origin, that is, a Hamiltonian cycle with minimal weight.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"To model the problem, we introduce a binary variable, x_ij in 01  forall i j in V, that indicates if edge (ij) is part of the tour or not. Using these variables, the Traveling Salesperson Problem can be modeled as the following integer linear program.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_objective","page":"Traveling Salesperson Problem","title":"Objective Function","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The objective is to minimize the length of the tour (due to the assumed symmetry, the second sum only contains ij):","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"textmin  sum_i in V  sum_j in V i  j d_ij x_ij","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Note that it is also possible to use the following objective function instead:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"textmin  sum_i in V  sum_j in V dfracd_ij x_ij2","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_constraints","page":"Traveling Salesperson Problem","title":"Constraints","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"There are four classes of constraints in our formulation.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"First, due to the presumed symmetry, the following constraints must hold:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"x_ij = x_ji quad forall ij in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Second, for each vertex i, exactly two edges must be selected that connect it to other vertices j in the graph G:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"sum_j in V x_ij = 2 quad forall i in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Third, we do not permit loops to occur:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"x_ii = 0 quad forall i in V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The fourth constraint is more complicated. A major difficulty of the Traveling Salesperson Problem arises from the fact that we need to prevent subtours, that is, several distinct Hamiltonian cycles existing on subgraphs of G.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Note that the previous constraints do not guarantee that the solution will be free of subtours. To this end, by S we label a subset of vertices. Then, for each proper subset S subset V, the following constraints guarantee that no subtour may occur:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"sum_i in S sum_j in S i  j x_ij leq vert S vert - 1 quad forall S subset V","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Problematically, we require exponentially many of these constraints as vert V vert increases. Therefore, we will add these constraints only when necessary.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#tsp_implementation","page":"Traveling Salesperson Problem","title":"Implementation","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"There are two ways we can eliminate subtours in JuMP, both of which will be shown in what follows:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"iteratively solving a new model that incorporates previously identified subtours,\nor adding violated subtours as lazy constraints.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Data","page":"Traveling Salesperson Problem","title":"Data","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"The vertices are assumed to be randomly distributed in the Euclidean space; thus, the weight (distance) of each edge is defined as follows.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function generate_distance_matrix(n; random_seed = 1)\n    rng = Random.MersenneTwister(random_seed)\n    X = 100 * rand(rng, n)\n    Y = 100 * rand(rng, n)\n    d = [sqrt((X[i] - X[j])^2 + (Y[i] - Y[j])^2) for i in 1:n, j in 1:n]\n    return X, Y, d\nend\n\nn = 40\nX, Y, d = generate_distance_matrix(n)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"For the JuMP model, we first initialize the model object. Then, we create the binary decision variables and add the objective function and constraints. By defining the x matrix as Symmetric, we do not need to add explicit constraints that x[i, j] == x[j, i].","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function build_tsp_model(d, n)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:n, 1:n], Bin, Symmetric)\n    @objective(model, Min, sum(d .* x) / 2)\n    @constraint(model, [i in 1:n], sum(x[i, :]) == 2)\n    @constraint(model, [i in 1:n], x[i, i] == 0)\n    return model\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"To search for violated constraints, based on the edges that are currently in the solution (that is, those that have value x_ij = 1), we identify the shortest cycle through the function subtour(). Whenever a subtour has been identified, a constraint corresponding to the form above can be added to the model.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function subtour(edges::Vector{Tuple{Int,Int}}, n)\n    shortest_subtour, unvisited = collect(1:n), Set(collect(1:n))\n    while !isempty(unvisited)\n        this_cycle, neighbors = Int[], unvisited\n        while !isempty(neighbors)\n            current = pop!(neighbors)\n            push!(this_cycle, current)\n            if length(this_cycle) > 1\n                pop!(unvisited, current)\n            end\n            neighbors =\n                [j for (i, j) in edges if i == current && j in unvisited]\n        end\n        if length(this_cycle) < length(shortest_subtour)\n            shortest_subtour = this_cycle\n        end\n    end\n    return shortest_subtour\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Let us declare a helper function selected_edges() that will be repeatedly used in what follows.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function selected_edges(x::Matrix{Float64}, n)\n    return Tuple{Int,Int}[(i, j) for i in 1:n, j in 1:n if x[i, j] > 0.5]\nend","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Other helper functions for computing subtours:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"subtour(x::Matrix{Float64}) = subtour(selected_edges(x, size(x, 1)), size(x, 1))\nsubtour(x::AbstractMatrix{VariableRef}) = subtour(value.(x))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Iterative-method","page":"Traveling Salesperson Problem","title":"Iterative method","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"An iterative way of eliminating subtours is the following.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"However, it is reasonable to assume that this is not the most efficient way: whenever a new subtour elimination constraint is added to the model, the optimization has to start from the very beginning.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"That way, the solver will repeatedly discard useful information encountered during previous solves (for example, all cuts, the incumbent solution, or lower bounds).","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"info: Info\nNote that, in principle, it would also be feasible to add all subtours (instead of just the shortest one) to the model. However, preventing just the shortest cycle is often sufficient for breaking other subtours and will keep the model size smaller.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"iterative_model = build_tsp_model(d, n)\noptimize!(iterative_model)\ntime_iterated = solve_time(iterative_model)\ncycle = subtour(iterative_model[:x])\nwhile 1 < length(cycle) < n\n    println(\"Found cycle of length $(length(cycle))\")\n    S = [(i, j) for (i, j) in Iterators.product(cycle, cycle) if i < j]\n    @constraint(\n        iterative_model,\n        sum(iterative_model[:x][i, j] for (i, j) in S) <= length(cycle) - 1,\n    )\n    optimize!(iterative_model)\n    global time_iterated += solve_time(iterative_model)\n    global cycle = subtour(iterative_model[:x])\nend\n\nobjective_value(iterative_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"time_iterated","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"As a quick sanity check, we visualize the optimal tour to verify that no subtour is present:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"function plot_tour(X, Y, x)\n    plot = Plots.plot()\n    for (i, j) in selected_edges(x, size(x, 1))\n        Plots.plot!([X[i], X[j]], [Y[i], Y[j]]; legend = false)\n    end\n    return plot\nend\n\nplot_tour(X, Y, value.(iterative_model[:x]))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/#Lazy-constraint-method","page":"Traveling Salesperson Problem","title":"Lazy constraint method","text":"","category":"section"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"A more sophisticated approach makes use of lazy constraints. To be more precise, we do this through the subtour_elimination_callback() below, which is only run whenever we encounter a new integer-feasible solution.","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"lazy_model = build_tsp_model(d, n)\nfunction subtour_elimination_callback(cb_data)\n    status = callback_node_status(cb_data, lazy_model)\n    if status != MOI.CALLBACK_NODE_STATUS_INTEGER\n        return  # Only run at integer solutions\n    end\n    cycle = subtour(callback_value.(cb_data, lazy_model[:x]))\n    if !(1 < length(cycle) < n)\n        return  # Only add a constraint if there is a cycle\n    end\n    println(\"Found cycle of length $(length(cycle))\")\n    S = [(i, j) for (i, j) in Iterators.product(cycle, cycle) if i < j]\n    con = @build_constraint(\n        sum(lazy_model[:x][i, j] for (i, j) in S) <= length(cycle) - 1,\n    )\n    MOI.submit(lazy_model, MOI.LazyConstraint(cb_data), con)\n    return\nend\nset_attribute(\n    lazy_model,\n    MOI.LazyConstraintCallback(),\n    subtour_elimination_callback,\n)\noptimize!(lazy_model)\nobjective_value(lazy_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"This finds the same optimal tour:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"plot_tour(X, Y, value.(lazy_model[:x]))","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"Surprisingly, for this particular model with GLPK, the solution time is worse than the iterative method:","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"time_lazy = solve_time(lazy_model)","category":"page"},{"location":"tutorials/algorithms/tsp_lazy_constraints/","page":"Traveling Salesperson Problem","title":"Traveling Salesperson Problem","text":"In most other cases and solvers, however, the lazy time should be faster than the iterative method.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"CurrentModule = JuMP","category":"page"},{"location":"changelog/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"changelog/#Version-1.16.0-(October-24,-2023)","page":"Release notes","title":"Version 1.16.0 (October 24, 2023)","text":"","category":"section"},{"location":"changelog/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added := operator for Boolean satisfiability problems (#3530)","category":"page"},{"location":"changelog/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed text/latex printing of MOI.Interval sets (#3537)\nFixed tests with duplicate function names (#3539)","category":"page"},{"location":"changelog/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated documentation list of supported solvers (#3527) (#3529) (#3538) (#3542) (#3545) (#3546)\nUpdated to Documenter@1.1 (#3528)\nFixed various tutorials (#3534) (#3532)\nFixed Project.toml compat bounds for standard libraries (#3544)","category":"page"},{"location":"changelog/#Version-1.15.1-(September-24,-2023)","page":"Release notes","title":"Version 1.15.1 (September 24, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for single argument min and max operators (#3522)\nFixed error message for add_to_expression! when called with a GenericNonlinearExpr (#3506)\nFixed constraint tags with broadcasted constraints (#3515)\nFixed MethodError in MA.scaling (#3518)\nFixed support for arrays of Parameter variables (#3524)","category":"page"},{"location":"changelog/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated to Documenter@1 (#3501)\nFixed links to data in tutorials (#3512)\nFixed typo in TSP tutorial (#3516)\nImproved error message for VariableNotOwned errors (#3520)\nFixed various JET errors (#3519)","category":"page"},{"location":"changelog/#Version-1.15.0-(September-15,-2023)","page":"Release notes","title":"Version 1.15.0 (September 15, 2023)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it adds an entirely new data structure and API path for working with nonlinear programs. The previous nonlinear interface remains unchanged and is documented at Nonlinear Modeling (Legacy). The new interface is a treated as a non-breaking feature addition and is documented at Nonlinear Modeling.","category":"page"},{"location":"changelog/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Although the new nonlinear interface is a feature addition, there are two changes which might be breaking for a very small number of users.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The syntax inside JuMP macros is parsed using a different code path, even for linear and quadratic expressions. We made this change to unify how we parse linear, quadratic, and nonlinear expressions. In all cases, the new code returns equivalent expressions, but because of the different order of operations, there are three changes to be aware of when updating:\nThe printed form of the expression may change, for example from x * y to y * x. This can cause tests which test the String representation of a model to fail.\nSome coefficients may change slightly due to floating point round-off error.\nParticularly when working with a JuMP extension, you may encounter a MethodError due to a missing or ambiguous method. These errors are due to previously existing bugs that were not triggered by the previous parsing code. If you encounter such an error, please open a GitHub issue.\nThe methods for Base.:^(x::VariableRef, n::Integer) and Base.:^(x::AffExpr, n::Integer) have changed. Previously, these methods supported only n = 0, 1, 2 and they always returned a QuadExpr, even for the case when n = 0 or n = 1. Now:\nx^0 returns one(T), where T is the value_type of the model (defaults to Float64)\nx^1 returns x\nx^2 returns a QuadExpr\nx^n where !(0 <= n <= 2) returns a NonlinearExpr.\nWe made this change to support nonlinear expressions and to align the mathematical definition of the operation with their return type. (Previously, users were surprised that x^1 returned a QuadExpr.) As a consequence of this change, the methods are now not type-stable. This means that the compiler cannot prove that x^2 returns a QuadExpr. If benchmarking shows that this is a performance problem, you can use the type-stable x * x instead of x^2.","category":"page"},{"location":"changelog/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added triangle_vec which simplifies adding MOI.LogDetConeTriangle and MOI.RootDetConeTriangle constraints (#3456)\nAdded the new nonlinear interface. This is a very large change. See the documentation at Nonlinear Modeling and the (long) discussion in JuMP.jl#3106. Related PRs are (#3468) (#3472) (#3475) (#3483) (#3487) (#3488) (#3489) (#3504) (#3509)","category":"page"},{"location":"changelog/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed uses of @nospecialize which cause precompilation failures in Julia v1.6.0 and v1.6.1. (#3464)\nFixed adding a container of Parameter (#3473)\nFixed return type of x^0 and x^1 to no longer return QuadExpr (see note in Breaking section above) (#3474)\nFixed error messages in LowerBoundRef, UpperBoundRef, FixRef, IntegerRef, BinaryRef, ParameterRef and related functions (#3494)\nFixed type inference of empty containers in JuMP macros (#3500)","category":"page"},{"location":"changelog/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added GAMS to solver documentation (#3357)\nUpdated various tutorials (#3459) (#3460) (#3462) (#3463) (#3465) (#3490) (#3492) (#3503)\nAdded The network multi-commodity flow problem tutorial (#3491)\nAdded Two-stage stochastic programs tutorial (#3466)\nAdded better error messages for unsupported operations in LinearAlgebra (#3476)\nUpdated to the latest version of Documenter (#3484) (#3495) (#3497)\nUpdated GitHub action versions (#3507)","category":"page"},{"location":"changelog/#Version-1.14.1-(September-2,-2023)","page":"Release notes","title":"Version 1.14.1 (September 2, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix links in Documentation (#3478)","category":"page"},{"location":"changelog/#Version-1.14.0-(August-27,-2023)","page":"Release notes","title":"Version 1.14.0 (August 27, 2023)","text":"","category":"section"},{"location":"changelog/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added DimensionalData.jl extension (#3413)\nAdded syntactic sugar for the MOI.Parameter set (#3443)\nParameter\nParameterRef\nis_parameter\nparameter_value\nset_parameter_value","category":"page"},{"location":"changelog/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed model_convert for BridgeableConstraint (#3437)\nFixed printing models with integer coefficients larger than typemax(Int) (#3447)\nFixed support for constant left-hand side functions in a complementarity constraint (#3452)","category":"page"},{"location":"changelog/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Updated packages used in documentation (#3444) (#3455)\nFixed docstring tests (#3445)\nFixed printing change for MathOptInterface (#3446)\nFixed typos in documentation (#3448) (#3457)\nAdded SCIP to callback documentation (#3449)","category":"page"},{"location":"changelog/#Version-1.13.0-(July-27,-2023)","page":"Release notes","title":"Version 1.13.0 (July 27, 2023)","text":"","category":"section"},{"location":"changelog/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for generic number types (#3377) (#3385)\nAdded fallback for MOI.AbstractSymmetricMatrixSetTriangle and MOI.AbstractSymmetricMatrixSetSquare (#3424)","category":"page"},{"location":"changelog/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed set_start_values with MOI.Bridges.Objective.SlackBridge (#3422)\nFixed flakey doctest in variables.md (#3425)\nFixed names on CITATION.bib (#3423)","category":"page"},{"location":"changelog/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Loraine.jl to the installation table (#3426)\nRemoved Penopt.jl from packages.toml (#3428)\nImproved problem statement in cannery example of tutorial (#3430)\nMinor cleanups in Containers.DenseAxisArray implementation (#3429)\nChanged nested_problems.jl: outer/inner to upper/lower (#3433)\nRemoved second SDP relaxation in OPF tutorial (#3432)","category":"page"},{"location":"changelog/#Version-1.12.0-(June-19,-2023)","page":"Release notes","title":"Version 1.12.0 (June 19, 2023)","text":"","category":"section"},{"location":"changelog/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added coefficient_type keyword argument to add_bridge and remove_bridge (#3394)","category":"page"},{"location":"changelog/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed error message for matrix in HermitianPSDCone (#3369)\nFixed EditURL for custom documentation pages (#3373)\nFixed return type annotations for MOI.ConstraintPrimal and MOI.ConstraintDual (#3381)\nFixed printing change in Julia nightly (#3391)\nFixed printing of Complex coefficients (#3397)\nFixed printing of constraints in text/latex mode (#3405)\nFixed performance issue in Containers.rowtable (#3410)\nFixed bug when variables added to set of wrong dimension (#3411)","category":"page"},{"location":"changelog/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added more solver READMEs to the documentation (#3358) (#3360) (#3364) (#3365) (#3366) (#3368) (#3372) (#3374) (#3376) (#3379) (#3387) (#3389)\nAdded StatusSwitchingQP.jl to the installation table (#3354)\nUpdated checklist for adding a new solver (#3370)\nUpdated extension-tests.yml action (#3371) (#3375)\nColor logs in GitHub actions (#3392)\nAdded new tutorials\nOptimal power flow (#3395) (#3412)\nLovász numbers (#3399)\nDualization (#3402)\nUpdated JuMP paper citation (#3400)\nChanged GitHub action to upload LaTeX logs when building documentation (#3403)\nFixed printing of SCS log in documentation (#3406)\nUpdated solver versions (#3407)\nUpdated documentation to use Julia v1.9 (#3398)\nReplaced _value_type with MOI.Utilities.value_type (#3414)\nFixed a typo in docstring (#3415)\nRefactored API documentation (#3386)\nUpdated SCIP license (#3420)","category":"page"},{"location":"changelog/#Version-1.11.1-(May-19,-2023)","page":"Release notes","title":"Version 1.11.1 (May 19, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a poor error message when sum(::DenseAxisArray; dims) was called (#3338)\nFixed support for dependent sets in the @variable macro (#3344)\nFixed a performance bug in constraints with sparse symmetric matrices (#3349)","category":"page"},{"location":"changelog/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the printing of complex numbers (#3332)\nWhen printing, sets which contain constants ending in .0 now print as integers. This follows the behavior of constants in functions (#3341)\nAdded InfiniteOpt to the extensions documentation (#3343)\nAdded more documentation for the exponential cone (#3345) (#3347)\nAdded checklists for developers (#3346) (#3355)\nFixed test support upcoming Julia nightly (#3351)\nFixed extension-tests.yml action (#3353)\nAdd more solvers to the documentation (#3359) (#3361) (#3362)","category":"page"},{"location":"changelog/#Version-1.11.0-(May-3,-2023)","page":"Release notes","title":"Version 1.11.0 (May 3, 2023)","text":"","category":"section"},{"location":"changelog/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added new methods to print_active_bridges for printing a particular objective, constraint, or variable (#3316)","category":"page"},{"location":"changelog/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed tests for MOI v1.14.0 release (#3312)\nFixed indexing containers when an axis is Vector{Any} that contains a Vector{Any} element (#3280)\nFixed getindex(::AbstractJuMPScalar) which is called for an expression like x[] (#3314)\nFixed bug in set_string_names_on_creation with a vector of variables (#3322)\nFixed bug in memoize function in nonlinear documentation (#3337)","category":"page"},{"location":"changelog/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#3317) (#3318) (#3328)\nAdded a test for the order of setting start values (#3315)\nAdded READMEs of solvers and extensions to the docs (#3309) (#3320) (#3327) (#3329) (#3333)\nStyle improvements to src/variables.jl (#3324)\nClarify that column generation does not find global optimum (#3325)\nAdd a GitHub actions workflow for testing extensions prior to release (#3331)\nDocument the release process for JuMP (#3334)\nFix links to discourse and chatroom (#3335)","category":"page"},{"location":"changelog/#Version-1.10.0-(April-3,-2023)","page":"Release notes","title":"Version 1.10.0 (April 3, 2023)","text":"","category":"section"},{"location":"changelog/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Nonnegatives, Nonpositives and Zeros, and support vector-valued inequality syntax in the JuMP macros (#3273)\nAdded special support for LinearAlgebra.Symmetric and LinearAlgebra.Hermitian matrices in Zeros constraints (#3281) (#3296)\nAdded HermitianMatrixSpace and the Hermitian tag for generating a matrix of variables that is Hermitian (#3292) (#3293)\nAdded Semicontinuous and Semiinteger (#3302)\nAdded support for keyword indexing of containers (#3237)","category":"page"},{"location":"changelog/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed [compat] bound for MathOptInterface in Project.toml (#3272)","category":"page"},{"location":"changelog/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Split out the Nested optimization problems tutorial (#3274)\nUpdated doctests to ensure none have hidden state (#3275) (#3276)\nClarified how lazy constraints may revisit points (#3278)\nAdded P-Norm example (#3282)\nClarified docs that macros create new bindings (#3284)\nFixed threading example (#3283)\nAdded plot to The minimum distortion problem (#3288)\nAdded Google style rules for Vale and fixed warnings (#3285)\nAdded citation for the JuMP 1.0 paper (#3294)\nUpdated package versions in the documentation (#3298)\nAdded comment for the order in which start values must be set (#3303)\nImproved error message for unrecognized constraint operators (#3311)","category":"page"},{"location":"changelog/#Version-1.9.0-(March-7,-2023)","page":"Release notes","title":"Version 1.9.0 (March 7, 2023)","text":"","category":"section"},{"location":"changelog/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added get_attribute and set_attribute. These replace get_optimizer_attribute and set_optimizer_attribute, although the _optimizer_ functions remain for backward compatibility. (#3219)\nAdded set_start_values for setting all supported start values in a model (#3238)\nAdd remove_bridge and print_active_bridges (#3259)","category":"page"},{"location":"changelog/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The matrix returned by a variable in HermitianPSDCone is now a LinearAlgebra.Hermitian matrix. This is potentially breaking if you have written code to assume the return is a Matrix. (#3245) (#3246)\nFixed missing support for Base.isreal of expressions (#3252)","category":"page"},{"location":"changelog/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a thread safety issue in the Parallelism tutorial (#3240) (#3243)\nImproved the error message when unsupported operators are used in @NL macros (#3236)\nClarified the documentation to say that matrices in HermitianPSDCone must be LinearAlgebra.Hermitian (#3241)\nMinor style fixes to internal macro code (#3247)\nAdd Quantum state discrimination tutorial (#3250)\nImprove error message when begin...end not passed to plural macros (#3255)\nDocument how to register function with varying number of input arguments (#3258)\nTidy tests by removing unneeded JuMP. prefixes (#3260)\nClarified the introduction to the Complex number support tutorial (#3262)\nFixed typos in the Documentation (#3263) (#3266) (#3268) (#3269)","category":"page"},{"location":"changelog/#Version-1.8.2-(February-27,-2023)","page":"Release notes","title":"Version 1.8.2 (February 27, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed dot product between complex JuMP expression and number (#3244)","category":"page"},{"location":"changelog/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Polish simple SDP examples (#3232)","category":"page"},{"location":"changelog/#Version-1.8.1-(February-23,-2023)","page":"Release notes","title":"Version 1.8.1 (February 23, 2023)","text":"","category":"section"},{"location":"changelog/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for init in nonlinear generator expressions (#3226)","category":"page"},{"location":"changelog/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Use and document import MathOptInterface as MOI (#3222)\nRemoved references in documentation to multiobjective optimization being unsupported (#3223)\nAdded tutorial on multi-objective portfolio optimization (#3227)\nRefactored some of the conic tutorials (#3229)\nFixed typos in the documentation (#3230)\nAdded tutorial on parallelism (#3231)","category":"page"},{"location":"changelog/#Version-1.8.0-(February-16,-2023)","page":"Release notes","title":"Version 1.8.0 (February 16, 2023)","text":"","category":"section"},{"location":"changelog/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added --> syntax support for indicator constraints. The old syntax of => remains supported (#3207)\nAdded <--> syntax for reified constraints. For now, few solvers support reified constraints (#3206)\nAdded fix_discrete_variables. This is most useful for computing the dual of a mixed-integer program (#3208)\nAdded support for vector-valued objectives. For details, see the Multi-objective knapsack tutorial (#3176)","category":"page"},{"location":"changelog/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report by switching to an explicit LU factorization of the basis matrix (#3182)\nFixed a bug that prevented [; kwarg] arguments in macros (#3220)","category":"page"},{"location":"changelog/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Minor fixes to the documentation (#3200) (#3201) (#3203) (#3210)\nAdded tutorial Constraint programming (#3202)\nAdded more examples to Tips and Tricks\nRemove _distance_to_set in favor of MOI.Utilities.distance_to_set (#3209)\nImprove The diet problem tutorial by adding the variable as a column in the dataframe (#3213)\nImprove The knapsack problem example tutorial (#3216) (#3217)\nAdded the Ellipsoid approximation tutorial (#3218)","category":"page"},{"location":"changelog/#Version-1.7.0-(January-25,-2023)","page":"Release notes","title":"Version 1.7.0 (January 25, 2023)","text":"","category":"section"},{"location":"changelog/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for view of a Containers.DenseAxisArray (#3152) (#3180)\nAdded support for containers of variables in ComplexPlane (#3184)\nAdded support for minimum and maximum generators in nonlinear expressions (#3189)\nAdded SnoopPrecompile statements that reduce the time-to-first-solve in Julia 1.9 (#3193) (#3195) (#3196) (#3197)","category":"page"},{"location":"changelog/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Large refactoring of the tests (#3166) (#3167) (#3168) (#3169) (#3170) (#3171)\nRemove unreachable code due to VERSION checks (#3172)\nDocument how to test JuMP extensions (#3174)\nFix method ambiguities in Containers (#3173)\nImprove error message that is thrown when = is used instead of == in the @constraint macro (#3178)\nImprove the error message when Bool is used instead of Bin in the @variable macro (#3180)\nUpdate versions of the documentation (#3185)\nTidy the import of packages and remove unnecessary prefixes (#3186) (#3187)\nRefactor src/JuMP.jl by moving methods into more relevant files (#3188)\nFix docstring of Model not appearing in the documentation (#3198)","category":"page"},{"location":"changelog/#Version-1.6.0-(January-1,-2023)","page":"Release notes","title":"Version 1.6.0 (January 1, 2023)","text":"","category":"section"},{"location":"changelog/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a result keyword argument to solution_summary to allow summarizing models with multiple solutions (#3138)\nAdded relax_with_penalty!, which is a useful tool when debugging infeasible models (#3140)\nAdded has_start_value (#3157)\nAdded support for HermitianPSDCone in constraints (#3154)","category":"page"},{"location":"changelog/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed promotion of complex expressions (#3150) (#3164)","category":"page"},{"location":"changelog/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Benders tutorial with in-place resolves (#3145)\nAdded more Tips and tricks for linear programs (#3144) (#3163)\nClarified documentation that start can depend on the indices of a variable container (#3148)\nReplace instances of length and size by the recommended eachindex and axes (#3149)\nAdded a warning explaining why the model is dirty when accessing solution results from a modified model (#3156)\nClarify documentation that PSD ensures a symmetric matrix (#3159)\nMaintenance of the JuMP test suite (#3146) (#3158) (#3162)","category":"page"},{"location":"changelog/#Version-1.5.0-(December-8,-2022)","page":"Release notes","title":"Version 1.5.0 (December 8, 2022)","text":"","category":"section"},{"location":"changelog/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for complex-valued variables:\nHermitianPSDCone (#3109)\nComplexPlane and ComplexVariable (#3134)\nAdd support for MOI.OptimizerWithAttributes in set_optimizer_attribute and get_optimizer_attribute (#3129)","category":"page"},{"location":"changelog/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed error message for vectorized interval constraints (#3123)\nFixed passing AbstractString to set_optimizer_attribute (#3127)","category":"page"},{"location":"changelog/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update package versions used in docs (#3119) (#3133) (#3139)\nFixed output of diet tutorial (#3120)\nExplain how to use Dates.period in set_time_limit_sec (#3121)\nUpdate to JuliaFormatter v1.0.15 (#3130)\nFixed HTTP server example in web_app.jl (#3131)\nUpdate docs to build with Documenter#master (#3094)\nAdd tests for LinearAlgebra operations (#3132)\nTidy these release notes (#3135)\nAdded documentation for Complex number support (#3141)\nRemoved the \"workforce scheduling\" and \"steelT3\" tutorials (#3143)","category":"page"},{"location":"changelog/#Version-1.4.0-(October-29,-2022)","page":"Release notes","title":"Version 1.4.0 (October 29, 2022)","text":"","category":"section"},{"location":"changelog/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Containers.rowtable which converts a container into a vector of NamedTuples to support the Tables.jl interface. This simplifies converting Containers.DenseAxisArray and Containers.SparseAxisArray objects into tabular forms such as a DataFrame (#3104)\nAdded a new method to Containers.container so that index names are passed to the container (#3088)","category":"page"},{"location":"changelog/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in copy_to(dest::Model, src::MOI.ModelLike) when src has nonlinear components (#3101)\nFixed the printing of (-1.0 + 0.0im) coefficients in complex expressions (#3112)\nFixed a parsing bug in nonlinear expressions with generator statements that contain multiple for statements (#3116)","category":"page"},{"location":"changelog/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Converted the multi-commodity flow tutorial to use an SQLite database (#3098)\nFixed a number of typos in the documentation (#3103) (#3107) (#3018)\nImproved various style aspects of the PDF documentation (#3095) (#3098) (#3102)","category":"page"},{"location":"changelog/#Version-1.3.1-(September-28,-2022)","page":"Release notes","title":"Version 1.3.1 (September 28, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in relax_integrality (#3087)\nFixed the type stability of operators with Complex arguments (#3072)\nFixed a bug which added additional +() terms to some nonlinear expressions (#3091)\nFixed potential method ambiguities with AffExpr and QuadExpr objects (#3092)","category":"page"},{"location":"changelog/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added vale as a linter for the documentation (#3080)\nAdded a tutorial on debugging JuMP models (#3043)\nFixed a number of typos in the documentation (#3079) (#3083)\nMany other small tweaks to the documentation (#3068) (#3073) (#3074) (#3075) (#3076) (#3077) (#3078) (#3081) (#3082) (#3084) (#3085) (#3089)","category":"page"},{"location":"changelog/#Version-1.3.0-(September-5,-2022)","page":"Release notes","title":"Version 1.3.0 (September 5, 2022)","text":"","category":"section"},{"location":"changelog/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support slicing in SparseAxisArray (#3031)","category":"page"},{"location":"changelog/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug introduced in v1.2.0 that prevented DenseAxisArrays with Vector keys (#3064)","category":"page"},{"location":"changelog/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Released the JuMP logos under the CC BY 4.0 license (#3063)\nMinor tweaks to the documentation (#3054) (#3056) (#3057) (#3060) (#3061) (#3065)\nImproved code coverage of a number of files (#3048) (#3049) (#3050) (#3051) (#3052) (#3053) (#3058) (#3059)","category":"page"},{"location":"changelog/#Version-1.2.1-(August-22,-2022)","page":"Release notes","title":"Version 1.2.1 (August 22, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug when parsing two-sided nonlinear constraints (#3045)","category":"page"},{"location":"changelog/#Version-1.2.0-(August-16,-2022)","page":"Release notes","title":"Version 1.2.0 (August 16, 2022)","text":"","category":"section"},{"location":"changelog/#Breaking-2","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"This is a large minor release because it significantly refactors the internal code for handling nonlinear programs to use the MathOptInterface.Nonlinear submodule that was introduced in MathOptInterface v1.3.0. As a consequence, the internal datastructure in model.nlp_data has been removed, as has the JuMP._Derivatives submodule. Despite the changes, the public API for nonlinear programming has not changed, and any code that uses only the public API and that worked with v1.1.1 will continue to work with v1.2.0.","category":"page"},{"location":"changelog/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added all_constraints(model; include_variable_in_set_constraints) which simplifies returning a list of all constraint indices in the model.\nAdded the ability to delete nonlinear constraints via delete(::Model, ::NonlinearConstraintRef).\nAdded the ability to provide an explicit Hessian for a multivariate user-defined function.\nAdded support for querying the primal value of a nonlinear constraint via value(::NonlinearConstraintRef)","category":"page"},{"location":"changelog/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Containers.DenseAxisArray so that it now supports indexing with keys that hash to the same value, even if they are different types, for example, Int32 and Int64.\nFixed a bug printing the model when the solver does not support MOI.Name.","category":"page"},{"location":"changelog/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a constraint programming formulation to the Sudoku tutorial.\nAdded newly supported solvers Pajarito, Clarabel, and COPT to the installation table.\nFixed a variety of other miscellaneous issues in the documentation.","category":"page"},{"location":"changelog/#Version-1.1.1-(June-14,-2022)","page":"Release notes","title":"Version 1.1.1 (June 14, 2022)","text":"","category":"section"},{"location":"changelog/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed problem displaying LaTeX in the documentation\nMinor updates to the style guide\nUpdated to MOI v1.4.0 in the documentation","category":"page"},{"location":"changelog/#Version-1.1.0-(May-25,-2022)","page":"Release notes","title":"Version 1.1.0 (May 25, 2022)","text":"","category":"section"},{"location":"changelog/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added num_constraints(::Model; count_variable_in_set_constraints) to simplify the process of counting the number of constraints in a model\nAdded VariableRef(::ConstraintRef) for querying the variable associated with a bound or integrality constraint.\nAdded set_normalized_coefficients for modifying the variable coefficients of a vector-valued constraint.\nAdded set_string_names_on_creation to disable creating String names for variables and constraints. This can improve performance.","category":"page"},{"location":"changelog/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug passing nothing to the start keyword of @variable","category":"page"},{"location":"changelog/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"New tutorials:\nSensitivity analysis of a linear program\nServing web apps\nMinimal ellipse SDP tutorial refactored and improved\nDocs updated to the latest version of each package\nLots of minor fixes and improvements to the documentation","category":"page"},{"location":"changelog/#Version-1.0.0-(March-24,-2022)","page":"Release notes","title":"Version 1.0.0 (March 24, 2022)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Read more about this release, along with an acknowledgement of all the contributors in our JuMP 1.0.0 is released blog post.","category":"page"},{"location":"changelog/#Breaking-3","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The previously deprecated functions (v0.23.0, v0.23.1) have been removed. Deprecation was to improve consistency of function names:\nnum_nl_constraints (see num_nonlinear_constraints)\nall_nl_constraints (see all_nonlinear_constraints)\nadd_NL_expression (see add_nonlinear_expression)\nset_NL_objective  (see set_nonlinear_objective)\nadd_NL_constraint (see add_nonlinear_constraint)\nnl_expr_string (see nonlinear_expr_string)\nnl_constraint_string (see nonlinear_constraint_string)\nSymMatrixSpace (see SymmetricMatrixSpace)\nThe unintentionally exported variable JuMP.op_hint has been renamed to the unexported JuMP._OP_HINT","category":"page"},{"location":"changelog/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .nl files\nFixed a bug broadcasting SparseAxisArrays","category":"page"},{"location":"changelog/#Version-0.23.2-(March-14,-2022)","page":"Release notes","title":"Version 0.23.2 (March 14, 2022)","text":"","category":"section"},{"location":"changelog/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added relative_gap to solution_summary\nregister now throws an informative error if the function is not differentiable using ForwardDiff. In some cases, the check in register will encounter a false negative, and the informative error will be thrown at run-time. This usually happens when the function is non-differentiable in a subset of the domain.","category":"page"},{"location":"changelog/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a scoping issue when extending the container keyword of containers","category":"page"},{"location":"changelog/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Docs updated to the latest version of each package","category":"page"},{"location":"changelog/#Version-0.23.1-(March-2,-2022)","page":"Release notes","title":"Version 0.23.1 (March 2, 2022)","text":"","category":"section"},{"location":"changelog/#Deprecated","page":"Release notes","title":"Deprecated","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"nl_expr_string and nl_constraint_string have been renamed to nonlinear_expr_string and nonlinear_constraint_string. The old methods still exist with deprecation warnings. This change should impact very few users because to call them you must rely on private internals of the nonlinear API. Users are encouraged to use sprint(show, x) instead, where x is the nonlinear expression or constraint of interest.","category":"page"},{"location":"changelog/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for Base.abs2(x) where x is a variable or affine expression. This is mainly useful for complex-valued constraints.","category":"page"},{"location":"changelog/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed addition of complex and real affine expressions\nFixed arithmetic for Complex-valued quadratic expressions\nFixed variable bounds passed as Rational{Int}(Inf)\nFixed printing of the coefficient (0 + 1im)\nFixed a bug when solution_summary is called prior to optimize!","category":"page"},{"location":"changelog/#Version-0.23.0-(February-25,-2022)","page":"Release notes","title":"Version 0.23.0 (February 25, 2022)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP v0.23.0 is a breaking release. It is also a release-candidate for JuMP v1.0.0. That is, if no issues are found with the v0.23.0 release, then it will be re-tagged as v1.0.0.","category":"page"},{"location":"changelog/#Breaking-4","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nMathOptInterface has been updated to v1.0.0\nAll previously deprecated functionality has been removed\nPrintMode, REPLMode and IJuliaMode have been removed in favor of the MIME types MIME\"text/plain\" and MIME\"text/latex\". Replace instances of ::Type{REPLMode} with ::MIME\"text/plain\", REPLMode with MIME(\"text/plain\"), ::Type{IJuliaMode} with ::MIME\"text/latex\", and IJuliaMode with MIME(\"text/latex\").\nFunctions containing the nl_ acronym have been renamed to the more explicit nonlinear_. For example, num_nl_constraints is now num_nonlinear_constraints and set_NL_objective is now set_nonlinear_objective. Calls to the old functions throw an error explaining the new name.\nSymMatrixSpace has been renamed to SymmetricMatrixSpace","category":"page"},{"location":"changelog/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added nonlinear_dual_start_value and set_nonlinear_dual_start_value\nAdded preliminary support for Complex coefficient types","category":"page"},{"location":"changelog/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary","category":"page"},{"location":"changelog/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"MILP examples have been migrated from GLPK to HiGHS\nFixed various typos\nImproved section on setting constraint start values","category":"page"},{"location":"changelog/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated functionality. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Then install JuMP v0.22.3:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add JuMP@0.22.3\"","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"changelog/#Version-0.22.3-(February-10,-2022)","page":"Release notes","title":"Version 0.22.3 (February 10, 2022)","text":"","category":"section"},{"location":"changelog/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a reproducibility issue in the TSP tutorial\nFixed a reproducibility issue in the max_cut_sdp tutorial\nFixed a bug broadcasting an empty SparseAxisArray","category":"page"},{"location":"changelog/#Other-25","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added a warning and improved documentation for the modify-then-query case\nFixed a typo in the docstring of RotatedSecondOrderCone\nAdded Aqua.jl as a check for code health\nAdded introductions to each section of the tutorials\nImproved the column generation and Benders decomposition tutorials\nUpdated documentation to MOI v0.10.8\nUpdated JuliaFormatter to v0.22.2","category":"page"},{"location":"changelog/#Version-0.22.2-(January-10,-2022)","page":"Release notes","title":"Version 0.22.2 (January 10, 2022)","text":"","category":"section"},{"location":"changelog/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The function all_nl_constraints now returns all nonlinear constraints in a model\nstart_value and set_start_value can now be used to get and set the primal start for constraint references\nPlural macros now return a tuple containing the elements that were defined instead of nothing\nAnonymous variables are now printed as _[i] where i is the index of the variable instead of noname. Calling name(x) still returns \"\" so this is non-breaking.","category":"page"},{"location":"changelog/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed handling of min and max in nonlinear expressions\nCartesianIndex is no longer allowed as a key for DenseAxisArrays.","category":"page"},{"location":"changelog/#Other-26","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the performance of GenericAffExpr\nAdded a tutorial on the Travelling Salesperson Problem\nAdded a tutorial on querying the Hessian of a nonlinear program\nAdded documentation on using custom solver binaries.","category":"page"},{"location":"changelog/#Version-0.22.1-(November-29,-2021)","page":"Release notes","title":"Version 0.22.1 (November 29, 2021)","text":"","category":"section"},{"location":"changelog/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Export OptimizationSense enum, with instances: MIN_SENSE, MAX_SENSE, and FEASIBILITY_SENSE\nAdd Base.isempty(::Model) to match Base.empty(::Model)","category":"page"},{"location":"changelog/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in container with tuples as indices\nFix bug in set_time_limit_sec","category":"page"},{"location":"changelog/#Other-27","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add tutorial \"Design patterns for larger models\"\nRemove release notes section from PDF\nGeneral edits of the documentation and error messages","category":"page"},{"location":"changelog/#Version-0.22.0-(November-10,-2021)","page":"Release notes","title":"Version 0.22.0 (November 10, 2021)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP v0.22 is a breaking release","category":"page"},{"location":"changelog/#Breaking-5","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP 0.22 contains a number of breaking changes. However, these should be invisible for the majority of users. You will mostly encounter these breaking changes if you: wrote a JuMP extension, accessed backend(model), or called @SDconstraint.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The breaking changes are as follows:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"MathOptInterface has been updated to v0.10.4. For users who have interacted with the MOI backend, this contains a large number of breaking changes. Read the MathOptInterface release notes for more details.\nThe bridge_constraints keyword argument to Model and set_optimizer has been renamed add_bridges to reflect that more thing were bridged than just constraints.\nThe backend(model) field now contains a concrete instance of a MOI.Utilities.CachingOptimizer instead of one with an abstractly typed optimizer field. In most cases, this will lead to improved performance. However, calling set_optimizer after backend invalidates the old backend. For example:\nmodel = Model()\nb = backend(model)\nset_optimizer(model, GLPK.Optimizer)\n@variable(model, x)\n# b is not updated with `x`! Get a new b by calling `backend` again.\nnew_b = backend(model)\nAll usages of @SDconstraint are deprecated. The new syntax is @constraint(model, X >= Y, PSDCone()).\nCreating a DenseAxisArray with a Number as an axis will now display a warning. This catches a common error in which users write @variable(model, x[length(S)]) instead of @variable(model, x[1:length(S)]).\nThe caching_mode argument to Model, for example, Model(caching_mode = MOIU.MANUAL) mode has been removed. For more control over the optimizer, use direct_model instead.\nThe previously deprecated lp_objective_perturbation_range and lp_rhs_perturbation_range functions have been removed. Use lp_sensitivity_report instead.\nThe .m fields of NonlinearExpression and NonlinearParameter have been renamed to .model.\nInfinite variable bounds are now ignored. Thus, @variable(model, x <= Inf) will show has_upper_bound(x) == false. Previously, these bounds were passed through to the solvers which caused numerical issues for solvers expecting finite bounds.\nThe variable_type and constraint_type functions were removed. This should only affect users who previously wrote JuMP extensions. The functions can be deleted without consequence.\nThe internal functions moi_mode, moi_bridge_constraints, moi_add_constraint, and moi_add_to_function_constant are no longer exported.\nThe un-used method Containers.generate_container has been deleted.\nThe Containers API has been refactored, and _build_ref_sets is now public as Containers.build_ref_sets.\nThe parse_constraint_ methods for extending @constraint at parse time have been refactored in a breaking way. Consult the Extensions documentation for more details and examples.","category":"page"},{"location":"changelog/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The TerminationStatusCode and ResultStatusCode enums are now exported by JuMP. Prefer termination_status(model) == OPTIMAL instead of == MOI.OPTIMAL, although the MOI. prefix way still works.\nCopy a x::DenseAxisArray to an Array by calling Array(x).\nNonlinearExpression is now a subtype of AbstractJuMPScalar\nConstraints such as @constraint(model, x + 1 in MOI.Integer()) are now supported.\nprimal_feasibility_report now accepts a function as the first argument.\nScalar variables @variable(model, x[1:2] in MOI.Integer()) creates two variables, both of which are constrained to be in the set MOI.Integer.\nConic constraints can now be specified as inequalities under a different partial ordering. So @constraint(model, x - y in MOI.Nonnegatives()) can now be written as @constraint(model, x >= y, MOI.Nonnegatives()).\nNames are now set for vectorized constraints.","category":"page"},{"location":"changelog/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue when show was called on a SparseAxisArray with a large number of elements.\nFixed a bug displaying barrier and simplex iterations in solution_summary.\nFixed a bug by implementing hash for DenseAxisArray and SparseAxisArray.\nNames are now only set if the solver supports them. Previously, this prevented solvers such as Ipopt from being used with direct_model.\nMutableArithmetics.Zero is converted into a 0.0 before being returned to the user. Previously, some calls to @expression would return the undocumented MutableArithmetics.Zero() object. One example is summing over an empty set @expression(model, sum(x[i] for i in 1:0)). You will now get 0.0 instead.\nAffExpr and QuadExpr can now be used with == 0 instead of iszero. This fixes a number of issues relating to Julia standard libraries such as LinearAlgebra and SparseArrays.\nFixed a bug when registering a user-defined function with splatting.","category":"page"},{"location":"changelog/#Other-28","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The documentation is now available as a PDF.\nThe documentation now includes a full copy of the MathOptInterface documentation to make it easy to link concepts between the docs. (The MathOptInterface documentation has also been significantly improved.)\nThe documentation contains a large number of improvements and clarifications on a range of topics. Thanks to @sshin23, @DilumAluthge, and @jlwether.\nThe documentation is now built with Julia 1.6 instead of 1.0.\nVarious error messages have been improved to be more readable.","category":"page"},{"location":"changelog/#Version-0.21.10-(September-4,-2021)","page":"Release notes","title":"Version 0.21.10 (September 4, 2021)","text":"","category":"section"},{"location":"changelog/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added add_NL_expression\nadd_NL_xxx functions now support AffExpr and QuadExpr as terms","category":"page"},{"location":"changelog/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in solution_summary\nFixed a bug in relax_integrality","category":"page"},{"location":"changelog/#Other-29","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved error message in lp_sensitivity_report","category":"page"},{"location":"changelog/#Version-0.21.9-(August-1,-2021)","page":"Release notes","title":"Version 0.21.9 (August 1, 2021)","text":"","category":"section"},{"location":"changelog/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Containers now support arbitrary container types by passing the type to the container keyword and overloading Containers.container.\nis_valid now supports nonlinear constraints\nAdded unsafe_backend for querying the inner-most optimizer of a JuMP model.\nNonlinear parameters now support the plural @NLparameters macro.\nContainers (for example, DenseAxisArray) can now be used in vector-valued constraints.","category":"page"},{"location":"changelog/#Other-30","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Various improvements to the documentation.","category":"page"},{"location":"changelog/#Version-0.21.8-(May-8,-2021)","page":"Release notes","title":"Version 0.21.8 (May 8, 2021)","text":"","category":"section"},{"location":"changelog/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The @constraint macro is now extendable in the same way as @variable.\nAffExpr and QuadExpr can now be used in nonlinear macros.","category":"page"},{"location":"changelog/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in lp_sensitivity_report.\nFixed an inference issue when creating empty SparseAxisArrays.","category":"page"},{"location":"changelog/#Version-0.21.7-(April-12,-2021)","page":"Release notes","title":"Version 0.21.7 (April 12, 2021)","text":"","category":"section"},{"location":"changelog/#Added-26","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added primal_feasibility_report, which can be used to check whether a primal point satisfies primal feasibility.\nAdded coefficient, which returns the coefficient associated with a variable in affine and quadratic expressions.\nAdded copy_conflict, which returns the IIS of an infeasible model.\nAdded solution_summary, which returns (and prints) a struct containing a summary of the solution.\nAllow AbstractVector in vector constraints instead of just Vector.\nAdded latex_formulation(model) which returns an object representing the latex formulation of a model. Use print(latex_formulation(model)) to print the formulation as a string.\nUser-defined functions in nonlinear expressions are now automatically registered to aid quick model prototyping. However, a warning is printed to encourage the manual registration.\nDenseAxisArray's now support broadcasting over multiple arrays.\nContainer indices can now be iterators of Base.SizeUnknown.","category":"page"},{"location":"changelog/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug in rad2deg and deg2rad in nonlinear expressions.\nFixed a MethodError bug in Containers when forcing container type.\nAllow partial slicing of a DenseAxisArray, resolving an issue from 2014.\nFixed a bug printing variable names in IJulia.\nEnding an IJulia cell with model now prints a summary of the model (like in the REPL) not the latex formulation. Use print(model) to print the latex formulation.\nFixed a bug when copying models containing nested arrays.","category":"page"},{"location":"changelog/#Other-31","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Tutorials are now part of the documentation, and more refactoring has taken place.\nAdded JuliaFormatter added as a code formatter.\nAdded some precompilation statements to reduce initial latency.\nVarious improvements to error messages to make them more helpful.\nImproved performance of value(::NonlinearExpression).\nImproved performance of fix(::VariableRef).","category":"page"},{"location":"changelog/#Version-0.21.6-(January-29,-2021)","page":"Release notes","title":"Version 0.21.6 (January 29, 2021)","text":"","category":"section"},{"location":"changelog/#Added-27","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for skew symmetric variables via @variable(model, X[1:2, 1:2] in SkewSymmetricMatrixSpace()).\nlp_sensitivity_report has been added which significantly improves the performance of querying the sensitivity summary of an LP. lp_objective_perturbation_range and lp_rhs_perturbation_range are deprecated.\nDual warm-starts are now supported with set_dual_start_value and dual_start_value.\n∈ (\\in<tab>) can now be used in macros instead of = or in.\nUse haskey(model::Model, key::Symbol) to check if a name key is registered in a model.\nAdded unregister(model::Model, key::Symbol) to unregister a name key from model.\nAdded callback_node_status for use in callbacks.\nAdded print_bridge_graph to visualize the bridging graph generated by MathOptInterface.\nImproved error message for containers with duplicate indices.","category":"page"},{"location":"changelog/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Various fixes to pass tests on Julia 1.6.\nFixed a bug in the printing of nonlinear expressions in IJulia.\nFixed a bug when nonlinear expressions are passed to user-defined functions.\nSome internal functions that were previously exported are now no longer exported.\nFixed a bug when relaxing a fixed binary variable.\nFixed a StackOverflowError that occurred when SparseAxisArrays had a large number of elements.\nRemoved an unnecessary type assertion in list_of_constraint_types.\nFixed a bug when copying models with registered expressions.","category":"page"},{"location":"changelog/#Other-32","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The documentation has been significantly overhauled. It now has distinct sections for the manual, API reference, and examples. The existing examples in /examples have now been moved to /docs/src/examples and rewritten using Literate.jl, and they are now included in the documentation.\nJuliaFormatter has been applied to most of the codebase. This will continue to roll out over time, as we fix upstream issues in the formatter, and will eventually become compulsory.\nThe root cause of a large number of method invalidations has been resolved.\nWe switched continuous integration from Travis and Appveyor to GitHub Actions.","category":"page"},{"location":"changelog/#Version-0.21.5-(September-18,-2020)","page":"Release notes","title":"Version 0.21.5 (September 18, 2020)","text":"","category":"section"},{"location":"changelog/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed deprecation warnings\nThrow DimensionMismatch for incompatibly sized functions and sets\nUnify treatment of keys(x) on JuMP containers","category":"page"},{"location":"changelog/#Version-0.21.4-(September-14,-2020)","page":"Release notes","title":"Version 0.21.4 (September 14, 2020)","text":"","category":"section"},{"location":"changelog/#Added-28","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add debug info when adding unsupported constraints\nAdd relax_integrality for solving continuous relaxation\nAllow querying constraint conflicts","category":"page"},{"location":"changelog/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Dispatch on Real for MOI.submit\nImplement copy for CustomSet in tests\nDon't export private macros\nFix invalid assertion in nonlinear\nError if constraint has NaN right-hand side\nImprove speed of tests\nLots of work modularizing files in /test\nImprove line numbers in macro error messages\nPrint nonlinear subexpressions\nVarious documentation updates\nDependency updates:\nDatastructures 0.18\nMathOptFormat v0.5\nPrep for MathOptInterface 0.9.15","category":"page"},{"location":"changelog/#Version-0.21.3-(June-18,-2020)","page":"Release notes","title":"Version 0.21.3 (June 18, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added Special Order Sets (SOS1 and SOS2) to JuMP with default weights to ease the creation of such constraints (#2212).\nAdded functions simplex_iterations, barrier_iterations and node_count (#2201).\nAdded function reduced_cost (#2205).\nImplemented callback_value for affine and quadratic expressions (#2231).\nSupport MutableArithmetics.Zero in objective and constraints (#2219).\nDocumentation improvements:\nMention tutorials in the docs (#2223).\nUpdate COIN-OR links (#2242).\nExplicit link to the documentation of MOI.FileFormats (#2253).\nTypo fixes (#2261).\nContainers improvements:\nFix Base.map for DenseAxisArray (#2235).\nThrow BoundsError if number of indices is incorrect for DenseAxisArray and SparseAxisArray (#2240).\nExtensibility improvements:\nImplement a set_objective method fallback that redirects to set_objective_sense and set_objective_function (#2247).\nAdd parse_constraint method with arbitrary number of arguments (#2051).\nAdd parse_constraint_expr and parse_constraint_head (#2228).","category":"page"},{"location":"changelog/#Version-0.21.2-(April-2,-2020)","page":"Release notes","title":"Version 0.21.2 (April 2, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added relative_gap() to access MOI.RelativeGap() attribute (#2199).\nDocumentation fixes:\nAdded link to source for docstrings in the documentation (#2207).\nAdded docstring for @variables macro (#2216).\nTypo fixes (#2177, #2184, #2182).\nImplementation of methods for Base functions:\nImplemented Base.empty! for JuMP.Model (#2198).\nImplemented Base.conj for JuMP scalar types (#2209).","category":"page"},{"location":"changelog/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixed sum of expression with scalar product in macro (#2178).\nFixed writing of nonlinear models to MathOptFormat (#2181).\nFixed construction of empty SparseAxisArray (#2179).\nFixed constraint with zero function (#2188).","category":"page"},{"location":"changelog/#Version-0.21.1-(Feb-18,-2020)","page":"Release notes","title":"Version 0.21.1 (Feb 18, 2020)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved the clarity of the with_optimizer deprecation warning.","category":"page"},{"location":"changelog/#Version-0.21.0-(Feb-16,-2020)","page":"Release notes","title":"Version 0.21.0 (Feb 16, 2020)","text":"","category":"section"},{"location":"changelog/#Breaking-6","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Deprecated with_optimizer (#2090, #2084, #2141). You can replace with_optimizer by either nothing, optimizer_with_attributes or a closure:\nreplace with_optimizer(Ipopt.Optimizer) by Ipopt.Optimizer.\nreplace with_optimizer(Ipopt.Optimizer, max_cpu_time=60.0) by optimizer_with_attributes(Ipopt.Optimizer, \"max_cpu_time\" => 60.0).\nreplace with_optimizer(Gurobi.Optimizer, env) by () -> Gurobi.Optimizer(env).\nreplace with_optimizer(Gurobi.Optimizer, env, Presolve=0) by optimizer_with_attributes(() -> Gurobi.Optimizer(env), \"Presolve\" => 0).\nalternatively to optimizer_with_attributes, you can also set the attributes separately with set_optimizer_attribute.\nRenamed set_parameter and set_parameters to set_optimizer_attribute and set_optimizer_attributes (#2150).\nBroadcast should now be explicit inside macros. @SDconstraint(model, x >= 1) and @constraint(model, x + 1 in SecondOrderCone()) now throw an error instead of broadcasting 1 along the dimension of x (#2107).\n@SDconstraint(model, x >= 0) is now equivalent to @constraint(model, x in PSDCone()) instead of @constraint(model, (x .- 0) in PSDCone()) (#2107).\nThe macros now create the containers with map instead of for loops, as a consequence, containers created by @expression can now have any element type and containers of constraint references now have concrete element types when possible. This fixes a long-standing issue where @expression could only be used to generate a collection of linear expressions. Now it works for quadratic expressions as well (#2070).\nCalling deepcopy(::AbstractModel) now throws an error.\nThe constraint name is now printed in the model string (#2108).","category":"page"},{"location":"changelog/#Added-29","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Added support for solver-independent and solver-specific callbacks (#2101).\nAdded write_to_file and read_from_file, supported formats are CBF, LP, MathOptFormat, MPS and SDPA (#2114).\nAdded support for complementarity constraints (#2132).\nAdded support for indicator constraints (#2092).\nAdded support for querying multiple solutions with the result keyword (#2100).\nAdded support for constraining variables on creation (#2128).\nAdded method delete that deletes a vector of variables at once if it is supported by the underlying solver (#2135).\nThe arithmetic between JuMP expression has be refactored into the MutableArithmetics package (#2107).\nImproved error on complex values in NLP (#1978).\nAdded an example of column generation (#2010).","category":"page"},{"location":"changelog/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Incorrect coefficients generated when using Symmetric variables (#2102)","category":"page"},{"location":"changelog/#Version-0.20.1-(Oct-18,-2019)","page":"Release notes","title":"Version 0.20.1 (Oct 18, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add sections on @variables and @constraints in the documentation (#2062).\nFixed product of sparse matrices for Julia v1.3 (#2063).\nAdded set_objective_coefficient to modify the coefficient of a linear term of the objective function (#2008).\nAdded set_time_limit_sec, unset_time_limit_sec and time_limit_sec to set and query the time limit for the solver in seconds (#2053).","category":"page"},{"location":"changelog/#Version-0.20.0-(Aug-24,-2019)","page":"Release notes","title":"Version 0.20.0 (Aug 24, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Documentation updates.\nNumerous bug fixes.\nBetter error messages (#1977, #1978, #1997, #2017).\nPerformance improvements (#1947, #2032).\nAdded LP sensitivity summary functions lp_objective_perturbation_range and lp_rhs_perturbation_range (#1917).\nAdded functions dual_objective_value, raw_status and set_parameter.\nAdded function set_objective_coefficient to modify the coefficient of a linear term of the objective (#2008).\nAdded functions set_normalized_rhs, normalized_rhs, and add_to_function_constant to modify and get the constant part of a constraint (#1935, #1960).\nAdded functions set_normalized_coefficient and normalized_coefficient to modify and get the coefficient of a linear term of a constraint (#1935, #1960).\nNumerous other improvements in MOI 0.9, see the NEWS.md file of MOI for more details.","category":"page"},{"location":"changelog/#Version-0.19.2-(June-8,-2019)","page":"Release notes","title":"Version 0.19.2 (June 8, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in derivatives that could arise in models with nested nonlinear subexpressions.","category":"page"},{"location":"changelog/#Version-0.19.1-(May-12,-2019)","page":"Release notes","title":"Version 0.19.1 (May 12, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Usability and performance improvements.\nBug fixes.","category":"page"},{"location":"changelog/#Version-0.19.0-(February-15,-2019)","page":"Release notes","title":"Version 0.19.0 (February 15, 2019)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP 0.19 contains significant breaking changes.","category":"page"},{"location":"changelog/#Breaking-7","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"JuMP's abstraction layer for communicating with solvers changed from MathProgBase (MPB) to MathOptInterface (MOI). MOI addresses many longstanding design issues. (See @mlubin's slides from JuMP-dev 2018.) JuMP 0.19 is compatible only with solvers that have been updated for MOI. See the installation guide for a list of solvers that have and have not yet been updated.\nMost solvers have been renamed to PackageName.Optimizer. For example, GurobiSolver() is now Gurobi.Optimizer.\nSolvers are no longer added to a model via Model(solver = XXX(kwargs...)). Instead use Model(with_optimizer(XXX, kwargs...)). For example, Model(with_optimizer(Gurobi.Optimizer, OutputFlag=0)).\nJuMP containers (for example, the objects returned by @variable) have been redesigned. Containers.SparseAxisArray replaces JuMPDict, JuMPArray was rewritten (inspired by AxisArrays) and renamed Containers.DenseAxisArray, and you can now request a container type with the container= keyword to the macros. See the corresponding documentation for more details.\nThe statuses returned by solvers have changed. See the possible status values here. The MOI statuses are much richer than the MPB statuses and can be used to distinguish between previously indistinguishable cases (for example, did the solver have a feasible solution when it stopped because of the time limit?).\nStarting values are separate from result values. Use value to query the value of a variable in a solution. Use start_value and set_start_value to get and set an initial starting point provided to the solver. The solutions from previous solves are no longer automatically set as the starting points for the next solve.\nThe data structures for affine and quadratic expressions AffExpr and QuadExpr have changed. Internally, terms are stored in dictionaries instead of lists. Duplicate coefficients can no longer exist. Accessors and iteration methods have changed.\nJuMPNLPEvaluator no longer includes the linear and quadratic parts of the model in the evaluation calls. These are now handled separately to allow NLP solvers that support various types of constraints.\nJuMP solver-independent callbacks have been replaced by solver-specific callbacks. See your favorite solver for more details. (See the note below: No solver-specific callbacks are implemented yet.)\nThe norm() syntax is no longer recognized inside macros. Use the SecondOrderCone() set instead.\nJuMP no longer performs automatic transformation between special quadratic forms and second-order cone constraints. Support for these constraint classes depends on the solver.\nThe symbols :Min and :Max are no longer used as optimization senses. Instead, JuMP uses the OptimizationSense enum from MathOptInterface. @objective(model, Max, ...), @objective(model, Min, ...), @NLobjective(model, Max, ...), and @objective(model, Min, ...) remain valid, but @objective(m, :Max, ...) is no longer accepted.\nThe sign conventions for duals has changed in some cases for consistency with conic duality (see the documentation). The shadow_price helper method returns duals with signs that match conventional LP interpretations of dual values as sensitivities of the objective value to relaxations of constraints.\n@constraintref is no longer defined. Instead, create the appropriate container to hold constraint references manually. For example,\nconstraints = Dict() # Optionally, specify types for improved performance.\nfor i in 1:N\n  constraints[i] = @constraint(model, ...)\nend\nThe lowerbound, upperbound, and basename keyword arguments to the @variable macro have been renamed to lower_bound, upper_bound, and base_name, for consistency with JuMP's new style recommendations.\nWe rely on broadcasting syntax to apply accessors to collections of variables, for example, value.(x) instead of getvalue(x) for collections. (Use value(x) when x is a scalar object.)","category":"page"},{"location":"changelog/#Added-30","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Splatting (like f(x...)) is recognized in restricted settings in nonlinear expressions.\nSupport for deleting constraints and variables.\nThe documentation has been completely rewritten using docstrings and Documenter.\nSupport for modeling mixed conic and quadratic models (for example, conic models with quadratic objectives and bi-linear matrix inequalities).\nSignificantly improved support for modeling new types of constraints and for extending JuMP's macros.\nSupport for providing dual warm starts.\nImproved support for accessing solver-specific attributes (for example, the irreducible inconsistent subsystem).\nExplicit control of whether symmetry-enforcing constraints are added to PSD constraints.\nSupport for modeling exponential cones.\nSignificant improvements in internal code quality and testing.\nStyle and naming guidelines.\nDirect mode and manual mode provide explicit control over when copies of a model are stored or regenerated. See the corresponding documentation.","category":"page"},{"location":"changelog/#Regressions","page":"Release notes","title":"Regressions","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"There are known regressions from JuMP 0.18 that will be addressed in a future release (0.19.x or later):","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Performance regressions in model generation (issue). Please file an issue anyway if you notice a significant performance regression. We have plans to address a number of performance issues, but we might not be aware of all of them.\nFast incremental NLP solves are not yet reimplemented (issue).\nWe do not yet have an implementation of solver-specific callbacks.\nThe column generation syntax in @variable has been removed (that is, the objective, coefficients, and inconstraints keyword arguments). Support for column generation will be re-introduced in a future release.\nThe ability to solve the continuous relaxation (that is, via solve(model; relaxation = true)) is not yet reimplemented (issue).","category":"page"},{"location":"changelog/#Version-0.18.5-(December-1,-2018)","page":"Release notes","title":"Version 0.18.5 (December 1, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support views in some derivative evaluation functions.\nImproved compatibility with PackageCompiler.","category":"page"},{"location":"changelog/#Version-0.18.4-(October-8,-2018)","page":"Release notes","title":"Version 0.18.4 (October 8, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in model printing on Julia 0.7 and 1.0.","category":"page"},{"location":"changelog/#Version-0.18.3-(October-1,-2018)","page":"Release notes","title":"Version 0.18.3 (October 1, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for Julia v1.0 (Thanks @ExpandingMan)\nFix matrix expressions with quadratic functions (#1508)","category":"page"},{"location":"changelog/#Version-0.18.2-(June-10,-2018)","page":"Release notes","title":"Version 0.18.2 (June 10, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in second-order derivatives when expressions are present (#1319)\nFix a bug in @constraintref (#1330)","category":"page"},{"location":"changelog/#Version-0.18.1-(April-9,-2018)","page":"Release notes","title":"Version 0.18.1 (April 9, 2018)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for nested tuple destructuring (#1193)\nPreserve internal model when relaxation=true (#1209)\nMinor bug fixes and updates for example","category":"page"},{"location":"changelog/#Version-0.18.0-(July-27,-2017)","page":"Release notes","title":"Version 0.18.0 (July 27, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Drop support for Julia 0.5.\nUpdate for ForwardDiff 0.5.\nMinor bug fixes.","category":"page"},{"location":"changelog/#Version-0.17.1-(June-9,-2017)","page":"Release notes","title":"Version 0.17.1 (June 9, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Use of constructconstraint! in @SDconstraint.\nMinor bug fixes.","category":"page"},{"location":"changelog/#Version-0.17.0-(May-27,-2017)","page":"Release notes","title":"Version 0.17.0 (May 27, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: Mixing quadratic and conic constraints is no longer supported.\nBreaking change: The getvariable and getconstraint functions are replaced by indexing on the corresponding symbol. For instance, to access the variable with name x, one should now write m[:x] instead of getvariable(m, :x). As a consequence, creating a variable and constraint with the same name now triggers a warning, and accessing one of them afterwards throws an error. This change is breaking only in the latter case.\nAddition of the getobjectivebound function that mirrors the functionality of the MathProgBase getobjbound function except that it takes into account transformations performed by JuMP.\nMinor bug fixes.","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The following changes are primarily of interest to developers of JuMP extensions:","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The new syntax @constraint(model, expr in Cone) creates the constraint ensuring that expr is inside Cone. The Cone argument is passed to constructconstraint! which enables the call to the dispatched to an extension.\nThe @variable macro now calls constructvariable! instead of directly calling the Variable constructor. Extra arguments and keyword arguments passed to @variable are passed to constructvariable! which enables the call to be dispatched to an extension.\nRefactor the internal function conicdata (used build the MathProgBase conic model) into smaller sub-functions to make these parts reusable by extensions.","category":"page"},{"location":"changelog/#Version-0.16.2-(March-28,-2017)","page":"Release notes","title":"Version 0.16.2 (March 28, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Minor bug fixes and printing tweaks\nAddress deprecation warnings for Julia 0.6","category":"page"},{"location":"changelog/#Version-0.16.1-(March-7,-2017)","page":"Release notes","title":"Version 0.16.1 (March 7, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Better support for AbstractArray in JuMP (Thanks @tkoolen)\nMinor bug fixes","category":"page"},{"location":"changelog/#Version-0.16.0-(February-23,-2017)","page":"Release notes","title":"Version 0.16.0 (February 23, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: JuMP no longer has a mechanism for selecting solvers by default (the previous mechanism was flawed and incompatible with Julia 0.6). Not specifying a solver before calling solve() will result in an error.\nBreaking change: User-defined functions are no longer global. The first argument to JuMP.register is now a JuMP Model object within whose scope the function will be registered. Calling JuMP.register without a Model now produces an error.\nBreaking change: Use the new JuMP.fix method to fix a variable to a value or to update the value to which a variable is fixed. Calling setvalue on a fixed variable now results in an error in order to avoid silent behavior changes. (Thanks @joaquimg)\nNonlinear expressions now print out similarly to linear/quadratic expressions (useful for debugging!)\nNew category keyword to @variable. Used for specifying categories of anonymous variables.\nCompatibility with Julia 0.6-dev.\nMinor fixes and improvements (Thanks @cossio, @ccoffrin, @blegat)","category":"page"},{"location":"changelog/#Version-0.15.1-(January-31,-2017)","page":"Release notes","title":"Version 0.15.1 (January 31, 2017)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bugfix for @LinearConstraints and friends","category":"page"},{"location":"changelog/#Version-0.15.0-(December-22,-2016)","page":"Release notes","title":"Version 0.15.0 (December 22, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.5.0 is the minimum required version for this release.\nDocument support for BARON solver\nEnable info callbacks in more states than before, for example, for recording solutions. New when argument to addinfocallback (#814, thanks @yeesian)\nImproved support for anonymous variables. This includes new warnings for potentially confusing use of the traditional non-anonymous syntax:\nWhen multiple variables in a model are given the same name\nWhen non-symbols are used as names, for example, @variable(m, x[1][1:N])\nImprovements in iterating over JuMP containers (#836, thanks @IssamT)\nSupport for writing variable names in .lp file output (Thanks @leethargo)\nSupport for querying duals to SDP problems (Thanks @blegat)\nThe comprehension syntax with curly braces sum{}, prod{}, and norm2{} has been deprecated in favor of Julia's native comprehension syntax sum(), prod() and norm() as previously announced. (For early adopters of the new syntax, norm2() was renamed to norm() without deprecation.)\nUnit tests rewritten to use Base.Test instead of FactCheck\nImproved support for operations with matrices of JuMP types (Thanks @ExpandingMan)\nThe syntax to halt a solver from inside a callback has changed from throw(CallbackAbort()) to return JuMP.StopTheSolver\nMinor bug fixes","category":"page"},{"location":"changelog/#Version-0.14.2-(December-12,-2016)","page":"Release notes","title":"Version 0.14.2 (December 12, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Allow singleton anonymous variables (includes bugfix)","category":"page"},{"location":"changelog/#Version-0.14.1-(September-12,-2016)","page":"Release notes","title":"Version 0.14.1 (September 12, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"More consistent handling of states in informational callbacks, includes a new when parameter to addinfocallback for specifying in which state an informational callback should be called.","category":"page"},{"location":"changelog/#Version-0.14.0-(August-7,-2016)","page":"Release notes","title":"Version 0.14.0 (August 7, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility with Julia 0.5 and ForwardDiff 0.2\nSupport for \"anonymous\" variables, constraints, expressions, and parameters, for example, x = @variable(m, [1:N]) instead of @variable(m, x[1:N])\nSupport for retrieving constraints from a model by name via getconstraint\n@NLconstraint now returns constraint references (as expected).\nSupport for vectorized expressions within lazy constraints\nOn Julia 0.5, parse new comprehension syntax sum(x[i] for i in 1:N if isodd(i)) instead of sum{ x[i], i in 1:N; isodd(i) }. The old syntax with curly braces will be deprecated in JuMP 0.15.\nNow possible to provide nonlinear expressions as \"raw\" Julia Expr objects instead of using JuMP's nonlinear macros. This input format is useful for programmatically generated expressions.\ns/Mathematical Programming/Mathematical Optimization/\nSupport for local cuts (Thanks to @madanim, Mehdi Madani)\nDocument Xpress interface developed by @joaquimg, Joaquim Dias Garcia\nMinor bug and deprecation fixes (Thanks @odow, @jrevels)","category":"page"},{"location":"changelog/#Version-0.13.2-(May-16,-2016)","page":"Release notes","title":"Version 0.13.2 (May 16, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility update for MathProgBase","category":"page"},{"location":"changelog/#Version-0.13.1-(May-3,-2016)","page":"Release notes","title":"Version 0.13.1 (May 3, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix broken deprecation for registerNLfunction.","category":"page"},{"location":"changelog/#Version-0.13.0-(April-29,-2016)","page":"Release notes","title":"Version 0.13.0 (April 29, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Most exported methods and macros have been renamed to avoid camelCase. See the list of changes here. There is a 1-1 mapping from the old names to the new, and it is safe to simply replace the names to update existing models.\nSpecify variable lower/upper bounds in @variable using the lowerbound and upperbound keyword arguments.\nChange name printed for variable using the basename keyword argument to @variable.\nNew @variables macro allows multi-line declaration of groups of variables.\nA number of solver methods previously available only through MathProgBase are now exposed directly in JuMP. The fix was recorded live.\nCompatibility fixes with Julia 0.5.\nThe \"end\" indexing syntax is no longer supported within JuMPArrays which do not use 1-based indexing until upstream issues are resolved, see here.","category":"page"},{"location":"changelog/#Version-0.12.2-(March-9,-2016)","page":"Release notes","title":"Version 0.12.2 (March 9, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Small fixes for nonlinear optimization","category":"page"},{"location":"changelog/#Version-0.12.1-(March-1,-2016)","page":"Release notes","title":"Version 0.12.1 (March 1, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a regression in slicing for JuMPArrays (when not using 1-based indexing)","category":"page"},{"location":"changelog/#Version-0.12.0-(February-27,-2016)","page":"Release notes","title":"Version 0.12.0 (February 27, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"The automatic differentiation functionality has been completely rewritten with a number of user-facing changes:\n@defExpr and @defNLExpr now take the model as the first argument. The previous one-argument version of @defExpr is deprecated; all expressions should be named. For example, replace @defExpr(2x+y) with @defExpr(jump_model, my_expr, 2x+y).\nJuMP no longer uses Julia's variable binding rules for efficiently re-solving a sequence of nonlinear models. Instead, we have introduced nonlinear parameters. This is a breaking change, so we have added a warning message when we detect models that may depend on the old behavior.\nSupport for user-defined functions integrated within nonlinear JuMP expressions.\nReplaced iteration over AffExpr with Number-like scalar iteration; previous iteration behavior is now available via linearterms(::AffExpr).\nStopping the solver via throw(CallbackAbort()) from a callback no longer triggers an exception. Instead, solve() returns UserLimit status.\ngetDual() now works for conic problems (Thanks @emreyamangil.)","category":"page"},{"location":"changelog/#Version-0.11.3-(February-4,-2016)","page":"Release notes","title":"Version 0.11.3 (February 4, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bug-fix for problems with quadratic objectives and semidefinite constraints","category":"page"},{"location":"changelog/#Version-0.11.2-(January-14,-2016)","page":"Release notes","title":"Version 0.11.2 (January 14, 2016)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Compatibility update for Mosek","category":"page"},{"location":"changelog/#Version-0.11.1-(December-1,-2015)","page":"Release notes","title":"Version 0.11.1 (December 1, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Remove usage of @compat in tests.\nFix updating quadratic objectives for nonlinear models.","category":"page"},{"location":"changelog/#Version-0.11.0-(November-30,-2015)","page":"Release notes","title":"Version 0.11.0 (November 30, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.4.0 is the minimum required version for this release.\nFix for scoping semantics of index variables in sum{}. Index variables no longer leak into the surrounding scope.\nAddition of the solve(m::Model, relaxation=true) keyword argument to solve the standard continuous relaxation of model m\nThe getConstraintBounds() method allows access to the lower and upper bounds of all constraints in a (nonlinear) model.\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"changelog/#Version-0.10.3-(November-20,-2015)","page":"Release notes","title":"Version 0.10.3 (November 20, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a rare error when parsing quadratic expressions\nFix Variable() constructor with default arguments\nDetect unrecognized keywords in solve()","category":"page"},{"location":"changelog/#Version-0.10.2-(September-28,-2015)","page":"Release notes","title":"Version 0.10.2 (September 28, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for deprecation warnings","category":"page"},{"location":"changelog/#Version-0.10.1-(September-3,-2015)","page":"Release notes","title":"Version 0.10.1 (September 3, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixes for ambiguity warnings.\nFix for breaking change in precompilation syntax in Julia 0.4-pre","category":"page"},{"location":"changelog/#Version-0.10.0-(August-31,-2015)","page":"Release notes","title":"Version 0.10.0 (August 31, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support (on Julia 0.4 and later) for conditions in indexing @defVar and @addConstraint constructs, for example, @defVar(m, x[i=1:5,j=1:5; i+j >= 3])\nSupport for vectorized operations on Variables and expressions. See the documentation for details.\nNew getVar() method to access variables in a model by name\nSupport for semidefinite programming.\nDual solutions are now available for general nonlinear problems. You may call getDual on a reference object for a nonlinear constraint, and getDual on a variable object for Lagrange multipliers from active bounds.\nIntroduce warnings for two common performance traps: too many calls to getValue() on a collection of variables and use of the + operator in a loop to sum expressions.\nSecond-order cone constraints can be written directly with the norm() and norm2{} syntax.\nImplement MathProgBase interface for querying Hessian-vector products.\nIteration over JuMPContainers is deprecated; instead, use the keys and values functions, and zip(keys(d),values(d)) for the old behavior.\n@defVar returns Array{Variable,N} when each of N index sets are of the form 1:nᵢ.\nModule precompilation: on Julia 0.4 and later, using JuMP is now much faster.","category":"page"},{"location":"changelog/#Version-0.9.3-(August-11,-2015)","page":"Release notes","title":"Version 0.9.3 (August 11, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fixes for FactCheck testing on julia v0.4.","category":"page"},{"location":"changelog/#Version-0.9.2-(June-27,-2015)","page":"Release notes","title":"Version 0.9.2 (June 27, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in @addConstraints.","category":"page"},{"location":"changelog/#Version-0.9.1-(April-25,-2015)","page":"Release notes","title":"Version 0.9.1 (April 25, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix for Julia 0.4-dev.\nSmall infrastructure improvements for extensions.","category":"page"},{"location":"changelog/#Version-0.9.0-(April-18,-2015)","page":"Release notes","title":"Version 0.9.0 (April 18, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Comparison operators for constructing constraints (for example, 2x >= 1) have been deprecated. Instead, construct the constraints explicitly in the @addConstraint macro to add them to the model, or in the @LinearConstraint macro to create a stand-alone linear constraint instance.\ngetValue() method implemented to compute the value of a nonlinear subexpression\nJuMP is now released under the Mozilla Public License version 2.0 (was previously LGPL). MPL is a copyleft license which is less restrictive than LGPL, especially for embedding JuMP within other applications.\nA number of performance improvements in ReverseDiffSparse for computing derivatives.\nMathProgBase.getsolvetime(m) now returns the solution time reported by the solver, if available. (Thanks @odow, Oscar Dowson)\nFormatting fix for LP format output. (Thanks @sbebo, Leonardo Taccari).","category":"page"},{"location":"changelog/#Version-0.8.0-(February-17,-2015)","page":"Release notes","title":"Version 0.8.0 (February 17, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Nonlinear subexpressions now supported with the @defNLExpr macro.\nSCS supported for solving second-order conic problems.\nsetXXXCallback family deprecated in favor of addXXXCallback.\nMultiple callbacks of the same type can be registered.\nAdded support for informational callbacks via addInfoCallback.\nA CallbackAbort exception can be thrown from callback to safely exit optimization.","category":"page"},{"location":"changelog/#Version-0.7.4-(February-4,-2015)","page":"Release notes","title":"Version 0.7.4 (February 4, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Reduced costs and linear constraint duals are now accessible when quadratic constraints are present.\nTwo-sided nonlinear constraints are supported.\nMethods for accessing the number of variables and constraints in a model are renamed.\nNew default procedure for setting initial values in nonlinear optimization: project zero onto the variable bounds.\nSmall bug fixes.","category":"page"},{"location":"changelog/#Version-0.7.3-(January-14,-2015)","page":"Release notes","title":"Version 0.7.3 (January 14, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a method ambiguity conflict with Compose.jl (cosmetic fix)","category":"page"},{"location":"changelog/#Version-0.7.2-(January-9,-2015)","page":"Release notes","title":"Version 0.7.2 (January 9, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in sum(::JuMPDict)\nAdded the setCategory function to change a variables category (for example, continuous or binary)","category":"page"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"after construction, and getCategory to retrieve the variable category.","category":"page"},{"location":"changelog/#Version-0.7.1-(January-2,-2015)","page":"Release notes","title":"Version 0.7.1 (January 2, 2015)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in parsing linear expressions in macros. Affects only Julia 0.4 and later.","category":"page"},{"location":"changelog/#Version-0.7.0-(December-29,-2014)","page":"Release notes","title":"Version 0.7.0 (December 29, 2014)","text":"","category":"section"},{"location":"changelog/#Linear/quadratic/conic-programming","page":"Release notes","title":"Linear/quadratic/conic programming","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Breaking change: The syntax for column-wise model generation has been changed to use keyword arguments in @defVar.\nOn Julia 0.4 and later, variables and coefficients may be multiplied in any order within macros. That is, variable*coefficient is now valid syntax.\nECOS supported for solving second-order conic problems.","category":"page"},{"location":"changelog/#_nonlinear_programming_release_notes","page":"Release notes","title":"Nonlinear programming","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for skipping model generation when solving a sequence of nonlinear models with changing data.\nFix a memory leak when solving a sequence of nonlinear models.\nThe @addNLConstraint macro now supports the three-argument version to define sets of nonlinear constraints.\nKNITRO supported as a nonlinear solver.\nSpeed improvements for model generation.\nThe @addNLConstraints macro supports adding multiple (groups of) constraints at once. Syntax is similar to @addConstraints.\nDiscrete variables allowed in nonlinear problems for solvers which support them (currently only KNITRO).","category":"page"},{"location":"changelog/#General","page":"Release notes","title":"General","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Starting values for variables may now be specified with @defVar(m, x, start=value).\nThe setSolver function allows users to change the solver subsequent to model creation.\nSupport for \"fixed\" variables via the @defVar(m, x == 1) syntax.\nUnit tests rewritten to use FactCheck.jl, improved testing across solvers.","category":"page"},{"location":"changelog/#Version-0.6.3-(October-19,-2014)","page":"Release notes","title":"Version 0.6.3 (October 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in multiplying two AffExpr objects.","category":"page"},{"location":"changelog/#Version-0.6.2-(October-11,-2014)","page":"Release notes","title":"Version 0.6.2 (October 11, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Further improvements and bug fixes for printing.\nFixed a bug in @defExpr.\nSupport for accessing expression graphs through the MathProgBase NLP interface.","category":"page"},{"location":"changelog/#Version-0.6.1-(September-19,-2014)","page":"Release notes","title":"Version 0.6.1 (September 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improvements and bug fixes for printing.","category":"page"},{"location":"changelog/#Version-0.6.0-(September-9,-2014)","page":"Release notes","title":"Version 0.6.0 (September 9, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Julia 0.3.0 is the minimum required version for this release.\nbuildInternalModel(m::Model) added to build solver-level model in memory without optimizing.\nDeprecate load_model_only keyword argument to solve.\nAdd groups of constraints with @addConstraints macro.\nUnicode operators now supported, including ∑ for sum, ∏ for prod, and ≤/≥\nQuadratic constraints supported in @addConstraint macro.\nQuadratic objectives supported in @setObjective macro.\nMathProgBase solver-independent interface replaces Ipopt-specific interface for nonlinear problems\nBreaking change: IpoptOptions no longer supported to specify solver options, use m = Model(solver=IpoptSolver(options...)) instead.\nNew solver interfaces: ECOS, NLopt, and nonlinear support for MOSEK\nNew option to control whether the lazy constraint callback is executed at each node in the B&B tree or just when feasible solutions are found\nAdd support for semicontinuous and semi-integer variables for those solvers that support them.\nAdd support for index dependencies (for example, triangular indexing) in @defVar, @addConstraint, and @defExpr (for example, @defVar(m, x[i=1:10,j=i:10])).\nThis required some changes to the internal structure of JuMP containers, which may break code that explicitly stored JuMPDict objects.","category":"page"},{"location":"changelog/#Version-0.5.8-(September-24,-2014)","page":"Release notes","title":"Version 0.5.8 (September 24, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug with specifying solvers (affects Julia 0.2 only)","category":"page"},{"location":"changelog/#Version-0.5.7-(September-5,-2014)","page":"Release notes","title":"Version 0.5.7 (September 5, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in printing models","category":"page"},{"location":"changelog/#Version-0.5.6-(September-2,-2014)","page":"Release notes","title":"Version 0.5.6 (September 2, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Add support for semicontinuous and semi-integer variables for those solvers that support them.\nBreaking change: Syntax for Variable() constructor has changed (use of this interface remains discouraged)\nUpdate for breaking changes in MathProgBase","category":"page"},{"location":"changelog/#Version-0.5.5-(July-6,-2014)","page":"Release notes","title":"Version 0.5.5 (July 6, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug with problem modification: adding variables that did not appear in existing constraints or objective.","category":"page"},{"location":"changelog/#Version-0.5.4-(June-19,-2014)","page":"Release notes","title":"Version 0.5.4 (June 19, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for breaking change in MathProgBase which reduces loading times for using JuMP\nFix error when MIPs not solved to optimality","category":"page"},{"location":"changelog/#Version-0.5.3-(May-21,-2014)","page":"Release notes","title":"Version 0.5.3 (May 21, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for breaking change in ReverseDiffSparse","category":"page"},{"location":"changelog/#Version-0.5.2-(May-9,-2014)","page":"Release notes","title":"Version 0.5.2 (May 9, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix compatibility with Julia 0.3 prerelease","category":"page"},{"location":"changelog/#Version-0.5.1-(May-5,-2014)","page":"Release notes","title":"Version 0.5.1 (May 5, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix a bug in coefficient handling inside lazy constraints and user cuts","category":"page"},{"location":"changelog/#Version-0.5.0-(May-2,-2014)","page":"Release notes","title":"Version 0.5.0 (May 2, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for nonlinear optimization with exact, sparse second-order derivatives automatically computed. Ipopt is currently the only solver supported.\ngetValue for AffExpr and QuadExpr\nBreaking change: getSolverModel replaced by getInternalModel, which returns the internal MathProgBase-level model\nGroups of constraints can be specified with @addConstraint (see documentation for details). This is not a breaking change.\ndot(::JuMPDict{Variable},::JuMPDict{Variable}) now returns the corresponding quadratic expression.","category":"page"},{"location":"changelog/#Version-0.4.1-(March-24,-2014)","page":"Release notes","title":"Version 0.4.1 (March 24, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Fix bug where change in objective sense was ignored when re-solving a model.\nFix issue with handling zero coefficients in AffExpr.","category":"page"},{"location":"changelog/#Version-0.4.0-(March-10,-2014)","page":"Release notes","title":"Version 0.4.0 (March 10, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Support for SOS1 and SOS2 constraints.\nSolver-independent callback for user heuristics.\ndot and sum implemented for JuMPDict objects. Now you can say @addConstraint(m, dot(a,x) <= b).\nDevelopers: support for extensions to JuMP. See definition of Model in src/JuMP.jl for more details.\nOption to construct the low-level model before optimizing.","category":"page"},{"location":"changelog/#Version-0.3.2-(February-17,-2014)","page":"Release notes","title":"Version 0.3.2 (February 17, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Improved model printing\nPreliminary support for IJulia output","category":"page"},{"location":"changelog/#Version-0.3.1-(January-30,-2014)","page":"Release notes","title":"Version 0.3.1 (January 30, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Documentation updates\nSupport for MOSEK\nCPLEXLink renamed to CPLEX","category":"page"},{"location":"changelog/#Version-0.3.0-(January-21,-2014)","page":"Release notes","title":"Version 0.3.0 (January 21, 2014)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Unbounded/infeasibility rays: getValue() will return the corresponding components of an unbounded ray when a model is unbounded, if supported by the selected solver. getDual() will return an infeasibility ray (Farkas proof) if a model is infeasible and the selected solver supports this feature.\nSolver-independent callbacks for user generated cuts.\nUse new interface for solver-independent QCQP.\nsetlazycallback renamed to setLazyCallback for consistency.","category":"page"},{"location":"changelog/#Version-0.2.0-(December-15,-2013)","page":"Release notes","title":"Version 0.2.0 (December 15, 2013)","text":"","category":"section"},{"location":"changelog/#Breaking-8","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Objective sense is specified in setObjective instead of in the Model constructor.\nlpsolver and mipsolver merged into single solver option.","category":"page"},{"location":"changelog/#Added-31","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Problem modification with efficient LP restarts and MIP warm-starts.\nRelatedly, column-wise modeling now supported.\nSolver-independent callbacks supported. Currently we support only a \"lazy constraint\" callback, which works with Gurobi, CPLEX, and GLPK. More callbacks coming soon.","category":"page"},{"location":"changelog/#Version-0.1.2-(November-16,-2013)","page":"Release notes","title":"Version 0.1.2 (November 16, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Bug fixes for printing, improved error messages.\nAllow AffExpr to be used in macros; for example, ex = y + z; @addConstraint(m, x + 2*ex <= 3)","category":"page"},{"location":"changelog/#Version-0.1.1-(October-23,-2013)","page":"Release notes","title":"Version 0.1.1 (October 23, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Update for solver specification API changes in MathProgBase.","category":"page"},{"location":"changelog/#Version-0.1.0-(October-3,-2013)","page":"Release notes","title":"Version 0.1.0 (October 3, 2013)","text":"","category":"section"},{"location":"changelog/","page":"Release notes","title":"Release notes","text":"Initial public release.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"EditURL = \"https://github.com/JuliaPolyhedra/CDDLib.jl/blob/9700752f849845bc8d8477928baf3df920daadd6/README.md\"","category":"page"},{"location":"packages/CDDLib/#CDDLib","page":"JuliaPolyhedra/CDDLib.jl","title":"CDDLib","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"(Image: ) (Image: ) (Image: )","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl is a wrapper for cddlib.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl can be used with C API of cddlib, the higher level interface of Polyhedra.jl, or as a linear programming solver with JuMP or MathOptInterface.","category":"page"},{"location":"packages/CDDLib/#Problem-description","page":"JuliaPolyhedra/CDDLib.jl","title":"Problem description","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"As written in the README of cddlib:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"The C-library  cddlib is a C implementation of the Double Description Method of Motzkin et al. for generating all vertices (that is, extreme points) and extreme rays of a general convex polyhedron in R^d given by a system of linear inequalities:P = { x=(x1, ..., xd)^T :  b - A  x  >= 0 }where  A  is a given m x d real matrix, b is a given m-vector and 0 is the m-vector of all zeros.The program can be used for the reverse operation (that is, convex hull computation).  This means that  one can move back and forth between an inequality representation  and a generator (that is, vertex and ray) representation of a polyhedron with cdd.  Also, cdd can solve a linear programming problem, that is, a problem of maximizing and minimizing a linear function over P.","category":"page"},{"location":"packages/CDDLib/#License","page":"JuliaPolyhedra/CDDLib.jl","title":"License","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl is licensed under the GPL v2 license.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"The underlying solver, cddlib/cddlib is also licensed under the GPL v2 license.","category":"page"},{"location":"packages/CDDLib/#Installation","page":"JuliaPolyhedra/CDDLib.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Install CDDLib.jl using the Julia package manager:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"import Pkg\nPkg.add(\"CDDLib\")","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Building the package will download binaries of cddlib that are provided by cddlib_jll.jl.","category":"page"},{"location":"packages/CDDLib/#Use-with-JuMP","page":"JuliaPolyhedra/CDDLib.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"Use CDDLib.Optimizer{Float64} to use CDDLib.jl with JuMP:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"using JuMP, CDDLib\nmodel = Model(CDDLib.Optimizer{Float64})","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"When using CDDLib.jl with MathOptInterface, you can pass a different number type:","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"using MathOptInterface, CDDLib\nmodel = CDDLib.Optimizer{Rational{BigInt}}()","category":"page"},{"location":"packages/CDDLib/#Debugging","page":"JuliaPolyhedra/CDDLib.jl","title":"Debugging","text":"","category":"section"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"CDDLib.jl uses two global Boolean variables to enable debugging outputs: debug and log.","category":"page"},{"location":"packages/CDDLib/","page":"JuliaPolyhedra/CDDLib.jl","title":"JuliaPolyhedra/CDDLib.jl","text":"You can query the value of debug and log with get_debug and get_log, and set their values with set_debug and set_log.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"EditURL = \"https://github.com/scipopt/SCIP.jl/blob/v0.11.14/README.md\"","category":"page"},{"location":"packages/SCIP/#SCIP.jl","page":"scipopt/SCIP.jl","title":"SCIP.jl","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"(Image: Build Status) (Image: codecov) (Image: Genie Downloads)","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"SCIP.jl is a Julia interface to the SCIP solver.","category":"page"},{"location":"packages/SCIP/#Affiliation","page":"scipopt/SCIP.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"This wrapper is maintained by the SCIP project with the help of the JuMP community.","category":"page"},{"location":"packages/SCIP/#License","page":"scipopt/SCIP.jl","title":"License","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"SCIP.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The underlying solver, scipopt/scip, is licensed under the Apache 2.0 license.","category":"page"},{"location":"packages/SCIP/#Installation","page":"scipopt/SCIP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Install SCIP.jl using Pkg.add:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"import Pkg\nPkg.add(\"SCIP\")","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"On MacOS and Linux, installing the SCIP Julia package will work out of the box and install the SCIP_jll.jl and SCIP_PaPILO_jll.jl dependencies.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"On Windows, a separate installation of SCIP is still mandatory, as detailed in the \"Custom installation\" section below.","category":"page"},{"location":"packages/SCIP/#Custom-installations","page":"scipopt/SCIP.jl","title":"Custom installations","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"If you use an older Julia version, Windows, or you want a custom SCIP installation, you must manually install the SCIP binaries.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Once installed, set the SCIPOPTDIR environment variable to point to the installation path, that is, depending on your operating system, $SCIPOPTDIR/lib/libscip.so, $SCIPOPTDIR/lib/libscip.dylib, or $SCIPOPTDIR/bin/scip.dll must exist.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Then, install SCIP.jl using Pkg.add and Pkg.build:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"ENV[\"SCIPOPTDIR\"] = \"/Users/Oscar/code/SCIP\"\nimport Pkg\nPkg.add(\"SCIP\")\nPkg.build(\"SCIP\")","category":"page"},{"location":"packages/SCIP/#Use-with-JuMP","page":"scipopt/SCIP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Use SCIP with JuMP as follows:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"using JuMP, SCIP\nmodel = Model(SCIP.Optimizer)\nset_attribute(model, \"display/verblevel\", 0)\nset_attribute(model, \"limits/gap\", 0.05)","category":"page"},{"location":"packages/SCIP/#Options","page":"scipopt/SCIP.jl","title":"Options","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"See the SCIP documentation for a list of supported options.","category":"page"},{"location":"packages/SCIP/#MathOptInterface-API","page":"scipopt/SCIP.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The SCIP optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.Reals","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"MOI.NLPBlock()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/SCIP/#Design-considerations","page":"scipopt/SCIP.jl","title":"Design considerations","text":"","category":"section"},{"location":"packages/SCIP/#Wrapping-the-public-API","page":"scipopt/SCIP.jl","title":"Wrapping the public API","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"All of the public API methods are wrapped and available within the SCIP package. This includes the scip_*.h and pub_*.h headers that are collected in scip.h, as well as all default constraint handlers (cons_*.h.) ","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The wrapped functions do not transform any data structures and work on the raw pointers (for example, SCIP* in C, Ptr{SCIP_} in Julia). Convenience wrapper functions based on Julia types are added as needed.","category":"page"},{"location":"packages/SCIP/#Memory-management","page":"scipopt/SCIP.jl","title":"Memory management","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Programming with SCIP requires dealing with variable and constraint objects that use reference counting for memory management.","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"The SCIP.Optimizer wrapper type collects lists of SCIP_VAR* and SCIP_CONS* under the hood, and it releases all references when it is garbage collected itself (via finalize).","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"When adding a variable (add_variable) or a constraint (add_linear_constraint), an integer index is returned. This index can be used to retrieve the SCIP_VAR* or SCIP_CONS* pointer via get_var and get_cons respectively.","category":"page"},{"location":"packages/SCIP/#Supported-nonlinear-operators","page":"scipopt/SCIP.jl","title":"Supported nonlinear operators","text":"","category":"section"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"Supported operators in nonlinear expressions are as follows:","category":"page"},{"location":"packages/SCIP/","page":"scipopt/SCIP.jl","title":"scipopt/SCIP.jl","text":"+\n-\n*\n/\n^\nsqrt\nexp\nlog\nabs\ncos\nsin","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"EditURL = \"nested_problems.jl\"","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Nested-optimization-problems","page":"Nested optimization problems","title":"Nested optimization problems","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"In this tutorial we explain how to write a nested optimization problem, where an upper problem uses the results from the optimization of a lower subproblem.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"To model the problem, we define a user-defined operator to handle the decomposition of the lower problem inside the upper one. Finally, we show how to improve the performance by using a cache that avoids resolving the lower problem.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"For a simpler example of writing a user-defined operator, see the User-defined Hessians tutorial.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"using JuMP\nimport Ipopt","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"In the rest of this tutorial, our goal is to solve the bilevel optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\nminlimits_xz  x_1^2 + x_2^2 + z \nst             beginarrayr l\n                      z = maxlimits_y  x_1^2 y_1 + x_2^2 y_2  - x_1 y_1^4 - 2 x_2 y_2^4 \n                      st                   (y_1 - 10)^2 + (y_2 - 10)^2 le 25\n                  endarray \n                 x ge 0\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This bilevel optimization problem is composed of two nested optimization problems. An upper level, involving variables x, and a lower level, involving variables y. From the perspective of the lower-level problem, the values of x are fixed parameters, and so the model optimizes y given those fixed parameters. Simultaneously, the upper-level problem optimizes x and z given the response of y.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Decomposition","page":"Nested optimization problems","title":"Decomposition","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"There are a few ways to solve this problem, but we are going to use a nonlinear decomposition method. The first step is to write a function to compute the lower-level problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\n  V(x_1 x_2) = maxlimits_y  x_1^2 y_1 + x_2^2 y_2  - x_1 y_1^4 - 2 x_2 y_2^4 \n                           st  (y_1 - 10)^2 + (y_2 - 10)^2 le 25\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function solve_lower_level(x...)\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, y[1:2])\n    @objective(\n        model,\n        Max,\n        x[1]^2 * y[1] + x[2]^2 * y[2] - x[1] * y[1]^4 - 2 * x[2] * y[2]^4,\n    )\n    @constraint(model, (y[1] - 10)^2 + (y[2] - 10)^2 <= 25)\n    optimize!(model)\n    @assert termination_status(model) == LOCALLY_SOLVED\n    return objective_value(model), value.(y)\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"The next function takes a value of x and returns the optimal lower-level objective-value and the optimal response y. The reason why we need both the objective and the optimal y will be made clear shortly, but for now let us define:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function V(x...)\n    f, _ = solve_lower_level(x...)\n    return f\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Then, we can substitute V into our full problem to create:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"beginarrayr l\nminlimits_x  x_1^2 + x_2^2 + V(x_1 x_2) \nst             x ge 0\nendarray","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"This looks like a nonlinear optimization problem with a user-defined operator V! However, because V solves an optimization problem internally, we can't use automatic differentiation to compute the first and second derivatives. Instead, we can use JuMP's ability to pass callback functions for the gradient and Hessian instead.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"First up, we need to define the gradient of V with respect to x. In general, this may be difficult to compute, but because x appears only in the objective, we can just differentiate the objective function with respect to x, giving:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function ∇V(g::AbstractVector, x...)\n    _, y = solve_lower_level(x...)\n    g[1] = 2 * x[1] * y[1] - y[1]^4\n    g[2] = 2 * x[2] * y[2] - 2 * y[2]^4\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Second, we need to define the Hessian of V with respect to x. This is a symmetric matrix, but in our example only the diagonal elements are non-zero:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function ∇²V(H::AbstractMatrix, x...)\n    _, y = solve_lower_level(x...)\n    H[1, 1] = 2 * y[1]\n    H[2, 2] = 2 * y[2]\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"info: Info\nProviding an explicit Hessian function is optional if first derivatives are already available.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"We now have enough to define our bilevel optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2] >= 0)\n@operator(model, op_V, 2, V, ∇V, ∇²V)\n@objective(model, Min, x[1]^2 + x[2]^2 + op_V(x[1], x[2]))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"The optimal objective value is:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"objective_value(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"and the optimal upper-level decision variables x are:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"value.(x)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"To compute the optimal lower-level decision variables, we need to call solve_lower_level with the optimal upper-level decision variables:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"_, y = solve_lower_level(value.(x)...)\ny","category":"page"},{"location":"tutorials/nonlinear/nested_problems/#Improving-performance","page":"Nested optimization problems","title":"Improving performance","text":"","category":"section"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Our solution approach works, but it has a performance problem: every time we need to compute the value, gradient, or Hessian of V, we have to re-solve the lower-level optimization problem. This is wasteful, because we will often call the gradient and Hessian at the same point, and so solving the problem twice with the same input repeats work unnecessarily.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"We can work around this by using a cache:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"mutable struct Cache\n    x::Any\n    f::Float64\n    y::Vector{Float64}\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"with a function to update the cache if needed:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function _update_if_needed(cache::Cache, x...)\n    if cache.x !== x\n        cache.f, cache.y = solve_lower_level(x...)\n        cache.x = x\n    end\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Then, we define cached versions of out three functions which call _updated_if_needed and return values from the cache.","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"function cached_f(cache::Cache, x...)\n    _update_if_needed(cache, x...)\n    return cache.f\nend\n\nfunction cached_∇f(cache::Cache, g::AbstractVector, x...)\n    _update_if_needed(cache, x...)\n    g[1] = 2 * x[1] * cache.y[1] - cache.y[1]^4\n    g[2] = 2 * x[2] * cache.y[2] - 2 * cache.y[2]^4\n    return\nend\n\nfunction cached_∇²f(cache::Cache, H::AbstractMatrix, x...)\n    _update_if_needed(cache, x...)\n    H[1, 1] = 2 * cache.y[1]\n    H[2, 2] = 2 * cache.y[2]\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"Now we're ready to setup and solve the upper level optimization problem:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2] >= 0)\ncache = Cache(Float64[], NaN, Float64[])\n@operator(\n    model,\n    op_cached_f,\n    2,\n    (x...) -> cached_f(cache, x...),\n    (g, x...) -> cached_∇f(cache, g, x...),\n    (H, x...) -> cached_∇²f(cache, H, x...),\n)\n@objective(model, Min, x[1]^2 + x[2]^2 + op_cached_f(x[1], x[2]))\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"an we can check we get the same objective value:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"objective_value(model)","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"and upper-level decision variable x:","category":"page"},{"location":"tutorials/nonlinear/nested_problems/","page":"Nested optimization problems","title":"Nested optimization problems","text":"value.(x)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"EditURL = \"https://github.com/jump-dev/PolyJuMP.jl/blob/6b280429865850cd278f0d74d0a176d53da2980e/README.md\"","category":"page"},{"location":"packages/PolyJuMP/#PolyJuMP.jl","page":"jump-dev/PolyJuMP.jl","title":"PolyJuMP.jl","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"PolyJuMP.jl is a JuMP extension for formulating and solving polynomial optimization problems. This extension includes the following:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Polynomial functions on JuMP decisions variables. These can be solved with the PolyJuMP.QCQP.Optimizer or PolyJuMP.KKT.Optimizer.\nConstraints that a polynomial is nonnegative where the coefficients of the polynomials depend on JuMP decision variables. These nonnegativity constraints can be reformulated using sufficient conditions using PolyJuMP.RelativeEntropy submodule or SumOfSquares.jl.","category":"page"},{"location":"packages/PolyJuMP/#License","page":"jump-dev/PolyJuMP.jl","title":"License","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"PolyJuMP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/PolyJuMP/#Installation","page":"jump-dev/PolyJuMP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Install PolyJuMP using Pkg.add:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import Pkg\nPkg.add(\"PolyJuMP\")","category":"page"},{"location":"packages/PolyJuMP/#Use-with-JuMP","page":"jump-dev/PolyJuMP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use QCQP solver with JuMP, use a nonconvex QCQP solver, for example Gurobi.Optimizer, and PolyJuMP.QCQP.Optimizer:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using JuMP, PolyJuMP, Gurobi\nmodel = Model(() -> PolyJuMP.QCQP.Optimizer(Gurobi.Optimizer))","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use KKT solver with JuMP, use solver of algebraic systems of equations implementing the SemialgebraicSets interface, for example HomotopyContinuation.SemialgebraicSetsHCSolver, and PolyJuMP.KKT.Optimizer:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using JuMP, PolyJuMP, HomotopyContinuation\nmodel = Model(optimizer_with_attributes(\n    PolyJuMP.KKT.Optimizer,\n    \"solver\" => HomotopyContinuation.SemialgebraicSetsHCSolver(),\n))","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"For a nonnegativity constraint on a polynomial, for example","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"using DynamicPolynomials\n@polyvar x y\nusing JuMP\nmodel = Model()\n@variable(model, a)\n@constraint(model, a * x * y^2 + y^3 >= a * x)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"you need to specify how to interpret this nonnegativity constraint. To use Sum-of-Arithmetic-Geometric-Exponentials (SAGE), use","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import PolyJuMP\nPolyJuMP.setpolymodule!(model, PolyJuMP.SAGE)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"To use Sum-of-Squares (SOS), use","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"import SumOfSquares\nPolyJuMP.setpolymodule!(model, SumOfSquares)","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"or replace model = Model() by model = SOSModel().","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Alternatively, the nonnegativity constraint can be explicit:","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"@constraint(model, a * x * y^2 + y^3 - a * x in PolyJuMP.SAGE.Polynomials())\n@constraint(model, a * x * y^2 + y^3 - a * x in SumOfSquares.SOSCone())","category":"page"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"This allows mixing SAGE and SOS constraints in the same model.","category":"page"},{"location":"packages/PolyJuMP/#Documentation","page":"jump-dev/PolyJuMP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/PolyJuMP/","page":"jump-dev/PolyJuMP.jl","title":"jump-dev/PolyJuMP.jl","text":"Documentation for PolyJuMP.jl is included in the documentation for SumOfSquares.jl.","category":"page"},{"location":"extensions/introduction/#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"This section of the documentation contains brief documentation for some popular JuMP extensions. The list of extensions is not exhaustive, but instead is intended to help you discover popular JuMP extensions, and to give you an overview of the types of extensions that are possible to write with JuMP.","category":"page"},{"location":"extensions/introduction/#Affiliation","page":"Introduction","title":"Affiliation","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Packages beginning with jump-dev/ are developed and maintained by the JuMP developers.","category":"page"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Packages that do not begin with jump-dev/ are developed independently. The developers of these packages requested or consented to the inclusion of their README contents in the JuMP documentation for the benefit of users.","category":"page"},{"location":"extensions/introduction/#Adding-new-extensions","page":"Introduction","title":"Adding new extensions","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Written an extension? Add it to this section of the JuMP documentation by making a pull request to the docs/packages.toml file.","category":"page"},{"location":"extensions/introduction/#Weak-dependencies","page":"Introduction","title":"Weak dependencies","text":"","category":"section"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"Some extensions listed in this section are implemented using the weak dependency feature added to Julia in v1.9. These extensions are activated if and only if you have JuMP and the other package loaded into your current scope with using or import.","category":"page"},{"location":"extensions/introduction/","page":"Introduction","title":"Introduction","text":"compat: Compat\nUsing a weak dependency requires Julia v1.9 or later.","category":"page"},{"location":"installation/#Installation-Guide","page":"Installation Guide","title":"Installation Guide","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This guide explains how to install Julia and  JuMP. If you have installation troubles, read the Common installation issues section below.","category":"page"},{"location":"installation/#Install-Julia","page":"Installation Guide","title":"Install Julia","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMP is a package for Julia. To use JuMP, first download and install Julia.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nIf you are new to Julia, read our Getting started with Julia tutorial.","category":"page"},{"location":"installation/#Choosing-a-version","page":"Installation Guide","title":"Choosing a version","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"You can install the \"Current stable release\" or the \"Long-term support (LTS) release.\"","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"The \"Current stable release\" is the latest release of Julia. It has access to newer features, and is likely faster.\nThe \"Long-term support release\" is an older version of Julia that has continued to receive bug and security fixes. However, it may not have the latest features or performance improvements.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"For most users, you should install the \"Current stable release,\" and whenever Julia releases a new version of the current stable release, you should update your version of Julia. Note that any code you write on one version of the current stable release will continue to work on all subsequent releases.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"For users in restricted software environments (for example, your enterprise IT controls what software you can install), you may be better off installing the long-term support release because you will not have to update Julia as frequently.","category":"page"},{"location":"installation/#Install-JuMP","page":"Installation Guide","title":"Install JuMP","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"From Julia, JuMP is installed using the built-in package manager:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nWe recommend you create a Pkg environment for each project you use JuMP for, instead of adding lots of packages to the global environment. The Pkg manager documentation has more information on this topic.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"When we release a new version of JuMP, you can update with:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.update(\"JuMP\")","category":"page"},{"location":"installation/#Install-a-solver","page":"Installation Guide","title":"Install a solver","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMP depends on solvers to solve optimization problems. Therefore, you will need to install one before you can solve problems with JuMP.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Install a solver using the Julia package manager, replacing \"Clp\" by the Julia package name as appropriate.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"import Pkg\nPkg.add(\"Clp\")","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Once installed, you can use Clp as a solver with JuMP as follows, using set_attribute to set solver-specific options:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"using JuMP\nusing Clp\nmodel = Model(Clp.Optimizer)\nset_attribute(model, \"LogLevel\" => 1)\nset_attribute(model, \"PrimalTolerance\" => 1e-7)","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nMost packages follow the ModuleName.Optimizer naming convention, but exceptions may exist. See the README of the Julia package's GitHub repository for more details on how to use a particular solver, including any solver-specific options.","category":"page"},{"location":"installation/#Supported-solvers","page":"Installation Guide","title":"Supported solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Most solvers are not written in Julia, and some require commercial licenses to use, so installation is often more complex.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"If a solver has Manual in the Installation column, the solver requires a manual installation step, such as downloading and installing a binary, or obtaining a commercial license. Consult the README of the relevant Julia package for more information.\nIf the solver has Manualᴹ in the Installation column, the solver requires an installation of MATLAB.\nIf the Installation column is missing an entry, installing the Julia package will download and install any relevant solver binaries automatically, and you shouldn't need to do anything other than Pkg.add.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Solvers with a missing entry in the Julia Package column are written in Julia. The link in the Solver column is the corresponding Julia package.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Solver Julia Package Installation License Supports\nAlpine.jl   Triad NS (MI)NLP\nArtelys Knitro KNITRO.jl Manual Comm. (MI)LP, (MI)SOCP, (MI)NLP\nBARON BARON.jl Manual Comm. (MI)NLP\nBonmin AmplNLWriter.jl  EPL (MI)NLP\nCbc Cbc.jl  EPL (MI)LP\nCDCS CDCS.jl Manualᴹ GPL LP, SOCP, SDP\nCDD CDDLib.jl  GPL LP\nClarabel.jl   Apache LP, QP, SOCP, SDP\nClp Clp.jl  EPL LP\nCOPT COPT.jl  Comm. (MI)LP, SOCP, SDP\nCOSMO.jl   Apache LP, QP, SOCP, SDP\nCouenne AmplNLWriter.jl  EPL (MI)NLP\nCPLEX CPLEX.jl Manual Comm. (MI)LP, (MI)SOCP\nCSDP CSDP.jl  EPL LP, SDP\nDAQP DAQP.jl  MIT (Mixed-binary) QP\nEAGO.jl   MIT (MI)NLP\nECOS ECOS.jl  GPL LP, SOCP\nFICO Xpress Xpress.jl Manual Comm. (MI)LP, (MI)SOCP\nGLPK GLPK.jl  GPL (MI)LP\nGurobi Gurobi.jl Manual Comm. (MI)LP, (MI)SOCP\nHiGHS HiGHS.jl  MIT (MI)LP, QP\nHypatia.jl   MIT LP, SOCP, SDP\nIpopt Ipopt.jl  EPL LP, QP, NLP\nJuniper.jl   MIT (MI)SOCP, (MI)NLP\nLoraine.jl   MIT LP, SDP\nMadNLP.jl   MIT LP, QP, NLP\nMiniZinc MiniZinc.jl Manual MPL-2 CP-SAT\nMinotaur AmplNLWriter.jl Manual BSD-like (MI)NLP\nMOSEK MosekTools.jl Manual Comm. (MI)LP, (MI)SOCP, SDP\nNLopt NLopt.jl  GPL LP, QP, NLP\nOcteract AmplNLWriter.jl  Comm. (MI)NLP\nOSQP OSQP.jl  Apache LP, QP\nPATH PATHSolver.jl  MIT MCP\nPajarito.jl   MPL-2 (MI)NLP, (MI)SOCP, (MI)SDP\nPavito.jl   MPL-2 (MI)NLP\nPenbmi Penopt.jl  Comm. Bilinear SDP\nPolyJuMP.KKT PolyJuMP.jl  MIT NLP\nPolyJuMP.QCQP PolyJuMP.jl  MIT NLP\nProxSDP.jl   MIT LP, SOCP, SDP\nRAPOSa AmplNLWriter.jl Manual RAPOSa (MI)NLP\nSCIP SCIP.jl  Apache (MI)LP, (MI)NLP\nSCS SCS.jl  MIT LP, QP, SOCP, SDP\nSDPA SDPA.jl, SDPAFamily.jl  GPL LP, SDP\nSDPNAL SDPNAL.jl Manualᴹ CC BY-SA LP, SDP\nSDPT3 SDPT3.jl Manualᴹ GPL LP, SOCP, SDP\nSeDuMi SeDuMi.jl Manualᴹ GPL LP, SOCP, SDP\nStatusSwitchingQP.jl   MIT LP, QP\nTulip.jl   MPL-2 LP","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Where:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"LP = Linear programming\nQP = Quadratic programming\nSOCP = Second-order conic programming (including problems with convex quadratic constraints or objective)\nMCP = Mixed-complementarity programming\nNLP = Nonlinear programming\nSDP = Semidefinite programming\n(MI)XXX = Mixed-integer equivalent of problem type XXX\nCP-SAT = Constraint programming and Boolean satisfiability","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nDeveloped a solver or solver wrapper? This table is open for new contributions. Edit the installation.md file, and use the checklist Adding a new solver to the documentation when opening the pull request.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nDeveloping a solver or solver wrapper? See Models and the MathOptInterface docs for more details on how JuMP interacts with solvers. Please get in touch via the Developer Chatroom with any questions about connecting new solvers with JuMP.","category":"page"},{"location":"installation/#AMPL-based-solvers","page":"Installation Guide","title":"AMPL-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use AmplNLWriter to access solvers that support the NL format.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Some solvers, such as Bonmin and Couenne can be installed via the Julia package manager. Others need to be manually installed.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Consult the AMPL documentation for a complete list of supported solvers.","category":"page"},{"location":"installation/#GAMS-based-solvers","page":"Installation Guide","title":"GAMS-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use GAMS.jl to access solvers available through GAMS. Such solvers include: AlphaECP, Antigone, BARON, CONOPT, Couenne, LocalSolver, PATHNLP, SHOT, SNOPT, SoPlex. See a complete list here.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"note: Note\nGAMS.jl requires an installation of the commercial software GAMS for which a free community license exists.","category":"page"},{"location":"installation/#NEOS-based-solvers","page":"Installation Guide","title":"NEOS-based solvers","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Use NEOSServer.jl to access solvers available through the NEOS Server.","category":"page"},{"location":"installation/#Common-installation-issues","page":"Installation Guide","title":"Common installation issues","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nWhen in doubt, run import Pkg; Pkg.update() to see if updating your packages fixes the issue. Remember you will need to exit Julia and start a new session for the changes to take effect.","category":"page"},{"location":"installation/#Check-the-version-of-your-packages","page":"Installation Guide","title":"Check the version of your packages","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Each package is versioned with a three-part number of the form vX.Y.Z. You can check which versions you have installed with import Pkg; Pkg.status().","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This should almost always be the most-recent release. You can check the releases of a package by going to the relevant GitHub page, and navigating to the \"releases\" page. For example, the list of JuMP releases is available at: https://github.com/jump-dev/JuMP.jl/releases.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"If you post on the community forum, please include the output of Pkg.status().","category":"page"},{"location":"installation/#Unsatisfiable-requirements-detected","page":"Installation Guide","title":"Unsatisfiable requirements detected","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Did you get an error like Unsatisfiable requirements detected for package JuMP? The Pkg documentation has a section on how to understand and manage these conflicts.","category":"page"},{"location":"installation/#Installing-new-packages-can-make-JuMP-downgrade-to-an-earlier-version","page":"Installation Guide","title":"Installing new packages can make JuMP downgrade to an earlier version","text":"","category":"section"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"Another common complaint is that after adding a new package, code that previously worked no longer works.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"This usually happens because the new package is not compatible with the latest version of JuMP. Therefore, the package manager rolls-back JuMP to an earlier version. Here's an example.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"First, we add JuMP:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"(jump_example) pkg> add JuMP\n  Resolving package versions...\nUpdating `~/jump_example/Project.toml`\n  [4076af6c] + JuMP v0.21.5\nUpdating `~/jump_example/Manifest.toml`\n  ... lines omitted ...","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"The + JuMP v0.21.5 line indicates that JuMP has been added at version 0.21.5. However, watch what happens when we add JuMPeR:","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"(jump_example) pkg> add JuMPeR\n  Resolving package versions...\nUpdating `~/jump_example/Project.toml`\n  [4076af6c] ↓ JuMP v0.21.5 ⇒ v0.18.6\n  [707a9f91] + JuMPeR v0.6.0\nUpdating `~/jump_example/Manifest.toml`\n  ... lines omitted ...","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"JuMPeR gets added at version 0.6.0 (+ JuMPeR v0.6.0), but JuMP gets downgraded from 0.21.5 to 0.18.6 (↓ JuMP v0.21.5 ⇒ v0.18.6)! The reason for this is that JuMPeR doesn't support a version of JuMP newer than 0.18.6.","category":"page"},{"location":"installation/","page":"Installation Guide","title":"Installation Guide","text":"tip: Tip\nPay careful attention to the output of the package manager when adding new packages, especially when you see a package being downgraded.","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/errors.md\"","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/errors/#Errors","page":"Errors","title":"Errors","text":"","category":"section"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an MOI call fails on a model, precise errors should be thrown when possible instead of simply calling error with a message. The docstrings for the respective methods describe the errors that the implementation should throw in certain situations. This error-reporting system allows code to distinguish between internal errors (that should be shown to the user) and unsupported operations which may have automatic workarounds.","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an invalid index is used in an MOI call, an InvalidIndex is thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"InvalidIndex","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.InvalidIndex","page":"Errors","title":"MathOptInterface.InvalidIndex","text":"struct InvalidIndex{IndexType<:Index} <: Exception\n    index::IndexType\nend\n\nAn error indicating that the index index is invalid.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"When an invalid result index is used to retrieve an attribute, a ResultIndexBoundsError is thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"ResultIndexBoundsError\ncheck_result_index_bounds","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.ResultIndexBoundsError","page":"Errors","title":"MathOptInterface.ResultIndexBoundsError","text":"struct ResultIndexBoundsError{AttrType} <: Exception\n    attr::AttrType\n    result_count::Int\nend\n\nAn error indicating that the requested attribute attr could not be retrieved, because the solver returned too few results compared to what was requested. For instance, the user tries to retrieve VariablePrimal(2) when only one solution is available, or when the model is infeasible and has no solution.\n\nSee also: check_result_index_bounds.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.check_result_index_bounds","page":"Errors","title":"MathOptInterface.check_result_index_bounds","text":"check_result_index_bounds(model::ModelLike, attr)\n\nThis function checks whether enough results are available in the model for the requested attr, using its result_index field. If the model does not have sufficient results to answer the query, it throws a ResultIndexBoundsError.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"As discussed in JuMP mapping, for scalar constraint with a nonzero function constant, a ScalarFunctionConstantNotZero exception may be thrown:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"ScalarFunctionConstantNotZero","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.ScalarFunctionConstantNotZero","page":"Errors","title":"MathOptInterface.ScalarFunctionConstantNotZero","text":"struct ScalarFunctionConstantNotZero{T, F, S} <: Exception\n    constant::T\nend\n\nAn error indicating that the constant part of the function in the constraint F-in-S is nonzero.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Some VariableIndex constraints cannot be combined on the same variable:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"LowerBoundAlreadySet\nUpperBoundAlreadySet","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.LowerBoundAlreadySet","page":"Errors","title":"MathOptInterface.LowerBoundAlreadySet","text":"LowerBoundAlreadySet{S1, S2}\n\nError thrown when setting a VariableIndex-in-S2 when a VariableIndex-in-S1 has already been added and the sets S1, S2 both set a lower bound, i.e. they are EqualTo, GreaterThan, Interval, Semicontinuous or Semiinteger.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UpperBoundAlreadySet","page":"Errors","title":"MathOptInterface.UpperBoundAlreadySet","text":"UpperBoundAlreadySet{S1, S2}\n\nError thrown when setting a VariableIndex-in-S2 when a VariableIndex-in-S1 has already been added and the sets S1, S2 both set an upper bound, i.e. they are EqualTo, LessThan, Interval, Semicontinuous or Semiinteger.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"As discussed in AbstractCallback, trying to get attributes inside a callback may throw:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"OptimizeInProgress","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.OptimizeInProgress","page":"Errors","title":"MathOptInterface.OptimizeInProgress","text":"struct OptimizeInProgress{AttrType<:AnyAttribute} <: Exception\n    attr::AttrType\nend\n\nError thrown from optimizer when MOI.get(optimizer, attr) is called inside an AbstractCallback while it is only defined once optimize! has completed. This can only happen when is_set_by_optimize(attr) is true.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Trying to submit the wrong type of AbstractSubmittable inside an AbstractCallback (for example, a UserCut inside a LazyConstraintCallback) will throw:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"InvalidCallbackUsage","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.InvalidCallbackUsage","page":"Errors","title":"MathOptInterface.InvalidCallbackUsage","text":"struct InvalidCallbackUsage{C, S} <: Exception\n    callback::C\n    submittable::S\nend\n\nAn error indicating that submittable cannot be submitted inside callback.\n\nFor example, UserCut cannot be submitted inside LazyConstraintCallback.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"The rest of the errors defined in MOI fall in two categories represented by the following two abstract types:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"UnsupportedError\nNotAllowedError","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedError","page":"Errors","title":"MathOptInterface.UnsupportedError","text":"UnsupportedError <: Exception\n\nAbstract type for error thrown when an element is not supported by the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.NotAllowedError","page":"Errors","title":"MathOptInterface.NotAllowedError","text":"NotAllowedError <: Exception\n\nAbstract type for error thrown when an operation is supported but cannot be applied in the current state of the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"The different UnsupportedError and NotAllowedError are the following errors:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"UnsupportedAttribute\nSetAttributeNotAllowed\nAddVariableNotAllowed\nUnsupportedConstraint\nAddConstraintNotAllowed\nModifyConstraintNotAllowed\nModifyObjectiveNotAllowed\nDeleteNotAllowed\nUnsupportedSubmittable\nSubmitNotAllowed\nUnsupportedNonlinearOperator","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedAttribute","page":"Errors","title":"MathOptInterface.UnsupportedAttribute","text":"struct UnsupportedAttribute{AttrType} <: UnsupportedError\n    attr::AttrType\n    message::String\nend\n\nAn error indicating that the attribute attr is not supported by the model, i.e. that supports returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.SetAttributeNotAllowed","page":"Errors","title":"MathOptInterface.SetAttributeNotAllowed","text":"struct SetAttributeNotAllowed{AttrType} <: NotAllowedError\n    attr::AttrType\n    message::String\nend\n\nAn error indicating that the attribute attr is supported (see supports) but cannot be set for some reason (see the error string).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.AddVariableNotAllowed","page":"Errors","title":"MathOptInterface.AddVariableNotAllowed","text":"struct AddVariableNotAllowed <: NotAllowedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that variables cannot be added to the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedConstraint","page":"Errors","title":"MathOptInterface.UnsupportedConstraint","text":"struct UnsupportedConstraint{F<:AbstractFunction, S<:AbstractSet} <: UnsupportedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that constraints of type F-in-S are not supported by the model, i.e. that supports_constraint returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.AddConstraintNotAllowed","page":"Errors","title":"MathOptInterface.AddConstraintNotAllowed","text":"struct AddConstraintNotAllowed{F<:AbstractFunction, S<:AbstractSet} <: NotAllowedError\n    message::String # Human-friendly explanation why the attribute cannot be set\nend\n\nAn error indicating that constraints of type F-in-S are supported (see supports_constraint) but cannot be added.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.ModifyConstraintNotAllowed","page":"Errors","title":"MathOptInterface.ModifyConstraintNotAllowed","text":"struct ModifyConstraintNotAllowed{F<:AbstractFunction, S<:AbstractSet,\n                                  C<:AbstractFunctionModification} <: NotAllowedError\n    constraint_index::ConstraintIndex{F, S}\n    change::C\n    message::String\nend\n\nAn error indicating that the constraint modification change cannot be applied to the constraint of index ci.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.ModifyObjectiveNotAllowed","page":"Errors","title":"MathOptInterface.ModifyObjectiveNotAllowed","text":"struct ModifyObjectiveNotAllowed{C<:AbstractFunctionModification} <: NotAllowedError\n    change::C\n    message::String\nend\n\nAn error indicating that the objective modification change cannot be applied to the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.DeleteNotAllowed","page":"Errors","title":"MathOptInterface.DeleteNotAllowed","text":"struct DeleteNotAllowed{IndexType <: Index} <: NotAllowedError\n    index::IndexType\n    message::String\nend\n\nAn error indicating that the index index cannot be deleted.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedSubmittable","page":"Errors","title":"MathOptInterface.UnsupportedSubmittable","text":"struct UnsupportedSubmittable{SubmitType} <: UnsupportedError\n    sub::SubmitType\n    message::String\nend\n\nAn error indicating that the submittable sub is not supported by the model, i.e. that supports returns false.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.SubmitNotAllowed","page":"Errors","title":"MathOptInterface.SubmitNotAllowed","text":"struct SubmitNotAllowed{SubmitTyp<:AbstractSubmittable} <: NotAllowedError\n    sub::SubmitType\n    message::String\nend\n\nAn error indicating that the submittable sub is supported (see supports) but cannot be added for some reason (see the error string).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/#MathOptInterface.UnsupportedNonlinearOperator","page":"Errors","title":"MathOptInterface.UnsupportedNonlinearOperator","text":"UnsupportedNonlinearOperator(head::Symbol[, message::String]) <: UnsupportedError\n\nAn error thrown by optimizers if they do not support the operator head in a ScalarNonlinearFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> throw(MOI.UnsupportedNonlinearOperator(:black_box))\nERROR: MathOptInterface.UnsupportedNonlinearOperator: The nonlinear operator `:black_box` is not supported by the model.\nStacktrace:\n[...]\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"Note that setting the ConstraintFunction of a VariableIndex constraint is not allowed:","category":"page"},{"location":"moi/reference/errors/","page":"Errors","title":"Errors","text":"SettingVariableIndexNotAllowed","category":"page"},{"location":"moi/reference/errors/#MathOptInterface.SettingVariableIndexNotAllowed","page":"Errors","title":"MathOptInterface.SettingVariableIndexNotAllowed","text":"SettingVariableIndexNotAllowed()\n\nError type that should be thrown when the user calls set to change the ConstraintFunction of a VariableIndex constraint.\n\n\n\n\n\n","category":"type"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"EditURL = \"https://github.com/jump-dev/CPLEX.jl/blob/v1.0.0/README.md\"","category":"page"},{"location":"packages/CPLEX/#CPLEX.jl","page":"jump-dev/CPLEX.jl","title":"CPLEX.jl","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl is a wrapper for the IBM® ILOG® CPLEX® Optimization Studio.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl has two components:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The C API can be accessed via CPLEX.CPXxx functions, where the names and arguments are identical to the C API. See the CPLEX documentation for details.","category":"page"},{"location":"packages/CPLEX/#Affiliation","page":"jump-dev/CPLEX.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"This wrapper is maintained by the JuMP community and is not officially supported by IBM. However, we thank IBM for providing us with a CPLEX license to test CPLEX.jl on GitHub. If you are a commercial customer interested in official support for CPLEX in Julia, let them know.","category":"page"},{"location":"packages/CPLEX/#License","page":"jump-dev/CPLEX.jl","title":"License","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl is licensed under the MIT License.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Free CPLEX licenses are available for academics and students.","category":"page"},{"location":"packages/CPLEX/#Installation","page":"jump-dev/CPLEX.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl requires CPLEX version 12.10, 20.1, or 22.1.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"First, obtain a license of CPLEX and install CPLEX solver, following the instructions on IBM's website.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Once installed, set the CPLEX_STUDIO_BINARIES environment variable as appropriate and run Pkg.add(\"CPLEX\"). For example:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"# On Windows, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"C:\\\\Program Files\\\\CPLEX_Studio1210\\\\cplex\\\\bin\\\\x86-64_win\\\\\"\n# On OSX, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"/Applications/CPLEX_Studio1210/cplex/bin/x86-64_osx/\"\n# On Unix, this might be:\nENV[\"CPLEX_STUDIO_BINARIES\"] = \"/opt/CPLEX_Studio1210/cplex/bin/x86-64_linux/\"\n\nimport Pkg\nPkg.add(\"CPLEX\")","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"note: Note\nThe exact path may differ. Check which folder you installed CPLEX in, and update the path accordingly.","category":"page"},{"location":"packages/CPLEX/#Use-with-JuMP","page":"jump-dev/CPLEX.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Use CPLEX.jl with JuMP as follows:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX\nmodel = Model(CPLEX.Optimizer)\nset_attribute(model, \"CPX_PARAM_EPINT\", 1e-8)","category":"page"},{"location":"packages/CPLEX/#MathOptInterface-API","page":"jump-dev/CPLEX.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"The CPLEX optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}\nMOI.ObjectiveFunction{MOI.VectorAffineFunction{Float64}}","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.Reals","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"MOI.ConflictStatus()\nMOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/CPLEX/#Options","page":"jump-dev/CPLEX.jl","title":"Options","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Options match those of the C API in the CPLEX documentation.","category":"page"},{"location":"packages/CPLEX/#Callbacks","page":"jump-dev/CPLEX.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"CPLEX.jl provides a solver-specific callback to CPLEX:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX, Test\n\nmodel = direct_model(CPLEX.Optimizer())\nset_silent(model)\n\n# This is very, very important!!! Only use callbacks in single-threaded mode.\nMOI.set(model, MOI.NumberOfThreads(), 1)\n\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\ncb_calls = Clong[]\nfunction my_callback_function(cb_data::CPLEX.CallbackContext, context_id::Clong)\n    # You can reference variables outside the function as normal\n    push!(cb_calls, context_id)\n    # You can select where the callback is run\n    if context_id != CPX_CALLBACKCONTEXT_CANDIDATE\n        return\n    end\n    ispoint_p = Ref{Cint}()\n    ret = CPXcallbackcandidateispoint(cb_data, ispoint_p)\n    if ret != 0 || ispoint_p[] == 0\n        return  # No candidate point available or error\n    end\n    # You can query CALLBACKINFO items\n    valueP = Ref{Cdouble}()\n    ret = CPXcallbackgetinfodbl(cb_data, CPXCALLBACKINFO_BEST_BND, valueP)\n    @info \"Best bound is currently: $(valueP[])\"\n    # As well as any other C API\n    x_p = Vector{Cdouble}(undef, 2)\n    obj_p = Ref{Cdouble}()\n    ret = CPXcallbackgetincumbent(cb_data, x_p, 0, 1, obj_p)\n    if ret == 0\n        @info \"Objective incumbent is: $(obj_p[])\"\n        @info \"Incumbent solution is: $(x_p)\"\n        # Use CPLEX.column to map between variable references and the 1-based\n        # column.\n        x_col = CPLEX.column(cb_data, index(x))\n        @info \"x = $(x_p[x_col])\"\n    else\n        # Unable to query incumbent.\n    end\n\n    # Before querying `callback_value`, you must call:\n    CPLEX.load_callback_variable_primal(cb_data, context_id)\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    # You can submit solver-independent MathOptInterface attributes such as\n    # lazy constraints, user-cuts, and heuristic solutions.\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y + x <= 3)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nMOI.set(model, CPLEX.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2","category":"page"},{"location":"packages/CPLEX/#Annotations-for-automatic-Benders'-decomposition","page":"jump-dev/CPLEX.jl","title":"Annotations for automatic Benders' decomposition","text":"","category":"section"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"Here is an example of using the annotation feature for automatic Benders' decomposition:","category":"page"},{"location":"packages/CPLEX/","page":"jump-dev/CPLEX.jl","title":"jump-dev/CPLEX.jl","text":"using JuMP, CPLEX\n\nfunction add_annotation(\n    model::JuMP.Model,\n    variable_classification::Dict;\n    all_variables::Bool = true,\n)\n    num_variables = sum(length(it) for it in values(variable_classification))\n    if all_variables\n        @assert num_variables == JuMP.num_variables(model)\n    end\n    indices, annotations = CPXINT[], CPXLONG[]\n    for (key, value) in variable_classification\n        for variable_ref in value\n            push!(indices, variable_ref.index.value - 1)\n            push!(annotations, CPX_BENDERS_MASTERVALUE + key)\n        end\n    end\n    cplex = backend(model)\n    index_p = Ref{CPXINT}()\n    CPXnewlongannotation(\n        cplex.env,\n        cplex.lp,\n        CPX_BENDERS_ANNOTATION,\n        CPX_BENDERS_MASTERVALUE,\n    )\n    CPXgetlongannotationindex(\n        cplex.env,\n        cplex.lp,\n        CPX_BENDERS_ANNOTATION,\n        index_p,\n    )\n    CPXsetlongannotations(\n        cplex.env,\n        cplex.lp,\n        index_p[],\n        CPX_ANNOTATIONOBJ_COL,\n        length(indices),\n        indices,\n        annotations,\n    )\n    return\nend\n\n# Problem\n\nfunction illustrate_full_annotation()\n    c_1, c_2 = [1, 4], [2, 3]\n    dim_x, dim_y = length(c_1), length(c_2)\n    b = [-2; -3]\n    A_1, A_2 = [1 -3; -1 -3], [1 -2; -1 -1]\n    model = JuMP.direct_model(CPLEX.Optimizer())\n    set_optimizer_attribute(model, \"CPXPARAM_Benders_Strategy\", 1)\n    @variable(model, x[1:dim_x] >= 0, Bin)\n    @variable(model, y[1:dim_y] >= 0)\n    variable_classification = Dict(0 => [x[1], x[2]], 1 => [y[1], y[2]])\n    @constraint(model, A_2 * y + A_1 * x .<= b)\n    @objective(model, Min, c_1' * x + c_2' * y)\n    add_annotation(model, variable_classification)\n    optimize!(model)\n    x_optimal = value.(x)\n    y_optimal = value.(y)\n    println(\"x: $(x_optimal), y: $(y_optimal)\")\nend\n\nfunction illustrate_partial_annotation()\n    c_1, c_2 = [1, 4], [2, 3]\n    dim_x, dim_y = length(c_1), length(c_2)\n    b = [-2; -3]\n    A_1, A_2 = [1 -3; -1 -3], [1 -2; -1 -1]\n    model = JuMP.direct_model(CPLEX.Optimizer())\n    # Note that the \"CPXPARAM_Benders_Strategy\" has to be set to 2 if partial\n    # annotation is provided. If \"CPXPARAM_Benders_Strategy\" is set to 1, then\n    # the following error will be thrown:\n    # `CPLEX Error  2002: Invalid Benders decomposition.`\n    set_optimizer_attribute(model, \"CPXPARAM_Benders_Strategy\", 2)\n    @variable(model, x[1:dim_x] >= 0, Bin)\n    @variable(model, y[1:dim_y] >= 0)\n    variable_classification = Dict(0 => [x[1]], 1 => [y[1], y[2]])\n    @constraint(model, A_2 * y + A_1 * x .<= b)\n    @objective(model, Min, c_1' * x + c_2' * y)\n    add_annotation(model, variable_classification; all_variables = false)\n    optimize!(model)\n    x_optimal = value.(x)\n    y_optimal = value.(y)\n    println(\"x: $(x_optimal), y: $(y_optimal)\")\nend","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"EditURL = nothing","category":"page"},{"location":"api/JuMP/#DocumenterReference_JuMP","page":"JuMP","title":"JuMP","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"This page lists the public API of JuMP.","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"info: Info\nThis page is an unstructured list of the JuMP API. For a more structured overview, read the Manual or Tutorial parts of this documentation.","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"Load all of the public the API into the current scope with:","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"using JuMP","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"Alternatively, load only the module with:","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"import JuMP","category":"page"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"and then prefix all calls with JuMP. to create JuMP.<NAME>.","category":"page"},{"location":"api/JuMP/#@NLconstraint","page":"JuMP","title":"@NLconstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLconstraint","category":"page"},{"location":"api/JuMP/#JuMP.@NLconstraint","page":"JuMP","title":"JuMP.@NLconstraint","text":"@NLconstraint(model::GenericModel, expr)\n\nAdd a constraint described by the nonlinear expression expr. See also @constraint. For example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @NLconstraint(model, sin(x) <= 1)\nsin(x) - 1.0 ≤ 0\n\njulia> @NLconstraint(model, [i = 1:3], sin(i * x) <= 1 / i)\n3-element Vector{NonlinearConstraintRef{ScalarShape}}:\n (sin(1.0 * x) - 1.0 / 1.0) - 0.0 ≤ 0\n (sin(2.0 * x) - 1.0 / 2.0) - 0.0 ≤ 0\n (sin(3.0 * x) - 1.0 / 3.0) - 0.0 ≤ 0\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLconstraints","page":"JuMP","title":"@NLconstraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLconstraints","category":"page"},{"location":"api/JuMP/#JuMP.@NLconstraints","page":"JuMP","title":"JuMP.@NLconstraints","text":"@NLconstraints(model, args...)\n\nAdds multiple nonlinear constraints to model at once, in the same fashion as the @NLconstraint macro.\n\nThe model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the constraints that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, t);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @NLconstraints(model, begin\n           t >= sqrt(x^2 + y^2)\n           [i = 1:2], z[i] <= log(a[i])\n       end)\n((t - sqrt(x ^ 2.0 + y ^ 2.0)) - 0.0 ≥ 0, NonlinearConstraintRef{ScalarShape}[(z[1] - log(4.0)) - 0.0 ≤ 0, (z[2] - log(5.0)) - 0.0 ≤ 0])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLexpression","page":"JuMP","title":"@NLexpression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLexpression","category":"page"},{"location":"api/JuMP/#JuMP.@NLexpression","page":"JuMP","title":"JuMP.@NLexpression","text":"@NLexpression(args...)\n\nEfficiently build a nonlinear expression which can then be inserted in other nonlinear constraints and the objective. See also [@expression]. For example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @NLexpression(model, my_expr, sin(x)^2 + cos(x^2))\nsubexpression[1]: sin(x) ^ 2.0 + cos(x ^ 2.0)\n\njulia> @NLconstraint(model, my_expr + y >= 5)\n(subexpression[1] + y) - 5.0 ≥ 0\n\njulia> @NLobjective(model, Min, my_expr)\n\nIndexing over sets and anonymous expressions are also supported:\n\njulia> @NLexpression(model, my_expr_1[i=1:3], sin(i * x))\n3-element Vector{NonlinearExpression}:\n subexpression[2]: sin(1.0 * x)\n subexpression[3]: sin(2.0 * x)\n subexpression[4]: sin(3.0 * x)\n\njulia> my_expr_2 = @NLexpression(model, log(1 + sum(exp(my_expr_1[i]) for i in 1:2)))\nsubexpression[5]: log(1.0 + (exp(subexpression[2]) + exp(subexpression[3])))\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLexpressions","page":"JuMP","title":"@NLexpressions","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLexpressions","category":"page"},{"location":"api/JuMP/#JuMP.@NLexpressions","page":"JuMP","title":"JuMP.@NLexpressions","text":"@NLexpressions(model, args...)\n\nAdds multiple nonlinear expressions to model at once, in the same fashion as the @NLexpression macro.\n\nThe model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the expressions that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @NLexpressions(model, begin\n           my_expr, sqrt(x^2 + y^2)\n           my_expr_1[i = 1:2], log(a[i]) - z[i]\n       end)\n(subexpression[1]: sqrt(x ^ 2.0 + y ^ 2.0), NonlinearExpression[subexpression[2]: log(4.0) - z[1], subexpression[3]: log(5.0) - z[2]])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLobjective","page":"JuMP","title":"@NLobjective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLobjective","category":"page"},{"location":"api/JuMP/#JuMP.@NLobjective","page":"JuMP","title":"JuMP.@NLobjective","text":"@NLobjective(model, sense, expression)\n\nAdd a nonlinear objective to model with optimization sense sense. sense must be Max or Min.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @NLobjective(model, Max, 2x + 1 + sin(x))\n\njulia> print(model)\nMax 2.0 * x + 1.0 + sin(x)\nSubject to\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLparameter","page":"JuMP","title":"@NLparameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLparameter","category":"page"},{"location":"api/JuMP/#JuMP.@NLparameter","page":"JuMP","title":"JuMP.@NLparameter","text":"@NLparameter(model, param == value)\n\nCreate and return a nonlinear parameter param attached to the model model with initial value set to value. Nonlinear parameters may be used only in nonlinear expressions.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, x == 10)\nx == 10.0\n\njulia> value(x)\n10.0\n\n@NLparameter(model, value = param_value)\n\nCreate and return an anonymous nonlinear parameter param attached to the model model with initial value set to param_value. Nonlinear parameters may be used only in nonlinear expressions.\n\nExample\n\njulia> model = Model();\n\njulia> x = @NLparameter(model, value = 10)\nparameter[1] == 10.0\n\njulia> value(x)\n10.0\n\n@NLparameter(model, param_collection[...] == value_expr)\n\nCreate and return a collection of nonlinear parameters param_collection attached to the model model with initial value set to value_expr (may depend on index sets). Uses the same syntax for specifying index sets as @variable.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, y[i = 1:3] == 2 * i)\n3-element Vector{NonlinearParameter}:\n parameter[1] == 2.0\n parameter[2] == 4.0\n parameter[3] == 6.0\n\njulia> value(y[2])\n4.0\n\n@NLparameter(model, [...] == value_expr)\n\nCreate and return an anonymous collection of nonlinear parameters attached to the model model with initial value set to value_expr (may depend on index sets). Uses the same syntax for specifying index sets as @variable.\n\nExample\n\njulia> model = Model();\n\njulia> y = @NLparameter(model, [i = 1:3] == 2 * i)\n3-element Vector{NonlinearParameter}:\n parameter[1] == 2.0\n parameter[2] == 4.0\n parameter[3] == 6.0\n\njulia> value(y[2])\n4.0\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@NLparameters","page":"JuMP","title":"@NLparameters","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@NLparameters","category":"page"},{"location":"api/JuMP/#JuMP.@NLparameters","page":"JuMP","title":"JuMP.@NLparameters","text":" @NLparameters(model, args...)\n\nCreate and return multiple nonlinear parameters attached to model model, in the same fashion as @NLparameter macro.\n\nThe model must be the first argument, and multiple parameters can be added on multiple lines wrapped in a begin ... end block. Distinct parameters need to be placed on separate lines as in the following example.\n\nThe macro returns a tuple containing the parameters that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameters(model, begin\n           x == 10\n           b == 156\n       end);\n\njulia> value(x)\n10.0\n\n\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@build_constraint","page":"JuMP","title":"@build_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@build_constraint","category":"page"},{"location":"api/JuMP/#JuMP.@build_constraint","page":"JuMP","title":"JuMP.@build_constraint","text":"@build_constraint(constraint_expr)\n\nConstructs a ScalarConstraint or VectorConstraint using the same machinery as @constraint but without adding the constraint to a model.\n\nConstraints using broadcast operators like x .<= 1 are also supported and will create arrays of ScalarConstraint or VectorConstraint.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @build_constraint(2x >= 1)\nScalarConstraint{AffExpr, MathOptInterface.GreaterThan{Float64}}(2 x, MathOptInterface.GreaterThan{Float64}(1.0))\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@constraint","page":"JuMP","title":"@constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@constraint","category":"page"},{"location":"api/JuMP/#JuMP.@constraint","page":"JuMP","title":"JuMP.@constraint","text":"@constraint(m::GenericModel, expr, kw_args...)\n\nAdd a constraint described by the expression expr.\n\n@constraint(m::GenericModel, ref[i=..., j=..., ...], expr, kw_args...)\n\nAdd a group of constraints described by the expression expr parametrized by i, j, ...\n\nThe expression expr can either be\n\nof the form func in set constraining the function func to belong to the set set which is either a MOI.AbstractSet or one of the JuMP shortcuts SecondOrderCone, RotatedSecondOrderCone and PSDCone, e.g. @constraint(model, [1, x-1, y-2] in SecondOrderCone()) constrains the norm of [x-1, y-2] be less than 1;\nof the form a sign b, where sign is one of ==, ≥, >=, ≤ and <= building the single constraint enforcing the comparison to hold for the expression a and b, e.g. @constraint(m, x^2 + y^2 == 1) constrains x and y to lie on the unit circle;\nof the form a ≤ b ≤ c or a ≥ b ≥ c (where ≤ and <= (resp. ≥ and >=) can be used interchangeably) constraining the paired the expression b to lie between a and c;\nof the forms @constraint(m, a .sign b) or @constraint(m, a .sign b .sign c) which broadcast the constraint creation to each element of the vectors.\n\nThe recognized keyword arguments in kw_args are the following:\n\nbase_name: Sets the name prefix used to generate constraint names. It corresponds to the constraint name for scalar constraints, otherwise, the constraint names are set to base_name[...] for each index ... of the axes axes.\ncontainer: Specify the container type.\nset_string_name::Bool = true: control whether to set the MOI.ConstraintName attribute. Passing set_string_name = false can improve performance.\n\nNote for extending the constraint macro\n\nEach constraint will be created using add_constraint(m, build_constraint(_error, func, set)) where\n\n_error is an error function showing the constraint call in addition to the error message given as argument,\nfunc is the expression that is constrained\nand set is the set in which it is constrained to belong.\n\nFor expr of the first type (i.e. @constraint(m, func in set)), func and set are passed unchanged to build_constraint but for the other types, they are determined from the expressions and signs. For instance, @constraint(m, x^2 + y^2 == 1) is transformed into add_constraint(m, build_constraint(_error, x^2 + y^2, MOI.EqualTo(1.0))).\n\nTo extend JuMP to accept new constraints of this form, it is necessary to add the corresponding methods to build_constraint. Note that this will likely mean that either func or set will be some custom type, rather than e.g. a Symbol, since we will likely want to dispatch on the type of the function or set appearing in the constraint.\n\nFor extensions that need to create constraints with more information than just func and set, an additional positional argument can be specified to @constraint that will then be passed on build_constraint. Hence, we can enable this syntax by defining extensions of build_constraint(_error, func, set, my_arg; kw_args...). This produces the user syntax: @constraint(model, ref[...], expr, my_arg, kw_args...).\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@constraints","page":"JuMP","title":"@constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@constraints","category":"page"},{"location":"api/JuMP/#JuMP.@constraints","page":"JuMP","title":"JuMP.@constraints","text":"@constraints(model, args...)\n\nAdds groups of constraints at once, in the same fashion as the @constraint macro.\n\nThe model must be the first argument, and multiple constraints can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the constraints that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, w);\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:3]);\n\njulia> @constraints(model, begin\n           x >= 1\n           y - w <= 2\n           sum_to_one[i=1:3], z[i] + y == 1\n       end);\n\njulia> print(model)\nFeasibility\nSubject to\n sum_to_one[1] : y + z[1] = 1\n sum_to_one[2] : y + z[2] = 1\n sum_to_one[3] : y + z[3] = 1\n x ≥ 1\n -w + y ≤ 2\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@expression","page":"JuMP","title":"@expression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@expression","category":"page"},{"location":"api/JuMP/#JuMP.@expression","page":"JuMP","title":"JuMP.@expression","text":"@expression(args...)\n\nEfficiently builds a linear or quadratic expression but does not add to model immediately. Instead, returns the expression which can then be inserted in other constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:5]);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z);\n\njulia> @expression(model, shared, sum(i * x[i] for i in 1:5))\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5]\n\njulia> @constraint(model, shared + y >= 5)\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5] + y ≥ 5\n\njulia> @constraint(model, shared + z <= 10)\nx[1] + 2 x[2] + 3 x[3] + 4 x[4] + 5 x[5] + z ≤ 10\n\nThe ref accepts index sets in the same way as @variable, and those indices can be used in the construction of the expressions:\n\njulia> @expression(model, expr[i = 1:3], i * sum(x[j] for j in 1:3))\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[1] + 2 x[2] + 2 x[3]\n 3 x[1] + 3 x[2] + 3 x[3]\n\nAnonymous syntax is also supported:\n\njulia> expr = @expression(model, [i in 1:3], i * sum(x[j] for j in 1:3))\n3-element Vector{AffExpr}:\n x[1] + x[2] + x[3]\n 2 x[1] + 2 x[2] + 2 x[3]\n 3 x[1] + 3 x[2] + 3 x[3]\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@expressions","page":"JuMP","title":"@expressions","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@expressions","category":"page"},{"location":"api/JuMP/#JuMP.@expressions","page":"JuMP","title":"JuMP.@expressions","text":"@expressions(model, args...)\n\nAdds multiple expressions to model at once, in the same fashion as the @expression macro.\n\nThe model must be the first argument, and multiple expressions can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the expressions that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @variable(model, z[1:2]);\n\njulia> a = [4, 5];\n\njulia> @expressions(model, begin\n           my_expr, x^2 + y^2\n           my_expr_1[i = 1:2], a[i] - z[i]\n       end)\n(x² + y², AffExpr[-z[1] + 4, -z[2] + 5])\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@objective","page":"JuMP","title":"@objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@objective","category":"page"},{"location":"api/JuMP/#JuMP.@objective","page":"JuMP","title":"JuMP.@objective","text":"@objective(model::GenericModel, sense, func)\n\nSet the objective sense to sense and objective function to func. The objective sense can be either Min, Max, MOI.MIN_SENSE, MOI.MAX_SENSE or MOI.FEASIBILITY_SENSE; see MOI.ObjectiveSense.\n\nIn order to set the sense programmatically, i.e., when sense is a Julia variable whose value is the sense, one of the three MOI.ObjectiveSense values should be used.\n\nExample\n\nTo minimize the value of the variable x, do as follows:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @objective(model, Min, x)\nx\n\nTo maximize the value of the affine expression 2x - 1, do as follows:\n\njulia> @objective(model, Max, 2x - 1)\n2 x - 1\n\nTo set a quadratic objective and set the objective sense programmatically, do as follows:\n\njulia> sense = MIN_SENSE\nMIN_SENSE::OptimizationSense = 0\n\njulia> @objective(model, sense, x^2 - 2x + 1)\nx² - 2 x + 1\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@operator","page":"JuMP","title":"@operator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@operator","category":"page"},{"location":"api/JuMP/#JuMP.@operator","page":"JuMP","title":"JuMP.@operator","text":"@operator(model, operator, dim, f[, ∇f[, ∇²f]])\n\nAdd the nonlinear operator operator in model with dim arguments, and create a new NonlinearOperator object called operator in the current scope.\n\nThe function f evaluates the operator and must return a scalar.\n\nThe optional function ∇f evaluates the first derivative, and the optional function ∇²f evaluates the second derivative.\n\n∇²f may be provided only if ∇f is also provided.\n\nUnivariate syntax\n\nIf dim == 1, then the method signatures of each function must be:\n\nf(::T)::T where {T<:Real}\n∇f(::T)::T where {T<:Real}\n∇²f(::T)::T where {T<:Real}\n\nMultivariate syntax\n\nIf dim > 1, then the method signatures of each function must be:\n\nf(x::T...)::T where {T<:Real}\n∇f(g::AbstractVector{T}, x::T...)::Nothing where {T<:Real}\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T<:Real}\n\nWhere the gradient vector g and Hessian matrix H are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :op_f)\n\njulia> @objective(model, Min, op_f(x))\nop_f(x)\n\njulia> op_f(2.0)\n4.0\n\njulia> model[:op_f]\nNonlinearOperator(f, :op_f)\n\njulia> model[:op_f](x)\nop_f(x)\n\nNon-macro version\n\nThis macro is provided as helpful syntax that matches the style of the rest of the JuMP macros. However, you may also add operators outside the macro using add_nonlinear_operator. For example:\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f)\nNonlinearOperator(f, :op_f)\n\nis equivalent to\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> op_f = model[:op_f] = add_nonlinear_operator(model, 1, f; name = :op_f)\nNonlinearOperator(f, :op_f)\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@variable","page":"JuMP","title":"@variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@variable","category":"page"},{"location":"api/JuMP/#JuMP.@variable","page":"JuMP","title":"JuMP.@variable","text":"@variable(model, expr, args..., kw_args...)\n\nAdd a variable to the model model described by the expression expr, the positional arguments args and the keyword arguments kw_args.\n\nAnonymous and named variables\n\nexpr must be one of the forms:\n\nOmitted, like @variable(model), which creates an anonymous variable\nA single symbol like @variable(model, x)\nA container expression like @variable(model, x[i=1:3])\nAn anoymous container expression like @variable(model, [i=1:3])\n\nBounds\n\nIn addition, the expression can have bounds, such as:\n\n@variable(model, x >= 0)\n@variable(model, x <= 0)\n@variable(model, x == 0)\n@variable(model, 0 <= x <= 1)\n\nand bounds can depend on the indices of the container expressions:\n\n@variable(model, -i <= x[i=1:3] <= i)\n\nSets\n\nYou can explicitly specify the set to which the variable belongs:\n\n@variable(model, x in MOI.Interval(0.0, 1.0))\n\nFor more information on this syntax, read Variables constrained on creation.\n\nPositional arguments\n\nThe recognized positional arguments in args are the following:\n\nBin: restricts the variable to the MOI.ZeroOne set, that is, {0, 1}. For example, @variable(model, x, Bin). Note: you cannot use @variable(model, Bin), use the binary keyword instead.\nInt: restricts the variable to the set of integers, that is, ..., -2, -1,  0, 1, 2, ... For example, @variable(model, x, Int). Note: you cannot use  @variable(model, Int), use the integer keyword instead.\nSymmetric: Only available when creating a square matrix of variables, i.e., when expr is of the form varname[1:n,1:n] or varname[i=1:n,j=1:n], it creates a symmetric matrix of variables.\nPSD: A restrictive extension to Symmetric which constraints a square matrix of variables to Symmetric and constrains to be positive semidefinite.\n\nKeyword arguments\n\nFour keyword arguments are useful in all cases:\n\nbase_name: Sets the name prefix used to generate variable names. It corresponds to the variable name for scalar variable, otherwise, the variable names are set to base_name[...] for each index ... of the axes axes.\nstart::Float64: specify the value passed to set_start_value for each variable\ncontainer: specify the container type. See Forcing the container type for more information.\nset_string_name::Bool = true: control whether to set the MOI.VariableName attribute. Passing set_string_name = false can improve performance.\n\nOther keyword arguments are needed to disambiguate sitations with anonymous variables:\n\nlower_bound::Float64: an alternative to x >= lb, sets the value of the variable lower bound.\nupper_bound::Float64: an alternative to x <= ub, sets the value of the variable upper bound.\nbinary::Bool: an alternative to passing Bin, sets whether the variable is binary or not.\ninteger::Bool: an alternative to passing Int, sets whether the variable is integer or not.\nset::MOI.AbstractSet: an alternative to using x in set\nvariable_type: used by JuMP extensions. See Extend @variable for more information.\n\nExample\n\nThe following are equivalent ways of creating a variable x of name x with lower bound 0:\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0)\nx\n\njulia> model = Model();\n\njulia> @variable(model, x, lower_bound = 0)\nx\n\njulia> model = Model();\n\njulia> x = @variable(model, base_name = \"x\", lower_bound = 0)\nx\n\nOther examples:\n\njulia> model = Model();\n\njulia> @variable(model, x[i=1:3] <= i, Int, start = sqrt(i), lower_bound = -i)\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @variable(model, y[i=1:3], container = DenseAxisArray, set = MOI.ZeroOne())\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, Base.OneTo(3)\nAnd data, a 3-element Vector{VariableRef}:\n y[1]\n y[2]\n y[3]\n\njulia> @variable(model, z[i=1:3], set_string_name = false)\n3-element Vector{VariableRef}:\n _[7]\n _[8]\n _[9]\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#@variables","page":"JuMP","title":"@variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.@variables","category":"page"},{"location":"api/JuMP/#JuMP.@variables","page":"JuMP","title":"JuMP.@variables","text":"@variables(model, args...)\n\nAdds multiple variables to model at once, in the same fashion as the @variable macro.\n\nThe model must be the first argument, and multiple variables can be added on multiple lines wrapped in a begin ... end block.\n\nThe macro returns a tuple containing the variables that were defined.\n\nExample\n\njulia> model = Model();\n\njulia> @variables(model, begin\n           x\n           y[i = 1:2] >= 0, (start = i)\n           z, Bin, (start = 0, base_name = \"Z\")\n       end)\n(x, VariableRef[y[1], y[2]], Z)\n\nnote: Note\nKeyword arguments must be contained within parentheses (refer to the example above).\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP/#add_bridge","page":"JuMP","title":"add_bridge","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_bridge","category":"page"},{"location":"api/JuMP/#JuMP.add_bridge","page":"JuMP","title":"JuMP.add_bridge","text":"add_bridge(\n    model::GenericModel{T},\n    BT::Type{<:MOI.Bridges.AbstractBridge};\n    coefficient_type::Type{S} = T,\n) where {T,S}\n\nAdd BT{T} to the list of bridges that can be used to transform unsupported constraints into an equivalent formulation using only constraints supported by the optimizer.\n\nSee also: remove_bridge.\n\nExample\n\njulia> model = Model();\n\njulia> add_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> add_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64}\n       )\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_constraint","page":"JuMP","title":"add_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_constraint","category":"page"},{"location":"api/JuMP/#JuMP.add_constraint","page":"JuMP","title":"JuMP.add_constraint","text":"add_constraint(model::GenericModel, con::AbstractConstraint, name::String=\"\")\n\nAdd a constraint con to Model model and sets its name.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_constraint","page":"JuMP","title":"add_nonlinear_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_constraint","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_constraint","page":"JuMP","title":"JuMP.add_nonlinear_constraint","text":"add_nonlinear_constraint(model::Model, expr::Expr)\n\nAdd a nonlinear constraint described by the Julia expression ex to model.\n\nThis function is most useful if the expression ex is generated programmatically, and you cannot use @NLconstraint.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> add_nonlinear_constraint(model, :($(x) + $(x)^2 <= 1))\n(x + x ^ 2.0) - 1.0 ≤ 0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_expression","page":"JuMP","title":"add_nonlinear_expression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_expression","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_expression","page":"JuMP","title":"JuMP.add_nonlinear_expression","text":"add_nonlinear_expression(model::Model, expr::Expr)\n\nAdd a nonlinear expression expr to model.\n\nThis function is most useful if the expression expr is generated programmatically, and you cannot use @NLexpression.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> add_nonlinear_expression(model, :($(x) + $(x)^2))\nsubexpression[1]: x + x ^ 2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_operator","page":"JuMP","title":"add_nonlinear_operator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_operator","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_operator","page":"JuMP","title":"JuMP.add_nonlinear_operator","text":"add_nonlinear_operator(\n    model::Model,\n    dim::Int,\n    f::Function,\n    [∇f::Function,]\n    [∇²f::Function];\n    [name::Symbol = Symbol(f),]\n)\n\nAdd a new nonlinear operator with dim input arguments to model and associate it with the name name.\n\nThe function f evaluates the operator and must return a scalar.\n\nThe optional function ∇f evaluates the first derivative, and the optional function ∇²f evaluates the second derivative.\n\n∇²f may be provided only if ∇f is also provided.\n\nUnivariate syntax\n\nIf dim == 1, then the method signatures of each function must be:\n\nf(::T)::T where {T<:Real}\n∇f(::T)::T where {T<:Real}\n∇²f(::T)::T where {T<:Real}\n\nMultivariate syntax\n\nIf dim > 1, then the method signatures of each function must be:\n\nf(x::T...)::T where {T<:Real}\n∇f(g::AbstractVector{T}, x::T...)::Nothing where {T<:Real}\n∇²f(H::AbstractMatrix{T}, x::T...)::Nothing where {T<:Real}\n\nWhere the gradient vector g and Hessian matrix H are filled in-place. For the Hessian, you must fill in the non-zero lower-triangular entries only. Setting an off-diagonal upper-triangular element may error.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> op_f = add_nonlinear_operator(model, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :f)\n\njulia> @objective(model, Min, op_f(x))\nf(x)\n\njulia> op_f(2.0)\n4.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_nonlinear_parameter","page":"JuMP","title":"add_nonlinear_parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_nonlinear_parameter","category":"page"},{"location":"api/JuMP/#JuMP.add_nonlinear_parameter","page":"JuMP","title":"JuMP.add_nonlinear_parameter","text":"add_nonlinear_parameter(model::Model, value::Real)\n\nAdd an anonymous parameter to the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_to_expression!","page":"JuMP","title":"add_to_expression!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_to_expression!","category":"page"},{"location":"api/JuMP/#JuMP.add_to_expression!","page":"JuMP","title":"JuMP.add_to_expression!","text":"add_to_expression!(expression, terms...)\n\nUpdates expression in place to expression + (*)(terms...). This is typically much more efficient than expression += (*)(terms...). For example, add_to_expression!(expression, a, b) produces the same result as expression += a*b, and add_to_expression!(expression, a) produces the same result as expression += a.\n\nOnly a few methods are defined, mostly for internal use, and only for the cases when (1) they can be implemented efficiently and (2) expression is capable of storing the result. For example, add_to_expression!(::AffExpr, ::GenericVariableRef, ::GenericVariableRef) is not defined because a GenericAffExpr cannot store the product of two variables.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_to_function_constant","page":"JuMP","title":"add_to_function_constant","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_to_function_constant","category":"page"},{"location":"api/JuMP/#JuMP.add_to_function_constant","page":"JuMP","title":"JuMP.add_to_function_constant","text":"add_to_function_constant(constraint::ConstraintRef, value)\n\nAdd value to the function constant term.\n\nNote that for scalar constraints, JuMP will aggregate all constant terms onto the right-hand side of the constraint so instead of modifying the function, the set will be translated by -value. For example, given a constraint 2x <= 3, add_to_function_constant(c, 4) will modify it to 2x <= -1.\n\nExample\n\nFor scalar constraints, the set is translated by -value:\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 0 <= 2x - 1 <= 2)\ncon : 2 x ∈ [1, 3]\n\njulia> add_to_function_constant(con, 4)\n\njulia> con\ncon : 2 x ∈ [-3, -1]\n\nFor vector constraints, the constant is added to the function:\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, con, [x + y, x, y] in SecondOrderCone())\ncon : [x + y, x, y] ∈ MathOptInterface.SecondOrderCone(3)\n\njulia> add_to_function_constant(con, [1, 2, 2])\n\njulia> con\ncon : [x + y + 1, x + 2, y + 2] ∈ MathOptInterface.SecondOrderCone(3)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#add_variable","page":"JuMP","title":"add_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.add_variable","category":"page"},{"location":"api/JuMP/#JuMP.add_variable","page":"JuMP","title":"JuMP.add_variable","text":"add_variable(m::GenericModel, v::AbstractVariable, name::String=\"\")\n\nAdd a variable v to Model m and sets its name.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_constraints","page":"JuMP","title":"all_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_constraints","category":"page"},{"location":"api/JuMP/#JuMP.all_constraints","page":"JuMP","title":"JuMP.all_constraints","text":"all_constraints(model::GenericModel, function_type, set_type)::Vector{<:ConstraintRef}\n\nReturn a list of all constraints currently in the model where the function has type function_type and the set has type set_type. The constraints are ordered by creation time.\n\nSee also list_of_constraint_types and num_constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> all_constraints(model, VariableRef, MOI.GreaterThan{Float64})\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n x ≥ 0\n\njulia> all_constraints(model, VariableRef, MOI.ZeroOne)\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}, ScalarShape}}:\n x binary\n\njulia> all_constraints(model, AffExpr, MOI.LessThan{Float64})\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n 2 x ≤ 1\n\n\n\n\n\nall_constraints(\n    model::GenericModel;\n    include_variable_in_set_constraints::Bool,\n)::Vector{ConstraintRef}\n\nReturn a list of all constraints in model.\n\nIf include_variable_in_set_constraints == true, then VariableRef constraints such as VariableRef-in-Integer are included. To return only the structural constraints (e.g., the rows in the constraint matrix of a linear program), pass include_variable_in_set_constraints = false.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Int);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> @NLconstraint(model, x^2 <= 1);\n\njulia> all_constraints(model; include_variable_in_set_constraints = true)\n4-element Vector{ConstraintRef}:\n 2 x ≤ 1\n x ≥ 0\n x integer\n x ^ 2.0 - 1.0 ≤ 0\n\njulia> all_constraints(model; include_variable_in_set_constraints = false)\n2-element Vector{ConstraintRef}:\n 2 x ≤ 1\n x ^ 2.0 - 1.0 ≤ 0\n\nPerformance considerations\n\nNote that this function is type-unstable because it returns an abstractly typed vector. If performance is a problem, consider using list_of_constraint_types and a function barrier. See the Performance tips for extensions section of the documentation for more details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_nonlinear_constraints","page":"JuMP","title":"all_nonlinear_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_nonlinear_constraints","category":"page"},{"location":"api/JuMP/#JuMP.all_nonlinear_constraints","page":"JuMP","title":"JuMP.all_nonlinear_constraints","text":"all_nonlinear_constraints(model::GenericModel)\n\nReturn a vector of all nonlinear constraint references in the model in the order they were added to the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#all_variables","page":"JuMP","title":"all_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.all_variables","category":"page"},{"location":"api/JuMP/#JuMP.all_variables","page":"JuMP","title":"JuMP.all_variables","text":"all_variables(model::GenericModel{T})::Vector{GenericVariableRef{T}} where {T}\n\nReturns a list of all variables currently in the model. The variables are ordered by creation time.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, y);\n\njulia> all_variables(model)\n2-element Vector{VariableRef}:\n x\n y\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#anonymous_name","page":"JuMP","title":"anonymous_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.anonymous_name","category":"page"},{"location":"api/JuMP/#JuMP.anonymous_name","page":"JuMP","title":"JuMP.anonymous_name","text":"anonymous_name(::MIME, x::AbstractVariableRef)\n\nThe name to use for an anonymous variable x when printing.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#backend","page":"JuMP","title":"backend","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.backend","category":"page"},{"location":"api/JuMP/#JuMP.backend","page":"JuMP","title":"JuMP.backend","text":"backend(model::GenericModel)\n\nReturn the lower-level MathOptInterface model that sits underneath JuMP. This model depends on which operating mode JuMP is in (see mode).\n\nIf JuMP is in DIRECT mode (i.e., the model was created using direct_model), the backend will be the optimizer passed to direct_model.\nIf JuMP is in MANUAL or AUTOMATIC mode, the backend is a MOI.Utilities.CachingOptimizer.\n\nThis function should only be used by advanced users looking to access low-level MathOptInterface or solver-specific functionality.\n\nNotes\n\nIf JuMP is not in DIRECT mode, the type returned by backend may change between any JuMP releases. Therefore, only use the public API exposed by MathOptInterface, and do not access internal fields. If you require access to the innermost optimizer, see unsafe_backend. Alternatively, use direct_model to create a JuMP model in DIRECT mode.\n\nSee also: unsafe_backend.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#barrier_iterations","page":"JuMP","title":"barrier_iterations","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.barrier_iterations","category":"page"},{"location":"api/JuMP/#JuMP.barrier_iterations","page":"JuMP","title":"JuMP.barrier_iterations","text":"barrier_iterations(model::GenericModel)\n\nGets the cumulative number of barrier iterations during the most recent optimization.\n\nSolvers must implement MOI.BarrierIterations() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#bridge_constraints","page":"JuMP","title":"bridge_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.bridge_constraints","category":"page"},{"location":"api/JuMP/#JuMP.bridge_constraints","page":"JuMP","title":"JuMP.bridge_constraints","text":"bridge_constraints(model::GenericModel)\n\nWhen in direct mode, return false.\n\nWhen in manual or automatic mode, return a Bool indicating whether the optimizer is set and unsupported constraints are automatically bridged to equivalent supported constraints when an appropriate transformation is available.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#build_constraint","page":"JuMP","title":"build_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.build_constraint","category":"page"},{"location":"api/JuMP/#JuMP.build_constraint","page":"JuMP","title":"JuMP.build_constraint","text":"build_constraint(\n    _error::Function,\n    Q::LinearAlgebra.Symmetric{V, M},\n    ::PSDCone,\n) where {V<:AbstractJuMPScalar,M<:AbstractMatrix{V}}\n\nReturn a VectorConstraint of shape SymmetricMatrixShape constraining the matrix Q to be positive semidefinite.\n\nThis function is used by the @constraint macros as follows:\n\njulia> import LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, LinearAlgebra.Symmetric(Q) in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ PSDCone()\n\nThe form above is usually used when the entries of Q are affine or quadratic expressions, but it can also be used when the entries are variables to get the reference of the semidefinite constraint, e.g.,\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\njulia> @constraint(model, Q in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ PSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    Q::AbstractMatrix{<:AbstractJuMPScalar},\n    ::PSDCone,\n)\n\nReturn a VectorConstraint of shape SquareMatrixShape constraining the matrix Q to be symmetric and positive semidefinite.\n\nThis function is used by the @constraint macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, Q in PSDCone())\n[Q[1,1]  Q[1,2];\n Q[2,1]  Q[2,2]] ∈ PSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    Q::LinearAlgebra.Hermitian{V,M},\n    ::HermitianPSDCone,\n) where {V<:AbstractJuMPScalar,M<:AbstractMatrix{V}}\n\nReturn a VectorConstraint of shape HermitianMatrixShape constraining the matrix Q to be Hermitian positive semidefinite.\n\nThis function is used by the @constraint macros as follows:\n\njulia> import LinearAlgebra\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2]);\n\njulia> @constraint(model, LinearAlgebra.Hermitian(Q) in HermitianPSDCone())\n[Q[1,1]  Q[1,2];\n Q[1,2]  Q[2,2]] ∈ HermitianPSDCone()\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    f::AbstractVector{<:AbstractJuMPScalar},\n    ::Nonnegatives,\n    extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},\n)\n\nA helper method that re-writes\n\n@constraint(model, X >= Y, extra)\n\ninto\n\n@constraint(model, X - Y in extra)\n\n\n\n\n\nbuild_constraint(\n    _error::Function,\n    f::AbstractVector{<:AbstractJuMPScalar},\n    ::Nonpositives,\n    extra::Union{MOI.AbstractVectorSet,AbstractVectorSet},\n)\n\nA helper method that re-writes\n\n@constraint(model, Y <= X, extra)\n\ninto\n\n@constraint(model, X - Y in extra)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#build_variable","page":"JuMP","title":"build_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.build_variable","category":"page"},{"location":"api/JuMP/#JuMP.build_variable","page":"JuMP","title":"JuMP.build_variable","text":"build_variable(\n    _error::Function,\n    info::VariableInfo,\n    args...;\n    kwargs...,\n)\n\nReturn a new AbstractVariable object.\n\nThis method should only be implemented by developers creating JuMP extensions. It should never be called by users of JuMP.\n\nArguments\n\n_error: a function to call instead of error. _error annotates the error message with additional information for the user.\ninfo: an instance of VariableInfo. This has a variety of fields relating to the variable such as info.lower_bound and info.binary.\nargs: optional additional positional arguments for extending the @variable macro.\nkwargs: optional keyword arguments for extending the @variable macro.\n\nSee also: @variable\n\nwarning: Warning\nExtensions should define a method with ONE positional argument to dispatch the call to a different method. Creating an extension that relies on multiple positional arguments leads to MethodErrors if the user passes the arguments in the wrong order.\n\nExample\n\n@variable(model, x, Foo)\n\nwill call\n\nbuild_variable(_error::Function, info::VariableInfo, ::Type{Foo})\n\nPassing special-case positional arguments such as Bin, Int, and PSD is okay, along with keyword arguments:\n\n@variable(model, x, Int, Foo(), mykwarg = true)\n# or\n@variable(model, x, Foo(), Int, mykwarg = true)\n\nwill call\n\nbuild_variable(_error::Function, info::VariableInfo, ::Foo; mykwarg)\n\nand info.integer will be true.\n\nNote that the order of the positional arguments does not matter.\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::SymmetricMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape SymmetricMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::SkewSymmetricMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape SkewSymmetricMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        Q[1,2]\n -Q[1,2]  0\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::HermitianMatrixSpace)\n\nReturn a VariablesConstrainedOnCreation of shape HermitianMatrixShape creating variables in MOI.Reals, i.e. \"free\" variables unless they are constrained after their creation.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im\n real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])\n\n\n\n\n\nbuild_variable(_error::Function, variables, ::PSDCone)\n\nReturn a VariablesConstrainedOnCreation of shape SymmetricMatrixShape constraining the variables to be positive semidefinite.\n\nThis function is used by the @variable macro as follows:\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2], PSD)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#callback_node_status","page":"JuMP","title":"callback_node_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.callback_node_status","category":"page"},{"location":"api/JuMP/#JuMP.callback_node_status","page":"JuMP","title":"JuMP.callback_node_status","text":"callback_node_status(cb_data, model::GenericModel)\n\nReturn an MOI.CallbackNodeStatusCode enum, indicating if the current primal solution available from callback_value is integer feasible.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#callback_value","page":"JuMP","title":"callback_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.callback_value","category":"page"},{"location":"api/JuMP/#JuMP.callback_value","page":"JuMP","title":"JuMP.callback_value","text":"callback_value(cb_data, x::GenericVariableRef)\n\nReturn the primal solution of a variable inside a callback.\n\ncb_data is the argument to the callback function, and the type is dependent on the solver.\n\n\n\n\n\ncallback_value(cb_data, expr::Union{GenericAffExpr, GenericQuadExpr})\n\nReturn the primal solution of an affine or quadratic expression inside a callback by getting the value for each variable appearing in the expression.\n\ncb_data is the argument to the callback function, and the type is dependent on the solver.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#check_belongs_to_model","page":"JuMP","title":"check_belongs_to_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.check_belongs_to_model","category":"page"},{"location":"api/JuMP/#JuMP.check_belongs_to_model","page":"JuMP","title":"JuMP.check_belongs_to_model","text":"check_belongs_to_model(func::AbstractJuMPScalar, model::AbstractModel)\n\nThrow VariableNotOwned if the owner_model of one of the variables of the function func is not model.\n\ncheck_belongs_to_model(constraint::AbstractConstraint, model::AbstractModel)\n\nThrow VariableNotOwned if the owner_model of one of the variables of the constraint constraint is not model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#coefficient","page":"JuMP","title":"coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.coefficient","category":"page"},{"location":"api/JuMP/#JuMP.coefficient","page":"JuMP","title":"JuMP.coefficient","text":"coefficient(v1::GenericVariableRef{T}, v2::GenericVariableRef{T}) where {T}\n\nReturn one(T) if v1 == v2, and zero(T) otherwise.\n\nThis is a fallback for other coefficient methods to simplify code in which the expression may be a single variable.\n\n\n\n\n\ncoefficient(a::GenericAffExpr{C,V}, v::V) where {C,V}\n\nReturn the coefficient associated with variable v in the affine expression a.\n\n\n\n\n\ncoefficient(a::GenericAffExpr{C,V}, v1::V, v2::V) where {C,V}\n\nReturn the coefficient associated with the term v1 * v2 in the quadratic expression a.\n\nNote that coefficient(a, v1, v2) is the same as coefficient(a, v2, v1).\n\n\n\n\n\ncoefficient(a::GenericQuadExpr{C,V}, v::V) where {C,V}\n\nReturn the coefficient associated with variable v in the affine component of a.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#compute_conflict!","page":"JuMP","title":"compute_conflict!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.compute_conflict!","category":"page"},{"location":"api/JuMP/#JuMP.compute_conflict!","page":"JuMP","title":"JuMP.compute_conflict!","text":"compute_conflict!(model::GenericModel)\n\nCompute a conflict if the model is infeasible. If an optimizer has not been set yet (see set_optimizer), a NoOptimizer error is thrown.\n\nThe status of the conflict can be checked with the MOI.ConflictStatus model attribute. Then, the status for each constraint can be queried with the MOI.ConstraintConflictStatus attribute.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constant","page":"JuMP","title":"constant","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constant","category":"page"},{"location":"api/JuMP/#JuMP.constant","page":"JuMP","title":"JuMP.constant","text":"constant(aff::GenericAffExpr{C, V})::C\n\nReturn the constant of the affine expression.\n\n\n\n\n\nconstant(aff::GenericQuadExpr{C, V})::C\n\nReturn the constant of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_by_name","page":"JuMP","title":"constraint_by_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_by_name","category":"page"},{"location":"api/JuMP/#JuMP.constraint_by_name","page":"JuMP","title":"JuMP.constraint_by_name","text":"constraint_by_name(model::AbstractModel,\n                   name::String)::Union{ConstraintRef, Nothing}\n\nReturn the reference of the constraint with name attribute name or Nothing if no constraint has this name attribute. Throws an error if several constraints have name as their name attribute.\n\nconstraint_by_name(model::AbstractModel,\n                   name::String,\n                   F::Type{<:Union{AbstractJuMPScalar,\n                                   Vector{<:AbstractJuMPScalar},\n                                   MOI.AbstactFunction}},\n                   S::Type{<:MOI.AbstractSet})::Union{ConstraintRef, Nothing}\n\nSimilar to the method above, except that it throws an error if the constraint is not an F-in-S contraint where F is either the JuMP or MOI type of the function, and S is the MOI type of the set. This method is recommended if you know the type of the function and set since its returned type can be inferred while for the method above (i.e. without F and S), the exact return type of the constraint index cannot be inferred.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, x^2 == 1)\ncon : x² = 1\n\njulia> constraint_by_name(model, \"kon\")\n\njulia> constraint_by_name(model, \"con\")\ncon : x² = 1\n\njulia> constraint_by_name(model, \"con\", AffExpr, MOI.EqualTo{Float64})\n\njulia> constraint_by_name(model, \"con\", QuadExpr, MOI.EqualTo{Float64})\ncon : x² = 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_object","page":"JuMP","title":"constraint_object","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_object","category":"page"},{"location":"api/JuMP/#JuMP.constraint_object","page":"JuMP","title":"JuMP.constraint_object","text":"constraint_object(con_ref::ConstraintRef)\n\nReturn the underlying constraint data for the constraint referenced by ref.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_ref_with_index","page":"JuMP","title":"constraint_ref_with_index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_ref_with_index","category":"page"},{"location":"api/JuMP/#JuMP.constraint_ref_with_index","page":"JuMP","title":"JuMP.constraint_ref_with_index","text":"constraint_ref_with_index(model::AbstractModel, index::MOI.ConstraintIndex)\n\nReturn a ConstraintRef of model corresponding to index.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraint_string","page":"JuMP","title":"constraint_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraint_string","category":"page"},{"location":"api/JuMP/#JuMP.constraint_string","page":"JuMP","title":"JuMP.constraint_string","text":"constraint_string(\n    mode::MIME,\n    ref::ConstraintRef;\n    in_math_mode::Bool = false)\n\nReturn a string representation of the constraint ref, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#constraints_string","page":"JuMP","title":"constraints_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.constraints_string","category":"page"},{"location":"api/JuMP/#JuMP.constraints_string","page":"JuMP","title":"JuMP.constraints_string","text":"constraints_string(mode, model::AbstractModel)::Vector{String}\n\nReturn a list of Strings describing each constraint of the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_conflict","page":"JuMP","title":"copy_conflict","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_conflict","category":"page"},{"location":"api/JuMP/#JuMP.copy_conflict","page":"JuMP","title":"JuMP.copy_conflict","text":"copy_conflict(model::GenericModel)\n\nReturn a copy of the current conflict for the model model and a GenericReferenceMap that can be used to obtain the variable and constraint reference of the new model corresponding to a given model's reference.\n\nThis is a convenience function that provides a filtering function for copy_model.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and two constraints c1 and c2. This model has no solution, as the two constraints are mutually exclusive. The solver is asked to compute a conflict with compute_conflict!. The parts of model participating in the conflict are then copied into a model iis_model.\n\njulia> using JuMP\n\njulia> import Gurobi\n\njulia> model = Model(Gurobi.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, c1, x >= 2)\nc1 : x ≥ 2\n\njulia> @constraint(model, c2, x <= 1)\nc2 : x ≤ 1\n\njulia> optimize!(model)\n\njulia> compute_conflict!(model)\n\njulia> if get_attribute(model, MOI.ConflictStatus()) == MOI.CONFLICT_FOUND\n           iis_model, reference_map = copy_conflict(model)\n           print(iis_model)\n       end\nFeasibility\nSubject to\n c1 : x ≥ 2\n c2 : x ≤ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_extension_data","page":"JuMP","title":"copy_extension_data","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_extension_data","category":"page"},{"location":"api/JuMP/#JuMP.copy_extension_data","page":"JuMP","title":"JuMP.copy_extension_data","text":"copy_extension_data(data, new_model::AbstractModel, model::AbstractModel)\n\nReturn a copy of the extension data data of the model model to the extension data of the new model new_model.\n\nA method should be added for any JuMP extension storing data in the ext field.\n\nwarning: Warning\nDo not engage in type piracy by implementing this method for types of data that you did not define! JuMP extensions should store types that they define in model.ext, rather than regular Julia types.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#copy_model","page":"JuMP","title":"copy_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.copy_model","category":"page"},{"location":"api/JuMP/#JuMP.copy_model","page":"JuMP","title":"JuMP.copy_model","text":"copy_model(model::GenericModel; filter_constraints::Union{Nothing, Function}=nothing)\n\nReturn a copy of the model model and a GenericReferenceMap that can be used to obtain the variable and constraint reference of the new model corresponding to a given model's reference. A Base.copy(::AbstractModel) method has also been implemented, it is similar to copy_model but does not return the reference map.\n\nIf the filter_constraints argument is given, only the constraints for which this function returns true will be copied. This function is given a constraint reference as argument.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and a constraint cref. It is then copied into a model new_model with the new references assigned to x_new and cref_new.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, cref, x == 2)\ncref : x = 2\n\njulia> new_model, reference_map = copy_model(model);\n\njulia> x_new = reference_map[x]\nx\n\njulia> cref_new = reference_map[cref]\ncref : x = 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete","page":"JuMP","title":"delete","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete","category":"page"},{"location":"api/JuMP/#JuMP.delete","page":"JuMP","title":"JuMP.delete","text":"delete(model::GenericModel, con_ref::ConstraintRef)\n\nDelete the constraint associated with constraint_ref from the model model.\n\nNote that delete does not unregister the name from the model, so adding a new constraint of the same name will throw an error. Use unregister to unregister the name after deletion.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, c, 2x <= 1)\nc : 2 x ≤ 1\n\njulia> delete(model, c)\n\njulia> unregister(model, :c)\n\njulia> print(model)\nFeasibility\nSubject to\n\njulia> model[:c]\nERROR: KeyError: key :c not found\nStacktrace:\n[...]\n\n\n\n\n\ndelete(model::GenericModel, con_refs::Vector{<:ConstraintRef})\n\nDelete the constraints associated with con_refs from the model model. Solvers may implement specialized methods for deleting multiple constraints of the same concrete type, i.e., when isconcretetype(eltype(con_refs)). These may be more efficient than repeatedly calling the single constraint delete method.\n\nSee also: unregister\n\n\n\n\n\ndelete(model::GenericModel, variable_ref::GenericVariableRef)\n\nDelete the variable associated with variable_ref from the model model.\n\nNote that delete does not unregister the name from the model, so adding a new variable of the same name will throw an error. Use unregister to unregister the name after deletion.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> delete(model, x)\n\njulia> unregister(model, :x)\n\njulia> print(model)\nFeasibility\nSubject to\n\njulia> model[:x]\nERROR: KeyError: key :x not found\nStacktrace:\n[...]\n\n\n\n\n\ndelete(model::GenericModel, variable_refs::Vector{<:GenericVariableRef})\n\nDelete the variables associated with variable_refs from the model model. Solvers may implement methods for deleting multiple variables that are more efficient than repeatedly calling the single variable delete method.\n\nSee also: unregister\n\n\n\n\n\ndelete(model::Model, c::NonlinearConstraintRef)\n\nDelete the nonlinear constraint c from model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete_lower_bound","page":"JuMP","title":"delete_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.delete_lower_bound","page":"JuMP","title":"JuMP.delete_lower_bound","text":"delete_lower_bound(v::GenericVariableRef)\n\nDelete the lower bound constraint of a variable.\n\nSee also LowerBoundRef, has_lower_bound, lower_bound, set_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> has_lower_bound(x)\ntrue\n\njulia> delete_lower_bound(x)\n\njulia> has_lower_bound(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#delete_upper_bound","page":"JuMP","title":"delete_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.delete_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.delete_upper_bound","page":"JuMP","title":"JuMP.delete_upper_bound","text":"delete_upper_bound(v::GenericVariableRef)\n\nDelete the upper bound constraint of a variable.\n\nErrors if one does not exist.\n\nSee also UpperBoundRef, has_upper_bound, upper_bound, set_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> has_upper_bound(x)\ntrue\n\njulia> delete_upper_bound(x)\n\njulia> has_upper_bound(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#direct_generic_model","page":"JuMP","title":"direct_generic_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.direct_generic_model","category":"page"},{"location":"api/JuMP/#JuMP.direct_generic_model","page":"JuMP","title":"JuMP.direct_generic_model","text":"direct_generic_model(\n    value_type::Type{T},\n    backend::MOI.ModelLike;\n) where {T<:Real}\n\nReturn a new JuMP model using backend to store the model and solve it.\n\nAs opposed to the Model constructor, no cache of the model is stored outside of backend and no bridges are automatically applied to backend.\n\nNotes\n\nThe absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this direct mode:\n\nWhen backend does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the Model constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when optimize! is called.\nNo constraint bridging is supported by default.\nThe optimizer used cannot be changed the model is constructed.\nThe model created cannot be copied.\n\n\n\n\n\ndirect_generic_model(::Type{T}, factory::MOI.OptimizerWithAttributes)\n\nCreate a direct_generic_model using factory, a MOI.OptimizerWithAttributes object created by optimizer_with_attributes.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer,\n           \"presolve\" => \"off\",\n           MOI.Silent() => true,\n       );\n\njulia> model = direct_generic_model(Float64, optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = direct_generic_model(Float64, HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#direct_model","page":"JuMP","title":"direct_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.direct_model","category":"page"},{"location":"api/JuMP/#JuMP.direct_model","page":"JuMP","title":"JuMP.direct_model","text":"direct_model(backend::MOI.ModelLike)\n\nReturn a new JuMP model using backend to store the model and solve it.\n\nAs opposed to the Model constructor, no cache of the model is stored outside of backend and no bridges are automatically applied to backend.\n\nNotes\n\nThe absence of a cache reduces the memory footprint but, it is important to bear in mind the following implications of creating models using this direct mode:\n\nWhen backend does not support an operation, such as modifying constraints or adding variables/constraints after solving, an error is thrown. For models created using the Model constructor, such situations can be dealt with by storing the modifications in a cache and loading them into the optimizer when optimize! is called.\nNo constraint bridging is supported by default.\nThe optimizer used cannot be changed the model is constructed.\nThe model created cannot be copied.\n\n\n\n\n\ndirect_model(factory::MOI.OptimizerWithAttributes)\n\nCreate a direct_model using factory, a MOI.OptimizerWithAttributes object created by optimizer_with_attributes.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer,\n           \"presolve\" => \"off\",\n           MOI.Silent() => true,\n       );\n\njulia> model = direct_model(optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = direct_model(HiGHS.Optimizer())\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: DIRECT\nSolver name: HiGHS\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#drop_zeros!","page":"JuMP","title":"drop_zeros!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.drop_zeros!","category":"page"},{"location":"api/JuMP/#JuMP.drop_zeros!","page":"JuMP","title":"JuMP.drop_zeros!","text":"drop_zeros!(expr::GenericAffExpr)\n\nRemove terms in the affine expression with 0 coefficients.\n\n\n\n\n\ndrop_zeros!(expr::GenericQuadExpr)\n\nRemove terms in the quadratic expression with 0 coefficients.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual","page":"JuMP","title":"dual","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual","category":"page"},{"location":"api/JuMP/#JuMP.dual","page":"JuMP","title":"JuMP.dual","text":"dual(con_ref::ConstraintRef; result::Int = 1)\n\nReturn the dual value of constraint con_ref associated with result index result of the most-recent solution returned by the solver.\n\nUse has_dual to check if a result exists before asking for values.\n\nSee also: result_count, shadow_price.\n\n\n\n\n\ndual(c::NonlinearConstraintRef)\n\nReturn the dual of the nonlinear constraint c.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_objective_value","page":"JuMP","title":"dual_objective_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_objective_value","category":"page"},{"location":"api/JuMP/#JuMP.dual_objective_value","page":"JuMP","title":"JuMP.dual_objective_value","text":"dual_objective_value(model::GenericModel; result::Int = 1)\n\nReturn the value of the objective of the dual problem associated with result index result of the most-recent solution returned by the solver.\n\nThrows MOI.UnsupportedAttribute{MOI.DualObjectiveValue} if the solver does not support this attribute.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_shape","page":"JuMP","title":"dual_shape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_shape","category":"page"},{"location":"api/JuMP/#JuMP.dual_shape","page":"JuMP","title":"JuMP.dual_shape","text":"dual_shape(shape::AbstractShape)::AbstractShape\n\nReturns the shape of the dual space of the space of objects of shape shape. By default, the dual_shape of a shape is itself. See the examples section below for an example for which this is not the case.\n\nExample\n\nConsider polynomial constraints for which the dual is moment constraints and moment constraints for which the dual is polynomial constraints. Shapes for polynomials can be defined as follows:\n\nstruct Polynomial\n    coefficients::Vector{Float64}\n    monomials::Vector{Monomial}\nend\nstruct PolynomialShape <: AbstractShape\n    monomials::Vector{Monomial}\nend\nJuMP.reshape_vector(x::Vector, shape::PolynomialShape) = Polynomial(x, shape.monomials)\n\nand a shape for moments can be defined as follows:\n\nstruct Moments\n    coefficients::Vector{Float64}\n    monomials::Vector{Monomial}\nend\nstruct MomentsShape <: AbstractShape\n    monomials::Vector{Monomial}\nend\nJuMP.reshape_vector(x::Vector, shape::MomentsShape) = Moments(x, shape.monomials)\n\nThen dual_shape allows the definition of the shape of the dual of polynomial and moment constraints:\n\ndual_shape(shape::PolynomialShape) = MomentsShape(shape.monomials)\ndual_shape(shape::MomentsShape) = PolynomialShape(shape.monomials)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_start_value","page":"JuMP","title":"dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.dual_start_value","page":"JuMP","title":"JuMP.dual_start_value","text":"dual_start_value(con_ref::ConstraintRef)\n\nReturn the dual start value (MOI attribute ConstraintDualStart) of the constraint con_ref.\n\nNote: If no dual start value has been set, dual_start_value will return nothing.\n\nSee also set_dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#dual_status","page":"JuMP","title":"dual_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.dual_status","category":"page"},{"location":"api/JuMP/#JuMP.dual_status","page":"JuMP","title":"JuMP.dual_status","text":"dual_status(model::GenericModel; result::Int = 1)\n\nReturn a MOI.ResultStatusCode describing the status of the most recent dual solution of the solver (i.e., the MOI.DualStatus attribute) associated with the result index result.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#error_if_direct_mode","page":"JuMP","title":"error_if_direct_mode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.error_if_direct_mode","category":"page"},{"location":"api/JuMP/#JuMP.error_if_direct_mode","page":"JuMP","title":"JuMP.error_if_direct_mode","text":"error_if_direct_mode(model::GenericModel, func::Symbol)\n\nErrors if model is in direct mode during a call from the function named func.\n\nUsed internally within JuMP, or by JuMP extensions who do not want to support models in direct mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix","page":"JuMP","title":"fix","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix","category":"page"},{"location":"api/JuMP/#JuMP.fix","page":"JuMP","title":"JuMP.fix","text":"fix(v::GenericVariableRef, value::Number; force::Bool = false)\n\nFix a variable to a value. Update the fixing constraint if one exists, otherwise create a new one.\n\nIf the variable already has variable bounds and force=false, calling fix will throw an error. If force=true, existing variable bounds will be deleted, and the fixing constraint will be added. Note a variable will have no bounds after a call to unfix.\n\nSee also FixRef, is_fixed, fix_value, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0)\n\njulia> is_fixed(x)\ntrue\n\njulia> model = Model();\n\njulia> @variable(model, 0 <= x <= 1);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0; force = true)\n\njulia> is_fixed(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix_discrete_variables","page":"JuMP","title":"fix_discrete_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix_discrete_variables","category":"page"},{"location":"api/JuMP/#JuMP.fix_discrete_variables","page":"JuMP","title":"JuMP.fix_discrete_variables","text":"fix_discrete_variables([var_value::Function = value,] model::GenericModel)\n\nModifies model to convert all binary and integer variables to continuous variables with fixed bounds of var_value(x).\n\nReturn\n\nReturns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.\n\nNotes\n\nAn error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).\nAll other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin, start = 1);\n\njulia> @variable(model, 1 <= y <= 10, Int, start = 2);\n\njulia> @objective(model, Min, x + y);\n\njulia> undo_relax = fix_discrete_variables(start_value, model);\n\njulia> print(model)\nMin x + y\nSubject to\n x = 1\n y = 2\n\njulia> undo_relax()\n\njulia> print(model)\nMin x + y\nSubject to\n y ≥ 1\n y ≤ 10\n y integer\n x binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#fix_value","page":"JuMP","title":"fix_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.fix_value","category":"page"},{"location":"api/JuMP/#JuMP.fix_value","page":"JuMP","title":"JuMP.fix_value","text":"fix_value(v::GenericVariableRef)\n\nReturn the value to which a variable is fixed.\n\nError if one does not exist.\n\nSee also FixRef, is_fixed, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> fix_value(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#flatten!","page":"JuMP","title":"flatten!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.flatten!","category":"page"},{"location":"api/JuMP/#JuMP.flatten!","page":"JuMP","title":"JuMP.flatten!","text":"flatten!(expr::GenericNonlinearExpr)\n\nFlatten a nonlinear expression in-place by lifting nested + and * nodes into a single n-ary operation.\n\nMotivation\n\nNonlinear expressions created using operator overloading can be deeply nested and unbalanced. For example, prod(x for i in 1:4) creates *(x, *(x, *(x, x))) instead of the more preferable *(x, x, x, x).\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> y = prod(x for i in 1:4)\n((x²) * x) * x\n\njulia> flatten!(y)\n(x²) * x * x\n\njulia> flatten!(sin(prod(x for i in 1:4)))\nsin((x²) * x * x)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#function_string","page":"JuMP","title":"function_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.function_string","category":"page"},{"location":"api/JuMP/#JuMP.function_string","page":"JuMP","title":"JuMP.function_string","text":"function_string(\n    mode::MIME,\n    func::Union{JuMP.AbstractJuMPScalar,Vector{<:JuMP.AbstractJuMPScalar}},\n)\n\nReturn a String representing the function func using print mode mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#get_attribute","page":"JuMP","title":"get_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.get_attribute","category":"page"},{"location":"api/JuMP/#JuMP.get_attribute","page":"JuMP","title":"JuMP.get_attribute","text":"get_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute)\nget_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute)\nget_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute)\n\nGet the value of a solver-specifc attribute attr.\n\nThis is equivalent to calling MOI.get with the associated MOI model and, for variables and constraints, with the associated MOI.VariableIndex or MOI.ConstraintIndex.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2 * x <= 1)\nc : 2 x ≤ 1\n\njulia> get_attribute(model, MOI.Name())\n\"\"\n\njulia> get_attribute(x, MOI.VariableName())\n\"x\"\n\njulia> get_attribute(c, MOI.ConstraintName())\n\"c\"\n\n\n\n\n\nget_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n)\n\nGet the value of a solver-specifc attribute attr.\n\nThis is equivalent to calling MOI.get with the associated MOI model.\n\nIf attr is an AbstractString, it is converted to MOI.RawOptimizerAttribute.\n\nExample\n\njulia> import HiGHS\n\njulia> opt = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => true);\n\njulia> model = Model(opt);\n\njulia> get_attribute(model, \"output_flag\")\ntrue\n\njulia> get_attribute(model, MOI.RawOptimizerAttribute(\"output_flag\"))\ntrue\n\njulia> get_attribute(opt, \"output_flag\")\ntrue\n\njulia> get_attribute(opt, MOI.RawOptimizerAttribute(\"output_flag\"))\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#get_optimizer_attribute","page":"JuMP","title":"get_optimizer_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.get_optimizer_attribute","category":"page"},{"location":"api/JuMP/#JuMP.get_optimizer_attribute","page":"JuMP","title":"JuMP.get_optimizer_attribute","text":"get_optimizer_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n)\n\nReturn the value associated with the solver-specific attribute attr.\n\nIf attr is an AbstractString, this is equivalent to get_optimizer_attribute(model, MOI.RawOptimizerAttribute(name)).\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using get_attribute instead.\n\nSee also: set_optimizer_attribute, set_optimizer_attributes.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> get_optimizer_attribute(model, MOI.Silent())\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_duals","page":"JuMP","title":"has_duals","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_duals","category":"page"},{"location":"api/JuMP/#JuMP.has_duals","page":"JuMP","title":"JuMP.has_duals","text":"has_duals(model::GenericModel; result::Int = 1)\n\nReturn true if the solver has a dual solution in result index result available to query, otherwise return false.\n\nSee also dual, shadow_price, and result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_lower_bound","page":"JuMP","title":"has_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.has_lower_bound","page":"JuMP","title":"JuMP.has_lower_bound","text":"has_lower_bound(v::GenericVariableRef)\n\nReturn true if v has a lower bound. If true, the lower bound can be queried with lower_bound.\n\nSee also LowerBoundRef, lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> has_lower_bound(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_start_value","page":"JuMP","title":"has_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_start_value","category":"page"},{"location":"api/JuMP/#JuMP.has_start_value","page":"JuMP","title":"JuMP.has_start_value","text":"has_start_value(variable::AbstractVariableRef)\n\nReturn true if the variable has a start value set otherwise return false.\n\nSee also set_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_upper_bound","page":"JuMP","title":"has_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.has_upper_bound","page":"JuMP","title":"JuMP.has_upper_bound","text":"has_upper_bound(v::GenericVariableRef)\n\nReturn true if v has a upper bound. If true, the upper bound can be queried with upper_bound.\n\nSee also UpperBoundRef, upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> has_upper_bound(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#has_values","page":"JuMP","title":"has_values","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.has_values","category":"page"},{"location":"api/JuMP/#JuMP.has_values","page":"JuMP","title":"JuMP.has_values","text":"has_values(model::GenericModel; result::Int = 1)\n\nReturn true if the solver has a primal solution in result index result available to query, otherwise return false.\n\nSee also value and result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#in_set_string","page":"JuMP","title":"in_set_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.in_set_string","category":"page"},{"location":"api/JuMP/#JuMP.in_set_string","page":"JuMP","title":"JuMP.in_set_string","text":"in_set_string(mode::MIME, set)\n\nReturn a String representing the membership to the set set using print mode mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#index","page":"JuMP","title":"index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.index","category":"page"},{"location":"api/JuMP/#JuMP.index","page":"JuMP","title":"JuMP.index","text":"index(cr::ConstraintRef)::MOI.ConstraintIndex\n\nReturn the index of the constraint that corresponds to cr in the MOI backend.\n\n\n\n\n\nindex(v::GenericVariableRef)::MOI.VariableIndex\n\nReturn the index of the variable that corresponds to v in the MOI backend.\n\n\n\n\n\nindex(p::NonlinearParameter)::MOI.Nonlinear.ParameterIndex\n\nReturn the index of the nonlinear parameter associated with p.\n\n\n\n\n\nindex(ex::NonlinearExpression)::MOI.Nonlinear.ExpressionIndex\n\nReturn the index of the nonlinear expression associated with ex.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_binary","page":"JuMP","title":"is_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_binary","category":"page"},{"location":"api/JuMP/#JuMP.is_binary","page":"JuMP","title":"JuMP.is_binary","text":"is_binary(v::GenericVariableRef)\n\nReturn true if v is constrained to be binary.\n\nSee also BinaryRef, set_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> is_binary(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_fixed","page":"JuMP","title":"is_fixed","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_fixed","category":"page"},{"location":"api/JuMP/#JuMP.is_fixed","page":"JuMP","title":"JuMP.is_fixed","text":"is_fixed(v::GenericVariableRef)\n\nReturn true if v is a fixed variable. If true, the fixed value can be queried with fix_value.\n\nSee also FixRef, fix_value, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_fixed(x)\nfalse\n\njulia> fix(x, 1.0)\n\njulia> is_fixed(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_integer","page":"JuMP","title":"is_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_integer","category":"page"},{"location":"api/JuMP/#JuMP.is_integer","page":"JuMP","title":"JuMP.is_integer","text":"is_integer(v::GenericVariableRef)\n\nReturn true if v is constrained to be integer.\n\nSee also IntegerRef, set_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_integer(x)\nfalse\n\njulia> set_integer(x)\n\njulia> is_integer(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_parameter","page":"JuMP","title":"is_parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_parameter","category":"page"},{"location":"api/JuMP/#JuMP.is_parameter","page":"JuMP","title":"JuMP.is_parameter","text":"is_parameter(x::GenericVariableRef)::Bool\n\nReturn true if x is constrained to be a parameter.\n\nSee also ParameterRef, set_parameter_value, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> is_parameter(p)\ntrue\n\njulia> @variable(model, x)\nx\n\njulia> is_parameter(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#is_valid","page":"JuMP","title":"is_valid","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.is_valid","category":"page"},{"location":"api/JuMP/#JuMP.is_valid","page":"JuMP","title":"JuMP.is_valid","text":"is_valid(model::GenericModel, con_ref::ConstraintRef{<:AbstractModel})\n\nReturn true if constraint_ref refers to a valid constraint in model.\n\n\n\n\n\nis_valid(model::GenericModel, variable_ref::GenericVariableRef)\n\nReturn true if variable refers to a valid variable in model.\n\n\n\n\n\nis_valid(model::Model, c::NonlinearConstraintRef)\n\nReturn true if c refers to a valid nonlinear constraint in model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#isequal_canonical","page":"JuMP","title":"isequal_canonical","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.isequal_canonical","category":"page"},{"location":"api/JuMP/#JuMP.isequal_canonical","page":"JuMP","title":"JuMP.isequal_canonical","text":"isequal_canonical(\n    aff::GenericAffExpr{C,V},\n    other::GenericAffExpr{C,V}\n) where {C,V}\n\nReturn true if aff is equal to other after dropping zeros and disregarding the order. Mainly useful for testing.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#jump_function","page":"JuMP","title":"jump_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.jump_function","category":"page"},{"location":"api/JuMP/#JuMP.jump_function","page":"JuMP","title":"JuMP.jump_function","text":"jump_function(x)\n\nGiven an MathOptInterface object x, return the JuMP equivalent.\n\nSee also: moi_function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#jump_function_type","page":"JuMP","title":"jump_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.jump_function_type","category":"page"},{"location":"api/JuMP/#JuMP.jump_function_type","page":"JuMP","title":"JuMP.jump_function_type","text":"jump_function_type(::Type{T}) where {T}\n\nGiven an MathOptInterface object type T, return the JuMP equivalent.\n\nSee also: moi_function_type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#latex_formulation","page":"JuMP","title":"latex_formulation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.latex_formulation","category":"page"},{"location":"api/JuMP/#JuMP.latex_formulation","page":"JuMP","title":"JuMP.latex_formulation","text":"latex_formulation(model::AbstractModel)\n\nWrap model in a type so that it can be pretty-printed as text/latex in a notebook like IJulia, or in Documenter.\n\nTo render the model, end the cell with latex_formulation(model), or call display(latex_formulation(model)) in to force the display of the model from inside a function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#linear_terms","page":"JuMP","title":"linear_terms","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.linear_terms","category":"page"},{"location":"api/JuMP/#JuMP.linear_terms","page":"JuMP","title":"JuMP.linear_terms","text":"linear_terms(aff::GenericAffExpr{C, V})\n\nProvides an iterator over coefficient-variable tuples (a_i::C, x_i::V) in the linear part of the affine expression.\n\n\n\n\n\nlinear_terms(quad::GenericQuadExpr{C, V})\n\nProvides an iterator over tuples (coefficient::C, variable::V) in the linear part of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#list_of_constraint_types","page":"JuMP","title":"list_of_constraint_types","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.list_of_constraint_types","category":"page"},{"location":"api/JuMP/#JuMP.list_of_constraint_types","page":"JuMP","title":"JuMP.list_of_constraint_types","text":"list_of_constraint_types(model::GenericModel)::Vector{Tuple{Type,Type}}\n\nReturn a list of tuples of the form (F, S) where F is a JuMP function type and S is an MOI set type such that all_constraints(model, F, S) returns a nonempty list.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> list_of_constraint_types(model)\n3-element Vector{Tuple{Type, Type}}:\n (AffExpr, MathOptInterface.LessThan{Float64})\n (VariableRef, MathOptInterface.GreaterThan{Float64})\n (VariableRef, MathOptInterface.ZeroOne)\n\nPerformance considerations\n\nIterating over the list of function and set types is a type-unstable operation. Consider using a function barrier. See the Performance tips for extensions section of the documentation for more details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#lower_bound","page":"JuMP","title":"lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.lower_bound","page":"JuMP","title":"JuMP.lower_bound","text":"lower_bound(v::GenericVariableRef)\n\nReturn the lower bound of a variable. Error if one does not exist.\n\nSee also LowerBoundRef, has_lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> lower_bound(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#lp_sensitivity_report","page":"JuMP","title":"lp_sensitivity_report","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.lp_sensitivity_report","category":"page"},{"location":"api/JuMP/#JuMP.lp_sensitivity_report","page":"JuMP","title":"JuMP.lp_sensitivity_report","text":"lp_sensitivity_report(model::GenericModel{T}; atol::T = Base.rtoldefault(T))::SensitivityReport{T} where {T}\n\nGiven a linear program model with a current optimal basis, return a SensitivityReport object, which maps:\n\nEvery variable reference to a tuple (d_lo, d_hi)::Tuple{T,T}, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.\nEvery constraint reference to a tuple (d_lo, d_hi)::Tuple{T,T}, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.\n\nBoth tuples are relative, rather than absolute. So given a objective coefficient of 1.0 and a tuple (-0.5, 0.5), the objective coefficient can range between 1.0 - 0.5 an 1.0 + 0.5.\n\natol is the primal/dual optimality tolerance, and should match the tolerance of the solver used to compute the basis.\n\nNote: interval constraints are NOT supported.\n\nExample\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, -1 <= x <= 2)\nx\n\njulia> @objective(model, Min, x)\nx\n\njulia> optimize!(model)\n\njulia> report = lp_sensitivity_report(model; atol = 1e-7);\n\njulia> dx_lo, dx_hi = report[x]\n(-1.0, Inf)\n\njulia> println(\n           \"The objective coefficient of `x` can decrease by $dx_lo or \" *\n           \"increase by $dx_hi.\"\n       )\nThe objective coefficient of `x` can decrease by -1.0 or increase by Inf.\n\njulia> dRHS_lo, dRHS_hi = report[LowerBoundRef(x)]\n(-Inf, 3.0)\n\njulia> println(\n           \"The lower bound of `x` can decrease by $dRHS_lo or increase \" *\n           \"by $dRHS_hi.\"\n       )\nThe lower bound of `x` can decrease by -Inf or increase by 3.0.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#map_coefficients","page":"JuMP","title":"map_coefficients","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.map_coefficients","category":"page"},{"location":"api/JuMP/#JuMP.map_coefficients","page":"JuMP","title":"JuMP.map_coefficients","text":"map_coefficients(f::Function, a::GenericAffExpr)\n\nApply f to the coefficients and constant term of an GenericAffExpr a and return a new expression.\n\nSee also: map_coefficients_inplace!\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = GenericAffExpr(1.0, x => 1.0)\nx + 1\n\njulia> map_coefficients(c -> 2 * c, a)\n2 x + 2\n\njulia> a\nx + 1\n\n\n\n\n\nmap_coefficients(f::Function, a::GenericQuadExpr)\n\nApply f to the coefficients and constant term of an GenericQuadExpr a and return a new expression.\n\nSee also: map_coefficients_inplace!\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = @expression(model, x^2 + x + 1)\nx² + x + 1\n\njulia> map_coefficients(c -> 2 * c, a)\n2 x² + 2 x + 2\n\njulia> a\nx² + x + 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#map_coefficients_inplace!","page":"JuMP","title":"map_coefficients_inplace!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.map_coefficients_inplace!","category":"page"},{"location":"api/JuMP/#JuMP.map_coefficients_inplace!","page":"JuMP","title":"JuMP.map_coefficients_inplace!","text":"map_coefficients_inplace!(f::Function, a::GenericAffExpr)\n\nApply f to the coefficients and constant term of an GenericAffExpr a and update them in-place.\n\nSee also: map_coefficients\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = GenericAffExpr(1.0, x => 1.0)\nx + 1\n\njulia> map_coefficients_inplace!(c -> 2 * c, a)\n2 x + 2\n\njulia> a\n2 x + 2\n\n\n\n\n\nmap_coefficients_inplace!(f::Function, a::GenericQuadExpr)\n\nApply f to the coefficients and constant term of an GenericQuadExpr a and update them in-place.\n\nSee also: map_coefficients\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> a = @expression(model, x^2 + x + 1)\nx² + x + 1\n\njulia> map_coefficients_inplace!(c -> 2 * c, a)\n2 x² + 2 x + 2\n\njulia> a\n2 x² + 2 x + 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#mode","page":"JuMP","title":"mode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.mode","category":"page"},{"location":"api/JuMP/#JuMP.mode","page":"JuMP","title":"JuMP.mode","text":"mode(model::GenericModel)\n\nReturn the ModelMode (DIRECT, AUTOMATIC, or MANUAL) of model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#model_convert","page":"JuMP","title":"model_convert","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.model_convert","category":"page"},{"location":"api/JuMP/#JuMP.model_convert","page":"JuMP","title":"JuMP.model_convert","text":"model_convert(\n    model::AbstractModel,\n    rhs::Union{\n        AbstractConstraint,\n        Number,\n        AbstractJuMPScalar,\n        MOI.AbstractSet,\n    },\n)\n\nConvert the coefficients and constants of functions and sets in the rhs to the coefficient type value_type(typeof(model)).\n\nPurpose\n\nCreating and adding a constraint is a two-step process. The first step calls build_constraint, and the result of that is passed to add_constraint.\n\nHowever, because build_constraint does not take the model as an argument, the coefficients and constants of the function or set might be different than value_type(typeof(model)).\n\nTherefore, the result of build_constraint is converted in a call to model_convert before the result is passed to add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#model_string","page":"JuMP","title":"model_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.model_string","category":"page"},{"location":"api/JuMP/#JuMP.model_string","page":"JuMP","title":"JuMP.model_string","text":"model_string(mode::MIME, model::AbstractModel)\n\nReturn a String representation of model given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_function","page":"JuMP","title":"moi_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_function","category":"page"},{"location":"api/JuMP/#JuMP.moi_function","page":"JuMP","title":"JuMP.moi_function","text":"moi_function(x)\n\nGiven a JuMP object x, return the MathOptInterface equivalent.\n\nSee also: jump_function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_function_type","page":"JuMP","title":"moi_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_function_type","category":"page"},{"location":"api/JuMP/#JuMP.moi_function_type","page":"JuMP","title":"JuMP.moi_function_type","text":"moi_function_type(::Type{T}) where {T}\n\nGiven a JuMP object type T, return the MathOptInterface equivalent.\n\nSee also: jump_function_type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#moi_set","page":"JuMP","title":"moi_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.moi_set","category":"page"},{"location":"api/JuMP/#JuMP.moi_set","page":"JuMP","title":"JuMP.moi_set","text":"moi_set(constraint::AbstractConstraint)\n\nReturn the set of the constraint constraint in the function-in-set form as a MathOptInterface.AbstractSet.\n\nmoi_set(s::AbstractVectorSet, dim::Int)\n\nReturns the MOI set of dimension dim corresponding to the JuMP set s.\n\nmoi_set(s::AbstractScalarSet)\n\nReturns the MOI set corresponding to the JuMP set s.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#name","page":"JuMP","title":"name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.name","category":"page"},{"location":"api/JuMP/#JuMP.name","page":"JuMP","title":"JuMP.name","text":"name(con_ref::ConstraintRef)\n\nGet a constraint's name attribute.\n\n\n\n\n\nname(v::GenericVariableRef)::String\n\nGet a variable's name attribute.\n\n\n\n\n\nname(model::AbstractModel)\n\nReturn the MOI.Name attribute of model's backend, or a default if empty.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#node_count","page":"JuMP","title":"node_count","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.node_count","category":"page"},{"location":"api/JuMP/#JuMP.node_count","page":"JuMP","title":"JuMP.node_count","text":"node_count(model::GenericModel)\n\nGets the total number of branch-and-bound nodes explored during the most recent optimization in a Mixed Integer Program.\n\nSolvers must implement MOI.NodeCount() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_constraint_string","page":"JuMP","title":"nonlinear_constraint_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_constraint_string","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_constraint_string","page":"JuMP","title":"JuMP.nonlinear_constraint_string","text":"nonlinear_constraint_string(\n    model::GenericModel,\n    mode::MIME,\n    c::_NonlinearConstraint,\n)\n\nReturn a string representation of the nonlinear constraint c belonging to model, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_dual_start_value","page":"JuMP","title":"nonlinear_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_dual_start_value","page":"JuMP","title":"JuMP.nonlinear_dual_start_value","text":"nonlinear_dual_start_value(model::Model)\n\nReturn the current value of the MOI attribute MOI.NLPBlockDualStart.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_expr_string","page":"JuMP","title":"nonlinear_expr_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_expr_string","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_expr_string","page":"JuMP","title":"JuMP.nonlinear_expr_string","text":"nonlinear_expr_string(\n    model::GenericModel,\n    mode::MIME,\n    c::MOI.Nonlinear.Expression,\n)\n\nReturn a string representation of the nonlinear expression c belonging to model, given the mode.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#nonlinear_model","page":"JuMP","title":"nonlinear_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.nonlinear_model","category":"page"},{"location":"api/JuMP/#JuMP.nonlinear_model","page":"JuMP","title":"JuMP.nonlinear_model","text":"nonlinear_model(\n    model::GenericModel;\n    force::Bool = false,\n)::Union{MOI.Nonlinear.Model,Nothing}\n\nIf model has nonlinear components, return a MOI.Nonlinear.Model, otherwise return nothing.\n\nIf force, always return a MOI.Nonlinear.Model, and if one does not exist for the model, create an empty one.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#normalized_coefficient","page":"JuMP","title":"normalized_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.normalized_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.normalized_coefficient","page":"JuMP","title":"JuMP.normalized_coefficient","text":"normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef)\n\nReturn the coefficient associated with variable in constraint after JuMP has normalized the constraint into its standard form. See also set_normalized_coefficient.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#normalized_rhs","page":"JuMP","title":"normalized_rhs","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.normalized_rhs","category":"page"},{"location":"api/JuMP/#JuMP.normalized_rhs","page":"JuMP","title":"JuMP.normalized_rhs","text":"normalized_rhs(con_ref::ConstraintRef)\n\nReturn the right-hand side term of con_ref after JuMP has converted the constraint into its normalized form. See also set_normalized_rhs.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_constraints","page":"JuMP","title":"num_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_constraints","category":"page"},{"location":"api/JuMP/#JuMP.num_constraints","page":"JuMP","title":"JuMP.num_constraints","text":"num_constraints(model::GenericModel, function_type, set_type)::Int64\n\nReturn the number of constraints currently in the model where the function has type function_type and the set has type set_type.\n\nSee also list_of_constraint_types and all_constraints.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Bin);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, y in MOI.GreaterThan(1.0));\n\njulia> @constraint(model, y <= 1.0);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> num_constraints(model, VariableRef, MOI.GreaterThan{Float64})\n2\n\njulia> num_constraints(model, VariableRef, MOI.ZeroOne)\n1\n\njulia> num_constraints(model, AffExpr, MOI.LessThan{Float64})\n2\n\n\n\n\n\nnum_constraints(model::GenericModel; count_variable_in_set_constraints::Bool)\n\nReturn the number of constraints in model.\n\nIf count_variable_in_set_constraints == true, then VariableRef constraints such as VariableRef-in-Integer are included. To count only the number of structural constraints (e.g., the rows in the constraint matrix of a linear program), pass count_variable_in_set_constraints = false.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x >= 0, Int);\n\njulia> @constraint(model, 2x <= 1);\n\njulia> num_constraints(model; count_variable_in_set_constraints = true)\n3\n\njulia> num_constraints(model; count_variable_in_set_constraints = false)\n1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_nonlinear_constraints","page":"JuMP","title":"num_nonlinear_constraints","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_nonlinear_constraints","category":"page"},{"location":"api/JuMP/#JuMP.num_nonlinear_constraints","page":"JuMP","title":"JuMP.num_nonlinear_constraints","text":"num_nonlinear_constraints(model::GenericModel)\n\nReturns the number of nonlinear constraints associated with the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#num_variables","page":"JuMP","title":"num_variables","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.num_variables","category":"page"},{"location":"api/JuMP/#JuMP.num_variables","page":"JuMP","title":"JuMP.num_variables","text":"num_variables(model::GenericModel)::Int64\n\nReturns number of variables in model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#object_dictionary","page":"JuMP","title":"object_dictionary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.object_dictionary","category":"page"},{"location":"api/JuMP/#JuMP.object_dictionary","page":"JuMP","title":"JuMP.object_dictionary","text":"object_dictionary(model::GenericModel)\n\nReturn the dictionary that maps the symbol name of a variable, constraint, or expression to the corresponding object.\n\nObjects are registered to a specific symbol in the macros. For example, @variable(model, x[1:2, 1:2]) registers the array of variables x to the symbol :x.\n\nThis method should be defined for any subtype of AbstractModel.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_bound","page":"JuMP","title":"objective_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_bound","category":"page"},{"location":"api/JuMP/#JuMP.objective_bound","page":"JuMP","title":"JuMP.objective_bound","text":"objective_bound(model::GenericModel)\n\nReturn the best known bound on the optimal objective value after a call to optimize!(model).\n\nFor scalar-valued objectives, this function returns a Float64. For vector-valued objectives, it returns a Vector{Float64}.\n\nIn the case of a vector-valued objective, this returns the ideal point, that is, the point obtained if each objective was optimized independently.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function","page":"JuMP","title":"objective_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function","category":"page"},{"location":"api/JuMP/#JuMP.objective_function","page":"JuMP","title":"JuMP.objective_function","text":"objective_function(\n    model::GenericModel,\n    T::Type = objective_function_type(model),\n)\n\nReturn an object of type T representing the objective function.\n\nError if the objective is not convertible to type T.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @objective(model, Min, 2x + 1)\n2 x + 1\n\njulia> objective_function(model, AffExpr)\n2 x + 1\n\njulia> objective_function(model, QuadExpr)\n2 x + 1\n\njulia> typeof(objective_function(model, QuadExpr))\nQuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})\n\nWe see with the last two commands that even if the objective function is affine, as it is convertible to a quadratic function, it can be queried as a quadratic function and the result is quadratic.\n\nHowever, it is not convertible to a variable.\n\njulia> objective_function(model, VariableRef)\nERROR: InexactError: convert(MathOptInterface.VariableIndex, 1.0 + 2.0 MOI.VariableIndex(1))\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function_string","page":"JuMP","title":"objective_function_string","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function_string","category":"page"},{"location":"api/JuMP/#JuMP.objective_function_string","page":"JuMP","title":"JuMP.objective_function_string","text":"objective_function_string(mode, model::AbstractModel)::String\n\nReturn a String describing the objective function of the model.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_function_type","page":"JuMP","title":"objective_function_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_function_type","category":"page"},{"location":"api/JuMP/#JuMP.objective_function_type","page":"JuMP","title":"JuMP.objective_function_type","text":"objective_function_type(model::GenericModel)::AbstractJuMPScalar\n\nReturn the type of the objective function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_sense","page":"JuMP","title":"objective_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_sense","category":"page"},{"location":"api/JuMP/#JuMP.objective_sense","page":"JuMP","title":"JuMP.objective_sense","text":"objective_sense(model::GenericModel)::MOI.OptimizationSense\n\nReturn the objective sense.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#objective_value","page":"JuMP","title":"objective_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.objective_value","category":"page"},{"location":"api/JuMP/#JuMP.objective_value","page":"JuMP","title":"JuMP.objective_value","text":"objective_value(model::GenericModel; result::Int = 1)\n\nReturn the objective value associated with result index result of the most-recent solution returned by the solver.\n\nFor scalar-valued objectives, this function returns a Float64. For vector-valued objectives, it returns a Vector{Float64}.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#op_ifelse","page":"JuMP","title":"op_ifelse","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_ifelse","category":"page"},{"location":"api/JuMP/#JuMP.op_ifelse","page":"JuMP","title":"JuMP.op_ifelse","text":"op_ifelse(a, x, y)\n\nA function that falls back to ifelse(a, x, y), but when called with a JuMP variables or expression in the first argument, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_ifelse(true, 1.0, 2.0)\n1.0\n\njulia> op_ifelse(x, 1.0, 2.0)\nifelse(x, 1.0, 2.0)\n\njulia> op_ifelse(true, x, 2.0)\nx\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#operator_to_set","page":"JuMP","title":"operator_to_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.operator_to_set","category":"page"},{"location":"api/JuMP/#JuMP.operator_to_set","page":"JuMP","title":"JuMP.operator_to_set","text":"operator_to_set(_error::Function, ::Val{sense_symbol})\n\nConverts a sense symbol to a set set such that @constraint(model, func sense_symbol 0) is equivalent to @constraint(model, func in set) for any func::AbstractJuMPScalar.\n\nExample\n\nOnce a custom set is defined you can directly create a JuMP constraint with it:\n\njulia> struct CustomSet{T} <: MOI.AbstractScalarSet\n           value::T\n       end\n\njulia> Base.copy(x::CustomSet) = CustomSet(x.value)\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> cref = @constraint(model, x in CustomSet(1.0))\nx ∈ CustomSet{Float64}(1.0)\n\nHowever, there might be an appropriate sign that could be used in order to provide a more convenient syntax:\n\njulia> JuMP.operator_to_set(::Function, ::Val{:⊰}) = CustomSet(0.0)\n\njulia> MOIU.supports_shift_constant(::Type{<:CustomSet}) = true\n\njulia> MOIU.shift_constant(set::CustomSet, value) = CustomSet(set.value + value)\n\njulia> cref = @constraint(model, x ⊰ 1)\nx ∈ CustomSet{Float64}(1.0)\n\nNote that the whole function is first moved to the right-hand side, then the sign is transformed into a set with zero constant and finally the constant is moved to the set with MOIU.shift_constant.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#operator_warn","page":"JuMP","title":"operator_warn","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.operator_warn","category":"page"},{"location":"api/JuMP/#JuMP.operator_warn","page":"JuMP","title":"JuMP.operator_warn","text":"operator_warn(model::AbstractModel)\noperator_warn(model::GenericModel)\n\nThis function is called on the model whenever two affine expressions are added together without using destructive_add!, and at least one of the two expressions has more than 50 terms.\n\nFor the case of Model, if this function is called more than 20,000 times then a warning is generated once.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimize!","page":"JuMP","title":"optimize!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimize!","category":"page"},{"location":"api/JuMP/#JuMP.optimize!","page":"JuMP","title":"JuMP.optimize!","text":"optimize!(\n    model::GenericModel;\n    ignore_optimize_hook = (model.optimize_hook === nothing),\n    _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =\n        MOI.Nonlinear.SparseReverseMode(),\n    kwargs...,\n)\n\nOptimize the model.\n\nIf an optimizer has not been set yet (see set_optimizer), a NoOptimizer error is thrown.\n\nIf ignore_optimize_hook == true, the optimize hook is ignored and the model is solved as if the hook was not set. Keyword arguments kwargs are passed to the optimize_hook. An error is thrown if optimize_hook is nothing and keyword arguments are provided.\n\nExperimental features\n\nThese features may change or be removed in any future version of JuMP.\n\nPass _differentiation_backend to set the MOI.Nonlinear.AbstractAutomaticDifferentiation backend used to compute derivatives of nonlinear programs.\n\nIf you require only :ExprGraph, it is more efficient to pass _differentiation_backend = MOI.Nonlinear.ExprGraphOnly().\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimizer_index","page":"JuMP","title":"optimizer_index","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimizer_index","category":"page"},{"location":"api/JuMP/#JuMP.optimizer_index","page":"JuMP","title":"JuMP.optimizer_index","text":"optimizer_index(x::GenericVariableRef)::MOI.VariableIndex\noptimizer_index(x::ConstraintRef{<:GenericModel})::MOI.ConstraintIndex\n\nReturn the index that corresponds to x in the optimizer model.\n\nThrows NoOptimizer if no optimizer is set, and throws an ErrorException if the optimizer is set but is not attached.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#optimizer_with_attributes","page":"JuMP","title":"optimizer_with_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.optimizer_with_attributes","category":"page"},{"location":"api/JuMP/#JuMP.optimizer_with_attributes","page":"JuMP","title":"JuMP.optimizer_with_attributes","text":"optimizer_with_attributes(optimizer_constructor, attrs::Pair...)\n\nGroups an optimizer constructor with the list of attributes attrs. Note that it is equivalent to MOI.OptimizerWithAttributes.\n\nWhen provided to the Model constructor or to set_optimizer, it creates an optimizer by calling optimizer_constructor(), and then sets the attributes using set_attribute.\n\nSee also: set_attribute, get_attribute.\n\nNote\n\nThe string names of the attributes are specific to each solver. One should consult the solver's documentation to find the attributes of interest.\n\nExample\n\njulia> import HiGHS\n\njulia> optimizer = optimizer_with_attributes(\n           HiGHS.Optimizer, \"presolve\" => \"off\", MOI.Silent() => true,\n       );\n\njulia> model = Model(optimizer);\n\nis equivalent to:\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_attribute(model, \"presolve\", \"off\")\n\njulia> set_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#owner_model","page":"JuMP","title":"owner_model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.owner_model","category":"page"},{"location":"api/JuMP/#JuMP.owner_model","page":"JuMP","title":"JuMP.owner_model","text":"owner_model(s::AbstractJuMPScalar)\n\nReturn the model owning the scalar s.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parameter_value","page":"JuMP","title":"parameter_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parameter_value","category":"page"},{"location":"api/JuMP/#JuMP.parameter_value","page":"JuMP","title":"JuMP.parameter_value","text":"parameter_value(x::GenericVariableRef)\n\nReturn the value of the parameter x.\n\nErrors if x is not a parameter.\n\nSee also ParameterRef, is_parameter, set_parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> parameter_value(p)\n2.0\n\njulia> set_parameter_value(p, 2.5)\n\njulia> parameter_value(p)\n2.5\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint","page":"JuMP","title":"parse_constraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint","page":"JuMP","title":"JuMP.parse_constraint","text":"parse_constraint(_error::Function, expr::Expr)\n\nThe entry-point for all constraint-related parsing.\n\nArguments\n\nThe _error function is passed everywhere to provide better error messages\nexpr comes from the @constraint macro. There are two possibilities:\n@constraint(model, expr)\n@constraint(model, name[args], expr)\nIn both cases, expr is the main component of the constraint.\n\nSupported syntax\n\nJuMP currently supports the following expr objects:\n\nlhs <= rhs\nlhs == rhs\nlhs >= rhs\nl <= body <= u\nu >= body >= l\nlhs ⟂ rhs\nlhs in rhs\nlhs ∈ rhs\nz => {constraint}\n!z => {constraint}\n\nas well as all broadcasted variants.\n\nExtensions\n\nThe infrastructure behind parse_constraint is extendable. See parse_constraint_head and parse_constraint_call for details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint_call","page":"JuMP","title":"parse_constraint_call","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint_call","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint_call","page":"JuMP","title":"JuMP.parse_constraint_call","text":"parse_constraint_call(\n    _error::Function,\n    is_vectorized::Bool,\n    ::Val{op},\n    args...,\n)\n\nImplement this method to intercept the parsing of a :call expression with operator op.\n\nwarning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.\n\nArguments\n\n_error: a function that accepts a String and throws the string as an error, along with some descriptive information of the macro from which it was thrown.\nis_vectorized: a boolean to indicate if op should be broadcast or not\nop: the first element of the .args field of the Expr to intercept\nargs...: the .args field of the Expr.\n\nReturns\n\nThis function must return:\n\nparse_code::Expr: an expression containing any setup or rewriting code that needs to be called before build_constraint\nbuild_code::Expr: an expression that calls build_constraint( or build_constraint.( depending on is_vectorized.\n\nSee also: parse_constraint_head, build_constraint\n\n\n\n\n\nparse_constraint_call(\n    _error::Function,\n    vectorized::Bool,\n    ::Val{op},\n    lhs,\n    rhs,\n) where {op}\n\nFallback handler for binary operators. These might be infix operators like @constraint(model, lhs op rhs), or normal operators like @constraint(model, op(lhs, rhs)).\n\nIn both cases, we rewrite as lhs - rhs in operator_to_set(_error, op).\n\nSee operator_to_set for details.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_constraint_head","page":"JuMP","title":"parse_constraint_head","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_constraint_head","category":"page"},{"location":"api/JuMP/#JuMP.parse_constraint_head","page":"JuMP","title":"JuMP.parse_constraint_head","text":"parse_constraint_head(_error::Function, ::Val{head}, args...)\n\nImplement this method to intercept the parsing of an expression with head head.\n\nwarning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.\n\nArguments\n\n_error: a function that accepts a String and throws the string as an error, along with some descriptive information of the macro from which it was thrown.\nhead: the .head field of the Expr to intercept\nargs...: the .args field of the Expr.\n\nReturns\n\nThis function must return:\n\nis_vectorized::Bool: whether the expression represents a broadcasted expression like x .<= 1\nparse_code::Expr: an expression containing any setup or rewriting code that needs to be called before build_constraint\nbuild_code::Expr: an expression that calls build_constraint( or build_constraint.( depending on is_vectorized.\n\nExisting implementations\n\nJuMP currently implements:\n\n::Val{:call}, which forwards calls to parse_constraint_call\n::Val{:comparison}, which handles the special case of l <= body <= u.\n\nSee also: parse_constraint_call, build_constraint\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_one_operator_variable","page":"JuMP","title":"parse_one_operator_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_one_operator_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_one_operator_variable","page":"JuMP","title":"JuMP.parse_one_operator_variable","text":"parse_one_operator_variable(_error::Function, infoexpr::_VariableInfoExpr, sense::Val{S}, value) where S\n\nUpdate infoexr for a variable expression in the @variable macro of the form variable name S value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_ternary_variable","page":"JuMP","title":"parse_ternary_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_ternary_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_ternary_variable","page":"JuMP","title":"JuMP.parse_ternary_variable","text":"parse_ternary_variable(_error, variable_info, lhs_sense, lhs, rhs_sense, rhs)\n\nA hook for JuMP extensiosn to intercept the parsing of a :comparison expression, which has the form lhs lhs_sense variable rhs_sense rhs.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#parse_variable","page":"JuMP","title":"parse_variable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.parse_variable","category":"page"},{"location":"api/JuMP/#JuMP.parse_variable","page":"JuMP","title":"JuMP.parse_variable","text":"parse_variable(_error::Function, ::_VariableInfoExpr, args...)\n\nA hook for extensions to intercept the parsing of inequality constraints in the @variable macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#primal_feasibility_report","page":"JuMP","title":"primal_feasibility_report","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.primal_feasibility_report","category":"page"},{"location":"api/JuMP/#JuMP.primal_feasibility_report","page":"JuMP","title":"JuMP.primal_feasibility_report","text":"primal_feasibility_report(\n    model::GenericModel{T},\n    point::AbstractDict{GenericVariableRef{T},T} = _last_primal_solution(model),\n    atol::T = zero(T),\n    skip_missing::Bool = false,\n)::Dict{Any,T}\n\nGiven a dictionary point, which maps variables to primal values, return a dictionary whose keys are the constraints with an infeasibility greater than the supplied tolerance atol. The value corresponding to each key is the respective infeasibility. Infeasibility is defined as the distance between the primal value of the constraint (see MOI.ConstraintPrimal) and the nearest point by Euclidean distance in the corresponding set.\n\nNotes\n\nIf skip_missing = true, constraints containing variables that are not in point will be ignored.\nIf skip_missing = false and a partial primal solution is provided, an error will be thrown.\nIf no point is provided, the primal solution from the last time the model was solved is used.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, 0.5 <= x <= 1);\n\njulia> primal_feasibility_report(model, Dict(x => 0.2))\nDict{Any, Float64} with 1 entry:\n  x ≥ 0.5 => 0.3\n\n\n\n\n\nprimal_feasibility_report(\n    point::Function,\n    model::GenericModel{T};\n    atol::T = zero(T),\n    skip_missing::Bool = false,\n) where {T}\n\nA form of primal_feasibility_report where a function is passed as the first argument instead of a dictionary as the second argument.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, 0.5 <= x <= 1, start = 1.3);\n\njulia> primal_feasibility_report(model) do v\n           return start_value(v)\n       end\nDict{Any, Float64} with 1 entry:\n  x ≤ 1 => 0.3\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#primal_status","page":"JuMP","title":"primal_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.primal_status","category":"page"},{"location":"api/JuMP/#JuMP.primal_status","page":"JuMP","title":"JuMP.primal_status","text":"primal_status(model::GenericModel; result::Int = 1)\n\nReturn a MOI.ResultStatusCode describing the status of the most recent primal solution of the solver (i.e., the MOI.PrimalStatus attribute) associated with the result index result.\n\nSee also: result_count.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#print_active_bridges","page":"JuMP","title":"print_active_bridges","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.print_active_bridges","category":"page"},{"location":"api/JuMP/#JuMP.print_active_bridges","page":"JuMP","title":"JuMP.print_active_bridges","text":"print_active_bridges([io::IO = stdout,] model::GenericModel)\n\nPrint a list of the variable, constraint, and objective bridges that are currently used in the model.\n\n\n\n\n\nprint_active_bridges([io::IO = stdout,] model::GenericModel, ::Type{F}) where {F}\n\nPrint a list of bridges required for an objective function of type F.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO = stdout,]\n    model::GenericModel,\n    F::Type,\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint a list of bridges required for a constraint of type F-in-S.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO = stdout,]\n    model::GenericModel,\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint a list of bridges required to add a variable constrained to the set S.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#print_bridge_graph","page":"JuMP","title":"print_bridge_graph","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.print_bridge_graph","category":"page"},{"location":"api/JuMP/#JuMP.print_bridge_graph","page":"JuMP","title":"JuMP.print_bridge_graph","text":" print_bridge_graph([io::IO,] model::GenericModel)\n\nPrint the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model.\n\nwarning: Warning\nThis function is intended for advanced users. If you want to see only the bridges that are currently used, use print_active_bridges instead.\n\nExplanation of output\n\nEach node in the hyper-graph corresponds to a variable, constraint, or objective type.\n\nVariable nodes are indicated by [ ]\nConstraint nodes are indicated by ( )\nObjective nodes are indicated by | |\n\nThe number inside each pair of brackets is an index of the node in the hyper-graph.\n\nNote that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.\n\nFor more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  \"MathOptInterface: a data structure for mathematical optimization problems.\" URL: https://arxiv.org/abs/2002.03447\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#quad_terms","page":"JuMP","title":"quad_terms","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.quad_terms","category":"page"},{"location":"api/JuMP/#JuMP.quad_terms","page":"JuMP","title":"JuMP.quad_terms","text":"quad_terms(quad::GenericQuadExpr{C, V})\n\nProvides an iterator over tuples (coefficient::C, var_1::V, var_2::V) in the quadratic part of the quadratic expression.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#raw_status","page":"JuMP","title":"raw_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.raw_status","category":"page"},{"location":"api/JuMP/#JuMP.raw_status","page":"JuMP","title":"JuMP.raw_status","text":"raw_status(model::GenericModel)\n\nReturn the reason why the solver stopped in its own words (i.e., the MathOptInterface model attribute RawStatusString).\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#read_from_file","page":"JuMP","title":"read_from_file","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.read_from_file","category":"page"},{"location":"api/JuMP/#JuMP.read_from_file","page":"JuMP","title":"JuMP.read_from_file","text":"read_from_file(\n    filename::String;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,\n    kwargs...,\n)\n\nReturn a JuMP model read from filename in the format format.\n\nIf the filename ends in .gz, it will be uncompressed using Gzip. If the filename ends in .bz2, it will be uncompressed using BZip2.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reduced_cost","page":"JuMP","title":"reduced_cost","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reduced_cost","category":"page"},{"location":"api/JuMP/#JuMP.reduced_cost","page":"JuMP","title":"JuMP.reduced_cost","text":"reduced_cost(x::GenericVariableRef{T})::T where {T}\n\nReturn the reduced cost associated with variable x.\n\nEquivalent to querying the shadow price of the active variable bound (if one exists and is active).\n\nSee also: shadow_price.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#register","page":"JuMP","title":"register","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.register","category":"page"},{"location":"api/JuMP/#JuMP.register","page":"JuMP","title":"JuMP.register","text":"register(\n    model::Model,\n    op::Symbol,\n    dimension::Integer,\n    f::Function;\n    autodiff:Bool = false,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol op.\n\nThe function f must support all subtypes of Real as arguments. Do not assume that the inputs are Float64.\n\nNotes\n\nFor this method, you must explicitly set autodiff = true, because no user-provided gradient function ∇f is given.\nSecond-derivative information is only computed if dimension == 1.\nop does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::T) where {T<:Real} = x^2\nf (generic function with 1 method)\n\njulia> register(model, :foo, 1, f; autodiff = true)\n\njulia> @NLobjective(model, Min, foo(x))\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> g(x::T, y::T) where {T<:Real} = x * y\ng (generic function with 1 method)\n\njulia> register(model, :g, 2, g; autodiff = true)\n\njulia> @NLobjective(model, Min, g(x[1], x[2]))\n\n\n\n\n\nregister(\n    model::Model,\n    s::Symbol,\n    dimension::Integer,\n    f::Function,\n    ∇f::Function;\n    autodiff:Bool = false,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol s. In addition, provide a gradient function ∇f.\n\nThe functions fand ∇f must support all subtypes of Real as arguments. Do not assume that the inputs are Float64.\n\nNotes\n\nIf the function f is univariate (i.e., dimension == 1), ∇f must return a number which represents the first-order derivative of the function f.\nIf the function f is multi-variate, ∇f must have a signature matching ∇f(g::AbstractVector{T}, args::T...) where {T<:Real}, where the first argument is a vector g that is modified in-place with the gradient.\nIf autodiff = true and dimension == 1, use automatic differentiation to compute the second-order derivative information. If autodiff = false, only first-order derivative information will be used.\ns does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::T) where {T<:Real} = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::T) where {T<:Real} = 2 * x\n∇f (generic function with 1 method)\n\njulia> register(model, :foo, 1, f, ∇f; autodiff = true)\n\njulia> @NLobjective(model, Min, foo(x))\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> g(x::T, y::T) where {T<:Real} = x * y\ng (generic function with 1 method)\n\njulia> function ∇g(g::AbstractVector{T}, x::T, y::T) where {T<:Real}\n           g[1] = y\n           g[2] = x\n           return\n       end\n∇g (generic function with 1 method)\n\njulia> register(model, :g, 2, g, ∇g)\n\njulia> @NLobjective(model, Min, g(x[1], x[2]))\n\n\n\n\n\nregister(\n    model::Model,\n    s::Symbol,\n    dimension::Integer,\n    f::Function,\n    ∇f::Function,\n    ∇²f::Function,\n)\n\nRegister the user-defined function f that takes dimension arguments in model as the symbol s. In addition, provide a gradient function ∇f and a hessian function ∇²f.\n\n∇f and ∇²f must return numbers corresponding to the first- and second-order derivatives of the function f respectively.\n\nNotes\n\nBecause automatic differentiation is not used, you can assume the inputs are all Float64.\nThis method will throw an error if dimension > 1.\ns does not have to be the same symbol as f, but it is generally more readable if it is.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> register(model, :foo, 1, f, ∇f, ∇²f)\n\njulia> @NLobjective(model, Min, foo(x))\n\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relative_gap","page":"JuMP","title":"relative_gap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relative_gap","category":"page"},{"location":"api/JuMP/#JuMP.relative_gap","page":"JuMP","title":"JuMP.relative_gap","text":"relative_gap(model::GenericModel)\n\nReturn the final relative optimality gap after a call to optimize!(model). Exact value depends upon implementation of MathOptInterface.RelativeGap() by the particular solver used for optimization.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relax_integrality","page":"JuMP","title":"relax_integrality","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relax_integrality","category":"page"},{"location":"api/JuMP/#JuMP.relax_integrality","page":"JuMP","title":"JuMP.relax_integrality","text":"relax_integrality(model::GenericModel)\n\nModifies model to \"relax\" all binary and integrality constraints on variables. Specifically,\n\nBinary constraints are deleted, and variable bounds are tightened if necessary to ensure the variable is constrained to the interval 0 1.\nIntegrality constraints are deleted without modifying variable bounds.\nAn error is thrown if semi-continuous or semi-integer constraints are present (support may be added for these in the future).\nAll other constraints are ignored (left in place). This includes discrete constraints like SOS and indicator constraints.\n\nReturns a function that can be called without any arguments to restore the original model. The behavior of this function is undefined if additional changes are made to the affected variables in the meantime.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> @variable(model, 1 <= y <= 10, Int);\n\njulia> @objective(model, Min, x + y);\n\njulia> undo_relax = relax_integrality(model);\n\njulia> print(model)\nMin x + y\nSubject to\n x ≥ 0\n y ≥ 1\n x ≤ 1\n y ≤ 10\n\njulia> undo_relax()\n\njulia> print(model)\nMin x + y\nSubject to\n y ≥ 1\n y ≤ 10\n y integer\n x binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#relax_with_penalty!","page":"JuMP","title":"relax_with_penalty!","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.relax_with_penalty!","category":"page"},{"location":"api/JuMP/#JuMP.relax_with_penalty!","page":"JuMP","title":"JuMP.relax_with_penalty!","text":"relax_with_penalty!(\n    model::GenericModel{T},\n    [penalties::Dict{ConstraintRef,T}];\n    [default::Union{Nothing,Real} = nothing,]\n) where {T}\n\nDestructively modify the model in-place to create a penalized relaxation of the constraints.\n\nwarning: Warning\nThis is a destructive routine that modifies the model in-place. If you don't want to modify the original model, use copy_model to create a copy before calling relax_with_penalty!.\n\nReformulation\n\nSee MOI.Utilities.ScalarPenaltyRelaxation for details of the reformulation.\n\nFor each constraint ci, the penalty passed to MOI.Utilities.ScalarPenaltyRelaxation is get(penalties, ci, default). If the value is nothing, because ci does not exist in penalties and default = nothing, then the constraint is skipped.\n\nReturn value\n\nThis function returns a Dict{ConstraintRef,AffExpr} that maps each constraint index to the corresponding y + z as an AffExpr. In an optimal solution, query the value of these functions to compute the violation of each constraint.\n\nRelax a subset of constraints\n\nTo relax a subset of constraints, pass a penalties dictionary and set default = nothing.\n\nExample\n\njulia> function new_model()\n           model = Model()\n           @variable(model, x)\n           @objective(model, Max, 2x + 1)\n           @constraint(model, c1, 2x - 1 <= -2)\n           @constraint(model, c2, 3x >= 0)\n           return model\n       end\nnew_model (generic function with 1 method)\n\njulia> model_1 = new_model();\n\njulia> penalty_map = relax_with_penalty!(model_1; default = 2.0);\n\njulia> penalty_map[model_1[:c1]]\n_[3]\n\njulia> penalty_map[model_1[:c2]]\n_[2]\n\njulia> print(model_1)\nMax 2 x - 2 _[2] - 2 _[3] + 1\nSubject to\n c2 : 3 x + _[2] ≥ 0\n c1 : 2 x - _[3] ≤ -1\n _[2] ≥ 0\n _[3] ≥ 0\n\njulia> model_2 = new_model();\n\njulia> relax_with_penalty!(model_2, Dict(model_2[:c2] => 3.0))\nDict{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.GreaterThan{Float64}}, ScalarShape}, AffExpr} with 1 entry:\n  c2 : 3 x + _[2] ≥ 0 => _[2]\n\njulia> print(model_2)\nMax 2 x - 3 _[2] + 1\nSubject to\n c2 : 3 x + _[2] ≥ 0\n c1 : 2 x ≤ -1\n _[2] ≥ 0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#remove_bridge","page":"JuMP","title":"remove_bridge","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.remove_bridge","category":"page"},{"location":"api/JuMP/#JuMP.remove_bridge","page":"JuMP","title":"JuMP.remove_bridge","text":"remove_bridge(\n    model::GenericModel{S},\n    BT::Type{<:MOI.Bridges.AbstractBridge};\n    coefficient_type::Type{T} = S,\n) where {S,T}\n\nRemove BT{T} from the list of bridges that can be used to transform unsupported constraints into an equivalent formulation using only constraints supported by the optimizer.\n\nSee also: add_bridge.\n\nExample\n\njulia> model = Model();\n\njulia> add_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> remove_bridge(model, MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge)\n\njulia> add_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64},\n       )\n\njulia> remove_bridge(\n           model,\n           MOI.Bridges.Constraint.NumberConversionBridge;\n           coefficient_type = Complex{Float64},\n       )\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reshape_set","page":"JuMP","title":"reshape_set","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reshape_set","category":"page"},{"location":"api/JuMP/#JuMP.reshape_set","page":"JuMP","title":"JuMP.reshape_set","text":"reshape_set(vectorized_set::MOI.AbstractSet, shape::AbstractShape)\n\nReturn a set in its original shape shape given its vectorized form vectorized_form.\n\nExample\n\nGiven a SymmetricMatrixShape of vectorized form [1, 2, 3] in MOI.PositiveSemidefinieConeTriangle(2), the following code returns the set of the original constraint Symmetric(Matrix[1 2; 2 3]) in PSDCone():\n\njulia> reshape_set(MOI.PositiveSemidefiniteConeTriangle(2), SymmetricMatrixShape(2))\nPSDCone()\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reshape_vector","page":"JuMP","title":"reshape_vector","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reshape_vector","category":"page"},{"location":"api/JuMP/#JuMP.reshape_vector","page":"JuMP","title":"JuMP.reshape_vector","text":"reshape_vector(vectorized_form::Vector, shape::AbstractShape)\n\nReturn an object in its original shape shape given its vectorized form vectorized_form.\n\nExample\n\nGiven a SymmetricMatrixShape of vectorized form [1, 2, 3], the following code returns the matrix Symmetric(Matrix[1 2; 2 3]):\n\njulia> reshape_vector([1, 2, 3], SymmetricMatrixShape(2))\n2×2 LinearAlgebra.Symmetric{Int64, Matrix{Int64}}:\n 1  2\n 2  3\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#result_count","page":"JuMP","title":"result_count","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.result_count","category":"page"},{"location":"api/JuMP/#JuMP.result_count","page":"JuMP","title":"JuMP.result_count","text":"result_count(model::GenericModel)\n\nReturn the number of results available to query after a call to optimize!.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#reverse_sense","page":"JuMP","title":"reverse_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.reverse_sense","category":"page"},{"location":"api/JuMP/#JuMP.reverse_sense","page":"JuMP","title":"JuMP.reverse_sense","text":"reverse_sense(::Val{T}) where {T}\n\nGiven an (in)equality symbol T, return a new Val object with the opposite (in)equality symbol.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_attribute","page":"JuMP","title":"set_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_attribute","category":"page"},{"location":"api/JuMP/#JuMP.set_attribute","page":"JuMP","title":"JuMP.set_attribute","text":"set_attribute(model::GenericModel, attr::MOI.AbstractModelAttribute, value)\nset_attribute(x::GenericVariableRef, attr::MOI.AbstractVariableAttribute, value)\nset_attribute(cr::ConstraintRef, attr::MOI.AbstractConstraintAttribute, value)\n\nSet the value of a solver-specifc attribute attr to value.\n\nThis is equivalent to calling MOI.set with the associated MOI model and, for variables and constraints, with the associated MOI.VariableIndex or MOI.ConstraintIndex.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, c, 2 * x <= 1)\nc : 2 x ≤ 1\n\njulia> set_attribute(model, MOI.Name(), \"model_new\")\n\njulia> set_attribute(x, MOI.VariableName(), \"x_new\")\n\njulia> set_attribute(c, MOI.ConstraintName(), \"c_new\")\n\n\n\n\n\nset_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n    value,\n)\n\nSet the value of a solver-specifc attribute attr to value.\n\nThis is equivalent to calling MOI.set with the associated MOI model.\n\nIf attr is an AbstractString, it is converted to MOI.RawOptimizerAttribute.\n\nExample\n\njulia> import HiGHS\n\njulia> opt = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false);\n\njulia> model = Model(opt);\n\njulia> set_attribute(model, \"output_flag\", false)\n\njulia> set_attribute(model, MOI.RawOptimizerAttribute(\"output_flag\"), true)\n\njulia> set_attribute(opt, \"output_flag\", true)\n\njulia> set_attribute(opt, MOI.RawOptimizerAttribute(\"output_flag\"), false)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_attributes","page":"JuMP","title":"set_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_attributes","category":"page"},{"location":"api/JuMP/#JuMP.set_attributes","page":"JuMP","title":"JuMP.set_attributes","text":"set_attributes(\n    destination::Union{\n        GenericModel,\n        MOI.OptimizerWithAttributes,\n        GenericVariableRef,\n        ConstraintRef,\n    },\n    pairs::Pair...,\n)\n\nGiven a list of attribute => value pairs, calls set_attribute(destination, attribute, value) for each pair.\n\nSee also: set_attribute, get_attribute.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_attributes(model, \"tol\" => 1e-4, \"max_iter\" => 100)\n\nis equivalent to:\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_attribute(model, \"tol\", 1e-4)\n\njulia> set_attribute(model, \"max_iter\", 100)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_binary","page":"JuMP","title":"set_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_binary","category":"page"},{"location":"api/JuMP/#JuMP.set_binary","page":"JuMP","title":"JuMP.set_binary","text":"set_binary(v::GenericVariableRef)\n\nAdd a constraint on the variable v that it must take values in the set 01.\n\nSee also BinaryRef, is_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_binary(x)\nfalse\n\njulia> set_binary(x)\n\njulia> is_binary(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_dual_start_value","page":"JuMP","title":"set_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_dual_start_value","page":"JuMP","title":"JuMP.set_dual_start_value","text":"set_dual_start_value(con_ref::ConstraintRef, value)\n\nSet the dual start value (MOI attribute ConstraintDualStart) of the constraint con_ref to value. To remove a dual start value set it to nothing.\n\nSee also dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_integer","page":"JuMP","title":"set_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_integer","category":"page"},{"location":"api/JuMP/#JuMP.set_integer","page":"JuMP","title":"JuMP.set_integer","text":"set_integer(variable_ref::GenericVariableRef)\n\nAdd an integrality constraint on the variable variable_ref.\n\nSee also IntegerRef, is_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> is_integer(x)\nfalse\n\njulia> set_integer(x)\n\njulia> is_integer(x)\ntrue\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_lower_bound","page":"JuMP","title":"set_lower_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_lower_bound","category":"page"},{"location":"api/JuMP/#JuMP.set_lower_bound","page":"JuMP","title":"JuMP.set_lower_bound","text":"set_lower_bound(v::GenericVariableRef, lower::Number)\n\nSet the lower bound of a variable. If one does not exist, create a new lower bound constraint.\n\nSee also LowerBoundRef, has_lower_bound, lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> lower_bound(x)\n1.0\n\njulia> set_lower_bound(x, 2.0)\n\njulia> lower_bound(x)\n2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_name","page":"JuMP","title":"set_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_name","category":"page"},{"location":"api/JuMP/#JuMP.set_name","page":"JuMP","title":"JuMP.set_name","text":"set_name(con_ref::ConstraintRef, s::AbstractString)\n\nSet a constraint's name attribute.\n\n\n\n\n\nset_name(v::GenericVariableRef, s::AbstractString)\n\nSet a variable's name attribute.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_nonlinear_dual_start_value","page":"JuMP","title":"set_nonlinear_dual_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_nonlinear_dual_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_nonlinear_dual_start_value","page":"JuMP","title":"JuMP.set_nonlinear_dual_start_value","text":"set_nonlinear_dual_start_value(\n    model::Model,\n    start::Union{Nothing,Vector{Float64}},\n)\n\nSet the value of the MOI attribute MOI.NLPBlockDualStart.\n\nThe start vector corresponds to the Lagrangian duals of the nonlinear constraints, in the order given by all_nonlinear_constraints. That is, you must pass a single start vector corresponding to all of the nonlinear constraints in a single function call; you cannot set the dual start value of nonlinear constraints one-by-one. The example below demonstrates how to use all_nonlinear_constraints to create a mapping between the nonlinear constraint references and the start vector.\n\nPass nothing to unset a previous start.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> nl1 = @NLconstraint(model, x[1] <= sqrt(x[2]));\n\njulia> nl2 = @NLconstraint(model, x[1] >= exp(x[2]));\n\njulia> start = Dict(nl1 => -1.0, nl2 => 1.0);\n\njulia> start_vector = [start[con] for con in all_nonlinear_constraints(model)]\n2-element Vector{Float64}:\n -1.0\n  1.0\n\njulia> set_nonlinear_dual_start_value(model, start_vector)\n\njulia> nonlinear_dual_start_value(model)\n2-element Vector{Float64}:\n -1.0\n  1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_nonlinear_objective","page":"JuMP","title":"set_nonlinear_objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_nonlinear_objective","category":"page"},{"location":"api/JuMP/#JuMP.set_nonlinear_objective","page":"JuMP","title":"JuMP.set_nonlinear_objective","text":"set_nonlinear_objective(\n    model::Model,\n    sense::MOI.OptimizationSense,\n    expr::Expr,\n)\n\nSet the nonlinear objective of model to the expression expr, with the optimization sense sense.\n\nThis function is most useful if the expression expr is generated programmatically, and you cannot use @NLobjective.\n\nNotes\n\nYou must interpolate the variables directly into the expression expr.\nYou must use MIN_SENSE or MAX_SENSE instead of Min and Max.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> set_nonlinear_objective(model, MIN_SENSE, :($(x) + $(x)^2))\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_coefficient","page":"JuMP","title":"set_normalized_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_coefficient","page":"JuMP","title":"JuMP.set_normalized_coefficient","text":"set_normalized_coefficient(con_ref::ConstraintRef, variable::GenericVariableRef, value)\n\nSet the coefficient of variable in the constraint constraint to value.\n\nNote that prior to this step, JuMP will aggregate multiple terms containing the same variable. For example, given a constraint 2x + 3x <= 2, set_normalized_coefficient(con, x, 4) will create the constraint 4x <= 2.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, 2x + 3x <= 2)\ncon : 5 x ≤ 2\n\njulia> set_normalized_coefficient(con, x, 4)\n\njulia> con\ncon : 4 x ≤ 2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_coefficients","page":"JuMP","title":"set_normalized_coefficients","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_coefficients","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_coefficients","page":"JuMP","title":"JuMP.set_normalized_coefficients","text":"set_normalized_coefficients(\n    con_ref::ConstraintRef,\n    variable,\n    new_coefficients::Vector{Tuple{Int64,T}},\n)\n\nSet the coefficients of variable in the constraint con_ref to new_coefficients, where each element in new_coefficients is a tuple which maps the row to a new coefficient.\n\nNote that prior to this step, during constraint creation, JuMP will aggregate multiple terms containing the same variable.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, con, [2x + 3x, 4x] in MOI.Nonnegatives(2))\ncon : [5 x, 4 x] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> set_normalized_coefficients(con, x, [(1, 2.0), (2, 5.0)])\n\njulia> con\ncon : [2 x, 5 x] ∈ MathOptInterface.Nonnegatives(2)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_normalized_rhs","page":"JuMP","title":"set_normalized_rhs","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_normalized_rhs","category":"page"},{"location":"api/JuMP/#JuMP.set_normalized_rhs","page":"JuMP","title":"JuMP.set_normalized_rhs","text":"set_normalized_rhs(con_ref::ConstraintRef, value)\n\nSet the right-hand side term of constraint to value.\n\nNote that prior to this step, JuMP will aggregate all constant terms onto the right-hand side of the constraint. For example, given a constraint 2x + 1 <= 2, set_normalized_rhs(con, 4) will create the constraint 2x <= 4, not 2x + 1 <= 4.\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, con, 2x + 1 <= 2)\ncon : 2 x ≤ 1\n\njulia> set_normalized_rhs(con, 4)\n\njulia> con\ncon : 2 x ≤ 4\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective","page":"JuMP","title":"set_objective","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective","category":"page"},{"location":"api/JuMP/#JuMP.set_objective","page":"JuMP","title":"JuMP.set_objective","text":"set_objective(model::AbstractModel, sense::MOI.OptimizationSense, func)\n\nThe functional equivalent of the @objective macro.\n\nSets the objective sense and objective function simultaneously, and is equivalent to calling set_objective_sense and set_objective_function separately.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_objective(model, MIN_SENSE, x)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_coefficient","page":"JuMP","title":"set_objective_coefficient","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_coefficient","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_coefficient","page":"JuMP","title":"JuMP.set_objective_coefficient","text":"set_objective_coefficient(model::GenericModel, variable::GenericVariableRef, coefficient::Real)\n\nSet the linear objective coefficient associated with Variable to coefficient.\n\nNote: this function will throw an error if a nonlinear objective is set.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_function","page":"JuMP","title":"set_objective_function","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_function","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_function","page":"JuMP","title":"JuMP.set_objective_function","text":"set_objective_function(model::GenericModel, func::MOI.AbstractFunction)\nset_objective_function(model::GenericModel, func::AbstractJuMPScalar)\nset_objective_function(model::GenericModel, func::Real)\nset_objective_function(model::GenericModel, func::Vector{<:AbstractJuMPScalar})\n\nSets the objective function of the model to the given function. See set_objective_sense to set the objective sense. These are low-level functions; the recommended way to set the objective is with the @objective macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_objective_sense","page":"JuMP","title":"set_objective_sense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_objective_sense","category":"page"},{"location":"api/JuMP/#JuMP.set_objective_sense","page":"JuMP","title":"JuMP.set_objective_sense","text":"set_objective_sense(model::GenericModel, sense::MOI.OptimizationSense)\n\nSets the objective sense of the model to the given sense. See set_objective_function to set the objective function. These are low-level functions; the recommended way to set the objective is with the @objective macro.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimize_hook","page":"JuMP","title":"set_optimize_hook","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimize_hook","category":"page"},{"location":"api/JuMP/#JuMP.set_optimize_hook","page":"JuMP","title":"JuMP.set_optimize_hook","text":"set_optimize_hook(model::GenericModel, f::Union{Function,Nothing})\n\nSet the function f as the optimize hook for model.\n\nf should have a signature f(model::GenericModel; kwargs...), where the kwargs are those passed to optimize!.\n\nNotes\n\nThe optimize hook should generally modify the model, or some external state in some way, and then call optimize!(model; ignore_optimize_hook = true) to optimize the problem, bypassing the hook.\nUse set_optimize_hook(model, nothing) to unset an optimize hook.\n\nExample\n\njulia> model = Model();\n\njulia> function my_hook(model::Model; kwargs...)\n           println(kwargs)\n           println(\"Calling with `ignore_optimize_hook = true`\")\n           optimize!(model; ignore_optimize_hook = true)\n           return\n       end\nmy_hook (generic function with 1 method)\n\njulia> set_optimize_hook(model, my_hook)\nmy_hook (generic function with 1 method)\n\njulia> optimize!(model; test_arg = true)\nBase.Pairs{Symbol, Bool, Tuple{Symbol}, NamedTuple{(:test_arg,), Tuple{Bool}}}(:test_arg => 1)\nCalling with `ignore_optimize_hook = true`\nERROR: NoOptimizer()\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer","page":"JuMP","title":"set_optimizer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer","page":"JuMP","title":"JuMP.set_optimizer","text":"set_optimizer(\n    model::GenericModel,\n    optimizer_factory;\n    add_bridges::Bool = true,\n)\n\nCreates an empty MathOptInterface.AbstractOptimizer instance by calling optimizer_factory() and sets it as the optimizer of model. Specifically, optimizer_factory must be callable with zero arguments and return an empty MathOptInterface.AbstractOptimizer.\n\nIf add_bridges is true, constraints and objectives that are not supported by the optimizer are automatically bridged to equivalent supported formulation. Passing add_bridges = false can improve performance if the solver natively supports all of the elements in model.\n\nSee set_attribute for setting solver-specific parameters of the optimizer.\n\nExample\n\njulia> import HiGHS\n\njulia> model = Model();\n\njulia> set_optimizer(model, () -> HiGHS.Optimizer())\n\njulia> set_optimizer(model, HiGHS.Optimizer; add_bridges = false)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer_attribute","page":"JuMP","title":"set_optimizer_attribute","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer_attribute","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer_attribute","page":"JuMP","title":"JuMP.set_optimizer_attribute","text":"set_optimizer_attribute(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    attr::Union{AbstractString,MOI.AbstractOptimizerAttribute},\n    value,\n)\n\nSet the solver-specific attribute attr in model to value.\n\nIf attr is an AbstractString, this is equivalent to set_optimizer_attribute(model, MOI.RawOptimizerAttribute(name), value).\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using set_attribute instead.\n\nSee also: set_optimizer_attributes, get_optimizer_attribute.\n\nExample\n\njulia> model = Model();\n\njulia> set_optimizer_attribute(model, MOI.Silent(), true)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_optimizer_attributes","page":"JuMP","title":"set_optimizer_attributes","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_optimizer_attributes","category":"page"},{"location":"api/JuMP/#JuMP.set_optimizer_attributes","page":"JuMP","title":"JuMP.set_optimizer_attributes","text":"set_optimizer_attributes(\n    model::Union{GenericModel,MOI.OptimizerWithAttributes},\n    pairs::Pair...,\n)\n\nGiven a list of attribute => value pairs, calls set_optimizer_attribute(model, attribute, value) for each pair.\n\ncompat: Compat\nThis method will remain in all v1.X releases of JuMP, but it may be removed in a future v2.0 release. We recommend using set_attributes instead.\n\nSee also: set_optimizer_attribute, get_optimizer_attribute.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_optimizer_attributes(model, \"tol\" => 1e-4, \"max_iter\" => 100)\n\nis equivalent to:\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> set_optimizer_attribute(model, \"tol\", 1e-4)\n\njulia> set_optimizer_attribute(model, \"max_iter\", 100)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_parameter_value","page":"JuMP","title":"set_parameter_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_parameter_value","category":"page"},{"location":"api/JuMP/#JuMP.set_parameter_value","page":"JuMP","title":"JuMP.set_parameter_value","text":"set_parameter_value(x::GenericVariableRef, value)\n\nUpdate the parameter constraint on the variable x to value.\n\nErrors if x is not a parameter.\n\nSee also ParameterRef, is_parameter, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> parameter_value(p)\n2.0\n\njulia> set_parameter_value(p, 2.5)\n\njulia> parameter_value(p)\n2.5\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_silent","page":"JuMP","title":"set_silent","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_silent","category":"page"},{"location":"api/JuMP/#JuMP.set_silent","page":"JuMP","title":"JuMP.set_silent","text":"set_silent(model::GenericModel)\n\nTakes precedence over any other attribute controlling verbosity and requires the solver to produce no output.\n\nSee also: unset_silent.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_start_value","page":"JuMP","title":"set_start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_start_value","category":"page"},{"location":"api/JuMP/#JuMP.set_start_value","page":"JuMP","title":"JuMP.set_start_value","text":"set_start_value(con_ref::ConstraintRef, value)\n\nSet the primal start value (MOI.ConstraintPrimalStart) of the constraint con_ref to value. To remove a primal start value set it to nothing.\n\nSee also start_value.\n\n\n\n\n\nset_start_value(variable::GenericVariableRef, value::Union{Real,Nothing})\n\nSet the start value (MOI attribute VariablePrimalStart) of the variable to value.\n\nPass nothing to unset the start value.\n\nNote: VariablePrimalStarts are sometimes called \"MIP-starts\" or \"warmstarts\".\n\nSee also start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_start_values","page":"JuMP","title":"set_start_values","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_start_values","category":"page"},{"location":"api/JuMP/#JuMP.set_start_values","page":"JuMP","title":"JuMP.set_start_values","text":"set_start_values(\n    model::GenericModel;\n    variable_primal_start::Union{Nothing,Function} = value,\n    constraint_primal_start::Union{Nothing,Function} = value,\n    constraint_dual_start::Union{Nothing,Function} = dual,\n    nonlinear_dual_start::Union{Nothing,Function} = nonlinear_dual_start_value,\n)\n\nSet the primal and dual starting values in model using the functions provided.\n\nIf any keyword argument is nothing, the corresponding start value is skipped.\n\nIf the optimizer does not support setting the starting value, the value will be skipped.\n\nvariable_primal_start\n\nThis function controls the primal starting solution for the variables. It is equivalent to calling set_start_value for each variable, or setting the MOI.VariablePrimalStart attribute.\n\nIf it is a function, it must have the form variable_primal_start(x::VariableRef) that maps each variable x to the starting primal value.\n\nThe default is value.\n\nconstraint_primal_start\n\nThis function controls the primal starting solution for the constraints. It is equivalent to calling set_start_value for each constraint, or setting the MOI.ConstraintPrimalStart attribute.\n\nIf it is a function, it must have the form constraint_primal_start(ci::ConstraintRef) that maps each constraint ci to the starting primal value.\n\nThe default is value.\n\nconstraint_dual_start\n\nThis function controls the dual starting solution for the constraints. It is equivalent to calling set_dual_start_value for each constraint, or setting the MOI.ConstraintDualStart attribute.\n\nIf it is a function, it must have the form constraint_dual_start(ci::ConstraintRef) that maps each constraint ci to the starting dual value.\n\nThe default is dual.\n\nnonlinear_dual_start\n\nThis function controls the dual starting solution for the nonlinear constraints It is equivalent to calling set_nonlinear_dual_start_value.\n\nIf it is a function, it must have the form nonlinear_dual_start(model::GenericModel) that returns a vector corresponding to the dual start of the constraints.\n\nThe default is nonlinear_dual_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_string_names_on_creation","page":"JuMP","title":"set_string_names_on_creation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_string_names_on_creation","category":"page"},{"location":"api/JuMP/#JuMP.set_string_names_on_creation","page":"JuMP","title":"JuMP.set_string_names_on_creation","text":"set_string_names_on_creation(model::GenericModel, value::Bool)\n\nSet the default argument of the set_string_name keyword in the @variable and @constraint macros to value. This is used to determine whether to assign String names to all variables and constraints in model.\n\nBy default, value is true. However, for larger models calling set_string_names_on_creation(model, false) can improve performance at the cost of reducing the readability of printing and solver log messages.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_time_limit_sec","page":"JuMP","title":"set_time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.set_time_limit_sec","page":"JuMP","title":"JuMP.set_time_limit_sec","text":"set_time_limit_sec(model::GenericModel, limit::Float64)\n\nSet the time limit (in seconds) of the solver.\n\nCan be unset using unset_time_limit_sec or with limit set to nothing.\n\nSee also: unset_time_limit_sec, time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_upper_bound","page":"JuMP","title":"set_upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.set_upper_bound","page":"JuMP","title":"JuMP.set_upper_bound","text":"set_upper_bound(v::GenericVariableRef, upper::Number)\n\nSet the upper bound of a variable. If one does not exist, create an upper bound constraint.\n\nSee also UpperBoundRef, has_upper_bound, upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> upper_bound(x)\n1.0\n\njulia> set_upper_bound(x, 2.0)\n\njulia> upper_bound(x)\n2.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#set_value","page":"JuMP","title":"set_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.set_value","category":"page"},{"location":"api/JuMP/#JuMP.set_value","page":"JuMP","title":"JuMP.set_value","text":"set_value(p::NonlinearParameter, v::Number)\n\nStore the value v in the nonlinear parameter p.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, p == 0)\np == 0.0\n\njulia> set_value(p, 5)\n5\n\njulia> value(p)\n5.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#shadow_price","page":"JuMP","title":"shadow_price","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.shadow_price","category":"page"},{"location":"api/JuMP/#JuMP.shadow_price","page":"JuMP","title":"JuMP.shadow_price","text":"shadow_price(con_ref::ConstraintRef)\n\nReturn the change in the objective from an infinitesimal relaxation of the constraint.\n\nThis value is computed from dual and can be queried only when has_duals is true and the objective sense is MIN_SENSE or MAX_SENSE (not FEASIBILITY_SENSE). For linear constraints, the shadow prices differ at most in sign from the dual value depending on the objective sense.\n\nSee also reduced_cost.\n\nNotes\n\nThe function simply translates signs from dual and does not validate the conditions needed to guarantee the sensitivity interpretation of the shadow price. The caller is responsible, e.g., for checking whether the solver converged to an optimal primal-dual pair or a proof of infeasibility.\nThe computation is based on the current objective sense of the model. If this has changed since the last solve, the results will be incorrect.\nRelaxation of equality constraints (and hence the shadow price) is defined based on which sense of the equality constraint is active.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#shape","page":"JuMP","title":"shape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.shape","category":"page"},{"location":"api/JuMP/#JuMP.shape","page":"JuMP","title":"JuMP.shape","text":"shape(c::AbstractConstraint)::AbstractShape\n\nReturn the shape of the constraint c.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_backend_summary","page":"JuMP","title":"show_backend_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_backend_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_backend_summary","page":"JuMP","title":"JuMP.show_backend_summary","text":"show_backend_summary(io::IO, model::GenericModel)\n\nPrint a summary of the optimizer backing model.\n\nAbstractModels should implement this method.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_constraints_summary","page":"JuMP","title":"show_constraints_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_constraints_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_constraints_summary","page":"JuMP","title":"JuMP.show_constraints_summary","text":"show_constraints_summary(io::IO, model::AbstractModel)\n\nWrite to io a summary of the number of constraints.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#show_objective_function_summary","page":"JuMP","title":"show_objective_function_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.show_objective_function_summary","category":"page"},{"location":"api/JuMP/#JuMP.show_objective_function_summary","page":"JuMP","title":"JuMP.show_objective_function_summary","text":"show_objective_function_summary(io::IO, model::AbstractModel)\n\nWrite to io a summary of the objective function type.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#simplex_iterations","page":"JuMP","title":"simplex_iterations","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.simplex_iterations","category":"page"},{"location":"api/JuMP/#JuMP.simplex_iterations","page":"JuMP","title":"JuMP.simplex_iterations","text":"simplex_iterations(model::GenericModel)\n\nGets the cumulative number of simplex iterations during the most-recent optimization.\n\nSolvers must implement MOI.SimplexIterations() to use this function.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solution_summary","page":"JuMP","title":"solution_summary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solution_summary","category":"page"},{"location":"api/JuMP/#JuMP.solution_summary","page":"JuMP","title":"JuMP.solution_summary","text":"solution_summary(model::GenericModel; result::Int = 1, verbose::Bool = false)\n\nReturn a struct that can be used print a summary of the solution in result result.\n\nIf verbose=true, write out the primal solution for every variable and the dual solution for every constraint, excluding those with empty names.\n\nExample\n\nWhen called at the REPL, the summary is automatically printed:\n\njulia> model = Model();\n\njulia> solution_summary(model)\n* Solver : No optimizer attached.\n\n* Status\n  Result count       : 0\n  Termination status : OPTIMIZE_NOT_CALLED\n  Message from the solver:\n  \"optimize not called\"\n\n* Candidate solution (result #1)\n  Primal status      : NO_SOLUTION\n  Dual status        : NO_SOLUTION\n\n* Work counters\n\nUse print to force the printing of the summary from inside a function:\n\njulia> model = Model();\n\njulia> function foo(model)\n           print(solution_summary(model))\n           return\n       end\nfoo (generic function with 1 method)\n\njulia> foo(model)\n* Solver : No optimizer attached.\n\n* Status\n  Result count       : 0\n  Termination status : OPTIMIZE_NOT_CALLED\n  Message from the solver:\n  \"optimize not called\"\n\n* Candidate solution (result #1)\n  Primal status      : NO_SOLUTION\n  Dual status        : NO_SOLUTION\n\n* Work counters\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solve_time","page":"JuMP","title":"solve_time","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solve_time","category":"page"},{"location":"api/JuMP/#JuMP.solve_time","page":"JuMP","title":"JuMP.solve_time","text":"solve_time(model::GenericModel)\n\nIf available, returns the solve time reported by the solver. Returns \"ArgumentError: ModelLike of type Solver.Optimizer does not support accessing the attribute MathOptInterface.SolveTimeSec()\" if the attribute is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#solver_name","page":"JuMP","title":"solver_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.solver_name","category":"page"},{"location":"api/JuMP/#JuMP.solver_name","page":"JuMP","title":"JuMP.solver_name","text":"solver_name(model::GenericModel)\n\nIf available, returns the SolverName property of the underlying optimizer.\n\nReturns \"No optimizer attached\" in AUTOMATIC or MANUAL modes when no optimizer is attached.\n\nReturns \"SolverName() attribute not implemented by the optimizer.\" if the attribute is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#start_value","page":"JuMP","title":"start_value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.start_value","category":"page"},{"location":"api/JuMP/#JuMP.start_value","page":"JuMP","title":"JuMP.start_value","text":"start_value(con_ref::ConstraintRef)\n\nReturn the primal start value (MOI.ConstraintPrimalStart) of the constraint con_ref.\n\nNote: If no primal start value has been set, start_value will return nothing.\n\nSee also set_start_value.\n\n\n\n\n\nstart_value(v::GenericVariableRef)\n\nReturn the start value (MOI attribute VariablePrimalStart) of the variable v.\n\nNote: VariablePrimalStarts are sometimes called \"MIP-starts\" or \"warmstarts\".\n\nSee also set_start_value.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#termination_status","page":"JuMP","title":"termination_status","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.termination_status","category":"page"},{"location":"api/JuMP/#JuMP.termination_status","page":"JuMP","title":"JuMP.termination_status","text":"termination_status(model::GenericModel)\n\nReturn a MOI.TerminationStatusCode describing why the solver stopped (i.e., the MOI.TerminationStatus attribute).\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#time_limit_sec","page":"JuMP","title":"time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.time_limit_sec","page":"JuMP","title":"JuMP.time_limit_sec","text":"time_limit_sec(model::GenericModel)\n\nReturn the time limit (in seconds) of the model.\n\nReturns nothing if unset.\n\nSee also: set_time_limit_sec, unset_time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#triangle_vec","page":"JuMP","title":"triangle_vec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.triangle_vec","category":"page"},{"location":"api/JuMP/#JuMP.triangle_vec","page":"JuMP","title":"JuMP.triangle_vec","text":"triangle_vec(matrix::Matrix)\n\nReturn the upper triangle of a matrix concatenated into a vector in the order required by JuMP and MathOptInterface for Triangle sets.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, X[1:3, 1:3], Symmetric);\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(3))\n[t, X[1,1], X[1,2], X[2,2], X[1,3], X[2,3], X[3,3]] ∈ MathOptInterface.RootDetConeTriangle(3)\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unfix","page":"JuMP","title":"unfix","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unfix","category":"page"},{"location":"api/JuMP/#JuMP.unfix","page":"JuMP","title":"JuMP.unfix","text":"unfix(v::GenericVariableRef)\n\nDelete the fixing constraint of a variable.\n\nError if one does not exist.\n\nSee also FixRef, is_fixed, fix_value, fix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> is_fixed(x)\ntrue\n\njulia> unfix(x)\n\njulia> is_fixed(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unregister","page":"JuMP","title":"unregister","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unregister","category":"page"},{"location":"api/JuMP/#JuMP.unregister","page":"JuMP","title":"JuMP.unregister","text":"unregister(model::GenericModel, key::Symbol)\n\nUnregister the name key from model so that a new variable, constraint, or expression can be created with the same key.\n\nNote that this will not delete the object model[key]; it will just remove the reference at model[key]. To delete the object, use delete as well.\n\nSee also: delete, object_dictionary.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n\nStacktrace:\n[...]\n\njulia> num_variables(model)\n1\n\njulia> unregister(model, :x)\n\njulia> @variable(model, x)\nx\n\njulia> num_variables(model)\n2\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unsafe_backend","page":"JuMP","title":"unsafe_backend","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unsafe_backend","category":"page"},{"location":"api/JuMP/#JuMP.unsafe_backend","page":"JuMP","title":"JuMP.unsafe_backend","text":"unsafe_backend(model::GenericModel)\n\nReturn the innermost optimizer associated with the JuMP model model.\n\nThis function should only be used by advanced users looking to access low-level solver-specific functionality. It has a high-risk of incorrect usage. We strongly suggest you use the alternative suggested below.\n\nSee also: backend.\n\nUnsafe behavior\n\nThis function is unsafe for two main reasons.\n\nFirst, the formulation and order of variables and constraints in the unsafe backend may be different to the variables and constraints in model. This can happen because of bridges, or because the solver requires the variables or constraints in a specific order. In addition, the variable or constraint index returned by index at the JuMP level may be different to the index of the corresponding variable or constraint in the unsafe_backend. There is no solution to this. Use the alternative suggested below instead.\n\nSecond, the unsafe_backend may be empty, or lack some modifications made to the JuMP model. Thus, before calling unsafe_backend you should first call MOI.Utilities.attach_optimizer to ensure that the backend is synchronized with the JuMP model.\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: HiGHS\n\njulia> MOI.Utilities.attach_optimizer(model)\n\njulia> inner = unsafe_backend(model)\nA HiGHS model with 0 columns and 0 rows.\n\nMoreover, if you modify the JuMP model, the reference you have to the backend (i.e., inner in the example above) may be out-dated, and you should call MOI.Utilities.attach_optimizer again.\n\nThis function is also unsafe in the reverse direction: if you modify the unsafe backend, e.g., by adding a new constraint to inner, the changes may be silently discarded by JuMP when the JuMP model is modified or solved.\n\nAlternative\n\nInstead of unsafe_backend, create a model using direct_model and call backend instead.\n\nFor example, instead of:\n\njulia> import HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> MOI.Utilities.attach_optimizer(model)\n\njulia> highs = unsafe_backend(model)\nA HiGHS model with 1 columns and 0 rows.\n\nUse:\n\njulia> import HiGHS\n\njulia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> highs = backend(model)  # No need to call `attach_optimizer`.\nA HiGHS model with 1 columns and 0 rows.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_binary","page":"JuMP","title":"unset_binary","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_binary","category":"page"},{"location":"api/JuMP/#JuMP.unset_binary","page":"JuMP","title":"JuMP.unset_binary","text":"unset_binary(variable_ref::GenericVariableRef)\n\nRemove the binary constraint on the variable variable_ref.\n\nSee also BinaryRef, is_binary, set_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> is_binary(x)\ntrue\n\njulia> unset_binary(x)\n\njulia> is_binary(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_integer","page":"JuMP","title":"unset_integer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_integer","category":"page"},{"location":"api/JuMP/#JuMP.unset_integer","page":"JuMP","title":"JuMP.unset_integer","text":"unset_integer(variable_ref::GenericVariableRef)\n\nRemove the integrality constraint on the variable variable_ref.\n\nErrors if one does not exist.\n\nSee also IntegerRef, is_integer, set_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Int);\n\njulia> is_integer(x)\ntrue\n\njulia> unset_integer(x)\n\njulia> is_integer(x)\nfalse\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_silent","page":"JuMP","title":"unset_silent","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_silent","category":"page"},{"location":"api/JuMP/#JuMP.unset_silent","page":"JuMP","title":"JuMP.unset_silent","text":"unset_silent(model::GenericModel)\n\nNeutralize the effect of the set_silent function and let the solver attributes control the verbosity.\n\nSee also: set_silent.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#unset_time_limit_sec","page":"JuMP","title":"unset_time_limit_sec","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.unset_time_limit_sec","category":"page"},{"location":"api/JuMP/#JuMP.unset_time_limit_sec","page":"JuMP","title":"JuMP.unset_time_limit_sec","text":"unset_time_limit_sec(model::GenericModel)\n\nUnset the time limit of the solver.\n\nSee also: set_time_limit_sec, time_limit_sec.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#upper_bound","page":"JuMP","title":"upper_bound","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.upper_bound","category":"page"},{"location":"api/JuMP/#JuMP.upper_bound","page":"JuMP","title":"JuMP.upper_bound","text":"upper_bound(v::GenericVariableRef)\n\nReturn the upper bound of a variable.\n\nError if one does not exist.\n\nSee also UpperBoundRef, has_upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> upper_bound(x)\n1.0\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#value","page":"JuMP","title":"value","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.value","category":"page"},{"location":"api/JuMP/#JuMP.value","page":"JuMP","title":"JuMP.value","text":"value(con_ref::ConstraintRef; result::Int = 1)\n\nReturn the primal value of constraint con_ref associated with result index result of the most-recent solution returned by the solver.\n\nThat is, if con_ref is the reference of a constraint func-in-set, it returns the value of func evaluated at the value of the variables (given by value(::GenericVariableRef)).\n\nUse has_values to check if a result exists before asking for values.\n\nSee also: result_count.\n\nNote\n\nFor scalar constraints, the constant is moved to the set so it is not taken into account in the primal value of the constraint. For instance, the constraint @constraint(model, 2x + 3y + 1 == 5) is transformed into 2x + 3y-in-MOI.EqualTo(4) so the value returned by this function is the evaluation of 2x + 3y.\n\n\n\n\n\nvalue(var_value::Function, con_ref::ConstraintRef)\n\nEvaluate the primal value of the constraint con_ref using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericVariableRef; result = 1)\n\nReturn the value of variable v associated with result index result of the most-recent returned by the solver.\n\nUse has_values to check if a result exists before asking for values.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, v::GenericVariableRef)\n\nEvaluate the value of the variable v as var_value(v).\n\n\n\n\n\nvalue(var_value::Function, ex::GenericAffExpr)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericAffExpr; result::Int = 1)\n\nReturn the value of the GenericAffExpr v associated with result index result of the most-recent solution returned by the solver.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, ex::GenericQuadExpr)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(v::GenericQuadExpr; result::Int = 1)\n\nReturn the value of the GenericQuadExpr v associated with result index result of the most-recent solution returned by the solver.\n\nReplaces getvalue for most use cases.\n\nSee also: result_count.\n\n\n\n\n\nvalue(p::NonlinearParameter)\n\nReturn the current value stored in the nonlinear parameter p.\n\nExample\n\njulia> model = Model();\n\njulia> @NLparameter(model, p == 10)\np == 10.0\n\njulia> value(p)\n10.0\n\n\n\n\n\nvalue(ex::NonlinearExpression; result::Int = 1)\n\nReturn the value of the NonlinearExpression ex associated with result index result of the most-recent solution returned by the solver.\n\nReplaces getvalue for most use cases.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, ex::NonlinearExpression)\n\nEvaluate ex using var_value(v) as the value for each variable v.\n\n\n\n\n\nvalue(c::NonlinearConstraintRef; result::Int = 1)\n\nReturn the value of the NonlinearConstraintRef c associated with result index result of the most-recent solution returned by the solver.\n\nSee also: result_count.\n\n\n\n\n\nvalue(var_value::Function, c::NonlinearConstraintRef)\n\nEvaluate c using var_value(v) as the value for each variable v.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#value_type","page":"JuMP","title":"value_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.value_type","category":"page"},{"location":"api/JuMP/#JuMP.value_type","page":"JuMP","title":"JuMP.value_type","text":"value_type(::Type{<:Union{AbstractModel,AbstractVariableRef}})\n\nReturn the return type of value for variables of that model. It defaults to Float64 if it is not implemented.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#variable_by_name","page":"JuMP","title":"variable_by_name","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.variable_by_name","category":"page"},{"location":"api/JuMP/#JuMP.variable_by_name","page":"JuMP","title":"JuMP.variable_by_name","text":"variable_by_name(\n    model::AbstractModel,\n    name::String,\n)::Union{AbstractVariableRef,Nothing}\n\nReturns the reference of the variable with name attribute name or Nothing if no variable has this name attribute. Throws an error if several variables have name as their name attribute.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> variable_by_name(model, \"x\")\nx\n\njulia> @variable(model, base_name=\"x\")\nx\n\njulia> variable_by_name(model, \"x\")\nERROR: Multiple variables have the name x.\nStacktrace:\n [1] error(::String) at ./error.jl:33\n [2] get(::MOIU.Model{Float64}, ::Type{MathOptInterface.VariableIndex}, ::String) at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/model.jl:222\n [3] get at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/universalfallback.jl:201 [inlined]\n [4] get(::MathOptInterface.Utilities.CachingOptimizer{MathOptInterface.AbstractOptimizer,MathOptInterface.Utilities.UniversalFallback{MOIU.Model{Float64}}}, ::Type{MathOptInterface.VariableIndex}, ::String) at /home/blegat/.julia/dev/MathOptInterface/src/Utilities/cachingoptimizer.jl:490\n [5] variable_by_name(::GenericModel, ::String) at /home/blegat/.julia/dev/JuMP/src/variables.jl:268\n [6] top-level scope at none:0\n\njulia> var = @variable(model, base_name=\"y\")\ny\n\njulia> variable_by_name(model, \"y\")\ny\n\njulia> set_name(var, \"z\")\n\njulia> variable_by_name(model, \"y\")\n\njulia> variable_by_name(model, \"z\")\nz\n\njulia> @variable(model, u[1:2])\n2-element Vector{VariableRef}:\n u[1]\n u[2]\n\njulia> variable_by_name(model, \"u[2]\")\nu[2]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#variable_ref_type","page":"JuMP","title":"variable_ref_type","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.variable_ref_type","category":"page"},{"location":"api/JuMP/#JuMP.variable_ref_type","page":"JuMP","title":"JuMP.variable_ref_type","text":"variable_ref_type(::Union{F,Type{F}}) where {F}\n\nA helper function used internally by JuMP and some JuMP extensions. Returns the variable type associated with the model or expression type F.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#vectorize","page":"JuMP","title":"vectorize","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.vectorize","category":"page"},{"location":"api/JuMP/#JuMP.vectorize","page":"JuMP","title":"JuMP.vectorize","text":"vectorize(matrix::AbstractMatrix, ::Shape)\n\nConvert the matrix into a vector according to Shape.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#write_to_file","page":"JuMP","title":"write_to_file","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.write_to_file","category":"page"},{"location":"api/JuMP/#JuMP.write_to_file","page":"JuMP","title":"JuMP.write_to_file","text":"write_to_file(\n    model::GenericModel,\n    filename::String;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_AUTOMATIC,\n    kwargs...,\n)\n\nWrite the JuMP model model to filename in the format format.\n\nIf the filename ends in .gz, it will be compressed using Gzip. If the filename ends in .bz2, it will be compressed using BZip2.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#AbstractConstraint","page":"JuMP","title":"AbstractConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractConstraint","category":"page"},{"location":"api/JuMP/#JuMP.AbstractConstraint","page":"JuMP","title":"JuMP.AbstractConstraint","text":"abstract type AbstractConstraint\n\nAn abstract base type for all constraint types. AbstractConstraints store the function and set directly, unlike ConstraintRefs that are merely references to constraints stored in a model. AbstractConstraints do not need to be attached to a model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractJuMPScalar","page":"JuMP","title":"AbstractJuMPScalar","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractJuMPScalar","category":"page"},{"location":"api/JuMP/#JuMP.AbstractJuMPScalar","page":"JuMP","title":"JuMP.AbstractJuMPScalar","text":"AbstractJuMPScalar <: MutableArithmetics.AbstractMutable\n\nAbstract base type for all scalar types\n\nThe subtyping of AbstractMutable will allow calls of some Base functions to be redirected to a method in MA that handles type promotion more carefully (e.g. the promotion in sparse matrix products in SparseArrays usually does not work for JuMP types) and exploits the mutability of AffExpr and QuadExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractModel","page":"JuMP","title":"AbstractModel","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractModel","category":"page"},{"location":"api/JuMP/#JuMP.AbstractModel","page":"JuMP","title":"JuMP.AbstractModel","text":"AbstractModel\n\nAn abstract type that should be subtyped for users creating JuMP extensions.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractScalarSet","page":"JuMP","title":"AbstractScalarSet","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractScalarSet","category":"page"},{"location":"api/JuMP/#JuMP.AbstractScalarSet","page":"JuMP","title":"JuMP.AbstractScalarSet","text":"AbstractScalarSet\n\nAn abstract type for defining new scalar sets in JuMP.\n\nImplement moi_set(::AbstractScalarSet) to convert the type into an MOI set.\n\nSee also: moi_set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractShape","page":"JuMP","title":"AbstractShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractShape","category":"page"},{"location":"api/JuMP/#JuMP.AbstractShape","page":"JuMP","title":"JuMP.AbstractShape","text":"AbstractShape\n\nAbstract vectorizable shape. Given a flat vector form of an object of shape shape, the original object can be obtained by reshape_vector.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVariable","page":"JuMP","title":"AbstractVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVariable","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVariable","page":"JuMP","title":"JuMP.AbstractVariable","text":"AbstractVariable\n\nVariable returned by build_variable. It represents a variable that has not been added yet to any model. It can be added to a given model with add_variable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVariableRef","page":"JuMP","title":"AbstractVariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVariableRef","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVariableRef","page":"JuMP","title":"JuMP.AbstractVariableRef","text":"AbstractVariableRef\n\nVariable returned by add_variable. Affine (resp. quadratic) operations with variables of type V<:AbstractVariableRef and coefficients of type T     create a GenericAffExpr{T,V} (resp. GenericQuadExpr{T,V}).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AbstractVectorSet","page":"JuMP","title":"AbstractVectorSet","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AbstractVectorSet","category":"page"},{"location":"api/JuMP/#JuMP.AbstractVectorSet","page":"JuMP","title":"JuMP.AbstractVectorSet","text":"AbstractVectorSet\n\nAn abstract type for defining new sets in JuMP.\n\nImplement moi_set(::AbstractVectorSet, dim::Int) to convert the type into an MOI set.\n\nSee also: moi_set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#AffExpr","page":"JuMP","title":"AffExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AffExpr","category":"page"},{"location":"api/JuMP/#JuMP.AffExpr","page":"JuMP","title":"JuMP.AffExpr","text":"AffExpr\n\nAlias for GenericAffExpr{Float64,VariableRef}, the specific GenericAffExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#BinaryRef","page":"JuMP","title":"BinaryRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.BinaryRef","category":"page"},{"location":"api/JuMP/#JuMP.BinaryRef","page":"JuMP","title":"JuMP.BinaryRef","text":"BinaryRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining v to be binary. Errors if one does not exist.\n\nSee also is_binary, set_binary, unset_binary.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Bin);\n\njulia> BinaryRef(x)\nx binary\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#BridgeableConstraint","page":"JuMP","title":"BridgeableConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.BridgeableConstraint","category":"page"},{"location":"api/JuMP/#JuMP.BridgeableConstraint","page":"JuMP","title":"JuMP.BridgeableConstraint","text":"BridgeableConstraint(\n    constraint::C,\n    bridge_type::B;\n    coefficient_type::Type{T} = Float64,\n) where {C<:AbstractConstraint,B<:Type{<:MOI.Bridges.AbstractBridge},T}\n\nAn AbstractConstraint representinng that constraint that can be bridged by the bridge of type bridge_type{coefficient_type}.\n\nAdding a BridgeableConstraint to a model is equivalent to:\n\nadd_bridge(model, bridge_type; coefficient_type = coefficient_type)\nadd_constraint(model, constraint)\n\nExample\n\nGiven a new scalar set type CustomSet with a bridge CustomBridge that can bridge F-in-CustomSet constraints, when the user does:\n\nmodel = Model()\n@variable(model, x)\n@constraint(model, x + 1 in CustomSet())\noptimize!(model)\n\nwith an optimizer that does not support F-in-CustomSet constraints, the constraint will not be bridged unless they first call add_bridge(model, CustomBridge).\n\nIn order to automatically add the CustomBridge to any model to which an F-in-CustomSet is added, add the following method:\n\nfunction JuMP.build_constraint(\n    _error::Function,\n    func::AbstractJuMPScalar,\n    set::CustomSet,\n)\n    constraint = ScalarConstraint(func, set)\n    return BridgeableConstraint(constraint, CustomBridge)\nend\n\nNote\n\nJuMP extensions should extend JuMP.build_constraint only if they also defined CustomSet, for three reasons:\n\nIt is problematic if multiple extensions overload the same JuMP method.\nA missing method will not inform the users that they forgot to load the extension module defining the build_constraint method.\nDefining a method where neither the function nor any of the argument types are defined in the package is called type piracy and is discouraged in the Julia style guide.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ComplexPlane","page":"JuMP","title":"ComplexPlane","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ComplexPlane","category":"page"},{"location":"api/JuMP/#JuMP.ComplexPlane","page":"JuMP","title":"JuMP.ComplexPlane","text":"ComplexPlane\n\nComplex plane object that can be used to create a complex variable in the @variable macro.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, x in ComplexPlane())\nreal(x) + imag(x) im\n\njulia> all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\nWe see in the output of the last command that two real variables were created. The Julia variable x binds to an affine expression in terms of these two variables that parametrize the complex plane.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ComplexVariable","page":"JuMP","title":"ComplexVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ComplexVariable","category":"page"},{"location":"api/JuMP/#JuMP.ComplexVariable","page":"JuMP","title":"JuMP.ComplexVariable","text":"ComplexVariable{S,T,U,V} <: AbstractVariable\n\nA struct used when adding complex variables.\n\nSee also: ComplexPlane.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ConstraintNotOwned","page":"JuMP","title":"ConstraintNotOwned","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ConstraintNotOwned","category":"page"},{"location":"api/JuMP/#JuMP.ConstraintNotOwned","page":"JuMP","title":"JuMP.ConstraintNotOwned","text":"struct ConstraintNotOwned{C <: ConstraintRef} <: Exception\n    constraint_ref::C\nend\n\nThe constraint constraint_ref was used in a model different to owner_model(constraint_ref).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ConstraintRef","page":"JuMP","title":"ConstraintRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ConstraintRef","category":"page"},{"location":"api/JuMP/#JuMP.ConstraintRef","page":"JuMP","title":"JuMP.ConstraintRef","text":"ConstraintRef\n\nHolds a reference to the model and the corresponding MOI.ConstraintIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#FixRef","page":"JuMP","title":"FixRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FixRef","category":"page"},{"location":"api/JuMP/#JuMP.FixRef","page":"JuMP","title":"JuMP.FixRef","text":"FixRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint fixing the value of v.\n\nErrors if one does not exist.\n\nSee also is_fixed, fix_value, fix, unfix.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x == 1);\n\njulia> FixRef(x)\nx = 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#GenericAffExpr","page":"JuMP","title":"GenericAffExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericAffExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericAffExpr","page":"JuMP","title":"JuMP.GenericAffExpr","text":"mutable struct GenericAffExpr{CoefType,VarType} <: AbstractJuMPScalar\n    constant::CoefType\n    terms::OrderedDict{VarType,CoefType}\nend\n\nAn expression type representing an affine expression of the form: sum a_i x_i + c.\n\nFields\n\n.constant: the constant c in the expression.\n.terms: an OrderedDict, with keys of VarType and values of CoefType describing the sparse vector a.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericModel","page":"JuMP","title":"GenericModel","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericModel","category":"page"},{"location":"api/JuMP/#JuMP.GenericModel","page":"JuMP","title":"JuMP.GenericModel","text":"GenericModel{T}(\n    [optimizer_factory;]\n    add_bridges::Bool = true,\n) where {T<:Real}\n\nCreate a new instance of a JuMP model.\n\nIf optimizer_factory is provided, the model is initialized with the optimizer returned by MOI.instantiate(optimizer_factory).\n\nIf optimizer_factory is not provided, use set_optimizer to set the optimizer before calling optimize!.\n\nIf add_bridges, JuMP adds a MOI.Bridges.LazyBridgeOptimizer to automatically reformulate the problem into a form supported by the optimizer.\n\nValue type T\n\nPassing a type other than Float64 as the value type T is an advanced operation. The value type must match that expected by the chosen optimizer. Consult the optimizers documentation for details.\n\nIf not documented, assume that the optimizer supports only Float64.\n\nChoosing an unsupported value type will throw an MOI.UnsupportedConstraint or an MOI.UnsupportedAttribute error, the timing of which (during the model construction or during a call to optimize!) depends on how the solver is interfaced to JuMP.\n\nExample\n\njulia> model = GenericModel{BigFloat}();\n\njulia> typeof(model)\nGenericModel{BigFloat}\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericNonlinearExpr","page":"JuMP","title":"GenericNonlinearExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericNonlinearExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericNonlinearExpr","page":"JuMP","title":"JuMP.GenericNonlinearExpr","text":"GenericNonlinearExpr{V}(head::Symbol, args::Vector{Any})\nGenericNonlinearExpr{V}(head::Symbol, args::Any...)\n\nThe scalar-valued nonlinear function head(args...), represented as a symbolic expression tree, with the call operator head and ordered arguments in args.\n\nV is the type of AbstractVariableRef present in the expression, and is used to help dispatch JuMP extensions.\n\nhead\n\nThe head::Symbol must be an operator supported by the model.\n\nThe default list of supported univariate operators is given by:\n\nMOI.Nonlinear.DEFAULT_UNIVARIATE_OPERATORS\n\nand the default list of supported multivariate operators is given by:\n\nMOI.Nonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n\nAdditional operators can be add using @operator.\n\nSee the full list of operators supported by a MOI.ModelLike by querying the MOI.ListOfSupportedNonlinearOperators attribute.\n\nargs\n\nThe vector args contains the arguments to the nonlinear function. If the operator is univariate, it must contain one element. Otherwise, it may contain multiple elements.\n\nGiven a subtype of AbstractVariableRef, V, for GenericNonlinearExpr{V}, each element must be one of the following:\n\nA constant value of type <:Real\nA V\nA GenericAffExpr{T,V}\nA GenericQuadExpr{T,V}\nA GenericNonlinearExpr{V}\n\nwhere T<:Real and T == value_type(V).\n\nUnsupported operators\n\nIf the optimizer does not support head, an MOI.UnsupportedNonlinearOperator error will be thrown.\n\nThere is no guarantee about when this error will be thrown; it may be thrown when the function is first added to the model, or it may be thrown when optimize! is called.\n\nExample\n\nTo represent the function f(x) = sin(x)^2, do:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f = sin(x)^2\nsin(x) ^ 2.0\n\njulia> f = GenericNonlinearExpr{VariableRef}(\n           :^,\n           GenericNonlinearExpr{VariableRef}(:sin, x),\n           2.0,\n       )\nsin(x) ^ 2.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericQuadExpr","page":"JuMP","title":"GenericQuadExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericQuadExpr","category":"page"},{"location":"api/JuMP/#JuMP.GenericQuadExpr","page":"JuMP","title":"JuMP.GenericQuadExpr","text":"mutable struct GenericQuadExpr{CoefType,VarType} <: AbstractJuMPScalar\n    aff::GenericAffExpr{CoefType,VarType}\n    terms::OrderedDict{UnorderedPair{VarType}, CoefType}\nend\n\nAn expression type representing an quadratic expression of the form: sum q_ij x_i x_j + sum a_i x_i + c.\n\nFields\n\n.aff: an GenericAffExpr representing the affine portion of the expression.\n.terms: an OrderedDict, with keys of UnorderedPair{VarType} and values of CoefType, describing the sparse list of terms q.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericReferenceMap","page":"JuMP","title":"GenericReferenceMap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericReferenceMap","category":"page"},{"location":"api/JuMP/#JuMP.GenericReferenceMap","page":"JuMP","title":"JuMP.GenericReferenceMap","text":"GenericReferenceMap{T}\n\nMapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#GenericVariableRef","page":"JuMP","title":"GenericVariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.GenericVariableRef","category":"page"},{"location":"api/JuMP/#JuMP.GenericVariableRef","page":"JuMP","title":"JuMP.GenericVariableRef","text":"GenericVariableRef{T} <: AbstractVariableRef\n\nHolds a reference to the model and the corresponding MOI.VariableIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianMatrixShape","page":"JuMP","title":"HermitianMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.HermitianMatrixShape","page":"JuMP","title":"JuMP.HermitianMatrixShape","text":"HermitianMatrixShape\n\nShape object for a Hermitian square matrix of side_dimension rows and columns. The vectorized form corresponds to MOI.HermitianPositiveSemidefiniteConeTriangle.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianMatrixSpace","page":"JuMP","title":"HermitianMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.HermitianMatrixSpace","page":"JuMP","title":"JuMP.HermitianMatrixSpace","text":"HermitianMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be hermitian.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(Q[1,1])                    real(Q[1,2]) + imag(Q[1,2]) im\n real(Q[1,2]) - imag(Q[1,2]) im  real(Q[2,2])\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#HermitianPSDCone","page":"JuMP","title":"HermitianPSDCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.HermitianPSDCone","category":"page"},{"location":"api/JuMP/#JuMP.HermitianPSDCone","page":"JuMP","title":"JuMP.HermitianPSDCone","text":"HermitianPSDCone\n\nHermitian positive semidefinite cone object that can be used to create a Hermitian positive semidefinite square matrix in the @variable and @constraint macros.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, H[1:3, 1:3] in HermitianPSDCone())\n3×3 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    …  real(H[1,3]) + imag(H[1,3]) im\n real(H[1,2]) - imag(H[1,2]) im     real(H[2,3]) + imag(H[2,3]) im\n real(H[1,3]) - imag(H[1,3]) im     real(H[3,3])\n\njulia> all_variables(model)\n9-element Vector{VariableRef}:\n real(H[1,1])\n real(H[1,2])\n real(H[2,2])\n real(H[1,3])\n real(H[2,3])\n real(H[3,3])\n imag(H[1,2])\n imag(H[1,3])\n imag(H[2,3])\n\njulia> all_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle)\n1-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.HermitianPositiveSemidefiniteConeTriangle}}}:\n [real(H[1,1]), real(H[1,2]), real(H[2,2]), real(H[1,3]), real(H[2,3]), real(H[3,3]), imag(H[1,2]), imag(H[1,3]), imag(H[2,3])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(3)\n\nWe see in the output of the last commands that 9 real variables were created. The matrix H contrains affine expressions in terms of these 9 variables that parametrize a Hermitian matrix.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#IntegerRef","page":"JuMP","title":"IntegerRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.IntegerRef","category":"page"},{"location":"api/JuMP/#JuMP.IntegerRef","page":"JuMP","title":"JuMP.IntegerRef","text":"IntegerRef(v::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining v to be integer.\n\nErrors if one does not exist.\n\nSee also is_integer, set_integer, unset_integer.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x, Int);\n\njulia> IntegerRef(x)\nx integer\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#LinearTermIterator","page":"JuMP","title":"LinearTermIterator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LinearTermIterator","category":"page"},{"location":"api/JuMP/#JuMP.LinearTermIterator","page":"JuMP","title":"JuMP.LinearTermIterator","text":"LinearTermIterator{GAE<:GenericAffExpr}\n\nA struct that implements the iterate protocol in order to iterate over tuples of (coefficient, variable) in the GenericAffExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#LowerBoundRef","page":"JuMP","title":"LowerBoundRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LowerBoundRef","category":"page"},{"location":"api/JuMP/#JuMP.LowerBoundRef","page":"JuMP","title":"JuMP.LowerBoundRef","text":"LowerBoundRef(v::GenericVariableRef)\n\nReturn a constraint reference to the lower bound constraint of v.\n\nErrors if one does not exist.\n\nSee also has_lower_bound, lower_bound, set_lower_bound, delete_lower_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x >= 1.0);\n\njulia> LowerBoundRef(x)\nx ≥ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#Model","page":"JuMP","title":"Model","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Model","category":"page"},{"location":"api/JuMP/#JuMP.Model","page":"JuMP","title":"JuMP.Model","text":"Model([optimizer_factory;] add_bridges::Bool = true)\n\nCreate a new instance of a JuMP model.\n\nIf optimizer_factory is provided, the model is initialized with thhe optimizer returned by MOI.instantiate(optimizer_factory).\n\nIf optimizer_factory is not provided, use set_optimizer to set the optimizer before calling optimize!.\n\nIf add_bridges, JuMP adds a MOI.Bridges.LazyBridgeOptimizer to automatically reformulate the problem into a form supported by the optimizer.\n\nExample\n\njulia> import Ipopt\n\njulia> model = Model(Ipopt.Optimizer);\n\njulia> solver_name(model)\n\"Ipopt\"\n\njulia> import HiGHS\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer); add_bridges = false);\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ModelMode","page":"JuMP","title":"ModelMode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ModelMode","category":"page"},{"location":"api/JuMP/#JuMP.ModelMode","page":"JuMP","title":"JuMP.ModelMode","text":"ModelMode\n\nAn enum to describe the state of the CachingOptimizer inside a JuMP model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NLPEvaluator","page":"JuMP","title":"NLPEvaluator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NLPEvaluator","category":"page"},{"location":"api/JuMP/#JuMP.NLPEvaluator","page":"JuMP","title":"JuMP.NLPEvaluator","text":"NLPEvaluator(\n    model::Model,\n    _differentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation =\n        MOI.Nonlinear.SparseReverseMode(),\n)\n\nReturn an MOI.AbstractNLPEvaluator constructed from model\n\nwarning: Warning\nBefore using, you must initialize the evaluator using MOI.initialize.\n\nExperimental\n\nThese features may change or be removed in any future version of JuMP.\n\nPass _differentiation_backend to specify the differentiation backend used to compute derivatives.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#NoOptimizer","page":"JuMP","title":"NoOptimizer","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NoOptimizer","category":"page"},{"location":"api/JuMP/#JuMP.NoOptimizer","page":"JuMP","title":"JuMP.NoOptimizer","text":"struct NoOptimizer <: Exception end\n\nNo optimizer is set. The optimizer can be provided to the Model constructor or by calling set_optimizer.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearConstraintIndex","page":"JuMP","title":"NonlinearConstraintIndex","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearConstraintIndex","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearConstraintIndex","page":"JuMP","title":"JuMP.NonlinearConstraintIndex","text":"ConstraintIndex\n\nAn index to a nonlinear constraint that is returned by add_constraint.\n\nGiven data::Model and c::ConstraintIndex, use data[c] to retrieve the corresponding Constraint.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearConstraintRef","page":"JuMP","title":"NonlinearConstraintRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearConstraintRef","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearConstraintRef","page":"JuMP","title":"JuMP.NonlinearConstraintRef","text":"NonlinearConstraintRef\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearExpr","page":"JuMP","title":"NonlinearExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearExpr","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearExpr","page":"JuMP","title":"JuMP.NonlinearExpr","text":"NonlinearExpr\n\nAlias for GenericNonlinearExpr{VariableRef}, the specific GenericNonlinearExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearExpression","page":"JuMP","title":"NonlinearExpression","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearExpression","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearExpression","page":"JuMP","title":"JuMP.NonlinearExpression","text":"NonlinearExpression <: AbstractJuMPScalar\n\nA struct to represent a nonlinear expression.\n\nCreate an expression using @NLexpression.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearOperator","page":"JuMP","title":"NonlinearOperator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearOperator","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearOperator","page":"JuMP","title":"JuMP.NonlinearOperator","text":"NonlinearOperator(func::Function, head::Symbol)\n\nA callable struct (functor) representing a function named head.\n\nWhen called with AbstractJuMPScalars, the struct returns a GenericNonlinearExpr.\n\nWhen called with non-JuMP types, the struct returns the evaluation of func(args...).\n\nUnless head is special-cased by the optimizer, the operator must have already been added to the model using add_nonlinear_operator or @operator.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> f(x::Float64) = x^2\nf (generic function with 1 method)\n\njulia> ∇f(x::Float64) = 2 * x\n∇f (generic function with 1 method)\n\njulia> ∇²f(x::Float64) = 2.0\n∇²f (generic function with 1 method)\n\njulia> @operator(model, op_f, 1, f, ∇f, ∇²f)\nNonlinearOperator(f, :op_f)\n\njulia> bar = NonlinearOperator(f, :op_f)\nNonlinearOperator(f, :op_f)\n\njulia> @objective(model, Min, bar(x))\nop_f(x)\n\njulia> bar(2.0)\n4.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#NonlinearParameter","page":"JuMP","title":"NonlinearParameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NonlinearParameter","category":"page"},{"location":"api/JuMP/#JuMP.NonlinearParameter","page":"JuMP","title":"JuMP.NonlinearParameter","text":"NonlinearParameter <: AbstractJuMPScalar\n\nA struct to represent a nonlinear parameter.\n\nCreate a parameter using @NLparameter.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Nonnegatives","page":"JuMP","title":"Nonnegatives","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Nonnegatives","category":"page"},{"location":"api/JuMP/#JuMP.Nonnegatives","page":"JuMP","title":"JuMP.Nonnegatives","text":"Nonnegatives()\n\nThe JuMP equivalent of the MOI.Nonnegatives set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Nonnegatives())\n[x[1], x[2]] ∈ MathOptInterface.Nonnegatives(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x >= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonnegatives(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Nonpositives","page":"JuMP","title":"Nonpositives","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Nonpositives","category":"page"},{"location":"api/JuMP/#JuMP.Nonpositives","page":"JuMP","title":"JuMP.Nonpositives","text":"Nonpositives()\n\nThe JuMP equivalent of the MOI.Nonpositives set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Nonpositives())\n[x[1], x[2]] ∈ MathOptInterface.Nonpositives(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x <= b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Nonpositives(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#OptimizationSense","page":"JuMP","title":"OptimizationSense","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OptimizationSense","category":"page"},{"location":"api/JuMP/#JuMP.OptimizationSense","page":"JuMP","title":"JuMP.OptimizationSense","text":"OptimizationSense\n\nAn enum for the value of the ObjectiveSense attribute.\n\nValues\n\nPossible values are:\n\nMIN_SENSE: the goal is to minimize the objective function\nMAX_SENSE: the goal is to maximize the objective function\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#OptimizeNotCalled","page":"JuMP","title":"OptimizeNotCalled","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OptimizeNotCalled","category":"page"},{"location":"api/JuMP/#JuMP.OptimizeNotCalled","page":"JuMP","title":"JuMP.OptimizeNotCalled","text":"struct OptimizeNotCalled <: Exception end\n\nA result attribute cannot be queried before optimize! is called.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#PSDCone","page":"JuMP","title":"PSDCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.PSDCone","category":"page"},{"location":"api/JuMP/#JuMP.PSDCone","page":"JuMP","title":"JuMP.PSDCone","text":"PSDCone\n\nPositive semidefinite cone object that can be used to constrain a square matrix to be positive semidefinite in the @constraint macro. If the matrix has type Symmetric then the columns vectorization (the vector obtained by concatenating the columns) of its upper triangular part is constrained to belong to the MOI.PositiveSemidefiniteConeTriangle set, otherwise its column vectorization is constrained to belong to the MOI.PositiveSemidefiniteConeSquare set.\n\nExample\n\nConsider the following example:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> a = [ x 2x\n            2x  x];\n\njulia> b = [1 2\n            2 4];\n\njulia> cref = @constraint(model, a >= b, PSDCone())\n[x - 1    2 x - 2;\n 2 x - 2  x - 4] ∈ PSDCone()\n\njulia> jump_function(constraint_object(cref))\n4-element Vector{AffExpr}:\n x - 1\n 2 x - 2\n 2 x - 2\n x - 4\n\njulia> moi_set(constraint_object(cref))\nMathOptInterface.PositiveSemidefiniteConeSquare(2)\n\nWe see in the output of the last command that the vectorization of the matrix is constrained to belong to the PositiveSemidefiniteConeSquare.\n\njulia> using LinearAlgebra # For Symmetric\n\njulia> cref = @constraint(model, Symmetric(a - b) in PSDCone())\n[x - 1    2 x - 2;\n 2 x - 2  x - 4] ∈ PSDCone()\n\njulia> jump_function(constraint_object(cref))\n3-element Vector{AffExpr}:\n x - 1\n 2 x - 2\n x - 4\n\njulia> moi_set(constraint_object(cref))\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\nAs we see in the output of the last command, the vectorization of only the upper triangular part of the matrix is constrained to belong to the PositiveSemidefiniteConeSquare.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Parameter","page":"JuMP","title":"Parameter","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Parameter","category":"page"},{"location":"api/JuMP/#JuMP.Parameter","page":"JuMP","title":"JuMP.Parameter","text":"Parameter(value)\n\nA short-cut for the MOI.Parameter set.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Parameter(2))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Parameter{Float64}(2.0)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ParameterRef","page":"JuMP","title":"ParameterRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ParameterRef","category":"page"},{"location":"api/JuMP/#JuMP.ParameterRef","page":"JuMP","title":"JuMP.ParameterRef","text":"ParameterRef(x::GenericVariableRef)\n\nReturn a constraint reference to the constraint constraining x to be a parameter.\n\nErrors if one does not exist.\n\nSee also is_parameter, set_parameter_value, parameter_value.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, p in Parameter(2))\np\n\njulia> ParameterRef(p)\np ∈ MathOptInterface.Parameter{Float64}(2.0)\n\njulia> @variable(model, x);\n\njulia> ParameterRef(x)\nERROR: Variable x is not a parameter.\nStacktrace:\n[...]\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#QuadExpr","page":"JuMP","title":"QuadExpr","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.QuadExpr","category":"page"},{"location":"api/JuMP/#JuMP.QuadExpr","page":"JuMP","title":"JuMP.QuadExpr","text":"QuadExpr\n\nAn alias for GenericQuadExpr{Float64,VariableRef}, the specific GenericQuadExpr used by JuMP.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#QuadTermIterator","page":"JuMP","title":"QuadTermIterator","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.QuadTermIterator","category":"page"},{"location":"api/JuMP/#JuMP.QuadTermIterator","page":"JuMP","title":"JuMP.QuadTermIterator","text":"QuadTermIterator{GQE<:GenericQuadExpr}\n\nA struct that implements the iterate protocol in order to iterate over tuples of (coefficient, variable, variable) in the GenericQuadExpr.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ReferenceMap","page":"JuMP","title":"ReferenceMap","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ReferenceMap","category":"page"},{"location":"api/JuMP/#JuMP.ReferenceMap","page":"JuMP","title":"JuMP.ReferenceMap","text":"GenericReferenceMap{T}\n\nMapping between variable and constraint reference of a model and its copy. The reference of the copied model can be obtained by indexing the map with the reference of the corresponding reference of the original model.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ResultStatusCode","page":"JuMP","title":"ResultStatusCode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ResultStatusCode","category":"page"},{"location":"api/JuMP/#JuMP.ResultStatusCode","page":"JuMP","title":"JuMP.ResultStatusCode","text":"ResultStatusCode\n\nAn Enum of possible values for the PrimalStatus and DualStatus attributes.\n\nThe values indicate how to interpret the result vector.\n\nValues\n\nPossible values are:\n\nNO_SOLUTION: the result vector is empty.\nFEASIBLE_POINT: the result vector is a feasible point.\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\nINFEASIBLE_POINT: the result vector is an infeasible point.\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#RotatedSecondOrderCone","page":"JuMP","title":"RotatedSecondOrderCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.RotatedSecondOrderCone","category":"page"},{"location":"api/JuMP/#JuMP.RotatedSecondOrderCone","page":"JuMP","title":"JuMP.RotatedSecondOrderCone","text":"RotatedSecondOrderCone\n\nRotated second order cone object that can be used to constrain the square of the euclidean norm of a vector x to be less than or equal to 2tu where t and u are nonnegative scalars. This is a shortcut for the MOI.RotatedSecondOrderCone.\n\nExample\n\nThe following constrains (x-1 x-2)^2_2 le 2tx and t x ge 0:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t, x, x-1, x-2] in RotatedSecondOrderCone())\n[t, x, x - 1, x - 2] ∈ MathOptInterface.RotatedSecondOrderCone(4)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SOS1","page":"JuMP","title":"SOS1","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOS1","category":"page"},{"location":"api/JuMP/#JuMP.SOS1","page":"JuMP","title":"JuMP.SOS1","text":"SOS1\n\nSOS1 (Special Ordered Sets type 1) object than can be used to constrain a vector x to a set where at most 1 variable can take a non-zero value, all others being at 0. The weights, when specified, induce an ordering of the variables; as such, they should be unique values. The kth element in the set corresponds to the kth weight in weights. See here for a description of SOS constraints and their potential uses. This is a shortcut for the MathOptInterface.SOS1 set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SOS2","page":"JuMP","title":"SOS2","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOS2","category":"page"},{"location":"api/JuMP/#JuMP.SOS2","page":"JuMP","title":"JuMP.SOS2","text":"SOS2\n\nSOS1 (Special Ordered Sets type 2) object than can be used to constrain a vector x to a set where at most 2 variables can take a non-zero value, all others being at 0. In addition, if two are non-zero these must be consecutive in their ordering. The weights induce an ordering of the variables; as such, they should be unique values. The kth element in the set corresponds to the kth weight in weights. See here for a description of SOS constraints and their potential uses. This is a shortcut for the MathOptInterface.SOS2 set.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarConstraint","page":"JuMP","title":"ScalarConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarConstraint","category":"page"},{"location":"api/JuMP/#JuMP.ScalarConstraint","page":"JuMP","title":"JuMP.ScalarConstraint","text":"struct ScalarConstraint\n\nThe data for a scalar constraint. The func field contains a JuMP object representing the function and the set field contains the MOI set. See also the documentation on JuMP's representation of constraints for more background.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarShape","page":"JuMP","title":"ScalarShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarShape","category":"page"},{"location":"api/JuMP/#JuMP.ScalarShape","page":"JuMP","title":"JuMP.ScalarShape","text":"ScalarShape\n\nShape of scalar constraints.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ScalarVariable","page":"JuMP","title":"ScalarVariable","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ScalarVariable","category":"page"},{"location":"api/JuMP/#JuMP.ScalarVariable","page":"JuMP","title":"JuMP.ScalarVariable","text":"ScalarVariable{S,T,U,V} <: AbstractVariable\n\nA struct used when adding variables.\n\nSee also: add_variable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SecondOrderCone","page":"JuMP","title":"SecondOrderCone","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SecondOrderCone","category":"page"},{"location":"api/JuMP/#JuMP.SecondOrderCone","page":"JuMP","title":"JuMP.SecondOrderCone","text":"SecondOrderCone\n\nSecond order cone object that can be used to constrain the euclidean norm of a vector x to be less than or equal to a nonnegative scalar t. This is a shortcut for the MOI.SecondOrderCone.\n\nExample\n\nThe following constrains (x-1 x-2)_2 le t and t ge 0:\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, t)\nt\n\njulia> @constraint(model, [t, x-1, x-2] in SecondOrderCone())\n[t, x - 1, x - 2] ∈ MathOptInterface.SecondOrderCone(3)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Semicontinuous","page":"JuMP","title":"Semicontinuous","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Semicontinuous","category":"page"},{"location":"api/JuMP/#JuMP.Semicontinuous","page":"JuMP","title":"JuMP.Semicontinuous","text":"Semicontinuous(lower, upper)\n\nA short-cut for the MOI.Semicontinuous set.\n\nThis short-cut is useful because it automatically promotes lower and upper to the same type, and converts them into the element type supported by the JuMP model.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Semicontinuous(1, 2))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Semicontinuous{Int64}(1, 2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Semiinteger","page":"JuMP","title":"Semiinteger","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Semiinteger","category":"page"},{"location":"api/JuMP/#JuMP.Semiinteger","page":"JuMP","title":"JuMP.Semiinteger","text":"Semiinteger(lower, upper)\n\nA short-cut for the MOI.Semiinteger set.\n\nThis short-cut is useful because it automatically promotes lower and upper to the same type, and converts them into the element type supported by the JuMP model.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x in Semiinteger(3, 5))\nx\n\njulia> print(model)\nFeasibility\nSubject to\n x ∈ MathOptInterface.Semiinteger{Int64}(3, 5)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SensitivityReport","page":"JuMP","title":"SensitivityReport","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SensitivityReport","category":"page"},{"location":"api/JuMP/#JuMP.SensitivityReport","page":"JuMP","title":"JuMP.SensitivityReport","text":"SensitivityReport\n\nSee lp_sensitivity_report.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkewSymmetricMatrixShape","page":"JuMP","title":"SkewSymmetricMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkewSymmetricMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SkewSymmetricMatrixShape","page":"JuMP","title":"JuMP.SkewSymmetricMatrixShape","text":"SkewSymmetricMatrixShape\n\nShape object for a skew symmetric square matrix of side_dimension rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix (without the diagonal) given column by column (or equivalently, the entries of the lower-left triangular part given row by row). The diagonal is zero.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkewSymmetricMatrixSpace","page":"JuMP","title":"SkewSymmetricMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkewSymmetricMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.SkewSymmetricMatrixSpace","page":"JuMP","title":"JuMP.SkewSymmetricMatrixSpace","text":"SkewSymmetricMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be skew-symmetric.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        Q[1,2]\n -Q[1,2]  0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SkipModelConvertScalarSetWrapper","page":"JuMP","title":"SkipModelConvertScalarSetWrapper","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SkipModelConvertScalarSetWrapper","category":"page"},{"location":"api/JuMP/#JuMP.SkipModelConvertScalarSetWrapper","page":"JuMP","title":"JuMP.SkipModelConvertScalarSetWrapper","text":"SkipModelConvertScalarSetWrapper(set::MOI.AbstractScalarSet)\n\nJuMP uses model_convert to automatically promote MOI.AbstractScalarSet sets to the same value_type as the model.\n\nIn cases there this is undesirable, wrap the set in SkipModelConvertScalarSetWrapper to pass the set un-changed to the solver.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @constraint(model, x in MOI.EqualTo(1 // 2))\nx = 0.5\n\njulia> @constraint(model, x in SkipModelConvertScalarSetWrapper(MOI.EqualTo(1 // 2)))\nx = 1//2\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SquareMatrixShape","page":"JuMP","title":"SquareMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SquareMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SquareMatrixShape","page":"JuMP","title":"JuMP.SquareMatrixShape","text":"SquareMatrixShape\n\nShape object for a square matrix of side_dimension rows and columns. The vectorized form contains the entries of the the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SymmetricMatrixShape","page":"JuMP","title":"SymmetricMatrixShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SymmetricMatrixShape","category":"page"},{"location":"api/JuMP/#JuMP.SymmetricMatrixShape","page":"JuMP","title":"JuMP.SymmetricMatrixShape","text":"SymmetricMatrixShape\n\nShape object for a symmetric square matrix of side_dimension rows and columns. The vectorized form contains the entries of the upper-right triangular part of the matrix given column by column (or equivalently, the entries of the lower-left triangular part given row by row).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#SymmetricMatrixSpace","page":"JuMP","title":"SymmetricMatrixSpace","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SymmetricMatrixSpace","category":"page"},{"location":"api/JuMP/#JuMP.SymmetricMatrixSpace","page":"JuMP","title":"JuMP.SymmetricMatrixSpace","text":"SymmetricMatrixSpace()\n\nUse in the @variable macro to constrain a matrix of variables to be symmetric.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, Q[1:2, 1:2] in SymmetricMatrixSpace())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n Q[1,1]  Q[1,2]\n Q[1,2]  Q[2,2]\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#TerminationStatusCode","page":"JuMP","title":"TerminationStatusCode","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.TerminationStatusCode","category":"page"},{"location":"api/JuMP/#JuMP.TerminationStatusCode","page":"JuMP","title":"JuMP.TerminationStatusCode","text":"TerminationStatusCode\n\nAn Enum of possible values for the TerminationStatus attribute. This attribute is meant to explain the reason why the optimizer stopped executing in the most recent call to optimize!.\n\nValues\n\nPossible values are:\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\nOPTIMAL: The algorithm found a globally optimal solution.\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\nINVALID_MODEL: The algorithm stopped because the model is invalid.\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#UnorderedPair","page":"JuMP","title":"UnorderedPair","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UnorderedPair","category":"page"},{"location":"api/JuMP/#JuMP.UnorderedPair","page":"JuMP","title":"JuMP.UnorderedPair","text":"UnorderedPair(a::T, b::T)\n\nA wrapper type used by GenericQuadExpr with fields .a and .b.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#UpperBoundRef","page":"JuMP","title":"UpperBoundRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UpperBoundRef","category":"page"},{"location":"api/JuMP/#JuMP.UpperBoundRef","page":"JuMP","title":"JuMP.UpperBoundRef","text":"UpperBoundRef(v::GenericVariableRef)\n\nReturn a constraint reference to the upper bound constraint of v.\n\nErrors if one does not exist.\n\nSee also has_upper_bound, upper_bound, set_upper_bound, delete_upper_bound.\n\nExamples\n\njulia> model = Model();\n\njulia> @variable(model, x <= 1.0);\n\njulia> UpperBoundRef(x)\nx ≤ 1\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP/#VariableConstrainedOnCreation","page":"JuMP","title":"VariableConstrainedOnCreation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableConstrainedOnCreation","category":"page"},{"location":"api/JuMP/#JuMP.VariableConstrainedOnCreation","page":"JuMP","title":"JuMP.VariableConstrainedOnCreation","text":"VariableConstrainedOnCreation <: AbstractVariable\n\nVariable scalar_variables constrained to belong to set.\n\nAdding this variable can be understood as doing:\n\nfunction JuMP.add_variable(\n    model::GenericModel,\n    variable::VariableConstrainedOnCreation,\n    names,\n)\n    var_ref = add_variable(model, variable.scalar_variable, name)\n    add_constraint(model, VectorConstraint(var_ref, variable.set))\n    return var_ref\nend\n\nbut adds the variables with MOI.add_constrained_variable(model, variable.set) instead. See the MOI documentation for the difference between adding the variables with MOI.add_constrained_variable and adding them with MOI.add_variable and adding the constraint separately.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableInfo","page":"JuMP","title":"VariableInfo","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableInfo","category":"page"},{"location":"api/JuMP/#JuMP.VariableInfo","page":"JuMP","title":"JuMP.VariableInfo","text":"VariableInfo{S,T,U,V}\n\nA struct by JuMP internally when creating variables. This may also be used by JuMP extensions to create new types of variables.\n\nSee also: ScalarVariable.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableNotOwned","page":"JuMP","title":"VariableNotOwned","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableNotOwned","category":"page"},{"location":"api/JuMP/#JuMP.VariableNotOwned","page":"JuMP","title":"JuMP.VariableNotOwned","text":"struct VariableNotOwned{V<:AbstractVariableRef} <: Exception\n    variable::V\nend\n\nThe variable variable was used in a model different to owner_model(variable).\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariableRef","page":"JuMP","title":"VariableRef","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariableRef","category":"page"},{"location":"api/JuMP/#JuMP.VariableRef","page":"JuMP","title":"JuMP.VariableRef","text":"GenericVariableRef{T} <: AbstractVariableRef\n\nHolds a reference to the model and the corresponding MOI.VariableIndex.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VariablesConstrainedOnCreation","page":"JuMP","title":"VariablesConstrainedOnCreation","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VariablesConstrainedOnCreation","category":"page"},{"location":"api/JuMP/#JuMP.VariablesConstrainedOnCreation","page":"JuMP","title":"JuMP.VariablesConstrainedOnCreation","text":"VariablesConstrainedOnCreation <: AbstractVariable\n\nVector of variables scalar_variables constrained to belong to set. Adding this variable can be thought as doing:\n\nfunction JuMP.add_variable(\n    model::GenericModel,\n    variable::VariablesConstrainedOnCreation,\n    names,\n)\n    v_names = vectorize(names, variable.shape)\n    var_refs = add_variable.(model, variable.scalar_variables, v_names)\n    add_constraint(model, VectorConstraint(var_refs, variable.set))\n    return reshape_vector(var_refs, variable.shape)\nend\n\nbut adds the variables with MOI.add_constrained_variables(model, variable.set) instead. See the MOI documentation for the difference between adding the variables with MOI.add_constrained_variables and adding them with MOI.add_variables and adding the constraint separately.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VectorConstraint","page":"JuMP","title":"VectorConstraint","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VectorConstraint","category":"page"},{"location":"api/JuMP/#JuMP.VectorConstraint","page":"JuMP","title":"JuMP.VectorConstraint","text":"struct VectorConstraint\n\nThe data for a vector constraint. The func field contains a JuMP object representing the function and the set field contains the MOI set. The shape field contains an AbstractShape matching the form in which the constraint was constructed (e.g., by using matrices or flat vectors). See also the documentation on JuMP's representation of constraints.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#VectorShape","page":"JuMP","title":"VectorShape","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.VectorShape","category":"page"},{"location":"api/JuMP/#JuMP.VectorShape","page":"JuMP","title":"JuMP.VectorShape","text":"VectorShape\n\nVector for which the vectorized form corresponds exactly to the vector given.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#Zeros","page":"JuMP","title":"Zeros","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Zeros","category":"page"},{"location":"api/JuMP/#JuMP.Zeros","page":"JuMP","title":"JuMP.Zeros","text":"Zeros()\n\nThe JuMP equivalent of the MOI.Zeros set, in which the dimension is inferred from the corresponding function.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> @constraint(model, x in Zeros())\n[x[1], x[2]] ∈ MathOptInterface.Zeros(2)\n\njulia> A = [1 2; 3 4];\n\njulia> b = [5, 6];\n\njulia> @constraint(model, A * x == b)\n[x[1] + 2 x[2] - 5, 3 x[1] + 4 x[2] - 6] ∈ MathOptInterface.Zeros(2)\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP/#ALMOST_DUAL_INFEASIBLE","page":"JuMP","title":"ALMOST_DUAL_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_DUAL_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_DUAL_INFEASIBLE","page":"JuMP","title":"JuMP.ALMOST_DUAL_INFEASIBLE","text":"ALMOST_DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_DUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_INFEASIBLE","page":"JuMP","title":"ALMOST_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_INFEASIBLE","page":"JuMP","title":"JuMP.ALMOST_INFEASIBLE","text":"ALMOST_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_INFEASIBLE: The algorithm concluded that no feasible solution exists within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_LOCALLY_SOLVED","page":"JuMP","title":"ALMOST_LOCALLY_SOLVED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_LOCALLY_SOLVED","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_LOCALLY_SOLVED","page":"JuMP","title":"JuMP.ALMOST_LOCALLY_SOLVED","text":"ALMOST_LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_LOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, or could not find directions for improvement within relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ALMOST_OPTIMAL","page":"JuMP","title":"ALMOST_OPTIMAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ALMOST_OPTIMAL","category":"page"},{"location":"api/JuMP/#JuMP.ALMOST_OPTIMAL","page":"JuMP","title":"JuMP.ALMOST_OPTIMAL","text":"ALMOST_OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nALMOST_OPTIMAL: The algorithm found a globally optimal solution to relaxed tolerances.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#AUTOMATIC","page":"JuMP","title":"AUTOMATIC","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.AUTOMATIC","category":"page"},{"location":"api/JuMP/#JuMP.AUTOMATIC","page":"JuMP","title":"JuMP.AUTOMATIC","text":"moi_backend field holds a CachingOptimizer in AUTOMATIC mode.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#DIRECT","page":"JuMP","title":"DIRECT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.DIRECT","category":"page"},{"location":"api/JuMP/#JuMP.DIRECT","page":"JuMP","title":"JuMP.DIRECT","text":"moi_backend field holds an AbstractOptimizer. No extra copy of the model is stored. The moi_backend must support add_constraint etc.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#DUAL_INFEASIBLE","page":"JuMP","title":"DUAL_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.DUAL_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.DUAL_INFEASIBLE","page":"JuMP","title":"JuMP.DUAL_INFEASIBLE","text":"DUAL_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nDUAL_INFEASIBLE: The algorithm concluded that no dual bound exists for the problem. If, additionally, a feasible (primal) solution is known to exist, this status typically implies that the problem is unbounded, with some technical exceptions.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#FEASIBILITY_SENSE","page":"JuMP","title":"FEASIBILITY_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FEASIBILITY_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.FEASIBILITY_SENSE","page":"JuMP","title":"JuMP.FEASIBILITY_SENSE","text":"FEASIBILITY_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nFEASIBILITY_SENSE: the model does not have an objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#FEASIBLE_POINT","page":"JuMP","title":"FEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.FEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.FEASIBLE_POINT","page":"JuMP","title":"JuMP.FEASIBLE_POINT","text":"FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nFEASIBLE_POINT: the result vector is a feasible point.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"INFEASIBILITY_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBILITY_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"JuMP.INFEASIBILITY_CERTIFICATE","text":"INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBILITY_CERTIFICATE: the result vector is an infeasibility certificate. If the PrimalStatus is INFEASIBILITY_CERTIFICATE, then the primal result vector is a certificate of dual infeasibility. If the DualStatus is INFEASIBILITY_CERTIFICATE, then the dual result vector is a proof of primal infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE","page":"JuMP","title":"INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE","page":"JuMP","title":"JuMP.INFEASIBLE","text":"INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE: The algorithm concluded that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE_OR_UNBOUNDED","page":"JuMP","title":"INFEASIBLE_OR_UNBOUNDED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE_OR_UNBOUNDED","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE_OR_UNBOUNDED","page":"JuMP","title":"JuMP.INFEASIBLE_OR_UNBOUNDED","text":"INFEASIBLE_OR_UNBOUNDED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINFEASIBLE_OR_UNBOUNDED: The algorithm stopped because it decided that the problem is infeasible or unbounded; this occasionally happens during MIP presolve.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INFEASIBLE_POINT","page":"JuMP","title":"INFEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INFEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.INFEASIBLE_POINT","page":"JuMP","title":"JuMP.INFEASIBLE_POINT","text":"INFEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nINFEASIBLE_POINT: the result vector is an infeasible point.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INTERRUPTED","page":"JuMP","title":"INTERRUPTED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INTERRUPTED","category":"page"},{"location":"api/JuMP/#JuMP.INTERRUPTED","page":"JuMP","title":"JuMP.INTERRUPTED","text":"INTERRUPTED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINTERRUPTED: The algorithm stopped because of an interrupt signal.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INVALID_MODEL","page":"JuMP","title":"INVALID_MODEL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INVALID_MODEL","category":"page"},{"location":"api/JuMP/#JuMP.INVALID_MODEL","page":"JuMP","title":"JuMP.INVALID_MODEL","text":"INVALID_MODEL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_MODEL: The algorithm stopped because the model is invalid.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#INVALID_OPTION","page":"JuMP","title":"INVALID_OPTION","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.INVALID_OPTION","category":"page"},{"location":"api/JuMP/#JuMP.INVALID_OPTION","page":"JuMP","title":"JuMP.INVALID_OPTION","text":"INVALID_OPTION::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nINVALID_OPTION: The algorithm stopped because it was provided an invalid option.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#ITERATION_LIMIT","page":"JuMP","title":"ITERATION_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.ITERATION_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.ITERATION_LIMIT","page":"JuMP","title":"JuMP.ITERATION_LIMIT","text":"ITERATION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nITERATION_LIMIT: An iterative algorithm stopped after conducting the maximum number of iterations.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#LOCALLY_INFEASIBLE","page":"JuMP","title":"LOCALLY_INFEASIBLE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LOCALLY_INFEASIBLE","category":"page"},{"location":"api/JuMP/#JuMP.LOCALLY_INFEASIBLE","page":"JuMP","title":"JuMP.LOCALLY_INFEASIBLE","text":"LOCALLY_INFEASIBLE::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_INFEASIBLE: The algorithm converged to an infeasible point or otherwise completed its search without finding a feasible solution, without guarantees that no feasible solution exists.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#LOCALLY_SOLVED","page":"JuMP","title":"LOCALLY_SOLVED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.LOCALLY_SOLVED","category":"page"},{"location":"api/JuMP/#JuMP.LOCALLY_SOLVED","page":"JuMP","title":"JuMP.LOCALLY_SOLVED","text":"LOCALLY_SOLVED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nLOCALLY_SOLVED: The algorithm converged to a stationary point, local optimal solution, could not find directions for improvement, or otherwise completed its search without global guarantees.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MANUAL","page":"JuMP","title":"MANUAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MANUAL","category":"page"},{"location":"api/JuMP/#JuMP.MANUAL","page":"JuMP","title":"JuMP.MANUAL","text":"moi_backend field holds a CachingOptimizer in MANUAL mode.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MAX_SENSE","page":"JuMP","title":"MAX_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MAX_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.MAX_SENSE","page":"JuMP","title":"JuMP.MAX_SENSE","text":"MAX_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMAX_SENSE: the goal is to maximize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MEMORY_LIMIT","page":"JuMP","title":"MEMORY_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MEMORY_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.MEMORY_LIMIT","page":"JuMP","title":"JuMP.MEMORY_LIMIT","text":"MEMORY_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nMEMORY_LIMIT: The algorithm stopped because it ran out of memory.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#MIN_SENSE","page":"JuMP","title":"MIN_SENSE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MIN_SENSE","category":"page"},{"location":"api/JuMP/#JuMP.MIN_SENSE","page":"JuMP","title":"JuMP.MIN_SENSE","text":"MIN_SENSE::OptimizationSense\n\nAn instance of the OptimizationSense enum.\n\nMIN_SENSE: the goal is to minimize the objective function\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_FEASIBLE_POINT","page":"JuMP","title":"NEARLY_FEASIBLE_POINT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_FEASIBLE_POINT","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_FEASIBLE_POINT","page":"JuMP","title":"JuMP.NEARLY_FEASIBLE_POINT","text":"NEARLY_FEASIBLE_POINT::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_FEASIBLE_POINT: the result vector is feasible if some constraint tolerances are relaxed.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"NEARLY_INFEASIBILITY_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","page":"JuMP","title":"JuMP.NEARLY_INFEASIBILITY_CERTIFICATE","text":"NEARLY_INFEASIBILITY_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_INFEASIBILITY_CERTIFICATE: the result satisfies a relaxed criterion for a certificate of infeasibility.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NEARLY_REDUCTION_CERTIFICATE","page":"JuMP","title":"NEARLY_REDUCTION_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NEARLY_REDUCTION_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.NEARLY_REDUCTION_CERTIFICATE","page":"JuMP","title":"JuMP.NEARLY_REDUCTION_CERTIFICATE","text":"NEARLY_REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNEARLY_REDUCTION_CERTIFICATE: the result satisfies a relaxed criterion for an ill-posed certificate.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NODE_LIMIT","page":"JuMP","title":"NODE_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NODE_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.NODE_LIMIT","page":"JuMP","title":"JuMP.NODE_LIMIT","text":"NODE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNODE_LIMIT: A branch-and-bound algorithm stopped because it explored a maximum number of nodes in the branch-and-bound tree.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NORM_LIMIT","page":"JuMP","title":"NORM_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NORM_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.NORM_LIMIT","page":"JuMP","title":"JuMP.NORM_LIMIT","text":"NORM_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNORM_LIMIT: The algorithm stopped because the norm of an iterate became too large.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NO_SOLUTION","page":"JuMP","title":"NO_SOLUTION","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NO_SOLUTION","category":"page"},{"location":"api/JuMP/#JuMP.NO_SOLUTION","page":"JuMP","title":"JuMP.NO_SOLUTION","text":"NO_SOLUTION::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nNO_SOLUTION: the result vector is empty.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#NUMERICAL_ERROR","page":"JuMP","title":"NUMERICAL_ERROR","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.NUMERICAL_ERROR","category":"page"},{"location":"api/JuMP/#JuMP.NUMERICAL_ERROR","page":"JuMP","title":"JuMP.NUMERICAL_ERROR","text":"NUMERICAL_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nNUMERICAL_ERROR: The algorithm stopped because it encountered unrecoverable numerical error.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OBJECTIVE_LIMIT","page":"JuMP","title":"OBJECTIVE_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OBJECTIVE_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.OBJECTIVE_LIMIT","page":"JuMP","title":"JuMP.OBJECTIVE_LIMIT","text":"OBJECTIVE_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOBJECTIVE_LIMIT: The algorithm stopped because it found a solution better than a minimum limit set by the user.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OPTIMAL","page":"JuMP","title":"OPTIMAL","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OPTIMAL","category":"page"},{"location":"api/JuMP/#JuMP.OPTIMAL","page":"JuMP","title":"JuMP.OPTIMAL","text":"OPTIMAL::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMAL: The algorithm found a globally optimal solution.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OPTIMIZE_NOT_CALLED","page":"JuMP","title":"OPTIMIZE_NOT_CALLED","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OPTIMIZE_NOT_CALLED","category":"page"},{"location":"api/JuMP/#JuMP.OPTIMIZE_NOT_CALLED","page":"JuMP","title":"JuMP.OPTIMIZE_NOT_CALLED","text":"OPTIMIZE_NOT_CALLED::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOPTIMIZE_NOT_CALLED: The algorithm has not started.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_ERROR","page":"JuMP","title":"OTHER_ERROR","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_ERROR","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_ERROR","page":"JuMP","title":"JuMP.OTHER_ERROR","text":"OTHER_ERROR::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_ERROR: The algorithm stopped because of an error not covered by one of the statuses defined above.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_LIMIT","page":"JuMP","title":"OTHER_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_LIMIT","page":"JuMP","title":"JuMP.OTHER_LIMIT","text":"OTHER_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nOTHER_LIMIT: The algorithm stopped due to a limit not covered by one of the _LIMIT_ statuses above.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#OTHER_RESULT_STATUS","page":"JuMP","title":"OTHER_RESULT_STATUS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.OTHER_RESULT_STATUS","category":"page"},{"location":"api/JuMP/#JuMP.OTHER_RESULT_STATUS","page":"JuMP","title":"JuMP.OTHER_RESULT_STATUS","text":"OTHER_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nOTHER_RESULT_STATUS: the result vector contains a solution with an interpretation not covered by one of the statuses defined above\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#REDUCTION_CERTIFICATE","page":"JuMP","title":"REDUCTION_CERTIFICATE","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.REDUCTION_CERTIFICATE","category":"page"},{"location":"api/JuMP/#JuMP.REDUCTION_CERTIFICATE","page":"JuMP","title":"JuMP.REDUCTION_CERTIFICATE","text":"REDUCTION_CERTIFICATE::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nREDUCTION_CERTIFICATE: the result vector is an ill-posed certificate; see this article for details. If the PrimalStatus is REDUCTION_CERTIFICATE, then the primal result vector is a proof that the dual problem is ill-posed. If the DualStatus is REDUCTION_CERTIFICATE, then the dual result vector is a proof that the primal is ill-posed.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#SLOW_PROGRESS","page":"JuMP","title":"SLOW_PROGRESS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SLOW_PROGRESS","category":"page"},{"location":"api/JuMP/#JuMP.SLOW_PROGRESS","page":"JuMP","title":"JuMP.SLOW_PROGRESS","text":"SLOW_PROGRESS::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSLOW_PROGRESS: The algorithm stopped because it was unable to continue making progress towards the solution.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#SOLUTION_LIMIT","page":"JuMP","title":"SOLUTION_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.SOLUTION_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.SOLUTION_LIMIT","page":"JuMP","title":"JuMP.SOLUTION_LIMIT","text":"SOLUTION_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nSOLUTION_LIMIT: The algorithm stopped because it found the required number of solutions. This is often used in MIPs to get the solver to return the first feasible solution it encounters.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#TIME_LIMIT","page":"JuMP","title":"TIME_LIMIT","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.TIME_LIMIT","category":"page"},{"location":"api/JuMP/#JuMP.TIME_LIMIT","page":"JuMP","title":"JuMP.TIME_LIMIT","text":"TIME_LIMIT::TerminationStatusCode\n\nAn instance of the TerminationStatusCode enum.\n\nTIME_LIMIT: The algorithm stopped after a user-specified computation time.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#UNKNOWN_RESULT_STATUS","page":"JuMP","title":"UNKNOWN_RESULT_STATUS","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.UNKNOWN_RESULT_STATUS","category":"page"},{"location":"api/JuMP/#JuMP.UNKNOWN_RESULT_STATUS","page":"JuMP","title":"JuMP.UNKNOWN_RESULT_STATUS","text":"UNKNOWN_RESULT_STATUS::ResultStatusCode\n\nAn instance of the ResultStatusCode enum.\n\nUNKNOWN_RESULT_STATUS: the result vector contains a solution with an unknown interpretation.\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_and","page":"JuMP","title":"op_and","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_and","category":"page"},{"location":"api/JuMP/#JuMP.op_and","page":"JuMP","title":"JuMP.op_and","text":"op_and(x, y)\n\nA function that falls back to x & y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_and(true, false)\nfalse\n\njulia> op_and(true, x)\ntrue && x\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_equal_to","page":"JuMP","title":"op_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_equal_to","page":"JuMP","title":"JuMP.op_equal_to","text":"op_equal_to(x, y)\n\nA function that falls back to x == y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_equal_to(2, 2)\ntrue\n\njulia> op_equal_to(x, 2)\nx == 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_greater_than_or_equal_to","page":"JuMP","title":"op_greater_than_or_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_greater_than_or_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_greater_than_or_equal_to","page":"JuMP","title":"JuMP.op_greater_than_or_equal_to","text":"op_greater_than_or_equal_to(x, y)\n\nA function that falls back to x >= y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_greater_than_or_equal_to(2, 2)\ntrue\n\njulia> op_greater_than_or_equal_to(x, 2)\nx >= 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_less_than_or_equal_to","page":"JuMP","title":"op_less_than_or_equal_to","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_less_than_or_equal_to","category":"page"},{"location":"api/JuMP/#JuMP.op_less_than_or_equal_to","page":"JuMP","title":"JuMP.op_less_than_or_equal_to","text":"op_less_than_or_equal_to(x, y)\n\nA function that falls back to x <= y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_less_than_or_equal_to(2, 2)\ntrue\n\njulia> op_less_than_or_equal_to(x, 2)\nx <= 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_or","page":"JuMP","title":"op_or","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_or","category":"page"},{"location":"api/JuMP/#JuMP.op_or","page":"JuMP","title":"JuMP.op_or","text":"op_or(x, y)\n\nA function that falls back to x | y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_or(true, false)\ntrue\n\njulia> op_or(true, x)\ntrue || x\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_strictly_greater_than","page":"JuMP","title":"op_strictly_greater_than","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_strictly_greater_than","category":"page"},{"location":"api/JuMP/#JuMP.op_strictly_greater_than","page":"JuMP","title":"JuMP.op_strictly_greater_than","text":"op_strictly_greater_than(x, y)\n\nA function that falls back to x > y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_strictly_greater_than(1, 2)\nfalse\n\njulia> op_strictly_greater_than(x, 2)\nx > 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#op_strictly_less_than","page":"JuMP","title":"op_strictly_less_than","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.op_strictly_less_than","category":"page"},{"location":"api/JuMP/#JuMP.op_strictly_less_than","page":"JuMP","title":"JuMP.op_strictly_less_than","text":"op_strictly_less_than(x, y)\n\nA function that falls back to x < y, but when called with JuMP variables or expressions, returns a GenericNonlinearExpr.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x);\n\njulia> op_strictly_less_than(1, 2)\ntrue\n\njulia> op_strictly_less_than(x, 2)\nx < 2\n\n\n\n\n\n","category":"constant"},{"location":"api/JuMP/#Base.empty!(::GenericModel)","page":"JuMP","title":"Base.empty!(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.empty!(::GenericModel)","category":"page"},{"location":"api/JuMP/#Base.empty!-Tuple{GenericModel}","page":"JuMP","title":"Base.empty!","text":"empty!(model::GenericModel)::GenericModel\n\nEmpty the model, that is, remove all variables, constraints and model attributes but not optimizer attributes. Always return the argument.\n\nNote: removes extensions data.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.isempty(::GenericModel)","page":"JuMP","title":"Base.isempty(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.isempty(::GenericModel)","category":"page"},{"location":"api/JuMP/#Base.isempty-Tuple{GenericModel}","page":"JuMP","title":"Base.isempty","text":"isempty(model::GenericModel)\n\nVerifies whether the model is empty, that is, whether the MOI backend is empty and whether the model is in the same state as at its creation apart from optimizer attributes.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.copy(::AbstractModel)","page":"JuMP","title":"Base.copy(::AbstractModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.copy(::AbstractModel)","category":"page"},{"location":"api/JuMP/#Base.copy-Tuple{AbstractModel}","page":"JuMP","title":"Base.copy","text":"copy(model::AbstractModel)\n\nReturn a copy of the model model. It is similar to copy_model except that it does not return the mapping between the references of model and its copy.\n\nNote\n\nModel copy is not supported in DIRECT mode, i.e. when a model is constructed using the direct_model constructor instead of the Model constructor. Moreover, independently on whether an optimizer was provided at model construction, the new model will have no optimizer, i.e., an optimizer will have to be provided to the new model in the optimize! call.\n\nExample\n\nIn the following example, a model model is constructed with a variable x and a constraint cref. It is then copied into a model new_model with the new references assigned to x_new and cref_new.\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @constraint(model, cref, x == 2)\ncref : x = 2\n\njulia> new_model = copy(model);\n\njulia> x_new = model[:x]\nx\n\njulia> cref_new = model[:cref]\ncref : x = 2\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#Base.write(::IO,-::GenericModel;-::MOI.FileFormats.FileFormat)","page":"JuMP","title":"Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.Base.write(::IO, ::GenericModel; ::MOI.FileFormats.FileFormat)","category":"page"},{"location":"api/JuMP/#Base.write-Tuple{IO, GenericModel}","page":"JuMP","title":"Base.write","text":"Base.write(\n    io::IO,\n    model::GenericModel;\n    format::MOI.FileFormats.FileFormat = MOI.FileFormats.FORMAT_MOF,\n    kwargs...,\n)\n\nWrite the JuMP model model to io in the format format.\n\nOther kwargs are passed to the Model constructor of the chosen format.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.reset_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.reset_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.reset_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.reset_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.reset_optimizer","text":"MOIU.reset_optimizer(model::GenericModel)\n\nCall MOIU.reset_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.drop_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.drop_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.drop_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.drop_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.drop_optimizer","text":"MOIU.drop_optimizer(model::GenericModel)\n\nCall MOIU.drop_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"api/JuMP/#MOI.Utilities.attach_optimizer(::GenericModel)","page":"JuMP","title":"MOI.Utilities.attach_optimizer(::GenericModel)","text":"","category":"section"},{"location":"api/JuMP/","page":"JuMP","title":"JuMP","text":"JuMP.MOI.Utilities.attach_optimizer(::GenericModel)","category":"page"},{"location":"api/JuMP/#MathOptInterface.Utilities.attach_optimizer-Tuple{GenericModel}","page":"JuMP","title":"MathOptInterface.Utilities.attach_optimizer","text":"MOIU.attach_optimizer(model::GenericModel)\n\nCall MOIU.attach_optimizer on the backend of model.\n\nCannot be called in direct mode.\n\n\n\n\n\n","category":"method"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"EditURL = \"constraint_programming.jl\"","category":"page"},{"location":"tutorials/linear/constraint_programming/#Constraint-programming","page":"Constraint programming","title":"Constraint programming","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"JuMP supports a range of constraint-programming type constraints via the corresponding sets in MathOptInterface. For most constraints, there are reformulations built-in that convert the constraint programming constraint into a mixed-integer programming equivalent.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Because of this reformulation, all variables must be integer, and they must typically have finite bounds. An error will be thrown if the reformulation requires finiteness and you have a variable with non-finite bounds.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/constraint_programming/#AllDifferent","page":"Constraint programming","title":"AllDifferent","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.AllDifferent set ensures that every element in a list takes a different integer value.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 1 <= x[1:4] <= 4, Int)\n@constraint(model, x in MOI.AllDifferent(4))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#BinPacking","page":"Constraint programming","title":"BinPacking","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.BinPacking set can be used to divide up a set of items into different groups, such that the sum of their weights does not exceed the capacity of a bin.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"weights, capacity = Float64[1, 1, 2, 2, 3], 3.0;\nnumber_of_bins = 3\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 1 <= x[1:length(weights)] <= number_of_bins, Int)\n@constraint(model, x in MOI.BinPacking(capacity, weights))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Here, the value of x[i] is the bin that item i was placed into.","category":"page"},{"location":"tutorials/linear/constraint_programming/#Circuit","page":"Constraint programming","title":"Circuit","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.Circuit set is used to construct a tour of a list of N variables. They will each be assigned an integer from 1 to N, that describes the successor to each variable in the list:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[1:4], Int)\n@constraint(model, x in MOI.Circuit(4))\noptimize!(model)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Let's see what tour was found, starting at node number 1:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"y = round.(Int, value.(x))\ntour = Int[1]\nwhile length(tour) < length(y)\n    push!(tour, y[tour[end]])\nend\ntour","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountAtLeast","page":"Constraint programming","title":"CountAtLeast","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountAtLeast set is used to ensure that at least n elements in a set of variables belong to a set of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, here is a model with three variables, constrained between 0 and 5:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[1:3] <= 5, Int)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"If we want to ensure that at least one element of each set {x[1], x[2]} and {x[2], x[3]} is in the set {3}, then we create a list of variables by concatenating the sets together:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"variables = [x[1], x[2], x[2], x[3]]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Then we need a partition list that contains the number of elements in each set of variables:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"partitions = [2, 2]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Finally, we need a set of values that the elements must be a part of:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"values = Set([3])","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"And the number of elements that must be part of the set values:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"n = 1","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The constraint is:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"@constraint(model, variables in MOI.CountAtLeast(n, partitions, values))","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"To ensure the uniqueness of the solution, we'll add a constraint that x[2] must be <= 2. This ensures that the only feasible solution is for x[1] and x[3] to be 3:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"@constraint(model, x[2] <= 2)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Let's check that we found a valid solution:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"optimize!(model)\nvalue.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountBelongs","page":"Constraint programming","title":"CountBelongs","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountBelongs set is used to count how many elements in a set of variables belong to a set of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, to count how many elements in a set of 4 variables belong to the set {2, 3}, do:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@objective(model, Max, sum(x))\nset = Set([2, 3])\n@constraint(model, [n; x] in MOI.CountBelongs(1 + length(x), set))\noptimize!(model)\nvalue(n), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountDistinct","page":"Constraint programming","title":"CountDistinct","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountDistinct set is used to count the number of distinct elements in a set of variables.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@objective(model, Max, sum(x))\n@constraint(model, [n; x] in MOI.CountDistinct(1 + length(x)))\noptimize!(model)\nvalue(n), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/#CountGreaterThan","page":"Constraint programming","title":"CountGreaterThan","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.CountGreaterThan set is used to strictly upper-bound the number of distinct elements in a set of variables that have a value equal to another variable.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, to count the number n of times that y appears in the vector x, use:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, 0 <= x[i = 1:4] <= i, Int)\n@variable(model, n, Int)\n@variable(model, 3 <= y <= 4, Int)\n@objective(model, Max, sum(x))\n@constraint(model, [n; y; x] in MOI.CountGreaterThan(1 + 1 + length(x)))\noptimize!(model)\nvalue(n), value(y), value.(x)","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"Here n is strictly greater than the count, and there is no limit on how large n could be. For example, n = 100 is also a feasible solution. The only constraint is that n cannot be equal to or smaller than the number of times that y appears.","category":"page"},{"location":"tutorials/linear/constraint_programming/#Table","page":"Constraint programming","title":"Table","text":"","category":"section"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"The MOI.Table set is used to select a single row from a matrix of values.","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"For example, given a matrix:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"table = Float64[1 1 0; 0 1 1; 1 0 1; 1 1 1]","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"we can constraint a 3-element vector x to equal one of the rows in table via:","category":"page"},{"location":"tutorials/linear/constraint_programming/","page":"Constraint programming","title":"Constraint programming","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[i = 1:3], Int)\n@constraint(model, x in MOI.Table(table))\noptimize!(model)\nvalue.(x)","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"EditURL = \"https://github.com/oxfordcontrol/COSMO.jl/blob/v0.8.8/README.md\"","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<div style=\"display: flex; justify-content: flex-end\" margin=0px>\n    <img src=\"https://github.com/migarstka/COSMO_assets/blob/master/star_badge_3.png?raw=true\" align=\"right\" width=6%>\n</div>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<h1 align=\"center\" margin=0px>\n  <img src=\"https://github.com/migarstka/COSMO_assets/blob/master/cosmo_rocket_with_convex_set.png?raw=true\" width=70%>\n</h1>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n   <a href=\"https://github.com/oxfordcontrol/COSMO.jl/actions\"><img src=\"https://github.com/oxfordcontrol/COSMO.jl/workflows/ci/badge.svg?branch=master\"></a>\n  <a href=\"https://codecov.io/gh/oxfordcontrol/COSMO.jl\"><img src=\"https://codecov.io/gh/oxfordcontrol/COSMO.jl/branch/master/graph/badge.svg?raw=true\"></a>\n  <a href=\"https://oxfordcontrol.github.io/COSMO.jl/stable\"><img src=\"https://img.shields.io/badge/Documentation-stable-purple.svg?raw=true\"></a>\n  <a href=\"https://opensource.org/licenses/Apache-2.0\"><img src=\"https://img.shields.io/badge/License-Apache%202.0-blue.svg?raw=true\"></a>\n  <a href=\"https://github.com/oxfordcontrol/COSMO.jl/releases\"><img src=\"https://img.shields.io/badge/Release-v0.8.8-blue.svg?raw=true\"></a>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n  <a href=\"#features\">Features</a> •\n  <a href=\"#installation\">Installation</a> •\n  <a href=\"NEWS.md\">News</a> •\n  <a href=\"#citing-\">Citing</a> •\n  <a href=\"#contributing\">Contributing</a>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"This is a Julia implementation of the Conic operator splitting method (COSMO) solver. It can solve large convex conic optimization problems of the following form:","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<p align=\"center\">\n<img src=\"https://github.com/migarstka/COSMO_assets/blob/master/cosmo_format.png?raw=true\" width=220px>\n</p>","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"with decision variables x ϵ R^n, s ϵ R^m and data matrices P=P'>=0, q ϵ R^n, A ϵ R^(m×n), and b ϵ R^m. The convex set K is a composition of convex sets and cones.","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"For more information take a look at the COSMO.jl Documentation (stable |  dev).","category":"page"},{"location":"packages/COSMO/#Features","page":"oxfordcontrol/COSMO.jl","title":"Features","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"Versatile: COSMO solves linear programs, quadratic programs, second-order cone programs, semidefinite programs and problems involving exponential and power cones\nQuad SDPs: Positive semidefinite programs with quadratic objective functions are natively supported\nSafeguarded acceleration: robust and faster convergence to higher precision using COSMOAccelerators\nInfeasibility detection: Infeasible problems are detected without a homogeneous self-dual embedding of the problem\nJuMP / Convex.jl support: We provide an interface to MathOptInterface (MOI), which allows you to describe your problem in JuMP and Convex.jl.\nWarm starting: COSMO supports warm starting of the decision variables\nCustom sets and linear solver: Customize COSMO's components by defining your own convex constraint sets and by choosing from a number of direct and indirect linear system solvers, for example, QDLDL, Pardiso, Conjugate Gradient and MINRES\nArbitrary precision types: You can solve problems with any floating point precision.\nOpen Source: Our code is free to use and distributed under the Apache 2.0 Licence\nChordal decomposition: COSMO tries to decompose large structured PSD constraints using chordal decomposition techniques. This often results in a significant speedup compared to the original problem.\nSmart clique merging: After an initial decomposition of a structured SDP, COSMO recombines overlapping cliques/blocks to speed up the algorithm.","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"<div align=\"center\" margin=0px>\n  <img src=\"https://github.com/oxfordcontrol/COSMO.jl/blob/master/docs/src/assets/example_decomposition.gif?raw=true\" width=45%>\n</div>","category":"page"},{"location":"packages/COSMO/#Installation","page":"oxfordcontrol/COSMO.jl","title":"Installation","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"COSMO can be added via the Julia package manager (type ]): pkg> add COSMO","category":"page"},{"location":"packages/COSMO/#Citing","page":"oxfordcontrol/COSMO.jl","title":"Citing","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"If you find COSMO useful in your project, we kindly request that you cite the following paper:","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"@Article{Garstka_2021,\n  author  = {Michael Garstka and Mark Cannon and Paul Goulart},\n  journal = {Journal of Optimization Theory and Applications},\n  title   = {{COSMO}: A Conic Operator Splitting Method for Convex Conic Problems},\n  volume  = {190},\n  number  = {3},\n  pages   = {779--810},\n  year    = {2021},\n  publisher = {Springer},\n  doi     = {10.1007/s10957-021-01896-x},\n  url     = {https://doi.org/10.1007/s10957-021-01896-x}\n}","category":"page"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"The article is available under Open Access here.","category":"page"},{"location":"packages/COSMO/#Contributing","page":"oxfordcontrol/COSMO.jl","title":"Contributing","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"Contributions are always welcome. Our style guide can be found here.\nCurrent issues, tasks and future ideas are listed in Issues. Please report any issues or bugs that you encounter.\nAs an open source project we are also interested in any projects and applications that use COSMO. Please let us know by opening a GitHub issue.","category":"page"},{"location":"packages/COSMO/#Python-Interface","page":"oxfordcontrol/COSMO.jl","title":"Python - Interface","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"COSMO can also be called from Python. Take a look at: cosmo-python","category":"page"},{"location":"packages/COSMO/#Licence","page":"oxfordcontrol/COSMO.jl","title":"Licence 🔍","text":"","category":"section"},{"location":"packages/COSMO/","page":"oxfordcontrol/COSMO.jl","title":"oxfordcontrol/COSMO.jl","text":"This project is licensed under the Apache License - see the LICENSE.md file for details.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"EditURL = \"https://github.com/chriscoey/Hypatia.jl/blob/v0.7.3/README.md\"","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"<img src=\"https://github.com/chriscoey/Hypatia.jl/wiki/hypatia_logo.png?raw=true\" alt=\"Hypatia logo\" width=\"358\"/>","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"(Image: Build Status) (Image: codecov) (Image: ) (Image: )","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is a highly customizable open source interior point solver for generic conic optimization problems, written in Julia.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"For more information on Hypatia, please see:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"documentation for Hypatia's conic form, predefined cones, and interfaces\ncones reference for cone definitions and oracles\nexamples folder for applied examples and instances\nbenchmarks folder for scripts used to run and analyze various computational benchmarks","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"and preprints of our papers:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Solving natural conic formulations with Hypatia.jl for computational arguments for expanding the class of cones recognized by conic solvers\nPerformance enhancements for a generic conic interior point algorithm for a description of Hypatia's algorithm and our enhanced stepping procedures\nSum of squares generalizations for conic sets for barriers and computational techniques for our generalized polynomial sum of squares cones\nConic optimization with spectral functions on Euclidean Jordan algebras for barriers and computational techniques for many of our epigraph/hypograph cones","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"and corresponding raw results CSV files generated by our run scripts in the benchmarks folder.","category":"page"},{"location":"packages/Hypatia/#License","page":"chriscoey/Hypatia.jl","title":"License","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is licensed under the MIT License (see LICENSE).","category":"page"},{"location":"packages/Hypatia/#Installation","page":"chriscoey/Hypatia.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"To use Hypatia, install Julia, then at the Julia REPL, type:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"using Hypatia\nusing Pkg\nPkg.add(\"Hypatia\")","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia is an experimental solver and a work in progress, and may not run with older releases of Julia. Default options/parameters are not well-tuned, so we encourage you to experiment with these.","category":"page"},{"location":"packages/Hypatia/#Usage","page":"chriscoey/Hypatia.jl","title":"Usage","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Hypatia can be accessed through a low-level native Julia interface or through open-source modeling tools such as JuMP and Convex.jl. The native interface is more expressive, allowing Hypatia to solve conic models expressed with generic real floating point types and structured matrices or linear operators, for example. However, it is typically sufficient and more convenient to use JuMP.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Using JuMP, we can model a simple D-optimal experiment design problem and call Hypatia:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"using LinearAlgebra\nusing JuMP\nusing Hypatia\n\nmodel = Model(() -> Hypatia.Optimizer(verbose = false))\n@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) == 5)\n@variable(model, hypo)\n@objective(model, Max, hypo)\nV = rand(2, 3)\nQ = V * diagm(x) * V'\naff = vcat(hypo, [Q[i, j] for i in 1:2 for j in 1:i]...)\n@constraint(model, aff in MOI.RootDetConeTriangle(2))\n\n# solve and query solution\noptimize!(model)\ntermination_status(model)\nobjective_value(model)\nvalue.(x)","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"See our D-optimal design example for more information and references.","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Many more examples using the native interface or JuMP can be found in the examples folder.","category":"page"},{"location":"packages/Hypatia/#Contributing","page":"chriscoey/Hypatia.jl","title":"Contributing","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"Comments, questions, suggestions, and improvements/extensions to the code or documentation are welcomed. Please reach out on Discourse, or submit an issue or contribute a PR on our GitHub. If contributing code, try to maintain consistent style and add docstrings or comments for clarity. New examples are welcomed and should be implemented similarly to the existing examples.","category":"page"},{"location":"packages/Hypatia/#Acknowledgements","page":"chriscoey/Hypatia.jl","title":"Acknowledgements","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"This work has been partially funded by the National Science Foundation under grant OAC-1835443 and the Office of Naval Research under grant N00014-18-1-2079.","category":"page"},{"location":"packages/Hypatia/#Citing-Hypatia","page":"chriscoey/Hypatia.jl","title":"Citing Hypatia","text":"","category":"section"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"If you find Hypatia solver useful, please cite our solver paper:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"@article{coey2022solving,\n    title={Solving natural conic formulations with {H}ypatia.jl},\n    author={Chris Coey and Lea Kapelevich and Juan Pablo Vielma},\n    year={2022},\n    journal={INFORMS Journal on Computing},\n    publisher={INFORMS},\n    volume={34},\n    number={5},\n    pages={2686--2699},\n    doi={https://doi.org/10.1287/ijoc.2022.1202}\n}","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"If you find aspects of Hypatia's IPM implementation useful, please cite our algorithm paper:","category":"page"},{"location":"packages/Hypatia/","page":"chriscoey/Hypatia.jl","title":"chriscoey/Hypatia.jl","text":"@article{coey2022performance,\n    title={Performance enhancements for a generic conic interior point algorithm},\n    author={Chris Coey and Lea Kapelevich and Juan Pablo Vielma},\n    year={2023},\n    journal={Mathematical Programming Computation},\n    publisher={Springer},\n    volume={15},\n    pages={53--101},\n    doi={https://doi.org/10.1007/s12532-022-00226-0}\n}","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"EditURL = \"arbitrary_precision.jl\"","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Arbitrary-precision-arithmetic","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The purpose of this tutorial is to explain how to use a solver which supports arithmetic using a number type other than Float64.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"using JuMP\nimport CDDLib\nimport Clarabel","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Higher-precision-arithmetic","page":"Arbitrary precision arithmetic","title":"Higher-precision arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"To create a model with a number type other than Float64, use GenericModel with an optimizer which supports the same number type:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"model = GenericModel{BigFloat}(Clarabel.Optimizer{BigFloat})","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding decision variables is the same as a normal JuMP model, except that values are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@variable(model, -1 <= x[1:2] <= sqrt(big\"2\"))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Note that each x is now a GenericVariableRef{BigFloat}, which means that the value of x in a solution will be a BigFloat.","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The lower and upper bounds of the decision variables are also BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"lower_bound(x[1])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(lower_bound(x[1]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"upper_bound(x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(upper_bound(x[2]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding constraints is the same as a normal JuMP model, except that coefficients are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c, x[1] == big\"2\" * x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The function is a GenericAffExpr with BigFloat for the coefficient and variable types;","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"constraint = constraint_object(c)\ntypeof(constraint.func)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and the set is a MOI.EqualTo{BigFloat}:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(constraint.set)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The syntax for adding and objective is the same as a normal JuMP model, except that coefficients are converted to BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@objective(model, Min, 3x[1]^2 + 2x[2]^2 - x[1] - big\"4\" * x[2])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(objective_function(model))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Here's the model we have built:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"print(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Let's solve and inspect the solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The value of each decision variable is a BigFloat:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"as well as other solution attributes like the objective value:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and dual solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"dual(c)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"This problem has an analytic solution of x = [3//7, 3//14]. Currently, our solution has an error of approximately 1e-9:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x) .- [3 // 7, 3 // 14]","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"But by reducing the tolerances, we can obtain a more accurate solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"set_attribute(model, \"tol_gap_abs\", 1e-32)\nset_attribute(model, \"tol_gap_rel\", 1e-32)\noptimize!(model)\nvalue.(x) .- [3 // 7, 3 // 14]","category":"page"},{"location":"tutorials/conic/arbitrary_precision/#Rational-arithmetic","page":"Arbitrary precision arithmetic","title":"Rational arithmetic","text":"","category":"section"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"In addition to higher-precision floating point number types like BigFloat, JuMP also supports solvers with exact rational arithmetic. One example is CDDLib.jl, which supports the Rational{BigInt} number type:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"model = GenericModel{Rational{BigInt}}(CDDLib.Optimizer{Rational{BigInt}})","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"As before, we can create variables using rational bounds:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@variable(model, 1 // 7 <= x[1:2] <= 2 // 3)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"lower_bound(x[1])","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"typeof(lower_bound(x[1]))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"As well as constraints:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c1, (2 // 1) * x[1] + x[2] <= 1)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@constraint(model, c2, x[1] + 3x[2] <= 9 // 4)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"and objective functions:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"@objective(model, Max, sum(x))","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Here's the model we have built:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"print(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"Let's solve and inspect the solution:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"The optimal values are given in exact rational arithmetic:","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value.(x)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/arbitrary_precision/","page":"Arbitrary precision arithmetic","title":"Arbitrary precision arithmetic","text":"value(c2)","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/standard_form.md\"","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/standard_form/#Standard-form-problem","page":"Standard form problem","title":"Standard form problem","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"MathOptInterface represents optimization problems in the standard form:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"where:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"the functions f_0 f_1 ldots f_m are specified by AbstractFunction objects\nthe sets mathcalS_1 ldots mathcalS_m are specified by AbstractSet objects","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"tip: Tip\nFor more information on this standard form, read our paper.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"MOI defines some commonly used functions and sets, but the interface is extensible to other sets recognized by the solver.","category":"page"},{"location":"moi/manual/standard_form/#Functions","page":"Standard form problem","title":"Functions","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The function types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Function Description\nVariableIndex x_j, the projection onto a single coordinate defined by a variable index j.\nVectorOfVariables The projection onto multiple coordinates (that is, extracting a sub-vector).\nScalarAffineFunction a^T x + b, where a is a vector and b scalar.\nScalarNonlinearFunction f(x), where f is a nonlinear function.\nVectorAffineFunction A x + b, where A is a matrix and b is a vector.\nScalarQuadraticFunction frac12 x^T Q x + a^T x + b, where Q is a symmetric matrix, a is a vector, and b is a constant.\nVectorQuadraticFunction A vector of scalar-valued quadratic functions.\nVectorNonlinearFunction f(x), where f is a vector-valued nonlinear function.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Extensions for nonlinear programming are present but not yet well documented.","category":"page"},{"location":"moi/manual/standard_form/#One-dimensional-sets","page":"Standard form problem","title":"One-dimensional sets","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The one-dimensional set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nLessThan(u) (-infty u\nGreaterThan(l) l infty)\nEqualTo(v) v\nInterval(l, u) l u\nInteger() mathbbZ\nZeroOne()  0 1 \nSemicontinuous(l, u)  0 cup l u\nSemiinteger(l, u)  0 cup ll+1ldotsu-1u","category":"page"},{"location":"moi/manual/standard_form/#Vector-cones","page":"Standard form problem","title":"Vector cones","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The vector-valued set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nReals(d) mathbbR^d\nZeros(d) 0^d\nNonnegatives(d)  x in mathbbR^d  x ge 0 \nNonpositives(d)  x in mathbbR^d  x le 0 \nSecondOrderCone(d)  (tx) in mathbbR^d  t ge lVert x rVert_2 \nRotatedSecondOrderCone(d)  (tux) in mathbbR^d  2tu ge lVert x rVert_2^2 t ge 0u ge 0 \nExponentialCone()  (xyz) in mathbbR^3  y exp (xy) le z y  0 \nDualExponentialCone()  (uvw) in mathbbR^3  -u exp (vu) le exp(1) w u  0 \nGeometricMeanCone(d)  (tx) in mathbbR^1+n  x ge 0 t le sqrtnx_1 x_2 cdots x_n  where n is d - 1\nPowerCone(α)  (xyz) in mathbbR^3  x^alpha y^1-alpha ge z x ge 0y ge 0 \nDualPowerCone(α)  (uvw) in mathbbR^3  left(fracualpharight(^alphaleft(fracv1-alpharight)^1-alpha ge w uv ge 0 \nNormOneCone(d)  (tx) in mathbbR^d  t ge sum_i lvert x_i rvert \nNormInfinityCone(d)  (tx) in mathbbR^d  t ge max_i lvert x_i rvert \nRelativeEntropyCone(d)  (u v w) in mathbbR^d  u ge sum_i w_i log (fracw_iv_i) v_i ge 0 w_i ge 0 \nHyperRectangle(l, u) x in barmathbbR^d x_i in l_i u_i forall i=1ldotsd\nNormCone(p, d) ``{ (t,x) \\in \\mathbb{R}^{d} : t \\ge \\left(\\sum\\limits_i","category":"page"},{"location":"moi/manual/standard_form/#Matrix-cones","page":"Standard form problem","title":"Matrix cones","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"The matrix-valued set types implemented in MathOptInterface.jl are:","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nRootDetConeTriangle(d)  (tX) in mathbbR^1+d(1+d)2  t le det(X)^1d X mbox is the upper triangle of a PSD matrix \nRootDetConeSquare(d)  (tX) in mathbbR^1+d^2  t le det(X)^1d X mbox is a PSD matrix \nPositiveSemidefiniteConeTriangle(d)  X in mathbbR^d(d+1)2  X mbox is the upper triangle of a PSD matrix \nPositiveSemidefiniteConeSquare(d)  X in mathbbR^d^2  X mbox is a PSD matrix \nLogDetConeTriangle(d)  (tuX) in mathbbR^2+d(1+d)2  t le ulog(det(Xu)) X mbox is the upper triangle of a PSD matrix u  0  \nLogDetConeSquare(d)  (tuX) in mathbbR^2+d^2  t le u log(det(Xu)) X mbox is a PSD matrix u  0 \nNormSpectralCone(r, c)  (t X) in mathbbR^1 + r times c  t ge sigma_1(X) X mbox is a  rtimes cmbox matrix \nNormNuclearCone(r, c)  (t X) in mathbbR^1 + r times c  t ge sum_i sigma_i(X) X mbox is a  rtimes cmbox matrix \nHermitianPositiveSemidefiniteConeTriangle(d) The cone of Hermitian positive semidefinite matrices, with\nside_dimension rows and columns. \nScaled(S) The set S scaled so that Utilities.set_dot corresponds to LinearAlgebra.dot","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Some of these cones can take two forms: XXXConeTriangle and XXXConeSquare.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"In XXXConeTriangle sets, the matrix is assumed to be symmetric, and the elements are provided by a vector, in which the entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row).","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"In XXXConeSquare sets, the entries of the matrix are given column by column (or equivalently, row by row), and the matrix is constrained to be symmetric. As an example, given a 2-by-2 matrix of variables X and a one-dimensional variable t, we can specify a root-det constraint as [t, X11, X12, X22] ∈ RootDetConeTriangle or [t, X11, X12, X21, X22] ∈ RootDetConeSquare.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"We provide both forms to enable flexibility for solvers who may natively support one or the other. Transformations between XXXConeTriangle and XXXConeSquare are handled by bridges, which removes the chance of conversion mistakes by users or solver developers.","category":"page"},{"location":"moi/manual/standard_form/#Multi-dimensional-sets-with-combinatorial-structure","page":"Standard form problem","title":"Multi-dimensional sets with combinatorial structure","text":"","category":"section"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Other sets are vector-valued, with a particular combinatorial structure. Read their docstrings for more information on how to interpret them.","category":"page"},{"location":"moi/manual/standard_form/","page":"Standard form problem","title":"Standard form problem","text":"Set Description\nSOS1 A Special Ordered Set (SOS) of Type I\nSOS2 A Special Ordered Set (SOS) of Type II\nIndicator A set to specify an indicator constraint\nComplements A set to specify a mixed complementarity constraint\nAllDifferent The all_different global constraint\nBinPacking The bin_packing global constraint\nCircuit The circuit global constraint\nCountAtLeast The at_least global constraint\nCountBelongs The nvalue global constraint\nCountDistinct The distinct global constraint\nCountGreaterThan The count_gt global constraint\nCumulative The cumulative global constraint\nPath The path global constraint\nTable The table global constraint","category":"page"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/variables.md\"","category":"page"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/variables/#Variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"moi/reference/variables/#Functions","page":"Variables","title":"Functions","text":"","category":"section"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"add_variable\nadd_variables\nadd_constrained_variable\nadd_constrained_variables\nsupports_add_constrained_variable\nsupports_add_constrained_variables\nis_valid(::ModelLike,::VariableIndex)\ndelete(::ModelLike, ::VariableIndex)\ndelete(::ModelLike, ::Vector{VariableIndex})","category":"page"},{"location":"moi/reference/variables/#MathOptInterface.add_variable","page":"Variables","title":"MathOptInterface.add_variable","text":"add_variable(model::ModelLike)::VariableIndex\n\nAdd a scalar variable to the model, returning a variable index.\n\nA AddVariableNotAllowed error is thrown if adding variables cannot be done in the current state of the model model.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_variables","page":"Variables","title":"MathOptInterface.add_variables","text":"add_variables(model::ModelLike, n::Int)::Vector{VariableIndex}\n\nAdd n scalar variables to the model, returning a vector of variable indices.\n\nA AddVariableNotAllowed error is thrown if adding variables cannot be done in the current state of the model model.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_constrained_variable","page":"Variables","title":"MathOptInterface.add_constrained_variable","text":"add_constrained_variable(\n    model::ModelLike,\n    set::AbstractScalarSet\n)::Tuple{MOI.VariableIndex,\n         MOI.ConstraintIndex{MOI.VariableIndex, typeof(set)}}\n\nAdd to model a scalar variable constrained to belong to set, returning the index of the variable created and the index of the constraint constraining the variable to belong to set.\n\nBy default, this function falls back to creating a free variable with add_variable and then constraining it to belong to set with add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.add_constrained_variables","page":"Variables","title":"MathOptInterface.add_constrained_variables","text":"add_constrained_variables(\n    model::ModelLike,\n    sets::AbstractVector{<:AbstractScalarSet}\n)::Tuple{\n    Vector{MOI.VariableIndex},\n    Vector{MOI.ConstraintIndex{MOI.VariableIndex,eltype(sets)}},\n}\n\nAdd to model scalar variables constrained to belong to sets, returning the indices of the variables created and the indices of the constraints constraining the variables to belong to each set in sets. That is, if it returns variables and constraints, constraints[i] is the index of the constraint constraining variable[i] to belong to sets[i].\n\nBy default, this function falls back to calling add_constrained_variable on each set.\n\n\n\n\n\nadd_constrained_variables(\n    model::ModelLike,\n    set::AbstractVectorSet,\n)::Tuple{\n    Vector{MOI.VariableIndex},\n    MOI.ConstraintIndex{MOI.VectorOfVariables,typeof(set)},\n}\n\nAdd to model a vector of variables constrained to belong to set, returning the indices of the variables created and the index of the constraint constraining the vector of variables to belong to set.\n\nBy default, this function falls back to creating free variables with add_variables and then constraining it to belong to set with add_constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.supports_add_constrained_variable","page":"Variables","title":"MathOptInterface.supports_add_constrained_variable","text":"supports_add_constrained_variable(\n    model::ModelLike,\n    S::Type{<:AbstractScalarSet}\n)::Bool\n\nReturn a Bool indicating whether model supports constraining a variable to belong to a set of type S either on creation of the variable with add_constrained_variable or after the variable is created with add_constraint.\n\nBy default, this function falls back to supports_add_constrained_variables(model, Reals) && supports_constraint(model, MOI.VariableIndex, S) which is the correct definition for most models.\n\nExample\n\nSuppose that a solver supports only two kind of variables: binary variables and continuous variables with a lower bound. If the solver decides not to support VariableIndex-in-Binary and VariableIndex-in-GreaterThan constraints, it only has to implement add_constrained_variable for these two sets which prevents the user to add both a binary constraint and a lower bound on the same variable. Moreover, if the user adds a VariableIndex-in-GreaterThan constraint, implementing this interface (i.e., supports_add_constrained_variables) enables the constraint to be transparently bridged into a supported constraint.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.supports_add_constrained_variables","page":"Variables","title":"MathOptInterface.supports_add_constrained_variables","text":"supports_add_constrained_variables(\n    model::ModelLike,\n    S::Type{<:AbstractVectorSet}\n)::Bool\n\nReturn a Bool indicating whether model supports constraining a vector of variables to belong to a set of type S either on creation of the vector of variables with add_constrained_variables or after the variable is created with add_constraint.\n\nBy default, if S is Reals then this function returns true and otherwise, it falls back to supports_add_constrained_variables(model, Reals) && supports_constraint(model, MOI.VectorOfVariables, S) which is the correct definition for most models.\n\nExample\n\nIn the standard conic form (see Duality), the variables are grouped into several cones and the constraints are affine equality constraints. If Reals is not one of the cones supported by the solvers then it needs to implement supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false as free variables are not supported. The solvers should then implement supports_add_constrained_variables(::Optimizer, ::Type{<:SupportedCones}) = true where SupportedCones is the union of all cone types that are supported; it does not have to implement the method supports_constraint(::Type{VectorOfVariables}, Type{<:SupportedCones}) as it should return false and it's the default. This prevents the user to constrain the same variable in two different cones. When a VectorOfVariables-in-S is added, the variables of the vector have already been created so they already belong to given cones. If bridges are enabled, the constraint will therefore be bridged by adding slack variables in S and equality constraints ensuring that the slack variables are equal to the corresponding variables of the given constraint function.\n\nNote that there may also be sets for which !supports_add_constrained_variables(model, S) and supports_constraint(model, MOI.VectorOfVariables, S). For instance, suppose a solver supports positive semidefinite variable constraints and two types of variables: binary variables and nonnegative variables. Then the solver should support adding VectorOfVariables-in-PositiveSemidefiniteConeTriangle constraints, but it should not support creating variables constrained to belong to the PositiveSemidefiniteConeTriangle because the variables in PositiveSemidefiniteConeTriangle should first be created as either binary or non-negative.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/variables/#MathOptInterface.is_valid-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}","page":"Variables","title":"MathOptInterface.is_valid","text":"is_valid(model::ModelLike, index::Index)::Bool\n\nReturn a Bool indicating whether this index refers to a valid object in the model model.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, MathOptInterface.VariableIndex}","page":"Variables","title":"MathOptInterface.delete","text":"delete(model::ModelLike, index::Index)\n\nDelete the referenced object from the model. Throw DeleteNotAllowed if if index cannot be deleted.\n\nThe following modifications also take effect if Index is VariableIndex:\n\nIf index used in the objective function, it is removed from the function, i.e., it is substituted for zero.\nFor each func-in-set constraint of the model:\nIf func isa VariableIndex and func == index then the constraint is deleted.\nIf func isa VectorOfVariables and index in func.variables then\nif length(func.variables) == 1 is one, the constraint is deleted;\nif length(func.variables) > 1 and supports_dimension_update(set) then then the variable is removed from func and set is replaced by update_dimension(set, MOI.dimension(set) - 1).\nOtherwise, a DeleteNotAllowed error is thrown.\nOtherwise, the variable is removed from func, i.e., it is substituted for zero.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#MathOptInterface.delete-Tuple{MathOptInterface.ModelLike, Vector{MathOptInterface.VariableIndex}}","page":"Variables","title":"MathOptInterface.delete","text":"delete(model::ModelLike, indices::Vector{R<:Index}) where {R}\n\nDelete the referenced objects in the vector indices from the model. It may be assumed that R is a concrete type. The default fallback sequentially deletes the individual items in indices, although specialized implementations may be more efficient.\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/variables/#Attributes","page":"Variables","title":"Attributes","text":"","category":"section"},{"location":"moi/reference/variables/","page":"Variables","title":"Variables","text":"AbstractVariableAttribute\nVariableName\nVariablePrimalStart\nVariablePrimal\nVariableBasisStatus","category":"page"},{"location":"moi/reference/variables/#MathOptInterface.AbstractVariableAttribute","page":"Variables","title":"MathOptInterface.AbstractVariableAttribute","text":"AbstractVariableAttribute\n\nAbstract supertype for attribute objects that can be used to set or get attributes (properties) of variables in the model.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariableName","page":"Variables","title":"MathOptInterface.VariableName","text":"VariableName()\n\nA variable attribute for a string identifying the variable. It is valid for two variables to have the same name; however, variables with duplicate names cannot be looked up using get. It has a default value of \"\" if not set`.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariablePrimalStart","page":"Variables","title":"MathOptInterface.VariablePrimalStart","text":"VariablePrimalStart()\n\nA variable attribute for the initial assignment to some primal variable's value that the optimizer may use to warm-start the solve. May be a number or nothing (unset).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariablePrimal","page":"Variables","title":"MathOptInterface.VariablePrimal","text":"VariablePrimal(result_index::Int = 1)\n\nA variable attribute for the assignment to some primal variable's value in result result_index. If result_index is omitted, it is 1 by default.\n\nIf the solver does not have a primal value for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the VariablePrimal attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/variables/#MathOptInterface.VariableBasisStatus","page":"Variables","title":"MathOptInterface.VariableBasisStatus","text":"VariableBasisStatus(result_index::Int = 1)\n\nA variable attribute for the BasisStatusCode of a variable in result result_index, with respect to an available optimal solution basis.\n\nIf the solver does not have a basis statue for the variable because the result_index is beyond the available solutions (whose number is indicated by the ResultCount attribute), getting this attribute must throw a ResultIndexBoundsError. Otherwise, if the result is unavailable for another reason (for instance, only a dual solution is available), the result is undefined. Users should first check PrimalStatus before accessing the VariableBasisStatus attribute.\n\nSee ResultCount for information on how the results are ordered.\n\n\n\n\n\n","category":"type"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"EditURL = \"https://github.com/jump-dev/Ipopt.jl/blob/v1.5.0/README.md\"","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"(Image: )","category":"page"},{"location":"packages/Ipopt/#Ipopt.jl","page":"jump-dev/Ipopt.jl","title":"Ipopt.jl","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl is a wrapper for the Ipopt solver.","category":"page"},{"location":"packages/Ipopt/#Affiliation","page":"jump-dev/Ipopt.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/Ipopt/#License","page":"jump-dev/Ipopt.jl","title":"License","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The underlying solver, coin-or/Ipopt, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/Ipopt/#Installation","page":"jump-dev/Ipopt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Install Ipopt.jl using the Julia package manager:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import Pkg\nPkg.add(\"Ipopt\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"In addition to installing the Ipopt.jl package, this will also download and install the Ipopt binaries. You do not need to install Ipopt separately.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"For details on using a different linear solver, see the Linear Solvers section below. You do not need a custom binary to change the linear solver.","category":"page"},{"location":"packages/Ipopt/#Use-with-JuMP","page":"jump-dev/Ipopt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"You can use Ipopt with JuMP as follows:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"max_cpu_time\", 60.0)\nset_attribute(model, \"print_level\", 0)","category":"page"},{"location":"packages/Ipopt/#MathOptInterface-API","page":"jump-dev/Ipopt.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The Ipopt optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"MOI.BarrierIterations\nMOI.NLPBlock\nMOI.NLPBlockDualStart\nMOI.Name\nMOI.ObjectiveSense\nMOI.SolveTimeSec","category":"page"},{"location":"packages/Ipopt/#Options","page":"jump-dev/Ipopt.jl","title":"Options","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Supported options are listed in the Ipopt documentation.","category":"page"},{"location":"packages/Ipopt/#Solver-specific-callbacks","page":"jump-dev/Ipopt.jl","title":"Solver-specific callbacks","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt provides a callback that can be used to log the status of the optimization during a solve. It can also be used to terminate the optimization by returning false. Here is an example:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt, Test\nmodel = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x >= 1)\n@objective(model, Min, x + 0.5)\nx_vals = Float64[]\nfunction my_callback(\n   alg_mod::Cint,\n   iter_count::Cint,\n   obj_value::Float64,\n   inf_pr::Float64,\n   inf_du::Float64,\n   mu::Float64,\n   d_norm::Float64,\n   regularization_size::Float64,\n   alpha_du::Float64,\n   alpha_pr::Float64,\n   ls_trials::Cint,\n)\n   push!(x_vals, callback_value(model, x))\n   @test isapprox(obj_value, 1.0 * x_vals[end] + 0.5, atol = 1e-1)\n   # return `true` to keep going, or `false` to terminate the optimization.\n   return iter_count < 1\nend\nMOI.set(model, Ipopt.CallbackFunction(), my_callback)\noptimize!(model)\n@test MOI.get(model, MOI.TerminationStatus()) == MOI.INTERRUPTED\n@test length(x_vals) == 2","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"See the Ipopt documentation for an explanation of the arguments to the callback. They are identical to the output contained in the logging table printed to the screen.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To access the current solution and primal, dual, and complementarity violations of each iteration, use Ipopt.GetIpoptCurrentViolations and Ipopt.GetIpoptCurrentIterate. The two functions are identical to the ones in the Ipopt C interface.","category":"page"},{"location":"packages/Ipopt/#C-API","page":"jump-dev/Ipopt.jl","title":"C API","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Ipopt.jl wraps the Ipopt C interface with minimal modifications.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"A complete example is available in the test/C_wrapper.jl file.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"For simplicity, the five callbacks required by Ipopt are slightly different to the C interface. They are as follows:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"\"\"\"\n   eval_f(x::Vector{Float64})::Float64\n\nReturns the objective value `f(x)`.\n\"\"\"\nfunction eval_f end\n\n\"\"\"\n   eval_grad_f(x::Vector{Float64}, grad_f::Vector{Float64})::Nothing\n\nFills `grad_f` in-place with the gradient of the objective function evaluated at\n`x`.\n\"\"\"\nfunction eval_grad_f end\n\n\"\"\"\n   eval_g(x::Vector{Float64}, g::Vector{Float64})::Nothing\n\nFills `g` in-place with the value of the constraints evaluated at `x`.\n\"\"\"\nfunction eval_g end\n\n\"\"\"\n   eval_jac_g(\n      x::Vector{Float64},\n      rows::Vector{Cint},\n      cols::Vector{Cint},\n      values::Union{Nothing,Vector{Float64}},\n   )::Nothing\n\nCompute the Jacobian matrix.\n\n* If `values === nothing`\n   - Fill `rows` and `cols` with the 1-indexed sparsity structure\n* Otherwise:\n   - Fill `values` with the elements of the Jacobian matrix according to the\n     sparsity structure.\n\n!!! warning\n    If `values === nothing`, `x` is an undefined object. Accessing any elements\n    in it will cause Julia to segfault.\n\"\"\"\nfunction eval_jac_g end\n\n\"\"\"\n   eval_h(\n      x::Vector{Float64},\n      rows::Vector{Cint},\n      cols::Vector{Cint},\n      obj_factor::Float64,\n      lambda::Float64,\n      values::Union{Nothing,Vector{Float64}},\n   )::Nothing\n\nCompute the Hessian-of-the-Lagrangian matrix.\n\n* If `values === nothing`\n   - Fill `rows` and `cols` with the 1-indexed sparsity structure\n* Otherwise:\n   - Fill `values` with the Hessian matrix according to the sparsity structure.\n\n!!! warning\n    If `values === nothing`, `x` is an undefined object. Accessing any elements\n    in it will cause Julia to segfault.\n\"\"\"\nfunction eval_h end","category":"page"},{"location":"packages/Ipopt/#INVALID_MODEL-error","page":"jump-dev/Ipopt.jl","title":"INVALID_MODEL error","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you get a termination status MOI.INVALID_MODEL, it is probably because you have some undefined value in your model, for example, a division by zero. Fix this by removing the division, or by imposing variable bounds so that you cut off the undefined region.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Instead of","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x)\n@NLobjective(model, 1 / x)","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"do","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x >= 0.0001)\n@NLobjective(model, 1 / x)","category":"page"},{"location":"packages/Ipopt/#Linear-Solvers","page":"jump-dev/Ipopt.jl","title":"Linear Solvers","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"To improve performance, Ipopt supports a number of linear solvers.","category":"page"},{"location":"packages/Ipopt/#HSL","page":"jump-dev/Ipopt.jl","title":"HSL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Obtain a license and download HSL_jll.jl from https://licences.stfc.ac.uk/product/julia-hsl.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"There are two versions available: LBT and OpenBLAS. LBT is the recommended option for Julia ≥ v1.9.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Install this download into your current environment using:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import Pkg\nPkg.develop(path = \"/full/path/to/HSL_jll.jl\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Then, use a linear solver in HSL by setting the hsllib and linear_solver attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nimport HSL_jll\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"hsllib\", HSL_jll.libhsl_path)\nset_attribute(model, \"linear_solver\", \"ma86\")","category":"page"},{"location":"packages/Ipopt/#macOS-users","page":"jump-dev/Ipopt.jl","title":"macOS users","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Due to the security policy of macOS, Mac users may need to delete the quarantine attribute of the ZIP archive before extracting. For example:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"xattr -d com.apple.quarantine lbt_HSL_jll.jl-2023.5.26.zip\nxattr -d com.apple.quarantine openblas_HSL_jll.jl-2023.5.26.zip","category":"page"},{"location":"packages/Ipopt/#Pardiso","page":"jump-dev/Ipopt.jl","title":"Pardiso","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Download Pardiso from https://www.pardiso-project.org. Save the shared library somewhere, and record the filename.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Then, use Pardiso by setting the pardisolib and linear_solver attributes:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"pardisolib\", \"/full/path/to/libpardiso\")\nset_attribute(model, \"linear_solver\", \"pardiso\")","category":"page"},{"location":"packages/Ipopt/#SPRAL","page":"jump-dev/Ipopt.jl","title":"SPRAL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you use Ipopt.jl with Julia ≥ v1.9, the linear solver SPRAL is available. You can use it by setting the linear_solver attribute:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_attribute(model, \"linear_solver\", \"spral\")","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Note that the following environment variables must be set before starting Julia:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"export OMP_CANCELLATION=TRUE\nexport OMP_PROC_BIND=TRUE","category":"page"},{"location":"packages/Ipopt/#BLAS-and-LAPACK","page":"jump-dev/Ipopt.jl","title":"BLAS and LAPACK","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"With Julia v1.9 or later, Ipopt and the linear solvers MUMPS (default), SPRAL, and HSL are compiled with libblastrampoline (LBT), a library that can change between BLAS and LAPACK backends at runtime.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"The default BLAS and LAPACK backend is OpenBLAS.","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Using LBT, we can also switch dynamically to other BLAS backends such as Intel MKL and Apple Accelerate. Because Ipopt and the linear solvers heavily rely on BLAS and LAPACK routines, using an optimized backend for a particular platform can improve the performance.","category":"page"},{"location":"packages/Ipopt/#MKL","page":"jump-dev/Ipopt.jl","title":"MKL","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you have MKL.jl installed, switch to MKL by adding using MKL to your code:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using MKL  # Replace OpenBLAS by Intel MKL\nusing Ipopt","category":"page"},{"location":"packages/Ipopt/#AppleAccelerate","page":"jump-dev/Ipopt.jl","title":"AppleAccelerate","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"If you are using macOS ≥ v13.4 and you have AppleAccelerate.jl installed, add using AppleAccelerate to your code:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"using AppleAccelerate  # Replace OpenBLAS by Apple Accelerate\nusing Ipopt","category":"page"},{"location":"packages/Ipopt/#Display-backends","page":"jump-dev/Ipopt.jl","title":"Display backends","text":"","category":"section"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"Check what backends are loaded using:","category":"page"},{"location":"packages/Ipopt/","page":"jump-dev/Ipopt.jl","title":"jump-dev/Ipopt.jl","text":"import LinearAlgebra\nLinearAlgebra.BLAS.lbt_get_config()","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/release_notes.md\"","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"CurrentModule = MathOptInterface","category":"page"},{"location":"moi/release_notes/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"moi/release_notes/#[v1.20.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.20.1)-(September-24,-2023)","page":"Release notes","title":"v1.20.1 (September 24, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Removed incorrect kwargs in some copy_to methods (#2272)\nFixed ConstraintDualStart for Bridges.Constraint.SplitIntervalBridge (#2275)\nFixed MethodError when trying to modify a variable objective (#2278)\nFixed stack-overflow in Utilities.operate(+,...) with many arguments (#2285) (#2290)","category":"page"},{"location":"moi/release_notes/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MathOptSetDistances to solver-tests.yml (#2265)\nUpdated Documenter (#2266)\nFixed various JET errors (#2267) (#2269) (#2270) (#2271) (#2276) (#2277) (#2289)\nVarious style improvements\nReplaced using Package with import Package where possible (#2274)\nRemoved Utilities.EMPTYSTRING (#2283)\nRemoved unnecessary const acronyms in Utilities (#2280) (#2281)\nRemoved invalid and unused method (#2286)\nRefactored src/Utilities/model.jl (#2287)","category":"page"},{"location":"moi/release_notes/#[v1.20.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.20.0)-(September-7,-2023)","page":"Release notes","title":"v1.20.0 (September 7, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Scaled set (#2237) (#2263) (#2264)\nAdded ObjectiveLimit attribute (#2257)","category":"page"},{"location":"moi/release_notes/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Updated dependencies (#2258)\nImproved performance of ScalarNonlinearFunction utilities (#2259)\nFixed docstrings (#2261)","category":"page"},{"location":"moi/release_notes/#[v1.19.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.19.0)-(August-15,-2023)","page":"Release notes","title":"v1.19.0 (August 15, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added VectorNonlinearFunction (#2201)\nAdded new bridges\nBridges.Constraint.IntegerToZeroOneBridge (#2205)\nBridges.Constraint.ToScalarQuadraticBridge (#2235)\nBridges.Constraint.ToVectorQuadraticBridge (#2235)\nBridges.Constraint.ToScalarNonlinearBridge (#2233) (#2235)\nBridges.Constraint.FunctionConversionBridge (#2235)\nAdded Bridges.bridging_cost for setting a bridge-specific cost in the bridging graph (#2235)\nAdded Utilities.eval_variables support for ScalarNonlinearFunction (#2218) (#2219)\nAdded support for ScalarNonlinearFunction in FileFormats.NL (#2228) (#2231)\nAdded support for writing non-Float64 functions in FileFormats.MOF\nAdded Utilities.lazy_map (#2254)","category":"page"},{"location":"moi/release_notes/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed method ambiguities in operate (#2224)\nFixed reading LP file with a -infinity <= x <= +infinity variable (#2225)\nFixed missing require in Test.test_nonlinear_duals (#2230)\nFixed bug in ConstraintPrimal of Bridges.Constraint.QuadtoSOCBridge (#2240)","category":"page"},{"location":"moi/release_notes/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added extensions to solver-tests.yml (#2229)\nRefactored test/Benchmarks (#2234)\nFixed warnings in tests (#2241) (#2243)\nSmall refactoring of bridges for upcoming VectorNonlinearFunction (#2244) (#2245)\nFixed various typos (#2251) (#2255)\nPartitioned how we run the tests on GitHub actions (#2252) (#2253)","category":"page"},{"location":"moi/release_notes/#[v1.18.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.18.0)-(June-23,-2023)","page":"Release notes","title":"v1.18.0 (June 23, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.SlackBridgePrimalDualStart (#2194)\nAdded constraint_gradient_structure and eval_constraint_gradient (#2200)","category":"page"},{"location":"moi/release_notes/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a missing @require in MOI.Test (#2195) (#2196)\nFixed incorrect usage of Utilities.operate! in bridges (#2207) (#2216)\nFixed splatting nonlinear expression with univariate operator (#2221)","category":"page"},{"location":"moi/release_notes/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Removed unused argument names (#2199)\nReduced memory requirement for tests (#2204)\nRefactored Utilities.promote_operation (#2206)\nImproved code style in Utilities/mutable_arithmetics.jl (#2209)\nRefactored various methods in Utilities/functions.jl (#2208) (#2212) (#2213) (#2214) (#2215)","category":"page"},{"location":"moi/release_notes/#[v1.17.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.17.1)-(June-6,-2023)","page":"Release notes","title":"v1.17.1 (June 6, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed comparison of ScalarFunctionConstantNotZero (#2190)","category":"page"},{"location":"moi/release_notes/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added documentation for enum instances (#2186)\nUpdated chatroom links in documentation (#2188)\nChanged the documentation to build on Julia v1.9 (#2191)","category":"page"},{"location":"moi/release_notes/#[v1.17.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.17.0)-(June-1,-2023)","page":"Release notes","title":"v1.17.0 (June 1, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added ScalarNonlinearFunction (#2059)\nAdded a variety of tests for NormSpectralCone, NormNuclearCone, and HermitianPositiveSemidefiniteConeTriangle (#2174)\nAdded Utilities.value_type for computing the value type of a function (#2176)","category":"page"},{"location":"moi/release_notes/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for external sets in Utilities.loadfromstring! (#2177)\nFixed promote_operation for ScalarNonlinearFunction (#2179)\nFixed two issues in FileFormats.LP when reading files with quadratic functions (#2182) (#2184)","category":"page"},{"location":"moi/release_notes/#[v1.16.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.16.0)-(May-16,-2023)","page":"Release notes","title":"v1.16.0 (May 16, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for MathOptFormat v1.3 and v1.4 (#2158) (#2169)\nAdded new method to the nonlinear API (#2162) (#2164)\neval_hessian_constraint\neval_hessian_objective\nhessian_constraint_structure\nhessian_objective_structure\nAdded new sets\nNormCone (#2119)\nScaledPositiveSemidefiniteConeTriangle (#2154)","category":"page"},{"location":"moi/release_notes/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for Julia v1.9 to work around a bug in the upstream Julia compiler (#2161) (#2163)\nFixed a correctness bug in Bridges.Constraint.HermitianToSymmetricPSDBridge (#2171)\nFixed convert(::VariableIndex, ::ScalarAffineFunction) when the function has terms with 0coefficients (#2173)","category":"page"},{"location":"moi/release_notes/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed solver-tests.yml (#2157)\nUpdated documentation links to developer chatroom (#2160)\nAdded various tests for bridges (#2156)\nAdded checklists to the developer documentation (#2167) (#2168)","category":"page"},{"location":"moi/release_notes/#[v1.15.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.15.1)-(April-25,-2023)","page":"Release notes","title":"v1.15.1 (April 25, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed deleting a variable in a bridged objective (#2150)","category":"page"},{"location":"moi/release_notes/#[v1.15.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.15.0)-(April-19,-2023)","page":"Release notes","title":"v1.15.0 (April 19, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.VectorFunctionizeBridge (#2139)","category":"page"},{"location":"moi/release_notes/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed support for Rational in Bridges.Constraint.SplitIntervalBridge (#2137)\nFixed printing of LaTeX models (#2141)\nFixed modify in Bridges.Objective.VectorSlackBridge (#2144)\nFixed NAME record with spaces in FileFormats.MPS (#2146)\nFixed deleting a variable in a bridged objective (#2147)","category":"page"},{"location":"moi/release_notes/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add a test for variables in one-sided open Interval sets (#2133)\nMinor style fixes in the source code (#2148)","category":"page"},{"location":"moi/release_notes/#[v1.14.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.14.1)-(April-6,-2023)","page":"Release notes","title":"v1.14.1 (April 6, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Bridges.print_active_bridges (#2135)","category":"page"},{"location":"moi/release_notes/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a warning when an ambiguous string is passed to exclude in Test.runtests (#2136)","category":"page"},{"location":"moi/release_notes/#[v1.14.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.14.0)-(April-4,-2023)","page":"Release notes","title":"v1.14.0 (April 4, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for starting values in Bridges.Constraint.QuadtoSOCBridge (#2115)\nAdded support for Regex in the include and exclude arguments to Test.runtests (#2129)\nAdded Bridges.print_active_bridges methods for individual objectives and constraints (#2128)","category":"page"},{"location":"moi/release_notes/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ResultCount when parsing .sol files in FileFormats.NL (#2130)","category":"page"},{"location":"moi/release_notes/#[v1.13.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.2)-(March-21,-2023)","page":"Release notes","title":"v1.13.2 (March 21, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed splatting of containers in MOI.Nonlinear (#2120)\nFixed a bug reading LP files with default bounds (#2121)\nFixed a bug in which Bridges.Constraint.HermitianToSymmetricPSDBridge was not enabled by default (#2123)","category":"page"},{"location":"moi/release_notes/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#2114)\nFunctions now print to the REPL in algebraic form. This is potentially breaking if you have tests which rely on a specific String form of MOI functions. (#2112) (#2126)","category":"page"},{"location":"moi/release_notes/#[v1.13.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.1)-(March-3,-2023)","page":"Release notes","title":"v1.13.1 (March 3, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added the Google style guide to the documentation linter Vale, and fixed the resulting warnings (#2110)\nImproved the docstrings in src/functions.jl (#2108)","category":"page"},{"location":"moi/release_notes/#[v1.13.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.13.0)-(February-28,-2023)","page":"Release notes","title":"v1.13.0 (February 28, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.Constraint.NumberConversionBridge (#2091)\nAdded Parameter set (#2095) (#2105) (#2106) (#2109)\nAdded with_cache_type argument to instantiate (#2097)\nAdded support for HermitianPositiveSemidefiniteConeTriangle in Utilities.Model (#2100)","category":"page"},{"location":"moi/release_notes/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug when Utilities.@product_of_sets is empty (#2101)\nFixed Bridges.print_active_bridges when variable bridge is an AbstractScalarSet (#2107)","category":"page"},{"location":"moi/release_notes/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added tests for vector-valued objective functions in FileFormats.MOF (#2093)\nUsed and documented preference for import MathOptInterface as MOI (#2096)\nFix and test links in the documentation with linkcheck = true (#2098)\nImproved docstrings of sets in src/sets.jl (#2099)\nSkip checking flakey links in documentation with linkcheck_ignore  (#2103)","category":"page"},{"location":"moi/release_notes/#[v1.12.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.12.0)-(February-10,-2023)","page":"Release notes","title":"v1.12.0 (February 10, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for vector-valued objective functions (#2070)\nAdded a Utilities.distance_to_set method for SecondOrderCone (#2060)","category":"page"},{"location":"moi/release_notes/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a number of constraint bridges so that Bridges.final_touch can be called multiple times without forcing a rebuild of the reformulation (#2089)","category":"page"},{"location":"moi/release_notes/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new tests that a set ObjectiveFunction appears in ListOfModelAttributesSet (#2085)\nImproved the docstrings of a number of constraint-programming related sets (#2087)","category":"page"},{"location":"moi/release_notes/#[v1.11.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.5)-(January-24,-2023)","page":"Release notes","title":"v1.11.5 (January 24, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .lp files with an off-diagonal quadratic objective (#2082)","category":"page"},{"location":"moi/release_notes/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added SnoopPrecompile directives for reduced time-to-first-X in Julia v1.9 (#2080)","category":"page"},{"location":"moi/release_notes/#[v1.11.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.4)-(January-12,-2023)","page":"Release notes","title":"v1.11.4 (January 12, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .lp files with an Integer section (#2078)","category":"page"},{"location":"moi/release_notes/#[v1.11.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.3)-(January-12,-2023)","page":"Release notes","title":"v1.11.3 (January 12, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance bug when deleting a vector of constraints (#2072)\nFixed a bug reading .lp files with terms like x -1 y (#2076)","category":"page"},{"location":"moi/release_notes/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Documented the two-argument method of optimize! (#2074)","category":"page"},{"location":"moi/release_notes/#[v1.11.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.2)-(January-2,-2023)","page":"Release notes","title":"v1.11.2 (January 2, 2023)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with ConstraintName set for VariableIndex constraints (#2066)\nFixed a bug reading .mof.json files with nonlinear objectives and no constraints (#2068)","category":"page"},{"location":"moi/release_notes/#[v1.11.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.1)-(December-22,-2022)","page":"Release notes","title":"v1.11.1 (December 22, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with integer coefficients for affine and quadratic functions (#2063)","category":"page"},{"location":"moi/release_notes/#[v1.11.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.11.0)-(December-2,-2022)","page":"Release notes","title":"v1.11.0 (December 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Utilities.PenaltyRelaxation and Utilities.ScalarPenaltyRelaxation (#1995)\nAdded Utilities.distance_to_set (#2048)\nAdded support for ConstraintPrimalStart and ConstraintDualStart in FileFormats.MOF (#2056)","category":"page"},{"location":"moi/release_notes/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Tidied these release notes (#2055)","category":"page"},{"location":"moi/release_notes/#[v1.10.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.10.0)-(November-22,-2022)","page":"Release notes","title":"v1.10.0 (November 22, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new methods set(::OptimizerWithAttributes, ::RawOptimizerAttribute, value) and get(::OptimizerWithAttributes, ::RawOptimizerAttribute) (#2049)\nAdded new methods Utilities.DoubleDicts.outer_keys and Utilities.DoubleDicts.nonempty_outer_keys (#2052)","category":"page"},{"location":"moi/release_notes/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed Bridges.Objective.SlackBridge when the objective function is complex-valued (#2036) (#2038)\nFixed docstring of Test.runtests to clarify the warn_unsupported argument (#2037)\nFixed reading of free variables in FileFormats.LP (#2044)\nFixed numerous edge cases reading files from QPLIB using FileFormats.LP (#2042) (#2044)\nFixed situations in which x^y returns a complex value in Nonlinear (#2050)","category":"page"},{"location":"moi/release_notes/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved the error message thrown when a user-defined nonlinear function does not accept splatted input (#2032)\nRemoved specialized iterators for keys and values in Utilities.CleverDicts (#2051)","category":"page"},{"location":"moi/release_notes/#[v1.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.9.0)-(October-29,-2022)","page":"Release notes","title":"v1.9.0 (October 29, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added default fallback for getting ListOfConstraintIndices and NumberOfConstraints when the constraint type is unsupported by the model (#2021)\nAdded support for min and max in nonlinear expressions (#2023)\nAdded support for Indicator{EqualTo{T}} constraints in FileFormats.MPS (#2022)\nAdded default fallback for write_to_file and read_from_file (#2029)","category":"page"},{"location":"moi/release_notes/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed Constraint.ZeroOneBridge by adding new bounds as affine constraints instead of variable bounds (#1879)\nFixed reading free rows in FileFormats.MPS files (#2009)\nFixed parsing of OBJSENSE blocks in FileFormats.MPS files (#2016) (#2019)\nFixed the parsing of deeply nested nonlinear expressions by removing the use of recursion (#2020)\nFixed the requirements check in Test.test_constrainnt_get_ConstraintIndex (#2024)","category":"page"},{"location":"moi/release_notes/#[v1.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.2)-(September-20,-2022)","page":"Release notes","title":"v1.8.2 (September 20, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Documentation","page":"Release notes","title":"Documentation","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added vale as a documentation linter (#2002)\nImproved styling of code blocks in the PDF (#1999) (#2000)\nFixed a number of typos in the documentation (#2001) (#2003)","category":"page"},{"location":"moi/release_notes/#[v1.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.1)-(September-12,-2022)","page":"Release notes","title":"v1.8.1 (September 12, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in supports(::AbstractBridgeOptimizer for constraint attributes (#1991) (#1992)","category":"page"},{"location":"moi/release_notes/#[v1.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.8.0)-(September-1,-2022)","page":"Release notes","title":"v1.8.0 (September 1, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new sets\nHyperRectangle (#1961)\nReified (#1955)\nAdded new bridges (#1955)\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.SplitHyperRectangleBridge\nAdded support for atan(y, x) in Nonlinear (#1987)","category":"page"},{"location":"moi/release_notes/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Lazily construct expressions in Nonlinear so that expressions are updated when Nonlinear.Parameter values are updated (#1984)\nAllow NORM_LIMIT as a TerminationStatus for unbounded problems in Test (#1990)","category":"page"},{"location":"moi/release_notes/#[v1.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.7.0)-(August-16,-2022)","page":"Release notes","title":"v1.7.0 (August 16, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new sets\nHermitianPositiveSemidefiniteConeTriangle\nAdded new optimizer-independent options\nRelativeGapTolerance\nAbsoluteGapTolerance\nAdded new bridges\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Objective.QuadratizeBridge\nAdded support for generic number types in Utilities.loadfromstring!\nUpdated FileFormats.MOF to MathOptFormat v1.1, enabling support for constraint programming sets in the MOF file format\nAdded support in various FileFormats for\nindicator constraints in FileFormats.MPS\nquadratic constraints and an objective in FileFormats.LP\nquadratic constraints and an objective in FileFormats.MPS","category":"page"},{"location":"moi/release_notes/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed some missing promotion rules","category":"page"},{"location":"moi/release_notes/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved the performance of Jacobian products in Nonlinear\nRemoved an un-needed copy in Utilities.modify_function\nVarious clean-ups in Bridges/bridge_optimizer.jl","category":"page"},{"location":"moi/release_notes/#[v1.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.6.1)-(July-23,-2022)","page":"Release notes","title":"v1.6.1 (July 23, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added support for ExponentialCone in MatrixOfConstraints\nFix PSDSquare_3 test to reflect a previously fixed bug getting the ConstraintDual of a PositiveSemidefiniteConeSquare constraint","category":"page"},{"location":"moi/release_notes/#[v1.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.6.0)-(July-2,-2022)","page":"Release notes","title":"v1.6.0 (July 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Bridges.needs_final_touch and Bridges.final_touch\nAdded new bridges from constraint programming sets to mixed-integer linear programs:\nAllDifferentToCountDistinctBridge\nCountAtLeastToCountBelongsBridge\nCountBelongsToMILPBridge\nCountDistinctToMILPBridge\nCountGreaterThanToMILPBridge\nCircuitToMILPBridge","category":"page"},{"location":"moi/release_notes/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Relax an instance of ::Vector to ::AbstractVector in MOI.Nonlinear\nFix BinPackingToMILPBridge to respect variable bounds\nFix SemiToBinaryBridge to throw error if other bounds are set","category":"page"},{"location":"moi/release_notes/#[v1.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.5.0)-(June-27,-2022)","page":"Release notes","title":"v1.5.0 (June 27, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added GetAttributeNotAllowed for solvers to indicate when getting an attribute encounters an error\nAdded Utilities.get_fallback support for ObjectiveValue and DualObjectiveValue\nAdded new bridges:\nRootDetConeSquare to RootDetConeTriangle\nLogDetConeSquare to LogDetConeTriangle\nBinPacking to a mixed-integer linear program\nTable to a mixed-integer linear program\nAdded Bridges.print_active_bridges to display the current optimal hyper-path in a Bridges.LazyBridgeOptimizer","category":"page"},{"location":"moi/release_notes/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ZeroOne tests with lower and upper bounds\nFixed error in FileFormats.LP when reading a malformed file\nFixed reading of nonlinear programs in FileFormats.MOF\nFixed bug in ConstraintDual when using SquareBridge","category":"page"},{"location":"moi/release_notes/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved documentation of nonlinear API\nDocumented duality convention for PositiveSemidefiniteConeSquare sets\nFixed typo in Bridges.Constraint.QuadToSOCBridge docstring","category":"page"},{"location":"moi/release_notes/#[v1.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.4.0)-(June-9,-2022)","page":"Release notes","title":"v1.4.0 (June 9, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a number of sets for constraint programming:\nAllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nTable\nAdded support for user-defined hessians in Nonlinear\nAdded Bridges.runtests to simplify the testing of bridge implementations","category":"page"},{"location":"moi/release_notes/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in FileFormats.NL when writing univariate *","category":"page"},{"location":"moi/release_notes/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Began a large refactoring of the Bridges submodule, with greatly improved documentation.","category":"page"},{"location":"moi/release_notes/#[v1.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.3.0)-(May-27,-2022)","page":"Release notes","title":"v1.3.0 (May 27, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add MOI.Nonlinear submodule. This is a large new submodule that has been refactored from code that was in JuMP. For now, it should be considered experimental.\nAdd FileFormats.NL.SolFileResults(::IO, ::Model)\nAdd FileFormats.NL.read!(::IO, ::Model)\nAdd MOI.modify that accepts a vector of modifications","category":"page"},{"location":"moi/release_notes/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Test which attempted to include non-.jl files\nFixed a bug in FileFormats for models with open interval constraints","category":"page"},{"location":"moi/release_notes/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in Utilities.DoubleDict\nVarious minor improvements to the documentation","category":"page"},{"location":"moi/release_notes/#[v1.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.2.0)-(April-25,-2022)","page":"Release notes","title":"v1.2.0 (April 25, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add support for the FORMAT_REW/.rew file format in FileFormats.","category":"page"},{"location":"moi/release_notes/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug handling of default variable bounds in FileFormats.LP\nFix FileFormats.MPS to not write OBJSENSE by default since this is only supported by some readers.","category":"page"},{"location":"moi/release_notes/#[v1.1.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.2)-(March-31,-2022)","page":"Release notes","title":"v1.1.2 (March 31, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix a range of bugs in FileFormats.LP\nFix reading of problem dimensions in FileFormats.SDPA","category":"page"},{"location":"moi/release_notes/#[v1.1.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.1)-(March-23,-2022)","page":"Release notes","title":"v1.1.1 (March 23, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix bug in test_model_UpperBoundAlreadySet\nFix bug in test_infeasible_ tests\nFix bug in test_objective_ObjectiveFunction_blank\nRelax restriction of MOI.AbstractOptimizer to MOI.ModelLike in Utilities.CachingOptimizer and instantiate.","category":"page"},{"location":"moi/release_notes/#New-tests","page":"Release notes","title":"New tests","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add test_conic_empty_matrix that checks conic solvers support problems with no variables.","category":"page"},{"location":"moi/release_notes/#[v1.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.1.0)-(March-2,-2022)","page":"Release notes","title":"v1.1.0 (March 2, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.throw_unsupported(::UniversalFallback) for simplifying solver wrappers which copy from a UniversalFallback.","category":"page"},{"location":"moi/release_notes/#[v1.0.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.2)-(March-1,-2022)","page":"Release notes","title":"v1.0.2 (March 1, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in the test_model_ScalarFunctionConstantNotZero test\nFixed the error type when an AbstractFunctionConversionBridge cannot get or set an attribute\nIdentified a correctness bug in RSOCtoPSDBridge. We now thrown an error instead of returning an incorrect result.","category":"page"},{"location":"moi/release_notes/#[v1.0.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.1)-(February-25,-2022)","page":"Release notes","title":"v1.0.1 (February 25, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in which OptimizerAttributes were not copied in CachingOptimizer\nFixed a bug in which shift_constant did not promote mixed types of coefficients\nFixed a bug in which deleting a constraint of a bridged variable threw ErrorException instead of MOI.DeleteNotAllowed\nFixed a bug in which add_constraint in MatrixOfConstraints did not canonicalize the function\nFixed a bug when modifying scalar constants of a function containing a bridged variable\nFixed a bug in which final_touch was not always called with a CachingOptimizer","category":"page"},{"location":"moi/release_notes/#[v1.0.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v1.0.0)-(February-17,-2022)","page":"Release notes","title":"v1.0.0 (February 17, 2022)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Although tagged as a breaking release, v1.0.0 is v0.10.9 with deprecations removed, similar to how Julia 1.0 was Julia 0.7 with deprecations removed.","category":"page"},{"location":"moi/release_notes/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nAll deprecations have been removed","category":"page"},{"location":"moi/release_notes/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated features. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Then install MathOptInterface v0.10.9:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add MathOptInterface@0.10\"","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"moi/release_notes/#[v0.10.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.9)-(February-16,-2022)","page":"Release notes","title":"v0.10.9 (February 16, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.FreeVariables as a new VariablesConstrainer for conic solvers\nAdded MOI.default_cache for specifying the model used in CachingOptimizer","category":"page"},{"location":"moi/release_notes/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed LaTeX printing of MOI.Interval sets","category":"page"},{"location":"moi/release_notes/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added Aqua.jl as a CI check, and fixed suggested issues\nThe constructors of GeoMeanBridge, StructOfConstraints, and CachingOptimizer were changed from outer to inner constructors. This change is technically breaking, but does not impact users who followed the documented API.","category":"page"},{"location":"moi/release_notes/#[v0.10.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.8)-(February-3,-2022)","page":"Release notes","title":"v0.10.8 (February 3, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added a Base.read! for FileFormats.LP.","category":"page"},{"location":"moi/release_notes/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a bug in MutableSparseMatrix\nFixed a bug when calling operate!(vcat, ...) with Number arguments\nRemoved unintended export of deprecated symbols\nFixed a bug with PowerCone and DualPowerCone in MatrixOfConstraints.","category":"page"},{"location":"moi/release_notes/#[v0.10.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.7)-(January-5,-2022)","page":"Release notes","title":"v0.10.7 (January 5, 2022)","text":"","category":"section"},{"location":"moi/release_notes/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added test for modifying the constant vector in a VectorAffineFunction-in-Zeros constraint.","category":"page"},{"location":"moi/release_notes/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed the order in which sets are added to a LazyBridgeOptimizer. Compared to v0.10.6, this may result in bridged models being created with a different number (and order) of variables and constraints. However, it was necessary to fix cases which were previously rejected as unsupported, even though there was a valid bridge transformation.\nFixed an error message in FileFormats.CBF\nFixed comparison in test_linear_integration_Interval\nFixed errors for ConstraintPrimal in a CachingOptimizer\nFixed printing of models with non-Float64 coefficients.","category":"page"},{"location":"moi/release_notes/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements to reduce time-to-first-solve latency\nImproved error message when an optimizer does not support compute_conflict!","category":"page"},{"location":"moi/release_notes/#[v0.10.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.6)-(November-30,-2021)","page":"Release notes","title":"v0.10.6 (November 30, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added new documentation and tests for infeasibility certificates\nAdded a version control system for the tests in MOI.Test.runtests. Pass exclude_tests_after = v\"0.10.5\" to run tests added in v0.10.5 and earlier.\nMOI.Test.runtests now supports generic number types. To specify the number type T, pass MOI.Test.Config(T).\nAdded infeasible_status to MOI.Test.Config for solvers which return LOCALLY_INFEASIBLE\nCachingOptimizers now use a fallback for ConstraintPrimal. This should enable solvers using a CachingOptimizer to pass tests requiring ConstraintPrimal.","category":"page"},{"location":"moi/release_notes/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed a StackOverflow bug in copy_to\nFixed error thrown when nonconvex quadratic constraints cannot be bridged\nFixed a bug in copy_to for FileFormats.NL.Model\nFixed a bug in FileFormats.NL when printing large integers\nRemove a common test failure for LowerBoundAlreadySet tests\nUtilities.num_rows is now exported\nRemove parts of failing test_model_copy_to_xxx tests due to bridges","category":"page"},{"location":"moi/release_notes/#[v0.10.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.5)-(November-7,-2021)","page":"Release notes","title":"v0.10.5 (November 7, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed getter in UniversalFallback\nFixed test_solve_conflict_zeroone_ii","category":"page"},{"location":"moi/release_notes/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Make normalize_and_add_constraint more flexible\nUpdate paper BibTeX","category":"page"},{"location":"moi/release_notes/#[v0.10.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.4)-(October-26,-2021)","page":"Release notes","title":"v0.10.4 (October 26, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add SolverVersion attribute\nAdd new tests:\ntest_solve_conflict_zeroone_ii\ntest_nonlinear_objective\nUtilities.VariablesContainer now supports ConstraintFunction and ConstraintSet\nThe documentation is now available as a PDF","category":"page"},{"location":"moi/release_notes/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Update to MutableArithmetics 0.3\nVarious improvements to the documentation","category":"page"},{"location":"moi/release_notes/#[v0.10.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.3)-(September-18,-2021)","page":"Release notes","title":"v0.10.3 (September 18, 2021)","text":"","category":"section"},{"location":"moi/release_notes/#Fixed-39","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed bug which prevented callbacks from working through a CachingOptimizer\nFixed bug in Test submodule","category":"page"},{"location":"moi/release_notes/#[v0.10.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.2)-(September-16,-2021)","page":"Release notes","title":"v0.10.2 (September 16, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Updated MathOptFormat to v1.0\nUpdated JSONSchema to v1.0\nAdded Utilities.set_with_dimension\nAdded two-argument optimize!(::AbstractOptimizer, ::ModelLike)\nThe experimental feature copy_to_and_optimize! has been removed\nDet bridges now support getting ConstraintFunction and ConstraintSet\nVarious minor bug fixes identified by improved testing","category":"page"},{"location":"moi/release_notes/#[v0.10.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.1)-(September-8,-2021)","page":"Release notes","title":"v0.10.1 (September 8, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes to MOI.Test","category":"page"},{"location":"moi/release_notes/#[v0.10.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.10.0)-(September-6,-2021)","page":"Release notes","title":"v0.10.0 (September 6, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"MOI v0.10 is a significant breaking release. There are a large number of user-visible breaking changes and code refactors, as well as a substantial number of new features.","category":"page"},{"location":"moi/release_notes/#Breaking-in-MOI","page":"Release notes","title":"Breaking in MOI","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"SingleVariable has been removed; use VariableIndex instead\nSingleVariableConstraintNameError has been renamed to VariableIndexConstraintNameError\nSettingSingleVariableFunctionNotAllowed has been renamed to SettingVariableIndexFunctionNotAllowed\nVariableIndex constraints should not support ConstraintName\nVariableIndex constraints should not support ConstraintBasisStatus; implement VariableBasisStatus instead\nListOfConstraints has been renamed to ListOfConstraintTypesPresent\nListOfConstraintTypesPresent should now return Tuple{Type,Type} instead of Tuple{DataType,DataType}\nSolveTime has been renamed to SolveTimeSec\nIndicatorSet has been renamed to Indicator\nRawParameter has been renamed to RawOptimizerAttribute and now takes String instead of Any as the only argument\nThe .N field in result attributes has been renamed to .result_index\nThe .variable_index field in ScalarAffineTerm has been renamed to .variable\nThe .variable_index_1 field in ScalarQuadraticTerm has been renamed to .variable_1\nThe .variable_index_2 field in ScalarQuadraticTerm has been renamed to .variable_2\nThe order of affine_terms and quadratic_terms in ScalarQuadraticFunction and VectorQuadraticFunction have been reversed. Both functions now accept quadratic, affine, and constant terms in that order.\nThe index_value function has been removed. Use .value instead.\nisapprox has been removed for SOS1 and SOS2.\nThe dimension argument to Complements(dimension::Int) should now be the length of the corresponding function, instead of half the length. An ArgumentError is thrown if dimension is not even.\ncopy_to no longer takes keyword arguments:\ncopy_names: now copy names if they are supported by the destination solver\nfilter_constraints: use Utilities.ModelFilter instead\nwarn_attributes: never warn about optimizer attributes","category":"page"},{"location":"moi/release_notes/#Breaking-in-Bridges","page":"Release notes","title":"Breaking in Bridges","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Constraint.RSOCBridge has been renamed to Constraint.RSOCtoSOCBridge\nConstraint.SOCRBridge has been renamed to Constraint.SOCtoRSOCBridge\nBridges now return vectors that can be modified by the user. Previously, some bridges returned views instead of copies.\nBridges.IndexInVector has been unified into a single type. Previously, there was a different type for each submodule within Bridges\nThe signature of indicator bridges has been fixed. Use MOI.Bridges.Constraint.IndicatortoSOS1{Float64}(model).","category":"page"},{"location":"moi/release_notes/#Breaking-in-FileFormats","page":"Release notes","title":"Breaking in FileFormats","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"FileFormats.MOF.Model no longer accepts validate argument. Use the JSONSchema package to validate the MOF file. See the documentation for more information.","category":"page"},{"location":"moi/release_notes/#Breaking-in-Utilities","page":"Release notes","title":"Breaking in Utilities","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The datastructure of Utilities.Model (and models created with Utilities.@model) has been significantly refactored in a breaking way. This includes the way that objective functions and variable-related information is stored.\nUtilities.supports_default_copy has been renamed to supports_incremental_interface\nUtilities.automatic_copy_to has been renamed to Utilities.default_copy_to\nThe allocate-load API has been removed\nCachingOptimizers are now initialized as EMPTY_OPTIMIZER instead of ATTACHED_OPTIMIZER. If your code relies on the optimizer being attached, call MOIU.attach_optimizer(model) after creation.\nThe field names of Utilities.IndexMap have been renamed to var_map and con_map. Accessing these fields directly is considered a private detail that may change. Use the public getindex and setindex! API instead.\nThe size argument to Utilities.CleverDicts.CleverDict(::Integer) has been removed.\nThe size argument to Utilities.IndexMap(::Integer) has been removed.\nUtilities.DoubleDicts have been significantly refactored. Consult the source code for details.\nUtilities.test_models_equal has been moved to MOI.Test","category":"page"},{"location":"moi/release_notes/#Breaking-in-Test","page":"Release notes","title":"Breaking in Test","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"MOI.Test has been renamed to MOI.DeprecatedTest\nAn entirely new MOI.Test submodule has been written. See the documentation for details. The new MOI.Test submodule may find many bugs in the implementations of existing solvers that were previously untested.","category":"page"},{"location":"moi/release_notes/#Other-changes:","page":"Release notes","title":"Other changes:","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"attribute_value_type has been added\ncopy_to_and_optimize! has been added\nVariableBasisStatus has been added\nprint(model) now prints a human-readable description of the model\nVarious improvements to the FileFormats submodule\nFileFormats.CBF was refactored and received bugfixes\nSupport for MathOptFormat v0.6 was added in FileFormats.MOF\nFileFormats.MPS has had bugfixes and support for more features such as OBJSENSE and objective constants.\nFileFormats.NL has been added to support nonlinear files\nImproved type inference throughout to reduce latency","category":"page"},{"location":"moi/release_notes/#Updating","page":"Release notes","title":"Updating","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"A helpful script when updating is:","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"for (root, dirs, files) in walkdir(\".\")\n    for file in files\n        if !endswith(file, \".jl\")\n            continue\n        end\n        path = joinpath(root, file)\n        s = read(path, String)\n        for pair in [\n            \".variable_index\" => \".variable\",\n            \"RawParameter\" => \"RawOptimizerAttribute\",\n            \"ListOfConstraints\" => \"ListOfConstraintTypesPresent\",\n            \"TestConfig\" => \"Config\",\n            \"attr.N\" => \"attr.result_index\",\n            \"SolveTime\" => \"SolveTimeSec\",\n            \"DataType\" => \"Type\",\n            \"Utilities.supports_default_copy_to\" =>\n                \"supports_incremental_interface\",\n            \"SingleVariableConstraintNameError\" =>\n                \"VariableIndexConstraintNameError\",\n            \"SettingSingleVariableFunctionNotAllowed\" =>\n                \"SettingVariableIndexFunctionNotAllowed\",\n            \"automatic_copy_to\" => \"default_copy_to\",\n        ]\n            s = replace(s, pair)\n        end\n        write(path, s)\n    end\nend","category":"page"},{"location":"moi/release_notes/#[v0.9.22](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.22)-(May-22,-2021)","page":"Release notes","title":"v0.9.22 (May 22, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"This release contains backports from the ongoing development of the v0.10 release.","category":"page"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved type inference in Utilities, Bridges and FileFormats submodules to reduce latency.\nImproved performance of Utilities.is_canonical.\nFixed Utilities.pass_nonvariable_constraints with bridged variables.\nFixed performance regression of Utilities.Model.\nFixed ordering of objective setting in parser.","category":"page"},{"location":"moi/release_notes/#[v0.9.21](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.21)-(April-23,-2021)","page":"Release notes","title":"v0.9.21 (April 23, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added supports_shift_constant.\nImprove performance of bridging quadratic constraints.\nAdd precompilation statements.\nLarge improvements to the documentation.\nFix a variety of inference issues, benefiting precompilation and reducing initial latency.\nRawParameters are now ignored when resetting a CachingOptimizer. Previously, changing the underlying optimizer after RawParameters were set would throw an error.\nUtilities.AbstractModel is being refactored. This may break users interacting with private fields of a model generated using @model.","category":"page"},{"location":"moi/release_notes/#[v0.9.20](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.20)-(February-20,-2021)","page":"Release notes","title":"v0.9.20 (February 20, 2021)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improved performance of Utilities.ScalarFunctionIterator\nAdded support for compute_conflict to MOI layers\nAdded test with zero off-diagonal quadratic term in objective\nFixed double deletion of nested bridged SingleVariable/VectorOfVariables constraints\nFixed modification of un-set objective\nFixed function modification with duplicate terms\nMade unit tests abort without failing if the problem class is not supported\nFormatted code with JuliaFormatter\nClarified BasisStatusCode's docstring","category":"page"},{"location":"moi/release_notes/#[v0.9.19](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.19)-(December-1,-2020)","page":"Release notes","title":"v0.9.19 (December 1, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added CallbackNodeStatus attribute\nAdded bridge from GreaterThan or LessThan to Interval\nAdded tests for infeasibility certificates and double optimize\nFixed support for Julia v1.6\nRe-organized MOI docs and added documentation for adding a test","category":"page"},{"location":"moi/release_notes/#[v0.9.18](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.18)-(November-3,-2020)","page":"Release notes","title":"v0.9.18 (November 3, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various improvements for working with complex numbers\nAdded GeoMeantoRelEntrBridge to bridge a GeometricMeanCone constraint to a relative entropy constraint","category":"page"},{"location":"moi/release_notes/#[v0.9.17](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.17)-(September-21,-2020)","page":"Release notes","title":"v0.9.17 (September 21, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed CleverDict with variable of negative index value\nImplement supports_add_constrained_variable for MockOptimizer","category":"page"},{"location":"moi/release_notes/#[v0.9.16](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.16)-(September-17,-2020)","page":"Release notes","title":"v0.9.16 (September 17, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Various fixes:\n32-bit support\nCleverDict with abstract value type\nChecks in test suite","category":"page"},{"location":"moi/release_notes/#[v0.9.15](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.15)-(September-14,-2020)","page":"Release notes","title":"v0.9.15 (September 14, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Bridges improvements:\n(R)SOCtoNonConvexQuad bridge\nZeroOne bridge\nUse supports_add_constrained_variable in LazyBridgeOptimizer\nExposed VariableBridgeCost and ConstraintBridgeCost attributes\nPrioritize constraining variables on creation according to these costs\nRefactor bridge debugging\nLarge performance improvements across all submodules\nLots of documentation improvements\nFileFormats improvements:\nUpdate MathOptFormat to v0.5\nFix supported objectives in FileFormats\nTesting improvements:\nAdd name option for basic_constraint_test\nBug fixes and missing methods\nAdd length for iterators\nFix bug with duplicate terms\nFix order of LinearOfConstraintIndices","category":"page"},{"location":"moi/release_notes/#[v0.9.14](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.14)-(May-30,-2020)","page":"Release notes","title":"v0.9.14 (May 30, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).\nBump JSONSchema dependency from v0.2 to v0.3 (#1090).\nDocumentation improvements:\nFix typos (#1054, #1060, #1061, #1064, #1069, #1070).\nRemove the outdated recommendation for a package implementing MOI for a solver XXX to be called MathOptInterfaceXXX (#1087).\nUtilities improvements:\nFix is_canonical for quadratic functions (#1081, #1089).\nImplement add_constrained_variable[s] for CachingOptimizer so that it is added as constrained variables to the underlying optimizer (#1084).\nAdd support for custom objective functions for UniversalFallback (#1086).\nDeterministic ordering of constraints in UniversalFallback (#1088).\nTesting improvements:\nAdd NormOneCone/NormInfinityCone tests (#1045).\nBridges improvements:\nAdd bridges from Semiinteger and Semicontinuous (#1059).\nImplement getting ConstraintSet for Variable.FlipSignBridge (#1066).\nFix setting ConstraintFunction for Constraint.ScalarizeBridge (#1093).\nFix NormOne/NormInf bridges with nonzero constants (#1045).\nFix StackOverflow in debug (#1063).\nFileFormats improvements:\n[SDPA] Implement the extension for integer variables (#1079).\n[SDPA] Ignore comments after m and nblocks and detect dat-s extension (#1077).\n[SDPA] No scaling of off-diagonal coefficient (#1076).\n[SDPA] Add missing negation of constant (#1075).","category":"page"},{"location":"moi/release_notes/#[v0.9.13](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.13)-(March-24,-2020)","page":"Release notes","title":"v0.9.13 (March 24, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added tests for Semicontinuous and Semiinteger variables (#1033).\nAdded tests for using ExprGraphs from NLP evaluators (#1043).\nUpdate version compatibilities of dependencies (#1034, #1051, #1052).\nFixed typos in documentation (#1044).","category":"page"},{"location":"moi/release_notes/#[v0.9.12](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.12)-(February-28,-2020)","page":"Release notes","title":"v0.9.12 (February 28, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed writing NLPBlock in MathOptFormat (#1037).\nFixed MockOptimizer for result attributes with non-one result index (#1039).\nUpdated test template with instantiate (#1032).","category":"page"},{"location":"moi/release_notes/#[v0.9.11](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.11)-(February-21,-2020)","page":"Release notes","title":"v0.9.11 (February 21, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add an option for the model created by Utilities.@model to be a subtype of AbstractOptimizer (#1031).\nDescribed dual cone in docstrings of GeoMeanCone and RelativeEntropyCone (#1018, #1028).\nFixed typos in documentation (#1022, #1024).\nFixed warning of unsupported attribute (#1027).\nAdded more rootdet/logdet conic tests (#1026).\nImplemented ConstraintDual for Constraint.GeoMeanBridge, Constraint.RootDetBridge and Constraint.LogDetBridge and test duals in tests with GeoMeanCone and RootDetConeTriangle and LogDetConeTriangle cones (#1025, #1026).","category":"page"},{"location":"moi/release_notes/#[v0.9.10](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.10)-(January-31,-2020)","page":"Release notes","title":"v0.9.10 (January 31, 2020)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added OptimizerWithAttributes grouping an optimizer constructor and a list of optimizer attributes (#1008).\nAdded RelativeEntropyCone with corresponding bridge into exponential cone constraints (#993).\nAdded NormSpectralCone and NormNuclearCone with corresponding bridges into positive semidefinite constraints (#976).\nAdded supports_constrained_variable(s) (#1004).\nAdded dual_set_type (#1002).\nAdded tests for vector specialized version of delete (#989, #1011).\nAdded PSD3 test (#1007).\nClarified dual solution of Tests.pow1v and Tests.pow1f (#1013).\nAdded support for EqualTo and Zero in Bridges.Constraint.SplitIntervalBridge (#1005).\nFixed Utilities.vectorize for empty vector (#1003).\nFixed free variables in LP writer (#1006).","category":"page"},{"location":"moi/release_notes/#[v0.9.9](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.9)-(December-29,-2019)","page":"Release notes","title":"v0.9.9 (December 29, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).\nImproved performance of deletion of vector of variables in MOI.Utilities.Model (#983).\nUpdated to MutableArithmetics v0.2 (#981).\nAdded MutableArithmetics.promote_operation allocation tests (#975).\nFixed inference issue on Julia v1.1 (#982).","category":"page"},{"location":"moi/release_notes/#[v0.9.8](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.8)-(December-19,-2019)","page":"Release notes","title":"v0.9.8 (December 19, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented MutableArithmetics API (#924).\nFixed callbacks with CachingOptimizer (#959).\nFixed MOI.dimension for MOI.Complements (#948).\nAdded fallback for add_variables (#972).\nAdded is_diagonal_vectorized_index utility (#965).\nImproved linear constraints display in manual (#963, #964).\nBridges improvements:\nAdded IndicatorSet to SOS1 bridge (#877).\nAdded support for starting values for Variable.VectorizeBridge (#944).\nFixed MOI.add_constraints with non-bridged variable constraint on bridged variable (#951).\nFixed corner cases and docstring of GeoMeanBridge (#961, #962, #966).\nFixed choice between variable or constraint bridges for constrained variables (#973).\nImprove performance of bridge shortest path (#945, #946, #956).\nAdded docstring for test_delete_bridge (#954).\nAdded Variable bridge tests (#952).","category":"page"},{"location":"moi/release_notes/#[v0.9.7](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.7)-(October-30,-2019)","page":"Release notes","title":"v0.9.7 (October 30, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented _result_index_field for NLPBlockDual (#934).\nFixed copy of model with starting values for vector constraints (#941).\nBridges improvements:\nImproved performance of add_bridge and added has_bridge (#935).\nAdded AbstractSetMapBridge for bridges between sets S1, S2 such that there is a linear map A such that A*S1 = S2 (#933).\nAdded support for starting values for FlipSignBridge, VectorizeBridge, ScalarizeBridge, SlackBridge, SplitIntervalBridge, RSOCBridge, SOCRBridge NormInfinityBridge, SOCtoPSDBridge and RSOCtoPSDBridge (#933, #936, #937, #938, #939).","category":"page"},{"location":"moi/release_notes/#[v0.9.6](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.6)-(October-25,-2019)","page":"Release notes","title":"v0.9.6 (October 25, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added complementarity constraints (#913).\nAllowed ModelLike objects as value of attributes (#928).\nTesting improvements:\nAdded dual_objective_value option to MOI.Test.TestConfig (#922).\nAdded InvalidIndex tests in basic_constraint_tests (#921).\nAdded tests for the constant term in indicator constraint (#929).\nBridges improvements:\nAdded support for starting values for Functionize bridges (#923).\nAdded variable indices context to variable bridges (#920).\nFixed a typo in printing o debug_supports (#927).","category":"page"},{"location":"moi/release_notes/#[v0.9.5](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.5)-(October-9,-2019)","page":"Release notes","title":"v0.9.5 (October 9, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Clarified PrimalStatus/DualStatus to be NO_SOLUTION if result_index is out of bounds (#912).\nAdded tolerance for checks  and use ResultCount + 1 for the result_index in MOI.Test.solve_result_status (#910, #917).\nUse 0.5 instead of 2.0 for power in PowerCone in basic_constraint_test (#916).\nBridges improvements:\nAdded debug utilities for unsupported variable/constraint/objective (#861).\nFixed deletion of variables in bridged VectorOfVariables constraints (#909).\nFixed result_index with objective bridges (#911).","category":"page"},{"location":"moi/release_notes/#[v0.9.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.4)-(October-2,-2019)","page":"Release notes","title":"v0.9.4 (October 2, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Added solver-independent MIP callbacks (#782).\nImplements submit for Utilities.CachingOptimizer and Bridges.AbstractBridgeOptimizer (#906).\nAdded tests for result count of solution attributes (#901, #904).\nAdded NumberOfThreads attribute (#892).\nAdded Utilities.get_bounds to get the bounds on a variable (#890).\nAdded a note on duplicate coefficients in documentation (#581).\nAdded result index in ConstraintBasisStatus (#898).\nAdded extension dictionary to Utilities.Model (#884, #895).\nFixed deletion of constrained variables for CachingOptimizer (#905).\nImplemented Utilities.shift_constraint for Test.UnknownScalarSet (#896).\nBridges improvements:\nAdded Variable.RSOCtoSOCBridge (#907).\nImplemented MOI.get for ConstraintFunction/ConstraintSet for Bridges.Constraint.SquareBridge (#899).","category":"page"},{"location":"moi/release_notes/#[v0.9.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.3)-(September-20,-2019)","page":"Release notes","title":"v0.9.3 (September 20, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixed ambiguity detected in Julia v1.3 (#891, #893).\nFixed missing sets from ListOfSupportedConstraints (#880).\nFixed copy of VectorOfVariables constraints with duplicate indices (#886).\nAdded extension dictionary to MOIU.Model (#884).\nImplemented MOI.get for function and set for GeoMeanBridge (#888).\nUpdated documentation for SingleVariable indices and bridges (#885).\nTesting improvements:\nAdded more comprehensive tests for names (#882).\nAdded tests for SingleVariable duals (#883).\nAdded tests for DualExponentialCone and DualPowerCone (#873).\nImprovements for arbitrary coefficient type:\nFixed == for sets with mutable fields (#887).\nRemoved some Float64 assumptions in bridges (#878).\nAutomatic selection of Constraint.[Scalar|Vector]FunctionizeBridge (#889).","category":"page"},{"location":"moi/release_notes/#[v0.9.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.2)-(September-5,-2019)","page":"Release notes","title":"v0.9.2 (September 5, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Implemented model printing for MOI.ModelLike and specialized it for models defined in MOI (864).\nGeneralized contlinear tests for arbitrary coefficient type (#855).\nFixed supports_constraint for Semiinteger and Semicontinuous and supports for ObjectiveFunction (#859).\nFixed Allocate-Load copy for single variable constraints (#856).\nBridges improvements:\nAdd objective bridges (#789).\nFixed Variable.RSOCtoPSDBridge for dimension 2 (#869).\nAdded Variable.SOCtoRSOCBridge (#865).\nAdded Constraint.SOCRBridge and disable MOI.Bridges.Constraint.SOCtoPSDBridge (#751).\nFixed added_constraint_types for Contraint.LogDetBridge and Constraint.RootDetBridge (#870).","category":"page"},{"location":"moi/release_notes/#[v0.9.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.1)-(August-22,-2019)","page":"Release notes","title":"v0.9.1 (August 22, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix support for Julia v1.2 (#834).\nL1 and L∞ norm epigraph cones and corresponding bridges to LP were added (#818).\nAdded tests to MOI.Test.nametest (#833).\nFix MOI.Test.soc3test for solvers not supporting infeasibility certificates (#839).\nImplements operate for operators * and / between vector function and constant (#837).\nImplements show for MOI.Utilities.IndexMap (#847).\nFix corner cases for mapping of variables in MOI.Utilities.CachingOptimizer and substitution of variables in MOI.Bridges.AbstractBridgeOptimizer (#848).\nFix transformation of constant terms for MOI.Bridges.Constraint.SOCtoPSDBridge and MOI.Bridges.Constraint.RSOCtoPSDBridge (#840).","category":"page"},{"location":"moi/release_notes/#[v0.9.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.9.0)-(August-13,-2019)","page":"Release notes","title":"v0.9.0 (August 13, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Support for Julia v0.6 and v0.7 was dropped (#714, #717).\nA MOI.Utilities.Model implementation of ModelLike, this should replace most use cases of MOI.Utilities.@model (#781).\nadd_constrained_variable and add_constrained_variables were added (#759).\nSupport for indicator constraints was added (#709, #712).\nDualObjectiveValue attribute was added (#473).\nRawParameter attribute was added (#733).\nA dual_set function was added (#804).\nA Benchmarks submodule was added to facilitate solver benchmarking (#769).\nA submit function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).\nThe field of ObjectiveValue was renamed to result_index (#729).\nThe _constant and Utilities.getconstant function were renamed to constant\nREDUCTION_CERTIFICATE result status was added (#734).\nAbstract matrix sets were added (#731).\nTesting improvements:\nThe testing guideline was updated (#728).\nQuadratic tests were added (#697).\nUnit tests for RawStatusString, SolveTime, Silent and SolverName were added (#726, #741).\nA rotated second-order cone test was added (#759).\nA power cone test was added (#768).\nTests for ZeroOne variables with variable bounds were added (#772).\nAn unbounded test was added (#773).\nExisting tests had a few updates (#702, #703, #763).\nDocumentation improvements:\nAdded a section on CachingOptimizer (#777).\nAdded a section on UniversalFallback, Model and @model (#762).\nTransition the knapsack example to a doctest with MockOptimizer (#786).\nUtilities improvements:\nA CleverDict utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).\nThe Utilities.constant function was renamed to Utilities.constant_vector (#740).\nImplement optimizer attributes for CachingOptimizer (#745).\nRename Utilities.add_scalar_constraint to Utilities.normalize_and_add_constraint (#801).\noperate with vcat, SingleVariable and VectorOfVariables now returns a VectorOfVariables (#616).\nFix a type piracy of operate (#784).\nThe load_constraint fallback signature was fixed (#760).\nThe set_dot function was extended to work with sparse arrays (#805).\nBridges improvements:\nThe bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement ConstraintFunction and ConstraintSet if the user wants to recover them. As a consequence, the @bridge macro was removed (#722).\nBridge are now instantiated with a bridge_constraint function instead of using a constructor (#730).\nFix constraint attributes for bridges (#699).\nConstraint bridges were moved to the Bridges/Constraint submodule so they should now inherit from MOI.Bridges.Constraint.Abstract and should implement MOI.Bridges.Constraint.concrete_bridge_type instead of MOI.Bridges.concrete_bridge_type (#756).\nVariable bridges were added in (#759).\nVarious improvements (#746, #747).","category":"page"},{"location":"moi/release_notes/#[v0.8.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.4)-(March-13,-2019)","page":"Release notes","title":"v0.8.4 (March 13, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Performance improvement in default_copy_to and bridge optimizer (#696).\nAdd Silent and implement setting optimizer attributes in caching and mock optimizers (#695).\nAdd Functionize bridges (SingleVariable and VectorOfVariables) (#659).\nMinor typo fixes (#694).","category":"page"},{"location":"moi/release_notes/#[v0.8.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.3)-(March-6,-2019)","page":"Release notes","title":"v0.8.3 (March 6, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Use zero constant in scalar constraint function of MOI.Test.copytest (#691).\nFix variable deletion with SingleVariable objective function (#690).\nFix LazyBridgeOptimizer with bridges that add no constraints (#689).\nError message improvements (#673, #685, #686, #688).\nDocumentation improvements (#682, #683, #687).\nBasis status:\nRemove VariableBasisStatus (#679).\nTest ConstraintBasisStatus and implement it in bridges (#678).\nFix inference of NumberOfVariables and NumberOfConstraints (#677).\nImplement division between a quadratic function and a number (#675).","category":"page"},{"location":"moi/release_notes/#[v0.8.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.2)-(February-7,-2019)","page":"Release notes","title":"v0.8.2 (February 7, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add RawStatusString attribute (#629).\nDo not set names to the optimizer but only to the cache in CachingOptimizer (#638).\nMake scalar MOI functions act as scalars in broadcast (#646).\nAdd function utilities:\nImplement Base.zero (#634), Base.iszero (#643), add missing arithmetic operations (#644, #645) and fix division (#648).\nAdd a vectorize function that turns a vector of ScalarAffineFunction into a VectorAffineFunction (#642).\nImprove support for starting values:\nShow a warning in copy when starting values are not supported instead of throwing an error (#630).\nFix UniversalFallback for getting an variable or constraint attribute set to no indices (#623).\nAdd a test in contlineartest with partially set VariablePrimalStart.\nBridges improvements:\nFix StackOverFlow in LazyBridgeOptimizer when there is a cycle in the graph of bridges.\nAdd Slack bridges (#610, #650).\nAdd FlipSign bridges (#658).\nAdd tests with duplicate coefficients in ScalarAffineFunction and VectorAffineFunction (#639).\nUse tolerance to compare VariablePrimal in rotatedsoc1 test (#632).\nUse a zero constant in ScalarAffineFunction of constraints in psdt2 (#622).","category":"page"},{"location":"moi/release_notes/#[v0.8.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.1)-(January-7,-2019)","page":"Release notes","title":"v0.8.1 (January 7, 2019)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Adding an NLP objective now overrides any objective set using the ObjectiveFunction attribute (#619).\nRename fullbridgeoptimizer into full_bridge_optimizer (#621).\nAllow custom constraint types with full_bridge_optimizer (#617).\nAdd Vectorize bridge which transforms scalar linear constraints into vector linear constraints (#615).","category":"page"},{"location":"moi/release_notes/#[v0.8.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.8.0)-(December-18,-2018)","page":"Release notes","title":"v0.8.0 (December 18, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Rename all enum values to follow the JuMP naming guidelines for constants, for example, Optimal becomes OPTIMAL, and DualInfeasible becomes DUAL_INFEASIBLE.\nRename CachingOptimizer methods for style compliance.\nAdd an MOI.TerminationStatusCode called ALMOST_DUAL_INFEASIBLE.","category":"page"},{"location":"moi/release_notes/#[v0.7.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.7.0)-(December-13,-2018)","page":"Release notes","title":"v0.7.0 (December 13, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Test that MOI.TerminationStatus is MOI.OptimizeNotCalled before MOI.optimize! is called.\nCheck supports_default_copy_to in tests (#594).\nKey pieces of information like optimality, infeasibility, etc., are now reported through TerminationStatusCode. It is typically no longer necessary to check the result statuses in addition to the termination status.\nAdd perspective dimension to log-det cone (#593).","category":"page"},{"location":"moi/release_notes/#[v0.6.4](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.4)-(November-27,-2018)","page":"Release notes","title":"v0.6.4 (November 27, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Add OptimizeNotCalled termination status (#577) and improve documentation of other statuses (#575).\nAdd a solver naming guideline (#578).\nMake FeasibilitySense the default ObjectiveSense (#579).\nFix Utilities.@model and Bridges.@bridge macros for functions and sets defined outside MOI (#582).\nDocument solver-specific attributes (#580) and implement them in Utilities.CachingOptimizer (#565).","category":"page"},{"location":"moi/release_notes/#[v0.6.3](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.3)-(November-16,-2018)","page":"Release notes","title":"v0.6.3 (November 16, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)\nAttributes may now be partially set (some values could be nothing). (#563)\nPerformance improvements in Utilities.Model (#549, #567, #568)\nFix bug in QuadtoSOC (#558).\nNew supports_default_copy_to method that optimizers should implement to control caching behavior.\nDocumentation improvements.","category":"page"},{"location":"moi/release_notes/#[v0.6.2](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.2)-(October-26,-2018)","page":"Release notes","title":"v0.6.2 (October 26, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improve hygiene of @model macro (#544).\nFix bug in copy tests (#543).\nFix bug in UniversalFallback attribute getter (#540).\nAllow all correct solutions for solve_blank_obj unit test (#537).\nAdd errors for Allocate-Load and bad constraints (#534).\n[performance] Add specialized implementation of hash for VariableIndex (#533).\n[performance] Construct the name to object dictionaries lazily in model (#535).\nAdd the QuadtoSOC bridge which transforms ScalarQuadraticFunction constraints into RotatedSecondOrderCone (#483).","category":"page"},{"location":"moi/release_notes/#[v0.6.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.1)-(September-22,-2018)","page":"Release notes","title":"v0.6.1 (September 22, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Enable PositiveSemidefiniteConeSquare set and quadratic functions in MOIB.fullbridgeoptimizer (#524).\nAdd warning in the bridge between PositiveSemidefiniteConeSquare and PositiveSemidefiniteConeTriangle when the matrix is almost symmetric (#522).\nModify MOIT.copytest to not add multiples constraints on the same variable (#521).\nAdd missing keyword argument in one of MOIU.add_scalar_constraint methods (#520).","category":"page"},{"location":"moi/release_notes/#[v0.6.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.6.0)-(August-30,-2018)","page":"Release notes","title":"v0.6.0 (August 30, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"The MOIU.@model and MOIB.@bridge macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.\nRename functions according to the JuMP style guide:\ncopy! with keyword arguments copynames and warnattributes -> copy_to with keyword arguments copy_names and warn_attributes;\nset! -> set;\naddvariable[s]! -> add_variable[s];\nsupportsconstraint -> supports_constraint;\naddconstraint[s]! -> add_constraint[s];\nisvalid -> is_valid;\nisempty -> is_empty;\nBase.delete! -> delete;\nmodify! -> modify;\ntransform! -> transform;\ninitialize! -> initialize;\nwrite -> write_to_file; and\nread! -> read_from_file.\nRemove free! (use Base.finalize instead).\nAdd the SquarePSD bridge which transforms PositiveSemidefiniteConeTriangle constraints into PositiveSemidefiniteConeTriangle.\nAdd result fallback for ConstraintDual of variable-wise constraint, ConstraintPrimal and ObjectiveValue.\nAdd tests for ObjectiveBound.\nAdd test for empty rows in vector linear constraint.\nRework errors: CannotError has been renamed NotAllowedError and the distinction between UnsupportedError and NotAllowedError is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)\ncanget is removed. NoSolution is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).","category":"page"},{"location":"moi/release_notes/#[v0.5.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.5.0)-(August-5,-2018)","page":"Release notes","title":"v0.5.0 (August 5, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fix names with CachingOptimizer.\nCleanup thanks to @mohamed82008.\nAdded a universal fallback for constraints.\nFast utilities for function canonicalization thanks to @rdeits.\nRenamed dimension field to side_dimension in the context of matrix-like sets.\nNew and improved tests for cases like duplicate terms and ObjectiveBound.\nRemoved cantransform, canaddconstraint, canaddvariable, canset, canmodify, and candelete functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of UnsupportedError indicate unsupported operations, while subtypes of CannotError indicate operations that cannot be performed in the current state.\nThe API for copy! is updated to remove the CopyResult type.\nUpdates for the new JuMP style guide.","category":"page"},{"location":"moi/release_notes/#[v0.4.1](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.1)-(June-28,-2018)","page":"Release notes","title":"v0.4.1 (June 28, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Fixes vector function modification on 32 bits.\nFixes Bellman-Ford algorithm for bridges.\nAdded an NLP test with FeasibilitySense.\nUpdate modification documentation.","category":"page"},{"location":"moi/release_notes/#[v0.4.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.4.0)-(June-23,-2018)","page":"Release notes","title":"v0.4.0 (June 23, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Helper constructors for VectorAffineTerm and VectorQuadraticTerm.\nAdded modify_lhs to TestConfig.\nAdditional unit tests for optimizers.\nAdded a type parameter to CachingOptimizer for the optimizer field.\nNew API for problem modification (#388)\nTests pass without deprecation warnings on Julia 0.7.\nSmall fixes and documentation updates.","category":"page"},{"location":"moi/release_notes/#[v0.3.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.3.0)-(May-25,-2018)","page":"Release notes","title":"v0.3.0 (May 25, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.\nImprovements to MockOptimizer.\nSignificant changes to Bridges.\nNew and improved unit tests.\nFixes for Julia 0.7.","category":"page"},{"location":"moi/release_notes/#[v0.2.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.2.0)-(April-24,-2018)","page":"Release notes","title":"v0.2.0 (April 24, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Improvements to and better coverage of Tests.\nDocumentation fixes.\nSolverName attribute.\nChanges to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).\nAddition of NLP tests.\nIntroduction of UniversalFallback.\ncopynames keyword argument to MOI.copy!.\nAdd Bridges submodule.","category":"page"},{"location":"moi/release_notes/#[v0.1.0](https://github.com/jump-dev/MathOptInterface.jl/releases/tag/v0.1.0)-(February-28,-2018)","page":"Release notes","title":"v0.1.0 (February 28, 2018)","text":"","category":"section"},{"location":"moi/release_notes/","page":"Release notes","title":"Release notes","text":"Initial public release.\nThe framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"developers/extensions/#extensions_manual","page":"Extensions","title":"Extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP provides a variety of ways to extend the basic modeling functionality.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nThis documentation in this section is still a work-in-progress. The best place to look for ideas and help when writing a new JuMP extension are existing JuMP extensions. Examples include:BilevelJuMP.jl\nColuna.jl\nInfiniteOpt.jl\nPlasmo.jl\nPolyJuMP.jl\nSDDP.jl\nStochasticPrograms.jl\nSumOfSquares.jl\nvOptGeneric.jl","category":"page"},{"location":"developers/extensions/#Compatibility","page":"Extensions","title":"Compatibility","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"When writing JuMP extensions, you should carefully consider the compatibility guarantees that JuMP makes. In particular:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"All functions, structs, and constants which do not begin with an underscore (_) are public. These are always safe to use, and they should all have corresponding documentation.\nAll identifiers beginning with an underscore (_) are private. These are not safe to use, because they may break in any JuMP release, including patch releases.\nUnless explicitly mentioned in the documentation, all fields of a struct are private. These are not safe to use, because they may break in any JuMP release, including patch releases. An example of a field which is safe to use is the model.ext extension dictionary, which is documented in The extension dictionary.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"In general, we strongly encourage you to use only the public API of JuMP. If you are missing a feature, please open a GitHub issue.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"However, if you do use the private API (for example, because your feature request has not been implemented yet), then you must carefully restrict the versions of JuMP that your package is compatible with in the Project.toml file. The easiest way to do this is via the hyphen specifiers. For example, if your package supports all JuMP versions between v1.0.0 and v1.1.1, do:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP = \"1.0.0 - 1.1.1\"","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Then, whenever JuMP releases a new version, you should check if your package is still compatible and update the bound accordingly.","category":"page"},{"location":"developers/extensions/#Define-a-new-set","page":"Extensions","title":"Define a new set","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To define a new set for JuMP, subtype MOI.AbstractScalarSet or MOI.AbstractVectorSet and implement Base.copy for the set.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct NewMOIVectorSet <: MOI.AbstractVectorSet\n           dimension::Int\n       end\n\njulia> Base.copy(x::NewMOIVectorSet) = x\n\njulia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, x in NewMOIVectorSet(2))\n[x[1], x[2]] ∈ NewMOIVectorSet(2)","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"However, for vector-sets, this requires the user to specify the dimension argument to their set, even though we could infer it from the length of x!","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"You can make a more user-friendly set by subtyping AbstractVectorSet and implementing moi_set.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct NewVectorSet <: JuMP.AbstractVectorSet end\n\njulia> JuMP.moi_set(::NewVectorSet, dim::Int) = NewMOIVectorSet(dim)\n\njulia> @constraint(model, x in NewVectorSet())\n[x[1], x[2]] ∈ NewMOIVectorSet(2)","category":"page"},{"location":"developers/extensions/#extend_variable_macro","page":"Extensions","title":"Extend @variable","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Just as Bin and Int create binary and integer variables, you can extend the @variable macro to create new types of variables. Here is an explanation by example, where we create a AddTwice type, that creates a tuple of two JuMP variables instead of a single variable.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"First, create a new struct. This can be anything. Our struct holds a VariableInfo object that stores bound information, and whether the variable is binary or integer.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct AddTwice\n           info::JuMP.VariableInfo\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Second, implement build_variable, which takes ::Type{AddTwice} as an argument, and returns an instance of AddTwice. Note that you can also receive keyword arguments.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function JuMP.build_variable(\n           _err::Function,\n           info::JuMP.VariableInfo,\n           ::Type{AddTwice};\n           kwargs...\n       )\n           println(\"Can also use $kwargs here.\")\n           return AddTwice(info)\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Third, implement add_variable, which takes the instance of AddTwice from the previous step, and returns something. Typically, you will want to call add_variable here. For example, our AddTwice call is going to add two JuMP variables.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function JuMP.add_variable(\n           model::JuMP.Model,\n           duplicate::AddTwice,\n           name::String,\n       )\n           a = JuMP.add_variable(\n               model,\n               JuMP.ScalarVariable(duplicate.info),\n               \"$(name)_a\",\n            )\n           b = JuMP.add_variable(\n               model,\n               JuMP.ScalarVariable(duplicate.info),\n               \"$(name)_b\",\n            )\n           return (a, b)\n       end","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Now AddTwice can be passed to @variable similar to Bin or Int, or through the variable_type keyword. However, now it adds two variables instead of one.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2], variable_type = AddTwice, kw = i)\nCan also use Base.Pairs(:kw => 1) here.\nCan also use Base.Pairs(:kw => 2) here.\n2-element Vector{Tuple{VariableRef, VariableRef}}:\n (x[1]_a, x[1]_b)\n (x[2]_a, x[2]_b)\n\njulia> num_variables(model)\n4\n\njulia> first(x[1])\nx[1]_a\n\njulia> last(x[2])\nx[2]_b","category":"page"},{"location":"developers/extensions/#Extend-[@constraint](@ref)","page":"Extensions","title":"Extend @constraint","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The @constraint macro has three steps that can be intercepted and extended: parse time, build time, and add time.","category":"page"},{"location":"developers/extensions/#Parse","page":"Extensions","title":"Parse","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To extend the @constraint macro at parse time, implement one of the following methods:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_head\nparse_constraint_call","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nExtending the constraint macro at parse time is an advanced operation and has the potential to interfere with existing JuMP syntax. Please discuss with the developer chatroom before publishing any code that implements these methods.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_head should be implemented to intercept an expression based on the .head field of Base.Expr. For example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP\n\njulia> const MutableArithmetics = JuMP._MA;\n\njulia> model = Model(); @variable(model, x);\n\njulia> function JuMP.parse_constraint_head(\n           _error::Function,\n           ::Val{:≔},\n           lhs,\n           rhs,\n       )\n           println(\"Rewriting ≔ as ==\")\n           new_lhs, parse_code = MutableArithmetics.rewrite(lhs)\n           build_code = :(\n               build_constraint($(_error), $(new_lhs), MOI.EqualTo($(rhs)))\n           )\n           return false, parse_code, build_code\n       end\n\njulia> @constraint(model, x + x ≔ 1.0)\nRewriting ≔ as ==\n2 x = 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"parse_constraint_call should be implemented to intercept an expression of the form Expr(:call, op, args...). For example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP\n\njulia> const MutableArithmetics = JuMP._MA;\n\njulia> model = Model(); @variable(model, x);\n\njulia> function JuMP.parse_constraint_call(\n           _error::Function,\n           is_vectorized::Bool,\n           ::Val{:my_equal_to},\n           lhs,\n           rhs,\n       )\n           println(\"Rewriting my_equal_to to ==\")\n           new_lhs, parse_code = MutableArithmetics.rewrite(lhs)\n           build_code = if is_vectorized\n               :(build_constraint($(_error), $(new_lhs), MOI.EqualTo($(rhs)))\n           )\n           else\n               :(build_constraint.($(_error), $(new_lhs), MOI.EqualTo($(rhs))))\n           end\n           return parse_code, build_code\n       end\n\njulia> @constraint(model, my_equal_to(x + x, 1.0))\nRewriting my_equal_to to ==\n2 x = 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nWhen parsing a constraint you can recurse into sub-constraint (for example, the {expr} in z => {x <= 1}) by calling parse_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To prevent JuMP from promoting the set to the same value type as the model, use SkipModelConvertScalarSetWrapper.","category":"page"},{"location":"developers/extensions/#Build","page":"Extensions","title":"Build","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"To extend the @constraint macro at build time, implement a new build_constraint method.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"This may mean implementing a method for a specific function or set created at parse time, or it may mean implementing a method which handles additional positional arguments.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"build_constraint must return an AbstractConstraint, which can either be an AbstractConstraint already supported by JuMP, for example, ScalarConstraint or VectorConstraint, or a custom AbstractConstraint with a corresponding add_constraint method (see Add).","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"tip: Tip\nThe easiest way to extend @constraint is via an additional positional argument to build_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here is an example of adding extra arguments to build_constraint:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model(); @variable(model, x);\n\njulia> struct MyConstrType end\n\njulia> function JuMP.build_constraint(\n            _error::Function,\n            f::JuMP.GenericAffExpr,\n            set::MOI.EqualTo,\n            extra::Type{MyConstrType};\n            d = 0,\n       )\n            new_set = MOI.LessThan(set.value + d)\n            return JuMP.build_constraint(_error, f, new_set)\n       end\n\njulia> @constraint(model, my_con, x == 0, MyConstrType, d = 2)\nmy_con : x ≤ 2","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"note: Note\nOnly a single positional argument can be given to a particular constraint. Extensions that seek to pass multiple arguments (for example, Foo and Bar) should combine them into one argument type (for example, FooBar).","category":"page"},{"location":"developers/extensions/#extension_add_constraint","page":"Extensions","title":"Add","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"build_constraint returns an AbstractConstraint object. To extend @constraint at add time, define a subtype of AbstractConstraint, implement build_constraint to return an instance of the new type, and then implement add_constraint.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here is an example:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> model = Model(); @variable(model, x);\n\njulia> struct MyTag\n           name::String\n       end\n\njulia> struct MyConstraint{S} <: AbstractConstraint\n           name::String\n           f::AffExpr\n           s::S\n       end\n\njulia> function JuMP.build_constraint(\n            _error::Function,\n            f::AffExpr,\n            set::MOI.AbstractScalarSet,\n            extra::MyTag,\n       )\n            return MyConstraint(extra.name, f, set)\n       end\n\njulia> function JuMP.add_constraint(\n            model::Model,\n            con::MyConstraint,\n            name::String,\n       )\n            return add_constraint(\n                model,\n                ScalarConstraint(con.f, con.s),\n                \"$(con.name)[$(name)]\",\n            )\n       end\n\njulia> @constraint(model, my_con, 2x <= 1, MyTag(\"my_prefix\"))\nmy_prefix[my_con] : 2 x - 1 ≤ 0","category":"page"},{"location":"developers/extensions/#The-extension-dictionary","page":"Extensions","title":"The extension dictionary","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Every JuMP model has a field .ext::Dict{Symbol,Any} that can be used by extensions. This is useful if your extensions to @variable and @constraint need to store information between calls.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The most common way to initialize a model with information in the .ext dictionary is to provide a new constructor:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> function MyModel()\n           model = Model()\n           model.ext[:MyModel] = 1\n           return model\n       end\nMyModel (generic function with 1 method)\n\njulia> model = MyModel()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> model.ext\nDict{Symbol, Any} with 1 entry:\n  :MyModel => 1","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If you define extension data, implement copy_extension_data to support copy_model.","category":"page"},{"location":"developers/extensions/#Defining-new-JuMP-models","page":"Extensions","title":"Defining new JuMP models","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If extending individual calls to @variable and @constraint is not sufficient, it is possible to implement a new model via a subtype of AbstractModel. You can also define new AbstractVariableRefs to create different types of JuMP variables.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nExtending JuMP in this manner is an advanced operation. We strongly encourage you to consider how you can use the methods mentioned in the previous sections to achieve your aims instead of defining new model and variable types. Consult the developer chatroom before starting work on this.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"If you define new types, you will need to implement a considerable number of methods, and doing so will require a detailed understanding of the JuMP internals. Therefore, the list of methods to implement is currently undocumented.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The easiest way to extend JuMP by defining a new model type is to follow an existing example. A simple example to follow is the JuMPExtension module in the JuMP test suite. The best example of an external JuMP extension that implements an AbstractModel is InfiniteOpt.jl.","category":"page"},{"location":"developers/extensions/#Testing-JuMP-extensions","page":"Extensions","title":"Testing JuMP extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The JuMP test suite contains a large number of tests for JuMP extensions. You can run these tests by copying the MIT-licensed Kokako.jl file in the JuMP tests into your /test folder, and then adding this snippet to your /test/runtests.jl file:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"using MyJuMPExtension\nimport JuMP\ninclude(\"Kokako.jl\")\nconst MODULES_TO_TEST = Kokako.include_modules_to_test(JuMP)\nKokako.run_tests(\n    MODULES_TO_TEST,\n    MyJuMPExtension.MyModel,\n    MyJuMPExtension.MyVariableRef;\n    test_prefix = \"test_extension_\",\n)","category":"page"},{"location":"developers/extensions/#Set-an-optimize!-hook","page":"Extensions","title":"Set an optimize! hook","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Some extensions require modification to the problem after the user has finished constructing the problem, but before optimize! is called. For these situations, JuMP provides set_optimize_hook, which lets you intercept the optimize! call.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"Here's a simple example of adding an optimize hook that extends optimize! to take a keyword argument silent:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> using JuMP, HiGHS\n\njulia> model = Model(HiGHS.Optimizer);\n\njulia> @variable(model, x >= 1.5, Int);\n\njulia> @objective(model, Min, x);\n\njulia> function silent_hook(model; silent::Bool)\n           if silent\n               set_silent(model)\n           else\n               unset_silent(model)\n           end\n           ## Make sure you set ignore_optimize_hook = true, or we'll\n           ## recursively enter the optimize hook!\n           return optimize!(model; ignore_optimize_hook = true)\n       end\nsilent_hook (generic function with 1 method)\n\njulia> set_optimize_hook(model, silent_hook)\nsilent_hook (generic function with 1 method)\n\njulia> optimize!(model; silent = true)\n\njulia> optimize!(model; silent = false)\nSolution has               num          max          sum\nCol     infeasibilities      0            0            0\nInteger infeasibilities      0            0            0\nRow     infeasibilities      0            0            0\nRow     residuals            0            0            0\nPresolving model\n0 rows, 0 cols, 0 nonzeros\n0 rows, 0 cols, 0 nonzeros\nPresolve: Optimal\n\nSolving report\n  Status            Optimal\n  Primal bound      2\n  Dual bound        2\n  Gap               0% (tolerance: 0.01%)\n  Solution status   feasible\n                    2 (objective)\n                    0 (bound viol.)\n                    0 (int. viol.)\n                    0 (row viol.)\n  Timing            0.00 (total)\n                    0.00 (presolve)\n                    0.00 (postsolve)\n  Nodes             0\n  LP iterations     0 (total)\n                    0 (strong br.)\n                    0 (separation)\n                    0 (heuristics)","category":"page"},{"location":"developers/extensions/#Creating-new-container-types","page":"Extensions","title":"Creating new container types","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"JuMP macros (for example, @variable) accept a container keyword argument to force the type of container that is chosen. By default, JuMP supports container = Array, container = DenseAxisArray, container = SparseAxisArray and container = Auto. You can extend support to user-defined types by implementing Containers.container.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"For example, here is a container that reverses the order of the indices:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"julia> struct Foo end\n\njulia> function Containers.container(f::Function, indices, ::Type{Foo})\n           return reverse([f(i...) for i in indices])\n       end\n\njulia> model = Model();\n\njulia> @variable(model, x[1:3], container = Foo)\n3-element Vector{VariableRef}:\n x[3]\n x[2]\n x[1]\n\njulia> x[1]\nx[3]\n\njulia> @variable(model, y[1:3, 1:2], container = Foo)\n3×2 Matrix{VariableRef}:\n y[3,2]  y[3,1]\n y[2,2]  y[2,1]\n y[1,2]  y[1,1]\n\njulia> y[1, 1]\ny[3,2]\n\njulia> @variable(model, z[i=1:3; isodd(i)], container = Foo)\n2-element Vector{VariableRef}:\n z[3]\n z[1]\n\njulia> z[2]\nz[1]","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"warning: Warning\nIf you are a general user, you should not need to create a new container type. Instead, consider following User-defined containers and create a new container using standard Julia syntax. For example:julia> model = Model();\n\njulia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> y = reverse(x)\n3-element Vector{VariableRef}:\n x[3]\n x[2]\n x[1]","category":"page"},{"location":"developers/extensions/#Performance-tips-for-extensions","page":"Extensions","title":"Performance tips for extensions","text":"","category":"section"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"The function-in-set design of MathOptInterface causes type stability issues in Julia if you try to iterate over all of the constraints in a model. The easiest way to fix this is to use a function barrier.","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"For example, instead of:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"function all_names_slow(model)\n    names = Set{String}()\n    for ci in all_constraints(model)\n        push!(names, name(ci))\n    end\n    return names\nend","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"use:","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"function _function_barrier(names, model, ::Type{F}, ::Type{S}) where {F,S}\n    for ci in all_constraints(model, F, S)\n        push!(names, name(ci))\n    end\n    return\nend\n\nfunction all_names_fast(model)\n    names = Set{String}()\n    for (F, S) in list_of_constraint_types(model)\n        _function_barrier(names, model, F, S)\n    end\n    return names\nend","category":"page"},{"location":"developers/extensions/","page":"Extensions","title":"Extensions","text":"note: Note\nIt is important to explicitly type the F and S arguments. If you leave them untyped, for example, function _function_barrier(names, model, F, S), Julia will not specialize the function calls and performance will not be improved.","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/changelog.md\"","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"CurrentModule = MathOptInterface","category":"page"},{"location":"moi/changelog/#Release-notes","page":"Release notes","title":"Release notes","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.","category":"page"},{"location":"moi/changelog/#v1.20.1-(September-24,-2023)","page":"Release notes","title":"v1.20.1 (September 24, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Removed incorrect kwargs in some copy_to methods (#2272)\nFixed ConstraintDualStart for Bridges.Constraint.SplitIntervalBridge (#2275)\nFixed MethodError when trying to modify a variable objective (#2278)\nFixed stack-overflow in Utilities.operate(+,...) with many arguments (#2285) (#2290)","category":"page"},{"location":"moi/changelog/#Other","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MathOptSetDistances to solver-tests.yml (#2265)\nUpdated Documenter (#2266)\nFixed various JET errors (#2267) (#2269) (#2270) (#2271) (#2276) (#2277) (#2289)\nVarious style improvements\nReplaced using Package with import Package where possible (#2274)\nRemoved Utilities.EMPTYSTRING (#2283)\nRemoved unnecessary const acronyms in Utilities (#2280) (#2281)\nRemoved invalid and unused method (#2286)\nRefactored src/Utilities/model.jl (#2287)","category":"page"},{"location":"moi/changelog/#v1.20.0-(September-7,-2023)","page":"Release notes","title":"v1.20.0 (September 7, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Scaled set (#2237) (#2263) (#2264)\nAdded ObjectiveLimit attribute (#2257)","category":"page"},{"location":"moi/changelog/#Other-2","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Updated dependencies (#2258)\nImproved performance of ScalarNonlinearFunction utilities (#2259)\nFixed docstrings (#2261)","category":"page"},{"location":"moi/changelog/#v1.19.0-(August-15,-2023)","page":"Release notes","title":"v1.19.0 (August 15, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-2","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added VectorNonlinearFunction (#2201)\nAdded new bridges\nBridges.Constraint.IntegerToZeroOneBridge (#2205)\nBridges.Constraint.ToScalarQuadraticBridge (#2235)\nBridges.Constraint.ToVectorQuadraticBridge (#2235)\nBridges.Constraint.ToScalarNonlinearBridge (#2233) (#2235)\nBridges.Constraint.FunctionConversionBridge (#2235)\nAdded Bridges.bridging_cost for setting a bridge-specific cost in the bridging graph (#2235)\nAdded Utilities.eval_variables support for ScalarNonlinearFunction (#2218) (#2219)\nAdded support for ScalarNonlinearFunction in FileFormats.NL (#2228) (#2231)\nAdded support for writing non-Float64 functions in FileFormats.MOF\nAdded Utilities.lazy_map (#2254)","category":"page"},{"location":"moi/changelog/#Fixed-2","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed method ambiguities in operate (#2224)\nFixed reading LP file with a -infinity <= x <= +infinity variable (#2225)\nFixed missing require in Test.test_nonlinear_duals (#2230)\nFixed bug in ConstraintPrimal of Bridges.Constraint.QuadtoSOCBridge (#2240)","category":"page"},{"location":"moi/changelog/#Other-3","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added extensions to solver-tests.yml (#2229)\nRefactored test/Benchmarks (#2234)\nFixed warnings in tests (#2241) (#2243)\nSmall refactoring of bridges for upcoming VectorNonlinearFunction (#2244) (#2245)\nFixed various typos (#2251) (#2255)\nPartitioned how we run the tests on GitHub actions (#2252) (#2253)","category":"page"},{"location":"moi/changelog/#v1.18.0-(June-23,-2023)","page":"Release notes","title":"v1.18.0 (June 23, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-3","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.SlackBridgePrimalDualStart (#2194)\nAdded constraint_gradient_structure and eval_constraint_gradient (#2200)","category":"page"},{"location":"moi/changelog/#Fixed-3","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a missing @require in MOI.Test (#2195) (#2196)\nFixed incorrect usage of Utilities.operate! in bridges (#2207) (#2216)\nFixed splatting nonlinear expression with univariate operator (#2221)","category":"page"},{"location":"moi/changelog/#Other-4","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Removed unused argument names (#2199)\nReduced memory requirement for tests (#2204)\nRefactored Utilities.promote_operation (#2206)\nImproved code style in Utilities/mutable_arithmetics.jl (#2209)\nRefactored various methods in Utilities/functions.jl (#2208) (#2212) (#2213) (#2214) (#2215)","category":"page"},{"location":"moi/changelog/#v1.17.1-(June-6,-2023)","page":"Release notes","title":"v1.17.1 (June 6, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-4","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed comparison of ScalarFunctionConstantNotZero (#2190)","category":"page"},{"location":"moi/changelog/#Other-5","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added documentation for enum instances (#2186)\nUpdated chatroom links in documentation (#2188)\nChanged the documentation to build on Julia v1.9 (#2191)","category":"page"},{"location":"moi/changelog/#v1.17.0-(June-1,-2023)","page":"Release notes","title":"v1.17.0 (June 1, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-4","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added ScalarNonlinearFunction (#2059)\nAdded a variety of tests for NormSpectralCone, NormNuclearCone, and HermitianPositiveSemidefiniteConeTriangle (#2174)\nAdded Utilities.value_type for computing the value type of a function (#2176)","category":"page"},{"location":"moi/changelog/#Fixed-5","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for external sets in Utilities.loadfromstring! (#2177)\nFixed promote_operation for ScalarNonlinearFunction (#2179)\nFixed two issues in FileFormats.LP when reading files with quadratic functions (#2182) (#2184)","category":"page"},{"location":"moi/changelog/#v1.16.0-(May-16,-2023)","page":"Release notes","title":"v1.16.0 (May 16, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-5","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for MathOptFormat v1.3 and v1.4 (#2158) (#2169)\nAdded new method to the nonlinear API (#2162) (#2164)\neval_hessian_constraint\neval_hessian_objective\nhessian_constraint_structure\nhessian_objective_structure\nAdded new sets\nNormCone (#2119)\nScaledPositiveSemidefiniteConeTriangle (#2154)","category":"page"},{"location":"moi/changelog/#Fixed-6","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for Julia v1.9 to work around a bug in the upstream Julia compiler (#2161) (#2163)\nFixed a correctness bug in Bridges.Constraint.HermitianToSymmetricPSDBridge (#2171)\nFixed convert(::VariableIndex, ::ScalarAffineFunction) when the function has terms with 0coefficients (#2173)","category":"page"},{"location":"moi/changelog/#Other-6","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed solver-tests.yml (#2157)\nUpdated documentation links to developer chatroom (#2160)\nAdded various tests for bridges (#2156)\nAdded checklists to the developer documentation (#2167) (#2168)","category":"page"},{"location":"moi/changelog/#v1.15.1-(April-25,-2023)","page":"Release notes","title":"v1.15.1 (April 25, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-7","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed deleting a variable in a bridged objective (#2150)","category":"page"},{"location":"moi/changelog/#v1.15.0-(April-19,-2023)","page":"Release notes","title":"v1.15.0 (April 19, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-6","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Objective.VectorFunctionizeBridge (#2139)","category":"page"},{"location":"moi/changelog/#Fixed-8","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed support for Rational in Bridges.Constraint.SplitIntervalBridge (#2137)\nFixed printing of LaTeX models (#2141)\nFixed modify in Bridges.Objective.VectorSlackBridge (#2144)\nFixed NAME record with spaces in FileFormats.MPS (#2146)\nFixed deleting a variable in a bridged objective (#2147)","category":"page"},{"location":"moi/changelog/#Other-7","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add a test for variables in one-sided open Interval sets (#2133)\nMinor style fixes in the source code (#2148)","category":"page"},{"location":"moi/changelog/#v1.14.1-(April-6,-2023)","page":"Release notes","title":"v1.14.1 (April 6, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-9","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Bridges.print_active_bridges (#2135)","category":"page"},{"location":"moi/changelog/#Other-8","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a warning when an ambiguous string is passed to exclude in Test.runtests (#2136)","category":"page"},{"location":"moi/changelog/#v1.14.0-(April-4,-2023)","page":"Release notes","title":"v1.14.0 (April 4, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-7","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for starting values in Bridges.Constraint.QuadtoSOCBridge (#2115)\nAdded support for Regex in the include and exclude arguments to Test.runtests (#2129)\nAdded Bridges.print_active_bridges methods for individual objectives and constraints (#2128)","category":"page"},{"location":"moi/changelog/#Fixed-10","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ResultCount when parsing .sol files in FileFormats.NL (#2130)","category":"page"},{"location":"moi/changelog/#v1.13.2-(March-21,-2023)","page":"Release notes","title":"v1.13.2 (March 21, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-11","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed splatting of containers in MOI.Nonlinear (#2120)\nFixed a bug reading LP files with default bounds (#2121)\nFixed a bug in which Bridges.Constraint.HermitianToSymmetricPSDBridge was not enabled by default (#2123)","category":"page"},{"location":"moi/changelog/#Other-9","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed typos in the documentation (#2114)\nFunctions now print to the REPL in algebraic form. This is potentially breaking if you have tests which rely on a specific String form of MOI functions. (#2112) (#2126)","category":"page"},{"location":"moi/changelog/#v1.13.1-(March-3,-2023)","page":"Release notes","title":"v1.13.1 (March 3, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Other-10","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added the Google style guide to the documentation linter Vale, and fixed the resulting warnings (#2110)\nImproved the docstrings in src/functions.jl (#2108)","category":"page"},{"location":"moi/changelog/#v1.13.0-(February-28,-2023)","page":"Release notes","title":"v1.13.0 (February 28, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-8","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.Constraint.NumberConversionBridge (#2091)\nAdded Parameter set (#2095) (#2105) (#2106) (#2109)\nAdded with_cache_type argument to instantiate (#2097)\nAdded support for HermitianPositiveSemidefiniteConeTriangle in Utilities.Model (#2100)","category":"page"},{"location":"moi/changelog/#Fixed-12","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug when Utilities.@product_of_sets is empty (#2101)\nFixed Bridges.print_active_bridges when variable bridge is an AbstractScalarSet (#2107)","category":"page"},{"location":"moi/changelog/#Other-11","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added tests for vector-valued objective functions in FileFormats.MOF (#2093)\nUsed and documented preference for import MathOptInterface as MOI (#2096)\nFix and test links in the documentation with linkcheck = true (#2098)\nImproved docstrings of sets in src/sets.jl (#2099)\nSkip checking flakey links in documentation with linkcheck_ignore  (#2103)","category":"page"},{"location":"moi/changelog/#v1.12.0-(February-10,-2023)","page":"Release notes","title":"v1.12.0 (February 10, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Added-9","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for vector-valued objective functions (#2070)\nAdded a Utilities.distance_to_set method for SecondOrderCone (#2060)","category":"page"},{"location":"moi/changelog/#Fixed-13","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a number of constraint bridges so that Bridges.final_touch can be called multiple times without forcing a rebuild of the reformulation (#2089)","category":"page"},{"location":"moi/changelog/#Other-12","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new tests that a set ObjectiveFunction appears in ListOfModelAttributesSet (#2085)\nImproved the docstrings of a number of constraint-programming related sets (#2087)","category":"page"},{"location":"moi/changelog/#v1.11.5-(January-24,-2023)","page":"Release notes","title":"v1.11.5 (January 24, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-14","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug writing .lp files with an off-diagonal quadratic objective (#2082)","category":"page"},{"location":"moi/changelog/#Other-13","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added SnoopPrecompile directives for reduced time-to-first-X in Julia v1.9 (#2080)","category":"page"},{"location":"moi/changelog/#v1.11.4-(January-12,-2023)","page":"Release notes","title":"v1.11.4 (January 12, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-15","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .lp files with an Integer section (#2078)","category":"page"},{"location":"moi/changelog/#v1.11.3-(January-12,-2023)","page":"Release notes","title":"v1.11.3 (January 12, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-16","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance bug when deleting a vector of constraints (#2072)\nFixed a bug reading .lp files with terms like x -1 y (#2076)","category":"page"},{"location":"moi/changelog/#Other-14","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Documented the two-argument method of optimize! (#2074)","category":"page"},{"location":"moi/changelog/#v1.11.2-(January-2,-2023)","page":"Release notes","title":"v1.11.2 (January 2, 2023)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-17","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with ConstraintName set for VariableIndex constraints (#2066)\nFixed a bug reading .mof.json files with nonlinear objectives and no constraints (#2068)","category":"page"},{"location":"moi/changelog/#v1.11.1-(December-22,-2022)","page":"Release notes","title":"v1.11.1 (December 22, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-18","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug reading .mof.json files with integer coefficients for affine and quadratic functions (#2063)","category":"page"},{"location":"moi/changelog/#v1.11.0-(December-2,-2022)","page":"Release notes","title":"v1.11.0 (December 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-10","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Utilities.PenaltyRelaxation and Utilities.ScalarPenaltyRelaxation (#1995)\nAdded Utilities.distance_to_set (#2048)\nAdded support for ConstraintPrimalStart and ConstraintDualStart in FileFormats.MOF (#2056)","category":"page"},{"location":"moi/changelog/#Other-15","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Tidied these release notes (#2055)","category":"page"},{"location":"moi/changelog/#v1.10.0-(November-22,-2022)","page":"Release notes","title":"v1.10.0 (November 22, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-11","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new methods set(::OptimizerWithAttributes, ::RawOptimizerAttribute, value) and get(::OptimizerWithAttributes, ::RawOptimizerAttribute) (#2049)\nAdded new methods Utilities.DoubleDicts.outer_keys and Utilities.DoubleDicts.nonempty_outer_keys (#2052)","category":"page"},{"location":"moi/changelog/#Fixed-19","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed Bridges.Objective.SlackBridge when the objective function is complex-valued (#2036) (#2038)\nFixed docstring of Test.runtests to clarify the warn_unsupported argument (#2037)\nFixed reading of free variables in FileFormats.LP (#2044)\nFixed numerous edge cases reading files from QPLIB using FileFormats.LP (#2042) (#2044)\nFixed situations in which x^y returns a complex value in Nonlinear (#2050)","category":"page"},{"location":"moi/changelog/#Other-16","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved the error message thrown when a user-defined nonlinear function does not accept splatted input (#2032)\nRemoved specialized iterators for keys and values in Utilities.CleverDicts (#2051)","category":"page"},{"location":"moi/changelog/#v1.9.0-(October-29,-2022)","page":"Release notes","title":"v1.9.0 (October 29, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-12","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added default fallback for getting ListOfConstraintIndices and NumberOfConstraints when the constraint type is unsupported by the model (#2021)\nAdded support for min and max in nonlinear expressions (#2023)\nAdded support for Indicator{EqualTo{T}} constraints in FileFormats.MPS (#2022)\nAdded default fallback for write_to_file and read_from_file (#2029)","category":"page"},{"location":"moi/changelog/#Fixed-20","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed Constraint.ZeroOneBridge by adding new bounds as affine constraints instead of variable bounds (#1879)\nFixed reading free rows in FileFormats.MPS files (#2009)\nFixed parsing of OBJSENSE blocks in FileFormats.MPS files (#2016) (#2019)\nFixed the parsing of deeply nested nonlinear expressions by removing the use of recursion (#2020)\nFixed the requirements check in Test.test_constrainnt_get_ConstraintIndex (#2024)","category":"page"},{"location":"moi/changelog/#v1.8.2-(September-20,-2022)","page":"Release notes","title":"v1.8.2 (September 20, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Documentation","page":"Release notes","title":"Documentation","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added vale as a documentation linter (#2002)\nImproved styling of code blocks in the PDF (#1999) (#2000)\nFixed a number of typos in the documentation (#2001) (#2003)","category":"page"},{"location":"moi/changelog/#v1.8.1-(September-12,-2022)","page":"Release notes","title":"v1.8.1 (September 12, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-21","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in supports(::AbstractBridgeOptimizer for constraint attributes (#1991) (#1992)","category":"page"},{"location":"moi/changelog/#v1.8.0-(September-1,-2022)","page":"Release notes","title":"v1.8.0 (September 1, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-13","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new sets\nHyperRectangle (#1961)\nReified (#1955)\nAdded new bridges (#1955)\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.SplitHyperRectangleBridge\nAdded support for atan(y, x) in Nonlinear (#1987)","category":"page"},{"location":"moi/changelog/#Fixed-22","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Lazily construct expressions in Nonlinear so that expressions are updated when Nonlinear.Parameter values are updated (#1984)\nAllow NORM_LIMIT as a TerminationStatus for unbounded problems in Test (#1990)","category":"page"},{"location":"moi/changelog/#v1.7.0-(August-16,-2022)","page":"Release notes","title":"v1.7.0 (August 16, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-14","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new sets\nHermitianPositiveSemidefiniteConeTriangle\nAdded new optimizer-independent options\nRelativeGapTolerance\nAbsoluteGapTolerance\nAdded new bridges\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Objective.QuadratizeBridge\nAdded support for generic number types in Utilities.loadfromstring!\nUpdated FileFormats.MOF to MathOptFormat v1.1, enabling support for constraint programming sets in the MOF file format\nAdded support in various FileFormats for\nindicator constraints in FileFormats.MPS\nquadratic constraints and an objective in FileFormats.LP\nquadratic constraints and an objective in FileFormats.MPS","category":"page"},{"location":"moi/changelog/#Fixed-23","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed some missing promotion rules","category":"page"},{"location":"moi/changelog/#Other-17","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved the performance of Jacobian products in Nonlinear\nRemoved an un-needed copy in Utilities.modify_function\nVarious clean-ups in Bridges/bridge_optimizer.jl","category":"page"},{"location":"moi/changelog/#v1.6.1-(July-23,-2022)","page":"Release notes","title":"v1.6.1 (July 23, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-24","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added support for ExponentialCone in MatrixOfConstraints\nFix PSDSquare_3 test to reflect a previously fixed bug getting the ConstraintDual of a PositiveSemidefiniteConeSquare constraint","category":"page"},{"location":"moi/changelog/#v1.6.0-(July-2,-2022)","page":"Release notes","title":"v1.6.0 (July 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-15","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Bridges.needs_final_touch and Bridges.final_touch\nAdded new bridges from constraint programming sets to mixed-integer linear programs:\nAllDifferentToCountDistinctBridge\nCountAtLeastToCountBelongsBridge\nCountBelongsToMILPBridge\nCountDistinctToMILPBridge\nCountGreaterThanToMILPBridge\nCircuitToMILPBridge","category":"page"},{"location":"moi/changelog/#Fixed-25","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Relax an instance of ::Vector to ::AbstractVector in MOI.Nonlinear\nFix BinPackingToMILPBridge to respect variable bounds\nFix SemiToBinaryBridge to throw error if other bounds are set","category":"page"},{"location":"moi/changelog/#v1.5.0-(June-27,-2022)","page":"Release notes","title":"v1.5.0 (June 27, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-16","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added GetAttributeNotAllowed for solvers to indicate when getting an attribute encounters an error\nAdded Utilities.get_fallback support for ObjectiveValue and DualObjectiveValue\nAdded new bridges:\nRootDetConeSquare to RootDetConeTriangle\nLogDetConeSquare to LogDetConeTriangle\nBinPacking to a mixed-integer linear program\nTable to a mixed-integer linear program\nAdded Bridges.print_active_bridges to display the current optimal hyper-path in a Bridges.LazyBridgeOptimizer","category":"page"},{"location":"moi/changelog/#Fixed-26","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ZeroOne tests with lower and upper bounds\nFixed error in FileFormats.LP when reading a malformed file\nFixed reading of nonlinear programs in FileFormats.MOF\nFixed bug in ConstraintDual when using SquareBridge","category":"page"},{"location":"moi/changelog/#Other-18","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved documentation of nonlinear API\nDocumented duality convention for PositiveSemidefiniteConeSquare sets\nFixed typo in Bridges.Constraint.QuadToSOCBridge docstring","category":"page"},{"location":"moi/changelog/#v1.4.0-(June-9,-2022)","page":"Release notes","title":"v1.4.0 (June 9, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-17","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a number of sets for constraint programming:\nAllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nTable\nAdded support for user-defined hessians in Nonlinear\nAdded Bridges.runtests to simplify the testing of bridge implementations","category":"page"},{"location":"moi/changelog/#Fixed-27","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in FileFormats.NL when writing univariate *","category":"page"},{"location":"moi/changelog/#Other-19","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Began a large refactoring of the Bridges submodule, with greatly improved documentation.","category":"page"},{"location":"moi/changelog/#v1.3.0-(May-27,-2022)","page":"Release notes","title":"v1.3.0 (May 27, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-18","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add MOI.Nonlinear submodule. This is a large new submodule that has been refactored from code that was in JuMP. For now, it should be considered experimental.\nAdd FileFormats.NL.SolFileResults(::IO, ::Model)\nAdd FileFormats.NL.read!(::IO, ::Model)\nAdd MOI.modify that accepts a vector of modifications","category":"page"},{"location":"moi/changelog/#Fixed-28","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in Test which attempted to include non-.jl files\nFixed a bug in FileFormats for models with open interval constraints","category":"page"},{"location":"moi/changelog/#Other-20","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a performance issue in Utilities.DoubleDict\nVarious minor improvements to the documentation","category":"page"},{"location":"moi/changelog/#v1.2.0-(April-25,-2022)","page":"Release notes","title":"v1.2.0 (April 25, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-19","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add support for the FORMAT_REW/.rew file format in FileFormats.","category":"page"},{"location":"moi/changelog/#Fixed-29","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix bug handling of default variable bounds in FileFormats.LP\nFix FileFormats.MPS to not write OBJSENSE by default since this is only supported by some readers.","category":"page"},{"location":"moi/changelog/#v1.1.2-(March-31,-2022)","page":"Release notes","title":"v1.1.2 (March 31, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-30","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix a range of bugs in FileFormats.LP\nFix reading of problem dimensions in FileFormats.SDPA","category":"page"},{"location":"moi/changelog/#v1.1.1-(March-23,-2022)","page":"Release notes","title":"v1.1.1 (March 23, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-31","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix bug in test_model_UpperBoundAlreadySet\nFix bug in test_infeasible_ tests\nFix bug in test_objective_ObjectiveFunction_blank\nRelax restriction of MOI.AbstractOptimizer to MOI.ModelLike in Utilities.CachingOptimizer and instantiate.","category":"page"},{"location":"moi/changelog/#New-tests","page":"Release notes","title":"New tests","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add test_conic_empty_matrix that checks conic solvers support problems with no variables.","category":"page"},{"location":"moi/changelog/#v1.1.0-(March-2,-2022)","page":"Release notes","title":"v1.1.0 (March 2, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-20","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.throw_unsupported(::UniversalFallback) for simplifying solver wrappers which copy from a UniversalFallback.","category":"page"},{"location":"moi/changelog/#v1.0.2-(March-1,-2022)","page":"Release notes","title":"v1.0.2 (March 1, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-32","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in the test_model_ScalarFunctionConstantNotZero test\nFixed the error type when an AbstractFunctionConversionBridge cannot get or set an attribute\nIdentified a correctness bug in RSOCtoPSDBridge. We now thrown an error instead of returning an incorrect result.","category":"page"},{"location":"moi/changelog/#v1.0.1-(February-25,-2022)","page":"Release notes","title":"v1.0.1 (February 25, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-33","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in which OptimizerAttributes were not copied in CachingOptimizer\nFixed a bug in which shift_constant did not promote mixed types of coefficients\nFixed a bug in which deleting a constraint of a bridged variable threw ErrorException instead of MOI.DeleteNotAllowed\nFixed a bug in which add_constraint in MatrixOfConstraints did not canonicalize the function\nFixed a bug when modifying scalar constants of a function containing a bridged variable\nFixed a bug in which final_touch was not always called with a CachingOptimizer","category":"page"},{"location":"moi/changelog/#v1.0.0-(February-17,-2022)","page":"Release notes","title":"v1.0.0 (February 17, 2022)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Although tagged as a breaking release, v1.0.0 is v0.10.9 with deprecations removed, similar to how Julia 1.0 was Julia 0.7 with deprecations removed.","category":"page"},{"location":"moi/changelog/#Breaking","page":"Release notes","title":"Breaking","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Julia 1.6 is now the minimum supported version\nAll deprecations have been removed","category":"page"},{"location":"moi/changelog/#Troubleshooting-problems-when-updating","page":"Release notes","title":"Troubleshooting problems when updating","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"If you experience problems when updating, you are likely using previously deprecated features. (By default, Julia does not warn when you use deprecated features.)","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"To find the deprecated features you are using, start Julia with --depwarn=yes:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"$ julia --depwarn=yes","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Then install MathOptInterface v0.10.9:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"julia> using Pkg\njulia> pkg\"add MathOptInterface@0.10\"","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"And then run your code. Apply any suggestions, or search the release notes below for advice on updating a specific deprecated feature.","category":"page"},{"location":"moi/changelog/#v0.10.9-(February-16,-2022)","page":"Release notes","title":"v0.10.9 (February 16, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-21","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added MOI.Utilities.FreeVariables as a new VariablesConstrainer for conic solvers\nAdded MOI.default_cache for specifying the model used in CachingOptimizer","category":"page"},{"location":"moi/changelog/#Fixed-34","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed LaTeX printing of MOI.Interval sets","category":"page"},{"location":"moi/changelog/#Other-21","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added Aqua.jl as a CI check, and fixed suggested issues\nThe constructors of GeoMeanBridge, StructOfConstraints, and CachingOptimizer were changed from outer to inner constructors. This change is technically breaking, but does not impact users who followed the documented API.","category":"page"},{"location":"moi/changelog/#v0.10.8-(February-3,-2022)","page":"Release notes","title":"v0.10.8 (February 3, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-22","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added a Base.read! for FileFormats.LP.","category":"page"},{"location":"moi/changelog/#Fixed-35","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a bug in MutableSparseMatrix\nFixed a bug when calling operate!(vcat, ...) with Number arguments\nRemoved unintended export of deprecated symbols\nFixed a bug with PowerCone and DualPowerCone in MatrixOfConstraints.","category":"page"},{"location":"moi/changelog/#v0.10.7-(January-5,-2022)","page":"Release notes","title":"v0.10.7 (January 5, 2022)","text":"","category":"section"},{"location":"moi/changelog/#Added-23","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added test for modifying the constant vector in a VectorAffineFunction-in-Zeros constraint.","category":"page"},{"location":"moi/changelog/#Fixed-36","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed the order in which sets are added to a LazyBridgeOptimizer. Compared to v0.10.6, this may result in bridged models being created with a different number (and order) of variables and constraints. However, it was necessary to fix cases which were previously rejected as unsupported, even though there was a valid bridge transformation.\nFixed an error message in FileFormats.CBF\nFixed comparison in test_linear_integration_Interval\nFixed errors for ConstraintPrimal in a CachingOptimizer\nFixed printing of models with non-Float64 coefficients.","category":"page"},{"location":"moi/changelog/#Other-22","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various improvements to reduce time-to-first-solve latency\nImproved error message when an optimizer does not support compute_conflict!","category":"page"},{"location":"moi/changelog/#v0.10.6-(November-30,-2021)","page":"Release notes","title":"v0.10.6 (November 30, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Added-24","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added new documentation and tests for infeasibility certificates\nAdded a version control system for the tests in MOI.Test.runtests. Pass exclude_tests_after = v\"0.10.5\" to run tests added in v0.10.5 and earlier.\nMOI.Test.runtests now supports generic number types. To specify the number type T, pass MOI.Test.Config(T).\nAdded infeasible_status to MOI.Test.Config for solvers which return LOCALLY_INFEASIBLE\nCachingOptimizers now use a fallback for ConstraintPrimal. This should enable solvers using a CachingOptimizer to pass tests requiring ConstraintPrimal.","category":"page"},{"location":"moi/changelog/#Fixed-37","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed a StackOverflow bug in copy_to\nFixed error thrown when nonconvex quadratic constraints cannot be bridged\nFixed a bug in copy_to for FileFormats.NL.Model\nFixed a bug in FileFormats.NL when printing large integers\nRemove a common test failure for LowerBoundAlreadySet tests\nUtilities.num_rows is now exported\nRemove parts of failing test_model_copy_to_xxx tests due to bridges","category":"page"},{"location":"moi/changelog/#v0.10.5-(November-7,-2021)","page":"Release notes","title":"v0.10.5 (November 7, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-38","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed getter in UniversalFallback\nFixed test_solve_conflict_zeroone_ii","category":"page"},{"location":"moi/changelog/#Other-23","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Make normalize_and_add_constraint more flexible\nUpdate paper BibTeX","category":"page"},{"location":"moi/changelog/#v0.10.4-(October-26,-2021)","page":"Release notes","title":"v0.10.4 (October 26, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Added-25","page":"Release notes","title":"Added","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add SolverVersion attribute\nAdd new tests:\ntest_solve_conflict_zeroone_ii\ntest_nonlinear_objective\nUtilities.VariablesContainer now supports ConstraintFunction and ConstraintSet\nThe documentation is now available as a PDF","category":"page"},{"location":"moi/changelog/#Other-24","page":"Release notes","title":"Other","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Update to MutableArithmetics 0.3\nVarious improvements to the documentation","category":"page"},{"location":"moi/changelog/#v0.10.3-(September-18,-2021)","page":"Release notes","title":"v0.10.3 (September 18, 2021)","text":"","category":"section"},{"location":"moi/changelog/#Fixed-39","page":"Release notes","title":"Fixed","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed bug which prevented callbacks from working through a CachingOptimizer\nFixed bug in Test submodule","category":"page"},{"location":"moi/changelog/#v0.10.2-(September-16,-2021)","page":"Release notes","title":"v0.10.2 (September 16, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Updated MathOptFormat to v1.0\nUpdated JSONSchema to v1.0\nAdded Utilities.set_with_dimension\nAdded two-argument optimize!(::AbstractOptimizer, ::ModelLike)\nThe experimental feature copy_to_and_optimize! has been removed\nDet bridges now support getting ConstraintFunction and ConstraintSet\nVarious minor bug fixes identified by improved testing","category":"page"},{"location":"moi/changelog/#v0.10.1-(September-8,-2021)","page":"Release notes","title":"v0.10.1 (September 8, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various fixes to MOI.Test","category":"page"},{"location":"moi/changelog/#v0.10.0-(September-6,-2021)","page":"Release notes","title":"v0.10.0 (September 6, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"MOI v0.10 is a significant breaking release. There are a large number of user-visible breaking changes and code refactors, as well as a substantial number of new features.","category":"page"},{"location":"moi/changelog/#Breaking-in-MOI","page":"Release notes","title":"Breaking in MOI","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"SingleVariable has been removed; use VariableIndex instead\nSingleVariableConstraintNameError has been renamed to VariableIndexConstraintNameError\nSettingSingleVariableFunctionNotAllowed has been renamed to SettingVariableIndexFunctionNotAllowed\nVariableIndex constraints should not support ConstraintName\nVariableIndex constraints should not support ConstraintBasisStatus; implement VariableBasisStatus instead\nListOfConstraints has been renamed to ListOfConstraintTypesPresent\nListOfConstraintTypesPresent should now return Tuple{Type,Type} instead of Tuple{DataType,DataType}\nSolveTime has been renamed to SolveTimeSec\nIndicatorSet has been renamed to Indicator\nRawParameter has been renamed to RawOptimizerAttribute and now takes String instead of Any as the only argument\nThe .N field in result attributes has been renamed to .result_index\nThe .variable_index field in ScalarAffineTerm has been renamed to .variable\nThe .variable_index_1 field in ScalarQuadraticTerm has been renamed to .variable_1\nThe .variable_index_2 field in ScalarQuadraticTerm has been renamed to .variable_2\nThe order of affine_terms and quadratic_terms in ScalarQuadraticFunction and VectorQuadraticFunction have been reversed. Both functions now accept quadratic, affine, and constant terms in that order.\nThe index_value function has been removed. Use .value instead.\nisapprox has been removed for SOS1 and SOS2.\nThe dimension argument to Complements(dimension::Int) should now be the length of the corresponding function, instead of half the length. An ArgumentError is thrown if dimension is not even.\ncopy_to no longer takes keyword arguments:\ncopy_names: now copy names if they are supported by the destination solver\nfilter_constraints: use Utilities.ModelFilter instead\nwarn_attributes: never warn about optimizer attributes","category":"page"},{"location":"moi/changelog/#Breaking-in-Bridges","page":"Release notes","title":"Breaking in Bridges","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Constraint.RSOCBridge has been renamed to Constraint.RSOCtoSOCBridge\nConstraint.SOCRBridge has been renamed to Constraint.SOCtoRSOCBridge\nBridges now return vectors that can be modified by the user. Previously, some bridges returned views instead of copies.\nBridges.IndexInVector has been unified into a single type. Previously, there was a different type for each submodule within Bridges\nThe signature of indicator bridges has been fixed. Use MOI.Bridges.Constraint.IndicatortoSOS1{Float64}(model).","category":"page"},{"location":"moi/changelog/#Breaking-in-FileFormats","page":"Release notes","title":"Breaking in FileFormats","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"FileFormats.MOF.Model no longer accepts validate argument. Use the JSONSchema package to validate the MOF file. See the documentation for more information.","category":"page"},{"location":"moi/changelog/#Breaking-in-Utilities","page":"Release notes","title":"Breaking in Utilities","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The datastructure of Utilities.Model (and models created with Utilities.@model) has been significantly refactored in a breaking way. This includes the way that objective functions and variable-related information is stored.\nUtilities.supports_default_copy has been renamed to supports_incremental_interface\nUtilities.automatic_copy_to has been renamed to Utilities.default_copy_to\nThe allocate-load API has been removed\nCachingOptimizers are now initialized as EMPTY_OPTIMIZER instead of ATTACHED_OPTIMIZER. If your code relies on the optimizer being attached, call MOIU.attach_optimizer(model) after creation.\nThe field names of Utilities.IndexMap have been renamed to var_map and con_map. Accessing these fields directly is considered a private detail that may change. Use the public getindex and setindex! API instead.\nThe size argument to Utilities.CleverDicts.CleverDict(::Integer) has been removed.\nThe size argument to Utilities.IndexMap(::Integer) has been removed.\nUtilities.DoubleDicts have been significantly refactored. Consult the source code for details.\nUtilities.test_models_equal has been moved to MOI.Test","category":"page"},{"location":"moi/changelog/#Breaking-in-Test","page":"Release notes","title":"Breaking in Test","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"MOI.Test has been renamed to MOI.DeprecatedTest\nAn entirely new MOI.Test submodule has been written. See the documentation for details. The new MOI.Test submodule may find many bugs in the implementations of existing solvers that were previously untested.","category":"page"},{"location":"moi/changelog/#Other-changes:","page":"Release notes","title":"Other changes:","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"attribute_value_type has been added\ncopy_to_and_optimize! has been added\nVariableBasisStatus has been added\nprint(model) now prints a human-readable description of the model\nVarious improvements to the FileFormats submodule\nFileFormats.CBF was refactored and received bugfixes\nSupport for MathOptFormat v0.6 was added in FileFormats.MOF\nFileFormats.MPS has had bugfixes and support for more features such as OBJSENSE and objective constants.\nFileFormats.NL has been added to support nonlinear files\nImproved type inference throughout to reduce latency","category":"page"},{"location":"moi/changelog/#Updating","page":"Release notes","title":"Updating","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"A helpful script when updating is:","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"for (root, dirs, files) in walkdir(\".\")\n    for file in files\n        if !endswith(file, \".jl\")\n            continue\n        end\n        path = joinpath(root, file)\n        s = read(path, String)\n        for pair in [\n            \".variable_index\" => \".variable\",\n            \"RawParameter\" => \"RawOptimizerAttribute\",\n            \"ListOfConstraints\" => \"ListOfConstraintTypesPresent\",\n            \"TestConfig\" => \"Config\",\n            \"attr.N\" => \"attr.result_index\",\n            \"SolveTime\" => \"SolveTimeSec\",\n            \"DataType\" => \"Type\",\n            \"Utilities.supports_default_copy_to\" =>\n                \"supports_incremental_interface\",\n            \"SingleVariableConstraintNameError\" =>\n                \"VariableIndexConstraintNameError\",\n            \"SettingSingleVariableFunctionNotAllowed\" =>\n                \"SettingVariableIndexFunctionNotAllowed\",\n            \"automatic_copy_to\" => \"default_copy_to\",\n        ]\n            s = replace(s, pair)\n        end\n        write(path, s)\n    end\nend","category":"page"},{"location":"moi/changelog/#v0.9.22-(May-22,-2021)","page":"Release notes","title":"v0.9.22 (May 22, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"This release contains backports from the ongoing development of the v0.10 release.","category":"page"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved type inference in Utilities, Bridges and FileFormats submodules to reduce latency.\nImproved performance of Utilities.is_canonical.\nFixed Utilities.pass_nonvariable_constraints with bridged variables.\nFixed performance regression of Utilities.Model.\nFixed ordering of objective setting in parser.","category":"page"},{"location":"moi/changelog/#v0.9.21-(April-23,-2021)","page":"Release notes","title":"v0.9.21 (April 23, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added supports_shift_constant.\nImprove performance of bridging quadratic constraints.\nAdd precompilation statements.\nLarge improvements to the documentation.\nFix a variety of inference issues, benefiting precompilation and reducing initial latency.\nRawParameters are now ignored when resetting a CachingOptimizer. Previously, changing the underlying optimizer after RawParameters were set would throw an error.\nUtilities.AbstractModel is being refactored. This may break users interacting with private fields of a model generated using @model.","category":"page"},{"location":"moi/changelog/#v0.9.20-(February-20,-2021)","page":"Release notes","title":"v0.9.20 (February 20, 2021)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improved performance of Utilities.ScalarFunctionIterator\nAdded support for compute_conflict to MOI layers\nAdded test with zero off-diagonal quadratic term in objective\nFixed double deletion of nested bridged SingleVariable/VectorOfVariables constraints\nFixed modification of un-set objective\nFixed function modification with duplicate terms\nMade unit tests abort without failing if the problem class is not supported\nFormatted code with JuliaFormatter\nClarified BasisStatusCode's docstring","category":"page"},{"location":"moi/changelog/#v0.9.19-(December-1,-2020)","page":"Release notes","title":"v0.9.19 (December 1, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added CallbackNodeStatus attribute\nAdded bridge from GreaterThan or LessThan to Interval\nAdded tests for infeasibility certificates and double optimize\nFixed support for Julia v1.6\nRe-organized MOI docs and added documentation for adding a test","category":"page"},{"location":"moi/changelog/#v0.9.18-(November-3,-2020)","page":"Release notes","title":"v0.9.18 (November 3, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various improvements for working with complex numbers\nAdded GeoMeantoRelEntrBridge to bridge a GeometricMeanCone constraint to a relative entropy constraint","category":"page"},{"location":"moi/changelog/#v0.9.17-(September-21,-2020)","page":"Release notes","title":"v0.9.17 (September 21, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed CleverDict with variable of negative index value\nImplement supports_add_constrained_variable for MockOptimizer","category":"page"},{"location":"moi/changelog/#v0.9.16-(September-17,-2020)","page":"Release notes","title":"v0.9.16 (September 17, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Various fixes:\n32-bit support\nCleverDict with abstract value type\nChecks in test suite","category":"page"},{"location":"moi/changelog/#v0.9.15-(September-14,-2020)","page":"Release notes","title":"v0.9.15 (September 14, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Bridges improvements:\n(R)SOCtoNonConvexQuad bridge\nZeroOne bridge\nUse supports_add_constrained_variable in LazyBridgeOptimizer\nExposed VariableBridgeCost and ConstraintBridgeCost attributes\nPrioritize constraining variables on creation according to these costs\nRefactor bridge debugging\nLarge performance improvements across all submodules\nLots of documentation improvements\nFileFormats improvements:\nUpdate MathOptFormat to v0.5\nFix supported objectives in FileFormats\nTesting improvements:\nAdd name option for basic_constraint_test\nBug fixes and missing methods\nAdd length for iterators\nFix bug with duplicate terms\nFix order of LinearOfConstraintIndices","category":"page"},{"location":"moi/changelog/#v0.9.14-(May-30,-2020)","page":"Release notes","title":"v0.9.14 (May 30, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add a solver-independent interface for accessing the set of conflicting constraints an Irreducible Inconsistent Subsystem (#1056).\nBump JSONSchema dependency from v0.2 to v0.3 (#1090).\nDocumentation improvements:\nFix typos (#1054, #1060, #1061, #1064, #1069, #1070).\nRemove the outdated recommendation for a package implementing MOI for a solver XXX to be called MathOptInterfaceXXX (#1087).\nUtilities improvements:\nFix is_canonical for quadratic functions (#1081, #1089).\nImplement add_constrained_variable[s] for CachingOptimizer so that it is added as constrained variables to the underlying optimizer (#1084).\nAdd support for custom objective functions for UniversalFallback (#1086).\nDeterministic ordering of constraints in UniversalFallback (#1088).\nTesting improvements:\nAdd NormOneCone/NormInfinityCone tests (#1045).\nBridges improvements:\nAdd bridges from Semiinteger and Semicontinuous (#1059).\nImplement getting ConstraintSet for Variable.FlipSignBridge (#1066).\nFix setting ConstraintFunction for Constraint.ScalarizeBridge (#1093).\nFix NormOne/NormInf bridges with nonzero constants (#1045).\nFix StackOverflow in debug (#1063).\nFileFormats improvements:\n[SDPA] Implement the extension for integer variables (#1079).\n[SDPA] Ignore comments after m and nblocks and detect dat-s extension (#1077).\n[SDPA] No scaling of off-diagonal coefficient (#1076).\n[SDPA] Add missing negation of constant (#1075).","category":"page"},{"location":"moi/changelog/#v0.9.13-(March-24,-2020)","page":"Release notes","title":"v0.9.13 (March 24, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added tests for Semicontinuous and Semiinteger variables (#1033).\nAdded tests for using ExprGraphs from NLP evaluators (#1043).\nUpdate version compatibilities of dependencies (#1034, #1051, #1052).\nFixed typos in documentation (#1044).","category":"page"},{"location":"moi/changelog/#v0.9.12-(February-28,-2020)","page":"Release notes","title":"v0.9.12 (February 28, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed writing NLPBlock in MathOptFormat (#1037).\nFixed MockOptimizer for result attributes with non-one result index (#1039).\nUpdated test template with instantiate (#1032).","category":"page"},{"location":"moi/changelog/#v0.9.11-(February-21,-2020)","page":"Release notes","title":"v0.9.11 (February 21, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add an option for the model created by Utilities.@model to be a subtype of AbstractOptimizer (#1031).\nDescribed dual cone in docstrings of GeoMeanCone and RelativeEntropyCone (#1018, #1028).\nFixed typos in documentation (#1022, #1024).\nFixed warning of unsupported attribute (#1027).\nAdded more rootdet/logdet conic tests (#1026).\nImplemented ConstraintDual for Constraint.GeoMeanBridge, Constraint.RootDetBridge and Constraint.LogDetBridge and test duals in tests with GeoMeanCone and RootDetConeTriangle and LogDetConeTriangle cones (#1025, #1026).","category":"page"},{"location":"moi/changelog/#v0.9.10-(January-31,-2020)","page":"Release notes","title":"v0.9.10 (January 31, 2020)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added OptimizerWithAttributes grouping an optimizer constructor and a list of optimizer attributes (#1008).\nAdded RelativeEntropyCone with corresponding bridge into exponential cone constraints (#993).\nAdded NormSpectralCone and NormNuclearCone with corresponding bridges into positive semidefinite constraints (#976).\nAdded supports_constrained_variable(s) (#1004).\nAdded dual_set_type (#1002).\nAdded tests for vector specialized version of delete (#989, #1011).\nAdded PSD3 test (#1007).\nClarified dual solution of Tests.pow1v and Tests.pow1f (#1013).\nAdded support for EqualTo and Zero in Bridges.Constraint.SplitIntervalBridge (#1005).\nFixed Utilities.vectorize for empty vector (#1003).\nFixed free variables in LP writer (#1006).","category":"page"},{"location":"moi/changelog/#v0.9.9-(December-29,-2019)","page":"Release notes","title":"v0.9.9 (December 29, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Incorporated MathOptFormat.jl as the FileFormats submodule. FileFormats provides readers and writers for a number of standard file formats and MOF, a file format specialized for MOI (#969).\nImproved performance of deletion of vector of variables in MOI.Utilities.Model (#983).\nUpdated to MutableArithmetics v0.2 (#981).\nAdded MutableArithmetics.promote_operation allocation tests (#975).\nFixed inference issue on Julia v1.1 (#982).","category":"page"},{"location":"moi/changelog/#v0.9.8-(December-19,-2019)","page":"Release notes","title":"v0.9.8 (December 19, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented MutableArithmetics API (#924).\nFixed callbacks with CachingOptimizer (#959).\nFixed MOI.dimension for MOI.Complements (#948).\nAdded fallback for add_variables (#972).\nAdded is_diagonal_vectorized_index utility (#965).\nImproved linear constraints display in manual (#963, #964).\nBridges improvements:\nAdded IndicatorSet to SOS1 bridge (#877).\nAdded support for starting values for Variable.VectorizeBridge (#944).\nFixed MOI.add_constraints with non-bridged variable constraint on bridged variable (#951).\nFixed corner cases and docstring of GeoMeanBridge (#961, #962, #966).\nFixed choice between variable or constraint bridges for constrained variables (#973).\nImprove performance of bridge shortest path (#945, #946, #956).\nAdded docstring for test_delete_bridge (#954).\nAdded Variable bridge tests (#952).","category":"page"},{"location":"moi/changelog/#v0.9.7-(October-30,-2019)","page":"Release notes","title":"v0.9.7 (October 30, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented _result_index_field for NLPBlockDual (#934).\nFixed copy of model with starting values for vector constraints (#941).\nBridges improvements:\nImproved performance of add_bridge and added has_bridge (#935).\nAdded AbstractSetMapBridge for bridges between sets S1, S2 such that there is a linear map A such that A*S1 = S2 (#933).\nAdded support for starting values for FlipSignBridge, VectorizeBridge, ScalarizeBridge, SlackBridge, SplitIntervalBridge, RSOCBridge, SOCRBridge NormInfinityBridge, SOCtoPSDBridge and RSOCtoPSDBridge (#933, #936, #937, #938, #939).","category":"page"},{"location":"moi/changelog/#v0.9.6-(October-25,-2019)","page":"Release notes","title":"v0.9.6 (October 25, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added complementarity constraints (#913).\nAllowed ModelLike objects as value of attributes (#928).\nTesting improvements:\nAdded dual_objective_value option to MOI.Test.TestConfig (#922).\nAdded InvalidIndex tests in basic_constraint_tests (#921).\nAdded tests for the constant term in indicator constraint (#929).\nBridges improvements:\nAdded support for starting values for Functionize bridges (#923).\nAdded variable indices context to variable bridges (#920).\nFixed a typo in printing o debug_supports (#927).","category":"page"},{"location":"moi/changelog/#v0.9.5-(October-9,-2019)","page":"Release notes","title":"v0.9.5 (October 9, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Clarified PrimalStatus/DualStatus to be NO_SOLUTION if result_index is out of bounds (#912).\nAdded tolerance for checks  and use ResultCount + 1 for the result_index in MOI.Test.solve_result_status (#910, #917).\nUse 0.5 instead of 2.0 for power in PowerCone in basic_constraint_test (#916).\nBridges improvements:\nAdded debug utilities for unsupported variable/constraint/objective (#861).\nFixed deletion of variables in bridged VectorOfVariables constraints (#909).\nFixed result_index with objective bridges (#911).","category":"page"},{"location":"moi/changelog/#v0.9.4-(October-2,-2019)","page":"Release notes","title":"v0.9.4 (October 2, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Added solver-independent MIP callbacks (#782).\nImplements submit for Utilities.CachingOptimizer and Bridges.AbstractBridgeOptimizer (#906).\nAdded tests for result count of solution attributes (#901, #904).\nAdded NumberOfThreads attribute (#892).\nAdded Utilities.get_bounds to get the bounds on a variable (#890).\nAdded a note on duplicate coefficients in documentation (#581).\nAdded result index in ConstraintBasisStatus (#898).\nAdded extension dictionary to Utilities.Model (#884, #895).\nFixed deletion of constrained variables for CachingOptimizer (#905).\nImplemented Utilities.shift_constraint for Test.UnknownScalarSet (#896).\nBridges improvements:\nAdded Variable.RSOCtoSOCBridge (#907).\nImplemented MOI.get for ConstraintFunction/ConstraintSet for Bridges.Constraint.SquareBridge (#899).","category":"page"},{"location":"moi/changelog/#v0.9.3-(September-20,-2019)","page":"Release notes","title":"v0.9.3 (September 20, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixed ambiguity detected in Julia v1.3 (#891, #893).\nFixed missing sets from ListOfSupportedConstraints (#880).\nFixed copy of VectorOfVariables constraints with duplicate indices (#886).\nAdded extension dictionary to MOIU.Model (#884).\nImplemented MOI.get for function and set for GeoMeanBridge (#888).\nUpdated documentation for SingleVariable indices and bridges (#885).\nTesting improvements:\nAdded more comprehensive tests for names (#882).\nAdded tests for SingleVariable duals (#883).\nAdded tests for DualExponentialCone and DualPowerCone (#873).\nImprovements for arbitrary coefficient type:\nFixed == for sets with mutable fields (#887).\nRemoved some Float64 assumptions in bridges (#878).\nAutomatic selection of Constraint.[Scalar|Vector]FunctionizeBridge (#889).","category":"page"},{"location":"moi/changelog/#v0.9.2-(September-5,-2019)","page":"Release notes","title":"v0.9.2 (September 5, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Implemented model printing for MOI.ModelLike and specialized it for models defined in MOI (864).\nGeneralized contlinear tests for arbitrary coefficient type (#855).\nFixed supports_constraint for Semiinteger and Semicontinuous and supports for ObjectiveFunction (#859).\nFixed Allocate-Load copy for single variable constraints (#856).\nBridges improvements:\nAdd objective bridges (#789).\nFixed Variable.RSOCtoPSDBridge for dimension 2 (#869).\nAdded Variable.SOCtoRSOCBridge (#865).\nAdded Constraint.SOCRBridge and disable MOI.Bridges.Constraint.SOCtoPSDBridge (#751).\nFixed added_constraint_types for Contraint.LogDetBridge and Constraint.RootDetBridge (#870).","category":"page"},{"location":"moi/changelog/#v0.9.1-(August-22,-2019)","page":"Release notes","title":"v0.9.1 (August 22, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix support for Julia v1.2 (#834).\nL1 and L∞ norm epigraph cones and corresponding bridges to LP were added (#818).\nAdded tests to MOI.Test.nametest (#833).\nFix MOI.Test.soc3test for solvers not supporting infeasibility certificates (#839).\nImplements operate for operators * and / between vector function and constant (#837).\nImplements show for MOI.Utilities.IndexMap (#847).\nFix corner cases for mapping of variables in MOI.Utilities.CachingOptimizer and substitution of variables in MOI.Bridges.AbstractBridgeOptimizer (#848).\nFix transformation of constant terms for MOI.Bridges.Constraint.SOCtoPSDBridge and MOI.Bridges.Constraint.RSOCtoPSDBridge (#840).","category":"page"},{"location":"moi/changelog/#v0.9.0-(August-13,-2019)","page":"Release notes","title":"v0.9.0 (August 13, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Support for Julia v0.6 and v0.7 was dropped (#714, #717).\nA MOI.Utilities.Model implementation of ModelLike, this should replace most use cases of MOI.Utilities.@model (#781).\nadd_constrained_variable and add_constrained_variables were added (#759).\nSupport for indicator constraints was added (#709, #712).\nDualObjectiveValue attribute was added (#473).\nRawParameter attribute was added (#733).\nA dual_set function was added (#804).\nA Benchmarks submodule was added to facilitate solver benchmarking (#769).\nA submit function was added, this may for instance allow the user to submit solutions or cuts to the solver from a callback (#775).\nThe field of ObjectiveValue was renamed to result_index (#729).\nThe _constant and Utilities.getconstant function were renamed to constant\nREDUCTION_CERTIFICATE result status was added (#734).\nAbstract matrix sets were added (#731).\nTesting improvements:\nThe testing guideline was updated (#728).\nQuadratic tests were added (#697).\nUnit tests for RawStatusString, SolveTime, Silent and SolverName were added (#726, #741).\nA rotated second-order cone test was added (#759).\nA power cone test was added (#768).\nTests for ZeroOne variables with variable bounds were added (#772).\nAn unbounded test was added (#773).\nExisting tests had a few updates (#702, #703, #763).\nDocumentation improvements:\nAdded a section on CachingOptimizer (#777).\nAdded a section on UniversalFallback, Model and @model (#762).\nTransition the knapsack example to a doctest with MockOptimizer (#786).\nUtilities improvements:\nA CleverDict utility was added for a vector that automatically transform into a dictionary once a first index is removed (#767).\nThe Utilities.constant function was renamed to Utilities.constant_vector (#740).\nImplement optimizer attributes for CachingOptimizer (#745).\nRename Utilities.add_scalar_constraint to Utilities.normalize_and_add_constraint (#801).\noperate with vcat, SingleVariable and VectorOfVariables now returns a VectorOfVariables (#616).\nFix a type piracy of operate (#784).\nThe load_constraint fallback signature was fixed (#760).\nThe set_dot function was extended to work with sparse arrays (#805).\nBridges improvements:\nThe bridges no longer store the constraint function and set before it is bridged, the bridges now have to implement ConstraintFunction and ConstraintSet if the user wants to recover them. As a consequence, the @bridge macro was removed (#722).\nBridge are now instantiated with a bridge_constraint function instead of using a constructor (#730).\nFix constraint attributes for bridges (#699).\nConstraint bridges were moved to the Bridges/Constraint submodule so they should now inherit from MOI.Bridges.Constraint.Abstract and should implement MOI.Bridges.Constraint.concrete_bridge_type instead of MOI.Bridges.concrete_bridge_type (#756).\nVariable bridges were added in (#759).\nVarious improvements (#746, #747).","category":"page"},{"location":"moi/changelog/#v0.8.4-(March-13,-2019)","page":"Release notes","title":"v0.8.4 (March 13, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Performance improvement in default_copy_to and bridge optimizer (#696).\nAdd Silent and implement setting optimizer attributes in caching and mock optimizers (#695).\nAdd Functionize bridges (SingleVariable and VectorOfVariables) (#659).\nMinor typo fixes (#694).","category":"page"},{"location":"moi/changelog/#v0.8.3-(March-6,-2019)","page":"Release notes","title":"v0.8.3 (March 6, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Use zero constant in scalar constraint function of MOI.Test.copytest (#691).\nFix variable deletion with SingleVariable objective function (#690).\nFix LazyBridgeOptimizer with bridges that add no constraints (#689).\nError message improvements (#673, #685, #686, #688).\nDocumentation improvements (#682, #683, #687).\nBasis status:\nRemove VariableBasisStatus (#679).\nTest ConstraintBasisStatus and implement it in bridges (#678).\nFix inference of NumberOfVariables and NumberOfConstraints (#677).\nImplement division between a quadratic function and a number (#675).","category":"page"},{"location":"moi/changelog/#v0.8.2-(February-7,-2019)","page":"Release notes","title":"v0.8.2 (February 7, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add RawStatusString attribute (#629).\nDo not set names to the optimizer but only to the cache in CachingOptimizer (#638).\nMake scalar MOI functions act as scalars in broadcast (#646).\nAdd function utilities:\nImplement Base.zero (#634), Base.iszero (#643), add missing arithmetic operations (#644, #645) and fix division (#648).\nAdd a vectorize function that turns a vector of ScalarAffineFunction into a VectorAffineFunction (#642).\nImprove support for starting values:\nShow a warning in copy when starting values are not supported instead of throwing an error (#630).\nFix UniversalFallback for getting an variable or constraint attribute set to no indices (#623).\nAdd a test in contlineartest with partially set VariablePrimalStart.\nBridges improvements:\nFix StackOverFlow in LazyBridgeOptimizer when there is a cycle in the graph of bridges.\nAdd Slack bridges (#610, #650).\nAdd FlipSign bridges (#658).\nAdd tests with duplicate coefficients in ScalarAffineFunction and VectorAffineFunction (#639).\nUse tolerance to compare VariablePrimal in rotatedsoc1 test (#632).\nUse a zero constant in ScalarAffineFunction of constraints in psdt2 (#622).","category":"page"},{"location":"moi/changelog/#v0.8.1-(January-7,-2019)","page":"Release notes","title":"v0.8.1 (January 7, 2019)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Adding an NLP objective now overrides any objective set using the ObjectiveFunction attribute (#619).\nRename fullbridgeoptimizer into full_bridge_optimizer (#621).\nAllow custom constraint types with full_bridge_optimizer (#617).\nAdd Vectorize bridge which transforms scalar linear constraints into vector linear constraints (#615).","category":"page"},{"location":"moi/changelog/#v0.8.0-(December-18,-2018)","page":"Release notes","title":"v0.8.0 (December 18, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Rename all enum values to follow the JuMP naming guidelines for constants, for example, Optimal becomes OPTIMAL, and DualInfeasible becomes DUAL_INFEASIBLE.\nRename CachingOptimizer methods for style compliance.\nAdd an MOI.TerminationStatusCode called ALMOST_DUAL_INFEASIBLE.","category":"page"},{"location":"moi/changelog/#v0.7.0-(December-13,-2018)","page":"Release notes","title":"v0.7.0 (December 13, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Test that MOI.TerminationStatus is MOI.OptimizeNotCalled before MOI.optimize! is called.\nCheck supports_default_copy_to in tests (#594).\nKey pieces of information like optimality, infeasibility, etc., are now reported through TerminationStatusCode. It is typically no longer necessary to check the result statuses in addition to the termination status.\nAdd perspective dimension to log-det cone (#593).","category":"page"},{"location":"moi/changelog/#v0.6.4-(November-27,-2018)","page":"Release notes","title":"v0.6.4 (November 27, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Add OptimizeNotCalled termination status (#577) and improve documentation of other statuses (#575).\nAdd a solver naming guideline (#578).\nMake FeasibilitySense the default ObjectiveSense (#579).\nFix Utilities.@model and Bridges.@bridge macros for functions and sets defined outside MOI (#582).\nDocument solver-specific attributes (#580) and implement them in Utilities.CachingOptimizer (#565).","category":"page"},{"location":"moi/changelog/#v0.6.3-(November-16,-2018)","page":"Release notes","title":"v0.6.3 (November 16, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Variables and constraints are now allowed to have duplicate names. An error is thrown only on lookup. This change breaks some existing tests. (#549)\nAttributes may now be partially set (some values could be nothing). (#563)\nPerformance improvements in Utilities.Model (#549, #567, #568)\nFix bug in QuadtoSOC (#558).\nNew supports_default_copy_to method that optimizers should implement to control caching behavior.\nDocumentation improvements.","category":"page"},{"location":"moi/changelog/#v0.6.2-(October-26,-2018)","page":"Release notes","title":"v0.6.2 (October 26, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improve hygiene of @model macro (#544).\nFix bug in copy tests (#543).\nFix bug in UniversalFallback attribute getter (#540).\nAllow all correct solutions for solve_blank_obj unit test (#537).\nAdd errors for Allocate-Load and bad constraints (#534).\n[performance] Add specialized implementation of hash for VariableIndex (#533).\n[performance] Construct the name to object dictionaries lazily in model (#535).\nAdd the QuadtoSOC bridge which transforms ScalarQuadraticFunction constraints into RotatedSecondOrderCone (#483).","category":"page"},{"location":"moi/changelog/#v0.6.1-(September-22,-2018)","page":"Release notes","title":"v0.6.1 (September 22, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Enable PositiveSemidefiniteConeSquare set and quadratic functions in MOIB.fullbridgeoptimizer (#524).\nAdd warning in the bridge between PositiveSemidefiniteConeSquare and PositiveSemidefiniteConeTriangle when the matrix is almost symmetric (#522).\nModify MOIT.copytest to not add multiples constraints on the same variable (#521).\nAdd missing keyword argument in one of MOIU.add_scalar_constraint methods (#520).","category":"page"},{"location":"moi/changelog/#v0.6.0-(August-30,-2018)","page":"Release notes","title":"v0.6.0 (August 30, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"The MOIU.@model and MOIB.@bridge macros now support functions and sets defined in external modules. As a consequence, function and set names in the macro arguments need to be prefixed by module name.\nRename functions according to the JuMP style guide:\ncopy! with keyword arguments copynames and warnattributes -> copy_to with keyword arguments copy_names and warn_attributes;\nset! -> set;\naddvariable[s]! -> add_variable[s];\nsupportsconstraint -> supports_constraint;\naddconstraint[s]! -> add_constraint[s];\nisvalid -> is_valid;\nisempty -> is_empty;\nBase.delete! -> delete;\nmodify! -> modify;\ntransform! -> transform;\ninitialize! -> initialize;\nwrite -> write_to_file; and\nread! -> read_from_file.\nRemove free! (use Base.finalize instead).\nAdd the SquarePSD bridge which transforms PositiveSemidefiniteConeTriangle constraints into PositiveSemidefiniteConeTriangle.\nAdd result fallback for ConstraintDual of variable-wise constraint, ConstraintPrimal and ObjectiveValue.\nAdd tests for ObjectiveBound.\nAdd test for empty rows in vector linear constraint.\nRework errors: CannotError has been renamed NotAllowedError and the distinction between UnsupportedError and NotAllowedError is now about whether the element is not supported (for example, it cannot be copied a model containing this element) or the operation is not allowed (either because it is not implemented, because it cannot be performed in the current state of the model, or because it cannot be performed for a specific index)\ncanget is removed. NoSolution is added as a result status to indicate that the solver does not have either a primal or dual solution available (See #479).","category":"page"},{"location":"moi/changelog/#v0.5.0-(August-5,-2018)","page":"Release notes","title":"v0.5.0 (August 5, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fix names with CachingOptimizer.\nCleanup thanks to @mohamed82008.\nAdded a universal fallback for constraints.\nFast utilities for function canonicalization thanks to @rdeits.\nRenamed dimension field to side_dimension in the context of matrix-like sets.\nNew and improved tests for cases like duplicate terms and ObjectiveBound.\nRemoved cantransform, canaddconstraint, canaddvariable, canset, canmodify, and candelete functions from the API. They are replaced by a new set of errors that are thrown: Subtypes of UnsupportedError indicate unsupported operations, while subtypes of CannotError indicate operations that cannot be performed in the current state.\nThe API for copy! is updated to remove the CopyResult type.\nUpdates for the new JuMP style guide.","category":"page"},{"location":"moi/changelog/#v0.4.1-(June-28,-2018)","page":"Release notes","title":"v0.4.1 (June 28, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Fixes vector function modification on 32 bits.\nFixes Bellman-Ford algorithm for bridges.\nAdded an NLP test with FeasibilitySense.\nUpdate modification documentation.","category":"page"},{"location":"moi/changelog/#v0.4.0-(June-23,-2018)","page":"Release notes","title":"v0.4.0 (June 23, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Helper constructors for VectorAffineTerm and VectorQuadraticTerm.\nAdded modify_lhs to TestConfig.\nAdditional unit tests for optimizers.\nAdded a type parameter to CachingOptimizer for the optimizer field.\nNew API for problem modification (#388)\nTests pass without deprecation warnings on Julia 0.7.\nSmall fixes and documentation updates.","category":"page"},{"location":"moi/changelog/#v0.3.0-(May-25,-2018)","page":"Release notes","title":"v0.3.0 (May 25, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Functions have been redefined to use arrays-of-structs instead of structs-of-arrays.\nImprovements to MockOptimizer.\nSignificant changes to Bridges.\nNew and improved unit tests.\nFixes for Julia 0.7.","category":"page"},{"location":"moi/changelog/#v0.2.0-(April-24,-2018)","page":"Release notes","title":"v0.2.0 (April 24, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Improvements to and better coverage of Tests.\nDocumentation fixes.\nSolverName attribute.\nChanges to the NLP interface (new definition of variable order and arrays of structs for bound pairs and sparsity patterns).\nAddition of NLP tests.\nIntroduction of UniversalFallback.\ncopynames keyword argument to MOI.copy!.\nAdd Bridges submodule.","category":"page"},{"location":"moi/changelog/#v0.1.0-(February-28,-2018)","page":"Release notes","title":"v0.1.0 (February 28, 2018)","text":"","category":"section"},{"location":"moi/changelog/","page":"Release notes","title":"Release notes","text":"Initial public release.\nThe framework for MOI was developed at the JuMP-dev workshop at MIT in June 2017 as a sorely needed replacement for MathProgBase.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/overview.md\"","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/overview/#The-Bridges-submodule","page":"Overview","title":"The Bridges submodule","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The Bridges module simplifies the process of converting models between equivalent formulations.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nRead our paper for more details on how bridges are implemented.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Why-bridges?","page":"Overview","title":"Why bridges?","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"A constraint can often be written in a number of equivalent formulations. For example, the constraint l le a^top x le u (ScalarAffineFunction-in-Interval) could be re-formulated as two constraints: a^top x ge l (ScalarAffineFunction-in-GreaterThan) and a^top x le u (ScalarAffineFunction-in-LessThan). An alternative re-formulation is to add a dummy variable y with the constraints l le y le u (VariableIndex-in-Interval) and a^top x - y = 0 (ScalarAffineFunction-in-EqualTo).","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"To avoid each solver having to code these transformations manually, MathOptInterface provides bridges.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"A bridge is a small transformation from one constraint type to another (potentially collection of) constraint type.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Because these bridges are included in MathOptInterface, they can be re-used by any optimizer. Some bridges also implement constraint modifications and constraint primal and dual translations.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Several bridges can be used in combination to transform a single constraint into a form that the solver may understand. Choosing the bridges to use takes the form of finding a shortest path in the hyper-graph of bridges. The methodology is detailed in the MOI paper.","category":"page"},{"location":"moi/submodules/Bridges/overview/#The-three-types-of-bridges","page":"Overview","title":"The three types of bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"There are three types of bridges in MathOptInterface:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Constraint bridges\nVariable bridges\nObjective bridges","category":"page"},{"location":"moi/submodules/Bridges/overview/#Constraint-bridges","page":"Overview","title":"Constraint bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Constraint bridges convert constraints formulated by the user into an equivalent form supported by the solver. Constraint bridges are subtypes of Bridges.Constraint.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"In particular, constraint bridges can focus on rewriting the function of a constraint, and do not change the set. Function bridges are subtypes of Bridges.Constraint.AbstractFunctionConversionBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented constraint bridges for more details on the types of transformations that are available. Function bridges are Bridges.Constraint.ScalarFunctionizeBridge and Bridges.Constraint.VectorFunctionizeBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/#variable_bridges","page":"Overview","title":"Variable bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Variable bridges convert variables added by the user, either free with add_variable/add_variables, or constrained with add_constrained_variable/add_constrained_variables, into an equivalent form supported by the solver. Variable bridges are subtypes of Bridges.Variable.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented variable bridges for more details on the types of transformations that are available.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Objective-bridges","page":"Overview","title":"Objective bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Objective bridges convert the ObjectiveFunction set by the user into an equivalent form supported by the solver. Objective bridges are subtypes of Bridges.Objective.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"The equivalent formulation may add constraints (and possibly also variables) in the underlying model.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Read the list of implemented objective bridges for more details on the types of transformations that are available.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Bridges.full_bridge_optimizer","page":"Overview","title":"Bridges.full_bridge_optimizer","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nUnless you have an advanced use-case, this is probably the only function you need to care about.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"To enable the full power of MathOptInterface's bridges, wrap an optimizer in a Bridges.full_bridge_optimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.full_bridge_optimizer(inner_optimizer, Float64)\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Now, use optimizer as normal, and bridging will happen lazily behind the scenes. By lazily, we mean that bridging will happen if and only if the constraint is not supported by the inner_optimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"info: Info\nMost bridges are added by default in Bridges.full_bridge_optimizer. However, for technical reasons, some bridges are not added by default. Three examples include Bridges.Constraint.SOCtoPSDBridge, Bridges.Constraint.SOCtoNonConvexQuadBridge and Bridges.Constraint.RSOCtoNonConvexQuadBridge. See the docs of those bridges for more information.","category":"page"},{"location":"moi/submodules/Bridges/overview/#Add-a-single-bridge","page":"Overview","title":"Add a single bridge","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"If you don't want to use Bridges.full_bridge_optimizer, you can wrap an optimizer in a single bridge.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"However, this will force the constraint to be bridged, even if the inner_optimizer supports it.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.Constraint.SplitInterval{Float64}(inner_optimizer)\nMOIB.Constraint.SingleBridgeOptimizer{MOIB.Constraint.SplitIntervalBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\njulia> x = MOI.add_variable(optimizer)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(optimizer, x, MOI.Interval(0.0, 1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\njulia> MOI.get(optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})\n\njulia> MOI.get(inner_optimizer, MOI.ListOfConstraintTypesPresent())\n2-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64})\n (MathOptInterface.VariableIndex, MathOptInterface.LessThan{Float64})","category":"page"},{"location":"moi/submodules/Bridges/overview/#Bridges.LazyBridgeOptimizer","page":"Overview","title":"Bridges.LazyBridgeOptimizer","text":"","category":"section"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"If you don't want to use Bridges.full_bridge_optimizer, but you need more than a single bridge (or you want the bridging to happen lazily), you can manually construct a Bridges.LazyBridgeOptimizer.","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"First, wrap an inner optimizer:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> inner_optimizer = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> optimizer = MOI.Bridges.LazyBridgeOptimizer(inner_optimizer)\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Then use Bridges.add_bridge to add individual bridges:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Bridges.add_bridge(optimizer, MOI.Bridges.Constraint.SplitIntervalBridge{Float64})\n\njulia> MOI.Bridges.add_bridge(optimizer, MOI.Bridges.Objective.FunctionizeBridge{Float64})","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"Now the constraints will be bridged only if needed:","category":"page"},{"location":"moi/submodules/Bridges/overview/","page":"Overview","title":"Overview","text":"julia> x = MOI.add_variable(optimizer)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(optimizer, x, MOI.Interval(0.0, 1.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\njulia> MOI.get(optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})\n\njulia> MOI.get(inner_optimizer, MOI.ListOfConstraintTypesPresent())\n1-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64})","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"EditURL = \"getting_started_with_data_and_plotting.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Getting-started-with-data-and-plotting","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"In this tutorial we will learn how to read tabular data into Julia, and some of the basics of plotting.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"If you're new to Julia, start by reading Getting started with Julia and Getting started with JuMP first.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"note: Note\nThere are multiple ways to read the same kind of data into Julia. This tutorial focuses on DataFrames.jl because it provides the ecosystem to work with most of the required file types in a straightforward manner.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Before we get started, we need this constant to point to where the data files are.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import JuMP\nconst DATA_DIR = joinpath(\n    dirname(pathof(JuMP)),\n    joinpath(\"..\", \"docs\", \"src\", \"tutorials\", \"getting_started\", \"data\"),\n);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Where-to-get-help","page":"Getting started with data and plotting","title":"Where to get help","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Read the documentation","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.jl: http://docs.juliaplots.org/latest/\nCSV.jl: http://csv.juliadata.org/stable\nDataFrames.jl: https://dataframes.juliadata.org/stable/","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Preliminaries","page":"Getting started with data and plotting","title":"Preliminaries","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To get started, we need to install some packages.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#DataFrames.jl","page":"Getting started with data and plotting","title":"DataFrames.jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The DataFrames package provides a set of tools for working with tabular data. It is available through the Julia package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using Pkg\nPkg.add(\"DataFrames\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import DataFrames","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"info: What is a DataFrame?\nA DataFrame is a data structure like a table or spreadsheet. You can use it for storing and exploring a set of related data values. Think of it as a smarter array for holding tabular data.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Plots.jl","page":"Getting started with data and plotting","title":"Plots.jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The Plots package provides a set of tools for plotting. It is available through the Julia package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using Pkg\nPkg.add(\"Plots\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import Plots","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#CSV-.jl","page":"Getting started with data and plotting","title":"CSV .jl","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"CSV and other delimited text files can be read by the CSV.jl package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Pkg.add(\"CSV\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"import CSV","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#DataFrame-basics","page":"Getting started with data and plotting","title":"DataFrame basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To read a CSV file into a DataFrame, we use the CSV.read function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df = CSV.read(joinpath(DATA_DIR, \"StarWars.csv\"), DataFrames.DataFrame)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's try plotting some of this data","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.scatter(\n    csv_df.Weight,\n    csv_df.Height;\n    xlabel = \"Weight\",\n    ylabel = \"Height\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"That doesn't look right. What happened? If you look at the dataframe above, it read Weight in as a String column because there are \"NA\" fields. Let's correct that, by telling CSV to consider \"NA\" as missing.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df = CSV.read(\n    joinpath(DATA_DIR, \"StarWars.csv\"),\n    DataFrames.DataFrame;\n    missingstring = \"NA\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then let's re-plot our data","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Plots.scatter(\n    csv_df.Weight,\n    csv_df.Height;\n    title = \"Height vs Weight of StarWars characters\",\n    xlabel = \"Weight\",\n    ylabel = \"Height\",\n    label = false,\n    ylims = (0, 3),\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"That looks better.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"tip: Tip\nRead the CSV documentation for other parsing options.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.jl supports manipulation using functions similar to pandas. For example, split the dataframe into groups based on eye-color:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"by_eyecolor = DataFrames.groupby(csv_df, :Eyecolor)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then recombine into a single dataframe based on a function operating over the split dataframes:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"eyecolor_count = DataFrames.combine(by_eyecolor) do df\n    return DataFrames.nrow(df)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can rename columns:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.rename!(eyecolor_count, :x1 => :count)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Drop some missing rows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.dropmissing!(eyecolor_count, :Eyecolor)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then we can visualize the data:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"sort!(eyecolor_count, :count; rev = true)\nPlots.bar(\n    eyecolor_count.Eyecolor,\n    eyecolor_count.count;\n    xlabel = \"Eye color\",\n    ylabel = \"Number of characters\",\n    label = false,\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Other-Delimited-Files","page":"Getting started with data and plotting","title":"Other Delimited Files","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also use the CSV.jl package to read any other delimited text file format.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"By default, CSV.File will try to detect a file's delimiter from the first 10 lines of the file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Candidate delimiters include ',', '\\t', ' ', '|', ';', and ':'. If it can't auto-detect the delimiter, it will assume ','.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's take the example of space separated data.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"ss_df = CSV.read(joinpath(DATA_DIR, \"Cereal.txt\"), DataFrames.DataFrame)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also specify the delimiter as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"delim_df = CSV.read(\n    joinpath(DATA_DIR, \"Soccer.txt\"),\n    DataFrames.DataFrame;\n    delim = \"::\",\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Working-with-DataFrames","page":"Getting started with data and plotting","title":"Working with DataFrames","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Now that we have read the required data into a DataFrame, let us look at some basic operations we can perform on it.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Querying-Basic-Information","page":"Getting started with data and plotting","title":"Querying Basic Information","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The size function gets us the dimensions of the DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.size(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can also use the nrow and ncol functions to get the number of rows and columns respectively:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.nrow(ss_df), DataFrames.ncol(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The describe function gives basic summary statistics of data in a DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.describe(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Names of every column can be obtained by the names function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"DataFrames.names(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Corresponding data types are obtained using the broadcasted eltype function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"eltype.(ss_df)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Accessing-the-Data","page":"Getting started with data and plotting","title":"Accessing the Data","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Similar to regular arrays, we use numerical indexing to access elements of a DataFrame:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The following are different ways to access a column:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[!, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[!, :Name]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df.Name","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[:, 1] # Note that this creates a copy.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The following are different ways to access a row:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1:1, :]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1, :] # This produces a DataFrameRow.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can change the values just as we normally assign values.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Assign a range to scalar:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[1:3, :Height] .= 1.83","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Assign a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df[4:6, :Height] = [1.8, 1.6, 1.8]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"csv_df","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"tip: Tip\nThere are a lot more things which can be done with a DataFrame. Read the docs for more information.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"For information on dplyr-type syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Read the DataFrames.jl documentation\nCheck out DataFramesMeta.jl","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Example:-the-passport-problem","page":"Getting started with data and plotting","title":"Example: the passport problem","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's now apply what we have learned to solve a real problem.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Data-manipulation","page":"Getting started with data and plotting","title":"Data manipulation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The Passport Index Dataset lists travel visa requirements for 199 countries, in .csv format. Our task is to find the minimum number of passports required to visit all countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"passport_data = CSV.read(\n    joinpath(DATA_DIR, \"passport-index-matrix.csv\"),\n    DataFrames.DataFrame,\n);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"In this dataset, the first column represents a passport (=from) and each remaining column represents a foreign country (=to).","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values in each cell are as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"3 = visa-free travel\n2 = eTA is required\n1 = visa can be obtained on arrival\n0 = visa is required\n-1 is for all instances where passport and destination are the same","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Our task is to find out the minimum number of passports needed to visit every country without requiring a visa.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values we are interested in are -1 and 3. Let's modify the dataframe so that the -1 and 3 are 1 (true), and all others are 0 (false):","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"function modifier(x)\n    if x == -1 || x == 3\n        return 1\n    else\n        return 0\n    end\nend\n\nfor country in passport_data.Passport\n    passport_data[!, country] = modifier.(passport_data[!, country])\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"The values in the cells now represent:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"1 = no visa required for travel\n0 = visa required for travel","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#JuMP-Modeling","page":"Getting started with data and plotting","title":"JuMP Modeling","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"To model the problem as a mixed-integer linear program, we need a binary decision variable x_c for each country c. x_c is 1 if we select passport c and 0 otherwise. Our objective is to minimize the sum sum x_c over all countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Since we wish to visit all the countries, for every country, we must own at least one passport that lets us travel to that country visa free. For one destination, this can be mathematically represented as sum_c in C a_cd cdot x_d geq 1, where a is the passport_data dataframe.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Thus, we can represent this problem using the following model:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"beginaligned\nmin  sum_c in C x_c \ntextst  sum_c in C a_cd x_c geq 1  forall d in C \n x_c in 01  forall c in C\nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We'll now solve the problem using JuMP:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"First, create the set of countries:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"C = passport_data.Passport","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Then, create the model and initialize the decision variables:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x[C], Bin)\n@objective(model, Min, sum(x))\n@constraint(model, [d in C], passport_data[!, d]' * x >= 1)\nmodel","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Now optimize:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"optimize!(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We can use the solution_summary function to get an overview of the solution:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"solution_summary(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/#Solution","page":"Getting started with data and plotting","title":"Solution","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"Let's  have a look at the solution in more detail:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"println(\"Minimum number of passports needed: \", objective_value(model))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"println(\"Optimal passports:\")\nfor c in C\n    if value(x[c]) > 0.5\n        println(\" * \", c)\n    end\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"We need some passports, like New Zealand and the United States, which have widespread access to a large number of countries. However, we also need passports like North Korea which only have visa-free access to a very limited number of countries.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_data_and_plotting/","page":"Getting started with data and plotting","title":"Getting started with data and plotting","text":"note: Note\nWe use value(x[c]) > 0.5 rather than value(x[c]) == 1 to avoid excluding solutions like x[c] = 0.99999 that are \"1\" to some tolerance.","category":"page"},{"location":"developers/roadmap/#Development-roadmap","page":"Roadmap","title":"Development roadmap","text":"","category":"section"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"The JuMP developers have compiled this roadmap document to share their plans and goals with the JuMP community. Contributions to roadmap issues are especially invited.","category":"page"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"Most of these issues will require changes to both JuMP and MathOptInterface, and are non-trivial in their implementation. They are in no particular order, but represent broad themes that we see as areas in which JuMP could be improved.","category":"page"},{"location":"developers/roadmap/","page":"Roadmap","title":"Roadmap","text":"Done #3106 Make nonlinear programming a first-class citizen. There have been many issues and discussions about this: currently nonlinear constraints are handled through a MOI.NLPBlock and have various limitations and restrictions.\nhttps://github.com/jump-dev/JuMP.jl/issues/1185\nhttps://github.com/jump-dev/JuMP.jl/issues/1198\nhttps://github.com/jump-dev/JuMP.jl/issues/2788\nhttps://github.com/jump-dev/MathOptInterface.jl/issues/846\nhttps://github.com/jump-dev/MathOptInterface.jl/issues/1397\nDone #3385 Add support for coefficient types other than Float64: https://github.com/jump-dev/JuMP.jl/issues/2025 Since the very beginning, JuMP has hard-coded the coefficient type as Float64. This has made it impossible to support solvers which can use other types such as BigFloat or Rational{BigInt}.\nAdd support for constraint programming: https://github.com/jump-dev/JuMP.jl/issues/2227 JuMP has a strong focus on linear, conic and nonlinear optimization problems. We want to add better support for constraint programming.\nDone #3176 Add support for multiobjective problems: https://github.com/jump-dev/JuMP.jl/issues/2099 JuMP is restricted to problems with scalar-valued objectives. We want to extend this to vector-valued problems.\nRefactor the internal code of JuMP's macros. The code in src/macros.jl is some of the oldest part of JuMP and is difficult to read, modify, and extend. We should overhaul the internals of JuMP's macros–-without making user-visible breaking changes–-to improve their long-term maintainability.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"EditURL = \"simple_examples.jl\"","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Simple-examples","page":"Simple examples","title":"Simple examples","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This tutorial is a collection of examples of small nonlinear programs. It uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"using JuMP\nimport Ipopt\nimport Random\nimport Statistics\nimport Test","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#The-Rosenbrock-function","page":"Simple examples","title":"The Rosenbrock function","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"A nonlinear example of the classical Rosenbrock function.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_rosenbrock()\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, x)\n    @variable(model, y)\n    @objective(model, Min, (1 - x)^2 + 100 * (y - x^2)^2)\n    optimize!(model)\n    Test.@test termination_status(model) == LOCALLY_SOLVED\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 0.0 atol = 1e-10\n    Test.@test value(x) ≈ 1.0\n    Test.@test value(y) ≈ 1.0\n    return\nend\n\nexample_rosenbrock()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#The-clnlbeam-problem","page":"Simple examples","title":"The clnlbeam problem","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Based on an AMPL model by Hande Y. Benson","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Copyright (C) 2001 Princeton University All Rights Reserved","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Permission to use, copy, modify, and distribute this software and its documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appear in all copies and that the copyright notice and this permission notice appear in all supporting documentation.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"Source:","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"H. Maurer and H.D. Mittelman, \"The non-linear beam via optimal control with bound state variables,\" Optimal Control Applications and Methods 12, pp. 19-31, 1991.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_clnlbeam()\n    N = 1000\n    h = 1 / N\n    alpha = 350\n    model = Model(Ipopt.Optimizer)\n    @variables(model, begin\n        -1 <= t[1:(N+1)] <= 1\n        -0.05 <= x[1:(N+1)] <= 0.05\n        u[1:(N+1)]\n    end)\n    @objective(\n        model,\n        Min,\n        sum(\n            0.5 * h * (u[i+1]^2 + u[i]^2) +\n            0.5 * alpha * h * (cos(t[i+1]) + cos(t[i])) for i in 1:N\n        ),\n    )\n    @constraint(\n        model,\n        [i = 1:N],\n        x[i+1] - x[i] - 0.5 * h * (sin(t[i+1]) + sin(t[i])) == 0,\n    )\n    @constraint(\n        model,\n        [i = 1:N],\n        t[i+1] - t[i] - 0.5 * h * u[i+1] - 0.5 * h * u[i] == 0,\n    )\n    optimize!(model)\n    println(\"\"\"\n    termination_status = $(termination_status(model))\n    primal_status      = $(primal_status(model))\n    objective_value    = $(objective_value(model))\n    \"\"\")\n    return\nend\n\nexample_clnlbeam()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Maximum-likelihood-estimation","page":"Simple examples","title":"Maximum likelihood estimation","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"This example uses nonlinear optimization to compute the maximum likelihood estimate (MLE) of the parameters of a normal distribution, a.k.a., the sample mean and variance.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_mle()\n    n = 1_000\n    Random.seed!(1234)\n    data = randn(n)\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, μ, start = 0.0)\n    @variable(model, σ >= 0.0, start = 1.0)\n    @objective(\n        model,\n        Max,\n        n / 2 * log(1 / (2 * π * σ^2)) -\n        sum((data[i] - μ)^2 for i in 1:n) / (2 * σ^2)\n    )\n    optimize!(model)\n    println(\"μ             = \", value(μ))\n    println(\"mean(data)    = \", Statistics.mean(data))\n    println(\"σ^2           = \", value(σ)^2)\n    println(\"var(data)     = \", Statistics.var(data))\n    println(\"MLE objective = \", objective_value(model))\n    Test.@test value(μ) ≈ Statistics.mean(data) atol = 1e-3\n    Test.@test value(σ)^2 ≈ Statistics.var(data) atol = 1e-2\n    # You can even do constrained MLE!\n    @constraint(model, μ == σ^2)\n    optimize!(model)\n    Test.@test value(μ) ≈ value(σ)^2\n    println()\n    println(\"With constraint μ == σ^2:\")\n    println(\"μ                         = \", value(μ))\n    println(\"σ^2                       = \", value(σ)^2)\n    println(\"Constrained MLE objective = \", objective_value(model))\n    return\nend\n\nexample_mle()","category":"page"},{"location":"tutorials/nonlinear/simple_examples/#Quadratically-constrained-programs","page":"Simple examples","title":"Quadratically constrained programs","text":"","category":"section"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"A simple quadratically constrained program based on an example from Gurobi.","category":"page"},{"location":"tutorials/nonlinear/simple_examples/","page":"Simple examples","title":"Simple examples","text":"function example_qcp()\n    model = Model(Ipopt.Optimizer)\n    set_silent(model)\n    @variable(model, x)\n    @variable(model, y >= 0)\n    @variable(model, z >= 0)\n    @objective(model, Max, x)\n    @constraint(model, x + y + z == 1)\n    @constraint(model, x * x + y * y - z * z <= 0)\n    @constraint(model, x * x - y * z <= 0)\n    optimize!(model)\n    print(model)\n    println(\"Objective value: \", objective_value(model))\n    println(\"x = \", value(x))\n    println(\"y = \", value(y))\n    Test.@test termination_status(model) == LOCALLY_SOLVED\n    Test.@test primal_status(model) == FEASIBLE_POINT\n    Test.@test objective_value(model) ≈ 0.32699 atol = 1e-5\n    Test.@test value(x) ≈ 0.32699 atol = 1e-5\n    Test.@test value(y) ≈ 0.25707 atol = 1e-5\n    return\nend\n\nexample_qcp()","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"EditURL = \"https://github.com/jump-dev/GLPK.jl/blob/v1.1.3/README.md\"","category":"page"},{"location":"packages/GLPK/#GLPK.jl","page":"jump-dev/GLPK.jl","title":"GLPK.jl","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK.jl is a wrapper for the GNU Linear Programming Kit library.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/GLPK/#Affiliation","page":"jump-dev/GLPK.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"This wrapper is maintained by the JuMP community and is not an GNU project.","category":"page"},{"location":"packages/GLPK/#License","page":"jump-dev/GLPK.jl","title":"License","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK.jl is licensed under the GPL v3 license.","category":"page"},{"location":"packages/GLPK/#Installation","page":"jump-dev/GLPK.jl","title":"Installation","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Install GLPK using Pkg.add:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"import Pkg\nPkg.add(\"GLPK\")","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"In addition to installing the GLPK.jl package, this will also download and install the GLPK binaries. You do not need to install GLPK separately.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/GLPK/#Use-with-JuMP","page":"jump-dev/GLPK.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"To use GLPK with JuMP, use GLPK.Optimizer:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"using JuMP, GLPK\nmodel = Model(GLPK.Optimizer)\nset_attribute(model, \"tm_lim\", 60 * 1_000)\nset_attribute(model, \"msg_lev\", GLPK.GLP_MSG_OFF)","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"If the model is primal or dual infeasible, GLPK will attempt to find a certificate of infeasibility. This can be expensive, particularly if you do not intend to use the certificate. If this is the case, use:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"model = Model(() -> GLPK.Optimizer(; want_infeasibility_certificates = false))","category":"page"},{"location":"packages/GLPK/#MathOptInterface-API","page":"jump-dev/GLPK.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The GLPK optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.Reals","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"MOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/GLPK/#Options","page":"jump-dev/GLPK.jl","title":"Options","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Options for GLPK are comprehensively documented in the PDF documentation, but they are hard to find.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Options when solving a linear program are defined in Section 2.8.1\nOptions when solving a mixed-integer program are defined in Section 2.10.5","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"However, the following options are likely to be the most useful:","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Parameter Example Explanation\nmsg_lev GLPK.GLP_MSG_ALL Message level for terminal output\npresolve GLPK.GLP_ON Turn presolve on or off\ntol_int 1e-5 Absolute tolerance for integer feasibility\ntol_obj 1e-7 Relative objective tolerance for mixed-integer programs","category":"page"},{"location":"packages/GLPK/#Callbacks","page":"jump-dev/GLPK.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"Here is an example using GLPK's solver-specific callbacks.","category":"page"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"using JuMP, GLPK, Test\n\nmodel = Model(GLPK.Optimizer)\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\nreasons = UInt8[]\nfunction my_callback_function(cb_data)\n    reason = GLPK.glp_ios_reason(cb_data.tree)\n    push!(reasons, reason)\n    if reason != GLPK.GLP_IROWGEN\n        return\n    end\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\nend\nMOI.set(model, GLPK.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2\n@show reasons","category":"page"},{"location":"packages/GLPK/#C-API","page":"jump-dev/GLPK.jl","title":"C API","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"The C API can be accessed via GLPK.glp_XXX functions, where the names and arguments are identical to the C API. See the /tests folder for inspiration.","category":"page"},{"location":"packages/GLPK/#Thread-safety","page":"jump-dev/GLPK.jl","title":"Thread safety","text":"","category":"section"},{"location":"packages/GLPK/","page":"jump-dev/GLPK.jl","title":"jump-dev/GLPK.jl","text":"GLPK is not thread-safe and should not be used with multithreading.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"EditURL = \"callbacks.jl\"","category":"page"},{"location":"tutorials/linear/callbacks/#callbacks_tutorial","page":"Callbacks","title":"Callbacks","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"The purpose of the tutorial is to demonstrate the various solver-independent and solver-dependent callbacks that are supported by JuMP.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"The tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"using JuMP\nimport GLPK\nimport Random","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"info: Info\nThis tutorial uses the MathOptInterface API. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/linear/callbacks/#Lazy-constraints","page":"Callbacks","title":"Lazy constraints","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a lazy constraint callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_lazy_constraint()\n    model = Model(GLPK.Optimizer)\n    @variable(model, 0 <= x <= 2.5, Int)\n    @variable(model, 0 <= y <= 2.5, Int)\n    @objective(model, Max, y)\n    lazy_called = false\n    function my_callback_function(cb_data)\n        lazy_called = true\n        x_val = callback_value(cb_data, x)\n        y_val = callback_value(cb_data, y)\n        println(\"Called from (x, y) = ($x_val, $y_val)\")\n        status = callback_node_status(cb_data, model)\n        if status == MOI.CALLBACK_NODE_STATUS_FRACTIONAL\n            println(\" - Solution is integer infeasible!\")\n        elseif status == MOI.CALLBACK_NODE_STATUS_INTEGER\n            println(\" - Solution is integer feasible!\")\n        else\n            @assert status == MOI.CALLBACK_NODE_STATUS_UNKNOWN\n            println(\" - I don't know if the solution is integer feasible :(\")\n        end\n        if y_val - x_val > 1 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        elseif y_val + x_val > 3 + 1e-6\n            con = @build_constraint(y + x <= 3)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        end\n    end\n    set_attribute(model, MOI.LazyConstraintCallback(), my_callback_function)\n    optimize!(model)\n    println(\"Optimal solution (x, y) = ($(value(x)), $(value(y)))\")\n    return\nend\n\nexample_lazy_constraint()","category":"page"},{"location":"tutorials/linear/callbacks/#User-cuts","page":"Callbacks","title":"User-cuts","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a user-cut callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_user_cut_constraint()\n    Random.seed!(1)\n    N = 30\n    item_weights, item_values = rand(N), rand(N)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @constraint(model, sum(item_weights[i] * x[i] for i in 1:N) <= 10)\n    @objective(model, Max, sum(item_values[i] * x[i] for i in 1:N))\n    callback_called = false\n    function my_callback_function(cb_data)\n        callback_called = true\n        x_vals = callback_value.(Ref(cb_data), x)\n        accumulated = sum(item_weights[i] for i in 1:N if x_vals[i] > 1e-4)\n        println(\"Called with accumulated = $(accumulated)\")\n        n_terms = sum(1 for i in 1:N if x_vals[i] > 1e-4)\n        if accumulated > 10\n            con = @build_constraint(\n                sum(x[i] for i in 1:N if x_vals[i] > 0.5) <= n_terms - 1\n            )\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.UserCut(cb_data), con)\n        end\n    end\n    set_attribute(model, MOI.UserCutCallback(), my_callback_function)\n    optimize!(model)\n    @show callback_called\n    return\nend\n\nexample_user_cut_constraint()","category":"page"},{"location":"tutorials/linear/callbacks/#Heuristic-solutions","page":"Callbacks","title":"Heuristic solutions","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using a heuristic solution callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_heuristic_solution()\n    Random.seed!(1)\n    N = 30\n    item_weights, item_values = rand(N), rand(N)\n    model = Model(GLPK.Optimizer)\n    @variable(model, x[1:N], Bin)\n    @constraint(model, sum(item_weights[i] * x[i] for i in 1:N) <= 10)\n    @objective(model, Max, sum(item_values[i] * x[i] for i in 1:N))\n    callback_called = false\n    function my_callback_function(cb_data)\n        callback_called = true\n        x_vals = callback_value.(Ref(cb_data), x)\n        ret =\n            MOI.submit(model, MOI.HeuristicSolution(cb_data), x, floor.(x_vals))\n        println(\"Heuristic solution status = $(ret)\")\n    end\n    set_attribute(model, MOI.HeuristicCallback(), my_callback_function)\n    optimize!(model)\n    return\nend\n\nexample_heuristic_solution()","category":"page"},{"location":"tutorials/linear/callbacks/#GLPK-solver-dependent-callback","page":"Callbacks","title":"GLPK solver-dependent callback","text":"","category":"section"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"An example using GLPK's solver-dependent callback.","category":"page"},{"location":"tutorials/linear/callbacks/","page":"Callbacks","title":"Callbacks","text":"function example_solver_dependent_callback()\n    model = Model(GLPK.Optimizer)\n    @variable(model, 0 <= x <= 2.5, Int)\n    @variable(model, 0 <= y <= 2.5, Int)\n    @objective(model, Max, y)\n    lazy_called = false\n    function my_callback_function(cb_data)\n        lazy_called = true\n        reason = GLPK.glp_ios_reason(cb_data.tree)\n        println(\"Called from reason = $(reason)\")\n        if reason != GLPK.GLP_IROWGEN\n            return\n        end\n        x_val = callback_value(cb_data, x)\n        y_val = callback_value(cb_data, y)\n        if y_val - x_val > 1 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        elseif y_val + x_val > 3 + 1e-6\n            con = @build_constraint(y - x <= 1)\n            println(\"Adding $(con)\")\n            MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n        end\n    end\n    set_attribute(model, GLPK.CallbackFunction(), my_callback_function)\n    optimize!(model)\n    return\nend\n\nexample_solver_dependent_callback()","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/nlp/#Nonlinear-Modeling-(Legacy)","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nThis page describes the legacy nonlinear interface to JuMP. It has a number of quirks and limitations that prompted the development of a new nonlinear interface. The new interface is documented at Nonlinear Modeling. This legacy interface will remain for all future v1.X releases of JuMP. The two nonlinear interfaces cannot be combined.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP has support for general smooth nonlinear (convex and nonconvex) optimization problems. JuMP is able to provide exact, sparse second-order derivatives to solvers. This information can improve solver accuracy and performance.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"There are three main changes to solve nonlinear programs in JuMP.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLobjective instead of @objective\nUse @NLconstraint instead of @constraint\nUse @NLexpression instead of @expression","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nThere are some restrictions on what syntax you can use in the @NLxxx macros. Make sure to read the Syntax notes.","category":"page"},{"location":"manual/nlp/#Set-a-nonlinear-objective","page":"Nonlinear Modeling (Legacy)","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLobjective to set a nonlinear objective.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @NLobjective(model, Min, exp(x[1]) - sqrt(x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"To modify a nonlinear objective, call @NLobjective again.","category":"page"},{"location":"manual/nlp/#Add-a-nonlinear-constraint","page":"Nonlinear Modeling (Legacy)","title":"Add a nonlinear constraint","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLconstraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @NLconstraint(model, exp(x[1]) <= 1)\nexp(x[1]) - 1.0 ≤ 0\n\njulia> @NLconstraint(model, [i = 1:2], x[i]^i >= i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n x[1] ^ 1.0 - 1.0 ≥ 0\n x[2] ^ 2.0 - 2.0 ≥ 0\n\njulia> @NLconstraint(model, con[i = 1:2], prod(x[j] for j = 1:i) == i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n (*)(x[1]) - 1.0 = 0\n x[1] * x[2] - 2.0 = 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nYou can only create nonlinear constraints with <=, >=, and ==. More general Nonlinear-in-Set constraints are not supported.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Delete a nonlinear constraint using delete:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> delete(model, con[1])","category":"page"},{"location":"manual/nlp/#Create-a-nonlinear-expression","page":"Nonlinear Modeling (Legacy)","title":"Create a nonlinear expression","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use @NLexpression to create nonlinear expression objects. The syntax is identical to @expression, except that the expression can contain nonlinear terms.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> expr = @NLexpression(model, exp(x[1]) + sqrt(x[2]))\nsubexpression[1]: exp(x[1]) + sqrt(x[2])\n\njulia> my_anon_expr = @NLexpression(model, [i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[2]: sin(x[1])\n subexpression[3]: sin(x[2])\n\njulia> @NLexpression(model, my_expr[i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[4]: sin(x[1])\n subexpression[5]: sin(x[2])","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear expression can be used in @NLobjective, @NLconstraint, and even nested in other @NLexpressions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @NLobjective(model, Min, expr^2 + 1)\n\njulia> @NLconstraint(model, [i = 1:2], my_expr[i] <= i)\n2-element Vector{NonlinearConstraintRef{ScalarShape}}:\n subexpression[4] - 1.0 ≤ 0\n subexpression[5] - 2.0 ≤ 0\n\njulia> @NLexpression(model, nested[i = 1:2], sin(my_expr[i]))\n2-element Vector{NonlinearExpression}:\n subexpression[6]: sin(subexpression[4])\n subexpression[7]: sin(subexpression[5])","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use value to query the value of a nonlinear expression:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> set_start_value(x[1], 1.0)\n\njulia> value(start_value, nested[1])\n0.7456241416655579\n\njulia> sin(sin(1.0))\n0.7456241416655579","category":"page"},{"location":"manual/nlp/#Create-a-nonlinear-parameter","page":"Nonlinear Modeling (Legacy)","title":"Create a nonlinear parameter","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For nonlinear models only, JuMP offers a syntax for explicit \"parameter\" objects, which are constants in the model that can be efficiently updated between solves.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters are declared by using the @NLparameter macro and may be indexed by arbitrary sets analogously to JuMP variables and expressions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The initial value of the parameter must be provided on the right-hand side of the == sign.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLparameter(model, p[i = 1:2] == i)\n2-element Vector{NonlinearParameter}:\n parameter[1] == 1.0\n parameter[2] == 2.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Create anonymous parameters using the value keyword:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> anon_parameter = @NLparameter(model, value = 1)\nparameter[3] == 1.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nA parameter is not an optimization variable. It must be fixed to a value with ==. If you want a parameter that is <= or >=, create a variable instead using @variable.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use value and set_value to query or update the value of a parameter.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> value.(p)\n2-element Vector{Float64}:\n 1.0\n 2.0\n\njulia> set_value(p[2], 3.0)\n3.0\n\njulia> value.(p)\n2-element Vector{Float64}:\n 1.0\n 3.0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters must be used within nonlinear macros only.","category":"page"},{"location":"manual/nlp/#When-to-use-a-parameter","page":"Nonlinear Modeling (Legacy)","title":"When to use a parameter","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Nonlinear parameters are useful when solving nonlinear models in a sequence:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, z)\n@NLparameter(model, x == 1.0)\n@NLobjective(model, Min, (z - x)^2)\noptimize!(model)\n@show value(z) # Equals 1.0.\n\n# Now, update the value of x to solve a different problem.\nset_value(x, 5.0)\noptimize!(model)\n@show value(z) # Equals 5.0\nnothing #hide","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nUsing nonlinear parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.","category":"page"},{"location":"manual/nlp/#Syntax-notes","page":"Nonlinear Modeling (Legacy)","title":"Syntax notes","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The syntax accepted in nonlinear macros is more restricted than the syntax for linear and quadratic macros. We note some important points below.","category":"page"},{"location":"manual/nlp/#Scalar-operations-only","page":"Nonlinear Modeling (Legacy)","title":"Scalar operations only","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Except for the splatting syntax discussed below, all expressions must be simple scalar operations. You cannot use dot, matrix-vector products, vector slices, etc.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @variable(model, y);\n\njulia> c = [1, 2];\n\njulia> @NLobjective(model, Min, c' * x + 3y)\nERROR: Unexpected array [1 2] in nonlinear expression. Nonlinear expressions may contain only scalar expressions.\n[...]","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Translate vector operations into explicit sum() operations:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @NLobjective(model, Min, sum(c[i] * x[i] for i = 1:2) + 3y)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Or use an @expression:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> @expression(model, expr, c' * x)\nx[1] + 2 x[2]\n\njulia> @NLobjective(model, Min, expr + 3y)\n","category":"page"},{"location":"manual/nlp/#Splatting","page":"Nonlinear Modeling (Legacy)","title":"Splatting","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The splatting operator   ... is recognized in a very restricted setting for expanding function   arguments. The expression splatted can be only a symbol. More complex   expressions are not recognized.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x[1:3]);\n\njulia> @NLconstraint(model, *(x...) <= 1.0)\nx[1] * x[2] * x[3] - 1.0 ≤ 0\n\njulia> @NLconstraint(model, *((x / 2)...) <= 0.0)\nERROR: Unsupported use of the splatting operator. JuMP supports splatting only symbols. For example, `x...` is ok, but `(x + 1)...`, `[x; y]...` and `g(f(y)...)` are not.","category":"page"},{"location":"manual/nlp/#User-defined-Functions","page":"Nonlinear Modeling (Legacy)","title":"User-defined Functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP natively supports the set of univariate and multivariate functions recognized by the MOI.Nonlinear submodule. In addition to this list of functions, it is possible to register custom user-defined nonlinear functions. User-defined functions can be used anywhere in @NLobjective, @NLconstraint, and @NLexpression.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP will attempt to automatically register functions it detects in your nonlinear expressions, which usually means manually registering a function is not needed. Two exceptions are if you want to provide custom derivatives, or if the function is not available in the scope of the nonlinear expression.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nUser-defined functions must return a scalar output. For a work-around, see User-defined operators with vector outputs.","category":"page"},{"location":"manual/nlp/#Automatic-differentiation","page":"Nonlinear Modeling (Legacy)","title":"Automatic differentiation","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP does not support black-box optimization, so all user-defined functions must provide derivatives in some form. Fortunately, JuMP supports automatic differentiation of user-defined functions, a feature to our knowledge not available in any comparable modeling systems.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"info: Info\nAutomatic differentiation is not finite differencing. JuMP's automatically computed derivatives are not subject to approximation error.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"JuMP uses ForwardDiff.jl to perform automatic differentiation; see the ForwardDiff.jl documentation for a description of how to write a function suitable for automatic differentiation.","category":"page"},{"location":"manual/nlp/#Common-mistakes-when-writing-a-user-defined-function","page":"Nonlinear Modeling (Legacy)","title":"Common mistakes when writing a user-defined function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nGet an error like No method matching Float64(::ForwardDiff.Dual)? Read this section, and see the guidelines at ForwardDiff.jl.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The most common error is that your user-defined function is not generic with respect to the number type, that is, don't assume that the input to the function is Float64.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x::Float64) = 2 * x  # This will not work.\nf(x::Real)    = 2 * x  # This is good.\nf(x)          = 2 * x  # This is also good.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Another reason you may encounter this error is if you create arrays inside your function which are Float64.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"function bad_f(x...)\n    y = zeros(length(x))  # This constructs an array of `Float64`!\n    for i = 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend\n\nfunction good_f(x::T...) where {T<:Real}\n    y = zeros(T, length(x))  # Construct an array of type `T` instead!\n    for i = 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend","category":"page"},{"location":"manual/nlp/#Register-a-function","page":"Nonlinear Modeling (Legacy)","title":"Register a function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"To register a user-defined function with derivatives computed by automatic differentiation, use the register method as in the following example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\n\nmodel = Model()\n\nregister(model, :square, 1, square; autodiff = true)\nregister(model, :my_f, 2, f; autodiff = true)\n\n@variable(model, x[1:2] >= 0.5)\n@NLobjective(model, Min, my_f(x[1], square(x[2])))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The above code creates a JuMP model with the objective function (x[1] - 1)^2 + (x[2]^2 - 2)^2. The arguments to register are:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The model for which the functions are registered.\nA Julia symbol object which serves as the name of the user-defined function in JuMP expressions.\nThe number of input arguments that the function takes.\nThe Julia method which computes the function\nA flag to instruct JuMP to compute exact gradients automatically.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"tip: Tip\nThe symbol :my_f doesn't have to match the name of the function f. However, it's more readable if it does. Make sure you use my_f and not f in the macros.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nUser-defined functions cannot be re-registered and will not update if you modify the underlying Julia function. If you want to change a user-defined function between solves, rebuild the model or use a different name. To use a different name programmatically, see Raw expression input.","category":"page"},{"location":"manual/nlp/#Register-a-function-and-gradient","page":"Nonlinear Modeling (Legacy)","title":"Register a function and gradient","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Forward-mode automatic differentiation as implemented by ForwardDiff.jl has a computational cost that scales linearly with the number of input dimensions. As such, it is not the most efficient way to compute gradients of user-defined functions if the number of input arguments is large. In this case, users may want to provide their own routines for evaluating gradients.","category":"page"},{"location":"manual/nlp/#Univariate-functions","page":"Nonlinear Modeling (Legacy)","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For univariate functions, the gradient function ∇f returns a number that represents the first-order derivative:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x) = x^2\n∇f(x) = 2x\nmodel = Model()\nregister(model, :my_square, 1, f, ∇f; autodiff = true)\n@variable(model, x >= 0)\n@NLobjective(model, Min, my_square(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"If autodiff = true, JuMP will use automatic differentiation to compute the hessian.","category":"page"},{"location":"manual/nlp/#Multivariate-functions","page":"Nonlinear Modeling (Legacy)","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For multivariate functions, the gradient function ∇f must take a gradient vector as the first argument that is filled in-place:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x, y) = (x - 1)^2 + (y - 2)^2\nfunction ∇f(g::AbstractVector{T}, x::T, y::T) where {T}\n    g[1] = 2 * (x - 1)\n    g[2] = 2 * (y - 2)\n    return\nend\n\nmodel = Model()\nregister(model, :my_square, 2, f, ∇f)\n@variable(model, x[1:2] >= 0)\n@NLobjective(model, Min, my_square(x[1], x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nMake sure the first argument to ∇f supports an AbstractVector, and do not assume the input is Float64.","category":"page"},{"location":"manual/nlp/#Register-a-function,-gradient,-and-hessian","page":"Nonlinear Modeling (Legacy)","title":"Register a function, gradient, and hessian","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"You can also register a function with the second-order derivative information, which is a scalar for univariate functions, and a symmetric matrix for multivariate functions.","category":"page"},{"location":"manual/nlp/#Univariate-functions-2","page":"Nonlinear Modeling (Legacy)","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Pass a function which returns a number representing the second-order derivative:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x) = x^2\n∇f(x) = 2x\n∇²f(x) = 2\nmodel = Model()\nregister(model, :my_square, 1, f, ∇f, ∇²f)\n@variable(model, x >= 0)\n@NLobjective(model, Min, my_square(x))","category":"page"},{"location":"manual/nlp/#Multivariate-functions-2","page":"Nonlinear Modeling (Legacy)","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For multivariate functions, the hessian function ∇²f must take an AbstractMatrix as the first argument, the lower-triangular of which is filled in-place:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nf(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2\nfunction ∇f(g, x...)\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend\nfunction ∇²f(H, x...)\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1]  <-- Not needed. Fill the lower-triangular only.\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend\n\nmodel = Model()\nregister(model, :rosenbrock, 2, f, ∇f, ∇²f)\n@variable(model, x[1:2])\n@NLobjective(model, Min, rosenbrock(x[1], x[2]))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nYou may assume the Hessian matrix H is initialized with zeros, and because H is symmetric, you need only to fill in the non-zero of the lower-triangular terms. The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). However, you may assume only that H supports size(H) and setindex!. Finally, the matrix is treated as dense, so the performance will be poor on functions with high-dimensional input.","category":"page"},{"location":"manual/nlp/#User-defined-functions-with-vector-inputs","page":"Nonlinear Modeling (Legacy)","title":"User-defined functions with vector inputs","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"User-defined functions which take vectors as input arguments (for example, f(x::Vector)) are not supported. Instead, use Julia's splatting syntax to create a function with scalar arguments. For example, instead of","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x::Vector) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"define:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"f(x...) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This function f can be used in a JuMP model as follows:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"using JuMP #hide\nmodel = Model()\n@variable(model, x[1:5] >= 0)\nf(x...) = sum(x[i]^i for i in 1:length(x))\nregister(model, :f, 5, f; autodiff = true)\n@NLobjective(model, Min, f(x...))","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"tip: Tip\nMake sure to read the syntax restrictions of Splatting.","category":"page"},{"location":"manual/nlp/#Factors-affecting-solution-time","page":"Nonlinear Modeling (Legacy)","title":"Factors affecting solution time","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The execution time when solving a nonlinear programming problem can be divided into two parts, the time spent in the optimization algorithm (the solver) and the time spent evaluating the nonlinear functions and corresponding derivatives. Ipopt explicitly displays these two timings in its output, for example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Total CPU secs in IPOPT (w/o function evaluations)   =      7.412\nTotal CPU secs in NLP function evaluations           =      2.083","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For Ipopt in particular, one can improve the performance by installing advanced sparse linear algebra packages, see Installation Guide. For other solvers, see their respective documentation for performance tips.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The function evaluation time, on the other hand, is the responsibility of the modeling language. JuMP computes derivatives by using reverse-mode automatic differentiation with graph coloring methods for exploiting sparsity of the Hessian matrix. As a conservative bound, JuMP's performance here currently may be expected to be within a factor of 5 of AMPL's. Our paper in SIAM Review has more details.","category":"page"},{"location":"manual/nlp/#Querying-derivatives-from-a-JuMP-model","page":"Nonlinear Modeling (Legacy)","title":"Querying derivatives from a JuMP model","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For some advanced use cases, one may want to directly query the derivatives of a JuMP model instead of handing the problem off to a solver. Internally, JuMP implements the MOI.AbstractNLPEvaluator interface. To obtain an NLP evaluator object from a JuMP model, use NLPEvaluator. index returns the MOI.VariableIndex corresponding to a JuMP variable. MOI.VariableIndex itself is a type-safe wrapper for Int64 (stored in the .value field.)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> raw_index(v::MOI.VariableIndex) = v.value\nraw_index (generic function with 1 method)\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, y)\ny\n\njulia> @NLobjective(model, Min, sin(x) + sin(y))\n\njulia> values = zeros(2)\n2-element Vector{Float64}:\n 0.0\n 0.0\n\njulia> x_index = raw_index(JuMP.index(x))\n1\n\njulia> y_index = raw_index(JuMP.index(y))\n2\n\njulia> values[x_index] = 2.0\n2.0\n\njulia> values[y_index] = 3.0\n3.0\n\njulia> d = NLPEvaluator(model)\nNonlinear.Evaluator with available features:\n  * :Grad\n  * :Jac\n  * :JacVec\n  * :Hess\n  * :HessVec\n  * :ExprGraph\n\njulia> MOI.initialize(d, [:Grad])\n\njulia> MOI.eval_objective(d, values)\n1.0504174348855488\n\njulia> sin(2.0) + sin(3.0)\n1.0504174348855488\n\njulia> ∇f = zeros(2)\n2-element Vector{Float64}:\n 0.0\n 0.0\n\njulia> MOI.eval_objective_gradient(d, ∇f, values)\n\njulia> ∇f[x_index], ∇f[y_index]\n(-0.4161468365471424, -0.9899924966004454)\n\njulia> cos(2.0), cos(3.0)\n(-0.4161468365471424, -0.9899924966004454)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Only nonlinear constraints (those added with @NLconstraint), and nonlinear objectives (added with @NLobjective) exist in the scope of the NLPEvaluator.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The NLPEvaluator does not evaluate derivatives of linear or quadratic constraints or objectives.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The index method applied to a nonlinear constraint reference object returns its index as a MOI.Nonlinear.ConstraintIndex. For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLconstraint(model, cons1, sin(x) <= 1);\n\njulia> @NLconstraint(model, cons2, x + 5 == 10);\n\njulia> typeof(cons1)\nNonlinearConstraintRef{ScalarShape} (alias for ConstraintRef{GenericModel{Float64}, MathOptInterface.Nonlinear.ConstraintIndex, ScalarShape})\n\njulia> index(cons1)\nMathOptInterface.Nonlinear.ConstraintIndex(1)\n\njulia> index(cons2)\nMathOptInterface.Nonlinear.ConstraintIndex(2)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"# TODO: Provide a link for how to access the linear and quadratic parts of the\n# model.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Note that for one-sided nonlinear constraints, JuMP subtracts any values on the right-hand side when computing expressions. In other words, one-sided nonlinear constraints are always transformed to have a right-hand side of zero.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This method of querying derivatives directly from a JuMP model is convenient for interacting with the model in a structured way, for example, for accessing derivatives of specific variables. For example, in statistical maximum likelihood estimation problems, one is often interested in the Hessian matrix at the optimal solution, which can be queried using the NLPEvaluator.","category":"page"},{"location":"manual/nlp/#Raw-expression-input","page":"Nonlinear Modeling (Legacy)","title":"Raw expression input","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"warning: Warning\nThis section requires advanced knowledge of Julia's Expr. You should read the Expressions and evaluation section of the Julia documentation first.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"In addition to the @NLexpression, @NLobjective and @NLconstraint macros, it is also possible to provide Julia Expr objects directly by using add_nonlinear_expression, set_nonlinear_objective and add_nonlinear_constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This input form may be useful if the expressions are generated programmatically, or if you experience compilation issues with the macro input (see Known performance issues for more information).","category":"page"},{"location":"manual/nlp/#Add-a-nonlinear-expression","page":"Nonlinear Modeling (Legacy)","title":"Add a nonlinear expression","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use add_nonlinear_expression to add a nonlinear expression to the model.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> expr = :($(x) + sin($(x)^2))\n:(x + sin(x ^ 2))\n\njulia> expr_ref = add_nonlinear_expression(model, expr)\nsubexpression[1]: x + sin(x ^ 2.0)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr_ref = @NLexpression(model, x + sin(x^2))\nsubexpression[1]: x + sin(x ^ 2.0)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"note: Note\nYou must interpolate the variables directly into the expression expr.","category":"page"},{"location":"manual/nlp/#Set-the-objective-function","page":"Nonlinear Modeling (Legacy)","title":"Set the objective function","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use set_nonlinear_objective to set a nonlinear objective.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = :($(x) + $(x)^2)\n:(x + x ^ 2)\n\njulia> set_nonlinear_objective(model, MIN_SENSE, expr)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLobjective(model, Min, x + x^2)","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"note: Note\nYou must use MIN_SENSE or MAX_SENSE instead of Min and Max.","category":"page"},{"location":"manual/nlp/#Add-a-constraint","page":"Nonlinear Modeling (Legacy)","title":"Add a constraint","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Use add_nonlinear_constraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = :($(x) + $(x)^2)\n:(x + x ^ 2)\n\njulia> add_nonlinear_constraint(model, :($(expr) <= 1))\n(x + x ^ 2.0) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"This is equivalent to","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @NLconstraint(model, Min, x + x^2 <= 1)\n(x + x ^ 2.0) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#More-complicated-examples","page":"Nonlinear Modeling (Legacy)","title":"More complicated examples","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"Raw expression input is most useful when the expressions are generated programmatically, often in conjunction with user-defined functions.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As an example, we construct a model with the nonlinear constraints f(x) <= 1, where f(x) = x^2 and f(x) = sin(x)^2:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> function main(functions::Vector{Function})\n           model = Model()\n           @variable(model, x)\n           for (i, f) in enumerate(functions)\n               f_sym = Symbol(\"f_$(i)\")\n               register(model, f_sym, 1, f; autodiff = true)\n               add_nonlinear_constraint(model, :($(f_sym)($(x)) <= 1))\n           end\n           print(model)\n           return\n       end\nmain (generic function with 1 method)\n\njulia> main([x -> x^2, x -> sin(x)^2])\nFeasibility\nSubject to\n f_1(x) - 1.0 ≤ 0\n f_2(x) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As another example, we construct a model with the constraint x^2 + sin(x)^2 <= 1:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> function main(functions::Vector{Function})\n           model = Model()\n           @variable(model, x)\n           expr = Expr(:call, :+)\n           for (i, f) in enumerate(functions)\n               f_sym = Symbol(\"f_$(i)\")\n               register(model, f_sym, 1, f; autodiff = true)\n               push!(expr.args, :($(f_sym)($(x))))\n           end\n           add_nonlinear_constraint(model, :($(expr) <= 1))\n           print(model)\n           return\n       end\nmain (generic function with 1 method)\n\njulia> main([x -> x^2, x -> sin(x)^2])\nFeasibility\nSubject to\n (f_1(x) + f_2(x)) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#Registered-functions-with-a-variable-number-of-arguments","page":"Nonlinear Modeling (Legacy)","title":"Registered functions with a variable number of arguments","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"User defined functions require a fixed number of input arguments. However, sometimes you will want to use a registered function like:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> f(x...) = sum(exp(x[i]^2) for i in 1:length(x));","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"with different numbers of arguments.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The solution is to register the same function f for each unique number of input arguments, making sure to use a unique name each time. For example:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"julia> A = [[1], [1, 2], [2, 3, 4], [1, 3, 4, 5]];\n\njulia> model = Model();\n\njulia> @variable(model, x[1:5]);\n\njulia> funcs = Set{Symbol}();\n\njulia> for a in A\n           key = Symbol(\"f$(length(a))\")\n           if !(key in funcs)\n               push!(funcs, key)\n               register(model, key, length(a), f; autodiff = true)\n           end\n           add_nonlinear_constraint(model, :($key($(x[a]...)) <= 1))\n       end\n\njulia> print(model)\nFeasibility\nSubject to\n f1(x[1]) - 1.0 ≤ 0\n f2(x[1], x[2]) - 1.0 ≤ 0\n f3(x[2], x[3], x[4]) - 1.0 ≤ 0\n f4(x[1], x[3], x[4], x[5]) - 1.0 ≤ 0","category":"page"},{"location":"manual/nlp/#Known-performance-issues","page":"Nonlinear Modeling (Legacy)","title":"Known performance issues","text":"","category":"section"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The macro-based input to JuMP's nonlinear interface can cause a performance issue if you:","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"write a macro with a large number (hundreds) of terms\ncall that macro from within a function instead of from the top-level in global scope.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The first issue does not depend on the number of resulting terms in the mathematical expression, but rather the number of terms in the Julia Expr representation of that expression. For example, the expression sum(x[i] for i in 1:1_000_000) contains one million mathematical terms, but the Expr representation is just a single sum.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"The most common cause, other than a lot of tedious typing, is if you write a program that automatically writes a JuMP model as a text file, which you later execute. One example is MINLPlib.jl which automatically transpiled models in the GAMS scalar format into JuMP examples.","category":"page"},{"location":"manual/nlp/","page":"Nonlinear Modeling (Legacy)","title":"Nonlinear Modeling (Legacy)","text":"As a rule of thumb, if you are writing programs to automatically generate expressions for the JuMP macros, you should target the Raw expression input instead. For more information, read MathOptInterface Issue#1997.","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"EditURL = \"https://github.com/jump-dev/Dualization.jl/blob/v0.5.8/README.md\"","category":"page"},{"location":"packages/Dualization/#Dualization.jl","page":"jump-dev/Dualization.jl","title":"Dualization.jl","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"(Image: Build Status) (Image: codecov) (Image: DOI)","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl is an extension package for MathOptInterface.jl that formulates the dual of conic optimization problems.","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl has two main features:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"The Dualization.dualize function that computes the dual formulation of either a MathOptInterface.jl or a JuMP model.\nThe Dualization.dual_optimizer function that creates a MathOptInterface-compatible optimizer that solves the dual of the problem instead of the primal.","category":"page"},{"location":"packages/Dualization/#License","page":"jump-dev/Dualization.jl","title":"License","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Dualization.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Dualization/#Installation","page":"jump-dev/Dualization.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"Install Dualization using Pkg.add:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"import Pkg\nPkg.add(\"Dualization\")","category":"page"},{"location":"packages/Dualization/#Use-with-JuMP","page":"jump-dev/Dualization.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"To compute the dual formulation of a JuMP model, use dualize:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"using JuMP, Dualization\nmodel = Model()\n# ... build model ...\ndual_model = dualize(model)","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"To solve the dual formulation of a JuMP model, create a dual_optimizer:","category":"page"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"using JuMP, Dualization, SCS\nmodel = Model(dual_optimizer(SCS.Optimizer))\n# ... build model ...\noptimize!(model)  # Solves the dual instead of the primal","category":"page"},{"location":"packages/Dualization/#Documentation","page":"jump-dev/Dualization.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Dualization/","page":"jump-dev/Dualization.jl","title":"jump-dev/Dualization.jl","text":"The documentation for Dualization.jl includes a detailed description of the dual reformulation, along with examples and an API reference.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/constraints.md\"","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/constraints/#Constraints","page":"Constraints","title":"Constraints","text":"","category":"section"},{"location":"moi/manual/constraints/#Add-a-constraint","page":"Constraints","title":"Add a constraint","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use add_constraint to add a single constraint.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> c = MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonnegatives(2))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"add_constraint returns a ConstraintIndex type, which is used to refer to the added constraint in other calls.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Check if a ConstraintIndex is valid using is_valid.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.is_valid(model, c)\ntrue","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use add_constraints to add a number of constraints of the same type.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> c = MOI.add_constraints(\n           model,\n           [x[1], x[2]],\n           [MOI.GreaterThan(0.0), MOI.GreaterThan(1.0)]\n       )\n2-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(2)","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"This time, a vector of ConstraintIndex are returned.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use supports_constraint to check if the model supports adding a constraint type.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.supports_constraint(\n           model,\n           MOI.VariableIndex,\n           MOI.GreaterThan{Float64},\n       )\ntrue","category":"page"},{"location":"moi/manual/constraints/#Delete-a-constraint","page":"Constraints","title":"Delete a constraint","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Use delete to delete a constraint.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.delete(model, c)\n\njulia> MOI.is_valid(model, c)\nfalse","category":"page"},{"location":"moi/manual/constraints/#Constraint-attributes","page":"Constraints","title":"Constraint attributes","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"The following attributes are available for constraints:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"ConstraintName\nConstraintPrimalStart\nConstraintDualStart\nConstraintPrimal\nConstraintDual\nConstraintBasisStatus\nConstraintFunction\nCanonicalConstraintFunction\nConstraintSet","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Get and set these attributes using get and set.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"julia> MOI.set(model, MOI.ConstraintName(), c, \"con_c\")\n\njulia> MOI.get(model, MOI.ConstraintName(), c)\n\"con_c\"","category":"page"},{"location":"moi/manual/constraints/#Constraints-by-function-set-pairs","page":"Constraints","title":"Constraints by function-set pairs","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Below is a list of common constraint types and how they are represented as function-set pairs in MOI. In the notation below, x is a vector of decision variables, x_i is a scalar decision variable, alpha beta are scalar constants, a b are constant vectors, A is a constant matrix and mathbbR_+ (resp. mathbbR_-) is the set of non-negative (resp. non-positive) real numbers.","category":"page"},{"location":"moi/manual/constraints/#Linear-constraints","page":"Constraints","title":"Linear constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\na^Tx le beta ScalarAffineFunction LessThan\na^Tx ge alpha ScalarAffineFunction GreaterThan\na^Tx = beta ScalarAffineFunction EqualTo\nalpha le a^Tx le beta ScalarAffineFunction Interval\nx_i le beta VariableIndex LessThan\nx_i ge alpha VariableIndex GreaterThan\nx_i = beta VariableIndex EqualTo\nalpha le x_i le beta VariableIndex Interval\nAx + b in mathbbR_+^n VectorAffineFunction Nonnegatives\nAx + b in mathbbR_-^n VectorAffineFunction Nonpositives\nAx + b = 0 VectorAffineFunction Zeros","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"By convention, solvers are not expected to support nonzero constant terms in the ScalarAffineFunctions the first four rows of the preceding table because they are redundant with the parameters of the sets. For example, encode 2x + 1 le 2 as 2x le 1.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Constraints with VariableIndex in LessThan, GreaterThan, EqualTo, or Interval sets have a natural interpretation as variable bounds. As such, it is typically not natural to impose multiple lower- or upper-bounds on the same variable, and the solver interfaces will throw respectively LowerBoundAlreadySet or UpperBoundAlreadySet.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Moreover, adding two VariableIndex constraints on the same variable with the same set is impossible because they share the same index as it is the index of the variable, see ConstraintIndex.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"It is natural, however, to impose upper- and lower-bounds separately as two different constraints on a single variable. The difference between imposing bounds by using a single Interval constraint and by using separate LessThan and GreaterThan constraints is that the latter will allow the solver to return separate dual multipliers for the two bounds, while the former will allow the solver to return only a single dual for the interval constraint.","category":"page"},{"location":"moi/manual/constraints/#Conic-constraints","page":"Constraints","title":"Conic constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nlVert Ax + brVert_2 le c^Tx + d VectorAffineFunction SecondOrderCone\ny ge lVert x rVert_2 VectorOfVariables SecondOrderCone\n2yz ge lVert x rVert_2^2 yz ge 0 VectorOfVariables RotatedSecondOrderCone\n(a_1^Tx + b_1a_2^Tx + b_2a_3^Tx + b_3) in mathcalE VectorAffineFunction ExponentialCone\nA(x) in mathcalS_+ VectorAffineFunction PositiveSemidefiniteConeTriangle\nB(x) in mathcalS_+ VectorAffineFunction PositiveSemidefiniteConeSquare\nx in mathcalS_+ VectorOfVariables PositiveSemidefiniteConeTriangle\nx in mathcalS_+ VectorOfVariables PositiveSemidefiniteConeSquare","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"where mathcalE is the exponential cone (see ExponentialCone), mathcalS_+ is the set of positive semidefinite symmetric matrices, A is an affine map that outputs symmetric matrices and B is an affine map that outputs square matrices.","category":"page"},{"location":"moi/manual/constraints/#Quadratic-constraints","page":"Constraints","title":"Quadratic constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nfrac12x^TQx + a^Tx + b ge 0 ScalarQuadraticFunction GreaterThan\nfrac12x^TQx + a^Tx + b le 0 ScalarQuadraticFunction LessThan\nfrac12x^TQx + a^Tx + b = 0 ScalarQuadraticFunction EqualTo\nBilinear matrix inequality VectorQuadraticFunction PositiveSemidefiniteCone...","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"note: Note\nFor more details on the internal format of the quadratic functions see ScalarQuadraticFunction or VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/constraints/#Discrete-and-logical-constraints","page":"Constraints","title":"Discrete and logical constraints","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Mathematical Constraint MOI Function MOI Set\nx_i in mathbbZ VariableIndex Integer\nx_i in 01 VariableIndex ZeroOne\nx_i in 0 cup lu VariableIndex Semicontinuous\nx_i in 0 cup ll+1ldotsu-1u VariableIndex Semiinteger\nAt most one component of x can be nonzero VectorOfVariables SOS1\nAt most two components of x can be nonzero, and if so they must be adjacent components VectorOfVariables SOS2\ny = 1 implies a^T x in S VectorAffineFunction Indicator","category":"page"},{"location":"moi/manual/constraints/#JuMP-mapping","page":"Constraints","title":"JuMP mapping","text":"","category":"section"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"The following bullet points show examples of how JuMP constraints are translated into MOI function-set pairs:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@constraint(m, 2x + y <= 10) becomes ScalarAffineFunction-in-LessThan\n@constraint(m, 2x + y >= 10) becomes ScalarAffineFunction-in-GreaterThan\n@constraint(m, 2x + y == 10) becomes ScalarAffineFunction-in-EqualTo\n@constraint(m, 0 <= 2x + y <= 10) becomes ScalarAffineFunction-in-Interval\n@constraint(m, 2x + y in ArbitrarySet()) becomes ScalarAffineFunction-in-ArbitrarySet.","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"Variable bounds are handled in a similar fashion:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@variable(m, x <= 1) becomes VariableIndex-in-LessThan\n@variable(m, x >= 1) becomes VariableIndex-in-GreaterThan","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"One notable difference is that a variable with an upper and lower bound is translated into two constraints, rather than an interval, that is:","category":"page"},{"location":"moi/manual/constraints/","page":"Constraints","title":"Constraints","text":"@variable(m, 0 <= x <= 1) becomes VariableIndex-in-LessThan and  VariableIndex-in-GreaterThan.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"EditURL = \"sudoku.jl\"","category":"page"},{"location":"tutorials/linear/sudoku/#Sudoku","page":"Sudoku","title":"Sudoku","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"This tutorial was originally contributed by Iain Dunning.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Sudoku is a popular number puzzle. The goal is to place the digits 1 to 9 on a nine-by-nine grid, with some of the digits already filled in. Your solution must satisfy the following rules:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The numbers 1 to 9 must appear in each 3x3 square\nThe numbers 1 to 9 must appear in each row\nThe numbers 1 to 9 must appear in each column","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Here is a partially solved Sudoku problem:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"(Image: Partially solved Sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Solving a Sudoku isn't an optimization problem with an objective; its actually a feasibility problem: we wish to find a feasible solution that satisfies these rules. You can think of it as an optimization problem with an objective of 0.","category":"page"},{"location":"tutorials/linear/sudoku/#Mixed-integer-linear-programming-formulation","page":"Sudoku","title":"Mixed-integer linear programming formulation","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We can model this problem using 0-1 integer programming: a problem where all the decision variables are binary. We'll use JuMP to create the model, and then we can solve it with any integer programming solver.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"using JuMP\nusing HiGHS","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We will define a binary variable (a variable that is either 0 or 1) for each possible number in each possible cell. The meaning of each variable is as follows: x[i,j,k] = 1  if and only if cell (i,j) has number k, where i is the row and j is the column.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create a model","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sudoku = Model(HiGHS.Optimizer)\nset_silent(sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create our variables","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@variable(sudoku, x[i = 1:9, j = 1:9, k = 1:9], Bin);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Now we can begin to add our constraints. We'll actually start with something obvious to us as humans, but what we need to enforce: that there can be only one number per cell.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:9  # For each row\n    for j in 1:9  # and each column\n        # Sum across all the possible digits. One and only one of the digits\n        # can be in this cell, so the sum must be equal to one.\n        @constraint(sudoku, sum(x[i, j, k] for k in 1:9) == 1)\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Next we'll add the constraints for the rows and the columns. These constraints are all very similar, so much so that we can actually add them at the same time.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for ind in 1:9  # Each row, OR each column\n    for k in 1:9  # Each digit\n        # Sum across columns (j) - row constraint\n        @constraint(sudoku, sum(x[ind, j, k] for j in 1:9) == 1)\n        # Sum across rows (i) - column constraint\n        @constraint(sudoku, sum(x[i, ind, k] for i in 1:9) == 1)\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Finally, we have the to enforce the constraint that each digit appears once in each of the nine 3x3 sub-grids. Our strategy will be to index over the top-left corners of each 3x3 square with for loops, then sum over the squares.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:3:7\n    for j in 1:3:7\n        for k in 1:9\n            # i is the top left row, j is the top left column.\n            # We'll sum from i to i+2, for example, i=4, r=4, 5, 6.\n            @constraint(\n                sudoku,\n                sum(x[r, c, k] for r in i:(i+2), c in j:(j+2)) == 1\n            )\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The final step is to add the initial solution as a set of constraints. We'll solve the problem that is in the picture at the start of the tutorial. We'll put a 0 if there is no digit in that location.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"The given digits","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"init_sol = [\n    5 3 0 0 7 0 0 0 0\n    6 0 0 1 9 5 0 0 0\n    0 9 8 0 0 0 0 6 0\n    8 0 0 0 6 0 0 0 3\n    4 0 0 8 0 3 0 0 1\n    7 0 0 0 2 0 0 0 6\n    0 6 0 0 0 0 2 8 0\n    0 0 0 4 1 9 0 0 5\n    0 0 0 0 8 0 0 7 9\n]\nfor i in 1:9\n    for j in 1:9\n        # If the space isn't empty\n        if init_sol[i, j] != 0\n            # Then the corresponding variable for that digit and location must\n            # be 1.\n            fix(x[i, j, init_sol[i, j]], 1; force = true)\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"solve problem","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"optimize!(sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Extract the values of x","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"x_val = value.(x);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Create a matrix to store the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol = zeros(Int, 9, 9)  # 9x9 matrix of integers\nfor i in 1:9\n    for j in 1:9\n        for k in 1:9\n            # Integer programs are solved as a series of linear programs so the\n            # values might not be precisely 0 and 1. We can round them to\n            # the nearest integer to make it easier.\n            if round(Int, x_val[i, j, k]) == 1\n                sol[i, j] = k\n            end\n        end\n    end\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Display the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Which is the correct solution:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"(Image: Solved Sudoku)","category":"page"},{"location":"tutorials/linear/sudoku/#Constraint-programming-formulation","page":"Sudoku","title":"Constraint programming formulation","text":"","category":"section"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"We can also model this problem using constraint programming and the all-different constraint, which says that no two elements of a vector can take the same value.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Because of the reformulation system in MathOptInterface, we can still solve this problem using HiGHS.","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n# HiGHS v1.2 has a bug in presolve which causes the problem to be classified as\n# infeasible.\nset_attribute(model, \"presolve\", \"off\")","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Instead of the binary variables, we directly define a 9x9 grid of integer values between 1 and 9:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@variable(model, 1 <= x[1:9, 1:9] <= 9, Int);\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Then, we enforce that the values in each row must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@constraint(model, [i = 1:9], x[i, :] in MOI.AllDifferent(9));\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"That the values in each column must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"@constraint(model, [j = 1:9], x[:, j] in MOI.AllDifferent(9));\nnothing #hide","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"And that the values in each 3x3 sub-grid must be all-different:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in (0, 3, 6), j in (0, 3, 6)\n    @constraint(model, vec(x[i.+(1:3), j.+(1:3)]) in MOI.AllDifferent(9))\nend","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Finally, as before we set the initial solution and optimize:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"for i in 1:9, j in 1:9\n    if init_sol[i, j] != 0\n        fix(x[i, j], init_sol[i, j]; force = true)\n    end\nend\n\noptimize!(model)","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Display the solution","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"csp_sol = round.(Int, value.(x))","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"Which is the same as we found before:","category":"page"},{"location":"tutorials/linear/sudoku/","page":"Sudoku","title":"Sudoku","text":"sol == csp_sol","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"EditURL = \"performance_tips.jl\"","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Performance-tips","page":"Performance tips","title":"Performance tips","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"By now you should have read the other \"getting started\" tutorials. You're almost ready to write your own models, but before you do so there are some important things to be aware of.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Read-the-Julia-performance-tips","page":"Performance tips","title":"Read the Julia performance tips","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"The first thing to do is read the Performance tips section of the Julia manual. The most important rule is to avoid global variables. This is particularly important if you're learning JuMP after using a language like MATLAB.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#The-\"time-to-first-solve\"-issue","page":"Performance tips","title":"The \"time-to-first-solve\" issue","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Similar to the infamous time-to-first-plot plotting problem, JuMP suffers from time-to-first-solve latency. This latency occurs because the first time you call JuMP code in each session, Julia needs to compile a lot of code specific to your problem. This issue is actively being worked on, but there are a few things you can do to improve things.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-1:-don't-call-JuMP-from-the-command-line","page":"Performance tips","title":"Suggestion 1: don't call JuMP from the command line","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"In other languages, you might be used to a workflow like:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ julia my_script.jl","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"This doesn't work for JuMP, because we have to pay the compilation latency every time you run the script. Instead, use one of the suggested workflows from the Julia documentation.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-2:-disable-bridges-if-none-are-being-used","page":"Performance tips","title":"Suggestion 2: disable bridges if none are being used","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"At present, the majority of the latency problems are caused by JuMP's bridging mechanism. If you only use constraints that are natively supported by the solver, you can disable bridges by passing add_bridges = false to Model.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model(HiGHS.Optimizer; add_bridges = false)","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Suggestion-3:-use-PackageCompiler","page":"Performance tips","title":"Suggestion 3: use PackageCompiler","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"As an example of compilation latency, consider the following linear program with two variables and two constraints:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using JuMP, HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\noptimize!(model)\nopen(\"model.log\", \"w\") do io\n    print(io, solution_summary(model; verbose = true))\n    return\nend","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Saving the problem in model.jl and calling from the command line results in:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ time julia model.jl\n15.78s user 0.48s system 100% cpu 16.173 total","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Clearly, 16 seconds is a large overhead to pay for solving this trivial model. However, the compilation latency is independent on the problem size, and so 16 seconds of additional overhead may be tolerable for larger models that take minutes or hours to solve.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"In cases where the compilation latency is intolerable, JuMP is compatible with the PackageCompiler.jl package, which makes it easy to generate a custom sysimage (a binary extension to Julia that caches compiled code) that dramatically reduces the compilation latency. A custom image for our problem can be created as follows:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"using PackageCompiler, Libdl\nPackageCompiler.create_sysimage(\n    [\"JuMP\", \"HiGHS\"],\n    sysimage_path = \"customimage.\" * Libdl.dlext,\n    precompile_execution_file = \"model.jl\",\n)","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"When Julia is run with the custom image, the run time is now 0.7 seconds instead of 16:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"$ time julia --sysimage customimage model.jl\n0.68s user 0.22s system 153% cpu 0.587 total","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Other performance tweaks, such as disabling bridges or using direct mode can reduce this time further.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"note: Note\ncreate_sysimage only needs to be run once, and the same sysimage can be used–to a slight detriment of performance–even if we modify model.jl or run a different file.","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Use-macros-to-build-expressions","page":"Performance tips","title":"Use macros to build expressions","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Use JuMP's macros (or add_to_expression!) to build expressions. Avoid constructing expressions outside the macros.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Constructing an expression outside the macro results in intermediate copies of the expression. For example,","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"x[1] + x[2] + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"is equivalent to","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"a = x[1]\nb = a + x[2]\nc = b + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Since we only care about c, the a and b expressions are not needed and constructing them slows the program down.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"JuMP's macros rewrite the expressions to operate in-place and avoid these extra copies. Because they allocate less memory, they are faster, particularly for large expressions.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Here's an example.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model()\n@variable(model, x[1:3])","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Here's what happens if we construct the expression outside the macro:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"@allocated x[1] + x[2] + x[3]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"info: Info\nThe @allocated measures how many bytes were allocated during the evaluation of an expression. Fewer is better.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"If we use the @expression macro, we get many fewer allocations:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"@allocated @expression(model, x[1] + x[2] + x[3])","category":"page"},{"location":"tutorials/getting_started/performance_tips/#Disable-string-names","page":"Performance tips","title":"Disable string names","text":"","category":"section"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"By default, JuMP creates String names for variables and constraints and passes these to the solver. The benefit of passing names is that it improves the readability of log messages from the solver (for example, \"variable x has invalid bounds\" instead of \"variable v1203 has invalid bounds\"), but for larger models the overhead of passing names can be non-trivial.","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Disable the creation of String names by setting set_string_name = false in the @variable and @constraint macros, or by calling set_string_names_on_creation to disable all names for a particular model:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"model = Model();\nset_string_names_on_creation(model, false)\n@variable(model, x)\n@constraint(model, c, 2x <= 1)","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"Note that this doesn't change how symbolic names and bindings are stored:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"x\nmodel[:x]\nx === model[:x]","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"But you can no longer look up the variable by the string name:","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"variable_by_name(model, \"x\") === nothing","category":"page"},{"location":"tutorials/getting_started/performance_tips/","page":"Performance tips","title":"Performance tips","text":"info: Info\nFor more information on the difference between string names, symbolic names, and bindings, see String names, symbolic names, and bindings.","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"EditURL = \"https://github.com/jump-dev/ParametricOptInterface.jl/blob/v0.5.1/README.md\"","category":"page"},{"location":"packages/ParametricOptInterface/#ParametricOptInterface.jl","page":"jump-dev/ParametricOptInterface.jl","title":"ParametricOptInterface.jl","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"(Image: stable docs) (Image: development docs) (Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface.jl is a package that adds parameters to models in JuMP and MathOptInterface.","category":"page"},{"location":"packages/ParametricOptInterface/#License","page":"jump-dev/ParametricOptInterface.jl","title":"License","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface.jl is licensed under the MIT License.","category":"page"},{"location":"packages/ParametricOptInterface/#Installation","page":"jump-dev/ParametricOptInterface.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"Install ParametricOptInterface using Pkg.add:","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"import Pkg\nPkg.add(\"ParametricOptInterface\")","category":"page"},{"location":"packages/ParametricOptInterface/#Documentation","page":"jump-dev/ParametricOptInterface.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"The documentation for ParametricOptInterface.jl includes a detailed description of the theory behind the package, along with examples, tutorials, and an API reference.","category":"page"},{"location":"packages/ParametricOptInterface/#Use-with-JuMP","page":"jump-dev/ParametricOptInterface.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"Use ParametricOptInterface with JuMP by following this brief example:","category":"page"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"using JuMP, HiGHS\nimport ParametricOptInterface as POI\nmodel = Model(() -> POI.Optimizer(HiGHS.Optimizer()))\n@variable(model, x)\n@variable(model, p in POI.Parameter(1.0))\n@constraint(model, cons, x + p >= 3)\n@objective(model, Min, 2x)\noptimize!(model)\nMOI.set(model, POI.ParameterValue(), p, 2.0)\noptimize!(model)","category":"page"},{"location":"packages/ParametricOptInterface/#GSOC2020","page":"jump-dev/ParametricOptInterface.jl","title":"GSOC2020","text":"","category":"section"},{"location":"packages/ParametricOptInterface/","page":"jump-dev/ParametricOptInterface.jl","title":"jump-dev/ParametricOptInterface.jl","text":"ParametricOptInterface began as a NumFOCUS sponsored Google Summer of Code (2020) project.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"EditURL = \"quantum_discrimination.jl\"","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Quantum-state-discrimination","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial solves the problem of quantum state discrimination.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The purpose of this tutorial to demonstrate how to solve problems involving complex-valued decision variables and the HermitianPSDCone. See Complex number support for more details.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Required-packages","page":"Quantum state discrimination","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This tutorial makes use of the following packages:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"using JuMP\nimport LinearAlgebra\nimport SCS","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Formulation","page":"Quantum state discrimination","title":"Formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"A d-dimensional quantum state, rho, can be defined by a complex-valued Hermitian matrix with a trace of 1. Assume we have N d-dimensional quantum states, rho_i_i=1^N, each of which is equally likely.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The goal of the quantum state discrimination problem is to choose a positive operator-valued measure (POVM),  E_i _i=1^N, such that if we observe E_i then the most probable state that we are in is rho_i.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Each POVM element, E_i, is a complex-valued Hermitian matrix, and there is a requirement that sumlimits_i=1^N E_i = mathbfI.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To choose a POVM, we want to maximize the probability that we guess the quantum state correctly. This can be formulated as the following optimization problem:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"beginaligned\nmaxlimits_E   frac1N sumlimits_i=1^N operatornametr(rho_i  E_i) \ntextst       sumlimits_i=1^N E_i = mathbfI \n                      E_i succeq 0  forall i = 1ldotsN\nendaligned","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Data","page":"Quantum state discrimination","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To setup our problem, we need N d-dimensional quantum states. To keep the problem simple, we use N = 2 and d = 2.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"N, d = 2, 2","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"We then generated N random d-dimensional quantum states:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"function random_state(d)\n    x = randn(ComplexF64, (d, d))\n    y = x * x'\n    return LinearAlgebra.Hermitian(y / LinearAlgebra.tr(y))\nend\n\nρ = [random_state(d) for i in 1:N]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#JuMP-formulation","page":"Quantum state discrimination","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"To model the problem in JuMP, we need a solver that supports positive semidefinite matrices:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"model = Model(SCS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Then, we construct our set of E variables:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"E = [@variable(model, [1:d, 1:d] in HermitianPSDCone()) for i in 1:N]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Here we have created a vector of matrices. This is different to other modeling languages such as YALMIP, which allow you to create a multi-dimensional array in which 2-dimensional slices of the array are Hermitian matrices.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"We also need to enforce the constraint that sumlimits_i=1^N E_i = mathbfI:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@constraint(model, sum(E) == LinearAlgebra.I)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"This constraint is a complex-valued equality constraint. In the solver, it will be decomposed onto two types of equality constraints: one to enforce equality of the real components, and one to enforce equality of the imaginary components.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Our objective is to maximize the expected probability of guessing correctly:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@objective(\n    model,\n    Max,\n    sum(real(LinearAlgebra.tr(ρ[i] * E[i])) for i in 1:N) / N,\n)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Now we optimize:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The probability of guessing correctly is:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"objective_value(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"When N = 2, there is a known analytical solution of:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"0.5 + 0.25 * sum(LinearAlgebra.svdvals(ρ[1] - ρ[2]))","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"proving that we found the optimal solution.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Finally, the optimal POVM is:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"solution = [value.(e) for e in E]","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"tip: Tip\nDuality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual of this problem instead of the primal. If performance is an issue, see the Dualization tutorial for more details.","category":"page"},{"location":"tutorials/conic/quantum_discrimination/#Alternative-formulation","page":"Quantum state discrimination","title":"Alternative formulation","text":"","category":"section"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The formulation above includes N Hermitian matrices and a set of linear equality constraints. We can simplify the problem by replacing E_N with E_N = I - sumlimits_i=1^N-1 E_i. This results in:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"model = Model(SCS.Optimizer)\nset_silent(model)\nE = [@variable(model, [1:d, 1:d] in HermitianPSDCone()) for i in 1:N-1]\nE_N = LinearAlgebra.Hermitian(LinearAlgebra.I - sum(E))\n@constraint(model, E_N in HermitianPSDCone())\npush!(E, E_N)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"The objective can also be simplified, by observing that it is equivalent to:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"@objective(model, Max, real(LinearAlgebra.dot(ρ, E)) / N)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"Then we can check that we get the same solution:","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/quantum_discrimination/","page":"Quantum state discrimination","title":"Quantum state discrimination","text":"objective_value(model)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/implementing.md\"","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/implementing/#Implementing-a-solver-interface","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"This guide outlines the basic steps to implement an interface to MathOptInterface for a new solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"danger: Danger\nImplementing an interface to MathOptInterface for a new solver is a lot of work. Before starting, we recommend that you join the Developer chatroom and explain a little bit about the solver you are wrapping. If you have questions that are not answered by this guide, please ask them in the Developer chatroom so we can improve this guide.","category":"page"},{"location":"moi/tutorials/implementing/#A-note-on-the-API","page":"Implementing a solver interface","title":"A note on the API","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The API of MathOptInterface is large and varied. In order to support the diversity of solvers and use-cases, we make heavy use of duck-typing. That is, solvers are not expected to implement the full API, nor is there a well-defined minimal subset of what must be implemented. Instead, you should implement the API as necessary to make the solver function as you require.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The main reason for using duck-typing is that solvers work in different ways and target different use-cases.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Some solvers support incremental problem construction, support modification after a solve, and have native support for things like variable names.\nOther solvers are \"one-shot\" solvers that require all of the problem data to construct and solve the problem in a single function call. They do not support modification or things like variable names.\nOther \"solvers\" are not solvers at all, but things like file readers. These may only support functions like read_from_file, and may not even support the ability to add variables or constraints directly.\nFinally, some \"solvers\" are layers which take a problem as input, transform it according to some rules, and pass the transformed problem to an inner solver.","category":"page"},{"location":"moi/tutorials/implementing/#Preliminaries","page":"Implementing a solver interface","title":"Preliminaries","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before starting on your wrapper, you should do some background research and make the solver accessible via Julia.","category":"page"},{"location":"moi/tutorials/implementing/#Decide-if-MathOptInterface-is-right-for-you","page":"Implementing a solver interface","title":"Decide if MathOptInterface is right for you","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The first step in writing a wrapper is to decide whether implementing an interface is the right thing to do.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface is an abstraction layer for unifying constrained mathematical optimization solvers. If your solver doesn't fit in the category, for example, it implements a derivative-free algorithm for unconstrained objective functions, MathOptInterface may not be the right tool for the job.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf you're not sure whether you should write an interface, ask in the Developer chatroom.","category":"page"},{"location":"moi/tutorials/implementing/#Find-a-similar-solver-already-wrapped","page":"Implementing a solver interface","title":"Find a similar solver already wrapped","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to find (if possible) a similar solver that is already wrapped. Although not strictly necessary, this will be a good place to look for inspiration when implementing your wrapper.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The JuMP documentation has a good list of solvers, along with the problem classes they support.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf you're not sure which solver is most similar, ask in the Developer chatroom.","category":"page"},{"location":"moi/tutorials/implementing/#Create-a-low-level-interface","page":"Implementing a solver interface","title":"Create a low-level interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before writing a MathOptInterface wrapper, you first need to be able to call the solver from Julia.","category":"page"},{"location":"moi/tutorials/implementing/#Wrapping-solvers-written-in-Julia","page":"Implementing a solver interface","title":"Wrapping solvers written in Julia","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver is written in Julia, there's nothing to do here. Go to the next section.","category":"page"},{"location":"moi/tutorials/implementing/#Wrapping-solvers-written-in-C","page":"Implementing a solver interface","title":"Wrapping solvers written in C","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Julia is well suited to wrapping solvers written in C.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nThis is not true for C++. If you have a solver written in C++, first write a C interface, then wrap the C interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Before writing a MathOptInterface wrapper, there are a few extra steps.","category":"page"},{"location":"moi/tutorials/implementing/#Create-a-JLL","page":"Implementing a solver interface","title":"Create a JLL","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If the C code is publicly available under an open source license, create a JLL package via Yggdrasil. The easiest way to do this is to copy an existing solver. Good examples to follow are the COIN-OR solvers.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nBuilding the solver via Yggdrasil is non-trivial. Please ask the Developer chatroom for help.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If the code is commercial or not publicly available, the user will need to manually install the solver. See Gurobi.jl or CPLEX.jl for examples of how to structure this.","category":"page"},{"location":"moi/tutorials/implementing/#implement_clang","page":"Implementing a solver interface","title":"Use Clang.jl to wrap the C API","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to use Clang.jl to automatically wrap the C API. The easiest way to do this is to follow an example. Good examples to follow are Cbc.jl and HiGHS.jl.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Sometimes, you will need to make manual modifications to the resulting files.","category":"page"},{"location":"moi/tutorials/implementing/#Solvers-written-in-other-languages","page":"Implementing a solver interface","title":"Solvers written in other languages","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Ask the Developer chatroom for advice. You may be able to use one of the JuliaInterop packages to call out to the solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, SeDuMi.jl uses MATLAB.jl to call the SeDuMi solver written in MATLAB.","category":"page"},{"location":"moi/tutorials/implementing/#Structuring-the-package","page":"Implementing a solver interface","title":"Structuring the package","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Structure your wrapper as a Julia package. Consult the Julia documentation if you haven't done this before.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI solver interfaces may be in the same package as the solver itself (either the C wrapper if the solver is accessible through C, or the Julia code if the solver is written in Julia, for example), or in a separate package which depends on the solver package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"note: Note\nThe JuMP core contributors request that you do not use \"JuMP\" in the name of your package without prior consent.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Your package should have the following structure:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"/.github\n    /workflows\n        ci.yml\n        format_check.yml\n        TagBot.yml\n/gen\n    gen.jl  # Code to wrap the C API\n/src\n    NewSolver.jl\n    /gen\n        libnewsolver_api.jl\n        libnewsolver_common.jl\n    /MOI_wrapper\n        MOI_wrapper.jl\n        other_files.jl\n/test\n    runtests.jl\n    /MOI_wrapper\n        MOI_wrapper.jl\n.gitignore\n.JuliaFormatter.toml\nREADME.md\nLICENSE.md\nProject.toml","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The /.github folder contains the scripts for GitHub actions. The easiest way to write these is to copy the ones from an existing solver.\nThe /gen and /src/gen folders are only needed if you are wrapping a solver written in C.\nThe /src/MOI_wrapper folder contains the Julia code for the MOI wrapper.\nThe /test folder contains code for testing your package. See Setup tests for more information.\nThe .JuliaFormatter.toml and .github/workflows/format_check.yml enforce code formatting using JuliaFormatter.jl. Check existing solvers or JuMP.jl for details.","category":"page"},{"location":"moi/tutorials/implementing/#Documentation","page":"Implementing a solver interface","title":"Documentation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Your package must include documentation explaining how to use the package. The easiest approach is to include documentation in your README.md. A more involved option is to use Documenter.jl.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Examples of packages with README-based documentation include:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Cbc.jl\nHiGHS.jl\nSCS.jl","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Examples of packages with Documenter-based documentation include:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Alpine.jl\nCOSMO.jl\nJuniper.jl","category":"page"},{"location":"moi/tutorials/implementing/#Setup-tests","page":"Implementing a solver interface","title":"Setup tests","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The best way to implement an interface to MathOptInterface is via test-driven development.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The MOI.Test submodule contains a large test suite to help check that you have implemented things correctly.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Follow the guide How to test a solver to set up the tests for your package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nRun the tests frequently when developing. However, at the start there is going to be a lot of errors. Start by excluding large classes of tests (for example, exclude = [\"test_basic_\", \"test_model_\"], implement any missing methods until the tests pass, then remove an exclusion and repeat.","category":"page"},{"location":"moi/tutorials/implementing/#Initial-code","page":"Implementing a solver interface","title":"Initial code","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By this point, you should have a package setup with tests, formatting, and access to the underlying solver. Now it's time to start writing the wrapper.","category":"page"},{"location":"moi/tutorials/implementing/#The-Optimizer-object","page":"Implementing a solver interface","title":"The Optimizer object","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The first object to create is a subtype of AbstractOptimizer. This type is going to store everything related to the problem.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By convention, these optimizers should not be exported and should be named PackageName.Optimizer.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"import MathOptInterface as MOI\n\nstruct Optimizer <: MOI.AbstractOptimizer\n    # Fields go here\nend","category":"page"},{"location":"moi/tutorials/implementing/#Optimizer-objects-for-C-solvers","page":"Implementing a solver interface","title":"Optimizer objects for C solvers","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nThis section is important if you wrap a solver written in C.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Wrapping a solver written in C will require the use of pointers, and for you to manually free the solver's memory when the Optimizer is garbage collected by Julia.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Never pass a pointer directly to a Julia ccall function.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Instead, store the pointer as a field in your Optimizer, and implement Base.cconvert and Base.unsafe_convert. Then you can pass Optimizer to any ccall function that expects the pointer.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In addition, make sure you implement a finalizer for each model you create.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If newsolver_createProblem() is the low-level function that creates the problem pointer in C, and newsolver_freeProblem(::Ptr{Cvoid}) is the low-level function that frees memory associated with the pointer, your Optimizer() function should look like this:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"struct Optimizer <: MOI.AbstractOptimizer\n    ptr::Ptr{Cvoid}\n\n    function Optimizer()\n        ptr = newsolver_createProblem()\n        model = Optimizer(ptr)\n        finalizer(model) do m\n            newsolver_freeProblem(m)\n            return\n        end\n        return model\n    end\nend\n\nBase.cconvert(::Type{Ptr{Cvoid}}, model::Optimizer) = model\nBase.unsafe_convert(::Type{Ptr{Cvoid}}, model::Optimizer) = model.ptr","category":"page"},{"location":"moi/tutorials/implementing/#Implement-methods-for-Optimizer","page":"Implementing a solver interface","title":"Implement methods for Optimizer","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All Optimizers must implement the following methods:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"empty!\nis_empty","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Other methods, detailed below, are optional or depend on how you implement the interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nFor this and all future methods, read the docstrings to understand what each method does, what it expects as input, and what it produces as output. If it isn't clear, let us know and we will improve the docstrings. It is also very helpful to look at an existing wrapper for a similar solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You should also implement Base.show(::IO, ::Optimizer) to print a nice string when someone prints your model. For example","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function Base.show(io::IO, model::Optimizer)\n    return print(io, \"NewSolver with the pointer $(model.ptr)\")\nend","category":"page"},{"location":"moi/tutorials/implementing/#Implement-attributes","page":"Implementing a solver interface","title":"Implement attributes","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface uses attributes to manage different aspects of the problem.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For each attribute","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"get gets the current value of the attribute\nset sets a new value of the attribute. Not all attributes can be set. For example, the user can't modify the SolverName.\nsupports returns a Bool indicating whether the solver supports the attribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nUse attribute_value_type to check the value expected by a given attribute. You should make sure that your get function correctly infers to this type (or a subtype of it).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Each column in the table indicates whether you need to implement the particular method for each attribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nSolverName Yes No No\nSolverVersion Yes No No\nRawSolver Yes No No\nName Yes Yes Yes\nSilent Yes Yes Yes\nTimeLimitSec Yes Yes Yes\nObjectiveLimit Yes Yes Yes\nRawOptimizerAttribute Yes Yes Yes\nNumberOfThreads Yes Yes Yes\nAbsoluteGapTolerance Yes Yes Yes\nRelativeGapTolerance Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.get(model::Optimizer, ::MOI.Silent)\n    return # true if MOI.Silent is set\nend\n\nfunction MOI.set(model::Optimizer, ::MOI.Silent, v::Bool)\n    if v\n        # Set a parameter to turn off printing\n    else\n        # Restore the default printing\n    end\n    return\nend\n\nMOI.supports(::Optimizer, ::MOI.Silent) = true","category":"page"},{"location":"moi/tutorials/implementing/#Define-supports_constraint","page":"Implementing a solver interface","title":"Define supports_constraint","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The next step is to define which constraints and objective functions you plan to support.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For each function-set constraint pair, define supports_constraint:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.supports_constraint(\n    ::Optimizer,\n    ::Type{MOI.VariableIndex},\n    ::Type{MOI.ZeroOne},\n)\n    return true\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"To make this easier, you may want to use Unions:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.supports_constraint(\n    ::Optimizer,\n    ::Type{MOI.VariableIndex},\n    ::Type{<:Union{MOI.LessThan,MOI.GreaterThan,MOI.EqualTo}},\n)\n    return true\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nOnly support a constraint if your solver has native support for it.","category":"page"},{"location":"moi/tutorials/implementing/#The-big-decision:-incremental-modification?","page":"Implementing a solver interface","title":"The big decision: incremental modification?","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Now you need to decide whether to support incremental modification or not.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Incremental modification means that the user can add variables and constraints one-by-one without needing to rebuild the entire problem, and they can modify the problem data after an optimize! call. Supporting incremental modification means implementing functions like add_variable and add_constraint.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"The alternative is to accept the problem data in a single optimize! or copy_to function call. Because these functions see all of the data at once, it can typically call a more efficient function to load data into the underlying solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Good examples of solvers supporting incremental modification are MILP solvers like GLPK.jl and Gurobi.jl. Examples of non-incremental solvers are AmplNLWriter.jl and SCS.jl","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"It is possible for a solver to implement both approaches, but you should probably start with one for simplicity.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nOnly support incremental modification if your solver has native support for it.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In general, supporting incremental modification is more work, and it usually requires some extra book-keeping. However, it provides a more efficient interface to the solver if the problem is going to be resolved multiple times with small modifications. Moreover, once you've implemented incremental modification, it's usually not much extra work to add a copy_to interface. The converse is not true.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nIf this is your first time writing an interface, start with the one-shot optimize!.","category":"page"},{"location":"moi/tutorials/implementing/#The-non-incremental-interface","page":"Implementing a solver interface","title":"The non-incremental interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"There are two ways to implement the non-incremental interface. The first uses a two-argument version of optimize!, the second implements copy_to followed by the one-argument version of optimize!.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver does not support modification, and requires all data to solve the problem in a single function call, you should implement the \"one-shot\" optimize!.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"optimize!(::ModelLike, ::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver separates data loading and the actual optimization into separate steps, implement the copy_to interface.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"copy_to(::ModelLike, ::ModelLike)\noptimize!(::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/#The-incremental-interface","page":"Implementing a solver interface","title":"The incremental interface","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nWriting this interface is a lot of work. The easiest way is to consult the source code of a similar solver.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"To implement the incremental interface, implement the following functions:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"add_variable\nadd_variables\nadd_constraint\nadd_constraints\nis_valid\ndelete\noptimize!(::ModelLike)","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nSolvers do not have to support AbstractScalarFunction in GreaterThan, LessThan, EqualTo, or Interval with a nonzero constant in the function. Throw ScalarFunctionConstantNotZero if the function constant is not zero.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In addition, you should implement the following model attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfModelAttributesSet Yes No No\nObjectiveFunctionType Yes No No\nObjectiveFunction Yes Yes Yes\nObjectiveSense Yes Yes Yes\nName Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Variable-related attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfVariableAttributesSet Yes No No\nNumberOfVariables Yes No No\nListOfVariableIndices Yes No No","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Constraint-related attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nListOfConstraintAttributesSet Yes No No\nNumberOfConstraints Yes No No\nListOfConstraintTypesPresent Yes No No\nConstraintFunction Yes Yes No\nConstraintSet Yes Yes No","category":"page"},{"location":"moi/tutorials/implementing/#Modifications","page":"Implementing a solver interface","title":"Modifications","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver supports modifying data in-place, implement modify for the following AbstractModifications:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ScalarConstantChange\nScalarCoefficientChange\nVectorConstantChange\nMultirowChange","category":"page"},{"location":"moi/tutorials/implementing/#Variables-constrained-on-creation","page":"Implementing a solver interface","title":"Variables constrained on creation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Some solvers require variables be associated with a set when they are created. This conflicts with the incremental modification approach, since you cannot first add a free variable and then constrain it to the set.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If this is the case, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"add_constrained_variable\nadd_constrained_variables\nsupports_add_constrained_variables","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"By default, MathOptInterface assumes solvers support free variables. If your solver does not support free variables, define:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI.supports_add_constrained_variables(::Optimizer, ::Type{Reals}) = false","category":"page"},{"location":"moi/tutorials/implementing/#Incremental-and-copy_to","page":"Implementing a solver interface","title":"Incremental and copy_to","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you implement the incremental interface, you have the option of also implementing copy_to.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you don't want to implement copy_to, for example, because the solver has no API for building the problem in a single function call, define the following fallback:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MOI.supports_incremental_interface(::Optimizer) = true\n\nfunction MOI.copy_to(dest::Optimizer, src::MOI.ModelLike)\n    return MOI.Utilities.default_copy_to(dest, src)\nend","category":"page"},{"location":"moi/tutorials/implementing/#implement_names","page":"Implementing a solver interface","title":"Names","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Regardless of which interface you implement, you have the option of implementing the Name attribute for variables and constraints:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Attribute get set supports\nVariableName Yes Yes Yes\nConstraintName Yes Yes Yes","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If you implement names, you must also implement the following three methods:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"function MOI.get(model::Optimizer, ::Type{MOI.VariableIndex}, name::String)\n    return # The variable named `name`.\nend\n\nfunction MOI.get(model::Optimizer, ::Type{MOI.ConstraintIndex}, name::String)\n    return # The constraint any type named `name`.\nend\n\nfunction MOI.get(\n    model::Optimizer,\n    ::Type{MOI.ConstraintIndex{F,S}},\n    name::String,\n) where {F,S}\n    return # The constraint of type F-in-S named `name`.\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"These methods have the following rules:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If there is no variable or constraint with the name, return nothing\nIf there is a single variable or constraint with that name, return the variable or constraint\nIf there are multiple variables or constraints with the name, throw an error.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"warning: Warning\nYou should not implement ConstraintName for VariableIndex constraints. If you implement ConstraintName for other constraints, you can add the following two methods to disable ConstraintName for VariableIndex constraints.function MOI.supports(\n    ::Optimizer,\n    ::MOI.ConstraintName,\n    ::Type{<:MOI.ConstraintIndex{MOI.VariableIndex,<:MOI.AbstractScalarSet}},\n)\n    return throw(MOI.VariableIndexConstraintNameError())\nend\nfunction MOI.set(\n    ::Optimizer,\n    ::MOI.ConstraintName,\n    ::MOI.ConstraintIndex{MOI.VariableIndex,<:MOI.AbstractScalarSet},\n    ::String,\n)\n    return throw(MOI.VariableIndexConstraintNameError())\nend","category":"page"},{"location":"moi/tutorials/implementing/#Solutions","page":"Implementing a solver interface","title":"Solutions","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Implement optimize! to solve the model:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"optimize!","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All Optimizers must implement the following attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"DualStatus\nPrimalStatus\nRawStatusString\nResultCount\nTerminationStatus","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"info: Info\nYou only need to implement get for solution attributes. Don't implement set or supports.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"note: Note\nSolver wrappers should document how the low-level statuses map to the MOI statuses. Statuses like NEARLY_FEASIBLE_POINT and INFEASIBLE_POINT, are designed to be used when the solver explicitly indicates that relaxed tolerances are satisfied or the returned point is infeasible, respectively.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You should also implement the following attributes:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ObjectiveValue\nSolveTimeSec\nVariablePrimal","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"tip: Tip\nAttributes like VariablePrimal and ObjectiveValue are indexed by the result count. Use MOI.check_result_index_bounds(model, attr) to throw an error if the attribute is not available.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver returns dual solutions, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ConstraintDual\nDualObjectiveValue","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For integer solvers, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ObjectiveBound\nRelativeGap","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If applicable, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"SimplexIterations\nBarrierIterations\nNodeCount","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver uses the Simplex method, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"ConstraintBasisStatus","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"If your solver accepts primal or dual warm-starts, implement:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"VariablePrimalStart\nConstraintDualStart","category":"page"},{"location":"moi/tutorials/implementing/#Other-tips","page":"Implementing a solver interface","title":"Other tips","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Here are some other points to be aware of when writing your wrapper.","category":"page"},{"location":"moi/tutorials/implementing/#Unsupported-constraints-at-runtime","page":"Implementing a solver interface","title":"Unsupported constraints at runtime","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In some cases, your solver may support a particular type of constraint (for example, quadratic constraints), but only if the data meets some condition (for example, it is convex).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"In this case, declare that you support the constraint, and throw AddConstraintNotAllowed.","category":"page"},{"location":"moi/tutorials/implementing/#Dealing-with-multiple-variable-bounds","page":"Implementing a solver interface","title":"Dealing with multiple variable bounds","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"MathOptInterface uses VariableIndex constraints to represent variable bounds. Defining multiple variable bounds on a single variable is not allowed.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Throw LowerBoundAlreadySet or UpperBoundAlreadySet if the user adds a constraint that results in multiple bounds.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Only throw if the constraints conflict. It is okay to add VariableIndex-in-GreaterThan and then VariableIndex-in-LessThan, but not VariableIndex-in-Interval and then VariableIndex-in-LessThan,","category":"page"},{"location":"moi/tutorials/implementing/#Expect-duplicate-coefficients","page":"Implementing a solver interface","title":"Expect duplicate coefficients","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Solvers must expect that functions such as ScalarAffineFunction and VectorQuadraticFunction may contain duplicate coefficients.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, ScalarAffineFunction([ScalarAffineTerm(x, 1), ScalarAffineTerm(x, 1)], 0.0).","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Use Utilities.canonical to return a new function with the duplicate coefficients aggregated together.","category":"page"},{"location":"moi/tutorials/implementing/#Don't-modify-user-data","page":"Implementing a solver interface","title":"Don't modify user-data","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"All data passed to the solver must be copied immediately to internal data structures. Solvers may not modify any input vectors and must assume that input vectors will not be modified by users in the future.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"This applies, for example, to the terms vector in ScalarAffineFunction. Vectors returned to the user, for example, via ObjectiveFunction or ConstraintFunction attributes, must not be modified by the solver afterwards. The in-place version of get! can be used by users to avoid extra copies in this case.","category":"page"},{"location":"moi/tutorials/implementing/#Column-Generation","page":"Implementing a solver interface","title":"Column Generation","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"There is no special interface for column generation. If the solver has a special API for setting coefficients in existing constraints when adding a new variable, it is possible to queue modifications and new variables and then call the solver's API once all of the new coefficients are known.","category":"page"},{"location":"moi/tutorials/implementing/#Solver-specific-attributes","page":"Implementing a solver interface","title":"Solver-specific attributes","text":"","category":"section"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"You don't need to restrict yourself to the attributes defined in the MathOptInterface.jl package.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Solver-specific attributes should be specified by creating an appropriate subtype of AbstractModelAttribute, AbstractOptimizerAttribute, AbstractVariableAttribute, or AbstractConstraintAttribute.","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"For example, Gurobi.jl adds attributes for multiobjective optimization by defining:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"struct NumberOfObjectives <: MOI.AbstractModelAttribute end\n\nfunction MOI.set(model::Optimizer, ::NumberOfObjectives, n::Integer)\n    # Code to set NumberOfObjectives\n    return\nend\n\nfunction MOI.get(model::Optimizer, ::NumberOfObjectives)\n    n = # Code to get NumberOfObjectives\n    return n\nend","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"Then, the user can write:","category":"page"},{"location":"moi/tutorials/implementing/","page":"Implementing a solver interface","title":"Implementing a solver interface","text":"model = Gurobi.Optimizer()\nMOI.set(model, Gurobi.NumberofObjectives(), 3)","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Benchmarks/reference.md\"","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Benchmarks/reference/#Benchmarks","page":"API Reference","title":"Benchmarks","text":"","category":"section"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"Functions to help benchmark the performance of solver wrappers. See The Benchmarks submodule for more details.","category":"page"},{"location":"moi/submodules/Benchmarks/reference/","page":"API Reference","title":"API Reference","text":"Benchmarks.suite\nBenchmarks.create_baseline\nBenchmarks.compare_against_baseline","category":"page"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.suite","page":"API Reference","title":"MathOptInterface.Benchmarks.suite","text":"suite(\n    new_model::Function;\n    exclude::Vector{Regex} = Regex[]\n)\n\nCreate a suite of benchmarks. new_model should be a function that takes no arguments, and returns a new instance of the optimizer you wish to benchmark.\n\nUse exclude to exclude a subset of benchmarks.\n\nExamples\n\nsuite() do\n    GLPK.Optimizer()\nend\nsuite(exclude = [r\"delete\"]) do\n    Gurobi.Optimizer(OutputFlag=0)\nend\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.create_baseline","page":"API Reference","title":"MathOptInterface.Benchmarks.create_baseline","text":"create_baseline(suite, name::String; directory::String = \"\"; kwargs...)\n\nRun all benchmarks in suite and save to files called name in directory.\n\nExtra kwargs are based to BenchmarkTools.run.\n\nExamples\n\nmy_suite = suite(() -> GLPK.Optimizer())\ncreate_baseline(my_suite, \"glpk_master\"; directory = \"/tmp\", verbose = true)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Benchmarks/reference/#MathOptInterface.Benchmarks.compare_against_baseline","page":"API Reference","title":"MathOptInterface.Benchmarks.compare_against_baseline","text":"compare_against_baseline(\n    suite, name::String; directory::String = \"\",\n    report_filename::String = \"report.txt\"\n)\n\nRun all benchmarks in suite and compare against files called name in directory that were created by a call to create_baseline.\n\nA report summarizing the comparison is written to report_filename in directory.\n\nExtra kwargs are based to BenchmarkTools.run.\n\nExamples\n\nmy_suite = suite(() -> GLPK.Optimizer())\ncompare_against_baseline(\n    my_suite, \"glpk_master\"; directory = \"/tmp\", verbose = true\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/bridging_constraint.md\"","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Implementing-a-constraint-bridge","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This guide outlines the basic steps to create a new bridge from a constraint expressed in the formalism Function-in-Set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Preliminaries","page":"Implementing a constraint bridge","title":"Preliminaries","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"First, decide on the set you want to bridge. Then, study its properties: the most important one is whether the set is scalar or vector, which impacts the dimensionality of the functions that can be used with the set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A scalar function only has one dimension. MOI defines three types of scalar functions: a variable (VariableIndex), an affine function (ScalarAffineFunction), or a quadratic function (ScalarQuadraticFunction).\nA vector function has several dimensions (at least one). MOI defines three types of vector functions: several variables (VectorOfVariables), an affine function (VectorAffineFunction), or a quadratic function (VectorQuadraticFunction). The main difference with scalar functions is that the order of dimensions can be very important: for instance, in an indicator constraint (Indicator), the first dimension indicates whether the constraint about the second dimension is active.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To explain how to implement a bridge, we present the example of Bridges.Constraint.FlipSignBridge. This bridge maps <= (LessThan) constraints to >= (GreaterThan) constraints. This corresponds to reversing the sign of the inequality. We focus on scalar affine functions (we disregard the cases of a single variable or of quadratic functions). This example is a simplified version of the code included in MOI.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Four-mandatory-parts-in-a-constraint-bridge","page":"Implementing a constraint bridge","title":"Four mandatory parts in a constraint bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The first part of a constraint bridge is a new concrete subtype of Bridges.Constraint.AbstractBridge. This type must have fields to store all the new variables and constraints that the bridge will add. Typically, these types are parametrized by the type of the coefficients in the model.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Then, three sets of functions must be defined:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Bridges.Constraint.bridge_constraint: this function implements the bridge and creates the required variables and constraints.\nsupports_constraint: these functions must return true when the combination of function and set is supported by the bridge. By default, the base implementation always returns false and the bridge does not have to provide this implementation.\nBridges.added_constrained_variable_types and Bridges.added_constraint_types: these functions return the types of variables and constraints that this bridge adds. They are used to compute the set of other bridges that are required to use the one you are defining, if need be.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"More functions can be implemented, for instance to retrieve properties from the bridge or deleting a bridged constraint.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Structure-for-the-bridge","page":"Implementing a constraint bridge","title":"1. Structure for the bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A typical struct behind a bridge depends on the type of the coefficients that are used for the model (typically Float64, but coefficients might also be integers or complex numbers).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This structure must hold a reference to all the variables and the constraints that are created as part of the bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The type of this structure is used throughout MOI as an identifier for the bridge. It is passed as argument to most functions related to bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The best practice is to have the name of this type end with Bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, the bridge maps any ScalarAffineFunction{T}-in-LessThan{T} constraint to a single ScalarAffineFunction{T}-in-GreaterThan{T} constraint. The affine function has coefficients of type T. The bridge is parametrized with T, so that the constraint that the bridge creates also has coefficients of type T.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"struct SignBridge{T<:Number} <: Bridges.Constraint.AbstractBridge\n    constraint::ConstraintIndex{ScalarAffineFunction{T}, GreaterThan{T}}\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Bridge-creation","page":"Implementing a constraint bridge","title":"2. Bridge creation","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The function Bridges.Constraint.bridge_constraint is called whenever the bridge is instantiated for a specific model, with the given function and set. The arguments to bridge_constraint are similar to add_constraint, with the exception of the first argument: it is the Type of the struct defined in the first step (for our example, Type{SignBridge{T}}).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"bridge_constraint returns an instance of the struct defined in the first step. the first step.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, the bridge constraint could be defined as:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.Constraint.bridge_constraint(\n    ::Type{SignBridge{T}}, # Bridge to use.\n    model::ModelLike, # Model to which the constraint is being added.\n    f::ScalarAffineFunction{T}, # Function to rewrite.\n    s::LessThan{T}, # Set to rewrite.\n) where {T}\n    # Create the variables and constraints required for the bridge.\n    con = add_constraint(model, -f, GreaterThan(-s.upper))\n\n    # Return an instance of the bridge type with a reference to all the\n    # variables and constraints that were created in this function.\n    return SignBridge(con)\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Supported-constraint-types","page":"Implementing a constraint bridge","title":"3. Supported constraint types","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The function supports_constraint determines whether the bridge type supports a given combination of function and set.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"This function must closely match bridge_constraint, because it will not be called if supports_constraint returns false.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function supports_constraint(\n    ::Type{SignBridge{T}}, # Bridge to use.\n    ::Type{ScalarAffineFunction{T}}, # Function to rewrite.\n    ::Type{LessThan{T}}, # Set to rewrite.\n) where {T}\n    # Do some computation to ensure that the constraint is supported.\n    # Typically, you can directly return true.\n    return true\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#.-Metadata-about-the-bridge","page":"Implementing a constraint bridge","title":"4. Metadata about the bridge","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To determine whether a bridge can be used, MOI uses a shortest-path algorithm that uses the variable types and the constraints that the bridge can create. This information is communicated from the bridge to MOI using the functions Bridges.added_constrained_variable_types and Bridges.added_constraint_types. Both return lists of tuples: either a list of 1-tuples containing the variable types (typically, ZeroOne or Integer) or a list of 2-tuples contained the functions and sets (like ScalarAffineFunction{T}-GreaterThan).","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For our example, the bridge does not create any constrained variables, and only ScalarAffineFunction{T}-in-GreaterThan{T} constraints:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.added_constrained_variable_types(::Type{SignBridge{T}}) where {T}\n    # The bridge does not create variables, return an empty list of tuples:\n    return Tuple{Type}[]\nend\n\nfunction Bridges.added_constraint_types(::Type{SignBridge{T}}) where {T}\n    return Tuple{Type,Type}[\n        # One element per F-in-S the bridge creates.\n        (ScalarAffineFunction{T}, GreaterThan{T}),\n    ]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"A bridge that creates binary variables would rather have this definition of added_constrained_variable_types:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function Bridges.added_constrained_variable_types(::Type{SomeBridge{T}}) where {T}\n    # The bridge only creates binary variables:\n    return Tuple{Type}[(ZeroOne,)]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nIf you declare the creation of constrained variables in added_constrained_variable_types, the corresponding constraint type VariableIndex must not be indicated in added_constraint_types. This would restrict the use of the bridge to solvers that can add such a constraint after the variable is created.More concretely, if you declare in added_constrained_variable_types that your bridge creates binary variables (ZeroOne), and if you never add such a constraint afterward (you do not call add_constraint(model, var, ZeroOne())), then you must not list (VariableIndex, ZeroOne) in added_constraint_types.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Typically, the function Bridges.Constraint.concrete_bridge_type does not have to be defined for most bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-registration","page":"Implementing a constraint bridge","title":"Bridge registration","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For a bridge to be used by MOI, it must be known by MOI.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#SingleBridgeOptimizer","page":"Implementing a constraint bridge","title":"SingleBridgeOptimizer","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"The first way to do so is to create a single-bridge optimizer. This type of optimizer wraps another optimizer and adds the possibility to use only one bridge. It is especially useful when unit testing bridges.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"It is common practice to use the same name as the type defined for the bridge (SignBridge, in our example) without the suffix Bridge.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"const Sign{T,OT<: ModelLike} =\n    SingleBridgeOptimizer{SignBridge{T}, OT}","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In the context of unit tests, this bridge is used in conjunction with a Utilities.MockOptimizer:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"mock = Utilities.MockOptimizer(\n    Utilities.UniversalFallback(Utilities.Model{Float64}()),\n)\nbridged_mock = Sign{Float64}(mock)","category":"page"},{"location":"moi/tutorials/bridging_constraint/#New-bridge-for-a-LazyBridgeOptimizer","page":"Implementing a constraint bridge","title":"New bridge for a LazyBridgeOptimizer","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Typical user-facing models for MOI are based on Bridges.LazyBridgeOptimizer. For instance, this type of model is returned by Bridges.full_bridge_optimizer. These models can be added more bridges by using Bridges.add_bridge:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"inner_optimizer = Utilities.Model{Float64}()\noptimizer = Bridges.full_bridge_optimizer(inner_optimizer, Float64)\nBridges.add_bridge(optimizer, SignBridge{Float64})","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-improvements","page":"Implementing a constraint bridge","title":"Bridge improvements","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/#Attribute-retrieval","page":"Implementing a constraint bridge","title":"Attribute retrieval","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"Like models, bridges have attributes that can be retrieved using get and set. The most important ones are the number of variables and constraints, but also the lists of variables and constraints.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our example, we only have one constraint and only have to implement the NumberOfConstraints and ListOfConstraintIndices attributes:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    ::SignBridge{T},\n    ::NumberOfConstraints{\n        ScalarAffineFunction{T},\n        GreaterThan{T},\n    },\n) where {T}\n    return 1\nend\n\nfunction get(\n    bridge::SignBridge{T},\n    ::ListOfConstraintIndices{\n        ScalarAffineFunction{T},\n        GreaterThan{T},\n    },\n) where {T}\n    return [bridge.constraint]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"You must implement one such pair of functions for each type of constraint the bridge adds to the model.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nAvoid returning a list from the bridge object without copying it. Users must be able to change the contents of the returned list without altering the bridge object.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"For variables, the situation is simpler. If your bridge creates new variables, you must implement the NumberOfVariables and ListOfVariableIndices attributes. However, these attributes do not have parameters, unlike their constraint counterparts. Only two functions suffice:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    ::SignBridge{T},\n    ::NumberOfVariables,\n) where {T}\n    return 0\nend\n\nfunction get(\n    ::SignBridge{T},\n    ::ListOfVariableIndices,\n) where {T}\n    return VariableIndex[]\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In order for the user to be able to access the function and set of the original constraint, the bridge needs to implement getters for the ConstraintFunction and ConstraintSet attributes:","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function get(\n    model::MOI.ModelLike,\n    attr::MOI.ConstraintFunction,\n    bridge::SignBridge,\n)\n    return -MOI.get(model, attr, bridge.constraint)\nend\n\nfunction get(\n    model::MOI.ModelLike,\n    attr::MOI.ConstraintSet,\n    bridge::SignBridge,\n)\n    set = MOI.get(model, attr, bridge.constraint)\n    return MOI.LessThan(-set.lower)\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"warning: Warning\nAlternatively, one could store the original function and set in SignBridge during Bridges.Constraint.bridge_constraint to make these getters simpler and more efficient. On the other hand, this will increase the memory footprint of the bridges as the garbage collector won't be able to delete that object. The convention is to not store the function in the bridge and not care too much about the efficiency of these getters. If the user needs efficient getters for ConstraintFunction then they should use a Utilities.CachingOptimizer.","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Model-modifications","page":"Implementing a constraint bridge","title":"Model modifications","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"To avoid copying the model when the user request to change a constraint, MOI provides modify. Bridges can also implement this API to allow certain changes, such as coefficient changes.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"In our case, a modification of a coefficient in the original constraint (for example, replacing the value of the coefficient of a variable in the affine function) must be transmitted to the constraint created by the bridge, but with a sign change.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function modify(\n    model::ModelLike,\n    bridge::SignBridge,\n    change::ScalarCoefficientChange,\n)\n    modify(\n        model,\n        bridge.constraint,\n        ScalarCoefficientChange(change.variable, -change.new_coefficient),\n    )\n    return\nend","category":"page"},{"location":"moi/tutorials/bridging_constraint/#Bridge-deletion","page":"Implementing a constraint bridge","title":"Bridge deletion","text":"","category":"section"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"When a bridge is deleted, the constraints it added must be deleted too.","category":"page"},{"location":"moi/tutorials/bridging_constraint/","page":"Implementing a constraint bridge","title":"Implementing a constraint bridge","text":"function delete(model::ModelLike, bridge::SignBridge)\n    delete(model, bridge.constraint)\n    return\nend","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"EditURL = \"experiment_design.jl\"","category":"page"},{"location":"tutorials/conic/experiment_design/#Experiment-design","page":"Experiment design","title":"Experiment design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial was originally contributed by Arpit Bhatia and Chris Coey.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"This tutorial covers experiment design examples (D-optimal, A-optimal, and E-optimal) from section 7.5 of the book Convex Optimization by Boyd and Vandenberghe.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The tutorial uses the following packages","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"using JuMP\nimport SCS\nimport LinearAlgebra\nimport Random","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"info: Info\nThis tutorial uses sets from MathOptInterface. By default, JuMP exports the MOI symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:import MathOptInterface as MOI","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"We set a seed so the random numbers are repeatable:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Random.seed!(1234)","category":"page"},{"location":"tutorials/conic/experiment_design/#The-relaxed-experiment-design-problem","page":"Experiment design","title":"The relaxed experiment design problem","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The basic experiment design problem is as follows.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Given the menu of possible choices for experiments, v_1 ldots v_p, and the total number m of experiments to be carried out, choose the numbers of each type of experiment, that is, m_1 ldots m_p to make the error covariance E small (in  some sense).","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The variables m_1 ldots m_p must, of course, be integers and sum to m the given total number of experiments. This leads to the optimization problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nminleft(mathrmwrt mathbfS_+^nright)  E=left(sum_j=1^p m_j v_j v_j^Tright)^-1 \ntextsubject to  m_i geq 0 \n sumlimits_i=1^p m_i = m \n m_i in mathbbZquad i=1ldotsp\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The basic experiment design problem can be a hard combinatorial problem when m, the total number of experiments, is comparable to n, since in this case the m_i are all small integers.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In the case when m is large compared to n, however, a good approximate solution can be found by ignoring, or relaxing, the constraint that the m_i are integers.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Let lambda_i=m_i  m which is the fraction of the total number of experiments for which a_j=v_i or the relative frequency of experiment i. We can express the error covariance in terms of lambda_i as:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"E=frac1mleft(sum_i=1^p lambda_i v_i v_i^Tright)^-1","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The vector lambda in mathbfR^p satisfies lambda succeq 0 mathbf1^T lambda=1 and also, each lambda_i is an integer multiple of 1  m. By ignoring this last constraint, we arrive at the problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nminleft(mathrmwrt mathbfS_+^nright)  E=(1  m)left(sum_i=1^p lambda_i v_i v_i^Tright)^-1 \ntextsubject toquad  lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Several scalarizations have been proposed for the experiment design problem, which is a vector optimization problem over the positive semidefinite cone.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"q = 4 # dimension of estimate space\np = 8 # number of experimental vectors\nn_max = 3 # upper bound on lambda\nn = 12\n\nV = randn(q, p)\n\neye = Matrix{Float64}(LinearAlgebra.I, q, q);\nnothing #hide","category":"page"},{"location":"tutorials/conic/experiment_design/#A-optimal-design","page":"Experiment design","title":"A-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In A-optimal experiment design, we minimize tr E, the trace of the covariance matrix. This objective is simply the mean of the norm of the error squared:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"mathbfEe_2^2=mathbfE operatornametrleft(e e^Tright)=operatornametr E","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The A-optimal experiment design problem in SDP form is","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  mathbf1^T u \ntextsubject to  leftbeginalignedsum_i=1^p lambda_i v_i v_i^T  e_k  e_k^T  u_kendalignedright succeq 0 quad k=1 ldots n \n lambda succeq 0 \n  mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"aOpt = Model(SCS.Optimizer)\nset_silent(aOpt)\n@variable(aOpt, np[1:p], lower_bound = 0, upper_bound = n_max)\n@variable(aOpt, u[1:q], lower_bound = 0)\n@constraint(aOpt, sum(np) <= n)\nfor i in 1:q\n    matrix = [\n        V*LinearAlgebra.diagm(0 => np ./ n)*V' eye[:, i]\n        eye[i, :]' u[i]\n    ]\n    @constraint(aOpt, matrix >= 0, PSDCone())\nend\n@objective(aOpt, Min, sum(u))\noptimize!(aOpt)\nobjective_value(aOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"tutorials/conic/experiment_design/#E-optimal-design","page":"Experiment design","title":"E-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"In E -optimal design, we minimize the norm of the error covariance matrix, that is, the maximum eigenvalue of E.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"Since the diameter (twice the longest semi-axis) of the confidence ellipsoid mathcalE is proportional to E_2^1  2, minimizing E_2 can be interpreted geometrically as minimizing the diameter of the confidence ellipsoid.","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"E-optimal design can also be interpreted as minimizing the maximum variance of q^T e, over all q with q_2=1. The E-optimal experiment design problem in SDP form is:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  t \ntextsubject to  sum_i=1^p lambda_i v_i v_i^T succeq t I \n lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"eOpt = Model(SCS.Optimizer)\nset_silent(eOpt)\n@variable(eOpt, 0 <= np[1:p] <= n_max)\n@variable(eOpt, t)\n@constraint(\n    eOpt,\n    V * LinearAlgebra.diagm(0 => np ./ n) * V' - (t .* eye) >= 0,\n    PSDCone(),\n)\n@constraint(eOpt, sum(np) <= n)\n@objective(eOpt, Max, t)\noptimize!(eOpt)\nobjective_value(eOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"tutorials/conic/experiment_design/#D-optimal-design","page":"Experiment design","title":"D-optimal design","text":"","category":"section"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"The most widely used scalarization is called D -optimal design, in which we minimize the determinant of the error covariance matrix E. This corresponds to designing the experiment to minimize the volume of the resulting confidence ellipsoid (for a fixed confidence level). Ignoring the constant factor 1  m in E, and taking the logarithm of the objective, we can pose this problem as convex optimization problem:","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"beginaligned\nmin  log operatornamedetleft(sum_i=1^p lambda_i v_i v_i^Tright)^-1 \ntextsubject to  lambda succeq 0 \n mathbf1^T lambda=1\nendaligned","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"dOpt = Model(SCS.Optimizer)\nset_silent(dOpt)\n@variable(dOpt, np[1:p], lower_bound = 0, upper_bound = n_max)\n@variable(dOpt, t)\n@objective(dOpt, Max, t)\n@constraint(dOpt, sum(np) <= n)\nE = V * LinearAlgebra.diagm(0 => np ./ n) * V'\n@constraint(dOpt, [t; 1; triangle_vec(E)] in MOI.LogDetConeTriangle(q))\noptimize!(dOpt)\nobjective_value(dOpt)","category":"page"},{"location":"tutorials/conic/experiment_design/","page":"Experiment design","title":"Experiment design","text":"value.(np)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/nonlinear/#Nonlinear-Modeling","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nThis page describes a new nonlinear interface to JuMP. It replaces the legacy @NL interface, which is documented at Nonlinear Modeling (Legacy). The API described below is stable, and it will not break with future 1.X releases of JuMP. However, solver support may be limited, and there may be gaps in functionality compared with the legacy interface. To report a bug, or request a missing feature, please open an issue.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP has support for nonlinear (convex and nonconvex) optimization problems. JuMP is able to automatically provide exact, sparse second-order derivatives to solvers. This information can improve solver accuracy and performance.","category":"page"},{"location":"manual/nonlinear/#Set-a-nonlinear-objective","page":"Nonlinear Modeling","title":"Set a nonlinear objective","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @objective to set a nonlinear objective.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @objective(model, Min, exp(x[1]) - sqrt(x[2]))\nexp(x[1]) - sqrt(x[2])","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"To modify a nonlinear objective, call @objective again.","category":"page"},{"location":"manual/nonlinear/#Add-a-nonlinear-constraint","page":"Nonlinear Modeling","title":"Add a nonlinear constraint","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @constraint to add a nonlinear constraint.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, exp(x[1]) <= 1)\nexp(x[1]) - 1.0 ≤ 0\n\njulia> @constraint(model, con[i = 1:2], 2^x[i] >= i)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarNonlinearFunction, MathOptInterface.GreaterThan{Float64}}, ScalarShape}}:\n con[1] : (2.0 ^ x[1]) - 1.0 ≥ 0\n con[2] : (2.0 ^ x[2]) - 2.0 ≥ 0","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Delete a nonlinear constraint using delete:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> delete(model, con[1])","category":"page"},{"location":"manual/nonlinear/#Create-a-nonlinear-expression","page":"Nonlinear Modeling","title":"Create a nonlinear expression","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use @expression to create nonlinear expression objects:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2]);\n\njulia> expr = @expression(model, exp(x[1]) + sqrt(x[2]))\nexp(x[1]) + sqrt(x[2])\n\njulia> my_anon_expr = @expression(model, [i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpr}:\n sin(x[1])\n sin(x[2])\n\njulia> @expression(model, my_expr[i = 1:2], sin(x[i]))\n2-element Vector{NonlinearExpr}:\n sin(x[1])\n sin(x[2])","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"A NonlinearExpr can be used in @objective, @constraint, and even nested in other @expressions.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> @objective(model, Min, expr^2 + 1)\n((exp(x[1]) + sqrt(x[2])) ^ 2.0) + 1.0\n\njulia> @constraint(model, [i = 1:2], my_expr[i] <= i)\n2-element Vector{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarNonlinearFunction, MathOptInterface.LessThan{Float64}}, ScalarShape}}:\n sin(x[1]) - 1.0 ≤ 0\n sin(x[2]) - 2.0 ≤ 0\n\njulia> @expression(model, nested[i = 1:2], sin(my_expr[i]))\n2-element Vector{NonlinearExpr}:\n sin(sin(x[1]))\n sin(sin(x[2]))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Use value to query the value of a nonlinear expression:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> set_start_value(x[1], 1.0)\n\njulia> value(start_value, nested[1])\n0.7456241416655579\n\njulia> sin(sin(1.0))\n0.7456241416655579","category":"page"},{"location":"manual/nonlinear/#Nonlinear-expressions-in-detail","page":"Nonlinear Modeling","title":"Nonlinear expressions in detail","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions in JuMP are represented by a NonlinearExpr object.","category":"page"},{"location":"manual/nonlinear/#Constructors","page":"Nonlinear Modeling","title":"Constructors","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions can be created using the NonlinearExpr constructors:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = NonlinearExpr(:sin, Any[x])\nsin(x)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"or via operator overloading:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = sin(x)\nsin(x)","category":"page"},{"location":"manual/nonlinear/#Supported-arguments","page":"Nonlinear Modeling","title":"Supported arguments","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Nonlinear expressions can contain a mix of numbers, AffExpr, QuadExpr, and other NonlinearExpr:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> aff = x + 1;\n\njulia> quad = x^2 + x;\n\njulia> expr = cos(x) * sin(quad) + aff\n(cos(x) * sin(x² + x)) + (x + 1)","category":"page"},{"location":"manual/nonlinear/#Supported-operators","page":"Nonlinear Modeling","title":"Supported operators","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The list of supported operators may vary between solvers. Given an optimizer, query the list of supported operators using MOI.ListOfSupportedNonlinearOperators:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> import Ipopt\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.get(Ipopt.Optimizer(), MOI.ListOfSupportedNonlinearOperators())\n85-element Vector{Symbol}:\n :+\n :-\n :abs\n :sqrt\n :cbrt\n :abs2\n :inv\n :log\n :log10\n :log2\n ⋮\n :min\n :max\n :&&\n :||\n :<=\n :(==)\n :>=\n :<\n :>","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"In some univariate cases, the operator is defined in SpecialFunctions.jl. To use these functions, you must explicitly import SpecialFunctions.jl","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> import Ipopt\n\njulia> op = MOI.get(Ipopt.Optimizer(), MOI.ListOfSupportedNonlinearOperators());\n\njulia> :erfcx in op\ntrue\n\njulia> :dawson in op\ntrue\n\njulia> import SpecialFunctions\n\njulia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @expression(model, SpecialFunctions.erfcx(x))\nerfcx(x)\n\njulia> @expression(model, SpecialFunctions.dawson(x))\ndawson(x)","category":"page"},{"location":"manual/nonlinear/#Limitations","page":"Nonlinear Modeling","title":"Limitations","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Some nonlinear expressions cannot be created via operator overloading. For example, to minimize the likelihood of bugs in user-code, we have not overloaded comparisons such as < and >= between JuMP objects:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> x < 1\nERROR: Cannot evaluate `<` between a variable and a number.\n[...]","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Instead, wrap the expression in the @expression macro:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = @expression(model, x < 1)\nx < 1","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For technical reasons, other operators that are not overloaded include ||, &&, and ifelse.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = @expression(model, ifelse(x < -1 || x >= 1, x^2, 0.0))\nifelse((x < -1) || (x >= 1), x², 0.0)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"As an alternative, use the JuMP.op_ functions, which fallback to the various comparison and logical operators:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> expr = op_ifelse(\n           op_or(op_strictly_less_than(x, -1), op_greater_than_or_equal_to(x, 1)),\n           x^2,\n           0.0,\n       )\nifelse((x < -1) || (x >= 1), x², 0.0)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The available functions are:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP function Julia function\nop_ifelse ifelse\nop_and &&\nop_or ||\nop_greater_than_or_equal_to >=\nop_less_than_or_equal_to <=\nop_equal_to ==\nop_strictly_greater_than >\nop_strictly_less_than <","category":"page"},{"location":"manual/nonlinear/#Fields","page":"Nonlinear Modeling","title":"Fields","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Each NonlinearExpr has two fields.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The .head field is a Symbol that represents the operator being called:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> expr.head\n:sin","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The .args field is a Vector{Any} containing the arguments to the operator:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> expr.args\n1-element Vector{Any}:\n x","category":"page"},{"location":"manual/nonlinear/#jump_user_defined_operators","page":"Nonlinear Modeling","title":"User-defined operators","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"In addition to a standard list of univariate and multivariate operators recognized by the MOI.Nonlinear submodule, JuMP supports user-defined Julia operators.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nUser-defined operators must return a scalar output. For a work-around, see User-defined operators with vector outputs.","category":"page"},{"location":"manual/nonlinear/#Add-an-operator","page":"Nonlinear Modeling","title":"Add an operator","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Add a user-defined operator using the @operator macro:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\nmodel = Model();\n@operator(model, op_square, 1, square)\n@operator(model, op_f, 2, f)\n@variable(model, x[1:2]);\n@objective(model, Min, op_f(x[1], op_square(x[2])))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The arguments to @operator are:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The model to which the operator is added.\nA Julia symbol object which serves as the name of the user-defined operator in JuMP expressions. This name must not be the same as that of the function.\nThe number of scalar input arguments that the function takes.\nA Julia method which computes the function.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nUser-defined operators cannot be deleted.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"You can obtain a reference to the operator using the model[:key] syntax:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nmodel = Model();\n@operator(model, op_square, 1, square)\nop_square_2 = model[:op_square]","category":"page"},{"location":"manual/nonlinear/#Add-an-operator-without-macros","page":"Nonlinear Modeling","title":"Add an operator without macros","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The @operator macro is syntactic sugar for add_nonlinear_operator. Thus, the non-macro version of the preceding example is:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nsquare(x) = x^2\nf(x, y) = (x - 1)^2 + (y - 2)^2\nmodel = Model();\nop_square = add_nonlinear_operator(model, 1, square; name = :op_square)\nmodel[:op_square] = op_square\nop_f = add_nonlinear_operator(model, 2, f; name = :op_f)\nmodel[:op_f] = op_f\n@variable(model, x[1:2]);\n@objective(model, Min, op_f(x[1], op_square(x[2])))","category":"page"},{"location":"manual/nonlinear/#Operators-with-the-same-name-as-an-existing-function","page":"Nonlinear Modeling","title":"Operators with the same name as an existing function","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"A common error encountered is the following:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> using JuMP\n\njulia> model = Model();\n\njulia> f(x) = x^2\nf (generic function with 1 method)\n\njulia> @operator(model, f, 1, f)\nERROR: Unable to add the nonlinear operator `:f` with the same name as\nan existing function.\n[...]","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"This error occurs because @operator(model, f, 1, f) is equivalent to:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> f = add_nonlinear_operator(model, 1, f; name = :f)","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"but f already exists as a Julia function.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"If you evaluate the function without adding it as an operator, JuMP will trace the function using operator overloading:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> @variable(model, x);\n\njulia> f(x)\nx²","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"To force JuMP to treat f as a user-defined operator and not trace it, add the operator using add_nonlinear_operator and define a new method which manually creates a NonlinearExpr:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"julia> _ = add_nonlinear_operator(model, 1, f; name = :f)\nNonlinearOperator(f, :f)\n\njulia> f(x::AbstractJuMPScalar) = NonlinearExpr(:f, Any[x])\nf (generic function with 2 methods)\n\njulia> @expression(model, log(f(x)))\nlog(f(x))","category":"page"},{"location":"manual/nonlinear/#Gradients-and-Hessians","page":"Nonlinear Modeling","title":"Gradients and Hessians","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"By default, JuMP will use automatic differentiation to compute the gradient and Hessian of user-defined operators. If your function is not amenable to automatic differentiation, or you can compute analytic derivatives, you may pass additional arguments to @operator to compute the first- and second-derivatives.","category":"page"},{"location":"manual/nonlinear/#Univariate-functions","page":"Nonlinear Modeling","title":"Univariate functions","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For univariate functions, a gradient function ∇f returns a number that represents the first-order derivative. You may, in addition, pass a third function which returns a number representing the second-order derivative:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nf(x) = x^2\n∇f(x) = 2x\n∇²f(x) = 2\nmodel = Model();\n@operator(model, op_f, 1, f, ∇f, ∇²f)  # Providing ∇²f is optional\n@variable(model, x)\n@objective(model, Min, op_f(x))","category":"page"},{"location":"manual/nonlinear/#Multivariate-functions","page":"Nonlinear Modeling","title":"Multivariate functions","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"For multivariate functions, the gradient function ∇f must take an AbstractVector as the first argument that is filled in-place. The Hessian function, ∇²f, must take an AbstractMatrix as the first argument, the lower-triangular of which is filled in-place:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nf(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2\nfunction ∇f(g::AbstractVector{T}, x::T...) where {T}\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend\nfunction ∇²f(H::AbstractMatrix{T}, x::T...) where {T}\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1]  <-- Not needed. Fill the lower-triangular only.\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend\nmodel = Model();\n@operator(model, rosenbrock, 2, f, ∇f, ∇²f)  # Providing ∇²f is optional\n@variable(model, x[1:2])\n@objective(model, Min, rosenbrock(x[1], x[2]))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"You may assume the Hessian matrix H is initialized with zeros, and because H is symmetric, you need only to fill in the non-zero lower-triangular terms. The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). Moreover, you may assume only that H supports size(H) and setindex!. Finally, the matrix is treated as dense, so the performance will be poor on functions with high-dimensional input.","category":"page"},{"location":"manual/nonlinear/#User-defined-operators-with-vector-inputs","page":"Nonlinear Modeling","title":"User-defined operators with vector inputs","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"User-defined operators which take vectors as input arguments (for example, f(x::Vector)) are not supported. Instead, use Julia's splatting syntax to create a function with scalar arguments. For example, instead of:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x::Vector) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"define:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x...) = sum(x[i]^i for i in 1:length(x))","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Another approach is to define the splatted function as an anonymous function:","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"using JuMP\nmodel = Model();\n@variable(model, x[1:5])\nf(x::Vector) = sum(x[i]^i for i in 1:length(x))\n@operator(model, op_f, 5, (x...) -> f(collect(x)))\n@objective(model, Min, op_f(x...))","category":"page"},{"location":"manual/nonlinear/#Automatic-differentiation","page":"Nonlinear Modeling","title":"Automatic differentiation","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP does not support black-box optimization, so all user-defined operators must provide derivatives in some form. Fortunately, JuMP supports automatic differentiation of user-defined operators.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"info: Info\nAutomatic differentiation is not finite differencing. JuMP's automatically computed derivatives are not subject to approximation error.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"JuMP uses ForwardDiff.jl to perform automatic differentiation of user-defined operators; see the ForwardDiff.jl documentation for a description of how to write a function suitable for automatic differentiation.","category":"page"},{"location":"manual/nonlinear/#Common-mistakes-when-writing-a-user-defined-operator","page":"Nonlinear Modeling","title":"Common mistakes when writing a user-defined operator","text":"","category":"section"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"warning: Warning\nGet an error like No method matching Float64(::ForwardDiff.Dual)? Read this section, and see the guidelines at ForwardDiff.jl.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"The most common error is that your user-defined operator is not generic with respect to the number type, that is, don't assume that the input to the function is Float64.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"f(x::Float64) = 2 * x  # This will not work.\nf(x::Real)    = 2 * x  # This is good.\nf(x)          = 2 * x  # This is also good.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"Another reason you may encounter this error is if you create arrays inside your function which are Float64.","category":"page"},{"location":"manual/nonlinear/","page":"Nonlinear Modeling","title":"Nonlinear Modeling","text":"function bad_f(x...)\n    y = zeros(length(x))  # This constructs an array of `Float64`!\n    for i in 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend\n\nfunction good_f(x::T...) where {T<:Real}\n    y = zeros(T, length(x))  # Construct an array of type `T` instead!\n    for i in 1:length(x)\n        y[i] = x[i]^i\n    end\n    return sum(y)\nend","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/background/naming_conventions.md\"","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/background/naming_conventions/#Naming-conventions","page":"Naming conventions","title":"Naming conventions","text":"","category":"section"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"MOI follows several conventions for naming functions and structures. These should also be followed by packages extending MOI.","category":"page"},{"location":"moi/background/naming_conventions/#Sets","page":"Naming conventions","title":"Sets","text":"","category":"section"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"Sets encode the structure of constraints. Their names should follow the following conventions:","category":"page"},{"location":"moi/background/naming_conventions/","page":"Naming conventions","title":"Naming conventions","text":"Abstract types in the set hierarchy should begin with Abstract and end in Set, for example, AbstractScalarSet, AbstractVectorSet.\nVector-valued conic sets should end with Cone, for example, NormInfinityCone, SecondOrderCone.\nVector-valued Cartesian products should be plural and not end in Cone, for example, Nonnegatives, not NonnegativeCone.\nMatrix-valued conic sets should provide two representations: ConeSquare and ConeTriangle, for example, RootDetConeTriangle and RootDetConeSquare. See Matrix cones for more details.\nScalar sets should be singular, not plural, for example, Integer, not Integers.\nAs much as possible, the names should follow established conventions in the domain where this set is used: for instance, convex sets should have names close to those of CVX, and constraint-programming sets should follow MiniZinc's constraints.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"EditURL = \"https://github.com/oxfordcontrol/CDCS.jl/blob/6d165e8e8f59b14e36e99c8efb19f0e775bbd292/README.md\"","category":"page"},{"location":"packages/CDCS/#CDCS.jl","page":"oxfordcontrol/CDCS.jl","title":"CDCS.jl","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"CDCS.jl is a wrapper for the CDCS solver. ","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"an exported cdcs function that is a thin wrapper on top of the cdcs MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/CDCS/#License","page":"oxfordcontrol/CDCS.jl","title":"License","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"CDCS.jl is licensed under the MIT license.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The underlying solver oxfordcontrol/CDCS is licensed under the LGPL v3 license.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"In addition, CDCS requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/CDCS/#Installation","page":"oxfordcontrol/CDCS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package and that the CDCS software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"Then, install CDCS.jl using Pkg.add:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"import Pkg\nPkg.add(\"CDCS\")","category":"page"},{"location":"packages/CDCS/#Use-with-JuMP","page":"oxfordcontrol/CDCS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"To use CDCS with JuMP, do:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"using JuMP, CDCS\nmodel = Model(CDCS.Optimizer)\nset_attribute(model, \"verbose\", 0)","category":"page"},{"location":"packages/CDCS/#Troubleshooting","page":"oxfordcontrol/CDCS.jl","title":"Troubleshooting","text":"","category":"section"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"If you get the error:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"Undefined function or variable 'cdcs'.\n\nError using save\nVariable 'jx_cdcs_arg_out_1' not found.\n\nLinear Programming example: Error During Test at /home/blegat/.julia/dev/CDCS/test/lp.jl:5\n  Got exception outside of a @test\n  MATLAB.MEngineError(\"failed to get variable jx_cdcs_arg_out_1 from MATLAB session\")\n  Stacktrace:\n    [1] get_mvariable(session::MATLAB.MSession, name::Symbol)\n      @ MATLAB ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:164\n    [2] mxcall(::MATLAB.MSession, ::Symbol, ::Int64, ::Matrix{Float64}, ::Vararg{Any})\n      @ MATLAB ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:297\n    [3] mxcall\n      @ ~/.julia/packages/MATLAB/SVjnA/src/engine.jl:317 [inlined]\n    [4] cdcs(A::Matrix{Float64}, b::Vector{Float64}, c::Vector{Float64}, K::CDCS.Cone; kws::Base.Pairs{Symbol, Int64, Tuple{Symbol}, NamedTuple{(:verbose,), Tuple{Int64}}})","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"The error means that we could not find the cdcs function with one output argument using the MATLAB C API.","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"This most likely means that you did not add CDCS to the MATLAB's path (that is, the toolbox/local/pathdef.m file).","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work where /path/to/CDCS/ is the directory where the CDCS folder is located:","category":"page"},{"location":"packages/CDCS/","page":"oxfordcontrol/CDCS.jl","title":"oxfordcontrol/CDCS.jl","text":"julia> using MATLAB\n\njulia> cd(\"/path/to/CDCS/\") do\n           mat\"cdcsInstall\"\n       end\n\njulia> mat\"savepath\"","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"EditURL = \"https://github.com/jump-dev/MultiObjectiveAlgorithms.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"<img src=\"https://raw.githubusercontent.com/jump-dev/MultiObjectiveAlgorithms.jl/master/moa.png?raw=true\" alt=\"An image of the Moa bird. Licensed into the Public Domain by https://freesvg.org/moa\" width=\"100px\"/>","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#MultiObjectiveAlgorithms.jl","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"MultiObjectiveAlgorithms.jl","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MultiObjectiveAlgorithms.jl (MOA) is a collection of algorithms for multi-objective optimization.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#License","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"License","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MultiObjectiveAlgorithms.jl is licensed under the MPL 2.0 License.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Installation","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Install MOA using Pkg.add:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"import Pkg\nPkg.add(\"MultiObjectiveAlgorithms\")","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Use-with-JuMP","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Use MultiObjectiveAlgorithms with JuMP as follows:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA\nmodel = JuMP.Model(() -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.Dichotomy())\nset_attribute(model, MOA.SolutionLimit(), 4)","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Replace HiGHS.Optimizer with an optimizer capable of solving a single-objective instance of your optimization problem.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"You may set additional optimizer attributes, the supported attributes depend on the choice of solution algorithm.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Algorithm","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Algorithm","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Set the algorithm using the MOA.Algorithm() attribute.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"The value must be one of the algorithms supported by MOA:","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MOA.Chalmet()\nMOA.Dichotomy()\nMOA.DominguezRios()\nMOA.EpsilonConstraint()\nMOA.Hierarchical()\nMOA.KirlikSayin()\nMOA.Lexicographic() [default]\nMOA.TambyVanderpooten()","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Consult their docstrings for details.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/#Other-optimizer-attributes","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"Other optimizer attributes","text":"","category":"section"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"There are a number of optimizer attributes supported by the algorithms in MOA.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"Each algorithm supports only a subset of the attributes. Consult the algorithm's docstring for details on which attributes it supports, and how it uses them in the solution process.","category":"page"},{"location":"packages/MultiObjectiveAlgorithms/","page":"jump-dev/MultiObjectiveAlgorithms.jl","title":"jump-dev/MultiObjectiveAlgorithms.jl","text":"MOA.EpsilonConstraintStep()\nMOA.LexicographicAllPermutations()\nMOA.ObjectiveAbsoluteTolerance(index::Int)\nMOA.ObjectivePriority(index::Int)\nMOA.ObjectiveRelativeTolerance(index::Int)\nMOA.ObjectiveWeight(index::Int)\nMOA.SolutionLimit()\nMOI.TimeLimitSec()","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"EditURL = \"multi_commodity_network.jl\"","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#The-network-multi-commodity-flow-problem","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial is a variation of The multi-commodity flow problem where the graph is a network instead of a bipartite graph.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate a style of modeling that uses relational algebra.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Required-packages","page":"The network multi-commodity flow problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"using JuMP\nimport DataFrames\nimport HiGHS\nimport SQLite\nimport SQLite.DBInterface","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Formulation","page":"The network multi-commodity flow problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The network multi-commondity flow problem is an extension of the The multi-commodity flow problem, where instead of having a bipartite graph of supply and demand nodes, the graph can contains a set of nodes, i in mathcalN , which each have a (potentially zero) supply capacity, u^s_ip, and (potentially zero) a demand, d_ip for each commodity p in P. The nodes are connected by a set of edges (i j) in mathcalE, which have a shipment cost c^x_ijp and a total flow capacity of u^x_ij.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Our take is to choose an optimal supply for each node s_ip, as well as the optimal transshipment x_ijp that minimizes the total cost.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The mathematical formulation is:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"beginaligned\nmin   sum_(ij)inmathcalE p in P c^x_ijp x_ijp + sum_iinmathcalN p in P c^s_ip s_ip \nst   s_ip + sum_(j i) in mathcalE x_jip - sum_(ij) in mathcalE x_ijp = d_ip  forall i in mathcalN p in P \n           x_ijp ge 0            forall (i j) in mathcalE p in P \n           sum_p in P x_ijp le u^x_ij            forall (i j) in mathcalE \n           0 le s_ip le u^s_ip  forall i in mathcalN p in P\nendaligned","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The purpose of this tutorial is to demonstrate how this model can be built using relational algebra instead of a direct math-to-code translation of the summations.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Data","page":"The network multi-commodity flow problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the purpose of this tutorial, the JuMP repository contains an example database called commodity_nz.db:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"filename = joinpath(@__DIR__, \"commodity_nz.db\");\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"To run locally, download commodity_nz.db and update filename appropriately.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Load the database using SQLite.DB:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"db = SQLite.DB(filename)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"A quick way to see the schema of the database is via SQLite.tables:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"SQLite.tables(db)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We interact with the database by executing queries and then loading the results into a DataFrame:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"function get_table(db, table)\n    query = DBInterface.execute(db, \"SELECT * FROM $table\")\n    return DataFrames.DataFrame(query)\nend","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The shipping table contains the set of arcs and their distances:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping = get_table(db, \"shipping\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The products table contains the shipping cost per kilometer of each product:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_products = get_table(db, \"products\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The supply table contains the supply capacity of each node, as well as the cost:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_supply = get_table(db, \"supply\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"The demand table contains the demand of each node:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_demand = get_table(db, \"demand\")","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#JuMP-formulation","page":"The network multi-commodity flow problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We start by creating a model and our decision variables:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the shipping decisions, we create a new column in df_shipping called x_flow, which has one non-negative decision variable for each row:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping.x_flow = @variable(model, x[1:size(df_shipping, 1)] >= 0)\ndf_shipping","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the supply, we add a variable to each row, and then set the upper bound to the capacity of each supply node:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_supply.x_supply = @variable(model, s[1:size(df_supply, 1)] >= 0)\nset_upper_bound.(df_supply.x_supply, df_supply.capacity)\ndf_supply","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Our objective is to minimize the shipping cost plus the supply cost. To compute the flow cost, we need to join the shipping table, which contains distance_km with the products table, which contains cost_per_km:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_cost = DataFrames.leftjoin(df_shipping, df_products; on = [:product])\ndf_cost.flow_cost = df_cost.cost_per_km .* df_cost.distance_km\ndf_cost","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Then we can use linear algebra to compute the inner product between two columns:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"@objective(\n    model,\n    Min,\n    df_cost.flow_cost' * df_shipping.x_flow +\n    df_supply.cost' * df_supply.x_supply\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For the flow capacities on each arc, we use DataFrames.groupby to partition the flow variables based on :origin and :destination, and then we constrain their sum to be less than a fixed capacity.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"capacity = 30\nfor df in DataFrames.groupby(df_shipping, [:origin, :destination])\n    @constraint(model, sum(df.x_flow) <= capacity)\nend","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"For each node in the graph, we need to compute a mass balance constraint which says that for each product, the supply, plus the flow into the node, and less the flow out of the node is equal to the demand.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can compute an expression for the flow out of each node using DataFrames.groupby on the origin and product columns of the df_shipping table:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_flow_out = DataFrames.DataFrame(\n    (node = i.origin, product = i.product, x_flow_out = sum(df.x_flow)) for\n    (i, df) in pairs(DataFrames.groupby(df_shipping, [:origin, :product]))\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can compute an expression for the flow into each node using DataFrames.groupby on the destination and product columns of the df_shipping table:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_flow_in = DataFrames.DataFrame(\n    (node = i.destination, product = i.product, x_flow_in = sum(df.x_flow))\n    for (i, df) in\n    pairs(DataFrames.groupby(df_shipping, [:destination, :product]))\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"We can join the two tables together using DataFrames.outerjoin. We need to use outerjoin here because there might be missing rows.","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.outerjoin(df_flow_in, df_flow_out; on = [:node, :product])","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Next, we need to join the supply column:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.leftjoin(\n    df,\n    DataFrames.select(df_supply, [:origin, :product, :x_supply]);\n    on = [:node => :origin, :product],\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"and then the demand column","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df = DataFrames.leftjoin(\n    df,\n    DataFrames.select(df_demand, [:destination, :product, :demand]);\n    on = [:node => :destination, :product],\n)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Now we're ready to add our mass balance constraint. Because some rows contain missing values, we need to use coalesce to convert any missing into a numeric value:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"@constraint(\n    model,\n    [r in eachrow(df)],\n    coalesce(r.x_supply, 0.0) + coalesce(r.x_flow_in, 0.0) -\n    coalesce(r.x_flow_out, 0.0) == coalesce(r.demand, 0.0),\n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/#Solution","page":"The network multi-commodity flow problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"Finally, we can optimize the model:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"update the solution in the DataFrames:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"df_shipping.x_flow = value.(df_shipping.x_flow)\ndf_supply.x_supply = value.(df_supply.x_supply);\nnothing #hide","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"and display the optimal solution for flows:","category":"page"},{"location":"tutorials/linear/multi_commodity_network/","page":"The network multi-commodity flow problem","title":"The network multi-commodity flow problem","text":"DataFrames.select(\n    filter!(row -> row.x_flow > 0.0, df_shipping),\n    [:origin, :destination, :product, :x_flow],\n)","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/standard_form.md\"","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/standard_form/#Standard-form","page":"Standard form","title":"Standard form","text":"","category":"section"},{"location":"moi/reference/standard_form/#Functions","page":"Standard form","title":"Functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractFunction\noutput_dimension\nconstant","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractFunction","page":"Standard form","title":"MathOptInterface.AbstractFunction","text":"AbstractFunction\n\nAbstract supertype for function objects.\n\nRequired methods\n\nAll functions must implement:\n\nBase.copy\nBase.isapprox\nconstant\n\nAbstract subtypes of AbstractFunction may require additional methods to be implemented.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.output_dimension","page":"Standard form","title":"MathOptInterface.output_dimension","text":"output_dimension(f::AbstractFunction)\n\nReturn 1 if f is an AbstractScalarFunction, or the number of output components if f is an AbstractVectorFunction.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.constant","page":"Standard form","title":"MathOptInterface.constant","text":"constant(f::AbstractFunction[, ::Type{T}]) where {T}\n\nReturns the constant term of a scalar-valued function, or the constant vector of a vector-valued function.\n\nIf f is untyped and T is provided, returns zero(T).\n\n\n\n\n\nconstant(set::Union{EqualTo,GreaterThan,LessThan,Parameter})\n\nReturns the constant term of the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.constant(MOI.GreaterThan(1.0))\n1.0\n\njulia> MOI.constant(MOI.LessThan(2.5))\n2.5\n\njulia> MOI.constant(MOI.EqualTo(3))\n3\n\njulia> MOI.constant(MOI.Parameter(4.5))\n4.5\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#Scalar-functions","page":"Standard form","title":"Scalar functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractScalarFunction\nVariableIndex\nScalarAffineTerm\nScalarAffineFunction\nScalarQuadraticTerm\nScalarQuadraticFunction\nScalarNonlinearFunction","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractScalarFunction","page":"Standard form","title":"MathOptInterface.AbstractScalarFunction","text":"abstract type AbstractScalarFunction <: AbstractFunction\n\nAbstract supertype for scalar-valued AbstractFunctions.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VariableIndex","page":"Standard form","title":"MathOptInterface.VariableIndex","text":"VariableIndex\n\nA type-safe wrapper for Int64 for use in referencing variables in a model. To allow for deletion, indices need not be consecutive.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarAffineTerm","page":"Standard form","title":"MathOptInterface.ScalarAffineTerm","text":"ScalarAffineTerm{T}(coefficient::T, variable::VariableIndex) where {T}\n\nRepresents the scalar-valued term coefficient * variable.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarAffineTerm(2.0, x)\nMathOptInterface.ScalarAffineTerm{Float64}(2.0, MOI.VariableIndex(1))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarAffineFunction","page":"Standard form","title":"MathOptInterface.ScalarAffineFunction","text":"ScalarAffineFunction{T}(terms::ScalarAffineTerm{T}, constant::T) where {T}\n\nRepresents the scalar-valued affine function a^top x + b, where:\n\na^top x is represented by the vector of ScalarAffineTerms\nb is a scalar constant::T\n\nDuplicates\n\nDuplicate variable indices in terms are accepted, and the corresponding coefficients are summed together.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> terms = [MOI.ScalarAffineTerm(2.0, x), MOI.ScalarAffineTerm(3.0, x)]\n2-element Vector{MathOptInterface.ScalarAffineTerm{Float64}}:\n MathOptInterface.ScalarAffineTerm{Float64}(2.0, MOI.VariableIndex(1))\n MathOptInterface.ScalarAffineTerm{Float64}(3.0, MOI.VariableIndex(1))\n\njulia> f = MOI.ScalarAffineFunction(terms, 4.0)\n4.0 + 2.0 MOI.VariableIndex(1) + 3.0 MOI.VariableIndex(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarQuadraticTerm","page":"Standard form","title":"MathOptInterface.ScalarQuadraticTerm","text":"ScalarQuadraticTerm{T}(\n    coefficient::T,\n    variable_1::VariableIndex,\n    variable_2::VariableIndex,\n) where {T}\n\nRepresents the scalar-valued term c x_i x_j where c is coefficient, x_i is variable_1 and x_j is variable_2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarQuadraticTerm(2.0, x, x)\nMathOptInterface.ScalarQuadraticTerm{Float64}(2.0, MOI.VariableIndex(1), MOI.VariableIndex(1))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarQuadraticFunction","page":"Standard form","title":"MathOptInterface.ScalarQuadraticFunction","text":"ScalarQuadraticFunction{T}(\n    quadratic_terms::Vector{ScalarQuadraticTerm{T}},\n    affine_terms::Vector{ScalarAffineTerm{T}},\n    constant::T,\n) wher {T}\n\nThe scalar-valued quadratic function frac12x^top Q x + a^top x + b, where:\n\nQ is the symmetric matrix given by the vector of ScalarQuadraticTerms\na^top x is a sparse vector given by the vector of ScalarAffineTerms\nb is the scalar constant::T.\n\nDuplicates\n\nDuplicate indices in quadratic_terms or affine_terms are accepted, and the corresponding coefficients are summed together.\n\nIn quadratic_terms, \"mirrored\" indices, (q, r) and (r, q) where r and q are VariableIndexes, are considered duplicates; only one needs to be specified.\n\nThe 0.5 factor\n\nCoupled with the interpretation of mirrored indices, the 0.5 factor in front of the Q matrix is a common source of bugs.\n\nAs a rule, to represent a * x^2 + b * x * y:\n\nThe coefficient a in front of squared variables (diagonal elements in Q) must be doubled when creating a ScalarQuadraticTerm\nThe coefficient b in front of off-diagonal elements in Q should be left as b, be cause the mirrored index b * y * x will be implicitly added.\n\nExample\n\nTo represent the function f(x y) = 2 * x^2 + 3 * x * y + 4 * x + 5, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> y = MOI.VariableIndex(2);\n\njulia> constant = 5.0;\n\njulia> affine_terms = [MOI.ScalarAffineTerm(4.0, x)];\n\njulia> quadratic_terms = [\n           MOI.ScalarQuadraticTerm(4.0, x, x),  # Note the changed coefficient\n           MOI.ScalarQuadraticTerm(3.0, x, y),\n       ]\n2-element Vector{MathOptInterface.ScalarQuadraticTerm{Float64}}:\n MathOptInterface.ScalarQuadraticTerm{Float64}(4.0, MOI.VariableIndex(1), MOI.VariableIndex(1))\n MathOptInterface.ScalarQuadraticTerm{Float64}(3.0, MOI.VariableIndex(1), MOI.VariableIndex(2))\n\njulia> f = MOI.ScalarQuadraticFunction(quadratic_terms, affine_terms, constant)\n5.0 + 4.0 MOI.VariableIndex(1) + 2.0 MOI.VariableIndex(1)² + 3.0 MOI.VariableIndex(1)*MOI.VariableIndex(2)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ScalarNonlinearFunction","page":"Standard form","title":"MathOptInterface.ScalarNonlinearFunction","text":"ScalarNonlinearFunction(head::Symbol, args::Vector{Any})\n\nThe scalar-valued nonlinear function head(args...), represented as a symbolic expression tree, with the call operator head and ordered arguments in args.\n\nhead\n\nThe head::Symbol must be an operator supported by the model.\n\nThe default list of supported univariate operators is given by:\n\nNonlinear.DEFAULT_UNIVARIATE_OPERATORS\n\nand the default list of supported multivariate operators is given by:\n\nNonlinear.DEFAULT_MULTIVARIATE_OPERATORS\n\nAdditional operators can be registered by setting a UserDefinedFunction attribute.\n\nSee the full list of operators supported by a ModelLike by querying ListOfSupportedNonlinearOperators.\n\nargs\n\nThe vector args contains the arguments to the nonlinear function. If the operator is univariate, it must contain one element. Otherwise, it may contain multiple elements.\n\nEach element must be one of the following:\n\nA constant value of type T<:Real\nA VariableIndex\nA ScalarAffineFunction\nA ScalarQuadraticFunction\nA ScalarNonlinearFunction\n\nUnsupported operators\n\nIf the optimizer does not support head, an UnsupportedNonlinearOperator error will be thrown.\n\nThere is no guarantee about when this error will be thrown; it may be thrown when the function is first added to the model, or it may be thrown when optimize! is called.\n\nExample\n\nTo represent the function f(x) = sin(x)^2, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> MOI.ScalarNonlinearFunction(\n           :^,\n           Any[MOI.ScalarNonlinearFunction(:sin, Any[x]), 2],\n       )\n^(sin(MOI.VariableIndex(1)), (2))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Vector-functions","page":"Standard form","title":"Vector functions","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractVectorFunction\nVectorOfVariables\nVectorAffineTerm\nVectorAffineFunction\nVectorQuadraticTerm\nVectorQuadraticFunction\nVectorNonlinearFunction","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractVectorFunction","page":"Standard form","title":"MathOptInterface.AbstractVectorFunction","text":"abstract type AbstractVectorFunction <: AbstractFunction\n\nAbstract supertype for vector-valued AbstractFunctions.\n\nRequired methods\n\nAll subtypes of AbstractVectorFunction must implement:\n\noutput_dimension\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorOfVariables","page":"Standard form","title":"MathOptInterface.VectorOfVariables","text":"VectorOfVariables(variables::Vector{VariableIndex}) <: AbstractVectorFunction\n\nThe vector-valued function f(x) = variables, where variables is a subset of VariableIndexes in the model.\n\nThe list of variables may contain duplicates.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex.(1:2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> f = MOI.VectorOfVariables([x[1], x[2], x[1]])\n┌                    ┐\n│MOI.VariableIndex(1)│\n│MOI.VariableIndex(2)│\n│MOI.VariableIndex(1)│\n└                    ┘\n\njulia> MOI.output_dimension(f)\n3\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorAffineTerm","page":"Standard form","title":"MathOptInterface.VectorAffineTerm","text":"VectorAffineTerm{T}(\n    output_index::Int64,\n    scalar_term::ScalarAffineTerm{T},\n) where {T}\n\nA VectorAffineTerm is a scalar_term that appears in the output_index row of the vector-valued VectorAffineFunction or VectorQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x))\nMathOptInterface.VectorAffineTerm{Float64}(2, MathOptInterface.ScalarAffineTerm{Float64}(3.0, MOI.VariableIndex(1)))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorAffineFunction","page":"Standard form","title":"MathOptInterface.VectorAffineFunction","text":"VectorAffineFunction{T}(\n    terms::Vector{VectorAffineTerm{T}},\n    constants::Vector{T},\n) where {T}\n\nThe vector-valued affine function A x + b, where:\n\nA x is the sparse matrix given by the vector of VectorAffineTerms\nb is the vector constants\n\nDuplicates\n\nDuplicate indices in the A are accepted, and the corresponding coefficients are summed together.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> terms = [\n           MOI.VectorAffineTerm(Int64(1), MOI.ScalarAffineTerm(2.0, x)),\n           MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x)),\n       ];\n\njulia> f = MOI.VectorAffineFunction(terms, [4.0, 5.0])\n┌                              ┐\n│4.0 + 2.0 MOI.VariableIndex(1)│\n│5.0 + 3.0 MOI.VariableIndex(1)│\n└                              ┘\n\njulia> MOI.output_dimension(f)\n2\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorQuadraticTerm","page":"Standard form","title":"MathOptInterface.VectorQuadraticTerm","text":"VectorQuadraticTerm{T}(\n    output_index::Int64,\n    scalar_term::ScalarQuadraticTerm{T},\n) where {T}\n\nA VectorQuadraticTerm is a ScalarQuadraticTerm scalar_term that appears in the output_index row of the vector-valued VectorQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> MOI.VectorQuadraticTerm(Int64(2), MOI.ScalarQuadraticTerm(3.0, x, x))\nMathOptInterface.VectorQuadraticTerm{Float64}(2, MathOptInterface.ScalarQuadraticTerm{Float64}(3.0, MOI.VariableIndex(1), MOI.VariableIndex(1)))\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorQuadraticFunction","page":"Standard form","title":"MathOptInterface.VectorQuadraticFunction","text":"VectorQuadraticFunction{T}(\n    quadratic_terms::Vector{VectorQuadraticTerm{T}},\n    affine_terms::Vector{VectorAffineTerm{T}},\n    constants::Vector{T},\n) where {T}\n\nThe vector-valued quadratic function with ith component (\"output index\") defined as frac12x^top Q_i x + a_i^top x + b_i, where:\n\nfrac12x^top Q_i x is the symmetric matrix given by the VectorQuadraticTerm elements in quadratic_terms with output_index == i\na_i^top x is the sparse vector given by the VectorAffineTerm elements in affine_terms with output_index == i\nb_i is a scalar given by constants[i]\n\nDuplicates\n\nDuplicate indices in quadratic_terms and affine_terms with the same output_index are handled in the same manner as duplicates in ScalarQuadraticFunction.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1);\n\njulia> y = MOI.VariableIndex(2);\n\njulia> constants = [4.0, 5.0];\n\njulia> affine_terms = [\n           MOI.VectorAffineTerm(Int64(1), MOI.ScalarAffineTerm(2.0, x)),\n           MOI.VectorAffineTerm(Int64(2), MOI.ScalarAffineTerm(3.0, x)),\n       ];\n\njulia> quad_terms = [\n        MOI.VectorQuadraticTerm(Int64(1), MOI.ScalarQuadraticTerm(2.0, x, x)),\n        MOI.VectorQuadraticTerm(Int64(2), MOI.ScalarQuadraticTerm(3.0, x, y)),\n           ];\n\njulia> f = MOI.VectorQuadraticFunction(quad_terms, affine_terms, constants)\n┌                                                                              ┐\n│4.0 + 2.0 MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(1)²                    │\n│5.0 + 3.0 MOI.VariableIndex(1) + 3.0 MOI.VariableIndex(1)*MOI.VariableIndex(2)│\n└                                                                              ┘\n\njulia> MOI.output_dimension(f)\n2\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.VectorNonlinearFunction","page":"Standard form","title":"MathOptInterface.VectorNonlinearFunction","text":"VectorNonlinearFunction(args::Vector{ScalarNonlinearFunction})\n\nThe vector-valued nonlinear function composed of a vector of ScalarNonlinearFunction.\n\nargs\n\nThe vector args contains the scalar elements of the nonlinear function. Each element must be a ScalarNonlinearFunction, but if you pass a Vector{Any}, the elements can be automatically converted from one of the following:\n\nA constant value of type T<:Real\nA VariableIndex\nA ScalarAffineFunction\nA ScalarQuadraticFunction\nA ScalarNonlinearFunction\n\nExample\n\nTo represent the function f(x) = sin(x)^2 x, do:\n\njulia> import MathOptInterface as MOI\n\njulia> x = MOI.VariableIndex(1)\nMOI.VariableIndex(1)\n\njulia> g = MOI.ScalarNonlinearFunction(\n           :^,\n           Any[MOI.ScalarNonlinearFunction(:sin, Any[x]), 2.0],\n       )\n^(sin(MOI.VariableIndex(1)), 2.0)\n\njulia> MOI.VectorNonlinearFunction([g, x])\n┌                                 ┐\n│^(sin(MOI.VariableIndex(1)), 2.0)│\n│+(MOI.VariableIndex(1))          │\n└                                 ┘\n\nNote the automatic conversion from x to +(x).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Sets","page":"Standard form","title":"Sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractSet\nAbstractScalarSet\nAbstractVectorSet","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSet","page":"Standard form","title":"MathOptInterface.AbstractSet","text":"AbstractSet\n\nAbstract supertype for set objects used to encode constraints.\n\nRequired methods\n\nFor sets of type S with isbitstype(S) == false, you must implement:\n\nBase.copy(set::S)\nBase.:(==)(x::S, y::S)\n\nSubtypes of AbstractSet such as AbstractScalarSet and AbstractVectorSet may prescribe additional required methods.\n\nOptional methods\n\nYou may optionally implement:\n\ndual_set\ndual_set_type\n\nNote for developers\n\nWhen creating a new set, the set struct must not contain any VariableIndex or ConstraintIndex objects.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractScalarSet","page":"Standard form","title":"MathOptInterface.AbstractScalarSet","text":"AbstractScalarSet\n\nAbstract supertype for subsets of mathbbR.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractVectorSet","page":"Standard form","title":"MathOptInterface.AbstractVectorSet","text":"AbstractVectorSet\n\nAbstract supertype for subsets of mathbbR^n for some n.\n\nRequired methods\n\nAll AbstractVectorSets of type S must implement:\n\ndimension, unless the dimension is stored in the set.dimension field\nUtilities.set_dot, unless the dot product between two vectors in the set is equivalent to LinearAlgebra.dot.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Utilities","page":"Standard form","title":"Utilities","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"dimension\ndual_set\ndual_set_type\nconstant(s::EqualTo)\nsupports_dimension_update\nupdate_dimension","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.dimension","page":"Standard form","title":"MathOptInterface.dimension","text":"dimension(set::AbstractSet)\n\nReturn the output_dimension that an AbstractFunction should have to be used with the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dimension(MOI.Reals(4))\n4\n\njulia> MOI.dimension(MOI.LessThan(3.0))\n1\n\njulia> MOI.dimension(MOI.PositiveSemidefiniteConeTriangle(2))\n3\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.dual_set","page":"Standard form","title":"MathOptInterface.dual_set","text":"dual_set(set::AbstractSet)\n\nReturn the dual set of set, that is the dual cone of the set. This follows the definition of duality discussed in Duality.\n\nSee Dual cone for more information.\n\nIf the dual cone is not defined it returns an error.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dual_set(MOI.Reals(4))\nMathOptInterface.Zeros(4)\n\njulia> MOI.dual_set(MOI.SecondOrderCone(5))\nMathOptInterface.SecondOrderCone(5)\n\njulia> MOI.dual_set(MOI.ExponentialCone())\nMathOptInterface.DualExponentialCone()\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.dual_set_type","page":"Standard form","title":"MathOptInterface.dual_set_type","text":"dual_set_type(S::Type{<:AbstractSet})\n\nReturn the type of dual set of sets of type S, as returned by dual_set. If the dual cone is not defined it returns an error.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.dual_set_type(MOI.Reals)\nMathOptInterface.Zeros\n\njulia> MOI.dual_set_type(MOI.SecondOrderCone)\nMathOptInterface.SecondOrderCone\n\njulia> MOI.dual_set_type(MOI.ExponentialCone)\nMathOptInterface.DualExponentialCone\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.constant-Tuple{MathOptInterface.EqualTo}","page":"Standard form","title":"MathOptInterface.constant","text":"constant(set::Union{EqualTo,GreaterThan,LessThan,Parameter})\n\nReturns the constant term of the set set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> MOI.constant(MOI.GreaterThan(1.0))\n1.0\n\njulia> MOI.constant(MOI.LessThan(2.5))\n2.5\n\njulia> MOI.constant(MOI.EqualTo(3))\n3\n\njulia> MOI.constant(MOI.Parameter(4.5))\n4.5\n\n\n\n\n\n","category":"method"},{"location":"moi/reference/standard_form/#MathOptInterface.supports_dimension_update","page":"Standard form","title":"MathOptInterface.supports_dimension_update","text":"supports_dimension_update(S::Type{<:MOI.AbstractVectorSet})\n\nReturn a Bool indicating whether the elimination of any dimension of n-dimensional sets of type S give an n-1-dimensional set S. By default, this function returns false so it should only be implemented for sets that supports dimension update.\n\nFor instance, supports_dimension_update(MOI.Nonnegatives) is true because the elimination of any dimension of the n-dimensional nonnegative orthant gives the n-1-dimensional nonnegative orthant. However supports_dimension_update(MOI.ExponentialCone) is false.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.update_dimension","page":"Standard form","title":"MathOptInterface.update_dimension","text":"update_dimension(s::AbstractVectorSet, new_dim::Int)\n\nReturns a set with the dimension modified to new_dim.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#Scalar-sets","page":"Standard form","title":"Scalar sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized scalar sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"GreaterThan\nLessThan\nEqualTo\nInterval\nInteger\nZeroOne\nSemicontinuous\nSemiinteger\nParameter","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.GreaterThan","page":"Standard form","title":"MathOptInterface.GreaterThan","text":"GreaterThan{T<:Real}(lower::T)\n\nThe set lower infty) subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.GreaterThan(0.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LessThan","page":"Standard form","title":"MathOptInterface.LessThan","text":"LessThan{T<:Real}(upper::T)\n\nThe set (-infty upper subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.LessThan(2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.LessThan{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.EqualTo","page":"Standard form","title":"MathOptInterface.EqualTo","text":"EqualTo{T<:Number}(value::T)\n\nThe set containing the single point value subseteq mathbbR.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.EqualTo(2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.EqualTo{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Interval","page":"Standard form","title":"MathOptInterface.Interval","text":"Interval{T<:Real}(lower::T, upper::T)\n\nThe interval lower upper subseteq mathbbR cup -infty +infty.\n\nIf lower or upper is -Inf or Inf, respectively, the set is interpreted as a one-sided interval.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Interval(1.0, 2.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Integer","page":"Standard form","title":"MathOptInterface.Integer","text":"Integer()\n\nThe set of integers, mathbbZ.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Integer())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ZeroOne","page":"Standard form","title":"MathOptInterface.ZeroOne","text":"ZeroOne()\n\nThe set 0 1.\n\nVariables belonging to the ZeroOne set are also known as \"binary\" variables.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.ZeroOne())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Semicontinuous","page":"Standard form","title":"MathOptInterface.Semicontinuous","text":"Semicontinuous{T<:Real}(lower::T, upper::T)\n\nThe set 0 cup lower upper.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Semicontinuous(2.0, 3.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Semicontinuous{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Semiinteger","page":"Standard form","title":"MathOptInterface.Semiinteger","text":"Semiinteger{T<:Real}(lower::T, upper::T)\n\nThe set 0 cup lower lower+1 ldots upper-1 upper.\n\nNote that if lower and upper are not equivalent to an integer, then the solver may throw an error, or it may round up lower and round down upper to the nearest integers.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.add_constraint(model, x, MOI.Semiinteger(2.0, 3.0))\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Semiinteger{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Parameter","page":"Standard form","title":"MathOptInterface.Parameter","text":"Parameter{T<:Number}(value::T)\n\nThe set containing the single point value subseteq mathbbR.\n\nThe Parameter set is conceptually similar to the EqualTo set, except that a variable constrained to the Parameter set cannot have other constraints added to it, and the Parameter set can never be deleted. Thus, solvers are free to treat the variable as a constant, and they need not add it as a decision variable to the model.\n\nBecause of this behavior, you must add parameters using add_constrained_variable, and solvers should declare supports_add_constrained_variable and not supports_constraint for the Parameter set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> p, ci = MOI.add_constrained_variable(model, MOI.Parameter(2.5))\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Parameter{Float64}}(1))\n\njulia> MOI.set(model, MOI.ConstraintSet(), ci, MOI.Parameter(3.0))\n\njulia> MOI.get(model, MOI.ConstraintSet(), ci)\nMathOptInterface.Parameter{Float64}(3.0)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Vector-sets","page":"Standard form","title":"Vector sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized vector sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"Reals\nZeros\nNonnegatives\nNonpositives\nNormInfinityCone\nNormOneCone\nNormCone\nSecondOrderCone\nRotatedSecondOrderCone\nGeometricMeanCone\nExponentialCone\nDualExponentialCone\nPowerCone\nDualPowerCone\nRelativeEntropyCone\nNormSpectralCone\nNormNuclearCone\nSOS1\nSOS2\nIndicator\nComplements\nHyperRectangle\nScaled","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.Reals","page":"Standard form","title":"MathOptInterface.Reals","text":"Reals(dimension::Int)\n\nThe set mathbbR^dimension (containing all points) of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Reals(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Reals}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Zeros","page":"Standard form","title":"MathOptInterface.Zeros","text":"Zeros(dimension::Int)\n\nThe set  0 ^dimension (containing only the origin) of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Zeros(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Zeros}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Nonnegatives","page":"Standard form","title":"MathOptInterface.Nonnegatives","text":"Nonnegatives(dimension::Int)\n\nThe nonnegative orthant  x in mathbbR^dimension  x ge 0  of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonnegatives(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Nonpositives","page":"Standard form","title":"MathOptInterface.Nonpositives","text":"Nonpositives(dimension::Int)\n\nThe nonpositive orthant  x in mathbbR^dimension  x le 0  of non-negative dimension dimension.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Nonpositives(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonpositives}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormInfinityCone","page":"Standard form","title":"MathOptInterface.NormInfinityCone","text":"NormInfinityCone(dimension::Int)\n\nThe ell_infty-norm cone  (tx) in mathbbR^dimension  t ge lVert x rVert_infty = max_i lvert x_i rvert  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormInfinityCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormInfinityCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormOneCone","page":"Standard form","title":"MathOptInterface.NormOneCone","text":"NormOneCone(dimension::Int)\n\nThe ell_1-norm cone  (tx) in mathbbR^dimension  t ge lVert x rVert_1 = sum_i lvert x_i rvert  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormOneCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormOneCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormCone","page":"Standard form","title":"MathOptInterface.NormCone","text":"NormCone(p::Float64, dimension::Int)\n\nThe ell_p-norm cone  (tx) in mathbbR^dimension  t ge left(sumlimits_i x_i^pright)^frac1p  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.NormCone(3, 4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SecondOrderCone","page":"Standard form","title":"MathOptInterface.SecondOrderCone","text":"SecondOrderCone(dimension::Int)\n\nThe second-order cone (or Lorenz cone or ell_2-norm cone)  (tx) in mathbbR^dimension  t ge lVert x rVert_2  of dimension dimension.\n\nThe dimension must be at least 1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([t; x]), MOI.SecondOrderCone(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SecondOrderCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RotatedSecondOrderCone","page":"Standard form","title":"MathOptInterface.RotatedSecondOrderCone","text":"RotatedSecondOrderCone(dimension::Int)\n\nThe rotated second-order cone  (tux) in mathbbR^dimension  2tu ge lVert x rVert_2^2 tu ge 0  of dimension dimension.\n\nThe dimension must be at least 2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> u = MOI.add_variable(model)\nMOI.VariableIndex(2)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; u; x]),\n           MOI.RotatedSecondOrderCone(5),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RotatedSecondOrderCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.GeometricMeanCone","page":"Standard form","title":"MathOptInterface.GeometricMeanCone","text":"GeometricMeanCone(dimension::Int)\n\nThe geometric mean cone  (tx) in mathbbR^n+1  x ge 0 t le sqrtnx_1 x_2 cdots x_n , where dimension = n + 1 >= 2.\n\nDuality note\n\nThe dual of the geometric mean cone is  (u v) in mathbbR^n+1  u le 0 v ge 0 -u le n sqrtnprod_i v_i , where dimension = n + 1 >= 2.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; x]),\n           MOI.GeometricMeanCone(4),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.GeometricMeanCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.ExponentialCone","page":"Standard form","title":"MathOptInterface.ExponentialCone","text":"ExponentialCone()\n\nThe 3-dimensional exponential cone  (xyz) in mathbbR^3  y exp (xy) le z y  0 .\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.ExponentialCone())\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.ExponentialCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.DualExponentialCone","page":"Standard form","title":"MathOptInterface.DualExponentialCone","text":"DualExponentialCone()\n\nThe 3-dimensional dual exponential cone  (uvw) in mathbbR^3  -u exp (vu) le exp(1) w u  0 .\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.DualExponentialCone())\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.DualExponentialCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.PowerCone","page":"Standard form","title":"MathOptInterface.PowerCone","text":"PowerCone{T<:Real}(exponent::T)\n\nThe 3-dimensional power cone  (xyz) in mathbbR^3  x^exponent y^1-exponent ge z x ge 0 y ge 0  with parameter exponent.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.PowerCone(0.5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.PowerCone{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.DualPowerCone","page":"Standard form","title":"MathOptInterface.DualPowerCone","text":"DualPowerCone{T<:Real}(exponent::T)\n\nThe 3-dimensional power cone  (uvw) in mathbbR^3  (fracuexponent)^exponent (fracv1-exponent)^1-exponent ge w u ge 0 v ge 0  with parameter exponent.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.DualPowerCone(0.5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.DualPowerCone{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RelativeEntropyCone","page":"Standard form","title":"MathOptInterface.RelativeEntropyCone","text":"RelativeEntropyCone(dimension::Int)\n\nThe relative entropy cone  (u v w) in mathbbR^1+2n  u ge sum_i=1^n w_i log(fracw_iv_i) v_i ge 0 w_i ge 0 , where dimension = 2n + 1 >= 1.\n\nDuality note\n\nThe dual of the relative entropy cone is  (u v w) in mathbbR^1+2n  forall i w_i ge u (log (fracuv_i) - 1) v_i ge 0 u  0  of dimension dimension=2n+1.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> u = MOI.add_variable(model);\n\njulia> v = MOI.add_variables(model, 3);\n\njulia> w = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([u; v; w]),\n           MOI.RelativeEntropyCone(7),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RelativeEntropyCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormSpectralCone","page":"Standard form","title":"MathOptInterface.NormSpectralCone","text":"NormSpectralCone(row_dim::Int, column_dim::Int)\n\nThe epigraph of the matrix spectral norm (maximum singular value function)  (t X) in mathbbR^1 + row_dim times column_dim  t ge sigma_1(X) , where sigma_i is the ith singular value of the matrix X of non-negative row dimension row_dim and column dimension column_dim.\n\nThe matrix X is vectorized by stacking the columns, matching the behavior of Julia's vec function.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 6), 2, 3)\n2×3 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)  MOI.VariableIndex(6)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)  MOI.VariableIndex(7)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.NormSpectralCone(2, 3),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormSpectralCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.NormNuclearCone","page":"Standard form","title":"MathOptInterface.NormNuclearCone","text":"NormNuclearCone(row_dim::Int, column_dim::Int)\n\nThe epigraph of the matrix nuclear norm (sum of singular values function)  (t X) in mathbbR^1 + row_dim times column_dim  t ge sum_i sigma_i(X) , where sigma_i is the ith singular value of the matrix X of non-negative row dimension row_dim and column dimension column_dim.\n\nThe matrix X is vectorized by stacking the columns, matching the behavior of Julia's vec function.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 6), 2, 3)\n2×3 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)  MOI.VariableIndex(6)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)  MOI.VariableIndex(7)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.NormNuclearCone(2, 3),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.NormNuclearCone}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SOS1","page":"Standard form","title":"MathOptInterface.SOS1","text":"SOS1{T<:Real}(weights::Vector{T})\n\nThe set corresponding to the Special Ordered Set (SOS) constraint of Type I.\n\nOf the variables in the set, at most one can be nonzero.\n\nThe weights induce an ordering of the variables such that the kth element in the set corresponds to the kth weight in weights. Solvers may use these weights to improve the efficiency of the solution process, but the ordering does not change the set of feasible solutions.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.SOS1([1.0, 3.0, 2.5]),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SOS1{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.SOS2","page":"Standard form","title":"MathOptInterface.SOS2","text":"SOS2{T<:Real}(weights::Vector{T})\n\nThe set corresponding to the Special Ordered Set (SOS) constraint of Type II.\n\nThe weights induce an ordering of the variables such that the kth element in the set corresponds to the kth weight in weights. Therefore, the weights must be unique.\n\nOf the variables in the set, at most two can be nonzero, and if two are nonzero, they must be adjacent in the ordering of the set.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.SOS2([1.0, 3.0, 2.5]),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.SOS2{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Indicator","page":"Standard form","title":"MathOptInterface.Indicator","text":"Indicator{A<:ActivationCondition,S<:AbstractScalarSet}(set::S)\n\nThe set corresponding to an indicator constraint.\n\nWhen A is ACTIVATE_ON_ZERO, this means: (y x) in 0 1 times mathbbR^n  y = 0 implies x in set\n\nWhen A is ACTIVATE_ON_ONE, this means: (y x) in 0 1 times mathbbR^n  y = 1 implies x in set\n\nNotes\n\nMost solvers expect that the first row of the function is interpretable as a variable index x_i (e.g., 1.0 * x + 0.0). An error will be thrown if this is not the case.\n\nExample\n\nThe constraint (y x) in 0 1 times mathbbR^2  y = 1 implies x_1 + x_2 leq 9  is defined as\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 2)\n2-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n\njulia> y, _ = MOI.add_constrained_variable(model, MOI.ZeroOne())\n(MOI.VariableIndex(3), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(3))\n\njulia> f = MOI.VectorAffineFunction(\n           [\n               MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, y)),\n               MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(1.0, x[1])),\n               MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(1.0, x[2])),\n           ],\n           [0.0, 0.0],\n       )\n┌                                                         ┐\n│0.0 + 1.0 MOI.VariableIndex(3)                           │\n│0.0 + 1.0 MOI.VariableIndex(1) + 1.0 MOI.VariableIndex(2)│\n└                                                         ┘\n\njulia> s = MOI.Indicator{MOI.ACTIVATE_ON_ONE}(MOI.LessThan(9.0))\nMathOptInterface.Indicator{MathOptInterface.ACTIVATE_ON_ONE, MathOptInterface.LessThan{Float64}}(MathOptInterface.LessThan{Float64}(9.0))\n\njulia> MOI.add_constraint(model, f, s)\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Indicator{MathOptInterface.ACTIVATE_ON_ONE, MathOptInterface.LessThan{Float64}}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Complements","page":"Standard form","title":"MathOptInterface.Complements","text":"Complements(dimension::Base.Integer)\n\nThe set corresponding to a mixed complementarity constraint.\n\nComplementarity constraints should be specified with an AbstractVectorFunction-in-Complements(dimension) constraint.\n\nThe dimension of the vector-valued function F must be dimension. This defines a complementarity constraint between the scalar function F[i] and the variable in F[i + dimension/2]. Thus, F[i + dimension/2] must be interpretable as a single variable x_i (e.g., 1.0 * x + 0.0), and dimension must be even.\n\nThe mixed complementarity problem consists of finding x_i in the interval [lb, ub] (i.e., in the set Interval(lb, ub)), such that the following holds:\n\nF_i(x) == 0 if lb_i < x_i < ub_i\nF_i(x) >= 0 if lb_i == x_i\nF_i(x) <= 0 if x_i == ub_i\n\nClassically, the bounding set for x_i is Interval(0, Inf), which recovers: 0 <= F_i(x) ⟂ x_i >= 0, where the ⟂ operator implies F_i(x) * x_i = 0.\n\nExample\n\nThe problem:\n\nx -in- Interval(-1, 1)\n[-4 * x - 3, x] -in- Complements(2)\n\ndefines the mixed complementarity problem where the following holds:\n\n-4 * x - 3 == 0 if -1 < x < 1\n-4 * x - 3 >= 0 if x == -1\n-4 * x - 3 <= 0 if x == 1\n\nThere are three solutions:\n\nx = -3/4 with F(x) = 0\nx = -1 with F(x) = 1\nx = 1 with F(x) = -7\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x, _ = MOI.add_constrained_variable(model, MOI.Interval(-1.0, 1.0));\n\njulia> MOI.add_constraint(\n            model,\n            MOI.Utilities.vectorize([-4.0 * x - 3.0, x]),\n            MOI.Complements(2),\n        )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Complements}(1)\n\nThe function F can also be defined in terms of single variables. For example, the problem:\n\n[x_3, x_4] -in- Nonnegatives(2)\n[x_1, x_2, x_3, x_4] -in- Complements(4)\n\ndefines the complementarity problem where 0 <= x_1 ⟂ x_3 >= 0 and 0 <= x_2 ⟂ x_4 >= 0.\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 4);\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x[3:4]), MOI.Nonnegatives(2))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.add_constraint(\n            model,\n            MOI.VectorOfVariables(x),\n            MOI.Complements(4),\n        )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Complements}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.HyperRectangle","page":"Standard form","title":"MathOptInterface.HyperRectangle","text":"HyperRectangle(lower::Vector{T}, upper::Vector{T}) where {T}\n\nThe set x in barmathbbR^d x_i in lower_i upper_i forall i=1ldotsd.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3)\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables(x),\n           MOI.HyperRectangle(zeros(3), ones(3)),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.HyperRectangle{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Scaled","page":"Standard form","title":"MathOptInterface.Scaled","text":"struct Scaled{S<:AbstractVectorSet} <: AbstractVectorSet\n    set::S\nend\n\nGiven a vector a in mathbbR^d and a set representing the set mathcalS in mathbbR^d such that Utilities.set_dot for x in mathcalS and y in mathcalS^* is\n\nsum_i=1^d a_i x_i y_i\n\nthe set Scaled(set) is defined as\n\n (sqrta_1 x_1 sqrta_2 x_2 ldots sqrta_d x_d)  x in S \n\nExample\n\nThis can be used to scale a vector of numbers\n\njulia> import MathOptInterface as MOI\n\njulia> set = MOI.PositiveSemidefiniteConeTriangle(2)\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\njulia> a = MOI.Utilities.SetDotScalingVector{Float64}(set)\n3-element MathOptInterface.Utilities.SetDotScalingVector{Float64, MathOptInterface.PositiveSemidefiniteConeTriangle}:\n 1.0\n 1.4142135623730951\n 1.0\n\njulia> using LinearAlgebra\n\njulia> MOI.Utilities.operate(*, Float64, Diagonal(a), ones(3))\n3-element Vector{Float64}:\n 1.0\n 1.4142135623730951\n 1.0\n\nIt can be also used to scale a vector of function\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3);\n\njulia> func = MOI.VectorOfVariables(x)\n┌                    ┐\n│MOI.VariableIndex(1)│\n│MOI.VariableIndex(2)│\n│MOI.VariableIndex(3)│\n└                    ┘\n\njulia> set = MOI.PositiveSemidefiniteConeTriangle(2)\nMathOptInterface.PositiveSemidefiniteConeTriangle(2)\n\njulia> MOI.Utilities.operate(*, Float64, Diagonal(a), func)\n┌                                             ┐\n│0.0 + 1.0 MOI.VariableIndex(1)               │\n│0.0 + 1.4142135623730951 MOI.VariableIndex(2)│\n│0.0 + 1.0 MOI.VariableIndex(3)               │\n└                                             ┘\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Constraint-programming-sets","page":"Standard form","title":"Constraint programming sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AllDifferent\nBinPacking\nCircuit\nCountAtLeast\nCountBelongs\nCountDistinct\nCountGreaterThan\nCumulative\nPath\nReified\nTable","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AllDifferent","page":"Standard form","title":"MathOptInterface.AllDifferent","text":"AllDifferent(dimension::Int)\n\nThe set x in mathbbZ^d such that no two elements in x take the same value and dimension = d.\n\nAlso known as\n\nThis constraint is called all_different in MiniZinc, and is sometimes also called distinct.\n\nExample\n\nTo enforce x[1] != x[2] AND x[1] != x[3] AND x[2] != x[3]:\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.AllDifferent(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.AllDifferent}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.BinPacking","page":"Standard form","title":"MathOptInterface.BinPacking","text":"BinPacking(c::T, w::Vector{T}) where {T}\n\nThe set x in mathbbZ^d where d = length(w), such that each item i in 1:d of weight w[i] is put into bin x[i], and the total weight of each bin does not exceed c.\n\nThere are additional assumptions that the capacity, c, and the weights, w, must all be non-negative.\n\nThe bin numbers depend on the bounds of x, so they may be something other than the integers 1:d.\n\nAlso known as\n\nThis constraint is called bin_packing in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> bins = MOI.add_variables(model, 5)\n5-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n\njulia> weights = Float64[1, 1, 2, 2, 3]\n5-element Vector{Float64}:\n 1.0\n 1.0\n 2.0\n 2.0\n 3.0\n\njulia> MOI.add_constraint.(model, bins, MOI.Integer())\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(5)\n\njulia> MOI.add_constraint.(model, bins, MOI.Interval(4.0, 6.0))\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(1)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(2)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}}(5)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(bins), MOI.BinPacking(3.0, weights))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.BinPacking{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Circuit","page":"Standard form","title":"MathOptInterface.Circuit","text":"Circuit(dimension::Int)\n\nThe set x in 1d^d that constraints x to be a circuit, such that x_i = j means that j is the successor of i, and dimension = d.\n\nGraphs with multiple independent circuits, such as [2, 1, 3] and [2, 1, 4, 3], are not valid.\n\nAlso known as\n\nThis constraint is called circuit in MiniZinc, and it is equivalent to forming a (potentially sub-optimal) tour in the travelling salesperson problem.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Circuit(3))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Circuit}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountAtLeast","page":"Standard form","title":"MathOptInterface.CountAtLeast","text":"CountAtLeast(n::Int, d::Vector{Int}, set::Set{Int})\n\nThe set x in mathbbZ^d_1 + d_2 + ldots d_N, where x is partitioned into N subsets (x_1  ldots x_d_1, x_d_1 + 1  ldots x_d_1 + d_2 and so on), and at least n elements of each subset take one of the values in set.\n\nAlso known as\n\nThis constraint is called at_least in MiniZinc.\n\nExample\n\nTo ensure that 3 appears at least once in each of the subsets {a, b} and {b, c}:\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> a, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> b, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> c, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(3), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(3))\n\njulia> x, d, set = [a, b, b, c], [2, 2], [3]\n(MathOptInterface.VariableIndex[MOI.VariableIndex(1), MOI.VariableIndex(2), MOI.VariableIndex(2), MOI.VariableIndex(3)], [2, 2], [3])\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.CountAtLeast(1, d, Set(set)))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountAtLeast}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountBelongs","page":"Standard form","title":"MathOptInterface.CountBelongs","text":"CountBelongs(dimenson::Int, set::Set{Int})\n\nThe set (n x) in mathbbZ^1+d, such that n elements of the vector x take on of the values in set and dimension = 1 + d.\n\nAlso known as\n\nThis constraint is called among by MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> n, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n\njulia> set = Set([3, 4, 5])\nSet{Int64} with 3 elements:\n  5\n  4\n  3\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([n; x]), MOI.CountBelongs(4, set))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountBelongs}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountDistinct","page":"Standard form","title":"MathOptInterface.CountDistinct","text":"CountDistinct(dimension::Int)\n\nThe set (n x) in mathbbZ^1+d, such that the number of distinct values in x is n and dimension = 1 + d.\n\nAlso known as\n\nThis constraint is called nvalues in MiniZinc.\n\nExample\n\nTo model:\n\nif n == 1`, thenx[1] == x[2] == x[3]`\nif n == 2, then\nx[1] == x[2] != x[3] or\nx[1] != x[2] == x[3] or\nx[1] == x[3] != x[2]\nif n == 3, then x[1] != x[2], x[2] != x[3] and x[3] != x[1]\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> n, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(vcat(n, x)), MOI.CountDistinct(4))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountDistinct}(1)\n\nRelationship to AllDifferent\n\nWhen the first element is d, CountDistinct is equivalent to an AllDifferent constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.CountGreaterThan","page":"Standard form","title":"MathOptInterface.CountGreaterThan","text":"CountGreaterThan(dimension::Int)\n\nThe set (c y x) in mathbbZ^1+1+d, such that c is strictly greater than the number of occurances of y in x and dimension = 1 + 1 + d.\n\nAlso known as\n\nThis constraint is called count_gt in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> c, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> y, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> x = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([c; y; x]), MOI.CountGreaterThan(5))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.CountGreaterThan}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Cumulative","page":"Standard form","title":"MathOptInterface.Cumulative","text":"Cumulative(dimension::Int)\n\nThe set (s d r b) in mathbbZ^3n+1, representing the cumulative global constraint, where n == length(s) == length(r) == length(b) and dimension = 3n + 1.\n\nCumulative requires that a set of tasks given by start times s, durations d, and resource requirements r, never requires more than the global resource bound b at any one time.\n\nAlso known as\n\nThis constraint is called cumulative in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> s = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> d = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n MOI.VariableIndex(6)\n\njulia> r = [MOI.add_constrained_variable(model, MOI.Integer())[1] for _ in 1:3]\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(7)\n MOI.VariableIndex(8)\n MOI.VariableIndex(9)\n\njulia> b, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(10), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(10))\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([s; d; r; b]), MOI.Cumulative(10))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Cumulative}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Path","page":"Standard form","title":"MathOptInterface.Path","text":"Path(from::Vector{Int}, to::Vector{Int})\n\nGiven a graph comprised of a set of nodes 1..N and a set of arcs 1..E represented by an edge from node from[i] to node to[i], Path constrains the set (s t ns es) in (1N)times(1E)times01^Ntimes01^E, to form subgraph that is a path from node s to node t, where node n is in the path if ns[n] is 1, and edge e is in the path if es[e] is 1.\n\nThe path must be acyclic, and it must traverse all nodes n for which ns[n] is 1, and all edges e for which es[e] is 1.\n\nAlso known as\n\nThis constraint is called path in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> N, E = 4, 5\n(4, 5)\n\njulia> from = [1, 1, 2, 2, 3]\n5-element Vector{Int64}:\n 1\n 1\n 2\n 2\n 3\n\njulia> to = [2, 3, 3, 4, 4]\n5-element Vector{Int64}:\n 2\n 3\n 3\n 4\n 4\n\njulia> s, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(1))\n\njulia> t, _ = MOI.add_constrained_variable(model, MOI.Integer())\n(MOI.VariableIndex(2), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.Integer}(2))\n\njulia> ns = MOI.add_variables(model, N)\n4-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(3)\n MOI.VariableIndex(4)\n MOI.VariableIndex(5)\n MOI.VariableIndex(6)\n\njulia> MOI.add_constraint.(model, ns, MOI.ZeroOne())\n4-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(3)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(4)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(5)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(6)\n\njulia> es = MOI.add_variables(model, E)\n5-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(7)\n MOI.VariableIndex(8)\n MOI.VariableIndex(9)\n MOI.VariableIndex(10)\n MOI.VariableIndex(11)\n\njulia> MOI.add_constraint.(model, es, MOI.ZeroOne())\n5-element Vector{MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}}:\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(7)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(8)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(9)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(10)\n MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(11)\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables([s; t; ns; es]), MOI.Path(from, to))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Path}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Reified","page":"Standard form","title":"MathOptInterface.Reified","text":"Reified(set::AbstractSet)\n\nThe constraint z f(x) in Reified(S) ensures that f(x) in S if and only if z == 1, where z in 0 1.\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}\n\njulia> z, _ = MOI.add_constrained_variable(model, MOI.ZeroOne())\n(MOI.VariableIndex(1), MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1))\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(2)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.Utilities.vectorize([z, 2.0 * x]),\n           MOI.Reified(MOI.GreaterThan(1.0)),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Reified{MathOptInterface.GreaterThan{Float64}}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.Table","page":"Standard form","title":"MathOptInterface.Table","text":"Table(table::Matrix{T}) where {T}\n\nThe set x in mathbbR^d where d = size(table, 2), such that x belongs to one row of table. That is, there exists some j in 1:size(table, 1), such that x[i] = table[j, i] for all i=1:size(table, 2).\n\nAlso known as\n\nThis constraint is called table in MiniZinc.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> x = MOI.add_variables(model, 3)\n3-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n MOI.VariableIndex(2)\n MOI.VariableIndex(3)\n\njulia> table = Float64[1 1 0; 0 1 1; 1 0 1; 1 1 1]\n4×3 Matrix{Float64}:\n 1.0  1.0  0.0\n 0.0  1.0  1.0\n 1.0  0.0  1.0\n 1.0  1.0  1.0\n\njulia> MOI.add_constraint(model, MOI.VectorOfVariables(x), MOI.Table(table))\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.Table{Float64}}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#Matrix-sets","page":"Standard form","title":"Matrix sets","text":"","category":"section"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"Matrix sets are vectorized to be subtypes of AbstractVectorSet.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"For sets of symmetric matrices, storing both the (i, j) and (j, i) elements is redundant. Use the AbstractSymmetricMatrixSetTriangle set to represent only the vectorization of the upper triangular part of the matrix.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"When the matrix of expressions constrained to be in the set is not symmetric, and hence additional constraints are needed to force the equality of the (i, j) and (j, i) elements, use the AbstractSymmetricMatrixSetSquare set.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"The Bridges.Constraint.SquareBridge can transform a set from the square form to the triangular_form by adding appropriate constraints if the (i, j) and (j, i) expressions are different.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"AbstractSymmetricMatrixSetTriangle\nAbstractSymmetricMatrixSetSquare\nside_dimension\ntriangular_form","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSymmetricMatrixSetTriangle","page":"Standard form","title":"MathOptInterface.AbstractSymmetricMatrixSetTriangle","text":"abstract type AbstractSymmetricMatrixSetTriangle <: AbstractVectorSet end\n\nAbstract supertype for subsets of the (vectorized) cone of symmetric matrices, with side_dimension rows and columns. The entries of the upper-right triangular part of the matrix are given column by column (or equivalently, the entries of the lower-left triangular part are given row by row). A vectorized cone of dimension n corresponds to a square matrix with side dimension sqrt14 + 2 n - 12. (Because a d times d matrix has d(d + 1)  2 elements in the upper or lower triangle.)\n\nExample\n\nThe matrix\n\nbeginbmatrix\n  1  2  4\n  2  3  5\n  4  5  6\nendbmatrix\n\nhas side_dimension 3 and vectorization (1 2 3 4 5 6).\n\nNote\n\nTwo packed storage formats exist for symmetric matrices, the respective orders of the entries are:\n\nupper triangular column by column (or lower triangular row by row);\nlower triangular column by column (or upper triangular row by row).\n\nThe advantage of the first format is the mapping between the (i, j) matrix indices and the k index of the vectorized form. It is simpler and does not depend on the side dimension of the matrix. Indeed,\n\nthe entry of matrix indices (i, j) has vectorized index k = div((j - 1) * j, 2) + i if i leq j and k = div((i - 1) * i, 2) + j if j leq i;\nand the entry with vectorized index k has matrix indices i = div(1 + isqrt(8k - 7), 2) and j = k - div((i - 1) * i, 2) or j = div(1 + isqrt(8k - 7), 2) and i = k - div((j - 1) * j, 2).\n\nDuality note\n\nThe scalar product for the symmetric matrix in its vectorized form is the sum of the pairwise product of the diagonal entries plus twice the sum of the pairwise product of the upper diagonal entries; see [p. 634, 1]. This has important consequence for duality.\n\nConsider for example the following problem (PositiveSemidefiniteConeTriangle is a subtype of AbstractSymmetricMatrixSetTriangle)\n\nbeginalign*\n     max_x in mathbbR  x\n    \n     textst \n    (1 -x 1)  in textPositiveSemidefiniteConeTriangle(2)\nendalign*\n\nThe dual is the following problem\n\nbeginalign*\n     min_x in mathbbR^3  y_1 + y_3\n    \n     textst  2y_2  = 1\n      y  in textPositiveSemidefiniteConeTriangle(2)\nendalign*\n\nWhy do we use 2y_2 in the dual constraint instead of y_2 ? The reason is that 2y_2 is the scalar product between y and the symmetric matrix whose vectorized form is (0 1 0). Indeed, with our modified scalar products we have\n\nlangle\n(0 1 0)\n(y_1 y_2 y_3)\nrangle\n=\nmathrmtrace\nbeginpmatrix\n  0  1\n  1  0\nendpmatrix\nbeginpmatrix\n  y_1  y_2\n  y_2  y_3\nendpmatrix\n= 2y_2\n\nReferences\n\n[1] Boyd, S. and Vandenberghe, L.. Convex optimization. Cambridge university     press, 2004.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.AbstractSymmetricMatrixSetSquare","page":"Standard form","title":"MathOptInterface.AbstractSymmetricMatrixSetSquare","text":"abstract type AbstractSymmetricMatrixSetSquare <: AbstractVectorSet end\n\nAbstract supertype for subsets of the (vectorized) cone of symmetric matrices, with side_dimension rows and columns. The entries of the matrix are given column by column (or equivalently, row by row). The matrix is both constrained to be symmetric and to have its triangular_form belong to the corresponding set. That is, if the functions in entries (i j) and (j i) are different, then a constraint will be added to make sure that the entries are equal.\n\nExample\n\nPositiveSemidefiniteConeSquare is a subtype of AbstractSymmetricMatrixSetSquare and constraining the matrix\n\nbeginbmatrix\n  1  -y\n  -z  0\nendbmatrix\n\nto be symmetric positive semidefinite can be achieved by constraining the vector (1 -z -y 0) (or (1 -y -z 0)) to belong to the PositiveSemidefiniteConeSquare(2). It both constrains y = z and (1 -y 0) (or (1 -z 0)) to be in PositiveSemidefiniteConeTriangle(2), since triangular_form(PositiveSemidefiniteConeSquare) is PositiveSemidefiniteConeTriangle.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.side_dimension","page":"Standard form","title":"MathOptInterface.side_dimension","text":"side_dimension(\n    set::Union{\n        AbstractSymmetricMatrixSetTriangle,\n        AbstractSymmetricMatrixSetSquare,\n        HermitianPositiveSemidefiniteConeTriangle,\n    },\n)\n\nSide dimension of the matrices in set.\n\nConvention\n\nBy convention, the side dimension should be stored in the side_dimension field. If this is not the case for a subtype of AbstractSymmetricMatrixSetTriangle, or AbstractSymmetricMatrixSetSquare you must implement this method.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/#MathOptInterface.triangular_form","page":"Standard form","title":"MathOptInterface.triangular_form","text":"triangular_form(S::Type{<:AbstractSymmetricMatrixSetSquare})\ntriangular_form(set::AbstractSymmetricMatrixSetSquare)\n\nReturn the AbstractSymmetricMatrixSetTriangle corresponding to the vectorization of the upper triangular part of matrices in the AbstractSymmetricMatrixSetSquare set.\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"List of recognized matrix sets.","category":"page"},{"location":"moi/reference/standard_form/","page":"Standard form","title":"Standard form","text":"PositiveSemidefiniteConeTriangle\nPositiveSemidefiniteConeSquare\nHermitianPositiveSemidefiniteConeTriangle\nLogDetConeTriangle\nLogDetConeSquare\nRootDetConeTriangle\nRootDetConeSquare","category":"page"},{"location":"moi/reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle","page":"Standard form","title":"MathOptInterface.PositiveSemidefiniteConeTriangle","text":"PositiveSemidefiniteConeTriangle(side_dimension::Int) <: AbstractSymmetricMatrixSetTriangle\n\nThe (vectorized) cone of symmetric positive semidefinite matrices, with non-negative side_dimension rows and columns.\n\nSee AbstractSymmetricMatrixSetTriangle for more details on the vectorized form.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeSquare","page":"Standard form","title":"MathOptInterface.PositiveSemidefiniteConeSquare","text":"PositiveSemidefiniteConeSquare(side_dimension::Int) <: AbstractSymmetricMatrixSetSquare\n\nThe cone of symmetric positive semidefinite matrices, with non-negative side length side_dimension.\n\nSee AbstractSymmetricMatrixSetSquare for more details on the vectorized form.\n\nThe entries of the matrix are given column by column (or equivalently, row by row).\n\nThe matrix is both constrained to be symmetric and to be positive semidefinite. That is, if the functions in entries (i j) and (j i) are different, then a constraint will be added to make sure that the entries are equal.\n\nExample\n\nConstraining the matrix\n\nbeginbmatrix\n  1  -y\n  -z  0\nendbmatrix\n\nto be symmetric positive semidefinite can be achieved by constraining the vector (1 -z -y 0) (or (1 -y -z 0)) to belong to the PositiveSemidefiniteConeSquare(2).\n\nIt both constrains y = z and (1 -y 0) (or (1 -z 0)) to be in PositiveSemidefiniteConeTriangle(2).\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.HermitianPositiveSemidefiniteConeTriangle","page":"Standard form","title":"MathOptInterface.HermitianPositiveSemidefiniteConeTriangle","text":"HermitianPositiveSemidefiniteConeTriangle(side_dimension::Int) <: AbstractVectorSet\n\nThe (vectorized) cone of Hermitian positive semidefinite matrices, with non-negative side_dimension rows and columns.\n\nBecaue the matrix is Hermitian, the diagonal elements are real, and the complex-valued lower triangular entries are obtained as the conjugate of corresponding upper triangular entries.\n\nVectorization format\n\nThe vectorized form starts with real part of the entries of the upper triangular part of the matrix, given column by column as explained in AbstractSymmetricMatrixSetSquare.\n\nIt is then followed by the imaginary part of the off-diagonal entries of the upper triangular part, also given column by column.\n\nFor example, the matrix\n\nbeginbmatrix\n  1  2 + 7im  4 + 8im\n  2 - 7im  3  5 + 9im\n  4 - 8im  5 - 9im  6\nendbmatrix\n\nhas side_dimension 3 and is represented as the vector 1 2 3 4 5 6 7 8 9.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LogDetConeTriangle","page":"Standard form","title":"MathOptInterface.LogDetConeTriangle","text":"LogDetConeTriangle(side_dimension::Int)\n\nThe log-determinant cone  (t u X) in mathbbR^2 + d(d+1)2  t le u log(det(Xu)) u  0 , where the matrix X is represented in the same symmetric packed format as in the PositiveSemidefiniteConeTriangle.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; X]),\n           MOI.LogDetConeTriangle(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.LogDetConeTriangle}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.LogDetConeSquare","page":"Standard form","title":"MathOptInterface.LogDetConeSquare","text":"LogDetConeSquare(side_dimension::Int)\n\nThe log-determinant cone  (t u X) in mathbbR^2 + d^2  t le u log(det(Xu)) X text symmetric u  0 , where the matrix X is represented in the same format as in the PositiveSemidefiniteConeSquare.\n\nSimilarly to PositiveSemidefiniteConeSquare, constraints are added to ensure that X is symmetric.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 4), 2, 2)\n2×2 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.LogDetConeSquare(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.LogDetConeSquare}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RootDetConeTriangle","page":"Standard form","title":"MathOptInterface.RootDetConeTriangle","text":"RootDetConeTriangle(side_dimension::Int)\n\nThe root-determinant cone  (t X) in mathbbR^1 + d(d+1)2  t le det(X)^1d , where the matrix X is represented in the same symmetric packed format as in the PositiveSemidefiniteConeTriangle.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = MOI.add_variables(model, 3);\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; X]),\n           MOI.RootDetConeTriangle(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeTriangle}(1)\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/standard_form/#MathOptInterface.RootDetConeSquare","page":"Standard form","title":"MathOptInterface.RootDetConeSquare","text":"RootDetConeSquare(side_dimension::Int)\n\nThe root-determinant cone  (t X) in mathbbR^1 + d^2  t le det(X)^1d X text symmetric , where the matrix X is represented in the same format as PositiveSemidefiniteConeSquare.\n\nSimilarly to PositiveSemidefiniteConeSquare, constraints are added to ensure that X is symmetric.\n\nThe non-negative argument side_dimension is the side dimension of the matrix X, i.e., its number of rows or columns.\n\nExample\n\njulia> import MathOptInterface as MOI\n\njulia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> t = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> X = reshape(MOI.add_variables(model, 4), 2, 2)\n2×2 Matrix{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(2)  MOI.VariableIndex(4)\n MOI.VariableIndex(3)  MOI.VariableIndex(5)\n\njulia> MOI.add_constraint(\n           model,\n           MOI.VectorOfVariables([t; vec(X)]),\n           MOI.RootDetConeSquare(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorOfVariables, MathOptInterface.RootDetConeSquare}(1)\n\n\n\n\n\n","category":"type"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"EditURL = \"https://github.com/darnstrom/DAQP.jl/blob/v0.5.1/README.md\"","category":"page"},{"location":"packages/DAQP/#DAQP.jl","page":"darnstrom/DAQP.jl","title":"DAQP.jl","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"(Image: ) (Image: )","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"DAQP.jl is a Julia wrapper for the Quadratic Programming solver DAQP.","category":"page"},{"location":"packages/DAQP/#License","page":"darnstrom/DAQP.jl","title":"License","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"DAQP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"The underlying solver, darnstrom/daqp is licensed under the MIT license.","category":"page"},{"location":"packages/DAQP/#Installation","page":"darnstrom/DAQP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"Install DAQP.jl using the Julia package manager:","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"import Pkg\nPkg.add(\"DAQP\")","category":"page"},{"location":"packages/DAQP/#Use-with-JuMP","page":"darnstrom/DAQP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"To use DAQP with JuMP, do:","category":"page"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"using JuMP, DAQP\nmodel = Model(DAQP.Optimizer)","category":"page"},{"location":"packages/DAQP/#Documentation","page":"darnstrom/DAQP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/DAQP/","page":"darnstrom/DAQP.jl","title":"darnstrom/DAQP.jl","text":"General information about the solver is available at https://darnstrom.github.io/daqp/, and specifics for the Julia interface are available at https://darnstrom.github.io/daqp/start/julia.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"EditURL = \"portfolio.jl\"","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Portfolio-optimization","page":"Portfolio optimization","title":"Portfolio optimization","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Optimization models play an increasingly important role in financial decisions. Many computational finance problems can be solved efficiently using modern optimization techniques.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial solves the famous Markowitz Portfolio Optimization problem with data from lecture notes from a course taught at Georgia Tech by Shabbir Ahmed.","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Required-packages","page":"Portfolio optimization","title":"Required packages","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"This tutorial uses the following packages","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"using JuMP\nimport DataFrames\nimport Ipopt\nimport MultiObjectiveAlgorithms as MOA\nimport Plots\nimport Statistics\nimport StatsPlots","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Formulation","page":"Portfolio optimization","title":"Formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Suppose we are considering investing 1000 dollars in three non-dividend paying stocks, IBM (IBM), Walmart (WMT), and Southern Electric (SEHI), for a one-month period.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We will use the initial money to buy shares of the three stocks at the current market prices, hold these for one month, and sell the shares off at the prevailing market prices at the end of the month.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"As a rational investor, we hope to make some profit out of this endeavor, that is, the return on our investment should be positive.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Suppose we bought a stock at p dollars per share in the beginning of the month, and sold it off at s dollars per share at the end of the month. Then the one-month return on a share of the stock is fracs-pp.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Since the stock prices are quite uncertain, so is the end-of-month return on our investment. Our goal is to invest in such a way that the expected end-of-month return is at least $50 or 5%. Furthermore, we want to make sure that the “risk” of not achieving our desired return is minimum.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Note that we are solving the problem under the following assumptions:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We can trade any continuum of shares.\nNo short-selling is allowed.\nThere are no transaction costs.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We model this problem by taking decision variables x_i i=123 denoting the dollars invested in each of the 3 stocks.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Let us denote by tilder_i the random variable corresponding to the monthly return (increase in the stock price) per dollar for stock i.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Then, the return (or profit) on x_i dollars invested in stock i is tilder_i x_i and the total (random) return on our investment is sum_i=1^3 tilder_i x_i The expected return on our investment is then mathbbEleftsum_i=1^3 tilder_i x_iright=sum_i=1^3 overliner_i x_i where overliner_i is the expected value of the tilder_i","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Now we need to quantify the notion of “risk” in our investment.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Markowitz, in his Nobel prize winning work, showed that a rational investor’s notion of minimizing risk can be closely approximated by minimizing the variance of the return of the investment portfolio. This variance is given by:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"operatornameVarleftsum_i=1^3 tilder_i x_iright = sum_i=1^3 sum_j=1^3 x_i x_j sigma_i j","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"where sigma_i j is the covariance of the return of stock i with stock j.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Note that the right hand side of the equation is the most reduced form of the expression and we have not shown the intermediate steps involved in getting to this form. We can also write this equation as:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"operatornameVarleftsum_i=1^3 tilder_i x_iright =x^top Q x","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Where Q is the covariance matrix for the random vector tilder.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Finally, we can write the model as:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"beginaligned\nmin x^top Q x \ntext  st  sum_i=1^3 x_i leq 1000 \noverliner^top x geq 50 \nx geq 0\nendaligned","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Data","page":"Portfolio optimization","title":"Data","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"For the data in our problem, we use the stock prices given below, in monthly values from November 2000, through November 2001.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"df = DataFrames.DataFrame(\n    [\n        93.043 51.826 1.063\n        84.585 52.823 0.938\n        111.453 56.477 1.000\n        99.525 49.805 0.938\n        95.819 50.287 1.438\n        114.708 51.521 1.700\n        111.515 51.531 2.540\n        113.211 48.664 2.390\n        104.942 55.744 3.120\n        99.827 47.916 2.980\n        91.607 49.438 1.900\n        107.937 51.336 1.750\n        115.590 55.081 1.800\n    ],\n    [:IBM, :WMT, :SEHI],\n)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Next, we compute the percentage return for the stock in each month:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"returns = diff(Matrix(df); dims = 1) ./ Matrix(df[1:end-1, :])","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The expected monthly return is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"r = vec(Statistics.mean(returns; dims = 1))","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"and the covariance matrix is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Q = Statistics.cov(returns)","category":"page"},{"location":"tutorials/nonlinear/portfolio/#JuMP-formulation","page":"Portfolio optimization","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, x[1:3] >= 0)\n@objective(model, Min, x' * Q * x)\n@constraint(model, sum(x) <= 1000)\n@constraint(model, r' * x >= 50)\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The optimal allocation of our assets is:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"value.(x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"So we spend $497 on IBM, and $503 on SEHI. This results in a variance of:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"scalar_variance = value(x' * Q * x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"and an expected return of:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"scalar_return = value(r' * x)","category":"page"},{"location":"tutorials/nonlinear/portfolio/#Multi-objective-portfolio-optimization","page":"Portfolio optimization","title":"Multi-objective portfolio optimization","text":"","category":"section"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The previous model returned a single solution that minimized the variance, ensuring that our expected return was at least $50. In practice, we might be willing to accept a slightly higher variance if it meant a much increased expected return. To explore this problem space, we can instead formulate our portfolio optimization problem with two objectives:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"to minimize the variance\nto maximize the expected return","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The solution to this bi-objective problem is the efficient frontier of modern portfolio theory, and each point in the solution is a point with the best return for a fixed level of risk.","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"model = Model(() -> MOA.Optimizer(Ipopt.Optimizer))\nset_silent(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"We also need to choose a solution algorithm for MOA. For our problem, the efficient frontier will have an infinite number of solutions. Since we cannot find all of the solutions, we choose an approximation algorithm and limit the number of solution points that are returned:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"set_optimizer_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\nset_optimizer_attribute(model, MOA.SolutionLimit(), 50)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Now we can define the rest of the model:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) <= 1000)\n@expression(model, variance, x' * Q * x)\n@expression(model, expected_return, r' * x)\n# We want to minimize variance and maximize expected return, but we must pick\n# a single objective sense `Min`, and negate any `Max` objectives:\n@objective(model, Min, [variance, -expected_return])\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"The algorithm found 50 different solutions. Let's plot them to see how they differ:","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"objective_space = Plots.hline(\n    [scalar_return];\n    label = \"Single-objective solution\",\n    linecolor = :red,\n)\nPlots.vline!(objective_space, [scalar_variance]; label = \"\", linecolor = :red)\nPlots.scatter!(\n    objective_space,\n    [value(variance; result = i) for i in 1:result_count(model)],\n    [value(expected_return; result = i) for i in 1:result_count(model)];\n    xlabel = \"Variance\",\n    ylabel = \"Expected Return\",\n    label = \"\",\n    title = \"Objective space\",\n    markercolor = \"white\",\n    markersize = 5,\n    legend = :bottomright,\n)\nfor i in 1:result_count(model)\n    y = objective_value(model; result = i)\n    Plots.annotate!(objective_space, y[1], -y[2], (i, 3))\nend\n\ndecision_space = StatsPlots.groupedbar(\n    vcat([value.(x; result = i)' for i in 1:result_count(model)]...);\n    bar_position = :stack,\n    label = [\"IBM\" \"WMT\" \"SEHI\"],\n    xlabel = \"Solution #\",\n    ylabel = \"Investment (\\$)\",\n    title = \"Decision space\",\n)\nPlots.plot(objective_space, decision_space; layout = (2, 1), size = (600, 600))","category":"page"},{"location":"tutorials/nonlinear/portfolio/","page":"Portfolio optimization","title":"Portfolio optimization","text":"Perhaps our trade-off wasn't so bad after all. Our original solution corresponded to picking a solution #17. If we buy more SEHI, we can increase the return, but the variance also increases. If we buy less SEHI, such as a solution like #5 or #6, then we can achieve the corresponding return without deploying all of our capital. We should also note that at no point should we buy WMT.","category":"page"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"EditURL = \"https://github.com/joaquimg/BilevelJuMP.jl/blob/v0.6.2/README.md\"","category":"page"},{"location":"packages/BilevelJuMP/#BilevelJuMP.jl","page":"joaquimg/BilevelJuMP.jl","title":"BilevelJuMP.jl","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"(Image: Build Status) (Image: Codecov branch) (Image: Docs dev) (Image: Docs stable)","category":"page"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"BilevelJuMP.jl is a JuMP extension for modeling and solving bilevel optimization problems.","category":"page"},{"location":"packages/BilevelJuMP/#License","page":"joaquimg/BilevelJuMP.jl","title":"License","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"BilevelJuMP.jl is licensed under the MIT license.","category":"page"},{"location":"packages/BilevelJuMP/#Documentation","page":"joaquimg/BilevelJuMP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"You can find the documentation at https://joaquimg.github.io/BilevelJuMP.jl/stable/.","category":"page"},{"location":"packages/BilevelJuMP/#Help","page":"joaquimg/BilevelJuMP.jl","title":"Help","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"If you need help, please open a GitHub issue.","category":"page"},{"location":"packages/BilevelJuMP/#Example","page":"joaquimg/BilevelJuMP.jl","title":"Example","text":"","category":"section"},{"location":"packages/BilevelJuMP/#Install","page":"joaquimg/BilevelJuMP.jl","title":"Install","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"import Pkg\nPkg.add(\"BilevelJuMP\")\nPkg.add(\"HiGHS\")","category":"page"},{"location":"packages/BilevelJuMP/#Run","page":"joaquimg/BilevelJuMP.jl","title":"Run","text":"","category":"section"},{"location":"packages/BilevelJuMP/","page":"joaquimg/BilevelJuMP.jl","title":"joaquimg/BilevelJuMP.jl","text":"using BilevelJuMP, HiGHS\n\nmodel = BilevelModel(\n    HiGHS.Optimizer,\n    mode = BilevelJuMP.FortunyAmatMcCarlMode(primal_big_M = 100, dual_big_M = 100)\n)\n\n@variable(Lower(model), x)\n@variable(Upper(model), y)\n\n@objective(Upper(model), Min, 3x + y)\n@constraints(Upper(model), begin\n    x <= 5\n    y <= 8\n    y >= 0\nend)\n\n@objective(Lower(model), Min, -x)\n@constraints(Lower(model), begin\n     x +  y <= 8\n    4x +  y >= 8\n    2x +  y <= 13\n    2x - 7y <= 0\nend)\n\noptimize!(model)\n\nobjective_value(model) # = 3 * (3.5 * 8/15) + 8/15 # = 6.13...\nvalue(x) # = 3.5 * 8/15 # = 1.86...\nvalue(y) # = 8/15 # = 0.53...","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"EditURL = \"https://github.com/oxfordcontrol/Clarabel.jl/blob/v0.6.0/README.md\"","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<h1 align=\"center\" margin=0px>\n  <img src=\"https://github.com/oxfordcontrol/ClarabelDocs/blob/main/docs/src/assets/logo-banner-light-jl.png?raw=true\" class=\"display-light-only\" width=60%>\n  <img src=\"https://github.com/oxfordcontrol/ClarabelDocs/blob/main/docs/src/assets/logo-banner-dark-jl.png?raw=true\" class=\"display-dark-only\"   width=60%>\n  <br>\nInterior Point Conic Optimization for Julia\n</h1>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<p align=\"center\">\n   <a href=\"https://github.com/oxfordcontrol/Clarabel.jl/actions\"><img src=\"https://github.com/oxfordcontrol/Clarabel.jl/workflows/ci/badge.svg?branch=main\"></a>\n  <a href=\"https://codecov.io/gh/oxfordcontrol/Clarabel.jl\"><img src=\"https://codecov.io/gh/oxfordcontrol/Clarabel.jl/branch/main/graph/badge.svg?raw=true\"></a>\n  <a href=\"https://oxfordcontrol.github.io/ClarabelDocs/stable\"><img src=\"https://img.shields.io/badge/Documentation-stable-purple.svg?raw=true\"></a>\n  <a href=\"https://opensource.org/licenses/Apache-2.0\"><img src=\"https://img.shields.io/badge/License-Apache%202.0-blue.svg?raw=true\"></a>\n  <a href=\"https://github.com/oxfordcontrol/Clarabel.jl/releases\"><img src=\"https://img.shields.io/badge/Release-v0.6.0-blue.svg?raw=true\"></a>\n</p>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"<p align=\"center\">\n  <a href=\"#features\">Features</a> •\n  <a href=\"#installation\">Installation</a> •\n  <a href=\"#license-\">License</a> •\n  <a href=\"https://oxfordcontrol.github.io/ClarabelDocs/stable\">Documentation</a>\n</p>","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel.jl is a Julia implementation of an interior point numerical solver for convex optimization problems using a novel homogeneous embedding.  Clarabel.jl solves the following problem:","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"beginarrayr\ntextminimize  frac12x^T P x + q^T x2ex\ntextsubject to  Ax + s = b 1ex\n         s in mathcalK\nendarray","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"with decision variables x in mathbbR^n, s in mathbbR^m and data matrices P=P^top succeq 0, q in mathbbR^n, A in mathbbR^m times n, and b in mathbbR^m. The convex set mathcalK is a composition of convex cones.","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"For more information see the Clarabel Documentation (stable |  dev).","category":"page"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel is also available in a Rust implementation with additional language interfaces.  See here.","category":"page"},{"location":"packages/Clarabel/#Features","page":"oxfordcontrol/Clarabel.jl","title":"Features","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Versatile: Clarabel.jl solves linear programs (LPs), quadratic programs (QPs), second-order cone programs (SOCPs) and semidefinite programs (SDPs). It also solves problems with exponential, power cone and generalized power cone constraints.\nQuadratic objectives: Unlike interior point solvers based on the standard homogeneous self-dual embedding (HSDE), Clarabel.jl handles quadratic objectives without requiring any epigraphical reformulation of the objective.   It can therefore be significantly faster than other HSDE-based solvers for problems with quadratic objective functions.\nInfeasibility detection: Infeasible problems are detected using a homogeneous embedding technique.\nJuMP / Convex.jl support: We provide an interface to MathOptInterface (MOI), which allows you to describe your problem in JuMP and Convex.jl.\nArbitrary precision types: You can solve problems with any floating point precision, for example, Float32 or Julia's BigFloat type, using either the native interface, or via MathOptInterface / Convex.jl.\nOpen Source: Our code is available on GitHub and distributed under the Apache 2.0 License","category":"page"},{"location":"packages/Clarabel/#Installation","page":"oxfordcontrol/Clarabel.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"Clarabel.jl can be added via the Julia package manager (type ]): pkg> add Clarabel","category":"page"},{"location":"packages/Clarabel/#License","page":"oxfordcontrol/Clarabel.jl","title":"License 🔍","text":"","category":"section"},{"location":"packages/Clarabel/","page":"oxfordcontrol/Clarabel.jl","title":"oxfordcontrol/Clarabel.jl","text":"This project is licensed under the Apache License  2.0 - see the LICENSE.md file for details.","category":"page"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/reference/modification.md\"","category":"page"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/reference/modification/#Modifications","page":"Modifications","title":"Modifications","text":"","category":"section"},{"location":"moi/reference/modification/","page":"Modifications","title":"Modifications","text":"modify\nAbstractFunctionModification\nScalarConstantChange\nVectorConstantChange\nScalarCoefficientChange\nMultirowChange","category":"page"},{"location":"moi/reference/modification/#MathOptInterface.modify","page":"Modifications","title":"MathOptInterface.modify","text":"Constraint Function\n\nmodify(model::ModelLike, ci::ConstraintIndex, change::AbstractFunctionModification)\n\nApply the modification specified by change to the function of constraint ci.\n\nAn ModifyConstraintNotAllowed error is thrown if modifying constraints is not supported by the model model.\n\nExamples\n\nmodify(model, ci, ScalarConstantChange(10.0))\n\nObjective Function\n\nmodify(model::ModelLike, ::ObjectiveFunction, change::AbstractFunctionModification)\n\nApply the modification specified by change to the objective function of model. To change the function completely, call set instead.\n\nAn ModifyObjectiveNotAllowed error is thrown if modifying objectives is not supported by the model model.\n\nExamples\n\nmodify(model, ObjectiveFunction{ScalarAffineFunction{Float64}}(), ScalarConstantChange(10.0))\n\nMultiple modifications in Constraint Functions\n\nmodify(\n    model::ModelLike,\n    cis::AbstractVector{<:ConstraintIndex},\n    changes::AbstractVector{<:AbstractFunctionModification},\n)\n\nApply multiple modifications specified by changes to the functions of constraints cis.\n\nA ModifyConstraintNotAllowed error is thrown if modifying constraints is not supported by model.\n\nExamples\n\nmodify(\n    model,\n    [ci, ci],\n    [\n        ScalarCoefficientChange{Float64}(VariableIndex(1), 1.0),\n        ScalarCoefficientChange{Float64}(VariableIndex(2), 0.5),\n    ],\n)\n\nMultiple modifications in the Objective Function\n\nmodify(\n    model::ModelLike,\n    attr::ObjectiveFunction,\n    changes::AbstractVector{<:AbstractFunctionModification},\n)\n\nApply multiple modifications specified by changes to the functions of constraints cis.\n\nA ModifyObjectiveNotAllowed error is thrown if modifying objective coefficients is not supported by model.\n\nExamples\n\nmodify(\n    model,\n    ObjectiveFunction{ScalarAffineFunction{Float64}}(),\n    [\n        ScalarCoefficientChange{Float64}(VariableIndex(1), 1.0),\n        ScalarCoefficientChange{Float64}(VariableIndex(2), 0.5),\n    ],\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/reference/modification/#MathOptInterface.AbstractFunctionModification","page":"Modifications","title":"MathOptInterface.AbstractFunctionModification","text":"AbstractFunctionModification\n\nAn abstract supertype for structs which specify partial modifications to functions, to be used for making small modifications instead of replacing the functions entirely.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.ScalarConstantChange","page":"Modifications","title":"MathOptInterface.ScalarConstantChange","text":"ScalarConstantChange{T}(new_constant::T)\n\nA struct used to request a change in the constant term of a scalar-valued function.\n\nApplicable to ScalarAffineFunction and ScalarQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.VectorConstantChange","page":"Modifications","title":"MathOptInterface.VectorConstantChange","text":"VectorConstantChange{T}(new_constant::Vector{T})\n\nA struct used to request a change in the constant vector of a vector-valued function.\n\nApplicable to VectorAffineFunction and VectorQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.ScalarCoefficientChange","page":"Modifications","title":"MathOptInterface.ScalarCoefficientChange","text":"ScalarCoefficientChange{T}(variable::VariableIndex, new_coefficient::T)\n\nA struct used to request a change in the linear coefficient of a single variable in a scalar-valued function.\n\nApplicable to ScalarAffineFunction and ScalarQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"moi/reference/modification/#MathOptInterface.MultirowChange","page":"Modifications","title":"MathOptInterface.MultirowChange","text":"MultirowChange{T}(\n    variable::VariableIndex,\n    new_coefficients::Vector{Tuple{Int64,T}},\n) where {T}\n\nA struct used to request a change in the linear coefficients of a single variable in a vector-valued function.\n\nNew coefficients are specified by (output_index, coefficient) tuples.\n\nApplicable to VectorAffineFunction and VectorQuadraticFunction.\n\n\n\n\n\n","category":"type"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"EditURL = \"https://github.com/mariohsouto/ProxSDP.jl/blob/de648be90e63a69b8d8e4f82371760065d242fe9/README.md\"","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"<img src=\"https://raw.githubusercontent.com/mariohsouto/ProxSDP_aux/19f261a912e29c77cb68ee2957a5b41128824f22/logo_proxSDP.png?raw=true\" alt=\"logo\"/>","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"(Image: ) (Image: ) (Image: )","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is an open-source semidefinite programming (SDP) solver based on the paper \"Exploiting Low-Rank Structure in Semidefinite Programming by Approximate Operator Splitting\".","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The main advantage of ProxSDP over other state-of-the-art solvers is the ability to exploit the low-rank structure inherent to several SDP problems.","category":"page"},{"location":"packages/ProxSDP/#Overview-of-problems-ProxSDP-can-solve","page":"mariohsouto/ProxSDP.jl","title":"Overview of problems ProxSDP can solve","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"General conic convex optimization problems with the presence of the positive semidefinite cone, second-order cone and positive orthant;\nSemidefinite relaxation of nonconvex problems, for example, max-cut, binary MIMO, optimal power flow, sensor localization, sum-of-squares;\nControl theory problems with LMI constraints;\nNuclear norm minimization problems, for example, matrix completion;","category":"page"},{"location":"packages/ProxSDP/#License","page":"mariohsouto/ProxSDP.jl","title":"License","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is licensed under the MIT License.","category":"page"},{"location":"packages/ProxSDP/#Installation","page":"mariohsouto/ProxSDP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Install ProxSDP using Julia package manager:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"import Pkg\nPkg.add(\"ProxSDP\")","category":"page"},{"location":"packages/ProxSDP/#Documentation","page":"mariohsouto/ProxSDP.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The online documentation is available at https://mariohsouto.github.io/ProxSDP.jl/latest/.","category":"page"},{"location":"packages/ProxSDP/#Usage","page":"mariohsouto/ProxSDP.jl","title":"Usage","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Consider the semidefinite programming relaxation of the max-cut problem, as in:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"max   0.25 * W•X\ns.t.  diag(X) = 1,\n      X ≽ 0,","category":"page"},{"location":"packages/ProxSDP/#JuMP","page":"mariohsouto/ProxSDP.jl","title":"JuMP","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"This problem can be solved by the following code using ProxSDP and JuMP.","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"# Load packages\nusing ProxSDP, JuMP, LinearAlgebra\n# Number of vertices\nn = 4\n# Graph weights\nW = [\n    18.0  -5.0  -7.0  -6.0\n    -5.0   6.0   0.0  -1.0\n    -7.0   0.0   8.0  -1.0\n    -6.0  -1.0  -1.0   8.0\n]\n# Build Max-Cut SDP relaxation via JuMP\nmodel = Model(ProxSDP.Optimizer)\nset_optimizer_attribute(model, \"log_verbose\", true)\nset_optimizer_attribute(model, \"tol_gap\", 1e-4)\nset_optimizer_attribute(model, \"tol_feasibility\", 1e-4)\n@variable(model, X[1:n, 1:n], PSD)\n@objective(model, Max, 0.25 * LinearAlgebra.dot(W, X))\n@constraint(model, LinearAlgebra.diag(X) .== 1)\n# Solve optimization problem with ProxSDP\noptimize!(model)\n# Retrieve solution\nXsol = value.(X)","category":"page"},{"location":"packages/ProxSDP/#Convex.jl","page":"mariohsouto/ProxSDP.jl","title":"Convex.jl","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Another alternative is to use ProxSDP via Convex.jl:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"# Load packages\nusing Convex, ProxSDP\n# Number of vertices\nn = 4\n# Graph weights\nW = [\n    18.0  -5.0  -7.0  -6.0\n    -5.0   6.0   0.0  -1.0\n    -7.0   0.0   8.0  -1.0\n    -6.0  -1.0  -1.0   8.0\n]\n# Define optimization problem\nX = Semidefinite(n)\nproblem = maximize(0.25 * dot(W, X), diag(X) == 1)\n# Solve optimization problem with ProxSDP\nsolve!(problem, ProxSDP.Optimizer(log_verbose=true, tol_gap=1e-4, tol_feasibility=1e-4))\n# Get the objective value\nproblem.optval\n# Retrieve solution\nevaluate(X)","category":"page"},{"location":"packages/ProxSDP/#Citing-this-package","page":"mariohsouto/ProxSDP.jl","title":"Citing this package","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The published version of the paper can be found here and the arXiv version here.","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"We kindly request that you cite the paper as:","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"@article{souto2020exploiting,\n  author = {Mario Souto and Joaquim D. Garcia and \\'Alvaro Veiga},\n  title = {Exploiting low-rank structure in semidefinite programming by approximate operator splitting},\n  journal = {Optimization},\n  pages = {1-28},\n  year  = {2020},\n  publisher = {Taylor & Francis},\n  doi = {10.1080/02331934.2020.1823387},\n  URL = {https://doi.org/10.1080/02331934.2020.1823387}\n}","category":"page"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"The preprint version of the paper can be found here.","category":"page"},{"location":"packages/ProxSDP/#Disclaimer","page":"mariohsouto/ProxSDP.jl","title":"Disclaimer","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"ProxSDP is a research software, therefore it should not be used in production.\nPlease open an issue if you find any problems, developers will try to fix and find alternatives.\nThere is no continuous development for 32-bit systems, the package should work, but might reach some issues.\nProxSDP assumes primal and dual feasibility.","category":"page"},{"location":"packages/ProxSDP/#ROAD-MAP","page":"mariohsouto/ProxSDP.jl","title":"ROAD MAP","text":"","category":"section"},{"location":"packages/ProxSDP/","page":"mariohsouto/ProxSDP.jl","title":"mariohsouto/ProxSDP.jl","text":"Support for exponential and power cones\nWarm start","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"EditURL = \"https://github.com/jump-dev/HiGHS.jl/blob/v1.7.4/README.md\"","category":"page"},{"location":"packages/HiGHS/#HiGHS.jl","page":"jump-dev/HiGHS.jl","title":"HiGHS.jl","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"HiGHS.jl is a wrapper for the HiGHS solver.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"It has two components:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/HiGHS/#Affiliation","page":"jump-dev/HiGHS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"This wrapper is maintained by the JuMP community and is not an official project of the HiGHS developers.","category":"page"},{"location":"packages/HiGHS/#License","page":"jump-dev/HiGHS.jl","title":"License","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"HiGHS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The underlying solver, ERGO-Code/HiGHS, is licensed under the MIT license.","category":"page"},{"location":"packages/HiGHS/#Installation","page":"jump-dev/HiGHS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"Install HiGHS as follows:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"import Pkg\nPkg.add(\"HiGHS\")","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"In addition to installing the HiGHS.jl package, this will also download and install the HiGHS binaries. You do not need to install HiGHS separately.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/HiGHS/#Use-with-JuMP","page":"jump-dev/HiGHS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"To use HiGHS with JuMP, use HiGHS.Optimizer:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"using JuMP, HiGHS\nmodel = Model(HiGHS.Optimizer)\nset_attribute(model, \"presolve\", \"on\")\nset_attribute(model, \"time_limit\", 60.0)","category":"page"},{"location":"packages/HiGHS/#MathOptInterface-API","page":"jump-dev/HiGHS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The HiGHS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"MOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/HiGHS/#Options","page":"jump-dev/HiGHS.jl","title":"Options","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"See the HiGHS documentation for a full list of the available options.","category":"page"},{"location":"packages/HiGHS/#C-API","page":"jump-dev/HiGHS.jl","title":"C API","text":"","category":"section"},{"location":"packages/HiGHS/","page":"jump-dev/HiGHS.jl","title":"jump-dev/HiGHS.jl","text":"The C API can be accessed via HiGHS.Highs_xxx functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/FileFormats/reference.md\"","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/FileFormats/reference/#File-Formats","page":"API Reference","title":"File Formats","text":"","category":"section"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"Functions to help read and write MOI models to/from various file formats. See The FileFormats submodule for more details.","category":"page"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"FileFormats.Model\nFileFormats.FileFormat\nFileFormats.CBF.Model\nFileFormats.LP.Model\nFileFormats.MOF.Model\nFileFormats.MPS.Model\nFileFormats.NL.Model\nFileFormats.SDPA.Model","category":"page"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.Model","page":"API Reference","title":"MathOptInterface.FileFormats.Model","text":"Model(\n    ;\n    format::FileFormat = FORMAT_AUTOMATIC,\n    filename::Union{Nothing, String} = nothing,\n    kwargs...\n)\n\nReturn model corresponding to the FileFormat format, or, if format == FORMAT_AUTOMATIC, guess the format from filename.\n\nThe filename argument is only needed if format == FORMAT_AUTOMATIC.\n\nkwargs are passed to the underlying model constructor.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.FileFormat","page":"API Reference","title":"MathOptInterface.FileFormats.FileFormat","text":"FileFormat\n\nList of accepted export formats.\n\nFORMAT_AUTOMATIC: try to detect the file format based on the file name\nFORMAT_CBF: the Conic Benchmark format\nFORMAT_LP: the LP file format\nFORMAT_MOF: the MathOptFormat file format\nFORMAT_MPS: the MPS file format\nFORMAT_NL: the AMPL .nl file format\nFORMAT_REW: the .rew file format, which is MPS with generic names\nFORMAT_SDPA: the SemiDefinite Programming Algorithm format\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.CBF.Model","page":"API Reference","title":"MathOptInterface.FileFormats.CBF.Model","text":"Model()\n\nCreate an empty instance of FileFormats.CBF.Model.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.LP.Model","page":"API Reference","title":"MathOptInterface.FileFormats.LP.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.LP.Model.\n\nKeyword arguments are:\n\nmaximum_length::Int=255: the maximum length for the name of a variable. lp_solve 5.0 allows only 16 characters, while CPLEX 12.5+ allow 255.\nwarn::Bool=false: print a warning when variables or constraints are renamed.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.MOF.Model","page":"API Reference","title":"MathOptInterface.FileFormats.MOF.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.MOF.Model.\n\nKeyword arguments are:\n\nprint_compact::Bool=false: print the JSON file in a compact format without spaces or newlines.\nwarn::Bool=false: print a warning when variables or constraints are renamed\ndifferentiation_backend::MOI.Nonlinear.AbstractAutomaticDifferentiation = MOI.Nonlinear.SparseReverseMode(): automatic differentiation backend to use when reading models with nonlinear constraints and objectives.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.MPS.Model","page":"API Reference","title":"MathOptInterface.FileFormats.MPS.Model","text":"Model(; kwargs...)\n\nCreate an empty instance of FileFormats.MPS.Model.\n\nKeyword arguments are:\n\nwarn::Bool=false: print a warning when variables or constraints are renamed.\nprint_objsense::Bool=false: print the OBJSENSE section when writing\ngeneric_names::Bool=false: strip all names in the model and replace them with the generic names C$i and R$i for the i'th column and row respectively.\nquadratic_format::QuadraticFormat = kQuadraticFormatGurobi: specify the solver-specific extension used when writing the quadratic components of the model. Options are kQuadraticFormatGurobi, kQuadraticFormatCPLEX, and kQuadraticFormatMosek.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.NL.Model","page":"API Reference","title":"MathOptInterface.FileFormats.NL.Model","text":"Model(; use_nlp_block::Bool = true)\n\nCreate a new Optimizer object.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.SDPA.Model","page":"API Reference","title":"MathOptInterface.FileFormats.SDPA.Model","text":"Model(; number_type::Type = Float64)\n\nCreate an empty instance of FileFormats.SDPA.Model{number_type}.\n\nIt is important to be aware that the SDPA file format is interpreted in geometric form and not standard conic form. The standard conic form and geometric conic form are two dual standard forms for semidefinite programs (SDPs). The geometric conic form of an SDP is as follows:\n\nbeginalign\n min_y in mathbbR^m  b^T y\n\n textst  sum_i=1^m A_i y_i - C  in mathbbK\nendalign\n\nwhere mathcalK is a cartesian product of nonnegative orthant and positive semidefinite matrices that align with a block diagonal structure shared with the matrices A_i and C.\n\nIn other words, the geometric conic form contains free variables and affine constraints in either the nonnegative orthant or the positive semidefinite cone. That is, in the MathOptInterface's terminology, MOI.VectorAffineFunction-in-MOI.Nonnegatives and MOI.VectorAffineFunction-in-MOI.PositiveSemidefiniteConeTriangle constraints.\n\nThe corresponding standard conic form of the dual SDP is as follows:\n\nbeginalign\n max_X in mathbbK  texttr(CX)\n\n textst  texttr(A_iX)  = b_i  i = 1 ldots m\nendalign\n\nIn other words, the standard conic form contains nonnegative and positive semidefinite variables with equality constraints. That is, in the MathOptInterface's terminology, MOI.VectorOfVariables-in-MOI.Nonnegatives, MOI.VectorOfVariables-in-MOI.PositiveSemidefiniteConeTriangle and MOI.ScalarAffineFunction-in-MOI.EqualTo constraints.\n\nIf a model is in standard conic form, use Dualization.jl to transform it into the geometric conic form before writting it. Otherwise, the nonnegative (resp. positive semidefinite) variables will be bridged into free variables with affine constraints constraining them to belong to the nonnegative orthant (resp. positive semidefinite cone) by the MOI.Bridges.Constraint.VectorFunctionizeBridge. Moreover, equality constraints will be bridged into pairs of affine constraints in the nonnegative orthant by the MOI.Bridges.Constraint.SplitIntervalBridge and then the MOI.Bridges.Constraint.VectorizeBridge.\n\nIf a solver is in standard conic form, use Dualization.jl to transform the model read into standard conic form before copying it to the solver. Otherwise, the free variables will be bridged into pairs of variables in the nonnegative orthant by the MOI.Bridges.Variable.FreeBridge and affine constraints will be bridged into equality constraints by creating a slack variable by the MOI.Bridges.Constraint.VectorSlackBridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/FileFormats/reference/#Other-helpers","page":"API Reference","title":"Other helpers","text":"","category":"section"},{"location":"moi/submodules/FileFormats/reference/","page":"API Reference","title":"API Reference","text":"FileFormats.NL.SolFileResults","category":"page"},{"location":"moi/submodules/FileFormats/reference/#MathOptInterface.FileFormats.NL.SolFileResults","page":"API Reference","title":"MathOptInterface.FileFormats.NL.SolFileResults","text":"SolFileResults(filename::String, model::Model)\n\nParse the .sol file filename created by solving model and return a SolFileResults struct.\n\nThe returned struct supports the MOI.get API for querying result attributes such as MOI.TerminationStatus, MOI.VariablePrimal, and MOI.ConstraintDual.\n\n\n\n\n\nSolFileResults(\n    raw_status::String,\n    termination_status::MOI.TerminationStatusCode,\n)\n\nReturn a SolFileResults struct with MOI.RawStatusString set to raw_status, MOI.TerminationStatus set to termination_status, and MOI.PrimalStatus and MOI.DualStatus set to NO_SOLUTION.\n\nAll other attributes are un-set.\n\n\n\n\n\n","category":"type"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"EditURL = \"https://github.com/COPT-Public/COPT.jl/blob/v1.1.11/README.md\"","category":"page"},{"location":"packages/COPT/#COPT.jl","page":"COPT-Public/COPT.jl","title":"COPT.jl","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT.jl is a wrapper for the COPT (Cardinal Optimizer), a mathematical optimization solver for large-scale optimization problems.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT includes high-performance solvers for LP, MIP, SOCP, convex QP/QCP and SDP.","category":"page"},{"location":"packages/COPT/#License","page":"COPT-Public/COPT.jl","title":"License","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"COPT.jl is licensed under the MIT license.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/COPT/#Note","page":"COPT-Public/COPT.jl","title":"Note","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"When COPT is upgraded to a newer version, you may see an error message such as ERROR: COPT error 4: Unable to create COPT environment, which indicates that you will need to reapply and upgrade your COPT license files as well.","category":"page"},{"location":"packages/COPT/#Installation","page":"COPT-Public/COPT.jl","title":"Installation","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"Install COPT using the Julia package manager","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"import Pkg\nPkg.add(\"COPT\")","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"When there is no local version of COPT installed, installing COPT.jl will automatically download the necessary solver binaries.","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"Without a license, you can solve small models for non-commercial purpose. We strongly recommend that you apply for a license by following the link above.","category":"page"},{"location":"packages/COPT/#Note-2","page":"COPT-Public/COPT.jl","title":"Note","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"MacOS Apple M1/ARM: on MacOS with Apple M1 chips, Intel based programs can run via Rosetta. When you installed the COPT binaries manually, then please make sure that the COPT build matches the Julia build. We recommend the Intel based COPT and Julia build, as the Apple M1/ARM build of Julia is experimental.","category":"page"},{"location":"packages/COPT/#Use-with-JuMP","page":"COPT-Public/COPT.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"To use COPT with JuMP, use COPT.Optimizer:","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"using JuMP\nusing COPT\nmodel = Model(COPT.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\nprint(model)\noptimize!(model)\n@show termination_status(model)\n@show primal_status(model)\n@show dual_status(model)\n@show objective_value(model)\n@show value(x)\n@show value(y)\n@show shadow_price(c1)\n@show shadow_price(c2)","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"To use the semidefinite programming solver in COPT with JuMP, use COPT.ConeOptimizer:","category":"page"},{"location":"packages/COPT/","page":"COPT-Public/COPT.jl","title":"COPT-Public/COPT.jl","text":"using JuMP\nusing COPT\nusing LinearAlgebra\nmodel = Model(COPT.ConeOptimizer)\nC = [1.0 -1.0; -1.0 2.0]\n@variable(model, X[1:2, 1:2], PSD)\n@variable(model, z[1:2] >= 0)\n@objective(model, Min, C ⋅ X)\n@constraint(model, c1, X[1, 1] - z[1] == 1)\n@constraint(model, c2, X[2, 2] - z[2] == 1)\noptimize!(model)\n@show termination_status(model)\n@show primal_status(model)\n@show dual_status(model)\n@show objective_value(model)\n@show value.(X)\n@show value.(z)\n@show shadow_price(c1)\n@show shadow_price(c2)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"EditURL = \"cannery.jl\"","category":"page"},{"location":"tutorials/linear/cannery/#The-cannery-problem","page":"The cannery problem","title":"The cannery problem","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial was originally contributed by Louis Luangkesorn.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial solves the cannery problem from Dantzig, Linear Programming and Extensions, Princeton University Press, Princeton, NJ, 1963. This class of problem is known as a transshipment problem.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"The purpose of this tutorial is to demonstrate how to use JSON data in the formulation of a JuMP model.","category":"page"},{"location":"tutorials/linear/cannery/#Required-packages","page":"The cannery problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"using JuMP\nimport HiGHS\nimport JSON","category":"page"},{"location":"tutorials/linear/cannery/#Formulation","page":"The cannery problem","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"The cannery problem assumes we are optimizing the shipment of cases of cans from production plants p in P to markets m in M.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Each production plant p has a capacity c_p, and each market m has a demand d_m. The shipping cost per case of cans from plant p to market m is d_pm.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We wish to find the distribution plan x_pm, the number of cases of cans to ship from plant p to market m, for p in P and m in M that minimizes the shipping costs. We can formulate our problem as the following linear program:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"beginaligned\nmin  sumlimits_p in Psumlimits_m in M d_pm x_pm \ntextst  sumlimits_m in M x_pm le c_p  forall p in P \n             sumlimits_p in P x_pm ge d_m  forall m in M \n             x_pm ge 0  forall p in P m in M\nendaligned","category":"page"},{"location":"tutorials/linear/cannery/#Data","page":"The cannery problem","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"A key feature of the tutorial is to demonstrate how to load data from JSON.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"For simplicity, we've hard-coded it below. But if the data was available as a .json file, we could use data = JSON.parsefile(filename) to read in the data.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"data = JSON.parse(\"\"\"\n{\n    \"plants\": {\n        \"Seattle\": {\"capacity\": 350},\n        \"San-Diego\": {\"capacity\": 600}\n    },\n    \"markets\": {\n        \"New-York\": {\"demand\": 300},\n        \"Chicago\": {\"demand\": 300},\n        \"Topeka\": {\"demand\": 300}\n    },\n    \"distances\": {\n        \"Seattle => New-York\": 2.5,\n        \"Seattle => Chicago\": 1.7,\n        \"Seattle => Topeka\": 1.8,\n        \"San-Diego => New-York\": 2.5,\n        \"San-Diego => Chicago\": 1.8,\n        \"San-Diego => Topeka\": 1.4\n    }\n}\n\"\"\")","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Create the set of plants:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"P = keys(data[\"plants\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Create the set of markets:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"M = keys(data[\"markets\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We also need a function to compute the distance from plant to market:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"distance(p::String, m::String) = data[\"distances\"][\"$(p) => $(m)\"]","category":"page"},{"location":"tutorials/linear/cannery/#JuMP-formulation","page":"The cannery problem","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Now we're ready to convert our mathematical formulation into a JuMP model.","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"First, create a new JuMP model. Since we have a linear program, we'll use HiGHS as our optimizer:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Our decision variables are indexed over the set of plants and markets:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@variable(model, x[P, M] >= 0)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"We need a constraint that each plant can ship no more than its capacity:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@constraint(model, [p in P], sum(x[p, :]) <= data[\"plants\"][p][\"capacity\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"and each market must receive at least its demand:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@constraint(model, [m in M], sum(x[:, m]) >= data[\"markets\"][m][\"demand\"])","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Finally, our objective is to minimize the transportation distance:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"@objective(model, Min, sum(distance(p, m) * x[p, m] for p in P, m in M));\nnothing #hide","category":"page"},{"location":"tutorials/linear/cannery/#Solution","page":"The cannery problem","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"Let's optimize and look at the solution:","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"What's the optimal shipment?","category":"page"},{"location":"tutorials/linear/cannery/","page":"The cannery problem","title":"The cannery problem","text":"for p in P, m in M\n    println(p, \" => \", m, \": \", value(x[p, m]))\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Utilities/overview.md\"","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Utilities/overview/#The-Utilities-submodule","page":"Overview","title":"The Utilities submodule","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The Utilities submodule provides a variety of functions and datastructures for managing MOI.ModelLike objects.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.Model","page":"Overview","title":"Utilities.Model","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities.Model provides an implementation of a ModelLike that efficiently supports all functions and sets defined within MOI. However, given the extensibility of MOI, this might not cover all use cases.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Create a model as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.UniversalFallback","page":"Overview","title":"Utilities.UniversalFallback","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities.UniversalFallback is a layer that sits on top of any ModelLike and provides non-specialized (slower) fallbacks for constraints and attributes that the underlying ModelLike does not support.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For example, Utilities.Model doesn't support some variable attributes like VariablePrimalStart, so JuMP uses a combination of Universal fallback and Utilities.Model as a generic problem cache:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}())\nMOIU.UniversalFallback{MOIU.Model{Float64}}\nfallback for MOIU.Model{Float64}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nAdding a UniversalFallback means that your model will now support all constraints, even if the inner-model does not. This can lead to unexpected behavior.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.@model","page":"Overview","title":"Utilities.@model","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For advanced use cases that need efficient support for functions and sets defined outside of MOI (but still known at compile time), we provide the Utilities.@model macro.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The @model macro takes a name (for a new type, which must not exist yet), eight tuples specifying the types of constraints that are supported, and then a Bool indicating the type is a subtype of MOI.AbstractOptimizer (if true) or MOI.ModelLike (if false).","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The eight tuples are in the following order:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Un-typed scalar sets, for example, Integer\nTyped scalar sets, for example, LessThan\nUn-typed vector sets, for example, Nonnegatives\nTyped vector sets, for example, PowerCone\nUn-typed scalar functions, for example, VariableIndex\nTyped scalar functions, for example, ScalarAffineFunction\nUn-typed vector functions, for example, VectorOfVariables\nTyped vector functions, for example, VectorAffineFunction","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The tuples can contain more than one element. Typed-sets must be specified without their type parameter, for example, MOI.LessThan, not MOI.LessThan{Float64}.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Here is an example:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.@model(\n           MyNewModel,\n           (MOI.Integer,),                  # Un-typed scalar sets\n           (MOI.GreaterThan,),              # Typed scalar sets\n           (MOI.Nonnegatives,),             # Un-typed vector sets\n           (MOI.PowerCone,),                # Typed vector sets\n           (MOI.VariableIndex,),            # Un-typed scalar functions\n           (MOI.ScalarAffineFunction,),     # Typed scalar functions\n           (MOI.VectorOfVariables,),        # Un-typed vector functions\n           (MOI.VectorAffineFunction,),     # Typed vector functions\n           true,                            # <:MOI.AbstractOptimizer?\n       )\nMathOptInterface.Utilities.GenericOptimizer{T, MathOptInterface.Utilities.ObjectiveContainer{T}, MathOptInterface.Utilities.VariablesContainer{T}, MyNewModelFunctionConstraints{T}} where T\n\njulia> model = MyNewModel{Float64}()\nMOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MyNewModelFunctionConstraints{Float64}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nMyNewModel supports every VariableIndex-in-Set constraint, as well as VariableIndex, ScalarAffineFunction, and ScalarQuadraticFunction objective functions. Implement MOI.supports as needed to forbid constraint and objective function combinations.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"As another example, PATHSolver, which only supports VectorAffineFunction-in-Complements defines its optimizer as:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.@model(\n           PathOptimizer,\n           (),  # Scalar sets\n           (),  # Typed scalar sets\n           (MOI.Complements,),  # Vector sets\n           (),  # Typed vector sets\n           (),  # Scalar functions\n           (),  # Typed scalar functions\n           (),  # Vector functions\n           (MOI.VectorAffineFunction,),  # Typed vector functions\n           true,  # is_optimizer\n       )\nMathOptInterface.Utilities.GenericOptimizer{T, MathOptInterface.Utilities.ObjectiveContainer{T}, MathOptInterface.Utilities.VariablesContainer{T}, MathOptInterface.Utilities.VectorOfConstraints{MathOptInterface.VectorAffineFunction{T}, MathOptInterface.Complements}} where T","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"However, PathOptimizer does not support some VariableIndex-in-Set constraints, so we must explicitly define:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> function MOI.supports_constraint(\n           ::PathOptimizer,\n           ::Type{MOI.VariableIndex},\n           ::Type{Union{<:MOI.Semiinteger,MOI.Semicontinuous,MOI.ZeroOne,MOI.Integer}}\n       )\n           return false\n       end","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Finally, PATH doesn't support an objective function, so we need to add:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.supports(::PathOptimizer, ::MOI.ObjectiveFunction) = false","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"warning: Warning\nThis macro creates a new type, so it must be called from the top-level of a module, for example, it cannot be called from inside a function.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.CachingOptimizer","page":"Overview","title":"Utilities.CachingOptimizer","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A [Utilities.CachingOptimizer] is an MOI layer that abstracts the difference between solvers that support incremental modification (for example, they support adding variables one-by-one), and solvers that require the entire problem in a single API call (for example, they only accept the A, b and c matrices of a linear program).","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"It has two parts:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A cache, where the model can be built and modified incrementally\nAn optimizer, which is used to solve the problem","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.CachingOptimizer(\n           MOI.Utilities.Model{Float64}(),\n           PathOptimizer{Float64}(),\n       )\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"A Utilities.CachingOptimizer may be in one of three possible states:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"NO_OPTIMIZER: The CachingOptimizer does not have any optimizer.\nEMPTY_OPTIMIZER: The CachingOptimizer has an empty optimizer, and it is not synchronized with the cached model. Modifications are forwarded to the cache, but not to the optimizer.\nATTACHED_OPTIMIZER: The CachingOptimizer has an optimizer, and it is synchronized with the cached model. Modifications are forwarded to the optimizer. If the optimizer does not support modifications, and error will be thrown.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.attach_optimizer to go from EMPTY_OPTIMIZER to ATTACHED_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.attach_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state ATTACHED_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"info: Info\nYou must be in ATTACHED_OPTIMIZER to use optimize!.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.reset_optimizer to go from ATTACHED_OPTIMIZER to EMPTY_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.reset_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"info: Info\nCalling MOI.empty!(model) also resets the state to EMPTY_OPTIMIZER. So after emptying a model, the modification will only be applied to the cache.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.drop_optimizer to go from any state to NO_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.drop_optimizer(model)\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state NO_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer nothing","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Pass an empty optimizer to Utilities.reset_optimizer to go from NO_OPTIMIZER to EMPTY_OPTIMIZER:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.reset_optimizer(model, PathOptimizer{Float64}())\n\njulia> model\nMOIU.CachingOptimizer{MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode AUTOMATIC\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Deciding when to attach and reset the optimizer is tedious, and you will often write code like this:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"try\n    # modification\ncatch\n    MOI.Utilities.reset_optimizer(model)\n    # Re-try modification\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"To make this easier, Utilities.CachingOptimizer has two modes of operation:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"AUTOMATIC: The CachingOptimizer changes its state when necessary. Attempting to add a constraint or perform a modification not supported by the optimizer results in a drop to EMPTY_OPTIMIZER mode.\nMANUAL: The user must change the state of the CachingOptimizer. Attempting to perform an operation in the incorrect state results in an error.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"By default, AUTOMATIC mode is chosen. However, you can create a CachingOptimizer in MANUAL mode as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.CachingOptimizer(\n           MOI.Utilities.Model{Float64}(),\n           MOI.Utilities.MANUAL,\n       )\nMOIU.CachingOptimizer{MOI.AbstractOptimizer, MOIU.Model{Float64}}\nin state NO_OPTIMIZER\nin mode MANUAL\nwith model cache MOIU.Model{Float64}\nwith optimizer nothing\n\njulia> MOI.Utilities.reset_optimizer(model, PathOptimizer{Float64}())\n\njulia> model\nMOIU.CachingOptimizer{MOI.AbstractOptimizer, MOIU.Model{Float64}}\nin state EMPTY_OPTIMIZER\nin mode MANUAL\nwith model cache MOIU.Model{Float64}\nwith optimizer MOIU.GenericOptimizer{Float64, MOIU.ObjectiveContainer{Float64}, MOIU.VariablesContainer{Float64}, MOIU.VectorOfConstraints{MOI.VectorAffineFunction{Float64}, MOI.Complements}}","category":"page"},{"location":"moi/submodules/Utilities/overview/#Printing","page":"Overview","title":"Printing","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use print to print the formulation of the model.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model)\nMOI.VariableIndex(1)\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x_var\")\n\njulia> MOI.add_constraint(model, x, MOI.ZeroOne())\nMathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.ZeroOne}(1)\n\njulia> MOI.set(model, MOI.ObjectiveFunction{typeof(x)}(), x)\n\njulia> MOI.set(model, MOI.ObjectiveSense(), MOI.MAX_SENSE)\n\njulia> print(model)\nMaximize VariableIndex:\n x_var\n\nSubject to:\n\nVariableIndex-in-ZeroOne\n x_var ∈ {0, 1}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Use Utilities.latex_formulation to display the model in LaTeX form:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> MOI.Utilities.latex_formulation(model)\n$$ \\begin{aligned}\n\\max\\quad & x\\_var \\\\\n\\text{Subject to}\\\\\n & \\text{VariableIndex-in-ZeroOne} \\\\\n & x\\_var \\in \\{0, 1\\} \\\\\n\\end{aligned} $$","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"tip: Tip\nIn IJulia, calling print or ending a cell with Utilities.latex_formulation will render the model in LaTeX.","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.PenaltyRelaxation","page":"Overview","title":"Utilities.PenaltyRelaxation","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Pass Utilities.PenaltyRelaxation to modify to relax the problem by adding penalized slack variables to the constraints. This is helpful when debugging sources of infeasible models.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x\")\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> map = MOI.modify(model, MOI.Utilities.PenaltyRelaxation(Dict(c => 2.0)));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 x - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> map[c]\n0.0 + 1.0 MOI.VariableIndex(2)","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"You can also modify a single constraint using Utilities.ScalarPenaltyRelaxation:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.Utilities.Model{Float64}();\n\njulia> x = MOI.add_variable(model);\n\njulia> MOI.set(model, MOI.VariableName(), x, \"x\")\n\njulia> c = MOI.add_constraint(model, 1.0 * x, MOI.LessThan(2.0));\n\njulia> f = MOI.modify(model, c, MOI.Utilities.ScalarPenaltyRelaxation(2.0));\n\njulia> print(model)\nMinimize ScalarAffineFunction{Float64}:\n 0.0 + 2.0 v[2]\n\nSubject to:\n\nScalarAffineFunction{Float64}-in-LessThan{Float64}\n 0.0 + 1.0 x - 1.0 v[2] <= 2.0\n\nVariableIndex-in-GreaterThan{Float64}\n v[2] >= 0.0\n\njulia> f\n0.0 + 1.0 MOI.VariableIndex(2)","category":"page"},{"location":"moi/submodules/Utilities/overview/#Utilities.MatrixOfConstraints","page":"Overview","title":"Utilities.MatrixOfConstraints","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The constraints of Utilities.Model are stored as a vector of tuples of function and set in a Utilities.VectorOfConstraints. Other representations can be used by parameterizing the type Utilities.GenericModel (resp. Utilities.GenericOptimizer). For instance, if all non-VariableIndex constraints are affine, the coefficients of all the constraints can be stored in a single sparse matrix using Utilities.MatrixOfConstraints. The constraints storage can even be customized up to a point where it exactly matches the storage of the solver of interest, in which case copy_to can be implemented for the solver by calling copy_to to this custom model.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For instance, Clp defines the following model:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"MOI.Utilities.@product_of_scalar_sets(LP, MOI.EqualTo{T}, MOI.LessThan{T}, MOI.GreaterThan{T})\nconst Model = MOI.Utilities.GenericModel{\n    Float64,\n    MOI.Utilities.MatrixOfConstraints{\n        Float64,\n        MOI.Utilities.MutableSparseMatrixCSC{Float64,Cint,MOI.Utilities.ZeroBasedIndexing},\n        MOI.Utilities.Hyperrectangle{Float64},\n        LP{Float64},\n    },\n}","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The copy_to operation can now be implemented as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function _copy_to(dest::Optimizer, src::Model)\n    @assert MOI.is_empty(dest)\n    A = src.constraints.coefficients\n    row_bounds = src.constraints.constants\n    Clp_loadProblem(\n        dest,\n        A.n,\n        A.m,\n        A.colptr,\n        A.rowval,\n        A.nzval,\n        src.lower_bound,\n        src.upper_bound,\n        # (...) objective vector (omitted),\n        row_bounds.lower,\n        row_bounds.upper,\n    )\n    # Set objective sense and constant (omitted)\n    return\nend\n\nfunction MOI.copy_to(dest::Optimizer, src::Model)\n    _copy_to(dest, src)\n    return MOI.Utilities.identity_index_map(src)\nend\n\nfunction MOI.copy_to(\n    dest::Optimizer,\n    src::MOI.Utilities.UniversalFallback{Model},\n)\n    # Copy attributes from `src` to `dest` and error in case any unsupported\n    # constraints or attributes are set in `UniversalFallback`.\n    return MOI.copy_to(dest, src.model)\nend\n\nfunction MOI.copy_to(\n    dest::Optimizer,\n    src::MOI.ModelLike,\n)\n    model = Model()\n    index_map = MOI.copy_to(model, src)\n    _copy_to(dest, model)\n    return index_map\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/#ModelFilter","page":"Overview","title":"ModelFilter","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"Utilities provides Utilities.ModelFilter as a useful tool to copy a subset of a model. For example, given an infeasible model, we can copy the irreducible infeasible subsystem (for models implementing ConstraintConflictStatus) as follows:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"my_filter(::Any) = true\nfunction my_filter(ci::MOI.ConstraintIndex)\n    status = MOI.get(dest, MOI.ConstraintConflictStatus(), ci)\n    return status != MOI.NOT_IN_CONFLICT\nend\nfiltered_src = MOI.Utilities.ModelFilter(my_filter, src)\nindex_map = MOI.copy_to(dest, filtered_src)","category":"page"},{"location":"moi/submodules/Utilities/overview/#Fallbacks","page":"Overview","title":"Fallbacks","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The value of some attributes can be inferred from the value of other attributes.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"For example, the value of ObjectiveValue can be computed using ObjectiveFunction and VariablePrimal.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"When a solver gives direct access to an attribute, it is better to return this value. However, if this is not the case, Utilities.get_fallback can be used instead. For example:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function MOI.get(model::Optimizer, attr::MOI.ObjectiveFunction)\n    return MOI.Utilities.get_fallback(model, attr)\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/#DoubleDicts","page":"Overview","title":"DoubleDicts","text":"","category":"section"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"When writing MOI interfaces, we often need to handle situations in which we map ConstraintIndexs to different values. For example, to a string for ConstraintName.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"One option is to use a dictionary like Dict{MOI.ConstraintIndex,String}. However, this incurs a performance cost because the key is not a concrete type.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The DoubleDicts submodule helps this situation by providing two types main types Utilities.DoubleDicts.DoubleDict and Utilities.DoubleDicts.IndexDoubleDict. These types act like normal dictionaries, but internally they use more efficient dictionaries specialized to the type of the function-set pair.","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"The most common usage of a DoubleDict is in the index_map returned by copy_to. Performance can be improved, by using a function barrier. That is, instead of code like:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"index_map = MOI.copy_to(dest, src)\nfor (F, S) in MOI.get(src, MOI.ListOfConstraintTypesPresent())\n    for ci in MOI.get(src, MOI.ListOfConstraintIndices{F,S}())\n        dest_ci = index_map[ci]\n        # ...\n    end\nend","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"use instead:","category":"page"},{"location":"moi/submodules/Utilities/overview/","page":"Overview","title":"Overview","text":"function function_barrier(\n    dest,\n    src,\n    index_map::MOI.Utilities.DoubleDicts.IndexDoubleDictInner{F,S},\n) where {F,S}\n    for ci in MOI.get(src, MOI.ListOfConstraintIndices{F,S}())\n        dest_ci = index_map[ci]\n        # ...\n    end\n    return\nend\n\nindex_map = MOI.copy_to(dest, src)\nfor (F, S) in MOI.get(src, MOI.ListOfConstraintTypesPresent())\n    function_barrier(dest, src, index_map[F, S])\nend","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Test/reference.md\"","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Test/reference/#test_reference","page":"API Reference","title":"The Test submodule","text":"","category":"section"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"Functions to help test implementations of MOI. See The Test submodule for more details.","category":"page"},{"location":"moi/submodules/Test/reference/","page":"API Reference","title":"API Reference","text":"Test.Config\nTest.runtests\nTest.setup_test\nTest.version_added\nTest.@requires\nTest.RequirementUnmet","category":"page"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.Config","page":"API Reference","title":"MathOptInterface.Test.Config","text":"Config(\n    ::Type{T} = Float64;\n    atol::Real = Base.rtoldefault(T),\n    rtol::Real = Base.rtoldefault(T),\n    optimal_status::MOI.TerminationStatusCode = MOI.OPTIMAL,\n    infeasible_status::MOI.TerminationStatusCode = MOI.INFEASIBLE,\n    exclude::Vector{Any} = Any[],\n) where {T}\n\nReturn an object that is used to configure various tests.\n\nConfiguration arguments\n\natol::Real = Base.rtoldefault(T): Control the absolute tolerance used when comparing solutions.\nrtol::Real = Base.rtoldefault(T): Control the relative tolerance used when comparing solutions.\noptimal_status = MOI.OPTIMAL: Set to MOI.LOCALLY_SOLVED if the solver cannot prove global optimality.\ninfeasible_status = MOI.INFEASIBLE: Set to MOI.LOCALLY_INFEASIBLE if the solver cannot prove global infeasibility.\nexclude = Vector{Any}: Pass attributes or functions to exclude to skip parts of tests that require certain functionality. Common arguments include:\nMOI.delete to skip deletion-related tests\nMOI.optimize! to skip optimize-related tests\nMOI.ConstraintDual to skip dual-related tests\nMOI.VariableName to skip setting variable names\nMOI.ConstraintName to skip setting constraint names\n\nExamples\n\nFor a nonlinear solver that finds local optima and does not support finding dual variables or constraint names:\n\nConfig(\n    Float64;\n    optimal_status = MOI.LOCALLY_SOLVED,\n    exclude = Any[\n        MOI.ConstraintDual,\n        MOI.VariableName,\n        MOI.ConstraintName,\n        MOI.delete,\n    ],\n)\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.runtests","page":"API Reference","title":"MathOptInterface.Test.runtests","text":"runtests(\n    model::MOI.ModelLike,\n    config::Config;\n    include::Vector{Union{String,Regex}} = String[],\n    exclude::Vector{Union{String,Regex}} = String[],\n    warn_unsupported::Bool = false,\n    exclude_tests_after::VersionNumber = v\"999.0.0\",\n)\n\nRun all tests in MathOptInterface.Test on model.\n\nConfiguration arguments\n\nconfig is a Test.Config object that can be used to modify the behavior of tests.\nIf include is not empty, only run tests if an element from include occursin the name of the test.\nIf exclude is not empty, skip tests if an element from exclude occursin the name of the test.\nexclude takes priority over include.\nIf warn_unsupported is false, runtests will silently skip tests that fail with a MOI.NotAllowedError, MOI.UnsupportedError, or RequirementUnmet error. (The latter is thrown when an @requires statement returns false.) When warn_unsupported is true, a warning will be printed. For most cases the default behavior, false, is what you want, since these tests likely test functionality that is not supported by model. However, it can be useful to run  warn_unsupported = true to check you are not skipping tests due to a missing supports_constraint method or equivalent.\nexclude_tests_after is a version number that excludes any tests to MOI added after that version number. This is useful for solvers who can declare a fixed set of tests, and not cause their tests to break if a new patch of MOI is released with a new test.\n\nSee also: setup_test.\n\nExample\n\nconfig = MathOptInterface.Test.Config()\nMathOptInterface.Test.runtests(\n    model,\n    config;\n    include = [\"test_linear_\", r\"^test_model_Name$\"],\n    exclude = [\"VariablePrimalStart\"],\n    warn_unsupported = true,\n    exclude_tests_after = v\"0.10.5\",\n)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.setup_test","page":"API Reference","title":"MathOptInterface.Test.setup_test","text":"setup_test(::typeof(f), model::MOI.ModelLike, config::Config)\n\nOverload this method to modify model before running the test function f on model with config. You can also modify the fields in config (e.g., to loosen the default tolerances).\n\nThis function should either return nothing, or return a function which, when called with zero arguments, undoes the setup to return the model to its previous state. You do not need to undo any modifications to config.\n\nThis function is most useful when writing new tests of the tests for MOI, but it can also be used to set test-specific tolerances, etc.\n\nSee also: runtests\n\nExample\n\nfunction MOI.Test.setup_test(\n    ::typeof(MOI.Test.test_linear_VariablePrimalStart_partial),\n    mock::MOIU.MockOptimizer,\n    ::MOI.Test.Config,\n)\n    MOIU.set_mock_optimize!(\n        mock,\n        (mock::MOIU.MockOptimizer) -> MOIU.mock_optimize!(mock, [1.0, 0.0]),\n    )\n    mock.eval_variable_constraint_dual = false\n\n    function reset_function()\n        mock.eval_variable_constraint_dual = true\n        return\n    end\n    return reset_function\nend\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.version_added","page":"API Reference","title":"MathOptInterface.Test.version_added","text":"version_added(::typeof(function_name))\n\nReturns the version of MOI in which the test function_name was added.\n\nThis method should be implemented for all new tests.\n\nSee the exclude_tests_after keyword of runtests for more details.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.@requires","page":"API Reference","title":"MathOptInterface.Test.@requires","text":"@requires(x)\n\nCheck that the condition x is true. Otherwise, throw an RequirementUnmet error to indicate that the model does not support something required by the test function.\n\nExamples\n\n@requires MOI.supports(model, MOI.Silent())\n@test MOI.get(model, MOI.Silent())\n\n\n\n\n\n","category":"macro"},{"location":"moi/submodules/Test/reference/#MathOptInterface.Test.RequirementUnmet","page":"API Reference","title":"MathOptInterface.Test.RequirementUnmet","text":"RequirementUnmet(msg::String) <: Exception\n\nAn error for throwing in tests to indicate that the model does not support some requirement expected by the test function.\n\n\n\n\n\n","category":"type"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"EditURL = \"https://github.com/MadNLP/MadNLP.jl/blob/389561354a774441cd5e6b3aa5fffee102ed222e/README.md\"","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"<img src=\"https://github.com/MadNLP/MadNLP.jl/blob/master/logo-full.svg?raw=true\"/>","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Documentation Build Status Coverage DOI\n(Image: doc) (Image: build) (Image: codecov) (Image: DOI)","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is a nonlinear programming (NLP) solver, purely implemented in Julia. MadNLP implements a filter line-search algorithm, as that used in Ipopt. MadNLP seeks to streamline the development of modeling and algorithmic paradigms in order to exploit structures and to make efficient use of high-performance computers.","category":"page"},{"location":"packages/MadNLP/#License","page":"MadNLP/MadNLP.jl","title":"License","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is available under the MIT license.","category":"page"},{"location":"packages/MadNLP/#Installation","page":"MadNLP/MadNLP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"pkg> add MadNLP","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Optionally, various extension packages can be installed together:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"pkg> add MadNLPHSL, MadNLPPardiso, MadNLPMumps, MadNLPGPU, MadNLPGraph, MadNLPKrylov","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"These packages are stored in the lib subdirectory within the main MadNLP repository. Some extension packages may require additional dependencies or specific hardware. For the instructions for the build procedure, see the following links:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLPHSL\nMadNLPPardiso\nMadNLPGPU","category":"page"},{"location":"packages/MadNLP/#Usage","page":"MadNLP/MadNLP.jl","title":"Usage","text":"","category":"section"},{"location":"packages/MadNLP/#Interfaces","page":"MadNLP/MadNLP.jl","title":"Interfaces","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is interfaced with modeling packages:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"JuMP\nPlasmo\nNLPModels.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Users can pass various options to MadNLP also through the modeling packages. The interface-specific syntax are shown below. To see the list of MadNLP solver options, check the OPTIONS.md file.","category":"page"},{"location":"packages/MadNLP/#JuMP-interface","page":"MadNLP/MadNLP.jl","title":"JuMP interface","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, JuMP\nmodel = Model(()->MadNLP.Optimizer(print_level=MadNLP.INFO, max_iter=100))\n@variable(model, x, start = 0.0)\n@variable(model, y, start = 0.0)\n@NLobjective(model, Min, (1 - x)^2 + 100 * (y - x^2)^2)\noptimize!(model)","category":"page"},{"location":"packages/MadNLP/#NLPModels-interface","page":"MadNLP/MadNLP.jl","title":"NLPModels interface","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, CUTEst\nmodel = CUTEstModel(\"PRIMALC1\")\nmadnlp(model, print_level=MadNLP.WARN, max_wall_time=3600)","category":"page"},{"location":"packages/MadNLP/#Plasmo-interface-(requires-extension-MadNLPGraph)","page":"MadNLP/MadNLP.jl","title":"Plasmo interface (requires extension MadNLPGraph)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGraph, Plasmo\ngraph = OptiGraph()\n@optinode(graph,n1)\n@optinode(graph,n2)\n@variable(n1,0 <= x <= 2)\n@variable(n1,0 <= y <= 3)\n@constraint(n1,x+y <= 4)\n@objective(n1,Min,x)\n@variable(n2,x)\n@NLnodeconstraint(n2,exp(x) >= 2)\n@linkconstraint(graph,n1[:x] == n2[:x])\nMadNLP.optimize!(graph; print_level=MadNLP.DEBUG, max_iter=100)","category":"page"},{"location":"packages/MadNLP/#Linear-Solvers","page":"MadNLP/MadNLP.jl","title":"Linear Solvers","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"MadNLP is interfaced with non-Julia sparse/dense linear solvers:","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Umfpack\nMKL-Pardiso\nMKL-Lapack\nHSL solvers (requires extension)\nPardiso (requires extension)\nMumps  (requires extension)\ncuSOLVER (requires extension)","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Each linear solver in MadNLP is a Julia type, and the linear_solver option should be specified by the actual type. Note that the linear solvers are always exported to Main.","category":"page"},{"location":"packages/MadNLP/#Built-in-Solvers:-Umfpack,-PardisoMKL,-LapackCPU","page":"MadNLP/MadNLP.jl","title":"Built-in Solvers: Umfpack, PardisoMKL, LapackCPU","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=UmfpackSolver)) # default\nmodel = Model(()->MadNLP.Optimizer(linear_solver=LapackCPUSolver))","category":"page"},{"location":"packages/MadNLP/#HSL-(requires-extension-MadNLPHSL)","page":"MadNLP/MadNLP.jl","title":"HSL (requires extension MadNLPHSL)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPHSL, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma27Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma57Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma77Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma86Solver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=Ma97Solver))","category":"page"},{"location":"packages/MadNLP/#Mumps-(requires-extension-MadNLPMumps)","page":"MadNLP/MadNLP.jl","title":"Mumps (requires extension MadNLPMumps)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPMumps, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MumpsSolver))","category":"page"},{"location":"packages/MadNLP/#Pardiso-(requires-extension-MadNLPPardiso)","page":"MadNLP/MadNLP.jl","title":"Pardiso (requires extension MadNLPPardiso)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPPardiso, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=PardisoSolver))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=PardisoMKLSolver))","category":"page"},{"location":"packages/MadNLP/#LapackGPU-(requires-extension-MadNLPGPU)","page":"MadNLP/MadNLP.jl","title":"LapackGPU (requires extension MadNLPGPU)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGPU, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=LapackGPUSolver))","category":"page"},{"location":"packages/MadNLP/#Schur-and-Schwarz-(requires-extension-MadNLPGraph)","page":"MadNLP/MadNLP.jl","title":"Schur and Schwarz (requires extension MadNLPGraph)","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"using MadNLP, MadNLPGraph, JuMP\n# ...\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MadNLPSchwarz))\nmodel = Model(()->MadNLP.Optimizer(linear_solver=MadNLPSchur))","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"The solvers in MadNLPGraph (Schur and Schwarz) use multi-thread parallelism; thus, Julia session should be started with -t flag.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"julia -t 16 # to use 16 threads","category":"page"},{"location":"packages/MadNLP/#Citing-MadNLP.jl","page":"MadNLP/MadNLP.jl","title":"Citing MadNLP.jl","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"If you use MadNLP.jl in your research, we would greatly appreciate your citing it.","category":"page"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"@article{shin2020graph,\n  title={Graph-Based Modeling and Decomposition of Energy Infrastructures},\n  author={Shin, Sungho and Coffrin, Carleton and Sundar, Kaarthik and Zavala, Victor M},\n  journal={arXiv preprint arXiv:2010.02404},\n  year={2020}\n}","category":"page"},{"location":"packages/MadNLP/#Bug-reports-and-support","page":"MadNLP/MadNLP.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/MadNLP/","page":"MadNLP/MadNLP.jl","title":"MadNLP/MadNLP.jl","text":"Please report issues and feature requests via the GitHub issue tracker.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\nend","category":"page"},{"location":"manual/variables/#jump_variables","page":"Variables","title":"Variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The term variable in mathematical optimization has many meanings. For example, optimization variables (also called decision variables) are the unknowns x that we are solving for in the problem:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"beginalign\n     min_x in mathbbR^n  f_0(x)\n    \n     textst  f_i(x)  in mathcalS_i  i = 1 ldots m\nendalign","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To complicate things, Julia uses variable to mean a binding between a name and a value. For example, in the statement:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x = 1\n1","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"x is a variable that stores the value 1.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP uses variable in a third way, to mean an instance of the VariableRef struct. JuMP variables are the link between Julia and the optimization variables inside a JuMP model.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This page explains how to create and manage JuMP variables in a variety of contexts.","category":"page"},{"location":"manual/variables/#Create-a-variable","page":"Variables","title":"Create a variable","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create variables using the @variable macro:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> typeof(x)\nVariableRef (alias for GenericVariableRef{Float64})\n\njulia> num_variables(model)\n1","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here x is a Julia variable that is bound to a VariableRef object, and we have added 1 decision variable to our model.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To make the binding more explicit, we could have written:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model, x)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"but there is no need to in general; the macro does it for us.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When creating a variable, you can also specify variable bounds:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x_free)\nx_free\n\njulia> @variable(model, x_lower >= 0)\nx_lower\n\njulia> @variable(model, x_upper <= 1)\nx_upper\n\njulia> @variable(model, 2 <= x_interval <= 3)\nx_interval\n\njulia> @variable(model, x_fixed == 4)\nx_fixed\n\njulia> print(model)\nFeasibility\nSubject to\n x_fixed = 4\n x_lower ≥ 0\n x_interval ≥ 2\n x_upper ≤ 1\n x_interval ≤ 3","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nWhen creating a variable with a single lower- or upper-bound, and the value of the bound is not a numeric literal (for example, 1 or 1.0), the name of the variable must appear on the left-hand side. Putting the name on the right-hand side is an error. For example, to create a variable x:a = 1\n@variable(model, x >= 1)      # ✓ Okay\n@variable(model, 1.0 <= x)    # ✓ Okay\n@variable(model, x >= a)      # ✓ Okay\n@variable(model, a <= x)      # × Not okay\n@variable(model, x >= 1 / 2)  # ✓ Okay\n@variable(model, 1 / 2 <= x)  # × Not okay","category":"page"},{"location":"manual/variables/#Containers-of-variables","page":"Variables","title":"Containers of variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The @variable macro also supports creating collections of JuMP variables. We'll cover some brief syntax here; read the Variable containers section for more details.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"You can create arrays of JuMP variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]\n\njulia> x[1, 2]\nx[1,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Index sets can be named, and bounds can depend on those names:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, sqrt(i) <= x[i = 1:3] <= i^2)\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> x[2]\nx[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Sets can be any Julia type that supports iteration:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i = 2:3, j = 1:2:3, [\"red\", \"blue\"]] >= 0)\n3-dimensional DenseAxisArray{VariableRef,3,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\n    Dimension 3, [\"red\", \"blue\"]\nAnd data, a 2×2×2 Array{VariableRef, 3}:\n[:, :, \"red\"] =\n x[2,1,red]  x[2,3,red]\n x[3,1,red]  x[3,3,red]\n\n[:, :, \"blue\"] =\n x[2,1,blue]  x[2,3,blue]\n x[3,1,blue]  x[3,3,blue]\n\njulia> x[2, 1, \"red\"]\nx[2,1,red]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Sets can depend upon previous indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, u[i = 1:2, j = i:3])\nJuMP.Containers.SparseAxisArray{VariableRef, 2, Tuple{Int64, Int64}} with 5 entries:\n  [1, 1]  =  u[1,1]\n  [1, 2]  =  u[1,2]\n  [1, 3]  =  u[1,3]\n  [2, 2]  =  u[2,2]\n  [2, 3]  =  u[2,3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"and we can filter elements in the sets using the ; syntax:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, v[i = 1:9; mod(i, 3) == 0])\nJuMP.Containers.SparseAxisArray{VariableRef, 1, Tuple{Int64}} with 3 entries:\n  [3]  =  v[3]\n  [6]  =  v[6]\n  [9]  =  v[9]","category":"page"},{"location":"manual/variables/#Registered-variables","page":"Variables","title":"Registered variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When you create variables, JuMP registers them inside the model using their corresponding symbol. Get a registered name using model[:key]:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model()\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> @variable(model, x)\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x\n\njulia> model[:x] === x\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Registered names are most useful when you start to write larger models and want to break up the model construction into functions:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> function set_objective(model::Model)\n           @objective(model, Min, 2 * model[:my_x] + 1)\n           return\n       end\nset_objective (generic function with 1 method)\n\njulia> model = Model();\n\njulia> @variable(model, my_x);\n\njulia> set_objective(model)\n\njulia> print(model)\nMin 2 my_x + 1\nSubject to","category":"page"},{"location":"manual/variables/#anonymous_variables","page":"Variables","title":"Anonymous variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To reduce the likelihood of accidental bugs, and because JuMP registers variables inside a model, creating two variables with the same name is an error:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n[...]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"A common reason for encountering this error is adding variables in a loop.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"As a work-around, JuMP provides anonymous variables. Create a scalar valued anonymous variable by omitting the name argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model)\n_[1]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Anonymous variables get printed as an underscore followed by a unique index of the variable.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nThe index of the variable may not correspond to the column of the variable in the solver.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a container of anonymous JuMP variables by dropping the name in front of the [:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> y = @variable(model, [1:2])\n2-element Vector{VariableRef}:\n _[1]\n _[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The <= and >= short-hand cannot be used to set bounds on scalar-valued anonymous JuMP variables. Instead, use the lower_bound and upper_bound keywords:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x_lower = @variable(model, lower_bound = 1.0)\n_[1]\n\njulia> x_upper = @variable(model, upper_bound = 2.0)\n_[2]\n\njulia> x_interval = @variable(model, lower_bound = 3.0, upper_bound = 4.0)\n_[3]","category":"page"},{"location":"manual/variables/#Variable-names","page":"Variables","title":"Variable names","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"In addition to the symbol that variables are registered with, JuMP variables have a String name that is used for printing and writing to file formats.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Get and set the name of a variable using name and set_name:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> name(x)\n\"x\"\n\njulia> set_name(x, \"my_x_name\")\n\njulia> x\nmy_x_name","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Override the default choice of name using the base_name keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Note that names apply to each element of the container, not to the container of variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> name(x[1])\n\"my_var[1]\"\n\njulia> set_name(x[1], \"my_x\")\n\njulia> x\n2-element Vector{VariableRef}:\n my_x\n my_var[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nFor some models, setting the string name of each variable can take a non-trivial portion of the total time required to build the model. Turn off String names by passing set_string_name = false to @variable:julia> model = Model();\n\njulia> @variable(model, x, set_string_name = false)\n_[1]See Disable string names for more information.","category":"page"},{"location":"manual/variables/#Retrieve-a-variable-by-name","page":"Variables","title":"Retrieve a variable by name","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Retrieve a variable from a model using variable_by_name:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> variable_by_name(model, \"my_x\")\nmy_x","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If the name is not present, nothing will be returned:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> variable_by_name(model, \"bad_name\")","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"You can only look up individual variables using variable_by_name. Something like this will not work:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, [i = 1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]\n\njulia> variable_by_name(model, \"my_var\")","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"To look up a collection of variables, do not use variable_by_name. Instead, register them using the model[:key] = value syntax:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> model[:x] = @variable(model, [i = 1:2], base_name = \"my_var\")\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]\n\njulia> model[:x]\n2-element Vector{VariableRef}:\n my_var[1]\n my_var[2]","category":"page"},{"location":"manual/variables/#variable_names_and_bindings","page":"Variables","title":"String names, symbolic names, and bindings","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"It's common for new users to experience confusion relating to JuMP variables. Part of the problem is the overloaded use of \"variable\" in mathematical optimization, along with the difference between the name that a variable is registered under and the String name used for printing.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here's a summary of the differences:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP variables are created using @variable.\nJuMP variables can be named or anonymous.\nNamed JuMP variables have the form @variable(model, x). For named variables:\nThe String name of the variable is set to \"x\".\nA Julia variable x is created that binds x to  the JuMP variable.\nThe name :x is registered as a key in the model with the value x.\nAnonymous JuMP variables have the form x = @variable(model). For anonymous variables:\nThe String name of the variable is set to \"\". When printed, this is replaced with \"_[i]\" where i is the index of the variable.\nYou control the name of the Julia variable used as the binding.\nNo name is registered as a key in the model.\nThe base_name keyword can override the String name of the variable.\nYou can manually register names in the model via model[:key] = value","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Here's an example that should make things clearer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x_binding = @variable(model, base_name = \"x\")\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\n\njulia> x\nERROR: UndefVarError: `x` not defined\n\njulia> x_binding\nx\n\njulia> name(x_binding)\n\"x\"\n\njulia> model[:x_register] = x_binding\nx\n\njulia> model\nA JuMP Model\nFeasibility problem with:\nVariable: 1\nModel mode: AUTOMATIC\nCachingOptimizer state: NO_OPTIMIZER\nSolver name: No optimizer attached.\nNames registered in the model: x_register\n\njulia> model[:x_register]\nx\n\njulia> model[:x_register] === x_binding\ntrue\n\njulia> x\nERROR: UndefVarError: `x` not defined","category":"page"},{"location":"manual/variables/#Create,-delete,-and-modify-variable-bounds","page":"Variables","title":"Create, delete, and modify variable bounds","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Query whether a variable has a bound using has_lower_bound, has_upper_bound, and is_fixed:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> has_lower_bound(x_free)\nfalse\n\njulia> has_upper_bound(x_upper)\ntrue\n\njulia> is_fixed(x_fixed)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If a variable has a particular bound, query the value of it using lower_bound, upper_bound, and fix_value:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> lower_bound(x_interval)\n2.0\n\njulia> upper_bound(x_interval)\n3.0\n\njulia> fix_value(x_fixed)\n4.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Querying the value of a bound that does not exist will result in an error.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete variable bounds using delete_lower_bound, delete_upper_bound, and unfix:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> delete_lower_bound(x_lower)\n\njulia> has_lower_bound(x_lower)\nfalse\n\njulia> delete_upper_bound(x_upper)\n\njulia> has_upper_bound(x_upper)\nfalse\n\njulia> unfix(x_fixed)\n\njulia> is_fixed(x_fixed)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Set or update variable bounds using set_lower_bound, set_upper_bound, and fix:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> set_lower_bound(x_lower, 1.1)\n\njulia> set_upper_bound(x_upper, 2.1)\n\njulia> fix(x_fixed, 4.1)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Fixing a variable with existing bounds will throw an error. To delete the bounds prior to fixing, use fix(variable, value; force = true).","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x >= 1)\nx\n\njulia> fix(x, 2)\nERROR: Unable to fix x to 2 because it has existing variable bounds. Consider calling `JuMP.fix(variable, value; force=true)` which will delete existing bounds before fixing the variable.\n\njulia> fix(x, 2; force = true)\n\n\njulia> fix_value(x)\n2.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nUse fix instead of @constraint(model, x == 2). The former modifies variable bounds, while the latter adds a new linear constraint to the problem.","category":"page"},{"location":"manual/variables/#Binary-variables","page":"Variables","title":"Binary variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Binary variables are constrained to the set x in 0 1.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a binary variable by passing Bin as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, Bin)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Check if a variable is binary using is_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> is_binary(x)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete a binary constraint using unset_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unset_binary(x)\n\njulia> is_binary(x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Binary variables can also be created by setting the binary keyword to true:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, binary=true)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"or by using set_binary:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_binary(x)","category":"page"},{"location":"manual/variables/#Integer-variables","page":"Variables","title":"Integer variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Integer variables are constrained to the set x in mathbbZ.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create an integer variable by passing Int as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, Int)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Check if a variable is integer using is_integer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> is_integer(x)\ntrue","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Delete an integer constraint using unset_integer.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unset_integer(x)\n\njulia> is_integer(x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Integer variables can also be created by setting the integer keyword to true:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x, integer=true)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"or by using set_integer:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_integer(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nThe relax_integrality function relaxes all integrality constraints in the model, returning a function that can be called to undo the operation later on.","category":"page"},{"location":"manual/variables/#Semi-integer-and-semi-continuous-variables","page":"Variables","title":"Semi-integer and semi-continuous variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Semi-continuous variables are constrained to the set x in 0 cup l u.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a semi-continuous variable using the Semicontinuous set:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x in Semicontinuous(1.5, 3.5))\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Semi-integer variables  are constrained to the set x in 0 cup l l+1 dots u.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create a semi-integer variable using the Semiinteger set:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x in Semiinteger(1.0, 3.0))\nx","category":"page"},{"location":"manual/variables/#Start-values","page":"Variables","title":"Start values","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"There are two ways to provide a primal starting solution (also called MIP-start or a warmstart) for each variable:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using the start keyword in the @variable macro\nusing set_start_value","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The starting value of a variable can be queried using start_value. If no start value has been set, start_value will return nothing.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> start_value(x)\n\njulia> @variable(model, y, start = 1)\ny\n\njulia> start_value(y)\n1.0\n\njulia> set_start_value(y, 2)\n\njulia> start_value(y)\n2.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The start keyword argument can depend on the indices of a variable container:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, z[i = 1:2], start = i^2)\n2-element Vector{VariableRef}:\n z[1]\n z[2]\n\njulia> start_value.(z)\n2-element Vector{Float64}:\n 1.0\n 4.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"warning: Warning\nSome solvers do not support start values. If a solver does not support start values, an MathOptInterface.UnsupportedAttribute{MathOptInterface.VariablePrimalStart} error will be thrown.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nTo set the optimal solution from a previous solve as a new starting value, use all_variables to get a vector of all the variables in the model, then run:x = all_variables(model)\nx_solution = value.(x)\nset_start_value.(x, x_solution)Alternatively, use set_start_values.","category":"page"},{"location":"manual/variables/#delete_a_variable","page":"Variables","title":"Delete a variable","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use delete to delete a variable from a model. Use is_valid to check if a variable belongs to a model and has not been deleted.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> is_valid(model, x)\ntrue\n\njulia> delete(model, x)\n\njulia> is_valid(model, x)\nfalse","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Deleting a variable does not unregister the corresponding name from the model. Therefore, creating a new variable of the same name will throw an error:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, x)\nERROR: An object of name x is already attached to this model. If this\n    is intended, consider using the anonymous construction syntax, e.g.,\n    `x = @variable(model, [1:N], ...)` where the name of the object does\n    not appear inside the macro.\n\n    Alternatively, use `unregister(model, :x)` to first unregister\n    the existing name from the model. Note that this will not delete the\n    object; it will just remove the reference at `model[:x]`.\n[...]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"After calling delete, call unregister to remove the symbolic reference:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> unregister(model, :x)\n\njulia> @variable(model, x)\nx","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"info: Info\ndelete does not automatically unregister because we do not distinguish between names that are automatically registered by JuMP macros and names that are manually registered by the user by setting values in object_dictionary. In addition, deleting a variable and then adding a new variable of the same name is an easy way to introduce bugs into your code.","category":"page"},{"location":"manual/variables/#Variable-containers","page":"Variables","title":"Variable containers","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP provides a mechanism for creating collections of variables in three types of data structures, which we refer to as containers.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The three types are Arrays, DenseAxisArrays, and SparseAxisArrays. We explain each of these in the following.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nYou can read more about containers in the Containers section.","category":"page"},{"location":"manual/variables/#Arrays","page":"Variables","title":"Arrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We have already seen the creation of an array of JuMP variables with the x[1:2] syntax. This can be extended to create multi-dimensional arrays of JuMP variables. For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2])\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Arrays of JuMP variables can be indexed and sliced as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x[1, 2]\nx[1,2]\n\njulia> x[2, :]\n2-element Vector{VariableRef}:\n x[2,1]\n x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Variable bounds can depend upon the indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=1:2] >= 2i + j)\n2×2 Matrix{VariableRef}:\n x[1,1]  x[1,2]\n x[2,1]  x[2,2]\n\njulia> lower_bound.(x)\n2×2 Matrix{Float64}:\n 3.0  4.0\n 5.0  6.0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP will form an Array of JuMP variables when it can determine at compile time that the indices are one-based integer ranges. Therefore x[1:b] will create an Array of JuMP variables, but x[a:b] will not. If JuMP cannot determine that the indices are one-based integer ranges (for example, in the case of x[a:b]), JuMP will create a DenseAxisArray instead.","category":"page"},{"location":"manual/variables/#variable_jump_arrays","page":"Variables","title":"DenseAxisArrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We often want to create arrays where the indices are not one-based integer ranges. For example, we may want to create a variable indexed by the name of a product or a location. The syntax is the same as that above, except with an arbitrary vector as an index as opposed to a one-based range. The biggest difference is that instead of returning an Array of JuMP variables, JuMP will return a DenseAxisArray. For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, [:A,:B]])\n2-dimensional DenseAxisArray{VariableRef,2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [:A, :B]\nAnd data, a 2×2 Matrix{VariableRef}:\n x[1,A]  x[1,B]\n x[2,A]  x[2,B]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"DenseAxisArrays can be indexed and sliced as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> x[1, :A]\nx[1,A]\n\njulia> x[2, :]\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, [:A, :B]\nAnd data, a 2-element Vector{VariableRef}:\n x[2,A]\n x[2,B]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Bounds can depend upon indices:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=2:3, j=1:2:3] >= 0.5i + j)\n2-dimensional DenseAxisArray{VariableRef,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\nAnd data, a 2×2 Matrix{VariableRef}:\n x[2,1]  x[2,3]\n x[3,1]  x[3,3]\n\njulia> lower_bound.(x)\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, 1:2:3\nAnd data, a 2×2 Matrix{Float64}:\n 2.0  4.0\n 2.5  4.5","category":"page"},{"location":"manual/variables/#variable_sparseaxisarrays","page":"Variables","title":"SparseAxisArrays","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The third container type that JuMP natively supports is SparseAxisArray. These arrays are created when the indices do not form a rectangular set. For example, this applies when indices have a dependence upon previous indices (called triangular indexing). JuMP supports this as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=i:2])\nJuMP.Containers.SparseAxisArray{VariableRef, 2, Tuple{Int64, Int64}} with 3 entries:\n  [1, 1]  =  x[1,1]\n  [1, 2]  =  x[1,2]\n  [2, 2]  =  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We can also conditionally create variables via a JuMP-specific syntax. This syntax appends a comparison check that depends upon the named indices and is separated from the indices by a semi-colon (;). For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[i=1:4; mod(i, 2)==0])\nJuMP.Containers.SparseAxisArray{VariableRef, 1, Tuple{Int64}} with 2 entries:\n  [2]  =  x[2]\n  [4]  =  x[4]","category":"page"},{"location":"manual/variables/#Performance-considerations","page":"Variables","title":"Performance considerations","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When using the semi-colon as a filter, JuMP iterates over all indices and evaluates the conditional for each combination. If there are many index dimensions and a large amount of sparsity, this can be inefficient.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> N = 10\n10\n\njulia> S = [(1, 1, 1), (N, N, N)]\n2-element Vector{Tuple{Int64, Int64, Int64}}:\n (1, 1, 1)\n (10, 10, 10)\n\njulia> @time @variable(model, x1[i=1:N, j=1:N, k=1:N; (i, j, k) in S])\n  0.203861 seconds (392.22 k allocations: 23.977 MiB, 99.10% compilation time)\nJuMP.Containers.SparseAxisArray{VariableRef, 3, Tuple{Int64, Int64, Int64}} with 2 entries:\n  [1, 1, 1   ]  =  x1[1,1,1]\n  [10, 10, 10]  =  x1[10,10,10]\n\njulia> @time @variable(model, x2[S])\n  0.045407 seconds (65.24 k allocations: 3.771 MiB, 99.15% compilation time)\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, [(1, 1, 1), (10, 10, 10)]\nAnd data, a 2-element Vector{VariableRef}:\n x2[(1, 1, 1)]\n x2[(10, 10, 10)]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The first option is slower because it is equivalent to:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x1 = Dict{NTuple{3,Int},VariableRef}()\nDict{Tuple{Int64, Int64, Int64}, VariableRef}()\n\njulia> for i in 1:N\n           for j in 1:N\n               for k in 1:N\n                   if (i, j, k) in S\n                       x1[i, j, k] = @variable(model, base_name = \"x1[$i,$j,$k]\")\n                   end\n               end\n           end\n       end\n\njulia> x1\nDict{Tuple{Int64, Int64, Int64}, VariableRef} with 2 entries:\n  (1, 1, 1)    => x1[1,1,1]\n  (10, 10, 10) => x1[10,10,10]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If performance is a concern, explicitly construct the set of indices instead of using the filtering syntax.","category":"page"},{"location":"manual/variables/#variable_forcing","page":"Variables","title":"Forcing the container type","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"When creating a container of JuMP variables, JuMP will attempt to choose the tightest container type that can store the JuMP variables. Thus, it will prefer to create an Array before a DenseAxisArray and a DenseAxisArray before a SparseAxisArray. However, because this happens at compile time, JuMP does not always make the best choice. To illustrate this, consider the following example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> A = 1:2\n1:2\n\njulia> @variable(model, x[A])\n1-dimensional DenseAxisArray{VariableRef,1,...} with index sets:\n    Dimension 1, 1:2\nAnd data, a 2-element Vector{VariableRef}:\n x[1]\n x[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Since the value (and type) of A is unknown at parsing time, JuMP is unable to infer that A is a one-based integer range. Therefore, JuMP creates a DenseAxisArray, even though it could store these two variables in a standard one-dimensional Array.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"We can share our knowledge that it is possible to store these JuMP variables as an array by setting the container keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, y[A], container=Array)\n2-element Vector{VariableRef}:\n y[1]\n y[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP now creates a vector of JuMP variables instead of a DenseAxisArray. Choosing an invalid container type will throw an error.","category":"page"},{"location":"manual/variables/#User-defined-containers","page":"Variables","title":"User-defined containers","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"In addition to the built-in container types, you can create your own collections of JuMP variables.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"tip: Tip\nThis is a point that users often overlook: you are not restricted to the built-in container types in JuMP.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, the following code creates a dictionary with symmetric matrices as the values:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> variables = Dict{Symbol,Array{VariableRef,2}}(\n           key => @variable(model, [1:2, 1:2], Symmetric, base_name = \"$(key)\")\n           for key in [:A, :B]\n       )\nDict{Symbol, Matrix{VariableRef}} with 2 entries:\n  :A => [A[1,1] A[1,2]; A[1,2] A[2,2]]\n  :B => [B[1,1] B[1,2]; B[1,2] B[2,2]]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Another common scenario is a request to add variables to existing containers, for example:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using JuMP\nmodel = Model()\n@variable(model, x[1:2] >= 0)\n# Later I want to add\n@variable(model, x[3:4] >= 0)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This is not possible with the built-in JuMP container types. However, you can use regular Julia types instead:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = model[:x] = @variable(model, [1:2], lower_bound = 0, base_name = \"x\")\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> append!(x, @variable(model, [1:2], lower_bound = 0, base_name = \"y\"));\n\njulia> model[:x]\n4-element Vector{VariableRef}:\n x[1]\n x[2]\n y[1]\n y[2]","category":"page"},{"location":"manual/variables/#Semidefinite-variables","page":"Variables","title":"Semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a square matrix of JuMP variables to be positive semidefinite by passing PSD as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], PSD)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This will ensure that x is symmetric, and that all of its eigenvalues are nonnegative.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nx must be a square 2-dimensional Array of JuMP variables; it cannot be a DenseAxisArray or a SparseAxisArray.","category":"page"},{"location":"manual/variables/#Symmetric-variables","page":"Variables","title":"Symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a square matrix of JuMP variables to be symmetric (but not necessarily positive semidefinite) by passing Symmetric  as an optional positional argument:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], Symmetric)\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#variables","page":"Variables","title":"The @variables macro","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"If you have many @variable calls, JuMP provides the macro @variables that can improve readability:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variables(model, begin\n           x\n           y[i=1:2] >= i, (start = i, base_name = \"Y_$i\")\n           z, Bin\n       end)\n(x, VariableRef[Y_1[1], Y_2[2]], z)\n\njulia> print(model)\nFeasibility\nSubject to\n Y_1[1] ≥ 1\n Y_2[2] ≥ 2\n z binary","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The @variables macro returns a tuple of the variables that were defined.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nKeyword arguments must be contained within parentheses.","category":"page"},{"location":"manual/variables/#jump_variables_on_creation","page":"Variables","title":"Variables constrained on creation","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"All uses of the @variable macro documented so far translate into separate calls for variable creation and the adding of any bound or integrality constraints.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, @variable(model, x >= 0, Int), is equivalent to:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x)\nx\n\njulia> set_lower_bound(x, 0.0)\n\njulia> set_integer(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Importantly, the bound and integrality constraints are added after the variable has been created.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"However, some solvers require a set specifying the variable domain to be given when the variable is first created. We say that these variables are constrained on creation.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use in within @variable to access the special syntax for constraining variables on creation.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For example, the following creates a vector of variables that belong to the SecondOrderCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, y[1:3] in SecondOrderCone())\n3-element Vector{VariableRef}:\n y[1]\n y[2]\n y[3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For contrast, the standard syntax is as follows:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> @variable(model, x[1:3])\n3-element Vector{VariableRef}:\n x[1]\n x[2]\n x[3]\n\njulia> @constraint(model, x in SecondOrderCone())\n[x[1], x[2], x[3]] ∈ MathOptInterface.SecondOrderCone(3)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"An alternate syntax to x in Set is to use the set keyword of @variable. This is most useful when creating anonymous variables:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> x = @variable(model, [1:3], set = SecondOrderCone())\n3-element Vector{VariableRef}:\n _[1]\n _[2]\n _[3]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nYou cannot delete the constraint associated with a variable constrained on creation.","category":"page"},{"location":"manual/variables/#Example:-positive-semidefinite-variables","page":"Variables","title":"Example: positive semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"An alternative to the syntax in Semidefinite variables, declare a matrix of JuMP variables to be positive semidefinite using PSDCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in PSDCone())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#Example:-symmetric-variables","page":"Variables","title":"Example: symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"As an alternative to the syntax in Symmetric variables, declare a matrix of JuMP variables to be symmetric using SymmetricMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in SymmetricMatrixSpace())\n2×2 LinearAlgebra.Symmetric{VariableRef, Matrix{VariableRef}}:\n x[1,1]  x[1,2]\n x[1,2]  x[2,2]","category":"page"},{"location":"manual/variables/#Example:-skew-symmetric-variables","page":"Variables","title":"Example: skew-symmetric variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be skew-symmetric using SkewSymmetricMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in SkewSymmetricMatrixSpace())\n2×2 Matrix{AffExpr}:\n 0        x[1,2]\n -x[1,2]  0","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"note: Note\nEven though x is a 2 by 2 matrix, only one decision variable is added to model; the remaining elements in x are linear transformations of the single variable.","category":"page"},{"location":"manual/variables/#Example:-Hermitian-positive-semidefinite-variables","page":"Variables","title":"Example: Hermitian positive semidefinite variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be Hermitian positive semidefinite using HermitianPSDCone:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, H[1:2, 1:2] in HermitianPSDCone())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    real(H[1,2]) + imag(H[1,2]) im\n real(H[1,2]) - imag(H[1,2]) im  real(H[2,2])","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This adds 4 real variables in the MOI.HermitianPositiveSemidefiniteConeTriangle:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> first(all_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle))\n[real(H[1,1]), real(H[1,2]), real(H[2,2]), imag(H[1,2])] ∈ MathOptInterface.HermitianPositiveSemidefiniteConeTriangle(2)","category":"page"},{"location":"manual/variables/#Example:-Hermitian-variables","page":"Variables","title":"Example: Hermitian variables","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Declare a matrix of JuMP variables to be Hermitian using the Hermitian tag:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2], Hermitian)\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(x[1,1])                    real(x[1,2]) + imag(x[1,2]) im\n real(x[1,2]) - imag(x[1,2]) im  real(x[2,2])","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"This is equivalent to declaring the variable in HermitianMatrixSpace:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2, 1:2] in HermitianMatrixSpace())\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(x[1,1])                    real(x[1,2]) + imag(x[1,2]) im\n real(x[1,2]) - imag(x[1,2]) im  real(x[2,2])","category":"page"},{"location":"manual/variables/#Why-use-variables-constrained-on-creation?","page":"Variables","title":"Why use variables constrained on creation?","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"For most users, it does not matter if you use the constrained on creation syntax. Therefore, use whatever syntax you find most convenient.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"However, if you use direct_model, you may be forced to use the constrained on creation syntax.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The technical difference between variables constrained on creation and the standard JuMP syntax is that variables constrained on creation calls MOI.add_constrained_variables, while the standard JuMP syntax calls MOI.add_variables and then MOI.add_constraint.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Consult the implementation of solver package you are using to see if your solver requires MOI.add_constrained_variables.","category":"page"},{"location":"manual/variables/#Parameters","page":"Variables","title":"Parameters","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Some solvers have explicit support for parameters, which are constants in the model that can be efficiently updated between solves.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"JuMP implements parameters by a decision variable constrained on creation to the Parameter set.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, p[i = 1:2] in Parameter(i))\n2-element Vector{VariableRef}:\n p[1]\n p[2]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Create anonymous parameters using the set keyword:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> anon_parameter = @variable(model, set = Parameter(1.0))\n_[4]","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Use parameter_value and set_parameter_value to query or update the value of a parameter.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> parameter_value.(p)\n2-element Vector{Float64}:\n 1.0\n 2.0\n\njulia> set_parameter_value(p[2], 3.0)\n\njulia> parameter_value.(p)\n2-element Vector{Float64}:\n 1.0\n 3.0","category":"page"},{"location":"manual/variables/#Limitations","page":"Variables","title":"Limitations","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Parameters are implemented as decision variables belonging to the Parameter set. If the solver supports the MOI.Parameter set, it may decide to replace all instances of the parameter variable by the associated constant. If the solver does not support parameters, it will add the parameter as a decision variable with fixed bounds.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"The most important implication of this design is that JuMP treats a parameter multiplied by a decision variable as a quadratic expression, even though it is equivalent to a linear expression.","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"julia> model = Model();\n\njulia> @variable(model, x);\n\njulia> @variable(model, p in Parameter(2));\n\njulia> px = @expression(model, p * x)\np*x\n\njulia> typeof(px)\nQuadExpr (alias for GenericQuadExpr{Float64, GenericVariableRef{Float64}})","category":"page"},{"location":"manual/variables/#When-to-use-a-parameter","page":"Variables","title":"When to use a parameter","text":"","category":"section"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Parameters are most useful when solving nonlinear models in a sequence:","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"using JuMP, Ipopt\nmodel = Model(Ipopt.Optimizer);\nset_silent(model)\n@variable(model, x)\n@variable(model, p in Parameter(1.0))\n@objective(model, Min, (x - p)^2)\noptimize!(model)\nvalue(x)\nset_parameter_value(p, 5.0)\noptimize!(model)\nvalue(x)","category":"page"},{"location":"manual/variables/","page":"Variables","title":"Variables","text":"Using parameters can be faster than creating a new model from scratch with updated data because JuMP is able to avoid repeating a number of steps in processing the model before handing it off to the solver.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/FileFormats/overview.md\"","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/FileFormats/overview/#The-FileFormats-submodule","page":"Overview","title":"The FileFormats submodule","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The FileFormats module provides functions for reading and writing MOI models using write_to_file and read_from_file.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Supported-file-types","page":"Overview","title":"Supported file types","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"You must read and write files to a FileFormats.Model object. Specific the file-type by passing a FileFormats.FileFormat enum. For example:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The Conic Benchmark Format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_CBF)\nA Conic Benchmark Format (CBF) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The LP file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_LP)\nA .LP-file model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The MathOptFormat file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The MPS file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The NL file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_NL)\nAn AMPL (.nl) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The REW file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_REW)\nA Mathematical Programming System (MPS) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Note that the REW format is identical to the MPS file format, except that all names are replaced with generic identifiers.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"The SDPA file format","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> model = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_SDPA)\nA SemiDefinite Programming Algorithm Format (SDPA) model","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Write-to-file","page":"Overview","title":"Write to file","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"To write a model src to a MathOptFormat file, use:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> MOI.add_variable(src)\nMOI.VariableIndex(1)\n\njulia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap with 1 entry:\n  MOI.VariableIndex(1) => MOI.VariableIndex(1)\n\njulia> MOI.write_to_file(dest, \"file.mof.json\")\n\njulia> print(read(\"file.mof.json\", String))\n{\n  \"name\": \"MathOptFormat Model\",\n  \"version\": {\n    \"major\": 1,\n    \"minor\": 5\n  },\n  \"variables\": [\n    {\n      \"name\": \"x1\"\n    }\n  ],\n  \"objective\": {\n    \"sense\": \"feasibility\"\n  },\n  \"constraints\": []\n}","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Read-from-file","page":"Overview","title":"Read from file","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"To read a MathOptFormat file, use:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MOF)\nA MathOptFormat Model\n\njulia> MOI.read_from_file(dest, \"file.mof.json\")\n\njulia> MOI.get(dest, MOI.ListOfVariableIndices())\n1-element Vector{MathOptInterface.VariableIndex}:\n MOI.VariableIndex(1)\n\njulia> rm(\"file.mof.json\")  # Clean up after ourselves.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Detecting-the-file-type-automatically","page":"Overview","title":"Detecting the file-type automatically","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Instead of the format keyword, you can also use the filename keyword argument to FileFormats.Model. This will attempt to automatically guess the format from the file extension. For example:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> MOI.write_to_file(dest, \"file.cbf.gz\")\n\njulia> src_2 = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> MOI.write_to_file(dest, \"file.cbf.gz\")\n\njulia> src_2 = MOI.FileFormats.Model(filename = \"file.cbf.gz\")\nA Conic Benchmark Format (CBF) model\n\njulia> MOI.read_from_file(src_2, \"file.cbf.gz\")\n\njulia> rm(\"file.cbf.gz\")  # Clean up after ourselves.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Note how the compression format (GZip) is also automatically detected from the filename.","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Unsupported-constraints","page":"Overview","title":"Unsupported constraints","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"In some cases src may contain constraints that are not supported by the file format (for example, the CBF format supports integer variables but not binary). If so, copy src to a bridged model using Bridges.full_bridge_optimizer:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"src = MOI.Utilities.Model{Float64}()\nx = MOI.add_variable(model)\nMOI.add_constraint(model, x, MOI.ZeroOne())\ndest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_CBF)\nbridged = MOI.Bridges.full_bridge_optimizer(dest, Float64)\nMOI.copy_to(bridged, src)\nMOI.write_to_file(dest, \"my_model.cbf\")","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"note: Note\nEven after bridging, it may still not be possible to write the model to file because of unsupported constraints (for example, PSD variables in the LP file format).","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Read-and-write-to-io","page":"Overview","title":"Read and write to io","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"In addition to write_to_file and read_from_file, you can read and write directly from IO streams using Base.write and Base.read!:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> src = MOI.Utilities.Model{Float64}()\nMOIU.Model{Float64}\n\njulia> dest = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model\n\njulia> MOI.copy_to(dest, src)\nMathOptInterface.Utilities.IndexMap()\n\njulia> io = IOBuffer();\n\njulia> write(io, dest)\n\njulia> seekstart(io);\n\njulia> src_2 = MOI.FileFormats.Model(format = MOI.FileFormats.FORMAT_MPS)\nA Mathematical Programming System (MPS) model\n\njulia> read!(io, src_2);","category":"page"},{"location":"moi/submodules/FileFormats/overview/#Validating-MOF-files","page":"Overview","title":"Validating MOF files","text":"","category":"section"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"MathOptFormat files are governed by a schema. Use JSONSchema.jl to check if a .mof.json file satisfies the schema.","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"First, construct the schema object as follows:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> import JSON, JSONSchema\n\njulia> schema = JSONSchema.Schema(JSON.parsefile(MOI.FileFormats.MOF.SCHEMA_PATH))\nA JSONSchema","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Then, check if a model file is valid using isvalid:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> good_model = JSON.parse(\"\"\"\n       {\n         \"version\": {\n           \"major\": 1,\n           \"minor\": 5\n         },\n         \"variables\": [{\"name\": \"x\"}],\n         \"objective\": {\"sense\": \"feasibility\"},\n         \"constraints\": []\n       }\n       \"\"\");\n\njulia> isvalid(schema, good_model)\ntrue","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"If we construct an invalid file, for example by mis-typing name as NaMe, the validation fails:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> bad_model = JSON.parse(\"\"\"\n       {\n         \"version\": {\n           \"major\": 1,\n           \"minor\": 5\n         },\n         \"variables\": [{\"NaMe\": \"x\"}],\n         \"objective\": {\"sense\": \"feasibility\"},\n         \"constraints\": []\n       }\n       \"\"\");\n\njulia> isvalid(schema, bad_model)\nfalse","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"Use JSONSchema.validate to obtain more insight into why the validation failed:","category":"page"},{"location":"moi/submodules/FileFormats/overview/","page":"Overview","title":"Overview","text":"julia> JSONSchema.validate(schema, bad_model)\nValidation failed:\npath:         [variables][1]\ninstance:     Dict{String, Any}(\"NaMe\" => \"x\")\nschema key:   required\nschema value: Any[\"name\"]","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"EditURL = \"https://github.com/lanl-ansi/Alpine.jl/blob/v0.5.5/README.md\"","category":"page"},{"location":"packages/Alpine/#Alpine,-a-global-solver-for-non-convex-MINLPs","page":"lanl-ansi/Alpine.jl","title":"Alpine, a global solver for non-convex MINLPs","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"(Image: CI) (Image: codecov) (Image: Documentation)  (Image: version)","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"ALPINE (glob(AL) o(P)timization for mixed-(I)nteger programs with (N)onlinear (E)quations), is a novel global optimization solver that uses an adaptive, piecewise convexification scheme and constraint programming methods to solve non-convex Mixed-Integer Non-Linear Programs (MINLPs) efficiently. MINLPs are typically \"hard\" optimization problems which appear in numerous applications (see MINLPLib.jl). ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine is entirely built upon JuMP and MathOptInterface in Julia, which provides incredible flexibility for usage and further development.","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine globally solves a given MINLP by:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Analyzing the problem's expressions (objective & constraints) and applies appropriate convex relaxations and polyhedral outer-approximations\nPerforming sequential optimization-based bound tightening (OBBT) and an iterative MIP-based adaptive partitioning scheme via piecewise polyhedral relaxations with a guarantee of global convergence","category":"page"},{"location":"packages/Alpine/#Installation","page":"lanl-ansi/Alpine.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Install Alpine using the Julia package manager:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"import Pkg\nPkg.add(\"Alpine\")","category":"page"},{"location":"packages/Alpine/#Usage-with-JuMP","page":"lanl-ansi/Alpine.jl","title":"Usage with JuMP","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Use Alpine with JuMP as follows:","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"using JuMP, Alpine, Ipopt, HiGHS\nipopt = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0)\nhighs = optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false)\nmodel = Model(\n    optimizer_with_attributes(\n        Juniper.Optimizer,\n        \"nlp_solver\" => ipopt,\n        \"mip_solver\" => highs,\n    ),\n)","category":"page"},{"location":"packages/Alpine/#Documentation","page":"lanl-ansi/Alpine.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"For more details, see the online documentation.","category":"page"},{"location":"packages/Alpine/#Support-problem-types","page":"lanl-ansi/Alpine.jl","title":"Support problem types","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Alpine can currently handle MINLPs with polynomials in constraints and/or in the objective. Currently, there is no support for exponential cones and Positive Semi-Definite (PSD) cones in MINLPs. Alpine is also a good fit for subsets of the MINLP family, for example, Mixed-Integer Quadratically Constrained Quadratic Programs (MIQCQPs), Non-Linear Programs (NLPs), etc.","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"For more details, check out this video on Alpine.jl at JuMP-dev 2018.","category":"page"},{"location":"packages/Alpine/#Underlying-solvers","page":"lanl-ansi/Alpine.jl","title":"Underlying solvers","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Though the MIP-based bounding algorithm implemented in Alpine is quite involved, most of the computational bottleneck arises in the underlying MIP solvers. Since every iteration of Alpine solves an MIP sub-problem, which is typically a convex MILP/MIQCQP, Alpine's run time heavily depends on the run-time of these solvers. For the best performance of Alpine, we recommend using the commercial solver Gurobi, which is available free for academic purposes. However, due to the flexibility offered by JuMP, the following MIP and NLP solvers are supported in Alpine: ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Solver Julia Package\nCPLEX CPLEX.jl\nCbc Cbc.jl\nGurobi Gurobi.jl\nIpopt Ipopt.jl\nBonmin Bonmin.jl\nArtelys KNITRO KNITRO.jl\nXpress Xpress.jl\nHiGHS HiGHS.jl","category":"page"},{"location":"packages/Alpine/#Bug-reports-and-support","page":"lanl-ansi/Alpine.jl","title":"Bug reports and support","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"Please report any issues via the GitHub issue tracker. All types of issues are welcome and encouraged; this includes bug reports, documentation typos, feature requests, etc. ","category":"page"},{"location":"packages/Alpine/#Challenging-Problems","page":"lanl-ansi/Alpine.jl","title":"Challenging Problems","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"We are seeking out hard benchmark instances for MINLPs. Please get in touch either by opening an issue or privately if you would like to share any hard instances.","category":"page"},{"location":"packages/Alpine/#Citing-Alpine","page":"lanl-ansi/Alpine.jl","title":"Citing Alpine","text":"","category":"section"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"If you find Alpine useful in your work, we kindly request that you cite the following papers (PDF, PDF)","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"@article{alpine_JOGO2019,\n  title = {An adaptive, multivariate partitioning algorithm for global optimization of nonconvex programs},\n  author = {Nagarajan, Harsha and Lu, Mowen and Wang, Site and Bent, Russell and Sundar, Kaarthik},\n  journal = {Journal of Global Optimization},\n  year = {2019},\n  issn = {1573-2916},\n  doi = {10.1007/s10898-018-00734-1},\n}\n\n@inproceedings{alpine_CP2016,\n  title = {Tightening {McCormick} relaxations for nonlinear programs via dynamic multivariate partitioning},\n  author = {Nagarajan, Harsha and Lu, Mowen and Yamangil, Emre and Bent, Russell},\n  booktitle = {International Conference on Principles and Practice of Constraint Programming},\n  pages = {369--387},\n  year = {2016},\n  organization = {Springer},\n  doi = {10.1007/978-3-319-44953-1_24},\n}","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"If you find the underlying piecewise polyhedral formulations implemented in Alpine useful in your work, we kindly request that you cite the following papers (link-1, link-2): ","category":"page"},{"location":"packages/Alpine/","page":"lanl-ansi/Alpine.jl","title":"lanl-ansi/Alpine.jl","text":"@article{alpine_ORL2021,\n  title = {Piecewise polyhedral formulations for a multilinear term},\n  author = {Sundar, Kaarthik and Nagarajan, Harsha and Linderoth, Jeff and Wang, Site and Bent, Russell},\n  journal = {Operations Research Letters},\n  volume = {49},\n  number = {1},\n  pages = {144--149},\n  year = {2021},\n  publisher = {Elsevier}\n}\n\n@article{alpine_OptOnline2022,\n  title={Piecewise Polyhedral Relaxations of Multilinear Optimization},\n  author={Kim, Jongeun and Richard, Jean-Philippe P. and Tawarmalani, Mohit},\n  eprinttype={Optimization Online},\n  date={2022}\n}","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"EditURL = \"https://github.com/odow/NEOSServer.jl/blob/v1.1.0/README.md\"","category":"page"},{"location":"packages/NEOSServer/#NEOSServer.jl","page":"odow/NEOSServer.jl","title":"NEOSServer.jl","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOSServer.jl is a wrapper for the NEOS Server, a free internet-based service for solving numerical optimization problems.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"See here for the full list of solvers and input formats that NEOS supports.","category":"page"},{"location":"packages/NEOSServer/#License","page":"odow/NEOSServer.jl","title":"License","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOSServer.jl is licensed under the MIT License.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"However, use of the NEOS Server requires you to comply with NEOS Server terms of use. In particular, the commercial solvers are to be used solely for academic, non-commercial research purposes.","category":"page"},{"location":"packages/NEOSServer/#Installation","page":"odow/NEOSServer.jl","title":"Installation","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Install NEOSServer.jl using the package manager:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"import Pkg\nPkg.add(\"NEOSServer\")","category":"page"},{"location":"packages/NEOSServer/#The-NEOS-API","page":"odow/NEOSServer.jl","title":"The NEOS API","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"This package contains an interface for the NEOS XML-RPC API.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"The following example shows how you can interact with the API. Wrapped XML-RPC functions begin with neos_ and are exported.","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"using NEOSServer\n\n# Create a server. You must supply a valid email:\nserver = NEOSServer.Server(\"me@mydomain.com\")\n\n# Print the NEOS welcome message:\nprintln(neos_welcome(server))\n\n# Get an XML template:\nxml_string = neos_getSolverTemplate(server, \"milp\", \"Cbc\", \"AMPL\")\n\n# Modify template with problem data...\n\n# Submit the XML job to NEOS:\njob = neos_submitJob(server, xml_string)\n\n# Get the status of the Job from NEOS:\nstatus = neos_getJobStatus(server, job)\n\n# Get the final results:\nresults = neos_getFinalResults(server, job)","category":"page"},{"location":"packages/NEOSServer/#Use-with-JuMP","page":"odow/NEOSServer.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Use NEOSServer.jl with JuMP as follows:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"using JuMP, NEOSServer\n\nmodel = Model() do\n    NEOSServer.Optimizer(email=\"me@mydomain.com\", solver=\"Ipopt\")\nend","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"Note: NEOSServer.Optimizer is limited to the following solvers:","category":"page"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"\"CPLEX\"\n\"FICO-Xpress\"\n\"Ipopt\"\n\"Knitro\"\n\"MOSEK\"\n\"OCTERACT\"\n\"SNOPT\"","category":"page"},{"location":"packages/NEOSServer/#NEOS-Limits","page":"odow/NEOSServer.jl","title":"NEOS Limits","text":"","category":"section"},{"location":"packages/NEOSServer/","page":"odow/NEOSServer.jl","title":"odow/NEOSServer.jl","text":"NEOS currently limits jobs to an 8 hour time limit, 3 GB of memory, and a 16 MB submission file. If your model exceeds these limits, NEOSServer.jl may be unable to return useful information to the user.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"EditURL = \"https://github.com/jump-dev/SDPA.jl/blob/v0.5.0/README.md\"","category":"page"},{"location":"packages/SDPA/#SDPA.jl","page":"jump-dev/SDPA.jl","title":"SDPA.jl","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA.jl is a wrapper for the SDPA semidefinite programming solver in double precision floating point arithmetic.","category":"page"},{"location":"packages/SDPA/#Affiliation","page":"jump-dev/SDPA.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"This wrapper is maintained by the JuMP community and is not a product of the SDPA developers.","category":"page"},{"location":"packages/SDPA/#License","page":"jump-dev/SDPA.jl","title":"License","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The underlying solver, SDPA is licensed under the GPL v2 license.","category":"page"},{"location":"packages/SDPA/#Installation","page":"jump-dev/SDPA.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Install SDPA using Pkg.add:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"import Pkg\nPkg.add(\"SDPA\")","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"In addition to installing the SDPA.jl package, this will also download and install the SDPA binaries. (You do not need to install SDPA separately.)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"If you see an error similar to:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"INFO: Precompiling module GZip.\nERROR: LoadError: LoadError: error compiling anonymous: could not load library \"libz\"","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"please see GZip.jl#54 or Flux.jl#343. In particular, in Ubuntu this issue may be resolved by running","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"sudo apt-get install zlib1g-dev","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"See SDPAFamily for the other solvers, SDPA-GMP, SDPA-DD, and SDPA-QD of the family.","category":"page"},{"location":"packages/SDPA/#Use-with-JuMP","page":"jump-dev/SDPA.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_DEFAULT)","category":"page"},{"location":"packages/SDPA/#MathOptInterface-API","page":"jump-dev/SDPA.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The SDPA optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.Nonnegatives\nMOI.PositiveSemidefiniteConeTriangle","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/SDPA/#Options","page":"jump-dev/SDPA.jl","title":"Options","text":"","category":"section"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"SDPA has three modes that give default value to all ten parameters.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The following table gives the default values for each parameter and mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Parameter PARAMETER_DEFAULT PARAMETER_UNSTABLE_BUT_FAST PARAMETER_STABLE_BUT_SLOW\nMaxIteration 100 100 1000\nEpsilonStar 1.0e-7 1.0e-7 1.0e-7\nLambdaStar 1.0e+2 1.0e+2 1.0e+4\nOmegaStar 2.0 2.0 2.0\nLowerBound 1.0e+5 1.0e+5 1.0e+5\nUpperBound 1.0e+5 1.0e+5 1.0e+5\nBetaStar 0.1 0.01 0.1\nBetaBar 0.2 0.02 0.3\nGammaStar 0.9 0.95 0.8\nEpsilonDash 1.0e-7 1.0e-7 1.0e-7","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"By default, we put SDPA in the SDPA.PARAMETER_DEFAULT mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Change the mode using the \"Mode\" option:","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_STABLE_BUT_SLOW)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"Note that the parameters are set in the order they are given, so you can set a mode and then modify parameters from this mode.","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"using JuMP, SDPA\nmodel = Model(SDPA.Optimizer)\nset_attribute(model, \"Mode\", SDPA.PARAMETER_STABLE_BUT_SLOW)\nset_attribute(model, \"MaxIteration\", 100)","category":"page"},{"location":"packages/SDPA/","page":"jump-dev/SDPA.jl","title":"jump-dev/SDPA.jl","text":"The choice of parameter mode has a large impact on the performance and stability of SDPA, and not necessarily in the way implied by the names of the modes; for example, PARAMETER_UNSTABLE_BUT_FAST can be more stable than the other modes for some problems. You should try each mode to see how it performs on your specific problem. See SDPA.jl#17 for more details.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/mathprogbase.md\"","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-from-MathProgBase","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface is a replacement for MathProgBase.jl. However, it is not a direct replacement.","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-a-solver-interface","page":"Transitioning from MathProgBase","title":"Transitioning a solver interface","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface is more extensive than MathProgBase which may make its implementation seem daunting at first. There are however numerous utilities in MathOptInterface that the simplify implementation process.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"For more information, read Implementing a solver interface.","category":"page"},{"location":"moi/tutorials/mathprogbase/#Transitioning-the-high-level-functions","page":"Transitioning from MathProgBase","title":"Transitioning the high-level functions","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"MathOptInterface doesn't provide replacements for the high-level interfaces in MathProgBase. We recommend you use JuMP as a modeling interface instead.","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"tip: Tip\nIf you haven't used JuMP before, start with the tutorial Getting started with JuMP","category":"page"},{"location":"moi/tutorials/mathprogbase/#linprog","page":"Transitioning from MathProgBase","title":"linprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from linprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction linprog(c, A, sense, b, l, u, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, l[i] <= x[i=1:N] <= u[i])\n    @objective(model, Min, c' * x)\n    eq_rows, ge_rows, le_rows = sense .== '=', sense .== '>', sense .== '<'\n    @constraint(model, A[eq_rows, :] * x .== b[eq_rows])\n    @constraint(model, A[ge_rows, :] * x .>= b[ge_rows])\n    @constraint(model, A[le_rows, :] * x .<= b[le_rows])\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/tutorials/mathprogbase/#mixintprog","page":"Transitioning from MathProgBase","title":"mixintprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from mixintprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction mixintprog(c, A, rowlb, rowub, vartypes, lb, ub, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, lb[i] <= x[i=1:N] <= ub[i])\n    for i in 1:N\n        if vartypes[i] == :Bin\n            set_binary(x[i])\n        elseif vartypes[i] == :Int\n            set_integer(x[i])\n        end\n    end\n    @objective(model, Min, c' * x)\n    @constraint(model, rowlb .<= A * x .<= rowub)\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/tutorials/mathprogbase/#quadprog","page":"Transitioning from MathProgBase","title":"quadprog","text":"","category":"section"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"Here is one way of transitioning from quadprog:","category":"page"},{"location":"moi/tutorials/mathprogbase/","page":"Transitioning from MathProgBase","title":"Transitioning from MathProgBase","text":"using JuMP\n\nfunction quadprog(c, Q, A, rowlb, rowub, lb, ub, solver)\n    N = length(c)\n    model = Model(solver)\n    @variable(model, lb[i] <= x[i=1:N] <= ub[i])\n    @objective(model, Min, c' * x + 0.5 * x' * Q * x)\n    @constraint(model, rowlb .<= A * x .<= rowub)\n    optimize!(model)\n    return (\n        status = termination_status(model),\n        objval = objective_value(model),\n        sol = value.(x)\n    )\nend","category":"page"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/reference.md\"","category":"page"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/reference/#Bridges","page":"API Reference","title":"Bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/#AbstractBridge-API","page":"API Reference","title":"AbstractBridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.AbstractBridge\nBridges.added_constrained_variable_types\nBridges.added_constraint_types\nget(::Bridges.AbstractBridge, ::NumberOfVariables)\nget(::Bridges.AbstractBridge, ::ListOfVariableIndices)\nget(::Bridges.AbstractBridge, ::NumberOfConstraints)\nget(::Bridges.AbstractBridge, ::ListOfConstraintIndices)\nBridges.needs_final_touch\nBridges.final_touch\nBridges.bridging_cost","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.AbstractBridge","text":"abstract type AbstractBridge end\n\nAn abstract type representing a bridged constraint or variable in a MOI.Bridges.AbstractBridgeOptimizer.\n\nAll bridges must implement:\n\nadded_constrained_variable_types\nadded_constraint_types\nMOI.get(::AbstractBridge, ::MOI.NumberOfVariables)\nMOI.get(::AbstractBridge, ::MOI.ListOfVariableIndices)\nMOI.get(::AbstractBridge, ::MOI.NumberOfConstraints)\nMOI.get(::AbstractBridge, ::MOI.ListOfConstraintIndices)\n\nSubtypes of AbstractBridge may have additional requirements. Consult their docstrings for details.\n\nIn addition, all subtypes may optionally implement the following constraint attributes with the bridge in place of the constraint index:\n\nMOI.ConstraintDual\nMOI.ConstraintPrimal\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.added_constrained_variable_types","page":"API Reference","title":"MathOptInterface.Bridges.added_constrained_variable_types","text":"added_constrained_variable_types(\n    BT::Type{<:AbstractBridge},\n)::Vector{Tuple{Type}}\n\nReturn a list of the types of constrained variables that bridges of concrete type BT add.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value. If the bridge may add a constrained variable, the type must be included in the return vector.\nIf the bridge adds a free variable via MOI.add_variable or MOI.add_variables, the return vector must include (MOI.Reals,).\n\nExample\n\njulia> MOI.Bridges.added_constrained_variable_types(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n       )\n1-element Vector{Tuple{Type}}:\n (MathOptInterface.Nonnegatives,)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.added_constraint_types","page":"API Reference","title":"MathOptInterface.Bridges.added_constraint_types","text":"added_constraint_types(\n    BT::Type{<:AbstractBridge},\n)::Vector{Tuple{Type,Type}}\n\nReturn a list of the types of constraints that bridges of concrete type BT add.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value. If the bridge may add a constraint, the type must be included in the return vector.\n\nExample\n\njulia> MOI.Bridges.added_constraint_types(\n           MOI.Bridges.Constraint.ZeroOneBridge{Float64},\n       )\n2-element Vector{Tuple{Type, Type}}:\n (MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.Interval{Float64})\n (MathOptInterface.VariableIndex, MathOptInterface.Integer)\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.NumberOfVariables}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.NumberOfVariables)::Int64\n\nReturn the number of variables created by the bridge b in the model.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.ListOfVariableIndices}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.ListOfVariableIndices)\n\nReturn the list of variables created by the bridge b.\n\nSee also MOI.ListOfVariableIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this if the bridge adds new variables.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.NumberOfConstraints}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.NumberOfConstraints{F,S})::Int64 where {F,S}\n\nReturn the number of constraints of the type F-in-S created by the bridge b.\n\nSee also MOI.NumberOfConstraints.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.get-Tuple{MathOptInterface.Bridges.AbstractBridge, MathOptInterface.ListOfConstraintIndices}","page":"API Reference","title":"MathOptInterface.get","text":"MOI.get(b::AbstractBridge, ::MOI.ListOfConstraintIndices{F,S}) where {F,S}\n\nReturn a  Vector{ConstraintIndex{F,S}} with indices of all constraints of type F-in-S created by the bride b.\n\nSee also MOI.ListOfConstraintIndices.\n\nImplementation notes\n\nThere is a default fallback, so you need only implement this for the constraint types returned by added_constraint_types.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.needs_final_touch","page":"API Reference","title":"MathOptInterface.Bridges.needs_final_touch","text":"needs_final_touch(bridge::AbstractBridge)::Bool\n\nReturn whether final_touch is implemented by bridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.final_touch","page":"API Reference","title":"MathOptInterface.Bridges.final_touch","text":"final_touch(bridge::AbstractBridge, model::MOI.ModelLike)::Nothing\n\nA function that is called immediately prior to MOI.optimize! to allow bridges to modify their reformulations with repsect to other variables and constraints in model.\n\nFor example, if the correctness of bridge depends on the bounds of a variable or the fact that variables are integer, then the bridge can implement final_touch to check assumptions immediately before a call to MOI.optimize!.\n\nIf you implement this method, you must also implement needs_final_touch.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridging_cost","page":"API Reference","title":"MathOptInterface.Bridges.bridging_cost","text":"bridging_cost(b::AbstractBridgeOptimizer, S::Type{<:MOI.AbstractSet}})\n\nReturn the cost of bridging variables constrained in S on creation, is_bridged(b, S) is assumed to be true.\n\nbridging_cost(\n    b::AbstractBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)\n\nReturn the cost of bridging F-in-S constraints.\n\nis_bridged(b, S) is assumed to be true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Constraint-bridge-API","page":"API Reference","title":"Constraint bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Constraint.AbstractBridge\nsupports_constraint(::Type{<:Bridges.Constraint.AbstractBridge}, ::Type{<:AbstractFunction}, ::Type{<:AbstractSet})\nBridges.Constraint.concrete_bridge_type\nBridges.Constraint.bridge_constraint\nBridges.Constraint.AbstractFunctionConversionBridge\nBridges.Constraint.SingleBridgeOptimizer\nBridges.Constraint.add_all_bridges\nBridges.Constraint.FlipSignBridge\nBridges.Constraint.AbstractToIntervalBridge\nBridges.Constraint.SetMapBridge\nBridges.Constraint.conversion_cost","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractType\n\nSubtype of MOI.Bridges.AbstractBridge for constraint bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nMOI.supports_constraint(::Type{<:AbstractBridge}, ::Type{<:MOI.AbstractFunction}, ::Type{<:MOI.AbstractSet})\nconcrete_bridge_type\nbridge_constraint\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.supports_constraint-Tuple{Type{<:MathOptInterface.Bridges.Constraint.AbstractBridge}, Type{<:MathOptInterface.AbstractFunction}, Type{<:MathOptInterface.AbstractSet}}","page":"API Reference","title":"MathOptInterface.supports_constraint","text":"MOI.supports_constraint(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging F-in-S constraints.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method for constraint types that the bridge implements.\n\n\n\n\n\n","category":"method"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet}\n)::Type\n\nReturn the concrete type of the bridge supporting F-in-S constraints.\n\nThis function can only be called if MOI.supports_constraint(BT, F, S) is true.\n\nExample\n\nThe SplitIntervalBridge bridges a MOI.VariableIndex-in-MOI.Interval constraint into a MOI.VariableIndex-in-MOI.GreaterThan and a MOI.VariableIndex-in-MOI.LessThan constraint.\n\njulia> MOI.Bridges.Constraint.concrete_bridge_type(\n           MOI.Bridges.Constraint.SplitIntervalBridge{Float64},\n           MOI.VariableIndex,\n           MOI.Interval{Float64},\n       )\nMathOptInterface.Bridges.Constraint.SplitIntervalBridge{Float64, MathOptInterface.VariableIndex, MathOptInterface.Interval{Float64}, MathOptInterface.GreaterThan{Float64}, MathOptInterface.LessThan{Float64}}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.bridge_constraint","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.bridge_constraint","text":"bridge_constraint(\n    BT::Type{<:AbstractBridge},\n    model::MOI.ModelLike,\n    func::AbstractFunction,\n    set::MOI.AbstractSet,\n)::BT\n\nBridge the constraint func-in-set using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT should be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractFunctionConversionBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractFunctionConversionBridge","text":"abstract type AbstractFunctionConversionBridge{F,S} <: AbstractBridge end\n\nAbstract type to support writing bridges in which the function changes but the set does not.\n\nBy convention, the transformed function is stored in the .constraint field.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn AbstractBridgeOptimizer that always bridges any objective function supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the objective function if it is supported by the bridge BT and unsupported by model.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Constraint.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Constraint.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Constraint.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 constraint bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.add_all_bridges","text":"add_all_bridges(bridged_model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Constraint submodule to bridged_model. The coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.FlipSignBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.FlipSignBridge","text":"FlipSignBridge{T,S1,S2,F,G}\n\nAn abstract type that simplifies the creation of other bridges.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.AbstractToIntervalBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.AbstractToIntervalBridge","text":"AbstractToIntervalBridge{T<:AbstractFloat,S,F}\n\nAn abstract type that simplifies the creation of other bridges.\n\nwarning: Warning\nT must be a AbstractFloat type because otherwise typemin and typemax would either be not implemented (e.g. BigInt), or would not give infinite value (e.g. Int). For this reason, this bridge is only added to MOI.Bridges.full_bridge_optimizer when T is a subtype of AbstractFloat.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.SetMapBridge","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.SetMapBridge","text":"abstract type SetMapBridge{T,S2,S1,F,G} <: AbstractBridge end\n\nConsider two type of sets, S1 and S2, and a linear mapping A such that the image of a set of type S1 under A is a set of type S2.\n\nA SetMapBridge{T,S2,S1,F,G} is a bridge that maps G-in-S1 constraints into F-in-S2 by mapping the function through A.\n\nThe linear map A is described by;\n\nMOI.Bridges.map_set\nMOI.Bridges.map_function.\n\nImplementing a method for these two functions is sufficient to bridge constraints. However, in order for the getters and setters of attributes such as dual solutions and starting values to work as well, a method for the following functions must be implemented:\n\nMOI.Bridges.inverse_map_set\nMOI.Bridges.inverse_map_function\nMOI.Bridges.adjoint_map_function\nMOI.Bridges.inverse_adjoint_map_function\n\nSee the docstrings of each function to see which feature would be missing if it was not implemented for a given bridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Constraint.conversion_cost","page":"API Reference","title":"MathOptInterface.Bridges.Constraint.conversion_cost","text":"conversion_cost(\n    F::Type{<:MOI.AbstractFunction},\n    G::Type{<:MOI.AbstractFunction},\n)::Float64\n\nReturn a Float64 returning the cost of converting any function of type G to a function of type F with convert.\n\nThis cost is used to compute MOI.Bridges.bridging_cost.\n\nThe default cost is Inf, which means that MOI.Bridges.Constraint.FunctionConversionBridge should not attempt the conversion.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Objective-bridge-API","page":"API Reference","title":"Objective bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Objective.AbstractBridge\nBridges.Objective.supports_objective_function\nBridges.set_objective_function_type\nBridges.Objective.concrete_bridge_type\nBridges.Objective.bridge_objective\nBridges.Objective.SingleBridgeOptimizer\nBridges.Objective.add_all_bridges","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Objective.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractBridge end\n\nSubtype of MOI.Bridges.AbstractBridge for objective bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nsupports_objective_function\nconcrete_bridge_type\nbridge_objective\nMOI.Bridges.set_objective_function_type\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.supports_objective_function","page":"API Reference","title":"MathOptInterface.Bridges.Objective.supports_objective_function","text":"supports_objective_function(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging objective functions of type F.\n\nImplementation notes\n\nThis method depends only on the type of the inputs, not the runtime values.\nThere is a default fallback, so you need only implement this method For objective functions that the bridge implements.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.set_objective_function_type","page":"API Reference","title":"MathOptInterface.Bridges.set_objective_function_type","text":"set_objective_function_type(\n    BT::Type{<:Objective.AbstractBridge},\n)::Type{<:MOI.AbstractFunction}\n\nReturn the type of objective function that bridges of concrete type BT set.\n\nImplementation notes\n\nThis method depends only on the type of the bridge, not the runtime value.\n\nExample\n\njulia> MOI.Bridges.set_objective_function_type(\n           MOI.Bridges.Objective.FunctionizeBridge{Float64},\n       )\nMathOptInterface.ScalarAffineFunction{Float64}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Objective.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    F::Type{<:MOI.AbstractFunction},\n)::Type\n\nReturn the concrete type of the bridge supporting objective functions of type F.\n\nThis function can only be called if MOI.supports_objective_function(BT, F) is true.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.bridge_objective","page":"API Reference","title":"MathOptInterface.Bridges.Objective.bridge_objective","text":"bridge_objective(\n    BT::Type{<:MOI.Bridges.Objective.AbstractBridge},\n    model::MOI.ModelLike,\n    func::MOI.AbstractFunction,\n)::BT\n\nBridge the objective function func using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT must be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Objective.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn AbstractBridgeOptimizer that always bridges any objective function supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the objective function if it is supported by the bridge BT and unsupported by model.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Objective.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Objective.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Objective.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Objective.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Objective.add_all_bridges","text":"add_all_bridges(model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Objective submodule to model.\n\nThe coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#ref_variable_bridges","page":"API Reference","title":"Variable bridge API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Variable.AbstractBridge\nBridges.Variable.supports_constrained_variable\nBridges.Variable.concrete_bridge_type\nBridges.Variable.bridge_constrained_variable\nBridges.Variable.SingleBridgeOptimizer\nBridges.Variable.add_all_bridges\nBridges.Variable.FlipSignBridge\nBridges.Variable.SetMapBridge\nBridges.Variable.unbridged_map","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.AbstractBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.AbstractBridge","text":"abstract type AbstractBridge <: MOI.Bridges.AbstractBridge end\n\nSubtype of MOI.Bridges.AbstractBridge for variable bridges.\n\nIn addition to the required implementation described in MOI.Bridges.AbstractBridge, subtypes of AbstractBridge must additionally implement:\n\nsupports_constrained_variable\nconcrete_bridge_type\nbridge_constrained_variable\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.supports_constrained_variable","page":"API Reference","title":"MathOptInterface.Bridges.Variable.supports_constrained_variable","text":"supports_constrained_variable(\n    BT::Type{<:AbstractBridge},\n    S::Type{<:MOI.AbstractSet},\n)::Bool\n\nReturn a Bool indicating whether the bridges of type BT support bridging constrained variables in S. That is, it returns true if the bridge of type BT converts constrained variables of type S into a form supported by the solver.\n\nImplementation notes\n\nThis method depends only on the type of the bridge and set, not the runtime values.\nThere is a default fallback, so you need only implement this method for sets that the bridge implements.\n\nExample\n\njulia> MOI.Bridges.Variable.supports_constrained_variable(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n           MOI.Nonpositives,\n       )\ntrue\n\njulia> MOI.Bridges.Variable.supports_constrained_variable(\n           MOI.Bridges.Variable.NonposToNonnegBridge{Float64},\n           MOI.Nonnegatives,\n       )\nfalse\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.concrete_bridge_type","page":"API Reference","title":"MathOptInterface.Bridges.Variable.concrete_bridge_type","text":"concrete_bridge_type(\n    BT::Type{<:AbstractBridge},\n    S::Type{<:MOI.AbstractSet},\n)::Type\n\nReturn the concrete type of the bridge supporting variables in S constraints.\n\nThis function can only be called if MOI.supports_constrained_variable(BT, S) is true.\n\nExamples\n\nAs a variable in MOI.GreaterThan is bridged into variables in MOI.Nonnegatives by the VectorizeBridge:\n\njulia> MOI.Bridges.Variable.concrete_bridge_type(\n           MOI.Bridges.Variable.VectorizeBridge{Float64},\n           MOI.GreaterThan{Float64},\n       )\nMathOptInterface.Bridges.Variable.VectorizeBridge{Float64, MathOptInterface.Nonnegatives}\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.bridge_constrained_variable","page":"API Reference","title":"MathOptInterface.Bridges.Variable.bridge_constrained_variable","text":"bridge_constrained_variable(\n    BT::Type{<:AbstractBridge},\n    model::MOI.ModelLike,\n    set::MOI.AbstractSet,\n)::BT\n\nBridge the constrained variable in set using bridge BT to model and returns a bridge object of type BT.\n\nImplementation notes\n\nThe bridge type BT must be a concrete type, that is, all the type parameters of the bridge must be set.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.SingleBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.Variable.SingleBridgeOptimizer","text":"SingleBridgeOptimizer{BT<:AbstractBridge}(model::MOI.ModelLike)\n\nReturn MOI.Bridges.AbstractBridgeOptimizer that always bridges any variables constrained on creation supported by the bridge BT.\n\nThis is in contrast with the MOI.Bridges.LazyBridgeOptimizer, which only bridges the variables constrained on creation if they are supported by the bridge BT and unsupported by model.\n\nwarning: Warning\nTwo SingleBridgeOptimizers cannot be used together as both of them assume that the underlying model only returns variable indices with nonnegative values. Use MOI.Bridges.LazyBridgeOptimizer instead.\n\nExample\n\njulia> struct MyNewBridge{T} <: MOI.Bridges.Variable.AbstractBridge end\n\njulia> bridge = MOI.Bridges.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}}(\n           MOI.Utilities.Model{Float64}(),\n       )\nMOIB.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 variable bridges\nwith inner model MOIU.Model{Float64}\n\nImplementation notes\n\nAll bridges should simplify the creation of SingleBridgeOptimizers by defining a constant that wraps the bridge in a SingleBridgeOptimizer.\n\njulia> const MyNewBridgeModel{T,OT<:MOI.ModelLike} =\n           MOI.Bridges.Variable.SingleBridgeOptimizer{MyNewBridge{T},OT};\n\nThis enables users to create bridged models as follows:\n\njulia> MyNewBridgeModel{Float64}(MOI.Utilities.Model{Float64}())\nMOIB.Variable.SingleBridgeOptimizer{MyNewBridge{Float64}, MOIU.Model{Float64}}\nwith 0 variable bridges\nwith inner model MOIU.Model{Float64}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.add_all_bridges","page":"API Reference","title":"MathOptInterface.Bridges.Variable.add_all_bridges","text":"add_all_bridges(model, ::Type{T}) where {T}\n\nAdd all bridges defined in the Bridges.Variable submodule to model.\n\nThe coefficient type used is T.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.FlipSignBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.FlipSignBridge","text":"abstract type FlipSignBridge{T,S1,S2} <: SetMapBridge{T,S2,S1} end\n\nAn abstract type that simplifies the creation of other bridges.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.SetMapBridge","page":"API Reference","title":"MathOptInterface.Bridges.Variable.SetMapBridge","text":"abstract type SetMapBridge{T,S1,S2} <: AbstractBridge end\n\nConsider two type of sets, S1 and S2, and a linear mapping A such that the image of a set of type S1 under A is a set of type S2.\n\nA SetMapBridge{T,S1,S2} is a bridge that substitutes constrained variables in S2 into the image through A of constrained variables in S1.\n\nThe linear map A is described by:\n\nMOI.Bridges.map_set\nMOI.Bridges.map_function\n\nImplementing a method for these two functions is sufficient to bridge constrained variables. However, in order for the getters and setters of attributes such as dual solutions and starting values to work as well, a method for the following functions must be implemented:\n\nMOI.Bridges.inverse_map_set\nMOI.Bridges.inverse_map_function\nMOI.Bridges.adjoint_map_function\nMOI.Bridges.inverse_adjoint_map_function.\n\nSee the docstrings of each function to see which feature would be missing if it was not implemented for a given bridge.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Variable.unbridged_map","page":"API Reference","title":"MathOptInterface.Bridges.Variable.unbridged_map","text":"unbridged_map(\n   bridge::MOI.Bridges.Variable.AbstractBridge,\n    vi::MOI.VariableIndex,\n)\n\nFor a bridged variable in a scalar set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vi.\n\nunbridged_map(\n    bridge::MOI.Bridges.Variable.AbstractBridge,\n    vis::Vector{MOI.VariableIndex},\n)\n\nFor a bridged variable in a vector set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vis. If this method is not implemented, it falls back to calling the following method for every variable of vis.\n\nunbridged_map(\n    bridge::MOI.Bridges.Variable.AbstractBridge,\n    vi::MOI.VariableIndex,\n    i::MOI.Bridges.IndexInVector,\n)\n\nFor a bridged variable in a vector set, return a tuple of pairs mapping the variables created by the bridge to an affine expression in terms of the bridged variable vi corresponding to the ith variable of the vector.\n\nIf there is no way to recover the expression in terms of the bridged variable(s) vi(s), return nothing. See ZerosBridge for an example of bridge returning nothing.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#AbstractBridgeOptimizer-API","page":"API Reference","title":"AbstractBridgeOptimizer API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.AbstractBridgeOptimizer\nBridges.bridged_variable_function\nBridges.unbridged_variable_function\nBridges.bridged_function\nBridges.supports_constraint_bridges\nBridges.recursive_model","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.AbstractBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.AbstractBridgeOptimizer","text":"abstract type AbstractBridgeOptimizer <: MOI.AbstractOptimizer end\n\nAn abstract type that implements generic functions for bridges.\n\nImplementation notes\n\nBy convention, the inner optimizer should be stored in a model field. If not, the optimizer must implement MOI.optimize!.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridged_variable_function","page":"API Reference","title":"MathOptInterface.Bridges.bridged_variable_function","text":"bridged_variable_function(\n    b::AbstractBridgeOptimizer,\n    vi::MOI.VariableIndex,\n)\n\nReturn a MOI.AbstractScalarFunction of variables of b.model that equals vi. That is, if the variable vi is bridged, it returns its expression in terms of the variables of b.model. Otherwise, it returns vi.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.unbridged_variable_function","page":"API Reference","title":"MathOptInterface.Bridges.unbridged_variable_function","text":"unbridged_variable_function(\n    b::AbstractBridgeOptimizer,\n    vi::MOI.VariableIndex,\n)\n\nReturn a MOI.AbstractScalarFunction of variables of b that equals vi. That is, if the variable vi is an internal variable of b.model created by a bridge but not visible to the user, it returns its expression in terms of the variables of bridged variables. Otherwise, it returns vi.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridged_function","page":"API Reference","title":"MathOptInterface.Bridges.bridged_function","text":"bridged_function(b::AbstractBridgeOptimizer, value)::typeof(value)\n\nSubstitute any bridged MOI.VariableIndex in value by an equivalent expression in terms of variables of b.model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.supports_constraint_bridges","page":"API Reference","title":"MathOptInterface.Bridges.supports_constraint_bridges","text":"supports_constraint_bridges(b::AbstractBridgeOptimizer)::Bool\n\nReturn a Bool indicating if b supports MOI.Bridges.Constraint.AbstractBridge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.recursive_model","page":"API Reference","title":"MathOptInterface.Bridges.recursive_model","text":"recursive_model(b::AbstractBridgeOptimizer)\n\nIf a variable, constraint, or objective is bridged, return the context of the inner variables. For most optimizers, this should be b.model.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#LazyBridgeOptimizer-API","page":"API Reference","title":"LazyBridgeOptimizer API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.LazyBridgeOptimizer\nBridges.full_bridge_optimizer\nBridges.ListOfNonstandardBridges\nBridges.add_bridge\nBridges.remove_bridge\nBridges.has_bridge\nBridges.print_active_bridges\nBridges.print_graph\nBridges.debug_supports_constraint\nBridges.debug_supports","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.LazyBridgeOptimizer","page":"API Reference","title":"MathOptInterface.Bridges.LazyBridgeOptimizer","text":"LazyBridgeOptimizer(model::MOI.ModelLike)\n\nThe LazyBridgeOptimizer is a bridge optimizer that supports multiple bridges, and only bridges things which are not supported by the internal model.\n\nInternally, the LazyBridgeOptimizer solves a shortest hyper-path problem to determine which bridges to use.\n\nIn general, you should use full_bridge_optimizer instead of this constructor because full_bridge_optimizer automatically adds a large number of supported bridges.\n\nSee also: add_bridge, remove_bridge, has_bridge and full_bridge_optimizer.\n\nExample\n\njulia> model = MOI.Bridges.LazyBridgeOptimizer(MOI.Utilities.Model{Float64}())\nMOIB.LazyBridgeOptimizer{MOIU.Model{Float64}}\nwith 0 variable bridges\nwith 0 constraint bridges\nwith 0 objective bridges\nwith inner model MOIU.Model{Float64}\n\njulia> MOI.Bridges.add_bridge(model, MOI.Bridges.Variable.FreeBridge{Float64})\n\njulia> MOI.Bridges.has_bridge(model, MOI.Bridges.Variable.FreeBridge{Float64})\ntrue\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.full_bridge_optimizer","page":"API Reference","title":"MathOptInterface.Bridges.full_bridge_optimizer","text":"full_bridge_optimizer(model::MOI.ModelLike, ::Type{T}) where {T}\n\nReturns a LazyBridgeOptimizer bridging model for every bridge defined in this package (see below for the few exceptions) and for the coefficient type T, as well as the bridges in the list returned by the ListOfNonstandardBridges attribute.\n\nExample\n\njulia> model = MOI.Utilities.Model{Float64}();\n\njulia> bridged_model = MOI.Bridges.full_bridge_optimizer(model, Float64);\n\nExceptions\n\nThe following bridges are not added by full_bridge_optimizer, except if they are in the list returned by the ListOfNonstandardBridges attribute:\n\nConstraint.SOCtoNonConvexQuadBridge\nConstraint.RSOCtoNonConvexQuadBridge](@ref)\nConstraint.SOCtoPSDBridge\nIf T is not a subtype of AbstractFloat, subtypes of Constraint.AbstractToIntervalBridge\nConstraint.GreaterToIntervalBridge\nConstraint.LessToIntervalBridge)\n\nSee the docstring of the each bridge for the reason they are not added.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ListOfNonstandardBridges","page":"API Reference","title":"MathOptInterface.Bridges.ListOfNonstandardBridges","text":"ListOfNonstandardBridges{T}() <: MOI.AbstractOptimizerAttribute\n\nAny optimizer can be wrapped in a LazyBridgeOptimizer using full_bridge_optimizer. However, by default LazyBridgeOptimizer uses a limited set of bridges that are:\n\nimplemented in MOI.Bridges\ngenerally applicable for all optimizers.\n\nFor some optimizers however, it is useful to add additional bridges, such as those that are implemented in external packages (e.g., within the solver package itself) or only apply in certain circumstances (e.g., Constraint.SOCtoNonConvexQuadBridge).\n\nSuch optimizers should implement the ListOfNonstandardBridges attribute to return a vector of bridge types that are added by full_bridge_optimizer in addition to the list of default bridges.\n\nNote that optimizers implementing ListOfNonstandardBridges may require package-specific functions or sets to be used if the non-standard bridges are not added. Therefore, you are recommended to use model = MOI.instantiate(Package.Optimizer; with_bridge_type = T) instead of model = MOI.instantiate(Package.Optimizer). See MOI.instantiate.\n\nExamples\n\nAn optimizer using a non-default bridge in MOI.Bridges\n\nSolvers supporting MOI.ScalarQuadraticFunction can support MOI.SecondOrderCone and MOI.RotatedSecondOrderCone by defining:\n\nfunction MOI.get(::MyQuadraticOptimizer, ::ListOfNonstandardBridges{Float64})\n    return Type[\n        MOI.Bridges.Constraint.SOCtoNonConvexQuadBridge{Float64},\n        MOI.Bridges.Constraint.RSOCtoNonConvexQuadBridge{Float64},\n    ]\nend\n\nAn optimizer defining an internal bridge\n\nSuppose an optimizer can exploit specific structure of a constraint, e.g., it can exploit the structure of the matrix A in the linear system of equations A * x = b.\n\nThe optimizer can define the function:\n\nstruct MatrixAffineFunction{T} <: MOI.AbstractVectorFunction\n    A::SomeStructuredMatrixType{T}\n    b::Vector{T}\nend\n\nand then a bridge\n\nstruct MatrixAffineFunctionBridge{T} <: MOI.Constraint.AbstractBridge\n    # ...\nend\n# ...\n\nfrom VectorAffineFunction{T} to the MatrixAffineFunction. Finally, it defines:\n\nfunction MOI.get(::Optimizer{T}, ::ListOfNonstandardBridges{T}) where {T}\n    return Type[MatrixAffineFunctionBridge{T}]\nend\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_bridge","page":"API Reference","title":"MathOptInterface.Bridges.add_bridge","text":"add_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nEnable the use of the bridges of type BT by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.remove_bridge","page":"API Reference","title":"MathOptInterface.Bridges.remove_bridge","text":"remove_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nDisable the use of the bridges of type BT by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.has_bridge","page":"API Reference","title":"MathOptInterface.Bridges.has_bridge","text":"has_bridge(b::LazyBridgeOptimizer, BT::Type{<:AbstractBridge})\n\nReturn a Bool indicating whether the bridges of type BT are used by b.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.print_active_bridges","page":"API Reference","title":"MathOptInterface.Bridges.print_active_bridges","text":"print_active_bridges([io::IO=stdout,] b::MOI.Bridges.LazyBridgeOptimizer)\n\nPrint the set of bridges that are active in the model b.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction}\n)\n\nPrint the set of bridges required for an objective function of type F.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet},\n)\n\nPrint the set of bridges required for a constraint of type F-in-S.\n\n\n\n\n\nprint_active_bridges(\n    [io::IO=stdout,]\n    b::MOI.Bridges.LazyBridgeOptimizer,\n    S::Type{<:MOI.AbstractSet}\n)\n\nPrint the set of bridges required for a variable constrained to set S.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.print_graph","page":"API Reference","title":"MathOptInterface.Bridges.print_graph","text":"print_graph([io::IO = stdout,] b::LazyBridgeOptimizer)\n\nPrint the hyper-graph containing all variable, constraint, and objective types that could be obtained by bridging the variables, constraints, and objectives that are present in the model by all the bridges added to b.\n\nEach node in the hyper-graph corresponds to a variable, constraint, or objective type.\n\nVariable nodes are indicated by [ ]\nConstraint nodes are indicated by ( )\nObjective nodes are indicated by | |\n\nThe number inside each pair of brackets is an index of the node in the hyper-graph.\n\nNote that this hyper-graph is the full list of possible transformations. When the bridged model is created, we select the shortest hyper-path(s) from this graph, so many nodes may be un-used.\n\nTo see which nodes are used, call print_active_bridges.\n\nFor more information, see Legat, B., Dowson, O., Garcia, J., and Lubin, M. (2020).  \"MathOptInterface: a data structure for mathematical optimization problems.\" URL: https://arxiv.org/abs/2002.03447\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.debug_supports_constraint","page":"API Reference","title":"MathOptInterface.Bridges.debug_supports_constraint","text":"debug_supports_constraint(\n    b::LazyBridgeOptimizer,\n    F::Type{<:MOI.AbstractFunction},\n    S::Type{<:MOI.AbstractSet};\n    io::IO = Base.stdout,\n)\n\nPrints to io explanations for the value of MOI.supports_constraint with the same arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.debug_supports","page":"API Reference","title":"MathOptInterface.Bridges.debug_supports","text":"debug_supports(\n    b::LazyBridgeOptimizer,\n    ::MOI.ObjectiveFunction{F};\n    io::IO = Base.stdout,\n) where F\n\nPrints to io explanations for the value of MOI.supports with the same arguments.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#constraint_set_map","page":"API Reference","title":"SetMap API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.map_set\nBridges.inverse_map_set\nBridges.map_function\nBridges.inverse_map_function\nBridges.adjoint_map_function\nBridges.inverse_adjoint_map_function","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.map_set","page":"API Reference","title":"MathOptInterface.Bridges.map_set","text":"map_set(::Type{BT}, set) where {BT}\n\nReturn the image of set through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for bridging the constraint and setting the MOI.ConstraintSet.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_map_set","page":"API Reference","title":"MathOptInterface.Bridges.inverse_map_set","text":"inverse_map_set(::Type{BT}, set) where {BT}\n\nReturn the preimage of set through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintSet.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.map_function","page":"API Reference","title":"MathOptInterface.Bridges.map_function","text":"map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintPrimal of variable bridges. For constraint bridges, this is used for bridging the constraint, setting the MOI.ConstraintFunction and MOI.ConstraintPrimalStart and modifying the function with MOI.modify.\n\nmap_function(::Type{BT}, func, i::IndexInVector) where {BT}\n\nReturn the scalar function at the ith index of the vector function that would be returned by map_function(BT, func) except that it may compute the ith element. This is used by bridged_function and for getting the MOI.VariablePrimal and MOI.VariablePrimalStart of variable bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_map_function","page":"API Reference","title":"MathOptInterface.Bridges.inverse_map_function","text":"inverse_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the inverse of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used by Variable.unbridged_map and for setting the MOI.VariablePrimalStart of variable bridges and for getting the MOI.ConstraintFunction, the MOI.ConstraintPrimal and the MOI.ConstraintPrimalStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.adjoint_map_function","page":"API Reference","title":"MathOptInterface.Bridges.adjoint_map_function","text":"adjoint_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the adjoint of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintDual and MOI.ConstraintDualStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.inverse_adjoint_map_function","page":"API Reference","title":"MathOptInterface.Bridges.inverse_adjoint_map_function","text":"inverse_adjoint_map_function(::Type{BT}, func) where {BT}\n\nReturn the image of func through the inverse of the adjoint of the linear map A defined in Variable.SetMapBridge and Constraint.SetMapBridge. This is used for getting the MOI.ConstraintDual of variable bridges and setting the MOI.ConstraintDualStart of constraint bridges.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#Bridging-graph-API","page":"API Reference","title":"Bridging graph API","text":"","category":"section"},{"location":"moi/submodules/Bridges/reference/","page":"API Reference","title":"API Reference","text":"Bridges.Graph\nBridges.VariableNode\nBridges.ConstraintNode\nBridges.ObjectiveNode\nBridges.Edge\nBridges.ObjectiveEdge\nBridges.add_node\nBridges.add_edge\nBridges.set_variable_constraint_node\nBridges.bridge_index\nBridges.is_variable_edge_best","category":"page"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Graph","page":"API Reference","title":"MathOptInterface.Bridges.Graph","text":"Graph()\n\nA type-stable datastructure for computing the shortest hyperpath problem.\n\nNodes\n\nThere are three types of nodes in the graph:\n\nVariableNode\nConstraintNode\nObjectiveNode\n\nAdd nodes to the graph using add_node.\n\nEdges\n\nThere are two types of edges in the graph:\n\nEdge\nObjectiveEdge\n\nAdd edges to the graph using add_edge.\n\nFor the ability to add a variable constrained  on creation as a free variable followed by a constraint, use set_variable_constraint_node.\n\nOptimal hyper-edges\n\nUse bridge_index to compute the minimum-cost bridge leaving a node.\n\nNote that bridge_index lazy runs a Bellman-Ford algorithm to compute the set of minimum cost edges. Thus, the first call to bridge_index after adding new nodes or edges will take longer than subsequent calls.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.VariableNode","page":"API Reference","title":"MathOptInterface.Bridges.VariableNode","text":"VariableNode(index::Int)\n\nA node in Graph representing a variable constrained on creation.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ConstraintNode","page":"API Reference","title":"MathOptInterface.Bridges.ConstraintNode","text":"ConstraintNode(index::Int)\n\nA node in Graph representing a constraint.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ObjectiveNode","page":"API Reference","title":"MathOptInterface.Bridges.ObjectiveNode","text":"ObjectiveNode(index::Int)\n\nA node in Graph representing an objective function.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.Edge","page":"API Reference","title":"MathOptInterface.Bridges.Edge","text":"Edge(\n    bridge_index::Int,\n    added_variables::Vector{VariableNode},\n    added_constraints::Vector{ConstraintNode},\n    cost::Float64 = 1.0,\n)\n\nReturn a new datastructure representing an edge in Graph that starts at a VariableNode or a ConstraintNode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.ObjectiveEdge","page":"API Reference","title":"MathOptInterface.Bridges.ObjectiveEdge","text":"ObjectiveEdge(\n    bridge_index::Int,\n    added_variables::Vector{VariableNode},\n    added_constraints::Vector{ConstraintNode},\n)\n\nReturn a new datastructure representing an edge in Graph that starts at an ObjectiveNode.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_node","page":"API Reference","title":"MathOptInterface.Bridges.add_node","text":"add_node(graph::Graph, ::Type{VariableNode})::VariableNode\nadd_node(graph::Graph, ::Type{ConstraintNode})::ConstraintNode\nadd_node(graph::Graph, ::Type{ObjectiveNode})::ObjectiveNode\n\nAdd a new node to graph.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.add_edge","page":"API Reference","title":"MathOptInterface.Bridges.add_edge","text":"add_edge(graph::Graph, node::VariableNode, edge::Edge)::Nothing\nadd_edge(graph::Graph, node::ConstraintNode, edge::Edge)::Nothing\nadd_edge(graph::Graph, node::ObjectiveNode, edge::ObjectiveEdge)::Nothing\n\nAdd edge to graph, where edge starts at node and connects to the nodes defined in edge.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.set_variable_constraint_node","page":"API Reference","title":"MathOptInterface.Bridges.set_variable_constraint_node","text":"set_variable_constraint_node(\n    graph::Graph,\n    variable_node::VariableNode,\n    constraint_node::ConstraintNode,\n    cost::Int,\n)\n\nAs an alternative to variable_node, add a virtual edge to graph that represents adding a free variable, followed by a constraint of type constraint_node, with bridging cost cost.\n\nWhy is this needed?\n\nVariables can either be added as a variable constrained on creation, or as a free variable which then has a constraint added to it.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.bridge_index","page":"API Reference","title":"MathOptInterface.Bridges.bridge_index","text":"bridge_index(graph::Graph, node::VariableNode)::Int\nbridge_index(graph::Graph, node::ConstraintNode)::Int\nbridge_index(graph::Graph, node::ObjectiveNode)::Int\n\nReturn the optimal index of the bridge to chose from node.\n\n\n\n\n\n","category":"function"},{"location":"moi/submodules/Bridges/reference/#MathOptInterface.Bridges.is_variable_edge_best","page":"API Reference","title":"MathOptInterface.Bridges.is_variable_edge_best","text":"is_variable_edge_best(graph::Graph, node::VariableNode)::Bool\n\nReturn a Bool indicating whether node should be added as a variable constrained on creation, or as a free variable followed by a constraint.\n\n\n\n\n\n","category":"function"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"EditURL = \"https://github.com/kocvara/Loraine.jl/blob/v0.1.2/README.md\"","category":"page"},{"location":"packages/Loraine/#Loraine.jl","page":"kocvara/Loraine.jl","title":"Loraine.jl","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Sweet Lor(r)aine, let the party carry on[1]...","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"[1]: https://www.youtube.com/watch?v=0D2wNf1lVrI","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine.jl is a Julia implementation of an interior point method algorithm for linear semidefinite optimization problems. ","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The special feature of Loraine is the iterative solver for linear systems. This is to be used for problems with (very) low rank solution matrix.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Standard (non-low-rank) problems and linear programs can be solved using the direct solver; then the user gets a standard IP method akin SDPT3.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"There is also a MATLAB version of the code at kocvara/Loraine.m.","category":"page"},{"location":"packages/Loraine/#License-and-Original-Contributors","page":"kocvara/Loraine.jl","title":"License and Original Contributors","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine is licensed under the MIT License.","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Loraine was developed by Soodeh Habibi and Michal Kočvara, University of Birmingham, and Michael Stingl, University of Erlangen, for H2020 ITN POEMA. ","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The JuMP interface was provided by Benoît Legat. His help is greatly acknowledged.","category":"page"},{"location":"packages/Loraine/#Installation","page":"kocvara/Loraine.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"Install Loraine using Pkg.add:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"import Pkg\nPkg.add(\"Loraine\")","category":"page"},{"location":"packages/Loraine/#Use-with-JuMP","page":"kocvara/Loraine.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"To use Loraine with JuMP, use Loraine.Optimizer:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"using JuMP, Loraine\nmodel = Model(Loraine.Optimizer)\nset_attribute(model, \"maxit\", 100)","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"To solve an SDP problem stored in SDPA format, do","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"using JuMP, Loraine\nmodel = read_from_file(\"examples/data/theta1.dat-s\")\nset_optimizer(model, Loraine.Optimizer)\noptimize!(model)","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"For more examples, the folder examples includes a few examples of how to use Loraine via JuMP; in particular, solve_sdpa.jl reads an SDP in the SDPA input format and solves it by Loraine. A few sample problems can be found in folder examples/data.","category":"page"},{"location":"packages/Loraine/#Rank-one-data","page":"kocvara/Loraine.jl","title":"Rank-one data","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"If the solution does not have low rank, it is recommended to use a direct  solver kit = 0. However, if you know that your data matrices are all rank-one,  use the option datarank = -1 to get a significant reduction in the complexity  (and CPU time). Examples of such problems are maxG11 and thetaG11 from the  SDPLIB collection.","category":"page"},{"location":"packages/Loraine/#Options","page":"kocvara/Loraine.jl","title":"Options","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"The list of options:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"kit             # kit = 0 for direct solver; kit = 1 for CG [0]\ntol_cg          # tolerance for CG solver [1.0e-2]\ntol_cg_up       # tolerance update [0.5]\ntol_cg_min      # minimal tolerance for CG solver [1.0e-6]\neDIMACS         # epsilon for DIMACS error stopping criterion [1.0e-5]\npreconditioner  # 0...no; 1...H_alpha; 2...H_beta; 4...hybrid [1]\nerank           # estimated rank [1]\naamat           # 0..A^TA; 1..diag(A^TA); 2..identity [2]\nverb            # 2..full output; 1..short output; 0..no output [1]\ndatarank        # 0..full rank matrices expected [0]\n                # -1..rank-1 matrices expected, converted to vectors, if possible\n                # (TBD) 1..vectors expected for low-rank data matrices\ninitpoint       # 0..Loraine heuristics, 1..SDPT3-like heuristics [0]\ntiming          # 1..yes, 0..no\nmaxit           # maximal number of global iterations [200]","category":"page"},{"location":"packages/Loraine/#Citing","page":"kocvara/Loraine.jl","title":"Citing","text":"","category":"section"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"If you find Loraine useful, please cite the following paper:","category":"page"},{"location":"packages/Loraine/","page":"kocvara/Loraine.jl","title":"kocvara/Loraine.jl","text":"@article{loraine2023,\n  title={Loraine-An interior-point solver for low-rank semidefinite programming},\n  author={Habibi, Soodeh and Ko{\\v{c}}vara, Michal and Stingl, Michael},\n  www={https://hal.science/hal-04076509/}\n  note={Preprint hal-04076509}\n  year={2023}\n}","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"EditURL = nothing","category":"page"},{"location":"api/JuMP.Containers/#DocumenterReference_JuMP.Containers","page":"JuMP.Containers","title":"JuMP.Containers","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"This page lists the public API of JuMP.Containers.","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"info: Info\nThis page is an unstructured list of the JuMP.Containers API. For a more structured overview, read the Manual or Tutorial parts of this documentation.","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"Load all of the public the API into the current scope with:","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"using JuMP.Containers","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"Alternatively, load only the module with:","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"import JuMP.Containers","category":"page"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"and then prefix all calls with JuMP.Containers. to create JuMP.Containers.<NAME>.","category":"page"},{"location":"api/JuMP.Containers/#DenseAxisArray","page":"JuMP.Containers","title":"DenseAxisArray","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.DenseAxisArray","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.DenseAxisArray","page":"JuMP.Containers","title":"JuMP.Containers.DenseAxisArray","text":"DenseAxisArray(data::Array{T, N}, axes...) where {T, N}\n\nConstruct a JuMP array with the underlying data specified by the data array and the given axes. Exactly N axes must be provided, and their lengths must match size(data) in the corresponding dimensions.\n\nExample\n\njulia> array = Containers.DenseAxisArray([1 2; 3 4], [:a, :b], 2:3)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 2:3\nAnd data, a 2×2 Matrix{Int64}:\n 1  2\n 3  4\n\njulia> array[:b, 3]\n4\n\n\n\n\n\nDenseAxisArray{T}(undef, axes...) where T\n\nConstruct an uninitialized DenseAxisArray with element-type T indexed over the given axes.\n\nExample\n\njulia> array = Containers.DenseAxisArray{Float64}(undef, [:a, :b], 1:2);\n\njulia> fill!(array, 1.0)\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 1:2\nAnd data, a 2×2 Matrix{Float64}:\n 1.0  1.0\n 1.0  1.0\n\njulia> array[:a, 2] = 5.0\n5.0\n\njulia> array[:a, 2]\n5.0\n\njulia> array\n2-dimensional DenseAxisArray{Float64,2,...} with index sets:\n    Dimension 1, [:a, :b]\n    Dimension 2, 1:2\nAnd data, a 2×2 Matrix{Float64}:\n 1.0  5.0\n 1.0  1.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#SparseAxisArray","page":"JuMP.Containers","title":"SparseAxisArray","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.SparseAxisArray","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.SparseAxisArray","page":"JuMP.Containers","title":"JuMP.Containers.SparseAxisArray","text":"struct SparseAxisArray{T,N,K<:NTuple{N, Any}} <: AbstractArray{T,N}\n    data::Dict{K,T}\nend\n\nN-dimensional array with elements of type T where only a subset of the entries are defined. The entries with indices idx = (i1, i2, ..., iN) in keys(data) has value data[idx]. Note that as opposed to SparseArrays.AbstractSparseArray, the missing entries are not assumed to be zero(T), they are simply not part of the array. This means that the result of map(f, sa::SparseAxisArray) or f.(sa::SparseAxisArray) has the same sparsity structure than sa even if f(zero(T)) is not zero.\n\nExample\n\njulia> dict = Dict((:a, 2) => 1.0, (:a, 3) => 2.0, (:b, 3) => 3.0)\nDict{Tuple{Symbol, Int64}, Float64} with 3 entries:\n  (:a, 3) => 2.0\n  (:b, 3) => 3.0\n  (:a, 2) => 1.0\n\njulia> array = Containers.SparseAxisArray(dict)\nSparseAxisArray{Float64, 2, Tuple{Symbol, Int64}} with 3 entries:\n  [a, 2]  =  1.0\n  [a, 3]  =  2.0\n  [b, 3]  =  3.0\n\njulia> array[:b, 3]\n3.0\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.@container","page":"JuMP.Containers","title":"Containers.@container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.@container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.@container","page":"JuMP.Containers","title":"JuMP.Containers.@container","text":"@container([i=..., j=..., ...], expr[, container = :Auto])\n\nCreate a container with indices i, j, ... and values given by expr that may depend on the value of the indices.\n\n@container(ref[i=..., j=..., ...], expr[, container = :Auto])\n\nSame as above but the container is assigned to the variable of name ref.\n\nThe type of container can be controlled by the container keyword.\n\nnote: Note\nWhen the index set is explicitly given as 1:n for any expression n, it is transformed to Base.OneTo(n) before being given to container.\n\nExample\n\njulia> Containers.@container([i = 1:3, j = 1:3], i + j)\n3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> I = 1:3\n1:3\n\njulia> Containers.@container(x[i = I, j = I], i + j);\n\njulia> x\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, 1:3\nAnd data, a 3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.@container([i = 2:3, j = 1:3], i + j)\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, Base.OneTo(3)\nAnd data, a 2×3 Matrix{Int64}:\n 3  4  5\n 4  5  6\n\njulia> Containers.@container([i = 1:3, j = 1:3; i <= j], i + j)\nSparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 6 entries:\n  [1, 1]  =  2\n  [1, 2]  =  3\n  [1, 3]  =  4\n  [2, 2]  =  4\n  [2, 3]  =  5\n  [3, 3]  =  6\n\n\n\n\n\n","category":"macro"},{"location":"api/JuMP.Containers/#Containers.container","page":"JuMP.Containers","title":"Containers.container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.container","page":"JuMP.Containers","title":"JuMP.Containers.container","text":"container(f::Function, indices[[, ::Type{C} = AutoContainerType], names])\n\nCreate a container of type C with index names names, indices indices and values at given indices given by f.\n\nIf the method with names is not specialized on Type{C}, it falls back to calling  container(f, indices, c) for backwards compatibility with containers not supporting index names.\n\nExample\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(Base.OneTo(3), Base.OneTo(3)))\n3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(1:3, 1:3))\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 1:3\n    Dimension 2, 1:3\nAnd data, a 3×3 Matrix{Int64}:\n 2  3  4\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.vectorized_product(2:3, Base.OneTo(3)))\n2-dimensional DenseAxisArray{Int64,2,...} with index sets:\n    Dimension 1, 2:3\n    Dimension 2, Base.OneTo(3)\nAnd data, a 2×3 Matrix{Int64}:\n 3  4  5\n 4  5  6\n\njulia> Containers.container((i, j) -> i + j, Containers.nested(() -> 1:3, i -> i:3, condition = (i, j) -> isodd(i) || isodd(j)))\nSparseAxisArray{Int64, 2, Tuple{Int64, Int64}} with 5 entries:\n  [1, 1]  =  2\n  [1, 2]  =  3\n  [1, 3]  =  4\n  [2, 3]  =  5\n  [3, 3]  =  6\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.rowtable","page":"JuMP.Containers","title":"Containers.rowtable","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.rowtable","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.rowtable","page":"JuMP.Containers","title":"JuMP.Containers.rowtable","text":"rowtable([f::Function=identity,] x; [header::Vector{Symbol} = Symbol[]])\n\nApplies the function f to all elements of the variable container x, returning the result as a Vector of NamedTuples, where header is a vector containing the corresponding axis names.\n\nIf x is an N-dimensional array, there must be N+1 names, so that the last name corresponds to the result of f(x[i]).\n\nIf header is left empty, then the default header is [:x1, :x2, ..., :xN, :y].\n\ninfo: Info\nA Vector of NamedTuples implements the Tables.jl interface, and so the result can be used as input for any function that consumes a 'Tables.jl' compatible source.\n\nExample\n\njulia> model = Model();\n\njulia> @variable(model, x[i=1:2, j=i:2] >= 0, start = i+j);\n\njulia> Containers.rowtable(start_value, x; header = [:i, :j, :start])\n3-element Vector{NamedTuple{(:i, :j, :start), Tuple{Int64, Int64, Float64}}}:\n (i = 1, j = 2, start = 3.0)\n (i = 1, j = 1, start = 2.0)\n (i = 2, j = 2, start = 4.0)\n\njulia> Containers.rowtable(x)\n3-element Vector{NamedTuple{(:x1, :x2, :y), Tuple{Int64, Int64, VariableRef}}}:\n (x1 = 1, x2 = 2, y = x[1,2])\n (x1 = 1, x2 = 1, y = x[1,1])\n (x1 = 2, x2 = 2, y = x[2,2])\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.default_container","page":"JuMP.Containers","title":"Containers.default_container","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.default_container","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.default_container","page":"JuMP.Containers","title":"JuMP.Containers.default_container","text":"default_container(indices)\n\nIf indices is a NestedIterator, return a SparseAxisArray. Otherwise, indices should be a VectorizedProductIterator and the function returns Array if all iterators of the product are Base.OneTo and retunrs DenseAxisArray otherwise.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.nested","page":"JuMP.Containers","title":"Containers.nested","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.nested","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.nested","page":"JuMP.Containers","title":"JuMP.Containers.nested","text":"nested(iterators...; condition = (args...) -> true)\n\nCreate a NestedIterator.\n\nExample\n\njulia> iterator = Containers.nested(\n           () -> 1:2,\n           (i,) -> [\"A\", \"B\"];\n           condition = (i, j) -> isodd(i) || j == \"B\",\n       );\n\njulia> collect(iterator)\n3-element Vector{Tuple{Int64, String}}:\n (1, \"A\")\n (1, \"B\")\n (2, \"B\")\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.vectorized_product","page":"JuMP.Containers","title":"Containers.vectorized_product","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.vectorized_product","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.vectorized_product","page":"JuMP.Containers","title":"JuMP.Containers.vectorized_product","text":"vectorized_product(iterators...)\n\nCreated a VectorizedProductIterator.\n\nExample\n\njulia> iterator = Containers.vectorized_product(1:2, [\"A\", \"B\"]);\n\njulia> collect(iterator)\n2×2 Matrix{Tuple{Int64, String}}:\n (1, \"A\")  (1, \"B\")\n (2, \"A\")  (2, \"B\")\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.build_ref_sets","page":"JuMP.Containers","title":"Containers.build_ref_sets","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.build_ref_sets","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.build_ref_sets","page":"JuMP.Containers","title":"JuMP.Containers.build_ref_sets","text":"build_ref_sets(_error::Function, expr)\n\nHelper function for macros to construct container objects.\n\nwarning: Warning\nThis function is for advanced users implementing JuMP extensions. See container_code for more details.\n\nArguments\n\n_error: a function that takes a String and throws an error, potentially annotating the input string with extra information such as from which macro it was thrown from. Use error if you do not want a modified error message.\nexpr: an Expr that specifies the container, e.g., :(x[i = 1:3, [:red, :blue], k = S; i + k <= 6])\n\nReturns\n\nindex_vars: a Vector{Any} of names for the index variables, e.g., [:i, gensym(), :k]. These may also be expressions, like :((i, j)) from a call like :(x[(i, j) in S]).\nindices: an iterator over the indices, for example, Containers.NestedIterator\n\nExample\n\nSee container_code for a worked example.\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.container_code","page":"JuMP.Containers","title":"Containers.container_code","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.container_code","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.container_code","page":"JuMP.Containers","title":"JuMP.Containers.container_code","text":"container_code(\n    index_vars::Vector{Any},\n    indices::Expr,\n    code,\n    requested_container::Union{Symbol,Expr},\n)\n\nUsed in macros to construct a call to container. This should be used in conjunction with build_ref_sets.\n\nArguments\n\nindex_vars::Vector{Any}: a vector of names for the indices of the container. These may also be expressions, like :((i, j)) from a call like :(x[(i, j) in S]).\nindices::Expr: an expression that evaluates to an iterator of the indices.\ncode: an expression or literal constant for the value to be stored in the container as a function of the named index_vars.\nrequested_container: passed to the third argument of container. For built-in JuMP types, choose one of :Array, :DenseAxisArray, :SparseAxisArray, or :Auto. For a user-defined container, this expression must evaluate to the correct type.\n\nwarning: Warning\nIn most cases, you should esc(code) before passing it to container_code.\n\nExample\n\njulia> macro foo(ref_sets, code)\n           index_vars, indices = Containers.build_ref_sets(error, ref_sets)\n           return Containers.container_code(\n               index_vars,\n               indices,\n               esc(code),\n               :Auto,\n            )\n       end\n@foo (macro with 1 method)\n\njulia> @foo(x[i=1:2, j=[\"A\", \"B\"]], j^i)\n2-dimensional DenseAxisArray{String,2,...} with index sets:\n    Dimension 1, Base.OneTo(2)\n    Dimension 2, [\"A\", \"B\"]\nAnd data, a 2×2 Matrix{String}:\n \"A\"   \"B\"\n \"AA\"  \"BB\"\n\n\n\n\n\n","category":"function"},{"location":"api/JuMP.Containers/#Containers.AutoContainerType","page":"JuMP.Containers","title":"Containers.AutoContainerType","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.AutoContainerType","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.AutoContainerType","page":"JuMP.Containers","title":"JuMP.Containers.AutoContainerType","text":"AutoContainerType\n\nPass AutoContainerType to container to let the container type be chosen based on the type of the indices using default_container.\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.NestedIterator","page":"JuMP.Containers","title":"Containers.NestedIterator","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.NestedIterator","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.NestedIterator","page":"JuMP.Containers","title":"JuMP.Containers.NestedIterator","text":"struct NestedIterator{T}\n    iterators::T # Tuple of functions\n    condition::Function\nend\n\nIterators over the tuples that are produced by a nested for loop.\n\nConstruct a NestedIterator using nested.\n\nExample\n\njulia> iterators = (() -> 1:2, (i,) -> [\"A\", \"B\"]);\n\njulia> condition = (i, j) -> isodd(i) || j == \"B\";\n\njulia> x = Containers.NestedIterator(iterators, condition);\n\njulia> for (i, j) in x\n           println((i, j))\n       end\n(1, \"A\")\n(1, \"B\")\n(2, \"B\")\n\nis the same as\n\njulia> for i in iterators[1]()\n           for j in iterators[2](i)\n               if condition(i, j)\n                   println((i, j))\n               end\n           end\n       end\n(1, \"A\")\n(1, \"B\")\n(2, \"B\")\n\n\n\n\n\n","category":"type"},{"location":"api/JuMP.Containers/#Containers.VectorizedProductIterator","page":"JuMP.Containers","title":"Containers.VectorizedProductIterator","text":"","category":"section"},{"location":"api/JuMP.Containers/","page":"JuMP.Containers","title":"JuMP.Containers","text":"JuMP.Containers.Containers.VectorizedProductIterator","category":"page"},{"location":"api/JuMP.Containers/#JuMP.Containers.VectorizedProductIterator","page":"JuMP.Containers","title":"JuMP.Containers.VectorizedProductIterator","text":"struct VectorizedProductIterator{T}\n    prod::Iterators.ProductIterator{T}\nend\n\nA wrapper type for Iterators.ProuctIterator that discards shape information and returns a Vector.\n\nConstruct a VectorizedProductIterator using vectorized_product.\n\n\n\n\n\n","category":"type"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"EditURL = \"multi_objective_knapsack.jl\"","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Multi-objective-knapsack","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The purpose of this tutorial is to demonstrate how to create and solve a multi-objective linear program. In addition, it demonstrates how to work with solvers which return multiple solutions.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Required-packages","page":"Multi-objective knapsack","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA\nimport Plots\nimport Test  #hide","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl is a package which implements a variety of algorithms for solving multi-objective optimization problems. Because it is a long package name, we import it instead as MOA.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Formulation","page":"Multi-objective knapsack","title":"Formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The knapsack problem is a classic problem in mixed-integer programming. Given a collection of items i in I, each of which has an associated weight, w_i, and profit, p_i, the knapsack problem determines which profit-maximizing subset of items to pack into a knapsack such that the total weight is less than a capacity c. The mathematical formulation is:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"beginaligned\nmax  sumlimits_i in I p_i x_i \ntextst   sumlimits_i in I w_i x_i le c\n x_i in 0 1  forall i in I\nendaligned","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"where x_i is 1 if we pack item i into the knapsack and 0 otherwise.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"For this tutorial, we extend the single-objective knapsack problem by adding another objective: given a desirability rating, r_i, we wish to maximize the total desirability of the items in our knapsack. Thus, our mathematical formulation is now:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"beginaligned\nmax  sumlimits_i in I p_i x_i \n      sumlimits_i in I r_i x_i \ntextst   sumlimits_i in I w_i x_i le c\n x_i in 0 1  forall i in I\nendaligned","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Data","page":"Multi-objective knapsack","title":"Data","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The data for this example was taken from vOptGeneric, and the original author was @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"profit = [77, 94, 71, 63, 96, 82, 85, 75, 72, 91, 99, 63, 84, 87, 79, 94, 90]\ndesire = [65, 90, 90, 77, 95, 84, 70, 94, 66, 92, 74, 97, 60, 60, 65, 97, 93]\nweight = [80, 87, 68, 72, 66, 77, 99, 85, 70, 93, 98, 72, 100, 89, 67, 86, 91]\ncapacity = 900\nN = length(profit)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Comparing the capacity to the total weight of all the items:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"capacity / sum(weight)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"shows that we can take approximately 64% of the items.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Plotting the items, we see that there are a range of items with different profits and desirability. Some items have a high profit and a high desirability, others have a low profit and a high desirability (and vice versa).","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Plots.scatter(\n    profit,\n    desire;\n    xlabel = \"Profit\",\n    ylabel = \"Desire\",\n    legend = false,\n)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The goal of the bi-objective knapsack problem is to choose a subset which maximizes both objectives.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#JuMP-formulation","page":"Multi-objective knapsack","title":"JuMP formulation","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Our JuMP formulation is a direct translation of the mathematical formulation:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"model = Model()\n@variable(model, x[1:N], Bin)\n@constraint(model, sum(weight[i] * x[i] for i in 1:N) <= capacity)\n@expression(model, profit_expr, sum(profit[i] * x[i] for i in 1:N))\n@expression(model, desire_expr, sum(desire[i] * x[i] for i in 1:N))\n@objective(model, Max, [profit_expr, desire_expr])","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Note how we form a multi-objective program by passing a vector of scalar objective functions.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Solution","page":"Multi-objective knapsack","title":"Solution","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"To solve our model, we need an optimizer which supports multi-objective linear programs. One option is to use the MultiObjectiveAlgorithms.jl package.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"set_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_silent(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl supports many different algorithms for solving multiobjective optimization problems. One option is the epsilon-constraint method:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"set_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Let's solve the problem and see the solution","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"There are 9 solutions available. We can also use result_count to see how many solutions are available:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"result_count(model)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Accessing-multiple-solutions","page":"Multi-objective knapsack","title":"Accessing multiple solutions","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Access the nine different solutions in the model using the result keyword to solution_summary, value, and objective_value:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"solution_summary(model; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"objective_value(model; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Note that because we set a vector of two objective functions, the objective value is a vector with two elements. We can also query the value of each objective separately:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"value(profit_expr; result = 5)","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Visualizing-objective-space","page":"Multi-objective knapsack","title":"Visualizing objective space","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Unlike single-objective optimization problems, multi-objective optimization problems do not have a single optimal solution. Instead, the solutions returned represent possible trade-offs that the decision maker can choose between the two objectives. A common way to visualize this is by plotting the objective values of each of the solutions:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"plot = Plots.scatter(\n    [value(profit_expr; result = i) for i in 1:result_count(model)],\n    [value(desire_expr; result = i) for i in 1:result_count(model)];\n    xlabel = \"Profit\",\n    ylabel = \"Desire\",\n    title = \"Objective space\",\n    label = \"\",\n    xlims = (915, 960),\n)\nfor i in 1:result_count(model)\n    y = objective_value(model; result = i)\n    Plots.annotate!(y[1] - 1, y[2], (i, 10))\nend\nideal_point = objective_bound(model)\nPlots.scatter!([ideal_point[1]], [ideal_point[2]]; label = \"Ideal point\")","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"Visualizing the objective space lets the decision maker choose a solution that suits their personal preferences. For example, result #7 is close to the maximum value of profit, but offers significantly higher desirability compared with solutions #8 and #9.","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"The set of items that are chosen in solution #7 are:","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"items_chosen = [i for i in 1:N if value(x[i]; result = 7) > 0.9]","category":"page"},{"location":"tutorials/linear/multi_objective_knapsack/#Next-steps","page":"Multi-objective knapsack","title":"Next steps","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_knapsack/","page":"Multi-objective knapsack","title":"Multi-objective knapsack","text":"MultiObjectiveAlgorithms.jl implements a number of different algorithms. Try solving the same problem using MOA.Dichotomy(). Does it find the same solution?","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"EditURL = \"https://github.com/chkwon/PATHSolver.jl/blob/v1.6.1/README.md\"","category":"page"},{"location":"packages/PATHSolver/#PATHSolver.jl","page":"chkwon/PATHSolver.jl","title":"PATHSolver.jl","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl is a wrapper for the PATH solver.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"a thin wrapper around the C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"You can solve any complementarity problem using the wrapper around the C API, although you must manually provide the callback functions, including the Jacobian.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The MathOptInterface wrapper is more limited, supporting only linear complementarity problems, but it enables PATHSolver to be used with JuMP.","category":"page"},{"location":"packages/PATHSolver/#Affiliation","page":"chkwon/PATHSolver.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of PATH. However, we are in close contact with the PATH developers, and they have given us permission to re-distribute the PATH binaries for automatic installation.","category":"page"},{"location":"packages/PATHSolver/#License","page":"chkwon/PATHSolver.jl","title":"License","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl is licensed under the MIT License.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The underlying solver, path is closed source and requires a license.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Without a license, the PATH Solver can solve problem instances up to with up to 300 variables and 2000 non-zeros. For larger problems, this web page provides a temporary license that is valid for a year.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"You can either store the license in the PATH_LICENSE_STRING environment variable, or you can use the PATHSolver.c_api_License_SetString function immediately after importing the PATHSolver package:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"import PATHSolver\nPATHSolver.c_api_License_SetString(\"<LICENSE STRING>\")","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"where <LICENSE STRING> is replaced by the current license string.","category":"page"},{"location":"packages/PATHSolver/#Installation","page":"chkwon/PATHSolver.jl","title":"Installation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Install PATHSolver.jl as follows:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"import Pkg\nPkg.add(\"PATHSolver\")","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default, PATHSolver.jl will download a copy of the underlying PATH solver. To use a different version of PATH, see the Manual Installation section below.","category":"page"},{"location":"packages/PATHSolver/#Use-with-JuMP","page":"chkwon/PATHSolver.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"julia> using JuMP, PATHSolver\n\njulia> M = [\n           0  0 -1 -1\n           0  0  1 -2\n           1 -1  2 -2\n           1  2 -2  4\n       ]\n4×4 Array{Int64,2}:\n 0   0  -1  -1\n 0   0   1  -2\n 1  -1   2  -2\n 1   2  -2   4\n\njulia> q = [2, 2, -2, -6]\n4-element Array{Int64,1}:\n  2\n  2\n -2\n -6\n\njulia> model = Model(PATHSolver.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: Path 5.0.00\n\njulia> set_optimizer_attribute(model, \"output\", \"no\")\n\njulia> @variable(model, x[1:4] >= 0)\n4-element Array{VariableRef,1}:\n x[1]\n x[2]\n x[3]\n x[4]\n\njulia> @constraint(model, M * x .+ q ⟂ x)\n[-x[3] - x[4] + 2, x[3] - 2 x[4] + 2, x[1] - x[2] + 2 x[3] - 2 x[4] - 2, x[1] + 2 x[2] - 2 x[3] + 4 x[4] - 6, x[1], x[2], x[3], x[4]] ∈ MOI.Complements(4)\n\njulia> optimize!(model)\nReading options file /var/folders/bg/dzq_hhvx1dxgy6gb5510pxj80000gn/T/tmpiSsCRO\nRead of options file complete.\n\nPath 5.0.00 (Mon Aug 19 10:57:18 2019)\nWritten by Todd Munson, Steven Dirkse, Youngdae Kim, and Michael Ferris\n\njulia> value.(x)\n4-element Array{Float64,1}:\n 2.8\n 0.0\n 0.7999999999999998\n 1.2\n\njulia> termination_status(model)\nLOCALLY_SOLVED::TerminationStatusCode = 4","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Note that options are set using JuMP.set_optimizer_attribute.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The list of options supported by PATH can be found here: https://pages.cs.wisc.edu/~ferris/path/options.pdf","category":"page"},{"location":"packages/PATHSolver/#MathOptInterface-API","page":"chkwon/PATHSolver.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"The Path 5.0.03 optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.Reals","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VectorOfVariables in MOI.Complements\nMOI.VectorAffineFunction{Float64} in MOI.Complements\nMOI.VectorQuadraticFunction{Float64} in MOI.Complements\nMOI.VectorNonlinearFunction in MOI.Complements","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"MOI.Name()","category":"page"},{"location":"packages/PATHSolver/#Use-with-the-C-API","page":"chkwon/PATHSolver.jl","title":"Use with the C API","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATHSolver.jl wraps the PATH C API using PATHSolver.c_api_XXX for the C method XXX. However, using the C API directly from Julia can be challenging, particularly with respect to avoiding issues with Julia's garbage collector.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Instead, we recommend that you use the PATHSolver.solve_mcp function, which wrappers the C API into a single call. See the docstring of PATHSolver.solve_mcp for a detailed description of the arguments.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"Here is the same example using PATHSolver.solve_mcp. Note that you must manually construct the sparse Jacobian callback.","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"julia> import PATHSolver\n\njulia> M = [\n           0  0 -1 -1\n           0  0  1 -2\n           1 -1  2 -2\n           1  2 -2  4\n       ]\n4×4 Matrix{Int64}:\n 0   0  -1  -1\n 0   0   1  -2\n 1  -1   2  -2\n 1   2  -2   4\n\njulia> q = [2, 2, -2, -6]\n4-element Vector{Int64}:\n  2\n  2\n -2\n -6\n\njulia> function F(n::Cint, x::Vector{Cdouble}, f::Vector{Cdouble})\n           @assert n == length(x) == length(f)\n           f .= M * x .+ q\n           return Cint(0)\n       end\nF (generic function with 1 method)\n\njulia> function J(\n           n::Cint,\n           nnz::Cint,\n           x::Vector{Cdouble},\n           col::Vector{Cint},\n           len::Vector{Cint},\n           row::Vector{Cint},\n           data::Vector{Cdouble},\n       )\n           @assert n == length(x) == length(col) == length(len) == 4\n           @assert nnz == length(row) == length(data)\n           i = 1\n           for c in 1:n\n               col[c], len[c] = i, 0\n               for r in 1:n\n                   if !iszero(M[r, c])\n                       row[i], data[i] = r, M[r, c]\n                       len[c] += 1\n                       i += 1\n                   end\n               end\n           end\n           return Cint(0)\n       end\nJ (generic function with 1 method)\n\njulia> status, z, info = PATHSolver.solve_mcp(\n           F,\n           J,\n           fill(0.0, 4),  # Lower bounds\n           fill(Inf, 4),  # Upper bounds\n           fill(0.0, 4);  # Starting point\n           nnz = 12,      # Number of nonzeros in the Jacobian\n           output = \"yes\",\n       )\nReading options file /var/folders/bg/dzq_hhvx1dxgy6gb5510pxj80000gn/T/jl_iftYBS\n > output yes\nRead of options file complete.\n\nPath 5.0.03 (Fri Jun 26 09:58:07 2020)\nWritten by Todd Munson, Steven Dirkse, Youngdae Kim, and Michael Ferris\n\nCrash Log\nmajor  func  diff  size  residual    step       prox   (label)\n    0     0             1.2649e+01             0.0e+00 (f[    4])\n    1     2     4     2 1.0535e+01  8.0e-01    0.0e+00 (f[    1])\n    2     3     2     4 8.4815e-01  1.0e+00    0.0e+00 (f[    4])\n    3     4     0     3 4.4409e-16  1.0e+00    0.0e+00 (f[    3])\npn_search terminated: no basis change.\n\nMajor Iteration Log\nmajor minor  func  grad  residual    step  type prox    inorm  (label)\n    0     0     5     4 4.4409e-16           I 0.0e+00 4.4e-16 (f[    3])\n\nMajor Iterations. . . . 0\nMinor Iterations. . . . 0\nRestarts. . . . . . . . 0\nCrash Iterations. . . . 3\nGradient Steps. . . . . 0\nFunction Evaluations. . 5\nGradient Evaluations. . 4\nBasis Time. . . . . . . 0.000016\nTotal Time. . . . . . . 0.044383\nResidual. . . . . . . . 4.440892e-16\n(PATHSolver.MCP_Solved, [2.8, 0.0, 0.8, 1.2], PATHSolver.Information(4.4408920985006247e-16, 0.0, 0.0, 0.044383, 1.6e-5, 0.0, 0, 0, 3, 5, 4, 0, 0, 0, 0, false, false, false, true, false, false, false))\n\njulia> status\nMCP_Solved::MCP_Termination = 1\n\njulia> z\n4-element Vector{Float64}:\n 2.8\n 0.0\n 0.8\n 1.2","category":"page"},{"location":"packages/PATHSolver/#Thread-safety","page":"chkwon/PATHSolver.jl","title":"Thread safety","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"PATH is not thread-safe and there are no known work-arounds. Do not run it in parallel using Threads.@threads. See issue #62 for more details.","category":"page"},{"location":"packages/PATHSolver/#Factorization-methods","page":"chkwon/PATHSolver.jl","title":"Factorization methods","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default, PATHSolver.jl will download the LUSOL shared library. To use LUSOL, set the following options:","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"model = Model(PATHSolver.Optimizer)\nset_optimizer_attribute(model, \"factorization_method\", \"blu_lusol\")\nset_optimizer_attribute(model, \"factorization_library_name\", PATHSolver.LUSOL_LIBRARY_PATH)","category":"page"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"To use factorization_method umfpack you will need the umfpack shared library that is available directly from the developers of that code for academic use.","category":"page"},{"location":"packages/PATHSolver/#Manual-installation","page":"chkwon/PATHSolver.jl","title":"Manual installation","text":"","category":"section"},{"location":"packages/PATHSolver/","page":"chkwon/PATHSolver.jl","title":"chkwon/PATHSolver.jl","text":"By default PATHSolver.jl will download a copy of the libpath library. If you already have one installed and want to use that, set the PATH_JL_LOCATION environment variable to point to the libpath50.xx library.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"EditURL = \"multi_objective_examples.jl\"","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Simple-multi-objective-examples","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial contains a number of examples of multi-objective programs from the literature.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Required-packages","page":"Simple multi-objective examples","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"using JuMP\nimport HiGHS\nimport MultiObjectiveAlgorithms as MOA","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-linear-problem","page":"Simple multi-objective examples","title":"Bi-objective linear problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Example 6.3 (from Steuer, 1985), page 154 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"model = Model()\nset_silent(model)\n@variable(model, x1 >= 0)\n@variable(model, 0 <= x2 <= 3)\n@objective(model, Min, [3x1 + x2, -x1 - 2x2])\n@constraint(model, 3x1 - x2 <= 6)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.Lexicographic())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    println(\"x = \", value.([x1, x2]; result = i))\nend","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-linear-assignment-problem","page":"Simple multi-objective examples","title":"Bi-objective linear assignment problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Example 9.38 (from Ulungu and Teghem, 1994), page 255 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"C1 = [5 1 4 7; 6 2 2 6; 2 8 4 4; 3 5 7 1]\nC2 = [3 6 4 2; 1 3 8 3; 5 2 2 3; 4 2 3 5]\nn = size(C2, 1)\nmodel = Model()\nset_silent(model)\n@variable(model, x[1:n, 1:n], Bin)\n@objective(model, Min, [sum(C1 .* x), sum(C2 .* x)])\n@constraint(model, [i = 1:n], sum(x[i, :]) == 1)\n@constraint(model, [j = 1:n], sum(x[:, j]) == 1)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    println(\"x = \", round.(Int, value.(x; result = i)))\nend","category":"page"},{"location":"tutorials/linear/multi_objective_examples/#Bi-objective-shortest-path-problem","page":"Simple multi-objective examples","title":"Bi-objective shortest path problem","text":"","category":"section"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"This example is taken from Exercise 9.5 page 269 of Ehrgott, M. (2005). Multicriteria Optimization. Springer, Berlin. The code was adapted from an example in vOptGeneric by @xgandibleux.","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"M = 50\nC1 = [\n    M 4 5 M M M\n    M M 2 1 2 7\n    M M M 5 2 M\n    M M 5 M M 3\n    M M M M M 4\n    M M M M M M\n]\nC2 = [\n    M 3 1 M M M\n    M M 1 4 2 2\n    M M M 1 7 M\n    M M 1 M M 2\n    M M M M M 2\n    M M M M M M\n]\nn = size(C2, 1)\nmodel = Model()\nset_silent(model)\n@variable(model, x[1:n, 1:n], Bin)\n@objective(model, Min, [sum(C1 .* x), sum(C2 .* x)])\n@constraint(model, sum(x[1, :]) == 1)\n@constraint(model, sum(x[:, n]) == 1)\n@constraint(model, [i = 2:n-1], sum(x[i, :]) - sum(x[:, i]) == 0)\nset_optimizer(model, () -> MOA.Optimizer(HiGHS.Optimizer))\nset_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())\noptimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/linear/multi_objective_examples/","page":"Simple multi-objective examples","title":"Simple multi-objective examples","text":"for i in 1:result_count(model)\n    print(i, \": z = \", round.(Int, objective_value(model; result = i)), \" | \")\n    X = round.(Int, value.(x; result = i))\n    print(\"Path:\")\n    for ind in findall(val -> val ≈ 1, X)\n        i, j = ind.I\n        print(\" $i->$j\")\n    end\n    println()\nend","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"EditURL = \"network_flows.jl\"","category":"page"},{"location":"tutorials/linear/network_flows/#Network-flow-problems","page":"Network flow problems","title":"Network flow problems","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial was originally contributed by Arpit Bhatia.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"In graph theory, a flow network (also known as a transportation network) is a directed graph where each edge has a capacity and each edge receives a flow. The amount of flow on an edge cannot exceed the capacity of the edge.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Often in operations research, a directed graph is called a network, the vertices are called nodes and the edges are called arcs.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"A flow must satisfy the restriction that the amount of flow into a node equals the amount of flow out of it,  unless it is a source, which has only outgoing flow, or sink, which has only incoming flow.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"A network can be used to model traffic in a computer network, circulation with demands, fluids in pipes,  currents in an electrical circuit, or anything similar in which something travels through a network of nodes.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/network_flows/#The-shortest-path-problem","page":"Network flow problems","title":"The shortest path problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Suppose that each arc (i j) of a graph is assigned a scalar cost a_ij, and suppose that we define the cost of a forward path to be the sum of the costs of its arcs.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Given a pair of nodes, the shortest path problem is to find a forward path that connects these nodes and has minimum cost.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmin  sum_forall e(ij) in E a_ij times x_ij \nst  sum_j x_ij - sum_k x_ki = b_i  forall i\n x_e in 01  forall e in E\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"where b_i is 1 if i is the starting node, -1 if i is the ending node, and 0 otherwise.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    0 100 30 0 0\n    0 0 20 0 0\n    0 0 0 10 60\n    0 15 0 0 50\n    0 0 0 0 0\n]\nn = size(G)[1]\nb = [1, -1, 0, 0, 0]\nshortest_path = Model(HiGHS.Optimizer)\nset_silent(shortest_path)\n@variable(shortest_path, x[1:n, 1:n], Bin)\n# Arcs with zero cost are not a part of the path as they do no exist\n@constraint(shortest_path, [i = 1:n, j = 1:n; G[i, j] == 0], x[i, j] == 0)\n# Flow conservation constraint\n@constraint(shortest_path, [i = 1:n], sum(x[i, :]) - sum(x[:, i]) == b[i],)\n@objective(shortest_path, Min, sum(G .* x))\noptimize!(shortest_path)\nobjective_value(shortest_path)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(x)","category":"page"},{"location":"tutorials/linear/network_flows/#The-assignment-problem","page":"Network flow problems","title":"The assignment problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Suppose that there are n persons and n objects that we have to match on a one-to-one basis. There is a benefit or value a_ij for matching person i with object j, and we want to assign persons to objects so as to maximize the total benefit.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"There is also a restriction that person i can be assigned to object j only if (i j) belongs to a given set of pairs A.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"Mathematically, we want to find a set of person-object pairs (1 j_1) (n j_n) from A such that the objects j_1j_n are all distinct, and the total benefit sum_i=1^y a_ij_i is maximized.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmax  sum_(ij) in A a_ij times y_ij \nst  sum_j(ij) in A y_ij = 1  forall i = 12n \n sum_i(ij) in A y_ij = 1  forall j = 12n \n y_ij in 01  forall (ij) in 12k\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    6 4 5 0\n    0 3 6 0\n    5 0 4 3\n    7 5 5 5\n]\nn = size(G)[1]\nassignment = Model(HiGHS.Optimizer)\nset_silent(assignment)\n@variable(assignment, y[1:n, 1:n], Bin)\n# One person can only be assigned to one object\n@constraint(assignment, [i = 1:n], sum(y[:, i]) == 1)\n# One object can only be assigned to one person\n@constraint(assignment, [j = 1:n], sum(y[j, :]) == 1)\n@objective(assignment, Max, sum(G .* y))\noptimize!(assignment)\nobjective_value(assignment)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(y)","category":"page"},{"location":"tutorials/linear/network_flows/#The-max-flow-problem","page":"Network flow problems","title":"The max-flow problem","text":"","category":"section"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"In the max-flow problem, we have a graph with two special nodes: the source, denoted by s, and the sink, denoted by t.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"The objective is to move as much flow as possible from s into t while observing the capacity constraints.","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"beginaligned\nmax  sum_v(sv) in E f(sv) \nst  sum_u(uv) in E f(uv)  = sum_w(vw) in E f(vw)  forall v in V - st \n f(uv) leq c(uv)  forall (uv) in E \n f(uv) geq 0  forall (uv) in E\nendaligned","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"G = [\n    0 3 2 2 0 0 0 0\n    0 0 0 0 5 1 0 0\n    0 0 0 0 1 3 1 0\n    0 0 0 0 0 1 0 0\n    0 0 0 0 0 0 0 4\n    0 0 0 0 0 0 0 2\n    0 0 0 0 0 0 0 4\n    0 0 0 0 0 0 0 0\n]\nn = size(G)[1]\nmax_flow = Model(HiGHS.Optimizer)\n@variable(max_flow, f[1:n, 1:n] >= 0)\n# Capacity constraints\n@constraint(max_flow, [i = 1:n, j = 1:n], f[i, j] <= G[i, j])\n# Flow conservation constraints\n@constraint(max_flow, [i = 1:n; i != 1 && i != 8], sum(f[i, :]) == sum(f[:, i]))\n@objective(max_flow, Max, sum(f[1, :]))\noptimize!(max_flow)\nobjective_value(max_flow)","category":"page"},{"location":"tutorials/linear/network_flows/","page":"Network flow problems","title":"Network flow problems","text":"value.(f)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"EditURL = \"optimal_power_flow.jl\"","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Optimal-power-flow","page":"Optimal power flow","title":"Optimal power flow","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial was originally contributed by James Foster (@jd-foster).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial formulates and solves an alternating current optimal power flow (AC-OPF) problem, a much-studied nonlinear problem from the field of electrical engineering.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Once we've formulated and solved the nonlinear problem, we will turn our focus to obtaining a good estimate of the objective value at the global optimum through the use of semidefinite programming.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"One main purpose of this tutorial is to highlight JuMP's ability to directly formulate problems involving complex-valued decision variables and complex matrix cones such as the  HermitianPSDCone object.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For another example of modeling with complex decision variables, see the Quantum state discrimination tutorial, and see the Complex number support section of the manual for more details.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial takes a matrix-oriented approach focused on network nodes that simplifies the construction of semidefinite programs. Another approach is to formulate the problem focusing on network lines where it is easier to work with flow constraints. A general approach is provided by the Julia/JuMP package PowerModels.jl, an open-source framework to a broad range of power flow model formulations along with utilities for working with detailed network data.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Required-packages","page":"Optimal power flow","title":"Required packages","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"using JuMP\nimport Clarabel\nimport DataFrames\nimport Ipopt\nimport LinearAlgebra\nimport SparseArrays","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Initial-formulation","page":"Optimal power flow","title":"Initial formulation","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Optimal power flow problems for electrical transmission typically pose the following question: what is the most cost-effective operation of electricity generators while meeting constraints on the safe limits of network components?","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We'll use the 9-node network test case case9mod to explore this problem. The graph of the network, shown here, has three nodes (or buses) each for the different purposes of generation G (nodes 1, 2, and 3), trans-shipment (nodes 4, 6, and 8), and demand D (nodes 5, 7, and 9).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(Image: Nine Nodes)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This example is a modified version of the MATPOWER test case case9 (archive) created by Bukhsh et al. (2013) for their test case archive of optimal power flow problems with local optima. This test case is also extensively evaluated in Krasko and Rebennack (2017).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Here bus and network node are taken as analogous terms, as are branch and transmission line.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For future reference, let's name the number of nodes in the network:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"N = 9;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The network data can be summarised using a small number of arrays. Using the sparsevec function from the SparseArrays standard library package, we can give the indices and values of the non-zero data points:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"# Real generation: lower (`lb`) and upper (`ub`) bounds\nP_Gen_lb = SparseArrays.sparsevec([1, 2, 3], [10, 10, 10], N)\nP_Gen_ub = SparseArrays.sparsevec([1, 2, 3], [250, 300, 270], N)\n# Reactive generation: lower (`lb`) and upper (`ub`) bounds\nQ_Gen_lb = SparseArrays.sparsevec([1, 2, 3], [-5, -5, -5], N)\nQ_Gen_ub = SparseArrays.sparsevec([1, 2, 3], [300, 300, 300], N)\n# Power demand levels (real, reactive, and complex form)\nP_Demand = SparseArrays.sparsevec([5, 7, 9], [54, 60, 75], N)\nQ_Demand = SparseArrays.sparsevec([5, 7, 9], [18, 21, 30], N)\nS_Demand = P_Demand + im * Q_Demand","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The key decision variables are the real power injections P^G and reactive power injections Q^Gover the allowed range of the generators. All other buses must restrict their generation variables to 0. On the other hand, these non-generator nodes have a fixed  real and reactive power demand, denoted P^D and Q^D respectively (these are fixed at 0 in the case of trans-shipment and generator nodes).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The cost of operating each generator is modeled as a quadratic function of its real power output; in our specific test case, the objective function to minimize is:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginalign\n    min          011  (P^G_1)^2 +   5 P^G_1 + 150  \n               +  0085  (P^G_2)^2 + 12 P^G_2 + 600  \n               + 01225   (P^G_3)^2 +     P^G_3 + 335 \nendalign","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's create an initial JuMP model with some of this data:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)\n@variable(model, P_Gen_lb[i] <= P_G[i in 1:N] <= P_Gen_ub[i])\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Even before solving an optimization problem, we can estimate a lower bound on the best objective value by substituting the lower bound on each generator's real power range (all 10, as it turns out in this case):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"basic_lower_bound = value(lower_bound, objective_function(model));\nprintln(\"Objective value (basic lower bound) : $basic_lower_bound\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"to see that we can do no better than an objective cost of 1188.75.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(Direct substitution works because a quadratic function of a single variable x with positive coefficients is strictly increasing for all x geq 0.)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"In fact, we can get a quick but even better estimate from the direct observation that the real power generated must meet or exceed the real power demand:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, sum(P_G) >= sum(P_Demand))\noptimize!(model)\nbetter_lower_bound = round(objective_value(model); digits = 2)\nprintln(\"Objective value (better lower bound): $better_lower_bound\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"However, there are additional power flow constraints that must be satisfied.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Power must flow from one or more generation nodes through the transmission lines and end up at a demand node. The state variables of our steady-state alternating current (AC) electrical network are complex-valued voltage variables V_1 ldots V_N. Voltages capture both a magnitude and phase of the node's electrical state in relation to the rest of the system. An AC power system also extends the notion of resistance in wires found in a direct current (DC) circuit to a  complex quantity, known as the impedance, of each transmission line. The reciprocal of impedance is known as admittance. Together, these complex quantities are used to express a complex version of Ohm's law: current flow through a line is proportional to the difference in voltages on each end of the line, multiplied by the admittance.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Network-data","page":"Optimal power flow","title":"Network data","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's assemble the data we need for writing the complex power flow constraints. The data for the problem consists of a list of the real and imaginary parts of the line impedance. We obtain the following data table from the branch data section of the case9mod MATPOWER format file:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"branch_data = DataFrames.DataFrame([\n    (1, 4, 0.0, 0.0576, 0.0),\n    (4, 5, 0.017, 0.092, 0.158),\n    (6, 5, 0.039, 0.17, 0.358),\n    (3, 6, 0.0, 0.0586, 0.0),\n    (6, 7, 0.0119, 0.1008, 0.209),\n    (8, 7, 0.0085, 0.072, 0.149),\n    (2, 8, 0.0, 0.0625, 0.0),\n    (8, 9, 0.032, 0.161, 0.306),\n    (4, 9, 0.01, 0.085, 0.176),\n]);\nDataFrames.rename!(branch_data, [:F_BUS, :T_BUS, :BR_R, :BR_X, :BR_Bc])","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The first two columns describe the network, supplying the from and to connection points of the lines. The last three columns give the branch resistance, branch reactance and line-charging susceptance.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We will also need to reference the base_MVA number (used for re-scaling):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"base_MVA = 100;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and the number of lines:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"M = size(branch_data, 1)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"From the first two columns of the branch data table, we can create a sparse incidence matrix that simplifies handling of the network layout:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"A =\n    SparseArrays.sparse(branch_data.F_BUS, 1:M, 1, N, M) +\n    SparseArrays.sparse(branch_data.T_BUS, 1:M, -1, N, M)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We form the network impedance vector from the next two columns","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"z = (branch_data.BR_R .+ im * branch_data.BR_X) / base_MVA;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and calculate it's corresponding bus admittance matrix as","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y_0 = A * SparseArrays.spdiagm(1 ./ z) * A';\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"while the last column gives the branch line-charging susceptance","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"y_sh = 1 / 2 * (im * branch_data.BR_Bc) * base_MVA;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and leads to the shunt admittance matrix","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y_sh = SparseArrays.spdiagm(\n    LinearAlgebra.diag(A * SparseArrays.spdiagm(y_sh) * A'),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"(The construction of the shunt admittance matrix Y_sh looks somewhat more complicated than Y_0 because we only want to add the diagonal elements in the calculation; the line-charging is used only in the nodal voltage terms and not the line voltage terms.)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The full bus admittance matrix Y is then defined as","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Y = Y_0 + Y_sh;\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#JuMP-model","page":"Optimal power flow","title":"JuMP model","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Now we're ready to write the complex power flow constraints we need to more accurately model the electricity system.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We'll introduce a number of constraints that model both the physics and operational requirements.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's start by initializing a new model:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Ipopt.Optimizer)\nset_silent(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Then we'll create the nodal power generation variables:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@variable(\n    model,\n    S_G[i in 1:N] in ComplexPlane(),\n    lower_bound = P_Gen_lb[i] + Q_Gen_lb[i] * im,\n    upper_bound = P_Gen_ub[i] + Q_Gen_ub[i] * im,\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We need complex nodal voltages (the system state variables):","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@variable(model, V[1:N] in ComplexPlane(), start = 1.0 + 0.0im)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and operational constraints for maintaining voltage magnitude levels:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, [i in 1:N], 0.9^2 <= real(V[i])^2 + imag(V[i])^2 <= 1.1^2)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We also need to fix an origin or reference angle from which all other complex voltage angles (arguments) are determined. Here we will use node 1 as the nominated reference bus. Fixing the imaginary component of a reference bus to zero sets its complex voltage angle to 0, while constraining the real part to be non-negative disallows equivalent solutions that are just a reflection by 180 degrees:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, imag(V[1]) == 0);\n@constraint(model, real(V[1]) >= 0);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The power flow equations express a conservation of energy (power) principle, where power generated less the power consumed must balance the power exchanged with the network:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"@constraint(model, S_G - S_Demand .== V .* conj(Y * V))","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"As above, the objective function is a quadratic cost of real power:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"P_G = real(S_G)\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We're finally ready to solve our nonlinear AC-OPF problem:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"objval_solution = round(objective_value(model); digits = 2)\nprintln(\"Objective value (feasible solution) : $(objval_solution)\")","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The solution's power generation (in rectangular form) and complex voltage values (in polar form using degrees) are:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"DataFrames.DataFrame(;\n    Bus = 1:N,\n    ComplexPowerGen = round.(value.(S_G); digits = 2),\n    VoltageMagnitude = round.(abs.(value.(V)); digits = 2),\n    VoltageAngle_Deg = round.(rad2deg.(angle.(value.(V))); digits = 2),\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#Relaxations-and-better-objective-bounds","page":"Optimal power flow","title":"Relaxations and better objective bounds","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The Ipopt solver uses an interior-point algorithm. It has local optimality guarantees, but is unable to certify whether the solution is globally optimal. The solution we found is indeed globally optimal. The work to verify this has been done in Bukhsh et al. (2013) and Krasko and Rebennack (2017), and different solvers (such as Gurobi, SCIP and GLOMIQO) are also able to verify this.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The techniques of convex relaxations can also be used to improve on our current best lower bound:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"better_lower_bound","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"To this end, observe that the nonlinear constraints in the AC-OPF formulation are quadratic equalities for power flow along with quadratic voltage inequalities.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Let's linearize these constraints by first making the substitution W = V V^*, where:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W = V V^* quad iff  quad W_ii =  V_i ^2 quad W_ik = V_i  overlineV_k quad forall i  k in  1 ldots N ","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and where V^* is the conjugate transpose of V.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"On the face of it, this turns a quadratic voltage bound constraint like:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"v_L leq V_i ^2  leq v_U quad i  in  1 ldots N ","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"for some real v_L and v_U into a simple two-sided bound:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"v_L leq W_ii  leq v_U","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"while each quadratic expression for the nodal power term:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"    S^Node_i = V_i overline(YV)_i","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"becomes the linear combination:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"    S^Node_i = (E_ii Y^T) bullet W","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Here A bullet B = operatornametr(A^* B) is the Frobenius inner product of two complex matrices, while E_kn denotes the matrix unit with a single nonzero entry of 1 in row k and column n.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"E(k, n) = SparseArrays.sparse([k], [n], 1, N, N);\nnothing #hide","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Of course, we've shifted the nonlinearity into the equality constraint W = V V^*: it is this constraint we will now relax using a semidefinite programming approach.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We will make use of complex voltages and relax W = V V^* to;","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W succeq V V^*","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"where the relation succeq is the ordering in the Hermitian positive semidefinite cone.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"The above constraint is equivalent to:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginbmatrix 1  V^*  V  W  endbmatrix succeq 0","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"by the theory of the Schur complement. This matrix inequality implies a number of second-order cone constraints by taking certain 2 times 2 minors of the matrix for each i  in  1 ldots N :","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"beginbmatrix 1  V_i^*  V_i  W_ii  endbmatrix succeq 0","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"which is equivalent to the real second-order cone inequality:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"operatornamereal(W_ii) geq operatornamereal(V_i)^2 + operatornameimag(V_i)^2","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We include these implied constraints as well for demonstration purposes.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Putting it all together we get the following semidefinite relaxation of the AC-OPF problem:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"model = Model(Clarabel.Optimizer)\nset_attribute(model, \"tol_gap_rel\", 1e-3)\nset_attribute(model, \"tol_feas\", 1e-3)\nset_attribute(model, \"tol_ktratio\", 1e-3)\n@variable(\n    model,\n    S_G[i in 1:N] in ComplexPlane(),\n    lower_bound = P_Gen_lb[i] + Q_Gen_lb[i] * im,\n    upper_bound = P_Gen_ub[i] + Q_Gen_ub[i] * im,\n)\n@variable(model, W[1:N, 1:N] in HermitianPSDCone())\n@variable(model, V[1:N] in ComplexPlane(), start = 1.0 + 0.0im)\n@constraint(model, [i in 1:N], 0.9^2 <= real(W[i, i]) <= 1.1^2)\n@constraint(model, real(V[1]) >= 0)\n@constraint(model, imag(V[1]) == 0)\n@constraint(model, 0.9 <= real(V[1]) <= 1.1)\n@constraint(model, LinearAlgebra.Hermitian([1 V'; V W]) in HermitianPSDCone())\n# 2 x 2 minor inequalities:\n@constraint(\n    model,\n    [i in 1:N],\n    [0.5, real(W[i, i]), real(V[i]), imag(V[i])] in RotatedSecondOrderCone()\n)\n@constraint(\n    model,\n    [i in 1:N],\n    S_G[i] - S_Demand[i] == LinearAlgebra.tr((conj(Y) * E(i, i)) * W),\n)\nP_G = real(S_G)\n@objective(\n    model,\n    Min,\n    (0.11 * P_G[1]^2 + 5 * P_G[1] + 150) +\n    (0.085 * P_G[2]^2 + 1.2 * P_G[2] + 600) +\n    (0.1225 * P_G[3]^2 + P_G[3] + 335),\n)\noptimize!(model)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"sdp_relaxation_lower_bound = round(objective_value(model); digits = 2)\nprintln(\n    \"Objective value (W & V relax. lower bound): $sdp_relaxation_lower_bound\",\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"We can more easily see solution values by rounding out noisy data:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"W_1 = SparseArrays.sparse(round.(value.(W); digits = 2))","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"and recover an approximation to the voltage variables as:","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"DataFrames.DataFrame(;\n    Bus = 1:N,\n    Magnitude = round.(abs.(value.(V)); digits = 2),\n    AngleDeg = round.(rad2deg.(angle.(value.(V))); digits = 2),\n)","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"For further information on exploiting sparsity see Jabr (2012).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"This relaxation has the advantage that we can work directly with complex voltages to extend the formulation, strengthen the relaxation and gain additional approximate information about the voltage variables.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/#References-and-further-resources","page":"Optimal power flow","title":"References and further resources","text":"","category":"section"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Bukhsh, W. A., Grothey, A., McKinnon, K. I., & Trodden, P. A. Local solutions of the optimal power flow problem. IEEE Transactions on Power Systems, 28(4), 4780-4788 (2013).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Jabr, R. A. Exploiting sparsity in SDP relaxations of the OPF problem. IEEE Transactions on Power Systems, 27(2), 1138-1139 (2011).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Krasko, V., & S. Rebennack. Chapter 15: Global Optimization: Optimal Power Flow Problem. In Advances and Trends in Optimization with Engineering Applications, 187—205. MOS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics, 2017.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"Test case case9mod: from the Test Case Archive of Optimal Power Flow (OPF) Problems with Local Optima.","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"R. D. Zimmerman, C. E. Murillo-Sanchez, & R. J. Thomas, MATPOWER: Steady-State Operations, Planning and Analysis Tools for Power Systems Research and Education, IEEE Transactions on Power Systems, 26(1), 12-19 (2011).","category":"page"},{"location":"tutorials/applications/optimal_power_flow/","page":"Optimal power flow","title":"Optimal power flow","text":"MATPOWER data format: MATPOWER manual; see especially Appendix B \"Data File Format\" and Table B-3.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Bridges/list_of_bridges.md\"","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#List-of-bridges","page":"List of bridges","title":"List of bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"This section describes the Bridges.AbstractBridges that are implemented in MathOptInterface.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#constraint_bridges_ref","page":"List of bridges","title":"Constraint bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Constraint.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Constraint.GreaterToIntervalBridge\nBridges.Constraint.LessToIntervalBridge\nBridges.Constraint.GreaterToLessBridge\nBridges.Constraint.LessToGreaterBridge\nBridges.Constraint.NonnegToNonposBridge\nBridges.Constraint.NonposToNonnegBridge\nBridges.Constraint.VectorizeBridge\nBridges.Constraint.ScalarizeBridge\nBridges.Constraint.ScalarSlackBridge\nBridges.Constraint.VectorSlackBridge\nBridges.Constraint.ScalarFunctionizeBridge\nBridges.Constraint.VectorFunctionizeBridge\nBridges.Constraint.ToScalarQuadraticBridge\nBridges.Constraint.ToVectorQuadraticBridge\nBridges.Constraint.ToScalarNonlinearBridge\nBridges.Constraint.FunctionConversionBridge\nBridges.Constraint.SplitComplexEqualToBridge\nBridges.Constraint.SplitComplexZerosBridge\nBridges.Constraint.SplitHyperRectangleBridge\nBridges.Constraint.SplitIntervalBridge\nBridges.Constraint.SOCtoRSOCBridge\nBridges.Constraint.RSOCtoSOCBridge\nBridges.Constraint.SOCtoNonConvexQuadBridge\nBridges.Constraint.RSOCtoNonConvexQuadBridge\nBridges.Constraint.QuadtoSOCBridge\nBridges.Constraint.SOCtoPSDBridge\nBridges.Constraint.RSOCtoPSDBridge\nBridges.Constraint.NormInfinityBridge\nBridges.Constraint.NormOneBridge\nBridges.Constraint.NormToPowerBridge\nBridges.Constraint.NormOneConeToNormConeBridge\nBridges.Constraint.SecondOrderConeToNormConeBridge\nBridges.Constraint.NormInfinityConeToNormConeBridge\nBridges.Constraint.GeoMeantoRelEntrBridge\nBridges.Constraint.GeoMeanToPowerBridge\nBridges.Constraint.GeoMeanBridge\nBridges.Constraint.RelativeEntropyBridge\nBridges.Constraint.NormSpectralBridge\nBridges.Constraint.NormNuclearBridge\nBridges.Constraint.SquareBridge\nBridges.Constraint.HermitianToSymmetricPSDBridge\nBridges.Constraint.SetDotScalingBridge\nBridges.Constraint.SetDotInverseScalingBridge\nBridges.Constraint.RootDetBridge\nBridges.Constraint.LogDetBridge\nBridges.Constraint.IndicatorActiveOnFalseBridge\nBridges.Constraint.IndicatorGreaterToLessThanBridge\nBridges.Constraint.IndicatorLessToGreaterThanBridge\nBridges.Constraint.IndicatorSOS1Bridge\nBridges.Constraint.SemiToBinaryBridge\nBridges.Constraint.ZeroOneBridge\nBridges.Constraint.IntegerToZeroOneBridge\nBridges.Constraint.NumberConversionBridge\nBridges.Constraint.AllDifferentToCountDistinctBridge\nBridges.Constraint.ReifiedAllDifferentToCountDistinctBridge\nBridges.Constraint.BinPackingToMILPBridge\nBridges.Constraint.CircuitToMILPBridge\nBridges.Constraint.CountAtLeastToCountBelongsBridge\nBridges.Constraint.CountBelongsToMILPBridge\nBridges.Constraint.CountDistinctToMILPBridge\nBridges.Constraint.ReifiedCountDistinctToMILPBridge\nBridges.Constraint.CountGreaterThanToMILPBridge\nBridges.Constraint.TableToMILPBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GreaterToIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GreaterToIntervalBridge","text":"GreaterToIntervalBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nGreaterToIntervalBridge implements the following reformulations:\n\nf(x) ge l into f(x) in l infty)\n\nSource node\n\nGreaterToIntervalBridge supports:\n\nF in MOI.GreaterThan{T}\n\nTarget nodes\n\nGreaterToIntervalBridge creates:\n\nF in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LessToIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LessToIntervalBridge","text":"LessToIntervalBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nLessToIntervalBridge implements the following reformulations:\n\nf(x) le u into f(x) in (-infty u\n\nSource node\n\nLessToIntervalBridge supports:\n\nF in MOI.LessThan{T}\n\nTarget nodes\n\nLessToIntervalBridge creates:\n\nF in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GreaterToLessBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GreaterToLessBridge","text":"GreaterToLessBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nGreaterToLessBridge implements the following reformulation:\n\nf(x) ge l into -f(x) le -l\n\nSource node\n\nGreaterToLessBridge supports:\n\nG in MOI.GreaterThan{T}\n\nTarget nodes\n\nGreaterToLessBridge creates:\n\nF in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LessToGreaterBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LessToGreaterBridge","text":"LessToGreaterBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nLessToGreaterBridge implements the following reformulation:\n\nf(x) le u into -f(x) ge -u\n\nSource node\n\nLessToGreaterBridge supports:\n\nG in MOI.LessThan{T}\n\nTarget nodes\n\nLessToGreaterBridge creates:\n\nF in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NonnegToNonposBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NonnegToNonposBridge","text":"NonnegToNonposBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNonnegToNonposBridge implements the following reformulation:\n\nf(x) in mathbbR_+ into -f(x) in mathbbR_-\n\nSource node\n\nNonnegToNonposBridge supports:\n\nG in MOI.Nonnegatives\n\nTarget nodes\n\nNonnegToNonposBridge creates:\n\nF in MOI.Nonpositives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NonposToNonnegBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NonposToNonnegBridge","text":"NonposToNonnegBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNonposToNonnegBridge implements the following reformulation:\n\nf(x) in mathbbR_- into -f(x) in mathbbR_+\n\nSource node\n\nNonposToNonnegBridge supports:\n\nG in MOI.Nonpositives\n\nTarget nodes\n\nNonposToNonnegBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorizeBridge","text":"VectorizeBridge{T,F,S,G} <: Bridges.Constraint.AbstractBridge\n\nVectorizeBridge implements the following reformulations:\n\ng(x) ge a into g(x) - a in mathbbR_+\ng(x) le a into g(x) - a in mathbbR_-\ng(x) == a into g(x) - a in 0\n\nwhere T is the coefficient type of g(x) - a.\n\nSource node\n\nVectorizeBridge supports:\n\nG in MOI.GreaterThan{T}\nG in MOI.LessThan{T}\nG in MOI.EqualTo{T}\n\nTarget nodes\n\nVectorizeBridge creates:\n\nF in S, where S is one of MOI.Nonnegatives, MOI.Nonpositives, MOI.Zeros depending on the type of the input set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarizeBridge","text":"ScalarizeBridge{T,F,S}\n\nScalarizeBridge implements the following reformulations:\n\nf(x) - a in mathbbR_+ into f_i(x) ge a_i for all i\nf(x) - a in mathbbR_- into f_i(x) le a_i for all i\nf(x) - a in 0 into f_i(x) == a_i for all i\n\nSource node\n\nScalarizeBridge supports:\n\nG in MOI.Nonnegatives{T}\nG in MOI.Nonpositives{T}\nG in MOI.Zeros{T}\n\nTarget nodes\n\nScalarizeBridge creates:\n\nF in S, where S is one of MOI.GreaterThan{T}, MOI.LessThan{T}, and MOI.EqualTo{T}, depending on the type of the input set.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarSlackBridge","text":"ScalarSlackBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nScalarSlackBridge implements the following reformulation:\n\nf(x) in S into f(x) - y == 0 and y in S\n\nSource node\n\nScalarSlackBridge supports:\n\nG in S, where G is not MOI.VariableIndex and S is not MOI.EqualTo\n\nTarget nodes\n\nScalarSlackBridge creates:\n\nF in MOI.EqualTo{T}\nMOI.VariableIndex in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorSlackBridge","text":"VectorSlackBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nVectorSlackBridge implements the following reformulation:\n\nf(x) in S into f(x) - y in 0 and y in S\n\nSource node\n\nVectorSlackBridge supports:\n\nG in S, where G is not MOI.VectorOfVariables and S is not MOI.Zeros\n\nTarget nodes\n\nVectorSlackBridge creates:\n\nF in MOI.Zeros\nMOI.VectorOfVariables in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ScalarFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ScalarFunctionizeBridge","text":"ScalarFunctionizeBridge{T,S} = FunctionConversionBridge{T,MOI.ScalarAffineFunction{T},MOI.VariableIndex,S}\n\nScalarFunctionizeBridge implements the following reformulations:\n\nx in S into 1x + 0 in S\n\nSource node\n\nScalarFunctionizeBridge supports:\n\nMOI.VariableIndex in S\n\nTarget nodes\n\nScalarFunctionizeBridge creates:\n\nMOI.ScalarAffineFunction{T} in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.VectorFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.VectorFunctionizeBridge","text":"VectorFunctionizeBridge{T,S} = FunctionConversionBridge{T,MOI.VectorAffineFunction{T},S}\n\nVectorFunctionizeBridge implements the following reformulations:\n\nx in S into Ix + 0 in S\n\nSource node\n\nVectorFunctionizeBridge supports:\n\nMOI.VectorOfVariables in S\n\nTarget nodes\n\nVectorFunctionizeBridge creates:\n\nMOI.VectorAffineFunction{T} in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToScalarQuadraticBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToScalarQuadraticBridge","text":"ToScalarQuadraticBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToScalarQuadraticBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract scalar function and f is a MOI.ScalarQuadraticFunction.\n\nSource node\n\nToScalarQuadraticBridge supports:\n\nG<:AbstractScalarFunction in S\n\nTarget nodes\n\nToScalarQuadraticBridge creates:\n\nMOI.ScalarQuadraticFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToVectorQuadraticBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToVectorQuadraticBridge","text":"ToVectorQuadraticBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToVectorQuadraticBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract vector function and f is a MOI.VectorQuadraticFunction.\n\nSource node\n\nToVectorQuadraticBridge supports:\n\nG<:AbstractVectorFunction in S\n\nTarget nodes\n\nToVectorQuadraticBridge creates:\n\nMOI.VectorQuadraticFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ToScalarNonlinearBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ToScalarNonlinearBridge","text":"ToScalarNonlinearBridge{T,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nToScalarNonlinearBridge implements the following reformulation:\n\ng(x) in S into f(x) in S\n\nwhere g is an abstract scalar function and f is a MOI.ScalarNonlinearFunction.\n\nSource node\n\nToScalarNonlinearBridge supports:\n\nG<:AbstractScalarFunction in S\n\nTarget nodes\n\nToScalarNonlinearBridge creates:\n\nMOI.ScalarNonlinearFunction in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.FunctionConversionBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.FunctionConversionBridge","text":"FunctionConversionBridge{T,F,G,S} <: AbstractFunctionConversionBridge{G,S}\n\nFunctionConversionBridge implements the following reformulations:\n\ng(x) in S into f(x) in S\n\nfor these pairs of functions:\n\nMOI.ScalarAffineFunctionto [MOI.ScalarQuadraticFunction`](@ref)\nMOI.ScalarQuadraticFunction  to MOI.ScalarNonlinearFunction\nMOI.VectorAffineFunction to MOI.VectorQuadraticFunction\n\nSource node\n\nFunctionConversionBridge supports:\n\nG in S\n\nTarget nodes\n\nFunctionConversionBridge creates:\n\nF in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitComplexEqualToBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitComplexEqualToBridge","text":"SplitComplexEqualToBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSplitComplexEqualToBridge implements the following reformulation:\n\nf(x) + g(x) * im = a + b * im into f(x) = a and g(x) = b\n\nSource node\n\nSplitComplexEqualToBridge supports:\n\nG in MOI.EqualTo{Complex{T}}\n\nwhere G is a function with Complex coefficients.\n\nTarget nodes\n\nSplitComplexEqualToBridge creates:\n\nF in MOI.EqualTo{T}\n\nwhere F is the type of the real/imaginary part of G.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitComplexZerosBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitComplexZerosBridge","text":"SplitComplexZerosBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSplitComplexZerosBridge implements the following reformulation:\n\nf(x) in 0^n into textRe(f(x)) in 0^n and textIm(f(x)) in 0^n\n\nSource node\n\nSplitComplexZerosBridge supports:\n\nG in MOI.Zeros\n\nwhere G is a function with Complex coefficients.\n\nTarget nodes\n\nSplitComplexZerosBridge creates:\n\nF in MOI.Zeros\n\nwhere F is the type of the real/imaginary part of G.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitHyperRectangleBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitHyperRectangleBridge","text":"SplitHyperRectangleBridge{T,G,F} <: Bridges.Constraint.AbstractBridge\n\nSplitHyperRectangleBridge implements the following reformulation:\n\nf(x) in textsfHyperRectangle(l u) to f(x) - l u - f(x) in mathbbR_+.\n\nSource node\n\nSplitHyperRectangleBridge supports:\n\nF in MOI.HyperRectangle\n\nTarget nodes\n\nSplitHyperRectangleBridge creates:\n\nG in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SplitIntervalBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SplitIntervalBridge","text":"SplitIntervalBridge{T,F,S,LS,US} <: Bridges.Constraint.AbstractBridge\n\nSplitIntervalBridge implements the following reformulations:\n\nl le f(x) le u into f(x) ge l and f(x) le u\nf(x) = b into f(x) ge b and f(x) le b\nf(x) in 0 into f(x) in mathbbR_+ and f(x) in mathbbR_-\n\nSource node\n\nSplitIntervalBridge supports:\n\nF in MOI.Interval{T}\nF in MOI.EqualTo{T}\nF in MOI.Zeros\n\nTarget nodes\n\nSplitIntervalBridge creates:\n\nF in MOI.LessThan{T}\nF in MOI.GreaterThan{T}\n\nor\n\nF in MOI.Nonnegatives\nF in MOI.Nonpositives\n\nnote: Note\nIf T<:AbstractFloat and S is MOI.Interval{T} then no lower (resp. upper) bound constraint is created if the lower (resp. upper) bound is typemin(T) (resp. typemax(T)). Similarly, when MOI.ConstraintSet is set, a lower or upper bound constraint may be deleted or created accordingly.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoRSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoRSOCBridge","text":"SOCtoRSOCBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSOCtoRSOCBridge implements the following reformulation:\n\nx_2 le t into (t+x_1)(t-x_1)ge (x_2ldotsx_N)_2^2\n\nAssumptions\n\nSOCtoRSOCBridge assumes that the length of x is at least one.\n\nSource node\n\nSOCtoRSOCBridge supports:\n\nG in MOI.SecondOrderCone\n\nTarget node\n\nSOCtoRSOCBridge creates:\n\nF in MOI.RotatedSecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoSOCBridge","text":"RSOCtoSOCBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoSOCBridge implements the following reformulation:\n\nx_2^2 le 2tu into fract - usqrt 2 x_2 le  fract + usqrt 2\n\nSource node\n\nRSOCtoSOCBridge supports:\n\nG in MOI.RotatedSecondOrderCone\n\nTarget node\n\nRSOCtoSOCBridge creates:\n\nF in MOI.SecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoNonConvexQuadBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoNonConvexQuadBridge","text":"SOCtoNonConvexQuadBridge{T} <: Bridges.Constraint.AbstractBridge\n\nSOCtoNonConvexQuadBridge implements the following reformulations:\n\nx_2 le t into sum x^2 - t^2 le 0 and 1t + 0 ge 0\n\nThe MOI.ScalarAffineFunction 1t + 0 is used in case the variable has other bound constraints.\n\nwarning: Warning\nThis transformation starts from a convex constraint and creates a non-convex constraint. Unless the solver has explicit support for detecting second-order cones in quadratic form, this may (wrongly) be interpreted by the solver as being non-convex. Therefore, this bridge is not added automatically by MOI.Bridges.full_bridge_optimizer. Care is recommended when adding this bridge to a optimizer.\n\nSource node\n\nSOCtoNonConvexQuadBridge supports:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\nTarget nodes\n\nSOCtoNonConvexQuadBridge creates:\n\nMOI.ScalarQuadraticFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoNonConvexQuadBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoNonConvexQuadBridge","text":"RSOCtoNonConvexQuadBridge{T} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoNonConvexQuadBridge implements the following reformulations:\n\nx_2^2 le 2tu into sum x^2 - 2tu le 0, 1t + 0 ge 0, and 1u + 0 ge 0.\n\nThe MOI.ScalarAffineFunctions 1t + 0 and 1u + 0 are used in case the variables have other bound constraints.\n\nwarning: Warning\nThis transformation starts from a convex constraint and creates a non-convex constraint. Unless the solver has explicit support for detecting rotated second-order cones in quadratic form, this may (wrongly) be interpreted by the solver as being non-convex. Therefore, this bridge is not added automatically by MOI.Bridges.full_bridge_optimizer. Care is recommended when adding this bridge to a optimizer.\n\nSource node\n\nRSOCtoNonConvexQuadBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoNonConvexQuadBridge creates:\n\nMOI.ScalarQuadraticFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.QuadtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.QuadtoSOCBridge","text":"QuadtoSOCBridge{T} <: Bridges.Constraint.AbstractBridge\n\nQuadtoSOCBridge converts quadratic inequalities\n\nfrac12x^T Q x + a^T x le ub\n\ninto MOI.RotatedSecondOrderCone constraints, but it only applies when Q is positive definite.\n\nThis is because, if Q is positive definite, there exists U such that Q = U^T U, and so the inequality can then be rewritten as;\n\nU x_2^2 le 2 (-a^T x + ub)\n\nTherefore, QuadtoSOCBridge implements the following reformulations:\n\nfrac12x^T Q x + a^T x le ub into (1 -a^T x + ub Ux) in RotatedSecondOrderCone where Q = U^T U\nfrac12x^T Q x + a^T x ge lb into (1 a^T x - lb Ux) in RotatedSecondOrderCone where -Q = U^T U\n\nSource node\n\nQuadtoSOCBridge supports:\n\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\nTarget nodes\n\nRelativeEntropyBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.RotatedSecondOrderCone\n\nErrors\n\nThis bridge errors if Q is not positive definite.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SOCtoPSDBridge","text":"SOCtoPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nSOCtoPSDBridge implements the following reformulation:\n\nx_2 le t into leftbeginarrayc ct  x^top  x  t mathbfIendarrayrightsucceq 0\n\nwarning: Warning\nThis bridge is not added by default by MOI.Bridges.full_bridge_optimizer because bridging second order cone constraints to semidefinite constraints can be achieved by the SOCtoRSOCBridge followed by the RSOCtoPSDBridge, while creating a smaller semidefinite constraint.\n\nSource node\n\nSOCtoPSDBridge supports:\n\nG in MOI.SecondOrderCone\n\nTarget nodes\n\nSOCtoPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RSOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RSOCtoPSDBridge","text":"RSOCtoPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nRSOCtoPSDBridge implements the following reformulation:\n\nx_2^2 le 2tcdot u into leftbeginarrayc ct  x^top  x  2tu mathbfIendarrayrightsucceq 0\n\nSource node\n\nRSOCtoPSDBridge supports:\n\nG in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormInfinityBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormInfinityBridge","text":"NormInfinityBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormInfinityBridge implements the following reformulation:\n\nx_infty le t into t - x_i t + x_i in mathbbR_+.\n\nSource node\n\nNormInfinityBridge supports:\n\nG in MOI.NormInfinityCone{T}\n\nTarget nodes\n\nNormInfinityBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormOneBridge","text":"NormOneBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormOneBridge implements the following reformulation:\n\nsum x_i le t into t - sum y_i y_i - x_i y_i + x_i in mathbbR_+.\n\nSource node\n\nNormOneBridge supports:\n\nG in MOI.NormOneCone{T}\n\nTarget nodes\n\nNormOneBridge creates:\n\nF in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormToPowerBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormToPowerBridge","text":"NormToPowerBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormToPowerBridge implements the following reformulation:\n\n(t x) in NormCone(p 1+d) into (r_i t x_i) in PowerCone(1  p) for all i, and sumlimits_i r_i == t.\n\nFor details, see Alizadeh, F., and Goldfarb, D. (2001). \"Second-order cone programming.\" Mathematical Programming, Series B, 95:3-51.\n\nSource node\n\nNormToPowerBridge supports:\n\nF in MOI.NormCone\n\nTarget nodes\n\nNormToPowerBridge creates:\n\nF in MOI.PowerCone{T}\nMOI.ScalarAffineFunction in MOI.EqualTo\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormOneConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormOneConeToNormConeBridge","text":"NormOneConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormOneConeToNormConeBridge implements the following reformulations:\n\n(t x) in NormOneCone(d) into (t x) in NormCone(1 d)\n\nSource node\n\nNormOneConeToNormConeBridge supports:\n\nF in MOI.NormOneCone\n\nTarget nodes\n\nNormOneConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SecondOrderConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SecondOrderConeToNormConeBridge","text":"SecondOrderConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nSecondOrderConeToNormConeBridge implements the following reformulations:\n\n(t x) in SecondOrderCone(d) into (t x) in NormCone(2 d)\n\nSource node\n\nSecondOrderConeToNormConeBridge supports:\n\nF in MOI.SecondOrderCone\n\nTarget nodes\n\nSecondOrderConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormInfinityConeToNormConeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormInfinityConeToNormConeBridge","text":"NormInfinityConeToNormConeBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nNormInfinityConeToNormConeBridge implements the following reformulations:\n\n(t x) in NormInfinityCone(d) into (t x) in NormCone(Inf d)\n\nSource node\n\nNormInfinityConeToNormConeBridge supports:\n\nF in MOI.NormInfinityCone\n\nTarget nodes\n\nNormInfinityConeToNormConeBridge creates:\n\nF in MOI.NormCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeantoRelEntrBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeantoRelEntrBridge","text":"GeoMeantoRelEntrBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nGeoMeantoRelEntrBridge implements the following reformulation:\n\n(u w) in GeometricMeanCone into (0 w (u + y)mathbf1)in RelativeEntropyCone and y ge 0\n\nSource node\n\nGeoMeantoRelEntrBridge supports:\n\nH in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeantoRelEntrBridge creates:\n\nG in MOI.RelativeEntropyCone\nF in MOI.Nonnegatives\n\nDerivation\n\nThe derivation of the bridge is as follows:\n\nbeginaligned\n(u w) in GeometricMeanCone iff  u le left(prod_i=1^n w_iright)^1n \niff  0 le u + y le left(prod_i=1^n w_iright)^1n y ge 0 \niff  1 le fracleft(prod_i=1^n w_iright)^1nu + y y ge 0 \niff  1 le left(prod_i=1^n fracw_iu + yright)^1n y ge 0 \niff  0 le sum_i=1^n logleft(fracw_iu + yright) y ge 0 \niff  0 ge sum_i=1^n logleft(fracu + yw_iright) y ge 0 \niff  0 ge sum_i=1^n (u + y) logleft(fracu + yw_iright) y ge 0 \niff  (0 w (u + y)mathbf1) in RelativeEntropyCone y ge 0 \nendaligned\n\nThis derivation assumes that u + y  0, which is enforced by the relative entropy cone.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge","text":"GeoMeanToPowerBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nGeoMeanToPowerBridge implements the following reformulation:\n\n(y x) in GeometricMeanCone(1+d) into (x_1 t y) in PowerCone(1d) and (t x_2  x_d) in GeometricMeanCone(d), which is then recursively expanded into more PowerCone constraints.\n\nSource node\n\nGeoMeanToPowerBridge supports:\n\nF in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeanToPowerBridge creates:\n\nF in MOI.PowerCone{T}\nMOI.VectorOfVariables in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.GeoMeanBridge","text":"GeoMeanBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nGeoMeanBridge implements a reformulation from MOI.GeometricMeanCone into MOI.RotatedSecondOrderCone.\n\nThe reformulation is best described in an example.\n\nConsider the cone of dimension 4:\n\nt le sqrt3x_1 x_2 x_3\n\nThis can be rewritten as exists y ge 0 such that:\n\nbeginalign*\n  t  le y\n  y^4  le x_1 x_2 x_3 y\nendalign*\n\nNote that we need to create y and not use t^4 directly because t is allowed to be negative.\n\nThis is equivalent to:\n\nbeginalign*\n    t  le fracy_1sqrt4\n    y_1^2  le 2y_2 y_3\n    y_2^2  le 2x_1 x_2 \n    y_3^2  le 2x_3(y_1sqrt4) \n    y      ge 0\nendalign*\n\nMore generally, you can show how the geometric mean code is recursively expanded into a set of new variables y in MOI.Nonnegatives, a set of MOI.RotatedSecondOrderCone constraints, and a MOI.LessThan constraint between t and y_1.\n\nSource node\n\nGeoMeanBridge supports:\n\nH in MOI.GeometricMeanCone\n\nTarget nodes\n\nGeoMeanBridge creates:\n\nF in MOI.LessThan{T}\nG in MOI.RotatedSecondOrderCone\nG in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RelativeEntropyBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RelativeEntropyBridge","text":"RelativeEntropyBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nRelativeEntropyBridge implements the following reformulation that converts a MOI.RelativeEntropyCone into an MOI.ExponentialCone:\n\nu ge sum_i=1^n w_i log left(fracw_iv_iright) into y_i ge 0, u ge sum_i=1^n y_i, and (-y_i w_i v_i) in ExponentialCone.\n\nSource node\n\nRelativeEntropyBridge supports:\n\nH in MOI.RelativeEntropyCone\n\nTarget nodes\n\nRelativeEntropyBridge creates:\n\nF in MOI.GreaterThan{T}\nG in MOI.ExponentialCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormSpectralBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormSpectralBridge","text":"NormSpectralBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nNormSpectralBridge implements the following reformulation:\n\nt ge sigma_1(X) into leftbeginarrayc ctmathbfI  X^top  X  t mathbfIendarrayright succeq 0\n\nSource node\n\nNormSpectralBridge supports:\n\nG in MOI.NormSpectralCone\n\nTarget nodes\n\nNormSpectralBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NormNuclearBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NormNuclearBridge","text":"NormNuclearBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nNormNuclearBridge implements the following reformulation:\n\nt ge sum_i sigma_i (X) into leftbeginarrayc cU  X^top  X  Vendarrayright succeq 0 and 2t ge tr(U) + tr(V).\n\nSource node\n\nNormNuclearBridge supports:\n\nH in MOI.NormNuclearCone\n\nTarget nodes\n\nNormNuclearBridge creates:\n\nF in MOI.GreaterThan{T}\nG in MOI.PositiveSemidefiniteConeTriangle\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SquareBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SquareBridge","text":"SquareBridge{T,F,G,TT,ST} <: Bridges.Constraint.AbstractBridge\n\nSquareBridge implements the following reformulations:\n\n(t u X) in LogDetConeSquare into (t u Y) in LogDetConeTriangle\n(t X) in RootDetConeSquare into (t Y) in RootDetConeTriangle\nX in AbstractSymmetricMatrixSetSquare into Y in AbstractSymmetricMatrixSetTriangle\n\nwhere Y is the upper triangluar component of X.\n\nIn addition, constraints are added as necessary to constrain the matrix X to be symmetric. For example, the constraint for the matrix:\n\nbeginpmatrix\n  1       1 + x  2 - 3x\n  1 +  x  2 + x  3 -  x\n  2 - 3x  2 + x      2x\nendpmatrix\n\ncan be broken down to the constraint of the symmetric matrix\n\nbeginpmatrix\n  1       1 + x  2 - 3x\n  cdot  2 + x  3 -  x\n  cdot  cdot     2x\nendpmatrix\n\nand the equality constraint between the off-diagonal entries (2, 3) and (3, 2) 3 - x == 2 + x. Note that no symmetrization constraint needs to be added between the off-diagonal entries (1, 2) and (2, 1) or between (1, 3) and (3, 1) because the expressions are the same.\n\nSource node\n\nSquareBridge supports:\n\nF in ST\n\nTarget nodes\n\nSquareBridge creates:\n\nG in TT\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.HermitianToSymmetricPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.HermitianToSymmetricPSDBridge","text":"HermitianToSymmetricPSDBridge{T,F,G} <: Bridges.Constraint.AbstractBridge\n\nHermitianToSymmetricPSDBridge implements the following reformulation:\n\nHermitian positive semidefinite n x n complex matrix to a symmetric positive semidefinite 2n x 2n real matrix.\n\nSee also MOI.Bridges.Variable.HermitianToSymmetricPSDBridge.\n\nSource node\n\nHermitianToSymmetricPSDBridge supports:\n\nG in MOI.HermitianPositiveSemidefiniteConeTriangle\n\nTarget node\n\nHermitianToSymmetricPSDBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\n\nReformulation\n\nThe reformulation is best described by example.\n\nThe Hermitian matrix:\n\nbeginbmatrix\n  x_11             x_12 + y_12im  x_13 + y_13im\n  x_12 - y_12im  x_22             x_23 + y_23im\n  x_13 - y_13im  x_23 - y_23im  x_33\nendbmatrix\n\nis positive semidefinite if and only if the symmetric matrix:\n\nbeginbmatrix\n    x_11  x_12  x_13  0        y_12   y_13 \n            x_22  x_23  -y_12  0        y_23 \n                    x_33  -y_13  -y_23  0      \n                            x_11   x_12   x_13 \n                                     x_22   x_23 \n                                              x_33\nendbmatrix\n\nis positive semidefinite.\n\nThe bridge achieves this reformulation by constraining the above matrix to belong to the MOI.PositiveSemidefiniteConeTriangle(6).\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SetDotScalingBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SetDotScalingBridge","text":"SetDotScalingBridge{T,S,F,G} <: Bridges.Constraint.AbstractBridge\n\nSetDotScalingBridge implements the reformulation from constraints in S to constraints in MOI.Scaled{S}.\n\nSource node\n\nSetDotScalingBridge supports:\n\nG in S\n\nTarget node\n\nSetDotScalingBridge creates:\n\nF in MOI.Scaled{S}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SetDotInverseScalingBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SetDotInverseScalingBridge","text":"SetDotInverseScalingBridge{T,S,F,G} <: Bridges.Constraint.AbstractBridge\n\nSetDotInverseScalingBridge implements the reformulation from constraints in the MOI.Scaled{S} to constraints in the S.\n\nSource node\n\nSetDotInverseScalingBridge supports:\n\nG in MOI.Scaled{S}\n\nTarget node\n\nSetDotInverseScalingBridge creates:\n\nF in S\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.RootDetBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.RootDetBridge","text":"RootDetBridge{T,F,G,H} <: Bridges.Constraint.AbstractBridge\n\nThe MOI.RootDetConeTriangle is representable by MOI.PositiveSemidefiniteConeTriangle and MOI.GeometricMeanCone constraints, see [1, p. 149].\n\nIndeed, t le det(X)^1n if and only if there exists a lower triangular matrix Δ such that:\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  (t mathrmDiag(Δ))  in GeometricMeanCone\nendalign*\n\nSource node\n\nRootDetBridge supports:\n\nI in MOI.RootDetConeTriangle\n\nTarget nodes\n\nRootDetBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\nG in MOI.GeometricMeanCone\n\n[1] Ben-Tal, Aharon, and Arkadi Nemirovski. Lectures on modern convex     optimization: analysis, algorithms, and engineering applications. Society     for Industrial and Applied Mathematics, 2001.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.LogDetBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.LogDetBridge","text":"LogDetBridge{T,F,G,H,I} <: Bridges.Constraint.AbstractBridge\n\nThe MOI.LogDetConeTriangle is representable by MOI.PositiveSemidefiniteConeTriangle and MOI.ExponentialCone constraints.\n\nIndeed, logdet(X) = sumlimits_i=1^n log(delta_i) where delta_i are the eigenvalues of X.\n\nAdapting the method from [1, p. 149], we see that t le u log(det(Xu)) for u  0 if and only if there exists a lower triangular matrix Δ such that\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  t - sum_i=1^n u logleft(fracΔ_iiuright)  le 0\nendalign*\n\nWhich we reformulate further into\n\nbeginalign*\n  beginpmatrix\n    X  Δ\n    Δ^top  mathrmDiag(Δ)\n  endpmatrix  succeq 0\n  (l_i u  Δ_ii)  in ExponentialConequad forall i \n  t - sum_i=1^n l_i  le 0\nendalign*\n\nSource node\n\nLogDetBridge supports:\n\nI in MOI.LogDetConeTriangle\n\nTarget nodes\n\nLogDetBridge creates:\n\nF in MOI.PositiveSemidefiniteConeTriangle\nG in MOI.ExponentialCone\nH in MOI.LessThan{T}\n\n[1] Ben-Tal, Aharon, and Arkadi Nemirovski. Lectures on modern convex     optimization: analysis, algorithms, and engineering applications. Society     for Industrial and Applied Mathematics, 2001.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorActiveOnFalseBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorActiveOnFalseBridge","text":"IndicatorActiveOnFalseBridge{T,F,S} <: Bridges.Constraint.AbstractBridge\n\nIndicatorActiveOnFalseBridge implements the following reformulation:\n\nneg z implies f(x) in S into y implies f(x) in S, z + y = 1, and y in 0 1\n\nSource node\n\nIndicatorActiveOnFalseBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ZERO,S}\n\nTarget nodes\n\nIndicatorActiveOnFalseBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ONE,S}\nMOI.ScalarAffineFunction{T} in MOI.EqualTo\nMOI.VariableIndex in MOI.ZeroOne\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorGreaterToLessThanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorGreaterToLessThanBridge","text":"IndicatorGreaterToLessThanBridge{T,A} <: Bridges.Constraint.AbstractBridge\n\nIndicatorGreaterToLessThanBridge implements the following reformulation:\n\nz implies f(x) ge l into z implies -f(x) le -l\n\nSource node\n\nIndicatorGreaterToLessThanBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.GreaterThan{T}}\n\nTarget nodes\n\nIndicatorGreaterToLessThanBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.LessThan{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorLessToGreaterThanBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorLessToGreaterThanBridge","text":"IndicatorLessToGreaterThanBridge{T,A} <: Bridges.Constraint.AbstractBridge\n\nIndicatorLessToGreaterThanBridge implements the following reformulations:\n\nz implies f(x) le u into z implies -f(x) ge -u\n\nSource node\n\nIndicatorLessToGreaterThanBridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.LessThan{T}}\n\nTarget nodes\n\nIndicatorLessToGreaterThanBridge creates:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{A,MOI.GreaterThan{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IndicatorSOS1Bridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IndicatorSOS1Bridge","text":"IndicatorSOS1Bridge{T,S} <: Bridges.Constraint.AbstractBridge\n\nIndicatorSOS1Bridge implements the following reformulation:\n\nz implies f(x) in S into f(x) + y in S, SOS1(y z)\n\nwarning: Warning\nThis bridge assumes that the solver supports MOI.SOS1{T} constraints in which one of the variables (y) is continuous.\n\nSource node\n\nIndicatorSOS1Bridge supports:\n\nMOI.VectorAffineFunction{T} in MOI.Indicator{MOI.ACTIVATE_ON_ONE,S}\n\nTarget nodes\n\nIndicatorSOS1Bridge creates:\n\nMOI.ScalarAffineFunction{T} in S\nMOI.VectorOfVariables in MOI.SOS1{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.SemiToBinaryBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.SemiToBinaryBridge","text":"SemiToBinaryBridge{T,S} <: Bridges.Constraint.AbstractBridge\n\nSemiToBinaryBridge implements the following reformulations:\n\nx in 0 cup l u into\nbeginaligned\nx leq z u \nx geq z l \nz in 0 1\nendaligned\nx in 0 cup l ldots u into\nbeginaligned\nx leq z u \nx geq z l \nz in 0 1 \nx in mathbbZ\nendaligned\n\nSource node\n\nSemiToBinaryBridge supports:\n\nMOI.VariableIndex in MOI.Semicontinuous{T}\nMOI.VariableIndex in MOI.Semiinteger{T}\n\nTarget nodes\n\nSemiToBinaryBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\nMOI.VariableIndex{T} in MOI.Integer (if S is MOI.Semiinteger{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ZeroOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ZeroOneBridge","text":"ZeroOneBridge{T} <: Bridges.Constraint.AbstractBridge\n\nZeroOneBridge implements the following reformulation:\n\nx in 0 1 into x in mathbbZ, 1x in 0 1.\n\nnote: Note\nZeroOneBridge adds a linear constraint instead of adding variable bounds to avoid conflicting with bounds set by the user.\n\nSource node\n\nZeroOneBridge supports:\n\nMOI.VariableIndex in MOI.ZeroOne\n\nTarget nodes\n\nZeroOneBridge creates:\n\nMOI.VariableIndex in MOI.Integer\nMOI.ScalarAffineFunction{T} in MOI.Interval{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.IntegerToZeroOneBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.IntegerToZeroOneBridge","text":"IntegerToZeroOneBridge{T} <: Bridges.Constraint.AbstractBridge\n\nIntegerToZeroOneBridge implements the following reformulation:\n\nx in mathbfZ into y_i in 0 1, x == lb + sum 2^i-1 y_i.\n\nSource node\n\nIntegerToZeroOneBridge supports:\n\nVariableIndex in MOI.Integer\n\nTarget nodes\n\nIntegerToZeroOneBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\nDeveloper note\n\nThis bridge is implemented as a constraint bridge instead of a variable bridge because we don't want to substitute the linear combination of y for every instance of x. Doing so would be expensive and greatly reduce the sparsity of the constraints.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.NumberConversionBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.NumberConversionBridge","text":"NumberConversionBridge{T,F1,S1,F2,S2} <: Bridges.Constraint.AbstractBridge\n\nNumberConversionBridge implements the following reformulation:\n\nf1(x) in S1 to f2(x) in S2\n\nwhere f and S are the same functional form, but differ in their coefficient type.\n\nSource node\n\nNumberConversionBridge supports:\n\nF1 in S1\n\nTarget node\n\nNumberConversionBridge creates:\n\nF2 in S2\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.AllDifferentToCountDistinctBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.AllDifferentToCountDistinctBridge","text":"AllDifferentToCountDistinctBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nAllDifferentToCountDistinctBridge implements the following reformulations:\n\nx in textsfAllDifferent(d) to (n x) in textsfCountDistinct(1+d) and n = d\nf(x) in textsfAllDifferent(d) to (d f(x)) in textsfCountDistinct(1+d)\n\nSource node\n\nAllDifferentToCountDistinctBridge supports:\n\nF in MOI.AllDifferent\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nAllDifferentToCountDistinctBridge creates:\n\nF in MOI.CountDistinct\nMOI.VariableIndex in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ReifiedAllDifferentToCountDistinctBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ReifiedAllDifferentToCountDistinctBridge","text":"ReifiedAllDifferentToCountDistinctBridge{T,F} <:\nBridges.Constraint.AbstractBridge\n\nReifiedAllDifferentToCountDistinctBridge implements the following reformulations:\n\nr iff x in textsfAllDifferent(d) to r iff (n x) in textsfCountDistinct(1+d) and n = d\nr iff f(x) in textsfAllDifferent(d) to r iff (d f(x)) in textsfCountDistinct(1+d)\n\nSource node\n\nReifiedAllDifferentToCountDistinctBridge supports:\n\nF in MOI.Reified{MOI.AllDifferent}\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nReifiedAllDifferentToCountDistinctBridge creates:\n\nF in MOI.Reified{MOI.CountDistinct}\nMOI.VariableIndex in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.BinPackingToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.BinPackingToMILPBridge","text":"BinPackingToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nBinPackingToMILPBridge implements the following reformulation:\n\nx in BinPacking(c w) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we add the capacity constraint for all possible bins j:\n\nsumlimits_i w_i z_ij le c forall j in bigcup_i=1ldotsd S_i\n\nSource node\n\nBinPackingToMILPBridge supports:\n\nF in MOI.BinPacking{T}\n\nTarget nodes\n\nBinPackingToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CircuitToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CircuitToMILPBridge","text":"CircuitToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCircuitToMILPBridge implements the following reformulation:\n\nx in textsfCircuit(d) to the Miller-Tucker-Zemlin formulation of the Traveling Salesperson Problem.\n\nSource node\n\nCircuitToMILPBridge supports:\n\nF in MOI.Circuit\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCircuitToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{T}\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountAtLeastToCountBelongsBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountAtLeastToCountBelongsBridge","text":"CountAtLeastToCountBelongsBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountAtLeastToCountBelongsBridge implements the following reformulation:\n\nx in textsfCountAtLeast(n d mathcalS) to (n_i x_d_i) in textsfCountBelongs(1+d mathcalS) and n_i ge n for all i.\n\nSource node\n\nCountAtLeastToCountBelongsBridge supports:\n\nF in MOI.CountAtLeast\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountAtLeastToCountBelongsBridge creates:\n\nF in MOI.CountBelongs\nMOI.VariableIndex in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountBelongsToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountBelongsToMILPBridge","text":"CountBelongsToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountBelongsToMILPBridge implements the following reformulation:\n\n(n x) in textsfCountBelongs(1+d mathcalS) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nFinally, n is constrained to be the number of z_ij elements that are in mathcalS:\n\nn - sumlimits_iin 1ldots d j in mathcalS z_ij = 0\n\nSource node\n\nCountBelongsToMILPBridge supports:\n\nF in MOI.CountBelongs\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountBelongsToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountDistinctToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountDistinctToMILPBridge","text":"CountDistinctToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountDistinctToMILPBridge implements the following reformulation:\n\n(n x) in textsfCountDistinct(1+d) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we introduce new binary variables y_j, which are 1 if a variable takes the value j in the optimal solution and 0 otherwise.\n\nbeginaligned\ny_j in 0 1                          forall j in bigcup_i=1ldotsd S_i \ny_j le sumlimits_i in 1ldots d j in S_i z_ij le M y_j   forall j in bigcup_i=1ldotsd S_i\nendaligned\n\nFinally, n is constrained to be the number of y_j elements that are non-zero:\n\nn - sumlimits_j in bigcup_i=1ldotsd S_i y_j = 0\n\nSource node\n\nCountDistinctToMILPBridge supports:\n\nF in MOI.CountDistinct\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nCountDistinctToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.ReifiedCountDistinctToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.ReifiedCountDistinctToMILPBridge","text":"ReifiedCountDistinctToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nReifiedCountDistinctToMILPBridge implements the following reformulation:\n\nr iff (n x) in textsfCountDistinct(1+d) into a mixed-integer linear program.\n\nReformulation\n\nThe reformulation is non-trivial, and it depends on the finite domain of each variable x_i, which we as define S_i = l_ildotsu_i.\n\nFirst, we introduce new binary variables z_ij, which are 1 if variable x_i takes the value j in the optimal solution and 0 otherwise:\n\nbeginaligned\nz_ij in 0 1                                forall i in 1ldots d j in S_i  \nx_i - sumlimits_jin S_i j cdot z_ij = 0   forall i in 1ldots d              \nsumlimits_jin S_i z_ij = 1                  forall i in 1ldots d              \nendaligned\n\nThen, we introduce new binary variables y_j, which are 1 if a variable takes the value j in the optimal solution and 0 otherwise.\n\nbeginaligned\ny_j in 0 1                          forall j in bigcup_i=1ldotsd S_i \ny_j le sumlimits_i in 1ldots d j in S_i z_ij le M y_j   forall j in bigcup_i=1ldotsd S_i\nendaligned\n\nFinally, n is constrained to be the number of y_j elements that are non-zero, with some slack:\n\nn - sumlimits_j in bigcup_i=1ldotsd S_i y_j = delta^+ - delta^-\n\nAnd then the slack is constrained to respect the reif variable r:\n\nbeginaligned\nd_1 le delta^+ le M d_1 \nd_2 le delta^- le M d_s \nd_1 + d_2 + r = 1             \nd_1 d_2 in 0 1\nendaligned\n\nSource node\n\nReifiedCountDistinctToMILPBridge supports:\n\nF in MOI.Reified{MOI.CountDistinct}\n\nwhere F is MOI.VectorOfVariables or MOI.VectorAffineFunction{T}.\n\nTarget nodes\n\nReifiedCountDistinctToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.LessThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.CountGreaterThanToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.CountGreaterThanToMILPBridge","text":"CountGreaterThanToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nCountGreaterThanToMILPBridge implements the following reformulation:\n\n(c y x) in CountGreaterThan() into a mixed-integer linear program.\n\nSource node\n\nCountGreaterThanToMILPBridge supports:\n\nF in MOI.CountGreaterThan\n\nTarget nodes\n\nCountGreaterThanToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\nMOI.ScalarAffineFunction{T} in MOI.GreaterThan{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.TableToMILPBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Constraint.TableToMILPBridge","text":"TableToMILPBridge{T,F} <: Bridges.Constraint.AbstractBridge\n\nTableToMILPBridge implements the following reformulation:\n\nx in Table(t) into\nbeginaligned\nz_j in 0 1                      quad forall i j \nsumlimits_j=1^n z_j = 1                           \nsumlimits_j=1^n t_ij z_j = x_i  quad forall i\nendaligned\n\nSource node\n\nTableToMILPBridge supports:\n\nF in MOI.Table{T}\n\nTarget nodes\n\nTableToMILPBridge creates:\n\nMOI.VariableIndex in MOI.ZeroOne\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#objective_bridges_ref","page":"List of bridges","title":"Objective bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Objective.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Objective.FunctionizeBridge\nBridges.Objective.QuadratizeBridge\nBridges.Objective.SlackBridge\nBridges.Objective.VectorFunctionizeBridge\nBridges.Objective.VectorSlackBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.FunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.FunctionizeBridge","text":"FunctionizeBridge{T}\n\nFunctionizeBridge implements the following reformulations:\n\nmin x into min1x + 0\nmax x into max1x + 0\n\nwhere T is the coefficient type of 1 and 0.\n\nSource node\n\nFunctionizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.VariableIndex}\n\nTarget nodes\n\nFunctionizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.QuadratizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.QuadratizeBridge","text":"QuadratizeBridge{T}\n\nQuadratizeBridge implements the following reformulations:\n\nmin a^top x + b into minx^top mathbf0 x + a^top x + b\nmax a^top x + b into maxx^top mathbf0 x + a^top x + b\n\nwhere T is the coefficient type of 0.\n\nSource node\n\nQuadratizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.ScalarAffineFunction{T}}\n\nTarget nodes\n\nQuadratizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.SlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.SlackBridge","text":"SlackBridge{T,F,G}\n\nSlackBridge implements the following reformulations:\n\nminf(x) into miny f(x) - y le 0\nmaxf(x) into maxy f(x) - y ge 0\n\nwhere F is the type of f(x) - y, G is the type of f(x), and T is the coefficient type of f(x).\n\nSource node\n\nSlackBridge supports:\n\nMOI.ObjectiveFunction{G}\n\nTarget nodes\n\nSlackBridge creates:\n\nOne variable node: MOI.VariableIndex in MOI.Reals\nOne objective node: MOI.ObjectiveFunction{MOI.VariableIndex}\nOne constraint node, that depends on the MOI.ObjectiveSense:\nF-in-MOI.LessThan if MIN_SENSE\nF-in-MOI.GreaterThan if MAX_SENSE\n\nwarning: Warning\nWhen using this bridge, changing the optimization sense is not supported. Set the sense to MOI.FEASIBILITY_SENSE first to delete the bridge, then set MOI.ObjectiveSense and re-add the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.VectorFunctionizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.VectorFunctionizeBridge","text":"VectorFunctionizeBridge{T}\n\nVectorFunctionizeBridge implements the following reformulations:\n\nmin x into min1x + 0\nmax x into max1x + 0\n\nwhere T is the coefficient type of 1 and 0.\n\nSource node\n\nVectorFunctionizeBridge supports:\n\nMOI.ObjectiveFunction{MOI.VectorOfVariables}\n\nTarget nodes\n\nVectorFunctionizeBridge creates:\n\nOne objective node: MOI.ObjectiveFunction{MOI.VectorAffineFunction{T}}\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Objective.VectorSlackBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Objective.VectorSlackBridge","text":"VectorSlackBridge{T,F,G}\n\nVectorSlackBridge implements the following reformulations:\n\nminf(x) into miny y - f(x) in mathbbR_+ \nmaxf(x) into maxy f(x) - y in mathbbR_+ \n\nwhere F is the type of f(x) - y, G is the type of f(x), and T is the coefficient type of f(x).\n\nSource node\n\nVectorSlackBridge supports:\n\nMOI.ObjectiveFunction{G}\n\nTarget nodes\n\nVectorSlackBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Reals\nOne objective node: MOI.ObjectiveFunction{MOI.VectorOfVariables}\nOne constraint node: F-in-MOI.Nonnegatives\n\nwarning: Warning\nWhen using this bridge, changing the optimization sense is not supported. Set the sense to MOI.FEASIBILITY_SENSE first to delete the bridge, then set MOI.ObjectiveSense and re-add the objective.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#variable_bridges_ref","page":"List of bridges","title":"Variable bridges","text":"","category":"section"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"These bridges are subtypes of Bridges.Variable.AbstractBridge.","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/","page":"List of bridges","title":"List of bridges","text":"Bridges.Variable.FreeBridge\nBridges.Variable.NonposToNonnegBridge\nBridges.Variable.RSOCtoPSDBridge\nBridges.Variable.RSOCtoSOCBridge\nBridges.Variable.SOCtoRSOCBridge\nBridges.Variable.VectorizeBridge\nBridges.Variable.ZerosBridge\nBridges.Variable.HermitianToSymmetricPSDBridge","category":"page"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.FreeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.FreeBridge","text":"FreeBridge{T} <: Bridges.Variable.AbstractBridge\n\nFreeBridge implements the following reformulation:\n\nx in mathbbR into y z ge 0 with the substitution rule x = y - z,\n\nwhere T is the coefficient type of y - z.\n\nSource node\n\nFreeBridge supports:\n\nMOI.VectorOfVariables in MOI.Reals\n\nTarget nodes\n\nFreeBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Nonnegatives\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.NonposToNonnegBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.NonposToNonnegBridge","text":"NonposToNonnegBridge{T} <: Bridges.Variable.AbstractBridge\n\nNonposToNonnegBridge implements the following reformulation:\n\nx in mathbbR_- into y in mathbbR_+ with the substitution rule x = -y,\n\nwhere T is the coefficient type of -y.\n\nSource node\n\nNonposToNonnegBridge supports:\n\nMOI.VectorOfVariables in MOI.Nonpositives\n\nTarget nodes\n\nNonposToNonnegBridge creates:\n\nOne variable node: MOI.VectorOfVariables in MOI.Nonnegatives,\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.RSOCtoPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.RSOCtoPSDBridge","text":"RSOCtoPSDBridge{T} <: Bridges.Variable.AbstractBridge\n\nRSOCtoPSDBridge implements the following reformulation:\n\nx_2^2 le 2tu where t u ge 0 into Y succeq 0, with the substitution rule: Y = leftbeginarrayc ct  x^top  x  2u mathbfIendarrayright\n\nAdditional bounds are added to ensure the off-diagonals of the 2uI submatrix are 0, and linear constraints are added to ennsure the diagonal of 2uI takes the same values.\n\nAs a special case, if x = 0, then RSOCtoPSDBridge reformulates into (t u) in mathbbR_+.\n\nSource node\n\nRSOCtoPSDBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget nodes\n\nRSOCtoPSDBridge creates:\n\nOne variable node that depends on the input dimension:\nMOI.VectorOfVariables in MOI.Nonnegatives if dimension is 1 or 2\nMOI.VectorOfVariables in\nMOI.PositiveSemidefiniteConeTriangle otherwise\nThe constraint node MOI.VariableIndex in MOI.EqualTo\nThe constrant node MOI.ScalarAffineFunction in MOI.EqualTo\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.RSOCtoSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.RSOCtoSOCBridge","text":"RSOCtoSOCBridge{T} <: Bridges.Variable.AbstractBridge\n\nRSOCtoSOCBridge implements the following reformulation:\n\nx_2^2 le 2tu into v_2 le w, with the substitution rules t = fracwsqrt 2 + fracv_1sqrt 2, u = fracwsqrt 2 - fracv_1sqrt 2, and x = (v_2ldotsv_N).\n\nSource node\n\nRSOCtoSOCBridge supports:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\nTarget node\n\nRSOCtoSOCBridge creates:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.SOCtoRSOCBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.SOCtoRSOCBridge","text":"SOCtoRSOCBridge{T} <: Bridges.Variable.AbstractBridge\n\nSOCtoRSOCBridge implements the following reformulation:\n\nx_2 le t into 2uv ge w_2^2, with the substitution rules t = fracusqrt 2 + fracvsqrt 2, x = (fracusqrt 2 - fracvsqrt 2 w).\n\nAssumptions\n\nSOCtoRSOCBridge assumes that x ge 1.\n\nSource node\n\nSOCtoRSOCBridge supports:\n\nMOI.VectorOfVariables in MOI.SecondOrderCone\n\nTarget node\n\nSOCtoRSOCBridge creates:\n\nMOI.VectorOfVariables in MOI.RotatedSecondOrderCone\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.VectorizeBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.VectorizeBridge","text":"VectorizeBridge{T,S} <: Bridges.Variable.AbstractBridge\n\nVectorizeBridge implements the following reformulations:\n\nx ge a into y in mathbbR_+ with the substitution rule x = a + y\nx le a into y in mathbbR_- with the substitution rule x = a + y\nx == a into y in 0 with the substitution rule x = a + y\n\nwhere T is the coefficient type of a + y.\n\nSource node\n\nVectorizeBridge supports:\n\nMOI.VariableIndex in MOI.GreaterThan{T}\nMOI.VariableIndex in MOI.LessThan{T}\nMOI.VariableIndex in MOI.EqualTo{T}\n\nTarget nodes\n\nVectorizeBridge creates:\n\nOne variable node: MOI.VectorOfVariables in S, where S is one of MOI.Nonnegatives, MOI.Nonpositives, MOI.Zeros depending on the type of S.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.ZerosBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.ZerosBridge","text":"ZerosBridge{T} <: Bridges.Variable.AbstractBridge\n\nZerosBridge implements the following reformulation:\n\nx in 0 into the substitution rule x = 0,\n\nwhere T is the coefficient type of 0.\n\nSource node\n\nZerosBridge supports:\n\nMOI.VectorOfVariables in MOI.Zeros\n\nTarget nodes\n\nZerosBridge does not create target nodes. It replaces all instances of x with 0 via substitution. This means that no variables are created in the underlying model.\n\nCaveats\n\nThe bridged variables are similar to parameters with zero values. Parameters with non-zero values can be created with constrained variables in MOI.EqualTo by combining a VectorizeBridge and this bridge.\n\nHowever, functions modified by ZerosBridge cannot be unbridged. That is, for a given function, we cannot determine if the bridged variables were used.\n\nA related implication is that this bridge does not support MOI.ConstraintDual. However, if a MOI.Utilities.CachingOptimizer is used, the dual can be determined by the bridged optimizer using MOI.Utilities.get_fallback because the caching optimizer records the unbridged function.\n\n\n\n\n\n","category":"type"},{"location":"moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Variable.HermitianToSymmetricPSDBridge","page":"List of bridges","title":"MathOptInterface.Bridges.Variable.HermitianToSymmetricPSDBridge","text":"HermitianToSymmetricPSDBridge{T} <: Bridges.Variable.AbstractBridge\n\nHermitianToSymmetricPSDBridge implements the following reformulation:\n\nHermitian positive semidefinite n x n complex matrix to a symmetric positive semidefinite 2n x 2n real matrix satisfying equality constraints described below.\n\nSource node\n\nHermitianToSymmetricPSDBridge supports:\n\nMOI.VectorOfVariables in MOI.HermitianPositiveSemidefiniteConeTriangle\n\nTarget node\n\nHermitianToSymmetricPSDBridge creates:\n\nMOI.VectorOfVariables in MOI.PositiveSemidefiniteConeTriangle\nMOI.ScalarAffineFunction{T} in MOI.EqualTo{T}\n\nReformulation\n\nThe reformulation is best described by example.\n\nThe Hermitian matrix:\n\nbeginbmatrix\n  x_11             x_12 + y_12im  x_13 + y_13im\n  x_12 - y_12im  x_22             x_23 + y_23im\n  x_13 - y_13im  x_23 - y_23im  x_33\nendbmatrix\n\nis positive semidefinite if and only if the symmetric matrix:\n\nbeginbmatrix\n    x_11  x_12  x_13  0        y_12   y_13 \n            x_22  x_23  -y_12  0        y_23 \n                    x_33  -y_13  -y_23  0      \n                            x_11   x_12   x_13 \n                                     x_22   x_23 \n                                              x_33\nendbmatrix\n\nis positive semidefinite.\n\nThe bridge achieves this reformulation by adding a new set of variables in MOI.PositiveSemidefiniteConeTriangle(6), and then adding three groups of equality constraints to:\n\nconstrain the two x blocks to be equal\nforce the diagonal of the y blocks to be 0\nforce the lower triangular of the y block to be the negative of the upper triangle.\n\n\n\n\n\n","category":"type"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"EditURL = \"https://github.com/jump-dev/SeDuMi.jl/blob/v0.4.2/README.md\"","category":"page"},{"location":"packages/SeDuMi/#SeDuMi.jl","page":"jump-dev/SeDuMi.jl","title":"SeDuMi.jl","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"SeDuMi.jl is wrapper for the SeDuMi solver.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"an exported sedumi function that is a thin wrapper on top of the sedumi MATLAB function\nan interface to MathOptInterface","category":"page"},{"location":"packages/SeDuMi/#Affiliation","page":"jump-dev/SeDuMi.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"This wrapper is maintained by the JuMP community and is not an official wrapper of SeDuMi.","category":"page"},{"location":"packages/SeDuMi/#License","page":"jump-dev/SeDuMi.jl","title":"License","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"SeDuMi.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The underlying solver, sqlp/sedumi is licensed under the GPL v2 license.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"In addition, SeDuMi requires an installation of MATLAB, which is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/SeDuMi/#Use-with-JuMP","page":"jump-dev/SeDuMi.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"To use SeDuMi with JuMP, do:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"using JuMP, SeDuMi\nmodel = Model(SeDuMi.Optimizer)\nset_attribute(model, \"fid\", 0)","category":"page"},{"location":"packages/SeDuMi/#Installation","page":"jump-dev/SeDuMi.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"First, make sure that you satisfy the requirements of the MATLAB.jl Julia package, and that the SeDuMi software is installed in your MATLAB™ installation.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"Then, install SeDuMi.jl using Pkg.add:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"import Pkg\nPkg.add(\"SeDuMi\")","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"If you get the error:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"Undefined function or variable 'sedumi'.\n\nError using save\nVariable 'jx_sedumi_arg_out_1' not found.\n\nERROR: LoadError: MATLAB.MEngineError(\"failed to get variable jx_sedumi_arg_out_1 from MATLAB session\")","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"The error means that we couldn't find the sedumi function with one output argument using the MATLAB C API.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"This most likely means that you did not add SeDuMi to the MATLAB's path, that is, the toolbox/local/pathdef.m file.","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"If modifying toolbox/local/pathdef.m does not work, the following should work, where /path/to/sedumi/ is the directory where the sedumi folder is located:","category":"page"},{"location":"packages/SeDuMi/","page":"jump-dev/SeDuMi.jl","title":"jump-dev/SeDuMi.jl","text":"julia> import MATLAB\n\njulia> cd(\"/path/to/sedumi/\") do\n           MATLAB.mat\"install_sedumi\"\n       end\n\njulia> MATLAB.mat\"savepath\"","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"EditURL = \"space_shuttle_reentry_trajectory.jl\"","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial was originally contributed by Henrique Ferrolho.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"This tutorial demonstrates how to compute a reentry trajectory for the Space Shuttle, by formulating and solving a nonlinear programming problem. The problem was drawn from Chapter 6 of \"Practical Methods for Optimal Control and Estimation Using Nonlinear Programming\", by John T. Betts.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"tip: Tip\nThis tutorial is a more-complicated version of the Rocket Control example. If you are new to solving nonlinear programs in JuMP, you may want to start there instead.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The motion of the vehicle is defined by the following set of DAEs:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\ndoth  = v sin gamma  \ndotphi  = fracvr cos gamma sin psi  cos theta  \ndottheta  = fracvr cos gamma cos psi  \ndotv  = -fracDm - g sin gamma  \ndotgamma  = fracLm v cos(beta) + cos gamma left ( fracvr - fracgv right )  \ndotpsi  = frac1m v cos gamma L sin(beta) + fracvr cos theta cos gamma sin psi sin theta  \nq  le q_U  \nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"where the aerodynamic heating on the vehicle wing leading edge is q = q_a q_r and the dynamic variables are","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n     h  quad textaltitude (ft)     qquad   gamma quad textflight path angle (rad) \n  phi  quad textlongitude (rad)   qquad   psi   quad textazimuth (rad)           \ntheta  quad textlatitude (rad)    qquad   alpha quad textangle of attack (rad)   \n     v  quad textvelocity (ftsec) qquad   beta  quad textbank angle (rad)\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The aerodynamic and atmospheric forces on the vehicle are specified by the following quantities (English units):","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n           D  = frac12 c_D S rho v^2                  qquad  a_0  = -020704 \n           L  = frac12 c_L S rho v^2                  qquad  a_1  =  0029244 \n           g  = mu  r^2                                   qquad  mu  =  014076539 times 10^17 \n           r  = R_e + h                                     qquad  b_0  =  007854 \n        rho  = rho_0 exp-hh_r                         qquad  b_1  = -061592  times 10^-2 \n      rho_0  = 0002378                                    qquad  b_2  =  0621408 times 10^-3 \n         h_r  = 23800                                       qquad  q_r  =  17700 sqrtrho (00001 v)^307 \n         c_L  = a_0 + a_1 hatalpha                      qquad  q_a  =  c_0 + c_1 hatalpha + c_2 hatalpha^2 + c_3 hatalpha^3 \n         c_D  = b_0 + b_1 hatalpha + b_2 hatalpha^2 qquad  c_0  =  10672181 \nhatalpha  = 180 alpha  pi                            qquad  c_1  = -019213774 times 10^-1 \n         R_e  = 20902900                                    qquad  c_2  =  021286289 times 10^-3 \n           S  = 2690                                        qquad  c_3  = -010117249 times 10^-5\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The reentry trajectory begins at an altitude where the aerodynamic forces are quite small with a weight of w = 203000 (lb) and mass m = w  g_0 (slug), where g_0 = 32174 (ft/sec^2). The initial conditions are as follows:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"beginaligned\n     h  = 260000 text ft  qquad  v       = 25600 text ftsec \n  phi  = 0      text deg qquad  gamma  = -1    text deg    \ntheta  = 0      text deg qquad  psi    = 90    text deg\nendaligned","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"The final point on the reentry trajectory occurs at the unknown (free) time t_F, at the so-called terminal area energy management (TAEM) interface, which is defined by the conditions","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"h = 80000 text ft qquad v = 2500 text ftsec qquad gamma = -5 text deg","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"As explained in the book, our goal is to maximize the final cross-range, which is equivalent to maximizing the final latitude of the vehicle, that is, J = theta(t_F).","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"(Image: Max cross-range shuttle reentry)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Approach","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Approach","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"We will use a discretized model of time, with a fixed number of discretized points, n. The decision variables at each point are going to be the state of the vehicle and the controls commanded to it. In addition, we will also make each time step size Delta t a decision variable; that way, we can either fix the time step size easily, or allow the solver to fine-tune the duration between each adjacent pair of points. Finally, in order to approximate the derivatives of the problem dynamics, we will use either rectangular or trapezoidal integration.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"warning: Warning\nDo not try to actually land a Space Shuttle using this notebook. There's no mesh refinement going on, which can lead to unrealistic trajectories having position and velocity errors with orders of magnitude 10^4 ft and 10^2 ft/sec, respectively.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"using JuMP\nimport Interpolations\nimport Ipopt\n\n# Global variables\nconst w = 203000.0  # weight (lb)\nconst g₀ = 32.174    # acceleration (ft/sec^2)\nconst m = w / g₀    # mass (slug)\n\n# Aerodynamic and atmospheric forces on the vehicle\nconst ρ₀ = 0.002378\nconst hᵣ = 23800.0\nconst Rₑ = 20902900.0\nconst μ = 0.14076539e17\nconst S = 2690.0\nconst a₀ = -0.20704\nconst a₁ = 0.029244\nconst b₀ = 0.07854\nconst b₁ = -0.61592e-2\nconst b₂ = 0.621408e-3\nconst c₀ = 1.0672181\nconst c₁ = -0.19213774e-1\nconst c₂ = 0.21286289e-3\nconst c₃ = -0.10117249e-5\n\n# Initial conditions\nconst h_s = 2.6          # altitude (ft) / 1e5\nconst ϕ_s = deg2rad(0)   # longitude (rad)\nconst θ_s = deg2rad(0)   # latitude (rad)\nconst v_s = 2.56         # velocity (ft/sec) / 1e4\nconst γ_s = deg2rad(-1)  # flight path angle (rad)\nconst ψ_s = deg2rad(90)  # azimuth (rad)\nconst α_s = deg2rad(0)   # angle of attack (rad)\nconst β_s = deg2rad(0)   # bank angle (rad)\nconst t_s = 1.00         # time step (sec)\n\n# Final conditions, the so-called Terminal Area Energy Management (TAEM)\nconst h_t = 0.8          # altitude (ft) / 1e5\nconst v_t = 0.25         # velocity (ft/sec) / 1e4\nconst γ_t = deg2rad(-5)  # flight path angle (rad)\n\n# Number of mesh points (knots) to be used\nconst n = 503\n\n# Integration scheme to be used for the dynamics\nconst integration_rule = \"rectangular\"\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"tip: Choose a good linear solver\nPicking a good linear solver is extremely important to maximize the performance of nonlinear solvers. For the best results, it is advised to experiment different linear solvers.For example, the linear solver MA27 is outdated and can be quite slow. MA57 is a much better alternative, especially for highly-sparse problems (such as trajectory optimization problems).","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"# Uncomment the lines below to pass user options to the solver\nuser_options = (\n# \"mu_strategy\" => \"monotone\",\n# \"linear_solver\" => \"ma27\",\n)\n\n# Create JuMP model, using Ipopt as the solver\nmodel = Model(optimizer_with_attributes(Ipopt.Optimizer, user_options...))\n\n@variables(model, begin\n    0 ≤ scaled_h[1:n]                # altitude (ft) / 1e5\n    ϕ[1:n]                # longitude (rad)\n    deg2rad(-89) ≤ θ[1:n] ≤ deg2rad(89)  # latitude (rad)\n    1e-4 ≤ scaled_v[1:n]                # velocity (ft/sec) / 1e4\n    deg2rad(-89) ≤ γ[1:n] ≤ deg2rad(89)  # flight path angle (rad)\n    ψ[1:n]                # azimuth (rad)\n    deg2rad(-90) ≤ α[1:n] ≤ deg2rad(90)  # angle of attack (rad)\n    deg2rad(-89) ≤ β[1:n] ≤ deg2rad(1)  # bank angle (rad)\n    #        3.5 ≤       Δt[1:n] ≤ 4.5          # time step (sec)\n    Δt[1:n] == 4.0         # time step (sec)\nend);\nnothing #hide","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"info: Info\nAbove you can find two alternatives for the Δt variables.The first one, 3.5 ≤ Δt[1:n] ≤ 4.5 (currently commented), allows some wiggle room for the solver to adjust the time step size between pairs of mesh points. This is neat because it allows the solver to figure out which parts of the flight require more dense discretization than others. (Remember, the number of discretized points is fixed, and this example does not implement mesh refinement.) However, this makes the problem more complex to solve, and therefore leads to a longer computation time.The second line, Δt[1:n] == 4.0, fixes the duration of every time step to exactly 4.0 seconds. This allows the problem to be solved faster. However, to do this we need to know beforehand that the close-to-optimal total duration of the flight is ~2009 seconds. Therefore, if we split the total duration in slices of 4.0 seconds, we know that we require n = 503 knots to discretize the whole trajectory.","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"# Fix initial conditions\nfix(scaled_h[1], h_s; force = true)\nfix(ϕ[1], ϕ_s; force = true)\nfix(θ[1], θ_s; force = true)\nfix(scaled_v[1], v_s; force = true)\nfix(γ[1], γ_s; force = true)\nfix(ψ[1], ψ_s; force = true)\n\n# Fix final conditions\nfix(scaled_h[n], h_t; force = true)\nfix(scaled_v[n], v_t; force = true)\nfix(γ[n], γ_t; force = true)\n\n# Initial guess: linear interpolation between boundary conditions\nx_s = [h_s, ϕ_s, θ_s, v_s, γ_s, ψ_s, α_s, β_s, t_s]\nx_t = [h_t, ϕ_s, θ_s, v_t, γ_t, ψ_s, α_s, β_s, t_s]\ninterp_linear = Interpolations.LinearInterpolation([1, n], [x_s, x_t])\ninitial_guess = mapreduce(transpose, vcat, interp_linear.(1:n))\nset_start_value.(all_variables(model), vec(initial_guess))\n\n# Functions to restore `h` and `v` to their true scale\n@expression(model, h[j = 1:n], scaled_h[j] * 1e5)\n@expression(model, v[j = 1:n], scaled_v[j] * 1e4)\n\n# Helper functions\n@expression(model, c_L[j = 1:n], a₀ + a₁ * rad2deg(α[j]))\n@expression(model, c_D[j = 1:n], b₀ + b₁ * rad2deg(α[j]) + b₂ * rad2deg(α[j])^2)\n@expression(model, ρ[j = 1:n], ρ₀ * exp(-h[j] / hᵣ))\n@expression(model, D[j = 1:n], 0.5 * c_D[j] * S * ρ[j] * v[j]^2)\n@expression(model, L[j = 1:n], 0.5 * c_L[j] * S * ρ[j] * v[j]^2)\n@expression(model, r[j = 1:n], Rₑ + h[j])\n@expression(model, g[j = 1:n], μ / r[j]^2)\n\n# Motion of the vehicle as a differential-algebraic system of equations (DAEs)\n@expression(model, δh[j = 1:n], v[j] * sin(γ[j]))\n@expression(\n    model,\n    δϕ[j = 1:n],\n    (v[j] / r[j]) * cos(γ[j]) * sin(ψ[j]) / cos(θ[j])\n)\n@expression(model, δθ[j = 1:n], (v[j] / r[j]) * cos(γ[j]) * cos(ψ[j]))\n@expression(model, δv[j = 1:n], -(D[j] / m) - g[j] * sin(γ[j]))\n@expression(\n    model,\n    δγ[j = 1:n],\n    (L[j] / (m * v[j])) * cos(β[j]) +\n    cos(γ[j]) * ((v[j] / r[j]) - (g[j] / v[j]))\n)\n@expression(\n    model,\n    δψ[j = 1:n],\n    (1 / (m * v[j] * cos(γ[j]))) * L[j] * sin(β[j]) +\n    (v[j] / (r[j] * cos(θ[j]))) * cos(γ[j]) * sin(ψ[j]) * sin(θ[j])\n)\n\n# System dynamics\nfor j in 2:n\n    i = j - 1  # index of previous knot\n\n    if integration_rule == \"rectangular\"\n        # Rectangular integration\n        @constraint(model, h[j] == h[i] + Δt[i] * δh[i])\n        @constraint(model, ϕ[j] == ϕ[i] + Δt[i] * δϕ[i])\n        @constraint(model, θ[j] == θ[i] + Δt[i] * δθ[i])\n        @constraint(model, v[j] == v[i] + Δt[i] * δv[i])\n        @constraint(model, γ[j] == γ[i] + Δt[i] * δγ[i])\n        @constraint(model, ψ[j] == ψ[i] + Δt[i] * δψ[i])\n    elseif integration_rule == \"trapezoidal\"\n        # Trapezoidal integration\n        @constraint(model, h[j] == h[i] + 0.5 * Δt[i] * (δh[j] + δh[i]))\n        @constraint(model, ϕ[j] == ϕ[i] + 0.5 * Δt[i] * (δϕ[j] + δϕ[i]))\n        @constraint(model, θ[j] == θ[i] + 0.5 * Δt[i] * (δθ[j] + δθ[i]))\n        @constraint(model, v[j] == v[i] + 0.5 * Δt[i] * (δv[j] + δv[i]))\n        @constraint(model, γ[j] == γ[i] + 0.5 * Δt[i] * (δγ[j] + δγ[i]))\n        @constraint(model, ψ[j] == ψ[i] + 0.5 * Δt[i] * (δψ[j] + δψ[i]))\n    else\n        @error \"Unexpected integration rule '$(integration_rule)'\"\n    end\nend\n\n# Objective: Maximize cross-range\n@objective(model, Max, θ[n])\n\nset_silent(model)  # Hide solver's verbose output\noptimize!(model)  # Solve for the control and state\n@assert termination_status(model) == LOCALLY_SOLVED\n\n# Show final cross-range of the solution\nprintln(\n    \"Final latitude θ = \",\n    round(objective_value(model) |> rad2deg; digits = 2),\n    \"°\",\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/#Plotting-the-results","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Plotting the results","text":"","category":"section"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"Let's plot the results to visualize the optimal trajectory:","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"using Plots\nts = cumsum([0; value.(Δt)])[1:end-1]\nplt_altitude = plot(\n    ts,\n    value.(scaled_h);\n    legend = nothing,\n    title = \"Altitude (100,000 ft)\",\n)\nplt_longitude =\n    plot(ts, rad2deg.(value.(ϕ)); legend = nothing, title = \"Longitude (deg)\")\nplt_latitude =\n    plot(ts, rad2deg.(value.(θ)); legend = nothing, title = \"Latitude (deg)\")\nplt_velocity = plot(\n    ts,\n    value.(scaled_v);\n    legend = nothing,\n    title = \"Velocity (1000 ft/sec)\",\n)\nplt_flight_path =\n    plot(ts, rad2deg.(value.(γ)); legend = nothing, title = \"Flight Path (deg)\")\nplt_azimuth =\n    plot(ts, rad2deg.(value.(ψ)); legend = nothing, title = \"Azimuth (deg)\")\n\nplot(\n    plt_altitude,\n    plt_velocity,\n    plt_longitude,\n    plt_flight_path,\n    plt_latitude,\n    plt_azimuth;\n    layout = grid(3, 2),\n    linewidth = 2,\n    size = (700, 700),\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"function q(h, v, a)\n    ρ(h) = ρ₀ * exp(-h / hᵣ)\n    qᵣ(h, v) = 17700 * √ρ(h) * (0.0001 * v)^3.07\n    qₐ(a) = c₀ + c₁ * rad2deg(a) + c₂ * rad2deg(a)^2 + c₃ * rad2deg(a)^3\n    # Aerodynamic heating on the vehicle wing leading edge\n    return qₐ(a) * qᵣ(h, v)\nend\n\nplt_attack_angle = plot(\n    ts[1:end-1],\n    rad2deg.(value.(α)[1:end-1]);\n    legend = nothing,\n    title = \"Angle of Attack (deg)\",\n)\nplt_bank_angle = plot(\n    ts[1:end-1],\n    rad2deg.(value.(β)[1:end-1]);\n    legend = nothing,\n    title = \"Bank Angle (deg)\",\n)\nplt_heating = plot(\n    ts,\n    q.(value.(scaled_h) * 1e5, value.(scaled_v) * 1e4, value.(α));\n    legend = nothing,\n    title = \"Heating (BTU/ft/ft/sec)\",\n)\n\nplot(\n    plt_attack_angle,\n    plt_bank_angle,\n    plt_heating;\n    layout = grid(3, 1),\n    linewidth = 2,\n    size = (700, 700),\n)","category":"page"},{"location":"tutorials/nonlinear/space_shuttle_reentry_trajectory/","page":"Optimal control for a Space Shuttle reentry trajectory","title":"Optimal control for a Space Shuttle reentry trajectory","text":"plot(\n    rad2deg.(value.(ϕ)),\n    rad2deg.(value.(θ)),\n    value.(scaled_h);\n    linewidth = 2,\n    legend = nothing,\n    title = \"Space Shuttle Reentry Trajectory\",\n    xlabel = \"Longitude (deg)\",\n    ylabel = \"Latitude (deg)\",\n    zlabel = \"Altitude (100,000 ft)\",\n)","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"background/algebraic_modeling_languages/#algebraic-modeling-language","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP is an algebraic modeling language for mathematical optimization written in the Julia language. In this page, we explain what an algebraic modeling language actually is.","category":"page"},{"location":"background/algebraic_modeling_languages/#What-is-an-algebraic-modeling-language?","page":"Algebraic modeling languages","title":"What is an algebraic modeling language?","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"If you have taken a class in mixed-integer linear programming, you will have seen a formulation like:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmin       c^top x \ntextst  A x = b  \n             x ge 0  \n             x_i in mathbbZ quad forall i in mathcalI\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"where c, A, and b are appropriately sized vectors and matrices of data, and mathcalI denotes the set of variables that are integer.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Solvers expect problems in a standard form like this because it limits the types of constraints that they need to consider. This makes writing a solver much easier.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"info: What is a solver?\nA solver is a software package that computes solutions to one or more classes of problems.For example, HiGHS is a solver for linear programming (LP) and mixed integer programming (MIP) problems. It incorporates algorithms such as the simplex method and the interior-point method.JuMP currently supports a number of open-source and commercial solvers, which can be viewed in the Supported-solvers table.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Despite the textbook view of a linear program, you probably formulated problems algebraically like so:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmax       sumlimits_i = 1^n c_i x_i                   \ntextst  sumlimits_i = 1^n w_i x_i le b             \n             x_i ge 0 quad forall i = 1ldotsn          \n             x_i in mathbbZ quad forall i = 1ldotsn\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"info: Info\nDo you recognize this formulation? It's the knapsack problem.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Users prefer to write problems in algebraic form because it is more convenient. For example, we used le b, even though the standard form only supported constraints of the form Ax = b.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"We could convert our knapsack problem into the standard form by adding a new slack variable x_0:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"beginaligned\nmax       sumlimits_i = 1^n c_i x_i            \ntextst  x_0 + sumlimits_i = 1^n w_i x_i = b  \n             x_i ge 0 quad forall i = 0ldotsn   \n             x_i in mathbbZ quad forall i = 1ldotsn\nendaligned","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"However, as models get more complicated, this manual conversion becomes more and more error-prone.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"An algebraic modeling language is a tool that simplifies the translation between the algebraic form of the modeler, and the standard form of the solver.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Each algebraic modeling language has two main parts:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"A domain specific language for the user to write down problems in algebraic form.\nA converter from the algebraic form into a standard form supported by the solver (and back again).","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Part 2 is less trivial than it might seem, because each solver has a unique application programming interface (API) and data structure for representing optimization models and obtaining results.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP uses the MathOptInterface.jl package to abstract these differences between solvers.","category":"page"},{"location":"background/algebraic_modeling_languages/#What-is-MathOptInterface?","page":"Algebraic modeling languages","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"MathOptInterface (MOI) is an abstraction layer designed to provide an interface to mathematical optimization solvers so that users do not need to understand multiple solver-specific APIs. MOI can be used directly, or through a higher-level modeling interface like JuMP.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"There are three main parts to MathOptInterface:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"A solver-independent API that abstracts concepts such as adding and deleting variables and constraints, setting and getting parameters, and querying results. For more information on the MathOptInterface API, read the documentation.\nAn automatic rewriting system based on equivalent formulations of a constraint. For more information on this rewriting system, read the LazyBridgeOptimizer section of the manual, and our paper on arXiv.\nUtilities for managing how and when models are copied to solvers. For more information on this, read the CachingOptimizer section of the manual.","category":"page"},{"location":"background/algebraic_modeling_languages/#From-user-to-solver","page":"Algebraic modeling languages","title":"From user to solver","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"This section provides a brief summary of the steps that happen in order to translate the model that the user writes into a model that the solver understands.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-I:-writing-in-algebraic-form","page":"Algebraic modeling languages","title":"Step I: writing in algebraic form","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"JuMP provides the first part of an algebraic modeling language using the @variable, @objective, and @constraint macros.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"For example, here's how we write the knapsack problem in JuMP:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using JuMP, HiGHS\n\njulia> function algebraic_knapsack(c, w, b)\n           n = length(c)\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x[1:n] >= 0, Int)\n           @objective(model, Max, sum(c[i] * x[i] for i = 1:n))\n           @constraint(model, sum(w[i] * x[i] for i = 1:n) <= b)\n           optimize!(model)\n           return value.(x)\n       end\nalgebraic_knapsack (generic function with 1 method)\n\njulia> algebraic_knapsack([1, 2], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"This formulation is compact, and it closely matches the algebraic formulation of the model we wrote out above.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-II:-algebraic-to-functional","page":"Algebraic modeling languages","title":"Step II: algebraic to functional","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"For the next step, JuMP's macros re-write the variables and constraints into a functional form. Here's what the JuMP code looks like after this step:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using JuMP, HiGHS\n\njulia> function nonalgebraic_knapsack(c, w, b)\n           n = length(c)\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           x = [VariableRef(model) for i = 1:n]\n           for i = 1:n\n               set_lower_bound(x[i], 0)\n               set_integer(x[i])\n               set_name(x[i], \"x[$i]\")\n           end\n           obj = AffExpr(0.0)\n           for i = 1:n\n               add_to_expression!(obj, c[i], x[i])\n           end\n           set_objective(model, MAX_SENSE, obj)\n           lhs = AffExpr(0.0)\n           for i = 1:n\n               add_to_expression!(lhs, w[i], x[i])\n           end\n           con = build_constraint(error, lhs, MOI.LessThan(b))\n           add_constraint(model, con)\n           optimize!(model)\n           return value.(x)\n       end\nnonalgebraic_knapsack (generic function with 1 method)\n\njulia> nonalgebraic_knapsack([1, 2], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"Hopefully you agree that the macro version is much easier to read.","category":"page"},{"location":"background/algebraic_modeling_languages/#Part-III:-JuMP-to-MathOptInterface","page":"Algebraic modeling languages","title":"Part III: JuMP to MathOptInterface","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"In the third step, JuMP converts the functional form of the problem, that is, nonalgebraic_knapsack, into the MathOptInterface API:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> import MathOptInterface as MOI\n\njulia> import HiGHS\n\njulia> function mathoptinterface_knapsack(optimizer, c, w, b)\n           n = length(c)\n           model = MOI.instantiate(optimizer)\n           MOI.set(model, MOI.Silent(), true)\n           x = MOI.add_variables(model, n)\n           for i in 1:n\n               MOI.add_constraint(model, x[i], MOI.GreaterThan(0.0))\n               MOI.add_constraint(model, x[i], MOI.Integer())\n               MOI.set(model, MOI.VariableName(), x[i], \"x[$i]\")\n           end\n           MOI.set(model, MOI.ObjectiveSense(), MOI.MAX_SENSE)\n           obj = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(c, x), 0.0)\n           MOI.set(model, MOI.ObjectiveFunction{typeof(obj)}(), obj)\n           MOI.add_constraint(\n               model,\n               MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(w, x), 0.0),\n               MOI.LessThan(b),\n           )\n           MOI.optimize!(model)\n           return MOI.get.(model, MOI.VariablePrimal(), x)\n       end\nmathoptinterface_knapsack (generic function with 1 method)\n\njulia> mathoptinterface_knapsack(HiGHS.Optimizer, [1.0, 2.0], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"The code is becoming more verbose and looking less like the mathematical formulation that we started with.","category":"page"},{"location":"background/algebraic_modeling_languages/#Step-IV:-MathOptInterface-to-HiGHS","page":"Algebraic modeling languages","title":"Step IV: MathOptInterface to HiGHS","text":"","category":"section"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"As a final step, the HiGHS.jl package converts the MathOptInterface form, that is, mathoptinterface_knapsack, into a HiGHS-specific API:","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"julia> using HiGHS\n\njulia> function highs_knapsack(c, w, b)\n           n = length(c)\n           model = Highs_create()\n           Highs_setBoolOptionValue(model, \"output_flag\", false)\n           for i in 1:n\n               Highs_addCol(model, c[i], 0.0, Inf, 0, C_NULL, C_NULL)\n               Highs_changeColIntegrality(model, i-1, 1)\n           end\n           Highs_changeObjectiveSense(model, -1)\n           Highs_addRow(\n               model,\n               -Inf,\n               b,\n               Cint(length(w)),\n               collect(Cint(0):Cint(n-1)),\n               w,\n           )\n           Highs_run(model)\n           x = fill(NaN, 2)\n           Highs_getSolution(model, x, C_NULL, C_NULL, C_NULL)\n           Highs_destroy(model)\n           return x\n       end\nhighs_knapsack (generic function with 1 method)\n\njulia> highs_knapsack([1.0, 2.0], [0.5, 0.5], 1.25)\n2-element Vector{Float64}:\n 0.0\n 2.0","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"We've now gone from a algebraic model that looked identical to the mathematical model we started with, to a verbose function that uses HiGHS-specific functionality.","category":"page"},{"location":"background/algebraic_modeling_languages/","page":"Algebraic modeling languages","title":"Algebraic modeling languages","text":"The difference between algebraic_knapsack and highs_knapsack highlights the benefit that algebraic modeling languages provide to users. Moreover, if we used a different solver, the solver-specific function would be entirely different. A key benefit of an algebraic modeling language is that you can change the solver without needing to rewrite the model.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"EditURL = \"https://github.com/jump-dev/AmplNLWriter.jl/blob/v1.2.0/README.md\"","category":"page"},{"location":"packages/AmplNLWriter/#AmplNLWriter.jl","page":"jump-dev/AmplNLWriter.jl","title":"AmplNLWriter.jl","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"(Image: Build Status) (Image: MINLPTests) (Image: codecov)","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"AmplNLWriter.jl is an interface between MathOptInterface.jl and AMPL-enabled solvers.","category":"page"},{"location":"packages/AmplNLWriter/#Affiliation","page":"jump-dev/AmplNLWriter.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"This wrapper is maintained by the JuMP community and has no official connection with the AMPL modeling language or AMPL Optimization Inc.","category":"page"},{"location":"packages/AmplNLWriter/#Installation","page":"jump-dev/AmplNLWriter.jl","title":"Installation","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Install AmplNLWriter using Pkg.add:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"import Pkg\nPkg.add(\"AmplNLWriter\")","category":"page"},{"location":"packages/AmplNLWriter/#Use-with-JuMP","page":"jump-dev/AmplNLWriter.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"AmplNLWriter requires an AMPL compatible solver binary to function.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Pass a string pointing to any AMPL-compatible solver binary as the first positional argument to AmplNLWriter.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"For example, if the bonmin executable is on the system path, use:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter\nmodel = Model(() -> AmplNLWriter.Optimizer(\"bonmin\"))","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"If the solver is not on the system path, pass the full path to the solver:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter\nmodel = Model(() -> AmplNLWriter.Optimizer(\"/Users/Oscar/ampl.macos64/bonmin\"))","category":"page"},{"location":"packages/AmplNLWriter/#Precompiled-binaries","page":"jump-dev/AmplNLWriter.jl","title":"Precompiled binaries","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"To simplify the process of installing solver binaries, a number of Julia packages provide precompiled binaries that are compatible with AmplNLWriter. These are generally the name of the solver, followed by _jll. For example, bomin is provided by the Bonmin_jll package.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"To call Bonmin via AmplNLWriter.jl, install the Bonmin_jll package, then run:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP, AmplNLWriter, Bonmin_jll\nmodel = Model(() -> AmplNLWriter.Optimizer(Bonmin_jll.amplexe))","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Supported packages include:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Solver Julia Package Executable\nBonmin Bonmin_jll.jl Bomin_jll.amplexe\nCouenne Couenne_jll.jl Couenne_jll.amplexe\nIpopt Ipopt_jll.jl Ipopt_jll.amplexe\nSHOT SHOT_jll.jl SHOT_jll.amplexe\nKNITRO KNITRO.jl KNITRO.amplexe","category":"page"},{"location":"packages/AmplNLWriter/#MathOptInterface-API","page":"jump-dev/AmplNLWriter.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"The AmplNLWriter optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.Reals","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"MOI.NLPBlock()\nMOI.Name()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Note that some solver executables may not support the full list of constraint types. For example, Ipopt_jll does not support MOI.Integer or MOI.ZeroOne constraints.","category":"page"},{"location":"packages/AmplNLWriter/#Options","page":"jump-dev/AmplNLWriter.jl","title":"Options","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"A list of available options for each solver can be found here:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Bonmin (plus Ipopt options)\nCouenne (plus Ipopt and Bonmin options)\nIpopt\nSHOT","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Set an option using set_attribute. For example, to set the \"bonmin.nlp_log_level\" option to 0 in Bonmin, use:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"using JuMP\nimport AmplNLWriter\nimport Bonmin_jll\nmodel = Model(() -> AmplNLWriter.Optimizer(Bonmin_jll.amplexe))\nset_attribute(model, \"bonmin.nlp_log_level\", 0)","category":"page"},{"location":"packages/AmplNLWriter/#opt-files","page":"jump-dev/AmplNLWriter.jl","title":"opt files","text":"","category":"section"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"Some options need to be specified via an .opt file.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"This file must be located in the current working directory whenever the model is solved.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"The .opt file must be named after the name of the solver, for example, bonmin.opt, and each line must contain an option name and the desired value, separated by a space.","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"For example, to set the absolute and relative tolerances in Couenne to 1 and 0.05 respectively, the couenne.opt file should contain:","category":"page"},{"location":"packages/AmplNLWriter/","page":"jump-dev/AmplNLWriter.jl","title":"jump-dev/AmplNLWriter.jl","text":"allowable_gap 1\nallowable_fraction_gap 0.05","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"<img class=\"display-light-only\" src=\"assets/logo-with-text.svg\" alt=\"JuMP logo\"/>\n<img class=\"display-dark-only\" src=\"assets/logo-dark-with-text.svg\" alt=\"JuMP logo\"/>","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"# These comments do not display in the HTML output.\n# See https://github.com/JuliaDocs/Documenter.jl/issues/674.","category":"page"},{"location":"#Introduction","page":"Introduction","title":"Introduction","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"\\begin{figure}[h]\n  \\includegraphics[width=0.5\\textwidth]{assets/logo-with-text.pdf}\n\\end{figure}","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Welcome to the documentation for JuMP.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"note: Note\nThis documentation is also available in PDF format: JuMP.pdf.","category":"page"},{"location":"#What-is-JuMP?","page":"Introduction","title":"What is JuMP?","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is a domain-specific modeling language for mathematical optimization embedded in Julia. It currently supports a number of open-source and commercial solvers for a variety of problem classes, including linear, mixed-integer, second-order conic, semidefinite, and nonlinear programming.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"tip: Tip\nIf you aren't sure if you should use JuMP, read Should you use JuMP?.","category":"page"},{"location":"#Resources-for-getting-started","page":"Introduction","title":"Resources for getting started","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"There are few ways to get started with JuMP:","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Read the Installation Guide.\nRead the introductory tutorials Getting started with Julia and Getting started with JuMP.\nBrowse some of our modeling tutorials, including classics such as The diet problem, or the Maximum likelihood estimation problem using nonlinear programming.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"tip: Tip\nNeed help? Join the community forum to search for answers to commonly asked questions.Before asking a question, make sure to read the post make it easier to help you, which contains a number of tips on how to ask a good question.","category":"page"},{"location":"#How-the-documentation-is-structured","page":"Introduction","title":"How the documentation is structured","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"Having a high-level overview of how this documentation is structured will help you know where to look for certain things.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Tutorials contain worked examples of solving problems with JuMP. Start here if you are new to JuMP, or you have a particular problem class you want to model.\nThe Manual contains short code-snippets that explain how to achieve specific tasks in JuMP. Look here if you want to know how to achieve a particular task, such as how to Delete a variable or how to Modify an objective coefficient.\nThe API Reference contains a complete list of the functions you can use in JuMP. Look here if you want to know how to use a particular function.\nThe Background information section contains background reading material to provide context to JuMP. Look here if you want an understanding of what JuMP is and why we created it, rather than how to use it.\nThe Developer docs section contains information for people contributing to JuMP development or writing JuMP extensions. Don't worry about this section if you are using JuMP to formulate and solve problems as a user.\nThe MathOptInterface section is a self-contained copy of the documentation for MathOptInterface. Look here for functions and constants beginning with MOI., as well as for general information on how MathOptInterface works.","category":"page"},{"location":"#Citing-JuMP","page":"Introduction","title":"Citing JuMP","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"If you find JuMP useful in your work, we kindly request that you cite the following paper (preprint):","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"@article{Lubin2023,\n    author = {Miles Lubin and Oscar Dowson and Joaquim {Dias Garcia} and Joey Huchette and Beno{\\^i}t Legat and Juan Pablo Vielma},\n    title = {{JuMP} 1.0: {R}ecent improvements to a modeling language for mathematical optimization},\n    journal = {Mathematical Programming Computation},\n    year = {2023},\n    doi = {10.1007/s12532-023-00239-3}\n}","category":"page"},{"location":"#NumFOCUS","page":"Introduction","title":"NumFOCUS","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"(Image: NumFOCUS logo)","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is a Sponsored Project of NumFOCUS, a 501(c)(3) nonprofit charity in the United States. NumFOCUS provides JuMP with fiscal, legal, and administrative support to help ensure the health and sustainability of the project. Visit numfocus.org for more information.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"You can support JuMP by donating.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"Donations to JuMP are managed by NumFOCUS. For donors in the United States, your gift is tax-deductible to the extent provided by law. As with any donation, you should consult with your tax adviser about your particular tax situation.","category":"page"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP's largest expense is the annual JuMP-dev workshop. Donations will help us provide travel support for JuMP-dev attendees and take advantage of other opportunities that arise to support JuMP development.","category":"page"},{"location":"#License","page":"Introduction","title":"License","text":"","category":"section"},{"location":"","page":"Introduction","title":"Introduction","text":"JuMP is licensed under the MPL-2.0 software license. Consult the license and the Mozilla FAQ for more information. In addition, JuMP is typically used in conjunction with solver packages and extensions which have their own licences. Consult their package repositories for the specific licenses that apply.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"EditURL = \"https://github.com/infiniteopt/InfiniteOpt.jl/blob/v0.5.8/docs/jump/README.md\"","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"(Image: Logo)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt.jl is a JuMP extension for expressing and solving infinite-dimensional optimization problems. Such areas include stochastic programming, dynamic programming, space-time optimization, and more. InfiniteOpt serves as an easy-to-use modeling interface for these advanced problem types that can be used by those with little background in these areas. It also it contains a wealth of capabilities making it a powerful and convenient tool for advanced users.  ","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Current Version Documentation Build Status Citation\n(Image: ) (Image: ) (Image: Build Status) (Image: codecov.io) (Image: DOI)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt builds upon JuMP to add support for many complex modeling objects which  include:","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Infinite parameters (for example, time, space, and/or uncertainty)\nFinite parameters (similar to ParameterJuMP)\nInfinite variables (decision functions) (for example, y(t x))\nDerivatives (for example, fracpartial y(t x)partial t)\nMeasures (for example, int_t in Ty(t x)dt and mathbbEy(xi))\nFirst class nonlinear modeling","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"The unifying modeling abstraction behind InfiniteOpt captures a wide spectrum  of disciplines which include dynamic, PDE, stochastic, and semi-infinite  optimization. Moreover, we facilitate transferring techniques between these  to synthesize new optimization paradigms.","category":"page"},{"location":"packages/InfiniteOpt/#License","page":"infiniteopt/InfiniteOpt.jl","title":"License","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt is licensed under the MIT \"Expat\" license.","category":"page"},{"location":"packages/InfiniteOpt/#Installation","page":"infiniteopt/InfiniteOpt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"InfiniteOpt.jl is a registered Julia package and  can be installed by entering the following in the REPL.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"julia> import Pkg; Pkg.add(\"InfiniteOpt\")","category":"page"},{"location":"packages/InfiniteOpt/#Documentation","page":"infiniteopt/InfiniteOpt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"Please visit our documentation pages  to learn more. These pages are quite extensive and feature overviews, guides, manuals, tutorials, examples, and more.","category":"page"},{"location":"packages/InfiniteOpt/#Questions","page":"infiniteopt/InfiniteOpt.jl","title":"Questions","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"For additional help please visit and post in our  discussion forum.","category":"page"},{"location":"packages/InfiniteOpt/#Citing","page":"infiniteopt/InfiniteOpt.jl","title":"Citing","text":"","category":"section"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"(Image: DOI)  (Image: DOI)","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"If you use InfiniteOpt.jl in your research, we would greatly appreciate your  citing it.","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"@article{pulsipher2022unifying,\n      title = {A unifying modeling abstraction for infinite-dimensional optimization},\n      journal = {Computers & Chemical Engineering},\n      volume = {156},\n      year = {2022},\n      issn = {0098-1354},\n      doi = {https://doi.org/10.1016/j.compchemeng.2021.107567},\n      url = {https://www.sciencedirect.com/science/article/pii/S0098135421003458},\n      author = {Joshua L. Pulsipher and Weiqi Zhang and Tyler J. Hongisto and Victor M. Zavala},\n}","category":"page"},{"location":"packages/InfiniteOpt/","page":"infiniteopt/InfiniteOpt.jl","title":"infiniteopt/InfiniteOpt.jl","text":"A pre-print version is freely available though arXiv.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"EditURL = \"getting_started_with_sets_and_indexing.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Getting-started-with-sets-and-indexing","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Most introductory courses to linear programming will teach you to identify sets over which the decision variables and constraints are indexed. Therefore, it is common to write variables such as x_i for all i in I.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"A common stumbling block for new users to JuMP is that JuMP does not provide specialized syntax for constructing and manipulating these sets.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We made this decision because Julia already provides a wealth of data structures for working with sets.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"In contrast, because tools like AMPL are stand-alone software packages, they had to define their own syntax for set construction and manipulation. Indeed, the AMPL Book has two entire chapters devoted to sets and indexing (Chapter 5, \"Simple Sets and Indexing,\" and Chapter 6, \"Compound Sets and Indexing\").","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The purpose of this tutorial is to demonstrate a variety of ways in which you can construct and manipulate sets for optimization models.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"If you haven't already, you should first read Getting started with JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"using JuMP","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Unordered-sets","page":"Getting started with sets and indexing","title":"Unordered sets","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Unordered sets are useful to describe non-numeric indices, such as the names of cities or types of products.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The most common way to construct a set is by creating a vector:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"animals = [\"dog\", \"cat\", \"chicken\", \"cow\", \"pig\"]\nmodel = Model()\n@variable(model, x[animals])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We can also use things like the keys of a dictionary:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"weight_of_animals = Dict(\n    \"dog\" => 20.0,\n    \"cat\" => 5.0,\n    \"chicken\" => 2.0,\n    \"cow\" => 720.0,\n    \"pig\" => 150.0,\n)\nanimal_keys = keys(weight_of_animals)\nmodel = Model()\n@variable(model, x[animal_keys])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"A third option is to use Julia's Set object.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"animal_set = Set()\nfor animal in keys(weight_of_animals)\n    push!(animal_set, animal)\nend\nanimal_set","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The nice thing about Sets is that they automatically remove duplicates:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"push!(animal_set, \"dog\")\nanimal_set","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Note how dog does not appear twice.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[animal_set])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-of-numbers","page":"Getting started with sets and indexing","title":"Sets of numbers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Sets of numbers are useful to describe sets that are ordered, such as years or elements in a vector. The easiest way to create sets of numbers is to use Julia's range syntax.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"These can start at 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[1:4])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"but they don't have to:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[2012:2021])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Ranges also have a start:step:stop syntax. So the Olympic years are:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[1896:4:2020])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-of-other-things","page":"Getting started with sets and indexing","title":"Sets of other things","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"An important observation is that you can have any Julia type as the element of a set. It doesn't have to be a String or a Number. For example, you can have tuples:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"sources = [\"A\", \"B\", \"C\"]\nsinks = [\"D\", \"E\"]\nS = [(source, sink) for source in sources, sink in sinks]\nmodel = Model()\n@variable(model, x[S])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"x[(\"A\", \"D\")]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"For multi-dimensional sets, you can use JuMP's syntax for constructing Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[sources, sinks])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"x[\"A\", \"D\"]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"info: Info\nNote how we indexed x[\"A\", \"D\"] instead of x[(\"A\", \"D\")] as above.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Sets-to-watch-out-for","page":"Getting started with sets and indexing","title":"Sets to watch out for","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"JuMP supports any sets which are iterable, that is, the set set supports a for-loop like: [i for i in set]. This causes a few common errors.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"First, if T = 3, you may pass the integer T by mistake instead of a range like 1:T:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\nT = 3\n@variable(model, x[T])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This results in a single variable being created, instead of three as desired. Because this is a common error, a warning is printed, advising you to pass a Vector{Int} instead:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, x_fixed[[T]])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Second, because Strings are iterable, passing a \"index\" as a singleton index is the same as passing ['i', 'n', 'd', 'e', 'x']:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, y[\"index\"])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This time, a warning is not printed, but the work-around is similar, pass a Vector{String} instead:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"@variable(model, y_fixed[[\"index\"]])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"tip: Tip\nAs a rule of thumb, if you want an index with one element, avoid confusion by passing [index] instead of index.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Set-operations","page":"Getting started with sets and indexing","title":"Set operations","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Julia has built-in support for set operations such as union, intersect, and setdiff.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Therefore, to create a set of all years in which the summer Olympics were held, we can use:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"baseline = 1896:4:2020\ncancelled = [1916, 1940, 1944, 2020]\noff_year = [2021]\nolympic_years = union(setdiff(baseline, cancelled), off_year)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"You can also find the number of elements (that is, the cardinality) in a set using length:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"length(olympic_years)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Set-membership-operations","page":"Getting started with sets and indexing","title":"Set membership operations","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"To compute membership of sets, use the in function.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"2000 in olympic_years","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"2001 in olympic_years","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Indexing-expressions","page":"Getting started with sets and indexing","title":"Indexing expressions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Use Julia's generator syntax to compute new sets, such as the list of Olympic years that are divisible by 3:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"olympic_3_years = [year for year in olympic_years if mod(year, 3) == 0]\nmodel = Model()\n@variable(model, x[olympic_3_years])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Alternatively, use JuMP's syntax for constructing Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[year in olympic_years; mod(year, 3) == 0])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Compound-sets","page":"Getting started with sets and indexing","title":"Compound sets","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Consider a transportation problem in which we need to ship goods between cities. We have been provided a list of cities:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"cities = [\"Auckland\", \"Wellington\", \"Christchurch\", \"Dunedin\"]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"and a distance matrix which records the shipping distance between pairs of cities. If we can't ship between two cities, the distance is 0.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"distances = [0 643 1071 1426; 0 0 436 790; 0 0 0 360; 1426 0 0 0]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Let's have a look at ways we could write a model with an objective function to minimize the total shipping cost. For simplicity, we'll ignore all constraints.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Fix-unused-variables","page":"Getting started with sets and indexing","title":"Fix unused variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"One approach is to fix all variables that we can't use to zero. Most solvers are smart-enough to remove these during a presolve phase, so it has a very small impact on performance:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[1:N, 1:N] >= 0)\nfor i in 1:N, j in 1:N\n    if distances[i, j] == 0\n        fix(x[i, j], 0.0; force = true)\n    end\nend\n@objective(model, Min, sum(distances[i, j] * x[i, j] for i in 1:N, j in 1:N))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Filtered-summation","page":"Getting started with sets and indexing","title":"Filtered summation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Another approach is to define filters whenever we want to sum over our decision variables:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[1:N, 1:N] >= 0)\n@objective(\n    model,\n    Min,\n    sum(\n        distances[i, j] * x[i, j] for i in 1:N, j in 1:N if distances[i, j] > 0\n    ),\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Filtered-indexing","page":"Getting started with sets and indexing","title":"Filtered indexing","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"We could also use JuMP's support for Containers:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"N = length(cities)\nmodel = Model()\n@variable(model, x[i = 1:N, j = 1:N; distances[i, j] > 0])\n@objective(model, Min, sum(distances[i...] * x[i] for i in eachindex(x)))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"note: Note\nThe i... is called  a \"splat.\" It converts a tuple like (1, 2) into two indices like distances[1, 2].","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Converting-to-a-different-data-structure","page":"Getting started with sets and indexing","title":"Converting to a different data structure","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Another approach, and one that is often the most readable, is to convert the data you have into something that is easier to work with. Originally, we had a vector of strings and a matrix of distances. What we really need is something that maps usable origin-destination pairs to distances. A dictionary is an obvious choice:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"routes = Dict(\n    (a, b) => distances[i, j] for\n    (i, a) in enumerate(cities), (j, b) in enumerate(cities) if\n    distances[i, j] > 0\n)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Then, we can create our model like so:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"model = Model()\n@variable(model, x[keys(routes)])\n@objective(model, Min, sum(v * x[k] for (k, v) in routes))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This has a number of benefits over the other approaches, including a compacter algebraic model and variables that are named in a more meaningful way.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"tip: Tip\nIf you're struggling to formulate a problem using the available syntax in JuMP, it's probably a sign that you should convert your data into a different form.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/#Next-steps","page":"Getting started with sets and indexing","title":"Next steps","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"The purpose of this tutorial was to show how JuMP does not have specialized syntax for set creation and manipulation. Instead, you should use the tools provided by Julia itself.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"This is both an opportunity and a challenge, because you are free to pick the syntax and data structures that best suit your problem, but for new users it can be daunting to decide which structure to use.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_sets_and_indexing/","page":"Getting started with sets and indexing","title":"Getting started with sets and indexing","text":"Read through some of the other JuMP tutorials to get inspiration and ideas for how you can use Julia's syntax and data structures to your advantage.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"EditURL = \"https://github.com/jump-dev/Gurobi.jl/blob/v1.0.4/README.md\"","category":"page"},{"location":"packages/Gurobi/#Gurobi.jl","page":"jump-dev/Gurobi.jl","title":"Gurobi.jl","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi.jl is a wrapper for the Gurobi Optimizer.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"It has two components:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Gurobi/#Affiliation","page":"jump-dev/Gurobi.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This wrapper is maintained by the JuMP community with help from Gurobi. ","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you encounter a problem with this interface, please either open an issue in this repository directly or create a topic in the Julia Discourse with the gurobi tag. ","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you encounter a problem with the Gurobi solver, please post in Gurobi’s Community Forum, or if you are a commercial customer, please contact Gurobi directly through the Gurobi Help Center.","category":"page"},{"location":"packages/Gurobi/#License","page":"jump-dev/Gurobi.jl","title":"License","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The underlying solver is a closed-source commercial product for which you must obtain a license.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Free Gurobi licenses are available for academics and students.","category":"page"},{"location":"packages/Gurobi/#Installation","page":"jump-dev/Gurobi.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"First, obtain a license of Gurobi and install Gurobi solver.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Then, set the GUROBI_HOME environment variable as appropriate and run Pkg.add(\"Gurobi\"):","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"# On Windows, this might be\nENV[\"GUROBI_HOME\"] = \"C:\\\\Program Files\\\\gurobi1000\\\\win64\"\n# ... or perhaps ...\nENV[\"GUROBI_HOME\"] = \"C:\\\\gurobi1000\\\\win64\"\n# On Mac, this might be\nENV[\"GUROBI_HOME\"] = \"/Library/gurobi1000/mac64\"\n\nimport Pkg\nPkg.add(\"Gurobi\")","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Note: your path may differ. Check which folder you installed Gurobi in, and update the path accordingly.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"By default, building Gurobi.jl will fail if the Gurobi library is not found. This may not be desirable in certain cases, for example when part of a package's test suite uses Gurobi as an optional test dependency, but Gurobi cannot be installed on a CI server running the test suite. To support this use case, the GUROBI_JL_SKIP_LIB_CHECK environment variable may be set (to any value) to make Gurobi.jl installable (but not usable).","category":"page"},{"location":"packages/Gurobi/#Use-with-JuMP","page":"jump-dev/Gurobi.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"To use Gurobi with JuMP, use Gurobi.Optimizer:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nmodel = Model(Gurobi.Optimizer)\nset_attribute(model, \"TimeLimit\", 100)\nset_attribute(model, \"Presolve\", 0)","category":"page"},{"location":"packages/Gurobi/#MathOptInterface-API","page":"jump-dev/Gurobi.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The Gurobi optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}\nMOI.ObjectiveFunction{MOI.VectorAffineFunction{Float64}}","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.Reals","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.Semicontinuous{Float64}\nMOI.VariableIndex in MOI.Semiinteger{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorOfVariables in MOI.SOS1{Float64}\nMOI.VectorOfVariables in MOI.SOS2{Float64}\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"MOI.HeuristicCallback()\nMOI.LazyConstraintCallback()\nMOI.Name()\nMOI.ObjectiveSense()\nMOI.UserCutCallback()","category":"page"},{"location":"packages/Gurobi/#Options","page":"jump-dev/Gurobi.jl","title":"Options","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi Documentation for a list and description of allowable parameters.","category":"page"},{"location":"packages/Gurobi/#C-API","page":"jump-dev/Gurobi.jl","title":"C API","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The C API can be accessed via Gurobi.GRBxx functions, where the names and arguments are identical to the C API.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi documentation for details.","category":"page"},{"location":"packages/Gurobi/#Reusing-the-same-Gurobi-environment-for-multiple-solves","page":"jump-dev/Gurobi.jl","title":"Reusing the same Gurobi environment for multiple solves","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"When using this package via other packages such as JuMP.jl, the default behavior is to obtain a new Gurobi license token every time a model is created. If you are using Gurobi in a setting where the number of concurrent Gurobi uses is limited (for example, \"Single-Use\" or \"Floating-Use\" licenses), you might instead prefer to obtain a single license token that is shared by all models that your program solves.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"You can do this by passing a Gurobi.Env() object as the first parameter to Gurobi.Optimizer. For example:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nconst GRB_ENV = Gurobi.Env()\n\nmodel_1 = Model(() -> Gurobi.Optimizer(GRB_ENV))\n\n# The solvers can have different options too\nmodel_2 = direct_model(Gurobi.Optimizer(GRB_ENV))\nset_attribute(model_2, \"OutputFlag\", 0)","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"If you create a module with a Gurobi.Env as a module-level constant, use an __init__ function to ensure that a new environment is created each time the module is loaded:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"module MyModule\n\nimport Gurobi\n\nconst GRB_ENV_REF = Ref{Gurobi.Env}()\n\nfunction __init__()\n    global GRB_ENV_REF\n    GRB_ENV_REF[] = Gurobi.Env()\n    return\nend\n\n# Note the need for GRB_ENV_REF[] not GRB_ENV_REF\ncreate_optimizer() = Gurobi.Optimizer(GRB_ENV_REF[])\n\nend","category":"page"},{"location":"packages/Gurobi/#Accessing-Gurobi-specific-attributes","page":"jump-dev/Gurobi.jl","title":"Accessing Gurobi-specific attributes","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Get and set Gurobi-specific variable, constraint, and model attributes as follows:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi\nmodel = direct_model(Gurobi.Optimizer())\n@variable(model, x >= 0)\n@constraint(model, c, 2x >= 1)\n@objective(model, Min, x)\nMOI.set(model, Gurobi.ConstraintAttribute(\"Lazy\"), c, 2)\noptimize!(model)\nMOI.get(model, Gurobi.VariableAttribute(\"LB\"), x)  # Returns 0.0\nMOI.get(model, Gurobi.ModelAttribute(\"NumConstrs\")) # Returns 1","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"A complete list of supported Gurobi attributes can be found in their online documentation.","category":"page"},{"location":"packages/Gurobi/#Callbacks","page":"jump-dev/Gurobi.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Here is an example using Gurobi's solver-specific callbacks.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"using JuMP, Gurobi, Test\n\nmodel = direct_model(Gurobi.Optimizer())\n@variable(model, 0 <= x <= 2.5, Int)\n@variable(model, 0 <= y <= 2.5, Int)\n@objective(model, Max, y)\ncb_calls = Cint[]\nfunction my_callback_function(cb_data, cb_where::Cint)\n    # You can reference variables outside the function as normal\n    push!(cb_calls, cb_where)\n    # You can select where the callback is run\n    if cb_where != GRB_CB_MIPSOL && cb_where != GRB_CB_MIPNODE\n        return\n    end\n    # You can query a callback attribute using GRBcbget\n    if cb_where == GRB_CB_MIPNODE\n        resultP = Ref{Cint}()\n        GRBcbget(cb_data, cb_where, GRB_CB_MIPNODE_STATUS, resultP)\n        if resultP[] != GRB_OPTIMAL\n            return  # Solution is something other than optimal.\n        end\n    end\n    # Before querying `callback_value`, you must call:\n    Gurobi.load_callback_variable_primal(cb_data, cb_where)\n    x_val = callback_value(cb_data, x)\n    y_val = callback_value(cb_data, y)\n    # You can submit solver-independent MathOptInterface attributes such as\n    # lazy constraints, user-cuts, and heuristic solutions.\n    if y_val - x_val > 1 + 1e-6\n        con = @build_constraint(y - x <= 1)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    elseif y_val + x_val > 3 + 1e-6\n        con = @build_constraint(y + x <= 3)\n        MOI.submit(model, MOI.LazyConstraint(cb_data), con)\n    end\n    if rand() < 0.1\n        # You can terminate the callback as follows:\n        GRBterminate(backend(model))\n    end\n    return\nend\n# You _must_ set this parameter if using lazy constraints.\nMOI.set(model, MOI.RawOptimizerAttribute(\"LazyConstraints\"), 1)\nMOI.set(model, Gurobi.CallbackFunction(), my_callback_function)\noptimize!(model)\n@test termination_status(model) == MOI.OPTIMAL\n@test primal_status(model) == MOI.FEASIBLE_POINT\n@test value(x) == 1\n@test value(y) == 2","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"See the Gurobi documentation for other information that can be queried with GRBcbget.","category":"page"},{"location":"packages/Gurobi/#Common-Performance-Pitfall-with-JuMP","page":"jump-dev/Gurobi.jl","title":"Common Performance Pitfall with JuMP","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Gurobi's API works differently than most solvers. Any changes to the model are not applied immediately, but instead go sit in a internal buffer (making any modifications appear to be instantaneous) waiting for a call to GRBupdatemodel (where the work is  done).","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This leads to a common performance pitfall that has the following message as its main symptom:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Warning: excessive time spent in model updates. Consider calling update less frequently.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"This often means the JuMP program was structured in such a way that Gurobi.jl ends up calling GRBupdatemodel in each iteration of a loop.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Usually, it is possible (and easy) to restructure the JuMP program in a way it stays ssolver-agnostic and has a close-to-ideal performance with Gurobi.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"To guide such restructuring it is good to keep in mind the following bits of information:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"GRBupdatemodel is only called if changes were done since last GRBupdatemodel (that is, if the internal buffer is not empty).\nGRBupdatemodel is called when JuMP.optimize! is called, but this often is not the source of the problem.\nGRBupdatemodel may be called when any model attribute is queried, even if that specific attribute was not changed. This often the source of the problem.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"The worst-case scenario is, therefore, a loop of modify-query-modify-query, even if what is being modified and what is being queried are two completely distinct things.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"As an example, instead of:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\n@variable(model, x[1:100] >= 0)\nfor i in 1:100\n    set_upper_bound(x[i], i)\n    # `GRBupdatemodel` called on each iteration of this loop.\n    println(lower_bound(x[i]))\nend","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"do","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\n@variable(model, x[1:100] >= 0)\n# All modifications are done before any queries.\nfor i in 1:100\n    set_upper_bound(x[i], i)\nend\nfor i in 1:100\n    # Only the first `lower_bound` query may trigger an `GRBupdatemodel`.\n    println(lower_bound(x[i]))\nend","category":"page"},{"location":"packages/Gurobi/#Common-errors","page":"jump-dev/Gurobi.jl","title":"Common errors","text":"","category":"section"},{"location":"packages/Gurobi/#Using-Gurobi-v9.0-and-you-got-an-error-like-Q-not-PSD?","page":"jump-dev/Gurobi.jl","title":"Using Gurobi v9.0 and you got an error like Q not PSD?","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"You need to set the NonConvex parameter:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"model = Model(Gurobi.Optimizer)\nset_optimizer_attribute(model, \"NonConvex\", 2)","category":"page"},{"location":"packages/Gurobi/#Gurobi-Error-1009:-Version-number-is-XX.X,-license-is-for-version-XX.X","page":"jump-dev/Gurobi.jl","title":"Gurobi Error 1009: Version number is XX.X, license is for version XX.X","text":"","category":"section"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Make sure that your license is correct for your Gurobi version. See the Gurobi documentation for details.","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"Once you are sure that the license and Gurobi versions match, re-install Gurobi.jl by running:","category":"page"},{"location":"packages/Gurobi/","page":"jump-dev/Gurobi.jl","title":"jump-dev/Gurobi.jl","text":"import Pkg\nPkg.build(\"Gurobi\")","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/submodules/Benchmarks/overview.md\"","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/submodules/Benchmarks/overview/#The-Benchmarks-submodule","page":"Overview","title":"The Benchmarks submodule","text":"","category":"section"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"To aid the development of efficient solver wrappers, MathOptInterface provides benchmarking capability. Benchmarking a wrapper follows a two-step process.","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"First, prior to making changes, create a baseline for the benchmark results on a given benchmark suite as follows:","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"using SolverPackage  # Replace with your choice of solver.\nimport MathOptInterface as MOI\n\nsuite = MOI.Benchmarks.suite() do\n    SolverPackage.Optimizer()\nend\n\nMOI.Benchmarks.create_baseline(\n    suite, \"current\"; directory = \"/tmp\", verbose = true\n)","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"Use the exclude argument to Benchmarks.suite to exclude benchmarks that the solver doesn't support.","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"Second, after making changes to the package, re-run the benchmark suite and compare to the prior saved results:","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"using SolverPackage\nimport MathOptInterface as MOI\n\nsuite = MOI.Benchmarks.suite() do\n    SolverPackage.Optimizer()\nend\n\nMOI.Benchmarks.compare_against_baseline(\n    suite, \"current\"; directory = \"/tmp\", verbose = true\n)","category":"page"},{"location":"moi/submodules/Benchmarks/overview/","page":"Overview","title":"Overview","text":"This comparison will create a report detailing improvements and regressions.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/latency.md\"","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/latency/#Latency","page":"Latency","title":"Latency","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"MathOptInterface suffers the \"time-to-first-solve\" problem of start-up latency.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This hurts both the user- and developer-experience of MathOptInterface. In the first case, because simple models have a multi-second delay before solving, and in the latter, because our tests take so long to run.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This page contains some advice on profiling and fixing latency-related problems in the MathOptInterface.jl repository.","category":"page"},{"location":"moi/tutorials/latency/#Background","page":"Latency","title":"Background","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Before reading this part of the documentation, you should familiarize yourself with the reasons for latency in Julia and how to fix them.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Read the blogposts on julialang.org on precompilation and SnoopCompile\nRead the SnoopCompile documentation.\nWatch Tim Holy's talk at JuliaCon 2021\nWatch the package development workshop at JuliaCon 2021","category":"page"},{"location":"moi/tutorials/latency/#Causes","page":"Latency","title":"Causes","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are three main causes of latency in MathOptInterface:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"A large number of types\nLack of method ownership\nType-instability in the bridge layer","category":"page"},{"location":"moi/tutorials/latency/#A-large-number-of-types","page":"Latency","title":"A large number of types","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Julia is very good at specializing method calls based on the input type. Each specialization has a compilation cost, but the benefit of faster run-time performance.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"The best-case scenario is for a method to be called a large number of times with a single set of argument types. The worst-case scenario is for a method to be called a single time for a large set of argument types.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Because of MathOptInterface's function-in-set formulation, we fall into the worst-case situation.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This is a fundamental limitation of Julia, so there isn't much we can do about it. However, if we can precompile MathOptInterface, much of the cost can be shifted from start-up latency to the time it takes to precompile a package on installation.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"However, there are two things which make MathOptInterface hard to precompile.","category":"page"},{"location":"moi/tutorials/latency/#Lack-of-method-ownership","page":"Latency","title":"Lack of method ownership","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Lack of method ownership happens when a call is made using a mix of structs and methods from different modules. Because of this, no single module \"owns\" the method that is being dispatched, and so it cannot be precompiled.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"tip: Tip\nThis is a slightly simplified explanation. Read the precompilation tutorial for a more in-depth discussion on back-edges.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Unfortunately, the design of MOI means that this is a frequent occurrence: we have a bunch of types in MOI.Utilities that wrap types defined in external packages (for example, the Optimizers), which implement methods of functions defined in MOI (for example, add_variable, add_constraint).","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Here's a simple example of method-ownership in practice:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"module MyMOI\nstruct Wrapper{T}\n    inner::T\nend\noptimize!(x::Wrapper) = optimize!(x.inner)\nend  # MyMOI\n\nmodule MyOptimizer\nusing ..MyMOI\nstruct Optimizer end\nMyMOI.optimize!(x::Optimizer) = 1\nend  # MyOptimizer\n\nusing SnoopCompile\nmodel = MyMOI.Wrapper(MyOptimizer.Optimizer())\n\njulia> tinf = @snoopi_deep MyMOI.optimize!(model)\nInferenceTimingNode: 0.008256/0.008543 on InferenceFrameInfo for Core.Compiler.Timings.ROOT() with 1 direct children","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"The result is that there was one method that required type inference. If we visualize tinf:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"using ProfileView\nProfileView.view(flamegraph(tinf))","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"we see a flamegraph with a large red-bar indicating that the method MyMOI.optimize(MyMOI.Wrapper{MyOptimizer.Optimizer}) cannot be precompiled.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"To fix this, we need to designate a module to \"own\" that method (that is, create a back-edge). The easiest way to do this is for MyOptimizer to call MyMOI.optimize(MyMOI.Wrapper{MyOptimizer.Optimizer}) during using MyOptimizer. Let's see that in practice:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"module MyMOI\nstruct Wrapper{T}\n    inner::T\nend\noptimize(x::Wrapper) = optimize(x.inner)\nend  # MyMOI\n\nmodule MyOptimizer\nusing ..MyMOI\nstruct Optimizer end\nMyMOI.optimize(x::Optimizer) = 1\n# The syntax of this let-while loop is very particular:\n#  * `let ... end` keeps everything local to avoid polluting the MyOptimizer\n#    namespace\n#  * `while true ... break end` runs the code once, and forces Julia to compile\n#    the inner loop, rather than interpret it.\nlet\n    while true\n        model = MyMOI.Wrapper(Optimizer())\n        MyMOI.optimize(model)\n        break\n    end\nend\nend  # MyOptimizer\n\nusing SnoopCompile\nmodel = MyMOI.Wrapper(MyOptimizer.Optimizer())\n\njulia> tinf = @snoopi_deep MyMOI.optimize(model)\nInferenceTimingNode: 0.006822/0.006822 on InferenceFrameInfo for Core.Compiler.Timings.ROOT() with 0 direct children","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are now 0 direct children that required type inference because the method was already stored in MyOptimizer!","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Unfortunately, this trick only works if the call-chain is fully inferrable. If there are breaks (due to type instability), then the benefit of doing this is reduced. And unfortunately for us, the design of MathOptInterface has a lot of type instabilities.","category":"page"},{"location":"moi/tutorials/latency/#Type-instability-in-the-bridge-layer","page":"Latency","title":"Type instability in the bridge layer","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Most of MathOptInterface is pretty good at ensuring type-stability. However, a key component is not type stable, and that is the bridging layer.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"In particular, the bridging layer defines Bridges.LazyBridgeOptimizer, which has fields like:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"struct LazyBridgeOptimizer\n    constraint_bridge_types::Vector{Any}\n    constraint_node::Dict{Tuple{Type,Type},ConstraintNode}\n    constraint_types::Vector{Tuple{Type,Type}}\nend","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"This is because the LazyBridgeOptimizer needs to be able to deal with any function-in-set type passed to it, and we also allow users to pass additional bridges that they defined in external packages.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"So to recap, MathOptInterface suffers package latency because:","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"there are a large number of types and functions\nand these are split between multiple modules, including external packages\nand there are type-instabilities like those in the bridging layer.","category":"page"},{"location":"moi/tutorials/latency/#Resolutions","page":"Latency","title":"Resolutions","text":"","category":"section"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are no magic solutions to reduce latency. Issue #1313 tracks progress on reducing latency in MathOptInterface.","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"A useful script is the following (replace GLPK as needed):","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"import GLPK\nimport MathOptInterface as MOI\n\nfunction example_diet(optimizer, bridge)\n    category_data = [\n        1800.0 2200.0;\n          91.0    Inf;\n           0.0   65.0;\n           0.0 1779.0\n    ]\n    cost = [2.49, 2.89, 1.50, 1.89, 2.09, 1.99, 2.49, 0.89, 1.59]\n    food_data = [\n        410 24 26 730;\n        420 32 10 1190;\n        560 20 32 1800;\n        380  4 19 270;\n        320 12 10 930;\n        320 15 12 820;\n        320 31 12 1230;\n        100  8 2.5 125;\n        330  8 10 180\n    ]\n    bridge_model = if bridge\n        MOI.instantiate(optimizer; with_bridge_type=Float64)\n    else\n        MOI.instantiate(optimizer)\n    end\n    model = MOI.Utilities.CachingOptimizer(\n        MOI.Utilities.UniversalFallback(MOI.Utilities.Model{Float64}()),\n        MOI.Utilities.AUTOMATIC,\n    )\n    MOI.Utilities.reset_optimizer(model, bridge_model)\n    MOI.set(model, MOI.Silent(), true)\n    nutrition = MOI.add_variables(model, size(category_data, 1))\n    for (i, v) in enumerate(nutrition)\n        MOI.add_constraint(model, v, MOI.GreaterThan(category_data[i, 1]))\n        MOI.add_constraint(model, v, MOI.LessThan(category_data[i, 2]))\n    end\n    buy = MOI.add_variables(model, size(food_data, 1))\n    MOI.add_constraint.(model, buy, MOI.GreaterThan(0.0))\n    MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)\n    f = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(cost, buy), 0.0)\n    MOI.set(model, MOI.ObjectiveFunction{typeof(f)}(), f)\n    for (j, n) in enumerate(nutrition)\n        f = MOI.ScalarAffineFunction(\n            MOI.ScalarAffineTerm.(food_data[:, j], buy),\n            0.0,\n        )\n        push!(f.terms, MOI.ScalarAffineTerm(-1.0, n))\n        MOI.add_constraint(model, f, MOI.EqualTo(0.0))\n    end\n    MOI.optimize!(model)\n    term_status = MOI.get(model, MOI.TerminationStatus())\n    @assert term_status == MOI.OPTIMAL\n    MOI.add_constraint(\n        model,\n        MOI.ScalarAffineFunction(\n            MOI.ScalarAffineTerm.(1.0, [buy[end-1], buy[end]]),\n            0.0,\n        ),\n        MOI.LessThan(6.0),\n    )\n    MOI.optimize!(model)\n    @assert MOI.get(model, MOI.TerminationStatus()) == MOI.INFEASIBLE\n    return\nend\n\nif length(ARGS) > 0\n    bridge = get(ARGS, 2, \"\") != \"--no-bridge\"\n    println(\"Running: $(ARGS[1]) $(get(ARGS, 2, \"\"))\")\n    @time example_diet(GLPK.Optimizer, bridge)\n    @time example_diet(GLPK.Optimizer, bridge)\n    exit(0)\nend","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"You can create a flame-graph via","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"using SnoopComile\ntinf = @snoopi_deep example_diet(GLPK.Optimizer, true)\nusing ProfileView\nProfileView.view(flamegraph(tinf))","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"Here's how things looked in mid-August 2021: (Image: flamegraph)","category":"page"},{"location":"moi/tutorials/latency/","page":"Latency","title":"Latency","text":"There are a few opportunities for improvement (non-red flames, particularly on the right). But the main problem is a large red (non-precompilable due to method ownership) flame.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"EditURL = \"https://github.com/jump-dev/SCS.jl/blob/v1.3.1/README.md\"","category":"page"},{"location":"packages/SCS/#SCS.jl","page":"jump-dev/SCS.jl","title":"SCS.jl","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl is a wrapper for the SCS splitting cone solver.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS can solve linear programs, second-order cone programs, semidefinite programs, exponential cone programs, and power cone programs.","category":"page"},{"location":"packages/SCS/#Affiliation","page":"jump-dev/SCS.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This wrapper is maintained by the JuMP community and is not a project of the SCS developers.","category":"page"},{"location":"packages/SCS/#License","page":"jump-dev/SCS.jl","title":"License","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl is licensed under the MIT License.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The underlying solver, cvxgrp/scs, is licensed under the MIT License.","category":"page"},{"location":"packages/SCS/#Installation","page":"jump-dev/SCS.jl","title":"Installation","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"Install SCS.jl using the Julia package manager:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"import Pkg\nPkg.add(\"SCS\")","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"In addition to installing the SCS.jl package, this will also download and install the SCS binaries. (You do not need to install SCS separately.)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To use a custom binary, read the Custom solver binaries section of the JuMP documentation.","category":"page"},{"location":"packages/SCS/#Use-with-Convex.jl","page":"jump-dev/SCS.jl","title":"Use with Convex.jl","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This example shows how we can model a simple knapsack problem with Convex and use SCS to solve it.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"using Convex, SCS\nitems  = [:Gold, :Silver, :Bronze]\nvalues = [5.0, 3.0, 1.0]\nweights = [2.0, 1.5, 0.3]\n\n# Define a variable of size 3, each index representing an item\nx = Variable(3)\np = maximize(x' * values, 0 <= x, x <= 1, x' * weights <= 3)\nsolve!(p, SCS.Optimizer)\nprintln([items x.value])\n# [:Gold 0.9999971880377178\n#  :Silver 0.46667637765641057\n#  :Bronze 0.9999998036351865]","category":"page"},{"location":"packages/SCS/#Use-with-JuMP","page":"jump-dev/SCS.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This example shows how we can model a simple knapsack problem with JuMP and use SCS to solve it.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"using JuMP, SCS\nitems  = [:Gold, :Silver, :Bronze]\nvalues = Dict(:Gold => 5.0,  :Silver => 3.0,  :Bronze => 1.0)\nweight = Dict(:Gold => 2.0,  :Silver => 1.5,  :Bronze => 0.3)\n\nmodel = Model(SCS.Optimizer)\n@variable(model, 0 <= take[items] <= 1)  # Define a variable for each item\n@objective(model, Max, sum(values[item] * take[item] for item in items))\n@constraint(model, sum(weight[item] * take[item] for item in items) <= 3)\noptimize!(model)\nprintln(value.(take))\n# 1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n#     Dimension 1, Symbol[:Gold, :Silver, :Bronze]\n# And data, a 3-element Array{Float64,1}:\n#  1.0000002002226671\n#  0.4666659513182934\n#  1.0000007732744878","category":"page"},{"location":"packages/SCS/#MathOptInterface-API","page":"jump-dev/SCS.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The SCS optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.Reals","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.VectorAffineFunction{Float64} in MOI.DualExponentialCone\nMOI.VectorAffineFunction{Float64} in MOI.DualPowerCone{Float64}\nMOI.VectorAffineFunction{Float64} in MOI.ExponentialCone\nMOI.VectorAffineFunction{Float64} in MOI.Nonnegatives\nMOI.VectorAffineFunction{Float64} in MOI.PowerCone{Float64}\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorAffineFunction{Float64} in MOI.Zeros\nMOI.VectorAffineFunction{Float64} in SCS.ScaledPSDCone","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/SCS/#Options","page":"jump-dev/SCS.jl","title":"Options","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"All SCS solver options can be set through Convex.jl or MathOptInterface.jl.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"For example:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"model = Model(optimizer_with_attributes(SCS.Optimizer, \"max_iters\" => 10))\n\n# via MathOptInterface:\noptimizer = SCS.Optimizer()\nMOI.set(optimizer, MOI.RawOptimizerAttribute(\"max_iters\"), 10)\nMOI.set(optimizer, MOI.RawOptimizerAttribute(\"verbose\"), 0)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"Common options are:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"max_iters: the maximum number of iterations to take\nverbose: turn printing on (1) or off (0)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"See the glbopts.h header for other options.","category":"page"},{"location":"packages/SCS/#Linear-solvers","page":"jump-dev/SCS.jl","title":"Linear solvers","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS uses a linear solver internally, see this section of SCS documentation. SCS.jl ships with","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.DirectSolver (sparse direct, the default) and\nSCS.LinearSolver (sparse indirect, by conjugate gradient)","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"enabled.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The find currently available linear solvers one can inspect SCS.available_solvers:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> using SCS\n\njulia> SCS.available_solvers\n2-element Vector{DataType}:\n SCS.DirectSolver\n SCS.IndirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To select the linear solver of choice:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"pass the linear_solver option to optimizer_with_attributes, or to MOI.OptimizerWithAttributes;\nspecify the solver as the first argument when using scs_solve directly (see the low-level wrapper section below).","category":"page"},{"location":"packages/SCS/#SCS-with-MKL-Pardiso-linear-solver","page":"jump-dev/SCS.jl","title":"SCS with MKL Pardiso linear solver","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To enable the MKL Pardiso (direct sparse) solver one needs to load MKL_jll before SCS:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> import Pkg\n\njulia> Pkg.add(Pkg.PackageSpec(name = \"MKL_jll\", version = \"2022.2\"))\n\njulia> using MKL_jll    # This must be called before `using SCS`.\n\njulia> using SCS\n\njulia> SCS.available_solvers\n3-element Vector{DataType}:\n SCS.DirectSolver\n SCS.IndirectSolver\n SCS.MKLDirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The MKLDirectSolver is available on Linux x86_64 platform only.","category":"page"},{"location":"packages/SCS/#SCS-with-Sparse-GPU-indirect-solver-(CUDA-only)","page":"jump-dev/SCS.jl","title":"SCS with Sparse GPU indirect solver (CUDA only)","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To enable the indirect linear solver on GPU one needs to load CUDA_jll before SCS:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"julia> import Pkg\n\njulia> Pkg.add(Pkg.PackageSpec(name = \"CUDA_jll\", version = \"10.1\"))\n\njulia> using CUDA_jll  # This must be called before `using SCS`.\n\njulia> using SCS\n\njulia> SCS.available_solvers\n3-element Array{DataType,1}:\n SCS.DirectSolver\n SCS.IndirectSolver\n SCS.GpuIndirectSolver","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"The GpuIndirectSolver is available on Linux x86_64 platform only.","category":"page"},{"location":"packages/SCS/#Low-level-wrapper","page":"jump-dev/SCS.jl","title":"Low-level wrapper","text":"","category":"section"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS.jl provides a low-level interface to solve a problem directly, without interfacing through MathOptInterface.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"This is an advanced interface with a risk of incorrect usage. For new users, we recommend that you use the JuMP or Convex interfaces instead.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"SCS solves a problem of the form:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"minimize        1/2 * x' * P * x + c' * x\nsubject to      A * x + s = b\n                s in K","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"where K is a product cone of:","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"zero cone\npositive orthant { x | x ≥ 0 }\nbox cone { (t,x) | t*l ≤ x ≤ t*u}\nsecond-order cone (SOC) { (t,x) | ||x||_2 ≤ t }\nsemi-definite cone (SDC) { X | X is psd }\nexponential cone { (x,y,z) | y e^(x/y) ≤ z, y>0 }\npower cone { (x,y,z) | x^a * y^(1-a) ≥ |z|, x ≥ 0, y ≥ 0 }\ndual power cone { (u,v,w) | (u/a)^a * (v/(1-a))^(1-a) ≥ |w|, u ≥ 0, v ≥ 0 }.","category":"page"},{"location":"packages/SCS/","page":"jump-dev/SCS.jl","title":"jump-dev/SCS.jl","text":"To solve this problem with SCS, call SCS.scs_solve; see the docstring for details.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"EditURL = \"lp_sensitivity.jl\"","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Sensitivity-analysis-of-a-linear-program","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial explains how to use the lp_sensitivity_report function to create sensitivity reports like those that are produced by the Excel Solver. This is most often used in introductory classes to linear programming.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"In brief, sensitivity analysis of a linear program is about asking two questions:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Given an optimal solution, how much can the objective coefficients change by before a different solution becomes optimal?\nGiven an optimal solution, how much can the right-hand side of a linear constraint change by before a different solution becomes optimal?","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"JuMP provides a function, lp_sensitivity_report, to help us compute these values, but this tutorial extends that to create more informative tables in the form of a DataFrame.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Setup","page":"Sensitivity analysis of a linear program","title":"Setup","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"using JuMP\nimport HiGHS\nimport DataFrames","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"as well as this small linear program:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"model = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@variable(model, z <= 1)\n@objective(model, Min, 12x + 20y - z)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\n@constraint(model, c3, x + y <= 20)\noptimize!(model)\nsolution_summary(model; verbose = true)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Can you identify:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"The objective coefficient of each variable?\nThe right-hand side of each constraint?\nThe optimal primal and dual solutions?","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Sensitivity-reports","page":"Sensitivity analysis of a linear program","title":"Sensitivity reports","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Now let's call lp_sensitivity_report:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report = lp_sensitivity_report(model)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"It returns a SensitivityReport object, which maps:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Every variable reference to a tuple (d_lo, d_hi)::Tuple{Float64,Float64}, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.\nEvery constraint reference to a tuple (d_lo, d_hi)::Tuple{Float64,Float64}, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Both tuples are relative, rather than absolute. So, given an objective coefficient of 1.0 and a tuple (-0.5, 0.5), the objective coefficient can range between 1.0 - 0.5 an 1.0 + 0.5.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"For example:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report[x]","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"indicates that the objective coefficient on x, that is, 12, can decrease by -0.333 or increase by 3.0 and the primal solution (15, 1.25) will remain optimal. In addition:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"report[c1]","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"means that the right-hand side of the c1 constraint (100), can decrease by 4 units, or increase by 2.85 units, and the primal solution (15, 1.25) will remain optimal.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Variable-sensitivity","page":"Sensitivity analysis of a linear program","title":"Variable sensitivity","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"By themselves, the tuples aren't informative. Let's put them in context by collating a range of other information about a variable:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"function variable_report(xi)\n    return (\n        name = name(xi),\n        lower_bound = has_lower_bound(xi) ? lower_bound(xi) : -Inf,\n        value = value(xi),\n        upper_bound = has_upper_bound(xi) ? upper_bound(xi) : Inf,\n        reduced_cost = reduced_cost(xi),\n        obj_coefficient = coefficient(objective_function(model), xi),\n        allowed_decrease = report[xi][1],\n        allowed_increase = report[xi][2],\n    )\nend","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Calling our function on x:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"x_report = variable_report(x)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"That's a bit hard to read, so let's call this on every variable in the model and put things into a DataFrame:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"variable_df =\n    DataFrames.DataFrame(variable_report(xi) for xi in all_variables(model))","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Constraint-sensitivity","page":"Sensitivity analysis of a linear program","title":"Constraint sensitivity","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"We can do something similar with constraints:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"function constraint_report(c::ConstraintRef)\n    return (\n        name = name(c),\n        value = value(c),\n        rhs = normalized_rhs(c),\n        slack = normalized_rhs(c) - value(c),\n        shadow_price = shadow_price(c),\n        allowed_decrease = report[c][1],\n        allowed_increase = report[c][2],\n    )\nend\n\nc1_report = constraint_report(c1)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"That's a bit hard to read, so let's call this on every variable in the model and put things into a DataFrame:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"constraint_df = DataFrames.DataFrame(\n    constraint_report(ci) for (F, S) in list_of_constraint_types(model) for\n    ci in all_constraints(model, F, S) if F == AffExpr\n)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/#Analysis-questions","page":"Sensitivity analysis of a linear program","title":"Analysis questions","text":"","category":"section"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"Now we can use these dataframes to ask questions of the solution.","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"For example, we can find basic variables by looking for variables with a reduced cost of 0:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"basic = filter(row -> iszero(row.reduced_cost), variable_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"and non-basic variables by looking for non-zero reduced costs:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"non_basic = filter(row -> !iszero(row.reduced_cost), variable_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"we can also find constraints that are binding by looking for zero slacks:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"binding = filter(row -> iszero(row.slack), constraint_df)","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"or non-zero shadow prices:","category":"page"},{"location":"tutorials/linear/lp_sensitivity/","page":"Sensitivity analysis of a linear program","title":"Sensitivity analysis of a linear program","text":"binding2 = filter(row -> !iszero(row.shadow_price), constraint_df)","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/tutorials/example.md\"","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/tutorials/example/#Solving-a-problem-using-MathOptInterface","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"In this tutorial we demonstrate how to use MathOptInterface to solve the binary-constrained knapsack problem:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"beginaligned\nmax   c^top x       \nst   w^top x le C \n         x_i in 01quad forall i=1ldotsn\nendaligned","category":"page"},{"location":"moi/tutorials/example/#Required-packages","page":"Solving a problem using MathOptInterface","title":"Required packages","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Load the MathOptInterface module and define the shorthand MOI:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"import MathOptInterface as MOI","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"As an optimizer, we choose GLPK:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"using GLPK\noptimizer = GLPK.Optimizer()","category":"page"},{"location":"moi/tutorials/example/#Define-the-data","page":"Solving a problem using MathOptInterface","title":"Define the data","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"We first define the constants of the problem:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> c = [1.0, 2.0, 3.0]\n3-element Vector{Float64}:\n 1.0\n 2.0\n 3.0\n\njulia> w = [0.3, 0.5, 1.0]\n3-element Vector{Float64}:\n 0.3\n 0.5\n 1.0\n\njulia> C = 3.2\n3.2","category":"page"},{"location":"moi/tutorials/example/#Add-the-variables","page":"Solving a problem using MathOptInterface","title":"Add the variables","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> x = MOI.add_variables(optimizer, length(c));","category":"page"},{"location":"moi/tutorials/example/#set_objective_example","page":"Solving a problem using MathOptInterface","title":"Set the objective","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.set(\n           optimizer,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(c, x), 0.0),\n       );\n\njulia> MOI.set(optimizer, MOI.ObjectiveSense(), MOI.MAX_SENSE)","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"tip: Tip\nMOI.ScalarAffineTerm.(c, x) is a shortcut for [MOI.ScalarAffineTerm(c[i], x[i]) for i = 1:3]. This is Julia's broadcast syntax in action, and is used quite often throughout MOI.","category":"page"},{"location":"moi/tutorials/example/#Add-the-constraints","page":"Solving a problem using MathOptInterface","title":"Add the constraints","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"We add the knapsack constraint and integrality constraints:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.add_constraint(\n           optimizer,\n           MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(w, x), 0.0),\n           MOI.LessThan(C),\n       );","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Add integrality constraints:","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> for x_i in x\n           MOI.add_constraint(optimizer, x_i, MOI.ZeroOne())\n       end","category":"page"},{"location":"moi/tutorials/example/#Optimize-the-model","page":"Solving a problem using MathOptInterface","title":"Optimize the model","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.optimize!(optimizer)","category":"page"},{"location":"moi/tutorials/example/#Understand-why-the-solver-stopped","page":"Solving a problem using MathOptInterface","title":"Understand why the solver stopped","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"The first thing to check after optimization is why the solver stopped, for example, did it stop because of a time limit or did it stop because it found the optimal solution?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.TerminationStatus())\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"Looks like we found an optimal solution.","category":"page"},{"location":"moi/tutorials/example/#Understand-what-solution-was-returned","page":"Solving a problem using MathOptInterface","title":"Understand what solution was returned","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.ResultCount())\n1\n\njulia> MOI.get(optimizer, MOI.PrimalStatus())\nFEASIBLE_POINT::ResultStatusCode = 1\n\njulia> MOI.get(optimizer, MOI.DualStatus())\nNO_SOLUTION::ResultStatusCode = 0","category":"page"},{"location":"moi/tutorials/example/#Query-the-objective","page":"Solving a problem using MathOptInterface","title":"Query the objective","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"What is its objective value?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.ObjectiveValue())\n6.0","category":"page"},{"location":"moi/tutorials/example/#Query-the-primal-solution","page":"Solving a problem using MathOptInterface","title":"Query the primal solution","text":"","category":"section"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"And what is the value of the variables x?","category":"page"},{"location":"moi/tutorials/example/","page":"Solving a problem using MathOptInterface","title":"Solving a problem using MathOptInterface","text":"julia> MOI.get(optimizer, MOI.VariablePrimal(), x)\n3-element Vector{Float64}:\n 1.0\n 1.0\n 1.0","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"EditURL = \"https://github.com/jump-dev/DiffOpt.jl/blob/v0.4.2/README.md\"","category":"page"},{"location":"packages/DiffOpt/#DiffOpt.jl","page":"jump-dev/DiffOpt.jl","title":"DiffOpt.jl","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"(Image: stable docs) (Image: development docs) (Image: Build Status) (Image: Coverage)","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt.jl is a package for differentiating convex optimization programs with respect to the program parameters. DiffOpt currently supports linear, quadratic, and conic programs.","category":"page"},{"location":"packages/DiffOpt/#License","page":"jump-dev/DiffOpt.jl","title":"License","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt.jl is licensed under the MIT License.","category":"page"},{"location":"packages/DiffOpt/#Installation","page":"jump-dev/DiffOpt.jl","title":"Installation","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"Install DiffOpt using Pkg.add:","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"import Pkg\nPkg.add(\"DiffOpt\")","category":"page"},{"location":"packages/DiffOpt/#Documentation","page":"jump-dev/DiffOpt.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"The documentation for DiffOpt.jl includes a detailed description of the theory behind the package, along with examples, tutorials, and an API reference.","category":"page"},{"location":"packages/DiffOpt/#Use-with-JuMP","page":"jump-dev/DiffOpt.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"Use DiffOpt with JuMP by following this brief example:","category":"page"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"using JuMP, DiffOpt, HiGHS\n# Create a model using the wrapper\nmodel = Model(() -> DiffOpt.diff_optimizer(HiGHS.Optimizer))\n# Define your model and solve it\n@variable(model, x)\n@constraint(model, cons, x >= 3)\n@objective(model, Min, 2x)\noptimize!(model)\n# Choose the problem parameters to differentiate with respect to, and set their\n# perturbations.\nMOI.set(model, DiffOpt.ReverseVariablePrimal(), x, 1.0)\n# Differentiate the model\nDiffOpt.reverse_differentiate!(model)\n# fetch the gradients\ngrad_exp = MOI.get(model, DiffOpt.ReverseConstraintFunction(), cons)  # -3 x - 1\nconstant(grad_exp)        # -1\ncoefficient(grad_exp, x)  # -3","category":"page"},{"location":"packages/DiffOpt/#GSOC2020","page":"jump-dev/DiffOpt.jl","title":"GSOC2020","text":"","category":"section"},{"location":"packages/DiffOpt/","page":"jump-dev/DiffOpt.jl","title":"jump-dev/DiffOpt.jl","text":"DiffOpt began as a NumFOCUS sponsored Google Summer of Code (2020) project","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"EditURL = \"https://github.com/jump-dev/KNITRO.jl/blob/4c56de7684c42dd3c83c5fbc515ae1f424eed524/README.md\"","category":"page"},{"location":"packages/KNITRO/#KNITRO.jl","page":"jump-dev/KNITRO.jl","title":"KNITRO.jl","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl is a wrapper for the Artelys Knitro solver.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"It has two components:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"a thin wrapper around the C API\nan interface to MathOptInterface.","category":"page"},{"location":"packages/KNITRO/#Affiliation","page":"jump-dev/KNITRO.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"This wrapper is maintained by the JuMP community with help from Artelys.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Contact Artelys support if you encounter any problem with this interface or the solver.","category":"page"},{"location":"packages/KNITRO/#License","page":"jump-dev/KNITRO.jl","title":"License","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl is licensed under the MIT License.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/KNITRO/#Installation","page":"jump-dev/KNITRO.jl","title":"Installation","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"First, obtain a license and install a copy of KNITRO from Artelys.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Then, install KNITRO.jl using the Julia package manager:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"import Pkg\nPkg.add(\"KNITRO\")","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"If you are having trouble installing KNITRO.jl, here are several things to try:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Make sure that you have defined your global variables correctly, for example with export KNITRODIR=\"/path/to/knitro-vXXX-$OS-64\" and export LD_LIBRARY_PATH=\"$LD_LIBRARY_PATH:$KNITRODIR/lib\". You can check that KNITRO.jl sees your library with using KNITRO; KNITRO.has_knitro().\nIf KNITRO.has_knitro() returns false but you are confident that your paths are correct, try running Pkg.build(\"KNITRO\") and restarting Julia. In at least one user's experience, installing and using KNITRO in a temporary Julia environment (activated with ] activate --temp) does not work and the need to manually build is likely the reason why.","category":"page"},{"location":"packages/KNITRO/#Use-with-JuMP","page":"jump-dev/KNITRO.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"To use KNITRO with JuMP, use KNITRO.Optimizer:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"using JuMP, KNITRO\nmodel = Model(KNITRO.Optimizer)\nset_attribute(model, \"outlev\", 1)\nset_attribute(model, \"algorithm\", 4)","category":"page"},{"location":"packages/KNITRO/#Use-with-AMPL","page":"jump-dev/KNITRO.jl","title":"Use with AMPL","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"To use KNITRO with AmplNLWriter.jl, use KNITRO.amplexe:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"using JuMP\nimport AmplNLWriter\nimport KNITRO\nmodel = Model(() -> AmplNLWriter.Optimizer(KNITRO.amplexe, [\"outlev=3\"]))","category":"page"},{"location":"packages/KNITRO/#Use-with-other-packages","page":"jump-dev/KNITRO.jl","title":"Use with other packages","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"A variety of packages extend KNITRO.jl to support other optimization modeling systems. These include:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"NLPModelsKnitro\nOptimization.jl","category":"page"},{"location":"packages/KNITRO/#MathOptInterface-API","page":"jump-dev/KNITRO.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"The Knitro optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}\nMOI.ObjectiveFunction{MOI.ScalarQuadraticFunction{Float64}}\nMOI.ObjectiveFunction{MOI.VariableIndex}","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.Reals","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarAffineFunction{Float64} in MOI.LessThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.EqualTo{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.GreaterThan{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.Interval{Float64}\nMOI.ScalarQuadraticFunction{Float64} in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.EqualTo{Float64}\nMOI.VariableIndex in MOI.GreaterThan{Float64}\nMOI.VariableIndex in MOI.Integer\nMOI.VariableIndex in MOI.Interval{Float64}\nMOI.VariableIndex in MOI.LessThan{Float64}\nMOI.VariableIndex in MOI.ZeroOne\nMOI.VectorAffineFunction{Float64} in MOI.SecondOrderCone\nMOI.VectorOfVariables in MOI.Complements\nMOI.VectorOfVariables in MOI.SecondOrderCone","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"MOI.NLPBlock()\nMOI.NLPBlockDualStart()\nMOI.ObjectiveSense()","category":"page"},{"location":"packages/KNITRO/#Options","page":"jump-dev/KNITRO.jl","title":"Options","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"A list of available options is provided in the KNITRO reference manual.","category":"page"},{"location":"packages/KNITRO/#Low-level-wrapper","page":"jump-dev/KNITRO.jl","title":"Low-level wrapper","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"KNITRO.jl implements most of Knitro's functionalities. If you aim at using part of Knitro's API that are not implemented in the MathOptInterface/JuMP ecosystem, you can refer to the low-level API, which wraps Knitro's C API (whose templates are specified in the file knitro.h).","category":"page"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Extensive examples using the C wrapper can be found in examples/.","category":"page"},{"location":"packages/KNITRO/#Multi-threading","page":"jump-dev/KNITRO.jl","title":"Multi-threading","text":"","category":"section"},{"location":"packages/KNITRO/","page":"jump-dev/KNITRO.jl","title":"jump-dev/KNITRO.jl","text":"Due to limitations in the interaction between Julia and C, KNITRO.jl disables multi-threading if the problem is nonlinear. This will override any options such as par_numthreads that you may have set. Read GitHub issue #93 for more details.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"EditURL = \"https://github.com/jump-dev/MathOptInterface.jl/blob/v1.20.1/docs/src/manual/modification.md\"","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"CurrentModule = MathOptInterface\nDocTestSetup = quote\n    import MathOptInterface as MOI\nend\nDocTestFilters = [r\"MathOptInterface|MOI\"]","category":"page"},{"location":"moi/manual/modification/#Problem-modification","page":"Problem modification","title":"Problem modification","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"In addition to adding and deleting constraints and variables, MathOptInterface supports modifying, in-place, coefficients in the constraints and the objective function of a model.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"These modifications can be grouped into two categories:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"modifications which replace the set of function of a constraint with a new set or function\nmodifications which change, in-place, a component of a function","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"warning: Warning\nSome ModelLike objects do not support problem modification.","category":"page"},{"location":"moi/manual/modification/#Modify-the-set-of-a-constraint","page":"Problem modification","title":"Modify the set of a constraint","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use set and ConstraintSet to modify the set of a constraint by replacing it with a new instance of the same type.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.set(model, MOI.ConstraintSet(), c, MOI.EqualTo(2.0));\n\njulia> MOI.get(model, MOI.ConstraintSet(), c) == MOI.EqualTo(2.0)\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"However, the following will fail as the new set is of a different type to the original set:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(model, MOI.ConstraintSet(), c, MOI.GreaterThan(2.0))\nERROR: [...]","category":"page"},{"location":"moi/manual/modification/#Special-cases:-set-transforms","page":"Problem modification","title":"Special cases: set transforms","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"If our constraint is an affine inequality, then this corresponds to modifying the right-hand side of a constraint in linear programming.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"In some special cases, solvers may support efficiently changing the set of a constraint (for example, from LessThan to GreaterThan). For these cases, MathOptInterface provides the transform method.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"The transform function returns a new constraint index, and the old constraint index (that is, c) is no longer valid.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.LessThan(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}(1)\n\njulia> new_c = MOI.transform(model, c, MOI.GreaterThan(2.0));\n\njulia> MOI.is_valid(model, c)\nfalse\n\njulia> MOI.is_valid(model, new_c)\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"note: Note\ntransform cannot be called with a set of the same type. Use set instead.","category":"page"},{"location":"moi/manual/modification/#Modify-the-function-of-a-constraint","page":"Problem modification","title":"Modify the function of a constraint","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use set and ConstraintFunction to modify the function of a constraint by replacing it with a new instance of the same type.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(2.0, x)], 1.0);\n\njulia> MOI.set(model, MOI.ConstraintFunction(), c, new_f);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"However, the following will fail as the new function is of a different type to the original function:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(model, MOI.ConstraintFunction(), c, x)\nERROR: [...]","category":"page"},{"location":"moi/manual/modification/#Modify-constant-term-in-a-scalar-function","page":"Problem modification","title":"Modify constant term in a scalar function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and ScalarConstantChange to modify the constant term in a ScalarAffineFunction or ScalarQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.modify(model, c, MOI.ScalarConstantChange(1.0));\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 1.0);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"ScalarConstantChange can also be used to modify the objective function by passing an instance of ObjectiveFunction:","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> MOI.set(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           new_f,\n       );\n\njulia> MOI.modify(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n           MOI.ScalarConstantChange(-1.0)\n       );\n\njulia> MOI.get(\n           model,\n           MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),\n       ) ≈ MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], -1.0)\ntrue","category":"page"},{"location":"moi/manual/modification/#Modify-constant-terms-in-a-vector-function","page":"Problem modification","title":"Modify constant terms in a vector function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and VectorConstantChange to modify the constant vector in a VectorAffineFunction or VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.VectorAffineFunction([\n                   MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n                   MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n               ],\n               [0.0, 0.0],\n           ),\n           MOI.Nonnegatives(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.modify(model, c, MOI.VectorConstantChange([3.0, 4.0]));\n\njulia> new_f = MOI.VectorAffineFunction(\n           [\n        MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n        MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n           ],\n           [3.0, 4.0],\n       );\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/#Modify-affine-coefficients-in-a-scalar-function","page":"Problem modification","title":"Modify affine coefficients in a scalar function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and ScalarCoefficientChange to modify the affine coefficient of a ScalarAffineFunction or ScalarQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(1.0, x)], 0.0),\n           MOI.EqualTo(1.0),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.EqualTo{Float64}}(1)\n\njulia> MOI.modify(model, c, MOI.ScalarCoefficientChange(x, 2.0));\n\njulia> new_f = MOI.ScalarAffineFunction([MOI.ScalarAffineTerm(2.0, x)], 0.0);\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"ScalarCoefficientChange can also be used to modify the objective function by passing an instance of ObjectiveFunction.","category":"page"},{"location":"moi/manual/modification/#Modify-affine-coefficients-in-a-vector-function","page":"Problem modification","title":"Modify affine coefficients in a vector function","text":"","category":"section"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"Use modify and MultirowChange to modify a vector of affine coefficients in a VectorAffineFunction or a VectorQuadraticFunction.","category":"page"},{"location":"moi/manual/modification/","page":"Problem modification","title":"Problem modification","text":"julia> c = MOI.add_constraint(\n           model,\n           MOI.VectorAffineFunction([\n                   MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(1.0, x)),\n                   MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(2.0, x)),\n               ],\n               [0.0, 0.0],\n           ),\n           MOI.Nonnegatives(2),\n       )\nMathOptInterface.ConstraintIndex{MathOptInterface.VectorAffineFunction{Float64}, MathOptInterface.Nonnegatives}(1)\n\njulia> MOI.modify(model, c, MOI.MultirowChange(x, [(1, 3.0), (2, 4.0)]));\n\njulia> new_f = MOI.VectorAffineFunction(\n           [\n        MOI.VectorAffineTerm(1, MOI.ScalarAffineTerm(3.0, x)),\n        MOI.VectorAffineTerm(2, MOI.ScalarAffineTerm(4.0, x)),\n           ],\n           [0.0, 0.0],\n       );\n\njulia> MOI.get(model, MOI.ConstraintFunction(), c) ≈ new_f\ntrue","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"EditURL = \"facility_location.jl\"","category":"page"},{"location":"tutorials/linear/facility_location/#The-facility-location-problem","page":"The facility location problem","title":"The facility location problem","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial was originally contributed by Mathieu Tanneau and Alexis Montoison.","category":"page"},{"location":"tutorials/linear/facility_location/#Required-packages","page":"The facility location problem","title":"Required packages","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"This tutorial requires the following packages:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"using JuMP\nimport HiGHS\nimport LinearAlgebra\nimport Plots\nimport Random","category":"page"},{"location":"tutorials/linear/facility_location/#Uncapacitated-facility-location","page":"The facility location problem","title":"Uncapacitated facility location","text":"","category":"section"},{"location":"tutorials/linear/facility_location/#Problem-description","page":"The facility location problem","title":"Problem description","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"We are given","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"A set M=1 dots m of clients\nA set N= 1 dots n of sites where a facility can be built","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Decision variables Decision variables are split into two categories:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Binary variable y_j indicates whether facility j is built or not\nBinary variable x_i j indicates whether client i is assigned to facility j","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Objective The objective is to minimize the total cost of serving all clients. This costs breaks down into two components:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Fixed cost of building a facility.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"In this example, this cost is f_j = 1  forall j.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Cost of serving clients from the assigned facility.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"In this example, the cost c_i j of serving client i from facility j is the Euclidean distance between the two.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Constraints","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Each customer must be served by exactly one facility\nA facility cannot serve any client unless it is open","category":"page"},{"location":"tutorials/linear/facility_location/#MILP-formulation","page":"The facility location problem","title":"MILP formulation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The problem can be formulated as the following MILP:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nmin_x y    \nsum_i j c_i j x_i j +\nsum_j f_j y_j \nst \nsum_j x_i j = 1  forall i in M \n x_i j leq y_j  forall i in M j in N \n x_i j y_j in 0 1  forall i in M j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"where the first set of constraints ensures that each client is served exactly once, and the second set of constraints ensures that no client is served from an unopened facility.","category":"page"},{"location":"tutorials/linear/facility_location/#Problem-data","page":"The facility location problem","title":"Problem data","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"To ensure reproducibility, we set the random number seed:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Random.seed!(314)","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Here's the data we need:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"# Number of clients\nm = 12\n# Number of facility locations\nn = 5\n\n# Clients' locations\nx_c, y_c = rand(m), rand(m)\n\n# Facilities' potential locations\nx_f, y_f = rand(n), rand(n)\n\n# Fixed costs\nf = ones(n);\n\n# Distance\nc = zeros(m, n)\nfor i in 1:m\n    for j in 1:n\n        c[i, j] = LinearAlgebra.norm([x_c[i] - x_f[j], y_c[i] - y_f[j]], 2)\n    end\nend","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the data","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Plots.scatter(\n    x_c,\n    y_c;\n    label = \"Clients\",\n    markershape = :circle,\n    markercolor = :blue,\n)\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = \"Facility\",\n    markershape = :square,\n    markercolor = :white,\n    markersize = 6,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)","category":"page"},{"location":"tutorials/linear/facility_location/#JuMP-implementation","page":"The facility location problem","title":"JuMP implementation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Create a JuMP model","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, y[1:n], Bin);\n@variable(model, x[1:m, 1:n], Bin);\n# Each client is served exactly once\n@constraint(model, client_service[i in 1:m], sum(x[i, j] for j in 1:n) == 1);\n# A facility must be open to serve a client\n@constraint(model, open_facility[i in 1:m, j in 1:n], x[i, j] <= y[j]);\n@objective(model, Min, f' * y + sum(c .* x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Solve the uncapacitated facility location problem with HiGHS","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"optimize!(model)\nprintln(\"Optimal value: \", objective_value(model))","category":"page"},{"location":"tutorials/linear/facility_location/#Visualizing-the-solution","page":"The facility location problem","title":"Visualizing the solution","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The threshold 1e-5 ensure that edges between clients and facilities are drawn when x[i, j] ≈ 1.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"x_is_selected = isapprox.(value.(x), 1; atol = 1e-5);\ny_is_selected = isapprox.(value.(y), 1; atol = 1e-5);\n\np = Plots.scatter(\n    x_c,\n    y_c;\n    markershape = :circle,\n    markercolor = :blue,\n    label = nothing,\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    markershape = :square,\n    markercolor = [(y_is_selected[j] ? :red : :white) for j in 1:n],\n    markersize = 6,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n    label = nothing,\n)\n\nfor i in 1:m, j in 1:n\n    if x_is_selected[i, j]\n        Plots.plot!(\n            [x_c[i], x_f[j]],\n            [y_c[i], y_f[j]];\n            color = :black,\n            label = nothing,\n        )\n    end\nend\n\np","category":"page"},{"location":"tutorials/linear/facility_location/#Capacitated-facility-location","page":"The facility location problem","title":"Capacitated facility location","text":"","category":"section"},{"location":"tutorials/linear/facility_location/#Problem-formulation","page":"The facility location problem","title":"Problem formulation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The capacitated variant introduces a capacity constraint on each facility, that is, clients have a certain level of demand to be served, while each facility only has finite capacity which cannot be exceeded.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Specifically,","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The demand of client i is denoted by a_i geq 0\nThe capacity of facility j is denoted by q_j geq 0","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The capacity constraints then write","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nsum_i a_i x_i j leq q_j y_j  forall j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Note that, if y_j is set to 0, the capacity constraint above automatically forces x_i j to 0.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Thus, the capacitated facility location can be formulated as follows","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"beginaligned\nmin_x y    \nsum_i j c_i j x_i j +\nsum_j f_j y_j \nst \nsum_j x_i j = 1  forall i in M \n sum_i a_i x_i j leq q_j y_j  forall j in N \n x_i j y_j in 0 1  forall i in M j in N\nendaligned","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"For simplicity, we will assume that there is enough capacity to serve the demand, that is, there exists at least one feasible solution.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"We need some new data:","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"# Demands\na = rand(1:3, m);\n\n# Capacities\nq = rand(5:10, n);\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the data","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Plots.scatter(\n    x_c,\n    y_c;\n    label = nothing,\n    markershape = :circle,\n    markercolor = :blue,\n    markersize = 2 .* (2 .+ a),\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = nothing,\n    markershape = :rect,\n    markercolor = :white,\n    markersize = q,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)","category":"page"},{"location":"tutorials/linear/facility_location/#JuMP-implementation-2","page":"The facility location problem","title":"JuMP implementation","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Create a JuMP model","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"model = Model(HiGHS.Optimizer)\nset_silent(model)\n@variable(model, y[1:n], Bin);\n@variable(model, x[1:m, 1:n], Bin);\n# Each client is served exactly once\n@constraint(model, client_service[i in 1:m], sum(x[i, :]) == 1);\n# Capacity constraint\n@constraint(model, capacity, x' * a .<= (q .* y));\n# Objective\n@objective(model, Min, f' * y + sum(c .* x));\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Solve the problem","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"optimize!(model)\nprintln(\"Optimal value: \", objective_value(model))","category":"page"},{"location":"tutorials/linear/facility_location/#Visualizing-the-solution-2","page":"The facility location problem","title":"Visualizing the solution","text":"","category":"section"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"The threshold 1e-5 ensure that edges between clients and facilities are drawn when x[i, j] ≈ 1.","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"x_is_selected = isapprox.(value.(x), 1; atol = 1e-5);\ny_is_selected = isapprox.(value.(y), 1; atol = 1e-5);\nnothing #hide","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"Display the solution","category":"page"},{"location":"tutorials/linear/facility_location/","page":"The facility location problem","title":"The facility location problem","text":"p = Plots.scatter(\n    x_c,\n    y_c;\n    label = nothing,\n    markershape = :circle,\n    markercolor = :blue,\n    markersize = 2 .* (2 .+ a),\n)\n\nPlots.scatter!(\n    x_f,\n    y_f;\n    label = nothing,\n    markershape = :rect,\n    markercolor = [(y_is_selected[j] ? :red : :white) for j in 1:n],\n    markersize = q,\n    markerstrokecolor = :red,\n    markerstrokewidth = 2,\n)\n\nfor i in 1:m, j in 1:n\n    if x_is_selected[i, j]\n        Plots.plot!(\n            [x_c[i], x_f[j]],\n            [y_c[i], y_f[j]];\n            color = :black,\n            label = nothing,\n        )\n    end\nend\n\np","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"EditURL = \"https://github.com/jump-dev/Xpress.jl/blob/v0.16.1/README.md\"","category":"page"},{"location":"packages/Xpress/#Xpress.jl","page":"jump-dev/Xpress.jl","title":"Xpress.jl","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Xpress.jl is a wrapper for the FICO Xpress Solver.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"It has two components:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"a thin wrapper around the complete C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/Xpress/#Affiliation","page":"jump-dev/Xpress.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The Xpress wrapper for Julia is community driven and not officially supported by FICO Xpress. If you are a commercial customer interested in official support for Julia from FICO Xpress, let them know.","category":"page"},{"location":"packages/Xpress/#License","page":"jump-dev/Xpress.jl","title":"License","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Xpress.jl is licensed under the MIT License.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The underlying solver is a closed-source commercial product for which you must purchase a license.","category":"page"},{"location":"packages/Xpress/#Installation","page":"jump-dev/Xpress.jl","title":"Installation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"First, obtain a license of Xpress and install Xpress solver, following the instructions on the FICO website.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Then, install this package using:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"import Pkg\nPkg.add(\"Xpress\")","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"If you encounter an error, make sure that the XPRESSDIR environmental variable is set to the path of the Xpress directory. This should be part of a standard installation. The Xpress library will be searched for in XPRESSDIR/lib on Unix platforms and XPRESSDIR/bin on Windows.","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For example, on macOS, you may need:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"ENV[\"XPRESSDIR\"] = \"/Applications/FICO Xpress/xpressmp/\"\nimport Pkg\nPkg.add(\"Xpress\")","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"By default, building Xpress.jl will fail if the Xpress library is not found. This may not be desirable in certain cases, for example when part of a package's test suite uses Xpress as an optional test dependency, but Xpress cannot be installed on a CI server running the test suite. To support this use case, the XPRESS_JL_SKIP_LIB_CHECK environment variable may be set (to any value) to make Xpress.jl installable (but not usable).","category":"page"},{"location":"packages/Xpress/#Use-with-JuMP","page":"jump-dev/Xpress.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"To use Xpress with JuMP, use:","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"using JuMP, Xpress\nmodel = Model(Xpress.Optimizer)\nset_optimizer(model, \"PRESOLVE\", 0)","category":"page"},{"location":"packages/Xpress/#Options","page":"jump-dev/Xpress.jl","title":"Options","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For other parameters use Xpress Optimizer manual or type julia -e \"using Xpress; println(keys(Xpress.XPRS_ATTRIBUTES))\".","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"If logfile is set to \"\", the log file is disabled and output is printed to the console (there might be issues with console output on windows (it is manually implemented with callbacks)). If logfile is set to a file's path, output is printed to that file. By default, logfile = \"\" (console).","category":"page"},{"location":"packages/Xpress/#Callbacks","page":"jump-dev/Xpress.jl","title":"Callbacks","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Solver specific and solver independent callbacks are working in MathOptInterface and, consequently, in JuMP. However, the current implementation should be considered experimental.","category":"page"},{"location":"packages/Xpress/#Environment-variables","page":"jump-dev/Xpress.jl","title":"Environment variables","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"XPRESS_JL_SKIP_LIB_CHECK: Used to skip build lib check as previously described.\nXPRESS_JL_NO_INFO: Disable license info log.\nXPRESS_JL_NO_DEPS_ERROR: Disable error when do deps.jl file is found.\nXPRESS_JL_NO_AUTO_INIT: Disable automatic run of Xpress.initialize().","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"Specially useful for explicitly loading the dynamic library.","category":"page"},{"location":"packages/Xpress/#Skipping-Xpress.postsolve","page":"jump-dev/Xpress.jl","title":"Skipping Xpress.postsolve","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"In older versions of Xpress, the command XPRSpostsolve throws an error in infeasible models. In these older versions the post solve should not be executed. To do this, one can use the MOI.RawOptimizerAttribute(\"MOI_POST_SOLVE\") to skip this routine.","category":"page"},{"location":"packages/Xpress/#C-API","page":"jump-dev/Xpress.jl","title":"C API","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"The C API can be accessed via Xpress.Lib.XPRSxx functions, where the names and arguments are identical to the C API. ","category":"page"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"See the Xpress documentation for details.","category":"page"},{"location":"packages/Xpress/#Documentation","page":"jump-dev/Xpress.jl","title":"Documentation","text":"","category":"section"},{"location":"packages/Xpress/","page":"jump-dev/Xpress.jl","title":"jump-dev/Xpress.jl","text":"For more information, consult the FICO optimizer manual.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"EditURL = \"mip_duality.jl\"","category":"page"},{"location":"tutorials/linear/mip_duality/#Computing-the-duals-of-a-mixed-integer-program","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial explains how to compute the duals of a mixed-integer linear program by fixing the discrete variables to their optimal solution and resolving as a linear program.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"using JuMP\nimport HiGHS","category":"page"},{"location":"tutorials/linear/mip_duality/#The-model","page":"Computing the duals of a mixed-integer program","title":"The model","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Our example model is the unit commitment example from Unit commitment. The details are unimportant, other than to note that there are two types of continuous variables, g and w, representing the quantity of generation from thermal and wind plants, and a discrete variable dispatch, which is 1 if plant i is operating, and 0 if not.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"We are interested in the \"dual\" of the power_balance constraint, because it represents the marginal price of electricity that consumers should pay for their consumption.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"generators = [\n    (min = 0.0, max = 1000.0, fixed_cost = 1000.0, variable_cost = 50.0),\n    (min = 300.0, max = 1000.0, fixed_cost = 0.0, variable_cost = 100.0),\n]\nN = length(generators)\nmodel = Model(HiGHS.Optimizer)\nset_silent(model)\n@variables(model, begin\n    generators[i].min <= g[i = 1:N] <= generators[i].max\n    0 <= w <= 200\n    dispatch[i = 1:N], Bin\nend)\n@constraints(model, begin\n    power_balance, sum(g[i] for i in 1:N) + w == 1500\n    [i = 1:N], g[i] <= generators[i].max * dispatch[i]\n    [i = 1:N], g[i] >= generators[i].min * dispatch[i]\nend)\n@objective(\n    model,\n    Min,\n    sum(\n        generators[i].fixed_cost * dispatch[i] +\n        generators[i].variable_cost * g[i] for i in 1:N\n    )\n)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/#Manually-fix-the-variables","page":"Computing the duals of a mixed-integer program","title":"Manually fix the variables","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"If we optimize this model, we obtain a dual_status of NO_SOLUTION:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"This is because HiGHS cannot compute the duals of a mixed-integer program. We can work around this problem by fixing the integer variables to their optimal solution, relaxing integrality, and re-solving as a linear program.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"discrete_values = value.(dispatch)\nfix.(dispatch, discrete_values; force = true)\nunset_binary.(dispatch)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Now if we re-solve the problem, we obtain a FEASIBLE_POINT for the dual:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"and a marginal price of electricity of $100/MWh:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"dual(power_balance)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"To reset the problem back to a mixed-integer linear program, we need to unfix and call set_binary:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"unfix.(dispatch)\nset_binary.(dispatch)\nprint(model)","category":"page"},{"location":"tutorials/linear/mip_duality/#Use-fix_discrete_variables","page":"Computing the duals of a mixed-integer program","title":"Use fix_discrete_variables","text":"","category":"section"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Manually choosing the variables to relax and fix works for our small example, but it becomes more difficult in problems with a larger number of binary and integer variables. To automate the process we just did manually, JuMP provides the fix_discrete_variables function:","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"undo = fix_discrete_variables(model);\nnothing #hide","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Here undo is a function that, when called with no arguments, returns the model to the original mixed-integer formulation.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"tip: Tip\nAfter calling fix_discrete_variables, you can set a new solver with set_optimizer if your mixed-integer solver does not support computing a dual solution.","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"print(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"optimize!(model)\ndual_status(model)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"dual(power_balance)","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"Finally, call undo to revert the reformulation","category":"page"},{"location":"tutorials/linear/mip_duality/","page":"Computing the duals of a mixed-integer program","title":"Computing the duals of a mixed-integer program","text":"undo()\nprint(model)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"EditURL = \"min_ellipse.jl\"","category":"page"},{"location":"tutorials/conic/min_ellipse/#Minimal-ellipses","page":"Minimal ellipses","title":"Minimal ellipses","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This example comes from section 8.4.1 of the book Convex Optimization by Boyd and Vandenberghe (2004).","category":"page"},{"location":"tutorials/conic/min_ellipse/#Formulation","page":"Minimal ellipses","title":"Formulation","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Given a set of m ellipses of the form:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"E(A b c) =  x  x^top A x + 2 b^top x + c leq 0 ","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"the minimal ellipse problem finds an ellipse with the minimum area that encloses the given ellipses.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"It is convenient to parameterize the minimal enclosing ellipse as","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":" x   Px + q  leq 1 ","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Then the optimal P and q are given by the convex semidefinite program;","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"beginaligned\ntextmaximize   quad log(det(P))  \ntextsubject to   quad tau_i geq 0  i = 1 ldots m \n quadbeginbmatrix\n    P^2 - tau_i A_i          P q - tau_i b_i       0        \n    (P q - tau_i b_i)^top    -1 - tau_i c_i     (P q)^top \n                0                     (P q)       - P^2       \nendbmatrix\npreceq 0 text (PSD)    i=1 ldots m\nendaligned","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"with helper variables tau.","category":"page"},{"location":"tutorials/conic/min_ellipse/#Required-packages","page":"Minimal ellipses","title":"Required packages","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"using JuMP\nimport LinearAlgebra\nimport Plots\nimport SCS","category":"page"},{"location":"tutorials/conic/min_ellipse/#Data","page":"Minimal ellipses","title":"Data","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"First, define the m input ellipses (here m = 6), parameterized as x^T A_i x + 2 b_i^T x + c leq 0:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"struct Ellipse\n    A::Matrix{Float64}\n    b::Vector{Float64}\n    c::Float64\n    function Ellipse(A::Matrix{Float64}, b::Vector{Float64}, c::Float64)\n        @assert isreal(A) && LinearAlgebra.issymmetric(A)\n        return new(A, b, c)\n    end\nend\n\nellipses = [\n    Ellipse([1.2576 -0.3873; -0.3873 0.3467], [0.2722, 0.1969], 0.1831),\n    Ellipse([1.4125 -2.1777; -2.1777 6.7775], [-1.228, -0.0521], 0.3295),\n    Ellipse([1.7018 0.8141; 0.8141 1.7538], [-0.4049, 1.5713], 0.2077),\n    Ellipse([0.9742 -0.7202; -0.7202 1.5444], [0.0265, 0.5623], 0.2362),\n    Ellipse([0.6798 -0.1424; -0.1424 0.6871], [-0.4301, -1.0157], 0.3284),\n    Ellipse([0.1796 -0.1423; -0.1423 2.6181], [-0.3286, 0.557], 0.4931),\n];\nnothing #hide","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"We visualise the ellipses using the Plots package:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"function plot_ellipse(plot, ellipse::Ellipse)\n    A, b, c = ellipse.A, ellipse.b, ellipse.c\n    θ = range(0, 2pi + 0.05; step = 0.05)\n    # Some linear algebra to convert θ into (x,y) coordinates.\n    x_y = √A \\ (√(b' * (A \\ b) - c) .* hcat(cos.(θ), sin.(θ)) .- (√A \\ b)')'\n    Plots.plot!(plot, x_y[1, :], x_y[2, :]; label = nothing, c = :navy)\n    return\nend\n\nplot = Plots.plot(; size = (600, 600))\nfor ellipse in ellipses\n    plot_ellipse(plot, ellipse)\nend\nplot","category":"page"},{"location":"tutorials/conic/min_ellipse/#Build-the-model","page":"Minimal ellipses","title":"Build the model","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Now let's build the model, using the change-of-variables P² = P^2 and P_q = P q. We'll recover the true value of P and q after the solve.","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"model = Model(SCS.Optimizer)\n# We need to use a tighter tolerance for this example, otherwise the bounding\n# ellipse won't actually be bounding...\nset_attribute(model, \"eps_rel\", 1e-6)\nset_silent(model)\nm, n = length(ellipses), size(first(ellipses).A, 1)\n@variable(model, τ[1:m] >= 0)\n@variable(model, P²[1:n, 1:n], PSD)\n@variable(model, P_q[1:n])\n\nfor (i, ellipse) in enumerate(ellipses)\n    A, b, c = ellipse.A, ellipse.b, ellipse.c\n    X = [\n        #! format: off\n        (P² - τ[i] * A)   (P_q - τ[i] * b) zeros(n, n)\n        (P_q - τ[i] * b)' (-1 - τ[i] * c)  P_q'\n        zeros(n, n)       P_q              -P²\n        #! format: on\n    ]\n    @constraint(model, LinearAlgebra.Symmetric(X) <= 0, PSDCone())\nend","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"We cannot directly represent the objective log(det(P)), so we introduce the conic reformulation:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"@variable(model, log_det_P)\n@constraint(model, [log_det_P; 1; vec(P²)] in MOI.LogDetConeSquare(n))\n@objective(model, Max, log_det_P)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Now, solve the program:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"optimize!(model)\nsolution_summary(model)","category":"page"},{"location":"tutorials/conic/min_ellipse/#Results","page":"Minimal ellipses","title":"Results","text":"","category":"section"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"After solving the model to optimality we can recover the solution in terms of P and q:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"P = sqrt(value.(P²))\nq = P \\ value.(P_q)","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Finally, overlaying the solution in the plot we see the minimal area enclosing ellipsoid:","category":"page"},{"location":"tutorials/conic/min_ellipse/","page":"Minimal ellipses","title":"Minimal ellipses","text":"Plots.plot!(\n    plot,\n    [tuple(P \\ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)];\n    c = :crimson,\n    label = nothing,\n)","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"EditURL = \"https://github.com/jump-dev/CSDP.jl/blob/3a2e4c7c48af7660a5ba258a4cb6549ab0970366/README.md\"","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"(Image: )","category":"page"},{"location":"packages/CSDP/#CSDP.jl","page":"jump-dev/CSDP.jl","title":"CSDP.jl","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"(Image: Build Status) (Image: codecov)","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP.jl is a wrapper for the COIN-OR SemiDefinite Programming solver.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The wrapper has two components:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"a thin wrapper around the low-level C API\nan interface to MathOptInterface","category":"page"},{"location":"packages/CSDP/#Affiliation","page":"jump-dev/CSDP.jl","title":"Affiliation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"This wrapper is maintained by the JuMP community and is not a COIN-OR project.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The original algorithm is described by B. Borchers (1999). CSDP, A C Library for Semidefinite Programming. Optimization Methods and Software. 11(1), 613-623. [preprint]","category":"page"},{"location":"packages/CSDP/#License","page":"jump-dev/CSDP.jl","title":"License","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP.jl is licensed under the MIT License.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The underlying solver, coin-or/Csdp, is licensed under the Eclipse public license.","category":"page"},{"location":"packages/CSDP/#Installation","page":"jump-dev/CSDP.jl","title":"Installation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"Install CSDP using Pkg.add:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"import Pkg\nPkg.add(\"CSDP\")","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In addition to installing the CSDP.jl package, this will also download and install the CSDP binaries. You do not need to install CSDP separately.","category":"page"},{"location":"packages/CSDP/#Use-with-JuMP","page":"jump-dev/CSDP.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"To use CSDP with JuMP, use CSDP.Optimizer:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"using JuMP, CSDP\nmodel = Model(CSDP.Optimizer)\nset_attribute(model, \"maxiter\", 1000)","category":"page"},{"location":"packages/CSDP/#MathOptInterface-API","page":"jump-dev/CSDP.jl","title":"MathOptInterface API","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The CSDP optimizer supports the following constraints and attributes.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported objective functions:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported variable types:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.Nonnegatives\nMOI.PositiveSemidefiniteConeTriangle","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported constraint types:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ScalarAffineFunction{Float64} in MOI.EqualTo{Float64}","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"List of supported model attributes:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"MOI.ObjectiveSense()","category":"page"},{"location":"packages/CSDP/#Options","page":"jump-dev/CSDP.jl","title":"Options","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The CSDP options are listed in the table below.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"Name Default Value Explanation\naxtol 1.0e-8 Tolerance for primal feasibility\natytol 1.0e-8 Tolerance for dual feasibility\nobjtol 1.0e-8 Tolerance for relative duality gap\npinftol 1.0e8 Tolerance for determining primal infeasibility\ndinftol 1.0e8 Tolerance for determining dual infeasibility\nmaxiter 100 Limit for the total number of iterations\nminstepfrac 0.90 The minstepfrac and maxstepfrac parameters determine how close to the edge of the feasible region CSDP will step\nmaxstepfrac 0.97 The minstepfrac and maxstepfrac parameters determine how close to the edge of the feasible region CSDP will step\nminstepp 1.0e-8 If the primal step is shorter than minstepp then CSDP declares a line search failure\nminstepd 1.0e-8 If the dual step is shorter than minstepd then CSDP declares a line search failure\nusexzgap 1 If usexzgap is 0 then CSDP will use the objective duality gap d - p instead of the XY duality gap ⟨Z, X⟩\ntweakgap 0 If tweakgap is set to 1, and usexzgap is set to 0, then CSDP will attempt to \"fix\" negative duality gaps\naffine 0 If affine is set to 1, then CSDP will take only primal-dual affine steps and not make use of the barrier term. This can be useful for some problems that do not have feasible solutions that are strictly in the interior of the cone of semidefinite matrices\nperturbobj 1 The perturbobj parameter determines whether the objective function will be perturbed to help deal with problems that have unbounded optimal solution sets. If perturbobj is 0, then the objective will not be perturbed. If perturbobj is 1, then the objective function will be perturbed by a default amount. Larger values of perturbobj (for example, 100) increase the size of the perturbation. This can be helpful in solving some difficult problems.\nfastmode 0 The fastmode parameter determines whether or not CSDP will skip certain time consuming operations that slightly improve the accuracy of the solutions. If fastmode is set to 1, then CSDP may be somewhat faster, but also somewhat less accurate\nprintlevel 1 The printlevel parameter determines how much debugging information is output. Use a printlevel of 0 for no output and a printlevel of 1 for normal output. Higher values of printlevel will generate more debugging output","category":"page"},{"location":"packages/CSDP/#Problem-representation","page":"jump-dev/CSDP.jl","title":"Problem representation","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The primal is represented internally by CSDP as follows:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"max ⟨C, X⟩\n      A(X) = a\n         X ⪰ 0","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"where A(X) = [⟨A_1, X⟩, ..., ⟨A_m, X⟩]. The corresponding dual is:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"min ⟨a, y⟩\n     A'(y) - C = Z\n             Z ⪰ 0","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"where A'(y) = y_1A_1 + ... + y_mA_m","category":"page"},{"location":"packages/CSDP/#Termination-criteria","page":"jump-dev/CSDP.jl","title":"Termination criteria","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP will terminate successfully (or partially) in the following cases:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"If CSDP finds X, Z ⪰ 0 such that the following 3 tolerances are satisfied:\nprimal feasibility tolerance: ||A(x) - a||_2 / (1 + ||a||_2) < axtol\ndual feasibility tolerance: ||A'(y) - C - Z||_F / (1 + ||C||_F) < atytol\nrelative duality gap tolerance: gap / (1 + |⟨a, y⟩| + |⟨C, X⟩|) < objtol\nobjective duality gap: if usexygap is 0, gap = ⟨a, y⟩ - ⟨C, X⟩\nXY duality gap: if usexygap is 1, gap = ⟨Z, X⟩\nthen it returns 0.\nIf CSDP finds y and Z ⪰ 0 such that -⟨a, y⟩ / ||A'(y) - Z||_F > pinftol, it returns 1 with y such that ⟨a, y⟩ = -1.\nIf CSDP finds X ⪰ 0 such that ⟨C, X⟩ / ||A(X)||_2 > dinftol, it returns 2 with X such that ⟨C, X⟩ = 1.\nIf CSDP finds X, Z ⪰ 0 such that the following 3 tolerances are satisfied with 1000*axtol, 1000*atytol and 1000*objtol but at least one of them is not satisfied with axtol, atytol and objtol and cannot make progress, then it returns 3.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In addition, if the printlevel option is at least 1, the following will be printed:","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"If the return code is 1, CSDP will print ⟨a, y⟩ and ||A'(y) - Z||_F\nIf the return code is 2, CSDP will print ⟨C, X⟩ and ||A(X)||_F\nOtherwise, CSDP will print\nthe primal/dual objective value,\nthe relative primal/dual infeasibility,\nthe objective duality gap ⟨a, y⟩ - ⟨C, X⟩ and objective relative duality gap (⟨a, y⟩ - ⟨C, X⟩) / (1 + |⟨a, y⟩| + |⟨C, X⟩|),\nthe XY duality gap ⟨Z, X⟩ and XY relative duality gap ⟨Z, X⟩ / (1 + |⟨a, y⟩| + |⟨C, X⟩|)\nand the DIMACS error measures.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"In theory, for feasible primal and dual solutions, ⟨a, y⟩ - ⟨C, X⟩ = ⟨Z, X⟩, so the objective and XY duality gap should be equivalent. However, in practice, there are sometimes solution which satisfy primal and dual feasibility tolerances but have objective duality gap which are not close to XY duality gap. In some cases, the objective duality gap may even become negative (hence the tweakgap option). This is the reason usexygap is 1 by default.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP considers that X ⪰ 0 (resp. Z ⪰ 0) is satisfied when the Cholesky factorizations can be computed. In practice, this is somewhat more conservative than simply requiring all eigenvalues to be nonnegative.","category":"page"},{"location":"packages/CSDP/#Status","page":"jump-dev/CSDP.jl","title":"Status","text":"","category":"section"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"The table below shows how the different CSDP statuses are converted to the MathOptInterface statuses.","category":"page"},{"location":"packages/CSDP/","page":"jump-dev/CSDP.jl","title":"jump-dev/CSDP.jl","text":"CSDP code State Description MOI status\n0 Success SDP solved MOI.OPTIMAL\n1 Success The problem is primal infeasible, and we have a certificate MOI.INFEASIBLE\n2 Success The problem is dual infeasible, and we have a certificate MOI.DUAL_INFEASIBLE\n3 Partial Success A solution has been found, but full accuracy was not achieved MOI.ALMOST_OPTIMAL\n4 Failure Maximum iterations reached MOI.ITERATION_LIMIT\n5 Failure Stuck at edge of primal feasibility MOI.SLOW_PROGRESS\n6 Failure Stuck at edge of dual infeasibility MOI.SLOW_PROGRESS\n7 Failure Lack of progress MOI.SLOW_PROGRESS\n8 Failure X, Z, or O was singular MOI.NUMERICAL_ERROR\n9 Failure Detected NaN or Inf values MOI.NUMERICAL_ERROR","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"EditURL = \"start_values.jl\"","category":"page"},{"location":"tutorials/conic/start_values/#Primal-and-dual-warm-starts","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"","category":"section"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Some conic solvers have the ability to set warm-starts for the primal and dual solution. This can improve performance, particularly if you are repeatedly solving a sequence of related problems.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"tip: Tip\nSee set_start_values for a generic implementation of this function that was added to JuMP after this tutorial was written.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"In this tutorial, we demonstrate how to write a function that sets the primal and dual starts as the optimal solution stored in a model. It is intended to be a starting point for which you can modify if you want to do something similar in your own code.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"warning: Warning\nThis tutorial does not set start values for nonlinear models.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"using JuMP\nimport SCS","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"The main component of this tutorial is the following function. The most important observation is that we cache all of the solution values first, and then we modify the model second. (Alternating between querying a value and modifying the model is not allowed in JuMP.)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"function set_optimal_start_values(model::Model)\n    # Store a mapping of the variable primal solution\n    variable_primal = Dict(x => value(x) for x in all_variables(model))\n    # In the following, we loop through every constraint and store a mapping\n    # from the constraint index to a tuple containing the primal and dual\n    # solutions.\n    constraint_solution = Dict()\n    for (F, S) in list_of_constraint_types(model)\n        # We add a try-catch here because some constraint types might not\n        # support getting the primal or dual solution.\n        try\n            for ci in all_constraints(model, F, S)\n                constraint_solution[ci] = (value(ci), dual(ci))\n            end\n        catch\n            @info(\"Something went wrong getting $F-in-$S. Skipping\")\n        end\n    end\n    # Now we can loop through our cached solutions and set the starting values.\n    for (x, primal_start) in variable_primal\n        set_start_value(x, primal_start)\n    end\n    for (ci, (primal_start, dual_start)) in constraint_solution\n        set_start_value(ci, primal_start)\n        set_dual_start_value(ci, dual_start)\n    end\n    return\nend","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"To test our function, we use the following linear program:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"model = Model(SCS.Optimizer)\n@variable(model, x[1:3] >= 0)\n@constraint(model, sum(x) <= 1)\n@objective(model, Max, sum(i * x[i] for i in 1:3))\noptimize!(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"By looking at the log, we can see that SCS took 75 iterations to find the optimal solution. Now we set the optimal solution as our starting point:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"set_optimal_start_values(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"and we re-optimize:","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"optimize!(model)","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Now the optimization terminates after 0 iterations because our starting point is already optimal.","category":"page"},{"location":"tutorials/conic/start_values/","page":"Primal and dual warm-starts","title":"Primal and dual warm-starts","text":"Note that some solvers do not support setting some parts of the starting solution, for example, they may support only set_start_value for variables. If you encounter an UnsupportedSupported attribute error for MOI.VariablePrimalStart, MOI.ConstraintPrimalStart, or MOI.ConstraintDualStart, comment out the corresponding part of the set_optimal_start_values function.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP, HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/solutions/#jump_solutions","page":"Solutions","title":"Solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"This section of the manual describes how to access a solved solution to a problem. It uses the following model as an example:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x >= 0)\n           @variable(model, y[[:a, :b]] <= 1)\n           @objective(model, Max, -12x - 20y[:a])\n           @expression(model, my_expr, 6x + 8y[:a])\n           @constraint(model, my_expr >= 100)\n           @constraint(model, c1, 7x + 12y[:a] >= 120)\n           optimize!(model)\n           print(model)\n       end\nMax -12 x - 20 y[a]\nSubject to\n 6 x + 8 y[a] ≥ 100\n c1 : 7 x + 12 y[a] ≥ 120\n x ≥ 0\n y[a] ≤ 1\n y[b] ≤ 1","category":"page"},{"location":"manual/solutions/#Solutions-summary","page":"Solutions","title":"Solutions summary","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"solution_summary can be used for checking the summary of the optimization solutions.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> solution_summary(model)\n* Solver : HiGHS\n\n* Status\n  Result count       : 1\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"kHighsModelStatusOptimal\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : FEASIBLE_POINT\n  Objective value    : -2.05143e+02\n  Objective bound    : -0.00000e+00\n  Relative gap       : Inf\n  Dual objective value : -2.05143e+02\n\n* Work counters\n  Solve time (sec)   : 6.70987e-04\n  Simplex iterations : 2\n  Barrier iterations : 0\n  Node count         : -1\n\njulia> solution_summary(model; verbose = true)\n* Solver : HiGHS\n\n* Status\n  Result count       : 1\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"kHighsModelStatusOptimal\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : FEASIBLE_POINT\n  Objective value    : -2.05143e+02\n  Objective bound    : -0.00000e+00\n  Relative gap       : Inf\n  Dual objective value : -2.05143e+02\n  Primal solution :\n    x : 1.54286e+01\n    y[a] : 1.00000e+00\n    y[b] : 1.00000e+00\n  Dual solution :\n    c1 : 1.71429e+00\n\n* Work counters\n  Solve time (sec)   : 6.70987e-04\n  Simplex iterations : 2\n  Barrier iterations : 0\n  Node count         : -1","category":"page"},{"location":"manual/solutions/#Why-did-the-solver-stop?","page":"Solutions","title":"Why did the solver stop?","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Usetermination_status to understand why the solver stopped.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The MOI.TerminationStatusCode enum describes the full list of statuses that could be returned.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Common return values include OPTIMAL, LOCALLY_SOLVED, INFEASIBLE, DUAL_INFEASIBLE, and TIME_LIMIT.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nA return status of OPTIMAL means the solver found (and proved) a globally optimal solution. A return status of LOCALLY_SOLVED means the solver found a locally optimal solution (which may also be globally optimal, but it could not prove so).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nA return status of DUAL_INFEASIBLE does not guarantee that the primal is unbounded. When the dual is infeasible, the primal is unbounded if there exists a feasible primal solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use raw_status to get a solver-specific string explaining why the optimization stopped:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> raw_status(model)\n\"kHighsModelStatusOptimal\"","category":"page"},{"location":"manual/solutions/#Primal-solutions","page":"Solutions","title":"Primal solutions","text":"","category":"section"},{"location":"manual/solutions/#Primal-solution-status","page":"Solutions","title":"Primal solution status","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use primal_status to return an MOI.ResultStatusCode enum describing the status of the primal solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_status(model)\nFEASIBLE_POINT::ResultStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Other common returns are NO_SOLUTION, and INFEASIBILITY_CERTIFICATE. The first means that the solver doesn't have a solution to return, and the second means that the primal solution is a certificate of dual infeasibility (a primal unbounded ray).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use has_values, which returns true if there is a solution that can be queried, and false otherwise.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> has_values(model)\ntrue","category":"page"},{"location":"manual/solutions/#Objective-values","page":"Solutions","title":"Objective values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The objective value of a solved problem can be obtained via objective_value. The best known bound on the optimal objective value can be obtained via objective_bound. If the solver supports it, the value of the dual objective can be obtained via dual_objective_value.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> objective_value(model)\n-205.14285714285714\n\njulia> objective_bound(model)  # HiGHS only implements objective bound for MIPs\n-0.0\n\njulia> dual_objective_value(model)\n-205.1428571428571","category":"page"},{"location":"manual/solutions/#Primal-solution-values","page":"Solutions","title":"Primal solution values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the solver has a primal solution to return, use value to access it:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(x)\n15.428571428571429","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Broadcast value over containers:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value.(y)\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n 1.0\n 1.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"value also works on expressions:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(my_expr)\n100.57142857142857","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"and constraints:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> value(c1)\n120.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"info: Info\nCalling value on a constraint returns the constraint function evaluated at the solution.","category":"page"},{"location":"manual/solutions/#Dual-solutions","page":"Solutions","title":"Dual solutions","text":"","category":"section"},{"location":"manual/solutions/#Dual-solution-status","page":"Solutions","title":"Dual solution status","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Use dual_status to return an MOI.ResultStatusCode enum describing the status of the dual solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual_status(model)\nFEASIBLE_POINT::ResultStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Other common returns are NO_SOLUTION, and INFEASIBILITY_CERTIFICATE. The first means that the solver doesn't have a solution to return, and the second means that the dual solution is a certificate of primal infeasibility (a dual unbounded ray).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use has_duals, which returns true if there is a solution that can be queried, and false otherwise.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> has_duals(model)\ntrue","category":"page"},{"location":"manual/solutions/#Dual-solution-values","page":"Solutions","title":"Dual solution values","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the solver has a dual solution to return, use dual to access it:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual(c1)\n1.7142857142857142","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Query the duals of variable bounds using LowerBoundRef, UpperBoundRef, and FixRef:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> dual(LowerBoundRef(x))\n0.0\n\njulia> dual.(UpperBoundRef.(y))\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n -0.5714285714285694\n  0.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nJuMP's definition of duality is independent of the objective sense. That is, the sign of feasible duals associated with a constraint depends on the direction of the constraint and not whether the problem is maximization or minimization. This is a different convention from linear programming duality in some common textbooks. If you have a linear program, and you want the textbook definition, you probably want to use shadow_price and reduced_cost instead.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> shadow_price(c1)\n1.7142857142857142\n\njulia> reduced_cost(x)\n-0.0\n\njulia> reduced_cost.(y)\n1-dimensional DenseAxisArray{Float64,1,...} with index sets:\n    Dimension 1, [:a, :b]\nAnd data, a 2-element Vector{Float64}:\n  0.5714285714285694\n -0.0","category":"page"},{"location":"manual/solutions/#Recommended-workflow","page":"Solutions","title":"Recommended workflow","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The recommended workflow for solving a model and querying the solution is something like the following:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           if termination_status(model) == OPTIMAL\n               println(\"Solution is optimal\")\n           elseif termination_status(model) == TIME_LIMIT && has_values(model)\n               println(\"Solution is suboptimal due to a time limit, but a primal solution is available\")\n           else\n               error(\"The model was not solved correctly.\")\n           end\n           println(\"  objective value = \", objective_value(model))\n           if primal_status(model) == FEASIBLE_POINT\n               println(\"  primal solution: x = \", value(x))\n           end\n           if dual_status(model) == FEASIBLE_POINT\n               println(\"  dual solution: c1 = \", dual(c1))\n           end\n       end\nSolution is optimal\n  objective value = -205.14285714285714\n  primal solution: x = 15.428571428571429\n  dual solution: c1 = 1.7142857142857142","category":"page"},{"location":"manual/solutions/#OptimizeNotCalled-errors","page":"Solutions","title":"OptimizeNotCalled errors","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Due to differences in how solvers cache solutions internally, modifying a model after calling optimize! will reset the model into the MOI.OPTIMIZE_NOT_CALLED state. If you then attempt to query solution information, an OptimizeNotCalled error will be thrown.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you are iteratively querying solution information and modifying a model, query all the results first, then modify the problem.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"For example, instead of:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> x_val = value(x)\n┌ Warning: The model has been modified since the last call to `optimize!` (or `optimize!` has not been called yet). If you are iteratively querying solution information and modifying a model, query all the results first, then modify the model.\n└ @ JuMP ~/work/JuMP.jl/JuMP.jl/src/optimizer_interface.jl:712\nERROR: OptimizeNotCalled()\nStacktrace:\n[...]\n\njulia> termination_status(model)\nOPTIMIZE_NOT_CALLED::TerminationStatusCode = 0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"do","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model);\n\njulia> x_val = value(x)\n0.0\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> set_lower_bound(x, x_val)\n\njulia> termination_status(model)\nOPTIMIZE_NOT_CALLED::TerminationStatusCode = 0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you know that your particular solver supports querying solution information after modifications, you can use direct_model to bypass the MOI.OPTIMIZE_NOT_CALLED state:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = direct_model(HiGHS.Optimizer());\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0);\n\njulia> optimize!(model)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1\n\njulia> set_upper_bound(x, 1)\n\njulia> x_val = value(x)\n0.0\n\njulia> set_lower_bound(x, x_val)\n\njulia> termination_status(model)\nOPTIMAL::TerminationStatusCode = 1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nBe careful doing this. If your particular solver does not support querying solution information after modification, it may silently return incorrect solutions or throw an error.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"# TODO: How to accurately measure the solve time.","category":"page"},{"location":"manual/solutions/#Accessing-attributes","page":"Solutions","title":"Accessing attributes","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"MathOptInterface defines many model attributes that can be queried. Some attributes can be directly accessed by getter functions. These include:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"solve_time\nrelative_gap\nsimplex_iterations\nbarrier_iterations\nnode_count","category":"page"},{"location":"manual/solutions/#Sensitivity-analysis-for-LP","page":"Solutions","title":"Sensitivity analysis for LP","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Given an LP problem and an optimal solution corresponding to a basis, we can question how much an objective coefficient or standard form right-hand side coefficient (c.f., normalized_rhs) can change without violating primal or dual feasibility of the basic solution.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Note that not all solvers compute the basis, and for sensitivity analysis, the solver interface must implement MOI.ConstraintBasisStatus.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"tip: Tip\nRead the Sensitivity analysis of a linear program for more information on sensitivity analysis.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To give a simple example, we could analyze the sensitivity of the optimal solution to the following (non-degenerate) LP problem:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> begin\n           model = Model(HiGHS.Optimizer)\n           set_silent(model)\n           @variable(model, x[1:2])\n           set_lower_bound(x[2], -0.5)\n           set_upper_bound(x[2], 0.5)\n           @constraint(model, c1, x[1] + x[2] <= 1)\n           @constraint(model, c2, x[1] - x[2] <= 1)\n           @objective(model, Max, x[1])\n           print(model)\n       end\nMax x[1]\nSubject to\n c1 : x[1] + x[2] ≤ 1\n c2 : x[1] - x[2] ≤ 1\n x[2] ≥ -0.5\n x[2] ≤ 0.5","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To analyze the sensitivity of the problem we could check the allowed perturbation ranges of, for example, the cost coefficients and the right-hand side coefficient of the constraint c1 as follows:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n  1.0\n -0.0\n\njulia> report = lp_sensitivity_report(model);\n\njulia> x1_lo, x1_hi = report[x[1]]\n(-1.0, Inf)\n\njulia> println(\"The objective coefficient of x[1] could decrease by $(x1_lo) or increase by $(x1_hi).\")\nThe objective coefficient of x[1] could decrease by -1.0 or increase by Inf.\n\njulia> x2_lo, x2_hi = report[x[2]]\n(-1.0, 1.0)\n\njulia> println(\"The objective coefficient of x[2] could decrease by $(x2_lo) or increase by $(x2_hi).\")\nThe objective coefficient of x[2] could decrease by -1.0 or increase by 1.0.\n\njulia> c_lo, c_hi = report[c1]\n(-1.0, 1.0)\n\njulia> println(\"The RHS of c1 could decrease by $(c_lo) or increase by $(c_hi).\")\nThe RHS of c1 could decrease by -1.0 or increase by 1.0.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The range associated with a variable is the range of the allowed perturbation of the corresponding objective coefficient. Note that the current primal solution remains optimal within this range; however the corresponding dual solution might change since a cost coefficient is perturbed. Similarly, the range associated with a constraint is the range of the allowed perturbation of the corresponding right-hand side coefficient. In this range the current dual solution remains optimal, but the optimal primal solution might change.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the problem is degenerate, there are multiple optimal bases and hence these ranges might not be as intuitive and seem too narrow, for example, a larger cost coefficient perturbation might not invalidate the optimality of the current primal solution. Moreover, if a problem is degenerate, due to finite precision, it can happen that, for example, a perturbation seems to invalidate a basis even though it doesn't (again providing too narrow ranges). To prevent this, increase the atol keyword argument to lp_sensitivity_report. Note that this might make the ranges too wide for numerically challenging instances. Thus, do not blindly trust these ranges, especially not for highly degenerate or numerically unstable instances.","category":"page"},{"location":"manual/solutions/#Conflicts","page":"Solutions","title":"Conflicts","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"When the model you input is infeasible, some solvers can help you find the cause of this infeasibility by offering a conflict, that is, a subset of the constraints that create this infeasibility. Depending on the solver, this can also be called an IIS (irreducible inconsistent subsystem).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If supported by the solver, use compute_conflict! to trigger the computation of a conflict. Once this process is finished, query the MOI.ConflictStatus attribute to check if a conflict was found.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If found, copy the IIS to a new model using copy_conflict, which you can then print or write to a file for easier debugging:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> using JuMP\n\njulia> import Gurobi\n\njulia> model = Model(Gurobi.Optimizer)\nA JuMP Model\nFeasibility problem with:\nVariables: 0\nModel mode: AUTOMATIC\nCachingOptimizer state: EMPTY_OPTIMIZER\nSolver name: Gurobi\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 0)\nx\n\njulia> @constraint(model, c1, x >= 2)\nc1 : x ≥ 2.0\n\njulia> @constraint(model, c2, x <= 1)\nc2 : x ≤ 1.0\n\njulia> optimize!(model)\n\njulia> compute_conflict!(model)\n\njulia> if get_attribute(model, MOI.ConflictStatus()) == MOI.CONFLICT_FOUND\n           iis_model, _ = copy_conflict(model)\n           print(iis_model)\n       end\nFeasibility\nSubject to\n c1 : x ≥ 2.0\n c2 : x ≤ 1.0","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If you need more control over the list of constraints that appear in the conflict, iterate over the list of constraints and query the MOI.ConstraintConflictStatus attribute:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> list_of_conflicting_constraints = ConstraintRef[]\nConstraintRef[]\n\njulia> for (F, S) in list_of_constraint_types(model)\n           for con in all_constraints(model, F, S)\n               if get_attribute(con, MOI.ConstraintConflictStatus()) == MOI.IN_CONFLICT\n                   push!(list_of_conflicting_constraints, con)\n               end\n           end\n       end\n\njulia> list_of_conflicting_constraints\n2-element Vector{ConstraintRef}:\n c1 : x ≥ 2.0\n c2 : x ≤ 1.0","category":"page"},{"location":"manual/solutions/#Multiple-solutions","page":"Solutions","title":"Multiple solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Some solvers support returning multiple solutions. You can check how many solutions are available to query using result_count.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Functions for querying the solutions, for example, primal_status, dual_status, value, dual, and solution_summary all take an additional keyword argument result which can be used to specify which result to return.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nEven if termination_status is OPTIMAL, some of the returned solutions may be suboptimal. However, if the solver found at least one optimal solution, then result = 1 will always return an optimal solution. Use objective_value to assess the quality of the remaining solutions.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> using JuMP\n\njulia> import MultiObjectiveAlgorithms as MOA\n\njulia> import HiGHS\n\njulia> model = Model(() -> MOA.Optimizer(HiGHS.Optimizer));\n\njulia> set_attribute(model, MOA.Algorithm(), MOA.Dichotomy())\n\njulia> set_silent(model)\n\njulia> @variable(model, x1 >= 0)\nx1\n\njulia> @variable(model, 0 <= x2 <= 3)\nx2\n\njulia> @objective(model, Min, [3x1 + x2, -x1 - 2x2])\n2-element Vector{AffExpr}:\n 3 x1 + x2\n -x1 - 2 x2\n\njulia> @constraint(model, 3x1 - x2 <= 6)\n3 x1 - x2 ≤ 6\n\njulia> optimize!(model)\n\njulia> solution_summary(model; result = 1)\n* Solver : MOA[algorithm=MultiObjectiveAlgorithms.Dichotomy, optimizer=HiGHS]\n\n* Status\n  Result count       : 3\n  Termination status : OPTIMAL\n  Message from the solver:\n  \"Solve complete. Found 3 solution(s)\"\n\n* Candidate solution (result #1)\n  Primal status      : FEASIBLE_POINT\n  Dual status        : NO_SOLUTION\n  Objective value    : [0.00000e+00,0.00000e+00]\n  Objective bound    : [0.00000e+00,-9.00000e+00]\n  Relative gap       : Inf\n  Dual objective value : -9.00000e+00\n\n* Work counters\n  Solve time (sec)   : 1.82720e-03\n  Simplex iterations : 1\n  Barrier iterations : 0\n  Node count         : -1\n\njulia> for i in 1:result_count(model)\n           println(\"Solution $i\")\n           println(\"   x = \", value.([x1, x2]; result = i))\n           println(\" obj = \", objective_value(model; result = i))\n       end\nSolution 1\n   x = [0.0, 0.0]\n obj = [0.0, 0.0]\nSolution 2\n   x = [0.0, 3.0]\n obj = [3.0, -6.0]\nSolution 3\n   x = [3.0, 3.0]\n obj = [12.0, -9.0]","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"tip: Tip\nThe Multi-objective knapsack tutorial provides more examples of querying multiple solutions.","category":"page"},{"location":"manual/solutions/#Checking-feasibility-of-solutions","page":"Solutions","title":"Checking feasibility of solutions","text":"","category":"section"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To check the feasibility of a primal solution, use primal_feasibility_report, which takes a model, a dictionary mapping each variable to a primal solution value (defaults to the last solved solution), and a tolerance atol (defaults to 0.0).","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"The function returns a dictionary which maps the infeasible constraint references to the distance between the primal value of the constraint and the nearest point in the corresponding set. A point is classed as infeasible if the distance is greater than the supplied tolerance atol.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"# Add a filter here because the output of the dictionary is not ordered, and\n# changes in printing order will cause the doctest to fail.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x >= 1, Int);\n\njulia> @variable(model, y);\n\njulia> @constraint(model, c1, x + y <= 1.95);\n\njulia> point = Dict(x => 1.9, y => 0.06);\n\njulia> primal_feasibility_report(model, point)\nDict{Any, Float64} with 2 entries:\n  x integer         => 0.1\n  c1 : x + y ≤ 1.95 => 0.01\n\njulia> primal_feasibility_report(model, point; atol = 0.02)\nDict{Any, Float64} with 1 entry:\n  x integer => 0.1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"If the point is feasible, an empty dictionary is returned:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_feasibility_report(model, Dict(x => 1.0, y => 0.0))\nDict{Any, Float64}()","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"To use the primal solution from a solve, omit the point argument:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> primal_feasibility_report(model; atol = 0.0)\nDict{Any, Float64}()","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Calling primal_feasibility_report without the point argument is useful when primal_status is FEASIBLE_POINT or NEARLY_FEASIBLE_POINT, and you want to assess the solution quality.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"warning: Warning\nTo apply primal_feasibility_report to infeasible models, you must also provide a candidate point (solvers generally do not provide one). To diagnose the source of infeasibility, see Conflicts.","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"Pass skip_mising = true to skip constraints which contain variables that are not in point:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> primal_feasibility_report(model, Dict(x => 2.1); skip_missing = true)\nDict{Any, Float64} with 1 entry:\n  x integer => 0.1","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"You can also use the functional form, where the first argument is a function that maps variables to their primal values:","category":"page"},{"location":"manual/solutions/","page":"Solutions","title":"Solutions","text":"julia> optimize!(model)\n\njulia> primal_feasibility_report(v -> value(v), model)\nDict{Any, Float64}()","category":"page"},{"location":"extensions/DimensionalData/#DimensionalData.jl","page":"rafaqz/DimensionalData.jl","title":"DimensionalData.jl","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"DimensionalData.jl provides tools and abstractions for working with rectangular arrays that have named dimensions.","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"compat: Compat\nUsing the DimensionalData extension with JuMP requires Julia v1.9 or later.","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"The DimensionalData extension in JuMP lets you construct a DimensionalData.DimArray as an alternative to Containers.DenseAxisArray in the JuMP macros.","category":"page"},{"location":"extensions/DimensionalData/#License","page":"rafaqz/DimensionalData.jl","title":"License","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"DimensionalData.jl is licensed under the MIT license.","category":"page"},{"location":"extensions/DimensionalData/#Installation","page":"rafaqz/DimensionalData.jl","title":"Installation","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Install DimensionalData using Pkg.add:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"import Pkg\nPkg.add(\"DimensionalData\")","category":"page"},{"location":"extensions/DimensionalData/#Use-with-JuMP","page":"rafaqz/DimensionalData.jl","title":"Use with JuMP","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Activate the extension by loading both JuMP and DimensionalData:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> using JuMP, DimensionalData","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Then, pass container = DimensionalData.DimArray in the @variable, @constraint, or @expression macros:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> model = Model();\n\njulia> @variable(\n           model,\n           x[i = 2:4, j = [\"a\", \"b\"]] >= i,\n           container = DimensionalData.DimArray,\n       )\n3×2 DimArray{VariableRef,2} with dimensions:\n  Dim{:i} Sampled{Int64} 2:4 ForwardOrdered Regular Points,\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n    \"a\"     \"b\"\n 2  x[2,a]  x[2,b]\n 3  x[3,a]  x[3,b]\n 4  x[4,a]  x[4,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"Here x is a DimensionalData.Dim array object, so indexing uses the DimensionalData syntax:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> x[At(2), At(\"a\")]\nx[2,a]\n\njulia> x[2, 2]\nx[3,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"You can use container = DimensionalData.DimArray in the @expression macro:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> @expression(\n           model,\n           expr[j = [\"a\", \"b\"]],\n           sum(x[At(i), At(j)] for i in 2:4),\n           container = DimensionalData.DimArray,\n       )\n2-element DimArray{AffExpr,1} with dimensions:\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n \"a\"  x[2,a] + x[3,a] + x[4,a]\n \"b\"  x[2,b] + x[3,b] + x[4,b]","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"and in @constraint:","category":"page"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"julia> @constraint(\n           model,\n           [j = [\"a\", \"b\"]],\n           expr[At(j)] <= 1,\n           container = DimensionalData.DimArray,\n       )\n2-element DimArray{ConstraintRef{Model, MathOptInterface.ConstraintIndex{MathOptInterface.ScalarAffineFunction{Float64}, MathOptInterface.LessThan{Float64}}, ScalarShape},1} with dimensions:\n  Dim{:j} Categorical{String} String[\"a\", \"b\"] ForwardOrdered\n \"a\"  x[2,a] + x[3,a] + x[4,a] ≤ 1\n \"b\"  x[2,b] + x[3,b] + x[4,b] ≤ 1","category":"page"},{"location":"extensions/DimensionalData/#Documentation","page":"rafaqz/DimensionalData.jl","title":"Documentation","text":"","category":"section"},{"location":"extensions/DimensionalData/","page":"rafaqz/DimensionalData.jl","title":"rafaqz/DimensionalData.jl","text":"See the DimensionalData.jl documentation for more details on the syntax and features of DimensionalData.DimArray.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"EditURL = \"getting_started_with_JuMP.jl\"","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Getting-started-with-JuMP","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"This tutorial is aimed at providing a quick introduction to writing and solving optimization models with JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"If you're new to Julia, start by reading Getting started with Julia.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-JuMP?","page":"Getting started with JuMP","title":"What is JuMP?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP (\"Julia for Mathematical Programming\") is an open-source modeling language that is embedded in Julia. It allows users to formulate various classes of optimization problems (linear, mixed-integer, quadratic, conic quadratic, semidefinite, and nonlinear) with easy-to-read code. These problems can then be solved using state-of-the-art open-source and commercial solvers.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP also makes advanced optimization techniques easily accessible from a high-level language.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-a-solver?","page":"Getting started with JuMP","title":"What is a solver?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"A solver is a software package that incorporates algorithms for finding solutions to one or more classes of problem.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"For example, HiGHS is a solver for linear programming (LP) and mixed integer programming (MIP) problems. It incorporates algorithms such as the simplex method and the interior-point method.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"The Supported-solvers table lists the open-source and commercial solvers that JuMP currently supports.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#What-is-MathOptInterface?","page":"Getting started with JuMP","title":"What is MathOptInterface?","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Each solver has its own concepts and data structures for representing optimization models and obtaining results.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"MathOptInterface (MOI) is an abstraction layer that JuMP uses to convert from the problem written in JuMP to the solver-specific data structures for each solver.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"MOI can be used directly, or through a higher-level modeling interface like JuMP.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Because JuMP is built on top of MOI, you'll often see the MathOptInterface. prefix displayed when JuMP types are printed. However, you'll only need to understand and interact with MOI to accomplish advanced tasks such as creating solver-independent callbacks.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Installation","page":"Getting started with JuMP","title":"Installation","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP is a package for Julia. From Julia, JuMP is installed by using the built-in package manager.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"import Pkg\nPkg.add(\"JuMP\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"You also need to include a Julia package which provides an appropriate solver. One such solver is HiGHS.Optimizer, which is provided by the HiGHS.jl package.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"import Pkg\nPkg.add(\"HiGHS\")","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"See Installation Guide for a list of other solvers you can use.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#An-example","page":"Getting started with JuMP","title":"An example","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Let's solve the following linear programming problem using JuMP and HiGHS. We will first look at the complete code to solve the problem and then go through it step by step.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Here's the problem:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"beginaligned\n min  12x + 20y \n textst  6x + 8y geq 100 \n  7x + 12y geq 120 \n  x geq 0 \n  y in 0 3 \nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"And here's the code to solve this problem:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using JuMP\nusing HiGHS\nmodel = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, 0 <= y <= 3)\n@objective(model, Min, 12x + 20y)\n@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)\nprint(model)\noptimize!(model)\ntermination_status(model)\nprimal_status(model)\ndual_status(model)\nobjective_value(model)\nvalue(x)\nvalue(y)\nshadow_price(c1)\nshadow_price(c2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Step-by-step","page":"Getting started with JuMP","title":"Step-by-step","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Once JuMP is installed, to use JuMP in your programs write:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using JuMP","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We also need to include a Julia package which provides an appropriate solver. We want to use HiGHS.Optimizer here which is provided by the HiGHS.jl package:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"using HiGHS","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP builds problems incrementally in a Model object. Create a model by passing an optimizer to the Model function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Variables are modeled using @variable:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, x >= 0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"info: Info\nThe macro creates a new Julia object, x, in the current scope. We could have made this more explicit by writing:x = @variable(model, x >= 0)but the macro does this automatically for us to save writing x twice.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Variables can have lower and upper bounds:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, 0 <= y <= 30)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"The objective is set using @objective:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@objective(model, Min, 12x + 20y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Constraints are modeled using @constraint. Here, c1 and c2 are the names of our constraint:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, c1, 6x + 8y >= 100)\n@constraint(model, c2, 7x + 12y >= 120)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Call print to display the model:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"print(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"To solve the optimization problem, call the optimize! function:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"optimize!(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"info: Info\nThe ! after optimize is part of the name. It's nothing special. Julia has a convention that functions which mutate their arguments should end in !. A common example is push!.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Now let's see what information we can query about the solution.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"termination_status tells us why the solver stopped:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"termination_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"In this case, the solver found an optimal solution.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Check primal_status to see if the solver found a primal feasible point:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"primal_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"and dual_status to see if the solver found a dual feasible point:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"dual_status(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Now we know that our solver found an optimal solution, and that it has a primal and a dual solution to query.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Query the objective value using objective_value:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"objective_value(model)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"the primal solution using value:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"value(x)\nvalue(y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"and the dual solution using shadow_price:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"shadow_price(c1)\nshadow_price(c2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"That's it for our simple model. In the rest of this tutorial, we expand on some of the basic JuMP operations.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Model-basics","page":"Getting started with JuMP","title":"Model basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a model by passing an optimizer:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Alternatively, call set_optimizer at any point before calling optimize!:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()\nset_optimizer(model, HiGHS.Optimizer)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"For some solvers, you can also use direct_model, which offers a more efficient connection to the underlying solver:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = direct_model(HiGHS.Optimizer())","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"warning: Warning\nSome solvers do not support direct_model!","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Solver-Options","page":"Getting started with JuMP","title":"Solver Options","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Pass options to solvers with optimizer_with_attributes:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model =\n    Model(optimizer_with_attributes(HiGHS.Optimizer, \"output_flag\" => false))","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"note: Note\nThese options are solver-specific. To find out the various options available, see the GitHub README of the individual solver packages. The link to each solver's GitHub page is in the Supported solvers table.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"You can also pass options with set_attribute:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)\nset_attribute(model, \"output_flag\", false)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Solution-basics","page":"Getting started with JuMP","title":"Solution basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We saw above how to use termination_status and primal_status to understand the solution returned by the solver.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"However, only query solution attributes like value and objective_value if there is an available solution. Here's a recommended way to check:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"function solve_infeasible()\n    model = Model(HiGHS.Optimizer)\n    @variable(model, 0 <= x <= 1)\n    @variable(model, 0 <= y <= 1)\n    @constraint(model, x + y >= 3)\n    @objective(model, Max, x + 2y)\n    optimize!(model)\n    if termination_status(model) != OPTIMAL\n        @warn(\"The model was not solved correctly.\")\n        return\n    end\n    return value(x), value(y)\nend\n\nsolve_infeasible()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Variable-basics","page":"Getting started with JuMP","title":"Variable basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Let's create a new empty model to explain some of the variable syntax:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Variable-bounds","page":"Getting started with JuMP","title":"Variable bounds","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"All of the variables we have created till now have had a bound. We can also create a free variable.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, free_x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"While creating a variable, instead of using the <= and >= syntax, we can also use the lower_bound and upper_bound keyword arguments.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, keyword_x, lower_bound = 1, upper_bound = 2)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can query whether a variable has a bound using the has_lower_bound and has_upper_bound functions. The values of the bound can be obtained using the lower_bound and upper_bound functions.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"has_upper_bound(keyword_x)\nupper_bound(keyword_x)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Note querying the value of a bound that does not exist will result in an error.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"try                         #hide\n    lower_bound(free_x)\ncatch err                   #hide\n    showerror(stderr, err)  #hide\nend                         #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#tutorial_variable_container","page":"Getting started with JuMP","title":"Containers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We have already seen how to add a single variable to a model using the @variable macro. Now let's look at ways to add multiple variables to a model.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP provides data structures for adding collections of variables to a model. These data structures are referred to as containers and are of three types: Arrays, DenseAxisArrays, and SparseAxisArrays.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Arrays","page":"Getting started with JuMP","title":"Arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP arrays are created when you have integer indices that start at 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, a[1:2, 1:2])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"a[1, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"a[2, :]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an n-dimensional variable x in R^n with bounds l le x le u (l u in R^n) as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"n = 10\nl = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10];\nu = [10, 11, 12, 13, 14, 15, 16, 17, 18, 19];\n\n@variable(model, l[i] <= x[i = 1:n] <= u[i])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also create variable bounds that depend upon the indices:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, y[i = 1:2, j = 1:2] >= 2i + j)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#DenseAxisArrays","page":"Getting started with JuMP","title":"DenseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"DenseAxisArrays are used when the indices are not one-based integer ranges. The syntax is similar except with an arbitrary vector as an index as opposed to a one-based range:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, z[i = 2:3, j = 1:2:3] >= 0)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Indices do not have to be integers. They can be any Julia type:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, w[1:5, [\"red\", \"blue\"]] <= 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a DenseAxisArray as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"z[2, 1]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"w[2:3, [\"red\", \"blue\"]]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"See Forcing the container type for more details.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#SparseAxisArrays","page":"Getting started with JuMP","title":"SparseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"SparseAxisArrays are created when the indices do not form a Cartesian product. For example, this applies when indices have a dependence upon previous indices (called triangular indexing):","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, u[i = 1:2, j = i:3])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also conditionally create variables by adding a comparison check that depends upon the named indices and is separated from the indices by a semi-colon ;:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, v[i = 1:9; mod(i, 3) == 0])","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Index elements in a DenseAxisArray as follows:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"u[1, 2]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"v[[3, 6]]","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Integrality","page":"Getting started with JuMP","title":"Integrality","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"JuMP can create binary and integer variables. Binary variables are constrained to the set  0 1, and integer variables are constrained to the set mathbbZ.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Integer-variables","page":"Getting started with JuMP","title":"Integer variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an integer variable by passing  Int:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, integer_x, Int)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or setting the integer keyword to true:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, integer_z, integer = true)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Binary-variables","page":"Getting started with JuMP","title":"Binary variables","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a binary variable by passing Bin:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, binary_x, Bin)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or setting the binary keyword to true:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@variable(model, binary_z, binary = true)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Constraint-basics","page":"Getting started with JuMP","title":"Constraint basics","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We'll need a new model to explain some of the constraint  basics:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model()\n@variable(model, x)\n@variable(model, y)\n@variable(model, z[1:10]);\nnothing #hide","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#tutorial_constraint_container","page":"Getting started with JuMP","title":"Containers","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Just as we had containers for variables, JuMP also provides Arrays, DenseAxisArrays, and SparseAxisArrays for storing collections of constraints. Examples for each container type are given below.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Arrays-2","page":"Getting started with JuMP","title":"Arrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an Array of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:3], i * x <= i + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#DenseAxisArrays-2","page":"Getting started with JuMP","title":"DenseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an DenseAxisArray of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:2, j = 2:3], i * x <= j + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#SparseAxisArrays-2","page":"Getting started with JuMP","title":"SparseAxisArrays","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create an SparseAxisArray of constraints:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i = 1:2, j = 1:2; i != j], i * x <= j + 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Constraints-in-a-loop","page":"Getting started with JuMP","title":"Constraints in a loop","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can add constraints using regular Julia loops:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"for i in 1:3\n    @constraint(model, 6x + 4y >= 5i)\nend","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"or use for each loops inside the @constraint macro:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, [i in 1:3], 6x + 4y >= 5i)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also create constraints such as sum _i = 1^10 z_i leq 1:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@constraint(model, sum(z[i] for i in 1:10) <= 1)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Objective-functions","page":"Getting started with JuMP","title":"Objective functions","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Set an objective function with @objective:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"model = Model(HiGHS.Optimizer)\n@variable(model, x >= 0)\n@variable(model, y >= 0)\n@objective(model, Min, 2x + y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"Create a maximization objective using Max:","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"@objective(model, Max, 2x + y)","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"tip: Tip\nCalling @objective multiple times will over-write the previous objective. This can be useful when you want to solve the same problem with different objectives.","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/#Vectorized-syntax","page":"Getting started with JuMP","title":"Vectorized syntax","text":"","category":"section"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"We can also add constraints and an objective to JuMP using vectorized linear algebra. We'll illustrate this by solving an LP in standard form that is,","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"beginaligned\n min  c^T x \n textst  A x = b \n  x ge 0\nendaligned","category":"page"},{"location":"tutorials/getting_started/getting_started_with_JuMP/","page":"Getting started with JuMP","title":"Getting started with JuMP","text":"vector_model = Model(HiGHS.Optimizer)\nA = [1 1 9 5; 3 5 0 8; 2 0 6 13]\nb = [7, 3, 5]\nc = [1, 3, 5, 2]\n@variable(vector_model, x[1:4] >= 0)\n@constraint(vector_model, A * x .== b)\n@objective(vector_model, Min, c' * x)\noptimize!(vector_model)\nobjective_value(vector_model)","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"CurrentModule = JuMP\nDocTestSetup = quote\n    using JuMP\n    import HiGHS\nend\nDocTestFilters = [r\"≤|<=\", r\"≥|>=\", r\" == | = \", r\" ∈ | in \", r\"MathOptInterface|MOI\"]","category":"page"},{"location":"manual/complex/#Complex-number-support","page":"Complex number support","title":"Complex number support","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"This page explains the complex-valued variables and constraints that JuMP supports. For a worked-example using these features, read the Quantum state discrimination tutorial.","category":"page"},{"location":"manual/complex/#Complex-valued-variables","page":"Complex number support","title":"Complex-valued variables","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Create a complex-valued variable using ComplexPlane:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, x in ComplexPlane())\nreal(x) + imag(x) im","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Note that x is not a VariableRef; instead, it is an affine expression with Complex{Float64}-valued coefficients:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(x)\nGenericAffExpr{ComplexF64, VariableRef}","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Behind the scenes, JuMP has created two real-valued variables, with names \"real(x)\" and \"imag(x)\":","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\njulia> name.(all_variables(model))\n2-element Vector{String}:\n \"real(x)\"\n \"imag(x)\"","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Use the real and imag functions on x to return a real-valued affine expression representing each variable:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(real(x))\nAffExpr (alias for GenericAffExpr{Float64, GenericVariableRef{Float64}})\n\njulia> typeof(imag(x))\nAffExpr (alias for GenericAffExpr{Float64, GenericVariableRef{Float64}})","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"To create an anonymous variable, use the set keyword argument:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> x = @variable(model, set = ComplexPlane())\n_[1] + _[2] im","category":"page"},{"location":"manual/complex/#Complex-valued-variable-bounds","page":"Complex number support","title":"Complex-valued variable bounds","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Because complex-valued variables lack a total ordering, the definition of a variable bound for a complex-valued variable is ambiguous. If you pass a real- or complex-valued argument to keywords such as lower_bound, upper_bound, and start_value, JuMP will apply the real and imaginary parts to the associated real-valued variables.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(\n           model,\n           x in ComplexPlane(),\n           lower_bound = 1.0,\n           upper_bound = 2.0 + 3.0im,\n           start = 4im,\n       )\nreal(x) + imag(x) im\n\njulia> vars = all_variables(model)\n2-element Vector{VariableRef}:\n real(x)\n imag(x)\n\njulia> lower_bound.(vars)\n2-element Vector{Float64}:\n 1.0\n 0.0\n\njulia> upper_bound.(vars)\n2-element Vector{Float64}:\n 2.0\n 3.0\n\njulia> start_value.(vars)\n2-element Vector{Float64}:\n 0.0\n 4.0","category":"page"},{"location":"manual/complex/#Complex-valued-equality-constraints","page":"Complex number support","title":"Complex-valued equality constraints","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"JuMP reformulates complex-valued equality constraints into two real-valued constraints: one representing the real part, and one representing the imaginary part. Thus, complex-valued equality constraints can be solved any solver that supports the real-valued constraint type.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"For example:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, (1 + 2im) * x[1] + 3 * x[2] == 4 + 5im)\n(1 + 2im) x[1] + 3 x[2] = (4 + 5im)\n\njulia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n 2.5\n 0.5","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"is equivalent to","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x[1:2]);\n\njulia> @constraint(model, 1 * x[1] + 3 * x[2] == 4)  # real component\nx[1] + 3 x[2] = 4\n\njulia> @constraint(model, 2 * x[1] == 5)  # imag component\n2 x[1] = 5\n\njulia> optimize!(model)\n\njulia> value.(x)\n2-element Vector{Float64}:\n 2.5\n 0.5","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"This also applies if the variables are complex-valued:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x in ComplexPlane());\n\njulia> @constraint(model, (1 + 2im) * x + 3 * x == 4 + 5im)\n(4 + 2im) real(x) + (-2 + 4im) imag(x) = (4 + 5im)\n\njulia> optimize!(model)\n\njulia> value(x)\n1.3 + 0.6000000000000001im","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"which is equivalent to","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model(HiGHS.Optimizer);\n\njulia> set_silent(model)\n\njulia> @variable(model, x_real);\n\njulia> @variable(model, x_imag);\n\njulia> @constraint(model, x_real - 2 * x_imag + 3 * x_real == 4)\n4 x_real - 2 x_imag = 4\n\njulia> @constraint(model, x_imag + 2 * x_real + 3 * x_imag == 5)\n2 x_real + 4 x_imag = 5\n\njulia> optimize!(model)\n\njulia> value(x_real) + value(x_imag) * im\n1.3 + 0.6000000000000001im","category":"page"},{"location":"manual/complex/#Hermitian-PSD-Cones","page":"Complex number support","title":"Hermitian PSD Cones","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"JuMP supports creating matrices where are Hermitian.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, H[1:3, 1:3] in HermitianPSDCone())\n3×3 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n real(H[1,1])                    …  real(H[1,3]) + imag(H[1,3]) im\n real(H[1,2]) - imag(H[1,2]) im     real(H[2,3]) + imag(H[2,3]) im\n real(H[1,3]) - imag(H[1,3]) im     real(H[3,3])","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Behind the scenes, JuMP has created nine real-valued decision variables:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> all_variables(model)\n9-element Vector{VariableRef}:\n real(H[1,1])\n real(H[1,2])\n real(H[2,2])\n real(H[1,3])\n real(H[2,3])\n real(H[3,3])\n imag(H[1,2])\n imag(H[1,3])\n imag(H[2,3])","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"and a Vector{VariableRef}-in-MOI.HermitianPositiveSemidefiniteConeTriangle constraint:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> num_constraints(model, Vector{VariableRef}, MOI.HermitianPositiveSemidefiniteConeTriangle)\n1","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"The MOI.HermitianPositiveSemidefiniteConeTriangle set can be efficiently bridged to MOI.PositiveSemidefiniteConeTriangle, so it can be solved by any solver that supports PSD constraints.","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"Each element of H is an affine expression with Complex{Float64}-valued coefficients:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> typeof(H[1, 1])\nGenericAffExpr{ComplexF64, VariableRef}\n\njulia> typeof(H[2, 1])\nGenericAffExpr{ComplexF64, VariableRef}","category":"page"},{"location":"manual/complex/#Hermitian-PSD-constraints","page":"Complex number support","title":"Hermitian PSD constraints","text":"","category":"section"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"The HermitianPSDCone can also be used in the @constraint macro:","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"julia> model = Model();\n\njulia> @variable(model, x[1:2])\n2-element Vector{VariableRef}:\n x[1]\n x[2]\n\njulia> import LinearAlgebra\n\njulia> H = LinearAlgebra.Hermitian([x[1] 1im; -1im -x[2]])\n2×2 LinearAlgebra.Hermitian{GenericAffExpr{ComplexF64, VariableRef}, Matrix{GenericAffExpr{ComplexF64, VariableRef}}}:\n x[1]  im\n -im   -x[2]\n\njulia> @constraint(model, H in HermitianPSDCone())\n[x[1]  im;\n -im   -x[2]] ∈ HermitianPSDCone()","category":"page"},{"location":"manual/complex/","page":"Complex number support","title":"Complex number support","text":"note: Note\nThe matrix H in H in HermitianPSDCone() must be a LinearAlgebra.Hermitian matrix type. A build_constraint error will be thrown if the matrix is a different matrix type.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"EditURL = \"user_defined_hessians.jl\"","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/#User-defined-Hessians","page":"User-defined Hessians","title":"User-defined Hessians","text":"","category":"section"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"This tutorial was generated using Literate.jl. Download the source as a .jl file.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"In this tutorial, we explain how to write a user-defined operator (see User-defined operators) with a Hessian matrix explicitly provided by the user.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"For a more advanced example, see Nested optimization problems.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"This tutorial uses the following packages:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"using JuMP\nimport Ipopt","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/#Rosenbrock-example","page":"User-defined Hessians","title":"Rosenbrock example","text":"","category":"section"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"As a simple example, we consider the Rosenbrock function:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"rosenbrock(x...) = (1 - x[1])^2 + 100 * (x[2] - x[1]^2)^2","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"which has the gradient vector:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"function ∇rosenbrock(g::AbstractVector, x...)\n    g[1] = 400 * x[1]^3 - 400 * x[1] * x[2] + 2 * x[1] - 2\n    g[2] = 200 * (x[2] - x[1]^2)\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"and the Hessian matrix:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"function ∇²rosenbrock(H::AbstractMatrix, x...)\n    H[1, 1] = 1200 * x[1]^2 - 400 * x[2] + 2\n    # H[1, 2] = -400 * x[1] <-- not needed because Hessian is symmetric\n    H[2, 1] = -400 * x[1]\n    H[2, 2] = 200.0\n    return\nend","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"You may assume the Hessian matrix H is initialized with zeros, and because it is symmetric you need only to fill in the non-zero of the lower-triangular terms.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"The matrix type passed in as H depends on the automatic differentiation system, so make sure the first argument to the Hessian function supports an AbstractMatrix (it may be something other than Matrix{Float64}). However, you may assume only that H supports size(H) and setindex!.","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"Now that we have the function, its gradient, and its Hessian, we can construct a JuMP model, add the operator, and use it in a macro:","category":"page"},{"location":"tutorials/nonlinear/user_defined_hessians/","page":"User-defined Hessians","title":"User-defined Hessians","text":"model = Model(Ipopt.Optimizer)\n@variable(model, x[1:2])\n@operator(model, op_rosenbrock, 2, rosenbrock, ∇rosenbrock, ∇²rosenbrock)\n@objective(model, Min, op_rosenbrock(x[1], x[2]))\noptimize!(model)\nsolution_summary(model; verbose = true)","category":"page"}]
 }
diff --git a/previews/PR3562/should_i_use/index.html b/previews/PR3562/should_i_use/index.html
index f552d20f784..94a7a4087b6 100644
--- a/previews/PR3562/should_i_use/index.html
+++ b/previews/PR3562/should_i_use/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL=".."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../assets/documenter.js"></script><script src="../search_index.js"></script><script src="../siteinfo.js"></script><script src="../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../assets/themeswap.js"></script><link href="../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../"><img class="docs-light-only" src="../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox" checked/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../">Introduction</a></li><li class="is-active"><a class="tocitem" href>Should you use JuMP?</a><ul class="internal"><li><a class="tocitem" href="#When-should-you-use-JuMP?"><span>When should you use JuMP?</span></a></li><li><a class="tocitem" href="#When-should-you-not-use-JuMP?"><span>When should you not use JuMP?</span></a></li></ul></li><li><a class="tocitem" href="../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../manual/models/">Models</a></li><li><a class="tocitem" href="../manual/variables/">Variables</a></li><li><a class="tocitem" href="../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../manual/containers/">Containers</a></li><li><a class="tocitem" href="../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/">Introduction</a></li><li><a class="tocitem" href="../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Introduction</a></li><li class="is-active"><a href>Should you use JuMP?</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Should you use JuMP?</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/should_i_use.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Should-you-use-JuMP?"><a class="docs-heading-anchor" href="#Should-you-use-JuMP?">Should you use JuMP?</a><a id="Should-you-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#Should-you-use-JuMP?" title="Permalink"></a></h1><p>JuMP is an <a href="../background/algebraic_modeling_languages/#algebraic-modeling-language">algebraic modeling language</a> for mathematical optimization written in the <a href="https://julialang.org">Julia language</a>.</p><p>This page explains when you should consider using JuMP, and importantly, when you should <em>not</em> use JuMP.</p><h2 id="When-should-you-use-JuMP?"><a class="docs-heading-anchor" href="#When-should-you-use-JuMP?">When should you use JuMP?</a><a id="When-should-you-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#When-should-you-use-JuMP?" title="Permalink"></a></h2><p>You should use JuMP if you have a constrained optimization problem for which you can formulate using the language of mathematical programming, that is:</p><ul><li>a set of decision variables</li><li>a scalar- or vector-valued objective function</li><li>a set of constraints.</li></ul><p>Key reasons to use JuMP include:</p><ul><li>User friendliness<ul><li>JuMP has syntax that mimics natural mathematical expressions. (See the section on <a href="../background/algebraic_modeling_languages/#algebraic-modeling-language">algebraic modeling languages</a>.)</li></ul></li><li>Solver independence<ul><li>JuMP uses a generic solver-independent interface provided by the <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a> package, making it easy to change between a number of open-source and commercial optimization software packages (&quot;solvers&quot;). The <a href="../installation/#Supported-solvers">Supported solvers</a> section contains a table of the currently supported solvers.</li></ul></li><li>Ease of embedding<ul><li>JuMP itself is written purely in Julia. Solvers are the only binary dependencies.</li><li>JuMP provides automatic installation of many open-source solvers. This is different to modeling languages in Python which require you to download and install a solver yourself.</li><li>Because it is embedded in a general-purpose programming language, JuMP makes it easy to solve optimization problems as part of a larger workflow, for example, inside a simulation, behind a web server, or as a subproblem in a decomposition algorithm. As a trade-off, JuMP&#39;s syntax is constrained by the syntax and functionality available in Julia.</li><li>JuMP is <a href="https://www.mozilla.org/MPL/2.0/">MPL</a> licensed, meaning that it can be embedded in commercial software that complies with the terms of the license.</li></ul></li><li>Speed<ul><li>Benchmarking has shown that JuMP can create problems at similar speeds to special-purpose modeling languages such as <a href="https://ampl.com/">AMPL</a>.</li><li>JuMP communicates with most solvers in memory, avoiding the need to write intermediary files.</li></ul></li><li>Access to advanced algorithmic techniques<ul><li>JuMP supports efficient <em>in-memory</em> re-solves of linear programs, which previously required using solver-specific or low-level C++ libraries.</li><li>JuMP provides access to solver-independent and solver-dependent <a href="../manual/callbacks/#callbacks_manual">Callbacks</a>.</li></ul></li></ul><h2 id="When-should-you-not-use-JuMP?"><a class="docs-heading-anchor" href="#When-should-you-not-use-JuMP?">When should you not use JuMP?</a><a id="When-should-you-not-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#When-should-you-not-use-JuMP?" title="Permalink"></a></h2><p>JuMP supports a broad range of optimization classes. However, there are still some that it doesn&#39;t support, or that are better supported by other software packages.</p><h3 id="You-want-to-optimize-a-complicated-Julia-function"><a class="docs-heading-anchor" href="#You-want-to-optimize-a-complicated-Julia-function">You want to optimize a complicated Julia function</a><a id="You-want-to-optimize-a-complicated-Julia-function-1"></a><a class="docs-heading-anchor-permalink" href="#You-want-to-optimize-a-complicated-Julia-function" title="Permalink"></a></h3><p>Packages in Julia compose well. It&#39;s common for people to pick two unrelated packages and use them in conjunction to create novel behavior. JuMP isn&#39;t one of those packages.</p><p>If you want to optimize an ordinary differential equation from <a href="https://github.com/SciML/DifferentialEquations.jl">DifferentialEquations.jl</a> or tune a neural network from <a href="https://github.com/FluxML/Flux.jl">Flux.jl</a>, consider using other packages such as:</p><ul><li><a href="https://github.com/JuliaNLSolvers/Optim.jl">Optim.jl</a></li><li><a href="https://github.com/SciML/Optimization.jl">Optimization.jl</a></li><li><a href="https://github.com/JuliaSmoothOptimizers/NLPModels.jl">NLPModels.jl</a></li><li><a href="https://github.com/JuliaNonconvex/Nonconvex.jl">Nonconvex.jl</a></li></ul><h3 id="Black-box,-derivative-free,-or-unconstrained-optimization"><a class="docs-heading-anchor" href="#Black-box,-derivative-free,-or-unconstrained-optimization">Black-box, derivative free, or unconstrained optimization</a><a id="Black-box,-derivative-free,-or-unconstrained-optimization-1"></a><a class="docs-heading-anchor-permalink" href="#Black-box,-derivative-free,-or-unconstrained-optimization" title="Permalink"></a></h3><p>JuMP does support nonlinear programs with constraints and objectives containing user-defined operators. However, the functions must be automatically differentiable, or need to provide explicit derivatives. (See <a href="../manual/nonlinear/#jump_user_defined_operators">User-defined operators</a> for more information.)</p><p>If your function is a black-box that is non-differentiable (for example, it is the output of a simulation written in C++), JuMP is not the right tool for the job. This also applies if you want to use a derivative free method.</p><p>Even if your problem is differentiable, if it is unconstrained there is limited benefit (and downsides in the form of more overhead) to using JuMP over tools which are only concerned with function minimization.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/JuliaNLSolvers/Optim.jl">Optim.jl</a></li><li><a href="https://github.com/SciML/Optimization.jl">Optimization.jl</a></li><li><a href="https://github.com/JuliaOpt/NLopt.jl">NLopt.jl</a></li></ul><h3 id="Optimal-control-problems"><a class="docs-heading-anchor" href="#Optimal-control-problems">Optimal control problems</a><a id="Optimal-control-problems-1"></a><a class="docs-heading-anchor-permalink" href="#Optimal-control-problems" title="Permalink"></a></h3><p>JuMP supports formulating optimal control problems as large nonlinear programs (see, for example, <a href="../tutorials/nonlinear/space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory">Optimal control for a Space Shuttle reentry trajectory</a>). However, the nonlinear interface has a number of limitations (for example, the need to write out the dynamics in algebraic form) that mean JuMP might not be the right tool for the job.</p><p>Alternatives to consider are:</p><ul><li><a href="https://web.casadi.org">CasADi [MATLAB/Python]</a>, <a href="https://github.com/ichatzinikolaidis/CasADi.jl">CasADi.jl</a></li><li><a href="https://github.com/infiniteopt/InfiniteOpt.jl">InfiniteOpt.jl</a></li><li><a href="https://pyomo.readthedocs.io/en/stable/modeling_extensions/dae.html">pyomo.DAE [Python]</a></li></ul><h3 id="Disciplined-convex-programming"><a class="docs-heading-anchor" href="#Disciplined-convex-programming">Disciplined convex programming</a><a id="Disciplined-convex-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Disciplined-convex-programming" title="Permalink"></a></h3><p>JuMP does not support <a href="https://dcp.stanford.edu">disciplined convex programming (DCP)</a>.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a></li><li><a href="https://github.com/cvxpy/cvxpy">CVXPY [Python]</a></li><li><a href="https://yalmip.github.io">YALMIP [MATLAB]</a></li></ul><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p><code>Convex.jl</code> is also built on MathOptInterface, and shares the same set of underlying solvers. However, you input problems differently, and Convex.jl checks that the problem is DCP.</p></div></div><h3 id="Stochastic-programming"><a class="docs-heading-anchor" href="#Stochastic-programming">Stochastic programming</a><a id="Stochastic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Stochastic-programming" title="Permalink"></a></h3><p>JuMP requires deterministic input data.</p><p>If you have stochastic input data, consider using a JuMP extension such as:</p><ul><li><a href="https://github.com/infiniteopt/InfiniteOpt.jl">InfiniteOpt.jl</a></li><li><a href="https://github.com/martinbiel/StochasticPrograms.jl">StochasticPrograms.jl</a></li><li><a href="https://github.com/odow/SDDP.jl">SDDP.jl</a></li></ul><h3 id="Polyhedral-computations"><a class="docs-heading-anchor" href="#Polyhedral-computations">Polyhedral computations</a><a id="Polyhedral-computations-1"></a><a class="docs-heading-anchor-permalink" href="#Polyhedral-computations" title="Permalink"></a></h3><p>JuMP does not provide tools for working with the polyhedron formed by the set of linear constraints.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/JuliaPolyhedra/Polyhedra.jl">Polyhedra.jl</a> (See the <a href="https://juliapolyhedra.github.io/Polyhedra.jl/v0.7.6/optimization/#Creating-a-polyhedron-from-the-feasible-set-of-a-JuMP-model">documentation</a> to create a polyhedron from a JuMP model.)</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../">« Introduction</a><a class="docs-footer-nextpage" href="../installation/">Installation Guide »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL=".."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../assets/documenter.js"></script><script src="../search_index.js"></script><script src="../siteinfo.js"></script><script src="../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../assets/themeswap.js"></script><link href="../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../"><img class="docs-light-only" src="../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox" checked/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../">Introduction</a></li><li class="is-active"><a class="tocitem" href>Should you use JuMP?</a><ul class="internal"><li><a class="tocitem" href="#When-should-you-use-JuMP?"><span>When should you use JuMP?</span></a></li><li><a class="tocitem" href="#When-should-you-not-use-JuMP?"><span>When should you not use JuMP?</span></a></li></ul></li><li><a class="tocitem" href="../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox"/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../tutorials/getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../tutorials/getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../tutorials/getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../tutorials/getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../tutorials/linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../tutorials/linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../tutorials/linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../tutorials/linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../tutorials/linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../tutorials/linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../tutorials/linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../tutorials/linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../tutorials/linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../tutorials/linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../tutorials/linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../tutorials/linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../tutorials/linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../tutorials/linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../tutorials/linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../tutorials/linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../tutorials/linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../tutorials/linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../tutorials/linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../tutorials/nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../tutorials/nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../tutorials/nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../tutorials/nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../tutorials/nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../tutorials/nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../tutorials/nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../tutorials/nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../tutorials/conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../tutorials/conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../tutorials/conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../tutorials/conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../tutorials/conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../tutorials/conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../tutorials/conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../tutorials/conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../tutorials/conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../tutorials/conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../tutorials/algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../tutorials/algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../tutorials/algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../tutorials/applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../tutorials/applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../tutorials/applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../tutorials/applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../manual/models/">Models</a></li><li><a class="tocitem" href="../manual/variables/">Variables</a></li><li><a class="tocitem" href="../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../manual/containers/">Containers</a></li><li><a class="tocitem" href="../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/">Introduction</a></li><li><a class="tocitem" href="../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Introduction</a></li><li class="is-active"><a href>Should you use JuMP?</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Should you use JuMP?</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/should_i_use.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Should-you-use-JuMP?"><a class="docs-heading-anchor" href="#Should-you-use-JuMP?">Should you use JuMP?</a><a id="Should-you-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#Should-you-use-JuMP?" title="Permalink"></a></h1><p>JuMP is an <a href="../background/algebraic_modeling_languages/#algebraic-modeling-language">algebraic modeling language</a> for mathematical optimization written in the <a href="https://julialang.org">Julia language</a>.</p><p>This page explains when you should consider using JuMP, and importantly, when you should <em>not</em> use JuMP.</p><h2 id="When-should-you-use-JuMP?"><a class="docs-heading-anchor" href="#When-should-you-use-JuMP?">When should you use JuMP?</a><a id="When-should-you-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#When-should-you-use-JuMP?" title="Permalink"></a></h2><p>You should use JuMP if you have a constrained optimization problem for which you can formulate using the language of mathematical programming, that is:</p><ul><li>a set of decision variables</li><li>a scalar- or vector-valued objective function</li><li>a set of constraints.</li></ul><p>Key reasons to use JuMP include:</p><ul><li>User friendliness<ul><li>JuMP has syntax that mimics natural mathematical expressions. (See the section on <a href="../background/algebraic_modeling_languages/#algebraic-modeling-language">algebraic modeling languages</a>.)</li></ul></li><li>Solver independence<ul><li>JuMP uses a generic solver-independent interface provided by the <a href="https://github.com/jump-dev/MathOptInterface.jl">MathOptInterface</a> package, making it easy to change between a number of open-source and commercial optimization software packages (&quot;solvers&quot;). The <a href="../installation/#Supported-solvers">Supported solvers</a> section contains a table of the currently supported solvers.</li></ul></li><li>Ease of embedding<ul><li>JuMP itself is written purely in Julia. Solvers are the only binary dependencies.</li><li>JuMP provides automatic installation of many open-source solvers. This is different to modeling languages in Python which require you to download and install a solver yourself.</li><li>Because it is embedded in a general-purpose programming language, JuMP makes it easy to solve optimization problems as part of a larger workflow, for example, inside a simulation, behind a web server, or as a subproblem in a decomposition algorithm. As a trade-off, JuMP&#39;s syntax is constrained by the syntax and functionality available in Julia.</li><li>JuMP is <a href="https://www.mozilla.org/MPL/2.0/">MPL</a> licensed, meaning that it can be embedded in commercial software that complies with the terms of the license.</li></ul></li><li>Speed<ul><li>Benchmarking has shown that JuMP can create problems at similar speeds to special-purpose modeling languages such as <a href="https://ampl.com/">AMPL</a>.</li><li>JuMP communicates with most solvers in memory, avoiding the need to write intermediary files.</li></ul></li><li>Access to advanced algorithmic techniques<ul><li>JuMP supports efficient <em>in-memory</em> re-solves of linear programs, which previously required using solver-specific or low-level C++ libraries.</li><li>JuMP provides access to solver-independent and solver-dependent <a href="../manual/callbacks/#callbacks_manual">Callbacks</a>.</li></ul></li></ul><h2 id="When-should-you-not-use-JuMP?"><a class="docs-heading-anchor" href="#When-should-you-not-use-JuMP?">When should you not use JuMP?</a><a id="When-should-you-not-use-JuMP?-1"></a><a class="docs-heading-anchor-permalink" href="#When-should-you-not-use-JuMP?" title="Permalink"></a></h2><p>JuMP supports a broad range of optimization classes. However, there are still some that it doesn&#39;t support, or that are better supported by other software packages.</p><h3 id="You-want-to-optimize-a-complicated-Julia-function"><a class="docs-heading-anchor" href="#You-want-to-optimize-a-complicated-Julia-function">You want to optimize a complicated Julia function</a><a id="You-want-to-optimize-a-complicated-Julia-function-1"></a><a class="docs-heading-anchor-permalink" href="#You-want-to-optimize-a-complicated-Julia-function" title="Permalink"></a></h3><p>Packages in Julia compose well. It&#39;s common for people to pick two unrelated packages and use them in conjunction to create novel behavior. JuMP isn&#39;t one of those packages.</p><p>If you want to optimize an ordinary differential equation from <a href="https://github.com/SciML/DifferentialEquations.jl">DifferentialEquations.jl</a> or tune a neural network from <a href="https://github.com/FluxML/Flux.jl">Flux.jl</a>, consider using other packages such as:</p><ul><li><a href="https://github.com/JuliaNLSolvers/Optim.jl">Optim.jl</a></li><li><a href="https://github.com/SciML/Optimization.jl">Optimization.jl</a></li><li><a href="https://github.com/JuliaSmoothOptimizers/NLPModels.jl">NLPModels.jl</a></li><li><a href="https://github.com/JuliaNonconvex/Nonconvex.jl">Nonconvex.jl</a></li></ul><h3 id="Black-box,-derivative-free,-or-unconstrained-optimization"><a class="docs-heading-anchor" href="#Black-box,-derivative-free,-or-unconstrained-optimization">Black-box, derivative free, or unconstrained optimization</a><a id="Black-box,-derivative-free,-or-unconstrained-optimization-1"></a><a class="docs-heading-anchor-permalink" href="#Black-box,-derivative-free,-or-unconstrained-optimization" title="Permalink"></a></h3><p>JuMP does support nonlinear programs with constraints and objectives containing user-defined operators. However, the functions must be automatically differentiable, or need to provide explicit derivatives. (See <a href="../manual/nonlinear/#jump_user_defined_operators">User-defined operators</a> for more information.)</p><p>If your function is a black-box that is non-differentiable (for example, it is the output of a simulation written in C++), JuMP is not the right tool for the job. This also applies if you want to use a derivative free method.</p><p>Even if your problem is differentiable, if it is unconstrained there is limited benefit (and downsides in the form of more overhead) to using JuMP over tools which are only concerned with function minimization.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/JuliaNLSolvers/Optim.jl">Optim.jl</a></li><li><a href="https://github.com/SciML/Optimization.jl">Optimization.jl</a></li><li><a href="https://github.com/JuliaOpt/NLopt.jl">NLopt.jl</a></li></ul><h3 id="Optimal-control-problems"><a class="docs-heading-anchor" href="#Optimal-control-problems">Optimal control problems</a><a id="Optimal-control-problems-1"></a><a class="docs-heading-anchor-permalink" href="#Optimal-control-problems" title="Permalink"></a></h3><p>JuMP supports formulating optimal control problems as large nonlinear programs (see, for example, <a href="../tutorials/nonlinear/space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory">Optimal control for a Space Shuttle reentry trajectory</a>). However, the nonlinear interface has a number of limitations (for example, the need to write out the dynamics in algebraic form) that mean JuMP might not be the right tool for the job.</p><p>Alternatives to consider are:</p><ul><li><a href="https://web.casadi.org">CasADi [MATLAB/Python]</a>, <a href="https://github.com/ichatzinikolaidis/CasADi.jl">CasADi.jl</a></li><li><a href="https://github.com/infiniteopt/InfiniteOpt.jl">InfiniteOpt.jl</a></li><li><a href="https://pyomo.readthedocs.io/en/stable/modeling_extensions/dae.html">pyomo.DAE [Python]</a></li></ul><h3 id="Disciplined-convex-programming"><a class="docs-heading-anchor" href="#Disciplined-convex-programming">Disciplined convex programming</a><a id="Disciplined-convex-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Disciplined-convex-programming" title="Permalink"></a></h3><p>JuMP does not support <a href="https://dcp.stanford.edu">disciplined convex programming (DCP)</a>.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/jump-dev/Convex.jl">Convex.jl</a></li><li><a href="https://github.com/cvxpy/cvxpy">CVXPY [Python]</a></li><li><a href="https://yalmip.github.io">YALMIP [MATLAB]</a></li></ul><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p><code>Convex.jl</code> is also built on MathOptInterface, and shares the same set of underlying solvers. However, you input problems differently, and Convex.jl checks that the problem is DCP.</p></div></div><h3 id="Stochastic-programming"><a class="docs-heading-anchor" href="#Stochastic-programming">Stochastic programming</a><a id="Stochastic-programming-1"></a><a class="docs-heading-anchor-permalink" href="#Stochastic-programming" title="Permalink"></a></h3><p>JuMP requires deterministic input data.</p><p>If you have stochastic input data, consider using a JuMP extension such as:</p><ul><li><a href="https://github.com/infiniteopt/InfiniteOpt.jl">InfiniteOpt.jl</a></li><li><a href="https://github.com/martinbiel/StochasticPrograms.jl">StochasticPrograms.jl</a></li><li><a href="https://github.com/odow/SDDP.jl">SDDP.jl</a></li></ul><h3 id="Polyhedral-computations"><a class="docs-heading-anchor" href="#Polyhedral-computations">Polyhedral computations</a><a id="Polyhedral-computations-1"></a><a class="docs-heading-anchor-permalink" href="#Polyhedral-computations" title="Permalink"></a></h3><p>JuMP does not provide tools for working with the polyhedron formed by the set of linear constraints.</p><p>Alternatives to consider are:</p><ul><li><a href="https://github.com/JuliaPolyhedra/Polyhedra.jl">Polyhedra.jl</a> (See the <a href="https://juliapolyhedra.github.io/Polyhedra.jl/v0.7.6/optimization/#Creating-a-polyhedron-from-the-feasible-set-of-a-JuMP-model">documentation</a> to create a polyhedron from a JuMP model.)</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../">« Introduction</a><a class="docs-footer-nextpage" href="../installation/">Installation Guide »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/algorithms/benders_decomposition/index.html b/previews/PR3562/tutorials/algorithms/benders_decomposition/index.html
index d9fbd72a52e..040a8b9a0d7 100644
--- a/previews/PR3562/tutorials/algorithms/benders_decomposition/index.html
+++ b/previews/PR3562/tutorials/algorithms/benders_decomposition/index.html
@@ -175,4 +175,4 @@
  1.0</code></pre><pre><code class="language-julia hljs">optimal_ret = solve_subproblem(subproblem, x_optimal)
 y_optimal = optimal_ret.y</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
  0.0
- 0.0</code></pre><p>For larger problems, the benefit of re-using the same subproblem and not needing to multiply the duals by <code>A_1</code> in the cut coefficient usually outweights the cost of needing a copy of the <code>x</code> variables in the subproblem.</p><p>As a secondary benefit, because we no longer need an explicit representation of <code>A_1</code> in the cut, we can build the <code>model</code> and <code>subproblem</code> formulations using arbitrary JuMP syntax; they do not need to be in matrix form.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../conic/quantum_discrimination/">« Quantum state discrimination</a><a class="docs-footer-nextpage" href="../cutting_stock_column_generation/">Column generation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 0.0</code></pre><p>For larger problems, the benefit of re-using the same subproblem and not needing to multiply the duals by <code>A_1</code> in the cut coefficient usually outweights the cost of needing a copy of the <code>x</code> variables in the subproblem.</p><p>As a secondary benefit, because we no longer need an explicit representation of <code>A_1</code> in the cut, we can build the <code>model</code> and <code>subproblem</code> formulations using arbitrary JuMP syntax; they do not need to be in matrix form.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../conic/quantum_discrimination/">« Quantum state discrimination</a><a class="docs-footer-nextpage" href="../cutting_stock_column_generation/">Column generation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/ccf7f2c9.svg b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/74fb9305.svg
similarity index 69%
rename from previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/ccf7f2c9.svg
rename to previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/74fb9305.svg
index 7357eda0668..2cc9224db49 100644
--- a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/ccf7f2c9.svg
+++ b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/74fb9305.svg
@@ -1,516 +1,516 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip470">
+  <clipPath id="clip290">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip470)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip290)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip471">
+  <clipPath id="clip291">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip470)" d="M232.389 1423.18 L2352.76 1423.18 L2352.76 47.2441 L232.389 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip290)" d="M232.389 1423.18 L2352.76 1423.18 L2352.76 47.2441 L232.389 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip472">
+  <clipPath id="clip292">
     <rect x="232" y="47" width="2121" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="821.38,1423.18 821.38,47.2441 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.37,1423.18 1410.37,47.2441 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1999.36,1423.18 1999.36,47.2441 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,1404.28 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="821.38,1423.18 821.38,1404.28 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.37,1423.18 1410.37,1404.28 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.36,1423.18 1999.36,1404.28 "/>
-<path clip-path="url(#clip470)" d="M232.389 1454.1 Q228.778 1454.1 226.95 1457.66 Q225.144 1461.2 225.144 1468.33 Q225.144 1475.44 226.95 1479.01 Q228.778 1482.55 232.389 1482.55 Q236.024 1482.55 237.829 1479.01 Q239.658 1475.44 239.658 1468.33 Q239.658 1461.2 237.829 1457.66 Q236.024 1454.1 232.389 1454.1 M232.389 1450.39 Q238.2 1450.39 241.255 1455 Q244.334 1459.58 244.334 1468.33 Q244.334 1477.06 241.255 1481.67 Q238.2 1486.25 232.389 1486.25 Q226.579 1486.25 223.501 1481.67 Q220.445 1477.06 220.445 1468.33 Q220.445 1459.58 223.501 1455 Q226.579 1450.39 232.389 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M796.068 1481.64 L803.707 1481.64 L803.707 1455.28 L795.396 1456.95 L795.396 1452.69 L803.66 1451.02 L808.336 1451.02 L808.336 1481.64 L815.975 1481.64 L815.975 1485.58 L796.068 1485.58 L796.068 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M835.419 1454.1 Q831.808 1454.1 829.98 1457.66 Q828.174 1461.2 828.174 1468.33 Q828.174 1475.44 829.98 1479.01 Q831.808 1482.55 835.419 1482.55 Q839.054 1482.55 840.859 1479.01 Q842.688 1475.44 842.688 1468.33 Q842.688 1461.2 840.859 1457.66 Q839.054 1454.1 835.419 1454.1 M835.419 1450.39 Q841.229 1450.39 844.285 1455 Q847.364 1459.58 847.364 1468.33 Q847.364 1477.06 844.285 1481.67 Q841.229 1486.25 835.419 1486.25 Q829.609 1486.25 826.53 1481.67 Q823.475 1477.06 823.475 1468.33 Q823.475 1459.58 826.53 1455 Q829.609 1450.39 835.419 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1389.14 1481.64 L1405.46 1481.64 L1405.46 1485.58 L1383.52 1485.58 L1383.52 1481.64 Q1386.18 1478.89 1390.76 1474.26 Q1395.37 1469.61 1396.55 1468.27 Q1398.8 1465.74 1399.68 1464.01 Q1400.58 1462.25 1400.58 1460.56 Q1400.58 1457.8 1398.63 1456.07 Q1396.71 1454.33 1393.61 1454.33 Q1391.41 1454.33 1388.96 1455.09 Q1386.53 1455.86 1383.75 1457.41 L1383.75 1452.69 Q1386.57 1451.55 1389.03 1450.97 Q1391.48 1450.39 1393.52 1450.39 Q1398.89 1450.39 1402.08 1453.08 Q1405.28 1455.77 1405.28 1460.26 Q1405.28 1462.39 1404.47 1464.31 Q1403.68 1466.2 1401.57 1468.8 Q1401 1469.47 1397.89 1472.69 Q1394.79 1475.88 1389.14 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1425.28 1454.1 Q1421.67 1454.1 1419.84 1457.66 Q1418.03 1461.2 1418.03 1468.33 Q1418.03 1475.44 1419.84 1479.01 Q1421.67 1482.55 1425.28 1482.55 Q1428.91 1482.55 1430.72 1479.01 Q1432.55 1475.44 1432.55 1468.33 Q1432.55 1461.2 1430.72 1457.66 Q1428.91 1454.1 1425.28 1454.1 M1425.28 1450.39 Q1431.09 1450.39 1434.14 1455 Q1437.22 1459.58 1437.22 1468.33 Q1437.22 1477.06 1434.14 1481.67 Q1431.09 1486.25 1425.28 1486.25 Q1419.47 1486.25 1416.39 1481.67 Q1413.33 1477.06 1413.33 1468.33 Q1413.33 1459.58 1416.39 1455 Q1419.47 1450.39 1425.28 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1988.2 1466.95 Q1991.56 1467.66 1993.44 1469.93 Q1995.33 1472.2 1995.33 1475.53 Q1995.33 1480.65 1991.82 1483.45 Q1988.3 1486.25 1981.82 1486.25 Q1979.64 1486.25 1977.32 1485.81 Q1975.03 1485.39 1972.58 1484.54 L1972.58 1480.02 Q1974.52 1481.16 1976.84 1481.74 Q1979.15 1482.32 1981.68 1482.32 Q1986.07 1482.32 1988.37 1480.58 Q1990.68 1478.84 1990.68 1475.53 Q1990.68 1472.48 1988.53 1470.77 Q1986.4 1469.03 1982.58 1469.03 L1978.55 1469.03 L1978.55 1465.19 L1982.76 1465.19 Q1986.21 1465.19 1988.04 1463.82 Q1989.87 1462.43 1989.87 1459.84 Q1989.87 1457.18 1987.97 1455.77 Q1986.1 1454.33 1982.58 1454.33 Q1980.66 1454.33 1978.46 1454.75 Q1976.26 1455.16 1973.62 1456.04 L1973.62 1451.88 Q1976.28 1451.14 1978.6 1450.77 Q1980.94 1450.39 1983 1450.39 Q1988.32 1450.39 1991.42 1452.83 Q1994.52 1455.23 1994.52 1459.35 Q1994.52 1462.22 1992.88 1464.21 Q1991.24 1466.18 1988.2 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2014.2 1454.1 Q2010.59 1454.1 2008.76 1457.66 Q2006.95 1461.2 2006.95 1468.33 Q2006.95 1475.44 2008.76 1479.01 Q2010.59 1482.55 2014.2 1482.55 Q2017.83 1482.55 2019.64 1479.01 Q2021.47 1475.44 2021.47 1468.33 Q2021.47 1461.2 2019.64 1457.66 Q2017.83 1454.1 2014.2 1454.1 M2014.2 1450.39 Q2020.01 1450.39 2023.07 1455 Q2026.14 1459.58 2026.14 1468.33 Q2026.14 1477.06 2023.07 1481.67 Q2020.01 1486.25 2014.2 1486.25 Q2008.39 1486.25 2005.31 1481.67 Q2002.25 1477.06 2002.25 1468.33 Q2002.25 1459.58 2005.31 1455 Q2008.39 1450.39 2014.2 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1187.71 1525.81 L1187.71 1543.66 L1195.8 1543.66 Q1200.29 1543.66 1202.74 1541.34 Q1205.19 1539.02 1205.19 1534.72 Q1205.19 1530.45 1202.74 1528.13 Q1200.29 1525.81 1195.8 1525.81 L1187.71 1525.81 M1181.28 1520.52 L1195.8 1520.52 Q1203.79 1520.52 1207.86 1524.15 Q1211.97 1527.75 1211.97 1534.72 Q1211.97 1541.75 1207.86 1545.35 Q1203.79 1548.95 1195.8 1548.95 L1187.71 1548.95 L1187.71 1568.04 L1181.28 1568.04 L1181.28 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1233.61 1550.12 Q1226.51 1550.12 1223.78 1551.75 Q1221.04 1553.37 1221.04 1557.29 Q1221.04 1560.4 1223.08 1562.25 Q1225.14 1564.07 1228.68 1564.07 Q1233.55 1564.07 1236.47 1560.63 Q1239.43 1557.16 1239.43 1551.43 L1239.43 1550.12 L1233.61 1550.12 M1245.29 1547.71 L1245.29 1568.04 L1239.43 1568.04 L1239.43 1562.63 Q1237.43 1565.88 1234.44 1567.44 Q1231.45 1568.97 1227.12 1568.97 Q1221.64 1568.97 1218.4 1565.91 Q1215.18 1562.82 1215.18 1557.67 Q1215.18 1551.65 1219.19 1548.6 Q1223.23 1545.54 1231.22 1545.54 L1239.43 1545.54 L1239.43 1544.97 Q1239.43 1540.93 1236.76 1538.73 Q1234.12 1536.5 1229.31 1536.5 Q1226.26 1536.5 1223.36 1537.23 Q1220.47 1537.97 1217.79 1539.43 L1217.79 1534.02 Q1221.01 1532.78 1224.03 1532.17 Q1227.05 1531.54 1229.92 1531.54 Q1237.65 1531.54 1241.47 1535.55 Q1245.29 1539.56 1245.29 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1263.15 1522.27 L1263.15 1532.4 L1275.21 1532.4 L1275.21 1536.95 L1263.15 1536.95 L1263.15 1556.3 Q1263.15 1560.66 1264.32 1561.9 Q1265.53 1563.14 1269.19 1563.14 L1275.21 1563.14 L1275.21 1568.04 L1269.19 1568.04 Q1262.42 1568.04 1259.84 1565.53 Q1257.26 1562.98 1257.26 1556.3 L1257.26 1536.95 L1252.96 1536.95 L1252.96 1532.4 L1257.26 1532.4 L1257.26 1522.27 L1263.15 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1288.71 1522.27 L1288.71 1532.4 L1300.77 1532.4 L1300.77 1536.95 L1288.71 1536.95 L1288.71 1556.3 Q1288.71 1560.66 1289.88 1561.9 Q1291.09 1563.14 1294.75 1563.14 L1300.77 1563.14 L1300.77 1568.04 L1294.75 1568.04 Q1287.97 1568.04 1285.4 1565.53 Q1282.82 1562.98 1282.82 1556.3 L1282.82 1536.95 L1278.52 1536.95 L1278.52 1532.4 L1282.82 1532.4 L1282.82 1522.27 L1288.71 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1338.96 1548.76 L1338.96 1551.62 L1312.04 1551.62 Q1312.42 1557.67 1315.66 1560.85 Q1318.94 1564 1324.77 1564 Q1328.14 1564 1331.29 1563.17 Q1334.47 1562.35 1337.59 1560.69 L1337.59 1566.23 Q1334.44 1567.57 1331.13 1568.27 Q1327.82 1568.97 1324.42 1568.97 Q1315.89 1568.97 1310.89 1564 Q1305.92 1559.04 1305.92 1550.57 Q1305.92 1541.82 1310.64 1536.69 Q1315.38 1531.54 1323.4 1531.54 Q1330.59 1531.54 1334.76 1536.18 Q1338.96 1540.8 1338.96 1548.76 M1333.11 1547.04 Q1333.04 1542.23 1330.4 1539.37 Q1327.79 1536.5 1323.46 1536.5 Q1318.56 1536.5 1315.6 1539.27 Q1312.67 1542.04 1312.23 1547.07 L1333.11 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1369.23 1537.87 Q1368.24 1537.3 1367.07 1537.04 Q1365.92 1536.76 1364.52 1536.76 Q1359.56 1536.76 1356.88 1540 Q1354.24 1543.22 1354.24 1549.27 L1354.24 1568.04 L1348.35 1568.04 L1348.35 1532.4 L1354.24 1532.4 L1354.24 1537.93 Q1356.09 1534.69 1359.05 1533.13 Q1362.01 1531.54 1366.24 1531.54 Q1366.84 1531.54 1367.58 1531.63 Q1368.31 1531.7 1369.2 1531.85 L1369.23 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1403.86 1546.53 L1403.86 1568.04 L1398 1568.04 L1398 1546.72 Q1398 1541.66 1396.03 1539.14 Q1394.06 1536.63 1390.11 1536.63 Q1385.37 1536.63 1382.63 1539.65 Q1379.89 1542.68 1379.89 1547.9 L1379.89 1568.04 L1374.01 1568.04 L1374.01 1532.4 L1379.89 1532.4 L1379.89 1537.93 Q1381.99 1534.72 1384.83 1533.13 Q1387.69 1531.54 1391.42 1531.54 Q1397.56 1531.54 1400.71 1535.36 Q1403.86 1539.14 1403.86 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1147.99 2352.76,1147.99 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,872.806 2352.76,872.806 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,597.618 2352.76,597.618 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,322.431 2352.76,322.431 "/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,47.2441 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 251.287,1423.18 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1147.99 251.287,1147.99 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,872.806 251.287,872.806 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,597.618 251.287,597.618 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,322.431 251.287,322.431 "/>
-<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,47.2441 251.287,47.2441 "/>
-<path clip-path="url(#clip470)" d="M184.445 1408.98 Q180.834 1408.98 179.005 1412.54 Q177.2 1416.08 177.2 1423.21 Q177.2 1430.32 179.005 1433.89 Q180.834 1437.43 184.445 1437.43 Q188.079 1437.43 189.885 1433.89 Q191.713 1430.32 191.713 1423.21 Q191.713 1416.08 189.885 1412.54 Q188.079 1408.98 184.445 1408.98 M184.445 1405.27 Q190.255 1405.27 193.311 1409.88 Q196.389 1414.46 196.389 1423.21 Q196.389 1431.94 193.311 1436.55 Q190.255 1441.13 184.445 1441.13 Q178.635 1441.13 175.556 1436.55 Q172.501 1431.94 172.501 1423.21 Q172.501 1414.46 175.556 1409.88 Q178.635 1405.27 184.445 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M148.311 1161.34 L164.63 1161.34 L164.63 1165.27 L142.686 1165.27 L142.686 1161.34 Q145.348 1158.58 149.931 1153.95 Q154.538 1149.3 155.718 1147.96 Q157.964 1145.43 158.843 1143.7 Q159.746 1141.94 159.746 1140.25 Q159.746 1137.5 157.802 1135.76 Q155.88 1134.02 152.778 1134.02 Q150.579 1134.02 148.126 1134.79 Q145.695 1135.55 142.917 1137.1 L142.917 1132.38 Q145.741 1131.25 148.195 1130.67 Q150.649 1130.09 152.686 1130.09 Q158.056 1130.09 161.251 1132.77 Q164.445 1135.46 164.445 1139.95 Q164.445 1142.08 163.635 1144 Q162.848 1145.9 160.741 1148.49 Q160.163 1149.16 157.061 1152.38 Q153.959 1155.57 148.311 1161.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 1133.79 Q180.834 1133.79 179.005 1137.36 Q177.2 1140.9 177.2 1148.03 Q177.2 1155.13 179.005 1158.7 Q180.834 1162.24 184.445 1162.24 Q188.079 1162.24 189.885 1158.7 Q191.713 1155.13 191.713 1148.03 Q191.713 1140.9 189.885 1137.36 Q188.079 1133.79 184.445 1133.79 M184.445 1130.09 Q190.255 1130.09 193.311 1134.69 Q196.389 1139.28 196.389 1148.03 Q196.389 1156.75 193.311 1161.36 Q190.255 1165.94 184.445 1165.94 Q178.635 1165.94 175.556 1161.36 Q172.501 1156.75 172.501 1148.03 Q172.501 1139.28 175.556 1134.69 Q178.635 1130.09 184.445 1130.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M157.13 859.6 L145.325 878.049 L157.13 878.049 L157.13 859.6 M155.903 855.526 L161.783 855.526 L161.783 878.049 L166.714 878.049 L166.714 881.937 L161.783 881.937 L161.783 890.086 L157.13 890.086 L157.13 881.937 L141.529 881.937 L141.529 877.424 L155.903 855.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 858.604 Q180.834 858.604 179.005 862.169 Q177.2 865.711 177.2 872.84 Q177.2 879.947 179.005 883.512 Q180.834 887.053 184.445 887.053 Q188.079 887.053 189.885 883.512 Q191.713 879.947 191.713 872.84 Q191.713 865.711 189.885 862.169 Q188.079 858.604 184.445 858.604 M184.445 854.901 Q190.255 854.901 193.311 859.507 Q196.389 864.09 196.389 872.84 Q196.389 881.567 193.311 886.174 Q190.255 890.757 184.445 890.757 Q178.635 890.757 175.556 886.174 Q172.501 881.567 172.501 872.84 Q172.501 864.09 175.556 859.507 Q178.635 854.901 184.445 854.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M154.862 595.755 Q151.714 595.755 149.862 597.908 Q148.033 600.061 148.033 603.81 Q148.033 607.537 149.862 609.713 Q151.714 611.866 154.862 611.866 Q158.01 611.866 159.839 609.713 Q161.69 607.537 161.69 603.81 Q161.69 600.061 159.839 597.908 Q158.01 595.755 154.862 595.755 M164.144 581.102 L164.144 585.362 Q162.385 584.528 160.579 584.088 Q158.797 583.649 157.038 583.649 Q152.408 583.649 149.954 586.774 Q147.524 589.899 147.177 596.218 Q148.542 594.204 150.603 593.139 Q152.663 592.051 155.14 592.051 Q160.348 592.051 163.357 595.223 Q166.39 598.371 166.39 603.81 Q166.39 609.135 163.241 612.352 Q160.093 615.57 154.862 615.57 Q148.866 615.57 145.695 610.986 Q142.524 606.38 142.524 597.653 Q142.524 589.459 146.413 584.598 Q150.302 579.713 156.853 579.713 Q158.612 579.713 160.394 580.061 Q162.2 580.408 164.144 581.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 583.417 Q180.834 583.417 179.005 586.982 Q177.2 590.524 177.2 597.653 Q177.2 604.76 179.005 608.324 Q180.834 611.866 184.445 611.866 Q188.079 611.866 189.885 608.324 Q191.713 604.76 191.713 597.653 Q191.713 590.524 189.885 586.982 Q188.079 583.417 184.445 583.417 M184.445 579.713 Q190.255 579.713 193.311 584.32 Q196.389 588.903 196.389 597.653 Q196.389 606.38 193.311 610.986 Q190.255 615.57 184.445 615.57 Q178.635 615.57 175.556 610.986 Q172.501 606.38 172.501 597.653 Q172.501 588.903 175.556 584.32 Q178.635 579.713 184.445 579.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M154.283 323.299 Q150.95 323.299 149.028 325.082 Q147.13 326.864 147.13 329.989 Q147.13 333.114 149.028 334.896 Q150.95 336.679 154.283 336.679 Q157.616 336.679 159.538 334.896 Q161.459 333.091 161.459 329.989 Q161.459 326.864 159.538 325.082 Q157.64 323.299 154.283 323.299 M149.607 321.309 Q146.598 320.568 144.908 318.508 Q143.241 316.447 143.241 313.485 Q143.241 309.341 146.181 306.934 Q149.144 304.526 154.283 304.526 Q159.445 304.526 162.385 306.934 Q165.325 309.341 165.325 313.485 Q165.325 316.447 163.635 318.508 Q161.968 320.568 158.982 321.309 Q162.362 322.096 164.237 324.387 Q166.135 326.679 166.135 329.989 Q166.135 335.012 163.056 337.697 Q160.001 340.383 154.283 340.383 Q148.566 340.383 145.487 337.697 Q142.431 335.012 142.431 329.989 Q142.431 326.679 144.329 324.387 Q146.228 322.096 149.607 321.309 M147.894 313.924 Q147.894 316.61 149.561 318.114 Q151.251 319.619 154.283 319.619 Q157.292 319.619 158.982 318.114 Q160.695 316.61 160.695 313.924 Q160.695 311.239 158.982 309.735 Q157.292 308.23 154.283 308.23 Q151.251 308.23 149.561 309.735 Q147.894 311.239 147.894 313.924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 308.23 Q180.834 308.23 179.005 311.795 Q177.2 315.336 177.2 322.466 Q177.2 329.572 179.005 333.137 Q180.834 336.679 184.445 336.679 Q188.079 336.679 189.885 333.137 Q191.713 329.572 191.713 322.466 Q191.713 315.336 189.885 311.795 Q188.079 308.23 184.445 308.23 M184.445 304.526 Q190.255 304.526 193.311 309.133 Q196.389 313.716 196.389 322.466 Q196.389 331.193 193.311 335.799 Q190.255 340.383 184.445 340.383 Q178.635 340.383 175.556 335.799 Q172.501 331.193 172.501 322.466 Q172.501 313.716 175.556 309.133 Q178.635 304.526 184.445 304.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M114.931 60.5889 L122.57 60.5889 L122.57 34.2233 L114.26 35.89 L114.26 31.6308 L122.524 29.9641 L127.2 29.9641 L127.2 60.5889 L134.839 60.5889 L134.839 64.5241 L114.931 64.5241 L114.931 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M154.283 33.0428 Q150.672 33.0428 148.843 36.6076 Q147.038 40.1492 147.038 47.2788 Q147.038 54.3853 148.843 57.9501 Q150.672 61.4917 154.283 61.4917 Q157.917 61.4917 159.723 57.9501 Q161.552 54.3853 161.552 47.2788 Q161.552 40.1492 159.723 36.6076 Q157.917 33.0428 154.283 33.0428 M154.283 29.3391 Q160.093 29.3391 163.149 33.9456 Q166.227 38.5289 166.227 47.2788 Q166.227 56.0056 163.149 60.6121 Q160.093 65.1954 154.283 65.1954 Q148.473 65.1954 145.394 60.6121 Q142.339 56.0056 142.339 47.2788 Q142.339 38.5289 145.394 33.9456 Q148.473 29.3391 154.283 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 33.0428 Q180.834 33.0428 179.005 36.6076 Q177.2 40.1492 177.2 47.2788 Q177.2 54.3853 179.005 57.9501 Q180.834 61.4917 184.445 61.4917 Q188.079 61.4917 189.885 57.9501 Q191.713 54.3853 191.713 47.2788 Q191.713 40.1492 189.885 36.6076 Q188.079 33.0428 184.445 33.0428 M184.445 29.3391 Q190.255 29.3391 193.311 33.9456 Q196.389 38.5289 196.389 47.2788 Q196.389 56.0056 193.311 60.6121 Q190.255 65.1954 184.445 65.1954 Q178.635 65.1954 175.556 60.6121 Q172.501 56.0056 172.501 47.2788 Q172.501 38.5289 175.556 33.9456 Q178.635 29.3391 184.445 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M41.7242 880.191 Q42.4244 878.122 44.7161 876.181 Q47.0077 874.207 51.0181 872.234 L64.0042 865.709 L64.0042 872.616 L51.8138 878.695 Q47.0395 881.05 45.48 883.278 Q43.9204 885.474 43.9204 889.294 L43.9204 896.296 L64.0042 896.296 L64.0042 902.726 L16.4842 902.726 L16.4842 888.212 Q16.4842 880.064 19.8898 876.053 Q23.2955 872.043 30.1704 872.043 Q34.6582 872.043 37.6183 874.144 Q40.5784 876.212 41.7242 880.191 M21.7677 896.296 L38.6368 896.296 L38.6368 888.212 Q38.6368 883.565 36.5043 881.209 Q34.34 878.822 30.1704 878.822 Q26.0009 878.822 23.9002 881.209 Q21.7677 883.565 21.7677 888.212 L21.7677 896.296 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M32.4621 846.803 Q32.4621 851.513 36.1542 854.251 Q39.8145 856.988 46.212 856.988 Q52.6095 856.988 56.3017 854.283 Q59.9619 851.545 59.9619 846.803 Q59.9619 842.124 56.2698 839.387 Q52.5777 836.65 46.212 836.65 Q39.8781 836.65 36.186 839.387 Q32.4621 842.124 32.4621 846.803 M27.4968 846.803 Q27.4968 839.164 32.4621 834.803 Q37.4273 830.443 46.212 830.443 Q54.9649 830.443 59.9619 834.803 Q64.9272 839.164 64.9272 846.803 Q64.9272 854.473 59.9619 858.834 Q54.9649 863.163 46.212 863.163 Q37.4273 863.163 32.4621 858.834 Q27.4968 854.473 27.4968 846.803 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M14.479 820.735 L14.479 814.879 L64.0042 814.879 L64.0042 820.735 L14.479 820.735 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M14.479 802.625 L14.479 796.768 L64.0042 796.768 L64.0042 802.625 L14.479 802.625 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M14.479 763.794 L14.479 757.938 L64.0042 757.938 L64.0042 763.794 L14.479 763.794 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M44.7161 715.192 L47.5806 715.192 L47.5806 742.119 Q53.6281 741.737 56.8109 738.49 Q59.9619 735.212 59.9619 729.387 Q59.9619 726.014 59.1344 722.862 Q58.3069 719.68 56.6518 716.56 L62.1899 716.56 Q63.5267 719.711 64.227 723.022 Q64.9272 726.332 64.9272 729.737 Q64.9272 738.268 59.9619 743.265 Q54.9967 748.23 46.5303 748.23 Q37.7774 748.23 32.6531 743.519 Q27.4968 738.777 27.4968 730.756 Q27.4968 723.563 32.1438 719.393 Q36.7589 715.192 44.7161 715.192 M42.9973 721.048 Q38.1912 721.112 35.3266 723.754 Q32.4621 726.364 32.4621 730.692 Q32.4621 735.594 35.2312 738.554 Q38.0002 741.482 43.0292 741.928 L42.9973 721.048 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M42.4881 675.947 L64.0042 675.947 L64.0042 681.804 L42.679 681.804 Q37.6183 681.804 35.1038 683.777 Q32.5894 685.75 32.5894 689.697 Q32.5894 694.44 35.6131 697.177 Q38.6368 699.914 43.8567 699.914 L64.0042 699.914 L64.0042 705.802 L28.3562 705.802 L28.3562 699.914 L33.8944 699.914 Q30.6797 697.813 29.0883 694.981 Q27.4968 692.116 27.4968 688.392 Q27.4968 682.249 31.3163 679.098 Q35.1038 675.947 42.4881 675.947 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M45.7664 640.809 Q39.4007 640.809 35.8996 643.45 Q32.3984 646.06 32.3984 650.803 Q32.3984 655.513 35.8996 658.155 Q39.4007 660.765 45.7664 660.765 Q52.1003 660.765 55.6014 658.155 Q59.1026 655.513 59.1026 650.803 Q59.1026 646.06 55.6014 643.45 Q52.1003 640.809 45.7664 640.809 M59.58 634.952 Q68.683 634.952 73.1071 638.994 Q77.5631 643.037 77.5631 651.376 Q77.5631 654.463 77.0857 657.2 Q76.6401 659.937 75.6852 662.516 L69.9879 662.516 Q71.3884 659.937 72.0568 657.423 Q72.7252 654.909 72.7252 652.299 Q72.7252 646.538 69.7015 643.673 Q66.7096 640.809 60.6303 640.809 L57.7339 640.809 Q60.885 642.623 62.4446 645.456 Q64.0042 648.288 64.0042 652.235 Q64.0042 658.792 59.0071 662.802 Q54.01 666.812 45.7664 666.812 Q37.491 666.812 32.4939 662.802 Q27.4968 658.792 27.4968 652.235 Q27.4968 648.288 29.0564 645.456 Q30.616 642.623 33.7671 640.809 L28.3562 640.809 L28.3562 634.952 L59.58 634.952 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M18.2347 617.096 L28.3562 617.096 L28.3562 605.033 L32.9077 605.033 L32.9077 617.096 L52.2594 617.096 Q56.6199 617.096 57.8613 615.919 Q59.1026 614.709 59.1026 611.049 L59.1026 605.033 L64.0042 605.033 L64.0042 611.049 Q64.0042 617.828 61.4897 620.406 Q58.9434 622.985 52.2594 622.985 L32.9077 622.985 L32.9077 627.281 L28.3562 627.281 L28.3562 622.985 L18.2347 622.985 L18.2347 617.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M42.4881 567.698 L64.0042 567.698 L64.0042 573.555 L42.679 573.555 Q37.6183 573.555 35.1038 575.528 Q32.5894 577.502 32.5894 581.448 Q32.5894 586.191 35.6131 588.928 Q38.6368 591.665 43.8567 591.665 L64.0042 591.665 L64.0042 597.554 L14.479 597.554 L14.479 591.665 L33.8944 591.665 Q30.6797 589.565 29.0883 586.732 Q27.4968 583.867 27.4968 580.143 Q27.4968 574 31.3163 570.849 Q35.1038 567.698 42.4881 567.698 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip472)" d="M273.619 391.228 L273.619 1423.18 L308.958 1423.18 L308.958 391.228 L273.619 391.228 L273.619 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="273.619,391.228 273.619,1423.18 308.958,1423.18 308.958,391.228 273.619,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="291.288" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M332.518 391.228 L332.518 1423.18 L367.857 1423.18 L367.857 391.228 L332.518 391.228 L332.518 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="332.518,391.228 332.518,1423.18 367.857,1423.18 367.857,391.228 332.518,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="350.187" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M391.417 391.228 L391.417 1423.18 L426.756 1423.18 L426.756 391.228 L391.417 391.228 L391.417 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="391.417,391.228 391.417,1423.18 426.756,1423.18 426.756,391.228 391.417,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="409.087" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M450.316 391.228 L450.316 1423.18 L485.655 1423.18 L485.655 391.228 L450.316 391.228 L450.316 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="450.316,391.228 450.316,1423.18 485.655,1423.18 485.655,391.228 450.316,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="467.986" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M509.215 391.228 L509.215 1423.18 L544.554 1423.18 L544.554 391.228 L509.215 391.228 L509.215 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="509.215,391.228 509.215,1423.18 544.554,1423.18 544.554,391.228 509.215,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="526.885" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M568.114 682.926 L568.114 1423.18 L603.453 1423.18 L603.453 682.926 L568.114 682.926 L568.114 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="568.114,682.926 568.114,1423.18 603.453,1423.18 603.453,682.926 568.114,682.926 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="585.784" cy="682.926" r="2"/>
-<path clip-path="url(#clip472)" d="M627.013 693.934 L627.013 1423.18 L662.353 1423.18 L662.353 693.934 L627.013 693.934 L627.013 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="627.013,693.934 627.013,1423.18 662.353,1423.18 662.353,693.934 627.013,693.934 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="644.683" cy="693.934" r="2"/>
-<path clip-path="url(#clip472)" d="M685.912 721.453 L685.912 1423.18 L721.252 1423.18 L721.252 721.453 L685.912 721.453 L685.912 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="685.912,721.453 685.912,1423.18 721.252,1423.18 721.252,721.453 685.912,721.453 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="703.582" cy="721.453" r="2"/>
-<path clip-path="url(#clip472)" d="M744.811 732.46 L744.811 1423.18 L780.151 1423.18 L780.151 732.46 L744.811 732.46 L744.811 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="744.811,732.46 744.811,1423.18 780.151,1423.18 780.151,732.46 744.811,732.46 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="762.481" cy="732.46" r="2"/>
-<path clip-path="url(#clip472)" d="M803.71 94.0259 L803.71 1423.18 L839.05 1423.18 L839.05 94.0259 L803.71 94.0259 L803.71 94.0259  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,94.0259 803.71,1423.18 839.05,1423.18 839.05,94.0259 803.71,94.0259 "/>
-<path clip-path="url(#clip472)" d="M803.71 537.077 L803.71 1423.18 L839.05 1423.18 L839.05 537.077 L803.71 537.077 L803.71 537.077  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,537.077 803.71,1423.18 839.05,1423.18 839.05,537.077 803.71,537.077 "/>
-<path clip-path="url(#clip472)" d="M803.71 980.129 L803.71 1423.18 L839.05 1423.18 L839.05 980.129 L803.71 980.129 L803.71 980.129  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,980.129 803.71,1423.18 839.05,1423.18 839.05,980.129 803.71,980.129 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="94.0259" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="537.077" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="980.129" r="2"/>
-<path clip-path="url(#clip472)" d="M862.609 151.815 L862.609 1423.18 L897.949 1423.18 L897.949 151.815 L862.609 151.815 L862.609 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,151.815 862.609,1423.18 897.949,1423.18 897.949,151.815 862.609,151.815 "/>
-<path clip-path="url(#clip472)" d="M862.609 575.603 L862.609 1423.18 L897.949 1423.18 L897.949 575.603 L862.609 575.603 L862.609 575.603  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,575.603 862.609,1423.18 897.949,1423.18 897.949,575.603 862.609,575.603 "/>
-<path clip-path="url(#clip472)" d="M862.609 999.392 L862.609 1423.18 L897.949 1423.18 L897.949 999.392 L862.609 999.392 L862.609 999.392  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,999.392 862.609,1423.18 897.949,1423.18 897.949,999.392 862.609,999.392 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="151.815" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="575.603" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="999.392" r="2"/>
-<path clip-path="url(#clip472)" d="M921.508 193.093 L921.508 1423.18 L956.848 1423.18 L956.848 193.093 L921.508 193.093 L921.508 193.093  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,193.093 921.508,1423.18 956.848,1423.18 956.848,193.093 921.508,193.093 "/>
-<path clip-path="url(#clip472)" d="M921.508 603.122 L921.508 1423.18 L956.848 1423.18 L956.848 603.122 L921.508 603.122 L921.508 603.122  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,603.122 921.508,1423.18 956.848,1423.18 956.848,603.122 921.508,603.122 "/>
-<path clip-path="url(#clip472)" d="M921.508 1013.15 L921.508 1423.18 L956.848 1423.18 L956.848 1013.15 L921.508 1013.15 L921.508 1013.15  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,1013.15 921.508,1423.18 956.848,1423.18 956.848,1013.15 921.508,1013.15 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="193.093" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="603.122" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="1013.15" r="2"/>
-<path clip-path="url(#clip472)" d="M980.408 316.928 L980.408 1423.18 L1015.75 1423.18 L1015.75 316.928 L980.408 316.928 L980.408 316.928  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,316.928 980.408,1423.18 1015.75,1423.18 1015.75,316.928 980.408,316.928 "/>
-<path clip-path="url(#clip472)" d="M980.408 593.491 L980.408 1423.18 L1015.75 1423.18 L1015.75 593.491 L980.408 593.491 L980.408 593.491  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,593.491 980.408,1423.18 1015.75,1423.18 1015.75,593.491 980.408,593.491 "/>
-<path clip-path="url(#clip472)" d="M980.408 870.054 L980.408 1423.18 L1015.75 1423.18 L1015.75 870.054 L980.408 870.054 L980.408 870.054  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,870.054 980.408,1423.18 1015.75,1423.18 1015.75,870.054 980.408,870.054 "/>
-<path clip-path="url(#clip472)" d="M980.408 1146.62 L980.408 1423.18 L1015.75 1423.18 L1015.75 1146.62 L980.408 1146.62 L980.408 1146.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,1146.62 980.408,1423.18 1015.75,1423.18 1015.75,1146.62 980.408,1146.62 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="316.928" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="593.491" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="870.054" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="1146.62" r="2"/>
-<path clip-path="url(#clip472)" d="M1039.31 85.7703 L1039.31 1423.18 L1074.65 1423.18 L1074.65 85.7703 L1039.31 85.7703 L1039.31 85.7703  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,85.7703 1039.31,1423.18 1074.65,1423.18 1074.65,85.7703 1039.31,85.7703 "/>
-<path clip-path="url(#clip472)" d="M1039.31 308.672 L1039.31 1423.18 L1074.65 1423.18 L1074.65 308.672 L1039.31 308.672 L1039.31 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,308.672 1039.31,1423.18 1074.65,1423.18 1074.65,308.672 1039.31,308.672 "/>
-<path clip-path="url(#clip472)" d="M1039.31 531.573 L1039.31 1423.18 L1074.65 1423.18 L1074.65 531.573 L1039.31 531.573 L1039.31 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,531.573 1039.31,1423.18 1074.65,1423.18 1074.65,531.573 1039.31,531.573 "/>
-<path clip-path="url(#clip472)" d="M1039.31 754.475 L1039.31 1423.18 L1074.65 1423.18 L1074.65 754.475 L1039.31 754.475 L1039.31 754.475  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,754.475 1039.31,1423.18 1074.65,1423.18 1074.65,754.475 1039.31,754.475 "/>
-<path clip-path="url(#clip472)" d="M1039.31 977.377 L1039.31 1423.18 L1074.65 1423.18 L1074.65 977.377 L1039.31 977.377 L1039.31 977.377  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,977.377 1039.31,1423.18 1074.65,1423.18 1074.65,977.377 1039.31,977.377 "/>
-<path clip-path="url(#clip472)" d="M1039.31 1200.28 L1039.31 1423.18 L1074.65 1423.18 L1074.65 1200.28 L1039.31 1200.28 L1039.31 1200.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,1200.28 1039.31,1423.18 1074.65,1423.18 1074.65,1200.28 1039.31,1200.28 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="85.7703" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="308.672" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="531.573" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="754.475" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="977.377" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="1200.28" r="2"/>
-<path clip-path="url(#clip472)" d="M1098.21 226.116 L1098.21 1423.18 L1133.55 1423.18 L1133.55 226.116 L1098.21 226.116 L1098.21 226.116  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,226.116 1098.21,1423.18 1133.55,1423.18 1133.55,226.116 1098.21,226.116 "/>
-<path clip-path="url(#clip472)" d="M1098.21 425.626 L1098.21 1423.18 L1133.55 1423.18 L1133.55 425.626 L1098.21 425.626 L1098.21 425.626  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,425.626 1098.21,1423.18 1133.55,1423.18 1133.55,425.626 1098.21,425.626 "/>
-<path clip-path="url(#clip472)" d="M1098.21 625.137 L1098.21 1423.18 L1133.55 1423.18 L1133.55 625.137 L1098.21 625.137 L1098.21 625.137  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,625.137 1098.21,1423.18 1133.55,1423.18 1133.55,625.137 1098.21,625.137 "/>
-<path clip-path="url(#clip472)" d="M1098.21 824.648 L1098.21 1423.18 L1133.55 1423.18 L1133.55 824.648 L1098.21 824.648 L1098.21 824.648  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,824.648 1098.21,1423.18 1133.55,1423.18 1133.55,824.648 1098.21,824.648 "/>
-<path clip-path="url(#clip472)" d="M1098.21 1024.16 L1098.21 1423.18 L1133.55 1423.18 L1133.55 1024.16 L1098.21 1024.16 L1098.21 1024.16  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,1024.16 1098.21,1423.18 1133.55,1423.18 1133.55,1024.16 1098.21,1024.16 "/>
-<path clip-path="url(#clip472)" d="M1098.21 1223.67 L1098.21 1423.18 L1133.55 1423.18 L1133.55 1223.67 L1098.21 1223.67 L1098.21 1223.67  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,1223.67 1098.21,1423.18 1133.55,1423.18 1133.55,1223.67 1098.21,1223.67 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="226.116" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="425.626" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="625.137" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="824.648" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="1024.16" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="1223.67" r="2"/>
-<path clip-path="url(#clip472)" d="M1157.1 61.0035 L1157.1 1423.18 L1192.44 1423.18 L1192.44 61.0035 L1157.1 61.0035 L1157.1 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,61.0035 1157.1,1423.18 1192.44,1423.18 1192.44,61.0035 1157.1,61.0035 "/>
-<path clip-path="url(#clip472)" d="M1157.1 212.356 L1157.1 1423.18 L1192.44 1423.18 L1192.44 212.356 L1157.1 212.356 L1157.1 212.356  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,212.356 1157.1,1423.18 1192.44,1423.18 1192.44,212.356 1157.1,212.356 "/>
-<path clip-path="url(#clip472)" d="M1157.1 363.709 L1157.1 1423.18 L1192.44 1423.18 L1192.44 363.709 L1157.1 363.709 L1157.1 363.709  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,363.709 1157.1,1423.18 1192.44,1423.18 1192.44,363.709 1157.1,363.709 "/>
-<path clip-path="url(#clip472)" d="M1157.1 515.062 L1157.1 1423.18 L1192.44 1423.18 L1192.44 515.062 L1157.1 515.062 L1157.1 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,515.062 1157.1,1423.18 1192.44,1423.18 1192.44,515.062 1157.1,515.062 "/>
-<path clip-path="url(#clip472)" d="M1157.1 666.415 L1157.1 1423.18 L1192.44 1423.18 L1192.44 666.415 L1157.1 666.415 L1157.1 666.415  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,666.415 1157.1,1423.18 1192.44,1423.18 1192.44,666.415 1157.1,666.415 "/>
-<path clip-path="url(#clip472)" d="M1157.1 817.768 L1157.1 1423.18 L1192.44 1423.18 L1192.44 817.768 L1157.1 817.768 L1157.1 817.768  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,817.768 1157.1,1423.18 1192.44,1423.18 1192.44,817.768 1157.1,817.768 "/>
-<path clip-path="url(#clip472)" d="M1157.1 969.121 L1157.1 1423.18 L1192.44 1423.18 L1192.44 969.121 L1157.1 969.121 L1157.1 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,969.121 1157.1,1423.18 1192.44,1423.18 1192.44,969.121 1157.1,969.121 "/>
-<path clip-path="url(#clip472)" d="M1157.1 1120.47 L1157.1 1423.18 L1192.44 1423.18 L1192.44 1120.47 L1157.1 1120.47 L1157.1 1120.47  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,1120.47 1157.1,1423.18 1192.44,1423.18 1192.44,1120.47 1157.1,1120.47 "/>
-<path clip-path="url(#clip472)" d="M1157.1 1271.83 L1157.1 1423.18 L1192.44 1423.18 L1192.44 1271.83 L1157.1 1271.83 L1157.1 1271.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,1271.83 1157.1,1423.18 1192.44,1423.18 1192.44,1271.83 1157.1,1271.83 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="61.0035" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="212.356" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="363.709" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="515.062" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="666.415" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="817.768" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="969.121" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="1120.47" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="1271.83" r="2"/>
-<path clip-path="url(#clip472)" d="M1216 121.545 L1216 1423.18 L1251.34 1423.18 L1251.34 121.545 L1216 121.545 L1216 121.545  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,121.545 1216,1423.18 1251.34,1423.18 1251.34,121.545 1216,121.545 "/>
-<path clip-path="url(#clip472)" d="M1216 239.875 L1216 1423.18 L1251.34 1423.18 L1251.34 239.875 L1216 239.875 L1216 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,239.875 1216,1423.18 1251.34,1423.18 1251.34,239.875 1216,239.875 "/>
-<path clip-path="url(#clip472)" d="M1216 358.206 L1216 1423.18 L1251.34 1423.18 L1251.34 358.206 L1216 358.206 L1216 358.206  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,358.206 1216,1423.18 1251.34,1423.18 1251.34,358.206 1216,358.206 "/>
-<path clip-path="url(#clip472)" d="M1216 476.536 L1216 1423.18 L1251.34 1423.18 L1251.34 476.536 L1216 476.536 L1216 476.536  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,476.536 1216,1423.18 1251.34,1423.18 1251.34,476.536 1216,476.536 "/>
-<path clip-path="url(#clip472)" d="M1216 594.867 L1216 1423.18 L1251.34 1423.18 L1251.34 594.867 L1216 594.867 L1216 594.867  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,594.867 1216,1423.18 1251.34,1423.18 1251.34,594.867 1216,594.867 "/>
-<path clip-path="url(#clip472)" d="M1216 713.197 L1216 1423.18 L1251.34 1423.18 L1251.34 713.197 L1216 713.197 L1216 713.197  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,713.197 1216,1423.18 1251.34,1423.18 1251.34,713.197 1216,713.197 "/>
-<path clip-path="url(#clip472)" d="M1216 831.527 L1216 1423.18 L1251.34 1423.18 L1251.34 831.527 L1216 831.527 L1216 831.527  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,831.527 1216,1423.18 1251.34,1423.18 1251.34,831.527 1216,831.527 "/>
-<path clip-path="url(#clip472)" d="M1216 949.858 L1216 1423.18 L1251.34 1423.18 L1251.34 949.858 L1216 949.858 L1216 949.858  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,949.858 1216,1423.18 1251.34,1423.18 1251.34,949.858 1216,949.858 "/>
-<path clip-path="url(#clip472)" d="M1216 1068.19 L1216 1423.18 L1251.34 1423.18 L1251.34 1068.19 L1216 1068.19 L1216 1068.19  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1068.19 1216,1423.18 1251.34,1423.18 1251.34,1068.19 1216,1068.19 "/>
-<path clip-path="url(#clip472)" d="M1216 1186.52 L1216 1423.18 L1251.34 1423.18 L1251.34 1186.52 L1216 1186.52 L1216 1186.52  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1186.52 1216,1423.18 1251.34,1423.18 1251.34,1186.52 1216,1186.52 "/>
-<path clip-path="url(#clip472)" d="M1216 1304.85 L1216 1423.18 L1251.34 1423.18 L1251.34 1304.85 L1216 1304.85 L1216 1304.85  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1304.85 1216,1423.18 1251.34,1423.18 1251.34,1304.85 1216,1304.85 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="121.545" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="239.875" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="358.206" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="476.536" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="594.867" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="713.197" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="831.527" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="949.858" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1068.19" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1186.52" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1304.85" r="2"/>
-<path clip-path="url(#clip472)" d="M1274.9 69.2591 L1274.9 1423.18 L1310.24 1423.18 L1310.24 69.2591 L1274.9 69.2591 L1274.9 69.2591  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,69.2591 1274.9,1423.18 1310.24,1423.18 1310.24,69.2591 1274.9,69.2591 "/>
-<path clip-path="url(#clip472)" d="M1274.9 182.086 L1274.9 1423.18 L1310.24 1423.18 L1310.24 182.086 L1274.9 182.086 L1274.9 182.086  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,182.086 1274.9,1423.18 1310.24,1423.18 1310.24,182.086 1274.9,182.086 "/>
-<path clip-path="url(#clip472)" d="M1274.9 294.913 L1274.9 1423.18 L1310.24 1423.18 L1310.24 294.913 L1274.9 294.913 L1274.9 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,294.913 1274.9,1423.18 1310.24,1423.18 1310.24,294.913 1274.9,294.913 "/>
-<path clip-path="url(#clip472)" d="M1274.9 407.739 L1274.9 1423.18 L1310.24 1423.18 L1310.24 407.739 L1274.9 407.739 L1274.9 407.739  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,407.739 1274.9,1423.18 1310.24,1423.18 1310.24,407.739 1274.9,407.739 "/>
-<path clip-path="url(#clip472)" d="M1274.9 520.566 L1274.9 1423.18 L1310.24 1423.18 L1310.24 520.566 L1274.9 520.566 L1274.9 520.566  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,520.566 1274.9,1423.18 1310.24,1423.18 1310.24,520.566 1274.9,520.566 "/>
-<path clip-path="url(#clip472)" d="M1274.9 633.393 L1274.9 1423.18 L1310.24 1423.18 L1310.24 633.393 L1274.9 633.393 L1274.9 633.393  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,633.393 1274.9,1423.18 1310.24,1423.18 1310.24,633.393 1274.9,633.393 "/>
-<path clip-path="url(#clip472)" d="M1274.9 746.219 L1274.9 1423.18 L1310.24 1423.18 L1310.24 746.219 L1274.9 746.219 L1274.9 746.219  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,746.219 1274.9,1423.18 1310.24,1423.18 1310.24,746.219 1274.9,746.219 "/>
-<path clip-path="url(#clip472)" d="M1274.9 859.046 L1274.9 1423.18 L1310.24 1423.18 L1310.24 859.046 L1274.9 859.046 L1274.9 859.046  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,859.046 1274.9,1423.18 1310.24,1423.18 1310.24,859.046 1274.9,859.046 "/>
-<path clip-path="url(#clip472)" d="M1274.9 971.873 L1274.9 1423.18 L1310.24 1423.18 L1310.24 971.873 L1274.9 971.873 L1274.9 971.873  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,971.873 1274.9,1423.18 1310.24,1423.18 1310.24,971.873 1274.9,971.873 "/>
-<path clip-path="url(#clip472)" d="M1274.9 1084.7 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1084.7 L1274.9 1084.7 L1274.9 1084.7  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1084.7 1274.9,1423.18 1310.24,1423.18 1310.24,1084.7 1274.9,1084.7 "/>
-<path clip-path="url(#clip472)" d="M1274.9 1197.53 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1197.53 L1274.9 1197.53 L1274.9 1197.53  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1197.53 1274.9,1423.18 1310.24,1423.18 1310.24,1197.53 1274.9,1197.53 "/>
-<path clip-path="url(#clip472)" d="M1274.9 1310.35 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1310.35 L1274.9 1310.35 L1274.9 1310.35  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1310.35 1274.9,1423.18 1310.24,1423.18 1310.24,1310.35 1274.9,1310.35 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="69.2591" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="182.086" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="294.913" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="407.739" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="520.566" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="633.393" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="746.219" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="859.046" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="971.873" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1084.7" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1197.53" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1310.35" r="2"/>
-<path clip-path="url(#clip472)" d="M1333.8 61.0035 L1333.8 1423.18 L1369.14 1423.18 L1369.14 61.0035 L1333.8 61.0035 L1333.8 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,61.0035 1333.8,1423.18 1369.14,1423.18 1369.14,61.0035 1333.8,61.0035 "/>
-<path clip-path="url(#clip472)" d="M1333.8 151.815 L1333.8 1423.18 L1369.14 1423.18 L1369.14 151.815 L1333.8 151.815 L1333.8 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,151.815 1333.8,1423.18 1369.14,1423.18 1369.14,151.815 1333.8,151.815 "/>
-<path clip-path="url(#clip472)" d="M1333.8 242.627 L1333.8 1423.18 L1369.14 1423.18 L1369.14 242.627 L1333.8 242.627 L1333.8 242.627  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,242.627 1333.8,1423.18 1369.14,1423.18 1369.14,242.627 1333.8,242.627 "/>
-<path clip-path="url(#clip472)" d="M1333.8 333.439 L1333.8 1423.18 L1369.14 1423.18 L1369.14 333.439 L1333.8 333.439 L1333.8 333.439  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,333.439 1333.8,1423.18 1369.14,1423.18 1369.14,333.439 1333.8,333.439 "/>
-<path clip-path="url(#clip472)" d="M1333.8 424.25 L1333.8 1423.18 L1369.14 1423.18 L1369.14 424.25 L1333.8 424.25 L1333.8 424.25  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,424.25 1333.8,1423.18 1369.14,1423.18 1369.14,424.25 1333.8,424.25 "/>
-<path clip-path="url(#clip472)" d="M1333.8 515.062 L1333.8 1423.18 L1369.14 1423.18 L1369.14 515.062 L1333.8 515.062 L1333.8 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,515.062 1333.8,1423.18 1369.14,1423.18 1369.14,515.062 1333.8,515.062 "/>
-<path clip-path="url(#clip472)" d="M1333.8 605.874 L1333.8 1423.18 L1369.14 1423.18 L1369.14 605.874 L1333.8 605.874 L1333.8 605.874  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,605.874 1333.8,1423.18 1369.14,1423.18 1369.14,605.874 1333.8,605.874 "/>
-<path clip-path="url(#clip472)" d="M1333.8 696.686 L1333.8 1423.18 L1369.14 1423.18 L1369.14 696.686 L1333.8 696.686 L1333.8 696.686  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,696.686 1333.8,1423.18 1369.14,1423.18 1369.14,696.686 1333.8,696.686 "/>
-<path clip-path="url(#clip472)" d="M1333.8 787.498 L1333.8 1423.18 L1369.14 1423.18 L1369.14 787.498 L1333.8 787.498 L1333.8 787.498  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,787.498 1333.8,1423.18 1369.14,1423.18 1369.14,787.498 1333.8,787.498 "/>
-<path clip-path="url(#clip472)" d="M1333.8 878.309 L1333.8 1423.18 L1369.14 1423.18 L1369.14 878.309 L1333.8 878.309 L1333.8 878.309  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,878.309 1333.8,1423.18 1369.14,1423.18 1369.14,878.309 1333.8,878.309 "/>
-<path clip-path="url(#clip472)" d="M1333.8 969.121 L1333.8 1423.18 L1369.14 1423.18 L1369.14 969.121 L1333.8 969.121 L1333.8 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,969.121 1333.8,1423.18 1369.14,1423.18 1369.14,969.121 1333.8,969.121 "/>
-<path clip-path="url(#clip472)" d="M1333.8 1059.93 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1059.93 L1333.8 1059.93 L1333.8 1059.93  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1059.93 1333.8,1423.18 1369.14,1423.18 1369.14,1059.93 1333.8,1059.93 "/>
-<path clip-path="url(#clip472)" d="M1333.8 1150.74 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1150.74 L1333.8 1150.74 L1333.8 1150.74  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1150.74 1333.8,1423.18 1369.14,1423.18 1369.14,1150.74 1333.8,1150.74 "/>
-<path clip-path="url(#clip472)" d="M1333.8 1241.56 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1241.56 L1333.8 1241.56 L1333.8 1241.56  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1241.56 1333.8,1423.18 1369.14,1423.18 1369.14,1241.56 1333.8,1241.56 "/>
-<path clip-path="url(#clip472)" d="M1333.8 1332.37 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1332.37 L1333.8 1332.37 L1333.8 1332.37  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1332.37 1333.8,1423.18 1369.14,1423.18 1369.14,1332.37 1333.8,1332.37 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="61.0035" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="151.815" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="242.627" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="333.439" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="424.25" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="515.062" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="605.874" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="696.686" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="787.498" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="878.309" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="969.121" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1059.93" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1150.74" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1241.56" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1332.37" r="2"/>
-<path clip-path="url(#clip472)" d="M1392.7 89.8981 L1392.7 1423.18 L1428.04 1423.18 L1428.04 89.8981 L1392.7 89.8981 L1392.7 89.8981  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,89.8981 1392.7,1423.18 1428.04,1423.18 1428.04,89.8981 1392.7,89.8981 "/>
-<path clip-path="url(#clip472)" d="M1392.7 160.071 L1392.7 1423.18 L1428.04 1423.18 L1428.04 160.071 L1392.7 160.071 L1392.7 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,160.071 1392.7,1423.18 1428.04,1423.18 1428.04,160.071 1392.7,160.071 "/>
-<path clip-path="url(#clip472)" d="M1392.7 230.244 L1392.7 1423.18 L1428.04 1423.18 L1428.04 230.244 L1392.7 230.244 L1392.7 230.244  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,230.244 1392.7,1423.18 1428.04,1423.18 1428.04,230.244 1392.7,230.244 "/>
-<path clip-path="url(#clip472)" d="M1392.7 300.416 L1392.7 1423.18 L1428.04 1423.18 L1428.04 300.416 L1392.7 300.416 L1392.7 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,300.416 1392.7,1423.18 1428.04,1423.18 1428.04,300.416 1392.7,300.416 "/>
-<path clip-path="url(#clip472)" d="M1392.7 370.589 L1392.7 1423.18 L1428.04 1423.18 L1428.04 370.589 L1392.7 370.589 L1392.7 370.589  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,370.589 1392.7,1423.18 1428.04,1423.18 1428.04,370.589 1392.7,370.589 "/>
-<path clip-path="url(#clip472)" d="M1392.7 440.762 L1392.7 1423.18 L1428.04 1423.18 L1428.04 440.762 L1392.7 440.762 L1392.7 440.762  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,440.762 1392.7,1423.18 1428.04,1423.18 1428.04,440.762 1392.7,440.762 "/>
-<path clip-path="url(#clip472)" d="M1392.7 510.934 L1392.7 1423.18 L1428.04 1423.18 L1428.04 510.934 L1392.7 510.934 L1392.7 510.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,510.934 1392.7,1423.18 1428.04,1423.18 1428.04,510.934 1392.7,510.934 "/>
-<path clip-path="url(#clip472)" d="M1392.7 581.107 L1392.7 1423.18 L1428.04 1423.18 L1428.04 581.107 L1392.7 581.107 L1392.7 581.107  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,581.107 1392.7,1423.18 1428.04,1423.18 1428.04,581.107 1392.7,581.107 "/>
-<path clip-path="url(#clip472)" d="M1392.7 651.28 L1392.7 1423.18 L1428.04 1423.18 L1428.04 651.28 L1392.7 651.28 L1392.7 651.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,651.28 1392.7,1423.18 1428.04,1423.18 1428.04,651.28 1392.7,651.28 "/>
-<path clip-path="url(#clip472)" d="M1392.7 721.453 L1392.7 1423.18 L1428.04 1423.18 L1428.04 721.453 L1392.7 721.453 L1392.7 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,721.453 1392.7,1423.18 1428.04,1423.18 1428.04,721.453 1392.7,721.453 "/>
-<path clip-path="url(#clip472)" d="M1392.7 791.625 L1392.7 1423.18 L1428.04 1423.18 L1428.04 791.625 L1392.7 791.625 L1392.7 791.625  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,791.625 1392.7,1423.18 1428.04,1423.18 1428.04,791.625 1392.7,791.625 "/>
-<path clip-path="url(#clip472)" d="M1392.7 861.798 L1392.7 1423.18 L1428.04 1423.18 L1428.04 861.798 L1392.7 861.798 L1392.7 861.798  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,861.798 1392.7,1423.18 1428.04,1423.18 1428.04,861.798 1392.7,861.798 "/>
-<path clip-path="url(#clip472)" d="M1392.7 931.971 L1392.7 1423.18 L1428.04 1423.18 L1428.04 931.971 L1392.7 931.971 L1392.7 931.971  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,931.971 1392.7,1423.18 1428.04,1423.18 1428.04,931.971 1392.7,931.971 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1002.14 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1002.14 L1392.7 1002.14 L1392.7 1002.14  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1002.14 1392.7,1423.18 1428.04,1423.18 1428.04,1002.14 1392.7,1002.14 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1072.32 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1072.32 L1392.7 1072.32 L1392.7 1072.32  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1072.32 1392.7,1423.18 1428.04,1423.18 1428.04,1072.32 1392.7,1072.32 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1142.49 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1142.49 L1392.7 1142.49 L1392.7 1142.49  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1142.49 1392.7,1423.18 1428.04,1423.18 1428.04,1142.49 1392.7,1142.49 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1212.66 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1212.66 L1392.7 1212.66 L1392.7 1212.66  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1212.66 1392.7,1423.18 1428.04,1423.18 1428.04,1212.66 1392.7,1212.66 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1282.83 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1282.83 L1392.7 1282.83 L1392.7 1282.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1282.83 1392.7,1423.18 1428.04,1423.18 1428.04,1282.83 1392.7,1282.83 "/>
-<path clip-path="url(#clip472)" d="M1392.7 1353.01 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1353.01 L1392.7 1353.01 L1392.7 1353.01  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1353.01 1392.7,1423.18 1428.04,1423.18 1428.04,1353.01 1392.7,1353.01 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="89.8981" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="160.071" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="230.244" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="300.416" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="370.589" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="440.762" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="510.934" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="581.107" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="651.28" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="721.453" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="791.625" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="861.798" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="931.971" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1002.14" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1072.32" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1142.49" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1212.66" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1282.83" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1353.01" r="2"/>
-<path clip-path="url(#clip472)" d="M1451.6 61.0035 L1451.6 1423.18 L1486.94 1423.18 L1486.94 61.0035 L1451.6 61.0035 L1451.6 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,61.0035 1451.6,1423.18 1486.94,1423.18 1486.94,61.0035 1451.6,61.0035 "/>
-<path clip-path="url(#clip472)" d="M1451.6 179.334 L1451.6 1423.18 L1486.94 1423.18 L1486.94 179.334 L1451.6 179.334 L1451.6 179.334  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,179.334 1451.6,1423.18 1486.94,1423.18 1486.94,179.334 1451.6,179.334 "/>
-<path clip-path="url(#clip472)" d="M1451.6 455.897 L1451.6 1423.18 L1486.94 1423.18 L1486.94 455.897 L1451.6 455.897 L1451.6 455.897  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,455.897 1451.6,1423.18 1486.94,1423.18 1486.94,455.897 1451.6,455.897 "/>
-<path clip-path="url(#clip472)" d="M1451.6 732.46 L1451.6 1423.18 L1486.94 1423.18 L1486.94 732.46 L1451.6 732.46 L1451.6 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,732.46 1451.6,1423.18 1486.94,1423.18 1486.94,732.46 1451.6,732.46 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="61.0035" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="179.334" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="455.897" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="732.46" r="2"/>
-<path clip-path="url(#clip472)" d="M1510.5 52.7478 L1510.5 1423.18 L1545.84 1423.18 L1545.84 52.7478 L1510.5 52.7478 L1510.5 52.7478  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,52.7478 1510.5,1423.18 1545.84,1423.18 1545.84,52.7478 1510.5,52.7478 "/>
-<path clip-path="url(#clip472)" d="M1510.5 122.921 L1510.5 1423.18 L1545.84 1423.18 L1545.84 122.921 L1510.5 122.921 L1510.5 122.921  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,122.921 1510.5,1423.18 1545.84,1423.18 1545.84,122.921 1510.5,122.921 "/>
-<path clip-path="url(#clip472)" d="M1510.5 322.431 L1510.5 1423.18 L1545.84 1423.18 L1545.84 322.431 L1510.5 322.431 L1510.5 322.431  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,322.431 1510.5,1423.18 1545.84,1423.18 1545.84,322.431 1510.5,322.431 "/>
-<path clip-path="url(#clip472)" d="M1510.5 521.942 L1510.5 1423.18 L1545.84 1423.18 L1545.84 521.942 L1510.5 521.942 L1510.5 521.942  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,521.942 1510.5,1423.18 1545.84,1423.18 1545.84,521.942 1510.5,521.942 "/>
-<path clip-path="url(#clip472)" d="M1510.5 721.453 L1510.5 1423.18 L1545.84 1423.18 L1545.84 721.453 L1510.5 721.453 L1510.5 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,721.453 1510.5,1423.18 1545.84,1423.18 1545.84,721.453 1510.5,721.453 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="52.7478" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="122.921" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="322.431" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="521.942" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="721.453" r="2"/>
-<path clip-path="url(#clip472)" d="M1569.4 47.2441 L1569.4 1423.18 L1604.74 1423.18 L1604.74 47.2441 L1569.4 47.2441 L1569.4 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,47.2441 1569.4,1423.18 1604.74,1423.18 1604.74,47.2441 1569.4,47.2441 "/>
-<path clip-path="url(#clip472)" d="M1569.4 165.575 L1569.4 1423.18 L1604.74 1423.18 L1604.74 165.575 L1569.4 165.575 L1569.4 165.575  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,165.575 1569.4,1423.18 1604.74,1423.18 1604.74,165.575 1569.4,165.575 "/>
-<path clip-path="url(#clip472)" d="M1569.4 283.905 L1569.4 1423.18 L1604.74 1423.18 L1604.74 283.905 L1569.4 283.905 L1569.4 283.905  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,283.905 1569.4,1423.18 1604.74,1423.18 1604.74,283.905 1569.4,283.905 "/>
-<path clip-path="url(#clip472)" d="M1569.4 693.934 L1569.4 1423.18 L1604.74 1423.18 L1604.74 693.934 L1569.4 693.934 L1569.4 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,693.934 1569.4,1423.18 1604.74,1423.18 1604.74,693.934 1569.4,693.934 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="165.575" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="283.905" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="693.934" r="2"/>
-<path clip-path="url(#clip472)" d="M1628.3 48.62 L1628.3 1423.18 L1663.64 1423.18 L1663.64 48.62 L1628.3 48.62 L1628.3 48.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,48.62 1628.3,1423.18 1663.64,1423.18 1663.64,48.62 1628.3,48.62 "/>
-<path clip-path="url(#clip472)" d="M1628.3 118.793 L1628.3 1423.18 L1663.64 1423.18 L1663.64 118.793 L1628.3 118.793 L1628.3 118.793  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,118.793 1628.3,1423.18 1663.64,1423.18 1663.64,118.793 1628.3,118.793 "/>
-<path clip-path="url(#clip472)" d="M1628.3 209.605 L1628.3 1423.18 L1663.64 1423.18 L1663.64 209.605 L1628.3 209.605 L1628.3 209.605  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,209.605 1628.3,1423.18 1663.64,1423.18 1663.64,209.605 1628.3,209.605 "/>
-<path clip-path="url(#clip472)" d="M1628.3 322.431 L1628.3 1423.18 L1663.64 1423.18 L1663.64 322.431 L1628.3 322.431 L1628.3 322.431  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,322.431 1628.3,1423.18 1663.64,1423.18 1663.64,322.431 1628.3,322.431 "/>
-<path clip-path="url(#clip472)" d="M1628.3 732.46 L1628.3 1423.18 L1663.64 1423.18 L1663.64 732.46 L1628.3 732.46 L1628.3 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,732.46 1628.3,1423.18 1663.64,1423.18 1663.64,732.46 1628.3,732.46 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="48.62" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="118.793" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="209.605" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="322.431" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="732.46" r="2"/>
-<path clip-path="url(#clip472)" d="M1687.2 55.4997 L1687.2 1423.18 L1722.54 1423.18 L1722.54 55.4997 L1687.2 55.4997 L1687.2 55.4997  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,55.4997 1687.2,1423.18 1722.54,1423.18 1722.54,55.4997 1687.2,55.4997 "/>
-<path clip-path="url(#clip472)" d="M1687.2 146.311 L1687.2 1423.18 L1722.54 1423.18 L1722.54 146.311 L1687.2 146.311 L1687.2 146.311  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,146.311 1687.2,1423.18 1722.54,1423.18 1722.54,146.311 1687.2,146.311 "/>
-<path clip-path="url(#clip472)" d="M1687.2 297.664 L1687.2 1423.18 L1722.54 1423.18 L1722.54 297.664 L1687.2 297.664 L1687.2 297.664  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,297.664 1687.2,1423.18 1722.54,1423.18 1722.54,297.664 1687.2,297.664 "/>
-<path clip-path="url(#clip472)" d="M1687.2 721.453 L1687.2 1423.18 L1722.54 1423.18 L1722.54 721.453 L1687.2 721.453 L1687.2 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,721.453 1687.2,1423.18 1722.54,1423.18 1722.54,721.453 1687.2,721.453 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="55.4997" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="146.311" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="297.664" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="721.453" r="2"/>
-<path clip-path="url(#clip472)" d="M1746.1 47.2441 L1746.1 1423.18 L1781.43 1423.18 L1781.43 47.2441 L1746.1 47.2441 L1746.1 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,47.2441 1746.1,1423.18 1781.43,1423.18 1781.43,47.2441 1746.1,47.2441 "/>
-<path clip-path="url(#clip472)" d="M1746.1 270.146 L1746.1 1423.18 L1781.43 1423.18 L1781.43 270.146 L1746.1 270.146 L1746.1 270.146  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,270.146 1746.1,1423.18 1781.43,1423.18 1781.43,270.146 1746.1,270.146 "/>
-<path clip-path="url(#clip472)" d="M1746.1 693.934 L1746.1 1423.18 L1781.43 1423.18 L1781.43 693.934 L1746.1 693.934 L1746.1 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,693.934 1746.1,1423.18 1781.43,1423.18 1781.43,693.934 1746.1,693.934 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="270.146" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="693.934" r="2"/>
-<path clip-path="url(#clip472)" d="M1804.99 59.6275 L1804.99 1423.18 L1840.33 1423.18 L1840.33 59.6275 L1804.99 59.6275 L1804.99 59.6275  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,59.6275 1804.99,1423.18 1840.33,1423.18 1840.33,59.6275 1804.99,59.6275 "/>
-<path clip-path="url(#clip472)" d="M1804.99 129.8 L1804.99 1423.18 L1840.33 1423.18 L1840.33 129.8 L1804.99 129.8 L1804.99 129.8  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,129.8 1804.99,1423.18 1840.33,1423.18 1840.33,129.8 1804.99,129.8 "/>
-<path clip-path="url(#clip472)" d="M1804.99 406.363 L1804.99 1423.18 L1840.33 1423.18 L1840.33 406.363 L1804.99 406.363 L1804.99 406.363  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,406.363 1804.99,1423.18 1840.33,1423.18 1840.33,406.363 1804.99,406.363 "/>
-<path clip-path="url(#clip472)" d="M1804.99 682.926 L1804.99 1423.18 L1840.33 1423.18 L1840.33 682.926 L1804.99 682.926 L1804.99 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,682.926 1804.99,1423.18 1840.33,1423.18 1840.33,682.926 1804.99,682.926 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="59.6275" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="129.8" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="406.363" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="682.926" r="2"/>
-<path clip-path="url(#clip472)" d="M1863.89 58.2516 L1863.89 1423.18 L1899.23 1423.18 L1899.23 58.2516 L1863.89 58.2516 L1863.89 58.2516  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,58.2516 1863.89,1423.18 1899.23,1423.18 1899.23,58.2516 1863.89,58.2516 "/>
-<path clip-path="url(#clip472)" d="M1863.89 176.582 L1863.89 1423.18 L1899.23 1423.18 L1899.23 176.582 L1863.89 176.582 L1863.89 176.582  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,176.582 1863.89,1423.18 1899.23,1423.18 1899.23,176.582 1863.89,176.582 "/>
-<path clip-path="url(#clip472)" d="M1863.89 294.913 L1863.89 1423.18 L1899.23 1423.18 L1899.23 294.913 L1863.89 294.913 L1863.89 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,294.913 1863.89,1423.18 1899.23,1423.18 1899.23,294.913 1863.89,294.913 "/>
-<path clip-path="url(#clip472)" d="M1863.89 413.243 L1863.89 1423.18 L1899.23 1423.18 L1899.23 413.243 L1863.89 413.243 L1863.89 413.243  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,413.243 1863.89,1423.18 1899.23,1423.18 1899.23,413.243 1863.89,413.243 "/>
-<path clip-path="url(#clip472)" d="M1863.89 531.573 L1863.89 1423.18 L1899.23 1423.18 L1899.23 531.573 L1863.89 531.573 L1863.89 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,531.573 1863.89,1423.18 1899.23,1423.18 1899.23,531.573 1863.89,531.573 "/>
-<path clip-path="url(#clip472)" d="M1863.89 682.926 L1863.89 1423.18 L1899.23 1423.18 L1899.23 682.926 L1863.89 682.926 L1863.89 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,682.926 1863.89,1423.18 1899.23,1423.18 1899.23,682.926 1863.89,682.926 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="58.2516" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="176.582" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="294.913" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="413.243" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="531.573" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="682.926" r="2"/>
-<path clip-path="url(#clip472)" d="M1922.79 55.4997 L1922.79 1423.18 L1958.13 1423.18 L1958.13 55.4997 L1922.79 55.4997 L1922.79 55.4997  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,55.4997 1922.79,1423.18 1958.13,1423.18 1958.13,55.4997 1922.79,55.4997 "/>
-<path clip-path="url(#clip472)" d="M1922.79 278.401 L1922.79 1423.18 L1958.13 1423.18 L1958.13 278.401 L1922.79 278.401 L1922.79 278.401  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,278.401 1922.79,1423.18 1958.13,1423.18 1958.13,278.401 1922.79,278.401 "/>
-<path clip-path="url(#clip472)" d="M1922.79 721.453 L1922.79 1423.18 L1958.13 1423.18 L1958.13 721.453 L1922.79 721.453 L1922.79 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,721.453 1922.79,1423.18 1958.13,1423.18 1958.13,721.453 1922.79,721.453 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="55.4997" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="278.401" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="721.453" r="2"/>
-<path clip-path="url(#clip472)" d="M1981.69 47.2441 L1981.69 1423.18 L2017.03 1423.18 L2017.03 47.2441 L1981.69 47.2441 L1981.69 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,47.2441 1981.69,1423.18 2017.03,1423.18 2017.03,47.2441 1981.69,47.2441 "/>
-<path clip-path="url(#clip472)" d="M1981.69 160.071 L1981.69 1423.18 L2017.03 1423.18 L2017.03 160.071 L1981.69 160.071 L1981.69 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,160.071 1981.69,1423.18 2017.03,1423.18 2017.03,160.071 1981.69,160.071 "/>
-<path clip-path="url(#clip472)" d="M1981.69 272.898 L1981.69 1423.18 L2017.03 1423.18 L2017.03 272.898 L1981.69 272.898 L1981.69 272.898  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,272.898 1981.69,1423.18 2017.03,1423.18 2017.03,272.898 1981.69,272.898 "/>
-<path clip-path="url(#clip472)" d="M1981.69 682.926 L1981.69 1423.18 L2017.03 1423.18 L2017.03 682.926 L1981.69 682.926 L1981.69 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,682.926 1981.69,1423.18 2017.03,1423.18 2017.03,682.926 1981.69,682.926 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="160.071" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="272.898" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="682.926" r="2"/>
-<path clip-path="url(#clip472)" d="M2040.59 88.5222 L2040.59 1423.18 L2075.93 1423.18 L2075.93 88.5222 L2040.59 88.5222 L2040.59 88.5222  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,88.5222 2040.59,1423.18 2075.93,1423.18 2075.93,88.5222 2040.59,88.5222 "/>
-<path clip-path="url(#clip472)" d="M2040.59 239.875 L2040.59 1423.18 L2075.93 1423.18 L2075.93 239.875 L2040.59 239.875 L2040.59 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,239.875 2040.59,1423.18 2075.93,1423.18 2075.93,239.875 2040.59,239.875 "/>
-<path clip-path="url(#clip472)" d="M2040.59 682.926 L2040.59 1423.18 L2075.93 1423.18 L2075.93 682.926 L2040.59 682.926 L2040.59 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,682.926 2040.59,1423.18 2075.93,1423.18 2075.93,682.926 2040.59,682.926 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="88.5222" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="239.875" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="682.926" r="2"/>
-<path clip-path="url(#clip472)" d="M2099.49 49.996 L2099.49 1423.18 L2134.83 1423.18 L2134.83 49.996 L2099.49 49.996 L2099.49 49.996  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,49.996 2099.49,1423.18 2134.83,1423.18 2134.83,49.996 2099.49,49.996 "/>
-<path clip-path="url(#clip472)" d="M2099.49 168.326 L2099.49 1423.18 L2134.83 1423.18 L2134.83 168.326 L2099.49 168.326 L2099.49 168.326  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,168.326 2099.49,1423.18 2134.83,1423.18 2134.83,168.326 2099.49,168.326 "/>
-<path clip-path="url(#clip472)" d="M2099.49 391.228 L2099.49 1423.18 L2134.83 1423.18 L2134.83 391.228 L2099.49 391.228 L2099.49 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,391.228 2099.49,1423.18 2134.83,1423.18 2134.83,391.228 2099.49,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="49.996" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="168.326" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M2158.39 114.665 L2158.39 1423.18 L2193.73 1423.18 L2193.73 114.665 L2158.39 114.665 L2158.39 114.665  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2158.39,114.665 2158.39,1423.18 2193.73,1423.18 2193.73,114.665 2158.39,114.665 "/>
-<path clip-path="url(#clip472)" d="M2158.39 391.228 L2158.39 1423.18 L2193.73 1423.18 L2193.73 391.228 L2158.39 391.228 L2158.39 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2158.39,391.228 2158.39,1423.18 2193.73,1423.18 2193.73,391.228 2158.39,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2176.06" cy="114.665" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2176.06" cy="391.228" r="2"/>
-<path clip-path="url(#clip472)" d="M2217.29 109.161 L2217.29 1423.18 L2252.63 1423.18 L2252.63 109.161 L2217.29 109.161 L2217.29 109.161  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,109.161 2217.29,1423.18 2252.63,1423.18 2252.63,109.161 2217.29,109.161 "/>
-<path clip-path="url(#clip472)" d="M2217.29 308.672 L2217.29 1423.18 L2252.63 1423.18 L2252.63 308.672 L2217.29 308.672 L2217.29 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,308.672 2217.29,1423.18 2252.63,1423.18 2252.63,308.672 2217.29,308.672 "/>
-<path clip-path="url(#clip472)" d="M2217.29 732.46 L2217.29 1423.18 L2252.63 1423.18 L2252.63 732.46 L2217.29 732.46 L2217.29 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,732.46 2217.29,1423.18 2252.63,1423.18 2252.63,732.46 2217.29,732.46 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="109.161" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="308.672" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="732.46" r="2"/>
-<path clip-path="url(#clip472)" d="M2276.19 48.62 L2276.19 1423.18 L2311.53 1423.18 L2311.53 48.62 L2276.19 48.62 L2276.19 48.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,48.62 2276.19,1423.18 2311.53,1423.18 2311.53,48.62 2276.19,48.62 "/>
-<path clip-path="url(#clip472)" d="M2276.19 118.793 L2276.19 1423.18 L2311.53 1423.18 L2311.53 118.793 L2276.19 118.793 L2276.19 118.793  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,118.793 2276.19,1423.18 2311.53,1423.18 2311.53,118.793 2276.19,118.793 "/>
-<path clip-path="url(#clip472)" d="M2276.19 209.605 L2276.19 1423.18 L2311.53 1423.18 L2311.53 209.605 L2276.19 209.605 L2276.19 209.605  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,209.605 2276.19,1423.18 2311.53,1423.18 2311.53,209.605 2276.19,209.605 "/>
-<path clip-path="url(#clip472)" d="M2276.19 300.416 L2276.19 1423.18 L2311.53 1423.18 L2311.53 300.416 L2276.19 300.416 L2276.19 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,300.416 2276.19,1423.18 2311.53,1423.18 2311.53,300.416 2276.19,300.416 "/>
-<path clip-path="url(#clip472)" d="M2276.19 391.228 L2276.19 1423.18 L2311.53 1423.18 L2311.53 391.228 L2276.19 391.228 L2276.19 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,391.228 2276.19,1423.18 2311.53,1423.18 2311.53,391.228 2276.19,391.228 "/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="48.62" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="118.793" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="209.605" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="300.416" r="2"/>
-<circle clip-path="url(#clip472)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="391.228" r="2"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="821.38,1423.18 821.38,47.2441 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.37,1423.18 1410.37,47.2441 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1999.36,1423.18 1999.36,47.2441 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,1404.28 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="821.38,1423.18 821.38,1404.28 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.37,1423.18 1410.37,1404.28 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.36,1423.18 1999.36,1404.28 "/>
+<path clip-path="url(#clip290)" d="M232.389 1454.1 Q228.778 1454.1 226.95 1457.66 Q225.144 1461.2 225.144 1468.33 Q225.144 1475.44 226.95 1479.01 Q228.778 1482.55 232.389 1482.55 Q236.024 1482.55 237.829 1479.01 Q239.658 1475.44 239.658 1468.33 Q239.658 1461.2 237.829 1457.66 Q236.024 1454.1 232.389 1454.1 M232.389 1450.39 Q238.2 1450.39 241.255 1455 Q244.334 1459.58 244.334 1468.33 Q244.334 1477.06 241.255 1481.67 Q238.2 1486.25 232.389 1486.25 Q226.579 1486.25 223.501 1481.67 Q220.445 1477.06 220.445 1468.33 Q220.445 1459.58 223.501 1455 Q226.579 1450.39 232.389 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M796.068 1481.64 L803.707 1481.64 L803.707 1455.28 L795.396 1456.95 L795.396 1452.69 L803.66 1451.02 L808.336 1451.02 L808.336 1481.64 L815.975 1481.64 L815.975 1485.58 L796.068 1485.58 L796.068 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M835.419 1454.1 Q831.808 1454.1 829.98 1457.66 Q828.174 1461.2 828.174 1468.33 Q828.174 1475.44 829.98 1479.01 Q831.808 1482.55 835.419 1482.55 Q839.054 1482.55 840.859 1479.01 Q842.688 1475.44 842.688 1468.33 Q842.688 1461.2 840.859 1457.66 Q839.054 1454.1 835.419 1454.1 M835.419 1450.39 Q841.229 1450.39 844.285 1455 Q847.364 1459.58 847.364 1468.33 Q847.364 1477.06 844.285 1481.67 Q841.229 1486.25 835.419 1486.25 Q829.609 1486.25 826.53 1481.67 Q823.475 1477.06 823.475 1468.33 Q823.475 1459.58 826.53 1455 Q829.609 1450.39 835.419 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1389.14 1481.64 L1405.46 1481.64 L1405.46 1485.58 L1383.52 1485.58 L1383.52 1481.64 Q1386.18 1478.89 1390.76 1474.26 Q1395.37 1469.61 1396.55 1468.27 Q1398.8 1465.74 1399.68 1464.01 Q1400.58 1462.25 1400.58 1460.56 Q1400.58 1457.8 1398.63 1456.07 Q1396.71 1454.33 1393.61 1454.33 Q1391.41 1454.33 1388.96 1455.09 Q1386.53 1455.86 1383.75 1457.41 L1383.75 1452.69 Q1386.57 1451.55 1389.03 1450.97 Q1391.48 1450.39 1393.52 1450.39 Q1398.89 1450.39 1402.08 1453.08 Q1405.28 1455.77 1405.28 1460.26 Q1405.28 1462.39 1404.47 1464.31 Q1403.68 1466.2 1401.57 1468.8 Q1401 1469.47 1397.89 1472.69 Q1394.79 1475.88 1389.14 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1425.28 1454.1 Q1421.67 1454.1 1419.84 1457.66 Q1418.03 1461.2 1418.03 1468.33 Q1418.03 1475.44 1419.84 1479.01 Q1421.67 1482.55 1425.28 1482.55 Q1428.91 1482.55 1430.72 1479.01 Q1432.55 1475.44 1432.55 1468.33 Q1432.55 1461.2 1430.72 1457.66 Q1428.91 1454.1 1425.28 1454.1 M1425.28 1450.39 Q1431.09 1450.39 1434.14 1455 Q1437.22 1459.58 1437.22 1468.33 Q1437.22 1477.06 1434.14 1481.67 Q1431.09 1486.25 1425.28 1486.25 Q1419.47 1486.25 1416.39 1481.67 Q1413.33 1477.06 1413.33 1468.33 Q1413.33 1459.58 1416.39 1455 Q1419.47 1450.39 1425.28 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1988.2 1466.95 Q1991.56 1467.66 1993.44 1469.93 Q1995.33 1472.2 1995.33 1475.53 Q1995.33 1480.65 1991.82 1483.45 Q1988.3 1486.25 1981.82 1486.25 Q1979.64 1486.25 1977.32 1485.81 Q1975.03 1485.39 1972.58 1484.54 L1972.58 1480.02 Q1974.52 1481.16 1976.84 1481.74 Q1979.15 1482.32 1981.68 1482.32 Q1986.07 1482.32 1988.37 1480.58 Q1990.68 1478.84 1990.68 1475.53 Q1990.68 1472.48 1988.53 1470.77 Q1986.4 1469.03 1982.58 1469.03 L1978.55 1469.03 L1978.55 1465.19 L1982.76 1465.19 Q1986.21 1465.19 1988.04 1463.82 Q1989.87 1462.43 1989.87 1459.84 Q1989.87 1457.18 1987.97 1455.77 Q1986.1 1454.33 1982.58 1454.33 Q1980.66 1454.33 1978.46 1454.75 Q1976.26 1455.16 1973.62 1456.04 L1973.62 1451.88 Q1976.28 1451.14 1978.6 1450.77 Q1980.94 1450.39 1983 1450.39 Q1988.32 1450.39 1991.42 1452.83 Q1994.52 1455.23 1994.52 1459.35 Q1994.52 1462.22 1992.88 1464.21 Q1991.24 1466.18 1988.2 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M2014.2 1454.1 Q2010.59 1454.1 2008.76 1457.66 Q2006.95 1461.2 2006.95 1468.33 Q2006.95 1475.44 2008.76 1479.01 Q2010.59 1482.55 2014.2 1482.55 Q2017.83 1482.55 2019.64 1479.01 Q2021.47 1475.44 2021.47 1468.33 Q2021.47 1461.2 2019.64 1457.66 Q2017.83 1454.1 2014.2 1454.1 M2014.2 1450.39 Q2020.01 1450.39 2023.07 1455 Q2026.14 1459.58 2026.14 1468.33 Q2026.14 1477.06 2023.07 1481.67 Q2020.01 1486.25 2014.2 1486.25 Q2008.39 1486.25 2005.31 1481.67 Q2002.25 1477.06 2002.25 1468.33 Q2002.25 1459.58 2005.31 1455 Q2008.39 1450.39 2014.2 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1187.71 1525.81 L1187.71 1543.66 L1195.8 1543.66 Q1200.29 1543.66 1202.74 1541.34 Q1205.19 1539.02 1205.19 1534.72 Q1205.19 1530.45 1202.74 1528.13 Q1200.29 1525.81 1195.8 1525.81 L1187.71 1525.81 M1181.28 1520.52 L1195.8 1520.52 Q1203.79 1520.52 1207.86 1524.15 Q1211.97 1527.75 1211.97 1534.72 Q1211.97 1541.75 1207.86 1545.35 Q1203.79 1548.95 1195.8 1548.95 L1187.71 1548.95 L1187.71 1568.04 L1181.28 1568.04 L1181.28 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1233.61 1550.12 Q1226.51 1550.12 1223.78 1551.75 Q1221.04 1553.37 1221.04 1557.29 Q1221.04 1560.4 1223.08 1562.25 Q1225.14 1564.07 1228.68 1564.07 Q1233.55 1564.07 1236.47 1560.63 Q1239.43 1557.16 1239.43 1551.43 L1239.43 1550.12 L1233.61 1550.12 M1245.29 1547.71 L1245.29 1568.04 L1239.43 1568.04 L1239.43 1562.63 Q1237.43 1565.88 1234.44 1567.44 Q1231.45 1568.97 1227.12 1568.97 Q1221.64 1568.97 1218.4 1565.91 Q1215.18 1562.82 1215.18 1557.67 Q1215.18 1551.65 1219.19 1548.6 Q1223.23 1545.54 1231.22 1545.54 L1239.43 1545.54 L1239.43 1544.97 Q1239.43 1540.93 1236.76 1538.73 Q1234.12 1536.5 1229.31 1536.5 Q1226.26 1536.5 1223.36 1537.23 Q1220.47 1537.97 1217.79 1539.43 L1217.79 1534.02 Q1221.01 1532.78 1224.03 1532.17 Q1227.05 1531.54 1229.92 1531.54 Q1237.65 1531.54 1241.47 1535.55 Q1245.29 1539.56 1245.29 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1263.15 1522.27 L1263.15 1532.4 L1275.21 1532.4 L1275.21 1536.95 L1263.15 1536.95 L1263.15 1556.3 Q1263.15 1560.66 1264.32 1561.9 Q1265.53 1563.14 1269.19 1563.14 L1275.21 1563.14 L1275.21 1568.04 L1269.19 1568.04 Q1262.42 1568.04 1259.84 1565.53 Q1257.26 1562.98 1257.26 1556.3 L1257.26 1536.95 L1252.96 1536.95 L1252.96 1532.4 L1257.26 1532.4 L1257.26 1522.27 L1263.15 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1288.71 1522.27 L1288.71 1532.4 L1300.77 1532.4 L1300.77 1536.95 L1288.71 1536.95 L1288.71 1556.3 Q1288.71 1560.66 1289.88 1561.9 Q1291.09 1563.14 1294.75 1563.14 L1300.77 1563.14 L1300.77 1568.04 L1294.75 1568.04 Q1287.97 1568.04 1285.4 1565.53 Q1282.82 1562.98 1282.82 1556.3 L1282.82 1536.95 L1278.52 1536.95 L1278.52 1532.4 L1282.82 1532.4 L1282.82 1522.27 L1288.71 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1338.96 1548.76 L1338.96 1551.62 L1312.04 1551.62 Q1312.42 1557.67 1315.66 1560.85 Q1318.94 1564 1324.77 1564 Q1328.14 1564 1331.29 1563.17 Q1334.47 1562.35 1337.59 1560.69 L1337.59 1566.23 Q1334.44 1567.57 1331.13 1568.27 Q1327.82 1568.97 1324.42 1568.97 Q1315.89 1568.97 1310.89 1564 Q1305.92 1559.04 1305.92 1550.57 Q1305.92 1541.82 1310.64 1536.69 Q1315.38 1531.54 1323.4 1531.54 Q1330.59 1531.54 1334.76 1536.18 Q1338.96 1540.8 1338.96 1548.76 M1333.11 1547.04 Q1333.04 1542.23 1330.4 1539.37 Q1327.79 1536.5 1323.46 1536.5 Q1318.56 1536.5 1315.6 1539.27 Q1312.67 1542.04 1312.23 1547.07 L1333.11 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1369.23 1537.87 Q1368.24 1537.3 1367.07 1537.04 Q1365.92 1536.76 1364.52 1536.76 Q1359.56 1536.76 1356.88 1540 Q1354.24 1543.22 1354.24 1549.27 L1354.24 1568.04 L1348.35 1568.04 L1348.35 1532.4 L1354.24 1532.4 L1354.24 1537.93 Q1356.09 1534.69 1359.05 1533.13 Q1362.01 1531.54 1366.24 1531.54 Q1366.84 1531.54 1367.58 1531.63 Q1368.31 1531.7 1369.2 1531.85 L1369.23 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M1403.86 1546.53 L1403.86 1568.04 L1398 1568.04 L1398 1546.72 Q1398 1541.66 1396.03 1539.14 Q1394.06 1536.63 1390.11 1536.63 Q1385.37 1536.63 1382.63 1539.65 Q1379.89 1542.68 1379.89 1547.9 L1379.89 1568.04 L1374.01 1568.04 L1374.01 1532.4 L1379.89 1532.4 L1379.89 1537.93 Q1381.99 1534.72 1384.83 1533.13 Q1387.69 1531.54 1391.42 1531.54 Q1397.56 1531.54 1400.71 1535.36 Q1403.86 1539.14 1403.86 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1147.99 2352.76,1147.99 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,872.806 2352.76,872.806 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,597.618 2352.76,597.618 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,322.431 2352.76,322.431 "/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 251.287,1423.18 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1147.99 251.287,1147.99 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,872.806 251.287,872.806 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,597.618 251.287,597.618 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,322.431 251.287,322.431 "/>
+<polyline clip-path="url(#clip290)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,47.2441 251.287,47.2441 "/>
+<path clip-path="url(#clip290)" d="M184.445 1408.98 Q180.834 1408.98 179.005 1412.54 Q177.2 1416.08 177.2 1423.21 Q177.2 1430.32 179.005 1433.89 Q180.834 1437.43 184.445 1437.43 Q188.079 1437.43 189.885 1433.89 Q191.713 1430.32 191.713 1423.21 Q191.713 1416.08 189.885 1412.54 Q188.079 1408.98 184.445 1408.98 M184.445 1405.27 Q190.255 1405.27 193.311 1409.88 Q196.389 1414.46 196.389 1423.21 Q196.389 1431.94 193.311 1436.55 Q190.255 1441.13 184.445 1441.13 Q178.635 1441.13 175.556 1436.55 Q172.501 1431.94 172.501 1423.21 Q172.501 1414.46 175.556 1409.88 Q178.635 1405.27 184.445 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M148.311 1161.34 L164.63 1161.34 L164.63 1165.27 L142.686 1165.27 L142.686 1161.34 Q145.348 1158.58 149.931 1153.95 Q154.538 1149.3 155.718 1147.96 Q157.964 1145.43 158.843 1143.7 Q159.746 1141.94 159.746 1140.25 Q159.746 1137.5 157.802 1135.76 Q155.88 1134.02 152.778 1134.02 Q150.579 1134.02 148.126 1134.79 Q145.695 1135.55 142.917 1137.1 L142.917 1132.38 Q145.741 1131.25 148.195 1130.67 Q150.649 1130.09 152.686 1130.09 Q158.056 1130.09 161.251 1132.77 Q164.445 1135.46 164.445 1139.95 Q164.445 1142.08 163.635 1144 Q162.848 1145.9 160.741 1148.49 Q160.163 1149.16 157.061 1152.38 Q153.959 1155.57 148.311 1161.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M184.445 1133.79 Q180.834 1133.79 179.005 1137.36 Q177.2 1140.9 177.2 1148.03 Q177.2 1155.13 179.005 1158.7 Q180.834 1162.24 184.445 1162.24 Q188.079 1162.24 189.885 1158.7 Q191.713 1155.13 191.713 1148.03 Q191.713 1140.9 189.885 1137.36 Q188.079 1133.79 184.445 1133.79 M184.445 1130.09 Q190.255 1130.09 193.311 1134.69 Q196.389 1139.28 196.389 1148.03 Q196.389 1156.75 193.311 1161.36 Q190.255 1165.94 184.445 1165.94 Q178.635 1165.94 175.556 1161.36 Q172.501 1156.75 172.501 1148.03 Q172.501 1139.28 175.556 1134.69 Q178.635 1130.09 184.445 1130.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M157.13 859.6 L145.325 878.049 L157.13 878.049 L157.13 859.6 M155.903 855.526 L161.783 855.526 L161.783 878.049 L166.714 878.049 L166.714 881.937 L161.783 881.937 L161.783 890.086 L157.13 890.086 L157.13 881.937 L141.529 881.937 L141.529 877.424 L155.903 855.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M184.445 858.604 Q180.834 858.604 179.005 862.169 Q177.2 865.711 177.2 872.84 Q177.2 879.947 179.005 883.512 Q180.834 887.053 184.445 887.053 Q188.079 887.053 189.885 883.512 Q191.713 879.947 191.713 872.84 Q191.713 865.711 189.885 862.169 Q188.079 858.604 184.445 858.604 M184.445 854.901 Q190.255 854.901 193.311 859.507 Q196.389 864.09 196.389 872.84 Q196.389 881.567 193.311 886.174 Q190.255 890.757 184.445 890.757 Q178.635 890.757 175.556 886.174 Q172.501 881.567 172.501 872.84 Q172.501 864.09 175.556 859.507 Q178.635 854.901 184.445 854.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M154.862 595.755 Q151.714 595.755 149.862 597.908 Q148.033 600.061 148.033 603.81 Q148.033 607.537 149.862 609.713 Q151.714 611.866 154.862 611.866 Q158.01 611.866 159.839 609.713 Q161.69 607.537 161.69 603.81 Q161.69 600.061 159.839 597.908 Q158.01 595.755 154.862 595.755 M164.144 581.102 L164.144 585.362 Q162.385 584.528 160.579 584.088 Q158.797 583.649 157.038 583.649 Q152.408 583.649 149.954 586.774 Q147.524 589.899 147.177 596.218 Q148.542 594.204 150.603 593.139 Q152.663 592.051 155.14 592.051 Q160.348 592.051 163.357 595.223 Q166.39 598.371 166.39 603.81 Q166.39 609.135 163.241 612.352 Q160.093 615.57 154.862 615.57 Q148.866 615.57 145.695 610.986 Q142.524 606.38 142.524 597.653 Q142.524 589.459 146.413 584.598 Q150.302 579.713 156.853 579.713 Q158.612 579.713 160.394 580.061 Q162.2 580.408 164.144 581.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M184.445 583.417 Q180.834 583.417 179.005 586.982 Q177.2 590.524 177.2 597.653 Q177.2 604.76 179.005 608.324 Q180.834 611.866 184.445 611.866 Q188.079 611.866 189.885 608.324 Q191.713 604.76 191.713 597.653 Q191.713 590.524 189.885 586.982 Q188.079 583.417 184.445 583.417 M184.445 579.713 Q190.255 579.713 193.311 584.32 Q196.389 588.903 196.389 597.653 Q196.389 606.38 193.311 610.986 Q190.255 615.57 184.445 615.57 Q178.635 615.57 175.556 610.986 Q172.501 606.38 172.501 597.653 Q172.501 588.903 175.556 584.32 Q178.635 579.713 184.445 579.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M154.283 323.299 Q150.95 323.299 149.028 325.082 Q147.13 326.864 147.13 329.989 Q147.13 333.114 149.028 334.896 Q150.95 336.679 154.283 336.679 Q157.616 336.679 159.538 334.896 Q161.459 333.091 161.459 329.989 Q161.459 326.864 159.538 325.082 Q157.64 323.299 154.283 323.299 M149.607 321.309 Q146.598 320.568 144.908 318.508 Q143.241 316.447 143.241 313.485 Q143.241 309.341 146.181 306.934 Q149.144 304.526 154.283 304.526 Q159.445 304.526 162.385 306.934 Q165.325 309.341 165.325 313.485 Q165.325 316.447 163.635 318.508 Q161.968 320.568 158.982 321.309 Q162.362 322.096 164.237 324.387 Q166.135 326.679 166.135 329.989 Q166.135 335.012 163.056 337.697 Q160.001 340.383 154.283 340.383 Q148.566 340.383 145.487 337.697 Q142.431 335.012 142.431 329.989 Q142.431 326.679 144.329 324.387 Q146.228 322.096 149.607 321.309 M147.894 313.924 Q147.894 316.61 149.561 318.114 Q151.251 319.619 154.283 319.619 Q157.292 319.619 158.982 318.114 Q160.695 316.61 160.695 313.924 Q160.695 311.239 158.982 309.735 Q157.292 308.23 154.283 308.23 Q151.251 308.23 149.561 309.735 Q147.894 311.239 147.894 313.924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M184.445 308.23 Q180.834 308.23 179.005 311.795 Q177.2 315.336 177.2 322.466 Q177.2 329.572 179.005 333.137 Q180.834 336.679 184.445 336.679 Q188.079 336.679 189.885 333.137 Q191.713 329.572 191.713 322.466 Q191.713 315.336 189.885 311.795 Q188.079 308.23 184.445 308.23 M184.445 304.526 Q190.255 304.526 193.311 309.133 Q196.389 313.716 196.389 322.466 Q196.389 331.193 193.311 335.799 Q190.255 340.383 184.445 340.383 Q178.635 340.383 175.556 335.799 Q172.501 331.193 172.501 322.466 Q172.501 313.716 175.556 309.133 Q178.635 304.526 184.445 304.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M114.931 60.5889 L122.57 60.5889 L122.57 34.2233 L114.26 35.89 L114.26 31.6308 L122.524 29.9641 L127.2 29.9641 L127.2 60.5889 L134.839 60.5889 L134.839 64.5241 L114.931 64.5241 L114.931 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M154.283 33.0428 Q150.672 33.0428 148.843 36.6076 Q147.038 40.1492 147.038 47.2788 Q147.038 54.3853 148.843 57.9501 Q150.672 61.4917 154.283 61.4917 Q157.917 61.4917 159.723 57.9501 Q161.552 54.3853 161.552 47.2788 Q161.552 40.1492 159.723 36.6076 Q157.917 33.0428 154.283 33.0428 M154.283 29.3391 Q160.093 29.3391 163.149 33.9456 Q166.227 38.5289 166.227 47.2788 Q166.227 56.0056 163.149 60.6121 Q160.093 65.1954 154.283 65.1954 Q148.473 65.1954 145.394 60.6121 Q142.339 56.0056 142.339 47.2788 Q142.339 38.5289 145.394 33.9456 Q148.473 29.3391 154.283 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M184.445 33.0428 Q180.834 33.0428 179.005 36.6076 Q177.2 40.1492 177.2 47.2788 Q177.2 54.3853 179.005 57.9501 Q180.834 61.4917 184.445 61.4917 Q188.079 61.4917 189.885 57.9501 Q191.713 54.3853 191.713 47.2788 Q191.713 40.1492 189.885 36.6076 Q188.079 33.0428 184.445 33.0428 M184.445 29.3391 Q190.255 29.3391 193.311 33.9456 Q196.389 38.5289 196.389 47.2788 Q196.389 56.0056 193.311 60.6121 Q190.255 65.1954 184.445 65.1954 Q178.635 65.1954 175.556 60.6121 Q172.501 56.0056 172.501 47.2788 Q172.501 38.5289 175.556 33.9456 Q178.635 29.3391 184.445 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M41.7242 880.191 Q42.4244 878.122 44.7161 876.181 Q47.0077 874.207 51.0181 872.234 L64.0042 865.709 L64.0042 872.616 L51.8138 878.695 Q47.0395 881.05 45.48 883.278 Q43.9204 885.474 43.9204 889.294 L43.9204 896.296 L64.0042 896.296 L64.0042 902.726 L16.4842 902.726 L16.4842 888.212 Q16.4842 880.064 19.8898 876.053 Q23.2955 872.043 30.1704 872.043 Q34.6582 872.043 37.6183 874.144 Q40.5784 876.212 41.7242 880.191 M21.7677 896.296 L38.6368 896.296 L38.6368 888.212 Q38.6368 883.565 36.5043 881.209 Q34.34 878.822 30.1704 878.822 Q26.0009 878.822 23.9002 881.209 Q21.7677 883.565 21.7677 888.212 L21.7677 896.296 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M32.4621 846.803 Q32.4621 851.513 36.1542 854.251 Q39.8145 856.988 46.212 856.988 Q52.6095 856.988 56.3017 854.283 Q59.9619 851.545 59.9619 846.803 Q59.9619 842.124 56.2698 839.387 Q52.5777 836.65 46.212 836.65 Q39.8781 836.65 36.186 839.387 Q32.4621 842.124 32.4621 846.803 M27.4968 846.803 Q27.4968 839.164 32.4621 834.803 Q37.4273 830.443 46.212 830.443 Q54.9649 830.443 59.9619 834.803 Q64.9272 839.164 64.9272 846.803 Q64.9272 854.473 59.9619 858.834 Q54.9649 863.163 46.212 863.163 Q37.4273 863.163 32.4621 858.834 Q27.4968 854.473 27.4968 846.803 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M14.479 820.735 L14.479 814.879 L64.0042 814.879 L64.0042 820.735 L14.479 820.735 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M14.479 802.625 L14.479 796.768 L64.0042 796.768 L64.0042 802.625 L14.479 802.625 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M14.479 763.794 L14.479 757.938 L64.0042 757.938 L64.0042 763.794 L14.479 763.794 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M44.7161 715.192 L47.5806 715.192 L47.5806 742.119 Q53.6281 741.737 56.8109 738.49 Q59.9619 735.212 59.9619 729.387 Q59.9619 726.014 59.1344 722.862 Q58.3069 719.68 56.6518 716.56 L62.1899 716.56 Q63.5267 719.711 64.227 723.022 Q64.9272 726.332 64.9272 729.737 Q64.9272 738.268 59.9619 743.265 Q54.9967 748.23 46.5303 748.23 Q37.7774 748.23 32.6531 743.519 Q27.4968 738.777 27.4968 730.756 Q27.4968 723.563 32.1438 719.393 Q36.7589 715.192 44.7161 715.192 M42.9973 721.048 Q38.1912 721.112 35.3266 723.754 Q32.4621 726.364 32.4621 730.692 Q32.4621 735.594 35.2312 738.554 Q38.0002 741.482 43.0292 741.928 L42.9973 721.048 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M42.4881 675.947 L64.0042 675.947 L64.0042 681.804 L42.679 681.804 Q37.6183 681.804 35.1038 683.777 Q32.5894 685.75 32.5894 689.697 Q32.5894 694.44 35.6131 697.177 Q38.6368 699.914 43.8567 699.914 L64.0042 699.914 L64.0042 705.802 L28.3562 705.802 L28.3562 699.914 L33.8944 699.914 Q30.6797 697.813 29.0883 694.981 Q27.4968 692.116 27.4968 688.392 Q27.4968 682.249 31.3163 679.098 Q35.1038 675.947 42.4881 675.947 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M45.7664 640.809 Q39.4007 640.809 35.8996 643.45 Q32.3984 646.06 32.3984 650.803 Q32.3984 655.513 35.8996 658.155 Q39.4007 660.765 45.7664 660.765 Q52.1003 660.765 55.6014 658.155 Q59.1026 655.513 59.1026 650.803 Q59.1026 646.06 55.6014 643.45 Q52.1003 640.809 45.7664 640.809 M59.58 634.952 Q68.683 634.952 73.1071 638.994 Q77.5631 643.037 77.5631 651.376 Q77.5631 654.463 77.0857 657.2 Q76.6401 659.937 75.6852 662.516 L69.9879 662.516 Q71.3884 659.937 72.0568 657.423 Q72.7252 654.909 72.7252 652.299 Q72.7252 646.538 69.7015 643.673 Q66.7096 640.809 60.6303 640.809 L57.7339 640.809 Q60.885 642.623 62.4446 645.456 Q64.0042 648.288 64.0042 652.235 Q64.0042 658.792 59.0071 662.802 Q54.01 666.812 45.7664 666.812 Q37.491 666.812 32.4939 662.802 Q27.4968 658.792 27.4968 652.235 Q27.4968 648.288 29.0564 645.456 Q30.616 642.623 33.7671 640.809 L28.3562 640.809 L28.3562 634.952 L59.58 634.952 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M18.2347 617.096 L28.3562 617.096 L28.3562 605.033 L32.9077 605.033 L32.9077 617.096 L52.2594 617.096 Q56.6199 617.096 57.8613 615.919 Q59.1026 614.709 59.1026 611.049 L59.1026 605.033 L64.0042 605.033 L64.0042 611.049 Q64.0042 617.828 61.4897 620.406 Q58.9434 622.985 52.2594 622.985 L32.9077 622.985 L32.9077 627.281 L28.3562 627.281 L28.3562 622.985 L18.2347 622.985 L18.2347 617.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip290)" d="M42.4881 567.698 L64.0042 567.698 L64.0042 573.555 L42.679 573.555 Q37.6183 573.555 35.1038 575.528 Q32.5894 577.502 32.5894 581.448 Q32.5894 586.191 35.6131 588.928 Q38.6368 591.665 43.8567 591.665 L64.0042 591.665 L64.0042 597.554 L14.479 597.554 L14.479 591.665 L33.8944 591.665 Q30.6797 589.565 29.0883 586.732 Q27.4968 583.867 27.4968 580.143 Q27.4968 574 31.3163 570.849 Q35.1038 567.698 42.4881 567.698 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip292)" d="M273.619 391.228 L273.619 1423.18 L308.958 1423.18 L308.958 391.228 L273.619 391.228 L273.619 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="273.619,391.228 273.619,1423.18 308.958,1423.18 308.958,391.228 273.619,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="291.288" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M332.518 391.228 L332.518 1423.18 L367.857 1423.18 L367.857 391.228 L332.518 391.228 L332.518 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="332.518,391.228 332.518,1423.18 367.857,1423.18 367.857,391.228 332.518,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="350.187" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M391.417 391.228 L391.417 1423.18 L426.756 1423.18 L426.756 391.228 L391.417 391.228 L391.417 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="391.417,391.228 391.417,1423.18 426.756,1423.18 426.756,391.228 391.417,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="409.087" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M450.316 391.228 L450.316 1423.18 L485.655 1423.18 L485.655 391.228 L450.316 391.228 L450.316 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="450.316,391.228 450.316,1423.18 485.655,1423.18 485.655,391.228 450.316,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="467.986" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M509.215 391.228 L509.215 1423.18 L544.554 1423.18 L544.554 391.228 L509.215 391.228 L509.215 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="509.215,391.228 509.215,1423.18 544.554,1423.18 544.554,391.228 509.215,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="526.885" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M568.114 682.926 L568.114 1423.18 L603.453 1423.18 L603.453 682.926 L568.114 682.926 L568.114 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="568.114,682.926 568.114,1423.18 603.453,1423.18 603.453,682.926 568.114,682.926 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="585.784" cy="682.926" r="2"/>
+<path clip-path="url(#clip292)" d="M627.013 693.934 L627.013 1423.18 L662.353 1423.18 L662.353 693.934 L627.013 693.934 L627.013 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="627.013,693.934 627.013,1423.18 662.353,1423.18 662.353,693.934 627.013,693.934 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="644.683" cy="693.934" r="2"/>
+<path clip-path="url(#clip292)" d="M685.912 721.453 L685.912 1423.18 L721.252 1423.18 L721.252 721.453 L685.912 721.453 L685.912 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="685.912,721.453 685.912,1423.18 721.252,1423.18 721.252,721.453 685.912,721.453 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="703.582" cy="721.453" r="2"/>
+<path clip-path="url(#clip292)" d="M744.811 732.46 L744.811 1423.18 L780.151 1423.18 L780.151 732.46 L744.811 732.46 L744.811 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="744.811,732.46 744.811,1423.18 780.151,1423.18 780.151,732.46 744.811,732.46 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="762.481" cy="732.46" r="2"/>
+<path clip-path="url(#clip292)" d="M803.71 94.0259 L803.71 1423.18 L839.05 1423.18 L839.05 94.0259 L803.71 94.0259 L803.71 94.0259  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,94.0259 803.71,1423.18 839.05,1423.18 839.05,94.0259 803.71,94.0259 "/>
+<path clip-path="url(#clip292)" d="M803.71 537.077 L803.71 1423.18 L839.05 1423.18 L839.05 537.077 L803.71 537.077 L803.71 537.077  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,537.077 803.71,1423.18 839.05,1423.18 839.05,537.077 803.71,537.077 "/>
+<path clip-path="url(#clip292)" d="M803.71 980.129 L803.71 1423.18 L839.05 1423.18 L839.05 980.129 L803.71 980.129 L803.71 980.129  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="803.71,980.129 803.71,1423.18 839.05,1423.18 839.05,980.129 803.71,980.129 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="94.0259" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="537.077" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="821.38" cy="980.129" r="2"/>
+<path clip-path="url(#clip292)" d="M862.609 151.815 L862.609 1423.18 L897.949 1423.18 L897.949 151.815 L862.609 151.815 L862.609 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,151.815 862.609,1423.18 897.949,1423.18 897.949,151.815 862.609,151.815 "/>
+<path clip-path="url(#clip292)" d="M862.609 575.603 L862.609 1423.18 L897.949 1423.18 L897.949 575.603 L862.609 575.603 L862.609 575.603  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,575.603 862.609,1423.18 897.949,1423.18 897.949,575.603 862.609,575.603 "/>
+<path clip-path="url(#clip292)" d="M862.609 999.392 L862.609 1423.18 L897.949 1423.18 L897.949 999.392 L862.609 999.392 L862.609 999.392  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="862.609,999.392 862.609,1423.18 897.949,1423.18 897.949,999.392 862.609,999.392 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="151.815" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="575.603" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="880.279" cy="999.392" r="2"/>
+<path clip-path="url(#clip292)" d="M921.508 193.093 L921.508 1423.18 L956.848 1423.18 L956.848 193.093 L921.508 193.093 L921.508 193.093  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,193.093 921.508,1423.18 956.848,1423.18 956.848,193.093 921.508,193.093 "/>
+<path clip-path="url(#clip292)" d="M921.508 603.122 L921.508 1423.18 L956.848 1423.18 L956.848 603.122 L921.508 603.122 L921.508 603.122  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,603.122 921.508,1423.18 956.848,1423.18 956.848,603.122 921.508,603.122 "/>
+<path clip-path="url(#clip292)" d="M921.508 1013.15 L921.508 1423.18 L956.848 1423.18 L956.848 1013.15 L921.508 1013.15 L921.508 1013.15  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="921.508,1013.15 921.508,1423.18 956.848,1423.18 956.848,1013.15 921.508,1013.15 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="193.093" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="603.122" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="1013.15" r="2"/>
+<path clip-path="url(#clip292)" d="M980.408 316.928 L980.408 1423.18 L1015.75 1423.18 L1015.75 316.928 L980.408 316.928 L980.408 316.928  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,316.928 980.408,1423.18 1015.75,1423.18 1015.75,316.928 980.408,316.928 "/>
+<path clip-path="url(#clip292)" d="M980.408 593.491 L980.408 1423.18 L1015.75 1423.18 L1015.75 593.491 L980.408 593.491 L980.408 593.491  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,593.491 980.408,1423.18 1015.75,1423.18 1015.75,593.491 980.408,593.491 "/>
+<path clip-path="url(#clip292)" d="M980.408 870.054 L980.408 1423.18 L1015.75 1423.18 L1015.75 870.054 L980.408 870.054 L980.408 870.054  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,870.054 980.408,1423.18 1015.75,1423.18 1015.75,870.054 980.408,870.054 "/>
+<path clip-path="url(#clip292)" d="M980.408 1146.62 L980.408 1423.18 L1015.75 1423.18 L1015.75 1146.62 L980.408 1146.62 L980.408 1146.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="980.408,1146.62 980.408,1423.18 1015.75,1423.18 1015.75,1146.62 980.408,1146.62 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="316.928" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="593.491" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="870.054" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="998.077" cy="1146.62" r="2"/>
+<path clip-path="url(#clip292)" d="M1039.31 85.7703 L1039.31 1423.18 L1074.65 1423.18 L1074.65 85.7703 L1039.31 85.7703 L1039.31 85.7703  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,85.7703 1039.31,1423.18 1074.65,1423.18 1074.65,85.7703 1039.31,85.7703 "/>
+<path clip-path="url(#clip292)" d="M1039.31 308.672 L1039.31 1423.18 L1074.65 1423.18 L1074.65 308.672 L1039.31 308.672 L1039.31 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,308.672 1039.31,1423.18 1074.65,1423.18 1074.65,308.672 1039.31,308.672 "/>
+<path clip-path="url(#clip292)" d="M1039.31 531.573 L1039.31 1423.18 L1074.65 1423.18 L1074.65 531.573 L1039.31 531.573 L1039.31 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,531.573 1039.31,1423.18 1074.65,1423.18 1074.65,531.573 1039.31,531.573 "/>
+<path clip-path="url(#clip292)" d="M1039.31 754.475 L1039.31 1423.18 L1074.65 1423.18 L1074.65 754.475 L1039.31 754.475 L1039.31 754.475  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,754.475 1039.31,1423.18 1074.65,1423.18 1074.65,754.475 1039.31,754.475 "/>
+<path clip-path="url(#clip292)" d="M1039.31 977.377 L1039.31 1423.18 L1074.65 1423.18 L1074.65 977.377 L1039.31 977.377 L1039.31 977.377  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,977.377 1039.31,1423.18 1074.65,1423.18 1074.65,977.377 1039.31,977.377 "/>
+<path clip-path="url(#clip292)" d="M1039.31 1200.28 L1039.31 1423.18 L1074.65 1423.18 L1074.65 1200.28 L1039.31 1200.28 L1039.31 1200.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1039.31,1200.28 1039.31,1423.18 1074.65,1423.18 1074.65,1200.28 1039.31,1200.28 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="85.7703" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="308.672" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="531.573" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="754.475" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="977.377" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1056.98" cy="1200.28" r="2"/>
+<path clip-path="url(#clip292)" d="M1098.21 226.116 L1098.21 1423.18 L1133.55 1423.18 L1133.55 226.116 L1098.21 226.116 L1098.21 226.116  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,226.116 1098.21,1423.18 1133.55,1423.18 1133.55,226.116 1098.21,226.116 "/>
+<path clip-path="url(#clip292)" d="M1098.21 425.626 L1098.21 1423.18 L1133.55 1423.18 L1133.55 425.626 L1098.21 425.626 L1098.21 425.626  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,425.626 1098.21,1423.18 1133.55,1423.18 1133.55,425.626 1098.21,425.626 "/>
+<path clip-path="url(#clip292)" d="M1098.21 625.137 L1098.21 1423.18 L1133.55 1423.18 L1133.55 625.137 L1098.21 625.137 L1098.21 625.137  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,625.137 1098.21,1423.18 1133.55,1423.18 1133.55,625.137 1098.21,625.137 "/>
+<path clip-path="url(#clip292)" d="M1098.21 824.648 L1098.21 1423.18 L1133.55 1423.18 L1133.55 824.648 L1098.21 824.648 L1098.21 824.648  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,824.648 1098.21,1423.18 1133.55,1423.18 1133.55,824.648 1098.21,824.648 "/>
+<path clip-path="url(#clip292)" d="M1098.21 1024.16 L1098.21 1423.18 L1133.55 1423.18 L1133.55 1024.16 L1098.21 1024.16 L1098.21 1024.16  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,1024.16 1098.21,1423.18 1133.55,1423.18 1133.55,1024.16 1098.21,1024.16 "/>
+<path clip-path="url(#clip292)" d="M1098.21 1223.67 L1098.21 1423.18 L1133.55 1423.18 L1133.55 1223.67 L1098.21 1223.67 L1098.21 1223.67  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.21,1223.67 1098.21,1423.18 1133.55,1423.18 1133.55,1223.67 1098.21,1223.67 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="226.116" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="425.626" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="625.137" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="824.648" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="1024.16" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1115.88" cy="1223.67" r="2"/>
+<path clip-path="url(#clip292)" d="M1157.1 61.0035 L1157.1 1423.18 L1192.44 1423.18 L1192.44 61.0035 L1157.1 61.0035 L1157.1 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,61.0035 1157.1,1423.18 1192.44,1423.18 1192.44,61.0035 1157.1,61.0035 "/>
+<path clip-path="url(#clip292)" d="M1157.1 212.356 L1157.1 1423.18 L1192.44 1423.18 L1192.44 212.356 L1157.1 212.356 L1157.1 212.356  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,212.356 1157.1,1423.18 1192.44,1423.18 1192.44,212.356 1157.1,212.356 "/>
+<path clip-path="url(#clip292)" d="M1157.1 363.709 L1157.1 1423.18 L1192.44 1423.18 L1192.44 363.709 L1157.1 363.709 L1157.1 363.709  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,363.709 1157.1,1423.18 1192.44,1423.18 1192.44,363.709 1157.1,363.709 "/>
+<path clip-path="url(#clip292)" d="M1157.1 515.062 L1157.1 1423.18 L1192.44 1423.18 L1192.44 515.062 L1157.1 515.062 L1157.1 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,515.062 1157.1,1423.18 1192.44,1423.18 1192.44,515.062 1157.1,515.062 "/>
+<path clip-path="url(#clip292)" d="M1157.1 666.415 L1157.1 1423.18 L1192.44 1423.18 L1192.44 666.415 L1157.1 666.415 L1157.1 666.415  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,666.415 1157.1,1423.18 1192.44,1423.18 1192.44,666.415 1157.1,666.415 "/>
+<path clip-path="url(#clip292)" d="M1157.1 817.768 L1157.1 1423.18 L1192.44 1423.18 L1192.44 817.768 L1157.1 817.768 L1157.1 817.768  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,817.768 1157.1,1423.18 1192.44,1423.18 1192.44,817.768 1157.1,817.768 "/>
+<path clip-path="url(#clip292)" d="M1157.1 969.121 L1157.1 1423.18 L1192.44 1423.18 L1192.44 969.121 L1157.1 969.121 L1157.1 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,969.121 1157.1,1423.18 1192.44,1423.18 1192.44,969.121 1157.1,969.121 "/>
+<path clip-path="url(#clip292)" d="M1157.1 1120.47 L1157.1 1423.18 L1192.44 1423.18 L1192.44 1120.47 L1157.1 1120.47 L1157.1 1120.47  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,1120.47 1157.1,1423.18 1192.44,1423.18 1192.44,1120.47 1157.1,1120.47 "/>
+<path clip-path="url(#clip292)" d="M1157.1 1271.83 L1157.1 1423.18 L1192.44 1423.18 L1192.44 1271.83 L1157.1 1271.83 L1157.1 1271.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1157.1,1271.83 1157.1,1423.18 1192.44,1423.18 1192.44,1271.83 1157.1,1271.83 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="61.0035" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="212.356" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="363.709" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="515.062" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="666.415" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="817.768" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="969.121" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="1120.47" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1174.77" cy="1271.83" r="2"/>
+<path clip-path="url(#clip292)" d="M1216 121.545 L1216 1423.18 L1251.34 1423.18 L1251.34 121.545 L1216 121.545 L1216 121.545  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,121.545 1216,1423.18 1251.34,1423.18 1251.34,121.545 1216,121.545 "/>
+<path clip-path="url(#clip292)" d="M1216 239.875 L1216 1423.18 L1251.34 1423.18 L1251.34 239.875 L1216 239.875 L1216 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,239.875 1216,1423.18 1251.34,1423.18 1251.34,239.875 1216,239.875 "/>
+<path clip-path="url(#clip292)" d="M1216 358.206 L1216 1423.18 L1251.34 1423.18 L1251.34 358.206 L1216 358.206 L1216 358.206  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,358.206 1216,1423.18 1251.34,1423.18 1251.34,358.206 1216,358.206 "/>
+<path clip-path="url(#clip292)" d="M1216 476.536 L1216 1423.18 L1251.34 1423.18 L1251.34 476.536 L1216 476.536 L1216 476.536  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,476.536 1216,1423.18 1251.34,1423.18 1251.34,476.536 1216,476.536 "/>
+<path clip-path="url(#clip292)" d="M1216 594.867 L1216 1423.18 L1251.34 1423.18 L1251.34 594.867 L1216 594.867 L1216 594.867  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,594.867 1216,1423.18 1251.34,1423.18 1251.34,594.867 1216,594.867 "/>
+<path clip-path="url(#clip292)" d="M1216 713.197 L1216 1423.18 L1251.34 1423.18 L1251.34 713.197 L1216 713.197 L1216 713.197  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,713.197 1216,1423.18 1251.34,1423.18 1251.34,713.197 1216,713.197 "/>
+<path clip-path="url(#clip292)" d="M1216 831.527 L1216 1423.18 L1251.34 1423.18 L1251.34 831.527 L1216 831.527 L1216 831.527  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,831.527 1216,1423.18 1251.34,1423.18 1251.34,831.527 1216,831.527 "/>
+<path clip-path="url(#clip292)" d="M1216 949.858 L1216 1423.18 L1251.34 1423.18 L1251.34 949.858 L1216 949.858 L1216 949.858  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,949.858 1216,1423.18 1251.34,1423.18 1251.34,949.858 1216,949.858 "/>
+<path clip-path="url(#clip292)" d="M1216 1068.19 L1216 1423.18 L1251.34 1423.18 L1251.34 1068.19 L1216 1068.19 L1216 1068.19  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1068.19 1216,1423.18 1251.34,1423.18 1251.34,1068.19 1216,1068.19 "/>
+<path clip-path="url(#clip292)" d="M1216 1186.52 L1216 1423.18 L1251.34 1423.18 L1251.34 1186.52 L1216 1186.52 L1216 1186.52  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1186.52 1216,1423.18 1251.34,1423.18 1251.34,1186.52 1216,1186.52 "/>
+<path clip-path="url(#clip292)" d="M1216 1304.85 L1216 1423.18 L1251.34 1423.18 L1251.34 1304.85 L1216 1304.85 L1216 1304.85  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1216,1304.85 1216,1423.18 1251.34,1423.18 1251.34,1304.85 1216,1304.85 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="121.545" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="239.875" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="358.206" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="476.536" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="594.867" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="713.197" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="831.527" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="949.858" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1068.19" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1186.52" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1233.67" cy="1304.85" r="2"/>
+<path clip-path="url(#clip292)" d="M1274.9 69.2591 L1274.9 1423.18 L1310.24 1423.18 L1310.24 69.2591 L1274.9 69.2591 L1274.9 69.2591  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,69.2591 1274.9,1423.18 1310.24,1423.18 1310.24,69.2591 1274.9,69.2591 "/>
+<path clip-path="url(#clip292)" d="M1274.9 182.086 L1274.9 1423.18 L1310.24 1423.18 L1310.24 182.086 L1274.9 182.086 L1274.9 182.086  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,182.086 1274.9,1423.18 1310.24,1423.18 1310.24,182.086 1274.9,182.086 "/>
+<path clip-path="url(#clip292)" d="M1274.9 294.913 L1274.9 1423.18 L1310.24 1423.18 L1310.24 294.913 L1274.9 294.913 L1274.9 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,294.913 1274.9,1423.18 1310.24,1423.18 1310.24,294.913 1274.9,294.913 "/>
+<path clip-path="url(#clip292)" d="M1274.9 407.739 L1274.9 1423.18 L1310.24 1423.18 L1310.24 407.739 L1274.9 407.739 L1274.9 407.739  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,407.739 1274.9,1423.18 1310.24,1423.18 1310.24,407.739 1274.9,407.739 "/>
+<path clip-path="url(#clip292)" d="M1274.9 520.566 L1274.9 1423.18 L1310.24 1423.18 L1310.24 520.566 L1274.9 520.566 L1274.9 520.566  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,520.566 1274.9,1423.18 1310.24,1423.18 1310.24,520.566 1274.9,520.566 "/>
+<path clip-path="url(#clip292)" d="M1274.9 633.393 L1274.9 1423.18 L1310.24 1423.18 L1310.24 633.393 L1274.9 633.393 L1274.9 633.393  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,633.393 1274.9,1423.18 1310.24,1423.18 1310.24,633.393 1274.9,633.393 "/>
+<path clip-path="url(#clip292)" d="M1274.9 746.219 L1274.9 1423.18 L1310.24 1423.18 L1310.24 746.219 L1274.9 746.219 L1274.9 746.219  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,746.219 1274.9,1423.18 1310.24,1423.18 1310.24,746.219 1274.9,746.219 "/>
+<path clip-path="url(#clip292)" d="M1274.9 859.046 L1274.9 1423.18 L1310.24 1423.18 L1310.24 859.046 L1274.9 859.046 L1274.9 859.046  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,859.046 1274.9,1423.18 1310.24,1423.18 1310.24,859.046 1274.9,859.046 "/>
+<path clip-path="url(#clip292)" d="M1274.9 971.873 L1274.9 1423.18 L1310.24 1423.18 L1310.24 971.873 L1274.9 971.873 L1274.9 971.873  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,971.873 1274.9,1423.18 1310.24,1423.18 1310.24,971.873 1274.9,971.873 "/>
+<path clip-path="url(#clip292)" d="M1274.9 1084.7 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1084.7 L1274.9 1084.7 L1274.9 1084.7  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1084.7 1274.9,1423.18 1310.24,1423.18 1310.24,1084.7 1274.9,1084.7 "/>
+<path clip-path="url(#clip292)" d="M1274.9 1197.53 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1197.53 L1274.9 1197.53 L1274.9 1197.53  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1197.53 1274.9,1423.18 1310.24,1423.18 1310.24,1197.53 1274.9,1197.53 "/>
+<path clip-path="url(#clip292)" d="M1274.9 1310.35 L1274.9 1423.18 L1310.24 1423.18 L1310.24 1310.35 L1274.9 1310.35 L1274.9 1310.35  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1274.9,1310.35 1274.9,1423.18 1310.24,1423.18 1310.24,1310.35 1274.9,1310.35 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="69.2591" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="182.086" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="294.913" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="407.739" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="520.566" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="633.393" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="746.219" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="859.046" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="971.873" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1084.7" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1197.53" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1292.57" cy="1310.35" r="2"/>
+<path clip-path="url(#clip292)" d="M1333.8 61.0035 L1333.8 1423.18 L1369.14 1423.18 L1369.14 61.0035 L1333.8 61.0035 L1333.8 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,61.0035 1333.8,1423.18 1369.14,1423.18 1369.14,61.0035 1333.8,61.0035 "/>
+<path clip-path="url(#clip292)" d="M1333.8 151.815 L1333.8 1423.18 L1369.14 1423.18 L1369.14 151.815 L1333.8 151.815 L1333.8 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,151.815 1333.8,1423.18 1369.14,1423.18 1369.14,151.815 1333.8,151.815 "/>
+<path clip-path="url(#clip292)" d="M1333.8 242.627 L1333.8 1423.18 L1369.14 1423.18 L1369.14 242.627 L1333.8 242.627 L1333.8 242.627  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,242.627 1333.8,1423.18 1369.14,1423.18 1369.14,242.627 1333.8,242.627 "/>
+<path clip-path="url(#clip292)" d="M1333.8 333.439 L1333.8 1423.18 L1369.14 1423.18 L1369.14 333.439 L1333.8 333.439 L1333.8 333.439  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,333.439 1333.8,1423.18 1369.14,1423.18 1369.14,333.439 1333.8,333.439 "/>
+<path clip-path="url(#clip292)" d="M1333.8 424.25 L1333.8 1423.18 L1369.14 1423.18 L1369.14 424.25 L1333.8 424.25 L1333.8 424.25  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,424.25 1333.8,1423.18 1369.14,1423.18 1369.14,424.25 1333.8,424.25 "/>
+<path clip-path="url(#clip292)" d="M1333.8 515.062 L1333.8 1423.18 L1369.14 1423.18 L1369.14 515.062 L1333.8 515.062 L1333.8 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,515.062 1333.8,1423.18 1369.14,1423.18 1369.14,515.062 1333.8,515.062 "/>
+<path clip-path="url(#clip292)" d="M1333.8 605.874 L1333.8 1423.18 L1369.14 1423.18 L1369.14 605.874 L1333.8 605.874 L1333.8 605.874  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,605.874 1333.8,1423.18 1369.14,1423.18 1369.14,605.874 1333.8,605.874 "/>
+<path clip-path="url(#clip292)" d="M1333.8 696.686 L1333.8 1423.18 L1369.14 1423.18 L1369.14 696.686 L1333.8 696.686 L1333.8 696.686  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,696.686 1333.8,1423.18 1369.14,1423.18 1369.14,696.686 1333.8,696.686 "/>
+<path clip-path="url(#clip292)" d="M1333.8 787.498 L1333.8 1423.18 L1369.14 1423.18 L1369.14 787.498 L1333.8 787.498 L1333.8 787.498  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,787.498 1333.8,1423.18 1369.14,1423.18 1369.14,787.498 1333.8,787.498 "/>
+<path clip-path="url(#clip292)" d="M1333.8 878.309 L1333.8 1423.18 L1369.14 1423.18 L1369.14 878.309 L1333.8 878.309 L1333.8 878.309  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,878.309 1333.8,1423.18 1369.14,1423.18 1369.14,878.309 1333.8,878.309 "/>
+<path clip-path="url(#clip292)" d="M1333.8 969.121 L1333.8 1423.18 L1369.14 1423.18 L1369.14 969.121 L1333.8 969.121 L1333.8 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,969.121 1333.8,1423.18 1369.14,1423.18 1369.14,969.121 1333.8,969.121 "/>
+<path clip-path="url(#clip292)" d="M1333.8 1059.93 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1059.93 L1333.8 1059.93 L1333.8 1059.93  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1059.93 1333.8,1423.18 1369.14,1423.18 1369.14,1059.93 1333.8,1059.93 "/>
+<path clip-path="url(#clip292)" d="M1333.8 1150.74 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1150.74 L1333.8 1150.74 L1333.8 1150.74  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1150.74 1333.8,1423.18 1369.14,1423.18 1369.14,1150.74 1333.8,1150.74 "/>
+<path clip-path="url(#clip292)" d="M1333.8 1241.56 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1241.56 L1333.8 1241.56 L1333.8 1241.56  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1241.56 1333.8,1423.18 1369.14,1423.18 1369.14,1241.56 1333.8,1241.56 "/>
+<path clip-path="url(#clip292)" d="M1333.8 1332.37 L1333.8 1423.18 L1369.14 1423.18 L1369.14 1332.37 L1333.8 1332.37 L1333.8 1332.37  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1333.8,1332.37 1333.8,1423.18 1369.14,1423.18 1369.14,1332.37 1333.8,1332.37 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="61.0035" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="151.815" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="242.627" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="333.439" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="424.25" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="515.062" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="605.874" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="696.686" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="787.498" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="878.309" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="969.121" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1059.93" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1150.74" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1241.56" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1351.47" cy="1332.37" r="2"/>
+<path clip-path="url(#clip292)" d="M1392.7 89.8981 L1392.7 1423.18 L1428.04 1423.18 L1428.04 89.8981 L1392.7 89.8981 L1392.7 89.8981  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,89.8981 1392.7,1423.18 1428.04,1423.18 1428.04,89.8981 1392.7,89.8981 "/>
+<path clip-path="url(#clip292)" d="M1392.7 160.071 L1392.7 1423.18 L1428.04 1423.18 L1428.04 160.071 L1392.7 160.071 L1392.7 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,160.071 1392.7,1423.18 1428.04,1423.18 1428.04,160.071 1392.7,160.071 "/>
+<path clip-path="url(#clip292)" d="M1392.7 230.244 L1392.7 1423.18 L1428.04 1423.18 L1428.04 230.244 L1392.7 230.244 L1392.7 230.244  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,230.244 1392.7,1423.18 1428.04,1423.18 1428.04,230.244 1392.7,230.244 "/>
+<path clip-path="url(#clip292)" d="M1392.7 300.416 L1392.7 1423.18 L1428.04 1423.18 L1428.04 300.416 L1392.7 300.416 L1392.7 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,300.416 1392.7,1423.18 1428.04,1423.18 1428.04,300.416 1392.7,300.416 "/>
+<path clip-path="url(#clip292)" d="M1392.7 370.589 L1392.7 1423.18 L1428.04 1423.18 L1428.04 370.589 L1392.7 370.589 L1392.7 370.589  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,370.589 1392.7,1423.18 1428.04,1423.18 1428.04,370.589 1392.7,370.589 "/>
+<path clip-path="url(#clip292)" d="M1392.7 440.762 L1392.7 1423.18 L1428.04 1423.18 L1428.04 440.762 L1392.7 440.762 L1392.7 440.762  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,440.762 1392.7,1423.18 1428.04,1423.18 1428.04,440.762 1392.7,440.762 "/>
+<path clip-path="url(#clip292)" d="M1392.7 510.934 L1392.7 1423.18 L1428.04 1423.18 L1428.04 510.934 L1392.7 510.934 L1392.7 510.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,510.934 1392.7,1423.18 1428.04,1423.18 1428.04,510.934 1392.7,510.934 "/>
+<path clip-path="url(#clip292)" d="M1392.7 581.107 L1392.7 1423.18 L1428.04 1423.18 L1428.04 581.107 L1392.7 581.107 L1392.7 581.107  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,581.107 1392.7,1423.18 1428.04,1423.18 1428.04,581.107 1392.7,581.107 "/>
+<path clip-path="url(#clip292)" d="M1392.7 651.28 L1392.7 1423.18 L1428.04 1423.18 L1428.04 651.28 L1392.7 651.28 L1392.7 651.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,651.28 1392.7,1423.18 1428.04,1423.18 1428.04,651.28 1392.7,651.28 "/>
+<path clip-path="url(#clip292)" d="M1392.7 721.453 L1392.7 1423.18 L1428.04 1423.18 L1428.04 721.453 L1392.7 721.453 L1392.7 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,721.453 1392.7,1423.18 1428.04,1423.18 1428.04,721.453 1392.7,721.453 "/>
+<path clip-path="url(#clip292)" d="M1392.7 791.625 L1392.7 1423.18 L1428.04 1423.18 L1428.04 791.625 L1392.7 791.625 L1392.7 791.625  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,791.625 1392.7,1423.18 1428.04,1423.18 1428.04,791.625 1392.7,791.625 "/>
+<path clip-path="url(#clip292)" d="M1392.7 861.798 L1392.7 1423.18 L1428.04 1423.18 L1428.04 861.798 L1392.7 861.798 L1392.7 861.798  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,861.798 1392.7,1423.18 1428.04,1423.18 1428.04,861.798 1392.7,861.798 "/>
+<path clip-path="url(#clip292)" d="M1392.7 931.971 L1392.7 1423.18 L1428.04 1423.18 L1428.04 931.971 L1392.7 931.971 L1392.7 931.971  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,931.971 1392.7,1423.18 1428.04,1423.18 1428.04,931.971 1392.7,931.971 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1002.14 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1002.14 L1392.7 1002.14 L1392.7 1002.14  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1002.14 1392.7,1423.18 1428.04,1423.18 1428.04,1002.14 1392.7,1002.14 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1072.32 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1072.32 L1392.7 1072.32 L1392.7 1072.32  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1072.32 1392.7,1423.18 1428.04,1423.18 1428.04,1072.32 1392.7,1072.32 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1142.49 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1142.49 L1392.7 1142.49 L1392.7 1142.49  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1142.49 1392.7,1423.18 1428.04,1423.18 1428.04,1142.49 1392.7,1142.49 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1212.66 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1212.66 L1392.7 1212.66 L1392.7 1212.66  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1212.66 1392.7,1423.18 1428.04,1423.18 1428.04,1212.66 1392.7,1212.66 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1282.83 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1282.83 L1392.7 1282.83 L1392.7 1282.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1282.83 1392.7,1423.18 1428.04,1423.18 1428.04,1282.83 1392.7,1282.83 "/>
+<path clip-path="url(#clip292)" d="M1392.7 1353.01 L1392.7 1423.18 L1428.04 1423.18 L1428.04 1353.01 L1392.7 1353.01 L1392.7 1353.01  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1392.7,1353.01 1392.7,1423.18 1428.04,1423.18 1428.04,1353.01 1392.7,1353.01 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="89.8981" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="160.071" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="230.244" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="300.416" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="370.589" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="440.762" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="510.934" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="581.107" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="651.28" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="721.453" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="791.625" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="861.798" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="931.971" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1002.14" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1072.32" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1142.49" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1212.66" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1282.83" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1410.37" cy="1353.01" r="2"/>
+<path clip-path="url(#clip292)" d="M1451.6 61.0035 L1451.6 1423.18 L1486.94 1423.18 L1486.94 61.0035 L1451.6 61.0035 L1451.6 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,61.0035 1451.6,1423.18 1486.94,1423.18 1486.94,61.0035 1451.6,61.0035 "/>
+<path clip-path="url(#clip292)" d="M1451.6 179.334 L1451.6 1423.18 L1486.94 1423.18 L1486.94 179.334 L1451.6 179.334 L1451.6 179.334  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,179.334 1451.6,1423.18 1486.94,1423.18 1486.94,179.334 1451.6,179.334 "/>
+<path clip-path="url(#clip292)" d="M1451.6 455.897 L1451.6 1423.18 L1486.94 1423.18 L1486.94 455.897 L1451.6 455.897 L1451.6 455.897  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,455.897 1451.6,1423.18 1486.94,1423.18 1486.94,455.897 1451.6,455.897 "/>
+<path clip-path="url(#clip292)" d="M1451.6 732.46 L1451.6 1423.18 L1486.94 1423.18 L1486.94 732.46 L1451.6 732.46 L1451.6 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1451.6,732.46 1451.6,1423.18 1486.94,1423.18 1486.94,732.46 1451.6,732.46 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="61.0035" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="179.334" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="455.897" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1469.27" cy="732.46" r="2"/>
+<path clip-path="url(#clip292)" d="M1510.5 52.7478 L1510.5 1423.18 L1545.84 1423.18 L1545.84 52.7478 L1510.5 52.7478 L1510.5 52.7478  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,52.7478 1510.5,1423.18 1545.84,1423.18 1545.84,52.7478 1510.5,52.7478 "/>
+<path clip-path="url(#clip292)" d="M1510.5 122.921 L1510.5 1423.18 L1545.84 1423.18 L1545.84 122.921 L1510.5 122.921 L1510.5 122.921  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,122.921 1510.5,1423.18 1545.84,1423.18 1545.84,122.921 1510.5,122.921 "/>
+<path clip-path="url(#clip292)" d="M1510.5 322.431 L1510.5 1423.18 L1545.84 1423.18 L1545.84 322.431 L1510.5 322.431 L1510.5 322.431  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,322.431 1510.5,1423.18 1545.84,1423.18 1545.84,322.431 1510.5,322.431 "/>
+<path clip-path="url(#clip292)" d="M1510.5 521.942 L1510.5 1423.18 L1545.84 1423.18 L1545.84 521.942 L1510.5 521.942 L1510.5 521.942  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,521.942 1510.5,1423.18 1545.84,1423.18 1545.84,521.942 1510.5,521.942 "/>
+<path clip-path="url(#clip292)" d="M1510.5 721.453 L1510.5 1423.18 L1545.84 1423.18 L1545.84 721.453 L1510.5 721.453 L1510.5 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1510.5,721.453 1510.5,1423.18 1545.84,1423.18 1545.84,721.453 1510.5,721.453 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="52.7478" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="122.921" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="322.431" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="521.942" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1528.17" cy="721.453" r="2"/>
+<path clip-path="url(#clip292)" d="M1569.4 47.2441 L1569.4 1423.18 L1604.74 1423.18 L1604.74 47.2441 L1569.4 47.2441 L1569.4 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,47.2441 1569.4,1423.18 1604.74,1423.18 1604.74,47.2441 1569.4,47.2441 "/>
+<path clip-path="url(#clip292)" d="M1569.4 165.575 L1569.4 1423.18 L1604.74 1423.18 L1604.74 165.575 L1569.4 165.575 L1569.4 165.575  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,165.575 1569.4,1423.18 1604.74,1423.18 1604.74,165.575 1569.4,165.575 "/>
+<path clip-path="url(#clip292)" d="M1569.4 283.905 L1569.4 1423.18 L1604.74 1423.18 L1604.74 283.905 L1569.4 283.905 L1569.4 283.905  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,283.905 1569.4,1423.18 1604.74,1423.18 1604.74,283.905 1569.4,283.905 "/>
+<path clip-path="url(#clip292)" d="M1569.4 693.934 L1569.4 1423.18 L1604.74 1423.18 L1604.74 693.934 L1569.4 693.934 L1569.4 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1569.4,693.934 1569.4,1423.18 1604.74,1423.18 1604.74,693.934 1569.4,693.934 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="165.575" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="283.905" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1587.07" cy="693.934" r="2"/>
+<path clip-path="url(#clip292)" d="M1628.3 48.62 L1628.3 1423.18 L1663.64 1423.18 L1663.64 48.62 L1628.3 48.62 L1628.3 48.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,48.62 1628.3,1423.18 1663.64,1423.18 1663.64,48.62 1628.3,48.62 "/>
+<path clip-path="url(#clip292)" d="M1628.3 118.793 L1628.3 1423.18 L1663.64 1423.18 L1663.64 118.793 L1628.3 118.793 L1628.3 118.793  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,118.793 1628.3,1423.18 1663.64,1423.18 1663.64,118.793 1628.3,118.793 "/>
+<path clip-path="url(#clip292)" d="M1628.3 209.605 L1628.3 1423.18 L1663.64 1423.18 L1663.64 209.605 L1628.3 209.605 L1628.3 209.605  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,209.605 1628.3,1423.18 1663.64,1423.18 1663.64,209.605 1628.3,209.605 "/>
+<path clip-path="url(#clip292)" d="M1628.3 322.431 L1628.3 1423.18 L1663.64 1423.18 L1663.64 322.431 L1628.3 322.431 L1628.3 322.431  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,322.431 1628.3,1423.18 1663.64,1423.18 1663.64,322.431 1628.3,322.431 "/>
+<path clip-path="url(#clip292)" d="M1628.3 732.46 L1628.3 1423.18 L1663.64 1423.18 L1663.64 732.46 L1628.3 732.46 L1628.3 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1628.3,732.46 1628.3,1423.18 1663.64,1423.18 1663.64,732.46 1628.3,732.46 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="48.62" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="118.793" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="209.605" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="322.431" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="732.46" r="2"/>
+<path clip-path="url(#clip292)" d="M1687.2 55.4997 L1687.2 1423.18 L1722.54 1423.18 L1722.54 55.4997 L1687.2 55.4997 L1687.2 55.4997  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,55.4997 1687.2,1423.18 1722.54,1423.18 1722.54,55.4997 1687.2,55.4997 "/>
+<path clip-path="url(#clip292)" d="M1687.2 146.311 L1687.2 1423.18 L1722.54 1423.18 L1722.54 146.311 L1687.2 146.311 L1687.2 146.311  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,146.311 1687.2,1423.18 1722.54,1423.18 1722.54,146.311 1687.2,146.311 "/>
+<path clip-path="url(#clip292)" d="M1687.2 297.664 L1687.2 1423.18 L1722.54 1423.18 L1722.54 297.664 L1687.2 297.664 L1687.2 297.664  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,297.664 1687.2,1423.18 1722.54,1423.18 1722.54,297.664 1687.2,297.664 "/>
+<path clip-path="url(#clip292)" d="M1687.2 721.453 L1687.2 1423.18 L1722.54 1423.18 L1722.54 721.453 L1687.2 721.453 L1687.2 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1687.2,721.453 1687.2,1423.18 1722.54,1423.18 1722.54,721.453 1687.2,721.453 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="55.4997" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="146.311" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="297.664" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1704.87" cy="721.453" r="2"/>
+<path clip-path="url(#clip292)" d="M1746.1 47.2441 L1746.1 1423.18 L1781.43 1423.18 L1781.43 47.2441 L1746.1 47.2441 L1746.1 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,47.2441 1746.1,1423.18 1781.43,1423.18 1781.43,47.2441 1746.1,47.2441 "/>
+<path clip-path="url(#clip292)" d="M1746.1 270.146 L1746.1 1423.18 L1781.43 1423.18 L1781.43 270.146 L1746.1 270.146 L1746.1 270.146  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,270.146 1746.1,1423.18 1781.43,1423.18 1781.43,270.146 1746.1,270.146 "/>
+<path clip-path="url(#clip292)" d="M1746.1 693.934 L1746.1 1423.18 L1781.43 1423.18 L1781.43 693.934 L1746.1 693.934 L1746.1 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.1,693.934 1746.1,1423.18 1781.43,1423.18 1781.43,693.934 1746.1,693.934 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="270.146" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1763.77" cy="693.934" r="2"/>
+<path clip-path="url(#clip292)" d="M1804.99 59.6275 L1804.99 1423.18 L1840.33 1423.18 L1840.33 59.6275 L1804.99 59.6275 L1804.99 59.6275  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,59.6275 1804.99,1423.18 1840.33,1423.18 1840.33,59.6275 1804.99,59.6275 "/>
+<path clip-path="url(#clip292)" d="M1804.99 129.8 L1804.99 1423.18 L1840.33 1423.18 L1840.33 129.8 L1804.99 129.8 L1804.99 129.8  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,129.8 1804.99,1423.18 1840.33,1423.18 1840.33,129.8 1804.99,129.8 "/>
+<path clip-path="url(#clip292)" d="M1804.99 406.363 L1804.99 1423.18 L1840.33 1423.18 L1840.33 406.363 L1804.99 406.363 L1804.99 406.363  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,406.363 1804.99,1423.18 1840.33,1423.18 1840.33,406.363 1804.99,406.363 "/>
+<path clip-path="url(#clip292)" d="M1804.99 682.926 L1804.99 1423.18 L1840.33 1423.18 L1840.33 682.926 L1804.99 682.926 L1804.99 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.99,682.926 1804.99,1423.18 1840.33,1423.18 1840.33,682.926 1804.99,682.926 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="59.6275" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="129.8" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="406.363" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1822.66" cy="682.926" r="2"/>
+<path clip-path="url(#clip292)" d="M1863.89 58.2516 L1863.89 1423.18 L1899.23 1423.18 L1899.23 58.2516 L1863.89 58.2516 L1863.89 58.2516  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,58.2516 1863.89,1423.18 1899.23,1423.18 1899.23,58.2516 1863.89,58.2516 "/>
+<path clip-path="url(#clip292)" d="M1863.89 176.582 L1863.89 1423.18 L1899.23 1423.18 L1899.23 176.582 L1863.89 176.582 L1863.89 176.582  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,176.582 1863.89,1423.18 1899.23,1423.18 1899.23,176.582 1863.89,176.582 "/>
+<path clip-path="url(#clip292)" d="M1863.89 294.913 L1863.89 1423.18 L1899.23 1423.18 L1899.23 294.913 L1863.89 294.913 L1863.89 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,294.913 1863.89,1423.18 1899.23,1423.18 1899.23,294.913 1863.89,294.913 "/>
+<path clip-path="url(#clip292)" d="M1863.89 413.243 L1863.89 1423.18 L1899.23 1423.18 L1899.23 413.243 L1863.89 413.243 L1863.89 413.243  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,413.243 1863.89,1423.18 1899.23,1423.18 1899.23,413.243 1863.89,413.243 "/>
+<path clip-path="url(#clip292)" d="M1863.89 531.573 L1863.89 1423.18 L1899.23 1423.18 L1899.23 531.573 L1863.89 531.573 L1863.89 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,531.573 1863.89,1423.18 1899.23,1423.18 1899.23,531.573 1863.89,531.573 "/>
+<path clip-path="url(#clip292)" d="M1863.89 682.926 L1863.89 1423.18 L1899.23 1423.18 L1899.23 682.926 L1863.89 682.926 L1863.89 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1863.89,682.926 1863.89,1423.18 1899.23,1423.18 1899.23,682.926 1863.89,682.926 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="58.2516" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="176.582" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="294.913" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="413.243" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="531.573" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1881.56" cy="682.926" r="2"/>
+<path clip-path="url(#clip292)" d="M1922.79 55.4997 L1922.79 1423.18 L1958.13 1423.18 L1958.13 55.4997 L1922.79 55.4997 L1922.79 55.4997  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,55.4997 1922.79,1423.18 1958.13,1423.18 1958.13,55.4997 1922.79,55.4997 "/>
+<path clip-path="url(#clip292)" d="M1922.79 278.401 L1922.79 1423.18 L1958.13 1423.18 L1958.13 278.401 L1922.79 278.401 L1922.79 278.401  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,278.401 1922.79,1423.18 1958.13,1423.18 1958.13,278.401 1922.79,278.401 "/>
+<path clip-path="url(#clip292)" d="M1922.79 721.453 L1922.79 1423.18 L1958.13 1423.18 L1958.13 721.453 L1922.79 721.453 L1922.79 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1922.79,721.453 1922.79,1423.18 1958.13,1423.18 1958.13,721.453 1922.79,721.453 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="55.4997" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="278.401" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1940.46" cy="721.453" r="2"/>
+<path clip-path="url(#clip292)" d="M1981.69 47.2441 L1981.69 1423.18 L2017.03 1423.18 L2017.03 47.2441 L1981.69 47.2441 L1981.69 47.2441  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,47.2441 1981.69,1423.18 2017.03,1423.18 2017.03,47.2441 1981.69,47.2441 "/>
+<path clip-path="url(#clip292)" d="M1981.69 160.071 L1981.69 1423.18 L2017.03 1423.18 L2017.03 160.071 L1981.69 160.071 L1981.69 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,160.071 1981.69,1423.18 2017.03,1423.18 2017.03,160.071 1981.69,160.071 "/>
+<path clip-path="url(#clip292)" d="M1981.69 272.898 L1981.69 1423.18 L2017.03 1423.18 L2017.03 272.898 L1981.69 272.898 L1981.69 272.898  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,272.898 1981.69,1423.18 2017.03,1423.18 2017.03,272.898 1981.69,272.898 "/>
+<path clip-path="url(#clip292)" d="M1981.69 682.926 L1981.69 1423.18 L2017.03 1423.18 L2017.03 682.926 L1981.69 682.926 L1981.69 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1981.69,682.926 1981.69,1423.18 2017.03,1423.18 2017.03,682.926 1981.69,682.926 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="160.071" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="272.898" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1999.36" cy="682.926" r="2"/>
+<path clip-path="url(#clip292)" d="M2040.59 88.5222 L2040.59 1423.18 L2075.93 1423.18 L2075.93 88.5222 L2040.59 88.5222 L2040.59 88.5222  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,88.5222 2040.59,1423.18 2075.93,1423.18 2075.93,88.5222 2040.59,88.5222 "/>
+<path clip-path="url(#clip292)" d="M2040.59 239.875 L2040.59 1423.18 L2075.93 1423.18 L2075.93 239.875 L2040.59 239.875 L2040.59 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,239.875 2040.59,1423.18 2075.93,1423.18 2075.93,239.875 2040.59,239.875 "/>
+<path clip-path="url(#clip292)" d="M2040.59 682.926 L2040.59 1423.18 L2075.93 1423.18 L2075.93 682.926 L2040.59 682.926 L2040.59 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2040.59,682.926 2040.59,1423.18 2075.93,1423.18 2075.93,682.926 2040.59,682.926 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="88.5222" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="239.875" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2058.26" cy="682.926" r="2"/>
+<path clip-path="url(#clip292)" d="M2099.49 49.996 L2099.49 1423.18 L2134.83 1423.18 L2134.83 49.996 L2099.49 49.996 L2099.49 49.996  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,49.996 2099.49,1423.18 2134.83,1423.18 2134.83,49.996 2099.49,49.996 "/>
+<path clip-path="url(#clip292)" d="M2099.49 168.326 L2099.49 1423.18 L2134.83 1423.18 L2134.83 168.326 L2099.49 168.326 L2099.49 168.326  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,168.326 2099.49,1423.18 2134.83,1423.18 2134.83,168.326 2099.49,168.326 "/>
+<path clip-path="url(#clip292)" d="M2099.49 391.228 L2099.49 1423.18 L2134.83 1423.18 L2134.83 391.228 L2099.49 391.228 L2099.49 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.49,391.228 2099.49,1423.18 2134.83,1423.18 2134.83,391.228 2099.49,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="49.996" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="168.326" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2117.16" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M2158.39 114.665 L2158.39 1423.18 L2193.73 1423.18 L2193.73 114.665 L2158.39 114.665 L2158.39 114.665  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2158.39,114.665 2158.39,1423.18 2193.73,1423.18 2193.73,114.665 2158.39,114.665 "/>
+<path clip-path="url(#clip292)" d="M2158.39 391.228 L2158.39 1423.18 L2193.73 1423.18 L2193.73 391.228 L2158.39 391.228 L2158.39 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2158.39,391.228 2158.39,1423.18 2193.73,1423.18 2193.73,391.228 2158.39,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2176.06" cy="114.665" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2176.06" cy="391.228" r="2"/>
+<path clip-path="url(#clip292)" d="M2217.29 109.161 L2217.29 1423.18 L2252.63 1423.18 L2252.63 109.161 L2217.29 109.161 L2217.29 109.161  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,109.161 2217.29,1423.18 2252.63,1423.18 2252.63,109.161 2217.29,109.161 "/>
+<path clip-path="url(#clip292)" d="M2217.29 308.672 L2217.29 1423.18 L2252.63 1423.18 L2252.63 308.672 L2217.29 308.672 L2217.29 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,308.672 2217.29,1423.18 2252.63,1423.18 2252.63,308.672 2217.29,308.672 "/>
+<path clip-path="url(#clip292)" d="M2217.29 732.46 L2217.29 1423.18 L2252.63 1423.18 L2252.63 732.46 L2217.29 732.46 L2217.29 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2217.29,732.46 2217.29,1423.18 2252.63,1423.18 2252.63,732.46 2217.29,732.46 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="109.161" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="308.672" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2234.96" cy="732.46" r="2"/>
+<path clip-path="url(#clip292)" d="M2276.19 48.62 L2276.19 1423.18 L2311.53 1423.18 L2311.53 48.62 L2276.19 48.62 L2276.19 48.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,48.62 2276.19,1423.18 2311.53,1423.18 2311.53,48.62 2276.19,48.62 "/>
+<path clip-path="url(#clip292)" d="M2276.19 118.793 L2276.19 1423.18 L2311.53 1423.18 L2311.53 118.793 L2276.19 118.793 L2276.19 118.793  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,118.793 2276.19,1423.18 2311.53,1423.18 2311.53,118.793 2276.19,118.793 "/>
+<path clip-path="url(#clip292)" d="M2276.19 209.605 L2276.19 1423.18 L2311.53 1423.18 L2311.53 209.605 L2276.19 209.605 L2276.19 209.605  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,209.605 2276.19,1423.18 2311.53,1423.18 2311.53,209.605 2276.19,209.605 "/>
+<path clip-path="url(#clip292)" d="M2276.19 300.416 L2276.19 1423.18 L2311.53 1423.18 L2311.53 300.416 L2276.19 300.416 L2276.19 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,300.416 2276.19,1423.18 2311.53,1423.18 2311.53,300.416 2276.19,300.416 "/>
+<path clip-path="url(#clip292)" d="M2276.19 391.228 L2276.19 1423.18 L2311.53 1423.18 L2311.53 391.228 L2276.19 391.228 L2276.19 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip292)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2276.19,391.228 2276.19,1423.18 2311.53,1423.18 2311.53,391.228 2276.19,391.228 "/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="48.62" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="118.793" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="209.605" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="300.416" r="2"/>
+<circle clip-path="url(#clip292)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2293.86" cy="391.228" r="2"/>
 </svg>
diff --git a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/50ad43f9.svg b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/80b5cf75.svg
similarity index 72%
rename from previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/50ad43f9.svg
rename to previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/80b5cf75.svg
index f953f416d8c..bab4189438e 100644
--- a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/50ad43f9.svg
+++ b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/80b5cf75.svg
@@ -1,350 +1,350 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip410">
+  <clipPath id="clip230">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip410)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip230)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip411">
+  <clipPath id="clip231">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip410)" d="M232.389 1423.18 L2352.76 1423.18 L2352.76 47.2441 L232.389 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip230)" d="M232.389 1423.18 L2352.76 1423.18 L2352.76 47.2441 L232.389 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip412">
+  <clipPath id="clip232">
     <rect x="232" y="47" width="2121" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="535.299,1423.18 535.299,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.208,1423.18 838.208,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1141.12,1423.18 1141.12,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1444.03,1423.18 1444.03,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1746.94,1423.18 1746.94,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2049.85,1423.18 2049.85,47.2441 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2352.76,1423.18 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="535.299,1423.18 535.299,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.208,1423.18 838.208,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1141.12,1423.18 1141.12,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1444.03,1423.18 1444.03,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.94,1423.18 1746.94,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2049.85,1423.18 2049.85,1404.28 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2352.76,1423.18 2352.76,1404.28 "/>
-<path clip-path="url(#clip410)" d="M232.389 1454.1 Q228.778 1454.1 226.95 1457.66 Q225.144 1461.2 225.144 1468.33 Q225.144 1475.44 226.95 1479.01 Q228.778 1482.55 232.389 1482.55 Q236.024 1482.55 237.829 1479.01 Q239.658 1475.44 239.658 1468.33 Q239.658 1461.2 237.829 1457.66 Q236.024 1454.1 232.389 1454.1 M232.389 1450.39 Q238.2 1450.39 241.255 1455 Q244.334 1459.58 244.334 1468.33 Q244.334 1477.06 241.255 1481.67 Q238.2 1486.25 232.389 1486.25 Q226.579 1486.25 223.501 1481.67 Q220.445 1477.06 220.445 1468.33 Q220.445 1459.58 223.501 1455 Q226.579 1450.39 232.389 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M539.547 1466.95 Q542.903 1467.66 544.778 1469.93 Q546.676 1472.2 546.676 1475.53 Q546.676 1480.65 543.158 1483.45 Q539.639 1486.25 533.158 1486.25 Q530.982 1486.25 528.667 1485.81 Q526.375 1485.39 523.922 1484.54 L523.922 1480.02 Q525.866 1481.16 528.181 1481.74 Q530.496 1482.32 533.019 1482.32 Q537.417 1482.32 539.709 1480.58 Q542.023 1478.84 542.023 1475.53 Q542.023 1472.48 539.871 1470.77 Q537.741 1469.03 533.922 1469.03 L529.894 1469.03 L529.894 1465.19 L534.107 1465.19 Q537.556 1465.19 539.384 1463.82 Q541.213 1462.43 541.213 1459.84 Q541.213 1457.18 539.315 1455.77 Q537.44 1454.33 533.922 1454.33 Q532 1454.33 529.801 1454.75 Q527.602 1455.16 524.963 1456.04 L524.963 1451.88 Q527.625 1451.14 529.94 1450.77 Q532.278 1450.39 534.338 1450.39 Q539.662 1450.39 542.764 1452.83 Q545.866 1455.23 545.866 1459.35 Q545.866 1462.22 544.222 1464.21 Q542.579 1466.18 539.547 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M838.613 1466.44 Q835.465 1466.44 833.613 1468.59 Q831.785 1470.74 831.785 1474.49 Q831.785 1478.22 833.613 1480.39 Q835.465 1482.55 838.613 1482.55 Q841.762 1482.55 843.59 1480.39 Q845.442 1478.22 845.442 1474.49 Q845.442 1470.74 843.59 1468.59 Q841.762 1466.44 838.613 1466.44 M847.896 1451.78 L847.896 1456.04 Q846.137 1455.21 844.331 1454.77 Q842.549 1454.33 840.789 1454.33 Q836.16 1454.33 833.706 1457.45 Q831.276 1460.58 830.928 1466.9 Q832.294 1464.89 834.354 1463.82 Q836.414 1462.73 838.891 1462.73 Q844.1 1462.73 847.109 1465.9 Q850.141 1469.05 850.141 1474.49 Q850.141 1479.82 846.993 1483.03 Q843.845 1486.25 838.613 1486.25 Q832.618 1486.25 829.447 1481.67 Q826.276 1477.06 826.276 1468.33 Q826.276 1460.14 830.164 1455.28 Q834.053 1450.39 840.604 1450.39 Q842.363 1450.39 844.146 1450.74 Q845.951 1451.09 847.896 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1131.42 1484.86 L1131.42 1480.6 Q1133.18 1481.44 1134.98 1481.88 Q1136.79 1482.32 1138.53 1482.32 Q1143.15 1482.32 1145.59 1479.21 Q1148.04 1476.09 1148.39 1469.75 Q1147.04 1471.74 1144.98 1472.8 Q1142.92 1473.87 1140.42 1473.87 Q1135.24 1473.87 1132.21 1470.74 Q1129.2 1467.59 1129.2 1462.15 Q1129.2 1456.83 1132.34 1453.61 Q1135.49 1450.39 1140.72 1450.39 Q1146.72 1450.39 1149.87 1455 Q1153.04 1459.58 1153.04 1468.33 Q1153.04 1476.51 1149.15 1481.39 Q1145.28 1486.25 1138.73 1486.25 Q1136.97 1486.25 1135.17 1485.9 Q1133.36 1485.56 1131.42 1484.86 M1140.72 1470.21 Q1143.87 1470.21 1145.7 1468.06 Q1147.55 1465.9 1147.55 1462.15 Q1147.55 1458.43 1145.7 1456.27 Q1143.87 1454.1 1140.72 1454.1 Q1137.58 1454.1 1135.72 1456.27 Q1133.9 1458.43 1133.9 1462.15 Q1133.9 1465.9 1135.72 1468.06 Q1137.58 1470.21 1140.72 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1419.51 1481.64 L1427.15 1481.64 L1427.15 1455.28 L1418.84 1456.95 L1418.84 1452.69 L1427.11 1451.02 L1431.78 1451.02 L1431.78 1481.64 L1439.42 1481.64 L1439.42 1485.58 L1419.51 1485.58 L1419.51 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1452.89 1481.64 L1469.21 1481.64 L1469.21 1485.58 L1447.27 1485.58 L1447.27 1481.64 Q1449.93 1478.89 1454.51 1474.26 Q1459.12 1469.61 1460.3 1468.27 Q1462.55 1465.74 1463.43 1464.01 Q1464.33 1462.25 1464.33 1460.56 Q1464.33 1457.8 1462.38 1456.07 Q1460.46 1454.33 1457.36 1454.33 Q1455.16 1454.33 1452.71 1455.09 Q1450.28 1455.86 1447.5 1457.41 L1447.5 1452.69 Q1450.32 1451.55 1452.78 1450.97 Q1455.23 1450.39 1457.27 1450.39 Q1462.64 1450.39 1465.83 1453.08 Q1469.03 1455.77 1469.03 1460.26 Q1469.03 1462.39 1468.22 1464.31 Q1467.43 1466.2 1465.32 1468.8 Q1464.74 1469.47 1461.64 1472.69 Q1458.54 1475.88 1452.89 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1722.12 1481.64 L1729.76 1481.64 L1729.76 1455.28 L1721.45 1456.95 L1721.45 1452.69 L1729.71 1451.02 L1734.39 1451.02 L1734.39 1481.64 L1742.03 1481.64 L1742.03 1485.58 L1722.12 1485.58 L1722.12 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1751.52 1451.02 L1769.88 1451.02 L1769.88 1454.96 L1755.8 1454.96 L1755.8 1463.43 Q1756.82 1463.08 1757.84 1462.92 Q1758.86 1462.73 1759.88 1462.73 Q1765.66 1462.73 1769.04 1465.9 Q1772.42 1469.08 1772.42 1474.49 Q1772.42 1480.07 1768.95 1483.17 Q1765.48 1486.25 1759.16 1486.25 Q1756.98 1486.25 1754.71 1485.88 Q1752.47 1485.51 1750.06 1484.77 L1750.06 1480.07 Q1752.15 1481.2 1754.37 1481.76 Q1756.59 1482.32 1759.07 1482.32 Q1763.07 1482.32 1765.41 1480.21 Q1767.75 1478.1 1767.75 1474.49 Q1767.75 1470.88 1765.41 1468.77 Q1763.07 1466.67 1759.07 1466.67 Q1757.19 1466.67 1755.32 1467.08 Q1753.46 1467.5 1751.52 1468.38 L1751.52 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2024.58 1481.64 L2032.22 1481.64 L2032.22 1455.28 L2023.91 1456.95 L2023.91 1452.69 L2032.17 1451.02 L2036.85 1451.02 L2036.85 1481.64 L2044.49 1481.64 L2044.49 1485.58 L2024.58 1485.58 L2024.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2063.93 1469.17 Q2060.6 1469.17 2058.68 1470.95 Q2056.78 1472.73 2056.78 1475.86 Q2056.78 1478.98 2058.68 1480.77 Q2060.6 1482.55 2063.93 1482.55 Q2067.27 1482.55 2069.19 1480.77 Q2071.11 1478.96 2071.11 1475.86 Q2071.11 1472.73 2069.19 1470.95 Q2067.29 1469.17 2063.93 1469.17 M2059.26 1467.18 Q2056.25 1466.44 2054.56 1464.38 Q2052.89 1462.32 2052.89 1459.35 Q2052.89 1455.21 2055.83 1452.8 Q2058.79 1450.39 2063.93 1450.39 Q2069.09 1450.39 2072.03 1452.8 Q2074.97 1455.21 2074.97 1459.35 Q2074.97 1462.32 2073.28 1464.38 Q2071.62 1466.44 2068.63 1467.18 Q2072.01 1467.96 2073.89 1470.26 Q2075.78 1472.55 2075.78 1475.86 Q2075.78 1480.88 2072.71 1483.57 Q2069.65 1486.25 2063.93 1486.25 Q2058.21 1486.25 2055.14 1483.57 Q2052.08 1480.88 2052.08 1475.86 Q2052.08 1472.55 2053.98 1470.26 Q2055.88 1467.96 2059.26 1467.18 M2057.54 1459.79 Q2057.54 1462.48 2059.21 1463.98 Q2060.9 1465.49 2063.93 1465.49 Q2066.94 1465.49 2068.63 1463.98 Q2070.34 1462.48 2070.34 1459.79 Q2070.34 1457.11 2068.63 1455.6 Q2066.94 1454.1 2063.93 1454.1 Q2060.9 1454.1 2059.21 1455.6 Q2057.54 1457.11 2057.54 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2332.14 1481.64 L2348.46 1481.64 L2348.46 1485.58 L2326.52 1485.58 L2326.52 1481.64 Q2329.18 1478.89 2333.76 1474.26 Q2338.37 1469.61 2339.55 1468.27 Q2341.8 1465.74 2342.67 1464.01 Q2343.58 1462.25 2343.58 1460.56 Q2343.58 1457.8 2341.63 1456.07 Q2339.71 1454.33 2336.61 1454.33 Q2334.41 1454.33 2331.96 1455.09 Q2329.53 1455.86 2326.75 1457.41 L2326.75 1452.69 Q2329.57 1451.55 2332.03 1450.97 Q2334.48 1450.39 2336.52 1450.39 Q2341.89 1450.39 2345.08 1453.08 Q2348.28 1455.77 2348.28 1460.26 Q2348.28 1462.39 2347.47 1464.31 Q2346.68 1466.2 2344.57 1468.8 Q2343.99 1469.47 2340.89 1472.69 Q2337.79 1475.88 2332.14 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2359.09 1481.64 L2366.73 1481.64 L2366.73 1455.28 L2358.42 1456.95 L2358.42 1452.69 L2366.68 1451.02 L2371.36 1451.02 L2371.36 1481.64 L2378.99 1481.64 L2378.99 1485.58 L2359.09 1485.58 L2359.09 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1187.71 1525.81 L1187.71 1543.66 L1195.8 1543.66 Q1200.29 1543.66 1202.74 1541.34 Q1205.19 1539.02 1205.19 1534.72 Q1205.19 1530.45 1202.74 1528.13 Q1200.29 1525.81 1195.8 1525.81 L1187.71 1525.81 M1181.28 1520.52 L1195.8 1520.52 Q1203.79 1520.52 1207.86 1524.15 Q1211.97 1527.75 1211.97 1534.72 Q1211.97 1541.75 1207.86 1545.35 Q1203.79 1548.95 1195.8 1548.95 L1187.71 1548.95 L1187.71 1568.04 L1181.28 1568.04 L1181.28 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1233.61 1550.12 Q1226.51 1550.12 1223.78 1551.75 Q1221.04 1553.37 1221.04 1557.29 Q1221.04 1560.4 1223.08 1562.25 Q1225.14 1564.07 1228.68 1564.07 Q1233.55 1564.07 1236.47 1560.63 Q1239.43 1557.16 1239.43 1551.43 L1239.43 1550.12 L1233.61 1550.12 M1245.29 1547.71 L1245.29 1568.04 L1239.43 1568.04 L1239.43 1562.63 Q1237.43 1565.88 1234.44 1567.44 Q1231.45 1568.97 1227.12 1568.97 Q1221.64 1568.97 1218.4 1565.91 Q1215.18 1562.82 1215.18 1557.67 Q1215.18 1551.65 1219.19 1548.6 Q1223.23 1545.54 1231.22 1545.54 L1239.43 1545.54 L1239.43 1544.97 Q1239.43 1540.93 1236.76 1538.73 Q1234.12 1536.5 1229.31 1536.5 Q1226.26 1536.5 1223.36 1537.23 Q1220.47 1537.97 1217.79 1539.43 L1217.79 1534.02 Q1221.01 1532.78 1224.03 1532.17 Q1227.05 1531.54 1229.92 1531.54 Q1237.65 1531.54 1241.47 1535.55 Q1245.29 1539.56 1245.29 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1263.15 1522.27 L1263.15 1532.4 L1275.21 1532.4 L1275.21 1536.95 L1263.15 1536.95 L1263.15 1556.3 Q1263.15 1560.66 1264.32 1561.9 Q1265.53 1563.14 1269.19 1563.14 L1275.21 1563.14 L1275.21 1568.04 L1269.19 1568.04 Q1262.42 1568.04 1259.84 1565.53 Q1257.26 1562.98 1257.26 1556.3 L1257.26 1536.95 L1252.96 1536.95 L1252.96 1532.4 L1257.26 1532.4 L1257.26 1522.27 L1263.15 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1288.71 1522.27 L1288.71 1532.4 L1300.77 1532.4 L1300.77 1536.95 L1288.71 1536.95 L1288.71 1556.3 Q1288.71 1560.66 1289.88 1561.9 Q1291.09 1563.14 1294.75 1563.14 L1300.77 1563.14 L1300.77 1568.04 L1294.75 1568.04 Q1287.97 1568.04 1285.4 1565.53 Q1282.82 1562.98 1282.82 1556.3 L1282.82 1536.95 L1278.52 1536.95 L1278.52 1532.4 L1282.82 1532.4 L1282.82 1522.27 L1288.71 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1338.96 1548.76 L1338.96 1551.62 L1312.04 1551.62 Q1312.42 1557.67 1315.66 1560.85 Q1318.94 1564 1324.77 1564 Q1328.14 1564 1331.29 1563.17 Q1334.47 1562.35 1337.59 1560.69 L1337.59 1566.23 Q1334.44 1567.57 1331.13 1568.27 Q1327.82 1568.97 1324.42 1568.97 Q1315.89 1568.97 1310.89 1564 Q1305.92 1559.04 1305.92 1550.57 Q1305.92 1541.82 1310.64 1536.69 Q1315.38 1531.54 1323.4 1531.54 Q1330.59 1531.54 1334.76 1536.18 Q1338.96 1540.8 1338.96 1548.76 M1333.11 1547.04 Q1333.04 1542.23 1330.4 1539.37 Q1327.79 1536.5 1323.46 1536.5 Q1318.56 1536.5 1315.6 1539.27 Q1312.67 1542.04 1312.23 1547.07 L1333.11 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1369.23 1537.87 Q1368.24 1537.3 1367.07 1537.04 Q1365.92 1536.76 1364.52 1536.76 Q1359.56 1536.76 1356.88 1540 Q1354.24 1543.22 1354.24 1549.27 L1354.24 1568.04 L1348.35 1568.04 L1348.35 1532.4 L1354.24 1532.4 L1354.24 1537.93 Q1356.09 1534.69 1359.05 1533.13 Q1362.01 1531.54 1366.24 1531.54 Q1366.84 1531.54 1367.58 1531.63 Q1368.31 1531.7 1369.2 1531.85 L1369.23 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1403.86 1546.53 L1403.86 1568.04 L1398 1568.04 L1398 1546.72 Q1398 1541.66 1396.03 1539.14 Q1394.06 1536.63 1390.11 1536.63 Q1385.37 1536.63 1382.63 1539.65 Q1379.89 1542.68 1379.89 1547.9 L1379.89 1568.04 L1374.01 1568.04 L1374.01 1532.4 L1379.89 1532.4 L1379.89 1537.93 Q1381.99 1534.72 1384.83 1533.13 Q1387.69 1531.54 1391.42 1531.54 Q1397.56 1531.54 1400.71 1535.36 Q1403.86 1539.14 1403.86 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1147.99 2352.76,1147.99 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,872.806 2352.76,872.806 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,597.618 2352.76,597.618 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,322.431 2352.76,322.431 "/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,47.2441 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 251.287,1423.18 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1147.99 251.287,1147.99 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,872.806 251.287,872.806 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,597.618 251.287,597.618 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,322.431 251.287,322.431 "/>
-<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,47.2441 251.287,47.2441 "/>
-<path clip-path="url(#clip410)" d="M184.445 1408.98 Q180.834 1408.98 179.005 1412.54 Q177.2 1416.08 177.2 1423.21 Q177.2 1430.32 179.005 1433.89 Q180.834 1437.43 184.445 1437.43 Q188.079 1437.43 189.885 1433.89 Q191.713 1430.32 191.713 1423.21 Q191.713 1416.08 189.885 1412.54 Q188.079 1408.98 184.445 1408.98 M184.445 1405.27 Q190.255 1405.27 193.311 1409.88 Q196.389 1414.46 196.389 1423.21 Q196.389 1431.94 193.311 1436.55 Q190.255 1441.13 184.445 1441.13 Q178.635 1441.13 175.556 1436.55 Q172.501 1431.94 172.501 1423.21 Q172.501 1414.46 175.556 1409.88 Q178.635 1405.27 184.445 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M148.311 1161.34 L164.63 1161.34 L164.63 1165.27 L142.686 1165.27 L142.686 1161.34 Q145.348 1158.58 149.931 1153.95 Q154.538 1149.3 155.718 1147.96 Q157.964 1145.43 158.843 1143.7 Q159.746 1141.94 159.746 1140.25 Q159.746 1137.5 157.802 1135.76 Q155.88 1134.02 152.778 1134.02 Q150.579 1134.02 148.126 1134.79 Q145.695 1135.55 142.917 1137.1 L142.917 1132.38 Q145.741 1131.25 148.195 1130.67 Q150.649 1130.09 152.686 1130.09 Q158.056 1130.09 161.251 1132.77 Q164.445 1135.46 164.445 1139.95 Q164.445 1142.08 163.635 1144 Q162.848 1145.9 160.741 1148.49 Q160.163 1149.16 157.061 1152.38 Q153.959 1155.57 148.311 1161.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M184.445 1133.79 Q180.834 1133.79 179.005 1137.36 Q177.2 1140.9 177.2 1148.03 Q177.2 1155.13 179.005 1158.7 Q180.834 1162.24 184.445 1162.24 Q188.079 1162.24 189.885 1158.7 Q191.713 1155.13 191.713 1148.03 Q191.713 1140.9 189.885 1137.36 Q188.079 1133.79 184.445 1133.79 M184.445 1130.09 Q190.255 1130.09 193.311 1134.69 Q196.389 1139.28 196.389 1148.03 Q196.389 1156.75 193.311 1161.36 Q190.255 1165.94 184.445 1165.94 Q178.635 1165.94 175.556 1161.36 Q172.501 1156.75 172.501 1148.03 Q172.501 1139.28 175.556 1134.69 Q178.635 1130.09 184.445 1130.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M157.13 859.6 L145.325 878.049 L157.13 878.049 L157.13 859.6 M155.903 855.526 L161.783 855.526 L161.783 878.049 L166.714 878.049 L166.714 881.937 L161.783 881.937 L161.783 890.086 L157.13 890.086 L157.13 881.937 L141.529 881.937 L141.529 877.424 L155.903 855.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M184.445 858.604 Q180.834 858.604 179.005 862.169 Q177.2 865.711 177.2 872.84 Q177.2 879.947 179.005 883.512 Q180.834 887.053 184.445 887.053 Q188.079 887.053 189.885 883.512 Q191.713 879.947 191.713 872.84 Q191.713 865.711 189.885 862.169 Q188.079 858.604 184.445 858.604 M184.445 854.901 Q190.255 854.901 193.311 859.507 Q196.389 864.09 196.389 872.84 Q196.389 881.567 193.311 886.174 Q190.255 890.757 184.445 890.757 Q178.635 890.757 175.556 886.174 Q172.501 881.567 172.501 872.84 Q172.501 864.09 175.556 859.507 Q178.635 854.901 184.445 854.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M154.862 595.755 Q151.714 595.755 149.862 597.908 Q148.033 600.061 148.033 603.81 Q148.033 607.537 149.862 609.713 Q151.714 611.866 154.862 611.866 Q158.01 611.866 159.839 609.713 Q161.69 607.537 161.69 603.81 Q161.69 600.061 159.839 597.908 Q158.01 595.755 154.862 595.755 M164.144 581.102 L164.144 585.362 Q162.385 584.528 160.579 584.088 Q158.797 583.649 157.038 583.649 Q152.408 583.649 149.954 586.774 Q147.524 589.899 147.177 596.218 Q148.542 594.204 150.603 593.139 Q152.663 592.051 155.14 592.051 Q160.348 592.051 163.357 595.223 Q166.39 598.371 166.39 603.81 Q166.39 609.135 163.241 612.352 Q160.093 615.57 154.862 615.57 Q148.866 615.57 145.695 610.986 Q142.524 606.38 142.524 597.653 Q142.524 589.459 146.413 584.598 Q150.302 579.713 156.853 579.713 Q158.612 579.713 160.394 580.061 Q162.2 580.408 164.144 581.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M184.445 583.417 Q180.834 583.417 179.005 586.982 Q177.2 590.524 177.2 597.653 Q177.2 604.76 179.005 608.324 Q180.834 611.866 184.445 611.866 Q188.079 611.866 189.885 608.324 Q191.713 604.76 191.713 597.653 Q191.713 590.524 189.885 586.982 Q188.079 583.417 184.445 583.417 M184.445 579.713 Q190.255 579.713 193.311 584.32 Q196.389 588.903 196.389 597.653 Q196.389 606.38 193.311 610.986 Q190.255 615.57 184.445 615.57 Q178.635 615.57 175.556 610.986 Q172.501 606.38 172.501 597.653 Q172.501 588.903 175.556 584.32 Q178.635 579.713 184.445 579.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M154.283 323.299 Q150.95 323.299 149.028 325.082 Q147.13 326.864 147.13 329.989 Q147.13 333.114 149.028 334.896 Q150.95 336.679 154.283 336.679 Q157.616 336.679 159.538 334.896 Q161.459 333.091 161.459 329.989 Q161.459 326.864 159.538 325.082 Q157.64 323.299 154.283 323.299 M149.607 321.309 Q146.598 320.568 144.908 318.508 Q143.241 316.447 143.241 313.485 Q143.241 309.341 146.181 306.934 Q149.144 304.526 154.283 304.526 Q159.445 304.526 162.385 306.934 Q165.325 309.341 165.325 313.485 Q165.325 316.447 163.635 318.508 Q161.968 320.568 158.982 321.309 Q162.362 322.096 164.237 324.387 Q166.135 326.679 166.135 329.989 Q166.135 335.012 163.056 337.697 Q160.001 340.383 154.283 340.383 Q148.566 340.383 145.487 337.697 Q142.431 335.012 142.431 329.989 Q142.431 326.679 144.329 324.387 Q146.228 322.096 149.607 321.309 M147.894 313.924 Q147.894 316.61 149.561 318.114 Q151.251 319.619 154.283 319.619 Q157.292 319.619 158.982 318.114 Q160.695 316.61 160.695 313.924 Q160.695 311.239 158.982 309.735 Q157.292 308.23 154.283 308.23 Q151.251 308.23 149.561 309.735 Q147.894 311.239 147.894 313.924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M184.445 308.23 Q180.834 308.23 179.005 311.795 Q177.2 315.336 177.2 322.466 Q177.2 329.572 179.005 333.137 Q180.834 336.679 184.445 336.679 Q188.079 336.679 189.885 333.137 Q191.713 329.572 191.713 322.466 Q191.713 315.336 189.885 311.795 Q188.079 308.23 184.445 308.23 M184.445 304.526 Q190.255 304.526 193.311 309.133 Q196.389 313.716 196.389 322.466 Q196.389 331.193 193.311 335.799 Q190.255 340.383 184.445 340.383 Q178.635 340.383 175.556 335.799 Q172.501 331.193 172.501 322.466 Q172.501 313.716 175.556 309.133 Q178.635 304.526 184.445 304.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M114.931 60.5889 L122.57 60.5889 L122.57 34.2233 L114.26 35.89 L114.26 31.6308 L122.524 29.9641 L127.2 29.9641 L127.2 60.5889 L134.839 60.5889 L134.839 64.5241 L114.931 64.5241 L114.931 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M154.283 33.0428 Q150.672 33.0428 148.843 36.6076 Q147.038 40.1492 147.038 47.2788 Q147.038 54.3853 148.843 57.9501 Q150.672 61.4917 154.283 61.4917 Q157.917 61.4917 159.723 57.9501 Q161.552 54.3853 161.552 47.2788 Q161.552 40.1492 159.723 36.6076 Q157.917 33.0428 154.283 33.0428 M154.283 29.3391 Q160.093 29.3391 163.149 33.9456 Q166.227 38.5289 166.227 47.2788 Q166.227 56.0056 163.149 60.6121 Q160.093 65.1954 154.283 65.1954 Q148.473 65.1954 145.394 60.6121 Q142.339 56.0056 142.339 47.2788 Q142.339 38.5289 145.394 33.9456 Q148.473 29.3391 154.283 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M184.445 33.0428 Q180.834 33.0428 179.005 36.6076 Q177.2 40.1492 177.2 47.2788 Q177.2 54.3853 179.005 57.9501 Q180.834 61.4917 184.445 61.4917 Q188.079 61.4917 189.885 57.9501 Q191.713 54.3853 191.713 47.2788 Q191.713 40.1492 189.885 36.6076 Q188.079 33.0428 184.445 33.0428 M184.445 29.3391 Q190.255 29.3391 193.311 33.9456 Q196.389 38.5289 196.389 47.2788 Q196.389 56.0056 193.311 60.6121 Q190.255 65.1954 184.445 65.1954 Q178.635 65.1954 175.556 60.6121 Q172.501 56.0056 172.501 47.2788 Q172.501 38.5289 175.556 33.9456 Q178.635 29.3391 184.445 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M41.7242 880.191 Q42.4244 878.122 44.7161 876.181 Q47.0077 874.207 51.0181 872.234 L64.0042 865.709 L64.0042 872.616 L51.8138 878.695 Q47.0395 881.05 45.48 883.278 Q43.9204 885.474 43.9204 889.294 L43.9204 896.296 L64.0042 896.296 L64.0042 902.726 L16.4842 902.726 L16.4842 888.212 Q16.4842 880.064 19.8898 876.053 Q23.2955 872.043 30.1704 872.043 Q34.6582 872.043 37.6183 874.144 Q40.5784 876.212 41.7242 880.191 M21.7677 896.296 L38.6368 896.296 L38.6368 888.212 Q38.6368 883.565 36.5043 881.209 Q34.34 878.822 30.1704 878.822 Q26.0009 878.822 23.9002 881.209 Q21.7677 883.565 21.7677 888.212 L21.7677 896.296 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M32.4621 846.803 Q32.4621 851.513 36.1542 854.251 Q39.8145 856.988 46.212 856.988 Q52.6095 856.988 56.3017 854.283 Q59.9619 851.545 59.9619 846.803 Q59.9619 842.124 56.2698 839.387 Q52.5777 836.65 46.212 836.65 Q39.8781 836.65 36.186 839.387 Q32.4621 842.124 32.4621 846.803 M27.4968 846.803 Q27.4968 839.164 32.4621 834.803 Q37.4273 830.443 46.212 830.443 Q54.9649 830.443 59.9619 834.803 Q64.9272 839.164 64.9272 846.803 Q64.9272 854.473 59.9619 858.834 Q54.9649 863.163 46.212 863.163 Q37.4273 863.163 32.4621 858.834 Q27.4968 854.473 27.4968 846.803 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M14.479 820.735 L14.479 814.879 L64.0042 814.879 L64.0042 820.735 L14.479 820.735 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M14.479 802.625 L14.479 796.768 L64.0042 796.768 L64.0042 802.625 L14.479 802.625 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M14.479 763.794 L14.479 757.938 L64.0042 757.938 L64.0042 763.794 L14.479 763.794 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M44.7161 715.192 L47.5806 715.192 L47.5806 742.119 Q53.6281 741.737 56.8109 738.49 Q59.9619 735.212 59.9619 729.387 Q59.9619 726.014 59.1344 722.862 Q58.3069 719.68 56.6518 716.56 L62.1899 716.56 Q63.5267 719.711 64.227 723.022 Q64.9272 726.332 64.9272 729.737 Q64.9272 738.268 59.9619 743.265 Q54.9967 748.23 46.5303 748.23 Q37.7774 748.23 32.6531 743.519 Q27.4968 738.777 27.4968 730.756 Q27.4968 723.563 32.1438 719.393 Q36.7589 715.192 44.7161 715.192 M42.9973 721.048 Q38.1912 721.112 35.3266 723.754 Q32.4621 726.364 32.4621 730.692 Q32.4621 735.594 35.2312 738.554 Q38.0002 741.482 43.0292 741.928 L42.9973 721.048 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M42.4881 675.947 L64.0042 675.947 L64.0042 681.804 L42.679 681.804 Q37.6183 681.804 35.1038 683.777 Q32.5894 685.75 32.5894 689.697 Q32.5894 694.44 35.6131 697.177 Q38.6368 699.914 43.8567 699.914 L64.0042 699.914 L64.0042 705.802 L28.3562 705.802 L28.3562 699.914 L33.8944 699.914 Q30.6797 697.813 29.0883 694.981 Q27.4968 692.116 27.4968 688.392 Q27.4968 682.249 31.3163 679.098 Q35.1038 675.947 42.4881 675.947 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M45.7664 640.809 Q39.4007 640.809 35.8996 643.45 Q32.3984 646.06 32.3984 650.803 Q32.3984 655.513 35.8996 658.155 Q39.4007 660.765 45.7664 660.765 Q52.1003 660.765 55.6014 658.155 Q59.1026 655.513 59.1026 650.803 Q59.1026 646.06 55.6014 643.45 Q52.1003 640.809 45.7664 640.809 M59.58 634.952 Q68.683 634.952 73.1071 638.994 Q77.5631 643.037 77.5631 651.376 Q77.5631 654.463 77.0857 657.2 Q76.6401 659.937 75.6852 662.516 L69.9879 662.516 Q71.3884 659.937 72.0568 657.423 Q72.7252 654.909 72.7252 652.299 Q72.7252 646.538 69.7015 643.673 Q66.7096 640.809 60.6303 640.809 L57.7339 640.809 Q60.885 642.623 62.4446 645.456 Q64.0042 648.288 64.0042 652.235 Q64.0042 658.792 59.0071 662.802 Q54.01 666.812 45.7664 666.812 Q37.491 666.812 32.4939 662.802 Q27.4968 658.792 27.4968 652.235 Q27.4968 648.288 29.0564 645.456 Q30.616 642.623 33.7671 640.809 L28.3562 640.809 L28.3562 634.952 L59.58 634.952 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M18.2347 617.096 L28.3562 617.096 L28.3562 605.033 L32.9077 605.033 L32.9077 617.096 L52.2594 617.096 Q56.6199 617.096 57.8613 615.919 Q59.1026 614.709 59.1026 611.049 L59.1026 605.033 L64.0042 605.033 L64.0042 611.049 Q64.0042 617.828 61.4897 620.406 Q58.9434 622.985 52.2594 622.985 L32.9077 622.985 L32.9077 627.281 L28.3562 627.281 L28.3562 622.985 L18.2347 622.985 L18.2347 617.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M42.4881 567.698 L64.0042 567.698 L64.0042 573.555 L42.679 573.555 Q37.6183 573.555 35.1038 575.528 Q32.5894 577.502 32.5894 581.448 Q32.5894 586.191 35.6131 588.928 Q38.6368 591.665 43.8567 591.665 L64.0042 591.665 L64.0042 597.554 L14.479 597.554 L14.479 591.665 L33.8944 591.665 Q30.6797 589.565 29.0883 586.732 Q27.4968 583.867 27.4968 580.143 Q27.4968 574 31.3163 570.849 Q35.1038 567.698 42.4881 567.698 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip412)" d="M303.068 391.228 L303.068 1423.18 L363.65 1423.18 L363.65 391.228 L303.068 391.228 L303.068 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.068,391.228 303.068,1423.18 363.65,1423.18 363.65,391.228 303.068,391.228 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="333.359" cy="391.228" r="2"/>
-<path clip-path="url(#clip412)" d="M404.038 391.228 L404.038 1423.18 L464.62 1423.18 L464.62 391.228 L404.038 391.228 L404.038 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="404.038,391.228 404.038,1423.18 464.62,1423.18 464.62,391.228 404.038,391.228 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="434.329" cy="391.228" r="2"/>
-<path clip-path="url(#clip412)" d="M505.008 391.228 L505.008 1423.18 L565.59 1423.18 L565.59 391.228 L505.008 391.228 L505.008 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="505.008,391.228 505.008,1423.18 565.59,1423.18 565.59,391.228 505.008,391.228 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="535.299" cy="391.228" r="2"/>
-<path clip-path="url(#clip412)" d="M605.978 391.228 L605.978 1423.18 L666.56 1423.18 L666.56 391.228 L605.978 391.228 L605.978 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="605.978,391.228 605.978,1423.18 666.56,1423.18 666.56,391.228 605.978,391.228 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="636.269" cy="391.228" r="2"/>
-<path clip-path="url(#clip412)" d="M706.948 391.228 L706.948 1423.18 L767.529 1423.18 L767.529 391.228 L706.948 391.228 L706.948 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="706.948,391.228 706.948,1423.18 767.529,1423.18 767.529,391.228 706.948,391.228 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="737.239" cy="391.228" r="2"/>
-<path clip-path="url(#clip412)" d="M807.917 682.926 L807.917 1423.18 L868.499 1423.18 L868.499 682.926 L807.917 682.926 L807.917 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="807.917,682.926 807.917,1423.18 868.499,1423.18 868.499,682.926 807.917,682.926 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="838.208" cy="682.926" r="2"/>
-<path clip-path="url(#clip412)" d="M908.887 693.934 L908.887 1423.18 L969.469 1423.18 L969.469 693.934 L908.887 693.934 L908.887 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="908.887,693.934 908.887,1423.18 969.469,1423.18 969.469,693.934 908.887,693.934 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="693.934" r="2"/>
-<path clip-path="url(#clip412)" d="M1009.86 721.453 L1009.86 1423.18 L1070.44 1423.18 L1070.44 721.453 L1009.86 721.453 L1009.86 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1009.86,721.453 1009.86,1423.18 1070.44,1423.18 1070.44,721.453 1009.86,721.453 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1040.15" cy="721.453" r="2"/>
-<path clip-path="url(#clip412)" d="M1110.83 732.46 L1110.83 1423.18 L1171.41 1423.18 L1171.41 732.46 L1110.83 732.46 L1110.83 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1110.83,732.46 1110.83,1423.18 1171.41,1423.18 1171.41,732.46 1110.83,732.46 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1141.12" cy="732.46" r="2"/>
-<path clip-path="url(#clip412)" d="M1211.8 94.0259 L1211.8 1423.18 L1272.38 1423.18 L1272.38 94.0259 L1211.8 94.0259 L1211.8 94.0259  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,94.0259 1211.8,1423.18 1272.38,1423.18 1272.38,94.0259 1211.8,94.0259 "/>
-<path clip-path="url(#clip412)" d="M1211.8 537.077 L1211.8 1423.18 L1272.38 1423.18 L1272.38 537.077 L1211.8 537.077 L1211.8 537.077  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,537.077 1211.8,1423.18 1272.38,1423.18 1272.38,537.077 1211.8,537.077 "/>
-<path clip-path="url(#clip412)" d="M1211.8 980.129 L1211.8 1423.18 L1272.38 1423.18 L1272.38 980.129 L1211.8 980.129 L1211.8 980.129  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,980.129 1211.8,1423.18 1272.38,1423.18 1272.38,980.129 1211.8,980.129 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="94.0259" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="537.077" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="980.129" r="2"/>
-<path clip-path="url(#clip412)" d="M1312.77 151.815 L1312.77 1423.18 L1373.35 1423.18 L1373.35 151.815 L1312.77 151.815 L1312.77 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,151.815 1312.77,1423.18 1373.35,1423.18 1373.35,151.815 1312.77,151.815 "/>
-<path clip-path="url(#clip412)" d="M1312.77 575.603 L1312.77 1423.18 L1373.35 1423.18 L1373.35 575.603 L1312.77 575.603 L1312.77 575.603  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,575.603 1312.77,1423.18 1373.35,1423.18 1373.35,575.603 1312.77,575.603 "/>
-<path clip-path="url(#clip412)" d="M1312.77 999.392 L1312.77 1423.18 L1373.35 1423.18 L1373.35 999.392 L1312.77 999.392 L1312.77 999.392  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,999.392 1312.77,1423.18 1373.35,1423.18 1373.35,999.392 1312.77,999.392 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="151.815" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="575.603" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="999.392" r="2"/>
-<path clip-path="url(#clip412)" d="M1413.74 193.093 L1413.74 1423.18 L1474.32 1423.18 L1474.32 193.093 L1413.74 193.093 L1413.74 193.093  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,193.093 1413.74,1423.18 1474.32,1423.18 1474.32,193.093 1413.74,193.093 "/>
-<path clip-path="url(#clip412)" d="M1413.74 603.122 L1413.74 1423.18 L1474.32 1423.18 L1474.32 603.122 L1413.74 603.122 L1413.74 603.122  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,603.122 1413.74,1423.18 1474.32,1423.18 1474.32,603.122 1413.74,603.122 "/>
-<path clip-path="url(#clip412)" d="M1413.74 1013.15 L1413.74 1423.18 L1474.32 1423.18 L1474.32 1013.15 L1413.74 1013.15 L1413.74 1013.15  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,1013.15 1413.74,1423.18 1474.32,1423.18 1474.32,1013.15 1413.74,1013.15 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="193.093" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="603.122" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="1013.15" r="2"/>
-<path clip-path="url(#clip412)" d="M1514.71 316.928 L1514.71 1423.18 L1575.29 1423.18 L1575.29 316.928 L1514.71 316.928 L1514.71 316.928  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,316.928 1514.71,1423.18 1575.29,1423.18 1575.29,316.928 1514.71,316.928 "/>
-<path clip-path="url(#clip412)" d="M1514.71 593.491 L1514.71 1423.18 L1575.29 1423.18 L1575.29 593.491 L1514.71 593.491 L1514.71 593.491  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,593.491 1514.71,1423.18 1575.29,1423.18 1575.29,593.491 1514.71,593.491 "/>
-<path clip-path="url(#clip412)" d="M1514.71 870.054 L1514.71 1423.18 L1575.29 1423.18 L1575.29 870.054 L1514.71 870.054 L1514.71 870.054  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,870.054 1514.71,1423.18 1575.29,1423.18 1575.29,870.054 1514.71,870.054 "/>
-<path clip-path="url(#clip412)" d="M1514.71 1146.62 L1514.71 1423.18 L1575.29 1423.18 L1575.29 1146.62 L1514.71 1146.62 L1514.71 1146.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,1146.62 1514.71,1423.18 1575.29,1423.18 1575.29,1146.62 1514.71,1146.62 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="316.928" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="593.491" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="870.054" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="1146.62" r="2"/>
-<path clip-path="url(#clip412)" d="M1615.68 85.7703 L1615.68 1423.18 L1676.26 1423.18 L1676.26 85.7703 L1615.68 85.7703 L1615.68 85.7703  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,85.7703 1615.68,1423.18 1676.26,1423.18 1676.26,85.7703 1615.68,85.7703 "/>
-<path clip-path="url(#clip412)" d="M1615.68 308.672 L1615.68 1423.18 L1676.26 1423.18 L1676.26 308.672 L1615.68 308.672 L1615.68 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,308.672 1615.68,1423.18 1676.26,1423.18 1676.26,308.672 1615.68,308.672 "/>
-<path clip-path="url(#clip412)" d="M1615.68 531.573 L1615.68 1423.18 L1676.26 1423.18 L1676.26 531.573 L1615.68 531.573 L1615.68 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,531.573 1615.68,1423.18 1676.26,1423.18 1676.26,531.573 1615.68,531.573 "/>
-<path clip-path="url(#clip412)" d="M1615.68 754.475 L1615.68 1423.18 L1676.26 1423.18 L1676.26 754.475 L1615.68 754.475 L1615.68 754.475  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,754.475 1615.68,1423.18 1676.26,1423.18 1676.26,754.475 1615.68,754.475 "/>
-<path clip-path="url(#clip412)" d="M1615.68 977.377 L1615.68 1423.18 L1676.26 1423.18 L1676.26 977.377 L1615.68 977.377 L1615.68 977.377  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,977.377 1615.68,1423.18 1676.26,1423.18 1676.26,977.377 1615.68,977.377 "/>
-<path clip-path="url(#clip412)" d="M1615.68 1200.28 L1615.68 1423.18 L1676.26 1423.18 L1676.26 1200.28 L1615.68 1200.28 L1615.68 1200.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,1200.28 1615.68,1423.18 1676.26,1423.18 1676.26,1200.28 1615.68,1200.28 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="85.7703" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="308.672" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="531.573" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="754.475" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="977.377" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="1200.28" r="2"/>
-<path clip-path="url(#clip412)" d="M1716.65 226.116 L1716.65 1423.18 L1777.23 1423.18 L1777.23 226.116 L1716.65 226.116 L1716.65 226.116  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,226.116 1716.65,1423.18 1777.23,1423.18 1777.23,226.116 1716.65,226.116 "/>
-<path clip-path="url(#clip412)" d="M1716.65 425.626 L1716.65 1423.18 L1777.23 1423.18 L1777.23 425.626 L1716.65 425.626 L1716.65 425.626  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,425.626 1716.65,1423.18 1777.23,1423.18 1777.23,425.626 1716.65,425.626 "/>
-<path clip-path="url(#clip412)" d="M1716.65 625.137 L1716.65 1423.18 L1777.23 1423.18 L1777.23 625.137 L1716.65 625.137 L1716.65 625.137  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,625.137 1716.65,1423.18 1777.23,1423.18 1777.23,625.137 1716.65,625.137 "/>
-<path clip-path="url(#clip412)" d="M1716.65 824.648 L1716.65 1423.18 L1777.23 1423.18 L1777.23 824.648 L1716.65 824.648 L1716.65 824.648  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,824.648 1716.65,1423.18 1777.23,1423.18 1777.23,824.648 1716.65,824.648 "/>
-<path clip-path="url(#clip412)" d="M1716.65 1024.16 L1716.65 1423.18 L1777.23 1423.18 L1777.23 1024.16 L1716.65 1024.16 L1716.65 1024.16  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,1024.16 1716.65,1423.18 1777.23,1423.18 1777.23,1024.16 1716.65,1024.16 "/>
-<path clip-path="url(#clip412)" d="M1716.65 1223.67 L1716.65 1423.18 L1777.23 1423.18 L1777.23 1223.67 L1716.65 1223.67 L1716.65 1223.67  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,1223.67 1716.65,1423.18 1777.23,1423.18 1777.23,1223.67 1716.65,1223.67 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="226.116" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="425.626" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="625.137" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="824.648" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="1024.16" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="1223.67" r="2"/>
-<path clip-path="url(#clip412)" d="M1817.62 61.0035 L1817.62 1423.18 L1878.2 1423.18 L1878.2 61.0035 L1817.62 61.0035 L1817.62 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,61.0035 1817.62,1423.18 1878.2,1423.18 1878.2,61.0035 1817.62,61.0035 "/>
-<path clip-path="url(#clip412)" d="M1817.62 212.356 L1817.62 1423.18 L1878.2 1423.18 L1878.2 212.356 L1817.62 212.356 L1817.62 212.356  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,212.356 1817.62,1423.18 1878.2,1423.18 1878.2,212.356 1817.62,212.356 "/>
-<path clip-path="url(#clip412)" d="M1817.62 363.709 L1817.62 1423.18 L1878.2 1423.18 L1878.2 363.709 L1817.62 363.709 L1817.62 363.709  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,363.709 1817.62,1423.18 1878.2,1423.18 1878.2,363.709 1817.62,363.709 "/>
-<path clip-path="url(#clip412)" d="M1817.62 515.062 L1817.62 1423.18 L1878.2 1423.18 L1878.2 515.062 L1817.62 515.062 L1817.62 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,515.062 1817.62,1423.18 1878.2,1423.18 1878.2,515.062 1817.62,515.062 "/>
-<path clip-path="url(#clip412)" d="M1817.62 666.415 L1817.62 1423.18 L1878.2 1423.18 L1878.2 666.415 L1817.62 666.415 L1817.62 666.415  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,666.415 1817.62,1423.18 1878.2,1423.18 1878.2,666.415 1817.62,666.415 "/>
-<path clip-path="url(#clip412)" d="M1817.62 817.768 L1817.62 1423.18 L1878.2 1423.18 L1878.2 817.768 L1817.62 817.768 L1817.62 817.768  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,817.768 1817.62,1423.18 1878.2,1423.18 1878.2,817.768 1817.62,817.768 "/>
-<path clip-path="url(#clip412)" d="M1817.62 969.121 L1817.62 1423.18 L1878.2 1423.18 L1878.2 969.121 L1817.62 969.121 L1817.62 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,969.121 1817.62,1423.18 1878.2,1423.18 1878.2,969.121 1817.62,969.121 "/>
-<path clip-path="url(#clip412)" d="M1817.62 1120.47 L1817.62 1423.18 L1878.2 1423.18 L1878.2 1120.47 L1817.62 1120.47 L1817.62 1120.47  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,1120.47 1817.62,1423.18 1878.2,1423.18 1878.2,1120.47 1817.62,1120.47 "/>
-<path clip-path="url(#clip412)" d="M1817.62 1271.83 L1817.62 1423.18 L1878.2 1423.18 L1878.2 1271.83 L1817.62 1271.83 L1817.62 1271.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,1271.83 1817.62,1423.18 1878.2,1423.18 1878.2,1271.83 1817.62,1271.83 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="61.0035" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="212.356" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="363.709" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="515.062" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="666.415" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="817.768" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="969.121" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="1120.47" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="1271.83" r="2"/>
-<path clip-path="url(#clip412)" d="M1918.59 121.545 L1918.59 1423.18 L1979.17 1423.18 L1979.17 121.545 L1918.59 121.545 L1918.59 121.545  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,121.545 1918.59,1423.18 1979.17,1423.18 1979.17,121.545 1918.59,121.545 "/>
-<path clip-path="url(#clip412)" d="M1918.59 239.875 L1918.59 1423.18 L1979.17 1423.18 L1979.17 239.875 L1918.59 239.875 L1918.59 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,239.875 1918.59,1423.18 1979.17,1423.18 1979.17,239.875 1918.59,239.875 "/>
-<path clip-path="url(#clip412)" d="M1918.59 358.206 L1918.59 1423.18 L1979.17 1423.18 L1979.17 358.206 L1918.59 358.206 L1918.59 358.206  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,358.206 1918.59,1423.18 1979.17,1423.18 1979.17,358.206 1918.59,358.206 "/>
-<path clip-path="url(#clip412)" d="M1918.59 476.536 L1918.59 1423.18 L1979.17 1423.18 L1979.17 476.536 L1918.59 476.536 L1918.59 476.536  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,476.536 1918.59,1423.18 1979.17,1423.18 1979.17,476.536 1918.59,476.536 "/>
-<path clip-path="url(#clip412)" d="M1918.59 594.867 L1918.59 1423.18 L1979.17 1423.18 L1979.17 594.867 L1918.59 594.867 L1918.59 594.867  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,594.867 1918.59,1423.18 1979.17,1423.18 1979.17,594.867 1918.59,594.867 "/>
-<path clip-path="url(#clip412)" d="M1918.59 713.197 L1918.59 1423.18 L1979.17 1423.18 L1979.17 713.197 L1918.59 713.197 L1918.59 713.197  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,713.197 1918.59,1423.18 1979.17,1423.18 1979.17,713.197 1918.59,713.197 "/>
-<path clip-path="url(#clip412)" d="M1918.59 831.527 L1918.59 1423.18 L1979.17 1423.18 L1979.17 831.527 L1918.59 831.527 L1918.59 831.527  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,831.527 1918.59,1423.18 1979.17,1423.18 1979.17,831.527 1918.59,831.527 "/>
-<path clip-path="url(#clip412)" d="M1918.59 949.858 L1918.59 1423.18 L1979.17 1423.18 L1979.17 949.858 L1918.59 949.858 L1918.59 949.858  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,949.858 1918.59,1423.18 1979.17,1423.18 1979.17,949.858 1918.59,949.858 "/>
-<path clip-path="url(#clip412)" d="M1918.59 1068.19 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1068.19 L1918.59 1068.19 L1918.59 1068.19  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1068.19 1918.59,1423.18 1979.17,1423.18 1979.17,1068.19 1918.59,1068.19 "/>
-<path clip-path="url(#clip412)" d="M1918.59 1186.52 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1186.52 L1918.59 1186.52 L1918.59 1186.52  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1186.52 1918.59,1423.18 1979.17,1423.18 1979.17,1186.52 1918.59,1186.52 "/>
-<path clip-path="url(#clip412)" d="M1918.59 1304.85 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1304.85 L1918.59 1304.85 L1918.59 1304.85  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1304.85 1918.59,1423.18 1979.17,1423.18 1979.17,1304.85 1918.59,1304.85 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="121.545" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="239.875" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="358.206" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="476.536" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="594.867" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="713.197" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="831.527" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="949.858" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1068.19" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1186.52" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1304.85" r="2"/>
-<path clip-path="url(#clip412)" d="M2019.56 69.2591 L2019.56 1423.18 L2080.14 1423.18 L2080.14 69.2591 L2019.56 69.2591 L2019.56 69.2591  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,69.2591 2019.56,1423.18 2080.14,1423.18 2080.14,69.2591 2019.56,69.2591 "/>
-<path clip-path="url(#clip412)" d="M2019.56 182.086 L2019.56 1423.18 L2080.14 1423.18 L2080.14 182.086 L2019.56 182.086 L2019.56 182.086  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,182.086 2019.56,1423.18 2080.14,1423.18 2080.14,182.086 2019.56,182.086 "/>
-<path clip-path="url(#clip412)" d="M2019.56 294.913 L2019.56 1423.18 L2080.14 1423.18 L2080.14 294.913 L2019.56 294.913 L2019.56 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,294.913 2019.56,1423.18 2080.14,1423.18 2080.14,294.913 2019.56,294.913 "/>
-<path clip-path="url(#clip412)" d="M2019.56 407.739 L2019.56 1423.18 L2080.14 1423.18 L2080.14 407.739 L2019.56 407.739 L2019.56 407.739  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,407.739 2019.56,1423.18 2080.14,1423.18 2080.14,407.739 2019.56,407.739 "/>
-<path clip-path="url(#clip412)" d="M2019.56 520.566 L2019.56 1423.18 L2080.14 1423.18 L2080.14 520.566 L2019.56 520.566 L2019.56 520.566  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,520.566 2019.56,1423.18 2080.14,1423.18 2080.14,520.566 2019.56,520.566 "/>
-<path clip-path="url(#clip412)" d="M2019.56 633.393 L2019.56 1423.18 L2080.14 1423.18 L2080.14 633.393 L2019.56 633.393 L2019.56 633.393  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,633.393 2019.56,1423.18 2080.14,1423.18 2080.14,633.393 2019.56,633.393 "/>
-<path clip-path="url(#clip412)" d="M2019.56 746.219 L2019.56 1423.18 L2080.14 1423.18 L2080.14 746.219 L2019.56 746.219 L2019.56 746.219  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,746.219 2019.56,1423.18 2080.14,1423.18 2080.14,746.219 2019.56,746.219 "/>
-<path clip-path="url(#clip412)" d="M2019.56 859.046 L2019.56 1423.18 L2080.14 1423.18 L2080.14 859.046 L2019.56 859.046 L2019.56 859.046  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,859.046 2019.56,1423.18 2080.14,1423.18 2080.14,859.046 2019.56,859.046 "/>
-<path clip-path="url(#clip412)" d="M2019.56 971.873 L2019.56 1423.18 L2080.14 1423.18 L2080.14 971.873 L2019.56 971.873 L2019.56 971.873  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,971.873 2019.56,1423.18 2080.14,1423.18 2080.14,971.873 2019.56,971.873 "/>
-<path clip-path="url(#clip412)" d="M2019.56 1084.7 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1084.7 L2019.56 1084.7 L2019.56 1084.7  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1084.7 2019.56,1423.18 2080.14,1423.18 2080.14,1084.7 2019.56,1084.7 "/>
-<path clip-path="url(#clip412)" d="M2019.56 1197.53 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1197.53 L2019.56 1197.53 L2019.56 1197.53  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1197.53 2019.56,1423.18 2080.14,1423.18 2080.14,1197.53 2019.56,1197.53 "/>
-<path clip-path="url(#clip412)" d="M2019.56 1310.35 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1310.35 L2019.56 1310.35 L2019.56 1310.35  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1310.35 2019.56,1423.18 2080.14,1423.18 2080.14,1310.35 2019.56,1310.35 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="69.2591" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="182.086" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="294.913" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="407.739" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="520.566" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="633.393" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="746.219" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="859.046" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="971.873" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1084.7" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1197.53" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1310.35" r="2"/>
-<path clip-path="url(#clip412)" d="M2120.53 61.0035 L2120.53 1423.18 L2181.11 1423.18 L2181.11 61.0035 L2120.53 61.0035 L2120.53 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,61.0035 2120.53,1423.18 2181.11,1423.18 2181.11,61.0035 2120.53,61.0035 "/>
-<path clip-path="url(#clip412)" d="M2120.53 151.815 L2120.53 1423.18 L2181.11 1423.18 L2181.11 151.815 L2120.53 151.815 L2120.53 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,151.815 2120.53,1423.18 2181.11,1423.18 2181.11,151.815 2120.53,151.815 "/>
-<path clip-path="url(#clip412)" d="M2120.53 242.627 L2120.53 1423.18 L2181.11 1423.18 L2181.11 242.627 L2120.53 242.627 L2120.53 242.627  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,242.627 2120.53,1423.18 2181.11,1423.18 2181.11,242.627 2120.53,242.627 "/>
-<path clip-path="url(#clip412)" d="M2120.53 333.439 L2120.53 1423.18 L2181.11 1423.18 L2181.11 333.439 L2120.53 333.439 L2120.53 333.439  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,333.439 2120.53,1423.18 2181.11,1423.18 2181.11,333.439 2120.53,333.439 "/>
-<path clip-path="url(#clip412)" d="M2120.53 424.25 L2120.53 1423.18 L2181.11 1423.18 L2181.11 424.25 L2120.53 424.25 L2120.53 424.25  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,424.25 2120.53,1423.18 2181.11,1423.18 2181.11,424.25 2120.53,424.25 "/>
-<path clip-path="url(#clip412)" d="M2120.53 515.062 L2120.53 1423.18 L2181.11 1423.18 L2181.11 515.062 L2120.53 515.062 L2120.53 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,515.062 2120.53,1423.18 2181.11,1423.18 2181.11,515.062 2120.53,515.062 "/>
-<path clip-path="url(#clip412)" d="M2120.53 605.874 L2120.53 1423.18 L2181.11 1423.18 L2181.11 605.874 L2120.53 605.874 L2120.53 605.874  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,605.874 2120.53,1423.18 2181.11,1423.18 2181.11,605.874 2120.53,605.874 "/>
-<path clip-path="url(#clip412)" d="M2120.53 696.686 L2120.53 1423.18 L2181.11 1423.18 L2181.11 696.686 L2120.53 696.686 L2120.53 696.686  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,696.686 2120.53,1423.18 2181.11,1423.18 2181.11,696.686 2120.53,696.686 "/>
-<path clip-path="url(#clip412)" d="M2120.53 787.498 L2120.53 1423.18 L2181.11 1423.18 L2181.11 787.498 L2120.53 787.498 L2120.53 787.498  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,787.498 2120.53,1423.18 2181.11,1423.18 2181.11,787.498 2120.53,787.498 "/>
-<path clip-path="url(#clip412)" d="M2120.53 878.309 L2120.53 1423.18 L2181.11 1423.18 L2181.11 878.309 L2120.53 878.309 L2120.53 878.309  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,878.309 2120.53,1423.18 2181.11,1423.18 2181.11,878.309 2120.53,878.309 "/>
-<path clip-path="url(#clip412)" d="M2120.53 969.121 L2120.53 1423.18 L2181.11 1423.18 L2181.11 969.121 L2120.53 969.121 L2120.53 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,969.121 2120.53,1423.18 2181.11,1423.18 2181.11,969.121 2120.53,969.121 "/>
-<path clip-path="url(#clip412)" d="M2120.53 1059.93 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1059.93 L2120.53 1059.93 L2120.53 1059.93  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1059.93 2120.53,1423.18 2181.11,1423.18 2181.11,1059.93 2120.53,1059.93 "/>
-<path clip-path="url(#clip412)" d="M2120.53 1150.74 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1150.74 L2120.53 1150.74 L2120.53 1150.74  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1150.74 2120.53,1423.18 2181.11,1423.18 2181.11,1150.74 2120.53,1150.74 "/>
-<path clip-path="url(#clip412)" d="M2120.53 1241.56 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1241.56 L2120.53 1241.56 L2120.53 1241.56  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1241.56 2120.53,1423.18 2181.11,1423.18 2181.11,1241.56 2120.53,1241.56 "/>
-<path clip-path="url(#clip412)" d="M2120.53 1332.37 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1332.37 L2120.53 1332.37 L2120.53 1332.37  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1332.37 2120.53,1423.18 2181.11,1423.18 2181.11,1332.37 2120.53,1332.37 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="61.0035" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="151.815" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="242.627" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="333.439" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="424.25" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="515.062" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="605.874" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="696.686" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="787.498" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="878.309" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="969.121" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1059.93" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1150.74" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1241.56" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1332.37" r="2"/>
-<path clip-path="url(#clip412)" d="M2221.5 89.8981 L2221.5 1423.18 L2282.08 1423.18 L2282.08 89.8981 L2221.5 89.8981 L2221.5 89.8981  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,89.8981 2221.5,1423.18 2282.08,1423.18 2282.08,89.8981 2221.5,89.8981 "/>
-<path clip-path="url(#clip412)" d="M2221.5 160.071 L2221.5 1423.18 L2282.08 1423.18 L2282.08 160.071 L2221.5 160.071 L2221.5 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,160.071 2221.5,1423.18 2282.08,1423.18 2282.08,160.071 2221.5,160.071 "/>
-<path clip-path="url(#clip412)" d="M2221.5 230.244 L2221.5 1423.18 L2282.08 1423.18 L2282.08 230.244 L2221.5 230.244 L2221.5 230.244  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,230.244 2221.5,1423.18 2282.08,1423.18 2282.08,230.244 2221.5,230.244 "/>
-<path clip-path="url(#clip412)" d="M2221.5 300.416 L2221.5 1423.18 L2282.08 1423.18 L2282.08 300.416 L2221.5 300.416 L2221.5 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,300.416 2221.5,1423.18 2282.08,1423.18 2282.08,300.416 2221.5,300.416 "/>
-<path clip-path="url(#clip412)" d="M2221.5 370.589 L2221.5 1423.18 L2282.08 1423.18 L2282.08 370.589 L2221.5 370.589 L2221.5 370.589  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,370.589 2221.5,1423.18 2282.08,1423.18 2282.08,370.589 2221.5,370.589 "/>
-<path clip-path="url(#clip412)" d="M2221.5 440.762 L2221.5 1423.18 L2282.08 1423.18 L2282.08 440.762 L2221.5 440.762 L2221.5 440.762  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,440.762 2221.5,1423.18 2282.08,1423.18 2282.08,440.762 2221.5,440.762 "/>
-<path clip-path="url(#clip412)" d="M2221.5 510.934 L2221.5 1423.18 L2282.08 1423.18 L2282.08 510.934 L2221.5 510.934 L2221.5 510.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,510.934 2221.5,1423.18 2282.08,1423.18 2282.08,510.934 2221.5,510.934 "/>
-<path clip-path="url(#clip412)" d="M2221.5 581.107 L2221.5 1423.18 L2282.08 1423.18 L2282.08 581.107 L2221.5 581.107 L2221.5 581.107  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,581.107 2221.5,1423.18 2282.08,1423.18 2282.08,581.107 2221.5,581.107 "/>
-<path clip-path="url(#clip412)" d="M2221.5 651.28 L2221.5 1423.18 L2282.08 1423.18 L2282.08 651.28 L2221.5 651.28 L2221.5 651.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,651.28 2221.5,1423.18 2282.08,1423.18 2282.08,651.28 2221.5,651.28 "/>
-<path clip-path="url(#clip412)" d="M2221.5 721.453 L2221.5 1423.18 L2282.08 1423.18 L2282.08 721.453 L2221.5 721.453 L2221.5 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,721.453 2221.5,1423.18 2282.08,1423.18 2282.08,721.453 2221.5,721.453 "/>
-<path clip-path="url(#clip412)" d="M2221.5 791.625 L2221.5 1423.18 L2282.08 1423.18 L2282.08 791.625 L2221.5 791.625 L2221.5 791.625  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,791.625 2221.5,1423.18 2282.08,1423.18 2282.08,791.625 2221.5,791.625 "/>
-<path clip-path="url(#clip412)" d="M2221.5 861.798 L2221.5 1423.18 L2282.08 1423.18 L2282.08 861.798 L2221.5 861.798 L2221.5 861.798  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,861.798 2221.5,1423.18 2282.08,1423.18 2282.08,861.798 2221.5,861.798 "/>
-<path clip-path="url(#clip412)" d="M2221.5 931.971 L2221.5 1423.18 L2282.08 1423.18 L2282.08 931.971 L2221.5 931.971 L2221.5 931.971  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,931.971 2221.5,1423.18 2282.08,1423.18 2282.08,931.971 2221.5,931.971 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1002.14 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1002.14 L2221.5 1002.14 L2221.5 1002.14  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1002.14 2221.5,1423.18 2282.08,1423.18 2282.08,1002.14 2221.5,1002.14 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1072.32 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1072.32 L2221.5 1072.32 L2221.5 1072.32  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1072.32 2221.5,1423.18 2282.08,1423.18 2282.08,1072.32 2221.5,1072.32 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1142.49 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1142.49 L2221.5 1142.49 L2221.5 1142.49  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1142.49 2221.5,1423.18 2282.08,1423.18 2282.08,1142.49 2221.5,1142.49 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1212.66 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1212.66 L2221.5 1212.66 L2221.5 1212.66  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1212.66 2221.5,1423.18 2282.08,1423.18 2282.08,1212.66 2221.5,1212.66 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1282.83 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1282.83 L2221.5 1282.83 L2221.5 1282.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1282.83 2221.5,1423.18 2282.08,1423.18 2282.08,1282.83 2221.5,1282.83 "/>
-<path clip-path="url(#clip412)" d="M2221.5 1353.01 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1353.01 L2221.5 1353.01 L2221.5 1353.01  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1353.01 2221.5,1423.18 2282.08,1423.18 2282.08,1353.01 2221.5,1353.01 "/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="89.8981" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="160.071" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="230.244" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="300.416" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="370.589" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="440.762" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="510.934" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="581.107" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="651.28" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="721.453" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="791.625" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="861.798" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="931.971" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1002.14" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1072.32" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1142.49" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1212.66" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1282.83" r="2"/>
-<circle clip-path="url(#clip412)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1353.01" r="2"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="535.299,1423.18 535.299,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.208,1423.18 838.208,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1141.12,1423.18 1141.12,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1444.03,1423.18 1444.03,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1746.94,1423.18 1746.94,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2049.85,1423.18 2049.85,47.2441 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2352.76,1423.18 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="535.299,1423.18 535.299,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.208,1423.18 838.208,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1141.12,1423.18 1141.12,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1444.03,1423.18 1444.03,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.94,1423.18 1746.94,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2049.85,1423.18 2049.85,1404.28 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2352.76,1423.18 2352.76,1404.28 "/>
+<path clip-path="url(#clip230)" d="M232.389 1454.1 Q228.778 1454.1 226.95 1457.66 Q225.144 1461.2 225.144 1468.33 Q225.144 1475.44 226.95 1479.01 Q228.778 1482.55 232.389 1482.55 Q236.024 1482.55 237.829 1479.01 Q239.658 1475.44 239.658 1468.33 Q239.658 1461.2 237.829 1457.66 Q236.024 1454.1 232.389 1454.1 M232.389 1450.39 Q238.2 1450.39 241.255 1455 Q244.334 1459.58 244.334 1468.33 Q244.334 1477.06 241.255 1481.67 Q238.2 1486.25 232.389 1486.25 Q226.579 1486.25 223.501 1481.67 Q220.445 1477.06 220.445 1468.33 Q220.445 1459.58 223.501 1455 Q226.579 1450.39 232.389 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M539.547 1466.95 Q542.903 1467.66 544.778 1469.93 Q546.676 1472.2 546.676 1475.53 Q546.676 1480.65 543.158 1483.45 Q539.639 1486.25 533.158 1486.25 Q530.982 1486.25 528.667 1485.81 Q526.375 1485.39 523.922 1484.54 L523.922 1480.02 Q525.866 1481.16 528.181 1481.74 Q530.496 1482.32 533.019 1482.32 Q537.417 1482.32 539.709 1480.58 Q542.023 1478.84 542.023 1475.53 Q542.023 1472.48 539.871 1470.77 Q537.741 1469.03 533.922 1469.03 L529.894 1469.03 L529.894 1465.19 L534.107 1465.19 Q537.556 1465.19 539.384 1463.82 Q541.213 1462.43 541.213 1459.84 Q541.213 1457.18 539.315 1455.77 Q537.44 1454.33 533.922 1454.33 Q532 1454.33 529.801 1454.75 Q527.602 1455.16 524.963 1456.04 L524.963 1451.88 Q527.625 1451.14 529.94 1450.77 Q532.278 1450.39 534.338 1450.39 Q539.662 1450.39 542.764 1452.83 Q545.866 1455.23 545.866 1459.35 Q545.866 1462.22 544.222 1464.21 Q542.579 1466.18 539.547 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M838.613 1466.44 Q835.465 1466.44 833.613 1468.59 Q831.785 1470.74 831.785 1474.49 Q831.785 1478.22 833.613 1480.39 Q835.465 1482.55 838.613 1482.55 Q841.762 1482.55 843.59 1480.39 Q845.442 1478.22 845.442 1474.49 Q845.442 1470.74 843.59 1468.59 Q841.762 1466.44 838.613 1466.44 M847.896 1451.78 L847.896 1456.04 Q846.137 1455.21 844.331 1454.77 Q842.549 1454.33 840.789 1454.33 Q836.16 1454.33 833.706 1457.45 Q831.276 1460.58 830.928 1466.9 Q832.294 1464.89 834.354 1463.82 Q836.414 1462.73 838.891 1462.73 Q844.1 1462.73 847.109 1465.9 Q850.141 1469.05 850.141 1474.49 Q850.141 1479.82 846.993 1483.03 Q843.845 1486.25 838.613 1486.25 Q832.618 1486.25 829.447 1481.67 Q826.276 1477.06 826.276 1468.33 Q826.276 1460.14 830.164 1455.28 Q834.053 1450.39 840.604 1450.39 Q842.363 1450.39 844.146 1450.74 Q845.951 1451.09 847.896 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1131.42 1484.86 L1131.42 1480.6 Q1133.18 1481.44 1134.98 1481.88 Q1136.79 1482.32 1138.53 1482.32 Q1143.15 1482.32 1145.59 1479.21 Q1148.04 1476.09 1148.39 1469.75 Q1147.04 1471.74 1144.98 1472.8 Q1142.92 1473.87 1140.42 1473.87 Q1135.24 1473.87 1132.21 1470.74 Q1129.2 1467.59 1129.2 1462.15 Q1129.2 1456.83 1132.34 1453.61 Q1135.49 1450.39 1140.72 1450.39 Q1146.72 1450.39 1149.87 1455 Q1153.04 1459.58 1153.04 1468.33 Q1153.04 1476.51 1149.15 1481.39 Q1145.28 1486.25 1138.73 1486.25 Q1136.97 1486.25 1135.17 1485.9 Q1133.36 1485.56 1131.42 1484.86 M1140.72 1470.21 Q1143.87 1470.21 1145.7 1468.06 Q1147.55 1465.9 1147.55 1462.15 Q1147.55 1458.43 1145.7 1456.27 Q1143.87 1454.1 1140.72 1454.1 Q1137.58 1454.1 1135.72 1456.27 Q1133.9 1458.43 1133.9 1462.15 Q1133.9 1465.9 1135.72 1468.06 Q1137.58 1470.21 1140.72 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1419.51 1481.64 L1427.15 1481.64 L1427.15 1455.28 L1418.84 1456.95 L1418.84 1452.69 L1427.11 1451.02 L1431.78 1451.02 L1431.78 1481.64 L1439.42 1481.64 L1439.42 1485.58 L1419.51 1485.58 L1419.51 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1452.89 1481.64 L1469.21 1481.64 L1469.21 1485.58 L1447.27 1485.58 L1447.27 1481.64 Q1449.93 1478.89 1454.51 1474.26 Q1459.12 1469.61 1460.3 1468.27 Q1462.55 1465.74 1463.43 1464.01 Q1464.33 1462.25 1464.33 1460.56 Q1464.33 1457.8 1462.38 1456.07 Q1460.46 1454.33 1457.36 1454.33 Q1455.16 1454.33 1452.71 1455.09 Q1450.28 1455.86 1447.5 1457.41 L1447.5 1452.69 Q1450.32 1451.55 1452.78 1450.97 Q1455.23 1450.39 1457.27 1450.39 Q1462.64 1450.39 1465.83 1453.08 Q1469.03 1455.77 1469.03 1460.26 Q1469.03 1462.39 1468.22 1464.31 Q1467.43 1466.2 1465.32 1468.8 Q1464.74 1469.47 1461.64 1472.69 Q1458.54 1475.88 1452.89 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1722.12 1481.64 L1729.76 1481.64 L1729.76 1455.28 L1721.45 1456.95 L1721.45 1452.69 L1729.71 1451.02 L1734.39 1451.02 L1734.39 1481.64 L1742.03 1481.64 L1742.03 1485.58 L1722.12 1485.58 L1722.12 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1751.52 1451.02 L1769.88 1451.02 L1769.88 1454.96 L1755.8 1454.96 L1755.8 1463.43 Q1756.82 1463.08 1757.84 1462.92 Q1758.86 1462.73 1759.88 1462.73 Q1765.66 1462.73 1769.04 1465.9 Q1772.42 1469.08 1772.42 1474.49 Q1772.42 1480.07 1768.95 1483.17 Q1765.48 1486.25 1759.16 1486.25 Q1756.98 1486.25 1754.71 1485.88 Q1752.47 1485.51 1750.06 1484.77 L1750.06 1480.07 Q1752.15 1481.2 1754.37 1481.76 Q1756.59 1482.32 1759.07 1482.32 Q1763.07 1482.32 1765.41 1480.21 Q1767.75 1478.1 1767.75 1474.49 Q1767.75 1470.88 1765.41 1468.77 Q1763.07 1466.67 1759.07 1466.67 Q1757.19 1466.67 1755.32 1467.08 Q1753.46 1467.5 1751.52 1468.38 L1751.52 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M2024.58 1481.64 L2032.22 1481.64 L2032.22 1455.28 L2023.91 1456.95 L2023.91 1452.69 L2032.17 1451.02 L2036.85 1451.02 L2036.85 1481.64 L2044.49 1481.64 L2044.49 1485.58 L2024.58 1485.58 L2024.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M2063.93 1469.17 Q2060.6 1469.17 2058.68 1470.95 Q2056.78 1472.73 2056.78 1475.86 Q2056.78 1478.98 2058.68 1480.77 Q2060.6 1482.55 2063.93 1482.55 Q2067.27 1482.55 2069.19 1480.77 Q2071.11 1478.96 2071.11 1475.86 Q2071.11 1472.73 2069.19 1470.95 Q2067.29 1469.17 2063.93 1469.17 M2059.26 1467.18 Q2056.25 1466.44 2054.56 1464.38 Q2052.89 1462.32 2052.89 1459.35 Q2052.89 1455.21 2055.83 1452.8 Q2058.79 1450.39 2063.93 1450.39 Q2069.09 1450.39 2072.03 1452.8 Q2074.97 1455.21 2074.97 1459.35 Q2074.97 1462.32 2073.28 1464.38 Q2071.62 1466.44 2068.63 1467.18 Q2072.01 1467.96 2073.89 1470.26 Q2075.78 1472.55 2075.78 1475.86 Q2075.78 1480.88 2072.71 1483.57 Q2069.65 1486.25 2063.93 1486.25 Q2058.21 1486.25 2055.14 1483.57 Q2052.08 1480.88 2052.08 1475.86 Q2052.08 1472.55 2053.98 1470.26 Q2055.88 1467.96 2059.26 1467.18 M2057.54 1459.79 Q2057.54 1462.48 2059.21 1463.98 Q2060.9 1465.49 2063.93 1465.49 Q2066.94 1465.49 2068.63 1463.98 Q2070.34 1462.48 2070.34 1459.79 Q2070.34 1457.11 2068.63 1455.6 Q2066.94 1454.1 2063.93 1454.1 Q2060.9 1454.1 2059.21 1455.6 Q2057.54 1457.11 2057.54 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M2332.14 1481.64 L2348.46 1481.64 L2348.46 1485.58 L2326.52 1485.58 L2326.52 1481.64 Q2329.18 1478.89 2333.76 1474.26 Q2338.37 1469.61 2339.55 1468.27 Q2341.8 1465.74 2342.67 1464.01 Q2343.58 1462.25 2343.58 1460.56 Q2343.58 1457.8 2341.63 1456.07 Q2339.71 1454.33 2336.61 1454.33 Q2334.41 1454.33 2331.96 1455.09 Q2329.53 1455.86 2326.75 1457.41 L2326.75 1452.69 Q2329.57 1451.55 2332.03 1450.97 Q2334.48 1450.39 2336.52 1450.39 Q2341.89 1450.39 2345.08 1453.08 Q2348.28 1455.77 2348.28 1460.26 Q2348.28 1462.39 2347.47 1464.31 Q2346.68 1466.2 2344.57 1468.8 Q2343.99 1469.47 2340.89 1472.69 Q2337.79 1475.88 2332.14 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M2359.09 1481.64 L2366.73 1481.64 L2366.73 1455.28 L2358.42 1456.95 L2358.42 1452.69 L2366.68 1451.02 L2371.36 1451.02 L2371.36 1481.64 L2378.99 1481.64 L2378.99 1485.58 L2359.09 1485.58 L2359.09 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1187.71 1525.81 L1187.71 1543.66 L1195.8 1543.66 Q1200.29 1543.66 1202.74 1541.34 Q1205.19 1539.02 1205.19 1534.72 Q1205.19 1530.45 1202.74 1528.13 Q1200.29 1525.81 1195.8 1525.81 L1187.71 1525.81 M1181.28 1520.52 L1195.8 1520.52 Q1203.79 1520.52 1207.86 1524.15 Q1211.97 1527.75 1211.97 1534.72 Q1211.97 1541.75 1207.86 1545.35 Q1203.79 1548.95 1195.8 1548.95 L1187.71 1548.95 L1187.71 1568.04 L1181.28 1568.04 L1181.28 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1233.61 1550.12 Q1226.51 1550.12 1223.78 1551.75 Q1221.04 1553.37 1221.04 1557.29 Q1221.04 1560.4 1223.08 1562.25 Q1225.14 1564.07 1228.68 1564.07 Q1233.55 1564.07 1236.47 1560.63 Q1239.43 1557.16 1239.43 1551.43 L1239.43 1550.12 L1233.61 1550.12 M1245.29 1547.71 L1245.29 1568.04 L1239.43 1568.04 L1239.43 1562.63 Q1237.43 1565.88 1234.44 1567.44 Q1231.45 1568.97 1227.12 1568.97 Q1221.64 1568.97 1218.4 1565.91 Q1215.18 1562.82 1215.18 1557.67 Q1215.18 1551.65 1219.19 1548.6 Q1223.23 1545.54 1231.22 1545.54 L1239.43 1545.54 L1239.43 1544.97 Q1239.43 1540.93 1236.76 1538.73 Q1234.12 1536.5 1229.31 1536.5 Q1226.26 1536.5 1223.36 1537.23 Q1220.47 1537.97 1217.79 1539.43 L1217.79 1534.02 Q1221.01 1532.78 1224.03 1532.17 Q1227.05 1531.54 1229.92 1531.54 Q1237.65 1531.54 1241.47 1535.55 Q1245.29 1539.56 1245.29 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1263.15 1522.27 L1263.15 1532.4 L1275.21 1532.4 L1275.21 1536.95 L1263.15 1536.95 L1263.15 1556.3 Q1263.15 1560.66 1264.32 1561.9 Q1265.53 1563.14 1269.19 1563.14 L1275.21 1563.14 L1275.21 1568.04 L1269.19 1568.04 Q1262.42 1568.04 1259.84 1565.53 Q1257.26 1562.98 1257.26 1556.3 L1257.26 1536.95 L1252.96 1536.95 L1252.96 1532.4 L1257.26 1532.4 L1257.26 1522.27 L1263.15 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1288.71 1522.27 L1288.71 1532.4 L1300.77 1532.4 L1300.77 1536.95 L1288.71 1536.95 L1288.71 1556.3 Q1288.71 1560.66 1289.88 1561.9 Q1291.09 1563.14 1294.75 1563.14 L1300.77 1563.14 L1300.77 1568.04 L1294.75 1568.04 Q1287.97 1568.04 1285.4 1565.53 Q1282.82 1562.98 1282.82 1556.3 L1282.82 1536.95 L1278.52 1536.95 L1278.52 1532.4 L1282.82 1532.4 L1282.82 1522.27 L1288.71 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1338.96 1548.76 L1338.96 1551.62 L1312.04 1551.62 Q1312.42 1557.67 1315.66 1560.85 Q1318.94 1564 1324.77 1564 Q1328.14 1564 1331.29 1563.17 Q1334.47 1562.35 1337.59 1560.69 L1337.59 1566.23 Q1334.44 1567.57 1331.13 1568.27 Q1327.82 1568.97 1324.42 1568.97 Q1315.89 1568.97 1310.89 1564 Q1305.92 1559.04 1305.92 1550.57 Q1305.92 1541.82 1310.64 1536.69 Q1315.38 1531.54 1323.4 1531.54 Q1330.59 1531.54 1334.76 1536.18 Q1338.96 1540.8 1338.96 1548.76 M1333.11 1547.04 Q1333.04 1542.23 1330.4 1539.37 Q1327.79 1536.5 1323.46 1536.5 Q1318.56 1536.5 1315.6 1539.27 Q1312.67 1542.04 1312.23 1547.07 L1333.11 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1369.23 1537.87 Q1368.24 1537.3 1367.07 1537.04 Q1365.92 1536.76 1364.52 1536.76 Q1359.56 1536.76 1356.88 1540 Q1354.24 1543.22 1354.24 1549.27 L1354.24 1568.04 L1348.35 1568.04 L1348.35 1532.4 L1354.24 1532.4 L1354.24 1537.93 Q1356.09 1534.69 1359.05 1533.13 Q1362.01 1531.54 1366.24 1531.54 Q1366.84 1531.54 1367.58 1531.63 Q1368.31 1531.7 1369.2 1531.85 L1369.23 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M1403.86 1546.53 L1403.86 1568.04 L1398 1568.04 L1398 1546.72 Q1398 1541.66 1396.03 1539.14 Q1394.06 1536.63 1390.11 1536.63 Q1385.37 1536.63 1382.63 1539.65 Q1379.89 1542.68 1379.89 1547.9 L1379.89 1568.04 L1374.01 1568.04 L1374.01 1532.4 L1379.89 1532.4 L1379.89 1537.93 Q1381.99 1534.72 1384.83 1533.13 Q1387.69 1531.54 1391.42 1531.54 Q1397.56 1531.54 1400.71 1535.36 Q1403.86 1539.14 1403.86 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,1147.99 2352.76,1147.99 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,872.806 2352.76,872.806 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,597.618 2352.76,597.618 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,322.431 2352.76,322.431 "/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="232.389,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 232.389,47.2441 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1423.18 251.287,1423.18 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,1147.99 251.287,1147.99 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,872.806 251.287,872.806 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,597.618 251.287,597.618 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,322.431 251.287,322.431 "/>
+<polyline clip-path="url(#clip230)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="232.389,47.2441 251.287,47.2441 "/>
+<path clip-path="url(#clip230)" d="M184.445 1408.98 Q180.834 1408.98 179.005 1412.54 Q177.2 1416.08 177.2 1423.21 Q177.2 1430.32 179.005 1433.89 Q180.834 1437.43 184.445 1437.43 Q188.079 1437.43 189.885 1433.89 Q191.713 1430.32 191.713 1423.21 Q191.713 1416.08 189.885 1412.54 Q188.079 1408.98 184.445 1408.98 M184.445 1405.27 Q190.255 1405.27 193.311 1409.88 Q196.389 1414.46 196.389 1423.21 Q196.389 1431.94 193.311 1436.55 Q190.255 1441.13 184.445 1441.13 Q178.635 1441.13 175.556 1436.55 Q172.501 1431.94 172.501 1423.21 Q172.501 1414.46 175.556 1409.88 Q178.635 1405.27 184.445 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M148.311 1161.34 L164.63 1161.34 L164.63 1165.27 L142.686 1165.27 L142.686 1161.34 Q145.348 1158.58 149.931 1153.95 Q154.538 1149.3 155.718 1147.96 Q157.964 1145.43 158.843 1143.7 Q159.746 1141.94 159.746 1140.25 Q159.746 1137.5 157.802 1135.76 Q155.88 1134.02 152.778 1134.02 Q150.579 1134.02 148.126 1134.79 Q145.695 1135.55 142.917 1137.1 L142.917 1132.38 Q145.741 1131.25 148.195 1130.67 Q150.649 1130.09 152.686 1130.09 Q158.056 1130.09 161.251 1132.77 Q164.445 1135.46 164.445 1139.95 Q164.445 1142.08 163.635 1144 Q162.848 1145.9 160.741 1148.49 Q160.163 1149.16 157.061 1152.38 Q153.959 1155.57 148.311 1161.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M184.445 1133.79 Q180.834 1133.79 179.005 1137.36 Q177.2 1140.9 177.2 1148.03 Q177.2 1155.13 179.005 1158.7 Q180.834 1162.24 184.445 1162.24 Q188.079 1162.24 189.885 1158.7 Q191.713 1155.13 191.713 1148.03 Q191.713 1140.9 189.885 1137.36 Q188.079 1133.79 184.445 1133.79 M184.445 1130.09 Q190.255 1130.09 193.311 1134.69 Q196.389 1139.28 196.389 1148.03 Q196.389 1156.75 193.311 1161.36 Q190.255 1165.94 184.445 1165.94 Q178.635 1165.94 175.556 1161.36 Q172.501 1156.75 172.501 1148.03 Q172.501 1139.28 175.556 1134.69 Q178.635 1130.09 184.445 1130.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M157.13 859.6 L145.325 878.049 L157.13 878.049 L157.13 859.6 M155.903 855.526 L161.783 855.526 L161.783 878.049 L166.714 878.049 L166.714 881.937 L161.783 881.937 L161.783 890.086 L157.13 890.086 L157.13 881.937 L141.529 881.937 L141.529 877.424 L155.903 855.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M184.445 858.604 Q180.834 858.604 179.005 862.169 Q177.2 865.711 177.2 872.84 Q177.2 879.947 179.005 883.512 Q180.834 887.053 184.445 887.053 Q188.079 887.053 189.885 883.512 Q191.713 879.947 191.713 872.84 Q191.713 865.711 189.885 862.169 Q188.079 858.604 184.445 858.604 M184.445 854.901 Q190.255 854.901 193.311 859.507 Q196.389 864.09 196.389 872.84 Q196.389 881.567 193.311 886.174 Q190.255 890.757 184.445 890.757 Q178.635 890.757 175.556 886.174 Q172.501 881.567 172.501 872.84 Q172.501 864.09 175.556 859.507 Q178.635 854.901 184.445 854.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M154.862 595.755 Q151.714 595.755 149.862 597.908 Q148.033 600.061 148.033 603.81 Q148.033 607.537 149.862 609.713 Q151.714 611.866 154.862 611.866 Q158.01 611.866 159.839 609.713 Q161.69 607.537 161.69 603.81 Q161.69 600.061 159.839 597.908 Q158.01 595.755 154.862 595.755 M164.144 581.102 L164.144 585.362 Q162.385 584.528 160.579 584.088 Q158.797 583.649 157.038 583.649 Q152.408 583.649 149.954 586.774 Q147.524 589.899 147.177 596.218 Q148.542 594.204 150.603 593.139 Q152.663 592.051 155.14 592.051 Q160.348 592.051 163.357 595.223 Q166.39 598.371 166.39 603.81 Q166.39 609.135 163.241 612.352 Q160.093 615.57 154.862 615.57 Q148.866 615.57 145.695 610.986 Q142.524 606.38 142.524 597.653 Q142.524 589.459 146.413 584.598 Q150.302 579.713 156.853 579.713 Q158.612 579.713 160.394 580.061 Q162.2 580.408 164.144 581.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M184.445 583.417 Q180.834 583.417 179.005 586.982 Q177.2 590.524 177.2 597.653 Q177.2 604.76 179.005 608.324 Q180.834 611.866 184.445 611.866 Q188.079 611.866 189.885 608.324 Q191.713 604.76 191.713 597.653 Q191.713 590.524 189.885 586.982 Q188.079 583.417 184.445 583.417 M184.445 579.713 Q190.255 579.713 193.311 584.32 Q196.389 588.903 196.389 597.653 Q196.389 606.38 193.311 610.986 Q190.255 615.57 184.445 615.57 Q178.635 615.57 175.556 610.986 Q172.501 606.38 172.501 597.653 Q172.501 588.903 175.556 584.32 Q178.635 579.713 184.445 579.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M154.283 323.299 Q150.95 323.299 149.028 325.082 Q147.13 326.864 147.13 329.989 Q147.13 333.114 149.028 334.896 Q150.95 336.679 154.283 336.679 Q157.616 336.679 159.538 334.896 Q161.459 333.091 161.459 329.989 Q161.459 326.864 159.538 325.082 Q157.64 323.299 154.283 323.299 M149.607 321.309 Q146.598 320.568 144.908 318.508 Q143.241 316.447 143.241 313.485 Q143.241 309.341 146.181 306.934 Q149.144 304.526 154.283 304.526 Q159.445 304.526 162.385 306.934 Q165.325 309.341 165.325 313.485 Q165.325 316.447 163.635 318.508 Q161.968 320.568 158.982 321.309 Q162.362 322.096 164.237 324.387 Q166.135 326.679 166.135 329.989 Q166.135 335.012 163.056 337.697 Q160.001 340.383 154.283 340.383 Q148.566 340.383 145.487 337.697 Q142.431 335.012 142.431 329.989 Q142.431 326.679 144.329 324.387 Q146.228 322.096 149.607 321.309 M147.894 313.924 Q147.894 316.61 149.561 318.114 Q151.251 319.619 154.283 319.619 Q157.292 319.619 158.982 318.114 Q160.695 316.61 160.695 313.924 Q160.695 311.239 158.982 309.735 Q157.292 308.23 154.283 308.23 Q151.251 308.23 149.561 309.735 Q147.894 311.239 147.894 313.924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M184.445 308.23 Q180.834 308.23 179.005 311.795 Q177.2 315.336 177.2 322.466 Q177.2 329.572 179.005 333.137 Q180.834 336.679 184.445 336.679 Q188.079 336.679 189.885 333.137 Q191.713 329.572 191.713 322.466 Q191.713 315.336 189.885 311.795 Q188.079 308.23 184.445 308.23 M184.445 304.526 Q190.255 304.526 193.311 309.133 Q196.389 313.716 196.389 322.466 Q196.389 331.193 193.311 335.799 Q190.255 340.383 184.445 340.383 Q178.635 340.383 175.556 335.799 Q172.501 331.193 172.501 322.466 Q172.501 313.716 175.556 309.133 Q178.635 304.526 184.445 304.526 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M114.931 60.5889 L122.57 60.5889 L122.57 34.2233 L114.26 35.89 L114.26 31.6308 L122.524 29.9641 L127.2 29.9641 L127.2 60.5889 L134.839 60.5889 L134.839 64.5241 L114.931 64.5241 L114.931 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M154.283 33.0428 Q150.672 33.0428 148.843 36.6076 Q147.038 40.1492 147.038 47.2788 Q147.038 54.3853 148.843 57.9501 Q150.672 61.4917 154.283 61.4917 Q157.917 61.4917 159.723 57.9501 Q161.552 54.3853 161.552 47.2788 Q161.552 40.1492 159.723 36.6076 Q157.917 33.0428 154.283 33.0428 M154.283 29.3391 Q160.093 29.3391 163.149 33.9456 Q166.227 38.5289 166.227 47.2788 Q166.227 56.0056 163.149 60.6121 Q160.093 65.1954 154.283 65.1954 Q148.473 65.1954 145.394 60.6121 Q142.339 56.0056 142.339 47.2788 Q142.339 38.5289 145.394 33.9456 Q148.473 29.3391 154.283 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M184.445 33.0428 Q180.834 33.0428 179.005 36.6076 Q177.2 40.1492 177.2 47.2788 Q177.2 54.3853 179.005 57.9501 Q180.834 61.4917 184.445 61.4917 Q188.079 61.4917 189.885 57.9501 Q191.713 54.3853 191.713 47.2788 Q191.713 40.1492 189.885 36.6076 Q188.079 33.0428 184.445 33.0428 M184.445 29.3391 Q190.255 29.3391 193.311 33.9456 Q196.389 38.5289 196.389 47.2788 Q196.389 56.0056 193.311 60.6121 Q190.255 65.1954 184.445 65.1954 Q178.635 65.1954 175.556 60.6121 Q172.501 56.0056 172.501 47.2788 Q172.501 38.5289 175.556 33.9456 Q178.635 29.3391 184.445 29.3391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M41.7242 880.191 Q42.4244 878.122 44.7161 876.181 Q47.0077 874.207 51.0181 872.234 L64.0042 865.709 L64.0042 872.616 L51.8138 878.695 Q47.0395 881.05 45.48 883.278 Q43.9204 885.474 43.9204 889.294 L43.9204 896.296 L64.0042 896.296 L64.0042 902.726 L16.4842 902.726 L16.4842 888.212 Q16.4842 880.064 19.8898 876.053 Q23.2955 872.043 30.1704 872.043 Q34.6582 872.043 37.6183 874.144 Q40.5784 876.212 41.7242 880.191 M21.7677 896.296 L38.6368 896.296 L38.6368 888.212 Q38.6368 883.565 36.5043 881.209 Q34.34 878.822 30.1704 878.822 Q26.0009 878.822 23.9002 881.209 Q21.7677 883.565 21.7677 888.212 L21.7677 896.296 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M32.4621 846.803 Q32.4621 851.513 36.1542 854.251 Q39.8145 856.988 46.212 856.988 Q52.6095 856.988 56.3017 854.283 Q59.9619 851.545 59.9619 846.803 Q59.9619 842.124 56.2698 839.387 Q52.5777 836.65 46.212 836.65 Q39.8781 836.65 36.186 839.387 Q32.4621 842.124 32.4621 846.803 M27.4968 846.803 Q27.4968 839.164 32.4621 834.803 Q37.4273 830.443 46.212 830.443 Q54.9649 830.443 59.9619 834.803 Q64.9272 839.164 64.9272 846.803 Q64.9272 854.473 59.9619 858.834 Q54.9649 863.163 46.212 863.163 Q37.4273 863.163 32.4621 858.834 Q27.4968 854.473 27.4968 846.803 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M14.479 820.735 L14.479 814.879 L64.0042 814.879 L64.0042 820.735 L14.479 820.735 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M14.479 802.625 L14.479 796.768 L64.0042 796.768 L64.0042 802.625 L14.479 802.625 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M14.479 763.794 L14.479 757.938 L64.0042 757.938 L64.0042 763.794 L14.479 763.794 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M44.7161 715.192 L47.5806 715.192 L47.5806 742.119 Q53.6281 741.737 56.8109 738.49 Q59.9619 735.212 59.9619 729.387 Q59.9619 726.014 59.1344 722.862 Q58.3069 719.68 56.6518 716.56 L62.1899 716.56 Q63.5267 719.711 64.227 723.022 Q64.9272 726.332 64.9272 729.737 Q64.9272 738.268 59.9619 743.265 Q54.9967 748.23 46.5303 748.23 Q37.7774 748.23 32.6531 743.519 Q27.4968 738.777 27.4968 730.756 Q27.4968 723.563 32.1438 719.393 Q36.7589 715.192 44.7161 715.192 M42.9973 721.048 Q38.1912 721.112 35.3266 723.754 Q32.4621 726.364 32.4621 730.692 Q32.4621 735.594 35.2312 738.554 Q38.0002 741.482 43.0292 741.928 L42.9973 721.048 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M42.4881 675.947 L64.0042 675.947 L64.0042 681.804 L42.679 681.804 Q37.6183 681.804 35.1038 683.777 Q32.5894 685.75 32.5894 689.697 Q32.5894 694.44 35.6131 697.177 Q38.6368 699.914 43.8567 699.914 L64.0042 699.914 L64.0042 705.802 L28.3562 705.802 L28.3562 699.914 L33.8944 699.914 Q30.6797 697.813 29.0883 694.981 Q27.4968 692.116 27.4968 688.392 Q27.4968 682.249 31.3163 679.098 Q35.1038 675.947 42.4881 675.947 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M45.7664 640.809 Q39.4007 640.809 35.8996 643.45 Q32.3984 646.06 32.3984 650.803 Q32.3984 655.513 35.8996 658.155 Q39.4007 660.765 45.7664 660.765 Q52.1003 660.765 55.6014 658.155 Q59.1026 655.513 59.1026 650.803 Q59.1026 646.06 55.6014 643.45 Q52.1003 640.809 45.7664 640.809 M59.58 634.952 Q68.683 634.952 73.1071 638.994 Q77.5631 643.037 77.5631 651.376 Q77.5631 654.463 77.0857 657.2 Q76.6401 659.937 75.6852 662.516 L69.9879 662.516 Q71.3884 659.937 72.0568 657.423 Q72.7252 654.909 72.7252 652.299 Q72.7252 646.538 69.7015 643.673 Q66.7096 640.809 60.6303 640.809 L57.7339 640.809 Q60.885 642.623 62.4446 645.456 Q64.0042 648.288 64.0042 652.235 Q64.0042 658.792 59.0071 662.802 Q54.01 666.812 45.7664 666.812 Q37.491 666.812 32.4939 662.802 Q27.4968 658.792 27.4968 652.235 Q27.4968 648.288 29.0564 645.456 Q30.616 642.623 33.7671 640.809 L28.3562 640.809 L28.3562 634.952 L59.58 634.952 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M18.2347 617.096 L28.3562 617.096 L28.3562 605.033 L32.9077 605.033 L32.9077 617.096 L52.2594 617.096 Q56.6199 617.096 57.8613 615.919 Q59.1026 614.709 59.1026 611.049 L59.1026 605.033 L64.0042 605.033 L64.0042 611.049 Q64.0042 617.828 61.4897 620.406 Q58.9434 622.985 52.2594 622.985 L32.9077 622.985 L32.9077 627.281 L28.3562 627.281 L28.3562 622.985 L18.2347 622.985 L18.2347 617.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip230)" d="M42.4881 567.698 L64.0042 567.698 L64.0042 573.555 L42.679 573.555 Q37.6183 573.555 35.1038 575.528 Q32.5894 577.502 32.5894 581.448 Q32.5894 586.191 35.6131 588.928 Q38.6368 591.665 43.8567 591.665 L64.0042 591.665 L64.0042 597.554 L14.479 597.554 L14.479 591.665 L33.8944 591.665 Q30.6797 589.565 29.0883 586.732 Q27.4968 583.867 27.4968 580.143 Q27.4968 574 31.3163 570.849 Q35.1038 567.698 42.4881 567.698 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip232)" d="M303.068 391.228 L303.068 1423.18 L363.65 1423.18 L363.65 391.228 L303.068 391.228 L303.068 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.068,391.228 303.068,1423.18 363.65,1423.18 363.65,391.228 303.068,391.228 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="333.359" cy="391.228" r="2"/>
+<path clip-path="url(#clip232)" d="M404.038 391.228 L404.038 1423.18 L464.62 1423.18 L464.62 391.228 L404.038 391.228 L404.038 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="404.038,391.228 404.038,1423.18 464.62,1423.18 464.62,391.228 404.038,391.228 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="434.329" cy="391.228" r="2"/>
+<path clip-path="url(#clip232)" d="M505.008 391.228 L505.008 1423.18 L565.59 1423.18 L565.59 391.228 L505.008 391.228 L505.008 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="505.008,391.228 505.008,1423.18 565.59,1423.18 565.59,391.228 505.008,391.228 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="535.299" cy="391.228" r="2"/>
+<path clip-path="url(#clip232)" d="M605.978 391.228 L605.978 1423.18 L666.56 1423.18 L666.56 391.228 L605.978 391.228 L605.978 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="605.978,391.228 605.978,1423.18 666.56,1423.18 666.56,391.228 605.978,391.228 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="636.269" cy="391.228" r="2"/>
+<path clip-path="url(#clip232)" d="M706.948 391.228 L706.948 1423.18 L767.529 1423.18 L767.529 391.228 L706.948 391.228 L706.948 391.228  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="706.948,391.228 706.948,1423.18 767.529,1423.18 767.529,391.228 706.948,391.228 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="737.239" cy="391.228" r="2"/>
+<path clip-path="url(#clip232)" d="M807.917 682.926 L807.917 1423.18 L868.499 1423.18 L868.499 682.926 L807.917 682.926 L807.917 682.926  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="807.917,682.926 807.917,1423.18 868.499,1423.18 868.499,682.926 807.917,682.926 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="838.208" cy="682.926" r="2"/>
+<path clip-path="url(#clip232)" d="M908.887 693.934 L908.887 1423.18 L969.469 1423.18 L969.469 693.934 L908.887 693.934 L908.887 693.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="908.887,693.934 908.887,1423.18 969.469,1423.18 969.469,693.934 908.887,693.934 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="939.178" cy="693.934" r="2"/>
+<path clip-path="url(#clip232)" d="M1009.86 721.453 L1009.86 1423.18 L1070.44 1423.18 L1070.44 721.453 L1009.86 721.453 L1009.86 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1009.86,721.453 1009.86,1423.18 1070.44,1423.18 1070.44,721.453 1009.86,721.453 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1040.15" cy="721.453" r="2"/>
+<path clip-path="url(#clip232)" d="M1110.83 732.46 L1110.83 1423.18 L1171.41 1423.18 L1171.41 732.46 L1110.83 732.46 L1110.83 732.46  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1110.83,732.46 1110.83,1423.18 1171.41,1423.18 1171.41,732.46 1110.83,732.46 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1141.12" cy="732.46" r="2"/>
+<path clip-path="url(#clip232)" d="M1211.8 94.0259 L1211.8 1423.18 L1272.38 1423.18 L1272.38 94.0259 L1211.8 94.0259 L1211.8 94.0259  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,94.0259 1211.8,1423.18 1272.38,1423.18 1272.38,94.0259 1211.8,94.0259 "/>
+<path clip-path="url(#clip232)" d="M1211.8 537.077 L1211.8 1423.18 L1272.38 1423.18 L1272.38 537.077 L1211.8 537.077 L1211.8 537.077  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,537.077 1211.8,1423.18 1272.38,1423.18 1272.38,537.077 1211.8,537.077 "/>
+<path clip-path="url(#clip232)" d="M1211.8 980.129 L1211.8 1423.18 L1272.38 1423.18 L1272.38 980.129 L1211.8 980.129 L1211.8 980.129  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1211.8,980.129 1211.8,1423.18 1272.38,1423.18 1272.38,980.129 1211.8,980.129 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="94.0259" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="537.077" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1242.09" cy="980.129" r="2"/>
+<path clip-path="url(#clip232)" d="M1312.77 151.815 L1312.77 1423.18 L1373.35 1423.18 L1373.35 151.815 L1312.77 151.815 L1312.77 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,151.815 1312.77,1423.18 1373.35,1423.18 1373.35,151.815 1312.77,151.815 "/>
+<path clip-path="url(#clip232)" d="M1312.77 575.603 L1312.77 1423.18 L1373.35 1423.18 L1373.35 575.603 L1312.77 575.603 L1312.77 575.603  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,575.603 1312.77,1423.18 1373.35,1423.18 1373.35,575.603 1312.77,575.603 "/>
+<path clip-path="url(#clip232)" d="M1312.77 999.392 L1312.77 1423.18 L1373.35 1423.18 L1373.35 999.392 L1312.77 999.392 L1312.77 999.392  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1312.77,999.392 1312.77,1423.18 1373.35,1423.18 1373.35,999.392 1312.77,999.392 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="151.815" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="575.603" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1343.06" cy="999.392" r="2"/>
+<path clip-path="url(#clip232)" d="M1413.74 193.093 L1413.74 1423.18 L1474.32 1423.18 L1474.32 193.093 L1413.74 193.093 L1413.74 193.093  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,193.093 1413.74,1423.18 1474.32,1423.18 1474.32,193.093 1413.74,193.093 "/>
+<path clip-path="url(#clip232)" d="M1413.74 603.122 L1413.74 1423.18 L1474.32 1423.18 L1474.32 603.122 L1413.74 603.122 L1413.74 603.122  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,603.122 1413.74,1423.18 1474.32,1423.18 1474.32,603.122 1413.74,603.122 "/>
+<path clip-path="url(#clip232)" d="M1413.74 1013.15 L1413.74 1423.18 L1474.32 1423.18 L1474.32 1013.15 L1413.74 1013.15 L1413.74 1013.15  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.74,1013.15 1413.74,1423.18 1474.32,1423.18 1474.32,1013.15 1413.74,1013.15 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="193.093" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="603.122" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1444.03" cy="1013.15" r="2"/>
+<path clip-path="url(#clip232)" d="M1514.71 316.928 L1514.71 1423.18 L1575.29 1423.18 L1575.29 316.928 L1514.71 316.928 L1514.71 316.928  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,316.928 1514.71,1423.18 1575.29,1423.18 1575.29,316.928 1514.71,316.928 "/>
+<path clip-path="url(#clip232)" d="M1514.71 593.491 L1514.71 1423.18 L1575.29 1423.18 L1575.29 593.491 L1514.71 593.491 L1514.71 593.491  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,593.491 1514.71,1423.18 1575.29,1423.18 1575.29,593.491 1514.71,593.491 "/>
+<path clip-path="url(#clip232)" d="M1514.71 870.054 L1514.71 1423.18 L1575.29 1423.18 L1575.29 870.054 L1514.71 870.054 L1514.71 870.054  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,870.054 1514.71,1423.18 1575.29,1423.18 1575.29,870.054 1514.71,870.054 "/>
+<path clip-path="url(#clip232)" d="M1514.71 1146.62 L1514.71 1423.18 L1575.29 1423.18 L1575.29 1146.62 L1514.71 1146.62 L1514.71 1146.62  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1514.71,1146.62 1514.71,1423.18 1575.29,1423.18 1575.29,1146.62 1514.71,1146.62 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="316.928" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="593.491" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="870.054" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1545" cy="1146.62" r="2"/>
+<path clip-path="url(#clip232)" d="M1615.68 85.7703 L1615.68 1423.18 L1676.26 1423.18 L1676.26 85.7703 L1615.68 85.7703 L1615.68 85.7703  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,85.7703 1615.68,1423.18 1676.26,1423.18 1676.26,85.7703 1615.68,85.7703 "/>
+<path clip-path="url(#clip232)" d="M1615.68 308.672 L1615.68 1423.18 L1676.26 1423.18 L1676.26 308.672 L1615.68 308.672 L1615.68 308.672  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,308.672 1615.68,1423.18 1676.26,1423.18 1676.26,308.672 1615.68,308.672 "/>
+<path clip-path="url(#clip232)" d="M1615.68 531.573 L1615.68 1423.18 L1676.26 1423.18 L1676.26 531.573 L1615.68 531.573 L1615.68 531.573  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,531.573 1615.68,1423.18 1676.26,1423.18 1676.26,531.573 1615.68,531.573 "/>
+<path clip-path="url(#clip232)" d="M1615.68 754.475 L1615.68 1423.18 L1676.26 1423.18 L1676.26 754.475 L1615.68 754.475 L1615.68 754.475  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,754.475 1615.68,1423.18 1676.26,1423.18 1676.26,754.475 1615.68,754.475 "/>
+<path clip-path="url(#clip232)" d="M1615.68 977.377 L1615.68 1423.18 L1676.26 1423.18 L1676.26 977.377 L1615.68 977.377 L1615.68 977.377  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,977.377 1615.68,1423.18 1676.26,1423.18 1676.26,977.377 1615.68,977.377 "/>
+<path clip-path="url(#clip232)" d="M1615.68 1200.28 L1615.68 1423.18 L1676.26 1423.18 L1676.26 1200.28 L1615.68 1200.28 L1615.68 1200.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1615.68,1200.28 1615.68,1423.18 1676.26,1423.18 1676.26,1200.28 1615.68,1200.28 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="85.7703" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="308.672" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="531.573" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="754.475" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="977.377" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1645.97" cy="1200.28" r="2"/>
+<path clip-path="url(#clip232)" d="M1716.65 226.116 L1716.65 1423.18 L1777.23 1423.18 L1777.23 226.116 L1716.65 226.116 L1716.65 226.116  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,226.116 1716.65,1423.18 1777.23,1423.18 1777.23,226.116 1716.65,226.116 "/>
+<path clip-path="url(#clip232)" d="M1716.65 425.626 L1716.65 1423.18 L1777.23 1423.18 L1777.23 425.626 L1716.65 425.626 L1716.65 425.626  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,425.626 1716.65,1423.18 1777.23,1423.18 1777.23,425.626 1716.65,425.626 "/>
+<path clip-path="url(#clip232)" d="M1716.65 625.137 L1716.65 1423.18 L1777.23 1423.18 L1777.23 625.137 L1716.65 625.137 L1716.65 625.137  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,625.137 1716.65,1423.18 1777.23,1423.18 1777.23,625.137 1716.65,625.137 "/>
+<path clip-path="url(#clip232)" d="M1716.65 824.648 L1716.65 1423.18 L1777.23 1423.18 L1777.23 824.648 L1716.65 824.648 L1716.65 824.648  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,824.648 1716.65,1423.18 1777.23,1423.18 1777.23,824.648 1716.65,824.648 "/>
+<path clip-path="url(#clip232)" d="M1716.65 1024.16 L1716.65 1423.18 L1777.23 1423.18 L1777.23 1024.16 L1716.65 1024.16 L1716.65 1024.16  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,1024.16 1716.65,1423.18 1777.23,1423.18 1777.23,1024.16 1716.65,1024.16 "/>
+<path clip-path="url(#clip232)" d="M1716.65 1223.67 L1716.65 1423.18 L1777.23 1423.18 L1777.23 1223.67 L1716.65 1223.67 L1716.65 1223.67  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1716.65,1223.67 1716.65,1423.18 1777.23,1423.18 1777.23,1223.67 1716.65,1223.67 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="226.116" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="425.626" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="625.137" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="824.648" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="1024.16" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1746.94" cy="1223.67" r="2"/>
+<path clip-path="url(#clip232)" d="M1817.62 61.0035 L1817.62 1423.18 L1878.2 1423.18 L1878.2 61.0035 L1817.62 61.0035 L1817.62 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,61.0035 1817.62,1423.18 1878.2,1423.18 1878.2,61.0035 1817.62,61.0035 "/>
+<path clip-path="url(#clip232)" d="M1817.62 212.356 L1817.62 1423.18 L1878.2 1423.18 L1878.2 212.356 L1817.62 212.356 L1817.62 212.356  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,212.356 1817.62,1423.18 1878.2,1423.18 1878.2,212.356 1817.62,212.356 "/>
+<path clip-path="url(#clip232)" d="M1817.62 363.709 L1817.62 1423.18 L1878.2 1423.18 L1878.2 363.709 L1817.62 363.709 L1817.62 363.709  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,363.709 1817.62,1423.18 1878.2,1423.18 1878.2,363.709 1817.62,363.709 "/>
+<path clip-path="url(#clip232)" d="M1817.62 515.062 L1817.62 1423.18 L1878.2 1423.18 L1878.2 515.062 L1817.62 515.062 L1817.62 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,515.062 1817.62,1423.18 1878.2,1423.18 1878.2,515.062 1817.62,515.062 "/>
+<path clip-path="url(#clip232)" d="M1817.62 666.415 L1817.62 1423.18 L1878.2 1423.18 L1878.2 666.415 L1817.62 666.415 L1817.62 666.415  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,666.415 1817.62,1423.18 1878.2,1423.18 1878.2,666.415 1817.62,666.415 "/>
+<path clip-path="url(#clip232)" d="M1817.62 817.768 L1817.62 1423.18 L1878.2 1423.18 L1878.2 817.768 L1817.62 817.768 L1817.62 817.768  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,817.768 1817.62,1423.18 1878.2,1423.18 1878.2,817.768 1817.62,817.768 "/>
+<path clip-path="url(#clip232)" d="M1817.62 969.121 L1817.62 1423.18 L1878.2 1423.18 L1878.2 969.121 L1817.62 969.121 L1817.62 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,969.121 1817.62,1423.18 1878.2,1423.18 1878.2,969.121 1817.62,969.121 "/>
+<path clip-path="url(#clip232)" d="M1817.62 1120.47 L1817.62 1423.18 L1878.2 1423.18 L1878.2 1120.47 L1817.62 1120.47 L1817.62 1120.47  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,1120.47 1817.62,1423.18 1878.2,1423.18 1878.2,1120.47 1817.62,1120.47 "/>
+<path clip-path="url(#clip232)" d="M1817.62 1271.83 L1817.62 1423.18 L1878.2 1423.18 L1878.2 1271.83 L1817.62 1271.83 L1817.62 1271.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1817.62,1271.83 1817.62,1423.18 1878.2,1423.18 1878.2,1271.83 1817.62,1271.83 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="61.0035" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="212.356" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="363.709" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="515.062" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="666.415" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="817.768" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="969.121" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="1120.47" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1847.91" cy="1271.83" r="2"/>
+<path clip-path="url(#clip232)" d="M1918.59 121.545 L1918.59 1423.18 L1979.17 1423.18 L1979.17 121.545 L1918.59 121.545 L1918.59 121.545  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,121.545 1918.59,1423.18 1979.17,1423.18 1979.17,121.545 1918.59,121.545 "/>
+<path clip-path="url(#clip232)" d="M1918.59 239.875 L1918.59 1423.18 L1979.17 1423.18 L1979.17 239.875 L1918.59 239.875 L1918.59 239.875  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,239.875 1918.59,1423.18 1979.17,1423.18 1979.17,239.875 1918.59,239.875 "/>
+<path clip-path="url(#clip232)" d="M1918.59 358.206 L1918.59 1423.18 L1979.17 1423.18 L1979.17 358.206 L1918.59 358.206 L1918.59 358.206  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,358.206 1918.59,1423.18 1979.17,1423.18 1979.17,358.206 1918.59,358.206 "/>
+<path clip-path="url(#clip232)" d="M1918.59 476.536 L1918.59 1423.18 L1979.17 1423.18 L1979.17 476.536 L1918.59 476.536 L1918.59 476.536  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,476.536 1918.59,1423.18 1979.17,1423.18 1979.17,476.536 1918.59,476.536 "/>
+<path clip-path="url(#clip232)" d="M1918.59 594.867 L1918.59 1423.18 L1979.17 1423.18 L1979.17 594.867 L1918.59 594.867 L1918.59 594.867  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,594.867 1918.59,1423.18 1979.17,1423.18 1979.17,594.867 1918.59,594.867 "/>
+<path clip-path="url(#clip232)" d="M1918.59 713.197 L1918.59 1423.18 L1979.17 1423.18 L1979.17 713.197 L1918.59 713.197 L1918.59 713.197  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,713.197 1918.59,1423.18 1979.17,1423.18 1979.17,713.197 1918.59,713.197 "/>
+<path clip-path="url(#clip232)" d="M1918.59 831.527 L1918.59 1423.18 L1979.17 1423.18 L1979.17 831.527 L1918.59 831.527 L1918.59 831.527  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,831.527 1918.59,1423.18 1979.17,1423.18 1979.17,831.527 1918.59,831.527 "/>
+<path clip-path="url(#clip232)" d="M1918.59 949.858 L1918.59 1423.18 L1979.17 1423.18 L1979.17 949.858 L1918.59 949.858 L1918.59 949.858  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,949.858 1918.59,1423.18 1979.17,1423.18 1979.17,949.858 1918.59,949.858 "/>
+<path clip-path="url(#clip232)" d="M1918.59 1068.19 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1068.19 L1918.59 1068.19 L1918.59 1068.19  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1068.19 1918.59,1423.18 1979.17,1423.18 1979.17,1068.19 1918.59,1068.19 "/>
+<path clip-path="url(#clip232)" d="M1918.59 1186.52 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1186.52 L1918.59 1186.52 L1918.59 1186.52  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1186.52 1918.59,1423.18 1979.17,1423.18 1979.17,1186.52 1918.59,1186.52 "/>
+<path clip-path="url(#clip232)" d="M1918.59 1304.85 L1918.59 1423.18 L1979.17 1423.18 L1979.17 1304.85 L1918.59 1304.85 L1918.59 1304.85  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1918.59,1304.85 1918.59,1423.18 1979.17,1423.18 1979.17,1304.85 1918.59,1304.85 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="121.545" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="239.875" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="358.206" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="476.536" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="594.867" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="713.197" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="831.527" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="949.858" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1068.19" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1186.52" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="1948.88" cy="1304.85" r="2"/>
+<path clip-path="url(#clip232)" d="M2019.56 69.2591 L2019.56 1423.18 L2080.14 1423.18 L2080.14 69.2591 L2019.56 69.2591 L2019.56 69.2591  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,69.2591 2019.56,1423.18 2080.14,1423.18 2080.14,69.2591 2019.56,69.2591 "/>
+<path clip-path="url(#clip232)" d="M2019.56 182.086 L2019.56 1423.18 L2080.14 1423.18 L2080.14 182.086 L2019.56 182.086 L2019.56 182.086  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,182.086 2019.56,1423.18 2080.14,1423.18 2080.14,182.086 2019.56,182.086 "/>
+<path clip-path="url(#clip232)" d="M2019.56 294.913 L2019.56 1423.18 L2080.14 1423.18 L2080.14 294.913 L2019.56 294.913 L2019.56 294.913  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,294.913 2019.56,1423.18 2080.14,1423.18 2080.14,294.913 2019.56,294.913 "/>
+<path clip-path="url(#clip232)" d="M2019.56 407.739 L2019.56 1423.18 L2080.14 1423.18 L2080.14 407.739 L2019.56 407.739 L2019.56 407.739  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,407.739 2019.56,1423.18 2080.14,1423.18 2080.14,407.739 2019.56,407.739 "/>
+<path clip-path="url(#clip232)" d="M2019.56 520.566 L2019.56 1423.18 L2080.14 1423.18 L2080.14 520.566 L2019.56 520.566 L2019.56 520.566  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,520.566 2019.56,1423.18 2080.14,1423.18 2080.14,520.566 2019.56,520.566 "/>
+<path clip-path="url(#clip232)" d="M2019.56 633.393 L2019.56 1423.18 L2080.14 1423.18 L2080.14 633.393 L2019.56 633.393 L2019.56 633.393  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,633.393 2019.56,1423.18 2080.14,1423.18 2080.14,633.393 2019.56,633.393 "/>
+<path clip-path="url(#clip232)" d="M2019.56 746.219 L2019.56 1423.18 L2080.14 1423.18 L2080.14 746.219 L2019.56 746.219 L2019.56 746.219  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,746.219 2019.56,1423.18 2080.14,1423.18 2080.14,746.219 2019.56,746.219 "/>
+<path clip-path="url(#clip232)" d="M2019.56 859.046 L2019.56 1423.18 L2080.14 1423.18 L2080.14 859.046 L2019.56 859.046 L2019.56 859.046  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,859.046 2019.56,1423.18 2080.14,1423.18 2080.14,859.046 2019.56,859.046 "/>
+<path clip-path="url(#clip232)" d="M2019.56 971.873 L2019.56 1423.18 L2080.14 1423.18 L2080.14 971.873 L2019.56 971.873 L2019.56 971.873  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,971.873 2019.56,1423.18 2080.14,1423.18 2080.14,971.873 2019.56,971.873 "/>
+<path clip-path="url(#clip232)" d="M2019.56 1084.7 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1084.7 L2019.56 1084.7 L2019.56 1084.7  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1084.7 2019.56,1423.18 2080.14,1423.18 2080.14,1084.7 2019.56,1084.7 "/>
+<path clip-path="url(#clip232)" d="M2019.56 1197.53 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1197.53 L2019.56 1197.53 L2019.56 1197.53  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1197.53 2019.56,1423.18 2080.14,1423.18 2080.14,1197.53 2019.56,1197.53 "/>
+<path clip-path="url(#clip232)" d="M2019.56 1310.35 L2019.56 1423.18 L2080.14 1423.18 L2080.14 1310.35 L2019.56 1310.35 L2019.56 1310.35  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.56,1310.35 2019.56,1423.18 2080.14,1423.18 2080.14,1310.35 2019.56,1310.35 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="69.2591" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="182.086" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="294.913" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="407.739" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="520.566" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="633.393" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="746.219" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="859.046" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="971.873" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1084.7" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1197.53" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2049.85" cy="1310.35" r="2"/>
+<path clip-path="url(#clip232)" d="M2120.53 61.0035 L2120.53 1423.18 L2181.11 1423.18 L2181.11 61.0035 L2120.53 61.0035 L2120.53 61.0035  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,61.0035 2120.53,1423.18 2181.11,1423.18 2181.11,61.0035 2120.53,61.0035 "/>
+<path clip-path="url(#clip232)" d="M2120.53 151.815 L2120.53 1423.18 L2181.11 1423.18 L2181.11 151.815 L2120.53 151.815 L2120.53 151.815  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,151.815 2120.53,1423.18 2181.11,1423.18 2181.11,151.815 2120.53,151.815 "/>
+<path clip-path="url(#clip232)" d="M2120.53 242.627 L2120.53 1423.18 L2181.11 1423.18 L2181.11 242.627 L2120.53 242.627 L2120.53 242.627  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,242.627 2120.53,1423.18 2181.11,1423.18 2181.11,242.627 2120.53,242.627 "/>
+<path clip-path="url(#clip232)" d="M2120.53 333.439 L2120.53 1423.18 L2181.11 1423.18 L2181.11 333.439 L2120.53 333.439 L2120.53 333.439  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,333.439 2120.53,1423.18 2181.11,1423.18 2181.11,333.439 2120.53,333.439 "/>
+<path clip-path="url(#clip232)" d="M2120.53 424.25 L2120.53 1423.18 L2181.11 1423.18 L2181.11 424.25 L2120.53 424.25 L2120.53 424.25  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,424.25 2120.53,1423.18 2181.11,1423.18 2181.11,424.25 2120.53,424.25 "/>
+<path clip-path="url(#clip232)" d="M2120.53 515.062 L2120.53 1423.18 L2181.11 1423.18 L2181.11 515.062 L2120.53 515.062 L2120.53 515.062  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,515.062 2120.53,1423.18 2181.11,1423.18 2181.11,515.062 2120.53,515.062 "/>
+<path clip-path="url(#clip232)" d="M2120.53 605.874 L2120.53 1423.18 L2181.11 1423.18 L2181.11 605.874 L2120.53 605.874 L2120.53 605.874  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,605.874 2120.53,1423.18 2181.11,1423.18 2181.11,605.874 2120.53,605.874 "/>
+<path clip-path="url(#clip232)" d="M2120.53 696.686 L2120.53 1423.18 L2181.11 1423.18 L2181.11 696.686 L2120.53 696.686 L2120.53 696.686  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,696.686 2120.53,1423.18 2181.11,1423.18 2181.11,696.686 2120.53,696.686 "/>
+<path clip-path="url(#clip232)" d="M2120.53 787.498 L2120.53 1423.18 L2181.11 1423.18 L2181.11 787.498 L2120.53 787.498 L2120.53 787.498  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,787.498 2120.53,1423.18 2181.11,1423.18 2181.11,787.498 2120.53,787.498 "/>
+<path clip-path="url(#clip232)" d="M2120.53 878.309 L2120.53 1423.18 L2181.11 1423.18 L2181.11 878.309 L2120.53 878.309 L2120.53 878.309  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,878.309 2120.53,1423.18 2181.11,1423.18 2181.11,878.309 2120.53,878.309 "/>
+<path clip-path="url(#clip232)" d="M2120.53 969.121 L2120.53 1423.18 L2181.11 1423.18 L2181.11 969.121 L2120.53 969.121 L2120.53 969.121  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,969.121 2120.53,1423.18 2181.11,1423.18 2181.11,969.121 2120.53,969.121 "/>
+<path clip-path="url(#clip232)" d="M2120.53 1059.93 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1059.93 L2120.53 1059.93 L2120.53 1059.93  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1059.93 2120.53,1423.18 2181.11,1423.18 2181.11,1059.93 2120.53,1059.93 "/>
+<path clip-path="url(#clip232)" d="M2120.53 1150.74 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1150.74 L2120.53 1150.74 L2120.53 1150.74  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1150.74 2120.53,1423.18 2181.11,1423.18 2181.11,1150.74 2120.53,1150.74 "/>
+<path clip-path="url(#clip232)" d="M2120.53 1241.56 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1241.56 L2120.53 1241.56 L2120.53 1241.56  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1241.56 2120.53,1423.18 2181.11,1423.18 2181.11,1241.56 2120.53,1241.56 "/>
+<path clip-path="url(#clip232)" d="M2120.53 1332.37 L2120.53 1423.18 L2181.11 1423.18 L2181.11 1332.37 L2120.53 1332.37 L2120.53 1332.37  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.53,1332.37 2120.53,1423.18 2181.11,1423.18 2181.11,1332.37 2120.53,1332.37 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="61.0035" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="151.815" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="242.627" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="333.439" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="424.25" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="515.062" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="605.874" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="696.686" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="787.498" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="878.309" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="969.121" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1059.93" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1150.74" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1241.56" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2150.82" cy="1332.37" r="2"/>
+<path clip-path="url(#clip232)" d="M2221.5 89.8981 L2221.5 1423.18 L2282.08 1423.18 L2282.08 89.8981 L2221.5 89.8981 L2221.5 89.8981  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,89.8981 2221.5,1423.18 2282.08,1423.18 2282.08,89.8981 2221.5,89.8981 "/>
+<path clip-path="url(#clip232)" d="M2221.5 160.071 L2221.5 1423.18 L2282.08 1423.18 L2282.08 160.071 L2221.5 160.071 L2221.5 160.071  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,160.071 2221.5,1423.18 2282.08,1423.18 2282.08,160.071 2221.5,160.071 "/>
+<path clip-path="url(#clip232)" d="M2221.5 230.244 L2221.5 1423.18 L2282.08 1423.18 L2282.08 230.244 L2221.5 230.244 L2221.5 230.244  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,230.244 2221.5,1423.18 2282.08,1423.18 2282.08,230.244 2221.5,230.244 "/>
+<path clip-path="url(#clip232)" d="M2221.5 300.416 L2221.5 1423.18 L2282.08 1423.18 L2282.08 300.416 L2221.5 300.416 L2221.5 300.416  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,300.416 2221.5,1423.18 2282.08,1423.18 2282.08,300.416 2221.5,300.416 "/>
+<path clip-path="url(#clip232)" d="M2221.5 370.589 L2221.5 1423.18 L2282.08 1423.18 L2282.08 370.589 L2221.5 370.589 L2221.5 370.589  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,370.589 2221.5,1423.18 2282.08,1423.18 2282.08,370.589 2221.5,370.589 "/>
+<path clip-path="url(#clip232)" d="M2221.5 440.762 L2221.5 1423.18 L2282.08 1423.18 L2282.08 440.762 L2221.5 440.762 L2221.5 440.762  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,440.762 2221.5,1423.18 2282.08,1423.18 2282.08,440.762 2221.5,440.762 "/>
+<path clip-path="url(#clip232)" d="M2221.5 510.934 L2221.5 1423.18 L2282.08 1423.18 L2282.08 510.934 L2221.5 510.934 L2221.5 510.934  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,510.934 2221.5,1423.18 2282.08,1423.18 2282.08,510.934 2221.5,510.934 "/>
+<path clip-path="url(#clip232)" d="M2221.5 581.107 L2221.5 1423.18 L2282.08 1423.18 L2282.08 581.107 L2221.5 581.107 L2221.5 581.107  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,581.107 2221.5,1423.18 2282.08,1423.18 2282.08,581.107 2221.5,581.107 "/>
+<path clip-path="url(#clip232)" d="M2221.5 651.28 L2221.5 1423.18 L2282.08 1423.18 L2282.08 651.28 L2221.5 651.28 L2221.5 651.28  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,651.28 2221.5,1423.18 2282.08,1423.18 2282.08,651.28 2221.5,651.28 "/>
+<path clip-path="url(#clip232)" d="M2221.5 721.453 L2221.5 1423.18 L2282.08 1423.18 L2282.08 721.453 L2221.5 721.453 L2221.5 721.453  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,721.453 2221.5,1423.18 2282.08,1423.18 2282.08,721.453 2221.5,721.453 "/>
+<path clip-path="url(#clip232)" d="M2221.5 791.625 L2221.5 1423.18 L2282.08 1423.18 L2282.08 791.625 L2221.5 791.625 L2221.5 791.625  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,791.625 2221.5,1423.18 2282.08,1423.18 2282.08,791.625 2221.5,791.625 "/>
+<path clip-path="url(#clip232)" d="M2221.5 861.798 L2221.5 1423.18 L2282.08 1423.18 L2282.08 861.798 L2221.5 861.798 L2221.5 861.798  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,861.798 2221.5,1423.18 2282.08,1423.18 2282.08,861.798 2221.5,861.798 "/>
+<path clip-path="url(#clip232)" d="M2221.5 931.971 L2221.5 1423.18 L2282.08 1423.18 L2282.08 931.971 L2221.5 931.971 L2221.5 931.971  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,931.971 2221.5,1423.18 2282.08,1423.18 2282.08,931.971 2221.5,931.971 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1002.14 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1002.14 L2221.5 1002.14 L2221.5 1002.14  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1002.14 2221.5,1423.18 2282.08,1423.18 2282.08,1002.14 2221.5,1002.14 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1072.32 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1072.32 L2221.5 1072.32 L2221.5 1072.32  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1072.32 2221.5,1423.18 2282.08,1423.18 2282.08,1072.32 2221.5,1072.32 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1142.49 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1142.49 L2221.5 1142.49 L2221.5 1142.49  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1142.49 2221.5,1423.18 2282.08,1423.18 2282.08,1142.49 2221.5,1142.49 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1212.66 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1212.66 L2221.5 1212.66 L2221.5 1212.66  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1212.66 2221.5,1423.18 2282.08,1423.18 2282.08,1212.66 2221.5,1212.66 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1282.83 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1282.83 L2221.5 1282.83 L2221.5 1282.83  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1282.83 2221.5,1423.18 2282.08,1423.18 2282.08,1282.83 2221.5,1282.83 "/>
+<path clip-path="url(#clip232)" d="M2221.5 1353.01 L2221.5 1423.18 L2282.08 1423.18 L2282.08 1353.01 L2221.5 1353.01 L2221.5 1353.01  Z" fill="#90caf9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip232)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2221.5,1353.01 2221.5,1423.18 2282.08,1423.18 2282.08,1353.01 2221.5,1353.01 "/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="89.8981" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="160.071" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="230.244" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="300.416" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="370.589" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="440.762" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="510.934" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="581.107" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="651.28" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="721.453" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="791.625" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="861.798" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="931.971" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1002.14" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1072.32" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1142.49" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1212.66" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1282.83" r="2"/>
+<circle clip-path="url(#clip232)" style="fill:#90caf9; stroke:none; fill-opacity:0" cx="2251.79" cy="1353.01" r="2"/>
 </svg>
diff --git a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/index.html b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/index.html
index eed2991db52..f8739ee37d5 100644
--- a/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/index.html
+++ b/previews/PR3562/tutorials/algorithms/cutting_stock_column_generation/index.html
@@ -117,8 +117,8 @@
   Relative gap       : 2.87105e-01
 
 * Work counters
-  Solve time (sec)   : 1.05345e+01
-  Simplex iterations : 20180
+  Solve time (sec)   : 5.00974e+00
+  Simplex iterations : 14227
   Barrier iterations : -1
   Node count         : 0
 </code></pre><p>However, there is a formulation that solves much faster, and that is to use a column generation scheme.</p><h2 id="Column-generation-theory"><a class="docs-heading-anchor" href="#Column-generation-theory">Column generation theory</a><a id="Column-generation-theory-1"></a><a class="docs-heading-anchor-permalink" href="#Column-generation-theory" title="Permalink"></a></h2><p>The key insight for column generation is to recognize that feasible columns in the <span>$x$</span> matrix of variables encode <em>cutting patterns</em>.</p><p>For example, if we look only at the roll <span>$j=1$</span>, then a feasible solution is:</p><ul><li><span>$x_{1,1} = 1$</span> (1 unit of piece #1)</li><li><span>$x_{13,1} = 1$</span> (1 unit of piece #13)</li><li>All other <span>$x_{i,1} = 0$</span></li></ul><p>Another solution is</p><ul><li><span>$x_{20,1} = 19$</span> (19 unit of piece #20)</li><li>All other <span>$x_{i,1} = 0$</span></li></ul><p>Cutting patterns like <span>$x_{1,1} = 1$</span> and <span>$x_{2,1} = 1$</span> are infeasible because the combined length is greater than <span>$W$</span>.</p><p>Since there are a finite number of ways that we could cut a roll into a valid cutting pattern, we could create a set of all possible cutting patterns <span>$p = 1,\ldots,P$</span>, with data <span>$a_{i,p}$</span> indicating how many units of piece <span>$i$</span> we cut in pattern <span>$p$</span>. Then, we can formulate our mixed-integer linear program as:</p><p class="math-container">\[\begin{align}
@@ -188,7 +188,7 @@
     return plot
 end
 
-plot_patterns(data, patterns)</code></pre><img src="50ad43f9.svg" alt="Example block output"/><h2 id="The-base-problem"><a class="docs-heading-anchor" href="#The-base-problem">The base problem</a><a id="The-base-problem-1"></a><a class="docs-heading-anchor-permalink" href="#The-base-problem" title="Permalink"></a></h2><p>Using the initial set of patterns, we can create and optimize our base model:</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
+plot_patterns(data, patterns)</code></pre><img src="80b5cf75.svg" alt="Example block output"/><h2 id="The-base-problem"><a class="docs-heading-anchor" href="#The-base-problem">The base problem</a><a id="The-base-problem-1"></a><a class="docs-heading-anchor-permalink" href="#The-base-problem" title="Permalink"></a></h2><p>Using the initial set of patterns, we can create and optimize our base model:</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
 set_silent(model)
 @variable(model, x[1:length(patterns)] &gt;= 0, Int)
 @objective(model, Min, sum(x))
@@ -210,7 +210,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 1.72615e-04
+  Solve time (sec)   : 2.00987e-04
   Simplex iterations : 0
   Barrier iterations : -1
   Node count         : 0
@@ -279,7 +279,7 @@
 <span class="sgr36"><span class="sgr1">[ Info: </span></span>No new patterns, terminating the algorithm.</code></pre><p>We found lots of new patterns. Here&#39;s pattern 21:</p><pre><code class="language-julia hljs">patterns[21]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">20-element SparseArrays.SparseVector{Int64, Int64} with 3 stored entries:
   [9 ]  =  1
   [13]  =  2
-  [17]  =  1</code></pre><p>Let&#39;s have a look at the patterns now:</p><pre><code class="language-julia hljs">plot_patterns(data, patterns)</code></pre><img src="ccf7f2c9.svg" alt="Example block output"/><h2 id="Looking-at-the-solution"><a class="docs-heading-anchor" href="#Looking-at-the-solution">Looking at the solution</a><a id="Looking-at-the-solution-1"></a><a class="docs-heading-anchor-permalink" href="#Looking-at-the-solution" title="Permalink"></a></h2><p>Let&#39;s see how many of each column we need:</p><pre><code class="language-julia hljs">solution = DataFrames.DataFrame([
+  [17]  =  1</code></pre><p>Let&#39;s have a look at the patterns now:</p><pre><code class="language-julia hljs">plot_patterns(data, patterns)</code></pre><img src="74fb9305.svg" alt="Example block output"/><h2 id="Looking-at-the-solution"><a class="docs-heading-anchor" href="#Looking-at-the-solution">Looking at the solution</a><a id="Looking-at-the-solution-1"></a><a class="docs-heading-anchor-permalink" href="#Looking-at-the-solution" title="Permalink"></a></h2><p>Let&#39;s see how many of each column we need:</p><pre><code class="language-julia hljs">solution = DataFrames.DataFrame([
     (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)
 ])
 filter!(row -&gt; row.rolls &gt; 0, solution)</code></pre><div><div style = "float: left;"><span>16×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">pattern</th><th style = "text-align: left;">rolls</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">38.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">44.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">30.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">21</td><td style = "text-align: right;">0.5</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">22</td><td style = "text-align: right;">10.2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">23</td><td style = "text-align: right;">14.65</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">24</td><td style = "text-align: right;">23.1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">25</td><td style = "text-align: right;">11.25</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">26</td><td style = "text-align: right;">21.35</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">28</td><td style = "text-align: right;">4.3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">29</td><td style = "text-align: right;">19.55</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">30</td><td style = "text-align: right;">11.25</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">31</td><td style = "text-align: right;">17.45</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">33</td><td style = "text-align: right;">36.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">34</td><td style = "text-align: right;">11.4</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">35</td><td style = "text-align: right;">41.0</td></tr></tbody></table></div><p>Since we solved a linear program, some of our columns have fractional solutions. We can create a integer feasible solution by rounding up the orders. This requires 341 rolls:</p><pre><code class="language-julia hljs">sum(ceil.(Int, solution.rolls))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">341</code></pre><p>Alternatively, we can re-introduce the integrality constraints and resolve the problem:</p><pre><code class="language-julia hljs">set_integer.(x)
@@ -287,4 +287,4 @@
 solution = DataFrames.DataFrame([
     (pattern = p, rolls = value(x_p)) for (p, x_p) in enumerate(x)
 ])
-filter!(row -&gt; row.rolls &gt; 0, solution)</code></pre><div><div style = "float: left;"><span>16×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">pattern</th><th style = "text-align: left;">rolls</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">38.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">44.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">30.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">21</td><td style = "text-align: right;">1.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">22</td><td style = "text-align: right;">9.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">23</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">24</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">25</td><td style = "text-align: right;">13.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">26</td><td style = "text-align: right;">17.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">28</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">29</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">30</td><td style = "text-align: right;">13.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">31</td><td style = "text-align: right;">18.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">33</td><td style = "text-align: right;">36.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">34</td><td style = "text-align: right;">15.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">35</td><td style = "text-align: right;">41.0</td></tr></tbody></table></div><p>This now requires 334 rolls:</p><pre><code class="language-julia hljs">sum(solution.rolls)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">333.99999999999994</code></pre><p>Note that this may not be the global minimum because we are not adding new columns during the solution of the mixed-integer problem <code>model</code> (an algorithm known as <a href="https://en.wikipedia.org/wiki/Branch_and_price">branch and price</a>). Nevertheless, the column generation algorithm typically finds good integer feasible solutions to an otherwise intractable optimization problem.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Our objective function is to minimize the total number of rolls. What is the total length of waste? How does that compare to the total demand?</li><li>Writing the optimization algorithm is only part of the challenge. Can you develop a better way to communicate the solution to stakeholders?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../benders_decomposition/">« Benders decomposition</a><a class="docs-footer-nextpage" href="../tsp_lazy_constraints/">Traveling Salesperson Problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+filter!(row -&gt; row.rolls &gt; 0, solution)</code></pre><div><div style = "float: left;"><span>16×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">pattern</th><th style = "text-align: left;">rolls</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">38.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">44.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">30.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">21</td><td style = "text-align: right;">1.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">22</td><td style = "text-align: right;">9.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">23</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">24</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">25</td><td style = "text-align: right;">13.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">26</td><td style = "text-align: right;">17.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">28</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">29</td><td style = "text-align: right;">19.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">30</td><td style = "text-align: right;">13.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">31</td><td style = "text-align: right;">18.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">33</td><td style = "text-align: right;">36.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">34</td><td style = "text-align: right;">15.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">35</td><td style = "text-align: right;">41.0</td></tr></tbody></table></div><p>This now requires 334 rolls:</p><pre><code class="language-julia hljs">sum(solution.rolls)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">333.99999999999994</code></pre><p>Note that this may not be the global minimum because we are not adding new columns during the solution of the mixed-integer problem <code>model</code> (an algorithm known as <a href="https://en.wikipedia.org/wiki/Branch_and_price">branch and price</a>). Nevertheless, the column generation algorithm typically finds good integer feasible solutions to an otherwise intractable optimization problem.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Our objective function is to minimize the total number of rolls. What is the total length of waste? How does that compare to the total demand?</li><li>Writing the optimization algorithm is only part of the challenge. Can you develop a better way to communicate the solution to stakeholders?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../benders_decomposition/">« Benders decomposition</a><a class="docs-footer-nextpage" href="../tsp_lazy_constraints/">Traveling Salesperson Problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/algorithms/parallelism/index.html b/previews/PR3562/tutorials/algorithms/parallelism/index.html
index 30254aadd6a..093a49794f2 100644
--- a/previews/PR3562/tutorials/algorithms/parallelism/index.html
+++ b/previews/PR3562/tutorials/algorithms/parallelism/index.html
@@ -126,4 +126,4 @@
   7.0
   8.0
   9.0
- 10.0</code></pre><h2 id="Other-types-of-parallelism"><a class="docs-heading-anchor" href="#Other-types-of-parallelism">Other types of parallelism</a><a id="Other-types-of-parallelism-1"></a><a class="docs-heading-anchor-permalink" href="#Other-types-of-parallelism" title="Permalink"></a></h2><h3 id="GPU"><a class="docs-heading-anchor" href="#GPU">GPU</a><a id="GPU-1"></a><a class="docs-heading-anchor-permalink" href="#GPU" title="Permalink"></a></h3><p>JuMP does not support GPU programming, and few solvers support execution on a GPU.</p><h3 id="Parallelism-within-the-solver"><a class="docs-heading-anchor" href="#Parallelism-within-the-solver">Parallelism within the solver</a><a id="Parallelism-within-the-solver-1"></a><a class="docs-heading-anchor-permalink" href="#Parallelism-within-the-solver" title="Permalink"></a></h3><p>Many solvers use parallelism internally. For example, commercial solvers like <a href="https://github.com/jump-dev/Gurobi.jl">Gurobi</a> and <a href="https://github.com/jump-dev/CPLEX.jl">CPLEX</a> both parallelize the search in branch-and-bound. Solvers supporting internal parallelism will typically support the <a href="../../../moi/reference/models/#MathOptInterface.NumberOfThreads"><code>MOI.NumberOfThreads</code></a> attribute, which you can set using <a href="../../../api/JuMP/#JuMP.set_attribute"><code>set_attribute</code></a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tsp_lazy_constraints/">« Traveling Salesperson Problem</a><a class="docs-footer-nextpage" href="../../applications/power_systems/">Power Systems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 10.0</code></pre><h2 id="Other-types-of-parallelism"><a class="docs-heading-anchor" href="#Other-types-of-parallelism">Other types of parallelism</a><a id="Other-types-of-parallelism-1"></a><a class="docs-heading-anchor-permalink" href="#Other-types-of-parallelism" title="Permalink"></a></h2><h3 id="GPU"><a class="docs-heading-anchor" href="#GPU">GPU</a><a id="GPU-1"></a><a class="docs-heading-anchor-permalink" href="#GPU" title="Permalink"></a></h3><p>JuMP does not support GPU programming, and few solvers support execution on a GPU.</p><h3 id="Parallelism-within-the-solver"><a class="docs-heading-anchor" href="#Parallelism-within-the-solver">Parallelism within the solver</a><a id="Parallelism-within-the-solver-1"></a><a class="docs-heading-anchor-permalink" href="#Parallelism-within-the-solver" title="Permalink"></a></h3><p>Many solvers use parallelism internally. For example, commercial solvers like <a href="https://github.com/jump-dev/Gurobi.jl">Gurobi</a> and <a href="https://github.com/jump-dev/CPLEX.jl">CPLEX</a> both parallelize the search in branch-and-bound. Solvers supporting internal parallelism will typically support the <a href="../../../moi/reference/models/#MathOptInterface.NumberOfThreads"><code>MOI.NumberOfThreads</code></a> attribute, which you can set using <a href="../../../api/JuMP/#JuMP.set_attribute"><code>set_attribute</code></a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tsp_lazy_constraints/">« Traveling Salesperson Problem</a><a class="docs-footer-nextpage" href="../../applications/power_systems/">Power Systems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/21585044.svg b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6096df34.svg
similarity index 75%
rename from previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/21585044.svg
rename to previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6096df34.svg
index 7ab64bc6b11..d9e2c5e0b38 100644
--- a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/21585044.svg
+++ b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6096df34.svg
@@ -1,122 +1,122 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip590">
+  <clipPath id="clip350">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip590)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip350)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip591">
+  <clipPath id="clip351">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip590)" d="M169.121 1486.45 L2352.76 1486.45 L2352.76 47.2441 L169.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip350)" d="M169.121 1486.45 L2352.76 1486.45 L2352.76 47.2441 L169.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip592">
+  <clipPath id="clip352">
     <rect x="169" y="47" width="2185" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="214.497,1486.45 214.497,47.2441 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="733.661,1486.45 733.661,47.2441 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1252.83,1486.45 1252.83,47.2441 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1771.99,1486.45 1771.99,47.2441 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2291.15,1486.45 2291.15,47.2441 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="214.497,1486.45 214.497,1467.55 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="733.661,1486.45 733.661,1467.55 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1252.83,1486.45 1252.83,1467.55 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1771.99,1486.45 1771.99,1467.55 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2291.15,1486.45 2291.15,1467.55 "/>
-<path clip-path="url(#clip590)" d="M214.497 1517.37 Q210.886 1517.37 209.058 1520.93 Q207.252 1524.47 207.252 1531.6 Q207.252 1538.71 209.058 1542.27 Q210.886 1545.82 214.497 1545.82 Q218.132 1545.82 219.937 1542.27 Q221.766 1538.71 221.766 1531.6 Q221.766 1524.47 219.937 1520.93 Q218.132 1517.37 214.497 1517.37 M214.497 1513.66 Q220.308 1513.66 223.363 1518.27 Q226.442 1522.85 226.442 1531.6 Q226.442 1540.33 223.363 1544.94 Q220.308 1549.52 214.497 1549.52 Q208.687 1549.52 205.609 1544.94 Q202.553 1540.33 202.553 1531.6 Q202.553 1522.85 205.609 1518.27 Q208.687 1513.66 214.497 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M712.932 1544.91 L729.252 1544.91 L729.252 1548.85 L707.307 1548.85 L707.307 1544.91 Q709.969 1542.16 714.553 1537.53 Q719.159 1532.88 720.34 1531.53 Q722.585 1529.01 723.465 1527.27 Q724.367 1525.51 724.367 1523.82 Q724.367 1521.07 722.423 1519.33 Q720.502 1517.6 717.4 1517.6 Q715.201 1517.6 712.747 1518.36 Q710.317 1519.13 707.539 1520.68 L707.539 1515.95 Q710.363 1514.82 712.817 1514.24 Q715.27 1513.66 717.307 1513.66 Q722.678 1513.66 725.872 1516.35 Q729.066 1519.03 729.066 1523.52 Q729.066 1525.65 728.256 1527.57 Q727.469 1529.47 725.363 1532.07 Q724.784 1532.74 721.682 1535.95 Q718.58 1539.15 712.932 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M739.113 1514.29 L757.469 1514.29 L757.469 1518.22 L743.395 1518.22 L743.395 1526.7 Q744.414 1526.35 745.432 1526.19 Q746.451 1526 747.469 1526 Q753.256 1526 756.636 1529.17 Q760.015 1532.34 760.015 1537.76 Q760.015 1543.34 756.543 1546.44 Q753.071 1549.52 746.752 1549.52 Q744.576 1549.52 742.307 1549.15 Q740.062 1548.78 737.654 1548.04 L737.654 1543.34 Q739.738 1544.47 741.96 1545.03 Q744.182 1545.58 746.659 1545.58 Q750.664 1545.58 753.001 1543.48 Q755.339 1541.37 755.339 1537.76 Q755.339 1534.15 753.001 1532.04 Q750.664 1529.94 746.659 1529.94 Q744.784 1529.94 742.909 1530.35 Q741.057 1530.77 739.113 1531.65 L739.113 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1227.52 1514.29 L1245.88 1514.29 L1245.88 1518.22 L1231.81 1518.22 L1231.81 1526.7 Q1232.83 1526.35 1233.84 1526.19 Q1234.86 1526 1235.88 1526 Q1241.67 1526 1245.05 1529.17 Q1248.43 1532.34 1248.43 1537.76 Q1248.43 1543.34 1244.95 1546.44 Q1241.48 1549.52 1235.16 1549.52 Q1232.99 1549.52 1230.72 1549.15 Q1228.47 1548.78 1226.07 1548.04 L1226.07 1543.34 Q1228.15 1544.47 1230.37 1545.03 Q1232.59 1545.58 1235.07 1545.58 Q1239.08 1545.58 1241.41 1543.48 Q1243.75 1541.37 1243.75 1537.76 Q1243.75 1534.15 1241.41 1532.04 Q1239.08 1529.94 1235.07 1529.94 Q1233.2 1529.94 1231.32 1530.35 Q1229.47 1530.77 1227.52 1531.65 L1227.52 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1267.64 1517.37 Q1264.03 1517.37 1262.2 1520.93 Q1260.39 1524.47 1260.39 1531.6 Q1260.39 1538.71 1262.2 1542.27 Q1264.03 1545.82 1267.64 1545.82 Q1271.27 1545.82 1273.08 1542.27 Q1274.91 1538.71 1274.91 1531.6 Q1274.91 1524.47 1273.08 1520.93 Q1271.27 1517.37 1267.64 1517.37 M1267.64 1513.66 Q1273.45 1513.66 1276.51 1518.27 Q1279.58 1522.85 1279.58 1531.6 Q1279.58 1540.33 1276.51 1544.94 Q1273.45 1549.52 1267.64 1549.52 Q1261.83 1549.52 1258.75 1544.94 Q1255.7 1540.33 1255.7 1531.6 Q1255.7 1522.85 1258.75 1518.27 Q1261.83 1513.66 1267.64 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1745.84 1514.29 L1768.07 1514.29 L1768.07 1516.28 L1755.52 1548.85 L1750.63 1548.85 L1762.44 1518.22 L1745.84 1518.22 L1745.84 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1777.23 1514.29 L1795.59 1514.29 L1795.59 1518.22 L1781.51 1518.22 L1781.51 1526.7 Q1782.53 1526.35 1783.55 1526.19 Q1784.57 1526 1785.59 1526 Q1791.38 1526 1794.76 1529.17 Q1798.13 1532.34 1798.13 1537.76 Q1798.13 1543.34 1794.66 1546.44 Q1791.19 1549.52 1784.87 1549.52 Q1782.7 1549.52 1780.43 1549.15 Q1778.18 1548.78 1775.77 1548.04 L1775.77 1543.34 Q1777.86 1544.47 1780.08 1545.03 Q1782.3 1545.58 1784.78 1545.58 Q1788.78 1545.58 1791.12 1543.48 Q1793.46 1541.37 1793.46 1537.76 Q1793.46 1534.15 1791.12 1532.04 Q1788.78 1529.94 1784.78 1529.94 Q1782.9 1529.94 1781.03 1530.35 Q1779.18 1530.77 1777.23 1531.65 L1777.23 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2250.76 1544.91 L2258.4 1544.91 L2258.4 1518.55 L2250.09 1520.21 L2250.09 1515.95 L2258.35 1514.29 L2263.03 1514.29 L2263.03 1544.91 L2270.67 1544.91 L2270.67 1548.85 L2250.76 1548.85 L2250.76 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2290.11 1517.37 Q2286.5 1517.37 2284.67 1520.93 Q2282.87 1524.47 2282.87 1531.6 Q2282.87 1538.71 2284.67 1542.27 Q2286.5 1545.82 2290.11 1545.82 Q2293.75 1545.82 2295.55 1542.27 Q2297.38 1538.71 2297.38 1531.6 Q2297.38 1524.47 2295.55 1520.93 Q2293.75 1517.37 2290.11 1517.37 M2290.11 1513.66 Q2295.92 1513.66 2298.98 1518.27 Q2302.06 1522.85 2302.06 1531.6 Q2302.06 1540.33 2298.98 1544.94 Q2295.92 1549.52 2290.11 1549.52 Q2284.3 1549.52 2281.22 1544.94 Q2278.17 1540.33 2278.17 1531.6 Q2278.17 1522.85 2281.22 1518.27 Q2284.3 1513.66 2290.11 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2320.27 1517.37 Q2316.66 1517.37 2314.83 1520.93 Q2313.03 1524.47 2313.03 1531.6 Q2313.03 1538.71 2314.83 1542.27 Q2316.66 1545.82 2320.27 1545.82 Q2323.91 1545.82 2325.71 1542.27 Q2327.54 1538.71 2327.54 1531.6 Q2327.54 1524.47 2325.71 1520.93 Q2323.91 1517.37 2320.27 1517.37 M2320.27 1513.66 Q2326.08 1513.66 2329.14 1518.27 Q2332.22 1522.85 2332.22 1531.6 Q2332.22 1540.33 2329.14 1544.94 Q2326.08 1549.52 2320.27 1549.52 Q2314.46 1549.52 2311.38 1544.94 Q2308.33 1540.33 2308.33 1531.6 Q2308.33 1522.85 2311.38 1518.27 Q2314.46 1513.66 2320.27 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1464.39 2352.76,1464.39 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1119.64 2352.76,1119.64 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,774.897 2352.76,774.897 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,430.151 2352.76,430.151 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,85.4059 2352.76,85.4059 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 169.121,47.2441 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1464.39 188.019,1464.39 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1119.64 188.019,1119.64 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,774.897 188.019,774.897 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,430.151 188.019,430.151 "/>
-<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,85.4059 188.019,85.4059 "/>
-<path clip-path="url(#clip590)" d="M121.177 1450.19 Q117.566 1450.19 115.737 1453.75 Q113.932 1457.29 113.932 1464.42 Q113.932 1471.53 115.737 1475.09 Q117.566 1478.64 121.177 1478.64 Q124.811 1478.64 126.617 1475.09 Q128.445 1471.53 128.445 1464.42 Q128.445 1457.29 126.617 1453.75 Q124.811 1450.19 121.177 1450.19 M121.177 1446.48 Q126.987 1446.48 130.043 1451.09 Q133.121 1455.67 133.121 1464.42 Q133.121 1473.15 130.043 1477.76 Q126.987 1482.34 121.177 1482.34 Q115.367 1482.34 112.288 1477.76 Q109.233 1473.15 109.233 1464.42 Q109.233 1455.67 112.288 1451.09 Q115.367 1446.48 121.177 1446.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M86.0382 1132.99 L102.358 1132.99 L102.358 1136.92 L80.4133 1136.92 L80.4133 1132.99 Q83.0753 1130.23 87.6586 1125.6 Q92.2651 1120.95 93.4456 1119.61 Q95.691 1117.08 96.5706 1115.35 Q97.4734 1113.59 97.4734 1111.9 Q97.4734 1109.14 95.5289 1107.41 Q93.6076 1105.67 90.5058 1105.67 Q88.3067 1105.67 85.8531 1106.44 Q83.4225 1107.2 80.6447 1108.75 L80.6447 1104.03 Q83.4688 1102.89 85.9225 1102.32 Q88.3762 1101.74 90.4132 1101.74 Q95.7836 1101.74 98.978 1104.42 Q102.172 1107.11 102.172 1111.6 Q102.172 1113.73 101.362 1115.65 Q100.575 1117.55 98.4687 1120.14 Q97.89 1120.81 94.7882 1124.03 Q91.6864 1127.22 86.0382 1132.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M112.219 1102.36 L130.575 1102.36 L130.575 1106.3 L116.501 1106.3 L116.501 1114.77 Q117.52 1114.42 118.538 1114.26 Q119.557 1114.08 120.575 1114.08 Q126.362 1114.08 129.742 1117.25 Q133.121 1120.42 133.121 1125.83 Q133.121 1131.41 129.649 1134.51 Q126.177 1137.59 119.857 1137.59 Q117.682 1137.59 115.413 1137.22 Q113.168 1136.85 110.76 1136.11 L110.76 1131.41 Q112.844 1132.55 115.066 1133.1 Q117.288 1133.66 119.765 1133.66 Q123.77 1133.66 126.107 1131.55 Q128.445 1129.45 128.445 1125.83 Q128.445 1122.22 126.107 1120.12 Q123.77 1118.01 119.765 1118.01 Q117.89 1118.01 116.015 1118.43 Q114.163 1118.84 112.219 1119.72 L112.219 1102.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M81.0614 757.617 L99.4178 757.617 L99.4178 761.552 L85.3438 761.552 L85.3438 770.024 Q86.3623 769.677 87.3808 769.515 Q88.3993 769.33 89.4178 769.33 Q95.2049 769.33 98.5845 772.501 Q101.964 775.672 101.964 781.089 Q101.964 786.668 98.4919 789.769 Q95.0197 792.848 88.7003 792.848 Q86.5243 792.848 84.2558 792.478 Q82.0105 792.107 79.6031 791.367 L79.6031 786.668 Q81.6864 787.802 83.9086 788.357 Q86.1308 788.913 88.6077 788.913 Q92.6123 788.913 94.9502 786.806 Q97.2882 784.7 97.2882 781.089 Q97.2882 777.478 94.9502 775.371 Q92.6123 773.265 88.6077 773.265 Q86.7327 773.265 84.8577 773.681 Q83.0058 774.098 81.0614 774.978 L81.0614 757.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M121.177 760.695 Q117.566 760.695 115.737 764.26 Q113.932 767.802 113.932 774.931 Q113.932 782.038 115.737 785.603 Q117.566 789.144 121.177 789.144 Q124.811 789.144 126.617 785.603 Q128.445 782.038 128.445 774.931 Q128.445 767.802 126.617 764.26 Q124.811 760.695 121.177 760.695 M121.177 756.992 Q126.987 756.992 130.043 761.598 Q133.121 766.182 133.121 774.931 Q133.121 783.658 130.043 788.265 Q126.987 792.848 121.177 792.848 Q115.367 792.848 112.288 788.265 Q109.233 783.658 109.233 774.931 Q109.233 766.182 112.288 761.598 Q115.367 756.992 121.177 756.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M80.8299 412.871 L103.052 412.871 L103.052 414.862 L90.5058 447.431 L85.6216 447.431 L97.4271 416.807 L80.8299 416.807 L80.8299 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M112.219 412.871 L130.575 412.871 L130.575 416.807 L116.501 416.807 L116.501 425.279 Q117.52 424.931 118.538 424.769 Q119.557 424.584 120.575 424.584 Q126.362 424.584 129.742 427.756 Q133.121 430.927 133.121 436.343 Q133.121 441.922 129.649 445.024 Q126.177 448.103 119.857 448.103 Q117.682 448.103 115.413 447.732 Q113.168 447.362 110.76 446.621 L110.76 441.922 Q112.844 443.056 115.066 443.612 Q117.288 444.167 119.765 444.167 Q123.77 444.167 126.107 442.061 Q128.445 439.955 128.445 436.343 Q128.445 432.732 126.107 430.626 Q123.77 428.519 119.765 428.519 Q117.89 428.519 116.015 428.936 Q114.163 429.353 112.219 430.232 L112.219 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M51.6634 98.7508 L59.3023 98.7508 L59.3023 72.3852 L50.9921 74.0518 L50.9921 69.7926 L59.256 68.1259 L63.9319 68.1259 L63.9319 98.7508 L71.5707 98.7508 L71.5707 102.686 L51.6634 102.686 L51.6634 98.7508 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M91.0151 71.2046 Q87.404 71.2046 85.5753 74.7694 Q83.7697 78.3111 83.7697 85.4407 Q83.7697 92.5471 85.5753 96.1119 Q87.404 99.6535 91.0151 99.6535 Q94.6493 99.6535 96.4548 96.1119 Q98.2835 92.5471 98.2835 85.4407 Q98.2835 78.3111 96.4548 74.7694 Q94.6493 71.2046 91.0151 71.2046 M91.0151 67.5009 Q96.8252 67.5009 99.8808 72.1074 Q102.959 76.6907 102.959 85.4407 Q102.959 94.1675 99.8808 98.7739 Q96.8252 103.357 91.0151 103.357 Q85.2049 103.357 82.1262 98.7739 Q79.0707 94.1675 79.0707 85.4407 Q79.0707 76.6907 82.1262 72.1074 Q85.2049 67.5009 91.0151 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M121.177 71.2046 Q117.566 71.2046 115.737 74.7694 Q113.932 78.3111 113.932 85.4407 Q113.932 92.5471 115.737 96.1119 Q117.566 99.6535 121.177 99.6535 Q124.811 99.6535 126.617 96.1119 Q128.445 92.5471 128.445 85.4407 Q128.445 78.3111 126.617 74.7694 Q124.811 71.2046 121.177 71.2046 M121.177 67.5009 Q126.987 67.5009 130.043 72.1074 Q133.121 76.6907 133.121 85.4407 Q133.121 94.1675 130.043 98.7739 Q126.987 103.357 121.177 103.357 Q115.367 103.357 112.288 98.7739 Q109.233 94.1675 109.233 85.4407 Q109.233 76.6907 112.288 72.1074 Q115.367 67.5009 121.177 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 704.657,952.826 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 704.657,952.826 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 934.094,231.349 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 934.094,231.349 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 863.882,570.733 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 863.882,570.733 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 230.922,889.298 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 230.922,889.298 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 1229.18,1265.03 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1229.18,1265.03 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 652.606,606.105 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 652.606,606.105 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2191.3,261.454 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2191.3,261.454 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 2290.95,740.457 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 2290.95,740.457 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 737.113,1131.24 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 737.113,1131.24 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 2263.46,244.214 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 2263.46,244.214 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1368.6,856.398 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1368.6,856.398 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1122.22,906.351 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 1122.22,906.351 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 1096.49,448.372 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1096.49,448.372 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1820.22,149.11 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 1820.22,149.11 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 798.433,1333.13 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 798.433,1333.13 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 649.499,747.765 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 649.499,747.765 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 736.525,1445.72 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 736.525,1445.72 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 256.809,1046.01 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 256.809,1046.01 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 811.954,495.574 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 811.954,495.574 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 1999.41,641.774 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 1999.41,641.774 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 374.298,583.308 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 374.298,583.308 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 1544.38,261.436 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 1544.38,261.436 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 2028.55,707.83 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2028.55,707.83 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 793.017,1101.73 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 793.017,1101.73 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1774.71,738.432 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 1774.71,738.432 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1553.7,823.134 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1553.7,823.134 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 376.116,1084.45 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 376.116,1084.45 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 1975.89,827.543 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1975.89,827.543 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 392.332,151.788 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 392.332,151.788 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 1363.31,1066.22 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 1363.31,1066.22 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1176.72,552.561 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1176.72,552.561 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 600.384,1196.08 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 600.384,1196.08 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 447.043,922.181 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 447.043,922.181 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2241.96,98.1744 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 2241.96,98.1744 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 321.683,705.487 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 321.683,705.487 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1331.8,663.5 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1331.8,663.5 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 1160.81,404.835 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1160.81,404.835 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 794.705,747.973 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 794.705,747.973 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 584.653,754.401 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 584.653,754.401 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 1354.55,87.9763 "/>
-<polyline clip-path="url(#clip592)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1354.55,87.9763 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="214.497,1486.45 214.497,47.2441 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="733.661,1486.45 733.661,47.2441 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1252.83,1486.45 1252.83,47.2441 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1771.99,1486.45 1771.99,47.2441 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2291.15,1486.45 2291.15,47.2441 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="214.497,1486.45 214.497,1467.55 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="733.661,1486.45 733.661,1467.55 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1252.83,1486.45 1252.83,1467.55 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1771.99,1486.45 1771.99,1467.55 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2291.15,1486.45 2291.15,1467.55 "/>
+<path clip-path="url(#clip350)" d="M214.497 1517.37 Q210.886 1517.37 209.058 1520.93 Q207.252 1524.47 207.252 1531.6 Q207.252 1538.71 209.058 1542.27 Q210.886 1545.82 214.497 1545.82 Q218.132 1545.82 219.937 1542.27 Q221.766 1538.71 221.766 1531.6 Q221.766 1524.47 219.937 1520.93 Q218.132 1517.37 214.497 1517.37 M214.497 1513.66 Q220.308 1513.66 223.363 1518.27 Q226.442 1522.85 226.442 1531.6 Q226.442 1540.33 223.363 1544.94 Q220.308 1549.52 214.497 1549.52 Q208.687 1549.52 205.609 1544.94 Q202.553 1540.33 202.553 1531.6 Q202.553 1522.85 205.609 1518.27 Q208.687 1513.66 214.497 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M712.932 1544.91 L729.252 1544.91 L729.252 1548.85 L707.307 1548.85 L707.307 1544.91 Q709.969 1542.16 714.553 1537.53 Q719.159 1532.88 720.34 1531.53 Q722.585 1529.01 723.465 1527.27 Q724.367 1525.51 724.367 1523.82 Q724.367 1521.07 722.423 1519.33 Q720.502 1517.6 717.4 1517.6 Q715.201 1517.6 712.747 1518.36 Q710.317 1519.13 707.539 1520.68 L707.539 1515.95 Q710.363 1514.82 712.817 1514.24 Q715.27 1513.66 717.307 1513.66 Q722.678 1513.66 725.872 1516.35 Q729.066 1519.03 729.066 1523.52 Q729.066 1525.65 728.256 1527.57 Q727.469 1529.47 725.363 1532.07 Q724.784 1532.74 721.682 1535.95 Q718.58 1539.15 712.932 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M739.113 1514.29 L757.469 1514.29 L757.469 1518.22 L743.395 1518.22 L743.395 1526.7 Q744.414 1526.35 745.432 1526.19 Q746.451 1526 747.469 1526 Q753.256 1526 756.636 1529.17 Q760.015 1532.34 760.015 1537.76 Q760.015 1543.34 756.543 1546.44 Q753.071 1549.52 746.752 1549.52 Q744.576 1549.52 742.307 1549.15 Q740.062 1548.78 737.654 1548.04 L737.654 1543.34 Q739.738 1544.47 741.96 1545.03 Q744.182 1545.58 746.659 1545.58 Q750.664 1545.58 753.001 1543.48 Q755.339 1541.37 755.339 1537.76 Q755.339 1534.15 753.001 1532.04 Q750.664 1529.94 746.659 1529.94 Q744.784 1529.94 742.909 1530.35 Q741.057 1530.77 739.113 1531.65 L739.113 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M1227.52 1514.29 L1245.88 1514.29 L1245.88 1518.22 L1231.81 1518.22 L1231.81 1526.7 Q1232.83 1526.35 1233.84 1526.19 Q1234.86 1526 1235.88 1526 Q1241.67 1526 1245.05 1529.17 Q1248.43 1532.34 1248.43 1537.76 Q1248.43 1543.34 1244.95 1546.44 Q1241.48 1549.52 1235.16 1549.52 Q1232.99 1549.52 1230.72 1549.15 Q1228.47 1548.78 1226.07 1548.04 L1226.07 1543.34 Q1228.15 1544.47 1230.37 1545.03 Q1232.59 1545.58 1235.07 1545.58 Q1239.08 1545.58 1241.41 1543.48 Q1243.75 1541.37 1243.75 1537.76 Q1243.75 1534.15 1241.41 1532.04 Q1239.08 1529.94 1235.07 1529.94 Q1233.2 1529.94 1231.32 1530.35 Q1229.47 1530.77 1227.52 1531.65 L1227.52 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M1267.64 1517.37 Q1264.03 1517.37 1262.2 1520.93 Q1260.39 1524.47 1260.39 1531.6 Q1260.39 1538.71 1262.2 1542.27 Q1264.03 1545.82 1267.64 1545.82 Q1271.27 1545.82 1273.08 1542.27 Q1274.91 1538.71 1274.91 1531.6 Q1274.91 1524.47 1273.08 1520.93 Q1271.27 1517.37 1267.64 1517.37 M1267.64 1513.66 Q1273.45 1513.66 1276.51 1518.27 Q1279.58 1522.85 1279.58 1531.6 Q1279.58 1540.33 1276.51 1544.94 Q1273.45 1549.52 1267.64 1549.52 Q1261.83 1549.52 1258.75 1544.94 Q1255.7 1540.33 1255.7 1531.6 Q1255.7 1522.85 1258.75 1518.27 Q1261.83 1513.66 1267.64 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M1745.84 1514.29 L1768.07 1514.29 L1768.07 1516.28 L1755.52 1548.85 L1750.63 1548.85 L1762.44 1518.22 L1745.84 1518.22 L1745.84 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M1777.23 1514.29 L1795.59 1514.29 L1795.59 1518.22 L1781.51 1518.22 L1781.51 1526.7 Q1782.53 1526.35 1783.55 1526.19 Q1784.57 1526 1785.59 1526 Q1791.38 1526 1794.76 1529.17 Q1798.13 1532.34 1798.13 1537.76 Q1798.13 1543.34 1794.66 1546.44 Q1791.19 1549.52 1784.87 1549.52 Q1782.7 1549.52 1780.43 1549.15 Q1778.18 1548.78 1775.77 1548.04 L1775.77 1543.34 Q1777.86 1544.47 1780.08 1545.03 Q1782.3 1545.58 1784.78 1545.58 Q1788.78 1545.58 1791.12 1543.48 Q1793.46 1541.37 1793.46 1537.76 Q1793.46 1534.15 1791.12 1532.04 Q1788.78 1529.94 1784.78 1529.94 Q1782.9 1529.94 1781.03 1530.35 Q1779.18 1530.77 1777.23 1531.65 L1777.23 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M2250.76 1544.91 L2258.4 1544.91 L2258.4 1518.55 L2250.09 1520.21 L2250.09 1515.95 L2258.35 1514.29 L2263.03 1514.29 L2263.03 1544.91 L2270.67 1544.91 L2270.67 1548.85 L2250.76 1548.85 L2250.76 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M2290.11 1517.37 Q2286.5 1517.37 2284.67 1520.93 Q2282.87 1524.47 2282.87 1531.6 Q2282.87 1538.71 2284.67 1542.27 Q2286.5 1545.82 2290.11 1545.82 Q2293.75 1545.82 2295.55 1542.27 Q2297.38 1538.71 2297.38 1531.6 Q2297.38 1524.47 2295.55 1520.93 Q2293.75 1517.37 2290.11 1517.37 M2290.11 1513.66 Q2295.92 1513.66 2298.98 1518.27 Q2302.06 1522.85 2302.06 1531.6 Q2302.06 1540.33 2298.98 1544.94 Q2295.92 1549.52 2290.11 1549.52 Q2284.3 1549.52 2281.22 1544.94 Q2278.17 1540.33 2278.17 1531.6 Q2278.17 1522.85 2281.22 1518.27 Q2284.3 1513.66 2290.11 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M2320.27 1517.37 Q2316.66 1517.37 2314.83 1520.93 Q2313.03 1524.47 2313.03 1531.6 Q2313.03 1538.71 2314.83 1542.27 Q2316.66 1545.82 2320.27 1545.82 Q2323.91 1545.82 2325.71 1542.27 Q2327.54 1538.71 2327.54 1531.6 Q2327.54 1524.47 2325.71 1520.93 Q2323.91 1517.37 2320.27 1517.37 M2320.27 1513.66 Q2326.08 1513.66 2329.14 1518.27 Q2332.22 1522.85 2332.22 1531.6 Q2332.22 1540.33 2329.14 1544.94 Q2326.08 1549.52 2320.27 1549.52 Q2314.46 1549.52 2311.38 1544.94 Q2308.33 1540.33 2308.33 1531.6 Q2308.33 1522.85 2311.38 1518.27 Q2314.46 1513.66 2320.27 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1464.39 2352.76,1464.39 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1119.64 2352.76,1119.64 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,774.897 2352.76,774.897 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,430.151 2352.76,430.151 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,85.4059 2352.76,85.4059 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 169.121,47.2441 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1464.39 188.019,1464.39 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1119.64 188.019,1119.64 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,774.897 188.019,774.897 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,430.151 188.019,430.151 "/>
+<polyline clip-path="url(#clip350)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,85.4059 188.019,85.4059 "/>
+<path clip-path="url(#clip350)" d="M121.177 1450.19 Q117.566 1450.19 115.737 1453.75 Q113.932 1457.29 113.932 1464.42 Q113.932 1471.53 115.737 1475.09 Q117.566 1478.64 121.177 1478.64 Q124.811 1478.64 126.617 1475.09 Q128.445 1471.53 128.445 1464.42 Q128.445 1457.29 126.617 1453.75 Q124.811 1450.19 121.177 1450.19 M121.177 1446.48 Q126.987 1446.48 130.043 1451.09 Q133.121 1455.67 133.121 1464.42 Q133.121 1473.15 130.043 1477.76 Q126.987 1482.34 121.177 1482.34 Q115.367 1482.34 112.288 1477.76 Q109.233 1473.15 109.233 1464.42 Q109.233 1455.67 112.288 1451.09 Q115.367 1446.48 121.177 1446.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M86.0382 1132.99 L102.358 1132.99 L102.358 1136.92 L80.4133 1136.92 L80.4133 1132.99 Q83.0753 1130.23 87.6586 1125.6 Q92.2651 1120.95 93.4456 1119.61 Q95.691 1117.08 96.5706 1115.35 Q97.4734 1113.59 97.4734 1111.9 Q97.4734 1109.14 95.5289 1107.41 Q93.6076 1105.67 90.5058 1105.67 Q88.3067 1105.67 85.8531 1106.44 Q83.4225 1107.2 80.6447 1108.75 L80.6447 1104.03 Q83.4688 1102.89 85.9225 1102.32 Q88.3762 1101.74 90.4132 1101.74 Q95.7836 1101.74 98.978 1104.42 Q102.172 1107.11 102.172 1111.6 Q102.172 1113.73 101.362 1115.65 Q100.575 1117.55 98.4687 1120.14 Q97.89 1120.81 94.7882 1124.03 Q91.6864 1127.22 86.0382 1132.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M112.219 1102.36 L130.575 1102.36 L130.575 1106.3 L116.501 1106.3 L116.501 1114.77 Q117.52 1114.42 118.538 1114.26 Q119.557 1114.08 120.575 1114.08 Q126.362 1114.08 129.742 1117.25 Q133.121 1120.42 133.121 1125.83 Q133.121 1131.41 129.649 1134.51 Q126.177 1137.59 119.857 1137.59 Q117.682 1137.59 115.413 1137.22 Q113.168 1136.85 110.76 1136.11 L110.76 1131.41 Q112.844 1132.55 115.066 1133.1 Q117.288 1133.66 119.765 1133.66 Q123.77 1133.66 126.107 1131.55 Q128.445 1129.45 128.445 1125.83 Q128.445 1122.22 126.107 1120.12 Q123.77 1118.01 119.765 1118.01 Q117.89 1118.01 116.015 1118.43 Q114.163 1118.84 112.219 1119.72 L112.219 1102.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M81.0614 757.617 L99.4178 757.617 L99.4178 761.552 L85.3438 761.552 L85.3438 770.024 Q86.3623 769.677 87.3808 769.515 Q88.3993 769.33 89.4178 769.33 Q95.2049 769.33 98.5845 772.501 Q101.964 775.672 101.964 781.089 Q101.964 786.668 98.4919 789.769 Q95.0197 792.848 88.7003 792.848 Q86.5243 792.848 84.2558 792.478 Q82.0105 792.107 79.6031 791.367 L79.6031 786.668 Q81.6864 787.802 83.9086 788.357 Q86.1308 788.913 88.6077 788.913 Q92.6123 788.913 94.9502 786.806 Q97.2882 784.7 97.2882 781.089 Q97.2882 777.478 94.9502 775.371 Q92.6123 773.265 88.6077 773.265 Q86.7327 773.265 84.8577 773.681 Q83.0058 774.098 81.0614 774.978 L81.0614 757.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M121.177 760.695 Q117.566 760.695 115.737 764.26 Q113.932 767.802 113.932 774.931 Q113.932 782.038 115.737 785.603 Q117.566 789.144 121.177 789.144 Q124.811 789.144 126.617 785.603 Q128.445 782.038 128.445 774.931 Q128.445 767.802 126.617 764.26 Q124.811 760.695 121.177 760.695 M121.177 756.992 Q126.987 756.992 130.043 761.598 Q133.121 766.182 133.121 774.931 Q133.121 783.658 130.043 788.265 Q126.987 792.848 121.177 792.848 Q115.367 792.848 112.288 788.265 Q109.233 783.658 109.233 774.931 Q109.233 766.182 112.288 761.598 Q115.367 756.992 121.177 756.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M80.8299 412.871 L103.052 412.871 L103.052 414.862 L90.5058 447.431 L85.6216 447.431 L97.4271 416.807 L80.8299 416.807 L80.8299 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M112.219 412.871 L130.575 412.871 L130.575 416.807 L116.501 416.807 L116.501 425.279 Q117.52 424.931 118.538 424.769 Q119.557 424.584 120.575 424.584 Q126.362 424.584 129.742 427.756 Q133.121 430.927 133.121 436.343 Q133.121 441.922 129.649 445.024 Q126.177 448.103 119.857 448.103 Q117.682 448.103 115.413 447.732 Q113.168 447.362 110.76 446.621 L110.76 441.922 Q112.844 443.056 115.066 443.612 Q117.288 444.167 119.765 444.167 Q123.77 444.167 126.107 442.061 Q128.445 439.955 128.445 436.343 Q128.445 432.732 126.107 430.626 Q123.77 428.519 119.765 428.519 Q117.89 428.519 116.015 428.936 Q114.163 429.353 112.219 430.232 L112.219 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M51.6634 98.7508 L59.3023 98.7508 L59.3023 72.3852 L50.9921 74.0518 L50.9921 69.7926 L59.256 68.1259 L63.9319 68.1259 L63.9319 98.7508 L71.5707 98.7508 L71.5707 102.686 L51.6634 102.686 L51.6634 98.7508 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M91.0151 71.2046 Q87.404 71.2046 85.5753 74.7694 Q83.7697 78.3111 83.7697 85.4407 Q83.7697 92.5471 85.5753 96.1119 Q87.404 99.6535 91.0151 99.6535 Q94.6493 99.6535 96.4548 96.1119 Q98.2835 92.5471 98.2835 85.4407 Q98.2835 78.3111 96.4548 74.7694 Q94.6493 71.2046 91.0151 71.2046 M91.0151 67.5009 Q96.8252 67.5009 99.8808 72.1074 Q102.959 76.6907 102.959 85.4407 Q102.959 94.1675 99.8808 98.7739 Q96.8252 103.357 91.0151 103.357 Q85.2049 103.357 82.1262 98.7739 Q79.0707 94.1675 79.0707 85.4407 Q79.0707 76.6907 82.1262 72.1074 Q85.2049 67.5009 91.0151 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip350)" d="M121.177 71.2046 Q117.566 71.2046 115.737 74.7694 Q113.932 78.3111 113.932 85.4407 Q113.932 92.5471 115.737 96.1119 Q117.566 99.6535 121.177 99.6535 Q124.811 99.6535 126.617 96.1119 Q128.445 92.5471 128.445 85.4407 Q128.445 78.3111 126.617 74.7694 Q124.811 71.2046 121.177 71.2046 M121.177 67.5009 Q126.987 67.5009 130.043 72.1074 Q133.121 76.6907 133.121 85.4407 Q133.121 94.1675 130.043 98.7739 Q126.987 103.357 121.177 103.357 Q115.367 103.357 112.288 98.7739 Q109.233 94.1675 109.233 85.4407 Q109.233 76.6907 112.288 72.1074 Q115.367 67.5009 121.177 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip352)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 704.657,952.826 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 704.657,952.826 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 934.094,231.349 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 934.094,231.349 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 863.882,570.733 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 863.882,570.733 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 230.922,889.298 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 230.922,889.298 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 1229.18,1265.03 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1229.18,1265.03 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 652.606,606.105 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 652.606,606.105 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2191.3,261.454 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2191.3,261.454 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 2290.95,740.457 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 2290.95,740.457 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 737.113,1131.24 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 737.113,1131.24 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 2263.46,244.214 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 2263.46,244.214 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1368.6,856.398 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1368.6,856.398 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1122.22,906.351 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 1122.22,906.351 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 1096.49,448.372 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1096.49,448.372 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1820.22,149.11 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 1820.22,149.11 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 798.433,1333.13 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 798.433,1333.13 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 649.499,747.765 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 649.499,747.765 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 736.525,1445.72 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 736.525,1445.72 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 256.809,1046.01 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 256.809,1046.01 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 811.954,495.574 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 811.954,495.574 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 1999.41,641.774 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 1999.41,641.774 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 374.298,583.308 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 374.298,583.308 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 1544.38,261.436 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 1544.38,261.436 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 2028.55,707.83 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2028.55,707.83 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 793.017,1101.73 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 793.017,1101.73 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1774.71,738.432 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 1774.71,738.432 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1553.7,823.134 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1553.7,823.134 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 376.116,1084.45 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 376.116,1084.45 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 1975.89,827.543 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1975.89,827.543 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 392.332,151.788 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 392.332,151.788 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 1363.31,1066.22 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 1363.31,1066.22 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1176.72,552.561 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1176.72,552.561 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 600.384,1196.08 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 600.384,1196.08 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 447.043,922.181 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 447.043,922.181 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2241.96,98.1744 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 2241.96,98.1744 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 321.683,705.487 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 321.683,705.487 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1331.8,663.5 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1331.8,663.5 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 1160.81,404.835 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1160.81,404.835 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 794.705,747.973 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 794.705,747.973 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 584.653,754.401 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 584.653,754.401 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 1354.55,87.9763 "/>
+<polyline clip-path="url(#clip352)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1354.55,87.9763 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/88a7f5c2.svg b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6e144cc3.svg
similarity index 75%
rename from previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/88a7f5c2.svg
rename to previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6e144cc3.svg
index 5cd68f3f490..9acc214eaa2 100644
--- a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/88a7f5c2.svg
+++ b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/6e144cc3.svg
@@ -1,122 +1,122 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip530">
+  <clipPath id="clip410">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip530)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip410)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip531">
+  <clipPath id="clip411">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip530)" d="M169.121 1486.45 L2352.76 1486.45 L2352.76 47.2441 L169.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip410)" d="M169.121 1486.45 L2352.76 1486.45 L2352.76 47.2441 L169.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip532">
+  <clipPath id="clip412">
     <rect x="169" y="47" width="2185" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="214.497,1486.45 214.497,47.2441 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="733.661,1486.45 733.661,47.2441 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1252.83,1486.45 1252.83,47.2441 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1771.99,1486.45 1771.99,47.2441 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2291.15,1486.45 2291.15,47.2441 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="214.497,1486.45 214.497,1467.55 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="733.661,1486.45 733.661,1467.55 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1252.83,1486.45 1252.83,1467.55 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1771.99,1486.45 1771.99,1467.55 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2291.15,1486.45 2291.15,1467.55 "/>
-<path clip-path="url(#clip530)" d="M214.497 1517.37 Q210.886 1517.37 209.058 1520.93 Q207.252 1524.47 207.252 1531.6 Q207.252 1538.71 209.058 1542.27 Q210.886 1545.82 214.497 1545.82 Q218.132 1545.82 219.937 1542.27 Q221.766 1538.71 221.766 1531.6 Q221.766 1524.47 219.937 1520.93 Q218.132 1517.37 214.497 1517.37 M214.497 1513.66 Q220.308 1513.66 223.363 1518.27 Q226.442 1522.85 226.442 1531.6 Q226.442 1540.33 223.363 1544.94 Q220.308 1549.52 214.497 1549.52 Q208.687 1549.52 205.609 1544.94 Q202.553 1540.33 202.553 1531.6 Q202.553 1522.85 205.609 1518.27 Q208.687 1513.66 214.497 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M712.932 1544.91 L729.252 1544.91 L729.252 1548.85 L707.307 1548.85 L707.307 1544.91 Q709.969 1542.16 714.553 1537.53 Q719.159 1532.88 720.34 1531.53 Q722.585 1529.01 723.465 1527.27 Q724.367 1525.51 724.367 1523.82 Q724.367 1521.07 722.423 1519.33 Q720.502 1517.6 717.4 1517.6 Q715.201 1517.6 712.747 1518.36 Q710.317 1519.13 707.539 1520.68 L707.539 1515.95 Q710.363 1514.82 712.817 1514.24 Q715.27 1513.66 717.307 1513.66 Q722.678 1513.66 725.872 1516.35 Q729.066 1519.03 729.066 1523.52 Q729.066 1525.65 728.256 1527.57 Q727.469 1529.47 725.363 1532.07 Q724.784 1532.74 721.682 1535.95 Q718.58 1539.15 712.932 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M739.113 1514.29 L757.469 1514.29 L757.469 1518.22 L743.395 1518.22 L743.395 1526.7 Q744.414 1526.35 745.432 1526.19 Q746.451 1526 747.469 1526 Q753.256 1526 756.636 1529.17 Q760.015 1532.34 760.015 1537.76 Q760.015 1543.34 756.543 1546.44 Q753.071 1549.52 746.752 1549.52 Q744.576 1549.52 742.307 1549.15 Q740.062 1548.78 737.654 1548.04 L737.654 1543.34 Q739.738 1544.47 741.96 1545.03 Q744.182 1545.58 746.659 1545.58 Q750.664 1545.58 753.001 1543.48 Q755.339 1541.37 755.339 1537.76 Q755.339 1534.15 753.001 1532.04 Q750.664 1529.94 746.659 1529.94 Q744.784 1529.94 742.909 1530.35 Q741.057 1530.77 739.113 1531.65 L739.113 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1227.52 1514.29 L1245.88 1514.29 L1245.88 1518.22 L1231.81 1518.22 L1231.81 1526.7 Q1232.83 1526.35 1233.84 1526.19 Q1234.86 1526 1235.88 1526 Q1241.67 1526 1245.05 1529.17 Q1248.43 1532.34 1248.43 1537.76 Q1248.43 1543.34 1244.95 1546.44 Q1241.48 1549.52 1235.16 1549.52 Q1232.99 1549.52 1230.72 1549.15 Q1228.47 1548.78 1226.07 1548.04 L1226.07 1543.34 Q1228.15 1544.47 1230.37 1545.03 Q1232.59 1545.58 1235.07 1545.58 Q1239.08 1545.58 1241.41 1543.48 Q1243.75 1541.37 1243.75 1537.76 Q1243.75 1534.15 1241.41 1532.04 Q1239.08 1529.94 1235.07 1529.94 Q1233.2 1529.94 1231.32 1530.35 Q1229.47 1530.77 1227.52 1531.65 L1227.52 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1267.64 1517.37 Q1264.03 1517.37 1262.2 1520.93 Q1260.39 1524.47 1260.39 1531.6 Q1260.39 1538.71 1262.2 1542.27 Q1264.03 1545.82 1267.64 1545.82 Q1271.27 1545.82 1273.08 1542.27 Q1274.91 1538.71 1274.91 1531.6 Q1274.91 1524.47 1273.08 1520.93 Q1271.27 1517.37 1267.64 1517.37 M1267.64 1513.66 Q1273.45 1513.66 1276.51 1518.27 Q1279.58 1522.85 1279.58 1531.6 Q1279.58 1540.33 1276.51 1544.94 Q1273.45 1549.52 1267.64 1549.52 Q1261.83 1549.52 1258.75 1544.94 Q1255.7 1540.33 1255.7 1531.6 Q1255.7 1522.85 1258.75 1518.27 Q1261.83 1513.66 1267.64 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1745.84 1514.29 L1768.07 1514.29 L1768.07 1516.28 L1755.52 1548.85 L1750.63 1548.85 L1762.44 1518.22 L1745.84 1518.22 L1745.84 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1777.23 1514.29 L1795.59 1514.29 L1795.59 1518.22 L1781.51 1518.22 L1781.51 1526.7 Q1782.53 1526.35 1783.55 1526.19 Q1784.57 1526 1785.59 1526 Q1791.38 1526 1794.76 1529.17 Q1798.13 1532.34 1798.13 1537.76 Q1798.13 1543.34 1794.66 1546.44 Q1791.19 1549.52 1784.87 1549.52 Q1782.7 1549.52 1780.43 1549.15 Q1778.18 1548.78 1775.77 1548.04 L1775.77 1543.34 Q1777.86 1544.47 1780.08 1545.03 Q1782.3 1545.58 1784.78 1545.58 Q1788.78 1545.58 1791.12 1543.48 Q1793.46 1541.37 1793.46 1537.76 Q1793.46 1534.15 1791.12 1532.04 Q1788.78 1529.94 1784.78 1529.94 Q1782.9 1529.94 1781.03 1530.35 Q1779.18 1530.77 1777.23 1531.65 L1777.23 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2250.76 1544.91 L2258.4 1544.91 L2258.4 1518.55 L2250.09 1520.21 L2250.09 1515.95 L2258.35 1514.29 L2263.03 1514.29 L2263.03 1544.91 L2270.67 1544.91 L2270.67 1548.85 L2250.76 1548.85 L2250.76 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2290.11 1517.37 Q2286.5 1517.37 2284.67 1520.93 Q2282.87 1524.47 2282.87 1531.6 Q2282.87 1538.71 2284.67 1542.27 Q2286.5 1545.82 2290.11 1545.82 Q2293.75 1545.82 2295.55 1542.27 Q2297.38 1538.71 2297.38 1531.6 Q2297.38 1524.47 2295.55 1520.93 Q2293.75 1517.37 2290.11 1517.37 M2290.11 1513.66 Q2295.92 1513.66 2298.98 1518.27 Q2302.06 1522.85 2302.06 1531.6 Q2302.06 1540.33 2298.98 1544.94 Q2295.92 1549.52 2290.11 1549.52 Q2284.3 1549.52 2281.22 1544.94 Q2278.17 1540.33 2278.17 1531.6 Q2278.17 1522.85 2281.22 1518.27 Q2284.3 1513.66 2290.11 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2320.27 1517.37 Q2316.66 1517.37 2314.83 1520.93 Q2313.03 1524.47 2313.03 1531.6 Q2313.03 1538.71 2314.83 1542.27 Q2316.66 1545.82 2320.27 1545.82 Q2323.91 1545.82 2325.71 1542.27 Q2327.54 1538.71 2327.54 1531.6 Q2327.54 1524.47 2325.71 1520.93 Q2323.91 1517.37 2320.27 1517.37 M2320.27 1513.66 Q2326.08 1513.66 2329.14 1518.27 Q2332.22 1522.85 2332.22 1531.6 Q2332.22 1540.33 2329.14 1544.94 Q2326.08 1549.52 2320.27 1549.52 Q2314.46 1549.52 2311.38 1544.94 Q2308.33 1540.33 2308.33 1531.6 Q2308.33 1522.85 2311.38 1518.27 Q2314.46 1513.66 2320.27 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1464.39 2352.76,1464.39 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1119.64 2352.76,1119.64 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,774.897 2352.76,774.897 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,430.151 2352.76,430.151 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,85.4059 2352.76,85.4059 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 169.121,47.2441 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1464.39 188.019,1464.39 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1119.64 188.019,1119.64 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,774.897 188.019,774.897 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,430.151 188.019,430.151 "/>
-<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,85.4059 188.019,85.4059 "/>
-<path clip-path="url(#clip530)" d="M121.177 1450.19 Q117.566 1450.19 115.737 1453.75 Q113.932 1457.29 113.932 1464.42 Q113.932 1471.53 115.737 1475.09 Q117.566 1478.64 121.177 1478.64 Q124.811 1478.64 126.617 1475.09 Q128.445 1471.53 128.445 1464.42 Q128.445 1457.29 126.617 1453.75 Q124.811 1450.19 121.177 1450.19 M121.177 1446.48 Q126.987 1446.48 130.043 1451.09 Q133.121 1455.67 133.121 1464.42 Q133.121 1473.15 130.043 1477.76 Q126.987 1482.34 121.177 1482.34 Q115.367 1482.34 112.288 1477.76 Q109.233 1473.15 109.233 1464.42 Q109.233 1455.67 112.288 1451.09 Q115.367 1446.48 121.177 1446.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M86.0382 1132.99 L102.358 1132.99 L102.358 1136.92 L80.4133 1136.92 L80.4133 1132.99 Q83.0753 1130.23 87.6586 1125.6 Q92.2651 1120.95 93.4456 1119.61 Q95.691 1117.08 96.5706 1115.35 Q97.4734 1113.59 97.4734 1111.9 Q97.4734 1109.14 95.5289 1107.41 Q93.6076 1105.67 90.5058 1105.67 Q88.3067 1105.67 85.8531 1106.44 Q83.4225 1107.2 80.6447 1108.75 L80.6447 1104.03 Q83.4688 1102.89 85.9225 1102.32 Q88.3762 1101.74 90.4132 1101.74 Q95.7836 1101.74 98.978 1104.42 Q102.172 1107.11 102.172 1111.6 Q102.172 1113.73 101.362 1115.65 Q100.575 1117.55 98.4687 1120.14 Q97.89 1120.81 94.7882 1124.03 Q91.6864 1127.22 86.0382 1132.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M112.219 1102.36 L130.575 1102.36 L130.575 1106.3 L116.501 1106.3 L116.501 1114.77 Q117.52 1114.42 118.538 1114.26 Q119.557 1114.08 120.575 1114.08 Q126.362 1114.08 129.742 1117.25 Q133.121 1120.42 133.121 1125.83 Q133.121 1131.41 129.649 1134.51 Q126.177 1137.59 119.857 1137.59 Q117.682 1137.59 115.413 1137.22 Q113.168 1136.85 110.76 1136.11 L110.76 1131.41 Q112.844 1132.55 115.066 1133.1 Q117.288 1133.66 119.765 1133.66 Q123.77 1133.66 126.107 1131.55 Q128.445 1129.45 128.445 1125.83 Q128.445 1122.22 126.107 1120.12 Q123.77 1118.01 119.765 1118.01 Q117.89 1118.01 116.015 1118.43 Q114.163 1118.84 112.219 1119.72 L112.219 1102.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M81.0614 757.617 L99.4178 757.617 L99.4178 761.552 L85.3438 761.552 L85.3438 770.024 Q86.3623 769.677 87.3808 769.515 Q88.3993 769.33 89.4178 769.33 Q95.2049 769.33 98.5845 772.501 Q101.964 775.672 101.964 781.089 Q101.964 786.668 98.4919 789.769 Q95.0197 792.848 88.7003 792.848 Q86.5243 792.848 84.2558 792.478 Q82.0105 792.107 79.6031 791.367 L79.6031 786.668 Q81.6864 787.802 83.9086 788.357 Q86.1308 788.913 88.6077 788.913 Q92.6123 788.913 94.9502 786.806 Q97.2882 784.7 97.2882 781.089 Q97.2882 777.478 94.9502 775.371 Q92.6123 773.265 88.6077 773.265 Q86.7327 773.265 84.8577 773.681 Q83.0058 774.098 81.0614 774.978 L81.0614 757.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M121.177 760.695 Q117.566 760.695 115.737 764.26 Q113.932 767.802 113.932 774.931 Q113.932 782.038 115.737 785.603 Q117.566 789.144 121.177 789.144 Q124.811 789.144 126.617 785.603 Q128.445 782.038 128.445 774.931 Q128.445 767.802 126.617 764.26 Q124.811 760.695 121.177 760.695 M121.177 756.992 Q126.987 756.992 130.043 761.598 Q133.121 766.182 133.121 774.931 Q133.121 783.658 130.043 788.265 Q126.987 792.848 121.177 792.848 Q115.367 792.848 112.288 788.265 Q109.233 783.658 109.233 774.931 Q109.233 766.182 112.288 761.598 Q115.367 756.992 121.177 756.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M80.8299 412.871 L103.052 412.871 L103.052 414.862 L90.5058 447.431 L85.6216 447.431 L97.4271 416.807 L80.8299 416.807 L80.8299 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M112.219 412.871 L130.575 412.871 L130.575 416.807 L116.501 416.807 L116.501 425.279 Q117.52 424.931 118.538 424.769 Q119.557 424.584 120.575 424.584 Q126.362 424.584 129.742 427.756 Q133.121 430.927 133.121 436.343 Q133.121 441.922 129.649 445.024 Q126.177 448.103 119.857 448.103 Q117.682 448.103 115.413 447.732 Q113.168 447.362 110.76 446.621 L110.76 441.922 Q112.844 443.056 115.066 443.612 Q117.288 444.167 119.765 444.167 Q123.77 444.167 126.107 442.061 Q128.445 439.955 128.445 436.343 Q128.445 432.732 126.107 430.626 Q123.77 428.519 119.765 428.519 Q117.89 428.519 116.015 428.936 Q114.163 429.353 112.219 430.232 L112.219 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M51.6634 98.7508 L59.3023 98.7508 L59.3023 72.3852 L50.9921 74.0518 L50.9921 69.7926 L59.256 68.1259 L63.9319 68.1259 L63.9319 98.7508 L71.5707 98.7508 L71.5707 102.686 L51.6634 102.686 L51.6634 98.7508 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M91.0151 71.2046 Q87.404 71.2046 85.5753 74.7694 Q83.7697 78.3111 83.7697 85.4407 Q83.7697 92.5471 85.5753 96.1119 Q87.404 99.6535 91.0151 99.6535 Q94.6493 99.6535 96.4548 96.1119 Q98.2835 92.5471 98.2835 85.4407 Q98.2835 78.3111 96.4548 74.7694 Q94.6493 71.2046 91.0151 71.2046 M91.0151 67.5009 Q96.8252 67.5009 99.8808 72.1074 Q102.959 76.6907 102.959 85.4407 Q102.959 94.1675 99.8808 98.7739 Q96.8252 103.357 91.0151 103.357 Q85.2049 103.357 82.1262 98.7739 Q79.0707 94.1675 79.0707 85.4407 Q79.0707 76.6907 82.1262 72.1074 Q85.2049 67.5009 91.0151 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M121.177 71.2046 Q117.566 71.2046 115.737 74.7694 Q113.932 78.3111 113.932 85.4407 Q113.932 92.5471 115.737 96.1119 Q117.566 99.6535 121.177 99.6535 Q124.811 99.6535 126.617 96.1119 Q128.445 92.5471 128.445 85.4407 Q128.445 78.3111 126.617 74.7694 Q124.811 71.2046 121.177 71.2046 M121.177 67.5009 Q126.987 67.5009 130.043 72.1074 Q133.121 76.6907 133.121 85.4407 Q133.121 94.1675 130.043 98.7739 Q126.987 103.357 121.177 103.357 Q115.367 103.357 112.288 98.7739 Q109.233 94.1675 109.233 85.4407 Q109.233 76.6907 112.288 72.1074 Q115.367 67.5009 121.177 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 704.657,952.826 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 704.657,952.826 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 934.094,231.349 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 934.094,231.349 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 863.882,570.733 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 863.882,570.733 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 230.922,889.298 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 230.922,889.298 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 1229.18,1265.03 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1229.18,1265.03 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 652.606,606.105 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 652.606,606.105 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2191.3,261.454 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2191.3,261.454 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 2290.95,740.457 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 2290.95,740.457 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 737.113,1131.24 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 737.113,1131.24 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 2263.46,244.214 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 2263.46,244.214 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1368.6,856.398 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1368.6,856.398 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1122.22,906.351 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 1122.22,906.351 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 1096.49,448.372 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1096.49,448.372 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1820.22,149.11 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 1820.22,149.11 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 798.433,1333.13 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 798.433,1333.13 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 649.499,747.765 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 649.499,747.765 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 736.525,1445.72 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 736.525,1445.72 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 256.809,1046.01 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 256.809,1046.01 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 811.954,495.574 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 811.954,495.574 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 1999.41,641.774 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 1999.41,641.774 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 374.298,583.308 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 374.298,583.308 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 1544.38,261.436 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 1544.38,261.436 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 2028.55,707.83 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2028.55,707.83 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 793.017,1101.73 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 793.017,1101.73 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1774.71,738.432 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 1774.71,738.432 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1553.7,823.134 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1553.7,823.134 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 376.116,1084.45 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 376.116,1084.45 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 1975.89,827.543 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1975.89,827.543 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 392.332,151.788 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 392.332,151.788 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 1363.31,1066.22 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 1363.31,1066.22 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1176.72,552.561 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1176.72,552.561 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 600.384,1196.08 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 600.384,1196.08 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 447.043,922.181 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 447.043,922.181 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2241.96,98.1744 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 2241.96,98.1744 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 321.683,705.487 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 321.683,705.487 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1331.8,663.5 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1331.8,663.5 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 1160.81,404.835 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1160.81,404.835 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 794.705,747.973 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 794.705,747.973 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 584.653,754.401 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 584.653,754.401 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 1354.55,87.9763 "/>
-<polyline clip-path="url(#clip532)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1354.55,87.9763 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="214.497,1486.45 214.497,47.2441 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="733.661,1486.45 733.661,47.2441 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1252.83,1486.45 1252.83,47.2441 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1771.99,1486.45 1771.99,47.2441 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2291.15,1486.45 2291.15,47.2441 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="214.497,1486.45 214.497,1467.55 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="733.661,1486.45 733.661,1467.55 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1252.83,1486.45 1252.83,1467.55 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1771.99,1486.45 1771.99,1467.55 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2291.15,1486.45 2291.15,1467.55 "/>
+<path clip-path="url(#clip410)" d="M214.497 1517.37 Q210.886 1517.37 209.058 1520.93 Q207.252 1524.47 207.252 1531.6 Q207.252 1538.71 209.058 1542.27 Q210.886 1545.82 214.497 1545.82 Q218.132 1545.82 219.937 1542.27 Q221.766 1538.71 221.766 1531.6 Q221.766 1524.47 219.937 1520.93 Q218.132 1517.37 214.497 1517.37 M214.497 1513.66 Q220.308 1513.66 223.363 1518.27 Q226.442 1522.85 226.442 1531.6 Q226.442 1540.33 223.363 1544.94 Q220.308 1549.52 214.497 1549.52 Q208.687 1549.52 205.609 1544.94 Q202.553 1540.33 202.553 1531.6 Q202.553 1522.85 205.609 1518.27 Q208.687 1513.66 214.497 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M712.932 1544.91 L729.252 1544.91 L729.252 1548.85 L707.307 1548.85 L707.307 1544.91 Q709.969 1542.16 714.553 1537.53 Q719.159 1532.88 720.34 1531.53 Q722.585 1529.01 723.465 1527.27 Q724.367 1525.51 724.367 1523.82 Q724.367 1521.07 722.423 1519.33 Q720.502 1517.6 717.4 1517.6 Q715.201 1517.6 712.747 1518.36 Q710.317 1519.13 707.539 1520.68 L707.539 1515.95 Q710.363 1514.82 712.817 1514.24 Q715.27 1513.66 717.307 1513.66 Q722.678 1513.66 725.872 1516.35 Q729.066 1519.03 729.066 1523.52 Q729.066 1525.65 728.256 1527.57 Q727.469 1529.47 725.363 1532.07 Q724.784 1532.74 721.682 1535.95 Q718.58 1539.15 712.932 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M739.113 1514.29 L757.469 1514.29 L757.469 1518.22 L743.395 1518.22 L743.395 1526.7 Q744.414 1526.35 745.432 1526.19 Q746.451 1526 747.469 1526 Q753.256 1526 756.636 1529.17 Q760.015 1532.34 760.015 1537.76 Q760.015 1543.34 756.543 1546.44 Q753.071 1549.52 746.752 1549.52 Q744.576 1549.52 742.307 1549.15 Q740.062 1548.78 737.654 1548.04 L737.654 1543.34 Q739.738 1544.47 741.96 1545.03 Q744.182 1545.58 746.659 1545.58 Q750.664 1545.58 753.001 1543.48 Q755.339 1541.37 755.339 1537.76 Q755.339 1534.15 753.001 1532.04 Q750.664 1529.94 746.659 1529.94 Q744.784 1529.94 742.909 1530.35 Q741.057 1530.77 739.113 1531.65 L739.113 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1227.52 1514.29 L1245.88 1514.29 L1245.88 1518.22 L1231.81 1518.22 L1231.81 1526.7 Q1232.83 1526.35 1233.84 1526.19 Q1234.86 1526 1235.88 1526 Q1241.67 1526 1245.05 1529.17 Q1248.43 1532.34 1248.43 1537.76 Q1248.43 1543.34 1244.95 1546.44 Q1241.48 1549.52 1235.16 1549.52 Q1232.99 1549.52 1230.72 1549.15 Q1228.47 1548.78 1226.07 1548.04 L1226.07 1543.34 Q1228.15 1544.47 1230.37 1545.03 Q1232.59 1545.58 1235.07 1545.58 Q1239.08 1545.58 1241.41 1543.48 Q1243.75 1541.37 1243.75 1537.76 Q1243.75 1534.15 1241.41 1532.04 Q1239.08 1529.94 1235.07 1529.94 Q1233.2 1529.94 1231.32 1530.35 Q1229.47 1530.77 1227.52 1531.65 L1227.52 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1267.64 1517.37 Q1264.03 1517.37 1262.2 1520.93 Q1260.39 1524.47 1260.39 1531.6 Q1260.39 1538.71 1262.2 1542.27 Q1264.03 1545.82 1267.64 1545.82 Q1271.27 1545.82 1273.08 1542.27 Q1274.91 1538.71 1274.91 1531.6 Q1274.91 1524.47 1273.08 1520.93 Q1271.27 1517.37 1267.64 1517.37 M1267.64 1513.66 Q1273.45 1513.66 1276.51 1518.27 Q1279.58 1522.85 1279.58 1531.6 Q1279.58 1540.33 1276.51 1544.94 Q1273.45 1549.52 1267.64 1549.52 Q1261.83 1549.52 1258.75 1544.94 Q1255.7 1540.33 1255.7 1531.6 Q1255.7 1522.85 1258.75 1518.27 Q1261.83 1513.66 1267.64 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1745.84 1514.29 L1768.07 1514.29 L1768.07 1516.28 L1755.52 1548.85 L1750.63 1548.85 L1762.44 1518.22 L1745.84 1518.22 L1745.84 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M1777.23 1514.29 L1795.59 1514.29 L1795.59 1518.22 L1781.51 1518.22 L1781.51 1526.7 Q1782.53 1526.35 1783.55 1526.19 Q1784.57 1526 1785.59 1526 Q1791.38 1526 1794.76 1529.17 Q1798.13 1532.34 1798.13 1537.76 Q1798.13 1543.34 1794.66 1546.44 Q1791.19 1549.52 1784.87 1549.52 Q1782.7 1549.52 1780.43 1549.15 Q1778.18 1548.78 1775.77 1548.04 L1775.77 1543.34 Q1777.86 1544.47 1780.08 1545.03 Q1782.3 1545.58 1784.78 1545.58 Q1788.78 1545.58 1791.12 1543.48 Q1793.46 1541.37 1793.46 1537.76 Q1793.46 1534.15 1791.12 1532.04 Q1788.78 1529.94 1784.78 1529.94 Q1782.9 1529.94 1781.03 1530.35 Q1779.18 1530.77 1777.23 1531.65 L1777.23 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2250.76 1544.91 L2258.4 1544.91 L2258.4 1518.55 L2250.09 1520.21 L2250.09 1515.95 L2258.35 1514.29 L2263.03 1514.29 L2263.03 1544.91 L2270.67 1544.91 L2270.67 1548.85 L2250.76 1548.85 L2250.76 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2290.11 1517.37 Q2286.5 1517.37 2284.67 1520.93 Q2282.87 1524.47 2282.87 1531.6 Q2282.87 1538.71 2284.67 1542.27 Q2286.5 1545.82 2290.11 1545.82 Q2293.75 1545.82 2295.55 1542.27 Q2297.38 1538.71 2297.38 1531.6 Q2297.38 1524.47 2295.55 1520.93 Q2293.75 1517.37 2290.11 1517.37 M2290.11 1513.66 Q2295.92 1513.66 2298.98 1518.27 Q2302.06 1522.85 2302.06 1531.6 Q2302.06 1540.33 2298.98 1544.94 Q2295.92 1549.52 2290.11 1549.52 Q2284.3 1549.52 2281.22 1544.94 Q2278.17 1540.33 2278.17 1531.6 Q2278.17 1522.85 2281.22 1518.27 Q2284.3 1513.66 2290.11 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M2320.27 1517.37 Q2316.66 1517.37 2314.83 1520.93 Q2313.03 1524.47 2313.03 1531.6 Q2313.03 1538.71 2314.83 1542.27 Q2316.66 1545.82 2320.27 1545.82 Q2323.91 1545.82 2325.71 1542.27 Q2327.54 1538.71 2327.54 1531.6 Q2327.54 1524.47 2325.71 1520.93 Q2323.91 1517.37 2320.27 1517.37 M2320.27 1513.66 Q2326.08 1513.66 2329.14 1518.27 Q2332.22 1522.85 2332.22 1531.6 Q2332.22 1540.33 2329.14 1544.94 Q2326.08 1549.52 2320.27 1549.52 Q2314.46 1549.52 2311.38 1544.94 Q2308.33 1540.33 2308.33 1531.6 Q2308.33 1522.85 2311.38 1518.27 Q2314.46 1513.66 2320.27 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1464.39 2352.76,1464.39 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,1119.64 2352.76,1119.64 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,774.897 2352.76,774.897 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,430.151 2352.76,430.151 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="169.121,85.4059 2352.76,85.4059 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1486.45 169.121,47.2441 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1464.39 188.019,1464.39 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,1119.64 188.019,1119.64 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,774.897 188.019,774.897 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,430.151 188.019,430.151 "/>
+<polyline clip-path="url(#clip410)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="169.121,85.4059 188.019,85.4059 "/>
+<path clip-path="url(#clip410)" d="M121.177 1450.19 Q117.566 1450.19 115.737 1453.75 Q113.932 1457.29 113.932 1464.42 Q113.932 1471.53 115.737 1475.09 Q117.566 1478.64 121.177 1478.64 Q124.811 1478.64 126.617 1475.09 Q128.445 1471.53 128.445 1464.42 Q128.445 1457.29 126.617 1453.75 Q124.811 1450.19 121.177 1450.19 M121.177 1446.48 Q126.987 1446.48 130.043 1451.09 Q133.121 1455.67 133.121 1464.42 Q133.121 1473.15 130.043 1477.76 Q126.987 1482.34 121.177 1482.34 Q115.367 1482.34 112.288 1477.76 Q109.233 1473.15 109.233 1464.42 Q109.233 1455.67 112.288 1451.09 Q115.367 1446.48 121.177 1446.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M86.0382 1132.99 L102.358 1132.99 L102.358 1136.92 L80.4133 1136.92 L80.4133 1132.99 Q83.0753 1130.23 87.6586 1125.6 Q92.2651 1120.95 93.4456 1119.61 Q95.691 1117.08 96.5706 1115.35 Q97.4734 1113.59 97.4734 1111.9 Q97.4734 1109.14 95.5289 1107.41 Q93.6076 1105.67 90.5058 1105.67 Q88.3067 1105.67 85.8531 1106.44 Q83.4225 1107.2 80.6447 1108.75 L80.6447 1104.03 Q83.4688 1102.89 85.9225 1102.32 Q88.3762 1101.74 90.4132 1101.74 Q95.7836 1101.74 98.978 1104.42 Q102.172 1107.11 102.172 1111.6 Q102.172 1113.73 101.362 1115.65 Q100.575 1117.55 98.4687 1120.14 Q97.89 1120.81 94.7882 1124.03 Q91.6864 1127.22 86.0382 1132.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M112.219 1102.36 L130.575 1102.36 L130.575 1106.3 L116.501 1106.3 L116.501 1114.77 Q117.52 1114.42 118.538 1114.26 Q119.557 1114.08 120.575 1114.08 Q126.362 1114.08 129.742 1117.25 Q133.121 1120.42 133.121 1125.83 Q133.121 1131.41 129.649 1134.51 Q126.177 1137.59 119.857 1137.59 Q117.682 1137.59 115.413 1137.22 Q113.168 1136.85 110.76 1136.11 L110.76 1131.41 Q112.844 1132.55 115.066 1133.1 Q117.288 1133.66 119.765 1133.66 Q123.77 1133.66 126.107 1131.55 Q128.445 1129.45 128.445 1125.83 Q128.445 1122.22 126.107 1120.12 Q123.77 1118.01 119.765 1118.01 Q117.89 1118.01 116.015 1118.43 Q114.163 1118.84 112.219 1119.72 L112.219 1102.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M81.0614 757.617 L99.4178 757.617 L99.4178 761.552 L85.3438 761.552 L85.3438 770.024 Q86.3623 769.677 87.3808 769.515 Q88.3993 769.33 89.4178 769.33 Q95.2049 769.33 98.5845 772.501 Q101.964 775.672 101.964 781.089 Q101.964 786.668 98.4919 789.769 Q95.0197 792.848 88.7003 792.848 Q86.5243 792.848 84.2558 792.478 Q82.0105 792.107 79.6031 791.367 L79.6031 786.668 Q81.6864 787.802 83.9086 788.357 Q86.1308 788.913 88.6077 788.913 Q92.6123 788.913 94.9502 786.806 Q97.2882 784.7 97.2882 781.089 Q97.2882 777.478 94.9502 775.371 Q92.6123 773.265 88.6077 773.265 Q86.7327 773.265 84.8577 773.681 Q83.0058 774.098 81.0614 774.978 L81.0614 757.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M121.177 760.695 Q117.566 760.695 115.737 764.26 Q113.932 767.802 113.932 774.931 Q113.932 782.038 115.737 785.603 Q117.566 789.144 121.177 789.144 Q124.811 789.144 126.617 785.603 Q128.445 782.038 128.445 774.931 Q128.445 767.802 126.617 764.26 Q124.811 760.695 121.177 760.695 M121.177 756.992 Q126.987 756.992 130.043 761.598 Q133.121 766.182 133.121 774.931 Q133.121 783.658 130.043 788.265 Q126.987 792.848 121.177 792.848 Q115.367 792.848 112.288 788.265 Q109.233 783.658 109.233 774.931 Q109.233 766.182 112.288 761.598 Q115.367 756.992 121.177 756.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M80.8299 412.871 L103.052 412.871 L103.052 414.862 L90.5058 447.431 L85.6216 447.431 L97.4271 416.807 L80.8299 416.807 L80.8299 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M112.219 412.871 L130.575 412.871 L130.575 416.807 L116.501 416.807 L116.501 425.279 Q117.52 424.931 118.538 424.769 Q119.557 424.584 120.575 424.584 Q126.362 424.584 129.742 427.756 Q133.121 430.927 133.121 436.343 Q133.121 441.922 129.649 445.024 Q126.177 448.103 119.857 448.103 Q117.682 448.103 115.413 447.732 Q113.168 447.362 110.76 446.621 L110.76 441.922 Q112.844 443.056 115.066 443.612 Q117.288 444.167 119.765 444.167 Q123.77 444.167 126.107 442.061 Q128.445 439.955 128.445 436.343 Q128.445 432.732 126.107 430.626 Q123.77 428.519 119.765 428.519 Q117.89 428.519 116.015 428.936 Q114.163 429.353 112.219 430.232 L112.219 412.871 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M51.6634 98.7508 L59.3023 98.7508 L59.3023 72.3852 L50.9921 74.0518 L50.9921 69.7926 L59.256 68.1259 L63.9319 68.1259 L63.9319 98.7508 L71.5707 98.7508 L71.5707 102.686 L51.6634 102.686 L51.6634 98.7508 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M91.0151 71.2046 Q87.404 71.2046 85.5753 74.7694 Q83.7697 78.3111 83.7697 85.4407 Q83.7697 92.5471 85.5753 96.1119 Q87.404 99.6535 91.0151 99.6535 Q94.6493 99.6535 96.4548 96.1119 Q98.2835 92.5471 98.2835 85.4407 Q98.2835 78.3111 96.4548 74.7694 Q94.6493 71.2046 91.0151 71.2046 M91.0151 67.5009 Q96.8252 67.5009 99.8808 72.1074 Q102.959 76.6907 102.959 85.4407 Q102.959 94.1675 99.8808 98.7739 Q96.8252 103.357 91.0151 103.357 Q85.2049 103.357 82.1262 98.7739 Q79.0707 94.1675 79.0707 85.4407 Q79.0707 76.6907 82.1262 72.1074 Q85.2049 67.5009 91.0151 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip410)" d="M121.177 71.2046 Q117.566 71.2046 115.737 74.7694 Q113.932 78.3111 113.932 85.4407 Q113.932 92.5471 115.737 96.1119 Q117.566 99.6535 121.177 99.6535 Q124.811 99.6535 126.617 96.1119 Q128.445 92.5471 128.445 85.4407 Q128.445 78.3111 126.617 74.7694 Q124.811 71.2046 121.177 71.2046 M121.177 67.5009 Q126.987 67.5009 130.043 72.1074 Q133.121 76.6907 133.121 85.4407 Q133.121 94.1675 130.043 98.7739 Q126.987 103.357 121.177 103.357 Q115.367 103.357 112.288 98.7739 Q109.233 94.1675 109.233 85.4407 Q109.233 76.6907 112.288 72.1074 Q115.367 67.5009 121.177 67.5009 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip412)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 704.657,952.826 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 704.657,952.826 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 934.094,231.349 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 934.094,231.349 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 863.882,570.733 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 863.882,570.733 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 230.922,889.298 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 230.922,889.298 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 1229.18,1265.03 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1229.18,1265.03 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 652.606,606.105 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 652.606,606.105 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2191.3,261.454 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2191.3,261.454 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 2290.95,740.457 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 2290.95,740.457 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.017,1101.73 737.113,1131.24 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 737.113,1131.24 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 2263.46,244.214 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 2263.46,244.214 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1368.6,856.398 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1368.6,856.398 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.31,1066.22 1122.22,906.351 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 1122.22,906.351 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="811.954,495.574 1096.49,448.372 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1096.49,448.372 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1820.22,149.11 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2241.96,98.1744 1820.22,149.11 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 798.433,1333.13 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 798.433,1333.13 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="794.705,747.973 649.499,747.765 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.653,754.401 649.499,747.765 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="798.433,1333.13 736.525,1445.72 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.384,1196.08 736.525,1445.72 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 256.809,1046.01 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 256.809,1046.01 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.882,570.733 811.954,495.574 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 811.954,495.574 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2191.3,261.454 1999.41,641.774 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2028.55,707.83 1999.41,641.774 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="392.332,151.788 374.298,583.308 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="321.683,705.487 374.298,583.308 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 1544.38,261.436 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1354.55,87.9763 1544.38,261.436 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 2028.55,707.83 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1999.41,641.774 2028.55,707.83 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 793.017,1101.73 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 793.017,1101.73 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1553.7,823.134 1774.71,738.432 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1975.89,827.543 1774.71,738.432 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1553.7,823.134 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1553.7,823.134 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="256.809,1046.01 376.116,1084.45 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="447.043,922.181 376.116,1084.45 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2290.95,740.457 1975.89,827.543 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1774.71,738.432 1975.89,827.543 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 392.332,151.788 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 392.332,151.788 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1229.18,1265.03 1363.31,1066.22 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 1363.31,1066.22 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1331.8,663.5 1176.72,552.561 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1160.81,404.835 1176.72,552.561 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="737.113,1131.24 600.384,1196.08 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="736.525,1445.72 600.384,1196.08 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="704.657,952.826 447.043,922.181 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="376.116,1084.45 447.043,922.181 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2263.46,244.214 2241.96,98.1744 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1820.22,149.11 2241.96,98.1744 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ac8d18; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="230.922,889.298 321.683,705.487 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a9ad; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="374.298,583.308 321.683,705.487 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#ed5d92; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1368.6,856.398 1331.8,663.5 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#c68125; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1331.8,663.5 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a98d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1096.49,448.372 1160.81,404.835 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#8e971d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1176.72,552.561 1160.81,404.835 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#00a8cb; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1122.22,906.351 794.705,747.973 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#9b7fe8; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 794.705,747.973 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#608cf6; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="652.606,606.105 584.653,754.401 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#f05f73; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="649.499,747.765 584.653,754.401 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#dd64b5; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="934.094,231.349 1354.55,87.9763 "/>
+<polyline clip-path="url(#clip412)" style="stroke:#6b9e32; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.38,261.436 1354.55,87.9763 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/index.html b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/index.html
index 37968c4bd09..11c33e49681 100644
--- a/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/index.html
+++ b/previews/PR3562/tutorials/algorithms/tsp_lazy_constraints/index.html
@@ -59,7 +59,7 @@
     global cycle = subtour(iterative_model[:x])
 end
 
-objective_value(iterative_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">525.7039004442727</code></pre><pre><code class="language-julia hljs">time_iterated</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.028665781021118164</code></pre><p>As a quick sanity check, we visualize the optimal tour to verify that no subtour is present:</p><pre><code class="language-julia hljs">function plot_tour(X, Y, x)
+objective_value(iterative_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">525.7039004442727</code></pre><pre><code class="language-julia hljs">time_iterated</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.04599452018737793</code></pre><p>As a quick sanity check, we visualize the optimal tour to verify that no subtour is present:</p><pre><code class="language-julia hljs">function plot_tour(X, Y, x)
     plot = Plots.plot()
     for (i, j) in selected_edges(x, size(x, 1))
         Plots.plot!([X[i], X[j]], [Y[i], Y[j]]; legend = false)
@@ -67,7 +67,7 @@
     return plot
 end
 
-plot_tour(X, Y, value.(iterative_model[:x]))</code></pre><img src="88a7f5c2.svg" alt="Example block output"/><h3 id="Lazy-constraint-method"><a class="docs-heading-anchor" href="#Lazy-constraint-method">Lazy constraint method</a><a id="Lazy-constraint-method-1"></a><a class="docs-heading-anchor-permalink" href="#Lazy-constraint-method" title="Permalink"></a></h3><p>A more sophisticated approach makes use of <em>lazy constraints</em>. To be more precise, we do this through the <code>subtour_elimination_callback()</code> below, which is only run whenever we encounter a new integer-feasible solution.</p><pre><code class="language-julia hljs">lazy_model = build_tsp_model(d, n)
+plot_tour(X, Y, value.(iterative_model[:x]))</code></pre><img src="6096df34.svg" alt="Example block output"/><h3 id="Lazy-constraint-method"><a class="docs-heading-anchor" href="#Lazy-constraint-method">Lazy constraint method</a><a id="Lazy-constraint-method-1"></a><a class="docs-heading-anchor-permalink" href="#Lazy-constraint-method" title="Permalink"></a></h3><p>A more sophisticated approach makes use of <em>lazy constraints</em>. To be more precise, we do this through the <code>subtour_elimination_callback()</code> below, which is only run whenever we encounter a new integer-feasible solution.</p><pre><code class="language-julia hljs">lazy_model = build_tsp_model(d, n)
 function subtour_elimination_callback(cb_data)
     status = callback_node_status(cb_data, lazy_model)
     if status != MOI.CALLBACK_NODE_STATUS_INTEGER
@@ -91,4 +91,4 @@
     subtour_elimination_callback,
 )
 optimize!(lazy_model)
-objective_value(lazy_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">525.7039004442727</code></pre><p>This finds the same optimal tour:</p><pre><code class="language-julia hljs">plot_tour(X, Y, value.(lazy_model[:x]))</code></pre><img src="21585044.svg" alt="Example block output"/><p>Surprisingly, for this particular model with GLPK, the solution time is worse than the iterative method:</p><pre><code class="language-julia hljs">time_lazy = solve_time(lazy_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.13727283477783203</code></pre><p>In most other cases and solvers, however, the lazy time should be faster than the iterative method.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../cutting_stock_column_generation/">« Column generation</a><a class="docs-footer-nextpage" href="../parallelism/">Parallelism »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+objective_value(lazy_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">525.7039004442727</code></pre><p>This finds the same optimal tour:</p><pre><code class="language-julia hljs">plot_tour(X, Y, value.(lazy_model[:x]))</code></pre><img src="6e144cc3.svg" alt="Example block output"/><p>Surprisingly, for this particular model with GLPK, the solution time is worse than the iterative method:</p><pre><code class="language-julia hljs">time_lazy = solve_time(lazy_model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.27559399604797363</code></pre><p>In most other cases and solvers, however, the lazy time should be faster than the iterative method.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../cutting_stock_column_generation/">« Column generation</a><a class="docs-footer-nextpage" href="../parallelism/">Parallelism »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/applications/optimal_power_flow/index.html b/previews/PR3562/tutorials/applications/optimal_power_flow/index.html
index b4613eadb89..aa43b177b49 100644
--- a/previews/PR3562/tutorials/applications/optimal_power_flow/index.html
+++ b/previews/PR3562/tutorials/applications/optimal_power_flow/index.html
@@ -126,7 +126,7 @@
   Objective value    : 3.08784e+03
 
 * Work counters
-  Solve time (sec)   : 7.86710e-03
+  Solve time (sec)   : 1.23641e-02
   Barrier iterations : 16
 </code></pre><pre><code class="language-julia hljs">objval_solution = round(objective_value(model); digits = 2)
 println(&quot;Objective value (feasible solution) : $(objval_solution)&quot;)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Objective value (feasible solution) : 3087.84</code></pre><p>The solution&#39;s power generation (in rectangular form) and complex voltage values (in polar form using degrees) are:</p><pre><code class="language-julia hljs">DataFrames.DataFrame(;
@@ -134,7 +134,7 @@
     ComplexPowerGen = round.(value.(S_G); digits = 2),
     VoltageMagnitude = round.(abs.(value.(V)); digits = 2),
     VoltageAngle_Deg = round.(rad2deg.(angle.(value.(V))); digits = 2),
-)</code></pre><div><div style = "float: left;"><span>9×4 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Bus</th><th style = "text-align: left;">ComplexPowerGen</th><th style = "text-align: left;">VoltageMagnitude</th><th style = "text-align: left;">VoltageAngle_Deg</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "ComplexF64" style = "text-align: left;">Complex…</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">10.0-5.0im</td><td style = "text-align: right;">0.91</td><td style = "text-align: right;">-0.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">125.37-5.0im</td><td style = "text-align: right;">0.92</td><td style = "text-align: right;">12.37</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">57.03-5.0im</td><td style = "text-align: right;">0.94</td><td style = "text-align: right;">7.01</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.91</td><td style = "text-align: right;">-0.4</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">5</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.92</td><td style = "text-align: right;">-0.73</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">6</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.94</td><td style = "text-align: right;">4.84</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">7</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.93</td><td style = "text-align: right;">4.52</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">8</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.93</td><td style = "text-align: right;">7.12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">9</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">-0.63</td></tr></tbody></table></div><h2 id="Relaxations-and-better-objective-bounds"><a class="docs-heading-anchor" href="#Relaxations-and-better-objective-bounds">Relaxations and better objective bounds</a><a id="Relaxations-and-better-objective-bounds-1"></a><a class="docs-heading-anchor-permalink" href="#Relaxations-and-better-objective-bounds" title="Permalink"></a></h2><p>The Ipopt solver uses an interior-point algorithm. It has local optimality guarantees, but is unable to certify whether the solution is globally optimal. The solution we found is indeed globally optimal. The work to verify this has been done in Bukhsh et al. (2013) and Krasko and Rebennack (2017), and different solvers (such as Gurobi, SCIP and GLOMIQO) are also able to verify this.</p><p>The techniques of <em>convex relaxations</em> can also be used to improve on our current best lower bound:</p><pre><code class="language-julia hljs">better_lower_bound</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2733.55</code></pre><p>To this end, observe that the nonlinear constraints in the AC-OPF formulation are quadratic equalities for power flow along with quadratic voltage inequalities.</p><p>Let&#39;s linearize these constraints by first making the substitution <span>$W = V V^*$</span>, where:</p><p class="math-container">\[W = V V^* \quad \iff  \quad W_{ii} = | V_i |^2, \quad W_{ik} = V_i \; \overline{V_k}, \quad \forall i, \, k \in \{ 1, \ldots, N \}\]</p><p>and where <span>$V^*$</span> is the <a href="https://en.wikipedia.org/wiki/Conjugate_transpose">conjugate transpose</a> of <span>$V$</span>.</p><p>On the face of it, this turns a quadratic voltage bound constraint like:</p><p class="math-container">\[v_L \leq |V_i |^2  \leq v_U, \quad i  \in \{ 1, \ldots, N \}\]</p><p>for some real <span>$v_L$</span> and <span>$v_U$</span> into a simple two-sided bound:</p><p class="math-container">\[v_L \leq W_{ii}  \leq v_U,\]</p><p>while each quadratic expression for the nodal power term:</p><p class="math-container">\[    S^{Node}_i = V_i \overline{(YV)_i}\]</p><p>becomes the linear combination:</p><p class="math-container">\[    S^{Node}_i = (E_{ii} Y^T) \bullet W.\]</p><p>Here <span>$A \bullet B = \operatorname{tr}(A^* B)$</span> is the <a href="https://en.wikipedia.org/wiki/Frobenius_inner_product">Frobenius inner product</a> of two complex matrices, while <span>$E_{kn}$</span> denotes the <em>matrix unit</em> with a single nonzero entry of 1 in row <span>$k$</span> and column <span>$n$</span>.</p><pre><code class="language-julia hljs">E(k, n) = SparseArrays.sparse([k], [n], 1, N, N);</code></pre><p>Of course, we&#39;ve shifted the nonlinearity into the equality constraint <span>$W = V V^*$</span>: it is this constraint we will now relax using a semidefinite programming approach.</p><p>We will make use of complex voltages and relax <span>$W = V V^*$</span> to;</p><p class="math-container">\[W \succeq V V^*,\]</p><p>where the relation <span>$\succeq$</span> is the ordering in the Hermitian positive semidefinite cone.</p><p>The above constraint is equivalent to:</p><p class="math-container">\[\begin{bmatrix} 1 &amp; V^* \\ V &amp; W \\ \end{bmatrix} \succeq 0\]</p><p>by the theory of the <a href="https://en.wikipedia.org/wiki/Schur_complement">Schur complement</a>. This matrix inequality implies a number of second-order cone constraints by taking certain <span>$2 \times 2$</span> minors of the matrix for each <span>$i  \in \{ 1, \ldots, N \}$</span>:</p><p class="math-container">\[\begin{bmatrix} 1 &amp; V_i^* \\ V_i &amp; W_{ii} \\ \end{bmatrix} \succeq 0,\]</p><p>which is equivalent to the real second-order cone inequality:</p><p class="math-container">\[\operatorname{real}(W_{ii}) \geq \operatorname{real}(V_i)^2 + \operatorname{imag}(V_i)^2.\]</p><p>We include these implied constraints as well for demonstration purposes.</p><p>Putting it all together we get the following semidefinite relaxation of the AC-OPF problem:</p><pre><code class="language-julia hljs">model = Model(Clarabel.Optimizer)
+)</code></pre><div><div style = "float: left;"><span>9×4 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Bus</th><th style = "text-align: left;">ComplexPowerGen</th><th style = "text-align: left;">VoltageMagnitude</th><th style = "text-align: left;">VoltageAngle_Deg</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "ComplexF64" style = "text-align: left;">Complex…</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">10.0-5.0im</td><td style = "text-align: right;">0.91</td><td style = "text-align: right;">0.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">125.37-5.0im</td><td style = "text-align: right;">0.92</td><td style = "text-align: right;">12.37</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">57.03-5.0im</td><td style = "text-align: right;">0.94</td><td style = "text-align: right;">7.01</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.91</td><td style = "text-align: right;">-0.4</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">5</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.92</td><td style = "text-align: right;">-0.73</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">6</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.94</td><td style = "text-align: right;">4.84</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">7</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.93</td><td style = "text-align: right;">4.52</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">8</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.93</td><td style = "text-align: right;">7.12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">9</td><td style = "text-align: right;">0.0+0.0im</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">-0.63</td></tr></tbody></table></div><h2 id="Relaxations-and-better-objective-bounds"><a class="docs-heading-anchor" href="#Relaxations-and-better-objective-bounds">Relaxations and better objective bounds</a><a id="Relaxations-and-better-objective-bounds-1"></a><a class="docs-heading-anchor-permalink" href="#Relaxations-and-better-objective-bounds" title="Permalink"></a></h2><p>The Ipopt solver uses an interior-point algorithm. It has local optimality guarantees, but is unable to certify whether the solution is globally optimal. The solution we found is indeed globally optimal. The work to verify this has been done in Bukhsh et al. (2013) and Krasko and Rebennack (2017), and different solvers (such as Gurobi, SCIP and GLOMIQO) are also able to verify this.</p><p>The techniques of <em>convex relaxations</em> can also be used to improve on our current best lower bound:</p><pre><code class="language-julia hljs">better_lower_bound</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2733.55</code></pre><p>To this end, observe that the nonlinear constraints in the AC-OPF formulation are quadratic equalities for power flow along with quadratic voltage inequalities.</p><p>Let&#39;s linearize these constraints by first making the substitution <span>$W = V V^*$</span>, where:</p><p class="math-container">\[W = V V^* \quad \iff  \quad W_{ii} = | V_i |^2, \quad W_{ik} = V_i \; \overline{V_k}, \quad \forall i, \, k \in \{ 1, \ldots, N \}\]</p><p>and where <span>$V^*$</span> is the <a href="https://en.wikipedia.org/wiki/Conjugate_transpose">conjugate transpose</a> of <span>$V$</span>.</p><p>On the face of it, this turns a quadratic voltage bound constraint like:</p><p class="math-container">\[v_L \leq |V_i |^2  \leq v_U, \quad i  \in \{ 1, \ldots, N \}\]</p><p>for some real <span>$v_L$</span> and <span>$v_U$</span> into a simple two-sided bound:</p><p class="math-container">\[v_L \leq W_{ii}  \leq v_U,\]</p><p>while each quadratic expression for the nodal power term:</p><p class="math-container">\[    S^{Node}_i = V_i \overline{(YV)_i}\]</p><p>becomes the linear combination:</p><p class="math-container">\[    S^{Node}_i = (E_{ii} Y^T) \bullet W.\]</p><p>Here <span>$A \bullet B = \operatorname{tr}(A^* B)$</span> is the <a href="https://en.wikipedia.org/wiki/Frobenius_inner_product">Frobenius inner product</a> of two complex matrices, while <span>$E_{kn}$</span> denotes the <em>matrix unit</em> with a single nonzero entry of 1 in row <span>$k$</span> and column <span>$n$</span>.</p><pre><code class="language-julia hljs">E(k, n) = SparseArrays.sparse([k], [n], 1, N, N);</code></pre><p>Of course, we&#39;ve shifted the nonlinearity into the equality constraint <span>$W = V V^*$</span>: it is this constraint we will now relax using a semidefinite programming approach.</p><p>We will make use of complex voltages and relax <span>$W = V V^*$</span> to;</p><p class="math-container">\[W \succeq V V^*,\]</p><p>where the relation <span>$\succeq$</span> is the ordering in the Hermitian positive semidefinite cone.</p><p>The above constraint is equivalent to:</p><p class="math-container">\[\begin{bmatrix} 1 &amp; V^* \\ V &amp; W \\ \end{bmatrix} \succeq 0\]</p><p>by the theory of the <a href="https://en.wikipedia.org/wiki/Schur_complement">Schur complement</a>. This matrix inequality implies a number of second-order cone constraints by taking certain <span>$2 \times 2$</span> minors of the matrix for each <span>$i  \in \{ 1, \ldots, N \}$</span>:</p><p class="math-container">\[\begin{bmatrix} 1 &amp; V_i^* \\ V_i &amp; W_{ii} \\ \end{bmatrix} \succeq 0,\]</p><p>which is equivalent to the real second-order cone inequality:</p><p class="math-container">\[\operatorname{real}(W_{ii}) \geq \operatorname{real}(V_i)^2 + \operatorname{imag}(V_i)^2.\]</p><p>We include these implied constraints as well for demonstration purposes.</p><p>Putting it all together we get the following semidefinite relaxation of the AC-OPF problem:</p><pre><code class="language-julia hljs">model = Model(Clarabel.Optimizer)
 set_attribute(model, &quot;tol_gap_rel&quot;, 1e-3)
 set_attribute(model, &quot;tol_feas&quot;, 1e-3)
 set_attribute(model, &quot;tol_ktratio&quot;, 1e-3)
@@ -209,23 +209,23 @@
   6   1.5768e+03   1.5579e+03  1.21e-02  5.34e-07  6.55e-05  9.28e-02  3.09e-01  9.19e-01
   7   1.5691e+03   1.5654e+03  2.35e-03  9.56e-08  1.22e-05  1.96e-02  5.73e-02  8.55e-01
   8   1.5657e+03   1.5640e+03  1.08e-03  3.99e-08  5.32e-06  8.54e-03  2.47e-02  7.64e-01
-  9   1.5653e+03   1.5641e+03  7.79e-04  2.84e-08  3.82e-06  6.06e-03  1.76e-02  4.20e-01
+  9   1.5654e+03   1.5641e+03  8.03e-04  2.94e-08  3.94e-06  6.27e-03  1.82e-02  3.85e-01
 ---------------------------------------------------------------------------------------------
 Terminated with status = solved
-solve time = 67.7ms</code></pre><pre><code class="language-julia hljs">sdp_relaxation_lower_bound = round(objective_value(model); digits = 2)
+solve time =  102ms</code></pre><pre><code class="language-julia hljs">sdp_relaxation_lower_bound = round(objective_value(model); digits = 2)
 println(
     &quot;Objective value (W &amp; V relax. lower bound): $sdp_relaxation_lower_bound&quot;,
-)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Objective value (W &amp; V relax. lower bound): 2754.09</code></pre><p>We can more easily see solution values by rounding out noisy data:</p><pre><code class="language-julia hljs">W_1 = SparseArrays.sparse(round.(value.(W); digits = 2))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">9×9 SparseArrays.SparseMatrixCSC{ComplexF64, Int64} with 81 stored entries:
- 1.19-0.0im   1.14-0.07im  1.13-0.05im  …  1.16-0.02im  1.14+0.05im
+)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Objective value (W &amp; V relax. lower bound): 2754.12</code></pre><p>We can more easily see solution values by rounding out noisy data:</p><pre><code class="language-julia hljs">W_1 = SparseArrays.sparse(round.(value.(W); digits = 2))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">9×9 SparseArrays.SparseMatrixCSC{ComplexF64, Int64} with 81 stored entries:
+ 1.19-0.0im   1.14-0.07im  1.13-0.05im  …  1.15-0.02im  1.14+0.05im
  1.14+0.07im   1.2-0.0im   1.15+0.02im     1.17+0.05im  1.14+0.12im
  1.13+0.05im  1.15-0.02im  1.19-0.0im      1.16+0.03im  1.14+0.1im
  1.14-0.03im  1.15-0.1im   1.15-0.08im     1.17-0.05im  1.15+0.03im
- 1.15-0.05im  1.15-0.12im  1.15-0.1im      1.17-0.07im  1.16+0.01im
- 1.15+0.01im  1.17-0.06im  1.16-0.03im  …  1.18-0.0im   1.16+0.07im
- 1.15-0.01im  1.16-0.08im  1.16-0.06im     1.18-0.03im  1.16+0.05im
- 1.16+0.02im  1.17-0.05im  1.16-0.03im     1.18-0.0im   1.16+0.07im
+ 1.14-0.05im  1.15-0.12im  1.15-0.1im      1.17-0.07im  1.15+0.01im
+ 1.15+0.01im  1.16-0.06im  1.16-0.03im  …  1.18-0.0im   1.16+0.07im
+ 1.15-0.01im  1.16-0.08im  1.15-0.06im     1.18-0.03im  1.16+0.05im
+ 1.15+0.02im  1.17-0.05im  1.16-0.03im     1.18-0.0im   1.16+0.07im
  1.14-0.05im  1.14-0.12im  1.14-0.1im      1.16-0.07im  1.15-0.0im</code></pre><p>and recover an approximation to the voltage variables as:</p><pre><code class="language-julia hljs">DataFrames.DataFrame(;
     Bus = 1:N,
     Magnitude = round.(abs.(value.(V)); digits = 2),
     AngleDeg = round.(rad2deg.(angle.(value.(V))); digits = 2),
-)</code></pre><div><div style = "float: left;"><span>9×3 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Bus</th><th style = "text-align: left;">Magnitude</th><th style = "text-align: left;">AngleDeg</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.95</td><td style = "text-align: right;">-0.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">0.84</td><td style = "text-align: right;">3.77</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">0.83</td><td style = "text-align: right;">2.67</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0.85</td><td style = "text-align: right;">-1.21</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">5</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">-2.12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">6</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">0.94</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">7</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">-0.17</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">8</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">1.14</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">9</td><td style = "text-align: right;">0.85</td><td style = "text-align: right;">-2.46</td></tr></tbody></table></div><p>For further information on exploiting sparsity see Jabr (2012).</p><p>This relaxation has the advantage that we can work directly with complex voltages to extend the formulation, strengthen the relaxation and gain additional approximate information about the voltage variables.</p><h2 id="References-and-further-resources"><a class="docs-heading-anchor" href="#References-and-further-resources">References and further resources</a><a id="References-and-further-resources-1"></a><a class="docs-heading-anchor-permalink" href="#References-and-further-resources" title="Permalink"></a></h2><p><strong>Bukhsh</strong>, W. A., Grothey, A., McKinnon, K. I., &amp; Trodden, P. A. <a href="https://doi.org/10.1109/TPWRS.2013.2274577"><em>Local solutions of the optimal power flow problem.</em></a> IEEE Transactions on Power Systems, 28(4), 4780-4788 (2013).</p><p><strong>Jabr</strong>, R. A. <a href="https://doi.org/10.1109/TPWRS.2011.2170772"><em>Exploiting sparsity in SDP relaxations of the OPF problem.</em></a> IEEE Transactions on Power Systems, 27(2), 1138-1139 (2011).</p><p><strong>Krasko</strong>, V., &amp; S. <strong>Rebennack</strong>. <a href="https://doi.org/10.1137/1.9781611974683.ch15"><em>Chapter 15: Global Optimization: Optimal Power Flow Problem.</em></a> In Advances and Trends in Optimization with Engineering Applications, 187—205. MOS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics, 2017.</p><p><a href="https://www.maths.ed.ac.uk/optenergy/LocalOpt/9busnetwork.html"><strong>Test case <code>case9mod</code></strong></a>: from the <a href="https://www.maths.ed.ac.uk/optenergy/LocalOpt/">Test Case Archive of Optimal Power Flow (OPF) Problems with Local Optima</a>.</p><p>R. D. Zimmerman, C. E. Murillo-Sanchez, &amp; R. J. Thomas, <a href="https://doi.org/10.1109/TPWRS.2010.2051168"><em>MATPOWER: Steady-State Operations, Planning and Analysis Tools for Power Systems Research and Education,</em></a> IEEE Transactions on Power Systems, 26(1), 12-19 (2011).</p><p><strong>MATPOWER data format</strong>: <a href="https://matpower.org/docs/MATPOWER-manual.pdf">MATPOWER manual</a>; see especially Appendix B &quot;Data File Format&quot; and Table B-3.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../power_systems/">« Power Systems</a><a class="docs-footer-nextpage" href="../web_app/">Serving web apps »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><div><div style = "float: left;"><span>9×3 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Bus</th><th style = "text-align: left;">Magnitude</th><th style = "text-align: left;">AngleDeg</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.95</td><td style = "text-align: right;">-0.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">2</td><td style = "text-align: right;">0.84</td><td style = "text-align: right;">3.79</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">3</td><td style = "text-align: right;">0.83</td><td style = "text-align: right;">2.68</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0.85</td><td style = "text-align: right;">-1.2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">5</td><td style = "text-align: right;">0.85</td><td style = "text-align: right;">-2.11</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">6</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">0.95</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">7</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">-0.16</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">8</td><td style = "text-align: right;">0.86</td><td style = "text-align: right;">1.16</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">9</td><td style = "text-align: right;">0.85</td><td style = "text-align: right;">-2.45</td></tr></tbody></table></div><p>For further information on exploiting sparsity see Jabr (2012).</p><p>This relaxation has the advantage that we can work directly with complex voltages to extend the formulation, strengthen the relaxation and gain additional approximate information about the voltage variables.</p><h2 id="References-and-further-resources"><a class="docs-heading-anchor" href="#References-and-further-resources">References and further resources</a><a id="References-and-further-resources-1"></a><a class="docs-heading-anchor-permalink" href="#References-and-further-resources" title="Permalink"></a></h2><p><strong>Bukhsh</strong>, W. A., Grothey, A., McKinnon, K. I., &amp; Trodden, P. A. <a href="https://doi.org/10.1109/TPWRS.2013.2274577"><em>Local solutions of the optimal power flow problem.</em></a> IEEE Transactions on Power Systems, 28(4), 4780-4788 (2013).</p><p><strong>Jabr</strong>, R. A. <a href="https://doi.org/10.1109/TPWRS.2011.2170772"><em>Exploiting sparsity in SDP relaxations of the OPF problem.</em></a> IEEE Transactions on Power Systems, 27(2), 1138-1139 (2011).</p><p><strong>Krasko</strong>, V., &amp; S. <strong>Rebennack</strong>. <a href="https://doi.org/10.1137/1.9781611974683.ch15"><em>Chapter 15: Global Optimization: Optimal Power Flow Problem.</em></a> In Advances and Trends in Optimization with Engineering Applications, 187—205. MOS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics, 2017.</p><p><a href="https://www.maths.ed.ac.uk/optenergy/LocalOpt/9busnetwork.html"><strong>Test case <code>case9mod</code></strong></a>: from the <a href="https://www.maths.ed.ac.uk/optenergy/LocalOpt/">Test Case Archive of Optimal Power Flow (OPF) Problems with Local Optima</a>.</p><p>R. D. Zimmerman, C. E. Murillo-Sanchez, &amp; R. J. Thomas, <a href="https://doi.org/10.1109/TPWRS.2010.2051168"><em>MATPOWER: Steady-State Operations, Planning and Analysis Tools for Power Systems Research and Education,</em></a> IEEE Transactions on Power Systems, 26(1), 12-19 (2011).</p><p><strong>MATPOWER data format</strong>: <a href="https://matpower.org/docs/MATPOWER-manual.pdf">MATPOWER manual</a>; see especially Appendix B &quot;Data File Format&quot; and Table B-3.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../power_systems/">« Power Systems</a><a class="docs-footer-nextpage" href="../web_app/">Serving web apps »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/applications/power_systems/2588df61.svg b/previews/PR3562/tutorials/applications/power_systems/0215129c.svg
similarity index 85%
rename from previews/PR3562/tutorials/applications/power_systems/2588df61.svg
rename to previews/PR3562/tutorials/applications/power_systems/0215129c.svg
index ea59417e004..f0283a30e6d 100644
--- a/previews/PR3562/tutorials/applications/power_systems/2588df61.svg
+++ b/previews/PR3562/tutorials/applications/power_systems/0215129c.svg
@@ -1,43 +1,43 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip770">
+  <clipPath id="clip590">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip770)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip590)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip771">
+  <clipPath id="clip591">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip770)" d="M234.125 1423.18 L2352.76 1423.18 L2352.76 47.2441 L234.125 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip590)" d="M234.125 1423.18 L2352.76 1423.18 L2352.76 47.2441 L234.125 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip772">
+  <clipPath id="clip592">
     <rect x="234" y="47" width="2120" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="294.087,1423.18 294.087,47.2441 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="793.764,1423.18 793.764,47.2441 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1293.44,1423.18 1293.44,47.2441 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1793.12,1423.18 1793.12,47.2441 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2292.79,1423.18 2292.79,47.2441 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="294.087,1423.18 294.087,1404.28 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.764,1423.18 793.764,1404.28 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1293.44,1423.18 1293.44,1404.28 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1793.12,1423.18 1793.12,1404.28 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2292.79,1423.18 2292.79,1404.28 "/>
-<path clip-path="url(#clip770)" d="M294.087 1454.1 Q290.476 1454.1 288.647 1457.66 Q286.841 1461.2 286.841 1468.33 Q286.841 1475.44 288.647 1479.01 Q290.476 1482.55 294.087 1482.55 Q297.721 1482.55 299.526 1479.01 Q301.355 1475.44 301.355 1468.33 Q301.355 1461.2 299.526 1457.66 Q297.721 1454.1 294.087 1454.1 M294.087 1450.39 Q299.897 1450.39 302.952 1455 Q306.031 1459.58 306.031 1468.33 Q306.031 1477.06 302.952 1481.67 Q299.897 1486.25 294.087 1486.25 Q288.277 1486.25 285.198 1481.67 Q282.142 1477.06 282.142 1468.33 Q282.142 1459.58 285.198 1455 Q288.277 1450.39 294.087 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M773.035 1481.64 L789.354 1481.64 L789.354 1485.58 L767.41 1485.58 L767.41 1481.64 Q770.072 1478.89 774.655 1474.26 Q779.261 1469.61 780.442 1468.27 Q782.687 1465.74 783.567 1464.01 Q784.47 1462.25 784.47 1460.56 Q784.47 1457.8 782.525 1456.07 Q780.604 1454.33 777.502 1454.33 Q775.303 1454.33 772.849 1455.09 Q770.419 1455.86 767.641 1457.41 L767.641 1452.69 Q770.465 1451.55 772.919 1450.97 Q775.373 1450.39 777.41 1450.39 Q782.78 1450.39 785.974 1453.08 Q789.169 1455.77 789.169 1460.26 Q789.169 1462.39 788.359 1464.31 Q787.572 1466.2 785.465 1468.8 Q784.886 1469.47 781.785 1472.69 Q778.683 1475.88 773.035 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M799.215 1451.02 L817.571 1451.02 L817.571 1454.96 L803.497 1454.96 L803.497 1463.43 Q804.516 1463.08 805.534 1462.92 Q806.553 1462.73 807.571 1462.73 Q813.358 1462.73 816.738 1465.9 Q820.118 1469.08 820.118 1474.49 Q820.118 1480.07 816.646 1483.17 Q813.173 1486.25 806.854 1486.25 Q804.678 1486.25 802.409 1485.88 Q800.164 1485.51 797.757 1484.77 L797.757 1480.07 Q799.84 1481.2 802.062 1481.76 Q804.284 1482.32 806.761 1482.32 Q810.766 1482.32 813.104 1480.21 Q815.442 1478.1 815.442 1474.49 Q815.442 1470.88 813.104 1468.77 Q810.766 1466.67 806.761 1466.67 Q804.886 1466.67 803.011 1467.08 Q801.159 1467.5 799.215 1468.38 L799.215 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1268.14 1451.02 L1286.5 1451.02 L1286.5 1454.96 L1272.42 1454.96 L1272.42 1463.43 Q1273.44 1463.08 1274.46 1462.92 Q1275.48 1462.73 1276.5 1462.73 Q1282.28 1462.73 1285.66 1465.9 Q1289.04 1469.08 1289.04 1474.49 Q1289.04 1480.07 1285.57 1483.17 Q1282.1 1486.25 1275.78 1486.25 Q1273.6 1486.25 1271.33 1485.88 Q1269.09 1485.51 1266.68 1484.77 L1266.68 1480.07 Q1268.76 1481.2 1270.99 1481.76 Q1273.21 1482.32 1275.69 1482.32 Q1279.69 1482.32 1282.03 1480.21 Q1284.37 1478.1 1284.37 1474.49 Q1284.37 1470.88 1282.03 1468.77 Q1279.69 1466.67 1275.69 1466.67 Q1273.81 1466.67 1271.94 1467.08 Q1270.08 1467.5 1268.14 1468.38 L1268.14 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1308.26 1454.1 Q1304.64 1454.1 1302.82 1457.66 Q1301.01 1461.2 1301.01 1468.33 Q1301.01 1475.44 1302.82 1479.01 Q1304.64 1482.55 1308.26 1482.55 Q1311.89 1482.55 1313.7 1479.01 Q1315.52 1475.44 1315.52 1468.33 Q1315.52 1461.2 1313.7 1457.66 Q1311.89 1454.1 1308.26 1454.1 M1308.26 1450.39 Q1314.07 1450.39 1317.12 1455 Q1320.2 1459.58 1320.2 1468.33 Q1320.2 1477.06 1317.12 1481.67 Q1314.07 1486.25 1308.26 1486.25 Q1302.45 1486.25 1299.37 1481.67 Q1296.31 1477.06 1296.31 1468.33 Q1296.31 1459.58 1299.37 1455 Q1302.45 1450.39 1308.26 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1766.97 1451.02 L1789.19 1451.02 L1789.19 1453.01 L1776.65 1485.58 L1771.76 1485.58 L1783.57 1454.96 L1766.97 1454.96 L1766.97 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1798.36 1451.02 L1816.72 1451.02 L1816.72 1454.96 L1802.64 1454.96 L1802.64 1463.43 Q1803.66 1463.08 1804.68 1462.92 Q1805.7 1462.73 1806.72 1462.73 Q1812.5 1462.73 1815.88 1465.9 Q1819.26 1469.08 1819.26 1474.49 Q1819.26 1480.07 1815.79 1483.17 Q1812.32 1486.25 1806 1486.25 Q1803.82 1486.25 1801.56 1485.88 Q1799.31 1485.51 1796.9 1484.77 L1796.9 1480.07 Q1798.99 1481.2 1801.21 1481.76 Q1803.43 1482.32 1805.91 1482.32 Q1809.91 1482.32 1812.25 1480.21 Q1814.59 1478.1 1814.59 1474.49 Q1814.59 1470.88 1812.25 1468.77 Q1809.91 1466.67 1805.91 1466.67 Q1804.03 1466.67 1802.16 1467.08 Q1800.31 1467.5 1798.36 1468.38 L1798.36 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2252.4 1481.64 L2260.04 1481.64 L2260.04 1455.28 L2251.73 1456.95 L2251.73 1452.69 L2259.99 1451.02 L2264.67 1451.02 L2264.67 1481.64 L2272.31 1481.64 L2272.31 1485.58 L2252.4 1485.58 L2252.4 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2291.75 1454.1 Q2288.14 1454.1 2286.31 1457.66 Q2284.51 1461.2 2284.51 1468.33 Q2284.51 1475.44 2286.31 1479.01 Q2288.14 1482.55 2291.75 1482.55 Q2295.39 1482.55 2297.19 1479.01 Q2299.02 1475.44 2299.02 1468.33 Q2299.02 1461.2 2297.19 1457.66 Q2295.39 1454.1 2291.75 1454.1 M2291.75 1450.39 Q2297.56 1450.39 2300.62 1455 Q2303.7 1459.58 2303.7 1468.33 Q2303.7 1477.06 2300.62 1481.67 Q2297.56 1486.25 2291.75 1486.25 Q2285.94 1486.25 2282.86 1481.67 Q2279.81 1477.06 2279.81 1468.33 Q2279.81 1459.58 2282.86 1455 Q2285.94 1450.39 2291.75 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2321.91 1454.1 Q2318.3 1454.1 2316.48 1457.66 Q2314.67 1461.2 2314.67 1468.33 Q2314.67 1475.44 2316.48 1479.01 Q2318.3 1482.55 2321.91 1482.55 Q2325.55 1482.55 2327.35 1479.01 Q2329.18 1475.44 2329.18 1468.33 Q2329.18 1461.2 2327.35 1457.66 Q2325.55 1454.1 2321.91 1454.1 M2321.91 1450.39 Q2327.73 1450.39 2330.78 1455 Q2333.86 1459.58 2333.86 1468.33 Q2333.86 1477.06 2330.78 1481.67 Q2327.73 1486.25 2321.91 1486.25 Q2316.1 1486.25 2313.03 1481.67 Q2309.97 1477.06 2309.97 1468.33 Q2309.97 1459.58 2313.03 1455 Q2316.1 1450.39 2321.91 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1261.91 1524.18 L1261.91 1530.96 Q1258.67 1527.94 1254.98 1526.44 Q1251.32 1524.95 1247.18 1524.95 Q1239.03 1524.95 1234.7 1529.95 Q1230.37 1534.91 1230.37 1544.33 Q1230.37 1553.72 1234.7 1558.72 Q1239.03 1563.68 1247.18 1563.68 Q1251.32 1563.68 1254.98 1562.19 Q1258.67 1560.69 1261.91 1557.67 L1261.91 1564.38 Q1258.54 1566.68 1254.75 1567.82 Q1251 1568.97 1246.8 1568.97 Q1236.01 1568.97 1229.8 1562.38 Q1223.59 1555.76 1223.59 1544.33 Q1223.59 1532.87 1229.8 1526.28 Q1236.01 1519.66 1246.8 1519.66 Q1251.06 1519.66 1254.82 1520.81 Q1258.6 1521.92 1261.91 1524.18 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1285.4 1536.5 Q1280.69 1536.5 1277.96 1540.19 Q1275.22 1543.85 1275.22 1550.25 Q1275.22 1556.65 1277.92 1560.34 Q1280.66 1564 1285.4 1564 Q1290.08 1564 1292.82 1560.31 Q1295.56 1556.62 1295.56 1550.25 Q1295.56 1543.92 1292.82 1540.23 Q1290.08 1536.5 1285.4 1536.5 M1285.4 1531.54 Q1293.04 1531.54 1297.4 1536.5 Q1301.76 1541.47 1301.76 1550.25 Q1301.76 1559 1297.4 1564 Q1293.04 1568.97 1285.4 1568.97 Q1277.73 1568.97 1273.37 1564 Q1269.04 1559 1269.04 1550.25 Q1269.04 1541.47 1273.37 1536.5 Q1277.73 1531.54 1285.4 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1334.2 1533.45 L1334.2 1538.98 Q1331.71 1537.71 1329.04 1537.07 Q1326.37 1536.44 1323.5 1536.44 Q1319.14 1536.44 1316.95 1537.77 Q1314.78 1539.11 1314.78 1541.79 Q1314.78 1543.82 1316.34 1545 Q1317.9 1546.15 1322.61 1547.2 L1324.62 1547.64 Q1330.86 1548.98 1333.47 1551.43 Q1336.11 1553.85 1336.11 1558.21 Q1336.11 1563.17 1332.16 1566.07 Q1328.25 1568.97 1321.37 1568.97 Q1318.51 1568.97 1315.39 1568.39 Q1312.3 1567.85 1308.86 1566.74 L1308.86 1560.69 Q1312.11 1562.38 1315.26 1563.24 Q1318.41 1564.07 1321.5 1564.07 Q1325.64 1564.07 1327.86 1562.66 Q1330.09 1561.23 1330.09 1558.65 Q1330.09 1556.27 1328.47 1554.99 Q1326.88 1553.72 1321.43 1552.54 L1319.4 1552.07 Q1313.95 1550.92 1311.54 1548.56 Q1309.12 1546.18 1309.12 1542.04 Q1309.12 1537.01 1312.68 1534.27 Q1316.25 1531.54 1322.8 1531.54 Q1326.05 1531.54 1328.91 1532.01 Q1331.78 1532.49 1334.2 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1351.23 1522.27 L1351.23 1532.4 L1363.29 1532.4 L1363.29 1536.95 L1351.23 1536.95 L1351.23 1556.3 Q1351.23 1560.66 1352.4 1561.9 Q1353.61 1563.14 1357.27 1563.14 L1363.29 1563.14 L1363.29 1568.04 L1357.27 1568.04 Q1350.49 1568.04 1347.92 1565.53 Q1345.34 1562.98 1345.34 1556.3 L1345.34 1536.95 L1341.04 1536.95 L1341.04 1532.4 L1345.34 1532.4 L1345.34 1522.27 L1351.23 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,1410.55 2352.76,1410.55 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,1079.46 2352.76,1079.46 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,748.368 2352.76,748.368 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,417.277 2352.76,417.277 "/>
-<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,86.1857 2352.76,86.1857 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1423.18 234.125,47.2441 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1410.55 253.023,1410.55 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1079.46 253.023,1079.46 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,748.368 253.023,748.368 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,417.277 253.023,417.277 "/>
-<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,86.1857 253.023,86.1857 "/>
-<path clip-path="url(#clip770)" d="M186.181 1396.35 Q182.57 1396.35 180.741 1399.91 Q178.936 1403.45 178.936 1410.58 Q178.936 1417.69 180.741 1421.26 Q182.57 1424.8 186.181 1424.8 Q189.815 1424.8 191.621 1421.26 Q193.45 1417.69 193.45 1410.58 Q193.45 1403.45 191.621 1399.91 Q189.815 1396.35 186.181 1396.35 M186.181 1392.64 Q191.991 1392.64 195.047 1397.25 Q198.125 1401.83 198.125 1410.58 Q198.125 1419.31 195.047 1423.92 Q191.991 1428.5 186.181 1428.5 Q180.371 1428.5 177.292 1423.92 Q174.237 1419.31 174.237 1410.58 Q174.237 1401.83 177.292 1397.25 Q180.371 1392.64 186.181 1392.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M146.066 1062.18 L164.422 1062.18 L164.422 1066.11 L150.348 1066.11 L150.348 1074.59 Q151.366 1074.24 152.385 1074.08 Q153.403 1073.89 154.422 1073.89 Q160.209 1073.89 163.589 1077.06 Q166.968 1080.23 166.968 1085.65 Q166.968 1091.23 163.496 1094.33 Q160.024 1097.41 153.704 1097.41 Q151.528 1097.41 149.26 1097.04 Q147.015 1096.67 144.607 1095.93 L144.607 1091.23 Q146.691 1092.36 148.913 1092.92 Q151.135 1093.47 153.612 1093.47 Q157.616 1093.47 159.954 1091.37 Q162.292 1089.26 162.292 1085.65 Q162.292 1082.04 159.954 1079.93 Q157.616 1077.83 153.612 1077.83 Q151.737 1077.83 149.862 1078.24 Q148.01 1078.66 146.066 1079.54 L146.066 1062.18 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M186.181 1065.26 Q182.57 1065.26 180.741 1068.82 Q178.936 1072.36 178.936 1079.49 Q178.936 1086.6 180.741 1090.16 Q182.57 1093.71 186.181 1093.71 Q189.815 1093.71 191.621 1090.16 Q193.45 1086.6 193.45 1079.49 Q193.45 1072.36 191.621 1068.82 Q189.815 1065.26 186.181 1065.26 M186.181 1061.55 Q191.991 1061.55 195.047 1066.16 Q198.125 1070.74 198.125 1079.49 Q198.125 1088.22 195.047 1092.83 Q191.991 1097.41 186.181 1097.41 Q180.371 1097.41 177.292 1092.83 Q174.237 1088.22 174.237 1079.49 Q174.237 1070.74 177.292 1066.16 Q180.371 1061.55 186.181 1061.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M116.668 761.713 L124.306 761.713 L124.306 735.347 L115.996 737.014 L115.996 732.754 L124.26 731.088 L128.936 731.088 L128.936 761.713 L136.575 761.713 L136.575 765.648 L116.668 765.648 L116.668 761.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M156.019 734.166 Q152.408 734.166 150.579 737.731 Q148.774 741.273 148.774 748.402 Q148.774 755.509 150.579 759.074 Q152.408 762.615 156.019 762.615 Q159.653 762.615 161.459 759.074 Q163.288 755.509 163.288 748.402 Q163.288 741.273 161.459 737.731 Q159.653 734.166 156.019 734.166 M156.019 730.463 Q161.829 730.463 164.885 735.069 Q167.964 739.653 167.964 748.402 Q167.964 757.129 164.885 761.736 Q161.829 766.319 156.019 766.319 Q150.209 766.319 147.13 761.736 Q144.075 757.129 144.075 748.402 Q144.075 739.653 147.13 735.069 Q150.209 730.463 156.019 730.463 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M186.181 734.166 Q182.57 734.166 180.741 737.731 Q178.936 741.273 178.936 748.402 Q178.936 755.509 180.741 759.074 Q182.57 762.615 186.181 762.615 Q189.815 762.615 191.621 759.074 Q193.45 755.509 193.45 748.402 Q193.45 741.273 191.621 737.731 Q189.815 734.166 186.181 734.166 M186.181 730.463 Q191.991 730.463 195.047 735.069 Q198.125 739.653 198.125 748.402 Q198.125 757.129 195.047 761.736 Q191.991 766.319 186.181 766.319 Q180.371 766.319 177.292 761.736 Q174.237 757.129 174.237 748.402 Q174.237 739.653 177.292 735.069 Q180.371 730.463 186.181 730.463 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M116.668 430.622 L124.306 430.622 L124.306 404.256 L115.996 405.923 L115.996 401.663 L124.26 399.997 L128.936 399.997 L128.936 430.622 L136.575 430.622 L136.575 434.557 L116.668 434.557 L116.668 430.622 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M146.066 399.997 L164.422 399.997 L164.422 403.932 L150.348 403.932 L150.348 412.404 Q151.366 412.057 152.385 411.895 Q153.403 411.71 154.422 411.71 Q160.209 411.71 163.589 414.881 Q166.968 418.052 166.968 423.469 Q166.968 429.047 163.496 432.149 Q160.024 435.228 153.704 435.228 Q151.528 435.228 149.26 434.858 Q147.015 434.487 144.607 433.747 L144.607 429.047 Q146.691 430.182 148.913 430.737 Q151.135 431.293 153.612 431.293 Q157.616 431.293 159.954 429.186 Q162.292 427.08 162.292 423.469 Q162.292 419.858 159.954 417.751 Q157.616 415.645 153.612 415.645 Q151.737 415.645 149.862 416.061 Q148.01 416.478 146.066 417.358 L146.066 399.997 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M186.181 403.075 Q182.57 403.075 180.741 406.64 Q178.936 410.182 178.936 417.311 Q178.936 424.418 180.741 427.983 Q182.57 431.524 186.181 431.524 Q189.815 431.524 191.621 427.983 Q193.45 424.418 193.45 417.311 Q193.45 410.182 191.621 406.64 Q189.815 403.075 186.181 403.075 M186.181 399.372 Q191.991 399.372 195.047 403.978 Q198.125 408.561 198.125 417.311 Q198.125 426.038 195.047 430.645 Q191.991 435.228 186.181 435.228 Q180.371 435.228 177.292 430.645 Q174.237 426.038 174.237 417.311 Q174.237 408.561 177.292 403.978 Q180.371 399.372 186.181 399.372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M119.885 99.5305 L136.204 99.5305 L136.204 103.466 L114.26 103.466 L114.26 99.5305 Q116.922 96.7759 121.505 92.1463 Q126.112 87.4935 127.292 86.151 Q129.538 83.6278 130.417 81.8917 Q131.32 80.1325 131.32 78.4427 Q131.32 75.6881 129.376 73.952 Q127.455 72.2159 124.353 72.2159 Q122.154 72.2159 119.7 72.9797 Q117.269 73.7436 114.492 75.2945 L114.492 70.5723 Q117.316 69.4381 119.769 68.8594 Q122.223 68.2807 124.26 68.2807 Q129.63 68.2807 132.825 70.9659 Q136.019 73.651 136.019 78.1417 Q136.019 80.2714 135.209 82.1927 Q134.422 84.0908 132.316 86.6834 Q131.737 87.3547 128.635 90.5722 Q125.533 93.7667 119.885 99.5305 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M156.019 71.9844 Q152.408 71.9844 150.579 75.5492 Q148.774 79.0908 148.774 86.2204 Q148.774 93.3269 150.579 96.8916 Q152.408 100.433 156.019 100.433 Q159.653 100.433 161.459 96.8916 Q163.288 93.3269 163.288 86.2204 Q163.288 79.0908 161.459 75.5492 Q159.653 71.9844 156.019 71.9844 M156.019 68.2807 Q161.829 68.2807 164.885 72.8871 Q167.964 77.4705 167.964 86.2204 Q167.964 94.9472 164.885 99.5537 Q161.829 104.137 156.019 104.137 Q150.209 104.137 147.13 99.5537 Q144.075 94.9472 144.075 86.2204 Q144.075 77.4705 147.13 72.8871 Q150.209 68.2807 156.019 68.2807 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M186.181 71.9844 Q182.57 71.9844 180.741 75.5492 Q178.936 79.0908 178.936 86.2204 Q178.936 93.3269 180.741 96.8916 Q182.57 100.433 186.181 100.433 Q189.815 100.433 191.621 96.8916 Q193.45 93.3269 193.45 86.2204 Q193.45 79.0908 191.621 75.5492 Q189.815 71.9844 186.181 71.9844 M186.181 68.2807 Q191.991 68.2807 195.047 72.8871 Q198.125 77.4705 198.125 86.2204 Q198.125 94.9472 195.047 99.5537 Q191.991 104.137 186.181 104.137 Q180.371 104.137 177.292 99.5537 Q174.237 94.9472 174.237 86.2204 Q174.237 77.4705 177.292 72.8871 Q180.371 68.2807 186.181 68.2807 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M21.7677 962.07 L58.7206 962.07 L58.7206 954.304 Q58.7206 944.469 54.2646 939.917 Q49.8086 935.334 40.1964 935.334 Q30.6479 935.334 26.2237 939.917 Q21.7677 944.469 21.7677 954.304 L21.7677 962.07 M16.4842 968.499 L16.4842 955.29 Q16.4842 941.477 22.2451 935.016 Q27.9743 928.554 40.1964 928.554 Q52.4822 928.554 58.2432 935.047 Q64.0042 941.54 64.0042 955.29 L64.0042 968.499 L16.4842 968.499 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M28.3562 918.56 L28.3562 912.704 L64.0042 912.704 L64.0042 918.56 L28.3562 918.56 M14.479 918.56 L14.479 912.704 L21.895 912.704 L21.895 918.56 L14.479 918.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M29.4065 877.724 L34.9447 877.724 Q33.6716 880.207 33.035 882.88 Q32.3984 885.554 32.3984 888.419 Q32.3984 892.779 33.7352 894.975 Q35.072 897.14 37.7456 897.14 Q39.7826 897.14 40.9603 895.58 Q42.1061 894.02 43.1565 889.31 L43.6021 887.305 Q44.9389 881.066 47.3897 878.456 Q49.8086 875.815 54.1691 875.815 Q59.1344 875.815 62.0308 879.761 Q64.9272 883.676 64.9272 890.551 Q64.9272 893.416 64.3543 896.535 Q63.8132 899.622 62.6992 903.06 L56.6518 903.06 Q58.3387 899.813 59.198 896.662 Q60.0256 893.511 60.0256 890.424 Q60.0256 886.286 58.6251 884.058 Q57.1929 881.83 54.6147 881.83 Q52.2276 881.83 50.9545 883.453 Q49.6813 885.045 48.5037 890.487 L48.0262 892.524 Q46.8804 897.967 44.5251 900.386 Q42.138 902.805 38.0002 902.805 Q32.9713 902.805 30.2341 899.24 Q27.4968 895.676 27.4968 889.119 Q27.4968 885.872 27.9743 883.008 Q28.4517 880.143 29.4065 877.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M58.657 860.823 L77.5631 860.823 L77.5631 866.712 L28.3562 866.712 L28.3562 860.823 L33.7671 860.823 Q30.5842 858.977 29.0564 856.176 Q27.4968 853.344 27.4968 849.429 Q27.4968 842.936 32.6531 838.893 Q37.8093 834.819 46.212 834.819 Q54.6147 834.819 59.771 838.893 Q64.9272 842.936 64.9272 849.429 Q64.9272 853.344 63.3994 856.176 Q61.8398 858.977 58.657 860.823 M46.212 840.899 Q39.7508 840.899 36.0905 843.572 Q32.3984 846.214 32.3984 850.861 Q32.3984 855.508 36.0905 858.182 Q39.7508 860.823 46.212 860.823 Q52.6732 860.823 56.3653 858.182 Q60.0256 855.508 60.0256 850.861 Q60.0256 846.214 56.3653 843.572 Q52.6732 840.899 46.212 840.899 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M46.0847 808.911 Q46.0847 816.009 47.7079 818.746 Q49.3312 821.483 53.2461 821.483 Q56.3653 821.483 58.2114 819.446 Q60.0256 817.377 60.0256 813.844 Q60.0256 808.975 56.5881 806.046 Q53.1188 803.086 47.3897 803.086 L46.0847 803.086 L46.0847 808.911 M43.6657 797.23 L64.0042 797.23 L64.0042 803.086 L58.5933 803.086 Q61.8398 805.092 63.3994 808.083 Q64.9272 811.075 64.9272 815.404 Q64.9272 820.878 61.8716 824.125 Q58.7843 827.34 53.6281 827.34 Q47.6125 827.34 44.5569 823.329 Q41.5014 819.287 41.5014 811.298 L41.5014 803.086 L40.9285 803.086 Q36.8862 803.086 34.6901 805.76 Q32.4621 808.402 32.4621 813.208 Q32.4621 816.263 33.1941 819.16 Q33.9262 822.056 35.3903 824.73 L29.9795 824.73 Q28.7381 821.515 28.1334 818.491 Q27.4968 815.468 27.4968 812.603 Q27.4968 804.869 31.5072 801.049 Q35.5176 797.23 43.6657 797.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M18.2347 779.374 L28.3562 779.374 L28.3562 767.311 L32.9077 767.311 L32.9077 779.374 L52.2594 779.374 Q56.6199 779.374 57.8613 778.196 Q59.1026 776.987 59.1026 773.327 L59.1026 767.311 L64.0042 767.311 L64.0042 773.327 Q64.0042 780.106 61.4897 782.684 Q58.9434 785.262 52.2594 785.262 L32.9077 785.262 L32.9077 789.559 L28.3562 789.559 L28.3562 785.262 L18.2347 785.262 L18.2347 779.374 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M29.7248 733.955 L35.1993 733.955 Q33.8307 736.437 33.1623 738.952 Q32.4621 741.434 32.4621 743.981 Q32.4621 749.678 36.0905 752.829 Q39.6872 755.98 46.212 755.98 Q52.7369 755.98 56.3653 752.829 Q59.9619 749.678 59.9619 743.981 Q59.9619 741.434 59.2935 738.952 Q58.5933 736.437 57.2247 733.955 L62.6355 733.955 Q63.7814 736.406 64.3543 739.047 Q64.9272 741.657 64.9272 744.617 Q64.9272 752.67 59.8664 757.412 Q54.8057 762.155 46.212 762.155 Q37.491 762.155 32.4939 757.381 Q27.4968 752.574 27.4968 744.235 Q27.4968 741.53 28.0697 738.952 Q28.6108 736.374 29.7248 733.955 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M42.4881 694.137 L64.0042 694.137 L64.0042 699.994 L42.679 699.994 Q37.6183 699.994 35.1038 701.967 Q32.5894 703.94 32.5894 707.887 Q32.5894 712.63 35.6131 715.367 Q38.6368 718.104 43.8567 718.104 L64.0042 718.104 L64.0042 723.992 L14.479 723.992 L14.479 718.104 L33.8944 718.104 Q30.6797 716.003 29.0883 713.171 Q27.4968 710.306 27.4968 706.582 Q27.4968 700.439 31.3163 697.288 Q35.1038 694.137 42.4881 694.137 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M14.479 662.277 L14.479 648.782 L19.0304 648.782 L19.0304 656.42 L68.0464 656.42 L68.0464 648.782 L72.5979 648.782 L72.5979 662.277 L14.479 662.277 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M16.4842 636.05 L16.4842 626.47 L48.8219 614.343 L16.4842 602.153 L16.4842 592.572 L64.0042 592.572 L64.0042 598.843 L22.277 598.843 L54.8694 611.097 L54.8694 617.558 L22.277 629.812 L64.0042 629.812 L64.0042 636.05 L16.4842 636.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M16.4842 584.042 L16.4842 577.549 L56.6518 567.555 L16.4842 557.593 L16.4842 550.368 L56.6518 540.374 L16.4842 530.411 L16.4842 523.886 L64.0042 535.822 L64.0042 543.907 L22.7544 553.933 L64.0042 564.054 L64.0042 572.138 L16.4842 584.042 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M14.479 501.925 L72.5979 501.925 L72.5979 515.42 L68.0464 515.42 L68.0464 507.813 L19.0304 507.813 L19.0304 515.42 L14.479 515.42 L14.479 501.925 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip772)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="294.087,86.1857 314.074,86.1857 334.061,86.1857 354.048,86.1857 374.035,86.1857 394.022,86.1857 414.009,86.1857 433.996,86.1857 453.983,86.1857 473.97,86.1857 493.957,86.1857 513.945,86.1857 533.932,86.1857 553.919,86.1857 573.906,121.282 593.893,283.964 613.88,417.69 633.867,528.946 653.854,622.498 673.841,701.913 693.828,769.901 713.815,828.554 733.802,879.506 753.79,924.049 773.777,963.213 793.764,997.832 813.751,1028.58 833.738,1056.02 853.725,1080.6 873.712,1102.72 893.699,1122.68 913.686,1140.76 933.673,1157.19 953.66,1172.16 973.647,1185.85 993.634,1198.39 1013.62,1209.91 1033.61,1220.51 1053.6,1230.3 1073.58,1239.35 1093.57,1247.73 1113.56,1255.51 1133.54,1262.75 1153.53,1269.49 1173.52,1275.78 1193.51,1281.66 1213.49,1287.17 1233.48,1292.33 1253.47,1297.17 1273.45,1301.72 1293.44,1306 1313.43,1310.03 1333.41,1313.84 1353.4,1317.43 1373.39,1320.83 1393.38,1324.04 1413.36,1327.08 1433.35,1329.97 1453.34,1332.71 1473.32,1335.31 1493.31,1337.78 1513.3,1340.14 1533.29,1342.38 1553.27,1344.52 1573.26,1346.55 1593.25,1348.5 1613.23,1350.35 1633.22,1352.13 1653.21,1353.83 1673.2,1355.45 1693.18,1357.01 1713.17,1358.5 1733.16,1359.93 1753.14,1361.3 1773.13,1362.61 1793.12,1363.88 1813.1,1365.09 1833.09,1366.26 1853.08,1367.38 1873.07,1368.46 1893.05,1369.51 1913.04,1370.51 1933.03,1371.48 1953.01,1372.41 1973,1373.31 1992.99,1374.18 2012.98,1375.01 2032.96,1375.82 2052.95,1376.6 2072.94,1377.36 2092.92,1378.09 2112.91,1378.8 2132.9,1379.48 2152.89,1380.14 2172.87,1380.78 2192.86,1381.41 2212.85,1382.01 2232.83,1382.59 2252.82,1383.16 2272.81,1383.71 2292.79,1384.24 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="294.087,1423.18 294.087,47.2441 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="793.764,1423.18 793.764,47.2441 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1293.44,1423.18 1293.44,47.2441 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1793.12,1423.18 1793.12,47.2441 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2292.79,1423.18 2292.79,47.2441 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="294.087,1423.18 294.087,1404.28 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="793.764,1423.18 793.764,1404.28 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1293.44,1423.18 1293.44,1404.28 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1793.12,1423.18 1793.12,1404.28 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2292.79,1423.18 2292.79,1404.28 "/>
+<path clip-path="url(#clip590)" d="M294.087 1454.1 Q290.476 1454.1 288.647 1457.66 Q286.841 1461.2 286.841 1468.33 Q286.841 1475.44 288.647 1479.01 Q290.476 1482.55 294.087 1482.55 Q297.721 1482.55 299.526 1479.01 Q301.355 1475.44 301.355 1468.33 Q301.355 1461.2 299.526 1457.66 Q297.721 1454.1 294.087 1454.1 M294.087 1450.39 Q299.897 1450.39 302.952 1455 Q306.031 1459.58 306.031 1468.33 Q306.031 1477.06 302.952 1481.67 Q299.897 1486.25 294.087 1486.25 Q288.277 1486.25 285.198 1481.67 Q282.142 1477.06 282.142 1468.33 Q282.142 1459.58 285.198 1455 Q288.277 1450.39 294.087 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M773.035 1481.64 L789.354 1481.64 L789.354 1485.58 L767.41 1485.58 L767.41 1481.64 Q770.072 1478.89 774.655 1474.26 Q779.261 1469.61 780.442 1468.27 Q782.687 1465.74 783.567 1464.01 Q784.47 1462.25 784.47 1460.56 Q784.47 1457.8 782.525 1456.07 Q780.604 1454.33 777.502 1454.33 Q775.303 1454.33 772.849 1455.09 Q770.419 1455.86 767.641 1457.41 L767.641 1452.69 Q770.465 1451.55 772.919 1450.97 Q775.373 1450.39 777.41 1450.39 Q782.78 1450.39 785.974 1453.08 Q789.169 1455.77 789.169 1460.26 Q789.169 1462.39 788.359 1464.31 Q787.572 1466.2 785.465 1468.8 Q784.886 1469.47 781.785 1472.69 Q778.683 1475.88 773.035 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M799.215 1451.02 L817.571 1451.02 L817.571 1454.96 L803.497 1454.96 L803.497 1463.43 Q804.516 1463.08 805.534 1462.92 Q806.553 1462.73 807.571 1462.73 Q813.358 1462.73 816.738 1465.9 Q820.118 1469.08 820.118 1474.49 Q820.118 1480.07 816.646 1483.17 Q813.173 1486.25 806.854 1486.25 Q804.678 1486.25 802.409 1485.88 Q800.164 1485.51 797.757 1484.77 L797.757 1480.07 Q799.84 1481.2 802.062 1481.76 Q804.284 1482.32 806.761 1482.32 Q810.766 1482.32 813.104 1480.21 Q815.442 1478.1 815.442 1474.49 Q815.442 1470.88 813.104 1468.77 Q810.766 1466.67 806.761 1466.67 Q804.886 1466.67 803.011 1467.08 Q801.159 1467.5 799.215 1468.38 L799.215 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1268.14 1451.02 L1286.5 1451.02 L1286.5 1454.96 L1272.42 1454.96 L1272.42 1463.43 Q1273.44 1463.08 1274.46 1462.92 Q1275.48 1462.73 1276.5 1462.73 Q1282.28 1462.73 1285.66 1465.9 Q1289.04 1469.08 1289.04 1474.49 Q1289.04 1480.07 1285.57 1483.17 Q1282.1 1486.25 1275.78 1486.25 Q1273.6 1486.25 1271.33 1485.88 Q1269.09 1485.51 1266.68 1484.77 L1266.68 1480.07 Q1268.76 1481.2 1270.99 1481.76 Q1273.21 1482.32 1275.69 1482.32 Q1279.69 1482.32 1282.03 1480.21 Q1284.37 1478.1 1284.37 1474.49 Q1284.37 1470.88 1282.03 1468.77 Q1279.69 1466.67 1275.69 1466.67 Q1273.81 1466.67 1271.94 1467.08 Q1270.08 1467.5 1268.14 1468.38 L1268.14 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1308.26 1454.1 Q1304.64 1454.1 1302.82 1457.66 Q1301.01 1461.2 1301.01 1468.33 Q1301.01 1475.44 1302.82 1479.01 Q1304.64 1482.55 1308.26 1482.55 Q1311.89 1482.55 1313.7 1479.01 Q1315.52 1475.44 1315.52 1468.33 Q1315.52 1461.2 1313.7 1457.66 Q1311.89 1454.1 1308.26 1454.1 M1308.26 1450.39 Q1314.07 1450.39 1317.12 1455 Q1320.2 1459.58 1320.2 1468.33 Q1320.2 1477.06 1317.12 1481.67 Q1314.07 1486.25 1308.26 1486.25 Q1302.45 1486.25 1299.37 1481.67 Q1296.31 1477.06 1296.31 1468.33 Q1296.31 1459.58 1299.37 1455 Q1302.45 1450.39 1308.26 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1766.97 1451.02 L1789.19 1451.02 L1789.19 1453.01 L1776.65 1485.58 L1771.76 1485.58 L1783.57 1454.96 L1766.97 1454.96 L1766.97 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1798.36 1451.02 L1816.72 1451.02 L1816.72 1454.96 L1802.64 1454.96 L1802.64 1463.43 Q1803.66 1463.08 1804.68 1462.92 Q1805.7 1462.73 1806.72 1462.73 Q1812.5 1462.73 1815.88 1465.9 Q1819.26 1469.08 1819.26 1474.49 Q1819.26 1480.07 1815.79 1483.17 Q1812.32 1486.25 1806 1486.25 Q1803.82 1486.25 1801.56 1485.88 Q1799.31 1485.51 1796.9 1484.77 L1796.9 1480.07 Q1798.99 1481.2 1801.21 1481.76 Q1803.43 1482.32 1805.91 1482.32 Q1809.91 1482.32 1812.25 1480.21 Q1814.59 1478.1 1814.59 1474.49 Q1814.59 1470.88 1812.25 1468.77 Q1809.91 1466.67 1805.91 1466.67 Q1804.03 1466.67 1802.16 1467.08 Q1800.31 1467.5 1798.36 1468.38 L1798.36 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2252.4 1481.64 L2260.04 1481.64 L2260.04 1455.28 L2251.73 1456.95 L2251.73 1452.69 L2259.99 1451.02 L2264.67 1451.02 L2264.67 1481.64 L2272.31 1481.64 L2272.31 1485.58 L2252.4 1485.58 L2252.4 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2291.75 1454.1 Q2288.14 1454.1 2286.31 1457.66 Q2284.51 1461.2 2284.51 1468.33 Q2284.51 1475.44 2286.31 1479.01 Q2288.14 1482.55 2291.75 1482.55 Q2295.39 1482.55 2297.19 1479.01 Q2299.02 1475.44 2299.02 1468.33 Q2299.02 1461.2 2297.19 1457.66 Q2295.39 1454.1 2291.75 1454.1 M2291.75 1450.39 Q2297.56 1450.39 2300.62 1455 Q2303.7 1459.58 2303.7 1468.33 Q2303.7 1477.06 2300.62 1481.67 Q2297.56 1486.25 2291.75 1486.25 Q2285.94 1486.25 2282.86 1481.67 Q2279.81 1477.06 2279.81 1468.33 Q2279.81 1459.58 2282.86 1455 Q2285.94 1450.39 2291.75 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M2321.91 1454.1 Q2318.3 1454.1 2316.48 1457.66 Q2314.67 1461.2 2314.67 1468.33 Q2314.67 1475.44 2316.48 1479.01 Q2318.3 1482.55 2321.91 1482.55 Q2325.55 1482.55 2327.35 1479.01 Q2329.18 1475.44 2329.18 1468.33 Q2329.18 1461.2 2327.35 1457.66 Q2325.55 1454.1 2321.91 1454.1 M2321.91 1450.39 Q2327.73 1450.39 2330.78 1455 Q2333.86 1459.58 2333.86 1468.33 Q2333.86 1477.06 2330.78 1481.67 Q2327.73 1486.25 2321.91 1486.25 Q2316.1 1486.25 2313.03 1481.67 Q2309.97 1477.06 2309.97 1468.33 Q2309.97 1459.58 2313.03 1455 Q2316.1 1450.39 2321.91 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1261.91 1524.18 L1261.91 1530.96 Q1258.67 1527.94 1254.98 1526.44 Q1251.32 1524.95 1247.18 1524.95 Q1239.03 1524.95 1234.7 1529.95 Q1230.37 1534.91 1230.37 1544.33 Q1230.37 1553.72 1234.7 1558.72 Q1239.03 1563.68 1247.18 1563.68 Q1251.32 1563.68 1254.98 1562.19 Q1258.67 1560.69 1261.91 1557.67 L1261.91 1564.38 Q1258.54 1566.68 1254.75 1567.82 Q1251 1568.97 1246.8 1568.97 Q1236.01 1568.97 1229.8 1562.38 Q1223.59 1555.76 1223.59 1544.33 Q1223.59 1532.87 1229.8 1526.28 Q1236.01 1519.66 1246.8 1519.66 Q1251.06 1519.66 1254.82 1520.81 Q1258.6 1521.92 1261.91 1524.18 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1285.4 1536.5 Q1280.69 1536.5 1277.96 1540.19 Q1275.22 1543.85 1275.22 1550.25 Q1275.22 1556.65 1277.92 1560.34 Q1280.66 1564 1285.4 1564 Q1290.08 1564 1292.82 1560.31 Q1295.56 1556.62 1295.56 1550.25 Q1295.56 1543.92 1292.82 1540.23 Q1290.08 1536.5 1285.4 1536.5 M1285.4 1531.54 Q1293.04 1531.54 1297.4 1536.5 Q1301.76 1541.47 1301.76 1550.25 Q1301.76 1559 1297.4 1564 Q1293.04 1568.97 1285.4 1568.97 Q1277.73 1568.97 1273.37 1564 Q1269.04 1559 1269.04 1550.25 Q1269.04 1541.47 1273.37 1536.5 Q1277.73 1531.54 1285.4 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1334.2 1533.45 L1334.2 1538.98 Q1331.71 1537.71 1329.04 1537.07 Q1326.37 1536.44 1323.5 1536.44 Q1319.14 1536.44 1316.95 1537.77 Q1314.78 1539.11 1314.78 1541.79 Q1314.78 1543.82 1316.34 1545 Q1317.9 1546.15 1322.61 1547.2 L1324.62 1547.64 Q1330.86 1548.98 1333.47 1551.43 Q1336.11 1553.85 1336.11 1558.21 Q1336.11 1563.17 1332.16 1566.07 Q1328.25 1568.97 1321.37 1568.97 Q1318.51 1568.97 1315.39 1568.39 Q1312.3 1567.85 1308.86 1566.74 L1308.86 1560.69 Q1312.11 1562.38 1315.26 1563.24 Q1318.41 1564.07 1321.5 1564.07 Q1325.64 1564.07 1327.86 1562.66 Q1330.09 1561.23 1330.09 1558.65 Q1330.09 1556.27 1328.47 1554.99 Q1326.88 1553.72 1321.43 1552.54 L1319.4 1552.07 Q1313.95 1550.92 1311.54 1548.56 Q1309.12 1546.18 1309.12 1542.04 Q1309.12 1537.01 1312.68 1534.27 Q1316.25 1531.54 1322.8 1531.54 Q1326.05 1531.54 1328.91 1532.01 Q1331.78 1532.49 1334.2 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M1351.23 1522.27 L1351.23 1532.4 L1363.29 1532.4 L1363.29 1536.95 L1351.23 1536.95 L1351.23 1556.3 Q1351.23 1560.66 1352.4 1561.9 Q1353.61 1563.14 1357.27 1563.14 L1363.29 1563.14 L1363.29 1568.04 L1357.27 1568.04 Q1350.49 1568.04 1347.92 1565.53 Q1345.34 1562.98 1345.34 1556.3 L1345.34 1536.95 L1341.04 1536.95 L1341.04 1532.4 L1345.34 1532.4 L1345.34 1522.27 L1351.23 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,1410.55 2352.76,1410.55 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,1079.46 2352.76,1079.46 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,748.368 2352.76,748.368 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,417.277 2352.76,417.277 "/>
+<polyline clip-path="url(#clip592)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.125,86.1857 2352.76,86.1857 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1423.18 234.125,47.2441 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1410.55 253.023,1410.55 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,1079.46 253.023,1079.46 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,748.368 253.023,748.368 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,417.277 253.023,417.277 "/>
+<polyline clip-path="url(#clip590)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.125,86.1857 253.023,86.1857 "/>
+<path clip-path="url(#clip590)" d="M186.181 1396.35 Q182.57 1396.35 180.741 1399.91 Q178.936 1403.45 178.936 1410.58 Q178.936 1417.69 180.741 1421.26 Q182.57 1424.8 186.181 1424.8 Q189.815 1424.8 191.621 1421.26 Q193.45 1417.69 193.45 1410.58 Q193.45 1403.45 191.621 1399.91 Q189.815 1396.35 186.181 1396.35 M186.181 1392.64 Q191.991 1392.64 195.047 1397.25 Q198.125 1401.83 198.125 1410.58 Q198.125 1419.31 195.047 1423.92 Q191.991 1428.5 186.181 1428.5 Q180.371 1428.5 177.292 1423.92 Q174.237 1419.31 174.237 1410.58 Q174.237 1401.83 177.292 1397.25 Q180.371 1392.64 186.181 1392.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M146.066 1062.18 L164.422 1062.18 L164.422 1066.11 L150.348 1066.11 L150.348 1074.59 Q151.366 1074.24 152.385 1074.08 Q153.403 1073.89 154.422 1073.89 Q160.209 1073.89 163.589 1077.06 Q166.968 1080.23 166.968 1085.65 Q166.968 1091.23 163.496 1094.33 Q160.024 1097.41 153.704 1097.41 Q151.528 1097.41 149.26 1097.04 Q147.015 1096.67 144.607 1095.93 L144.607 1091.23 Q146.691 1092.36 148.913 1092.92 Q151.135 1093.47 153.612 1093.47 Q157.616 1093.47 159.954 1091.37 Q162.292 1089.26 162.292 1085.65 Q162.292 1082.04 159.954 1079.93 Q157.616 1077.83 153.612 1077.83 Q151.737 1077.83 149.862 1078.24 Q148.01 1078.66 146.066 1079.54 L146.066 1062.18 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M186.181 1065.26 Q182.57 1065.26 180.741 1068.82 Q178.936 1072.36 178.936 1079.49 Q178.936 1086.6 180.741 1090.16 Q182.57 1093.71 186.181 1093.71 Q189.815 1093.71 191.621 1090.16 Q193.45 1086.6 193.45 1079.49 Q193.45 1072.36 191.621 1068.82 Q189.815 1065.26 186.181 1065.26 M186.181 1061.55 Q191.991 1061.55 195.047 1066.16 Q198.125 1070.74 198.125 1079.49 Q198.125 1088.22 195.047 1092.83 Q191.991 1097.41 186.181 1097.41 Q180.371 1097.41 177.292 1092.83 Q174.237 1088.22 174.237 1079.49 Q174.237 1070.74 177.292 1066.16 Q180.371 1061.55 186.181 1061.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M116.668 761.713 L124.306 761.713 L124.306 735.347 L115.996 737.014 L115.996 732.754 L124.26 731.088 L128.936 731.088 L128.936 761.713 L136.575 761.713 L136.575 765.648 L116.668 765.648 L116.668 761.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M156.019 734.166 Q152.408 734.166 150.579 737.731 Q148.774 741.273 148.774 748.402 Q148.774 755.509 150.579 759.074 Q152.408 762.615 156.019 762.615 Q159.653 762.615 161.459 759.074 Q163.288 755.509 163.288 748.402 Q163.288 741.273 161.459 737.731 Q159.653 734.166 156.019 734.166 M156.019 730.463 Q161.829 730.463 164.885 735.069 Q167.964 739.653 167.964 748.402 Q167.964 757.129 164.885 761.736 Q161.829 766.319 156.019 766.319 Q150.209 766.319 147.13 761.736 Q144.075 757.129 144.075 748.402 Q144.075 739.653 147.13 735.069 Q150.209 730.463 156.019 730.463 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M186.181 734.166 Q182.57 734.166 180.741 737.731 Q178.936 741.273 178.936 748.402 Q178.936 755.509 180.741 759.074 Q182.57 762.615 186.181 762.615 Q189.815 762.615 191.621 759.074 Q193.45 755.509 193.45 748.402 Q193.45 741.273 191.621 737.731 Q189.815 734.166 186.181 734.166 M186.181 730.463 Q191.991 730.463 195.047 735.069 Q198.125 739.653 198.125 748.402 Q198.125 757.129 195.047 761.736 Q191.991 766.319 186.181 766.319 Q180.371 766.319 177.292 761.736 Q174.237 757.129 174.237 748.402 Q174.237 739.653 177.292 735.069 Q180.371 730.463 186.181 730.463 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M116.668 430.622 L124.306 430.622 L124.306 404.256 L115.996 405.923 L115.996 401.663 L124.26 399.997 L128.936 399.997 L128.936 430.622 L136.575 430.622 L136.575 434.557 L116.668 434.557 L116.668 430.622 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M146.066 399.997 L164.422 399.997 L164.422 403.932 L150.348 403.932 L150.348 412.404 Q151.366 412.057 152.385 411.895 Q153.403 411.71 154.422 411.71 Q160.209 411.71 163.589 414.881 Q166.968 418.052 166.968 423.469 Q166.968 429.047 163.496 432.149 Q160.024 435.228 153.704 435.228 Q151.528 435.228 149.26 434.858 Q147.015 434.487 144.607 433.747 L144.607 429.047 Q146.691 430.182 148.913 430.737 Q151.135 431.293 153.612 431.293 Q157.616 431.293 159.954 429.186 Q162.292 427.08 162.292 423.469 Q162.292 419.858 159.954 417.751 Q157.616 415.645 153.612 415.645 Q151.737 415.645 149.862 416.061 Q148.01 416.478 146.066 417.358 L146.066 399.997 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M186.181 403.075 Q182.57 403.075 180.741 406.64 Q178.936 410.182 178.936 417.311 Q178.936 424.418 180.741 427.983 Q182.57 431.524 186.181 431.524 Q189.815 431.524 191.621 427.983 Q193.45 424.418 193.45 417.311 Q193.45 410.182 191.621 406.64 Q189.815 403.075 186.181 403.075 M186.181 399.372 Q191.991 399.372 195.047 403.978 Q198.125 408.561 198.125 417.311 Q198.125 426.038 195.047 430.645 Q191.991 435.228 186.181 435.228 Q180.371 435.228 177.292 430.645 Q174.237 426.038 174.237 417.311 Q174.237 408.561 177.292 403.978 Q180.371 399.372 186.181 399.372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M119.885 99.5305 L136.204 99.5305 L136.204 103.466 L114.26 103.466 L114.26 99.5305 Q116.922 96.7759 121.505 92.1463 Q126.112 87.4935 127.292 86.151 Q129.538 83.6278 130.417 81.8917 Q131.32 80.1325 131.32 78.4427 Q131.32 75.6881 129.376 73.952 Q127.455 72.2159 124.353 72.2159 Q122.154 72.2159 119.7 72.9797 Q117.269 73.7436 114.492 75.2945 L114.492 70.5723 Q117.316 69.4381 119.769 68.8594 Q122.223 68.2807 124.26 68.2807 Q129.63 68.2807 132.825 70.9659 Q136.019 73.651 136.019 78.1417 Q136.019 80.2714 135.209 82.1927 Q134.422 84.0908 132.316 86.6834 Q131.737 87.3547 128.635 90.5722 Q125.533 93.7667 119.885 99.5305 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M156.019 71.9844 Q152.408 71.9844 150.579 75.5492 Q148.774 79.0908 148.774 86.2204 Q148.774 93.3269 150.579 96.8916 Q152.408 100.433 156.019 100.433 Q159.653 100.433 161.459 96.8916 Q163.288 93.3269 163.288 86.2204 Q163.288 79.0908 161.459 75.5492 Q159.653 71.9844 156.019 71.9844 M156.019 68.2807 Q161.829 68.2807 164.885 72.8871 Q167.964 77.4705 167.964 86.2204 Q167.964 94.9472 164.885 99.5537 Q161.829 104.137 156.019 104.137 Q150.209 104.137 147.13 99.5537 Q144.075 94.9472 144.075 86.2204 Q144.075 77.4705 147.13 72.8871 Q150.209 68.2807 156.019 68.2807 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M186.181 71.9844 Q182.57 71.9844 180.741 75.5492 Q178.936 79.0908 178.936 86.2204 Q178.936 93.3269 180.741 96.8916 Q182.57 100.433 186.181 100.433 Q189.815 100.433 191.621 96.8916 Q193.45 93.3269 193.45 86.2204 Q193.45 79.0908 191.621 75.5492 Q189.815 71.9844 186.181 71.9844 M186.181 68.2807 Q191.991 68.2807 195.047 72.8871 Q198.125 77.4705 198.125 86.2204 Q198.125 94.9472 195.047 99.5537 Q191.991 104.137 186.181 104.137 Q180.371 104.137 177.292 99.5537 Q174.237 94.9472 174.237 86.2204 Q174.237 77.4705 177.292 72.8871 Q180.371 68.2807 186.181 68.2807 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M21.7677 962.07 L58.7206 962.07 L58.7206 954.304 Q58.7206 944.469 54.2646 939.917 Q49.8086 935.334 40.1964 935.334 Q30.6479 935.334 26.2237 939.917 Q21.7677 944.469 21.7677 954.304 L21.7677 962.07 M16.4842 968.499 L16.4842 955.29 Q16.4842 941.477 22.2451 935.016 Q27.9743 928.554 40.1964 928.554 Q52.4822 928.554 58.2432 935.047 Q64.0042 941.54 64.0042 955.29 L64.0042 968.499 L16.4842 968.499 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M28.3562 918.56 L28.3562 912.704 L64.0042 912.704 L64.0042 918.56 L28.3562 918.56 M14.479 918.56 L14.479 912.704 L21.895 912.704 L21.895 918.56 L14.479 918.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M29.4065 877.724 L34.9447 877.724 Q33.6716 880.207 33.035 882.88 Q32.3984 885.554 32.3984 888.419 Q32.3984 892.779 33.7352 894.975 Q35.072 897.14 37.7456 897.14 Q39.7826 897.14 40.9603 895.58 Q42.1061 894.02 43.1565 889.31 L43.6021 887.305 Q44.9389 881.066 47.3897 878.456 Q49.8086 875.815 54.1691 875.815 Q59.1344 875.815 62.0308 879.761 Q64.9272 883.676 64.9272 890.551 Q64.9272 893.416 64.3543 896.535 Q63.8132 899.622 62.6992 903.06 L56.6518 903.06 Q58.3387 899.813 59.198 896.662 Q60.0256 893.511 60.0256 890.424 Q60.0256 886.286 58.6251 884.058 Q57.1929 881.83 54.6147 881.83 Q52.2276 881.83 50.9545 883.453 Q49.6813 885.045 48.5037 890.487 L48.0262 892.524 Q46.8804 897.967 44.5251 900.386 Q42.138 902.805 38.0002 902.805 Q32.9713 902.805 30.2341 899.24 Q27.4968 895.676 27.4968 889.119 Q27.4968 885.872 27.9743 883.008 Q28.4517 880.143 29.4065 877.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M58.657 860.823 L77.5631 860.823 L77.5631 866.712 L28.3562 866.712 L28.3562 860.823 L33.7671 860.823 Q30.5842 858.977 29.0564 856.176 Q27.4968 853.344 27.4968 849.429 Q27.4968 842.936 32.6531 838.893 Q37.8093 834.819 46.212 834.819 Q54.6147 834.819 59.771 838.893 Q64.9272 842.936 64.9272 849.429 Q64.9272 853.344 63.3994 856.176 Q61.8398 858.977 58.657 860.823 M46.212 840.899 Q39.7508 840.899 36.0905 843.572 Q32.3984 846.214 32.3984 850.861 Q32.3984 855.508 36.0905 858.182 Q39.7508 860.823 46.212 860.823 Q52.6732 860.823 56.3653 858.182 Q60.0256 855.508 60.0256 850.861 Q60.0256 846.214 56.3653 843.572 Q52.6732 840.899 46.212 840.899 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M46.0847 808.911 Q46.0847 816.009 47.7079 818.746 Q49.3312 821.483 53.2461 821.483 Q56.3653 821.483 58.2114 819.446 Q60.0256 817.377 60.0256 813.844 Q60.0256 808.975 56.5881 806.046 Q53.1188 803.086 47.3897 803.086 L46.0847 803.086 L46.0847 808.911 M43.6657 797.23 L64.0042 797.23 L64.0042 803.086 L58.5933 803.086 Q61.8398 805.092 63.3994 808.083 Q64.9272 811.075 64.9272 815.404 Q64.9272 820.878 61.8716 824.125 Q58.7843 827.34 53.6281 827.34 Q47.6125 827.34 44.5569 823.329 Q41.5014 819.287 41.5014 811.298 L41.5014 803.086 L40.9285 803.086 Q36.8862 803.086 34.6901 805.76 Q32.4621 808.402 32.4621 813.208 Q32.4621 816.263 33.1941 819.16 Q33.9262 822.056 35.3903 824.73 L29.9795 824.73 Q28.7381 821.515 28.1334 818.491 Q27.4968 815.468 27.4968 812.603 Q27.4968 804.869 31.5072 801.049 Q35.5176 797.23 43.6657 797.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M18.2347 779.374 L28.3562 779.374 L28.3562 767.311 L32.9077 767.311 L32.9077 779.374 L52.2594 779.374 Q56.6199 779.374 57.8613 778.196 Q59.1026 776.987 59.1026 773.327 L59.1026 767.311 L64.0042 767.311 L64.0042 773.327 Q64.0042 780.106 61.4897 782.684 Q58.9434 785.262 52.2594 785.262 L32.9077 785.262 L32.9077 789.559 L28.3562 789.559 L28.3562 785.262 L18.2347 785.262 L18.2347 779.374 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M29.7248 733.955 L35.1993 733.955 Q33.8307 736.437 33.1623 738.952 Q32.4621 741.434 32.4621 743.981 Q32.4621 749.678 36.0905 752.829 Q39.6872 755.98 46.212 755.98 Q52.7369 755.98 56.3653 752.829 Q59.9619 749.678 59.9619 743.981 Q59.9619 741.434 59.2935 738.952 Q58.5933 736.437 57.2247 733.955 L62.6355 733.955 Q63.7814 736.406 64.3543 739.047 Q64.9272 741.657 64.9272 744.617 Q64.9272 752.67 59.8664 757.412 Q54.8057 762.155 46.212 762.155 Q37.491 762.155 32.4939 757.381 Q27.4968 752.574 27.4968 744.235 Q27.4968 741.53 28.0697 738.952 Q28.6108 736.374 29.7248 733.955 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M42.4881 694.137 L64.0042 694.137 L64.0042 699.994 L42.679 699.994 Q37.6183 699.994 35.1038 701.967 Q32.5894 703.94 32.5894 707.887 Q32.5894 712.63 35.6131 715.367 Q38.6368 718.104 43.8567 718.104 L64.0042 718.104 L64.0042 723.992 L14.479 723.992 L14.479 718.104 L33.8944 718.104 Q30.6797 716.003 29.0883 713.171 Q27.4968 710.306 27.4968 706.582 Q27.4968 700.439 31.3163 697.288 Q35.1038 694.137 42.4881 694.137 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M14.479 662.277 L14.479 648.782 L19.0304 648.782 L19.0304 656.42 L68.0464 656.42 L68.0464 648.782 L72.5979 648.782 L72.5979 662.277 L14.479 662.277 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M16.4842 636.05 L16.4842 626.47 L48.8219 614.343 L16.4842 602.153 L16.4842 592.572 L64.0042 592.572 L64.0042 598.843 L22.277 598.843 L54.8694 611.097 L54.8694 617.558 L22.277 629.812 L64.0042 629.812 L64.0042 636.05 L16.4842 636.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M16.4842 584.042 L16.4842 577.549 L56.6518 567.555 L16.4842 557.593 L16.4842 550.368 L56.6518 540.374 L16.4842 530.411 L16.4842 523.886 L64.0042 535.822 L64.0042 543.907 L22.7544 553.933 L64.0042 564.054 L64.0042 572.138 L16.4842 584.042 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip590)" d="M14.479 501.925 L72.5979 501.925 L72.5979 515.42 L68.0464 515.42 L68.0464 507.813 L19.0304 507.813 L19.0304 515.42 L14.479 515.42 L14.479 501.925 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip592)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="294.087,86.1857 314.074,86.1857 334.061,86.1857 354.048,86.1857 374.035,86.1857 394.022,86.1857 414.009,86.1857 433.996,86.1857 453.983,86.1857 473.97,86.1857 493.957,86.1857 513.945,86.1857 533.932,86.1857 553.919,86.1857 573.906,121.282 593.893,283.964 613.88,417.69 633.867,528.946 653.854,622.498 673.841,701.913 693.828,769.901 713.815,828.554 733.802,879.506 753.79,924.049 773.777,963.213 793.764,997.832 813.751,1028.58 833.738,1056.02 853.725,1080.6 873.712,1102.72 893.699,1122.68 913.686,1140.76 933.673,1157.19 953.66,1172.16 973.647,1185.85 993.634,1198.39 1013.62,1209.91 1033.61,1220.51 1053.6,1230.3 1073.58,1239.35 1093.57,1247.73 1113.56,1255.51 1133.54,1262.75 1153.53,1269.49 1173.52,1275.78 1193.51,1281.66 1213.49,1287.17 1233.48,1292.33 1253.47,1297.17 1273.45,1301.72 1293.44,1306 1313.43,1310.03 1333.41,1313.84 1353.4,1317.43 1373.39,1320.83 1393.38,1324.04 1413.36,1327.08 1433.35,1329.97 1453.34,1332.71 1473.32,1335.31 1493.31,1337.78 1513.3,1340.14 1533.29,1342.38 1553.27,1344.52 1573.26,1346.55 1593.25,1348.5 1613.23,1350.35 1633.22,1352.13 1653.21,1353.83 1673.2,1355.45 1693.18,1357.01 1713.17,1358.5 1733.16,1359.93 1753.14,1361.3 1773.13,1362.61 1793.12,1363.88 1813.1,1365.09 1833.09,1366.26 1853.08,1367.38 1873.07,1368.46 1893.05,1369.51 1913.04,1370.51 1933.03,1371.48 1953.01,1372.41 1973,1373.31 1992.99,1374.18 2012.98,1375.01 2032.96,1375.82 2052.95,1376.6 2072.94,1377.36 2092.92,1378.09 2112.91,1378.8 2132.9,1379.48 2152.89,1380.14 2172.87,1380.78 2192.86,1381.41 2212.85,1382.01 2232.83,1382.59 2252.82,1383.16 2272.81,1383.71 2292.79,1384.24 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/applications/power_systems/472a82a6.svg b/previews/PR3562/tutorials/applications/power_systems/d9e67996.svg
similarity index 86%
rename from previews/PR3562/tutorials/applications/power_systems/472a82a6.svg
rename to previews/PR3562/tutorials/applications/power_systems/d9e67996.svg
index 529042a552f..e4e13b89c48 100644
--- a/previews/PR3562/tutorials/applications/power_systems/472a82a6.svg
+++ b/previews/PR3562/tutorials/applications/power_systems/d9e67996.svg
@@ -1,78 +1,78 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip650">
+  <clipPath id="clip470">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip650)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip470)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip651">
+  <clipPath id="clip471">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip650)" d="M262.551 1423.18 L1166.97 1423.18 L1166.97 123.472 L262.551 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip470)" d="M262.551 1423.18 L1166.97 1423.18 L1166.97 123.472 L262.551 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip652">
+  <clipPath id="clip472">
     <rect x="262" y="123" width="905" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="382.951,1423.18 382.951,123.472 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="619.957,1423.18 619.957,123.472 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="856.964,1423.18 856.964,123.472 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1093.97,1423.18 1093.97,123.472 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1423.18 1166.97,1423.18 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="382.951,1423.18 382.951,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="619.957,1423.18 619.957,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="856.964,1423.18 856.964,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1093.97,1423.18 1093.97,1404.28 "/>
-<path clip-path="url(#clip650)" d="M342.569 1451.02 L360.925 1451.02 L360.925 1454.96 L346.851 1454.96 L346.851 1463.43 Q347.87 1463.08 348.888 1462.92 Q349.907 1462.73 350.925 1462.73 Q356.712 1462.73 360.092 1465.9 Q363.472 1469.08 363.472 1474.49 Q363.472 1480.07 359.999 1483.17 Q356.527 1486.25 350.208 1486.25 Q348.032 1486.25 345.763 1485.88 Q343.518 1485.51 341.111 1484.77 L341.111 1480.07 Q343.194 1481.2 345.416 1481.76 Q347.638 1482.32 350.115 1482.32 Q354.12 1482.32 356.458 1480.21 Q358.796 1478.1 358.796 1474.49 Q358.796 1470.88 356.458 1468.77 Q354.12 1466.67 350.115 1466.67 Q348.24 1466.67 346.365 1467.08 Q344.513 1467.5 342.569 1468.38 L342.569 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M382.684 1454.1 Q379.073 1454.1 377.245 1457.66 Q375.439 1461.2 375.439 1468.33 Q375.439 1475.44 377.245 1479.01 Q379.073 1482.55 382.684 1482.55 Q386.319 1482.55 388.124 1479.01 Q389.953 1475.44 389.953 1468.33 Q389.953 1461.2 388.124 1457.66 Q386.319 1454.1 382.684 1454.1 M382.684 1450.39 Q388.495 1450.39 391.55 1455 Q394.629 1459.58 394.629 1468.33 Q394.629 1477.06 391.55 1481.67 Q388.495 1486.25 382.684 1486.25 Q376.874 1486.25 373.796 1481.67 Q370.74 1477.06 370.74 1468.33 Q370.74 1459.58 373.796 1455 Q376.874 1450.39 382.684 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M412.846 1454.1 Q409.235 1454.1 407.407 1457.66 Q405.601 1461.2 405.601 1468.33 Q405.601 1475.44 407.407 1479.01 Q409.235 1482.55 412.846 1482.55 Q416.481 1482.55 418.286 1479.01 Q420.115 1475.44 420.115 1468.33 Q420.115 1461.2 418.286 1457.66 Q416.481 1454.1 412.846 1454.1 M412.846 1450.39 Q418.656 1450.39 421.712 1455 Q424.791 1459.58 424.791 1468.33 Q424.791 1477.06 421.712 1481.67 Q418.656 1486.25 412.846 1486.25 Q407.036 1486.25 403.957 1481.67 Q400.902 1477.06 400.902 1468.33 Q400.902 1459.58 403.957 1455 Q407.036 1450.39 412.846 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M564.483 1481.64 L572.122 1481.64 L572.122 1455.28 L563.812 1456.95 L563.812 1452.69 L572.076 1451.02 L576.752 1451.02 L576.752 1481.64 L584.39 1481.64 L584.39 1485.58 L564.483 1485.58 L564.483 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M603.835 1454.1 Q600.224 1454.1 598.395 1457.66 Q596.589 1461.2 596.589 1468.33 Q596.589 1475.44 598.395 1479.01 Q600.224 1482.55 603.835 1482.55 Q607.469 1482.55 609.275 1479.01 Q611.103 1475.44 611.103 1468.33 Q611.103 1461.2 609.275 1457.66 Q607.469 1454.1 603.835 1454.1 M603.835 1450.39 Q609.645 1450.39 612.7 1455 Q615.779 1459.58 615.779 1468.33 Q615.779 1477.06 612.7 1481.67 Q609.645 1486.25 603.835 1486.25 Q598.025 1486.25 594.946 1481.67 Q591.89 1477.06 591.89 1468.33 Q591.89 1459.58 594.946 1455 Q598.025 1450.39 603.835 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M633.997 1454.1 Q630.386 1454.1 628.557 1457.66 Q626.751 1461.2 626.751 1468.33 Q626.751 1475.44 628.557 1479.01 Q630.386 1482.55 633.997 1482.55 Q637.631 1482.55 639.436 1479.01 Q641.265 1475.44 641.265 1468.33 Q641.265 1461.2 639.436 1457.66 Q637.631 1454.1 633.997 1454.1 M633.997 1450.39 Q639.807 1450.39 642.862 1455 Q645.941 1459.58 645.941 1468.33 Q645.941 1477.06 642.862 1481.67 Q639.807 1486.25 633.997 1486.25 Q628.186 1486.25 625.108 1481.67 Q622.052 1477.06 622.052 1468.33 Q622.052 1459.58 625.108 1455 Q628.186 1450.39 633.997 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M664.158 1454.1 Q660.547 1454.1 658.719 1457.66 Q656.913 1461.2 656.913 1468.33 Q656.913 1475.44 658.719 1479.01 Q660.547 1482.55 664.158 1482.55 Q667.793 1482.55 669.598 1479.01 Q671.427 1475.44 671.427 1468.33 Q671.427 1461.2 669.598 1457.66 Q667.793 1454.1 664.158 1454.1 M664.158 1450.39 Q669.969 1450.39 673.024 1455 Q676.103 1459.58 676.103 1468.33 Q676.103 1477.06 673.024 1481.67 Q669.969 1486.25 664.158 1486.25 Q658.348 1486.25 655.27 1481.67 Q652.214 1477.06 652.214 1468.33 Q652.214 1459.58 655.27 1455 Q658.348 1450.39 664.158 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M801.49 1481.64 L809.129 1481.64 L809.129 1455.28 L800.818 1456.95 L800.818 1452.69 L809.082 1451.02 L813.758 1451.02 L813.758 1481.64 L821.397 1481.64 L821.397 1485.58 L801.49 1485.58 L801.49 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M830.888 1451.02 L849.244 1451.02 L849.244 1454.96 L835.17 1454.96 L835.17 1463.43 Q836.189 1463.08 837.207 1462.92 Q838.226 1462.73 839.244 1462.73 Q845.031 1462.73 848.411 1465.9 Q851.79 1469.08 851.79 1474.49 Q851.79 1480.07 848.318 1483.17 Q844.846 1486.25 838.527 1486.25 Q836.351 1486.25 834.082 1485.88 Q831.837 1485.51 829.429 1484.77 L829.429 1480.07 Q831.513 1481.2 833.735 1481.76 Q835.957 1482.32 838.434 1482.32 Q842.439 1482.32 844.777 1480.21 Q847.115 1478.1 847.115 1474.49 Q847.115 1470.88 844.777 1468.77 Q842.439 1466.67 838.434 1466.67 Q836.559 1466.67 834.684 1467.08 Q832.832 1467.5 830.888 1468.38 L830.888 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M871.003 1454.1 Q867.392 1454.1 865.564 1457.66 Q863.758 1461.2 863.758 1468.33 Q863.758 1475.44 865.564 1479.01 Q867.392 1482.55 871.003 1482.55 Q874.638 1482.55 876.443 1479.01 Q878.272 1475.44 878.272 1468.33 Q878.272 1461.2 876.443 1457.66 Q874.638 1454.1 871.003 1454.1 M871.003 1450.39 Q876.813 1450.39 879.869 1455 Q882.948 1459.58 882.948 1468.33 Q882.948 1477.06 879.869 1481.67 Q876.813 1486.25 871.003 1486.25 Q865.193 1486.25 862.114 1481.67 Q859.059 1477.06 859.059 1468.33 Q859.059 1459.58 862.114 1455 Q865.193 1450.39 871.003 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M901.165 1454.1 Q897.554 1454.1 895.725 1457.66 Q893.92 1461.2 893.92 1468.33 Q893.92 1475.44 895.725 1479.01 Q897.554 1482.55 901.165 1482.55 Q904.799 1482.55 906.605 1479.01 Q908.434 1475.44 908.434 1468.33 Q908.434 1461.2 906.605 1457.66 Q904.799 1454.1 901.165 1454.1 M901.165 1450.39 Q906.975 1450.39 910.031 1455 Q913.11 1459.58 913.11 1468.33 Q913.11 1477.06 910.031 1481.67 Q906.975 1486.25 901.165 1486.25 Q895.355 1486.25 892.276 1481.67 Q889.221 1477.06 889.221 1468.33 Q889.221 1459.58 892.276 1455 Q895.355 1450.39 901.165 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1042.58 1481.64 L1058.9 1481.64 L1058.9 1485.58 L1036.96 1485.58 L1036.96 1481.64 Q1039.62 1478.89 1044.2 1474.26 Q1048.81 1469.61 1049.99 1468.27 Q1052.23 1465.74 1053.11 1464.01 Q1054.02 1462.25 1054.02 1460.56 Q1054.02 1457.8 1052.07 1456.07 Q1050.15 1454.33 1047.05 1454.33 Q1044.85 1454.33 1042.4 1455.09 Q1039.97 1455.86 1037.19 1457.41 L1037.19 1452.69 Q1040.01 1451.55 1042.47 1450.97 Q1044.92 1450.39 1046.96 1450.39 Q1052.33 1450.39 1055.52 1453.08 Q1058.72 1455.77 1058.72 1460.26 Q1058.72 1462.39 1057.91 1464.31 Q1057.12 1466.2 1055.01 1468.8 Q1054.43 1469.47 1051.33 1472.69 Q1048.23 1475.88 1042.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1078.72 1454.1 Q1075.11 1454.1 1073.28 1457.66 Q1071.47 1461.2 1071.47 1468.33 Q1071.47 1475.44 1073.28 1479.01 Q1075.11 1482.55 1078.72 1482.55 Q1082.35 1482.55 1084.16 1479.01 Q1085.98 1475.44 1085.98 1468.33 Q1085.98 1461.2 1084.16 1457.66 Q1082.35 1454.1 1078.72 1454.1 M1078.72 1450.39 Q1084.53 1450.39 1087.58 1455 Q1090.66 1459.58 1090.66 1468.33 Q1090.66 1477.06 1087.58 1481.67 Q1084.53 1486.25 1078.72 1486.25 Q1072.91 1486.25 1069.83 1481.67 Q1066.77 1477.06 1066.77 1468.33 Q1066.77 1459.58 1069.83 1455 Q1072.91 1450.39 1078.72 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1108.88 1454.1 Q1105.27 1454.1 1103.44 1457.66 Q1101.63 1461.2 1101.63 1468.33 Q1101.63 1475.44 1103.44 1479.01 Q1105.27 1482.55 1108.88 1482.55 Q1112.51 1482.55 1114.32 1479.01 Q1116.15 1475.44 1116.15 1468.33 Q1116.15 1461.2 1114.32 1457.66 Q1112.51 1454.1 1108.88 1454.1 M1108.88 1450.39 Q1114.69 1450.39 1117.74 1455 Q1120.82 1459.58 1120.82 1468.33 Q1120.82 1477.06 1117.74 1481.67 Q1114.69 1486.25 1108.88 1486.25 Q1103.07 1486.25 1099.99 1481.67 Q1096.93 1477.06 1096.93 1468.33 Q1096.93 1459.58 1099.99 1455 Q1103.07 1450.39 1108.88 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1139.04 1454.1 Q1135.43 1454.1 1133.6 1457.66 Q1131.79 1461.2 1131.79 1468.33 Q1131.79 1475.44 1133.6 1479.01 Q1135.43 1482.55 1139.04 1482.55 Q1142.67 1482.55 1144.48 1479.01 Q1146.31 1475.44 1146.31 1468.33 Q1146.31 1461.2 1144.48 1457.66 Q1142.67 1454.1 1139.04 1454.1 M1139.04 1450.39 Q1144.85 1450.39 1147.91 1455 Q1150.98 1459.58 1150.98 1468.33 Q1150.98 1477.06 1147.91 1481.67 Q1144.85 1486.25 1139.04 1486.25 Q1133.23 1486.25 1130.15 1481.67 Q1127.1 1477.06 1127.1 1468.33 Q1127.1 1459.58 1130.15 1455 Q1133.23 1450.39 1139.04 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M589.133 1525.81 L589.133 1562.76 L596.899 1562.76 Q606.734 1562.76 611.285 1558.3 Q615.869 1553.85 615.869 1544.24 Q615.869 1534.69 611.285 1530.26 Q606.734 1525.81 596.899 1525.81 L589.133 1525.81 M582.703 1520.52 L595.912 1520.52 Q609.726 1520.52 616.187 1526.28 Q622.648 1532.01 622.648 1544.24 Q622.648 1556.52 616.155 1562.28 Q609.662 1568.04 595.912 1568.04 L582.703 1568.04 L582.703 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M663.134 1548.76 L663.134 1551.62 L636.207 1551.62 Q636.589 1557.67 639.836 1560.85 Q643.114 1564 648.939 1564 Q652.312 1564 655.463 1563.17 Q658.646 1562.35 661.765 1560.69 L661.765 1566.23 Q658.614 1567.57 655.304 1568.27 Q651.994 1568.97 648.588 1568.97 Q640.058 1568.97 635.061 1564 Q630.096 1559.04 630.096 1550.57 Q630.096 1541.82 634.807 1536.69 Q639.549 1531.54 647.57 1531.54 Q654.763 1531.54 658.933 1536.18 Q663.134 1540.8 663.134 1548.76 M657.278 1547.04 Q657.214 1542.23 654.572 1539.37 Q651.962 1536.5 647.634 1536.5 Q642.732 1536.5 639.772 1539.27 Q636.844 1542.04 636.398 1547.07 L657.278 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M700.501 1539.24 Q702.697 1535.29 705.753 1533.41 Q708.808 1531.54 712.946 1531.54 Q718.516 1531.54 721.539 1535.45 Q724.563 1539.33 724.563 1546.53 L724.563 1568.04 L718.675 1568.04 L718.675 1546.72 Q718.675 1541.59 716.861 1539.11 Q715.046 1536.63 711.323 1536.63 Q706.771 1536.63 704.129 1539.65 Q701.488 1542.68 701.488 1547.9 L701.488 1568.04 L695.599 1568.04 L695.599 1546.72 Q695.599 1541.56 693.785 1539.11 Q691.971 1536.63 688.183 1536.63 Q683.695 1536.63 681.054 1539.68 Q678.412 1542.71 678.412 1547.9 L678.412 1568.04 L672.524 1568.04 L672.524 1532.4 L678.412 1532.4 L678.412 1537.93 Q680.417 1534.66 683.218 1533.1 Q686.019 1531.54 689.87 1531.54 Q693.753 1531.54 696.459 1533.51 Q699.196 1535.48 700.501 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M752.445 1550.12 Q745.347 1550.12 742.61 1551.75 Q739.873 1553.37 739.873 1557.29 Q739.873 1560.4 741.91 1562.25 Q743.979 1564.07 747.512 1564.07 Q752.381 1564.07 755.31 1560.63 Q758.27 1557.16 758.27 1551.43 L758.27 1550.12 L752.445 1550.12 M764.126 1547.71 L764.126 1568.04 L758.27 1568.04 L758.27 1562.63 Q756.264 1565.88 753.273 1567.44 Q750.281 1568.97 745.952 1568.97 Q740.477 1568.97 737.231 1565.91 Q734.016 1562.82 734.016 1557.67 Q734.016 1551.65 738.027 1548.6 Q742.069 1545.54 750.058 1545.54 L758.27 1545.54 L758.27 1544.97 Q758.27 1540.93 755.596 1538.73 Q752.954 1536.5 748.148 1536.5 Q745.093 1536.5 742.196 1537.23 Q739.3 1537.97 736.626 1539.43 L736.626 1534.02 Q739.841 1532.78 742.865 1532.17 Q745.888 1531.54 748.753 1531.54 Q756.487 1531.54 760.307 1535.55 Q764.126 1539.56 764.126 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M805.821 1546.53 L805.821 1568.04 L799.965 1568.04 L799.965 1546.72 Q799.965 1541.66 797.992 1539.14 Q796.018 1536.63 792.072 1536.63 Q787.329 1536.63 784.592 1539.65 Q781.855 1542.68 781.855 1547.9 L781.855 1568.04 L775.966 1568.04 L775.966 1532.4 L781.855 1532.4 L781.855 1537.93 Q783.955 1534.72 786.788 1533.13 Q789.653 1531.54 793.376 1531.54 Q799.519 1531.54 802.67 1535.36 Q805.821 1539.14 805.821 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M840.96 1537.81 L840.96 1518.52 L846.817 1518.52 L846.817 1568.04 L840.96 1568.04 L840.96 1562.7 Q839.114 1565.88 836.281 1567.44 Q833.48 1568.97 829.534 1568.97 Q823.073 1568.97 818.998 1563.81 Q814.956 1558.65 814.956 1550.25 Q814.956 1541.85 818.998 1536.69 Q823.073 1531.54 829.534 1531.54 Q833.48 1531.54 836.281 1533.1 Q839.114 1534.62 840.96 1537.81 M821.004 1550.25 Q821.004 1556.71 823.645 1560.4 Q826.319 1564.07 830.966 1564.07 Q835.613 1564.07 838.287 1560.4 Q840.96 1556.71 840.96 1550.25 Q840.96 1543.79 838.287 1540.13 Q835.613 1536.44 830.966 1536.44 Q826.319 1536.44 823.645 1540.13 Q821.004 1543.79 821.004 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,1386.4 1166.97,1386.4 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,1079.86 1166.97,1079.86 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,773.326 1166.97,773.326 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,466.791 1166.97,466.791 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,160.256 1166.97,160.256 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1423.18 262.551,123.472 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1386.4 281.449,1386.4 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1079.86 281.449,1079.86 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,773.326 281.449,773.326 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,466.791 281.449,466.791 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,160.256 281.449,160.256 "/>
-<path clip-path="url(#clip650)" d="M214.607 1372.19 Q210.996 1372.19 209.167 1375.76 Q207.362 1379.3 207.362 1386.43 Q207.362 1393.54 209.167 1397.1 Q210.996 1400.64 214.607 1400.64 Q218.241 1400.64 220.047 1397.1 Q221.875 1393.54 221.875 1386.43 Q221.875 1379.3 220.047 1375.76 Q218.241 1372.19 214.607 1372.19 M214.607 1368.49 Q220.417 1368.49 223.473 1373.1 Q226.551 1377.68 226.551 1386.43 Q226.551 1395.16 223.473 1399.76 Q220.417 1404.35 214.607 1404.35 Q208.797 1404.35 205.718 1399.76 Q202.662 1395.16 202.662 1386.43 Q202.662 1377.68 205.718 1373.1 Q208.797 1368.49 214.607 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M148.311 1093.21 L164.63 1093.21 L164.63 1097.14 L142.686 1097.14 L142.686 1093.21 Q145.348 1090.45 149.931 1085.82 Q154.538 1081.17 155.718 1079.83 Q157.964 1077.3 158.843 1075.57 Q159.746 1073.81 159.746 1072.12 Q159.746 1069.36 157.802 1067.63 Q155.88 1065.89 152.778 1065.89 Q150.579 1065.89 148.126 1066.65 Q145.695 1067.42 142.917 1068.97 L142.917 1064.25 Q145.741 1063.11 148.195 1062.53 Q150.649 1061.96 152.686 1061.96 Q158.056 1061.96 161.251 1064.64 Q164.445 1067.33 164.445 1071.82 Q164.445 1073.95 163.635 1075.87 Q162.848 1077.77 160.741 1080.36 Q160.163 1081.03 157.061 1084.25 Q153.959 1087.44 148.311 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M174.491 1062.58 L192.848 1062.58 L192.848 1066.52 L178.774 1066.52 L178.774 1074.99 Q179.792 1074.64 180.811 1074.48 Q181.829 1074.29 182.848 1074.29 Q188.635 1074.29 192.014 1077.47 Q195.394 1080.64 195.394 1086.05 Q195.394 1091.63 191.922 1094.73 Q188.45 1097.81 182.13 1097.81 Q179.954 1097.81 177.686 1097.44 Q175.44 1097.07 173.033 1096.33 L173.033 1091.63 Q175.116 1092.77 177.339 1093.32 Q179.561 1093.88 182.038 1093.88 Q186.042 1093.88 188.38 1091.77 Q190.718 1089.66 190.718 1086.05 Q190.718 1082.44 188.38 1080.34 Q186.042 1078.23 182.038 1078.23 Q180.163 1078.23 178.288 1078.65 Q176.436 1079.06 174.491 1079.94 L174.491 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M214.607 1065.66 Q210.996 1065.66 209.167 1069.22 Q207.362 1072.77 207.362 1079.9 Q207.362 1087 209.167 1090.57 Q210.996 1094.11 214.607 1094.11 Q218.241 1094.11 220.047 1090.57 Q221.875 1087 221.875 1079.9 Q221.875 1072.77 220.047 1069.22 Q218.241 1065.66 214.607 1065.66 M214.607 1061.96 Q220.417 1061.96 223.473 1066.56 Q226.551 1071.15 226.551 1079.9 Q226.551 1088.62 223.473 1093.23 Q220.417 1097.81 214.607 1097.81 Q208.797 1097.81 205.718 1093.23 Q202.662 1088.62 202.662 1079.9 Q202.662 1071.15 205.718 1066.56 Q208.797 1061.96 214.607 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M144.329 756.046 L162.686 756.046 L162.686 759.981 L148.612 759.981 L148.612 768.453 Q149.63 768.106 150.649 767.944 Q151.667 767.759 152.686 767.759 Q158.473 767.759 161.852 770.93 Q165.232 774.101 165.232 779.518 Q165.232 785.097 161.76 788.199 Q158.288 791.277 151.968 791.277 Q149.792 791.277 147.524 790.907 Q145.279 790.537 142.871 789.796 L142.871 785.097 Q144.954 786.231 147.177 786.787 Q149.399 787.342 151.876 787.342 Q155.88 787.342 158.218 785.236 Q160.556 783.129 160.556 779.518 Q160.556 775.907 158.218 773.801 Q155.88 771.694 151.876 771.694 Q150.001 771.694 148.126 772.111 Q146.274 772.527 144.329 773.407 L144.329 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M184.445 759.125 Q180.834 759.125 179.005 762.689 Q177.2 766.231 177.2 773.361 Q177.2 780.467 179.005 784.032 Q180.834 787.574 184.445 787.574 Q188.079 787.574 189.885 784.032 Q191.713 780.467 191.713 773.361 Q191.713 766.231 189.885 762.689 Q188.079 759.125 184.445 759.125 M184.445 755.421 Q190.255 755.421 193.311 760.027 Q196.389 764.611 196.389 773.361 Q196.389 782.088 193.311 786.694 Q190.255 791.277 184.445 791.277 Q178.635 791.277 175.556 786.694 Q172.501 782.088 172.501 773.361 Q172.501 764.611 175.556 760.027 Q178.635 755.421 184.445 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M214.607 759.125 Q210.996 759.125 209.167 762.689 Q207.362 766.231 207.362 773.361 Q207.362 780.467 209.167 784.032 Q210.996 787.574 214.607 787.574 Q218.241 787.574 220.047 784.032 Q221.875 780.467 221.875 773.361 Q221.875 766.231 220.047 762.689 Q218.241 759.125 214.607 759.125 M214.607 755.421 Q220.417 755.421 223.473 760.027 Q226.551 764.611 226.551 773.361 Q226.551 782.088 223.473 786.694 Q220.417 791.277 214.607 791.277 Q208.797 791.277 205.718 786.694 Q202.662 782.088 202.662 773.361 Q202.662 764.611 205.718 760.027 Q208.797 755.421 214.607 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M143.103 449.511 L165.325 449.511 L165.325 451.502 L152.778 484.071 L147.894 484.071 L159.7 453.446 L143.103 453.446 L143.103 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M174.491 449.511 L192.848 449.511 L192.848 453.446 L178.774 453.446 L178.774 461.918 Q179.792 461.571 180.811 461.409 Q181.829 461.224 182.848 461.224 Q188.635 461.224 192.014 464.395 Q195.394 467.567 195.394 472.983 Q195.394 478.562 191.922 481.664 Q188.45 484.742 182.13 484.742 Q179.954 484.742 177.686 484.372 Q175.44 484.002 173.033 483.261 L173.033 478.562 Q175.116 479.696 177.339 480.252 Q179.561 480.807 182.038 480.807 Q186.042 480.807 188.38 478.701 Q190.718 476.594 190.718 472.983 Q190.718 469.372 188.38 467.266 Q186.042 465.159 182.038 465.159 Q180.163 465.159 178.288 465.576 Q176.436 465.993 174.491 466.872 L174.491 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M214.607 452.59 Q210.996 452.59 209.167 456.155 Q207.362 459.696 207.362 466.826 Q207.362 473.932 209.167 477.497 Q210.996 481.039 214.607 481.039 Q218.241 481.039 220.047 477.497 Q221.875 473.932 221.875 466.826 Q221.875 459.696 220.047 456.155 Q218.241 452.59 214.607 452.59 M214.607 448.886 Q220.417 448.886 223.473 453.493 Q226.551 458.076 226.551 466.826 Q226.551 475.553 223.473 480.159 Q220.417 484.742 214.607 484.742 Q208.797 484.742 205.718 480.159 Q202.662 475.553 202.662 466.826 Q202.662 458.076 205.718 453.493 Q208.797 448.886 214.607 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M114.931 173.601 L122.57 173.601 L122.57 147.236 L114.26 148.902 L114.26 144.643 L122.524 142.976 L127.2 142.976 L127.2 173.601 L134.839 173.601 L134.839 177.536 L114.931 177.536 L114.931 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M154.283 146.055 Q150.672 146.055 148.843 149.62 Q147.038 153.161 147.038 160.291 Q147.038 167.397 148.843 170.962 Q150.672 174.504 154.283 174.504 Q157.917 174.504 159.723 170.962 Q161.552 167.397 161.552 160.291 Q161.552 153.161 159.723 149.62 Q157.917 146.055 154.283 146.055 M154.283 142.351 Q160.093 142.351 163.149 146.958 Q166.227 151.541 166.227 160.291 Q166.227 169.018 163.149 173.624 Q160.093 178.208 154.283 178.208 Q148.473 178.208 145.394 173.624 Q142.339 169.018 142.339 160.291 Q142.339 151.541 145.394 146.958 Q148.473 142.351 154.283 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M184.445 146.055 Q180.834 146.055 179.005 149.62 Q177.2 153.161 177.2 160.291 Q177.2 167.397 179.005 170.962 Q180.834 174.504 184.445 174.504 Q188.079 174.504 189.885 170.962 Q191.713 167.397 191.713 160.291 Q191.713 153.161 189.885 149.62 Q188.079 146.055 184.445 146.055 M184.445 142.351 Q190.255 142.351 193.311 146.958 Q196.389 151.541 196.389 160.291 Q196.389 169.018 193.311 173.624 Q190.255 178.208 184.445 178.208 Q178.635 178.208 175.556 173.624 Q172.501 169.018 172.501 160.291 Q172.501 151.541 175.556 146.958 Q178.635 142.351 184.445 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M214.607 146.055 Q210.996 146.055 209.167 149.62 Q207.362 153.161 207.362 160.291 Q207.362 167.397 209.167 170.962 Q210.996 174.504 214.607 174.504 Q218.241 174.504 220.047 170.962 Q221.875 167.397 221.875 160.291 Q221.875 153.161 220.047 149.62 Q218.241 146.055 214.607 146.055 M214.607 142.351 Q220.417 142.351 223.473 146.958 Q226.551 151.541 226.551 160.291 Q226.551 169.018 223.473 173.624 Q220.417 178.208 214.607 178.208 Q208.797 178.208 205.718 173.624 Q202.662 169.018 202.662 160.291 Q202.662 151.541 205.718 146.958 Q208.797 142.351 214.607 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M21.7677 1000.18 L58.7206 1000.18 L58.7206 992.418 Q58.7206 982.583 54.2646 978.031 Q49.8086 973.448 40.1964 973.448 Q30.6479 973.448 26.2237 978.031 Q21.7677 982.583 21.7677 992.418 L21.7677 1000.18 M16.4842 1006.61 L16.4842 993.404 Q16.4842 979.591 22.2451 973.13 Q27.9743 966.668 40.1964 966.668 Q52.4822 966.668 58.2432 973.161 Q64.0042 979.654 64.0042 993.404 L64.0042 1006.61 L16.4842 1006.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M28.3562 956.674 L28.3562 950.818 L64.0042 950.818 L64.0042 956.674 L28.3562 956.674 M14.479 956.674 L14.479 950.818 L21.895 950.818 L21.895 956.674 L14.479 956.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M29.4065 915.838 L34.9447 915.838 Q33.6716 918.321 33.035 920.994 Q32.3984 923.668 32.3984 926.533 Q32.3984 930.893 33.7352 933.089 Q35.072 935.254 37.7456 935.254 Q39.7826 935.254 40.9603 933.694 Q42.1061 932.134 43.1565 927.424 L43.6021 925.419 Q44.9389 919.18 47.3897 916.57 Q49.8086 913.929 54.1691 913.929 Q59.1344 913.929 62.0308 917.875 Q64.9272 921.79 64.9272 928.665 Q64.9272 931.53 64.3543 934.649 Q63.8132 937.736 62.6992 941.174 L56.6518 941.174 Q58.3387 937.927 59.198 934.776 Q60.0256 931.625 60.0256 928.538 Q60.0256 924.4 58.6251 922.172 Q57.1929 919.944 54.6147 919.944 Q52.2276 919.944 50.9545 921.567 Q49.6813 923.159 48.5037 928.601 L48.0262 930.639 Q46.8804 936.081 44.5251 938.5 Q42.138 940.919 38.0002 940.919 Q32.9713 940.919 30.2341 937.354 Q27.4968 933.79 27.4968 927.233 Q27.4968 923.986 27.9743 921.122 Q28.4517 918.257 29.4065 915.838 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M58.657 898.937 L77.5631 898.937 L77.5631 904.826 L28.3562 904.826 L28.3562 898.937 L33.7671 898.937 Q30.5842 897.091 29.0564 894.29 Q27.4968 891.458 27.4968 887.543 Q27.4968 881.05 32.6531 877.007 Q37.8093 872.933 46.212 872.933 Q54.6147 872.933 59.771 877.007 Q64.9272 881.05 64.9272 887.543 Q64.9272 891.458 63.3994 894.29 Q61.8398 897.091 58.657 898.937 M46.212 879.013 Q39.7508 879.013 36.0905 881.686 Q32.3984 884.328 32.3984 888.975 Q32.3984 893.622 36.0905 896.296 Q39.7508 898.937 46.212 898.937 Q52.6732 898.937 56.3653 896.296 Q60.0256 893.622 60.0256 888.975 Q60.0256 884.328 56.3653 881.686 Q52.6732 879.013 46.212 879.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M46.0847 847.025 Q46.0847 854.123 47.7079 856.86 Q49.3312 859.597 53.2461 859.597 Q56.3653 859.597 58.2114 857.56 Q60.0256 855.491 60.0256 851.958 Q60.0256 847.089 56.5881 844.16 Q53.1188 841.2 47.3897 841.2 L46.0847 841.2 L46.0847 847.025 M43.6657 835.344 L64.0042 835.344 L64.0042 841.2 L58.5933 841.2 Q61.8398 843.206 63.3994 846.197 Q64.9272 849.189 64.9272 853.518 Q64.9272 858.992 61.8716 862.239 Q58.7843 865.454 53.6281 865.454 Q47.6125 865.454 44.5569 861.443 Q41.5014 857.401 41.5014 849.412 L41.5014 841.2 L40.9285 841.2 Q36.8862 841.2 34.6901 843.874 Q32.4621 846.516 32.4621 851.322 Q32.4621 854.377 33.1941 857.274 Q33.9262 860.17 35.3903 862.844 L29.9795 862.844 Q28.7381 859.629 28.1334 856.605 Q27.4968 853.582 27.4968 850.717 Q27.4968 842.983 31.5072 839.163 Q35.5176 835.344 43.6657 835.344 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M18.2347 817.488 L28.3562 817.488 L28.3562 805.425 L32.9077 805.425 L32.9077 817.488 L52.2594 817.488 Q56.6199 817.488 57.8613 816.31 Q59.1026 815.101 59.1026 811.441 L59.1026 805.425 L64.0042 805.425 L64.0042 811.441 Q64.0042 818.22 61.4897 820.798 Q58.9434 823.376 52.2594 823.376 L32.9077 823.376 L32.9077 827.673 L28.3562 827.673 L28.3562 823.376 L18.2347 823.376 L18.2347 817.488 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M29.7248 772.069 L35.1993 772.069 Q33.8307 774.551 33.1623 777.066 Q32.4621 779.548 32.4621 782.095 Q32.4621 787.792 36.0905 790.943 Q39.6872 794.094 46.212 794.094 Q52.7369 794.094 56.3653 790.943 Q59.9619 787.792 59.9619 782.095 Q59.9619 779.548 59.2935 777.066 Q58.5933 774.551 57.2247 772.069 L62.6355 772.069 Q63.7814 774.52 64.3543 777.161 Q64.9272 779.771 64.9272 782.731 Q64.9272 790.784 59.8664 795.526 Q54.8057 800.269 46.212 800.269 Q37.491 800.269 32.4939 795.495 Q27.4968 790.688 27.4968 782.349 Q27.4968 779.644 28.0697 777.066 Q28.6108 774.488 29.7248 772.069 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M42.4881 732.251 L64.0042 732.251 L64.0042 738.108 L42.679 738.108 Q37.6183 738.108 35.1038 740.081 Q32.5894 742.054 32.5894 746.001 Q32.5894 750.744 35.6131 753.481 Q38.6368 756.218 43.8567 756.218 L64.0042 756.218 L64.0042 762.106 L14.479 762.106 L14.479 756.218 L33.8944 756.218 Q30.6797 754.117 29.0883 751.285 Q27.4968 748.42 27.4968 744.696 Q27.4968 738.553 31.3163 735.402 Q35.1038 732.251 42.4881 732.251 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M14.479 700.391 L14.479 686.896 L19.0304 686.896 L19.0304 694.534 L68.0464 694.534 L68.0464 686.896 L72.5979 686.896 L72.5979 700.391 L14.479 700.391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M16.4842 674.164 L16.4842 664.584 L48.8219 652.457 L16.4842 640.267 L16.4842 630.686 L64.0042 630.686 L64.0042 636.957 L22.277 636.957 L54.8694 649.211 L54.8694 655.672 L22.277 667.926 L64.0042 667.926 L64.0042 674.164 L16.4842 674.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M16.4842 622.156 L16.4842 615.663 L56.6518 605.669 L16.4842 595.707 L16.4842 588.482 L56.6518 578.488 L16.4842 568.525 L16.4842 562 L64.0042 573.936 L64.0042 582.021 L22.7544 592.047 L64.0042 602.168 L64.0042 610.252 L16.4842 622.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M14.479 540.039 L72.5979 540.039 L72.5979 553.534 L68.0464 553.534 L68.0464 545.927 L19.0304 545.927 L19.0304 553.534 L14.479 553.534 L14.479 540.039 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M351.961 12.096 L403.124 12.096 L403.124 18.9825 L381.654 18.9825 L381.654 72.576 L373.431 72.576 L373.431 18.9825 L351.961 18.9825 L351.961 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M448.413 45.1919 L448.413 72.576 L440.959 72.576 L440.959 45.4349 Q440.959 38.994 438.448 35.7938 Q435.936 32.5936 430.913 32.5936 Q424.877 32.5936 421.394 36.4419 Q417.91 40.2903 417.91 46.9338 L417.91 72.576 L410.416 72.576 L410.416 9.54393 L417.91 9.54393 L417.91 34.2544 Q420.583 30.163 424.189 28.1376 Q427.834 26.1121 432.574 26.1121 Q440.392 26.1121 444.403 30.9732 Q448.413 35.7938 448.413 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M502.088 48.0275 L502.088 51.6733 L467.817 51.6733 Q468.303 59.3701 472.435 63.421 Q476.607 67.4314 484.021 67.4314 Q488.314 67.4314 492.325 66.3781 Q496.376 65.3249 500.346 63.2184 L500.346 70.267 Q496.335 71.9684 492.122 72.8596 Q487.909 73.7508 483.575 73.7508 Q472.718 73.7508 466.359 67.4314 Q460.039 61.1119 460.039 50.3365 Q460.039 39.1965 466.034 32.6746 Q472.07 26.1121 482.279 26.1121 Q491.434 26.1121 496.74 32.0264 Q502.088 37.9003 502.088 48.0275 M494.634 45.84 Q494.553 39.7232 491.191 36.0774 Q487.869 32.4315 482.36 32.4315 Q476.121 32.4315 472.354 35.9558 Q468.627 39.4801 468.06 45.8805 L494.634 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M540.612 34.1734 Q539.356 33.4443 537.857 33.1202 Q536.399 32.7556 534.616 32.7556 Q528.297 32.7556 524.894 36.8875 Q521.532 40.9789 521.532 48.6757 L521.532 72.576 L514.038 72.576 L514.038 27.2059 L521.532 27.2059 L521.532 34.2544 Q523.881 30.1225 527.649 28.1376 Q531.416 26.1121 536.804 26.1121 Q537.573 26.1121 538.505 26.2337 Q539.437 26.3147 540.571 26.5172 L540.612 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M582.295 35.9153 Q585.091 30.8922 588.979 28.5022 Q592.868 26.1121 598.134 26.1121 Q605.224 26.1121 609.072 31.0947 Q612.92 36.0368 612.92 45.1919 L612.92 72.576 L605.426 72.576 L605.426 45.4349 Q605.426 38.913 603.117 35.7533 Q600.808 32.5936 596.069 32.5936 Q590.276 32.5936 586.913 36.4419 Q583.551 40.2903 583.551 46.9338 L583.551 72.576 L576.057 72.576 L576.057 45.4349 Q576.057 38.8725 573.748 35.7533 Q571.439 32.5936 566.618 32.5936 Q560.907 32.5936 557.544 36.4824 Q554.182 40.3308 554.182 46.9338 L554.182 72.576 L546.688 72.576 L546.688 27.2059 L554.182 27.2059 L554.182 34.2544 Q556.734 30.082 560.299 28.0971 Q563.864 26.1121 568.765 26.1121 Q573.708 26.1121 577.151 28.6237 Q580.635 31.1352 582.295 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M648.406 49.7694 Q639.373 49.7694 635.889 51.8354 Q632.405 53.9013 632.405 58.8839 Q632.405 62.8538 634.998 65.2034 Q637.631 67.5124 642.127 67.5124 Q648.325 67.5124 652.052 63.1374 Q655.819 58.7219 655.819 51.4303 L655.819 49.7694 L648.406 49.7694 M663.273 46.6907 L663.273 72.576 L655.819 72.576 L655.819 65.6895 Q653.267 69.8214 649.459 71.8063 Q645.652 73.7508 640.142 73.7508 Q633.175 73.7508 629.043 69.8619 Q624.951 65.9325 624.951 59.3701 Q624.951 51.7138 630.056 47.825 Q635.2 43.9361 645.368 43.9361 L655.819 43.9361 L655.819 43.2069 Q655.819 38.0623 652.417 35.2672 Q649.054 32.4315 642.937 32.4315 Q639.049 32.4315 635.362 33.3632 Q631.676 34.295 628.273 36.1584 L628.273 29.2718 Q632.365 27.692 636.213 26.9223 Q640.061 26.1121 643.707 26.1121 Q653.551 26.1121 658.412 31.2163 Q663.273 36.3204 663.273 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M678.626 9.54393 L686.08 9.54393 L686.08 72.576 L678.626 72.576 L678.626 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M736.554 18.8205 L736.554 65.8515 L746.438 65.8515 Q758.955 65.8515 764.748 60.1802 Q770.581 54.509 770.581 42.2752 Q770.581 30.1225 764.748 24.4918 Q758.955 18.8205 746.438 18.8205 L736.554 18.8205 M728.371 12.096 L745.182 12.096 Q762.763 12.096 770.987 19.4281 Q779.21 26.7198 779.21 42.2752 Q779.21 57.9117 770.946 65.2439 Q762.682 72.576 745.182 72.576 L728.371 72.576 L728.371 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M791.93 27.2059 L799.383 27.2059 L799.383 72.576 L791.93 72.576 L791.93 27.2059 M791.93 9.54393 L799.383 9.54393 L799.383 18.9825 L791.93 18.9825 L791.93 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M843.903 28.5427 L843.903 35.5912 Q840.743 33.9709 837.34 33.1607 Q833.938 32.3505 830.292 32.3505 Q824.742 32.3505 821.947 34.0519 Q819.192 35.7533 819.192 39.156 Q819.192 41.7486 821.177 43.2475 Q823.162 44.7058 829.158 46.0426 L831.71 46.6097 Q839.649 48.3111 842.971 51.4303 Q846.333 54.509 846.333 60.0587 Q846.333 66.3781 841.31 70.0644 Q836.328 73.7508 827.578 73.7508 Q823.932 73.7508 819.962 73.0216 Q816.033 72.3329 811.658 70.9151 L811.658 63.2184 Q815.79 65.3654 819.8 66.4591 Q823.81 67.5124 827.74 67.5124 Q833.006 67.5124 835.842 65.73 Q838.677 63.9071 838.677 60.6258 Q838.677 57.5877 836.611 55.9673 Q834.586 54.3469 827.659 52.8481 L825.066 52.2405 Q818.139 50.7821 815.06 47.7845 Q811.982 44.7463 811.982 39.4801 Q811.982 33.0797 816.519 29.5959 Q821.056 26.1121 829.401 26.1121 Q833.533 26.1121 837.178 26.7198 Q840.824 27.3274 843.903 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M865.413 65.7705 L865.413 89.8329 L857.919 89.8329 L857.919 27.2059 L865.413 27.2059 L865.413 34.0924 Q867.763 30.0415 871.327 28.0971 Q874.933 26.1121 879.915 26.1121 Q888.179 26.1121 893.324 32.6746 Q898.509 39.2371 898.509 49.9314 Q898.509 60.6258 893.324 67.1883 Q888.179 73.7508 879.915 73.7508 Q874.933 73.7508 871.327 71.8063 Q867.763 69.8214 865.413 65.7705 M890.772 49.9314 Q890.772 41.7081 887.369 37.0496 Q884.007 32.3505 878.092 32.3505 Q872.178 32.3505 868.775 37.0496 Q865.413 41.7081 865.413 49.9314 Q865.413 58.1548 868.775 62.8538 Q872.178 67.5124 878.092 67.5124 Q884.007 67.5124 887.369 62.8538 Q890.772 58.1548 890.772 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M931.483 49.7694 Q922.45 49.7694 918.966 51.8354 Q915.482 53.9013 915.482 58.8839 Q915.482 62.8538 918.075 65.2034 Q920.708 67.5124 925.204 67.5124 Q931.402 67.5124 935.129 63.1374 Q938.897 58.7219 938.897 51.4303 L938.897 49.7694 L931.483 49.7694 M946.35 46.6907 L946.35 72.576 L938.897 72.576 L938.897 65.6895 Q936.344 69.8214 932.537 71.8063 Q928.729 73.7508 923.22 73.7508 Q916.252 73.7508 912.12 69.8619 Q908.029 65.9325 908.029 59.3701 Q908.029 51.7138 913.133 47.825 Q918.277 43.9361 928.445 43.9361 L938.897 43.9361 L938.897 43.2069 Q938.897 38.0623 935.494 35.2672 Q932.132 32.4315 926.015 32.4315 Q922.126 32.4315 918.439 33.3632 Q914.753 34.295 911.35 36.1584 L911.35 29.2718 Q915.442 27.692 919.29 26.9223 Q923.139 26.1121 926.784 26.1121 Q936.628 26.1121 941.489 31.2163 Q946.35 36.3204 946.35 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M969.076 14.324 L969.076 27.2059 L984.429 27.2059 L984.429 32.9987 L969.076 32.9987 L969.076 57.6282 Q969.076 63.1779 970.575 64.7578 Q972.114 66.3376 976.772 66.3376 L984.429 66.3376 L984.429 72.576 L976.772 72.576 Q968.144 72.576 964.863 69.3758 Q961.582 66.1351 961.582 57.6282 L961.582 32.9987 L956.113 32.9987 L956.113 27.2059 L961.582 27.2059 L961.582 14.324 L969.076 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1026.88 28.9478 L1026.88 35.9153 Q1023.72 34.1734 1020.52 33.3227 Q1017.36 32.4315 1014.12 32.4315 Q1006.87 32.4315 1002.86 37.0496 Q998.85 41.6271 998.85 49.9314 Q998.85 58.2358 1002.86 62.8538 Q1006.87 67.4314 1014.12 67.4314 Q1017.36 67.4314 1020.52 66.5807 Q1023.72 65.6895 1026.88 63.9476 L1026.88 70.8341 Q1023.76 72.2924 1020.4 73.0216 Q1017.08 73.7508 1013.31 73.7508 Q1003.06 73.7508 997.027 67.3098 Q990.991 60.8689 990.991 49.9314 Q990.991 38.832 997.068 32.472 Q1003.18 26.1121 1013.8 26.1121 Q1017.24 26.1121 1020.52 26.8413 Q1023.8 27.5299 1026.88 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1077.56 45.1919 L1077.56 72.576 L1070.11 72.576 L1070.11 45.4349 Q1070.11 38.994 1067.59 35.7938 Q1065.08 32.5936 1060.06 32.5936 Q1054.02 32.5936 1050.54 36.4419 Q1047.06 40.2903 1047.06 46.9338 L1047.06 72.576 L1039.56 72.576 L1039.56 9.54393 L1047.06 9.54393 L1047.06 34.2544 Q1049.73 30.163 1053.33 28.1376 Q1056.98 26.1121 1061.72 26.1121 Q1069.54 26.1121 1073.55 30.9732 Q1077.56 35.7938 1077.56 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip652)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="288.148,1386.4 359.25,1202.47 430.352,1018.55 501.454,834.633 572.556,650.712 643.658,466.791 714.76,282.87 785.862,344.177 856.964,160.256 928.066,160.256 999.168,160.256 1070.27,160.256 1141.37,160.256 "/>
-<polyline clip-path="url(#clip652)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="288.148,1018.55 359.25,1018.55 430.352,1018.55 501.454,1018.55 572.556,1018.55 643.658,1018.55 714.76,1018.55 785.862,1018.55 856.964,1018.55 928.066,834.633 999.168,650.712 1070.27,466.791 1141.37,282.87 "/>
-<path clip-path="url(#clip650)" d="M986.519 1379.86 L1136.82 1379.86 L1136.82 1224.34 L986.519 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="986.519,1379.86 1136.82,1379.86 1136.82,1224.34 986.519,1224.34 986.519,1379.86 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="996.568,1276.18 1056.86,1276.18 "/>
-<path clip-path="url(#clip650)" d="M1092.47 1288.53 L1092.47 1279.24 L1084.83 1279.24 L1084.83 1275.4 L1097.1 1275.4 L1097.1 1290.24 Q1094.39 1292.16 1091.12 1293.16 Q1087.86 1294.13 1084.16 1294.13 Q1076.06 1294.13 1071.47 1289.41 Q1066.91 1284.66 1066.91 1276.21 Q1066.91 1267.74 1071.47 1263.02 Q1076.06 1258.27 1084.16 1258.27 Q1087.54 1258.27 1090.57 1259.1 Q1093.62 1259.94 1096.19 1261.56 L1096.19 1266.54 Q1093.6 1264.34 1090.68 1263.22 Q1087.77 1262.11 1084.55 1262.11 Q1078.21 1262.11 1075.01 1265.66 Q1071.84 1269.2 1071.84 1276.21 Q1071.84 1283.2 1075.01 1286.74 Q1078.21 1290.28 1084.55 1290.28 Q1087.03 1290.28 1088.97 1289.87 Q1090.92 1289.43 1092.47 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1106.87 1289.52 L1114.5 1289.52 L1114.5 1263.16 L1106.19 1264.82 L1106.19 1260.56 L1114.46 1258.9 L1119.13 1258.9 L1119.13 1289.52 L1126.77 1289.52 L1126.77 1293.46 L1106.87 1293.46 L1106.87 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip650)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="996.568,1328.02 1056.86,1328.02 "/>
-<path clip-path="url(#clip650)" d="M1092.47 1340.37 L1092.47 1331.08 L1084.83 1331.08 L1084.83 1327.24 L1097.1 1327.24 L1097.1 1342.08 Q1094.39 1344 1091.12 1345 Q1087.86 1345.97 1084.16 1345.97 Q1076.06 1345.97 1071.47 1341.25 Q1066.91 1336.5 1066.91 1328.05 Q1066.91 1319.58 1071.47 1314.86 Q1076.06 1310.11 1084.16 1310.11 Q1087.54 1310.11 1090.57 1310.94 Q1093.62 1311.78 1096.19 1313.4 L1096.19 1318.38 Q1093.6 1316.18 1090.68 1315.06 Q1087.77 1313.95 1084.55 1313.95 Q1078.21 1313.95 1075.01 1317.5 Q1071.84 1321.04 1071.84 1328.05 Q1071.84 1335.04 1075.01 1338.58 Q1078.21 1342.12 1084.55 1342.12 Q1087.03 1342.12 1088.97 1341.71 Q1090.92 1341.27 1092.47 1340.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1110.08 1341.36 L1126.4 1341.36 L1126.4 1345.3 L1104.46 1345.3 L1104.46 1341.36 Q1107.12 1338.61 1111.7 1333.98 Q1116.31 1329.32 1117.49 1327.98 Q1119.74 1325.46 1120.62 1323.72 Q1121.52 1321.96 1121.52 1320.27 Q1121.52 1317.52 1119.57 1315.78 Q1117.65 1314.05 1114.55 1314.05 Q1112.35 1314.05 1109.9 1314.81 Q1107.47 1315.57 1104.69 1317.13 L1104.69 1312.4 Q1107.51 1311.27 1109.97 1310.69 Q1112.42 1310.11 1114.46 1310.11 Q1119.83 1310.11 1123.02 1312.8 Q1126.22 1315.48 1126.22 1319.97 Q1126.22 1322.1 1125.41 1324.02 Q1124.62 1325.92 1122.51 1328.51 Q1121.93 1329.19 1118.83 1332.4 Q1115.73 1335.6 1110.08 1341.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1448.34 1423.18 L2352.76 1423.18 L2352.76 123.472 L1448.34 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="382.951,1423.18 382.951,123.472 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="619.957,1423.18 619.957,123.472 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="856.964,1423.18 856.964,123.472 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1093.97,1423.18 1093.97,123.472 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1423.18 1166.97,1423.18 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="382.951,1423.18 382.951,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="619.957,1423.18 619.957,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="856.964,1423.18 856.964,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1093.97,1423.18 1093.97,1404.28 "/>
+<path clip-path="url(#clip470)" d="M342.569 1451.02 L360.925 1451.02 L360.925 1454.96 L346.851 1454.96 L346.851 1463.43 Q347.87 1463.08 348.888 1462.92 Q349.907 1462.73 350.925 1462.73 Q356.712 1462.73 360.092 1465.9 Q363.472 1469.08 363.472 1474.49 Q363.472 1480.07 359.999 1483.17 Q356.527 1486.25 350.208 1486.25 Q348.032 1486.25 345.763 1485.88 Q343.518 1485.51 341.111 1484.77 L341.111 1480.07 Q343.194 1481.2 345.416 1481.76 Q347.638 1482.32 350.115 1482.32 Q354.12 1482.32 356.458 1480.21 Q358.796 1478.1 358.796 1474.49 Q358.796 1470.88 356.458 1468.77 Q354.12 1466.67 350.115 1466.67 Q348.24 1466.67 346.365 1467.08 Q344.513 1467.5 342.569 1468.38 L342.569 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M382.684 1454.1 Q379.073 1454.1 377.245 1457.66 Q375.439 1461.2 375.439 1468.33 Q375.439 1475.44 377.245 1479.01 Q379.073 1482.55 382.684 1482.55 Q386.319 1482.55 388.124 1479.01 Q389.953 1475.44 389.953 1468.33 Q389.953 1461.2 388.124 1457.66 Q386.319 1454.1 382.684 1454.1 M382.684 1450.39 Q388.495 1450.39 391.55 1455 Q394.629 1459.58 394.629 1468.33 Q394.629 1477.06 391.55 1481.67 Q388.495 1486.25 382.684 1486.25 Q376.874 1486.25 373.796 1481.67 Q370.74 1477.06 370.74 1468.33 Q370.74 1459.58 373.796 1455 Q376.874 1450.39 382.684 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M412.846 1454.1 Q409.235 1454.1 407.407 1457.66 Q405.601 1461.2 405.601 1468.33 Q405.601 1475.44 407.407 1479.01 Q409.235 1482.55 412.846 1482.55 Q416.481 1482.55 418.286 1479.01 Q420.115 1475.44 420.115 1468.33 Q420.115 1461.2 418.286 1457.66 Q416.481 1454.1 412.846 1454.1 M412.846 1450.39 Q418.656 1450.39 421.712 1455 Q424.791 1459.58 424.791 1468.33 Q424.791 1477.06 421.712 1481.67 Q418.656 1486.25 412.846 1486.25 Q407.036 1486.25 403.957 1481.67 Q400.902 1477.06 400.902 1468.33 Q400.902 1459.58 403.957 1455 Q407.036 1450.39 412.846 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M564.483 1481.64 L572.122 1481.64 L572.122 1455.28 L563.812 1456.95 L563.812 1452.69 L572.076 1451.02 L576.752 1451.02 L576.752 1481.64 L584.39 1481.64 L584.39 1485.58 L564.483 1485.58 L564.483 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M603.835 1454.1 Q600.224 1454.1 598.395 1457.66 Q596.589 1461.2 596.589 1468.33 Q596.589 1475.44 598.395 1479.01 Q600.224 1482.55 603.835 1482.55 Q607.469 1482.55 609.275 1479.01 Q611.103 1475.44 611.103 1468.33 Q611.103 1461.2 609.275 1457.66 Q607.469 1454.1 603.835 1454.1 M603.835 1450.39 Q609.645 1450.39 612.7 1455 Q615.779 1459.58 615.779 1468.33 Q615.779 1477.06 612.7 1481.67 Q609.645 1486.25 603.835 1486.25 Q598.025 1486.25 594.946 1481.67 Q591.89 1477.06 591.89 1468.33 Q591.89 1459.58 594.946 1455 Q598.025 1450.39 603.835 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M633.997 1454.1 Q630.386 1454.1 628.557 1457.66 Q626.751 1461.2 626.751 1468.33 Q626.751 1475.44 628.557 1479.01 Q630.386 1482.55 633.997 1482.55 Q637.631 1482.55 639.436 1479.01 Q641.265 1475.44 641.265 1468.33 Q641.265 1461.2 639.436 1457.66 Q637.631 1454.1 633.997 1454.1 M633.997 1450.39 Q639.807 1450.39 642.862 1455 Q645.941 1459.58 645.941 1468.33 Q645.941 1477.06 642.862 1481.67 Q639.807 1486.25 633.997 1486.25 Q628.186 1486.25 625.108 1481.67 Q622.052 1477.06 622.052 1468.33 Q622.052 1459.58 625.108 1455 Q628.186 1450.39 633.997 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M664.158 1454.1 Q660.547 1454.1 658.719 1457.66 Q656.913 1461.2 656.913 1468.33 Q656.913 1475.44 658.719 1479.01 Q660.547 1482.55 664.158 1482.55 Q667.793 1482.55 669.598 1479.01 Q671.427 1475.44 671.427 1468.33 Q671.427 1461.2 669.598 1457.66 Q667.793 1454.1 664.158 1454.1 M664.158 1450.39 Q669.969 1450.39 673.024 1455 Q676.103 1459.58 676.103 1468.33 Q676.103 1477.06 673.024 1481.67 Q669.969 1486.25 664.158 1486.25 Q658.348 1486.25 655.27 1481.67 Q652.214 1477.06 652.214 1468.33 Q652.214 1459.58 655.27 1455 Q658.348 1450.39 664.158 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M801.49 1481.64 L809.129 1481.64 L809.129 1455.28 L800.818 1456.95 L800.818 1452.69 L809.082 1451.02 L813.758 1451.02 L813.758 1481.64 L821.397 1481.64 L821.397 1485.58 L801.49 1485.58 L801.49 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M830.888 1451.02 L849.244 1451.02 L849.244 1454.96 L835.17 1454.96 L835.17 1463.43 Q836.189 1463.08 837.207 1462.92 Q838.226 1462.73 839.244 1462.73 Q845.031 1462.73 848.411 1465.9 Q851.79 1469.08 851.79 1474.49 Q851.79 1480.07 848.318 1483.17 Q844.846 1486.25 838.527 1486.25 Q836.351 1486.25 834.082 1485.88 Q831.837 1485.51 829.429 1484.77 L829.429 1480.07 Q831.513 1481.2 833.735 1481.76 Q835.957 1482.32 838.434 1482.32 Q842.439 1482.32 844.777 1480.21 Q847.115 1478.1 847.115 1474.49 Q847.115 1470.88 844.777 1468.77 Q842.439 1466.67 838.434 1466.67 Q836.559 1466.67 834.684 1467.08 Q832.832 1467.5 830.888 1468.38 L830.888 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M871.003 1454.1 Q867.392 1454.1 865.564 1457.66 Q863.758 1461.2 863.758 1468.33 Q863.758 1475.44 865.564 1479.01 Q867.392 1482.55 871.003 1482.55 Q874.638 1482.55 876.443 1479.01 Q878.272 1475.44 878.272 1468.33 Q878.272 1461.2 876.443 1457.66 Q874.638 1454.1 871.003 1454.1 M871.003 1450.39 Q876.813 1450.39 879.869 1455 Q882.948 1459.58 882.948 1468.33 Q882.948 1477.06 879.869 1481.67 Q876.813 1486.25 871.003 1486.25 Q865.193 1486.25 862.114 1481.67 Q859.059 1477.06 859.059 1468.33 Q859.059 1459.58 862.114 1455 Q865.193 1450.39 871.003 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M901.165 1454.1 Q897.554 1454.1 895.725 1457.66 Q893.92 1461.2 893.92 1468.33 Q893.92 1475.44 895.725 1479.01 Q897.554 1482.55 901.165 1482.55 Q904.799 1482.55 906.605 1479.01 Q908.434 1475.44 908.434 1468.33 Q908.434 1461.2 906.605 1457.66 Q904.799 1454.1 901.165 1454.1 M901.165 1450.39 Q906.975 1450.39 910.031 1455 Q913.11 1459.58 913.11 1468.33 Q913.11 1477.06 910.031 1481.67 Q906.975 1486.25 901.165 1486.25 Q895.355 1486.25 892.276 1481.67 Q889.221 1477.06 889.221 1468.33 Q889.221 1459.58 892.276 1455 Q895.355 1450.39 901.165 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1042.58 1481.64 L1058.9 1481.64 L1058.9 1485.58 L1036.96 1485.58 L1036.96 1481.64 Q1039.62 1478.89 1044.2 1474.26 Q1048.81 1469.61 1049.99 1468.27 Q1052.23 1465.74 1053.11 1464.01 Q1054.02 1462.25 1054.02 1460.56 Q1054.02 1457.8 1052.07 1456.07 Q1050.15 1454.33 1047.05 1454.33 Q1044.85 1454.33 1042.4 1455.09 Q1039.97 1455.86 1037.19 1457.41 L1037.19 1452.69 Q1040.01 1451.55 1042.47 1450.97 Q1044.92 1450.39 1046.96 1450.39 Q1052.33 1450.39 1055.52 1453.08 Q1058.72 1455.77 1058.72 1460.26 Q1058.72 1462.39 1057.91 1464.31 Q1057.12 1466.2 1055.01 1468.8 Q1054.43 1469.47 1051.33 1472.69 Q1048.23 1475.88 1042.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1078.72 1454.1 Q1075.11 1454.1 1073.28 1457.66 Q1071.47 1461.2 1071.47 1468.33 Q1071.47 1475.44 1073.28 1479.01 Q1075.11 1482.55 1078.72 1482.55 Q1082.35 1482.55 1084.16 1479.01 Q1085.98 1475.44 1085.98 1468.33 Q1085.98 1461.2 1084.16 1457.66 Q1082.35 1454.1 1078.72 1454.1 M1078.72 1450.39 Q1084.53 1450.39 1087.58 1455 Q1090.66 1459.58 1090.66 1468.33 Q1090.66 1477.06 1087.58 1481.67 Q1084.53 1486.25 1078.72 1486.25 Q1072.91 1486.25 1069.83 1481.67 Q1066.77 1477.06 1066.77 1468.33 Q1066.77 1459.58 1069.83 1455 Q1072.91 1450.39 1078.72 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1108.88 1454.1 Q1105.27 1454.1 1103.44 1457.66 Q1101.63 1461.2 1101.63 1468.33 Q1101.63 1475.44 1103.44 1479.01 Q1105.27 1482.55 1108.88 1482.55 Q1112.51 1482.55 1114.32 1479.01 Q1116.15 1475.44 1116.15 1468.33 Q1116.15 1461.2 1114.32 1457.66 Q1112.51 1454.1 1108.88 1454.1 M1108.88 1450.39 Q1114.69 1450.39 1117.74 1455 Q1120.82 1459.58 1120.82 1468.33 Q1120.82 1477.06 1117.74 1481.67 Q1114.69 1486.25 1108.88 1486.25 Q1103.07 1486.25 1099.99 1481.67 Q1096.93 1477.06 1096.93 1468.33 Q1096.93 1459.58 1099.99 1455 Q1103.07 1450.39 1108.88 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1139.04 1454.1 Q1135.43 1454.1 1133.6 1457.66 Q1131.79 1461.2 1131.79 1468.33 Q1131.79 1475.44 1133.6 1479.01 Q1135.43 1482.55 1139.04 1482.55 Q1142.67 1482.55 1144.48 1479.01 Q1146.31 1475.44 1146.31 1468.33 Q1146.31 1461.2 1144.48 1457.66 Q1142.67 1454.1 1139.04 1454.1 M1139.04 1450.39 Q1144.85 1450.39 1147.91 1455 Q1150.98 1459.58 1150.98 1468.33 Q1150.98 1477.06 1147.91 1481.67 Q1144.85 1486.25 1139.04 1486.25 Q1133.23 1486.25 1130.15 1481.67 Q1127.1 1477.06 1127.1 1468.33 Q1127.1 1459.58 1130.15 1455 Q1133.23 1450.39 1139.04 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M589.133 1525.81 L589.133 1562.76 L596.899 1562.76 Q606.734 1562.76 611.285 1558.3 Q615.869 1553.85 615.869 1544.24 Q615.869 1534.69 611.285 1530.26 Q606.734 1525.81 596.899 1525.81 L589.133 1525.81 M582.703 1520.52 L595.912 1520.52 Q609.726 1520.52 616.187 1526.28 Q622.648 1532.01 622.648 1544.24 Q622.648 1556.52 616.155 1562.28 Q609.662 1568.04 595.912 1568.04 L582.703 1568.04 L582.703 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M663.134 1548.76 L663.134 1551.62 L636.207 1551.62 Q636.589 1557.67 639.836 1560.85 Q643.114 1564 648.939 1564 Q652.312 1564 655.463 1563.17 Q658.646 1562.35 661.765 1560.69 L661.765 1566.23 Q658.614 1567.57 655.304 1568.27 Q651.994 1568.97 648.588 1568.97 Q640.058 1568.97 635.061 1564 Q630.096 1559.04 630.096 1550.57 Q630.096 1541.82 634.807 1536.69 Q639.549 1531.54 647.57 1531.54 Q654.763 1531.54 658.933 1536.18 Q663.134 1540.8 663.134 1548.76 M657.278 1547.04 Q657.214 1542.23 654.572 1539.37 Q651.962 1536.5 647.634 1536.5 Q642.732 1536.5 639.772 1539.27 Q636.844 1542.04 636.398 1547.07 L657.278 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M700.501 1539.24 Q702.697 1535.29 705.753 1533.41 Q708.808 1531.54 712.946 1531.54 Q718.516 1531.54 721.539 1535.45 Q724.563 1539.33 724.563 1546.53 L724.563 1568.04 L718.675 1568.04 L718.675 1546.72 Q718.675 1541.59 716.861 1539.11 Q715.046 1536.63 711.323 1536.63 Q706.771 1536.63 704.129 1539.65 Q701.488 1542.68 701.488 1547.9 L701.488 1568.04 L695.599 1568.04 L695.599 1546.72 Q695.599 1541.56 693.785 1539.11 Q691.971 1536.63 688.183 1536.63 Q683.695 1536.63 681.054 1539.68 Q678.412 1542.71 678.412 1547.9 L678.412 1568.04 L672.524 1568.04 L672.524 1532.4 L678.412 1532.4 L678.412 1537.93 Q680.417 1534.66 683.218 1533.1 Q686.019 1531.54 689.87 1531.54 Q693.753 1531.54 696.459 1533.51 Q699.196 1535.48 700.501 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M752.445 1550.12 Q745.347 1550.12 742.61 1551.75 Q739.873 1553.37 739.873 1557.29 Q739.873 1560.4 741.91 1562.25 Q743.979 1564.07 747.512 1564.07 Q752.381 1564.07 755.31 1560.63 Q758.27 1557.16 758.27 1551.43 L758.27 1550.12 L752.445 1550.12 M764.126 1547.71 L764.126 1568.04 L758.27 1568.04 L758.27 1562.63 Q756.264 1565.88 753.273 1567.44 Q750.281 1568.97 745.952 1568.97 Q740.477 1568.97 737.231 1565.91 Q734.016 1562.82 734.016 1557.67 Q734.016 1551.65 738.027 1548.6 Q742.069 1545.54 750.058 1545.54 L758.27 1545.54 L758.27 1544.97 Q758.27 1540.93 755.596 1538.73 Q752.954 1536.5 748.148 1536.5 Q745.093 1536.5 742.196 1537.23 Q739.3 1537.97 736.626 1539.43 L736.626 1534.02 Q739.841 1532.78 742.865 1532.17 Q745.888 1531.54 748.753 1531.54 Q756.487 1531.54 760.307 1535.55 Q764.126 1539.56 764.126 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M805.821 1546.53 L805.821 1568.04 L799.965 1568.04 L799.965 1546.72 Q799.965 1541.66 797.992 1539.14 Q796.018 1536.63 792.072 1536.63 Q787.329 1536.63 784.592 1539.65 Q781.855 1542.68 781.855 1547.9 L781.855 1568.04 L775.966 1568.04 L775.966 1532.4 L781.855 1532.4 L781.855 1537.93 Q783.955 1534.72 786.788 1533.13 Q789.653 1531.54 793.376 1531.54 Q799.519 1531.54 802.67 1535.36 Q805.821 1539.14 805.821 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M840.96 1537.81 L840.96 1518.52 L846.817 1518.52 L846.817 1568.04 L840.96 1568.04 L840.96 1562.7 Q839.114 1565.88 836.281 1567.44 Q833.48 1568.97 829.534 1568.97 Q823.073 1568.97 818.998 1563.81 Q814.956 1558.65 814.956 1550.25 Q814.956 1541.85 818.998 1536.69 Q823.073 1531.54 829.534 1531.54 Q833.48 1531.54 836.281 1533.1 Q839.114 1534.62 840.96 1537.81 M821.004 1550.25 Q821.004 1556.71 823.645 1560.4 Q826.319 1564.07 830.966 1564.07 Q835.613 1564.07 838.287 1560.4 Q840.96 1556.71 840.96 1550.25 Q840.96 1543.79 838.287 1540.13 Q835.613 1536.44 830.966 1536.44 Q826.319 1536.44 823.645 1540.13 Q821.004 1543.79 821.004 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,1386.4 1166.97,1386.4 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,1079.86 1166.97,1079.86 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,773.326 1166.97,773.326 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,466.791 1166.97,466.791 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="262.551,160.256 1166.97,160.256 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1423.18 262.551,123.472 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1386.4 281.449,1386.4 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,1079.86 281.449,1079.86 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,773.326 281.449,773.326 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,466.791 281.449,466.791 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="262.551,160.256 281.449,160.256 "/>
+<path clip-path="url(#clip470)" d="M214.607 1372.19 Q210.996 1372.19 209.167 1375.76 Q207.362 1379.3 207.362 1386.43 Q207.362 1393.54 209.167 1397.1 Q210.996 1400.64 214.607 1400.64 Q218.241 1400.64 220.047 1397.1 Q221.875 1393.54 221.875 1386.43 Q221.875 1379.3 220.047 1375.76 Q218.241 1372.19 214.607 1372.19 M214.607 1368.49 Q220.417 1368.49 223.473 1373.1 Q226.551 1377.68 226.551 1386.43 Q226.551 1395.16 223.473 1399.76 Q220.417 1404.35 214.607 1404.35 Q208.797 1404.35 205.718 1399.76 Q202.662 1395.16 202.662 1386.43 Q202.662 1377.68 205.718 1373.1 Q208.797 1368.49 214.607 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M148.311 1093.21 L164.63 1093.21 L164.63 1097.14 L142.686 1097.14 L142.686 1093.21 Q145.348 1090.45 149.931 1085.82 Q154.538 1081.17 155.718 1079.83 Q157.964 1077.3 158.843 1075.57 Q159.746 1073.81 159.746 1072.12 Q159.746 1069.36 157.802 1067.63 Q155.88 1065.89 152.778 1065.89 Q150.579 1065.89 148.126 1066.65 Q145.695 1067.42 142.917 1068.97 L142.917 1064.25 Q145.741 1063.11 148.195 1062.53 Q150.649 1061.96 152.686 1061.96 Q158.056 1061.96 161.251 1064.64 Q164.445 1067.33 164.445 1071.82 Q164.445 1073.95 163.635 1075.87 Q162.848 1077.77 160.741 1080.36 Q160.163 1081.03 157.061 1084.25 Q153.959 1087.44 148.311 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M174.491 1062.58 L192.848 1062.58 L192.848 1066.52 L178.774 1066.52 L178.774 1074.99 Q179.792 1074.64 180.811 1074.48 Q181.829 1074.29 182.848 1074.29 Q188.635 1074.29 192.014 1077.47 Q195.394 1080.64 195.394 1086.05 Q195.394 1091.63 191.922 1094.73 Q188.45 1097.81 182.13 1097.81 Q179.954 1097.81 177.686 1097.44 Q175.44 1097.07 173.033 1096.33 L173.033 1091.63 Q175.116 1092.77 177.339 1093.32 Q179.561 1093.88 182.038 1093.88 Q186.042 1093.88 188.38 1091.77 Q190.718 1089.66 190.718 1086.05 Q190.718 1082.44 188.38 1080.34 Q186.042 1078.23 182.038 1078.23 Q180.163 1078.23 178.288 1078.65 Q176.436 1079.06 174.491 1079.94 L174.491 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M214.607 1065.66 Q210.996 1065.66 209.167 1069.22 Q207.362 1072.77 207.362 1079.9 Q207.362 1087 209.167 1090.57 Q210.996 1094.11 214.607 1094.11 Q218.241 1094.11 220.047 1090.57 Q221.875 1087 221.875 1079.9 Q221.875 1072.77 220.047 1069.22 Q218.241 1065.66 214.607 1065.66 M214.607 1061.96 Q220.417 1061.96 223.473 1066.56 Q226.551 1071.15 226.551 1079.9 Q226.551 1088.62 223.473 1093.23 Q220.417 1097.81 214.607 1097.81 Q208.797 1097.81 205.718 1093.23 Q202.662 1088.62 202.662 1079.9 Q202.662 1071.15 205.718 1066.56 Q208.797 1061.96 214.607 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M144.329 756.046 L162.686 756.046 L162.686 759.981 L148.612 759.981 L148.612 768.453 Q149.63 768.106 150.649 767.944 Q151.667 767.759 152.686 767.759 Q158.473 767.759 161.852 770.93 Q165.232 774.101 165.232 779.518 Q165.232 785.097 161.76 788.199 Q158.288 791.277 151.968 791.277 Q149.792 791.277 147.524 790.907 Q145.279 790.537 142.871 789.796 L142.871 785.097 Q144.954 786.231 147.177 786.787 Q149.399 787.342 151.876 787.342 Q155.88 787.342 158.218 785.236 Q160.556 783.129 160.556 779.518 Q160.556 775.907 158.218 773.801 Q155.88 771.694 151.876 771.694 Q150.001 771.694 148.126 772.111 Q146.274 772.527 144.329 773.407 L144.329 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 759.125 Q180.834 759.125 179.005 762.689 Q177.2 766.231 177.2 773.361 Q177.2 780.467 179.005 784.032 Q180.834 787.574 184.445 787.574 Q188.079 787.574 189.885 784.032 Q191.713 780.467 191.713 773.361 Q191.713 766.231 189.885 762.689 Q188.079 759.125 184.445 759.125 M184.445 755.421 Q190.255 755.421 193.311 760.027 Q196.389 764.611 196.389 773.361 Q196.389 782.088 193.311 786.694 Q190.255 791.277 184.445 791.277 Q178.635 791.277 175.556 786.694 Q172.501 782.088 172.501 773.361 Q172.501 764.611 175.556 760.027 Q178.635 755.421 184.445 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M214.607 759.125 Q210.996 759.125 209.167 762.689 Q207.362 766.231 207.362 773.361 Q207.362 780.467 209.167 784.032 Q210.996 787.574 214.607 787.574 Q218.241 787.574 220.047 784.032 Q221.875 780.467 221.875 773.361 Q221.875 766.231 220.047 762.689 Q218.241 759.125 214.607 759.125 M214.607 755.421 Q220.417 755.421 223.473 760.027 Q226.551 764.611 226.551 773.361 Q226.551 782.088 223.473 786.694 Q220.417 791.277 214.607 791.277 Q208.797 791.277 205.718 786.694 Q202.662 782.088 202.662 773.361 Q202.662 764.611 205.718 760.027 Q208.797 755.421 214.607 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M143.103 449.511 L165.325 449.511 L165.325 451.502 L152.778 484.071 L147.894 484.071 L159.7 453.446 L143.103 453.446 L143.103 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M174.491 449.511 L192.848 449.511 L192.848 453.446 L178.774 453.446 L178.774 461.918 Q179.792 461.571 180.811 461.409 Q181.829 461.224 182.848 461.224 Q188.635 461.224 192.014 464.395 Q195.394 467.567 195.394 472.983 Q195.394 478.562 191.922 481.664 Q188.45 484.742 182.13 484.742 Q179.954 484.742 177.686 484.372 Q175.44 484.002 173.033 483.261 L173.033 478.562 Q175.116 479.696 177.339 480.252 Q179.561 480.807 182.038 480.807 Q186.042 480.807 188.38 478.701 Q190.718 476.594 190.718 472.983 Q190.718 469.372 188.38 467.266 Q186.042 465.159 182.038 465.159 Q180.163 465.159 178.288 465.576 Q176.436 465.993 174.491 466.872 L174.491 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M214.607 452.59 Q210.996 452.59 209.167 456.155 Q207.362 459.696 207.362 466.826 Q207.362 473.932 209.167 477.497 Q210.996 481.039 214.607 481.039 Q218.241 481.039 220.047 477.497 Q221.875 473.932 221.875 466.826 Q221.875 459.696 220.047 456.155 Q218.241 452.59 214.607 452.59 M214.607 448.886 Q220.417 448.886 223.473 453.493 Q226.551 458.076 226.551 466.826 Q226.551 475.553 223.473 480.159 Q220.417 484.742 214.607 484.742 Q208.797 484.742 205.718 480.159 Q202.662 475.553 202.662 466.826 Q202.662 458.076 205.718 453.493 Q208.797 448.886 214.607 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M114.931 173.601 L122.57 173.601 L122.57 147.236 L114.26 148.902 L114.26 144.643 L122.524 142.976 L127.2 142.976 L127.2 173.601 L134.839 173.601 L134.839 177.536 L114.931 177.536 L114.931 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M154.283 146.055 Q150.672 146.055 148.843 149.62 Q147.038 153.161 147.038 160.291 Q147.038 167.397 148.843 170.962 Q150.672 174.504 154.283 174.504 Q157.917 174.504 159.723 170.962 Q161.552 167.397 161.552 160.291 Q161.552 153.161 159.723 149.62 Q157.917 146.055 154.283 146.055 M154.283 142.351 Q160.093 142.351 163.149 146.958 Q166.227 151.541 166.227 160.291 Q166.227 169.018 163.149 173.624 Q160.093 178.208 154.283 178.208 Q148.473 178.208 145.394 173.624 Q142.339 169.018 142.339 160.291 Q142.339 151.541 145.394 146.958 Q148.473 142.351 154.283 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M184.445 146.055 Q180.834 146.055 179.005 149.62 Q177.2 153.161 177.2 160.291 Q177.2 167.397 179.005 170.962 Q180.834 174.504 184.445 174.504 Q188.079 174.504 189.885 170.962 Q191.713 167.397 191.713 160.291 Q191.713 153.161 189.885 149.62 Q188.079 146.055 184.445 146.055 M184.445 142.351 Q190.255 142.351 193.311 146.958 Q196.389 151.541 196.389 160.291 Q196.389 169.018 193.311 173.624 Q190.255 178.208 184.445 178.208 Q178.635 178.208 175.556 173.624 Q172.501 169.018 172.501 160.291 Q172.501 151.541 175.556 146.958 Q178.635 142.351 184.445 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M214.607 146.055 Q210.996 146.055 209.167 149.62 Q207.362 153.161 207.362 160.291 Q207.362 167.397 209.167 170.962 Q210.996 174.504 214.607 174.504 Q218.241 174.504 220.047 170.962 Q221.875 167.397 221.875 160.291 Q221.875 153.161 220.047 149.62 Q218.241 146.055 214.607 146.055 M214.607 142.351 Q220.417 142.351 223.473 146.958 Q226.551 151.541 226.551 160.291 Q226.551 169.018 223.473 173.624 Q220.417 178.208 214.607 178.208 Q208.797 178.208 205.718 173.624 Q202.662 169.018 202.662 160.291 Q202.662 151.541 205.718 146.958 Q208.797 142.351 214.607 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M21.7677 1000.18 L58.7206 1000.18 L58.7206 992.418 Q58.7206 982.583 54.2646 978.031 Q49.8086 973.448 40.1964 973.448 Q30.6479 973.448 26.2237 978.031 Q21.7677 982.583 21.7677 992.418 L21.7677 1000.18 M16.4842 1006.61 L16.4842 993.404 Q16.4842 979.591 22.2451 973.13 Q27.9743 966.668 40.1964 966.668 Q52.4822 966.668 58.2432 973.161 Q64.0042 979.654 64.0042 993.404 L64.0042 1006.61 L16.4842 1006.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M28.3562 956.674 L28.3562 950.818 L64.0042 950.818 L64.0042 956.674 L28.3562 956.674 M14.479 956.674 L14.479 950.818 L21.895 950.818 L21.895 956.674 L14.479 956.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M29.4065 915.838 L34.9447 915.838 Q33.6716 918.321 33.035 920.994 Q32.3984 923.668 32.3984 926.533 Q32.3984 930.893 33.7352 933.089 Q35.072 935.254 37.7456 935.254 Q39.7826 935.254 40.9603 933.694 Q42.1061 932.134 43.1565 927.424 L43.6021 925.419 Q44.9389 919.18 47.3897 916.57 Q49.8086 913.929 54.1691 913.929 Q59.1344 913.929 62.0308 917.875 Q64.9272 921.79 64.9272 928.665 Q64.9272 931.53 64.3543 934.649 Q63.8132 937.736 62.6992 941.174 L56.6518 941.174 Q58.3387 937.927 59.198 934.776 Q60.0256 931.625 60.0256 928.538 Q60.0256 924.4 58.6251 922.172 Q57.1929 919.944 54.6147 919.944 Q52.2276 919.944 50.9545 921.567 Q49.6813 923.159 48.5037 928.601 L48.0262 930.639 Q46.8804 936.081 44.5251 938.5 Q42.138 940.919 38.0002 940.919 Q32.9713 940.919 30.2341 937.354 Q27.4968 933.79 27.4968 927.233 Q27.4968 923.986 27.9743 921.122 Q28.4517 918.257 29.4065 915.838 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M58.657 898.937 L77.5631 898.937 L77.5631 904.826 L28.3562 904.826 L28.3562 898.937 L33.7671 898.937 Q30.5842 897.091 29.0564 894.29 Q27.4968 891.458 27.4968 887.543 Q27.4968 881.05 32.6531 877.007 Q37.8093 872.933 46.212 872.933 Q54.6147 872.933 59.771 877.007 Q64.9272 881.05 64.9272 887.543 Q64.9272 891.458 63.3994 894.29 Q61.8398 897.091 58.657 898.937 M46.212 879.013 Q39.7508 879.013 36.0905 881.686 Q32.3984 884.328 32.3984 888.975 Q32.3984 893.622 36.0905 896.296 Q39.7508 898.937 46.212 898.937 Q52.6732 898.937 56.3653 896.296 Q60.0256 893.622 60.0256 888.975 Q60.0256 884.328 56.3653 881.686 Q52.6732 879.013 46.212 879.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M46.0847 847.025 Q46.0847 854.123 47.7079 856.86 Q49.3312 859.597 53.2461 859.597 Q56.3653 859.597 58.2114 857.56 Q60.0256 855.491 60.0256 851.958 Q60.0256 847.089 56.5881 844.16 Q53.1188 841.2 47.3897 841.2 L46.0847 841.2 L46.0847 847.025 M43.6657 835.344 L64.0042 835.344 L64.0042 841.2 L58.5933 841.2 Q61.8398 843.206 63.3994 846.197 Q64.9272 849.189 64.9272 853.518 Q64.9272 858.992 61.8716 862.239 Q58.7843 865.454 53.6281 865.454 Q47.6125 865.454 44.5569 861.443 Q41.5014 857.401 41.5014 849.412 L41.5014 841.2 L40.9285 841.2 Q36.8862 841.2 34.6901 843.874 Q32.4621 846.516 32.4621 851.322 Q32.4621 854.377 33.1941 857.274 Q33.9262 860.17 35.3903 862.844 L29.9795 862.844 Q28.7381 859.629 28.1334 856.605 Q27.4968 853.582 27.4968 850.717 Q27.4968 842.983 31.5072 839.163 Q35.5176 835.344 43.6657 835.344 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M18.2347 817.488 L28.3562 817.488 L28.3562 805.425 L32.9077 805.425 L32.9077 817.488 L52.2594 817.488 Q56.6199 817.488 57.8613 816.31 Q59.1026 815.101 59.1026 811.441 L59.1026 805.425 L64.0042 805.425 L64.0042 811.441 Q64.0042 818.22 61.4897 820.798 Q58.9434 823.376 52.2594 823.376 L32.9077 823.376 L32.9077 827.673 L28.3562 827.673 L28.3562 823.376 L18.2347 823.376 L18.2347 817.488 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M29.7248 772.069 L35.1993 772.069 Q33.8307 774.551 33.1623 777.066 Q32.4621 779.548 32.4621 782.095 Q32.4621 787.792 36.0905 790.943 Q39.6872 794.094 46.212 794.094 Q52.7369 794.094 56.3653 790.943 Q59.9619 787.792 59.9619 782.095 Q59.9619 779.548 59.2935 777.066 Q58.5933 774.551 57.2247 772.069 L62.6355 772.069 Q63.7814 774.52 64.3543 777.161 Q64.9272 779.771 64.9272 782.731 Q64.9272 790.784 59.8664 795.526 Q54.8057 800.269 46.212 800.269 Q37.491 800.269 32.4939 795.495 Q27.4968 790.688 27.4968 782.349 Q27.4968 779.644 28.0697 777.066 Q28.6108 774.488 29.7248 772.069 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M42.4881 732.251 L64.0042 732.251 L64.0042 738.108 L42.679 738.108 Q37.6183 738.108 35.1038 740.081 Q32.5894 742.054 32.5894 746.001 Q32.5894 750.744 35.6131 753.481 Q38.6368 756.218 43.8567 756.218 L64.0042 756.218 L64.0042 762.106 L14.479 762.106 L14.479 756.218 L33.8944 756.218 Q30.6797 754.117 29.0883 751.285 Q27.4968 748.42 27.4968 744.696 Q27.4968 738.553 31.3163 735.402 Q35.1038 732.251 42.4881 732.251 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M14.479 700.391 L14.479 686.896 L19.0304 686.896 L19.0304 694.534 L68.0464 694.534 L68.0464 686.896 L72.5979 686.896 L72.5979 700.391 L14.479 700.391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M16.4842 674.164 L16.4842 664.584 L48.8219 652.457 L16.4842 640.267 L16.4842 630.686 L64.0042 630.686 L64.0042 636.957 L22.277 636.957 L54.8694 649.211 L54.8694 655.672 L22.277 667.926 L64.0042 667.926 L64.0042 674.164 L16.4842 674.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M16.4842 622.156 L16.4842 615.663 L56.6518 605.669 L16.4842 595.707 L16.4842 588.482 L56.6518 578.488 L16.4842 568.525 L16.4842 562 L64.0042 573.936 L64.0042 582.021 L22.7544 592.047 L64.0042 602.168 L64.0042 610.252 L16.4842 622.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M14.479 540.039 L72.5979 540.039 L72.5979 553.534 L68.0464 553.534 L68.0464 545.927 L19.0304 545.927 L19.0304 553.534 L14.479 553.534 L14.479 540.039 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M351.961 12.096 L403.124 12.096 L403.124 18.9825 L381.654 18.9825 L381.654 72.576 L373.431 72.576 L373.431 18.9825 L351.961 18.9825 L351.961 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M448.413 45.1919 L448.413 72.576 L440.959 72.576 L440.959 45.4349 Q440.959 38.994 438.448 35.7938 Q435.936 32.5936 430.913 32.5936 Q424.877 32.5936 421.394 36.4419 Q417.91 40.2903 417.91 46.9338 L417.91 72.576 L410.416 72.576 L410.416 9.54393 L417.91 9.54393 L417.91 34.2544 Q420.583 30.163 424.189 28.1376 Q427.834 26.1121 432.574 26.1121 Q440.392 26.1121 444.403 30.9732 Q448.413 35.7938 448.413 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M502.088 48.0275 L502.088 51.6733 L467.817 51.6733 Q468.303 59.3701 472.435 63.421 Q476.607 67.4314 484.021 67.4314 Q488.314 67.4314 492.325 66.3781 Q496.376 65.3249 500.346 63.2184 L500.346 70.267 Q496.335 71.9684 492.122 72.8596 Q487.909 73.7508 483.575 73.7508 Q472.718 73.7508 466.359 67.4314 Q460.039 61.1119 460.039 50.3365 Q460.039 39.1965 466.034 32.6746 Q472.07 26.1121 482.279 26.1121 Q491.434 26.1121 496.74 32.0264 Q502.088 37.9003 502.088 48.0275 M494.634 45.84 Q494.553 39.7232 491.191 36.0774 Q487.869 32.4315 482.36 32.4315 Q476.121 32.4315 472.354 35.9558 Q468.627 39.4801 468.06 45.8805 L494.634 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M540.612 34.1734 Q539.356 33.4443 537.857 33.1202 Q536.399 32.7556 534.616 32.7556 Q528.297 32.7556 524.894 36.8875 Q521.532 40.9789 521.532 48.6757 L521.532 72.576 L514.038 72.576 L514.038 27.2059 L521.532 27.2059 L521.532 34.2544 Q523.881 30.1225 527.649 28.1376 Q531.416 26.1121 536.804 26.1121 Q537.573 26.1121 538.505 26.2337 Q539.437 26.3147 540.571 26.5172 L540.612 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M582.295 35.9153 Q585.091 30.8922 588.979 28.5022 Q592.868 26.1121 598.134 26.1121 Q605.224 26.1121 609.072 31.0947 Q612.92 36.0368 612.92 45.1919 L612.92 72.576 L605.426 72.576 L605.426 45.4349 Q605.426 38.913 603.117 35.7533 Q600.808 32.5936 596.069 32.5936 Q590.276 32.5936 586.913 36.4419 Q583.551 40.2903 583.551 46.9338 L583.551 72.576 L576.057 72.576 L576.057 45.4349 Q576.057 38.8725 573.748 35.7533 Q571.439 32.5936 566.618 32.5936 Q560.907 32.5936 557.544 36.4824 Q554.182 40.3308 554.182 46.9338 L554.182 72.576 L546.688 72.576 L546.688 27.2059 L554.182 27.2059 L554.182 34.2544 Q556.734 30.082 560.299 28.0971 Q563.864 26.1121 568.765 26.1121 Q573.708 26.1121 577.151 28.6237 Q580.635 31.1352 582.295 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M648.406 49.7694 Q639.373 49.7694 635.889 51.8354 Q632.405 53.9013 632.405 58.8839 Q632.405 62.8538 634.998 65.2034 Q637.631 67.5124 642.127 67.5124 Q648.325 67.5124 652.052 63.1374 Q655.819 58.7219 655.819 51.4303 L655.819 49.7694 L648.406 49.7694 M663.273 46.6907 L663.273 72.576 L655.819 72.576 L655.819 65.6895 Q653.267 69.8214 649.459 71.8063 Q645.652 73.7508 640.142 73.7508 Q633.175 73.7508 629.043 69.8619 Q624.951 65.9325 624.951 59.3701 Q624.951 51.7138 630.056 47.825 Q635.2 43.9361 645.368 43.9361 L655.819 43.9361 L655.819 43.2069 Q655.819 38.0623 652.417 35.2672 Q649.054 32.4315 642.937 32.4315 Q639.049 32.4315 635.362 33.3632 Q631.676 34.295 628.273 36.1584 L628.273 29.2718 Q632.365 27.692 636.213 26.9223 Q640.061 26.1121 643.707 26.1121 Q653.551 26.1121 658.412 31.2163 Q663.273 36.3204 663.273 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M678.626 9.54393 L686.08 9.54393 L686.08 72.576 L678.626 72.576 L678.626 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M736.554 18.8205 L736.554 65.8515 L746.438 65.8515 Q758.955 65.8515 764.748 60.1802 Q770.581 54.509 770.581 42.2752 Q770.581 30.1225 764.748 24.4918 Q758.955 18.8205 746.438 18.8205 L736.554 18.8205 M728.371 12.096 L745.182 12.096 Q762.763 12.096 770.987 19.4281 Q779.21 26.7198 779.21 42.2752 Q779.21 57.9117 770.946 65.2439 Q762.682 72.576 745.182 72.576 L728.371 72.576 L728.371 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M791.93 27.2059 L799.383 27.2059 L799.383 72.576 L791.93 72.576 L791.93 27.2059 M791.93 9.54393 L799.383 9.54393 L799.383 18.9825 L791.93 18.9825 L791.93 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M843.903 28.5427 L843.903 35.5912 Q840.743 33.9709 837.34 33.1607 Q833.938 32.3505 830.292 32.3505 Q824.742 32.3505 821.947 34.0519 Q819.192 35.7533 819.192 39.156 Q819.192 41.7486 821.177 43.2475 Q823.162 44.7058 829.158 46.0426 L831.71 46.6097 Q839.649 48.3111 842.971 51.4303 Q846.333 54.509 846.333 60.0587 Q846.333 66.3781 841.31 70.0644 Q836.328 73.7508 827.578 73.7508 Q823.932 73.7508 819.962 73.0216 Q816.033 72.3329 811.658 70.9151 L811.658 63.2184 Q815.79 65.3654 819.8 66.4591 Q823.81 67.5124 827.74 67.5124 Q833.006 67.5124 835.842 65.73 Q838.677 63.9071 838.677 60.6258 Q838.677 57.5877 836.611 55.9673 Q834.586 54.3469 827.659 52.8481 L825.066 52.2405 Q818.139 50.7821 815.06 47.7845 Q811.982 44.7463 811.982 39.4801 Q811.982 33.0797 816.519 29.5959 Q821.056 26.1121 829.401 26.1121 Q833.533 26.1121 837.178 26.7198 Q840.824 27.3274 843.903 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M865.413 65.7705 L865.413 89.8329 L857.919 89.8329 L857.919 27.2059 L865.413 27.2059 L865.413 34.0924 Q867.763 30.0415 871.327 28.0971 Q874.933 26.1121 879.915 26.1121 Q888.179 26.1121 893.324 32.6746 Q898.509 39.2371 898.509 49.9314 Q898.509 60.6258 893.324 67.1883 Q888.179 73.7508 879.915 73.7508 Q874.933 73.7508 871.327 71.8063 Q867.763 69.8214 865.413 65.7705 M890.772 49.9314 Q890.772 41.7081 887.369 37.0496 Q884.007 32.3505 878.092 32.3505 Q872.178 32.3505 868.775 37.0496 Q865.413 41.7081 865.413 49.9314 Q865.413 58.1548 868.775 62.8538 Q872.178 67.5124 878.092 67.5124 Q884.007 67.5124 887.369 62.8538 Q890.772 58.1548 890.772 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M931.483 49.7694 Q922.45 49.7694 918.966 51.8354 Q915.482 53.9013 915.482 58.8839 Q915.482 62.8538 918.075 65.2034 Q920.708 67.5124 925.204 67.5124 Q931.402 67.5124 935.129 63.1374 Q938.897 58.7219 938.897 51.4303 L938.897 49.7694 L931.483 49.7694 M946.35 46.6907 L946.35 72.576 L938.897 72.576 L938.897 65.6895 Q936.344 69.8214 932.537 71.8063 Q928.729 73.7508 923.22 73.7508 Q916.252 73.7508 912.12 69.8619 Q908.029 65.9325 908.029 59.3701 Q908.029 51.7138 913.133 47.825 Q918.277 43.9361 928.445 43.9361 L938.897 43.9361 L938.897 43.2069 Q938.897 38.0623 935.494 35.2672 Q932.132 32.4315 926.015 32.4315 Q922.126 32.4315 918.439 33.3632 Q914.753 34.295 911.35 36.1584 L911.35 29.2718 Q915.442 27.692 919.29 26.9223 Q923.139 26.1121 926.784 26.1121 Q936.628 26.1121 941.489 31.2163 Q946.35 36.3204 946.35 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M969.076 14.324 L969.076 27.2059 L984.429 27.2059 L984.429 32.9987 L969.076 32.9987 L969.076 57.6282 Q969.076 63.1779 970.575 64.7578 Q972.114 66.3376 976.772 66.3376 L984.429 66.3376 L984.429 72.576 L976.772 72.576 Q968.144 72.576 964.863 69.3758 Q961.582 66.1351 961.582 57.6282 L961.582 32.9987 L956.113 32.9987 L956.113 27.2059 L961.582 27.2059 L961.582 14.324 L969.076 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1026.88 28.9478 L1026.88 35.9153 Q1023.72 34.1734 1020.52 33.3227 Q1017.36 32.4315 1014.12 32.4315 Q1006.87 32.4315 1002.86 37.0496 Q998.85 41.6271 998.85 49.9314 Q998.85 58.2358 1002.86 62.8538 Q1006.87 67.4314 1014.12 67.4314 Q1017.36 67.4314 1020.52 66.5807 Q1023.72 65.6895 1026.88 63.9476 L1026.88 70.8341 Q1023.76 72.2924 1020.4 73.0216 Q1017.08 73.7508 1013.31 73.7508 Q1003.06 73.7508 997.027 67.3098 Q990.991 60.8689 990.991 49.9314 Q990.991 38.832 997.068 32.472 Q1003.18 26.1121 1013.8 26.1121 Q1017.24 26.1121 1020.52 26.8413 Q1023.8 27.5299 1026.88 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1077.56 45.1919 L1077.56 72.576 L1070.11 72.576 L1070.11 45.4349 Q1070.11 38.994 1067.59 35.7938 Q1065.08 32.5936 1060.06 32.5936 Q1054.02 32.5936 1050.54 36.4419 Q1047.06 40.2903 1047.06 46.9338 L1047.06 72.576 L1039.56 72.576 L1039.56 9.54393 L1047.06 9.54393 L1047.06 34.2544 Q1049.73 30.163 1053.33 28.1376 Q1056.98 26.1121 1061.72 26.1121 Q1069.54 26.1121 1073.55 30.9732 Q1077.56 35.7938 1077.56 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip472)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="288.148,1386.4 359.25,1202.47 430.352,1018.55 501.454,834.633 572.556,650.712 643.658,466.791 714.76,282.87 785.862,344.177 856.964,160.256 928.066,160.256 999.168,160.256 1070.27,160.256 1141.37,160.256 "/>
+<polyline clip-path="url(#clip472)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="288.148,1018.55 359.25,1018.55 430.352,1018.55 501.454,1018.55 572.556,1018.55 643.658,1018.55 714.76,1018.55 785.862,1018.55 856.964,1018.55 928.066,834.633 999.168,650.712 1070.27,466.791 1141.37,282.87 "/>
+<path clip-path="url(#clip470)" d="M986.519 1379.86 L1136.82 1379.86 L1136.82 1224.34 L986.519 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="986.519,1379.86 1136.82,1379.86 1136.82,1224.34 986.519,1224.34 986.519,1379.86 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="996.568,1276.18 1056.86,1276.18 "/>
+<path clip-path="url(#clip470)" d="M1092.47 1288.53 L1092.47 1279.24 L1084.83 1279.24 L1084.83 1275.4 L1097.1 1275.4 L1097.1 1290.24 Q1094.39 1292.16 1091.12 1293.16 Q1087.86 1294.13 1084.16 1294.13 Q1076.06 1294.13 1071.47 1289.41 Q1066.91 1284.66 1066.91 1276.21 Q1066.91 1267.74 1071.47 1263.02 Q1076.06 1258.27 1084.16 1258.27 Q1087.54 1258.27 1090.57 1259.1 Q1093.62 1259.94 1096.19 1261.56 L1096.19 1266.54 Q1093.6 1264.34 1090.68 1263.22 Q1087.77 1262.11 1084.55 1262.11 Q1078.21 1262.11 1075.01 1265.66 Q1071.84 1269.2 1071.84 1276.21 Q1071.84 1283.2 1075.01 1286.74 Q1078.21 1290.28 1084.55 1290.28 Q1087.03 1290.28 1088.97 1289.87 Q1090.92 1289.43 1092.47 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1106.87 1289.52 L1114.5 1289.52 L1114.5 1263.16 L1106.19 1264.82 L1106.19 1260.56 L1114.46 1258.9 L1119.13 1258.9 L1119.13 1289.52 L1126.77 1289.52 L1126.77 1293.46 L1106.87 1293.46 L1106.87 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip470)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="996.568,1328.02 1056.86,1328.02 "/>
+<path clip-path="url(#clip470)" d="M1092.47 1340.37 L1092.47 1331.08 L1084.83 1331.08 L1084.83 1327.24 L1097.1 1327.24 L1097.1 1342.08 Q1094.39 1344 1091.12 1345 Q1087.86 1345.97 1084.16 1345.97 Q1076.06 1345.97 1071.47 1341.25 Q1066.91 1336.5 1066.91 1328.05 Q1066.91 1319.58 1071.47 1314.86 Q1076.06 1310.11 1084.16 1310.11 Q1087.54 1310.11 1090.57 1310.94 Q1093.62 1311.78 1096.19 1313.4 L1096.19 1318.38 Q1093.6 1316.18 1090.68 1315.06 Q1087.77 1313.95 1084.55 1313.95 Q1078.21 1313.95 1075.01 1317.5 Q1071.84 1321.04 1071.84 1328.05 Q1071.84 1335.04 1075.01 1338.58 Q1078.21 1342.12 1084.55 1342.12 Q1087.03 1342.12 1088.97 1341.71 Q1090.92 1341.27 1092.47 1340.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1110.08 1341.36 L1126.4 1341.36 L1126.4 1345.3 L1104.46 1345.3 L1104.46 1341.36 Q1107.12 1338.61 1111.7 1333.98 Q1116.31 1329.32 1117.49 1327.98 Q1119.74 1325.46 1120.62 1323.72 Q1121.52 1321.96 1121.52 1320.27 Q1121.52 1317.52 1119.57 1315.78 Q1117.65 1314.05 1114.55 1314.05 Q1112.35 1314.05 1109.9 1314.81 Q1107.47 1315.57 1104.69 1317.13 L1104.69 1312.4 Q1107.51 1311.27 1109.97 1310.69 Q1112.42 1310.11 1114.46 1310.11 Q1119.83 1310.11 1123.02 1312.8 Q1126.22 1315.48 1126.22 1319.97 Q1126.22 1322.1 1125.41 1324.02 Q1124.62 1325.92 1122.51 1328.51 Q1121.93 1329.19 1118.83 1332.4 Q1115.73 1335.6 1110.08 1341.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1448.34 1423.18 L2352.76 1423.18 L2352.76 123.472 L1448.34 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip653">
+  <clipPath id="clip473">
     <rect x="1448" y="123" width="905" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1568.74,1423.18 1568.74,123.472 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1805.74,1423.18 1805.74,123.472 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2042.75,1423.18 2042.75,123.472 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2279.76,1423.18 2279.76,123.472 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1568.74,1423.18 1568.74,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1805.74,1423.18 1805.74,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2042.75,1423.18 2042.75,1404.28 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2279.76,1423.18 2279.76,1404.28 "/>
-<path clip-path="url(#clip650)" d="M1528.36 1451.02 L1546.71 1451.02 L1546.71 1454.96 L1532.64 1454.96 L1532.64 1463.43 Q1533.66 1463.08 1534.68 1462.92 Q1535.69 1462.73 1536.71 1462.73 Q1542.5 1462.73 1545.88 1465.9 Q1549.26 1469.08 1549.26 1474.49 Q1549.26 1480.07 1545.79 1483.17 Q1542.31 1486.25 1535.99 1486.25 Q1533.82 1486.25 1531.55 1485.88 Q1529.31 1485.51 1526.9 1484.77 L1526.9 1480.07 Q1528.98 1481.2 1531.2 1481.76 Q1533.43 1482.32 1535.9 1482.32 Q1539.91 1482.32 1542.24 1480.21 Q1544.58 1478.1 1544.58 1474.49 Q1544.58 1470.88 1542.24 1468.77 Q1539.91 1466.67 1535.9 1466.67 Q1534.03 1466.67 1532.15 1467.08 Q1530.3 1467.5 1528.36 1468.38 L1528.36 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1568.47 1454.1 Q1564.86 1454.1 1563.03 1457.66 Q1561.23 1461.2 1561.23 1468.33 Q1561.23 1475.44 1563.03 1479.01 Q1564.86 1482.55 1568.47 1482.55 Q1572.11 1482.55 1573.91 1479.01 Q1575.74 1475.44 1575.74 1468.33 Q1575.74 1461.2 1573.91 1457.66 Q1572.11 1454.1 1568.47 1454.1 M1568.47 1450.39 Q1574.28 1450.39 1577.34 1455 Q1580.42 1459.58 1580.42 1468.33 Q1580.42 1477.06 1577.34 1481.67 Q1574.28 1486.25 1568.47 1486.25 Q1562.66 1486.25 1559.58 1481.67 Q1556.53 1477.06 1556.53 1468.33 Q1556.53 1459.58 1559.58 1455 Q1562.66 1450.39 1568.47 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1598.63 1454.1 Q1595.02 1454.1 1593.19 1457.66 Q1591.39 1461.2 1591.39 1468.33 Q1591.39 1475.44 1593.19 1479.01 Q1595.02 1482.55 1598.63 1482.55 Q1602.27 1482.55 1604.07 1479.01 Q1605.9 1475.44 1605.9 1468.33 Q1605.9 1461.2 1604.07 1457.66 Q1602.27 1454.1 1598.63 1454.1 M1598.63 1450.39 Q1604.44 1450.39 1607.5 1455 Q1610.58 1459.58 1610.58 1468.33 Q1610.58 1477.06 1607.5 1481.67 Q1604.44 1486.25 1598.63 1486.25 Q1592.82 1486.25 1589.74 1481.67 Q1586.69 1477.06 1586.69 1468.33 Q1586.69 1459.58 1589.74 1455 Q1592.82 1450.39 1598.63 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1750.27 1481.64 L1757.91 1481.64 L1757.91 1455.28 L1749.6 1456.95 L1749.6 1452.69 L1757.86 1451.02 L1762.54 1451.02 L1762.54 1481.64 L1770.18 1481.64 L1770.18 1485.58 L1750.27 1485.58 L1750.27 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1789.62 1454.1 Q1786.01 1454.1 1784.18 1457.66 Q1782.38 1461.2 1782.38 1468.33 Q1782.38 1475.44 1784.18 1479.01 Q1786.01 1482.55 1789.62 1482.55 Q1793.26 1482.55 1795.06 1479.01 Q1796.89 1475.44 1796.89 1468.33 Q1796.89 1461.2 1795.06 1457.66 Q1793.26 1454.1 1789.62 1454.1 M1789.62 1450.39 Q1795.43 1450.39 1798.49 1455 Q1801.57 1459.58 1801.57 1468.33 Q1801.57 1477.06 1798.49 1481.67 Q1795.43 1486.25 1789.62 1486.25 Q1783.81 1486.25 1780.73 1481.67 Q1777.68 1477.06 1777.68 1468.33 Q1777.68 1459.58 1780.73 1455 Q1783.81 1450.39 1789.62 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1819.78 1454.1 Q1816.17 1454.1 1814.34 1457.66 Q1812.54 1461.2 1812.54 1468.33 Q1812.54 1475.44 1814.34 1479.01 Q1816.17 1482.55 1819.78 1482.55 Q1823.42 1482.55 1825.22 1479.01 Q1827.05 1475.44 1827.05 1468.33 Q1827.05 1461.2 1825.22 1457.66 Q1823.42 1454.1 1819.78 1454.1 M1819.78 1450.39 Q1825.59 1450.39 1828.65 1455 Q1831.73 1459.58 1831.73 1468.33 Q1831.73 1477.06 1828.65 1481.67 Q1825.59 1486.25 1819.78 1486.25 Q1813.97 1486.25 1810.89 1481.67 Q1807.84 1477.06 1807.84 1468.33 Q1807.84 1459.58 1810.89 1455 Q1813.97 1450.39 1819.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1849.95 1454.1 Q1846.33 1454.1 1844.51 1457.66 Q1842.7 1461.2 1842.7 1468.33 Q1842.7 1475.44 1844.51 1479.01 Q1846.33 1482.55 1849.95 1482.55 Q1853.58 1482.55 1855.39 1479.01 Q1857.21 1475.44 1857.21 1468.33 Q1857.21 1461.2 1855.39 1457.66 Q1853.58 1454.1 1849.95 1454.1 M1849.95 1450.39 Q1855.76 1450.39 1858.81 1455 Q1861.89 1459.58 1861.89 1468.33 Q1861.89 1477.06 1858.81 1481.67 Q1855.76 1486.25 1849.95 1486.25 Q1844.14 1486.25 1841.06 1481.67 Q1838 1477.06 1838 1468.33 Q1838 1459.58 1841.06 1455 Q1844.14 1450.39 1849.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1987.28 1481.64 L1994.92 1481.64 L1994.92 1455.28 L1986.61 1456.95 L1986.61 1452.69 L1994.87 1451.02 L1999.55 1451.02 L1999.55 1481.64 L2007.18 1481.64 L2007.18 1485.58 L1987.28 1485.58 L1987.28 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2016.67 1451.02 L2035.03 1451.02 L2035.03 1454.96 L2020.96 1454.96 L2020.96 1463.43 Q2021.98 1463.08 2022.99 1462.92 Q2024.01 1462.73 2025.03 1462.73 Q2030.82 1462.73 2034.2 1465.9 Q2037.58 1469.08 2037.58 1474.49 Q2037.58 1480.07 2034.11 1483.17 Q2030.63 1486.25 2024.31 1486.25 Q2022.14 1486.25 2019.87 1485.88 Q2017.62 1485.51 2015.22 1484.77 L2015.22 1480.07 Q2017.3 1481.2 2019.52 1481.76 Q2021.74 1482.32 2024.22 1482.32 Q2028.23 1482.32 2030.56 1480.21 Q2032.9 1478.1 2032.9 1474.49 Q2032.9 1470.88 2030.56 1468.77 Q2028.23 1466.67 2024.22 1466.67 Q2022.35 1466.67 2020.47 1467.08 Q2018.62 1467.5 2016.67 1468.38 L2016.67 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2056.79 1454.1 Q2053.18 1454.1 2051.35 1457.66 Q2049.55 1461.2 2049.55 1468.33 Q2049.55 1475.44 2051.35 1479.01 Q2053.18 1482.55 2056.79 1482.55 Q2060.42 1482.55 2062.23 1479.01 Q2064.06 1475.44 2064.06 1468.33 Q2064.06 1461.2 2062.23 1457.66 Q2060.42 1454.1 2056.79 1454.1 M2056.79 1450.39 Q2062.6 1450.39 2065.66 1455 Q2068.73 1459.58 2068.73 1468.33 Q2068.73 1477.06 2065.66 1481.67 Q2062.6 1486.25 2056.79 1486.25 Q2050.98 1486.25 2047.9 1481.67 Q2044.85 1477.06 2044.85 1468.33 Q2044.85 1459.58 2047.9 1455 Q2050.98 1450.39 2056.79 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2086.95 1454.1 Q2083.34 1454.1 2081.51 1457.66 Q2079.71 1461.2 2079.71 1468.33 Q2079.71 1475.44 2081.51 1479.01 Q2083.34 1482.55 2086.95 1482.55 Q2090.59 1482.55 2092.39 1479.01 Q2094.22 1475.44 2094.22 1468.33 Q2094.22 1461.2 2092.39 1457.66 Q2090.59 1454.1 2086.95 1454.1 M2086.95 1450.39 Q2092.76 1450.39 2095.82 1455 Q2098.9 1459.58 2098.9 1468.33 Q2098.9 1477.06 2095.82 1481.67 Q2092.76 1486.25 2086.95 1486.25 Q2081.14 1486.25 2078.06 1481.67 Q2075.01 1477.06 2075.01 1468.33 Q2075.01 1459.58 2078.06 1455 Q2081.14 1450.39 2086.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2228.37 1481.64 L2244.69 1481.64 L2244.69 1485.58 L2222.74 1485.58 L2222.74 1481.64 Q2225.41 1478.89 2229.99 1474.26 Q2234.6 1469.61 2235.78 1468.27 Q2238.02 1465.74 2238.9 1464.01 Q2239.8 1462.25 2239.8 1460.56 Q2239.8 1457.8 2237.86 1456.07 Q2235.94 1454.33 2232.84 1454.33 Q2230.64 1454.33 2228.18 1455.09 Q2225.75 1455.86 2222.98 1457.41 L2222.98 1452.69 Q2225.8 1451.55 2228.25 1450.97 Q2230.71 1450.39 2232.74 1450.39 Q2238.11 1450.39 2241.31 1453.08 Q2244.5 1455.77 2244.5 1460.26 Q2244.5 1462.39 2243.69 1464.31 Q2242.91 1466.2 2240.8 1468.8 Q2240.22 1469.47 2237.12 1472.69 Q2234.02 1475.88 2228.37 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2264.5 1454.1 Q2260.89 1454.1 2259.06 1457.66 Q2257.26 1461.2 2257.26 1468.33 Q2257.26 1475.44 2259.06 1479.01 Q2260.89 1482.55 2264.5 1482.55 Q2268.14 1482.55 2269.94 1479.01 Q2271.77 1475.44 2271.77 1468.33 Q2271.77 1461.2 2269.94 1457.66 Q2268.14 1454.1 2264.5 1454.1 M2264.5 1450.39 Q2270.31 1450.39 2273.37 1455 Q2276.45 1459.58 2276.45 1468.33 Q2276.45 1477.06 2273.37 1481.67 Q2270.31 1486.25 2264.5 1486.25 Q2258.69 1486.25 2255.61 1481.67 Q2252.56 1477.06 2252.56 1468.33 Q2252.56 1459.58 2255.61 1455 Q2258.69 1450.39 2264.5 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2294.67 1454.1 Q2291.05 1454.1 2289.23 1457.66 Q2287.42 1461.2 2287.42 1468.33 Q2287.42 1475.44 2289.23 1479.01 Q2291.05 1482.55 2294.67 1482.55 Q2298.3 1482.55 2300.1 1479.01 Q2301.93 1475.44 2301.93 1468.33 Q2301.93 1461.2 2300.1 1457.66 Q2298.3 1454.1 2294.67 1454.1 M2294.67 1450.39 Q2300.48 1450.39 2303.53 1455 Q2306.61 1459.58 2306.61 1468.33 Q2306.61 1477.06 2303.53 1481.67 Q2300.48 1486.25 2294.67 1486.25 Q2288.86 1486.25 2285.78 1481.67 Q2282.72 1477.06 2282.72 1468.33 Q2282.72 1459.58 2285.78 1455 Q2288.86 1450.39 2294.67 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2324.83 1454.1 Q2321.22 1454.1 2319.39 1457.66 Q2317.58 1461.2 2317.58 1468.33 Q2317.58 1475.44 2319.39 1479.01 Q2321.22 1482.55 2324.83 1482.55 Q2328.46 1482.55 2330.27 1479.01 Q2332.1 1475.44 2332.1 1468.33 Q2332.1 1461.2 2330.27 1457.66 Q2328.46 1454.1 2324.83 1454.1 M2324.83 1450.39 Q2330.64 1450.39 2333.69 1455 Q2336.77 1459.58 2336.77 1468.33 Q2336.77 1477.06 2333.69 1481.67 Q2330.64 1486.25 2324.83 1486.25 Q2319.02 1486.25 2315.94 1481.67 Q2312.88 1477.06 2312.88 1468.33 Q2312.88 1459.58 2315.94 1455 Q2319.02 1450.39 2324.83 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1678.62 1525.81 L1678.62 1562.76 L1686.39 1562.76 Q1696.22 1562.76 1700.78 1558.3 Q1705.36 1553.85 1705.36 1544.24 Q1705.36 1534.69 1700.78 1530.26 Q1696.22 1525.81 1686.39 1525.81 L1678.62 1525.81 M1672.19 1520.52 L1685.4 1520.52 Q1699.22 1520.52 1705.68 1526.28 Q1712.14 1532.01 1712.14 1544.24 Q1712.14 1556.52 1705.65 1562.28 Q1699.15 1568.04 1685.4 1568.04 L1672.19 1568.04 L1672.19 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1752.62 1548.76 L1752.62 1551.62 L1725.7 1551.62 Q1726.08 1557.67 1729.33 1560.85 Q1732.6 1564 1738.43 1564 Q1741.8 1564 1744.95 1563.17 Q1748.14 1562.35 1751.26 1560.69 L1751.26 1566.23 Q1748.1 1567.57 1744.79 1568.27 Q1741.48 1568.97 1738.08 1568.97 Q1729.55 1568.97 1724.55 1564 Q1719.59 1559.04 1719.59 1550.57 Q1719.59 1541.82 1724.3 1536.69 Q1729.04 1531.54 1737.06 1531.54 Q1744.25 1531.54 1748.42 1536.18 Q1752.62 1540.8 1752.62 1548.76 M1746.77 1547.04 Q1746.7 1542.23 1744.06 1539.37 Q1741.45 1536.5 1737.12 1536.5 Q1732.22 1536.5 1729.26 1539.27 Q1726.33 1542.04 1725.89 1547.07 L1746.77 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1789.99 1539.24 Q1792.19 1535.29 1795.24 1533.41 Q1798.3 1531.54 1802.44 1531.54 Q1808.01 1531.54 1811.03 1535.45 Q1814.05 1539.33 1814.05 1546.53 L1814.05 1568.04 L1808.16 1568.04 L1808.16 1546.72 Q1808.16 1541.59 1806.35 1539.11 Q1804.54 1536.63 1800.81 1536.63 Q1796.26 1536.63 1793.62 1539.65 Q1790.98 1542.68 1790.98 1547.9 L1790.98 1568.04 L1785.09 1568.04 L1785.09 1546.72 Q1785.09 1541.56 1783.27 1539.11 Q1781.46 1536.63 1777.67 1536.63 Q1773.19 1536.63 1770.54 1539.68 Q1767.9 1542.71 1767.9 1547.9 L1767.9 1568.04 L1762.01 1568.04 L1762.01 1532.4 L1767.9 1532.4 L1767.9 1537.93 Q1769.91 1534.66 1772.71 1533.1 Q1775.51 1531.54 1779.36 1531.54 Q1783.24 1531.54 1785.95 1533.51 Q1788.69 1535.48 1789.99 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1841.93 1550.12 Q1834.84 1550.12 1832.1 1551.75 Q1829.36 1553.37 1829.36 1557.29 Q1829.36 1560.4 1831.4 1562.25 Q1833.47 1564.07 1837 1564.07 Q1841.87 1564.07 1844.8 1560.63 Q1847.76 1557.16 1847.76 1551.43 L1847.76 1550.12 L1841.93 1550.12 M1853.62 1547.71 L1853.62 1568.04 L1847.76 1568.04 L1847.76 1562.63 Q1845.75 1565.88 1842.76 1567.44 Q1839.77 1568.97 1835.44 1568.97 Q1829.97 1568.97 1826.72 1565.91 Q1823.51 1562.82 1823.51 1557.67 Q1823.51 1551.65 1827.52 1548.6 Q1831.56 1545.54 1839.55 1545.54 L1847.76 1545.54 L1847.76 1544.97 Q1847.76 1540.93 1845.09 1538.73 Q1842.44 1536.5 1837.64 1536.5 Q1834.58 1536.5 1831.69 1537.23 Q1828.79 1537.97 1826.12 1539.43 L1826.12 1534.02 Q1829.33 1532.78 1832.35 1532.17 Q1835.38 1531.54 1838.24 1531.54 Q1845.98 1531.54 1849.8 1535.55 Q1853.62 1539.56 1853.62 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1895.31 1546.53 L1895.31 1568.04 L1889.45 1568.04 L1889.45 1546.72 Q1889.45 1541.66 1887.48 1539.14 Q1885.51 1536.63 1881.56 1536.63 Q1876.82 1536.63 1874.08 1539.65 Q1871.34 1542.68 1871.34 1547.9 L1871.34 1568.04 L1865.46 1568.04 L1865.46 1532.4 L1871.34 1532.4 L1871.34 1537.93 Q1873.45 1534.72 1876.28 1533.13 Q1879.14 1531.54 1882.87 1531.54 Q1889.01 1531.54 1892.16 1535.36 Q1895.31 1539.14 1895.31 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1930.45 1537.81 L1930.45 1518.52 L1936.31 1518.52 L1936.31 1568.04 L1930.45 1568.04 L1930.45 1562.7 Q1928.6 1565.88 1925.77 1567.44 Q1922.97 1568.97 1919.02 1568.97 Q1912.56 1568.97 1908.49 1563.81 Q1904.45 1558.65 1904.45 1550.25 Q1904.45 1541.85 1908.49 1536.69 Q1912.56 1531.54 1919.02 1531.54 Q1922.97 1531.54 1925.77 1533.1 Q1928.6 1534.62 1930.45 1537.81 M1910.49 1550.25 Q1910.49 1556.71 1913.14 1560.4 Q1915.81 1564.07 1920.46 1564.07 Q1925.1 1564.07 1927.78 1560.4 Q1930.45 1556.71 1930.45 1550.25 Q1930.45 1543.79 1927.78 1540.13 Q1925.1 1536.44 1920.46 1536.44 Q1915.81 1536.44 1913.14 1540.13 Q1910.49 1543.79 1910.49 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1968.55 1518.52 L1982.04 1518.52 L1982.04 1523.07 L1974.41 1523.07 L1974.41 1572.09 L1982.04 1572.09 L1982.04 1576.64 L1968.55 1576.64 L1968.55 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1994.78 1520.52 L2004.36 1520.52 L2016.48 1552.86 L2028.67 1520.52 L2038.25 1520.52 L2038.25 1568.04 L2031.98 1568.04 L2031.98 1526.32 L2019.73 1558.91 L2013.27 1558.91 L2001.01 1526.32 L2001.01 1568.04 L1994.78 1568.04 L1994.78 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2046.78 1520.52 L2053.28 1520.52 L2063.27 1560.69 L2073.23 1520.52 L2080.46 1520.52 L2090.45 1560.69 L2100.41 1520.52 L2106.94 1520.52 L2095 1568.04 L2086.92 1568.04 L2076.89 1526.79 L2066.77 1568.04 L2058.69 1568.04 L2046.78 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2128.9 1518.52 L2128.9 1576.64 L2115.41 1576.64 L2115.41 1572.09 L2123.01 1572.09 L2123.01 1523.07 L2115.41 1523.07 L2115.41 1518.52 L2128.9 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,1386.4 2352.76,1386.4 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,1079.86 2352.76,1079.86 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,773.326 2352.76,773.326 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,466.791 2352.76,466.791 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,160.256 2352.76,160.256 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1423.18 1448.34,123.472 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1386.4 1467.24,1386.4 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1079.86 1467.24,1079.86 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,773.326 1467.24,773.326 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,466.791 1467.24,466.791 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,160.256 1467.24,160.256 "/>
-<path clip-path="url(#clip650)" d="M1400.39 1372.19 Q1396.78 1372.19 1394.95 1375.76 Q1393.15 1379.3 1393.15 1386.43 Q1393.15 1393.54 1394.95 1397.1 Q1396.78 1400.64 1400.39 1400.64 Q1404.03 1400.64 1405.83 1397.1 Q1407.66 1393.54 1407.66 1386.43 Q1407.66 1379.3 1405.83 1375.76 Q1404.03 1372.19 1400.39 1372.19 M1400.39 1368.49 Q1406.2 1368.49 1409.26 1373.1 Q1412.34 1377.68 1412.34 1386.43 Q1412.34 1395.16 1409.26 1399.76 Q1406.2 1404.35 1400.39 1404.35 Q1394.58 1404.35 1391.51 1399.76 Q1388.45 1395.16 1388.45 1386.43 Q1388.45 1377.68 1391.51 1373.1 Q1394.58 1368.49 1400.39 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1360.28 1062.58 L1378.63 1062.58 L1378.63 1066.52 L1364.56 1066.52 L1364.56 1074.99 Q1365.58 1074.64 1366.6 1074.48 Q1367.62 1074.29 1368.63 1074.29 Q1374.42 1074.29 1377.8 1077.47 Q1381.18 1080.64 1381.18 1086.05 Q1381.18 1091.63 1377.71 1094.73 Q1374.24 1097.81 1367.92 1097.81 Q1365.74 1097.81 1363.47 1097.44 Q1361.23 1097.07 1358.82 1096.33 L1358.82 1091.63 Q1360.9 1092.77 1363.13 1093.32 Q1365.35 1093.88 1367.82 1093.88 Q1371.83 1093.88 1374.17 1091.77 Q1376.51 1089.66 1376.51 1086.05 Q1376.51 1082.44 1374.17 1080.34 Q1371.83 1078.23 1367.82 1078.23 Q1365.95 1078.23 1364.07 1078.65 Q1362.22 1079.06 1360.28 1079.94 L1360.28 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1400.39 1065.66 Q1396.78 1065.66 1394.95 1069.22 Q1393.15 1072.77 1393.15 1079.9 Q1393.15 1087 1394.95 1090.57 Q1396.78 1094.11 1400.39 1094.11 Q1404.03 1094.11 1405.83 1090.57 Q1407.66 1087 1407.66 1079.9 Q1407.66 1072.77 1405.83 1069.22 Q1404.03 1065.66 1400.39 1065.66 M1400.39 1061.96 Q1406.2 1061.96 1409.26 1066.56 Q1412.34 1071.15 1412.34 1079.9 Q1412.34 1088.62 1409.26 1093.23 Q1406.2 1097.81 1400.39 1097.81 Q1394.58 1097.81 1391.51 1093.23 Q1388.45 1088.62 1388.45 1079.9 Q1388.45 1071.15 1391.51 1066.56 Q1394.58 1061.96 1400.39 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1330.88 786.671 L1338.52 786.671 L1338.52 760.305 L1330.21 761.972 L1330.21 757.713 L1338.47 756.046 L1343.15 756.046 L1343.15 786.671 L1350.79 786.671 L1350.79 790.606 L1330.88 790.606 L1330.88 786.671 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1370.23 759.125 Q1366.62 759.125 1364.79 762.689 Q1362.99 766.231 1362.99 773.361 Q1362.99 780.467 1364.79 784.032 Q1366.62 787.574 1370.23 787.574 Q1373.87 787.574 1375.67 784.032 Q1377.5 780.467 1377.5 773.361 Q1377.5 766.231 1375.67 762.689 Q1373.87 759.125 1370.23 759.125 M1370.23 755.421 Q1376.04 755.421 1379.1 760.027 Q1382.18 764.611 1382.18 773.361 Q1382.18 782.088 1379.1 786.694 Q1376.04 791.277 1370.23 791.277 Q1364.42 791.277 1361.34 786.694 Q1358.29 782.088 1358.29 773.361 Q1358.29 764.611 1361.34 760.027 Q1364.42 755.421 1370.23 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1400.39 759.125 Q1396.78 759.125 1394.95 762.689 Q1393.15 766.231 1393.15 773.361 Q1393.15 780.467 1394.95 784.032 Q1396.78 787.574 1400.39 787.574 Q1404.03 787.574 1405.83 784.032 Q1407.66 780.467 1407.66 773.361 Q1407.66 766.231 1405.83 762.689 Q1404.03 759.125 1400.39 759.125 M1400.39 755.421 Q1406.2 755.421 1409.26 760.027 Q1412.34 764.611 1412.34 773.361 Q1412.34 782.088 1409.26 786.694 Q1406.2 791.277 1400.39 791.277 Q1394.58 791.277 1391.51 786.694 Q1388.45 782.088 1388.45 773.361 Q1388.45 764.611 1391.51 760.027 Q1394.58 755.421 1400.39 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1330.88 480.136 L1338.52 480.136 L1338.52 453.77 L1330.21 455.437 L1330.21 451.178 L1338.47 449.511 L1343.15 449.511 L1343.15 480.136 L1350.79 480.136 L1350.79 484.071 L1330.88 484.071 L1330.88 480.136 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1360.28 449.511 L1378.63 449.511 L1378.63 453.446 L1364.56 453.446 L1364.56 461.918 Q1365.58 461.571 1366.6 461.409 Q1367.62 461.224 1368.63 461.224 Q1374.42 461.224 1377.8 464.395 Q1381.18 467.567 1381.18 472.983 Q1381.18 478.562 1377.71 481.664 Q1374.24 484.742 1367.92 484.742 Q1365.74 484.742 1363.47 484.372 Q1361.23 484.002 1358.82 483.261 L1358.82 478.562 Q1360.9 479.696 1363.13 480.252 Q1365.35 480.807 1367.82 480.807 Q1371.83 480.807 1374.17 478.701 Q1376.51 476.594 1376.51 472.983 Q1376.51 469.372 1374.17 467.266 Q1371.83 465.159 1367.82 465.159 Q1365.95 465.159 1364.07 465.576 Q1362.22 465.993 1360.28 466.872 L1360.28 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1400.39 452.59 Q1396.78 452.59 1394.95 456.155 Q1393.15 459.696 1393.15 466.826 Q1393.15 473.932 1394.95 477.497 Q1396.78 481.039 1400.39 481.039 Q1404.03 481.039 1405.83 477.497 Q1407.66 473.932 1407.66 466.826 Q1407.66 459.696 1405.83 456.155 Q1404.03 452.59 1400.39 452.59 M1400.39 448.886 Q1406.2 448.886 1409.26 453.493 Q1412.34 458.076 1412.34 466.826 Q1412.34 475.553 1409.26 480.159 Q1406.2 484.742 1400.39 484.742 Q1394.58 484.742 1391.51 480.159 Q1388.45 475.553 1388.45 466.826 Q1388.45 458.076 1391.51 453.493 Q1394.58 448.886 1400.39 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1334.1 173.601 L1350.42 173.601 L1350.42 177.536 L1328.47 177.536 L1328.47 173.601 Q1331.14 170.847 1335.72 166.217 Q1340.32 161.564 1341.51 160.222 Q1343.75 157.698 1344.63 155.962 Q1345.53 154.203 1345.53 152.513 Q1345.53 149.759 1343.59 148.023 Q1341.67 146.286 1338.57 146.286 Q1336.37 146.286 1333.91 147.05 Q1331.48 147.814 1328.7 149.365 L1328.7 144.643 Q1331.53 143.509 1333.98 142.93 Q1336.44 142.351 1338.47 142.351 Q1343.84 142.351 1347.04 145.036 Q1350.23 147.722 1350.23 152.212 Q1350.23 154.342 1349.42 156.263 Q1348.63 158.161 1346.53 160.754 Q1345.95 161.425 1342.85 164.643 Q1339.75 167.837 1334.1 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1370.23 146.055 Q1366.62 146.055 1364.79 149.62 Q1362.99 153.161 1362.99 160.291 Q1362.99 167.397 1364.79 170.962 Q1366.62 174.504 1370.23 174.504 Q1373.87 174.504 1375.67 170.962 Q1377.5 167.397 1377.5 160.291 Q1377.5 153.161 1375.67 149.62 Q1373.87 146.055 1370.23 146.055 M1370.23 142.351 Q1376.04 142.351 1379.1 146.958 Q1382.18 151.541 1382.18 160.291 Q1382.18 169.018 1379.1 173.624 Q1376.04 178.208 1370.23 178.208 Q1364.42 178.208 1361.34 173.624 Q1358.29 169.018 1358.29 160.291 Q1358.29 151.541 1361.34 146.958 Q1364.42 142.351 1370.23 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1400.39 146.055 Q1396.78 146.055 1394.95 149.62 Q1393.15 153.161 1393.15 160.291 Q1393.15 167.397 1394.95 170.962 Q1396.78 174.504 1400.39 174.504 Q1404.03 174.504 1405.83 170.962 Q1407.66 167.397 1407.66 160.291 Q1407.66 153.161 1405.83 149.62 Q1404.03 146.055 1400.39 146.055 M1400.39 142.351 Q1406.2 142.351 1409.26 146.958 Q1412.34 151.541 1412.34 160.291 Q1412.34 169.018 1409.26 173.624 Q1406.2 178.208 1400.39 178.208 Q1394.58 178.208 1391.51 173.624 Q1388.45 169.018 1388.45 160.291 Q1388.45 151.541 1391.51 146.958 Q1394.58 142.351 1400.39 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1230.7 977.57 L1230.7 947.524 L1236.11 947.524 L1236.11 971.14 L1250.18 971.14 L1250.18 948.51 L1255.59 948.51 L1255.59 971.14 L1272.81 971.14 L1272.81 946.951 L1278.22 946.951 L1278.22 977.57 L1230.7 977.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1256.7 907.006 L1278.22 907.006 L1278.22 912.862 L1256.89 912.862 Q1251.83 912.862 1249.32 914.836 Q1246.8 916.809 1246.8 920.756 Q1246.8 925.498 1249.83 928.235 Q1252.85 930.973 1258.07 930.973 L1278.22 930.973 L1278.22 936.861 L1242.57 936.861 L1242.57 930.973 L1248.11 930.973 Q1244.89 928.872 1243.3 926.039 Q1241.71 923.175 1241.71 919.451 Q1241.71 913.308 1245.53 910.157 Q1249.32 907.006 1256.7 907.006 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1258.93 864.833 L1261.79 864.833 L1261.79 891.76 Q1267.84 891.378 1271.02 888.132 Q1274.17 884.853 1274.17 879.029 Q1274.17 875.655 1273.35 872.504 Q1272.52 869.321 1270.86 866.202 L1276.4 866.202 Q1277.74 869.353 1278.44 872.663 Q1279.14 875.973 1279.14 879.379 Q1279.14 887.909 1274.17 892.906 Q1269.21 897.871 1260.74 897.871 Q1251.99 897.871 1246.87 893.16 Q1241.71 888.418 1241.71 880.397 Q1241.71 873.204 1246.36 869.034 Q1250.97 864.833 1258.93 864.833 M1257.21 870.689 Q1252.4 870.753 1249.54 873.395 Q1246.67 876.005 1246.67 880.334 Q1246.67 885.235 1249.44 888.195 Q1252.21 891.123 1257.24 891.569 L1257.21 870.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1248.04 834.564 Q1247.47 835.551 1247.22 836.728 Q1246.93 837.874 1246.93 839.275 Q1246.93 844.24 1250.18 846.914 Q1253.39 849.555 1259.44 849.555 L1278.22 849.555 L1278.22 855.444 L1242.57 855.444 L1242.57 849.555 L1248.11 849.555 Q1244.86 847.709 1243.3 844.749 Q1241.71 841.789 1241.71 837.556 Q1241.71 836.951 1241.81 836.219 Q1241.87 835.487 1242.03 834.596 L1248.04 834.564 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1259.98 806.109 Q1253.61 806.109 1250.11 808.751 Q1246.61 811.361 1246.61 816.104 Q1246.61 820.814 1250.11 823.456 Q1253.61 826.066 1259.98 826.066 Q1266.31 826.066 1269.81 823.456 Q1273.32 820.814 1273.32 816.104 Q1273.32 811.361 1269.81 808.751 Q1266.31 806.109 1259.98 806.109 M1273.79 800.253 Q1282.9 800.253 1287.32 804.295 Q1291.78 808.337 1291.78 816.676 Q1291.78 819.764 1291.3 822.501 Q1290.85 825.238 1289.9 827.816 L1284.2 827.816 Q1285.6 825.238 1286.27 822.724 Q1286.94 820.209 1286.94 817.599 Q1286.94 811.839 1283.91 808.974 Q1280.92 806.109 1274.84 806.109 L1271.95 806.109 Q1275.1 807.924 1276.66 810.756 Q1278.22 813.589 1278.22 817.536 Q1278.22 824.092 1273.22 828.103 Q1268.22 832.113 1259.98 832.113 Q1251.7 832.113 1246.71 828.103 Q1241.71 824.092 1241.71 817.536 Q1241.71 813.589 1243.27 810.756 Q1244.83 807.924 1247.98 806.109 L1242.57 806.109 L1242.57 800.253 L1273.79 800.253 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1281.53 773.358 Q1287.89 775.84 1289.83 778.196 Q1291.78 780.551 1291.78 784.498 L1291.78 789.177 L1286.87 789.177 L1286.87 785.739 Q1286.87 783.32 1285.73 781.983 Q1284.58 780.647 1280.32 779.023 L1277.64 777.973 L1242.57 792.391 L1242.57 786.185 L1270.45 775.045 L1242.57 763.905 L1242.57 757.698 L1281.53 773.358 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1228.69 729.434 L1228.69 715.939 L1233.24 715.939 L1233.24 723.578 L1282.26 723.578 L1282.26 715.939 L1286.81 715.939 L1286.81 729.434 L1228.69 729.434 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1230.7 703.208 L1230.7 693.627 L1263.03 681.501 L1230.7 669.31 L1230.7 659.73 L1278.22 659.73 L1278.22 666 L1236.49 666 L1269.08 678.254 L1269.08 684.715 L1236.49 696.969 L1278.22 696.969 L1278.22 703.208 L1230.7 703.208 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1230.7 651.2 L1230.7 644.707 L1270.86 634.713 L1230.7 624.75 L1230.7 617.525 L1270.86 607.531 L1230.7 597.569 L1230.7 591.044 L1278.22 602.98 L1278.22 611.064 L1236.97 621.09 L1278.22 631.212 L1278.22 639.296 L1230.7 651.2 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1228.69 569.082 L1286.81 569.082 L1286.81 582.578 L1282.26 582.578 L1282.26 574.971 L1233.24 574.971 L1233.24 582.578 L1228.69 582.578 L1228.69 569.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1801.44 12.096 L1809.71 12.096 L1822.43 63.2184 L1835.1 12.096 L1844.3 12.096 L1857.02 63.2184 L1869.7 12.096 L1878 12.096 L1862.81 72.576 L1852.52 72.576 L1839.76 20.0763 L1826.88 72.576 L1816.59 72.576 L1801.44 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1886.71 27.2059 L1894.17 27.2059 L1894.17 72.576 L1886.71 72.576 L1886.71 27.2059 M1886.71 9.54393 L1894.17 9.54393 L1894.17 18.9825 L1886.71 18.9825 L1886.71 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1947.48 45.1919 L1947.48 72.576 L1940.02 72.576 L1940.02 45.4349 Q1940.02 38.994 1937.51 35.7938 Q1935 32.5936 1929.98 32.5936 Q1923.94 32.5936 1920.46 36.4419 Q1916.97 40.2903 1916.97 46.9338 L1916.97 72.576 L1909.48 72.576 L1909.48 27.2059 L1916.97 27.2059 L1916.97 34.2544 Q1919.65 30.163 1923.25 28.1376 Q1926.9 26.1121 1931.64 26.1121 Q1939.46 26.1121 1943.47 30.9732 Q1947.48 35.7938 1947.48 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1992.2 34.0924 L1992.2 9.54393 L1999.65 9.54393 L1999.65 72.576 L1992.2 72.576 L1992.2 65.7705 Q1989.85 69.8214 1986.24 71.8063 Q1982.68 73.7508 1977.66 73.7508 Q1969.43 73.7508 1964.25 67.1883 Q1959.1 60.6258 1959.1 49.9314 Q1959.1 39.2371 1964.25 32.6746 Q1969.43 26.1121 1977.66 26.1121 Q1982.68 26.1121 1986.24 28.0971 Q1989.85 30.0415 1992.2 34.0924 M1966.8 49.9314 Q1966.8 58.1548 1970.16 62.8538 Q1973.56 67.5124 1979.48 67.5124 Q1985.39 67.5124 1988.8 62.8538 Q1992.2 58.1548 1992.2 49.9314 Q1992.2 41.7081 1988.8 37.0496 Q1985.39 32.3505 1979.48 32.3505 Q1973.56 32.3505 1970.16 37.0496 Q1966.8 41.7081 1966.8 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip653)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1473.94,1386.4 1545.04,1386.4 1616.14,1386.4 1687.24,1386.4 1758.34,1386.4 1829.45,1386.4 1900.55,1386.4 1971.65,160.256 2042.75,160.256 2113.85,160.256 2184.96,160.256 2256.06,160.256 2327.16,160.256 "/>
-<polyline clip-path="url(#clip653)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1473.94,160.256 1545.04,160.256 1616.14,160.256 1687.24,160.256 1758.34,160.256 1829.45,160.256 1900.55,160.256 1971.65,1386.4 2042.75,1386.4 2113.85,1386.4 2184.96,1386.4 2256.06,1386.4 2327.16,1386.4 "/>
-<path clip-path="url(#clip650)" d="M2032.63 1379.86 L2322.61 1379.86 L2322.61 1224.34 L2032.63 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2032.63,1379.86 2322.61,1379.86 2322.61,1224.34 2032.63,1224.34 2032.63,1379.86 "/>
-<polyline clip-path="url(#clip650)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2042.68,1276.18 2102.97,1276.18 "/>
-<path clip-path="url(#clip650)" d="M2117.7 1262.74 L2117.7 1289.61 L2123.35 1289.61 Q2130.5 1289.61 2133.81 1286.37 Q2137.14 1283.13 2137.14 1276.14 Q2137.14 1269.2 2133.81 1265.98 Q2130.5 1262.74 2123.35 1262.74 L2117.7 1262.74 M2113.02 1258.9 L2122.63 1258.9 Q2132.68 1258.9 2137.38 1263.09 Q2142.07 1267.25 2142.07 1276.14 Q2142.07 1285.08 2137.35 1289.27 Q2132.63 1293.46 2122.63 1293.46 L2113.02 1293.46 L2113.02 1258.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2149.34 1267.53 L2153.6 1267.53 L2153.6 1293.46 L2149.34 1293.46 L2149.34 1267.53 M2149.34 1257.44 L2153.6 1257.44 L2153.6 1262.83 L2149.34 1262.83 L2149.34 1257.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2179.04 1268.29 L2179.04 1272.32 Q2177.24 1271.4 2175.29 1270.93 Q2173.35 1270.47 2171.26 1270.47 Q2168.09 1270.47 2166.5 1271.44 Q2164.92 1272.41 2164.92 1274.36 Q2164.92 1275.84 2166.06 1276.7 Q2167.19 1277.53 2170.62 1278.29 L2172.07 1278.62 Q2176.61 1279.59 2178.51 1281.37 Q2180.43 1283.13 2180.43 1286.3 Q2180.43 1289.91 2177.56 1292.02 Q2174.71 1294.13 2169.71 1294.13 Q2167.63 1294.13 2165.36 1293.71 Q2163.12 1293.32 2160.62 1292.51 L2160.62 1288.11 Q2162.98 1289.34 2165.27 1289.96 Q2167.56 1290.56 2169.81 1290.56 Q2172.81 1290.56 2174.44 1289.54 Q2176.06 1288.5 2176.06 1286.63 Q2176.06 1284.89 2174.88 1283.97 Q2173.72 1283.04 2169.76 1282.18 L2168.28 1281.84 Q2164.32 1281 2162.56 1279.29 Q2160.8 1277.55 2160.8 1274.54 Q2160.8 1270.89 2163.39 1268.9 Q2165.99 1266.91 2170.75 1266.91 Q2173.12 1266.91 2175.2 1267.25 Q2177.28 1267.6 2179.04 1268.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2191.33 1289.57 L2191.33 1303.32 L2187.05 1303.32 L2187.05 1267.53 L2191.33 1267.53 L2191.33 1271.47 Q2192.68 1269.15 2194.71 1268.04 Q2196.77 1266.91 2199.62 1266.91 Q2204.34 1266.91 2207.28 1270.66 Q2210.25 1274.41 2210.25 1280.52 Q2210.25 1286.63 2207.28 1290.38 Q2204.34 1294.13 2199.62 1294.13 Q2196.77 1294.13 2194.71 1293.02 Q2192.68 1291.88 2191.33 1289.57 M2205.82 1280.52 Q2205.82 1275.82 2203.88 1273.16 Q2201.96 1270.47 2198.58 1270.47 Q2195.2 1270.47 2193.25 1273.16 Q2191.33 1275.82 2191.33 1280.52 Q2191.33 1285.22 2193.25 1287.9 Q2195.2 1290.56 2198.58 1290.56 Q2201.96 1290.56 2203.88 1287.9 Q2205.82 1285.22 2205.82 1280.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2229.09 1280.42 Q2223.93 1280.42 2221.94 1281.6 Q2219.94 1282.79 2219.94 1285.63 Q2219.94 1287.9 2221.43 1289.24 Q2222.93 1290.56 2225.5 1290.56 Q2229.04 1290.56 2231.17 1288.06 Q2233.32 1285.54 2233.32 1281.37 L2233.32 1280.42 L2229.09 1280.42 M2237.58 1278.66 L2237.58 1293.46 L2233.32 1293.46 L2233.32 1289.52 Q2231.87 1291.88 2229.69 1293.02 Q2227.51 1294.13 2224.37 1294.13 Q2220.38 1294.13 2218.02 1291.91 Q2215.69 1289.66 2215.69 1285.91 Q2215.69 1281.54 2218.6 1279.31 Q2221.54 1277.09 2227.35 1277.09 L2233.32 1277.09 L2233.32 1276.67 Q2233.32 1273.73 2231.38 1272.14 Q2229.46 1270.52 2225.96 1270.52 Q2223.74 1270.52 2221.63 1271.05 Q2219.53 1271.58 2217.58 1272.65 L2217.58 1268.71 Q2219.92 1267.81 2222.12 1267.37 Q2224.32 1266.91 2226.4 1266.91 Q2232.03 1266.91 2234.81 1269.82 Q2237.58 1272.74 2237.58 1278.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2250.57 1260.17 L2250.57 1267.53 L2259.34 1267.53 L2259.34 1270.84 L2250.57 1270.84 L2250.57 1284.91 Q2250.57 1288.09 2251.43 1288.99 Q2252.31 1289.89 2254.97 1289.89 L2259.34 1289.89 L2259.34 1293.46 L2254.97 1293.46 Q2250.04 1293.46 2248.16 1291.63 Q2246.29 1289.78 2246.29 1284.91 L2246.29 1270.84 L2243.16 1270.84 L2243.16 1267.53 L2246.29 1267.53 L2246.29 1260.17 L2250.57 1260.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2283.6 1268.53 L2283.6 1272.51 Q2281.8 1271.51 2279.97 1271.03 Q2278.16 1270.52 2276.31 1270.52 Q2272.17 1270.52 2269.87 1273.16 Q2267.58 1275.77 2267.58 1280.52 Q2267.58 1285.26 2269.87 1287.9 Q2272.17 1290.52 2276.31 1290.52 Q2278.16 1290.52 2279.97 1290.03 Q2281.8 1289.52 2283.6 1288.53 L2283.6 1292.46 Q2281.82 1293.29 2279.9 1293.71 Q2278 1294.13 2275.85 1294.13 Q2269.99 1294.13 2266.54 1290.45 Q2263.09 1286.77 2263.09 1280.52 Q2263.09 1274.17 2266.56 1270.54 Q2270.06 1266.91 2276.12 1266.91 Q2278.09 1266.91 2279.97 1267.32 Q2281.84 1267.72 2283.6 1268.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2312.56 1277.81 L2312.56 1293.46 L2308.3 1293.46 L2308.3 1277.95 Q2308.3 1274.27 2306.87 1272.44 Q2305.43 1270.61 2302.56 1270.61 Q2299.11 1270.61 2297.12 1272.81 Q2295.13 1275.01 2295.13 1278.8 L2295.13 1293.46 L2290.85 1293.46 L2290.85 1257.44 L2295.13 1257.44 L2295.13 1271.56 Q2296.66 1269.22 2298.72 1268.06 Q2300.8 1266.91 2303.51 1266.91 Q2307.98 1266.91 2310.27 1269.68 Q2312.56 1272.44 2312.56 1277.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip650)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2042.68,1328.02 2102.97,1328.02 "/>
-<path clip-path="url(#clip650)" d="M2135.27 1311.87 L2135.27 1316.43 Q2132.61 1315.16 2130.25 1314.53 Q2127.88 1313.91 2125.69 1313.91 Q2121.87 1313.91 2119.78 1315.39 Q2117.72 1316.87 2117.72 1319.6 Q2117.72 1321.89 2119.09 1323.07 Q2120.48 1324.23 2124.32 1324.95 L2127.14 1325.53 Q2132.38 1326.52 2134.85 1329.05 Q2137.35 1331.55 2137.35 1335.76 Q2137.35 1340.78 2133.97 1343.37 Q2130.62 1345.97 2124.11 1345.97 Q2121.66 1345.97 2118.88 1345.41 Q2116.13 1344.86 2113.16 1343.77 L2113.16 1338.95 Q2116.01 1340.55 2118.74 1341.36 Q2121.47 1342.17 2124.11 1342.17 Q2128.12 1342.17 2130.29 1340.6 Q2132.47 1339.02 2132.47 1336.11 Q2132.47 1333.56 2130.89 1332.13 Q2129.34 1330.69 2125.78 1329.97 L2122.93 1329.42 Q2117.7 1328.38 2115.36 1326.15 Q2113.02 1323.93 2113.02 1319.97 Q2113.02 1315.39 2116.24 1312.75 Q2119.48 1310.11 2125.15 1310.11 Q2127.58 1310.11 2130.11 1310.55 Q2132.63 1310.99 2135.27 1311.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2148.58 1341.41 L2148.58 1355.16 L2144.3 1355.16 L2144.3 1319.37 L2148.58 1319.37 L2148.58 1323.31 Q2149.92 1320.99 2151.96 1319.88 Q2154.02 1318.75 2156.87 1318.75 Q2161.59 1318.75 2164.53 1322.5 Q2167.49 1326.25 2167.49 1332.36 Q2167.49 1338.47 2164.53 1342.22 Q2161.59 1345.97 2156.87 1345.97 Q2154.02 1345.97 2151.96 1344.86 Q2149.92 1343.72 2148.58 1341.41 M2163.07 1332.36 Q2163.07 1327.66 2161.13 1325 Q2159.2 1322.31 2155.82 1322.31 Q2152.44 1322.31 2150.5 1325 Q2148.58 1327.66 2148.58 1332.36 Q2148.58 1337.06 2150.5 1339.74 Q2152.44 1342.4 2155.82 1342.4 Q2159.2 1342.4 2161.13 1339.74 Q2163.07 1337.06 2163.07 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2174.55 1319.37 L2178.81 1319.37 L2178.81 1345.3 L2174.55 1345.3 L2174.55 1319.37 M2174.55 1309.28 L2178.81 1309.28 L2178.81 1314.67 L2174.55 1314.67 L2174.55 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2187.72 1309.28 L2191.98 1309.28 L2191.98 1345.3 L2187.72 1345.3 L2187.72 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2200.89 1309.28 L2205.15 1309.28 L2205.15 1345.3 L2200.89 1345.3 L2200.89 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2225.85 1332.26 Q2220.69 1332.26 2218.69 1333.44 Q2216.7 1334.63 2216.7 1337.47 Q2216.7 1339.74 2218.19 1341.08 Q2219.69 1342.4 2222.26 1342.4 Q2225.8 1342.4 2227.93 1339.9 Q2230.08 1337.38 2230.08 1333.21 L2230.08 1332.26 L2225.85 1332.26 M2234.34 1330.5 L2234.34 1345.3 L2230.08 1345.3 L2230.08 1341.36 Q2228.62 1343.72 2226.45 1344.86 Q2224.27 1345.97 2221.12 1345.97 Q2217.14 1345.97 2214.78 1343.75 Q2212.44 1341.5 2212.44 1337.75 Q2212.44 1333.38 2215.36 1331.15 Q2218.3 1328.93 2224.11 1328.93 L2230.08 1328.93 L2230.08 1328.51 Q2230.08 1325.57 2228.14 1323.98 Q2226.22 1322.36 2222.72 1322.36 Q2220.5 1322.36 2218.39 1322.89 Q2216.29 1323.42 2214.34 1324.49 L2214.34 1320.55 Q2216.68 1319.65 2218.88 1319.21 Q2221.08 1318.75 2223.16 1318.75 Q2228.79 1318.75 2231.56 1321.66 Q2234.34 1324.58 2234.34 1330.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2260.18 1332.03 Q2260.18 1327.4 2258.25 1324.86 Q2256.36 1322.31 2252.91 1322.31 Q2249.48 1322.31 2247.56 1324.86 Q2245.66 1327.4 2245.66 1332.03 Q2245.66 1336.64 2247.56 1339.19 Q2249.48 1341.73 2252.91 1341.73 Q2256.36 1341.73 2258.25 1339.19 Q2260.18 1336.64 2260.18 1332.03 M2264.43 1342.08 Q2264.43 1348.7 2261.5 1351.92 Q2258.56 1355.16 2252.49 1355.16 Q2250.25 1355.16 2248.25 1354.81 Q2246.26 1354.49 2244.39 1353.79 L2244.39 1349.65 Q2246.26 1350.67 2248.09 1351.15 Q2249.92 1351.64 2251.82 1351.64 Q2256.01 1351.64 2258.09 1349.44 Q2260.18 1347.26 2260.18 1342.84 L2260.18 1340.74 Q2258.86 1343.03 2256.8 1344.16 Q2254.74 1345.3 2251.87 1345.3 Q2247.1 1345.3 2244.18 1341.66 Q2241.26 1338.03 2241.26 1332.03 Q2241.26 1326.01 2244.18 1322.38 Q2247.1 1318.75 2251.87 1318.75 Q2254.74 1318.75 2256.8 1319.88 Q2258.86 1321.01 2260.18 1323.31 L2260.18 1319.37 L2264.43 1319.37 L2264.43 1342.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2295.38 1331.27 L2295.38 1333.35 L2275.8 1333.35 Q2276.08 1337.75 2278.44 1340.06 Q2280.82 1342.36 2285.06 1342.36 Q2287.51 1342.36 2289.81 1341.75 Q2292.12 1341.15 2294.39 1339.95 L2294.39 1343.98 Q2292.1 1344.95 2289.69 1345.46 Q2287.28 1345.97 2284.81 1345.97 Q2278.6 1345.97 2274.97 1342.36 Q2271.36 1338.75 2271.36 1332.59 Q2271.36 1326.22 2274.78 1322.5 Q2278.23 1318.75 2284.06 1318.75 Q2289.3 1318.75 2292.33 1322.13 Q2295.38 1325.48 2295.38 1331.27 M2291.12 1330.02 Q2291.08 1326.52 2289.16 1324.44 Q2287.26 1322.36 2284.11 1322.36 Q2280.55 1322.36 2278.39 1324.37 Q2276.26 1326.38 2275.94 1330.04 L2291.12 1330.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1568.74,1423.18 1568.74,123.472 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1805.74,1423.18 1805.74,123.472 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2042.75,1423.18 2042.75,123.472 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2279.76,1423.18 2279.76,123.472 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1568.74,1423.18 1568.74,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1805.74,1423.18 1805.74,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2042.75,1423.18 2042.75,1404.28 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2279.76,1423.18 2279.76,1404.28 "/>
+<path clip-path="url(#clip470)" d="M1528.36 1451.02 L1546.71 1451.02 L1546.71 1454.96 L1532.64 1454.96 L1532.64 1463.43 Q1533.66 1463.08 1534.68 1462.92 Q1535.69 1462.73 1536.71 1462.73 Q1542.5 1462.73 1545.88 1465.9 Q1549.26 1469.08 1549.26 1474.49 Q1549.26 1480.07 1545.79 1483.17 Q1542.31 1486.25 1535.99 1486.25 Q1533.82 1486.25 1531.55 1485.88 Q1529.31 1485.51 1526.9 1484.77 L1526.9 1480.07 Q1528.98 1481.2 1531.2 1481.76 Q1533.43 1482.32 1535.9 1482.32 Q1539.91 1482.32 1542.24 1480.21 Q1544.58 1478.1 1544.58 1474.49 Q1544.58 1470.88 1542.24 1468.77 Q1539.91 1466.67 1535.9 1466.67 Q1534.03 1466.67 1532.15 1467.08 Q1530.3 1467.5 1528.36 1468.38 L1528.36 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1568.47 1454.1 Q1564.86 1454.1 1563.03 1457.66 Q1561.23 1461.2 1561.23 1468.33 Q1561.23 1475.44 1563.03 1479.01 Q1564.86 1482.55 1568.47 1482.55 Q1572.11 1482.55 1573.91 1479.01 Q1575.74 1475.44 1575.74 1468.33 Q1575.74 1461.2 1573.91 1457.66 Q1572.11 1454.1 1568.47 1454.1 M1568.47 1450.39 Q1574.28 1450.39 1577.34 1455 Q1580.42 1459.58 1580.42 1468.33 Q1580.42 1477.06 1577.34 1481.67 Q1574.28 1486.25 1568.47 1486.25 Q1562.66 1486.25 1559.58 1481.67 Q1556.53 1477.06 1556.53 1468.33 Q1556.53 1459.58 1559.58 1455 Q1562.66 1450.39 1568.47 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1598.63 1454.1 Q1595.02 1454.1 1593.19 1457.66 Q1591.39 1461.2 1591.39 1468.33 Q1591.39 1475.44 1593.19 1479.01 Q1595.02 1482.55 1598.63 1482.55 Q1602.27 1482.55 1604.07 1479.01 Q1605.9 1475.44 1605.9 1468.33 Q1605.9 1461.2 1604.07 1457.66 Q1602.27 1454.1 1598.63 1454.1 M1598.63 1450.39 Q1604.44 1450.39 1607.5 1455 Q1610.58 1459.58 1610.58 1468.33 Q1610.58 1477.06 1607.5 1481.67 Q1604.44 1486.25 1598.63 1486.25 Q1592.82 1486.25 1589.74 1481.67 Q1586.69 1477.06 1586.69 1468.33 Q1586.69 1459.58 1589.74 1455 Q1592.82 1450.39 1598.63 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1750.27 1481.64 L1757.91 1481.64 L1757.91 1455.28 L1749.6 1456.95 L1749.6 1452.69 L1757.86 1451.02 L1762.54 1451.02 L1762.54 1481.64 L1770.18 1481.64 L1770.18 1485.58 L1750.27 1485.58 L1750.27 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1789.62 1454.1 Q1786.01 1454.1 1784.18 1457.66 Q1782.38 1461.2 1782.38 1468.33 Q1782.38 1475.44 1784.18 1479.01 Q1786.01 1482.55 1789.62 1482.55 Q1793.26 1482.55 1795.06 1479.01 Q1796.89 1475.44 1796.89 1468.33 Q1796.89 1461.2 1795.06 1457.66 Q1793.26 1454.1 1789.62 1454.1 M1789.62 1450.39 Q1795.43 1450.39 1798.49 1455 Q1801.57 1459.58 1801.57 1468.33 Q1801.57 1477.06 1798.49 1481.67 Q1795.43 1486.25 1789.62 1486.25 Q1783.81 1486.25 1780.73 1481.67 Q1777.68 1477.06 1777.68 1468.33 Q1777.68 1459.58 1780.73 1455 Q1783.81 1450.39 1789.62 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1819.78 1454.1 Q1816.17 1454.1 1814.34 1457.66 Q1812.54 1461.2 1812.54 1468.33 Q1812.54 1475.44 1814.34 1479.01 Q1816.17 1482.55 1819.78 1482.55 Q1823.42 1482.55 1825.22 1479.01 Q1827.05 1475.44 1827.05 1468.33 Q1827.05 1461.2 1825.22 1457.66 Q1823.42 1454.1 1819.78 1454.1 M1819.78 1450.39 Q1825.59 1450.39 1828.65 1455 Q1831.73 1459.58 1831.73 1468.33 Q1831.73 1477.06 1828.65 1481.67 Q1825.59 1486.25 1819.78 1486.25 Q1813.97 1486.25 1810.89 1481.67 Q1807.84 1477.06 1807.84 1468.33 Q1807.84 1459.58 1810.89 1455 Q1813.97 1450.39 1819.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1849.95 1454.1 Q1846.33 1454.1 1844.51 1457.66 Q1842.7 1461.2 1842.7 1468.33 Q1842.7 1475.44 1844.51 1479.01 Q1846.33 1482.55 1849.95 1482.55 Q1853.58 1482.55 1855.39 1479.01 Q1857.21 1475.44 1857.21 1468.33 Q1857.21 1461.2 1855.39 1457.66 Q1853.58 1454.1 1849.95 1454.1 M1849.95 1450.39 Q1855.76 1450.39 1858.81 1455 Q1861.89 1459.58 1861.89 1468.33 Q1861.89 1477.06 1858.81 1481.67 Q1855.76 1486.25 1849.95 1486.25 Q1844.14 1486.25 1841.06 1481.67 Q1838 1477.06 1838 1468.33 Q1838 1459.58 1841.06 1455 Q1844.14 1450.39 1849.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1987.28 1481.64 L1994.92 1481.64 L1994.92 1455.28 L1986.61 1456.95 L1986.61 1452.69 L1994.87 1451.02 L1999.55 1451.02 L1999.55 1481.64 L2007.18 1481.64 L2007.18 1485.58 L1987.28 1485.58 L1987.28 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2016.67 1451.02 L2035.03 1451.02 L2035.03 1454.96 L2020.96 1454.96 L2020.96 1463.43 Q2021.98 1463.08 2022.99 1462.92 Q2024.01 1462.73 2025.03 1462.73 Q2030.82 1462.73 2034.2 1465.9 Q2037.58 1469.08 2037.58 1474.49 Q2037.58 1480.07 2034.11 1483.17 Q2030.63 1486.25 2024.31 1486.25 Q2022.14 1486.25 2019.87 1485.88 Q2017.62 1485.51 2015.22 1484.77 L2015.22 1480.07 Q2017.3 1481.2 2019.52 1481.76 Q2021.74 1482.32 2024.22 1482.32 Q2028.23 1482.32 2030.56 1480.21 Q2032.9 1478.1 2032.9 1474.49 Q2032.9 1470.88 2030.56 1468.77 Q2028.23 1466.67 2024.22 1466.67 Q2022.35 1466.67 2020.47 1467.08 Q2018.62 1467.5 2016.67 1468.38 L2016.67 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2056.79 1454.1 Q2053.18 1454.1 2051.35 1457.66 Q2049.55 1461.2 2049.55 1468.33 Q2049.55 1475.44 2051.35 1479.01 Q2053.18 1482.55 2056.79 1482.55 Q2060.42 1482.55 2062.23 1479.01 Q2064.06 1475.44 2064.06 1468.33 Q2064.06 1461.2 2062.23 1457.66 Q2060.42 1454.1 2056.79 1454.1 M2056.79 1450.39 Q2062.6 1450.39 2065.66 1455 Q2068.73 1459.58 2068.73 1468.33 Q2068.73 1477.06 2065.66 1481.67 Q2062.6 1486.25 2056.79 1486.25 Q2050.98 1486.25 2047.9 1481.67 Q2044.85 1477.06 2044.85 1468.33 Q2044.85 1459.58 2047.9 1455 Q2050.98 1450.39 2056.79 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2086.95 1454.1 Q2083.34 1454.1 2081.51 1457.66 Q2079.71 1461.2 2079.71 1468.33 Q2079.71 1475.44 2081.51 1479.01 Q2083.34 1482.55 2086.95 1482.55 Q2090.59 1482.55 2092.39 1479.01 Q2094.22 1475.44 2094.22 1468.33 Q2094.22 1461.2 2092.39 1457.66 Q2090.59 1454.1 2086.95 1454.1 M2086.95 1450.39 Q2092.76 1450.39 2095.82 1455 Q2098.9 1459.58 2098.9 1468.33 Q2098.9 1477.06 2095.82 1481.67 Q2092.76 1486.25 2086.95 1486.25 Q2081.14 1486.25 2078.06 1481.67 Q2075.01 1477.06 2075.01 1468.33 Q2075.01 1459.58 2078.06 1455 Q2081.14 1450.39 2086.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2228.37 1481.64 L2244.69 1481.64 L2244.69 1485.58 L2222.74 1485.58 L2222.74 1481.64 Q2225.41 1478.89 2229.99 1474.26 Q2234.6 1469.61 2235.78 1468.27 Q2238.02 1465.74 2238.9 1464.01 Q2239.8 1462.25 2239.8 1460.56 Q2239.8 1457.8 2237.86 1456.07 Q2235.94 1454.33 2232.84 1454.33 Q2230.64 1454.33 2228.18 1455.09 Q2225.75 1455.86 2222.98 1457.41 L2222.98 1452.69 Q2225.8 1451.55 2228.25 1450.97 Q2230.71 1450.39 2232.74 1450.39 Q2238.11 1450.39 2241.31 1453.08 Q2244.5 1455.77 2244.5 1460.26 Q2244.5 1462.39 2243.69 1464.31 Q2242.91 1466.2 2240.8 1468.8 Q2240.22 1469.47 2237.12 1472.69 Q2234.02 1475.88 2228.37 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2264.5 1454.1 Q2260.89 1454.1 2259.06 1457.66 Q2257.26 1461.2 2257.26 1468.33 Q2257.26 1475.44 2259.06 1479.01 Q2260.89 1482.55 2264.5 1482.55 Q2268.14 1482.55 2269.94 1479.01 Q2271.77 1475.44 2271.77 1468.33 Q2271.77 1461.2 2269.94 1457.66 Q2268.14 1454.1 2264.5 1454.1 M2264.5 1450.39 Q2270.31 1450.39 2273.37 1455 Q2276.45 1459.58 2276.45 1468.33 Q2276.45 1477.06 2273.37 1481.67 Q2270.31 1486.25 2264.5 1486.25 Q2258.69 1486.25 2255.61 1481.67 Q2252.56 1477.06 2252.56 1468.33 Q2252.56 1459.58 2255.61 1455 Q2258.69 1450.39 2264.5 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2294.67 1454.1 Q2291.05 1454.1 2289.23 1457.66 Q2287.42 1461.2 2287.42 1468.33 Q2287.42 1475.44 2289.23 1479.01 Q2291.05 1482.55 2294.67 1482.55 Q2298.3 1482.55 2300.1 1479.01 Q2301.93 1475.44 2301.93 1468.33 Q2301.93 1461.2 2300.1 1457.66 Q2298.3 1454.1 2294.67 1454.1 M2294.67 1450.39 Q2300.48 1450.39 2303.53 1455 Q2306.61 1459.58 2306.61 1468.33 Q2306.61 1477.06 2303.53 1481.67 Q2300.48 1486.25 2294.67 1486.25 Q2288.86 1486.25 2285.78 1481.67 Q2282.72 1477.06 2282.72 1468.33 Q2282.72 1459.58 2285.78 1455 Q2288.86 1450.39 2294.67 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2324.83 1454.1 Q2321.22 1454.1 2319.39 1457.66 Q2317.58 1461.2 2317.58 1468.33 Q2317.58 1475.44 2319.39 1479.01 Q2321.22 1482.55 2324.83 1482.55 Q2328.46 1482.55 2330.27 1479.01 Q2332.1 1475.44 2332.1 1468.33 Q2332.1 1461.2 2330.27 1457.66 Q2328.46 1454.1 2324.83 1454.1 M2324.83 1450.39 Q2330.64 1450.39 2333.69 1455 Q2336.77 1459.58 2336.77 1468.33 Q2336.77 1477.06 2333.69 1481.67 Q2330.64 1486.25 2324.83 1486.25 Q2319.02 1486.25 2315.94 1481.67 Q2312.88 1477.06 2312.88 1468.33 Q2312.88 1459.58 2315.94 1455 Q2319.02 1450.39 2324.83 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1678.62 1525.81 L1678.62 1562.76 L1686.39 1562.76 Q1696.22 1562.76 1700.78 1558.3 Q1705.36 1553.85 1705.36 1544.24 Q1705.36 1534.69 1700.78 1530.26 Q1696.22 1525.81 1686.39 1525.81 L1678.62 1525.81 M1672.19 1520.52 L1685.4 1520.52 Q1699.22 1520.52 1705.68 1526.28 Q1712.14 1532.01 1712.14 1544.24 Q1712.14 1556.52 1705.65 1562.28 Q1699.15 1568.04 1685.4 1568.04 L1672.19 1568.04 L1672.19 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1752.62 1548.76 L1752.62 1551.62 L1725.7 1551.62 Q1726.08 1557.67 1729.33 1560.85 Q1732.6 1564 1738.43 1564 Q1741.8 1564 1744.95 1563.17 Q1748.14 1562.35 1751.26 1560.69 L1751.26 1566.23 Q1748.1 1567.57 1744.79 1568.27 Q1741.48 1568.97 1738.08 1568.97 Q1729.55 1568.97 1724.55 1564 Q1719.59 1559.04 1719.59 1550.57 Q1719.59 1541.82 1724.3 1536.69 Q1729.04 1531.54 1737.06 1531.54 Q1744.25 1531.54 1748.42 1536.18 Q1752.62 1540.8 1752.62 1548.76 M1746.77 1547.04 Q1746.7 1542.23 1744.06 1539.37 Q1741.45 1536.5 1737.12 1536.5 Q1732.22 1536.5 1729.26 1539.27 Q1726.33 1542.04 1725.89 1547.07 L1746.77 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1789.99 1539.24 Q1792.19 1535.29 1795.24 1533.41 Q1798.3 1531.54 1802.44 1531.54 Q1808.01 1531.54 1811.03 1535.45 Q1814.05 1539.33 1814.05 1546.53 L1814.05 1568.04 L1808.16 1568.04 L1808.16 1546.72 Q1808.16 1541.59 1806.35 1539.11 Q1804.54 1536.63 1800.81 1536.63 Q1796.26 1536.63 1793.62 1539.65 Q1790.98 1542.68 1790.98 1547.9 L1790.98 1568.04 L1785.09 1568.04 L1785.09 1546.72 Q1785.09 1541.56 1783.27 1539.11 Q1781.46 1536.63 1777.67 1536.63 Q1773.19 1536.63 1770.54 1539.68 Q1767.9 1542.71 1767.9 1547.9 L1767.9 1568.04 L1762.01 1568.04 L1762.01 1532.4 L1767.9 1532.4 L1767.9 1537.93 Q1769.91 1534.66 1772.71 1533.1 Q1775.51 1531.54 1779.36 1531.54 Q1783.24 1531.54 1785.95 1533.51 Q1788.69 1535.48 1789.99 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1841.93 1550.12 Q1834.84 1550.12 1832.1 1551.75 Q1829.36 1553.37 1829.36 1557.29 Q1829.36 1560.4 1831.4 1562.25 Q1833.47 1564.07 1837 1564.07 Q1841.87 1564.07 1844.8 1560.63 Q1847.76 1557.16 1847.76 1551.43 L1847.76 1550.12 L1841.93 1550.12 M1853.62 1547.71 L1853.62 1568.04 L1847.76 1568.04 L1847.76 1562.63 Q1845.75 1565.88 1842.76 1567.44 Q1839.77 1568.97 1835.44 1568.97 Q1829.97 1568.97 1826.72 1565.91 Q1823.51 1562.82 1823.51 1557.67 Q1823.51 1551.65 1827.52 1548.6 Q1831.56 1545.54 1839.55 1545.54 L1847.76 1545.54 L1847.76 1544.97 Q1847.76 1540.93 1845.09 1538.73 Q1842.44 1536.5 1837.64 1536.5 Q1834.58 1536.5 1831.69 1537.23 Q1828.79 1537.97 1826.12 1539.43 L1826.12 1534.02 Q1829.33 1532.78 1832.35 1532.17 Q1835.38 1531.54 1838.24 1531.54 Q1845.98 1531.54 1849.8 1535.55 Q1853.62 1539.56 1853.62 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1895.31 1546.53 L1895.31 1568.04 L1889.45 1568.04 L1889.45 1546.72 Q1889.45 1541.66 1887.48 1539.14 Q1885.51 1536.63 1881.56 1536.63 Q1876.82 1536.63 1874.08 1539.65 Q1871.34 1542.68 1871.34 1547.9 L1871.34 1568.04 L1865.46 1568.04 L1865.46 1532.4 L1871.34 1532.4 L1871.34 1537.93 Q1873.45 1534.72 1876.28 1533.13 Q1879.14 1531.54 1882.87 1531.54 Q1889.01 1531.54 1892.16 1535.36 Q1895.31 1539.14 1895.31 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1930.45 1537.81 L1930.45 1518.52 L1936.31 1518.52 L1936.31 1568.04 L1930.45 1568.04 L1930.45 1562.7 Q1928.6 1565.88 1925.77 1567.44 Q1922.97 1568.97 1919.02 1568.97 Q1912.56 1568.97 1908.49 1563.81 Q1904.45 1558.65 1904.45 1550.25 Q1904.45 1541.85 1908.49 1536.69 Q1912.56 1531.54 1919.02 1531.54 Q1922.97 1531.54 1925.77 1533.1 Q1928.6 1534.62 1930.45 1537.81 M1910.49 1550.25 Q1910.49 1556.71 1913.14 1560.4 Q1915.81 1564.07 1920.46 1564.07 Q1925.1 1564.07 1927.78 1560.4 Q1930.45 1556.71 1930.45 1550.25 Q1930.45 1543.79 1927.78 1540.13 Q1925.1 1536.44 1920.46 1536.44 Q1915.81 1536.44 1913.14 1540.13 Q1910.49 1543.79 1910.49 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1968.55 1518.52 L1982.04 1518.52 L1982.04 1523.07 L1974.41 1523.07 L1974.41 1572.09 L1982.04 1572.09 L1982.04 1576.64 L1968.55 1576.64 L1968.55 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1994.78 1520.52 L2004.36 1520.52 L2016.48 1552.86 L2028.67 1520.52 L2038.25 1520.52 L2038.25 1568.04 L2031.98 1568.04 L2031.98 1526.32 L2019.73 1558.91 L2013.27 1558.91 L2001.01 1526.32 L2001.01 1568.04 L1994.78 1568.04 L1994.78 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2046.78 1520.52 L2053.28 1520.52 L2063.27 1560.69 L2073.23 1520.52 L2080.46 1520.52 L2090.45 1560.69 L2100.41 1520.52 L2106.94 1520.52 L2095 1568.04 L2086.92 1568.04 L2076.89 1526.79 L2066.77 1568.04 L2058.69 1568.04 L2046.78 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2128.9 1518.52 L2128.9 1576.64 L2115.41 1576.64 L2115.41 1572.09 L2123.01 1572.09 L2123.01 1523.07 L2115.41 1523.07 L2115.41 1518.52 L2128.9 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,1386.4 2352.76,1386.4 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,1079.86 2352.76,1079.86 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,773.326 2352.76,773.326 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,466.791 2352.76,466.791 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.34,160.256 2352.76,160.256 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1423.18 1448.34,123.472 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1386.4 1467.24,1386.4 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,1079.86 1467.24,1079.86 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,773.326 1467.24,773.326 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,466.791 1467.24,466.791 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.34,160.256 1467.24,160.256 "/>
+<path clip-path="url(#clip470)" d="M1400.39 1372.19 Q1396.78 1372.19 1394.95 1375.76 Q1393.15 1379.3 1393.15 1386.43 Q1393.15 1393.54 1394.95 1397.1 Q1396.78 1400.64 1400.39 1400.64 Q1404.03 1400.64 1405.83 1397.1 Q1407.66 1393.54 1407.66 1386.43 Q1407.66 1379.3 1405.83 1375.76 Q1404.03 1372.19 1400.39 1372.19 M1400.39 1368.49 Q1406.2 1368.49 1409.26 1373.1 Q1412.34 1377.68 1412.34 1386.43 Q1412.34 1395.16 1409.26 1399.76 Q1406.2 1404.35 1400.39 1404.35 Q1394.58 1404.35 1391.51 1399.76 Q1388.45 1395.16 1388.45 1386.43 Q1388.45 1377.68 1391.51 1373.1 Q1394.58 1368.49 1400.39 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1360.28 1062.58 L1378.63 1062.58 L1378.63 1066.52 L1364.56 1066.52 L1364.56 1074.99 Q1365.58 1074.64 1366.6 1074.48 Q1367.62 1074.29 1368.63 1074.29 Q1374.42 1074.29 1377.8 1077.47 Q1381.18 1080.64 1381.18 1086.05 Q1381.18 1091.63 1377.71 1094.73 Q1374.24 1097.81 1367.92 1097.81 Q1365.74 1097.81 1363.47 1097.44 Q1361.23 1097.07 1358.82 1096.33 L1358.82 1091.63 Q1360.9 1092.77 1363.13 1093.32 Q1365.35 1093.88 1367.82 1093.88 Q1371.83 1093.88 1374.17 1091.77 Q1376.51 1089.66 1376.51 1086.05 Q1376.51 1082.44 1374.17 1080.34 Q1371.83 1078.23 1367.82 1078.23 Q1365.95 1078.23 1364.07 1078.65 Q1362.22 1079.06 1360.28 1079.94 L1360.28 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1400.39 1065.66 Q1396.78 1065.66 1394.95 1069.22 Q1393.15 1072.77 1393.15 1079.9 Q1393.15 1087 1394.95 1090.57 Q1396.78 1094.11 1400.39 1094.11 Q1404.03 1094.11 1405.83 1090.57 Q1407.66 1087 1407.66 1079.9 Q1407.66 1072.77 1405.83 1069.22 Q1404.03 1065.66 1400.39 1065.66 M1400.39 1061.96 Q1406.2 1061.96 1409.26 1066.56 Q1412.34 1071.15 1412.34 1079.9 Q1412.34 1088.62 1409.26 1093.23 Q1406.2 1097.81 1400.39 1097.81 Q1394.58 1097.81 1391.51 1093.23 Q1388.45 1088.62 1388.45 1079.9 Q1388.45 1071.15 1391.51 1066.56 Q1394.58 1061.96 1400.39 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1330.88 786.671 L1338.52 786.671 L1338.52 760.305 L1330.21 761.972 L1330.21 757.713 L1338.47 756.046 L1343.15 756.046 L1343.15 786.671 L1350.79 786.671 L1350.79 790.606 L1330.88 790.606 L1330.88 786.671 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1370.23 759.125 Q1366.62 759.125 1364.79 762.689 Q1362.99 766.231 1362.99 773.361 Q1362.99 780.467 1364.79 784.032 Q1366.62 787.574 1370.23 787.574 Q1373.87 787.574 1375.67 784.032 Q1377.5 780.467 1377.5 773.361 Q1377.5 766.231 1375.67 762.689 Q1373.87 759.125 1370.23 759.125 M1370.23 755.421 Q1376.04 755.421 1379.1 760.027 Q1382.18 764.611 1382.18 773.361 Q1382.18 782.088 1379.1 786.694 Q1376.04 791.277 1370.23 791.277 Q1364.42 791.277 1361.34 786.694 Q1358.29 782.088 1358.29 773.361 Q1358.29 764.611 1361.34 760.027 Q1364.42 755.421 1370.23 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1400.39 759.125 Q1396.78 759.125 1394.95 762.689 Q1393.15 766.231 1393.15 773.361 Q1393.15 780.467 1394.95 784.032 Q1396.78 787.574 1400.39 787.574 Q1404.03 787.574 1405.83 784.032 Q1407.66 780.467 1407.66 773.361 Q1407.66 766.231 1405.83 762.689 Q1404.03 759.125 1400.39 759.125 M1400.39 755.421 Q1406.2 755.421 1409.26 760.027 Q1412.34 764.611 1412.34 773.361 Q1412.34 782.088 1409.26 786.694 Q1406.2 791.277 1400.39 791.277 Q1394.58 791.277 1391.51 786.694 Q1388.45 782.088 1388.45 773.361 Q1388.45 764.611 1391.51 760.027 Q1394.58 755.421 1400.39 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1330.88 480.136 L1338.52 480.136 L1338.52 453.77 L1330.21 455.437 L1330.21 451.178 L1338.47 449.511 L1343.15 449.511 L1343.15 480.136 L1350.79 480.136 L1350.79 484.071 L1330.88 484.071 L1330.88 480.136 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1360.28 449.511 L1378.63 449.511 L1378.63 453.446 L1364.56 453.446 L1364.56 461.918 Q1365.58 461.571 1366.6 461.409 Q1367.62 461.224 1368.63 461.224 Q1374.42 461.224 1377.8 464.395 Q1381.18 467.567 1381.18 472.983 Q1381.18 478.562 1377.71 481.664 Q1374.24 484.742 1367.92 484.742 Q1365.74 484.742 1363.47 484.372 Q1361.23 484.002 1358.82 483.261 L1358.82 478.562 Q1360.9 479.696 1363.13 480.252 Q1365.35 480.807 1367.82 480.807 Q1371.83 480.807 1374.17 478.701 Q1376.51 476.594 1376.51 472.983 Q1376.51 469.372 1374.17 467.266 Q1371.83 465.159 1367.82 465.159 Q1365.95 465.159 1364.07 465.576 Q1362.22 465.993 1360.28 466.872 L1360.28 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1400.39 452.59 Q1396.78 452.59 1394.95 456.155 Q1393.15 459.696 1393.15 466.826 Q1393.15 473.932 1394.95 477.497 Q1396.78 481.039 1400.39 481.039 Q1404.03 481.039 1405.83 477.497 Q1407.66 473.932 1407.66 466.826 Q1407.66 459.696 1405.83 456.155 Q1404.03 452.59 1400.39 452.59 M1400.39 448.886 Q1406.2 448.886 1409.26 453.493 Q1412.34 458.076 1412.34 466.826 Q1412.34 475.553 1409.26 480.159 Q1406.2 484.742 1400.39 484.742 Q1394.58 484.742 1391.51 480.159 Q1388.45 475.553 1388.45 466.826 Q1388.45 458.076 1391.51 453.493 Q1394.58 448.886 1400.39 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1334.1 173.601 L1350.42 173.601 L1350.42 177.536 L1328.47 177.536 L1328.47 173.601 Q1331.14 170.847 1335.72 166.217 Q1340.32 161.564 1341.51 160.222 Q1343.75 157.698 1344.63 155.962 Q1345.53 154.203 1345.53 152.513 Q1345.53 149.759 1343.59 148.023 Q1341.67 146.286 1338.57 146.286 Q1336.37 146.286 1333.91 147.05 Q1331.48 147.814 1328.7 149.365 L1328.7 144.643 Q1331.53 143.509 1333.98 142.93 Q1336.44 142.351 1338.47 142.351 Q1343.84 142.351 1347.04 145.036 Q1350.23 147.722 1350.23 152.212 Q1350.23 154.342 1349.42 156.263 Q1348.63 158.161 1346.53 160.754 Q1345.95 161.425 1342.85 164.643 Q1339.75 167.837 1334.1 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1370.23 146.055 Q1366.62 146.055 1364.79 149.62 Q1362.99 153.161 1362.99 160.291 Q1362.99 167.397 1364.79 170.962 Q1366.62 174.504 1370.23 174.504 Q1373.87 174.504 1375.67 170.962 Q1377.5 167.397 1377.5 160.291 Q1377.5 153.161 1375.67 149.62 Q1373.87 146.055 1370.23 146.055 M1370.23 142.351 Q1376.04 142.351 1379.1 146.958 Q1382.18 151.541 1382.18 160.291 Q1382.18 169.018 1379.1 173.624 Q1376.04 178.208 1370.23 178.208 Q1364.42 178.208 1361.34 173.624 Q1358.29 169.018 1358.29 160.291 Q1358.29 151.541 1361.34 146.958 Q1364.42 142.351 1370.23 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1400.39 146.055 Q1396.78 146.055 1394.95 149.62 Q1393.15 153.161 1393.15 160.291 Q1393.15 167.397 1394.95 170.962 Q1396.78 174.504 1400.39 174.504 Q1404.03 174.504 1405.83 170.962 Q1407.66 167.397 1407.66 160.291 Q1407.66 153.161 1405.83 149.62 Q1404.03 146.055 1400.39 146.055 M1400.39 142.351 Q1406.2 142.351 1409.26 146.958 Q1412.34 151.541 1412.34 160.291 Q1412.34 169.018 1409.26 173.624 Q1406.2 178.208 1400.39 178.208 Q1394.58 178.208 1391.51 173.624 Q1388.45 169.018 1388.45 160.291 Q1388.45 151.541 1391.51 146.958 Q1394.58 142.351 1400.39 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1230.7 977.57 L1230.7 947.524 L1236.11 947.524 L1236.11 971.14 L1250.18 971.14 L1250.18 948.51 L1255.59 948.51 L1255.59 971.14 L1272.81 971.14 L1272.81 946.951 L1278.22 946.951 L1278.22 977.57 L1230.7 977.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1256.7 907.006 L1278.22 907.006 L1278.22 912.862 L1256.89 912.862 Q1251.83 912.862 1249.32 914.836 Q1246.8 916.809 1246.8 920.756 Q1246.8 925.498 1249.83 928.235 Q1252.85 930.973 1258.07 930.973 L1278.22 930.973 L1278.22 936.861 L1242.57 936.861 L1242.57 930.973 L1248.11 930.973 Q1244.89 928.872 1243.3 926.039 Q1241.71 923.175 1241.71 919.451 Q1241.71 913.308 1245.53 910.157 Q1249.32 907.006 1256.7 907.006 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1258.93 864.833 L1261.79 864.833 L1261.79 891.76 Q1267.84 891.378 1271.02 888.132 Q1274.17 884.853 1274.17 879.029 Q1274.17 875.655 1273.35 872.504 Q1272.52 869.321 1270.86 866.202 L1276.4 866.202 Q1277.74 869.353 1278.44 872.663 Q1279.14 875.973 1279.14 879.379 Q1279.14 887.909 1274.17 892.906 Q1269.21 897.871 1260.74 897.871 Q1251.99 897.871 1246.87 893.16 Q1241.71 888.418 1241.71 880.397 Q1241.71 873.204 1246.36 869.034 Q1250.97 864.833 1258.93 864.833 M1257.21 870.689 Q1252.4 870.753 1249.54 873.395 Q1246.67 876.005 1246.67 880.334 Q1246.67 885.235 1249.44 888.195 Q1252.21 891.123 1257.24 891.569 L1257.21 870.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1248.04 834.564 Q1247.47 835.551 1247.22 836.728 Q1246.93 837.874 1246.93 839.275 Q1246.93 844.24 1250.18 846.914 Q1253.39 849.555 1259.44 849.555 L1278.22 849.555 L1278.22 855.444 L1242.57 855.444 L1242.57 849.555 L1248.11 849.555 Q1244.86 847.709 1243.3 844.749 Q1241.71 841.789 1241.71 837.556 Q1241.71 836.951 1241.81 836.219 Q1241.87 835.487 1242.03 834.596 L1248.04 834.564 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1259.98 806.109 Q1253.61 806.109 1250.11 808.751 Q1246.61 811.361 1246.61 816.104 Q1246.61 820.814 1250.11 823.456 Q1253.61 826.066 1259.98 826.066 Q1266.31 826.066 1269.81 823.456 Q1273.32 820.814 1273.32 816.104 Q1273.32 811.361 1269.81 808.751 Q1266.31 806.109 1259.98 806.109 M1273.79 800.253 Q1282.9 800.253 1287.32 804.295 Q1291.78 808.337 1291.78 816.676 Q1291.78 819.764 1291.3 822.501 Q1290.85 825.238 1289.9 827.816 L1284.2 827.816 Q1285.6 825.238 1286.27 822.724 Q1286.94 820.209 1286.94 817.599 Q1286.94 811.839 1283.91 808.974 Q1280.92 806.109 1274.84 806.109 L1271.95 806.109 Q1275.1 807.924 1276.66 810.756 Q1278.22 813.589 1278.22 817.536 Q1278.22 824.092 1273.22 828.103 Q1268.22 832.113 1259.98 832.113 Q1251.7 832.113 1246.71 828.103 Q1241.71 824.092 1241.71 817.536 Q1241.71 813.589 1243.27 810.756 Q1244.83 807.924 1247.98 806.109 L1242.57 806.109 L1242.57 800.253 L1273.79 800.253 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1281.53 773.358 Q1287.89 775.84 1289.83 778.196 Q1291.78 780.551 1291.78 784.498 L1291.78 789.177 L1286.87 789.177 L1286.87 785.739 Q1286.87 783.32 1285.73 781.983 Q1284.58 780.647 1280.32 779.023 L1277.64 777.973 L1242.57 792.391 L1242.57 786.185 L1270.45 775.045 L1242.57 763.905 L1242.57 757.698 L1281.53 773.358 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1228.69 729.434 L1228.69 715.939 L1233.24 715.939 L1233.24 723.578 L1282.26 723.578 L1282.26 715.939 L1286.81 715.939 L1286.81 729.434 L1228.69 729.434 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1230.7 703.208 L1230.7 693.627 L1263.03 681.501 L1230.7 669.31 L1230.7 659.73 L1278.22 659.73 L1278.22 666 L1236.49 666 L1269.08 678.254 L1269.08 684.715 L1236.49 696.969 L1278.22 696.969 L1278.22 703.208 L1230.7 703.208 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1230.7 651.2 L1230.7 644.707 L1270.86 634.713 L1230.7 624.75 L1230.7 617.525 L1270.86 607.531 L1230.7 597.569 L1230.7 591.044 L1278.22 602.98 L1278.22 611.064 L1236.97 621.09 L1278.22 631.212 L1278.22 639.296 L1230.7 651.2 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1228.69 569.082 L1286.81 569.082 L1286.81 582.578 L1282.26 582.578 L1282.26 574.971 L1233.24 574.971 L1233.24 582.578 L1228.69 582.578 L1228.69 569.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1801.44 12.096 L1809.71 12.096 L1822.43 63.2184 L1835.1 12.096 L1844.3 12.096 L1857.02 63.2184 L1869.7 12.096 L1878 12.096 L1862.81 72.576 L1852.52 72.576 L1839.76 20.0763 L1826.88 72.576 L1816.59 72.576 L1801.44 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1886.71 27.2059 L1894.17 27.2059 L1894.17 72.576 L1886.71 72.576 L1886.71 27.2059 M1886.71 9.54393 L1894.17 9.54393 L1894.17 18.9825 L1886.71 18.9825 L1886.71 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1947.48 45.1919 L1947.48 72.576 L1940.02 72.576 L1940.02 45.4349 Q1940.02 38.994 1937.51 35.7938 Q1935 32.5936 1929.98 32.5936 Q1923.94 32.5936 1920.46 36.4419 Q1916.97 40.2903 1916.97 46.9338 L1916.97 72.576 L1909.48 72.576 L1909.48 27.2059 L1916.97 27.2059 L1916.97 34.2544 Q1919.65 30.163 1923.25 28.1376 Q1926.9 26.1121 1931.64 26.1121 Q1939.46 26.1121 1943.47 30.9732 Q1947.48 35.7938 1947.48 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M1992.2 34.0924 L1992.2 9.54393 L1999.65 9.54393 L1999.65 72.576 L1992.2 72.576 L1992.2 65.7705 Q1989.85 69.8214 1986.24 71.8063 Q1982.68 73.7508 1977.66 73.7508 Q1969.43 73.7508 1964.25 67.1883 Q1959.1 60.6258 1959.1 49.9314 Q1959.1 39.2371 1964.25 32.6746 Q1969.43 26.1121 1977.66 26.1121 Q1982.68 26.1121 1986.24 28.0971 Q1989.85 30.0415 1992.2 34.0924 M1966.8 49.9314 Q1966.8 58.1548 1970.16 62.8538 Q1973.56 67.5124 1979.48 67.5124 Q1985.39 67.5124 1988.8 62.8538 Q1992.2 58.1548 1992.2 49.9314 Q1992.2 41.7081 1988.8 37.0496 Q1985.39 32.3505 1979.48 32.3505 Q1973.56 32.3505 1970.16 37.0496 Q1966.8 41.7081 1966.8 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip473)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1473.94,1386.4 1545.04,1386.4 1616.14,1386.4 1687.24,1386.4 1758.34,1386.4 1829.45,1386.4 1900.55,1386.4 1971.65,160.256 2042.75,160.256 2113.85,160.256 2184.96,160.256 2256.06,160.256 2327.16,160.256 "/>
+<polyline clip-path="url(#clip473)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1473.94,160.256 1545.04,160.256 1616.14,160.256 1687.24,160.256 1758.34,160.256 1829.45,160.256 1900.55,160.256 1971.65,1386.4 2042.75,1386.4 2113.85,1386.4 2184.96,1386.4 2256.06,1386.4 2327.16,1386.4 "/>
+<path clip-path="url(#clip470)" d="M2032.63 1379.86 L2322.61 1379.86 L2322.61 1224.34 L2032.63 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip470)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2032.63,1379.86 2322.61,1379.86 2322.61,1224.34 2032.63,1224.34 2032.63,1379.86 "/>
+<polyline clip-path="url(#clip470)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2042.68,1276.18 2102.97,1276.18 "/>
+<path clip-path="url(#clip470)" d="M2117.7 1262.74 L2117.7 1289.61 L2123.35 1289.61 Q2130.5 1289.61 2133.81 1286.37 Q2137.14 1283.13 2137.14 1276.14 Q2137.14 1269.2 2133.81 1265.98 Q2130.5 1262.74 2123.35 1262.74 L2117.7 1262.74 M2113.02 1258.9 L2122.63 1258.9 Q2132.68 1258.9 2137.38 1263.09 Q2142.07 1267.25 2142.07 1276.14 Q2142.07 1285.08 2137.35 1289.27 Q2132.63 1293.46 2122.63 1293.46 L2113.02 1293.46 L2113.02 1258.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2149.34 1267.53 L2153.6 1267.53 L2153.6 1293.46 L2149.34 1293.46 L2149.34 1267.53 M2149.34 1257.44 L2153.6 1257.44 L2153.6 1262.83 L2149.34 1262.83 L2149.34 1257.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2179.04 1268.29 L2179.04 1272.32 Q2177.24 1271.4 2175.29 1270.93 Q2173.35 1270.47 2171.26 1270.47 Q2168.09 1270.47 2166.5 1271.44 Q2164.92 1272.41 2164.92 1274.36 Q2164.92 1275.84 2166.06 1276.7 Q2167.19 1277.53 2170.62 1278.29 L2172.07 1278.62 Q2176.61 1279.59 2178.51 1281.37 Q2180.43 1283.13 2180.43 1286.3 Q2180.43 1289.91 2177.56 1292.02 Q2174.71 1294.13 2169.71 1294.13 Q2167.63 1294.13 2165.36 1293.71 Q2163.12 1293.32 2160.62 1292.51 L2160.62 1288.11 Q2162.98 1289.34 2165.27 1289.96 Q2167.56 1290.56 2169.81 1290.56 Q2172.81 1290.56 2174.44 1289.54 Q2176.06 1288.5 2176.06 1286.63 Q2176.06 1284.89 2174.88 1283.97 Q2173.72 1283.04 2169.76 1282.18 L2168.28 1281.84 Q2164.32 1281 2162.56 1279.29 Q2160.8 1277.55 2160.8 1274.54 Q2160.8 1270.89 2163.39 1268.9 Q2165.99 1266.91 2170.75 1266.91 Q2173.12 1266.91 2175.2 1267.25 Q2177.28 1267.6 2179.04 1268.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2191.33 1289.57 L2191.33 1303.32 L2187.05 1303.32 L2187.05 1267.53 L2191.33 1267.53 L2191.33 1271.47 Q2192.68 1269.15 2194.71 1268.04 Q2196.77 1266.91 2199.62 1266.91 Q2204.34 1266.91 2207.28 1270.66 Q2210.25 1274.41 2210.25 1280.52 Q2210.25 1286.63 2207.28 1290.38 Q2204.34 1294.13 2199.62 1294.13 Q2196.77 1294.13 2194.71 1293.02 Q2192.68 1291.88 2191.33 1289.57 M2205.82 1280.52 Q2205.82 1275.82 2203.88 1273.16 Q2201.96 1270.47 2198.58 1270.47 Q2195.2 1270.47 2193.25 1273.16 Q2191.33 1275.82 2191.33 1280.52 Q2191.33 1285.22 2193.25 1287.9 Q2195.2 1290.56 2198.58 1290.56 Q2201.96 1290.56 2203.88 1287.9 Q2205.82 1285.22 2205.82 1280.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2229.09 1280.42 Q2223.93 1280.42 2221.94 1281.6 Q2219.94 1282.79 2219.94 1285.63 Q2219.94 1287.9 2221.43 1289.24 Q2222.93 1290.56 2225.5 1290.56 Q2229.04 1290.56 2231.17 1288.06 Q2233.32 1285.54 2233.32 1281.37 L2233.32 1280.42 L2229.09 1280.42 M2237.58 1278.66 L2237.58 1293.46 L2233.32 1293.46 L2233.32 1289.52 Q2231.87 1291.88 2229.69 1293.02 Q2227.51 1294.13 2224.37 1294.13 Q2220.38 1294.13 2218.02 1291.91 Q2215.69 1289.66 2215.69 1285.91 Q2215.69 1281.54 2218.6 1279.31 Q2221.54 1277.09 2227.35 1277.09 L2233.32 1277.09 L2233.32 1276.67 Q2233.32 1273.73 2231.38 1272.14 Q2229.46 1270.52 2225.96 1270.52 Q2223.74 1270.52 2221.63 1271.05 Q2219.53 1271.58 2217.58 1272.65 L2217.58 1268.71 Q2219.92 1267.81 2222.12 1267.37 Q2224.32 1266.91 2226.4 1266.91 Q2232.03 1266.91 2234.81 1269.82 Q2237.58 1272.74 2237.58 1278.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2250.57 1260.17 L2250.57 1267.53 L2259.34 1267.53 L2259.34 1270.84 L2250.57 1270.84 L2250.57 1284.91 Q2250.57 1288.09 2251.43 1288.99 Q2252.31 1289.89 2254.97 1289.89 L2259.34 1289.89 L2259.34 1293.46 L2254.97 1293.46 Q2250.04 1293.46 2248.16 1291.63 Q2246.29 1289.78 2246.29 1284.91 L2246.29 1270.84 L2243.16 1270.84 L2243.16 1267.53 L2246.29 1267.53 L2246.29 1260.17 L2250.57 1260.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2283.6 1268.53 L2283.6 1272.51 Q2281.8 1271.51 2279.97 1271.03 Q2278.16 1270.52 2276.31 1270.52 Q2272.17 1270.52 2269.87 1273.16 Q2267.58 1275.77 2267.58 1280.52 Q2267.58 1285.26 2269.87 1287.9 Q2272.17 1290.52 2276.31 1290.52 Q2278.16 1290.52 2279.97 1290.03 Q2281.8 1289.52 2283.6 1288.53 L2283.6 1292.46 Q2281.82 1293.29 2279.9 1293.71 Q2278 1294.13 2275.85 1294.13 Q2269.99 1294.13 2266.54 1290.45 Q2263.09 1286.77 2263.09 1280.52 Q2263.09 1274.17 2266.56 1270.54 Q2270.06 1266.91 2276.12 1266.91 Q2278.09 1266.91 2279.97 1267.32 Q2281.84 1267.72 2283.6 1268.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2312.56 1277.81 L2312.56 1293.46 L2308.3 1293.46 L2308.3 1277.95 Q2308.3 1274.27 2306.87 1272.44 Q2305.43 1270.61 2302.56 1270.61 Q2299.11 1270.61 2297.12 1272.81 Q2295.13 1275.01 2295.13 1278.8 L2295.13 1293.46 L2290.85 1293.46 L2290.85 1257.44 L2295.13 1257.44 L2295.13 1271.56 Q2296.66 1269.22 2298.72 1268.06 Q2300.8 1266.91 2303.51 1266.91 Q2307.98 1266.91 2310.27 1269.68 Q2312.56 1272.44 2312.56 1277.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip470)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2042.68,1328.02 2102.97,1328.02 "/>
+<path clip-path="url(#clip470)" d="M2135.27 1311.87 L2135.27 1316.43 Q2132.61 1315.16 2130.25 1314.53 Q2127.88 1313.91 2125.69 1313.91 Q2121.87 1313.91 2119.78 1315.39 Q2117.72 1316.87 2117.72 1319.6 Q2117.72 1321.89 2119.09 1323.07 Q2120.48 1324.23 2124.32 1324.95 L2127.14 1325.53 Q2132.38 1326.52 2134.85 1329.05 Q2137.35 1331.55 2137.35 1335.76 Q2137.35 1340.78 2133.97 1343.37 Q2130.62 1345.97 2124.11 1345.97 Q2121.66 1345.97 2118.88 1345.41 Q2116.13 1344.86 2113.16 1343.77 L2113.16 1338.95 Q2116.01 1340.55 2118.74 1341.36 Q2121.47 1342.17 2124.11 1342.17 Q2128.12 1342.17 2130.29 1340.6 Q2132.47 1339.02 2132.47 1336.11 Q2132.47 1333.56 2130.89 1332.13 Q2129.34 1330.69 2125.78 1329.97 L2122.93 1329.42 Q2117.7 1328.38 2115.36 1326.15 Q2113.02 1323.93 2113.02 1319.97 Q2113.02 1315.39 2116.24 1312.75 Q2119.48 1310.11 2125.15 1310.11 Q2127.58 1310.11 2130.11 1310.55 Q2132.63 1310.99 2135.27 1311.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2148.58 1341.41 L2148.58 1355.16 L2144.3 1355.16 L2144.3 1319.37 L2148.58 1319.37 L2148.58 1323.31 Q2149.92 1320.99 2151.96 1319.88 Q2154.02 1318.75 2156.87 1318.75 Q2161.59 1318.75 2164.53 1322.5 Q2167.49 1326.25 2167.49 1332.36 Q2167.49 1338.47 2164.53 1342.22 Q2161.59 1345.97 2156.87 1345.97 Q2154.02 1345.97 2151.96 1344.86 Q2149.92 1343.72 2148.58 1341.41 M2163.07 1332.36 Q2163.07 1327.66 2161.13 1325 Q2159.2 1322.31 2155.82 1322.31 Q2152.44 1322.31 2150.5 1325 Q2148.58 1327.66 2148.58 1332.36 Q2148.58 1337.06 2150.5 1339.74 Q2152.44 1342.4 2155.82 1342.4 Q2159.2 1342.4 2161.13 1339.74 Q2163.07 1337.06 2163.07 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2174.55 1319.37 L2178.81 1319.37 L2178.81 1345.3 L2174.55 1345.3 L2174.55 1319.37 M2174.55 1309.28 L2178.81 1309.28 L2178.81 1314.67 L2174.55 1314.67 L2174.55 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2187.72 1309.28 L2191.98 1309.28 L2191.98 1345.3 L2187.72 1345.3 L2187.72 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2200.89 1309.28 L2205.15 1309.28 L2205.15 1345.3 L2200.89 1345.3 L2200.89 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2225.85 1332.26 Q2220.69 1332.26 2218.69 1333.44 Q2216.7 1334.63 2216.7 1337.47 Q2216.7 1339.74 2218.19 1341.08 Q2219.69 1342.4 2222.26 1342.4 Q2225.8 1342.4 2227.93 1339.9 Q2230.08 1337.38 2230.08 1333.21 L2230.08 1332.26 L2225.85 1332.26 M2234.34 1330.5 L2234.34 1345.3 L2230.08 1345.3 L2230.08 1341.36 Q2228.62 1343.72 2226.45 1344.86 Q2224.27 1345.97 2221.12 1345.97 Q2217.14 1345.97 2214.78 1343.75 Q2212.44 1341.5 2212.44 1337.75 Q2212.44 1333.38 2215.36 1331.15 Q2218.3 1328.93 2224.11 1328.93 L2230.08 1328.93 L2230.08 1328.51 Q2230.08 1325.57 2228.14 1323.98 Q2226.22 1322.36 2222.72 1322.36 Q2220.5 1322.36 2218.39 1322.89 Q2216.29 1323.42 2214.34 1324.49 L2214.34 1320.55 Q2216.68 1319.65 2218.88 1319.21 Q2221.08 1318.75 2223.16 1318.75 Q2228.79 1318.75 2231.56 1321.66 Q2234.34 1324.58 2234.34 1330.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2260.18 1332.03 Q2260.18 1327.4 2258.25 1324.86 Q2256.36 1322.31 2252.91 1322.31 Q2249.48 1322.31 2247.56 1324.86 Q2245.66 1327.4 2245.66 1332.03 Q2245.66 1336.64 2247.56 1339.19 Q2249.48 1341.73 2252.91 1341.73 Q2256.36 1341.73 2258.25 1339.19 Q2260.18 1336.64 2260.18 1332.03 M2264.43 1342.08 Q2264.43 1348.7 2261.5 1351.92 Q2258.56 1355.16 2252.49 1355.16 Q2250.25 1355.16 2248.25 1354.81 Q2246.26 1354.49 2244.39 1353.79 L2244.39 1349.65 Q2246.26 1350.67 2248.09 1351.15 Q2249.92 1351.64 2251.82 1351.64 Q2256.01 1351.64 2258.09 1349.44 Q2260.18 1347.26 2260.18 1342.84 L2260.18 1340.74 Q2258.86 1343.03 2256.8 1344.16 Q2254.74 1345.3 2251.87 1345.3 Q2247.1 1345.3 2244.18 1341.66 Q2241.26 1338.03 2241.26 1332.03 Q2241.26 1326.01 2244.18 1322.38 Q2247.1 1318.75 2251.87 1318.75 Q2254.74 1318.75 2256.8 1319.88 Q2258.86 1321.01 2260.18 1323.31 L2260.18 1319.37 L2264.43 1319.37 L2264.43 1342.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip470)" d="M2295.38 1331.27 L2295.38 1333.35 L2275.8 1333.35 Q2276.08 1337.75 2278.44 1340.06 Q2280.82 1342.36 2285.06 1342.36 Q2287.51 1342.36 2289.81 1341.75 Q2292.12 1341.15 2294.39 1339.95 L2294.39 1343.98 Q2292.1 1344.95 2289.69 1345.46 Q2287.28 1345.97 2284.81 1345.97 Q2278.6 1345.97 2274.97 1342.36 Q2271.36 1338.75 2271.36 1332.59 Q2271.36 1326.22 2274.78 1322.5 Q2278.23 1318.75 2284.06 1318.75 Q2289.3 1318.75 2292.33 1322.13 Q2295.38 1325.48 2295.38 1331.27 M2291.12 1330.02 Q2291.08 1326.52 2289.16 1324.44 Q2287.26 1322.36 2284.11 1322.36 Q2280.55 1322.36 2278.39 1324.37 Q2276.26 1326.38 2275.94 1330.04 L2291.12 1330.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/power_systems/e09dc019.svg b/previews/PR3562/tutorials/applications/power_systems/e02e8c47.svg
similarity index 86%
rename from previews/PR3562/tutorials/applications/power_systems/e09dc019.svg
rename to previews/PR3562/tutorials/applications/power_systems/e02e8c47.svg
index d5b206bc7fa..385d741cb1e 100644
--- a/previews/PR3562/tutorials/applications/power_systems/e09dc019.svg
+++ b/previews/PR3562/tutorials/applications/power_systems/e02e8c47.svg
@@ -1,80 +1,80 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip710">
+  <clipPath id="clip530">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip710)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip530)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip711">
+  <clipPath id="clip531">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip710)" d="M249.542 1423.18 L1146.25 1423.18 L1146.25 123.472 L249.542 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip530)" d="M249.542 1423.18 L1146.25 1423.18 L1146.25 123.472 L249.542 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip712">
+  <clipPath id="clip532">
     <rect x="249" y="123" width="898" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="368.915,1423.18 368.915,123.472 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="603.902,1423.18 603.902,123.472 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.889,1423.18 838.889,123.472 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1073.88,1423.18 1073.88,123.472 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1423.18 1146.25,1423.18 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="368.915,1423.18 368.915,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="603.902,1423.18 603.902,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.889,1423.18 838.889,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1073.88,1423.18 1073.88,1404.28 "/>
-<path clip-path="url(#clip710)" d="M328.534 1451.02 L346.89 1451.02 L346.89 1454.96 L332.816 1454.96 L332.816 1463.43 Q333.834 1463.08 334.853 1462.92 Q335.871 1462.73 336.89 1462.73 Q342.677 1462.73 346.057 1465.9 Q349.436 1469.08 349.436 1474.49 Q349.436 1480.07 345.964 1483.17 Q342.492 1486.25 336.172 1486.25 Q333.996 1486.25 331.728 1485.88 Q329.483 1485.51 327.075 1484.77 L327.075 1480.07 Q329.159 1481.2 331.381 1481.76 Q333.603 1482.32 336.08 1482.32 Q340.084 1482.32 342.422 1480.21 Q344.76 1478.1 344.76 1474.49 Q344.76 1470.88 342.422 1468.77 Q340.084 1466.67 336.08 1466.67 Q334.205 1466.67 332.33 1467.08 Q330.478 1467.5 328.534 1468.38 L328.534 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M368.649 1454.1 Q365.038 1454.1 363.209 1457.66 Q361.404 1461.2 361.404 1468.33 Q361.404 1475.44 363.209 1479.01 Q365.038 1482.55 368.649 1482.55 Q372.283 1482.55 374.089 1479.01 Q375.918 1475.44 375.918 1468.33 Q375.918 1461.2 374.089 1457.66 Q372.283 1454.1 368.649 1454.1 M368.649 1450.39 Q374.459 1450.39 377.515 1455 Q380.593 1459.58 380.593 1468.33 Q380.593 1477.06 377.515 1481.67 Q374.459 1486.25 368.649 1486.25 Q362.839 1486.25 359.76 1481.67 Q356.705 1477.06 356.705 1468.33 Q356.705 1459.58 359.76 1455 Q362.839 1450.39 368.649 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M398.811 1454.1 Q395.2 1454.1 393.371 1457.66 Q391.566 1461.2 391.566 1468.33 Q391.566 1475.44 393.371 1479.01 Q395.2 1482.55 398.811 1482.55 Q402.445 1482.55 404.251 1479.01 Q406.079 1475.44 406.079 1468.33 Q406.079 1461.2 404.251 1457.66 Q402.445 1454.1 398.811 1454.1 M398.811 1450.39 Q404.621 1450.39 407.677 1455 Q410.755 1459.58 410.755 1468.33 Q410.755 1477.06 407.677 1481.67 Q404.621 1486.25 398.811 1486.25 Q393.001 1486.25 389.922 1481.67 Q386.867 1477.06 386.867 1468.33 Q386.867 1459.58 389.922 1455 Q393.001 1450.39 398.811 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M548.428 1481.64 L556.067 1481.64 L556.067 1455.28 L547.756 1456.95 L547.756 1452.69 L556.02 1451.02 L560.696 1451.02 L560.696 1481.64 L568.335 1481.64 L568.335 1485.58 L548.428 1485.58 L548.428 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M587.779 1454.1 Q584.168 1454.1 582.34 1457.66 Q580.534 1461.2 580.534 1468.33 Q580.534 1475.44 582.34 1479.01 Q584.168 1482.55 587.779 1482.55 Q591.414 1482.55 593.219 1479.01 Q595.048 1475.44 595.048 1468.33 Q595.048 1461.2 593.219 1457.66 Q591.414 1454.1 587.779 1454.1 M587.779 1450.39 Q593.59 1450.39 596.645 1455 Q599.724 1459.58 599.724 1468.33 Q599.724 1477.06 596.645 1481.67 Q593.59 1486.25 587.779 1486.25 Q581.969 1486.25 578.891 1481.67 Q575.835 1477.06 575.835 1468.33 Q575.835 1459.58 578.891 1455 Q581.969 1450.39 587.779 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M617.941 1454.1 Q614.33 1454.1 612.501 1457.66 Q610.696 1461.2 610.696 1468.33 Q610.696 1475.44 612.501 1479.01 Q614.33 1482.55 617.941 1482.55 Q621.576 1482.55 623.381 1479.01 Q625.21 1475.44 625.21 1468.33 Q625.21 1461.2 623.381 1457.66 Q621.576 1454.1 617.941 1454.1 M617.941 1450.39 Q623.751 1450.39 626.807 1455 Q629.886 1459.58 629.886 1468.33 Q629.886 1477.06 626.807 1481.67 Q623.751 1486.25 617.941 1486.25 Q612.131 1486.25 609.052 1481.67 Q605.997 1477.06 605.997 1468.33 Q605.997 1459.58 609.052 1455 Q612.131 1450.39 617.941 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M648.103 1454.1 Q644.492 1454.1 642.663 1457.66 Q640.858 1461.2 640.858 1468.33 Q640.858 1475.44 642.663 1479.01 Q644.492 1482.55 648.103 1482.55 Q651.737 1482.55 653.543 1479.01 Q655.372 1475.44 655.372 1468.33 Q655.372 1461.2 653.543 1457.66 Q651.737 1454.1 648.103 1454.1 M648.103 1450.39 Q653.913 1450.39 656.969 1455 Q660.048 1459.58 660.048 1468.33 Q660.048 1477.06 656.969 1481.67 Q653.913 1486.25 648.103 1486.25 Q642.293 1486.25 639.214 1481.67 Q636.159 1477.06 636.159 1468.33 Q636.159 1459.58 639.214 1455 Q642.293 1450.39 648.103 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M783.414 1481.64 L791.053 1481.64 L791.053 1455.28 L782.743 1456.95 L782.743 1452.69 L791.007 1451.02 L795.683 1451.02 L795.683 1481.64 L803.322 1481.64 L803.322 1485.58 L783.414 1485.58 L783.414 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M812.812 1451.02 L831.169 1451.02 L831.169 1454.96 L817.095 1454.96 L817.095 1463.43 Q818.113 1463.08 819.132 1462.92 Q820.15 1462.73 821.169 1462.73 Q826.956 1462.73 830.336 1465.9 Q833.715 1469.08 833.715 1474.49 Q833.715 1480.07 830.243 1483.17 Q826.771 1486.25 820.451 1486.25 Q818.275 1486.25 816.007 1485.88 Q813.762 1485.51 811.354 1484.77 L811.354 1480.07 Q813.437 1481.2 815.66 1481.76 Q817.882 1482.32 820.359 1482.32 Q824.363 1482.32 826.701 1480.21 Q829.039 1478.1 829.039 1474.49 Q829.039 1470.88 826.701 1468.77 Q824.363 1466.67 820.359 1466.67 Q818.484 1466.67 816.609 1467.08 Q814.757 1467.5 812.812 1468.38 L812.812 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M852.928 1454.1 Q849.317 1454.1 847.488 1457.66 Q845.683 1461.2 845.683 1468.33 Q845.683 1475.44 847.488 1479.01 Q849.317 1482.55 852.928 1482.55 Q856.562 1482.55 858.368 1479.01 Q860.196 1475.44 860.196 1468.33 Q860.196 1461.2 858.368 1457.66 Q856.562 1454.1 852.928 1454.1 M852.928 1450.39 Q858.738 1450.39 861.794 1455 Q864.872 1459.58 864.872 1468.33 Q864.872 1477.06 861.794 1481.67 Q858.738 1486.25 852.928 1486.25 Q847.118 1486.25 844.039 1481.67 Q840.984 1477.06 840.984 1468.33 Q840.984 1459.58 844.039 1455 Q847.118 1450.39 852.928 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M883.09 1454.1 Q879.479 1454.1 877.65 1457.66 Q875.845 1461.2 875.845 1468.33 Q875.845 1475.44 877.65 1479.01 Q879.479 1482.55 883.09 1482.55 Q886.724 1482.55 888.53 1479.01 Q890.358 1475.44 890.358 1468.33 Q890.358 1461.2 888.53 1457.66 Q886.724 1454.1 883.09 1454.1 M883.09 1450.39 Q888.9 1450.39 891.956 1455 Q895.034 1459.58 895.034 1468.33 Q895.034 1477.06 891.956 1481.67 Q888.9 1486.25 883.09 1486.25 Q877.28 1486.25 874.201 1481.67 Q871.145 1477.06 871.145 1468.33 Q871.145 1459.58 874.201 1455 Q877.28 1450.39 883.09 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1022.49 1481.64 L1038.81 1481.64 L1038.81 1485.58 L1016.86 1485.58 L1016.86 1481.64 Q1019.52 1478.89 1024.11 1474.26 Q1028.71 1469.61 1029.89 1468.27 Q1032.14 1465.74 1033.02 1464.01 Q1033.92 1462.25 1033.92 1460.56 Q1033.92 1457.8 1031.98 1456.07 Q1030.06 1454.33 1026.95 1454.33 Q1024.76 1454.33 1022.3 1455.09 Q1019.87 1455.86 1017.09 1457.41 L1017.09 1452.69 Q1019.92 1451.55 1022.37 1450.97 Q1024.82 1450.39 1026.86 1450.39 Q1032.23 1450.39 1035.43 1453.08 Q1038.62 1455.77 1038.62 1460.26 Q1038.62 1462.39 1037.81 1464.31 Q1037.02 1466.2 1034.92 1468.8 Q1034.34 1469.47 1031.24 1472.69 Q1028.13 1475.88 1022.49 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1058.62 1454.1 Q1055.01 1454.1 1053.18 1457.66 Q1051.38 1461.2 1051.38 1468.33 Q1051.38 1475.44 1053.18 1479.01 Q1055.01 1482.55 1058.62 1482.55 Q1062.26 1482.55 1064.06 1479.01 Q1065.89 1475.44 1065.89 1468.33 Q1065.89 1461.2 1064.06 1457.66 Q1062.26 1454.1 1058.62 1454.1 M1058.62 1450.39 Q1064.43 1450.39 1067.49 1455 Q1070.57 1459.58 1070.57 1468.33 Q1070.57 1477.06 1067.49 1481.67 Q1064.43 1486.25 1058.62 1486.25 Q1052.81 1486.25 1049.73 1481.67 Q1046.68 1477.06 1046.68 1468.33 Q1046.68 1459.58 1049.73 1455 Q1052.81 1450.39 1058.62 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1088.78 1454.1 Q1085.17 1454.1 1083.34 1457.66 Q1081.54 1461.2 1081.54 1468.33 Q1081.54 1475.44 1083.34 1479.01 Q1085.17 1482.55 1088.78 1482.55 Q1092.42 1482.55 1094.22 1479.01 Q1096.05 1475.44 1096.05 1468.33 Q1096.05 1461.2 1094.22 1457.66 Q1092.42 1454.1 1088.78 1454.1 M1088.78 1450.39 Q1094.59 1450.39 1097.65 1455 Q1100.73 1459.58 1100.73 1468.33 Q1100.73 1477.06 1097.65 1481.67 Q1094.59 1486.25 1088.78 1486.25 Q1082.97 1486.25 1079.89 1481.67 Q1076.84 1477.06 1076.84 1468.33 Q1076.84 1459.58 1079.89 1455 Q1082.97 1450.39 1088.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1118.94 1454.1 Q1115.33 1454.1 1113.5 1457.66 Q1111.7 1461.2 1111.7 1468.33 Q1111.7 1475.44 1113.5 1479.01 Q1115.33 1482.55 1118.94 1482.55 Q1122.58 1482.55 1124.38 1479.01 Q1126.21 1475.44 1126.21 1468.33 Q1126.21 1461.2 1124.38 1457.66 Q1122.58 1454.1 1118.94 1454.1 M1118.94 1450.39 Q1124.75 1450.39 1127.81 1455 Q1130.89 1459.58 1130.89 1468.33 Q1130.89 1477.06 1127.81 1481.67 Q1124.75 1486.25 1118.94 1486.25 Q1113.13 1486.25 1110.06 1481.67 Q1107 1477.06 1107 1468.33 Q1107 1459.58 1110.06 1455 Q1113.13 1450.39 1118.94 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M475.972 1525.81 L475.972 1562.76 L483.738 1562.76 Q493.573 1562.76 498.125 1558.3 Q502.708 1553.85 502.708 1544.24 Q502.708 1534.69 498.125 1530.26 Q493.573 1525.81 483.738 1525.81 L475.972 1525.81 M469.543 1520.52 L482.752 1520.52 Q496.565 1520.52 503.026 1526.28 Q509.488 1532.01 509.488 1544.24 Q509.488 1556.52 502.995 1562.28 Q496.502 1568.04 482.752 1568.04 L469.543 1568.04 L469.543 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M549.974 1548.76 L549.974 1551.62 L523.047 1551.62 Q523.429 1557.67 526.675 1560.85 Q529.953 1564 535.778 1564 Q539.152 1564 542.303 1563.17 Q545.486 1562.35 548.605 1560.69 L548.605 1566.23 Q545.454 1567.57 542.144 1568.27 Q538.834 1568.97 535.428 1568.97 Q526.898 1568.97 521.901 1564 Q516.936 1559.04 516.936 1550.57 Q516.936 1541.82 521.646 1536.69 Q526.389 1531.54 534.409 1531.54 Q541.603 1531.54 545.772 1536.18 Q549.974 1540.8 549.974 1548.76 M544.117 1547.04 Q544.053 1542.23 541.412 1539.37 Q538.802 1536.5 534.473 1536.5 Q529.571 1536.5 526.611 1539.27 Q523.683 1542.04 523.238 1547.07 L544.117 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M587.34 1539.24 Q589.536 1535.29 592.592 1533.41 Q595.648 1531.54 599.785 1531.54 Q605.355 1531.54 608.379 1535.45 Q611.403 1539.33 611.403 1546.53 L611.403 1568.04 L605.514 1568.04 L605.514 1546.72 Q605.514 1541.59 603.7 1539.11 Q601.886 1536.63 598.162 1536.63 Q593.61 1536.63 590.969 1539.65 Q588.327 1542.68 588.327 1547.9 L588.327 1568.04 L582.439 1568.04 L582.439 1546.72 Q582.439 1541.56 580.624 1539.11 Q578.81 1536.63 575.023 1536.63 Q570.535 1536.63 567.893 1539.68 Q565.251 1542.71 565.251 1547.9 L565.251 1568.04 L559.363 1568.04 L559.363 1532.4 L565.251 1532.4 L565.251 1537.93 Q567.256 1534.66 570.057 1533.1 Q572.858 1531.54 576.71 1531.54 Q580.593 1531.54 583.298 1533.51 Q586.035 1535.48 587.34 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M639.284 1550.12 Q632.187 1550.12 629.449 1551.75 Q626.712 1553.37 626.712 1557.29 Q626.712 1560.4 628.749 1562.25 Q630.818 1564.07 634.351 1564.07 Q639.221 1564.07 642.149 1560.63 Q645.109 1557.16 645.109 1551.43 L645.109 1550.12 L639.284 1550.12 M650.965 1547.71 L650.965 1568.04 L645.109 1568.04 L645.109 1562.63 Q643.104 1565.88 640.112 1567.44 Q637.12 1568.97 632.791 1568.97 Q627.317 1568.97 624.07 1565.91 Q620.856 1562.82 620.856 1557.67 Q620.856 1551.65 624.866 1548.6 Q628.908 1545.54 636.897 1545.54 L645.109 1545.54 L645.109 1544.97 Q645.109 1540.93 642.435 1538.73 Q639.794 1536.5 634.988 1536.5 Q631.932 1536.5 629.036 1537.23 Q626.139 1537.97 623.466 1539.43 L623.466 1534.02 Q626.68 1532.78 629.704 1532.17 Q632.728 1531.54 635.592 1531.54 Q643.327 1531.54 647.146 1535.55 Q650.965 1539.56 650.965 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M692.661 1546.53 L692.661 1568.04 L686.804 1568.04 L686.804 1546.72 Q686.804 1541.66 684.831 1539.14 Q682.858 1536.63 678.911 1536.63 Q674.169 1536.63 671.431 1539.65 Q668.694 1542.68 668.694 1547.9 L668.694 1568.04 L662.806 1568.04 L662.806 1532.4 L668.694 1532.4 L668.694 1537.93 Q670.795 1534.72 673.627 1533.13 Q676.492 1531.54 680.216 1531.54 Q686.359 1531.54 689.51 1535.36 Q692.661 1539.14 692.661 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M727.8 1537.81 L727.8 1518.52 L733.656 1518.52 L733.656 1568.04 L727.8 1568.04 L727.8 1562.7 Q725.954 1565.88 723.121 1567.44 Q720.32 1568.97 716.373 1568.97 Q709.912 1568.97 705.838 1563.81 Q701.796 1558.65 701.796 1550.25 Q701.796 1541.85 705.838 1536.69 Q709.912 1531.54 716.373 1531.54 Q720.32 1531.54 723.121 1533.1 Q725.954 1534.62 727.8 1537.81 M707.843 1550.25 Q707.843 1556.71 710.485 1560.4 Q713.158 1564.07 717.805 1564.07 Q722.452 1564.07 725.126 1560.4 Q727.8 1556.71 727.8 1550.25 Q727.8 1543.79 725.126 1540.13 Q722.452 1536.44 717.805 1536.44 Q713.158 1536.44 710.485 1540.13 Q707.843 1543.79 707.843 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M765.898 1518.52 L779.394 1518.52 L779.394 1523.07 L771.755 1523.07 L771.755 1572.09 L779.394 1572.09 L779.394 1576.64 L765.898 1576.64 L765.898 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M792.125 1520.52 L801.705 1520.52 L813.832 1552.86 L826.022 1520.52 L835.603 1520.52 L835.603 1568.04 L829.333 1568.04 L829.333 1526.32 L817.079 1558.91 L810.617 1558.91 L798.363 1526.32 L798.363 1568.04 L792.125 1568.04 L792.125 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M844.133 1520.52 L850.626 1520.52 L860.62 1560.69 L870.582 1520.52 L877.807 1520.52 L887.802 1560.69 L897.764 1520.52 L904.289 1520.52 L892.353 1568.04 L884.269 1568.04 L874.243 1526.79 L864.121 1568.04 L856.037 1568.04 L844.133 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M926.25 1518.52 L926.25 1576.64 L912.755 1576.64 L912.755 1572.09 L920.362 1572.09 L920.362 1523.07 L912.755 1523.07 L912.755 1518.52 L926.25 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,1386.4 1146.25,1386.4 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,1079.86 1146.25,1079.86 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,773.326 1146.25,773.326 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,466.791 1146.25,466.791 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,160.256 1146.25,160.256 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1423.18 249.542,123.472 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1386.4 268.44,1386.4 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1079.86 268.44,1079.86 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,773.326 268.44,773.326 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,466.791 268.44,466.791 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,160.256 268.44,160.256 "/>
-<path clip-path="url(#clip710)" d="M126.205 1372.19 Q122.593 1372.19 120.765 1375.76 Q118.959 1379.3 118.959 1386.43 Q118.959 1393.54 120.765 1397.1 Q122.593 1400.64 126.205 1400.64 Q129.839 1400.64 131.644 1397.1 Q133.473 1393.54 133.473 1386.43 Q133.473 1379.3 131.644 1375.76 Q129.839 1372.19 126.205 1372.19 M126.205 1368.49 Q132.015 1368.49 135.07 1373.1 Q138.149 1377.68 138.149 1386.43 Q138.149 1395.16 135.07 1399.76 Q132.015 1404.35 126.205 1404.35 Q120.394 1404.35 117.316 1399.76 Q114.26 1395.16 114.26 1386.43 Q114.26 1377.68 117.316 1373.1 Q120.394 1368.49 126.205 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M146.366 1397.8 L151.251 1397.8 L151.251 1403.68 L146.366 1403.68 L146.366 1397.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M171.436 1372.19 Q167.825 1372.19 165.996 1375.76 Q164.19 1379.3 164.19 1386.43 Q164.19 1393.54 165.996 1397.1 Q167.825 1400.64 171.436 1400.64 Q175.07 1400.64 176.876 1397.1 Q178.704 1393.54 178.704 1386.43 Q178.704 1379.3 176.876 1375.76 Q175.07 1372.19 171.436 1372.19 M171.436 1368.49 Q177.246 1368.49 180.301 1373.1 Q183.38 1377.68 183.38 1386.43 Q183.38 1395.16 180.301 1399.76 Q177.246 1404.35 171.436 1404.35 Q165.626 1404.35 162.547 1399.76 Q159.491 1395.16 159.491 1386.43 Q159.491 1377.68 162.547 1373.1 Q165.626 1368.49 171.436 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M201.598 1372.19 Q197.987 1372.19 196.158 1375.76 Q194.352 1379.3 194.352 1386.43 Q194.352 1393.54 196.158 1397.1 Q197.987 1400.64 201.598 1400.64 Q205.232 1400.64 207.037 1397.1 Q208.866 1393.54 208.866 1386.43 Q208.866 1379.3 207.037 1375.76 Q205.232 1372.19 201.598 1372.19 M201.598 1368.49 Q207.408 1368.49 210.463 1373.1 Q213.542 1377.68 213.542 1386.43 Q213.542 1395.16 210.463 1399.76 Q207.408 1404.35 201.598 1404.35 Q195.787 1404.35 192.709 1399.76 Q189.653 1395.16 189.653 1386.43 Q189.653 1377.68 192.709 1373.1 Q195.787 1368.49 201.598 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M127.2 1065.66 Q123.589 1065.66 121.76 1069.22 Q119.955 1072.77 119.955 1079.9 Q119.955 1087 121.76 1090.57 Q123.589 1094.11 127.2 1094.11 Q130.834 1094.11 132.64 1090.57 Q134.468 1087 134.468 1079.9 Q134.468 1072.77 132.64 1069.22 Q130.834 1065.66 127.2 1065.66 M127.2 1061.96 Q133.01 1061.96 136.066 1066.56 Q139.144 1071.15 139.144 1079.9 Q139.144 1088.62 136.066 1093.23 Q133.01 1097.81 127.2 1097.81 Q121.39 1097.81 118.311 1093.23 Q115.256 1088.62 115.256 1079.9 Q115.256 1071.15 118.311 1066.56 Q121.39 1061.96 127.2 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M147.362 1091.26 L152.246 1091.26 L152.246 1097.14 L147.362 1097.14 L147.362 1091.26 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M166.459 1093.21 L182.778 1093.21 L182.778 1097.14 L160.834 1097.14 L160.834 1093.21 Q163.496 1090.45 168.079 1085.82 Q172.686 1081.17 173.866 1079.83 Q176.112 1077.3 176.991 1075.57 Q177.894 1073.81 177.894 1072.12 Q177.894 1069.36 175.95 1067.63 Q174.028 1065.89 170.927 1065.89 Q168.727 1065.89 166.274 1066.65 Q163.843 1067.42 161.065 1068.97 L161.065 1064.25 Q163.89 1063.11 166.343 1062.53 Q168.797 1061.96 170.834 1061.96 Q176.204 1061.96 179.399 1064.64 Q182.593 1067.33 182.593 1071.82 Q182.593 1073.95 181.783 1075.87 Q180.996 1077.77 178.889 1080.36 Q178.311 1081.03 175.209 1084.25 Q172.107 1087.44 166.459 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M192.639 1062.58 L210.996 1062.58 L210.996 1066.52 L196.922 1066.52 L196.922 1074.99 Q197.94 1074.64 198.959 1074.48 Q199.977 1074.29 200.996 1074.29 Q206.783 1074.29 210.162 1077.47 Q213.542 1080.64 213.542 1086.05 Q213.542 1091.63 210.07 1094.73 Q206.598 1097.81 200.278 1097.81 Q198.102 1097.81 195.834 1097.44 Q193.588 1097.07 191.181 1096.33 L191.181 1091.63 Q193.264 1092.77 195.487 1093.32 Q197.709 1093.88 200.186 1093.88 Q204.19 1093.88 206.528 1091.77 Q208.866 1089.66 208.866 1086.05 Q208.866 1082.44 206.528 1080.34 Q204.19 1078.23 200.186 1078.23 Q198.311 1078.23 196.436 1078.65 Q194.584 1079.06 192.639 1079.94 L192.639 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M126.205 759.125 Q122.593 759.125 120.765 762.689 Q118.959 766.231 118.959 773.361 Q118.959 780.467 120.765 784.032 Q122.593 787.574 126.205 787.574 Q129.839 787.574 131.644 784.032 Q133.473 780.467 133.473 773.361 Q133.473 766.231 131.644 762.689 Q129.839 759.125 126.205 759.125 M126.205 755.421 Q132.015 755.421 135.07 760.027 Q138.149 764.611 138.149 773.361 Q138.149 782.088 135.07 786.694 Q132.015 791.277 126.205 791.277 Q120.394 791.277 117.316 786.694 Q114.26 782.088 114.26 773.361 Q114.26 764.611 117.316 760.027 Q120.394 755.421 126.205 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M146.366 784.726 L151.251 784.726 L151.251 790.606 L146.366 790.606 L146.366 784.726 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M161.482 756.046 L179.839 756.046 L179.839 759.981 L165.765 759.981 L165.765 768.453 Q166.783 768.106 167.802 767.944 Q168.82 767.759 169.839 767.759 Q175.626 767.759 179.005 770.93 Q182.385 774.101 182.385 779.518 Q182.385 785.097 178.913 788.199 Q175.44 791.277 169.121 791.277 Q166.945 791.277 164.677 790.907 Q162.431 790.537 160.024 789.796 L160.024 785.097 Q162.107 786.231 164.329 786.787 Q166.552 787.342 169.028 787.342 Q173.033 787.342 175.371 785.236 Q177.709 783.129 177.709 779.518 Q177.709 775.907 175.371 773.801 Q173.033 771.694 169.028 771.694 Q167.153 771.694 165.278 772.111 Q163.427 772.527 161.482 773.407 L161.482 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M201.598 759.125 Q197.987 759.125 196.158 762.689 Q194.352 766.231 194.352 773.361 Q194.352 780.467 196.158 784.032 Q197.987 787.574 201.598 787.574 Q205.232 787.574 207.037 784.032 Q208.866 780.467 208.866 773.361 Q208.866 766.231 207.037 762.689 Q205.232 759.125 201.598 759.125 M201.598 755.421 Q207.408 755.421 210.463 760.027 Q213.542 764.611 213.542 773.361 Q213.542 782.088 210.463 786.694 Q207.408 791.277 201.598 791.277 Q195.787 791.277 192.709 786.694 Q189.653 782.088 189.653 773.361 Q189.653 764.611 192.709 760.027 Q195.787 755.421 201.598 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M127.2 452.59 Q123.589 452.59 121.76 456.155 Q119.955 459.696 119.955 466.826 Q119.955 473.932 121.76 477.497 Q123.589 481.039 127.2 481.039 Q130.834 481.039 132.64 477.497 Q134.468 473.932 134.468 466.826 Q134.468 459.696 132.64 456.155 Q130.834 452.59 127.2 452.59 M127.2 448.886 Q133.01 448.886 136.066 453.493 Q139.144 458.076 139.144 466.826 Q139.144 475.553 136.066 480.159 Q133.01 484.742 127.2 484.742 Q121.39 484.742 118.311 480.159 Q115.256 475.553 115.256 466.826 Q115.256 458.076 118.311 453.493 Q121.39 448.886 127.2 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M147.362 478.192 L152.246 478.192 L152.246 484.071 L147.362 484.071 L147.362 478.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M161.251 449.511 L183.473 449.511 L183.473 451.502 L170.927 484.071 L166.042 484.071 L177.848 453.446 L161.251 453.446 L161.251 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M192.639 449.511 L210.996 449.511 L210.996 453.446 L196.922 453.446 L196.922 461.918 Q197.94 461.571 198.959 461.409 Q199.977 461.224 200.996 461.224 Q206.783 461.224 210.162 464.395 Q213.542 467.567 213.542 472.983 Q213.542 478.562 210.07 481.664 Q206.598 484.742 200.278 484.742 Q198.102 484.742 195.834 484.372 Q193.588 484.002 191.181 483.261 L191.181 478.562 Q193.264 479.696 195.487 480.252 Q197.709 480.807 200.186 480.807 Q204.19 480.807 206.528 478.701 Q208.866 476.594 208.866 472.983 Q208.866 469.372 206.528 467.266 Q204.19 465.159 200.186 465.159 Q198.311 465.159 196.436 465.576 Q194.584 465.993 192.639 466.872 L192.639 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M117.015 173.601 L124.654 173.601 L124.654 147.236 L116.343 148.902 L116.343 144.643 L124.607 142.976 L129.283 142.976 L129.283 173.601 L136.922 173.601 L136.922 177.536 L117.015 177.536 L117.015 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M146.366 171.657 L151.251 171.657 L151.251 177.536 L146.366 177.536 L146.366 171.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M171.436 146.055 Q167.825 146.055 165.996 149.62 Q164.19 153.161 164.19 160.291 Q164.19 167.397 165.996 170.962 Q167.825 174.504 171.436 174.504 Q175.07 174.504 176.876 170.962 Q178.704 167.397 178.704 160.291 Q178.704 153.161 176.876 149.62 Q175.07 146.055 171.436 146.055 M171.436 142.351 Q177.246 142.351 180.301 146.958 Q183.38 151.541 183.38 160.291 Q183.38 169.018 180.301 173.624 Q177.246 178.208 171.436 178.208 Q165.626 178.208 162.547 173.624 Q159.491 169.018 159.491 160.291 Q159.491 151.541 162.547 146.958 Q165.626 142.351 171.436 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M201.598 146.055 Q197.987 146.055 196.158 149.62 Q194.352 153.161 194.352 160.291 Q194.352 167.397 196.158 170.962 Q197.987 174.504 201.598 174.504 Q205.232 174.504 207.037 170.962 Q208.866 167.397 208.866 160.291 Q208.866 153.161 207.037 149.62 Q205.232 146.055 201.598 146.055 M201.598 142.351 Q207.408 142.351 210.463 146.958 Q213.542 151.541 213.542 160.291 Q213.542 169.018 210.463 173.624 Q207.408 178.208 201.598 178.208 Q195.787 178.208 192.709 173.624 Q189.653 169.018 189.653 160.291 Q189.653 151.541 192.709 146.958 Q195.787 142.351 201.598 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M20.1444 1201.1 L26.9239 1201.1 Q23.9002 1204.35 22.4043 1208.04 Q20.9083 1211.7 20.9083 1215.84 Q20.9083 1223.99 25.9054 1228.31 Q30.8707 1232.64 40.2919 1232.64 Q49.6813 1232.64 54.6784 1228.31 Q59.6436 1223.99 59.6436 1215.84 Q59.6436 1211.7 58.1477 1208.04 Q56.6518 1204.35 53.6281 1201.1 L60.3439 1201.1 Q62.6355 1204.48 63.7814 1208.26 Q64.9272 1212.02 64.9272 1216.22 Q64.9272 1227.01 58.3387 1233.22 Q51.7183 1239.42 40.2919 1239.42 Q28.8336 1239.42 22.2451 1233.22 Q15.6248 1227.01 15.6248 1216.22 Q15.6248 1211.95 16.7706 1208.2 Q17.8846 1204.41 20.1444 1201.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M32.4621 1177.61 Q32.4621 1182.32 36.1542 1185.06 Q39.8145 1187.8 46.212 1187.8 Q52.6095 1187.8 56.3017 1185.09 Q59.9619 1182.35 59.9619 1177.61 Q59.9619 1172.93 56.2698 1170.2 Q52.5777 1167.46 46.212 1167.46 Q39.8781 1167.46 36.186 1170.2 Q32.4621 1172.93 32.4621 1177.61 M27.4968 1177.61 Q27.4968 1169.97 32.4621 1165.61 Q37.4273 1161.25 46.212 1161.25 Q54.9649 1161.25 59.9619 1165.61 Q64.9272 1169.97 64.9272 1177.61 Q64.9272 1185.28 59.9619 1189.64 Q54.9649 1193.97 46.212 1193.97 Q37.4273 1193.97 32.4621 1189.64 Q27.4968 1185.28 27.4968 1177.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M35.1993 1123.79 Q31.2526 1121.59 29.3747 1118.54 Q27.4968 1115.48 27.4968 1111.35 Q27.4968 1105.78 31.4117 1102.75 Q35.2948 1099.73 42.4881 1099.73 L64.0042 1099.73 L64.0042 1105.62 L42.679 1105.62 Q37.5546 1105.62 35.072 1107.43 Q32.5894 1109.24 32.5894 1112.97 Q32.5894 1117.52 35.6131 1120.16 Q38.6368 1122.8 43.8567 1122.8 L64.0042 1122.8 L64.0042 1128.69 L42.679 1128.69 Q37.5228 1128.69 35.072 1130.51 Q32.5894 1132.32 32.5894 1136.11 Q32.5894 1140.6 35.6449 1143.24 Q38.6686 1145.88 43.8567 1145.88 L64.0042 1145.88 L64.0042 1151.77 L28.3562 1151.77 L28.3562 1145.88 L33.8944 1145.88 Q30.616 1143.87 29.0564 1141.07 Q27.4968 1138.27 27.4968 1134.42 Q27.4968 1130.54 29.4702 1127.83 Q31.4436 1125.09 35.1993 1123.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M35.1993 1060.29 Q31.2526 1058.1 29.3747 1055.04 Q27.4968 1051.98 27.4968 1047.85 Q27.4968 1042.28 31.4117 1039.25 Q35.2948 1036.23 42.4881 1036.23 L64.0042 1036.23 L64.0042 1042.12 L42.679 1042.12 Q37.5546 1042.12 35.072 1043.93 Q32.5894 1045.75 32.5894 1049.47 Q32.5894 1054.02 35.6131 1056.66 Q38.6368 1059.31 43.8567 1059.31 L64.0042 1059.31 L64.0042 1065.19 L42.679 1065.19 Q37.5228 1065.19 35.072 1067.01 Q32.5894 1068.82 32.5894 1072.61 Q32.5894 1077.1 35.6449 1079.74 Q38.6686 1082.38 43.8567 1082.38 L64.0042 1082.38 L64.0042 1088.27 L28.3562 1088.27 L28.3562 1082.38 L33.8944 1082.38 Q30.616 1080.38 29.0564 1077.57 Q27.4968 1074.77 27.4968 1070.92 Q27.4968 1067.04 29.4702 1064.33 Q31.4436 1061.6 35.1993 1060.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M28.3562 1024.55 L28.3562 1018.69 L64.0042 1018.69 L64.0042 1024.55 L28.3562 1024.55 M14.479 1024.55 L14.479 1018.69 L21.895 1018.69 L21.895 1024.55 L14.479 1024.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M18.2347 1000.65 L28.3562 1000.65 L28.3562 988.582 L32.9077 988.582 L32.9077 1000.65 L52.2594 1000.65 Q56.6199 1000.65 57.8613 999.468 Q59.1026 998.258 59.1026 994.598 L59.1026 988.582 L64.0042 988.582 L64.0042 994.598 Q64.0042 1001.38 61.4897 1003.96 Q58.9434 1006.53 52.2594 1006.53 L32.9077 1006.53 L32.9077 1010.83 L28.3562 1010.83 L28.3562 1006.53 L18.2347 1006.53 L18.2347 1000.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M35.1993 953.125 Q31.2526 950.929 29.3747 947.874 Q27.4968 944.818 27.4968 940.68 Q27.4968 935.11 31.4117 932.087 Q35.2948 929.063 42.4881 929.063 L64.0042 929.063 L64.0042 934.951 L42.679 934.951 Q37.5546 934.951 35.072 936.766 Q32.5894 938.58 32.5894 942.304 Q32.5894 946.855 35.6131 949.497 Q38.6368 952.139 43.8567 952.139 L64.0042 952.139 L64.0042 958.027 L42.679 958.027 Q37.5228 958.027 35.072 959.841 Q32.5894 961.655 32.5894 965.443 Q32.5894 969.931 35.6449 972.573 Q38.6686 975.214 43.8567 975.214 L64.0042 975.214 L64.0042 981.103 L28.3562 981.103 L28.3562 975.214 L33.8944 975.214 Q30.616 973.209 29.0564 970.408 Q27.4968 967.607 27.4968 963.756 Q27.4968 959.873 29.4702 957.168 Q31.4436 954.43 35.1993 953.125 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M44.7161 886.89 L47.5806 886.89 L47.5806 913.817 Q53.6281 913.435 56.8109 910.189 Q59.9619 906.91 59.9619 901.086 Q59.9619 897.712 59.1344 894.561 Q58.3069 891.378 56.6518 888.259 L62.1899 888.259 Q63.5267 891.41 64.227 894.72 Q64.9272 898.03 64.9272 901.436 Q64.9272 909.966 59.9619 914.963 Q54.9967 919.928 46.5303 919.928 Q37.7774 919.928 32.6531 915.218 Q27.4968 910.475 27.4968 902.454 Q27.4968 895.261 32.1438 891.092 Q36.7589 886.89 44.7161 886.89 M42.9973 892.747 Q38.1912 892.81 35.3266 895.452 Q32.4621 898.062 32.4621 902.391 Q32.4621 907.292 35.2312 910.252 Q38.0002 913.181 43.0292 913.626 L42.9973 892.747 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M42.4881 847.646 L64.0042 847.646 L64.0042 853.502 L42.679 853.502 Q37.6183 853.502 35.1038 855.475 Q32.5894 857.449 32.5894 861.396 Q32.5894 866.138 35.6131 868.875 Q38.6368 871.612 43.8567 871.612 L64.0042 871.612 L64.0042 877.501 L28.3562 877.501 L28.3562 871.612 L33.8944 871.612 Q30.6797 869.512 29.0883 866.679 Q27.4968 863.815 27.4968 860.091 Q27.4968 853.948 31.3163 850.797 Q35.1038 847.646 42.4881 847.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M18.2347 830.172 L28.3562 830.172 L28.3562 818.109 L32.9077 818.109 L32.9077 830.172 L52.2594 830.172 Q56.6199 830.172 57.8613 828.994 Q59.1026 827.785 59.1026 824.124 L59.1026 818.109 L64.0042 818.109 L64.0042 824.124 Q64.0042 830.904 61.4897 833.482 Q58.9434 836.06 52.2594 836.06 L32.9077 836.06 L32.9077 840.357 L28.3562 840.357 L28.3562 836.06 L18.2347 836.06 L18.2347 830.172 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M33.7671 766.228 L14.479 766.228 L14.479 760.372 L64.0042 760.372 L64.0042 766.228 L58.657 766.228 Q61.8398 768.074 63.3994 770.907 Q64.9272 773.708 64.9272 777.655 Q64.9272 784.116 59.771 788.19 Q54.6147 792.232 46.212 792.232 Q37.8093 792.232 32.6531 788.19 Q27.4968 784.116 27.4968 777.655 Q27.4968 773.708 29.0564 770.907 Q30.5842 768.074 33.7671 766.228 M46.212 786.185 Q52.6732 786.185 56.3653 783.543 Q60.0256 780.869 60.0256 776.222 Q60.0256 771.575 56.3653 768.902 Q52.6732 766.228 46.212 766.228 Q39.7508 766.228 36.0905 768.902 Q32.3984 771.575 32.3984 776.222 Q32.3984 780.869 36.0905 783.543 Q39.7508 786.185 46.212 786.185 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M44.7161 717.817 L47.5806 717.817 L47.5806 744.744 Q53.6281 744.362 56.8109 741.116 Q59.9619 737.837 59.9619 732.013 Q59.9619 728.639 59.1344 725.488 Q58.3069 722.305 56.6518 719.186 L62.1899 719.186 Q63.5267 722.337 64.227 725.647 Q64.9272 728.957 64.9272 732.363 Q64.9272 740.893 59.9619 745.89 Q54.9967 750.855 46.5303 750.855 Q37.7774 750.855 32.6531 746.144 Q27.4968 741.402 27.4968 733.381 Q27.4968 726.188 32.1438 722.018 Q36.7589 717.817 44.7161 717.817 M42.9973 723.673 Q38.1912 723.737 35.3266 726.379 Q32.4621 728.989 32.4621 733.318 Q32.4621 738.219 35.2312 741.179 Q38.0002 744.107 43.0292 744.553 L42.9973 723.673 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M29.7248 682.551 L35.1993 682.551 Q33.8307 685.034 33.1623 687.548 Q32.4621 690.031 32.4621 692.577 Q32.4621 698.274 36.0905 701.425 Q39.6872 704.576 46.212 704.576 Q52.7369 704.576 56.3653 701.425 Q59.9619 698.274 59.9619 692.577 Q59.9619 690.031 59.2935 687.548 Q58.5933 685.034 57.2247 682.551 L62.6355 682.551 Q63.7814 685.002 64.3543 687.644 Q64.9272 690.254 64.9272 693.214 Q64.9272 701.266 59.8664 706.009 Q54.8057 710.751 46.212 710.751 Q37.491 710.751 32.4939 705.977 Q27.4968 701.171 27.4968 692.832 Q27.4968 690.126 28.0697 687.548 Q28.6108 684.97 29.7248 682.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M28.3562 672.366 L28.3562 666.509 L64.0042 666.509 L64.0042 672.366 L28.3562 672.366 M14.479 672.366 L14.479 666.509 L21.895 666.509 L21.895 672.366 L14.479 672.366 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M29.4065 631.53 L34.9447 631.53 Q33.6716 634.012 33.035 636.686 Q32.3984 639.36 32.3984 642.224 Q32.3984 646.585 33.7352 648.781 Q35.072 650.945 37.7456 650.945 Q39.7826 650.945 40.9603 649.386 Q42.1061 647.826 43.1565 643.115 L43.6021 641.11 Q44.9389 634.872 47.3897 632.262 Q49.8086 629.62 54.1691 629.62 Q59.1344 629.62 62.0308 633.567 Q64.9272 637.482 64.9272 644.357 Q64.9272 647.221 64.3543 650.341 Q63.8132 653.428 62.6992 656.865 L56.6518 656.865 Q58.3387 653.619 59.198 650.468 Q60.0256 647.317 60.0256 644.229 Q60.0256 640.092 58.6251 637.864 Q57.1929 635.636 54.6147 635.636 Q52.2276 635.636 50.9545 637.259 Q49.6813 638.85 48.5037 644.293 L48.0262 646.33 Q46.8804 651.773 44.5251 654.192 Q42.138 656.611 38.0002 656.611 Q32.9713 656.611 30.2341 653.046 Q27.4968 649.481 27.4968 642.924 Q27.4968 639.678 27.9743 636.813 Q28.4517 633.949 29.4065 631.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M28.3562 620.294 L28.3562 614.438 L64.0042 614.438 L64.0042 620.294 L28.3562 620.294 M14.479 620.294 L14.479 614.438 L21.895 614.438 L21.895 620.294 L14.479 620.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M32.4621 588.37 Q32.4621 593.081 36.1542 595.818 Q39.8145 598.556 46.212 598.556 Q52.6095 598.556 56.3017 595.85 Q59.9619 593.113 59.9619 588.37 Q59.9619 583.692 56.2698 580.954 Q52.5777 578.217 46.212 578.217 Q39.8781 578.217 36.186 580.954 Q32.4621 583.692 32.4621 588.37 M27.4968 588.37 Q27.4968 580.732 32.4621 576.371 Q37.4273 572.011 46.212 572.011 Q54.9649 572.011 59.9619 576.371 Q64.9272 580.732 64.9272 588.37 Q64.9272 596.041 59.9619 600.402 Q54.9649 604.73 46.212 604.73 Q37.4273 604.73 32.4621 600.402 Q27.4968 596.041 27.4968 588.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M42.4881 532.67 L64.0042 532.67 L64.0042 538.527 L42.679 538.527 Q37.6183 538.527 35.1038 540.5 Q32.5894 542.474 32.5894 546.42 Q32.5894 551.163 35.6131 553.9 Q38.6368 556.637 43.8567 556.637 L64.0042 556.637 L64.0042 562.526 L28.3562 562.526 L28.3562 556.637 L33.8944 556.637 Q30.6797 554.537 29.0883 551.704 Q27.4968 548.839 27.4968 545.115 Q27.4968 538.972 31.3163 535.821 Q35.1038 532.67 42.4881 532.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M70.0516 473.087 L74.6349 473.087 L74.6349 475.061 Q74.6349 482.986 72.2796 485.692 Q69.9243 488.365 62.8902 488.365 L55.2831 488.365 Q50.477 488.365 48.631 490.084 Q46.7849 491.803 46.7849 496.322 L46.7849 498.264 L42.2334 498.264 L42.2334 496.322 Q42.2334 491.771 40.4192 490.084 Q38.5732 488.365 33.8307 488.365 L26.1919 488.365 Q19.1578 488.365 16.8343 485.692 Q14.479 482.986 14.479 475.061 L14.479 473.087 L19.0304 473.087 L19.0304 475.252 Q19.0304 479.74 20.4309 481.108 Q21.8314 482.477 26.3192 482.477 L34.2127 482.477 Q39.2097 482.477 41.4696 483.941 Q43.7294 485.373 44.5251 488.874 Q45.3845 485.341 47.6443 483.909 Q49.9041 482.477 54.8694 482.477 L62.7628 482.477 Q67.2507 482.477 68.6511 481.108 Q70.0516 479.74 70.0516 475.252 L70.0516 473.087 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M20.7174 444.219 Q20.7174 449.184 25.6189 451.699 Q30.4887 454.181 40.2919 454.181 Q50.0633 454.181 54.9649 451.699 Q59.8346 449.184 59.8346 444.219 Q59.8346 439.222 54.9649 436.739 Q50.0633 434.225 40.2919 434.225 Q30.4887 434.225 25.6189 436.739 Q20.7174 439.222 20.7174 444.219 M15.6248 444.219 Q15.6248 436.23 21.9587 432.029 Q28.2607 427.795 40.2919 427.795 Q52.2913 427.795 58.6251 432.029 Q64.9272 436.23 64.9272 444.219 Q64.9272 452.208 58.6251 456.441 Q52.2913 460.642 40.2919 460.642 Q28.2607 460.642 21.9587 456.441 Q15.6248 452.208 15.6248 444.219 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M55.9197 415.828 L55.9197 409.112 L61.3942 409.112 L71.5793 414.332 L71.5793 418.438 L61.3942 415.828 L55.9197 415.828 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M58.5933 373.942 L58.5933 363.438 L22.3406 363.438 L24.6323 374.865 L18.7758 374.865 L16.4842 363.502 L16.4842 357.072 L58.5933 357.072 L58.5933 346.569 L64.0042 346.569 L64.0042 373.942 L58.5933 373.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M70.0516 332.405 L70.0516 330.177 Q70.0516 325.721 68.683 324.385 Q67.3143 323.016 62.7628 323.016 L54.8694 323.016 Q49.9041 323.016 47.6443 321.584 Q45.3845 320.151 44.5251 316.618 Q43.7294 320.151 41.4696 321.584 Q39.2097 323.016 34.2127 323.016 L26.3192 323.016 Q21.7995 323.016 20.4309 324.385 Q19.0304 325.721 19.0304 330.177 L19.0304 332.405 L14.479 332.405 L14.479 330.4 Q14.479 322.475 16.8343 319.833 Q19.1578 317.159 26.1919 317.159 L33.8307 317.159 Q38.5732 317.159 40.4192 315.441 Q42.2334 313.722 42.2334 309.202 L42.2334 307.229 L46.7849 307.229 L46.7849 309.202 Q46.7849 313.722 48.631 315.441 Q50.477 317.159 55.2831 317.159 L62.8902 317.159 Q69.9243 317.159 72.2796 319.833 Q74.6349 322.475 74.6349 330.4 L74.6349 332.405 L70.0516 332.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M478.56 16.7545 L478.56 25.383 Q474.429 21.5346 469.729 19.6307 Q465.071 17.7268 459.805 17.7268 Q449.434 17.7268 443.925 24.0867 Q438.416 30.4061 438.416 42.3968 Q438.416 54.3469 443.925 60.7069 Q449.434 67.0263 459.805 67.0263 Q465.071 67.0263 469.729 65.1223 Q474.429 63.2184 478.56 59.3701 L478.56 67.9175 Q474.267 70.8341 469.446 72.2924 Q464.666 73.7508 459.319 73.7508 Q445.586 73.7508 437.687 65.3654 Q429.788 56.9395 429.788 42.3968 Q429.788 27.8135 437.687 19.4281 Q445.586 11.0023 459.319 11.0023 Q464.747 11.0023 469.527 12.4606 Q474.348 13.8784 478.56 16.7545 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M508.456 32.4315 Q502.461 32.4315 498.977 37.1306 Q495.493 41.7891 495.493 49.9314 Q495.493 58.0738 498.937 62.7728 Q502.42 67.4314 508.456 67.4314 Q514.411 67.4314 517.895 62.7323 Q521.379 58.0333 521.379 49.9314 Q521.379 41.8701 517.895 37.1711 Q514.411 32.4315 508.456 32.4315 M508.456 26.1121 Q518.178 26.1121 523.728 32.4315 Q529.278 38.7509 529.278 49.9314 Q529.278 61.0714 523.728 67.4314 Q518.178 73.7508 508.456 73.7508 Q498.693 73.7508 493.144 67.4314 Q487.634 61.0714 487.634 49.9314 Q487.634 38.7509 493.144 32.4315 Q498.693 26.1121 508.456 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M576.957 35.9153 Q579.752 30.8922 583.641 28.5022 Q587.53 26.1121 592.796 26.1121 Q599.885 26.1121 603.733 31.0947 Q607.582 36.0368 607.582 45.1919 L607.582 72.576 L600.088 72.576 L600.088 45.4349 Q600.088 38.913 597.779 35.7533 Q595.47 32.5936 590.73 32.5936 Q584.937 32.5936 581.575 36.4419 Q578.213 40.2903 578.213 46.9338 L578.213 72.576 L570.719 72.576 L570.719 45.4349 Q570.719 38.8725 568.41 35.7533 Q566.101 32.5936 561.28 32.5936 Q555.568 32.5936 552.206 36.4824 Q548.844 40.3308 548.844 46.9338 L548.844 72.576 L541.349 72.576 L541.349 27.2059 L548.844 27.2059 L548.844 34.2544 Q551.396 30.082 554.961 28.0971 Q558.525 26.1121 563.427 26.1121 Q568.369 26.1121 571.812 28.6237 Q575.296 31.1352 576.957 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M657.772 35.9153 Q660.568 30.8922 664.456 28.5022 Q668.345 26.1121 673.612 26.1121 Q680.701 26.1121 684.549 31.0947 Q688.397 36.0368 688.397 45.1919 L688.397 72.576 L680.903 72.576 L680.903 45.4349 Q680.903 38.913 678.594 35.7533 Q676.285 32.5936 671.546 32.5936 Q665.753 32.5936 662.39 36.4419 Q659.028 40.2903 659.028 46.9338 L659.028 72.576 L651.534 72.576 L651.534 45.4349 Q651.534 38.8725 649.225 35.7533 Q646.916 32.5936 642.095 32.5936 Q636.384 32.5936 633.021 36.4824 Q629.659 40.3308 629.659 46.9338 L629.659 72.576 L622.165 72.576 L622.165 27.2059 L629.659 27.2059 L629.659 34.2544 Q632.211 30.082 635.776 28.0971 Q639.341 26.1121 644.242 26.1121 Q649.185 26.1121 652.628 28.6237 Q656.112 31.1352 657.772 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M703.264 27.2059 L710.718 27.2059 L710.718 72.576 L703.264 72.576 L703.264 27.2059 M703.264 9.54393 L710.718 9.54393 L710.718 18.9825 L703.264 18.9825 L703.264 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M733.686 14.324 L733.686 27.2059 L749.039 27.2059 L749.039 32.9987 L733.686 32.9987 L733.686 57.6282 Q733.686 63.1779 735.185 64.7578 Q736.725 66.3376 741.383 66.3376 L749.039 66.3376 L749.039 72.576 L741.383 72.576 Q732.755 72.576 729.473 69.3758 Q726.192 66.1351 726.192 57.6282 L726.192 32.9987 L720.724 32.9987 L720.724 27.2059 L726.192 27.2059 L726.192 14.324 L733.686 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M794.166 35.9153 Q796.962 30.8922 800.85 28.5022 Q804.739 26.1121 810.005 26.1121 Q817.095 26.1121 820.943 31.0947 Q824.791 36.0368 824.791 45.1919 L824.791 72.576 L817.297 72.576 L817.297 45.4349 Q817.297 38.913 814.988 35.7533 Q812.679 32.5936 807.939 32.5936 Q802.147 32.5936 798.784 36.4419 Q795.422 40.2903 795.422 46.9338 L795.422 72.576 L787.928 72.576 L787.928 45.4349 Q787.928 38.8725 785.619 35.7533 Q783.31 32.5936 778.489 32.5936 Q772.778 32.5936 769.415 36.4824 Q766.053 40.3308 766.053 46.9338 L766.053 72.576 L758.559 72.576 L758.559 27.2059 L766.053 27.2059 L766.053 34.2544 Q768.605 30.082 772.17 28.0971 Q775.735 26.1121 780.636 26.1121 Q785.578 26.1121 789.022 28.6237 Q792.506 31.1352 794.166 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M878.466 48.0275 L878.466 51.6733 L844.195 51.6733 Q844.681 59.3701 848.813 63.421 Q852.986 67.4314 860.399 67.4314 Q864.693 67.4314 868.703 66.3781 Q872.754 65.3249 876.724 63.2184 L876.724 70.267 Q872.713 71.9684 868.501 72.8596 Q864.288 73.7508 859.953 73.7508 Q849.097 73.7508 842.737 67.4314 Q836.417 61.1119 836.417 50.3365 Q836.417 39.1965 842.413 32.6746 Q848.449 26.1121 858.657 26.1121 Q867.812 26.1121 873.119 32.0264 Q878.466 37.9003 878.466 48.0275 M871.012 45.84 Q870.931 39.7232 867.569 36.0774 Q864.247 32.4315 858.738 32.4315 Q852.499 32.4315 848.732 35.9558 Q845.005 39.4801 844.438 45.8805 L871.012 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M928.413 45.1919 L928.413 72.576 L920.96 72.576 L920.96 45.4349 Q920.96 38.994 918.448 35.7938 Q915.937 32.5936 910.913 32.5936 Q904.878 32.5936 901.394 36.4419 Q897.91 40.2903 897.91 46.9338 L897.91 72.576 L890.416 72.576 L890.416 27.2059 L897.91 27.2059 L897.91 34.2544 Q900.584 30.163 904.189 28.1376 Q907.835 26.1121 912.574 26.1121 Q920.393 26.1121 924.403 30.9732 Q928.413 35.7938 928.413 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M950.653 14.324 L950.653 27.2059 L966.006 27.2059 L966.006 32.9987 L950.653 32.9987 L950.653 57.6282 Q950.653 63.1779 952.152 64.7578 Q953.691 66.3376 958.35 66.3376 L966.006 66.3376 L966.006 72.576 L958.35 72.576 Q949.721 72.576 946.44 69.3758 Q943.159 66.1351 943.159 57.6282 L943.159 32.9987 L937.69 32.9987 L937.69 27.2059 L943.159 27.2059 L943.159 14.324 L950.653 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip712)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="274.921,160.256 345.417,160.256 415.913,160.256 486.409,160.256 556.905,160.256 627.401,160.256 697.897,160.256 768.393,160.256 838.889,160.256 909.385,160.256 979.881,160.256 1050.38,160.256 1120.87,160.256 "/>
-<polyline clip-path="url(#clip712)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="274.921,1386.4 345.417,1386.4 415.913,1386.4 486.409,1386.4 556.905,1386.4 627.401,1386.4 697.897,1386.4 768.393,160.256 838.889,160.256 909.385,160.256 979.881,160.256 1050.38,160.256 1120.87,160.256 "/>
-<path clip-path="url(#clip710)" d="M966.829 1379.86 L1116.36 1379.86 L1116.36 1224.34 L966.829 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="966.829,1379.86 1116.36,1379.86 1116.36,1224.34 966.829,1224.34 966.829,1379.86 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="976.793,1276.18 1036.57,1276.18 "/>
-<path clip-path="url(#clip710)" d="M1072.09 1288.53 L1072.09 1279.24 L1064.45 1279.24 L1064.45 1275.4 L1076.72 1275.4 L1076.72 1290.24 Q1074.01 1292.16 1070.75 1293.16 Q1067.49 1294.13 1063.78 1294.13 Q1055.68 1294.13 1051.1 1289.41 Q1046.54 1284.66 1046.54 1276.21 Q1046.54 1267.74 1051.1 1263.02 Q1055.68 1258.27 1063.78 1258.27 Q1067.16 1258.27 1070.19 1259.1 Q1073.25 1259.94 1075.82 1261.56 L1075.82 1266.54 Q1073.23 1264.34 1070.31 1263.22 Q1067.39 1262.11 1064.18 1262.11 Q1057.83 1262.11 1054.64 1265.66 Q1051.47 1269.2 1051.47 1276.21 Q1051.47 1283.2 1054.64 1286.74 Q1057.83 1290.28 1064.18 1290.28 Q1066.65 1290.28 1068.6 1289.87 Q1070.54 1289.43 1072.09 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1086.49 1289.52 L1094.13 1289.52 L1094.13 1263.16 L1085.82 1264.82 L1085.82 1260.56 L1094.08 1258.9 L1098.76 1258.9 L1098.76 1289.52 L1106.4 1289.52 L1106.4 1293.46 L1086.49 1293.46 L1086.49 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip710)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="976.793,1328.02 1036.57,1328.02 "/>
-<path clip-path="url(#clip710)" d="M1072.09 1340.37 L1072.09 1331.08 L1064.45 1331.08 L1064.45 1327.24 L1076.72 1327.24 L1076.72 1342.08 Q1074.01 1344 1070.75 1345 Q1067.49 1345.97 1063.78 1345.97 Q1055.68 1345.97 1051.1 1341.25 Q1046.54 1336.5 1046.54 1328.05 Q1046.54 1319.58 1051.1 1314.86 Q1055.68 1310.11 1063.78 1310.11 Q1067.16 1310.11 1070.19 1310.94 Q1073.25 1311.78 1075.82 1313.4 L1075.82 1318.38 Q1073.23 1316.18 1070.31 1315.06 Q1067.39 1313.95 1064.18 1313.95 Q1057.83 1313.95 1054.64 1317.5 Q1051.47 1321.04 1051.47 1328.05 Q1051.47 1335.04 1054.64 1338.58 Q1057.83 1342.12 1064.18 1342.12 Q1066.65 1342.12 1068.6 1341.71 Q1070.54 1341.27 1072.09 1340.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1089.71 1341.36 L1106.03 1341.36 L1106.03 1345.3 L1084.08 1345.3 L1084.08 1341.36 Q1086.74 1338.61 1091.33 1333.98 Q1095.93 1329.32 1097.12 1327.98 Q1099.36 1325.46 1100.24 1323.72 Q1101.14 1321.96 1101.14 1320.27 Q1101.14 1317.52 1099.2 1315.78 Q1097.28 1314.05 1094.18 1314.05 Q1091.98 1314.05 1089.52 1314.81 Q1087.09 1315.57 1084.31 1317.13 L1084.31 1312.4 Q1087.14 1311.27 1089.59 1310.69 Q1092.05 1310.11 1094.08 1310.11 Q1099.45 1310.11 1102.65 1312.8 Q1105.84 1315.48 1105.84 1319.97 Q1105.84 1322.1 1105.03 1324.02 Q1104.24 1325.92 1102.14 1328.51 Q1101.56 1329.19 1098.46 1332.4 Q1095.36 1335.6 1089.71 1341.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1456.05 1423.18 L2352.76 1423.18 L2352.76 123.472 L1456.05 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="368.915,1423.18 368.915,123.472 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="603.902,1423.18 603.902,123.472 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.889,1423.18 838.889,123.472 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1073.88,1423.18 1073.88,123.472 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1423.18 1146.25,1423.18 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="368.915,1423.18 368.915,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="603.902,1423.18 603.902,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.889,1423.18 838.889,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1073.88,1423.18 1073.88,1404.28 "/>
+<path clip-path="url(#clip530)" d="M328.534 1451.02 L346.89 1451.02 L346.89 1454.96 L332.816 1454.96 L332.816 1463.43 Q333.834 1463.08 334.853 1462.92 Q335.871 1462.73 336.89 1462.73 Q342.677 1462.73 346.057 1465.9 Q349.436 1469.08 349.436 1474.49 Q349.436 1480.07 345.964 1483.17 Q342.492 1486.25 336.172 1486.25 Q333.996 1486.25 331.728 1485.88 Q329.483 1485.51 327.075 1484.77 L327.075 1480.07 Q329.159 1481.2 331.381 1481.76 Q333.603 1482.32 336.08 1482.32 Q340.084 1482.32 342.422 1480.21 Q344.76 1478.1 344.76 1474.49 Q344.76 1470.88 342.422 1468.77 Q340.084 1466.67 336.08 1466.67 Q334.205 1466.67 332.33 1467.08 Q330.478 1467.5 328.534 1468.38 L328.534 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M368.649 1454.1 Q365.038 1454.1 363.209 1457.66 Q361.404 1461.2 361.404 1468.33 Q361.404 1475.44 363.209 1479.01 Q365.038 1482.55 368.649 1482.55 Q372.283 1482.55 374.089 1479.01 Q375.918 1475.44 375.918 1468.33 Q375.918 1461.2 374.089 1457.66 Q372.283 1454.1 368.649 1454.1 M368.649 1450.39 Q374.459 1450.39 377.515 1455 Q380.593 1459.58 380.593 1468.33 Q380.593 1477.06 377.515 1481.67 Q374.459 1486.25 368.649 1486.25 Q362.839 1486.25 359.76 1481.67 Q356.705 1477.06 356.705 1468.33 Q356.705 1459.58 359.76 1455 Q362.839 1450.39 368.649 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M398.811 1454.1 Q395.2 1454.1 393.371 1457.66 Q391.566 1461.2 391.566 1468.33 Q391.566 1475.44 393.371 1479.01 Q395.2 1482.55 398.811 1482.55 Q402.445 1482.55 404.251 1479.01 Q406.079 1475.44 406.079 1468.33 Q406.079 1461.2 404.251 1457.66 Q402.445 1454.1 398.811 1454.1 M398.811 1450.39 Q404.621 1450.39 407.677 1455 Q410.755 1459.58 410.755 1468.33 Q410.755 1477.06 407.677 1481.67 Q404.621 1486.25 398.811 1486.25 Q393.001 1486.25 389.922 1481.67 Q386.867 1477.06 386.867 1468.33 Q386.867 1459.58 389.922 1455 Q393.001 1450.39 398.811 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M548.428 1481.64 L556.067 1481.64 L556.067 1455.28 L547.756 1456.95 L547.756 1452.69 L556.02 1451.02 L560.696 1451.02 L560.696 1481.64 L568.335 1481.64 L568.335 1485.58 L548.428 1485.58 L548.428 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M587.779 1454.1 Q584.168 1454.1 582.34 1457.66 Q580.534 1461.2 580.534 1468.33 Q580.534 1475.44 582.34 1479.01 Q584.168 1482.55 587.779 1482.55 Q591.414 1482.55 593.219 1479.01 Q595.048 1475.44 595.048 1468.33 Q595.048 1461.2 593.219 1457.66 Q591.414 1454.1 587.779 1454.1 M587.779 1450.39 Q593.59 1450.39 596.645 1455 Q599.724 1459.58 599.724 1468.33 Q599.724 1477.06 596.645 1481.67 Q593.59 1486.25 587.779 1486.25 Q581.969 1486.25 578.891 1481.67 Q575.835 1477.06 575.835 1468.33 Q575.835 1459.58 578.891 1455 Q581.969 1450.39 587.779 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M617.941 1454.1 Q614.33 1454.1 612.501 1457.66 Q610.696 1461.2 610.696 1468.33 Q610.696 1475.44 612.501 1479.01 Q614.33 1482.55 617.941 1482.55 Q621.576 1482.55 623.381 1479.01 Q625.21 1475.44 625.21 1468.33 Q625.21 1461.2 623.381 1457.66 Q621.576 1454.1 617.941 1454.1 M617.941 1450.39 Q623.751 1450.39 626.807 1455 Q629.886 1459.58 629.886 1468.33 Q629.886 1477.06 626.807 1481.67 Q623.751 1486.25 617.941 1486.25 Q612.131 1486.25 609.052 1481.67 Q605.997 1477.06 605.997 1468.33 Q605.997 1459.58 609.052 1455 Q612.131 1450.39 617.941 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M648.103 1454.1 Q644.492 1454.1 642.663 1457.66 Q640.858 1461.2 640.858 1468.33 Q640.858 1475.44 642.663 1479.01 Q644.492 1482.55 648.103 1482.55 Q651.737 1482.55 653.543 1479.01 Q655.372 1475.44 655.372 1468.33 Q655.372 1461.2 653.543 1457.66 Q651.737 1454.1 648.103 1454.1 M648.103 1450.39 Q653.913 1450.39 656.969 1455 Q660.048 1459.58 660.048 1468.33 Q660.048 1477.06 656.969 1481.67 Q653.913 1486.25 648.103 1486.25 Q642.293 1486.25 639.214 1481.67 Q636.159 1477.06 636.159 1468.33 Q636.159 1459.58 639.214 1455 Q642.293 1450.39 648.103 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M783.414 1481.64 L791.053 1481.64 L791.053 1455.28 L782.743 1456.95 L782.743 1452.69 L791.007 1451.02 L795.683 1451.02 L795.683 1481.64 L803.322 1481.64 L803.322 1485.58 L783.414 1485.58 L783.414 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M812.812 1451.02 L831.169 1451.02 L831.169 1454.96 L817.095 1454.96 L817.095 1463.43 Q818.113 1463.08 819.132 1462.92 Q820.15 1462.73 821.169 1462.73 Q826.956 1462.73 830.336 1465.9 Q833.715 1469.08 833.715 1474.49 Q833.715 1480.07 830.243 1483.17 Q826.771 1486.25 820.451 1486.25 Q818.275 1486.25 816.007 1485.88 Q813.762 1485.51 811.354 1484.77 L811.354 1480.07 Q813.437 1481.2 815.66 1481.76 Q817.882 1482.32 820.359 1482.32 Q824.363 1482.32 826.701 1480.21 Q829.039 1478.1 829.039 1474.49 Q829.039 1470.88 826.701 1468.77 Q824.363 1466.67 820.359 1466.67 Q818.484 1466.67 816.609 1467.08 Q814.757 1467.5 812.812 1468.38 L812.812 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M852.928 1454.1 Q849.317 1454.1 847.488 1457.66 Q845.683 1461.2 845.683 1468.33 Q845.683 1475.44 847.488 1479.01 Q849.317 1482.55 852.928 1482.55 Q856.562 1482.55 858.368 1479.01 Q860.196 1475.44 860.196 1468.33 Q860.196 1461.2 858.368 1457.66 Q856.562 1454.1 852.928 1454.1 M852.928 1450.39 Q858.738 1450.39 861.794 1455 Q864.872 1459.58 864.872 1468.33 Q864.872 1477.06 861.794 1481.67 Q858.738 1486.25 852.928 1486.25 Q847.118 1486.25 844.039 1481.67 Q840.984 1477.06 840.984 1468.33 Q840.984 1459.58 844.039 1455 Q847.118 1450.39 852.928 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M883.09 1454.1 Q879.479 1454.1 877.65 1457.66 Q875.845 1461.2 875.845 1468.33 Q875.845 1475.44 877.65 1479.01 Q879.479 1482.55 883.09 1482.55 Q886.724 1482.55 888.53 1479.01 Q890.358 1475.44 890.358 1468.33 Q890.358 1461.2 888.53 1457.66 Q886.724 1454.1 883.09 1454.1 M883.09 1450.39 Q888.9 1450.39 891.956 1455 Q895.034 1459.58 895.034 1468.33 Q895.034 1477.06 891.956 1481.67 Q888.9 1486.25 883.09 1486.25 Q877.28 1486.25 874.201 1481.67 Q871.145 1477.06 871.145 1468.33 Q871.145 1459.58 874.201 1455 Q877.28 1450.39 883.09 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1022.49 1481.64 L1038.81 1481.64 L1038.81 1485.58 L1016.86 1485.58 L1016.86 1481.64 Q1019.52 1478.89 1024.11 1474.26 Q1028.71 1469.61 1029.89 1468.27 Q1032.14 1465.74 1033.02 1464.01 Q1033.92 1462.25 1033.92 1460.56 Q1033.92 1457.8 1031.98 1456.07 Q1030.06 1454.33 1026.95 1454.33 Q1024.76 1454.33 1022.3 1455.09 Q1019.87 1455.86 1017.09 1457.41 L1017.09 1452.69 Q1019.92 1451.55 1022.37 1450.97 Q1024.82 1450.39 1026.86 1450.39 Q1032.23 1450.39 1035.43 1453.08 Q1038.62 1455.77 1038.62 1460.26 Q1038.62 1462.39 1037.81 1464.31 Q1037.02 1466.2 1034.92 1468.8 Q1034.34 1469.47 1031.24 1472.69 Q1028.13 1475.88 1022.49 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1058.62 1454.1 Q1055.01 1454.1 1053.18 1457.66 Q1051.38 1461.2 1051.38 1468.33 Q1051.38 1475.44 1053.18 1479.01 Q1055.01 1482.55 1058.62 1482.55 Q1062.26 1482.55 1064.06 1479.01 Q1065.89 1475.44 1065.89 1468.33 Q1065.89 1461.2 1064.06 1457.66 Q1062.26 1454.1 1058.62 1454.1 M1058.62 1450.39 Q1064.43 1450.39 1067.49 1455 Q1070.57 1459.58 1070.57 1468.33 Q1070.57 1477.06 1067.49 1481.67 Q1064.43 1486.25 1058.62 1486.25 Q1052.81 1486.25 1049.73 1481.67 Q1046.68 1477.06 1046.68 1468.33 Q1046.68 1459.58 1049.73 1455 Q1052.81 1450.39 1058.62 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1088.78 1454.1 Q1085.17 1454.1 1083.34 1457.66 Q1081.54 1461.2 1081.54 1468.33 Q1081.54 1475.44 1083.34 1479.01 Q1085.17 1482.55 1088.78 1482.55 Q1092.42 1482.55 1094.22 1479.01 Q1096.05 1475.44 1096.05 1468.33 Q1096.05 1461.2 1094.22 1457.66 Q1092.42 1454.1 1088.78 1454.1 M1088.78 1450.39 Q1094.59 1450.39 1097.65 1455 Q1100.73 1459.58 1100.73 1468.33 Q1100.73 1477.06 1097.65 1481.67 Q1094.59 1486.25 1088.78 1486.25 Q1082.97 1486.25 1079.89 1481.67 Q1076.84 1477.06 1076.84 1468.33 Q1076.84 1459.58 1079.89 1455 Q1082.97 1450.39 1088.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1118.94 1454.1 Q1115.33 1454.1 1113.5 1457.66 Q1111.7 1461.2 1111.7 1468.33 Q1111.7 1475.44 1113.5 1479.01 Q1115.33 1482.55 1118.94 1482.55 Q1122.58 1482.55 1124.38 1479.01 Q1126.21 1475.44 1126.21 1468.33 Q1126.21 1461.2 1124.38 1457.66 Q1122.58 1454.1 1118.94 1454.1 M1118.94 1450.39 Q1124.75 1450.39 1127.81 1455 Q1130.89 1459.58 1130.89 1468.33 Q1130.89 1477.06 1127.81 1481.67 Q1124.75 1486.25 1118.94 1486.25 Q1113.13 1486.25 1110.06 1481.67 Q1107 1477.06 1107 1468.33 Q1107 1459.58 1110.06 1455 Q1113.13 1450.39 1118.94 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M475.972 1525.81 L475.972 1562.76 L483.738 1562.76 Q493.573 1562.76 498.125 1558.3 Q502.708 1553.85 502.708 1544.24 Q502.708 1534.69 498.125 1530.26 Q493.573 1525.81 483.738 1525.81 L475.972 1525.81 M469.543 1520.52 L482.752 1520.52 Q496.565 1520.52 503.026 1526.28 Q509.488 1532.01 509.488 1544.24 Q509.488 1556.52 502.995 1562.28 Q496.502 1568.04 482.752 1568.04 L469.543 1568.04 L469.543 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M549.974 1548.76 L549.974 1551.62 L523.047 1551.62 Q523.429 1557.67 526.675 1560.85 Q529.953 1564 535.778 1564 Q539.152 1564 542.303 1563.17 Q545.486 1562.35 548.605 1560.69 L548.605 1566.23 Q545.454 1567.57 542.144 1568.27 Q538.834 1568.97 535.428 1568.97 Q526.898 1568.97 521.901 1564 Q516.936 1559.04 516.936 1550.57 Q516.936 1541.82 521.646 1536.69 Q526.389 1531.54 534.409 1531.54 Q541.603 1531.54 545.772 1536.18 Q549.974 1540.8 549.974 1548.76 M544.117 1547.04 Q544.053 1542.23 541.412 1539.37 Q538.802 1536.5 534.473 1536.5 Q529.571 1536.5 526.611 1539.27 Q523.683 1542.04 523.238 1547.07 L544.117 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M587.34 1539.24 Q589.536 1535.29 592.592 1533.41 Q595.648 1531.54 599.785 1531.54 Q605.355 1531.54 608.379 1535.45 Q611.403 1539.33 611.403 1546.53 L611.403 1568.04 L605.514 1568.04 L605.514 1546.72 Q605.514 1541.59 603.7 1539.11 Q601.886 1536.63 598.162 1536.63 Q593.61 1536.63 590.969 1539.65 Q588.327 1542.68 588.327 1547.9 L588.327 1568.04 L582.439 1568.04 L582.439 1546.72 Q582.439 1541.56 580.624 1539.11 Q578.81 1536.63 575.023 1536.63 Q570.535 1536.63 567.893 1539.68 Q565.251 1542.71 565.251 1547.9 L565.251 1568.04 L559.363 1568.04 L559.363 1532.4 L565.251 1532.4 L565.251 1537.93 Q567.256 1534.66 570.057 1533.1 Q572.858 1531.54 576.71 1531.54 Q580.593 1531.54 583.298 1533.51 Q586.035 1535.48 587.34 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M639.284 1550.12 Q632.187 1550.12 629.449 1551.75 Q626.712 1553.37 626.712 1557.29 Q626.712 1560.4 628.749 1562.25 Q630.818 1564.07 634.351 1564.07 Q639.221 1564.07 642.149 1560.63 Q645.109 1557.16 645.109 1551.43 L645.109 1550.12 L639.284 1550.12 M650.965 1547.71 L650.965 1568.04 L645.109 1568.04 L645.109 1562.63 Q643.104 1565.88 640.112 1567.44 Q637.12 1568.97 632.791 1568.97 Q627.317 1568.97 624.07 1565.91 Q620.856 1562.82 620.856 1557.67 Q620.856 1551.65 624.866 1548.6 Q628.908 1545.54 636.897 1545.54 L645.109 1545.54 L645.109 1544.97 Q645.109 1540.93 642.435 1538.73 Q639.794 1536.5 634.988 1536.5 Q631.932 1536.5 629.036 1537.23 Q626.139 1537.97 623.466 1539.43 L623.466 1534.02 Q626.68 1532.78 629.704 1532.17 Q632.728 1531.54 635.592 1531.54 Q643.327 1531.54 647.146 1535.55 Q650.965 1539.56 650.965 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M692.661 1546.53 L692.661 1568.04 L686.804 1568.04 L686.804 1546.72 Q686.804 1541.66 684.831 1539.14 Q682.858 1536.63 678.911 1536.63 Q674.169 1536.63 671.431 1539.65 Q668.694 1542.68 668.694 1547.9 L668.694 1568.04 L662.806 1568.04 L662.806 1532.4 L668.694 1532.4 L668.694 1537.93 Q670.795 1534.72 673.627 1533.13 Q676.492 1531.54 680.216 1531.54 Q686.359 1531.54 689.51 1535.36 Q692.661 1539.14 692.661 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M727.8 1537.81 L727.8 1518.52 L733.656 1518.52 L733.656 1568.04 L727.8 1568.04 L727.8 1562.7 Q725.954 1565.88 723.121 1567.44 Q720.32 1568.97 716.373 1568.97 Q709.912 1568.97 705.838 1563.81 Q701.796 1558.65 701.796 1550.25 Q701.796 1541.85 705.838 1536.69 Q709.912 1531.54 716.373 1531.54 Q720.32 1531.54 723.121 1533.1 Q725.954 1534.62 727.8 1537.81 M707.843 1550.25 Q707.843 1556.71 710.485 1560.4 Q713.158 1564.07 717.805 1564.07 Q722.452 1564.07 725.126 1560.4 Q727.8 1556.71 727.8 1550.25 Q727.8 1543.79 725.126 1540.13 Q722.452 1536.44 717.805 1536.44 Q713.158 1536.44 710.485 1540.13 Q707.843 1543.79 707.843 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M765.898 1518.52 L779.394 1518.52 L779.394 1523.07 L771.755 1523.07 L771.755 1572.09 L779.394 1572.09 L779.394 1576.64 L765.898 1576.64 L765.898 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M792.125 1520.52 L801.705 1520.52 L813.832 1552.86 L826.022 1520.52 L835.603 1520.52 L835.603 1568.04 L829.333 1568.04 L829.333 1526.32 L817.079 1558.91 L810.617 1558.91 L798.363 1526.32 L798.363 1568.04 L792.125 1568.04 L792.125 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M844.133 1520.52 L850.626 1520.52 L860.62 1560.69 L870.582 1520.52 L877.807 1520.52 L887.802 1560.69 L897.764 1520.52 L904.289 1520.52 L892.353 1568.04 L884.269 1568.04 L874.243 1526.79 L864.121 1568.04 L856.037 1568.04 L844.133 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M926.25 1518.52 L926.25 1576.64 L912.755 1576.64 L912.755 1572.09 L920.362 1572.09 L920.362 1523.07 L912.755 1523.07 L912.755 1518.52 L926.25 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,1386.4 1146.25,1386.4 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,1079.86 1146.25,1079.86 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,773.326 1146.25,773.326 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,466.791 1146.25,466.791 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="249.542,160.256 1146.25,160.256 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1423.18 249.542,123.472 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1386.4 268.44,1386.4 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,1079.86 268.44,1079.86 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,773.326 268.44,773.326 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,466.791 268.44,466.791 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="249.542,160.256 268.44,160.256 "/>
+<path clip-path="url(#clip530)" d="M126.205 1372.19 Q122.593 1372.19 120.765 1375.76 Q118.959 1379.3 118.959 1386.43 Q118.959 1393.54 120.765 1397.1 Q122.593 1400.64 126.205 1400.64 Q129.839 1400.64 131.644 1397.1 Q133.473 1393.54 133.473 1386.43 Q133.473 1379.3 131.644 1375.76 Q129.839 1372.19 126.205 1372.19 M126.205 1368.49 Q132.015 1368.49 135.07 1373.1 Q138.149 1377.68 138.149 1386.43 Q138.149 1395.16 135.07 1399.76 Q132.015 1404.35 126.205 1404.35 Q120.394 1404.35 117.316 1399.76 Q114.26 1395.16 114.26 1386.43 Q114.26 1377.68 117.316 1373.1 Q120.394 1368.49 126.205 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M146.366 1397.8 L151.251 1397.8 L151.251 1403.68 L146.366 1403.68 L146.366 1397.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M171.436 1372.19 Q167.825 1372.19 165.996 1375.76 Q164.19 1379.3 164.19 1386.43 Q164.19 1393.54 165.996 1397.1 Q167.825 1400.64 171.436 1400.64 Q175.07 1400.64 176.876 1397.1 Q178.704 1393.54 178.704 1386.43 Q178.704 1379.3 176.876 1375.76 Q175.07 1372.19 171.436 1372.19 M171.436 1368.49 Q177.246 1368.49 180.301 1373.1 Q183.38 1377.68 183.38 1386.43 Q183.38 1395.16 180.301 1399.76 Q177.246 1404.35 171.436 1404.35 Q165.626 1404.35 162.547 1399.76 Q159.491 1395.16 159.491 1386.43 Q159.491 1377.68 162.547 1373.1 Q165.626 1368.49 171.436 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M201.598 1372.19 Q197.987 1372.19 196.158 1375.76 Q194.352 1379.3 194.352 1386.43 Q194.352 1393.54 196.158 1397.1 Q197.987 1400.64 201.598 1400.64 Q205.232 1400.64 207.037 1397.1 Q208.866 1393.54 208.866 1386.43 Q208.866 1379.3 207.037 1375.76 Q205.232 1372.19 201.598 1372.19 M201.598 1368.49 Q207.408 1368.49 210.463 1373.1 Q213.542 1377.68 213.542 1386.43 Q213.542 1395.16 210.463 1399.76 Q207.408 1404.35 201.598 1404.35 Q195.787 1404.35 192.709 1399.76 Q189.653 1395.16 189.653 1386.43 Q189.653 1377.68 192.709 1373.1 Q195.787 1368.49 201.598 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M127.2 1065.66 Q123.589 1065.66 121.76 1069.22 Q119.955 1072.77 119.955 1079.9 Q119.955 1087 121.76 1090.57 Q123.589 1094.11 127.2 1094.11 Q130.834 1094.11 132.64 1090.57 Q134.468 1087 134.468 1079.9 Q134.468 1072.77 132.64 1069.22 Q130.834 1065.66 127.2 1065.66 M127.2 1061.96 Q133.01 1061.96 136.066 1066.56 Q139.144 1071.15 139.144 1079.9 Q139.144 1088.62 136.066 1093.23 Q133.01 1097.81 127.2 1097.81 Q121.39 1097.81 118.311 1093.23 Q115.256 1088.62 115.256 1079.9 Q115.256 1071.15 118.311 1066.56 Q121.39 1061.96 127.2 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M147.362 1091.26 L152.246 1091.26 L152.246 1097.14 L147.362 1097.14 L147.362 1091.26 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M166.459 1093.21 L182.778 1093.21 L182.778 1097.14 L160.834 1097.14 L160.834 1093.21 Q163.496 1090.45 168.079 1085.82 Q172.686 1081.17 173.866 1079.83 Q176.112 1077.3 176.991 1075.57 Q177.894 1073.81 177.894 1072.12 Q177.894 1069.36 175.95 1067.63 Q174.028 1065.89 170.927 1065.89 Q168.727 1065.89 166.274 1066.65 Q163.843 1067.42 161.065 1068.97 L161.065 1064.25 Q163.89 1063.11 166.343 1062.53 Q168.797 1061.96 170.834 1061.96 Q176.204 1061.96 179.399 1064.64 Q182.593 1067.33 182.593 1071.82 Q182.593 1073.95 181.783 1075.87 Q180.996 1077.77 178.889 1080.36 Q178.311 1081.03 175.209 1084.25 Q172.107 1087.44 166.459 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M192.639 1062.58 L210.996 1062.58 L210.996 1066.52 L196.922 1066.52 L196.922 1074.99 Q197.94 1074.64 198.959 1074.48 Q199.977 1074.29 200.996 1074.29 Q206.783 1074.29 210.162 1077.47 Q213.542 1080.64 213.542 1086.05 Q213.542 1091.63 210.07 1094.73 Q206.598 1097.81 200.278 1097.81 Q198.102 1097.81 195.834 1097.44 Q193.588 1097.07 191.181 1096.33 L191.181 1091.63 Q193.264 1092.77 195.487 1093.32 Q197.709 1093.88 200.186 1093.88 Q204.19 1093.88 206.528 1091.77 Q208.866 1089.66 208.866 1086.05 Q208.866 1082.44 206.528 1080.34 Q204.19 1078.23 200.186 1078.23 Q198.311 1078.23 196.436 1078.65 Q194.584 1079.06 192.639 1079.94 L192.639 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M126.205 759.125 Q122.593 759.125 120.765 762.689 Q118.959 766.231 118.959 773.361 Q118.959 780.467 120.765 784.032 Q122.593 787.574 126.205 787.574 Q129.839 787.574 131.644 784.032 Q133.473 780.467 133.473 773.361 Q133.473 766.231 131.644 762.689 Q129.839 759.125 126.205 759.125 M126.205 755.421 Q132.015 755.421 135.07 760.027 Q138.149 764.611 138.149 773.361 Q138.149 782.088 135.07 786.694 Q132.015 791.277 126.205 791.277 Q120.394 791.277 117.316 786.694 Q114.26 782.088 114.26 773.361 Q114.26 764.611 117.316 760.027 Q120.394 755.421 126.205 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M146.366 784.726 L151.251 784.726 L151.251 790.606 L146.366 790.606 L146.366 784.726 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M161.482 756.046 L179.839 756.046 L179.839 759.981 L165.765 759.981 L165.765 768.453 Q166.783 768.106 167.802 767.944 Q168.82 767.759 169.839 767.759 Q175.626 767.759 179.005 770.93 Q182.385 774.101 182.385 779.518 Q182.385 785.097 178.913 788.199 Q175.44 791.277 169.121 791.277 Q166.945 791.277 164.677 790.907 Q162.431 790.537 160.024 789.796 L160.024 785.097 Q162.107 786.231 164.329 786.787 Q166.552 787.342 169.028 787.342 Q173.033 787.342 175.371 785.236 Q177.709 783.129 177.709 779.518 Q177.709 775.907 175.371 773.801 Q173.033 771.694 169.028 771.694 Q167.153 771.694 165.278 772.111 Q163.427 772.527 161.482 773.407 L161.482 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M201.598 759.125 Q197.987 759.125 196.158 762.689 Q194.352 766.231 194.352 773.361 Q194.352 780.467 196.158 784.032 Q197.987 787.574 201.598 787.574 Q205.232 787.574 207.037 784.032 Q208.866 780.467 208.866 773.361 Q208.866 766.231 207.037 762.689 Q205.232 759.125 201.598 759.125 M201.598 755.421 Q207.408 755.421 210.463 760.027 Q213.542 764.611 213.542 773.361 Q213.542 782.088 210.463 786.694 Q207.408 791.277 201.598 791.277 Q195.787 791.277 192.709 786.694 Q189.653 782.088 189.653 773.361 Q189.653 764.611 192.709 760.027 Q195.787 755.421 201.598 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M127.2 452.59 Q123.589 452.59 121.76 456.155 Q119.955 459.696 119.955 466.826 Q119.955 473.932 121.76 477.497 Q123.589 481.039 127.2 481.039 Q130.834 481.039 132.64 477.497 Q134.468 473.932 134.468 466.826 Q134.468 459.696 132.64 456.155 Q130.834 452.59 127.2 452.59 M127.2 448.886 Q133.01 448.886 136.066 453.493 Q139.144 458.076 139.144 466.826 Q139.144 475.553 136.066 480.159 Q133.01 484.742 127.2 484.742 Q121.39 484.742 118.311 480.159 Q115.256 475.553 115.256 466.826 Q115.256 458.076 118.311 453.493 Q121.39 448.886 127.2 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M147.362 478.192 L152.246 478.192 L152.246 484.071 L147.362 484.071 L147.362 478.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M161.251 449.511 L183.473 449.511 L183.473 451.502 L170.927 484.071 L166.042 484.071 L177.848 453.446 L161.251 453.446 L161.251 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M192.639 449.511 L210.996 449.511 L210.996 453.446 L196.922 453.446 L196.922 461.918 Q197.94 461.571 198.959 461.409 Q199.977 461.224 200.996 461.224 Q206.783 461.224 210.162 464.395 Q213.542 467.567 213.542 472.983 Q213.542 478.562 210.07 481.664 Q206.598 484.742 200.278 484.742 Q198.102 484.742 195.834 484.372 Q193.588 484.002 191.181 483.261 L191.181 478.562 Q193.264 479.696 195.487 480.252 Q197.709 480.807 200.186 480.807 Q204.19 480.807 206.528 478.701 Q208.866 476.594 208.866 472.983 Q208.866 469.372 206.528 467.266 Q204.19 465.159 200.186 465.159 Q198.311 465.159 196.436 465.576 Q194.584 465.993 192.639 466.872 L192.639 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M117.015 173.601 L124.654 173.601 L124.654 147.236 L116.343 148.902 L116.343 144.643 L124.607 142.976 L129.283 142.976 L129.283 173.601 L136.922 173.601 L136.922 177.536 L117.015 177.536 L117.015 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M146.366 171.657 L151.251 171.657 L151.251 177.536 L146.366 177.536 L146.366 171.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M171.436 146.055 Q167.825 146.055 165.996 149.62 Q164.19 153.161 164.19 160.291 Q164.19 167.397 165.996 170.962 Q167.825 174.504 171.436 174.504 Q175.07 174.504 176.876 170.962 Q178.704 167.397 178.704 160.291 Q178.704 153.161 176.876 149.62 Q175.07 146.055 171.436 146.055 M171.436 142.351 Q177.246 142.351 180.301 146.958 Q183.38 151.541 183.38 160.291 Q183.38 169.018 180.301 173.624 Q177.246 178.208 171.436 178.208 Q165.626 178.208 162.547 173.624 Q159.491 169.018 159.491 160.291 Q159.491 151.541 162.547 146.958 Q165.626 142.351 171.436 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M201.598 146.055 Q197.987 146.055 196.158 149.62 Q194.352 153.161 194.352 160.291 Q194.352 167.397 196.158 170.962 Q197.987 174.504 201.598 174.504 Q205.232 174.504 207.037 170.962 Q208.866 167.397 208.866 160.291 Q208.866 153.161 207.037 149.62 Q205.232 146.055 201.598 146.055 M201.598 142.351 Q207.408 142.351 210.463 146.958 Q213.542 151.541 213.542 160.291 Q213.542 169.018 210.463 173.624 Q207.408 178.208 201.598 178.208 Q195.787 178.208 192.709 173.624 Q189.653 169.018 189.653 160.291 Q189.653 151.541 192.709 146.958 Q195.787 142.351 201.598 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M20.1444 1201.1 L26.9239 1201.1 Q23.9002 1204.35 22.4043 1208.04 Q20.9083 1211.7 20.9083 1215.84 Q20.9083 1223.99 25.9054 1228.31 Q30.8707 1232.64 40.2919 1232.64 Q49.6813 1232.64 54.6784 1228.31 Q59.6436 1223.99 59.6436 1215.84 Q59.6436 1211.7 58.1477 1208.04 Q56.6518 1204.35 53.6281 1201.1 L60.3439 1201.1 Q62.6355 1204.48 63.7814 1208.26 Q64.9272 1212.02 64.9272 1216.22 Q64.9272 1227.01 58.3387 1233.22 Q51.7183 1239.42 40.2919 1239.42 Q28.8336 1239.42 22.2451 1233.22 Q15.6248 1227.01 15.6248 1216.22 Q15.6248 1211.95 16.7706 1208.2 Q17.8846 1204.41 20.1444 1201.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M32.4621 1177.61 Q32.4621 1182.32 36.1542 1185.06 Q39.8145 1187.8 46.212 1187.8 Q52.6095 1187.8 56.3017 1185.09 Q59.9619 1182.35 59.9619 1177.61 Q59.9619 1172.93 56.2698 1170.2 Q52.5777 1167.46 46.212 1167.46 Q39.8781 1167.46 36.186 1170.2 Q32.4621 1172.93 32.4621 1177.61 M27.4968 1177.61 Q27.4968 1169.97 32.4621 1165.61 Q37.4273 1161.25 46.212 1161.25 Q54.9649 1161.25 59.9619 1165.61 Q64.9272 1169.97 64.9272 1177.61 Q64.9272 1185.28 59.9619 1189.64 Q54.9649 1193.97 46.212 1193.97 Q37.4273 1193.97 32.4621 1189.64 Q27.4968 1185.28 27.4968 1177.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M35.1993 1123.79 Q31.2526 1121.59 29.3747 1118.54 Q27.4968 1115.48 27.4968 1111.35 Q27.4968 1105.78 31.4117 1102.75 Q35.2948 1099.73 42.4881 1099.73 L64.0042 1099.73 L64.0042 1105.62 L42.679 1105.62 Q37.5546 1105.62 35.072 1107.43 Q32.5894 1109.24 32.5894 1112.97 Q32.5894 1117.52 35.6131 1120.16 Q38.6368 1122.8 43.8567 1122.8 L64.0042 1122.8 L64.0042 1128.69 L42.679 1128.69 Q37.5228 1128.69 35.072 1130.51 Q32.5894 1132.32 32.5894 1136.11 Q32.5894 1140.6 35.6449 1143.24 Q38.6686 1145.88 43.8567 1145.88 L64.0042 1145.88 L64.0042 1151.77 L28.3562 1151.77 L28.3562 1145.88 L33.8944 1145.88 Q30.616 1143.87 29.0564 1141.07 Q27.4968 1138.27 27.4968 1134.42 Q27.4968 1130.54 29.4702 1127.83 Q31.4436 1125.09 35.1993 1123.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M35.1993 1060.29 Q31.2526 1058.1 29.3747 1055.04 Q27.4968 1051.98 27.4968 1047.85 Q27.4968 1042.28 31.4117 1039.25 Q35.2948 1036.23 42.4881 1036.23 L64.0042 1036.23 L64.0042 1042.12 L42.679 1042.12 Q37.5546 1042.12 35.072 1043.93 Q32.5894 1045.75 32.5894 1049.47 Q32.5894 1054.02 35.6131 1056.66 Q38.6368 1059.31 43.8567 1059.31 L64.0042 1059.31 L64.0042 1065.19 L42.679 1065.19 Q37.5228 1065.19 35.072 1067.01 Q32.5894 1068.82 32.5894 1072.61 Q32.5894 1077.1 35.6449 1079.74 Q38.6686 1082.38 43.8567 1082.38 L64.0042 1082.38 L64.0042 1088.27 L28.3562 1088.27 L28.3562 1082.38 L33.8944 1082.38 Q30.616 1080.38 29.0564 1077.57 Q27.4968 1074.77 27.4968 1070.92 Q27.4968 1067.04 29.4702 1064.33 Q31.4436 1061.6 35.1993 1060.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M28.3562 1024.55 L28.3562 1018.69 L64.0042 1018.69 L64.0042 1024.55 L28.3562 1024.55 M14.479 1024.55 L14.479 1018.69 L21.895 1018.69 L21.895 1024.55 L14.479 1024.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M18.2347 1000.65 L28.3562 1000.65 L28.3562 988.582 L32.9077 988.582 L32.9077 1000.65 L52.2594 1000.65 Q56.6199 1000.65 57.8613 999.468 Q59.1026 998.258 59.1026 994.598 L59.1026 988.582 L64.0042 988.582 L64.0042 994.598 Q64.0042 1001.38 61.4897 1003.96 Q58.9434 1006.53 52.2594 1006.53 L32.9077 1006.53 L32.9077 1010.83 L28.3562 1010.83 L28.3562 1006.53 L18.2347 1006.53 L18.2347 1000.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M35.1993 953.125 Q31.2526 950.929 29.3747 947.874 Q27.4968 944.818 27.4968 940.68 Q27.4968 935.11 31.4117 932.087 Q35.2948 929.063 42.4881 929.063 L64.0042 929.063 L64.0042 934.951 L42.679 934.951 Q37.5546 934.951 35.072 936.766 Q32.5894 938.58 32.5894 942.304 Q32.5894 946.855 35.6131 949.497 Q38.6368 952.139 43.8567 952.139 L64.0042 952.139 L64.0042 958.027 L42.679 958.027 Q37.5228 958.027 35.072 959.841 Q32.5894 961.655 32.5894 965.443 Q32.5894 969.931 35.6449 972.573 Q38.6686 975.214 43.8567 975.214 L64.0042 975.214 L64.0042 981.103 L28.3562 981.103 L28.3562 975.214 L33.8944 975.214 Q30.616 973.209 29.0564 970.408 Q27.4968 967.607 27.4968 963.756 Q27.4968 959.873 29.4702 957.168 Q31.4436 954.43 35.1993 953.125 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M44.7161 886.89 L47.5806 886.89 L47.5806 913.817 Q53.6281 913.435 56.8109 910.189 Q59.9619 906.91 59.9619 901.086 Q59.9619 897.712 59.1344 894.561 Q58.3069 891.378 56.6518 888.259 L62.1899 888.259 Q63.5267 891.41 64.227 894.72 Q64.9272 898.03 64.9272 901.436 Q64.9272 909.966 59.9619 914.963 Q54.9967 919.928 46.5303 919.928 Q37.7774 919.928 32.6531 915.218 Q27.4968 910.475 27.4968 902.454 Q27.4968 895.261 32.1438 891.092 Q36.7589 886.89 44.7161 886.89 M42.9973 892.747 Q38.1912 892.81 35.3266 895.452 Q32.4621 898.062 32.4621 902.391 Q32.4621 907.292 35.2312 910.252 Q38.0002 913.181 43.0292 913.626 L42.9973 892.747 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M42.4881 847.646 L64.0042 847.646 L64.0042 853.502 L42.679 853.502 Q37.6183 853.502 35.1038 855.475 Q32.5894 857.449 32.5894 861.396 Q32.5894 866.138 35.6131 868.875 Q38.6368 871.612 43.8567 871.612 L64.0042 871.612 L64.0042 877.501 L28.3562 877.501 L28.3562 871.612 L33.8944 871.612 Q30.6797 869.512 29.0883 866.679 Q27.4968 863.815 27.4968 860.091 Q27.4968 853.948 31.3163 850.797 Q35.1038 847.646 42.4881 847.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M18.2347 830.172 L28.3562 830.172 L28.3562 818.109 L32.9077 818.109 L32.9077 830.172 L52.2594 830.172 Q56.6199 830.172 57.8613 828.994 Q59.1026 827.785 59.1026 824.124 L59.1026 818.109 L64.0042 818.109 L64.0042 824.124 Q64.0042 830.904 61.4897 833.482 Q58.9434 836.06 52.2594 836.06 L32.9077 836.06 L32.9077 840.357 L28.3562 840.357 L28.3562 836.06 L18.2347 836.06 L18.2347 830.172 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M33.7671 766.228 L14.479 766.228 L14.479 760.372 L64.0042 760.372 L64.0042 766.228 L58.657 766.228 Q61.8398 768.074 63.3994 770.907 Q64.9272 773.708 64.9272 777.655 Q64.9272 784.116 59.771 788.19 Q54.6147 792.232 46.212 792.232 Q37.8093 792.232 32.6531 788.19 Q27.4968 784.116 27.4968 777.655 Q27.4968 773.708 29.0564 770.907 Q30.5842 768.074 33.7671 766.228 M46.212 786.185 Q52.6732 786.185 56.3653 783.543 Q60.0256 780.869 60.0256 776.222 Q60.0256 771.575 56.3653 768.902 Q52.6732 766.228 46.212 766.228 Q39.7508 766.228 36.0905 768.902 Q32.3984 771.575 32.3984 776.222 Q32.3984 780.869 36.0905 783.543 Q39.7508 786.185 46.212 786.185 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M44.7161 717.817 L47.5806 717.817 L47.5806 744.744 Q53.6281 744.362 56.8109 741.116 Q59.9619 737.837 59.9619 732.013 Q59.9619 728.639 59.1344 725.488 Q58.3069 722.305 56.6518 719.186 L62.1899 719.186 Q63.5267 722.337 64.227 725.647 Q64.9272 728.957 64.9272 732.363 Q64.9272 740.893 59.9619 745.89 Q54.9967 750.855 46.5303 750.855 Q37.7774 750.855 32.6531 746.144 Q27.4968 741.402 27.4968 733.381 Q27.4968 726.188 32.1438 722.018 Q36.7589 717.817 44.7161 717.817 M42.9973 723.673 Q38.1912 723.737 35.3266 726.379 Q32.4621 728.989 32.4621 733.318 Q32.4621 738.219 35.2312 741.179 Q38.0002 744.107 43.0292 744.553 L42.9973 723.673 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M29.7248 682.551 L35.1993 682.551 Q33.8307 685.034 33.1623 687.548 Q32.4621 690.031 32.4621 692.577 Q32.4621 698.274 36.0905 701.425 Q39.6872 704.576 46.212 704.576 Q52.7369 704.576 56.3653 701.425 Q59.9619 698.274 59.9619 692.577 Q59.9619 690.031 59.2935 687.548 Q58.5933 685.034 57.2247 682.551 L62.6355 682.551 Q63.7814 685.002 64.3543 687.644 Q64.9272 690.254 64.9272 693.214 Q64.9272 701.266 59.8664 706.009 Q54.8057 710.751 46.212 710.751 Q37.491 710.751 32.4939 705.977 Q27.4968 701.171 27.4968 692.832 Q27.4968 690.126 28.0697 687.548 Q28.6108 684.97 29.7248 682.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M28.3562 672.366 L28.3562 666.509 L64.0042 666.509 L64.0042 672.366 L28.3562 672.366 M14.479 672.366 L14.479 666.509 L21.895 666.509 L21.895 672.366 L14.479 672.366 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M29.4065 631.53 L34.9447 631.53 Q33.6716 634.012 33.035 636.686 Q32.3984 639.36 32.3984 642.224 Q32.3984 646.585 33.7352 648.781 Q35.072 650.945 37.7456 650.945 Q39.7826 650.945 40.9603 649.386 Q42.1061 647.826 43.1565 643.115 L43.6021 641.11 Q44.9389 634.872 47.3897 632.262 Q49.8086 629.62 54.1691 629.62 Q59.1344 629.62 62.0308 633.567 Q64.9272 637.482 64.9272 644.357 Q64.9272 647.221 64.3543 650.341 Q63.8132 653.428 62.6992 656.865 L56.6518 656.865 Q58.3387 653.619 59.198 650.468 Q60.0256 647.317 60.0256 644.229 Q60.0256 640.092 58.6251 637.864 Q57.1929 635.636 54.6147 635.636 Q52.2276 635.636 50.9545 637.259 Q49.6813 638.85 48.5037 644.293 L48.0262 646.33 Q46.8804 651.773 44.5251 654.192 Q42.138 656.611 38.0002 656.611 Q32.9713 656.611 30.2341 653.046 Q27.4968 649.481 27.4968 642.924 Q27.4968 639.678 27.9743 636.813 Q28.4517 633.949 29.4065 631.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M28.3562 620.294 L28.3562 614.438 L64.0042 614.438 L64.0042 620.294 L28.3562 620.294 M14.479 620.294 L14.479 614.438 L21.895 614.438 L21.895 620.294 L14.479 620.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M32.4621 588.37 Q32.4621 593.081 36.1542 595.818 Q39.8145 598.556 46.212 598.556 Q52.6095 598.556 56.3017 595.85 Q59.9619 593.113 59.9619 588.37 Q59.9619 583.692 56.2698 580.954 Q52.5777 578.217 46.212 578.217 Q39.8781 578.217 36.186 580.954 Q32.4621 583.692 32.4621 588.37 M27.4968 588.37 Q27.4968 580.732 32.4621 576.371 Q37.4273 572.011 46.212 572.011 Q54.9649 572.011 59.9619 576.371 Q64.9272 580.732 64.9272 588.37 Q64.9272 596.041 59.9619 600.402 Q54.9649 604.73 46.212 604.73 Q37.4273 604.73 32.4621 600.402 Q27.4968 596.041 27.4968 588.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M42.4881 532.67 L64.0042 532.67 L64.0042 538.527 L42.679 538.527 Q37.6183 538.527 35.1038 540.5 Q32.5894 542.474 32.5894 546.42 Q32.5894 551.163 35.6131 553.9 Q38.6368 556.637 43.8567 556.637 L64.0042 556.637 L64.0042 562.526 L28.3562 562.526 L28.3562 556.637 L33.8944 556.637 Q30.6797 554.537 29.0883 551.704 Q27.4968 548.839 27.4968 545.115 Q27.4968 538.972 31.3163 535.821 Q35.1038 532.67 42.4881 532.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M70.0516 473.087 L74.6349 473.087 L74.6349 475.061 Q74.6349 482.986 72.2796 485.692 Q69.9243 488.365 62.8902 488.365 L55.2831 488.365 Q50.477 488.365 48.631 490.084 Q46.7849 491.803 46.7849 496.322 L46.7849 498.264 L42.2334 498.264 L42.2334 496.322 Q42.2334 491.771 40.4192 490.084 Q38.5732 488.365 33.8307 488.365 L26.1919 488.365 Q19.1578 488.365 16.8343 485.692 Q14.479 482.986 14.479 475.061 L14.479 473.087 L19.0304 473.087 L19.0304 475.252 Q19.0304 479.74 20.4309 481.108 Q21.8314 482.477 26.3192 482.477 L34.2127 482.477 Q39.2097 482.477 41.4696 483.941 Q43.7294 485.373 44.5251 488.874 Q45.3845 485.341 47.6443 483.909 Q49.9041 482.477 54.8694 482.477 L62.7628 482.477 Q67.2507 482.477 68.6511 481.108 Q70.0516 479.74 70.0516 475.252 L70.0516 473.087 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M20.7174 444.219 Q20.7174 449.184 25.6189 451.699 Q30.4887 454.181 40.2919 454.181 Q50.0633 454.181 54.9649 451.699 Q59.8346 449.184 59.8346 444.219 Q59.8346 439.222 54.9649 436.739 Q50.0633 434.225 40.2919 434.225 Q30.4887 434.225 25.6189 436.739 Q20.7174 439.222 20.7174 444.219 M15.6248 444.219 Q15.6248 436.23 21.9587 432.029 Q28.2607 427.795 40.2919 427.795 Q52.2913 427.795 58.6251 432.029 Q64.9272 436.23 64.9272 444.219 Q64.9272 452.208 58.6251 456.441 Q52.2913 460.642 40.2919 460.642 Q28.2607 460.642 21.9587 456.441 Q15.6248 452.208 15.6248 444.219 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M55.9197 415.828 L55.9197 409.112 L61.3942 409.112 L71.5793 414.332 L71.5793 418.438 L61.3942 415.828 L55.9197 415.828 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M58.5933 373.942 L58.5933 363.438 L22.3406 363.438 L24.6323 374.865 L18.7758 374.865 L16.4842 363.502 L16.4842 357.072 L58.5933 357.072 L58.5933 346.569 L64.0042 346.569 L64.0042 373.942 L58.5933 373.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M70.0516 332.405 L70.0516 330.177 Q70.0516 325.721 68.683 324.385 Q67.3143 323.016 62.7628 323.016 L54.8694 323.016 Q49.9041 323.016 47.6443 321.584 Q45.3845 320.151 44.5251 316.618 Q43.7294 320.151 41.4696 321.584 Q39.2097 323.016 34.2127 323.016 L26.3192 323.016 Q21.7995 323.016 20.4309 324.385 Q19.0304 325.721 19.0304 330.177 L19.0304 332.405 L14.479 332.405 L14.479 330.4 Q14.479 322.475 16.8343 319.833 Q19.1578 317.159 26.1919 317.159 L33.8307 317.159 Q38.5732 317.159 40.4192 315.441 Q42.2334 313.722 42.2334 309.202 L42.2334 307.229 L46.7849 307.229 L46.7849 309.202 Q46.7849 313.722 48.631 315.441 Q50.477 317.159 55.2831 317.159 L62.8902 317.159 Q69.9243 317.159 72.2796 319.833 Q74.6349 322.475 74.6349 330.4 L74.6349 332.405 L70.0516 332.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M478.56 16.7545 L478.56 25.383 Q474.429 21.5346 469.729 19.6307 Q465.071 17.7268 459.805 17.7268 Q449.434 17.7268 443.925 24.0867 Q438.416 30.4061 438.416 42.3968 Q438.416 54.3469 443.925 60.7069 Q449.434 67.0263 459.805 67.0263 Q465.071 67.0263 469.729 65.1223 Q474.429 63.2184 478.56 59.3701 L478.56 67.9175 Q474.267 70.8341 469.446 72.2924 Q464.666 73.7508 459.319 73.7508 Q445.586 73.7508 437.687 65.3654 Q429.788 56.9395 429.788 42.3968 Q429.788 27.8135 437.687 19.4281 Q445.586 11.0023 459.319 11.0023 Q464.747 11.0023 469.527 12.4606 Q474.348 13.8784 478.56 16.7545 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M508.456 32.4315 Q502.461 32.4315 498.977 37.1306 Q495.493 41.7891 495.493 49.9314 Q495.493 58.0738 498.937 62.7728 Q502.42 67.4314 508.456 67.4314 Q514.411 67.4314 517.895 62.7323 Q521.379 58.0333 521.379 49.9314 Q521.379 41.8701 517.895 37.1711 Q514.411 32.4315 508.456 32.4315 M508.456 26.1121 Q518.178 26.1121 523.728 32.4315 Q529.278 38.7509 529.278 49.9314 Q529.278 61.0714 523.728 67.4314 Q518.178 73.7508 508.456 73.7508 Q498.693 73.7508 493.144 67.4314 Q487.634 61.0714 487.634 49.9314 Q487.634 38.7509 493.144 32.4315 Q498.693 26.1121 508.456 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M576.957 35.9153 Q579.752 30.8922 583.641 28.5022 Q587.53 26.1121 592.796 26.1121 Q599.885 26.1121 603.733 31.0947 Q607.582 36.0368 607.582 45.1919 L607.582 72.576 L600.088 72.576 L600.088 45.4349 Q600.088 38.913 597.779 35.7533 Q595.47 32.5936 590.73 32.5936 Q584.937 32.5936 581.575 36.4419 Q578.213 40.2903 578.213 46.9338 L578.213 72.576 L570.719 72.576 L570.719 45.4349 Q570.719 38.8725 568.41 35.7533 Q566.101 32.5936 561.28 32.5936 Q555.568 32.5936 552.206 36.4824 Q548.844 40.3308 548.844 46.9338 L548.844 72.576 L541.349 72.576 L541.349 27.2059 L548.844 27.2059 L548.844 34.2544 Q551.396 30.082 554.961 28.0971 Q558.525 26.1121 563.427 26.1121 Q568.369 26.1121 571.812 28.6237 Q575.296 31.1352 576.957 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M657.772 35.9153 Q660.568 30.8922 664.456 28.5022 Q668.345 26.1121 673.612 26.1121 Q680.701 26.1121 684.549 31.0947 Q688.397 36.0368 688.397 45.1919 L688.397 72.576 L680.903 72.576 L680.903 45.4349 Q680.903 38.913 678.594 35.7533 Q676.285 32.5936 671.546 32.5936 Q665.753 32.5936 662.39 36.4419 Q659.028 40.2903 659.028 46.9338 L659.028 72.576 L651.534 72.576 L651.534 45.4349 Q651.534 38.8725 649.225 35.7533 Q646.916 32.5936 642.095 32.5936 Q636.384 32.5936 633.021 36.4824 Q629.659 40.3308 629.659 46.9338 L629.659 72.576 L622.165 72.576 L622.165 27.2059 L629.659 27.2059 L629.659 34.2544 Q632.211 30.082 635.776 28.0971 Q639.341 26.1121 644.242 26.1121 Q649.185 26.1121 652.628 28.6237 Q656.112 31.1352 657.772 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M703.264 27.2059 L710.718 27.2059 L710.718 72.576 L703.264 72.576 L703.264 27.2059 M703.264 9.54393 L710.718 9.54393 L710.718 18.9825 L703.264 18.9825 L703.264 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M733.686 14.324 L733.686 27.2059 L749.039 27.2059 L749.039 32.9987 L733.686 32.9987 L733.686 57.6282 Q733.686 63.1779 735.185 64.7578 Q736.725 66.3376 741.383 66.3376 L749.039 66.3376 L749.039 72.576 L741.383 72.576 Q732.755 72.576 729.473 69.3758 Q726.192 66.1351 726.192 57.6282 L726.192 32.9987 L720.724 32.9987 L720.724 27.2059 L726.192 27.2059 L726.192 14.324 L733.686 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M794.166 35.9153 Q796.962 30.8922 800.85 28.5022 Q804.739 26.1121 810.005 26.1121 Q817.095 26.1121 820.943 31.0947 Q824.791 36.0368 824.791 45.1919 L824.791 72.576 L817.297 72.576 L817.297 45.4349 Q817.297 38.913 814.988 35.7533 Q812.679 32.5936 807.939 32.5936 Q802.147 32.5936 798.784 36.4419 Q795.422 40.2903 795.422 46.9338 L795.422 72.576 L787.928 72.576 L787.928 45.4349 Q787.928 38.8725 785.619 35.7533 Q783.31 32.5936 778.489 32.5936 Q772.778 32.5936 769.415 36.4824 Q766.053 40.3308 766.053 46.9338 L766.053 72.576 L758.559 72.576 L758.559 27.2059 L766.053 27.2059 L766.053 34.2544 Q768.605 30.082 772.17 28.0971 Q775.735 26.1121 780.636 26.1121 Q785.578 26.1121 789.022 28.6237 Q792.506 31.1352 794.166 35.9153 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M878.466 48.0275 L878.466 51.6733 L844.195 51.6733 Q844.681 59.3701 848.813 63.421 Q852.986 67.4314 860.399 67.4314 Q864.693 67.4314 868.703 66.3781 Q872.754 65.3249 876.724 63.2184 L876.724 70.267 Q872.713 71.9684 868.501 72.8596 Q864.288 73.7508 859.953 73.7508 Q849.097 73.7508 842.737 67.4314 Q836.417 61.1119 836.417 50.3365 Q836.417 39.1965 842.413 32.6746 Q848.449 26.1121 858.657 26.1121 Q867.812 26.1121 873.119 32.0264 Q878.466 37.9003 878.466 48.0275 M871.012 45.84 Q870.931 39.7232 867.569 36.0774 Q864.247 32.4315 858.738 32.4315 Q852.499 32.4315 848.732 35.9558 Q845.005 39.4801 844.438 45.8805 L871.012 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M928.413 45.1919 L928.413 72.576 L920.96 72.576 L920.96 45.4349 Q920.96 38.994 918.448 35.7938 Q915.937 32.5936 910.913 32.5936 Q904.878 32.5936 901.394 36.4419 Q897.91 40.2903 897.91 46.9338 L897.91 72.576 L890.416 72.576 L890.416 27.2059 L897.91 27.2059 L897.91 34.2544 Q900.584 30.163 904.189 28.1376 Q907.835 26.1121 912.574 26.1121 Q920.393 26.1121 924.403 30.9732 Q928.413 35.7938 928.413 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M950.653 14.324 L950.653 27.2059 L966.006 27.2059 L966.006 32.9987 L950.653 32.9987 L950.653 57.6282 Q950.653 63.1779 952.152 64.7578 Q953.691 66.3376 958.35 66.3376 L966.006 66.3376 L966.006 72.576 L958.35 72.576 Q949.721 72.576 946.44 69.3758 Q943.159 66.1351 943.159 57.6282 L943.159 32.9987 L937.69 32.9987 L937.69 27.2059 L943.159 27.2059 L943.159 14.324 L950.653 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip532)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="274.921,160.256 345.417,160.256 415.913,160.256 486.409,160.256 556.905,160.256 627.401,160.256 697.897,160.256 768.393,160.256 838.889,160.256 909.385,160.256 979.881,160.256 1050.38,160.256 1120.87,160.256 "/>
+<polyline clip-path="url(#clip532)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="274.921,1386.4 345.417,1386.4 415.913,1386.4 486.409,1386.4 556.905,1386.4 627.401,1386.4 697.897,1386.4 768.393,160.256 838.889,160.256 909.385,160.256 979.881,160.256 1050.38,160.256 1120.87,160.256 "/>
+<path clip-path="url(#clip530)" d="M966.829 1379.86 L1116.36 1379.86 L1116.36 1224.34 L966.829 1224.34  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="966.829,1379.86 1116.36,1379.86 1116.36,1224.34 966.829,1224.34 966.829,1379.86 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="976.793,1276.18 1036.57,1276.18 "/>
+<path clip-path="url(#clip530)" d="M1072.09 1288.53 L1072.09 1279.24 L1064.45 1279.24 L1064.45 1275.4 L1076.72 1275.4 L1076.72 1290.24 Q1074.01 1292.16 1070.75 1293.16 Q1067.49 1294.13 1063.78 1294.13 Q1055.68 1294.13 1051.1 1289.41 Q1046.54 1284.66 1046.54 1276.21 Q1046.54 1267.74 1051.1 1263.02 Q1055.68 1258.27 1063.78 1258.27 Q1067.16 1258.27 1070.19 1259.1 Q1073.25 1259.94 1075.82 1261.56 L1075.82 1266.54 Q1073.23 1264.34 1070.31 1263.22 Q1067.39 1262.11 1064.18 1262.11 Q1057.83 1262.11 1054.64 1265.66 Q1051.47 1269.2 1051.47 1276.21 Q1051.47 1283.2 1054.64 1286.74 Q1057.83 1290.28 1064.18 1290.28 Q1066.65 1290.28 1068.6 1289.87 Q1070.54 1289.43 1072.09 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1086.49 1289.52 L1094.13 1289.52 L1094.13 1263.16 L1085.82 1264.82 L1085.82 1260.56 L1094.08 1258.9 L1098.76 1258.9 L1098.76 1289.52 L1106.4 1289.52 L1106.4 1293.46 L1086.49 1293.46 L1086.49 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip530)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="976.793,1328.02 1036.57,1328.02 "/>
+<path clip-path="url(#clip530)" d="M1072.09 1340.37 L1072.09 1331.08 L1064.45 1331.08 L1064.45 1327.24 L1076.72 1327.24 L1076.72 1342.08 Q1074.01 1344 1070.75 1345 Q1067.49 1345.97 1063.78 1345.97 Q1055.68 1345.97 1051.1 1341.25 Q1046.54 1336.5 1046.54 1328.05 Q1046.54 1319.58 1051.1 1314.86 Q1055.68 1310.11 1063.78 1310.11 Q1067.16 1310.11 1070.19 1310.94 Q1073.25 1311.78 1075.82 1313.4 L1075.82 1318.38 Q1073.23 1316.18 1070.31 1315.06 Q1067.39 1313.95 1064.18 1313.95 Q1057.83 1313.95 1054.64 1317.5 Q1051.47 1321.04 1051.47 1328.05 Q1051.47 1335.04 1054.64 1338.58 Q1057.83 1342.12 1064.18 1342.12 Q1066.65 1342.12 1068.6 1341.71 Q1070.54 1341.27 1072.09 1340.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1089.71 1341.36 L1106.03 1341.36 L1106.03 1345.3 L1084.08 1345.3 L1084.08 1341.36 Q1086.74 1338.61 1091.33 1333.98 Q1095.93 1329.32 1097.12 1327.98 Q1099.36 1325.46 1100.24 1323.72 Q1101.14 1321.96 1101.14 1320.27 Q1101.14 1317.52 1099.2 1315.78 Q1097.28 1314.05 1094.18 1314.05 Q1091.98 1314.05 1089.52 1314.81 Q1087.09 1315.57 1084.31 1317.13 L1084.31 1312.4 Q1087.14 1311.27 1089.59 1310.69 Q1092.05 1310.11 1094.08 1310.11 Q1099.45 1310.11 1102.65 1312.8 Q1105.84 1315.48 1105.84 1319.97 Q1105.84 1322.1 1105.03 1324.02 Q1104.24 1325.92 1102.14 1328.51 Q1101.56 1329.19 1098.46 1332.4 Q1095.36 1335.6 1089.71 1341.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1456.05 1423.18 L2352.76 1423.18 L2352.76 123.472 L1456.05 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip713">
+  <clipPath id="clip533">
     <rect x="1456" y="123" width="898" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1575.42,1423.18 1575.42,123.472 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1810.41,1423.18 1810.41,123.472 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2045.39,1423.18 2045.39,123.472 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2280.38,1423.18 2280.38,123.472 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1575.42,1423.18 1575.42,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1810.41,1423.18 1810.41,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2045.39,1423.18 2045.39,1404.28 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2280.38,1423.18 2280.38,1404.28 "/>
-<path clip-path="url(#clip710)" d="M1535.04 1451.02 L1553.39 1451.02 L1553.39 1454.96 L1539.32 1454.96 L1539.32 1463.43 Q1540.34 1463.08 1541.36 1462.92 Q1542.38 1462.73 1543.39 1462.73 Q1549.18 1462.73 1552.56 1465.9 Q1555.94 1469.08 1555.94 1474.49 Q1555.94 1480.07 1552.47 1483.17 Q1549 1486.25 1542.68 1486.25 Q1540.5 1486.25 1538.23 1485.88 Q1535.99 1485.51 1533.58 1484.77 L1533.58 1480.07 Q1535.66 1481.2 1537.89 1481.76 Q1540.11 1482.32 1542.58 1482.32 Q1546.59 1482.32 1548.93 1480.21 Q1551.26 1478.1 1551.26 1474.49 Q1551.26 1470.88 1548.93 1468.77 Q1546.59 1466.67 1542.58 1466.67 Q1540.71 1466.67 1538.83 1467.08 Q1536.98 1467.5 1535.04 1468.38 L1535.04 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1575.15 1454.1 Q1571.54 1454.1 1569.71 1457.66 Q1567.91 1461.2 1567.91 1468.33 Q1567.91 1475.44 1569.71 1479.01 Q1571.54 1482.55 1575.15 1482.55 Q1578.79 1482.55 1580.59 1479.01 Q1582.42 1475.44 1582.42 1468.33 Q1582.42 1461.2 1580.59 1457.66 Q1578.79 1454.1 1575.15 1454.1 M1575.15 1450.39 Q1580.96 1450.39 1584.02 1455 Q1587.1 1459.58 1587.1 1468.33 Q1587.1 1477.06 1584.02 1481.67 Q1580.96 1486.25 1575.15 1486.25 Q1569.34 1486.25 1566.26 1481.67 Q1563.21 1477.06 1563.21 1468.33 Q1563.21 1459.58 1566.26 1455 Q1569.34 1450.39 1575.15 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1605.32 1454.1 Q1601.7 1454.1 1599.88 1457.66 Q1598.07 1461.2 1598.07 1468.33 Q1598.07 1475.44 1599.88 1479.01 Q1601.7 1482.55 1605.32 1482.55 Q1608.95 1482.55 1610.76 1479.01 Q1612.58 1475.44 1612.58 1468.33 Q1612.58 1461.2 1610.76 1457.66 Q1608.95 1454.1 1605.32 1454.1 M1605.32 1450.39 Q1611.13 1450.39 1614.18 1455 Q1617.26 1459.58 1617.26 1468.33 Q1617.26 1477.06 1614.18 1481.67 Q1611.13 1486.25 1605.32 1486.25 Q1599.51 1486.25 1596.43 1481.67 Q1593.37 1477.06 1593.37 1468.33 Q1593.37 1459.58 1596.43 1455 Q1599.51 1450.39 1605.32 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1754.93 1481.64 L1762.57 1481.64 L1762.57 1455.28 L1754.26 1456.95 L1754.26 1452.69 L1762.52 1451.02 L1767.2 1451.02 L1767.2 1481.64 L1774.84 1481.64 L1774.84 1485.58 L1754.93 1485.58 L1754.93 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1794.28 1454.1 Q1790.67 1454.1 1788.84 1457.66 Q1787.04 1461.2 1787.04 1468.33 Q1787.04 1475.44 1788.84 1479.01 Q1790.67 1482.55 1794.28 1482.55 Q1797.92 1482.55 1799.72 1479.01 Q1801.55 1475.44 1801.55 1468.33 Q1801.55 1461.2 1799.72 1457.66 Q1797.92 1454.1 1794.28 1454.1 M1794.28 1450.39 Q1800.09 1450.39 1803.15 1455 Q1806.23 1459.58 1806.23 1468.33 Q1806.23 1477.06 1803.15 1481.67 Q1800.09 1486.25 1794.28 1486.25 Q1788.47 1486.25 1785.4 1481.67 Q1782.34 1477.06 1782.34 1468.33 Q1782.34 1459.58 1785.4 1455 Q1788.47 1450.39 1794.28 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1824.45 1454.1 Q1820.83 1454.1 1819.01 1457.66 Q1817.2 1461.2 1817.2 1468.33 Q1817.2 1475.44 1819.01 1479.01 Q1820.83 1482.55 1824.45 1482.55 Q1828.08 1482.55 1829.89 1479.01 Q1831.71 1475.44 1831.71 1468.33 Q1831.71 1461.2 1829.89 1457.66 Q1828.08 1454.1 1824.45 1454.1 M1824.45 1450.39 Q1830.26 1450.39 1833.31 1455 Q1836.39 1459.58 1836.39 1468.33 Q1836.39 1477.06 1833.31 1481.67 Q1830.26 1486.25 1824.45 1486.25 Q1818.64 1486.25 1815.56 1481.67 Q1812.5 1477.06 1812.5 1468.33 Q1812.5 1459.58 1815.56 1455 Q1818.64 1450.39 1824.45 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1854.61 1454.1 Q1851 1454.1 1849.17 1457.66 Q1847.36 1461.2 1847.36 1468.33 Q1847.36 1475.44 1849.17 1479.01 Q1851 1482.55 1854.61 1482.55 Q1858.24 1482.55 1860.05 1479.01 Q1861.88 1475.44 1861.88 1468.33 Q1861.88 1461.2 1860.05 1457.66 Q1858.24 1454.1 1854.61 1454.1 M1854.61 1450.39 Q1860.42 1450.39 1863.47 1455 Q1866.55 1459.58 1866.55 1468.33 Q1866.55 1477.06 1863.47 1481.67 Q1860.42 1486.25 1854.61 1486.25 Q1848.8 1486.25 1845.72 1481.67 Q1842.66 1477.06 1842.66 1468.33 Q1842.66 1459.58 1845.72 1455 Q1848.8 1450.39 1854.61 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1989.92 1481.64 L1997.56 1481.64 L1997.56 1455.28 L1989.25 1456.95 L1989.25 1452.69 L1997.51 1451.02 L2002.19 1451.02 L2002.19 1481.64 L2009.83 1481.64 L2009.83 1485.58 L1989.92 1485.58 L1989.92 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2019.32 1451.02 L2037.67 1451.02 L2037.67 1454.96 L2023.6 1454.96 L2023.6 1463.43 Q2024.62 1463.08 2025.64 1462.92 Q2026.65 1462.73 2027.67 1462.73 Q2033.46 1462.73 2036.84 1465.9 Q2040.22 1469.08 2040.22 1474.49 Q2040.22 1480.07 2036.75 1483.17 Q2033.28 1486.25 2026.96 1486.25 Q2024.78 1486.25 2022.51 1485.88 Q2020.27 1485.51 2017.86 1484.77 L2017.86 1480.07 Q2019.94 1481.2 2022.16 1481.76 Q2024.39 1482.32 2026.86 1482.32 Q2030.87 1482.32 2033.21 1480.21 Q2035.54 1478.1 2035.54 1474.49 Q2035.54 1470.88 2033.21 1468.77 Q2030.87 1466.67 2026.86 1466.67 Q2024.99 1466.67 2023.11 1467.08 Q2021.26 1467.5 2019.32 1468.38 L2019.32 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2059.43 1454.1 Q2055.82 1454.1 2053.99 1457.66 Q2052.19 1461.2 2052.19 1468.33 Q2052.19 1475.44 2053.99 1479.01 Q2055.82 1482.55 2059.43 1482.55 Q2063.07 1482.55 2064.87 1479.01 Q2066.7 1475.44 2066.7 1468.33 Q2066.7 1461.2 2064.87 1457.66 Q2063.07 1454.1 2059.43 1454.1 M2059.43 1450.39 Q2065.24 1450.39 2068.3 1455 Q2071.38 1459.58 2071.38 1468.33 Q2071.38 1477.06 2068.3 1481.67 Q2065.24 1486.25 2059.43 1486.25 Q2053.62 1486.25 2050.54 1481.67 Q2047.49 1477.06 2047.49 1468.33 Q2047.49 1459.58 2050.54 1455 Q2053.62 1450.39 2059.43 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2089.59 1454.1 Q2085.98 1454.1 2084.15 1457.66 Q2082.35 1461.2 2082.35 1468.33 Q2082.35 1475.44 2084.15 1479.01 Q2085.98 1482.55 2089.59 1482.55 Q2093.23 1482.55 2095.03 1479.01 Q2096.86 1475.44 2096.86 1468.33 Q2096.86 1461.2 2095.03 1457.66 Q2093.23 1454.1 2089.59 1454.1 M2089.59 1450.39 Q2095.4 1450.39 2098.46 1455 Q2101.54 1459.58 2101.54 1468.33 Q2101.54 1477.06 2098.46 1481.67 Q2095.4 1486.25 2089.59 1486.25 Q2083.78 1486.25 2080.71 1481.67 Q2077.65 1477.06 2077.65 1468.33 Q2077.65 1459.58 2080.71 1455 Q2083.78 1450.39 2089.59 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2228.99 1481.64 L2245.31 1481.64 L2245.31 1485.58 L2223.37 1485.58 L2223.37 1481.64 Q2226.03 1478.89 2230.61 1474.26 Q2235.22 1469.61 2236.4 1468.27 Q2238.64 1465.74 2239.52 1464.01 Q2240.43 1462.25 2240.43 1460.56 Q2240.43 1457.8 2238.48 1456.07 Q2236.56 1454.33 2233.46 1454.33 Q2231.26 1454.33 2228.81 1455.09 Q2226.38 1455.86 2223.6 1457.41 L2223.6 1452.69 Q2226.42 1451.55 2228.88 1450.97 Q2231.33 1450.39 2233.37 1450.39 Q2238.74 1450.39 2241.93 1453.08 Q2245.13 1455.77 2245.13 1460.26 Q2245.13 1462.39 2244.32 1464.31 Q2243.53 1466.2 2241.42 1468.8 Q2240.84 1469.47 2237.74 1472.69 Q2234.64 1475.88 2228.99 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2265.13 1454.1 Q2261.51 1454.1 2259.69 1457.66 Q2257.88 1461.2 2257.88 1468.33 Q2257.88 1475.44 2259.69 1479.01 Q2261.51 1482.55 2265.13 1482.55 Q2268.76 1482.55 2270.57 1479.01 Q2272.39 1475.44 2272.39 1468.33 Q2272.39 1461.2 2270.57 1457.66 Q2268.76 1454.1 2265.13 1454.1 M2265.13 1450.39 Q2270.94 1450.39 2273.99 1455 Q2277.07 1459.58 2277.07 1468.33 Q2277.07 1477.06 2273.99 1481.67 Q2270.94 1486.25 2265.13 1486.25 Q2259.32 1486.25 2256.24 1481.67 Q2253.18 1477.06 2253.18 1468.33 Q2253.18 1459.58 2256.24 1455 Q2259.32 1450.39 2265.13 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2295.29 1454.1 Q2291.68 1454.1 2289.85 1457.66 Q2288.04 1461.2 2288.04 1468.33 Q2288.04 1475.44 2289.85 1479.01 Q2291.68 1482.55 2295.29 1482.55 Q2298.92 1482.55 2300.73 1479.01 Q2302.56 1475.44 2302.56 1468.33 Q2302.56 1461.2 2300.73 1457.66 Q2298.92 1454.1 2295.29 1454.1 M2295.29 1450.39 Q2301.1 1450.39 2304.15 1455 Q2307.23 1459.58 2307.23 1468.33 Q2307.23 1477.06 2304.15 1481.67 Q2301.1 1486.25 2295.29 1486.25 Q2289.48 1486.25 2286.4 1481.67 Q2283.34 1477.06 2283.34 1468.33 Q2283.34 1459.58 2286.4 1455 Q2289.48 1450.39 2295.29 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2325.45 1454.1 Q2321.84 1454.1 2320.01 1457.66 Q2318.2 1461.2 2318.2 1468.33 Q2318.2 1475.44 2320.01 1479.01 Q2321.84 1482.55 2325.45 1482.55 Q2329.08 1482.55 2330.89 1479.01 Q2332.72 1475.44 2332.72 1468.33 Q2332.72 1461.2 2330.89 1457.66 Q2329.08 1454.1 2325.45 1454.1 M2325.45 1450.39 Q2331.26 1450.39 2334.31 1455 Q2337.39 1459.58 2337.39 1468.33 Q2337.39 1477.06 2334.31 1481.67 Q2331.26 1486.25 2325.45 1486.25 Q2319.64 1486.25 2316.56 1481.67 Q2313.5 1477.06 2313.5 1468.33 Q2313.5 1459.58 2316.56 1455 Q2319.64 1450.39 2325.45 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1778.77 1525.81 L1778.77 1562.76 L1786.54 1562.76 Q1796.38 1562.76 1800.93 1558.3 Q1805.51 1553.85 1805.51 1544.24 Q1805.51 1534.69 1800.93 1530.26 Q1796.38 1525.81 1786.54 1525.81 L1778.77 1525.81 M1772.34 1520.52 L1785.55 1520.52 Q1799.37 1520.52 1805.83 1526.28 Q1812.29 1532.01 1812.29 1544.24 Q1812.29 1556.52 1805.8 1562.28 Q1799.3 1568.04 1785.55 1568.04 L1772.34 1568.04 L1772.34 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1852.78 1548.76 L1852.78 1551.62 L1825.85 1551.62 Q1826.23 1557.67 1829.48 1560.85 Q1832.76 1564 1838.58 1564 Q1841.95 1564 1845.1 1563.17 Q1848.29 1562.35 1851.41 1560.69 L1851.41 1566.23 Q1848.26 1567.57 1844.95 1568.27 Q1841.64 1568.97 1838.23 1568.97 Q1829.7 1568.97 1824.7 1564 Q1819.74 1559.04 1819.74 1550.57 Q1819.74 1541.82 1824.45 1536.69 Q1829.19 1531.54 1837.21 1531.54 Q1844.4 1531.54 1848.57 1536.18 Q1852.78 1540.8 1852.78 1548.76 M1846.92 1547.04 Q1846.86 1542.23 1844.21 1539.37 Q1841.6 1536.5 1837.27 1536.5 Q1832.37 1536.5 1829.41 1539.27 Q1826.49 1542.04 1826.04 1547.07 L1846.92 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1890.14 1539.24 Q1892.34 1535.29 1895.39 1533.41 Q1898.45 1531.54 1902.59 1531.54 Q1908.16 1531.54 1911.18 1535.45 Q1914.2 1539.33 1914.2 1546.53 L1914.2 1568.04 L1908.32 1568.04 L1908.32 1546.72 Q1908.32 1541.59 1906.5 1539.11 Q1904.69 1536.63 1900.96 1536.63 Q1896.41 1536.63 1893.77 1539.65 Q1891.13 1542.68 1891.13 1547.9 L1891.13 1568.04 L1885.24 1568.04 L1885.24 1546.72 Q1885.24 1541.56 1883.43 1539.11 Q1881.61 1536.63 1877.82 1536.63 Q1873.34 1536.63 1870.69 1539.68 Q1868.05 1542.71 1868.05 1547.9 L1868.05 1568.04 L1862.16 1568.04 L1862.16 1532.4 L1868.05 1532.4 L1868.05 1537.93 Q1870.06 1534.66 1872.86 1533.1 Q1875.66 1531.54 1879.51 1531.54 Q1883.39 1531.54 1886.1 1533.51 Q1888.84 1535.48 1890.14 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1942.09 1550.12 Q1934.99 1550.12 1932.25 1551.75 Q1929.51 1553.37 1929.51 1557.29 Q1929.51 1560.4 1931.55 1562.25 Q1933.62 1564.07 1937.15 1564.07 Q1942.02 1564.07 1944.95 1560.63 Q1947.91 1557.16 1947.91 1551.43 L1947.91 1550.12 L1942.09 1550.12 M1953.77 1547.71 L1953.77 1568.04 L1947.91 1568.04 L1947.91 1562.63 Q1945.91 1565.88 1942.91 1567.44 Q1939.92 1568.97 1935.59 1568.97 Q1930.12 1568.97 1926.87 1565.91 Q1923.66 1562.82 1923.66 1557.67 Q1923.66 1551.65 1927.67 1548.6 Q1931.71 1545.54 1939.7 1545.54 L1947.91 1545.54 L1947.91 1544.97 Q1947.91 1540.93 1945.24 1538.73 Q1942.6 1536.5 1937.79 1536.5 Q1934.73 1536.5 1931.84 1537.23 Q1928.94 1537.97 1926.27 1539.43 L1926.27 1534.02 Q1929.48 1532.78 1932.51 1532.17 Q1935.53 1531.54 1938.39 1531.54 Q1946.13 1531.54 1949.95 1535.55 Q1953.77 1539.56 1953.77 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1995.46 1546.53 L1995.46 1568.04 L1989.61 1568.04 L1989.61 1546.72 Q1989.61 1541.66 1987.63 1539.14 Q1985.66 1536.63 1981.71 1536.63 Q1976.97 1536.63 1974.23 1539.65 Q1971.5 1542.68 1971.5 1547.9 L1971.5 1568.04 L1965.61 1568.04 L1965.61 1532.4 L1971.5 1532.4 L1971.5 1537.93 Q1973.6 1534.72 1976.43 1533.13 Q1979.29 1531.54 1983.02 1531.54 Q1989.16 1531.54 1992.31 1535.36 Q1995.46 1539.14 1995.46 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2030.6 1537.81 L2030.6 1518.52 L2036.46 1518.52 L2036.46 1568.04 L2030.6 1568.04 L2030.6 1562.7 Q2028.76 1565.88 2025.92 1567.44 Q2023.12 1568.97 2019.17 1568.97 Q2012.71 1568.97 2008.64 1563.81 Q2004.6 1558.65 2004.6 1550.25 Q2004.6 1541.85 2008.64 1536.69 Q2012.71 1531.54 2019.17 1531.54 Q2023.12 1531.54 2025.92 1533.1 Q2028.76 1534.62 2030.6 1537.81 M2010.64 1550.25 Q2010.64 1556.71 2013.29 1560.4 Q2015.96 1564.07 2020.61 1564.07 Q2025.25 1564.07 2027.93 1560.4 Q2030.6 1556.71 2030.6 1550.25 Q2030.6 1543.79 2027.93 1540.13 Q2025.25 1536.44 2020.61 1536.44 Q2015.96 1536.44 2013.29 1540.13 Q2010.64 1543.79 2010.64 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,1386.4 2352.76,1386.4 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,1079.86 2352.76,1079.86 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,773.326 2352.76,773.326 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,466.791 2352.76,466.791 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,160.256 2352.76,160.256 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1423.18 1456.05,123.472 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1386.4 1474.94,1386.4 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1079.86 1474.94,1079.86 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,773.326 1474.94,773.326 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,466.791 1474.94,466.791 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,160.256 1474.94,160.256 "/>
-<path clip-path="url(#clip710)" d="M1408.1 1372.19 Q1404.49 1372.19 1402.66 1375.76 Q1400.86 1379.3 1400.86 1386.43 Q1400.86 1393.54 1402.66 1397.1 Q1404.49 1400.64 1408.1 1400.64 Q1411.74 1400.64 1413.54 1397.1 Q1415.37 1393.54 1415.37 1386.43 Q1415.37 1379.3 1413.54 1375.76 Q1411.74 1372.19 1408.1 1372.19 M1408.1 1368.49 Q1413.91 1368.49 1416.97 1373.1 Q1420.05 1377.68 1420.05 1386.43 Q1420.05 1395.16 1416.97 1399.76 Q1413.91 1404.35 1408.1 1404.35 Q1402.29 1404.35 1399.21 1399.76 Q1396.16 1395.16 1396.16 1386.43 Q1396.16 1377.68 1399.21 1373.1 Q1402.29 1368.49 1408.1 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1341.81 1093.21 L1358.13 1093.21 L1358.13 1097.14 L1336.18 1097.14 L1336.18 1093.21 Q1338.84 1090.45 1343.43 1085.82 Q1348.03 1081.17 1349.21 1079.83 Q1351.46 1077.3 1352.34 1075.57 Q1353.24 1073.81 1353.24 1072.12 Q1353.24 1069.36 1351.3 1067.63 Q1349.38 1065.89 1346.27 1065.89 Q1344.07 1065.89 1341.62 1066.65 Q1339.19 1067.42 1336.41 1068.97 L1336.41 1064.25 Q1339.24 1063.11 1341.69 1062.53 Q1344.14 1061.96 1346.18 1061.96 Q1351.55 1061.96 1354.75 1064.64 Q1357.94 1067.33 1357.94 1071.82 Q1357.94 1073.95 1357.13 1075.87 Q1356.34 1077.77 1354.24 1080.36 Q1353.66 1081.03 1350.56 1084.25 Q1347.45 1087.44 1341.81 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1367.99 1062.58 L1386.34 1062.58 L1386.34 1066.52 L1372.27 1066.52 L1372.27 1074.99 Q1373.29 1074.64 1374.31 1074.48 Q1375.32 1074.29 1376.34 1074.29 Q1382.13 1074.29 1385.51 1077.47 Q1388.89 1080.64 1388.89 1086.05 Q1388.89 1091.63 1385.42 1094.73 Q1381.94 1097.81 1375.63 1097.81 Q1373.45 1097.81 1371.18 1097.44 Q1368.94 1097.07 1366.53 1096.33 L1366.53 1091.63 Q1368.61 1092.77 1370.83 1093.32 Q1373.06 1093.88 1375.53 1093.88 Q1379.54 1093.88 1381.88 1091.77 Q1384.21 1089.66 1384.21 1086.05 Q1384.21 1082.44 1381.88 1080.34 Q1379.54 1078.23 1375.53 1078.23 Q1373.66 1078.23 1371.78 1078.65 Q1369.93 1079.06 1367.99 1079.94 L1367.99 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1408.1 1065.66 Q1404.49 1065.66 1402.66 1069.22 Q1400.86 1072.77 1400.86 1079.9 Q1400.86 1087 1402.66 1090.57 Q1404.49 1094.11 1408.1 1094.11 Q1411.74 1094.11 1413.54 1090.57 Q1415.37 1087 1415.37 1079.9 Q1415.37 1072.77 1413.54 1069.22 Q1411.74 1065.66 1408.1 1065.66 M1408.1 1061.96 Q1413.91 1061.96 1416.97 1066.56 Q1420.05 1071.15 1420.05 1079.9 Q1420.05 1088.62 1416.97 1093.23 Q1413.91 1097.81 1408.1 1097.81 Q1402.29 1097.81 1399.21 1093.23 Q1396.16 1088.62 1396.16 1079.9 Q1396.16 1071.15 1399.21 1066.56 Q1402.29 1061.96 1408.1 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1337.82 756.046 L1356.18 756.046 L1356.18 759.981 L1342.11 759.981 L1342.11 768.453 Q1343.13 768.106 1344.14 767.944 Q1345.16 767.759 1346.18 767.759 Q1351.97 767.759 1355.35 770.93 Q1358.73 774.101 1358.73 779.518 Q1358.73 785.097 1355.26 788.199 Q1351.78 791.277 1345.46 791.277 Q1343.29 791.277 1341.02 790.907 Q1338.77 790.537 1336.37 789.796 L1336.37 785.097 Q1338.45 786.231 1340.67 786.787 Q1342.89 787.342 1345.37 787.342 Q1349.38 787.342 1351.71 785.236 Q1354.05 783.129 1354.05 779.518 Q1354.05 775.907 1351.71 773.801 Q1349.38 771.694 1345.37 771.694 Q1343.5 771.694 1341.62 772.111 Q1339.77 772.527 1337.82 773.407 L1337.82 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1377.94 759.125 Q1374.33 759.125 1372.5 762.689 Q1370.7 766.231 1370.7 773.361 Q1370.7 780.467 1372.5 784.032 Q1374.33 787.574 1377.94 787.574 Q1381.57 787.574 1383.38 784.032 Q1385.21 780.467 1385.21 773.361 Q1385.21 766.231 1383.38 762.689 Q1381.57 759.125 1377.94 759.125 M1377.94 755.421 Q1383.75 755.421 1386.81 760.027 Q1389.88 764.611 1389.88 773.361 Q1389.88 782.088 1386.81 786.694 Q1383.75 791.277 1377.94 791.277 Q1372.13 791.277 1369.05 786.694 Q1366 782.088 1366 773.361 Q1366 764.611 1369.05 760.027 Q1372.13 755.421 1377.94 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1408.1 759.125 Q1404.49 759.125 1402.66 762.689 Q1400.86 766.231 1400.86 773.361 Q1400.86 780.467 1402.66 784.032 Q1404.49 787.574 1408.1 787.574 Q1411.74 787.574 1413.54 784.032 Q1415.37 780.467 1415.37 773.361 Q1415.37 766.231 1413.54 762.689 Q1411.74 759.125 1408.1 759.125 M1408.1 755.421 Q1413.91 755.421 1416.97 760.027 Q1420.05 764.611 1420.05 773.361 Q1420.05 782.088 1416.97 786.694 Q1413.91 791.277 1408.1 791.277 Q1402.29 791.277 1399.21 786.694 Q1396.16 782.088 1396.16 773.361 Q1396.16 764.611 1399.21 760.027 Q1402.29 755.421 1408.1 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1336.6 449.511 L1358.82 449.511 L1358.82 451.502 L1346.27 484.071 L1341.39 484.071 L1353.2 453.446 L1336.6 453.446 L1336.6 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1367.99 449.511 L1386.34 449.511 L1386.34 453.446 L1372.27 453.446 L1372.27 461.918 Q1373.29 461.571 1374.31 461.409 Q1375.32 461.224 1376.34 461.224 Q1382.13 461.224 1385.51 464.395 Q1388.89 467.567 1388.89 472.983 Q1388.89 478.562 1385.42 481.664 Q1381.94 484.742 1375.63 484.742 Q1373.45 484.742 1371.18 484.372 Q1368.94 484.002 1366.53 483.261 L1366.53 478.562 Q1368.61 479.696 1370.83 480.252 Q1373.06 480.807 1375.53 480.807 Q1379.54 480.807 1381.88 478.701 Q1384.21 476.594 1384.21 472.983 Q1384.21 469.372 1381.88 467.266 Q1379.54 465.159 1375.53 465.159 Q1373.66 465.159 1371.78 465.576 Q1369.93 465.993 1367.99 466.872 L1367.99 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1408.1 452.59 Q1404.49 452.59 1402.66 456.155 Q1400.86 459.696 1400.86 466.826 Q1400.86 473.932 1402.66 477.497 Q1404.49 481.039 1408.1 481.039 Q1411.74 481.039 1413.54 477.497 Q1415.37 473.932 1415.37 466.826 Q1415.37 459.696 1413.54 456.155 Q1411.74 452.59 1408.1 452.59 M1408.1 448.886 Q1413.91 448.886 1416.97 453.493 Q1420.05 458.076 1420.05 466.826 Q1420.05 475.553 1416.97 480.159 Q1413.91 484.742 1408.1 484.742 Q1402.29 484.742 1399.21 480.159 Q1396.16 475.553 1396.16 466.826 Q1396.16 458.076 1399.21 453.493 Q1402.29 448.886 1408.1 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1308.43 173.601 L1316.07 173.601 L1316.07 147.236 L1307.76 148.902 L1307.76 144.643 L1316.02 142.976 L1320.7 142.976 L1320.7 173.601 L1328.33 173.601 L1328.33 177.536 L1308.43 177.536 L1308.43 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1347.78 146.055 Q1344.17 146.055 1342.34 149.62 Q1340.53 153.161 1340.53 160.291 Q1340.53 167.397 1342.34 170.962 Q1344.17 174.504 1347.78 174.504 Q1351.41 174.504 1353.22 170.962 Q1355.05 167.397 1355.05 160.291 Q1355.05 153.161 1353.22 149.62 Q1351.41 146.055 1347.78 146.055 M1347.78 142.351 Q1353.59 142.351 1356.64 146.958 Q1359.72 151.541 1359.72 160.291 Q1359.72 169.018 1356.64 173.624 Q1353.59 178.208 1347.78 178.208 Q1341.97 178.208 1338.89 173.624 Q1335.83 169.018 1335.83 160.291 Q1335.83 151.541 1338.89 146.958 Q1341.97 142.351 1347.78 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1377.94 146.055 Q1374.33 146.055 1372.5 149.62 Q1370.7 153.161 1370.7 160.291 Q1370.7 167.397 1372.5 170.962 Q1374.33 174.504 1377.94 174.504 Q1381.57 174.504 1383.38 170.962 Q1385.21 167.397 1385.21 160.291 Q1385.21 153.161 1383.38 149.62 Q1381.57 146.055 1377.94 146.055 M1377.94 142.351 Q1383.75 142.351 1386.81 146.958 Q1389.88 151.541 1389.88 160.291 Q1389.88 169.018 1386.81 173.624 Q1383.75 178.208 1377.94 178.208 Q1372.13 178.208 1369.05 173.624 Q1366 169.018 1366 160.291 Q1366 151.541 1369.05 146.958 Q1372.13 142.351 1377.94 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1408.1 146.055 Q1404.49 146.055 1402.66 149.62 Q1400.86 153.161 1400.86 160.291 Q1400.86 167.397 1402.66 170.962 Q1404.49 174.504 1408.1 174.504 Q1411.74 174.504 1413.54 170.962 Q1415.37 167.397 1415.37 160.291 Q1415.37 153.161 1413.54 149.62 Q1411.74 146.055 1408.1 146.055 M1408.1 142.351 Q1413.91 142.351 1416.97 146.958 Q1420.05 151.541 1420.05 160.291 Q1420.05 169.018 1416.97 173.624 Q1413.91 178.208 1408.1 178.208 Q1402.29 178.208 1399.21 173.624 Q1396.16 169.018 1396.16 160.291 Q1396.16 151.541 1399.21 146.958 Q1402.29 142.351 1408.1 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1215.26 1000.18 L1252.22 1000.18 L1252.22 992.418 Q1252.22 982.583 1247.76 978.031 Q1243.3 973.448 1233.69 973.448 Q1224.14 973.448 1219.72 978.031 Q1215.26 982.583 1215.26 992.418 L1215.26 1000.18 M1209.98 1006.61 L1209.98 993.404 Q1209.98 979.591 1215.74 973.13 Q1221.47 966.668 1233.69 966.668 Q1245.98 966.668 1251.74 973.161 Q1257.5 979.654 1257.5 993.404 L1257.5 1006.61 L1209.98 1006.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1221.85 956.674 L1221.85 950.818 L1257.5 950.818 L1257.5 956.674 L1221.85 956.674 M1207.97 956.674 L1207.97 950.818 L1215.39 950.818 L1215.39 956.674 L1207.97 956.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1222.9 915.838 L1228.44 915.838 Q1227.17 918.321 1226.53 920.994 Q1225.89 923.668 1225.89 926.533 Q1225.89 930.893 1227.23 933.089 Q1228.57 935.254 1231.24 935.254 Q1233.28 935.254 1234.46 933.694 Q1235.6 932.134 1236.65 927.424 L1237.1 925.419 Q1238.43 919.18 1240.89 916.57 Q1243.3 913.929 1247.66 913.929 Q1252.63 913.929 1255.53 917.875 Q1258.42 921.79 1258.42 928.665 Q1258.42 931.53 1257.85 934.649 Q1257.31 937.736 1256.19 941.174 L1250.15 941.174 Q1251.83 937.927 1252.69 934.776 Q1253.52 931.625 1253.52 928.538 Q1253.52 924.4 1252.12 922.172 Q1250.69 919.944 1248.11 919.944 Q1245.72 919.944 1244.45 921.567 Q1243.18 923.159 1242 928.601 L1241.52 930.639 Q1240.38 936.081 1238.02 938.5 Q1235.63 940.919 1231.5 940.919 Q1226.47 940.919 1223.73 937.354 Q1220.99 933.79 1220.99 927.233 Q1220.99 923.986 1221.47 921.122 Q1221.95 918.257 1222.9 915.838 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1252.15 898.937 L1271.06 898.937 L1271.06 904.826 L1221.85 904.826 L1221.85 898.937 L1227.26 898.937 Q1224.08 897.091 1222.55 894.29 Q1220.99 891.458 1220.99 887.543 Q1220.99 881.05 1226.15 877.007 Q1231.3 872.933 1239.71 872.933 Q1248.11 872.933 1253.27 877.007 Q1258.42 881.05 1258.42 887.543 Q1258.42 891.458 1256.89 894.29 Q1255.34 897.091 1252.15 898.937 M1239.71 879.013 Q1233.25 879.013 1229.59 881.686 Q1225.89 884.328 1225.89 888.975 Q1225.89 893.622 1229.59 896.296 Q1233.25 898.937 1239.71 898.937 Q1246.17 898.937 1249.86 896.296 Q1253.52 893.622 1253.52 888.975 Q1253.52 884.328 1249.86 881.686 Q1246.17 879.013 1239.71 879.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1239.58 847.025 Q1239.58 854.123 1241.2 856.86 Q1242.83 859.597 1246.74 859.597 Q1249.86 859.597 1251.71 857.56 Q1253.52 855.491 1253.52 851.958 Q1253.52 847.089 1250.08 844.16 Q1246.61 841.2 1240.89 841.2 L1239.58 841.2 L1239.58 847.025 M1237.16 835.344 L1257.5 835.344 L1257.5 841.2 L1252.09 841.2 Q1255.34 843.206 1256.89 846.197 Q1258.42 849.189 1258.42 853.518 Q1258.42 858.992 1255.37 862.239 Q1252.28 865.454 1247.12 865.454 Q1241.11 865.454 1238.05 861.443 Q1235 857.401 1235 849.412 L1235 841.2 L1234.42 841.2 Q1230.38 841.2 1228.19 843.874 Q1225.96 846.516 1225.96 851.322 Q1225.96 854.377 1226.69 857.274 Q1227.42 860.17 1228.89 862.844 L1223.47 862.844 Q1222.23 859.629 1221.63 856.605 Q1220.99 853.582 1220.99 850.717 Q1220.99 842.983 1225 839.163 Q1229.01 835.344 1237.16 835.344 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1211.73 817.488 L1221.85 817.488 L1221.85 805.425 L1226.4 805.425 L1226.4 817.488 L1245.75 817.488 Q1250.12 817.488 1251.36 816.31 Q1252.6 815.101 1252.6 811.441 L1252.6 805.425 L1257.5 805.425 L1257.5 811.441 Q1257.5 818.22 1254.99 820.798 Q1252.44 823.376 1245.75 823.376 L1226.4 823.376 L1226.4 827.673 L1221.85 827.673 L1221.85 823.376 L1211.73 823.376 L1211.73 817.488 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1223.22 772.069 L1228.69 772.069 Q1227.33 774.551 1226.66 777.066 Q1225.96 779.548 1225.96 782.095 Q1225.96 787.792 1229.59 790.943 Q1233.18 794.094 1239.71 794.094 Q1246.23 794.094 1249.86 790.943 Q1253.46 787.792 1253.46 782.095 Q1253.46 779.548 1252.79 777.066 Q1252.09 774.551 1250.72 772.069 L1256.13 772.069 Q1257.28 774.52 1257.85 777.161 Q1258.42 779.771 1258.42 782.731 Q1258.42 790.784 1253.36 795.526 Q1248.3 800.269 1239.71 800.269 Q1230.99 800.269 1225.99 795.495 Q1220.99 790.688 1220.99 782.349 Q1220.99 779.644 1221.57 777.066 Q1222.11 774.488 1223.22 772.069 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1235.98 732.251 L1257.5 732.251 L1257.5 738.108 L1236.17 738.108 Q1231.11 738.108 1228.6 740.081 Q1226.08 742.054 1226.08 746.001 Q1226.08 750.744 1229.11 753.481 Q1232.13 756.218 1237.35 756.218 L1257.5 756.218 L1257.5 762.106 L1207.97 762.106 L1207.97 756.218 L1227.39 756.218 Q1224.18 754.117 1222.58 751.285 Q1220.99 748.42 1220.99 744.696 Q1220.99 738.553 1224.81 735.402 Q1228.6 732.251 1235.98 732.251 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1207.97 700.391 L1207.97 686.896 L1212.53 686.896 L1212.53 694.534 L1261.54 694.534 L1261.54 686.896 L1266.09 686.896 L1266.09 700.391 L1207.97 700.391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1209.98 674.164 L1209.98 664.584 L1242.32 652.457 L1209.98 640.267 L1209.98 630.686 L1257.5 630.686 L1257.5 636.957 L1215.77 636.957 L1248.36 649.211 L1248.36 655.672 L1215.77 667.926 L1257.5 667.926 L1257.5 674.164 L1209.98 674.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1209.98 622.156 L1209.98 615.663 L1250.15 605.669 L1209.98 595.707 L1209.98 588.482 L1250.15 578.488 L1209.98 568.525 L1209.98 562 L1257.5 573.936 L1257.5 582.021 L1216.25 592.047 L1257.5 602.168 L1257.5 610.252 L1209.98 622.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1207.97 540.039 L1266.09 540.039 L1266.09 553.534 L1261.54 553.534 L1261.54 545.927 L1212.53 545.927 L1212.53 553.534 L1207.97 553.534 L1207.97 540.039 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1615.67 18.8205 L1615.67 65.8515 L1625.56 65.8515 Q1638.07 65.8515 1643.87 60.1802 Q1649.7 54.509 1649.7 42.2752 Q1649.7 30.1225 1643.87 24.4918 Q1638.07 18.8205 1625.56 18.8205 L1615.67 18.8205 M1607.49 12.096 L1624.3 12.096 Q1641.88 12.096 1650.11 19.4281 Q1658.33 26.7198 1658.33 42.2752 Q1658.33 57.9117 1650.07 65.2439 Q1641.8 72.576 1624.3 72.576 L1607.49 72.576 L1607.49 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1671.05 27.2059 L1678.5 27.2059 L1678.5 72.576 L1671.05 72.576 L1671.05 27.2059 M1671.05 9.54393 L1678.5 9.54393 L1678.5 18.9825 L1671.05 18.9825 L1671.05 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1723.02 28.5427 L1723.02 35.5912 Q1719.86 33.9709 1716.46 33.1607 Q1713.06 32.3505 1709.41 32.3505 Q1703.86 32.3505 1701.07 34.0519 Q1698.31 35.7533 1698.31 39.156 Q1698.31 41.7486 1700.3 43.2475 Q1702.28 44.7058 1708.28 46.0426 L1710.83 46.6097 Q1718.77 48.3111 1722.09 51.4303 Q1725.45 54.509 1725.45 60.0587 Q1725.45 66.3781 1720.43 70.0644 Q1715.45 73.7508 1706.7 73.7508 Q1703.05 73.7508 1699.08 73.0216 Q1695.15 72.3329 1690.78 70.9151 L1690.78 63.2184 Q1694.91 65.3654 1698.92 66.4591 Q1702.93 67.5124 1706.86 67.5124 Q1712.13 67.5124 1714.96 65.73 Q1717.8 63.9071 1717.8 60.6258 Q1717.8 57.5877 1715.73 55.9673 Q1713.7 54.3469 1706.78 52.8481 L1704.19 52.2405 Q1697.26 50.7821 1694.18 47.7845 Q1691.1 44.7463 1691.1 39.4801 Q1691.1 33.0797 1695.64 29.5959 Q1700.17 26.1121 1708.52 26.1121 Q1712.65 26.1121 1716.3 26.7198 Q1719.94 27.3274 1723.02 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1744.53 65.7705 L1744.53 89.8329 L1737.04 89.8329 L1737.04 27.2059 L1744.53 27.2059 L1744.53 34.0924 Q1746.88 30.0415 1750.45 28.0971 Q1754.05 26.1121 1759.03 26.1121 Q1767.3 26.1121 1772.44 32.6746 Q1777.63 39.2371 1777.63 49.9314 Q1777.63 60.6258 1772.44 67.1883 Q1767.3 73.7508 1759.03 73.7508 Q1754.05 73.7508 1750.45 71.8063 Q1746.88 69.8214 1744.53 65.7705 M1769.89 49.9314 Q1769.89 41.7081 1766.49 37.0496 Q1763.13 32.3505 1757.21 32.3505 Q1751.3 32.3505 1747.89 37.0496 Q1744.53 41.7081 1744.53 49.9314 Q1744.53 58.1548 1747.89 62.8538 Q1751.3 67.5124 1757.21 67.5124 Q1763.13 67.5124 1766.49 62.8538 Q1769.89 58.1548 1769.89 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1810.6 49.7694 Q1801.57 49.7694 1798.09 51.8354 Q1794.6 53.9013 1794.6 58.8839 Q1794.6 62.8538 1797.19 65.2034 Q1799.83 67.5124 1804.32 67.5124 Q1810.52 67.5124 1814.25 63.1374 Q1818.02 58.7219 1818.02 51.4303 L1818.02 49.7694 L1810.6 49.7694 M1825.47 46.6907 L1825.47 72.576 L1818.02 72.576 L1818.02 65.6895 Q1815.46 69.8214 1811.66 71.8063 Q1807.85 73.7508 1802.34 73.7508 Q1795.37 73.7508 1791.24 69.8619 Q1787.15 65.9325 1787.15 59.3701 Q1787.15 51.7138 1792.25 47.825 Q1797.4 43.9361 1807.56 43.9361 L1818.02 43.9361 L1818.02 43.2069 Q1818.02 38.0623 1814.61 35.2672 Q1811.25 32.4315 1805.13 32.4315 Q1801.24 32.4315 1797.56 33.3632 Q1793.87 34.295 1790.47 36.1584 L1790.47 29.2718 Q1794.56 27.692 1798.41 26.9223 Q1802.26 26.1121 1805.9 26.1121 Q1815.75 26.1121 1820.61 31.2163 Q1825.47 36.3204 1825.47 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1848.19 14.324 L1848.19 27.2059 L1863.55 27.2059 L1863.55 32.9987 L1848.19 32.9987 L1848.19 57.6282 Q1848.19 63.1779 1849.69 64.7578 Q1851.23 66.3376 1855.89 66.3376 L1863.55 66.3376 L1863.55 72.576 L1855.89 72.576 Q1847.26 72.576 1843.98 69.3758 Q1840.7 66.1351 1840.7 57.6282 L1840.7 32.9987 L1835.23 32.9987 L1835.23 27.2059 L1840.7 27.2059 L1840.7 14.324 L1848.19 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1906 28.9478 L1906 35.9153 Q1902.84 34.1734 1899.64 33.3227 Q1896.48 32.4315 1893.24 32.4315 Q1885.99 32.4315 1881.98 37.0496 Q1877.97 41.6271 1877.97 49.9314 Q1877.97 58.2358 1881.98 62.8538 Q1885.99 67.4314 1893.24 67.4314 Q1896.48 67.4314 1899.64 66.5807 Q1902.84 65.6895 1906 63.9476 L1906 70.8341 Q1902.88 72.2924 1899.52 73.0216 Q1896.2 73.7508 1892.43 73.7508 Q1882.18 73.7508 1876.15 67.3098 Q1870.11 60.8689 1870.11 49.9314 Q1870.11 38.832 1876.19 32.472 Q1882.3 26.1121 1892.92 26.1121 Q1896.36 26.1121 1899.64 26.8413 Q1902.92 27.5299 1906 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1956.68 45.1919 L1956.68 72.576 L1949.22 72.576 L1949.22 45.4349 Q1949.22 38.994 1946.71 35.7938 Q1944.2 32.5936 1939.18 32.5936 Q1933.14 32.5936 1929.66 36.4419 Q1926.17 40.2903 1926.17 46.9338 L1926.17 72.576 L1918.68 72.576 L1918.68 9.54393 L1926.17 9.54393 L1926.17 34.2544 Q1928.85 30.163 1932.45 28.1376 Q1936.1 26.1121 1940.84 26.1121 Q1948.66 26.1121 1952.67 30.9732 Q1956.68 35.7938 1956.68 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1997.23 9.54393 L2014.4 9.54393 L2014.4 15.3367 L2004.68 15.3367 L2004.68 77.7206 L2014.4 77.7206 L2014.4 83.5134 L1997.23 83.5134 L1997.23 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2030.61 12.096 L2042.8 12.096 L2058.23 53.2532 L2073.75 12.096 L2085.94 12.096 L2085.94 72.576 L2077.96 72.576 L2077.96 19.4686 L2062.37 60.9499 L2054.14 60.9499 L2038.55 19.4686 L2038.55 72.576 L2030.61 72.576 L2030.61 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2096.8 12.096 L2105.06 12.096 L2117.78 63.2184 L2130.46 12.096 L2139.66 12.096 L2152.38 63.2184 L2165.06 12.096 L2173.36 12.096 L2158.17 72.576 L2147.88 72.576 L2135.12 20.0763 L2122.24 72.576 L2111.95 72.576 L2096.8 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2201.31 9.54393 L2201.31 83.5134 L2184.14 83.5134 L2184.14 77.7206 L2193.82 77.7206 L2193.82 15.3367 L2184.14 15.3367 L2184.14 9.54393 L2201.31 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip713)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1263.78 1551.92,1079.86 1622.42,895.94 1692.91,712.019 1763.41,528.098 1833.91,344.177 1904.4,160.256 1974.9,344.177 2045.39,160.256 2115.89,160.256 2186.39,160.256 2256.88,160.256 2327.38,160.256 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1386.4 1551.92,1386.4 1622.42,1386.4 1692.91,1386.4 1763.41,1386.4 1833.91,1386.4 1904.4,1386.4 1974.9,1018.55 2045.39,1018.55 2115.89,834.633 2186.39,650.712 2256.88,466.791 2327.38,282.87 "/>
-<polyline clip-path="url(#clip713)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1141.17 1551.92,1141.17 1622.42,1141.17 1692.91,1141.17 1763.41,1141.17 1833.91,1141.17 1904.4,1141.17 1974.9,1141.17 2045.39,1141.17 2115.89,1141.17 2186.39,1141.17 2256.88,1141.17 2327.38,1141.17 "/>
-<path clip-path="url(#clip710)" d="M2119.93 1379.86 L2322.87 1379.86 L2322.87 1172.5 L2119.93 1172.5  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2119.93,1379.86 2322.87,1379.86 2322.87,1172.5 2119.93,1172.5 2119.93,1379.86 "/>
-<polyline clip-path="url(#clip710)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1224.34 2189.68,1224.34 "/>
-<path clip-path="url(#clip710)" d="M2225.19 1236.69 L2225.19 1227.4 L2217.56 1227.4 L2217.56 1223.56 L2229.82 1223.56 L2229.82 1238.4 Q2227.12 1240.32 2223.85 1241.32 Q2220.59 1242.29 2216.88 1242.29 Q2208.78 1242.29 2204.2 1237.57 Q2199.64 1232.82 2199.64 1224.37 Q2199.64 1215.9 2204.2 1211.18 Q2208.78 1206.43 2216.88 1206.43 Q2220.26 1206.43 2223.3 1207.26 Q2226.35 1208.1 2228.92 1209.72 L2228.92 1214.7 Q2226.33 1212.5 2223.41 1211.38 Q2220.5 1210.27 2217.28 1210.27 Q2210.94 1210.27 2207.74 1213.82 Q2204.57 1217.36 2204.57 1224.37 Q2204.57 1231.36 2207.74 1234.9 Q2210.94 1238.44 2217.28 1238.44 Q2219.75 1238.44 2221.7 1238.03 Q2223.64 1237.59 2225.19 1236.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2239.59 1237.68 L2247.23 1237.68 L2247.23 1211.32 L2238.92 1212.98 L2238.92 1208.72 L2247.18 1207.06 L2251.86 1207.06 L2251.86 1237.68 L2259.5 1237.68 L2259.5 1241.62 L2239.59 1241.62 L2239.59 1237.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip710)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1276.18 2189.68,1276.18 "/>
-<path clip-path="url(#clip710)" d="M2225.19 1288.53 L2225.19 1279.24 L2217.56 1279.24 L2217.56 1275.4 L2229.82 1275.4 L2229.82 1290.24 Q2227.12 1292.16 2223.85 1293.16 Q2220.59 1294.13 2216.88 1294.13 Q2208.78 1294.13 2204.2 1289.41 Q2199.64 1284.66 2199.64 1276.21 Q2199.64 1267.74 2204.2 1263.02 Q2208.78 1258.27 2216.88 1258.27 Q2220.26 1258.27 2223.3 1259.1 Q2226.35 1259.94 2228.92 1261.56 L2228.92 1266.54 Q2226.33 1264.34 2223.41 1263.22 Q2220.5 1262.11 2217.28 1262.11 Q2210.94 1262.11 2207.74 1265.66 Q2204.57 1269.2 2204.57 1276.21 Q2204.57 1283.2 2207.74 1286.74 Q2210.94 1290.28 2217.28 1290.28 Q2219.75 1290.28 2221.7 1289.87 Q2223.64 1289.43 2225.19 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2242.81 1289.52 L2259.13 1289.52 L2259.13 1293.46 L2237.18 1293.46 L2237.18 1289.52 Q2239.85 1286.77 2244.43 1282.14 Q2249.04 1277.48 2250.22 1276.14 Q2252.46 1273.62 2253.34 1271.88 Q2254.25 1270.12 2254.25 1268.43 Q2254.25 1265.68 2252.3 1263.94 Q2250.38 1262.21 2247.28 1262.21 Q2245.08 1262.21 2242.62 1262.97 Q2240.19 1263.73 2237.42 1265.29 L2237.42 1260.56 Q2240.24 1259.43 2242.69 1258.85 Q2245.15 1258.27 2247.18 1258.27 Q2252.56 1258.27 2255.75 1260.96 Q2258.94 1263.64 2258.94 1268.13 Q2258.94 1270.26 2258.13 1272.18 Q2257.35 1274.08 2255.24 1276.67 Q2254.66 1277.35 2251.56 1280.56 Q2248.46 1283.76 2242.81 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip710)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1328.02 2189.68,1328.02 "/>
-<path clip-path="url(#clip710)" d="M2199.64 1310.74 L2204.36 1310.74 L2211.63 1339.95 L2218.87 1310.74 L2224.13 1310.74 L2231.4 1339.95 L2238.64 1310.74 L2243.39 1310.74 L2234.71 1345.3 L2228.83 1345.3 L2221.54 1315.3 L2214.18 1345.3 L2208.3 1345.3 L2199.64 1310.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2248.37 1319.37 L2252.62 1319.37 L2252.62 1345.3 L2248.37 1345.3 L2248.37 1319.37 M2248.37 1309.28 L2252.62 1309.28 L2252.62 1314.67 L2248.37 1314.67 L2248.37 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2283.09 1329.65 L2283.09 1345.3 L2278.83 1345.3 L2278.83 1329.79 Q2278.83 1326.11 2277.39 1324.28 Q2275.96 1322.45 2273.09 1322.45 Q2269.64 1322.45 2267.65 1324.65 Q2265.66 1326.85 2265.66 1330.64 L2265.66 1345.3 L2261.37 1345.3 L2261.37 1319.37 L2265.66 1319.37 L2265.66 1323.4 Q2267.18 1321.06 2269.24 1319.9 Q2271.33 1318.75 2274.04 1318.75 Q2278.5 1318.75 2280.8 1321.52 Q2283.09 1324.28 2283.09 1329.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2308.64 1323.31 L2308.64 1309.28 L2312.9 1309.28 L2312.9 1345.3 L2308.64 1345.3 L2308.64 1341.41 Q2307.3 1343.72 2305.24 1344.86 Q2303.2 1345.97 2300.33 1345.97 Q2295.63 1345.97 2292.67 1342.22 Q2289.73 1338.47 2289.73 1332.36 Q2289.73 1326.25 2292.67 1322.5 Q2295.63 1318.75 2300.33 1318.75 Q2303.2 1318.75 2305.24 1319.88 Q2307.3 1320.99 2308.64 1323.31 M2294.13 1332.36 Q2294.13 1337.06 2296.05 1339.74 Q2297.99 1342.4 2301.37 1342.4 Q2304.75 1342.4 2306.7 1339.74 Q2308.64 1337.06 2308.64 1332.36 Q2308.64 1327.66 2306.7 1325 Q2304.75 1322.31 2301.37 1322.31 Q2297.99 1322.31 2296.05 1325 Q2294.13 1327.66 2294.13 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1575.42,1423.18 1575.42,123.472 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1810.41,1423.18 1810.41,123.472 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2045.39,1423.18 2045.39,123.472 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2280.38,1423.18 2280.38,123.472 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1575.42,1423.18 1575.42,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1810.41,1423.18 1810.41,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2045.39,1423.18 2045.39,1404.28 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2280.38,1423.18 2280.38,1404.28 "/>
+<path clip-path="url(#clip530)" d="M1535.04 1451.02 L1553.39 1451.02 L1553.39 1454.96 L1539.32 1454.96 L1539.32 1463.43 Q1540.34 1463.08 1541.36 1462.92 Q1542.38 1462.73 1543.39 1462.73 Q1549.18 1462.73 1552.56 1465.9 Q1555.94 1469.08 1555.94 1474.49 Q1555.94 1480.07 1552.47 1483.17 Q1549 1486.25 1542.68 1486.25 Q1540.5 1486.25 1538.23 1485.88 Q1535.99 1485.51 1533.58 1484.77 L1533.58 1480.07 Q1535.66 1481.2 1537.89 1481.76 Q1540.11 1482.32 1542.58 1482.32 Q1546.59 1482.32 1548.93 1480.21 Q1551.26 1478.1 1551.26 1474.49 Q1551.26 1470.88 1548.93 1468.77 Q1546.59 1466.67 1542.58 1466.67 Q1540.71 1466.67 1538.83 1467.08 Q1536.98 1467.5 1535.04 1468.38 L1535.04 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1575.15 1454.1 Q1571.54 1454.1 1569.71 1457.66 Q1567.91 1461.2 1567.91 1468.33 Q1567.91 1475.44 1569.71 1479.01 Q1571.54 1482.55 1575.15 1482.55 Q1578.79 1482.55 1580.59 1479.01 Q1582.42 1475.44 1582.42 1468.33 Q1582.42 1461.2 1580.59 1457.66 Q1578.79 1454.1 1575.15 1454.1 M1575.15 1450.39 Q1580.96 1450.39 1584.02 1455 Q1587.1 1459.58 1587.1 1468.33 Q1587.1 1477.06 1584.02 1481.67 Q1580.96 1486.25 1575.15 1486.25 Q1569.34 1486.25 1566.26 1481.67 Q1563.21 1477.06 1563.21 1468.33 Q1563.21 1459.58 1566.26 1455 Q1569.34 1450.39 1575.15 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1605.32 1454.1 Q1601.7 1454.1 1599.88 1457.66 Q1598.07 1461.2 1598.07 1468.33 Q1598.07 1475.44 1599.88 1479.01 Q1601.7 1482.55 1605.32 1482.55 Q1608.95 1482.55 1610.76 1479.01 Q1612.58 1475.44 1612.58 1468.33 Q1612.58 1461.2 1610.76 1457.66 Q1608.95 1454.1 1605.32 1454.1 M1605.32 1450.39 Q1611.13 1450.39 1614.18 1455 Q1617.26 1459.58 1617.26 1468.33 Q1617.26 1477.06 1614.18 1481.67 Q1611.13 1486.25 1605.32 1486.25 Q1599.51 1486.25 1596.43 1481.67 Q1593.37 1477.06 1593.37 1468.33 Q1593.37 1459.58 1596.43 1455 Q1599.51 1450.39 1605.32 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1754.93 1481.64 L1762.57 1481.64 L1762.57 1455.28 L1754.26 1456.95 L1754.26 1452.69 L1762.52 1451.02 L1767.2 1451.02 L1767.2 1481.64 L1774.84 1481.64 L1774.84 1485.58 L1754.93 1485.58 L1754.93 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1794.28 1454.1 Q1790.67 1454.1 1788.84 1457.66 Q1787.04 1461.2 1787.04 1468.33 Q1787.04 1475.44 1788.84 1479.01 Q1790.67 1482.55 1794.28 1482.55 Q1797.92 1482.55 1799.72 1479.01 Q1801.55 1475.44 1801.55 1468.33 Q1801.55 1461.2 1799.72 1457.66 Q1797.92 1454.1 1794.28 1454.1 M1794.28 1450.39 Q1800.09 1450.39 1803.15 1455 Q1806.23 1459.58 1806.23 1468.33 Q1806.23 1477.06 1803.15 1481.67 Q1800.09 1486.25 1794.28 1486.25 Q1788.47 1486.25 1785.4 1481.67 Q1782.34 1477.06 1782.34 1468.33 Q1782.34 1459.58 1785.4 1455 Q1788.47 1450.39 1794.28 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1824.45 1454.1 Q1820.83 1454.1 1819.01 1457.66 Q1817.2 1461.2 1817.2 1468.33 Q1817.2 1475.44 1819.01 1479.01 Q1820.83 1482.55 1824.45 1482.55 Q1828.08 1482.55 1829.89 1479.01 Q1831.71 1475.44 1831.71 1468.33 Q1831.71 1461.2 1829.89 1457.66 Q1828.08 1454.1 1824.45 1454.1 M1824.45 1450.39 Q1830.26 1450.39 1833.31 1455 Q1836.39 1459.58 1836.39 1468.33 Q1836.39 1477.06 1833.31 1481.67 Q1830.26 1486.25 1824.45 1486.25 Q1818.64 1486.25 1815.56 1481.67 Q1812.5 1477.06 1812.5 1468.33 Q1812.5 1459.58 1815.56 1455 Q1818.64 1450.39 1824.45 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1854.61 1454.1 Q1851 1454.1 1849.17 1457.66 Q1847.36 1461.2 1847.36 1468.33 Q1847.36 1475.44 1849.17 1479.01 Q1851 1482.55 1854.61 1482.55 Q1858.24 1482.55 1860.05 1479.01 Q1861.88 1475.44 1861.88 1468.33 Q1861.88 1461.2 1860.05 1457.66 Q1858.24 1454.1 1854.61 1454.1 M1854.61 1450.39 Q1860.42 1450.39 1863.47 1455 Q1866.55 1459.58 1866.55 1468.33 Q1866.55 1477.06 1863.47 1481.67 Q1860.42 1486.25 1854.61 1486.25 Q1848.8 1486.25 1845.72 1481.67 Q1842.66 1477.06 1842.66 1468.33 Q1842.66 1459.58 1845.72 1455 Q1848.8 1450.39 1854.61 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1989.92 1481.64 L1997.56 1481.64 L1997.56 1455.28 L1989.25 1456.95 L1989.25 1452.69 L1997.51 1451.02 L2002.19 1451.02 L2002.19 1481.64 L2009.83 1481.64 L2009.83 1485.58 L1989.92 1485.58 L1989.92 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2019.32 1451.02 L2037.67 1451.02 L2037.67 1454.96 L2023.6 1454.96 L2023.6 1463.43 Q2024.62 1463.08 2025.64 1462.92 Q2026.65 1462.73 2027.67 1462.73 Q2033.46 1462.73 2036.84 1465.9 Q2040.22 1469.08 2040.22 1474.49 Q2040.22 1480.07 2036.75 1483.17 Q2033.28 1486.25 2026.96 1486.25 Q2024.78 1486.25 2022.51 1485.88 Q2020.27 1485.51 2017.86 1484.77 L2017.86 1480.07 Q2019.94 1481.2 2022.16 1481.76 Q2024.39 1482.32 2026.86 1482.32 Q2030.87 1482.32 2033.21 1480.21 Q2035.54 1478.1 2035.54 1474.49 Q2035.54 1470.88 2033.21 1468.77 Q2030.87 1466.67 2026.86 1466.67 Q2024.99 1466.67 2023.11 1467.08 Q2021.26 1467.5 2019.32 1468.38 L2019.32 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2059.43 1454.1 Q2055.82 1454.1 2053.99 1457.66 Q2052.19 1461.2 2052.19 1468.33 Q2052.19 1475.44 2053.99 1479.01 Q2055.82 1482.55 2059.43 1482.55 Q2063.07 1482.55 2064.87 1479.01 Q2066.7 1475.44 2066.7 1468.33 Q2066.7 1461.2 2064.87 1457.66 Q2063.07 1454.1 2059.43 1454.1 M2059.43 1450.39 Q2065.24 1450.39 2068.3 1455 Q2071.38 1459.58 2071.38 1468.33 Q2071.38 1477.06 2068.3 1481.67 Q2065.24 1486.25 2059.43 1486.25 Q2053.62 1486.25 2050.54 1481.67 Q2047.49 1477.06 2047.49 1468.33 Q2047.49 1459.58 2050.54 1455 Q2053.62 1450.39 2059.43 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2089.59 1454.1 Q2085.98 1454.1 2084.15 1457.66 Q2082.35 1461.2 2082.35 1468.33 Q2082.35 1475.44 2084.15 1479.01 Q2085.98 1482.55 2089.59 1482.55 Q2093.23 1482.55 2095.03 1479.01 Q2096.86 1475.44 2096.86 1468.33 Q2096.86 1461.2 2095.03 1457.66 Q2093.23 1454.1 2089.59 1454.1 M2089.59 1450.39 Q2095.4 1450.39 2098.46 1455 Q2101.54 1459.58 2101.54 1468.33 Q2101.54 1477.06 2098.46 1481.67 Q2095.4 1486.25 2089.59 1486.25 Q2083.78 1486.25 2080.71 1481.67 Q2077.65 1477.06 2077.65 1468.33 Q2077.65 1459.58 2080.71 1455 Q2083.78 1450.39 2089.59 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2228.99 1481.64 L2245.31 1481.64 L2245.31 1485.58 L2223.37 1485.58 L2223.37 1481.64 Q2226.03 1478.89 2230.61 1474.26 Q2235.22 1469.61 2236.4 1468.27 Q2238.64 1465.74 2239.52 1464.01 Q2240.43 1462.25 2240.43 1460.56 Q2240.43 1457.8 2238.48 1456.07 Q2236.56 1454.33 2233.46 1454.33 Q2231.26 1454.33 2228.81 1455.09 Q2226.38 1455.86 2223.6 1457.41 L2223.6 1452.69 Q2226.42 1451.55 2228.88 1450.97 Q2231.33 1450.39 2233.37 1450.39 Q2238.74 1450.39 2241.93 1453.08 Q2245.13 1455.77 2245.13 1460.26 Q2245.13 1462.39 2244.32 1464.31 Q2243.53 1466.2 2241.42 1468.8 Q2240.84 1469.47 2237.74 1472.69 Q2234.64 1475.88 2228.99 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2265.13 1454.1 Q2261.51 1454.1 2259.69 1457.66 Q2257.88 1461.2 2257.88 1468.33 Q2257.88 1475.44 2259.69 1479.01 Q2261.51 1482.55 2265.13 1482.55 Q2268.76 1482.55 2270.57 1479.01 Q2272.39 1475.44 2272.39 1468.33 Q2272.39 1461.2 2270.57 1457.66 Q2268.76 1454.1 2265.13 1454.1 M2265.13 1450.39 Q2270.94 1450.39 2273.99 1455 Q2277.07 1459.58 2277.07 1468.33 Q2277.07 1477.06 2273.99 1481.67 Q2270.94 1486.25 2265.13 1486.25 Q2259.32 1486.25 2256.24 1481.67 Q2253.18 1477.06 2253.18 1468.33 Q2253.18 1459.58 2256.24 1455 Q2259.32 1450.39 2265.13 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2295.29 1454.1 Q2291.68 1454.1 2289.85 1457.66 Q2288.04 1461.2 2288.04 1468.33 Q2288.04 1475.44 2289.85 1479.01 Q2291.68 1482.55 2295.29 1482.55 Q2298.92 1482.55 2300.73 1479.01 Q2302.56 1475.44 2302.56 1468.33 Q2302.56 1461.2 2300.73 1457.66 Q2298.92 1454.1 2295.29 1454.1 M2295.29 1450.39 Q2301.1 1450.39 2304.15 1455 Q2307.23 1459.58 2307.23 1468.33 Q2307.23 1477.06 2304.15 1481.67 Q2301.1 1486.25 2295.29 1486.25 Q2289.48 1486.25 2286.4 1481.67 Q2283.34 1477.06 2283.34 1468.33 Q2283.34 1459.58 2286.4 1455 Q2289.48 1450.39 2295.29 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2325.45 1454.1 Q2321.84 1454.1 2320.01 1457.66 Q2318.2 1461.2 2318.2 1468.33 Q2318.2 1475.44 2320.01 1479.01 Q2321.84 1482.55 2325.45 1482.55 Q2329.08 1482.55 2330.89 1479.01 Q2332.72 1475.44 2332.72 1468.33 Q2332.72 1461.2 2330.89 1457.66 Q2329.08 1454.1 2325.45 1454.1 M2325.45 1450.39 Q2331.26 1450.39 2334.31 1455 Q2337.39 1459.58 2337.39 1468.33 Q2337.39 1477.06 2334.31 1481.67 Q2331.26 1486.25 2325.45 1486.25 Q2319.64 1486.25 2316.56 1481.67 Q2313.5 1477.06 2313.5 1468.33 Q2313.5 1459.58 2316.56 1455 Q2319.64 1450.39 2325.45 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1778.77 1525.81 L1778.77 1562.76 L1786.54 1562.76 Q1796.38 1562.76 1800.93 1558.3 Q1805.51 1553.85 1805.51 1544.24 Q1805.51 1534.69 1800.93 1530.26 Q1796.38 1525.81 1786.54 1525.81 L1778.77 1525.81 M1772.34 1520.52 L1785.55 1520.52 Q1799.37 1520.52 1805.83 1526.28 Q1812.29 1532.01 1812.29 1544.24 Q1812.29 1556.52 1805.8 1562.28 Q1799.3 1568.04 1785.55 1568.04 L1772.34 1568.04 L1772.34 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1852.78 1548.76 L1852.78 1551.62 L1825.85 1551.62 Q1826.23 1557.67 1829.48 1560.85 Q1832.76 1564 1838.58 1564 Q1841.95 1564 1845.1 1563.17 Q1848.29 1562.35 1851.41 1560.69 L1851.41 1566.23 Q1848.26 1567.57 1844.95 1568.27 Q1841.64 1568.97 1838.23 1568.97 Q1829.7 1568.97 1824.7 1564 Q1819.74 1559.04 1819.74 1550.57 Q1819.74 1541.82 1824.45 1536.69 Q1829.19 1531.54 1837.21 1531.54 Q1844.4 1531.54 1848.57 1536.18 Q1852.78 1540.8 1852.78 1548.76 M1846.92 1547.04 Q1846.86 1542.23 1844.21 1539.37 Q1841.6 1536.5 1837.27 1536.5 Q1832.37 1536.5 1829.41 1539.27 Q1826.49 1542.04 1826.04 1547.07 L1846.92 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1890.14 1539.24 Q1892.34 1535.29 1895.39 1533.41 Q1898.45 1531.54 1902.59 1531.54 Q1908.16 1531.54 1911.18 1535.45 Q1914.2 1539.33 1914.2 1546.53 L1914.2 1568.04 L1908.32 1568.04 L1908.32 1546.72 Q1908.32 1541.59 1906.5 1539.11 Q1904.69 1536.63 1900.96 1536.63 Q1896.41 1536.63 1893.77 1539.65 Q1891.13 1542.68 1891.13 1547.9 L1891.13 1568.04 L1885.24 1568.04 L1885.24 1546.72 Q1885.24 1541.56 1883.43 1539.11 Q1881.61 1536.63 1877.82 1536.63 Q1873.34 1536.63 1870.69 1539.68 Q1868.05 1542.71 1868.05 1547.9 L1868.05 1568.04 L1862.16 1568.04 L1862.16 1532.4 L1868.05 1532.4 L1868.05 1537.93 Q1870.06 1534.66 1872.86 1533.1 Q1875.66 1531.54 1879.51 1531.54 Q1883.39 1531.54 1886.1 1533.51 Q1888.84 1535.48 1890.14 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1942.09 1550.12 Q1934.99 1550.12 1932.25 1551.75 Q1929.51 1553.37 1929.51 1557.29 Q1929.51 1560.4 1931.55 1562.25 Q1933.62 1564.07 1937.15 1564.07 Q1942.02 1564.07 1944.95 1560.63 Q1947.91 1557.16 1947.91 1551.43 L1947.91 1550.12 L1942.09 1550.12 M1953.77 1547.71 L1953.77 1568.04 L1947.91 1568.04 L1947.91 1562.63 Q1945.91 1565.88 1942.91 1567.44 Q1939.92 1568.97 1935.59 1568.97 Q1930.12 1568.97 1926.87 1565.91 Q1923.66 1562.82 1923.66 1557.67 Q1923.66 1551.65 1927.67 1548.6 Q1931.71 1545.54 1939.7 1545.54 L1947.91 1545.54 L1947.91 1544.97 Q1947.91 1540.93 1945.24 1538.73 Q1942.6 1536.5 1937.79 1536.5 Q1934.73 1536.5 1931.84 1537.23 Q1928.94 1537.97 1926.27 1539.43 L1926.27 1534.02 Q1929.48 1532.78 1932.51 1532.17 Q1935.53 1531.54 1938.39 1531.54 Q1946.13 1531.54 1949.95 1535.55 Q1953.77 1539.56 1953.77 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1995.46 1546.53 L1995.46 1568.04 L1989.61 1568.04 L1989.61 1546.72 Q1989.61 1541.66 1987.63 1539.14 Q1985.66 1536.63 1981.71 1536.63 Q1976.97 1536.63 1974.23 1539.65 Q1971.5 1542.68 1971.5 1547.9 L1971.5 1568.04 L1965.61 1568.04 L1965.61 1532.4 L1971.5 1532.4 L1971.5 1537.93 Q1973.6 1534.72 1976.43 1533.13 Q1979.29 1531.54 1983.02 1531.54 Q1989.16 1531.54 1992.31 1535.36 Q1995.46 1539.14 1995.46 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2030.6 1537.81 L2030.6 1518.52 L2036.46 1518.52 L2036.46 1568.04 L2030.6 1568.04 L2030.6 1562.7 Q2028.76 1565.88 2025.92 1567.44 Q2023.12 1568.97 2019.17 1568.97 Q2012.71 1568.97 2008.64 1563.81 Q2004.6 1558.65 2004.6 1550.25 Q2004.6 1541.85 2008.64 1536.69 Q2012.71 1531.54 2019.17 1531.54 Q2023.12 1531.54 2025.92 1533.1 Q2028.76 1534.62 2030.6 1537.81 M2010.64 1550.25 Q2010.64 1556.71 2013.29 1560.4 Q2015.96 1564.07 2020.61 1564.07 Q2025.25 1564.07 2027.93 1560.4 Q2030.6 1556.71 2030.6 1550.25 Q2030.6 1543.79 2027.93 1540.13 Q2025.25 1536.44 2020.61 1536.44 Q2015.96 1536.44 2013.29 1540.13 Q2010.64 1543.79 2010.64 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,1386.4 2352.76,1386.4 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,1079.86 2352.76,1079.86 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,773.326 2352.76,773.326 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,466.791 2352.76,466.791 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1456.05,160.256 2352.76,160.256 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1423.18 1456.05,123.472 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1386.4 1474.94,1386.4 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,1079.86 1474.94,1079.86 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,773.326 1474.94,773.326 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,466.791 1474.94,466.791 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1456.05,160.256 1474.94,160.256 "/>
+<path clip-path="url(#clip530)" d="M1408.1 1372.19 Q1404.49 1372.19 1402.66 1375.76 Q1400.86 1379.3 1400.86 1386.43 Q1400.86 1393.54 1402.66 1397.1 Q1404.49 1400.64 1408.1 1400.64 Q1411.74 1400.64 1413.54 1397.1 Q1415.37 1393.54 1415.37 1386.43 Q1415.37 1379.3 1413.54 1375.76 Q1411.74 1372.19 1408.1 1372.19 M1408.1 1368.49 Q1413.91 1368.49 1416.97 1373.1 Q1420.05 1377.68 1420.05 1386.43 Q1420.05 1395.16 1416.97 1399.76 Q1413.91 1404.35 1408.1 1404.35 Q1402.29 1404.35 1399.21 1399.76 Q1396.16 1395.16 1396.16 1386.43 Q1396.16 1377.68 1399.21 1373.1 Q1402.29 1368.49 1408.1 1368.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1341.81 1093.21 L1358.13 1093.21 L1358.13 1097.14 L1336.18 1097.14 L1336.18 1093.21 Q1338.84 1090.45 1343.43 1085.82 Q1348.03 1081.17 1349.21 1079.83 Q1351.46 1077.3 1352.34 1075.57 Q1353.24 1073.81 1353.24 1072.12 Q1353.24 1069.36 1351.3 1067.63 Q1349.38 1065.89 1346.27 1065.89 Q1344.07 1065.89 1341.62 1066.65 Q1339.19 1067.42 1336.41 1068.97 L1336.41 1064.25 Q1339.24 1063.11 1341.69 1062.53 Q1344.14 1061.96 1346.18 1061.96 Q1351.55 1061.96 1354.75 1064.64 Q1357.94 1067.33 1357.94 1071.82 Q1357.94 1073.95 1357.13 1075.87 Q1356.34 1077.77 1354.24 1080.36 Q1353.66 1081.03 1350.56 1084.25 Q1347.45 1087.44 1341.81 1093.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1367.99 1062.58 L1386.34 1062.58 L1386.34 1066.52 L1372.27 1066.52 L1372.27 1074.99 Q1373.29 1074.64 1374.31 1074.48 Q1375.32 1074.29 1376.34 1074.29 Q1382.13 1074.29 1385.51 1077.47 Q1388.89 1080.64 1388.89 1086.05 Q1388.89 1091.63 1385.42 1094.73 Q1381.94 1097.81 1375.63 1097.81 Q1373.45 1097.81 1371.18 1097.44 Q1368.94 1097.07 1366.53 1096.33 L1366.53 1091.63 Q1368.61 1092.77 1370.83 1093.32 Q1373.06 1093.88 1375.53 1093.88 Q1379.54 1093.88 1381.88 1091.77 Q1384.21 1089.66 1384.21 1086.05 Q1384.21 1082.44 1381.88 1080.34 Q1379.54 1078.23 1375.53 1078.23 Q1373.66 1078.23 1371.78 1078.65 Q1369.93 1079.06 1367.99 1079.94 L1367.99 1062.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1408.1 1065.66 Q1404.49 1065.66 1402.66 1069.22 Q1400.86 1072.77 1400.86 1079.9 Q1400.86 1087 1402.66 1090.57 Q1404.49 1094.11 1408.1 1094.11 Q1411.74 1094.11 1413.54 1090.57 Q1415.37 1087 1415.37 1079.9 Q1415.37 1072.77 1413.54 1069.22 Q1411.74 1065.66 1408.1 1065.66 M1408.1 1061.96 Q1413.91 1061.96 1416.97 1066.56 Q1420.05 1071.15 1420.05 1079.9 Q1420.05 1088.62 1416.97 1093.23 Q1413.91 1097.81 1408.1 1097.81 Q1402.29 1097.81 1399.21 1093.23 Q1396.16 1088.62 1396.16 1079.9 Q1396.16 1071.15 1399.21 1066.56 Q1402.29 1061.96 1408.1 1061.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1337.82 756.046 L1356.18 756.046 L1356.18 759.981 L1342.11 759.981 L1342.11 768.453 Q1343.13 768.106 1344.14 767.944 Q1345.16 767.759 1346.18 767.759 Q1351.97 767.759 1355.35 770.93 Q1358.73 774.101 1358.73 779.518 Q1358.73 785.097 1355.26 788.199 Q1351.78 791.277 1345.46 791.277 Q1343.29 791.277 1341.02 790.907 Q1338.77 790.537 1336.37 789.796 L1336.37 785.097 Q1338.45 786.231 1340.67 786.787 Q1342.89 787.342 1345.37 787.342 Q1349.38 787.342 1351.71 785.236 Q1354.05 783.129 1354.05 779.518 Q1354.05 775.907 1351.71 773.801 Q1349.38 771.694 1345.37 771.694 Q1343.5 771.694 1341.62 772.111 Q1339.77 772.527 1337.82 773.407 L1337.82 756.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1377.94 759.125 Q1374.33 759.125 1372.5 762.689 Q1370.7 766.231 1370.7 773.361 Q1370.7 780.467 1372.5 784.032 Q1374.33 787.574 1377.94 787.574 Q1381.57 787.574 1383.38 784.032 Q1385.21 780.467 1385.21 773.361 Q1385.21 766.231 1383.38 762.689 Q1381.57 759.125 1377.94 759.125 M1377.94 755.421 Q1383.75 755.421 1386.81 760.027 Q1389.88 764.611 1389.88 773.361 Q1389.88 782.088 1386.81 786.694 Q1383.75 791.277 1377.94 791.277 Q1372.13 791.277 1369.05 786.694 Q1366 782.088 1366 773.361 Q1366 764.611 1369.05 760.027 Q1372.13 755.421 1377.94 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1408.1 759.125 Q1404.49 759.125 1402.66 762.689 Q1400.86 766.231 1400.86 773.361 Q1400.86 780.467 1402.66 784.032 Q1404.49 787.574 1408.1 787.574 Q1411.74 787.574 1413.54 784.032 Q1415.37 780.467 1415.37 773.361 Q1415.37 766.231 1413.54 762.689 Q1411.74 759.125 1408.1 759.125 M1408.1 755.421 Q1413.91 755.421 1416.97 760.027 Q1420.05 764.611 1420.05 773.361 Q1420.05 782.088 1416.97 786.694 Q1413.91 791.277 1408.1 791.277 Q1402.29 791.277 1399.21 786.694 Q1396.16 782.088 1396.16 773.361 Q1396.16 764.611 1399.21 760.027 Q1402.29 755.421 1408.1 755.421 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1336.6 449.511 L1358.82 449.511 L1358.82 451.502 L1346.27 484.071 L1341.39 484.071 L1353.2 453.446 L1336.6 453.446 L1336.6 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1367.99 449.511 L1386.34 449.511 L1386.34 453.446 L1372.27 453.446 L1372.27 461.918 Q1373.29 461.571 1374.31 461.409 Q1375.32 461.224 1376.34 461.224 Q1382.13 461.224 1385.51 464.395 Q1388.89 467.567 1388.89 472.983 Q1388.89 478.562 1385.42 481.664 Q1381.94 484.742 1375.63 484.742 Q1373.45 484.742 1371.18 484.372 Q1368.94 484.002 1366.53 483.261 L1366.53 478.562 Q1368.61 479.696 1370.83 480.252 Q1373.06 480.807 1375.53 480.807 Q1379.54 480.807 1381.88 478.701 Q1384.21 476.594 1384.21 472.983 Q1384.21 469.372 1381.88 467.266 Q1379.54 465.159 1375.53 465.159 Q1373.66 465.159 1371.78 465.576 Q1369.93 465.993 1367.99 466.872 L1367.99 449.511 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1408.1 452.59 Q1404.49 452.59 1402.66 456.155 Q1400.86 459.696 1400.86 466.826 Q1400.86 473.932 1402.66 477.497 Q1404.49 481.039 1408.1 481.039 Q1411.74 481.039 1413.54 477.497 Q1415.37 473.932 1415.37 466.826 Q1415.37 459.696 1413.54 456.155 Q1411.74 452.59 1408.1 452.59 M1408.1 448.886 Q1413.91 448.886 1416.97 453.493 Q1420.05 458.076 1420.05 466.826 Q1420.05 475.553 1416.97 480.159 Q1413.91 484.742 1408.1 484.742 Q1402.29 484.742 1399.21 480.159 Q1396.16 475.553 1396.16 466.826 Q1396.16 458.076 1399.21 453.493 Q1402.29 448.886 1408.1 448.886 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1308.43 173.601 L1316.07 173.601 L1316.07 147.236 L1307.76 148.902 L1307.76 144.643 L1316.02 142.976 L1320.7 142.976 L1320.7 173.601 L1328.33 173.601 L1328.33 177.536 L1308.43 177.536 L1308.43 173.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1347.78 146.055 Q1344.17 146.055 1342.34 149.62 Q1340.53 153.161 1340.53 160.291 Q1340.53 167.397 1342.34 170.962 Q1344.17 174.504 1347.78 174.504 Q1351.41 174.504 1353.22 170.962 Q1355.05 167.397 1355.05 160.291 Q1355.05 153.161 1353.22 149.62 Q1351.41 146.055 1347.78 146.055 M1347.78 142.351 Q1353.59 142.351 1356.64 146.958 Q1359.72 151.541 1359.72 160.291 Q1359.72 169.018 1356.64 173.624 Q1353.59 178.208 1347.78 178.208 Q1341.97 178.208 1338.89 173.624 Q1335.83 169.018 1335.83 160.291 Q1335.83 151.541 1338.89 146.958 Q1341.97 142.351 1347.78 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1377.94 146.055 Q1374.33 146.055 1372.5 149.62 Q1370.7 153.161 1370.7 160.291 Q1370.7 167.397 1372.5 170.962 Q1374.33 174.504 1377.94 174.504 Q1381.57 174.504 1383.38 170.962 Q1385.21 167.397 1385.21 160.291 Q1385.21 153.161 1383.38 149.62 Q1381.57 146.055 1377.94 146.055 M1377.94 142.351 Q1383.75 142.351 1386.81 146.958 Q1389.88 151.541 1389.88 160.291 Q1389.88 169.018 1386.81 173.624 Q1383.75 178.208 1377.94 178.208 Q1372.13 178.208 1369.05 173.624 Q1366 169.018 1366 160.291 Q1366 151.541 1369.05 146.958 Q1372.13 142.351 1377.94 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1408.1 146.055 Q1404.49 146.055 1402.66 149.62 Q1400.86 153.161 1400.86 160.291 Q1400.86 167.397 1402.66 170.962 Q1404.49 174.504 1408.1 174.504 Q1411.74 174.504 1413.54 170.962 Q1415.37 167.397 1415.37 160.291 Q1415.37 153.161 1413.54 149.62 Q1411.74 146.055 1408.1 146.055 M1408.1 142.351 Q1413.91 142.351 1416.97 146.958 Q1420.05 151.541 1420.05 160.291 Q1420.05 169.018 1416.97 173.624 Q1413.91 178.208 1408.1 178.208 Q1402.29 178.208 1399.21 173.624 Q1396.16 169.018 1396.16 160.291 Q1396.16 151.541 1399.21 146.958 Q1402.29 142.351 1408.1 142.351 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1215.26 1000.18 L1252.22 1000.18 L1252.22 992.418 Q1252.22 982.583 1247.76 978.031 Q1243.3 973.448 1233.69 973.448 Q1224.14 973.448 1219.72 978.031 Q1215.26 982.583 1215.26 992.418 L1215.26 1000.18 M1209.98 1006.61 L1209.98 993.404 Q1209.98 979.591 1215.74 973.13 Q1221.47 966.668 1233.69 966.668 Q1245.98 966.668 1251.74 973.161 Q1257.5 979.654 1257.5 993.404 L1257.5 1006.61 L1209.98 1006.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1221.85 956.674 L1221.85 950.818 L1257.5 950.818 L1257.5 956.674 L1221.85 956.674 M1207.97 956.674 L1207.97 950.818 L1215.39 950.818 L1215.39 956.674 L1207.97 956.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1222.9 915.838 L1228.44 915.838 Q1227.17 918.321 1226.53 920.994 Q1225.89 923.668 1225.89 926.533 Q1225.89 930.893 1227.23 933.089 Q1228.57 935.254 1231.24 935.254 Q1233.28 935.254 1234.46 933.694 Q1235.6 932.134 1236.65 927.424 L1237.1 925.419 Q1238.43 919.18 1240.89 916.57 Q1243.3 913.929 1247.66 913.929 Q1252.63 913.929 1255.53 917.875 Q1258.42 921.79 1258.42 928.665 Q1258.42 931.53 1257.85 934.649 Q1257.31 937.736 1256.19 941.174 L1250.15 941.174 Q1251.83 937.927 1252.69 934.776 Q1253.52 931.625 1253.52 928.538 Q1253.52 924.4 1252.12 922.172 Q1250.69 919.944 1248.11 919.944 Q1245.72 919.944 1244.45 921.567 Q1243.18 923.159 1242 928.601 L1241.52 930.639 Q1240.38 936.081 1238.02 938.5 Q1235.63 940.919 1231.5 940.919 Q1226.47 940.919 1223.73 937.354 Q1220.99 933.79 1220.99 927.233 Q1220.99 923.986 1221.47 921.122 Q1221.95 918.257 1222.9 915.838 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1252.15 898.937 L1271.06 898.937 L1271.06 904.826 L1221.85 904.826 L1221.85 898.937 L1227.26 898.937 Q1224.08 897.091 1222.55 894.29 Q1220.99 891.458 1220.99 887.543 Q1220.99 881.05 1226.15 877.007 Q1231.3 872.933 1239.71 872.933 Q1248.11 872.933 1253.27 877.007 Q1258.42 881.05 1258.42 887.543 Q1258.42 891.458 1256.89 894.29 Q1255.34 897.091 1252.15 898.937 M1239.71 879.013 Q1233.25 879.013 1229.59 881.686 Q1225.89 884.328 1225.89 888.975 Q1225.89 893.622 1229.59 896.296 Q1233.25 898.937 1239.71 898.937 Q1246.17 898.937 1249.86 896.296 Q1253.52 893.622 1253.52 888.975 Q1253.52 884.328 1249.86 881.686 Q1246.17 879.013 1239.71 879.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1239.58 847.025 Q1239.58 854.123 1241.2 856.86 Q1242.83 859.597 1246.74 859.597 Q1249.86 859.597 1251.71 857.56 Q1253.52 855.491 1253.52 851.958 Q1253.52 847.089 1250.08 844.16 Q1246.61 841.2 1240.89 841.2 L1239.58 841.2 L1239.58 847.025 M1237.16 835.344 L1257.5 835.344 L1257.5 841.2 L1252.09 841.2 Q1255.34 843.206 1256.89 846.197 Q1258.42 849.189 1258.42 853.518 Q1258.42 858.992 1255.37 862.239 Q1252.28 865.454 1247.12 865.454 Q1241.11 865.454 1238.05 861.443 Q1235 857.401 1235 849.412 L1235 841.2 L1234.42 841.2 Q1230.38 841.2 1228.19 843.874 Q1225.96 846.516 1225.96 851.322 Q1225.96 854.377 1226.69 857.274 Q1227.42 860.17 1228.89 862.844 L1223.47 862.844 Q1222.23 859.629 1221.63 856.605 Q1220.99 853.582 1220.99 850.717 Q1220.99 842.983 1225 839.163 Q1229.01 835.344 1237.16 835.344 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1211.73 817.488 L1221.85 817.488 L1221.85 805.425 L1226.4 805.425 L1226.4 817.488 L1245.75 817.488 Q1250.12 817.488 1251.36 816.31 Q1252.6 815.101 1252.6 811.441 L1252.6 805.425 L1257.5 805.425 L1257.5 811.441 Q1257.5 818.22 1254.99 820.798 Q1252.44 823.376 1245.75 823.376 L1226.4 823.376 L1226.4 827.673 L1221.85 827.673 L1221.85 823.376 L1211.73 823.376 L1211.73 817.488 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1223.22 772.069 L1228.69 772.069 Q1227.33 774.551 1226.66 777.066 Q1225.96 779.548 1225.96 782.095 Q1225.96 787.792 1229.59 790.943 Q1233.18 794.094 1239.71 794.094 Q1246.23 794.094 1249.86 790.943 Q1253.46 787.792 1253.46 782.095 Q1253.46 779.548 1252.79 777.066 Q1252.09 774.551 1250.72 772.069 L1256.13 772.069 Q1257.28 774.52 1257.85 777.161 Q1258.42 779.771 1258.42 782.731 Q1258.42 790.784 1253.36 795.526 Q1248.3 800.269 1239.71 800.269 Q1230.99 800.269 1225.99 795.495 Q1220.99 790.688 1220.99 782.349 Q1220.99 779.644 1221.57 777.066 Q1222.11 774.488 1223.22 772.069 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1235.98 732.251 L1257.5 732.251 L1257.5 738.108 L1236.17 738.108 Q1231.11 738.108 1228.6 740.081 Q1226.08 742.054 1226.08 746.001 Q1226.08 750.744 1229.11 753.481 Q1232.13 756.218 1237.35 756.218 L1257.5 756.218 L1257.5 762.106 L1207.97 762.106 L1207.97 756.218 L1227.39 756.218 Q1224.18 754.117 1222.58 751.285 Q1220.99 748.42 1220.99 744.696 Q1220.99 738.553 1224.81 735.402 Q1228.6 732.251 1235.98 732.251 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1207.97 700.391 L1207.97 686.896 L1212.53 686.896 L1212.53 694.534 L1261.54 694.534 L1261.54 686.896 L1266.09 686.896 L1266.09 700.391 L1207.97 700.391 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1209.98 674.164 L1209.98 664.584 L1242.32 652.457 L1209.98 640.267 L1209.98 630.686 L1257.5 630.686 L1257.5 636.957 L1215.77 636.957 L1248.36 649.211 L1248.36 655.672 L1215.77 667.926 L1257.5 667.926 L1257.5 674.164 L1209.98 674.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1209.98 622.156 L1209.98 615.663 L1250.15 605.669 L1209.98 595.707 L1209.98 588.482 L1250.15 578.488 L1209.98 568.525 L1209.98 562 L1257.5 573.936 L1257.5 582.021 L1216.25 592.047 L1257.5 602.168 L1257.5 610.252 L1209.98 622.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1207.97 540.039 L1266.09 540.039 L1266.09 553.534 L1261.54 553.534 L1261.54 545.927 L1212.53 545.927 L1212.53 553.534 L1207.97 553.534 L1207.97 540.039 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1615.67 18.8205 L1615.67 65.8515 L1625.56 65.8515 Q1638.07 65.8515 1643.87 60.1802 Q1649.7 54.509 1649.7 42.2752 Q1649.7 30.1225 1643.87 24.4918 Q1638.07 18.8205 1625.56 18.8205 L1615.67 18.8205 M1607.49 12.096 L1624.3 12.096 Q1641.88 12.096 1650.11 19.4281 Q1658.33 26.7198 1658.33 42.2752 Q1658.33 57.9117 1650.07 65.2439 Q1641.8 72.576 1624.3 72.576 L1607.49 72.576 L1607.49 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1671.05 27.2059 L1678.5 27.2059 L1678.5 72.576 L1671.05 72.576 L1671.05 27.2059 M1671.05 9.54393 L1678.5 9.54393 L1678.5 18.9825 L1671.05 18.9825 L1671.05 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1723.02 28.5427 L1723.02 35.5912 Q1719.86 33.9709 1716.46 33.1607 Q1713.06 32.3505 1709.41 32.3505 Q1703.86 32.3505 1701.07 34.0519 Q1698.31 35.7533 1698.31 39.156 Q1698.31 41.7486 1700.3 43.2475 Q1702.28 44.7058 1708.28 46.0426 L1710.83 46.6097 Q1718.77 48.3111 1722.09 51.4303 Q1725.45 54.509 1725.45 60.0587 Q1725.45 66.3781 1720.43 70.0644 Q1715.45 73.7508 1706.7 73.7508 Q1703.05 73.7508 1699.08 73.0216 Q1695.15 72.3329 1690.78 70.9151 L1690.78 63.2184 Q1694.91 65.3654 1698.92 66.4591 Q1702.93 67.5124 1706.86 67.5124 Q1712.13 67.5124 1714.96 65.73 Q1717.8 63.9071 1717.8 60.6258 Q1717.8 57.5877 1715.73 55.9673 Q1713.7 54.3469 1706.78 52.8481 L1704.19 52.2405 Q1697.26 50.7821 1694.18 47.7845 Q1691.1 44.7463 1691.1 39.4801 Q1691.1 33.0797 1695.64 29.5959 Q1700.17 26.1121 1708.52 26.1121 Q1712.65 26.1121 1716.3 26.7198 Q1719.94 27.3274 1723.02 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1744.53 65.7705 L1744.53 89.8329 L1737.04 89.8329 L1737.04 27.2059 L1744.53 27.2059 L1744.53 34.0924 Q1746.88 30.0415 1750.45 28.0971 Q1754.05 26.1121 1759.03 26.1121 Q1767.3 26.1121 1772.44 32.6746 Q1777.63 39.2371 1777.63 49.9314 Q1777.63 60.6258 1772.44 67.1883 Q1767.3 73.7508 1759.03 73.7508 Q1754.05 73.7508 1750.45 71.8063 Q1746.88 69.8214 1744.53 65.7705 M1769.89 49.9314 Q1769.89 41.7081 1766.49 37.0496 Q1763.13 32.3505 1757.21 32.3505 Q1751.3 32.3505 1747.89 37.0496 Q1744.53 41.7081 1744.53 49.9314 Q1744.53 58.1548 1747.89 62.8538 Q1751.3 67.5124 1757.21 67.5124 Q1763.13 67.5124 1766.49 62.8538 Q1769.89 58.1548 1769.89 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1810.6 49.7694 Q1801.57 49.7694 1798.09 51.8354 Q1794.6 53.9013 1794.6 58.8839 Q1794.6 62.8538 1797.19 65.2034 Q1799.83 67.5124 1804.32 67.5124 Q1810.52 67.5124 1814.25 63.1374 Q1818.02 58.7219 1818.02 51.4303 L1818.02 49.7694 L1810.6 49.7694 M1825.47 46.6907 L1825.47 72.576 L1818.02 72.576 L1818.02 65.6895 Q1815.46 69.8214 1811.66 71.8063 Q1807.85 73.7508 1802.34 73.7508 Q1795.37 73.7508 1791.24 69.8619 Q1787.15 65.9325 1787.15 59.3701 Q1787.15 51.7138 1792.25 47.825 Q1797.4 43.9361 1807.56 43.9361 L1818.02 43.9361 L1818.02 43.2069 Q1818.02 38.0623 1814.61 35.2672 Q1811.25 32.4315 1805.13 32.4315 Q1801.24 32.4315 1797.56 33.3632 Q1793.87 34.295 1790.47 36.1584 L1790.47 29.2718 Q1794.56 27.692 1798.41 26.9223 Q1802.26 26.1121 1805.9 26.1121 Q1815.75 26.1121 1820.61 31.2163 Q1825.47 36.3204 1825.47 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1848.19 14.324 L1848.19 27.2059 L1863.55 27.2059 L1863.55 32.9987 L1848.19 32.9987 L1848.19 57.6282 Q1848.19 63.1779 1849.69 64.7578 Q1851.23 66.3376 1855.89 66.3376 L1863.55 66.3376 L1863.55 72.576 L1855.89 72.576 Q1847.26 72.576 1843.98 69.3758 Q1840.7 66.1351 1840.7 57.6282 L1840.7 32.9987 L1835.23 32.9987 L1835.23 27.2059 L1840.7 27.2059 L1840.7 14.324 L1848.19 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1906 28.9478 L1906 35.9153 Q1902.84 34.1734 1899.64 33.3227 Q1896.48 32.4315 1893.24 32.4315 Q1885.99 32.4315 1881.98 37.0496 Q1877.97 41.6271 1877.97 49.9314 Q1877.97 58.2358 1881.98 62.8538 Q1885.99 67.4314 1893.24 67.4314 Q1896.48 67.4314 1899.64 66.5807 Q1902.84 65.6895 1906 63.9476 L1906 70.8341 Q1902.88 72.2924 1899.52 73.0216 Q1896.2 73.7508 1892.43 73.7508 Q1882.18 73.7508 1876.15 67.3098 Q1870.11 60.8689 1870.11 49.9314 Q1870.11 38.832 1876.19 32.472 Q1882.3 26.1121 1892.92 26.1121 Q1896.36 26.1121 1899.64 26.8413 Q1902.92 27.5299 1906 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1956.68 45.1919 L1956.68 72.576 L1949.22 72.576 L1949.22 45.4349 Q1949.22 38.994 1946.71 35.7938 Q1944.2 32.5936 1939.18 32.5936 Q1933.14 32.5936 1929.66 36.4419 Q1926.17 40.2903 1926.17 46.9338 L1926.17 72.576 L1918.68 72.576 L1918.68 9.54393 L1926.17 9.54393 L1926.17 34.2544 Q1928.85 30.163 1932.45 28.1376 Q1936.1 26.1121 1940.84 26.1121 Q1948.66 26.1121 1952.67 30.9732 Q1956.68 35.7938 1956.68 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M1997.23 9.54393 L2014.4 9.54393 L2014.4 15.3367 L2004.68 15.3367 L2004.68 77.7206 L2014.4 77.7206 L2014.4 83.5134 L1997.23 83.5134 L1997.23 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2030.61 12.096 L2042.8 12.096 L2058.23 53.2532 L2073.75 12.096 L2085.94 12.096 L2085.94 72.576 L2077.96 72.576 L2077.96 19.4686 L2062.37 60.9499 L2054.14 60.9499 L2038.55 19.4686 L2038.55 72.576 L2030.61 72.576 L2030.61 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2096.8 12.096 L2105.06 12.096 L2117.78 63.2184 L2130.46 12.096 L2139.66 12.096 L2152.38 63.2184 L2165.06 12.096 L2173.36 12.096 L2158.17 72.576 L2147.88 72.576 L2135.12 20.0763 L2122.24 72.576 L2111.95 72.576 L2096.8 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2201.31 9.54393 L2201.31 83.5134 L2184.14 83.5134 L2184.14 77.7206 L2193.82 77.7206 L2193.82 15.3367 L2184.14 15.3367 L2184.14 9.54393 L2201.31 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip533)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1263.78 1551.92,1079.86 1622.42,895.94 1692.91,712.019 1763.41,528.098 1833.91,344.177 1904.4,160.256 1974.9,344.177 2045.39,160.256 2115.89,160.256 2186.39,160.256 2256.88,160.256 2327.38,160.256 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1386.4 1551.92,1386.4 1622.42,1386.4 1692.91,1386.4 1763.41,1386.4 1833.91,1386.4 1904.4,1386.4 1974.9,1018.55 2045.39,1018.55 2115.89,834.633 2186.39,650.712 2256.88,466.791 2327.38,282.87 "/>
+<polyline clip-path="url(#clip533)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1481.43,1141.17 1551.92,1141.17 1622.42,1141.17 1692.91,1141.17 1763.41,1141.17 1833.91,1141.17 1904.4,1141.17 1974.9,1141.17 2045.39,1141.17 2115.89,1141.17 2186.39,1141.17 2256.88,1141.17 2327.38,1141.17 "/>
+<path clip-path="url(#clip530)" d="M2119.93 1379.86 L2322.87 1379.86 L2322.87 1172.5 L2119.93 1172.5  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip530)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2119.93,1379.86 2322.87,1379.86 2322.87,1172.5 2119.93,1172.5 2119.93,1379.86 "/>
+<polyline clip-path="url(#clip530)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1224.34 2189.68,1224.34 "/>
+<path clip-path="url(#clip530)" d="M2225.19 1236.69 L2225.19 1227.4 L2217.56 1227.4 L2217.56 1223.56 L2229.82 1223.56 L2229.82 1238.4 Q2227.12 1240.32 2223.85 1241.32 Q2220.59 1242.29 2216.88 1242.29 Q2208.78 1242.29 2204.2 1237.57 Q2199.64 1232.82 2199.64 1224.37 Q2199.64 1215.9 2204.2 1211.18 Q2208.78 1206.43 2216.88 1206.43 Q2220.26 1206.43 2223.3 1207.26 Q2226.35 1208.1 2228.92 1209.72 L2228.92 1214.7 Q2226.33 1212.5 2223.41 1211.38 Q2220.5 1210.27 2217.28 1210.27 Q2210.94 1210.27 2207.74 1213.82 Q2204.57 1217.36 2204.57 1224.37 Q2204.57 1231.36 2207.74 1234.9 Q2210.94 1238.44 2217.28 1238.44 Q2219.75 1238.44 2221.7 1238.03 Q2223.64 1237.59 2225.19 1236.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2239.59 1237.68 L2247.23 1237.68 L2247.23 1211.32 L2238.92 1212.98 L2238.92 1208.72 L2247.18 1207.06 L2251.86 1207.06 L2251.86 1237.68 L2259.5 1237.68 L2259.5 1241.62 L2239.59 1241.62 L2239.59 1237.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip530)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1276.18 2189.68,1276.18 "/>
+<path clip-path="url(#clip530)" d="M2225.19 1288.53 L2225.19 1279.24 L2217.56 1279.24 L2217.56 1275.4 L2229.82 1275.4 L2229.82 1290.24 Q2227.12 1292.16 2223.85 1293.16 Q2220.59 1294.13 2216.88 1294.13 Q2208.78 1294.13 2204.2 1289.41 Q2199.64 1284.66 2199.64 1276.21 Q2199.64 1267.74 2204.2 1263.02 Q2208.78 1258.27 2216.88 1258.27 Q2220.26 1258.27 2223.3 1259.1 Q2226.35 1259.94 2228.92 1261.56 L2228.92 1266.54 Q2226.33 1264.34 2223.41 1263.22 Q2220.5 1262.11 2217.28 1262.11 Q2210.94 1262.11 2207.74 1265.66 Q2204.57 1269.2 2204.57 1276.21 Q2204.57 1283.2 2207.74 1286.74 Q2210.94 1290.28 2217.28 1290.28 Q2219.75 1290.28 2221.7 1289.87 Q2223.64 1289.43 2225.19 1288.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2242.81 1289.52 L2259.13 1289.52 L2259.13 1293.46 L2237.18 1293.46 L2237.18 1289.52 Q2239.85 1286.77 2244.43 1282.14 Q2249.04 1277.48 2250.22 1276.14 Q2252.46 1273.62 2253.34 1271.88 Q2254.25 1270.12 2254.25 1268.43 Q2254.25 1265.68 2252.3 1263.94 Q2250.38 1262.21 2247.28 1262.21 Q2245.08 1262.21 2242.62 1262.97 Q2240.19 1263.73 2237.42 1265.29 L2237.42 1260.56 Q2240.24 1259.43 2242.69 1258.85 Q2245.15 1258.27 2247.18 1258.27 Q2252.56 1258.27 2255.75 1260.96 Q2258.94 1263.64 2258.94 1268.13 Q2258.94 1270.26 2258.13 1272.18 Q2257.35 1274.08 2255.24 1276.67 Q2254.66 1277.35 2251.56 1280.56 Q2248.46 1283.76 2242.81 1289.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip530)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="2129.89,1328.02 2189.68,1328.02 "/>
+<path clip-path="url(#clip530)" d="M2199.64 1310.74 L2204.36 1310.74 L2211.63 1339.95 L2218.87 1310.74 L2224.13 1310.74 L2231.4 1339.95 L2238.64 1310.74 L2243.39 1310.74 L2234.71 1345.3 L2228.83 1345.3 L2221.54 1315.3 L2214.18 1345.3 L2208.3 1345.3 L2199.64 1310.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2248.37 1319.37 L2252.62 1319.37 L2252.62 1345.3 L2248.37 1345.3 L2248.37 1319.37 M2248.37 1309.28 L2252.62 1309.28 L2252.62 1314.67 L2248.37 1314.67 L2248.37 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2283.09 1329.65 L2283.09 1345.3 L2278.83 1345.3 L2278.83 1329.79 Q2278.83 1326.11 2277.39 1324.28 Q2275.96 1322.45 2273.09 1322.45 Q2269.64 1322.45 2267.65 1324.65 Q2265.66 1326.85 2265.66 1330.64 L2265.66 1345.3 L2261.37 1345.3 L2261.37 1319.37 L2265.66 1319.37 L2265.66 1323.4 Q2267.18 1321.06 2269.24 1319.9 Q2271.33 1318.75 2274.04 1318.75 Q2278.5 1318.75 2280.8 1321.52 Q2283.09 1324.28 2283.09 1329.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip530)" d="M2308.64 1323.31 L2308.64 1309.28 L2312.9 1309.28 L2312.9 1345.3 L2308.64 1345.3 L2308.64 1341.41 Q2307.3 1343.72 2305.24 1344.86 Q2303.2 1345.97 2300.33 1345.97 Q2295.63 1345.97 2292.67 1342.22 Q2289.73 1338.47 2289.73 1332.36 Q2289.73 1326.25 2292.67 1322.5 Q2295.63 1318.75 2300.33 1318.75 Q2303.2 1318.75 2305.24 1319.88 Q2307.3 1320.99 2308.64 1323.31 M2294.13 1332.36 Q2294.13 1337.06 2296.05 1339.74 Q2297.99 1342.4 2301.37 1342.4 Q2304.75 1342.4 2306.7 1339.74 Q2308.64 1337.06 2308.64 1332.36 Q2308.64 1327.66 2306.7 1325 Q2304.75 1322.31 2301.37 1322.31 Q2297.99 1322.31 2296.05 1325 Q2294.13 1327.66 2294.13 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/power_systems/index.html b/previews/PR3562/tutorials/applications/power_systems/index.html
index 39a33495986..dd45bf69e5e 100644
--- a/previews/PR3562/tutorials/applications/power_systems/index.html
+++ b/previews/PR3562/tutorials/applications/power_systems/index.html
@@ -97,7 +97,7 @@
         (c_g1_scale, sol.g[1], sol.g[2], sol.w, sol.wind_spill, sol.total_cost),
     )
 end
-print(string(&quot;elapsed time: &quot;, time() - start, &quot; seconds&quot;))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">elapsed time: 0.11210799217224121 seconds</code></pre><pre><code class="language-julia hljs">c_g_scale_df</code></pre><div><div style = "float: left;"><span>26×6 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">scale</th><th style = "text-align: left;">dispatch_G1</th><th style = "text-align: left;">dispatch_G2</th><th style = "text-align: left;">dispatch_wind</th><th style = "text-align: left;">spillage_wind</th><th style = "text-align: left;">total_cost</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">65000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">0.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">70000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">0.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">75000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">0.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">80000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">85000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">90000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">1.1</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">1.2</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">100000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">1.3</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">105000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">1.4</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">110000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">115000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">1.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">120000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">1.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">125000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">1.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">130000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">1.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">135000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">140000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: right;">2.1</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">141500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: right;">2.2</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">143000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: right;">2.3</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">144500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: right;">2.4</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">146000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: right;">2.5</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">147500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: right;">2.6</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">149000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: right;">2.7</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">150500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">24</td><td style = "text-align: right;">2.8</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">152000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">25</td><td style = "text-align: right;">2.9</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">153500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">26</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">155000.0</td></tr></tbody></table></div><h2 id="Modifying-the-JuMP-model-in-place"><a class="docs-heading-anchor" href="#Modifying-the-JuMP-model-in-place">Modifying the JuMP model in-place</a><a id="Modifying-the-JuMP-model-in-place-1"></a><a class="docs-heading-anchor-permalink" href="#Modifying-the-JuMP-model-in-place" title="Permalink"></a></h2><p>Note that in the previous exercise we entirely rebuilt the optimization model at every iteration of the internal loop, which incurs an additional computational burden. This burden can be alleviated if instead of re-building the entire model, we modify the constraints or objective function, as it shown in the example below.</p><p>Compare the computing time in case of the above and below models.</p><pre><code class="language-julia hljs">function solve_economic_dispatch_inplace(
+print(string(&quot;elapsed time: &quot;, time() - start, &quot; seconds&quot;))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">elapsed time: 0.1521461009979248 seconds</code></pre><pre><code class="language-julia hljs">c_g_scale_df</code></pre><div><div style = "float: left;"><span>26×6 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">scale</th><th style = "text-align: left;">dispatch_G1</th><th style = "text-align: left;">dispatch_G2</th><th style = "text-align: left;">dispatch_wind</th><th style = "text-align: left;">spillage_wind</th><th style = "text-align: left;">total_cost</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">65000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">0.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">70000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">0.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">75000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">0.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">80000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">85000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">90000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">1.1</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">1.2</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">100000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">1.3</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">105000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">1.4</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">110000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">115000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">1.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">120000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">1.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">125000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">1.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">130000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">1.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">135000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">140000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: right;">2.1</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">141500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: right;">2.2</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">143000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: right;">2.3</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">144500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: right;">2.4</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">146000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: right;">2.5</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">147500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: right;">2.6</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">149000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: right;">2.7</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">150500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">24</td><td style = "text-align: right;">2.8</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">152000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">25</td><td style = "text-align: right;">2.9</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">153500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">26</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">155000.0</td></tr></tbody></table></div><h2 id="Modifying-the-JuMP-model-in-place"><a class="docs-heading-anchor" href="#Modifying-the-JuMP-model-in-place">Modifying the JuMP model in-place</a><a id="Modifying-the-JuMP-model-in-place-1"></a><a class="docs-heading-anchor-permalink" href="#Modifying-the-JuMP-model-in-place" title="Permalink"></a></h2><p>Note that in the previous exercise we entirely rebuilt the optimization model at every iteration of the internal loop, which incurs an additional computational burden. This burden can be alleviated if instead of re-building the entire model, we modify the constraints or objective function, as it shown in the example below.</p><p>Compare the computing time in case of the above and below models.</p><pre><code class="language-julia hljs">function solve_economic_dispatch_inplace(
     generators::Vector,
     wind,
     scenario,
@@ -153,7 +153,7 @@
     scenario,
     0.5:0.1:3.0,
 )
-print(string(&quot;elapsed time: &quot;, time() - start, &quot; seconds&quot;))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">elapsed time: 0.13160490989685059 seconds</code></pre><p>For small models, adjusting specific constraints or the objective function is sometimes faster and sometimes slower than re-building the entire model. However, as the problem size increases, updating the model in-place is usually faster.</p><pre><code class="language-julia hljs">inplace_df</code></pre><div><div style = "float: left;"><span>26×6 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">scale</th><th style = "text-align: left;">dispatch_G1</th><th style = "text-align: left;">dispatch_G2</th><th style = "text-align: left;">dispatch_wind</th><th style = "text-align: left;">spillage_wind</th><th style = "text-align: left;">total_cost</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">65000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">0.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">70000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">0.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">75000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">0.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">80000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">85000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">90000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">1.1</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">1.2</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">100000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">1.3</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">105000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">1.4</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">110000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">115000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">1.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">120000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">1.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">125000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">1.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">130000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">1.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">135000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">140000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: right;">2.1</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">141500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: right;">2.2</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">143000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: right;">2.3</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">144500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: right;">2.4</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">146000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: right;">2.5</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">147500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: right;">2.6</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">149000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: right;">2.7</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">150500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">24</td><td style = "text-align: right;">2.8</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">152000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">25</td><td style = "text-align: right;">2.9</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">153500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">26</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">155000.0</td></tr></tbody></table></div><h2 id="Inefficient-usage-of-wind-generators"><a class="docs-heading-anchor" href="#Inefficient-usage-of-wind-generators">Inefficient usage of wind generators</a><a id="Inefficient-usage-of-wind-generators-1"></a><a class="docs-heading-anchor-permalink" href="#Inefficient-usage-of-wind-generators" title="Permalink"></a></h2><p>The economic dispatch problem does not perform commitment decisions and, thus, assumes that all generators must be dispatched at least at their minimum power output limit. This approach is not cost efficient and may lead to absurd decisions. For example, if <span>$d = \sum_{i \in I} g^{\min}_{i}$</span>, the wind power injection must be zero, that is, all available wind generation is spilled, to meet the minimum power output constraints on generators.</p><p>In the following example, we adjust the total demand and observed how it affects wind spillage.</p><pre><code class="language-julia hljs">demand_scale_df = DataFrames.DataFrame(;
+print(string(&quot;elapsed time: &quot;, time() - start, &quot; seconds&quot;))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">elapsed time: 0.17378902435302734 seconds</code></pre><p>For small models, adjusting specific constraints or the objective function is sometimes faster and sometimes slower than re-building the entire model. However, as the problem size increases, updating the model in-place is usually faster.</p><pre><code class="language-julia hljs">inplace_df</code></pre><div><div style = "float: left;"><span>26×6 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">scale</th><th style = "text-align: left;">dispatch_G1</th><th style = "text-align: left;">dispatch_G2</th><th style = "text-align: left;">dispatch_wind</th><th style = "text-align: left;">spillage_wind</th><th style = "text-align: left;">total_cost</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">65000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: right;">0.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">70000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: right;">0.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">75000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: right;">0.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">80000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: right;">0.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">85000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">90000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: right;">1.1</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: right;">1.2</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">100000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: right;">1.3</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">105000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: right;">1.4</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">110000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">115000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: right;">1.6</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">120000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: right;">1.7</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">125000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: right;">1.8</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">130000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: right;">1.9</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">135000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">140000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: right;">2.1</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">141500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: right;">2.2</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">143000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: right;">2.3</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">144500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: right;">2.4</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">146000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: right;">2.5</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">147500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: right;">2.6</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">149000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: right;">2.7</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">150500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">24</td><td style = "text-align: right;">2.8</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">152000.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">25</td><td style = "text-align: right;">2.9</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">153500.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">26</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">300.0</td><td style = "text-align: right;">1000.0</td><td style = "text-align: right;">200.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">155000.0</td></tr></tbody></table></div><h2 id="Inefficient-usage-of-wind-generators"><a class="docs-heading-anchor" href="#Inefficient-usage-of-wind-generators">Inefficient usage of wind generators</a><a id="Inefficient-usage-of-wind-generators-1"></a><a class="docs-heading-anchor-permalink" href="#Inefficient-usage-of-wind-generators" title="Permalink"></a></h2><p>The economic dispatch problem does not perform commitment decisions and, thus, assumes that all generators must be dispatched at least at their minimum power output limit. This approach is not cost efficient and may lead to absurd decisions. For example, if <span>$d = \sum_{i \in I} g^{\min}_{i}$</span>, the wind power injection must be zero, that is, all available wind generation is spilled, to meet the minimum power output constraints on generators.</p><p>In the following example, we adjust the total demand and observed how it affects wind spillage.</p><pre><code class="language-julia hljs">demand_scale_df = DataFrames.DataFrame(;
     demand = Float64[],
     dispatch_G1 = Float64[],
     dispatch_G2 = Float64[],
@@ -210,7 +210,7 @@
     ),
 )
 
-Plots.plot(dispatch_plot, wind_plot)</code></pre><img src="472a82a6.svg" alt="Example block output"/><p>This particular drawback can be overcome by introducing binary decisions on the &quot;on/off&quot; status of generators. This model is called unit commitment and considered later in these notes.</p><p>For further reading on the interplay between wind generation and the minimum power output constraints of generators, we refer interested readers to R. Baldick, &quot;Wind and energy markets: a case study of Texas,&quot; IEEE Systems Journal, vol. 6, pp. 27-34, 2012.</p><h2 id="Unit-commitment"><a class="docs-heading-anchor" href="#Unit-commitment">Unit commitment</a><a id="Unit-commitment-1"></a><a class="docs-heading-anchor-permalink" href="#Unit-commitment" title="Permalink"></a></h2><p>The Unit Commitment (UC) model can be obtained from ED model by introducing binary variable associated with each generator. This binary variable can attain two values: if it is &quot;1,&quot; the generator is synchronized and, thus, can be dispatched, otherwise, that is, if the binary variable is &quot;0,&quot; that generator is not synchronized and its power output is set to 0.</p><p>To obtain the mathematical formulation of the UC model, we will modify the constraints of the ED model as follows:</p><p class="math-container">\[g^{\min}_{i} \cdot u_{t,i} \leq g_{i} \leq g^{\max}_{i} \cdot u_{t,i},\]</p><p>where <span>$u_{i} \in \{0,1\}.$</span> In this constraint, if <span>$u_{i} = 0$</span>, then <span>$g_{i}  = 0$</span>. On the other hand, if <span>$u_{i} = 1$</span>, then <span>$g^{min}_{i} \leq g_{i} \leq g^{max}_{i}$</span>.</p><p>For further reading on the UC problem we refer interested readers to G. Morales-Espana, J. M. Latorre, and A. Ramos, &quot;Tight and Compact MILP Formulation for the Thermal Unit Commitment Problem,&quot; IEEE Transactions on Power Systems, vol. 28, pp. 4897-4908, 2013.</p><p>In the following example we convert the ED model explained above to the UC model.</p><pre><code class="language-julia hljs">function solve_unit_commitment(generators::Vector, wind, scenario)
+Plots.plot(dispatch_plot, wind_plot)</code></pre><img src="d9e67996.svg" alt="Example block output"/><p>This particular drawback can be overcome by introducing binary decisions on the &quot;on/off&quot; status of generators. This model is called unit commitment and considered later in these notes.</p><p>For further reading on the interplay between wind generation and the minimum power output constraints of generators, we refer interested readers to R. Baldick, &quot;Wind and energy markets: a case study of Texas,&quot; IEEE Systems Journal, vol. 6, pp. 27-34, 2012.</p><h2 id="Unit-commitment"><a class="docs-heading-anchor" href="#Unit-commitment">Unit commitment</a><a id="Unit-commitment-1"></a><a class="docs-heading-anchor-permalink" href="#Unit-commitment" title="Permalink"></a></h2><p>The Unit Commitment (UC) model can be obtained from ED model by introducing binary variable associated with each generator. This binary variable can attain two values: if it is &quot;1,&quot; the generator is synchronized and, thus, can be dispatched, otherwise, that is, if the binary variable is &quot;0,&quot; that generator is not synchronized and its power output is set to 0.</p><p>To obtain the mathematical formulation of the UC model, we will modify the constraints of the ED model as follows:</p><p class="math-container">\[g^{\min}_{i} \cdot u_{t,i} \leq g_{i} \leq g^{\max}_{i} \cdot u_{t,i},\]</p><p>where <span>$u_{i} \in \{0,1\}.$</span> In this constraint, if <span>$u_{i} = 0$</span>, then <span>$g_{i}  = 0$</span>. On the other hand, if <span>$u_{i} = 1$</span>, then <span>$g^{min}_{i} \leq g_{i} \leq g^{max}_{i}$</span>.</p><p>For further reading on the UC problem we refer interested readers to G. Morales-Espana, J. M. Latorre, and A. Ramos, &quot;Tight and Compact MILP Formulation for the Thermal Unit Commitment Problem,&quot; IEEE Transactions on Power Systems, vol. 28, pp. 4897-4908, 2013.</p><p>In the following example we convert the ED model explained above to the UC model.</p><pre><code class="language-julia hljs">function solve_unit_commitment(generators::Vector, wind, scenario)
     model = Model(HiGHS.Optimizer)
     set_silent(model)
     N = length(generators)
@@ -322,7 +322,7 @@
     ),
 )
 
-Plots.plot(commitment_plot, dispatch_plot)</code></pre><img src="e09dc019.svg" alt="Example block output"/><h2 id="Nonlinear-economic-dispatch"><a class="docs-heading-anchor" href="#Nonlinear-economic-dispatch">Nonlinear economic dispatch</a><a id="Nonlinear-economic-dispatch-1"></a><a class="docs-heading-anchor-permalink" href="#Nonlinear-economic-dispatch" title="Permalink"></a></h2><p>As a final example, we modify our economic dispatch problem in two ways:</p><ul><li>The thermal cost function is user-defined</li><li>The output of the wind is only the square-root of the dispatch</li></ul><pre><code class="language-julia hljs">import Ipopt
+Plots.plot(commitment_plot, dispatch_plot)</code></pre><img src="e02e8c47.svg" alt="Example block output"/><h2 id="Nonlinear-economic-dispatch"><a class="docs-heading-anchor" href="#Nonlinear-economic-dispatch">Nonlinear economic dispatch</a><a id="Nonlinear-economic-dispatch-1"></a><a class="docs-heading-anchor-permalink" href="#Nonlinear-economic-dispatch" title="Permalink"></a></h2><p>As a final example, we modify our economic dispatch problem in two ways:</p><ul><li>The thermal cost function is user-defined</li><li>The output of the wind is only the square-root of the dispatch</li></ul><pre><code class="language-julia hljs">import Ipopt
 
 &quot;&quot;&quot;
     thermal_cost_function(g)
@@ -373,7 +373,7 @@
 end
 
 solution =
-    solve_nonlinear_economic_dispatch(generators, wind_generator, scenario)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(g = [847.3509933774712, 648.6754966887423], w = 15.788781193899027, wind_spill = 184.211218806101, total_cost = 190455.298013245)</code></pre><p>Now let&#39;s see how the wind is dispatched as a function of the cost:</p><pre><code class="language-julia hljs">wind_cost = 0.0:1:100
+    solve_nonlinear_economic_dispatch(generators, wind_generator, scenario)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(g = [847.3509933774712, 648.6754966887423], w = 15.78878119389903, wind_spill = 184.21121880610096, total_cost = 190455.298013245)</code></pre><p>Now let&#39;s see how the wind is dispatched as a function of the cost:</p><pre><code class="language-julia hljs">wind_cost = 0.0:1:100
 wind_dispatch = Float64[]
 for c in wind_cost
     sol = solve_nonlinear_economic_dispatch(
@@ -391,4 +391,4 @@
     xlabel = &quot;Cost&quot;,
     ylabel = &quot;Dispatch [MW]&quot;,
     label = false,
-)</code></pre><img src="2588df61.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../algorithms/parallelism/">« Parallelism</a><a class="docs-footer-nextpage" href="../optimal_power_flow/">Optimal power flow »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><img src="0215129c.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../algorithms/parallelism/">« Parallelism</a><a class="docs-footer-nextpage" href="../optimal_power_flow/">Optimal power flow »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/3b3c353b.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/3b3c353b.svg
new file mode 100644
index 00000000000..993f075cc08
--- /dev/null
+++ b/previews/PR3562/tutorials/applications/two_stage_stochastic/3b3c353b.svg
@@ -0,0 +1,80 @@
+<?xml version="1.0" encoding="utf-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
+<defs>
+  <clipPath id="clip770">
+    <rect x="0" y="0" width="2400" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip770)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip771">
+    <rect x="480" y="0" width="1681" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip770)" d="M205.121 1423.18 L2352.76 1423.18 L2352.76 47.2441 L205.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip772">
+    <rect x="205" y="47" width="2149" height="1377"/>
+  </clipPath>
+</defs>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="514.383,1423.18 514.383,47.2441 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="896.661,1423.18 896.661,47.2441 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1278.94,1423.18 1278.94,47.2441 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1661.22,1423.18 1661.22,47.2441 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2043.49,1423.18 2043.49,47.2441 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="514.383,1423.18 514.383,1404.28 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="896.661,1423.18 896.661,1404.28 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1278.94,1423.18 1278.94,1404.28 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1661.22,1423.18 1661.22,1404.28 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2043.49,1423.18 2043.49,1404.28 "/>
+<path clip-path="url(#clip770)" d="M487.474 1455.09 L475.668 1473.54 L487.474 1473.54 L487.474 1455.09 M486.247 1451.02 L492.127 1451.02 L492.127 1473.54 L497.057 1473.54 L497.057 1477.43 L492.127 1477.43 L492.127 1485.58 L487.474 1485.58 L487.474 1477.43 L471.872 1477.43 L471.872 1472.92 L486.247 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M514.789 1454.1 Q511.177 1454.1 509.349 1457.66 Q507.543 1461.2 507.543 1468.33 Q507.543 1475.44 509.349 1479.01 Q511.177 1482.55 514.789 1482.55 Q518.423 1482.55 520.228 1479.01 Q522.057 1475.44 522.057 1468.33 Q522.057 1461.2 520.228 1457.66 Q518.423 1454.1 514.789 1454.1 M514.789 1450.39 Q520.599 1450.39 523.654 1455 Q526.733 1459.58 526.733 1468.33 Q526.733 1477.06 523.654 1481.67 Q520.599 1486.25 514.789 1486.25 Q508.978 1486.25 505.9 1481.67 Q502.844 1477.06 502.844 1468.33 Q502.844 1459.58 505.9 1455 Q508.978 1450.39 514.789 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M544.95 1454.1 Q541.339 1454.1 539.511 1457.66 Q537.705 1461.2 537.705 1468.33 Q537.705 1475.44 539.511 1479.01 Q541.339 1482.55 544.95 1482.55 Q548.585 1482.55 550.39 1479.01 Q552.219 1475.44 552.219 1468.33 Q552.219 1461.2 550.39 1457.66 Q548.585 1454.1 544.95 1454.1 M544.95 1450.39 Q550.761 1450.39 553.816 1455 Q556.895 1459.58 556.895 1468.33 Q556.895 1477.06 553.816 1481.67 Q550.761 1486.25 544.95 1486.25 Q539.14 1486.25 536.062 1481.67 Q533.006 1477.06 533.006 1468.33 Q533.006 1459.58 536.062 1455 Q539.14 1450.39 544.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M869.751 1455.09 L857.946 1473.54 L869.751 1473.54 L869.751 1455.09 M868.525 1451.02 L874.404 1451.02 L874.404 1473.54 L879.335 1473.54 L879.335 1477.43 L874.404 1477.43 L874.404 1485.58 L869.751 1485.58 L869.751 1477.43 L854.15 1477.43 L854.15 1472.92 L868.525 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M887.112 1451.02 L905.469 1451.02 L905.469 1454.96 L891.395 1454.96 L891.395 1463.43 Q892.413 1463.08 893.432 1462.92 Q894.45 1462.73 895.469 1462.73 Q901.256 1462.73 904.635 1465.9 Q908.015 1469.08 908.015 1474.49 Q908.015 1480.07 904.543 1483.17 Q901.071 1486.25 894.751 1486.25 Q892.575 1486.25 890.307 1485.88 Q888.061 1485.51 885.654 1484.77 L885.654 1480.07 Q887.737 1481.2 889.96 1481.76 Q892.182 1482.32 894.659 1482.32 Q898.663 1482.32 901.001 1480.21 Q903.339 1478.1 903.339 1474.49 Q903.339 1470.88 901.001 1468.77 Q898.663 1466.67 894.659 1466.67 Q892.784 1466.67 890.909 1467.08 Q889.057 1467.5 887.112 1468.38 L887.112 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M927.228 1454.1 Q923.617 1454.1 921.788 1457.66 Q919.983 1461.2 919.983 1468.33 Q919.983 1475.44 921.788 1479.01 Q923.617 1482.55 927.228 1482.55 Q930.862 1482.55 932.668 1479.01 Q934.496 1475.44 934.496 1468.33 Q934.496 1461.2 932.668 1457.66 Q930.862 1454.1 927.228 1454.1 M927.228 1450.39 Q933.038 1450.39 936.094 1455 Q939.172 1459.58 939.172 1468.33 Q939.172 1477.06 936.094 1481.67 Q933.038 1486.25 927.228 1486.25 Q921.418 1486.25 918.339 1481.67 Q915.284 1477.06 915.284 1468.33 Q915.284 1459.58 918.339 1455 Q921.418 1450.39 927.228 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1238.56 1451.02 L1256.91 1451.02 L1256.91 1454.96 L1242.84 1454.96 L1242.84 1463.43 Q1243.86 1463.08 1244.88 1462.92 Q1245.89 1462.73 1246.91 1462.73 Q1252.7 1462.73 1256.08 1465.9 Q1259.46 1469.08 1259.46 1474.49 Q1259.46 1480.07 1255.99 1483.17 Q1252.51 1486.25 1246.2 1486.25 Q1244.02 1486.25 1241.75 1485.88 Q1239.51 1485.51 1237.1 1484.77 L1237.1 1480.07 Q1239.18 1481.2 1241.4 1481.76 Q1243.63 1482.32 1246.1 1482.32 Q1250.11 1482.32 1252.45 1480.21 Q1254.78 1478.1 1254.78 1474.49 Q1254.78 1470.88 1252.45 1468.77 Q1250.11 1466.67 1246.1 1466.67 Q1244.23 1466.67 1242.35 1467.08 Q1240.5 1467.5 1238.56 1468.38 L1238.56 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1278.67 1454.1 Q1275.06 1454.1 1273.23 1457.66 Q1271.43 1461.2 1271.43 1468.33 Q1271.43 1475.44 1273.23 1479.01 Q1275.06 1482.55 1278.67 1482.55 Q1282.31 1482.55 1284.11 1479.01 Q1285.94 1475.44 1285.94 1468.33 Q1285.94 1461.2 1284.11 1457.66 Q1282.31 1454.1 1278.67 1454.1 M1278.67 1450.39 Q1284.48 1450.39 1287.54 1455 Q1290.62 1459.58 1290.62 1468.33 Q1290.62 1477.06 1287.54 1481.67 Q1284.48 1486.25 1278.67 1486.25 Q1272.86 1486.25 1269.78 1481.67 Q1266.73 1477.06 1266.73 1468.33 Q1266.73 1459.58 1269.78 1455 Q1272.86 1450.39 1278.67 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1308.83 1454.1 Q1305.22 1454.1 1303.39 1457.66 Q1301.59 1461.2 1301.59 1468.33 Q1301.59 1475.44 1303.39 1479.01 Q1305.22 1482.55 1308.83 1482.55 Q1312.47 1482.55 1314.27 1479.01 Q1316.1 1475.44 1316.1 1468.33 Q1316.1 1461.2 1314.27 1457.66 Q1312.47 1454.1 1308.83 1454.1 M1308.83 1450.39 Q1314.64 1450.39 1317.7 1455 Q1320.78 1459.58 1320.78 1468.33 Q1320.78 1477.06 1317.7 1481.67 Q1314.64 1486.25 1308.83 1486.25 Q1303.02 1486.25 1299.95 1481.67 Q1296.89 1477.06 1296.89 1468.33 Q1296.89 1459.58 1299.95 1455 Q1303.02 1450.39 1308.83 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1620.83 1451.02 L1639.19 1451.02 L1639.19 1454.96 L1625.12 1454.96 L1625.12 1463.43 Q1626.14 1463.08 1627.15 1462.92 Q1628.17 1462.73 1629.19 1462.73 Q1634.98 1462.73 1638.36 1465.9 Q1641.74 1469.08 1641.74 1474.49 Q1641.74 1480.07 1638.26 1483.17 Q1634.79 1486.25 1628.47 1486.25 Q1626.3 1486.25 1624.03 1485.88 Q1621.78 1485.51 1619.38 1484.77 L1619.38 1480.07 Q1621.46 1481.2 1623.68 1481.76 Q1625.9 1482.32 1628.38 1482.32 Q1632.39 1482.32 1634.72 1480.21 Q1637.06 1478.1 1637.06 1474.49 Q1637.06 1470.88 1634.72 1468.77 Q1632.39 1466.67 1628.38 1466.67 Q1626.51 1466.67 1624.63 1467.08 Q1622.78 1467.5 1620.83 1468.38 L1620.83 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1651 1451.02 L1669.35 1451.02 L1669.35 1454.96 L1655.28 1454.96 L1655.28 1463.43 Q1656.3 1463.08 1657.32 1462.92 Q1658.33 1462.73 1659.35 1462.73 Q1665.14 1462.73 1668.52 1465.9 Q1671.9 1469.08 1671.9 1474.49 Q1671.9 1480.07 1668.43 1483.17 Q1664.95 1486.25 1658.63 1486.25 Q1656.46 1486.25 1654.19 1485.88 Q1651.95 1485.51 1649.54 1484.77 L1649.54 1480.07 Q1651.62 1481.2 1653.84 1481.76 Q1656.07 1482.32 1658.54 1482.32 Q1662.55 1482.32 1664.88 1480.21 Q1667.22 1478.1 1667.22 1474.49 Q1667.22 1470.88 1664.88 1468.77 Q1662.55 1466.67 1658.54 1466.67 Q1656.67 1466.67 1654.79 1467.08 Q1652.94 1467.5 1651 1468.38 L1651 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1691.11 1454.1 Q1687.5 1454.1 1685.67 1457.66 Q1683.87 1461.2 1683.87 1468.33 Q1683.87 1475.44 1685.67 1479.01 Q1687.5 1482.55 1691.11 1482.55 Q1694.75 1482.55 1696.55 1479.01 Q1698.38 1475.44 1698.38 1468.33 Q1698.38 1461.2 1696.55 1457.66 Q1694.75 1454.1 1691.11 1454.1 M1691.11 1450.39 Q1696.92 1450.39 1699.98 1455 Q1703.06 1459.58 1703.06 1468.33 Q1703.06 1477.06 1699.98 1481.67 Q1696.92 1486.25 1691.11 1486.25 Q1685.3 1486.25 1682.22 1481.67 Q1679.17 1477.06 1679.17 1468.33 Q1679.17 1459.58 1682.22 1455 Q1685.3 1450.39 1691.11 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2013.82 1466.44 Q2010.67 1466.44 2008.82 1468.59 Q2006.99 1470.74 2006.99 1474.49 Q2006.99 1478.22 2008.82 1480.39 Q2010.67 1482.55 2013.82 1482.55 Q2016.97 1482.55 2018.79 1480.39 Q2020.65 1478.22 2020.65 1474.49 Q2020.65 1470.74 2018.79 1468.59 Q2016.97 1466.44 2013.82 1466.44 M2023.1 1451.78 L2023.1 1456.04 Q2021.34 1455.21 2019.54 1454.77 Q2017.75 1454.33 2015.99 1454.33 Q2011.36 1454.33 2008.91 1457.45 Q2006.48 1460.58 2006.13 1466.9 Q2007.5 1464.89 2009.56 1463.82 Q2011.62 1462.73 2014.1 1462.73 Q2019.3 1462.73 2022.31 1465.9 Q2025.35 1469.05 2025.35 1474.49 Q2025.35 1479.82 2022.2 1483.03 Q2019.05 1486.25 2013.82 1486.25 Q2007.82 1486.25 2004.65 1481.67 Q2001.48 1477.06 2001.48 1468.33 Q2001.48 1460.14 2005.37 1455.28 Q2009.26 1450.39 2015.81 1450.39 Q2017.57 1450.39 2019.35 1450.74 Q2021.16 1451.09 2023.1 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2043.4 1454.1 Q2039.79 1454.1 2037.96 1457.66 Q2036.16 1461.2 2036.16 1468.33 Q2036.16 1475.44 2037.96 1479.01 Q2039.79 1482.55 2043.4 1482.55 Q2047.04 1482.55 2048.84 1479.01 Q2050.67 1475.44 2050.67 1468.33 Q2050.67 1461.2 2048.84 1457.66 Q2047.04 1454.1 2043.4 1454.1 M2043.4 1450.39 Q2049.21 1450.39 2052.27 1455 Q2055.35 1459.58 2055.35 1468.33 Q2055.35 1477.06 2052.27 1481.67 Q2049.21 1486.25 2043.4 1486.25 Q2037.59 1486.25 2034.51 1481.67 Q2031.46 1477.06 2031.46 1468.33 Q2031.46 1459.58 2034.51 1455 Q2037.59 1450.39 2043.4 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M2073.56 1454.1 Q2069.95 1454.1 2068.12 1457.66 Q2066.32 1461.2 2066.32 1468.33 Q2066.32 1475.44 2068.12 1479.01 Q2069.95 1482.55 2073.56 1482.55 Q2077.2 1482.55 2079 1479.01 Q2080.83 1475.44 2080.83 1468.33 Q2080.83 1461.2 2079 1457.66 Q2077.2 1454.1 2073.56 1454.1 M2073.56 1450.39 Q2079.37 1450.39 2082.43 1455 Q2085.51 1459.58 2085.51 1468.33 Q2085.51 1477.06 2082.43 1481.67 Q2079.37 1486.25 2073.56 1486.25 Q2067.75 1486.25 2064.67 1481.67 Q2061.62 1477.06 2061.62 1468.33 Q2061.62 1459.58 2064.67 1455 Q2067.75 1450.39 2073.56 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1149.83 1525.81 L1149.83 1543.66 L1157.91 1543.66 Q1162.4 1543.66 1164.85 1541.34 Q1167.3 1539.02 1167.3 1534.72 Q1167.3 1530.45 1164.85 1528.13 Q1162.4 1525.81 1157.91 1525.81 L1149.83 1525.81 M1143.4 1520.52 L1157.91 1520.52 Q1165.9 1520.52 1169.97 1524.15 Q1174.08 1527.75 1174.08 1534.72 Q1174.08 1541.75 1169.97 1545.35 Q1165.9 1548.95 1157.91 1548.95 L1149.83 1548.95 L1149.83 1568.04 L1143.4 1568.04 L1143.4 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1201.96 1537.87 Q1200.97 1537.3 1199.8 1537.04 Q1198.65 1536.76 1197.25 1536.76 Q1192.29 1536.76 1189.61 1540 Q1186.97 1543.22 1186.97 1549.27 L1186.97 1568.04 L1181.08 1568.04 L1181.08 1532.4 L1186.97 1532.4 L1186.97 1537.93 Q1188.82 1534.69 1191.78 1533.13 Q1194.74 1531.54 1198.97 1531.54 Q1199.57 1531.54 1200.31 1531.63 Q1201.04 1531.7 1201.93 1531.85 L1201.96 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1220.49 1536.5 Q1215.77 1536.5 1213.04 1540.19 Q1210.3 1543.85 1210.3 1550.25 Q1210.3 1556.65 1213.01 1560.34 Q1215.74 1564 1220.49 1564 Q1225.16 1564 1227.9 1560.31 Q1230.64 1556.62 1230.64 1550.25 Q1230.64 1543.92 1227.9 1540.23 Q1225.16 1536.5 1220.49 1536.5 M1220.49 1531.54 Q1228.12 1531.54 1232.48 1536.5 Q1236.85 1541.47 1236.85 1550.25 Q1236.85 1559 1232.48 1564 Q1228.12 1568.97 1220.49 1568.97 Q1212.81 1568.97 1208.45 1564 Q1204.13 1559 1204.13 1550.25 Q1204.13 1541.47 1208.45 1536.5 Q1212.81 1531.54 1220.49 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1264.6 1518.52 L1264.6 1523.39 L1259 1523.39 Q1255.85 1523.39 1254.61 1524.66 Q1253.4 1525.93 1253.4 1529.24 L1253.4 1532.4 L1263.04 1532.4 L1263.04 1536.95 L1253.4 1536.95 L1253.4 1568.04 L1247.51 1568.04 L1247.51 1536.95 L1241.91 1536.95 L1241.91 1532.4 L1247.51 1532.4 L1247.51 1529.91 Q1247.51 1523.96 1250.28 1521.26 Q1253.05 1518.52 1259.06 1518.52 L1264.6 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1269.5 1532.4 L1275.36 1532.4 L1275.36 1568.04 L1269.5 1568.04 L1269.5 1532.4 M1269.5 1518.52 L1275.36 1518.52 L1275.36 1525.93 L1269.5 1525.93 L1269.5 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1293.4 1522.27 L1293.4 1532.4 L1305.47 1532.4 L1305.47 1536.95 L1293.4 1536.95 L1293.4 1556.3 Q1293.4 1560.66 1294.58 1561.9 Q1295.79 1563.14 1299.45 1563.14 L1305.47 1563.14 L1305.47 1568.04 L1299.45 1568.04 Q1292.67 1568.04 1290.09 1565.53 Q1287.52 1562.98 1287.52 1556.3 L1287.52 1536.95 L1283.22 1536.95 L1283.22 1532.4 L1287.52 1532.4 L1287.52 1522.27 L1293.4 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1333.35 1518.52 L1346.84 1518.52 L1346.84 1523.07 L1339.21 1523.07 L1339.21 1572.09 L1346.84 1572.09 L1346.84 1576.64 L1333.35 1576.64 L1333.35 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1375.2 1577.62 L1372.02 1577.62 L1371.99 1568.04 Q1368.65 1567.98 1365.31 1567.28 Q1361.96 1566.55 1358.59 1565.12 L1358.59 1559.39 Q1361.84 1561.42 1365.15 1562.47 Q1368.49 1563.49 1372.02 1563.52 L1372.02 1549.01 Q1364.99 1547.86 1361.77 1545.13 Q1358.59 1542.39 1358.59 1537.62 Q1358.59 1532.43 1362.06 1529.44 Q1365.53 1526.44 1372.02 1526 L1372.02 1518.52 L1375.2 1518.52 L1375.2 1525.9 Q1378.16 1526.03 1380.93 1526.54 Q1383.7 1527.02 1386.34 1527.88 L1386.34 1533.45 Q1383.7 1532.11 1380.9 1531.38 Q1378.13 1530.65 1375.2 1530.52 L1375.2 1544.11 Q1382.43 1545.22 1385.83 1548.09 Q1389.24 1550.95 1389.24 1555.92 Q1389.24 1561.3 1385.61 1564.42 Q1382.02 1567.5 1375.2 1567.98 L1375.2 1577.62 M1372.02 1543.54 L1372.02 1530.49 Q1368.33 1530.9 1366.39 1532.59 Q1364.45 1534.27 1364.45 1537.07 Q1364.45 1539.81 1366.23 1541.34 Q1368.04 1542.87 1372.02 1543.54 M1375.2 1549.65 L1375.2 1563.43 Q1379.25 1562.89 1381.28 1561.14 Q1383.35 1559.39 1383.35 1556.52 Q1383.35 1553.72 1381.38 1552.07 Q1379.44 1550.41 1375.2 1549.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M1414.48 1518.52 L1414.48 1576.64 L1400.98 1576.64 L1400.98 1572.09 L1408.59 1572.09 L1408.59 1523.07 L1400.98 1523.07 L1400.98 1518.52 L1414.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1117.42 2352.76,1117.42 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,811.653 2352.76,811.653 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,505.889 2352.76,505.889 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,200.126 2352.76,200.126 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 205.121,47.2441 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 224.019,1423.18 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1117.42 224.019,1117.42 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,811.653 224.019,811.653 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,505.889 224.019,505.889 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,200.126 224.019,200.126 "/>
+<path clip-path="url(#clip770)" d="M157.177 1408.98 Q153.566 1408.98 151.737 1412.54 Q149.931 1416.08 149.931 1423.21 Q149.931 1430.32 151.737 1433.89 Q153.566 1437.43 157.177 1437.43 Q160.811 1437.43 162.616 1433.89 Q164.445 1430.32 164.445 1423.21 Q164.445 1416.08 162.616 1412.54 Q160.811 1408.98 157.177 1408.98 M157.177 1405.27 Q162.987 1405.27 166.042 1409.88 Q169.121 1414.46 169.121 1423.21 Q169.121 1431.94 166.042 1436.55 Q162.987 1441.13 157.177 1441.13 Q151.366 1441.13 148.288 1436.55 Q145.232 1431.94 145.232 1423.21 Q145.232 1414.46 148.288 1409.88 Q151.366 1405.27 157.177 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M117.825 1130.76 L125.464 1130.76 L125.464 1104.4 L117.154 1106.06 L117.154 1101.8 L125.418 1100.14 L130.093 1100.14 L130.093 1130.76 L137.732 1130.76 L137.732 1134.7 L117.825 1134.7 L117.825 1130.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M157.177 1103.22 Q153.566 1103.22 151.737 1106.78 Q149.931 1110.32 149.931 1117.45 Q149.931 1124.56 151.737 1128.12 Q153.566 1131.66 157.177 1131.66 Q160.811 1131.66 162.616 1128.12 Q164.445 1124.56 164.445 1117.45 Q164.445 1110.32 162.616 1106.78 Q160.811 1103.22 157.177 1103.22 M157.177 1099.51 Q162.987 1099.51 166.042 1104.12 Q169.121 1108.7 169.121 1117.45 Q169.121 1126.18 166.042 1130.78 Q162.987 1135.37 157.177 1135.37 Q151.366 1135.37 148.288 1130.78 Q145.232 1126.18 145.232 1117.45 Q145.232 1108.7 148.288 1104.12 Q151.366 1099.51 157.177 1099.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M121.043 824.998 L137.362 824.998 L137.362 828.933 L115.418 828.933 L115.418 824.998 Q118.08 822.243 122.663 817.613 Q127.269 812.961 128.45 811.618 Q130.695 809.095 131.575 807.359 Q132.478 805.6 132.478 803.91 Q132.478 801.155 130.533 799.419 Q128.612 797.683 125.51 797.683 Q123.311 797.683 120.857 798.447 Q118.427 799.211 115.649 800.762 L115.649 796.04 Q118.473 794.905 120.927 794.327 Q123.38 793.748 125.418 793.748 Q130.788 793.748 133.982 796.433 Q137.177 799.118 137.177 803.609 Q137.177 805.739 136.367 807.66 Q135.579 809.558 133.473 812.151 Q132.894 812.822 129.792 816.039 Q126.691 819.234 121.043 824.998 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M157.177 797.452 Q153.566 797.452 151.737 801.016 Q149.931 804.558 149.931 811.688 Q149.931 818.794 151.737 822.359 Q153.566 825.9 157.177 825.9 Q160.811 825.9 162.616 822.359 Q164.445 818.794 164.445 811.688 Q164.445 804.558 162.616 801.016 Q160.811 797.452 157.177 797.452 M157.177 793.748 Q162.987 793.748 166.042 798.354 Q169.121 802.938 169.121 811.688 Q169.121 820.414 166.042 825.021 Q162.987 829.604 157.177 829.604 Q151.366 829.604 148.288 825.021 Q145.232 820.414 145.232 811.688 Q145.232 802.938 148.288 798.354 Q151.366 793.748 157.177 793.748 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M131.181 504.535 Q134.538 505.253 136.413 507.521 Q138.311 509.79 138.311 513.123 Q138.311 518.239 134.792 521.04 Q131.274 523.841 124.793 523.841 Q122.617 523.841 120.302 523.401 Q118.01 522.984 115.556 522.128 L115.556 517.614 Q117.501 518.748 119.816 519.327 Q122.13 519.905 124.654 519.905 Q129.052 519.905 131.343 518.169 Q133.658 516.433 133.658 513.123 Q133.658 510.068 131.505 508.355 Q129.376 506.619 125.556 506.619 L121.529 506.619 L121.529 502.776 L125.742 502.776 Q129.191 502.776 131.019 501.41 Q132.848 500.021 132.848 497.429 Q132.848 494.767 130.95 493.355 Q129.075 491.92 125.556 491.92 Q123.635 491.92 121.436 492.336 Q119.237 492.753 116.598 493.632 L116.598 489.466 Q119.26 488.725 121.575 488.355 Q123.913 487.984 125.973 487.984 Q131.297 487.984 134.399 490.415 Q137.501 492.822 137.501 496.943 Q137.501 499.813 135.857 501.804 Q134.214 503.771 131.181 504.535 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M157.177 491.688 Q153.566 491.688 151.737 495.253 Q149.931 498.794 149.931 505.924 Q149.931 513.031 151.737 516.595 Q153.566 520.137 157.177 520.137 Q160.811 520.137 162.616 516.595 Q164.445 513.031 164.445 505.924 Q164.445 498.794 162.616 495.253 Q160.811 491.688 157.177 491.688 M157.177 487.984 Q162.987 487.984 166.042 492.591 Q169.121 497.174 169.121 505.924 Q169.121 514.651 166.042 519.257 Q162.987 523.841 157.177 523.841 Q151.366 523.841 148.288 519.257 Q145.232 514.651 145.232 505.924 Q145.232 497.174 148.288 492.591 Q151.366 487.984 157.177 487.984 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M129.862 186.92 L118.056 205.369 L129.862 205.369 L129.862 186.92 M128.635 182.846 L134.515 182.846 L134.515 205.369 L139.445 205.369 L139.445 209.258 L134.515 209.258 L134.515 217.406 L129.862 217.406 L129.862 209.258 L114.26 209.258 L114.26 204.744 L128.635 182.846 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M157.177 185.925 Q153.566 185.925 151.737 189.489 Q149.931 193.031 149.931 200.161 Q149.931 207.267 151.737 210.832 Q153.566 214.373 157.177 214.373 Q160.811 214.373 162.616 210.832 Q164.445 207.267 164.445 200.161 Q164.445 193.031 162.616 189.489 Q160.811 185.925 157.177 185.925 M157.177 182.221 Q162.987 182.221 166.042 186.827 Q169.121 191.411 169.121 200.161 Q169.121 208.887 166.042 213.494 Q162.987 218.077 157.177 218.077 Q151.366 218.077 148.288 213.494 Q145.232 208.887 145.232 200.161 Q145.232 191.411 148.288 186.827 Q151.366 182.221 157.177 182.221 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M16.4842 1073.5 L16.4842 1064.84 L56.238 1043.77 L16.4842 1043.77 L16.4842 1037.54 L64.0042 1037.54 L64.0042 1046.19 L24.2503 1067.26 L64.0042 1067.26 L64.0042 1073.5 L16.4842 1073.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M49.9359 1025.6 L28.3562 1025.6 L28.3562 1019.74 L49.7131 1019.74 Q54.7739 1019.74 57.3202 1017.77 Q59.8346 1015.8 59.8346 1011.85 Q59.8346 1007.11 56.8109 1004.37 Q53.7872 1001.6 48.5673 1001.6 L28.3562 1001.6 L28.3562 995.744 L64.0042 995.744 L64.0042 1001.6 L58.5296 1001.6 Q61.7762 1003.73 63.3676 1006.57 Q64.9272 1009.37 64.9272 1013.09 Q64.9272 1019.23 61.1078 1022.42 Q57.2883 1025.6 49.9359 1025.6 M27.4968 1010.86 L27.4968 1010.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M35.1993 955.927 Q31.2526 953.731 29.3747 950.675 Q27.4968 947.62 27.4968 943.482 Q27.4968 937.912 31.4117 934.888 Q35.2948 931.865 42.4881 931.865 L64.0042 931.865 L64.0042 937.753 L42.679 937.753 Q37.5546 937.753 35.072 939.567 Q32.5894 941.381 32.5894 945.105 Q32.5894 949.657 35.6131 952.298 Q38.6368 954.94 43.8567 954.94 L64.0042 954.94 L64.0042 960.829 L42.679 960.829 Q37.5228 960.829 35.072 962.643 Q32.5894 964.457 32.5894 968.245 Q32.5894 972.732 35.6449 975.374 Q38.6686 978.016 43.8567 978.016 L64.0042 978.016 L64.0042 983.904 L28.3562 983.904 L28.3562 978.016 L33.8944 978.016 Q30.616 976.011 29.0564 973.21 Q27.4968 970.409 27.4968 966.558 Q27.4968 962.675 29.4702 959.969 Q31.4436 957.232 35.1993 955.927 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M46.212 894.593 Q39.7508 894.593 36.0905 897.267 Q32.3984 899.909 32.3984 904.556 Q32.3984 909.203 36.0905 911.876 Q39.7508 914.518 46.212 914.518 Q52.6732 914.518 56.3653 911.876 Q60.0256 909.203 60.0256 904.556 Q60.0256 899.909 56.3653 897.267 Q52.6732 894.593 46.212 894.593 M33.7671 914.518 Q30.5842 912.672 29.0564 909.871 Q27.4968 907.038 27.4968 903.123 Q27.4968 896.63 32.6531 892.588 Q37.8093 888.514 46.212 888.514 Q54.6147 888.514 59.771 892.588 Q64.9272 896.63 64.9272 903.123 Q64.9272 907.038 63.3994 909.871 Q61.8398 912.672 58.657 914.518 L64.0042 914.518 L64.0042 920.406 L14.479 920.406 L14.479 914.518 L33.7671 914.518 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M44.7161 848.315 L47.5806 848.315 L47.5806 875.242 Q53.6281 874.86 56.8109 871.613 Q59.9619 868.335 59.9619 862.51 Q59.9619 859.136 59.1344 855.985 Q58.3069 852.802 56.6518 849.683 L62.1899 849.683 Q63.5267 852.834 64.227 856.144 Q64.9272 859.455 64.9272 862.86 Q64.9272 871.39 59.9619 876.387 Q54.9967 881.353 46.5303 881.353 Q37.7774 881.353 32.6531 876.642 Q27.4968 871.9 27.4968 863.879 Q27.4968 856.686 32.1438 852.516 Q36.7589 848.315 44.7161 848.315 M42.9973 854.171 Q38.1912 854.235 35.3266 856.877 Q32.4621 859.486 32.4621 863.815 Q32.4621 868.717 35.2312 871.677 Q38.0002 874.605 43.0292 875.051 L42.9973 854.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M33.8307 818.046 Q33.2578 819.032 33.0032 820.21 Q32.7167 821.356 32.7167 822.756 Q32.7167 827.722 35.9632 830.395 Q39.1779 833.037 45.2253 833.037 L64.0042 833.037 L64.0042 838.925 L28.3562 838.925 L28.3562 833.037 L33.8944 833.037 Q30.6479 831.191 29.0883 828.231 Q27.4968 825.271 27.4968 821.038 Q27.4968 820.433 27.5923 819.701 Q27.656 818.969 27.8151 818.078 L33.8307 818.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M32.4621 777.369 Q32.4621 782.079 36.1542 784.817 Q39.8145 787.554 46.212 787.554 Q52.6095 787.554 56.3017 784.849 Q59.9619 782.111 59.9619 777.369 Q59.9619 772.69 56.2698 769.953 Q52.5777 767.216 46.212 767.216 Q39.8781 767.216 36.186 769.953 Q32.4621 772.69 32.4621 777.369 M27.4968 777.369 Q27.4968 769.73 32.4621 765.37 Q37.4273 761.009 46.212 761.009 Q54.9649 761.009 59.9619 765.37 Q64.9272 769.73 64.9272 777.369 Q64.9272 785.04 59.9619 789.4 Q54.9649 793.729 46.212 793.729 Q37.4273 793.729 32.4621 789.4 Q27.4968 785.04 27.4968 777.369 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M14.479 733.255 L19.3487 733.255 L19.3487 738.856 Q19.3487 742.007 20.6219 743.249 Q21.895 744.458 25.2052 744.458 L28.3562 744.458 L28.3562 734.814 L32.9077 734.814 L32.9077 744.458 L64.0042 744.458 L64.0042 750.346 L32.9077 750.346 L32.9077 755.948 L28.3562 755.948 L28.3562 750.346 L25.8736 750.346 Q19.9216 750.346 17.2162 747.577 Q14.479 744.808 14.479 738.793 L14.479 733.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M32.4621 693.819 Q32.4621 698.53 36.1542 701.267 Q39.8145 704.004 46.212 704.004 Q52.6095 704.004 56.3017 701.299 Q59.9619 698.561 59.9619 693.819 Q59.9619 689.14 56.2698 686.403 Q52.5777 683.666 46.212 683.666 Q39.8781 683.666 36.186 686.403 Q32.4621 689.14 32.4621 693.819 M27.4968 693.819 Q27.4968 686.18 32.4621 681.82 Q37.4273 677.459 46.212 677.459 Q54.9649 677.459 59.9619 681.82 Q64.9272 686.18 64.9272 693.819 Q64.9272 701.49 59.9619 705.85 Q54.9649 710.179 46.212 710.179 Q37.4273 710.179 32.4621 705.85 Q27.4968 701.49 27.4968 693.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M49.9359 668.356 L28.3562 668.356 L28.3562 662.5 L49.7131 662.5 Q54.7739 662.5 57.3202 660.526 Q59.8346 658.553 59.8346 654.606 Q59.8346 649.864 56.8109 647.127 Q53.7872 644.357 48.5673 644.357 L28.3562 644.357 L28.3562 638.501 L64.0042 638.501 L64.0042 644.357 L58.5296 644.357 Q61.7762 646.49 63.3676 649.323 Q64.9272 652.124 64.9272 655.848 Q64.9272 661.99 61.1078 665.173 Q57.2883 668.356 49.9359 668.356 M27.4968 653.62 L27.4968 653.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M18.2347 620.645 L28.3562 620.645 L28.3562 608.582 L32.9077 608.582 L32.9077 620.645 L52.2594 620.645 Q56.6199 620.645 57.8613 619.468 Q59.1026 618.258 59.1026 614.598 L59.1026 608.582 L64.0042 608.582 L64.0042 614.598 Q64.0042 621.377 61.4897 623.955 Q58.9434 626.533 52.2594 626.533 L32.9077 626.533 L32.9077 630.83 L28.3562 630.83 L28.3562 626.533 L18.2347 626.533 L18.2347 620.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M29.7248 575.226 L35.1993 575.226 Q33.8307 577.708 33.1623 580.223 Q32.4621 582.706 32.4621 585.252 Q32.4621 590.949 36.0905 594.1 Q39.6872 597.251 46.212 597.251 Q52.7369 597.251 56.3653 594.1 Q59.9619 590.949 59.9619 585.252 Q59.9619 582.706 59.2935 580.223 Q58.5933 577.708 57.2247 575.226 L62.6355 575.226 Q63.7814 577.677 64.3543 580.318 Q64.9272 582.928 64.9272 585.888 Q64.9272 593.941 59.8664 598.683 Q54.8057 603.426 46.212 603.426 Q37.491 603.426 32.4939 598.652 Q27.4968 593.846 27.4968 585.506 Q27.4968 582.801 28.0697 580.223 Q28.6108 577.645 29.7248 575.226 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M32.4621 551.227 Q32.4621 555.938 36.1542 558.675 Q39.8145 561.412 46.212 561.412 Q52.6095 561.412 56.3017 558.707 Q59.9619 555.97 59.9619 551.227 Q59.9619 546.548 56.2698 543.811 Q52.5777 541.074 46.212 541.074 Q39.8781 541.074 36.186 543.811 Q32.4621 546.548 32.4621 551.227 M27.4968 551.227 Q27.4968 543.588 32.4621 539.228 Q37.4273 534.867 46.212 534.867 Q54.9649 534.867 59.9619 539.228 Q64.9272 543.588 64.9272 551.227 Q64.9272 558.898 59.9619 563.258 Q54.9649 567.587 46.212 567.587 Q37.4273 567.587 32.4621 563.258 Q27.4968 558.898 27.4968 551.227 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M35.1993 497.405 Q31.2526 495.209 29.3747 492.153 Q27.4968 489.098 27.4968 484.96 Q27.4968 479.39 31.4117 476.366 Q35.2948 473.343 42.4881 473.343 L64.0042 473.343 L64.0042 479.231 L42.679 479.231 Q37.5546 479.231 35.072 481.045 Q32.5894 482.859 32.5894 486.583 Q32.5894 491.135 35.6131 493.777 Q38.6368 496.418 43.8567 496.418 L64.0042 496.418 L64.0042 502.307 L42.679 502.307 Q37.5228 502.307 35.072 504.121 Q32.5894 505.935 32.5894 509.723 Q32.5894 514.211 35.6449 516.852 Q38.6686 519.494 43.8567 519.494 L64.0042 519.494 L64.0042 525.382 L28.3562 525.382 L28.3562 519.494 L33.8944 519.494 Q30.616 517.489 29.0564 514.688 Q27.4968 511.887 27.4968 508.036 Q27.4968 504.153 29.4702 501.447 Q31.4436 498.71 35.1993 497.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M44.7161 431.17 L47.5806 431.17 L47.5806 458.097 Q53.6281 457.715 56.8109 454.468 Q59.9619 451.19 59.9619 445.365 Q59.9619 441.992 59.1344 438.841 Q58.3069 435.658 56.6518 432.539 L62.1899 432.539 Q63.5267 435.69 64.227 439 Q64.9272 442.31 64.9272 445.716 Q64.9272 454.246 59.9619 459.243 Q54.9967 464.208 46.5303 464.208 Q37.7774 464.208 32.6531 459.497 Q27.4968 454.755 27.4968 446.734 Q27.4968 439.541 32.1438 435.371 Q36.7589 431.17 44.7161 431.17 M42.9973 437.026 Q38.1912 437.09 35.3266 439.732 Q32.4621 442.342 32.4621 446.67 Q32.4621 451.572 35.2312 454.532 Q38.0002 457.46 43.0292 457.906 L42.9973 437.026 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M29.4065 398.832 L34.9447 398.832 Q33.6716 401.315 33.035 403.988 Q32.3984 406.662 32.3984 409.527 Q32.3984 413.887 33.7352 416.083 Q35.072 418.248 37.7456 418.248 Q39.7826 418.248 40.9603 416.688 Q42.1061 415.128 43.1565 410.418 L43.6021 408.413 Q44.9389 402.174 47.3897 399.564 Q49.8086 396.922 54.1691 396.922 Q59.1344 396.922 62.0308 400.869 Q64.9272 404.784 64.9272 411.659 Q64.9272 414.524 64.3543 417.643 Q63.8132 420.73 62.6992 424.168 L56.6518 424.168 Q58.3387 420.921 59.198 417.77 Q60.0256 414.619 60.0256 411.532 Q60.0256 407.394 58.6251 405.166 Q57.1929 402.938 54.6147 402.938 Q52.2276 402.938 50.9545 404.561 Q49.6813 406.153 48.5037 411.595 L48.0262 413.632 Q46.8804 419.075 44.5251 421.494 Q42.138 423.913 38.0002 423.913 Q32.9713 423.913 30.2341 420.348 Q27.4968 416.783 27.4968 410.227 Q27.4968 406.98 27.9743 404.116 Q28.4517 401.251 29.4065 398.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip772)" d="M323.245 1392.6 L323.245 1423.18 L514.383 1423.18 L514.383 1392.6 L323.245 1392.6 L323.245 1392.6  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1392.6 323.245,1423.18 514.383,1423.18 514.383,1392.6 323.245,1392.6 "/>
+<path clip-path="url(#clip772)" d="M514.383 1331.45 L514.383 1423.18 L705.522 1423.18 L705.522 1331.45 L514.383 1331.45 L514.383 1331.45  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="514.383,1331.45 514.383,1423.18 705.522,1423.18 705.522,1331.45 514.383,1331.45 "/>
+<path clip-path="url(#clip772)" d="M705.522 1300.87 L705.522 1423.18 L896.661 1423.18 L896.661 1300.87 L705.522 1300.87 L705.522 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="705.522,1300.87 705.522,1423.18 896.661,1423.18 896.661,1300.87 705.522,1300.87 "/>
+<path clip-path="url(#clip772)" d="M896.661 1300.87 L896.661 1423.18 L1087.8 1423.18 L1087.8 1300.87 L896.661 1300.87 L896.661 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="896.661,1300.87 896.661,1423.18 1087.8,1423.18 1087.8,1300.87 896.661,1300.87 "/>
+<path clip-path="url(#clip772)" d="M1087.8 1300.87 L1087.8 1423.18 L1278.94 1423.18 L1278.94 1300.87 L1087.8 1300.87 L1087.8 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1087.8,1300.87 1087.8,1423.18 1278.94,1423.18 1278.94,1300.87 1087.8,1300.87 "/>
+<path clip-path="url(#clip772)" d="M1278.94 1178.57 L1278.94 1423.18 L1470.08 1423.18 L1470.08 1178.57 L1278.94 1178.57 L1278.94 1178.57  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1278.94,1178.57 1278.94,1423.18 1470.08,1423.18 1470.08,1178.57 1278.94,1178.57 "/>
+<path clip-path="url(#clip772)" d="M1470.08 1117.42 L1470.08 1423.18 L1661.22 1423.18 L1661.22 1117.42 L1470.08 1117.42 L1470.08 1117.42  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.08,1117.42 1470.08,1423.18 1661.22,1423.18 1661.22,1117.42 1470.08,1117.42 "/>
+<path clip-path="url(#clip772)" d="M1661.22 1086.84 L1661.22 1423.18 L1852.35 1423.18 L1852.35 1086.84 L1661.22 1086.84 L1661.22 1086.84  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1661.22,1086.84 1661.22,1423.18 1852.35,1423.18 1852.35,1086.84 1661.22,1086.84 "/>
+<path clip-path="url(#clip772)" d="M1852.35 1117.42 L1852.35 1423.18 L2043.49 1423.18 L2043.49 1117.42 L1852.35 1117.42 L1852.35 1117.42  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1852.35,1117.42 1852.35,1423.18 2043.49,1423.18 2043.49,1117.42 1852.35,1117.42 "/>
+<path clip-path="url(#clip772)" d="M2043.49 47.2441 L2043.49 1423.18 L2234.63 1423.18 L2234.63 47.2441 L2043.49 47.2441 L2043.49 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip772)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2043.49,47.2441 2043.49,1423.18 2234.63,1423.18 2234.63,47.2441 2043.49,47.2441 "/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="418.814" cy="1392.6" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="609.953" cy="1331.45" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="801.092" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="992.23" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1183.37" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1374.51" cy="1178.57" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1565.65" cy="1117.42" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1756.79" cy="1086.84" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1947.92" cy="1117.42" r="2"/>
+<circle clip-path="url(#clip772)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2139.06" cy="47.2441" r="2"/>
+<polyline clip-path="url(#clip772)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1757.69,2799.12 1757.69,-1328.69 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1391.87,2799.12 1391.87,-1328.69 "/>
+<polyline clip-path="url(#clip772)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="469.629,2799.12 469.629,-1328.69 "/>
+<path clip-path="url(#clip770)" d="M276.709 300.469 L660.893 300.469 L660.893 93.1086 L276.709 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip770)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="276.709,300.469 660.893,300.469 660.893,93.1086 276.709,93.1086 276.709,300.469 "/>
+<polyline clip-path="url(#clip770)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,144.949 443.747,144.949 "/>
+<path clip-path="url(#clip770)" d="M475.225 140.377 L480.897 156.395 L488.998 136.303 L493.512 136.303 L483.026 162.229 L483.026 172.09 L478.767 172.09 L478.767 162.229 L471.962 142.877 Q470.943 139.96 468.744 139.96 L467.61 139.96 L467.61 136.303 L469.23 136.303 Q473.79 136.303 475.225 140.377 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M515.017 140.701 L544.693 140.701 L544.693 144.59 L515.017 144.59 L515.017 140.701 M515.017 150.145 L544.693 150.145 L544.693 154.081 L515.017 154.081 L515.017 150.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M570.665 158.293 L578.304 158.293 L578.304 131.928 L569.993 133.595 L569.993 129.335 L578.257 127.669 L582.933 127.669 L582.933 158.293 L590.572 158.293 L590.572 162.229 L570.665 162.229 L570.665 158.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M600.016 156.349 L604.901 156.349 L604.901 162.229 L600.016 162.229 L600.016 156.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M625.086 130.747 Q621.475 130.747 619.646 134.312 Q617.84 137.854 617.84 144.983 Q617.84 152.09 619.646 155.655 Q621.475 159.196 625.086 159.196 Q628.72 159.196 630.525 155.655 Q632.354 152.09 632.354 144.983 Q632.354 137.854 630.525 134.312 Q628.72 130.747 625.086 130.747 M625.086 127.044 Q630.896 127.044 633.951 131.65 Q637.03 136.233 637.03 144.983 Q637.03 153.71 633.951 158.317 Q630.896 162.9 625.086 162.9 Q619.276 162.9 616.197 158.317 Q613.141 153.71 613.141 144.983 Q613.141 136.233 616.197 131.65 Q619.276 127.044 625.086 127.044 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip770)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,196.789 443.747,196.789 "/>
+<path clip-path="url(#clip770)" d="M475.225 192.217 L480.897 208.235 L488.998 188.143 L493.512 188.143 L483.026 214.069 L483.026 223.93 L478.767 223.93 L478.767 214.069 L471.962 194.717 Q470.943 191.8 468.744 191.8 L467.61 191.8 L467.61 188.143 L469.23 188.143 Q473.79 188.143 475.225 192.217 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M515.017 192.541 L544.693 192.541 L544.693 196.43 L515.017 196.43 L515.017 192.541 M515.017 201.985 L544.693 201.985 L544.693 205.921 L515.017 205.921 L515.017 201.985 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M579.854 182.587 Q576.243 182.587 574.415 186.152 Q572.609 189.694 572.609 196.823 Q572.609 203.93 574.415 207.495 Q576.243 211.036 579.854 211.036 Q583.489 211.036 585.294 207.495 Q587.123 203.93 587.123 196.823 Q587.123 189.694 585.294 186.152 Q583.489 182.587 579.854 182.587 M579.854 178.884 Q585.665 178.884 588.72 183.49 Q591.799 188.073 591.799 196.823 Q591.799 205.55 588.72 210.157 Q585.665 214.74 579.854 214.74 Q574.044 214.74 570.966 210.157 Q567.91 205.55 567.91 196.823 Q567.91 188.073 570.966 183.49 Q574.044 178.884 579.854 178.884 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M600.016 208.189 L604.901 208.189 L604.901 214.069 L600.016 214.069 L600.016 208.189 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M615.132 179.509 L633.488 179.509 L633.488 183.444 L619.414 183.444 L619.414 191.916 Q620.433 191.569 621.451 191.407 Q622.47 191.222 623.488 191.222 Q629.275 191.222 632.655 194.393 Q636.035 197.564 636.035 202.981 Q636.035 208.559 632.563 211.661 Q629.09 214.74 622.771 214.74 Q620.595 214.74 618.326 214.37 Q616.081 213.999 613.674 213.258 L613.674 208.559 Q615.757 209.694 617.979 210.249 Q620.201 210.805 622.678 210.805 Q626.683 210.805 629.021 208.698 Q631.359 206.592 631.359 202.981 Q631.359 199.37 629.021 197.263 Q626.683 195.157 622.678 195.157 Q620.803 195.157 618.928 195.573 Q617.076 195.99 615.132 196.87 L615.132 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip770)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,248.629 443.747,248.629 "/>
+<path clip-path="url(#clip770)" d="M475.225 244.057 L480.897 260.075 L488.998 239.983 L493.512 239.983 L483.026 265.909 L483.026 275.77 L478.767 275.77 L478.767 265.909 L471.962 246.557 Q470.943 243.64 468.744 243.64 L467.61 243.64 L467.61 239.983 L469.23 239.983 Q473.79 239.983 475.225 244.057 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M515.017 244.381 L544.693 244.381 L544.693 248.27 L515.017 248.27 L515.017 244.381 M515.017 253.825 L544.693 253.825 L544.693 257.761 L515.017 257.761 L515.017 253.825 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M579.854 234.427 Q576.243 234.427 574.415 237.992 Q572.609 241.534 572.609 248.663 Q572.609 255.77 574.415 259.335 Q576.243 262.876 579.854 262.876 Q583.489 262.876 585.294 259.335 Q587.123 255.77 587.123 248.663 Q587.123 241.534 585.294 237.992 Q583.489 234.427 579.854 234.427 M579.854 230.724 Q585.665 230.724 588.72 235.33 Q591.799 239.913 591.799 248.663 Q591.799 257.39 588.72 261.997 Q585.665 266.58 579.854 266.58 Q574.044 266.58 570.966 261.997 Q567.91 257.39 567.91 248.663 Q567.91 239.913 570.966 235.33 Q574.044 230.724 579.854 230.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M600.016 260.029 L604.901 260.029 L604.901 265.909 L600.016 265.909 L600.016 260.029 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip770)" d="M625.086 234.427 Q621.475 234.427 619.646 237.992 Q617.84 241.534 617.84 248.663 Q617.84 255.77 619.646 259.335 Q621.475 262.876 625.086 262.876 Q628.72 262.876 630.525 259.335 Q632.354 255.77 632.354 248.663 Q632.354 241.534 630.525 237.992 Q628.72 234.427 625.086 234.427 M625.086 230.724 Q630.896 230.724 633.951 235.33 Q637.03 239.913 637.03 248.663 Q637.03 257.39 633.951 261.997 Q630.896 266.58 625.086 266.58 Q619.276 266.58 616.197 261.997 Q613.141 257.39 613.141 248.663 Q613.141 239.913 616.197 235.33 Q619.276 230.724 625.086 230.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/5becafc9.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/5becafc9.svg
new file mode 100644
index 00000000000..5e3b1895b90
--- /dev/null
+++ b/previews/PR3562/tutorials/applications/two_stage_stochastic/5becafc9.svg
@@ -0,0 +1,106 @@
+<?xml version="1.0" encoding="utf-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
+<defs>
+  <clipPath id="clip830">
+    <rect x="0" y="0" width="2400" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip830)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip831">
+    <rect x="480" y="0" width="1681" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip830)" d="M141.853 1486.45 L2352.76 1486.45 L2352.76 47.2441 L141.853 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip832">
+    <rect x="141" y="47" width="2212" height="1440"/>
+  </clipPath>
+</defs>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="460.226,1486.45 460.226,47.2441 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="853.765,1486.45 853.765,47.2441 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1247.3,1486.45 1247.3,47.2441 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1640.84,1486.45 1640.84,47.2441 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2034.38,1486.45 2034.38,47.2441 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="460.226,1486.45 460.226,1467.55 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="853.765,1486.45 853.765,1467.55 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1247.3,1486.45 1247.3,1467.55 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1640.84,1486.45 1640.84,1467.55 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2034.38,1486.45 2034.38,1467.55 "/>
+<path clip-path="url(#clip830)" d="M433.317 1518.36 L421.511 1536.81 L433.317 1536.81 L433.317 1518.36 M432.09 1514.29 L437.969 1514.29 L437.969 1536.81 L442.9 1536.81 L442.9 1540.7 L437.969 1540.7 L437.969 1548.85 L433.317 1548.85 L433.317 1540.7 L417.715 1540.7 L417.715 1536.19 L432.09 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M460.631 1517.37 Q457.02 1517.37 455.191 1520.93 Q453.386 1524.47 453.386 1531.6 Q453.386 1538.71 455.191 1542.27 Q457.02 1545.82 460.631 1545.82 Q464.265 1545.82 466.071 1542.27 Q467.9 1538.71 467.9 1531.6 Q467.9 1524.47 466.071 1520.93 Q464.265 1517.37 460.631 1517.37 M460.631 1513.66 Q466.441 1513.66 469.497 1518.27 Q472.576 1522.85 472.576 1531.6 Q472.576 1540.33 469.497 1544.94 Q466.441 1549.52 460.631 1549.52 Q454.821 1549.52 451.742 1544.94 Q448.687 1540.33 448.687 1531.6 Q448.687 1522.85 451.742 1518.27 Q454.821 1513.66 460.631 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M490.793 1517.37 Q487.182 1517.37 485.353 1520.93 Q483.548 1524.47 483.548 1531.6 Q483.548 1538.71 485.353 1542.27 Q487.182 1545.82 490.793 1545.82 Q494.427 1545.82 496.233 1542.27 Q498.062 1538.71 498.062 1531.6 Q498.062 1524.47 496.233 1520.93 Q494.427 1517.37 490.793 1517.37 M490.793 1513.66 Q496.603 1513.66 499.659 1518.27 Q502.737 1522.85 502.737 1531.6 Q502.737 1540.33 499.659 1544.94 Q496.603 1549.52 490.793 1549.52 Q484.983 1549.52 481.904 1544.94 Q478.849 1540.33 478.849 1531.6 Q478.849 1522.85 481.904 1518.27 Q484.983 1513.66 490.793 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M826.856 1518.36 L815.05 1536.81 L826.856 1536.81 L826.856 1518.36 M825.629 1514.29 L831.508 1514.29 L831.508 1536.81 L836.439 1536.81 L836.439 1540.7 L831.508 1540.7 L831.508 1548.85 L826.856 1548.85 L826.856 1540.7 L811.254 1540.7 L811.254 1536.19 L825.629 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M844.217 1514.29 L862.573 1514.29 L862.573 1518.22 L848.499 1518.22 L848.499 1526.7 Q849.518 1526.35 850.536 1526.19 Q851.555 1526 852.573 1526 Q858.36 1526 861.74 1529.17 Q865.119 1532.34 865.119 1537.76 Q865.119 1543.34 861.647 1546.44 Q858.175 1549.52 851.856 1549.52 Q849.68 1549.52 847.411 1549.15 Q845.166 1548.78 842.758 1548.04 L842.758 1543.34 Q844.842 1544.47 847.064 1545.03 Q849.286 1545.58 851.763 1545.58 Q855.768 1545.58 858.106 1543.48 Q860.443 1541.37 860.443 1537.76 Q860.443 1534.15 858.106 1532.04 Q855.768 1529.94 851.763 1529.94 Q849.888 1529.94 848.013 1530.35 Q846.161 1530.77 844.217 1531.65 L844.217 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M884.332 1517.37 Q880.721 1517.37 878.892 1520.93 Q877.087 1524.47 877.087 1531.6 Q877.087 1538.71 878.892 1542.27 Q880.721 1545.82 884.332 1545.82 Q887.966 1545.82 889.772 1542.27 Q891.601 1538.71 891.601 1531.6 Q891.601 1524.47 889.772 1520.93 Q887.966 1517.37 884.332 1517.37 M884.332 1513.66 Q890.142 1513.66 893.198 1518.27 Q896.277 1522.85 896.277 1531.6 Q896.277 1540.33 893.198 1544.94 Q890.142 1549.52 884.332 1549.52 Q878.522 1549.52 875.443 1544.94 Q872.388 1540.33 872.388 1531.6 Q872.388 1522.85 875.443 1518.27 Q878.522 1513.66 884.332 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1206.92 1514.29 L1225.28 1514.29 L1225.28 1518.22 L1211.21 1518.22 L1211.21 1526.7 Q1212.22 1526.35 1213.24 1526.19 Q1214.26 1526 1215.28 1526 Q1221.07 1526 1224.45 1529.17 Q1227.83 1532.34 1227.83 1537.76 Q1227.83 1543.34 1224.35 1546.44 Q1220.88 1549.52 1214.56 1549.52 Q1212.39 1549.52 1210.12 1549.15 Q1207.87 1548.78 1205.46 1548.04 L1205.46 1543.34 Q1207.55 1544.47 1209.77 1545.03 Q1211.99 1545.58 1214.47 1545.58 Q1218.47 1545.58 1220.81 1543.48 Q1223.15 1541.37 1223.15 1537.76 Q1223.15 1534.15 1220.81 1532.04 Q1218.47 1529.94 1214.47 1529.94 Q1212.59 1529.94 1210.72 1530.35 Q1208.87 1530.77 1206.92 1531.65 L1206.92 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1247.04 1517.37 Q1243.43 1517.37 1241.6 1520.93 Q1239.79 1524.47 1239.79 1531.6 Q1239.79 1538.71 1241.6 1542.27 Q1243.43 1545.82 1247.04 1545.82 Q1250.67 1545.82 1252.48 1542.27 Q1254.31 1538.71 1254.31 1531.6 Q1254.31 1524.47 1252.48 1520.93 Q1250.67 1517.37 1247.04 1517.37 M1247.04 1513.66 Q1252.85 1513.66 1255.9 1518.27 Q1258.98 1522.85 1258.98 1531.6 Q1258.98 1540.33 1255.9 1544.94 Q1252.85 1549.52 1247.04 1549.52 Q1241.23 1549.52 1238.15 1544.94 Q1235.09 1540.33 1235.09 1531.6 Q1235.09 1522.85 1238.15 1518.27 Q1241.23 1513.66 1247.04 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1277.2 1517.37 Q1273.59 1517.37 1271.76 1520.93 Q1269.95 1524.47 1269.95 1531.6 Q1269.95 1538.71 1271.76 1542.27 Q1273.59 1545.82 1277.2 1545.82 Q1280.83 1545.82 1282.64 1542.27 Q1284.47 1538.71 1284.47 1531.6 Q1284.47 1524.47 1282.64 1520.93 Q1280.83 1517.37 1277.2 1517.37 M1277.2 1513.66 Q1283.01 1513.66 1286.07 1518.27 Q1289.14 1522.85 1289.14 1531.6 Q1289.14 1540.33 1286.07 1544.94 Q1283.01 1549.52 1277.2 1549.52 Q1271.39 1549.52 1268.31 1544.94 Q1265.26 1540.33 1265.26 1531.6 Q1265.26 1522.85 1268.31 1518.27 Q1271.39 1513.66 1277.2 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1600.46 1514.29 L1618.82 1514.29 L1618.82 1518.22 L1604.74 1518.22 L1604.74 1526.7 Q1605.76 1526.35 1606.78 1526.19 Q1607.8 1526 1608.82 1526 Q1614.61 1526 1617.98 1529.17 Q1621.36 1532.34 1621.36 1537.76 Q1621.36 1543.34 1617.89 1546.44 Q1614.42 1549.52 1608.1 1549.52 Q1605.92 1549.52 1603.66 1549.15 Q1601.41 1548.78 1599 1548.04 L1599 1543.34 Q1601.09 1544.47 1603.31 1545.03 Q1605.53 1545.58 1608.01 1545.58 Q1612.01 1545.58 1614.35 1543.48 Q1616.69 1541.37 1616.69 1537.76 Q1616.69 1534.15 1614.35 1532.04 Q1612.01 1529.94 1608.01 1529.94 Q1606.13 1529.94 1604.26 1530.35 Q1602.41 1530.77 1600.46 1531.65 L1600.46 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1630.62 1514.29 L1648.98 1514.29 L1648.98 1518.22 L1634.91 1518.22 L1634.91 1526.7 Q1635.92 1526.35 1636.94 1526.19 Q1637.96 1526 1638.98 1526 Q1644.77 1526 1648.15 1529.17 Q1651.53 1532.34 1651.53 1537.76 Q1651.53 1543.34 1648.05 1546.44 Q1644.58 1549.52 1638.26 1549.52 Q1636.09 1549.52 1633.82 1549.15 Q1631.57 1548.78 1629.17 1548.04 L1629.17 1543.34 Q1631.25 1544.47 1633.47 1545.03 Q1635.69 1545.58 1638.17 1545.58 Q1642.17 1545.58 1644.51 1543.48 Q1646.85 1541.37 1646.85 1537.76 Q1646.85 1534.15 1644.51 1532.04 Q1642.17 1529.94 1638.17 1529.94 Q1636.29 1529.94 1634.42 1530.35 Q1632.57 1530.77 1630.62 1531.65 L1630.62 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1670.74 1517.37 Q1667.13 1517.37 1665.3 1520.93 Q1663.49 1524.47 1663.49 1531.6 Q1663.49 1538.71 1665.3 1542.27 Q1667.13 1545.82 1670.74 1545.82 Q1674.37 1545.82 1676.18 1542.27 Q1678.01 1538.71 1678.01 1531.6 Q1678.01 1524.47 1676.18 1520.93 Q1674.37 1517.37 1670.74 1517.37 M1670.74 1513.66 Q1676.55 1513.66 1679.6 1518.27 Q1682.68 1522.85 1682.68 1531.6 Q1682.68 1540.33 1679.6 1544.94 Q1676.55 1549.52 1670.74 1549.52 Q1664.93 1549.52 1661.85 1544.94 Q1658.79 1540.33 1658.79 1531.6 Q1658.79 1522.85 1661.85 1518.27 Q1664.93 1513.66 1670.74 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M2004.71 1529.7 Q2001.56 1529.7 1999.71 1531.86 Q1997.88 1534.01 1997.88 1537.76 Q1997.88 1541.49 1999.71 1543.66 Q2001.56 1545.82 2004.71 1545.82 Q2007.86 1545.82 2009.68 1543.66 Q2011.54 1541.49 2011.54 1537.76 Q2011.54 1534.01 2009.68 1531.86 Q2007.86 1529.7 2004.71 1529.7 M2013.99 1515.05 L2013.99 1519.31 Q2012.23 1518.48 2010.42 1518.04 Q2008.64 1517.6 2006.88 1517.6 Q2002.25 1517.6 1999.8 1520.72 Q1997.37 1523.85 1997.02 1530.17 Q1998.39 1528.15 2000.45 1527.09 Q2002.51 1526 2004.98 1526 Q2010.19 1526 2013.2 1529.17 Q2016.23 1532.32 2016.23 1537.76 Q2016.23 1543.08 2013.09 1546.3 Q2009.94 1549.52 2004.71 1549.52 Q1998.71 1549.52 1995.54 1544.94 Q1992.37 1540.33 1992.37 1531.6 Q1992.37 1523.41 1996.26 1518.55 Q2000.15 1513.66 2006.7 1513.66 Q2008.46 1513.66 2010.24 1514.01 Q2012.04 1514.36 2013.99 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M2034.29 1517.37 Q2030.68 1517.37 2028.85 1520.93 Q2027.04 1524.47 2027.04 1531.6 Q2027.04 1538.71 2028.85 1542.27 Q2030.68 1545.82 2034.29 1545.82 Q2037.92 1545.82 2039.73 1542.27 Q2041.56 1538.71 2041.56 1531.6 Q2041.56 1524.47 2039.73 1520.93 Q2037.92 1517.37 2034.29 1517.37 M2034.29 1513.66 Q2040.1 1513.66 2043.16 1518.27 Q2046.23 1522.85 2046.23 1531.6 Q2046.23 1540.33 2043.16 1544.94 Q2040.1 1549.52 2034.29 1549.52 Q2028.48 1549.52 2025.4 1544.94 Q2022.35 1540.33 2022.35 1531.6 Q2022.35 1522.85 2025.4 1518.27 Q2028.48 1513.66 2034.29 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M2064.45 1517.37 Q2060.84 1517.37 2059.01 1520.93 Q2057.21 1524.47 2057.21 1531.6 Q2057.21 1538.71 2059.01 1542.27 Q2060.84 1545.82 2064.45 1545.82 Q2068.09 1545.82 2069.89 1542.27 Q2071.72 1538.71 2071.72 1531.6 Q2071.72 1524.47 2069.89 1520.93 Q2068.09 1517.37 2064.45 1517.37 M2064.45 1513.66 Q2070.26 1513.66 2073.32 1518.27 Q2076.4 1522.85 2076.4 1531.6 Q2076.4 1540.33 2073.32 1544.94 Q2070.26 1549.52 2064.45 1549.52 Q2058.64 1549.52 2055.56 1544.94 Q2052.51 1540.33 2052.51 1531.6 Q2052.51 1522.85 2055.56 1518.27 Q2058.64 1513.66 2064.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,1122.09 2352.76,1122.09 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,757.737 2352.76,757.737 "/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,393.382 2352.76,393.382 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 141.853,47.2441 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 160.751,1486.45 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1122.09 160.751,1122.09 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,757.737 160.751,757.737 "/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,393.382 160.751,393.382 "/>
+<path clip-path="url(#clip830)" d="M93.9086 1472.25 Q90.2975 1472.25 88.4688 1475.81 Q86.6632 1479.35 86.6632 1486.48 Q86.6632 1493.59 88.4688 1497.15 Q90.2975 1500.7 93.9086 1500.7 Q97.5428 1500.7 99.3483 1497.15 Q101.177 1493.59 101.177 1486.48 Q101.177 1479.35 99.3483 1475.81 Q97.5428 1472.25 93.9086 1472.25 M93.9086 1468.54 Q99.7187 1468.54 102.774 1473.15 Q105.853 1477.73 105.853 1486.48 Q105.853 1495.21 102.774 1499.82 Q99.7187 1504.4 93.9086 1504.4 Q88.0984 1504.4 85.0197 1499.82 Q81.9642 1495.21 81.9642 1486.48 Q81.9642 1477.73 85.0197 1473.15 Q88.0984 1468.54 93.9086 1468.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M57.7745 1135.44 L74.0939 1135.44 L74.0939 1139.37 L52.1495 1139.37 L52.1495 1135.44 Q54.8115 1132.68 59.3949 1128.05 Q64.0013 1123.4 65.1819 1122.06 Q67.4272 1119.53 68.3068 1117.8 Q69.2096 1116.04 69.2096 1114.35 Q69.2096 1111.59 67.2652 1109.86 Q65.3439 1108.12 62.2421 1108.12 Q60.043 1108.12 57.5893 1108.89 Q55.1588 1109.65 52.381 1111.2 L52.381 1106.48 Q55.2051 1105.34 57.6588 1104.77 Q60.1124 1104.19 62.1495 1104.19 Q67.5198 1104.19 70.7142 1106.87 Q73.9087 1109.56 73.9087 1114.05 Q73.9087 1116.18 73.0985 1118.1 Q72.3115 1120 70.205 1122.59 Q69.6263 1123.26 66.5245 1126.48 Q63.4226 1129.67 57.7745 1135.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M93.9086 1107.89 Q90.2975 1107.89 88.4688 1111.46 Q86.6632 1115 86.6632 1122.13 Q86.6632 1129.23 88.4688 1132.8 Q90.2975 1136.34 93.9086 1136.34 Q97.5428 1136.34 99.3483 1132.8 Q101.177 1129.23 101.177 1122.13 Q101.177 1115 99.3483 1111.46 Q97.5428 1107.89 93.9086 1107.89 M93.9086 1104.19 Q99.7187 1104.19 102.774 1108.79 Q105.853 1113.38 105.853 1122.13 Q105.853 1130.85 102.774 1135.46 Q99.7187 1140.04 93.9086 1140.04 Q88.0984 1140.04 85.0197 1135.46 Q81.9642 1130.85 81.9642 1122.13 Q81.9642 1113.38 85.0197 1108.79 Q88.0984 1104.19 93.9086 1104.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M66.5939 744.531 L54.7884 762.98 L66.5939 762.98 L66.5939 744.531 M65.367 740.457 L71.2466 740.457 L71.2466 762.98 L76.1772 762.98 L76.1772 766.869 L71.2466 766.869 L71.2466 775.017 L66.5939 775.017 L66.5939 766.869 L50.9921 766.869 L50.9921 762.355 L65.367 740.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M93.9086 743.536 Q90.2975 743.536 88.4688 747.101 Q86.6632 750.642 86.6632 757.772 Q86.6632 764.878 88.4688 768.443 Q90.2975 771.985 93.9086 771.985 Q97.5428 771.985 99.3483 768.443 Q101.177 764.878 101.177 757.772 Q101.177 750.642 99.3483 747.101 Q97.5428 743.536 93.9086 743.536 M93.9086 739.832 Q99.7187 739.832 102.774 744.439 Q105.853 749.022 105.853 757.772 Q105.853 766.499 102.774 771.105 Q99.7187 775.688 93.9086 775.688 Q88.0984 775.688 85.0197 771.105 Q81.9642 766.499 81.9642 757.772 Q81.9642 749.022 85.0197 744.439 Q88.0984 739.832 93.9086 739.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M64.3254 391.518 Q61.1773 391.518 59.3254 393.671 Q57.4967 395.824 57.4967 399.574 Q57.4967 403.301 59.3254 405.477 Q61.1773 407.629 64.3254 407.629 Q67.4735 407.629 69.3022 405.477 Q71.1541 403.301 71.1541 399.574 Q71.1541 395.824 69.3022 393.671 Q67.4735 391.518 64.3254 391.518 M73.6077 376.866 L73.6077 381.125 Q71.8485 380.292 70.0429 379.852 Q68.2606 379.412 66.5013 379.412 Q61.8717 379.412 59.418 382.537 Q56.9875 385.662 56.6402 391.981 Q58.006 389.967 60.0662 388.903 Q62.1263 387.815 64.6032 387.815 Q69.8115 387.815 72.8207 390.986 Q75.8531 394.134 75.8531 399.574 Q75.8531 404.898 72.705 408.115 Q69.5568 411.333 64.3254 411.333 Q58.33 411.333 55.1588 406.75 Q51.9875 402.143 51.9875 393.416 Q51.9875 385.222 55.8764 380.361 Q59.7652 375.477 66.3161 375.477 Q68.0754 375.477 69.8578 375.824 Q71.6633 376.171 73.6077 376.866 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M93.9086 379.18 Q90.2975 379.18 88.4688 382.745 Q86.6632 386.287 86.6632 393.416 Q86.6632 400.523 88.4688 404.088 Q90.2975 407.629 93.9086 407.629 Q97.5428 407.629 99.3483 404.088 Q101.177 400.523 101.177 393.416 Q101.177 386.287 99.3483 382.745 Q97.5428 379.18 93.9086 379.18 M93.9086 375.477 Q99.7187 375.477 102.774 380.083 Q105.853 384.666 105.853 393.416 Q105.853 402.143 102.774 406.75 Q99.7187 411.333 93.9086 411.333 Q88.0984 411.333 85.0197 406.75 Q81.9642 402.143 81.9642 393.416 Q81.9642 384.666 85.0197 380.083 Q88.0984 375.477 93.9086 375.477 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip832)" d="M263.457 1468.23 L263.457 1486.45 L460.226 1486.45 L460.226 1468.23 L263.457 1468.23 L263.457 1468.23  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="263.457,1468.23 263.457,1486.45 460.226,1486.45 460.226,1468.23 263.457,1468.23 "/>
+<path clip-path="url(#clip832)" d="M460.226 1431.79 L460.226 1486.45 L656.996 1486.45 L656.996 1431.79 L460.226 1431.79 L460.226 1431.79  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="460.226,1431.79 460.226,1486.45 656.996,1486.45 656.996,1431.79 460.226,1431.79 "/>
+<path clip-path="url(#clip832)" d="M656.996 1413.58 L656.996 1486.45 L853.765 1486.45 L853.765 1413.58 L656.996 1413.58 L656.996 1413.58  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="656.996,1413.58 656.996,1486.45 853.765,1486.45 853.765,1413.58 656.996,1413.58 "/>
+<path clip-path="url(#clip832)" d="M853.765 1413.58 L853.765 1486.45 L1050.53 1486.45 L1050.53 1413.58 L853.765 1413.58 L853.765 1413.58  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="853.765,1413.58 853.765,1486.45 1050.53,1486.45 1050.53,1413.58 853.765,1413.58 "/>
+<path clip-path="url(#clip832)" d="M1050.53 1413.58 L1050.53 1486.45 L1247.3 1486.45 L1247.3 1413.58 L1050.53 1413.58 L1050.53 1413.58  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1050.53,1413.58 1050.53,1486.45 1247.3,1486.45 1247.3,1413.58 1050.53,1413.58 "/>
+<path clip-path="url(#clip832)" d="M1247.3 1340.71 L1247.3 1486.45 L1444.07 1486.45 L1444.07 1340.71 L1247.3 1340.71 L1247.3 1340.71  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1247.3,1340.71 1247.3,1486.45 1444.07,1486.45 1444.07,1340.71 1247.3,1340.71 "/>
+<path clip-path="url(#clip832)" d="M1444.07 1304.27 L1444.07 1486.45 L1640.84 1486.45 L1640.84 1304.27 L1444.07 1304.27 L1444.07 1304.27  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1444.07,1304.27 1444.07,1486.45 1640.84,1486.45 1640.84,1304.27 1444.07,1304.27 "/>
+<path clip-path="url(#clip832)" d="M1640.84 1286.05 L1640.84 1486.45 L1837.61 1486.45 L1837.61 1286.05 L1640.84 1286.05 L1640.84 1286.05  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1640.84,1286.05 1640.84,1486.45 1837.61,1486.45 1837.61,1286.05 1640.84,1286.05 "/>
+<path clip-path="url(#clip832)" d="M1837.61 1304.27 L1837.61 1486.45 L2034.38 1486.45 L2034.38 1304.27 L1837.61 1304.27 L1837.61 1304.27  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1837.61,1304.27 1837.61,1486.45 2034.38,1486.45 2034.38,1304.27 1837.61,1304.27 "/>
+<path clip-path="url(#clip832)" d="M2034.38 666.648 L2034.38 1486.45 L2231.15 1486.45 L2231.15 666.648 L2034.38 666.648 L2034.38 666.648  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2034.38,666.648 2034.38,1486.45 2231.15,1486.45 2231.15,666.648 2034.38,666.648 "/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="361.841" cy="1468.23" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="558.611" cy="1431.79" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="755.38" cy="1413.58" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="952.15" cy="1413.58" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1148.92" cy="1413.58" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1345.69" cy="1340.71" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1542.46" cy="1304.27" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1739.23" cy="1286.05" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1936" cy="1304.27" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2132.77" cy="666.648" r="2"/>
+<path clip-path="url(#clip832)" d="M263.457 1486.45 L263.457 1486.45 L460.226 1486.45 L460.226 1486.45 L263.457 1486.45 L263.457 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="263.457,1486.45 263.457,1486.45 460.226,1486.45 263.457,1486.45 "/>
+<path clip-path="url(#clip832)" d="M460.226 1486.45 L460.226 1486.45 L656.996 1486.45 L656.996 1486.45 L460.226 1486.45 L460.226 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="460.226,1486.45 460.226,1486.45 656.996,1486.45 460.226,1486.45 "/>
+<path clip-path="url(#clip832)" d="M656.996 1431.79 L656.996 1486.45 L853.765 1486.45 L853.765 1431.79 L656.996 1431.79 L656.996 1431.79  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="656.996,1431.79 656.996,1486.45 853.765,1486.45 853.765,1431.79 656.996,1431.79 "/>
+<path clip-path="url(#clip832)" d="M853.765 1431.79 L853.765 1486.45 L1050.53 1486.45 L1050.53 1431.79 L853.765 1431.79 L853.765 1431.79  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="853.765,1431.79 853.765,1486.45 1050.53,1486.45 1050.53,1431.79 853.765,1431.79 "/>
+<path clip-path="url(#clip832)" d="M1050.53 1395.36 L1050.53 1486.45 L1247.3 1486.45 L1247.3 1395.36 L1050.53 1395.36 L1050.53 1395.36  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1050.53,1395.36 1050.53,1486.45 1247.3,1486.45 1247.3,1395.36 1050.53,1395.36 "/>
+<path clip-path="url(#clip832)" d="M1247.3 1450.01 L1247.3 1486.45 L1444.07 1486.45 L1444.07 1450.01 L1247.3 1450.01 L1247.3 1450.01  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1247.3,1450.01 1247.3,1486.45 1444.07,1486.45 1444.07,1450.01 1247.3,1450.01 "/>
+<path clip-path="url(#clip832)" d="M1444.07 1340.71 L1444.07 1486.45 L1640.84 1486.45 L1640.84 1340.71 L1444.07 1340.71 L1444.07 1340.71  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1444.07,1340.71 1444.07,1486.45 1640.84,1486.45 1640.84,1340.71 1444.07,1340.71 "/>
+<path clip-path="url(#clip832)" d="M1640.84 47.2441 L1640.84 1486.45 L1837.61 1486.45 L1837.61 47.2441 L1640.84 47.2441 L1640.84 47.2441  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1640.84,47.2441 1640.84,1486.45 1837.61,1486.45 1837.61,47.2441 1640.84,47.2441 "/>
+<path clip-path="url(#clip832)" d="M1837.61 1486.45 L1837.61 1486.45 L2034.38 1486.45 L2034.38 1486.45 L1837.61 1486.45 L1837.61 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1837.61,1486.45 1837.61,1486.45 2034.38,1486.45 1837.61,1486.45 "/>
+<path clip-path="url(#clip832)" d="M2034.38 1486.45 L2034.38 1486.45 L2231.15 1486.45 L2231.15 1486.45 L2034.38 1486.45 L2034.38 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="2034.38,1486.45 2034.38,1486.45 2231.15,1486.45 2034.38,1486.45 "/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="361.841" cy="1486.45" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="558.611" cy="1486.45" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="755.38" cy="1431.79" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="952.15" cy="1431.79" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1148.92" cy="1395.36" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1345.69" cy="1450.01" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1542.46" cy="1340.71" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1739.23" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1936" cy="1486.45" r="2"/>
+<circle clip-path="url(#clip832)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2132.77" cy="1486.45" r="2"/>
+<path clip-path="url(#clip830)" d="M215.55 250.738 L709.74 250.738 L709.74 95.2176 L215.55 95.2176  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="215.55,250.738 709.74,250.738 709.74,95.2176 215.55,95.2176 215.55,250.738 "/>
+<path clip-path="url(#clip830)" d="M240.115 167.794 L387.509 167.794 L387.509 126.322 L240.115 126.322 L240.115 167.794  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="240.115,167.794 387.509,167.794 387.509,126.322 240.115,126.322 240.115,167.794 "/>
+<path clip-path="url(#clip830)" d="M412.074 129.778 L433.926 129.778 L433.926 133.713 L416.75 133.713 L416.75 143.944 L433.209 143.944 L433.209 147.879 L416.75 147.879 L416.75 160.402 L434.343 160.402 L434.343 164.338 L412.074 164.338 L412.074 129.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M463.394 138.412 L454.019 151.027 L463.88 164.338 L458.857 164.338 L451.31 154.152 L443.764 164.338 L438.741 164.338 L448.81 150.773 L439.597 138.412 L444.621 138.412 L451.496 147.648 L458.37 138.412 L463.394 138.412 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M474.019 160.449 L474.019 174.199 L469.736 174.199 L469.736 138.412 L474.019 138.412 L474.019 142.347 Q475.361 140.032 477.398 138.921 Q479.458 137.787 482.306 137.787 Q487.028 137.787 489.968 141.537 Q492.93 145.287 492.93 151.398 Q492.93 157.509 489.968 161.259 Q487.028 165.009 482.306 165.009 Q479.458 165.009 477.398 163.898 Q475.361 162.763 474.019 160.449 M488.509 151.398 Q488.509 146.699 486.565 144.037 Q484.643 141.352 481.264 141.352 Q477.884 141.352 475.94 144.037 Q474.019 146.699 474.019 151.398 Q474.019 156.097 475.94 158.782 Q477.884 161.444 481.264 161.444 Q484.643 161.444 486.565 158.782 Q488.509 156.097 488.509 151.398 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M522.166 150.31 L522.166 152.393 L502.583 152.393 Q502.861 156.791 505.222 159.106 Q507.606 161.398 511.842 161.398 Q514.296 161.398 516.588 160.796 Q518.903 160.194 521.171 158.99 L521.171 163.018 Q518.879 163.99 516.472 164.5 Q514.065 165.009 511.588 165.009 Q505.384 165.009 501.75 161.398 Q498.139 157.787 498.139 151.629 Q498.139 145.264 501.565 141.537 Q505.014 137.787 510.847 137.787 Q516.078 137.787 519.111 141.166 Q522.166 144.523 522.166 150.31 M517.907 149.06 Q517.861 145.565 515.94 143.481 Q514.041 141.398 510.893 141.398 Q507.329 141.398 505.176 143.412 Q503.046 145.426 502.722 149.083 L517.907 149.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M547.814 139.407 L547.814 143.389 Q546.009 142.393 544.18 141.907 Q542.375 141.398 540.523 141.398 Q536.379 141.398 534.088 144.037 Q531.796 146.652 531.796 151.398 Q531.796 156.143 534.088 158.782 Q536.379 161.398 540.523 161.398 Q542.375 161.398 544.18 160.912 Q546.009 160.402 547.814 159.407 L547.814 163.342 Q546.032 164.176 544.111 164.592 Q542.213 165.009 540.06 165.009 Q534.203 165.009 530.754 161.328 Q527.305 157.648 527.305 151.398 Q527.305 145.055 530.777 141.421 Q534.273 137.787 540.338 137.787 Q542.305 137.787 544.18 138.203 Q546.055 138.597 547.814 139.407 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M559.435 131.051 L559.435 138.412 L568.208 138.412 L568.208 141.722 L559.435 141.722 L559.435 155.796 Q559.435 158.967 560.291 159.87 Q561.171 160.773 563.833 160.773 L568.208 160.773 L568.208 164.338 L563.833 164.338 Q558.902 164.338 557.027 162.509 Q555.152 160.657 555.152 155.796 L555.152 141.722 L552.027 141.722 L552.027 138.412 L555.152 138.412 L555.152 131.051 L559.435 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M585.592 151.305 Q580.43 151.305 578.439 152.486 Q576.449 153.666 576.449 156.514 Q576.449 158.782 577.93 160.125 Q579.435 161.444 582.004 161.444 Q585.546 161.444 587.675 158.944 Q589.828 156.421 589.828 152.254 L589.828 151.305 L585.592 151.305 M594.087 149.546 L594.087 164.338 L589.828 164.338 L589.828 160.402 Q588.37 162.763 586.194 163.898 Q584.018 165.009 580.87 165.009 Q576.888 165.009 574.527 162.787 Q572.189 160.541 572.189 156.791 Q572.189 152.416 575.106 150.194 Q578.046 147.972 583.856 147.972 L589.828 147.972 L589.828 147.555 Q589.828 144.615 587.884 143.018 Q585.962 141.398 582.467 141.398 Q580.245 141.398 578.138 141.93 Q576.032 142.463 574.087 143.527 L574.087 139.592 Q576.425 138.69 578.624 138.25 Q580.824 137.787 582.907 137.787 Q588.532 137.787 591.31 140.703 Q594.087 143.62 594.087 149.546 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M607.073 131.051 L607.073 138.412 L615.846 138.412 L615.846 141.722 L607.073 141.722 L607.073 155.796 Q607.073 158.967 607.93 159.87 Q608.809 160.773 611.471 160.773 L615.846 160.773 L615.846 164.338 L611.471 164.338 Q606.541 164.338 604.666 162.509 Q602.791 160.657 602.791 155.796 L602.791 141.722 L599.666 141.722 L599.666 138.412 L602.791 138.412 L602.791 131.051 L607.073 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M621.448 138.412 L625.708 138.412 L625.708 164.338 L621.448 164.338 L621.448 138.412 M621.448 128.319 L625.708 128.319 L625.708 133.713 L621.448 133.713 L621.448 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M644.666 141.398 Q641.24 141.398 639.249 144.083 Q637.258 146.745 637.258 151.398 Q637.258 156.051 639.226 158.736 Q641.217 161.398 644.666 161.398 Q648.069 161.398 650.059 158.713 Q652.05 156.027 652.05 151.398 Q652.05 146.791 650.059 144.106 Q648.069 141.398 644.666 141.398 M644.666 137.787 Q650.221 137.787 653.393 141.398 Q656.564 145.009 656.564 151.398 Q656.564 157.764 653.393 161.398 Q650.221 165.009 644.666 165.009 Q639.087 165.009 635.916 161.398 Q632.768 157.764 632.768 151.398 Q632.768 145.009 635.916 141.398 Q639.087 137.787 644.666 137.787 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M685.175 148.689 L685.175 164.338 L680.916 164.338 L680.916 148.828 Q680.916 145.148 679.48 143.319 Q678.045 141.49 675.175 141.49 Q671.726 141.49 669.735 143.69 Q667.744 145.889 667.744 149.685 L667.744 164.338 L663.462 164.338 L663.462 138.412 L667.744 138.412 L667.744 142.44 Q669.272 140.102 671.332 138.944 Q673.416 137.787 676.124 137.787 Q680.591 137.787 682.883 140.565 Q685.175 143.319 685.175 148.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M240.115 219.634 L387.509 219.634 L387.509 178.162 L240.115 178.162 L240.115 219.634  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
+<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="240.115,219.634 387.509,219.634 387.509,178.162 240.115,178.162 240.115,219.634 "/>
+<path clip-path="url(#clip830)" d="M439.945 184.28 L439.945 189.21 Q437.584 187.011 434.898 185.923 Q432.236 184.835 429.227 184.835 Q423.301 184.835 420.153 188.469 Q417.005 192.08 417.005 198.932 Q417.005 205.761 420.153 209.395 Q423.301 213.006 429.227 213.006 Q432.236 213.006 434.898 211.918 Q437.584 210.83 439.945 208.631 L439.945 213.516 Q437.491 215.182 434.736 216.016 Q432.005 216.849 428.949 216.849 Q421.102 216.849 416.588 212.057 Q412.074 207.242 412.074 198.932 Q412.074 190.599 416.588 185.807 Q421.102 180.993 428.949 180.993 Q432.051 180.993 434.783 181.826 Q437.537 182.636 439.945 184.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M456.079 216.178 L442.884 181.618 L447.769 181.618 L458.718 210.715 L469.69 181.618 L474.551 181.618 L461.38 216.178 L456.079 216.178 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M492.583 203.747 Q492.583 207.057 494.227 208.955 Q495.87 210.83 498.741 210.83 Q501.588 210.83 503.208 208.932 Q504.852 207.034 504.852 203.747 Q504.852 200.506 503.185 198.608 Q501.518 196.687 498.694 196.687 Q495.893 196.687 494.227 198.585 Q492.583 200.483 492.583 203.747 M505.199 210.668 Q503.81 212.451 502.004 213.307 Q500.222 214.141 497.838 214.141 Q493.856 214.141 491.356 211.27 Q488.88 208.377 488.88 203.747 Q488.88 199.117 491.38 196.224 Q493.88 193.33 497.838 193.33 Q500.222 193.33 502.028 194.21 Q503.833 195.067 505.199 196.826 L505.199 193.793 L508.509 193.793 L508.509 210.83 Q511.889 210.321 513.787 207.752 Q515.708 205.159 515.708 201.062 Q515.708 198.585 514.967 196.409 Q514.25 194.233 512.768 192.381 Q510.361 189.349 506.889 187.752 Q503.44 186.131 499.366 186.131 Q496.518 186.131 493.903 186.895 Q491.287 187.636 489.065 189.118 Q485.431 191.479 483.37 195.321 Q481.333 199.141 481.333 203.608 Q481.333 207.289 482.653 210.506 Q483.995 213.724 486.518 216.178 Q488.949 218.585 492.143 219.835 Q495.338 221.108 498.972 221.108 Q501.958 221.108 504.829 220.09 Q507.722 219.094 510.129 217.219 L512.213 219.789 Q509.319 222.034 505.893 223.215 Q502.491 224.418 498.972 224.418 Q494.69 224.418 490.893 222.89 Q487.097 221.386 484.134 218.492 Q481.171 215.599 479.62 211.803 Q478.069 207.983 478.069 203.608 Q478.069 199.395 479.643 195.576 Q481.218 191.756 484.134 188.863 Q487.12 185.923 491.032 184.372 Q494.944 182.798 499.319 182.798 Q504.227 182.798 508.416 184.812 Q512.629 186.826 515.477 190.53 Q517.213 192.798 518.116 195.46 Q519.041 198.122 519.041 200.969 Q519.041 207.057 515.361 210.576 Q511.68 214.094 505.199 214.233 L505.199 210.668 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M543.393 199.974 Q544.898 200.483 546.31 202.15 Q547.745 203.817 549.18 206.733 L553.926 216.178 L548.902 216.178 L544.481 207.312 Q542.768 203.84 541.148 202.705 Q539.551 201.571 536.773 201.571 L531.68 201.571 L531.68 216.178 L527.004 216.178 L527.004 181.618 L537.56 181.618 Q543.486 181.618 546.402 184.094 Q549.319 186.571 549.319 191.571 Q549.319 194.835 547.791 196.988 Q546.287 199.141 543.393 199.974 M531.68 185.46 L531.68 197.729 L537.56 197.729 Q540.939 197.729 542.652 196.178 Q544.389 194.604 544.389 191.571 Q544.389 188.539 542.652 187.011 Q540.939 185.46 537.56 185.46 L531.68 185.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/60f511d7.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/60f511d7.svg
deleted file mode 100644
index 89de1728510..00000000000
--- a/previews/PR3562/tutorials/applications/two_stage_stochastic/60f511d7.svg
+++ /dev/null
@@ -1,100 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
-<defs>
-  <clipPath id="clip010">
-    <rect x="0" y="0" width="2400" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip010)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip011">
-    <rect x="480" y="0" width="1681" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip010)" d="M141.853 1486.45 L2352.76 1486.45 L2352.76 47.2441 L141.853 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip012">
-    <rect x="141" y="47" width="2212" height="1440"/>
-  </clipPath>
-</defs>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="263.457,1486.45 263.457,47.2441 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="700.722,1486.45 700.722,47.2441 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1137.99,1486.45 1137.99,47.2441 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1575.25,1486.45 1575.25,47.2441 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2012.52,1486.45 2012.52,47.2441 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="263.457,1486.45 263.457,1467.55 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="700.722,1486.45 700.722,1467.55 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1137.99,1486.45 1137.99,1467.55 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1575.25,1486.45 1575.25,1467.55 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2012.52,1486.45 2012.52,1467.55 "/>
-<path clip-path="url(#clip010)" d="M236.547 1518.36 L224.741 1536.81 L236.547 1536.81 L236.547 1518.36 M235.32 1514.29 L241.2 1514.29 L241.2 1536.81 L246.13 1536.81 L246.13 1540.7 L241.2 1540.7 L241.2 1548.85 L236.547 1548.85 L236.547 1540.7 L220.945 1540.7 L220.945 1536.19 L235.32 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M263.862 1517.37 Q260.251 1517.37 258.422 1520.93 Q256.616 1524.47 256.616 1531.6 Q256.616 1538.71 258.422 1542.27 Q260.251 1545.82 263.862 1545.82 Q267.496 1545.82 269.301 1542.27 Q271.13 1538.71 271.13 1531.6 Q271.13 1524.47 269.301 1520.93 Q267.496 1517.37 263.862 1517.37 M263.862 1513.66 Q269.672 1513.66 272.727 1518.27 Q275.806 1522.85 275.806 1531.6 Q275.806 1540.33 272.727 1544.94 Q269.672 1549.52 263.862 1549.52 Q258.051 1549.52 254.973 1544.94 Q251.917 1540.33 251.917 1531.6 Q251.917 1522.85 254.973 1518.27 Q258.051 1513.66 263.862 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M294.024 1517.37 Q290.412 1517.37 288.584 1520.93 Q286.778 1524.47 286.778 1531.6 Q286.778 1538.71 288.584 1542.27 Q290.412 1545.82 294.024 1545.82 Q297.658 1545.82 299.463 1542.27 Q301.292 1538.71 301.292 1531.6 Q301.292 1524.47 299.463 1520.93 Q297.658 1517.37 294.024 1517.37 M294.024 1513.66 Q299.834 1513.66 302.889 1518.27 Q305.968 1522.85 305.968 1531.6 Q305.968 1540.33 302.889 1544.94 Q299.834 1549.52 294.024 1549.52 Q288.213 1549.52 285.135 1544.94 Q282.079 1540.33 282.079 1531.6 Q282.079 1522.85 285.135 1518.27 Q288.213 1513.66 294.024 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M673.813 1518.36 L662.007 1536.81 L673.813 1536.81 L673.813 1518.36 M672.586 1514.29 L678.465 1514.29 L678.465 1536.81 L683.396 1536.81 L683.396 1540.7 L678.465 1540.7 L678.465 1548.85 L673.813 1548.85 L673.813 1540.7 L658.211 1540.7 L658.211 1536.19 L672.586 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M691.174 1514.29 L709.53 1514.29 L709.53 1518.22 L695.456 1518.22 L695.456 1526.7 Q696.475 1526.35 697.493 1526.19 Q698.512 1526 699.53 1526 Q705.317 1526 708.697 1529.17 Q712.076 1532.34 712.076 1537.76 Q712.076 1543.34 708.604 1546.44 Q705.132 1549.52 698.813 1549.52 Q696.637 1549.52 694.368 1549.15 Q692.123 1548.78 689.715 1548.04 L689.715 1543.34 Q691.799 1544.47 694.021 1545.03 Q696.243 1545.58 698.72 1545.58 Q702.725 1545.58 705.063 1543.48 Q707.4 1541.37 707.4 1537.76 Q707.4 1534.15 705.063 1532.04 Q702.725 1529.94 698.72 1529.94 Q696.845 1529.94 694.97 1530.35 Q693.118 1530.77 691.174 1531.65 L691.174 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M731.289 1517.37 Q727.678 1517.37 725.849 1520.93 Q724.044 1524.47 724.044 1531.6 Q724.044 1538.71 725.849 1542.27 Q727.678 1545.82 731.289 1545.82 Q734.923 1545.82 736.729 1542.27 Q738.558 1538.71 738.558 1531.6 Q738.558 1524.47 736.729 1520.93 Q734.923 1517.37 731.289 1517.37 M731.289 1513.66 Q737.099 1513.66 740.155 1518.27 Q743.234 1522.85 743.234 1531.6 Q743.234 1540.33 740.155 1544.94 Q737.099 1549.52 731.289 1549.52 Q725.479 1549.52 722.4 1544.94 Q719.345 1540.33 719.345 1531.6 Q719.345 1522.85 722.4 1518.27 Q725.479 1513.66 731.289 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1097.61 1514.29 L1115.96 1514.29 L1115.96 1518.22 L1101.89 1518.22 L1101.89 1526.7 Q1102.91 1526.35 1103.93 1526.19 Q1104.94 1526 1105.96 1526 Q1111.75 1526 1115.13 1529.17 Q1118.51 1532.34 1118.51 1537.76 Q1118.51 1543.34 1115.04 1546.44 Q1111.56 1549.52 1105.25 1549.52 Q1103.07 1549.52 1100.8 1549.15 Q1098.56 1548.78 1096.15 1548.04 L1096.15 1543.34 Q1098.23 1544.47 1100.45 1545.03 Q1102.68 1545.58 1105.15 1545.58 Q1109.16 1545.58 1111.5 1543.48 Q1113.83 1541.37 1113.83 1537.76 Q1113.83 1534.15 1111.5 1532.04 Q1109.16 1529.94 1105.15 1529.94 Q1103.28 1529.94 1101.4 1530.35 Q1099.55 1530.77 1097.61 1531.65 L1097.61 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1137.72 1517.37 Q1134.11 1517.37 1132.28 1520.93 Q1130.48 1524.47 1130.48 1531.6 Q1130.48 1538.71 1132.28 1542.27 Q1134.11 1545.82 1137.72 1545.82 Q1141.36 1545.82 1143.16 1542.27 Q1144.99 1538.71 1144.99 1531.6 Q1144.99 1524.47 1143.16 1520.93 Q1141.36 1517.37 1137.72 1517.37 M1137.72 1513.66 Q1143.53 1513.66 1146.59 1518.27 Q1149.67 1522.85 1149.67 1531.6 Q1149.67 1540.33 1146.59 1544.94 Q1143.53 1549.52 1137.72 1549.52 Q1131.91 1549.52 1128.83 1544.94 Q1125.78 1540.33 1125.78 1531.6 Q1125.78 1522.85 1128.83 1518.27 Q1131.91 1513.66 1137.72 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1167.88 1517.37 Q1164.27 1517.37 1162.44 1520.93 Q1160.64 1524.47 1160.64 1531.6 Q1160.64 1538.71 1162.44 1542.27 Q1164.27 1545.82 1167.88 1545.82 Q1171.52 1545.82 1173.32 1542.27 Q1175.15 1538.71 1175.15 1531.6 Q1175.15 1524.47 1173.32 1520.93 Q1171.52 1517.37 1167.88 1517.37 M1167.88 1513.66 Q1173.69 1513.66 1176.75 1518.27 Q1179.83 1522.85 1179.83 1531.6 Q1179.83 1540.33 1176.75 1544.94 Q1173.69 1549.52 1167.88 1549.52 Q1162.07 1549.52 1158.99 1544.94 Q1155.94 1540.33 1155.94 1531.6 Q1155.94 1522.85 1158.99 1518.27 Q1162.07 1513.66 1167.88 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1534.87 1514.29 L1553.23 1514.29 L1553.23 1518.22 L1539.15 1518.22 L1539.15 1526.7 Q1540.17 1526.35 1541.19 1526.19 Q1542.21 1526 1543.23 1526 Q1549.02 1526 1552.4 1529.17 Q1555.77 1532.34 1555.77 1537.76 Q1555.77 1543.34 1552.3 1546.44 Q1548.83 1549.52 1542.51 1549.52 Q1540.33 1549.52 1538.07 1549.15 Q1535.82 1548.78 1533.41 1548.04 L1533.41 1543.34 Q1535.5 1544.47 1537.72 1545.03 Q1539.94 1545.58 1542.42 1545.58 Q1546.42 1545.58 1548.76 1543.48 Q1551.1 1541.37 1551.1 1537.76 Q1551.1 1534.15 1548.76 1532.04 Q1546.42 1529.94 1542.42 1529.94 Q1540.54 1529.94 1538.67 1530.35 Q1536.82 1530.77 1534.87 1531.65 L1534.87 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1565.03 1514.29 L1583.39 1514.29 L1583.39 1518.22 L1569.32 1518.22 L1569.32 1526.7 Q1570.33 1526.35 1571.35 1526.19 Q1572.37 1526 1573.39 1526 Q1579.18 1526 1582.56 1529.17 Q1585.94 1532.34 1585.94 1537.76 Q1585.94 1543.34 1582.46 1546.44 Q1578.99 1549.52 1572.67 1549.52 Q1570.5 1549.52 1568.23 1549.15 Q1565.98 1548.78 1563.58 1548.04 L1563.58 1543.34 Q1565.66 1544.47 1567.88 1545.03 Q1570.1 1545.58 1572.58 1545.58 Q1576.58 1545.58 1578.92 1543.48 Q1581.26 1541.37 1581.26 1537.76 Q1581.26 1534.15 1578.92 1532.04 Q1576.58 1529.94 1572.58 1529.94 Q1570.71 1529.94 1568.83 1530.35 Q1566.98 1530.77 1565.03 1531.65 L1565.03 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1605.15 1517.37 Q1601.54 1517.37 1599.71 1520.93 Q1597.9 1524.47 1597.9 1531.6 Q1597.9 1538.71 1599.71 1542.27 Q1601.54 1545.82 1605.15 1545.82 Q1608.78 1545.82 1610.59 1542.27 Q1612.42 1538.71 1612.42 1531.6 Q1612.42 1524.47 1610.59 1520.93 Q1608.78 1517.37 1605.15 1517.37 M1605.15 1513.66 Q1610.96 1513.66 1614.02 1518.27 Q1617.09 1522.85 1617.09 1531.6 Q1617.09 1540.33 1614.02 1544.94 Q1610.96 1549.52 1605.15 1549.52 Q1599.34 1549.52 1596.26 1544.94 Q1593.21 1540.33 1593.21 1531.6 Q1593.21 1522.85 1596.26 1518.27 Q1599.34 1513.66 1605.15 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1982.84 1529.7 Q1979.7 1529.7 1977.84 1531.86 Q1976.02 1534.01 1976.02 1537.76 Q1976.02 1541.49 1977.84 1543.66 Q1979.7 1545.82 1982.84 1545.82 Q1985.99 1545.82 1987.82 1543.66 Q1989.67 1541.49 1989.67 1537.76 Q1989.67 1534.01 1987.82 1531.86 Q1985.99 1529.7 1982.84 1529.7 M1992.13 1515.05 L1992.13 1519.31 Q1990.37 1518.48 1988.56 1518.04 Q1986.78 1517.6 1985.02 1517.6 Q1980.39 1517.6 1977.94 1520.72 Q1975.51 1523.85 1975.16 1530.17 Q1976.52 1528.15 1978.58 1527.09 Q1980.64 1526 1983.12 1526 Q1988.33 1526 1991.34 1529.17 Q1994.37 1532.32 1994.37 1537.76 Q1994.37 1543.08 1991.22 1546.3 Q1988.08 1549.52 1982.84 1549.52 Q1976.85 1549.52 1973.68 1544.94 Q1970.51 1540.33 1970.51 1531.6 Q1970.51 1523.41 1974.39 1518.55 Q1978.28 1513.66 1984.83 1513.66 Q1986.59 1513.66 1988.38 1514.01 Q1990.18 1514.36 1992.13 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2012.43 1517.37 Q2008.82 1517.37 2006.99 1520.93 Q2005.18 1524.47 2005.18 1531.6 Q2005.18 1538.71 2006.99 1542.27 Q2008.82 1545.82 2012.43 1545.82 Q2016.06 1545.82 2017.87 1542.27 Q2019.7 1538.71 2019.7 1531.6 Q2019.7 1524.47 2017.87 1520.93 Q2016.06 1517.37 2012.43 1517.37 M2012.43 1513.66 Q2018.24 1513.66 2021.29 1518.27 Q2024.37 1522.85 2024.37 1531.6 Q2024.37 1540.33 2021.29 1544.94 Q2018.24 1549.52 2012.43 1549.52 Q2006.62 1549.52 2003.54 1544.94 Q2000.48 1540.33 2000.48 1531.6 Q2000.48 1522.85 2003.54 1518.27 Q2006.62 1513.66 2012.43 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2042.59 1517.37 Q2038.98 1517.37 2037.15 1520.93 Q2035.34 1524.47 2035.34 1531.6 Q2035.34 1538.71 2037.15 1542.27 Q2038.98 1545.82 2042.59 1545.82 Q2046.22 1545.82 2048.03 1542.27 Q2049.86 1538.71 2049.86 1531.6 Q2049.86 1524.47 2048.03 1520.93 Q2046.22 1517.37 2042.59 1517.37 M2042.59 1513.66 Q2048.4 1513.66 2051.45 1518.27 Q2054.53 1522.85 2054.53 1531.6 Q2054.53 1540.33 2051.45 1544.94 Q2048.4 1549.52 2042.59 1549.52 Q2036.78 1549.52 2033.7 1544.94 Q2030.64 1540.33 2030.64 1531.6 Q2030.64 1522.85 2033.7 1518.27 Q2036.78 1513.66 2042.59 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,1112.63 2352.76,1112.63 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,738.81 2352.76,738.81 "/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="141.853,364.99 2352.76,364.99 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 141.853,47.2441 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1486.45 160.751,1486.45 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,1112.63 160.751,1112.63 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,738.81 160.751,738.81 "/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="141.853,364.99 160.751,364.99 "/>
-<path clip-path="url(#clip010)" d="M93.9086 1472.25 Q90.2975 1472.25 88.4688 1475.81 Q86.6632 1479.35 86.6632 1486.48 Q86.6632 1493.59 88.4688 1497.15 Q90.2975 1500.7 93.9086 1500.7 Q97.5428 1500.7 99.3483 1497.15 Q101.177 1493.59 101.177 1486.48 Q101.177 1479.35 99.3483 1475.81 Q97.5428 1472.25 93.9086 1472.25 M93.9086 1468.54 Q99.7187 1468.54 102.774 1473.15 Q105.853 1477.73 105.853 1486.48 Q105.853 1495.21 102.774 1499.82 Q99.7187 1504.4 93.9086 1504.4 Q88.0984 1504.4 85.0197 1499.82 Q81.9642 1495.21 81.9642 1486.48 Q81.9642 1477.73 85.0197 1473.15 Q88.0984 1468.54 93.9086 1468.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M57.7745 1125.97 L74.0939 1125.97 L74.0939 1129.91 L52.1495 1129.91 L52.1495 1125.97 Q54.8115 1123.22 59.3949 1118.59 Q64.0013 1113.94 65.1819 1112.59 Q67.4272 1110.07 68.3068 1108.33 Q69.2096 1106.58 69.2096 1104.89 Q69.2096 1102.13 67.2652 1100.39 Q65.3439 1098.66 62.2421 1098.66 Q60.043 1098.66 57.5893 1099.42 Q55.1588 1100.19 52.381 1101.74 L52.381 1097.02 Q55.2051 1095.88 57.6588 1095.3 Q60.1124 1094.72 62.1495 1094.72 Q67.5198 1094.72 70.7142 1097.41 Q73.9087 1100.09 73.9087 1104.58 Q73.9087 1106.71 73.0985 1108.64 Q72.3115 1110.53 70.205 1113.13 Q69.6263 1113.8 66.5245 1117.02 Q63.4226 1120.21 57.7745 1125.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M93.9086 1098.43 Q90.2975 1098.43 88.4688 1101.99 Q86.6632 1105.53 86.6632 1112.66 Q86.6632 1119.77 88.4688 1123.33 Q90.2975 1126.88 93.9086 1126.88 Q97.5428 1126.88 99.3483 1123.33 Q101.177 1119.77 101.177 1112.66 Q101.177 1105.53 99.3483 1101.99 Q97.5428 1098.43 93.9086 1098.43 M93.9086 1094.72 Q99.7187 1094.72 102.774 1099.33 Q105.853 1103.91 105.853 1112.66 Q105.853 1121.39 102.774 1126 Q99.7187 1130.58 93.9086 1130.58 Q88.0984 1130.58 85.0197 1126 Q81.9642 1121.39 81.9642 1112.66 Q81.9642 1103.91 85.0197 1099.33 Q88.0984 1094.72 93.9086 1094.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M66.5939 725.604 L54.7884 744.053 L66.5939 744.053 L66.5939 725.604 M65.367 721.53 L71.2466 721.53 L71.2466 744.053 L76.1772 744.053 L76.1772 747.941 L71.2466 747.941 L71.2466 756.09 L66.5939 756.09 L66.5939 747.941 L50.9921 747.941 L50.9921 743.428 L65.367 721.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M93.9086 724.608 Q90.2975 724.608 88.4688 728.173 Q86.6632 731.715 86.6632 738.844 Q86.6632 745.951 88.4688 749.516 Q90.2975 753.057 93.9086 753.057 Q97.5428 753.057 99.3483 749.516 Q101.177 745.951 101.177 738.844 Q101.177 731.715 99.3483 728.173 Q97.5428 724.608 93.9086 724.608 M93.9086 720.905 Q99.7187 720.905 102.774 725.511 Q105.853 730.094 105.853 738.844 Q105.853 747.571 102.774 752.178 Q99.7187 756.761 93.9086 756.761 Q88.0984 756.761 85.0197 752.178 Q81.9642 747.571 81.9642 738.844 Q81.9642 730.094 85.0197 725.511 Q88.0984 720.905 93.9086 720.905 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M64.3254 363.127 Q61.1773 363.127 59.3254 365.28 Q57.4967 367.433 57.4967 371.182 Q57.4967 374.909 59.3254 377.085 Q61.1773 379.238 64.3254 379.238 Q67.4735 379.238 69.3022 377.085 Q71.1541 374.909 71.1541 371.182 Q71.1541 367.433 69.3022 365.28 Q67.4735 363.127 64.3254 363.127 M73.6077 348.474 L73.6077 352.734 Q71.8485 351.9 70.0429 351.46 Q68.2606 351.021 66.5013 351.021 Q61.8717 351.021 59.418 354.146 Q56.9875 357.271 56.6402 363.59 Q58.006 361.576 60.0662 360.511 Q62.1263 359.423 64.6032 359.423 Q69.8115 359.423 72.8207 362.595 Q75.8531 365.743 75.8531 371.182 Q75.8531 376.507 72.705 379.724 Q69.5568 382.942 64.3254 382.942 Q58.33 382.942 55.1588 378.358 Q51.9875 373.752 51.9875 365.025 Q51.9875 356.831 55.8764 351.97 Q59.7652 347.085 66.3161 347.085 Q68.0754 347.085 69.8578 347.433 Q71.6633 347.78 73.6077 348.474 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M93.9086 350.789 Q90.2975 350.789 88.4688 354.354 Q86.6632 357.896 86.6632 365.025 Q86.6632 372.132 88.4688 375.696 Q90.2975 379.238 93.9086 379.238 Q97.5428 379.238 99.3483 375.696 Q101.177 372.132 101.177 365.025 Q101.177 357.896 99.3483 354.354 Q97.5428 350.789 93.9086 350.789 M93.9086 347.085 Q99.7187 347.085 102.774 351.692 Q105.853 356.275 105.853 365.025 Q105.853 373.752 102.774 378.358 Q99.7187 382.942 93.9086 382.942 Q88.0984 382.942 85.0197 378.358 Q81.9642 373.752 81.9642 365.025 Q81.9642 356.275 85.0197 351.692 Q88.0984 347.085 93.9086 347.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip012)" d="M263.457 1411.68 L263.457 1486.45 L482.089 1486.45 L482.089 1411.68 L263.457 1411.68 L263.457 1411.68  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="263.457,1411.68 263.457,1486.45 482.089,1486.45 482.089,1411.68 263.457,1411.68 "/>
-<path clip-path="url(#clip012)" d="M482.089 1392.99 L482.089 1486.45 L700.722 1486.45 L700.722 1392.99 L482.089 1392.99 L482.089 1392.99  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="482.089,1392.99 482.089,1486.45 700.722,1486.45 700.722,1392.99 482.089,1392.99 "/>
-<path clip-path="url(#clip012)" d="M700.722 1411.68 L700.722 1486.45 L919.355 1486.45 L919.355 1411.68 L700.722 1411.68 L700.722 1411.68  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="700.722,1411.68 700.722,1486.45 919.355,1486.45 919.355,1411.68 700.722,1411.68 "/>
-<path clip-path="url(#clip012)" d="M919.355 1336.92 L919.355 1486.45 L1137.99 1486.45 L1137.99 1336.92 L919.355 1336.92 L919.355 1336.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="919.355,1336.92 919.355,1486.45 1137.99,1486.45 1137.99,1336.92 919.355,1336.92 "/>
-<path clip-path="url(#clip012)" d="M1137.99 1374.3 L1137.99 1486.45 L1356.62 1486.45 L1356.62 1374.3 L1137.99 1374.3 L1137.99 1374.3  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1137.99,1374.3 1137.99,1486.45 1356.62,1486.45 1356.62,1374.3 1137.99,1374.3 "/>
-<path clip-path="url(#clip012)" d="M1356.62 1318.23 L1356.62 1486.45 L1575.25 1486.45 L1575.25 1318.23 L1356.62 1318.23 L1356.62 1318.23  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1356.62,1318.23 1356.62,1486.45 1575.25,1486.45 1575.25,1318.23 1356.62,1318.23 "/>
-<path clip-path="url(#clip012)" d="M1575.25 1262.16 L1575.25 1486.45 L1793.89 1486.45 L1793.89 1262.16 L1575.25 1262.16 L1575.25 1262.16  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1575.25,1262.16 1575.25,1486.45 1793.89,1486.45 1793.89,1262.16 1575.25,1262.16 "/>
-<path clip-path="url(#clip012)" d="M1793.89 1336.92 L1793.89 1486.45 L2012.52 1486.45 L2012.52 1336.92 L1793.89 1336.92 L1793.89 1336.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1793.89,1336.92 1793.89,1486.45 2012.52,1486.45 2012.52,1336.92 1793.89,1336.92 "/>
-<path clip-path="url(#clip012)" d="M2012.52 664.046 L2012.52 1486.45 L2231.15 1486.45 L2231.15 664.046 L2012.52 664.046 L2012.52 664.046  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2012.52,664.046 2012.52,1486.45 2231.15,1486.45 2231.15,664.046 2012.52,664.046 "/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="372.773" cy="1411.68" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="591.406" cy="1392.99" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="810.039" cy="1411.68" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1028.67" cy="1336.92" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1247.3" cy="1374.3" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1465.94" cy="1318.23" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1684.57" cy="1262.16" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1903.2" cy="1336.92" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2121.84" cy="664.046" r="2"/>
-<path clip-path="url(#clip012)" d="M263.457 1486.45 L263.457 1486.45 L482.089 1486.45 L482.089 1486.45 L263.457 1486.45 L263.457 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="263.457,1486.45 263.457,1486.45 482.089,1486.45 263.457,1486.45 "/>
-<path clip-path="url(#clip012)" d="M482.089 1449.07 L482.089 1486.45 L700.722 1486.45 L700.722 1449.07 L482.089 1449.07 L482.089 1449.07  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="482.089,1449.07 482.089,1486.45 700.722,1486.45 700.722,1449.07 482.089,1449.07 "/>
-<path clip-path="url(#clip012)" d="M700.722 1392.99 L700.722 1486.45 L919.355 1486.45 L919.355 1392.99 L700.722 1392.99 L700.722 1392.99  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="700.722,1392.99 700.722,1486.45 919.355,1486.45 919.355,1392.99 700.722,1392.99 "/>
-<path clip-path="url(#clip012)" d="M919.355 1411.68 L919.355 1486.45 L1137.99 1486.45 L1137.99 1411.68 L919.355 1411.68 L919.355 1411.68  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="919.355,1411.68 919.355,1486.45 1137.99,1486.45 1137.99,1411.68 919.355,1411.68 "/>
-<path clip-path="url(#clip012)" d="M1137.99 1336.92 L1137.99 1486.45 L1356.62 1486.45 L1356.62 1336.92 L1137.99 1336.92 L1137.99 1336.92  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1137.99,1336.92 1137.99,1486.45 1356.62,1486.45 1356.62,1336.92 1137.99,1336.92 "/>
-<path clip-path="url(#clip012)" d="M1356.62 1411.68 L1356.62 1486.45 L1575.25 1486.45 L1575.25 1411.68 L1356.62 1411.68 L1356.62 1411.68  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1356.62,1411.68 1356.62,1486.45 1575.25,1486.45 1575.25,1411.68 1356.62,1411.68 "/>
-<path clip-path="url(#clip012)" d="M1575.25 47.2441 L1575.25 1486.45 L1793.89 1486.45 L1793.89 47.2441 L1575.25 47.2441 L1575.25 47.2441  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1575.25,47.2441 1575.25,1486.45 1793.89,1486.45 1793.89,47.2441 1575.25,47.2441 "/>
-<path clip-path="url(#clip012)" d="M1793.89 1486.45 L1793.89 1486.45 L2012.52 1486.45 L2012.52 1486.45 L1793.89 1486.45 L1793.89 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="1793.89,1486.45 1793.89,1486.45 2012.52,1486.45 1793.89,1486.45 "/>
-<path clip-path="url(#clip012)" d="M2012.52 1486.45 L2012.52 1486.45 L2231.15 1486.45 L2231.15 1486.45 L2012.52 1486.45 L2012.52 1486.45  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:0.5; fill:none" points="2012.52,1486.45 2012.52,1486.45 2231.15,1486.45 2012.52,1486.45 "/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="372.773" cy="1486.45" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="591.406" cy="1449.07" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="810.039" cy="1392.99" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1028.67" cy="1411.68" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1247.3" cy="1336.92" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1465.94" cy="1411.68" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1684.57" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1903.2" cy="1486.45" r="2"/>
-<circle clip-path="url(#clip012)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2121.84" cy="1486.45" r="2"/>
-<path clip-path="url(#clip010)" d="M1784.87 250.738 L2279.06 250.738 L2279.06 95.2176 L1784.87 95.2176  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1784.87,250.738 2279.06,250.738 2279.06,95.2176 1784.87,95.2176 1784.87,250.738 "/>
-<path clip-path="url(#clip010)" d="M1809.43 167.794 L1956.83 167.794 L1956.83 126.322 L1809.43 126.322 L1809.43 167.794  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1809.43,167.794 1956.83,167.794 1956.83,126.322 1809.43,126.322 1809.43,167.794 "/>
-<path clip-path="url(#clip010)" d="M1981.39 129.778 L2003.24 129.778 L2003.24 133.713 L1986.07 133.713 L1986.07 143.944 L2002.53 143.944 L2002.53 147.879 L1986.07 147.879 L1986.07 160.402 L2003.66 160.402 L2003.66 164.338 L1981.39 164.338 L1981.39 129.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2032.71 138.412 L2023.34 151.027 L2033.2 164.338 L2028.18 164.338 L2020.63 154.152 L2013.08 164.338 L2008.06 164.338 L2018.13 150.773 L2008.92 138.412 L2013.94 138.412 L2020.81 147.648 L2027.69 138.412 L2032.71 138.412 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2043.34 160.449 L2043.34 174.199 L2039.05 174.199 L2039.05 138.412 L2043.34 138.412 L2043.34 142.347 Q2044.68 140.032 2046.72 138.921 Q2048.78 137.787 2051.62 137.787 Q2056.35 137.787 2059.29 141.537 Q2062.25 145.287 2062.25 151.398 Q2062.25 157.509 2059.29 161.259 Q2056.35 165.009 2051.62 165.009 Q2048.78 165.009 2046.72 163.898 Q2044.68 162.763 2043.34 160.449 M2057.83 151.398 Q2057.83 146.699 2055.88 144.037 Q2053.96 141.352 2050.58 141.352 Q2047.2 141.352 2045.26 144.037 Q2043.34 146.699 2043.34 151.398 Q2043.34 156.097 2045.26 158.782 Q2047.2 161.444 2050.58 161.444 Q2053.96 161.444 2055.88 158.782 Q2057.83 156.097 2057.83 151.398 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2091.49 150.31 L2091.49 152.393 L2071.9 152.393 Q2072.18 156.791 2074.54 159.106 Q2076.93 161.398 2081.16 161.398 Q2083.61 161.398 2085.91 160.796 Q2088.22 160.194 2090.49 158.99 L2090.49 163.018 Q2088.2 163.99 2085.79 164.5 Q2083.38 165.009 2080.91 165.009 Q2074.7 165.009 2071.07 161.398 Q2067.46 157.787 2067.46 151.629 Q2067.46 145.264 2070.88 141.537 Q2074.33 137.787 2080.17 137.787 Q2085.4 137.787 2088.43 141.166 Q2091.49 144.523 2091.49 150.31 M2087.23 149.06 Q2087.18 145.565 2085.26 143.481 Q2083.36 141.398 2080.21 141.398 Q2076.65 141.398 2074.49 143.412 Q2072.36 145.426 2072.04 149.083 L2087.23 149.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2117.13 139.407 L2117.13 143.389 Q2115.33 142.393 2113.5 141.907 Q2111.69 141.398 2109.84 141.398 Q2105.7 141.398 2103.41 144.037 Q2101.11 146.652 2101.11 151.398 Q2101.11 156.143 2103.41 158.782 Q2105.7 161.398 2109.84 161.398 Q2111.69 161.398 2113.5 160.912 Q2115.33 160.402 2117.13 159.407 L2117.13 163.342 Q2115.35 164.176 2113.43 164.592 Q2111.53 165.009 2109.38 165.009 Q2103.52 165.009 2100.07 161.328 Q2096.62 157.648 2096.62 151.398 Q2096.62 145.055 2100.1 141.421 Q2103.59 137.787 2109.66 137.787 Q2111.62 137.787 2113.5 138.203 Q2115.37 138.597 2117.13 139.407 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2128.75 131.051 L2128.75 138.412 L2137.53 138.412 L2137.53 141.722 L2128.75 141.722 L2128.75 155.796 Q2128.75 158.967 2129.61 159.87 Q2130.49 160.773 2133.15 160.773 L2137.53 160.773 L2137.53 164.338 L2133.15 164.338 Q2128.22 164.338 2126.35 162.509 Q2124.47 160.657 2124.47 155.796 L2124.47 141.722 L2121.35 141.722 L2121.35 138.412 L2124.47 138.412 L2124.47 131.051 L2128.75 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2154.91 151.305 Q2149.75 151.305 2147.76 152.486 Q2145.77 153.666 2145.77 156.514 Q2145.77 158.782 2147.25 160.125 Q2148.75 161.444 2151.32 161.444 Q2154.86 161.444 2156.99 158.944 Q2159.15 156.421 2159.15 152.254 L2159.15 151.305 L2154.91 151.305 M2163.41 149.546 L2163.41 164.338 L2159.15 164.338 L2159.15 160.402 Q2157.69 162.763 2155.51 163.898 Q2153.34 165.009 2150.19 165.009 Q2146.21 165.009 2143.85 162.787 Q2141.51 160.541 2141.51 156.791 Q2141.51 152.416 2144.42 150.194 Q2147.36 147.972 2153.17 147.972 L2159.15 147.972 L2159.15 147.555 Q2159.15 144.615 2157.2 143.018 Q2155.28 141.398 2151.79 141.398 Q2149.56 141.398 2147.46 141.93 Q2145.35 142.463 2143.41 143.527 L2143.41 139.592 Q2145.74 138.69 2147.94 138.25 Q2150.14 137.787 2152.23 137.787 Q2157.85 137.787 2160.63 140.703 Q2163.41 143.62 2163.41 149.546 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2176.39 131.051 L2176.39 138.412 L2185.17 138.412 L2185.17 141.722 L2176.39 141.722 L2176.39 155.796 Q2176.39 158.967 2177.25 159.87 Q2178.13 160.773 2180.79 160.773 L2185.17 160.773 L2185.17 164.338 L2180.79 164.338 Q2175.86 164.338 2173.98 162.509 Q2172.11 160.657 2172.11 155.796 L2172.11 141.722 L2168.98 141.722 L2168.98 138.412 L2172.11 138.412 L2172.11 131.051 L2176.39 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2190.77 138.412 L2195.03 138.412 L2195.03 164.338 L2190.77 164.338 L2190.77 138.412 M2190.77 128.319 L2195.03 128.319 L2195.03 133.713 L2190.77 133.713 L2190.77 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2213.98 141.398 Q2210.56 141.398 2208.57 144.083 Q2206.58 146.745 2206.58 151.398 Q2206.58 156.051 2208.54 158.736 Q2210.54 161.398 2213.98 161.398 Q2217.39 161.398 2219.38 158.713 Q2221.37 156.027 2221.37 151.398 Q2221.37 146.791 2219.38 144.106 Q2217.39 141.398 2213.98 141.398 M2213.98 137.787 Q2219.54 137.787 2222.71 141.398 Q2225.88 145.009 2225.88 151.398 Q2225.88 157.764 2222.71 161.398 Q2219.54 165.009 2213.98 165.009 Q2208.41 165.009 2205.23 161.398 Q2202.09 157.764 2202.09 151.398 Q2202.09 145.009 2205.23 141.398 Q2208.41 137.787 2213.98 137.787 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2254.49 148.689 L2254.49 164.338 L2250.23 164.338 L2250.23 148.828 Q2250.23 145.148 2248.8 143.319 Q2247.36 141.49 2244.49 141.49 Q2241.04 141.49 2239.05 143.69 Q2237.06 145.889 2237.06 149.685 L2237.06 164.338 L2232.78 164.338 L2232.78 138.412 L2237.06 138.412 L2237.06 142.44 Q2238.59 140.102 2240.65 138.944 Q2242.73 137.787 2245.44 137.787 Q2249.91 137.787 2252.2 140.565 Q2254.49 143.319 2254.49 148.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1809.43 219.634 L1956.83 219.634 L1956.83 178.162 L1809.43 178.162 L1809.43 219.634  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="0.5"/>
-<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1809.43,219.634 1956.83,219.634 1956.83,178.162 1809.43,178.162 1809.43,219.634 "/>
-<path clip-path="url(#clip010)" d="M2009.26 184.28 L2009.26 189.21 Q2006.9 187.011 2004.22 185.923 Q2001.56 184.835 1998.55 184.835 Q1992.62 184.835 1989.47 188.469 Q1986.32 192.08 1986.32 198.932 Q1986.32 205.761 1989.47 209.395 Q1992.62 213.006 1998.55 213.006 Q2001.56 213.006 2004.22 211.918 Q2006.9 210.83 2009.26 208.631 L2009.26 213.516 Q2006.81 215.182 2004.06 216.016 Q2001.32 216.849 1998.27 216.849 Q1990.42 216.849 1985.91 212.057 Q1981.39 207.242 1981.39 198.932 Q1981.39 190.599 1985.91 185.807 Q1990.42 180.993 1998.27 180.993 Q2001.37 180.993 2004.1 181.826 Q2006.86 182.636 2009.26 184.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2025.4 216.178 L2012.2 181.618 L2017.09 181.618 L2028.04 210.715 L2039.01 181.618 L2043.87 181.618 L2030.7 216.178 L2025.4 216.178 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2061.9 203.747 Q2061.9 207.057 2063.55 208.955 Q2065.19 210.83 2068.06 210.83 Q2070.91 210.83 2072.53 208.932 Q2074.17 207.034 2074.17 203.747 Q2074.17 200.506 2072.5 198.608 Q2070.84 196.687 2068.01 196.687 Q2065.21 196.687 2063.55 198.585 Q2061.9 200.483 2061.9 203.747 M2074.52 210.668 Q2073.13 212.451 2071.32 213.307 Q2069.54 214.141 2067.16 214.141 Q2063.18 214.141 2060.68 211.27 Q2058.2 208.377 2058.2 203.747 Q2058.2 199.117 2060.7 196.224 Q2063.2 193.33 2067.16 193.33 Q2069.54 193.33 2071.35 194.21 Q2073.15 195.067 2074.52 196.826 L2074.52 193.793 L2077.83 193.793 L2077.83 210.83 Q2081.21 210.321 2083.11 207.752 Q2085.03 205.159 2085.03 201.062 Q2085.03 198.585 2084.29 196.409 Q2083.57 194.233 2082.09 192.381 Q2079.68 189.349 2076.21 187.752 Q2072.76 186.131 2068.68 186.131 Q2065.84 186.131 2063.22 186.895 Q2060.61 187.636 2058.38 189.118 Q2054.75 191.479 2052.69 195.321 Q2050.65 199.141 2050.65 203.608 Q2050.65 207.289 2051.97 210.506 Q2053.31 213.724 2055.84 216.178 Q2058.27 218.585 2061.46 219.835 Q2064.66 221.108 2068.29 221.108 Q2071.28 221.108 2074.15 220.09 Q2077.04 219.094 2079.45 217.219 L2081.53 219.789 Q2078.64 222.034 2075.21 223.215 Q2071.81 224.418 2068.29 224.418 Q2064.01 224.418 2060.21 222.89 Q2056.42 221.386 2053.45 218.492 Q2050.49 215.599 2048.94 211.803 Q2047.39 207.983 2047.39 203.608 Q2047.39 199.395 2048.96 195.576 Q2050.54 191.756 2053.45 188.863 Q2056.44 185.923 2060.35 184.372 Q2064.26 182.798 2068.64 182.798 Q2073.55 182.798 2077.74 184.812 Q2081.95 186.826 2084.8 190.53 Q2086.53 192.798 2087.43 195.46 Q2088.36 198.122 2088.36 200.969 Q2088.36 207.057 2084.68 210.576 Q2081 214.094 2074.52 214.233 L2074.52 210.668 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2112.71 199.974 Q2114.22 200.483 2115.63 202.15 Q2117.06 203.817 2118.5 206.733 L2123.24 216.178 L2118.22 216.178 L2113.8 207.312 Q2112.09 203.84 2110.47 202.705 Q2108.87 201.571 2106.09 201.571 L2101 201.571 L2101 216.178 L2096.32 216.178 L2096.32 181.618 L2106.88 181.618 Q2112.8 181.618 2115.72 184.094 Q2118.64 186.571 2118.64 191.571 Q2118.64 194.835 2117.11 196.988 Q2115.61 199.141 2112.71 199.974 M2101 185.46 L2101 197.729 L2106.88 197.729 Q2110.26 197.729 2111.97 196.178 Q2113.71 194.604 2113.71 191.571 Q2113.71 188.539 2111.97 187.011 Q2110.26 185.46 2106.88 185.46 L2101 185.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/7add9a3e.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/7add9a3e.svg
deleted file mode 100644
index 9679afe1482..00000000000
--- a/previews/PR3562/tutorials/applications/two_stage_stochastic/7add9a3e.svg
+++ /dev/null
@@ -1,77 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
-<defs>
-  <clipPath id="clip950">
-    <rect x="0" y="0" width="2400" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip950)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip951">
-    <rect x="480" y="0" width="1681" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip950)" d="M205.121 1423.18 L2352.76 1423.18 L2352.76 47.2441 L205.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip952">
-    <rect x="205" y="47" width="2149" height="1377"/>
-  </clipPath>
-</defs>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="323.245,1423.18 323.245,47.2441 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="747.997,1423.18 747.997,47.2441 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1172.75,1423.18 1172.75,47.2441 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1597.5,1423.18 1597.5,47.2441 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2022.26,1423.18 2022.26,47.2441 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1423.18 323.245,1404.28 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.997,1423.18 747.997,1404.28 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1172.75,1423.18 1172.75,1404.28 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1597.5,1423.18 1597.5,1404.28 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2022.26,1423.18 2022.26,1404.28 "/>
-<path clip-path="url(#clip950)" d="M296.335 1455.09 L284.53 1473.54 L296.335 1473.54 L296.335 1455.09 M295.108 1451.02 L300.988 1451.02 L300.988 1473.54 L305.918 1473.54 L305.918 1477.43 L300.988 1477.43 L300.988 1485.58 L296.335 1485.58 L296.335 1477.43 L280.733 1477.43 L280.733 1472.92 L295.108 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M323.65 1454.1 Q320.039 1454.1 318.21 1457.66 Q316.404 1461.2 316.404 1468.33 Q316.404 1475.44 318.21 1479.01 Q320.039 1482.55 323.65 1482.55 Q327.284 1482.55 329.09 1479.01 Q330.918 1475.44 330.918 1468.33 Q330.918 1461.2 329.09 1457.66 Q327.284 1454.1 323.65 1454.1 M323.65 1450.39 Q329.46 1450.39 332.516 1455 Q335.594 1459.58 335.594 1468.33 Q335.594 1477.06 332.516 1481.67 Q329.46 1486.25 323.65 1486.25 Q317.84 1486.25 314.761 1481.67 Q311.705 1477.06 311.705 1468.33 Q311.705 1459.58 314.761 1455 Q317.84 1450.39 323.65 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M353.812 1454.1 Q350.201 1454.1 348.372 1457.66 Q346.566 1461.2 346.566 1468.33 Q346.566 1475.44 348.372 1479.01 Q350.201 1482.55 353.812 1482.55 Q357.446 1482.55 359.251 1479.01 Q361.08 1475.44 361.08 1468.33 Q361.08 1461.2 359.251 1457.66 Q357.446 1454.1 353.812 1454.1 M353.812 1450.39 Q359.622 1450.39 362.677 1455 Q365.756 1459.58 365.756 1468.33 Q365.756 1477.06 362.677 1481.67 Q359.622 1486.25 353.812 1486.25 Q348.002 1486.25 344.923 1481.67 Q341.867 1477.06 341.867 1468.33 Q341.867 1459.58 344.923 1455 Q348.002 1450.39 353.812 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M721.088 1455.09 L709.282 1473.54 L721.088 1473.54 L721.088 1455.09 M719.861 1451.02 L725.741 1451.02 L725.741 1473.54 L730.671 1473.54 L730.671 1477.43 L725.741 1477.43 L725.741 1485.58 L721.088 1485.58 L721.088 1477.43 L705.486 1477.43 L705.486 1472.92 L719.861 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M738.449 1451.02 L756.805 1451.02 L756.805 1454.96 L742.731 1454.96 L742.731 1463.43 Q743.75 1463.08 744.768 1462.92 Q745.787 1462.73 746.805 1462.73 Q752.592 1462.73 755.972 1465.9 Q759.352 1469.08 759.352 1474.49 Q759.352 1480.07 755.879 1483.17 Q752.407 1486.25 746.088 1486.25 Q743.912 1486.25 741.643 1485.88 Q739.398 1485.51 736.991 1484.77 L736.991 1480.07 Q739.074 1481.2 741.296 1481.76 Q743.518 1482.32 745.995 1482.32 Q750 1482.32 752.338 1480.21 Q754.676 1478.1 754.676 1474.49 Q754.676 1470.88 752.338 1468.77 Q750 1466.67 745.995 1466.67 Q744.12 1466.67 742.245 1467.08 Q740.393 1467.5 738.449 1468.38 L738.449 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M778.564 1454.1 Q774.953 1454.1 773.125 1457.66 Q771.319 1461.2 771.319 1468.33 Q771.319 1475.44 773.125 1479.01 Q774.953 1482.55 778.564 1482.55 Q782.199 1482.55 784.004 1479.01 Q785.833 1475.44 785.833 1468.33 Q785.833 1461.2 784.004 1457.66 Q782.199 1454.1 778.564 1454.1 M778.564 1450.39 Q784.375 1450.39 787.43 1455 Q790.509 1459.58 790.509 1468.33 Q790.509 1477.06 787.43 1481.67 Q784.375 1486.25 778.564 1486.25 Q772.754 1486.25 769.676 1481.67 Q766.62 1477.06 766.62 1468.33 Q766.62 1459.58 769.676 1455 Q772.754 1450.39 778.564 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1132.37 1451.02 L1150.72 1451.02 L1150.72 1454.96 L1136.65 1454.96 L1136.65 1463.43 Q1137.67 1463.08 1138.69 1462.92 Q1139.71 1462.73 1140.72 1462.73 Q1146.51 1462.73 1149.89 1465.9 Q1153.27 1469.08 1153.27 1474.49 Q1153.27 1480.07 1149.8 1483.17 Q1146.33 1486.25 1140.01 1486.25 Q1137.83 1486.25 1135.56 1485.88 Q1133.32 1485.51 1130.91 1484.77 L1130.91 1480.07 Q1132.99 1481.2 1135.22 1481.76 Q1137.44 1482.32 1139.91 1482.32 Q1143.92 1482.32 1146.26 1480.21 Q1148.6 1478.1 1148.6 1474.49 Q1148.6 1470.88 1146.26 1468.77 Q1143.92 1466.67 1139.91 1466.67 Q1138.04 1466.67 1136.16 1467.08 Q1134.31 1467.5 1132.37 1468.38 L1132.37 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1172.48 1454.1 Q1168.87 1454.1 1167.04 1457.66 Q1165.24 1461.2 1165.24 1468.33 Q1165.24 1475.44 1167.04 1479.01 Q1168.87 1482.55 1172.48 1482.55 Q1176.12 1482.55 1177.92 1479.01 Q1179.75 1475.44 1179.75 1468.33 Q1179.75 1461.2 1177.92 1457.66 Q1176.12 1454.1 1172.48 1454.1 M1172.48 1450.39 Q1178.29 1450.39 1181.35 1455 Q1184.43 1459.58 1184.43 1468.33 Q1184.43 1477.06 1181.35 1481.67 Q1178.29 1486.25 1172.48 1486.25 Q1166.67 1486.25 1163.6 1481.67 Q1160.54 1477.06 1160.54 1468.33 Q1160.54 1459.58 1163.6 1455 Q1166.67 1450.39 1172.48 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1202.65 1454.1 Q1199.03 1454.1 1197.21 1457.66 Q1195.4 1461.2 1195.4 1468.33 Q1195.4 1475.44 1197.21 1479.01 Q1199.03 1482.55 1202.65 1482.55 Q1206.28 1482.55 1208.09 1479.01 Q1209.91 1475.44 1209.91 1468.33 Q1209.91 1461.2 1208.09 1457.66 Q1206.28 1454.1 1202.65 1454.1 M1202.65 1450.39 Q1208.46 1450.39 1211.51 1455 Q1214.59 1459.58 1214.59 1468.33 Q1214.59 1477.06 1211.51 1481.67 Q1208.46 1486.25 1202.65 1486.25 Q1196.84 1486.25 1193.76 1481.67 Q1190.7 1477.06 1190.7 1468.33 Q1190.7 1459.58 1193.76 1455 Q1196.84 1450.39 1202.65 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1557.12 1451.02 L1575.48 1451.02 L1575.48 1454.96 L1561.4 1454.96 L1561.4 1463.43 Q1562.42 1463.08 1563.44 1462.92 Q1564.46 1462.73 1565.48 1462.73 Q1571.26 1462.73 1574.64 1465.9 Q1578.02 1469.08 1578.02 1474.49 Q1578.02 1480.07 1574.55 1483.17 Q1571.08 1486.25 1564.76 1486.25 Q1562.58 1486.25 1560.32 1485.88 Q1558.07 1485.51 1555.66 1484.77 L1555.66 1480.07 Q1557.75 1481.2 1559.97 1481.76 Q1562.19 1482.32 1564.67 1482.32 Q1568.67 1482.32 1571.01 1480.21 Q1573.35 1478.1 1573.35 1474.49 Q1573.35 1470.88 1571.01 1468.77 Q1568.67 1466.67 1564.67 1466.67 Q1562.79 1466.67 1560.92 1467.08 Q1559.07 1467.5 1557.12 1468.38 L1557.12 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1587.28 1451.02 L1605.64 1451.02 L1605.64 1454.96 L1591.57 1454.96 L1591.57 1463.43 Q1592.58 1463.08 1593.6 1462.92 Q1594.62 1462.73 1595.64 1462.73 Q1601.43 1462.73 1604.81 1465.9 Q1608.19 1469.08 1608.19 1474.49 Q1608.19 1480.07 1604.71 1483.17 Q1601.24 1486.25 1594.92 1486.25 Q1592.75 1486.25 1590.48 1485.88 Q1588.23 1485.51 1585.82 1484.77 L1585.82 1480.07 Q1587.91 1481.2 1590.13 1481.76 Q1592.35 1482.32 1594.83 1482.32 Q1598.83 1482.32 1601.17 1480.21 Q1603.51 1478.1 1603.51 1474.49 Q1603.51 1470.88 1601.17 1468.77 Q1598.83 1466.67 1594.83 1466.67 Q1592.95 1466.67 1591.08 1467.08 Q1589.23 1467.5 1587.28 1468.38 L1587.28 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1627.4 1454.1 Q1623.79 1454.1 1621.96 1457.66 Q1620.15 1461.2 1620.15 1468.33 Q1620.15 1475.44 1621.96 1479.01 Q1623.79 1482.55 1627.4 1482.55 Q1631.03 1482.55 1632.84 1479.01 Q1634.67 1475.44 1634.67 1468.33 Q1634.67 1461.2 1632.84 1457.66 Q1631.03 1454.1 1627.4 1454.1 M1627.4 1450.39 Q1633.21 1450.39 1636.26 1455 Q1639.34 1459.58 1639.34 1468.33 Q1639.34 1477.06 1636.26 1481.67 Q1633.21 1486.25 1627.4 1486.25 Q1621.59 1486.25 1618.51 1481.67 Q1615.45 1477.06 1615.45 1468.33 Q1615.45 1459.58 1618.51 1455 Q1621.59 1450.39 1627.4 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1992.58 1466.44 Q1989.43 1466.44 1987.58 1468.59 Q1985.75 1470.74 1985.75 1474.49 Q1985.75 1478.22 1987.58 1480.39 Q1989.43 1482.55 1992.58 1482.55 Q1995.73 1482.55 1997.56 1480.39 Q1999.41 1478.22 1999.41 1474.49 Q1999.41 1470.74 1997.56 1468.59 Q1995.73 1466.44 1992.58 1466.44 M2001.86 1451.78 L2001.86 1456.04 Q2000.1 1455.21 1998.3 1454.77 Q1996.52 1454.33 1994.76 1454.33 Q1990.13 1454.33 1987.67 1457.45 Q1985.24 1460.58 1984.89 1466.9 Q1986.26 1464.89 1988.32 1463.82 Q1990.38 1462.73 1992.86 1462.73 Q1998.07 1462.73 2001.08 1465.9 Q2004.11 1469.05 2004.11 1474.49 Q2004.11 1479.82 2000.96 1483.03 Q1997.81 1486.25 1992.58 1486.25 Q1986.58 1486.25 1983.41 1481.67 Q1980.24 1477.06 1980.24 1468.33 Q1980.24 1460.14 1984.13 1455.28 Q1988.02 1450.39 1994.57 1450.39 Q1996.33 1450.39 1998.11 1450.74 Q1999.92 1451.09 2001.86 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M2022.16 1454.1 Q2018.55 1454.1 2016.72 1457.66 Q2014.92 1461.2 2014.92 1468.33 Q2014.92 1475.44 2016.72 1479.01 Q2018.55 1482.55 2022.16 1482.55 Q2025.8 1482.55 2027.6 1479.01 Q2029.43 1475.44 2029.43 1468.33 Q2029.43 1461.2 2027.6 1457.66 Q2025.8 1454.1 2022.16 1454.1 M2022.16 1450.39 Q2027.97 1450.39 2031.03 1455 Q2034.11 1459.58 2034.11 1468.33 Q2034.11 1477.06 2031.03 1481.67 Q2027.97 1486.25 2022.16 1486.25 Q2016.35 1486.25 2013.27 1481.67 Q2010.22 1477.06 2010.22 1468.33 Q2010.22 1459.58 2013.27 1455 Q2016.35 1450.39 2022.16 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M2052.33 1454.1 Q2048.71 1454.1 2046.89 1457.66 Q2045.08 1461.2 2045.08 1468.33 Q2045.08 1475.44 2046.89 1479.01 Q2048.71 1482.55 2052.33 1482.55 Q2055.96 1482.55 2057.76 1479.01 Q2059.59 1475.44 2059.59 1468.33 Q2059.59 1461.2 2057.76 1457.66 Q2055.96 1454.1 2052.33 1454.1 M2052.33 1450.39 Q2058.14 1450.39 2061.19 1455 Q2064.27 1459.58 2064.27 1468.33 Q2064.27 1477.06 2061.19 1481.67 Q2058.14 1486.25 2052.33 1486.25 Q2046.51 1486.25 2043.44 1481.67 Q2040.38 1477.06 2040.38 1468.33 Q2040.38 1459.58 2043.44 1455 Q2046.51 1450.39 2052.33 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1149.83 1525.81 L1149.83 1543.66 L1157.91 1543.66 Q1162.4 1543.66 1164.85 1541.34 Q1167.3 1539.02 1167.3 1534.72 Q1167.3 1530.45 1164.85 1528.13 Q1162.4 1525.81 1157.91 1525.81 L1149.83 1525.81 M1143.4 1520.52 L1157.91 1520.52 Q1165.9 1520.52 1169.97 1524.15 Q1174.08 1527.75 1174.08 1534.72 Q1174.08 1541.75 1169.97 1545.35 Q1165.9 1548.95 1157.91 1548.95 L1149.83 1548.95 L1149.83 1568.04 L1143.4 1568.04 L1143.4 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1201.96 1537.87 Q1200.97 1537.3 1199.8 1537.04 Q1198.65 1536.76 1197.25 1536.76 Q1192.29 1536.76 1189.61 1540 Q1186.97 1543.22 1186.97 1549.27 L1186.97 1568.04 L1181.08 1568.04 L1181.08 1532.4 L1186.97 1532.4 L1186.97 1537.93 Q1188.82 1534.69 1191.78 1533.13 Q1194.74 1531.54 1198.97 1531.54 Q1199.57 1531.54 1200.31 1531.63 Q1201.04 1531.7 1201.93 1531.85 L1201.96 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1220.49 1536.5 Q1215.77 1536.5 1213.04 1540.19 Q1210.3 1543.85 1210.3 1550.25 Q1210.3 1556.65 1213.01 1560.34 Q1215.74 1564 1220.49 1564 Q1225.16 1564 1227.9 1560.31 Q1230.64 1556.62 1230.64 1550.25 Q1230.64 1543.92 1227.9 1540.23 Q1225.16 1536.5 1220.49 1536.5 M1220.49 1531.54 Q1228.12 1531.54 1232.48 1536.5 Q1236.85 1541.47 1236.85 1550.25 Q1236.85 1559 1232.48 1564 Q1228.12 1568.97 1220.49 1568.97 Q1212.81 1568.97 1208.45 1564 Q1204.13 1559 1204.13 1550.25 Q1204.13 1541.47 1208.45 1536.5 Q1212.81 1531.54 1220.49 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1264.6 1518.52 L1264.6 1523.39 L1259 1523.39 Q1255.85 1523.39 1254.61 1524.66 Q1253.4 1525.93 1253.4 1529.24 L1253.4 1532.4 L1263.04 1532.4 L1263.04 1536.95 L1253.4 1536.95 L1253.4 1568.04 L1247.51 1568.04 L1247.51 1536.95 L1241.91 1536.95 L1241.91 1532.4 L1247.51 1532.4 L1247.51 1529.91 Q1247.51 1523.96 1250.28 1521.26 Q1253.05 1518.52 1259.06 1518.52 L1264.6 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1269.5 1532.4 L1275.36 1532.4 L1275.36 1568.04 L1269.5 1568.04 L1269.5 1532.4 M1269.5 1518.52 L1275.36 1518.52 L1275.36 1525.93 L1269.5 1525.93 L1269.5 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1293.4 1522.27 L1293.4 1532.4 L1305.47 1532.4 L1305.47 1536.95 L1293.4 1536.95 L1293.4 1556.3 Q1293.4 1560.66 1294.58 1561.9 Q1295.79 1563.14 1299.45 1563.14 L1305.47 1563.14 L1305.47 1568.04 L1299.45 1568.04 Q1292.67 1568.04 1290.09 1565.53 Q1287.52 1562.98 1287.52 1556.3 L1287.52 1536.95 L1283.22 1536.95 L1283.22 1532.4 L1287.52 1532.4 L1287.52 1522.27 L1293.4 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1333.35 1518.52 L1346.84 1518.52 L1346.84 1523.07 L1339.21 1523.07 L1339.21 1572.09 L1346.84 1572.09 L1346.84 1576.64 L1333.35 1576.64 L1333.35 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1375.2 1577.62 L1372.02 1577.62 L1371.99 1568.04 Q1368.65 1567.98 1365.31 1567.28 Q1361.96 1566.55 1358.59 1565.12 L1358.59 1559.39 Q1361.84 1561.42 1365.15 1562.47 Q1368.49 1563.49 1372.02 1563.52 L1372.02 1549.01 Q1364.99 1547.86 1361.77 1545.13 Q1358.59 1542.39 1358.59 1537.62 Q1358.59 1532.43 1362.06 1529.44 Q1365.53 1526.44 1372.02 1526 L1372.02 1518.52 L1375.2 1518.52 L1375.2 1525.9 Q1378.16 1526.03 1380.93 1526.54 Q1383.7 1527.02 1386.34 1527.88 L1386.34 1533.45 Q1383.7 1532.11 1380.9 1531.38 Q1378.13 1530.65 1375.2 1530.52 L1375.2 1544.11 Q1382.43 1545.22 1385.83 1548.09 Q1389.24 1550.95 1389.24 1555.92 Q1389.24 1561.3 1385.61 1564.42 Q1382.02 1567.5 1375.2 1567.98 L1375.2 1577.62 M1372.02 1543.54 L1372.02 1530.49 Q1368.33 1530.9 1366.39 1532.59 Q1364.45 1534.27 1364.45 1537.07 Q1364.45 1539.81 1366.23 1541.34 Q1368.04 1542.87 1372.02 1543.54 M1375.2 1549.65 L1375.2 1563.43 Q1379.25 1562.89 1381.28 1561.14 Q1383.35 1559.39 1383.35 1556.52 Q1383.35 1553.72 1381.38 1552.07 Q1379.44 1550.41 1375.2 1549.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1414.48 1518.52 L1414.48 1576.64 L1400.98 1576.64 L1400.98 1572.09 L1408.59 1572.09 L1408.59 1523.07 L1400.98 1523.07 L1400.98 1518.52 L1414.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1110.47 2352.76,1110.47 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,797.755 2352.76,797.755 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,485.042 2352.76,485.042 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,172.329 2352.76,172.329 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 205.121,47.2441 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 224.019,1423.18 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1110.47 224.019,1110.47 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,797.755 224.019,797.755 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,485.042 224.019,485.042 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,172.329 224.019,172.329 "/>
-<path clip-path="url(#clip950)" d="M157.177 1408.98 Q153.566 1408.98 151.737 1412.54 Q149.931 1416.08 149.931 1423.21 Q149.931 1430.32 151.737 1433.89 Q153.566 1437.43 157.177 1437.43 Q160.811 1437.43 162.616 1433.89 Q164.445 1430.32 164.445 1423.21 Q164.445 1416.08 162.616 1412.54 Q160.811 1408.98 157.177 1408.98 M157.177 1405.27 Q162.987 1405.27 166.042 1409.88 Q169.121 1414.46 169.121 1423.21 Q169.121 1431.94 166.042 1436.55 Q162.987 1441.13 157.177 1441.13 Q151.366 1441.13 148.288 1436.55 Q145.232 1431.94 145.232 1423.21 Q145.232 1414.46 148.288 1409.88 Q151.366 1405.27 157.177 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M117.825 1123.81 L125.464 1123.81 L125.464 1097.45 L117.154 1099.11 L117.154 1094.85 L125.418 1093.19 L130.093 1093.19 L130.093 1123.81 L137.732 1123.81 L137.732 1127.75 L117.825 1127.75 L117.825 1123.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M157.177 1096.27 Q153.566 1096.27 151.737 1099.83 Q149.931 1103.37 149.931 1110.5 Q149.931 1117.61 151.737 1121.17 Q153.566 1124.71 157.177 1124.71 Q160.811 1124.71 162.616 1121.17 Q164.445 1117.61 164.445 1110.5 Q164.445 1103.37 162.616 1099.83 Q160.811 1096.27 157.177 1096.27 M157.177 1092.56 Q162.987 1092.56 166.042 1097.17 Q169.121 1101.75 169.121 1110.5 Q169.121 1119.23 166.042 1123.84 Q162.987 1128.42 157.177 1128.42 Q151.366 1128.42 148.288 1123.84 Q145.232 1119.23 145.232 1110.5 Q145.232 1101.75 148.288 1097.17 Q151.366 1092.56 157.177 1092.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M121.043 811.099 L137.362 811.099 L137.362 815.035 L115.418 815.035 L115.418 811.099 Q118.08 808.345 122.663 803.715 Q127.269 799.062 128.45 797.72 Q130.695 795.197 131.575 793.461 Q132.478 791.701 132.478 790.011 Q132.478 787.257 130.533 785.521 Q128.612 783.785 125.51 783.785 Q123.311 783.785 120.857 784.549 Q118.427 785.312 115.649 786.863 L115.649 782.141 Q118.473 781.007 120.927 780.428 Q123.38 779.85 125.418 779.85 Q130.788 779.85 133.982 782.535 Q137.177 785.22 137.177 789.711 Q137.177 791.84 136.367 793.761 Q135.579 795.66 133.473 798.252 Q132.894 798.923 129.792 802.141 Q126.691 805.335 121.043 811.099 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M157.177 783.553 Q153.566 783.553 151.737 787.118 Q149.931 790.66 149.931 797.789 Q149.931 804.896 151.737 808.46 Q153.566 812.002 157.177 812.002 Q160.811 812.002 162.616 808.46 Q164.445 804.896 164.445 797.789 Q164.445 790.66 162.616 787.118 Q160.811 783.553 157.177 783.553 M157.177 779.85 Q162.987 779.85 166.042 784.456 Q169.121 789.039 169.121 797.789 Q169.121 806.516 166.042 811.122 Q162.987 815.706 157.177 815.706 Q151.366 815.706 148.288 811.122 Q145.232 806.516 145.232 797.789 Q145.232 789.039 148.288 784.456 Q151.366 779.85 157.177 779.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M131.181 483.688 Q134.538 484.405 136.413 486.674 Q138.311 488.942 138.311 492.276 Q138.311 497.391 134.792 500.192 Q131.274 502.993 124.793 502.993 Q122.617 502.993 120.302 502.553 Q118.01 502.137 115.556 501.28 L115.556 496.766 Q117.501 497.901 119.816 498.479 Q122.13 499.058 124.654 499.058 Q129.052 499.058 131.343 497.322 Q133.658 495.586 133.658 492.276 Q133.658 489.22 131.505 487.507 Q129.376 485.771 125.556 485.771 L121.529 485.771 L121.529 481.928 L125.742 481.928 Q129.191 481.928 131.019 480.563 Q132.848 479.174 132.848 476.581 Q132.848 473.919 130.95 472.507 Q129.075 471.072 125.556 471.072 Q123.635 471.072 121.436 471.489 Q119.237 471.905 116.598 472.785 L116.598 468.618 Q119.26 467.878 121.575 467.507 Q123.913 467.137 125.973 467.137 Q131.297 467.137 134.399 469.567 Q137.501 471.975 137.501 476.095 Q137.501 478.965 135.857 480.956 Q134.214 482.924 131.181 483.688 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M157.177 470.841 Q153.566 470.841 151.737 474.405 Q149.931 477.947 149.931 485.077 Q149.931 492.183 151.737 495.748 Q153.566 499.289 157.177 499.289 Q160.811 499.289 162.616 495.748 Q164.445 492.183 164.445 485.077 Q164.445 477.947 162.616 474.405 Q160.811 470.841 157.177 470.841 M157.177 467.137 Q162.987 467.137 166.042 471.743 Q169.121 476.327 169.121 485.077 Q169.121 493.803 166.042 498.41 Q162.987 502.993 157.177 502.993 Q151.366 502.993 148.288 498.41 Q145.232 493.803 145.232 485.077 Q145.232 476.327 148.288 471.743 Q151.366 467.137 157.177 467.137 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M129.862 159.123 L118.056 177.572 L129.862 177.572 L129.862 159.123 M128.635 155.049 L134.515 155.049 L134.515 177.572 L139.445 177.572 L139.445 181.461 L134.515 181.461 L134.515 189.609 L129.862 189.609 L129.862 181.461 L114.26 181.461 L114.26 176.947 L128.635 155.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M157.177 158.128 Q153.566 158.128 151.737 161.693 Q149.931 165.234 149.931 172.364 Q149.931 179.47 151.737 183.035 Q153.566 186.577 157.177 186.577 Q160.811 186.577 162.616 183.035 Q164.445 179.47 164.445 172.364 Q164.445 165.234 162.616 161.693 Q160.811 158.128 157.177 158.128 M157.177 154.424 Q162.987 154.424 166.042 159.031 Q169.121 163.614 169.121 172.364 Q169.121 181.091 166.042 185.697 Q162.987 190.28 157.177 190.28 Q151.366 190.28 148.288 185.697 Q145.232 181.091 145.232 172.364 Q145.232 163.614 148.288 159.031 Q151.366 154.424 157.177 154.424 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M16.4842 1073.5 L16.4842 1064.84 L56.238 1043.77 L16.4842 1043.77 L16.4842 1037.54 L64.0042 1037.54 L64.0042 1046.19 L24.2503 1067.26 L64.0042 1067.26 L64.0042 1073.5 L16.4842 1073.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M49.9359 1025.6 L28.3562 1025.6 L28.3562 1019.74 L49.7131 1019.74 Q54.7739 1019.74 57.3202 1017.77 Q59.8346 1015.8 59.8346 1011.85 Q59.8346 1007.11 56.8109 1004.37 Q53.7872 1001.6 48.5673 1001.6 L28.3562 1001.6 L28.3562 995.744 L64.0042 995.744 L64.0042 1001.6 L58.5296 1001.6 Q61.7762 1003.73 63.3676 1006.57 Q64.9272 1009.37 64.9272 1013.09 Q64.9272 1019.23 61.1078 1022.42 Q57.2883 1025.6 49.9359 1025.6 M27.4968 1010.86 L27.4968 1010.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M35.1993 955.927 Q31.2526 953.731 29.3747 950.675 Q27.4968 947.62 27.4968 943.482 Q27.4968 937.912 31.4117 934.888 Q35.2948 931.865 42.4881 931.865 L64.0042 931.865 L64.0042 937.753 L42.679 937.753 Q37.5546 937.753 35.072 939.567 Q32.5894 941.381 32.5894 945.105 Q32.5894 949.657 35.6131 952.298 Q38.6368 954.94 43.8567 954.94 L64.0042 954.94 L64.0042 960.829 L42.679 960.829 Q37.5228 960.829 35.072 962.643 Q32.5894 964.457 32.5894 968.245 Q32.5894 972.732 35.6449 975.374 Q38.6686 978.016 43.8567 978.016 L64.0042 978.016 L64.0042 983.904 L28.3562 983.904 L28.3562 978.016 L33.8944 978.016 Q30.616 976.011 29.0564 973.21 Q27.4968 970.409 27.4968 966.558 Q27.4968 962.675 29.4702 959.969 Q31.4436 957.232 35.1993 955.927 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M46.212 894.593 Q39.7508 894.593 36.0905 897.267 Q32.3984 899.909 32.3984 904.556 Q32.3984 909.203 36.0905 911.876 Q39.7508 914.518 46.212 914.518 Q52.6732 914.518 56.3653 911.876 Q60.0256 909.203 60.0256 904.556 Q60.0256 899.909 56.3653 897.267 Q52.6732 894.593 46.212 894.593 M33.7671 914.518 Q30.5842 912.672 29.0564 909.871 Q27.4968 907.038 27.4968 903.123 Q27.4968 896.63 32.6531 892.588 Q37.8093 888.514 46.212 888.514 Q54.6147 888.514 59.771 892.588 Q64.9272 896.63 64.9272 903.123 Q64.9272 907.038 63.3994 909.871 Q61.8398 912.672 58.657 914.518 L64.0042 914.518 L64.0042 920.406 L14.479 920.406 L14.479 914.518 L33.7671 914.518 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M44.7161 848.315 L47.5806 848.315 L47.5806 875.242 Q53.6281 874.86 56.8109 871.613 Q59.9619 868.335 59.9619 862.51 Q59.9619 859.136 59.1344 855.985 Q58.3069 852.802 56.6518 849.683 L62.1899 849.683 Q63.5267 852.834 64.227 856.144 Q64.9272 859.455 64.9272 862.86 Q64.9272 871.39 59.9619 876.387 Q54.9967 881.353 46.5303 881.353 Q37.7774 881.353 32.6531 876.642 Q27.4968 871.9 27.4968 863.879 Q27.4968 856.686 32.1438 852.516 Q36.7589 848.315 44.7161 848.315 M42.9973 854.171 Q38.1912 854.235 35.3266 856.877 Q32.4621 859.486 32.4621 863.815 Q32.4621 868.717 35.2312 871.677 Q38.0002 874.605 43.0292 875.051 L42.9973 854.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M33.8307 818.046 Q33.2578 819.032 33.0032 820.21 Q32.7167 821.356 32.7167 822.756 Q32.7167 827.722 35.9632 830.395 Q39.1779 833.037 45.2253 833.037 L64.0042 833.037 L64.0042 838.925 L28.3562 838.925 L28.3562 833.037 L33.8944 833.037 Q30.6479 831.191 29.0883 828.231 Q27.4968 825.271 27.4968 821.038 Q27.4968 820.433 27.5923 819.701 Q27.656 818.969 27.8151 818.078 L33.8307 818.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M32.4621 777.369 Q32.4621 782.079 36.1542 784.817 Q39.8145 787.554 46.212 787.554 Q52.6095 787.554 56.3017 784.849 Q59.9619 782.111 59.9619 777.369 Q59.9619 772.69 56.2698 769.953 Q52.5777 767.216 46.212 767.216 Q39.8781 767.216 36.186 769.953 Q32.4621 772.69 32.4621 777.369 M27.4968 777.369 Q27.4968 769.73 32.4621 765.37 Q37.4273 761.009 46.212 761.009 Q54.9649 761.009 59.9619 765.37 Q64.9272 769.73 64.9272 777.369 Q64.9272 785.04 59.9619 789.4 Q54.9649 793.729 46.212 793.729 Q37.4273 793.729 32.4621 789.4 Q27.4968 785.04 27.4968 777.369 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M14.479 733.255 L19.3487 733.255 L19.3487 738.856 Q19.3487 742.007 20.6219 743.249 Q21.895 744.458 25.2052 744.458 L28.3562 744.458 L28.3562 734.814 L32.9077 734.814 L32.9077 744.458 L64.0042 744.458 L64.0042 750.346 L32.9077 750.346 L32.9077 755.948 L28.3562 755.948 L28.3562 750.346 L25.8736 750.346 Q19.9216 750.346 17.2162 747.577 Q14.479 744.808 14.479 738.793 L14.479 733.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M32.4621 693.819 Q32.4621 698.53 36.1542 701.267 Q39.8145 704.004 46.212 704.004 Q52.6095 704.004 56.3017 701.299 Q59.9619 698.561 59.9619 693.819 Q59.9619 689.14 56.2698 686.403 Q52.5777 683.666 46.212 683.666 Q39.8781 683.666 36.186 686.403 Q32.4621 689.14 32.4621 693.819 M27.4968 693.819 Q27.4968 686.18 32.4621 681.82 Q37.4273 677.459 46.212 677.459 Q54.9649 677.459 59.9619 681.82 Q64.9272 686.18 64.9272 693.819 Q64.9272 701.49 59.9619 705.85 Q54.9649 710.179 46.212 710.179 Q37.4273 710.179 32.4621 705.85 Q27.4968 701.49 27.4968 693.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M49.9359 668.356 L28.3562 668.356 L28.3562 662.5 L49.7131 662.5 Q54.7739 662.5 57.3202 660.526 Q59.8346 658.553 59.8346 654.606 Q59.8346 649.864 56.8109 647.127 Q53.7872 644.357 48.5673 644.357 L28.3562 644.357 L28.3562 638.501 L64.0042 638.501 L64.0042 644.357 L58.5296 644.357 Q61.7762 646.49 63.3676 649.323 Q64.9272 652.124 64.9272 655.848 Q64.9272 661.99 61.1078 665.173 Q57.2883 668.356 49.9359 668.356 M27.4968 653.62 L27.4968 653.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M18.2347 620.645 L28.3562 620.645 L28.3562 608.582 L32.9077 608.582 L32.9077 620.645 L52.2594 620.645 Q56.6199 620.645 57.8613 619.468 Q59.1026 618.258 59.1026 614.598 L59.1026 608.582 L64.0042 608.582 L64.0042 614.598 Q64.0042 621.377 61.4897 623.955 Q58.9434 626.533 52.2594 626.533 L32.9077 626.533 L32.9077 630.83 L28.3562 630.83 L28.3562 626.533 L18.2347 626.533 L18.2347 620.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M29.7248 575.226 L35.1993 575.226 Q33.8307 577.708 33.1623 580.223 Q32.4621 582.706 32.4621 585.252 Q32.4621 590.949 36.0905 594.1 Q39.6872 597.251 46.212 597.251 Q52.7369 597.251 56.3653 594.1 Q59.9619 590.949 59.9619 585.252 Q59.9619 582.706 59.2935 580.223 Q58.5933 577.708 57.2247 575.226 L62.6355 575.226 Q63.7814 577.677 64.3543 580.318 Q64.9272 582.928 64.9272 585.888 Q64.9272 593.941 59.8664 598.683 Q54.8057 603.426 46.212 603.426 Q37.491 603.426 32.4939 598.652 Q27.4968 593.846 27.4968 585.506 Q27.4968 582.801 28.0697 580.223 Q28.6108 577.645 29.7248 575.226 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M32.4621 551.227 Q32.4621 555.938 36.1542 558.675 Q39.8145 561.412 46.212 561.412 Q52.6095 561.412 56.3017 558.707 Q59.9619 555.97 59.9619 551.227 Q59.9619 546.548 56.2698 543.811 Q52.5777 541.074 46.212 541.074 Q39.8781 541.074 36.186 543.811 Q32.4621 546.548 32.4621 551.227 M27.4968 551.227 Q27.4968 543.588 32.4621 539.228 Q37.4273 534.867 46.212 534.867 Q54.9649 534.867 59.9619 539.228 Q64.9272 543.588 64.9272 551.227 Q64.9272 558.898 59.9619 563.258 Q54.9649 567.587 46.212 567.587 Q37.4273 567.587 32.4621 563.258 Q27.4968 558.898 27.4968 551.227 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M35.1993 497.405 Q31.2526 495.209 29.3747 492.153 Q27.4968 489.098 27.4968 484.96 Q27.4968 479.39 31.4117 476.366 Q35.2948 473.343 42.4881 473.343 L64.0042 473.343 L64.0042 479.231 L42.679 479.231 Q37.5546 479.231 35.072 481.045 Q32.5894 482.859 32.5894 486.583 Q32.5894 491.135 35.6131 493.777 Q38.6368 496.418 43.8567 496.418 L64.0042 496.418 L64.0042 502.307 L42.679 502.307 Q37.5228 502.307 35.072 504.121 Q32.5894 505.935 32.5894 509.723 Q32.5894 514.211 35.6449 516.852 Q38.6686 519.494 43.8567 519.494 L64.0042 519.494 L64.0042 525.382 L28.3562 525.382 L28.3562 519.494 L33.8944 519.494 Q30.616 517.489 29.0564 514.688 Q27.4968 511.887 27.4968 508.036 Q27.4968 504.153 29.4702 501.447 Q31.4436 498.71 35.1993 497.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M44.7161 431.17 L47.5806 431.17 L47.5806 458.097 Q53.6281 457.715 56.8109 454.468 Q59.9619 451.19 59.9619 445.365 Q59.9619 441.992 59.1344 438.841 Q58.3069 435.658 56.6518 432.539 L62.1899 432.539 Q63.5267 435.69 64.227 439 Q64.9272 442.31 64.9272 445.716 Q64.9272 454.246 59.9619 459.243 Q54.9967 464.208 46.5303 464.208 Q37.7774 464.208 32.6531 459.497 Q27.4968 454.755 27.4968 446.734 Q27.4968 439.541 32.1438 435.371 Q36.7589 431.17 44.7161 431.17 M42.9973 437.026 Q38.1912 437.09 35.3266 439.732 Q32.4621 442.342 32.4621 446.67 Q32.4621 451.572 35.2312 454.532 Q38.0002 457.46 43.0292 457.906 L42.9973 437.026 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M29.4065 398.832 L34.9447 398.832 Q33.6716 401.315 33.035 403.988 Q32.3984 406.662 32.3984 409.527 Q32.3984 413.887 33.7352 416.083 Q35.072 418.248 37.7456 418.248 Q39.7826 418.248 40.9603 416.688 Q42.1061 415.128 43.1565 410.418 L43.6021 408.413 Q44.9389 402.174 47.3897 399.564 Q49.8086 396.922 54.1691 396.922 Q59.1344 396.922 62.0308 400.869 Q64.9272 404.784 64.9272 411.659 Q64.9272 414.524 64.3543 417.643 Q63.8132 420.73 62.6992 424.168 L56.6518 424.168 Q58.3387 420.921 59.198 417.77 Q60.0256 414.619 60.0256 411.532 Q60.0256 407.394 58.6251 405.166 Q57.1929 402.938 54.6147 402.938 Q52.2276 402.938 50.9545 404.561 Q49.6813 406.153 48.5037 411.595 L48.0262 413.632 Q46.8804 419.075 44.5251 421.494 Q42.138 423.913 38.0002 423.913 Q32.9713 423.913 30.2341 420.348 Q27.4968 416.783 27.4968 410.227 Q27.4968 406.98 27.9743 404.116 Q28.4517 401.251 29.4065 398.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip952)" d="M323.245 1298.09 L323.245 1423.18 L535.621 1423.18 L535.621 1298.09 L323.245 1298.09 L323.245 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1298.09 323.245,1423.18 535.621,1423.18 535.621,1298.09 323.245,1298.09 "/>
-<path clip-path="url(#clip952)" d="M535.621 1266.82 L535.621 1423.18 L747.997 1423.18 L747.997 1266.82 L535.621 1266.82 L535.621 1266.82  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="535.621,1266.82 535.621,1423.18 747.997,1423.18 747.997,1266.82 535.621,1266.82 "/>
-<path clip-path="url(#clip952)" d="M747.997 1298.09 L747.997 1423.18 L960.374 1423.18 L960.374 1298.09 L747.997 1298.09 L747.997 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.997,1298.09 747.997,1423.18 960.374,1423.18 960.374,1298.09 747.997,1298.09 "/>
-<path clip-path="url(#clip952)" d="M960.374 1173.01 L960.374 1423.18 L1172.75 1423.18 L1172.75 1173.01 L960.374 1173.01 L960.374 1173.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="960.374,1173.01 960.374,1423.18 1172.75,1423.18 1172.75,1173.01 960.374,1173.01 "/>
-<path clip-path="url(#clip952)" d="M1172.75 1235.55 L1172.75 1423.18 L1385.13 1423.18 L1385.13 1235.55 L1172.75 1235.55 L1172.75 1235.55  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1172.75,1235.55 1172.75,1423.18 1385.13,1423.18 1385.13,1235.55 1172.75,1235.55 "/>
-<path clip-path="url(#clip952)" d="M1385.13 1141.74 L1385.13 1423.18 L1597.5 1423.18 L1597.5 1141.74 L1385.13 1141.74 L1385.13 1141.74  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1385.13,1141.74 1385.13,1423.18 1597.5,1423.18 1597.5,1141.74 1385.13,1141.74 "/>
-<path clip-path="url(#clip952)" d="M1597.5 1047.92 L1597.5 1423.18 L1809.88 1423.18 L1809.88 1047.92 L1597.5 1047.92 L1597.5 1047.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1597.5,1047.92 1597.5,1423.18 1809.88,1423.18 1809.88,1047.92 1597.5,1047.92 "/>
-<path clip-path="url(#clip952)" d="M1809.88 1173.01 L1809.88 1423.18 L2022.26 1423.18 L2022.26 1173.01 L1809.88 1173.01 L1809.88 1173.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1809.88,1173.01 1809.88,1423.18 2022.26,1423.18 2022.26,1173.01 1809.88,1173.01 "/>
-<path clip-path="url(#clip952)" d="M2022.26 47.2441 L2022.26 1423.18 L2234.63 1423.18 L2234.63 47.2441 L2022.26 47.2441 L2022.26 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2022.26,47.2441 2022.26,1423.18 2234.63,1423.18 2234.63,47.2441 2022.26,47.2441 "/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="429.433" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="641.809" cy="1266.82" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="854.186" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1066.56" cy="1173.01" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1278.94" cy="1235.55" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1491.31" cy="1141.74" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1703.69" cy="1047.92" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1916.07" cy="1173.01" r="2"/>
-<circle clip-path="url(#clip952)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2128.44" cy="47.2441" r="2"/>
-<polyline clip-path="url(#clip952)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1638.4,2799.12 1638.4,-1328.69 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1241.72,2799.12 1241.72,-1328.69 "/>
-<polyline clip-path="url(#clip952)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="355.696,2799.12 355.696,-1328.69 "/>
-<path clip-path="url(#clip950)" d="M276.709 300.469 L660.893 300.469 L660.893 93.1086 L276.709 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="276.709,300.469 660.893,300.469 660.893,93.1086 276.709,93.1086 276.709,300.469 "/>
-<polyline clip-path="url(#clip950)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,144.949 443.747,144.949 "/>
-<path clip-path="url(#clip950)" d="M475.225 140.377 L480.897 156.395 L488.998 136.303 L493.512 136.303 L483.026 162.229 L483.026 172.09 L478.767 172.09 L478.767 162.229 L471.962 142.877 Q470.943 139.96 468.744 139.96 L467.61 139.96 L467.61 136.303 L469.23 136.303 Q473.79 136.303 475.225 140.377 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M515.017 140.701 L544.693 140.701 L544.693 144.59 L515.017 144.59 L515.017 140.701 M515.017 150.145 L544.693 150.145 L544.693 154.081 L515.017 154.081 L515.017 150.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M570.665 158.293 L578.304 158.293 L578.304 131.928 L569.993 133.595 L569.993 129.335 L578.257 127.669 L582.933 127.669 L582.933 158.293 L590.572 158.293 L590.572 162.229 L570.665 162.229 L570.665 158.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M600.016 156.349 L604.901 156.349 L604.901 162.229 L600.016 162.229 L600.016 156.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M625.086 130.747 Q621.475 130.747 619.646 134.312 Q617.84 137.854 617.84 144.983 Q617.84 152.09 619.646 155.655 Q621.475 159.196 625.086 159.196 Q628.72 159.196 630.525 155.655 Q632.354 152.09 632.354 144.983 Q632.354 137.854 630.525 134.312 Q628.72 130.747 625.086 130.747 M625.086 127.044 Q630.896 127.044 633.951 131.65 Q637.03 136.233 637.03 144.983 Q637.03 153.71 633.951 158.317 Q630.896 162.9 625.086 162.9 Q619.276 162.9 616.197 158.317 Q613.141 153.71 613.141 144.983 Q613.141 136.233 616.197 131.65 Q619.276 127.044 625.086 127.044 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip950)" style="stroke:#3da44d; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,196.789 443.747,196.789 "/>
-<path clip-path="url(#clip950)" d="M475.225 192.217 L480.897 208.235 L488.998 188.143 L493.512 188.143 L483.026 214.069 L483.026 223.93 L478.767 223.93 L478.767 214.069 L471.962 194.717 Q470.943 191.8 468.744 191.8 L467.61 191.8 L467.61 188.143 L469.23 188.143 Q473.79 188.143 475.225 192.217 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M515.017 192.541 L544.693 192.541 L544.693 196.43 L515.017 196.43 L515.017 192.541 M515.017 201.985 L544.693 201.985 L544.693 205.921 L515.017 205.921 L515.017 201.985 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M579.854 182.587 Q576.243 182.587 574.415 186.152 Q572.609 189.694 572.609 196.823 Q572.609 203.93 574.415 207.495 Q576.243 211.036 579.854 211.036 Q583.489 211.036 585.294 207.495 Q587.123 203.93 587.123 196.823 Q587.123 189.694 585.294 186.152 Q583.489 182.587 579.854 182.587 M579.854 178.884 Q585.665 178.884 588.72 183.49 Q591.799 188.073 591.799 196.823 Q591.799 205.55 588.72 210.157 Q585.665 214.74 579.854 214.74 Q574.044 214.74 570.966 210.157 Q567.91 205.55 567.91 196.823 Q567.91 188.073 570.966 183.49 Q574.044 178.884 579.854 178.884 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M600.016 208.189 L604.901 208.189 L604.901 214.069 L600.016 214.069 L600.016 208.189 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M615.132 179.509 L633.488 179.509 L633.488 183.444 L619.414 183.444 L619.414 191.916 Q620.433 191.569 621.451 191.407 Q622.47 191.222 623.488 191.222 Q629.275 191.222 632.655 194.393 Q636.035 197.564 636.035 202.981 Q636.035 208.559 632.563 211.661 Q629.09 214.74 622.771 214.74 Q620.595 214.74 618.326 214.37 Q616.081 213.999 613.674 213.258 L613.674 208.559 Q615.757 209.694 617.979 210.249 Q620.201 210.805 622.678 210.805 Q626.683 210.805 629.021 208.698 Q631.359 206.592 631.359 202.981 Q631.359 199.37 629.021 197.263 Q626.683 195.157 622.678 195.157 Q620.803 195.157 618.928 195.573 Q617.076 195.99 615.132 196.87 L615.132 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip950)" style="stroke:#c271d2; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,248.629 443.747,248.629 "/>
-<path clip-path="url(#clip950)" d="M475.225 244.057 L480.897 260.075 L488.998 239.983 L493.512 239.983 L483.026 265.909 L483.026 275.77 L478.767 275.77 L478.767 265.909 L471.962 246.557 Q470.943 243.64 468.744 243.64 L467.61 243.64 L467.61 239.983 L469.23 239.983 Q473.79 239.983 475.225 244.057 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M515.017 244.381 L544.693 244.381 L544.693 248.27 L515.017 248.27 L515.017 244.381 M515.017 253.825 L544.693 253.825 L544.693 257.761 L515.017 257.761 L515.017 253.825 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M579.854 234.427 Q576.243 234.427 574.415 237.992 Q572.609 241.534 572.609 248.663 Q572.609 255.77 574.415 259.335 Q576.243 262.876 579.854 262.876 Q583.489 262.876 585.294 259.335 Q587.123 255.77 587.123 248.663 Q587.123 241.534 585.294 237.992 Q583.489 234.427 579.854 234.427 M579.854 230.724 Q585.665 230.724 588.72 235.33 Q591.799 239.913 591.799 248.663 Q591.799 257.39 588.72 261.997 Q585.665 266.58 579.854 266.58 Q574.044 266.58 570.966 261.997 Q567.91 257.39 567.91 248.663 Q567.91 239.913 570.966 235.33 Q574.044 230.724 579.854 230.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M600.016 260.029 L604.901 260.029 L604.901 265.909 L600.016 265.909 L600.016 260.029 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M625.086 234.427 Q621.475 234.427 619.646 237.992 Q617.84 241.534 617.84 248.663 Q617.84 255.77 619.646 259.335 Q621.475 262.876 625.086 262.876 Q628.72 262.876 630.525 259.335 Q632.354 255.77 632.354 248.663 Q632.354 241.534 630.525 237.992 Q628.72 234.427 625.086 234.427 M625.086 230.724 Q630.896 230.724 633.951 235.33 Q637.03 239.913 637.03 248.663 Q637.03 257.39 633.951 261.997 Q630.896 266.58 625.086 266.58 Q619.276 266.58 616.197 261.997 Q613.141 257.39 613.141 248.663 Q613.141 239.913 616.197 235.33 Q619.276 230.724 625.086 230.724 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/7f7c697a.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/7f7c697a.svg
deleted file mode 100644
index 30bb55e8ad6..00000000000
--- a/previews/PR3562/tutorials/applications/two_stage_stochastic/7f7c697a.svg
+++ /dev/null
@@ -1,73 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
-<defs>
-  <clipPath id="clip890">
-    <rect x="0" y="0" width="2400" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip890)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip891">
-    <rect x="480" y="0" width="1681" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip890)" d="M205.121 1423.18 L2352.76 1423.18 L2352.76 47.2441 L205.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip892">
-    <rect x="205" y="47" width="2149" height="1377"/>
-  </clipPath>
-</defs>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="323.245,1423.18 323.245,47.2441 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="747.997,1423.18 747.997,47.2441 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1172.75,1423.18 1172.75,47.2441 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1597.5,1423.18 1597.5,47.2441 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2022.26,1423.18 2022.26,47.2441 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1423.18 323.245,1404.28 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.997,1423.18 747.997,1404.28 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1172.75,1423.18 1172.75,1404.28 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1597.5,1423.18 1597.5,1404.28 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2022.26,1423.18 2022.26,1404.28 "/>
-<path clip-path="url(#clip890)" d="M296.335 1455.09 L284.53 1473.54 L296.335 1473.54 L296.335 1455.09 M295.108 1451.02 L300.988 1451.02 L300.988 1473.54 L305.918 1473.54 L305.918 1477.43 L300.988 1477.43 L300.988 1485.58 L296.335 1485.58 L296.335 1477.43 L280.733 1477.43 L280.733 1472.92 L295.108 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M323.65 1454.1 Q320.039 1454.1 318.21 1457.66 Q316.404 1461.2 316.404 1468.33 Q316.404 1475.44 318.21 1479.01 Q320.039 1482.55 323.65 1482.55 Q327.284 1482.55 329.09 1479.01 Q330.918 1475.44 330.918 1468.33 Q330.918 1461.2 329.09 1457.66 Q327.284 1454.1 323.65 1454.1 M323.65 1450.39 Q329.46 1450.39 332.516 1455 Q335.594 1459.58 335.594 1468.33 Q335.594 1477.06 332.516 1481.67 Q329.46 1486.25 323.65 1486.25 Q317.84 1486.25 314.761 1481.67 Q311.705 1477.06 311.705 1468.33 Q311.705 1459.58 314.761 1455 Q317.84 1450.39 323.65 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M353.812 1454.1 Q350.201 1454.1 348.372 1457.66 Q346.566 1461.2 346.566 1468.33 Q346.566 1475.44 348.372 1479.01 Q350.201 1482.55 353.812 1482.55 Q357.446 1482.55 359.251 1479.01 Q361.08 1475.44 361.08 1468.33 Q361.08 1461.2 359.251 1457.66 Q357.446 1454.1 353.812 1454.1 M353.812 1450.39 Q359.622 1450.39 362.677 1455 Q365.756 1459.58 365.756 1468.33 Q365.756 1477.06 362.677 1481.67 Q359.622 1486.25 353.812 1486.25 Q348.002 1486.25 344.923 1481.67 Q341.867 1477.06 341.867 1468.33 Q341.867 1459.58 344.923 1455 Q348.002 1450.39 353.812 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M721.088 1455.09 L709.282 1473.54 L721.088 1473.54 L721.088 1455.09 M719.861 1451.02 L725.741 1451.02 L725.741 1473.54 L730.671 1473.54 L730.671 1477.43 L725.741 1477.43 L725.741 1485.58 L721.088 1485.58 L721.088 1477.43 L705.486 1477.43 L705.486 1472.92 L719.861 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M738.449 1451.02 L756.805 1451.02 L756.805 1454.96 L742.731 1454.96 L742.731 1463.43 Q743.75 1463.08 744.768 1462.92 Q745.787 1462.73 746.805 1462.73 Q752.592 1462.73 755.972 1465.9 Q759.352 1469.08 759.352 1474.49 Q759.352 1480.07 755.879 1483.17 Q752.407 1486.25 746.088 1486.25 Q743.912 1486.25 741.643 1485.88 Q739.398 1485.51 736.991 1484.77 L736.991 1480.07 Q739.074 1481.2 741.296 1481.76 Q743.518 1482.32 745.995 1482.32 Q750 1482.32 752.338 1480.21 Q754.676 1478.1 754.676 1474.49 Q754.676 1470.88 752.338 1468.77 Q750 1466.67 745.995 1466.67 Q744.12 1466.67 742.245 1467.08 Q740.393 1467.5 738.449 1468.38 L738.449 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M778.564 1454.1 Q774.953 1454.1 773.125 1457.66 Q771.319 1461.2 771.319 1468.33 Q771.319 1475.44 773.125 1479.01 Q774.953 1482.55 778.564 1482.55 Q782.199 1482.55 784.004 1479.01 Q785.833 1475.44 785.833 1468.33 Q785.833 1461.2 784.004 1457.66 Q782.199 1454.1 778.564 1454.1 M778.564 1450.39 Q784.375 1450.39 787.43 1455 Q790.509 1459.58 790.509 1468.33 Q790.509 1477.06 787.43 1481.67 Q784.375 1486.25 778.564 1486.25 Q772.754 1486.25 769.676 1481.67 Q766.62 1477.06 766.62 1468.33 Q766.62 1459.58 769.676 1455 Q772.754 1450.39 778.564 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1132.37 1451.02 L1150.72 1451.02 L1150.72 1454.96 L1136.65 1454.96 L1136.65 1463.43 Q1137.67 1463.08 1138.69 1462.92 Q1139.71 1462.73 1140.72 1462.73 Q1146.51 1462.73 1149.89 1465.9 Q1153.27 1469.08 1153.27 1474.49 Q1153.27 1480.07 1149.8 1483.17 Q1146.33 1486.25 1140.01 1486.25 Q1137.83 1486.25 1135.56 1485.88 Q1133.32 1485.51 1130.91 1484.77 L1130.91 1480.07 Q1132.99 1481.2 1135.22 1481.76 Q1137.44 1482.32 1139.91 1482.32 Q1143.92 1482.32 1146.26 1480.21 Q1148.6 1478.1 1148.6 1474.49 Q1148.6 1470.88 1146.26 1468.77 Q1143.92 1466.67 1139.91 1466.67 Q1138.04 1466.67 1136.16 1467.08 Q1134.31 1467.5 1132.37 1468.38 L1132.37 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1172.48 1454.1 Q1168.87 1454.1 1167.04 1457.66 Q1165.24 1461.2 1165.24 1468.33 Q1165.24 1475.44 1167.04 1479.01 Q1168.87 1482.55 1172.48 1482.55 Q1176.12 1482.55 1177.92 1479.01 Q1179.75 1475.44 1179.75 1468.33 Q1179.75 1461.2 1177.92 1457.66 Q1176.12 1454.1 1172.48 1454.1 M1172.48 1450.39 Q1178.29 1450.39 1181.35 1455 Q1184.43 1459.58 1184.43 1468.33 Q1184.43 1477.06 1181.35 1481.67 Q1178.29 1486.25 1172.48 1486.25 Q1166.67 1486.25 1163.6 1481.67 Q1160.54 1477.06 1160.54 1468.33 Q1160.54 1459.58 1163.6 1455 Q1166.67 1450.39 1172.48 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1202.65 1454.1 Q1199.03 1454.1 1197.21 1457.66 Q1195.4 1461.2 1195.4 1468.33 Q1195.4 1475.44 1197.21 1479.01 Q1199.03 1482.55 1202.65 1482.55 Q1206.28 1482.55 1208.09 1479.01 Q1209.91 1475.44 1209.91 1468.33 Q1209.91 1461.2 1208.09 1457.66 Q1206.28 1454.1 1202.65 1454.1 M1202.65 1450.39 Q1208.46 1450.39 1211.51 1455 Q1214.59 1459.58 1214.59 1468.33 Q1214.59 1477.06 1211.51 1481.67 Q1208.46 1486.25 1202.65 1486.25 Q1196.84 1486.25 1193.76 1481.67 Q1190.7 1477.06 1190.7 1468.33 Q1190.7 1459.58 1193.76 1455 Q1196.84 1450.39 1202.65 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1557.12 1451.02 L1575.48 1451.02 L1575.48 1454.96 L1561.4 1454.96 L1561.4 1463.43 Q1562.42 1463.08 1563.44 1462.92 Q1564.46 1462.73 1565.48 1462.73 Q1571.26 1462.73 1574.64 1465.9 Q1578.02 1469.08 1578.02 1474.49 Q1578.02 1480.07 1574.55 1483.17 Q1571.08 1486.25 1564.76 1486.25 Q1562.58 1486.25 1560.32 1485.88 Q1558.07 1485.51 1555.66 1484.77 L1555.66 1480.07 Q1557.75 1481.2 1559.97 1481.76 Q1562.19 1482.32 1564.67 1482.32 Q1568.67 1482.32 1571.01 1480.21 Q1573.35 1478.1 1573.35 1474.49 Q1573.35 1470.88 1571.01 1468.77 Q1568.67 1466.67 1564.67 1466.67 Q1562.79 1466.67 1560.92 1467.08 Q1559.07 1467.5 1557.12 1468.38 L1557.12 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1587.28 1451.02 L1605.64 1451.02 L1605.64 1454.96 L1591.57 1454.96 L1591.57 1463.43 Q1592.58 1463.08 1593.6 1462.92 Q1594.62 1462.73 1595.64 1462.73 Q1601.43 1462.73 1604.81 1465.9 Q1608.19 1469.08 1608.19 1474.49 Q1608.19 1480.07 1604.71 1483.17 Q1601.24 1486.25 1594.92 1486.25 Q1592.75 1486.25 1590.48 1485.88 Q1588.23 1485.51 1585.82 1484.77 L1585.82 1480.07 Q1587.91 1481.2 1590.13 1481.76 Q1592.35 1482.32 1594.83 1482.32 Q1598.83 1482.32 1601.17 1480.21 Q1603.51 1478.1 1603.51 1474.49 Q1603.51 1470.88 1601.17 1468.77 Q1598.83 1466.67 1594.83 1466.67 Q1592.95 1466.67 1591.08 1467.08 Q1589.23 1467.5 1587.28 1468.38 L1587.28 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1627.4 1454.1 Q1623.79 1454.1 1621.96 1457.66 Q1620.15 1461.2 1620.15 1468.33 Q1620.15 1475.44 1621.96 1479.01 Q1623.79 1482.55 1627.4 1482.55 Q1631.03 1482.55 1632.84 1479.01 Q1634.67 1475.44 1634.67 1468.33 Q1634.67 1461.2 1632.84 1457.66 Q1631.03 1454.1 1627.4 1454.1 M1627.4 1450.39 Q1633.21 1450.39 1636.26 1455 Q1639.34 1459.58 1639.34 1468.33 Q1639.34 1477.06 1636.26 1481.67 Q1633.21 1486.25 1627.4 1486.25 Q1621.59 1486.25 1618.51 1481.67 Q1615.45 1477.06 1615.45 1468.33 Q1615.45 1459.58 1618.51 1455 Q1621.59 1450.39 1627.4 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1992.58 1466.44 Q1989.43 1466.44 1987.58 1468.59 Q1985.75 1470.74 1985.75 1474.49 Q1985.75 1478.22 1987.58 1480.39 Q1989.43 1482.55 1992.58 1482.55 Q1995.73 1482.55 1997.56 1480.39 Q1999.41 1478.22 1999.41 1474.49 Q1999.41 1470.74 1997.56 1468.59 Q1995.73 1466.44 1992.58 1466.44 M2001.86 1451.78 L2001.86 1456.04 Q2000.1 1455.21 1998.3 1454.77 Q1996.52 1454.33 1994.76 1454.33 Q1990.13 1454.33 1987.67 1457.45 Q1985.24 1460.58 1984.89 1466.9 Q1986.26 1464.89 1988.32 1463.82 Q1990.38 1462.73 1992.86 1462.73 Q1998.07 1462.73 2001.08 1465.9 Q2004.11 1469.05 2004.11 1474.49 Q2004.11 1479.82 2000.96 1483.03 Q1997.81 1486.25 1992.58 1486.25 Q1986.58 1486.25 1983.41 1481.67 Q1980.24 1477.06 1980.24 1468.33 Q1980.24 1460.14 1984.13 1455.28 Q1988.02 1450.39 1994.57 1450.39 Q1996.33 1450.39 1998.11 1450.74 Q1999.92 1451.09 2001.86 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M2022.16 1454.1 Q2018.55 1454.1 2016.72 1457.66 Q2014.92 1461.2 2014.92 1468.33 Q2014.92 1475.44 2016.72 1479.01 Q2018.55 1482.55 2022.16 1482.55 Q2025.8 1482.55 2027.6 1479.01 Q2029.43 1475.44 2029.43 1468.33 Q2029.43 1461.2 2027.6 1457.66 Q2025.8 1454.1 2022.16 1454.1 M2022.16 1450.39 Q2027.97 1450.39 2031.03 1455 Q2034.11 1459.58 2034.11 1468.33 Q2034.11 1477.06 2031.03 1481.67 Q2027.97 1486.25 2022.16 1486.25 Q2016.35 1486.25 2013.27 1481.67 Q2010.22 1477.06 2010.22 1468.33 Q2010.22 1459.58 2013.27 1455 Q2016.35 1450.39 2022.16 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M2052.33 1454.1 Q2048.71 1454.1 2046.89 1457.66 Q2045.08 1461.2 2045.08 1468.33 Q2045.08 1475.44 2046.89 1479.01 Q2048.71 1482.55 2052.33 1482.55 Q2055.96 1482.55 2057.76 1479.01 Q2059.59 1475.44 2059.59 1468.33 Q2059.59 1461.2 2057.76 1457.66 Q2055.96 1454.1 2052.33 1454.1 M2052.33 1450.39 Q2058.14 1450.39 2061.19 1455 Q2064.27 1459.58 2064.27 1468.33 Q2064.27 1477.06 2061.19 1481.67 Q2058.14 1486.25 2052.33 1486.25 Q2046.51 1486.25 2043.44 1481.67 Q2040.38 1477.06 2040.38 1468.33 Q2040.38 1459.58 2043.44 1455 Q2046.51 1450.39 2052.33 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1149.83 1525.81 L1149.83 1543.66 L1157.91 1543.66 Q1162.4 1543.66 1164.85 1541.34 Q1167.3 1539.02 1167.3 1534.72 Q1167.3 1530.45 1164.85 1528.13 Q1162.4 1525.81 1157.91 1525.81 L1149.83 1525.81 M1143.4 1520.52 L1157.91 1520.52 Q1165.9 1520.52 1169.97 1524.15 Q1174.08 1527.75 1174.08 1534.72 Q1174.08 1541.75 1169.97 1545.35 Q1165.9 1548.95 1157.91 1548.95 L1149.83 1548.95 L1149.83 1568.04 L1143.4 1568.04 L1143.4 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1201.96 1537.87 Q1200.97 1537.3 1199.8 1537.04 Q1198.65 1536.76 1197.25 1536.76 Q1192.29 1536.76 1189.61 1540 Q1186.97 1543.22 1186.97 1549.27 L1186.97 1568.04 L1181.08 1568.04 L1181.08 1532.4 L1186.97 1532.4 L1186.97 1537.93 Q1188.82 1534.69 1191.78 1533.13 Q1194.74 1531.54 1198.97 1531.54 Q1199.57 1531.54 1200.31 1531.63 Q1201.04 1531.7 1201.93 1531.85 L1201.96 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1220.49 1536.5 Q1215.77 1536.5 1213.04 1540.19 Q1210.3 1543.85 1210.3 1550.25 Q1210.3 1556.65 1213.01 1560.34 Q1215.74 1564 1220.49 1564 Q1225.16 1564 1227.9 1560.31 Q1230.64 1556.62 1230.64 1550.25 Q1230.64 1543.92 1227.9 1540.23 Q1225.16 1536.5 1220.49 1536.5 M1220.49 1531.54 Q1228.12 1531.54 1232.48 1536.5 Q1236.85 1541.47 1236.85 1550.25 Q1236.85 1559 1232.48 1564 Q1228.12 1568.97 1220.49 1568.97 Q1212.81 1568.97 1208.45 1564 Q1204.13 1559 1204.13 1550.25 Q1204.13 1541.47 1208.45 1536.5 Q1212.81 1531.54 1220.49 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1264.6 1518.52 L1264.6 1523.39 L1259 1523.39 Q1255.85 1523.39 1254.61 1524.66 Q1253.4 1525.93 1253.4 1529.24 L1253.4 1532.4 L1263.04 1532.4 L1263.04 1536.95 L1253.4 1536.95 L1253.4 1568.04 L1247.51 1568.04 L1247.51 1536.95 L1241.91 1536.95 L1241.91 1532.4 L1247.51 1532.4 L1247.51 1529.91 Q1247.51 1523.96 1250.28 1521.26 Q1253.05 1518.52 1259.06 1518.52 L1264.6 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1269.5 1532.4 L1275.36 1532.4 L1275.36 1568.04 L1269.5 1568.04 L1269.5 1532.4 M1269.5 1518.52 L1275.36 1518.52 L1275.36 1525.93 L1269.5 1525.93 L1269.5 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1293.4 1522.27 L1293.4 1532.4 L1305.47 1532.4 L1305.47 1536.95 L1293.4 1536.95 L1293.4 1556.3 Q1293.4 1560.66 1294.58 1561.9 Q1295.79 1563.14 1299.45 1563.14 L1305.47 1563.14 L1305.47 1568.04 L1299.45 1568.04 Q1292.67 1568.04 1290.09 1565.53 Q1287.52 1562.98 1287.52 1556.3 L1287.52 1536.95 L1283.22 1536.95 L1283.22 1532.4 L1287.52 1532.4 L1287.52 1522.27 L1293.4 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1333.35 1518.52 L1346.84 1518.52 L1346.84 1523.07 L1339.21 1523.07 L1339.21 1572.09 L1346.84 1572.09 L1346.84 1576.64 L1333.35 1576.64 L1333.35 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1375.2 1577.62 L1372.02 1577.62 L1371.99 1568.04 Q1368.65 1567.98 1365.31 1567.28 Q1361.96 1566.55 1358.59 1565.12 L1358.59 1559.39 Q1361.84 1561.42 1365.15 1562.47 Q1368.49 1563.49 1372.02 1563.52 L1372.02 1549.01 Q1364.99 1547.86 1361.77 1545.13 Q1358.59 1542.39 1358.59 1537.62 Q1358.59 1532.43 1362.06 1529.44 Q1365.53 1526.44 1372.02 1526 L1372.02 1518.52 L1375.2 1518.52 L1375.2 1525.9 Q1378.16 1526.03 1380.93 1526.54 Q1383.7 1527.02 1386.34 1527.88 L1386.34 1533.45 Q1383.7 1532.11 1380.9 1531.38 Q1378.13 1530.65 1375.2 1530.52 L1375.2 1544.11 Q1382.43 1545.22 1385.83 1548.09 Q1389.24 1550.95 1389.24 1555.92 Q1389.24 1561.3 1385.61 1564.42 Q1382.02 1567.5 1375.2 1567.98 L1375.2 1577.62 M1372.02 1543.54 L1372.02 1530.49 Q1368.33 1530.9 1366.39 1532.59 Q1364.45 1534.27 1364.45 1537.07 Q1364.45 1539.81 1366.23 1541.34 Q1368.04 1542.87 1372.02 1543.54 M1375.2 1549.65 L1375.2 1563.43 Q1379.25 1562.89 1381.28 1561.14 Q1383.35 1559.39 1383.35 1556.52 Q1383.35 1553.72 1381.38 1552.07 Q1379.44 1550.41 1375.2 1549.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1414.48 1518.52 L1414.48 1576.64 L1400.98 1576.64 L1400.98 1572.09 L1408.59 1572.09 L1408.59 1523.07 L1400.98 1523.07 L1400.98 1518.52 L1414.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1110.47 2352.76,1110.47 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,797.755 2352.76,797.755 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,485.042 2352.76,485.042 "/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,172.329 2352.76,172.329 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 205.121,47.2441 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 224.019,1423.18 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1110.47 224.019,1110.47 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,797.755 224.019,797.755 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,485.042 224.019,485.042 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,172.329 224.019,172.329 "/>
-<path clip-path="url(#clip890)" d="M157.177 1408.98 Q153.566 1408.98 151.737 1412.54 Q149.931 1416.08 149.931 1423.21 Q149.931 1430.32 151.737 1433.89 Q153.566 1437.43 157.177 1437.43 Q160.811 1437.43 162.616 1433.89 Q164.445 1430.32 164.445 1423.21 Q164.445 1416.08 162.616 1412.54 Q160.811 1408.98 157.177 1408.98 M157.177 1405.27 Q162.987 1405.27 166.042 1409.88 Q169.121 1414.46 169.121 1423.21 Q169.121 1431.94 166.042 1436.55 Q162.987 1441.13 157.177 1441.13 Q151.366 1441.13 148.288 1436.55 Q145.232 1431.94 145.232 1423.21 Q145.232 1414.46 148.288 1409.88 Q151.366 1405.27 157.177 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M117.825 1123.81 L125.464 1123.81 L125.464 1097.45 L117.154 1099.11 L117.154 1094.85 L125.418 1093.19 L130.093 1093.19 L130.093 1123.81 L137.732 1123.81 L137.732 1127.75 L117.825 1127.75 L117.825 1123.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M157.177 1096.27 Q153.566 1096.27 151.737 1099.83 Q149.931 1103.37 149.931 1110.5 Q149.931 1117.61 151.737 1121.17 Q153.566 1124.71 157.177 1124.71 Q160.811 1124.71 162.616 1121.17 Q164.445 1117.61 164.445 1110.5 Q164.445 1103.37 162.616 1099.83 Q160.811 1096.27 157.177 1096.27 M157.177 1092.56 Q162.987 1092.56 166.042 1097.17 Q169.121 1101.75 169.121 1110.5 Q169.121 1119.23 166.042 1123.84 Q162.987 1128.42 157.177 1128.42 Q151.366 1128.42 148.288 1123.84 Q145.232 1119.23 145.232 1110.5 Q145.232 1101.75 148.288 1097.17 Q151.366 1092.56 157.177 1092.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M121.043 811.099 L137.362 811.099 L137.362 815.035 L115.418 815.035 L115.418 811.099 Q118.08 808.345 122.663 803.715 Q127.269 799.062 128.45 797.72 Q130.695 795.197 131.575 793.461 Q132.478 791.701 132.478 790.011 Q132.478 787.257 130.533 785.521 Q128.612 783.785 125.51 783.785 Q123.311 783.785 120.857 784.549 Q118.427 785.312 115.649 786.863 L115.649 782.141 Q118.473 781.007 120.927 780.428 Q123.38 779.85 125.418 779.85 Q130.788 779.85 133.982 782.535 Q137.177 785.22 137.177 789.711 Q137.177 791.84 136.367 793.761 Q135.579 795.66 133.473 798.252 Q132.894 798.923 129.792 802.141 Q126.691 805.335 121.043 811.099 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M157.177 783.553 Q153.566 783.553 151.737 787.118 Q149.931 790.66 149.931 797.789 Q149.931 804.896 151.737 808.46 Q153.566 812.002 157.177 812.002 Q160.811 812.002 162.616 808.46 Q164.445 804.896 164.445 797.789 Q164.445 790.66 162.616 787.118 Q160.811 783.553 157.177 783.553 M157.177 779.85 Q162.987 779.85 166.042 784.456 Q169.121 789.039 169.121 797.789 Q169.121 806.516 166.042 811.122 Q162.987 815.706 157.177 815.706 Q151.366 815.706 148.288 811.122 Q145.232 806.516 145.232 797.789 Q145.232 789.039 148.288 784.456 Q151.366 779.85 157.177 779.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M131.181 483.688 Q134.538 484.405 136.413 486.674 Q138.311 488.942 138.311 492.276 Q138.311 497.391 134.792 500.192 Q131.274 502.993 124.793 502.993 Q122.617 502.993 120.302 502.553 Q118.01 502.137 115.556 501.28 L115.556 496.766 Q117.501 497.901 119.816 498.479 Q122.13 499.058 124.654 499.058 Q129.052 499.058 131.343 497.322 Q133.658 495.586 133.658 492.276 Q133.658 489.22 131.505 487.507 Q129.376 485.771 125.556 485.771 L121.529 485.771 L121.529 481.928 L125.742 481.928 Q129.191 481.928 131.019 480.563 Q132.848 479.174 132.848 476.581 Q132.848 473.919 130.95 472.507 Q129.075 471.072 125.556 471.072 Q123.635 471.072 121.436 471.489 Q119.237 471.905 116.598 472.785 L116.598 468.618 Q119.26 467.878 121.575 467.507 Q123.913 467.137 125.973 467.137 Q131.297 467.137 134.399 469.567 Q137.501 471.975 137.501 476.095 Q137.501 478.965 135.857 480.956 Q134.214 482.924 131.181 483.688 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M157.177 470.841 Q153.566 470.841 151.737 474.405 Q149.931 477.947 149.931 485.077 Q149.931 492.183 151.737 495.748 Q153.566 499.289 157.177 499.289 Q160.811 499.289 162.616 495.748 Q164.445 492.183 164.445 485.077 Q164.445 477.947 162.616 474.405 Q160.811 470.841 157.177 470.841 M157.177 467.137 Q162.987 467.137 166.042 471.743 Q169.121 476.327 169.121 485.077 Q169.121 493.803 166.042 498.41 Q162.987 502.993 157.177 502.993 Q151.366 502.993 148.288 498.41 Q145.232 493.803 145.232 485.077 Q145.232 476.327 148.288 471.743 Q151.366 467.137 157.177 467.137 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M129.862 159.123 L118.056 177.572 L129.862 177.572 L129.862 159.123 M128.635 155.049 L134.515 155.049 L134.515 177.572 L139.445 177.572 L139.445 181.461 L134.515 181.461 L134.515 189.609 L129.862 189.609 L129.862 181.461 L114.26 181.461 L114.26 176.947 L128.635 155.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M157.177 158.128 Q153.566 158.128 151.737 161.693 Q149.931 165.234 149.931 172.364 Q149.931 179.47 151.737 183.035 Q153.566 186.577 157.177 186.577 Q160.811 186.577 162.616 183.035 Q164.445 179.47 164.445 172.364 Q164.445 165.234 162.616 161.693 Q160.811 158.128 157.177 158.128 M157.177 154.424 Q162.987 154.424 166.042 159.031 Q169.121 163.614 169.121 172.364 Q169.121 181.091 166.042 185.697 Q162.987 190.28 157.177 190.28 Q151.366 190.28 148.288 185.697 Q145.232 181.091 145.232 172.364 Q145.232 163.614 148.288 159.031 Q151.366 154.424 157.177 154.424 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M16.4842 1073.5 L16.4842 1064.84 L56.238 1043.77 L16.4842 1043.77 L16.4842 1037.54 L64.0042 1037.54 L64.0042 1046.19 L24.2503 1067.26 L64.0042 1067.26 L64.0042 1073.5 L16.4842 1073.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M49.9359 1025.6 L28.3562 1025.6 L28.3562 1019.74 L49.7131 1019.74 Q54.7739 1019.74 57.3202 1017.77 Q59.8346 1015.8 59.8346 1011.85 Q59.8346 1007.11 56.8109 1004.37 Q53.7872 1001.6 48.5673 1001.6 L28.3562 1001.6 L28.3562 995.744 L64.0042 995.744 L64.0042 1001.6 L58.5296 1001.6 Q61.7762 1003.73 63.3676 1006.57 Q64.9272 1009.37 64.9272 1013.09 Q64.9272 1019.23 61.1078 1022.42 Q57.2883 1025.6 49.9359 1025.6 M27.4968 1010.86 L27.4968 1010.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M35.1993 955.927 Q31.2526 953.731 29.3747 950.675 Q27.4968 947.62 27.4968 943.482 Q27.4968 937.912 31.4117 934.888 Q35.2948 931.865 42.4881 931.865 L64.0042 931.865 L64.0042 937.753 L42.679 937.753 Q37.5546 937.753 35.072 939.567 Q32.5894 941.381 32.5894 945.105 Q32.5894 949.657 35.6131 952.298 Q38.6368 954.94 43.8567 954.94 L64.0042 954.94 L64.0042 960.829 L42.679 960.829 Q37.5228 960.829 35.072 962.643 Q32.5894 964.457 32.5894 968.245 Q32.5894 972.732 35.6449 975.374 Q38.6686 978.016 43.8567 978.016 L64.0042 978.016 L64.0042 983.904 L28.3562 983.904 L28.3562 978.016 L33.8944 978.016 Q30.616 976.011 29.0564 973.21 Q27.4968 970.409 27.4968 966.558 Q27.4968 962.675 29.4702 959.969 Q31.4436 957.232 35.1993 955.927 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M46.212 894.593 Q39.7508 894.593 36.0905 897.267 Q32.3984 899.909 32.3984 904.556 Q32.3984 909.203 36.0905 911.876 Q39.7508 914.518 46.212 914.518 Q52.6732 914.518 56.3653 911.876 Q60.0256 909.203 60.0256 904.556 Q60.0256 899.909 56.3653 897.267 Q52.6732 894.593 46.212 894.593 M33.7671 914.518 Q30.5842 912.672 29.0564 909.871 Q27.4968 907.038 27.4968 903.123 Q27.4968 896.63 32.6531 892.588 Q37.8093 888.514 46.212 888.514 Q54.6147 888.514 59.771 892.588 Q64.9272 896.63 64.9272 903.123 Q64.9272 907.038 63.3994 909.871 Q61.8398 912.672 58.657 914.518 L64.0042 914.518 L64.0042 920.406 L14.479 920.406 L14.479 914.518 L33.7671 914.518 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M44.7161 848.315 L47.5806 848.315 L47.5806 875.242 Q53.6281 874.86 56.8109 871.613 Q59.9619 868.335 59.9619 862.51 Q59.9619 859.136 59.1344 855.985 Q58.3069 852.802 56.6518 849.683 L62.1899 849.683 Q63.5267 852.834 64.227 856.144 Q64.9272 859.455 64.9272 862.86 Q64.9272 871.39 59.9619 876.387 Q54.9967 881.353 46.5303 881.353 Q37.7774 881.353 32.6531 876.642 Q27.4968 871.9 27.4968 863.879 Q27.4968 856.686 32.1438 852.516 Q36.7589 848.315 44.7161 848.315 M42.9973 854.171 Q38.1912 854.235 35.3266 856.877 Q32.4621 859.486 32.4621 863.815 Q32.4621 868.717 35.2312 871.677 Q38.0002 874.605 43.0292 875.051 L42.9973 854.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M33.8307 818.046 Q33.2578 819.032 33.0032 820.21 Q32.7167 821.356 32.7167 822.756 Q32.7167 827.722 35.9632 830.395 Q39.1779 833.037 45.2253 833.037 L64.0042 833.037 L64.0042 838.925 L28.3562 838.925 L28.3562 833.037 L33.8944 833.037 Q30.6479 831.191 29.0883 828.231 Q27.4968 825.271 27.4968 821.038 Q27.4968 820.433 27.5923 819.701 Q27.656 818.969 27.8151 818.078 L33.8307 818.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M32.4621 777.369 Q32.4621 782.079 36.1542 784.817 Q39.8145 787.554 46.212 787.554 Q52.6095 787.554 56.3017 784.849 Q59.9619 782.111 59.9619 777.369 Q59.9619 772.69 56.2698 769.953 Q52.5777 767.216 46.212 767.216 Q39.8781 767.216 36.186 769.953 Q32.4621 772.69 32.4621 777.369 M27.4968 777.369 Q27.4968 769.73 32.4621 765.37 Q37.4273 761.009 46.212 761.009 Q54.9649 761.009 59.9619 765.37 Q64.9272 769.73 64.9272 777.369 Q64.9272 785.04 59.9619 789.4 Q54.9649 793.729 46.212 793.729 Q37.4273 793.729 32.4621 789.4 Q27.4968 785.04 27.4968 777.369 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M14.479 733.255 L19.3487 733.255 L19.3487 738.856 Q19.3487 742.007 20.6219 743.249 Q21.895 744.458 25.2052 744.458 L28.3562 744.458 L28.3562 734.814 L32.9077 734.814 L32.9077 744.458 L64.0042 744.458 L64.0042 750.346 L32.9077 750.346 L32.9077 755.948 L28.3562 755.948 L28.3562 750.346 L25.8736 750.346 Q19.9216 750.346 17.2162 747.577 Q14.479 744.808 14.479 738.793 L14.479 733.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M32.4621 693.819 Q32.4621 698.53 36.1542 701.267 Q39.8145 704.004 46.212 704.004 Q52.6095 704.004 56.3017 701.299 Q59.9619 698.561 59.9619 693.819 Q59.9619 689.14 56.2698 686.403 Q52.5777 683.666 46.212 683.666 Q39.8781 683.666 36.186 686.403 Q32.4621 689.14 32.4621 693.819 M27.4968 693.819 Q27.4968 686.18 32.4621 681.82 Q37.4273 677.459 46.212 677.459 Q54.9649 677.459 59.9619 681.82 Q64.9272 686.18 64.9272 693.819 Q64.9272 701.49 59.9619 705.85 Q54.9649 710.179 46.212 710.179 Q37.4273 710.179 32.4621 705.85 Q27.4968 701.49 27.4968 693.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M49.9359 668.356 L28.3562 668.356 L28.3562 662.5 L49.7131 662.5 Q54.7739 662.5 57.3202 660.526 Q59.8346 658.553 59.8346 654.606 Q59.8346 649.864 56.8109 647.127 Q53.7872 644.357 48.5673 644.357 L28.3562 644.357 L28.3562 638.501 L64.0042 638.501 L64.0042 644.357 L58.5296 644.357 Q61.7762 646.49 63.3676 649.323 Q64.9272 652.124 64.9272 655.848 Q64.9272 661.99 61.1078 665.173 Q57.2883 668.356 49.9359 668.356 M27.4968 653.62 L27.4968 653.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M18.2347 620.645 L28.3562 620.645 L28.3562 608.582 L32.9077 608.582 L32.9077 620.645 L52.2594 620.645 Q56.6199 620.645 57.8613 619.468 Q59.1026 618.258 59.1026 614.598 L59.1026 608.582 L64.0042 608.582 L64.0042 614.598 Q64.0042 621.377 61.4897 623.955 Q58.9434 626.533 52.2594 626.533 L32.9077 626.533 L32.9077 630.83 L28.3562 630.83 L28.3562 626.533 L18.2347 626.533 L18.2347 620.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M29.7248 575.226 L35.1993 575.226 Q33.8307 577.708 33.1623 580.223 Q32.4621 582.706 32.4621 585.252 Q32.4621 590.949 36.0905 594.1 Q39.6872 597.251 46.212 597.251 Q52.7369 597.251 56.3653 594.1 Q59.9619 590.949 59.9619 585.252 Q59.9619 582.706 59.2935 580.223 Q58.5933 577.708 57.2247 575.226 L62.6355 575.226 Q63.7814 577.677 64.3543 580.318 Q64.9272 582.928 64.9272 585.888 Q64.9272 593.941 59.8664 598.683 Q54.8057 603.426 46.212 603.426 Q37.491 603.426 32.4939 598.652 Q27.4968 593.846 27.4968 585.506 Q27.4968 582.801 28.0697 580.223 Q28.6108 577.645 29.7248 575.226 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M32.4621 551.227 Q32.4621 555.938 36.1542 558.675 Q39.8145 561.412 46.212 561.412 Q52.6095 561.412 56.3017 558.707 Q59.9619 555.97 59.9619 551.227 Q59.9619 546.548 56.2698 543.811 Q52.5777 541.074 46.212 541.074 Q39.8781 541.074 36.186 543.811 Q32.4621 546.548 32.4621 551.227 M27.4968 551.227 Q27.4968 543.588 32.4621 539.228 Q37.4273 534.867 46.212 534.867 Q54.9649 534.867 59.9619 539.228 Q64.9272 543.588 64.9272 551.227 Q64.9272 558.898 59.9619 563.258 Q54.9649 567.587 46.212 567.587 Q37.4273 567.587 32.4621 563.258 Q27.4968 558.898 27.4968 551.227 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M35.1993 497.405 Q31.2526 495.209 29.3747 492.153 Q27.4968 489.098 27.4968 484.96 Q27.4968 479.39 31.4117 476.366 Q35.2948 473.343 42.4881 473.343 L64.0042 473.343 L64.0042 479.231 L42.679 479.231 Q37.5546 479.231 35.072 481.045 Q32.5894 482.859 32.5894 486.583 Q32.5894 491.135 35.6131 493.777 Q38.6368 496.418 43.8567 496.418 L64.0042 496.418 L64.0042 502.307 L42.679 502.307 Q37.5228 502.307 35.072 504.121 Q32.5894 505.935 32.5894 509.723 Q32.5894 514.211 35.6449 516.852 Q38.6686 519.494 43.8567 519.494 L64.0042 519.494 L64.0042 525.382 L28.3562 525.382 L28.3562 519.494 L33.8944 519.494 Q30.616 517.489 29.0564 514.688 Q27.4968 511.887 27.4968 508.036 Q27.4968 504.153 29.4702 501.447 Q31.4436 498.71 35.1993 497.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M44.7161 431.17 L47.5806 431.17 L47.5806 458.097 Q53.6281 457.715 56.8109 454.468 Q59.9619 451.19 59.9619 445.365 Q59.9619 441.992 59.1344 438.841 Q58.3069 435.658 56.6518 432.539 L62.1899 432.539 Q63.5267 435.69 64.227 439 Q64.9272 442.31 64.9272 445.716 Q64.9272 454.246 59.9619 459.243 Q54.9967 464.208 46.5303 464.208 Q37.7774 464.208 32.6531 459.497 Q27.4968 454.755 27.4968 446.734 Q27.4968 439.541 32.1438 435.371 Q36.7589 431.17 44.7161 431.17 M42.9973 437.026 Q38.1912 437.09 35.3266 439.732 Q32.4621 442.342 32.4621 446.67 Q32.4621 451.572 35.2312 454.532 Q38.0002 457.46 43.0292 457.906 L42.9973 437.026 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M29.4065 398.832 L34.9447 398.832 Q33.6716 401.315 33.035 403.988 Q32.3984 406.662 32.3984 409.527 Q32.3984 413.887 33.7352 416.083 Q35.072 418.248 37.7456 418.248 Q39.7826 418.248 40.9603 416.688 Q42.1061 415.128 43.1565 410.418 L43.6021 408.413 Q44.9389 402.174 47.3897 399.564 Q49.8086 396.922 54.1691 396.922 Q59.1344 396.922 62.0308 400.869 Q64.9272 404.784 64.9272 411.659 Q64.9272 414.524 64.3543 417.643 Q63.8132 420.73 62.6992 424.168 L56.6518 424.168 Q58.3387 420.921 59.198 417.77 Q60.0256 414.619 60.0256 411.532 Q60.0256 407.394 58.6251 405.166 Q57.1929 402.938 54.6147 402.938 Q52.2276 402.938 50.9545 404.561 Q49.6813 406.153 48.5037 411.595 L48.0262 413.632 Q46.8804 419.075 44.5251 421.494 Q42.138 423.913 38.0002 423.913 Q32.9713 423.913 30.2341 420.348 Q27.4968 416.783 27.4968 410.227 Q27.4968 406.98 27.9743 404.116 Q28.4517 401.251 29.4065 398.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip892)" d="M323.245 1298.09 L323.245 1423.18 L535.621 1423.18 L535.621 1298.09 L323.245 1298.09 L323.245 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1298.09 323.245,1423.18 535.621,1423.18 535.621,1298.09 323.245,1298.09 "/>
-<path clip-path="url(#clip892)" d="M535.621 1266.82 L535.621 1423.18 L747.997 1423.18 L747.997 1266.82 L535.621 1266.82 L535.621 1266.82  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="535.621,1266.82 535.621,1423.18 747.997,1423.18 747.997,1266.82 535.621,1266.82 "/>
-<path clip-path="url(#clip892)" d="M747.997 1298.09 L747.997 1423.18 L960.374 1423.18 L960.374 1298.09 L747.997 1298.09 L747.997 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.997,1298.09 747.997,1423.18 960.374,1423.18 960.374,1298.09 747.997,1298.09 "/>
-<path clip-path="url(#clip892)" d="M960.374 1173.01 L960.374 1423.18 L1172.75 1423.18 L1172.75 1173.01 L960.374 1173.01 L960.374 1173.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="960.374,1173.01 960.374,1423.18 1172.75,1423.18 1172.75,1173.01 960.374,1173.01 "/>
-<path clip-path="url(#clip892)" d="M1172.75 1235.55 L1172.75 1423.18 L1385.13 1423.18 L1385.13 1235.55 L1172.75 1235.55 L1172.75 1235.55  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1172.75,1235.55 1172.75,1423.18 1385.13,1423.18 1385.13,1235.55 1172.75,1235.55 "/>
-<path clip-path="url(#clip892)" d="M1385.13 1141.74 L1385.13 1423.18 L1597.5 1423.18 L1597.5 1141.74 L1385.13 1141.74 L1385.13 1141.74  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1385.13,1141.74 1385.13,1423.18 1597.5,1423.18 1597.5,1141.74 1385.13,1141.74 "/>
-<path clip-path="url(#clip892)" d="M1597.5 1047.92 L1597.5 1423.18 L1809.88 1423.18 L1809.88 1047.92 L1597.5 1047.92 L1597.5 1047.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1597.5,1047.92 1597.5,1423.18 1809.88,1423.18 1809.88,1047.92 1597.5,1047.92 "/>
-<path clip-path="url(#clip892)" d="M1809.88 1173.01 L1809.88 1423.18 L2022.26 1423.18 L2022.26 1173.01 L1809.88 1173.01 L1809.88 1173.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1809.88,1173.01 1809.88,1423.18 2022.26,1423.18 2022.26,1173.01 1809.88,1173.01 "/>
-<path clip-path="url(#clip892)" d="M2022.26 47.2441 L2022.26 1423.18 L2234.63 1423.18 L2234.63 47.2441 L2022.26 47.2441 L2022.26 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2022.26,47.2441 2022.26,1423.18 2234.63,1423.18 2234.63,47.2441 2022.26,47.2441 "/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="429.433" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="641.809" cy="1266.82" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="854.186" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1066.56" cy="1173.01" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1278.94" cy="1235.55" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1491.31" cy="1141.74" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1703.69" cy="1047.92" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1916.07" cy="1173.01" r="2"/>
-<circle clip-path="url(#clip892)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2128.44" cy="47.2441" r="2"/>
-<polyline clip-path="url(#clip892)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1638.4,2799.12 1638.4,-1328.69 "/>
-<path clip-path="url(#clip890)" d="M276.709 196.789 L1017.7 196.789 L1017.7 93.1086 L276.709 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="276.709,196.789 1017.7,196.789 1017.7,93.1086 276.709,93.1086 276.709,196.789 "/>
-<polyline clip-path="url(#clip890)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,144.949 443.747,144.949 "/>
-<path clip-path="url(#clip890)" d="M467.61 127.669 L489.461 127.669 L489.461 131.604 L472.286 131.604 L472.286 141.835 L488.744 141.835 L488.744 145.77 L472.286 145.77 L472.286 158.293 L489.878 158.293 L489.878 162.229 L467.61 162.229 L467.61 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M518.929 136.303 L509.554 148.919 L519.415 162.229 L514.392 162.229 L506.846 152.043 L499.299 162.229 L494.276 162.229 L504.346 148.664 L495.133 136.303 L500.156 136.303 L507.031 145.539 L513.906 136.303 L518.929 136.303 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M529.554 158.34 L529.554 172.09 L525.271 172.09 L525.271 136.303 L529.554 136.303 L529.554 140.238 Q530.896 137.923 532.933 136.812 Q534.994 135.678 537.841 135.678 Q542.563 135.678 545.503 139.428 Q548.466 143.178 548.466 149.289 Q548.466 155.4 545.503 159.15 Q542.563 162.9 537.841 162.9 Q534.994 162.9 532.933 161.789 Q530.896 160.655 529.554 158.34 M544.044 149.289 Q544.044 144.59 542.1 141.928 Q540.179 139.243 536.799 139.243 Q533.42 139.243 531.475 141.928 Q529.554 144.59 529.554 149.289 Q529.554 153.988 531.475 156.673 Q533.42 159.335 536.799 159.335 Q540.179 159.335 542.1 156.673 Q544.044 153.988 544.044 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M577.702 148.201 L577.702 150.284 L558.118 150.284 Q558.396 154.682 560.757 156.997 Q563.142 159.289 567.378 159.289 Q569.831 159.289 572.123 158.687 Q574.438 158.085 576.706 156.881 L576.706 160.909 Q574.415 161.881 572.007 162.391 Q569.6 162.9 567.123 162.9 Q560.919 162.9 557.285 159.289 Q553.674 155.678 553.674 149.52 Q553.674 143.155 557.1 139.428 Q560.549 135.678 566.382 135.678 Q571.614 135.678 574.646 139.057 Q577.702 142.414 577.702 148.201 M573.442 146.951 Q573.396 143.456 571.475 141.372 Q569.577 139.289 566.429 139.289 Q562.864 139.289 560.711 141.303 Q558.581 143.317 558.257 146.974 L573.442 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M603.35 137.298 L603.35 141.28 Q601.544 140.284 599.715 139.798 Q597.91 139.289 596.058 139.289 Q591.915 139.289 589.623 141.928 Q587.331 144.544 587.331 149.289 Q587.331 154.034 589.623 156.673 Q591.915 159.289 596.058 159.289 Q597.91 159.289 599.715 158.803 Q601.544 158.293 603.35 157.298 L603.35 161.233 Q601.567 162.067 599.646 162.483 Q597.748 162.9 595.595 162.9 Q589.739 162.9 586.29 159.219 Q582.841 155.539 582.841 149.289 Q582.841 142.946 586.313 139.312 Q589.808 135.678 595.873 135.678 Q597.84 135.678 599.715 136.095 Q601.59 136.488 603.35 137.298 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M614.97 128.942 L614.97 136.303 L623.743 136.303 L623.743 139.613 L614.97 139.613 L614.97 153.687 Q614.97 156.858 615.826 157.761 Q616.706 158.664 619.368 158.664 L623.743 158.664 L623.743 162.229 L619.368 162.229 Q614.438 162.229 612.563 160.4 Q610.688 158.548 610.688 153.687 L610.688 139.613 L607.563 139.613 L607.563 136.303 L610.688 136.303 L610.688 128.942 L614.97 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M651.521 148.201 L651.521 150.284 L631.938 150.284 Q632.215 154.682 634.576 156.997 Q636.961 159.289 641.197 159.289 Q643.65 159.289 645.942 158.687 Q648.257 158.085 650.525 156.881 L650.525 160.909 Q648.234 161.881 645.826 162.391 Q643.419 162.9 640.942 162.9 Q634.738 162.9 631.104 159.289 Q627.493 155.678 627.493 149.52 Q627.493 143.155 630.919 139.428 Q634.368 135.678 640.201 135.678 Q645.433 135.678 648.465 139.057 Q651.521 142.414 651.521 148.201 M647.262 146.951 Q647.215 143.456 645.294 141.372 Q643.396 139.289 640.248 139.289 Q636.683 139.289 634.53 141.303 Q632.4 143.317 632.076 146.974 L647.262 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M675.572 140.238 L675.572 126.21 L679.831 126.21 L679.831 162.229 L675.572 162.229 L675.572 158.34 Q674.229 160.655 672.169 161.789 Q670.132 162.9 667.261 162.9 Q662.562 162.9 659.599 159.15 Q656.66 155.4 656.66 149.289 Q656.66 143.178 659.599 139.428 Q662.562 135.678 667.261 135.678 Q670.132 135.678 672.169 136.812 Q674.229 137.923 675.572 140.238 M661.058 149.289 Q661.058 153.988 662.979 156.673 Q664.923 159.335 668.303 159.335 Q671.683 159.335 673.627 156.673 Q675.572 153.988 675.572 149.289 Q675.572 144.59 673.627 141.928 Q671.683 139.243 668.303 139.243 Q664.923 139.243 662.979 141.928 Q661.058 144.59 661.058 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M707.794 158.34 L707.794 172.09 L703.511 172.09 L703.511 136.303 L707.794 136.303 L707.794 140.238 Q709.136 137.923 711.173 136.812 Q713.233 135.678 716.081 135.678 Q720.803 135.678 723.743 139.428 Q726.706 143.178 726.706 149.289 Q726.706 155.4 723.743 159.15 Q720.803 162.9 716.081 162.9 Q713.233 162.9 711.173 161.789 Q709.136 160.655 707.794 158.34 M722.284 149.289 Q722.284 144.59 720.34 141.928 Q718.419 139.243 715.039 139.243 Q711.659 139.243 709.715 141.928 Q707.794 144.59 707.794 149.289 Q707.794 153.988 709.715 156.673 Q711.659 159.335 715.039 159.335 Q718.419 159.335 720.34 156.673 Q722.284 153.988 722.284 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M748.789 140.284 Q748.071 139.868 747.215 139.682 Q746.381 139.474 745.363 139.474 Q741.752 139.474 739.807 141.835 Q737.886 144.173 737.886 148.571 L737.886 162.229 L733.604 162.229 L733.604 136.303 L737.886 136.303 L737.886 140.331 Q739.229 137.969 741.381 136.835 Q743.534 135.678 746.613 135.678 Q747.053 135.678 747.585 135.747 Q748.117 135.794 748.766 135.909 L748.789 140.284 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M762.261 139.289 Q758.835 139.289 756.844 141.974 Q754.854 144.636 754.854 149.289 Q754.854 153.942 756.821 156.627 Q758.812 159.289 762.261 159.289 Q765.664 159.289 767.654 156.604 Q769.645 153.918 769.645 149.289 Q769.645 144.682 767.654 141.997 Q765.664 139.289 762.261 139.289 M762.261 135.678 Q767.816 135.678 770.988 139.289 Q774.159 142.9 774.159 149.289 Q774.159 155.655 770.988 159.289 Q767.816 162.9 762.261 162.9 Q756.682 162.9 753.511 159.289 Q750.363 155.655 750.363 149.289 Q750.363 142.9 753.511 139.289 Q756.682 135.678 762.261 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M794.344 126.21 L794.344 129.752 L790.27 129.752 Q787.978 129.752 787.076 130.678 Q786.196 131.604 786.196 134.011 L786.196 136.303 L793.21 136.303 L793.21 139.613 L786.196 139.613 L786.196 162.229 L781.914 162.229 L781.914 139.613 L777.84 139.613 L777.84 136.303 L781.914 136.303 L781.914 134.497 Q781.914 130.169 783.927 128.201 Q785.941 126.21 790.316 126.21 L794.344 126.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M797.909 136.303 L802.168 136.303 L802.168 162.229 L797.909 162.229 L797.909 136.303 M797.909 126.21 L802.168 126.21 L802.168 131.604 L797.909 131.604 L797.909 126.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M815.293 128.942 L815.293 136.303 L824.066 136.303 L824.066 139.613 L815.293 139.613 L815.293 153.687 Q815.293 156.858 816.149 157.761 Q817.029 158.664 819.691 158.664 L824.066 158.664 L824.066 162.229 L819.691 162.229 Q814.761 162.229 812.886 160.4 Q811.011 158.548 811.011 153.687 L811.011 139.613 L807.886 139.613 L807.886 136.303 L811.011 136.303 L811.011 128.942 L815.293 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M854.969 126.257 Q851.867 131.581 850.362 136.789 Q848.858 141.997 848.858 147.344 Q848.858 152.692 850.362 157.946 Q851.89 163.178 854.969 168.479 L851.265 168.479 Q847.793 163.039 846.057 157.784 Q844.344 152.53 844.344 147.344 Q844.344 142.182 846.057 136.951 Q847.77 131.72 851.265 126.257 L854.969 126.257 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M874.783 169.196 L872.469 169.196 L872.445 162.229 Q870.015 162.182 867.584 161.673 Q865.154 161.141 862.7 160.099 L862.7 155.932 Q865.061 157.414 867.469 158.178 Q869.899 158.918 872.469 158.942 L872.469 148.386 Q867.353 147.553 865.015 145.562 Q862.7 143.571 862.7 140.099 Q862.7 136.326 865.223 134.15 Q867.746 131.974 872.469 131.65 L872.469 126.21 L874.783 126.21 L874.783 131.581 Q876.936 131.673 878.95 132.044 Q880.964 132.391 882.885 133.016 L882.885 137.067 Q880.964 136.095 878.927 135.562 Q876.913 135.03 874.783 134.937 L874.783 144.821 Q880.038 145.631 882.515 147.715 Q884.992 149.798 884.992 153.409 Q884.992 157.321 882.353 159.59 Q879.737 161.835 874.783 162.182 L874.783 169.196 M872.469 144.405 L872.469 134.914 Q869.783 135.215 868.371 136.442 Q866.959 137.669 866.959 139.706 Q866.959 141.696 868.256 142.807 Q869.575 143.919 872.469 144.405 M874.783 148.849 L874.783 158.872 Q877.723 158.479 879.205 157.205 Q880.709 155.932 880.709 153.849 Q880.709 151.812 879.274 150.608 Q877.862 149.405 874.783 148.849 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M894.043 127.669 L912.399 127.669 L912.399 131.604 L898.325 131.604 L898.325 140.076 Q899.343 139.729 900.362 139.567 Q901.381 139.382 902.399 139.382 Q908.186 139.382 911.566 142.553 Q914.945 145.724 914.945 151.141 Q914.945 156.719 911.473 159.821 Q908.001 162.9 901.681 162.9 Q899.506 162.9 897.237 162.53 Q894.992 162.159 892.584 161.418 L892.584 156.719 Q894.668 157.854 896.89 158.409 Q899.112 158.965 901.589 158.965 Q905.593 158.965 907.931 156.858 Q910.269 154.752 910.269 151.141 Q910.269 147.53 907.931 145.423 Q905.593 143.317 901.589 143.317 Q899.714 143.317 897.839 143.733 Q895.987 144.15 894.043 145.03 L894.043 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M924.204 127.669 L942.561 127.669 L942.561 131.604 L928.487 131.604 L928.487 140.076 Q929.505 139.729 930.524 139.567 Q931.542 139.382 932.561 139.382 Q938.348 139.382 941.728 142.553 Q945.107 145.724 945.107 151.141 Q945.107 156.719 941.635 159.821 Q938.163 162.9 931.843 162.9 Q929.667 162.9 927.399 162.53 Q925.154 162.159 922.746 161.418 L922.746 156.719 Q924.829 157.854 927.052 158.409 Q929.274 158.965 931.751 158.965 Q935.755 158.965 938.093 156.858 Q940.431 154.752 940.431 151.141 Q940.431 147.53 938.093 145.423 Q935.755 143.317 931.751 143.317 Q929.876 143.317 928.001 143.733 Q926.149 144.15 924.204 145.03 L924.204 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M954.366 127.669 L972.723 127.669 L972.723 131.604 L958.649 131.604 L958.649 140.076 Q959.667 139.729 960.686 139.567 Q961.704 139.382 962.723 139.382 Q968.51 139.382 971.889 142.553 Q975.269 145.724 975.269 151.141 Q975.269 156.719 971.797 159.821 Q968.325 162.9 962.005 162.9 Q959.829 162.9 957.561 162.53 Q955.315 162.159 952.908 161.418 L952.908 156.719 Q954.991 157.854 957.214 158.409 Q959.436 158.965 961.913 158.965 Q965.917 158.965 968.255 156.858 Q970.593 154.752 970.593 151.141 Q970.593 147.53 968.255 145.423 Q965.917 143.317 961.913 143.317 Q960.038 143.317 958.163 143.733 Q956.311 144.15 954.366 145.03 L954.366 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M983.209 126.257 L986.912 126.257 Q990.385 131.72 992.098 136.951 Q993.834 142.182 993.834 147.344 Q993.834 152.53 992.098 157.784 Q990.385 163.039 986.912 168.479 L983.209 168.479 Q986.287 163.178 987.792 157.946 Q989.32 152.692 989.32 147.344 Q989.32 141.997 987.792 136.789 Q986.287 131.581 983.209 126.257 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/8f63bbf6.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/8f63bbf6.svg
deleted file mode 100644
index 72dfcdcca5c..00000000000
--- a/previews/PR3562/tutorials/applications/two_stage_stochastic/8f63bbf6.svg
+++ /dev/null
@@ -1,96 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
-<defs>
-  <clipPath id="clip830">
-    <rect x="0" y="0" width="2400" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip830)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip831">
-    <rect x="480" y="0" width="1681" height="1600"/>
-  </clipPath>
-</defs>
-<path clip-path="url(#clip830)" d="M184.191 1423.18 L2352.76 1423.18 L2352.76 47.2441 L184.191 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<defs>
-  <clipPath id="clip832">
-    <rect x="184" y="47" width="2170" height="1377"/>
-  </clipPath>
-</defs>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="303.466,1423.18 303.466,47.2441 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="732.358,1423.18 732.358,47.2441 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1161.25,1423.18 1161.25,47.2441 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1590.14,1423.18 1590.14,47.2441 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2019.03,1423.18 2019.03,47.2441 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.466,1423.18 303.466,1404.28 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.358,1423.18 732.358,1404.28 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1161.25,1423.18 1161.25,1404.28 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1590.14,1423.18 1590.14,1404.28 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.03,1423.18 2019.03,1404.28 "/>
-<path clip-path="url(#clip830)" d="M263.072 1481.64 L270.711 1481.64 L270.711 1455.28 L262.401 1456.95 L262.401 1452.69 L270.665 1451.02 L275.341 1451.02 L275.341 1481.64 L282.98 1481.64 L282.98 1485.58 L263.072 1485.58 L263.072 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M303.003 1466.44 Q299.855 1466.44 298.003 1468.59 Q296.174 1470.74 296.174 1474.49 Q296.174 1478.22 298.003 1480.39 Q299.855 1482.55 303.003 1482.55 Q306.151 1482.55 307.979 1480.39 Q309.831 1478.22 309.831 1474.49 Q309.831 1470.74 307.979 1468.59 Q306.151 1466.44 303.003 1466.44 M312.285 1451.78 L312.285 1456.04 Q310.526 1455.21 308.72 1454.77 Q306.938 1454.33 305.179 1454.33 Q300.549 1454.33 298.095 1457.45 Q295.665 1460.58 295.318 1466.9 Q296.683 1464.89 298.743 1463.82 Q300.804 1462.73 303.28 1462.73 Q308.489 1462.73 311.498 1465.9 Q314.53 1469.05 314.53 1474.49 Q314.53 1479.82 311.382 1483.03 Q308.234 1486.25 303.003 1486.25 Q297.007 1486.25 293.836 1481.67 Q290.665 1477.06 290.665 1468.33 Q290.665 1460.14 294.554 1455.28 Q298.443 1450.39 304.993 1450.39 Q306.753 1450.39 308.535 1450.74 Q310.341 1451.09 312.285 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M332.586 1454.1 Q328.975 1454.1 327.146 1457.66 Q325.341 1461.2 325.341 1468.33 Q325.341 1475.44 327.146 1479.01 Q328.975 1482.55 332.586 1482.55 Q336.22 1482.55 338.026 1479.01 Q339.854 1475.44 339.854 1468.33 Q339.854 1461.2 338.026 1457.66 Q336.22 1454.1 332.586 1454.1 M332.586 1450.39 Q338.396 1450.39 341.452 1455 Q344.53 1459.58 344.53 1468.33 Q344.53 1477.06 341.452 1481.67 Q338.396 1486.25 332.586 1486.25 Q326.776 1486.25 323.697 1481.67 Q320.641 1477.06 320.641 1468.33 Q320.641 1459.58 323.697 1455 Q326.776 1450.39 332.586 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M691.965 1481.64 L699.603 1481.64 L699.603 1455.28 L691.293 1456.95 L691.293 1452.69 L699.557 1451.02 L704.233 1451.02 L704.233 1481.64 L711.872 1481.64 L711.872 1485.58 L691.965 1485.58 L691.965 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M731.316 1469.17 Q727.983 1469.17 726.062 1470.95 Q724.164 1472.73 724.164 1475.86 Q724.164 1478.98 726.062 1480.77 Q727.983 1482.55 731.316 1482.55 Q734.65 1482.55 736.571 1480.77 Q738.492 1478.96 738.492 1475.86 Q738.492 1472.73 736.571 1470.95 Q734.673 1469.17 731.316 1469.17 M726.64 1467.18 Q723.631 1466.44 721.941 1464.38 Q720.275 1462.32 720.275 1459.35 Q720.275 1455.21 723.214 1452.8 Q726.177 1450.39 731.316 1450.39 Q736.478 1450.39 739.418 1452.8 Q742.358 1455.21 742.358 1459.35 Q742.358 1462.32 740.668 1464.38 Q739.001 1466.44 736.015 1467.18 Q739.395 1467.96 741.27 1470.26 Q743.168 1472.55 743.168 1475.86 Q743.168 1480.88 740.089 1483.57 Q737.034 1486.25 731.316 1486.25 Q725.599 1486.25 722.52 1483.57 Q719.464 1480.88 719.464 1475.86 Q719.464 1472.55 721.363 1470.26 Q723.261 1467.96 726.64 1467.18 M724.927 1459.79 Q724.927 1462.48 726.594 1463.98 Q728.284 1465.49 731.316 1465.49 Q734.326 1465.49 736.015 1463.98 Q737.728 1462.48 737.728 1459.79 Q737.728 1457.11 736.015 1455.6 Q734.326 1454.1 731.316 1454.1 Q728.284 1454.1 726.594 1455.6 Q724.927 1457.11 724.927 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M761.478 1454.1 Q757.867 1454.1 756.038 1457.66 Q754.233 1461.2 754.233 1468.33 Q754.233 1475.44 756.038 1479.01 Q757.867 1482.55 761.478 1482.55 Q765.112 1482.55 766.918 1479.01 Q768.747 1475.44 768.747 1468.33 Q768.747 1461.2 766.918 1457.66 Q765.112 1454.1 761.478 1454.1 M761.478 1450.39 Q767.288 1450.39 770.344 1455 Q773.423 1459.58 773.423 1468.33 Q773.423 1477.06 770.344 1481.67 Q767.288 1486.25 761.478 1486.25 Q755.668 1486.25 752.589 1481.67 Q749.534 1477.06 749.534 1468.33 Q749.534 1459.58 752.589 1455 Q755.668 1450.39 761.478 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1124.94 1481.64 L1141.26 1481.64 L1141.26 1485.58 L1119.32 1485.58 L1119.32 1481.64 Q1121.98 1478.89 1126.56 1474.26 Q1131.17 1469.61 1132.35 1468.27 Q1134.6 1465.74 1135.47 1464.01 Q1136.38 1462.25 1136.38 1460.56 Q1136.38 1457.8 1134.43 1456.07 Q1132.51 1454.33 1129.41 1454.33 Q1127.21 1454.33 1124.76 1455.09 Q1122.33 1455.86 1119.55 1457.41 L1119.55 1452.69 Q1122.37 1451.55 1124.83 1450.97 Q1127.28 1450.39 1129.32 1450.39 Q1134.69 1450.39 1137.88 1453.08 Q1141.08 1455.77 1141.08 1460.26 Q1141.08 1462.39 1140.27 1464.31 Q1139.48 1466.2 1137.37 1468.8 Q1136.79 1469.47 1133.69 1472.69 Q1130.59 1475.88 1124.94 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1161.08 1454.1 Q1157.47 1454.1 1155.64 1457.66 Q1153.83 1461.2 1153.83 1468.33 Q1153.83 1475.44 1155.64 1479.01 Q1157.47 1482.55 1161.08 1482.55 Q1164.71 1482.55 1166.52 1479.01 Q1168.35 1475.44 1168.35 1468.33 Q1168.35 1461.2 1166.52 1457.66 Q1164.71 1454.1 1161.08 1454.1 M1161.08 1450.39 Q1166.89 1450.39 1169.94 1455 Q1173.02 1459.58 1173.02 1468.33 Q1173.02 1477.06 1169.94 1481.67 Q1166.89 1486.25 1161.08 1486.25 Q1155.27 1486.25 1152.19 1481.67 Q1149.13 1477.06 1149.13 1468.33 Q1149.13 1459.58 1152.19 1455 Q1155.27 1450.39 1161.08 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1191.24 1454.1 Q1187.63 1454.1 1185.8 1457.66 Q1183.99 1461.2 1183.99 1468.33 Q1183.99 1475.44 1185.8 1479.01 Q1187.63 1482.55 1191.24 1482.55 Q1194.87 1482.55 1196.68 1479.01 Q1198.51 1475.44 1198.51 1468.33 Q1198.51 1461.2 1196.68 1457.66 Q1194.87 1454.1 1191.24 1454.1 M1191.24 1450.39 Q1197.05 1450.39 1200.1 1455 Q1203.18 1459.58 1203.18 1468.33 Q1203.18 1477.06 1200.1 1481.67 Q1197.05 1486.25 1191.24 1486.25 Q1185.43 1486.25 1182.35 1481.67 Q1179.29 1477.06 1179.29 1468.33 Q1179.29 1459.58 1182.35 1455 Q1185.43 1450.39 1191.24 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1553.83 1481.64 L1570.15 1481.64 L1570.15 1485.58 L1548.21 1485.58 L1548.21 1481.64 Q1550.87 1478.89 1555.46 1474.26 Q1560.06 1469.61 1561.24 1468.27 Q1563.49 1465.74 1564.37 1464.01 Q1565.27 1462.25 1565.27 1460.56 Q1565.27 1457.8 1563.33 1456.07 Q1561.4 1454.33 1558.3 1454.33 Q1556.1 1454.33 1553.65 1455.09 Q1551.22 1455.86 1548.44 1457.41 L1548.44 1452.69 Q1551.27 1451.55 1553.72 1450.97 Q1556.17 1450.39 1558.21 1450.39 Q1563.58 1450.39 1566.77 1453.08 Q1569.97 1455.77 1569.97 1460.26 Q1569.97 1462.39 1569.16 1464.31 Q1568.37 1466.2 1566.27 1468.8 Q1565.69 1469.47 1562.58 1472.69 Q1559.48 1475.88 1553.83 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1584 1481.64 L1600.32 1481.64 L1600.32 1485.58 L1578.37 1485.58 L1578.37 1481.64 Q1581.03 1478.89 1585.62 1474.26 Q1590.22 1469.61 1591.4 1468.27 Q1593.65 1465.74 1594.53 1464.01 Q1595.43 1462.25 1595.43 1460.56 Q1595.43 1457.8 1593.49 1456.07 Q1591.57 1454.33 1588.46 1454.33 Q1586.27 1454.33 1583.81 1455.09 Q1581.38 1455.86 1578.6 1457.41 L1578.6 1452.69 Q1581.43 1451.55 1583.88 1450.97 Q1586.33 1450.39 1588.37 1450.39 Q1593.74 1450.39 1596.94 1453.08 Q1600.13 1455.77 1600.13 1460.26 Q1600.13 1462.39 1599.32 1464.31 Q1598.53 1466.2 1596.43 1468.8 Q1595.85 1469.47 1592.75 1472.69 Q1589.64 1475.88 1584 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1620.13 1454.1 Q1616.52 1454.1 1614.69 1457.66 Q1612.89 1461.2 1612.89 1468.33 Q1612.89 1475.44 1614.69 1479.01 Q1616.52 1482.55 1620.13 1482.55 Q1623.77 1482.55 1625.57 1479.01 Q1627.4 1475.44 1627.4 1468.33 Q1627.4 1461.2 1625.57 1457.66 Q1623.77 1454.1 1620.13 1454.1 M1620.13 1450.39 Q1625.94 1450.39 1629 1455 Q1632.08 1459.58 1632.08 1468.33 Q1632.08 1477.06 1629 1481.67 Q1625.94 1486.25 1620.13 1486.25 Q1614.32 1486.25 1611.24 1481.67 Q1608.19 1477.06 1608.19 1468.33 Q1608.19 1459.58 1611.24 1455 Q1614.32 1450.39 1620.13 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1982.73 1481.64 L1999.05 1481.64 L1999.05 1485.58 L1977.1 1485.58 L1977.1 1481.64 Q1979.76 1478.89 1984.35 1474.26 Q1988.95 1469.61 1990.13 1468.27 Q1992.38 1465.74 1993.26 1464.01 Q1994.16 1462.25 1994.16 1460.56 Q1994.16 1457.8 1992.22 1456.07 Q1990.3 1454.33 1987.19 1454.33 Q1985 1454.33 1982.54 1455.09 Q1980.11 1455.86 1977.33 1457.41 L1977.33 1452.69 Q1980.16 1451.55 1982.61 1450.97 Q1985.07 1450.39 1987.1 1450.39 Q1992.47 1450.39 1995.67 1453.08 Q1998.86 1455.77 1998.86 1460.26 Q1998.86 1462.39 1998.05 1464.31 Q1997.26 1466.2 1995.16 1468.8 Q1994.58 1469.47 1991.48 1472.69 Q1988.38 1475.88 1982.73 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M2021.71 1455.09 L2009.9 1473.54 L2021.71 1473.54 L2021.71 1455.09 M2020.48 1451.02 L2026.36 1451.02 L2026.36 1473.54 L2031.29 1473.54 L2031.29 1477.43 L2026.36 1477.43 L2026.36 1485.58 L2021.71 1485.58 L2021.71 1477.43 L2006.11 1477.43 L2006.11 1472.92 L2020.48 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M2049.02 1454.1 Q2045.41 1454.1 2043.58 1457.66 Q2041.78 1461.2 2041.78 1468.33 Q2041.78 1475.44 2043.58 1479.01 Q2045.41 1482.55 2049.02 1482.55 Q2052.66 1482.55 2054.46 1479.01 Q2056.29 1475.44 2056.29 1468.33 Q2056.29 1461.2 2054.46 1457.66 Q2052.66 1454.1 2049.02 1454.1 M2049.02 1450.39 Q2054.83 1450.39 2057.89 1455 Q2060.97 1459.58 2060.97 1468.33 Q2060.97 1477.06 2057.89 1481.67 Q2054.83 1486.25 2049.02 1486.25 Q2043.21 1486.25 2040.13 1481.67 Q2037.08 1477.06 2037.08 1468.33 Q2037.08 1459.58 2040.13 1455 Q2043.21 1450.39 2049.02 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1142.85 1525.81 L1142.85 1562.76 L1150.61 1562.76 Q1160.45 1562.76 1165 1558.3 Q1169.58 1553.85 1169.58 1544.24 Q1169.58 1534.69 1165 1530.26 Q1160.45 1525.81 1150.61 1525.81 L1142.85 1525.81 M1136.42 1520.52 L1149.63 1520.52 Q1163.44 1520.52 1169.9 1526.28 Q1176.36 1532.01 1176.36 1544.24 Q1176.36 1556.52 1169.87 1562.28 Q1163.38 1568.04 1149.63 1568.04 L1136.42 1568.04 L1136.42 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1216.85 1548.76 L1216.85 1551.62 L1189.92 1551.62 Q1190.3 1557.67 1193.55 1560.85 Q1196.83 1564 1202.65 1564 Q1206.03 1564 1209.18 1563.17 Q1212.36 1562.35 1215.48 1560.69 L1215.48 1566.23 Q1212.33 1567.57 1209.02 1568.27 Q1205.71 1568.97 1202.3 1568.97 Q1193.77 1568.97 1188.77 1564 Q1183.81 1559.04 1183.81 1550.57 Q1183.81 1541.82 1188.52 1536.69 Q1193.26 1531.54 1201.28 1531.54 Q1208.48 1531.54 1212.65 1536.18 Q1216.85 1540.8 1216.85 1548.76 M1210.99 1547.04 Q1210.93 1542.23 1208.29 1539.37 Q1205.68 1536.5 1201.35 1536.5 Q1196.45 1536.5 1193.49 1539.27 Q1190.56 1542.04 1190.11 1547.07 L1210.99 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1254.21 1539.24 Q1256.41 1535.29 1259.47 1533.41 Q1262.52 1531.54 1266.66 1531.54 Q1272.23 1531.54 1275.25 1535.45 Q1278.28 1539.33 1278.28 1546.53 L1278.28 1568.04 L1272.39 1568.04 L1272.39 1546.72 Q1272.39 1541.59 1270.57 1539.11 Q1268.76 1536.63 1265.04 1536.63 Q1260.48 1536.63 1257.84 1539.65 Q1255.2 1542.68 1255.2 1547.9 L1255.2 1568.04 L1249.31 1568.04 L1249.31 1546.72 Q1249.31 1541.56 1247.5 1539.11 Q1245.68 1536.63 1241.9 1536.63 Q1237.41 1536.63 1234.77 1539.68 Q1232.13 1542.71 1232.13 1547.9 L1232.13 1568.04 L1226.24 1568.04 L1226.24 1532.4 L1232.13 1532.4 L1232.13 1537.93 Q1234.13 1534.66 1236.93 1533.1 Q1239.73 1531.54 1243.58 1531.54 Q1247.47 1531.54 1250.17 1533.51 Q1252.91 1535.48 1254.21 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1306.16 1550.12 Q1299.06 1550.12 1296.32 1551.75 Q1293.59 1553.37 1293.59 1557.29 Q1293.59 1560.4 1295.62 1562.25 Q1297.69 1564.07 1301.22 1564.07 Q1306.09 1564.07 1309.02 1560.63 Q1311.98 1557.16 1311.98 1551.43 L1311.98 1550.12 L1306.16 1550.12 M1317.84 1547.71 L1317.84 1568.04 L1311.98 1568.04 L1311.98 1562.63 Q1309.98 1565.88 1306.99 1567.44 Q1303.99 1568.97 1299.67 1568.97 Q1294.19 1568.97 1290.94 1565.91 Q1287.73 1562.82 1287.73 1557.67 Q1287.73 1551.65 1291.74 1548.6 Q1295.78 1545.54 1303.77 1545.54 L1311.98 1545.54 L1311.98 1544.97 Q1311.98 1540.93 1309.31 1538.73 Q1306.67 1536.5 1301.86 1536.5 Q1298.81 1536.5 1295.91 1537.23 Q1293.01 1537.97 1290.34 1539.43 L1290.34 1534.02 Q1293.55 1532.78 1296.58 1532.17 Q1299.6 1531.54 1302.47 1531.54 Q1310.2 1531.54 1314.02 1535.55 Q1317.84 1539.56 1317.84 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1359.53 1546.53 L1359.53 1568.04 L1353.68 1568.04 L1353.68 1546.72 Q1353.68 1541.66 1351.7 1539.14 Q1349.73 1536.63 1345.78 1536.63 Q1341.04 1536.63 1338.31 1539.65 Q1335.57 1542.68 1335.57 1547.9 L1335.57 1568.04 L1329.68 1568.04 L1329.68 1532.4 L1335.57 1532.4 L1335.57 1537.93 Q1337.67 1534.72 1340.5 1533.13 Q1343.37 1531.54 1347.09 1531.54 Q1353.23 1531.54 1356.38 1535.36 Q1359.53 1539.14 1359.53 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M1394.67 1537.81 L1394.67 1518.52 L1400.53 1518.52 L1400.53 1568.04 L1394.67 1568.04 L1394.67 1562.7 Q1392.83 1565.88 1389.99 1567.44 Q1387.19 1568.97 1383.25 1568.97 Q1376.79 1568.97 1372.71 1563.81 Q1368.67 1558.65 1368.67 1550.25 Q1368.67 1541.85 1372.71 1536.69 Q1376.79 1531.54 1383.25 1531.54 Q1387.19 1531.54 1389.99 1533.1 Q1392.83 1534.62 1394.67 1537.81 M1374.72 1550.25 Q1374.72 1556.71 1377.36 1560.4 Q1380.03 1564.07 1384.68 1564.07 Q1389.33 1564.07 1392 1560.4 Q1394.67 1556.71 1394.67 1550.25 Q1394.67 1543.79 1392 1540.13 Q1389.33 1536.44 1384.68 1536.44 Q1380.03 1536.44 1377.36 1540.13 Q1374.72 1543.79 1374.72 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="184.191,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="184.191,1110.47 2352.76,1110.47 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="184.191,797.755 2352.76,797.755 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="184.191,485.042 2352.76,485.042 "/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="184.191,172.329 2352.76,172.329 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,1423.18 184.191,47.2441 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,1423.18 203.088,1423.18 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,1110.47 203.088,1110.47 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,797.755 203.088,797.755 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,485.042 203.088,485.042 "/>
-<polyline clip-path="url(#clip830)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="184.191,172.329 203.088,172.329 "/>
-<path clip-path="url(#clip830)" d="M91.0151 1408.98 Q87.404 1408.98 85.5753 1412.54 Q83.7697 1416.08 83.7697 1423.21 Q83.7697 1430.32 85.5753 1433.89 Q87.404 1437.43 91.0151 1437.43 Q94.6493 1437.43 96.4548 1433.89 Q98.2835 1430.32 98.2835 1423.21 Q98.2835 1416.08 96.4548 1412.54 Q94.6493 1408.98 91.0151 1408.98 M91.0151 1405.27 Q96.8252 1405.27 99.8808 1409.88 Q102.959 1414.46 102.959 1423.21 Q102.959 1431.94 99.8808 1436.55 Q96.8252 1441.13 91.0151 1441.13 Q85.2049 1441.13 82.1262 1436.55 Q79.0707 1431.94 79.0707 1423.21 Q79.0707 1414.46 82.1262 1409.88 Q85.2049 1405.27 91.0151 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M111.177 1434.58 L116.061 1434.58 L116.061 1440.46 L111.177 1440.46 L111.177 1434.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M136.246 1408.98 Q132.635 1408.98 130.807 1412.54 Q129.001 1416.08 129.001 1423.21 Q129.001 1430.32 130.807 1433.89 Q132.635 1437.43 136.246 1437.43 Q139.881 1437.43 141.686 1433.89 Q143.515 1430.32 143.515 1423.21 Q143.515 1416.08 141.686 1412.54 Q139.881 1408.98 136.246 1408.98 M136.246 1405.27 Q142.056 1405.27 145.112 1409.88 Q148.191 1414.46 148.191 1423.21 Q148.191 1431.94 145.112 1436.55 Q142.056 1441.13 136.246 1441.13 Q130.436 1441.13 127.357 1436.55 Q124.302 1431.94 124.302 1423.21 Q124.302 1414.46 127.357 1409.88 Q130.436 1405.27 136.246 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M86.0382 1123.81 L102.358 1123.81 L102.358 1127.75 L80.4133 1127.75 L80.4133 1123.81 Q83.0753 1121.06 87.6586 1116.43 Q92.2651 1111.78 93.4456 1110.43 Q95.691 1107.91 96.5706 1106.17 Q97.4734 1104.41 97.4734 1102.72 Q97.4734 1099.97 95.5289 1098.23 Q93.6076 1096.5 90.5058 1096.5 Q88.3067 1096.5 85.8531 1097.26 Q83.4225 1098.03 80.6447 1099.58 L80.6447 1094.85 Q83.4688 1093.72 85.9225 1093.14 Q88.3762 1092.56 90.4132 1092.56 Q95.7836 1092.56 98.978 1095.25 Q102.172 1097.93 102.172 1102.42 Q102.172 1104.55 101.362 1106.47 Q100.575 1108.37 98.4687 1110.96 Q97.89 1111.64 94.7882 1114.85 Q91.6864 1118.05 86.0382 1123.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M112.172 1121.87 L117.057 1121.87 L117.057 1127.75 L112.172 1127.75 L112.172 1121.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M127.288 1093.19 L145.644 1093.19 L145.644 1097.12 L131.57 1097.12 L131.57 1105.59 Q132.589 1105.25 133.607 1105.09 Q134.626 1104.9 135.644 1104.9 Q141.431 1104.9 144.811 1108.07 Q148.191 1111.24 148.191 1116.66 Q148.191 1122.24 144.718 1125.34 Q141.246 1128.42 134.927 1128.42 Q132.751 1128.42 130.482 1128.05 Q128.237 1127.68 125.83 1126.94 L125.83 1122.24 Q127.913 1123.37 130.135 1123.93 Q132.357 1124.48 134.834 1124.48 Q138.839 1124.48 141.177 1122.38 Q143.515 1120.27 143.515 1116.66 Q143.515 1113.05 141.177 1110.94 Q138.839 1108.84 134.834 1108.84 Q132.959 1108.84 131.084 1109.25 Q129.232 1109.67 127.288 1110.55 L127.288 1093.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M81.0614 780.475 L99.4178 780.475 L99.4178 784.41 L85.3438 784.41 L85.3438 792.882 Q86.3623 792.535 87.3808 792.373 Q88.3993 792.187 89.4178 792.187 Q95.2049 792.187 98.5845 795.359 Q101.964 798.53 101.964 803.947 Q101.964 809.525 98.4919 812.627 Q95.0197 815.706 88.7003 815.706 Q86.5243 815.706 84.2558 815.335 Q82.0105 814.965 79.6031 814.224 L79.6031 809.525 Q81.6864 810.66 83.9086 811.215 Q86.1308 811.771 88.6077 811.771 Q92.6123 811.771 94.9502 809.664 Q97.2882 807.558 97.2882 803.947 Q97.2882 800.336 94.9502 798.229 Q92.6123 796.123 88.6077 796.123 Q86.7327 796.123 84.8577 796.539 Q83.0058 796.956 81.0614 797.836 L81.0614 780.475 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M111.177 809.155 L116.061 809.155 L116.061 815.035 L111.177 815.035 L111.177 809.155 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M136.246 783.553 Q132.635 783.553 130.807 787.118 Q129.001 790.66 129.001 797.789 Q129.001 804.896 130.807 808.46 Q132.635 812.002 136.246 812.002 Q139.881 812.002 141.686 808.46 Q143.515 804.896 143.515 797.789 Q143.515 790.66 141.686 787.118 Q139.881 783.553 136.246 783.553 M136.246 779.85 Q142.056 779.85 145.112 784.456 Q148.191 789.039 148.191 797.789 Q148.191 806.516 145.112 811.122 Q142.056 815.706 136.246 815.706 Q130.436 815.706 127.357 811.122 Q124.302 806.516 124.302 797.789 Q124.302 789.039 127.357 784.456 Q130.436 779.85 136.246 779.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M80.8299 467.762 L103.052 467.762 L103.052 469.753 L90.5058 502.322 L85.6216 502.322 L97.4271 471.697 L80.8299 471.697 L80.8299 467.762 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M112.172 496.442 L117.057 496.442 L117.057 502.322 L112.172 502.322 L112.172 496.442 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M127.288 467.762 L145.644 467.762 L145.644 471.697 L131.57 471.697 L131.57 480.169 Q132.589 479.822 133.607 479.66 Q134.626 479.475 135.644 479.475 Q141.431 479.475 144.811 482.646 Q148.191 485.817 148.191 491.234 Q148.191 496.813 144.718 499.914 Q141.246 502.993 134.927 502.993 Q132.751 502.993 130.482 502.623 Q128.237 502.252 125.83 501.512 L125.83 496.813 Q127.913 497.947 130.135 498.502 Q132.357 499.058 134.834 499.058 Q138.839 499.058 141.177 496.951 Q143.515 494.845 143.515 491.234 Q143.515 487.623 141.177 485.516 Q138.839 483.41 134.834 483.41 Q132.959 483.41 131.084 483.827 Q129.232 484.243 127.288 485.123 L127.288 467.762 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M51.6634 185.674 L59.3023 185.674 L59.3023 159.308 L50.9921 160.975 L50.9921 156.716 L59.256 155.049 L63.9319 155.049 L63.9319 185.674 L71.5707 185.674 L71.5707 189.609 L51.6634 189.609 L51.6634 185.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M91.0151 158.128 Q87.404 158.128 85.5753 161.693 Q83.7697 165.234 83.7697 172.364 Q83.7697 179.47 85.5753 183.035 Q87.404 186.577 91.0151 186.577 Q94.6493 186.577 96.4548 183.035 Q98.2835 179.47 98.2835 172.364 Q98.2835 165.234 96.4548 161.693 Q94.6493 158.128 91.0151 158.128 M91.0151 154.424 Q96.8252 154.424 99.8808 159.031 Q102.959 163.614 102.959 172.364 Q102.959 181.091 99.8808 185.697 Q96.8252 190.28 91.0151 190.28 Q85.2049 190.28 82.1262 185.697 Q79.0707 181.091 79.0707 172.364 Q79.0707 163.614 82.1262 159.031 Q85.2049 154.424 91.0151 154.424 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M111.177 183.73 L116.061 183.73 L116.061 189.609 L111.177 189.609 L111.177 183.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip830)" d="M136.246 158.128 Q132.635 158.128 130.807 161.693 Q129.001 165.234 129.001 172.364 Q129.001 179.47 130.807 183.035 Q132.635 186.577 136.246 186.577 Q139.881 186.577 141.686 183.035 Q143.515 179.47 143.515 172.364 Q143.515 165.234 141.686 161.693 Q139.881 158.128 136.246 158.128 M136.246 154.424 Q142.056 154.424 145.112 159.031 Q148.191 163.614 148.191 172.364 Q148.191 181.091 145.112 185.697 Q142.056 190.28 136.246 190.28 Q130.436 190.28 127.357 185.697 Q124.302 181.091 124.302 172.364 Q124.302 163.614 127.357 159.031 Q130.436 154.424 136.246 154.424 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip832)" d="M303.466 1173.01 L303.466 1423.18 L410.689 1423.18 L410.689 1173.01 L303.466 1173.01 L303.466 1173.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.466,1173.01 303.466,1423.18 410.689,1423.18 410.689,1173.01 303.466,1173.01 "/>
-<path clip-path="url(#clip832)" d="M410.689 547.584 L410.689 1423.18 L517.912 1423.18 L517.912 547.584 L410.689 547.584 L410.689 547.584  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="410.689,547.584 410.689,1423.18 517.912,1423.18 517.912,547.584 410.689,547.584 "/>
-<path clip-path="url(#clip832)" d="M517.912 1047.92 L517.912 1423.18 L625.135 1423.18 L625.135 1047.92 L517.912 1047.92 L517.912 1047.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="517.912,1047.92 517.912,1423.18 625.135,1423.18 625.135,1047.92 517.912,1047.92 "/>
-<path clip-path="url(#clip832)" d="M625.135 422.499 L625.135 1423.18 L732.358 1423.18 L732.358 422.499 L625.135 422.499 L625.135 422.499  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="625.135,422.499 625.135,1423.18 732.358,1423.18 732.358,422.499 625.135,422.499 "/>
-<path clip-path="url(#clip832)" d="M732.358 797.755 L732.358 1423.18 L839.581 1423.18 L839.581 797.755 L732.358 797.755 L732.358 797.755  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.358,797.755 732.358,1423.18 839.581,1423.18 839.581,797.755 732.358,797.755 "/>
-<path clip-path="url(#clip832)" d="M839.581 47.2441 L839.581 1423.18 L946.804 1423.18 L946.804 47.2441 L839.581 47.2441 L839.581 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="839.581,47.2441 839.581,1423.18 946.804,1423.18 946.804,47.2441 839.581,47.2441 "/>
-<path clip-path="url(#clip832)" d="M946.804 47.2441 L946.804 1423.18 L1054.03 1423.18 L1054.03 47.2441 L946.804 47.2441 L946.804 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="946.804,47.2441 946.804,1423.18 1054.03,1423.18 1054.03,47.2441 946.804,47.2441 "/>
-<path clip-path="url(#clip832)" d="M1054.03 422.499 L1054.03 1423.18 L1161.25 1423.18 L1161.25 422.499 L1054.03 422.499 L1054.03 422.499  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1054.03,422.499 1054.03,1423.18 1161.25,1423.18 1161.25,422.499 1054.03,422.499 "/>
-<path clip-path="url(#clip832)" d="M1161.25 172.329 L1161.25 1423.18 L1268.47 1423.18 L1268.47 172.329 L1161.25 172.329 L1161.25 172.329  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1161.25,172.329 1161.25,1423.18 1268.47,1423.18 1268.47,172.329 1161.25,172.329 "/>
-<path clip-path="url(#clip832)" d="M1268.47 422.499 L1268.47 1423.18 L1375.7 1423.18 L1375.7 422.499 L1268.47 422.499 L1268.47 422.499  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1268.47,422.499 1268.47,1423.18 1375.7,1423.18 1375.7,422.499 1268.47,422.499 "/>
-<path clip-path="url(#clip832)" d="M1375.7 297.414 L1375.7 1423.18 L1482.92 1423.18 L1482.92 297.414 L1375.7 297.414 L1375.7 297.414  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1375.7,297.414 1375.7,1423.18 1482.92,1423.18 1482.92,297.414 1375.7,297.414 "/>
-<path clip-path="url(#clip832)" d="M1482.92 172.329 L1482.92 1423.18 L1590.14 1423.18 L1590.14 172.329 L1482.92 172.329 L1482.92 172.329  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1482.92,172.329 1482.92,1423.18 1590.14,1423.18 1590.14,172.329 1482.92,172.329 "/>
-<path clip-path="url(#clip832)" d="M1590.14 922.84 L1590.14 1423.18 L1697.37 1423.18 L1697.37 922.84 L1590.14 922.84 L1590.14 922.84  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1590.14,922.84 1590.14,1423.18 1697.37,1423.18 1697.37,922.84 1590.14,922.84 "/>
-<path clip-path="url(#clip832)" d="M1697.37 1298.09 L1697.37 1423.18 L1804.59 1423.18 L1804.59 1298.09 L1697.37 1298.09 L1697.37 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1697.37,1298.09 1697.37,1423.18 1804.59,1423.18 1804.59,1298.09 1697.37,1298.09 "/>
-<path clip-path="url(#clip832)" d="M1804.59 1298.09 L1804.59 1423.18 L1911.81 1423.18 L1911.81 1298.09 L1804.59 1298.09 L1804.59 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1804.59,1298.09 1804.59,1423.18 1911.81,1423.18 1911.81,1298.09 1804.59,1298.09 "/>
-<path clip-path="url(#clip832)" d="M1911.81 1423.18 L1911.81 1423.18 L2019.03 1423.18 L2019.03 1423.18 L1911.81 1423.18 L1911.81 1423.18  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1911.81,1423.18 1911.81,1423.18 2019.03,1423.18 1911.81,1423.18 "/>
-<path clip-path="url(#clip832)" d="M2019.03 1298.09 L2019.03 1423.18 L2126.26 1423.18 L2126.26 1298.09 L2019.03 1298.09 L2019.03 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2019.03,1298.09 2019.03,1423.18 2126.26,1423.18 2126.26,1298.09 2019.03,1298.09 "/>
-<path clip-path="url(#clip832)" d="M2126.26 1298.09 L2126.26 1423.18 L2233.48 1423.18 L2233.48 1298.09 L2126.26 1298.09 L2126.26 1298.09  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip832)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2126.26,1298.09 2126.26,1423.18 2233.48,1423.18 2233.48,1298.09 2126.26,1298.09 "/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="357.077" cy="1173.01" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="464.3" cy="547.584" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="571.523" cy="1047.92" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="678.746" cy="422.499" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="785.969" cy="797.755" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="893.193" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1000.42" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1107.64" cy="422.499" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1214.86" cy="172.329" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1322.08" cy="422.499" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1429.31" cy="297.414" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1536.53" cy="172.329" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1643.75" cy="922.84" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1750.98" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1858.2" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1965.42" cy="1423.18" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2072.65" cy="1298.09" r="2"/>
-<circle clip-path="url(#clip832)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2179.87" cy="1298.09" r="2"/>
-</svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/a4f8868e.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/a4f8868e.svg
new file mode 100644
index 00000000000..8fa2d7ccd01
--- /dev/null
+++ b/previews/PR3562/tutorials/applications/two_stage_stochastic/a4f8868e.svg
@@ -0,0 +1,94 @@
+<?xml version="1.0" encoding="utf-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
+<defs>
+  <clipPath id="clip650">
+    <rect x="0" y="0" width="2400" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip650)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip651">
+    <rect x="480" y="0" width="1681" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip650)" d="M138.959 1423.18 L2352.76 1423.18 L2352.76 47.2441 L138.959 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip652">
+    <rect x="138" y="47" width="2215" height="1377"/>
+  </clipPath>
+</defs>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="260.722,1423.18 260.722,47.2441 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="698.56,1423.18 698.56,47.2441 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1136.4,1423.18 1136.4,47.2441 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1574.24,1423.18 1574.24,47.2441 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2012.07,1423.18 2012.07,47.2441 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="260.722,1423.18 260.722,1404.28 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="698.56,1423.18 698.56,1404.28 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1136.4,1423.18 1136.4,1404.28 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1574.24,1423.18 1574.24,1404.28 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2012.07,1423.18 2012.07,1404.28 "/>
+<path clip-path="url(#clip650)" d="M220.329 1481.64 L227.968 1481.64 L227.968 1455.28 L219.658 1456.95 L219.658 1452.69 L227.921 1451.02 L232.597 1451.02 L232.597 1481.64 L240.236 1481.64 L240.236 1485.58 L220.329 1485.58 L220.329 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M260.259 1466.44 Q257.111 1466.44 255.259 1468.59 Q253.431 1470.74 253.431 1474.49 Q253.431 1478.22 255.259 1480.39 Q257.111 1482.55 260.259 1482.55 Q263.407 1482.55 265.236 1480.39 Q267.088 1478.22 267.088 1474.49 Q267.088 1470.74 265.236 1468.59 Q263.407 1466.44 260.259 1466.44 M269.542 1451.78 L269.542 1456.04 Q267.782 1455.21 265.977 1454.77 Q264.194 1454.33 262.435 1454.33 Q257.806 1454.33 255.352 1457.45 Q252.921 1460.58 252.574 1466.9 Q253.94 1464.89 256 1463.82 Q258.06 1462.73 260.537 1462.73 Q265.745 1462.73 268.755 1465.9 Q271.787 1469.05 271.787 1474.49 Q271.787 1479.82 268.639 1483.03 Q265.491 1486.25 260.259 1486.25 Q254.264 1486.25 251.093 1481.67 Q247.921 1477.06 247.921 1468.33 Q247.921 1460.14 251.81 1455.28 Q255.699 1450.39 262.25 1450.39 Q264.009 1450.39 265.792 1450.74 Q267.597 1451.09 269.542 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M289.842 1454.1 Q286.231 1454.1 284.403 1457.66 Q282.597 1461.2 282.597 1468.33 Q282.597 1475.44 284.403 1479.01 Q286.231 1482.55 289.842 1482.55 Q293.477 1482.55 295.282 1479.01 Q297.111 1475.44 297.111 1468.33 Q297.111 1461.2 295.282 1457.66 Q293.477 1454.1 289.842 1454.1 M289.842 1450.39 Q295.653 1450.39 298.708 1455 Q301.787 1459.58 301.787 1468.33 Q301.787 1477.06 298.708 1481.67 Q295.653 1486.25 289.842 1486.25 Q284.032 1486.25 280.954 1481.67 Q277.898 1477.06 277.898 1468.33 Q277.898 1459.58 280.954 1455 Q284.032 1450.39 289.842 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M658.167 1481.64 L665.806 1481.64 L665.806 1455.28 L657.496 1456.95 L657.496 1452.69 L665.759 1451.02 L670.435 1451.02 L670.435 1481.64 L678.074 1481.64 L678.074 1485.58 L658.167 1485.58 L658.167 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M697.519 1469.17 Q694.185 1469.17 692.264 1470.95 Q690.366 1472.73 690.366 1475.86 Q690.366 1478.98 692.264 1480.77 Q694.185 1482.55 697.519 1482.55 Q700.852 1482.55 702.773 1480.77 Q704.694 1478.96 704.694 1475.86 Q704.694 1472.73 702.773 1470.95 Q700.875 1469.17 697.519 1469.17 M692.843 1467.18 Q689.833 1466.44 688.144 1464.38 Q686.477 1462.32 686.477 1459.35 Q686.477 1455.21 689.417 1452.8 Q692.38 1450.39 697.519 1450.39 Q702.681 1450.39 705.62 1452.8 Q708.56 1455.21 708.56 1459.35 Q708.56 1462.32 706.87 1464.38 Q705.204 1466.44 702.218 1467.18 Q705.597 1467.96 707.472 1470.26 Q709.37 1472.55 709.37 1475.86 Q709.37 1480.88 706.292 1483.57 Q703.236 1486.25 697.519 1486.25 Q691.801 1486.25 688.722 1483.57 Q685.667 1480.88 685.667 1475.86 Q685.667 1472.55 687.565 1470.26 Q689.463 1467.96 692.843 1467.18 M691.13 1459.79 Q691.13 1462.48 692.796 1463.98 Q694.486 1465.49 697.519 1465.49 Q700.528 1465.49 702.218 1463.98 Q703.931 1462.48 703.931 1459.79 Q703.931 1457.11 702.218 1455.6 Q700.528 1454.1 697.519 1454.1 Q694.486 1454.1 692.796 1455.6 Q691.13 1457.11 691.13 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M727.68 1454.1 Q724.069 1454.1 722.241 1457.66 Q720.435 1461.2 720.435 1468.33 Q720.435 1475.44 722.241 1479.01 Q724.069 1482.55 727.68 1482.55 Q731.315 1482.55 733.12 1479.01 Q734.949 1475.44 734.949 1468.33 Q734.949 1461.2 733.12 1457.66 Q731.315 1454.1 727.68 1454.1 M727.68 1450.39 Q733.491 1450.39 736.546 1455 Q739.625 1459.58 739.625 1468.33 Q739.625 1477.06 736.546 1481.67 Q733.491 1486.25 727.68 1486.25 Q721.87 1486.25 718.792 1481.67 Q715.736 1477.06 715.736 1468.33 Q715.736 1459.58 718.792 1455 Q721.87 1450.39 727.68 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1100.09 1481.64 L1116.41 1481.64 L1116.41 1485.58 L1094.47 1485.58 L1094.47 1481.64 Q1097.13 1478.89 1101.71 1474.26 Q1106.32 1469.61 1107.5 1468.27 Q1109.74 1465.74 1110.62 1464.01 Q1111.53 1462.25 1111.53 1460.56 Q1111.53 1457.8 1109.58 1456.07 Q1107.66 1454.33 1104.56 1454.33 Q1102.36 1454.33 1099.91 1455.09 Q1097.47 1455.86 1094.7 1457.41 L1094.7 1452.69 Q1097.52 1451.55 1099.97 1450.97 Q1102.43 1450.39 1104.47 1450.39 Q1109.84 1450.39 1113.03 1453.08 Q1116.22 1455.77 1116.22 1460.26 Q1116.22 1462.39 1115.41 1464.31 Q1114.63 1466.2 1112.52 1468.8 Q1111.94 1469.47 1108.84 1472.69 Q1105.74 1475.88 1100.09 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1136.22 1454.1 Q1132.61 1454.1 1130.78 1457.66 Q1128.98 1461.2 1128.98 1468.33 Q1128.98 1475.44 1130.78 1479.01 Q1132.61 1482.55 1136.22 1482.55 Q1139.86 1482.55 1141.66 1479.01 Q1143.49 1475.44 1143.49 1468.33 Q1143.49 1461.2 1141.66 1457.66 Q1139.86 1454.1 1136.22 1454.1 M1136.22 1450.39 Q1142.03 1450.39 1145.09 1455 Q1148.17 1459.58 1148.17 1468.33 Q1148.17 1477.06 1145.09 1481.67 Q1142.03 1486.25 1136.22 1486.25 Q1130.41 1486.25 1127.34 1481.67 Q1124.28 1477.06 1124.28 1468.33 Q1124.28 1459.58 1127.34 1455 Q1130.41 1450.39 1136.22 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1166.39 1454.1 Q1162.78 1454.1 1160.95 1457.66 Q1159.14 1461.2 1159.14 1468.33 Q1159.14 1475.44 1160.95 1479.01 Q1162.78 1482.55 1166.39 1482.55 Q1170.02 1482.55 1171.83 1479.01 Q1173.65 1475.44 1173.65 1468.33 Q1173.65 1461.2 1171.83 1457.66 Q1170.02 1454.1 1166.39 1454.1 M1166.39 1450.39 Q1172.2 1450.39 1175.25 1455 Q1178.33 1459.58 1178.33 1468.33 Q1178.33 1477.06 1175.25 1481.67 Q1172.2 1486.25 1166.39 1486.25 Q1160.58 1486.25 1157.5 1481.67 Q1154.44 1477.06 1154.44 1468.33 Q1154.44 1459.58 1157.5 1455 Q1160.58 1450.39 1166.39 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1537.93 1481.64 L1554.25 1481.64 L1554.25 1485.58 L1532.3 1485.58 L1532.3 1481.64 Q1534.97 1478.89 1539.55 1474.26 Q1544.16 1469.61 1545.34 1468.27 Q1547.58 1465.74 1548.46 1464.01 Q1549.36 1462.25 1549.36 1460.56 Q1549.36 1457.8 1547.42 1456.07 Q1545.5 1454.33 1542.4 1454.33 Q1540.2 1454.33 1537.74 1455.09 Q1535.31 1455.86 1532.54 1457.41 L1532.54 1452.69 Q1535.36 1451.55 1537.81 1450.97 Q1540.27 1450.39 1542.3 1450.39 Q1547.67 1450.39 1550.87 1453.08 Q1554.06 1455.77 1554.06 1460.26 Q1554.06 1462.39 1553.25 1464.31 Q1552.47 1466.2 1550.36 1468.8 Q1549.78 1469.47 1546.68 1472.69 Q1543.58 1475.88 1537.93 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1568.09 1481.64 L1584.41 1481.64 L1584.41 1485.58 L1562.47 1485.58 L1562.47 1481.64 Q1565.13 1478.89 1569.71 1474.26 Q1574.32 1469.61 1575.5 1468.27 Q1577.74 1465.74 1578.62 1464.01 Q1579.53 1462.25 1579.53 1460.56 Q1579.53 1457.8 1577.58 1456.07 Q1575.66 1454.33 1572.56 1454.33 Q1570.36 1454.33 1567.91 1455.09 Q1565.47 1455.86 1562.7 1457.41 L1562.7 1452.69 Q1565.52 1451.55 1567.97 1450.97 Q1570.43 1450.39 1572.47 1450.39 Q1577.84 1450.39 1581.03 1453.08 Q1584.22 1455.77 1584.22 1460.26 Q1584.22 1462.39 1583.41 1464.31 Q1582.63 1466.2 1580.52 1468.8 Q1579.94 1469.47 1576.84 1472.69 Q1573.74 1475.88 1568.09 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1604.22 1454.1 Q1600.61 1454.1 1598.78 1457.66 Q1596.98 1461.2 1596.98 1468.33 Q1596.98 1475.44 1598.78 1479.01 Q1600.61 1482.55 1604.22 1482.55 Q1607.86 1482.55 1609.66 1479.01 Q1611.49 1475.44 1611.49 1468.33 Q1611.49 1461.2 1609.66 1457.66 Q1607.86 1454.1 1604.22 1454.1 M1604.22 1450.39 Q1610.03 1450.39 1613.09 1455 Q1616.17 1459.58 1616.17 1468.33 Q1616.17 1477.06 1613.09 1481.67 Q1610.03 1486.25 1604.22 1486.25 Q1598.41 1486.25 1595.34 1481.67 Q1592.28 1477.06 1592.28 1468.33 Q1592.28 1459.58 1595.34 1455 Q1598.41 1450.39 1604.22 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1975.77 1481.64 L1992.09 1481.64 L1992.09 1485.58 L1970.14 1485.58 L1970.14 1481.64 Q1972.8 1478.89 1977.39 1474.26 Q1981.99 1469.61 1983.17 1468.27 Q1985.42 1465.74 1986.3 1464.01 Q1987.2 1462.25 1987.2 1460.56 Q1987.2 1457.8 1985.26 1456.07 Q1983.34 1454.33 1980.23 1454.33 Q1978.03 1454.33 1975.58 1455.09 Q1973.15 1455.86 1970.37 1457.41 L1970.37 1452.69 Q1973.2 1451.55 1975.65 1450.97 Q1978.1 1450.39 1980.14 1450.39 Q1985.51 1450.39 1988.71 1453.08 Q1991.9 1455.77 1991.9 1460.26 Q1991.9 1462.39 1991.09 1464.31 Q1990.3 1466.2 1988.2 1468.8 Q1987.62 1469.47 1984.52 1472.69 Q1981.41 1475.88 1975.77 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2014.75 1455.09 L2002.94 1473.54 L2014.75 1473.54 L2014.75 1455.09 M2013.52 1451.02 L2019.4 1451.02 L2019.4 1473.54 L2024.33 1473.54 L2024.33 1477.43 L2019.4 1477.43 L2019.4 1485.58 L2014.75 1485.58 L2014.75 1477.43 L1999.15 1477.43 L1999.15 1472.92 L2013.52 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M2042.06 1454.1 Q2038.45 1454.1 2036.62 1457.66 Q2034.82 1461.2 2034.82 1468.33 Q2034.82 1475.44 2036.62 1479.01 Q2038.45 1482.55 2042.06 1482.55 Q2045.7 1482.55 2047.5 1479.01 Q2049.33 1475.44 2049.33 1468.33 Q2049.33 1461.2 2047.5 1457.66 Q2045.7 1454.1 2042.06 1454.1 M2042.06 1450.39 Q2047.87 1450.39 2050.93 1455 Q2054.01 1459.58 2054.01 1468.33 Q2054.01 1477.06 2050.93 1481.67 Q2047.87 1486.25 2042.06 1486.25 Q2036.25 1486.25 2033.17 1481.67 Q2030.12 1477.06 2030.12 1468.33 Q2030.12 1459.58 2033.17 1455 Q2036.25 1450.39 2042.06 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1120.23 1525.81 L1120.23 1562.76 L1128 1562.76 Q1137.83 1562.76 1142.38 1558.3 Q1146.97 1553.85 1146.97 1544.24 Q1146.97 1534.69 1142.38 1530.26 Q1137.83 1525.81 1128 1525.81 L1120.23 1525.81 M1113.8 1520.52 L1127.01 1520.52 Q1140.82 1520.52 1147.28 1526.28 Q1153.75 1532.01 1153.75 1544.24 Q1153.75 1556.52 1147.25 1562.28 Q1140.76 1568.04 1127.01 1568.04 L1113.8 1568.04 L1113.8 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1194.23 1548.76 L1194.23 1551.62 L1167.3 1551.62 Q1167.69 1557.67 1170.93 1560.85 Q1174.21 1564 1180.04 1564 Q1183.41 1564 1186.56 1563.17 Q1189.74 1562.35 1192.86 1560.69 L1192.86 1566.23 Q1189.71 1567.57 1186.4 1568.27 Q1183.09 1568.97 1179.69 1568.97 Q1171.16 1568.97 1166.16 1564 Q1161.19 1559.04 1161.19 1550.57 Q1161.19 1541.82 1165.9 1536.69 Q1170.65 1531.54 1178.67 1531.54 Q1185.86 1531.54 1190.03 1536.18 Q1194.23 1540.8 1194.23 1548.76 M1188.38 1547.04 Q1188.31 1542.23 1185.67 1539.37 Q1183.06 1536.5 1178.73 1536.5 Q1173.83 1536.5 1170.87 1539.27 Q1167.94 1542.04 1167.5 1547.07 L1188.38 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1231.6 1539.24 Q1233.79 1535.29 1236.85 1533.41 Q1239.91 1531.54 1244.04 1531.54 Q1249.61 1531.54 1252.64 1535.45 Q1255.66 1539.33 1255.66 1546.53 L1255.66 1568.04 L1249.77 1568.04 L1249.77 1546.72 Q1249.77 1541.59 1247.96 1539.11 Q1246.14 1536.63 1242.42 1536.63 Q1237.87 1536.63 1235.23 1539.65 Q1232.59 1542.68 1232.59 1547.9 L1232.59 1568.04 L1226.7 1568.04 L1226.7 1546.72 Q1226.7 1541.56 1224.88 1539.11 Q1223.07 1536.63 1219.28 1536.63 Q1214.79 1536.63 1212.15 1539.68 Q1209.51 1542.71 1209.51 1547.9 L1209.51 1568.04 L1203.62 1568.04 L1203.62 1532.4 L1209.51 1532.4 L1209.51 1537.93 Q1211.51 1534.66 1214.32 1533.1 Q1217.12 1531.54 1220.97 1531.54 Q1224.85 1531.54 1227.56 1533.51 Q1230.29 1535.48 1231.6 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1283.54 1550.12 Q1276.44 1550.12 1273.71 1551.75 Q1270.97 1553.37 1270.97 1557.29 Q1270.97 1560.4 1273.01 1562.25 Q1275.08 1564.07 1278.61 1564.07 Q1283.48 1564.07 1286.41 1560.63 Q1289.37 1557.16 1289.37 1551.43 L1289.37 1550.12 L1283.54 1550.12 M1295.22 1547.71 L1295.22 1568.04 L1289.37 1568.04 L1289.37 1562.63 Q1287.36 1565.88 1284.37 1567.44 Q1281.38 1568.97 1277.05 1568.97 Q1271.58 1568.97 1268.33 1565.91 Q1265.11 1562.82 1265.11 1557.67 Q1265.11 1551.65 1269.12 1548.6 Q1273.17 1545.54 1281.16 1545.54 L1289.37 1545.54 L1289.37 1544.97 Q1289.37 1540.93 1286.69 1538.73 Q1284.05 1536.5 1279.25 1536.5 Q1276.19 1536.5 1273.29 1537.23 Q1270.4 1537.97 1267.72 1539.43 L1267.72 1534.02 Q1270.94 1532.78 1273.96 1532.17 Q1276.99 1531.54 1279.85 1531.54 Q1287.58 1531.54 1291.4 1535.55 Q1295.22 1539.56 1295.22 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1336.92 1546.53 L1336.92 1568.04 L1331.06 1568.04 L1331.06 1546.72 Q1331.06 1541.66 1329.09 1539.14 Q1327.12 1536.63 1323.17 1536.63 Q1318.43 1536.63 1315.69 1539.65 Q1312.95 1542.68 1312.95 1547.9 L1312.95 1568.04 L1307.06 1568.04 L1307.06 1532.4 L1312.95 1532.4 L1312.95 1537.93 Q1315.05 1534.72 1317.89 1533.13 Q1320.75 1531.54 1324.47 1531.54 Q1330.62 1531.54 1333.77 1535.36 Q1336.92 1539.14 1336.92 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M1372.06 1537.81 L1372.06 1518.52 L1377.91 1518.52 L1377.91 1568.04 L1372.06 1568.04 L1372.06 1562.7 Q1370.21 1565.88 1367.38 1567.44 Q1364.58 1568.97 1360.63 1568.97 Q1354.17 1568.97 1350.1 1563.81 Q1346.05 1558.65 1346.05 1550.25 Q1346.05 1541.85 1350.1 1536.69 Q1354.17 1531.54 1360.63 1531.54 Q1364.58 1531.54 1367.38 1533.1 Q1370.21 1534.62 1372.06 1537.81 M1352.1 1550.25 Q1352.1 1556.71 1354.74 1560.4 Q1357.42 1564.07 1362.06 1564.07 Q1366.71 1564.07 1369.38 1560.4 Q1372.06 1556.71 1372.06 1550.25 Q1372.06 1543.79 1369.38 1540.13 Q1366.71 1536.44 1362.06 1536.44 Q1357.42 1536.44 1354.74 1540.13 Q1352.1 1543.79 1352.1 1550.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="138.959,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="138.959,964.535 2352.76,964.535 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="138.959,505.889 2352.76,505.889 "/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="138.959,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,1423.18 138.959,47.2441 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,1423.18 157.857,1423.18 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,964.535 157.857,964.535 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,505.889 157.857,505.889 "/>
+<polyline clip-path="url(#clip650)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="138.959,47.2441 157.857,47.2441 "/>
+<path clip-path="url(#clip650)" d="M91.0151 1408.98 Q87.404 1408.98 85.5753 1412.54 Q83.7697 1416.08 83.7697 1423.21 Q83.7697 1430.32 85.5753 1433.89 Q87.404 1437.43 91.0151 1437.43 Q94.6493 1437.43 96.4548 1433.89 Q98.2835 1430.32 98.2835 1423.21 Q98.2835 1416.08 96.4548 1412.54 Q94.6493 1408.98 91.0151 1408.98 M91.0151 1405.27 Q96.8252 1405.27 99.8808 1409.88 Q102.959 1414.46 102.959 1423.21 Q102.959 1431.94 99.8808 1436.55 Q96.8252 1441.13 91.0151 1441.13 Q85.2049 1441.13 82.1262 1436.55 Q79.0707 1431.94 79.0707 1423.21 Q79.0707 1414.46 82.1262 1409.88 Q85.2049 1405.27 91.0151 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M82.0568 947.255 L100.413 947.255 L100.413 951.19 L86.3392 951.19 L86.3392 959.662 Q87.3577 959.315 88.3762 959.153 Q89.3947 958.967 90.4132 958.967 Q96.2002 958.967 99.5798 962.139 Q102.959 965.31 102.959 970.727 Q102.959 976.305 99.4872 979.407 Q96.015 982.486 89.6956 982.486 Q87.5197 982.486 85.2512 982.116 Q83.0058 981.745 80.5984 981.004 L80.5984 976.305 Q82.6818 977.44 84.904 977.995 Q87.1262 978.551 89.603 978.551 Q93.6076 978.551 95.9456 976.444 Q98.2835 974.338 98.2835 970.727 Q98.2835 967.116 95.9456 965.009 Q93.6076 962.903 89.603 962.903 Q87.728 962.903 85.8531 963.319 Q84.0012 963.736 82.0568 964.616 L82.0568 947.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M51.6634 519.234 L59.3023 519.234 L59.3023 492.869 L50.9921 494.535 L50.9921 490.276 L59.256 488.609 L63.9319 488.609 L63.9319 519.234 L71.5707 519.234 L71.5707 523.169 L51.6634 523.169 L51.6634 519.234 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M91.0151 491.688 Q87.404 491.688 85.5753 495.253 Q83.7697 498.794 83.7697 505.924 Q83.7697 513.031 85.5753 516.595 Q87.404 520.137 91.0151 520.137 Q94.6493 520.137 96.4548 516.595 Q98.2835 513.031 98.2835 505.924 Q98.2835 498.794 96.4548 495.253 Q94.6493 491.688 91.0151 491.688 M91.0151 487.984 Q96.8252 487.984 99.8808 492.591 Q102.959 497.174 102.959 505.924 Q102.959 514.651 99.8808 519.257 Q96.8252 523.841 91.0151 523.841 Q85.2049 523.841 82.1262 519.257 Q79.0707 514.651 79.0707 505.924 Q79.0707 497.174 82.1262 492.591 Q85.2049 487.984 91.0151 487.984 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M52.6588 60.5889 L60.2976 60.5889 L60.2976 34.2233 L51.9875 35.89 L51.9875 31.6308 L60.2513 29.9641 L64.9272 29.9641 L64.9272 60.5889 L72.5661 60.5889 L72.5661 64.5241 L52.6588 64.5241 L52.6588 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip650)" d="M82.0568 29.9641 L100.413 29.9641 L100.413 33.8993 L86.3392 33.8993 L86.3392 42.3714 Q87.3577 42.0242 88.3762 41.8622 Q89.3947 41.677 90.4132 41.677 Q96.2002 41.677 99.5798 44.8483 Q102.959 48.0196 102.959 53.4362 Q102.959 59.0149 99.4872 62.1167 Q96.015 65.1954 89.6956 65.1954 Q87.5197 65.1954 85.2512 64.825 Q83.0058 64.4547 80.5984 63.7139 L80.5984 59.0149 Q82.6818 60.1491 84.904 60.7047 Q87.1262 61.2602 89.603 61.2602 Q93.6076 61.2602 95.9456 59.1538 Q98.2835 57.0473 98.2835 53.4362 Q98.2835 49.8251 95.9456 47.7186 Q93.6076 45.6122 89.603 45.6122 Q87.728 45.6122 85.8531 46.0288 Q84.0012 46.4455 82.0568 47.3251 L82.0568 29.9641 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip652)" d="M260.722 1147.99 L260.722 1423.18 L370.182 1423.18 L370.182 1147.99 L260.722 1147.99 L260.722 1147.99  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="260.722,1147.99 260.722,1423.18 370.182,1423.18 370.182,1147.99 260.722,1147.99 "/>
+<path clip-path="url(#clip652)" d="M370.182 1147.99 L370.182 1423.18 L479.641 1423.18 L479.641 1147.99 L370.182 1147.99 L370.182 1147.99  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="370.182,1147.99 370.182,1423.18 479.641,1423.18 479.641,1147.99 370.182,1147.99 "/>
+<path clip-path="url(#clip652)" d="M479.641 964.535 L479.641 1423.18 L589.101 1423.18 L589.101 964.535 L479.641 964.535 L479.641 964.535  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="479.641,964.535 479.641,1423.18 589.101,1423.18 589.101,964.535 479.641,964.535 "/>
+<path clip-path="url(#clip652)" d="M589.101 1239.72 L589.101 1423.18 L698.56 1423.18 L698.56 1239.72 L589.101 1239.72 L589.101 1239.72  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="589.101,1239.72 589.101,1423.18 698.56,1423.18 698.56,1239.72 589.101,1239.72 "/>
+<path clip-path="url(#clip652)" d="M698.56 597.618 L698.56 1423.18 L808.02 1423.18 L808.02 597.618 L698.56 597.618 L698.56 597.618  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="698.56,597.618 698.56,1423.18 808.02,1423.18 808.02,597.618 698.56,597.618 "/>
+<path clip-path="url(#clip652)" d="M808.02 781.076 L808.02 1423.18 L917.479 1423.18 L917.479 781.076 L808.02 781.076 L808.02 781.076  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="808.02,781.076 808.02,1423.18 917.479,1423.18 917.479,781.076 808.02,781.076 "/>
+<path clip-path="url(#clip652)" d="M917.479 138.973 L917.479 1423.18 L1026.94 1423.18 L1026.94 138.973 L917.479 138.973 L917.479 138.973  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="917.479,138.973 917.479,1423.18 1026.94,1423.18 1026.94,138.973 917.479,138.973 "/>
+<path clip-path="url(#clip652)" d="M1026.94 964.535 L1026.94 1423.18 L1136.4 1423.18 L1136.4 964.535 L1026.94 964.535 L1026.94 964.535  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1026.94,964.535 1026.94,1423.18 1136.4,1423.18 1136.4,964.535 1026.94,964.535 "/>
+<path clip-path="url(#clip652)" d="M1136.4 322.431 L1136.4 1423.18 L1245.86 1423.18 L1245.86 322.431 L1136.4 322.431 L1136.4 322.431  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1136.4,322.431 1136.4,1423.18 1245.86,1423.18 1245.86,322.431 1136.4,322.431 "/>
+<path clip-path="url(#clip652)" d="M1245.86 47.2441 L1245.86 1423.18 L1355.32 1423.18 L1355.32 47.2441 L1245.86 47.2441 L1245.86 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1245.86,47.2441 1245.86,1423.18 1355.32,1423.18 1355.32,47.2441 1245.86,47.2441 "/>
+<path clip-path="url(#clip652)" d="M1355.32 322.431 L1355.32 1423.18 L1464.78 1423.18 L1464.78 322.431 L1355.32 322.431 L1355.32 322.431  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1355.32,322.431 1355.32,1423.18 1464.78,1423.18 1464.78,322.431 1355.32,322.431 "/>
+<path clip-path="url(#clip652)" d="M1464.78 1331.45 L1464.78 1423.18 L1574.24 1423.18 L1574.24 1331.45 L1464.78 1331.45 L1464.78 1331.45  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1464.78,1331.45 1464.78,1423.18 1574.24,1423.18 1574.24,1331.45 1464.78,1331.45 "/>
+<path clip-path="url(#clip652)" d="M1574.24 1239.72 L1574.24 1423.18 L1683.7 1423.18 L1683.7 1239.72 L1574.24 1239.72 L1574.24 1239.72  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1574.24,1239.72 1574.24,1423.18 1683.7,1423.18 1683.7,1239.72 1574.24,1239.72 "/>
+<path clip-path="url(#clip652)" d="M1683.7 964.535 L1683.7 1423.18 L1793.16 1423.18 L1793.16 964.535 L1683.7 964.535 L1683.7 964.535  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1683.7,964.535 1683.7,1423.18 1793.16,1423.18 1793.16,964.535 1683.7,964.535 "/>
+<path clip-path="url(#clip652)" d="M1793.16 1423.18 L1793.16 1423.18 L1902.61 1423.18 L1902.61 1423.18 L1793.16 1423.18 L1793.16 1423.18  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1793.16,1423.18 1793.16,1423.18 1902.61,1423.18 1793.16,1423.18 "/>
+<path clip-path="url(#clip652)" d="M1902.61 1331.45 L1902.61 1423.18 L2012.07 1423.18 L2012.07 1331.45 L1902.61 1331.45 L1902.61 1331.45  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1902.61,1331.45 1902.61,1423.18 2012.07,1423.18 2012.07,1331.45 1902.61,1331.45 "/>
+<path clip-path="url(#clip652)" d="M2012.07 1147.99 L2012.07 1423.18 L2121.53 1423.18 L2121.53 1147.99 L2012.07 1147.99 L2012.07 1147.99  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2012.07,1147.99 2012.07,1423.18 2121.53,1423.18 2121.53,1147.99 2012.07,1147.99 "/>
+<path clip-path="url(#clip652)" d="M2121.53 1331.45 L2121.53 1423.18 L2230.99 1423.18 L2230.99 1331.45 L2121.53 1331.45 L2121.53 1331.45  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip652)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2121.53,1331.45 2121.53,1423.18 2230.99,1423.18 2230.99,1331.45 2121.53,1331.45 "/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="315.452" cy="1147.99" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="424.911" cy="1147.99" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="534.371" cy="964.535" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="643.83" cy="1239.72" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="753.29" cy="597.618" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="862.749" cy="781.076" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="972.209" cy="138.973" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1081.67" cy="964.535" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1191.13" cy="322.431" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1300.59" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1410.05" cy="322.431" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1519.51" cy="1331.45" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1628.97" cy="1239.72" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1738.43" cy="964.535" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1847.88" cy="1423.18" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1957.34" cy="1331.45" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2066.8" cy="1147.99" r="2"/>
+<circle clip-path="url(#clip652)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2176.26" cy="1331.45" r="2"/>
+</svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/bea383e2.svg b/previews/PR3562/tutorials/applications/two_stage_stochastic/bea383e2.svg
new file mode 100644
index 00000000000..390a9e54383
--- /dev/null
+++ b/previews/PR3562/tutorials/applications/two_stage_stochastic/bea383e2.svg
@@ -0,0 +1,76 @@
+<?xml version="1.0" encoding="utf-8"?>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
+<defs>
+  <clipPath id="clip710">
+    <rect x="0" y="0" width="2400" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip710)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip711">
+    <rect x="480" y="0" width="1681" height="1600"/>
+  </clipPath>
+</defs>
+<path clip-path="url(#clip710)" d="M205.121 1423.18 L2352.76 1423.18 L2352.76 47.2441 L205.121 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<defs>
+  <clipPath id="clip712">
+    <rect x="205" y="47" width="2149" height="1377"/>
+  </clipPath>
+</defs>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="514.383,1423.18 514.383,47.2441 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="896.661,1423.18 896.661,47.2441 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1278.94,1423.18 1278.94,47.2441 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1661.22,1423.18 1661.22,47.2441 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2043.49,1423.18 2043.49,47.2441 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="514.383,1423.18 514.383,1404.28 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="896.661,1423.18 896.661,1404.28 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1278.94,1423.18 1278.94,1404.28 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1661.22,1423.18 1661.22,1404.28 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2043.49,1423.18 2043.49,1404.28 "/>
+<path clip-path="url(#clip710)" d="M487.474 1455.09 L475.668 1473.54 L487.474 1473.54 L487.474 1455.09 M486.247 1451.02 L492.127 1451.02 L492.127 1473.54 L497.057 1473.54 L497.057 1477.43 L492.127 1477.43 L492.127 1485.58 L487.474 1485.58 L487.474 1477.43 L471.872 1477.43 L471.872 1472.92 L486.247 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M514.789 1454.1 Q511.177 1454.1 509.349 1457.66 Q507.543 1461.2 507.543 1468.33 Q507.543 1475.44 509.349 1479.01 Q511.177 1482.55 514.789 1482.55 Q518.423 1482.55 520.228 1479.01 Q522.057 1475.44 522.057 1468.33 Q522.057 1461.2 520.228 1457.66 Q518.423 1454.1 514.789 1454.1 M514.789 1450.39 Q520.599 1450.39 523.654 1455 Q526.733 1459.58 526.733 1468.33 Q526.733 1477.06 523.654 1481.67 Q520.599 1486.25 514.789 1486.25 Q508.978 1486.25 505.9 1481.67 Q502.844 1477.06 502.844 1468.33 Q502.844 1459.58 505.9 1455 Q508.978 1450.39 514.789 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M544.95 1454.1 Q541.339 1454.1 539.511 1457.66 Q537.705 1461.2 537.705 1468.33 Q537.705 1475.44 539.511 1479.01 Q541.339 1482.55 544.95 1482.55 Q548.585 1482.55 550.39 1479.01 Q552.219 1475.44 552.219 1468.33 Q552.219 1461.2 550.39 1457.66 Q548.585 1454.1 544.95 1454.1 M544.95 1450.39 Q550.761 1450.39 553.816 1455 Q556.895 1459.58 556.895 1468.33 Q556.895 1477.06 553.816 1481.67 Q550.761 1486.25 544.95 1486.25 Q539.14 1486.25 536.062 1481.67 Q533.006 1477.06 533.006 1468.33 Q533.006 1459.58 536.062 1455 Q539.14 1450.39 544.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M869.751 1455.09 L857.946 1473.54 L869.751 1473.54 L869.751 1455.09 M868.525 1451.02 L874.404 1451.02 L874.404 1473.54 L879.335 1473.54 L879.335 1477.43 L874.404 1477.43 L874.404 1485.58 L869.751 1485.58 L869.751 1477.43 L854.15 1477.43 L854.15 1472.92 L868.525 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M887.112 1451.02 L905.469 1451.02 L905.469 1454.96 L891.395 1454.96 L891.395 1463.43 Q892.413 1463.08 893.432 1462.92 Q894.45 1462.73 895.469 1462.73 Q901.256 1462.73 904.635 1465.9 Q908.015 1469.08 908.015 1474.49 Q908.015 1480.07 904.543 1483.17 Q901.071 1486.25 894.751 1486.25 Q892.575 1486.25 890.307 1485.88 Q888.061 1485.51 885.654 1484.77 L885.654 1480.07 Q887.737 1481.2 889.96 1481.76 Q892.182 1482.32 894.659 1482.32 Q898.663 1482.32 901.001 1480.21 Q903.339 1478.1 903.339 1474.49 Q903.339 1470.88 901.001 1468.77 Q898.663 1466.67 894.659 1466.67 Q892.784 1466.67 890.909 1467.08 Q889.057 1467.5 887.112 1468.38 L887.112 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M927.228 1454.1 Q923.617 1454.1 921.788 1457.66 Q919.983 1461.2 919.983 1468.33 Q919.983 1475.44 921.788 1479.01 Q923.617 1482.55 927.228 1482.55 Q930.862 1482.55 932.668 1479.01 Q934.496 1475.44 934.496 1468.33 Q934.496 1461.2 932.668 1457.66 Q930.862 1454.1 927.228 1454.1 M927.228 1450.39 Q933.038 1450.39 936.094 1455 Q939.172 1459.58 939.172 1468.33 Q939.172 1477.06 936.094 1481.67 Q933.038 1486.25 927.228 1486.25 Q921.418 1486.25 918.339 1481.67 Q915.284 1477.06 915.284 1468.33 Q915.284 1459.58 918.339 1455 Q921.418 1450.39 927.228 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1238.56 1451.02 L1256.91 1451.02 L1256.91 1454.96 L1242.84 1454.96 L1242.84 1463.43 Q1243.86 1463.08 1244.88 1462.92 Q1245.89 1462.73 1246.91 1462.73 Q1252.7 1462.73 1256.08 1465.9 Q1259.46 1469.08 1259.46 1474.49 Q1259.46 1480.07 1255.99 1483.17 Q1252.51 1486.25 1246.2 1486.25 Q1244.02 1486.25 1241.75 1485.88 Q1239.51 1485.51 1237.1 1484.77 L1237.1 1480.07 Q1239.18 1481.2 1241.4 1481.76 Q1243.63 1482.32 1246.1 1482.32 Q1250.11 1482.32 1252.45 1480.21 Q1254.78 1478.1 1254.78 1474.49 Q1254.78 1470.88 1252.45 1468.77 Q1250.11 1466.67 1246.1 1466.67 Q1244.23 1466.67 1242.35 1467.08 Q1240.5 1467.5 1238.56 1468.38 L1238.56 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1278.67 1454.1 Q1275.06 1454.1 1273.23 1457.66 Q1271.43 1461.2 1271.43 1468.33 Q1271.43 1475.44 1273.23 1479.01 Q1275.06 1482.55 1278.67 1482.55 Q1282.31 1482.55 1284.11 1479.01 Q1285.94 1475.44 1285.94 1468.33 Q1285.94 1461.2 1284.11 1457.66 Q1282.31 1454.1 1278.67 1454.1 M1278.67 1450.39 Q1284.48 1450.39 1287.54 1455 Q1290.62 1459.58 1290.62 1468.33 Q1290.62 1477.06 1287.54 1481.67 Q1284.48 1486.25 1278.67 1486.25 Q1272.86 1486.25 1269.78 1481.67 Q1266.73 1477.06 1266.73 1468.33 Q1266.73 1459.58 1269.78 1455 Q1272.86 1450.39 1278.67 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1308.83 1454.1 Q1305.22 1454.1 1303.39 1457.66 Q1301.59 1461.2 1301.59 1468.33 Q1301.59 1475.44 1303.39 1479.01 Q1305.22 1482.55 1308.83 1482.55 Q1312.47 1482.55 1314.27 1479.01 Q1316.1 1475.44 1316.1 1468.33 Q1316.1 1461.2 1314.27 1457.66 Q1312.47 1454.1 1308.83 1454.1 M1308.83 1450.39 Q1314.64 1450.39 1317.7 1455 Q1320.78 1459.58 1320.78 1468.33 Q1320.78 1477.06 1317.7 1481.67 Q1314.64 1486.25 1308.83 1486.25 Q1303.02 1486.25 1299.95 1481.67 Q1296.89 1477.06 1296.89 1468.33 Q1296.89 1459.58 1299.95 1455 Q1303.02 1450.39 1308.83 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1620.83 1451.02 L1639.19 1451.02 L1639.19 1454.96 L1625.12 1454.96 L1625.12 1463.43 Q1626.14 1463.08 1627.15 1462.92 Q1628.17 1462.73 1629.19 1462.73 Q1634.98 1462.73 1638.36 1465.9 Q1641.74 1469.08 1641.74 1474.49 Q1641.74 1480.07 1638.26 1483.17 Q1634.79 1486.25 1628.47 1486.25 Q1626.3 1486.25 1624.03 1485.88 Q1621.78 1485.51 1619.38 1484.77 L1619.38 1480.07 Q1621.46 1481.2 1623.68 1481.76 Q1625.9 1482.32 1628.38 1482.32 Q1632.39 1482.32 1634.72 1480.21 Q1637.06 1478.1 1637.06 1474.49 Q1637.06 1470.88 1634.72 1468.77 Q1632.39 1466.67 1628.38 1466.67 Q1626.51 1466.67 1624.63 1467.08 Q1622.78 1467.5 1620.83 1468.38 L1620.83 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1651 1451.02 L1669.35 1451.02 L1669.35 1454.96 L1655.28 1454.96 L1655.28 1463.43 Q1656.3 1463.08 1657.32 1462.92 Q1658.33 1462.73 1659.35 1462.73 Q1665.14 1462.73 1668.52 1465.9 Q1671.9 1469.08 1671.9 1474.49 Q1671.9 1480.07 1668.43 1483.17 Q1664.95 1486.25 1658.63 1486.25 Q1656.46 1486.25 1654.19 1485.88 Q1651.95 1485.51 1649.54 1484.77 L1649.54 1480.07 Q1651.62 1481.2 1653.84 1481.76 Q1656.07 1482.32 1658.54 1482.32 Q1662.55 1482.32 1664.88 1480.21 Q1667.22 1478.1 1667.22 1474.49 Q1667.22 1470.88 1664.88 1468.77 Q1662.55 1466.67 1658.54 1466.67 Q1656.67 1466.67 1654.79 1467.08 Q1652.94 1467.5 1651 1468.38 L1651 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1691.11 1454.1 Q1687.5 1454.1 1685.67 1457.66 Q1683.87 1461.2 1683.87 1468.33 Q1683.87 1475.44 1685.67 1479.01 Q1687.5 1482.55 1691.11 1482.55 Q1694.75 1482.55 1696.55 1479.01 Q1698.38 1475.44 1698.38 1468.33 Q1698.38 1461.2 1696.55 1457.66 Q1694.75 1454.1 1691.11 1454.1 M1691.11 1450.39 Q1696.92 1450.39 1699.98 1455 Q1703.06 1459.58 1703.06 1468.33 Q1703.06 1477.06 1699.98 1481.67 Q1696.92 1486.25 1691.11 1486.25 Q1685.3 1486.25 1682.22 1481.67 Q1679.17 1477.06 1679.17 1468.33 Q1679.17 1459.58 1682.22 1455 Q1685.3 1450.39 1691.11 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2013.82 1466.44 Q2010.67 1466.44 2008.82 1468.59 Q2006.99 1470.74 2006.99 1474.49 Q2006.99 1478.22 2008.82 1480.39 Q2010.67 1482.55 2013.82 1482.55 Q2016.97 1482.55 2018.79 1480.39 Q2020.65 1478.22 2020.65 1474.49 Q2020.65 1470.74 2018.79 1468.59 Q2016.97 1466.44 2013.82 1466.44 M2023.1 1451.78 L2023.1 1456.04 Q2021.34 1455.21 2019.54 1454.77 Q2017.75 1454.33 2015.99 1454.33 Q2011.36 1454.33 2008.91 1457.45 Q2006.48 1460.58 2006.13 1466.9 Q2007.5 1464.89 2009.56 1463.82 Q2011.62 1462.73 2014.1 1462.73 Q2019.3 1462.73 2022.31 1465.9 Q2025.35 1469.05 2025.35 1474.49 Q2025.35 1479.82 2022.2 1483.03 Q2019.05 1486.25 2013.82 1486.25 Q2007.82 1486.25 2004.65 1481.67 Q2001.48 1477.06 2001.48 1468.33 Q2001.48 1460.14 2005.37 1455.28 Q2009.26 1450.39 2015.81 1450.39 Q2017.57 1450.39 2019.35 1450.74 Q2021.16 1451.09 2023.1 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2043.4 1454.1 Q2039.79 1454.1 2037.96 1457.66 Q2036.16 1461.2 2036.16 1468.33 Q2036.16 1475.44 2037.96 1479.01 Q2039.79 1482.55 2043.4 1482.55 Q2047.04 1482.55 2048.84 1479.01 Q2050.67 1475.44 2050.67 1468.33 Q2050.67 1461.2 2048.84 1457.66 Q2047.04 1454.1 2043.4 1454.1 M2043.4 1450.39 Q2049.21 1450.39 2052.27 1455 Q2055.35 1459.58 2055.35 1468.33 Q2055.35 1477.06 2052.27 1481.67 Q2049.21 1486.25 2043.4 1486.25 Q2037.59 1486.25 2034.51 1481.67 Q2031.46 1477.06 2031.46 1468.33 Q2031.46 1459.58 2034.51 1455 Q2037.59 1450.39 2043.4 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M2073.56 1454.1 Q2069.95 1454.1 2068.12 1457.66 Q2066.32 1461.2 2066.32 1468.33 Q2066.32 1475.44 2068.12 1479.01 Q2069.95 1482.55 2073.56 1482.55 Q2077.2 1482.55 2079 1479.01 Q2080.83 1475.44 2080.83 1468.33 Q2080.83 1461.2 2079 1457.66 Q2077.2 1454.1 2073.56 1454.1 M2073.56 1450.39 Q2079.37 1450.39 2082.43 1455 Q2085.51 1459.58 2085.51 1468.33 Q2085.51 1477.06 2082.43 1481.67 Q2079.37 1486.25 2073.56 1486.25 Q2067.75 1486.25 2064.67 1481.67 Q2061.62 1477.06 2061.62 1468.33 Q2061.62 1459.58 2064.67 1455 Q2067.75 1450.39 2073.56 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1149.83 1525.81 L1149.83 1543.66 L1157.91 1543.66 Q1162.4 1543.66 1164.85 1541.34 Q1167.3 1539.02 1167.3 1534.72 Q1167.3 1530.45 1164.85 1528.13 Q1162.4 1525.81 1157.91 1525.81 L1149.83 1525.81 M1143.4 1520.52 L1157.91 1520.52 Q1165.9 1520.52 1169.97 1524.15 Q1174.08 1527.75 1174.08 1534.72 Q1174.08 1541.75 1169.97 1545.35 Q1165.9 1548.95 1157.91 1548.95 L1149.83 1548.95 L1149.83 1568.04 L1143.4 1568.04 L1143.4 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1201.96 1537.87 Q1200.97 1537.3 1199.8 1537.04 Q1198.65 1536.76 1197.25 1536.76 Q1192.29 1536.76 1189.61 1540 Q1186.97 1543.22 1186.97 1549.27 L1186.97 1568.04 L1181.08 1568.04 L1181.08 1532.4 L1186.97 1532.4 L1186.97 1537.93 Q1188.82 1534.69 1191.78 1533.13 Q1194.74 1531.54 1198.97 1531.54 Q1199.57 1531.54 1200.31 1531.63 Q1201.04 1531.7 1201.93 1531.85 L1201.96 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1220.49 1536.5 Q1215.77 1536.5 1213.04 1540.19 Q1210.3 1543.85 1210.3 1550.25 Q1210.3 1556.65 1213.01 1560.34 Q1215.74 1564 1220.49 1564 Q1225.16 1564 1227.9 1560.31 Q1230.64 1556.62 1230.64 1550.25 Q1230.64 1543.92 1227.9 1540.23 Q1225.16 1536.5 1220.49 1536.5 M1220.49 1531.54 Q1228.12 1531.54 1232.48 1536.5 Q1236.85 1541.47 1236.85 1550.25 Q1236.85 1559 1232.48 1564 Q1228.12 1568.97 1220.49 1568.97 Q1212.81 1568.97 1208.45 1564 Q1204.13 1559 1204.13 1550.25 Q1204.13 1541.47 1208.45 1536.5 Q1212.81 1531.54 1220.49 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1264.6 1518.52 L1264.6 1523.39 L1259 1523.39 Q1255.85 1523.39 1254.61 1524.66 Q1253.4 1525.93 1253.4 1529.24 L1253.4 1532.4 L1263.04 1532.4 L1263.04 1536.95 L1253.4 1536.95 L1253.4 1568.04 L1247.51 1568.04 L1247.51 1536.95 L1241.91 1536.95 L1241.91 1532.4 L1247.51 1532.4 L1247.51 1529.91 Q1247.51 1523.96 1250.28 1521.26 Q1253.05 1518.52 1259.06 1518.52 L1264.6 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1269.5 1532.4 L1275.36 1532.4 L1275.36 1568.04 L1269.5 1568.04 L1269.5 1532.4 M1269.5 1518.52 L1275.36 1518.52 L1275.36 1525.93 L1269.5 1525.93 L1269.5 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1293.4 1522.27 L1293.4 1532.4 L1305.47 1532.4 L1305.47 1536.95 L1293.4 1536.95 L1293.4 1556.3 Q1293.4 1560.66 1294.58 1561.9 Q1295.79 1563.14 1299.45 1563.14 L1305.47 1563.14 L1305.47 1568.04 L1299.45 1568.04 Q1292.67 1568.04 1290.09 1565.53 Q1287.52 1562.98 1287.52 1556.3 L1287.52 1536.95 L1283.22 1536.95 L1283.22 1532.4 L1287.52 1532.4 L1287.52 1522.27 L1293.4 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1333.35 1518.52 L1346.84 1518.52 L1346.84 1523.07 L1339.21 1523.07 L1339.21 1572.09 L1346.84 1572.09 L1346.84 1576.64 L1333.35 1576.64 L1333.35 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1375.2 1577.62 L1372.02 1577.62 L1371.99 1568.04 Q1368.65 1567.98 1365.31 1567.28 Q1361.96 1566.55 1358.59 1565.12 L1358.59 1559.39 Q1361.84 1561.42 1365.15 1562.47 Q1368.49 1563.49 1372.02 1563.52 L1372.02 1549.01 Q1364.99 1547.86 1361.77 1545.13 Q1358.59 1542.39 1358.59 1537.62 Q1358.59 1532.43 1362.06 1529.44 Q1365.53 1526.44 1372.02 1526 L1372.02 1518.52 L1375.2 1518.52 L1375.2 1525.9 Q1378.16 1526.03 1380.93 1526.54 Q1383.7 1527.02 1386.34 1527.88 L1386.34 1533.45 Q1383.7 1532.11 1380.9 1531.38 Q1378.13 1530.65 1375.2 1530.52 L1375.2 1544.11 Q1382.43 1545.22 1385.83 1548.09 Q1389.24 1550.95 1389.24 1555.92 Q1389.24 1561.3 1385.61 1564.42 Q1382.02 1567.5 1375.2 1567.98 L1375.2 1577.62 M1372.02 1543.54 L1372.02 1530.49 Q1368.33 1530.9 1366.39 1532.59 Q1364.45 1534.27 1364.45 1537.07 Q1364.45 1539.81 1366.23 1541.34 Q1368.04 1542.87 1372.02 1543.54 M1375.2 1549.65 L1375.2 1563.43 Q1379.25 1562.89 1381.28 1561.14 Q1383.35 1559.39 1383.35 1556.52 Q1383.35 1553.72 1381.38 1552.07 Q1379.44 1550.41 1375.2 1549.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M1414.48 1518.52 L1414.48 1576.64 L1400.98 1576.64 L1400.98 1572.09 L1408.59 1572.09 L1408.59 1523.07 L1400.98 1523.07 L1400.98 1518.52 L1414.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,1117.42 2352.76,1117.42 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,811.653 2352.76,811.653 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,505.889 2352.76,505.889 "/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="205.121,200.126 2352.76,200.126 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 205.121,47.2441 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1423.18 224.019,1423.18 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,1117.42 224.019,1117.42 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,811.653 224.019,811.653 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,505.889 224.019,505.889 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="205.121,200.126 224.019,200.126 "/>
+<path clip-path="url(#clip710)" d="M157.177 1408.98 Q153.566 1408.98 151.737 1412.54 Q149.931 1416.08 149.931 1423.21 Q149.931 1430.32 151.737 1433.89 Q153.566 1437.43 157.177 1437.43 Q160.811 1437.43 162.616 1433.89 Q164.445 1430.32 164.445 1423.21 Q164.445 1416.08 162.616 1412.54 Q160.811 1408.98 157.177 1408.98 M157.177 1405.27 Q162.987 1405.27 166.042 1409.88 Q169.121 1414.46 169.121 1423.21 Q169.121 1431.94 166.042 1436.55 Q162.987 1441.13 157.177 1441.13 Q151.366 1441.13 148.288 1436.55 Q145.232 1431.94 145.232 1423.21 Q145.232 1414.46 148.288 1409.88 Q151.366 1405.27 157.177 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M117.825 1130.76 L125.464 1130.76 L125.464 1104.4 L117.154 1106.06 L117.154 1101.8 L125.418 1100.14 L130.093 1100.14 L130.093 1130.76 L137.732 1130.76 L137.732 1134.7 L117.825 1134.7 L117.825 1130.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M157.177 1103.22 Q153.566 1103.22 151.737 1106.78 Q149.931 1110.32 149.931 1117.45 Q149.931 1124.56 151.737 1128.12 Q153.566 1131.66 157.177 1131.66 Q160.811 1131.66 162.616 1128.12 Q164.445 1124.56 164.445 1117.45 Q164.445 1110.32 162.616 1106.78 Q160.811 1103.22 157.177 1103.22 M157.177 1099.51 Q162.987 1099.51 166.042 1104.12 Q169.121 1108.7 169.121 1117.45 Q169.121 1126.18 166.042 1130.78 Q162.987 1135.37 157.177 1135.37 Q151.366 1135.37 148.288 1130.78 Q145.232 1126.18 145.232 1117.45 Q145.232 1108.7 148.288 1104.12 Q151.366 1099.51 157.177 1099.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M121.043 824.998 L137.362 824.998 L137.362 828.933 L115.418 828.933 L115.418 824.998 Q118.08 822.243 122.663 817.613 Q127.269 812.961 128.45 811.618 Q130.695 809.095 131.575 807.359 Q132.478 805.6 132.478 803.91 Q132.478 801.155 130.533 799.419 Q128.612 797.683 125.51 797.683 Q123.311 797.683 120.857 798.447 Q118.427 799.211 115.649 800.762 L115.649 796.04 Q118.473 794.905 120.927 794.327 Q123.38 793.748 125.418 793.748 Q130.788 793.748 133.982 796.433 Q137.177 799.118 137.177 803.609 Q137.177 805.739 136.367 807.66 Q135.579 809.558 133.473 812.151 Q132.894 812.822 129.792 816.039 Q126.691 819.234 121.043 824.998 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M157.177 797.452 Q153.566 797.452 151.737 801.016 Q149.931 804.558 149.931 811.688 Q149.931 818.794 151.737 822.359 Q153.566 825.9 157.177 825.9 Q160.811 825.9 162.616 822.359 Q164.445 818.794 164.445 811.688 Q164.445 804.558 162.616 801.016 Q160.811 797.452 157.177 797.452 M157.177 793.748 Q162.987 793.748 166.042 798.354 Q169.121 802.938 169.121 811.688 Q169.121 820.414 166.042 825.021 Q162.987 829.604 157.177 829.604 Q151.366 829.604 148.288 825.021 Q145.232 820.414 145.232 811.688 Q145.232 802.938 148.288 798.354 Q151.366 793.748 157.177 793.748 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M131.181 504.535 Q134.538 505.253 136.413 507.521 Q138.311 509.79 138.311 513.123 Q138.311 518.239 134.792 521.04 Q131.274 523.841 124.793 523.841 Q122.617 523.841 120.302 523.401 Q118.01 522.984 115.556 522.128 L115.556 517.614 Q117.501 518.748 119.816 519.327 Q122.13 519.905 124.654 519.905 Q129.052 519.905 131.343 518.169 Q133.658 516.433 133.658 513.123 Q133.658 510.068 131.505 508.355 Q129.376 506.619 125.556 506.619 L121.529 506.619 L121.529 502.776 L125.742 502.776 Q129.191 502.776 131.019 501.41 Q132.848 500.021 132.848 497.429 Q132.848 494.767 130.95 493.355 Q129.075 491.92 125.556 491.92 Q123.635 491.92 121.436 492.336 Q119.237 492.753 116.598 493.632 L116.598 489.466 Q119.26 488.725 121.575 488.355 Q123.913 487.984 125.973 487.984 Q131.297 487.984 134.399 490.415 Q137.501 492.822 137.501 496.943 Q137.501 499.813 135.857 501.804 Q134.214 503.771 131.181 504.535 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M157.177 491.688 Q153.566 491.688 151.737 495.253 Q149.931 498.794 149.931 505.924 Q149.931 513.031 151.737 516.595 Q153.566 520.137 157.177 520.137 Q160.811 520.137 162.616 516.595 Q164.445 513.031 164.445 505.924 Q164.445 498.794 162.616 495.253 Q160.811 491.688 157.177 491.688 M157.177 487.984 Q162.987 487.984 166.042 492.591 Q169.121 497.174 169.121 505.924 Q169.121 514.651 166.042 519.257 Q162.987 523.841 157.177 523.841 Q151.366 523.841 148.288 519.257 Q145.232 514.651 145.232 505.924 Q145.232 497.174 148.288 492.591 Q151.366 487.984 157.177 487.984 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M129.862 186.92 L118.056 205.369 L129.862 205.369 L129.862 186.92 M128.635 182.846 L134.515 182.846 L134.515 205.369 L139.445 205.369 L139.445 209.258 L134.515 209.258 L134.515 217.406 L129.862 217.406 L129.862 209.258 L114.26 209.258 L114.26 204.744 L128.635 182.846 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M157.177 185.925 Q153.566 185.925 151.737 189.489 Q149.931 193.031 149.931 200.161 Q149.931 207.267 151.737 210.832 Q153.566 214.373 157.177 214.373 Q160.811 214.373 162.616 210.832 Q164.445 207.267 164.445 200.161 Q164.445 193.031 162.616 189.489 Q160.811 185.925 157.177 185.925 M157.177 182.221 Q162.987 182.221 166.042 186.827 Q169.121 191.411 169.121 200.161 Q169.121 208.887 166.042 213.494 Q162.987 218.077 157.177 218.077 Q151.366 218.077 148.288 213.494 Q145.232 208.887 145.232 200.161 Q145.232 191.411 148.288 186.827 Q151.366 182.221 157.177 182.221 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M16.4842 1073.5 L16.4842 1064.84 L56.238 1043.77 L16.4842 1043.77 L16.4842 1037.54 L64.0042 1037.54 L64.0042 1046.19 L24.2503 1067.26 L64.0042 1067.26 L64.0042 1073.5 L16.4842 1073.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M49.9359 1025.6 L28.3562 1025.6 L28.3562 1019.74 L49.7131 1019.74 Q54.7739 1019.74 57.3202 1017.77 Q59.8346 1015.8 59.8346 1011.85 Q59.8346 1007.11 56.8109 1004.37 Q53.7872 1001.6 48.5673 1001.6 L28.3562 1001.6 L28.3562 995.744 L64.0042 995.744 L64.0042 1001.6 L58.5296 1001.6 Q61.7762 1003.73 63.3676 1006.57 Q64.9272 1009.37 64.9272 1013.09 Q64.9272 1019.23 61.1078 1022.42 Q57.2883 1025.6 49.9359 1025.6 M27.4968 1010.86 L27.4968 1010.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M35.1993 955.927 Q31.2526 953.731 29.3747 950.675 Q27.4968 947.62 27.4968 943.482 Q27.4968 937.912 31.4117 934.888 Q35.2948 931.865 42.4881 931.865 L64.0042 931.865 L64.0042 937.753 L42.679 937.753 Q37.5546 937.753 35.072 939.567 Q32.5894 941.381 32.5894 945.105 Q32.5894 949.657 35.6131 952.298 Q38.6368 954.94 43.8567 954.94 L64.0042 954.94 L64.0042 960.829 L42.679 960.829 Q37.5228 960.829 35.072 962.643 Q32.5894 964.457 32.5894 968.245 Q32.5894 972.732 35.6449 975.374 Q38.6686 978.016 43.8567 978.016 L64.0042 978.016 L64.0042 983.904 L28.3562 983.904 L28.3562 978.016 L33.8944 978.016 Q30.616 976.011 29.0564 973.21 Q27.4968 970.409 27.4968 966.558 Q27.4968 962.675 29.4702 959.969 Q31.4436 957.232 35.1993 955.927 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M46.212 894.593 Q39.7508 894.593 36.0905 897.267 Q32.3984 899.909 32.3984 904.556 Q32.3984 909.203 36.0905 911.876 Q39.7508 914.518 46.212 914.518 Q52.6732 914.518 56.3653 911.876 Q60.0256 909.203 60.0256 904.556 Q60.0256 899.909 56.3653 897.267 Q52.6732 894.593 46.212 894.593 M33.7671 914.518 Q30.5842 912.672 29.0564 909.871 Q27.4968 907.038 27.4968 903.123 Q27.4968 896.63 32.6531 892.588 Q37.8093 888.514 46.212 888.514 Q54.6147 888.514 59.771 892.588 Q64.9272 896.63 64.9272 903.123 Q64.9272 907.038 63.3994 909.871 Q61.8398 912.672 58.657 914.518 L64.0042 914.518 L64.0042 920.406 L14.479 920.406 L14.479 914.518 L33.7671 914.518 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M44.7161 848.315 L47.5806 848.315 L47.5806 875.242 Q53.6281 874.86 56.8109 871.613 Q59.9619 868.335 59.9619 862.51 Q59.9619 859.136 59.1344 855.985 Q58.3069 852.802 56.6518 849.683 L62.1899 849.683 Q63.5267 852.834 64.227 856.144 Q64.9272 859.455 64.9272 862.86 Q64.9272 871.39 59.9619 876.387 Q54.9967 881.353 46.5303 881.353 Q37.7774 881.353 32.6531 876.642 Q27.4968 871.9 27.4968 863.879 Q27.4968 856.686 32.1438 852.516 Q36.7589 848.315 44.7161 848.315 M42.9973 854.171 Q38.1912 854.235 35.3266 856.877 Q32.4621 859.486 32.4621 863.815 Q32.4621 868.717 35.2312 871.677 Q38.0002 874.605 43.0292 875.051 L42.9973 854.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M33.8307 818.046 Q33.2578 819.032 33.0032 820.21 Q32.7167 821.356 32.7167 822.756 Q32.7167 827.722 35.9632 830.395 Q39.1779 833.037 45.2253 833.037 L64.0042 833.037 L64.0042 838.925 L28.3562 838.925 L28.3562 833.037 L33.8944 833.037 Q30.6479 831.191 29.0883 828.231 Q27.4968 825.271 27.4968 821.038 Q27.4968 820.433 27.5923 819.701 Q27.656 818.969 27.8151 818.078 L33.8307 818.046 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M32.4621 777.369 Q32.4621 782.079 36.1542 784.817 Q39.8145 787.554 46.212 787.554 Q52.6095 787.554 56.3017 784.849 Q59.9619 782.111 59.9619 777.369 Q59.9619 772.69 56.2698 769.953 Q52.5777 767.216 46.212 767.216 Q39.8781 767.216 36.186 769.953 Q32.4621 772.69 32.4621 777.369 M27.4968 777.369 Q27.4968 769.73 32.4621 765.37 Q37.4273 761.009 46.212 761.009 Q54.9649 761.009 59.9619 765.37 Q64.9272 769.73 64.9272 777.369 Q64.9272 785.04 59.9619 789.4 Q54.9649 793.729 46.212 793.729 Q37.4273 793.729 32.4621 789.4 Q27.4968 785.04 27.4968 777.369 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M14.479 733.255 L19.3487 733.255 L19.3487 738.856 Q19.3487 742.007 20.6219 743.249 Q21.895 744.458 25.2052 744.458 L28.3562 744.458 L28.3562 734.814 L32.9077 734.814 L32.9077 744.458 L64.0042 744.458 L64.0042 750.346 L32.9077 750.346 L32.9077 755.948 L28.3562 755.948 L28.3562 750.346 L25.8736 750.346 Q19.9216 750.346 17.2162 747.577 Q14.479 744.808 14.479 738.793 L14.479 733.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M32.4621 693.819 Q32.4621 698.53 36.1542 701.267 Q39.8145 704.004 46.212 704.004 Q52.6095 704.004 56.3017 701.299 Q59.9619 698.561 59.9619 693.819 Q59.9619 689.14 56.2698 686.403 Q52.5777 683.666 46.212 683.666 Q39.8781 683.666 36.186 686.403 Q32.4621 689.14 32.4621 693.819 M27.4968 693.819 Q27.4968 686.18 32.4621 681.82 Q37.4273 677.459 46.212 677.459 Q54.9649 677.459 59.9619 681.82 Q64.9272 686.18 64.9272 693.819 Q64.9272 701.49 59.9619 705.85 Q54.9649 710.179 46.212 710.179 Q37.4273 710.179 32.4621 705.85 Q27.4968 701.49 27.4968 693.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M49.9359 668.356 L28.3562 668.356 L28.3562 662.5 L49.7131 662.5 Q54.7739 662.5 57.3202 660.526 Q59.8346 658.553 59.8346 654.606 Q59.8346 649.864 56.8109 647.127 Q53.7872 644.357 48.5673 644.357 L28.3562 644.357 L28.3562 638.501 L64.0042 638.501 L64.0042 644.357 L58.5296 644.357 Q61.7762 646.49 63.3676 649.323 Q64.9272 652.124 64.9272 655.848 Q64.9272 661.99 61.1078 665.173 Q57.2883 668.356 49.9359 668.356 M27.4968 653.62 L27.4968 653.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M18.2347 620.645 L28.3562 620.645 L28.3562 608.582 L32.9077 608.582 L32.9077 620.645 L52.2594 620.645 Q56.6199 620.645 57.8613 619.468 Q59.1026 618.258 59.1026 614.598 L59.1026 608.582 L64.0042 608.582 L64.0042 614.598 Q64.0042 621.377 61.4897 623.955 Q58.9434 626.533 52.2594 626.533 L32.9077 626.533 L32.9077 630.83 L28.3562 630.83 L28.3562 626.533 L18.2347 626.533 L18.2347 620.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M29.7248 575.226 L35.1993 575.226 Q33.8307 577.708 33.1623 580.223 Q32.4621 582.706 32.4621 585.252 Q32.4621 590.949 36.0905 594.1 Q39.6872 597.251 46.212 597.251 Q52.7369 597.251 56.3653 594.1 Q59.9619 590.949 59.9619 585.252 Q59.9619 582.706 59.2935 580.223 Q58.5933 577.708 57.2247 575.226 L62.6355 575.226 Q63.7814 577.677 64.3543 580.318 Q64.9272 582.928 64.9272 585.888 Q64.9272 593.941 59.8664 598.683 Q54.8057 603.426 46.212 603.426 Q37.491 603.426 32.4939 598.652 Q27.4968 593.846 27.4968 585.506 Q27.4968 582.801 28.0697 580.223 Q28.6108 577.645 29.7248 575.226 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M32.4621 551.227 Q32.4621 555.938 36.1542 558.675 Q39.8145 561.412 46.212 561.412 Q52.6095 561.412 56.3017 558.707 Q59.9619 555.97 59.9619 551.227 Q59.9619 546.548 56.2698 543.811 Q52.5777 541.074 46.212 541.074 Q39.8781 541.074 36.186 543.811 Q32.4621 546.548 32.4621 551.227 M27.4968 551.227 Q27.4968 543.588 32.4621 539.228 Q37.4273 534.867 46.212 534.867 Q54.9649 534.867 59.9619 539.228 Q64.9272 543.588 64.9272 551.227 Q64.9272 558.898 59.9619 563.258 Q54.9649 567.587 46.212 567.587 Q37.4273 567.587 32.4621 563.258 Q27.4968 558.898 27.4968 551.227 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M35.1993 497.405 Q31.2526 495.209 29.3747 492.153 Q27.4968 489.098 27.4968 484.96 Q27.4968 479.39 31.4117 476.366 Q35.2948 473.343 42.4881 473.343 L64.0042 473.343 L64.0042 479.231 L42.679 479.231 Q37.5546 479.231 35.072 481.045 Q32.5894 482.859 32.5894 486.583 Q32.5894 491.135 35.6131 493.777 Q38.6368 496.418 43.8567 496.418 L64.0042 496.418 L64.0042 502.307 L42.679 502.307 Q37.5228 502.307 35.072 504.121 Q32.5894 505.935 32.5894 509.723 Q32.5894 514.211 35.6449 516.852 Q38.6686 519.494 43.8567 519.494 L64.0042 519.494 L64.0042 525.382 L28.3562 525.382 L28.3562 519.494 L33.8944 519.494 Q30.616 517.489 29.0564 514.688 Q27.4968 511.887 27.4968 508.036 Q27.4968 504.153 29.4702 501.447 Q31.4436 498.71 35.1993 497.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M44.7161 431.17 L47.5806 431.17 L47.5806 458.097 Q53.6281 457.715 56.8109 454.468 Q59.9619 451.19 59.9619 445.365 Q59.9619 441.992 59.1344 438.841 Q58.3069 435.658 56.6518 432.539 L62.1899 432.539 Q63.5267 435.69 64.227 439 Q64.9272 442.31 64.9272 445.716 Q64.9272 454.246 59.9619 459.243 Q54.9967 464.208 46.5303 464.208 Q37.7774 464.208 32.6531 459.497 Q27.4968 454.755 27.4968 446.734 Q27.4968 439.541 32.1438 435.371 Q36.7589 431.17 44.7161 431.17 M42.9973 437.026 Q38.1912 437.09 35.3266 439.732 Q32.4621 442.342 32.4621 446.67 Q32.4621 451.572 35.2312 454.532 Q38.0002 457.46 43.0292 457.906 L42.9973 437.026 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M29.4065 398.832 L34.9447 398.832 Q33.6716 401.315 33.035 403.988 Q32.3984 406.662 32.3984 409.527 Q32.3984 413.887 33.7352 416.083 Q35.072 418.248 37.7456 418.248 Q39.7826 418.248 40.9603 416.688 Q42.1061 415.128 43.1565 410.418 L43.6021 408.413 Q44.9389 402.174 47.3897 399.564 Q49.8086 396.922 54.1691 396.922 Q59.1344 396.922 62.0308 400.869 Q64.9272 404.784 64.9272 411.659 Q64.9272 414.524 64.3543 417.643 Q63.8132 420.73 62.6992 424.168 L56.6518 424.168 Q58.3387 420.921 59.198 417.77 Q60.0256 414.619 60.0256 411.532 Q60.0256 407.394 58.6251 405.166 Q57.1929 402.938 54.6147 402.938 Q52.2276 402.938 50.9545 404.561 Q49.6813 406.153 48.5037 411.595 L48.0262 413.632 Q46.8804 419.075 44.5251 421.494 Q42.138 423.913 38.0002 423.913 Q32.9713 423.913 30.2341 420.348 Q27.4968 416.783 27.4968 410.227 Q27.4968 406.98 27.9743 404.116 Q28.4517 401.251 29.4065 398.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip712)" d="M323.245 1392.6 L323.245 1423.18 L514.383 1423.18 L514.383 1392.6 L323.245 1392.6 L323.245 1392.6  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="323.245,1392.6 323.245,1423.18 514.383,1423.18 514.383,1392.6 323.245,1392.6 "/>
+<path clip-path="url(#clip712)" d="M514.383 1331.45 L514.383 1423.18 L705.522 1423.18 L705.522 1331.45 L514.383 1331.45 L514.383 1331.45  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="514.383,1331.45 514.383,1423.18 705.522,1423.18 705.522,1331.45 514.383,1331.45 "/>
+<path clip-path="url(#clip712)" d="M705.522 1300.87 L705.522 1423.18 L896.661 1423.18 L896.661 1300.87 L705.522 1300.87 L705.522 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="705.522,1300.87 705.522,1423.18 896.661,1423.18 896.661,1300.87 705.522,1300.87 "/>
+<path clip-path="url(#clip712)" d="M896.661 1300.87 L896.661 1423.18 L1087.8 1423.18 L1087.8 1300.87 L896.661 1300.87 L896.661 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="896.661,1300.87 896.661,1423.18 1087.8,1423.18 1087.8,1300.87 896.661,1300.87 "/>
+<path clip-path="url(#clip712)" d="M1087.8 1300.87 L1087.8 1423.18 L1278.94 1423.18 L1278.94 1300.87 L1087.8 1300.87 L1087.8 1300.87  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1087.8,1300.87 1087.8,1423.18 1278.94,1423.18 1278.94,1300.87 1087.8,1300.87 "/>
+<path clip-path="url(#clip712)" d="M1278.94 1178.57 L1278.94 1423.18 L1470.08 1423.18 L1470.08 1178.57 L1278.94 1178.57 L1278.94 1178.57  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1278.94,1178.57 1278.94,1423.18 1470.08,1423.18 1470.08,1178.57 1278.94,1178.57 "/>
+<path clip-path="url(#clip712)" d="M1470.08 1117.42 L1470.08 1423.18 L1661.22 1423.18 L1661.22 1117.42 L1470.08 1117.42 L1470.08 1117.42  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.08,1117.42 1470.08,1423.18 1661.22,1423.18 1661.22,1117.42 1470.08,1117.42 "/>
+<path clip-path="url(#clip712)" d="M1661.22 1086.84 L1661.22 1423.18 L1852.35 1423.18 L1852.35 1086.84 L1661.22 1086.84 L1661.22 1086.84  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1661.22,1086.84 1661.22,1423.18 1852.35,1423.18 1852.35,1086.84 1661.22,1086.84 "/>
+<path clip-path="url(#clip712)" d="M1852.35 1117.42 L1852.35 1423.18 L2043.49 1423.18 L2043.49 1117.42 L1852.35 1117.42 L1852.35 1117.42  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1852.35,1117.42 1852.35,1423.18 2043.49,1423.18 2043.49,1117.42 1852.35,1117.42 "/>
+<path clip-path="url(#clip712)" d="M2043.49 47.2441 L2043.49 1423.18 L2234.63 1423.18 L2234.63 47.2441 L2043.49 47.2441 L2043.49 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip712)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2043.49,47.2441 2043.49,1423.18 2234.63,1423.18 2234.63,47.2441 2043.49,47.2441 "/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="418.814" cy="1392.6" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="609.953" cy="1331.45" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="801.092" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="992.23" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1183.37" cy="1300.87" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1374.51" cy="1178.57" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1565.65" cy="1117.42" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1756.79" cy="1086.84" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1947.92" cy="1117.42" r="2"/>
+<circle clip-path="url(#clip712)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2139.06" cy="47.2441" r="2"/>
+<polyline clip-path="url(#clip712)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="1757.69,2799.12 1757.69,-1328.69 "/>
+<path clip-path="url(#clip710)" d="M276.709 196.789 L1017.7 196.789 L1017.7 93.1086 L276.709 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip710)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="276.709,196.789 1017.7,196.789 1017.7,93.1086 276.709,93.1086 276.709,196.789 "/>
+<polyline clip-path="url(#clip710)" style="stroke:#e26f46; stroke-linecap:round; stroke-linejoin:round; stroke-width:12; stroke-opacity:1; fill:none" points="300.571,144.949 443.747,144.949 "/>
+<path clip-path="url(#clip710)" d="M467.61 127.669 L489.461 127.669 L489.461 131.604 L472.286 131.604 L472.286 141.835 L488.744 141.835 L488.744 145.77 L472.286 145.77 L472.286 158.293 L489.878 158.293 L489.878 162.229 L467.61 162.229 L467.61 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M518.929 136.303 L509.554 148.919 L519.415 162.229 L514.392 162.229 L506.846 152.043 L499.299 162.229 L494.276 162.229 L504.346 148.664 L495.133 136.303 L500.156 136.303 L507.031 145.539 L513.906 136.303 L518.929 136.303 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M529.554 158.34 L529.554 172.09 L525.271 172.09 L525.271 136.303 L529.554 136.303 L529.554 140.238 Q530.896 137.923 532.933 136.812 Q534.994 135.678 537.841 135.678 Q542.563 135.678 545.503 139.428 Q548.466 143.178 548.466 149.289 Q548.466 155.4 545.503 159.15 Q542.563 162.9 537.841 162.9 Q534.994 162.9 532.933 161.789 Q530.896 160.655 529.554 158.34 M544.044 149.289 Q544.044 144.59 542.1 141.928 Q540.179 139.243 536.799 139.243 Q533.42 139.243 531.475 141.928 Q529.554 144.59 529.554 149.289 Q529.554 153.988 531.475 156.673 Q533.42 159.335 536.799 159.335 Q540.179 159.335 542.1 156.673 Q544.044 153.988 544.044 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M577.702 148.201 L577.702 150.284 L558.118 150.284 Q558.396 154.682 560.757 156.997 Q563.142 159.289 567.378 159.289 Q569.831 159.289 572.123 158.687 Q574.438 158.085 576.706 156.881 L576.706 160.909 Q574.415 161.881 572.007 162.391 Q569.6 162.9 567.123 162.9 Q560.919 162.9 557.285 159.289 Q553.674 155.678 553.674 149.52 Q553.674 143.155 557.1 139.428 Q560.549 135.678 566.382 135.678 Q571.614 135.678 574.646 139.057 Q577.702 142.414 577.702 148.201 M573.442 146.951 Q573.396 143.456 571.475 141.372 Q569.577 139.289 566.429 139.289 Q562.864 139.289 560.711 141.303 Q558.581 143.317 558.257 146.974 L573.442 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M603.35 137.298 L603.35 141.28 Q601.544 140.284 599.715 139.798 Q597.91 139.289 596.058 139.289 Q591.915 139.289 589.623 141.928 Q587.331 144.544 587.331 149.289 Q587.331 154.034 589.623 156.673 Q591.915 159.289 596.058 159.289 Q597.91 159.289 599.715 158.803 Q601.544 158.293 603.35 157.298 L603.35 161.233 Q601.567 162.067 599.646 162.483 Q597.748 162.9 595.595 162.9 Q589.739 162.9 586.29 159.219 Q582.841 155.539 582.841 149.289 Q582.841 142.946 586.313 139.312 Q589.808 135.678 595.873 135.678 Q597.84 135.678 599.715 136.095 Q601.59 136.488 603.35 137.298 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M614.97 128.942 L614.97 136.303 L623.743 136.303 L623.743 139.613 L614.97 139.613 L614.97 153.687 Q614.97 156.858 615.826 157.761 Q616.706 158.664 619.368 158.664 L623.743 158.664 L623.743 162.229 L619.368 162.229 Q614.438 162.229 612.563 160.4 Q610.688 158.548 610.688 153.687 L610.688 139.613 L607.563 139.613 L607.563 136.303 L610.688 136.303 L610.688 128.942 L614.97 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M651.521 148.201 L651.521 150.284 L631.938 150.284 Q632.215 154.682 634.576 156.997 Q636.961 159.289 641.197 159.289 Q643.65 159.289 645.942 158.687 Q648.257 158.085 650.525 156.881 L650.525 160.909 Q648.234 161.881 645.826 162.391 Q643.419 162.9 640.942 162.9 Q634.738 162.9 631.104 159.289 Q627.493 155.678 627.493 149.52 Q627.493 143.155 630.919 139.428 Q634.368 135.678 640.201 135.678 Q645.433 135.678 648.465 139.057 Q651.521 142.414 651.521 148.201 M647.262 146.951 Q647.215 143.456 645.294 141.372 Q643.396 139.289 640.248 139.289 Q636.683 139.289 634.53 141.303 Q632.4 143.317 632.076 146.974 L647.262 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M675.572 140.238 L675.572 126.21 L679.831 126.21 L679.831 162.229 L675.572 162.229 L675.572 158.34 Q674.229 160.655 672.169 161.789 Q670.132 162.9 667.261 162.9 Q662.562 162.9 659.599 159.15 Q656.66 155.4 656.66 149.289 Q656.66 143.178 659.599 139.428 Q662.562 135.678 667.261 135.678 Q670.132 135.678 672.169 136.812 Q674.229 137.923 675.572 140.238 M661.058 149.289 Q661.058 153.988 662.979 156.673 Q664.923 159.335 668.303 159.335 Q671.683 159.335 673.627 156.673 Q675.572 153.988 675.572 149.289 Q675.572 144.59 673.627 141.928 Q671.683 139.243 668.303 139.243 Q664.923 139.243 662.979 141.928 Q661.058 144.59 661.058 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M707.794 158.34 L707.794 172.09 L703.511 172.09 L703.511 136.303 L707.794 136.303 L707.794 140.238 Q709.136 137.923 711.173 136.812 Q713.233 135.678 716.081 135.678 Q720.803 135.678 723.743 139.428 Q726.706 143.178 726.706 149.289 Q726.706 155.4 723.743 159.15 Q720.803 162.9 716.081 162.9 Q713.233 162.9 711.173 161.789 Q709.136 160.655 707.794 158.34 M722.284 149.289 Q722.284 144.59 720.34 141.928 Q718.419 139.243 715.039 139.243 Q711.659 139.243 709.715 141.928 Q707.794 144.59 707.794 149.289 Q707.794 153.988 709.715 156.673 Q711.659 159.335 715.039 159.335 Q718.419 159.335 720.34 156.673 Q722.284 153.988 722.284 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M748.789 140.284 Q748.071 139.868 747.215 139.682 Q746.381 139.474 745.363 139.474 Q741.752 139.474 739.807 141.835 Q737.886 144.173 737.886 148.571 L737.886 162.229 L733.604 162.229 L733.604 136.303 L737.886 136.303 L737.886 140.331 Q739.229 137.969 741.381 136.835 Q743.534 135.678 746.613 135.678 Q747.053 135.678 747.585 135.747 Q748.117 135.794 748.766 135.909 L748.789 140.284 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M762.261 139.289 Q758.835 139.289 756.844 141.974 Q754.854 144.636 754.854 149.289 Q754.854 153.942 756.821 156.627 Q758.812 159.289 762.261 159.289 Q765.664 159.289 767.654 156.604 Q769.645 153.918 769.645 149.289 Q769.645 144.682 767.654 141.997 Q765.664 139.289 762.261 139.289 M762.261 135.678 Q767.816 135.678 770.988 139.289 Q774.159 142.9 774.159 149.289 Q774.159 155.655 770.988 159.289 Q767.816 162.9 762.261 162.9 Q756.682 162.9 753.511 159.289 Q750.363 155.655 750.363 149.289 Q750.363 142.9 753.511 139.289 Q756.682 135.678 762.261 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M794.344 126.21 L794.344 129.752 L790.27 129.752 Q787.978 129.752 787.076 130.678 Q786.196 131.604 786.196 134.011 L786.196 136.303 L793.21 136.303 L793.21 139.613 L786.196 139.613 L786.196 162.229 L781.914 162.229 L781.914 139.613 L777.84 139.613 L777.84 136.303 L781.914 136.303 L781.914 134.497 Q781.914 130.169 783.927 128.201 Q785.941 126.21 790.316 126.21 L794.344 126.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M797.909 136.303 L802.168 136.303 L802.168 162.229 L797.909 162.229 L797.909 136.303 M797.909 126.21 L802.168 126.21 L802.168 131.604 L797.909 131.604 L797.909 126.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M815.293 128.942 L815.293 136.303 L824.066 136.303 L824.066 139.613 L815.293 139.613 L815.293 153.687 Q815.293 156.858 816.149 157.761 Q817.029 158.664 819.691 158.664 L824.066 158.664 L824.066 162.229 L819.691 162.229 Q814.761 162.229 812.886 160.4 Q811.011 158.548 811.011 153.687 L811.011 139.613 L807.886 139.613 L807.886 136.303 L811.011 136.303 L811.011 128.942 L815.293 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M854.969 126.257 Q851.867 131.581 850.362 136.789 Q848.858 141.997 848.858 147.344 Q848.858 152.692 850.362 157.946 Q851.89 163.178 854.969 168.479 L851.265 168.479 Q847.793 163.039 846.057 157.784 Q844.344 152.53 844.344 147.344 Q844.344 142.182 846.057 136.951 Q847.77 131.72 851.265 126.257 L854.969 126.257 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M874.783 169.196 L872.469 169.196 L872.445 162.229 Q870.015 162.182 867.584 161.673 Q865.154 161.141 862.7 160.099 L862.7 155.932 Q865.061 157.414 867.469 158.178 Q869.899 158.918 872.469 158.942 L872.469 148.386 Q867.353 147.553 865.015 145.562 Q862.7 143.571 862.7 140.099 Q862.7 136.326 865.223 134.15 Q867.746 131.974 872.469 131.65 L872.469 126.21 L874.783 126.21 L874.783 131.581 Q876.936 131.673 878.95 132.044 Q880.964 132.391 882.885 133.016 L882.885 137.067 Q880.964 136.095 878.927 135.562 Q876.913 135.03 874.783 134.937 L874.783 144.821 Q880.038 145.631 882.515 147.715 Q884.992 149.798 884.992 153.409 Q884.992 157.321 882.353 159.59 Q879.737 161.835 874.783 162.182 L874.783 169.196 M872.469 144.405 L872.469 134.914 Q869.783 135.215 868.371 136.442 Q866.959 137.669 866.959 139.706 Q866.959 141.696 868.256 142.807 Q869.575 143.919 872.469 144.405 M874.783 148.849 L874.783 158.872 Q877.723 158.479 879.205 157.205 Q880.709 155.932 880.709 153.849 Q880.709 151.812 879.274 150.608 Q877.862 149.405 874.783 148.849 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M894.043 127.669 L912.399 127.669 L912.399 131.604 L898.325 131.604 L898.325 140.076 Q899.343 139.729 900.362 139.567 Q901.381 139.382 902.399 139.382 Q908.186 139.382 911.566 142.553 Q914.945 145.724 914.945 151.141 Q914.945 156.719 911.473 159.821 Q908.001 162.9 901.681 162.9 Q899.506 162.9 897.237 162.53 Q894.992 162.159 892.584 161.418 L892.584 156.719 Q894.668 157.854 896.89 158.409 Q899.112 158.965 901.589 158.965 Q905.593 158.965 907.931 156.858 Q910.269 154.752 910.269 151.141 Q910.269 147.53 907.931 145.423 Q905.593 143.317 901.589 143.317 Q899.714 143.317 897.839 143.733 Q895.987 144.15 894.043 145.03 L894.043 127.669 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M934.737 143.085 Q931.589 143.085 929.737 145.238 Q927.908 147.391 927.908 151.141 Q927.908 154.868 929.737 157.043 Q931.589 159.196 934.737 159.196 Q937.885 159.196 939.714 157.043 Q941.565 154.868 941.565 151.141 Q941.565 147.391 939.714 145.238 Q937.885 143.085 934.737 143.085 M944.019 128.433 L944.019 132.692 Q942.26 131.858 940.454 131.419 Q938.672 130.979 936.913 130.979 Q932.283 130.979 929.829 134.104 Q927.399 137.229 927.052 143.548 Q928.417 141.534 930.478 140.469 Q932.538 139.382 935.015 139.382 Q940.223 139.382 943.232 142.553 Q946.265 145.701 946.265 151.141 Q946.265 156.465 943.116 159.682 Q939.968 162.9 934.737 162.9 Q928.741 162.9 925.57 158.317 Q922.399 153.71 922.399 144.983 Q922.399 136.789 926.288 131.928 Q930.177 127.044 936.728 127.044 Q938.487 127.044 940.269 127.391 Q942.075 127.738 944.019 128.433 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M968.487 143.594 Q971.843 144.312 973.718 146.581 Q975.616 148.849 975.616 152.182 Q975.616 157.298 972.098 160.099 Q968.579 162.9 962.098 162.9 Q959.922 162.9 957.607 162.46 Q955.315 162.043 952.862 161.187 L952.862 156.673 Q954.806 157.807 957.121 158.386 Q959.436 158.965 961.959 158.965 Q966.357 158.965 968.649 157.229 Q970.963 155.493 970.963 152.182 Q970.963 149.127 968.811 147.414 Q966.681 145.678 962.862 145.678 L958.834 145.678 L958.834 141.835 L963.047 141.835 Q966.496 141.835 968.325 140.469 Q970.153 139.081 970.153 136.488 Q970.153 133.826 968.255 132.414 Q966.38 130.979 962.862 130.979 Q960.94 130.979 958.741 131.395 Q956.542 131.812 953.903 132.692 L953.903 128.525 Q956.565 127.784 958.88 127.414 Q961.218 127.044 963.278 127.044 Q968.602 127.044 971.704 129.474 Q974.806 131.882 974.806 136.002 Q974.806 138.872 973.163 140.863 Q971.519 142.831 968.487 143.594 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip710)" d="M983.209 126.257 L986.912 126.257 Q990.385 131.72 992.098 136.951 Q993.834 142.182 993.834 147.344 Q993.834 152.53 992.098 157.784 Q990.385 163.039 986.912 168.479 L983.209 168.479 Q986.287 163.178 987.792 157.946 Q989.32 152.692 989.32 147.344 Q989.32 141.997 987.792 136.789 Q986.287 131.581 983.209 126.257 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/applications/two_stage_stochastic/index.html b/previews/PR3562/tutorials/applications/two_stage_stochastic/index.html
index ee14bddf0c8..4a9dc581540 100644
--- a/previews/PR3562/tutorials/applications/two_stage_stochastic/index.html
+++ b/previews/PR3562/tutorials/applications/two_stage_stochastic/index.html
@@ -18,7 +18,7 @@
 d = sort!(rand(D, N));
 Ω = 1:N
 P = fill(1 / N, N);
-StatsPlots.histogram(d; bins = 20, label = &quot;&quot;, xlabel = &quot;Demand&quot;)</code></pre><img src="8f63bbf6.svg" alt="Example block output"/><h2 id="JuMP-model"><a class="docs-heading-anchor" href="#JuMP-model">JuMP model</a><a id="JuMP-model-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-model" title="Permalink"></a></h2><p>The implementation of our two-stage stochastic program in JuMP is:</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
+StatsPlots.histogram(d; bins = 20, label = &quot;&quot;, xlabel = &quot;Demand&quot;)</code></pre><img src="a4f8868e.svg" alt="Example block output"/><h2 id="JuMP-model"><a class="docs-heading-anchor" href="#JuMP-model">JuMP model</a><a id="JuMP-model-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-model" title="Permalink"></a></h2><p>The implementation of our two-stage stochastic program in JuMP is:</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
 set_silent(model)
 @variable(model, x &gt;= 0)
 @variable(model, 0 &lt;= y[ω in Ω] &lt;= d[ω])
@@ -37,37 +37,37 @@
 * Candidate solution (result #1)
   Primal status      : FEASIBLE_POINT
   Dual status        : FEASIBLE_POINT
-  Objective value    : 5.54814e+02
-  Objective bound    : 5.54814e+02
+  Objective value    : 5.62618e+02
+  Objective bound    : 5.62618e+02
   Relative gap       : Inf
-  Dual objective value : 5.54814e+02
+  Dual objective value : 5.62618e+02
 
 * Work counters
-  Solve time (sec)   : 3.44992e-04
+  Solve time (sec)   : 5.21660e-04
   Simplex iterations : 42
   Barrier iterations : 0
   Node count         : -1
-</code></pre><p>The optimal number of pies to make is:</p><pre><code class="language-julia hljs">value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">200.7659142389702</code></pre><p>The distribution of total profit is:</p><pre><code class="language-julia hljs">total_profit = [-2 * value(x) + value(z[ω]) for ω in Ω]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">100-element Vector{Float64}:
- 403.82005744492255
- 411.2937877948948
- 420.3012332570629
- 420.7615562881516
- 425.8970465957061
- 430.2193116748226
- 436.39853108230693
- 442.2622692130742
- 444.3510335311323
- 453.53441461058185
+</code></pre><p>The optimal number of pies to make is:</p><pre><code class="language-julia hljs">value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">204.15598694708746</code></pre><p>The distribution of total profit is:</p><pre><code class="language-julia hljs">total_profit = [-2 * value(x) + value(z[ω]) for ω in Ω]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">100-element Vector{Float64}:
+ 394.1463843292481
+ 401.05980478842866
+ 406.1726644646086
+ 421.3215132637321
+ 427.67333529720327
+ 430.39640473509576
+ 442.19361139887053
+ 448.1715309159646
+ 452.33694864726226
+ 455.8737999523461
    ⋮
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104
- 602.2977427169104</code></pre><p>Let&#39;s plot it:</p><pre><code class="language-julia hljs">&quot;&quot;&quot;
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624
+ 612.4679608412624</code></pre><p>Let&#39;s plot it:</p><pre><code class="language-julia hljs">&quot;&quot;&quot;
     bin_distribution(x::Vector{Float64}, N::Int)
 
 A helper function that discretizes `x` into bins of width `N`.
@@ -88,7 +88,7 @@
     label = &quot;Expected profit (\$$(round(Int, μ)))&quot;,
     linewidth = 3,
 )
-plot</code></pre><img src="7f7c697a.svg" alt="Example block output"/><h2 id="Risk-measures"><a class="docs-heading-anchor" href="#Risk-measures">Risk measures</a><a id="Risk-measures-1"></a><a class="docs-heading-anchor-permalink" href="#Risk-measures" title="Permalink"></a></h2><p>A risk measure is a function which maps a random variable to a real number. Common risk measures include the mean (expectation), median, mode, and maximum. We need a risk measure to convert the distribution of second stage costs into a single number that can be optimized.</p><p>Our model currently uses the expectation risk measure, but others are possible too. One popular risk measure is the conditional value at risk (CVaR).</p><p>CVaR has a parameter <span>$\gamma$</span>, and it computes the expectation of the worst <span>$\gamma$</span> fraction of outcomes.</p><p>If we are maximizing, so that small outcomes are bad, the definition of CVaR is:</p><p class="math-container">\[CVaR_{\gamma}[Z] = \max\limits_{\xi} \;\; \xi - \frac{1}{\gamma}\mathbb{E}_\omega\left[(\xi - Z)_+\right]\]</p><p>which can be formulated as the linear program:</p><p class="math-container">\[\begin{aligned}
+plot</code></pre><img src="bea383e2.svg" alt="Example block output"/><h2 id="Risk-measures"><a class="docs-heading-anchor" href="#Risk-measures">Risk measures</a><a id="Risk-measures-1"></a><a class="docs-heading-anchor-permalink" href="#Risk-measures" title="Permalink"></a></h2><p>A risk measure is a function which maps a random variable to a real number. Common risk measures include the mean (expectation), median, mode, and maximum. We need a risk measure to convert the distribution of second stage costs into a single number that can be optimized.</p><p>Our model currently uses the expectation risk measure, but others are possible too. One popular risk measure is the conditional value at risk (CVaR).</p><p>CVaR has a parameter <span>$\gamma$</span>, and it computes the expectation of the worst <span>$\gamma$</span> fraction of outcomes.</p><p>If we are maximizing, so that small outcomes are bad, the definition of CVaR is:</p><p class="math-container">\[CVaR_{\gamma}[Z] = \max\limits_{\xi} \;\; \xi - \frac{1}{\gamma}\mathbb{E}_\omega\left[(\xi - Z)_+\right]\]</p><p>which can be formulated as the linear program:</p><p class="math-container">\[\begin{aligned}
 CVaR_{\gamma}[Z] = \max\limits_{\xi, z_\omega} \;\; &amp; \xi - \frac{1}{\gamma}\sum P_\omega z_\omega\\
  &amp; z_\omega \ge \xi - Z_\omega &amp; \quad \forall \omega \\
  &amp; z_\omega \ge 0 &amp; \quad \forall \omega.
@@ -103,7 +103,7 @@
     @objective(model, Max, ξ - 1 / γ * sum(P[i] * z[i] for i in 1:N))
     optimize!(model)
     return objective_value(model)
-end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">CVaR (generic function with 1 method)</code></pre><p>When <code>γ</code> is <code>1.0</code>, we compute the mean of the profit:</p><pre><code class="language-julia hljs">cvar_10 = CVaR(total_profit, P; γ = 1.0)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">554.8141066343148</code></pre><pre><code class="language-julia hljs">Statistics.mean(total_profit)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">554.8141066343147</code></pre><p>As <code>γ</code> approaches <code>0.0</code>, we compute the worst-case (minimum) profit:</p><pre><code class="language-julia hljs">cvar_00 = CVaR(total_profit, P; γ = 0.0001)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">403.82005744492255</code></pre><pre><code class="language-julia hljs">minimum(total_profit)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">403.82005744492255</code></pre><p>By varying <code>γ</code> between <code>0</code> and <code>1</code> we can compute some trade-off of these two extremes:</p><pre><code class="language-julia hljs">cvar_05 = CVaR(total_profit, P; γ = 0.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">508.11913007979473</code></pre><p>Let&#39;s plot these outcomes on our distribution:</p><pre><code class="language-julia hljs">plot = StatsPlots.histogram(
+end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">CVaR (generic function with 1 method)</code></pre><p>When <code>γ</code> is <code>1.0</code>, we compute the mean of the profit:</p><pre><code class="language-julia hljs">cvar_10 = CVaR(total_profit, P; γ = 1.0)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">562.6181327944481</code></pre><pre><code class="language-julia hljs">Statistics.mean(total_profit)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">562.6181327944482</code></pre><p>As <code>γ</code> approaches <code>0.0</code>, we compute the worst-case (minimum) profit:</p><pre><code class="language-julia hljs">cvar_00 = CVaR(total_profit, P; γ = 0.0001)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">394.1463843292481</code></pre><pre><code class="language-julia hljs">minimum(total_profit)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">394.1463843292481</code></pre><p>By varying <code>γ</code> between <code>0</code> and <code>1</code> we can compute some trade-off of these two extremes:</p><pre><code class="language-julia hljs">cvar_05 = CVaR(total_profit, P; γ = 0.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">514.7707702985415</code></pre><p>Let&#39;s plot these outcomes on our distribution:</p><pre><code class="language-julia hljs">plot = StatsPlots.histogram(
     total_profit;
     bins = bin_distribution(total_profit, 25),
     label = &quot;&quot;,
@@ -116,7 +116,7 @@
     label = [&quot;γ = 1.0&quot; &quot;γ = 0.5&quot; &quot;γ = 0.0&quot;],
     linewidth = 3,
 )
-plot</code></pre><img src="7add9a3e.svg" alt="Example block output"/><h2 id="Risk-averse-sample-average-approximation"><a class="docs-heading-anchor" href="#Risk-averse-sample-average-approximation">Risk averse sample average approximation</a><a id="Risk-averse-sample-average-approximation-1"></a><a class="docs-heading-anchor-permalink" href="#Risk-averse-sample-average-approximation" title="Permalink"></a></h2><p>Because CVaR can be formulated as a linear program, we can form a risk averse sample average approximation model by combining the two formulations:</p><pre><code class="language-julia hljs">γ = 0.4
+plot</code></pre><img src="3b3c353b.svg" alt="Example block output"/><h2 id="Risk-averse-sample-average-approximation"><a class="docs-heading-anchor" href="#Risk-averse-sample-average-approximation">Risk averse sample average approximation</a><a id="Risk-averse-sample-average-approximation-1"></a><a class="docs-heading-anchor-permalink" href="#Risk-averse-sample-average-approximation" title="Permalink"></a></h2><p>Because CVaR can be formulated as a linear program, we can form a risk averse sample average approximation model by combining the two formulations:</p><pre><code class="language-julia hljs">γ = 0.4
 model = Model(HiGHS.Optimizer)
 set_silent(model)
 @variable(model, x &gt;= 0)
@@ -127,7 +127,7 @@
 @variable(model, z[ω in Ω] &gt;= 0)
 @constraint(model, [ω in Ω], z[ω] &gt;= ξ - Z[ω])
 @objective(model, Max, -2x + ξ - 1 / γ * sum(P[ω] * z[ω] for ω in Ω))
-optimize!(model)</code></pre><p>When <span>$\gamma = 0.4$</span>, the optimal number of pies to bake is:</p><pre><code class="language-julia hljs">value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">183.3527169192272</code></pre><p>The distribution of total profit is:</p><pre><code class="language-julia hljs">risk_averse_total_profit = [value(-2x + Z[ω]) for ω in Ω]
+optimize!(model)</code></pre><p>When <span>$\gamma = 0.4$</span>, the optimal number of pies to bake is:</p><pre><code class="language-julia hljs">value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">186.92196235493182</code></pre><p>The distribution of total profit is:</p><pre><code class="language-julia hljs">risk_averse_total_profit = [value(-2x + Z[ω]) for ω in Ω]
 bins = bin_distribution([total_profit; risk_averse_total_profit], 25)
 plot = StatsPlots.histogram(total_profit; label = &quot;Expectation&quot;, bins = bins)
 StatsPlots.histogram!(
@@ -137,4 +137,4 @@
     bins = bins,
     alpha = 0.5,
 )
-plot</code></pre><img src="60f511d7.svg" alt="Example block output"/><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Try solving this problem for different numbers of samples and different distributions.</li><li>Refactor the example to avoid hard-coding the costs. What happens to the solution if the cost of disposing unsold pies increases?</li><li>Plot the optimal number of pies to make for different values of the risk aversion parameter <span>$\gamma$</span>. What is the relationship?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../web_app/">« Serving web apps</a><a class="docs-footer-nextpage" href="../../../manual/models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+plot</code></pre><img src="5becafc9.svg" alt="Example block output"/><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Try solving this problem for different numbers of samples and different distributions.</li><li>Refactor the example to avoid hard-coding the costs. What happens to the solution if the cost of disposing unsold pies increases?</li><li>Plot the optimal number of pies to make for different values of the risk aversion parameter <span>$\gamma$</span>. What is the relationship?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../web_app/">« Serving web apps</a><a class="docs-footer-nextpage" href="../../../manual/models/">Models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/applications/web_app/index.html b/previews/PR3562/tutorials/applications/web_app/index.html
index 9b49df3f8da..6172bea9d59 100644
--- a/previews/PR3562/tutorials/applications/web_app/index.html
+++ b/previews/PR3562/tutorials/applications/web_app/index.html
@@ -88,4 +88,4 @@
   &quot;status&quot; =&gt; &quot;failure&quot;
   &quot;reason&quot; =&gt; &quot;missing lower_bound param&quot;</code></pre><p>If we don&#39;t send a <code>lower_bound</code> that is a number, we get:</p><pre><code class="language-julia hljs">send_request(Dict(&quot;lower_bound&quot; =&gt; &quot;1.2&quot;))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Dict{String, Any} with 2 entries:
   &quot;status&quot; =&gt; &quot;failure&quot;
-  &quot;reason&quot; =&gt; &quot;lower_bound is not a number&quot;</code></pre><p>Finally, we can shutdown our HTTP server:</p><pre><code class="language-julia hljs">close(server)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"><span class="sgr36"><span class="sgr1">[ Info: </span></span>Server on 127.0.0.1:8080 closing</code></pre><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>For more complicated examples relating to HTTP servers, consult the <a href="https://juliaweb.github.io/HTTP.jl/stable/">HTTP.jl documentation</a>.</p><p>To see how you can integrate this with a larger JuMP model, read <a href="../../getting_started/design_patterns_for_larger_models/#Design-patterns-for-larger-models">Design patterns for larger models</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../optimal_power_flow/">« Optimal power flow</a><a class="docs-footer-nextpage" href="../two_stage_stochastic/">Two-stage stochastic programs »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  &quot;reason&quot; =&gt; &quot;lower_bound is not a number&quot;</code></pre><p>Finally, we can shutdown our HTTP server:</p><pre><code class="language-julia hljs">close(server)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"><span class="sgr36"><span class="sgr1">[ Info: </span></span>Server on 127.0.0.1:8080 closing</code></pre><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>For more complicated examples relating to HTTP servers, consult the <a href="https://juliaweb.github.io/HTTP.jl/stable/">HTTP.jl documentation</a>.</p><p>To see how you can integrate this with a larger JuMP model, read <a href="../../getting_started/design_patterns_for_larger_models/#Design-patterns-for-larger-models">Design patterns for larger models</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../optimal_power_flow/">« Optimal power flow</a><a class="docs-footer-nextpage" href="../two_stage_stochastic/">Two-stage stochastic programs »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/arbitrary_precision/index.html b/previews/PR3562/tutorials/conic/arbitrary_precision/index.html
index 5f1b40b27cf..13ddb7937c0 100644
--- a/previews/PR3562/tutorials/conic/arbitrary_precision/index.html
+++ b/previews/PR3562/tutorials/conic/arbitrary_precision/index.html
@@ -35,7 +35,7 @@
   Dual objective value : -6.42857e-01
 
 * Work counters
-  Solve time (sec)   : 1.70336e-03
+  Solve time (sec)   : 2.63522e-03
   Barrier iterations : 5
 </code></pre><p>The value of each decision variable is a <code>BigFloat</code>:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{BigFloat}:
  0.4285714246558161076147072906813123533593766450416896337912086518811186790735016
@@ -77,4 +77,4 @@
 * Work counters
 </code></pre><p>The optimal values are given in exact rational arithmetic:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Rational{BigInt}}:
  1//6
- 2//3</code></pre><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">5//6</code></pre><pre><code class="language-julia hljs">value(c2)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">13//6</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../dualization/">« Dualization</a><a class="docs-footer-nextpage" href="../logistic_regression/">Logistic regression »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 2//3</code></pre><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">5//6</code></pre><pre><code class="language-julia hljs">value(c2)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">13//6</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../dualization/">« Dualization</a><a class="docs-footer-nextpage" href="../logistic_regression/">Logistic regression »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/dualization/index.html b/previews/PR3562/tutorials/conic/dualization/index.html
index c9a559826ba..117ef295703 100644
--- a/previews/PR3562/tutorials/conic/dualization/index.html
+++ b/previews/PR3562/tutorials/conic/dualization/index.html
@@ -51,19 +51,19 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 1.65e+01  1.60e-01  5.09e+01 -2.91e+01  1.00e-01  6.37e-05
-    50| 1.74e-08  2.70e-10  4.88e-08 -4.00e+00  1.00e-01  1.15e-04
+     0| 1.65e+01  1.60e-01  5.09e+01 -2.91e+01  1.00e-01  8.09e-05
+    50| 1.74e-08  2.70e-10  4.88e-08 -4.00e+00  1.00e-01  1.52e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 1.16e-04s = setup: 4.02e-05s + solve: 7.54e-05s
-	 lin-sys: 7.79e-06s, cones: 4.17e-05s, accel: 2.36e-06s
+timings: total: 1.53e-04s = setup: 5.15e-05s + solve: 1.02e-04s
+	 lin-sys: 9.30e-06s, cones: 6.08e-05s, accel: 3.30e-06s
 ------------------------------------------------------------------
 objective = -4.000000
 ------------------------------------------------------------------</code></pre><p>(There are five rows in the constraint matrix because SCS expects problems in geometric conic form, and so JuMP has reformulated the <code>X, PSD</code> variable constraint into the affine constraint <code>X .+ 0 in PSDCone()</code>.)</p><p>The solution we obtain is:</p><pre><code class="language-julia hljs">value.(X)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2×2 Matrix{Float64}:
   1.0  -1.0
  -1.0   1.0</code></pre><pre><code class="language-julia hljs">dual.(primal_c)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 1.9999999997299085
- 1.9999999997299085</code></pre><pre><code class="language-julia hljs">objective_value(model_primal)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">3.9999999506359716</code></pre><p>When we solve <code>model_dual</code> with <code>SCS.Optimizer</code>, SCS reports two variables (<code>variables n: 2</code>), three rows in the constraint matrix (<code>constraints m: 3</code>), and two non-zeros in the matrix (<code>nnz(A): 2</code>):</p><pre><code class="language-julia hljs">set_optimizer(model_dual, SCS.Optimizer)
+ 1.9999999997299098
+ 1.9999999997299098</code></pre><pre><code class="language-julia hljs">objective_value(model_primal)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">3.9999999506359964</code></pre><p>When we solve <code>model_dual</code> with <code>SCS.Optimizer</code>, SCS reports two variables (<code>variables n: 2</code>), three rows in the constraint matrix (<code>constraints m: 3</code>), and two non-zeros in the matrix (<code>nnz(A): 2</code>):</p><pre><code class="language-julia hljs">set_optimizer(model_dual, SCS.Optimizer)
 optimize!(model_dual)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">------------------------------------------------------------------
 	       SCS v3.2.3 - Splitting Conic Solver
 	(c) Brendan O&#39;Donoghue, Stanford University, 2012
@@ -79,19 +79,19 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 1.23e+01  1.00e+00  2.73e+01 -9.03e+00  1.00e-01  5.69e-05
-    50| 1.13e-07  1.05e-09  3.23e-07  4.00e+00  1.00e-01  1.10e-04
+     0| 1.23e+01  1.00e+00  2.73e+01 -9.03e+00  1.00e-01  8.72e-05
+    50| 1.13e-07  1.05e-09  3.23e-07  4.00e+00  1.00e-01  1.75e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 1.10e-04s = setup: 4.03e-05s + solve: 7.00e-05s
-	 lin-sys: 6.26e-06s, cones: 3.94e-05s, accel: 2.27e-06s
+timings: total: 1.76e-04s = setup: 6.19e-05s + solve: 1.14e-04s
+	 lin-sys: 9.10e-06s, cones: 6.71e-05s, accel: 3.10e-06s
 ------------------------------------------------------------------
 objective = 4.000000
 ------------------------------------------------------------------</code></pre><p>and the solution we obtain is:</p><pre><code class="language-julia hljs">dual.(dual_c)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2×2 Matrix{Float64}:
   1.0  -1.0
  -1.0   1.0</code></pre><pre><code class="language-julia hljs">value.(y)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 2.000000159272681
- 2.000000159272681</code></pre><pre><code class="language-julia hljs">objective_value(model_dual)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">4.000000318545362</code></pre><p>This particular problem is small enough that it isn&#39;t meaningful to compare the solve times, but in general, we should expect <code>model_dual</code> to solve faster than <code>model_primal</code> because it contains fewer variables and constraints. The difference is particularly noticeable on large-scale optimization problems.</p><h2 id="dual_optimizer"><a class="docs-heading-anchor" href="#dual_optimizer"><code>dual_optimizer</code></a><a id="dual_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#dual_optimizer" title="Permalink"></a></h2><p>Manually deriving the conic dual is difficult and error-prone. The package <a href="../../../packages/Dualization/#Dualization.jl">Dualization.jl</a> provides the <code>Dualization.dual_optimizer</code> meta-solver, which wraps any MathOptInterface-compatible solver in an interface that automatically formulates and solves the dual of an input problem.</p><p>To demonstrate, we use <code>Dualization.dual_optimizer</code> to solve <code>model_primal</code>:</p><pre><code class="language-julia hljs">set_optimizer(model_primal, Dualization.dual_optimizer(SCS.Optimizer))
+ 2.0000001592726786
+ 2.0000001592726786</code></pre><pre><code class="language-julia hljs">objective_value(model_dual)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">4.000000318545357</code></pre><p>This particular problem is small enough that it isn&#39;t meaningful to compare the solve times, but in general, we should expect <code>model_dual</code> to solve faster than <code>model_primal</code> because it contains fewer variables and constraints. The difference is particularly noticeable on large-scale optimization problems.</p><h2 id="dual_optimizer"><a class="docs-heading-anchor" href="#dual_optimizer"><code>dual_optimizer</code></a><a id="dual_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#dual_optimizer" title="Permalink"></a></h2><p>Manually deriving the conic dual is difficult and error-prone. The package <a href="../../../packages/Dualization/#Dualization.jl">Dualization.jl</a> provides the <code>Dualization.dual_optimizer</code> meta-solver, which wraps any MathOptInterface-compatible solver in an interface that automatically formulates and solves the dual of an input problem.</p><p>To demonstrate, we use <code>Dualization.dual_optimizer</code> to solve <code>model_primal</code>:</p><pre><code class="language-julia hljs">set_optimizer(model_primal, Dualization.dual_optimizer(SCS.Optimizer))
 optimize!(model_primal)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">------------------------------------------------------------------
 	       SCS v3.2.3 - Splitting Conic Solver
 	(c) Brendan O&#39;Donoghue, Stanford University, 2012
@@ -107,19 +107,19 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 1.23e+01  1.00e+00  2.73e+01 -9.03e+00  1.00e-01  4.42e-05
-    50| 1.13e-07  1.05e-09  3.23e-07  4.00e+00  1.00e-01  9.54e-05
+     0| 1.23e+01  1.00e+00  2.73e+01 -9.03e+00  1.00e-01  8.84e-05
+    50| 1.13e-07  1.05e-09  3.23e-07  4.00e+00  1.00e-01  1.76e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 9.60e-05s = setup: 3.29e-05s + solve: 6.30e-05s
-	 lin-sys: 5.94e-06s, cones: 3.47e-05s, accel: 2.16e-06s
+timings: total: 1.77e-04s = setup: 6.38e-05s + solve: 1.13e-04s
+	 lin-sys: 8.80e-06s, cones: 5.78e-05s, accel: 1.39e-05s
 ------------------------------------------------------------------
 objective = 4.000000
 ------------------------------------------------------------------</code></pre><p>The performance is the same as if we solved <code>model_dual</code>, and the correct solution is returned to <code>X</code>:</p><pre><code class="language-julia hljs">value.(X)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2×2 Matrix{Float64}:
   1.0  -1.0
  -1.0   1.0</code></pre><pre><code class="language-julia hljs">dual.(primal_c)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 2.000000159272681
- 2.000000159272681</code></pre><p>Moreover, if we use <code>dual_optimizer</code> on <code>model_dual</code>, then we get the same performance as if we had solved <code>model_primal</code>:</p><pre><code class="language-julia hljs">set_optimizer(model_dual, Dualization.dual_optimizer(SCS.Optimizer))
+ 2.0000001592726786
+ 2.0000001592726786</code></pre><p>Moreover, if we use <code>dual_optimizer</code> on <code>model_dual</code>, then we get the same performance as if we had solved <code>model_primal</code>:</p><pre><code class="language-julia hljs">set_optimizer(model_dual, Dualization.dual_optimizer(SCS.Optimizer))
 optimize!(model_dual)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">------------------------------------------------------------------
 	       SCS v3.2.3 - Splitting Conic Solver
 	(c) Brendan O&#39;Donoghue, Stanford University, 2012
@@ -136,19 +136,19 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 1.65e+01  1.60e-01  5.09e+01 -2.91e+01  1.00e-01  6.11e-05
-    50| 1.74e-08  2.70e-10  4.88e-08 -4.00e+00  1.00e-01  1.27e-04
+     0| 1.65e+01  1.60e-01  5.09e+01 -2.91e+01  1.00e-01  1.15e-04
+    50| 1.74e-08  2.70e-10  4.88e-08 -4.00e+00  1.00e-01  1.94e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 1.28e-04s = setup: 4.17e-05s + solve: 8.65e-05s
-	 lin-sys: 7.70e-06s, cones: 5.31e-05s, accel: 2.38e-06s
+timings: total: 1.95e-04s = setup: 7.31e-05s + solve: 1.22e-04s
+	 lin-sys: 1.07e-05s, cones: 6.42e-05s, accel: 3.00e-06s
 ------------------------------------------------------------------
 objective = -4.000000
 ------------------------------------------------------------------</code></pre><pre><code class="language-julia hljs">dual.(dual_c)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2×2 Matrix{Float64}:
   1.0  -1.0
  -1.0   1.0</code></pre><pre><code class="language-julia hljs">value.(y)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 1.9999999997299085
- 1.9999999997299085</code></pre><h2 id="A-mixed-example"><a class="docs-heading-anchor" href="#A-mixed-example">A mixed example</a><a id="A-mixed-example-1"></a><a class="docs-heading-anchor-permalink" href="#A-mixed-example" title="Permalink"></a></h2><p>The <a href="../simple_examples/#Maximum-cut-via-SDP">Maximum cut via SDP</a> example is nicely defined because the primal is in standard conic form and the dual is in geometric conic form. However, many practical models contain a mix of the two formulations. One example is <a href="../simple_examples/#The-minimum-distortion-problem">The minimum distortion problem</a>:</p><pre><code class="language-julia hljs">D = [0 1 1 1; 1 0 2 2; 1 2 0 2; 1 2 2 0]
+ 1.9999999997299098
+ 1.9999999997299098</code></pre><h2 id="A-mixed-example"><a class="docs-heading-anchor" href="#A-mixed-example">A mixed example</a><a id="A-mixed-example-1"></a><a class="docs-heading-anchor-permalink" href="#A-mixed-example" title="Permalink"></a></h2><p>The <a href="../simple_examples/#Maximum-cut-via-SDP">Maximum cut via SDP</a> example is nicely defined because the primal is in standard conic form and the dual is in geometric conic form. However, many practical models contain a mix of the two formulations. One example is <a href="../simple_examples/#The-minimum-distortion-problem">The minimum distortion problem</a>:</p><pre><code class="language-julia hljs">D = [0 1 1 1; 1 0 2 2; 1 2 0 2; 1 2 2 0]
 model = Model()
 @variable(model, c²)
 @variable(model, Q[1:4, 1:4], PSD)
@@ -176,12 +176,12 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 4.73e+00  1.00e+00  2.92e+00  1.23e+00  1.00e-01  1.01e-04
-   150| 1.01e-04  3.07e-05  6.08e-05  1.33e+00  1.00e-01  4.90e-04
+     0| 4.73e+00  1.00e+00  2.92e+00  1.23e+00  1.00e-01  1.67e-04
+   150| 1.01e-04  3.07e-05  6.08e-05  1.33e+00  1.00e-01  9.51e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 4.91e-04s = setup: 6.83e-05s + solve: 4.22e-04s
-	 lin-sys: 6.95e-05s, cones: 2.61e-04s, accel: 2.82e-05s
+timings: total: 9.53e-04s = setup: 1.21e-04s + solve: 8.32e-04s
+	 lin-sys: 1.30e-04s, cones: 4.65e-04s, accel: 1.36e-04s
 ------------------------------------------------------------------
 objective = 1.333363
 ------------------------------------------------------------------</code></pre><pre><code class="language-julia hljs">set_optimizer(model, Dualization.dual_optimizer(SCS.Optimizer))
@@ -202,12 +202,12 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 3.71e+01  1.48e+00  2.23e+02 -1.13e+02  1.00e-01  7.73e-05
-   150| 1.57e-04  2.28e-05  2.08e-04 -1.33e+00  1.00e-01  4.94e-04
+     0| 3.71e+01  1.48e+00  2.23e+02 -1.13e+02  1.00e-01  1.57e-04
+   150| 1.57e-04  2.28e-05  2.08e-04 -1.33e+00  1.00e-01  8.36e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 4.95e-04s = setup: 5.63e-05s + solve: 4.38e-04s
-	 lin-sys: 7.23e-05s, cones: 2.82e-04s, accel: 1.89e-05s
+timings: total: 8.37e-04s = setup: 1.08e-04s + solve: 7.29e-04s
+	 lin-sys: 1.34e-04s, cones: 4.56e-04s, accel: 4.42e-05s
 ------------------------------------------------------------------
 objective = -1.333460
-------------------------------------------------------------------</code></pre><p>For this problem, SCS reports that the primal has <code>variables n: 11, constraints m: 24</code> and that the dual has <code>variables n: 14, constraints m: 24</code>. Therefore, we should probably use the primal formulation because it has fewer variables and the same number of constraints.</p><h2 id="When-to-use-dual_optimizer"><a class="docs-heading-anchor" href="#When-to-use-dual_optimizer">When to use <code>dual_optimizer</code></a><a id="When-to-use-dual_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#When-to-use-dual_optimizer" title="Permalink"></a></h2><p>Because it can make the problem larger or smaller, depending on the problem and the choice of solver, there is no definitive rule on when you should use <code>dual_optimizer</code>. However, you should try <code>dual_optimizer</code> if your conic optimization problem takes a long time to solve, or if you need to repeatedly solve similarly structured problems with different data. In some cases solving the dual instead of the primal can make a large difference.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../simple_examples/">« Simple semidefinite programming examples</a><a class="docs-footer-nextpage" href="../arbitrary_precision/">Arbitrary precision arithmetic »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+------------------------------------------------------------------</code></pre><p>For this problem, SCS reports that the primal has <code>variables n: 11, constraints m: 24</code> and that the dual has <code>variables n: 14, constraints m: 24</code>. Therefore, we should probably use the primal formulation because it has fewer variables and the same number of constraints.</p><h2 id="When-to-use-dual_optimizer"><a class="docs-heading-anchor" href="#When-to-use-dual_optimizer">When to use <code>dual_optimizer</code></a><a id="When-to-use-dual_optimizer-1"></a><a class="docs-heading-anchor-permalink" href="#When-to-use-dual_optimizer" title="Permalink"></a></h2><p>Because it can make the problem larger or smaller, depending on the problem and the choice of solver, there is no definitive rule on when you should use <code>dual_optimizer</code>. However, you should try <code>dual_optimizer</code> if your conic optimization problem takes a long time to solve, or if you need to repeatedly solve similarly structured problems with different data. In some cases solving the dual instead of the primal can make a large difference.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../simple_examples/">« Simple semidefinite programming examples</a><a class="docs-footer-nextpage" href="../arbitrary_precision/">Arbitrary precision arithmetic »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/ellipse_approx/053884ea.svg b/previews/PR3562/tutorials/conic/ellipse_approx/618bca67.svg
similarity index 58%
rename from previews/PR3562/tutorials/conic/ellipse_approx/053884ea.svg
rename to previews/PR3562/tutorials/conic/ellipse_approx/618bca67.svg
index 01c9d478692..89719faf3ac 100644
--- a/previews/PR3562/tutorials/conic/ellipse_approx/053884ea.svg
+++ b/previews/PR3562/tutorials/conic/ellipse_approx/618bca67.svg
@@ -1,2451 +1,2451 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="600" viewBox="0 0 2400 2400">
 <defs>
-  <clipPath id="clip130">
+  <clipPath id="clip950">
     <rect x="0" y="0" width="2400" height="2400"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip130)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip950)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip131">
+  <clipPath id="clip951">
     <rect x="480" y="240" width="1681" height="1681"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip130)" d="M178.867 2261.17 L2352.76 2261.17 L2352.76 47.2441 L178.867 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip950)" d="M178.867 2261.17 L2352.76 2261.17 L2352.76 47.2441 L178.867 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip132">
+  <clipPath id="clip952">
     <rect x="178" y="47" width="2175" height="2215"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="327.236,2261.17 327.236,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="640.095,2261.17 640.095,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="952.953,2261.17 952.953,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1265.81,2261.17 1265.81,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1578.67,2261.17 1578.67,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1891.53,2261.17 1891.53,47.2441 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2204.39,2261.17 2204.39,47.2441 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 2352.76,2261.17 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="327.236,2261.17 327.236,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="640.095,2261.17 640.095,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="952.953,2261.17 952.953,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1265.81,2261.17 1265.81,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1578.67,2261.17 1578.67,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1891.53,2261.17 1891.53,2242.27 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2204.39,2261.17 2204.39,2242.27 "/>
-<path clip-path="url(#clip130)" d="M281.299 2306.74 L310.975 2306.74 L310.975 2310.67 L281.299 2310.67 L281.299 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M335.234 2304.93 Q338.59 2305.65 340.465 2307.92 Q342.363 2310.19 342.363 2313.52 Q342.363 2318.64 338.845 2321.44 Q335.326 2324.24 328.845 2324.24 Q326.669 2324.24 324.354 2323.8 Q322.063 2323.38 319.609 2322.53 L319.609 2318.01 Q321.553 2319.15 323.868 2319.73 Q326.183 2320.3 328.706 2320.3 Q333.104 2320.3 335.396 2318.57 Q337.711 2316.83 337.711 2313.52 Q337.711 2310.47 335.558 2308.75 Q333.428 2307.02 329.609 2307.02 L325.581 2307.02 L325.581 2303.17 L329.794 2303.17 Q333.243 2303.17 335.072 2301.81 Q336.9 2300.42 336.9 2297.83 Q336.9 2295.17 335.002 2293.75 Q333.127 2292.32 329.609 2292.32 Q327.688 2292.32 325.488 2292.73 Q323.289 2293.15 320.651 2294.03 L320.651 2289.86 Q323.313 2289.12 325.627 2288.75 Q327.965 2288.38 330.025 2288.38 Q335.35 2288.38 338.451 2290.81 Q341.553 2293.22 341.553 2297.34 Q341.553 2300.21 339.91 2302.2 Q338.266 2304.17 335.234 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M361.229 2292.09 Q357.618 2292.09 355.789 2295.65 Q353.984 2299.19 353.984 2306.32 Q353.984 2313.43 355.789 2316.99 Q357.618 2320.54 361.229 2320.54 Q364.863 2320.54 366.669 2316.99 Q368.497 2313.43 368.497 2306.32 Q368.497 2299.19 366.669 2295.65 Q364.863 2292.09 361.229 2292.09 M361.229 2288.38 Q367.039 2288.38 370.095 2292.99 Q373.173 2297.57 373.173 2306.32 Q373.173 2315.05 370.095 2319.66 Q367.039 2324.24 361.229 2324.24 Q355.419 2324.24 352.34 2319.66 Q349.285 2315.05 349.285 2306.32 Q349.285 2297.57 352.34 2292.99 Q355.419 2288.38 361.229 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M594.157 2306.74 L623.833 2306.74 L623.833 2310.67 L594.157 2310.67 L594.157 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M637.953 2319.63 L654.273 2319.63 L654.273 2323.57 L632.328 2323.57 L632.328 2319.63 Q634.99 2316.88 639.574 2312.25 Q644.18 2307.6 645.361 2306.25 Q647.606 2303.73 648.486 2301.99 Q649.388 2300.23 649.388 2298.54 Q649.388 2295.79 647.444 2294.05 Q645.523 2292.32 642.421 2292.32 Q640.222 2292.32 637.768 2293.08 Q635.338 2293.85 632.56 2295.4 L632.56 2290.67 Q635.384 2289.54 637.838 2288.96 Q640.291 2288.38 642.328 2288.38 Q647.699 2288.38 650.893 2291.07 Q654.087 2293.75 654.087 2298.24 Q654.087 2300.37 653.277 2302.29 Q652.49 2304.19 650.384 2306.79 Q649.805 2307.46 646.703 2310.67 Q643.601 2313.87 637.953 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M674.087 2292.09 Q670.476 2292.09 668.648 2295.65 Q666.842 2299.19 666.842 2306.32 Q666.842 2313.43 668.648 2316.99 Q670.476 2320.54 674.087 2320.54 Q677.722 2320.54 679.527 2316.99 Q681.356 2313.43 681.356 2306.32 Q681.356 2299.19 679.527 2295.65 Q677.722 2292.09 674.087 2292.09 M674.087 2288.38 Q679.898 2288.38 682.953 2292.99 Q686.032 2297.57 686.032 2306.32 Q686.032 2315.05 682.953 2319.66 Q679.898 2324.24 674.087 2324.24 Q668.277 2324.24 665.199 2319.66 Q662.143 2315.05 662.143 2306.32 Q662.143 2297.57 665.199 2292.99 Q668.277 2288.38 674.087 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M907.016 2306.74 L936.691 2306.74 L936.691 2310.67 L907.016 2310.67 L907.016 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M947.594 2319.63 L955.233 2319.63 L955.233 2293.27 L946.923 2294.93 L946.923 2290.67 L955.187 2289.01 L959.863 2289.01 L959.863 2319.63 L967.501 2319.63 L967.501 2323.57 L947.594 2323.57 L947.594 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M986.946 2292.09 Q983.335 2292.09 981.506 2295.65 Q979.7 2299.19 979.7 2306.32 Q979.7 2313.43 981.506 2316.99 Q983.335 2320.54 986.946 2320.54 Q990.58 2320.54 992.386 2316.99 Q994.214 2313.43 994.214 2306.32 Q994.214 2299.19 992.386 2295.65 Q990.58 2292.09 986.946 2292.09 M986.946 2288.38 Q992.756 2288.38 995.811 2292.99 Q998.89 2297.57 998.89 2306.32 Q998.89 2315.05 995.811 2319.66 Q992.756 2324.24 986.946 2324.24 Q981.136 2324.24 978.057 2319.66 Q975.001 2315.05 975.001 2306.32 Q975.001 2297.57 978.057 2292.99 Q981.136 2288.38 986.946 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1265.81 2292.09 Q1262.2 2292.09 1260.37 2295.65 Q1258.57 2299.19 1258.57 2306.32 Q1258.57 2313.43 1260.37 2316.99 Q1262.2 2320.54 1265.81 2320.54 Q1269.45 2320.54 1271.25 2316.99 Q1273.08 2313.43 1273.08 2306.32 Q1273.08 2299.19 1271.25 2295.65 Q1269.45 2292.09 1265.81 2292.09 M1265.81 2288.38 Q1271.62 2288.38 1274.68 2292.99 Q1277.76 2297.57 1277.76 2306.32 Q1277.76 2315.05 1274.68 2319.66 Q1271.62 2324.24 1265.81 2324.24 Q1260 2324.24 1256.92 2319.66 Q1253.87 2315.05 1253.87 2306.32 Q1253.87 2297.57 1256.92 2292.99 Q1260 2288.38 1265.81 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1553.36 2319.63 L1561 2319.63 L1561 2293.27 L1552.69 2294.93 L1552.69 2290.67 L1560.95 2289.01 L1565.63 2289.01 L1565.63 2319.63 L1573.26 2319.63 L1573.26 2323.57 L1553.36 2323.57 L1553.36 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1592.71 2292.09 Q1589.1 2292.09 1587.27 2295.65 Q1585.46 2299.19 1585.46 2306.32 Q1585.46 2313.43 1587.27 2316.99 Q1589.1 2320.54 1592.71 2320.54 Q1596.34 2320.54 1598.15 2316.99 Q1599.98 2313.43 1599.98 2306.32 Q1599.98 2299.19 1598.15 2295.65 Q1596.34 2292.09 1592.71 2292.09 M1592.71 2288.38 Q1598.52 2288.38 1601.57 2292.99 Q1604.65 2297.57 1604.65 2306.32 Q1604.65 2315.05 1601.57 2319.66 Q1598.52 2324.24 1592.71 2324.24 Q1586.9 2324.24 1583.82 2319.66 Q1580.76 2315.05 1580.76 2306.32 Q1580.76 2297.57 1583.82 2292.99 Q1586.9 2288.38 1592.71 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1870.3 2319.63 L1886.62 2319.63 L1886.62 2323.57 L1864.68 2323.57 L1864.68 2319.63 Q1867.34 2316.88 1871.92 2312.25 Q1876.53 2307.6 1877.71 2306.25 Q1879.95 2303.73 1880.83 2301.99 Q1881.74 2300.23 1881.74 2298.54 Q1881.74 2295.79 1879.79 2294.05 Q1877.87 2292.32 1874.77 2292.32 Q1872.57 2292.32 1870.12 2293.08 Q1867.69 2293.85 1864.91 2295.4 L1864.91 2290.67 Q1867.73 2289.54 1870.19 2288.96 Q1872.64 2288.38 1874.68 2288.38 Q1880.05 2288.38 1883.24 2291.07 Q1886.44 2293.75 1886.44 2298.24 Q1886.44 2300.37 1885.63 2302.29 Q1884.84 2304.19 1882.73 2306.79 Q1882.15 2307.46 1879.05 2310.67 Q1875.95 2313.87 1870.3 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1906.44 2292.09 Q1902.82 2292.09 1901 2295.65 Q1899.19 2299.19 1899.19 2306.32 Q1899.19 2313.43 1901 2316.99 Q1902.82 2320.54 1906.44 2320.54 Q1910.07 2320.54 1911.88 2316.99 Q1913.7 2313.43 1913.7 2306.32 Q1913.7 2299.19 1911.88 2295.65 Q1910.07 2292.09 1906.44 2292.09 M1906.44 2288.38 Q1912.25 2288.38 1915.3 2292.99 Q1918.38 2297.57 1918.38 2306.32 Q1918.38 2315.05 1915.3 2319.66 Q1912.25 2324.24 1906.44 2324.24 Q1900.63 2324.24 1897.55 2319.66 Q1894.49 2315.05 1894.49 2306.32 Q1894.49 2297.57 1897.55 2292.99 Q1900.63 2288.38 1906.44 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M2193.23 2304.93 Q2196.59 2305.65 2198.46 2307.92 Q2200.36 2310.19 2200.36 2313.52 Q2200.36 2318.64 2196.84 2321.44 Q2193.32 2324.24 2186.84 2324.24 Q2184.66 2324.24 2182.35 2323.8 Q2180.06 2323.38 2177.6 2322.53 L2177.6 2318.01 Q2179.55 2319.15 2181.86 2319.73 Q2184.18 2320.3 2186.7 2320.3 Q2191.1 2320.3 2193.39 2318.57 Q2195.71 2316.83 2195.71 2313.52 Q2195.71 2310.47 2193.55 2308.75 Q2191.42 2307.02 2187.6 2307.02 L2183.58 2307.02 L2183.58 2303.17 L2187.79 2303.17 Q2191.24 2303.17 2193.07 2301.81 Q2194.9 2300.42 2194.9 2297.83 Q2194.9 2295.17 2193 2293.75 Q2191.12 2292.32 2187.6 2292.32 Q2185.68 2292.32 2183.48 2292.73 Q2181.28 2293.15 2178.65 2294.03 L2178.65 2289.86 Q2181.31 2289.12 2183.62 2288.75 Q2185.96 2288.38 2188.02 2288.38 Q2193.34 2288.38 2196.45 2290.81 Q2199.55 2293.22 2199.55 2297.34 Q2199.55 2300.21 2197.9 2302.2 Q2196.26 2304.17 2193.23 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M2219.22 2292.09 Q2215.61 2292.09 2213.78 2295.65 Q2211.98 2299.19 2211.98 2306.32 Q2211.98 2313.43 2213.78 2316.99 Q2215.61 2320.54 2219.22 2320.54 Q2222.86 2320.54 2224.66 2316.99 Q2226.49 2313.43 2226.49 2306.32 Q2226.49 2299.19 2224.66 2295.65 Q2222.86 2292.09 2219.22 2292.09 M2219.22 2288.38 Q2225.03 2288.38 2228.09 2292.99 Q2231.17 2297.57 2231.17 2306.32 Q2231.17 2315.05 2228.09 2319.66 Q2225.03 2324.24 2219.22 2324.24 Q2213.41 2324.24 2210.34 2319.66 Q2207.28 2315.05 2207.28 2306.32 Q2207.28 2297.57 2210.34 2292.99 Q2213.41 2288.38 2219.22 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,2110.07 2352.76,2110.07 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1791.45 2352.76,1791.45 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1472.83 2352.76,1472.83 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1154.21 2352.76,1154.21 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,835.586 2352.76,835.586 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,516.966 2352.76,516.966 "/>
-<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,198.346 2352.76,198.346 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 178.867,47.2441 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2110.07 197.764,2110.07 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1791.45 197.764,1791.45 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1472.83 197.764,1472.83 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1154.21 197.764,1154.21 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,835.586 197.764,835.586 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,516.966 197.764,516.966 "/>
-<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,198.346 197.764,198.346 "/>
-<path clip-path="url(#clip130)" d="M50.9921 2110.52 L80.6679 2110.52 L80.6679 2114.45 L50.9921 2114.45 L50.9921 2110.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M104.927 2108.71 Q108.283 2109.43 110.158 2111.7 Q112.057 2113.97 112.057 2117.3 Q112.057 2122.42 108.538 2125.22 Q105.02 2128.02 98.5382 2128.02 Q96.3623 2128.02 94.0475 2127.58 Q91.7558 2127.16 89.3021 2126.3 L89.3021 2121.79 Q91.2465 2122.92 93.5613 2123.5 Q95.8761 2124.08 98.3993 2124.08 Q102.797 2124.08 105.089 2122.35 Q107.404 2120.61 107.404 2117.3 Q107.404 2114.24 105.251 2112.53 Q103.121 2110.8 99.3021 2110.8 L95.2743 2110.8 L95.2743 2106.95 L99.4872 2106.95 Q102.936 2106.95 104.765 2105.59 Q106.594 2104.2 106.594 2101.61 Q106.594 2098.94 104.696 2097.53 Q102.821 2096.1 99.3021 2096.1 Q97.3808 2096.1 95.1817 2096.51 Q92.9826 2096.93 90.3438 2097.81 L90.3438 2093.64 Q93.0058 2092.9 95.3206 2092.53 Q97.6585 2092.16 99.7187 2092.16 Q105.043 2092.16 108.145 2094.59 Q111.246 2097 111.246 2101.12 Q111.246 2103.99 109.603 2105.98 Q107.959 2107.95 104.927 2108.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 2095.86 Q127.311 2095.86 125.482 2099.43 Q123.677 2102.97 123.677 2110.1 Q123.677 2117.21 125.482 2120.77 Q127.311 2124.31 130.922 2124.31 Q134.556 2124.31 136.362 2120.77 Q138.191 2117.21 138.191 2110.1 Q138.191 2102.97 136.362 2099.43 Q134.556 2095.86 130.922 2095.86 M130.922 2092.16 Q136.732 2092.16 139.788 2096.77 Q142.867 2101.35 142.867 2110.1 Q142.867 2118.83 139.788 2123.43 Q136.732 2128.02 130.922 2128.02 Q125.112 2128.02 122.033 2123.43 Q118.978 2118.83 118.978 2110.1 Q118.978 2101.35 122.033 2096.77 Q125.112 2092.16 130.922 2092.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M50.9921 1791.9 L80.6679 1791.9 L80.6679 1795.83 L50.9921 1795.83 L50.9921 1791.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M94.7882 1804.79 L111.108 1804.79 L111.108 1808.73 L89.1632 1808.73 L89.1632 1804.79 Q91.8252 1802.04 96.4085 1797.41 Q101.015 1792.75 102.196 1791.41 Q104.441 1788.89 105.321 1787.15 Q106.223 1785.39 106.223 1783.7 Q106.223 1780.95 104.279 1779.21 Q102.358 1777.48 99.2558 1777.48 Q97.0567 1777.48 94.603 1778.24 Q92.1725 1779 89.3947 1780.55 L89.3947 1775.83 Q92.2188 1774.7 94.6724 1774.12 Q97.1261 1773.54 99.1632 1773.54 Q104.534 1773.54 107.728 1776.23 Q110.922 1778.91 110.922 1783.4 Q110.922 1785.53 110.112 1787.45 Q109.325 1789.35 107.219 1791.94 Q106.64 1792.61 103.538 1795.83 Q100.436 1799.03 94.7882 1804.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 1777.24 Q127.311 1777.24 125.482 1780.81 Q123.677 1784.35 123.677 1791.48 Q123.677 1798.59 125.482 1802.15 Q127.311 1805.69 130.922 1805.69 Q134.556 1805.69 136.362 1802.15 Q138.191 1798.59 138.191 1791.48 Q138.191 1784.35 136.362 1780.81 Q134.556 1777.24 130.922 1777.24 M130.922 1773.54 Q136.732 1773.54 139.788 1778.15 Q142.867 1782.73 142.867 1791.48 Q142.867 1800.21 139.788 1804.81 Q136.732 1809.4 130.922 1809.4 Q125.112 1809.4 122.033 1804.81 Q118.978 1800.21 118.978 1791.48 Q118.978 1782.73 122.033 1778.15 Q125.112 1773.54 130.922 1773.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M50.9921 1473.28 L80.6679 1473.28 L80.6679 1477.21 L50.9921 1477.21 L50.9921 1473.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M91.5706 1486.17 L99.2095 1486.17 L99.2095 1459.81 L90.8993 1461.47 L90.8993 1457.21 L99.1632 1455.55 L103.839 1455.55 L103.839 1486.17 L111.478 1486.17 L111.478 1490.11 L91.5706 1490.11 L91.5706 1486.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 1458.62 Q127.311 1458.62 125.482 1462.19 Q123.677 1465.73 123.677 1472.86 Q123.677 1479.97 125.482 1483.53 Q127.311 1487.07 130.922 1487.07 Q134.556 1487.07 136.362 1483.53 Q138.191 1479.97 138.191 1472.86 Q138.191 1465.73 136.362 1462.19 Q134.556 1458.62 130.922 1458.62 M130.922 1454.92 Q136.732 1454.92 139.788 1459.53 Q142.867 1464.11 142.867 1472.86 Q142.867 1481.59 139.788 1486.19 Q136.732 1490.78 130.922 1490.78 Q125.112 1490.78 122.033 1486.19 Q118.978 1481.59 118.978 1472.86 Q118.978 1464.11 122.033 1459.53 Q125.112 1454.92 130.922 1454.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 1140 Q127.311 1140 125.482 1143.57 Q123.677 1147.11 123.677 1154.24 Q123.677 1161.35 125.482 1164.91 Q127.311 1168.45 130.922 1168.45 Q134.556 1168.45 136.362 1164.91 Q138.191 1161.35 138.191 1154.24 Q138.191 1147.11 136.362 1143.57 Q134.556 1140 130.922 1140 M130.922 1136.3 Q136.732 1136.3 139.788 1140.91 Q142.867 1145.49 142.867 1154.24 Q142.867 1162.97 139.788 1167.57 Q136.732 1172.16 130.922 1172.16 Q125.112 1172.16 122.033 1167.57 Q118.978 1162.97 118.978 1154.24 Q118.978 1145.49 122.033 1140.91 Q125.112 1136.3 130.922 1136.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M91.5706 848.931 L99.2095 848.931 L99.2095 822.565 L90.8993 824.232 L90.8993 819.973 L99.1632 818.306 L103.839 818.306 L103.839 848.931 L111.478 848.931 L111.478 852.866 L91.5706 852.866 L91.5706 848.931 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 821.385 Q127.311 821.385 125.482 824.949 Q123.677 828.491 123.677 835.621 Q123.677 842.727 125.482 846.292 Q127.311 849.834 130.922 849.834 Q134.556 849.834 136.362 846.292 Q138.191 842.727 138.191 835.621 Q138.191 828.491 136.362 824.949 Q134.556 821.385 130.922 821.385 M130.922 817.681 Q136.732 817.681 139.788 822.287 Q142.867 826.871 142.867 835.621 Q142.867 844.348 139.788 848.954 Q136.732 853.537 130.922 853.537 Q125.112 853.537 122.033 848.954 Q118.978 844.348 118.978 835.621 Q118.978 826.871 122.033 822.287 Q125.112 817.681 130.922 817.681 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M94.7882 530.311 L111.108 530.311 L111.108 534.246 L89.1632 534.246 L89.1632 530.311 Q91.8252 527.556 96.4085 522.927 Q101.015 518.274 102.196 516.931 Q104.441 514.408 105.321 512.672 Q106.223 510.913 106.223 509.223 Q106.223 506.468 104.279 504.732 Q102.358 502.996 99.2558 502.996 Q97.0567 502.996 94.603 503.76 Q92.1725 504.524 89.3947 506.075 L89.3947 501.353 Q92.2188 500.218 94.6724 499.64 Q97.1261 499.061 99.1632 499.061 Q104.534 499.061 107.728 501.746 Q110.922 504.431 110.922 508.922 Q110.922 511.052 110.112 512.973 Q109.325 514.871 107.219 517.464 Q106.64 518.135 103.538 521.353 Q100.436 524.547 94.7882 530.311 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 502.765 Q127.311 502.765 125.482 506.329 Q123.677 509.871 123.677 517.001 Q123.677 524.107 125.482 527.672 Q127.311 531.214 130.922 531.214 Q134.556 531.214 136.362 527.672 Q138.191 524.107 138.191 517.001 Q138.191 509.871 136.362 506.329 Q134.556 502.765 130.922 502.765 M130.922 499.061 Q136.732 499.061 139.788 503.667 Q142.867 508.251 142.867 517.001 Q142.867 525.727 139.788 530.334 Q136.732 534.917 130.922 534.917 Q125.112 534.917 122.033 530.334 Q118.978 525.727 118.978 517.001 Q118.978 508.251 122.033 503.667 Q125.112 499.061 130.922 499.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M104.927 196.992 Q108.283 197.709 110.158 199.978 Q112.057 202.246 112.057 205.58 Q112.057 210.695 108.538 213.496 Q105.02 216.297 98.5382 216.297 Q96.3623 216.297 94.0475 215.857 Q91.7558 215.441 89.3021 214.584 L89.3021 210.07 Q91.2465 211.205 93.5613 211.783 Q95.8761 212.362 98.3993 212.362 Q102.797 212.362 105.089 210.626 Q107.404 208.89 107.404 205.58 Q107.404 202.524 105.251 200.811 Q103.121 199.075 99.3021 199.075 L95.2743 199.075 L95.2743 195.233 L99.4872 195.233 Q102.936 195.233 104.765 193.867 Q106.594 192.478 106.594 189.885 Q106.594 187.223 104.696 185.811 Q102.821 184.376 99.3021 184.376 Q97.3808 184.376 95.1817 184.793 Q92.9826 185.209 90.3438 186.089 L90.3438 181.922 Q93.0058 181.182 95.3206 180.811 Q97.6585 180.441 99.7187 180.441 Q105.043 180.441 108.145 182.872 Q111.246 185.279 111.246 189.399 Q111.246 192.27 109.603 194.26 Q107.959 196.228 104.927 196.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M130.922 184.145 Q127.311 184.145 125.482 187.709 Q123.677 191.251 123.677 198.381 Q123.677 205.487 125.482 209.052 Q127.311 212.594 130.922 212.594 Q134.556 212.594 136.362 209.052 Q138.191 205.487 138.191 198.381 Q138.191 191.251 136.362 187.709 Q134.556 184.145 130.922 184.145 M130.922 180.441 Q136.732 180.441 139.788 185.047 Q142.867 189.631 142.867 198.381 Q142.867 207.107 139.788 211.714 Q136.732 216.297 130.922 216.297 Q125.112 216.297 122.033 211.714 Q118.978 207.107 118.978 198.381 Q118.978 189.631 122.033 185.047 Q125.112 180.441 130.922 180.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><line clip-path="url(#clip132)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1926.4" y1="856.923" x2="1910.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1926.4" y1="856.923" x2="1910.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1926.4" y1="856.923" x2="1942.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1926.4" y1="856.923" x2="1942.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="912.69" y1="1348.83" x2="896.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="912.69" y1="1348.83" x2="896.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="912.69" y1="1348.83" x2="928.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="912.69" y1="1348.83" x2="928.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="625.575" y1="1469.54" x2="609.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="625.575" y1="1469.54" x2="609.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="625.575" y1="1469.54" x2="641.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="625.575" y1="1469.54" x2="641.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.36" y1="1138.56" x2="990.356" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.36" y1="1138.56" x2="990.356" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="785.24" y1="1436.86" x2="769.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="785.24" y1="1436.86" x2="769.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="785.24" y1="1436.86" x2="801.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="785.24" y1="1436.86" x2="801.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.289" y1="1394.75" x2="761.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.289" y1="1394.75" x2="761.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.289" y1="1394.75" x2="793.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.289" y1="1394.75" x2="793.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.41" y1="1014.15" x2="1369.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.41" y1="1014.15" x2="1369.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.41" y1="1014.15" x2="1401.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.41" y1="1014.15" x2="1401.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1612.46" y1="913.718" x2="1596.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1612.46" y1="913.718" x2="1596.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1612.46" y1="913.718" x2="1628.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1612.46" y1="913.718" x2="1628.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="947.412" y1="1370.03" x2="931.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="947.412" y1="1370.03" x2="931.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="947.412" y1="1370.03" x2="963.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="947.412" y1="1370.03" x2="963.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.497" y1="1483.85" x2="672.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.497" y1="1483.85" x2="672.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.497" y1="1483.85" x2="704.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.497" y1="1483.85" x2="704.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.95" y1="869.353" x2="1529.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.95" y1="869.353" x2="1529.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.95" y1="869.353" x2="1561.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.95" y1="869.353" x2="1561.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="771.564" y1="1431.34" x2="755.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="771.564" y1="1431.34" x2="755.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="771.564" y1="1431.34" x2="787.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="771.564" y1="1431.34" x2="787.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="918.674" y1="1363.31" x2="902.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="918.674" y1="1363.31" x2="902.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="918.674" y1="1363.31" x2="934.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="918.674" y1="1363.31" x2="934.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1707.08" y1="826.86" x2="1691.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1707.08" y1="826.86" x2="1691.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1707.08" y1="826.86" x2="1723.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1707.08" y1="826.86" x2="1723.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.63" y1="875.887" x2="1561.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.63" y1="875.887" x2="1561.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.63" y1="875.887" x2="1593.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.63" y1="875.887" x2="1593.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="894.72" y1="1142.51" x2="878.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="894.72" y1="1142.51" x2="878.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="894.72" y1="1142.51" x2="910.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="894.72" y1="1142.51" x2="910.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1580.8" y1="990.127" x2="1564.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1580.8" y1="990.127" x2="1564.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1580.8" y1="990.127" x2="1596.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1580.8" y1="990.127" x2="1596.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="852.02" y1="1283.64" x2="836.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="852.02" y1="1283.64" x2="836.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="852.02" y1="1283.64" x2="868.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="852.02" y1="1283.64" x2="868.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.48" y1="962.993" x2="1589.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.48" y1="962.993" x2="1589.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.48" y1="962.993" x2="1621.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.48" y1="962.993" x2="1621.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1172.56" y1="1009.01" x2="1156.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1172.56" y1="1009.01" x2="1156.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1172.56" y1="1009.01" x2="1188.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1172.56" y1="1009.01" x2="1188.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.588" y1="1415.15" x2="672.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.588" y1="1415.15" x2="672.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.588" y1="1415.15" x2="704.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="688.588" y1="1415.15" x2="704.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="991.105" y1="1402.23" x2="975.105" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="991.105" y1="1402.23" x2="975.105" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="991.105" y1="1402.23" x2="1007.11" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="991.105" y1="1402.23" x2="1007.11" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1811.99" y1="919.23" x2="1795.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1811.99" y1="919.23" x2="1795.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1811.99" y1="919.23" x2="1827.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1811.99" y1="919.23" x2="1827.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.971" y1="1350.77" x2="829.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.971" y1="1350.77" x2="829.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.971" y1="1350.77" x2="861.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.971" y1="1350.77" x2="861.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="886.864" y1="1368.41" x2="870.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="886.864" y1="1368.41" x2="870.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="886.864" y1="1368.41" x2="902.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="886.864" y1="1368.41" x2="902.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.009" y1="1388.03" x2="761.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.009" y1="1388.03" x2="761.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.009" y1="1388.03" x2="793.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="777.009" y1="1388.03" x2="793.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.23" y1="998.156" x2="1277.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.23" y1="998.156" x2="1277.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.23" y1="998.156" x2="1309.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.23" y1="998.156" x2="1309.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="885.196" y1="1405.14" x2="869.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="885.196" y1="1405.14" x2="869.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="885.196" y1="1405.14" x2="901.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="885.196" y1="1405.14" x2="901.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.55" y1="853.956" x2="1441.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.55" y1="853.956" x2="1441.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.55" y1="853.956" x2="1473.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.55" y1="853.956" x2="1473.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="949.006" y1="1440.98" x2="933.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="949.006" y1="1440.98" x2="933.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="949.006" y1="1440.98" x2="965.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="949.006" y1="1440.98" x2="965.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.411" y1="1427.9" x2="829.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.411" y1="1427.9" x2="829.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.411" y1="1427.9" x2="861.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="845.411" y1="1427.9" x2="861.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.57" y1="868.321" x2="1530.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.57" y1="868.321" x2="1530.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.57" y1="868.321" x2="1562.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.57" y1="868.321" x2="1562.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.43" y1="1342.43" x2="834.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.43" y1="1342.43" x2="834.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.43" y1="1342.43" x2="866.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.43" y1="1342.43" x2="866.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.5" y1="829.178" x2="1603.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.5" y1="829.178" x2="1603.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.5" y1="829.178" x2="1635.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1619.5" y1="829.178" x2="1635.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1847.7" y1="804.547" x2="1831.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1847.7" y1="804.547" x2="1831.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1847.7" y1="804.547" x2="1863.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1847.7" y1="804.547" x2="1863.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="992.659" y1="1251.94" x2="976.659" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="992.659" y1="1251.94" x2="976.659" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="992.659" y1="1251.94" x2="1008.66" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="992.659" y1="1251.94" x2="1008.66" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1641.07" y1="860.861" x2="1625.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1641.07" y1="860.861" x2="1625.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1641.07" y1="860.861" x2="1657.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1641.07" y1="860.861" x2="1657.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1301.11" y1="987.604" x2="1285.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1301.11" y1="987.604" x2="1285.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1301.11" y1="987.604" x2="1317.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1301.11" y1="987.604" x2="1317.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1673.06" y1="901.944" x2="1657.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1673.06" y1="901.944" x2="1657.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1673.06" y1="901.944" x2="1689.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1673.06" y1="901.944" x2="1689.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.89" y1="961.872" x2="1595.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.89" y1="961.872" x2="1595.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.89" y1="961.872" x2="1627.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.89" y1="961.872" x2="1627.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="782.176" y1="1390.69" x2="766.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="782.176" y1="1390.69" x2="766.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="782.176" y1="1390.69" x2="798.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="782.176" y1="1390.69" x2="798.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.77" y1="921.537" x2="1659.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.77" y1="921.537" x2="1659.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.77" y1="921.537" x2="1691.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.77" y1="921.537" x2="1691.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="816.257" y1="1346.85" x2="800.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="816.257" y1="1346.85" x2="800.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="816.257" y1="1346.85" x2="832.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="816.257" y1="1346.85" x2="832.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="724.712" y1="1453.02" x2="708.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="724.712" y1="1453.02" x2="708.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="724.712" y1="1453.02" x2="740.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="724.712" y1="1453.02" x2="740.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.34" y1="968.47" x2="1659.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.34" y1="968.47" x2="1659.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.34" y1="968.47" x2="1691.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1675.34" y1="968.47" x2="1691.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.396" y1="1435.49" x2="858.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.396" y1="1435.49" x2="858.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.396" y1="1435.49" x2="890.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.396" y1="1435.49" x2="890.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1330.6" y1="1013.95" x2="1314.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1330.6" y1="1013.95" x2="1314.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1330.6" y1="1013.95" x2="1346.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1330.6" y1="1013.95" x2="1346.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1626.12" y1="969.424" x2="1610.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1626.12" y1="969.424" x2="1610.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1626.12" y1="969.424" x2="1642.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1626.12" y1="969.424" x2="1642.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="933.271" y1="1353.8" x2="917.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="933.271" y1="1353.8" x2="917.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="933.271" y1="1353.8" x2="949.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="933.271" y1="1353.8" x2="949.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="906.639" y1="1451.01" x2="890.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="906.639" y1="1451.01" x2="890.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="906.639" y1="1451.01" x2="922.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="906.639" y1="1451.01" x2="922.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="958.996" y1="1406.94" x2="942.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="958.996" y1="1406.94" x2="942.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="958.996" y1="1406.94" x2="974.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="958.996" y1="1406.94" x2="974.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.277" y1="1312.14" x2="858.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.277" y1="1312.14" x2="858.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.277" y1="1312.14" x2="890.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="874.277" y1="1312.14" x2="890.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.691" y1="1424.89" x2="911.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.691" y1="1424.89" x2="911.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.691" y1="1424.89" x2="943.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.691" y1="1424.89" x2="943.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="538.801" y1="1550.28" x2="522.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="538.801" y1="1550.28" x2="522.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="538.801" y1="1550.28" x2="554.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="538.801" y1="1550.28" x2="554.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="869.402" y1="1512" x2="853.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="869.402" y1="1512" x2="853.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="869.402" y1="1512" x2="885.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="869.402" y1="1512" x2="885.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.688" y1="1677.8" x2="553.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.688" y1="1677.8" x2="553.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.688" y1="1677.8" x2="585.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.688" y1="1677.8" x2="585.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="915.573" y1="1448.5" x2="899.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="915.573" y1="1448.5" x2="899.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="915.573" y1="1448.5" x2="931.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="915.573" y1="1448.5" x2="931.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1520.69" y1="868.39" x2="1504.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1520.69" y1="868.39" x2="1504.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1520.69" y1="868.39" x2="1536.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1520.69" y1="868.39" x2="1536.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.79" y1="1007.52" x2="1446.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.79" y1="1007.52" x2="1446.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.79" y1="1007.52" x2="1478.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1462.79" y1="1007.52" x2="1478.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.71" y1="986.213" x2="1427.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.71" y1="986.213" x2="1427.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.71" y1="986.213" x2="1459.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.71" y1="986.213" x2="1459.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="831.587" y1="1424.87" x2="815.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="831.587" y1="1424.87" x2="815.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="831.587" y1="1424.87" x2="847.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="831.587" y1="1424.87" x2="847.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.93" y1="1006.01" x2="1593.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.93" y1="1006.01" x2="1593.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.93" y1="1006.01" x2="1625.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.93" y1="1006.01" x2="1625.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1733.98" y1="723.632" x2="1717.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1733.98" y1="723.632" x2="1717.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1733.98" y1="723.632" x2="1749.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1733.98" y1="723.632" x2="1749.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.936" y1="1351.49" x2="834.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.936" y1="1351.49" x2="834.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.936" y1="1351.49" x2="866.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="850.936" y1="1351.49" x2="866.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.92" y1="896.324" x2="1752.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.92" y1="896.324" x2="1752.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.92" y1="896.324" x2="1784.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.92" y1="896.324" x2="1784.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.86" y1="1264.5" x2="984.861" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.86" y1="1264.5" x2="984.861" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.48" y1="973.842" x2="1509.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.48" y1="973.842" x2="1509.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.48" y1="973.842" x2="1541.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.48" y1="973.842" x2="1541.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="814.5" y1="1271.95" x2="798.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="814.5" y1="1271.95" x2="798.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="814.5" y1="1271.95" x2="830.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="814.5" y1="1271.95" x2="830.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1834.69" y1="763.508" x2="1818.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1834.69" y1="763.508" x2="1818.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1834.69" y1="763.508" x2="1850.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1834.69" y1="763.508" x2="1850.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.04" y1="1278.88" x2="990.041" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.04" y1="1278.88" x2="990.041" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1424" y1="1023.62" x2="1408" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1424" y1="1023.62" x2="1408" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1424" y1="1023.62" x2="1440" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1424" y1="1023.62" x2="1440" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1618.21" y1="979.458" x2="1602.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1618.21" y1="979.458" x2="1602.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1618.21" y1="979.458" x2="1634.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1618.21" y1="979.458" x2="1634.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1005.39" y1="1284.51" x2="989.392" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1005.39" y1="1284.51" x2="989.392" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.98" y1="894.258" x2="1593.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.98" y1="894.258" x2="1593.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.98" y1="894.258" x2="1625.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1609.98" y1="894.258" x2="1625.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1597.07" y1="924.794" x2="1581.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1597.07" y1="924.794" x2="1581.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1597.07" y1="924.794" x2="1613.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1597.07" y1="924.794" x2="1613.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="882.48" y1="1367.74" x2="866.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="882.48" y1="1367.74" x2="866.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="882.48" y1="1367.74" x2="898.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="882.48" y1="1367.74" x2="898.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1527.95" y1="988.132" x2="1511.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1527.95" y1="988.132" x2="1511.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1527.95" y1="988.132" x2="1543.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1527.95" y1="988.132" x2="1543.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1766.37" y1="756.534" x2="1750.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1766.37" y1="756.534" x2="1750.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1766.37" y1="756.534" x2="1782.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1766.37" y1="756.534" x2="1782.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1422.2" y1="999.716" x2="1406.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1422.2" y1="999.716" x2="1406.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1422.2" y1="999.716" x2="1438.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1422.2" y1="999.716" x2="1438.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1381.71" y1="975.726" x2="1365.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1381.71" y1="975.726" x2="1365.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1381.71" y1="975.726" x2="1397.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1381.71" y1="975.726" x2="1397.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.037" y1="1422.12" x2="859.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.037" y1="1422.12" x2="859.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.037" y1="1422.12" x2="891.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.037" y1="1422.12" x2="891.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="898.259" y1="1319.39" x2="882.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="898.259" y1="1319.39" x2="882.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="898.259" y1="1319.39" x2="914.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="898.259" y1="1319.39" x2="914.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.19" y1="815.289" x2="1746.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.19" y1="815.289" x2="1746.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.19" y1="815.289" x2="1778.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.19" y1="815.289" x2="1778.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="843.736" y1="1421.71" x2="827.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="843.736" y1="1421.71" x2="827.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="843.736" y1="1421.71" x2="859.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="843.736" y1="1421.71" x2="859.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1399.57" y1="992.596" x2="1383.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1399.57" y1="992.596" x2="1383.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1399.57" y1="992.596" x2="1415.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1399.57" y1="992.596" x2="1415.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="712.506" y1="1466.98" x2="696.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="712.506" y1="1466.98" x2="696.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="712.506" y1="1466.98" x2="728.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="712.506" y1="1466.98" x2="728.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="996.845" y1="1298.95" x2="980.845" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="996.845" y1="1298.95" x2="980.845" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="996.845" y1="1298.95" x2="1012.85" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="996.845" y1="1298.95" x2="1012.85" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="764.769" y1="1428.81" x2="748.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="764.769" y1="1428.81" x2="748.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="764.769" y1="1428.81" x2="780.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="764.769" y1="1428.81" x2="780.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1481.71" y1="979.429" x2="1465.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1481.71" y1="979.429" x2="1465.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1481.71" y1="979.429" x2="1497.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1481.71" y1="979.429" x2="1497.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="794.022" y1="1392.87" x2="778.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="794.022" y1="1392.87" x2="778.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="794.022" y1="1392.87" x2="810.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="794.022" y1="1392.87" x2="810.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="989.1" y1="1361.68" x2="973.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="989.1" y1="1361.68" x2="973.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="989.1" y1="1361.68" x2="1005.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="989.1" y1="1361.68" x2="1005.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="658.223" y1="1594.72" x2="642.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="658.223" y1="1594.72" x2="642.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="658.223" y1="1594.72" x2="674.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="658.223" y1="1594.72" x2="674.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.13" y1="922.6" x2="1595.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.13" y1="922.6" x2="1595.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.13" y1="922.6" x2="1627.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1611.13" y1="922.6" x2="1627.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="968.976" y1="1423.08" x2="952.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="968.976" y1="1423.08" x2="952.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="968.976" y1="1423.08" x2="984.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="968.976" y1="1423.08" x2="984.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1451.83" y1="926.379" x2="1435.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1451.83" y1="926.379" x2="1435.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1451.83" y1="926.379" x2="1467.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1451.83" y1="926.379" x2="1467.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="908.822" y1="1392.95" x2="892.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="908.822" y1="1392.95" x2="892.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="908.822" y1="1392.95" x2="924.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="908.822" y1="1392.95" x2="924.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.57" y1="1011.37" x2="1484.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.57" y1="1011.37" x2="1484.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.57" y1="1011.37" x2="1516.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1500.57" y1="1011.37" x2="1516.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1002.47" y1="1234.46" x2="986.471" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1002.47" y1="1234.46" x2="986.471" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.95" y1="1013.72" x2="1442.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.95" y1="1013.72" x2="1442.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.95" y1="1013.72" x2="1474.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1458.95" y1="1013.72" x2="1474.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="892.734" y1="1460.48" x2="876.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="892.734" y1="1460.48" x2="876.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="892.734" y1="1460.48" x2="908.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="892.734" y1="1460.48" x2="908.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.79" y1="924.706" x2="1529.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.79" y1="924.706" x2="1529.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.79" y1="924.706" x2="1561.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.79" y1="924.706" x2="1561.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1003.69" y1="1262.65" x2="987.692" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1003.69" y1="1262.65" x2="987.692" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="862.589" y1="1433.49" x2="846.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="862.589" y1="1433.49" x2="846.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="862.589" y1="1433.49" x2="878.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="862.589" y1="1433.49" x2="878.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="522.504" y1="1512.98" x2="506.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="522.504" y1="1512.98" x2="506.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="522.504" y1="1512.98" x2="538.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="522.504" y1="1512.98" x2="538.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.208" y1="1309.37" x2="956.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.208" y1="1309.37" x2="956.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.208" y1="1309.37" x2="988.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.208" y1="1309.37" x2="988.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.36" y1="1054" x2="1160.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.36" y1="1054" x2="1160.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.36" y1="1054" x2="1192.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.36" y1="1054" x2="1192.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.451" y1="1406.48" x2="728.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.451" y1="1406.48" x2="728.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.451" y1="1406.48" x2="760.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.451" y1="1406.48" x2="760.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1012.88" y1="1293.81" x2="996.879" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1012.88" y1="1293.81" x2="996.879" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="277.68" y1="1755.15" x2="261.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="277.68" y1="1755.15" x2="261.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="277.68" y1="1755.15" x2="293.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="277.68" y1="1755.15" x2="293.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1498.04" y1="1011.64" x2="1482.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1498.04" y1="1011.64" x2="1482.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1498.04" y1="1011.64" x2="1514.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1498.04" y1="1011.64" x2="1514.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.19" y1="842.118" x2="1764.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.19" y1="842.118" x2="1764.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.19" y1="842.118" x2="1796.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.19" y1="842.118" x2="1796.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="855.143" y1="1461.56" x2="839.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="855.143" y1="1461.56" x2="839.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="855.143" y1="1461.56" x2="871.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="855.143" y1="1461.56" x2="871.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="941.026" y1="1398.61" x2="925.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="941.026" y1="1398.61" x2="925.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="941.026" y1="1398.61" x2="957.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="941.026" y1="1398.61" x2="957.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1491.9" y1="862.19" x2="1475.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1491.9" y1="862.19" x2="1475.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1491.9" y1="862.19" x2="1507.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1491.9" y1="862.19" x2="1507.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1614.72" y1="834.262" x2="1598.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1614.72" y1="834.262" x2="1598.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1614.72" y1="834.262" x2="1630.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1614.72" y1="834.262" x2="1630.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.83" y1="933.105" x2="1599.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.83" y1="933.105" x2="1599.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.83" y1="933.105" x2="1631.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.83" y1="933.105" x2="1631.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.951" y1="1271.4" x2="883.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.951" y1="1271.4" x2="883.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.951" y1="1271.4" x2="915.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.951" y1="1271.4" x2="915.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1651.8" y1="933.534" x2="1635.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1651.8" y1="933.534" x2="1635.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1651.8" y1="933.534" x2="1667.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1651.8" y1="933.534" x2="1667.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.2" y1="982.773" x2="1530.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.2" y1="982.773" x2="1530.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.2" y1="982.773" x2="1562.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1546.2" y1="982.773" x2="1562.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="798.462" y1="1475.78" x2="782.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="798.462" y1="1475.78" x2="782.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="798.462" y1="1475.78" x2="814.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="798.462" y1="1475.78" x2="814.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1653.74" y1="920.543" x2="1637.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1653.74" y1="920.543" x2="1637.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1653.74" y1="920.543" x2="1669.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1653.74" y1="920.543" x2="1669.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1550.14" y1="1004.56" x2="1534.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1550.14" y1="1004.56" x2="1534.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1550.14" y1="1004.56" x2="1566.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1550.14" y1="1004.56" x2="1566.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="955.148" y1="1370.3" x2="939.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="955.148" y1="1370.3" x2="939.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="955.148" y1="1370.3" x2="971.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="955.148" y1="1370.3" x2="971.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1845.45" y1="659.705" x2="1829.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1845.45" y1="659.705" x2="1829.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1845.45" y1="659.705" x2="1861.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1845.45" y1="659.705" x2="1861.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1817.11" y1="731.36" x2="1801.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1817.11" y1="731.36" x2="1801.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1817.11" y1="731.36" x2="1833.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1817.11" y1="731.36" x2="1833.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1838.37" y1="806.858" x2="1822.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1838.37" y1="806.858" x2="1822.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1838.37" y1="806.858" x2="1854.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1838.37" y1="806.858" x2="1854.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244" y1="1263.65" x2="1228" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244" y1="1263.65" x2="1228" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244" y1="1263.65" x2="1260" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244" y1="1263.65" x2="1260" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1558.7" y1="965.028" x2="1542.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1558.7" y1="965.028" x2="1542.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1558.7" y1="965.028" x2="1574.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1558.7" y1="965.028" x2="1574.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.76" y1="970.455" x2="1458.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.76" y1="970.455" x2="1458.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.76" y1="970.455" x2="1490.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.76" y1="970.455" x2="1490.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1682.62" y1="820.381" x2="1666.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1682.62" y1="820.381" x2="1666.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1682.62" y1="820.381" x2="1698.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1682.62" y1="820.381" x2="1698.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="872.16" y1="1370.98" x2="856.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="872.16" y1="1370.98" x2="856.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="872.16" y1="1370.98" x2="888.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="872.16" y1="1370.98" x2="888.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="866.851" y1="1331.78" x2="850.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="866.851" y1="1331.78" x2="850.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="866.851" y1="1331.78" x2="882.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="866.851" y1="1331.78" x2="882.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1604.79" y1="991.27" x2="1588.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1604.79" y1="991.27" x2="1588.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1604.79" y1="991.27" x2="1620.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1604.79" y1="991.27" x2="1620.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="436.181" y1="1793.59" x2="420.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="436.181" y1="1793.59" x2="420.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="436.181" y1="1793.59" x2="452.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="436.181" y1="1793.59" x2="452.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="957.079" y1="1394.58" x2="941.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="957.079" y1="1394.58" x2="941.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="957.079" y1="1394.58" x2="973.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="957.079" y1="1394.58" x2="973.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.366" y1="1495.39" x2="775.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.366" y1="1495.39" x2="775.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.366" y1="1495.39" x2="807.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.366" y1="1495.39" x2="807.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1636.9" y1="890.133" x2="1620.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1636.9" y1="890.133" x2="1620.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1636.9" y1="890.133" x2="1652.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1636.9" y1="890.133" x2="1652.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1256" y1="1205.58" x2="1240" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1256" y1="1205.58" x2="1240" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1256" y1="1205.58" x2="1272" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1256" y1="1205.58" x2="1272" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="721.091" y1="1561.75" x2="705.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="721.091" y1="1561.75" x2="705.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="721.091" y1="1561.75" x2="737.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="721.091" y1="1561.75" x2="737.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="823.473" y1="1446.58" x2="807.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="823.473" y1="1446.58" x2="807.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="823.473" y1="1446.58" x2="839.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="823.473" y1="1446.58" x2="839.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="887.822" y1="1375.21" x2="871.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="887.822" y1="1375.21" x2="871.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="887.822" y1="1375.21" x2="903.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="887.822" y1="1375.21" x2="903.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.39" y1="1014.06" x2="1468.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.39" y1="1014.06" x2="1468.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.39" y1="1014.06" x2="1500.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1484.39" y1="1014.06" x2="1500.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1685.1" y1="924.047" x2="1669.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1685.1" y1="924.047" x2="1669.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1685.1" y1="924.047" x2="1701.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1685.1" y1="924.047" x2="1701.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="877.025" y1="1403.06" x2="861.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="877.025" y1="1403.06" x2="861.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="877.025" y1="1403.06" x2="893.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="877.025" y1="1403.06" x2="893.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1473.39" y1="1000.62" x2="1457.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1473.39" y1="1000.62" x2="1457.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1473.39" y1="1000.62" x2="1489.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1473.39" y1="1000.62" x2="1489.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="930.593" y1="1251.3" x2="914.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="930.593" y1="1251.3" x2="914.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="930.593" y1="1251.3" x2="946.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="930.593" y1="1251.3" x2="946.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.55" y1="972.554" x2="1599.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.55" y1="972.554" x2="1599.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.55" y1="972.554" x2="1631.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1615.55" y1="972.554" x2="1631.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1395.64" y1="960.36" x2="1379.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1395.64" y1="960.36" x2="1379.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1395.64" y1="960.36" x2="1411.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1395.64" y1="960.36" x2="1411.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.08" y1="1312.36" x2="984.075" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.08" y1="1312.36" x2="984.075" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="805.605" y1="1368.59" x2="789.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="805.605" y1="1368.59" x2="789.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="805.605" y1="1368.59" x2="821.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="805.605" y1="1368.59" x2="821.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="735.913" y1="1350.58" x2="719.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="735.913" y1="1350.58" x2="719.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="735.913" y1="1350.58" x2="751.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="735.913" y1="1350.58" x2="751.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1826.24" y1="921.284" x2="1810.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1826.24" y1="921.284" x2="1810.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1826.24" y1="921.284" x2="1842.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1826.24" y1="921.284" x2="1842.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1572.67" y1="995.733" x2="1556.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1572.67" y1="995.733" x2="1556.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1572.67" y1="995.733" x2="1588.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1572.67" y1="995.733" x2="1588.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1469" y1="974.22" x2="1453" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1469" y1="974.22" x2="1453" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1469" y1="974.22" x2="1485" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1469" y1="974.22" x2="1485" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.64" y1="1000.17" x2="1752.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.64" y1="1000.17" x2="1752.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.64" y1="1000.17" x2="1784.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1768.64" y1="1000.17" x2="1784.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.6" y1="944.573" x2="1470.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.6" y1="944.573" x2="1470.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.6" y1="944.573" x2="1502.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1486.6" y1="944.573" x2="1502.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.81" y1="1008.74" x2="1427.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.81" y1="1008.74" x2="1427.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.81" y1="1008.74" x2="1459.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1443.81" y1="1008.74" x2="1459.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1628.36" y1="926.189" x2="1612.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1628.36" y1="926.189" x2="1612.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1628.36" y1="926.189" x2="1644.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1628.36" y1="926.189" x2="1644.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.938" y1="1534.93" x2="553.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.938" y1="1534.93" x2="553.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.938" y1="1534.93" x2="585.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="569.938" y1="1534.93" x2="585.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.218" y1="1483.05" x2="728.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.218" y1="1483.05" x2="728.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.218" y1="1483.05" x2="760.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="744.218" y1="1483.05" x2="760.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1713.33" y1="929.721" x2="1697.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1713.33" y1="929.721" x2="1697.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1713.33" y1="929.721" x2="1729.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1713.33" y1="929.721" x2="1729.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="610.516" y1="1568.23" x2="594.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="610.516" y1="1568.23" x2="594.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="610.516" y1="1568.23" x2="626.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="610.516" y1="1568.23" x2="626.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1702.63" y1="949.089" x2="1686.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1702.63" y1="949.089" x2="1686.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1702.63" y1="949.089" x2="1718.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1702.63" y1="949.089" x2="1718.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1799.12" y1="921.26" x2="1783.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1799.12" y1="921.26" x2="1783.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1799.12" y1="921.26" x2="1815.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1799.12" y1="921.26" x2="1815.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1539.91" y1="992.445" x2="1523.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1539.91" y1="992.445" x2="1523.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1539.91" y1="992.445" x2="1555.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1539.91" y1="992.445" x2="1555.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.824" y1="1370.17" x2="947.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.824" y1="1370.17" x2="947.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.824" y1="1370.17" x2="979.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.824" y1="1370.17" x2="979.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="694.471" y1="1399.51" x2="678.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="694.471" y1="1399.51" x2="678.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="694.471" y1="1399.51" x2="710.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="694.471" y1="1399.51" x2="710.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.32" y1="993.037" x2="1462.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.32" y1="993.037" x2="1462.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.32" y1="993.037" x2="1494.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1478.32" y1="993.037" x2="1494.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.95" y1="1279" x2="1264.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.95" y1="1279" x2="1264.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.95" y1="1279" x2="1296.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.95" y1="1279" x2="1296.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1650.12" y1="918.653" x2="1634.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1650.12" y1="918.653" x2="1634.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1650.12" y1="918.653" x2="1666.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1650.12" y1="918.653" x2="1666.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1860.89" y1="778.201" x2="1844.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1860.89" y1="778.201" x2="1844.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1860.89" y1="778.201" x2="1876.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1860.89" y1="778.201" x2="1876.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1444.37" y1="1001.52" x2="1428.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1444.37" y1="1001.52" x2="1428.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1444.37" y1="1001.52" x2="1460.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1444.37" y1="1001.52" x2="1460.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.75" y1="1012.66" x2="1412.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.75" y1="1012.66" x2="1412.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.75" y1="1012.66" x2="1444.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.75" y1="1012.66" x2="1444.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1896.85" y1="631.005" x2="1880.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1896.85" y1="631.005" x2="1880.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1896.85" y1="631.005" x2="1912.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1896.85" y1="631.005" x2="1912.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1863.03" y1="689.163" x2="1847.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1863.03" y1="689.163" x2="1847.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1863.03" y1="689.163" x2="1879.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1863.03" y1="689.163" x2="1879.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="973.556" y1="1249.11" x2="957.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="973.556" y1="1249.11" x2="957.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="973.556" y1="1249.11" x2="989.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="973.556" y1="1249.11" x2="989.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.42" y1="974.507" x2="1399.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.42" y1="974.507" x2="1399.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.42" y1="974.507" x2="1431.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1415.42" y1="974.507" x2="1431.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="961.172" y1="1297.9" x2="945.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="961.172" y1="1297.9" x2="945.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="961.172" y1="1297.9" x2="977.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="961.172" y1="1297.9" x2="977.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="828.067" y1="1395.72" x2="812.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="828.067" y1="1395.72" x2="812.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="828.067" y1="1395.72" x2="844.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="828.067" y1="1395.72" x2="844.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="812.031" y1="1484.71" x2="796.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="812.031" y1="1484.71" x2="796.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="812.031" y1="1484.71" x2="828.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="812.031" y1="1484.71" x2="828.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="901.731" y1="1295.1" x2="885.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="901.731" y1="1295.1" x2="885.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="901.731" y1="1295.1" x2="917.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="901.731" y1="1295.1" x2="917.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.853" y1="1331.79" x2="956.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.853" y1="1331.79" x2="956.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.853" y1="1331.79" x2="988.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="972.853" y1="1331.79" x2="988.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.97" y1="912.119" x2="1722.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.97" y1="912.119" x2="1722.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.97" y1="912.119" x2="1754.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.97" y1="912.119" x2="1754.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1534.29" y1="956.26" x2="1518.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1534.29" y1="956.26" x2="1518.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1534.29" y1="956.26" x2="1550.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1534.29" y1="956.26" x2="1550.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.177" y1="1437.3" x2="863.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.177" y1="1437.3" x2="863.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.177" y1="1437.3" x2="895.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.177" y1="1437.3" x2="895.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.308" y1="1302.19" x2="775.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.308" y1="1302.19" x2="775.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.308" y1="1302.19" x2="807.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="791.308" y1="1302.19" x2="807.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.44" y1="890.561" x2="1574.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.44" y1="890.561" x2="1574.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.44" y1="890.561" x2="1606.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.44" y1="890.561" x2="1606.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1729.02" y1="853.811" x2="1713.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1729.02" y1="853.811" x2="1713.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1729.02" y1="853.811" x2="1745.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1729.02" y1="853.811" x2="1745.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="936.651" y1="1397.2" x2="920.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="936.651" y1="1397.2" x2="920.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="936.651" y1="1397.2" x2="952.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="936.651" y1="1397.2" x2="952.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="856.289" y1="1386.44" x2="840.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="856.289" y1="1386.44" x2="840.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="856.289" y1="1386.44" x2="872.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="856.289" y1="1386.44" x2="872.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1350.38" y1="930.437" x2="1334.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1350.38" y1="930.437" x2="1334.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1350.38" y1="930.437" x2="1366.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1350.38" y1="930.437" x2="1366.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1529.58" y1="996.113" x2="1513.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1529.58" y1="996.113" x2="1513.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1529.58" y1="996.113" x2="1545.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1529.58" y1="996.113" x2="1545.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.47" y1="955.973" x2="1497.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.47" y1="955.973" x2="1497.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.47" y1="955.973" x2="1529.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.47" y1="955.973" x2="1529.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1530.22" y1="1008.88" x2="1514.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1530.22" y1="1008.88" x2="1514.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1530.22" y1="1008.88" x2="1546.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1530.22" y1="1008.88" x2="1546.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.75" y1="828.151" x2="1574.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.75" y1="828.151" x2="1574.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.75" y1="828.151" x2="1606.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1590.75" y1="828.151" x2="1606.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1942.19" y1="883.328" x2="1926.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1942.19" y1="883.328" x2="1926.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1942.19" y1="883.328" x2="1958.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1942.19" y1="883.328" x2="1958.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="800.484" y1="1327.86" x2="784.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="800.484" y1="1327.86" x2="784.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="800.484" y1="1327.86" x2="816.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="800.484" y1="1327.86" x2="816.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.479" y1="1361.21" x2="863.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.479" y1="1361.21" x2="863.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.479" y1="1361.21" x2="895.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="879.479" y1="1361.21" x2="895.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1621.59" y1="888.584" x2="1605.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1621.59" y1="888.584" x2="1605.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1621.59" y1="888.584" x2="1637.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1621.59" y1="888.584" x2="1637.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.03" y1="917.837" x2="1589.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.03" y1="917.837" x2="1589.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.03" y1="917.837" x2="1621.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1605.03" y1="917.837" x2="1621.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="863.983" y1="1256.2" x2="847.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="863.983" y1="1256.2" x2="847.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="863.983" y1="1256.2" x2="879.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="863.983" y1="1256.2" x2="879.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1763.1" y1="789.55" x2="1747.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1763.1" y1="789.55" x2="1747.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1763.1" y1="789.55" x2="1779.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1763.1" y1="789.55" x2="1779.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1840.32" y1="839.458" x2="1824.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1840.32" y1="839.458" x2="1824.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1840.32" y1="839.458" x2="1856.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1840.32" y1="839.458" x2="1856.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1566.73" y1="937.472" x2="1550.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1566.73" y1="937.472" x2="1550.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1566.73" y1="937.472" x2="1582.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1566.73" y1="937.472" x2="1582.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.17" y1="997.851" x2="1441.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.17" y1="997.851" x2="1441.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.17" y1="997.851" x2="1473.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.17" y1="997.851" x2="1473.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="725.698" y1="1501.31" x2="709.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="725.698" y1="1501.31" x2="709.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="725.698" y1="1501.31" x2="741.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="725.698" y1="1501.31" x2="741.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="804.197" y1="1345.85" x2="788.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="804.197" y1="1345.85" x2="788.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="804.197" y1="1345.85" x2="820.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="804.197" y1="1345.85" x2="820.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.44" y1="832.688" x2="1722.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.44" y1="832.688" x2="1722.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.44" y1="832.688" x2="1754.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1738.44" y1="832.688" x2="1754.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1575.07" y1="927.362" x2="1559.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1575.07" y1="927.362" x2="1559.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1575.07" y1="927.362" x2="1591.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1575.07" y1="927.362" x2="1591.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.369" y1="1393.26" x2="947.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.369" y1="1393.26" x2="947.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.369" y1="1393.26" x2="979.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.369" y1="1393.26" x2="979.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1667.21" y1="967.122" x2="1651.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1667.21" y1="967.122" x2="1651.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1667.21" y1="967.122" x2="1683.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1667.21" y1="967.122" x2="1683.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1808.12" y1="803.64" x2="1792.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1808.12" y1="803.64" x2="1792.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1808.12" y1="803.64" x2="1824.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1808.12" y1="803.64" x2="1824.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.457" y1="1364.93" x2="911.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.457" y1="1364.93" x2="911.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.457" y1="1364.93" x2="943.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="927.457" y1="1364.93" x2="943.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.588" y1="1394.79" x2="938.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.588" y1="1394.79" x2="938.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.588" y1="1394.79" x2="970.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.588" y1="1394.79" x2="970.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1600.82" y1="848.267" x2="1584.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1600.82" y1="848.267" x2="1584.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1600.82" y1="848.267" x2="1616.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1600.82" y1="848.267" x2="1616.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1013.05" y1="1254.21" x2="997.055" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1013.05" y1="1254.21" x2="997.055" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="710.768" y1="1471.31" x2="694.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="710.768" y1="1471.31" x2="694.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="710.768" y1="1471.31" x2="726.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="710.768" y1="1471.31" x2="726.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.455" y1="1241.66" x2="938.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.455" y1="1241.66" x2="938.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.455" y1="1241.66" x2="970.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="954.455" y1="1241.66" x2="970.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.19" y1="997.572" x2="1412.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.19" y1="997.572" x2="1412.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.19" y1="997.572" x2="1444.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1428.19" y1="997.572" x2="1444.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="907.609" y1="1313.11" x2="891.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="907.609" y1="1313.11" x2="891.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="907.609" y1="1313.11" x2="923.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="907.609" y1="1313.11" x2="923.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="904.55" y1="1372.25" x2="888.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="904.55" y1="1372.25" x2="888.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="904.55" y1="1372.25" x2="920.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="904.55" y1="1372.25" x2="920.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1561.34" y1="914.945" x2="1545.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1561.34" y1="914.945" x2="1545.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1561.34" y1="914.945" x2="1577.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1561.34" y1="914.945" x2="1577.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1311.21" y1="973.579" x2="1295.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1311.21" y1="973.579" x2="1295.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1311.21" y1="973.579" x2="1327.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1311.21" y1="973.579" x2="1327.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="743.908" y1="1344.73" x2="727.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="743.908" y1="1344.73" x2="727.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="743.908" y1="1344.73" x2="759.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="743.908" y1="1344.73" x2="759.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.408" y1="1314.39" x2="947.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.408" y1="1314.39" x2="947.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.408" y1="1314.39" x2="979.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="963.408" y1="1314.39" x2="979.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="994.563" y1="1220.85" x2="978.563" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="994.563" y1="1220.85" x2="978.563" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="994.563" y1="1220.85" x2="1010.56" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="994.563" y1="1220.85" x2="1010.56" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.07" y1="993.346" x2="1764.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.07" y1="993.346" x2="1764.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.07" y1="993.346" x2="1796.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1780.07" y1="993.346" x2="1796.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="983.358" y1="1291.75" x2="967.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="983.358" y1="1291.75" x2="967.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="983.358" y1="1291.75" x2="999.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="983.358" y1="1291.75" x2="999.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1496.43" y1="924.203" x2="1480.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1496.43" y1="924.203" x2="1480.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1496.43" y1="924.203" x2="1512.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1496.43" y1="924.203" x2="1512.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="938.137" y1="1395.53" x2="922.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="938.137" y1="1395.53" x2="922.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="938.137" y1="1395.53" x2="954.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="938.137" y1="1395.53" x2="954.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="745.679" y1="1455.46" x2="729.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="745.679" y1="1455.46" x2="729.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="745.679" y1="1455.46" x2="761.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="745.679" y1="1455.46" x2="761.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1226.7" y1="1172" x2="1210.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1226.7" y1="1172" x2="1210.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1226.7" y1="1172" x2="1242.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1226.7" y1="1172" x2="1242.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.77" y1="884.258" x2="1746.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.77" y1="884.258" x2="1746.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.77" y1="884.258" x2="1778.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1762.77" y1="884.258" x2="1778.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="977.334" y1="1349.1" x2="961.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="977.334" y1="1349.1" x2="961.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="977.334" y1="1349.1" x2="993.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="977.334" y1="1349.1" x2="993.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.292" y1="1482.58" x2="859.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.292" y1="1482.58" x2="859.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.292" y1="1482.58" x2="891.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="875.292" y1="1482.58" x2="891.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.433" y1="1329.29" x2="883.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.433" y1="1329.29" x2="883.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.433" y1="1329.29" x2="915.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="899.433" y1="1329.29" x2="915.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1552.62" y1="936.348" x2="1536.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1552.62" y1="936.348" x2="1536.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1552.62" y1="936.348" x2="1568.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1552.62" y1="936.348" x2="1568.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.54" y1="984.865" x2="1456.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.54" y1="984.865" x2="1456.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.54" y1="984.865" x2="1488.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1472.54" y1="984.865" x2="1488.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165" y1="1027.31" x2="1149" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165" y1="1027.31" x2="1149" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165" y1="1027.31" x2="1181" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1165" y1="1027.31" x2="1181" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239" y1="1106.12" x2="1223" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239" y1="1106.12" x2="1223" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239" y1="1106.12" x2="1255" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1239" y1="1106.12" x2="1255" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="839.379" y1="1287.74" x2="823.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="839.379" y1="1287.74" x2="823.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="839.379" y1="1287.74" x2="855.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="839.379" y1="1287.74" x2="855.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414" y1="1170.91" x2="1398" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414" y1="1170.91" x2="1398" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414" y1="1170.91" x2="1430" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414" y1="1170.91" x2="1430" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="844.925" y1="1383.63" x2="828.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="844.925" y1="1383.63" x2="828.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="844.925" y1="1383.63" x2="860.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip132)" x1="844.925" y1="1383.63" x2="860.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<polyline clip-path="url(#clip132)" style="stroke:#dc143c; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2006.95,814.774 2025.44,792.795 2041.77,771.865 2055.91,752.038 2067.82,733.363 2077.48,715.886 2084.85,699.652 2089.91,684.701 2092.67,671.07 2093.1,658.793 2091.21,647.901 2087.01,638.422 2080.5,630.379 2071.7,623.791 2060.63,618.676 2047.32,615.047 2031.8,612.912 2014.11,612.276 1994.3,613.142 1972.42,615.507 1948.51,619.365 1922.64,624.707 1894.88,631.519 1865.29,639.784 1833.95,649.482 1800.94,660.588 1766.33,673.075 1730.22,686.911 1692.69,702.061 1653.85,718.489 1613.78,736.152 1572.58,755.007 1530.37,775.007 1487.24,796.101 1443.3,818.237 1398.66,841.36 1353.44,865.411 1307.75,890.331 1261.69,916.057 1215.4,942.525 1168.97,969.669 1122.53,997.421 1076.2,1025.71 1030.09,1054.47 984.309,1083.63 938.981,1113.1 894.216,1142.83 850.125,1172.73 806.82,1202.74 764.408,1232.76 722.995,1262.74 682.684,1292.6 643.578,1322.25 605.772,1351.63 569.362,1380.67 534.439,1409.29 501.091,1437.41 469.399,1464.97 439.444,1491.91 411.301,1518.15 385.04,1543.62 360.726,1568.27 338.42,1592.03 318.178,1614.84 300.051,1636.64 284.084,1657.39 270.316,1677.03 258.783,1695.51 249.513,1712.78 242.529,1728.8 237.849,1743.53 235.484,1756.93 235.44,1768.98 237.718,1779.64 242.312,1788.88 249.21,1796.68 258.394,1803.02 269.843,1807.88 283.527,1811.26 299.413,1813.14 317.46,1813.53 337.623,1812.41 359.852,1809.79 384.092,1805.68 410.282,1800.09 438.356,1793.04 468.244,1784.53 499.871,1774.59 533.16,1763.25 568.025,1750.54 604.381,1736.48 642.135,1721.11 681.195,1704.47 721.462,1686.6 762.835,1667.55 805.212,1647.37 848.486,1626.09 892.549,1603.79 937.291,1580.5 982.6,1556.3 1028.36,1531.24 1074.47,1505.39 1120.79,1478.8 1167.23,1451.55 1213.66,1423.7 1259.96,1395.33 1306.02,1366.5 1351.73,1337.28 1396.98,1307.76 1441.64,1278 1485.6,1248.07 1528.77,1218.06 1571.02,1188.03 1612.25,1158.07 1652.37,1128.24 1691.26,1098.63 1728.84,1069.3 1765,1040.33 1799.67,1011.79 1832.74,983.758 1864.15,956.295 1893.8,929.473 1921.64,903.361 1947.57,878.021 1971.56,853.519 1993.52,829.915 2013.41,807.269 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="327.236,2261.17 327.236,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="640.095,2261.17 640.095,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="952.953,2261.17 952.953,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1265.81,2261.17 1265.81,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1578.67,2261.17 1578.67,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1891.53,2261.17 1891.53,47.2441 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2204.39,2261.17 2204.39,47.2441 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 2352.76,2261.17 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="327.236,2261.17 327.236,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="640.095,2261.17 640.095,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="952.953,2261.17 952.953,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1265.81,2261.17 1265.81,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1578.67,2261.17 1578.67,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1891.53,2261.17 1891.53,2242.27 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2204.39,2261.17 2204.39,2242.27 "/>
+<path clip-path="url(#clip950)" d="M281.299 2306.74 L310.975 2306.74 L310.975 2310.67 L281.299 2310.67 L281.299 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M335.234 2304.93 Q338.59 2305.65 340.465 2307.92 Q342.363 2310.19 342.363 2313.52 Q342.363 2318.64 338.845 2321.44 Q335.326 2324.24 328.845 2324.24 Q326.669 2324.24 324.354 2323.8 Q322.063 2323.38 319.609 2322.53 L319.609 2318.01 Q321.553 2319.15 323.868 2319.73 Q326.183 2320.3 328.706 2320.3 Q333.104 2320.3 335.396 2318.57 Q337.711 2316.83 337.711 2313.52 Q337.711 2310.47 335.558 2308.75 Q333.428 2307.02 329.609 2307.02 L325.581 2307.02 L325.581 2303.17 L329.794 2303.17 Q333.243 2303.17 335.072 2301.81 Q336.9 2300.42 336.9 2297.83 Q336.9 2295.17 335.002 2293.75 Q333.127 2292.32 329.609 2292.32 Q327.688 2292.32 325.488 2292.73 Q323.289 2293.15 320.651 2294.03 L320.651 2289.86 Q323.313 2289.12 325.627 2288.75 Q327.965 2288.38 330.025 2288.38 Q335.35 2288.38 338.451 2290.81 Q341.553 2293.22 341.553 2297.34 Q341.553 2300.21 339.91 2302.2 Q338.266 2304.17 335.234 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M361.229 2292.09 Q357.618 2292.09 355.789 2295.65 Q353.984 2299.19 353.984 2306.32 Q353.984 2313.43 355.789 2316.99 Q357.618 2320.54 361.229 2320.54 Q364.863 2320.54 366.669 2316.99 Q368.497 2313.43 368.497 2306.32 Q368.497 2299.19 366.669 2295.65 Q364.863 2292.09 361.229 2292.09 M361.229 2288.38 Q367.039 2288.38 370.095 2292.99 Q373.173 2297.57 373.173 2306.32 Q373.173 2315.05 370.095 2319.66 Q367.039 2324.24 361.229 2324.24 Q355.419 2324.24 352.34 2319.66 Q349.285 2315.05 349.285 2306.32 Q349.285 2297.57 352.34 2292.99 Q355.419 2288.38 361.229 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M594.157 2306.74 L623.833 2306.74 L623.833 2310.67 L594.157 2310.67 L594.157 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M637.953 2319.63 L654.273 2319.63 L654.273 2323.57 L632.328 2323.57 L632.328 2319.63 Q634.99 2316.88 639.574 2312.25 Q644.18 2307.6 645.361 2306.25 Q647.606 2303.73 648.486 2301.99 Q649.388 2300.23 649.388 2298.54 Q649.388 2295.79 647.444 2294.05 Q645.523 2292.32 642.421 2292.32 Q640.222 2292.32 637.768 2293.08 Q635.338 2293.85 632.56 2295.4 L632.56 2290.67 Q635.384 2289.54 637.838 2288.96 Q640.291 2288.38 642.328 2288.38 Q647.699 2288.38 650.893 2291.07 Q654.087 2293.75 654.087 2298.24 Q654.087 2300.37 653.277 2302.29 Q652.49 2304.19 650.384 2306.79 Q649.805 2307.46 646.703 2310.67 Q643.601 2313.87 637.953 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M674.087 2292.09 Q670.476 2292.09 668.648 2295.65 Q666.842 2299.19 666.842 2306.32 Q666.842 2313.43 668.648 2316.99 Q670.476 2320.54 674.087 2320.54 Q677.722 2320.54 679.527 2316.99 Q681.356 2313.43 681.356 2306.32 Q681.356 2299.19 679.527 2295.65 Q677.722 2292.09 674.087 2292.09 M674.087 2288.38 Q679.898 2288.38 682.953 2292.99 Q686.032 2297.57 686.032 2306.32 Q686.032 2315.05 682.953 2319.66 Q679.898 2324.24 674.087 2324.24 Q668.277 2324.24 665.199 2319.66 Q662.143 2315.05 662.143 2306.32 Q662.143 2297.57 665.199 2292.99 Q668.277 2288.38 674.087 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M907.016 2306.74 L936.691 2306.74 L936.691 2310.67 L907.016 2310.67 L907.016 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M947.594 2319.63 L955.233 2319.63 L955.233 2293.27 L946.923 2294.93 L946.923 2290.67 L955.187 2289.01 L959.863 2289.01 L959.863 2319.63 L967.501 2319.63 L967.501 2323.57 L947.594 2323.57 L947.594 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M986.946 2292.09 Q983.335 2292.09 981.506 2295.65 Q979.7 2299.19 979.7 2306.32 Q979.7 2313.43 981.506 2316.99 Q983.335 2320.54 986.946 2320.54 Q990.58 2320.54 992.386 2316.99 Q994.214 2313.43 994.214 2306.32 Q994.214 2299.19 992.386 2295.65 Q990.58 2292.09 986.946 2292.09 M986.946 2288.38 Q992.756 2288.38 995.811 2292.99 Q998.89 2297.57 998.89 2306.32 Q998.89 2315.05 995.811 2319.66 Q992.756 2324.24 986.946 2324.24 Q981.136 2324.24 978.057 2319.66 Q975.001 2315.05 975.001 2306.32 Q975.001 2297.57 978.057 2292.99 Q981.136 2288.38 986.946 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1265.81 2292.09 Q1262.2 2292.09 1260.37 2295.65 Q1258.57 2299.19 1258.57 2306.32 Q1258.57 2313.43 1260.37 2316.99 Q1262.2 2320.54 1265.81 2320.54 Q1269.45 2320.54 1271.25 2316.99 Q1273.08 2313.43 1273.08 2306.32 Q1273.08 2299.19 1271.25 2295.65 Q1269.45 2292.09 1265.81 2292.09 M1265.81 2288.38 Q1271.62 2288.38 1274.68 2292.99 Q1277.76 2297.57 1277.76 2306.32 Q1277.76 2315.05 1274.68 2319.66 Q1271.62 2324.24 1265.81 2324.24 Q1260 2324.24 1256.92 2319.66 Q1253.87 2315.05 1253.87 2306.32 Q1253.87 2297.57 1256.92 2292.99 Q1260 2288.38 1265.81 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1553.36 2319.63 L1561 2319.63 L1561 2293.27 L1552.69 2294.93 L1552.69 2290.67 L1560.95 2289.01 L1565.63 2289.01 L1565.63 2319.63 L1573.26 2319.63 L1573.26 2323.57 L1553.36 2323.57 L1553.36 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1592.71 2292.09 Q1589.1 2292.09 1587.27 2295.65 Q1585.46 2299.19 1585.46 2306.32 Q1585.46 2313.43 1587.27 2316.99 Q1589.1 2320.54 1592.71 2320.54 Q1596.34 2320.54 1598.15 2316.99 Q1599.98 2313.43 1599.98 2306.32 Q1599.98 2299.19 1598.15 2295.65 Q1596.34 2292.09 1592.71 2292.09 M1592.71 2288.38 Q1598.52 2288.38 1601.57 2292.99 Q1604.65 2297.57 1604.65 2306.32 Q1604.65 2315.05 1601.57 2319.66 Q1598.52 2324.24 1592.71 2324.24 Q1586.9 2324.24 1583.82 2319.66 Q1580.76 2315.05 1580.76 2306.32 Q1580.76 2297.57 1583.82 2292.99 Q1586.9 2288.38 1592.71 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1870.3 2319.63 L1886.62 2319.63 L1886.62 2323.57 L1864.68 2323.57 L1864.68 2319.63 Q1867.34 2316.88 1871.92 2312.25 Q1876.53 2307.6 1877.71 2306.25 Q1879.95 2303.73 1880.83 2301.99 Q1881.74 2300.23 1881.74 2298.54 Q1881.74 2295.79 1879.79 2294.05 Q1877.87 2292.32 1874.77 2292.32 Q1872.57 2292.32 1870.12 2293.08 Q1867.69 2293.85 1864.91 2295.4 L1864.91 2290.67 Q1867.73 2289.54 1870.19 2288.96 Q1872.64 2288.38 1874.68 2288.38 Q1880.05 2288.38 1883.24 2291.07 Q1886.44 2293.75 1886.44 2298.24 Q1886.44 2300.37 1885.63 2302.29 Q1884.84 2304.19 1882.73 2306.79 Q1882.15 2307.46 1879.05 2310.67 Q1875.95 2313.87 1870.3 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M1906.44 2292.09 Q1902.82 2292.09 1901 2295.65 Q1899.19 2299.19 1899.19 2306.32 Q1899.19 2313.43 1901 2316.99 Q1902.82 2320.54 1906.44 2320.54 Q1910.07 2320.54 1911.88 2316.99 Q1913.7 2313.43 1913.7 2306.32 Q1913.7 2299.19 1911.88 2295.65 Q1910.07 2292.09 1906.44 2292.09 M1906.44 2288.38 Q1912.25 2288.38 1915.3 2292.99 Q1918.38 2297.57 1918.38 2306.32 Q1918.38 2315.05 1915.3 2319.66 Q1912.25 2324.24 1906.44 2324.24 Q1900.63 2324.24 1897.55 2319.66 Q1894.49 2315.05 1894.49 2306.32 Q1894.49 2297.57 1897.55 2292.99 Q1900.63 2288.38 1906.44 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M2193.23 2304.93 Q2196.59 2305.65 2198.46 2307.92 Q2200.36 2310.19 2200.36 2313.52 Q2200.36 2318.64 2196.84 2321.44 Q2193.32 2324.24 2186.84 2324.24 Q2184.66 2324.24 2182.35 2323.8 Q2180.06 2323.38 2177.6 2322.53 L2177.6 2318.01 Q2179.55 2319.15 2181.86 2319.73 Q2184.18 2320.3 2186.7 2320.3 Q2191.1 2320.3 2193.39 2318.57 Q2195.71 2316.83 2195.71 2313.52 Q2195.71 2310.47 2193.55 2308.75 Q2191.42 2307.02 2187.6 2307.02 L2183.58 2307.02 L2183.58 2303.17 L2187.79 2303.17 Q2191.24 2303.17 2193.07 2301.81 Q2194.9 2300.42 2194.9 2297.83 Q2194.9 2295.17 2193 2293.75 Q2191.12 2292.32 2187.6 2292.32 Q2185.68 2292.32 2183.48 2292.73 Q2181.28 2293.15 2178.65 2294.03 L2178.65 2289.86 Q2181.31 2289.12 2183.62 2288.75 Q2185.96 2288.38 2188.02 2288.38 Q2193.34 2288.38 2196.45 2290.81 Q2199.55 2293.22 2199.55 2297.34 Q2199.55 2300.21 2197.9 2302.2 Q2196.26 2304.17 2193.23 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M2219.22 2292.09 Q2215.61 2292.09 2213.78 2295.65 Q2211.98 2299.19 2211.98 2306.32 Q2211.98 2313.43 2213.78 2316.99 Q2215.61 2320.54 2219.22 2320.54 Q2222.86 2320.54 2224.66 2316.99 Q2226.49 2313.43 2226.49 2306.32 Q2226.49 2299.19 2224.66 2295.65 Q2222.86 2292.09 2219.22 2292.09 M2219.22 2288.38 Q2225.03 2288.38 2228.09 2292.99 Q2231.17 2297.57 2231.17 2306.32 Q2231.17 2315.05 2228.09 2319.66 Q2225.03 2324.24 2219.22 2324.24 Q2213.41 2324.24 2210.34 2319.66 Q2207.28 2315.05 2207.28 2306.32 Q2207.28 2297.57 2210.34 2292.99 Q2213.41 2288.38 2219.22 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,2110.07 2352.76,2110.07 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1791.45 2352.76,1791.45 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1472.83 2352.76,1472.83 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1154.21 2352.76,1154.21 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,835.586 2352.76,835.586 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,516.966 2352.76,516.966 "/>
+<polyline clip-path="url(#clip952)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,198.346 2352.76,198.346 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 178.867,47.2441 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2110.07 197.764,2110.07 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1791.45 197.764,1791.45 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1472.83 197.764,1472.83 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1154.21 197.764,1154.21 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,835.586 197.764,835.586 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,516.966 197.764,516.966 "/>
+<polyline clip-path="url(#clip950)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,198.346 197.764,198.346 "/>
+<path clip-path="url(#clip950)" d="M50.9921 2110.52 L80.6679 2110.52 L80.6679 2114.45 L50.9921 2114.45 L50.9921 2110.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M104.927 2108.71 Q108.283 2109.43 110.158 2111.7 Q112.057 2113.97 112.057 2117.3 Q112.057 2122.42 108.538 2125.22 Q105.02 2128.02 98.5382 2128.02 Q96.3623 2128.02 94.0475 2127.58 Q91.7558 2127.16 89.3021 2126.3 L89.3021 2121.79 Q91.2465 2122.92 93.5613 2123.5 Q95.8761 2124.08 98.3993 2124.08 Q102.797 2124.08 105.089 2122.35 Q107.404 2120.61 107.404 2117.3 Q107.404 2114.24 105.251 2112.53 Q103.121 2110.8 99.3021 2110.8 L95.2743 2110.8 L95.2743 2106.95 L99.4872 2106.95 Q102.936 2106.95 104.765 2105.59 Q106.594 2104.2 106.594 2101.61 Q106.594 2098.94 104.696 2097.53 Q102.821 2096.1 99.3021 2096.1 Q97.3808 2096.1 95.1817 2096.51 Q92.9826 2096.93 90.3438 2097.81 L90.3438 2093.64 Q93.0058 2092.9 95.3206 2092.53 Q97.6585 2092.16 99.7187 2092.16 Q105.043 2092.16 108.145 2094.59 Q111.246 2097 111.246 2101.12 Q111.246 2103.99 109.603 2105.98 Q107.959 2107.95 104.927 2108.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 2095.86 Q127.311 2095.86 125.482 2099.43 Q123.677 2102.97 123.677 2110.1 Q123.677 2117.21 125.482 2120.77 Q127.311 2124.31 130.922 2124.31 Q134.556 2124.31 136.362 2120.77 Q138.191 2117.21 138.191 2110.1 Q138.191 2102.97 136.362 2099.43 Q134.556 2095.86 130.922 2095.86 M130.922 2092.16 Q136.732 2092.16 139.788 2096.77 Q142.867 2101.35 142.867 2110.1 Q142.867 2118.83 139.788 2123.43 Q136.732 2128.02 130.922 2128.02 Q125.112 2128.02 122.033 2123.43 Q118.978 2118.83 118.978 2110.1 Q118.978 2101.35 122.033 2096.77 Q125.112 2092.16 130.922 2092.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M50.9921 1791.9 L80.6679 1791.9 L80.6679 1795.83 L50.9921 1795.83 L50.9921 1791.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M94.7882 1804.79 L111.108 1804.79 L111.108 1808.73 L89.1632 1808.73 L89.1632 1804.79 Q91.8252 1802.04 96.4085 1797.41 Q101.015 1792.75 102.196 1791.41 Q104.441 1788.89 105.321 1787.15 Q106.223 1785.39 106.223 1783.7 Q106.223 1780.95 104.279 1779.21 Q102.358 1777.48 99.2558 1777.48 Q97.0567 1777.48 94.603 1778.24 Q92.1725 1779 89.3947 1780.55 L89.3947 1775.83 Q92.2188 1774.7 94.6724 1774.12 Q97.1261 1773.54 99.1632 1773.54 Q104.534 1773.54 107.728 1776.23 Q110.922 1778.91 110.922 1783.4 Q110.922 1785.53 110.112 1787.45 Q109.325 1789.35 107.219 1791.94 Q106.64 1792.61 103.538 1795.83 Q100.436 1799.03 94.7882 1804.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 1777.24 Q127.311 1777.24 125.482 1780.81 Q123.677 1784.35 123.677 1791.48 Q123.677 1798.59 125.482 1802.15 Q127.311 1805.69 130.922 1805.69 Q134.556 1805.69 136.362 1802.15 Q138.191 1798.59 138.191 1791.48 Q138.191 1784.35 136.362 1780.81 Q134.556 1777.24 130.922 1777.24 M130.922 1773.54 Q136.732 1773.54 139.788 1778.15 Q142.867 1782.73 142.867 1791.48 Q142.867 1800.21 139.788 1804.81 Q136.732 1809.4 130.922 1809.4 Q125.112 1809.4 122.033 1804.81 Q118.978 1800.21 118.978 1791.48 Q118.978 1782.73 122.033 1778.15 Q125.112 1773.54 130.922 1773.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M50.9921 1473.28 L80.6679 1473.28 L80.6679 1477.21 L50.9921 1477.21 L50.9921 1473.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M91.5706 1486.17 L99.2095 1486.17 L99.2095 1459.81 L90.8993 1461.47 L90.8993 1457.21 L99.1632 1455.55 L103.839 1455.55 L103.839 1486.17 L111.478 1486.17 L111.478 1490.11 L91.5706 1490.11 L91.5706 1486.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 1458.62 Q127.311 1458.62 125.482 1462.19 Q123.677 1465.73 123.677 1472.86 Q123.677 1479.97 125.482 1483.53 Q127.311 1487.07 130.922 1487.07 Q134.556 1487.07 136.362 1483.53 Q138.191 1479.97 138.191 1472.86 Q138.191 1465.73 136.362 1462.19 Q134.556 1458.62 130.922 1458.62 M130.922 1454.92 Q136.732 1454.92 139.788 1459.53 Q142.867 1464.11 142.867 1472.86 Q142.867 1481.59 139.788 1486.19 Q136.732 1490.78 130.922 1490.78 Q125.112 1490.78 122.033 1486.19 Q118.978 1481.59 118.978 1472.86 Q118.978 1464.11 122.033 1459.53 Q125.112 1454.92 130.922 1454.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 1140 Q127.311 1140 125.482 1143.57 Q123.677 1147.11 123.677 1154.24 Q123.677 1161.35 125.482 1164.91 Q127.311 1168.45 130.922 1168.45 Q134.556 1168.45 136.362 1164.91 Q138.191 1161.35 138.191 1154.24 Q138.191 1147.11 136.362 1143.57 Q134.556 1140 130.922 1140 M130.922 1136.3 Q136.732 1136.3 139.788 1140.91 Q142.867 1145.49 142.867 1154.24 Q142.867 1162.97 139.788 1167.57 Q136.732 1172.16 130.922 1172.16 Q125.112 1172.16 122.033 1167.57 Q118.978 1162.97 118.978 1154.24 Q118.978 1145.49 122.033 1140.91 Q125.112 1136.3 130.922 1136.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M91.5706 848.931 L99.2095 848.931 L99.2095 822.565 L90.8993 824.232 L90.8993 819.973 L99.1632 818.306 L103.839 818.306 L103.839 848.931 L111.478 848.931 L111.478 852.866 L91.5706 852.866 L91.5706 848.931 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 821.385 Q127.311 821.385 125.482 824.949 Q123.677 828.491 123.677 835.621 Q123.677 842.727 125.482 846.292 Q127.311 849.834 130.922 849.834 Q134.556 849.834 136.362 846.292 Q138.191 842.727 138.191 835.621 Q138.191 828.491 136.362 824.949 Q134.556 821.385 130.922 821.385 M130.922 817.681 Q136.732 817.681 139.788 822.287 Q142.867 826.871 142.867 835.621 Q142.867 844.348 139.788 848.954 Q136.732 853.537 130.922 853.537 Q125.112 853.537 122.033 848.954 Q118.978 844.348 118.978 835.621 Q118.978 826.871 122.033 822.287 Q125.112 817.681 130.922 817.681 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M94.7882 530.311 L111.108 530.311 L111.108 534.246 L89.1632 534.246 L89.1632 530.311 Q91.8252 527.556 96.4085 522.927 Q101.015 518.274 102.196 516.931 Q104.441 514.408 105.321 512.672 Q106.223 510.913 106.223 509.223 Q106.223 506.468 104.279 504.732 Q102.358 502.996 99.2558 502.996 Q97.0567 502.996 94.603 503.76 Q92.1725 504.524 89.3947 506.075 L89.3947 501.353 Q92.2188 500.218 94.6724 499.64 Q97.1261 499.061 99.1632 499.061 Q104.534 499.061 107.728 501.746 Q110.922 504.431 110.922 508.922 Q110.922 511.052 110.112 512.973 Q109.325 514.871 107.219 517.464 Q106.64 518.135 103.538 521.353 Q100.436 524.547 94.7882 530.311 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 502.765 Q127.311 502.765 125.482 506.329 Q123.677 509.871 123.677 517.001 Q123.677 524.107 125.482 527.672 Q127.311 531.214 130.922 531.214 Q134.556 531.214 136.362 527.672 Q138.191 524.107 138.191 517.001 Q138.191 509.871 136.362 506.329 Q134.556 502.765 130.922 502.765 M130.922 499.061 Q136.732 499.061 139.788 503.667 Q142.867 508.251 142.867 517.001 Q142.867 525.727 139.788 530.334 Q136.732 534.917 130.922 534.917 Q125.112 534.917 122.033 530.334 Q118.978 525.727 118.978 517.001 Q118.978 508.251 122.033 503.667 Q125.112 499.061 130.922 499.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M104.927 196.992 Q108.283 197.709 110.158 199.978 Q112.057 202.246 112.057 205.58 Q112.057 210.695 108.538 213.496 Q105.02 216.297 98.5382 216.297 Q96.3623 216.297 94.0475 215.857 Q91.7558 215.441 89.3021 214.584 L89.3021 210.07 Q91.2465 211.205 93.5613 211.783 Q95.8761 212.362 98.3993 212.362 Q102.797 212.362 105.089 210.626 Q107.404 208.89 107.404 205.58 Q107.404 202.524 105.251 200.811 Q103.121 199.075 99.3021 199.075 L95.2743 199.075 L95.2743 195.233 L99.4872 195.233 Q102.936 195.233 104.765 193.867 Q106.594 192.478 106.594 189.885 Q106.594 187.223 104.696 185.811 Q102.821 184.376 99.3021 184.376 Q97.3808 184.376 95.1817 184.793 Q92.9826 185.209 90.3438 186.089 L90.3438 181.922 Q93.0058 181.182 95.3206 180.811 Q97.6585 180.441 99.7187 180.441 Q105.043 180.441 108.145 182.872 Q111.246 185.279 111.246 189.399 Q111.246 192.27 109.603 194.26 Q107.959 196.228 104.927 196.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip950)" d="M130.922 184.145 Q127.311 184.145 125.482 187.709 Q123.677 191.251 123.677 198.381 Q123.677 205.487 125.482 209.052 Q127.311 212.594 130.922 212.594 Q134.556 212.594 136.362 209.052 Q138.191 205.487 138.191 198.381 Q138.191 191.251 136.362 187.709 Q134.556 184.145 130.922 184.145 M130.922 180.441 Q136.732 180.441 139.788 185.047 Q142.867 189.631 142.867 198.381 Q142.867 207.107 139.788 211.714 Q136.732 216.297 130.922 216.297 Q125.112 216.297 122.033 211.714 Q118.978 207.107 118.978 198.381 Q118.978 189.631 122.033 185.047 Q125.112 180.441 130.922 180.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><line clip-path="url(#clip952)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1926.4" y1="856.923" x2="1910.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1926.4" y1="856.923" x2="1910.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1926.4" y1="856.923" x2="1942.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1926.4" y1="856.923" x2="1942.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="912.69" y1="1348.83" x2="896.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="912.69" y1="1348.83" x2="896.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="912.69" y1="1348.83" x2="928.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="912.69" y1="1348.83" x2="928.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="625.575" y1="1469.54" x2="609.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="625.575" y1="1469.54" x2="609.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="625.575" y1="1469.54" x2="641.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="625.575" y1="1469.54" x2="641.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.36" y1="1138.56" x2="990.356" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.36" y1="1138.56" x2="990.356" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="785.24" y1="1436.86" x2="769.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="785.24" y1="1436.86" x2="769.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="785.24" y1="1436.86" x2="801.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="785.24" y1="1436.86" x2="801.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.289" y1="1394.75" x2="761.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.289" y1="1394.75" x2="761.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.289" y1="1394.75" x2="793.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.289" y1="1394.75" x2="793.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.41" y1="1014.15" x2="1369.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.41" y1="1014.15" x2="1369.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.41" y1="1014.15" x2="1401.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.41" y1="1014.15" x2="1401.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1612.46" y1="913.718" x2="1596.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1612.46" y1="913.718" x2="1596.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1612.46" y1="913.718" x2="1628.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1612.46" y1="913.718" x2="1628.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="947.412" y1="1370.03" x2="931.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="947.412" y1="1370.03" x2="931.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="947.412" y1="1370.03" x2="963.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="947.412" y1="1370.03" x2="963.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.497" y1="1483.85" x2="672.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.497" y1="1483.85" x2="672.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.497" y1="1483.85" x2="704.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.497" y1="1483.85" x2="704.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.95" y1="869.353" x2="1529.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.95" y1="869.353" x2="1529.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.95" y1="869.353" x2="1561.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.95" y1="869.353" x2="1561.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="771.564" y1="1431.34" x2="755.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="771.564" y1="1431.34" x2="755.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="771.564" y1="1431.34" x2="787.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="771.564" y1="1431.34" x2="787.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="918.674" y1="1363.31" x2="902.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="918.674" y1="1363.31" x2="902.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="918.674" y1="1363.31" x2="934.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="918.674" y1="1363.31" x2="934.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1707.08" y1="826.86" x2="1691.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1707.08" y1="826.86" x2="1691.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1707.08" y1="826.86" x2="1723.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1707.08" y1="826.86" x2="1723.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.63" y1="875.887" x2="1561.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.63" y1="875.887" x2="1561.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.63" y1="875.887" x2="1593.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.63" y1="875.887" x2="1593.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="894.72" y1="1142.51" x2="878.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="894.72" y1="1142.51" x2="878.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="894.72" y1="1142.51" x2="910.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="894.72" y1="1142.51" x2="910.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1580.8" y1="990.127" x2="1564.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1580.8" y1="990.127" x2="1564.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1580.8" y1="990.127" x2="1596.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1580.8" y1="990.127" x2="1596.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="852.02" y1="1283.64" x2="836.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="852.02" y1="1283.64" x2="836.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="852.02" y1="1283.64" x2="868.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="852.02" y1="1283.64" x2="868.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.48" y1="962.993" x2="1589.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.48" y1="962.993" x2="1589.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.48" y1="962.993" x2="1621.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.48" y1="962.993" x2="1621.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1172.56" y1="1009.01" x2="1156.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1172.56" y1="1009.01" x2="1156.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1172.56" y1="1009.01" x2="1188.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1172.56" y1="1009.01" x2="1188.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.588" y1="1415.15" x2="672.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.588" y1="1415.15" x2="672.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.588" y1="1415.15" x2="704.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="688.588" y1="1415.15" x2="704.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="991.105" y1="1402.23" x2="975.105" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="991.105" y1="1402.23" x2="975.105" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="991.105" y1="1402.23" x2="1007.11" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="991.105" y1="1402.23" x2="1007.11" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1811.99" y1="919.23" x2="1795.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1811.99" y1="919.23" x2="1795.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1811.99" y1="919.23" x2="1827.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1811.99" y1="919.23" x2="1827.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.971" y1="1350.77" x2="829.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.971" y1="1350.77" x2="829.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.971" y1="1350.77" x2="861.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.971" y1="1350.77" x2="861.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="886.864" y1="1368.41" x2="870.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="886.864" y1="1368.41" x2="870.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="886.864" y1="1368.41" x2="902.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="886.864" y1="1368.41" x2="902.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.009" y1="1388.03" x2="761.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.009" y1="1388.03" x2="761.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.009" y1="1388.03" x2="793.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="777.009" y1="1388.03" x2="793.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.23" y1="998.156" x2="1277.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.23" y1="998.156" x2="1277.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.23" y1="998.156" x2="1309.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.23" y1="998.156" x2="1309.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="885.196" y1="1405.14" x2="869.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="885.196" y1="1405.14" x2="869.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="885.196" y1="1405.14" x2="901.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="885.196" y1="1405.14" x2="901.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.55" y1="853.956" x2="1441.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.55" y1="853.956" x2="1441.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.55" y1="853.956" x2="1473.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.55" y1="853.956" x2="1473.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="949.006" y1="1440.98" x2="933.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="949.006" y1="1440.98" x2="933.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="949.006" y1="1440.98" x2="965.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="949.006" y1="1440.98" x2="965.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.411" y1="1427.9" x2="829.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.411" y1="1427.9" x2="829.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.411" y1="1427.9" x2="861.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="845.411" y1="1427.9" x2="861.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.57" y1="868.321" x2="1530.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.57" y1="868.321" x2="1530.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.57" y1="868.321" x2="1562.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.57" y1="868.321" x2="1562.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.43" y1="1342.43" x2="834.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.43" y1="1342.43" x2="834.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.43" y1="1342.43" x2="866.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.43" y1="1342.43" x2="866.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.5" y1="829.178" x2="1603.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.5" y1="829.178" x2="1603.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.5" y1="829.178" x2="1635.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1619.5" y1="829.178" x2="1635.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1847.7" y1="804.547" x2="1831.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1847.7" y1="804.547" x2="1831.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1847.7" y1="804.547" x2="1863.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1847.7" y1="804.547" x2="1863.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="992.659" y1="1251.94" x2="976.659" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="992.659" y1="1251.94" x2="976.659" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="992.659" y1="1251.94" x2="1008.66" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="992.659" y1="1251.94" x2="1008.66" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1641.07" y1="860.861" x2="1625.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1641.07" y1="860.861" x2="1625.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1641.07" y1="860.861" x2="1657.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1641.07" y1="860.861" x2="1657.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1301.11" y1="987.604" x2="1285.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1301.11" y1="987.604" x2="1285.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1301.11" y1="987.604" x2="1317.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1301.11" y1="987.604" x2="1317.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1673.06" y1="901.944" x2="1657.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1673.06" y1="901.944" x2="1657.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1673.06" y1="901.944" x2="1689.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1673.06" y1="901.944" x2="1689.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.89" y1="961.872" x2="1595.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.89" y1="961.872" x2="1595.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.89" y1="961.872" x2="1627.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.89" y1="961.872" x2="1627.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="782.176" y1="1390.69" x2="766.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="782.176" y1="1390.69" x2="766.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="782.176" y1="1390.69" x2="798.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="782.176" y1="1390.69" x2="798.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.77" y1="921.537" x2="1659.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.77" y1="921.537" x2="1659.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.77" y1="921.537" x2="1691.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.77" y1="921.537" x2="1691.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="816.257" y1="1346.85" x2="800.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="816.257" y1="1346.85" x2="800.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="816.257" y1="1346.85" x2="832.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="816.257" y1="1346.85" x2="832.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="724.712" y1="1453.02" x2="708.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="724.712" y1="1453.02" x2="708.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="724.712" y1="1453.02" x2="740.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="724.712" y1="1453.02" x2="740.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.34" y1="968.47" x2="1659.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.34" y1="968.47" x2="1659.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.34" y1="968.47" x2="1691.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1675.34" y1="968.47" x2="1691.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.396" y1="1435.49" x2="858.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.396" y1="1435.49" x2="858.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.396" y1="1435.49" x2="890.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.396" y1="1435.49" x2="890.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1330.6" y1="1013.95" x2="1314.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1330.6" y1="1013.95" x2="1314.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1330.6" y1="1013.95" x2="1346.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1330.6" y1="1013.95" x2="1346.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1626.12" y1="969.424" x2="1610.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1626.12" y1="969.424" x2="1610.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1626.12" y1="969.424" x2="1642.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1626.12" y1="969.424" x2="1642.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="933.271" y1="1353.8" x2="917.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="933.271" y1="1353.8" x2="917.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="933.271" y1="1353.8" x2="949.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="933.271" y1="1353.8" x2="949.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="906.639" y1="1451.01" x2="890.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="906.639" y1="1451.01" x2="890.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="906.639" y1="1451.01" x2="922.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="906.639" y1="1451.01" x2="922.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="958.996" y1="1406.94" x2="942.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="958.996" y1="1406.94" x2="942.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="958.996" y1="1406.94" x2="974.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="958.996" y1="1406.94" x2="974.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.277" y1="1312.14" x2="858.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.277" y1="1312.14" x2="858.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.277" y1="1312.14" x2="890.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="874.277" y1="1312.14" x2="890.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.691" y1="1424.89" x2="911.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.691" y1="1424.89" x2="911.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.691" y1="1424.89" x2="943.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.691" y1="1424.89" x2="943.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="538.801" y1="1550.28" x2="522.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="538.801" y1="1550.28" x2="522.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="538.801" y1="1550.28" x2="554.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="538.801" y1="1550.28" x2="554.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="869.402" y1="1512" x2="853.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="869.402" y1="1512" x2="853.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="869.402" y1="1512" x2="885.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="869.402" y1="1512" x2="885.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.688" y1="1677.8" x2="553.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.688" y1="1677.8" x2="553.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.688" y1="1677.8" x2="585.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.688" y1="1677.8" x2="585.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="915.573" y1="1448.5" x2="899.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="915.573" y1="1448.5" x2="899.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="915.573" y1="1448.5" x2="931.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="915.573" y1="1448.5" x2="931.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1520.69" y1="868.39" x2="1504.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1520.69" y1="868.39" x2="1504.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1520.69" y1="868.39" x2="1536.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1520.69" y1="868.39" x2="1536.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.79" y1="1007.52" x2="1446.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.79" y1="1007.52" x2="1446.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.79" y1="1007.52" x2="1478.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1462.79" y1="1007.52" x2="1478.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.71" y1="986.213" x2="1427.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.71" y1="986.213" x2="1427.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.71" y1="986.213" x2="1459.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.71" y1="986.213" x2="1459.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="831.587" y1="1424.87" x2="815.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="831.587" y1="1424.87" x2="815.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="831.587" y1="1424.87" x2="847.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="831.587" y1="1424.87" x2="847.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.93" y1="1006.01" x2="1593.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.93" y1="1006.01" x2="1593.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.93" y1="1006.01" x2="1625.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.93" y1="1006.01" x2="1625.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1733.98" y1="723.632" x2="1717.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1733.98" y1="723.632" x2="1717.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1733.98" y1="723.632" x2="1749.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1733.98" y1="723.632" x2="1749.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.936" y1="1351.49" x2="834.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.936" y1="1351.49" x2="834.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.936" y1="1351.49" x2="866.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="850.936" y1="1351.49" x2="866.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.92" y1="896.324" x2="1752.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.92" y1="896.324" x2="1752.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.92" y1="896.324" x2="1784.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.92" y1="896.324" x2="1784.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.86" y1="1264.5" x2="984.861" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.86" y1="1264.5" x2="984.861" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.48" y1="973.842" x2="1509.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.48" y1="973.842" x2="1509.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.48" y1="973.842" x2="1541.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.48" y1="973.842" x2="1541.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="814.5" y1="1271.95" x2="798.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="814.5" y1="1271.95" x2="798.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="814.5" y1="1271.95" x2="830.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="814.5" y1="1271.95" x2="830.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1834.69" y1="763.508" x2="1818.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1834.69" y1="763.508" x2="1818.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1834.69" y1="763.508" x2="1850.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1834.69" y1="763.508" x2="1850.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.04" y1="1278.88" x2="990.041" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.04" y1="1278.88" x2="990.041" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1424" y1="1023.62" x2="1408" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1424" y1="1023.62" x2="1408" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1424" y1="1023.62" x2="1440" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1424" y1="1023.62" x2="1440" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1618.21" y1="979.458" x2="1602.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1618.21" y1="979.458" x2="1602.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1618.21" y1="979.458" x2="1634.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1618.21" y1="979.458" x2="1634.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1005.39" y1="1284.51" x2="989.392" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1005.39" y1="1284.51" x2="989.392" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.98" y1="894.258" x2="1593.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.98" y1="894.258" x2="1593.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.98" y1="894.258" x2="1625.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1609.98" y1="894.258" x2="1625.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1597.07" y1="924.794" x2="1581.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1597.07" y1="924.794" x2="1581.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1597.07" y1="924.794" x2="1613.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1597.07" y1="924.794" x2="1613.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="882.48" y1="1367.74" x2="866.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="882.48" y1="1367.74" x2="866.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="882.48" y1="1367.74" x2="898.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="882.48" y1="1367.74" x2="898.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1527.95" y1="988.132" x2="1511.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1527.95" y1="988.132" x2="1511.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1527.95" y1="988.132" x2="1543.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1527.95" y1="988.132" x2="1543.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1766.37" y1="756.534" x2="1750.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1766.37" y1="756.534" x2="1750.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1766.37" y1="756.534" x2="1782.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1766.37" y1="756.534" x2="1782.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1422.2" y1="999.716" x2="1406.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1422.2" y1="999.716" x2="1406.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1422.2" y1="999.716" x2="1438.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1422.2" y1="999.716" x2="1438.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1381.71" y1="975.726" x2="1365.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1381.71" y1="975.726" x2="1365.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1381.71" y1="975.726" x2="1397.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1381.71" y1="975.726" x2="1397.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.037" y1="1422.12" x2="859.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.037" y1="1422.12" x2="859.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.037" y1="1422.12" x2="891.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.037" y1="1422.12" x2="891.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="898.259" y1="1319.39" x2="882.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="898.259" y1="1319.39" x2="882.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="898.259" y1="1319.39" x2="914.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="898.259" y1="1319.39" x2="914.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.19" y1="815.289" x2="1746.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.19" y1="815.289" x2="1746.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.19" y1="815.289" x2="1778.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.19" y1="815.289" x2="1778.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="843.736" y1="1421.71" x2="827.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="843.736" y1="1421.71" x2="827.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="843.736" y1="1421.71" x2="859.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="843.736" y1="1421.71" x2="859.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1399.57" y1="992.596" x2="1383.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1399.57" y1="992.596" x2="1383.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1399.57" y1="992.596" x2="1415.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1399.57" y1="992.596" x2="1415.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="712.506" y1="1466.98" x2="696.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="712.506" y1="1466.98" x2="696.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="712.506" y1="1466.98" x2="728.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="712.506" y1="1466.98" x2="728.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="996.845" y1="1298.95" x2="980.845" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="996.845" y1="1298.95" x2="980.845" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="996.845" y1="1298.95" x2="1012.85" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="996.845" y1="1298.95" x2="1012.85" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="764.769" y1="1428.81" x2="748.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="764.769" y1="1428.81" x2="748.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="764.769" y1="1428.81" x2="780.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="764.769" y1="1428.81" x2="780.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1481.71" y1="979.429" x2="1465.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1481.71" y1="979.429" x2="1465.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1481.71" y1="979.429" x2="1497.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1481.71" y1="979.429" x2="1497.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="794.022" y1="1392.87" x2="778.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="794.022" y1="1392.87" x2="778.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="794.022" y1="1392.87" x2="810.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="794.022" y1="1392.87" x2="810.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="989.1" y1="1361.68" x2="973.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="989.1" y1="1361.68" x2="973.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="989.1" y1="1361.68" x2="1005.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="989.1" y1="1361.68" x2="1005.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="658.223" y1="1594.72" x2="642.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="658.223" y1="1594.72" x2="642.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="658.223" y1="1594.72" x2="674.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="658.223" y1="1594.72" x2="674.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.13" y1="922.6" x2="1595.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.13" y1="922.6" x2="1595.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.13" y1="922.6" x2="1627.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1611.13" y1="922.6" x2="1627.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="968.976" y1="1423.08" x2="952.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="968.976" y1="1423.08" x2="952.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="968.976" y1="1423.08" x2="984.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="968.976" y1="1423.08" x2="984.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1451.83" y1="926.379" x2="1435.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1451.83" y1="926.379" x2="1435.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1451.83" y1="926.379" x2="1467.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1451.83" y1="926.379" x2="1467.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="908.822" y1="1392.95" x2="892.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="908.822" y1="1392.95" x2="892.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="908.822" y1="1392.95" x2="924.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="908.822" y1="1392.95" x2="924.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.57" y1="1011.37" x2="1484.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.57" y1="1011.37" x2="1484.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.57" y1="1011.37" x2="1516.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1500.57" y1="1011.37" x2="1516.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1002.47" y1="1234.46" x2="986.471" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1002.47" y1="1234.46" x2="986.471" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.95" y1="1013.72" x2="1442.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.95" y1="1013.72" x2="1442.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.95" y1="1013.72" x2="1474.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1458.95" y1="1013.72" x2="1474.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="892.734" y1="1460.48" x2="876.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="892.734" y1="1460.48" x2="876.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="892.734" y1="1460.48" x2="908.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="892.734" y1="1460.48" x2="908.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.79" y1="924.706" x2="1529.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.79" y1="924.706" x2="1529.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.79" y1="924.706" x2="1561.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.79" y1="924.706" x2="1561.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1003.69" y1="1262.65" x2="987.692" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1003.69" y1="1262.65" x2="987.692" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="862.589" y1="1433.49" x2="846.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="862.589" y1="1433.49" x2="846.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="862.589" y1="1433.49" x2="878.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="862.589" y1="1433.49" x2="878.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="522.504" y1="1512.98" x2="506.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="522.504" y1="1512.98" x2="506.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="522.504" y1="1512.98" x2="538.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="522.504" y1="1512.98" x2="538.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.208" y1="1309.37" x2="956.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.208" y1="1309.37" x2="956.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.208" y1="1309.37" x2="988.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.208" y1="1309.37" x2="988.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.36" y1="1054" x2="1160.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.36" y1="1054" x2="1160.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.36" y1="1054" x2="1192.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.36" y1="1054" x2="1192.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.451" y1="1406.48" x2="728.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.451" y1="1406.48" x2="728.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.451" y1="1406.48" x2="760.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.451" y1="1406.48" x2="760.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1012.88" y1="1293.81" x2="996.879" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1012.88" y1="1293.81" x2="996.879" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="277.68" y1="1755.15" x2="261.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="277.68" y1="1755.15" x2="261.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="277.68" y1="1755.15" x2="293.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="277.68" y1="1755.15" x2="293.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1498.04" y1="1011.64" x2="1482.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1498.04" y1="1011.64" x2="1482.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1498.04" y1="1011.64" x2="1514.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1498.04" y1="1011.64" x2="1514.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.19" y1="842.118" x2="1764.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.19" y1="842.118" x2="1764.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.19" y1="842.118" x2="1796.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.19" y1="842.118" x2="1796.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="855.143" y1="1461.56" x2="839.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="855.143" y1="1461.56" x2="839.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="855.143" y1="1461.56" x2="871.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="855.143" y1="1461.56" x2="871.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="941.026" y1="1398.61" x2="925.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="941.026" y1="1398.61" x2="925.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="941.026" y1="1398.61" x2="957.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="941.026" y1="1398.61" x2="957.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1491.9" y1="862.19" x2="1475.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1491.9" y1="862.19" x2="1475.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1491.9" y1="862.19" x2="1507.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1491.9" y1="862.19" x2="1507.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1614.72" y1="834.262" x2="1598.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1614.72" y1="834.262" x2="1598.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1614.72" y1="834.262" x2="1630.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1614.72" y1="834.262" x2="1630.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.83" y1="933.105" x2="1599.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.83" y1="933.105" x2="1599.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.83" y1="933.105" x2="1631.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.83" y1="933.105" x2="1631.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.951" y1="1271.4" x2="883.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.951" y1="1271.4" x2="883.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.951" y1="1271.4" x2="915.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.951" y1="1271.4" x2="915.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1651.8" y1="933.534" x2="1635.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1651.8" y1="933.534" x2="1635.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1651.8" y1="933.534" x2="1667.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1651.8" y1="933.534" x2="1667.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.2" y1="982.773" x2="1530.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.2" y1="982.773" x2="1530.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.2" y1="982.773" x2="1562.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1546.2" y1="982.773" x2="1562.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="798.462" y1="1475.78" x2="782.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="798.462" y1="1475.78" x2="782.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="798.462" y1="1475.78" x2="814.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="798.462" y1="1475.78" x2="814.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1653.74" y1="920.543" x2="1637.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1653.74" y1="920.543" x2="1637.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1653.74" y1="920.543" x2="1669.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1653.74" y1="920.543" x2="1669.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1550.14" y1="1004.56" x2="1534.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1550.14" y1="1004.56" x2="1534.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1550.14" y1="1004.56" x2="1566.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1550.14" y1="1004.56" x2="1566.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="955.148" y1="1370.3" x2="939.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="955.148" y1="1370.3" x2="939.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="955.148" y1="1370.3" x2="971.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="955.148" y1="1370.3" x2="971.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1845.45" y1="659.705" x2="1829.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1845.45" y1="659.705" x2="1829.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1845.45" y1="659.705" x2="1861.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1845.45" y1="659.705" x2="1861.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1817.11" y1="731.36" x2="1801.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1817.11" y1="731.36" x2="1801.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1817.11" y1="731.36" x2="1833.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1817.11" y1="731.36" x2="1833.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1838.37" y1="806.858" x2="1822.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1838.37" y1="806.858" x2="1822.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1838.37" y1="806.858" x2="1854.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1838.37" y1="806.858" x2="1854.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244" y1="1263.65" x2="1228" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244" y1="1263.65" x2="1228" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244" y1="1263.65" x2="1260" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244" y1="1263.65" x2="1260" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1558.7" y1="965.028" x2="1542.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1558.7" y1="965.028" x2="1542.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1558.7" y1="965.028" x2="1574.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1558.7" y1="965.028" x2="1574.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.76" y1="970.455" x2="1458.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.76" y1="970.455" x2="1458.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.76" y1="970.455" x2="1490.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.76" y1="970.455" x2="1490.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1682.62" y1="820.381" x2="1666.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1682.62" y1="820.381" x2="1666.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1682.62" y1="820.381" x2="1698.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1682.62" y1="820.381" x2="1698.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="872.16" y1="1370.98" x2="856.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="872.16" y1="1370.98" x2="856.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="872.16" y1="1370.98" x2="888.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="872.16" y1="1370.98" x2="888.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="866.851" y1="1331.78" x2="850.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="866.851" y1="1331.78" x2="850.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="866.851" y1="1331.78" x2="882.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="866.851" y1="1331.78" x2="882.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1604.79" y1="991.27" x2="1588.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1604.79" y1="991.27" x2="1588.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1604.79" y1="991.27" x2="1620.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1604.79" y1="991.27" x2="1620.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="436.181" y1="1793.59" x2="420.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="436.181" y1="1793.59" x2="420.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="436.181" y1="1793.59" x2="452.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="436.181" y1="1793.59" x2="452.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="957.079" y1="1394.58" x2="941.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="957.079" y1="1394.58" x2="941.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="957.079" y1="1394.58" x2="973.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="957.079" y1="1394.58" x2="973.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.366" y1="1495.39" x2="775.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.366" y1="1495.39" x2="775.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.366" y1="1495.39" x2="807.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.366" y1="1495.39" x2="807.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1636.9" y1="890.133" x2="1620.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1636.9" y1="890.133" x2="1620.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1636.9" y1="890.133" x2="1652.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1636.9" y1="890.133" x2="1652.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1256" y1="1205.58" x2="1240" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1256" y1="1205.58" x2="1240" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1256" y1="1205.58" x2="1272" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1256" y1="1205.58" x2="1272" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="721.091" y1="1561.75" x2="705.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="721.091" y1="1561.75" x2="705.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="721.091" y1="1561.75" x2="737.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="721.091" y1="1561.75" x2="737.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="823.473" y1="1446.58" x2="807.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="823.473" y1="1446.58" x2="807.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="823.473" y1="1446.58" x2="839.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="823.473" y1="1446.58" x2="839.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="887.822" y1="1375.21" x2="871.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="887.822" y1="1375.21" x2="871.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="887.822" y1="1375.21" x2="903.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="887.822" y1="1375.21" x2="903.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.39" y1="1014.06" x2="1468.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.39" y1="1014.06" x2="1468.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.39" y1="1014.06" x2="1500.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1484.39" y1="1014.06" x2="1500.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1685.1" y1="924.047" x2="1669.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1685.1" y1="924.047" x2="1669.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1685.1" y1="924.047" x2="1701.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1685.1" y1="924.047" x2="1701.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="877.025" y1="1403.06" x2="861.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="877.025" y1="1403.06" x2="861.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="877.025" y1="1403.06" x2="893.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="877.025" y1="1403.06" x2="893.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1473.39" y1="1000.62" x2="1457.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1473.39" y1="1000.62" x2="1457.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1473.39" y1="1000.62" x2="1489.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1473.39" y1="1000.62" x2="1489.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="930.593" y1="1251.3" x2="914.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="930.593" y1="1251.3" x2="914.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="930.593" y1="1251.3" x2="946.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="930.593" y1="1251.3" x2="946.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.55" y1="972.554" x2="1599.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.55" y1="972.554" x2="1599.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.55" y1="972.554" x2="1631.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1615.55" y1="972.554" x2="1631.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1395.64" y1="960.36" x2="1379.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1395.64" y1="960.36" x2="1379.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1395.64" y1="960.36" x2="1411.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1395.64" y1="960.36" x2="1411.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.08" y1="1312.36" x2="984.075" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.08" y1="1312.36" x2="984.075" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="805.605" y1="1368.59" x2="789.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="805.605" y1="1368.59" x2="789.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="805.605" y1="1368.59" x2="821.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="805.605" y1="1368.59" x2="821.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="735.913" y1="1350.58" x2="719.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="735.913" y1="1350.58" x2="719.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="735.913" y1="1350.58" x2="751.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="735.913" y1="1350.58" x2="751.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1826.24" y1="921.284" x2="1810.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1826.24" y1="921.284" x2="1810.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1826.24" y1="921.284" x2="1842.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1826.24" y1="921.284" x2="1842.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1572.67" y1="995.733" x2="1556.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1572.67" y1="995.733" x2="1556.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1572.67" y1="995.733" x2="1588.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1572.67" y1="995.733" x2="1588.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1469" y1="974.22" x2="1453" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1469" y1="974.22" x2="1453" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1469" y1="974.22" x2="1485" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1469" y1="974.22" x2="1485" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.64" y1="1000.17" x2="1752.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.64" y1="1000.17" x2="1752.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.64" y1="1000.17" x2="1784.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1768.64" y1="1000.17" x2="1784.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.6" y1="944.573" x2="1470.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.6" y1="944.573" x2="1470.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.6" y1="944.573" x2="1502.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1486.6" y1="944.573" x2="1502.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.81" y1="1008.74" x2="1427.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.81" y1="1008.74" x2="1427.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.81" y1="1008.74" x2="1459.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1443.81" y1="1008.74" x2="1459.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1628.36" y1="926.189" x2="1612.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1628.36" y1="926.189" x2="1612.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1628.36" y1="926.189" x2="1644.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1628.36" y1="926.189" x2="1644.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.938" y1="1534.93" x2="553.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.938" y1="1534.93" x2="553.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.938" y1="1534.93" x2="585.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="569.938" y1="1534.93" x2="585.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.218" y1="1483.05" x2="728.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.218" y1="1483.05" x2="728.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.218" y1="1483.05" x2="760.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="744.218" y1="1483.05" x2="760.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1713.33" y1="929.721" x2="1697.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1713.33" y1="929.721" x2="1697.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1713.33" y1="929.721" x2="1729.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1713.33" y1="929.721" x2="1729.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="610.516" y1="1568.23" x2="594.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="610.516" y1="1568.23" x2="594.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="610.516" y1="1568.23" x2="626.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="610.516" y1="1568.23" x2="626.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1702.63" y1="949.089" x2="1686.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1702.63" y1="949.089" x2="1686.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1702.63" y1="949.089" x2="1718.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1702.63" y1="949.089" x2="1718.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1799.12" y1="921.26" x2="1783.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1799.12" y1="921.26" x2="1783.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1799.12" y1="921.26" x2="1815.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1799.12" y1="921.26" x2="1815.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1539.91" y1="992.445" x2="1523.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1539.91" y1="992.445" x2="1523.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1539.91" y1="992.445" x2="1555.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1539.91" y1="992.445" x2="1555.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.824" y1="1370.17" x2="947.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.824" y1="1370.17" x2="947.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.824" y1="1370.17" x2="979.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.824" y1="1370.17" x2="979.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="694.471" y1="1399.51" x2="678.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="694.471" y1="1399.51" x2="678.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="694.471" y1="1399.51" x2="710.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="694.471" y1="1399.51" x2="710.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.32" y1="993.037" x2="1462.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.32" y1="993.037" x2="1462.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.32" y1="993.037" x2="1494.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1478.32" y1="993.037" x2="1494.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.95" y1="1279" x2="1264.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.95" y1="1279" x2="1264.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.95" y1="1279" x2="1296.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.95" y1="1279" x2="1296.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1650.12" y1="918.653" x2="1634.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1650.12" y1="918.653" x2="1634.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1650.12" y1="918.653" x2="1666.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1650.12" y1="918.653" x2="1666.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1860.89" y1="778.201" x2="1844.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1860.89" y1="778.201" x2="1844.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1860.89" y1="778.201" x2="1876.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1860.89" y1="778.201" x2="1876.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1444.37" y1="1001.52" x2="1428.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1444.37" y1="1001.52" x2="1428.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1444.37" y1="1001.52" x2="1460.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1444.37" y1="1001.52" x2="1460.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.75" y1="1012.66" x2="1412.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.75" y1="1012.66" x2="1412.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.75" y1="1012.66" x2="1444.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.75" y1="1012.66" x2="1444.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1896.85" y1="631.005" x2="1880.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1896.85" y1="631.005" x2="1880.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1896.85" y1="631.005" x2="1912.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1896.85" y1="631.005" x2="1912.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1863.03" y1="689.163" x2="1847.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1863.03" y1="689.163" x2="1847.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1863.03" y1="689.163" x2="1879.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1863.03" y1="689.163" x2="1879.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="973.556" y1="1249.11" x2="957.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="973.556" y1="1249.11" x2="957.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="973.556" y1="1249.11" x2="989.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="973.556" y1="1249.11" x2="989.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.42" y1="974.507" x2="1399.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.42" y1="974.507" x2="1399.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.42" y1="974.507" x2="1431.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1415.42" y1="974.507" x2="1431.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="961.172" y1="1297.9" x2="945.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="961.172" y1="1297.9" x2="945.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="961.172" y1="1297.9" x2="977.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="961.172" y1="1297.9" x2="977.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="828.067" y1="1395.72" x2="812.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="828.067" y1="1395.72" x2="812.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="828.067" y1="1395.72" x2="844.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="828.067" y1="1395.72" x2="844.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="812.031" y1="1484.71" x2="796.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="812.031" y1="1484.71" x2="796.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="812.031" y1="1484.71" x2="828.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="812.031" y1="1484.71" x2="828.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="901.731" y1="1295.1" x2="885.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="901.731" y1="1295.1" x2="885.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="901.731" y1="1295.1" x2="917.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="901.731" y1="1295.1" x2="917.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.853" y1="1331.79" x2="956.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.853" y1="1331.79" x2="956.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.853" y1="1331.79" x2="988.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="972.853" y1="1331.79" x2="988.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.97" y1="912.119" x2="1722.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.97" y1="912.119" x2="1722.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.97" y1="912.119" x2="1754.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.97" y1="912.119" x2="1754.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1534.29" y1="956.26" x2="1518.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1534.29" y1="956.26" x2="1518.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1534.29" y1="956.26" x2="1550.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1534.29" y1="956.26" x2="1550.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.177" y1="1437.3" x2="863.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.177" y1="1437.3" x2="863.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.177" y1="1437.3" x2="895.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.177" y1="1437.3" x2="895.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.308" y1="1302.19" x2="775.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.308" y1="1302.19" x2="775.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.308" y1="1302.19" x2="807.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="791.308" y1="1302.19" x2="807.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.44" y1="890.561" x2="1574.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.44" y1="890.561" x2="1574.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.44" y1="890.561" x2="1606.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.44" y1="890.561" x2="1606.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1729.02" y1="853.811" x2="1713.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1729.02" y1="853.811" x2="1713.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1729.02" y1="853.811" x2="1745.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1729.02" y1="853.811" x2="1745.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="936.651" y1="1397.2" x2="920.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="936.651" y1="1397.2" x2="920.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="936.651" y1="1397.2" x2="952.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="936.651" y1="1397.2" x2="952.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="856.289" y1="1386.44" x2="840.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="856.289" y1="1386.44" x2="840.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="856.289" y1="1386.44" x2="872.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="856.289" y1="1386.44" x2="872.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1350.38" y1="930.437" x2="1334.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1350.38" y1="930.437" x2="1334.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1350.38" y1="930.437" x2="1366.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1350.38" y1="930.437" x2="1366.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1529.58" y1="996.113" x2="1513.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1529.58" y1="996.113" x2="1513.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1529.58" y1="996.113" x2="1545.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1529.58" y1="996.113" x2="1545.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.47" y1="955.973" x2="1497.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.47" y1="955.973" x2="1497.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.47" y1="955.973" x2="1529.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.47" y1="955.973" x2="1529.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1530.22" y1="1008.88" x2="1514.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1530.22" y1="1008.88" x2="1514.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1530.22" y1="1008.88" x2="1546.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1530.22" y1="1008.88" x2="1546.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.75" y1="828.151" x2="1574.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.75" y1="828.151" x2="1574.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.75" y1="828.151" x2="1606.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1590.75" y1="828.151" x2="1606.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1942.19" y1="883.328" x2="1926.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1942.19" y1="883.328" x2="1926.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1942.19" y1="883.328" x2="1958.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1942.19" y1="883.328" x2="1958.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="800.484" y1="1327.86" x2="784.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="800.484" y1="1327.86" x2="784.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="800.484" y1="1327.86" x2="816.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="800.484" y1="1327.86" x2="816.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.479" y1="1361.21" x2="863.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.479" y1="1361.21" x2="863.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.479" y1="1361.21" x2="895.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="879.479" y1="1361.21" x2="895.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1621.59" y1="888.584" x2="1605.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1621.59" y1="888.584" x2="1605.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1621.59" y1="888.584" x2="1637.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1621.59" y1="888.584" x2="1637.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.03" y1="917.837" x2="1589.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.03" y1="917.837" x2="1589.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.03" y1="917.837" x2="1621.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1605.03" y1="917.837" x2="1621.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="863.983" y1="1256.2" x2="847.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="863.983" y1="1256.2" x2="847.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="863.983" y1="1256.2" x2="879.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="863.983" y1="1256.2" x2="879.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1763.1" y1="789.55" x2="1747.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1763.1" y1="789.55" x2="1747.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1763.1" y1="789.55" x2="1779.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1763.1" y1="789.55" x2="1779.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1840.32" y1="839.458" x2="1824.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1840.32" y1="839.458" x2="1824.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1840.32" y1="839.458" x2="1856.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1840.32" y1="839.458" x2="1856.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1566.73" y1="937.472" x2="1550.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1566.73" y1="937.472" x2="1550.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1566.73" y1="937.472" x2="1582.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1566.73" y1="937.472" x2="1582.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.17" y1="997.851" x2="1441.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.17" y1="997.851" x2="1441.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.17" y1="997.851" x2="1473.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.17" y1="997.851" x2="1473.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="725.698" y1="1501.31" x2="709.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="725.698" y1="1501.31" x2="709.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="725.698" y1="1501.31" x2="741.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="725.698" y1="1501.31" x2="741.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="804.197" y1="1345.85" x2="788.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="804.197" y1="1345.85" x2="788.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="804.197" y1="1345.85" x2="820.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="804.197" y1="1345.85" x2="820.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.44" y1="832.688" x2="1722.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.44" y1="832.688" x2="1722.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.44" y1="832.688" x2="1754.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1738.44" y1="832.688" x2="1754.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1575.07" y1="927.362" x2="1559.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1575.07" y1="927.362" x2="1559.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1575.07" y1="927.362" x2="1591.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1575.07" y1="927.362" x2="1591.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.369" y1="1393.26" x2="947.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.369" y1="1393.26" x2="947.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.369" y1="1393.26" x2="979.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.369" y1="1393.26" x2="979.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1667.21" y1="967.122" x2="1651.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1667.21" y1="967.122" x2="1651.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1667.21" y1="967.122" x2="1683.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1667.21" y1="967.122" x2="1683.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1808.12" y1="803.64" x2="1792.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1808.12" y1="803.64" x2="1792.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1808.12" y1="803.64" x2="1824.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1808.12" y1="803.64" x2="1824.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.457" y1="1364.93" x2="911.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.457" y1="1364.93" x2="911.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.457" y1="1364.93" x2="943.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="927.457" y1="1364.93" x2="943.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.588" y1="1394.79" x2="938.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.588" y1="1394.79" x2="938.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.588" y1="1394.79" x2="970.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.588" y1="1394.79" x2="970.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1600.82" y1="848.267" x2="1584.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1600.82" y1="848.267" x2="1584.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1600.82" y1="848.267" x2="1616.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1600.82" y1="848.267" x2="1616.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1013.05" y1="1254.21" x2="997.055" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1013.05" y1="1254.21" x2="997.055" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="710.768" y1="1471.31" x2="694.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="710.768" y1="1471.31" x2="694.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="710.768" y1="1471.31" x2="726.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="710.768" y1="1471.31" x2="726.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.455" y1="1241.66" x2="938.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.455" y1="1241.66" x2="938.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.455" y1="1241.66" x2="970.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="954.455" y1="1241.66" x2="970.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.19" y1="997.572" x2="1412.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.19" y1="997.572" x2="1412.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.19" y1="997.572" x2="1444.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1428.19" y1="997.572" x2="1444.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="907.609" y1="1313.11" x2="891.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="907.609" y1="1313.11" x2="891.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="907.609" y1="1313.11" x2="923.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="907.609" y1="1313.11" x2="923.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="904.55" y1="1372.25" x2="888.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="904.55" y1="1372.25" x2="888.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="904.55" y1="1372.25" x2="920.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="904.55" y1="1372.25" x2="920.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1561.34" y1="914.945" x2="1545.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1561.34" y1="914.945" x2="1545.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1561.34" y1="914.945" x2="1577.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1561.34" y1="914.945" x2="1577.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1311.21" y1="973.579" x2="1295.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1311.21" y1="973.579" x2="1295.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1311.21" y1="973.579" x2="1327.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1311.21" y1="973.579" x2="1327.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="743.908" y1="1344.73" x2="727.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="743.908" y1="1344.73" x2="727.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="743.908" y1="1344.73" x2="759.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="743.908" y1="1344.73" x2="759.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.408" y1="1314.39" x2="947.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.408" y1="1314.39" x2="947.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.408" y1="1314.39" x2="979.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="963.408" y1="1314.39" x2="979.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="994.563" y1="1220.85" x2="978.563" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="994.563" y1="1220.85" x2="978.563" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="994.563" y1="1220.85" x2="1010.56" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="994.563" y1="1220.85" x2="1010.56" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.07" y1="993.346" x2="1764.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.07" y1="993.346" x2="1764.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.07" y1="993.346" x2="1796.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1780.07" y1="993.346" x2="1796.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="983.358" y1="1291.75" x2="967.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="983.358" y1="1291.75" x2="967.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="983.358" y1="1291.75" x2="999.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="983.358" y1="1291.75" x2="999.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1496.43" y1="924.203" x2="1480.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1496.43" y1="924.203" x2="1480.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1496.43" y1="924.203" x2="1512.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1496.43" y1="924.203" x2="1512.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="938.137" y1="1395.53" x2="922.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="938.137" y1="1395.53" x2="922.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="938.137" y1="1395.53" x2="954.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="938.137" y1="1395.53" x2="954.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="745.679" y1="1455.46" x2="729.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="745.679" y1="1455.46" x2="729.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="745.679" y1="1455.46" x2="761.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="745.679" y1="1455.46" x2="761.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1226.7" y1="1172" x2="1210.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1226.7" y1="1172" x2="1210.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1226.7" y1="1172" x2="1242.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1226.7" y1="1172" x2="1242.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.77" y1="884.258" x2="1746.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.77" y1="884.258" x2="1746.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.77" y1="884.258" x2="1778.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1762.77" y1="884.258" x2="1778.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="977.334" y1="1349.1" x2="961.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="977.334" y1="1349.1" x2="961.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="977.334" y1="1349.1" x2="993.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="977.334" y1="1349.1" x2="993.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.292" y1="1482.58" x2="859.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.292" y1="1482.58" x2="859.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.292" y1="1482.58" x2="891.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="875.292" y1="1482.58" x2="891.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.433" y1="1329.29" x2="883.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.433" y1="1329.29" x2="883.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.433" y1="1329.29" x2="915.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="899.433" y1="1329.29" x2="915.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1552.62" y1="936.348" x2="1536.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1552.62" y1="936.348" x2="1536.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1552.62" y1="936.348" x2="1568.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1552.62" y1="936.348" x2="1568.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.54" y1="984.865" x2="1456.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.54" y1="984.865" x2="1456.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.54" y1="984.865" x2="1488.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1472.54" y1="984.865" x2="1488.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165" y1="1027.31" x2="1149" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165" y1="1027.31" x2="1149" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165" y1="1027.31" x2="1181" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1165" y1="1027.31" x2="1181" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239" y1="1106.12" x2="1223" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239" y1="1106.12" x2="1223" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239" y1="1106.12" x2="1255" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1239" y1="1106.12" x2="1255" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="839.379" y1="1287.74" x2="823.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="839.379" y1="1287.74" x2="823.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="839.379" y1="1287.74" x2="855.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="839.379" y1="1287.74" x2="855.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414" y1="1170.91" x2="1398" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414" y1="1170.91" x2="1398" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414" y1="1170.91" x2="1430" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414" y1="1170.91" x2="1430" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="844.925" y1="1383.63" x2="828.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="844.925" y1="1383.63" x2="828.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="844.925" y1="1383.63" x2="860.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip952)" x1="844.925" y1="1383.63" x2="860.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<polyline clip-path="url(#clip952)" style="stroke:#dc143c; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2006.95,814.774 2025.44,792.795 2041.77,771.865 2055.91,752.038 2067.82,733.363 2077.48,715.886 2084.85,699.652 2089.91,684.701 2092.67,671.07 2093.1,658.793 2091.21,647.901 2087.01,638.422 2080.5,630.379 2071.7,623.791 2060.63,618.676 2047.32,615.047 2031.8,612.912 2014.11,612.276 1994.3,613.142 1972.42,615.507 1948.51,619.365 1922.64,624.707 1894.88,631.519 1865.29,639.784 1833.95,649.482 1800.94,660.588 1766.33,673.075 1730.22,686.911 1692.69,702.061 1653.85,718.489 1613.78,736.152 1572.58,755.007 1530.37,775.007 1487.24,796.101 1443.3,818.237 1398.66,841.36 1353.44,865.411 1307.75,890.331 1261.69,916.057 1215.4,942.525 1168.97,969.669 1122.53,997.421 1076.2,1025.71 1030.09,1054.47 984.309,1083.63 938.981,1113.1 894.216,1142.83 850.125,1172.73 806.82,1202.74 764.408,1232.76 722.995,1262.74 682.684,1292.6 643.578,1322.25 605.772,1351.63 569.362,1380.67 534.439,1409.29 501.091,1437.41 469.399,1464.97 439.444,1491.91 411.301,1518.15 385.04,1543.62 360.726,1568.27 338.42,1592.03 318.178,1614.84 300.051,1636.64 284.084,1657.39 270.316,1677.03 258.783,1695.51 249.513,1712.78 242.529,1728.8 237.849,1743.53 235.484,1756.93 235.44,1768.98 237.718,1779.64 242.312,1788.88 249.21,1796.68 258.394,1803.02 269.843,1807.88 283.527,1811.26 299.413,1813.14 317.46,1813.53 337.623,1812.41 359.852,1809.79 384.092,1805.68 410.282,1800.09 438.356,1793.04 468.244,1784.53 499.871,1774.59 533.16,1763.25 568.025,1750.54 604.381,1736.48 642.135,1721.11 681.195,1704.47 721.462,1686.6 762.835,1667.55 805.212,1647.37 848.486,1626.09 892.549,1603.79 937.291,1580.5 982.6,1556.3 1028.36,1531.24 1074.47,1505.39 1120.79,1478.8 1167.23,1451.55 1213.66,1423.7 1259.96,1395.33 1306.02,1366.5 1351.73,1337.28 1396.98,1307.76 1441.64,1278 1485.6,1248.07 1528.77,1218.06 1571.02,1188.03 1612.25,1158.07 1652.37,1128.24 1691.26,1098.63 1728.84,1069.3 1765,1040.33 1799.67,1011.79 1832.74,983.758 1864.15,956.295 1893.8,929.473 1921.64,903.361 1947.57,878.021 1971.56,853.519 1993.52,829.915 2013.41,807.269 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/conic/ellipse_approx/b2c746af.svg b/previews/PR3562/tutorials/conic/ellipse_approx/f9ac62e0.svg
similarity index 57%
rename from previews/PR3562/tutorials/conic/ellipse_approx/b2c746af.svg
rename to previews/PR3562/tutorials/conic/ellipse_approx/f9ac62e0.svg
index 0dcfbb63d15..3350b889067 100644
--- a/previews/PR3562/tutorials/conic/ellipse_approx/b2c746af.svg
+++ b/previews/PR3562/tutorials/conic/ellipse_approx/f9ac62e0.svg
@@ -1,2450 +1,2450 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="600" viewBox="0 0 2400 2400">
 <defs>
-  <clipPath id="clip070">
+  <clipPath id="clip890">
     <rect x="0" y="0" width="2400" height="2400"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip070)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip890)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip071">
+  <clipPath id="clip891">
     <rect x="480" y="240" width="1681" height="1681"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip070)" d="M178.867 2261.17 L2352.76 2261.17 L2352.76 47.2441 L178.867 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip890)" d="M178.867 2261.17 L2352.76 2261.17 L2352.76 47.2441 L178.867 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip072">
+  <clipPath id="clip892">
     <rect x="178" y="47" width="2175" height="2215"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="327.236,2261.17 327.236,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="640.095,2261.17 640.095,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="952.953,2261.17 952.953,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1265.81,2261.17 1265.81,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1578.67,2261.17 1578.67,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1891.53,2261.17 1891.53,47.2441 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2204.39,2261.17 2204.39,47.2441 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 2352.76,2261.17 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="327.236,2261.17 327.236,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="640.095,2261.17 640.095,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="952.953,2261.17 952.953,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1265.81,2261.17 1265.81,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1578.67,2261.17 1578.67,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1891.53,2261.17 1891.53,2242.27 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2204.39,2261.17 2204.39,2242.27 "/>
-<path clip-path="url(#clip070)" d="M281.299 2306.74 L310.975 2306.74 L310.975 2310.67 L281.299 2310.67 L281.299 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M335.234 2304.93 Q338.59 2305.65 340.465 2307.92 Q342.363 2310.19 342.363 2313.52 Q342.363 2318.64 338.845 2321.44 Q335.326 2324.24 328.845 2324.24 Q326.669 2324.24 324.354 2323.8 Q322.063 2323.38 319.609 2322.53 L319.609 2318.01 Q321.553 2319.15 323.868 2319.73 Q326.183 2320.3 328.706 2320.3 Q333.104 2320.3 335.396 2318.57 Q337.711 2316.83 337.711 2313.52 Q337.711 2310.47 335.558 2308.75 Q333.428 2307.02 329.609 2307.02 L325.581 2307.02 L325.581 2303.17 L329.794 2303.17 Q333.243 2303.17 335.072 2301.81 Q336.9 2300.42 336.9 2297.83 Q336.9 2295.17 335.002 2293.75 Q333.127 2292.32 329.609 2292.32 Q327.688 2292.32 325.488 2292.73 Q323.289 2293.15 320.651 2294.03 L320.651 2289.86 Q323.313 2289.12 325.627 2288.75 Q327.965 2288.38 330.025 2288.38 Q335.35 2288.38 338.451 2290.81 Q341.553 2293.22 341.553 2297.34 Q341.553 2300.21 339.91 2302.2 Q338.266 2304.17 335.234 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M361.229 2292.09 Q357.618 2292.09 355.789 2295.65 Q353.984 2299.19 353.984 2306.32 Q353.984 2313.43 355.789 2316.99 Q357.618 2320.54 361.229 2320.54 Q364.863 2320.54 366.669 2316.99 Q368.497 2313.43 368.497 2306.32 Q368.497 2299.19 366.669 2295.65 Q364.863 2292.09 361.229 2292.09 M361.229 2288.38 Q367.039 2288.38 370.095 2292.99 Q373.173 2297.57 373.173 2306.32 Q373.173 2315.05 370.095 2319.66 Q367.039 2324.24 361.229 2324.24 Q355.419 2324.24 352.34 2319.66 Q349.285 2315.05 349.285 2306.32 Q349.285 2297.57 352.34 2292.99 Q355.419 2288.38 361.229 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M594.157 2306.74 L623.833 2306.74 L623.833 2310.67 L594.157 2310.67 L594.157 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M637.953 2319.63 L654.273 2319.63 L654.273 2323.57 L632.328 2323.57 L632.328 2319.63 Q634.99 2316.88 639.574 2312.25 Q644.18 2307.6 645.361 2306.25 Q647.606 2303.73 648.486 2301.99 Q649.388 2300.23 649.388 2298.54 Q649.388 2295.79 647.444 2294.05 Q645.523 2292.32 642.421 2292.32 Q640.222 2292.32 637.768 2293.08 Q635.338 2293.85 632.56 2295.4 L632.56 2290.67 Q635.384 2289.54 637.838 2288.96 Q640.291 2288.38 642.328 2288.38 Q647.699 2288.38 650.893 2291.07 Q654.087 2293.75 654.087 2298.24 Q654.087 2300.37 653.277 2302.29 Q652.49 2304.19 650.384 2306.79 Q649.805 2307.46 646.703 2310.67 Q643.601 2313.87 637.953 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M674.087 2292.09 Q670.476 2292.09 668.648 2295.65 Q666.842 2299.19 666.842 2306.32 Q666.842 2313.43 668.648 2316.99 Q670.476 2320.54 674.087 2320.54 Q677.722 2320.54 679.527 2316.99 Q681.356 2313.43 681.356 2306.32 Q681.356 2299.19 679.527 2295.65 Q677.722 2292.09 674.087 2292.09 M674.087 2288.38 Q679.898 2288.38 682.953 2292.99 Q686.032 2297.57 686.032 2306.32 Q686.032 2315.05 682.953 2319.66 Q679.898 2324.24 674.087 2324.24 Q668.277 2324.24 665.199 2319.66 Q662.143 2315.05 662.143 2306.32 Q662.143 2297.57 665.199 2292.99 Q668.277 2288.38 674.087 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M907.016 2306.74 L936.691 2306.74 L936.691 2310.67 L907.016 2310.67 L907.016 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M947.594 2319.63 L955.233 2319.63 L955.233 2293.27 L946.923 2294.93 L946.923 2290.67 L955.187 2289.01 L959.863 2289.01 L959.863 2319.63 L967.501 2319.63 L967.501 2323.57 L947.594 2323.57 L947.594 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M986.946 2292.09 Q983.335 2292.09 981.506 2295.65 Q979.7 2299.19 979.7 2306.32 Q979.7 2313.43 981.506 2316.99 Q983.335 2320.54 986.946 2320.54 Q990.58 2320.54 992.386 2316.99 Q994.214 2313.43 994.214 2306.32 Q994.214 2299.19 992.386 2295.65 Q990.58 2292.09 986.946 2292.09 M986.946 2288.38 Q992.756 2288.38 995.811 2292.99 Q998.89 2297.57 998.89 2306.32 Q998.89 2315.05 995.811 2319.66 Q992.756 2324.24 986.946 2324.24 Q981.136 2324.24 978.057 2319.66 Q975.001 2315.05 975.001 2306.32 Q975.001 2297.57 978.057 2292.99 Q981.136 2288.38 986.946 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1265.81 2292.09 Q1262.2 2292.09 1260.37 2295.65 Q1258.57 2299.19 1258.57 2306.32 Q1258.57 2313.43 1260.37 2316.99 Q1262.2 2320.54 1265.81 2320.54 Q1269.45 2320.54 1271.25 2316.99 Q1273.08 2313.43 1273.08 2306.32 Q1273.08 2299.19 1271.25 2295.65 Q1269.45 2292.09 1265.81 2292.09 M1265.81 2288.38 Q1271.62 2288.38 1274.68 2292.99 Q1277.76 2297.57 1277.76 2306.32 Q1277.76 2315.05 1274.68 2319.66 Q1271.62 2324.24 1265.81 2324.24 Q1260 2324.24 1256.92 2319.66 Q1253.87 2315.05 1253.87 2306.32 Q1253.87 2297.57 1256.92 2292.99 Q1260 2288.38 1265.81 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1553.36 2319.63 L1561 2319.63 L1561 2293.27 L1552.69 2294.93 L1552.69 2290.67 L1560.95 2289.01 L1565.63 2289.01 L1565.63 2319.63 L1573.26 2319.63 L1573.26 2323.57 L1553.36 2323.57 L1553.36 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1592.71 2292.09 Q1589.1 2292.09 1587.27 2295.65 Q1585.46 2299.19 1585.46 2306.32 Q1585.46 2313.43 1587.27 2316.99 Q1589.1 2320.54 1592.71 2320.54 Q1596.34 2320.54 1598.15 2316.99 Q1599.98 2313.43 1599.98 2306.32 Q1599.98 2299.19 1598.15 2295.65 Q1596.34 2292.09 1592.71 2292.09 M1592.71 2288.38 Q1598.52 2288.38 1601.57 2292.99 Q1604.65 2297.57 1604.65 2306.32 Q1604.65 2315.05 1601.57 2319.66 Q1598.52 2324.24 1592.71 2324.24 Q1586.9 2324.24 1583.82 2319.66 Q1580.76 2315.05 1580.76 2306.32 Q1580.76 2297.57 1583.82 2292.99 Q1586.9 2288.38 1592.71 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1870.3 2319.63 L1886.62 2319.63 L1886.62 2323.57 L1864.68 2323.57 L1864.68 2319.63 Q1867.34 2316.88 1871.92 2312.25 Q1876.53 2307.6 1877.71 2306.25 Q1879.95 2303.73 1880.83 2301.99 Q1881.74 2300.23 1881.74 2298.54 Q1881.74 2295.79 1879.79 2294.05 Q1877.87 2292.32 1874.77 2292.32 Q1872.57 2292.32 1870.12 2293.08 Q1867.69 2293.85 1864.91 2295.4 L1864.91 2290.67 Q1867.73 2289.54 1870.19 2288.96 Q1872.64 2288.38 1874.68 2288.38 Q1880.05 2288.38 1883.24 2291.07 Q1886.44 2293.75 1886.44 2298.24 Q1886.44 2300.37 1885.63 2302.29 Q1884.84 2304.19 1882.73 2306.79 Q1882.15 2307.46 1879.05 2310.67 Q1875.95 2313.87 1870.3 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1906.44 2292.09 Q1902.82 2292.09 1901 2295.65 Q1899.19 2299.19 1899.19 2306.32 Q1899.19 2313.43 1901 2316.99 Q1902.82 2320.54 1906.44 2320.54 Q1910.07 2320.54 1911.88 2316.99 Q1913.7 2313.43 1913.7 2306.32 Q1913.7 2299.19 1911.88 2295.65 Q1910.07 2292.09 1906.44 2292.09 M1906.44 2288.38 Q1912.25 2288.38 1915.3 2292.99 Q1918.38 2297.57 1918.38 2306.32 Q1918.38 2315.05 1915.3 2319.66 Q1912.25 2324.24 1906.44 2324.24 Q1900.63 2324.24 1897.55 2319.66 Q1894.49 2315.05 1894.49 2306.32 Q1894.49 2297.57 1897.55 2292.99 Q1900.63 2288.38 1906.44 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M2193.23 2304.93 Q2196.59 2305.65 2198.46 2307.92 Q2200.36 2310.19 2200.36 2313.52 Q2200.36 2318.64 2196.84 2321.44 Q2193.32 2324.24 2186.84 2324.24 Q2184.66 2324.24 2182.35 2323.8 Q2180.06 2323.38 2177.6 2322.53 L2177.6 2318.01 Q2179.55 2319.15 2181.86 2319.73 Q2184.18 2320.3 2186.7 2320.3 Q2191.1 2320.3 2193.39 2318.57 Q2195.71 2316.83 2195.71 2313.52 Q2195.71 2310.47 2193.55 2308.75 Q2191.42 2307.02 2187.6 2307.02 L2183.58 2307.02 L2183.58 2303.17 L2187.79 2303.17 Q2191.24 2303.17 2193.07 2301.81 Q2194.9 2300.42 2194.9 2297.83 Q2194.9 2295.17 2193 2293.75 Q2191.12 2292.32 2187.6 2292.32 Q2185.68 2292.32 2183.48 2292.73 Q2181.28 2293.15 2178.65 2294.03 L2178.65 2289.86 Q2181.31 2289.12 2183.62 2288.75 Q2185.96 2288.38 2188.02 2288.38 Q2193.34 2288.38 2196.45 2290.81 Q2199.55 2293.22 2199.55 2297.34 Q2199.55 2300.21 2197.9 2302.2 Q2196.26 2304.17 2193.23 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M2219.22 2292.09 Q2215.61 2292.09 2213.78 2295.65 Q2211.98 2299.19 2211.98 2306.32 Q2211.98 2313.43 2213.78 2316.99 Q2215.61 2320.54 2219.22 2320.54 Q2222.86 2320.54 2224.66 2316.99 Q2226.49 2313.43 2226.49 2306.32 Q2226.49 2299.19 2224.66 2295.65 Q2222.86 2292.09 2219.22 2292.09 M2219.22 2288.38 Q2225.03 2288.38 2228.09 2292.99 Q2231.17 2297.57 2231.17 2306.32 Q2231.17 2315.05 2228.09 2319.66 Q2225.03 2324.24 2219.22 2324.24 Q2213.41 2324.24 2210.34 2319.66 Q2207.28 2315.05 2207.28 2306.32 Q2207.28 2297.57 2210.34 2292.99 Q2213.41 2288.38 2219.22 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,2110.07 2352.76,2110.07 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1791.45 2352.76,1791.45 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1472.83 2352.76,1472.83 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1154.21 2352.76,1154.21 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,835.586 2352.76,835.586 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,516.966 2352.76,516.966 "/>
-<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,198.346 2352.76,198.346 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 178.867,47.2441 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2110.07 197.764,2110.07 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1791.45 197.764,1791.45 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1472.83 197.764,1472.83 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1154.21 197.764,1154.21 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,835.586 197.764,835.586 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,516.966 197.764,516.966 "/>
-<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,198.346 197.764,198.346 "/>
-<path clip-path="url(#clip070)" d="M50.9921 2110.52 L80.6679 2110.52 L80.6679 2114.45 L50.9921 2114.45 L50.9921 2110.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M104.927 2108.71 Q108.283 2109.43 110.158 2111.7 Q112.057 2113.97 112.057 2117.3 Q112.057 2122.42 108.538 2125.22 Q105.02 2128.02 98.5382 2128.02 Q96.3623 2128.02 94.0475 2127.58 Q91.7558 2127.16 89.3021 2126.3 L89.3021 2121.79 Q91.2465 2122.92 93.5613 2123.5 Q95.8761 2124.08 98.3993 2124.08 Q102.797 2124.08 105.089 2122.35 Q107.404 2120.61 107.404 2117.3 Q107.404 2114.24 105.251 2112.53 Q103.121 2110.8 99.3021 2110.8 L95.2743 2110.8 L95.2743 2106.95 L99.4872 2106.95 Q102.936 2106.95 104.765 2105.59 Q106.594 2104.2 106.594 2101.61 Q106.594 2098.94 104.696 2097.53 Q102.821 2096.1 99.3021 2096.1 Q97.3808 2096.1 95.1817 2096.51 Q92.9826 2096.93 90.3438 2097.81 L90.3438 2093.64 Q93.0058 2092.9 95.3206 2092.53 Q97.6585 2092.16 99.7187 2092.16 Q105.043 2092.16 108.145 2094.59 Q111.246 2097 111.246 2101.12 Q111.246 2103.99 109.603 2105.98 Q107.959 2107.95 104.927 2108.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 2095.86 Q127.311 2095.86 125.482 2099.43 Q123.677 2102.97 123.677 2110.1 Q123.677 2117.21 125.482 2120.77 Q127.311 2124.31 130.922 2124.31 Q134.556 2124.31 136.362 2120.77 Q138.191 2117.21 138.191 2110.1 Q138.191 2102.97 136.362 2099.43 Q134.556 2095.86 130.922 2095.86 M130.922 2092.16 Q136.732 2092.16 139.788 2096.77 Q142.867 2101.35 142.867 2110.1 Q142.867 2118.83 139.788 2123.43 Q136.732 2128.02 130.922 2128.02 Q125.112 2128.02 122.033 2123.43 Q118.978 2118.83 118.978 2110.1 Q118.978 2101.35 122.033 2096.77 Q125.112 2092.16 130.922 2092.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M50.9921 1791.9 L80.6679 1791.9 L80.6679 1795.83 L50.9921 1795.83 L50.9921 1791.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M94.7882 1804.79 L111.108 1804.79 L111.108 1808.73 L89.1632 1808.73 L89.1632 1804.79 Q91.8252 1802.04 96.4085 1797.41 Q101.015 1792.75 102.196 1791.41 Q104.441 1788.89 105.321 1787.15 Q106.223 1785.39 106.223 1783.7 Q106.223 1780.95 104.279 1779.21 Q102.358 1777.48 99.2558 1777.48 Q97.0567 1777.48 94.603 1778.24 Q92.1725 1779 89.3947 1780.55 L89.3947 1775.83 Q92.2188 1774.7 94.6724 1774.12 Q97.1261 1773.54 99.1632 1773.54 Q104.534 1773.54 107.728 1776.23 Q110.922 1778.91 110.922 1783.4 Q110.922 1785.53 110.112 1787.45 Q109.325 1789.35 107.219 1791.94 Q106.64 1792.61 103.538 1795.83 Q100.436 1799.03 94.7882 1804.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 1777.24 Q127.311 1777.24 125.482 1780.81 Q123.677 1784.35 123.677 1791.48 Q123.677 1798.59 125.482 1802.15 Q127.311 1805.69 130.922 1805.69 Q134.556 1805.69 136.362 1802.15 Q138.191 1798.59 138.191 1791.48 Q138.191 1784.35 136.362 1780.81 Q134.556 1777.24 130.922 1777.24 M130.922 1773.54 Q136.732 1773.54 139.788 1778.15 Q142.867 1782.73 142.867 1791.48 Q142.867 1800.21 139.788 1804.81 Q136.732 1809.4 130.922 1809.4 Q125.112 1809.4 122.033 1804.81 Q118.978 1800.21 118.978 1791.48 Q118.978 1782.73 122.033 1778.15 Q125.112 1773.54 130.922 1773.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M50.9921 1473.28 L80.6679 1473.28 L80.6679 1477.21 L50.9921 1477.21 L50.9921 1473.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M91.5706 1486.17 L99.2095 1486.17 L99.2095 1459.81 L90.8993 1461.47 L90.8993 1457.21 L99.1632 1455.55 L103.839 1455.55 L103.839 1486.17 L111.478 1486.17 L111.478 1490.11 L91.5706 1490.11 L91.5706 1486.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 1458.62 Q127.311 1458.62 125.482 1462.19 Q123.677 1465.73 123.677 1472.86 Q123.677 1479.97 125.482 1483.53 Q127.311 1487.07 130.922 1487.07 Q134.556 1487.07 136.362 1483.53 Q138.191 1479.97 138.191 1472.86 Q138.191 1465.73 136.362 1462.19 Q134.556 1458.62 130.922 1458.62 M130.922 1454.92 Q136.732 1454.92 139.788 1459.53 Q142.867 1464.11 142.867 1472.86 Q142.867 1481.59 139.788 1486.19 Q136.732 1490.78 130.922 1490.78 Q125.112 1490.78 122.033 1486.19 Q118.978 1481.59 118.978 1472.86 Q118.978 1464.11 122.033 1459.53 Q125.112 1454.92 130.922 1454.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 1140 Q127.311 1140 125.482 1143.57 Q123.677 1147.11 123.677 1154.24 Q123.677 1161.35 125.482 1164.91 Q127.311 1168.45 130.922 1168.45 Q134.556 1168.45 136.362 1164.91 Q138.191 1161.35 138.191 1154.24 Q138.191 1147.11 136.362 1143.57 Q134.556 1140 130.922 1140 M130.922 1136.3 Q136.732 1136.3 139.788 1140.91 Q142.867 1145.49 142.867 1154.24 Q142.867 1162.97 139.788 1167.57 Q136.732 1172.16 130.922 1172.16 Q125.112 1172.16 122.033 1167.57 Q118.978 1162.97 118.978 1154.24 Q118.978 1145.49 122.033 1140.91 Q125.112 1136.3 130.922 1136.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M91.5706 848.931 L99.2095 848.931 L99.2095 822.565 L90.8993 824.232 L90.8993 819.973 L99.1632 818.306 L103.839 818.306 L103.839 848.931 L111.478 848.931 L111.478 852.866 L91.5706 852.866 L91.5706 848.931 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 821.385 Q127.311 821.385 125.482 824.949 Q123.677 828.491 123.677 835.621 Q123.677 842.727 125.482 846.292 Q127.311 849.834 130.922 849.834 Q134.556 849.834 136.362 846.292 Q138.191 842.727 138.191 835.621 Q138.191 828.491 136.362 824.949 Q134.556 821.385 130.922 821.385 M130.922 817.681 Q136.732 817.681 139.788 822.287 Q142.867 826.871 142.867 835.621 Q142.867 844.348 139.788 848.954 Q136.732 853.537 130.922 853.537 Q125.112 853.537 122.033 848.954 Q118.978 844.348 118.978 835.621 Q118.978 826.871 122.033 822.287 Q125.112 817.681 130.922 817.681 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M94.7882 530.311 L111.108 530.311 L111.108 534.246 L89.1632 534.246 L89.1632 530.311 Q91.8252 527.556 96.4085 522.927 Q101.015 518.274 102.196 516.931 Q104.441 514.408 105.321 512.672 Q106.223 510.913 106.223 509.223 Q106.223 506.468 104.279 504.732 Q102.358 502.996 99.2558 502.996 Q97.0567 502.996 94.603 503.76 Q92.1725 504.524 89.3947 506.075 L89.3947 501.353 Q92.2188 500.218 94.6724 499.64 Q97.1261 499.061 99.1632 499.061 Q104.534 499.061 107.728 501.746 Q110.922 504.431 110.922 508.922 Q110.922 511.052 110.112 512.973 Q109.325 514.871 107.219 517.464 Q106.64 518.135 103.538 521.353 Q100.436 524.547 94.7882 530.311 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 502.765 Q127.311 502.765 125.482 506.329 Q123.677 509.871 123.677 517.001 Q123.677 524.107 125.482 527.672 Q127.311 531.214 130.922 531.214 Q134.556 531.214 136.362 527.672 Q138.191 524.107 138.191 517.001 Q138.191 509.871 136.362 506.329 Q134.556 502.765 130.922 502.765 M130.922 499.061 Q136.732 499.061 139.788 503.667 Q142.867 508.251 142.867 517.001 Q142.867 525.727 139.788 530.334 Q136.732 534.917 130.922 534.917 Q125.112 534.917 122.033 530.334 Q118.978 525.727 118.978 517.001 Q118.978 508.251 122.033 503.667 Q125.112 499.061 130.922 499.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M104.927 196.992 Q108.283 197.709 110.158 199.978 Q112.057 202.246 112.057 205.58 Q112.057 210.695 108.538 213.496 Q105.02 216.297 98.5382 216.297 Q96.3623 216.297 94.0475 215.857 Q91.7558 215.441 89.3021 214.584 L89.3021 210.07 Q91.2465 211.205 93.5613 211.783 Q95.8761 212.362 98.3993 212.362 Q102.797 212.362 105.089 210.626 Q107.404 208.89 107.404 205.58 Q107.404 202.524 105.251 200.811 Q103.121 199.075 99.3021 199.075 L95.2743 199.075 L95.2743 195.233 L99.4872 195.233 Q102.936 195.233 104.765 193.867 Q106.594 192.478 106.594 189.885 Q106.594 187.223 104.696 185.811 Q102.821 184.376 99.3021 184.376 Q97.3808 184.376 95.1817 184.793 Q92.9826 185.209 90.3438 186.089 L90.3438 181.922 Q93.0058 181.182 95.3206 180.811 Q97.6585 180.441 99.7187 180.441 Q105.043 180.441 108.145 182.872 Q111.246 185.279 111.246 189.399 Q111.246 192.27 109.603 194.26 Q107.959 196.228 104.927 196.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M130.922 184.145 Q127.311 184.145 125.482 187.709 Q123.677 191.251 123.677 198.381 Q123.677 205.487 125.482 209.052 Q127.311 212.594 130.922 212.594 Q134.556 212.594 136.362 209.052 Q138.191 205.487 138.191 198.381 Q138.191 191.251 136.362 187.709 Q134.556 184.145 130.922 184.145 M130.922 180.441 Q136.732 180.441 139.788 185.047 Q142.867 189.631 142.867 198.381 Q142.867 207.107 139.788 211.714 Q136.732 216.297 130.922 216.297 Q125.112 216.297 122.033 211.714 Q118.978 207.107 118.978 198.381 Q118.978 189.631 122.033 185.047 Q125.112 180.441 130.922 180.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><line clip-path="url(#clip072)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1926.4" y1="856.923" x2="1910.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1926.4" y1="856.923" x2="1910.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1926.4" y1="856.923" x2="1942.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1926.4" y1="856.923" x2="1942.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="912.69" y1="1348.83" x2="896.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="912.69" y1="1348.83" x2="896.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="912.69" y1="1348.83" x2="928.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="912.69" y1="1348.83" x2="928.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="625.575" y1="1469.54" x2="609.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="625.575" y1="1469.54" x2="609.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="625.575" y1="1469.54" x2="641.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="625.575" y1="1469.54" x2="641.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.36" y1="1138.56" x2="990.356" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.36" y1="1138.56" x2="990.356" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="785.24" y1="1436.86" x2="769.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="785.24" y1="1436.86" x2="769.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="785.24" y1="1436.86" x2="801.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="785.24" y1="1436.86" x2="801.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.289" y1="1394.75" x2="761.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.289" y1="1394.75" x2="761.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.289" y1="1394.75" x2="793.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.289" y1="1394.75" x2="793.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.41" y1="1014.15" x2="1369.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.41" y1="1014.15" x2="1369.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.41" y1="1014.15" x2="1401.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.41" y1="1014.15" x2="1401.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1612.46" y1="913.718" x2="1596.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1612.46" y1="913.718" x2="1596.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1612.46" y1="913.718" x2="1628.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1612.46" y1="913.718" x2="1628.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="947.412" y1="1370.03" x2="931.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="947.412" y1="1370.03" x2="931.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="947.412" y1="1370.03" x2="963.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="947.412" y1="1370.03" x2="963.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.497" y1="1483.85" x2="672.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.497" y1="1483.85" x2="672.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.497" y1="1483.85" x2="704.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.497" y1="1483.85" x2="704.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.95" y1="869.353" x2="1529.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.95" y1="869.353" x2="1529.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.95" y1="869.353" x2="1561.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.95" y1="869.353" x2="1561.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="771.564" y1="1431.34" x2="755.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="771.564" y1="1431.34" x2="755.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="771.564" y1="1431.34" x2="787.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="771.564" y1="1431.34" x2="787.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="918.674" y1="1363.31" x2="902.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="918.674" y1="1363.31" x2="902.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="918.674" y1="1363.31" x2="934.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="918.674" y1="1363.31" x2="934.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1707.08" y1="826.86" x2="1691.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1707.08" y1="826.86" x2="1691.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1707.08" y1="826.86" x2="1723.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1707.08" y1="826.86" x2="1723.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.63" y1="875.887" x2="1561.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.63" y1="875.887" x2="1561.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.63" y1="875.887" x2="1593.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.63" y1="875.887" x2="1593.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="894.72" y1="1142.51" x2="878.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="894.72" y1="1142.51" x2="878.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="894.72" y1="1142.51" x2="910.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="894.72" y1="1142.51" x2="910.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1580.8" y1="990.127" x2="1564.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1580.8" y1="990.127" x2="1564.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1580.8" y1="990.127" x2="1596.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1580.8" y1="990.127" x2="1596.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="852.02" y1="1283.64" x2="836.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="852.02" y1="1283.64" x2="836.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="852.02" y1="1283.64" x2="868.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="852.02" y1="1283.64" x2="868.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.48" y1="962.993" x2="1589.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.48" y1="962.993" x2="1589.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.48" y1="962.993" x2="1621.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.48" y1="962.993" x2="1621.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1172.56" y1="1009.01" x2="1156.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1172.56" y1="1009.01" x2="1156.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1172.56" y1="1009.01" x2="1188.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1172.56" y1="1009.01" x2="1188.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.588" y1="1415.15" x2="672.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.588" y1="1415.15" x2="672.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.588" y1="1415.15" x2="704.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="688.588" y1="1415.15" x2="704.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="991.105" y1="1402.23" x2="975.105" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="991.105" y1="1402.23" x2="975.105" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="991.105" y1="1402.23" x2="1007.11" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="991.105" y1="1402.23" x2="1007.11" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1811.99" y1="919.23" x2="1795.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1811.99" y1="919.23" x2="1795.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1811.99" y1="919.23" x2="1827.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1811.99" y1="919.23" x2="1827.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.971" y1="1350.77" x2="829.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.971" y1="1350.77" x2="829.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.971" y1="1350.77" x2="861.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.971" y1="1350.77" x2="861.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="886.864" y1="1368.41" x2="870.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="886.864" y1="1368.41" x2="870.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="886.864" y1="1368.41" x2="902.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="886.864" y1="1368.41" x2="902.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.009" y1="1388.03" x2="761.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.009" y1="1388.03" x2="761.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.009" y1="1388.03" x2="793.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="777.009" y1="1388.03" x2="793.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.23" y1="998.156" x2="1277.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.23" y1="998.156" x2="1277.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.23" y1="998.156" x2="1309.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.23" y1="998.156" x2="1309.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="885.196" y1="1405.14" x2="869.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="885.196" y1="1405.14" x2="869.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="885.196" y1="1405.14" x2="901.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="885.196" y1="1405.14" x2="901.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.55" y1="853.956" x2="1441.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.55" y1="853.956" x2="1441.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.55" y1="853.956" x2="1473.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.55" y1="853.956" x2="1473.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="949.006" y1="1440.98" x2="933.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="949.006" y1="1440.98" x2="933.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="949.006" y1="1440.98" x2="965.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="949.006" y1="1440.98" x2="965.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.411" y1="1427.9" x2="829.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.411" y1="1427.9" x2="829.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.411" y1="1427.9" x2="861.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="845.411" y1="1427.9" x2="861.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.57" y1="868.321" x2="1530.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.57" y1="868.321" x2="1530.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.57" y1="868.321" x2="1562.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.57" y1="868.321" x2="1562.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.43" y1="1342.43" x2="834.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.43" y1="1342.43" x2="834.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.43" y1="1342.43" x2="866.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.43" y1="1342.43" x2="866.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.5" y1="829.178" x2="1603.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.5" y1="829.178" x2="1603.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.5" y1="829.178" x2="1635.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1619.5" y1="829.178" x2="1635.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1847.7" y1="804.547" x2="1831.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1847.7" y1="804.547" x2="1831.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1847.7" y1="804.547" x2="1863.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1847.7" y1="804.547" x2="1863.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="992.659" y1="1251.94" x2="976.659" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="992.659" y1="1251.94" x2="976.659" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="992.659" y1="1251.94" x2="1008.66" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="992.659" y1="1251.94" x2="1008.66" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1641.07" y1="860.861" x2="1625.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1641.07" y1="860.861" x2="1625.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1641.07" y1="860.861" x2="1657.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1641.07" y1="860.861" x2="1657.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1301.11" y1="987.604" x2="1285.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1301.11" y1="987.604" x2="1285.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1301.11" y1="987.604" x2="1317.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1301.11" y1="987.604" x2="1317.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1673.06" y1="901.944" x2="1657.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1673.06" y1="901.944" x2="1657.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1673.06" y1="901.944" x2="1689.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1673.06" y1="901.944" x2="1689.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.89" y1="961.872" x2="1595.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.89" y1="961.872" x2="1595.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.89" y1="961.872" x2="1627.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.89" y1="961.872" x2="1627.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="782.176" y1="1390.69" x2="766.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="782.176" y1="1390.69" x2="766.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="782.176" y1="1390.69" x2="798.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="782.176" y1="1390.69" x2="798.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.77" y1="921.537" x2="1659.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.77" y1="921.537" x2="1659.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.77" y1="921.537" x2="1691.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.77" y1="921.537" x2="1691.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="816.257" y1="1346.85" x2="800.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="816.257" y1="1346.85" x2="800.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="816.257" y1="1346.85" x2="832.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="816.257" y1="1346.85" x2="832.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="724.712" y1="1453.02" x2="708.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="724.712" y1="1453.02" x2="708.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="724.712" y1="1453.02" x2="740.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="724.712" y1="1453.02" x2="740.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.34" y1="968.47" x2="1659.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.34" y1="968.47" x2="1659.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.34" y1="968.47" x2="1691.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1675.34" y1="968.47" x2="1691.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.396" y1="1435.49" x2="858.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.396" y1="1435.49" x2="858.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.396" y1="1435.49" x2="890.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.396" y1="1435.49" x2="890.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1330.6" y1="1013.95" x2="1314.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1330.6" y1="1013.95" x2="1314.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1330.6" y1="1013.95" x2="1346.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1330.6" y1="1013.95" x2="1346.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1626.12" y1="969.424" x2="1610.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1626.12" y1="969.424" x2="1610.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1626.12" y1="969.424" x2="1642.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1626.12" y1="969.424" x2="1642.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="933.271" y1="1353.8" x2="917.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="933.271" y1="1353.8" x2="917.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="933.271" y1="1353.8" x2="949.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="933.271" y1="1353.8" x2="949.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="906.639" y1="1451.01" x2="890.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="906.639" y1="1451.01" x2="890.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="906.639" y1="1451.01" x2="922.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="906.639" y1="1451.01" x2="922.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="958.996" y1="1406.94" x2="942.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="958.996" y1="1406.94" x2="942.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="958.996" y1="1406.94" x2="974.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="958.996" y1="1406.94" x2="974.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.277" y1="1312.14" x2="858.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.277" y1="1312.14" x2="858.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.277" y1="1312.14" x2="890.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="874.277" y1="1312.14" x2="890.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.691" y1="1424.89" x2="911.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.691" y1="1424.89" x2="911.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.691" y1="1424.89" x2="943.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.691" y1="1424.89" x2="943.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="538.801" y1="1550.28" x2="522.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="538.801" y1="1550.28" x2="522.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="538.801" y1="1550.28" x2="554.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="538.801" y1="1550.28" x2="554.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="869.402" y1="1512" x2="853.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="869.402" y1="1512" x2="853.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="869.402" y1="1512" x2="885.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="869.402" y1="1512" x2="885.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.688" y1="1677.8" x2="553.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.688" y1="1677.8" x2="553.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.688" y1="1677.8" x2="585.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.688" y1="1677.8" x2="585.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="915.573" y1="1448.5" x2="899.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="915.573" y1="1448.5" x2="899.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="915.573" y1="1448.5" x2="931.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="915.573" y1="1448.5" x2="931.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1520.69" y1="868.39" x2="1504.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1520.69" y1="868.39" x2="1504.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1520.69" y1="868.39" x2="1536.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1520.69" y1="868.39" x2="1536.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.79" y1="1007.52" x2="1446.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.79" y1="1007.52" x2="1446.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.79" y1="1007.52" x2="1478.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1462.79" y1="1007.52" x2="1478.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.71" y1="986.213" x2="1427.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.71" y1="986.213" x2="1427.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.71" y1="986.213" x2="1459.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.71" y1="986.213" x2="1459.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="831.587" y1="1424.87" x2="815.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="831.587" y1="1424.87" x2="815.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="831.587" y1="1424.87" x2="847.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="831.587" y1="1424.87" x2="847.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.93" y1="1006.01" x2="1593.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.93" y1="1006.01" x2="1593.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.93" y1="1006.01" x2="1625.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.93" y1="1006.01" x2="1625.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1733.98" y1="723.632" x2="1717.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1733.98" y1="723.632" x2="1717.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1733.98" y1="723.632" x2="1749.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1733.98" y1="723.632" x2="1749.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.936" y1="1351.49" x2="834.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.936" y1="1351.49" x2="834.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.936" y1="1351.49" x2="866.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="850.936" y1="1351.49" x2="866.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.92" y1="896.324" x2="1752.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.92" y1="896.324" x2="1752.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.92" y1="896.324" x2="1784.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.92" y1="896.324" x2="1784.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.86" y1="1264.5" x2="984.861" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.86" y1="1264.5" x2="984.861" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.48" y1="973.842" x2="1509.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.48" y1="973.842" x2="1509.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.48" y1="973.842" x2="1541.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.48" y1="973.842" x2="1541.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="814.5" y1="1271.95" x2="798.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="814.5" y1="1271.95" x2="798.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="814.5" y1="1271.95" x2="830.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="814.5" y1="1271.95" x2="830.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1834.69" y1="763.508" x2="1818.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1834.69" y1="763.508" x2="1818.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1834.69" y1="763.508" x2="1850.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1834.69" y1="763.508" x2="1850.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.04" y1="1278.88" x2="990.041" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.04" y1="1278.88" x2="990.041" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1424" y1="1023.62" x2="1408" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1424" y1="1023.62" x2="1408" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1424" y1="1023.62" x2="1440" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1424" y1="1023.62" x2="1440" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1618.21" y1="979.458" x2="1602.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1618.21" y1="979.458" x2="1602.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1618.21" y1="979.458" x2="1634.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1618.21" y1="979.458" x2="1634.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1005.39" y1="1284.51" x2="989.392" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1005.39" y1="1284.51" x2="989.392" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.98" y1="894.258" x2="1593.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.98" y1="894.258" x2="1593.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.98" y1="894.258" x2="1625.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1609.98" y1="894.258" x2="1625.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1597.07" y1="924.794" x2="1581.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1597.07" y1="924.794" x2="1581.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1597.07" y1="924.794" x2="1613.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1597.07" y1="924.794" x2="1613.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="882.48" y1="1367.74" x2="866.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="882.48" y1="1367.74" x2="866.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="882.48" y1="1367.74" x2="898.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="882.48" y1="1367.74" x2="898.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1527.95" y1="988.132" x2="1511.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1527.95" y1="988.132" x2="1511.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1527.95" y1="988.132" x2="1543.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1527.95" y1="988.132" x2="1543.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1766.37" y1="756.534" x2="1750.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1766.37" y1="756.534" x2="1750.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1766.37" y1="756.534" x2="1782.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1766.37" y1="756.534" x2="1782.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1422.2" y1="999.716" x2="1406.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1422.2" y1="999.716" x2="1406.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1422.2" y1="999.716" x2="1438.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1422.2" y1="999.716" x2="1438.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1381.71" y1="975.726" x2="1365.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1381.71" y1="975.726" x2="1365.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1381.71" y1="975.726" x2="1397.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1381.71" y1="975.726" x2="1397.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.037" y1="1422.12" x2="859.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.037" y1="1422.12" x2="859.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.037" y1="1422.12" x2="891.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.037" y1="1422.12" x2="891.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="898.259" y1="1319.39" x2="882.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="898.259" y1="1319.39" x2="882.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="898.259" y1="1319.39" x2="914.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="898.259" y1="1319.39" x2="914.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.19" y1="815.289" x2="1746.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.19" y1="815.289" x2="1746.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.19" y1="815.289" x2="1778.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.19" y1="815.289" x2="1778.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="843.736" y1="1421.71" x2="827.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="843.736" y1="1421.71" x2="827.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="843.736" y1="1421.71" x2="859.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="843.736" y1="1421.71" x2="859.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1399.57" y1="992.596" x2="1383.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1399.57" y1="992.596" x2="1383.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1399.57" y1="992.596" x2="1415.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1399.57" y1="992.596" x2="1415.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="712.506" y1="1466.98" x2="696.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="712.506" y1="1466.98" x2="696.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="712.506" y1="1466.98" x2="728.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="712.506" y1="1466.98" x2="728.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="996.845" y1="1298.95" x2="980.845" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="996.845" y1="1298.95" x2="980.845" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="996.845" y1="1298.95" x2="1012.85" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="996.845" y1="1298.95" x2="1012.85" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="764.769" y1="1428.81" x2="748.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="764.769" y1="1428.81" x2="748.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="764.769" y1="1428.81" x2="780.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="764.769" y1="1428.81" x2="780.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1481.71" y1="979.429" x2="1465.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1481.71" y1="979.429" x2="1465.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1481.71" y1="979.429" x2="1497.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1481.71" y1="979.429" x2="1497.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="794.022" y1="1392.87" x2="778.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="794.022" y1="1392.87" x2="778.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="794.022" y1="1392.87" x2="810.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="794.022" y1="1392.87" x2="810.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="989.1" y1="1361.68" x2="973.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="989.1" y1="1361.68" x2="973.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="989.1" y1="1361.68" x2="1005.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="989.1" y1="1361.68" x2="1005.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="658.223" y1="1594.72" x2="642.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="658.223" y1="1594.72" x2="642.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="658.223" y1="1594.72" x2="674.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="658.223" y1="1594.72" x2="674.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.13" y1="922.6" x2="1595.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.13" y1="922.6" x2="1595.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.13" y1="922.6" x2="1627.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1611.13" y1="922.6" x2="1627.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="968.976" y1="1423.08" x2="952.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="968.976" y1="1423.08" x2="952.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="968.976" y1="1423.08" x2="984.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="968.976" y1="1423.08" x2="984.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1451.83" y1="926.379" x2="1435.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1451.83" y1="926.379" x2="1435.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1451.83" y1="926.379" x2="1467.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1451.83" y1="926.379" x2="1467.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="908.822" y1="1392.95" x2="892.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="908.822" y1="1392.95" x2="892.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="908.822" y1="1392.95" x2="924.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="908.822" y1="1392.95" x2="924.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.57" y1="1011.37" x2="1484.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.57" y1="1011.37" x2="1484.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.57" y1="1011.37" x2="1516.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1500.57" y1="1011.37" x2="1516.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1002.47" y1="1234.46" x2="986.471" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1002.47" y1="1234.46" x2="986.471" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.95" y1="1013.72" x2="1442.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.95" y1="1013.72" x2="1442.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.95" y1="1013.72" x2="1474.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1458.95" y1="1013.72" x2="1474.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="892.734" y1="1460.48" x2="876.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="892.734" y1="1460.48" x2="876.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="892.734" y1="1460.48" x2="908.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="892.734" y1="1460.48" x2="908.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.79" y1="924.706" x2="1529.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.79" y1="924.706" x2="1529.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.79" y1="924.706" x2="1561.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.79" y1="924.706" x2="1561.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1003.69" y1="1262.65" x2="987.692" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1003.69" y1="1262.65" x2="987.692" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="862.589" y1="1433.49" x2="846.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="862.589" y1="1433.49" x2="846.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="862.589" y1="1433.49" x2="878.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="862.589" y1="1433.49" x2="878.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="522.504" y1="1512.98" x2="506.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="522.504" y1="1512.98" x2="506.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="522.504" y1="1512.98" x2="538.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="522.504" y1="1512.98" x2="538.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.208" y1="1309.37" x2="956.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.208" y1="1309.37" x2="956.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.208" y1="1309.37" x2="988.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.208" y1="1309.37" x2="988.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.36" y1="1054" x2="1160.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.36" y1="1054" x2="1160.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.36" y1="1054" x2="1192.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.36" y1="1054" x2="1192.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.451" y1="1406.48" x2="728.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.451" y1="1406.48" x2="728.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.451" y1="1406.48" x2="760.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.451" y1="1406.48" x2="760.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1012.88" y1="1293.81" x2="996.879" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1012.88" y1="1293.81" x2="996.879" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="277.68" y1="1755.15" x2="261.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="277.68" y1="1755.15" x2="261.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="277.68" y1="1755.15" x2="293.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="277.68" y1="1755.15" x2="293.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1498.04" y1="1011.64" x2="1482.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1498.04" y1="1011.64" x2="1482.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1498.04" y1="1011.64" x2="1514.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1498.04" y1="1011.64" x2="1514.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.19" y1="842.118" x2="1764.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.19" y1="842.118" x2="1764.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.19" y1="842.118" x2="1796.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.19" y1="842.118" x2="1796.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="855.143" y1="1461.56" x2="839.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="855.143" y1="1461.56" x2="839.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="855.143" y1="1461.56" x2="871.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="855.143" y1="1461.56" x2="871.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="941.026" y1="1398.61" x2="925.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="941.026" y1="1398.61" x2="925.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="941.026" y1="1398.61" x2="957.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="941.026" y1="1398.61" x2="957.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1491.9" y1="862.19" x2="1475.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1491.9" y1="862.19" x2="1475.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1491.9" y1="862.19" x2="1507.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1491.9" y1="862.19" x2="1507.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1614.72" y1="834.262" x2="1598.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1614.72" y1="834.262" x2="1598.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1614.72" y1="834.262" x2="1630.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1614.72" y1="834.262" x2="1630.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.83" y1="933.105" x2="1599.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.83" y1="933.105" x2="1599.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.83" y1="933.105" x2="1631.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.83" y1="933.105" x2="1631.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.951" y1="1271.4" x2="883.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.951" y1="1271.4" x2="883.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.951" y1="1271.4" x2="915.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.951" y1="1271.4" x2="915.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1651.8" y1="933.534" x2="1635.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1651.8" y1="933.534" x2="1635.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1651.8" y1="933.534" x2="1667.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1651.8" y1="933.534" x2="1667.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.2" y1="982.773" x2="1530.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.2" y1="982.773" x2="1530.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.2" y1="982.773" x2="1562.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1546.2" y1="982.773" x2="1562.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="798.462" y1="1475.78" x2="782.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="798.462" y1="1475.78" x2="782.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="798.462" y1="1475.78" x2="814.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="798.462" y1="1475.78" x2="814.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1653.74" y1="920.543" x2="1637.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1653.74" y1="920.543" x2="1637.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1653.74" y1="920.543" x2="1669.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1653.74" y1="920.543" x2="1669.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1550.14" y1="1004.56" x2="1534.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1550.14" y1="1004.56" x2="1534.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1550.14" y1="1004.56" x2="1566.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1550.14" y1="1004.56" x2="1566.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="955.148" y1="1370.3" x2="939.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="955.148" y1="1370.3" x2="939.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="955.148" y1="1370.3" x2="971.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="955.148" y1="1370.3" x2="971.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1845.45" y1="659.705" x2="1829.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1845.45" y1="659.705" x2="1829.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1845.45" y1="659.705" x2="1861.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1845.45" y1="659.705" x2="1861.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1817.11" y1="731.36" x2="1801.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1817.11" y1="731.36" x2="1801.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1817.11" y1="731.36" x2="1833.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1817.11" y1="731.36" x2="1833.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1838.37" y1="806.858" x2="1822.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1838.37" y1="806.858" x2="1822.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1838.37" y1="806.858" x2="1854.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1838.37" y1="806.858" x2="1854.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244" y1="1263.65" x2="1228" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244" y1="1263.65" x2="1228" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244" y1="1263.65" x2="1260" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244" y1="1263.65" x2="1260" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1558.7" y1="965.028" x2="1542.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1558.7" y1="965.028" x2="1542.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1558.7" y1="965.028" x2="1574.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1558.7" y1="965.028" x2="1574.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.76" y1="970.455" x2="1458.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.76" y1="970.455" x2="1458.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.76" y1="970.455" x2="1490.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.76" y1="970.455" x2="1490.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1682.62" y1="820.381" x2="1666.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1682.62" y1="820.381" x2="1666.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1682.62" y1="820.381" x2="1698.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1682.62" y1="820.381" x2="1698.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="872.16" y1="1370.98" x2="856.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="872.16" y1="1370.98" x2="856.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="872.16" y1="1370.98" x2="888.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="872.16" y1="1370.98" x2="888.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="866.851" y1="1331.78" x2="850.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="866.851" y1="1331.78" x2="850.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="866.851" y1="1331.78" x2="882.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="866.851" y1="1331.78" x2="882.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1604.79" y1="991.27" x2="1588.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1604.79" y1="991.27" x2="1588.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1604.79" y1="991.27" x2="1620.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1604.79" y1="991.27" x2="1620.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="436.181" y1="1793.59" x2="420.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="436.181" y1="1793.59" x2="420.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="436.181" y1="1793.59" x2="452.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="436.181" y1="1793.59" x2="452.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="957.079" y1="1394.58" x2="941.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="957.079" y1="1394.58" x2="941.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="957.079" y1="1394.58" x2="973.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="957.079" y1="1394.58" x2="973.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.366" y1="1495.39" x2="775.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.366" y1="1495.39" x2="775.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.366" y1="1495.39" x2="807.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.366" y1="1495.39" x2="807.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1636.9" y1="890.133" x2="1620.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1636.9" y1="890.133" x2="1620.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1636.9" y1="890.133" x2="1652.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1636.9" y1="890.133" x2="1652.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1256" y1="1205.58" x2="1240" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1256" y1="1205.58" x2="1240" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1256" y1="1205.58" x2="1272" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1256" y1="1205.58" x2="1272" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="721.091" y1="1561.75" x2="705.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="721.091" y1="1561.75" x2="705.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="721.091" y1="1561.75" x2="737.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="721.091" y1="1561.75" x2="737.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="823.473" y1="1446.58" x2="807.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="823.473" y1="1446.58" x2="807.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="823.473" y1="1446.58" x2="839.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="823.473" y1="1446.58" x2="839.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="887.822" y1="1375.21" x2="871.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="887.822" y1="1375.21" x2="871.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="887.822" y1="1375.21" x2="903.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="887.822" y1="1375.21" x2="903.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.39" y1="1014.06" x2="1468.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.39" y1="1014.06" x2="1468.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.39" y1="1014.06" x2="1500.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1484.39" y1="1014.06" x2="1500.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1685.1" y1="924.047" x2="1669.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1685.1" y1="924.047" x2="1669.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1685.1" y1="924.047" x2="1701.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1685.1" y1="924.047" x2="1701.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="877.025" y1="1403.06" x2="861.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="877.025" y1="1403.06" x2="861.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="877.025" y1="1403.06" x2="893.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="877.025" y1="1403.06" x2="893.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1473.39" y1="1000.62" x2="1457.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1473.39" y1="1000.62" x2="1457.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1473.39" y1="1000.62" x2="1489.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1473.39" y1="1000.62" x2="1489.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="930.593" y1="1251.3" x2="914.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="930.593" y1="1251.3" x2="914.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="930.593" y1="1251.3" x2="946.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="930.593" y1="1251.3" x2="946.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.55" y1="972.554" x2="1599.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.55" y1="972.554" x2="1599.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.55" y1="972.554" x2="1631.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1615.55" y1="972.554" x2="1631.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1395.64" y1="960.36" x2="1379.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1395.64" y1="960.36" x2="1379.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1395.64" y1="960.36" x2="1411.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1395.64" y1="960.36" x2="1411.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.08" y1="1312.36" x2="984.075" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.08" y1="1312.36" x2="984.075" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="805.605" y1="1368.59" x2="789.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="805.605" y1="1368.59" x2="789.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="805.605" y1="1368.59" x2="821.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="805.605" y1="1368.59" x2="821.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="735.913" y1="1350.58" x2="719.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="735.913" y1="1350.58" x2="719.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="735.913" y1="1350.58" x2="751.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="735.913" y1="1350.58" x2="751.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1826.24" y1="921.284" x2="1810.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1826.24" y1="921.284" x2="1810.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1826.24" y1="921.284" x2="1842.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1826.24" y1="921.284" x2="1842.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1572.67" y1="995.733" x2="1556.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1572.67" y1="995.733" x2="1556.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1572.67" y1="995.733" x2="1588.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1572.67" y1="995.733" x2="1588.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1469" y1="974.22" x2="1453" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1469" y1="974.22" x2="1453" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1469" y1="974.22" x2="1485" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1469" y1="974.22" x2="1485" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.64" y1="1000.17" x2="1752.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.64" y1="1000.17" x2="1752.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.64" y1="1000.17" x2="1784.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1768.64" y1="1000.17" x2="1784.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.6" y1="944.573" x2="1470.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.6" y1="944.573" x2="1470.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.6" y1="944.573" x2="1502.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1486.6" y1="944.573" x2="1502.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.81" y1="1008.74" x2="1427.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.81" y1="1008.74" x2="1427.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.81" y1="1008.74" x2="1459.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1443.81" y1="1008.74" x2="1459.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1628.36" y1="926.189" x2="1612.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1628.36" y1="926.189" x2="1612.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1628.36" y1="926.189" x2="1644.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1628.36" y1="926.189" x2="1644.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.938" y1="1534.93" x2="553.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.938" y1="1534.93" x2="553.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.938" y1="1534.93" x2="585.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="569.938" y1="1534.93" x2="585.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.218" y1="1483.05" x2="728.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.218" y1="1483.05" x2="728.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.218" y1="1483.05" x2="760.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="744.218" y1="1483.05" x2="760.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1713.33" y1="929.721" x2="1697.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1713.33" y1="929.721" x2="1697.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1713.33" y1="929.721" x2="1729.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1713.33" y1="929.721" x2="1729.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="610.516" y1="1568.23" x2="594.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="610.516" y1="1568.23" x2="594.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="610.516" y1="1568.23" x2="626.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="610.516" y1="1568.23" x2="626.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1702.63" y1="949.089" x2="1686.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1702.63" y1="949.089" x2="1686.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1702.63" y1="949.089" x2="1718.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1702.63" y1="949.089" x2="1718.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1799.12" y1="921.26" x2="1783.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1799.12" y1="921.26" x2="1783.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1799.12" y1="921.26" x2="1815.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1799.12" y1="921.26" x2="1815.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1539.91" y1="992.445" x2="1523.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1539.91" y1="992.445" x2="1523.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1539.91" y1="992.445" x2="1555.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1539.91" y1="992.445" x2="1555.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.824" y1="1370.17" x2="947.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.824" y1="1370.17" x2="947.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.824" y1="1370.17" x2="979.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.824" y1="1370.17" x2="979.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="694.471" y1="1399.51" x2="678.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="694.471" y1="1399.51" x2="678.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="694.471" y1="1399.51" x2="710.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="694.471" y1="1399.51" x2="710.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.32" y1="993.037" x2="1462.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.32" y1="993.037" x2="1462.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.32" y1="993.037" x2="1494.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1478.32" y1="993.037" x2="1494.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.95" y1="1279" x2="1264.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.95" y1="1279" x2="1264.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.95" y1="1279" x2="1296.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.95" y1="1279" x2="1296.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1650.12" y1="918.653" x2="1634.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1650.12" y1="918.653" x2="1634.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1650.12" y1="918.653" x2="1666.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1650.12" y1="918.653" x2="1666.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1860.89" y1="778.201" x2="1844.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1860.89" y1="778.201" x2="1844.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1860.89" y1="778.201" x2="1876.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1860.89" y1="778.201" x2="1876.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1444.37" y1="1001.52" x2="1428.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1444.37" y1="1001.52" x2="1428.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1444.37" y1="1001.52" x2="1460.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1444.37" y1="1001.52" x2="1460.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.75" y1="1012.66" x2="1412.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.75" y1="1012.66" x2="1412.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.75" y1="1012.66" x2="1444.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.75" y1="1012.66" x2="1444.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1896.85" y1="631.005" x2="1880.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1896.85" y1="631.005" x2="1880.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1896.85" y1="631.005" x2="1912.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1896.85" y1="631.005" x2="1912.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1863.03" y1="689.163" x2="1847.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1863.03" y1="689.163" x2="1847.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1863.03" y1="689.163" x2="1879.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1863.03" y1="689.163" x2="1879.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="973.556" y1="1249.11" x2="957.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="973.556" y1="1249.11" x2="957.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="973.556" y1="1249.11" x2="989.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="973.556" y1="1249.11" x2="989.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.42" y1="974.507" x2="1399.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.42" y1="974.507" x2="1399.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.42" y1="974.507" x2="1431.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1415.42" y1="974.507" x2="1431.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="961.172" y1="1297.9" x2="945.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="961.172" y1="1297.9" x2="945.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="961.172" y1="1297.9" x2="977.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="961.172" y1="1297.9" x2="977.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="828.067" y1="1395.72" x2="812.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="828.067" y1="1395.72" x2="812.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="828.067" y1="1395.72" x2="844.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="828.067" y1="1395.72" x2="844.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="812.031" y1="1484.71" x2="796.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="812.031" y1="1484.71" x2="796.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="812.031" y1="1484.71" x2="828.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="812.031" y1="1484.71" x2="828.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="901.731" y1="1295.1" x2="885.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="901.731" y1="1295.1" x2="885.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="901.731" y1="1295.1" x2="917.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="901.731" y1="1295.1" x2="917.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.853" y1="1331.79" x2="956.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.853" y1="1331.79" x2="956.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.853" y1="1331.79" x2="988.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="972.853" y1="1331.79" x2="988.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.97" y1="912.119" x2="1722.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.97" y1="912.119" x2="1722.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.97" y1="912.119" x2="1754.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.97" y1="912.119" x2="1754.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1534.29" y1="956.26" x2="1518.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1534.29" y1="956.26" x2="1518.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1534.29" y1="956.26" x2="1550.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1534.29" y1="956.26" x2="1550.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.177" y1="1437.3" x2="863.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.177" y1="1437.3" x2="863.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.177" y1="1437.3" x2="895.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.177" y1="1437.3" x2="895.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.308" y1="1302.19" x2="775.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.308" y1="1302.19" x2="775.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.308" y1="1302.19" x2="807.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="791.308" y1="1302.19" x2="807.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.44" y1="890.561" x2="1574.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.44" y1="890.561" x2="1574.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.44" y1="890.561" x2="1606.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.44" y1="890.561" x2="1606.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1729.02" y1="853.811" x2="1713.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1729.02" y1="853.811" x2="1713.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1729.02" y1="853.811" x2="1745.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1729.02" y1="853.811" x2="1745.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="936.651" y1="1397.2" x2="920.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="936.651" y1="1397.2" x2="920.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="936.651" y1="1397.2" x2="952.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="936.651" y1="1397.2" x2="952.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="856.289" y1="1386.44" x2="840.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="856.289" y1="1386.44" x2="840.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="856.289" y1="1386.44" x2="872.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="856.289" y1="1386.44" x2="872.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1350.38" y1="930.437" x2="1334.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1350.38" y1="930.437" x2="1334.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1350.38" y1="930.437" x2="1366.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1350.38" y1="930.437" x2="1366.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1529.58" y1="996.113" x2="1513.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1529.58" y1="996.113" x2="1513.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1529.58" y1="996.113" x2="1545.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1529.58" y1="996.113" x2="1545.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.47" y1="955.973" x2="1497.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.47" y1="955.973" x2="1497.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.47" y1="955.973" x2="1529.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.47" y1="955.973" x2="1529.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1530.22" y1="1008.88" x2="1514.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1530.22" y1="1008.88" x2="1514.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1530.22" y1="1008.88" x2="1546.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1530.22" y1="1008.88" x2="1546.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.75" y1="828.151" x2="1574.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.75" y1="828.151" x2="1574.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.75" y1="828.151" x2="1606.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1590.75" y1="828.151" x2="1606.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1942.19" y1="883.328" x2="1926.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1942.19" y1="883.328" x2="1926.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1942.19" y1="883.328" x2="1958.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1942.19" y1="883.328" x2="1958.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="800.484" y1="1327.86" x2="784.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="800.484" y1="1327.86" x2="784.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="800.484" y1="1327.86" x2="816.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="800.484" y1="1327.86" x2="816.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.479" y1="1361.21" x2="863.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.479" y1="1361.21" x2="863.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.479" y1="1361.21" x2="895.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="879.479" y1="1361.21" x2="895.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1621.59" y1="888.584" x2="1605.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1621.59" y1="888.584" x2="1605.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1621.59" y1="888.584" x2="1637.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1621.59" y1="888.584" x2="1637.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.03" y1="917.837" x2="1589.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.03" y1="917.837" x2="1589.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.03" y1="917.837" x2="1621.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1605.03" y1="917.837" x2="1621.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="863.983" y1="1256.2" x2="847.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="863.983" y1="1256.2" x2="847.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="863.983" y1="1256.2" x2="879.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="863.983" y1="1256.2" x2="879.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1763.1" y1="789.55" x2="1747.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1763.1" y1="789.55" x2="1747.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1763.1" y1="789.55" x2="1779.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1763.1" y1="789.55" x2="1779.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1840.32" y1="839.458" x2="1824.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1840.32" y1="839.458" x2="1824.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1840.32" y1="839.458" x2="1856.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1840.32" y1="839.458" x2="1856.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1566.73" y1="937.472" x2="1550.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1566.73" y1="937.472" x2="1550.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1566.73" y1="937.472" x2="1582.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1566.73" y1="937.472" x2="1582.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.17" y1="997.851" x2="1441.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.17" y1="997.851" x2="1441.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.17" y1="997.851" x2="1473.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.17" y1="997.851" x2="1473.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="725.698" y1="1501.31" x2="709.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="725.698" y1="1501.31" x2="709.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="725.698" y1="1501.31" x2="741.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="725.698" y1="1501.31" x2="741.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="804.197" y1="1345.85" x2="788.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="804.197" y1="1345.85" x2="788.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="804.197" y1="1345.85" x2="820.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="804.197" y1="1345.85" x2="820.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.44" y1="832.688" x2="1722.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.44" y1="832.688" x2="1722.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.44" y1="832.688" x2="1754.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1738.44" y1="832.688" x2="1754.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1575.07" y1="927.362" x2="1559.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1575.07" y1="927.362" x2="1559.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1575.07" y1="927.362" x2="1591.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1575.07" y1="927.362" x2="1591.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.369" y1="1393.26" x2="947.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.369" y1="1393.26" x2="947.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.369" y1="1393.26" x2="979.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.369" y1="1393.26" x2="979.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1667.21" y1="967.122" x2="1651.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1667.21" y1="967.122" x2="1651.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1667.21" y1="967.122" x2="1683.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1667.21" y1="967.122" x2="1683.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1808.12" y1="803.64" x2="1792.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1808.12" y1="803.64" x2="1792.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1808.12" y1="803.64" x2="1824.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1808.12" y1="803.64" x2="1824.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.457" y1="1364.93" x2="911.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.457" y1="1364.93" x2="911.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.457" y1="1364.93" x2="943.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="927.457" y1="1364.93" x2="943.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.588" y1="1394.79" x2="938.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.588" y1="1394.79" x2="938.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.588" y1="1394.79" x2="970.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.588" y1="1394.79" x2="970.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1600.82" y1="848.267" x2="1584.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1600.82" y1="848.267" x2="1584.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1600.82" y1="848.267" x2="1616.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1600.82" y1="848.267" x2="1616.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1013.05" y1="1254.21" x2="997.055" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1013.05" y1="1254.21" x2="997.055" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="710.768" y1="1471.31" x2="694.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="710.768" y1="1471.31" x2="694.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="710.768" y1="1471.31" x2="726.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="710.768" y1="1471.31" x2="726.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.455" y1="1241.66" x2="938.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.455" y1="1241.66" x2="938.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.455" y1="1241.66" x2="970.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="954.455" y1="1241.66" x2="970.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.19" y1="997.572" x2="1412.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.19" y1="997.572" x2="1412.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.19" y1="997.572" x2="1444.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1428.19" y1="997.572" x2="1444.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="907.609" y1="1313.11" x2="891.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="907.609" y1="1313.11" x2="891.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="907.609" y1="1313.11" x2="923.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="907.609" y1="1313.11" x2="923.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="904.55" y1="1372.25" x2="888.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="904.55" y1="1372.25" x2="888.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="904.55" y1="1372.25" x2="920.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="904.55" y1="1372.25" x2="920.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1561.34" y1="914.945" x2="1545.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1561.34" y1="914.945" x2="1545.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1561.34" y1="914.945" x2="1577.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1561.34" y1="914.945" x2="1577.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1311.21" y1="973.579" x2="1295.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1311.21" y1="973.579" x2="1295.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1311.21" y1="973.579" x2="1327.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1311.21" y1="973.579" x2="1327.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="743.908" y1="1344.73" x2="727.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="743.908" y1="1344.73" x2="727.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="743.908" y1="1344.73" x2="759.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="743.908" y1="1344.73" x2="759.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.408" y1="1314.39" x2="947.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.408" y1="1314.39" x2="947.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.408" y1="1314.39" x2="979.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="963.408" y1="1314.39" x2="979.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="994.563" y1="1220.85" x2="978.563" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="994.563" y1="1220.85" x2="978.563" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="994.563" y1="1220.85" x2="1010.56" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="994.563" y1="1220.85" x2="1010.56" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.07" y1="993.346" x2="1764.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.07" y1="993.346" x2="1764.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.07" y1="993.346" x2="1796.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1780.07" y1="993.346" x2="1796.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="983.358" y1="1291.75" x2="967.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="983.358" y1="1291.75" x2="967.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="983.358" y1="1291.75" x2="999.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="983.358" y1="1291.75" x2="999.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1496.43" y1="924.203" x2="1480.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1496.43" y1="924.203" x2="1480.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1496.43" y1="924.203" x2="1512.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1496.43" y1="924.203" x2="1512.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="938.137" y1="1395.53" x2="922.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="938.137" y1="1395.53" x2="922.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="938.137" y1="1395.53" x2="954.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="938.137" y1="1395.53" x2="954.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="745.679" y1="1455.46" x2="729.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="745.679" y1="1455.46" x2="729.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="745.679" y1="1455.46" x2="761.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="745.679" y1="1455.46" x2="761.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1226.7" y1="1172" x2="1210.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1226.7" y1="1172" x2="1210.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1226.7" y1="1172" x2="1242.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1226.7" y1="1172" x2="1242.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.77" y1="884.258" x2="1746.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.77" y1="884.258" x2="1746.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.77" y1="884.258" x2="1778.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1762.77" y1="884.258" x2="1778.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="977.334" y1="1349.1" x2="961.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="977.334" y1="1349.1" x2="961.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="977.334" y1="1349.1" x2="993.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="977.334" y1="1349.1" x2="993.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.292" y1="1482.58" x2="859.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.292" y1="1482.58" x2="859.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.292" y1="1482.58" x2="891.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="875.292" y1="1482.58" x2="891.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.433" y1="1329.29" x2="883.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.433" y1="1329.29" x2="883.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.433" y1="1329.29" x2="915.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="899.433" y1="1329.29" x2="915.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1552.62" y1="936.348" x2="1536.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1552.62" y1="936.348" x2="1536.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1552.62" y1="936.348" x2="1568.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1552.62" y1="936.348" x2="1568.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.54" y1="984.865" x2="1456.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.54" y1="984.865" x2="1456.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.54" y1="984.865" x2="1488.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1472.54" y1="984.865" x2="1488.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165" y1="1027.31" x2="1149" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165" y1="1027.31" x2="1149" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165" y1="1027.31" x2="1181" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1165" y1="1027.31" x2="1181" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239" y1="1106.12" x2="1223" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239" y1="1106.12" x2="1223" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239" y1="1106.12" x2="1255" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1239" y1="1106.12" x2="1255" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="839.379" y1="1287.74" x2="823.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="839.379" y1="1287.74" x2="823.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="839.379" y1="1287.74" x2="855.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="839.379" y1="1287.74" x2="855.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414" y1="1170.91" x2="1398" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414" y1="1170.91" x2="1398" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414" y1="1170.91" x2="1430" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414" y1="1170.91" x2="1430" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="844.925" y1="1383.63" x2="828.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="844.925" y1="1383.63" x2="828.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="844.925" y1="1383.63" x2="860.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
-<line clip-path="url(#clip072)" x1="844.925" y1="1383.63" x2="860.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="327.236,2261.17 327.236,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="640.095,2261.17 640.095,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="952.953,2261.17 952.953,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1265.81,2261.17 1265.81,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1578.67,2261.17 1578.67,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1891.53,2261.17 1891.53,47.2441 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2204.39,2261.17 2204.39,47.2441 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 2352.76,2261.17 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="327.236,2261.17 327.236,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="640.095,2261.17 640.095,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="952.953,2261.17 952.953,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1265.81,2261.17 1265.81,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1578.67,2261.17 1578.67,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1891.53,2261.17 1891.53,2242.27 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2204.39,2261.17 2204.39,2242.27 "/>
+<path clip-path="url(#clip890)" d="M281.299 2306.74 L310.975 2306.74 L310.975 2310.67 L281.299 2310.67 L281.299 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M335.234 2304.93 Q338.59 2305.65 340.465 2307.92 Q342.363 2310.19 342.363 2313.52 Q342.363 2318.64 338.845 2321.44 Q335.326 2324.24 328.845 2324.24 Q326.669 2324.24 324.354 2323.8 Q322.063 2323.38 319.609 2322.53 L319.609 2318.01 Q321.553 2319.15 323.868 2319.73 Q326.183 2320.3 328.706 2320.3 Q333.104 2320.3 335.396 2318.57 Q337.711 2316.83 337.711 2313.52 Q337.711 2310.47 335.558 2308.75 Q333.428 2307.02 329.609 2307.02 L325.581 2307.02 L325.581 2303.17 L329.794 2303.17 Q333.243 2303.17 335.072 2301.81 Q336.9 2300.42 336.9 2297.83 Q336.9 2295.17 335.002 2293.75 Q333.127 2292.32 329.609 2292.32 Q327.688 2292.32 325.488 2292.73 Q323.289 2293.15 320.651 2294.03 L320.651 2289.86 Q323.313 2289.12 325.627 2288.75 Q327.965 2288.38 330.025 2288.38 Q335.35 2288.38 338.451 2290.81 Q341.553 2293.22 341.553 2297.34 Q341.553 2300.21 339.91 2302.2 Q338.266 2304.17 335.234 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M361.229 2292.09 Q357.618 2292.09 355.789 2295.65 Q353.984 2299.19 353.984 2306.32 Q353.984 2313.43 355.789 2316.99 Q357.618 2320.54 361.229 2320.54 Q364.863 2320.54 366.669 2316.99 Q368.497 2313.43 368.497 2306.32 Q368.497 2299.19 366.669 2295.65 Q364.863 2292.09 361.229 2292.09 M361.229 2288.38 Q367.039 2288.38 370.095 2292.99 Q373.173 2297.57 373.173 2306.32 Q373.173 2315.05 370.095 2319.66 Q367.039 2324.24 361.229 2324.24 Q355.419 2324.24 352.34 2319.66 Q349.285 2315.05 349.285 2306.32 Q349.285 2297.57 352.34 2292.99 Q355.419 2288.38 361.229 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M594.157 2306.74 L623.833 2306.74 L623.833 2310.67 L594.157 2310.67 L594.157 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M637.953 2319.63 L654.273 2319.63 L654.273 2323.57 L632.328 2323.57 L632.328 2319.63 Q634.99 2316.88 639.574 2312.25 Q644.18 2307.6 645.361 2306.25 Q647.606 2303.73 648.486 2301.99 Q649.388 2300.23 649.388 2298.54 Q649.388 2295.79 647.444 2294.05 Q645.523 2292.32 642.421 2292.32 Q640.222 2292.32 637.768 2293.08 Q635.338 2293.85 632.56 2295.4 L632.56 2290.67 Q635.384 2289.54 637.838 2288.96 Q640.291 2288.38 642.328 2288.38 Q647.699 2288.38 650.893 2291.07 Q654.087 2293.75 654.087 2298.24 Q654.087 2300.37 653.277 2302.29 Q652.49 2304.19 650.384 2306.79 Q649.805 2307.46 646.703 2310.67 Q643.601 2313.87 637.953 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M674.087 2292.09 Q670.476 2292.09 668.648 2295.65 Q666.842 2299.19 666.842 2306.32 Q666.842 2313.43 668.648 2316.99 Q670.476 2320.54 674.087 2320.54 Q677.722 2320.54 679.527 2316.99 Q681.356 2313.43 681.356 2306.32 Q681.356 2299.19 679.527 2295.65 Q677.722 2292.09 674.087 2292.09 M674.087 2288.38 Q679.898 2288.38 682.953 2292.99 Q686.032 2297.57 686.032 2306.32 Q686.032 2315.05 682.953 2319.66 Q679.898 2324.24 674.087 2324.24 Q668.277 2324.24 665.199 2319.66 Q662.143 2315.05 662.143 2306.32 Q662.143 2297.57 665.199 2292.99 Q668.277 2288.38 674.087 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M907.016 2306.74 L936.691 2306.74 L936.691 2310.67 L907.016 2310.67 L907.016 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M947.594 2319.63 L955.233 2319.63 L955.233 2293.27 L946.923 2294.93 L946.923 2290.67 L955.187 2289.01 L959.863 2289.01 L959.863 2319.63 L967.501 2319.63 L967.501 2323.57 L947.594 2323.57 L947.594 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M986.946 2292.09 Q983.335 2292.09 981.506 2295.65 Q979.7 2299.19 979.7 2306.32 Q979.7 2313.43 981.506 2316.99 Q983.335 2320.54 986.946 2320.54 Q990.58 2320.54 992.386 2316.99 Q994.214 2313.43 994.214 2306.32 Q994.214 2299.19 992.386 2295.65 Q990.58 2292.09 986.946 2292.09 M986.946 2288.38 Q992.756 2288.38 995.811 2292.99 Q998.89 2297.57 998.89 2306.32 Q998.89 2315.05 995.811 2319.66 Q992.756 2324.24 986.946 2324.24 Q981.136 2324.24 978.057 2319.66 Q975.001 2315.05 975.001 2306.32 Q975.001 2297.57 978.057 2292.99 Q981.136 2288.38 986.946 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1265.81 2292.09 Q1262.2 2292.09 1260.37 2295.65 Q1258.57 2299.19 1258.57 2306.32 Q1258.57 2313.43 1260.37 2316.99 Q1262.2 2320.54 1265.81 2320.54 Q1269.45 2320.54 1271.25 2316.99 Q1273.08 2313.43 1273.08 2306.32 Q1273.08 2299.19 1271.25 2295.65 Q1269.45 2292.09 1265.81 2292.09 M1265.81 2288.38 Q1271.62 2288.38 1274.68 2292.99 Q1277.76 2297.57 1277.76 2306.32 Q1277.76 2315.05 1274.68 2319.66 Q1271.62 2324.24 1265.81 2324.24 Q1260 2324.24 1256.92 2319.66 Q1253.87 2315.05 1253.87 2306.32 Q1253.87 2297.57 1256.92 2292.99 Q1260 2288.38 1265.81 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1553.36 2319.63 L1561 2319.63 L1561 2293.27 L1552.69 2294.93 L1552.69 2290.67 L1560.95 2289.01 L1565.63 2289.01 L1565.63 2319.63 L1573.26 2319.63 L1573.26 2323.57 L1553.36 2323.57 L1553.36 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1592.71 2292.09 Q1589.1 2292.09 1587.27 2295.65 Q1585.46 2299.19 1585.46 2306.32 Q1585.46 2313.43 1587.27 2316.99 Q1589.1 2320.54 1592.71 2320.54 Q1596.34 2320.54 1598.15 2316.99 Q1599.98 2313.43 1599.98 2306.32 Q1599.98 2299.19 1598.15 2295.65 Q1596.34 2292.09 1592.71 2292.09 M1592.71 2288.38 Q1598.52 2288.38 1601.57 2292.99 Q1604.65 2297.57 1604.65 2306.32 Q1604.65 2315.05 1601.57 2319.66 Q1598.52 2324.24 1592.71 2324.24 Q1586.9 2324.24 1583.82 2319.66 Q1580.76 2315.05 1580.76 2306.32 Q1580.76 2297.57 1583.82 2292.99 Q1586.9 2288.38 1592.71 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1870.3 2319.63 L1886.62 2319.63 L1886.62 2323.57 L1864.68 2323.57 L1864.68 2319.63 Q1867.34 2316.88 1871.92 2312.25 Q1876.53 2307.6 1877.71 2306.25 Q1879.95 2303.73 1880.83 2301.99 Q1881.74 2300.23 1881.74 2298.54 Q1881.74 2295.79 1879.79 2294.05 Q1877.87 2292.32 1874.77 2292.32 Q1872.57 2292.32 1870.12 2293.08 Q1867.69 2293.85 1864.91 2295.4 L1864.91 2290.67 Q1867.73 2289.54 1870.19 2288.96 Q1872.64 2288.38 1874.68 2288.38 Q1880.05 2288.38 1883.24 2291.07 Q1886.44 2293.75 1886.44 2298.24 Q1886.44 2300.37 1885.63 2302.29 Q1884.84 2304.19 1882.73 2306.79 Q1882.15 2307.46 1879.05 2310.67 Q1875.95 2313.87 1870.3 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M1906.44 2292.09 Q1902.82 2292.09 1901 2295.65 Q1899.19 2299.19 1899.19 2306.32 Q1899.19 2313.43 1901 2316.99 Q1902.82 2320.54 1906.44 2320.54 Q1910.07 2320.54 1911.88 2316.99 Q1913.7 2313.43 1913.7 2306.32 Q1913.7 2299.19 1911.88 2295.65 Q1910.07 2292.09 1906.44 2292.09 M1906.44 2288.38 Q1912.25 2288.38 1915.3 2292.99 Q1918.38 2297.57 1918.38 2306.32 Q1918.38 2315.05 1915.3 2319.66 Q1912.25 2324.24 1906.44 2324.24 Q1900.63 2324.24 1897.55 2319.66 Q1894.49 2315.05 1894.49 2306.32 Q1894.49 2297.57 1897.55 2292.99 Q1900.63 2288.38 1906.44 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M2193.23 2304.93 Q2196.59 2305.65 2198.46 2307.92 Q2200.36 2310.19 2200.36 2313.52 Q2200.36 2318.64 2196.84 2321.44 Q2193.32 2324.24 2186.84 2324.24 Q2184.66 2324.24 2182.35 2323.8 Q2180.06 2323.38 2177.6 2322.53 L2177.6 2318.01 Q2179.55 2319.15 2181.86 2319.73 Q2184.18 2320.3 2186.7 2320.3 Q2191.1 2320.3 2193.39 2318.57 Q2195.71 2316.83 2195.71 2313.52 Q2195.71 2310.47 2193.55 2308.75 Q2191.42 2307.02 2187.6 2307.02 L2183.58 2307.02 L2183.58 2303.17 L2187.79 2303.17 Q2191.24 2303.17 2193.07 2301.81 Q2194.9 2300.42 2194.9 2297.83 Q2194.9 2295.17 2193 2293.75 Q2191.12 2292.32 2187.6 2292.32 Q2185.68 2292.32 2183.48 2292.73 Q2181.28 2293.15 2178.65 2294.03 L2178.65 2289.86 Q2181.31 2289.12 2183.62 2288.75 Q2185.96 2288.38 2188.02 2288.38 Q2193.34 2288.38 2196.45 2290.81 Q2199.55 2293.22 2199.55 2297.34 Q2199.55 2300.21 2197.9 2302.2 Q2196.26 2304.17 2193.23 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M2219.22 2292.09 Q2215.61 2292.09 2213.78 2295.65 Q2211.98 2299.19 2211.98 2306.32 Q2211.98 2313.43 2213.78 2316.99 Q2215.61 2320.54 2219.22 2320.54 Q2222.86 2320.54 2224.66 2316.99 Q2226.49 2313.43 2226.49 2306.32 Q2226.49 2299.19 2224.66 2295.65 Q2222.86 2292.09 2219.22 2292.09 M2219.22 2288.38 Q2225.03 2288.38 2228.09 2292.99 Q2231.17 2297.57 2231.17 2306.32 Q2231.17 2315.05 2228.09 2319.66 Q2225.03 2324.24 2219.22 2324.24 Q2213.41 2324.24 2210.34 2319.66 Q2207.28 2315.05 2207.28 2306.32 Q2207.28 2297.57 2210.34 2292.99 Q2213.41 2288.38 2219.22 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,2110.07 2352.76,2110.07 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1791.45 2352.76,1791.45 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1472.83 2352.76,1472.83 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,1154.21 2352.76,1154.21 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,835.586 2352.76,835.586 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,516.966 2352.76,516.966 "/>
+<polyline clip-path="url(#clip892)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="178.867,198.346 2352.76,198.346 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2261.17 178.867,47.2441 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,2110.07 197.764,2110.07 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1791.45 197.764,1791.45 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1472.83 197.764,1472.83 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,1154.21 197.764,1154.21 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,835.586 197.764,835.586 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,516.966 197.764,516.966 "/>
+<polyline clip-path="url(#clip890)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="178.867,198.346 197.764,198.346 "/>
+<path clip-path="url(#clip890)" d="M50.9921 2110.52 L80.6679 2110.52 L80.6679 2114.45 L50.9921 2114.45 L50.9921 2110.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M104.927 2108.71 Q108.283 2109.43 110.158 2111.7 Q112.057 2113.97 112.057 2117.3 Q112.057 2122.42 108.538 2125.22 Q105.02 2128.02 98.5382 2128.02 Q96.3623 2128.02 94.0475 2127.58 Q91.7558 2127.16 89.3021 2126.3 L89.3021 2121.79 Q91.2465 2122.92 93.5613 2123.5 Q95.8761 2124.08 98.3993 2124.08 Q102.797 2124.08 105.089 2122.35 Q107.404 2120.61 107.404 2117.3 Q107.404 2114.24 105.251 2112.53 Q103.121 2110.8 99.3021 2110.8 L95.2743 2110.8 L95.2743 2106.95 L99.4872 2106.95 Q102.936 2106.95 104.765 2105.59 Q106.594 2104.2 106.594 2101.61 Q106.594 2098.94 104.696 2097.53 Q102.821 2096.1 99.3021 2096.1 Q97.3808 2096.1 95.1817 2096.51 Q92.9826 2096.93 90.3438 2097.81 L90.3438 2093.64 Q93.0058 2092.9 95.3206 2092.53 Q97.6585 2092.16 99.7187 2092.16 Q105.043 2092.16 108.145 2094.59 Q111.246 2097 111.246 2101.12 Q111.246 2103.99 109.603 2105.98 Q107.959 2107.95 104.927 2108.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 2095.86 Q127.311 2095.86 125.482 2099.43 Q123.677 2102.97 123.677 2110.1 Q123.677 2117.21 125.482 2120.77 Q127.311 2124.31 130.922 2124.31 Q134.556 2124.31 136.362 2120.77 Q138.191 2117.21 138.191 2110.1 Q138.191 2102.97 136.362 2099.43 Q134.556 2095.86 130.922 2095.86 M130.922 2092.16 Q136.732 2092.16 139.788 2096.77 Q142.867 2101.35 142.867 2110.1 Q142.867 2118.83 139.788 2123.43 Q136.732 2128.02 130.922 2128.02 Q125.112 2128.02 122.033 2123.43 Q118.978 2118.83 118.978 2110.1 Q118.978 2101.35 122.033 2096.77 Q125.112 2092.16 130.922 2092.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M50.9921 1791.9 L80.6679 1791.9 L80.6679 1795.83 L50.9921 1795.83 L50.9921 1791.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M94.7882 1804.79 L111.108 1804.79 L111.108 1808.73 L89.1632 1808.73 L89.1632 1804.79 Q91.8252 1802.04 96.4085 1797.41 Q101.015 1792.75 102.196 1791.41 Q104.441 1788.89 105.321 1787.15 Q106.223 1785.39 106.223 1783.7 Q106.223 1780.95 104.279 1779.21 Q102.358 1777.48 99.2558 1777.48 Q97.0567 1777.48 94.603 1778.24 Q92.1725 1779 89.3947 1780.55 L89.3947 1775.83 Q92.2188 1774.7 94.6724 1774.12 Q97.1261 1773.54 99.1632 1773.54 Q104.534 1773.54 107.728 1776.23 Q110.922 1778.91 110.922 1783.4 Q110.922 1785.53 110.112 1787.45 Q109.325 1789.35 107.219 1791.94 Q106.64 1792.61 103.538 1795.83 Q100.436 1799.03 94.7882 1804.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 1777.24 Q127.311 1777.24 125.482 1780.81 Q123.677 1784.35 123.677 1791.48 Q123.677 1798.59 125.482 1802.15 Q127.311 1805.69 130.922 1805.69 Q134.556 1805.69 136.362 1802.15 Q138.191 1798.59 138.191 1791.48 Q138.191 1784.35 136.362 1780.81 Q134.556 1777.24 130.922 1777.24 M130.922 1773.54 Q136.732 1773.54 139.788 1778.15 Q142.867 1782.73 142.867 1791.48 Q142.867 1800.21 139.788 1804.81 Q136.732 1809.4 130.922 1809.4 Q125.112 1809.4 122.033 1804.81 Q118.978 1800.21 118.978 1791.48 Q118.978 1782.73 122.033 1778.15 Q125.112 1773.54 130.922 1773.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M50.9921 1473.28 L80.6679 1473.28 L80.6679 1477.21 L50.9921 1477.21 L50.9921 1473.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M91.5706 1486.17 L99.2095 1486.17 L99.2095 1459.81 L90.8993 1461.47 L90.8993 1457.21 L99.1632 1455.55 L103.839 1455.55 L103.839 1486.17 L111.478 1486.17 L111.478 1490.11 L91.5706 1490.11 L91.5706 1486.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 1458.62 Q127.311 1458.62 125.482 1462.19 Q123.677 1465.73 123.677 1472.86 Q123.677 1479.97 125.482 1483.53 Q127.311 1487.07 130.922 1487.07 Q134.556 1487.07 136.362 1483.53 Q138.191 1479.97 138.191 1472.86 Q138.191 1465.73 136.362 1462.19 Q134.556 1458.62 130.922 1458.62 M130.922 1454.92 Q136.732 1454.92 139.788 1459.53 Q142.867 1464.11 142.867 1472.86 Q142.867 1481.59 139.788 1486.19 Q136.732 1490.78 130.922 1490.78 Q125.112 1490.78 122.033 1486.19 Q118.978 1481.59 118.978 1472.86 Q118.978 1464.11 122.033 1459.53 Q125.112 1454.92 130.922 1454.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 1140 Q127.311 1140 125.482 1143.57 Q123.677 1147.11 123.677 1154.24 Q123.677 1161.35 125.482 1164.91 Q127.311 1168.45 130.922 1168.45 Q134.556 1168.45 136.362 1164.91 Q138.191 1161.35 138.191 1154.24 Q138.191 1147.11 136.362 1143.57 Q134.556 1140 130.922 1140 M130.922 1136.3 Q136.732 1136.3 139.788 1140.91 Q142.867 1145.49 142.867 1154.24 Q142.867 1162.97 139.788 1167.57 Q136.732 1172.16 130.922 1172.16 Q125.112 1172.16 122.033 1167.57 Q118.978 1162.97 118.978 1154.24 Q118.978 1145.49 122.033 1140.91 Q125.112 1136.3 130.922 1136.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M91.5706 848.931 L99.2095 848.931 L99.2095 822.565 L90.8993 824.232 L90.8993 819.973 L99.1632 818.306 L103.839 818.306 L103.839 848.931 L111.478 848.931 L111.478 852.866 L91.5706 852.866 L91.5706 848.931 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 821.385 Q127.311 821.385 125.482 824.949 Q123.677 828.491 123.677 835.621 Q123.677 842.727 125.482 846.292 Q127.311 849.834 130.922 849.834 Q134.556 849.834 136.362 846.292 Q138.191 842.727 138.191 835.621 Q138.191 828.491 136.362 824.949 Q134.556 821.385 130.922 821.385 M130.922 817.681 Q136.732 817.681 139.788 822.287 Q142.867 826.871 142.867 835.621 Q142.867 844.348 139.788 848.954 Q136.732 853.537 130.922 853.537 Q125.112 853.537 122.033 848.954 Q118.978 844.348 118.978 835.621 Q118.978 826.871 122.033 822.287 Q125.112 817.681 130.922 817.681 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M94.7882 530.311 L111.108 530.311 L111.108 534.246 L89.1632 534.246 L89.1632 530.311 Q91.8252 527.556 96.4085 522.927 Q101.015 518.274 102.196 516.931 Q104.441 514.408 105.321 512.672 Q106.223 510.913 106.223 509.223 Q106.223 506.468 104.279 504.732 Q102.358 502.996 99.2558 502.996 Q97.0567 502.996 94.603 503.76 Q92.1725 504.524 89.3947 506.075 L89.3947 501.353 Q92.2188 500.218 94.6724 499.64 Q97.1261 499.061 99.1632 499.061 Q104.534 499.061 107.728 501.746 Q110.922 504.431 110.922 508.922 Q110.922 511.052 110.112 512.973 Q109.325 514.871 107.219 517.464 Q106.64 518.135 103.538 521.353 Q100.436 524.547 94.7882 530.311 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 502.765 Q127.311 502.765 125.482 506.329 Q123.677 509.871 123.677 517.001 Q123.677 524.107 125.482 527.672 Q127.311 531.214 130.922 531.214 Q134.556 531.214 136.362 527.672 Q138.191 524.107 138.191 517.001 Q138.191 509.871 136.362 506.329 Q134.556 502.765 130.922 502.765 M130.922 499.061 Q136.732 499.061 139.788 503.667 Q142.867 508.251 142.867 517.001 Q142.867 525.727 139.788 530.334 Q136.732 534.917 130.922 534.917 Q125.112 534.917 122.033 530.334 Q118.978 525.727 118.978 517.001 Q118.978 508.251 122.033 503.667 Q125.112 499.061 130.922 499.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M104.927 196.992 Q108.283 197.709 110.158 199.978 Q112.057 202.246 112.057 205.58 Q112.057 210.695 108.538 213.496 Q105.02 216.297 98.5382 216.297 Q96.3623 216.297 94.0475 215.857 Q91.7558 215.441 89.3021 214.584 L89.3021 210.07 Q91.2465 211.205 93.5613 211.783 Q95.8761 212.362 98.3993 212.362 Q102.797 212.362 105.089 210.626 Q107.404 208.89 107.404 205.58 Q107.404 202.524 105.251 200.811 Q103.121 199.075 99.3021 199.075 L95.2743 199.075 L95.2743 195.233 L99.4872 195.233 Q102.936 195.233 104.765 193.867 Q106.594 192.478 106.594 189.885 Q106.594 187.223 104.696 185.811 Q102.821 184.376 99.3021 184.376 Q97.3808 184.376 95.1817 184.793 Q92.9826 185.209 90.3438 186.089 L90.3438 181.922 Q93.0058 181.182 95.3206 180.811 Q97.6585 180.441 99.7187 180.441 Q105.043 180.441 108.145 182.872 Q111.246 185.279 111.246 189.399 Q111.246 192.27 109.603 194.26 Q107.959 196.228 104.927 196.992 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip890)" d="M130.922 184.145 Q127.311 184.145 125.482 187.709 Q123.677 191.251 123.677 198.381 Q123.677 205.487 125.482 209.052 Q127.311 212.594 130.922 212.594 Q134.556 212.594 136.362 209.052 Q138.191 205.487 138.191 198.381 Q138.191 191.251 136.362 187.709 Q134.556 184.145 130.922 184.145 M130.922 180.441 Q136.732 180.441 139.788 185.047 Q142.867 189.631 142.867 198.381 Q142.867 207.107 139.788 211.714 Q136.732 216.297 130.922 216.297 Q125.112 216.297 122.033 211.714 Q118.978 207.107 118.978 198.381 Q118.978 189.631 122.033 185.047 Q125.112 180.441 130.922 180.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><line clip-path="url(#clip892)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.44" y1="1186.36" x2="1375.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1202.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.44" y1="1186.36" x2="1407.44" y2="1170.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1372.17" y1="1098.14" x2="1356.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1114.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1372.17" y1="1098.14" x2="1388.17" y2="1082.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.37" y1="1262.61" x2="1095.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1278.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.37" y1="1262.61" x2="1127.37" y2="1246.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1229.97" x2="1288.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1245.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1229.97" x2="1320.99" y2="1213.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1053.12" y1="1313.7" x2="1037.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1329.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1053.12" y1="1313.7" x2="1069.12" y2="1297.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1388.9" y1="1173.08" x2="1372.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1189.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1388.9" y1="1173.08" x2="1404.9" y2="1157.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1926.4" y1="856.923" x2="1910.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1926.4" y1="856.923" x2="1910.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1926.4" y1="856.923" x2="1942.4" y2="872.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1926.4" y1="856.923" x2="1942.4" y2="840.923" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="912.69" y1="1348.83" x2="896.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="912.69" y1="1348.83" x2="896.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="912.69" y1="1348.83" x2="928.69" y2="1364.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="912.69" y1="1348.83" x2="928.69" y2="1332.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.77" y1="1174.01" x2="1343.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1190.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.77" y1="1174.01" x2="1375.77" y2="1158.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="625.575" y1="1469.54" x2="609.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="625.575" y1="1469.54" x2="609.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="625.575" y1="1469.54" x2="641.575" y2="1485.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="625.575" y1="1469.54" x2="641.575" y2="1453.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.02" y1="1148.47" x2="1438.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1164.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.02" y1="1148.47" x2="1470.02" y2="1132.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.94" y1="1144.11" x2="1399.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1160.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.94" y1="1144.11" x2="1431.94" y2="1128.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.61" y1="1071.18" x2="1413.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1087.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.61" y1="1071.18" x2="1445.61" y2="1055.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.18" y1="1043.75" x2="1535.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1059.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.18" y1="1043.75" x2="1567.18" y2="1027.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1383.83" y1="1070.02" x2="1367.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1086.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1383.83" y1="1070.02" x2="1399.83" y2="1054.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.68" y1="1259.28" x2="1120.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1275.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.68" y1="1259.28" x2="1152.68" y2="1243.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.69" y1="1027.81" x2="1325.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1043.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.69" y1="1027.81" x2="1357.69" y2="1011.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1260.55" y1="1177.07" x2="1244.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1193.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1260.55" y1="1177.07" x2="1276.55" y2="1161.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.36" y1="1138.56" x2="990.356" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.36" y1="1138.56" x2="990.356" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1154.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.36" y1="1138.56" x2="1022.36" y2="1122.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="785.24" y1="1436.86" x2="769.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="785.24" y1="1436.86" x2="769.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="785.24" y1="1436.86" x2="801.24" y2="1452.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="785.24" y1="1436.86" x2="801.24" y2="1420.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1271.22" y1="1086.79" x2="1255.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1102.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1271.22" y1="1086.79" x2="1287.22" y2="1070.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1019.17" y1="1201.94" x2="1003.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1217.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1019.17" y1="1201.94" x2="1035.17" y2="1185.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.289" y1="1394.75" x2="761.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.289" y1="1394.75" x2="761.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.289" y1="1394.75" x2="793.289" y2="1410.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.289" y1="1394.75" x2="793.289" y2="1378.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1511.54" y1="1080.45" x2="1495.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1096.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1511.54" y1="1080.45" x2="1527.54" y2="1064.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1461.76" y1="1072.86" x2="1445.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1088.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1461.76" y1="1072.86" x2="1477.76" y2="1056.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.41" y1="1014.15" x2="1369.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.41" y1="1014.15" x2="1369.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.41" y1="1014.15" x2="1401.41" y2="1030.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.41" y1="1014.15" x2="1401.41" y2="998.153" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.61" y1="1146.61" x2="1222.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.61" y1="1146.61" x2="1254.61" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1612.46" y1="913.718" x2="1596.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1612.46" y1="913.718" x2="1596.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1612.46" y1="913.718" x2="1628.46" y2="929.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1612.46" y1="913.718" x2="1628.46" y2="897.718" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.2" y1="1238.54" x2="1271.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1254.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.2" y1="1238.54" x2="1303.2" y2="1222.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.35" y1="1265.07" x2="1043.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1281.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.35" y1="1265.07" x2="1075.35" y2="1249.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="947.412" y1="1370.03" x2="931.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="947.412" y1="1370.03" x2="931.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="947.412" y1="1370.03" x2="963.412" y2="1386.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="947.412" y1="1370.03" x2="963.412" y2="1354.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1251.92" y1="1163.59" x2="1235.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1179.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1251.92" y1="1163.59" x2="1267.92" y2="1147.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.13" y1="1210.6" x2="1222.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1226.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.13" y1="1210.6" x2="1254.13" y2="1194.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.497" y1="1483.85" x2="672.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.497" y1="1483.85" x2="672.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.497" y1="1483.85" x2="704.497" y2="1499.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.497" y1="1483.85" x2="704.497" y2="1467.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1248.4" y1="1166.04" x2="1232.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1182.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1248.4" y1="1166.04" x2="1264.4" y2="1150.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.95" y1="869.353" x2="1529.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.95" y1="869.353" x2="1529.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.95" y1="869.353" x2="1561.95" y2="885.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.95" y1="869.353" x2="1561.95" y2="853.353" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.93" y1="1067.84" x2="1405.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1083.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.93" y1="1067.84" x2="1437.93" y2="1051.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.39" y1="1032.06" x2="1456.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1048.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.39" y1="1032.06" x2="1488.39" y2="1016.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.2" y1="1038.88" x2="1321.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1054.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.2" y1="1038.88" x2="1353.2" y2="1022.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.09" y1="1178.03" x2="1030.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1194.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.09" y1="1178.03" x2="1062.09" y2="1162.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.45" y1="1275.47" x2="1078.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1291.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.45" y1="1275.47" x2="1110.45" y2="1259.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="771.564" y1="1431.34" x2="755.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="771.564" y1="1431.34" x2="755.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="771.564" y1="1431.34" x2="787.564" y2="1447.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="771.564" y1="1431.34" x2="787.564" y2="1415.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="918.674" y1="1363.31" x2="902.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="918.674" y1="1363.31" x2="902.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="918.674" y1="1363.31" x2="934.674" y2="1379.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="918.674" y1="1363.31" x2="934.674" y2="1347.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.43" y1="1296.62" x2="1209.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1312.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.43" y1="1296.62" x2="1241.43" y2="1280.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.5" y1="1071.7" x2="1228.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1087.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.5" y1="1071.7" x2="1260.5" y2="1055.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.11" y1="1109.8" x2="1348.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1125.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.11" y1="1109.8" x2="1380.11" y2="1093.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1707.08" y1="826.86" x2="1691.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1707.08" y1="826.86" x2="1691.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1707.08" y1="826.86" x2="1723.08" y2="842.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1707.08" y1="826.86" x2="1723.08" y2="810.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.63" y1="875.887" x2="1561.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.63" y1="875.887" x2="1561.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.63" y1="875.887" x2="1593.63" y2="891.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.63" y1="875.887" x2="1593.63" y2="859.887" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1319.95" y1="1118.19" x2="1303.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1134.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1319.95" y1="1118.19" x2="1335.95" y2="1102.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="894.72" y1="1142.51" x2="878.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="894.72" y1="1142.51" x2="878.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="894.72" y1="1142.51" x2="910.72" y2="1158.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="894.72" y1="1142.51" x2="910.72" y2="1126.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.17" y1="1345.5" x2="1051.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1361.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.17" y1="1345.5" x2="1083.17" y2="1329.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1580.8" y1="990.127" x2="1564.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1580.8" y1="990.127" x2="1564.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1580.8" y1="990.127" x2="1596.8" y2="1006.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1580.8" y1="990.127" x2="1596.8" y2="974.127" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="852.02" y1="1283.64" x2="836.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="852.02" y1="1283.64" x2="836.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="852.02" y1="1283.64" x2="868.02" y2="1299.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="852.02" y1="1283.64" x2="868.02" y2="1267.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.48" y1="962.993" x2="1589.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.48" y1="962.993" x2="1589.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.48" y1="962.993" x2="1621.48" y2="978.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.48" y1="962.993" x2="1621.48" y2="946.993" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1283.38" y1="1220.65" x2="1267.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1236.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1283.38" y1="1220.65" x2="1299.38" y2="1204.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.15" y1="1016.63" x2="1446.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1032.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.15" y1="1016.63" x2="1478.15" y2="1000.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1172.56" y1="1009.01" x2="1156.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1172.56" y1="1009.01" x2="1156.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1172.56" y1="1009.01" x2="1188.56" y2="1025.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1172.56" y1="1009.01" x2="1188.56" y2="993.009" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.43" y1="1210.76" x2="1151.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1226.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.43" y1="1210.76" x2="1183.43" y2="1194.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.588" y1="1415.15" x2="672.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.588" y1="1415.15" x2="672.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.588" y1="1415.15" x2="704.588" y2="1431.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="688.588" y1="1415.15" x2="704.588" y2="1399.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.99" y1="1289.47" x2="1120.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1305.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1136.99" y1="1289.47" x2="1152.99" y2="1273.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="991.105" y1="1402.23" x2="975.105" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="991.105" y1="1402.23" x2="975.105" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="991.105" y1="1402.23" x2="1007.11" y2="1418.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="991.105" y1="1402.23" x2="1007.11" y2="1386.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1811.99" y1="919.23" x2="1795.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1811.99" y1="919.23" x2="1795.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1811.99" y1="919.23" x2="1827.99" y2="935.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1811.99" y1="919.23" x2="1827.99" y2="903.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.971" y1="1350.77" x2="829.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.971" y1="1350.77" x2="829.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.971" y1="1350.77" x2="861.971" y2="1366.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.971" y1="1350.77" x2="861.971" y2="1334.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1144.78" y1="1281.78" x2="1128.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1297.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1144.78" y1="1281.78" x2="1160.78" y2="1265.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1064.78" y1="1413.61" x2="1048.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1429.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1064.78" y1="1413.61" x2="1080.78" y2="1397.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="886.864" y1="1368.41" x2="870.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="886.864" y1="1368.41" x2="870.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="886.864" y1="1368.41" x2="902.864" y2="1384.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="886.864" y1="1368.41" x2="902.864" y2="1352.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.7" y1="1200.72" x2="1324.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1216.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.7" y1="1200.72" x2="1356.7" y2="1184.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.27" y1="1196.6" x2="1051.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1212.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.27" y1="1196.6" x2="1083.27" y2="1180.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.009" y1="1388.03" x2="761.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.009" y1="1388.03" x2="761.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.009" y1="1388.03" x2="793.009" y2="1404.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="777.009" y1="1388.03" x2="793.009" y2="1372.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.23" y1="998.156" x2="1277.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.23" y1="998.156" x2="1277.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.23" y1="998.156" x2="1309.23" y2="1014.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.23" y1="998.156" x2="1309.23" y2="982.156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="885.196" y1="1405.14" x2="869.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="885.196" y1="1405.14" x2="869.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="885.196" y1="1405.14" x2="901.196" y2="1421.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="885.196" y1="1405.14" x2="901.196" y2="1389.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1204.07" y1="1358.98" x2="1188.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1374.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1204.07" y1="1358.98" x2="1220.07" y2="1342.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1195.08" y1="1292.67" x2="1179.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1308.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1195.08" y1="1292.67" x2="1211.08" y2="1276.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.55" y1="853.956" x2="1441.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.55" y1="853.956" x2="1441.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.55" y1="853.956" x2="1473.55" y2="869.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.55" y1="853.956" x2="1473.55" y2="837.956" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.86" y1="1030.17" x2="1603.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1046.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.86" y1="1030.17" x2="1635.86" y2="1014.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="949.006" y1="1440.98" x2="933.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="949.006" y1="1440.98" x2="933.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="949.006" y1="1440.98" x2="965.006" y2="1456.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="949.006" y1="1440.98" x2="965.006" y2="1424.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.411" y1="1427.9" x2="829.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.411" y1="1427.9" x2="829.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.411" y1="1427.9" x2="861.411" y2="1443.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="845.411" y1="1427.9" x2="861.411" y2="1411.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.64" y1="1292.35" x2="1169.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1308.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.64" y1="1292.35" x2="1201.64" y2="1276.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.57" y1="868.321" x2="1530.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.57" y1="868.321" x2="1530.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.57" y1="868.321" x2="1562.57" y2="884.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.57" y1="868.321" x2="1562.57" y2="852.321" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.43" y1="1342.43" x2="834.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.43" y1="1342.43" x2="834.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.43" y1="1342.43" x2="866.43" y2="1358.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.43" y1="1342.43" x2="866.43" y2="1326.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.87" y1="1122.39" x2="1535.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1138.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1551.87" y1="1122.39" x2="1567.87" y2="1106.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.5" y1="829.178" x2="1603.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.5" y1="829.178" x2="1603.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.5" y1="829.178" x2="1635.5" y2="845.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1619.5" y1="829.178" x2="1635.5" y2="813.178" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1847.7" y1="804.547" x2="1831.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1847.7" y1="804.547" x2="1831.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1847.7" y1="804.547" x2="1863.7" y2="820.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1847.7" y1="804.547" x2="1863.7" y2="788.547" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="992.659" y1="1251.94" x2="976.659" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="992.659" y1="1251.94" x2="976.659" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="992.659" y1="1251.94" x2="1008.66" y2="1267.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="992.659" y1="1251.94" x2="1008.66" y2="1235.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.79" y1="1033.06" x2="1595.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1049.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.79" y1="1033.06" x2="1627.79" y2="1017.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1148.89" y1="1269.81" x2="1132.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1285.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1148.89" y1="1269.81" x2="1164.89" y2="1253.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1202.15" y1="1212.09" x2="1186.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1228.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1202.15" y1="1212.09" x2="1218.15" y2="1196.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1641.07" y1="860.861" x2="1625.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1641.07" y1="860.861" x2="1625.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1641.07" y1="860.861" x2="1657.07" y2="876.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1641.07" y1="860.861" x2="1657.07" y2="844.861" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.79" y1="1035.93" x2="1470.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1051.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.79" y1="1035.93" x2="1502.79" y2="1019.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1207.15" y1="1160.11" x2="1191.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1176.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1207.15" y1="1160.11" x2="1223.15" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1119.38" y1="1194.83" x2="1103.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1210.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1119.38" y1="1194.83" x2="1135.38" y2="1178.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1332.96" y1="1133.97" x2="1316.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1149.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1332.96" y1="1133.97" x2="1348.96" y2="1117.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1301.11" y1="987.604" x2="1285.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1301.11" y1="987.604" x2="1285.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1301.11" y1="987.604" x2="1317.11" y2="1003.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1301.11" y1="987.604" x2="1317.11" y2="971.604" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.19" y1="1037.37" x2="1299.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1053.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.19" y1="1037.37" x2="1331.19" y2="1021.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1300.37" y1="1049.99" x2="1284.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1065.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1300.37" y1="1049.99" x2="1316.37" y2="1033.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.97" y1="1034.41" x2="1347.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1050.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.97" y1="1034.41" x2="1379.97" y2="1018.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.47" y1="1159.14" x2="1328.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1175.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.47" y1="1159.14" x2="1360.47" y2="1143.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1673.06" y1="901.944" x2="1657.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1673.06" y1="901.944" x2="1657.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1673.06" y1="901.944" x2="1689.06" y2="917.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1673.06" y1="901.944" x2="1689.06" y2="885.944" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.89" y1="961.872" x2="1595.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.89" y1="961.872" x2="1595.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.89" y1="961.872" x2="1627.89" y2="977.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.89" y1="961.872" x2="1627.89" y2="945.872" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.83" y1="1129.04" x2="1175.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1145.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.83" y1="1129.04" x2="1207.83" y2="1113.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.08" y1="1021.63" x2="1231.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1037.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.08" y1="1021.63" x2="1263.08" y2="1005.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1040.17" y1="1275.11" x2="1024.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1291.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1040.17" y1="1275.11" x2="1056.17" y2="1259.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="782.176" y1="1390.69" x2="766.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="782.176" y1="1390.69" x2="766.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="782.176" y1="1390.69" x2="798.176" y2="1406.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="782.176" y1="1390.69" x2="798.176" y2="1374.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.77" y1="921.537" x2="1659.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.77" y1="921.537" x2="1659.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.77" y1="921.537" x2="1691.77" y2="937.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.77" y1="921.537" x2="1691.77" y2="905.537" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.19" y1="1173.52" x2="1214.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1189.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.19" y1="1173.52" x2="1246.19" y2="1157.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1092.98" y1="1136.22" x2="1076.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1152.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1092.98" y1="1136.22" x2="1108.98" y2="1120.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="816.257" y1="1346.85" x2="800.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="816.257" y1="1346.85" x2="800.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="816.257" y1="1346.85" x2="832.257" y2="1362.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="816.257" y1="1346.85" x2="832.257" y2="1330.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="724.712" y1="1453.02" x2="708.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="724.712" y1="1453.02" x2="708.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="724.712" y1="1453.02" x2="740.712" y2="1469.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="724.712" y1="1453.02" x2="740.712" y2="1437.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.34" y1="968.47" x2="1659.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.34" y1="968.47" x2="1659.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.34" y1="968.47" x2="1691.34" y2="984.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1675.34" y1="968.47" x2="1691.34" y2="952.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.82" y1="1198.32" x2="1130.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1214.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.82" y1="1198.32" x2="1162.82" y2="1182.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.2" y1="1016.21" x2="1484.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1032.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.2" y1="1016.21" x2="1516.2" y2="1000.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.396" y1="1435.49" x2="858.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.396" y1="1435.49" x2="858.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.396" y1="1435.49" x2="890.396" y2="1451.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.396" y1="1435.49" x2="890.396" y2="1419.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.29" y1="1203.85" x2="1271.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1219.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1287.29" y1="1203.85" x2="1303.29" y2="1187.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1098.58" y1="1260.55" x2="1082.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1276.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1098.58" y1="1260.55" x2="1114.58" y2="1244.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1330.6" y1="1013.95" x2="1314.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1330.6" y1="1013.95" x2="1314.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1330.6" y1="1013.95" x2="1346.6" y2="1029.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1330.6" y1="1013.95" x2="1346.6" y2="997.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.5" y1="1058.8" x2="1070.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1074.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.5" y1="1058.8" x2="1102.5" y2="1042.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1250.5" y1="1043.25" x2="1234.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1059.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1250.5" y1="1043.25" x2="1266.5" y2="1027.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1626.12" y1="969.424" x2="1610.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1626.12" y1="969.424" x2="1610.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1626.12" y1="969.424" x2="1642.12" y2="985.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1626.12" y1="969.424" x2="1642.12" y2="953.424" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.09" y1="1136.96" x2="1468.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1152.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.09" y1="1136.96" x2="1500.09" y2="1120.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="933.271" y1="1353.8" x2="917.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="933.271" y1="1353.8" x2="917.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="933.271" y1="1353.8" x2="949.271" y2="1369.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="933.271" y1="1353.8" x2="949.271" y2="1337.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1027.37" y1="1280.74" x2="1011.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1296.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1027.37" y1="1280.74" x2="1043.37" y2="1264.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1346.23" y1="1131.49" x2="1330.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1147.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1346.23" y1="1131.49" x2="1362.23" y2="1115.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="906.639" y1="1451.01" x2="890.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="906.639" y1="1451.01" x2="890.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="906.639" y1="1451.01" x2="922.639" y2="1467.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="906.639" y1="1451.01" x2="922.639" y2="1435.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.76" y1="1104.6" x2="1405.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1120.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.76" y1="1104.6" x2="1437.76" y2="1088.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="958.996" y1="1406.94" x2="942.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="958.996" y1="1406.94" x2="942.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="958.996" y1="1406.94" x2="974.996" y2="1422.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="958.996" y1="1406.94" x2="974.996" y2="1390.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.277" y1="1312.14" x2="858.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.277" y1="1312.14" x2="858.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.277" y1="1312.14" x2="890.277" y2="1328.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="874.277" y1="1312.14" x2="890.277" y2="1296.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.691" y1="1424.89" x2="911.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.691" y1="1424.89" x2="911.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.691" y1="1424.89" x2="943.691" y2="1440.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.691" y1="1424.89" x2="943.691" y2="1408.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1412.66" y1="1067.26" x2="1396.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1083.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1412.66" y1="1067.26" x2="1428.66" y2="1051.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="538.801" y1="1550.28" x2="522.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="538.801" y1="1550.28" x2="522.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="538.801" y1="1550.28" x2="554.801" y2="1566.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="538.801" y1="1550.28" x2="554.801" y2="1534.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1294.59" y1="1109.44" x2="1278.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1125.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1294.59" y1="1109.44" x2="1310.59" y2="1093.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="869.402" y1="1512" x2="853.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="869.402" y1="1512" x2="853.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="869.402" y1="1512" x2="885.402" y2="1528" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="869.402" y1="1512" x2="885.402" y2="1496" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.688" y1="1677.8" x2="553.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.688" y1="1677.8" x2="553.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.688" y1="1677.8" x2="585.688" y2="1693.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.688" y1="1677.8" x2="585.688" y2="1661.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1124.14" y1="1170.71" x2="1108.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1186.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1124.14" y1="1170.71" x2="1140.14" y2="1154.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="915.573" y1="1448.5" x2="899.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="915.573" y1="1448.5" x2="899.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="915.573" y1="1448.5" x2="931.573" y2="1464.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="915.573" y1="1448.5" x2="931.573" y2="1432.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1520.69" y1="868.39" x2="1504.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1520.69" y1="868.39" x2="1504.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1520.69" y1="868.39" x2="1536.69" y2="884.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1520.69" y1="868.39" x2="1536.69" y2="852.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1108.04" y1="1185.31" x2="1092.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1201.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1108.04" y1="1185.31" x2="1124.04" y2="1169.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.79" y1="1007.52" x2="1446.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.79" y1="1007.52" x2="1446.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.79" y1="1007.52" x2="1478.79" y2="1023.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1462.79" y1="1007.52" x2="1478.79" y2="991.518" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.71" y1="986.213" x2="1427.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.71" y1="986.213" x2="1427.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.71" y1="986.213" x2="1459.71" y2="1002.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.71" y1="986.213" x2="1459.71" y2="970.213" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="831.587" y1="1424.87" x2="815.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="831.587" y1="1424.87" x2="815.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="831.587" y1="1424.87" x2="847.587" y2="1440.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="831.587" y1="1424.87" x2="847.587" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.58" y1="1341.26" x2="1014.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1357.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.58" y1="1341.26" x2="1046.58" y2="1325.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.23" y1="1063.53" x2="1436.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1079.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.23" y1="1063.53" x2="1468.23" y2="1047.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.66" y1="1110.16" x2="1062.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1126.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.66" y1="1110.16" x2="1094.66" y2="1094.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.93" y1="1006.01" x2="1593.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.93" y1="1006.01" x2="1593.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.93" y1="1006.01" x2="1625.93" y2="1022.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.93" y1="1006.01" x2="1625.93" y2="990.013" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1733.98" y1="723.632" x2="1717.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1733.98" y1="723.632" x2="1717.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1733.98" y1="723.632" x2="1749.98" y2="739.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1733.98" y1="723.632" x2="1749.98" y2="707.632" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.47" y1="1228.99" x2="1265.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1244.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.47" y1="1228.99" x2="1297.47" y2="1212.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.93" y1="1191.01" x2="1213.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1207.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.93" y1="1191.01" x2="1245.93" y2="1175.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.48" y1="1238.33" x2="1224.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1254.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.48" y1="1238.33" x2="1256.48" y2="1222.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.936" y1="1351.49" x2="834.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.936" y1="1351.49" x2="834.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.936" y1="1351.49" x2="866.936" y2="1367.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="850.936" y1="1351.49" x2="866.936" y2="1335.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.92" y1="896.324" x2="1752.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.92" y1="896.324" x2="1752.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.92" y1="896.324" x2="1784.92" y2="912.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.92" y1="896.324" x2="1784.92" y2="880.324" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.86" y1="1264.5" x2="984.861" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.86" y1="1264.5" x2="984.861" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1280.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.86" y1="1264.5" x2="1016.86" y2="1248.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1022.7" y1="1343.63" x2="1006.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1359.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1022.7" y1="1343.63" x2="1038.7" y2="1327.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.15" y1="1293.76" x2="1051.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1309.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1067.15" y1="1293.76" x2="1083.15" y2="1277.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1109.19" y1="1130.27" x2="1093.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1146.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1109.19" y1="1130.27" x2="1125.19" y2="1114.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.88" y1="1116.12" x2="1399.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1132.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.88" y1="1116.12" x2="1431.88" y2="1100.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.02" y1="1086.3" x2="1442.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1102.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.02" y1="1086.3" x2="1474.02" y2="1070.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1138.81" y1="1184.29" x2="1122.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1200.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1138.81" y1="1184.29" x2="1154.81" y2="1168.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.12" y1="1087.76" x2="1183.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1103.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.12" y1="1087.76" x2="1215.12" y2="1071.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.48" y1="973.842" x2="1509.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.48" y1="973.842" x2="1509.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.48" y1="973.842" x2="1541.48" y2="989.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.48" y1="973.842" x2="1541.48" y2="957.842" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.75" y1="1324.55" x2="1091.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1340.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.75" y1="1324.55" x2="1123.75" y2="1308.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="814.5" y1="1271.95" x2="798.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="814.5" y1="1271.95" x2="798.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="814.5" y1="1271.95" x2="830.5" y2="1287.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="814.5" y1="1271.95" x2="830.5" y2="1255.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1369.5" y1="1060.06" x2="1353.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1076.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1369.5" y1="1060.06" x2="1385.5" y2="1044.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.95" y1="1174.4" x2="1227.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1190.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.95" y1="1174.4" x2="1259.95" y2="1158.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1181.71" y1="1301.58" x2="1165.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1317.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1181.71" y1="1301.58" x2="1197.71" y2="1285.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1834.69" y1="763.508" x2="1818.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1834.69" y1="763.508" x2="1818.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1834.69" y1="763.508" x2="1850.69" y2="779.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1834.69" y1="763.508" x2="1850.69" y2="747.508" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.65" y1="1160.58" x2="1462.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1176.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.65" y1="1160.58" x2="1494.65" y2="1144.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.04" y1="1278.88" x2="990.041" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.04" y1="1278.88" x2="990.041" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1294.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1006.04" y1="1278.88" x2="1022.04" y2="1262.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1268.84" y1="1258.35" x2="1252.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1274.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1268.84" y1="1258.35" x2="1284.84" y2="1242.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.05" y1="1167.1" x2="1279.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1183.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.05" y1="1167.1" x2="1311.05" y2="1151.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239.88" y1="1281.07" x2="1223.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1297.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239.88" y1="1281.07" x2="1255.88" y2="1265.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.19" y1="1165.26" x2="1296.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1181.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.19" y1="1165.26" x2="1328.19" y2="1149.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1424" y1="1023.62" x2="1408" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1424" y1="1023.62" x2="1408" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1424" y1="1023.62" x2="1440" y2="1039.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1424" y1="1023.62" x2="1440" y2="1007.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1028.72" y1="1432.67" x2="1012.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1448.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1028.72" y1="1432.67" x2="1044.72" y2="1416.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.75" y1="1174.67" x2="1292.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1190.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.75" y1="1174.67" x2="1324.75" y2="1158.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.12" y1="1148.69" x2="1221.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1164.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.12" y1="1148.69" x2="1253.12" y2="1132.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1618.21" y1="979.458" x2="1602.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1618.21" y1="979.458" x2="1602.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1618.21" y1="979.458" x2="1634.21" y2="995.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1618.21" y1="979.458" x2="1634.21" y2="963.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1279.69" y1="1082.34" x2="1263.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1279.69" y1="1082.34" x2="1295.69" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1393.4" y1="1152.94" x2="1377.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1168.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1393.4" y1="1152.94" x2="1409.4" y2="1136.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1005.39" y1="1284.51" x2="989.392" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1005.39" y1="1284.51" x2="989.392" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1300.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1005.39" y1="1284.51" x2="1021.39" y2="1268.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.58" y1="1254.71" x2="1222.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1270.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1238.58" y1="1254.71" x2="1254.58" y2="1238.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.98" y1="894.258" x2="1593.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.98" y1="894.258" x2="1593.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.98" y1="894.258" x2="1625.98" y2="910.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1609.98" y1="894.258" x2="1625.98" y2="878.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1497.3" y1="1096.59" x2="1481.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1112.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1497.3" y1="1096.59" x2="1513.3" y2="1080.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1387.69" y1="1064.12" x2="1371.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1080.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1387.69" y1="1064.12" x2="1403.69" y2="1048.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1471.44" y1="1131.3" x2="1455.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1147.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1471.44" y1="1131.3" x2="1487.44" y2="1115.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1597.07" y1="924.794" x2="1581.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1597.07" y1="924.794" x2="1581.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1597.07" y1="924.794" x2="1613.07" y2="940.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1597.07" y1="924.794" x2="1613.07" y2="908.794" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="882.48" y1="1367.74" x2="866.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="882.48" y1="1367.74" x2="866.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="882.48" y1="1367.74" x2="898.48" y2="1383.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="882.48" y1="1367.74" x2="898.48" y2="1351.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1285.26" y1="1113.9" x2="1269.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1129.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1285.26" y1="1113.9" x2="1301.26" y2="1097.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.2" y1="1043.06" x2="1413.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1059.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1429.2" y1="1043.06" x2="1445.2" y2="1027.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1413.65" y1="1215.29" x2="1397.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1413.65" y1="1215.29" x2="1429.65" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1527.95" y1="988.132" x2="1511.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1527.95" y1="988.132" x2="1511.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1527.95" y1="988.132" x2="1543.95" y2="1004.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1527.95" y1="988.132" x2="1543.95" y2="972.132" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.23" y1="1117.88" x2="1412.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1133.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.23" y1="1117.88" x2="1444.23" y2="1101.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.14" y1="1034.93" x2="1497.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1050.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.14" y1="1034.93" x2="1529.14" y2="1018.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1464.41" y1="1058.74" x2="1448.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1074.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1464.41" y1="1058.74" x2="1480.41" y2="1042.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1766.37" y1="756.534" x2="1750.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1766.37" y1="756.534" x2="1750.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1766.37" y1="756.534" x2="1782.37" y2="772.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1766.37" y1="756.534" x2="1782.37" y2="740.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.09" y1="1182.64" x2="1321.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1198.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.09" y1="1182.64" x2="1353.09" y2="1166.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.04" y1="1303.89" x2="1050.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1319.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.04" y1="1303.89" x2="1082.04" y2="1287.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.46" y1="1067.55" x2="1561.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1083.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1577.46" y1="1067.55" x2="1593.46" y2="1051.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.74" y1="1243.98" x2="1121.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1259.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.74" y1="1243.98" x2="1153.74" y2="1227.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.72" y1="1098.59" x2="1349.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1114.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.72" y1="1098.59" x2="1381.72" y2="1082.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1422.2" y1="999.716" x2="1406.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1422.2" y1="999.716" x2="1406.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1422.2" y1="999.716" x2="1438.2" y2="1015.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1422.2" y1="999.716" x2="1438.2" y2="983.716" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1381.71" y1="975.726" x2="1365.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1381.71" y1="975.726" x2="1365.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1381.71" y1="975.726" x2="1397.71" y2="991.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1381.71" y1="975.726" x2="1397.71" y2="959.726" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.037" y1="1422.12" x2="859.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.037" y1="1422.12" x2="859.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.037" y1="1422.12" x2="891.037" y2="1438.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.037" y1="1422.12" x2="891.037" y2="1406.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.48" y1="1128.32" x2="1100.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1144.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.48" y1="1128.32" x2="1132.48" y2="1112.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="898.259" y1="1319.39" x2="882.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="898.259" y1="1319.39" x2="882.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="898.259" y1="1319.39" x2="914.259" y2="1335.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="898.259" y1="1319.39" x2="914.259" y2="1303.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.31" y1="1035.06" x2="1415.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1051.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.31" y1="1035.06" x2="1447.31" y2="1019.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.19" y1="815.289" x2="1746.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.19" y1="815.289" x2="1746.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.19" y1="815.289" x2="1778.19" y2="831.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.19" y1="815.289" x2="1778.19" y2="799.289" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="843.736" y1="1421.71" x2="827.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="843.736" y1="1421.71" x2="827.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="843.736" y1="1421.71" x2="859.736" y2="1437.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="843.736" y1="1421.71" x2="859.736" y2="1405.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1329.12" y1="1076.38" x2="1313.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1092.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1329.12" y1="1076.38" x2="1345.12" y2="1060.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.51" y1="1094.73" x2="1325.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1110.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1341.51" y1="1094.73" x2="1357.51" y2="1078.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.39" y1="1147.76" x2="1209.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1163.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.39" y1="1147.76" x2="1241.39" y2="1131.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.09" y1="1248.47" x2="1075.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1264.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.09" y1="1248.47" x2="1107.09" y2="1232.47" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1399.57" y1="992.596" x2="1383.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1399.57" y1="992.596" x2="1383.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1399.57" y1="992.596" x2="1415.57" y2="1008.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1399.57" y1="992.596" x2="1415.57" y2="976.596" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="712.506" y1="1466.98" x2="696.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="712.506" y1="1466.98" x2="696.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="712.506" y1="1466.98" x2="728.506" y2="1482.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="712.506" y1="1466.98" x2="728.506" y2="1450.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1038.8" y1="1215.97" x2="1022.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1231.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1038.8" y1="1215.97" x2="1054.8" y2="1199.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="996.845" y1="1298.95" x2="980.845" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="996.845" y1="1298.95" x2="980.845" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="996.845" y1="1298.95" x2="1012.85" y2="1314.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="996.845" y1="1298.95" x2="1012.85" y2="1282.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="764.769" y1="1428.81" x2="748.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="764.769" y1="1428.81" x2="748.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="764.769" y1="1428.81" x2="780.769" y2="1444.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="764.769" y1="1428.81" x2="780.769" y2="1412.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1481.71" y1="979.429" x2="1465.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1481.71" y1="979.429" x2="1465.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1481.71" y1="979.429" x2="1497.71" y2="995.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1481.71" y1="979.429" x2="1497.71" y2="963.429" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="794.022" y1="1392.87" x2="778.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="794.022" y1="1392.87" x2="778.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="794.022" y1="1392.87" x2="810.022" y2="1408.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="794.022" y1="1392.87" x2="810.022" y2="1376.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.54" y1="1232.13" x2="1099.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1248.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.54" y1="1232.13" x2="1131.54" y2="1216.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="989.1" y1="1361.68" x2="973.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="989.1" y1="1361.68" x2="973.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="989.1" y1="1361.68" x2="1005.1" y2="1377.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="989.1" y1="1361.68" x2="1005.1" y2="1345.68" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="658.223" y1="1594.72" x2="642.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="658.223" y1="1594.72" x2="642.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="658.223" y1="1594.72" x2="674.223" y2="1610.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="658.223" y1="1594.72" x2="674.223" y2="1578.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.13" y1="922.6" x2="1595.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.13" y1="922.6" x2="1595.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.13" y1="922.6" x2="1627.13" y2="938.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1611.13" y1="922.6" x2="1627.13" y2="906.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1325.74" y1="1073.3" x2="1309.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1089.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1325.74" y1="1073.3" x2="1341.74" y2="1057.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.65" y1="1061.83" x2="1438.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1077.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.65" y1="1061.83" x2="1470.65" y2="1045.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1270.59" y1="1028.09" x2="1254.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1044.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1270.59" y1="1028.09" x2="1286.59" y2="1012.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.91" y1="1068.7" x2="1318.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1084.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.91" y1="1068.7" x2="1350.91" y2="1052.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.75" y1="1086.14" x2="1324.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1102.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1340.75" y1="1086.14" x2="1356.75" y2="1070.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="968.976" y1="1423.08" x2="952.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="968.976" y1="1423.08" x2="952.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="968.976" y1="1423.08" x2="984.976" y2="1439.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="968.976" y1="1423.08" x2="984.976" y2="1407.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1451.83" y1="926.379" x2="1435.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1451.83" y1="926.379" x2="1435.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1451.83" y1="926.379" x2="1467.83" y2="942.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1451.83" y1="926.379" x2="1467.83" y2="910.379" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1420.21" y1="1170.17" x2="1404.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1186.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1420.21" y1="1170.17" x2="1436.21" y2="1154.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.55" y1="1324.46" x2="1014.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1340.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1030.55" y1="1324.46" x2="1046.55" y2="1308.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.85" y1="1057.07" x2="1398.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1073.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.85" y1="1057.07" x2="1430.85" y2="1041.07" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1037.88" y1="1309.43" x2="1021.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1325.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1037.88" y1="1309.43" x2="1053.88" y2="1293.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1171.81" y1="1298.75" x2="1155.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1314.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1171.81" y1="1298.75" x2="1187.81" y2="1282.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.86" y1="1223.11" x2="1097.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1239.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.86" y1="1223.11" x2="1129.86" y2="1207.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.45" y1="1086.98" x2="1444.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1102.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.45" y1="1086.98" x2="1476.45" y2="1070.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1557.36" y1="1088.65" x2="1541.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1104.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1557.36" y1="1088.65" x2="1573.36" y2="1072.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="908.822" y1="1392.95" x2="892.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="908.822" y1="1392.95" x2="892.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="908.822" y1="1392.95" x2="924.822" y2="1408.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="908.822" y1="1392.95" x2="924.822" y2="1376.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.57" y1="1011.37" x2="1484.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.57" y1="1011.37" x2="1484.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.57" y1="1011.37" x2="1516.57" y2="1027.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1500.57" y1="1011.37" x2="1516.57" y2="995.374" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1002.47" y1="1234.46" x2="986.471" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1002.47" y1="1234.46" x2="986.471" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1250.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1002.47" y1="1234.46" x2="1018.47" y2="1218.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.95" y1="1013.72" x2="1442.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.95" y1="1013.72" x2="1442.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.95" y1="1013.72" x2="1474.95" y2="1029.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1458.95" y1="1013.72" x2="1474.95" y2="997.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.63" y1="1328.09" x2="1043.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1344.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1059.63" y1="1328.09" x2="1075.63" y2="1312.09" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1046.37" x2="1288.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1062.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1304.99" y1="1046.37" x2="1320.99" y2="1030.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1016.99" y1="1290.74" x2="1000.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1306.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1016.99" y1="1290.74" x2="1032.99" y2="1274.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="892.734" y1="1460.48" x2="876.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="892.734" y1="1460.48" x2="876.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="892.734" y1="1460.48" x2="908.734" y2="1476.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="892.734" y1="1460.48" x2="908.734" y2="1444.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.22" y1="1187.04" x2="1030.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1203.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.22" y1="1187.04" x2="1062.22" y2="1171.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.88" y1="1292.71" x2="1060.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1308.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.88" y1="1292.71" x2="1092.88" y2="1276.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.79" y1="924.706" x2="1529.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.79" y1="924.706" x2="1529.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.79" y1="924.706" x2="1561.79" y2="940.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.79" y1="924.706" x2="1561.79" y2="908.706" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1003.69" y1="1262.65" x2="987.692" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1003.69" y1="1262.65" x2="987.692" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1278.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1003.69" y1="1262.65" x2="1019.69" y2="1246.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1245.12" y1="1104.1" x2="1229.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1120.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1245.12" y1="1104.1" x2="1261.12" y2="1088.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.07" y1="1264.22" x2="1097.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1280.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1113.07" y1="1264.22" x2="1129.07" y2="1248.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="862.589" y1="1433.49" x2="846.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="862.589" y1="1433.49" x2="846.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="862.589" y1="1433.49" x2="878.589" y2="1449.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="862.589" y1="1433.49" x2="878.589" y2="1417.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1335.3" y1="1153.33" x2="1319.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1169.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1335.3" y1="1153.33" x2="1351.3" y2="1137.33" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1521.7" y1="1018.9" x2="1505.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1034.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1521.7" y1="1018.9" x2="1537.7" y2="1002.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1384.65" y1="1259.56" x2="1368.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1275.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1384.65" y1="1259.56" x2="1400.65" y2="1243.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1366.24" y1="1165.25" x2="1350.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1181.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1366.24" y1="1165.25" x2="1382.24" y2="1149.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="522.504" y1="1512.98" x2="506.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="522.504" y1="1512.98" x2="506.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="522.504" y1="1512.98" x2="538.504" y2="1528.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="522.504" y1="1512.98" x2="538.504" y2="1496.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.208" y1="1309.37" x2="956.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.208" y1="1309.37" x2="956.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.208" y1="1309.37" x2="988.208" y2="1325.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.208" y1="1309.37" x2="988.208" y2="1293.37" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.36" y1="1054" x2="1160.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.36" y1="1054" x2="1160.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.36" y1="1054" x2="1192.36" y2="1070" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.36" y1="1054" x2="1192.36" y2="1038" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.15" y1="1063.66" x2="1299.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1079.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1315.15" y1="1063.66" x2="1331.15" y2="1047.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1235.21" y1="1142.27" x2="1219.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1158.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1235.21" y1="1142.27" x2="1251.21" y2="1126.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.46" y1="1190.1" x2="1213.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1206.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1229.46" y1="1190.1" x2="1245.46" y2="1174.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.451" y1="1406.48" x2="728.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.451" y1="1406.48" x2="728.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.451" y1="1406.48" x2="760.451" y2="1422.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.451" y1="1406.48" x2="760.451" y2="1390.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.07" y1="1218.16" x2="1081.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1234.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.07" y1="1218.16" x2="1113.07" y2="1202.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1012.88" y1="1293.81" x2="996.879" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1012.88" y1="1293.81" x2="996.879" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1309.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1012.88" y1="1293.81" x2="1028.88" y2="1277.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="277.68" y1="1755.15" x2="261.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="277.68" y1="1755.15" x2="261.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="277.68" y1="1755.15" x2="293.68" y2="1771.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="277.68" y1="1755.15" x2="293.68" y2="1739.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1166.29" y1="1406.05" x2="1150.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1422.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1166.29" y1="1406.05" x2="1182.29" y2="1390.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1178.56" y1="1242.2" x2="1162.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1258.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1178.56" y1="1242.2" x2="1194.56" y2="1226.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1498.04" y1="1011.64" x2="1482.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1498.04" y1="1011.64" x2="1482.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1498.04" y1="1011.64" x2="1514.04" y2="1027.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1498.04" y1="1011.64" x2="1514.04" y2="995.641" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1380.98" y1="1081.87" x2="1364.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1097.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1380.98" y1="1081.87" x2="1396.98" y2="1065.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1189.27" y1="1225.02" x2="1173.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1241.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1189.27" y1="1225.02" x2="1205.27" y2="1209.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.33" y1="1091.79" x2="1337.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1107.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.33" y1="1091.79" x2="1369.33" y2="1075.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.57" y1="1214.17" x2="1175.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1230.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1191.57" y1="1214.17" x2="1207.57" y2="1198.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.19" y1="842.118" x2="1764.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.19" y1="842.118" x2="1764.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.19" y1="842.118" x2="1796.19" y2="858.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.19" y1="842.118" x2="1796.19" y2="826.118" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="855.143" y1="1461.56" x2="839.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="855.143" y1="1461.56" x2="839.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="855.143" y1="1461.56" x2="871.143" y2="1477.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="855.143" y1="1461.56" x2="871.143" y2="1445.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1487.2" y1="1215.29" x2="1471.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1231.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1487.2" y1="1215.29" x2="1503.2" y2="1199.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="941.026" y1="1398.61" x2="925.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="941.026" y1="1398.61" x2="925.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="941.026" y1="1398.61" x2="957.026" y2="1414.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="941.026" y1="1398.61" x2="957.026" y2="1382.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1491.9" y1="862.19" x2="1475.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1491.9" y1="862.19" x2="1475.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1491.9" y1="862.19" x2="1507.9" y2="878.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1491.9" y1="862.19" x2="1507.9" y2="846.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1045.21" y1="1271.31" x2="1029.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1287.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1045.21" y1="1271.31" x2="1061.21" y2="1255.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.16" y1="1192.34" x2="1262.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1208.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.16" y1="1192.34" x2="1294.16" y2="1176.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.72" y1="1184.27" x2="1262.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1200.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1278.72" y1="1184.27" x2="1294.72" y2="1168.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1614.72" y1="834.262" x2="1598.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1614.72" y1="834.262" x2="1598.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1614.72" y1="834.262" x2="1630.72" y2="850.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1614.72" y1="834.262" x2="1630.72" y2="818.262" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.83" y1="933.105" x2="1599.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.83" y1="933.105" x2="1599.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.83" y1="933.105" x2="1631.83" y2="949.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.83" y1="933.105" x2="1631.83" y2="917.105" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1306.51" y1="1052.44" x2="1290.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1068.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1306.51" y1="1052.44" x2="1322.51" y2="1036.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.55" y1="1095.87" x2="1376.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1111.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.55" y1="1095.87" x2="1408.55" y2="1079.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.951" y1="1271.4" x2="883.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.951" y1="1271.4" x2="883.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.951" y1="1271.4" x2="915.951" y2="1287.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.951" y1="1271.4" x2="915.951" y2="1255.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.68" y1="1069.21" x2="1376.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1085.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1392.68" y1="1069.21" x2="1408.68" y2="1053.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1651.8" y1="933.534" x2="1635.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1651.8" y1="933.534" x2="1635.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1651.8" y1="933.534" x2="1667.8" y2="949.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1651.8" y1="933.534" x2="1667.8" y2="917.534" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.2" y1="982.773" x2="1530.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.2" y1="982.773" x2="1530.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.2" y1="982.773" x2="1562.2" y2="998.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1546.2" y1="982.773" x2="1562.2" y2="966.773" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1197.19" y1="1159.24" x2="1181.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1175.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1197.19" y1="1159.24" x2="1213.19" y2="1143.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1378.06" y1="1142.61" x2="1362.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1158.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1378.06" y1="1142.61" x2="1394.06" y2="1126.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="798.462" y1="1475.78" x2="782.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="798.462" y1="1475.78" x2="782.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="798.462" y1="1475.78" x2="814.462" y2="1491.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="798.462" y1="1475.78" x2="814.462" y2="1459.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1653.74" y1="920.543" x2="1637.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1653.74" y1="920.543" x2="1637.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1653.74" y1="920.543" x2="1669.74" y2="936.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1653.74" y1="920.543" x2="1669.74" y2="904.543" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.58" y1="1157.39" x2="1062.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1173.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1078.58" y1="1157.39" x2="1094.58" y2="1141.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.44" y1="1151.58" x2="1405.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1167.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.44" y1="1151.58" x2="1437.44" y2="1135.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.19" y1="1152.96" x2="1277.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1168.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.19" y1="1152.96" x2="1309.19" y2="1136.96" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1307.01" y1="1135.59" x2="1291.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1151.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1307.01" y1="1135.59" x2="1323.01" y2="1119.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1157.68" y1="1240.17" x2="1141.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1256.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1157.68" y1="1240.17" x2="1173.68" y2="1224.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1322.34" y1="1098.52" x2="1306.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1114.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1322.34" y1="1098.52" x2="1338.34" y2="1082.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.24" y1="1200.4" x2="1231.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1216.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1247.24" y1="1200.4" x2="1263.24" y2="1184.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.59" y1="1055.14" x2="1348.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1071.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.59" y1="1055.14" x2="1380.59" y2="1039.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1640.29" y1="1056.93" x2="1624.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1072.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1640.29" y1="1056.93" x2="1656.29" y2="1040.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1286.58" y1="1059.6" x2="1270.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1075.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1286.58" y1="1059.6" x2="1302.58" y2="1043.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1550.14" y1="1004.56" x2="1534.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1550.14" y1="1004.56" x2="1534.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1550.14" y1="1004.56" x2="1566.14" y2="1020.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1550.14" y1="1004.56" x2="1566.14" y2="988.562" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="955.148" y1="1370.3" x2="939.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="955.148" y1="1370.3" x2="939.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="955.148" y1="1370.3" x2="971.148" y2="1386.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="955.148" y1="1370.3" x2="971.148" y2="1354.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1845.45" y1="659.705" x2="1829.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1845.45" y1="659.705" x2="1829.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1845.45" y1="659.705" x2="1861.45" y2="675.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1845.45" y1="659.705" x2="1861.45" y2="643.705" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1817.11" y1="731.36" x2="1801.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1817.11" y1="731.36" x2="1801.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1817.11" y1="731.36" x2="1833.11" y2="747.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1817.11" y1="731.36" x2="1833.11" y2="715.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1838.37" y1="806.858" x2="1822.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1838.37" y1="806.858" x2="1822.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1838.37" y1="806.858" x2="1854.37" y2="822.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1838.37" y1="806.858" x2="1854.37" y2="790.858" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1132.81" y1="1166.63" x2="1116.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1132.81" y1="1166.63" x2="1148.81" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244" y1="1263.65" x2="1228" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244" y1="1263.65" x2="1228" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244" y1="1263.65" x2="1260" y2="1279.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244" y1="1263.65" x2="1260" y2="1247.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.43" y1="1066.19" x2="1274.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.43" y1="1066.19" x2="1306.43" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1558.7" y1="965.028" x2="1542.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1558.7" y1="965.028" x2="1542.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1558.7" y1="965.028" x2="1574.7" y2="981.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1558.7" y1="965.028" x2="1574.7" y2="949.028" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.76" y1="970.455" x2="1458.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.76" y1="970.455" x2="1458.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.76" y1="970.455" x2="1490.76" y2="986.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.76" y1="970.455" x2="1490.76" y2="954.455" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1682.62" y1="820.381" x2="1666.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1682.62" y1="820.381" x2="1666.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1682.62" y1="820.381" x2="1698.62" y2="836.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1682.62" y1="820.381" x2="1698.62" y2="804.381" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="872.16" y1="1370.98" x2="856.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="872.16" y1="1370.98" x2="856.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="872.16" y1="1370.98" x2="888.16" y2="1386.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="872.16" y1="1370.98" x2="888.16" y2="1354.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1155.08" y1="1163.95" x2="1139.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1179.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1155.08" y1="1163.95" x2="1171.08" y2="1147.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1316.44" y1="1111.88" x2="1300.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1127.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1316.44" y1="1111.88" x2="1332.44" y2="1095.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="866.851" y1="1331.78" x2="850.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="866.851" y1="1331.78" x2="850.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="866.851" y1="1331.78" x2="882.851" y2="1347.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="866.851" y1="1331.78" x2="882.851" y2="1315.78" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.47" y1="1146.54" x2="1438.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1162.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1454.47" y1="1146.54" x2="1470.47" y2="1130.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1345.73" y1="1121.54" x2="1329.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1137.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1345.73" y1="1121.54" x2="1361.73" y2="1105.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1604.79" y1="991.27" x2="1588.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1604.79" y1="991.27" x2="1588.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1604.79" y1="991.27" x2="1620.79" y2="1007.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1604.79" y1="991.27" x2="1620.79" y2="975.27" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1425.4" y1="1018.53" x2="1409.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1034.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1425.4" y1="1018.53" x2="1441.4" y2="1002.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1177.09" y1="1221.26" x2="1161.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1237.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1177.09" y1="1221.26" x2="1193.09" y2="1205.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="436.181" y1="1793.59" x2="420.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="436.181" y1="1793.59" x2="420.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="436.181" y1="1793.59" x2="452.181" y2="1809.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="436.181" y1="1793.59" x2="452.181" y2="1777.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="957.079" y1="1394.58" x2="941.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="957.079" y1="1394.58" x2="941.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="957.079" y1="1394.58" x2="973.079" y2="1410.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="957.079" y1="1394.58" x2="973.079" y2="1378.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1427.73" y1="1093.18" x2="1411.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1109.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1427.73" y1="1093.18" x2="1443.73" y2="1077.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1348.36" y1="1249.7" x2="1332.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1265.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1348.36" y1="1249.7" x2="1364.36" y2="1233.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.366" y1="1495.39" x2="775.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.366" y1="1495.39" x2="775.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.366" y1="1495.39" x2="807.366" y2="1511.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.366" y1="1495.39" x2="807.366" y2="1479.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1636.9" y1="890.133" x2="1620.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1636.9" y1="890.133" x2="1620.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1636.9" y1="890.133" x2="1652.9" y2="906.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1636.9" y1="890.133" x2="1652.9" y2="874.133" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1256" y1="1205.58" x2="1240" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1256" y1="1205.58" x2="1240" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1256" y1="1205.58" x2="1272" y2="1221.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1256" y1="1205.58" x2="1272" y2="1189.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.44" y1="1065.3" x2="1412.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1081.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.44" y1="1065.3" x2="1444.44" y2="1049.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.03" y1="1294.95" x2="1030.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1310.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1046.03" y1="1294.95" x2="1062.03" y2="1278.95" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="721.091" y1="1561.75" x2="705.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="721.091" y1="1561.75" x2="705.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="721.091" y1="1561.75" x2="737.091" y2="1577.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="721.091" y1="1561.75" x2="737.091" y2="1545.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1085.36" y1="1148.29" x2="1069.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1164.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1085.36" y1="1148.29" x2="1101.36" y2="1132.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="823.473" y1="1446.58" x2="807.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="823.473" y1="1446.58" x2="807.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="823.473" y1="1446.58" x2="839.473" y2="1462.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="823.473" y1="1446.58" x2="839.473" y2="1430.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="887.822" y1="1375.21" x2="871.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="887.822" y1="1375.21" x2="871.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="887.822" y1="1375.21" x2="903.822" y2="1391.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="887.822" y1="1375.21" x2="903.822" y2="1359.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.18" y1="1169.39" x2="1147.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1185.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.18" y1="1169.39" x2="1179.18" y2="1153.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.39" y1="1014.06" x2="1468.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.39" y1="1014.06" x2="1468.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.39" y1="1014.06" x2="1500.39" y2="1030.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1484.39" y1="1014.06" x2="1500.39" y2="998.061" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1685.1" y1="924.047" x2="1669.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1685.1" y1="924.047" x2="1669.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1685.1" y1="924.047" x2="1701.1" y2="940.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1685.1" y1="924.047" x2="1701.1" y2="908.047" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1070.78" y1="1266.71" x2="1054.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1282.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1070.78" y1="1266.71" x2="1086.78" y2="1250.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1379.51" y1="1049.2" x2="1363.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1065.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1379.51" y1="1049.2" x2="1395.51" y2="1033.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="877.025" y1="1403.06" x2="861.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="877.025" y1="1403.06" x2="861.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="877.025" y1="1403.06" x2="893.025" y2="1419.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="877.025" y1="1403.06" x2="893.025" y2="1387.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.37" y1="1191.22" x2="1060.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1207.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1076.37" y1="1191.22" x2="1092.37" y2="1175.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1164.55" y1="1249.8" x2="1148.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1265.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1164.55" y1="1249.8" x2="1180.55" y2="1233.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1473.39" y1="1000.62" x2="1457.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1473.39" y1="1000.62" x2="1457.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1473.39" y1="1000.62" x2="1489.39" y2="1016.62" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1473.39" y1="1000.62" x2="1489.39" y2="984.624" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.54" y1="1202.19" x2="1279.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1218.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1295.54" y1="1202.19" x2="1311.54" y2="1186.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.58" y1="1267.5" x2="1100.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1283.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1116.58" y1="1267.5" x2="1132.58" y2="1251.5" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1435.37" y1="1059.7" x2="1419.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1075.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1435.37" y1="1059.7" x2="1451.37" y2="1043.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="930.593" y1="1251.3" x2="914.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="930.593" y1="1251.3" x2="914.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="930.593" y1="1251.3" x2="946.593" y2="1267.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="930.593" y1="1251.3" x2="946.593" y2="1235.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1125.99" y1="1256.74" x2="1109.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1272.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1125.99" y1="1256.74" x2="1141.99" y2="1240.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.3" y1="1267.31" x2="1105.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1283.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.3" y1="1267.31" x2="1137.3" y2="1251.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.55" y1="972.554" x2="1599.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.55" y1="972.554" x2="1599.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.55" y1="972.554" x2="1631.55" y2="988.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1615.55" y1="972.554" x2="1631.55" y2="956.554" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1395.64" y1="960.36" x2="1379.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1395.64" y1="960.36" x2="1379.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1395.64" y1="960.36" x2="1411.64" y2="976.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1395.64" y1="960.36" x2="1411.64" y2="944.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.08" y1="1312.36" x2="984.075" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.08" y1="1312.36" x2="984.075" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1328.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1000.08" y1="1312.36" x2="1016.08" y2="1296.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.75" y1="1068.03" x2="1337.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1084.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1353.75" y1="1068.03" x2="1369.75" y2="1052.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1126.95" y1="1234.81" x2="1110.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1250.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1126.95" y1="1234.81" x2="1142.95" y2="1218.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.72" y1="1159.69" x2="1347.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1175.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.72" y1="1159.69" x2="1379.72" y2="1143.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="805.605" y1="1368.59" x2="789.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="805.605" y1="1368.59" x2="789.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="805.605" y1="1368.59" x2="821.605" y2="1384.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="805.605" y1="1368.59" x2="821.605" y2="1352.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="735.913" y1="1350.58" x2="719.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="735.913" y1="1350.58" x2="719.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="735.913" y1="1350.58" x2="751.913" y2="1366.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="735.913" y1="1350.58" x2="751.913" y2="1334.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1057.78" y1="1211.94" x2="1041.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1227.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1057.78" y1="1211.94" x2="1073.78" y2="1195.94" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1826.24" y1="921.284" x2="1810.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1826.24" y1="921.284" x2="1810.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1826.24" y1="921.284" x2="1842.24" y2="937.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1826.24" y1="921.284" x2="1842.24" y2="905.284" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.2" y1="1094.53" x2="1277.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1110.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1293.2" y1="1094.53" x2="1309.2" y2="1078.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.65" y1="1361.41" x2="1091.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1377.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.65" y1="1361.41" x2="1123.65" y2="1345.41" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.07" y1="1240.9" x2="1160.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1256.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1176.07" y1="1240.9" x2="1192.07" y2="1224.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1215.2" y1="1190.81" x2="1199.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1206.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1215.2" y1="1190.81" x2="1231.2" y2="1174.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1572.67" y1="995.733" x2="1556.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1572.67" y1="995.733" x2="1556.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1572.67" y1="995.733" x2="1588.67" y2="1011.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1572.67" y1="995.733" x2="1588.67" y2="979.733" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.41" y1="1231.66" x2="1348.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1247.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1364.41" y1="1231.66" x2="1380.41" y2="1215.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1259.03" y1="1156.39" x2="1243.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1172.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1259.03" y1="1156.39" x2="1275.03" y2="1140.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1469" y1="974.22" x2="1453" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1469" y1="974.22" x2="1453" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1469" y1="974.22" x2="1485" y2="990.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1469" y1="974.22" x2="1485" y2="958.22" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1292.53" y1="1131.39" x2="1276.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1147.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1292.53" y1="1131.39" x2="1308.53" y2="1115.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.04" y1="1299.49" x2="1130.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1315.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1146.04" y1="1299.49" x2="1162.04" y2="1283.49" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.06" y1="1219.15" x2="1070.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1235.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1086.06" y1="1219.15" x2="1102.06" y2="1203.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.64" y1="1000.17" x2="1752.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.64" y1="1000.17" x2="1752.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.64" y1="1000.17" x2="1784.64" y2="1016.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1768.64" y1="1000.17" x2="1784.64" y2="984.169" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.35" y1="1164.06" x2="1349.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1180.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.35" y1="1164.06" x2="1381.35" y2="1148.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.6" y1="944.573" x2="1470.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.6" y1="944.573" x2="1470.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.6" y1="944.573" x2="1502.6" y2="960.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1486.6" y1="944.573" x2="1502.6" y2="928.573" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.16" y1="1210.82" x2="1257.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1226.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.16" y1="1210.82" x2="1289.16" y2="1194.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.92" y1="1149.92" x2="1246.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1165.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.92" y1="1149.92" x2="1278.92" y2="1133.92" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1122.29" y1="1146.61" x2="1106.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1162.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1122.29" y1="1146.61" x2="1138.29" y2="1130.61" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.41" y1="1171.23" x2="1275.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1187.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.41" y1="1171.23" x2="1307.41" y2="1155.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1448.29" y1="1128.98" x2="1432.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1144.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1448.29" y1="1128.98" x2="1464.29" y2="1112.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1130.44" y1="1149.23" x2="1114.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1165.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1130.44" y1="1149.23" x2="1146.44" y2="1133.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1039.51" y1="1295.52" x2="1023.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1311.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1039.51" y1="1295.52" x2="1055.51" y2="1279.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.81" y1="1008.74" x2="1427.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.81" y1="1008.74" x2="1427.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.81" y1="1008.74" x2="1459.81" y2="1024.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1443.81" y1="1008.74" x2="1459.81" y2="992.741" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.99" y1="1097.56" x2="1529.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1113.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1545.99" y1="1097.56" x2="1561.99" y2="1081.56" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.83" y1="1094.72" x2="1343.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1110.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1359.83" y1="1094.72" x2="1375.83" y2="1078.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.3" y1="1374.86" x2="1183.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1390.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1199.3" y1="1374.86" x2="1215.3" y2="1358.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1133.36" y1="1257.88" x2="1117.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1273.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1133.36" y1="1257.88" x2="1149.36" y2="1241.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1628.36" y1="926.189" x2="1612.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1628.36" y1="926.189" x2="1612.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1628.36" y1="926.189" x2="1644.36" y2="942.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1628.36" y1="926.189" x2="1644.36" y2="910.189" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1073.35" y1="1320.7" x2="1057.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1336.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1073.35" y1="1320.7" x2="1089.35" y2="1304.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.88" y1="1332.35" x2="1075.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1348.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1091.88" y1="1332.35" x2="1107.88" y2="1316.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1017.09" y1="1323.45" x2="1001.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1339.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1017.09" y1="1323.45" x2="1033.09" y2="1307.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.938" y1="1534.93" x2="553.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.938" y1="1534.93" x2="553.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.938" y1="1534.93" x2="585.938" y2="1550.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="569.938" y1="1534.93" x2="585.938" y2="1518.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1231.45" y1="1166.31" x2="1215.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1182.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1231.45" y1="1166.31" x2="1247.45" y2="1150.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.218" y1="1483.05" x2="728.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.218" y1="1483.05" x2="728.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.218" y1="1483.05" x2="760.218" y2="1499.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="744.218" y1="1483.05" x2="760.218" y2="1467.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.81" y1="1264.02" x2="1169.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1280.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1185.81" y1="1264.02" x2="1201.81" y2="1248.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1713.33" y1="929.721" x2="1697.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1713.33" y1="929.721" x2="1697.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1713.33" y1="929.721" x2="1729.33" y2="945.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1713.33" y1="929.721" x2="1729.33" y2="913.721" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.62" y1="1058.71" x2="1292.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1074.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1308.62" y1="1058.71" x2="1324.62" y2="1042.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="610.516" y1="1568.23" x2="594.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="610.516" y1="1568.23" x2="594.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="610.516" y1="1568.23" x2="626.516" y2="1584.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="610.516" y1="1568.23" x2="626.516" y2="1552.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1702.63" y1="949.089" x2="1686.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1702.63" y1="949.089" x2="1686.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1702.63" y1="949.089" x2="1718.63" y2="965.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1702.63" y1="949.089" x2="1718.63" y2="933.089" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1102.82" y1="1200.06" x2="1086.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1216.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1102.82" y1="1200.06" x2="1118.82" y2="1184.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.22" y1="1125.36" x2="1444.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1141.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1460.22" y1="1125.36" x2="1476.22" y2="1109.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1242.85" y1="1068.4" x2="1226.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1084.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1242.85" y1="1068.4" x2="1258.85" y2="1052.4" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1180.26" y1="1392.12" x2="1164.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1408.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1180.26" y1="1392.12" x2="1196.26" y2="1376.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1799.12" y1="921.26" x2="1783.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1799.12" y1="921.26" x2="1783.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1799.12" y1="921.26" x2="1815.12" y2="937.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1799.12" y1="921.26" x2="1815.12" y2="905.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1198.35" y1="1213.97" x2="1182.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1229.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1198.35" y1="1213.97" x2="1214.35" y2="1197.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1314.16" y1="1231.53" x2="1298.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1247.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1314.16" y1="1231.53" x2="1330.16" y2="1215.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1298.91" y1="1128.6" x2="1282.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1144.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1298.91" y1="1128.6" x2="1314.91" y2="1112.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1539.91" y1="992.445" x2="1523.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1539.91" y1="992.445" x2="1523.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1539.91" y1="992.445" x2="1555.91" y2="1008.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1539.91" y1="992.445" x2="1555.91" y2="976.445" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.49" y1="1165.67" x2="1192.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1181.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.49" y1="1165.67" x2="1224.49" y2="1149.67" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.4" y1="1164.76" x2="1275.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1180.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1291.4" y1="1164.76" x2="1307.4" y2="1148.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1509.7" y1="1053.86" x2="1493.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1069.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1509.7" y1="1053.86" x2="1525.7" y2="1037.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1586.87" y1="1027.39" x2="1570.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1043.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1586.87" y1="1027.39" x2="1602.87" y2="1011.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.89" y1="1264.65" x2="1095.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1280.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.89" y1="1264.65" x2="1127.89" y2="1248.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.824" y1="1370.17" x2="947.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.824" y1="1370.17" x2="947.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.824" y1="1370.17" x2="979.824" y2="1386.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.824" y1="1370.17" x2="979.824" y2="1354.17" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1269.64" y1="1053.76" x2="1253.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1069.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1269.64" y1="1053.76" x2="1285.64" y2="1037.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="694.471" y1="1399.51" x2="678.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="694.471" y1="1399.51" x2="678.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="694.471" y1="1399.51" x2="710.471" y2="1415.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="694.471" y1="1399.51" x2="710.471" y2="1383.51" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1318.45" y1="1163.46" x2="1302.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1179.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1318.45" y1="1163.46" x2="1334.45" y2="1147.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1386.71" y1="1019.77" x2="1370.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1035.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1386.71" y1="1019.77" x2="1402.71" y2="1003.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1547.26" y1="1096.34" x2="1531.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1112.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1547.26" y1="1096.34" x2="1563.26" y2="1080.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1054.12" y1="1246.84" x2="1038.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1262.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1054.12" y1="1246.84" x2="1070.12" y2="1230.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.9" y1="1186.98" x2="1257.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1202.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.9" y1="1186.98" x2="1289.9" y2="1170.98" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.32" y1="993.037" x2="1462.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.32" y1="993.037" x2="1462.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.32" y1="993.037" x2="1494.32" y2="1009.04" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1478.32" y1="993.037" x2="1494.32" y2="977.037" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.95" y1="1279" x2="1264.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.95" y1="1279" x2="1264.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.95" y1="1279" x2="1296.95" y2="1295" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.95" y1="1279" x2="1296.95" y2="1263" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.65" y1="1130.42" x2="1307.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1146.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.65" y1="1130.42" x2="1339.65" y2="1114.42" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.8" y1="1126.84" x2="1265.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1142.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1281.8" y1="1126.84" x2="1297.8" y2="1110.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.44" y1="1134.89" x2="1318.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1150.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1334.44" y1="1134.89" x2="1350.44" y2="1118.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1266.08" y1="1197.16" x2="1250.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1213.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1266.08" y1="1197.16" x2="1282.08" y2="1181.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1459.46" y1="1078.54" x2="1443.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1094.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1459.46" y1="1078.54" x2="1475.46" y2="1062.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1650.12" y1="918.653" x2="1634.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1650.12" y1="918.653" x2="1634.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1650.12" y1="918.653" x2="1666.12" y2="934.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1650.12" y1="918.653" x2="1666.12" y2="902.653" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1061.4" y1="1273.43" x2="1045.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1289.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1061.4" y1="1273.43" x2="1077.4" y2="1257.43" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.92" y1="1166.84" x2="1214.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1182.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1230.92" y1="1166.84" x2="1246.92" y2="1150.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1860.89" y1="778.201" x2="1844.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1860.89" y1="778.201" x2="1844.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1860.89" y1="778.201" x2="1876.89" y2="794.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1860.89" y1="778.201" x2="1876.89" y2="762.201" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.65" y1="1264.26" x2="1224.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1280.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.65" y1="1264.26" x2="1256.65" y2="1248.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.89" y1="1136.28" x2="1264.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1152.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.89" y1="1136.28" x2="1296.89" y2="1120.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1444.37" y1="1001.52" x2="1428.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1444.37" y1="1001.52" x2="1428.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1444.37" y1="1001.52" x2="1460.37" y2="1017.52" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1444.37" y1="1001.52" x2="1460.37" y2="985.521" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1455.79" y1="1193.99" x2="1439.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1209.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1455.79" y1="1193.99" x2="1471.79" y2="1177.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.41" y1="1124.9" x2="1105.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1140.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1121.41" y1="1124.9" x2="1137.41" y2="1108.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.75" y1="1012.66" x2="1412.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.75" y1="1012.66" x2="1412.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.75" y1="1012.66" x2="1444.75" y2="1028.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.75" y1="1012.66" x2="1444.75" y2="996.663" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.98" y1="1162.81" x2="1121.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1178.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1137.98" y1="1162.81" x2="1153.98" y2="1146.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1896.85" y1="631.005" x2="1880.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1896.85" y1="631.005" x2="1880.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1896.85" y1="631.005" x2="1912.85" y2="647.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1896.85" y1="631.005" x2="1912.85" y2="615.005" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1863.03" y1="689.163" x2="1847.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1863.03" y1="689.163" x2="1847.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1863.03" y1="689.163" x2="1879.03" y2="705.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1863.03" y1="689.163" x2="1879.03" y2="673.163" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1492.82" y1="1052.48" x2="1476.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1068.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1492.82" y1="1052.48" x2="1508.82" y2="1036.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.58" y1="1162.34" x2="1228.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1178.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1244.58" y1="1162.34" x2="1260.58" y2="1146.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.53" y1="1236.06" x2="1307.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1252.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1323.53" y1="1236.06" x2="1339.53" y2="1220.06" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.45" y1="1106.65" x2="1369.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1122.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.45" y1="1106.65" x2="1401.45" y2="1090.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="973.556" y1="1249.11" x2="957.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="973.556" y1="1249.11" x2="957.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="973.556" y1="1249.11" x2="989.556" y2="1265.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="973.556" y1="1249.11" x2="989.556" y2="1233.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.42" y1="974.507" x2="1399.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.42" y1="974.507" x2="1399.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.42" y1="974.507" x2="1431.42" y2="990.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1415.42" y1="974.507" x2="1431.42" y2="958.507" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.69" y1="1084.02" x2="1436.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1100.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1452.69" y1="1084.02" x2="1468.69" y2="1068.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1114.28" y1="1270.08" x2="1098.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1286.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1114.28" y1="1270.08" x2="1130.28" y2="1254.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.1" y1="1367.01" x2="1264.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1383.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1280.1" y1="1367.01" x2="1296.1" y2="1351.01" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="961.172" y1="1297.9" x2="945.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="961.172" y1="1297.9" x2="945.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="961.172" y1="1297.9" x2="977.172" y2="1313.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="961.172" y1="1297.9" x2="977.172" y2="1281.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="828.067" y1="1395.72" x2="812.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="828.067" y1="1395.72" x2="812.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="828.067" y1="1395.72" x2="844.067" y2="1411.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="828.067" y1="1395.72" x2="844.067" y2="1379.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="812.031" y1="1484.71" x2="796.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="812.031" y1="1484.71" x2="796.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="812.031" y1="1484.71" x2="828.031" y2="1500.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="812.031" y1="1484.71" x2="828.031" y2="1468.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="901.731" y1="1295.1" x2="885.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="901.731" y1="1295.1" x2="885.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="901.731" y1="1295.1" x2="917.731" y2="1311.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="901.731" y1="1295.1" x2="917.731" y2="1279.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1331.96" y1="1041.81" x2="1315.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1057.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1331.96" y1="1041.81" x2="1347.96" y2="1025.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.94" y1="1077.84" x2="1347.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1093.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1363.94" y1="1077.84" x2="1379.94" y2="1061.84" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.853" y1="1331.79" x2="956.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.853" y1="1331.79" x2="956.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.853" y1="1331.79" x2="988.853" y2="1347.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="972.853" y1="1331.79" x2="988.853" y2="1315.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.97" y1="912.119" x2="1722.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.97" y1="912.119" x2="1722.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.97" y1="912.119" x2="1754.97" y2="928.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.97" y1="912.119" x2="1754.97" y2="896.119" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1468.77" y1="1104.3" x2="1452.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1120.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1468.77" y1="1104.3" x2="1484.77" y2="1088.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.63" y1="1168.13" x2="1036.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1184.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.63" y1="1168.13" x2="1068.63" y2="1152.13" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.71" y1="1040.19" x2="1386.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1056.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.71" y1="1040.19" x2="1418.71" y2="1024.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1370.93" y1="1101.26" x2="1354.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1117.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1370.93" y1="1101.26" x2="1386.93" y2="1085.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1534.29" y1="956.26" x2="1518.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1534.29" y1="956.26" x2="1518.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1534.29" y1="956.26" x2="1550.29" y2="972.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1534.29" y1="956.26" x2="1550.29" y2="940.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1470.32" y1="1144.8" x2="1454.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1160.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1470.32" y1="1144.8" x2="1486.32" y2="1128.8" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.177" y1="1437.3" x2="863.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.177" y1="1437.3" x2="863.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.177" y1="1437.3" x2="895.177" y2="1453.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.177" y1="1437.3" x2="895.177" y2="1421.3" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.308" y1="1302.19" x2="775.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.308" y1="1302.19" x2="775.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.308" y1="1302.19" x2="807.308" y2="1318.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="791.308" y1="1302.19" x2="807.308" y2="1286.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1182.59" y1="1128.11" x2="1166.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1144.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1182.59" y1="1128.11" x2="1198.59" y2="1112.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.44" y1="890.561" x2="1574.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.44" y1="890.561" x2="1574.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.44" y1="890.561" x2="1606.44" y2="906.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.44" y1="890.561" x2="1606.44" y2="874.561" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1083.01" y1="1250.31" x2="1067.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1266.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1083.01" y1="1250.31" x2="1099.01" y2="1234.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1729.02" y1="853.811" x2="1713.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1729.02" y1="853.811" x2="1713.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1729.02" y1="853.811" x2="1745.02" y2="869.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1729.02" y1="853.811" x2="1745.02" y2="837.811" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="936.651" y1="1397.2" x2="920.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="936.651" y1="1397.2" x2="920.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="936.651" y1="1397.2" x2="952.651" y2="1413.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="936.651" y1="1397.2" x2="952.651" y2="1381.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1141.39" y1="1273.55" x2="1125.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1289.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1141.39" y1="1273.55" x2="1157.39" y2="1257.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.12" y1="1220.55" x2="1050.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1236.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1066.12" y1="1220.55" x2="1082.12" y2="1204.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="856.289" y1="1386.44" x2="840.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="856.289" y1="1386.44" x2="840.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="856.289" y1="1386.44" x2="872.289" y2="1402.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="856.289" y1="1386.44" x2="872.289" y2="1370.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1450.3" y1="1197.55" x2="1434.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1213.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1450.3" y1="1197.55" x2="1466.3" y2="1181.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1350.38" y1="930.437" x2="1334.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1350.38" y1="930.437" x2="1334.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1350.38" y1="930.437" x2="1366.38" y2="946.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1350.38" y1="930.437" x2="1366.38" y2="914.437" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1529.58" y1="996.113" x2="1513.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1529.58" y1="996.113" x2="1513.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1529.58" y1="996.113" x2="1545.58" y2="1012.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1529.58" y1="996.113" x2="1545.58" y2="980.113" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.87" y1="1216.87" x2="1036.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1232.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1052.87" y1="1216.87" x2="1068.87" y2="1200.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.28" y1="1112.64" x2="1209.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1128.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1225.28" y1="1112.64" x2="1241.28" y2="1096.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.47" y1="955.973" x2="1497.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.47" y1="955.973" x2="1497.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.47" y1="955.973" x2="1529.47" y2="971.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.47" y1="955.973" x2="1529.47" y2="939.973" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.93" y1="1082.34" x2="1458.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1098.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1474.93" y1="1082.34" x2="1490.93" y2="1066.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1530.22" y1="1008.88" x2="1514.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1530.22" y1="1008.88" x2="1514.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1530.22" y1="1008.88" x2="1546.22" y2="1024.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1530.22" y1="1008.88" x2="1546.22" y2="992.884" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.75" y1="828.151" x2="1574.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.75" y1="828.151" x2="1574.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.75" y1="828.151" x2="1606.75" y2="844.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1590.75" y1="828.151" x2="1606.75" y2="812.151" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.49" y1="1238.26" x2="1151.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1254.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1167.49" y1="1238.26" x2="1183.49" y2="1222.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.11" y1="1021.15" x2="1328.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1037.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1344.11" y1="1021.15" x2="1360.11" y2="1005.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1942.19" y1="883.328" x2="1926.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1942.19" y1="883.328" x2="1926.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1942.19" y1="883.328" x2="1958.19" y2="899.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1942.19" y1="883.328" x2="1958.19" y2="867.328" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="800.484" y1="1327.86" x2="784.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="800.484" y1="1327.86" x2="784.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="800.484" y1="1327.86" x2="816.484" y2="1343.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="800.484" y1="1327.86" x2="816.484" y2="1311.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.479" y1="1361.21" x2="863.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.479" y1="1361.21" x2="863.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.479" y1="1361.21" x2="895.479" y2="1377.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="879.479" y1="1361.21" x2="895.479" y2="1345.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1621.59" y1="888.584" x2="1605.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1621.59" y1="888.584" x2="1605.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1621.59" y1="888.584" x2="1637.59" y2="904.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1621.59" y1="888.584" x2="1637.59" y2="872.584" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1351.96" y1="1166.63" x2="1335.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1182.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1351.96" y1="1166.63" x2="1367.96" y2="1150.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.03" y1="917.837" x2="1589.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.03" y1="917.837" x2="1589.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.03" y1="917.837" x2="1621.03" y2="933.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1605.03" y1="917.837" x2="1621.03" y2="901.837" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.5" y1="1173.02" x2="1238.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1189.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.5" y1="1173.02" x2="1270.5" y2="1157.02" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.95" y1="1137.6" x2="1509.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1153.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1525.95" y1="1137.6" x2="1541.95" y2="1121.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="863.983" y1="1256.2" x2="847.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="863.983" y1="1256.2" x2="847.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="863.983" y1="1256.2" x2="879.983" y2="1272.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="863.983" y1="1256.2" x2="879.983" y2="1240.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1763.1" y1="789.55" x2="1747.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1763.1" y1="789.55" x2="1747.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1763.1" y1="789.55" x2="1779.1" y2="805.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1763.1" y1="789.55" x2="1779.1" y2="773.55" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.65" y1="1335.83" x2="1149.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1351.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.65" y1="1335.83" x2="1181.65" y2="1319.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.21" y1="1265.21" x2="1091.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1281.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1107.21" y1="1265.21" x2="1123.21" y2="1249.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1261.91" y1="1136.73" x2="1245.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1152.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1261.91" y1="1136.73" x2="1277.91" y2="1120.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1840.32" y1="839.458" x2="1824.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1840.32" y1="839.458" x2="1824.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1840.32" y1="839.458" x2="1856.32" y2="855.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1840.32" y1="839.458" x2="1856.32" y2="823.458" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1566.73" y1="937.472" x2="1550.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1566.73" y1="937.472" x2="1550.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1566.73" y1="937.472" x2="1582.73" y2="953.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1566.73" y1="937.472" x2="1582.73" y2="921.472" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.17" y1="997.851" x2="1441.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.17" y1="997.851" x2="1441.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.17" y1="997.851" x2="1473.17" y2="1013.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.17" y1="997.851" x2="1473.17" y2="981.851" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1518.09" y1="1097.6" x2="1502.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1113.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1518.09" y1="1097.6" x2="1534.09" y2="1081.6" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.74" y1="1166.18" x2="1246.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1182.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1262.74" y1="1166.18" x2="1278.74" y2="1150.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="725.698" y1="1501.31" x2="709.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="725.698" y1="1501.31" x2="709.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="725.698" y1="1501.31" x2="741.698" y2="1517.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="725.698" y1="1501.31" x2="741.698" y2="1485.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="804.197" y1="1345.85" x2="788.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="804.197" y1="1345.85" x2="788.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="804.197" y1="1345.85" x2="820.197" y2="1361.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="804.197" y1="1345.85" x2="820.197" y2="1329.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1068.71" y1="1367.71" x2="1052.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1383.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1068.71" y1="1367.71" x2="1084.71" y2="1351.71" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.44" y1="832.688" x2="1722.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.44" y1="832.688" x2="1722.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.44" y1="832.688" x2="1754.44" y2="848.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1738.44" y1="832.688" x2="1754.44" y2="816.688" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1575.07" y1="927.362" x2="1559.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1575.07" y1="927.362" x2="1559.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1575.07" y1="927.362" x2="1591.07" y2="943.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1575.07" y1="927.362" x2="1591.07" y2="911.362" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.369" y1="1393.26" x2="947.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.369" y1="1393.26" x2="947.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.369" y1="1393.26" x2="979.369" y2="1409.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.369" y1="1393.26" x2="979.369" y2="1377.26" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1338.29" y1="1101.82" x2="1322.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1117.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1338.29" y1="1101.82" x2="1354.29" y2="1085.82" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.77" y1="1157.2" x2="1326.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1173.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.77" y1="1157.2" x2="1358.77" y2="1141.2" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1465.22" y1="1059.54" x2="1449.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1075.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1465.22" y1="1059.54" x2="1481.22" y2="1043.54" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1667.21" y1="967.122" x2="1651.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1667.21" y1="967.122" x2="1651.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1667.21" y1="967.122" x2="1683.21" y2="983.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1667.21" y1="967.122" x2="1683.21" y2="951.122" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1808.12" y1="803.64" x2="1792.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1808.12" y1="803.64" x2="1792.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1808.12" y1="803.64" x2="1824.12" y2="819.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1808.12" y1="803.64" x2="1824.12" y2="787.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.84" y1="1113.91" x2="1349.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1129.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1365.84" y1="1113.91" x2="1381.84" y2="1097.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.457" y1="1364.93" x2="911.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.457" y1="1364.93" x2="911.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.457" y1="1364.93" x2="943.457" y2="1380.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="927.457" y1="1364.93" x2="943.457" y2="1348.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.588" y1="1394.79" x2="938.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.588" y1="1394.79" x2="938.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.588" y1="1394.79" x2="970.588" y2="1410.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.588" y1="1394.79" x2="970.588" y2="1378.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1600.82" y1="848.267" x2="1584.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1600.82" y1="848.267" x2="1584.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1600.82" y1="848.267" x2="1616.82" y2="864.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1600.82" y1="848.267" x2="1616.82" y2="832.267" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1013.05" y1="1254.21" x2="997.055" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1013.05" y1="1254.21" x2="997.055" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1270.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1013.05" y1="1254.21" x2="1029.05" y2="1238.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.35" y1="1098.11" x2="1147.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1114.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1163.35" y1="1098.11" x2="1179.35" y2="1082.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1400.1" y1="1101.9" x2="1384.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1117.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1400.1" y1="1101.9" x2="1416.1" y2="1085.9" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="710.768" y1="1471.31" x2="694.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="710.768" y1="1471.31" x2="694.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="710.768" y1="1471.31" x2="726.768" y2="1487.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="710.768" y1="1471.31" x2="726.768" y2="1455.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.455" y1="1241.66" x2="938.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.455" y1="1241.66" x2="938.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.455" y1="1241.66" x2="970.455" y2="1257.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="954.455" y1="1241.66" x2="970.455" y2="1225.66" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.63" y1="1188.99" x2="1081.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1204.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1097.63" y1="1188.99" x2="1113.63" y2="1172.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.96" y1="1069.03" x2="1369.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1085.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1385.96" y1="1069.03" x2="1401.96" y2="1053.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.19" y1="997.572" x2="1412.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.19" y1="997.572" x2="1412.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.19" y1="997.572" x2="1444.19" y2="1013.57" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1428.19" y1="997.572" x2="1444.19" y2="981.572" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1023.75" y1="1232.23" x2="1007.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1248.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1023.75" y1="1232.23" x2="1039.75" y2="1216.23" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1336.01" y1="1021.11" x2="1320.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1037.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1336.01" y1="1021.11" x2="1352.01" y2="1005.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.06" y1="1198.05" x2="1326.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1214.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1342.06" y1="1198.05" x2="1358.06" y2="1182.05" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1401.11" y1="1144.76" x2="1385.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1160.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1401.11" y1="1144.76" x2="1417.11" y2="1128.76" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.53" y1="1066.19" x2="1386.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1082.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1402.53" y1="1066.19" x2="1418.53" y2="1050.19" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1117.37" y1="1261.72" x2="1101.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1277.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1117.37" y1="1261.72" x2="1133.37" y2="1245.72" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.68" y1="1081.7" x2="1415.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1097.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1431.68" y1="1081.7" x2="1447.68" y2="1065.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="907.609" y1="1313.11" x2="891.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="907.609" y1="1313.11" x2="891.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="907.609" y1="1313.11" x2="923.609" y2="1329.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="907.609" y1="1313.11" x2="923.609" y2="1297.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1267.73" y1="1057.77" x2="1251.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1073.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1267.73" y1="1057.77" x2="1283.73" y2="1041.77" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="904.55" y1="1372.25" x2="888.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="904.55" y1="1372.25" x2="888.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="904.55" y1="1372.25" x2="920.55" y2="1388.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="904.55" y1="1372.25" x2="920.55" y2="1356.25" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.58" y1="1084.87" x2="1296.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1100.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1312.58" y1="1084.87" x2="1328.58" y2="1068.87" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1134.2" y1="1238.48" x2="1118.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1254.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1134.2" y1="1238.48" x2="1150.2" y2="1222.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1561.34" y1="914.945" x2="1545.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1561.34" y1="914.945" x2="1545.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1561.34" y1="914.945" x2="1577.34" y2="930.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1561.34" y1="914.945" x2="1577.34" y2="898.945" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.28" y1="1184.21" x2="1375.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1200.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1391.28" y1="1184.21" x2="1407.28" y2="1168.21" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1311.21" y1="973.579" x2="1295.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1311.21" y1="973.579" x2="1295.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1311.21" y1="973.579" x2="1327.21" y2="989.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1311.21" y1="973.579" x2="1327.21" y2="957.579" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.94" y1="1419.74" x2="1078.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1435.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1094.94" y1="1419.74" x2="1110.94" y2="1403.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1143.74" y1="1146.65" x2="1127.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1162.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1143.74" y1="1146.65" x2="1159.74" y2="1130.65" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="743.908" y1="1344.73" x2="727.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="743.908" y1="1344.73" x2="727.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="743.908" y1="1344.73" x2="759.908" y2="1360.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="743.908" y1="1344.73" x2="759.908" y2="1328.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.98" y1="1226.45" x2="1099.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1242.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1115.98" y1="1226.45" x2="1131.98" y2="1210.45" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1724.4" y1="1054.35" x2="1708.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1070.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1724.4" y1="1054.35" x2="1740.4" y2="1038.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.408" y1="1314.39" x2="947.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.408" y1="1314.39" x2="947.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.408" y1="1314.39" x2="979.408" y2="1330.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="963.408" y1="1314.39" x2="979.408" y2="1298.39" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.46" y1="1140.38" x2="1238.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1156.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1254.46" y1="1140.38" x2="1270.46" y2="1124.38" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="994.563" y1="1220.85" x2="978.563" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="994.563" y1="1220.85" x2="978.563" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="994.563" y1="1220.85" x2="1010.56" y2="1236.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="994.563" y1="1220.85" x2="1010.56" y2="1204.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1585.12" y1="1061.03" x2="1569.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1077.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1585.12" y1="1061.03" x2="1601.12" y2="1045.03" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.07" y1="993.346" x2="1764.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.07" y1="993.346" x2="1764.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.07" y1="993.346" x2="1796.07" y2="1009.35" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1780.07" y1="993.346" x2="1796.07" y2="977.346" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.71" y1="1150.44" x2="1192.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1166.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.71" y1="1150.44" x2="1224.71" y2="1134.44" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.86" y1="1171.34" x2="1355.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1187.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.86" y1="1171.34" x2="1387.86" y2="1155.34" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.82" y1="1306.89" x2="1094.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1322.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.82" y1="1306.89" x2="1126.82" y2="1290.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.33" y1="1287.85" x2="1192.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1303.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1208.33" y1="1287.85" x2="1224.33" y2="1271.85" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.49" y1="1169.31" x2="1149.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1185.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165.49" y1="1169.31" x2="1181.49" y2="1153.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.71" y1="1125.59" x2="1257.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1141.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1273.71" y1="1125.59" x2="1289.71" y2="1109.59" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1079.45" y1="1149.7" x2="1063.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1165.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1079.45" y1="1149.7" x2="1095.45" y2="1133.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1184.12" y1="1234.64" x2="1168.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1250.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1184.12" y1="1234.64" x2="1200.12" y2="1218.64" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.3" y1="1156.32" x2="1355.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1172.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1371.3" y1="1156.32" x2="1387.3" y2="1140.32" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.95" y1="1258.36" x2="1221.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1274.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1237.95" y1="1258.36" x2="1253.95" y2="1242.36" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="983.358" y1="1291.75" x2="967.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="983.358" y1="1291.75" x2="967.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="983.358" y1="1291.75" x2="999.358" y2="1307.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="983.358" y1="1291.75" x2="999.358" y2="1275.75" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1156.87" y1="1192.88" x2="1140.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1208.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1156.87" y1="1192.88" x2="1172.87" y2="1176.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1496.43" y1="924.203" x2="1480.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1496.43" y1="924.203" x2="1480.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1496.43" y1="924.203" x2="1512.43" y2="940.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1496.43" y1="924.203" x2="1512.43" y2="908.203" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="938.137" y1="1395.53" x2="922.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="938.137" y1="1395.53" x2="922.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="938.137" y1="1395.53" x2="954.137" y2="1411.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="938.137" y1="1395.53" x2="954.137" y2="1379.53" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="745.679" y1="1455.46" x2="729.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="745.679" y1="1455.46" x2="729.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="745.679" y1="1455.46" x2="761.679" y2="1471.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="745.679" y1="1455.46" x2="761.679" y2="1439.46" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1495.35" y1="1033.28" x2="1479.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1049.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1495.35" y1="1033.28" x2="1511.35" y2="1017.28" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1226.7" y1="1172" x2="1210.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1226.7" y1="1172" x2="1210.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1226.7" y1="1172" x2="1242.7" y2="1188" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1226.7" y1="1172" x2="1242.7" y2="1156" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.77" y1="884.258" x2="1746.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.77" y1="884.258" x2="1746.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.77" y1="884.258" x2="1778.77" y2="900.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1762.77" y1="884.258" x2="1778.77" y2="868.258" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="977.334" y1="1349.1" x2="961.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="977.334" y1="1349.1" x2="961.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="977.334" y1="1349.1" x2="993.334" y2="1365.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="977.334" y1="1349.1" x2="993.334" y2="1333.1" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1274.34" y1="1100.11" x2="1258.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1116.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1274.34" y1="1100.11" x2="1290.34" y2="1084.11" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.89" y1="1051.48" x2="1497.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1067.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1513.89" y1="1051.48" x2="1529.89" y2="1035.48" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.49" y1="1113.18" x2="1441.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1129.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1457.49" y1="1113.18" x2="1473.49" y2="1097.18" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.94" y1="1156.93" x2="1405.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1172.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1421.94" y1="1156.93" x2="1437.94" y2="1140.93" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1638.84" y1="1106.16" x2="1622.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1122.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1638.84" y1="1106.16" x2="1654.84" y2="1090.16" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.292" y1="1482.58" x2="859.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.292" y1="1482.58" x2="859.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.292" y1="1482.58" x2="891.292" y2="1498.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="875.292" y1="1482.58" x2="891.292" y2="1466.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.87" y1="1032.79" x2="1274.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1048.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1290.87" y1="1032.79" x2="1306.87" y2="1016.79" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1405.31" y1="1103.12" x2="1389.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1119.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1405.31" y1="1103.12" x2="1421.31" y2="1087.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1142.23" y1="1185.99" x2="1126.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1201.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1142.23" y1="1185.99" x2="1158.23" y2="1169.99" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.433" y1="1329.29" x2="883.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.433" y1="1329.29" x2="883.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.433" y1="1329.29" x2="915.433" y2="1345.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="899.433" y1="1329.29" x2="915.433" y2="1313.29" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.35" y1="1160.97" x2="1224.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1176.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1240.35" y1="1160.97" x2="1256.35" y2="1144.97" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1058.78" y1="1213.58" x2="1042.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1229.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1058.78" y1="1213.58" x2="1074.78" y2="1197.58" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1093.83" y1="1252.83" x2="1077.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1268.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1093.83" y1="1252.83" x2="1109.83" y2="1236.83" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1423.88" y1="1021.7" x2="1407.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1037.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1423.88" y1="1021.7" x2="1439.88" y2="1005.7" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1552.62" y1="936.348" x2="1536.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1552.62" y1="936.348" x2="1536.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1552.62" y1="936.348" x2="1568.62" y2="952.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1552.62" y1="936.348" x2="1568.62" y2="920.348" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.54" y1="984.865" x2="1456.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.54" y1="984.865" x2="1456.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.54" y1="984.865" x2="1488.54" y2="1000.86" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1472.54" y1="984.865" x2="1488.54" y2="968.865" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165" y1="1027.31" x2="1149" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165" y1="1027.31" x2="1149" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165" y1="1027.31" x2="1181" y2="1043.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1165" y1="1027.31" x2="1181" y2="1011.31" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.61" y1="1250.08" x2="1094.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1266.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1110.61" y1="1250.08" x2="1126.61" y2="1234.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1154.89" y1="1238.14" x2="1138.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1254.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1154.89" y1="1238.14" x2="1170.89" y2="1222.14" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.8" y1="1203.89" x2="1095.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1219.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1111.8" y1="1203.89" x2="1127.8" y2="1187.89" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239" y1="1106.12" x2="1223" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239" y1="1106.12" x2="1223" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239" y1="1106.12" x2="1255" y2="1122.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1239" y1="1106.12" x2="1255" y2="1090.12" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="839.379" y1="1287.74" x2="823.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="839.379" y1="1287.74" x2="823.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="839.379" y1="1287.74" x2="855.379" y2="1303.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="839.379" y1="1287.74" x2="855.379" y2="1271.74" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1100.71" y1="1170.24" x2="1084.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1186.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1100.71" y1="1170.24" x2="1116.71" y2="1154.24" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414" y1="1170.91" x2="1398" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414" y1="1170.91" x2="1398" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414" y1="1170.91" x2="1430" y2="1186.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414" y1="1170.91" x2="1430" y2="1154.91" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.86" y1="1217.88" x2="1398.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1233.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1414.86" y1="1217.88" x2="1430.86" y2="1201.88" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1194.86" y1="1244.81" x2="1178.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1260.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1194.86" y1="1244.81" x2="1210.86" y2="1228.81" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1257.94" y1="1026.73" x2="1241.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1042.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1257.94" y1="1026.73" x2="1273.94" y2="1010.73" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1296.85" y1="1204.15" x2="1280.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1220.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1296.85" y1="1204.15" x2="1312.85" y2="1188.15" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.11" y1="1134.69" x2="1227.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1150.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1243.11" y1="1134.69" x2="1259.11" y2="1118.69" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.52" y1="1257.08" x2="1321.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1273.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="1337.52" y1="1257.08" x2="1353.52" y2="1241.08" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="844.925" y1="1383.63" x2="828.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="844.925" y1="1383.63" x2="828.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="844.925" y1="1383.63" x2="860.925" y2="1399.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
+<line clip-path="url(#clip892)" x1="844.925" y1="1383.63" x2="860.925" y2="1367.63" style="stroke:#008000; stroke-width:4.8; stroke-opacity:1"/>
 </svg>
diff --git a/previews/PR3562/tutorials/conic/ellipse_approx/index.html b/previews/PR3562/tutorials/conic/ellipse_approx/index.html
index 0acab3509f1..fc66c07e012 100644
--- a/previews/PR3562/tutorials/conic/ellipse_approx/index.html
+++ b/previews/PR3562/tutorials/conic/ellipse_approx/index.html
@@ -35,7 +35,7 @@
     c = :green,
     shape = :x,
     size = (600, 600),
-)</code></pre><img src="b2c746af.svg" alt="Example block output"/><h2 id="JuMP-formulation"><a class="docs-heading-anchor" href="#JuMP-formulation">JuMP formulation</a><a id="JuMP-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-formulation" title="Permalink"></a></h2><p>Now let&#39;s build and the JuMP model. We&#39;ll compute <span>$D$</span> and <span>$c$</span> after the solve.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+)</code></pre><img src="f9ac62e0.svg" alt="Example block output"/><h2 id="JuMP-formulation"><a class="docs-heading-anchor" href="#JuMP-formulation">JuMP formulation</a><a id="JuMP-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-formulation" title="Permalink"></a></h2><p>Now let&#39;s build and the JuMP model. We&#39;ll compute <span>$D$</span> and <span>$c$</span> after the solve.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 # We need to use a tighter tolerance for this example, otherwise the bounding
 # ellipse won&#39;t actually be bounding...
 set_attribute(model, &quot;eps_rel&quot;, 1e-6)
@@ -69,15 +69,15 @@
   Dual objective value : 4.47966e-03
 
 * Work counters
-  Solve time (sec)   : 5.65044e-01
+  Solve time (sec)   : 9.05069e-01
 </code></pre><h2 id="Results"><a class="docs-heading-anchor" href="#Results">Results</a><a id="Results-1"></a><a class="docs-heading-anchor-permalink" href="#Results" title="Permalink"></a></h2><p>After solving the model to optimality we can recover the solution in terms of <span>$D$</span> and <span>$c$</span>:</p><pre><code class="language-julia hljs">D = value.(Z)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2×2 Matrix{Float64}:
   0.00707822  -0.0102173
  -0.0102173    0.0175624</code></pre><pre><code class="language-julia hljs">c = D \ value.(z)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- -3.2480287747786316
- -1.8428256866059616</code></pre><p>Finally, overlaying the solution in the plot we see the minimal volume approximating ellipsoid:</p><pre><code class="language-julia hljs">P = sqrt(D)
+ -3.2480287747852175
+ -1.8428256866083088</code></pre><p>Finally, overlaying the solution in the plot we see the minimal volume approximating ellipsoid:</p><pre><code class="language-julia hljs">P = sqrt(D)
 q = -P * c
 data = [tuple(P \ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)]
-Plots.plot!(plot, data; c = :crimson, label = nothing)</code></pre><img src="053884ea.svg" alt="Example block output"/><h2 id="Alternative-formulations"><a class="docs-heading-anchor" href="#Alternative-formulations">Alternative formulations</a><a id="Alternative-formulations-1"></a><a class="docs-heading-anchor-permalink" href="#Alternative-formulations" title="Permalink"></a></h2><p>The formulation of <code>model</code> uses <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>MOI.RootDetConeSquare</code></a>. However, because SCS does not natively support this cone, JuMP automatically reformulates the problem into an equivalent problem that SCS <em>does</em> support. You can see the reformulation that JuMP chose using <a href="../../../api/JuMP/#JuMP.print_active_bridges"><code>print_active_bridges</code></a>:</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr31">Unsupported objective: MOI.VariableIndex</span>
+Plots.plot!(plot, data; c = :crimson, label = nothing)</code></pre><img src="618bca67.svg" alt="Example block output"/><h2 id="Alternative-formulations"><a class="docs-heading-anchor" href="#Alternative-formulations">Alternative formulations</a><a id="Alternative-formulations-1"></a><a class="docs-heading-anchor-permalink" href="#Alternative-formulations" title="Permalink"></a></h2><p>The formulation of <code>model</code> uses <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>MOI.RootDetConeSquare</code></a>. However, because SCS does not natively support this cone, JuMP automatically reformulates the problem into an equivalent problem that SCS <em>does</em> support. You can see the reformulation that JuMP chose using <a href="../../../api/JuMP/#JuMP.print_active_bridges"><code>print_active_bridges</code></a>:</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr31">Unsupported objective: MOI.VariableIndex</span>
  |  bridged by:
  |   MOIB.Objective.FunctionizeBridge{Float64}
  |  may introduce:
@@ -256,7 +256,7 @@
 # The former @objective(model, Max, t)
 @objective(model, Max, 1 * t + 0)
 optimize!(model)
-solve_time_1 = solve_time(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.552436999</code></pre><p>This formulation gives the much smaller graph:</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr32">Supported objective: MOI.ScalarAffineFunction{Float64}</span>
+solve_time_1 = solve_time(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.921479728</code></pre><p>This formulation gives the much smaller graph:</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr32">Supported objective: MOI.ScalarAffineFunction{Float64}</span>
  * <span class="sgr32">Supported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.Nonnegatives</span>
  * <span class="sgr31">Unsupported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.PositiveSemidefiniteConeTriangle</span>
  |  bridged by:
@@ -294,7 +294,7 @@
  |   |   |   |  may introduce:
  |   |   |   |   * <span class="sgr32">Supported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.Nonnegatives</span>
  |   * <span class="sgr32">Supported variable: MOI.Reals</span></code></pre><p>The last bullet shows how JuMP reformulated the <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>MOI.RootDetConeTriangle</code></a> constraint by adding a mix of <a href="../../../moi/reference/standard_form/#MathOptInterface.PositiveSemidefiniteConeTriangle"><code>MOI.PositiveSemidefiniteConeTriangle</code></a> and <a href="../../../moi/reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>MOI.GeometricMeanCone</code></a> constraints.</p><p>Because SCS doesn&#39;t natively support the <a href="../../../moi/reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>MOI.GeometricMeanCone</code></a>, these constraints were further bridged using a <a href="../../../moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge"><code>MOI.Bridges.Constraint.GeoMeanToPowerBridge</code></a> to a series of <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a> constraints.</p><p>However, there are many other ways that a <a href="../../../moi/reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>MOI.GeometricMeanCone</code></a> can be reformulated into something that SCS supports. Let&#39;s see what happens if we use <a href="../../../api/JuMP/#JuMP.remove_bridge"><code>remove_bridge</code></a> to remove the <a href="../../../moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanToPowerBridge"><code>MOI.Bridges.Constraint.GeoMeanToPowerBridge</code></a>:</p><pre><code class="language-julia hljs">remove_bridge(model, MOI.Bridges.Constraint.GeoMeanToPowerBridge)
-optimize!(model)</code></pre><p>This time, the solve took:</p><pre><code class="language-julia hljs">solve_time_2 = solve_time(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.330904203</code></pre><p>where previously it took</p><pre><code class="language-julia hljs">solve_time_1</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.552436999</code></pre><p>Why was the solve time different?</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr32">Supported objective: MOI.ScalarAffineFunction{Float64}</span>
+optimize!(model)</code></pre><p>This time, the solve took:</p><pre><code class="language-julia hljs">solve_time_2 = solve_time(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.5831680170000001</code></pre><p>where previously it took</p><pre><code class="language-julia hljs">solve_time_1</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.921479728</code></pre><p>Why was the solve time different?</p><pre><code class="language-julia hljs">print_active_bridges(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi"> * <span class="sgr32">Supported objective: MOI.ScalarAffineFunction{Float64}</span>
  * <span class="sgr32">Supported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.Nonnegatives</span>
  * <span class="sgr31">Unsupported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.PositiveSemidefiniteConeTriangle</span>
  |  bridged by:
@@ -338,4 +338,4 @@
  |   |   |   * <span class="sgr32">Supported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.SecondOrderCone</span>
  |   |   * <span class="sgr32">Supported constraint: MOI.VectorAffineFunction{Float64}-in-MOI.Nonnegatives</span>
  |   |   * <span class="sgr32">Supported variable: MOI.Reals</span>
- |   * <span class="sgr32">Supported variable: MOI.Reals</span></code></pre><p>This time, JuMP used a <a href="../../../moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanBridge"><code>MOI.Bridges.Constraint.GeoMeanBridge</code></a> to reformulate the constraint into a set of <a href="../../../moi/reference/standard_form/#MathOptInterface.RotatedSecondOrderCone"><code>MOI.RotatedSecondOrderCone</code></a> constraints, which were further reformulated into a set of supported <a href="../../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a> constraints.</p><p>Since the two models are equivalent, we can conclude that for this particular model, the <a href="../../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a> formulation is more efficient.</p><p>In general though, the performance of a particular reformulation is problem- and solver-specific. Therefore, JuMP chooses to minimize the number of bridges in the default reformulation, leaving you to explore alternative formulations using the tools and techniques shown in this tutorial.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../min_ellipse/">« Minimal ellipses</a><a class="docs-footer-nextpage" href="../quantum_discrimination/">Quantum state discrimination »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ |   * <span class="sgr32">Supported variable: MOI.Reals</span></code></pre><p>This time, JuMP used a <a href="../../../moi/submodules/Bridges/list_of_bridges/#MathOptInterface.Bridges.Constraint.GeoMeanBridge"><code>MOI.Bridges.Constraint.GeoMeanBridge</code></a> to reformulate the constraint into a set of <a href="../../../moi/reference/standard_form/#MathOptInterface.RotatedSecondOrderCone"><code>MOI.RotatedSecondOrderCone</code></a> constraints, which were further reformulated into a set of supported <a href="../../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a> constraints.</p><p>Since the two models are equivalent, we can conclude that for this particular model, the <a href="../../../moi/reference/standard_form/#MathOptInterface.SecondOrderCone"><code>MOI.SecondOrderCone</code></a> formulation is more efficient.</p><p>In general though, the performance of a particular reformulation is problem- and solver-specific. Therefore, JuMP chooses to minimize the number of bridges in the default reformulation, leaving you to explore alternative formulations using the tools and techniques shown in this tutorial.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../min_ellipse/">« Minimal ellipses</a><a class="docs-footer-nextpage" href="../quantum_discrimination/">Quantum state discrimination »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/experiment_design/index.html b/previews/PR3562/tutorials/conic/experiment_design/index.html
index 694248023a6..2e737e423c0 100644
--- a/previews/PR3562/tutorials/conic/experiment_design/index.html
+++ b/previews/PR3562/tutorials/conic/experiment_design/index.html
@@ -41,15 +41,15 @@
 end
 @objective(aOpt, Min, sum(u))
 optimize!(aOpt)
-objective_value(aOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">5.103223362935152</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
- 2.9495211350241117
- 1.7790921964770656
- 8.104871340374779e-6
- 4.706163786116579e-6
- 2.1082669841842727
- 1.6983463798099407
- 1.2635010842547627
- 2.2012292330174965</code></pre><h2 id="E-optimal-design"><a class="docs-heading-anchor" href="#E-optimal-design">E-optimal design</a><a id="E-optimal-design-1"></a><a class="docs-heading-anchor-permalink" href="#E-optimal-design" title="Permalink"></a></h2><p>In <span>$E$</span> -optimal design, we minimize the norm of the error covariance matrix, that is, the maximum eigenvalue of <span>$E$</span>.</p><p>Since the diameter (twice the longest semi-axis) of the confidence ellipsoid <span>$\mathcal{E}$</span> is proportional to <span>$\|E\|_{2}^{1 / 2}$</span>, minimizing <span>$\|E\|_{2}$</span> can be interpreted geometrically as minimizing the diameter of the confidence ellipsoid.</p><p>E-optimal design can also be interpreted as minimizing the maximum variance of <span>$q^{T} e$</span>, over all <span>$q$</span> with <span>$\|q\|_{2}=1$</span>. The E-optimal experiment design problem in SDP form is:</p><p class="math-container">\[\begin{aligned}
+objective_value(aOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">5.103223362935134</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
+ 2.949521135024003
+ 1.7790921964770645
+ 8.10487134967608e-6
+ 4.706163786495679e-6
+ 2.1082669841842883
+ 1.6983463798099796
+ 1.2635010842548617
+ 2.201229233017432</code></pre><h2 id="E-optimal-design"><a class="docs-heading-anchor" href="#E-optimal-design">E-optimal design</a><a id="E-optimal-design-1"></a><a class="docs-heading-anchor-permalink" href="#E-optimal-design" title="Permalink"></a></h2><p>In <span>$E$</span> -optimal design, we minimize the norm of the error covariance matrix, that is, the maximum eigenvalue of <span>$E$</span>.</p><p>Since the diameter (twice the longest semi-axis) of the confidence ellipsoid <span>$\mathcal{E}$</span> is proportional to <span>$\|E\|_{2}^{1 / 2}$</span>, minimizing <span>$\|E\|_{2}$</span> can be interpreted geometrically as minimizing the diameter of the confidence ellipsoid.</p><p>E-optimal design can also be interpreted as minimizing the maximum variance of <span>$q^{T} e$</span>, over all <span>$q$</span> with <span>$\|q\|_{2}=1$</span>. The E-optimal experiment design problem in SDP form is:</p><p class="math-container">\[\begin{aligned}
 \min &amp; t \\
 \text{subject to} &amp; \sum_{i=1}^{p} \lambda_{i} v_{i} v_{i}^{T} \succeq t I \\
 &amp; \lambda \succeq 0 \\
@@ -66,15 +66,15 @@
 @constraint(eOpt, sum(np) &lt;= n)
 @objective(eOpt, Max, t)
 optimize!(eOpt)
-objective_value(eOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.43538638565427923</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
-  2.999998678490912
-  2.752810163966476
- -3.609586470621804e-7
- -1.4872470371469733e-6
-  2.181846279169085
-  2.3252635465673577
-  0.21896764651694434
-  1.5211163049621212</code></pre><h2 id="D-optimal-design"><a class="docs-heading-anchor" href="#D-optimal-design">D-optimal design</a><a id="D-optimal-design-1"></a><a class="docs-heading-anchor-permalink" href="#D-optimal-design" title="Permalink"></a></h2><p>The most widely used scalarization is called <span>$D$</span> -optimal design, in which we minimize the determinant of the error covariance matrix <span>$E$</span>. This corresponds to designing the experiment to minimize the volume of the resulting confidence ellipsoid (for a fixed confidence level). Ignoring the constant factor <span>$1 / m$</span> in <span>$E$</span>, and taking the logarithm of the objective, we can pose this problem as convex optimization problem:</p><p class="math-container">\[\begin{aligned}
+objective_value(eOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.4353863856688645</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
+  2.9999986784996944
+  2.752810163779928
+ -3.6095474331105857e-7
+ -1.4872411636950634e-6
+  2.1818462791954287
+  2.325263546428314
+  0.21896764666380808
+  1.5211163050754828</code></pre><h2 id="D-optimal-design"><a class="docs-heading-anchor" href="#D-optimal-design">D-optimal design</a><a id="D-optimal-design-1"></a><a class="docs-heading-anchor-permalink" href="#D-optimal-design" title="Permalink"></a></h2><p>The most widely used scalarization is called <span>$D$</span> -optimal design, in which we minimize the determinant of the error covariance matrix <span>$E$</span>. This corresponds to designing the experiment to minimize the volume of the resulting confidence ellipsoid (for a fixed confidence level). Ignoring the constant factor <span>$1 / m$</span> in <span>$E$</span>, and taking the logarithm of the objective, we can pose this problem as convex optimization problem:</p><p class="math-container">\[\begin{aligned}
 \min &amp; \log \operatorname{det}\left(\sum_{i=1}^{p} \lambda_{i} v_{i} v_{i}^{T}\right)^{-1} \\
 \text{subject to} &amp; \lambda \succeq 0 \\
 &amp; \mathbf{1}^{T} \lambda=1
@@ -87,12 +87,12 @@
 E = V * LinearAlgebra.diagm(0 =&gt; np ./ n) * V&#39;
 @constraint(dOpt, [t; 1; triangle_vec(E)] in MOI.LogDetConeTriangle(q))
 optimize!(dOpt)
-objective_value(dOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.3084539346728666</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
- 0.4275824266217608
- 2.9100163168942075
- 2.9168724521318788e-6
- 2.6291979699361115e-6
- 2.915780629983959
- 2.673375664592317
- 2.7353950122201236
- 0.33783880862535554</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../logistic_regression/">« Logistic regression</a><a class="docs-footer-nextpage" href="../min_ellipse/">Minimal ellipses »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+objective_value(dOpt)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.3084539346723429</code></pre><pre><code class="language-julia hljs">value.(np)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8-element Vector{Float64}:
+ 0.4275824266221196
+ 2.910016316893973
+ 2.9168724688641994e-6
+ 2.629197986378904e-6
+ 2.9157806299834754
+ 2.6733756645925353
+ 2.7353950122195503
+ 0.3378388086259769</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../logistic_regression/">« Logistic regression</a><a class="docs-footer-nextpage" href="../min_ellipse/">Minimal ellipses »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/introduction/index.html b/previews/PR3562/tutorials/conic/introduction/index.html
index 5fec4ddea9d..a9048501553 100644
--- a/previews/PR3562/tutorials/conic/introduction/index.html
+++ b/previews/PR3562/tutorials/conic/introduction/index.html
@@ -6,4 +6,4 @@
 </script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox" checked/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox"/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../getting_started/introduction/">Introduction</a></li><li><a class="tocitem" href="../../getting_started/getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../../getting_started/getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../../getting_started/getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../../getting_started/getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../../getting_started/debugging/">Debugging</a></li><li><a class="tocitem" href="../../getting_started/design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../../getting_started/performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox" checked/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#How-to-choose-a-solver"><span>How to choose a solver</span></a></li><li><a class="tocitem" href="#How-these-tutorials-are-structured"><span>How these tutorials are structured</span></a></li></ul></li><li><a class="tocitem" href="../start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../dualization/">Dualization</a></li><li><a class="tocitem" href="../arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/">Introduction</a></li><li><a class="tocitem" href="../../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Tutorials</a></li><li><a class="is-disabled">Conic programs</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/tutorials/conic/introduction.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p><a href="https://en.wikipedia.org/wiki/Conic_optimization">Conic programs</a> are a class of convex nonlinear optimization problems which use cones to represent the nonlinearities. They have the form:</p><p class="math-container">\[\begin{align}
     &amp; \min_{x \in \mathbb{R}^n} &amp; f_0(x) \\
     &amp; \;\;\text{s.t.} &amp; f_j(x) \in \mathcal{S}_j &amp; \;\; j = 1 \ldots m
-\end{align}\]</p><p>Mixed-integer conic programs (MICPs) are extensions of conic programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>JuMP supports a range of conic solvers, although support differs on what types of cones each solver supports. In the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>, &quot;SOCP&quot; denotes solvers supporting second-order cones and &quot;SDP&quot; denotes solvers supporting semidefinite cones. In addition, solvers such as SCS and Mosek have support for the exponential cone. Moreover, due to the bridging system in MathOptInterface, many of these solvers support a much wider range of exotic cones than they natively support. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Duality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual instead of the primal. If performance is an issue, see the <a href="../dualization/#Dualization">Dualization</a> tutorial for more details.</p></div></div><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../experiment_design/#Experiment-design">Experiment design</a></li><li><a href="../logistic_regression/#Logistic-regression">Logistic regression</a></li></ul></li><li>The <a href="../tips_and_tricks/#conic_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling conic programs. Look here if you are stuck trying to formulate a problem as a conic program.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../nonlinear/complementarity/">« Mixed complementarity problems</a><a class="docs-footer-nextpage" href="../start_values/">Primal and dual warm-starts »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>Mixed-integer conic programs (MICPs) are extensions of conic programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>JuMP supports a range of conic solvers, although support differs on what types of cones each solver supports. In the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>, &quot;SOCP&quot; denotes solvers supporting second-order cones and &quot;SDP&quot; denotes solvers supporting semidefinite cones. In addition, solvers such as SCS and Mosek have support for the exponential cone. Moreover, due to the bridging system in MathOptInterface, many of these solvers support a much wider range of exotic cones than they natively support. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Duality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual instead of the primal. If performance is an issue, see the <a href="../dualization/#Dualization">Dualization</a> tutorial for more details.</p></div></div><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../experiment_design/#Experiment-design">Experiment design</a></li><li><a href="../logistic_regression/#Logistic-regression">Logistic regression</a></li></ul></li><li>The <a href="../tips_and_tricks/#conic_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling conic programs. Look here if you are stuck trying to formulate a problem as a conic program.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../nonlinear/complementarity/">« Mixed complementarity problems</a><a class="docs-footer-nextpage" href="../start_values/">Primal and dual warm-starts »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/logistic_regression/index.html b/previews/PR3562/tutorials/conic/logistic_regression/index.html
index ed36fe4817f..b7808a6f772 100644
--- a/previews/PR3562/tutorials/conic/logistic_regression/index.html
+++ b/previews/PR3562/tutorials/conic/logistic_regression/index.html
@@ -59,17 +59,17 @@
 set_optimizer(model, SCS.Optimizer)
 set_silent(model)
 JuMP.optimize!(model)</code></pre><pre><code class="language-julia hljs">θ♯ = JuMP.value.(model[:θ])</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">11-element Vector{Float64}:
-  0.020413203134135874
-  0.16139903246564888
-  0.3570077183841772
- -0.30788082187579985
- -0.3939184218208197
- -0.05914040150988693
-  0.34717365911625764
- -0.8812123235442034
-  0.20125660911983972
-  0.5409398851900533
-  0.08090419699788672</code></pre><p>It appears that the speed of convergence is not that impacted by the correlation of the dataset, nor by the penalty <span>$\lambda$</span>.</p><h3 id="\\ell_1-regularized-logistic-regression"><a class="docs-heading-anchor" href="#\\ell_1-regularized-logistic-regression"><span>$\ell_1$</span> regularized logistic regression</a><a id="\\ell_1-regularized-logistic-regression-1"></a><a class="docs-heading-anchor-permalink" href="#\\ell_1-regularized-logistic-regression" title="Permalink"></a></h3><p>We now formulate the logistic problem with a <span>$\ell_1$</span> regularization term. The <span>$\ell_1$</span> regularization ensures sparsity in the optimal solution of the resulting optimization problem. Luckily, the <span>$\ell_1$</span> norm is implemented as a set in <code>MathOptInterface</code>. Thus, we could formulate the sparse logistic regression problem with the help of a <code>MOI.NormOneCone</code> set.</p><pre><code class="language-julia hljs">function build_sparse_logit_model(X, y, λ)
+  0.02041320313399188
+  0.16139903246568316
+  0.3570077183843268
+ -0.3078808218759439
+ -0.3939184218209453
+ -0.05914040150982515
+  0.3471736591163877
+ -0.8812123235444688
+  0.20125660912000326
+  0.5409398851901871
+  0.08090419699796532</code></pre><p>It appears that the speed of convergence is not that impacted by the correlation of the dataset, nor by the penalty <span>$\lambda$</span>.</p><h3 id="\\ell_1-regularized-logistic-regression"><a class="docs-heading-anchor" href="#\\ell_1-regularized-logistic-regression"><span>$\ell_1$</span> regularized logistic regression</a><a id="\\ell_1-regularized-logistic-regression-1"></a><a class="docs-heading-anchor-permalink" href="#\\ell_1-regularized-logistic-regression" title="Permalink"></a></h3><p>We now formulate the logistic problem with a <span>$\ell_1$</span> regularization term. The <span>$\ell_1$</span> regularization ensures sparsity in the optimal solution of the resulting optimization problem. Luckily, the <span>$\ell_1$</span> norm is implemented as a set in <code>MathOptInterface</code>. Thus, we could formulate the sparse logistic regression problem with the help of a <code>MOI.NormOneCone</code> set.</p><pre><code class="language-julia hljs">function build_sparse_logit_model(X, y, λ)
     n, p = size(X)
     model = Model()
     @variable(model, θ[1:p])
@@ -104,4 +104,4 @@
 )</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Number of non-zero components: 8 (out of 10 features)</code></pre><h3 id="Extensions"><a class="docs-heading-anchor" href="#Extensions">Extensions</a><a id="Extensions-1"></a><a class="docs-heading-anchor-permalink" href="#Extensions" title="Permalink"></a></h3><p>A direct extension would be to consider the sparse logistic regression with <em>hard</em> thresholding, which, on contrary to the <em>soft</em> version using a <span>$\ell_1$</span> regularization, adds an explicit cardinality constraint in its formulation:</p><p class="math-container">\[\begin{aligned}
 \min_{\theta} &amp; \; \sum_{i=1}^n \log(1 + \exp(-y_i \theta^\top x_i)) + \lambda \| \theta \|_2^2 \\
 \text{subject to } &amp; \quad \| \theta \|_0 &lt;= k
-\end{aligned}\]</p><p>where <span>$k$</span> is the maximum number of non-zero components in the vector <span>$\theta$</span>, and <span>$\|.\|_0$</span> is the <span>$\ell_0$</span> pseudo-norm:</p><p class="math-container">\[\| x\|_0 = \#\{i : \; x_i \neq 0\}\]</p><p>The cardinality constraint <span>$\|\theta\|_0 \leq k$</span> could be reformulated with binary variables. Thus the hard sparse regression problem could be solved by any solver supporting mixed integer conic problems.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../arbitrary_precision/">« Arbitrary precision arithmetic</a><a class="docs-footer-nextpage" href="../experiment_design/">Experiment design »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{aligned}\]</p><p>where <span>$k$</span> is the maximum number of non-zero components in the vector <span>$\theta$</span>, and <span>$\|.\|_0$</span> is the <span>$\ell_0$</span> pseudo-norm:</p><p class="math-container">\[\| x\|_0 = \#\{i : \; x_i \neq 0\}\]</p><p>The cardinality constraint <span>$\|\theta\|_0 \leq k$</span> could be reformulated with binary variables. Thus the hard sparse regression problem could be solved by any solver supporting mixed integer conic problems.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../arbitrary_precision/">« Arbitrary precision arithmetic</a><a class="docs-footer-nextpage" href="../experiment_design/">Experiment design »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/min_ellipse/1521bd4d.svg b/previews/PR3562/tutorials/conic/min_ellipse/8380bd2b.svg
similarity index 88%
rename from previews/PR3562/tutorials/conic/min_ellipse/1521bd4d.svg
rename to previews/PR3562/tutorials/conic/min_ellipse/8380bd2b.svg
index 53d24b4e4cb..86bc0750702 100644
--- a/previews/PR3562/tutorials/conic/min_ellipse/1521bd4d.svg
+++ b/previews/PR3562/tutorials/conic/min_ellipse/8380bd2b.svg
@@ -1,53 +1,53 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="600" viewBox="0 0 2400 2400">
 <defs>
-  <clipPath id="clip250">
+  <clipPath id="clip070">
     <rect x="0" y="0" width="2400" height="2400"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip250)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip070)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip251">
+  <clipPath id="clip071">
     <rect x="480" y="240" width="1681" height="1681"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip250)" d="M148.057 2261.17 L2352.76 2261.17 L2352.76 47.2441 L148.057 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip070)" d="M148.057 2261.17 L2352.76 2261.17 L2352.76 47.2441 L148.057 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip252">
+  <clipPath id="clip072">
     <rect x="148" y="47" width="2206" height="2215"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="406.43,2261.17 406.43,47.2441 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.578,2261.17 838.578,47.2441 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1270.73,2261.17 1270.73,47.2441 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1702.88,2261.17 1702.88,47.2441 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2135.02,2261.17 2135.02,47.2441 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2261.17 2352.76,2261.17 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="406.43,2261.17 406.43,2242.27 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.578,2261.17 838.578,2242.27 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1270.73,2261.17 1270.73,2242.27 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1702.88,2261.17 1702.88,2242.27 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.02,2261.17 2135.02,2242.27 "/>
-<path clip-path="url(#clip250)" d="M376.187 2306.74 L405.862 2306.74 L405.862 2310.67 L376.187 2310.67 L376.187 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M416.765 2319.63 L424.404 2319.63 L424.404 2293.27 L416.094 2294.93 L416.094 2290.67 L424.358 2289.01 L429.034 2289.01 L429.034 2319.63 L436.672 2319.63 L436.672 2323.57 L416.765 2323.57 L416.765 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M838.578 2292.09 Q834.967 2292.09 833.138 2295.65 Q831.333 2299.19 831.333 2306.32 Q831.333 2313.43 833.138 2316.99 Q834.967 2320.54 838.578 2320.54 Q842.212 2320.54 844.018 2316.99 Q845.847 2313.43 845.847 2306.32 Q845.847 2299.19 844.018 2295.65 Q842.212 2292.09 838.578 2292.09 M838.578 2288.38 Q844.388 2288.38 847.444 2292.99 Q850.523 2297.57 850.523 2306.32 Q850.523 2315.05 847.444 2319.66 Q844.388 2324.24 838.578 2324.24 Q832.768 2324.24 829.689 2319.66 Q826.634 2315.05 826.634 2306.32 Q826.634 2297.57 829.689 2292.99 Q832.768 2288.38 838.578 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1261.11 2319.63 L1268.75 2319.63 L1268.75 2293.27 L1260.44 2294.93 L1260.44 2290.67 L1268.7 2289.01 L1273.38 2289.01 L1273.38 2319.63 L1281.02 2319.63 L1281.02 2323.57 L1261.11 2323.57 L1261.11 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1697.53 2319.63 L1713.85 2319.63 L1713.85 2323.57 L1691.9 2323.57 L1691.9 2319.63 Q1694.57 2316.88 1699.15 2312.25 Q1703.75 2307.6 1704.94 2306.25 Q1707.18 2303.73 1708.06 2301.99 Q1708.96 2300.23 1708.96 2298.54 Q1708.96 2295.79 1707.02 2294.05 Q1705.1 2292.32 1702 2292.32 Q1699.8 2292.32 1697.34 2293.08 Q1694.91 2293.85 1692.13 2295.4 L1692.13 2290.67 Q1694.96 2289.54 1697.41 2288.96 Q1699.87 2288.38 1701.9 2288.38 Q1707.27 2288.38 1710.47 2291.07 Q1713.66 2293.75 1713.66 2298.24 Q1713.66 2300.37 1712.85 2302.29 Q1712.07 2304.19 1709.96 2306.79 Q1709.38 2307.46 1706.28 2310.67 Q1703.18 2313.87 1697.53 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2139.27 2304.93 Q2142.63 2305.65 2144.5 2307.92 Q2146.4 2310.19 2146.4 2313.52 Q2146.4 2318.64 2142.88 2321.44 Q2139.36 2324.24 2132.88 2324.24 Q2130.71 2324.24 2128.39 2323.8 Q2126.1 2323.38 2123.65 2322.53 L2123.65 2318.01 Q2125.59 2319.15 2127.91 2319.73 Q2130.22 2320.3 2132.74 2320.3 Q2137.14 2320.3 2139.43 2318.57 Q2141.75 2316.83 2141.75 2313.52 Q2141.75 2310.47 2139.6 2308.75 Q2137.47 2307.02 2133.65 2307.02 L2129.62 2307.02 L2129.62 2303.17 L2133.83 2303.17 Q2137.28 2303.17 2139.11 2301.81 Q2140.94 2300.42 2140.94 2297.83 Q2140.94 2295.17 2139.04 2293.75 Q2137.17 2292.32 2133.65 2292.32 Q2131.73 2292.32 2129.53 2292.73 Q2127.33 2293.15 2124.69 2294.03 L2124.69 2289.86 Q2127.35 2289.12 2129.67 2288.75 Q2132 2288.38 2134.06 2288.38 Q2139.39 2288.38 2142.49 2290.81 Q2145.59 2293.22 2145.59 2297.34 Q2145.59 2300.21 2143.95 2302.2 Q2142.3 2304.17 2139.27 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,2190 2352.76,2190 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1864.99 2352.76,1864.99 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1539.98 2352.76,1539.98 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1214.97 2352.76,1214.97 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,889.962 2352.76,889.962 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,564.952 2352.76,564.952 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,239.942 2352.76,239.942 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2261.17 148.057,47.2441 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2190 166.954,2190 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1864.99 166.954,1864.99 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1539.98 166.954,1539.98 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1214.97 166.954,1214.97 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,889.962 166.954,889.962 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,564.952 166.954,564.952 "/>
-<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,239.942 166.954,239.942 "/>
-<path clip-path="url(#clip250)" d="M50.9921 2190.45 L80.6679 2190.45 L80.6679 2194.39 L50.9921 2194.39 L50.9921 2190.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M104.927 2188.65 Q108.283 2189.36 110.158 2191.63 Q112.057 2193.9 112.057 2197.23 Q112.057 2202.35 108.538 2205.15 Q105.02 2207.95 98.5382 2207.95 Q96.3623 2207.95 94.0475 2207.51 Q91.7558 2207.1 89.3021 2206.24 L89.3021 2201.72 Q91.2465 2202.86 93.5613 2203.44 Q95.8761 2204.02 98.3993 2204.02 Q102.797 2204.02 105.089 2202.28 Q107.404 2200.54 107.404 2197.23 Q107.404 2194.18 105.251 2192.47 Q103.121 2190.73 99.3021 2190.73 L95.2743 2190.73 L95.2743 2186.89 L99.4872 2186.89 Q102.936 2186.89 104.765 2185.52 Q106.594 2184.13 106.594 2181.54 Q106.594 2178.88 104.696 2177.47 Q102.821 2176.03 99.3021 2176.03 Q97.3808 2176.03 95.1817 2176.45 Q92.9826 2176.86 90.3438 2177.74 L90.3438 2173.58 Q93.0058 2172.84 95.3206 2172.47 Q97.6585 2172.1 99.7187 2172.1 Q105.043 2172.1 108.145 2174.53 Q111.246 2176.93 111.246 2181.05 Q111.246 2183.92 109.603 2185.91 Q107.959 2187.88 104.927 2188.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M51.9412 1865.44 L81.617 1865.44 L81.617 1869.38 L51.9412 1869.38 L51.9412 1865.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M95.7373 1878.34 L112.057 1878.34 L112.057 1882.27 L90.1123 1882.27 L90.1123 1878.34 Q92.7743 1875.58 97.3576 1870.95 Q101.964 1866.3 103.145 1864.96 Q105.39 1862.43 106.27 1860.7 Q107.172 1858.94 107.172 1857.25 Q107.172 1854.49 105.228 1852.76 Q103.307 1851.02 100.205 1851.02 Q98.0058 1851.02 95.5521 1851.78 Q93.1215 1852.55 90.3438 1854.1 L90.3438 1849.38 Q93.1678 1848.24 95.6215 1847.66 Q98.0752 1847.09 100.112 1847.09 Q105.483 1847.09 108.677 1849.77 Q111.871 1852.46 111.871 1856.95 Q111.871 1859.08 111.061 1861 Q110.274 1862.9 108.168 1865.49 Q107.589 1866.16 104.487 1869.38 Q101.385 1872.57 95.7373 1878.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M51.5708 1540.43 L81.2466 1540.43 L81.2466 1544.37 L51.5708 1544.37 L51.5708 1540.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M92.1493 1553.33 L99.7882 1553.33 L99.7882 1526.96 L91.478 1528.63 L91.478 1524.37 L99.7419 1522.7 L104.418 1522.7 L104.418 1553.33 L112.057 1553.33 L112.057 1557.26 L92.1493 1557.26 L92.1493 1553.33 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M100.112 1200.77 Q96.5011 1200.77 94.6724 1204.33 Q92.8669 1207.88 92.8669 1215.01 Q92.8669 1222.11 94.6724 1225.68 Q96.5011 1229.22 100.112 1229.22 Q103.746 1229.22 105.552 1225.68 Q107.381 1222.11 107.381 1215.01 Q107.381 1207.88 105.552 1204.33 Q103.746 1200.77 100.112 1200.77 M100.112 1197.07 Q105.922 1197.07 108.978 1201.67 Q112.057 1206.26 112.057 1215.01 Q112.057 1223.73 108.978 1228.34 Q105.922 1232.92 100.112 1232.92 Q94.3021 1232.92 91.2234 1228.34 Q88.1679 1223.73 88.1679 1215.01 Q88.1679 1206.26 91.2234 1201.67 Q94.3021 1197.07 100.112 1197.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M92.1493 903.307 L99.7882 903.307 L99.7882 876.941 L91.478 878.608 L91.478 874.348 L99.7419 872.682 L104.418 872.682 L104.418 903.307 L112.057 903.307 L112.057 907.242 L92.1493 907.242 L92.1493 903.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M95.7373 578.297 L112.057 578.297 L112.057 582.232 L90.1123 582.232 L90.1123 578.297 Q92.7743 575.542 97.3576 570.913 Q101.964 566.26 103.145 564.917 Q105.39 562.394 106.27 560.658 Q107.172 558.899 107.172 557.209 Q107.172 554.454 105.228 552.718 Q103.307 550.982 100.205 550.982 Q98.0058 550.982 95.5521 551.746 Q93.1215 552.51 90.3438 554.061 L90.3438 549.339 Q93.1678 548.204 95.6215 547.626 Q98.0752 547.047 100.112 547.047 Q105.483 547.047 108.677 549.732 Q111.871 552.417 111.871 556.908 Q111.871 559.038 111.061 560.959 Q110.274 562.857 108.168 565.45 Q107.589 566.121 104.487 569.339 Q101.385 572.533 95.7373 578.297 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M104.927 238.588 Q108.283 239.306 110.158 241.574 Q112.057 243.843 112.057 247.176 Q112.057 252.292 108.538 255.093 Q105.02 257.894 98.5382 257.894 Q96.3623 257.894 94.0475 257.454 Q91.7558 257.037 89.3021 256.181 L89.3021 251.667 Q91.2465 252.801 93.5613 253.38 Q95.8761 253.959 98.3993 253.959 Q102.797 253.959 105.089 252.222 Q107.404 250.486 107.404 247.176 Q107.404 244.121 105.251 242.408 Q103.121 240.672 99.3021 240.672 L95.2743 240.672 L95.2743 236.829 L99.4872 236.829 Q102.936 236.829 104.765 235.463 Q106.594 234.074 106.594 231.482 Q106.594 228.82 104.696 227.408 Q102.821 225.973 99.3021 225.973 Q97.3808 225.973 95.1817 226.389 Q92.9826 226.806 90.3438 227.686 L90.3438 223.519 Q93.0058 222.778 95.3206 222.408 Q97.6585 222.037 99.7187 222.037 Q105.043 222.037 108.145 224.468 Q111.246 226.875 111.246 230.996 Q111.246 233.866 109.603 235.857 Q107.959 237.824 104.927 238.588 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="786.163,1548.29 790.416,1532.67 794.146,1517.27 797.343,1502.11 799.998,1487.23 802.105,1472.68 803.658,1458.49 804.655,1444.69 805.092,1431.32 804.968,1418.42 804.284,1406 803.042,1394.11 801.243,1382.78 798.894,1372.03 795.999,1361.89 792.567,1352.39 788.605,1343.54 784.123,1335.38 779.133,1327.92 773.648,1321.18 767.68,1315.17 761.245,1309.92 754.359,1305.43 747.039,1301.73 739.303,1298.8 731.171,1296.67 722.663,1295.34 713.801,1294.81 704.606,1295.08 695.101,1296.16 685.311,1298.03 675.259,1300.71 664.971,1304.16 654.473,1308.4 643.79,1313.41 632.95,1319.18 621.979,1325.68 610.906,1332.92 599.757,1340.86 588.56,1349.49 577.344,1358.78 566.137,1368.72 554.966,1379.28 543.86,1390.43 532.846,1402.14 521.951,1414.39 511.204,1427.15 500.63,1440.37 490.257,1454.04 480.11,1468.11 470.214,1482.55 460.595,1497.32 451.276,1512.39 442.281,1527.72 433.632,1543.28 425.351,1559.01 417.458,1574.89 409.973,1590.87 402.916,1606.91 396.302,1622.98 390.15,1639.02 384.474,1655.02 379.289,1670.91 374.607,1686.67 370.441,1702.25 366.8,1717.62 363.695,1732.73 361.131,1747.55 359.117,1762.04 357.657,1776.17 356.754,1789.9 356.411,1803.2 356.629,1816.02 357.407,1828.35 358.744,1840.15 360.635,1851.38 363.076,1862.03 366.062,1872.07 369.584,1881.46 373.635,1890.19 378.202,1898.24 383.276,1905.58 388.844,1912.2 394.892,1918.08 401.404,1923.2 408.364,1927.56 415.755,1931.14 423.559,1933.93 431.755,1935.92 440.324,1937.12 449.244,1937.52 458.493,1937.11 468.047,1935.9 477.883,1933.89 487.977,1931.08 498.302,1927.49 508.833,1923.12 519.544,1917.99 530.408,1912.1 541.398,1905.47 552.486,1898.11 563.645,1890.05 574.847,1881.31 586.063,1871.91 597.266,1861.86 608.428,1851.2 619.521,1839.95 630.517,1828.15 641.389,1815.81 652.109,1802.98 662.65,1789.68 672.987,1775.94 683.094,1761.81 692.945,1747.31 702.515,1732.48 711.781,1717.36 720.719,1701.99 729.308,1686.41 737.526,1670.65 745.351,1654.75 752.765,1638.76 759.75,1622.71 766.286,1606.64 772.36,1590.6 777.954,1574.62 783.055,1558.74 787.651,1543.01 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2268.85,906.823 2276.2,896.762 2281.84,887.035 2285.76,877.664 2287.95,868.675 2288.4,860.089 2287.12,851.929 2284.1,844.213 2279.35,836.963 2272.89,830.195 2264.73,823.926 2254.89,818.173 2243.4,812.95 2230.29,808.27 2215.58,804.144 2199.31,800.582 2181.53,797.595 2162.29,795.188 2141.61,793.369 2119.57,792.142 2096.21,791.509 2071.6,791.472 2045.79,792.033 2018.85,793.188 1990.84,794.935 1961.84,797.27 1931.92,800.187 1901.15,803.679 1869.62,807.737 1837.39,812.35 1804.55,817.508 1771.18,823.197 1737.37,829.403 1703.2,836.111 1668.76,843.304 1634.12,850.964 1599.39,859.071 1564.63,867.606 1529.95,876.547 1495.43,885.872 1461.15,895.558 1427.21,905.58 1393.67,915.913 1360.64,926.531 1328.19,937.408 1296.4,948.517 1265.35,959.83 1235.12,971.319 1205.79,982.954 1177.43,994.708 1150.1,1006.55 1123.88,1018.45 1098.84,1030.38 1075.03,1042.31 1052.52,1054.21 1031.35,1066.05 1011.6,1077.79 993.294,1089.42 976.49,1100.91 961.228,1112.21 947.545,1123.31 935.476,1134.17 925.052,1144.78 916.297,1155.1 909.235,1165.11 903.882,1174.78 900.252,1184.08 898.355,1193.01 898.194,1201.52 899.771,1209.61 903.081,1217.25 908.116,1224.42 914.863,1231.1 923.306,1237.29 933.424,1242.95 945.191,1248.08 958.578,1252.67 973.551,1256.7 990.073,1260.17 1008.1,1263.06 1027.59,1265.37 1048.5,1267.09 1070.77,1268.22 1094.34,1268.75 1119.16,1268.68 1145.17,1268.02 1172.29,1266.77 1200.47,1264.92 1229.63,1262.49 1259.7,1259.47 1290.6,1255.89 1322.26,1251.73 1354.59,1247.03 1387.53,1241.78 1420.97,1236 1454.85,1229.71 1489.07,1222.92 1523.56,1215.65 1558.21,1207.91 1592.96,1199.73 1627.71,1191.13 1662.37,1182.12 1696.85,1172.73 1731.08,1162.99 1764.96,1152.91 1798.42,1142.53 1831.36,1131.87 1863.71,1120.95 1895.38,1109.8 1926.29,1098.46 1956.38,1086.94 1985.55,1075.29 2013.75,1063.52 2040.89,1051.66 2066.91,1039.75 2091.75,1027.82 2115.35,1015.9 2137.64,1004.01 2158.57,992.18 2178.08,980.449 2196.14,968.843 2212.69,957.389 2227.69,946.118 2241.1,935.056 2252.89,924.232 2263.04,913.673 2271.51,903.404 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.02,1735.77 1738.68,1715.77 1730.02,1695.57 1720.05,1675.23 1708.8,1654.78 1696.31,1634.28 1682.6,1613.79 1667.7,1593.35 1651.66,1573.02 1634.5,1552.84 1616.29,1532.87 1597.06,1513.15 1576.85,1493.74 1555.73,1474.69 1533.74,1456.03 1510.94,1437.83 1487.38,1420.11 1463.12,1402.94 1438.23,1386.35 1412.77,1370.38 1386.8,1355.07 1360.38,1340.46 1333.59,1326.6 1306.48,1313.5 1279.12,1301.2 1251.6,1289.75 1223.96,1279.15 1196.29,1269.45 1168.65,1260.66 1141.11,1252.81 1113.74,1245.92 1086.6,1239.99 1059.77,1235.06 1033.32,1231.13 1007.3,1228.21 981.787,1226.3 956.841,1225.42 932.525,1225.56 908.9,1226.73 886.024,1228.92 863.955,1232.12 842.748,1236.34 822.456,1241.55 803.13,1247.74 784.818,1254.91 767.566,1263.02 751.417,1272.07 736.411,1282.02 722.585,1292.86 709.976,1304.55 698.613,1317.07 688.525,1330.39 679.738,1344.47 672.273,1359.27 666.149,1374.77 661.382,1390.92 657.983,1407.68 655.961,1425 655.321,1442.86 656.065,1461.19 658.19,1479.96 661.691,1499.12 666.56,1518.62 672.784,1538.42 680.348,1558.45 689.233,1578.68 699.417,1599.05 710.874,1619.51 723.576,1640 737.49,1660.49 752.583,1680.92 768.816,1701.23 786.149,1721.38 804.538,1741.31 823.938,1760.97 844.3,1780.33 865.573,1799.32 887.704,1817.9 910.638,1836.03 934.317,1853.65 958.682,1870.73 983.673,1887.22 1009.23,1903.08 1035.28,1918.27 1061.76,1932.76 1088.62,1946.5 1115.77,1959.47 1143.16,1971.62 1170.71,1982.93 1198.35,1993.38 1226.03,2002.93 1253.66,2011.56 1281.17,2019.26 1308.51,2025.99 1335.6,2031.74 1362.37,2036.51 1388.75,2040.27 1414.69,2043.02 1440.11,2044.76 1464.96,2045.47 1489.16,2045.15 1512.66,2043.81 1535.41,2041.45 1557.34,2038.08 1578.39,2033.7 1598.53,2028.32 1617.69,2021.96 1635.82,2014.63 1652.89,2006.36 1668.85,1997.16 1683.66,1987.06 1697.28,1976.07 1709.69,1964.24 1720.84,1951.58 1730.71,1938.13 1739.27,1923.93 1746.52,1909.01 1752.41,1893.4 1756.95,1877.14 1760.12,1860.29 1761.91,1842.87 1762.31,1824.93 1761.34,1806.52 1758.98,1787.68 1755.25,1768.46 1750.15,1748.91 1743.7,1729.07 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.932,1366.88 809.051,1362.15 810.756,1357.52 812.041,1353 812.904,1348.61 813.343,1344.34 813.356,1340.22 812.944,1336.25 812.108,1332.45 810.849,1328.82 809.17,1325.37 807.077,1322.12 804.574,1319.06 801.668,1316.21 798.365,1313.58 794.674,1311.16 790.604,1308.97 786.166,1307.02 781.37,1305.3 776.229,1303.82 770.755,1302.59 764.962,1301.6 758.864,1300.87 752.477,1300.39 745.817,1300.16 738.9,1300.19 731.743,1300.47 724.366,1301 716.785,1301.79 709.02,1302.82 701.09,1304.11 693.015,1305.64 684.816,1307.4 676.513,1309.41 668.126,1311.64 659.676,1314.1 651.185,1316.78 642.674,1319.68 634.164,1322.78 625.677,1326.07 617.233,1329.56 608.854,1333.22 600.56,1337.06 592.373,1341.06 584.313,1345.21 576.4,1349.5 568.653,1353.93 561.093,1358.47 553.737,1363.12 546.605,1367.87 539.715,1372.71 533.083,1377.61 526.725,1382.58 520.659,1387.6 514.899,1392.65 509.459,1397.72 504.354,1402.81 499.595,1407.89 495.195,1412.96 491.165,1417.99 487.514,1422.99 484.253,1427.94 481.388,1432.82 478.928,1437.62 476.879,1442.34 475.244,1446.95 474.03,1451.45 473.238,1455.83 472.871,1460.07 472.929,1464.16 473.412,1468.1 474.32,1471.88 475.65,1475.48 477.398,1478.89 479.561,1482.12 482.132,1485.14 485.106,1487.95 488.474,1490.55 492.23,1492.93 496.362,1495.08 500.861,1496.99 505.716,1498.67 510.914,1500.11 516.443,1501.3 522.288,1502.24 528.436,1502.93 534.87,1503.37 541.574,1503.56 548.533,1503.49 555.728,1503.17 563.141,1502.59 570.754,1501.76 578.548,1500.68 586.504,1499.36 594.601,1497.79 602.819,1495.98 611.138,1493.94 619.537,1491.67 627.995,1489.17 636.491,1486.45 645.003,1483.52 653.51,1480.39 661.992,1477.06 670.427,1473.55 678.793,1469.85 687.07,1465.98 695.237,1461.96 703.274,1457.78 711.161,1453.47 718.877,1449.02 726.404,1444.46 733.724,1439.79 740.816,1435.03 747.665,1430.18 754.252,1425.26 760.561,1420.29 766.577,1415.26 772.284,1410.21 777.668,1405.13 782.717,1400.05 787.416,1394.96 791.754,1389.9 795.721,1384.87 799.307,1379.88 802.502,1374.94 805.299,1370.07 807.691,1365.28 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1980.15,611.617 1983.01,584.925 1984.08,558.44 1983.35,532.23 1980.82,506.36 1976.5,480.894 1970.4,455.897 1962.54,431.43 1952.93,407.554 1941.59,384.33 1928.57,361.816 1913.89,340.067 1897.58,319.138 1879.69,299.082 1860.26,279.948 1839.34,261.785 1816.98,244.638 1793.24,228.549 1768.18,213.559 1741.86,199.705 1714.35,187.022 1685.71,175.542 1656.02,165.293 1625.35,156.301 1593.78,148.588 1561.38,142.174 1528.24,137.075 1494.44,133.303 1460.07,130.869 1425.21,129.777 1389.94,130.031 1354.36,131.63 1318.56,134.57 1282.62,138.844 1246.63,144.441 1210.69,151.347 1174.88,159.544 1139.29,169.013 1104.02,179.73 1069.15,191.668 1034.77,204.796 1000.96,219.083 967.805,234.492 935.394,250.985 903.805,268.521 873.117,287.056 843.407,306.544 814.749,326.935 787.214,348.18 760.872,370.224 735.788,393.013 712.026,416.489 689.643,440.595 668.697,465.269 649.24,490.451 631.319,516.077 614.981,542.083 600.266,568.404 587.211,594.975 575.848,621.728 566.205,648.598 558.308,675.517 552.175,702.417 547.822,729.232 545.26,755.894 544.495,782.336 545.529,808.494 548.36,834.301 552.98,859.692 559.378,884.605 567.538,908.977 577.439,932.747 589.057,955.855 602.363,978.245 617.324,999.859 633.901,1020.64 652.054,1040.55 671.738,1059.52 692.902,1077.52 715.495,1094.49 739.459,1110.4 764.735,1125.2 791.26,1138.86 818.967,1151.34 847.788,1162.62 877.649,1172.66 908.476,1181.43 940.193,1188.93 972.72,1195.12 1005.98,1200 1039.88,1203.55 1074.34,1205.76 1109.28,1206.62 1144.6,1206.14 1180.23,1204.32 1216.06,1201.15 1252.01,1196.66 1288,1190.84 1323.92,1183.71 1359.7,1175.3 1395.24,1165.62 1430.45,1154.7 1465.25,1142.56 1499.54,1129.23 1533.25,1114.75 1566.28,1099.16 1598.56,1082.49 1630,1064.78 1660.53,1046.08 1690.07,1026.44 1718.54,1005.9 1745.88,984.517 1772.01,962.344 1796.88,939.435 1820.42,915.848 1842.56,891.643 1863.26,866.879 1882.46,841.618 1900.12,815.924 1916.19,789.86 1930.63,763.492 1943.4,736.886 1954.47,710.108 1963.82,683.226 1971.43,656.306 1977.26,629.415 1981.31,602.621 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1982.85,1240.63 1983.21,1236.81 1982.59,1233.04 1980.98,1229.32 1978.39,1225.66 1974.82,1222.06 1970.29,1218.55 1964.81,1215.12 1958.38,1211.79 1951.03,1208.57 1942.78,1205.45 1933.64,1202.46 1923.64,1199.59 1912.8,1196.86 1901.15,1194.27 1888.72,1191.82 1875.55,1189.54 1861.65,1187.41 1847.07,1185.44 1831.85,1183.65 1816.01,1182.03 1799.62,1180.59 1782.69,1179.33 1765.28,1178.25 1747.42,1177.37 1729.17,1176.67 1710.57,1176.17 1691.67,1175.86 1672.5,1175.74 1653.13,1175.82 1633.6,1176.09 1613.95,1176.56 1594.24,1177.21 1574.52,1178.06 1554.83,1179.1 1535.23,1180.32 1515.76,1181.73 1496.47,1183.31 1477.42,1185.07 1458.64,1187 1440.19,1189.1 1422.11,1191.35 1404.44,1193.77 1387.24,1196.33 1370.54,1199.03 1354.38,1201.87 1338.81,1204.84 1323.86,1207.93 1309.58,1211.14 1295.99,1214.45 1283.14,1217.86 1271.05,1221.36 1259.75,1224.93 1249.28,1228.58 1239.65,1232.29 1230.9,1236.06 1223.04,1239.86 1216.1,1243.7 1210.08,1247.57 1205.01,1251.45 1200.91,1255.33 1197.77,1259.22 1195.61,1263.08 1194.43,1266.92 1194.23,1270.73 1195.02,1274.5 1196.8,1278.21 1199.56,1281.86 1203.28,1285.44 1207.98,1288.94 1213.62,1292.35 1220.2,1295.66 1227.7,1298.87 1236.11,1301.97 1245.39,1304.94 1255.53,1307.79 1266.51,1310.5 1278.29,1313.06 1290.85,1315.48 1304.15,1317.74 1318.17,1319.84 1332.85,1321.78 1348.18,1323.54 1364.11,1325.13 1380.61,1326.55 1397.62,1327.77 1415.11,1328.82 1433.03,1329.67 1451.34,1330.33 1470,1330.8 1488.95,1331.08 1508.15,1331.17 1527.55,1331.06 1547.11,1330.75 1566.77,1330.25 1586.48,1329.56 1606.2,1328.68 1625.88,1327.62 1645.46,1326.36 1664.91,1324.93 1684.16,1323.31 1703.17,1321.53 1721.9,1319.57 1740.29,1317.44 1758.3,1315.16 1775.9,1312.72 1793.02,1310.13 1809.63,1307.41 1825.69,1304.54 1841.16,1301.55 1855.99,1298.44 1870.16,1295.22 1883.63,1291.89 1896.36,1288.47 1908.32,1284.96 1919.48,1281.37 1929.81,1277.71 1939.29,1273.99 1947.89,1270.21 1955.6,1266.4 1962.39,1262.56 1968.25,1258.69 1973.15,1254.81 1977.1,1250.92 1980.07,1247.04 1982.07,1243.18 1983.09,1239.34 "/>
-<polyline clip-path="url(#clip252)" style="stroke:#dc143c; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2282.61,1056.92 2287.78,1005.07 2290.36,953.582 2290.34,902.599 2287.72,852.246 2282.51,802.647 2274.72,753.926 2264.37,706.206 2251.49,659.606 2236.1,614.242 2218.25,570.227 2197.98,527.672 2175.34,486.683 2150.4,447.362 2123.2,409.808 2093.82,374.115 2062.33,340.371 2028.82,308.661 1993.35,279.064 1956.04,251.655 1916.95,226.502 1876.21,203.667 1833.9,183.208 1790.13,165.176 1745.01,149.616 1698.65,136.567 1651.18,126.062 1602.7,118.126 1553.34,112.779 1503.23,110.036 1452.49,109.902 1401.24,112.379 1349.61,117.459 1297.73,125.131 1245.74,135.375 1193.76,148.165 1141.92,163.469 1090.35,181.25 1039.19,201.463 988.548,224.057 938.564,248.976 889.359,276.157 841.057,305.533 793.778,337.03 747.641,370.569 702.761,406.068 659.249,443.435 617.216,482.58 576.765,523.403 537.999,565.802 501.013,609.673 465.901,654.904 432.749,701.383 401.642,748.994 372.656,797.617 345.865,847.132 321.334,897.414 299.126,948.338 279.296,999.776 261.893,1051.6 246.962,1103.68 234.538,1155.89 224.654,1208.09 217.335,1260.16 212.597,1311.96 210.454,1363.37 210.91,1414.26 213.965,1464.49 219.61,1513.95 227.832,1562.51 238.611,1610.05 251.918,1656.45 267.721,1701.6 285.981,1745.37 306.652,1787.67 329.681,1828.39 355.013,1867.42 382.582,1904.67 412.321,1940.04 444.155,1973.45 478.004,2004.81 513.784,2034.04 551.406,2061.07 590.775,2085.84 631.792,2108.28 674.357,2128.34 718.361,2145.95 763.695,2161.09 810.246,2173.72 857.898,2183.79 906.531,2191.3 956.023,2196.2 1006.25,2198.51 1057.09,2198.2 1108.41,2195.29 1160.09,2189.77 1211.99,2181.67 1263.99,2170.99 1315.96,2157.78 1367.76,2142.05 1419.27,2123.86 1470.36,2103.24 1520.9,2080.25 1570.76,2054.95 1619.82,2027.4 1667.96,1997.66 1715.05,1965.81 1760.99,1931.94 1805.65,1896.12 1848.92,1858.44 1890.69,1819.01 1930.86,1777.92 1969.34,1735.26 2006.02,1691.15 2040.8,1645.71 2073.62,1599.03 2104.37,1551.24 2133,1502.46 2159.41,1452.8 2183.56,1402.41 2205.37,1351.39 2224.79,1299.87 2241.78,1248 2256.29,1195.89 2268.29,1143.67 2277.75,1091.48 2284.63,1039.45 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="406.43,2261.17 406.43,47.2441 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="838.578,2261.17 838.578,47.2441 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1270.73,2261.17 1270.73,47.2441 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1702.88,2261.17 1702.88,47.2441 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2135.02,2261.17 2135.02,47.2441 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2261.17 2352.76,2261.17 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="406.43,2261.17 406.43,2242.27 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="838.578,2261.17 838.578,2242.27 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1270.73,2261.17 1270.73,2242.27 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1702.88,2261.17 1702.88,2242.27 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.02,2261.17 2135.02,2242.27 "/>
+<path clip-path="url(#clip070)" d="M376.187 2306.74 L405.862 2306.74 L405.862 2310.67 L376.187 2310.67 L376.187 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M416.765 2319.63 L424.404 2319.63 L424.404 2293.27 L416.094 2294.93 L416.094 2290.67 L424.358 2289.01 L429.034 2289.01 L429.034 2319.63 L436.672 2319.63 L436.672 2323.57 L416.765 2323.57 L416.765 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M838.578 2292.09 Q834.967 2292.09 833.138 2295.65 Q831.333 2299.19 831.333 2306.32 Q831.333 2313.43 833.138 2316.99 Q834.967 2320.54 838.578 2320.54 Q842.212 2320.54 844.018 2316.99 Q845.847 2313.43 845.847 2306.32 Q845.847 2299.19 844.018 2295.65 Q842.212 2292.09 838.578 2292.09 M838.578 2288.38 Q844.388 2288.38 847.444 2292.99 Q850.523 2297.57 850.523 2306.32 Q850.523 2315.05 847.444 2319.66 Q844.388 2324.24 838.578 2324.24 Q832.768 2324.24 829.689 2319.66 Q826.634 2315.05 826.634 2306.32 Q826.634 2297.57 829.689 2292.99 Q832.768 2288.38 838.578 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1261.11 2319.63 L1268.75 2319.63 L1268.75 2293.27 L1260.44 2294.93 L1260.44 2290.67 L1268.7 2289.01 L1273.38 2289.01 L1273.38 2319.63 L1281.02 2319.63 L1281.02 2323.57 L1261.11 2323.57 L1261.11 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M1697.53 2319.63 L1713.85 2319.63 L1713.85 2323.57 L1691.9 2323.57 L1691.9 2319.63 Q1694.57 2316.88 1699.15 2312.25 Q1703.75 2307.6 1704.94 2306.25 Q1707.18 2303.73 1708.06 2301.99 Q1708.96 2300.23 1708.96 2298.54 Q1708.96 2295.79 1707.02 2294.05 Q1705.1 2292.32 1702 2292.32 Q1699.8 2292.32 1697.34 2293.08 Q1694.91 2293.85 1692.13 2295.4 L1692.13 2290.67 Q1694.96 2289.54 1697.41 2288.96 Q1699.87 2288.38 1701.9 2288.38 Q1707.27 2288.38 1710.47 2291.07 Q1713.66 2293.75 1713.66 2298.24 Q1713.66 2300.37 1712.85 2302.29 Q1712.07 2304.19 1709.96 2306.79 Q1709.38 2307.46 1706.28 2310.67 Q1703.18 2313.87 1697.53 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M2139.27 2304.93 Q2142.63 2305.65 2144.5 2307.92 Q2146.4 2310.19 2146.4 2313.52 Q2146.4 2318.64 2142.88 2321.44 Q2139.36 2324.24 2132.88 2324.24 Q2130.71 2324.24 2128.39 2323.8 Q2126.1 2323.38 2123.65 2322.53 L2123.65 2318.01 Q2125.59 2319.15 2127.91 2319.73 Q2130.22 2320.3 2132.74 2320.3 Q2137.14 2320.3 2139.43 2318.57 Q2141.75 2316.83 2141.75 2313.52 Q2141.75 2310.47 2139.6 2308.75 Q2137.47 2307.02 2133.65 2307.02 L2129.62 2307.02 L2129.62 2303.17 L2133.83 2303.17 Q2137.28 2303.17 2139.11 2301.81 Q2140.94 2300.42 2140.94 2297.83 Q2140.94 2295.17 2139.04 2293.75 Q2137.17 2292.32 2133.65 2292.32 Q2131.73 2292.32 2129.53 2292.73 Q2127.33 2293.15 2124.69 2294.03 L2124.69 2289.86 Q2127.35 2289.12 2129.67 2288.75 Q2132 2288.38 2134.06 2288.38 Q2139.39 2288.38 2142.49 2290.81 Q2145.59 2293.22 2145.59 2297.34 Q2145.59 2300.21 2143.95 2302.2 Q2142.3 2304.17 2139.27 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,2190 2352.76,2190 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1864.99 2352.76,1864.99 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1539.98 2352.76,1539.98 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,1214.97 2352.76,1214.97 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,889.962 2352.76,889.962 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,564.952 2352.76,564.952 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="148.057,239.942 2352.76,239.942 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2261.17 148.057,47.2441 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,2190 166.954,2190 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1864.99 166.954,1864.99 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1539.98 166.954,1539.98 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,1214.97 166.954,1214.97 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,889.962 166.954,889.962 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,564.952 166.954,564.952 "/>
+<polyline clip-path="url(#clip070)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="148.057,239.942 166.954,239.942 "/>
+<path clip-path="url(#clip070)" d="M50.9921 2190.45 L80.6679 2190.45 L80.6679 2194.39 L50.9921 2194.39 L50.9921 2190.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M104.927 2188.65 Q108.283 2189.36 110.158 2191.63 Q112.057 2193.9 112.057 2197.23 Q112.057 2202.35 108.538 2205.15 Q105.02 2207.95 98.5382 2207.95 Q96.3623 2207.95 94.0475 2207.51 Q91.7558 2207.1 89.3021 2206.24 L89.3021 2201.72 Q91.2465 2202.86 93.5613 2203.44 Q95.8761 2204.02 98.3993 2204.02 Q102.797 2204.02 105.089 2202.28 Q107.404 2200.54 107.404 2197.23 Q107.404 2194.18 105.251 2192.47 Q103.121 2190.73 99.3021 2190.73 L95.2743 2190.73 L95.2743 2186.89 L99.4872 2186.89 Q102.936 2186.89 104.765 2185.52 Q106.594 2184.13 106.594 2181.54 Q106.594 2178.88 104.696 2177.47 Q102.821 2176.03 99.3021 2176.03 Q97.3808 2176.03 95.1817 2176.45 Q92.9826 2176.86 90.3438 2177.74 L90.3438 2173.58 Q93.0058 2172.84 95.3206 2172.47 Q97.6585 2172.1 99.7187 2172.1 Q105.043 2172.1 108.145 2174.53 Q111.246 2176.93 111.246 2181.05 Q111.246 2183.92 109.603 2185.91 Q107.959 2187.88 104.927 2188.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M51.9412 1865.44 L81.617 1865.44 L81.617 1869.38 L51.9412 1869.38 L51.9412 1865.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M95.7373 1878.34 L112.057 1878.34 L112.057 1882.27 L90.1123 1882.27 L90.1123 1878.34 Q92.7743 1875.58 97.3576 1870.95 Q101.964 1866.3 103.145 1864.96 Q105.39 1862.43 106.27 1860.7 Q107.172 1858.94 107.172 1857.25 Q107.172 1854.49 105.228 1852.76 Q103.307 1851.02 100.205 1851.02 Q98.0058 1851.02 95.5521 1851.78 Q93.1215 1852.55 90.3438 1854.1 L90.3438 1849.38 Q93.1678 1848.24 95.6215 1847.66 Q98.0752 1847.09 100.112 1847.09 Q105.483 1847.09 108.677 1849.77 Q111.871 1852.46 111.871 1856.95 Q111.871 1859.08 111.061 1861 Q110.274 1862.9 108.168 1865.49 Q107.589 1866.16 104.487 1869.38 Q101.385 1872.57 95.7373 1878.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M51.5708 1540.43 L81.2466 1540.43 L81.2466 1544.37 L51.5708 1544.37 L51.5708 1540.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M92.1493 1553.33 L99.7882 1553.33 L99.7882 1526.96 L91.478 1528.63 L91.478 1524.37 L99.7419 1522.7 L104.418 1522.7 L104.418 1553.33 L112.057 1553.33 L112.057 1557.26 L92.1493 1557.26 L92.1493 1553.33 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M100.112 1200.77 Q96.5011 1200.77 94.6724 1204.33 Q92.8669 1207.88 92.8669 1215.01 Q92.8669 1222.11 94.6724 1225.68 Q96.5011 1229.22 100.112 1229.22 Q103.746 1229.22 105.552 1225.68 Q107.381 1222.11 107.381 1215.01 Q107.381 1207.88 105.552 1204.33 Q103.746 1200.77 100.112 1200.77 M100.112 1197.07 Q105.922 1197.07 108.978 1201.67 Q112.057 1206.26 112.057 1215.01 Q112.057 1223.73 108.978 1228.34 Q105.922 1232.92 100.112 1232.92 Q94.3021 1232.92 91.2234 1228.34 Q88.1679 1223.73 88.1679 1215.01 Q88.1679 1206.26 91.2234 1201.67 Q94.3021 1197.07 100.112 1197.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M92.1493 903.307 L99.7882 903.307 L99.7882 876.941 L91.478 878.608 L91.478 874.348 L99.7419 872.682 L104.418 872.682 L104.418 903.307 L112.057 903.307 L112.057 907.242 L92.1493 907.242 L92.1493 903.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M95.7373 578.297 L112.057 578.297 L112.057 582.232 L90.1123 582.232 L90.1123 578.297 Q92.7743 575.542 97.3576 570.913 Q101.964 566.26 103.145 564.917 Q105.39 562.394 106.27 560.658 Q107.172 558.899 107.172 557.209 Q107.172 554.454 105.228 552.718 Q103.307 550.982 100.205 550.982 Q98.0058 550.982 95.5521 551.746 Q93.1215 552.51 90.3438 554.061 L90.3438 549.339 Q93.1678 548.204 95.6215 547.626 Q98.0752 547.047 100.112 547.047 Q105.483 547.047 108.677 549.732 Q111.871 552.417 111.871 556.908 Q111.871 559.038 111.061 560.959 Q110.274 562.857 108.168 565.45 Q107.589 566.121 104.487 569.339 Q101.385 572.533 95.7373 578.297 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip070)" d="M104.927 238.588 Q108.283 239.306 110.158 241.574 Q112.057 243.843 112.057 247.176 Q112.057 252.292 108.538 255.093 Q105.02 257.894 98.5382 257.894 Q96.3623 257.894 94.0475 257.454 Q91.7558 257.037 89.3021 256.181 L89.3021 251.667 Q91.2465 252.801 93.5613 253.38 Q95.8761 253.959 98.3993 253.959 Q102.797 253.959 105.089 252.222 Q107.404 250.486 107.404 247.176 Q107.404 244.121 105.251 242.408 Q103.121 240.672 99.3021 240.672 L95.2743 240.672 L95.2743 236.829 L99.4872 236.829 Q102.936 236.829 104.765 235.463 Q106.594 234.074 106.594 231.482 Q106.594 228.82 104.696 227.408 Q102.821 225.973 99.3021 225.973 Q97.3808 225.973 95.1817 226.389 Q92.9826 226.806 90.3438 227.686 L90.3438 223.519 Q93.0058 222.778 95.3206 222.408 Q97.6585 222.037 99.7187 222.037 Q105.043 222.037 108.145 224.468 Q111.246 226.875 111.246 230.996 Q111.246 233.866 109.603 235.857 Q107.959 237.824 104.927 238.588 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="786.163,1548.29 790.416,1532.67 794.146,1517.27 797.343,1502.11 799.998,1487.23 802.105,1472.68 803.658,1458.49 804.655,1444.69 805.092,1431.32 804.968,1418.42 804.284,1406 803.042,1394.11 801.243,1382.78 798.894,1372.03 795.999,1361.89 792.567,1352.39 788.605,1343.54 784.123,1335.38 779.133,1327.92 773.648,1321.18 767.68,1315.17 761.245,1309.92 754.359,1305.43 747.039,1301.73 739.303,1298.8 731.171,1296.67 722.663,1295.34 713.801,1294.81 704.606,1295.08 695.101,1296.16 685.311,1298.03 675.259,1300.71 664.971,1304.16 654.473,1308.4 643.79,1313.41 632.95,1319.18 621.979,1325.68 610.906,1332.92 599.757,1340.86 588.56,1349.49 577.344,1358.78 566.137,1368.72 554.966,1379.28 543.86,1390.43 532.846,1402.14 521.951,1414.39 511.204,1427.15 500.63,1440.37 490.257,1454.04 480.11,1468.11 470.214,1482.55 460.595,1497.32 451.276,1512.39 442.281,1527.72 433.632,1543.28 425.351,1559.01 417.458,1574.89 409.973,1590.87 402.916,1606.91 396.302,1622.98 390.15,1639.02 384.474,1655.02 379.289,1670.91 374.607,1686.67 370.441,1702.25 366.8,1717.62 363.695,1732.73 361.131,1747.55 359.117,1762.04 357.657,1776.17 356.754,1789.9 356.411,1803.2 356.629,1816.02 357.407,1828.35 358.744,1840.15 360.635,1851.38 363.076,1862.03 366.062,1872.07 369.584,1881.46 373.635,1890.19 378.202,1898.24 383.276,1905.58 388.844,1912.2 394.892,1918.08 401.404,1923.2 408.364,1927.56 415.755,1931.14 423.559,1933.93 431.755,1935.92 440.324,1937.12 449.244,1937.52 458.493,1937.11 468.047,1935.9 477.883,1933.89 487.977,1931.08 498.302,1927.49 508.833,1923.12 519.544,1917.99 530.408,1912.1 541.398,1905.47 552.486,1898.11 563.645,1890.05 574.847,1881.31 586.063,1871.91 597.266,1861.86 608.428,1851.2 619.521,1839.95 630.517,1828.15 641.389,1815.81 652.109,1802.98 662.65,1789.68 672.987,1775.94 683.094,1761.81 692.945,1747.31 702.515,1732.48 711.781,1717.36 720.719,1701.99 729.308,1686.41 737.526,1670.65 745.351,1654.75 752.765,1638.76 759.75,1622.71 766.286,1606.64 772.36,1590.6 777.954,1574.62 783.055,1558.74 787.651,1543.01 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2268.85,906.823 2276.2,896.762 2281.84,887.035 2285.76,877.664 2287.95,868.675 2288.4,860.089 2287.12,851.929 2284.1,844.213 2279.35,836.963 2272.89,830.195 2264.73,823.926 2254.89,818.173 2243.4,812.95 2230.29,808.27 2215.58,804.144 2199.31,800.582 2181.53,797.595 2162.29,795.188 2141.61,793.369 2119.57,792.142 2096.21,791.509 2071.6,791.472 2045.79,792.033 2018.85,793.188 1990.84,794.935 1961.84,797.27 1931.92,800.187 1901.15,803.679 1869.62,807.737 1837.39,812.35 1804.55,817.508 1771.18,823.197 1737.37,829.403 1703.2,836.111 1668.76,843.304 1634.12,850.964 1599.39,859.071 1564.63,867.606 1529.95,876.547 1495.43,885.872 1461.15,895.558 1427.21,905.58 1393.67,915.913 1360.64,926.531 1328.19,937.408 1296.4,948.517 1265.35,959.83 1235.12,971.319 1205.79,982.954 1177.43,994.708 1150.1,1006.55 1123.88,1018.45 1098.84,1030.38 1075.03,1042.31 1052.52,1054.21 1031.35,1066.05 1011.6,1077.79 993.294,1089.42 976.49,1100.91 961.228,1112.21 947.545,1123.31 935.476,1134.17 925.052,1144.78 916.297,1155.1 909.235,1165.11 903.882,1174.78 900.252,1184.08 898.355,1193.01 898.194,1201.52 899.771,1209.61 903.081,1217.25 908.116,1224.42 914.863,1231.1 923.306,1237.29 933.424,1242.95 945.191,1248.08 958.578,1252.67 973.551,1256.7 990.073,1260.17 1008.1,1263.06 1027.59,1265.37 1048.5,1267.09 1070.77,1268.22 1094.34,1268.75 1119.16,1268.68 1145.17,1268.02 1172.29,1266.77 1200.47,1264.92 1229.63,1262.49 1259.7,1259.47 1290.6,1255.89 1322.26,1251.73 1354.59,1247.03 1387.53,1241.78 1420.97,1236 1454.85,1229.71 1489.07,1222.92 1523.56,1215.65 1558.21,1207.91 1592.96,1199.73 1627.71,1191.13 1662.37,1182.12 1696.85,1172.73 1731.08,1162.99 1764.96,1152.91 1798.42,1142.53 1831.36,1131.87 1863.71,1120.95 1895.38,1109.8 1926.29,1098.46 1956.38,1086.94 1985.55,1075.29 2013.75,1063.52 2040.89,1051.66 2066.91,1039.75 2091.75,1027.82 2115.35,1015.9 2137.64,1004.01 2158.57,992.18 2178.08,980.449 2196.14,968.843 2212.69,957.389 2227.69,946.118 2241.1,935.056 2252.89,924.232 2263.04,913.673 2271.51,903.404 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1746.02,1735.77 1738.68,1715.77 1730.02,1695.57 1720.05,1675.23 1708.8,1654.78 1696.31,1634.28 1682.6,1613.79 1667.7,1593.35 1651.66,1573.02 1634.5,1552.84 1616.29,1532.87 1597.06,1513.15 1576.85,1493.74 1555.73,1474.69 1533.74,1456.03 1510.94,1437.83 1487.38,1420.11 1463.12,1402.94 1438.23,1386.35 1412.77,1370.38 1386.8,1355.07 1360.38,1340.46 1333.59,1326.6 1306.48,1313.5 1279.12,1301.2 1251.6,1289.75 1223.96,1279.15 1196.29,1269.45 1168.65,1260.66 1141.11,1252.81 1113.74,1245.92 1086.6,1239.99 1059.77,1235.06 1033.32,1231.13 1007.3,1228.21 981.787,1226.3 956.841,1225.42 932.525,1225.56 908.9,1226.73 886.024,1228.92 863.955,1232.12 842.748,1236.34 822.456,1241.55 803.13,1247.74 784.818,1254.91 767.566,1263.02 751.417,1272.07 736.411,1282.02 722.585,1292.86 709.976,1304.55 698.613,1317.07 688.525,1330.39 679.738,1344.47 672.273,1359.27 666.149,1374.77 661.382,1390.92 657.983,1407.68 655.961,1425 655.321,1442.86 656.065,1461.19 658.19,1479.96 661.691,1499.12 666.56,1518.62 672.784,1538.42 680.348,1558.45 689.233,1578.68 699.417,1599.05 710.874,1619.51 723.576,1640 737.49,1660.49 752.583,1680.92 768.816,1701.23 786.149,1721.38 804.538,1741.31 823.938,1760.97 844.3,1780.33 865.573,1799.32 887.704,1817.9 910.638,1836.03 934.317,1853.65 958.682,1870.73 983.673,1887.22 1009.23,1903.08 1035.28,1918.27 1061.76,1932.76 1088.62,1946.5 1115.77,1959.47 1143.16,1971.62 1170.71,1982.93 1198.35,1993.38 1226.03,2002.93 1253.66,2011.56 1281.17,2019.26 1308.51,2025.99 1335.6,2031.74 1362.37,2036.51 1388.75,2040.27 1414.69,2043.02 1440.11,2044.76 1464.96,2045.47 1489.16,2045.15 1512.66,2043.81 1535.41,2041.45 1557.34,2038.08 1578.39,2033.7 1598.53,2028.32 1617.69,2021.96 1635.82,2014.63 1652.89,2006.36 1668.85,1997.16 1683.66,1987.06 1697.28,1976.07 1709.69,1964.24 1720.84,1951.58 1730.71,1938.13 1739.27,1923.93 1746.52,1909.01 1752.41,1893.4 1756.95,1877.14 1760.12,1860.29 1761.91,1842.87 1762.31,1824.93 1761.34,1806.52 1758.98,1787.68 1755.25,1768.46 1750.15,1748.91 1743.7,1729.07 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.932,1366.88 809.051,1362.15 810.756,1357.52 812.041,1353 812.904,1348.61 813.343,1344.34 813.356,1340.22 812.944,1336.25 812.108,1332.45 810.849,1328.82 809.17,1325.37 807.077,1322.12 804.574,1319.06 801.668,1316.21 798.365,1313.58 794.674,1311.16 790.604,1308.97 786.166,1307.02 781.37,1305.3 776.229,1303.82 770.755,1302.59 764.962,1301.6 758.864,1300.87 752.477,1300.39 745.817,1300.16 738.9,1300.19 731.743,1300.47 724.366,1301 716.785,1301.79 709.02,1302.82 701.09,1304.11 693.015,1305.64 684.816,1307.4 676.513,1309.41 668.126,1311.64 659.676,1314.1 651.185,1316.78 642.674,1319.68 634.164,1322.78 625.677,1326.07 617.233,1329.56 608.854,1333.22 600.56,1337.06 592.373,1341.06 584.313,1345.21 576.4,1349.5 568.653,1353.93 561.093,1358.47 553.737,1363.12 546.605,1367.87 539.715,1372.71 533.083,1377.61 526.725,1382.58 520.659,1387.6 514.899,1392.65 509.459,1397.72 504.354,1402.81 499.595,1407.89 495.195,1412.96 491.165,1417.99 487.514,1422.99 484.253,1427.94 481.389,1432.82 478.928,1437.62 476.879,1442.34 475.244,1446.95 474.03,1451.45 473.238,1455.83 472.871,1460.07 472.929,1464.16 473.412,1468.1 474.32,1471.88 475.65,1475.48 477.398,1478.89 479.561,1482.12 482.132,1485.14 485.106,1487.95 488.474,1490.55 492.23,1492.93 496.362,1495.08 500.861,1496.99 505.716,1498.67 510.914,1500.11 516.443,1501.3 522.288,1502.24 528.436,1502.93 534.87,1503.37 541.574,1503.56 548.533,1503.49 555.728,1503.17 563.141,1502.59 570.754,1501.76 578.548,1500.68 586.504,1499.36 594.601,1497.79 602.819,1495.98 611.138,1493.94 619.537,1491.67 627.995,1489.17 636.491,1486.45 645.003,1483.52 653.51,1480.39 661.992,1477.06 670.427,1473.55 678.793,1469.85 687.07,1465.98 695.237,1461.96 703.274,1457.78 711.161,1453.47 718.877,1449.02 726.404,1444.46 733.724,1439.79 740.816,1435.03 747.665,1430.18 754.252,1425.26 760.561,1420.29 766.577,1415.26 772.284,1410.21 777.668,1405.13 782.717,1400.05 787.416,1394.96 791.754,1389.9 795.721,1384.87 799.307,1379.88 802.502,1374.94 805.299,1370.07 807.691,1365.28 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1980.15,611.617 1983.01,584.925 1984.08,558.44 1983.35,532.23 1980.82,506.36 1976.5,480.894 1970.4,455.897 1962.54,431.43 1952.93,407.554 1941.59,384.33 1928.57,361.816 1913.89,340.067 1897.58,319.138 1879.69,299.082 1860.26,279.948 1839.34,261.785 1816.98,244.638 1793.24,228.549 1768.18,213.559 1741.86,199.705 1714.35,187.022 1685.71,175.542 1656.02,165.293 1625.35,156.301 1593.78,148.588 1561.38,142.174 1528.24,137.075 1494.44,133.303 1460.07,130.869 1425.21,129.777 1389.94,130.031 1354.36,131.63 1318.56,134.57 1282.62,138.844 1246.63,144.441 1210.69,151.347 1174.88,159.544 1139.29,169.013 1104.02,179.73 1069.15,191.668 1034.77,204.796 1000.96,219.083 967.805,234.492 935.394,250.985 903.805,268.521 873.117,287.056 843.407,306.544 814.749,326.935 787.214,348.18 760.872,370.224 735.788,393.013 712.026,416.489 689.643,440.595 668.697,465.269 649.24,490.451 631.319,516.077 614.981,542.083 600.266,568.404 587.211,594.975 575.848,621.728 566.205,648.598 558.308,675.517 552.175,702.417 547.822,729.232 545.26,755.894 544.495,782.336 545.529,808.494 548.36,834.301 552.98,859.692 559.378,884.605 567.538,908.977 577.439,932.747 589.057,955.855 602.363,978.245 617.324,999.859 633.901,1020.64 652.054,1040.55 671.738,1059.52 692.902,1077.52 715.495,1094.49 739.459,1110.4 764.735,1125.2 791.26,1138.86 818.967,1151.34 847.788,1162.62 877.649,1172.66 908.476,1181.43 940.193,1188.93 972.72,1195.12 1005.98,1200 1039.88,1203.55 1074.34,1205.76 1109.28,1206.62 1144.6,1206.14 1180.23,1204.32 1216.06,1201.15 1252.01,1196.66 1288,1190.84 1323.92,1183.71 1359.7,1175.3 1395.24,1165.62 1430.45,1154.7 1465.25,1142.56 1499.54,1129.23 1533.25,1114.75 1566.28,1099.16 1598.56,1082.49 1630,1064.78 1660.53,1046.08 1690.07,1026.44 1718.54,1005.9 1745.88,984.517 1772.01,962.344 1796.88,939.435 1820.42,915.848 1842.56,891.643 1863.26,866.879 1882.46,841.618 1900.12,815.924 1916.19,789.86 1930.63,763.492 1943.4,736.886 1954.47,710.108 1963.82,683.226 1971.43,656.306 1977.26,629.415 1981.31,602.621 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1982.85,1240.63 1983.21,1236.81 1982.59,1233.04 1980.98,1229.32 1978.39,1225.66 1974.82,1222.06 1970.29,1218.55 1964.81,1215.12 1958.38,1211.79 1951.03,1208.57 1942.78,1205.45 1933.64,1202.46 1923.64,1199.59 1912.8,1196.86 1901.15,1194.27 1888.72,1191.82 1875.55,1189.54 1861.65,1187.41 1847.07,1185.44 1831.85,1183.65 1816.01,1182.03 1799.62,1180.59 1782.69,1179.33 1765.28,1178.25 1747.42,1177.37 1729.17,1176.67 1710.57,1176.17 1691.67,1175.86 1672.5,1175.74 1653.13,1175.82 1633.6,1176.09 1613.95,1176.56 1594.24,1177.21 1574.52,1178.06 1554.83,1179.1 1535.23,1180.32 1515.76,1181.73 1496.47,1183.31 1477.42,1185.07 1458.64,1187 1440.19,1189.1 1422.11,1191.35 1404.44,1193.77 1387.24,1196.33 1370.54,1199.03 1354.38,1201.87 1338.81,1204.84 1323.86,1207.93 1309.58,1211.14 1295.99,1214.45 1283.14,1217.86 1271.05,1221.36 1259.75,1224.93 1249.28,1228.58 1239.65,1232.29 1230.9,1236.06 1223.04,1239.86 1216.1,1243.7 1210.08,1247.57 1205.01,1251.45 1200.91,1255.33 1197.77,1259.22 1195.61,1263.08 1194.43,1266.92 1194.23,1270.73 1195.02,1274.5 1196.8,1278.21 1199.56,1281.86 1203.28,1285.44 1207.98,1288.94 1213.62,1292.35 1220.2,1295.66 1227.7,1298.87 1236.11,1301.97 1245.39,1304.94 1255.53,1307.79 1266.51,1310.5 1278.29,1313.06 1290.85,1315.48 1304.15,1317.74 1318.17,1319.84 1332.85,1321.78 1348.18,1323.54 1364.11,1325.13 1380.61,1326.55 1397.62,1327.77 1415.11,1328.82 1433.03,1329.67 1451.34,1330.33 1470,1330.8 1488.95,1331.08 1508.15,1331.17 1527.55,1331.06 1547.11,1330.75 1566.77,1330.25 1586.48,1329.56 1606.2,1328.68 1625.88,1327.62 1645.46,1326.36 1664.91,1324.93 1684.16,1323.31 1703.17,1321.53 1721.9,1319.57 1740.29,1317.44 1758.3,1315.16 1775.9,1312.72 1793.02,1310.13 1809.63,1307.41 1825.69,1304.54 1841.16,1301.55 1855.99,1298.44 1870.16,1295.22 1883.63,1291.89 1896.36,1288.47 1908.32,1284.96 1919.48,1281.37 1929.81,1277.71 1939.29,1273.99 1947.89,1270.21 1955.6,1266.4 1962.39,1262.56 1968.25,1258.69 1973.15,1254.81 1977.1,1250.92 1980.07,1247.04 1982.07,1243.18 1983.09,1239.34 "/>
+<polyline clip-path="url(#clip072)" style="stroke:#dc143c; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2282.61,1056.92 2287.78,1005.07 2290.36,953.582 2290.34,902.599 2287.72,852.246 2282.51,802.647 2274.72,753.926 2264.37,706.206 2251.49,659.606 2236.1,614.242 2218.25,570.227 2197.98,527.672 2175.34,486.683 2150.4,447.362 2123.2,409.808 2093.82,374.115 2062.33,340.371 2028.82,308.661 1993.35,279.064 1956.04,251.655 1916.95,226.502 1876.21,203.667 1833.9,183.208 1790.13,165.176 1745.01,149.616 1698.65,136.567 1651.18,126.062 1602.7,118.126 1553.34,112.779 1503.23,110.036 1452.49,109.902 1401.24,112.379 1349.61,117.459 1297.73,125.131 1245.74,135.375 1193.76,148.165 1141.92,163.469 1090.35,181.25 1039.19,201.463 988.548,224.057 938.564,248.976 889.359,276.157 841.057,305.533 793.778,337.03 747.641,370.569 702.761,406.068 659.249,443.435 617.216,482.58 576.765,523.403 537.999,565.802 501.013,609.673 465.901,654.904 432.749,701.383 401.642,748.994 372.656,797.617 345.865,847.132 321.334,897.414 299.126,948.338 279.296,999.776 261.893,1051.6 246.962,1103.68 234.538,1155.89 224.654,1208.09 217.335,1260.16 212.597,1311.96 210.454,1363.37 210.91,1414.26 213.965,1464.49 219.61,1513.95 227.832,1562.51 238.611,1610.05 251.918,1656.45 267.721,1701.6 285.981,1745.37 306.652,1787.67 329.681,1828.39 355.013,1867.42 382.582,1904.67 412.321,1940.04 444.155,1973.45 478.004,2004.81 513.784,2034.04 551.406,2061.07 590.775,2085.84 631.792,2108.28 674.357,2128.34 718.361,2145.95 763.695,2161.09 810.246,2173.72 857.898,2183.79 906.531,2191.3 956.023,2196.2 1006.25,2198.51 1057.09,2198.2 1108.41,2195.29 1160.09,2189.77 1211.99,2181.67 1263.99,2170.99 1315.96,2157.78 1367.76,2142.05 1419.27,2123.86 1470.36,2103.24 1520.9,2080.25 1570.76,2054.95 1619.82,2027.4 1667.96,1997.66 1715.05,1965.81 1760.99,1931.94 1805.65,1896.12 1848.92,1858.44 1890.69,1819.01 1930.86,1777.92 1969.34,1735.26 2006.02,1691.15 2040.8,1645.71 2073.62,1599.03 2104.37,1551.24 2133,1502.46 2159.41,1452.8 2183.56,1402.41 2205.37,1351.39 2224.79,1299.87 2241.78,1248 2256.29,1195.89 2268.29,1143.67 2277.75,1091.48 2284.63,1039.45 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/conic/min_ellipse/1b95f9f7.svg b/previews/PR3562/tutorials/conic/min_ellipse/ca1157bb.svg
similarity index 88%
rename from previews/PR3562/tutorials/conic/min_ellipse/1b95f9f7.svg
rename to previews/PR3562/tutorials/conic/min_ellipse/ca1157bb.svg
index b3415d36efa..d2849298075 100644
--- a/previews/PR3562/tutorials/conic/min_ellipse/1b95f9f7.svg
+++ b/previews/PR3562/tutorials/conic/min_ellipse/ca1157bb.svg
@@ -1,50 +1,50 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="600" viewBox="0 0 2400 2400">
 <defs>
-  <clipPath id="clip190">
+  <clipPath id="clip010">
     <rect x="0" y="0" width="2400" height="2400"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip190)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip010)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip191">
+  <clipPath id="clip011">
     <rect x="480" y="240" width="1681" height="1681"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip190)" d="M147.478 2261.17 L2352.76 2261.17 L2352.76 47.2441 L147.478 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip010)" d="M147.478 2261.17 L2352.76 2261.17 L2352.76 47.2441 L147.478 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip192">
+  <clipPath id="clip012">
     <rect x="147" y="47" width="2206" height="2215"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="263.754,2261.17 263.754,47.2441 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.11,2261.17 729.11,47.2441 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1194.47,2261.17 1194.47,47.2441 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1659.82,2261.17 1659.82,47.2441 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2125.18,2261.17 2125.18,47.2441 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2261.17 2352.76,2261.17 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="263.754,2261.17 263.754,2242.27 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.11,2261.17 729.11,2242.27 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1194.47,2261.17 1194.47,2242.27 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1659.82,2261.17 1659.82,2242.27 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2125.18,2261.17 2125.18,2242.27 "/>
-<path clip-path="url(#clip190)" d="M233.511 2306.74 L263.186 2306.74 L263.186 2310.67 L233.511 2310.67 L233.511 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M274.089 2319.63 L281.728 2319.63 L281.728 2293.27 L273.418 2294.93 L273.418 2290.67 L281.682 2289.01 L286.358 2289.01 L286.358 2319.63 L293.996 2319.63 L293.996 2323.57 L274.089 2323.57 L274.089 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M729.11 2292.09 Q725.499 2292.09 723.67 2295.65 Q721.864 2299.19 721.864 2306.32 Q721.864 2313.43 723.67 2316.99 Q725.499 2320.54 729.11 2320.54 Q732.744 2320.54 734.55 2316.99 Q736.378 2313.43 736.378 2306.32 Q736.378 2299.19 734.55 2295.65 Q732.744 2292.09 729.11 2292.09 M729.11 2288.38 Q734.92 2288.38 737.975 2292.99 Q741.054 2297.57 741.054 2306.32 Q741.054 2315.05 737.975 2319.66 Q734.92 2324.24 729.11 2324.24 Q723.3 2324.24 720.221 2319.66 Q717.165 2315.05 717.165 2306.32 Q717.165 2297.57 720.221 2292.99 Q723.3 2288.38 729.11 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1184.85 2319.63 L1192.49 2319.63 L1192.49 2293.27 L1184.18 2294.93 L1184.18 2290.67 L1192.44 2289.01 L1197.12 2289.01 L1197.12 2319.63 L1204.76 2319.63 L1204.76 2323.57 L1184.85 2323.57 L1184.85 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1654.47 2319.63 L1670.79 2319.63 L1670.79 2323.57 L1648.85 2323.57 L1648.85 2319.63 Q1651.51 2316.88 1656.1 2312.25 Q1660.7 2307.6 1661.88 2306.25 Q1664.13 2303.73 1665.01 2301.99 Q1665.91 2300.23 1665.91 2298.54 Q1665.91 2295.79 1663.97 2294.05 Q1662.04 2292.32 1658.94 2292.32 Q1656.74 2292.32 1654.29 2293.08 Q1651.86 2293.85 1649.08 2295.4 L1649.08 2290.67 Q1651.91 2289.54 1654.36 2288.96 Q1656.81 2288.38 1658.85 2288.38 Q1664.22 2288.38 1667.41 2291.07 Q1670.61 2293.75 1670.61 2298.24 Q1670.61 2300.37 1669.8 2302.29 Q1669.01 2304.19 1666.91 2306.79 Q1666.33 2307.46 1663.22 2310.67 Q1660.12 2313.87 1654.47 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M2129.43 2304.93 Q2132.78 2305.65 2134.66 2307.92 Q2136.56 2310.19 2136.56 2313.52 Q2136.56 2318.64 2133.04 2321.44 Q2129.52 2324.24 2123.04 2324.24 Q2120.86 2324.24 2118.55 2323.8 Q2116.25 2323.38 2113.8 2322.53 L2113.8 2318.01 Q2115.75 2319.15 2118.06 2319.73 Q2120.37 2320.3 2122.9 2320.3 Q2127.3 2320.3 2129.59 2318.57 Q2131.9 2316.83 2131.9 2313.52 Q2131.9 2310.47 2129.75 2308.75 Q2127.62 2307.02 2123.8 2307.02 L2119.77 2307.02 L2119.77 2303.17 L2123.99 2303.17 Q2127.44 2303.17 2129.26 2301.81 Q2131.09 2300.42 2131.09 2297.83 Q2131.09 2295.17 2129.19 2293.75 Q2127.32 2292.32 2123.8 2292.32 Q2121.88 2292.32 2119.68 2292.73 Q2117.48 2293.15 2114.84 2294.03 L2114.84 2289.86 Q2117.5 2289.12 2119.82 2288.75 Q2122.16 2288.38 2124.22 2288.38 Q2129.54 2288.38 2132.64 2290.81 Q2135.75 2293.22 2135.75 2297.34 Q2135.75 2300.21 2134.1 2302.2 Q2132.46 2304.17 2129.43 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,2001.74 2352.76,2001.74 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,1647.4 2352.76,1647.4 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,1293.05 2352.76,1293.05 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,938.704 2352.76,938.704 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,584.358 2352.76,584.358 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,230.012 2352.76,230.012 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2261.17 147.478,47.2441 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2001.74 166.376,2001.74 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,1647.4 166.376,1647.4 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,1293.05 166.376,1293.05 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,938.704 166.376,938.704 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,584.358 166.376,584.358 "/>
-<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,230.012 166.376,230.012 "/>
-<path clip-path="url(#clip190)" d="M51.3625 2002.19 L81.0383 2002.19 L81.0383 2006.13 L51.3625 2006.13 L51.3625 2002.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M95.1586 2015.09 L111.478 2015.09 L111.478 2019.02 L89.5336 2019.02 L89.5336 2015.09 Q92.1956 2012.33 96.7789 2007.7 Q101.385 2003.05 102.566 2001.71 Q104.811 1999.18 105.691 1997.45 Q106.594 1995.69 106.594 1994 Q106.594 1991.25 104.649 1989.51 Q102.728 1987.77 99.6261 1987.77 Q97.4271 1987.77 94.9734 1988.54 Q92.5428 1989.3 89.7651 1990.85 L89.7651 1986.13 Q92.5891 1985 95.0428 1984.42 Q97.4965 1983.84 99.5335 1983.84 Q104.904 1983.84 108.098 1986.52 Q111.293 1989.21 111.293 1993.7 Q111.293 1995.83 110.483 1997.75 Q109.696 1999.65 107.589 2002.24 Q107.01 2002.91 103.909 2006.13 Q100.807 2009.32 95.1586 2015.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M50.9921 1647.85 L80.6679 1647.85 L80.6679 1651.78 L50.9921 1651.78 L50.9921 1647.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M91.5706 1660.74 L99.2095 1660.74 L99.2095 1634.38 L90.8993 1636.04 L90.8993 1631.78 L99.1632 1630.12 L103.839 1630.12 L103.839 1660.74 L111.478 1660.74 L111.478 1664.68 L91.5706 1664.68 L91.5706 1660.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M99.5335 1278.85 Q95.9224 1278.85 94.0937 1282.41 Q92.2882 1285.96 92.2882 1293.09 Q92.2882 1300.19 94.0937 1303.76 Q95.9224 1307.3 99.5335 1307.3 Q103.168 1307.3 104.973 1303.76 Q106.802 1300.19 106.802 1293.09 Q106.802 1285.96 104.973 1282.41 Q103.168 1278.85 99.5335 1278.85 M99.5335 1275.15 Q105.344 1275.15 108.399 1279.75 Q111.478 1284.34 111.478 1293.09 Q111.478 1301.81 108.399 1306.42 Q105.344 1311 99.5335 1311 Q93.7234 1311 90.6447 1306.42 Q87.5892 1301.81 87.5892 1293.09 Q87.5892 1284.34 90.6447 1279.75 Q93.7234 1275.15 99.5335 1275.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M91.5706 952.049 L99.2095 952.049 L99.2095 925.683 L90.8993 927.35 L90.8993 923.091 L99.1632 921.424 L103.839 921.424 L103.839 952.049 L111.478 952.049 L111.478 955.984 L91.5706 955.984 L91.5706 952.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M95.1586 597.703 L111.478 597.703 L111.478 601.638 L89.5336 601.638 L89.5336 597.703 Q92.1956 594.948 96.7789 590.319 Q101.385 585.666 102.566 584.323 Q104.811 581.8 105.691 580.064 Q106.594 578.305 106.594 576.615 Q106.594 573.86 104.649 572.124 Q102.728 570.388 99.6261 570.388 Q97.4271 570.388 94.9734 571.152 Q92.5428 571.916 89.7651 573.467 L89.7651 568.745 Q92.5891 567.61 95.0428 567.032 Q97.4965 566.453 99.5335 566.453 Q104.904 566.453 108.098 569.138 Q111.293 571.823 111.293 576.314 Q111.293 578.444 110.483 580.365 Q109.696 582.263 107.589 584.856 Q107.01 585.527 103.909 588.745 Q100.807 591.939 95.1586 597.703 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M104.348 228.658 Q107.705 229.375 109.58 231.644 Q111.478 233.912 111.478 237.246 Q111.478 242.361 107.959 245.162 Q104.441 247.963 97.9595 247.963 Q95.7836 247.963 93.4688 247.523 Q91.1771 247.107 88.7234 246.25 L88.7234 241.736 Q90.6678 242.87 92.9826 243.449 Q95.2974 244.028 97.8206 244.028 Q102.219 244.028 104.51 242.292 Q106.825 240.556 106.825 237.246 Q106.825 234.19 104.672 232.477 Q102.543 230.741 98.7234 230.741 L94.6956 230.741 L94.6956 226.898 L98.9085 226.898 Q102.358 226.898 104.186 225.533 Q106.015 224.144 106.015 221.551 Q106.015 218.889 104.117 217.477 Q102.242 216.042 98.7234 216.042 Q96.8021 216.042 94.603 216.459 Q92.4039 216.875 89.7651 217.755 L89.7651 213.588 Q92.4271 212.847 94.7419 212.477 Q97.0798 212.107 99.14 212.107 Q104.464 212.107 107.566 214.537 Q110.668 216.945 110.668 221.065 Q110.668 223.935 109.024 225.926 Q107.381 227.894 104.348 228.658 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="672.666,1656.45 677.247,1639.43 681.263,1622.63 684.705,1606.1 687.564,1589.89 689.833,1574.02 691.507,1558.55 692.58,1543.51 693.051,1528.93 692.917,1514.86 692.181,1501.32 690.843,1488.36 688.906,1476.01 686.376,1464.29 683.259,1453.23 679.563,1442.87 675.296,1433.23 670.47,1424.33 665.097,1416.19 659.19,1408.84 652.763,1402.3 645.834,1396.57 638.419,1391.68 630.536,1387.64 622.206,1384.45 613.449,1382.13 604.288,1380.67 594.744,1380.1 584.843,1380.39 574.608,1381.57 564.065,1383.61 553.241,1386.52 542.162,1390.29 530.857,1394.92 519.354,1400.38 507.681,1406.66 495.867,1413.76 483.942,1421.64 471.937,1430.3 459.88,1439.71 447.802,1449.84 435.733,1460.68 423.704,1472.19 411.744,1484.35 399.884,1497.12 388.152,1510.47 376.579,1524.38 365.193,1538.8 354.023,1553.7 343.096,1569.04 332.44,1584.78 322.082,1600.89 312.047,1617.32 302.36,1634.03 293.047,1650.99 284.129,1668.14 275.63,1685.45 267.57,1702.87 259.97,1720.37 252.848,1737.88 246.223,1755.38 240.111,1772.82 234.527,1790.14 229.486,1807.33 225,1824.31 221.079,1841.07 217.735,1857.54 214.974,1873.7 212.805,1889.5 211.233,1904.91 210.261,1919.88 209.891,1934.37 210.126,1948.36 210.964,1961.8 212.403,1974.66 214.44,1986.91 217.069,1998.52 220.284,2009.46 224.077,2019.7 228.438,2029.22 233.357,2037.99 238.821,2046 244.817,2053.21 251.329,2059.62 258.341,2065.21 265.836,2069.96 273.796,2073.86 282.199,2076.9 291.025,2079.08 300.253,2080.38 309.858,2080.82 319.818,2080.37 330.106,2079.05 340.698,2076.86 351.567,2073.8 362.685,2069.89 374.026,2065.12 385.56,2059.52 397.259,2053.1 409.093,2045.87 421.033,2037.85 433.05,2029.07 445.112,2019.54 457.191,2009.28 469.255,1998.33 481.274,1986.71 493.22,1974.45 505.061,1961.57 516.767,1948.13 528.311,1934.13 539.663,1919.63 550.794,1904.65 561.678,1889.24 572.285,1873.44 582.591,1857.27 592.569,1840.79 602.194,1824.03 611.443,1807.04 620.292,1789.86 628.719,1772.52 636.703,1755.09 644.224,1737.59 651.263,1720.07 657.803,1702.58 663.827,1685.16 669.32,1667.85 674.269,1650.7 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2269.29,957.088 2277.2,946.119 2283.28,935.513 2287.5,925.297 2289.85,915.496 2290.34,906.136 2288.96,897.238 2285.71,888.826 2280.6,880.921 2273.64,873.542 2264.85,866.708 2254.26,860.436 2241.89,854.741 2227.76,849.638 2211.92,845.14 2194.41,841.257 2175.26,838 2154.53,835.376 2132.27,833.393 2108.54,832.054 2083.38,831.365 2056.88,831.325 2029.08,831.935 2000.07,833.195 1969.91,835.1 1938.69,837.646 1906.47,840.826 1873.34,844.633 1839.38,849.057 1804.67,854.087 1769.31,859.71 1733.38,865.913 1696.97,872.679 1660.18,879.993 1623.08,887.835 1585.79,896.186 1548.38,905.025 1510.96,914.331 1473.61,924.079 1436.44,934.246 1399.53,944.805 1362.97,955.732 1326.86,966.997 1291.29,978.574 1256.34,990.433 1222.11,1002.55 1188.68,1014.88 1156.13,1027.4 1124.54,1040.09 1094,1052.9 1064.57,1065.82 1036.34,1078.79 1009.37,1091.8 983.731,1104.8 959.488,1117.78 936.7,1130.68 915.424,1143.49 895.714,1156.17 877.619,1168.69 861.184,1181.01 846.45,1193.11 833.454,1204.96 822.228,1216.52 812.801,1227.77 805.196,1238.68 799.432,1249.23 795.523,1259.37 793.48,1269.1 793.307,1278.39 795.005,1287.2 798.569,1295.53 803.991,1303.35 811.257,1310.64 820.349,1317.38 831.244,1323.56 843.915,1329.15 858.33,1334.15 874.454,1338.55 892.245,1342.33 911.66,1345.48 932.65,1348 955.163,1349.87 979.142,1351.1 1004.53,1351.68 1031.25,1351.61 1059.26,1350.89 1088.47,1349.52 1118.81,1347.51 1150.21,1344.86 1182.59,1341.57 1215.87,1337.66 1249.96,1333.13 1284.78,1328 1320.24,1322.28 1356.26,1315.98 1392.74,1309.12 1429.59,1301.72 1466.72,1293.79 1504.05,1285.35 1541.46,1276.43 1578.88,1267.05 1616.2,1257.23 1653.34,1247 1690.19,1236.38 1726.68,1225.39 1762.71,1214.07 1798.18,1202.44 1833.01,1190.54 1867.12,1178.39 1900.41,1166.02 1932.8,1153.47 1964.22,1140.76 1994.58,1127.92 2023.81,1115 2051.83,1102.02 2078.58,1089.01 2103.99,1076.01 2127.99,1063.04 2150.53,1050.15 2171.55,1037.36 2190.99,1024.71 2208.81,1012.22 2224.96,999.929 2239.41,987.869 2252.11,976.068 2263.03,964.555 2272.15,953.36 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1706.28,1860.86 1698.38,1839.06 1689.05,1817.03 1678.32,1794.85 1666.21,1772.56 1652.75,1750.21 1637.99,1727.87 1621.94,1705.58 1604.67,1683.41 1586.2,1661.41 1566.58,1639.64 1545.87,1618.14 1524.12,1596.98 1501.37,1576.21 1477.69,1555.87 1453.13,1536.02 1427.77,1516.71 1401.65,1497.99 1374.85,1479.9 1347.43,1462.49 1319.46,1445.8 1291.01,1429.87 1262.15,1414.75 1232.96,1400.47 1203.51,1387.07 1173.87,1374.58 1144.11,1363.03 1114.31,1352.45 1084.54,1342.87 1054.89,1334.31 1025.41,1326.79 996.192,1320.33 967.302,1314.95 938.813,1310.67 910.797,1307.48 883.323,1305.4 856.46,1304.44 830.276,1304.6 804.835,1305.87 780.201,1308.26 756.437,1311.75 733.6,1316.34 711.749,1322.03 690.938,1328.78 671.219,1336.59 652.641,1345.44 635.25,1355.3 619.091,1366.15 604.204,1377.97 590.625,1390.72 578.389,1404.37 567.526,1418.89 558.063,1434.24 550.025,1450.38 543.431,1467.27 538.297,1484.88 534.637,1503.15 532.46,1522.04 531.771,1541.51 532.571,1561.5 534.859,1581.96 538.63,1602.85 543.873,1624.11 550.575,1645.69 558.721,1667.53 568.289,1689.59 579.255,1711.79 591.592,1734.1 605.27,1756.45 620.254,1778.79 636.506,1801.05 653.987,1823.2 672.652,1845.16 692.454,1866.9 713.345,1888.34 735.271,1909.44 758.179,1930.14 782.011,1950.4 806.707,1970.16 832.205,1989.38 858.443,2008 885.354,2025.98 912.87,2043.27 940.924,2059.84 969.446,2075.63 998.363,2090.61 1027.6,2104.75 1057.09,2118 1086.76,2130.33 1116.53,2141.72 1146.33,2152.13 1176.08,2161.55 1205.72,2169.93 1235.15,2177.27 1264.32,2183.55 1293.15,2188.74 1321.56,2192.85 1349.49,2195.85 1376.87,2197.74 1403.62,2198.51 1429.69,2198.17 1455,2196.71 1479.49,2194.13 1503.1,2190.45 1525.77,2185.68 1547.45,2179.81 1568.09,2172.88 1587.62,2164.89 1606,2155.87 1623.18,2145.84 1639.13,2134.83 1653.8,2122.85 1667.16,2109.95 1679.16,2096.15 1689.79,2081.49 1699.02,2066 1706.82,2049.73 1713.16,2032.71 1718.05,2014.99 1721.46,1996.61 1723.39,1977.62 1723.83,1958.07 1722.78,1937.99 1720.24,1917.45 1716.22,1896.5 1710.73,1875.18 1703.79,1853.55 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.032,1458.68 697.314,1453.52 699.149,1448.47 700.533,1443.54 701.463,1438.75 701.935,1434.1 701.95,1429.61 701.506,1425.28 700.605,1421.13 699.249,1417.18 697.442,1413.42 695.188,1409.87 692.493,1406.54 689.363,1403.43 685.806,1400.56 681.832,1397.92 677.449,1395.54 672.67,1393.41 667.506,1391.53 661.969,1389.92 656.074,1388.58 649.836,1387.5 643.27,1386.7 636.392,1386.18 629.22,1385.93 621.772,1385.96 614.066,1386.26 606.121,1386.85 597.957,1387.7 589.596,1388.83 581.057,1390.23 572.361,1391.9 563.532,1393.83 554.591,1396.01 545.559,1398.45 536.46,1401.13 527.317,1404.05 518.152,1407.21 508.988,1410.59 499.848,1414.18 490.756,1417.98 481.733,1421.98 472.802,1426.16 463.986,1430.52 455.306,1435.05 446.785,1439.73 438.443,1444.55 430.301,1449.5 422.381,1454.58 414.701,1459.75 407.281,1465.02 400.139,1470.37 393.293,1475.79 386.761,1481.26 380.558,1486.76 374.7,1492.3 369.202,1497.84 364.078,1503.38 359.34,1508.91 355,1514.4 351.069,1519.85 347.557,1525.24 344.473,1530.56 341.823,1535.8 339.616,1540.94 337.856,1545.97 336.548,1550.88 335.696,1555.65 335.3,1560.27 335.363,1564.74 335.883,1569.03 336.861,1573.15 338.293,1577.07 340.176,1580.79 342.504,1584.31 345.273,1587.6 348.475,1590.67 352.103,1593.5 356.147,1596.1 360.597,1598.44 365.442,1600.53 370.67,1602.36 376.267,1603.93 382.221,1605.22 388.515,1606.25 395.135,1607.01 402.064,1607.48 409.283,1607.69 416.776,1607.61 424.524,1607.26 432.507,1606.63 440.705,1605.73 449.098,1604.55 457.665,1603.11 466.384,1601.4 475.234,1599.43 484.192,1597.2 493.237,1594.72 502.345,1592 511.493,1589.03 520.659,1585.84 529.821,1582.43 538.954,1578.8 548.037,1574.96 557.046,1570.94 565.959,1566.72 574.754,1562.33 583.408,1557.78 591.901,1553.07 600.211,1548.23 608.316,1543.25 616.198,1538.16 623.835,1532.97 631.21,1527.68 638.303,1522.32 645.097,1516.9 651.576,1511.42 657.721,1505.91 663.52,1500.37 668.956,1494.83 674.016,1489.29 678.688,1483.77 682.96,1478.28 686.821,1472.84 690.262,1467.46 693.273,1462.15 695.849,1456.93 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1958.4,635.235 1961.49,606.133 1962.64,577.258 1961.85,548.683 1959.13,520.477 1954.47,492.713 1947.91,465.459 1939.44,438.783 1929.09,412.753 1916.89,387.433 1902.86,362.886 1887.05,339.174 1869.49,316.356 1850.22,294.49 1829.3,273.629 1806.77,253.826 1782.7,235.131 1757.13,217.59 1730.15,201.246 1701.81,186.142 1672.18,172.314 1641.34,159.798 1609.37,148.624 1576.34,138.82 1542.34,130.412 1507.46,123.419 1471.77,117.859 1435.37,113.747 1398.36,111.093 1360.82,109.902 1322.84,110.179 1284.53,111.922 1245.97,115.128 1207.27,119.788 1168.52,125.89 1129.81,133.419 1091.25,142.357 1052.93,152.68 1014.95,164.364 977.4,177.379 940.374,191.693 903.966,207.269 868.267,224.07 833.365,242.051 799.349,261.17 766.303,281.378 734.31,302.625 703.449,324.857 673.799,348.019 645.433,372.053 618.421,396.899 592.833,422.494 568.73,448.776 546.174,475.678 525.222,503.132 505.925,531.071 488.331,559.425 472.485,588.122 458.427,617.091 446.19,646.259 435.807,675.554 427.303,704.903 420.699,734.231 416.011,763.466 413.252,792.535 412.428,821.364 413.542,849.883 416.59,878.019 421.565,905.702 428.455,932.864 437.242,959.435 447.904,985.351 460.415,1010.55 474.744,1034.96 490.854,1058.52 508.705,1081.18 528.253,1102.88 549.449,1123.57 572.24,1143.19 596.568,1161.7 622.374,1179.04 649.593,1195.18 678.156,1210.07 707.992,1223.68 739.027,1235.97 771.183,1246.92 804.379,1256.49 838.533,1264.66 873.56,1271.41 909.371,1276.73 945.878,1280.6 982.988,1283 1020.61,1283.95 1058.65,1283.42 1097.01,1281.44 1135.6,1277.98 1174.32,1273.08 1213.07,1266.74 1251.75,1258.97 1290.28,1249.8 1328.55,1239.24 1366.46,1227.33 1403.93,1214.1 1440.86,1199.57 1477.16,1183.78 1512.73,1166.78 1547.49,1148.61 1581.35,1129.3 1614.22,1108.92 1646.03,1087.5 1676.69,1065.11 1706.13,1041.79 1734.27,1017.62 1761.05,992.643 1786.39,966.928 1810.24,940.537 1832.53,913.538 1853.21,885.997 1872.22,857.983 1889.53,829.567 1905.07,800.819 1918.83,771.811 1930.75,742.617 1940.82,713.307 1949.01,683.957 1955.29,654.64 1959.66,625.427 "/>
-<polyline clip-path="url(#clip192)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1961.32,1321.02 1961.7,1316.86 1961.03,1312.75 1959.29,1308.69 1956.5,1304.7 1952.66,1300.78 1947.79,1296.95 1941.88,1293.22 1934.96,1289.59 1927.05,1286.07 1918.16,1282.67 1908.32,1279.41 1897.55,1276.28 1885.88,1273.3 1873.34,1270.48 1859.95,1267.81 1845.76,1265.32 1830.8,1263 1815.1,1260.86 1798.7,1258.9 1781.66,1257.13 1764,1255.56 1745.77,1254.19 1727.02,1253.02 1707.79,1252.05 1688.14,1251.3 1668.11,1250.75 1647.75,1250.41 1627.12,1250.28 1606.25,1250.37 1585.22,1250.66 1564.06,1251.17 1542.84,1251.89 1521.6,1252.81 1500.4,1253.94 1479.29,1255.27 1458.33,1256.8 1437.56,1258.53 1417.04,1260.45 1396.82,1262.55 1376.95,1264.84 1357.48,1267.3 1338.46,1269.93 1319.93,1272.72 1301.94,1275.67 1284.55,1278.77 1267.78,1282.01 1251.69,1285.38 1236.31,1288.87 1221.68,1292.48 1207.83,1296.2 1194.81,1300.01 1182.65,1303.91 1171.37,1307.89 1161.01,1311.93 1151.58,1316.04 1143.12,1320.19 1135.64,1324.38 1129.16,1328.59 1123.7,1332.82 1119.28,1337.06 1115.9,1341.29 1113.57,1345.5 1112.3,1349.69 1112.09,1353.84 1112.95,1357.95 1114.86,1362 1117.83,1365.98 1121.84,1369.88 1126.89,1373.7 1132.97,1377.41 1140.06,1381.03 1148.13,1384.53 1157.18,1387.9 1167.18,1391.14 1178.11,1394.24 1189.93,1397.2 1202.61,1400 1216.14,1402.63 1230.46,1405.1 1245.55,1407.39 1261.37,1409.5 1277.87,1411.42 1295.03,1413.16 1312.79,1414.7 1331.11,1416.03 1349.94,1417.17 1369.24,1418.1 1388.96,1418.82 1409.05,1419.34 1429.46,1419.64 1450.13,1419.73 1471.03,1419.61 1492.09,1419.28 1513.26,1418.74 1534.49,1417.99 1555.72,1417.03 1576.91,1415.86 1598,1414.5 1618.94,1412.93 1639.67,1411.17 1660.14,1409.22 1680.3,1407.09 1700.11,1404.77 1719.51,1402.28 1738.45,1399.62 1756.89,1396.8 1774.78,1393.83 1792.07,1390.71 1808.73,1387.45 1824.71,1384.06 1839.97,1380.54 1854.46,1376.91 1868.17,1373.18 1881.05,1369.35 1893.07,1365.44 1904.19,1361.45 1914.4,1357.39 1923.67,1353.28 1931.97,1349.12 1939.28,1344.93 1945.58,1340.71 1950.87,1336.48 1955.12,1332.24 1958.32,1328.01 1960.47,1323.8 1961.56,1319.62 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="263.754,2261.17 263.754,47.2441 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.11,2261.17 729.11,47.2441 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1194.47,2261.17 1194.47,47.2441 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1659.82,2261.17 1659.82,47.2441 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2125.18,2261.17 2125.18,47.2441 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2261.17 2352.76,2261.17 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="263.754,2261.17 263.754,2242.27 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.11,2261.17 729.11,2242.27 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1194.47,2261.17 1194.47,2242.27 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1659.82,2261.17 1659.82,2242.27 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2125.18,2261.17 2125.18,2242.27 "/>
+<path clip-path="url(#clip010)" d="M233.511 2306.74 L263.186 2306.74 L263.186 2310.67 L233.511 2310.67 L233.511 2306.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M274.089 2319.63 L281.728 2319.63 L281.728 2293.27 L273.418 2294.93 L273.418 2290.67 L281.682 2289.01 L286.358 2289.01 L286.358 2319.63 L293.996 2319.63 L293.996 2323.57 L274.089 2323.57 L274.089 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M729.11 2292.09 Q725.499 2292.09 723.67 2295.65 Q721.864 2299.19 721.864 2306.32 Q721.864 2313.43 723.67 2316.99 Q725.499 2320.54 729.11 2320.54 Q732.744 2320.54 734.55 2316.99 Q736.378 2313.43 736.378 2306.32 Q736.378 2299.19 734.55 2295.65 Q732.744 2292.09 729.11 2292.09 M729.11 2288.38 Q734.92 2288.38 737.975 2292.99 Q741.054 2297.57 741.054 2306.32 Q741.054 2315.05 737.975 2319.66 Q734.92 2324.24 729.11 2324.24 Q723.3 2324.24 720.221 2319.66 Q717.165 2315.05 717.165 2306.32 Q717.165 2297.57 720.221 2292.99 Q723.3 2288.38 729.11 2288.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1184.85 2319.63 L1192.49 2319.63 L1192.49 2293.27 L1184.18 2294.93 L1184.18 2290.67 L1192.44 2289.01 L1197.12 2289.01 L1197.12 2319.63 L1204.76 2319.63 L1204.76 2323.57 L1184.85 2323.57 L1184.85 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M1654.47 2319.63 L1670.79 2319.63 L1670.79 2323.57 L1648.85 2323.57 L1648.85 2319.63 Q1651.51 2316.88 1656.1 2312.25 Q1660.7 2307.6 1661.88 2306.25 Q1664.13 2303.73 1665.01 2301.99 Q1665.91 2300.23 1665.91 2298.54 Q1665.91 2295.79 1663.97 2294.05 Q1662.04 2292.32 1658.94 2292.32 Q1656.74 2292.32 1654.29 2293.08 Q1651.86 2293.85 1649.08 2295.4 L1649.08 2290.67 Q1651.91 2289.54 1654.36 2288.96 Q1656.81 2288.38 1658.85 2288.38 Q1664.22 2288.38 1667.41 2291.07 Q1670.61 2293.75 1670.61 2298.24 Q1670.61 2300.37 1669.8 2302.29 Q1669.01 2304.19 1666.91 2306.79 Q1666.33 2307.46 1663.22 2310.67 Q1660.12 2313.87 1654.47 2319.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M2129.43 2304.93 Q2132.78 2305.65 2134.66 2307.92 Q2136.56 2310.19 2136.56 2313.52 Q2136.56 2318.64 2133.04 2321.44 Q2129.52 2324.24 2123.04 2324.24 Q2120.86 2324.24 2118.55 2323.8 Q2116.25 2323.38 2113.8 2322.53 L2113.8 2318.01 Q2115.75 2319.15 2118.06 2319.73 Q2120.37 2320.3 2122.9 2320.3 Q2127.3 2320.3 2129.59 2318.57 Q2131.9 2316.83 2131.9 2313.52 Q2131.9 2310.47 2129.75 2308.75 Q2127.62 2307.02 2123.8 2307.02 L2119.77 2307.02 L2119.77 2303.17 L2123.99 2303.17 Q2127.44 2303.17 2129.26 2301.81 Q2131.09 2300.42 2131.09 2297.83 Q2131.09 2295.17 2129.19 2293.75 Q2127.32 2292.32 2123.8 2292.32 Q2121.88 2292.32 2119.68 2292.73 Q2117.48 2293.15 2114.84 2294.03 L2114.84 2289.86 Q2117.5 2289.12 2119.82 2288.75 Q2122.16 2288.38 2124.22 2288.38 Q2129.54 2288.38 2132.64 2290.81 Q2135.75 2293.22 2135.75 2297.34 Q2135.75 2300.21 2134.1 2302.2 Q2132.46 2304.17 2129.43 2304.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,2001.74 2352.76,2001.74 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,1647.4 2352.76,1647.4 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,1293.05 2352.76,1293.05 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,938.704 2352.76,938.704 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,584.358 2352.76,584.358 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="147.478,230.012 2352.76,230.012 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2261.17 147.478,47.2441 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,2001.74 166.376,2001.74 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,1647.4 166.376,1647.4 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,1293.05 166.376,1293.05 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,938.704 166.376,938.704 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,584.358 166.376,584.358 "/>
+<polyline clip-path="url(#clip010)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="147.478,230.012 166.376,230.012 "/>
+<path clip-path="url(#clip010)" d="M51.3625 2002.19 L81.0383 2002.19 L81.0383 2006.13 L51.3625 2006.13 L51.3625 2002.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M95.1586 2015.09 L111.478 2015.09 L111.478 2019.02 L89.5336 2019.02 L89.5336 2015.09 Q92.1956 2012.33 96.7789 2007.7 Q101.385 2003.05 102.566 2001.71 Q104.811 1999.18 105.691 1997.45 Q106.594 1995.69 106.594 1994 Q106.594 1991.25 104.649 1989.51 Q102.728 1987.77 99.6261 1987.77 Q97.4271 1987.77 94.9734 1988.54 Q92.5428 1989.3 89.7651 1990.85 L89.7651 1986.13 Q92.5891 1985 95.0428 1984.42 Q97.4965 1983.84 99.5335 1983.84 Q104.904 1983.84 108.098 1986.52 Q111.293 1989.21 111.293 1993.7 Q111.293 1995.83 110.483 1997.75 Q109.696 1999.65 107.589 2002.24 Q107.01 2002.91 103.909 2006.13 Q100.807 2009.32 95.1586 2015.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M50.9921 1647.85 L80.6679 1647.85 L80.6679 1651.78 L50.9921 1651.78 L50.9921 1647.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M91.5706 1660.74 L99.2095 1660.74 L99.2095 1634.38 L90.8993 1636.04 L90.8993 1631.78 L99.1632 1630.12 L103.839 1630.12 L103.839 1660.74 L111.478 1660.74 L111.478 1664.68 L91.5706 1664.68 L91.5706 1660.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M99.5335 1278.85 Q95.9224 1278.85 94.0937 1282.41 Q92.2882 1285.96 92.2882 1293.09 Q92.2882 1300.19 94.0937 1303.76 Q95.9224 1307.3 99.5335 1307.3 Q103.168 1307.3 104.973 1303.76 Q106.802 1300.19 106.802 1293.09 Q106.802 1285.96 104.973 1282.41 Q103.168 1278.85 99.5335 1278.85 M99.5335 1275.15 Q105.344 1275.15 108.399 1279.75 Q111.478 1284.34 111.478 1293.09 Q111.478 1301.81 108.399 1306.42 Q105.344 1311 99.5335 1311 Q93.7234 1311 90.6447 1306.42 Q87.5892 1301.81 87.5892 1293.09 Q87.5892 1284.34 90.6447 1279.75 Q93.7234 1275.15 99.5335 1275.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M91.5706 952.049 L99.2095 952.049 L99.2095 925.683 L90.8993 927.35 L90.8993 923.091 L99.1632 921.424 L103.839 921.424 L103.839 952.049 L111.478 952.049 L111.478 955.984 L91.5706 955.984 L91.5706 952.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M95.1586 597.703 L111.478 597.703 L111.478 601.638 L89.5336 601.638 L89.5336 597.703 Q92.1956 594.948 96.7789 590.319 Q101.385 585.666 102.566 584.323 Q104.811 581.8 105.691 580.064 Q106.594 578.305 106.594 576.615 Q106.594 573.86 104.649 572.124 Q102.728 570.388 99.6261 570.388 Q97.4271 570.388 94.9734 571.152 Q92.5428 571.916 89.7651 573.467 L89.7651 568.745 Q92.5891 567.61 95.0428 567.032 Q97.4965 566.453 99.5335 566.453 Q104.904 566.453 108.098 569.138 Q111.293 571.823 111.293 576.314 Q111.293 578.444 110.483 580.365 Q109.696 582.263 107.589 584.856 Q107.01 585.527 103.909 588.745 Q100.807 591.939 95.1586 597.703 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip010)" d="M104.348 228.658 Q107.705 229.375 109.58 231.644 Q111.478 233.912 111.478 237.246 Q111.478 242.361 107.959 245.162 Q104.441 247.963 97.9595 247.963 Q95.7836 247.963 93.4688 247.523 Q91.1771 247.107 88.7234 246.25 L88.7234 241.736 Q90.6678 242.87 92.9826 243.449 Q95.2974 244.028 97.8206 244.028 Q102.219 244.028 104.51 242.292 Q106.825 240.556 106.825 237.246 Q106.825 234.19 104.672 232.477 Q102.543 230.741 98.7234 230.741 L94.6956 230.741 L94.6956 226.898 L98.9085 226.898 Q102.358 226.898 104.186 225.533 Q106.015 224.144 106.015 221.551 Q106.015 218.889 104.117 217.477 Q102.242 216.042 98.7234 216.042 Q96.8021 216.042 94.603 216.459 Q92.4039 216.875 89.7651 217.755 L89.7651 213.588 Q92.4271 212.847 94.7419 212.477 Q97.0798 212.107 99.14 212.107 Q104.464 212.107 107.566 214.537 Q110.668 216.945 110.668 221.065 Q110.668 223.935 109.024 225.926 Q107.381 227.894 104.348 228.658 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="672.666,1656.45 677.247,1639.43 681.263,1622.63 684.705,1606.1 687.564,1589.89 689.833,1574.02 691.507,1558.55 692.58,1543.51 693.051,1528.93 692.917,1514.86 692.181,1501.32 690.843,1488.36 688.906,1476.01 686.376,1464.29 683.259,1453.23 679.563,1442.87 675.296,1433.23 670.47,1424.33 665.097,1416.19 659.19,1408.84 652.763,1402.3 645.834,1396.57 638.419,1391.68 630.536,1387.64 622.206,1384.45 613.449,1382.13 604.288,1380.67 594.744,1380.1 584.843,1380.39 574.608,1381.57 564.065,1383.61 553.241,1386.52 542.162,1390.29 530.857,1394.92 519.354,1400.38 507.681,1406.66 495.867,1413.76 483.942,1421.64 471.937,1430.3 459.88,1439.71 447.802,1449.84 435.733,1460.68 423.704,1472.19 411.744,1484.35 399.884,1497.12 388.152,1510.47 376.579,1524.38 365.193,1538.8 354.023,1553.7 343.096,1569.04 332.44,1584.78 322.082,1600.89 312.047,1617.32 302.36,1634.03 293.047,1650.99 284.129,1668.14 275.63,1685.45 267.57,1702.87 259.97,1720.37 252.848,1737.88 246.223,1755.38 240.111,1772.82 234.527,1790.14 229.486,1807.33 225,1824.31 221.079,1841.07 217.735,1857.54 214.974,1873.7 212.805,1889.5 211.233,1904.91 210.261,1919.88 209.891,1934.37 210.126,1948.36 210.964,1961.8 212.403,1974.66 214.44,1986.91 217.069,1998.52 220.284,2009.46 224.077,2019.7 228.438,2029.22 233.357,2037.99 238.821,2046 244.817,2053.21 251.329,2059.62 258.341,2065.21 265.836,2069.96 273.796,2073.86 282.199,2076.9 291.025,2079.08 300.253,2080.38 309.858,2080.82 319.818,2080.37 330.106,2079.05 340.698,2076.86 351.567,2073.8 362.685,2069.89 374.026,2065.12 385.56,2059.52 397.259,2053.1 409.093,2045.87 421.033,2037.85 433.05,2029.07 445.112,2019.54 457.191,2009.28 469.255,1998.33 481.274,1986.71 493.22,1974.45 505.061,1961.57 516.767,1948.13 528.311,1934.13 539.663,1919.63 550.794,1904.65 561.678,1889.24 572.285,1873.44 582.591,1857.27 592.569,1840.79 602.194,1824.03 611.443,1807.04 620.292,1789.86 628.719,1772.52 636.703,1755.09 644.224,1737.59 651.263,1720.07 657.803,1702.58 663.827,1685.16 669.32,1667.85 674.269,1650.7 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2269.29,957.088 2277.2,946.119 2283.28,935.513 2287.5,925.297 2289.85,915.496 2290.34,906.136 2288.96,897.238 2285.71,888.826 2280.6,880.921 2273.64,873.542 2264.85,866.708 2254.26,860.436 2241.89,854.741 2227.76,849.638 2211.92,845.14 2194.41,841.257 2175.26,838 2154.53,835.376 2132.27,833.393 2108.54,832.054 2083.38,831.365 2056.88,831.325 2029.08,831.935 2000.07,833.195 1969.91,835.1 1938.69,837.646 1906.47,840.826 1873.34,844.633 1839.38,849.057 1804.67,854.087 1769.31,859.71 1733.38,865.913 1696.97,872.679 1660.18,879.993 1623.08,887.835 1585.79,896.186 1548.38,905.025 1510.96,914.331 1473.61,924.079 1436.44,934.246 1399.53,944.805 1362.97,955.732 1326.86,966.997 1291.29,978.574 1256.34,990.433 1222.11,1002.55 1188.68,1014.88 1156.13,1027.4 1124.54,1040.09 1094,1052.9 1064.57,1065.82 1036.34,1078.79 1009.37,1091.8 983.731,1104.8 959.488,1117.78 936.7,1130.68 915.424,1143.49 895.714,1156.17 877.619,1168.69 861.184,1181.01 846.45,1193.11 833.454,1204.96 822.228,1216.52 812.801,1227.77 805.196,1238.68 799.432,1249.23 795.523,1259.37 793.48,1269.1 793.307,1278.39 795.005,1287.2 798.569,1295.53 803.991,1303.35 811.257,1310.64 820.349,1317.38 831.244,1323.56 843.915,1329.15 858.33,1334.15 874.454,1338.55 892.245,1342.33 911.66,1345.48 932.65,1348 955.163,1349.87 979.142,1351.1 1004.53,1351.68 1031.25,1351.61 1059.26,1350.89 1088.47,1349.52 1118.81,1347.51 1150.21,1344.86 1182.59,1341.57 1215.87,1337.66 1249.96,1333.13 1284.78,1328 1320.24,1322.28 1356.26,1315.98 1392.74,1309.12 1429.59,1301.72 1466.72,1293.79 1504.05,1285.35 1541.46,1276.43 1578.88,1267.05 1616.2,1257.23 1653.34,1247 1690.19,1236.38 1726.68,1225.39 1762.71,1214.07 1798.18,1202.44 1833.01,1190.54 1867.12,1178.39 1900.41,1166.02 1932.8,1153.47 1964.22,1140.76 1994.58,1127.92 2023.81,1115 2051.83,1102.02 2078.58,1089.01 2103.99,1076.01 2127.99,1063.04 2150.53,1050.15 2171.55,1037.36 2190.99,1024.71 2208.81,1012.22 2224.96,999.929 2239.41,987.869 2252.11,976.068 2263.03,964.555 2272.15,953.36 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1706.28,1860.86 1698.38,1839.06 1689.05,1817.03 1678.32,1794.85 1666.21,1772.56 1652.75,1750.21 1637.99,1727.87 1621.94,1705.58 1604.67,1683.41 1586.2,1661.41 1566.58,1639.64 1545.87,1618.14 1524.12,1596.98 1501.37,1576.21 1477.69,1555.87 1453.13,1536.02 1427.77,1516.71 1401.65,1497.99 1374.85,1479.9 1347.43,1462.49 1319.46,1445.8 1291.01,1429.87 1262.15,1414.75 1232.96,1400.47 1203.51,1387.07 1173.87,1374.58 1144.11,1363.03 1114.31,1352.45 1084.54,1342.87 1054.89,1334.31 1025.41,1326.79 996.192,1320.33 967.302,1314.95 938.813,1310.67 910.797,1307.48 883.323,1305.4 856.46,1304.44 830.276,1304.6 804.835,1305.87 780.201,1308.26 756.437,1311.75 733.6,1316.34 711.749,1322.03 690.938,1328.78 671.219,1336.59 652.641,1345.44 635.25,1355.3 619.091,1366.15 604.204,1377.97 590.625,1390.72 578.389,1404.37 567.526,1418.89 558.063,1434.24 550.025,1450.38 543.431,1467.27 538.297,1484.88 534.637,1503.15 532.46,1522.04 531.771,1541.51 532.571,1561.5 534.859,1581.96 538.63,1602.85 543.873,1624.11 550.575,1645.69 558.721,1667.53 568.289,1689.59 579.255,1711.79 591.592,1734.1 605.27,1756.45 620.254,1778.79 636.506,1801.05 653.987,1823.2 672.652,1845.16 692.454,1866.9 713.345,1888.34 735.271,1909.44 758.179,1930.14 782.011,1950.4 806.707,1970.16 832.205,1989.38 858.443,2008 885.354,2025.98 912.87,2043.27 940.924,2059.84 969.446,2075.63 998.363,2090.61 1027.6,2104.75 1057.09,2118 1086.76,2130.33 1116.53,2141.72 1146.33,2152.13 1176.08,2161.55 1205.72,2169.93 1235.15,2177.27 1264.32,2183.55 1293.15,2188.74 1321.56,2192.85 1349.49,2195.85 1376.87,2197.74 1403.62,2198.51 1429.69,2198.17 1455,2196.71 1479.49,2194.13 1503.1,2190.45 1525.77,2185.68 1547.45,2179.81 1568.09,2172.88 1587.62,2164.89 1606,2155.87 1623.18,2145.84 1639.13,2134.83 1653.8,2122.85 1667.16,2109.95 1679.16,2096.15 1689.79,2081.49 1699.02,2066 1706.82,2049.73 1713.16,2032.71 1718.05,2014.99 1721.46,1996.61 1723.39,1977.62 1723.83,1958.07 1722.78,1937.99 1720.24,1917.45 1716.22,1896.5 1710.73,1875.18 1703.79,1853.55 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.032,1458.68 697.314,1453.52 699.149,1448.47 700.533,1443.54 701.463,1438.75 701.935,1434.1 701.95,1429.61 701.506,1425.28 700.605,1421.13 699.249,1417.18 697.442,1413.42 695.188,1409.87 692.493,1406.54 689.363,1403.43 685.806,1400.56 681.832,1397.92 677.449,1395.54 672.67,1393.41 667.506,1391.53 661.969,1389.92 656.074,1388.58 649.836,1387.5 643.27,1386.7 636.392,1386.18 629.22,1385.93 621.772,1385.96 614.066,1386.26 606.121,1386.85 597.957,1387.7 589.596,1388.83 581.057,1390.23 572.361,1391.9 563.532,1393.83 554.591,1396.01 545.559,1398.45 536.46,1401.13 527.317,1404.05 518.152,1407.21 508.988,1410.59 499.848,1414.18 490.756,1417.98 481.733,1421.98 472.802,1426.16 463.986,1430.52 455.306,1435.05 446.785,1439.73 438.443,1444.55 430.301,1449.5 422.381,1454.58 414.701,1459.75 407.281,1465.02 400.139,1470.37 393.293,1475.79 386.761,1481.26 380.558,1486.76 374.7,1492.3 369.202,1497.84 364.078,1503.38 359.34,1508.91 355,1514.4 351.069,1519.85 347.557,1525.24 344.473,1530.56 341.823,1535.8 339.616,1540.94 337.856,1545.97 336.548,1550.88 335.696,1555.65 335.3,1560.27 335.363,1564.74 335.883,1569.03 336.861,1573.15 338.293,1577.07 340.176,1580.79 342.504,1584.31 345.273,1587.6 348.475,1590.67 352.103,1593.5 356.147,1596.1 360.597,1598.44 365.442,1600.53 370.67,1602.36 376.267,1603.93 382.221,1605.22 388.515,1606.25 395.135,1607.01 402.064,1607.48 409.283,1607.69 416.776,1607.61 424.524,1607.26 432.507,1606.63 440.705,1605.73 449.098,1604.55 457.665,1603.11 466.384,1601.4 475.234,1599.43 484.192,1597.2 493.237,1594.72 502.345,1592 511.493,1589.03 520.659,1585.84 529.821,1582.43 538.954,1578.8 548.037,1574.96 557.046,1570.94 565.959,1566.72 574.754,1562.33 583.408,1557.78 591.901,1553.07 600.211,1548.23 608.316,1543.25 616.198,1538.16 623.835,1532.97 631.21,1527.68 638.303,1522.32 645.097,1516.9 651.576,1511.42 657.721,1505.91 663.52,1500.37 668.956,1494.83 674.016,1489.29 678.688,1483.77 682.96,1478.28 686.821,1472.84 690.262,1467.46 693.273,1462.15 695.849,1456.93 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1958.4,635.235 1961.49,606.133 1962.64,577.258 1961.85,548.683 1959.13,520.477 1954.47,492.713 1947.91,465.459 1939.44,438.783 1929.09,412.753 1916.89,387.433 1902.86,362.886 1887.05,339.174 1869.49,316.356 1850.22,294.49 1829.3,273.629 1806.77,253.826 1782.7,235.131 1757.13,217.59 1730.15,201.246 1701.81,186.142 1672.18,172.314 1641.34,159.798 1609.37,148.624 1576.34,138.82 1542.34,130.412 1507.46,123.419 1471.77,117.859 1435.37,113.747 1398.36,111.093 1360.82,109.902 1322.84,110.179 1284.53,111.922 1245.97,115.128 1207.27,119.788 1168.52,125.89 1129.81,133.419 1091.25,142.357 1052.93,152.68 1014.95,164.364 977.4,177.379 940.374,191.693 903.966,207.269 868.267,224.07 833.365,242.051 799.349,261.17 766.303,281.378 734.31,302.625 703.449,324.857 673.799,348.019 645.433,372.053 618.421,396.899 592.833,422.494 568.73,448.776 546.174,475.678 525.222,503.132 505.925,531.071 488.331,559.425 472.485,588.122 458.427,617.091 446.19,646.259 435.807,675.554 427.303,704.903 420.699,734.231 416.011,763.466 413.252,792.535 412.428,821.364 413.542,849.883 416.59,878.019 421.565,905.702 428.455,932.864 437.242,959.435 447.904,985.351 460.415,1010.55 474.744,1034.96 490.854,1058.52 508.705,1081.18 528.253,1102.88 549.449,1123.57 572.24,1143.19 596.568,1161.7 622.374,1179.04 649.593,1195.18 678.156,1210.07 707.992,1223.68 739.027,1235.97 771.183,1246.92 804.379,1256.49 838.533,1264.66 873.56,1271.41 909.371,1276.73 945.878,1280.6 982.988,1283 1020.61,1283.95 1058.65,1283.42 1097.01,1281.44 1135.6,1277.98 1174.32,1273.08 1213.07,1266.74 1251.75,1258.97 1290.28,1249.8 1328.55,1239.24 1366.46,1227.33 1403.93,1214.1 1440.86,1199.57 1477.16,1183.78 1512.73,1166.78 1547.49,1148.61 1581.35,1129.3 1614.22,1108.92 1646.03,1087.5 1676.69,1065.11 1706.13,1041.79 1734.27,1017.62 1761.05,992.643 1786.39,966.928 1810.24,940.537 1832.53,913.538 1853.21,885.997 1872.22,857.983 1889.53,829.567 1905.07,800.819 1918.83,771.811 1930.75,742.617 1940.82,713.307 1949.01,683.957 1955.29,654.64 1959.66,625.427 "/>
+<polyline clip-path="url(#clip012)" style="stroke:#000080; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1961.32,1321.02 1961.7,1316.86 1961.03,1312.75 1959.29,1308.69 1956.5,1304.7 1952.66,1300.78 1947.79,1296.95 1941.88,1293.22 1934.96,1289.59 1927.05,1286.07 1918.16,1282.67 1908.32,1279.41 1897.55,1276.28 1885.88,1273.3 1873.34,1270.48 1859.95,1267.81 1845.76,1265.32 1830.8,1263 1815.1,1260.86 1798.7,1258.9 1781.66,1257.13 1764,1255.56 1745.77,1254.19 1727.02,1253.02 1707.79,1252.05 1688.14,1251.3 1668.11,1250.75 1647.75,1250.41 1627.12,1250.28 1606.25,1250.37 1585.22,1250.66 1564.06,1251.17 1542.84,1251.89 1521.6,1252.81 1500.4,1253.94 1479.29,1255.27 1458.33,1256.8 1437.56,1258.53 1417.04,1260.45 1396.82,1262.55 1376.95,1264.84 1357.48,1267.3 1338.46,1269.93 1319.93,1272.72 1301.94,1275.67 1284.55,1278.77 1267.78,1282.01 1251.69,1285.38 1236.31,1288.87 1221.68,1292.48 1207.83,1296.2 1194.81,1300.01 1182.65,1303.91 1171.37,1307.89 1161.01,1311.93 1151.58,1316.04 1143.12,1320.19 1135.64,1324.38 1129.16,1328.59 1123.7,1332.82 1119.28,1337.06 1115.9,1341.29 1113.57,1345.5 1112.3,1349.69 1112.09,1353.84 1112.95,1357.95 1114.86,1362 1117.83,1365.98 1121.84,1369.88 1126.89,1373.7 1132.97,1377.41 1140.06,1381.03 1148.13,1384.53 1157.18,1387.9 1167.18,1391.14 1178.11,1394.24 1189.93,1397.2 1202.61,1400 1216.14,1402.63 1230.46,1405.1 1245.55,1407.39 1261.37,1409.5 1277.87,1411.42 1295.03,1413.16 1312.79,1414.7 1331.11,1416.03 1349.94,1417.17 1369.24,1418.1 1388.96,1418.82 1409.05,1419.34 1429.46,1419.64 1450.13,1419.73 1471.03,1419.61 1492.09,1419.28 1513.26,1418.74 1534.49,1417.99 1555.72,1417.03 1576.91,1415.86 1598,1414.5 1618.94,1412.93 1639.67,1411.17 1660.14,1409.22 1680.3,1407.09 1700.11,1404.77 1719.51,1402.28 1738.45,1399.62 1756.89,1396.8 1774.78,1393.83 1792.07,1390.71 1808.73,1387.45 1824.71,1384.06 1839.97,1380.54 1854.46,1376.91 1868.17,1373.18 1881.05,1369.35 1893.07,1365.44 1904.19,1361.45 1914.4,1357.39 1923.67,1353.28 1931.97,1349.12 1939.28,1344.93 1945.58,1340.71 1950.87,1336.48 1955.12,1332.24 1958.32,1328.01 1960.47,1323.8 1961.56,1319.62 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/conic/min_ellipse/index.html b/previews/PR3562/tutorials/conic/min_ellipse/index.html
index 11337d10d26..61864d30854 100644
--- a/previews/PR3562/tutorials/conic/min_ellipse/index.html
+++ b/previews/PR3562/tutorials/conic/min_ellipse/index.html
@@ -45,7 +45,7 @@
 for ellipse in ellipses
     plot_ellipse(plot, ellipse)
 end
-plot</code></pre><img src="1b95f9f7.svg" alt="Example block output"/><h2 id="Build-the-model"><a class="docs-heading-anchor" href="#Build-the-model">Build the model</a><a id="Build-the-model-1"></a><a class="docs-heading-anchor-permalink" href="#Build-the-model" title="Permalink"></a></h2><p>Now let&#39;s build the model, using the change-of-variables <code>P²</code> = <span>$P^2$</span> and <code>P_q</code> = <span>$P q$</span>. We&#39;ll recover the true value of <code>P</code> and <code>q</code> after the solve.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+plot</code></pre><img src="ca1157bb.svg" alt="Example block output"/><h2 id="Build-the-model"><a class="docs-heading-anchor" href="#Build-the-model">Build the model</a><a id="Build-the-model-1"></a><a class="docs-heading-anchor-permalink" href="#Build-the-model" title="Permalink"></a></h2><p>Now let&#39;s build the model, using the change-of-variables <code>P²</code> = <span>$P^2$</span> and <code>P_q</code> = <span>$P q$</span>. We&#39;ll recover the true value of <code>P</code> and <code>q</code> after the solve.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 # We need to use a tighter tolerance for this example, otherwise the bounding
 # ellipse won&#39;t actually be bounding...
 set_attribute(model, &quot;eps_rel&quot;, 1e-6)
@@ -83,13 +83,13 @@
   Dual objective value : -4.04364e+00
 
 * Work counters
-  Solve time (sec)   : 1.50172e-01
+  Solve time (sec)   : 2.26980e-01
 </code></pre><h2 id="Results"><a class="docs-heading-anchor" href="#Results">Results</a><a id="Results-1"></a><a class="docs-heading-anchor-permalink" href="#Results" title="Permalink"></a></h2><p>After solving the model to optimality we can recover the solution in terms of <span>$P$</span> and <span>$q$</span>:</p><pre><code class="language-julia hljs">P = sqrt(value.(P²))
 q = P \ value.(P_q)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- -0.3964217695777314
- -0.021394178555636897</code></pre><p>Finally, overlaying the solution in the plot we see the minimal area enclosing ellipsoid:</p><pre><code class="language-julia hljs">Plots.plot!(
+ -0.39642176943764484
+ -0.021394178202894645</code></pre><p>Finally, overlaying the solution in the plot we see the minimal area enclosing ellipsoid:</p><pre><code class="language-julia hljs">Plots.plot!(
     plot,
     [tuple(P \ [cos(θ) - q[1], sin(θ) - q[2]]...) for θ in 0:0.05:(2pi+0.05)];
     c = :crimson,
     label = nothing,
-)</code></pre><img src="1521bd4d.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../experiment_design/">« Experiment design</a><a class="docs-footer-nextpage" href="../ellipse_approx/">Ellipsoid approximation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><img src="8380bd2b.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../experiment_design/">« Experiment design</a><a class="docs-footer-nextpage" href="../ellipse_approx/">Ellipsoid approximation »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/quantum_discrimination/index.html b/previews/PR3562/tutorials/conic/quantum_discrimination/index.html
index 3ae7af76a19..7d9e21f4e2b 100644
--- a/previews/PR3562/tutorials/conic/quantum_discrimination/index.html
+++ b/previews/PR3562/tutorials/conic/quantum_discrimination/index.html
@@ -43,10 +43,10 @@
   Dual objective value : 8.64062e-01
 
 * Work counters
-  Solve time (sec)   : 2.90212e-04
-</code></pre><p>The probability of guessing correctly is:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640614507314229</code></pre><p>When <code>N = 2</code>, there is a known analytical solution of:</p><pre><code class="language-julia hljs">0.5 + 0.25 * sum(LinearAlgebra.svdvals(ρ[1] - ρ[2]))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640627582954737</code></pre><p>proving that we found the optimal solution.</p><p>Finally, the optimal POVM is:</p><pre><code class="language-julia hljs">solution = [value.(e) for e in E]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Matrix{ComplexF64}}:
- [0.9495721399750005 - 0.0im 0.034424516039770876 + 0.21609731371190552im; 0.034424516039770876 - 0.21609731371190552im 0.05042785512985092 - 0.0im]
- [0.0504278551760218 - 0.0im -0.03442451605312506 - 0.21609731370614893im; -0.03442451605312506 + 0.21609731370614893im 0.9495721400119403 - 0.0im]</code></pre><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Duality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual of this problem instead of the primal. If performance is an issue, see the <a href="../dualization/#Dualization">Dualization</a> tutorial for more details.</p></div></div><h2 id="Alternative-formulation"><a class="docs-heading-anchor" href="#Alternative-formulation">Alternative formulation</a><a id="Alternative-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#Alternative-formulation" title="Permalink"></a></h2><p>The formulation above includes <code>N</code> Hermitian matrices and a set of linear equality constraints. We can simplify the problem by replacing <span>$E_N$</span> with <span>$E_N = I - \sum\limits_{i=1}^{N-1} E_i$</span>. This results in:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+  Solve time (sec)   : 4.60003e-04
+</code></pre><p>The probability of guessing correctly is:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640614507314223</code></pre><p>When <code>N = 2</code>, there is a known analytical solution of:</p><pre><code class="language-julia hljs">0.5 + 0.25 * sum(LinearAlgebra.svdvals(ρ[1] - ρ[2]))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640627582954737</code></pre><p>proving that we found the optimal solution.</p><p>Finally, the optimal POVM is:</p><pre><code class="language-julia hljs">solution = [value.(e) for e in E]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Matrix{ComplexF64}}:
+ [0.9495721399750018 - 0.0im 0.03442451603977107 + 0.216097313711905im; 0.03442451603977107 - 0.216097313711905im 0.050427855129854156 - 0.0im]
+ [0.050427855176020456 - 0.0im -0.034424516053125256 - 0.21609731370614849im; -0.034424516053125256 + 0.21609731370614849im 0.9495721400119368 - 0.0im]</code></pre><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Duality plays a large role in solving conic optimization models. Depending on the solver, it can be more efficient to solve the dual of this problem instead of the primal. If performance is an issue, see the <a href="../dualization/#Dualization">Dualization</a> tutorial for more details.</p></div></div><h2 id="Alternative-formulation"><a class="docs-heading-anchor" href="#Alternative-formulation">Alternative formulation</a><a id="Alternative-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#Alternative-formulation" title="Permalink"></a></h2><p>The formulation above includes <code>N</code> Hermitian matrices and a set of linear equality constraints. We can simplify the problem by replacing <span>$E_N$</span> with <span>$E_N = I - \sum\limits_{i=1}^{N-1} E_i$</span>. This results in:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 E = [@variable(model, [1:d, 1:d] in HermitianPSDCone()) for i in 1:N-1]
 E_N = LinearAlgebra.Hermitian(LinearAlgebra.I - sum(E))
@@ -69,5 +69,5 @@
   Dual objective value : 8.64062e-01
 
 * Work counters
-  Solve time (sec)   : 2.77499e-04
-</code></pre><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640596603179969</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ellipse_approx/">« Ellipsoid approximation</a><a class="docs-footer-nextpage" href="../../algorithms/benders_decomposition/">Benders decomposition »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  Solve time (sec)   : 4.37804e-04
+</code></pre><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.8640596603179971</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../ellipse_approx/">« Ellipsoid approximation</a><a class="docs-footer-nextpage" href="../../algorithms/benders_decomposition/">Benders decomposition »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/simple_examples/c9b2b8b7.svg b/previews/PR3562/tutorials/conic/simple_examples/4350a7ce.svg
similarity index 79%
rename from previews/PR3562/tutorials/conic/simple_examples/c9b2b8b7.svg
rename to previews/PR3562/tutorials/conic/simple_examples/4350a7ce.svg
index 8973b13e5a1..052d9ad39b8 100644
--- a/previews/PR3562/tutorials/conic/simple_examples/c9b2b8b7.svg
+++ b/previews/PR3562/tutorials/conic/simple_examples/4350a7ce.svg
@@ -1,49 +1,49 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip310">
+  <clipPath id="clip130">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip310)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip130)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip311">
+  <clipPath id="clip131">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
 <defs>
-  <clipPath id="clip312">
+  <clipPath id="clip132">
     <rect x="530" y="47" width="1440" height="1440"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip312)" d="M734.179 1024.82 L734.179 370.632 L1388.36 181.785 L1766.06 508.877 L1766.06 1163.06 L1111.87 1351.91 L734.179 1024.82  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="701.469,1066.97 1421.07,859.235 1421.07,139.634 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="795.893,1148.74 1515.49,941.008 1515.49,221.407 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="890.316,1230.51 1609.92,1022.78 1609.92,303.179 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="984.739,1312.29 1704.34,1104.55 1704.34,384.952 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1079.16,1394.06 1798.76,1186.33 1798.76,466.725 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,1050.61 1098.05,1410.41 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="701.469,1066.97 710.105,1064.47 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="795.893,1148.74 804.528,1146.25 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="890.316,1230.51 898.951,1228.02 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="984.739,1312.29 993.374,1309.79 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1079.16,1394.06 1087.8,1391.57 "/>
-<path clip-path="url(#clip310)" d="M575.672 1089.19 L605.347 1089.19 L605.347 1093.12 L575.672 1093.12 L575.672 1089.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M616.25 1102.08 L623.889 1102.08 L623.889 1075.72 L615.579 1077.38 L615.579 1073.12 L623.843 1071.46 L628.518 1071.46 L628.518 1102.08 L636.157 1102.08 L636.157 1106.02 L616.25 1106.02 L616.25 1102.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M645.602 1100.14 L650.486 1100.14 L650.486 1106.02 L645.602 1106.02 L645.602 1100.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M670.671 1074.54 Q667.06 1074.54 665.231 1078.1 Q663.426 1081.64 663.426 1088.77 Q663.426 1095.88 665.231 1099.44 Q667.06 1102.98 670.671 1102.98 Q674.305 1102.98 676.111 1099.44 Q677.94 1095.88 677.94 1088.77 Q677.94 1081.64 676.111 1078.1 Q674.305 1074.54 670.671 1074.54 M670.671 1070.83 Q676.481 1070.83 679.537 1075.44 Q682.615 1080.02 682.615 1088.77 Q682.615 1097.5 679.537 1102.11 Q676.481 1106.69 670.671 1106.69 Q664.861 1106.69 661.782 1102.11 Q658.727 1097.5 658.727 1088.77 Q658.727 1080.02 661.782 1075.44 Q664.861 1070.83 670.671 1070.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M671.09 1170.96 L700.766 1170.96 L700.766 1174.9 L671.09 1174.9 L671.09 1170.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M720.858 1156.31 Q717.247 1156.31 715.419 1159.87 Q713.613 1163.42 713.613 1170.55 Q713.613 1177.65 715.419 1181.22 Q717.247 1184.76 720.858 1184.76 Q724.493 1184.76 726.298 1181.22 Q728.127 1177.65 728.127 1170.55 Q728.127 1163.42 726.298 1159.87 Q724.493 1156.31 720.858 1156.31 M720.858 1152.61 Q726.669 1152.61 729.724 1157.21 Q732.803 1161.8 732.803 1170.55 Q732.803 1179.27 729.724 1183.88 Q726.669 1188.46 720.858 1188.46 Q715.048 1188.46 711.97 1183.88 Q708.914 1179.27 708.914 1170.55 Q708.914 1161.8 711.97 1157.21 Q715.048 1152.61 720.858 1152.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M741.02 1181.91 L745.905 1181.91 L745.905 1187.79 L741.02 1187.79 L741.02 1181.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M756.136 1153.23 L774.492 1153.23 L774.492 1157.17 L760.418 1157.17 L760.418 1165.64 Q761.437 1165.29 762.455 1165.13 Q763.474 1164.94 764.492 1164.94 Q770.279 1164.94 773.659 1168.11 Q777.039 1171.29 777.039 1176.7 Q777.039 1182.28 773.566 1185.38 Q770.094 1188.46 763.775 1188.46 Q761.599 1188.46 759.33 1188.09 Q757.085 1187.72 754.678 1186.98 L754.678 1182.28 Q756.761 1183.42 758.983 1183.97 Q761.205 1184.53 763.682 1184.53 Q767.687 1184.53 770.025 1182.42 Q772.363 1180.31 772.363 1176.7 Q772.363 1173.09 770.025 1170.98 Q767.687 1168.88 763.682 1168.88 Q761.807 1168.88 759.932 1169.3 Q758.08 1169.71 756.136 1170.59 L756.136 1153.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M814.286 1238.08 Q810.675 1238.08 808.847 1241.65 Q807.041 1245.19 807.041 1252.32 Q807.041 1259.42 808.847 1262.99 Q810.675 1266.53 814.286 1266.53 Q817.921 1266.53 819.726 1262.99 Q821.555 1259.42 821.555 1252.32 Q821.555 1245.19 819.726 1241.65 Q817.921 1238.08 814.286 1238.08 M814.286 1234.38 Q820.096 1234.38 823.152 1238.98 Q826.231 1243.57 826.231 1252.32 Q826.231 1261.04 823.152 1265.65 Q820.096 1270.23 814.286 1270.23 Q808.476 1270.23 805.397 1265.65 Q802.342 1261.04 802.342 1252.32 Q802.342 1243.57 805.397 1238.98 Q808.476 1234.38 814.286 1234.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M834.448 1263.68 L839.332 1263.68 L839.332 1269.56 L834.448 1269.56 L834.448 1263.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M859.518 1238.08 Q855.906 1238.08 854.078 1241.65 Q852.272 1245.19 852.272 1252.32 Q852.272 1259.42 854.078 1262.99 Q855.906 1266.53 859.518 1266.53 Q863.152 1266.53 864.957 1262.99 Q866.786 1259.42 866.786 1252.32 Q866.786 1245.19 864.957 1241.65 Q863.152 1238.08 859.518 1238.08 M859.518 1234.38 Q865.328 1234.38 868.383 1238.98 Q871.462 1243.57 871.462 1252.32 Q871.462 1261.04 868.383 1265.65 Q865.328 1270.23 859.518 1270.23 Q853.707 1270.23 850.629 1265.65 Q847.573 1261.04 847.573 1252.32 Q847.573 1243.57 850.629 1238.98 Q853.707 1234.38 859.518 1234.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M909.705 1319.85 Q906.094 1319.85 904.265 1323.42 Q902.46 1326.96 902.46 1334.09 Q902.46 1341.2 904.265 1344.76 Q906.094 1348.3 909.705 1348.3 Q913.339 1348.3 915.145 1344.76 Q916.973 1341.2 916.973 1334.09 Q916.973 1326.96 915.145 1323.42 Q913.339 1319.85 909.705 1319.85 M909.705 1316.15 Q915.515 1316.15 918.571 1320.76 Q921.649 1325.34 921.649 1334.09 Q921.649 1342.82 918.571 1347.42 Q915.515 1352.01 909.705 1352.01 Q903.895 1352.01 900.816 1347.42 Q897.761 1342.82 897.761 1334.09 Q897.761 1325.34 900.816 1320.76 Q903.895 1316.15 909.705 1316.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M929.867 1345.46 L934.751 1345.46 L934.751 1351.34 L929.867 1351.34 L929.867 1345.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M944.983 1316.78 L963.339 1316.78 L963.339 1320.71 L949.265 1320.71 L949.265 1329.18 Q950.283 1328.84 951.302 1328.67 Q952.32 1328.49 953.339 1328.49 Q959.126 1328.49 962.506 1331.66 Q965.885 1334.83 965.885 1340.25 Q965.885 1345.83 962.413 1348.93 Q958.941 1352.01 952.621 1352.01 Q950.445 1352.01 948.177 1351.64 Q945.932 1351.27 943.524 1350.53 L943.524 1345.83 Q945.608 1346.96 947.83 1347.52 Q950.052 1348.07 952.529 1348.07 Q956.533 1348.07 958.871 1345.97 Q961.209 1343.86 961.209 1340.25 Q961.209 1336.64 958.871 1334.53 Q956.533 1332.42 952.529 1332.42 Q950.654 1332.42 948.779 1332.84 Q946.927 1333.26 944.983 1334.14 L944.983 1316.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M993.943 1429.17 L1001.58 1429.17 L1001.58 1402.81 L993.272 1404.48 L993.272 1400.22 L1001.54 1398.55 L1006.21 1398.55 L1006.21 1429.17 L1013.85 1429.17 L1013.85 1433.11 L993.943 1433.11 L993.943 1429.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1023.29 1427.23 L1028.18 1427.23 L1028.18 1433.11 L1023.29 1433.11 L1023.29 1427.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1048.36 1401.63 Q1044.75 1401.63 1042.92 1405.19 Q1041.12 1408.73 1041.12 1415.86 Q1041.12 1422.97 1042.92 1426.54 Q1044.75 1430.08 1048.36 1430.08 Q1052 1430.08 1053.8 1426.54 Q1055.63 1422.97 1055.63 1415.86 Q1055.63 1408.73 1053.8 1405.19 Q1052 1401.63 1048.36 1401.63 M1048.36 1397.92 Q1054.17 1397.92 1057.23 1402.53 Q1060.31 1407.11 1060.31 1415.86 Q1060.31 1424.59 1057.23 1429.2 Q1054.17 1433.78 1048.36 1433.78 Q1042.55 1433.78 1039.48 1429.2 Q1036.42 1424.59 1036.42 1415.86 Q1036.42 1407.11 1039.48 1402.53 Q1042.55 1397.92 1048.36 1397.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1130.76,1400.97 715.294,1041.17 715.294,321.568 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1294.3,1353.76 878.84,993.958 878.84,274.356 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1457.85,1306.55 1042.39,946.746 1042.39,227.145 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1621.39,1259.34 1205.93,899.535 1205.93,179.933 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1784.94,1212.12 1369.48,852.323 1369.48,132.721 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.05,1410.41 1817.65,1202.68 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1130.76,1400.97 1125.77,1396.65 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1294.3,1353.76 1289.32,1349.44 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1457.85,1306.55 1452.86,1302.23 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1621.39,1259.34 1616.41,1255.02 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1784.94,1212.12 1779.95,1207.81 "/>
-<path clip-path="url(#clip310)" d="M1138.74 1429.09 L1168.42 1429.09 L1168.42 1433.03 L1138.74 1433.03 L1138.74 1429.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1179.32 1441.99 L1186.96 1441.99 L1186.96 1415.62 L1178.65 1417.29 L1178.65 1413.03 L1186.92 1411.36 L1191.59 1411.36 L1191.59 1441.99 L1199.23 1441.99 L1199.23 1445.92 L1179.32 1445.92 L1179.32 1441.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1208.67 1440.04 L1213.56 1440.04 L1213.56 1445.92 L1208.67 1445.92 L1208.67 1440.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1233.74 1414.44 Q1230.13 1414.44 1228.3 1418 Q1226.5 1421.55 1226.5 1428.68 Q1226.5 1435.78 1228.3 1439.35 Q1230.13 1442.89 1233.74 1442.89 Q1237.38 1442.89 1239.18 1439.35 Q1241.01 1435.78 1241.01 1428.68 Q1241.01 1421.55 1239.18 1418 Q1237.38 1414.44 1233.74 1414.44 M1233.74 1410.74 Q1239.55 1410.74 1242.61 1415.34 Q1245.69 1419.93 1245.69 1428.68 Q1245.69 1437.4 1242.61 1442.01 Q1239.55 1446.59 1233.74 1446.59 Q1227.93 1446.59 1224.85 1442.01 Q1221.8 1437.4 1221.8 1428.68 Q1221.8 1419.93 1224.85 1415.34 Q1227.93 1410.74 1233.74 1410.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1302.29 1381.88 L1331.97 1381.88 L1331.97 1385.82 L1302.29 1385.82 L1302.29 1381.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1352.06 1367.23 Q1348.45 1367.23 1346.62 1370.79 Q1344.81 1374.33 1344.81 1381.46 Q1344.81 1388.57 1346.62 1392.14 Q1348.45 1395.68 1352.06 1395.68 Q1355.69 1395.68 1357.5 1392.14 Q1359.33 1388.57 1359.33 1381.46 Q1359.33 1374.33 1357.5 1370.79 Q1355.69 1367.23 1352.06 1367.23 M1352.06 1363.52 Q1357.87 1363.52 1360.92 1368.13 Q1364 1372.71 1364 1381.46 Q1364 1390.19 1360.92 1394.8 Q1357.87 1399.38 1352.06 1399.38 Q1346.25 1399.38 1343.17 1394.8 Q1340.11 1390.19 1340.11 1381.46 Q1340.11 1372.71 1343.17 1368.13 Q1346.25 1363.52 1352.06 1363.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1372.22 1392.83 L1377.1 1392.83 L1377.1 1398.71 L1372.22 1398.71 L1372.22 1392.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1387.34 1364.15 L1405.69 1364.15 L1405.69 1368.08 L1391.62 1368.08 L1391.62 1376.56 Q1392.64 1376.21 1393.66 1376.05 Q1394.67 1375.86 1395.69 1375.86 Q1401.48 1375.86 1404.86 1379.03 Q1408.24 1382.2 1408.24 1387.62 Q1408.24 1393.2 1404.77 1396.3 Q1401.29 1399.38 1394.97 1399.38 Q1392.8 1399.38 1390.53 1399.01 Q1388.28 1398.64 1385.88 1397.9 L1385.88 1393.2 Q1387.96 1394.33 1390.18 1394.89 Q1392.41 1395.45 1394.88 1395.45 Q1398.89 1395.45 1401.22 1393.34 Q1403.56 1391.23 1403.56 1387.62 Q1403.56 1384.01 1401.22 1381.9 Q1398.89 1379.8 1394.88 1379.8 Q1393.01 1379.8 1391.13 1380.21 Q1389.28 1380.63 1387.34 1381.51 L1387.34 1364.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1477.78 1320.02 Q1474.17 1320.02 1472.34 1323.58 Q1470.53 1327.12 1470.53 1334.25 Q1470.53 1341.36 1472.34 1344.92 Q1474.17 1348.47 1477.78 1348.47 Q1481.41 1348.47 1483.22 1344.92 Q1485.05 1341.36 1485.05 1334.25 Q1485.05 1327.12 1483.22 1323.58 Q1481.41 1320.02 1477.78 1320.02 M1477.78 1316.31 Q1483.59 1316.31 1486.65 1320.92 Q1489.72 1325.5 1489.72 1334.25 Q1489.72 1342.98 1486.65 1347.59 Q1483.59 1352.17 1477.78 1352.17 Q1471.97 1352.17 1468.89 1347.59 Q1465.84 1342.98 1465.84 1334.25 Q1465.84 1325.5 1468.89 1320.92 Q1471.97 1316.31 1477.78 1316.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1497.94 1345.62 L1502.83 1345.62 L1502.83 1351.5 L1497.94 1351.5 L1497.94 1345.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1523.01 1320.02 Q1519.4 1320.02 1517.57 1323.58 Q1515.77 1327.12 1515.77 1334.25 Q1515.77 1341.36 1517.57 1344.92 Q1519.4 1348.47 1523.01 1348.47 Q1526.65 1348.47 1528.45 1344.92 Q1530.28 1341.36 1530.28 1334.25 Q1530.28 1327.12 1528.45 1323.58 Q1526.65 1320.02 1523.01 1320.02 M1523.01 1316.31 Q1528.82 1316.31 1531.88 1320.92 Q1534.96 1325.5 1534.96 1334.25 Q1534.96 1342.98 1531.88 1347.59 Q1528.82 1352.17 1523.01 1352.17 Q1517.2 1352.17 1514.12 1347.59 Q1511.07 1342.98 1511.07 1334.25 Q1511.07 1325.5 1514.12 1320.92 Q1517.2 1316.31 1523.01 1316.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1641.33 1272.8 Q1637.71 1272.8 1635.89 1276.37 Q1634.08 1279.91 1634.08 1287.04 Q1634.08 1294.15 1635.89 1297.71 Q1637.71 1301.25 1641.33 1301.25 Q1644.96 1301.25 1646.77 1297.71 Q1648.59 1294.15 1648.59 1287.04 Q1648.59 1279.91 1646.77 1276.37 Q1644.96 1272.8 1641.33 1272.8 M1641.33 1269.1 Q1647.14 1269.1 1650.19 1273.71 Q1653.27 1278.29 1653.27 1287.04 Q1653.27 1295.77 1650.19 1300.37 Q1647.14 1304.96 1641.33 1304.96 Q1635.52 1304.96 1632.44 1300.37 Q1629.38 1295.77 1629.38 1287.04 Q1629.38 1278.29 1632.44 1273.71 Q1635.52 1269.1 1641.33 1269.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1661.49 1298.41 L1666.37 1298.41 L1666.37 1304.29 L1661.49 1304.29 L1661.49 1298.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1676.6 1269.73 L1694.96 1269.73 L1694.96 1273.66 L1680.89 1273.66 L1680.89 1282.13 Q1681.9 1281.79 1682.92 1281.62 Q1683.94 1281.44 1684.96 1281.44 Q1690.75 1281.44 1694.13 1284.61 Q1697.51 1287.78 1697.51 1293.2 Q1697.51 1298.78 1694.03 1301.88 Q1690.56 1304.96 1684.24 1304.96 Q1682.07 1304.96 1679.8 1304.59 Q1677.55 1304.22 1675.15 1303.48 L1675.15 1298.78 Q1677.23 1299.91 1679.45 1300.47 Q1681.67 1301.02 1684.15 1301.02 Q1688.15 1301.02 1690.49 1298.92 Q1692.83 1296.81 1692.83 1293.2 Q1692.83 1289.59 1690.49 1287.48 Q1688.15 1285.37 1684.15 1285.37 Q1682.27 1285.37 1680.4 1285.79 Q1678.55 1286.21 1676.6 1287.09 L1676.6 1269.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1793.6 1253.14 L1801.24 1253.14 L1801.24 1226.77 L1792.93 1228.44 L1792.93 1224.18 L1801.19 1222.51 L1805.87 1222.51 L1805.87 1253.14 L1813.51 1253.14 L1813.51 1257.07 L1793.6 1257.07 L1793.6 1253.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1822.95 1251.19 L1827.83 1251.19 L1827.83 1257.07 L1822.95 1257.07 L1822.95 1251.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1848.02 1225.59 Q1844.41 1225.59 1842.58 1229.16 Q1840.77 1232.7 1840.77 1239.83 Q1840.77 1246.94 1842.58 1250.5 Q1844.41 1254.04 1848.02 1254.04 Q1851.65 1254.04 1853.46 1250.5 Q1855.29 1246.94 1855.29 1239.83 Q1855.29 1232.7 1853.46 1229.16 Q1851.65 1225.59 1848.02 1225.59 M1848.02 1221.89 Q1853.83 1221.89 1856.89 1226.5 Q1859.96 1231.08 1859.96 1239.83 Q1859.96 1248.56 1856.89 1253.16 Q1853.83 1257.75 1848.02 1257.75 Q1842.21 1257.75 1839.13 1253.16 Q1836.08 1248.56 1836.08 1239.83 Q1836.08 1231.08 1839.13 1226.5 Q1842.21 1221.89 1848.02 1221.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,906.692 1402.19,698.96 1817.65,1058.76 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,618.851 1402.19,411.12 1817.65,770.921 "/>
-<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,331.01 1402.19,123.279 1817.65,483.08 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,1050.61 682.585,331.01 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,906.692 691.22,904.199 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,618.851 691.22,616.358 "/>
-<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,331.01 691.22,328.517 "/>
-<path clip-path="url(#clip310)" d="M593.299 907.143 L622.975 907.143 L622.975 911.078 L593.299 911.078 L593.299 907.143 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M633.877 920.037 L641.516 920.037 L641.516 893.671 L633.206 895.338 L633.206 891.078 L641.47 889.412 L646.146 889.412 L646.146 920.037 L653.785 920.037 L653.785 923.972 L633.877 923.972 L633.877 920.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M641.84 604.65 Q638.229 604.65 636.401 608.214 Q634.595 611.756 634.595 618.886 Q634.595 625.992 636.401 629.557 Q638.229 633.099 641.84 633.099 Q645.475 633.099 647.28 629.557 Q649.109 625.992 649.109 618.886 Q649.109 611.756 647.28 608.214 Q645.475 604.65 641.84 604.65 M641.84 600.946 Q647.651 600.946 650.706 605.552 Q653.785 610.136 653.785 618.886 Q653.785 627.612 650.706 632.219 Q647.651 636.802 641.84 636.802 Q636.03 636.802 632.952 632.219 Q629.896 627.612 629.896 618.886 Q629.896 610.136 632.952 605.552 Q636.03 600.946 641.84 600.946 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M633.877 344.355 L641.516 344.355 L641.516 317.989 L633.206 319.656 L633.206 315.397 L641.47 313.73 L646.146 313.73 L646.146 344.355 L653.785 344.355 L653.785 348.29 L633.877 348.29 L633.877 344.355 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip312)" d="M1250.12 694.886 L1008.51 390.241 L1275.58 1028.6 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip312)" d="M1275.58 1028.6 L1471.09 663.774 L1008.51 390.241 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip312)" d="M1250.12 694.886 L1275.58 1028.6 L1471.09 663.774 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip312)" d="M1250.12 694.886 L1471.09 663.774 L1008.51 390.241 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /></svg>
+<path clip-path="url(#clip132)" d="M734.179 1024.82 L734.179 370.632 L1388.36 181.785 L1766.06 508.877 L1766.06 1163.06 L1111.87 1351.91 L734.179 1024.82  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="701.469,1066.97 1421.07,859.235 1421.07,139.634 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="795.893,1148.74 1515.49,941.008 1515.49,221.407 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="890.316,1230.51 1609.92,1022.78 1609.92,303.179 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="984.739,1312.29 1704.34,1104.55 1704.34,384.952 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1079.16,1394.06 1798.76,1186.33 1798.76,466.725 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,1050.61 1098.05,1410.41 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="701.469,1066.97 710.105,1064.47 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="795.893,1148.74 804.528,1146.25 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="890.316,1230.51 898.951,1228.02 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="984.739,1312.29 993.374,1309.79 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1079.16,1394.06 1087.8,1391.57 "/>
+<path clip-path="url(#clip130)" d="M575.672 1089.19 L605.347 1089.19 L605.347 1093.12 L575.672 1093.12 L575.672 1089.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M616.25 1102.08 L623.889 1102.08 L623.889 1075.72 L615.579 1077.38 L615.579 1073.12 L623.843 1071.46 L628.518 1071.46 L628.518 1102.08 L636.157 1102.08 L636.157 1106.02 L616.25 1106.02 L616.25 1102.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M645.602 1100.14 L650.486 1100.14 L650.486 1106.02 L645.602 1106.02 L645.602 1100.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M670.671 1074.54 Q667.06 1074.54 665.231 1078.1 Q663.426 1081.64 663.426 1088.77 Q663.426 1095.88 665.231 1099.44 Q667.06 1102.98 670.671 1102.98 Q674.305 1102.98 676.111 1099.44 Q677.94 1095.88 677.94 1088.77 Q677.94 1081.64 676.111 1078.1 Q674.305 1074.54 670.671 1074.54 M670.671 1070.83 Q676.481 1070.83 679.537 1075.44 Q682.615 1080.02 682.615 1088.77 Q682.615 1097.5 679.537 1102.11 Q676.481 1106.69 670.671 1106.69 Q664.861 1106.69 661.782 1102.11 Q658.727 1097.5 658.727 1088.77 Q658.727 1080.02 661.782 1075.44 Q664.861 1070.83 670.671 1070.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M671.09 1170.96 L700.766 1170.96 L700.766 1174.9 L671.09 1174.9 L671.09 1170.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M720.858 1156.31 Q717.247 1156.31 715.419 1159.87 Q713.613 1163.42 713.613 1170.55 Q713.613 1177.65 715.419 1181.22 Q717.247 1184.76 720.858 1184.76 Q724.493 1184.76 726.298 1181.22 Q728.127 1177.65 728.127 1170.55 Q728.127 1163.42 726.298 1159.87 Q724.493 1156.31 720.858 1156.31 M720.858 1152.61 Q726.669 1152.61 729.724 1157.21 Q732.803 1161.8 732.803 1170.55 Q732.803 1179.27 729.724 1183.88 Q726.669 1188.46 720.858 1188.46 Q715.048 1188.46 711.97 1183.88 Q708.914 1179.27 708.914 1170.55 Q708.914 1161.8 711.97 1157.21 Q715.048 1152.61 720.858 1152.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M741.02 1181.91 L745.905 1181.91 L745.905 1187.79 L741.02 1187.79 L741.02 1181.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M756.136 1153.23 L774.492 1153.23 L774.492 1157.17 L760.418 1157.17 L760.418 1165.64 Q761.437 1165.29 762.455 1165.13 Q763.474 1164.94 764.492 1164.94 Q770.279 1164.94 773.659 1168.11 Q777.039 1171.29 777.039 1176.7 Q777.039 1182.28 773.566 1185.38 Q770.094 1188.46 763.775 1188.46 Q761.599 1188.46 759.33 1188.09 Q757.085 1187.72 754.678 1186.98 L754.678 1182.28 Q756.761 1183.42 758.983 1183.97 Q761.205 1184.53 763.682 1184.53 Q767.687 1184.53 770.025 1182.42 Q772.363 1180.31 772.363 1176.7 Q772.363 1173.09 770.025 1170.98 Q767.687 1168.88 763.682 1168.88 Q761.807 1168.88 759.932 1169.3 Q758.08 1169.71 756.136 1170.59 L756.136 1153.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M814.286 1238.08 Q810.675 1238.08 808.847 1241.65 Q807.041 1245.19 807.041 1252.32 Q807.041 1259.42 808.847 1262.99 Q810.675 1266.53 814.286 1266.53 Q817.921 1266.53 819.726 1262.99 Q821.555 1259.42 821.555 1252.32 Q821.555 1245.19 819.726 1241.65 Q817.921 1238.08 814.286 1238.08 M814.286 1234.38 Q820.096 1234.38 823.152 1238.98 Q826.231 1243.57 826.231 1252.32 Q826.231 1261.04 823.152 1265.65 Q820.096 1270.23 814.286 1270.23 Q808.476 1270.23 805.397 1265.65 Q802.342 1261.04 802.342 1252.32 Q802.342 1243.57 805.397 1238.98 Q808.476 1234.38 814.286 1234.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M834.448 1263.68 L839.332 1263.68 L839.332 1269.56 L834.448 1269.56 L834.448 1263.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M859.518 1238.08 Q855.906 1238.08 854.078 1241.65 Q852.272 1245.19 852.272 1252.32 Q852.272 1259.42 854.078 1262.99 Q855.906 1266.53 859.518 1266.53 Q863.152 1266.53 864.957 1262.99 Q866.786 1259.42 866.786 1252.32 Q866.786 1245.19 864.957 1241.65 Q863.152 1238.08 859.518 1238.08 M859.518 1234.38 Q865.328 1234.38 868.383 1238.98 Q871.462 1243.57 871.462 1252.32 Q871.462 1261.04 868.383 1265.65 Q865.328 1270.23 859.518 1270.23 Q853.707 1270.23 850.629 1265.65 Q847.573 1261.04 847.573 1252.32 Q847.573 1243.57 850.629 1238.98 Q853.707 1234.38 859.518 1234.38 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M909.705 1319.85 Q906.094 1319.85 904.265 1323.42 Q902.46 1326.96 902.46 1334.09 Q902.46 1341.2 904.265 1344.76 Q906.094 1348.3 909.705 1348.3 Q913.339 1348.3 915.145 1344.76 Q916.973 1341.2 916.973 1334.09 Q916.973 1326.96 915.145 1323.42 Q913.339 1319.85 909.705 1319.85 M909.705 1316.15 Q915.515 1316.15 918.571 1320.76 Q921.649 1325.34 921.649 1334.09 Q921.649 1342.82 918.571 1347.42 Q915.515 1352.01 909.705 1352.01 Q903.895 1352.01 900.816 1347.42 Q897.761 1342.82 897.761 1334.09 Q897.761 1325.34 900.816 1320.76 Q903.895 1316.15 909.705 1316.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M929.867 1345.46 L934.751 1345.46 L934.751 1351.34 L929.867 1351.34 L929.867 1345.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M944.983 1316.78 L963.339 1316.78 L963.339 1320.71 L949.265 1320.71 L949.265 1329.18 Q950.283 1328.84 951.302 1328.67 Q952.32 1328.49 953.339 1328.49 Q959.126 1328.49 962.506 1331.66 Q965.885 1334.83 965.885 1340.25 Q965.885 1345.83 962.413 1348.93 Q958.941 1352.01 952.621 1352.01 Q950.445 1352.01 948.177 1351.64 Q945.932 1351.27 943.524 1350.53 L943.524 1345.83 Q945.608 1346.96 947.83 1347.52 Q950.052 1348.07 952.529 1348.07 Q956.533 1348.07 958.871 1345.97 Q961.209 1343.86 961.209 1340.25 Q961.209 1336.64 958.871 1334.53 Q956.533 1332.42 952.529 1332.42 Q950.654 1332.42 948.779 1332.84 Q946.927 1333.26 944.983 1334.14 L944.983 1316.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M993.943 1429.17 L1001.58 1429.17 L1001.58 1402.81 L993.272 1404.48 L993.272 1400.22 L1001.54 1398.55 L1006.21 1398.55 L1006.21 1429.17 L1013.85 1429.17 L1013.85 1433.11 L993.943 1433.11 L993.943 1429.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1023.29 1427.23 L1028.18 1427.23 L1028.18 1433.11 L1023.29 1433.11 L1023.29 1427.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1048.36 1401.63 Q1044.75 1401.63 1042.92 1405.19 Q1041.12 1408.73 1041.12 1415.86 Q1041.12 1422.97 1042.92 1426.54 Q1044.75 1430.08 1048.36 1430.08 Q1052 1430.08 1053.8 1426.54 Q1055.63 1422.97 1055.63 1415.86 Q1055.63 1408.73 1053.8 1405.19 Q1052 1401.63 1048.36 1401.63 M1048.36 1397.92 Q1054.17 1397.92 1057.23 1402.53 Q1060.31 1407.11 1060.31 1415.86 Q1060.31 1424.59 1057.23 1429.2 Q1054.17 1433.78 1048.36 1433.78 Q1042.55 1433.78 1039.48 1429.2 Q1036.42 1424.59 1036.42 1415.86 Q1036.42 1407.11 1039.48 1402.53 Q1042.55 1397.92 1048.36 1397.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1130.76,1400.97 715.294,1041.17 715.294,321.568 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1294.3,1353.76 878.84,993.958 878.84,274.356 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1457.85,1306.55 1042.39,946.746 1042.39,227.145 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1621.39,1259.34 1205.93,899.535 1205.93,179.933 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1784.94,1212.12 1369.48,852.323 1369.48,132.721 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1098.05,1410.41 1817.65,1202.68 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1130.76,1400.97 1125.77,1396.65 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1294.3,1353.76 1289.32,1349.44 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1457.85,1306.55 1452.86,1302.23 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1621.39,1259.34 1616.41,1255.02 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1784.94,1212.12 1779.95,1207.81 "/>
+<path clip-path="url(#clip130)" d="M1138.74 1429.09 L1168.42 1429.09 L1168.42 1433.03 L1138.74 1433.03 L1138.74 1429.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1179.32 1441.99 L1186.96 1441.99 L1186.96 1415.62 L1178.65 1417.29 L1178.65 1413.03 L1186.92 1411.36 L1191.59 1411.36 L1191.59 1441.99 L1199.23 1441.99 L1199.23 1445.92 L1179.32 1445.92 L1179.32 1441.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1208.67 1440.04 L1213.56 1440.04 L1213.56 1445.92 L1208.67 1445.92 L1208.67 1440.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1233.74 1414.44 Q1230.13 1414.44 1228.3 1418 Q1226.5 1421.55 1226.5 1428.68 Q1226.5 1435.78 1228.3 1439.35 Q1230.13 1442.89 1233.74 1442.89 Q1237.38 1442.89 1239.18 1439.35 Q1241.01 1435.78 1241.01 1428.68 Q1241.01 1421.55 1239.18 1418 Q1237.38 1414.44 1233.74 1414.44 M1233.74 1410.74 Q1239.55 1410.74 1242.61 1415.34 Q1245.69 1419.93 1245.69 1428.68 Q1245.69 1437.4 1242.61 1442.01 Q1239.55 1446.59 1233.74 1446.59 Q1227.93 1446.59 1224.85 1442.01 Q1221.8 1437.4 1221.8 1428.68 Q1221.8 1419.93 1224.85 1415.34 Q1227.93 1410.74 1233.74 1410.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1302.29 1381.88 L1331.97 1381.88 L1331.97 1385.82 L1302.29 1385.82 L1302.29 1381.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1352.06 1367.23 Q1348.45 1367.23 1346.62 1370.79 Q1344.81 1374.33 1344.81 1381.46 Q1344.81 1388.57 1346.62 1392.14 Q1348.45 1395.68 1352.06 1395.68 Q1355.69 1395.68 1357.5 1392.14 Q1359.33 1388.57 1359.33 1381.46 Q1359.33 1374.33 1357.5 1370.79 Q1355.69 1367.23 1352.06 1367.23 M1352.06 1363.52 Q1357.87 1363.52 1360.92 1368.13 Q1364 1372.71 1364 1381.46 Q1364 1390.19 1360.92 1394.8 Q1357.87 1399.38 1352.06 1399.38 Q1346.25 1399.38 1343.17 1394.8 Q1340.11 1390.19 1340.11 1381.46 Q1340.11 1372.71 1343.17 1368.13 Q1346.25 1363.52 1352.06 1363.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1372.22 1392.83 L1377.1 1392.83 L1377.1 1398.71 L1372.22 1398.71 L1372.22 1392.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1387.34 1364.15 L1405.69 1364.15 L1405.69 1368.08 L1391.62 1368.08 L1391.62 1376.56 Q1392.64 1376.21 1393.66 1376.05 Q1394.67 1375.86 1395.69 1375.86 Q1401.48 1375.86 1404.86 1379.03 Q1408.24 1382.2 1408.24 1387.62 Q1408.24 1393.2 1404.77 1396.3 Q1401.29 1399.38 1394.97 1399.38 Q1392.8 1399.38 1390.53 1399.01 Q1388.28 1398.64 1385.88 1397.9 L1385.88 1393.2 Q1387.96 1394.33 1390.18 1394.89 Q1392.41 1395.45 1394.88 1395.45 Q1398.89 1395.45 1401.22 1393.34 Q1403.56 1391.23 1403.56 1387.62 Q1403.56 1384.01 1401.22 1381.9 Q1398.89 1379.8 1394.88 1379.8 Q1393.01 1379.8 1391.13 1380.21 Q1389.28 1380.63 1387.34 1381.51 L1387.34 1364.15 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1477.78 1320.02 Q1474.17 1320.02 1472.34 1323.58 Q1470.53 1327.12 1470.53 1334.25 Q1470.53 1341.36 1472.34 1344.92 Q1474.17 1348.47 1477.78 1348.47 Q1481.41 1348.47 1483.22 1344.92 Q1485.05 1341.36 1485.05 1334.25 Q1485.05 1327.12 1483.22 1323.58 Q1481.41 1320.02 1477.78 1320.02 M1477.78 1316.31 Q1483.59 1316.31 1486.65 1320.92 Q1489.72 1325.5 1489.72 1334.25 Q1489.72 1342.98 1486.65 1347.59 Q1483.59 1352.17 1477.78 1352.17 Q1471.97 1352.17 1468.89 1347.59 Q1465.84 1342.98 1465.84 1334.25 Q1465.84 1325.5 1468.89 1320.92 Q1471.97 1316.31 1477.78 1316.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1497.94 1345.62 L1502.83 1345.62 L1502.83 1351.5 L1497.94 1351.5 L1497.94 1345.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1523.01 1320.02 Q1519.4 1320.02 1517.57 1323.58 Q1515.77 1327.12 1515.77 1334.25 Q1515.77 1341.36 1517.57 1344.92 Q1519.4 1348.47 1523.01 1348.47 Q1526.65 1348.47 1528.45 1344.92 Q1530.28 1341.36 1530.28 1334.25 Q1530.28 1327.12 1528.45 1323.58 Q1526.65 1320.02 1523.01 1320.02 M1523.01 1316.31 Q1528.82 1316.31 1531.88 1320.92 Q1534.96 1325.5 1534.96 1334.25 Q1534.96 1342.98 1531.88 1347.59 Q1528.82 1352.17 1523.01 1352.17 Q1517.2 1352.17 1514.12 1347.59 Q1511.07 1342.98 1511.07 1334.25 Q1511.07 1325.5 1514.12 1320.92 Q1517.2 1316.31 1523.01 1316.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1641.33 1272.8 Q1637.71 1272.8 1635.89 1276.37 Q1634.08 1279.91 1634.08 1287.04 Q1634.08 1294.15 1635.89 1297.71 Q1637.71 1301.25 1641.33 1301.25 Q1644.96 1301.25 1646.77 1297.71 Q1648.59 1294.15 1648.59 1287.04 Q1648.59 1279.91 1646.77 1276.37 Q1644.96 1272.8 1641.33 1272.8 M1641.33 1269.1 Q1647.14 1269.1 1650.19 1273.71 Q1653.27 1278.29 1653.27 1287.04 Q1653.27 1295.77 1650.19 1300.37 Q1647.14 1304.96 1641.33 1304.96 Q1635.52 1304.96 1632.44 1300.37 Q1629.38 1295.77 1629.38 1287.04 Q1629.38 1278.29 1632.44 1273.71 Q1635.52 1269.1 1641.33 1269.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1661.49 1298.41 L1666.37 1298.41 L1666.37 1304.29 L1661.49 1304.29 L1661.49 1298.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1676.6 1269.73 L1694.96 1269.73 L1694.96 1273.66 L1680.89 1273.66 L1680.89 1282.13 Q1681.9 1281.79 1682.92 1281.62 Q1683.94 1281.44 1684.96 1281.44 Q1690.75 1281.44 1694.13 1284.61 Q1697.51 1287.78 1697.51 1293.2 Q1697.51 1298.78 1694.03 1301.88 Q1690.56 1304.96 1684.24 1304.96 Q1682.07 1304.96 1679.8 1304.59 Q1677.55 1304.22 1675.15 1303.48 L1675.15 1298.78 Q1677.23 1299.91 1679.45 1300.47 Q1681.67 1301.02 1684.15 1301.02 Q1688.15 1301.02 1690.49 1298.92 Q1692.83 1296.81 1692.83 1293.2 Q1692.83 1289.59 1690.49 1287.48 Q1688.15 1285.37 1684.15 1285.37 Q1682.27 1285.37 1680.4 1285.79 Q1678.55 1286.21 1676.6 1287.09 L1676.6 1269.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1793.6 1253.14 L1801.24 1253.14 L1801.24 1226.77 L1792.93 1228.44 L1792.93 1224.18 L1801.19 1222.51 L1805.87 1222.51 L1805.87 1253.14 L1813.51 1253.14 L1813.51 1257.07 L1793.6 1257.07 L1793.6 1253.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1822.95 1251.19 L1827.83 1251.19 L1827.83 1257.07 L1822.95 1257.07 L1822.95 1251.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M1848.02 1225.59 Q1844.41 1225.59 1842.58 1229.16 Q1840.77 1232.7 1840.77 1239.83 Q1840.77 1246.94 1842.58 1250.5 Q1844.41 1254.04 1848.02 1254.04 Q1851.65 1254.04 1853.46 1250.5 Q1855.29 1246.94 1855.29 1239.83 Q1855.29 1232.7 1853.46 1229.16 Q1851.65 1225.59 1848.02 1225.59 M1848.02 1221.89 Q1853.83 1221.89 1856.89 1226.5 Q1859.96 1231.08 1859.96 1239.83 Q1859.96 1248.56 1856.89 1253.16 Q1853.83 1257.75 1848.02 1257.75 Q1842.21 1257.75 1839.13 1253.16 Q1836.08 1248.56 1836.08 1239.83 Q1836.08 1231.08 1839.13 1226.5 Q1842.21 1221.89 1848.02 1221.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,906.692 1402.19,698.96 1817.65,1058.76 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,618.851 1402.19,411.12 1817.65,770.921 "/>
+<polyline clip-path="url(#clip132)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.585,331.01 1402.19,123.279 1817.65,483.08 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,1050.61 682.585,331.01 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,906.692 691.22,904.199 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,618.851 691.22,616.358 "/>
+<polyline clip-path="url(#clip130)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.585,331.01 691.22,328.517 "/>
+<path clip-path="url(#clip130)" d="M593.299 907.143 L622.975 907.143 L622.975 911.078 L593.299 911.078 L593.299 907.143 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M633.877 920.037 L641.516 920.037 L641.516 893.671 L633.206 895.338 L633.206 891.078 L641.47 889.412 L646.146 889.412 L646.146 920.037 L653.785 920.037 L653.785 923.972 L633.877 923.972 L633.877 920.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M641.84 604.65 Q638.229 604.65 636.401 608.214 Q634.595 611.756 634.595 618.886 Q634.595 625.992 636.401 629.557 Q638.229 633.099 641.84 633.099 Q645.475 633.099 647.28 629.557 Q649.109 625.992 649.109 618.886 Q649.109 611.756 647.28 608.214 Q645.475 604.65 641.84 604.65 M641.84 600.946 Q647.651 600.946 650.706 605.552 Q653.785 610.136 653.785 618.886 Q653.785 627.612 650.706 632.219 Q647.651 636.802 641.84 636.802 Q636.03 636.802 632.952 632.219 Q629.896 627.612 629.896 618.886 Q629.896 610.136 632.952 605.552 Q636.03 600.946 641.84 600.946 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip130)" d="M633.877 344.355 L641.516 344.355 L641.516 317.989 L633.206 319.656 L633.206 315.397 L641.47 313.73 L646.146 313.73 L646.146 344.355 L653.785 344.355 L653.785 348.29 L633.877 348.29 L633.877 344.355 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip132)" d="M1250.12 694.886 L1008.51 390.241 L1275.58 1028.6 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip132)" d="M1275.58 1028.6 L1471.09 663.774 L1008.51 390.241 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip132)" d="M1250.12 694.886 L1275.58 1028.6 L1471.09 663.774 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /><path clip-path="url(#clip132)" d="M1250.12 694.886 L1471.09 663.774 L1008.51 390.241 Z" fill="#009afa" fill-rule="evenodd" fill-opacity="0.101961" stroke="#009af9" stroke-opacity="0.101961" stroke-linejoin="round" stroke-width="9.6" /></svg>
diff --git a/previews/PR3562/tutorials/conic/simple_examples/index.html b/previews/PR3562/tutorials/conic/simple_examples/index.html
index 47d6b248377..d160e754dc3 100644
--- a/previews/PR3562/tutorials/conic/simple_examples/index.html
+++ b/previews/PR3562/tutorials/conic/simple_examples/index.html
@@ -126,8 +126,8 @@
     return
 end
 
-example_correlation_problem()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">An upper bound for ρ_AC is 0.8719220302818699
-A lower bound for ρ_AC is -0.9779989594121374</code></pre><h2 id="The-minimum-distortion-problem"><a class="docs-heading-anchor" href="#The-minimum-distortion-problem">The minimum distortion problem</a><a id="The-minimum-distortion-problem-1"></a><a class="docs-heading-anchor-permalink" href="#The-minimum-distortion-problem" title="Permalink"></a></h2><p>This example arises from computational geometry, in particular the problem of embedding a general finite metric space into a Euclidean space.</p><p>It is known that the 4-point metric space defined by the star graph</p><pre><code class="language-raw hljs">  [1]
+example_correlation_problem()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">An upper bound for ρ_AC is 0.8719220302963583
+A lower bound for ρ_AC is -0.9779989594132403</code></pre><h2 id="The-minimum-distortion-problem"><a class="docs-heading-anchor" href="#The-minimum-distortion-problem">The minimum distortion problem</a><a id="The-minimum-distortion-problem-1"></a><a class="docs-heading-anchor-permalink" href="#The-minimum-distortion-problem" title="Permalink"></a></h2><p>This example arises from computational geometry, in particular the problem of embedding a general finite metric space into a Euclidean space.</p><p>It is known that the 4-point metric space defined by the star graph</p><pre><code class="language-raw hljs">  [1]
     \
      1
       \
@@ -176,7 +176,7 @@
     )
 end
 
-example_minimum_distortion()</code></pre><img src="c9b2b8b7.svg" alt="Example block output"/><h2 id="Lovász-numbers"><a class="docs-heading-anchor" href="#Lovász-numbers">Lovász numbers</a><a id="Lovász-numbers-1"></a><a class="docs-heading-anchor-permalink" href="#Lovász-numbers" title="Permalink"></a></h2><p>The <a href="https://en.wikipedia.org/wiki/Lov%C3%A1sz_number">Lovász number</a> of a graph, also known as Lovász&#39;s theta-function, is a number that lies between two important and related numbers that are computationally hard to determine, namely the chromatic and clique numbers of the graph. It is possible however to efficient compute the Lovász number as the optimal value of a semidefinite program.</p><p>Consider the pentagon graph:</p><pre><code class="language-raw hljs">     [5]
+example_minimum_distortion()</code></pre><img src="4350a7ce.svg" alt="Example block output"/><h2 id="Lovász-numbers"><a class="docs-heading-anchor" href="#Lovász-numbers">Lovász numbers</a><a id="Lovász-numbers-1"></a><a class="docs-heading-anchor-permalink" href="#Lovász-numbers" title="Permalink"></a></h2><p>The <a href="https://en.wikipedia.org/wiki/Lov%C3%A1sz_number">Lovász number</a> of a graph, also known as Lovász&#39;s theta-function, is a number that lies between two important and related numbers that are computationally hard to determine, namely the chromatic and clique numbers of the graph. It is possible however to efficient compute the Lovász number as the optimal value of a semidefinite program.</p><p>Consider the pentagon graph:</p><pre><code class="language-raw hljs">     [5]
     /   \
    /     \
  [1]     [4]
@@ -213,7 +213,7 @@
     return
 end
 
-example_theta_problem()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">The Lovász number is: 2.2360678617949126</code></pre><h2 id="Robust-uncertainty-sets"><a class="docs-heading-anchor" href="#Robust-uncertainty-sets">Robust uncertainty sets</a><a id="Robust-uncertainty-sets-1"></a><a class="docs-heading-anchor-permalink" href="#Robust-uncertainty-sets" title="Permalink"></a></h2><p>This example computes the Value at Risk for a data-driven uncertainty set. Closed-form expressions for the optimal value are available. For more details, see:</p><p>Bertsimas, D., Gupta, V., &amp; Kallus, N. (2018). <a href="https://doi.org/10.1007/s10107-017-1125-8"><em>Data-driven robust optimization.</em></a> Mathematical Programming, 167, 235-292.</p><pre><code class="language-julia hljs">function example_robust_uncertainty_sets()
+example_theta_problem()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">The Lovász number is: 2.236067861794848</code></pre><h2 id="Robust-uncertainty-sets"><a class="docs-heading-anchor" href="#Robust-uncertainty-sets">Robust uncertainty sets</a><a id="Robust-uncertainty-sets-1"></a><a class="docs-heading-anchor-permalink" href="#Robust-uncertainty-sets" title="Permalink"></a></h2><p>This example computes the Value at Risk for a data-driven uncertainty set. Closed-form expressions for the optimal value are available. For more details, see:</p><p>Bertsimas, D., Gupta, V., &amp; Kallus, N. (2018). <a href="https://doi.org/10.1007/s10107-017-1125-8"><em>Data-driven robust optimization.</em></a> Mathematical Programming, 167, 235-292.</p><pre><code class="language-julia hljs">function example_robust_uncertainty_sets()
     R, d, 𝛿, ɛ = 1, 3, 0.05, 0.05
     N = ceil((2 + 2 * log(2 / 𝛿))^2) + 1
     c, μhat, M = randn(d), rand(d), rand(d, d)
@@ -239,4 +239,4 @@
     return
 end
 
-example_robust_uncertainty_sets()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and Tricks</a><a class="docs-footer-nextpage" href="../dualization/">Dualization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+example_robust_uncertainty_sets()</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and Tricks</a><a class="docs-footer-nextpage" href="../dualization/">Dualization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/start_values/index.html b/previews/PR3562/tutorials/conic/start_values/index.html
index 3d5cbda45c5..de9b19cd289 100644
--- a/previews/PR3562/tutorials/conic/start_values/index.html
+++ b/previews/PR3562/tutorials/conic/start_values/index.html
@@ -50,12 +50,12 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 4.42e+01  1.00e+00  1.28e+02 -6.64e+01  1.00e-01  4.95e-05
-    75| 5.30e-07  2.63e-06  3.15e-07 -3.00e+00  1.00e-01  8.36e-05
+     0| 4.42e+01  1.00e+00  1.28e+02 -6.64e+01  1.00e-01  7.33e-05
+    75| 5.30e-07  2.63e-06  3.15e-07 -3.00e+00  1.00e-01  1.36e-04
 ------------------------------------------------------------------
 status:  solved
-timings: total: 8.45e-05s = setup: 3.83e-05s + solve: 4.62e-05s
-	 lin-sys: 1.02e-05s, cones: 5.44e-06s, accel: 3.02e-06s
+timings: total: 1.37e-04s = setup: 5.69e-05s + solve: 7.96e-05s
+	 lin-sys: 1.59e-05s, cones: 7.60e-06s, accel: 4.70e-06s
 ------------------------------------------------------------------
 objective = -2.999998
 ------------------------------------------------------------------</code></pre><p>By looking at the log, we can see that SCS took 75 iterations to find the optimal solution. Now we set the optimal solution as our starting point:</p><pre><code class="language-julia hljs">set_optimal_start_values(model)</code></pre><p>and we re-optimize:</p><pre><code class="language-julia hljs">optimize!(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">------------------------------------------------------------------
@@ -73,11 +73,11 @@
 ------------------------------------------------------------------
  iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)
 ------------------------------------------------------------------
-     0| 1.90e-05  1.56e-06  9.14e-05 -3.00e+00  1.00e-01  5.06e-05
+     0| 1.90e-05  1.56e-06  9.14e-05 -3.00e+00  1.00e-01  7.14e-05
 ------------------------------------------------------------------
 status:  solved
-timings: total: 5.13e-05s = setup: 3.90e-05s + solve: 1.24e-05s
-	 lin-sys: 7.61e-07s, cones: 1.42e-06s, accel: 2.00e-08s
+timings: total: 7.24e-05s = setup: 5.53e-05s + solve: 1.71e-05s
+	 lin-sys: 1.20e-06s, cones: 1.50e-06s, accel: 0.00e+00s
 ------------------------------------------------------------------
 objective = -3.000044
-------------------------------------------------------------------</code></pre><p>Now the optimization terminates after 0 iterations because our starting point is already optimal.</p><p>Note that some solvers do not support setting some parts of the starting solution, for example, they may support only <code>set_start_value</code> for variables. If you encounter an <code>UnsupportedSupported</code> attribute error for <a href="../../../moi/reference/variables/#MathOptInterface.VariablePrimalStart"><code>MOI.VariablePrimalStart</code></a>, <a href="../../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>, or <a href="../../../moi/reference/constraints/#MathOptInterface.ConstraintDualStart"><code>MOI.ConstraintDualStart</code></a>, comment out the corresponding part of the <code>set_optimal_start_values</code> function.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and Tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+------------------------------------------------------------------</code></pre><p>Now the optimization terminates after 0 iterations because our starting point is already optimal.</p><p>Note that some solvers do not support setting some parts of the starting solution, for example, they may support only <code>set_start_value</code> for variables. If you encounter an <code>UnsupportedSupported</code> attribute error for <a href="../../../moi/reference/variables/#MathOptInterface.VariablePrimalStart"><code>MOI.VariablePrimalStart</code></a>, <a href="../../../moi/reference/constraints/#MathOptInterface.ConstraintPrimalStart"><code>MOI.ConstraintPrimalStart</code></a>, or <a href="../../../moi/reference/constraints/#MathOptInterface.ConstraintDualStart"><code>MOI.ConstraintDualStart</code></a>, comment out the corresponding part of the <code>set_optimal_start_values</code> function.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and Tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/conic/tips_and_tricks/index.html b/previews/PR3562/tutorials/conic/tips_and_tricks/index.html
index 8b79691aed0..7c3fa4d89af 100644
--- a/previews/PR3562/tutorials/conic/tips_and_tricks/index.html
+++ b/previews/PR3562/tutorials/conic/tips_and_tricks/index.html
@@ -20,7 +20,7 @@
 @constraint(model, [t; x] in SecondOrderCone())
 @objective(model, Min, t)
 optimize!(model)
-value(t), value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.5773503148525415, [0.3333333333638126, 0.3333333333638124, 0.3333333333638124])</code></pre><h2 id="Rotated-Second-Order-Cone"><a class="docs-heading-anchor" href="#Rotated-Second-Order-Cone">Rotated Second-Order Cone</a><a id="Rotated-Second-Order-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Rotated-Second-Order-Cone" title="Permalink"></a></h2><p>A Second-Order Cone rotated by <span>$\pi/4$</span> in the <span>$(x_1,x_2)$</span> plane is called a <a href="../../../api/JuMP/#JuMP.RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a>. It is a cone of the form:</p><p class="math-container">\[K_{rsoc} = \{ (t,u,x) \in \mathbb{R}^n : 2tu \ge ||x||_2^2, t,u \ge 0 \}\]</p><p>When <code>u = 0.5</code>, it represents the sum of squares of a vector <span>$x$</span>:</p><pre><code class="language-julia hljs">data = [1.0, 2.0, 3.0, 4.0]
+value(t), value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.5773503148525505, [0.3333333333638094, 0.33333333336381227, 0.33333333336381227])</code></pre><h2 id="Rotated-Second-Order-Cone"><a class="docs-heading-anchor" href="#Rotated-Second-Order-Cone">Rotated Second-Order Cone</a><a id="Rotated-Second-Order-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Rotated-Second-Order-Cone" title="Permalink"></a></h2><p>A Second-Order Cone rotated by <span>$\pi/4$</span> in the <span>$(x_1,x_2)$</span> plane is called a <a href="../../../api/JuMP/#JuMP.RotatedSecondOrderCone"><code>RotatedSecondOrderCone</code></a>. It is a cone of the form:</p><p class="math-container">\[K_{rsoc} = \{ (t,u,x) \in \mathbb{R}^n : 2tu \ge ||x||_2^2, t,u \ge 0 \}\]</p><p>When <code>u = 0.5</code>, it represents the sum of squares of a vector <span>$x$</span>:</p><pre><code class="language-julia hljs">data = [1.0, 2.0, 3.0, 4.0]
 target = [0.45, 1.04, 1.51, 1.97]
 model = Model(SCS.Optimizer)
 set_silent(model)
@@ -30,21 +30,21 @@
 @constraint(model, [t; 0.5; residuals] in RotatedSecondOrderCone())
 @objective(model, Min, t)
 optimize!(model)
-value(θ), value(t)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.49800000000000544, 0.004979422159618532)</code></pre><h2 id="Exponential-Cone"><a class="docs-heading-anchor" href="#Exponential-Cone">Exponential Cone</a><a id="Exponential-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Exponential-Cone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.ExponentialCone"><code>MOI.ExponentialCone</code></a> is a set of the form:</p><p class="math-container">\[K_{exp} = \{ (x,y,z) \in \mathbb{R}^3 : y \exp (x/y) \le z, y &gt; 0 \}\]</p><p>It can be used to model problems involving <code>log</code> and <code>exp</code>.</p><h3 id="Exponential"><a class="docs-heading-anchor" href="#Exponential">Exponential</a><a id="Exponential-1"></a><a class="docs-heading-anchor-permalink" href="#Exponential" title="Permalink"></a></h3><p>To model <span>$\exp(x) \le z$</span>, use <code>(x, 1, z)</code>:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(θ), value(t)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.4980000000000055, 0.004979422159620655)</code></pre><h2 id="Exponential-Cone"><a class="docs-heading-anchor" href="#Exponential-Cone">Exponential Cone</a><a id="Exponential-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Exponential-Cone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.ExponentialCone"><code>MOI.ExponentialCone</code></a> is a set of the form:</p><p class="math-container">\[K_{exp} = \{ (x,y,z) \in \mathbb{R}^3 : y \exp (x/y) \le z, y &gt; 0 \}\]</p><p>It can be used to model problems involving <code>log</code> and <code>exp</code>.</p><h3 id="Exponential"><a class="docs-heading-anchor" href="#Exponential">Exponential</a><a id="Exponential-1"></a><a class="docs-heading-anchor-permalink" href="#Exponential" title="Permalink"></a></h3><p>To model <span>$\exp(x) \le z$</span>, use <code>(x, 1, z)</code>:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, x == 1.5)
 @variable(model, z)
 @objective(model, Min, z)
 @constraint(model, [x, 1, z] in MOI.ExponentialCone())
 optimize!(model)
-value(z), exp(1.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(4.481654619603659, 4.4816890703380645)</code></pre><h3 id="Logarithm"><a class="docs-heading-anchor" href="#Logarithm">Logarithm</a><a id="Logarithm-1"></a><a class="docs-heading-anchor-permalink" href="#Logarithm" title="Permalink"></a></h3><p>To model <span>$x \le \log(z)$</span>, use <code>(x, 1, z)</code>:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(z), exp(1.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(4.481654619603649, 4.4816890703380645)</code></pre><h3 id="Logarithm"><a class="docs-heading-anchor" href="#Logarithm">Logarithm</a><a id="Logarithm-1"></a><a class="docs-heading-anchor-permalink" href="#Logarithm" title="Permalink"></a></h3><p>To model <span>$x \le \log(z)$</span>, use <code>(x, 1, z)</code>:</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, x)
 @variable(model, z == 1.5)
 @objective(model, Max, x)
 @constraint(model, [x, 1, z] in MOI.ExponentialCone())
 optimize!(model)
-value(x), log(1.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.4055956586655673, 0.4054651081081644)</code></pre><h3 id="Log-sum-exp"><a class="docs-heading-anchor" href="#Log-sum-exp">Log-sum-exp</a><a id="Log-sum-exp-1"></a><a class="docs-heading-anchor-permalink" href="#Log-sum-exp" title="Permalink"></a></h3><p>To model <span>$t \ge \log\left(\sum e^{x_i}\right)$</span>, use:</p><pre><code class="language-julia hljs">N = 3
+value(x), log(1.5)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.4055956586655662, 0.4054651081081644)</code></pre><h3 id="Log-sum-exp"><a class="docs-heading-anchor" href="#Log-sum-exp">Log-sum-exp</a><a id="Log-sum-exp-1"></a><a class="docs-heading-anchor-permalink" href="#Log-sum-exp" title="Permalink"></a></h3><p>To model <span>$t \ge \log\left(\sum e^{x_i}\right)$</span>, use:</p><pre><code class="language-julia hljs">N = 3
 x0 = rand(N)
 model = Model(SCS.Optimizer)
 set_silent(model)
@@ -55,7 +55,7 @@
 @constraint(model, sum(u) &lt;= 1)
 @constraint(model, [i = 1:N], [x[i] - t, 1, u[i]] in MOI.ExponentialCone())
 optimize!(model)
-value(t), log(sum(exp.(x0)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.4726849472138905, 1.472772274699489)</code></pre><h3 id="Entropy"><a class="docs-heading-anchor" href="#Entropy">Entropy</a><a id="Entropy-1"></a><a class="docs-heading-anchor-permalink" href="#Entropy" title="Permalink"></a></h3><p>The entropy maximization problem consists of maximizing the entropy function, <span>$H(x) = -x\log{x}$</span> subject to linear inequality constraints.</p><p class="math-container">\[\begin{aligned}
+value(t), log(sum(exp.(x0)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.4726849472138828, 1.472772274699489)</code></pre><h3 id="Entropy"><a class="docs-heading-anchor" href="#Entropy">Entropy</a><a id="Entropy-1"></a><a class="docs-heading-anchor-permalink" href="#Entropy" title="Permalink"></a></h3><p>The entropy maximization problem consists of maximizing the entropy function, <span>$H(x) = -x\log{x}$</span> subject to linear inequality constraints.</p><p class="math-container">\[\begin{aligned}
 &amp; \max &amp; - \sum_{i=1}^n x_i \log x_i \\
 &amp; \;\;\text{s.t.} &amp; \mathbf{1}^\top x = 1 \\
 &amp; &amp; Ax \leq b
@@ -75,7 +75,7 @@
 @constraint(model, A * x .&lt;= b)
 @constraint(model, [i = 1:n], [t[i], x[i], 1] in MOI.ExponentialCone())
 optimize!(model)
-objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2.6632061721520914</code></pre><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.ExponentialCone"><code>MOI.ExponentialCone</code></a> has a dual, the <a href="../../../moi/reference/standard_form/#MathOptInterface.DualExponentialCone"><code>MOI.DualExponentialCone</code></a>, that offers an alternative formulation that can be more efficient for some formulations.</p><p>There is also the <a href="../../../moi/reference/standard_form/#MathOptInterface.RelativeEntropyCone"><code>MOI.RelativeEntropyCone</code></a> for explicitly encoding the relative entropy function</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2.663206172203935</code></pre><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.ExponentialCone"><code>MOI.ExponentialCone</code></a> has a dual, the <a href="../../../moi/reference/standard_form/#MathOptInterface.DualExponentialCone"><code>MOI.DualExponentialCone</code></a>, that offers an alternative formulation that can be more efficient for some formulations.</p><p>There is also the <a href="../../../moi/reference/standard_form/#MathOptInterface.RelativeEntropyCone"><code>MOI.RelativeEntropyCone</code></a> for explicitly encoding the relative entropy function</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, x[1:n])
@@ -84,14 +84,14 @@
 @constraint(model, A * x .&lt;= b)
 @constraint(model, [t; ones(n); x] in MOI.RelativeEntropyCone(2n + 1))
 optimize!(model)
-objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2.663173643175568</code></pre><h2 id="PowerCone"><a class="docs-heading-anchor" href="#PowerCone">PowerCone</a><a id="PowerCone-1"></a><a class="docs-heading-anchor-permalink" href="#PowerCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a> is a three-dimensional set parameterized by a scalar value <code>α</code>. It has the form:</p><p class="math-container">\[K_{p} = \{ (x,y,z) \in \mathbb{R}^3 : x^{\alpha} y^{1-\alpha} \ge |z|, x \ge 0, y \ge 0 \}\]</p><p>The power cone permits a number of reformulations. For example, when <span>$p &gt; 1$</span>, we can model <span>$t \ge x^p$</span> using the power cone <span>$(t, 1, x)$</span> with <span>$\alpha = 1 / p$</span>. Thus, to model <span>$t \ge x^3$</span> with <span>$x \ge 0$</span></p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2.6631736431749626</code></pre><h2 id="PowerCone"><a class="docs-heading-anchor" href="#PowerCone">PowerCone</a><a id="PowerCone-1"></a><a class="docs-heading-anchor-permalink" href="#PowerCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a> is a three-dimensional set parameterized by a scalar value <code>α</code>. It has the form:</p><p class="math-container">\[K_{p} = \{ (x,y,z) \in \mathbb{R}^3 : x^{\alpha} y^{1-\alpha} \ge |z|, x \ge 0, y \ge 0 \}\]</p><p>The power cone permits a number of reformulations. For example, when <span>$p &gt; 1$</span>, we can model <span>$t \ge x^p$</span> using the power cone <span>$(t, 1, x)$</span> with <span>$\alpha = 1 / p$</span>. Thus, to model <span>$t \ge x^3$</span> with <span>$x \ge 0$</span></p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, x &gt;= 1.5)
 @constraint(model, [t, 1, x] in MOI.PowerCone(1 / 3))
 @objective(model, Min, t)
 optimize!(model)
-value(t), value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(3.3747548357745134, 1.4999552263154623)</code></pre><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a> has a dual, the <a href="../../../moi/reference/standard_form/#MathOptInterface.DualPowerCone"><code>MOI.DualPowerCone</code></a>, that offers an alternative formulation that can be more efficient for some formulations.</p><h2 id="P-Norm"><a class="docs-heading-anchor" href="#P-Norm">P-Norm</a><a id="P-Norm-1"></a><a class="docs-heading-anchor-permalink" href="#P-Norm" title="Permalink"></a></h2><p>The p-norm <span>$||x||_p = \left(\sum\limits_{i} |x_i|^p\right)^{\frac{1}{p}}$</span> can be modeled using <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a>s. See the <a href="https://docs.mosek.com/modeling-cookbook/powo.html#p-norm-cones">Mosek Modeling Cookbook</a> for the derivation.</p><pre><code class="language-julia hljs">function p_norm(x::Vector, p)
+value(t), value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(3.374754835774048, 1.4999552263153844)</code></pre><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a> has a dual, the <a href="../../../moi/reference/standard_form/#MathOptInterface.DualPowerCone"><code>MOI.DualPowerCone</code></a>, that offers an alternative formulation that can be more efficient for some formulations.</p><h2 id="P-Norm"><a class="docs-heading-anchor" href="#P-Norm">P-Norm</a><a id="P-Norm-1"></a><a class="docs-heading-anchor-permalink" href="#P-Norm" title="Permalink"></a></h2><p>The p-norm <span>$||x||_p = \left(\sum\limits_{i} |x_i|^p\right)^{\frac{1}{p}}$</span> can be modeled using <a href="../../../moi/reference/standard_form/#MathOptInterface.PowerCone"><code>MOI.PowerCone</code></a>s. See the <a href="https://docs.mosek.com/modeling-cookbook/powo.html#p-norm-cones">Mosek Modeling Cookbook</a> for the derivation.</p><pre><code class="language-julia hljs">function p_norm(x::Vector, p)
     N = length(x)
     model = Model(SCS.Optimizer)
     set_silent(model)
@@ -105,7 +105,7 @@
 end
 
 x = rand(5);
-LinearAlgebra.norm(x, 4), p_norm(x, 4)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.9316922467512209, 0.9316875994951573)</code></pre><h2 id="Positive-Semidefinite-Cone"><a class="docs-heading-anchor" href="#Positive-Semidefinite-Cone">Positive Semidefinite Cone</a><a id="Positive-Semidefinite-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Positive-Semidefinite-Cone" title="Permalink"></a></h2><p>The set of positive semidefinite matrices (PSD) of dimension <span>$n$</span> form a cone in <span>$\mathbb{R}^n$</span>. We write this set mathematically as:</p><p class="math-container">\[\mathcal{S}_{+}^n = \{ X \in \mathcal{S}^n \mid z^T X z \geq 0, \: \forall z\in \mathbb{R}^n \}.\]</p><p>A PSD cone is represented in JuMP using the MOI sets <code>PositiveSemidefiniteConeTriangle</code> (for upper triangle of a PSD matrix) and <code>PositiveSemidefiniteConeSquare</code> (for a complete PSD matrix). However, it is preferable to use the <code>PSDCone</code> shortcut as illustrated below.</p><h4 id="Example:-largest-eigenvalue-of-a-symmetric-matrix"><a class="docs-heading-anchor" href="#Example:-largest-eigenvalue-of-a-symmetric-matrix">Example: largest eigenvalue of a symmetric matrix</a><a id="Example:-largest-eigenvalue-of-a-symmetric-matrix-1"></a><a class="docs-heading-anchor-permalink" href="#Example:-largest-eigenvalue-of-a-symmetric-matrix" title="Permalink"></a></h4><p>Suppose <span>$A$</span> has eigenvalues <span>$\lambda_{1} \geq \lambda_{2} \ldots \geq \lambda_{n}$</span>. Then the matrix <span>$t I-A$</span> has eigenvalues <span>$t-\lambda_{1}, t-\lambda_{2}, \ldots, t-\lambda_{n}$</span>. Note that <span>$t I-A$</span> is PSD exactly when all these eigenvalues are non-negative, and this happens for values <span>$t \geq \lambda_{1}$</span>. Thus, we can model the problem of finding the largest eigenvalue of a symmetric matrix as:</p><p class="math-container">\[\begin{aligned}
+LinearAlgebra.norm(x, 4), p_norm(x, 4)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.9316922467512209, 0.931687599495156)</code></pre><h2 id="Positive-Semidefinite-Cone"><a class="docs-heading-anchor" href="#Positive-Semidefinite-Cone">Positive Semidefinite Cone</a><a id="Positive-Semidefinite-Cone-1"></a><a class="docs-heading-anchor-permalink" href="#Positive-Semidefinite-Cone" title="Permalink"></a></h2><p>The set of positive semidefinite matrices (PSD) of dimension <span>$n$</span> form a cone in <span>$\mathbb{R}^n$</span>. We write this set mathematically as:</p><p class="math-container">\[\mathcal{S}_{+}^n = \{ X \in \mathcal{S}^n \mid z^T X z \geq 0, \: \forall z\in \mathbb{R}^n \}.\]</p><p>A PSD cone is represented in JuMP using the MOI sets <code>PositiveSemidefiniteConeTriangle</code> (for upper triangle of a PSD matrix) and <code>PositiveSemidefiniteConeSquare</code> (for a complete PSD matrix). However, it is preferable to use the <code>PSDCone</code> shortcut as illustrated below.</p><h4 id="Example:-largest-eigenvalue-of-a-symmetric-matrix"><a class="docs-heading-anchor" href="#Example:-largest-eigenvalue-of-a-symmetric-matrix">Example: largest eigenvalue of a symmetric matrix</a><a id="Example:-largest-eigenvalue-of-a-symmetric-matrix-1"></a><a class="docs-heading-anchor-permalink" href="#Example:-largest-eigenvalue-of-a-symmetric-matrix" title="Permalink"></a></h4><p>Suppose <span>$A$</span> has eigenvalues <span>$\lambda_{1} \geq \lambda_{2} \ldots \geq \lambda_{n}$</span>. Then the matrix <span>$t I-A$</span> has eigenvalues <span>$t-\lambda_{1}, t-\lambda_{2}, \ldots, t-\lambda_{n}$</span>. Note that <span>$t I-A$</span> is PSD exactly when all these eigenvalues are non-negative, and this happens for values <span>$t \geq \lambda_{1}$</span>. Thus, we can model the problem of finding the largest eigenvalue of a symmetric matrix as:</p><p class="math-container">\[\begin{aligned}
 \lambda_{1} = \min t \\
 \text { s.t. } t I-A \succeq 0
 \end{aligned}\]</p><pre><code class="language-julia hljs">A = [3 2 4; 2 0 2; 4 2 3]
@@ -116,14 +116,14 @@
 @objective(model, Min, t)
 @constraint(model, t .* I - A in PSDCone())
 optimize!(model)
-objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8.000003377698668</code></pre><h2 id="GeometricMeanCone"><a class="docs-heading-anchor" href="#GeometricMeanCone">GeometricMeanCone</a><a id="GeometricMeanCone-1"></a><a class="docs-heading-anchor-permalink" href="#GeometricMeanCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>MOI.GeometricMeanCone</code></a> is a cone of the form:</p><p class="math-container">\[K_{geo} = \{ (t, x) \in \mathbb{R}^n : x \ge 0, t \le \sqrt[n-1]{x_1 x_2 \cdots x_{n-1}} \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">8.000003377698652</code></pre><h2 id="GeometricMeanCone"><a class="docs-heading-anchor" href="#GeometricMeanCone">GeometricMeanCone</a><a id="GeometricMeanCone-1"></a><a class="docs-heading-anchor-permalink" href="#GeometricMeanCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.GeometricMeanCone"><code>MOI.GeometricMeanCone</code></a> is a cone of the form:</p><p class="math-container">\[K_{geo} = \{ (t, x) \in \mathbb{R}^n : x \ge 0, t \le \sqrt[n-1]{x_1 x_2 \cdots x_{n-1}} \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, x[1:4])
 @variable(model, t)
 @constraint(model, sum(x) == 1)
 @constraint(model, [t; x] in MOI.GeometricMeanCone(5))
 optimize!(model)
-value(t), value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.0, [0.25000000110304893, 0.250000001103052, 0.25000000110305065, 0.2500000011030498])</code></pre><h2 id="RootDetCone"><a class="docs-heading-anchor" href="#RootDetCone">RootDetCone</a><a id="RootDetCone-1"></a><a class="docs-heading-anchor-permalink" href="#RootDetCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>MOI.RootDetConeSquare</code></a> is a cone of the form:</p><p class="math-container">\[K = \{ (t, X) \in \mathbb{R}^{1+d^2} : t \le \det(X)^{\frac{1}{d}} \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(t), value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(0.0, [0.25000000110304854, 0.2500000011030519, 0.2500000011030504, 0.25000000110305026])</code></pre><h2 id="RootDetCone"><a class="docs-heading-anchor" href="#RootDetCone">RootDetCone</a><a id="RootDetCone-1"></a><a class="docs-heading-anchor-permalink" href="#RootDetCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeSquare"><code>MOI.RootDetConeSquare</code></a> is a cone of the form:</p><p class="math-container">\[K = \{ (t, X) \in \mathbb{R}^{1+d^2} : t \le \det(X)^{\frac{1}{d}} \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, X[1:2, 1:2])
@@ -131,7 +131,7 @@
 @constraint(model, [t; vec(X)] in MOI.RootDetConeSquare(2))
 @constraint(model, X .== [2 1; 1 3])
 optimize!(model)
-value(t), sqrt(LinearAlgebra.det(value.(X)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(2.236116364743911, 2.2360679574027222)</code></pre><p>If <code>X</code> is symmetric, then you can use <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>MOI.RootDetConeTriangle</code></a> instead. This can be more efficient because the solver does not need to add additional constraints to ensure <code>X</code> is symmetric.</p><p>When forming the function, use <a href="../../../api/JuMP/#JuMP.triangle_vec"><code>triangle_vec</code></a> to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(t), sqrt(LinearAlgebra.det(value.(X)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(2.2361163647438964, 2.236067957402722)</code></pre><p>If <code>X</code> is symmetric, then you can use <a href="../../../moi/reference/standard_form/#MathOptInterface.RootDetConeTriangle"><code>MOI.RootDetConeTriangle</code></a> instead. This can be more efficient because the solver does not need to add additional constraints to ensure <code>X</code> is symmetric.</p><p>When forming the function, use <a href="../../../api/JuMP/#JuMP.triangle_vec"><code>triangle_vec</code></a> to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, X[1:2, 1:2], Symmetric)
@@ -139,7 +139,7 @@
 @constraint(model, [t; triangle_vec(X)] in MOI.RootDetConeTriangle(2))
 @constraint(model, X .== [2 1; 1 3])
 optimize!(model)
-value(t), sqrt(LinearAlgebra.det(value.(X)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(2.2361792479821463, 2.2360679863316992)</code></pre><h2 id="LogDetCone"><a class="docs-heading-anchor" href="#LogDetCone">LogDetCone</a><a id="LogDetCone-1"></a><a class="docs-heading-anchor-permalink" href="#LogDetCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.LogDetConeSquare"><code>MOI.LogDetConeSquare</code></a> is a cone of the form:</p><p class="math-container">\[K = \{ (t, u, X) \in \mathbb{R}^{2+d^2} : t \le u \log(\det(X / u)) \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(t), sqrt(LinearAlgebra.det(value.(X)))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(2.2361792479821303, 2.2360679863316997)</code></pre><h2 id="LogDetCone"><a class="docs-heading-anchor" href="#LogDetCone">LogDetCone</a><a id="LogDetCone-1"></a><a class="docs-heading-anchor-permalink" href="#LogDetCone" title="Permalink"></a></h2><p>The <a href="../../../moi/reference/standard_form/#MathOptInterface.LogDetConeSquare"><code>MOI.LogDetConeSquare</code></a> is a cone of the form:</p><p class="math-container">\[K = \{ (t, u, X) \in \mathbb{R}^{2+d^2} : t \le u \log(\det(X / u)) \}\]</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, u)
@@ -149,7 +149,7 @@
 @constraint(model, X .== [2 1; 1 3])
 @constraint(model, u == 0.5)
 optimize!(model)
-value(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.497920445822069, 1.4978661006407137)</code></pre><p>If <code>X</code> is symmetric, then you can use <a href="../../../moi/reference/standard_form/#MathOptInterface.LogDetConeTriangle"><code>MOI.LogDetConeTriangle</code></a> instead. This can be more efficient because the solver does not need to add additional constraints to ensure <code>X</code> is symmetric.</p><p>When forming the function, use <a href="../../../api/JuMP/#JuMP.triangle_vec"><code>triangle_vec</code></a> to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
+value(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.4979204458220523, 1.4978661006407135)</code></pre><p>If <code>X</code> is symmetric, then you can use <a href="../../../moi/reference/standard_form/#MathOptInterface.LogDetConeTriangle"><code>MOI.LogDetConeTriangle</code></a> instead. This can be more efficient because the solver does not need to add additional constraints to ensure <code>X</code> is symmetric.</p><p>When forming the function, use <a href="../../../api/JuMP/#JuMP.triangle_vec"><code>triangle_vec</code></a> to obtain the column-wise upper triangle of the matrix as a vector in the order that JuMP requires.</p><pre><code class="language-julia hljs">model = Model(SCS.Optimizer)
 set_silent(model)
 @variable(model, t)
 @variable(model, u)
@@ -159,4 +159,4 @@
 @constraint(model, X .== [2 1; 1 3])
 @constraint(model, u == 0.5)
 optimize!(model)
-value(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.4979204481049417, 1.497866101098124)</code></pre><h2 id="Other-Cones-and-Functions"><a class="docs-heading-anchor" href="#Other-Cones-and-Functions">Other Cones and Functions</a><a id="Other-Cones-and-Functions-1"></a><a class="docs-heading-anchor-permalink" href="#Other-Cones-and-Functions" title="Permalink"></a></h2><p>For other cones supported by JuMP, check out the <a href="https://jump.dev/MathOptInterface.jl/stable">MathOptInterface Manual</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../start_values/">« Primal and dual warm-starts</a><a class="docs-footer-nextpage" href="../simple_examples/">Simple semidefinite programming examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+value(t), 0.5 * log(LinearAlgebra.det(value.(X) ./ 0.5))</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(1.4979204481049286, 1.497866101098124)</code></pre><h2 id="Other-Cones-and-Functions"><a class="docs-heading-anchor" href="#Other-Cones-and-Functions">Other Cones and Functions</a><a id="Other-Cones-and-Functions-1"></a><a class="docs-heading-anchor-permalink" href="#Other-Cones-and-Functions" title="Permalink"></a></h2><p>For other cones supported by JuMP, check out the <a href="https://jump.dev/MathOptInterface.jl/stable">MathOptInterface Manual</a>.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../start_values/">« Primal and dual warm-starts</a><a class="docs-footer-nextpage" href="../simple_examples/">Simple semidefinite programming examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/debugging/index.html b/previews/PR3562/tutorials/getting_started/debugging/index.html
index 688eb21e762..11adcaeefd3 100644
--- a/previews/PR3562/tutorials/getting_started/debugging/index.html
+++ b/previews/PR3562/tutorials/getting_started/debugging/index.html
@@ -20,4 +20,4 @@
            if abs(value(var)) &gt; 1e3
                println(&quot;Variable `$(name(var))` may be unbounded&quot;)
            end
-       end</code><code class="nohighlight hljs ansi" style="display:block;">Variable `x` may be unbounded</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_data_and_plotting/">« Getting started with data and plotting</a><a class="docs-footer-nextpage" href="../design_patterns_for_larger_models/">Design patterns for larger models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+       end</code><code class="nohighlight hljs ansi" style="display:block;">Variable `x` may be unbounded</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_data_and_plotting/">« Getting started with data and plotting</a><a class="docs-footer-nextpage" href="../design_patterns_for_larger_models/">Design patterns for larger models »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/design_patterns_for_larger_models/index.html b/previews/PR3562/tutorials/getting_started/design_patterns_for_larger_models/index.html
index 2051ed1980c..15a50e25388 100644
--- a/previews/PR3562/tutorials/getting_started/design_patterns_for_larger_models/index.html
+++ b/previews/PR3562/tutorials/getting_started/design_patterns_for_larger_models/index.html
@@ -514,4 +514,4 @@
         )
         @test x === nothing
     end
-end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Test.DefaultTestSet(&quot;KnapsackModel&quot;, Any[Test.DefaultTestSet(&quot;feasible_binary_knapsack&quot;, Any[], 5, false, false, true, 1.699305004541627e9, 1.69930500454738e9, false), Test.DefaultTestSet(&quot;feasible_integer_knapsack&quot;, Any[], 5, false, false, true, 1.699305004547422e9, 1.699305004661079e9, false), Test.DefaultTestSet(&quot;infeasible_binary_knapsack&quot;, Any[], 1, false, false, true, 1.699305004661117e9, 1.699305004662244e9, false)], 0, false, false, true, 1.699305004541587e9, 1.69930500466225e9, false)</code></pre><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Place these tests in a separate file <code>test_knapsack_model.jl</code> so that you can run the tests by adding <code>include(&quot;test_knapsack_model.jl&quot;)</code> to any file where needed.</p></div></div><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>We&#39;ve only briefly scratched the surface of ways to create and structure large JuMP models, so consider this tutorial a starting point, rather than a comprehensive list of all the possible ways to structure JuMP models.  If you are embarking on a large project that uses JuMP, a good next step is to look at ways people have written large JuMP projects &quot;in the wild.&quot;</p><p>Here are some good examples (all co-incidentally related to energy):</p><ul><li>AnyMOD.jl<ul><li><a href="https://www.youtube.com/watch?v=QE_tNDER0F4">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/leonardgoeke/AnyMOD.jl">source code</a></li></ul></li><li>PowerModels.jl<ul><li><a href="https://www.youtube.com/watch?v=POOt1FCA8LI">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/lanl-ansi/PowerModels.jl">source code</a></li></ul></li><li>PowerSimulations.jl<ul><li><a href="https://www.youtube.com/watch?v=-ZoO3npjwYU">JuliaCon 2021 talk</a></li><li><a href="https://github.com/NREL-SIIP/PowerSimulations.jl">source code</a></li></ul></li><li>UnitCommitment.jl<ul><li><a href="https://www.youtube.com/watch?v=rYUZK9kYeIY">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/ANL-CEEESA/UnitCommitment.jl">source code</a></li></ul></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../debugging/">« Debugging</a><a class="docs-footer-nextpage" href="../performance_tips/">Performance tips »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Test.DefaultTestSet(&quot;KnapsackModel&quot;, Any[Test.DefaultTestSet(&quot;feasible_binary_knapsack&quot;, Any[], 5, false, false, true, 1.69930524587289e9, 1.699305245880652e9, false), Test.DefaultTestSet(&quot;feasible_integer_knapsack&quot;, Any[], 5, false, false, true, 1.699305245880712e9, 1.699305246047624e9, false), Test.DefaultTestSet(&quot;infeasible_binary_knapsack&quot;, Any[], 1, false, false, true, 1.699305246047673e9, 1.699305246049347e9, false)], 0, false, false, true, 1.699305245872841e9, 1.699305246049357e9, false)</code></pre><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Place these tests in a separate file <code>test_knapsack_model.jl</code> so that you can run the tests by adding <code>include(&quot;test_knapsack_model.jl&quot;)</code> to any file where needed.</p></div></div><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>We&#39;ve only briefly scratched the surface of ways to create and structure large JuMP models, so consider this tutorial a starting point, rather than a comprehensive list of all the possible ways to structure JuMP models.  If you are embarking on a large project that uses JuMP, a good next step is to look at ways people have written large JuMP projects &quot;in the wild.&quot;</p><p>Here are some good examples (all co-incidentally related to energy):</p><ul><li>AnyMOD.jl<ul><li><a href="https://www.youtube.com/watch?v=QE_tNDER0F4">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/leonardgoeke/AnyMOD.jl">source code</a></li></ul></li><li>PowerModels.jl<ul><li><a href="https://www.youtube.com/watch?v=POOt1FCA8LI">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/lanl-ansi/PowerModels.jl">source code</a></li></ul></li><li>PowerSimulations.jl<ul><li><a href="https://www.youtube.com/watch?v=-ZoO3npjwYU">JuliaCon 2021 talk</a></li><li><a href="https://github.com/NREL-SIIP/PowerSimulations.jl">source code</a></li></ul></li><li>UnitCommitment.jl<ul><li><a href="https://www.youtube.com/watch?v=rYUZK9kYeIY">JuMP-dev 2021 talk</a></li><li><a href="https://github.com/ANL-CEEESA/UnitCommitment.jl">source code</a></li></ul></li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../debugging/">« Debugging</a><a class="docs-footer-nextpage" href="../performance_tips/">Performance tips »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_JuMP/index.html b/previews/PR3562/tutorials/getting_started/getting_started_with_JuMP/index.html
index 919e495a04c..c04539431a4 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_JuMP/index.html
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_JuMP/index.html
@@ -217,4 +217,4 @@
 Model   status      : Optimal
 Simplex   iterations: 4
 Objective value     :  4.9230769231e+00
-HiGHS run time      :          0.00</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; objective_value(vector_model)</code><code class="nohighlight hljs ansi" style="display:block;">4.923076923076922</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_julia/">« Getting started with Julia</a><a class="docs-footer-nextpage" href="../getting_started_with_sets_and_indexing/">Getting started with sets and indexing »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+HiGHS run time      :          0.00</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; objective_value(vector_model)</code><code class="nohighlight hljs ansi" style="display:block;">4.923076923076922</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_julia/">« Getting started with Julia</a><a class="docs-footer-nextpage" href="../getting_started_with_sets_and_indexing/">Getting started with sets and indexing »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/1e380636.svg b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/6cdf6131.svg
similarity index 84%
rename from previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/1e380636.svg
rename to previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/6cdf6131.svg
index f8a7c35619a..f2fa063fabf 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/1e380636.svg
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/6cdf6131.svg
@@ -1,62 +1,62 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip490">
+  <clipPath id="clip310">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip490)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip310)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip491">
+  <clipPath id="clip311">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip490)" d="M175.445 1423.18 L2352.76 1423.18 L2352.76 47.2441 L175.445 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip310)" d="M175.445 1423.18 L2352.76 1423.18 L2352.76 47.2441 L175.445 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip492">
+  <clipPath id="clip312">
     <rect x="175" y="47" width="2178" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="428.842,1423.18 428.842,47.2441 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="762.946,1423.18 762.946,47.2441 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1097.05,1423.18 1097.05,47.2441 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1431.15,1423.18 1431.15,47.2441 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1765.26,1423.18 1765.26,47.2441 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2099.36,1423.18 2099.36,47.2441 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="428.842,1423.18 428.842,1404.28 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="762.946,1423.18 762.946,1404.28 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1097.05,1423.18 1097.05,1404.28 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1431.15,1423.18 1431.15,1404.28 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1765.26,1423.18 1765.26,1404.28 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.36,1423.18 2099.36,1404.28 "/>
-<path clip-path="url(#clip490)" d="M378.6 1472.64 Q378.6 1467.94 376.655 1465.28 Q374.734 1462.59 371.354 1462.59 Q367.975 1462.59 366.03 1465.28 Q364.109 1467.94 364.109 1472.64 Q364.109 1477.34 366.03 1480.02 Q367.975 1482.69 371.354 1482.69 Q374.734 1482.69 376.655 1480.02 Q378.6 1477.34 378.6 1472.64 M364.109 1463.59 Q365.452 1461.27 367.489 1460.16 Q369.549 1459.03 372.396 1459.03 Q377.118 1459.03 380.058 1462.78 Q383.021 1466.53 383.021 1472.64 Q383.021 1478.75 380.058 1482.5 Q377.118 1486.25 372.396 1486.25 Q369.549 1486.25 367.489 1485.14 Q365.452 1484.01 364.109 1481.69 L364.109 1485.58 L359.827 1485.58 L359.827 1449.56 L364.109 1449.56 L364.109 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M405.104 1463.64 Q404.387 1463.22 403.53 1463.03 Q402.697 1462.83 401.678 1462.83 Q398.067 1462.83 396.123 1465.19 Q394.201 1467.52 394.201 1471.92 L394.201 1485.58 L389.919 1485.58 L389.919 1459.65 L394.201 1459.65 L394.201 1463.68 Q395.544 1461.32 397.697 1460.19 Q399.85 1459.03 402.928 1459.03 Q403.368 1459.03 403.9 1459.1 Q404.433 1459.14 405.081 1459.26 L405.104 1463.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M418.576 1462.64 Q415.15 1462.64 413.16 1465.33 Q411.169 1467.99 411.169 1472.64 Q411.169 1477.29 413.137 1479.98 Q415.127 1482.64 418.576 1482.64 Q421.979 1482.64 423.97 1479.95 Q425.961 1477.27 425.961 1472.64 Q425.961 1468.03 423.97 1465.35 Q421.979 1462.64 418.576 1462.64 M418.576 1459.03 Q424.132 1459.03 427.303 1462.64 Q430.474 1466.25 430.474 1472.64 Q430.474 1479.01 427.303 1482.64 Q424.132 1486.25 418.576 1486.25 Q412.998 1486.25 409.826 1482.64 Q406.678 1479.01 406.678 1472.64 Q406.678 1466.25 409.826 1462.64 Q412.998 1459.03 418.576 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M435.058 1459.65 L439.317 1459.65 L444.641 1479.89 L449.942 1459.65 L454.965 1459.65 L460.289 1479.89 L465.59 1459.65 L469.849 1459.65 L463.067 1485.58 L458.044 1485.58 L452.465 1464.33 L446.863 1485.58 L441.84 1485.58 L435.058 1459.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M497.858 1469.93 L497.858 1485.58 L493.599 1485.58 L493.599 1470.07 Q493.599 1466.39 492.164 1464.56 Q490.729 1462.73 487.858 1462.73 Q484.409 1462.73 482.419 1464.93 Q480.428 1467.13 480.428 1470.93 L480.428 1485.58 L476.145 1485.58 L476.145 1459.65 L480.428 1459.65 L480.428 1463.68 Q481.956 1461.34 484.016 1460.19 Q486.099 1459.03 488.807 1459.03 Q493.275 1459.03 495.567 1461.81 Q497.858 1464.56 497.858 1469.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M733.895 1472.64 Q733.895 1467.94 731.95 1465.28 Q730.029 1462.59 726.65 1462.59 Q723.27 1462.59 721.326 1465.28 Q719.404 1467.94 719.404 1472.64 Q719.404 1477.34 721.326 1480.02 Q723.27 1482.69 726.65 1482.69 Q730.029 1482.69 731.95 1480.02 Q733.895 1477.34 733.895 1472.64 M719.404 1463.59 Q720.747 1461.27 722.784 1460.16 Q724.844 1459.03 727.691 1459.03 Q732.413 1459.03 735.353 1462.78 Q738.316 1466.53 738.316 1472.64 Q738.316 1478.75 735.353 1482.5 Q732.413 1486.25 727.691 1486.25 Q724.844 1486.25 722.784 1485.14 Q720.747 1484.01 719.404 1481.69 L719.404 1485.58 L715.122 1485.58 L715.122 1449.56 L719.404 1449.56 L719.404 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M745.376 1449.56 L749.636 1449.56 L749.636 1485.58 L745.376 1485.58 L745.376 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M758.108 1475.35 L758.108 1459.65 L762.367 1459.65 L762.367 1475.19 Q762.367 1478.87 763.802 1480.72 Q765.237 1482.55 768.108 1482.55 Q771.557 1482.55 773.547 1480.35 Q775.561 1478.15 775.561 1474.35 L775.561 1459.65 L779.821 1459.65 L779.821 1485.58 L775.561 1485.58 L775.561 1481.6 Q774.01 1483.96 771.95 1485.12 Q769.913 1486.25 767.205 1486.25 Q762.737 1486.25 760.423 1483.47 Q758.108 1480.7 758.108 1475.35 M768.825 1459.03 L768.825 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M810.77 1471.55 L810.77 1473.64 L791.186 1473.64 Q791.464 1478.03 793.825 1480.35 Q796.209 1482.64 800.446 1482.64 Q802.899 1482.64 805.191 1482.04 Q807.506 1481.44 809.774 1480.23 L809.774 1484.26 Q807.483 1485.23 805.075 1485.74 Q802.668 1486.25 800.191 1486.25 Q793.987 1486.25 790.353 1482.64 Q786.742 1479.03 786.742 1472.87 Q786.742 1466.51 790.168 1462.78 Q793.617 1459.03 799.45 1459.03 Q804.682 1459.03 807.714 1462.41 Q810.77 1465.77 810.77 1471.55 M806.51 1470.3 Q806.464 1466.81 804.543 1464.72 Q802.645 1462.64 799.496 1462.64 Q795.932 1462.64 793.779 1464.65 Q791.649 1466.67 791.325 1470.33 L806.51 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1036.92 1487.99 Q1035.12 1492.62 1033.4 1494.03 Q1031.69 1495.44 1028.82 1495.44 L1025.42 1495.44 L1025.42 1491.88 L1027.92 1491.88 Q1029.68 1491.88 1030.65 1491.04 Q1031.62 1490.21 1032.8 1487.11 L1033.57 1485.16 L1023.08 1459.65 L1027.59 1459.65 L1035.7 1479.93 L1043.8 1459.65 L1048.31 1459.65 L1036.92 1487.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1076.37 1471.55 L1076.37 1473.64 L1056.78 1473.64 Q1057.06 1478.03 1059.42 1480.35 Q1061.81 1482.64 1066.04 1482.64 Q1068.5 1482.64 1070.79 1482.04 Q1073.1 1481.44 1075.37 1480.23 L1075.37 1484.26 Q1073.08 1485.23 1070.67 1485.74 Q1068.26 1486.25 1065.79 1486.25 Q1059.58 1486.25 1055.95 1482.64 Q1052.34 1479.03 1052.34 1472.87 Q1052.34 1466.51 1055.76 1462.78 Q1059.21 1459.03 1065.05 1459.03 Q1070.28 1459.03 1073.31 1462.41 Q1076.37 1465.77 1076.37 1471.55 M1072.11 1470.3 Q1072.06 1466.81 1070.14 1464.72 Q1068.24 1462.64 1065.09 1462.64 Q1061.53 1462.64 1059.38 1464.65 Q1057.25 1466.67 1056.92 1470.33 L1072.11 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1083.36 1449.56 L1087.62 1449.56 L1087.62 1485.58 L1083.36 1485.58 L1083.36 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1096.53 1449.56 L1100.79 1449.56 L1100.79 1485.58 L1096.53 1485.58 L1096.53 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1119.75 1462.64 Q1116.32 1462.64 1114.33 1465.33 Q1112.34 1467.99 1112.34 1472.64 Q1112.34 1477.29 1114.31 1479.98 Q1116.3 1482.64 1119.75 1482.64 Q1123.15 1482.64 1125.14 1479.95 Q1127.13 1477.27 1127.13 1472.64 Q1127.13 1468.03 1125.14 1465.35 Q1123.15 1462.64 1119.75 1462.64 M1119.75 1459.03 Q1125.3 1459.03 1128.47 1462.64 Q1131.64 1466.25 1131.64 1472.64 Q1131.64 1479.01 1128.47 1482.64 Q1125.3 1486.25 1119.75 1486.25 Q1114.17 1486.25 1111 1482.64 Q1107.85 1479.01 1107.85 1472.64 Q1107.85 1466.25 1111 1462.64 Q1114.17 1459.03 1119.75 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1136.23 1459.65 L1140.49 1459.65 L1145.81 1479.89 L1151.11 1459.65 L1156.13 1459.65 L1161.46 1479.89 L1166.76 1459.65 L1171.02 1459.65 L1164.24 1485.58 L1159.21 1485.58 L1153.63 1464.33 L1148.03 1485.58 L1143.01 1485.58 L1136.23 1459.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1348.2 1472.64 Q1348.2 1467.94 1346.26 1465.28 Q1344.34 1462.59 1340.96 1462.59 Q1337.58 1462.59 1335.63 1465.28 Q1333.71 1467.94 1333.71 1472.64 Q1333.71 1477.34 1335.63 1480.02 Q1337.58 1482.69 1340.96 1482.69 Q1344.34 1482.69 1346.26 1480.02 Q1348.2 1477.34 1348.2 1472.64 M1333.71 1463.59 Q1335.05 1461.27 1337.09 1460.16 Q1339.15 1459.03 1342 1459.03 Q1346.72 1459.03 1349.66 1462.78 Q1352.62 1466.53 1352.62 1472.64 Q1352.62 1478.75 1349.66 1482.5 Q1346.72 1486.25 1342 1486.25 Q1339.15 1486.25 1337.09 1485.14 Q1335.05 1484.01 1333.71 1481.69 L1333.71 1485.58 L1329.43 1485.58 L1329.43 1449.56 L1333.71 1449.56 L1333.71 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1359.68 1449.56 L1363.94 1449.56 L1363.94 1485.58 L1359.68 1485.58 L1359.68 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1372.41 1475.35 L1372.41 1459.65 L1376.67 1459.65 L1376.67 1475.19 Q1376.67 1478.87 1378.11 1480.72 Q1379.54 1482.55 1382.41 1482.55 Q1385.86 1482.55 1387.85 1480.35 Q1389.87 1478.15 1389.87 1474.35 L1389.87 1459.65 L1394.13 1459.65 L1394.13 1485.58 L1389.87 1485.58 L1389.87 1481.6 Q1388.32 1483.96 1386.26 1485.12 Q1384.22 1486.25 1381.51 1486.25 Q1377.04 1486.25 1374.73 1483.47 Q1372.41 1480.7 1372.41 1475.35 M1383.13 1459.03 L1383.13 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1425.08 1471.55 L1425.08 1473.64 L1405.49 1473.64 Q1405.77 1478.03 1408.13 1480.35 Q1410.52 1482.64 1414.75 1482.64 Q1417.21 1482.64 1419.5 1482.04 Q1421.81 1481.44 1424.08 1480.23 L1424.08 1484.26 Q1421.79 1485.23 1419.38 1485.74 Q1416.97 1486.25 1414.5 1486.25 Q1408.29 1486.25 1404.66 1482.64 Q1401.05 1479.03 1401.05 1472.87 Q1401.05 1466.51 1404.47 1462.78 Q1407.92 1459.03 1413.76 1459.03 Q1418.99 1459.03 1422.02 1462.41 Q1425.08 1465.77 1425.08 1471.55 M1420.82 1470.3 Q1420.77 1466.81 1418.85 1464.72 Q1416.95 1462.64 1413.8 1462.64 Q1410.24 1462.64 1408.09 1464.65 Q1405.96 1466.67 1405.63 1470.33 L1420.82 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1449.13 1472.32 Q1449.13 1467.69 1447.21 1465.14 Q1445.31 1462.59 1441.86 1462.59 Q1438.43 1462.59 1436.51 1465.14 Q1434.61 1467.69 1434.61 1472.32 Q1434.61 1476.92 1436.51 1479.47 Q1438.43 1482.02 1441.86 1482.02 Q1445.31 1482.02 1447.21 1479.47 Q1449.13 1476.92 1449.13 1472.32 M1453.39 1482.36 Q1453.39 1488.98 1450.45 1492.2 Q1447.51 1495.44 1441.44 1495.44 Q1439.2 1495.44 1437.21 1495.09 Q1435.21 1494.77 1433.34 1494.08 L1433.34 1489.93 Q1435.21 1490.95 1437.04 1491.44 Q1438.87 1491.92 1440.77 1491.92 Q1444.96 1491.92 1447.04 1489.72 Q1449.13 1487.55 1449.13 1483.13 L1449.13 1481.02 Q1447.81 1483.31 1445.75 1484.45 Q1443.69 1485.58 1440.82 1485.58 Q1436.05 1485.58 1433.13 1481.95 Q1430.21 1478.31 1430.21 1472.32 Q1430.21 1466.3 1433.13 1462.66 Q1436.05 1459.03 1440.82 1459.03 Q1443.69 1459.03 1445.75 1460.16 Q1447.81 1461.3 1449.13 1463.59 L1449.13 1459.65 L1453.39 1459.65 L1453.39 1482.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1477.18 1463.64 Q1476.46 1463.22 1475.61 1463.03 Q1474.77 1462.83 1473.76 1462.83 Q1470.15 1462.83 1468.2 1465.19 Q1466.28 1467.52 1466.28 1471.92 L1466.28 1485.58 L1462 1485.58 L1462 1459.65 L1466.28 1459.65 L1466.28 1463.68 Q1467.62 1461.32 1469.77 1460.19 Q1471.93 1459.03 1475.01 1459.03 Q1475.45 1459.03 1475.98 1459.1 Q1476.51 1459.14 1477.16 1459.26 L1477.18 1463.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1493.43 1472.55 Q1488.27 1472.55 1486.28 1473.73 Q1484.29 1474.91 1484.29 1477.76 Q1484.29 1480.02 1485.77 1481.37 Q1487.27 1482.69 1489.84 1482.69 Q1493.39 1482.69 1495.52 1480.19 Q1497.67 1477.66 1497.67 1473.5 L1497.67 1472.55 L1493.43 1472.55 M1501.93 1470.79 L1501.93 1485.58 L1497.67 1485.58 L1497.67 1481.64 Q1496.21 1484.01 1494.03 1485.14 Q1491.86 1486.25 1488.71 1486.25 Q1484.73 1486.25 1482.37 1484.03 Q1480.03 1481.78 1480.03 1478.03 Q1480.03 1473.66 1482.95 1471.44 Q1485.89 1469.21 1491.7 1469.21 L1497.67 1469.21 L1497.67 1468.8 Q1497.67 1465.86 1495.72 1464.26 Q1493.8 1462.64 1490.31 1462.64 Q1488.08 1462.64 1485.98 1463.17 Q1483.87 1463.7 1481.93 1464.77 L1481.93 1460.83 Q1484.27 1459.93 1486.46 1459.49 Q1488.66 1459.03 1490.75 1459.03 Q1496.37 1459.03 1499.15 1461.95 Q1501.93 1464.86 1501.93 1470.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1521.49 1487.99 Q1519.68 1492.62 1517.97 1494.03 Q1516.26 1495.44 1513.39 1495.44 L1509.98 1495.44 L1509.98 1491.88 L1512.48 1491.88 Q1514.24 1491.88 1515.21 1491.04 Q1516.19 1490.21 1517.37 1487.11 L1518.13 1485.16 L1507.64 1459.65 L1512.16 1459.65 L1520.26 1479.93 L1528.36 1459.65 L1532.88 1459.65 L1521.49 1487.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1736.45 1472.32 Q1736.45 1467.69 1734.53 1465.14 Q1732.63 1462.59 1729.18 1462.59 Q1725.75 1462.59 1723.83 1465.14 Q1721.93 1467.69 1721.93 1472.32 Q1721.93 1476.92 1723.83 1479.47 Q1725.75 1482.02 1729.18 1482.02 Q1732.63 1482.02 1734.53 1479.47 Q1736.45 1476.92 1736.45 1472.32 M1740.71 1482.36 Q1740.71 1488.98 1737.77 1492.2 Q1734.83 1495.44 1728.76 1495.44 Q1726.52 1495.44 1724.53 1495.09 Q1722.54 1494.77 1720.66 1494.08 L1720.66 1489.93 Q1722.54 1490.95 1724.36 1491.44 Q1726.19 1491.92 1728.09 1491.92 Q1732.28 1491.92 1734.36 1489.72 Q1736.45 1487.55 1736.45 1483.13 L1736.45 1481.02 Q1735.13 1483.31 1733.07 1484.45 Q1731.01 1485.58 1728.14 1485.58 Q1723.37 1485.58 1720.45 1481.95 Q1717.54 1478.31 1717.54 1472.32 Q1717.54 1466.3 1720.45 1462.66 Q1723.37 1459.03 1728.14 1459.03 Q1731.01 1459.03 1733.07 1460.16 Q1735.13 1461.3 1736.45 1463.59 L1736.45 1459.65 L1740.71 1459.65 L1740.71 1482.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1759.53 1462.64 Q1756.1 1462.64 1754.11 1465.33 Q1752.12 1467.99 1752.12 1472.64 Q1752.12 1477.29 1754.09 1479.98 Q1756.08 1482.64 1759.53 1482.64 Q1762.93 1482.64 1764.92 1479.95 Q1766.91 1477.27 1766.91 1472.64 Q1766.91 1468.03 1764.92 1465.35 Q1762.93 1462.64 1759.53 1462.64 M1759.53 1459.03 Q1765.08 1459.03 1768.25 1462.64 Q1771.42 1466.25 1771.42 1472.64 Q1771.42 1479.01 1768.25 1482.64 Q1765.08 1486.25 1759.53 1486.25 Q1753.95 1486.25 1750.78 1482.64 Q1747.63 1479.01 1747.63 1472.64 Q1747.63 1466.25 1750.78 1462.64 Q1753.95 1459.03 1759.53 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1778.48 1449.56 L1782.74 1449.56 L1782.74 1485.58 L1778.48 1485.58 L1778.48 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1808.72 1463.59 L1808.72 1449.56 L1812.98 1449.56 L1812.98 1485.58 L1808.72 1485.58 L1808.72 1481.69 Q1807.37 1484.01 1805.31 1485.14 Q1803.28 1486.25 1800.41 1486.25 Q1795.71 1486.25 1792.74 1482.5 Q1789.8 1478.75 1789.8 1472.64 Q1789.8 1466.53 1792.74 1462.78 Q1795.71 1459.03 1800.41 1459.03 Q1803.28 1459.03 1805.31 1460.16 Q1807.37 1461.27 1808.72 1463.59 M1794.2 1472.64 Q1794.2 1477.34 1796.12 1480.02 Q1798.07 1482.69 1801.45 1482.69 Q1804.83 1482.69 1806.77 1480.02 Q1808.72 1477.34 1808.72 1472.64 Q1808.72 1467.94 1806.77 1465.28 Q1804.83 1462.59 1801.45 1462.59 Q1798.07 1462.59 1796.12 1465.28 Q1794.2 1467.94 1794.2 1472.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2057.44 1472.64 Q2057.44 1467.94 2055.49 1465.28 Q2053.57 1462.59 2050.19 1462.59 Q2046.81 1462.59 2044.87 1465.28 Q2042.95 1467.94 2042.95 1472.64 Q2042.95 1477.34 2044.87 1480.02 Q2046.81 1482.69 2050.19 1482.69 Q2053.57 1482.69 2055.49 1480.02 Q2057.44 1477.34 2057.44 1472.64 M2042.95 1463.59 Q2044.29 1461.27 2046.33 1460.16 Q2048.39 1459.03 2051.23 1459.03 Q2055.96 1459.03 2058.9 1462.78 Q2061.86 1466.53 2061.86 1472.64 Q2061.86 1478.75 2058.9 1482.5 Q2055.96 1486.25 2051.23 1486.25 Q2048.39 1486.25 2046.33 1485.14 Q2044.29 1484.01 2042.95 1481.69 L2042.95 1485.58 L2038.66 1485.58 L2038.66 1449.56 L2042.95 1449.56 L2042.95 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2068.92 1449.56 L2073.18 1449.56 L2073.18 1485.58 L2068.92 1485.58 L2068.92 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2093.87 1472.55 Q2088.71 1472.55 2086.72 1473.73 Q2084.73 1474.91 2084.73 1477.76 Q2084.73 1480.02 2086.21 1481.37 Q2087.72 1482.69 2090.28 1482.69 Q2093.83 1482.69 2095.96 1480.19 Q2098.11 1477.66 2098.11 1473.5 L2098.11 1472.55 L2093.87 1472.55 M2102.37 1470.79 L2102.37 1485.58 L2098.11 1485.58 L2098.11 1481.64 Q2096.65 1484.01 2094.47 1485.14 Q2092.3 1486.25 2089.15 1486.25 Q2085.17 1486.25 2082.81 1484.03 Q2080.47 1481.78 2080.47 1478.03 Q2080.47 1473.66 2083.39 1471.44 Q2086.33 1469.21 2092.14 1469.21 L2098.11 1469.21 L2098.11 1468.8 Q2098.11 1465.86 2096.16 1464.26 Q2094.24 1462.64 2090.75 1462.64 Q2088.53 1462.64 2086.42 1463.17 Q2084.31 1463.7 2082.37 1464.77 L2082.37 1460.83 Q2084.71 1459.93 2086.91 1459.49 Q2089.1 1459.03 2091.19 1459.03 Q2096.81 1459.03 2099.59 1461.95 Q2102.37 1464.86 2102.37 1470.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2129.8 1460.65 L2129.8 1464.63 Q2127.99 1463.64 2126.16 1463.15 Q2124.36 1462.64 2122.51 1462.64 Q2118.36 1462.64 2116.07 1465.28 Q2113.78 1467.89 2113.78 1472.64 Q2113.78 1477.39 2116.07 1480.02 Q2118.36 1482.64 2122.51 1482.64 Q2124.36 1482.64 2126.16 1482.15 Q2127.99 1481.64 2129.8 1480.65 L2129.8 1484.58 Q2128.02 1485.42 2126.09 1485.83 Q2124.2 1486.25 2122.04 1486.25 Q2116.19 1486.25 2112.74 1482.57 Q2109.29 1478.89 2109.29 1472.64 Q2109.29 1466.3 2112.76 1462.66 Q2116.26 1459.03 2122.32 1459.03 Q2124.29 1459.03 2126.16 1459.45 Q2128.04 1459.84 2129.8 1460.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2137.04 1449.56 L2141.33 1449.56 L2141.33 1470.83 L2154.03 1459.65 L2159.47 1459.65 L2145.72 1471.78 L2160.05 1485.58 L2154.5 1485.58 L2141.33 1472.92 L2141.33 1485.58 L2137.04 1485.58 L2137.04 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1116.75 1520.52 L1146.8 1520.52 L1146.8 1525.93 L1123.18 1525.93 L1123.18 1540 L1145.81 1540 L1145.81 1545.41 L1123.18 1545.41 L1123.18 1562.63 L1147.37 1562.63 L1147.37 1568.04 L1116.75 1568.04 L1116.75 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1172.51 1571.35 Q1170.03 1577.72 1167.68 1579.66 Q1165.32 1581.6 1161.37 1581.6 L1156.7 1581.6 L1156.7 1576.7 L1160.13 1576.7 Q1162.55 1576.7 1163.89 1575.56 Q1165.23 1574.41 1166.85 1570.14 L1167.9 1567.47 L1153.48 1532.4 L1159.69 1532.4 L1170.83 1560.28 L1181.97 1532.4 L1188.17 1532.4 L1172.51 1571.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1226.75 1548.76 L1226.75 1551.62 L1199.82 1551.62 Q1200.2 1557.67 1203.45 1560.85 Q1206.73 1564 1212.55 1564 Q1215.93 1564 1219.08 1563.17 Q1222.26 1562.35 1225.38 1560.69 L1225.38 1566.23 Q1222.23 1567.57 1218.92 1568.27 Q1215.61 1568.97 1212.2 1568.97 Q1203.67 1568.97 1198.68 1564 Q1193.71 1559.04 1193.71 1550.57 Q1193.71 1541.82 1198.42 1536.69 Q1203.16 1531.54 1211.19 1531.54 Q1218.38 1531.54 1222.55 1536.18 Q1226.75 1540.8 1226.75 1548.76 M1220.89 1547.04 Q1220.83 1542.23 1218.19 1539.37 Q1215.58 1536.5 1211.25 1536.5 Q1206.35 1536.5 1203.39 1539.27 Q1200.46 1542.04 1200.01 1547.07 L1220.89 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1282.74 1533.76 L1282.74 1539.24 Q1280.25 1537.87 1277.74 1537.2 Q1275.26 1536.5 1272.71 1536.5 Q1267.01 1536.5 1263.86 1540.13 Q1260.71 1543.73 1260.71 1550.25 Q1260.71 1556.78 1263.86 1560.4 Q1267.01 1564 1272.71 1564 Q1275.26 1564 1277.74 1563.33 Q1280.25 1562.63 1282.74 1561.26 L1282.74 1566.68 Q1280.29 1567.82 1277.64 1568.39 Q1275.03 1568.97 1272.07 1568.97 Q1264.02 1568.97 1259.28 1563.91 Q1254.54 1558.85 1254.54 1550.25 Q1254.54 1541.53 1259.31 1536.53 Q1264.12 1531.54 1272.46 1531.54 Q1275.16 1531.54 1277.74 1532.11 Q1280.32 1532.65 1282.74 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1306.73 1536.5 Q1302.02 1536.5 1299.29 1540.19 Q1296.55 1543.85 1296.55 1550.25 Q1296.55 1556.65 1299.26 1560.34 Q1301.99 1564 1306.73 1564 Q1311.41 1564 1314.15 1560.31 Q1316.89 1556.62 1316.89 1550.25 Q1316.89 1543.92 1314.15 1540.23 Q1311.41 1536.5 1306.73 1536.5 M1306.73 1531.54 Q1314.37 1531.54 1318.73 1536.5 Q1323.09 1541.47 1323.09 1550.25 Q1323.09 1559 1318.73 1564 Q1314.37 1568.97 1306.73 1568.97 Q1299.06 1568.97 1294.7 1564 Q1290.38 1559 1290.38 1550.25 Q1290.38 1541.47 1294.7 1536.5 Q1299.06 1531.54 1306.73 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1332.8 1518.52 L1338.66 1518.52 L1338.66 1568.04 L1332.8 1568.04 L1332.8 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1364.73 1536.5 Q1360.02 1536.5 1357.28 1540.19 Q1354.54 1543.85 1354.54 1550.25 Q1354.54 1556.65 1357.25 1560.34 Q1359.98 1564 1364.73 1564 Q1369.41 1564 1372.14 1560.31 Q1374.88 1556.62 1374.88 1550.25 Q1374.88 1543.92 1372.14 1540.23 Q1369.41 1536.5 1364.73 1536.5 M1364.73 1531.54 Q1372.37 1531.54 1376.73 1536.5 Q1381.09 1541.47 1381.09 1550.25 Q1381.09 1559 1376.73 1564 Q1372.37 1568.97 1364.73 1568.97 Q1357.06 1568.97 1352.7 1564 Q1348.37 1559 1348.37 1550.25 Q1348.37 1541.47 1352.7 1536.5 Q1357.06 1531.54 1364.73 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1411.45 1537.87 Q1410.46 1537.3 1409.29 1537.04 Q1408.14 1536.76 1406.74 1536.76 Q1401.77 1536.76 1399.1 1540 Q1396.46 1543.22 1396.46 1549.27 L1396.46 1568.04 L1390.57 1568.04 L1390.57 1532.4 L1396.46 1532.4 L1396.46 1537.93 Q1398.31 1534.69 1401.27 1533.13 Q1404.23 1531.54 1408.46 1531.54 Q1409.06 1531.54 1409.8 1531.63 Q1410.53 1531.7 1411.42 1531.85 L1411.45 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,1079.2 2352.76,1079.2 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,735.212 2352.76,735.212 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,391.228 2352.76,391.228 "/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,47.2441 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 175.445,47.2441 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 194.343,1423.18 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1079.2 194.343,1079.2 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,735.212 194.343,735.212 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,391.228 194.343,391.228 "/>
-<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,47.2441 194.343,47.2441 "/>
-<path clip-path="url(#clip490)" d="M127.501 1408.98 Q123.89 1408.98 122.061 1412.54 Q120.255 1416.08 120.255 1423.21 Q120.255 1430.32 122.061 1433.89 Q123.89 1437.43 127.501 1437.43 Q131.135 1437.43 132.941 1433.89 Q134.769 1430.32 134.769 1423.21 Q134.769 1416.08 132.941 1412.54 Q131.135 1408.98 127.501 1408.98 M127.501 1405.27 Q133.311 1405.27 136.367 1409.88 Q139.445 1414.46 139.445 1423.21 Q139.445 1431.94 136.367 1436.55 Q133.311 1441.13 127.501 1441.13 Q121.691 1441.13 118.612 1436.55 Q115.556 1431.94 115.556 1423.21 Q115.556 1414.46 118.612 1409.88 Q121.691 1405.27 127.501 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M123.126 1092.54 L139.445 1092.54 L139.445 1096.48 L117.501 1096.48 L117.501 1092.54 Q120.163 1089.79 124.746 1085.16 Q129.353 1080.5 130.533 1079.16 Q132.779 1076.64 133.658 1074.9 Q134.561 1073.14 134.561 1071.45 Q134.561 1068.7 132.617 1066.96 Q130.695 1065.23 127.593 1065.23 Q125.394 1065.23 122.941 1065.99 Q120.51 1066.75 117.732 1068.3 L117.732 1063.58 Q120.556 1062.45 123.01 1061.87 Q125.464 1061.29 127.501 1061.29 Q132.871 1061.29 136.066 1063.98 Q139.26 1066.66 139.26 1071.15 Q139.26 1073.28 138.45 1075.2 Q137.663 1077.1 135.556 1079.69 Q134.978 1080.36 131.876 1083.58 Q128.774 1086.78 123.126 1092.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M129.862 722.006 L118.056 740.455 L129.862 740.455 L129.862 722.006 M128.635 717.932 L134.515 717.932 L134.515 740.455 L139.445 740.455 L139.445 744.344 L134.515 744.344 L134.515 752.492 L129.862 752.492 L129.862 744.344 L114.26 744.344 L114.26 739.83 L128.635 717.932 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M127.917 389.365 Q124.769 389.365 122.918 391.517 Q121.089 393.67 121.089 397.42 Q121.089 401.147 122.918 403.323 Q124.769 405.476 127.917 405.476 Q131.066 405.476 132.894 403.323 Q134.746 401.147 134.746 397.42 Q134.746 393.67 132.894 391.517 Q131.066 389.365 127.917 389.365 M137.2 374.712 L137.2 378.971 Q135.441 378.138 133.635 377.698 Q131.853 377.258 130.093 377.258 Q125.464 377.258 123.01 380.383 Q120.58 383.508 120.232 389.828 Q121.598 387.814 123.658 386.749 Q125.718 385.661 128.195 385.661 Q133.404 385.661 136.413 388.832 Q139.445 391.98 139.445 397.42 Q139.445 402.744 136.297 405.962 Q133.149 409.179 127.917 409.179 Q121.922 409.179 118.751 404.596 Q115.58 399.99 115.58 391.263 Q115.58 383.068 119.468 378.207 Q123.357 373.323 129.908 373.323 Q131.667 373.323 133.45 373.67 Q135.255 374.017 137.2 374.712 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M127.593 48.1121 Q124.26 48.1121 122.339 49.8945 Q120.441 51.6769 120.441 54.8019 Q120.441 57.9269 122.339 59.7093 Q124.26 61.4917 127.593 61.4917 Q130.927 61.4917 132.848 59.7093 Q134.769 57.9038 134.769 54.8019 Q134.769 51.6769 132.848 49.8945 Q130.95 48.1121 127.593 48.1121 M122.918 46.1214 Q119.908 45.3807 118.218 43.3205 Q116.552 41.2603 116.552 38.2974 Q116.552 34.1539 119.492 31.7465 Q122.455 29.3391 127.593 29.3391 Q132.755 29.3391 135.695 31.7465 Q138.635 34.1539 138.635 38.2974 Q138.635 41.2603 136.945 43.3205 Q135.279 45.3807 132.292 46.1214 Q135.672 46.9084 137.547 49.2001 Q139.445 51.4918 139.445 54.8019 Q139.445 59.825 136.367 62.5102 Q133.311 65.1954 127.593 65.1954 Q121.876 65.1954 118.797 62.5102 Q115.742 59.825 115.742 54.8019 Q115.742 51.4918 117.64 49.2001 Q119.538 46.9084 122.918 46.1214 M121.205 38.7372 Q121.205 41.4224 122.871 42.927 Q124.561 44.4316 127.593 44.4316 Q130.603 44.4316 132.292 42.927 Q134.005 41.4224 134.005 38.7372 Q134.005 36.052 132.292 34.5474 Q130.603 33.0428 127.593 33.0428 Q124.561 33.0428 122.871 34.5474 Q121.205 36.052 121.205 38.7372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M16.4842 1086.54 L16.4842 1077.88 L56.238 1056.81 L16.4842 1056.81 L16.4842 1050.57 L64.0042 1050.57 L64.0042 1059.23 L24.2503 1080.3 L64.0042 1080.3 L64.0042 1086.54 L16.4842 1086.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M49.9359 1038.63 L28.3562 1038.63 L28.3562 1032.78 L49.7131 1032.78 Q54.7739 1032.78 57.3202 1030.8 Q59.8346 1028.83 59.8346 1024.88 Q59.8346 1020.14 56.8109 1017.4 Q53.7872 1014.63 48.5673 1014.63 L28.3562 1014.63 L28.3562 1008.78 L64.0042 1008.78 L64.0042 1014.63 L58.5296 1014.63 Q61.7762 1016.77 63.3676 1019.6 Q64.9272 1022.4 64.9272 1026.12 Q64.9272 1032.27 61.1078 1035.45 Q57.2883 1038.63 49.9359 1038.63 M27.4968 1023.9 L27.4968 1023.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M35.1993 968.961 Q31.2526 966.765 29.3747 963.709 Q27.4968 960.653 27.4968 956.516 Q27.4968 950.946 31.4117 947.922 Q35.2948 944.898 42.4881 944.898 L64.0042 944.898 L64.0042 950.787 L42.679 950.787 Q37.5546 950.787 35.072 952.601 Q32.5894 954.415 32.5894 958.139 Q32.5894 962.69 35.6131 965.332 Q38.6368 967.974 43.8567 967.974 L64.0042 967.974 L64.0042 973.862 L42.679 973.862 Q37.5228 973.862 35.072 975.677 Q32.5894 977.491 32.5894 981.278 Q32.5894 985.766 35.6449 988.408 Q38.6686 991.05 43.8567 991.05 L64.0042 991.05 L64.0042 996.938 L28.3562 996.938 L28.3562 991.05 L33.8944 991.05 Q30.616 989.045 29.0564 986.244 Q27.4968 983.443 27.4968 979.591 Q27.4968 975.708 29.4702 973.003 Q31.4436 970.266 35.1993 968.961 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M46.212 907.627 Q39.7508 907.627 36.0905 910.301 Q32.3984 912.942 32.3984 917.589 Q32.3984 922.236 36.0905 924.91 Q39.7508 927.552 46.212 927.552 Q52.6732 927.552 56.3653 924.91 Q60.0256 922.236 60.0256 917.589 Q60.0256 912.942 56.3653 910.301 Q52.6732 907.627 46.212 907.627 M33.7671 927.552 Q30.5842 925.706 29.0564 922.905 Q27.4968 920.072 27.4968 916.157 Q27.4968 909.664 32.6531 905.622 Q37.8093 901.548 46.212 901.548 Q54.6147 901.548 59.771 905.622 Q64.9272 909.664 64.9272 916.157 Q64.9272 920.072 63.3994 922.905 Q61.8398 925.706 58.657 927.552 L64.0042 927.552 L64.0042 933.44 L14.479 933.44 L14.479 927.552 L33.7671 927.552 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M44.7161 861.348 L47.5806 861.348 L47.5806 888.275 Q53.6281 887.893 56.8109 884.647 Q59.9619 881.369 59.9619 875.544 Q59.9619 872.17 59.1344 869.019 Q58.3069 865.836 56.6518 862.717 L62.1899 862.717 Q63.5267 865.868 64.227 869.178 Q64.9272 872.488 64.9272 875.894 Q64.9272 884.424 59.9619 889.421 Q54.9967 894.386 46.5303 894.386 Q37.7774 894.386 32.6531 889.676 Q27.4968 884.933 27.4968 876.913 Q27.4968 869.719 32.1438 865.55 Q36.7589 861.348 44.7161 861.348 M42.9973 867.205 Q38.1912 867.269 35.3266 869.91 Q32.4621 872.52 32.4621 876.849 Q32.4621 881.751 35.2312 884.711 Q38.0002 887.639 43.0292 888.084 L42.9973 867.205 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M33.8307 831.08 Q33.2578 832.066 33.0032 833.244 Q32.7167 834.39 32.7167 835.79 Q32.7167 840.755 35.9632 843.429 Q39.1779 846.071 45.2253 846.071 L64.0042 846.071 L64.0042 851.959 L28.3562 851.959 L28.3562 846.071 L33.8944 846.071 Q30.6479 844.225 29.0883 841.265 Q27.4968 838.305 27.4968 834.071 Q27.4968 833.467 27.5923 832.735 Q27.656 832.003 27.8151 831.111 L33.8307 831.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M32.4621 790.403 Q32.4621 795.113 36.1542 797.851 Q39.8145 800.588 46.212 800.588 Q52.6095 800.588 56.3017 797.882 Q59.9619 795.145 59.9619 790.403 Q59.9619 785.724 56.2698 782.987 Q52.5777 780.249 46.212 780.249 Q39.8781 780.249 36.186 782.987 Q32.4621 785.724 32.4621 790.403 M27.4968 790.403 Q27.4968 782.764 32.4621 778.403 Q37.4273 774.043 46.212 774.043 Q54.9649 774.043 59.9619 778.403 Q64.9272 782.764 64.9272 790.403 Q64.9272 798.073 59.9619 802.434 Q54.9649 806.763 46.212 806.763 Q37.4273 806.763 32.4621 802.434 Q27.4968 798.073 27.4968 790.403 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M14.479 746.288 L19.3487 746.288 L19.3487 751.89 Q19.3487 755.041 20.6219 756.282 Q21.895 757.492 25.2052 757.492 L28.3562 757.492 L28.3562 747.848 L32.9077 747.848 L32.9077 757.492 L64.0042 757.492 L64.0042 763.38 L32.9077 763.38 L32.9077 768.982 L28.3562 768.982 L28.3562 763.38 L25.8736 763.38 Q19.9216 763.38 17.2162 760.611 Q14.479 757.842 14.479 751.826 L14.479 746.288 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M29.7248 695.013 L35.1993 695.013 Q33.8307 697.495 33.1623 700.01 Q32.4621 702.492 32.4621 705.039 Q32.4621 710.736 36.0905 713.887 Q39.6872 717.038 46.212 717.038 Q52.7369 717.038 56.3653 713.887 Q59.9619 710.736 59.9619 705.039 Q59.9619 702.492 59.2935 700.01 Q58.5933 697.495 57.2247 695.013 L62.6355 695.013 Q63.7814 697.463 64.3543 700.105 Q64.9272 702.715 64.9272 705.675 Q64.9272 713.728 59.8664 718.47 Q54.8057 723.213 46.212 723.213 Q37.491 723.213 32.4939 718.438 Q27.4968 713.632 27.4968 705.293 Q27.4968 702.588 28.0697 700.01 Q28.6108 697.431 29.7248 695.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M42.4881 655.195 L64.0042 655.195 L64.0042 661.051 L42.679 661.051 Q37.6183 661.051 35.1038 663.025 Q32.5894 664.998 32.5894 668.945 Q32.5894 673.687 35.6131 676.425 Q38.6368 679.162 43.8567 679.162 L64.0042 679.162 L64.0042 685.05 L14.479 685.05 L14.479 679.162 L33.8944 679.162 Q30.6797 677.061 29.0883 674.228 Q27.4968 671.364 27.4968 667.64 Q27.4968 661.497 31.3163 658.346 Q35.1038 655.195 42.4881 655.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M46.0847 627.313 Q46.0847 634.411 47.7079 637.148 Q49.3312 639.886 53.2461 639.886 Q56.3653 639.886 58.2114 637.848 Q60.0256 635.78 60.0256 632.247 Q60.0256 627.377 56.5881 624.449 Q53.1188 621.489 47.3897 621.489 L46.0847 621.489 L46.0847 627.313 M43.6657 615.632 L64.0042 615.632 L64.0042 621.489 L58.5933 621.489 Q61.8398 623.494 63.3994 626.486 Q64.9272 629.478 64.9272 633.806 Q64.9272 639.281 61.8716 642.527 Q58.7843 645.742 53.6281 645.742 Q47.6125 645.742 44.5569 641.732 Q41.5014 637.689 41.5014 629.7 L41.5014 621.489 L40.9285 621.489 Q36.8862 621.489 34.6901 624.162 Q32.4621 626.804 32.4621 631.61 Q32.4621 634.666 33.1941 637.562 Q33.9262 640.458 35.3903 643.132 L29.9795 643.132 Q28.7381 639.917 28.1334 636.894 Q27.4968 633.87 27.4968 631.005 Q27.4968 623.271 31.5072 619.452 Q35.5176 615.632 43.6657 615.632 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M33.8307 582.912 Q33.2578 583.899 33.0032 585.077 Q32.7167 586.223 32.7167 587.623 Q32.7167 592.588 35.9632 595.262 Q39.1779 597.904 45.2253 597.904 L64.0042 597.904 L64.0042 603.792 L28.3562 603.792 L28.3562 597.904 L33.8944 597.904 Q30.6479 596.058 29.0883 593.098 Q27.4968 590.138 27.4968 585.904 Q27.4968 585.3 27.5923 584.568 Q27.656 583.835 27.8151 582.944 L33.8307 582.912 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M46.0847 560.569 Q46.0847 567.667 47.7079 570.404 Q49.3312 573.141 53.2461 573.141 Q56.3653 573.141 58.2114 571.104 Q60.0256 569.035 60.0256 565.502 Q60.0256 560.632 56.5881 557.704 Q53.1188 554.744 47.3897 554.744 L46.0847 554.744 L46.0847 560.569 M43.6657 548.888 L64.0042 548.888 L64.0042 554.744 L58.5933 554.744 Q61.8398 556.749 63.3994 559.741 Q64.9272 562.733 64.9272 567.062 Q64.9272 572.536 61.8716 575.783 Q58.7843 578.998 53.6281 578.998 Q47.6125 578.998 44.5569 574.987 Q41.5014 570.945 41.5014 562.956 L41.5014 554.744 L40.9285 554.744 Q36.8862 554.744 34.6901 557.418 Q32.4621 560.06 32.4621 564.866 Q32.4621 567.921 33.1941 570.818 Q33.9262 573.714 35.3903 576.388 L29.9795 576.388 Q28.7381 573.173 28.1334 570.149 Q27.4968 567.125 27.4968 564.261 Q27.4968 556.527 31.5072 552.707 Q35.5176 548.888 43.6657 548.888 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M29.7248 511.171 L35.1993 511.171 Q33.8307 513.654 33.1623 516.168 Q32.4621 518.651 32.4621 521.197 Q32.4621 526.894 36.0905 530.045 Q39.6872 533.196 46.212 533.196 Q52.7369 533.196 56.3653 530.045 Q59.9619 526.894 59.9619 521.197 Q59.9619 518.651 59.2935 516.168 Q58.5933 513.654 57.2247 511.171 L62.6355 511.171 Q63.7814 513.622 64.3543 516.263 Q64.9272 518.873 64.9272 521.833 Q64.9272 529.886 59.8664 534.629 Q54.8057 539.371 46.212 539.371 Q37.491 539.371 32.4939 534.597 Q27.4968 529.791 27.4968 521.452 Q27.4968 518.746 28.0697 516.168 Q28.6108 513.59 29.7248 511.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M18.2347 495.193 L28.3562 495.193 L28.3562 483.13 L32.9077 483.13 L32.9077 495.193 L52.2594 495.193 Q56.6199 495.193 57.8613 494.015 Q59.1026 492.806 59.1026 489.146 L59.1026 483.13 L64.0042 483.13 L64.0042 489.146 Q64.0042 495.925 61.4897 498.503 Q58.9434 501.081 52.2594 501.081 L32.9077 501.081 L32.9077 505.378 L28.3562 505.378 L28.3562 501.081 L18.2347 501.081 L18.2347 495.193 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M44.7161 444.936 L47.5806 444.936 L47.5806 471.863 Q53.6281 471.481 56.8109 468.234 Q59.9619 464.956 59.9619 459.131 Q59.9619 455.757 59.1344 452.606 Q58.3069 449.424 56.6518 446.304 L62.1899 446.304 Q63.5267 449.455 64.227 452.766 Q64.9272 456.076 64.9272 459.481 Q64.9272 468.011 59.9619 473.009 Q54.9967 477.974 46.5303 477.974 Q37.7774 477.974 32.6531 473.263 Q27.4968 468.521 27.4968 460.5 Q27.4968 453.307 32.1438 449.137 Q36.7589 444.936 44.7161 444.936 M42.9973 450.792 Q38.1912 450.856 35.3266 453.498 Q32.4621 456.108 32.4621 460.436 Q32.4621 465.338 35.2312 468.298 Q38.0002 471.226 43.0292 471.672 L42.9973 450.792 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M33.8307 414.667 Q33.2578 415.653 33.0032 416.831 Q32.7167 417.977 32.7167 419.377 Q32.7167 424.343 35.9632 427.016 Q39.1779 429.658 45.2253 429.658 L64.0042 429.658 L64.0042 435.546 L28.3562 435.546 L28.3562 429.658 L33.8944 429.658 Q30.6479 427.812 29.0883 424.852 Q27.4968 421.892 27.4968 417.659 Q27.4968 417.054 27.5923 416.322 Q27.656 415.59 27.8151 414.699 L33.8307 414.667 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M29.4065 385.798 L34.9447 385.798 Q33.6716 388.281 33.035 390.955 Q32.3984 393.628 32.3984 396.493 Q32.3984 400.853 33.7352 403.049 Q35.072 405.214 37.7456 405.214 Q39.7826 405.214 40.9603 403.654 Q42.1061 402.095 43.1565 397.384 L43.6021 395.379 Q44.9389 389.14 47.3897 386.53 Q49.8086 383.889 54.1691 383.889 Q59.1344 383.889 62.0308 387.835 Q64.9272 391.75 64.9272 398.625 Q64.9272 401.49 64.3543 404.609 Q63.8132 407.696 62.6992 411.134 L56.6518 411.134 Q58.3387 407.887 59.198 404.736 Q60.0256 401.585 60.0256 398.498 Q60.0256 394.36 58.6251 392.132 Q57.1929 389.904 54.6147 389.904 Q52.2276 389.904 50.9545 391.527 Q49.6813 393.119 48.5037 398.562 L48.0262 400.599 Q46.8804 406.041 44.5251 408.46 Q42.138 410.879 38.0002 410.879 Q32.9713 410.879 30.2341 407.314 Q27.4968 403.75 27.4968 397.193 Q27.4968 393.946 27.9743 391.082 Q28.4517 388.217 29.4065 385.798 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip492)" d="M295.201 47.2441 L295.201 1423.18 L562.484 1423.18 L562.484 47.2441 L295.201 47.2441 L295.201 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="295.201,47.2441 295.201,1423.18 562.484,1423.18 562.484,47.2441 295.201,47.2441 "/>
-<path clip-path="url(#clip492)" d="M629.304 563.22 L629.304 1423.18 L896.587 1423.18 L896.587 563.22 L629.304 563.22 L629.304 563.22  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="629.304,563.22 629.304,1423.18 896.587,1423.18 896.587,563.22 629.304,563.22 "/>
-<path clip-path="url(#clip492)" d="M963.408 1079.2 L963.408 1423.18 L1230.69 1423.18 L1230.69 1079.2 L963.408 1079.2 L963.408 1079.2  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="963.408,1079.2 963.408,1423.18 1230.69,1423.18 1230.69,1079.2 963.408,1079.2 "/>
-<path clip-path="url(#clip492)" d="M1297.51 1251.19 L1297.51 1423.18 L1564.79 1423.18 L1564.79 1251.19 L1297.51 1251.19 L1297.51 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1297.51,1251.19 1297.51,1423.18 1564.79,1423.18 1564.79,1251.19 1297.51,1251.19 "/>
-<path clip-path="url(#clip492)" d="M1631.61 1251.19 L1631.61 1423.18 L1898.9 1423.18 L1898.9 1251.19 L1631.61 1251.19 L1631.61 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1631.61,1251.19 1631.61,1423.18 1898.9,1423.18 1898.9,1251.19 1631.61,1251.19 "/>
-<path clip-path="url(#clip492)" d="M1965.72 1251.19 L1965.72 1423.18 L2233 1423.18 L2233 1251.19 L1965.72 1251.19 L1965.72 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1965.72,1251.19 1965.72,1423.18 2233,1423.18 2233,1251.19 1965.72,1251.19 "/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="428.842" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="762.946" cy="563.22" r="2"/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1097.05" cy="1079.2" r="2"/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1431.15" cy="1251.19" r="2"/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1765.26" cy="1251.19" r="2"/>
-<circle clip-path="url(#clip492)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2099.36" cy="1251.19" r="2"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="428.842,1423.18 428.842,47.2441 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="762.946,1423.18 762.946,47.2441 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1097.05,1423.18 1097.05,47.2441 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1431.15,1423.18 1431.15,47.2441 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1765.26,1423.18 1765.26,47.2441 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2099.36,1423.18 2099.36,47.2441 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="428.842,1423.18 428.842,1404.28 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="762.946,1423.18 762.946,1404.28 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1097.05,1423.18 1097.05,1404.28 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1431.15,1423.18 1431.15,1404.28 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1765.26,1423.18 1765.26,1404.28 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2099.36,1423.18 2099.36,1404.28 "/>
+<path clip-path="url(#clip310)" d="M378.6 1472.64 Q378.6 1467.94 376.655 1465.28 Q374.734 1462.59 371.354 1462.59 Q367.975 1462.59 366.03 1465.28 Q364.109 1467.94 364.109 1472.64 Q364.109 1477.34 366.03 1480.02 Q367.975 1482.69 371.354 1482.69 Q374.734 1482.69 376.655 1480.02 Q378.6 1477.34 378.6 1472.64 M364.109 1463.59 Q365.452 1461.27 367.489 1460.16 Q369.549 1459.03 372.396 1459.03 Q377.118 1459.03 380.058 1462.78 Q383.021 1466.53 383.021 1472.64 Q383.021 1478.75 380.058 1482.5 Q377.118 1486.25 372.396 1486.25 Q369.549 1486.25 367.489 1485.14 Q365.452 1484.01 364.109 1481.69 L364.109 1485.58 L359.827 1485.58 L359.827 1449.56 L364.109 1449.56 L364.109 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M405.104 1463.64 Q404.387 1463.22 403.53 1463.03 Q402.697 1462.83 401.678 1462.83 Q398.067 1462.83 396.123 1465.19 Q394.201 1467.52 394.201 1471.92 L394.201 1485.58 L389.919 1485.58 L389.919 1459.65 L394.201 1459.65 L394.201 1463.68 Q395.544 1461.32 397.697 1460.19 Q399.85 1459.03 402.928 1459.03 Q403.368 1459.03 403.9 1459.1 Q404.433 1459.14 405.081 1459.26 L405.104 1463.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M418.576 1462.64 Q415.15 1462.64 413.16 1465.33 Q411.169 1467.99 411.169 1472.64 Q411.169 1477.29 413.137 1479.98 Q415.127 1482.64 418.576 1482.64 Q421.979 1482.64 423.97 1479.95 Q425.961 1477.27 425.961 1472.64 Q425.961 1468.03 423.97 1465.35 Q421.979 1462.64 418.576 1462.64 M418.576 1459.03 Q424.132 1459.03 427.303 1462.64 Q430.474 1466.25 430.474 1472.64 Q430.474 1479.01 427.303 1482.64 Q424.132 1486.25 418.576 1486.25 Q412.998 1486.25 409.826 1482.64 Q406.678 1479.01 406.678 1472.64 Q406.678 1466.25 409.826 1462.64 Q412.998 1459.03 418.576 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M435.058 1459.65 L439.317 1459.65 L444.641 1479.89 L449.942 1459.65 L454.965 1459.65 L460.289 1479.89 L465.59 1459.65 L469.849 1459.65 L463.067 1485.58 L458.044 1485.58 L452.465 1464.33 L446.863 1485.58 L441.84 1485.58 L435.058 1459.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M497.858 1469.93 L497.858 1485.58 L493.599 1485.58 L493.599 1470.07 Q493.599 1466.39 492.164 1464.56 Q490.729 1462.73 487.858 1462.73 Q484.409 1462.73 482.419 1464.93 Q480.428 1467.13 480.428 1470.93 L480.428 1485.58 L476.145 1485.58 L476.145 1459.65 L480.428 1459.65 L480.428 1463.68 Q481.956 1461.34 484.016 1460.19 Q486.099 1459.03 488.807 1459.03 Q493.275 1459.03 495.567 1461.81 Q497.858 1464.56 497.858 1469.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M733.895 1472.64 Q733.895 1467.94 731.95 1465.28 Q730.029 1462.59 726.65 1462.59 Q723.27 1462.59 721.326 1465.28 Q719.404 1467.94 719.404 1472.64 Q719.404 1477.34 721.326 1480.02 Q723.27 1482.69 726.65 1482.69 Q730.029 1482.69 731.95 1480.02 Q733.895 1477.34 733.895 1472.64 M719.404 1463.59 Q720.747 1461.27 722.784 1460.16 Q724.844 1459.03 727.691 1459.03 Q732.413 1459.03 735.353 1462.78 Q738.316 1466.53 738.316 1472.64 Q738.316 1478.75 735.353 1482.5 Q732.413 1486.25 727.691 1486.25 Q724.844 1486.25 722.784 1485.14 Q720.747 1484.01 719.404 1481.69 L719.404 1485.58 L715.122 1485.58 L715.122 1449.56 L719.404 1449.56 L719.404 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M745.376 1449.56 L749.636 1449.56 L749.636 1485.58 L745.376 1485.58 L745.376 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M758.108 1475.35 L758.108 1459.65 L762.367 1459.65 L762.367 1475.19 Q762.367 1478.87 763.802 1480.72 Q765.237 1482.55 768.108 1482.55 Q771.557 1482.55 773.547 1480.35 Q775.561 1478.15 775.561 1474.35 L775.561 1459.65 L779.821 1459.65 L779.821 1485.58 L775.561 1485.58 L775.561 1481.6 Q774.01 1483.96 771.95 1485.12 Q769.913 1486.25 767.205 1486.25 Q762.737 1486.25 760.423 1483.47 Q758.108 1480.7 758.108 1475.35 M768.825 1459.03 L768.825 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M810.77 1471.55 L810.77 1473.64 L791.186 1473.64 Q791.464 1478.03 793.825 1480.35 Q796.209 1482.64 800.446 1482.64 Q802.899 1482.64 805.191 1482.04 Q807.506 1481.44 809.774 1480.23 L809.774 1484.26 Q807.483 1485.23 805.075 1485.74 Q802.668 1486.25 800.191 1486.25 Q793.987 1486.25 790.353 1482.64 Q786.742 1479.03 786.742 1472.87 Q786.742 1466.51 790.168 1462.78 Q793.617 1459.03 799.45 1459.03 Q804.682 1459.03 807.714 1462.41 Q810.77 1465.77 810.77 1471.55 M806.51 1470.3 Q806.464 1466.81 804.543 1464.72 Q802.645 1462.64 799.496 1462.64 Q795.932 1462.64 793.779 1464.65 Q791.649 1466.67 791.325 1470.33 L806.51 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1036.92 1487.99 Q1035.12 1492.62 1033.4 1494.03 Q1031.69 1495.44 1028.82 1495.44 L1025.42 1495.44 L1025.42 1491.88 L1027.92 1491.88 Q1029.68 1491.88 1030.65 1491.04 Q1031.62 1490.21 1032.8 1487.11 L1033.57 1485.16 L1023.08 1459.65 L1027.59 1459.65 L1035.7 1479.93 L1043.8 1459.65 L1048.31 1459.65 L1036.92 1487.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1076.37 1471.55 L1076.37 1473.64 L1056.78 1473.64 Q1057.06 1478.03 1059.42 1480.35 Q1061.81 1482.64 1066.04 1482.64 Q1068.5 1482.64 1070.79 1482.04 Q1073.1 1481.44 1075.37 1480.23 L1075.37 1484.26 Q1073.08 1485.23 1070.67 1485.74 Q1068.26 1486.25 1065.79 1486.25 Q1059.58 1486.25 1055.95 1482.64 Q1052.34 1479.03 1052.34 1472.87 Q1052.34 1466.51 1055.76 1462.78 Q1059.21 1459.03 1065.05 1459.03 Q1070.28 1459.03 1073.31 1462.41 Q1076.37 1465.77 1076.37 1471.55 M1072.11 1470.3 Q1072.06 1466.81 1070.14 1464.72 Q1068.24 1462.64 1065.09 1462.64 Q1061.53 1462.64 1059.38 1464.65 Q1057.25 1466.67 1056.92 1470.33 L1072.11 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1083.36 1449.56 L1087.62 1449.56 L1087.62 1485.58 L1083.36 1485.58 L1083.36 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1096.53 1449.56 L1100.79 1449.56 L1100.79 1485.58 L1096.53 1485.58 L1096.53 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1119.75 1462.64 Q1116.32 1462.64 1114.33 1465.33 Q1112.34 1467.99 1112.34 1472.64 Q1112.34 1477.29 1114.31 1479.98 Q1116.3 1482.64 1119.75 1482.64 Q1123.15 1482.64 1125.14 1479.95 Q1127.13 1477.27 1127.13 1472.64 Q1127.13 1468.03 1125.14 1465.35 Q1123.15 1462.64 1119.75 1462.64 M1119.75 1459.03 Q1125.3 1459.03 1128.47 1462.64 Q1131.64 1466.25 1131.64 1472.64 Q1131.64 1479.01 1128.47 1482.64 Q1125.3 1486.25 1119.75 1486.25 Q1114.17 1486.25 1111 1482.64 Q1107.85 1479.01 1107.85 1472.64 Q1107.85 1466.25 1111 1462.64 Q1114.17 1459.03 1119.75 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1136.23 1459.65 L1140.49 1459.65 L1145.81 1479.89 L1151.11 1459.65 L1156.13 1459.65 L1161.46 1479.89 L1166.76 1459.65 L1171.02 1459.65 L1164.24 1485.58 L1159.21 1485.58 L1153.63 1464.33 L1148.03 1485.58 L1143.01 1485.58 L1136.23 1459.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1348.2 1472.64 Q1348.2 1467.94 1346.26 1465.28 Q1344.34 1462.59 1340.96 1462.59 Q1337.58 1462.59 1335.63 1465.28 Q1333.71 1467.94 1333.71 1472.64 Q1333.71 1477.34 1335.63 1480.02 Q1337.58 1482.69 1340.96 1482.69 Q1344.34 1482.69 1346.26 1480.02 Q1348.2 1477.34 1348.2 1472.64 M1333.71 1463.59 Q1335.05 1461.27 1337.09 1460.16 Q1339.15 1459.03 1342 1459.03 Q1346.72 1459.03 1349.66 1462.78 Q1352.62 1466.53 1352.62 1472.64 Q1352.62 1478.75 1349.66 1482.5 Q1346.72 1486.25 1342 1486.25 Q1339.15 1486.25 1337.09 1485.14 Q1335.05 1484.01 1333.71 1481.69 L1333.71 1485.58 L1329.43 1485.58 L1329.43 1449.56 L1333.71 1449.56 L1333.71 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1359.68 1449.56 L1363.94 1449.56 L1363.94 1485.58 L1359.68 1485.58 L1359.68 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1372.41 1475.35 L1372.41 1459.65 L1376.67 1459.65 L1376.67 1475.19 Q1376.67 1478.87 1378.11 1480.72 Q1379.54 1482.55 1382.41 1482.55 Q1385.86 1482.55 1387.85 1480.35 Q1389.87 1478.15 1389.87 1474.35 L1389.87 1459.65 L1394.13 1459.65 L1394.13 1485.58 L1389.87 1485.58 L1389.87 1481.6 Q1388.32 1483.96 1386.26 1485.12 Q1384.22 1486.25 1381.51 1486.25 Q1377.04 1486.25 1374.73 1483.47 Q1372.41 1480.7 1372.41 1475.35 M1383.13 1459.03 L1383.13 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1425.08 1471.55 L1425.08 1473.64 L1405.49 1473.64 Q1405.77 1478.03 1408.13 1480.35 Q1410.52 1482.64 1414.75 1482.64 Q1417.21 1482.64 1419.5 1482.04 Q1421.81 1481.44 1424.08 1480.23 L1424.08 1484.26 Q1421.79 1485.23 1419.38 1485.74 Q1416.97 1486.25 1414.5 1486.25 Q1408.29 1486.25 1404.66 1482.64 Q1401.05 1479.03 1401.05 1472.87 Q1401.05 1466.51 1404.47 1462.78 Q1407.92 1459.03 1413.76 1459.03 Q1418.99 1459.03 1422.02 1462.41 Q1425.08 1465.77 1425.08 1471.55 M1420.82 1470.3 Q1420.77 1466.81 1418.85 1464.72 Q1416.95 1462.64 1413.8 1462.64 Q1410.24 1462.64 1408.09 1464.65 Q1405.96 1466.67 1405.63 1470.33 L1420.82 1470.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1449.13 1472.32 Q1449.13 1467.69 1447.21 1465.14 Q1445.31 1462.59 1441.86 1462.59 Q1438.43 1462.59 1436.51 1465.14 Q1434.61 1467.69 1434.61 1472.32 Q1434.61 1476.92 1436.51 1479.47 Q1438.43 1482.02 1441.86 1482.02 Q1445.31 1482.02 1447.21 1479.47 Q1449.13 1476.92 1449.13 1472.32 M1453.39 1482.36 Q1453.39 1488.98 1450.45 1492.2 Q1447.51 1495.44 1441.44 1495.44 Q1439.2 1495.44 1437.21 1495.09 Q1435.21 1494.77 1433.34 1494.08 L1433.34 1489.93 Q1435.21 1490.95 1437.04 1491.44 Q1438.87 1491.92 1440.77 1491.92 Q1444.96 1491.92 1447.04 1489.72 Q1449.13 1487.55 1449.13 1483.13 L1449.13 1481.02 Q1447.81 1483.31 1445.75 1484.45 Q1443.69 1485.58 1440.82 1485.58 Q1436.05 1485.58 1433.13 1481.95 Q1430.21 1478.31 1430.21 1472.32 Q1430.21 1466.3 1433.13 1462.66 Q1436.05 1459.03 1440.82 1459.03 Q1443.69 1459.03 1445.75 1460.16 Q1447.81 1461.3 1449.13 1463.59 L1449.13 1459.65 L1453.39 1459.65 L1453.39 1482.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1477.18 1463.64 Q1476.46 1463.22 1475.61 1463.03 Q1474.77 1462.83 1473.76 1462.83 Q1470.15 1462.83 1468.2 1465.19 Q1466.28 1467.52 1466.28 1471.92 L1466.28 1485.58 L1462 1485.58 L1462 1459.65 L1466.28 1459.65 L1466.28 1463.68 Q1467.62 1461.32 1469.77 1460.19 Q1471.93 1459.03 1475.01 1459.03 Q1475.45 1459.03 1475.98 1459.1 Q1476.51 1459.14 1477.16 1459.26 L1477.18 1463.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1493.43 1472.55 Q1488.27 1472.55 1486.28 1473.73 Q1484.29 1474.91 1484.29 1477.76 Q1484.29 1480.02 1485.77 1481.37 Q1487.27 1482.69 1489.84 1482.69 Q1493.39 1482.69 1495.52 1480.19 Q1497.67 1477.66 1497.67 1473.5 L1497.67 1472.55 L1493.43 1472.55 M1501.93 1470.79 L1501.93 1485.58 L1497.67 1485.58 L1497.67 1481.64 Q1496.21 1484.01 1494.03 1485.14 Q1491.86 1486.25 1488.71 1486.25 Q1484.73 1486.25 1482.37 1484.03 Q1480.03 1481.78 1480.03 1478.03 Q1480.03 1473.66 1482.95 1471.44 Q1485.89 1469.21 1491.7 1469.21 L1497.67 1469.21 L1497.67 1468.8 Q1497.67 1465.86 1495.72 1464.26 Q1493.8 1462.64 1490.31 1462.64 Q1488.08 1462.64 1485.98 1463.17 Q1483.87 1463.7 1481.93 1464.77 L1481.93 1460.83 Q1484.27 1459.93 1486.46 1459.49 Q1488.66 1459.03 1490.75 1459.03 Q1496.37 1459.03 1499.15 1461.95 Q1501.93 1464.86 1501.93 1470.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1521.49 1487.99 Q1519.68 1492.62 1517.97 1494.03 Q1516.26 1495.44 1513.39 1495.44 L1509.98 1495.44 L1509.98 1491.88 L1512.48 1491.88 Q1514.24 1491.88 1515.21 1491.04 Q1516.19 1490.21 1517.37 1487.11 L1518.13 1485.16 L1507.64 1459.65 L1512.16 1459.65 L1520.26 1479.93 L1528.36 1459.65 L1532.88 1459.65 L1521.49 1487.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1736.45 1472.32 Q1736.45 1467.69 1734.53 1465.14 Q1732.63 1462.59 1729.18 1462.59 Q1725.75 1462.59 1723.83 1465.14 Q1721.93 1467.69 1721.93 1472.32 Q1721.93 1476.92 1723.83 1479.47 Q1725.75 1482.02 1729.18 1482.02 Q1732.63 1482.02 1734.53 1479.47 Q1736.45 1476.92 1736.45 1472.32 M1740.71 1482.36 Q1740.71 1488.98 1737.77 1492.2 Q1734.83 1495.44 1728.76 1495.44 Q1726.52 1495.44 1724.53 1495.09 Q1722.54 1494.77 1720.66 1494.08 L1720.66 1489.93 Q1722.54 1490.95 1724.36 1491.44 Q1726.19 1491.92 1728.09 1491.92 Q1732.28 1491.92 1734.36 1489.72 Q1736.45 1487.55 1736.45 1483.13 L1736.45 1481.02 Q1735.13 1483.31 1733.07 1484.45 Q1731.01 1485.58 1728.14 1485.58 Q1723.37 1485.58 1720.45 1481.95 Q1717.54 1478.31 1717.54 1472.32 Q1717.54 1466.3 1720.45 1462.66 Q1723.37 1459.03 1728.14 1459.03 Q1731.01 1459.03 1733.07 1460.16 Q1735.13 1461.3 1736.45 1463.59 L1736.45 1459.65 L1740.71 1459.65 L1740.71 1482.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1759.53 1462.64 Q1756.1 1462.64 1754.11 1465.33 Q1752.12 1467.99 1752.12 1472.64 Q1752.12 1477.29 1754.09 1479.98 Q1756.08 1482.64 1759.53 1482.64 Q1762.93 1482.64 1764.92 1479.95 Q1766.91 1477.27 1766.91 1472.64 Q1766.91 1468.03 1764.92 1465.35 Q1762.93 1462.64 1759.53 1462.64 M1759.53 1459.03 Q1765.08 1459.03 1768.25 1462.64 Q1771.42 1466.25 1771.42 1472.64 Q1771.42 1479.01 1768.25 1482.64 Q1765.08 1486.25 1759.53 1486.25 Q1753.95 1486.25 1750.78 1482.64 Q1747.63 1479.01 1747.63 1472.64 Q1747.63 1466.25 1750.78 1462.64 Q1753.95 1459.03 1759.53 1459.03 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1778.48 1449.56 L1782.74 1449.56 L1782.74 1485.58 L1778.48 1485.58 L1778.48 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1808.72 1463.59 L1808.72 1449.56 L1812.98 1449.56 L1812.98 1485.58 L1808.72 1485.58 L1808.72 1481.69 Q1807.37 1484.01 1805.31 1485.14 Q1803.28 1486.25 1800.41 1486.25 Q1795.71 1486.25 1792.74 1482.5 Q1789.8 1478.75 1789.8 1472.64 Q1789.8 1466.53 1792.74 1462.78 Q1795.71 1459.03 1800.41 1459.03 Q1803.28 1459.03 1805.31 1460.16 Q1807.37 1461.27 1808.72 1463.59 M1794.2 1472.64 Q1794.2 1477.34 1796.12 1480.02 Q1798.07 1482.69 1801.45 1482.69 Q1804.83 1482.69 1806.77 1480.02 Q1808.72 1477.34 1808.72 1472.64 Q1808.72 1467.94 1806.77 1465.28 Q1804.83 1462.59 1801.45 1462.59 Q1798.07 1462.59 1796.12 1465.28 Q1794.2 1467.94 1794.2 1472.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M2057.44 1472.64 Q2057.44 1467.94 2055.49 1465.28 Q2053.57 1462.59 2050.19 1462.59 Q2046.81 1462.59 2044.87 1465.28 Q2042.95 1467.94 2042.95 1472.64 Q2042.95 1477.34 2044.87 1480.02 Q2046.81 1482.69 2050.19 1482.69 Q2053.57 1482.69 2055.49 1480.02 Q2057.44 1477.34 2057.44 1472.64 M2042.95 1463.59 Q2044.29 1461.27 2046.33 1460.16 Q2048.39 1459.03 2051.23 1459.03 Q2055.96 1459.03 2058.9 1462.78 Q2061.86 1466.53 2061.86 1472.64 Q2061.86 1478.75 2058.9 1482.5 Q2055.96 1486.25 2051.23 1486.25 Q2048.39 1486.25 2046.33 1485.14 Q2044.29 1484.01 2042.95 1481.69 L2042.95 1485.58 L2038.66 1485.58 L2038.66 1449.56 L2042.95 1449.56 L2042.95 1463.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M2068.92 1449.56 L2073.18 1449.56 L2073.18 1485.58 L2068.92 1485.58 L2068.92 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M2093.87 1472.55 Q2088.71 1472.55 2086.72 1473.73 Q2084.73 1474.91 2084.73 1477.76 Q2084.73 1480.02 2086.21 1481.37 Q2087.72 1482.69 2090.28 1482.69 Q2093.83 1482.69 2095.96 1480.19 Q2098.11 1477.66 2098.11 1473.5 L2098.11 1472.55 L2093.87 1472.55 M2102.37 1470.79 L2102.37 1485.58 L2098.11 1485.58 L2098.11 1481.64 Q2096.65 1484.01 2094.47 1485.14 Q2092.3 1486.25 2089.15 1486.25 Q2085.17 1486.25 2082.81 1484.03 Q2080.47 1481.78 2080.47 1478.03 Q2080.47 1473.66 2083.39 1471.44 Q2086.33 1469.21 2092.14 1469.21 L2098.11 1469.21 L2098.11 1468.8 Q2098.11 1465.86 2096.16 1464.26 Q2094.24 1462.64 2090.75 1462.64 Q2088.53 1462.64 2086.42 1463.17 Q2084.31 1463.7 2082.37 1464.77 L2082.37 1460.83 Q2084.71 1459.93 2086.91 1459.49 Q2089.1 1459.03 2091.19 1459.03 Q2096.81 1459.03 2099.59 1461.95 Q2102.37 1464.86 2102.37 1470.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M2129.8 1460.65 L2129.8 1464.63 Q2127.99 1463.64 2126.16 1463.15 Q2124.36 1462.64 2122.51 1462.64 Q2118.36 1462.64 2116.07 1465.28 Q2113.78 1467.89 2113.78 1472.64 Q2113.78 1477.39 2116.07 1480.02 Q2118.36 1482.64 2122.51 1482.64 Q2124.36 1482.64 2126.16 1482.15 Q2127.99 1481.64 2129.8 1480.65 L2129.8 1484.58 Q2128.02 1485.42 2126.09 1485.83 Q2124.2 1486.25 2122.04 1486.25 Q2116.19 1486.25 2112.74 1482.57 Q2109.29 1478.89 2109.29 1472.64 Q2109.29 1466.3 2112.76 1462.66 Q2116.26 1459.03 2122.32 1459.03 Q2124.29 1459.03 2126.16 1459.45 Q2128.04 1459.84 2129.8 1460.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M2137.04 1449.56 L2141.33 1449.56 L2141.33 1470.83 L2154.03 1459.65 L2159.47 1459.65 L2145.72 1471.78 L2160.05 1485.58 L2154.5 1485.58 L2141.33 1472.92 L2141.33 1485.58 L2137.04 1485.58 L2137.04 1449.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1116.75 1520.52 L1146.8 1520.52 L1146.8 1525.93 L1123.18 1525.93 L1123.18 1540 L1145.81 1540 L1145.81 1545.41 L1123.18 1545.41 L1123.18 1562.63 L1147.37 1562.63 L1147.37 1568.04 L1116.75 1568.04 L1116.75 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1172.51 1571.35 Q1170.03 1577.72 1167.68 1579.66 Q1165.32 1581.6 1161.37 1581.6 L1156.7 1581.6 L1156.7 1576.7 L1160.13 1576.7 Q1162.55 1576.7 1163.89 1575.56 Q1165.23 1574.41 1166.85 1570.14 L1167.9 1567.47 L1153.48 1532.4 L1159.69 1532.4 L1170.83 1560.28 L1181.97 1532.4 L1188.17 1532.4 L1172.51 1571.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1226.75 1548.76 L1226.75 1551.62 L1199.82 1551.62 Q1200.2 1557.67 1203.45 1560.85 Q1206.73 1564 1212.55 1564 Q1215.93 1564 1219.08 1563.17 Q1222.26 1562.35 1225.38 1560.69 L1225.38 1566.23 Q1222.23 1567.57 1218.92 1568.27 Q1215.61 1568.97 1212.2 1568.97 Q1203.67 1568.97 1198.68 1564 Q1193.71 1559.04 1193.71 1550.57 Q1193.71 1541.82 1198.42 1536.69 Q1203.16 1531.54 1211.19 1531.54 Q1218.38 1531.54 1222.55 1536.18 Q1226.75 1540.8 1226.75 1548.76 M1220.89 1547.04 Q1220.83 1542.23 1218.19 1539.37 Q1215.58 1536.5 1211.25 1536.5 Q1206.35 1536.5 1203.39 1539.27 Q1200.46 1542.04 1200.01 1547.07 L1220.89 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1282.74 1533.76 L1282.74 1539.24 Q1280.25 1537.87 1277.74 1537.2 Q1275.26 1536.5 1272.71 1536.5 Q1267.01 1536.5 1263.86 1540.13 Q1260.71 1543.73 1260.71 1550.25 Q1260.71 1556.78 1263.86 1560.4 Q1267.01 1564 1272.71 1564 Q1275.26 1564 1277.74 1563.33 Q1280.25 1562.63 1282.74 1561.26 L1282.74 1566.68 Q1280.29 1567.82 1277.64 1568.39 Q1275.03 1568.97 1272.07 1568.97 Q1264.02 1568.97 1259.28 1563.91 Q1254.54 1558.85 1254.54 1550.25 Q1254.54 1541.53 1259.31 1536.53 Q1264.12 1531.54 1272.46 1531.54 Q1275.16 1531.54 1277.74 1532.11 Q1280.32 1532.65 1282.74 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1306.73 1536.5 Q1302.02 1536.5 1299.29 1540.19 Q1296.55 1543.85 1296.55 1550.25 Q1296.55 1556.65 1299.26 1560.34 Q1301.99 1564 1306.73 1564 Q1311.41 1564 1314.15 1560.31 Q1316.89 1556.62 1316.89 1550.25 Q1316.89 1543.92 1314.15 1540.23 Q1311.41 1536.5 1306.73 1536.5 M1306.73 1531.54 Q1314.37 1531.54 1318.73 1536.5 Q1323.09 1541.47 1323.09 1550.25 Q1323.09 1559 1318.73 1564 Q1314.37 1568.97 1306.73 1568.97 Q1299.06 1568.97 1294.7 1564 Q1290.38 1559 1290.38 1550.25 Q1290.38 1541.47 1294.7 1536.5 Q1299.06 1531.54 1306.73 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1332.8 1518.52 L1338.66 1518.52 L1338.66 1568.04 L1332.8 1568.04 L1332.8 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1364.73 1536.5 Q1360.02 1536.5 1357.28 1540.19 Q1354.54 1543.85 1354.54 1550.25 Q1354.54 1556.65 1357.25 1560.34 Q1359.98 1564 1364.73 1564 Q1369.41 1564 1372.14 1560.31 Q1374.88 1556.62 1374.88 1550.25 Q1374.88 1543.92 1372.14 1540.23 Q1369.41 1536.5 1364.73 1536.5 M1364.73 1531.54 Q1372.37 1531.54 1376.73 1536.5 Q1381.09 1541.47 1381.09 1550.25 Q1381.09 1559 1376.73 1564 Q1372.37 1568.97 1364.73 1568.97 Q1357.06 1568.97 1352.7 1564 Q1348.37 1559 1348.37 1550.25 Q1348.37 1541.47 1352.7 1536.5 Q1357.06 1531.54 1364.73 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M1411.45 1537.87 Q1410.46 1537.3 1409.29 1537.04 Q1408.14 1536.76 1406.74 1536.76 Q1401.77 1536.76 1399.1 1540 Q1396.46 1543.22 1396.46 1549.27 L1396.46 1568.04 L1390.57 1568.04 L1390.57 1532.4 L1396.46 1532.4 L1396.46 1537.93 Q1398.31 1534.69 1401.27 1533.13 Q1404.23 1531.54 1408.46 1531.54 Q1409.06 1531.54 1409.8 1531.63 Q1410.53 1531.7 1411.42 1531.85 L1411.45 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,1079.2 2352.76,1079.2 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,735.212 2352.76,735.212 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,391.228 2352.76,391.228 "/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="175.445,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 175.445,47.2441 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1423.18 194.343,1423.18 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,1079.2 194.343,1079.2 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,735.212 194.343,735.212 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,391.228 194.343,391.228 "/>
+<polyline clip-path="url(#clip310)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="175.445,47.2441 194.343,47.2441 "/>
+<path clip-path="url(#clip310)" d="M127.501 1408.98 Q123.89 1408.98 122.061 1412.54 Q120.255 1416.08 120.255 1423.21 Q120.255 1430.32 122.061 1433.89 Q123.89 1437.43 127.501 1437.43 Q131.135 1437.43 132.941 1433.89 Q134.769 1430.32 134.769 1423.21 Q134.769 1416.08 132.941 1412.54 Q131.135 1408.98 127.501 1408.98 M127.501 1405.27 Q133.311 1405.27 136.367 1409.88 Q139.445 1414.46 139.445 1423.21 Q139.445 1431.94 136.367 1436.55 Q133.311 1441.13 127.501 1441.13 Q121.691 1441.13 118.612 1436.55 Q115.556 1431.94 115.556 1423.21 Q115.556 1414.46 118.612 1409.88 Q121.691 1405.27 127.501 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M123.126 1092.54 L139.445 1092.54 L139.445 1096.48 L117.501 1096.48 L117.501 1092.54 Q120.163 1089.79 124.746 1085.16 Q129.353 1080.5 130.533 1079.16 Q132.779 1076.64 133.658 1074.9 Q134.561 1073.14 134.561 1071.45 Q134.561 1068.7 132.617 1066.96 Q130.695 1065.23 127.593 1065.23 Q125.394 1065.23 122.941 1065.99 Q120.51 1066.75 117.732 1068.3 L117.732 1063.58 Q120.556 1062.45 123.01 1061.87 Q125.464 1061.29 127.501 1061.29 Q132.871 1061.29 136.066 1063.98 Q139.26 1066.66 139.26 1071.15 Q139.26 1073.28 138.45 1075.2 Q137.663 1077.1 135.556 1079.69 Q134.978 1080.36 131.876 1083.58 Q128.774 1086.78 123.126 1092.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M129.862 722.006 L118.056 740.455 L129.862 740.455 L129.862 722.006 M128.635 717.932 L134.515 717.932 L134.515 740.455 L139.445 740.455 L139.445 744.344 L134.515 744.344 L134.515 752.492 L129.862 752.492 L129.862 744.344 L114.26 744.344 L114.26 739.83 L128.635 717.932 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M127.917 389.365 Q124.769 389.365 122.918 391.517 Q121.089 393.67 121.089 397.42 Q121.089 401.147 122.918 403.323 Q124.769 405.476 127.917 405.476 Q131.066 405.476 132.894 403.323 Q134.746 401.147 134.746 397.42 Q134.746 393.67 132.894 391.517 Q131.066 389.365 127.917 389.365 M137.2 374.712 L137.2 378.971 Q135.441 378.138 133.635 377.698 Q131.853 377.258 130.093 377.258 Q125.464 377.258 123.01 380.383 Q120.58 383.508 120.232 389.828 Q121.598 387.814 123.658 386.749 Q125.718 385.661 128.195 385.661 Q133.404 385.661 136.413 388.832 Q139.445 391.98 139.445 397.42 Q139.445 402.744 136.297 405.962 Q133.149 409.179 127.917 409.179 Q121.922 409.179 118.751 404.596 Q115.58 399.99 115.58 391.263 Q115.58 383.068 119.468 378.207 Q123.357 373.323 129.908 373.323 Q131.667 373.323 133.45 373.67 Q135.255 374.017 137.2 374.712 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M127.593 48.1121 Q124.26 48.1121 122.339 49.8945 Q120.441 51.6769 120.441 54.8019 Q120.441 57.9269 122.339 59.7093 Q124.26 61.4917 127.593 61.4917 Q130.927 61.4917 132.848 59.7093 Q134.769 57.9038 134.769 54.8019 Q134.769 51.6769 132.848 49.8945 Q130.95 48.1121 127.593 48.1121 M122.918 46.1214 Q119.908 45.3807 118.218 43.3205 Q116.552 41.2603 116.552 38.2974 Q116.552 34.1539 119.492 31.7465 Q122.455 29.3391 127.593 29.3391 Q132.755 29.3391 135.695 31.7465 Q138.635 34.1539 138.635 38.2974 Q138.635 41.2603 136.945 43.3205 Q135.279 45.3807 132.292 46.1214 Q135.672 46.9084 137.547 49.2001 Q139.445 51.4918 139.445 54.8019 Q139.445 59.825 136.367 62.5102 Q133.311 65.1954 127.593 65.1954 Q121.876 65.1954 118.797 62.5102 Q115.742 59.825 115.742 54.8019 Q115.742 51.4918 117.64 49.2001 Q119.538 46.9084 122.918 46.1214 M121.205 38.7372 Q121.205 41.4224 122.871 42.927 Q124.561 44.4316 127.593 44.4316 Q130.603 44.4316 132.292 42.927 Q134.005 41.4224 134.005 38.7372 Q134.005 36.052 132.292 34.5474 Q130.603 33.0428 127.593 33.0428 Q124.561 33.0428 122.871 34.5474 Q121.205 36.052 121.205 38.7372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M16.4842 1086.54 L16.4842 1077.88 L56.238 1056.81 L16.4842 1056.81 L16.4842 1050.57 L64.0042 1050.57 L64.0042 1059.23 L24.2503 1080.3 L64.0042 1080.3 L64.0042 1086.54 L16.4842 1086.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M49.9359 1038.63 L28.3562 1038.63 L28.3562 1032.78 L49.7131 1032.78 Q54.7739 1032.78 57.3202 1030.8 Q59.8346 1028.83 59.8346 1024.88 Q59.8346 1020.14 56.8109 1017.4 Q53.7872 1014.63 48.5673 1014.63 L28.3562 1014.63 L28.3562 1008.78 L64.0042 1008.78 L64.0042 1014.63 L58.5296 1014.63 Q61.7762 1016.77 63.3676 1019.6 Q64.9272 1022.4 64.9272 1026.12 Q64.9272 1032.27 61.1078 1035.45 Q57.2883 1038.63 49.9359 1038.63 M27.4968 1023.9 L27.4968 1023.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M35.1993 968.961 Q31.2526 966.765 29.3747 963.709 Q27.4968 960.653 27.4968 956.516 Q27.4968 950.946 31.4117 947.922 Q35.2948 944.898 42.4881 944.898 L64.0042 944.898 L64.0042 950.787 L42.679 950.787 Q37.5546 950.787 35.072 952.601 Q32.5894 954.415 32.5894 958.139 Q32.5894 962.69 35.6131 965.332 Q38.6368 967.974 43.8567 967.974 L64.0042 967.974 L64.0042 973.862 L42.679 973.862 Q37.5228 973.862 35.072 975.677 Q32.5894 977.491 32.5894 981.278 Q32.5894 985.766 35.6449 988.408 Q38.6686 991.05 43.8567 991.05 L64.0042 991.05 L64.0042 996.938 L28.3562 996.938 L28.3562 991.05 L33.8944 991.05 Q30.616 989.045 29.0564 986.244 Q27.4968 983.443 27.4968 979.591 Q27.4968 975.708 29.4702 973.003 Q31.4436 970.266 35.1993 968.961 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M46.212 907.627 Q39.7508 907.627 36.0905 910.301 Q32.3984 912.942 32.3984 917.589 Q32.3984 922.236 36.0905 924.91 Q39.7508 927.552 46.212 927.552 Q52.6732 927.552 56.3653 924.91 Q60.0256 922.236 60.0256 917.589 Q60.0256 912.942 56.3653 910.301 Q52.6732 907.627 46.212 907.627 M33.7671 927.552 Q30.5842 925.706 29.0564 922.905 Q27.4968 920.072 27.4968 916.157 Q27.4968 909.664 32.6531 905.622 Q37.8093 901.548 46.212 901.548 Q54.6147 901.548 59.771 905.622 Q64.9272 909.664 64.9272 916.157 Q64.9272 920.072 63.3994 922.905 Q61.8398 925.706 58.657 927.552 L64.0042 927.552 L64.0042 933.44 L14.479 933.44 L14.479 927.552 L33.7671 927.552 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M44.7161 861.348 L47.5806 861.348 L47.5806 888.275 Q53.6281 887.893 56.8109 884.647 Q59.9619 881.369 59.9619 875.544 Q59.9619 872.17 59.1344 869.019 Q58.3069 865.836 56.6518 862.717 L62.1899 862.717 Q63.5267 865.868 64.227 869.178 Q64.9272 872.488 64.9272 875.894 Q64.9272 884.424 59.9619 889.421 Q54.9967 894.386 46.5303 894.386 Q37.7774 894.386 32.6531 889.676 Q27.4968 884.933 27.4968 876.913 Q27.4968 869.719 32.1438 865.55 Q36.7589 861.348 44.7161 861.348 M42.9973 867.205 Q38.1912 867.269 35.3266 869.91 Q32.4621 872.52 32.4621 876.849 Q32.4621 881.751 35.2312 884.711 Q38.0002 887.639 43.0292 888.084 L42.9973 867.205 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M33.8307 831.08 Q33.2578 832.066 33.0032 833.244 Q32.7167 834.39 32.7167 835.79 Q32.7167 840.755 35.9632 843.429 Q39.1779 846.071 45.2253 846.071 L64.0042 846.071 L64.0042 851.959 L28.3562 851.959 L28.3562 846.071 L33.8944 846.071 Q30.6479 844.225 29.0883 841.265 Q27.4968 838.305 27.4968 834.071 Q27.4968 833.467 27.5923 832.735 Q27.656 832.003 27.8151 831.111 L33.8307 831.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M32.4621 790.403 Q32.4621 795.113 36.1542 797.851 Q39.8145 800.588 46.212 800.588 Q52.6095 800.588 56.3017 797.882 Q59.9619 795.145 59.9619 790.403 Q59.9619 785.724 56.2698 782.987 Q52.5777 780.249 46.212 780.249 Q39.8781 780.249 36.186 782.987 Q32.4621 785.724 32.4621 790.403 M27.4968 790.403 Q27.4968 782.764 32.4621 778.403 Q37.4273 774.043 46.212 774.043 Q54.9649 774.043 59.9619 778.403 Q64.9272 782.764 64.9272 790.403 Q64.9272 798.073 59.9619 802.434 Q54.9649 806.763 46.212 806.763 Q37.4273 806.763 32.4621 802.434 Q27.4968 798.073 27.4968 790.403 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M14.479 746.288 L19.3487 746.288 L19.3487 751.89 Q19.3487 755.041 20.6219 756.282 Q21.895 757.492 25.2052 757.492 L28.3562 757.492 L28.3562 747.848 L32.9077 747.848 L32.9077 757.492 L64.0042 757.492 L64.0042 763.38 L32.9077 763.38 L32.9077 768.982 L28.3562 768.982 L28.3562 763.38 L25.8736 763.38 Q19.9216 763.38 17.2162 760.611 Q14.479 757.842 14.479 751.826 L14.479 746.288 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M29.7248 695.013 L35.1993 695.013 Q33.8307 697.495 33.1623 700.01 Q32.4621 702.492 32.4621 705.039 Q32.4621 710.736 36.0905 713.887 Q39.6872 717.038 46.212 717.038 Q52.7369 717.038 56.3653 713.887 Q59.9619 710.736 59.9619 705.039 Q59.9619 702.492 59.2935 700.01 Q58.5933 697.495 57.2247 695.013 L62.6355 695.013 Q63.7814 697.463 64.3543 700.105 Q64.9272 702.715 64.9272 705.675 Q64.9272 713.728 59.8664 718.47 Q54.8057 723.213 46.212 723.213 Q37.491 723.213 32.4939 718.438 Q27.4968 713.632 27.4968 705.293 Q27.4968 702.588 28.0697 700.01 Q28.6108 697.431 29.7248 695.013 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M42.4881 655.195 L64.0042 655.195 L64.0042 661.051 L42.679 661.051 Q37.6183 661.051 35.1038 663.025 Q32.5894 664.998 32.5894 668.945 Q32.5894 673.687 35.6131 676.425 Q38.6368 679.162 43.8567 679.162 L64.0042 679.162 L64.0042 685.05 L14.479 685.05 L14.479 679.162 L33.8944 679.162 Q30.6797 677.061 29.0883 674.228 Q27.4968 671.364 27.4968 667.64 Q27.4968 661.497 31.3163 658.346 Q35.1038 655.195 42.4881 655.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M46.0847 627.313 Q46.0847 634.411 47.7079 637.148 Q49.3312 639.886 53.2461 639.886 Q56.3653 639.886 58.2114 637.848 Q60.0256 635.78 60.0256 632.247 Q60.0256 627.377 56.5881 624.449 Q53.1188 621.489 47.3897 621.489 L46.0847 621.489 L46.0847 627.313 M43.6657 615.632 L64.0042 615.632 L64.0042 621.489 L58.5933 621.489 Q61.8398 623.494 63.3994 626.486 Q64.9272 629.478 64.9272 633.806 Q64.9272 639.281 61.8716 642.527 Q58.7843 645.742 53.6281 645.742 Q47.6125 645.742 44.5569 641.732 Q41.5014 637.689 41.5014 629.7 L41.5014 621.489 L40.9285 621.489 Q36.8862 621.489 34.6901 624.162 Q32.4621 626.804 32.4621 631.61 Q32.4621 634.666 33.1941 637.562 Q33.9262 640.458 35.3903 643.132 L29.9795 643.132 Q28.7381 639.917 28.1334 636.894 Q27.4968 633.87 27.4968 631.005 Q27.4968 623.271 31.5072 619.452 Q35.5176 615.632 43.6657 615.632 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M33.8307 582.912 Q33.2578 583.899 33.0032 585.077 Q32.7167 586.223 32.7167 587.623 Q32.7167 592.588 35.9632 595.262 Q39.1779 597.904 45.2253 597.904 L64.0042 597.904 L64.0042 603.792 L28.3562 603.792 L28.3562 597.904 L33.8944 597.904 Q30.6479 596.058 29.0883 593.098 Q27.4968 590.138 27.4968 585.904 Q27.4968 585.3 27.5923 584.568 Q27.656 583.835 27.8151 582.944 L33.8307 582.912 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M46.0847 560.569 Q46.0847 567.667 47.7079 570.404 Q49.3312 573.141 53.2461 573.141 Q56.3653 573.141 58.2114 571.104 Q60.0256 569.035 60.0256 565.502 Q60.0256 560.632 56.5881 557.704 Q53.1188 554.744 47.3897 554.744 L46.0847 554.744 L46.0847 560.569 M43.6657 548.888 L64.0042 548.888 L64.0042 554.744 L58.5933 554.744 Q61.8398 556.749 63.3994 559.741 Q64.9272 562.733 64.9272 567.062 Q64.9272 572.536 61.8716 575.783 Q58.7843 578.998 53.6281 578.998 Q47.6125 578.998 44.5569 574.987 Q41.5014 570.945 41.5014 562.956 L41.5014 554.744 L40.9285 554.744 Q36.8862 554.744 34.6901 557.418 Q32.4621 560.06 32.4621 564.866 Q32.4621 567.921 33.1941 570.818 Q33.9262 573.714 35.3903 576.388 L29.9795 576.388 Q28.7381 573.173 28.1334 570.149 Q27.4968 567.125 27.4968 564.261 Q27.4968 556.527 31.5072 552.707 Q35.5176 548.888 43.6657 548.888 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M29.7248 511.171 L35.1993 511.171 Q33.8307 513.654 33.1623 516.168 Q32.4621 518.651 32.4621 521.197 Q32.4621 526.894 36.0905 530.045 Q39.6872 533.196 46.212 533.196 Q52.7369 533.196 56.3653 530.045 Q59.9619 526.894 59.9619 521.197 Q59.9619 518.651 59.2935 516.168 Q58.5933 513.654 57.2247 511.171 L62.6355 511.171 Q63.7814 513.622 64.3543 516.263 Q64.9272 518.873 64.9272 521.833 Q64.9272 529.886 59.8664 534.629 Q54.8057 539.371 46.212 539.371 Q37.491 539.371 32.4939 534.597 Q27.4968 529.791 27.4968 521.452 Q27.4968 518.746 28.0697 516.168 Q28.6108 513.59 29.7248 511.171 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M18.2347 495.193 L28.3562 495.193 L28.3562 483.13 L32.9077 483.13 L32.9077 495.193 L52.2594 495.193 Q56.6199 495.193 57.8613 494.015 Q59.1026 492.806 59.1026 489.146 L59.1026 483.13 L64.0042 483.13 L64.0042 489.146 Q64.0042 495.925 61.4897 498.503 Q58.9434 501.081 52.2594 501.081 L32.9077 501.081 L32.9077 505.378 L28.3562 505.378 L28.3562 501.081 L18.2347 501.081 L18.2347 495.193 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M44.7161 444.936 L47.5806 444.936 L47.5806 471.863 Q53.6281 471.481 56.8109 468.234 Q59.9619 464.956 59.9619 459.131 Q59.9619 455.757 59.1344 452.606 Q58.3069 449.424 56.6518 446.304 L62.1899 446.304 Q63.5267 449.455 64.227 452.766 Q64.9272 456.076 64.9272 459.481 Q64.9272 468.011 59.9619 473.009 Q54.9967 477.974 46.5303 477.974 Q37.7774 477.974 32.6531 473.263 Q27.4968 468.521 27.4968 460.5 Q27.4968 453.307 32.1438 449.137 Q36.7589 444.936 44.7161 444.936 M42.9973 450.792 Q38.1912 450.856 35.3266 453.498 Q32.4621 456.108 32.4621 460.436 Q32.4621 465.338 35.2312 468.298 Q38.0002 471.226 43.0292 471.672 L42.9973 450.792 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M33.8307 414.667 Q33.2578 415.653 33.0032 416.831 Q32.7167 417.977 32.7167 419.377 Q32.7167 424.343 35.9632 427.016 Q39.1779 429.658 45.2253 429.658 L64.0042 429.658 L64.0042 435.546 L28.3562 435.546 L28.3562 429.658 L33.8944 429.658 Q30.6479 427.812 29.0883 424.852 Q27.4968 421.892 27.4968 417.659 Q27.4968 417.054 27.5923 416.322 Q27.656 415.59 27.8151 414.699 L33.8307 414.667 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip310)" d="M29.4065 385.798 L34.9447 385.798 Q33.6716 388.281 33.035 390.955 Q32.3984 393.628 32.3984 396.493 Q32.3984 400.853 33.7352 403.049 Q35.072 405.214 37.7456 405.214 Q39.7826 405.214 40.9603 403.654 Q42.1061 402.095 43.1565 397.384 L43.6021 395.379 Q44.9389 389.14 47.3897 386.53 Q49.8086 383.889 54.1691 383.889 Q59.1344 383.889 62.0308 387.835 Q64.9272 391.75 64.9272 398.625 Q64.9272 401.49 64.3543 404.609 Q63.8132 407.696 62.6992 411.134 L56.6518 411.134 Q58.3387 407.887 59.198 404.736 Q60.0256 401.585 60.0256 398.498 Q60.0256 394.36 58.6251 392.132 Q57.1929 389.904 54.6147 389.904 Q52.2276 389.904 50.9545 391.527 Q49.6813 393.119 48.5037 398.562 L48.0262 400.599 Q46.8804 406.041 44.5251 408.46 Q42.138 410.879 38.0002 410.879 Q32.9713 410.879 30.2341 407.314 Q27.4968 403.75 27.4968 397.193 Q27.4968 393.946 27.9743 391.082 Q28.4517 388.217 29.4065 385.798 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip312)" d="M295.201 47.2441 L295.201 1423.18 L562.484 1423.18 L562.484 47.2441 L295.201 47.2441 L295.201 47.2441  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="295.201,47.2441 295.201,1423.18 562.484,1423.18 562.484,47.2441 295.201,47.2441 "/>
+<path clip-path="url(#clip312)" d="M629.304 563.22 L629.304 1423.18 L896.587 1423.18 L896.587 563.22 L629.304 563.22 L629.304 563.22  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="629.304,563.22 629.304,1423.18 896.587,1423.18 896.587,563.22 629.304,563.22 "/>
+<path clip-path="url(#clip312)" d="M963.408 1079.2 L963.408 1423.18 L1230.69 1423.18 L1230.69 1079.2 L963.408 1079.2 L963.408 1079.2  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="963.408,1079.2 963.408,1423.18 1230.69,1423.18 1230.69,1079.2 963.408,1079.2 "/>
+<path clip-path="url(#clip312)" d="M1297.51 1251.19 L1297.51 1423.18 L1564.79 1423.18 L1564.79 1251.19 L1297.51 1251.19 L1297.51 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1297.51,1251.19 1297.51,1423.18 1564.79,1423.18 1564.79,1251.19 1297.51,1251.19 "/>
+<path clip-path="url(#clip312)" d="M1631.61 1251.19 L1631.61 1423.18 L1898.9 1423.18 L1898.9 1251.19 L1631.61 1251.19 L1631.61 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1631.61,1251.19 1631.61,1423.18 1898.9,1423.18 1898.9,1251.19 1631.61,1251.19 "/>
+<path clip-path="url(#clip312)" d="M1965.72 1251.19 L1965.72 1423.18 L2233 1423.18 L2233 1251.19 L1965.72 1251.19 L1965.72 1251.19  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip312)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1965.72,1251.19 1965.72,1423.18 2233,1423.18 2233,1251.19 1965.72,1251.19 "/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="428.842" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="762.946" cy="563.22" r="2"/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1097.05" cy="1079.2" r="2"/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1431.15" cy="1251.19" r="2"/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1765.26" cy="1251.19" r="2"/>
+<circle clip-path="url(#clip312)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2099.36" cy="1251.19" r="2"/>
 </svg>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/054004dc.svg b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/76abfa9b.svg
similarity index 82%
rename from previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/054004dc.svg
rename to previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/76abfa9b.svg
index a0c52e8838a..5e855da3a0f 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/054004dc.svg
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/76abfa9b.svg
@@ -1,73 +1,73 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip370">
+  <clipPath id="clip190">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip370)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip190)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip371">
+  <clipPath id="clip191">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip370)" d="M219.033 1423.18 L2352.76 1423.18 L2352.76 47.2441 L219.033 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip190)" d="M219.033 1423.18 L2352.76 1423.18 L2352.76 47.2441 L219.033 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip372">
+  <clipPath id="clip192">
     <rect x="219" y="47" width="2135" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="413.618,1423.18 413.618,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.011,1423.18 682.011,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="950.403,1423.18 950.403,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1218.8,1423.18 1218.8,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1487.19,1423.18 1487.19,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1755.58,1423.18 1755.58,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2023.97,1423.18 2023.97,47.2441 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2292.37,1423.18 2292.37,47.2441 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="413.618,1423.18 413.618,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.011,1423.18 682.011,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="950.403,1423.18 950.403,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1218.8,1423.18 1218.8,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1487.19,1423.18 1487.19,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1755.58,1423.18 1755.58,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2023.97,1423.18 2023.97,1404.28 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2292.37,1423.18 2292.37,1404.28 "/>
-<path clip-path="url(#clip370)" d="M402.287 1455.09 L390.481 1473.54 L402.287 1473.54 L402.287 1455.09 M401.06 1451.02 L406.94 1451.02 L406.94 1473.54 L411.87 1473.54 L411.87 1477.43 L406.94 1477.43 L406.94 1485.58 L402.287 1485.58 L402.287 1477.43 L386.685 1477.43 L386.685 1472.92 L401.06 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M419.648 1451.02 L438.004 1451.02 L438.004 1454.96 L423.93 1454.96 L423.93 1463.43 Q424.949 1463.08 425.967 1462.92 Q426.986 1462.73 428.004 1462.73 Q433.791 1462.73 437.171 1465.9 Q440.55 1469.08 440.55 1474.49 Q440.55 1480.07 437.078 1483.17 Q433.606 1486.25 427.287 1486.25 Q425.111 1486.25 422.842 1485.88 Q420.597 1485.51 418.189 1484.77 L418.189 1480.07 Q420.273 1481.2 422.495 1481.76 Q424.717 1482.32 427.194 1482.32 Q431.199 1482.32 433.537 1480.21 Q435.875 1478.1 435.875 1474.49 Q435.875 1470.88 433.537 1468.77 Q431.199 1466.67 427.194 1466.67 Q425.319 1466.67 423.444 1467.08 Q421.592 1467.5 419.648 1468.38 L419.648 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M670.275 1455.09 L658.469 1473.54 L670.275 1473.54 L670.275 1455.09 M669.048 1451.02 L674.927 1451.02 L674.927 1473.54 L679.858 1473.54 L679.858 1477.43 L674.927 1477.43 L674.927 1485.58 L670.275 1485.58 L670.275 1477.43 L654.673 1477.43 L654.673 1472.92 L669.048 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M687.728 1484.86 L687.728 1480.6 Q689.487 1481.44 691.293 1481.88 Q693.098 1482.32 694.835 1482.32 Q699.464 1482.32 701.895 1479.21 Q704.348 1476.09 704.696 1469.75 Q703.353 1471.74 701.293 1472.8 Q699.233 1473.87 696.733 1473.87 Q691.548 1473.87 688.515 1470.74 Q685.506 1467.59 685.506 1462.15 Q685.506 1456.83 688.654 1453.61 Q691.802 1450.39 697.034 1450.39 Q703.029 1450.39 706.177 1455 Q709.348 1459.58 709.348 1468.33 Q709.348 1476.51 705.46 1481.39 Q701.594 1486.25 695.043 1486.25 Q693.284 1486.25 691.478 1485.9 Q689.673 1485.56 687.728 1484.86 M697.034 1470.21 Q700.182 1470.21 702.01 1468.06 Q703.862 1465.9 703.862 1462.15 Q703.862 1458.43 702.01 1456.27 Q700.182 1454.1 697.034 1454.1 Q693.885 1454.1 692.034 1456.27 Q690.205 1458.43 690.205 1462.15 Q690.205 1465.9 692.034 1468.06 Q693.885 1470.21 697.034 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M935.276 1469.17 Q931.943 1469.17 930.022 1470.95 Q928.123 1472.73 928.123 1475.86 Q928.123 1478.98 930.022 1480.77 Q931.943 1482.55 935.276 1482.55 Q938.609 1482.55 940.531 1480.77 Q942.452 1478.96 942.452 1475.86 Q942.452 1472.73 940.531 1470.95 Q938.633 1469.17 935.276 1469.17 M930.6 1467.18 Q927.591 1466.44 925.901 1464.38 Q924.235 1462.32 924.235 1459.35 Q924.235 1455.21 927.174 1452.8 Q930.137 1450.39 935.276 1450.39 Q940.438 1450.39 943.378 1452.8 Q946.318 1455.21 946.318 1459.35 Q946.318 1462.32 944.628 1464.38 Q942.961 1466.44 939.975 1467.18 Q943.355 1467.96 945.23 1470.26 Q947.128 1472.55 947.128 1475.86 Q947.128 1480.88 944.049 1483.57 Q940.994 1486.25 935.276 1486.25 Q929.559 1486.25 926.48 1483.57 Q923.424 1480.88 923.424 1475.86 Q923.424 1472.55 925.323 1470.26 Q927.221 1467.96 930.6 1467.18 M928.887 1459.79 Q928.887 1462.48 930.554 1463.98 Q932.244 1465.49 935.276 1465.49 Q938.285 1465.49 939.975 1463.98 Q941.688 1462.48 941.688 1459.79 Q941.688 1457.11 939.975 1455.6 Q938.285 1454.1 935.276 1454.1 Q932.244 1454.1 930.554 1455.6 Q928.887 1457.11 928.887 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M965.438 1454.1 Q961.827 1454.1 959.998 1457.66 Q958.193 1461.2 958.193 1468.33 Q958.193 1475.44 959.998 1479.01 Q961.827 1482.55 965.438 1482.55 Q969.072 1482.55 970.878 1479.01 Q972.707 1475.44 972.707 1468.33 Q972.707 1461.2 970.878 1457.66 Q969.072 1454.1 965.438 1454.1 M965.438 1450.39 Q971.248 1450.39 974.304 1455 Q977.382 1459.58 977.382 1468.33 Q977.382 1477.06 974.304 1481.67 Q971.248 1486.25 965.438 1486.25 Q959.628 1486.25 956.549 1481.67 Q953.494 1477.06 953.494 1468.33 Q953.494 1459.58 956.549 1455 Q959.628 1450.39 965.438 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1208.44 1466.95 Q1211.79 1467.66 1213.67 1469.93 Q1215.57 1472.2 1215.57 1475.53 Q1215.57 1480.65 1212.05 1483.45 Q1208.53 1486.25 1202.05 1486.25 Q1199.87 1486.25 1197.56 1485.81 Q1195.27 1485.39 1192.81 1484.54 L1192.81 1480.02 Q1194.76 1481.16 1197.07 1481.74 Q1199.39 1482.32 1201.91 1482.32 Q1206.31 1482.32 1208.6 1480.58 Q1210.91 1478.84 1210.91 1475.53 Q1210.91 1472.48 1208.76 1470.77 Q1206.63 1469.03 1202.81 1469.03 L1198.78 1469.03 L1198.78 1465.19 L1203 1465.19 Q1206.45 1465.19 1208.28 1463.82 Q1210.1 1462.43 1210.1 1459.84 Q1210.1 1457.18 1208.21 1455.77 Q1206.33 1454.33 1202.81 1454.33 Q1200.89 1454.33 1198.69 1454.75 Q1196.49 1455.16 1193.85 1456.04 L1193.85 1451.88 Q1196.52 1451.14 1198.83 1450.77 Q1201.17 1450.39 1203.23 1450.39 Q1208.55 1450.39 1211.66 1452.83 Q1214.76 1455.23 1214.76 1459.35 Q1214.76 1462.22 1213.11 1464.21 Q1211.47 1466.18 1208.44 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1228.46 1481.64 L1244.78 1481.64 L1244.78 1485.58 L1222.84 1485.58 L1222.84 1481.64 Q1225.5 1478.89 1230.08 1474.26 Q1234.69 1469.61 1235.87 1468.27 Q1238.11 1465.74 1238.99 1464.01 Q1239.9 1462.25 1239.9 1460.56 Q1239.9 1457.8 1237.95 1456.07 Q1236.03 1454.33 1232.93 1454.33 Q1230.73 1454.33 1228.28 1455.09 Q1225.84 1455.86 1223.07 1457.41 L1223.07 1452.69 Q1225.89 1451.55 1228.34 1450.97 Q1230.8 1450.39 1232.84 1450.39 Q1238.21 1450.39 1241.4 1453.08 Q1244.59 1455.77 1244.59 1460.26 Q1244.59 1462.39 1243.78 1464.31 Q1243 1466.2 1240.89 1468.8 Q1240.31 1469.47 1237.21 1472.69 Q1234.11 1475.88 1228.46 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1471.98 1469.17 Q1468.65 1469.17 1466.73 1470.95 Q1464.83 1472.73 1464.83 1475.86 Q1464.83 1478.98 1466.73 1480.77 Q1468.65 1482.55 1471.98 1482.55 Q1475.31 1482.55 1477.24 1480.77 Q1479.16 1478.96 1479.16 1475.86 Q1479.16 1472.73 1477.24 1470.95 Q1475.34 1469.17 1471.98 1469.17 M1467.3 1467.18 Q1464.3 1466.44 1462.61 1464.38 Q1460.94 1462.32 1460.94 1459.35 Q1460.94 1455.21 1463.88 1452.8 Q1466.84 1450.39 1471.98 1450.39 Q1477.14 1450.39 1480.08 1452.8 Q1483.02 1455.21 1483.02 1459.35 Q1483.02 1462.32 1481.33 1464.38 Q1479.67 1466.44 1476.68 1467.18 Q1480.06 1467.96 1481.93 1470.26 Q1483.83 1472.55 1483.83 1475.86 Q1483.83 1480.88 1480.75 1483.57 Q1477.7 1486.25 1471.98 1486.25 Q1466.26 1486.25 1463.18 1483.57 Q1460.13 1480.88 1460.13 1475.86 Q1460.13 1472.55 1462.03 1470.26 Q1463.93 1467.96 1467.3 1467.18 M1465.59 1459.79 Q1465.59 1462.48 1467.26 1463.98 Q1468.95 1465.49 1471.98 1465.49 Q1474.99 1465.49 1476.68 1463.98 Q1478.39 1462.48 1478.39 1459.79 Q1478.39 1457.11 1476.68 1455.6 Q1474.99 1454.1 1471.98 1454.1 Q1468.95 1454.1 1467.26 1455.6 Q1465.59 1457.11 1465.59 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1502.72 1466.44 Q1499.57 1466.44 1497.72 1468.59 Q1495.89 1470.74 1495.89 1474.49 Q1495.89 1478.22 1497.72 1480.39 Q1499.57 1482.55 1502.72 1482.55 Q1505.87 1482.55 1507.7 1480.39 Q1509.55 1478.22 1509.55 1474.49 Q1509.55 1470.74 1507.7 1468.59 Q1505.87 1466.44 1502.72 1466.44 M1512 1451.78 L1512 1456.04 Q1510.24 1455.21 1508.44 1454.77 Q1506.66 1454.33 1504.9 1454.33 Q1500.27 1454.33 1497.81 1457.45 Q1495.38 1460.58 1495.04 1466.9 Q1496.4 1464.89 1498.46 1463.82 Q1500.52 1462.73 1503 1462.73 Q1508.21 1462.73 1511.22 1465.9 Q1514.25 1469.05 1514.25 1474.49 Q1514.25 1479.82 1511.1 1483.03 Q1507.95 1486.25 1502.72 1486.25 Q1496.73 1486.25 1493.55 1481.67 Q1490.38 1477.06 1490.38 1468.33 Q1490.38 1460.14 1494.27 1455.28 Q1498.16 1450.39 1504.71 1450.39 Q1506.47 1450.39 1508.25 1450.74 Q1510.06 1451.09 1512 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1724.16 1451.02 L1730.45 1451.02 L1745.78 1479.93 L1745.78 1451.02 L1750.32 1451.02 L1750.32 1485.58 L1744.02 1485.58 L1728.7 1456.67 L1728.7 1485.58 L1724.16 1485.58 L1724.16 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1771.17 1455.63 L1764.83 1472.83 L1777.54 1472.83 L1771.17 1455.63 M1768.53 1451.02 L1773.83 1451.02 L1787.01 1485.58 L1782.14 1485.58 L1779 1476.71 L1763.42 1476.71 L1760.27 1485.58 L1755.34 1485.58 L1768.53 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1997.38 1451.02 L2019.6 1451.02 L2019.6 1453.01 L2007.05 1485.58 L2002.17 1485.58 L2013.97 1454.96 L1997.38 1454.96 L1997.38 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2038.72 1469.17 Q2035.39 1469.17 2033.47 1470.95 Q2031.57 1472.73 2031.57 1475.86 Q2031.57 1478.98 2033.47 1480.77 Q2035.39 1482.55 2038.72 1482.55 Q2042.05 1482.55 2043.97 1480.77 Q2045.9 1478.96 2045.9 1475.86 Q2045.9 1472.73 2043.97 1470.95 Q2042.08 1469.17 2038.72 1469.17 M2034.04 1467.18 Q2031.03 1466.44 2029.35 1464.38 Q2027.68 1462.32 2027.68 1459.35 Q2027.68 1455.21 2030.62 1452.8 Q2033.58 1450.39 2038.72 1450.39 Q2043.88 1450.39 2046.82 1452.8 Q2049.76 1455.21 2049.76 1459.35 Q2049.76 1462.32 2048.07 1464.38 Q2046.41 1466.44 2043.42 1467.18 Q2046.8 1467.96 2048.67 1470.26 Q2050.57 1472.55 2050.57 1475.86 Q2050.57 1480.88 2047.49 1483.57 Q2044.44 1486.25 2038.72 1486.25 Q2033 1486.25 2029.92 1483.57 Q2026.87 1480.88 2026.87 1475.86 Q2026.87 1472.55 2028.77 1470.26 Q2030.66 1467.96 2034.04 1467.18 M2032.33 1459.79 Q2032.33 1462.48 2034 1463.98 Q2035.69 1465.49 2038.72 1465.49 Q2041.73 1465.49 2043.42 1463.98 Q2045.13 1462.48 2045.13 1459.79 Q2045.13 1457.11 2043.42 1455.6 Q2041.73 1454.1 2038.72 1454.1 Q2035.69 1454.1 2034 1455.6 Q2032.33 1457.11 2032.33 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2267.07 1451.02 L2285.42 1451.02 L2285.42 1454.96 L2271.35 1454.96 L2271.35 1463.43 Q2272.37 1463.08 2273.39 1462.92 Q2274.4 1462.73 2275.42 1462.73 Q2281.21 1462.73 2284.59 1465.9 Q2287.97 1469.08 2287.97 1474.49 Q2287.97 1480.07 2284.5 1483.17 Q2281.02 1486.25 2274.71 1486.25 Q2272.53 1486.25 2270.26 1485.88 Q2268.02 1485.51 2265.61 1484.77 L2265.61 1480.07 Q2267.69 1481.2 2269.91 1481.76 Q2272.14 1482.32 2274.61 1482.32 Q2278.62 1482.32 2280.96 1480.21 Q2283.29 1478.1 2283.29 1474.49 Q2283.29 1470.88 2280.96 1468.77 Q2278.62 1466.67 2274.61 1466.67 Q2272.74 1466.67 2270.86 1467.08 Q2269.01 1467.5 2267.07 1468.38 L2267.07 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2307.18 1454.1 Q2303.57 1454.1 2301.74 1457.66 Q2299.94 1461.2 2299.94 1468.33 Q2299.94 1475.44 2301.74 1479.01 Q2303.57 1482.55 2307.18 1482.55 Q2310.82 1482.55 2312.62 1479.01 Q2314.45 1475.44 2314.45 1468.33 Q2314.45 1461.2 2312.62 1457.66 Q2310.82 1454.1 2307.18 1454.1 M2307.18 1450.39 Q2312.99 1450.39 2316.05 1455 Q2319.13 1459.58 2319.13 1468.33 Q2319.13 1477.06 2316.05 1481.67 Q2312.99 1486.25 2307.18 1486.25 Q2301.37 1486.25 2298.29 1481.67 Q2295.24 1477.06 2295.24 1468.33 Q2295.24 1459.58 2298.29 1455 Q2301.37 1450.39 2307.18 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1174.21 1520.52 L1180.7 1520.52 L1190.7 1560.69 L1200.66 1520.52 L1207.88 1520.52 L1217.88 1560.69 L1227.84 1520.52 L1234.36 1520.52 L1222.43 1568.04 L1214.34 1568.04 L1204.32 1526.79 L1194.2 1568.04 L1186.11 1568.04 L1174.21 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1269.31 1548.76 L1269.31 1551.62 L1242.38 1551.62 Q1242.77 1557.67 1246.01 1560.85 Q1249.29 1564 1255.12 1564 Q1258.49 1564 1261.64 1563.17 Q1264.82 1562.35 1267.94 1560.69 L1267.94 1566.23 Q1264.79 1567.57 1261.48 1568.27 Q1258.17 1568.97 1254.77 1568.97 Q1246.24 1568.97 1241.24 1564 Q1236.27 1559.04 1236.27 1550.57 Q1236.27 1541.82 1240.98 1536.69 Q1245.73 1531.54 1253.75 1531.54 Q1260.94 1531.54 1265.11 1536.18 Q1269.31 1540.8 1269.31 1548.76 M1263.46 1547.04 Q1263.39 1542.23 1260.75 1539.37 Q1258.14 1536.5 1253.81 1536.5 Q1248.91 1536.5 1245.95 1539.27 Q1243.02 1542.04 1242.58 1547.07 L1263.46 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1278.92 1532.4 L1284.78 1532.4 L1284.78 1568.04 L1278.92 1568.04 L1278.92 1532.4 M1278.92 1518.52 L1284.78 1518.52 L1284.78 1525.93 L1278.92 1525.93 L1278.92 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1320.49 1549.81 Q1320.49 1543.44 1317.85 1539.94 Q1315.24 1536.44 1310.5 1536.44 Q1305.79 1536.44 1303.15 1539.94 Q1300.54 1543.44 1300.54 1549.81 Q1300.54 1556.14 1303.15 1559.64 Q1305.79 1563.14 1310.5 1563.14 Q1315.24 1563.14 1317.85 1559.64 Q1320.49 1556.14 1320.49 1549.81 M1326.35 1563.62 Q1326.35 1572.72 1322.31 1577.15 Q1318.26 1581.6 1309.92 1581.6 Q1306.84 1581.6 1304.1 1581.13 Q1301.36 1580.68 1298.78 1579.72 L1298.78 1574.03 Q1301.36 1575.43 1303.88 1576.1 Q1306.39 1576.76 1309 1576.76 Q1314.76 1576.76 1317.63 1573.74 Q1320.49 1570.75 1320.49 1564.67 L1320.49 1561.77 Q1318.68 1564.92 1315.85 1566.48 Q1313.01 1568.04 1309.07 1568.04 Q1302.51 1568.04 1298.5 1563.05 Q1294.49 1558.05 1294.49 1549.81 Q1294.49 1541.53 1298.5 1536.53 Q1302.51 1531.54 1309.07 1531.54 Q1313.01 1531.54 1315.85 1533.1 Q1318.68 1534.66 1320.49 1537.81 L1320.49 1532.4 L1326.35 1532.4 L1326.35 1563.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1368.04 1546.53 L1368.04 1568.04 L1362.19 1568.04 L1362.19 1546.72 Q1362.19 1541.66 1360.21 1539.14 Q1358.24 1536.63 1354.29 1536.63 Q1349.55 1536.63 1346.81 1539.65 Q1344.08 1542.68 1344.08 1547.9 L1344.08 1568.04 L1338.19 1568.04 L1338.19 1518.52 L1344.08 1518.52 L1344.08 1537.93 Q1346.18 1534.72 1349.01 1533.13 Q1351.87 1531.54 1355.6 1531.54 Q1361.74 1531.54 1364.89 1535.36 Q1368.04 1539.14 1368.04 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1385.52 1522.27 L1385.52 1532.4 L1397.58 1532.4 L1397.58 1536.95 L1385.52 1536.95 L1385.52 1556.3 Q1385.52 1560.66 1386.7 1561.9 Q1387.9 1563.14 1391.57 1563.14 L1397.58 1563.14 L1397.58 1568.04 L1391.57 1568.04 Q1384.79 1568.04 1382.21 1565.53 Q1379.63 1562.98 1379.63 1556.3 L1379.63 1536.95 L1375.33 1536.95 L1375.33 1532.4 L1379.63 1532.4 L1379.63 1522.27 L1385.52 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,1248.02 2352.76,1248.02 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,1047.71 2352.76,1047.71 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,847.389 2352.76,847.389 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,647.073 2352.76,647.073 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,446.756 2352.76,446.756 "/>
-<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,246.439 2352.76,246.439 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1423.18 219.033,47.2441 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1248.02 237.931,1248.02 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1047.71 237.931,1047.71 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,847.389 237.931,847.389 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,647.073 237.931,647.073 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,446.756 237.931,446.756 "/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,246.439 237.931,246.439 "/>
-<path clip-path="url(#clip370)" d="M116.668 1261.37 L124.306 1261.37 L124.306 1235 L115.996 1236.67 L115.996 1232.41 L124.26 1230.74 L128.936 1230.74 L128.936 1261.37 L136.575 1261.37 L136.575 1265.3 L116.668 1265.3 L116.668 1261.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M146.019 1259.42 L150.903 1259.42 L150.903 1265.3 L146.019 1265.3 L146.019 1259.42 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M171.089 1233.82 Q167.477 1233.82 165.649 1237.39 Q163.843 1240.93 163.843 1248.06 Q163.843 1255.16 165.649 1258.73 Q167.477 1262.27 171.089 1262.27 Q174.723 1262.27 176.528 1258.73 Q178.357 1255.16 178.357 1248.06 Q178.357 1240.93 176.528 1237.39 Q174.723 1233.82 171.089 1233.82 M171.089 1230.12 Q176.899 1230.12 179.954 1234.72 Q183.033 1239.31 183.033 1248.06 Q183.033 1256.78 179.954 1261.39 Q176.899 1265.97 171.089 1265.97 Q165.278 1265.97 162.2 1261.39 Q159.144 1256.78 159.144 1248.06 Q159.144 1239.31 162.2 1234.72 Q165.278 1230.12 171.089 1230.12 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M117.663 1061.05 L125.302 1061.05 L125.302 1034.69 L116.992 1036.35 L116.992 1032.09 L125.255 1030.43 L129.931 1030.43 L129.931 1061.05 L137.57 1061.05 L137.57 1064.99 L117.663 1064.99 L117.663 1061.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M147.015 1059.11 L151.899 1059.11 L151.899 1064.99 L147.015 1064.99 L147.015 1059.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M162.13 1030.43 L180.487 1030.43 L180.487 1034.36 L166.413 1034.36 L166.413 1042.83 Q167.431 1042.49 168.45 1042.32 Q169.468 1042.14 170.487 1042.14 Q176.274 1042.14 179.653 1045.31 Q183.033 1048.48 183.033 1053.9 Q183.033 1059.48 179.561 1062.58 Q176.089 1065.66 169.769 1065.66 Q167.593 1065.66 165.325 1065.29 Q163.079 1064.92 160.672 1064.18 L160.672 1059.48 Q162.755 1060.61 164.977 1061.17 Q167.2 1061.72 169.677 1061.72 Q173.681 1061.72 176.019 1059.62 Q178.357 1057.51 178.357 1053.9 Q178.357 1050.29 176.019 1048.18 Q173.681 1046.07 169.677 1046.07 Q167.802 1046.07 165.927 1046.49 Q164.075 1046.91 162.13 1047.79 L162.13 1030.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M119.885 860.734 L136.204 860.734 L136.204 864.669 L114.26 864.669 L114.26 860.734 Q116.922 857.98 121.505 853.35 Q126.112 848.697 127.292 847.355 Q129.538 844.831 130.417 843.095 Q131.32 841.336 131.32 839.646 Q131.32 836.892 129.376 835.156 Q127.455 833.42 124.353 833.42 Q122.154 833.42 119.7 834.183 Q117.269 834.947 114.492 836.498 L114.492 831.776 Q117.316 830.642 119.769 830.063 Q122.223 829.484 124.26 829.484 Q129.63 829.484 132.825 832.17 Q136.019 834.855 136.019 839.345 Q136.019 841.475 135.209 843.396 Q134.422 845.294 132.316 847.887 Q131.737 848.558 128.635 851.776 Q125.533 854.97 119.885 860.734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M146.019 858.79 L150.903 858.79 L150.903 864.669 L146.019 864.669 L146.019 858.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M171.089 833.188 Q167.477 833.188 165.649 836.753 Q163.843 840.294 163.843 847.424 Q163.843 854.531 165.649 858.095 Q167.477 861.637 171.089 861.637 Q174.723 861.637 176.528 858.095 Q178.357 854.531 178.357 847.424 Q178.357 840.294 176.528 836.753 Q174.723 833.188 171.089 833.188 M171.089 829.484 Q176.899 829.484 179.954 834.091 Q183.033 838.674 183.033 847.424 Q183.033 856.151 179.954 860.757 Q176.899 865.341 171.089 865.341 Q165.278 865.341 162.2 860.757 Q159.144 856.151 159.144 847.424 Q159.144 838.674 162.2 834.091 Q165.278 829.484 171.089 829.484 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M120.88 660.417 L137.2 660.417 L137.2 664.353 L115.256 664.353 L115.256 660.417 Q117.918 657.663 122.501 653.033 Q127.107 648.38 128.288 647.038 Q130.533 644.515 131.413 642.779 Q132.316 641.019 132.316 639.33 Q132.316 636.575 130.371 634.839 Q128.45 633.103 125.348 633.103 Q123.149 633.103 120.695 633.867 Q118.265 634.631 115.487 636.181 L115.487 631.459 Q118.311 630.325 120.765 629.746 Q123.218 629.168 125.255 629.168 Q130.626 629.168 133.82 631.853 Q137.015 634.538 137.015 639.029 Q137.015 641.158 136.204 643.08 Q135.417 644.978 133.311 647.57 Q132.732 648.242 129.63 651.459 Q126.529 654.654 120.88 660.417 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M147.015 658.473 L151.899 658.473 L151.899 664.353 L147.015 664.353 L147.015 658.473 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M162.13 629.793 L180.487 629.793 L180.487 633.728 L166.413 633.728 L166.413 642.2 Q167.431 641.853 168.45 641.691 Q169.468 641.505 170.487 641.505 Q176.274 641.505 179.653 644.677 Q183.033 647.848 183.033 653.265 Q183.033 658.843 179.561 661.945 Q176.089 665.024 169.769 665.024 Q167.593 665.024 165.325 664.654 Q163.079 664.283 160.672 663.542 L160.672 658.843 Q162.755 659.978 164.977 660.533 Q167.2 661.089 169.677 661.089 Q173.681 661.089 176.019 658.982 Q178.357 656.876 178.357 653.265 Q178.357 649.654 176.019 647.547 Q173.681 645.441 169.677 645.441 Q167.802 645.441 165.927 645.857 Q164.075 646.274 162.13 647.154 L162.13 629.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M130.024 445.402 Q133.38 446.119 135.255 448.388 Q137.154 450.656 137.154 453.99 Q137.154 459.105 133.635 461.906 Q130.117 464.707 123.635 464.707 Q121.459 464.707 119.144 464.267 Q116.853 463.851 114.399 462.994 L114.399 458.48 Q116.343 459.615 118.658 460.193 Q120.973 460.772 123.496 460.772 Q127.894 460.772 130.186 459.036 Q132.501 457.3 132.501 453.99 Q132.501 450.934 130.348 449.221 Q128.218 447.485 124.399 447.485 L120.371 447.485 L120.371 443.642 L124.584 443.642 Q128.033 443.642 129.862 442.277 Q131.691 440.888 131.691 438.295 Q131.691 435.633 129.792 434.221 Q127.917 432.786 124.399 432.786 Q122.478 432.786 120.279 433.203 Q118.08 433.619 115.441 434.499 L115.441 430.332 Q118.103 429.592 120.418 429.221 Q122.755 428.851 124.816 428.851 Q130.14 428.851 133.242 431.281 Q136.343 433.689 136.343 437.809 Q136.343 440.679 134.7 442.67 Q133.056 444.638 130.024 445.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M146.019 458.156 L150.903 458.156 L150.903 464.036 L146.019 464.036 L146.019 458.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M171.089 432.555 Q167.477 432.555 165.649 436.119 Q163.843 439.661 163.843 446.791 Q163.843 453.897 165.649 457.462 Q167.477 461.003 171.089 461.003 Q174.723 461.003 176.528 457.462 Q178.357 453.897 178.357 446.791 Q178.357 439.661 176.528 436.119 Q174.723 432.555 171.089 432.555 M171.089 428.851 Q176.899 428.851 179.954 433.457 Q183.033 438.041 183.033 446.791 Q183.033 455.517 179.954 460.124 Q176.899 464.707 171.089 464.707 Q165.278 464.707 162.2 460.124 Q159.144 455.517 159.144 446.791 Q159.144 438.041 162.2 433.457 Q165.278 428.851 171.089 428.851 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M131.019 245.085 Q134.376 245.803 136.251 248.071 Q138.149 250.34 138.149 253.673 Q138.149 258.789 134.63 261.589 Q131.112 264.39 124.63 264.39 Q122.455 264.39 120.14 263.951 Q117.848 263.534 115.394 262.677 L115.394 258.164 Q117.339 259.298 119.654 259.877 Q121.968 260.455 124.492 260.455 Q128.89 260.455 131.181 258.719 Q133.496 256.983 133.496 253.673 Q133.496 250.617 131.343 248.904 Q129.214 247.168 125.394 247.168 L121.367 247.168 L121.367 243.326 L125.58 243.326 Q129.029 243.326 130.857 241.96 Q132.686 240.571 132.686 237.978 Q132.686 235.316 130.788 233.904 Q128.913 232.469 125.394 232.469 Q123.473 232.469 121.274 232.886 Q119.075 233.303 116.436 234.182 L116.436 230.016 Q119.098 229.275 121.413 228.904 Q123.751 228.534 125.811 228.534 Q131.135 228.534 134.237 230.965 Q137.339 233.372 137.339 237.492 Q137.339 240.363 135.695 242.353 Q134.052 244.321 131.019 245.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M147.015 257.839 L151.899 257.839 L151.899 263.719 L147.015 263.719 L147.015 257.839 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M162.13 229.159 L180.487 229.159 L180.487 233.094 L166.413 233.094 L166.413 241.566 Q167.431 241.219 168.45 241.057 Q169.468 240.872 170.487 240.872 Q176.274 240.872 179.653 244.043 Q183.033 247.215 183.033 252.631 Q183.033 258.21 179.561 261.312 Q176.089 264.39 169.769 264.39 Q167.593 264.39 165.325 264.02 Q163.079 263.65 160.672 262.909 L160.672 258.21 Q162.755 259.344 164.977 259.9 Q167.2 260.455 169.677 260.455 Q173.681 260.455 176.019 258.349 Q178.357 256.242 178.357 252.631 Q178.357 249.02 176.019 246.914 Q173.681 244.807 169.677 244.807 Q167.802 244.807 165.927 245.224 Q164.075 245.64 162.13 246.52 L162.13 229.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M16.4842 838.973 L16.4842 832.544 L35.9632 832.544 L35.9632 809.181 L16.4842 809.181 L16.4842 802.752 L64.0042 802.752 L64.0042 809.181 L41.3741 809.181 L41.3741 832.544 L64.0042 832.544 L64.0042 838.973 L16.4842 838.973 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M44.7161 759.72 L47.5806 759.72 L47.5806 786.647 Q53.6281 786.265 56.8109 783.018 Q59.9619 779.74 59.9619 773.915 Q59.9619 770.542 59.1344 767.391 Q58.3069 764.208 56.6518 761.089 L62.1899 761.089 Q63.5267 764.24 64.227 767.55 Q64.9272 770.86 64.9272 774.266 Q64.9272 782.796 59.9619 787.793 Q54.9967 792.758 46.5303 792.758 Q37.7774 792.758 32.6531 788.047 Q27.4968 783.305 27.4968 775.284 Q27.4968 768.091 32.1438 763.921 Q36.7589 759.72 44.7161 759.72 M42.9973 765.576 Q38.1912 765.64 35.3266 768.282 Q32.4621 770.892 32.4621 775.22 Q32.4621 780.122 35.2312 783.082 Q38.0002 786.01 43.0292 786.456 L42.9973 765.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M28.3562 750.108 L28.3562 744.251 L64.0042 744.251 L64.0042 750.108 L28.3562 750.108 M14.479 750.108 L14.479 744.251 L21.895 744.251 L21.895 750.108 L14.479 750.108 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M45.7664 708.54 Q39.4007 708.54 35.8996 711.181 Q32.3984 713.791 32.3984 718.534 Q32.3984 723.244 35.8996 725.886 Q39.4007 728.496 45.7664 728.496 Q52.1003 728.496 55.6014 725.886 Q59.1026 723.244 59.1026 718.534 Q59.1026 713.791 55.6014 711.181 Q52.1003 708.54 45.7664 708.54 M59.58 702.683 Q68.683 702.683 73.1071 706.725 Q77.5631 710.768 77.5631 719.107 Q77.5631 722.194 77.0857 724.931 Q76.6401 727.669 75.6852 730.247 L69.9879 730.247 Q71.3884 727.669 72.0568 725.154 Q72.7252 722.64 72.7252 720.03 Q72.7252 714.269 69.7015 711.404 Q66.7096 708.54 60.6303 708.54 L57.7339 708.54 Q60.885 710.354 62.4446 713.187 Q64.0042 716.019 64.0042 719.966 Q64.0042 726.523 59.0071 730.533 Q54.01 734.544 45.7664 734.544 Q37.491 734.544 32.4939 730.533 Q27.4968 726.523 27.4968 719.966 Q27.4968 716.019 29.0564 713.187 Q30.616 710.354 33.7671 708.54 L28.3562 708.54 L28.3562 702.683 L59.58 702.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M42.4881 660.988 L64.0042 660.988 L64.0042 666.844 L42.679 666.844 Q37.6183 666.844 35.1038 668.818 Q32.5894 670.791 32.5894 674.738 Q32.5894 679.48 35.6131 682.217 Q38.6368 684.955 43.8567 684.955 L64.0042 684.955 L64.0042 690.843 L14.479 690.843 L14.479 684.955 L33.8944 684.955 Q30.6797 682.854 29.0883 680.021 Q27.4968 677.157 27.4968 673.433 Q27.4968 667.29 31.3163 664.139 Q35.1038 660.988 42.4881 660.988 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M18.2347 643.514 L28.3562 643.514 L28.3562 631.451 L32.9077 631.451 L32.9077 643.514 L52.2594 643.514 Q56.6199 643.514 57.8613 642.336 Q59.1026 641.127 59.1026 637.467 L59.1026 631.451 L64.0042 631.451 L64.0042 637.467 Q64.0042 644.246 61.4897 646.824 Q58.9434 649.402 52.2594 649.402 L32.9077 649.402 L32.9077 653.699 L28.3562 653.699 L28.3562 649.402 L18.2347 649.402 L18.2347 643.514 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip372)" cx="279.421" cy="895.465" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="413.618" cy="987.611" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="547.814" cy="959.567" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="682.011" cy="1047.71" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="816.207" cy="875.434" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="547.814" cy="919.503" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="950.403" cy="927.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1084.6" cy="955.56" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1218.8" cy="1264.05" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1084.6" cy="979.598" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1352.99" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="950.403" cy="947.548" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1487.19" cy="875.434" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1621.39" cy="735.212" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1755.58" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1889.78" cy="935.529" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="2023.97" cy="915.497" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="1889.78" cy="915.497" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="2158.17" cy="783.288" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip372)" cx="2292.37" cy="1248.02" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip370)" d="M2015.83 196.789 L2281.63 196.789 L2281.63 93.1086 L2015.83 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2015.83,196.789 2281.63,196.789 2281.63,93.1086 2015.83,93.1086 2015.83,196.789 "/>
-<circle clip-path="url(#clip370)" cx="2110.66" cy="144.949" r="20.48" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
-<path clip-path="url(#clip370)" d="M2219.34 164.636 Q2217.53 169.266 2215.82 170.678 Q2214.1 172.09 2211.23 172.09 L2207.83 172.09 L2207.83 168.525 L2210.33 168.525 Q2212.09 168.525 2213.06 167.692 Q2214.04 166.858 2215.22 163.756 L2215.98 161.812 L2205.49 136.303 L2210.01 136.303 L2218.11 156.581 L2226.21 136.303 L2230.72 136.303 L2219.34 164.636 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2238.02 158.293 L2245.66 158.293 L2245.66 131.928 L2237.35 133.595 L2237.35 129.335 L2245.61 127.669 L2250.28 127.669 L2250.28 158.293 L2257.92 158.293 L2257.92 162.229 L2238.02 162.229 L2238.02 158.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="413.618,1423.18 413.618,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="682.011,1423.18 682.011,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="950.403,1423.18 950.403,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1218.8,1423.18 1218.8,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1487.19,1423.18 1487.19,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1755.58,1423.18 1755.58,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2023.97,1423.18 2023.97,47.2441 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2292.37,1423.18 2292.37,47.2441 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="413.618,1423.18 413.618,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="682.011,1423.18 682.011,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="950.403,1423.18 950.403,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1218.8,1423.18 1218.8,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1487.19,1423.18 1487.19,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1755.58,1423.18 1755.58,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2023.97,1423.18 2023.97,1404.28 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2292.37,1423.18 2292.37,1404.28 "/>
+<path clip-path="url(#clip190)" d="M402.287 1455.09 L390.481 1473.54 L402.287 1473.54 L402.287 1455.09 M401.06 1451.02 L406.94 1451.02 L406.94 1473.54 L411.87 1473.54 L411.87 1477.43 L406.94 1477.43 L406.94 1485.58 L402.287 1485.58 L402.287 1477.43 L386.685 1477.43 L386.685 1472.92 L401.06 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M419.648 1451.02 L438.004 1451.02 L438.004 1454.96 L423.93 1454.96 L423.93 1463.43 Q424.949 1463.08 425.967 1462.92 Q426.986 1462.73 428.004 1462.73 Q433.791 1462.73 437.171 1465.9 Q440.55 1469.08 440.55 1474.49 Q440.55 1480.07 437.078 1483.17 Q433.606 1486.25 427.287 1486.25 Q425.111 1486.25 422.842 1485.88 Q420.597 1485.51 418.189 1484.77 L418.189 1480.07 Q420.273 1481.2 422.495 1481.76 Q424.717 1482.32 427.194 1482.32 Q431.199 1482.32 433.537 1480.21 Q435.875 1478.1 435.875 1474.49 Q435.875 1470.88 433.537 1468.77 Q431.199 1466.67 427.194 1466.67 Q425.319 1466.67 423.444 1467.08 Q421.592 1467.5 419.648 1468.38 L419.648 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M670.275 1455.09 L658.469 1473.54 L670.275 1473.54 L670.275 1455.09 M669.048 1451.02 L674.927 1451.02 L674.927 1473.54 L679.858 1473.54 L679.858 1477.43 L674.927 1477.43 L674.927 1485.58 L670.275 1485.58 L670.275 1477.43 L654.673 1477.43 L654.673 1472.92 L669.048 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M687.728 1484.86 L687.728 1480.6 Q689.487 1481.44 691.293 1481.88 Q693.098 1482.32 694.835 1482.32 Q699.464 1482.32 701.895 1479.21 Q704.348 1476.09 704.696 1469.75 Q703.353 1471.74 701.293 1472.8 Q699.233 1473.87 696.733 1473.87 Q691.548 1473.87 688.515 1470.74 Q685.506 1467.59 685.506 1462.15 Q685.506 1456.83 688.654 1453.61 Q691.802 1450.39 697.034 1450.39 Q703.029 1450.39 706.177 1455 Q709.348 1459.58 709.348 1468.33 Q709.348 1476.51 705.46 1481.39 Q701.594 1486.25 695.043 1486.25 Q693.284 1486.25 691.478 1485.9 Q689.673 1485.56 687.728 1484.86 M697.034 1470.21 Q700.182 1470.21 702.01 1468.06 Q703.862 1465.9 703.862 1462.15 Q703.862 1458.43 702.01 1456.27 Q700.182 1454.1 697.034 1454.1 Q693.885 1454.1 692.034 1456.27 Q690.205 1458.43 690.205 1462.15 Q690.205 1465.9 692.034 1468.06 Q693.885 1470.21 697.034 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M935.276 1469.17 Q931.943 1469.17 930.022 1470.95 Q928.123 1472.73 928.123 1475.86 Q928.123 1478.98 930.022 1480.77 Q931.943 1482.55 935.276 1482.55 Q938.609 1482.55 940.531 1480.77 Q942.452 1478.96 942.452 1475.86 Q942.452 1472.73 940.531 1470.95 Q938.633 1469.17 935.276 1469.17 M930.6 1467.18 Q927.591 1466.44 925.901 1464.38 Q924.235 1462.32 924.235 1459.35 Q924.235 1455.21 927.174 1452.8 Q930.137 1450.39 935.276 1450.39 Q940.438 1450.39 943.378 1452.8 Q946.318 1455.21 946.318 1459.35 Q946.318 1462.32 944.628 1464.38 Q942.961 1466.44 939.975 1467.18 Q943.355 1467.96 945.23 1470.26 Q947.128 1472.55 947.128 1475.86 Q947.128 1480.88 944.049 1483.57 Q940.994 1486.25 935.276 1486.25 Q929.559 1486.25 926.48 1483.57 Q923.424 1480.88 923.424 1475.86 Q923.424 1472.55 925.323 1470.26 Q927.221 1467.96 930.6 1467.18 M928.887 1459.79 Q928.887 1462.48 930.554 1463.98 Q932.244 1465.49 935.276 1465.49 Q938.285 1465.49 939.975 1463.98 Q941.688 1462.48 941.688 1459.79 Q941.688 1457.11 939.975 1455.6 Q938.285 1454.1 935.276 1454.1 Q932.244 1454.1 930.554 1455.6 Q928.887 1457.11 928.887 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M965.438 1454.1 Q961.827 1454.1 959.998 1457.66 Q958.193 1461.2 958.193 1468.33 Q958.193 1475.44 959.998 1479.01 Q961.827 1482.55 965.438 1482.55 Q969.072 1482.55 970.878 1479.01 Q972.707 1475.44 972.707 1468.33 Q972.707 1461.2 970.878 1457.66 Q969.072 1454.1 965.438 1454.1 M965.438 1450.39 Q971.248 1450.39 974.304 1455 Q977.382 1459.58 977.382 1468.33 Q977.382 1477.06 974.304 1481.67 Q971.248 1486.25 965.438 1486.25 Q959.628 1486.25 956.549 1481.67 Q953.494 1477.06 953.494 1468.33 Q953.494 1459.58 956.549 1455 Q959.628 1450.39 965.438 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1208.44 1466.95 Q1211.79 1467.66 1213.67 1469.93 Q1215.57 1472.2 1215.57 1475.53 Q1215.57 1480.65 1212.05 1483.45 Q1208.53 1486.25 1202.05 1486.25 Q1199.87 1486.25 1197.56 1485.81 Q1195.27 1485.39 1192.81 1484.54 L1192.81 1480.02 Q1194.76 1481.16 1197.07 1481.74 Q1199.39 1482.32 1201.91 1482.32 Q1206.31 1482.32 1208.6 1480.58 Q1210.91 1478.84 1210.91 1475.53 Q1210.91 1472.48 1208.76 1470.77 Q1206.63 1469.03 1202.81 1469.03 L1198.78 1469.03 L1198.78 1465.19 L1203 1465.19 Q1206.45 1465.19 1208.28 1463.82 Q1210.1 1462.43 1210.1 1459.84 Q1210.1 1457.18 1208.21 1455.77 Q1206.33 1454.33 1202.81 1454.33 Q1200.89 1454.33 1198.69 1454.75 Q1196.49 1455.16 1193.85 1456.04 L1193.85 1451.88 Q1196.52 1451.14 1198.83 1450.77 Q1201.17 1450.39 1203.23 1450.39 Q1208.55 1450.39 1211.66 1452.83 Q1214.76 1455.23 1214.76 1459.35 Q1214.76 1462.22 1213.11 1464.21 Q1211.47 1466.18 1208.44 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1228.46 1481.64 L1244.78 1481.64 L1244.78 1485.58 L1222.84 1485.58 L1222.84 1481.64 Q1225.5 1478.89 1230.08 1474.26 Q1234.69 1469.61 1235.87 1468.27 Q1238.11 1465.74 1238.99 1464.01 Q1239.9 1462.25 1239.9 1460.56 Q1239.9 1457.8 1237.95 1456.07 Q1236.03 1454.33 1232.93 1454.33 Q1230.73 1454.33 1228.28 1455.09 Q1225.84 1455.86 1223.07 1457.41 L1223.07 1452.69 Q1225.89 1451.55 1228.34 1450.97 Q1230.8 1450.39 1232.84 1450.39 Q1238.21 1450.39 1241.4 1453.08 Q1244.59 1455.77 1244.59 1460.26 Q1244.59 1462.39 1243.78 1464.31 Q1243 1466.2 1240.89 1468.8 Q1240.31 1469.47 1237.21 1472.69 Q1234.11 1475.88 1228.46 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1471.98 1469.17 Q1468.65 1469.17 1466.73 1470.95 Q1464.83 1472.73 1464.83 1475.86 Q1464.83 1478.98 1466.73 1480.77 Q1468.65 1482.55 1471.98 1482.55 Q1475.31 1482.55 1477.24 1480.77 Q1479.16 1478.96 1479.16 1475.86 Q1479.16 1472.73 1477.24 1470.95 Q1475.34 1469.17 1471.98 1469.17 M1467.3 1467.18 Q1464.3 1466.44 1462.61 1464.38 Q1460.94 1462.32 1460.94 1459.35 Q1460.94 1455.21 1463.88 1452.8 Q1466.84 1450.39 1471.98 1450.39 Q1477.14 1450.39 1480.08 1452.8 Q1483.02 1455.21 1483.02 1459.35 Q1483.02 1462.32 1481.33 1464.38 Q1479.67 1466.44 1476.68 1467.18 Q1480.06 1467.96 1481.93 1470.26 Q1483.83 1472.55 1483.83 1475.86 Q1483.83 1480.88 1480.75 1483.57 Q1477.7 1486.25 1471.98 1486.25 Q1466.26 1486.25 1463.18 1483.57 Q1460.13 1480.88 1460.13 1475.86 Q1460.13 1472.55 1462.03 1470.26 Q1463.93 1467.96 1467.3 1467.18 M1465.59 1459.79 Q1465.59 1462.48 1467.26 1463.98 Q1468.95 1465.49 1471.98 1465.49 Q1474.99 1465.49 1476.68 1463.98 Q1478.39 1462.48 1478.39 1459.79 Q1478.39 1457.11 1476.68 1455.6 Q1474.99 1454.1 1471.98 1454.1 Q1468.95 1454.1 1467.26 1455.6 Q1465.59 1457.11 1465.59 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1502.72 1466.44 Q1499.57 1466.44 1497.72 1468.59 Q1495.89 1470.74 1495.89 1474.49 Q1495.89 1478.22 1497.72 1480.39 Q1499.57 1482.55 1502.72 1482.55 Q1505.87 1482.55 1507.7 1480.39 Q1509.55 1478.22 1509.55 1474.49 Q1509.55 1470.74 1507.7 1468.59 Q1505.87 1466.44 1502.72 1466.44 M1512 1451.78 L1512 1456.04 Q1510.24 1455.21 1508.44 1454.77 Q1506.66 1454.33 1504.9 1454.33 Q1500.27 1454.33 1497.81 1457.45 Q1495.38 1460.58 1495.04 1466.9 Q1496.4 1464.89 1498.46 1463.82 Q1500.52 1462.73 1503 1462.73 Q1508.21 1462.73 1511.22 1465.9 Q1514.25 1469.05 1514.25 1474.49 Q1514.25 1479.82 1511.1 1483.03 Q1507.95 1486.25 1502.72 1486.25 Q1496.73 1486.25 1493.55 1481.67 Q1490.38 1477.06 1490.38 1468.33 Q1490.38 1460.14 1494.27 1455.28 Q1498.16 1450.39 1504.71 1450.39 Q1506.47 1450.39 1508.25 1450.74 Q1510.06 1451.09 1512 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1724.16 1451.02 L1730.45 1451.02 L1745.78 1479.93 L1745.78 1451.02 L1750.32 1451.02 L1750.32 1485.58 L1744.02 1485.58 L1728.7 1456.67 L1728.7 1485.58 L1724.16 1485.58 L1724.16 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1771.17 1455.63 L1764.83 1472.83 L1777.54 1472.83 L1771.17 1455.63 M1768.53 1451.02 L1773.83 1451.02 L1787.01 1485.58 L1782.14 1485.58 L1779 1476.71 L1763.42 1476.71 L1760.27 1485.58 L1755.34 1485.58 L1768.53 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1997.38 1451.02 L2019.6 1451.02 L2019.6 1453.01 L2007.05 1485.58 L2002.17 1485.58 L2013.97 1454.96 L1997.38 1454.96 L1997.38 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M2038.72 1469.17 Q2035.39 1469.17 2033.47 1470.95 Q2031.57 1472.73 2031.57 1475.86 Q2031.57 1478.98 2033.47 1480.77 Q2035.39 1482.55 2038.72 1482.55 Q2042.05 1482.55 2043.97 1480.77 Q2045.9 1478.96 2045.9 1475.86 Q2045.9 1472.73 2043.97 1470.95 Q2042.08 1469.17 2038.72 1469.17 M2034.04 1467.18 Q2031.03 1466.44 2029.35 1464.38 Q2027.68 1462.32 2027.68 1459.35 Q2027.68 1455.21 2030.62 1452.8 Q2033.58 1450.39 2038.72 1450.39 Q2043.88 1450.39 2046.82 1452.8 Q2049.76 1455.21 2049.76 1459.35 Q2049.76 1462.32 2048.07 1464.38 Q2046.41 1466.44 2043.42 1467.18 Q2046.8 1467.96 2048.67 1470.26 Q2050.57 1472.55 2050.57 1475.86 Q2050.57 1480.88 2047.49 1483.57 Q2044.44 1486.25 2038.72 1486.25 Q2033 1486.25 2029.92 1483.57 Q2026.87 1480.88 2026.87 1475.86 Q2026.87 1472.55 2028.77 1470.26 Q2030.66 1467.96 2034.04 1467.18 M2032.33 1459.79 Q2032.33 1462.48 2034 1463.98 Q2035.69 1465.49 2038.72 1465.49 Q2041.73 1465.49 2043.42 1463.98 Q2045.13 1462.48 2045.13 1459.79 Q2045.13 1457.11 2043.42 1455.6 Q2041.73 1454.1 2038.72 1454.1 Q2035.69 1454.1 2034 1455.6 Q2032.33 1457.11 2032.33 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M2267.07 1451.02 L2285.42 1451.02 L2285.42 1454.96 L2271.35 1454.96 L2271.35 1463.43 Q2272.37 1463.08 2273.39 1462.92 Q2274.4 1462.73 2275.42 1462.73 Q2281.21 1462.73 2284.59 1465.9 Q2287.97 1469.08 2287.97 1474.49 Q2287.97 1480.07 2284.5 1483.17 Q2281.02 1486.25 2274.71 1486.25 Q2272.53 1486.25 2270.26 1485.88 Q2268.02 1485.51 2265.61 1484.77 L2265.61 1480.07 Q2267.69 1481.2 2269.91 1481.76 Q2272.14 1482.32 2274.61 1482.32 Q2278.62 1482.32 2280.96 1480.21 Q2283.29 1478.1 2283.29 1474.49 Q2283.29 1470.88 2280.96 1468.77 Q2278.62 1466.67 2274.61 1466.67 Q2272.74 1466.67 2270.86 1467.08 Q2269.01 1467.5 2267.07 1468.38 L2267.07 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M2307.18 1454.1 Q2303.57 1454.1 2301.74 1457.66 Q2299.94 1461.2 2299.94 1468.33 Q2299.94 1475.44 2301.74 1479.01 Q2303.57 1482.55 2307.18 1482.55 Q2310.82 1482.55 2312.62 1479.01 Q2314.45 1475.44 2314.45 1468.33 Q2314.45 1461.2 2312.62 1457.66 Q2310.82 1454.1 2307.18 1454.1 M2307.18 1450.39 Q2312.99 1450.39 2316.05 1455 Q2319.13 1459.58 2319.13 1468.33 Q2319.13 1477.06 2316.05 1481.67 Q2312.99 1486.25 2307.18 1486.25 Q2301.37 1486.25 2298.29 1481.67 Q2295.24 1477.06 2295.24 1468.33 Q2295.24 1459.58 2298.29 1455 Q2301.37 1450.39 2307.18 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1174.21 1520.52 L1180.7 1520.52 L1190.7 1560.69 L1200.66 1520.52 L1207.88 1520.52 L1217.88 1560.69 L1227.84 1520.52 L1234.36 1520.52 L1222.43 1568.04 L1214.34 1568.04 L1204.32 1526.79 L1194.2 1568.04 L1186.11 1568.04 L1174.21 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1269.31 1548.76 L1269.31 1551.62 L1242.38 1551.62 Q1242.77 1557.67 1246.01 1560.85 Q1249.29 1564 1255.12 1564 Q1258.49 1564 1261.64 1563.17 Q1264.82 1562.35 1267.94 1560.69 L1267.94 1566.23 Q1264.79 1567.57 1261.48 1568.27 Q1258.17 1568.97 1254.77 1568.97 Q1246.24 1568.97 1241.24 1564 Q1236.27 1559.04 1236.27 1550.57 Q1236.27 1541.82 1240.98 1536.69 Q1245.73 1531.54 1253.75 1531.54 Q1260.94 1531.54 1265.11 1536.18 Q1269.31 1540.8 1269.31 1548.76 M1263.46 1547.04 Q1263.39 1542.23 1260.75 1539.37 Q1258.14 1536.5 1253.81 1536.5 Q1248.91 1536.5 1245.95 1539.27 Q1243.02 1542.04 1242.58 1547.07 L1263.46 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1278.92 1532.4 L1284.78 1532.4 L1284.78 1568.04 L1278.92 1568.04 L1278.92 1532.4 M1278.92 1518.52 L1284.78 1518.52 L1284.78 1525.93 L1278.92 1525.93 L1278.92 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1320.49 1549.81 Q1320.49 1543.44 1317.85 1539.94 Q1315.24 1536.44 1310.5 1536.44 Q1305.79 1536.44 1303.15 1539.94 Q1300.54 1543.44 1300.54 1549.81 Q1300.54 1556.14 1303.15 1559.64 Q1305.79 1563.14 1310.5 1563.14 Q1315.24 1563.14 1317.85 1559.64 Q1320.49 1556.14 1320.49 1549.81 M1326.35 1563.62 Q1326.35 1572.72 1322.31 1577.15 Q1318.26 1581.6 1309.92 1581.6 Q1306.84 1581.6 1304.1 1581.13 Q1301.36 1580.68 1298.78 1579.72 L1298.78 1574.03 Q1301.36 1575.43 1303.88 1576.1 Q1306.39 1576.76 1309 1576.76 Q1314.76 1576.76 1317.63 1573.74 Q1320.49 1570.75 1320.49 1564.67 L1320.49 1561.77 Q1318.68 1564.92 1315.85 1566.48 Q1313.01 1568.04 1309.07 1568.04 Q1302.51 1568.04 1298.5 1563.05 Q1294.49 1558.05 1294.49 1549.81 Q1294.49 1541.53 1298.5 1536.53 Q1302.51 1531.54 1309.07 1531.54 Q1313.01 1531.54 1315.85 1533.1 Q1318.68 1534.66 1320.49 1537.81 L1320.49 1532.4 L1326.35 1532.4 L1326.35 1563.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1368.04 1546.53 L1368.04 1568.04 L1362.19 1568.04 L1362.19 1546.72 Q1362.19 1541.66 1360.21 1539.14 Q1358.24 1536.63 1354.29 1536.63 Q1349.55 1536.63 1346.81 1539.65 Q1344.08 1542.68 1344.08 1547.9 L1344.08 1568.04 L1338.19 1568.04 L1338.19 1518.52 L1344.08 1518.52 L1344.08 1537.93 Q1346.18 1534.72 1349.01 1533.13 Q1351.87 1531.54 1355.6 1531.54 Q1361.74 1531.54 1364.89 1535.36 Q1368.04 1539.14 1368.04 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M1385.52 1522.27 L1385.52 1532.4 L1397.58 1532.4 L1397.58 1536.95 L1385.52 1536.95 L1385.52 1556.3 Q1385.52 1560.66 1386.7 1561.9 Q1387.9 1563.14 1391.57 1563.14 L1397.58 1563.14 L1397.58 1568.04 L1391.57 1568.04 Q1384.79 1568.04 1382.21 1565.53 Q1379.63 1562.98 1379.63 1556.3 L1379.63 1536.95 L1375.33 1536.95 L1375.33 1532.4 L1379.63 1532.4 L1379.63 1522.27 L1385.52 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,1248.02 2352.76,1248.02 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,1047.71 2352.76,1047.71 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,847.389 2352.76,847.389 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,647.073 2352.76,647.073 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,446.756 2352.76,446.756 "/>
+<polyline clip-path="url(#clip192)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.033,246.439 2352.76,246.439 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1423.18 219.033,47.2441 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1248.02 237.931,1248.02 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,1047.71 237.931,1047.71 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,847.389 237.931,847.389 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,647.073 237.931,647.073 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,446.756 237.931,446.756 "/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.033,246.439 237.931,246.439 "/>
+<path clip-path="url(#clip190)" d="M116.668 1261.37 L124.306 1261.37 L124.306 1235 L115.996 1236.67 L115.996 1232.41 L124.26 1230.74 L128.936 1230.74 L128.936 1261.37 L136.575 1261.37 L136.575 1265.3 L116.668 1265.3 L116.668 1261.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M146.019 1259.42 L150.903 1259.42 L150.903 1265.3 L146.019 1265.3 L146.019 1259.42 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M171.089 1233.82 Q167.477 1233.82 165.649 1237.39 Q163.843 1240.93 163.843 1248.06 Q163.843 1255.16 165.649 1258.73 Q167.477 1262.27 171.089 1262.27 Q174.723 1262.27 176.528 1258.73 Q178.357 1255.16 178.357 1248.06 Q178.357 1240.93 176.528 1237.39 Q174.723 1233.82 171.089 1233.82 M171.089 1230.12 Q176.899 1230.12 179.954 1234.72 Q183.033 1239.31 183.033 1248.06 Q183.033 1256.78 179.954 1261.39 Q176.899 1265.97 171.089 1265.97 Q165.278 1265.97 162.2 1261.39 Q159.144 1256.78 159.144 1248.06 Q159.144 1239.31 162.2 1234.72 Q165.278 1230.12 171.089 1230.12 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M117.663 1061.05 L125.302 1061.05 L125.302 1034.69 L116.992 1036.35 L116.992 1032.09 L125.255 1030.43 L129.931 1030.43 L129.931 1061.05 L137.57 1061.05 L137.57 1064.99 L117.663 1064.99 L117.663 1061.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M147.015 1059.11 L151.899 1059.11 L151.899 1064.99 L147.015 1064.99 L147.015 1059.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M162.13 1030.43 L180.487 1030.43 L180.487 1034.36 L166.413 1034.36 L166.413 1042.83 Q167.431 1042.49 168.45 1042.32 Q169.468 1042.14 170.487 1042.14 Q176.274 1042.14 179.653 1045.31 Q183.033 1048.48 183.033 1053.9 Q183.033 1059.48 179.561 1062.58 Q176.089 1065.66 169.769 1065.66 Q167.593 1065.66 165.325 1065.29 Q163.079 1064.92 160.672 1064.18 L160.672 1059.48 Q162.755 1060.61 164.977 1061.17 Q167.2 1061.72 169.677 1061.72 Q173.681 1061.72 176.019 1059.62 Q178.357 1057.51 178.357 1053.9 Q178.357 1050.29 176.019 1048.18 Q173.681 1046.07 169.677 1046.07 Q167.802 1046.07 165.927 1046.49 Q164.075 1046.91 162.13 1047.79 L162.13 1030.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M119.885 860.734 L136.204 860.734 L136.204 864.669 L114.26 864.669 L114.26 860.734 Q116.922 857.98 121.505 853.35 Q126.112 848.697 127.292 847.355 Q129.538 844.831 130.417 843.095 Q131.32 841.336 131.32 839.646 Q131.32 836.892 129.376 835.156 Q127.455 833.42 124.353 833.42 Q122.154 833.42 119.7 834.183 Q117.269 834.947 114.492 836.498 L114.492 831.776 Q117.316 830.642 119.769 830.063 Q122.223 829.484 124.26 829.484 Q129.63 829.484 132.825 832.17 Q136.019 834.855 136.019 839.345 Q136.019 841.475 135.209 843.396 Q134.422 845.294 132.316 847.887 Q131.737 848.558 128.635 851.776 Q125.533 854.97 119.885 860.734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M146.019 858.79 L150.903 858.79 L150.903 864.669 L146.019 864.669 L146.019 858.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M171.089 833.188 Q167.477 833.188 165.649 836.753 Q163.843 840.294 163.843 847.424 Q163.843 854.531 165.649 858.095 Q167.477 861.637 171.089 861.637 Q174.723 861.637 176.528 858.095 Q178.357 854.531 178.357 847.424 Q178.357 840.294 176.528 836.753 Q174.723 833.188 171.089 833.188 M171.089 829.484 Q176.899 829.484 179.954 834.091 Q183.033 838.674 183.033 847.424 Q183.033 856.151 179.954 860.757 Q176.899 865.341 171.089 865.341 Q165.278 865.341 162.2 860.757 Q159.144 856.151 159.144 847.424 Q159.144 838.674 162.2 834.091 Q165.278 829.484 171.089 829.484 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M120.88 660.417 L137.2 660.417 L137.2 664.353 L115.256 664.353 L115.256 660.417 Q117.918 657.663 122.501 653.033 Q127.107 648.38 128.288 647.038 Q130.533 644.515 131.413 642.779 Q132.316 641.019 132.316 639.33 Q132.316 636.575 130.371 634.839 Q128.45 633.103 125.348 633.103 Q123.149 633.103 120.695 633.867 Q118.265 634.631 115.487 636.181 L115.487 631.459 Q118.311 630.325 120.765 629.746 Q123.218 629.168 125.255 629.168 Q130.626 629.168 133.82 631.853 Q137.015 634.538 137.015 639.029 Q137.015 641.158 136.204 643.08 Q135.417 644.978 133.311 647.57 Q132.732 648.242 129.63 651.459 Q126.529 654.654 120.88 660.417 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M147.015 658.473 L151.899 658.473 L151.899 664.353 L147.015 664.353 L147.015 658.473 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M162.13 629.793 L180.487 629.793 L180.487 633.728 L166.413 633.728 L166.413 642.2 Q167.431 641.853 168.45 641.691 Q169.468 641.505 170.487 641.505 Q176.274 641.505 179.653 644.677 Q183.033 647.848 183.033 653.265 Q183.033 658.843 179.561 661.945 Q176.089 665.024 169.769 665.024 Q167.593 665.024 165.325 664.654 Q163.079 664.283 160.672 663.542 L160.672 658.843 Q162.755 659.978 164.977 660.533 Q167.2 661.089 169.677 661.089 Q173.681 661.089 176.019 658.982 Q178.357 656.876 178.357 653.265 Q178.357 649.654 176.019 647.547 Q173.681 645.441 169.677 645.441 Q167.802 645.441 165.927 645.857 Q164.075 646.274 162.13 647.154 L162.13 629.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M130.024 445.402 Q133.38 446.119 135.255 448.388 Q137.154 450.656 137.154 453.99 Q137.154 459.105 133.635 461.906 Q130.117 464.707 123.635 464.707 Q121.459 464.707 119.144 464.267 Q116.853 463.851 114.399 462.994 L114.399 458.48 Q116.343 459.615 118.658 460.193 Q120.973 460.772 123.496 460.772 Q127.894 460.772 130.186 459.036 Q132.501 457.3 132.501 453.99 Q132.501 450.934 130.348 449.221 Q128.218 447.485 124.399 447.485 L120.371 447.485 L120.371 443.642 L124.584 443.642 Q128.033 443.642 129.862 442.277 Q131.691 440.888 131.691 438.295 Q131.691 435.633 129.792 434.221 Q127.917 432.786 124.399 432.786 Q122.478 432.786 120.279 433.203 Q118.08 433.619 115.441 434.499 L115.441 430.332 Q118.103 429.592 120.418 429.221 Q122.755 428.851 124.816 428.851 Q130.14 428.851 133.242 431.281 Q136.343 433.689 136.343 437.809 Q136.343 440.679 134.7 442.67 Q133.056 444.638 130.024 445.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M146.019 458.156 L150.903 458.156 L150.903 464.036 L146.019 464.036 L146.019 458.156 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M171.089 432.555 Q167.477 432.555 165.649 436.119 Q163.843 439.661 163.843 446.791 Q163.843 453.897 165.649 457.462 Q167.477 461.003 171.089 461.003 Q174.723 461.003 176.528 457.462 Q178.357 453.897 178.357 446.791 Q178.357 439.661 176.528 436.119 Q174.723 432.555 171.089 432.555 M171.089 428.851 Q176.899 428.851 179.954 433.457 Q183.033 438.041 183.033 446.791 Q183.033 455.517 179.954 460.124 Q176.899 464.707 171.089 464.707 Q165.278 464.707 162.2 460.124 Q159.144 455.517 159.144 446.791 Q159.144 438.041 162.2 433.457 Q165.278 428.851 171.089 428.851 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M131.019 245.085 Q134.376 245.803 136.251 248.071 Q138.149 250.34 138.149 253.673 Q138.149 258.789 134.63 261.589 Q131.112 264.39 124.63 264.39 Q122.455 264.39 120.14 263.951 Q117.848 263.534 115.394 262.677 L115.394 258.164 Q117.339 259.298 119.654 259.877 Q121.968 260.455 124.492 260.455 Q128.89 260.455 131.181 258.719 Q133.496 256.983 133.496 253.673 Q133.496 250.617 131.343 248.904 Q129.214 247.168 125.394 247.168 L121.367 247.168 L121.367 243.326 L125.58 243.326 Q129.029 243.326 130.857 241.96 Q132.686 240.571 132.686 237.978 Q132.686 235.316 130.788 233.904 Q128.913 232.469 125.394 232.469 Q123.473 232.469 121.274 232.886 Q119.075 233.303 116.436 234.182 L116.436 230.016 Q119.098 229.275 121.413 228.904 Q123.751 228.534 125.811 228.534 Q131.135 228.534 134.237 230.965 Q137.339 233.372 137.339 237.492 Q137.339 240.363 135.695 242.353 Q134.052 244.321 131.019 245.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M147.015 257.839 L151.899 257.839 L151.899 263.719 L147.015 263.719 L147.015 257.839 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M162.13 229.159 L180.487 229.159 L180.487 233.094 L166.413 233.094 L166.413 241.566 Q167.431 241.219 168.45 241.057 Q169.468 240.872 170.487 240.872 Q176.274 240.872 179.653 244.043 Q183.033 247.215 183.033 252.631 Q183.033 258.21 179.561 261.312 Q176.089 264.39 169.769 264.39 Q167.593 264.39 165.325 264.02 Q163.079 263.65 160.672 262.909 L160.672 258.21 Q162.755 259.344 164.977 259.9 Q167.2 260.455 169.677 260.455 Q173.681 260.455 176.019 258.349 Q178.357 256.242 178.357 252.631 Q178.357 249.02 176.019 246.914 Q173.681 244.807 169.677 244.807 Q167.802 244.807 165.927 245.224 Q164.075 245.64 162.13 246.52 L162.13 229.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M16.4842 838.973 L16.4842 832.544 L35.9632 832.544 L35.9632 809.181 L16.4842 809.181 L16.4842 802.752 L64.0042 802.752 L64.0042 809.181 L41.3741 809.181 L41.3741 832.544 L64.0042 832.544 L64.0042 838.973 L16.4842 838.973 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M44.7161 759.72 L47.5806 759.72 L47.5806 786.647 Q53.6281 786.265 56.8109 783.018 Q59.9619 779.74 59.9619 773.915 Q59.9619 770.542 59.1344 767.391 Q58.3069 764.208 56.6518 761.089 L62.1899 761.089 Q63.5267 764.24 64.227 767.55 Q64.9272 770.86 64.9272 774.266 Q64.9272 782.796 59.9619 787.793 Q54.9967 792.758 46.5303 792.758 Q37.7774 792.758 32.6531 788.047 Q27.4968 783.305 27.4968 775.284 Q27.4968 768.091 32.1438 763.921 Q36.7589 759.72 44.7161 759.72 M42.9973 765.576 Q38.1912 765.64 35.3266 768.282 Q32.4621 770.892 32.4621 775.22 Q32.4621 780.122 35.2312 783.082 Q38.0002 786.01 43.0292 786.456 L42.9973 765.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M28.3562 750.108 L28.3562 744.251 L64.0042 744.251 L64.0042 750.108 L28.3562 750.108 M14.479 750.108 L14.479 744.251 L21.895 744.251 L21.895 750.108 L14.479 750.108 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M45.7664 708.54 Q39.4007 708.54 35.8996 711.181 Q32.3984 713.791 32.3984 718.534 Q32.3984 723.244 35.8996 725.886 Q39.4007 728.496 45.7664 728.496 Q52.1003 728.496 55.6014 725.886 Q59.1026 723.244 59.1026 718.534 Q59.1026 713.791 55.6014 711.181 Q52.1003 708.54 45.7664 708.54 M59.58 702.683 Q68.683 702.683 73.1071 706.725 Q77.5631 710.768 77.5631 719.107 Q77.5631 722.194 77.0857 724.931 Q76.6401 727.669 75.6852 730.247 L69.9879 730.247 Q71.3884 727.669 72.0568 725.154 Q72.7252 722.64 72.7252 720.03 Q72.7252 714.269 69.7015 711.404 Q66.7096 708.54 60.6303 708.54 L57.7339 708.54 Q60.885 710.354 62.4446 713.187 Q64.0042 716.019 64.0042 719.966 Q64.0042 726.523 59.0071 730.533 Q54.01 734.544 45.7664 734.544 Q37.491 734.544 32.4939 730.533 Q27.4968 726.523 27.4968 719.966 Q27.4968 716.019 29.0564 713.187 Q30.616 710.354 33.7671 708.54 L28.3562 708.54 L28.3562 702.683 L59.58 702.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M42.4881 660.988 L64.0042 660.988 L64.0042 666.844 L42.679 666.844 Q37.6183 666.844 35.1038 668.818 Q32.5894 670.791 32.5894 674.738 Q32.5894 679.48 35.6131 682.217 Q38.6368 684.955 43.8567 684.955 L64.0042 684.955 L64.0042 690.843 L14.479 690.843 L14.479 684.955 L33.8944 684.955 Q30.6797 682.854 29.0883 680.021 Q27.4968 677.157 27.4968 673.433 Q27.4968 667.29 31.3163 664.139 Q35.1038 660.988 42.4881 660.988 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M18.2347 643.514 L28.3562 643.514 L28.3562 631.451 L32.9077 631.451 L32.9077 643.514 L52.2594 643.514 Q56.6199 643.514 57.8613 642.336 Q59.1026 641.127 59.1026 637.467 L59.1026 631.451 L64.0042 631.451 L64.0042 637.467 Q64.0042 644.246 61.4897 646.824 Q58.9434 649.402 52.2594 649.402 L32.9077 649.402 L32.9077 653.699 L28.3562 653.699 L28.3562 649.402 L18.2347 649.402 L18.2347 643.514 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip192)" cx="279.421" cy="895.465" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="413.618" cy="987.611" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="547.814" cy="959.567" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="682.011" cy="1047.71" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="816.207" cy="875.434" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="547.814" cy="919.503" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="950.403" cy="927.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1084.6" cy="955.56" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1218.8" cy="1264.05" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1084.6" cy="979.598" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1352.99" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="950.403" cy="947.548" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1487.19" cy="875.434" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1621.39" cy="735.212" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1755.58" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1889.78" cy="935.529" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="2023.97" cy="915.497" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="1889.78" cy="915.497" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="2158.17" cy="783.288" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip192)" cx="2292.37" cy="1248.02" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip190)" d="M2015.83 196.789 L2281.63 196.789 L2281.63 93.1086 L2015.83 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip190)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2015.83,196.789 2281.63,196.789 2281.63,93.1086 2015.83,93.1086 2015.83,196.789 "/>
+<circle clip-path="url(#clip190)" cx="2110.66" cy="144.949" r="20.48" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
+<path clip-path="url(#clip190)" d="M2219.34 164.636 Q2217.53 169.266 2215.82 170.678 Q2214.1 172.09 2211.23 172.09 L2207.83 172.09 L2207.83 168.525 L2210.33 168.525 Q2212.09 168.525 2213.06 167.692 Q2214.04 166.858 2215.22 163.756 L2215.98 161.812 L2205.49 136.303 L2210.01 136.303 L2218.11 156.581 L2226.21 136.303 L2230.72 136.303 L2219.34 164.636 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip190)" d="M2238.02 158.293 L2245.66 158.293 L2245.66 131.928 L2237.35 133.595 L2237.35 129.335 L2245.61 127.669 L2250.28 127.669 L2250.28 158.293 L2257.92 158.293 L2257.92 162.229 L2238.02 162.229 L2238.02 158.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c55aa536.svg b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c9b9ec62.svg
similarity index 85%
rename from previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c55aa536.svg
rename to previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c9b9ec62.svg
index 7ce15147d9e..04deb3de797 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c55aa536.svg
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/c9b9ec62.svg
@@ -1,59 +1,59 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip430">
+  <clipPath id="clip250">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip430)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip250)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip431">
+  <clipPath id="clip251">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip430)" d="M174.149 1423.18 L2352.76 1423.18 L2352.76 123.472 L174.149 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip250)" d="M174.149 1423.18 L2352.76 1423.18 L2352.76 123.472 L174.149 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip432">
+  <clipPath id="clip252">
     <rect x="174" y="123" width="2180" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="423.968,1423.18 423.968,123.472 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="858.184,1423.18 858.184,123.472 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1292.4,1423.18 1292.4,123.472 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1726.62,1423.18 1726.62,123.472 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2160.83,1423.18 2160.83,123.472 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="423.968,1423.18 423.968,1404.28 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="858.184,1423.18 858.184,1404.28 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1292.4,1423.18 1292.4,1404.28 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1726.62,1423.18 1726.62,1404.28 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.83,1423.18 2160.83,1404.28 "/>
-<path clip-path="url(#clip430)" d="M412.811 1466.95 Q416.167 1467.66 418.042 1469.93 Q419.94 1472.2 419.94 1475.53 Q419.94 1480.65 416.422 1483.45 Q412.903 1486.25 406.422 1486.25 Q404.246 1486.25 401.931 1485.81 Q399.639 1485.39 397.186 1484.54 L397.186 1480.02 Q399.13 1481.16 401.445 1481.74 Q403.76 1482.32 406.283 1482.32 Q410.681 1482.32 412.973 1480.58 Q415.287 1478.84 415.287 1475.53 Q415.287 1472.48 413.135 1470.77 Q411.005 1469.03 407.186 1469.03 L403.158 1469.03 L403.158 1465.19 L407.371 1465.19 Q410.82 1465.19 412.649 1463.82 Q414.477 1462.43 414.477 1459.84 Q414.477 1457.18 412.579 1455.77 Q410.704 1454.33 407.186 1454.33 Q405.264 1454.33 403.065 1454.75 Q400.866 1455.16 398.227 1456.04 L398.227 1451.88 Q400.889 1451.14 403.204 1450.77 Q405.542 1450.39 407.602 1450.39 Q412.926 1450.39 416.028 1452.83 Q419.13 1455.23 419.13 1459.35 Q419.13 1462.22 417.486 1464.21 Q415.843 1466.18 412.811 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M438.806 1454.1 Q435.195 1454.1 433.366 1457.66 Q431.56 1461.2 431.56 1468.33 Q431.56 1475.44 433.366 1479.01 Q435.195 1482.55 438.806 1482.55 Q442.44 1482.55 444.246 1479.01 Q446.074 1475.44 446.074 1468.33 Q446.074 1461.2 444.246 1457.66 Q442.44 1454.1 438.806 1454.1 M438.806 1450.39 Q444.616 1450.39 447.671 1455 Q450.75 1459.58 450.75 1468.33 Q450.75 1477.06 447.671 1481.67 Q444.616 1486.25 438.806 1486.25 Q432.996 1486.25 429.917 1481.67 Q426.861 1477.06 426.861 1468.33 Q426.861 1459.58 429.917 1455 Q432.996 1450.39 438.806 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M843.589 1466.44 Q840.441 1466.44 838.589 1468.59 Q836.761 1470.74 836.761 1474.49 Q836.761 1478.22 838.589 1480.39 Q840.441 1482.55 843.589 1482.55 Q846.737 1482.55 848.566 1480.39 Q850.418 1478.22 850.418 1474.49 Q850.418 1470.74 848.566 1468.59 Q846.737 1466.44 843.589 1466.44 M852.872 1451.78 L852.872 1456.04 Q851.112 1455.21 849.307 1454.77 Q847.524 1454.33 845.765 1454.33 Q841.136 1454.33 838.682 1457.45 Q836.251 1460.58 835.904 1466.9 Q837.27 1464.89 839.33 1463.82 Q841.39 1462.73 843.867 1462.73 Q849.075 1462.73 852.085 1465.9 Q855.117 1469.05 855.117 1474.49 Q855.117 1479.82 851.969 1483.03 Q848.821 1486.25 843.589 1486.25 Q837.594 1486.25 834.423 1481.67 Q831.251 1477.06 831.251 1468.33 Q831.251 1460.14 835.14 1455.28 Q839.029 1450.39 845.58 1450.39 Q847.339 1450.39 849.122 1450.74 Q850.927 1451.09 852.872 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M873.172 1454.1 Q869.561 1454.1 867.733 1457.66 Q865.927 1461.2 865.927 1468.33 Q865.927 1475.44 867.733 1479.01 Q869.561 1482.55 873.172 1482.55 Q876.807 1482.55 878.612 1479.01 Q880.441 1475.44 880.441 1468.33 Q880.441 1461.2 878.612 1457.66 Q876.807 1454.1 873.172 1454.1 M873.172 1450.39 Q878.983 1450.39 882.038 1455 Q885.117 1459.58 885.117 1468.33 Q885.117 1477.06 882.038 1481.67 Q878.983 1486.25 873.172 1486.25 Q867.362 1486.25 864.284 1481.67 Q861.228 1477.06 861.228 1468.33 Q861.228 1459.58 864.284 1455 Q867.362 1450.39 873.172 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1267.53 1484.86 L1267.53 1480.6 Q1269.29 1481.44 1271.09 1481.88 Q1272.9 1482.32 1274.63 1482.32 Q1279.26 1482.32 1281.69 1479.21 Q1284.15 1476.09 1284.5 1469.75 Q1283.15 1471.74 1281.09 1472.8 Q1279.03 1473.87 1276.53 1473.87 Q1271.35 1473.87 1268.31 1470.74 Q1265.31 1467.59 1265.31 1462.15 Q1265.31 1456.83 1268.45 1453.61 Q1271.6 1450.39 1276.83 1450.39 Q1282.83 1450.39 1285.98 1455 Q1289.15 1459.58 1289.15 1468.33 Q1289.15 1476.51 1285.26 1481.39 Q1281.39 1486.25 1274.84 1486.25 Q1273.08 1486.25 1271.28 1485.9 Q1269.47 1485.56 1267.53 1484.86 M1276.83 1470.21 Q1279.98 1470.21 1281.81 1468.06 Q1283.66 1465.9 1283.66 1462.15 Q1283.66 1458.43 1281.81 1456.27 Q1279.98 1454.1 1276.83 1454.1 Q1273.68 1454.1 1271.83 1456.27 Q1270 1458.43 1270 1462.15 Q1270 1465.9 1271.83 1468.06 Q1273.68 1470.21 1276.83 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1307.55 1454.1 Q1303.94 1454.1 1302.11 1457.66 Q1300.31 1461.2 1300.31 1468.33 Q1300.31 1475.44 1302.11 1479.01 Q1303.94 1482.55 1307.55 1482.55 Q1311.18 1482.55 1312.99 1479.01 Q1314.82 1475.44 1314.82 1468.33 Q1314.82 1461.2 1312.99 1457.66 Q1311.18 1454.1 1307.55 1454.1 M1307.55 1450.39 Q1313.36 1450.39 1316.42 1455 Q1319.49 1459.58 1319.49 1468.33 Q1319.49 1477.06 1316.42 1481.67 Q1313.36 1486.25 1307.55 1486.25 Q1301.74 1486.25 1298.66 1481.67 Q1295.61 1477.06 1295.61 1468.33 Q1295.61 1459.58 1298.66 1455 Q1301.74 1450.39 1307.55 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1686.22 1481.64 L1693.86 1481.64 L1693.86 1455.28 L1685.55 1456.95 L1685.55 1452.69 L1693.82 1451.02 L1698.49 1451.02 L1698.49 1481.64 L1706.13 1481.64 L1706.13 1485.58 L1686.22 1485.58 L1686.22 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1719.6 1481.64 L1735.92 1481.64 L1735.92 1485.58 L1713.98 1485.58 L1713.98 1481.64 Q1716.64 1478.89 1721.22 1474.26 Q1725.83 1469.61 1727.01 1468.27 Q1729.26 1465.74 1730.13 1464.01 Q1731.04 1462.25 1731.04 1460.56 Q1731.04 1457.8 1729.09 1456.07 Q1727.17 1454.33 1724.07 1454.33 Q1721.87 1454.33 1719.42 1455.09 Q1716.99 1455.86 1714.21 1457.41 L1714.21 1452.69 Q1717.03 1451.55 1719.49 1450.97 Q1721.94 1450.39 1723.98 1450.39 Q1729.35 1450.39 1732.54 1453.08 Q1735.74 1455.77 1735.74 1460.26 Q1735.74 1462.39 1734.93 1464.31 Q1734.14 1466.2 1732.03 1468.8 Q1731.45 1469.47 1728.35 1472.69 Q1725.25 1475.88 1719.6 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1755.74 1454.1 Q1752.13 1454.1 1750.3 1457.66 Q1748.49 1461.2 1748.49 1468.33 Q1748.49 1475.44 1750.3 1479.01 Q1752.13 1482.55 1755.74 1482.55 Q1759.37 1482.55 1761.18 1479.01 Q1763 1475.44 1763 1468.33 Q1763 1461.2 1761.18 1457.66 Q1759.37 1454.1 1755.74 1454.1 M1755.74 1450.39 Q1761.55 1450.39 1764.6 1455 Q1767.68 1459.58 1767.68 1468.33 Q1767.68 1477.06 1764.6 1481.67 Q1761.55 1486.25 1755.74 1486.25 Q1749.93 1486.25 1746.85 1481.67 Q1743.79 1477.06 1743.79 1468.33 Q1743.79 1459.58 1746.85 1455 Q1749.93 1450.39 1755.74 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2120.44 1481.64 L2128.08 1481.64 L2128.08 1455.28 L2119.77 1456.95 L2119.77 1452.69 L2128.03 1451.02 L2132.71 1451.02 L2132.71 1481.64 L2140.35 1481.64 L2140.35 1485.58 L2120.44 1485.58 L2120.44 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2149.84 1451.02 L2168.19 1451.02 L2168.19 1454.96 L2154.12 1454.96 L2154.12 1463.43 Q2155.14 1463.08 2156.16 1462.92 Q2157.17 1462.73 2158.19 1462.73 Q2163.98 1462.73 2167.36 1465.9 Q2170.74 1469.08 2170.74 1474.49 Q2170.74 1480.07 2167.27 1483.17 Q2163.8 1486.25 2157.48 1486.25 Q2155.3 1486.25 2153.03 1485.88 Q2150.79 1485.51 2148.38 1484.77 L2148.38 1480.07 Q2150.46 1481.2 2152.68 1481.76 Q2154.91 1482.32 2157.38 1482.32 Q2161.39 1482.32 2163.73 1480.21 Q2166.06 1478.1 2166.06 1474.49 Q2166.06 1470.88 2163.73 1468.77 Q2161.39 1466.67 2157.38 1466.67 Q2155.51 1466.67 2153.63 1467.08 Q2151.78 1467.5 2149.84 1468.38 L2149.84 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2189.95 1454.1 Q2186.34 1454.1 2184.51 1457.66 Q2182.71 1461.2 2182.71 1468.33 Q2182.71 1475.44 2184.51 1479.01 Q2186.34 1482.55 2189.95 1482.55 Q2193.59 1482.55 2195.39 1479.01 Q2197.22 1475.44 2197.22 1468.33 Q2197.22 1461.2 2195.39 1457.66 Q2193.59 1454.1 2189.95 1454.1 M2189.95 1450.39 Q2195.76 1450.39 2198.82 1455 Q2201.9 1459.58 2201.9 1468.33 Q2201.9 1477.06 2198.82 1481.67 Q2195.76 1486.25 2189.95 1486.25 Q2184.14 1486.25 2181.06 1481.67 Q2178.01 1477.06 2178.01 1468.33 Q2178.01 1459.58 2181.06 1455 Q2184.14 1450.39 2189.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1151.77 1520.52 L1158.26 1520.52 L1168.25 1560.69 L1178.22 1520.52 L1185.44 1520.52 L1195.43 1560.69 L1205.4 1520.52 L1211.92 1520.52 L1199.99 1568.04 L1191.9 1568.04 L1181.88 1526.79 L1171.75 1568.04 L1163.67 1568.04 L1151.77 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1246.87 1548.76 L1246.87 1551.62 L1219.94 1551.62 Q1220.32 1557.67 1223.57 1560.85 Q1226.85 1564 1232.67 1564 Q1236.05 1564 1239.2 1563.17 Q1242.38 1562.35 1245.5 1560.69 L1245.5 1566.23 Q1242.35 1567.57 1239.04 1568.27 Q1235.73 1568.97 1232.32 1568.97 Q1223.79 1568.97 1218.8 1564 Q1213.83 1559.04 1213.83 1550.57 Q1213.83 1541.82 1218.54 1536.69 Q1223.28 1531.54 1231.31 1531.54 Q1238.5 1531.54 1242.67 1536.18 Q1246.87 1540.8 1246.87 1548.76 M1241.01 1547.04 Q1240.95 1542.23 1238.31 1539.37 Q1235.7 1536.5 1231.37 1536.5 Q1226.47 1536.5 1223.51 1539.27 Q1220.58 1542.04 1220.13 1547.07 L1241.01 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1256.48 1532.4 L1262.34 1532.4 L1262.34 1568.04 L1256.48 1568.04 L1256.48 1532.4 M1256.48 1518.52 L1262.34 1518.52 L1262.34 1525.93 L1256.48 1525.93 L1256.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1298.05 1549.81 Q1298.05 1543.44 1295.41 1539.94 Q1292.8 1536.44 1288.06 1536.44 Q1283.35 1536.44 1280.7 1539.94 Q1278.09 1543.44 1278.09 1549.81 Q1278.09 1556.14 1280.7 1559.64 Q1283.35 1563.14 1288.06 1563.14 Q1292.8 1563.14 1295.41 1559.64 Q1298.05 1556.14 1298.05 1549.81 M1303.91 1563.62 Q1303.91 1572.72 1299.86 1577.15 Q1295.82 1581.6 1287.48 1581.6 Q1284.4 1581.6 1281.66 1581.13 Q1278.92 1580.68 1276.34 1579.72 L1276.34 1574.03 Q1278.92 1575.43 1281.44 1576.1 Q1283.95 1576.76 1286.56 1576.76 Q1292.32 1576.76 1295.19 1573.74 Q1298.05 1570.75 1298.05 1564.67 L1298.05 1561.77 Q1296.24 1564.92 1293.4 1566.48 Q1290.57 1568.04 1286.62 1568.04 Q1280.07 1568.04 1276.06 1563.05 Q1272.05 1558.05 1272.05 1549.81 Q1272.05 1541.53 1276.06 1536.53 Q1280.07 1531.54 1286.62 1531.54 Q1290.57 1531.54 1293.4 1533.1 Q1296.24 1534.66 1298.05 1537.81 L1298.05 1532.4 L1303.91 1532.4 L1303.91 1563.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1345.6 1546.53 L1345.6 1568.04 L1339.75 1568.04 L1339.75 1546.72 Q1339.75 1541.66 1337.77 1539.14 Q1335.8 1536.63 1331.85 1536.63 Q1327.11 1536.63 1324.37 1539.65 Q1321.63 1542.68 1321.63 1547.9 L1321.63 1568.04 L1315.75 1568.04 L1315.75 1518.52 L1321.63 1518.52 L1321.63 1537.93 Q1323.74 1534.72 1326.57 1533.13 Q1329.43 1531.54 1333.16 1531.54 Q1339.3 1531.54 1342.45 1535.36 Q1345.6 1539.14 1345.6 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1363.08 1522.27 L1363.08 1532.4 L1375.14 1532.4 L1375.14 1536.95 L1363.08 1536.95 L1363.08 1556.3 Q1363.08 1560.66 1364.25 1561.9 Q1365.46 1563.14 1369.12 1563.14 L1375.14 1563.14 L1375.14 1568.04 L1369.12 1568.04 Q1362.34 1568.04 1359.77 1565.53 Q1357.19 1562.98 1357.19 1556.3 L1357.19 1536.95 L1352.89 1536.95 L1352.89 1532.4 L1357.19 1532.4 L1357.19 1522.27 L1363.08 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,989.944 2352.76,989.944 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,556.708 2352.76,556.708 "/>
-<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,123.472 2352.76,123.472 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 174.149,123.472 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 193.047,1423.18 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,989.944 193.047,989.944 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,556.708 193.047,556.708 "/>
-<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,123.472 193.047,123.472 "/>
-<path clip-path="url(#clip430)" d="M126.205 1408.98 Q122.593 1408.98 120.765 1412.54 Q118.959 1416.08 118.959 1423.21 Q118.959 1430.32 120.765 1433.89 Q122.593 1437.43 126.205 1437.43 Q129.839 1437.43 131.644 1433.89 Q133.473 1430.32 133.473 1423.21 Q133.473 1416.08 131.644 1412.54 Q129.839 1408.98 126.205 1408.98 M126.205 1405.27 Q132.015 1405.27 135.07 1409.88 Q138.149 1414.46 138.149 1423.21 Q138.149 1431.94 135.07 1436.55 Q132.015 1441.13 126.205 1441.13 Q120.394 1441.13 117.316 1436.55 Q114.26 1431.94 114.26 1423.21 Q114.26 1414.46 117.316 1409.88 Q120.394 1405.27 126.205 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M118.242 1003.29 L125.88 1003.29 L125.88 976.923 L117.57 978.59 L117.57 974.331 L125.834 972.664 L130.51 972.664 L130.51 1003.29 L138.149 1003.29 L138.149 1007.22 L118.242 1007.22 L118.242 1003.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M121.83 570.053 L138.149 570.053 L138.149 573.988 L116.205 573.988 L116.205 570.053 Q118.867 567.298 123.45 562.669 Q128.056 558.016 129.237 556.673 Q131.482 554.15 132.362 552.414 Q133.265 550.655 133.265 548.965 Q133.265 546.21 131.32 544.474 Q129.399 542.738 126.297 542.738 Q124.098 542.738 121.644 543.502 Q119.214 544.266 116.436 545.817 L116.436 541.095 Q119.26 539.96 121.714 539.382 Q124.168 538.803 126.205 538.803 Q131.575 538.803 134.769 541.488 Q137.964 544.173 137.964 548.664 Q137.964 550.794 137.154 552.715 Q136.367 554.613 134.26 557.206 Q133.681 557.877 130.58 561.095 Q127.478 564.289 121.83 570.053 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M131.019 122.118 Q134.376 122.836 136.251 125.104 Q138.149 127.373 138.149 130.706 Q138.149 135.822 134.63 138.623 Q131.112 141.423 124.63 141.423 Q122.455 141.423 120.14 140.984 Q117.848 140.567 115.394 139.71 L115.394 135.197 Q117.339 136.331 119.654 136.91 Q121.968 137.488 124.492 137.488 Q128.89 137.488 131.181 135.752 Q133.496 134.016 133.496 130.706 Q133.496 127.65 131.343 125.937 Q129.214 124.201 125.394 124.201 L121.367 124.201 L121.367 120.359 L125.58 120.359 Q129.029 120.359 130.857 118.993 Q132.686 117.604 132.686 115.012 Q132.686 112.35 130.788 110.937 Q128.913 109.502 125.394 109.502 Q123.473 109.502 121.274 109.919 Q119.075 110.336 116.436 111.215 L116.436 107.049 Q119.098 106.308 121.413 105.937 Q123.751 105.567 125.811 105.567 Q131.135 105.567 134.237 107.998 Q137.339 110.405 137.339 114.525 Q137.339 117.396 135.695 119.386 Q134.052 121.354 131.019 122.118 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M16.4842 877.087 L16.4842 870.658 L35.9632 870.658 L35.9632 847.295 L16.4842 847.295 L16.4842 840.866 L64.0042 840.866 L64.0042 847.295 L41.3741 847.295 L41.3741 870.658 L64.0042 870.658 L64.0042 877.087 L16.4842 877.087 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M44.7161 797.834 L47.5806 797.834 L47.5806 824.761 Q53.6281 824.379 56.8109 821.132 Q59.9619 817.854 59.9619 812.029 Q59.9619 808.656 59.1344 805.505 Q58.3069 802.322 56.6518 799.203 L62.1899 799.203 Q63.5267 802.354 64.227 805.664 Q64.9272 808.974 64.9272 812.38 Q64.9272 820.91 59.9619 825.907 Q54.9967 830.872 46.5303 830.872 Q37.7774 830.872 32.6531 826.161 Q27.4968 821.419 27.4968 813.398 Q27.4968 806.205 32.1438 802.035 Q36.7589 797.834 44.7161 797.834 M42.9973 803.69 Q38.1912 803.754 35.3266 806.396 Q32.4621 809.006 32.4621 813.334 Q32.4621 818.236 35.2312 821.196 Q38.0002 824.124 43.0292 824.57 L42.9973 803.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M28.3562 788.222 L28.3562 782.365 L64.0042 782.365 L64.0042 788.222 L28.3562 788.222 M14.479 788.222 L14.479 782.365 L21.895 782.365 L21.895 788.222 L14.479 788.222 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M45.7664 746.654 Q39.4007 746.654 35.8996 749.295 Q32.3984 751.905 32.3984 756.648 Q32.3984 761.358 35.8996 764 Q39.4007 766.61 45.7664 766.61 Q52.1003 766.61 55.6014 764 Q59.1026 761.358 59.1026 756.648 Q59.1026 751.905 55.6014 749.295 Q52.1003 746.654 45.7664 746.654 M59.58 740.797 Q68.683 740.797 73.1071 744.839 Q77.5631 748.882 77.5631 757.221 Q77.5631 760.308 77.0857 763.045 Q76.6401 765.783 75.6852 768.361 L69.9879 768.361 Q71.3884 765.783 72.0568 763.268 Q72.7252 760.754 72.7252 758.144 Q72.7252 752.383 69.7015 749.518 Q66.7096 746.654 60.6303 746.654 L57.7339 746.654 Q60.885 748.468 62.4446 751.301 Q64.0042 754.133 64.0042 758.08 Q64.0042 764.637 59.0071 768.647 Q54.01 772.658 45.7664 772.658 Q37.491 772.658 32.4939 768.647 Q27.4968 764.637 27.4968 758.08 Q27.4968 754.133 29.0564 751.301 Q30.616 748.468 33.7671 746.654 L28.3562 746.654 L28.3562 740.797 L59.58 740.797 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M42.4881 699.102 L64.0042 699.102 L64.0042 704.958 L42.679 704.958 Q37.6183 704.958 35.1038 706.932 Q32.5894 708.905 32.5894 712.852 Q32.5894 717.594 35.6131 720.331 Q38.6368 723.069 43.8567 723.069 L64.0042 723.069 L64.0042 728.957 L14.479 728.957 L14.479 723.069 L33.8944 723.069 Q30.6797 720.968 29.0883 718.135 Q27.4968 715.271 27.4968 711.547 Q27.4968 705.404 31.3163 702.253 Q35.1038 699.102 42.4881 699.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M18.2347 681.628 L28.3562 681.628 L28.3562 669.565 L32.9077 669.565 L32.9077 681.628 L52.2594 681.628 Q56.6199 681.628 57.8613 680.45 Q59.1026 679.241 59.1026 675.581 L59.1026 669.565 L64.0042 669.565 L64.0042 675.581 Q64.0042 682.36 61.4897 684.938 Q58.9434 687.516 52.2594 687.516 L32.9077 687.516 L32.9077 691.813 L28.3562 691.813 L28.3562 687.516 L18.2347 687.516 L18.2347 681.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M428.075 12.096 L436.258 12.096 L436.258 36.8875 L465.991 36.8875 L465.991 12.096 L474.174 12.096 L474.174 72.576 L465.991 72.576 L465.991 43.7741 L436.258 43.7741 L436.258 72.576 L428.075 72.576 L428.075 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M528.942 48.0275 L528.942 51.6733 L494.672 51.6733 Q495.158 59.3701 499.29 63.421 Q503.462 67.4314 510.875 67.4314 Q515.169 67.4314 519.18 66.3781 Q523.231 65.3249 527.201 63.2184 L527.201 70.267 Q523.19 71.9684 518.977 72.8596 Q514.764 73.7508 510.43 73.7508 Q499.573 73.7508 493.213 67.4314 Q486.894 61.1119 486.894 50.3365 Q486.894 39.1965 492.889 32.6746 Q498.925 26.1121 509.134 26.1121 Q518.289 26.1121 523.595 32.0264 Q528.942 37.9003 528.942 48.0275 M521.489 45.84 Q521.408 39.7232 518.046 36.0774 Q514.724 32.4315 509.215 32.4315 Q502.976 32.4315 499.209 35.9558 Q495.482 39.4801 494.915 45.8805 L521.489 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M541.176 27.2059 L548.63 27.2059 L548.63 72.576 L541.176 72.576 L541.176 27.2059 M541.176 9.54393 L548.63 9.54393 L548.63 18.9825 L541.176 18.9825 L541.176 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M594.081 49.3643 Q594.081 41.2625 590.719 36.8065 Q587.397 32.3505 581.361 32.3505 Q575.366 32.3505 572.004 36.8065 Q568.682 41.2625 568.682 49.3643 Q568.682 57.4256 572.004 61.8816 Q575.366 66.3376 581.361 66.3376 Q587.397 66.3376 590.719 61.8816 Q594.081 57.4256 594.081 49.3643 M601.535 66.9452 Q601.535 78.5308 596.39 84.1616 Q591.245 89.8329 580.632 89.8329 Q576.703 89.8329 573.219 89.2252 Q569.735 88.6581 566.454 87.4428 L566.454 80.1917 Q569.735 81.9741 572.935 82.8248 Q576.135 83.6755 579.457 83.6755 Q586.789 83.6755 590.435 79.8271 Q594.081 76.0193 594.081 68.282 L594.081 64.5957 Q591.772 68.6061 588.167 70.5911 Q584.561 72.576 579.538 72.576 Q571.193 72.576 566.089 66.2161 Q560.985 59.8562 560.985 49.3643 Q560.985 38.832 566.089 32.472 Q571.193 26.1121 579.538 26.1121 Q584.561 26.1121 588.167 28.0971 Q591.772 30.082 594.081 34.0924 L594.081 27.2059 L601.535 27.2059 L601.535 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M654.602 45.1919 L654.602 72.576 L647.148 72.576 L647.148 45.4349 Q647.148 38.994 644.636 35.7938 Q642.125 32.5936 637.102 32.5936 Q631.066 32.5936 627.582 36.4419 Q624.098 40.2903 624.098 46.9338 L624.098 72.576 L616.604 72.576 L616.604 9.54393 L624.098 9.54393 L624.098 34.2544 Q626.772 30.163 630.377 28.1376 Q634.023 26.1121 638.762 26.1121 Q646.581 26.1121 650.591 30.9732 Q654.602 35.7938 654.602 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M676.841 14.324 L676.841 27.2059 L692.194 27.2059 L692.194 32.9987 L676.841 32.9987 L676.841 57.6282 Q676.841 63.1779 678.34 64.7578 Q679.879 66.3376 684.538 66.3376 L692.194 66.3376 L692.194 72.576 L684.538 72.576 Q675.909 72.576 672.628 69.3758 Q669.347 66.1351 669.347 57.6282 L669.347 32.9987 L663.878 32.9987 L663.878 27.2059 L669.347 27.2059 L669.347 14.324 L676.841 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M723.021 27.2059 L730.921 27.2059 L745.099 65.2844 L759.277 27.2059 L767.176 27.2059 L750.162 72.576 L740.035 72.576 L723.021 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M806.389 28.5427 L806.389 35.5912 Q803.229 33.9709 799.826 33.1607 Q796.424 32.3505 792.778 32.3505 Q787.228 32.3505 784.433 34.0519 Q781.678 35.7533 781.678 39.156 Q781.678 41.7486 783.663 43.2475 Q785.648 44.7058 791.644 46.0426 L794.196 46.6097 Q802.135 48.3111 805.457 51.4303 Q808.819 54.509 808.819 60.0587 Q808.819 66.3781 803.796 70.0644 Q798.814 73.7508 790.064 73.7508 Q786.418 73.7508 782.448 73.0216 Q778.519 72.3329 774.144 70.9151 L774.144 63.2184 Q778.276 65.3654 782.286 66.4591 Q786.296 67.5124 790.226 67.5124 Q795.492 67.5124 798.328 65.73 Q801.163 63.9071 801.163 60.6258 Q801.163 57.5877 799.097 55.9673 Q797.072 54.3469 790.145 52.8481 L787.552 52.2405 Q780.625 50.7821 777.546 47.7845 Q774.468 44.7463 774.468 39.4801 Q774.468 33.0797 779.005 29.5959 Q783.542 26.1121 791.887 26.1121 Q796.019 26.1121 799.664 26.7198 Q803.31 27.3274 806.389 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M841.996 12.096 L850.26 12.096 L862.98 63.2184 L875.659 12.096 L884.855 12.096 L897.575 63.2184 L910.254 12.096 L918.558 12.096 L903.368 72.576 L893.078 72.576 L880.318 20.0763 L867.436 72.576 L857.147 72.576 L841.996 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M963.037 48.0275 L963.037 51.6733 L928.767 51.6733 Q929.253 59.3701 933.385 63.421 Q937.557 67.4314 944.97 67.4314 Q949.264 67.4314 953.275 66.3781 Q957.326 65.3249 961.295 63.2184 L961.295 70.267 Q957.285 71.9684 953.072 72.8596 Q948.859 73.7508 944.525 73.7508 Q933.668 73.7508 927.308 67.4314 Q920.989 61.1119 920.989 50.3365 Q920.989 39.1965 926.984 32.6746 Q933.02 26.1121 943.228 26.1121 Q952.383 26.1121 957.69 32.0264 Q963.037 37.9003 963.037 48.0275 M955.584 45.84 Q955.503 39.7232 952.14 36.0774 Q948.819 32.4315 943.309 32.4315 Q937.071 32.4315 933.304 35.9558 Q929.577 39.4801 929.01 45.8805 L955.584 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M975.271 27.2059 L982.725 27.2059 L982.725 72.576 L975.271 72.576 L975.271 27.2059 M975.271 9.54393 L982.725 9.54393 L982.725 18.9825 L975.271 18.9825 L975.271 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1028.18 49.3643 Q1028.18 41.2625 1024.81 36.8065 Q1021.49 32.3505 1015.46 32.3505 Q1009.46 32.3505 1006.1 36.8065 Q1002.78 41.2625 1002.78 49.3643 Q1002.78 57.4256 1006.1 61.8816 Q1009.46 66.3376 1015.46 66.3376 Q1021.49 66.3376 1024.81 61.8816 Q1028.18 57.4256 1028.18 49.3643 M1035.63 66.9452 Q1035.63 78.5308 1030.48 84.1616 Q1025.34 89.8329 1014.73 89.8329 Q1010.8 89.8329 1007.31 89.2252 Q1003.83 88.6581 1000.55 87.4428 L1000.55 80.1917 Q1003.83 81.9741 1007.03 82.8248 Q1010.23 83.6755 1013.55 83.6755 Q1020.88 83.6755 1024.53 79.8271 Q1028.18 76.0193 1028.18 68.282 L1028.18 64.5957 Q1025.87 68.6061 1022.26 70.5911 Q1018.66 72.576 1013.63 72.576 Q1005.29 72.576 1000.18 66.2161 Q995.08 59.8562 995.08 49.3643 Q995.08 38.832 1000.18 32.472 Q1005.29 26.1121 1013.63 26.1121 Q1018.66 26.1121 1022.26 28.0971 Q1025.87 30.082 1028.18 34.0924 L1028.18 27.2059 L1035.63 27.2059 L1035.63 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1088.7 45.1919 L1088.7 72.576 L1081.24 72.576 L1081.24 45.4349 Q1081.24 38.994 1078.73 35.7938 Q1076.22 32.5936 1071.2 32.5936 Q1065.16 32.5936 1061.68 36.4419 Q1058.19 40.2903 1058.19 46.9338 L1058.19 72.576 L1050.7 72.576 L1050.7 9.54393 L1058.19 9.54393 L1058.19 34.2544 Q1060.87 30.163 1064.47 28.1376 Q1068.12 26.1121 1072.86 26.1121 Q1080.68 26.1121 1084.69 30.9732 Q1088.7 35.7938 1088.7 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1110.94 14.324 L1110.94 27.2059 L1126.29 27.2059 L1126.29 32.9987 L1110.94 32.9987 L1110.94 57.6282 Q1110.94 63.1779 1112.43 64.7578 Q1113.97 66.3376 1118.63 66.3376 L1126.29 66.3376 L1126.29 72.576 L1118.63 72.576 Q1110 72.576 1106.72 69.3758 Q1103.44 66.1351 1103.44 57.6282 L1103.44 32.9987 L1097.97 32.9987 L1097.97 27.2059 L1103.44 27.2059 L1103.44 14.324 L1110.94 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1180.04 32.4315 Q1174.05 32.4315 1170.57 37.1306 Q1167.08 41.7891 1167.08 49.9314 Q1167.08 58.0738 1170.52 62.7728 Q1174.01 67.4314 1180.04 67.4314 Q1186 67.4314 1189.48 62.7323 Q1192.97 58.0333 1192.97 49.9314 Q1192.97 41.8701 1189.48 37.1711 Q1186 32.4315 1180.04 32.4315 M1180.04 26.1121 Q1189.77 26.1121 1195.32 32.4315 Q1200.87 38.7509 1200.87 49.9314 Q1200.87 61.0714 1195.32 67.4314 Q1189.77 73.7508 1180.04 73.7508 Q1170.28 73.7508 1164.73 67.4314 Q1159.22 61.0714 1159.22 49.9314 Q1159.22 38.7509 1164.73 32.4315 Q1170.28 26.1121 1180.04 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1236.19 9.54393 L1236.19 15.7418 L1229.06 15.7418 Q1225.05 15.7418 1223.47 17.3622 Q1221.93 18.9825 1221.93 23.1955 L1221.93 27.2059 L1234.2 27.2059 L1234.2 32.9987 L1221.93 32.9987 L1221.93 72.576 L1214.44 72.576 L1214.44 32.9987 L1207.31 32.9987 L1207.31 27.2059 L1214.44 27.2059 L1214.44 24.0462 Q1214.44 16.471 1217.96 13.0277 Q1221.49 9.54393 1229.14 9.54393 L1236.19 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1305.38 14.0809 L1305.38 22.0612 Q1300.72 19.8332 1296.59 18.7395 Q1292.46 17.6457 1288.61 17.6457 Q1281.92 17.6457 1278.28 20.2383 Q1274.67 22.8309 1274.67 27.611 Q1274.67 31.6214 1277.06 33.6873 Q1279.49 35.7128 1286.22 36.9686 L1291.16 37.9813 Q1300.32 39.7232 1304.65 44.1387 Q1309.03 48.5136 1309.03 55.8863 Q1309.03 64.6767 1303.11 69.2137 Q1297.24 73.7508 1285.85 73.7508 Q1281.56 73.7508 1276.7 72.7785 Q1271.88 71.8063 1266.69 69.9024 L1266.69 61.4765 Q1271.68 64.2716 1276.46 65.6895 Q1281.24 67.1073 1285.85 67.1073 Q1292.86 67.1073 1296.67 64.3527 Q1300.48 61.598 1300.48 56.4939 Q1300.48 52.0379 1297.72 49.5264 Q1295.01 47.0148 1288.77 45.759 L1283.79 44.7868 Q1274.63 42.9639 1270.54 39.075 Q1266.45 35.1862 1266.45 28.2591 Q1266.45 20.2383 1272.08 15.6203 Q1277.75 11.0023 1287.68 11.0023 Q1291.93 11.0023 1296.35 11.7719 Q1300.76 12.5416 1305.38 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1328.83 14.324 L1328.83 27.2059 L1344.19 27.2059 L1344.19 32.9987 L1328.83 32.9987 L1328.83 57.6282 Q1328.83 63.1779 1330.33 64.7578 Q1331.87 66.3376 1336.53 66.3376 L1344.19 66.3376 L1344.19 72.576 L1336.53 72.576 Q1327.9 72.576 1324.62 69.3758 Q1321.34 66.1351 1321.34 57.6282 L1321.34 32.9987 L1315.87 32.9987 L1315.87 27.2059 L1321.34 27.2059 L1321.34 14.324 L1328.83 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1374.61 49.7694 Q1365.58 49.7694 1362.09 51.8354 Q1358.61 53.9013 1358.61 58.8839 Q1358.61 62.8538 1361.2 65.2034 Q1363.83 67.5124 1368.33 67.5124 Q1374.53 67.5124 1378.26 63.1374 Q1382.02 58.7219 1382.02 51.4303 L1382.02 49.7694 L1374.61 49.7694 M1389.48 46.6907 L1389.48 72.576 L1382.02 72.576 L1382.02 65.6895 Q1379.47 69.8214 1375.66 71.8063 Q1371.85 73.7508 1366.35 73.7508 Q1359.38 73.7508 1355.25 69.8619 Q1351.15 65.9325 1351.15 59.3701 Q1351.15 51.7138 1356.26 47.825 Q1361.4 43.9361 1371.57 43.9361 L1382.02 43.9361 L1382.02 43.2069 Q1382.02 38.0623 1378.62 35.2672 Q1375.26 32.4315 1369.14 32.4315 Q1365.25 32.4315 1361.57 33.3632 Q1357.88 34.295 1354.48 36.1584 L1354.48 29.2718 Q1358.57 27.692 1362.42 26.9223 Q1366.26 26.1121 1369.91 26.1121 Q1379.75 26.1121 1384.61 31.2163 Q1389.48 36.3204 1389.48 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1431.12 34.1734 Q1429.86 33.4443 1428.36 33.1202 Q1426.91 32.7556 1425.12 32.7556 Q1418.8 32.7556 1415.4 36.8875 Q1412.04 40.9789 1412.04 48.6757 L1412.04 72.576 L1404.55 72.576 L1404.55 27.2059 L1412.04 27.2059 L1412.04 34.2544 Q1414.39 30.1225 1418.16 28.1376 Q1421.92 26.1121 1427.31 26.1121 Q1428.08 26.1121 1429.01 26.2337 Q1429.94 26.3147 1431.08 26.5172 L1431.12 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1433.87 12.096 L1442.14 12.096 L1454.86 63.2184 L1467.54 12.096 L1476.73 12.096 L1489.45 63.2184 L1502.13 12.096 L1510.44 12.096 L1495.25 72.576 L1484.96 72.576 L1472.2 20.0763 L1459.31 72.576 L1449.02 72.576 L1433.87 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1536.28 49.7694 Q1527.25 49.7694 1523.76 51.8354 Q1520.28 53.9013 1520.28 58.8839 Q1520.28 62.8538 1522.87 65.2034 Q1525.51 67.5124 1530 67.5124 Q1536.2 67.5124 1539.93 63.1374 Q1543.69 58.7219 1543.69 51.4303 L1543.69 49.7694 L1536.28 49.7694 M1551.15 46.6907 L1551.15 72.576 L1543.69 72.576 L1543.69 65.6895 Q1541.14 69.8214 1537.33 71.8063 Q1533.53 73.7508 1528.02 73.7508 Q1521.05 73.7508 1516.92 69.8619 Q1512.83 65.9325 1512.83 59.3701 Q1512.83 51.7138 1517.93 47.825 Q1523.07 43.9361 1533.24 43.9361 L1543.69 43.9361 L1543.69 43.2069 Q1543.69 38.0623 1540.29 35.2672 Q1536.93 32.4315 1530.81 32.4315 Q1526.92 32.4315 1523.24 33.3632 Q1519.55 34.295 1516.15 36.1584 L1516.15 29.2718 Q1520.24 27.692 1524.09 26.9223 Q1527.94 26.1121 1531.58 26.1121 Q1541.43 26.1121 1546.29 31.2163 Q1551.15 36.3204 1551.15 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1592.79 34.1734 Q1591.54 33.4443 1590.04 33.1202 Q1588.58 32.7556 1586.8 32.7556 Q1580.48 32.7556 1577.07 36.8875 Q1573.71 40.9789 1573.71 48.6757 L1573.71 72.576 L1566.22 72.576 L1566.22 27.2059 L1573.71 27.2059 L1573.71 34.2544 Q1576.06 30.1225 1579.83 28.1376 Q1583.6 26.1121 1588.98 26.1121 Q1589.75 26.1121 1590.68 26.2337 Q1591.62 26.3147 1592.75 26.5172 L1592.79 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1629.53 28.5427 L1629.53 35.5912 Q1626.37 33.9709 1622.97 33.1607 Q1619.57 32.3505 1615.92 32.3505 Q1610.37 32.3505 1607.58 34.0519 Q1604.82 35.7533 1604.82 39.156 Q1604.82 41.7486 1606.81 43.2475 Q1608.79 44.7058 1614.79 46.0426 L1617.34 46.6097 Q1625.28 48.3111 1628.6 51.4303 Q1631.96 54.509 1631.96 60.0587 Q1631.96 66.3781 1626.94 70.0644 Q1621.96 73.7508 1613.21 73.7508 Q1609.56 73.7508 1605.59 73.0216 Q1601.66 72.3329 1597.29 70.9151 L1597.29 63.2184 Q1601.42 65.3654 1605.43 66.4591 Q1609.44 67.5124 1613.37 67.5124 Q1618.64 67.5124 1621.47 65.73 Q1624.31 63.9071 1624.31 60.6258 Q1624.31 57.5877 1622.24 55.9673 Q1620.22 54.3469 1613.29 52.8481 L1610.7 52.2405 Q1603.77 50.7821 1600.69 47.7845 Q1597.61 44.7463 1597.61 39.4801 Q1597.61 33.0797 1602.15 29.5959 Q1606.69 26.1121 1615.03 26.1121 Q1619.16 26.1121 1622.81 26.7198 Q1626.45 27.3274 1629.53 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1702.85 28.9478 L1702.85 35.9153 Q1699.69 34.1734 1696.49 33.3227 Q1693.33 32.4315 1690.09 32.4315 Q1682.84 32.4315 1678.83 37.0496 Q1674.82 41.6271 1674.82 49.9314 Q1674.82 58.2358 1678.83 62.8538 Q1682.84 67.4314 1690.09 67.4314 Q1693.33 67.4314 1696.49 66.5807 Q1699.69 65.6895 1702.85 63.9476 L1702.85 70.8341 Q1699.73 72.2924 1696.37 73.0216 Q1693.05 73.7508 1689.28 73.7508 Q1679.03 73.7508 1673 67.3098 Q1666.96 60.8689 1666.96 49.9314 Q1666.96 38.832 1673.04 32.472 Q1679.16 26.1121 1689.77 26.1121 Q1693.21 26.1121 1696.49 26.8413 Q1699.78 27.5299 1702.85 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1753.53 45.1919 L1753.53 72.576 L1746.08 72.576 L1746.08 45.4349 Q1746.08 38.994 1743.57 35.7938 Q1741.05 32.5936 1736.03 32.5936 Q1730 32.5936 1726.51 36.4419 Q1723.03 40.2903 1723.03 46.9338 L1723.03 72.576 L1715.53 72.576 L1715.53 9.54393 L1723.03 9.54393 L1723.03 34.2544 Q1725.7 30.163 1729.31 28.1376 Q1732.95 26.1121 1737.69 26.1121 Q1745.51 26.1121 1749.52 30.9732 Q1753.53 35.7938 1753.53 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1789.02 49.7694 Q1779.98 49.7694 1776.5 51.8354 Q1773.02 53.9013 1773.02 58.8839 Q1773.02 62.8538 1775.61 65.2034 Q1778.24 67.5124 1782.74 67.5124 Q1788.94 67.5124 1792.66 63.1374 Q1796.43 58.7219 1796.43 51.4303 L1796.43 49.7694 L1789.02 49.7694 M1803.88 46.6907 L1803.88 72.576 L1796.43 72.576 L1796.43 65.6895 Q1793.88 69.8214 1790.07 71.8063 Q1786.26 73.7508 1780.75 73.7508 Q1773.79 73.7508 1769.65 69.8619 Q1765.56 65.9325 1765.56 59.3701 Q1765.56 51.7138 1770.67 47.825 Q1775.81 43.9361 1785.98 43.9361 L1796.43 43.9361 L1796.43 43.2069 Q1796.43 38.0623 1793.03 35.2672 Q1789.66 32.4315 1783.55 32.4315 Q1779.66 32.4315 1775.97 33.3632 Q1772.29 34.295 1768.88 36.1584 L1768.88 29.2718 Q1772.98 27.692 1776.82 26.9223 Q1780.67 26.1121 1784.32 26.1121 Q1794.16 26.1121 1799.02 31.2163 Q1803.88 36.3204 1803.88 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1845.53 34.1734 Q1844.27 33.4443 1842.77 33.1202 Q1841.31 32.7556 1839.53 32.7556 Q1833.21 32.7556 1829.81 36.8875 Q1826.45 40.9789 1826.45 48.6757 L1826.45 72.576 L1818.95 72.576 L1818.95 27.2059 L1826.45 27.2059 L1826.45 34.2544 Q1828.8 30.1225 1832.56 28.1376 Q1836.33 26.1121 1841.72 26.1121 Q1842.49 26.1121 1843.42 26.2337 Q1844.35 26.3147 1845.49 26.5172 L1845.53 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1873.96 49.7694 Q1864.93 49.7694 1861.45 51.8354 Q1857.96 53.9013 1857.96 58.8839 Q1857.96 62.8538 1860.56 65.2034 Q1863.19 67.5124 1867.69 67.5124 Q1873.88 67.5124 1877.61 63.1374 Q1881.38 58.7219 1881.38 51.4303 L1881.38 49.7694 L1873.96 49.7694 M1888.83 46.6907 L1888.83 72.576 L1881.38 72.576 L1881.38 65.6895 Q1878.83 69.8214 1875.02 71.8063 Q1871.21 73.7508 1865.7 73.7508 Q1858.73 73.7508 1854.6 69.8619 Q1850.51 65.9325 1850.51 59.3701 Q1850.51 51.7138 1855.61 47.825 Q1860.76 43.9361 1870.93 43.9361 L1881.38 43.9361 L1881.38 43.2069 Q1881.38 38.0623 1877.97 35.2672 Q1874.61 32.4315 1868.5 32.4315 Q1864.61 32.4315 1860.92 33.3632 Q1857.23 34.295 1853.83 36.1584 L1853.83 29.2718 Q1857.92 27.692 1861.77 26.9223 Q1865.62 26.1121 1869.27 26.1121 Q1879.11 26.1121 1883.97 31.2163 Q1888.83 36.3204 1888.83 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1936.83 28.9478 L1936.83 35.9153 Q1933.67 34.1734 1930.47 33.3227 Q1927.31 32.4315 1924.07 32.4315 Q1916.82 32.4315 1912.81 37.0496 Q1908.8 41.6271 1908.8 49.9314 Q1908.8 58.2358 1912.81 62.8538 Q1916.82 67.4314 1924.07 67.4314 Q1927.31 67.4314 1930.47 66.5807 Q1933.67 65.6895 1936.83 63.9476 L1936.83 70.8341 Q1933.72 72.2924 1930.35 73.0216 Q1927.03 73.7508 1923.26 73.7508 Q1913.01 73.7508 1906.98 67.3098 Q1900.94 60.8689 1900.94 49.9314 Q1900.94 38.832 1907.02 32.472 Q1913.14 26.1121 1923.75 26.1121 Q1927.19 26.1121 1930.47 26.8413 Q1933.76 27.5299 1936.83 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1957.17 14.324 L1957.17 27.2059 L1972.52 27.2059 L1972.52 32.9987 L1957.17 32.9987 L1957.17 57.6282 Q1957.17 63.1779 1958.67 64.7578 Q1960.21 66.3376 1964.87 66.3376 L1972.52 66.3376 L1972.52 72.576 L1964.87 72.576 Q1956.24 72.576 1952.96 69.3758 Q1949.68 66.1351 1949.68 57.6282 L1949.68 32.9987 L1944.21 32.9987 L1944.21 27.2059 L1949.68 27.2059 L1949.68 14.324 L1957.17 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2021.13 48.0275 L2021.13 51.6733 L1986.86 51.6733 Q1987.35 59.3701 1991.48 63.421 Q1995.65 67.4314 2003.07 67.4314 Q2007.36 67.4314 2011.37 66.3781 Q2015.42 65.3249 2019.39 63.2184 L2019.39 70.267 Q2015.38 71.9684 2011.17 72.8596 Q2006.96 73.7508 2002.62 73.7508 Q1991.76 73.7508 1985.4 67.4314 Q1979.09 61.1119 1979.09 50.3365 Q1979.09 39.1965 1985.08 32.6746 Q1991.12 26.1121 2001.32 26.1121 Q2010.48 26.1121 2015.79 32.0264 Q2021.13 37.9003 2021.13 48.0275 M2013.68 45.84 Q2013.6 39.7232 2010.24 36.0774 Q2006.91 32.4315 2001.41 32.4315 Q1995.17 32.4315 1991.4 35.9558 Q1987.67 39.4801 1987.11 45.8805 L2013.68 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2059.66 34.1734 Q2058.4 33.4443 2056.9 33.1202 Q2055.44 32.7556 2053.66 32.7556 Q2047.34 32.7556 2043.94 36.8875 Q2040.58 40.9789 2040.58 48.6757 L2040.58 72.576 L2033.08 72.576 L2033.08 27.2059 L2040.58 27.2059 L2040.58 34.2544 Q2042.93 30.1225 2046.69 28.1376 Q2050.46 26.1121 2055.85 26.1121 Q2056.62 26.1121 2057.55 26.2337 Q2058.48 26.3147 2059.62 26.5172 L2059.66 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2096.4 28.5427 L2096.4 35.5912 Q2093.24 33.9709 2089.84 33.1607 Q2086.43 32.3505 2082.79 32.3505 Q2077.24 32.3505 2074.44 34.0519 Q2071.69 35.7533 2071.69 39.156 Q2071.69 41.7486 2073.67 43.2475 Q2075.66 44.7058 2081.65 46.0426 L2084.21 46.6097 Q2092.15 48.3111 2095.47 51.4303 Q2098.83 54.509 2098.83 60.0587 Q2098.83 66.3781 2093.81 70.0644 Q2088.82 73.7508 2080.07 73.7508 Q2076.43 73.7508 2072.46 73.0216 Q2068.53 72.3329 2064.15 70.9151 L2064.15 63.2184 Q2068.29 65.3654 2072.3 66.4591 Q2076.31 67.5124 2080.24 67.5124 Q2085.5 67.5124 2088.34 65.73 Q2091.17 63.9071 2091.17 60.6258 Q2091.17 57.5877 2089.11 55.9673 Q2087.08 54.3469 2080.16 52.8481 L2077.56 52.2405 Q2070.64 50.7821 2067.56 47.7845 Q2064.48 44.7463 2064.48 39.4801 Q2064.48 33.0797 2069.02 29.5959 Q2073.55 26.1121 2081.9 26.1121 Q2086.03 26.1121 2089.67 26.7198 Q2093.32 27.3274 2096.4 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip432)" cx="1205.56" cy="608.696" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="641.076" cy="708.341" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1104.24" cy="678.014" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="698.971" cy="773.326" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1270.69" cy="587.035" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1104.24" cy="634.69" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1147.66" cy="643.355" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1075.29" cy="673.682" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="452.916" cy="1007.27" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1075.29" cy="699.676" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="235.808" cy="1137.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1147.66" cy="665.017" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1234.5" cy="587.035" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1610.83" cy="435.402" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1133.19" cy="652.02" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1118.71" cy="630.358" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="1133.19" cy="630.358" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="2291.1" cy="487.39" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip432)" cx="713.445" cy="989.944" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="423.968,1423.18 423.968,123.472 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="858.184,1423.18 858.184,123.472 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1292.4,1423.18 1292.4,123.472 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1726.62,1423.18 1726.62,123.472 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2160.83,1423.18 2160.83,123.472 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="423.968,1423.18 423.968,1404.28 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="858.184,1423.18 858.184,1404.28 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1292.4,1423.18 1292.4,1404.28 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1726.62,1423.18 1726.62,1404.28 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.83,1423.18 2160.83,1404.28 "/>
+<path clip-path="url(#clip250)" d="M412.811 1466.95 Q416.167 1467.66 418.042 1469.93 Q419.94 1472.2 419.94 1475.53 Q419.94 1480.65 416.422 1483.45 Q412.903 1486.25 406.422 1486.25 Q404.246 1486.25 401.931 1485.81 Q399.639 1485.39 397.186 1484.54 L397.186 1480.02 Q399.13 1481.16 401.445 1481.74 Q403.76 1482.32 406.283 1482.32 Q410.681 1482.32 412.973 1480.58 Q415.287 1478.84 415.287 1475.53 Q415.287 1472.48 413.135 1470.77 Q411.005 1469.03 407.186 1469.03 L403.158 1469.03 L403.158 1465.19 L407.371 1465.19 Q410.82 1465.19 412.649 1463.82 Q414.477 1462.43 414.477 1459.84 Q414.477 1457.18 412.579 1455.77 Q410.704 1454.33 407.186 1454.33 Q405.264 1454.33 403.065 1454.75 Q400.866 1455.16 398.227 1456.04 L398.227 1451.88 Q400.889 1451.14 403.204 1450.77 Q405.542 1450.39 407.602 1450.39 Q412.926 1450.39 416.028 1452.83 Q419.13 1455.23 419.13 1459.35 Q419.13 1462.22 417.486 1464.21 Q415.843 1466.18 412.811 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M438.806 1454.1 Q435.195 1454.1 433.366 1457.66 Q431.56 1461.2 431.56 1468.33 Q431.56 1475.44 433.366 1479.01 Q435.195 1482.55 438.806 1482.55 Q442.44 1482.55 444.246 1479.01 Q446.074 1475.44 446.074 1468.33 Q446.074 1461.2 444.246 1457.66 Q442.44 1454.1 438.806 1454.1 M438.806 1450.39 Q444.616 1450.39 447.671 1455 Q450.75 1459.58 450.75 1468.33 Q450.75 1477.06 447.671 1481.67 Q444.616 1486.25 438.806 1486.25 Q432.996 1486.25 429.917 1481.67 Q426.861 1477.06 426.861 1468.33 Q426.861 1459.58 429.917 1455 Q432.996 1450.39 438.806 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M843.589 1466.44 Q840.441 1466.44 838.589 1468.59 Q836.761 1470.74 836.761 1474.49 Q836.761 1478.22 838.589 1480.39 Q840.441 1482.55 843.589 1482.55 Q846.737 1482.55 848.566 1480.39 Q850.418 1478.22 850.418 1474.49 Q850.418 1470.74 848.566 1468.59 Q846.737 1466.44 843.589 1466.44 M852.872 1451.78 L852.872 1456.04 Q851.112 1455.21 849.307 1454.77 Q847.524 1454.33 845.765 1454.33 Q841.136 1454.33 838.682 1457.45 Q836.251 1460.58 835.904 1466.9 Q837.27 1464.89 839.33 1463.82 Q841.39 1462.73 843.867 1462.73 Q849.075 1462.73 852.085 1465.9 Q855.117 1469.05 855.117 1474.49 Q855.117 1479.82 851.969 1483.03 Q848.821 1486.25 843.589 1486.25 Q837.594 1486.25 834.423 1481.67 Q831.251 1477.06 831.251 1468.33 Q831.251 1460.14 835.14 1455.28 Q839.029 1450.39 845.58 1450.39 Q847.339 1450.39 849.122 1450.74 Q850.927 1451.09 852.872 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M873.172 1454.1 Q869.561 1454.1 867.733 1457.66 Q865.927 1461.2 865.927 1468.33 Q865.927 1475.44 867.733 1479.01 Q869.561 1482.55 873.172 1482.55 Q876.807 1482.55 878.612 1479.01 Q880.441 1475.44 880.441 1468.33 Q880.441 1461.2 878.612 1457.66 Q876.807 1454.1 873.172 1454.1 M873.172 1450.39 Q878.983 1450.39 882.038 1455 Q885.117 1459.58 885.117 1468.33 Q885.117 1477.06 882.038 1481.67 Q878.983 1486.25 873.172 1486.25 Q867.362 1486.25 864.284 1481.67 Q861.228 1477.06 861.228 1468.33 Q861.228 1459.58 864.284 1455 Q867.362 1450.39 873.172 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1267.53 1484.86 L1267.53 1480.6 Q1269.29 1481.44 1271.09 1481.88 Q1272.9 1482.32 1274.63 1482.32 Q1279.26 1482.32 1281.69 1479.21 Q1284.15 1476.09 1284.5 1469.75 Q1283.15 1471.74 1281.09 1472.8 Q1279.03 1473.87 1276.53 1473.87 Q1271.35 1473.87 1268.31 1470.74 Q1265.31 1467.59 1265.31 1462.15 Q1265.31 1456.83 1268.45 1453.61 Q1271.6 1450.39 1276.83 1450.39 Q1282.83 1450.39 1285.98 1455 Q1289.15 1459.58 1289.15 1468.33 Q1289.15 1476.51 1285.26 1481.39 Q1281.39 1486.25 1274.84 1486.25 Q1273.08 1486.25 1271.28 1485.9 Q1269.47 1485.56 1267.53 1484.86 M1276.83 1470.21 Q1279.98 1470.21 1281.81 1468.06 Q1283.66 1465.9 1283.66 1462.15 Q1283.66 1458.43 1281.81 1456.27 Q1279.98 1454.1 1276.83 1454.1 Q1273.68 1454.1 1271.83 1456.27 Q1270 1458.43 1270 1462.15 Q1270 1465.9 1271.83 1468.06 Q1273.68 1470.21 1276.83 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1307.55 1454.1 Q1303.94 1454.1 1302.11 1457.66 Q1300.31 1461.2 1300.31 1468.33 Q1300.31 1475.44 1302.11 1479.01 Q1303.94 1482.55 1307.55 1482.55 Q1311.18 1482.55 1312.99 1479.01 Q1314.82 1475.44 1314.82 1468.33 Q1314.82 1461.2 1312.99 1457.66 Q1311.18 1454.1 1307.55 1454.1 M1307.55 1450.39 Q1313.36 1450.39 1316.42 1455 Q1319.49 1459.58 1319.49 1468.33 Q1319.49 1477.06 1316.42 1481.67 Q1313.36 1486.25 1307.55 1486.25 Q1301.74 1486.25 1298.66 1481.67 Q1295.61 1477.06 1295.61 1468.33 Q1295.61 1459.58 1298.66 1455 Q1301.74 1450.39 1307.55 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1686.22 1481.64 L1693.86 1481.64 L1693.86 1455.28 L1685.55 1456.95 L1685.55 1452.69 L1693.82 1451.02 L1698.49 1451.02 L1698.49 1481.64 L1706.13 1481.64 L1706.13 1485.58 L1686.22 1485.58 L1686.22 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1719.6 1481.64 L1735.92 1481.64 L1735.92 1485.58 L1713.98 1485.58 L1713.98 1481.64 Q1716.64 1478.89 1721.22 1474.26 Q1725.83 1469.61 1727.01 1468.27 Q1729.26 1465.74 1730.13 1464.01 Q1731.04 1462.25 1731.04 1460.56 Q1731.04 1457.8 1729.09 1456.07 Q1727.17 1454.33 1724.07 1454.33 Q1721.87 1454.33 1719.42 1455.09 Q1716.99 1455.86 1714.21 1457.41 L1714.21 1452.69 Q1717.03 1451.55 1719.49 1450.97 Q1721.94 1450.39 1723.98 1450.39 Q1729.35 1450.39 1732.54 1453.08 Q1735.74 1455.77 1735.74 1460.26 Q1735.74 1462.39 1734.93 1464.31 Q1734.14 1466.2 1732.03 1468.8 Q1731.45 1469.47 1728.35 1472.69 Q1725.25 1475.88 1719.6 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1755.74 1454.1 Q1752.13 1454.1 1750.3 1457.66 Q1748.49 1461.2 1748.49 1468.33 Q1748.49 1475.44 1750.3 1479.01 Q1752.13 1482.55 1755.74 1482.55 Q1759.37 1482.55 1761.18 1479.01 Q1763 1475.44 1763 1468.33 Q1763 1461.2 1761.18 1457.66 Q1759.37 1454.1 1755.74 1454.1 M1755.74 1450.39 Q1761.55 1450.39 1764.6 1455 Q1767.68 1459.58 1767.68 1468.33 Q1767.68 1477.06 1764.6 1481.67 Q1761.55 1486.25 1755.74 1486.25 Q1749.93 1486.25 1746.85 1481.67 Q1743.79 1477.06 1743.79 1468.33 Q1743.79 1459.58 1746.85 1455 Q1749.93 1450.39 1755.74 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2120.44 1481.64 L2128.08 1481.64 L2128.08 1455.28 L2119.77 1456.95 L2119.77 1452.69 L2128.03 1451.02 L2132.71 1451.02 L2132.71 1481.64 L2140.35 1481.64 L2140.35 1485.58 L2120.44 1485.58 L2120.44 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2149.84 1451.02 L2168.19 1451.02 L2168.19 1454.96 L2154.12 1454.96 L2154.12 1463.43 Q2155.14 1463.08 2156.16 1462.92 Q2157.17 1462.73 2158.19 1462.73 Q2163.98 1462.73 2167.36 1465.9 Q2170.74 1469.08 2170.74 1474.49 Q2170.74 1480.07 2167.27 1483.17 Q2163.8 1486.25 2157.48 1486.25 Q2155.3 1486.25 2153.03 1485.88 Q2150.79 1485.51 2148.38 1484.77 L2148.38 1480.07 Q2150.46 1481.2 2152.68 1481.76 Q2154.91 1482.32 2157.38 1482.32 Q2161.39 1482.32 2163.73 1480.21 Q2166.06 1478.1 2166.06 1474.49 Q2166.06 1470.88 2163.73 1468.77 Q2161.39 1466.67 2157.38 1466.67 Q2155.51 1466.67 2153.63 1467.08 Q2151.78 1467.5 2149.84 1468.38 L2149.84 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2189.95 1454.1 Q2186.34 1454.1 2184.51 1457.66 Q2182.71 1461.2 2182.71 1468.33 Q2182.71 1475.44 2184.51 1479.01 Q2186.34 1482.55 2189.95 1482.55 Q2193.59 1482.55 2195.39 1479.01 Q2197.22 1475.44 2197.22 1468.33 Q2197.22 1461.2 2195.39 1457.66 Q2193.59 1454.1 2189.95 1454.1 M2189.95 1450.39 Q2195.76 1450.39 2198.82 1455 Q2201.9 1459.58 2201.9 1468.33 Q2201.9 1477.06 2198.82 1481.67 Q2195.76 1486.25 2189.95 1486.25 Q2184.14 1486.25 2181.06 1481.67 Q2178.01 1477.06 2178.01 1468.33 Q2178.01 1459.58 2181.06 1455 Q2184.14 1450.39 2189.95 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1151.77 1520.52 L1158.26 1520.52 L1168.25 1560.69 L1178.22 1520.52 L1185.44 1520.52 L1195.43 1560.69 L1205.4 1520.52 L1211.92 1520.52 L1199.99 1568.04 L1191.9 1568.04 L1181.88 1526.79 L1171.75 1568.04 L1163.67 1568.04 L1151.77 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1246.87 1548.76 L1246.87 1551.62 L1219.94 1551.62 Q1220.32 1557.67 1223.57 1560.85 Q1226.85 1564 1232.67 1564 Q1236.05 1564 1239.2 1563.17 Q1242.38 1562.35 1245.5 1560.69 L1245.5 1566.23 Q1242.35 1567.57 1239.04 1568.27 Q1235.73 1568.97 1232.32 1568.97 Q1223.79 1568.97 1218.8 1564 Q1213.83 1559.04 1213.83 1550.57 Q1213.83 1541.82 1218.54 1536.69 Q1223.28 1531.54 1231.31 1531.54 Q1238.5 1531.54 1242.67 1536.18 Q1246.87 1540.8 1246.87 1548.76 M1241.01 1547.04 Q1240.95 1542.23 1238.31 1539.37 Q1235.7 1536.5 1231.37 1536.5 Q1226.47 1536.5 1223.51 1539.27 Q1220.58 1542.04 1220.13 1547.07 L1241.01 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1256.48 1532.4 L1262.34 1532.4 L1262.34 1568.04 L1256.48 1568.04 L1256.48 1532.4 M1256.48 1518.52 L1262.34 1518.52 L1262.34 1525.93 L1256.48 1525.93 L1256.48 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1298.05 1549.81 Q1298.05 1543.44 1295.41 1539.94 Q1292.8 1536.44 1288.06 1536.44 Q1283.35 1536.44 1280.7 1539.94 Q1278.09 1543.44 1278.09 1549.81 Q1278.09 1556.14 1280.7 1559.64 Q1283.35 1563.14 1288.06 1563.14 Q1292.8 1563.14 1295.41 1559.64 Q1298.05 1556.14 1298.05 1549.81 M1303.91 1563.62 Q1303.91 1572.72 1299.86 1577.15 Q1295.82 1581.6 1287.48 1581.6 Q1284.4 1581.6 1281.66 1581.13 Q1278.92 1580.68 1276.34 1579.72 L1276.34 1574.03 Q1278.92 1575.43 1281.44 1576.1 Q1283.95 1576.76 1286.56 1576.76 Q1292.32 1576.76 1295.19 1573.74 Q1298.05 1570.75 1298.05 1564.67 L1298.05 1561.77 Q1296.24 1564.92 1293.4 1566.48 Q1290.57 1568.04 1286.62 1568.04 Q1280.07 1568.04 1276.06 1563.05 Q1272.05 1558.05 1272.05 1549.81 Q1272.05 1541.53 1276.06 1536.53 Q1280.07 1531.54 1286.62 1531.54 Q1290.57 1531.54 1293.4 1533.1 Q1296.24 1534.66 1298.05 1537.81 L1298.05 1532.4 L1303.91 1532.4 L1303.91 1563.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1345.6 1546.53 L1345.6 1568.04 L1339.75 1568.04 L1339.75 1546.72 Q1339.75 1541.66 1337.77 1539.14 Q1335.8 1536.63 1331.85 1536.63 Q1327.11 1536.63 1324.37 1539.65 Q1321.63 1542.68 1321.63 1547.9 L1321.63 1568.04 L1315.75 1568.04 L1315.75 1518.52 L1321.63 1518.52 L1321.63 1537.93 Q1323.74 1534.72 1326.57 1533.13 Q1329.43 1531.54 1333.16 1531.54 Q1339.3 1531.54 1342.45 1535.36 Q1345.6 1539.14 1345.6 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1363.08 1522.27 L1363.08 1532.4 L1375.14 1532.4 L1375.14 1536.95 L1363.08 1536.95 L1363.08 1556.3 Q1363.08 1560.66 1364.25 1561.9 Q1365.46 1563.14 1369.12 1563.14 L1375.14 1563.14 L1375.14 1568.04 L1369.12 1568.04 Q1362.34 1568.04 1359.77 1565.53 Q1357.19 1562.98 1357.19 1556.3 L1357.19 1536.95 L1352.89 1536.95 L1352.89 1532.4 L1357.19 1532.4 L1357.19 1522.27 L1363.08 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,989.944 2352.76,989.944 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,556.708 2352.76,556.708 "/>
+<polyline clip-path="url(#clip252)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="174.149,123.472 2352.76,123.472 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 174.149,123.472 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,1423.18 193.047,1423.18 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,989.944 193.047,989.944 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,556.708 193.047,556.708 "/>
+<polyline clip-path="url(#clip250)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="174.149,123.472 193.047,123.472 "/>
+<path clip-path="url(#clip250)" d="M126.205 1408.98 Q122.593 1408.98 120.765 1412.54 Q118.959 1416.08 118.959 1423.21 Q118.959 1430.32 120.765 1433.89 Q122.593 1437.43 126.205 1437.43 Q129.839 1437.43 131.644 1433.89 Q133.473 1430.32 133.473 1423.21 Q133.473 1416.08 131.644 1412.54 Q129.839 1408.98 126.205 1408.98 M126.205 1405.27 Q132.015 1405.27 135.07 1409.88 Q138.149 1414.46 138.149 1423.21 Q138.149 1431.94 135.07 1436.55 Q132.015 1441.13 126.205 1441.13 Q120.394 1441.13 117.316 1436.55 Q114.26 1431.94 114.26 1423.21 Q114.26 1414.46 117.316 1409.88 Q120.394 1405.27 126.205 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M118.242 1003.29 L125.88 1003.29 L125.88 976.923 L117.57 978.59 L117.57 974.331 L125.834 972.664 L130.51 972.664 L130.51 1003.29 L138.149 1003.29 L138.149 1007.22 L118.242 1007.22 L118.242 1003.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M121.83 570.053 L138.149 570.053 L138.149 573.988 L116.205 573.988 L116.205 570.053 Q118.867 567.298 123.45 562.669 Q128.056 558.016 129.237 556.673 Q131.482 554.15 132.362 552.414 Q133.265 550.655 133.265 548.965 Q133.265 546.21 131.32 544.474 Q129.399 542.738 126.297 542.738 Q124.098 542.738 121.644 543.502 Q119.214 544.266 116.436 545.817 L116.436 541.095 Q119.26 539.96 121.714 539.382 Q124.168 538.803 126.205 538.803 Q131.575 538.803 134.769 541.488 Q137.964 544.173 137.964 548.664 Q137.964 550.794 137.154 552.715 Q136.367 554.613 134.26 557.206 Q133.681 557.877 130.58 561.095 Q127.478 564.289 121.83 570.053 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M131.019 122.118 Q134.376 122.836 136.251 125.104 Q138.149 127.373 138.149 130.706 Q138.149 135.822 134.63 138.623 Q131.112 141.423 124.63 141.423 Q122.455 141.423 120.14 140.984 Q117.848 140.567 115.394 139.71 L115.394 135.197 Q117.339 136.331 119.654 136.91 Q121.968 137.488 124.492 137.488 Q128.89 137.488 131.181 135.752 Q133.496 134.016 133.496 130.706 Q133.496 127.65 131.343 125.937 Q129.214 124.201 125.394 124.201 L121.367 124.201 L121.367 120.359 L125.58 120.359 Q129.029 120.359 130.857 118.993 Q132.686 117.604 132.686 115.012 Q132.686 112.35 130.788 110.937 Q128.913 109.502 125.394 109.502 Q123.473 109.502 121.274 109.919 Q119.075 110.336 116.436 111.215 L116.436 107.049 Q119.098 106.308 121.413 105.937 Q123.751 105.567 125.811 105.567 Q131.135 105.567 134.237 107.998 Q137.339 110.405 137.339 114.525 Q137.339 117.396 135.695 119.386 Q134.052 121.354 131.019 122.118 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M16.4842 877.087 L16.4842 870.658 L35.9632 870.658 L35.9632 847.295 L16.4842 847.295 L16.4842 840.866 L64.0042 840.866 L64.0042 847.295 L41.3741 847.295 L41.3741 870.658 L64.0042 870.658 L64.0042 877.087 L16.4842 877.087 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M44.7161 797.834 L47.5806 797.834 L47.5806 824.761 Q53.6281 824.379 56.8109 821.132 Q59.9619 817.854 59.9619 812.029 Q59.9619 808.656 59.1344 805.505 Q58.3069 802.322 56.6518 799.203 L62.1899 799.203 Q63.5267 802.354 64.227 805.664 Q64.9272 808.974 64.9272 812.38 Q64.9272 820.91 59.9619 825.907 Q54.9967 830.872 46.5303 830.872 Q37.7774 830.872 32.6531 826.161 Q27.4968 821.419 27.4968 813.398 Q27.4968 806.205 32.1438 802.035 Q36.7589 797.834 44.7161 797.834 M42.9973 803.69 Q38.1912 803.754 35.3266 806.396 Q32.4621 809.006 32.4621 813.334 Q32.4621 818.236 35.2312 821.196 Q38.0002 824.124 43.0292 824.57 L42.9973 803.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M28.3562 788.222 L28.3562 782.365 L64.0042 782.365 L64.0042 788.222 L28.3562 788.222 M14.479 788.222 L14.479 782.365 L21.895 782.365 L21.895 788.222 L14.479 788.222 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M45.7664 746.654 Q39.4007 746.654 35.8996 749.295 Q32.3984 751.905 32.3984 756.648 Q32.3984 761.358 35.8996 764 Q39.4007 766.61 45.7664 766.61 Q52.1003 766.61 55.6014 764 Q59.1026 761.358 59.1026 756.648 Q59.1026 751.905 55.6014 749.295 Q52.1003 746.654 45.7664 746.654 M59.58 740.797 Q68.683 740.797 73.1071 744.839 Q77.5631 748.882 77.5631 757.221 Q77.5631 760.308 77.0857 763.045 Q76.6401 765.783 75.6852 768.361 L69.9879 768.361 Q71.3884 765.783 72.0568 763.268 Q72.7252 760.754 72.7252 758.144 Q72.7252 752.383 69.7015 749.518 Q66.7096 746.654 60.6303 746.654 L57.7339 746.654 Q60.885 748.468 62.4446 751.301 Q64.0042 754.133 64.0042 758.08 Q64.0042 764.637 59.0071 768.647 Q54.01 772.658 45.7664 772.658 Q37.491 772.658 32.4939 768.647 Q27.4968 764.637 27.4968 758.08 Q27.4968 754.133 29.0564 751.301 Q30.616 748.468 33.7671 746.654 L28.3562 746.654 L28.3562 740.797 L59.58 740.797 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M42.4881 699.102 L64.0042 699.102 L64.0042 704.958 L42.679 704.958 Q37.6183 704.958 35.1038 706.932 Q32.5894 708.905 32.5894 712.852 Q32.5894 717.594 35.6131 720.331 Q38.6368 723.069 43.8567 723.069 L64.0042 723.069 L64.0042 728.957 L14.479 728.957 L14.479 723.069 L33.8944 723.069 Q30.6797 720.968 29.0883 718.135 Q27.4968 715.271 27.4968 711.547 Q27.4968 705.404 31.3163 702.253 Q35.1038 699.102 42.4881 699.102 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M18.2347 681.628 L28.3562 681.628 L28.3562 669.565 L32.9077 669.565 L32.9077 681.628 L52.2594 681.628 Q56.6199 681.628 57.8613 680.45 Q59.1026 679.241 59.1026 675.581 L59.1026 669.565 L64.0042 669.565 L64.0042 675.581 Q64.0042 682.36 61.4897 684.938 Q58.9434 687.516 52.2594 687.516 L32.9077 687.516 L32.9077 691.813 L28.3562 691.813 L28.3562 687.516 L18.2347 687.516 L18.2347 681.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M428.075 12.096 L436.258 12.096 L436.258 36.8875 L465.991 36.8875 L465.991 12.096 L474.174 12.096 L474.174 72.576 L465.991 72.576 L465.991 43.7741 L436.258 43.7741 L436.258 72.576 L428.075 72.576 L428.075 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M528.942 48.0275 L528.942 51.6733 L494.672 51.6733 Q495.158 59.3701 499.29 63.421 Q503.462 67.4314 510.875 67.4314 Q515.169 67.4314 519.18 66.3781 Q523.231 65.3249 527.201 63.2184 L527.201 70.267 Q523.19 71.9684 518.977 72.8596 Q514.764 73.7508 510.43 73.7508 Q499.573 73.7508 493.213 67.4314 Q486.894 61.1119 486.894 50.3365 Q486.894 39.1965 492.889 32.6746 Q498.925 26.1121 509.134 26.1121 Q518.289 26.1121 523.595 32.0264 Q528.942 37.9003 528.942 48.0275 M521.489 45.84 Q521.408 39.7232 518.046 36.0774 Q514.724 32.4315 509.215 32.4315 Q502.976 32.4315 499.209 35.9558 Q495.482 39.4801 494.915 45.8805 L521.489 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M541.176 27.2059 L548.63 27.2059 L548.63 72.576 L541.176 72.576 L541.176 27.2059 M541.176 9.54393 L548.63 9.54393 L548.63 18.9825 L541.176 18.9825 L541.176 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M594.081 49.3643 Q594.081 41.2625 590.719 36.8065 Q587.397 32.3505 581.361 32.3505 Q575.366 32.3505 572.004 36.8065 Q568.682 41.2625 568.682 49.3643 Q568.682 57.4256 572.004 61.8816 Q575.366 66.3376 581.361 66.3376 Q587.397 66.3376 590.719 61.8816 Q594.081 57.4256 594.081 49.3643 M601.535 66.9452 Q601.535 78.5308 596.39 84.1616 Q591.245 89.8329 580.632 89.8329 Q576.703 89.8329 573.219 89.2252 Q569.735 88.6581 566.454 87.4428 L566.454 80.1917 Q569.735 81.9741 572.935 82.8248 Q576.135 83.6755 579.457 83.6755 Q586.789 83.6755 590.435 79.8271 Q594.081 76.0193 594.081 68.282 L594.081 64.5957 Q591.772 68.6061 588.167 70.5911 Q584.561 72.576 579.538 72.576 Q571.193 72.576 566.089 66.2161 Q560.985 59.8562 560.985 49.3643 Q560.985 38.832 566.089 32.472 Q571.193 26.1121 579.538 26.1121 Q584.561 26.1121 588.167 28.0971 Q591.772 30.082 594.081 34.0924 L594.081 27.2059 L601.535 27.2059 L601.535 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M654.602 45.1919 L654.602 72.576 L647.148 72.576 L647.148 45.4349 Q647.148 38.994 644.636 35.7938 Q642.125 32.5936 637.102 32.5936 Q631.066 32.5936 627.582 36.4419 Q624.098 40.2903 624.098 46.9338 L624.098 72.576 L616.604 72.576 L616.604 9.54393 L624.098 9.54393 L624.098 34.2544 Q626.772 30.163 630.377 28.1376 Q634.023 26.1121 638.762 26.1121 Q646.581 26.1121 650.591 30.9732 Q654.602 35.7938 654.602 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M676.841 14.324 L676.841 27.2059 L692.194 27.2059 L692.194 32.9987 L676.841 32.9987 L676.841 57.6282 Q676.841 63.1779 678.34 64.7578 Q679.879 66.3376 684.538 66.3376 L692.194 66.3376 L692.194 72.576 L684.538 72.576 Q675.909 72.576 672.628 69.3758 Q669.347 66.1351 669.347 57.6282 L669.347 32.9987 L663.878 32.9987 L663.878 27.2059 L669.347 27.2059 L669.347 14.324 L676.841 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M723.021 27.2059 L730.921 27.2059 L745.099 65.2844 L759.277 27.2059 L767.176 27.2059 L750.162 72.576 L740.035 72.576 L723.021 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M806.389 28.5427 L806.389 35.5912 Q803.229 33.9709 799.826 33.1607 Q796.424 32.3505 792.778 32.3505 Q787.228 32.3505 784.433 34.0519 Q781.678 35.7533 781.678 39.156 Q781.678 41.7486 783.663 43.2475 Q785.648 44.7058 791.644 46.0426 L794.196 46.6097 Q802.135 48.3111 805.457 51.4303 Q808.819 54.509 808.819 60.0587 Q808.819 66.3781 803.796 70.0644 Q798.814 73.7508 790.064 73.7508 Q786.418 73.7508 782.448 73.0216 Q778.519 72.3329 774.144 70.9151 L774.144 63.2184 Q778.276 65.3654 782.286 66.4591 Q786.296 67.5124 790.226 67.5124 Q795.492 67.5124 798.328 65.73 Q801.163 63.9071 801.163 60.6258 Q801.163 57.5877 799.097 55.9673 Q797.072 54.3469 790.145 52.8481 L787.552 52.2405 Q780.625 50.7821 777.546 47.7845 Q774.468 44.7463 774.468 39.4801 Q774.468 33.0797 779.005 29.5959 Q783.542 26.1121 791.887 26.1121 Q796.019 26.1121 799.664 26.7198 Q803.31 27.3274 806.389 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M841.996 12.096 L850.26 12.096 L862.98 63.2184 L875.659 12.096 L884.855 12.096 L897.575 63.2184 L910.254 12.096 L918.558 12.096 L903.368 72.576 L893.078 72.576 L880.318 20.0763 L867.436 72.576 L857.147 72.576 L841.996 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M963.037 48.0275 L963.037 51.6733 L928.767 51.6733 Q929.253 59.3701 933.385 63.421 Q937.557 67.4314 944.97 67.4314 Q949.264 67.4314 953.275 66.3781 Q957.326 65.3249 961.295 63.2184 L961.295 70.267 Q957.285 71.9684 953.072 72.8596 Q948.859 73.7508 944.525 73.7508 Q933.668 73.7508 927.308 67.4314 Q920.989 61.1119 920.989 50.3365 Q920.989 39.1965 926.984 32.6746 Q933.02 26.1121 943.228 26.1121 Q952.383 26.1121 957.69 32.0264 Q963.037 37.9003 963.037 48.0275 M955.584 45.84 Q955.503 39.7232 952.14 36.0774 Q948.819 32.4315 943.309 32.4315 Q937.071 32.4315 933.304 35.9558 Q929.577 39.4801 929.01 45.8805 L955.584 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M975.271 27.2059 L982.725 27.2059 L982.725 72.576 L975.271 72.576 L975.271 27.2059 M975.271 9.54393 L982.725 9.54393 L982.725 18.9825 L975.271 18.9825 L975.271 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1028.18 49.3643 Q1028.18 41.2625 1024.81 36.8065 Q1021.49 32.3505 1015.46 32.3505 Q1009.46 32.3505 1006.1 36.8065 Q1002.78 41.2625 1002.78 49.3643 Q1002.78 57.4256 1006.1 61.8816 Q1009.46 66.3376 1015.46 66.3376 Q1021.49 66.3376 1024.81 61.8816 Q1028.18 57.4256 1028.18 49.3643 M1035.63 66.9452 Q1035.63 78.5308 1030.48 84.1616 Q1025.34 89.8329 1014.73 89.8329 Q1010.8 89.8329 1007.31 89.2252 Q1003.83 88.6581 1000.55 87.4428 L1000.55 80.1917 Q1003.83 81.9741 1007.03 82.8248 Q1010.23 83.6755 1013.55 83.6755 Q1020.88 83.6755 1024.53 79.8271 Q1028.18 76.0193 1028.18 68.282 L1028.18 64.5957 Q1025.87 68.6061 1022.26 70.5911 Q1018.66 72.576 1013.63 72.576 Q1005.29 72.576 1000.18 66.2161 Q995.08 59.8562 995.08 49.3643 Q995.08 38.832 1000.18 32.472 Q1005.29 26.1121 1013.63 26.1121 Q1018.66 26.1121 1022.26 28.0971 Q1025.87 30.082 1028.18 34.0924 L1028.18 27.2059 L1035.63 27.2059 L1035.63 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1088.7 45.1919 L1088.7 72.576 L1081.24 72.576 L1081.24 45.4349 Q1081.24 38.994 1078.73 35.7938 Q1076.22 32.5936 1071.2 32.5936 Q1065.16 32.5936 1061.68 36.4419 Q1058.19 40.2903 1058.19 46.9338 L1058.19 72.576 L1050.7 72.576 L1050.7 9.54393 L1058.19 9.54393 L1058.19 34.2544 Q1060.87 30.163 1064.47 28.1376 Q1068.12 26.1121 1072.86 26.1121 Q1080.68 26.1121 1084.69 30.9732 Q1088.7 35.7938 1088.7 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1110.94 14.324 L1110.94 27.2059 L1126.29 27.2059 L1126.29 32.9987 L1110.94 32.9987 L1110.94 57.6282 Q1110.94 63.1779 1112.43 64.7578 Q1113.97 66.3376 1118.63 66.3376 L1126.29 66.3376 L1126.29 72.576 L1118.63 72.576 Q1110 72.576 1106.72 69.3758 Q1103.44 66.1351 1103.44 57.6282 L1103.44 32.9987 L1097.97 32.9987 L1097.97 27.2059 L1103.44 27.2059 L1103.44 14.324 L1110.94 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1180.04 32.4315 Q1174.05 32.4315 1170.57 37.1306 Q1167.08 41.7891 1167.08 49.9314 Q1167.08 58.0738 1170.52 62.7728 Q1174.01 67.4314 1180.04 67.4314 Q1186 67.4314 1189.48 62.7323 Q1192.97 58.0333 1192.97 49.9314 Q1192.97 41.8701 1189.48 37.1711 Q1186 32.4315 1180.04 32.4315 M1180.04 26.1121 Q1189.77 26.1121 1195.32 32.4315 Q1200.87 38.7509 1200.87 49.9314 Q1200.87 61.0714 1195.32 67.4314 Q1189.77 73.7508 1180.04 73.7508 Q1170.28 73.7508 1164.73 67.4314 Q1159.22 61.0714 1159.22 49.9314 Q1159.22 38.7509 1164.73 32.4315 Q1170.28 26.1121 1180.04 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1236.19 9.54393 L1236.19 15.7418 L1229.06 15.7418 Q1225.05 15.7418 1223.47 17.3622 Q1221.93 18.9825 1221.93 23.1955 L1221.93 27.2059 L1234.2 27.2059 L1234.2 32.9987 L1221.93 32.9987 L1221.93 72.576 L1214.44 72.576 L1214.44 32.9987 L1207.31 32.9987 L1207.31 27.2059 L1214.44 27.2059 L1214.44 24.0462 Q1214.44 16.471 1217.96 13.0277 Q1221.49 9.54393 1229.14 9.54393 L1236.19 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1305.38 14.0809 L1305.38 22.0612 Q1300.72 19.8332 1296.59 18.7395 Q1292.46 17.6457 1288.61 17.6457 Q1281.92 17.6457 1278.28 20.2383 Q1274.67 22.8309 1274.67 27.611 Q1274.67 31.6214 1277.06 33.6873 Q1279.49 35.7128 1286.22 36.9686 L1291.16 37.9813 Q1300.32 39.7232 1304.65 44.1387 Q1309.03 48.5136 1309.03 55.8863 Q1309.03 64.6767 1303.11 69.2137 Q1297.24 73.7508 1285.85 73.7508 Q1281.56 73.7508 1276.7 72.7785 Q1271.88 71.8063 1266.69 69.9024 L1266.69 61.4765 Q1271.68 64.2716 1276.46 65.6895 Q1281.24 67.1073 1285.85 67.1073 Q1292.86 67.1073 1296.67 64.3527 Q1300.48 61.598 1300.48 56.4939 Q1300.48 52.0379 1297.72 49.5264 Q1295.01 47.0148 1288.77 45.759 L1283.79 44.7868 Q1274.63 42.9639 1270.54 39.075 Q1266.45 35.1862 1266.45 28.2591 Q1266.45 20.2383 1272.08 15.6203 Q1277.75 11.0023 1287.68 11.0023 Q1291.93 11.0023 1296.35 11.7719 Q1300.76 12.5416 1305.38 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1328.83 14.324 L1328.83 27.2059 L1344.19 27.2059 L1344.19 32.9987 L1328.83 32.9987 L1328.83 57.6282 Q1328.83 63.1779 1330.33 64.7578 Q1331.87 66.3376 1336.53 66.3376 L1344.19 66.3376 L1344.19 72.576 L1336.53 72.576 Q1327.9 72.576 1324.62 69.3758 Q1321.34 66.1351 1321.34 57.6282 L1321.34 32.9987 L1315.87 32.9987 L1315.87 27.2059 L1321.34 27.2059 L1321.34 14.324 L1328.83 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1374.61 49.7694 Q1365.58 49.7694 1362.09 51.8354 Q1358.61 53.9013 1358.61 58.8839 Q1358.61 62.8538 1361.2 65.2034 Q1363.83 67.5124 1368.33 67.5124 Q1374.53 67.5124 1378.26 63.1374 Q1382.02 58.7219 1382.02 51.4303 L1382.02 49.7694 L1374.61 49.7694 M1389.48 46.6907 L1389.48 72.576 L1382.02 72.576 L1382.02 65.6895 Q1379.47 69.8214 1375.66 71.8063 Q1371.85 73.7508 1366.35 73.7508 Q1359.38 73.7508 1355.25 69.8619 Q1351.15 65.9325 1351.15 59.3701 Q1351.15 51.7138 1356.26 47.825 Q1361.4 43.9361 1371.57 43.9361 L1382.02 43.9361 L1382.02 43.2069 Q1382.02 38.0623 1378.62 35.2672 Q1375.26 32.4315 1369.14 32.4315 Q1365.25 32.4315 1361.57 33.3632 Q1357.88 34.295 1354.48 36.1584 L1354.48 29.2718 Q1358.57 27.692 1362.42 26.9223 Q1366.26 26.1121 1369.91 26.1121 Q1379.75 26.1121 1384.61 31.2163 Q1389.48 36.3204 1389.48 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1431.12 34.1734 Q1429.86 33.4443 1428.36 33.1202 Q1426.91 32.7556 1425.12 32.7556 Q1418.8 32.7556 1415.4 36.8875 Q1412.04 40.9789 1412.04 48.6757 L1412.04 72.576 L1404.55 72.576 L1404.55 27.2059 L1412.04 27.2059 L1412.04 34.2544 Q1414.39 30.1225 1418.16 28.1376 Q1421.92 26.1121 1427.31 26.1121 Q1428.08 26.1121 1429.01 26.2337 Q1429.94 26.3147 1431.08 26.5172 L1431.12 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1433.87 12.096 L1442.14 12.096 L1454.86 63.2184 L1467.54 12.096 L1476.73 12.096 L1489.45 63.2184 L1502.13 12.096 L1510.44 12.096 L1495.25 72.576 L1484.96 72.576 L1472.2 20.0763 L1459.31 72.576 L1449.02 72.576 L1433.87 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1536.28 49.7694 Q1527.25 49.7694 1523.76 51.8354 Q1520.28 53.9013 1520.28 58.8839 Q1520.28 62.8538 1522.87 65.2034 Q1525.51 67.5124 1530 67.5124 Q1536.2 67.5124 1539.93 63.1374 Q1543.69 58.7219 1543.69 51.4303 L1543.69 49.7694 L1536.28 49.7694 M1551.15 46.6907 L1551.15 72.576 L1543.69 72.576 L1543.69 65.6895 Q1541.14 69.8214 1537.33 71.8063 Q1533.53 73.7508 1528.02 73.7508 Q1521.05 73.7508 1516.92 69.8619 Q1512.83 65.9325 1512.83 59.3701 Q1512.83 51.7138 1517.93 47.825 Q1523.07 43.9361 1533.24 43.9361 L1543.69 43.9361 L1543.69 43.2069 Q1543.69 38.0623 1540.29 35.2672 Q1536.93 32.4315 1530.81 32.4315 Q1526.92 32.4315 1523.24 33.3632 Q1519.55 34.295 1516.15 36.1584 L1516.15 29.2718 Q1520.24 27.692 1524.09 26.9223 Q1527.94 26.1121 1531.58 26.1121 Q1541.43 26.1121 1546.29 31.2163 Q1551.15 36.3204 1551.15 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1592.79 34.1734 Q1591.54 33.4443 1590.04 33.1202 Q1588.58 32.7556 1586.8 32.7556 Q1580.48 32.7556 1577.07 36.8875 Q1573.71 40.9789 1573.71 48.6757 L1573.71 72.576 L1566.22 72.576 L1566.22 27.2059 L1573.71 27.2059 L1573.71 34.2544 Q1576.06 30.1225 1579.83 28.1376 Q1583.6 26.1121 1588.98 26.1121 Q1589.75 26.1121 1590.68 26.2337 Q1591.62 26.3147 1592.75 26.5172 L1592.79 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1629.53 28.5427 L1629.53 35.5912 Q1626.37 33.9709 1622.97 33.1607 Q1619.57 32.3505 1615.92 32.3505 Q1610.37 32.3505 1607.58 34.0519 Q1604.82 35.7533 1604.82 39.156 Q1604.82 41.7486 1606.81 43.2475 Q1608.79 44.7058 1614.79 46.0426 L1617.34 46.6097 Q1625.28 48.3111 1628.6 51.4303 Q1631.96 54.509 1631.96 60.0587 Q1631.96 66.3781 1626.94 70.0644 Q1621.96 73.7508 1613.21 73.7508 Q1609.56 73.7508 1605.59 73.0216 Q1601.66 72.3329 1597.29 70.9151 L1597.29 63.2184 Q1601.42 65.3654 1605.43 66.4591 Q1609.44 67.5124 1613.37 67.5124 Q1618.64 67.5124 1621.47 65.73 Q1624.31 63.9071 1624.31 60.6258 Q1624.31 57.5877 1622.24 55.9673 Q1620.22 54.3469 1613.29 52.8481 L1610.7 52.2405 Q1603.77 50.7821 1600.69 47.7845 Q1597.61 44.7463 1597.61 39.4801 Q1597.61 33.0797 1602.15 29.5959 Q1606.69 26.1121 1615.03 26.1121 Q1619.16 26.1121 1622.81 26.7198 Q1626.45 27.3274 1629.53 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1702.85 28.9478 L1702.85 35.9153 Q1699.69 34.1734 1696.49 33.3227 Q1693.33 32.4315 1690.09 32.4315 Q1682.84 32.4315 1678.83 37.0496 Q1674.82 41.6271 1674.82 49.9314 Q1674.82 58.2358 1678.83 62.8538 Q1682.84 67.4314 1690.09 67.4314 Q1693.33 67.4314 1696.49 66.5807 Q1699.69 65.6895 1702.85 63.9476 L1702.85 70.8341 Q1699.73 72.2924 1696.37 73.0216 Q1693.05 73.7508 1689.28 73.7508 Q1679.03 73.7508 1673 67.3098 Q1666.96 60.8689 1666.96 49.9314 Q1666.96 38.832 1673.04 32.472 Q1679.16 26.1121 1689.77 26.1121 Q1693.21 26.1121 1696.49 26.8413 Q1699.78 27.5299 1702.85 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1753.53 45.1919 L1753.53 72.576 L1746.08 72.576 L1746.08 45.4349 Q1746.08 38.994 1743.57 35.7938 Q1741.05 32.5936 1736.03 32.5936 Q1730 32.5936 1726.51 36.4419 Q1723.03 40.2903 1723.03 46.9338 L1723.03 72.576 L1715.53 72.576 L1715.53 9.54393 L1723.03 9.54393 L1723.03 34.2544 Q1725.7 30.163 1729.31 28.1376 Q1732.95 26.1121 1737.69 26.1121 Q1745.51 26.1121 1749.52 30.9732 Q1753.53 35.7938 1753.53 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1789.02 49.7694 Q1779.98 49.7694 1776.5 51.8354 Q1773.02 53.9013 1773.02 58.8839 Q1773.02 62.8538 1775.61 65.2034 Q1778.24 67.5124 1782.74 67.5124 Q1788.94 67.5124 1792.66 63.1374 Q1796.43 58.7219 1796.43 51.4303 L1796.43 49.7694 L1789.02 49.7694 M1803.88 46.6907 L1803.88 72.576 L1796.43 72.576 L1796.43 65.6895 Q1793.88 69.8214 1790.07 71.8063 Q1786.26 73.7508 1780.75 73.7508 Q1773.79 73.7508 1769.65 69.8619 Q1765.56 65.9325 1765.56 59.3701 Q1765.56 51.7138 1770.67 47.825 Q1775.81 43.9361 1785.98 43.9361 L1796.43 43.9361 L1796.43 43.2069 Q1796.43 38.0623 1793.03 35.2672 Q1789.66 32.4315 1783.55 32.4315 Q1779.66 32.4315 1775.97 33.3632 Q1772.29 34.295 1768.88 36.1584 L1768.88 29.2718 Q1772.98 27.692 1776.82 26.9223 Q1780.67 26.1121 1784.32 26.1121 Q1794.16 26.1121 1799.02 31.2163 Q1803.88 36.3204 1803.88 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1845.53 34.1734 Q1844.27 33.4443 1842.77 33.1202 Q1841.31 32.7556 1839.53 32.7556 Q1833.21 32.7556 1829.81 36.8875 Q1826.45 40.9789 1826.45 48.6757 L1826.45 72.576 L1818.95 72.576 L1818.95 27.2059 L1826.45 27.2059 L1826.45 34.2544 Q1828.8 30.1225 1832.56 28.1376 Q1836.33 26.1121 1841.72 26.1121 Q1842.49 26.1121 1843.42 26.2337 Q1844.35 26.3147 1845.49 26.5172 L1845.53 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1873.96 49.7694 Q1864.93 49.7694 1861.45 51.8354 Q1857.96 53.9013 1857.96 58.8839 Q1857.96 62.8538 1860.56 65.2034 Q1863.19 67.5124 1867.69 67.5124 Q1873.88 67.5124 1877.61 63.1374 Q1881.38 58.7219 1881.38 51.4303 L1881.38 49.7694 L1873.96 49.7694 M1888.83 46.6907 L1888.83 72.576 L1881.38 72.576 L1881.38 65.6895 Q1878.83 69.8214 1875.02 71.8063 Q1871.21 73.7508 1865.7 73.7508 Q1858.73 73.7508 1854.6 69.8619 Q1850.51 65.9325 1850.51 59.3701 Q1850.51 51.7138 1855.61 47.825 Q1860.76 43.9361 1870.93 43.9361 L1881.38 43.9361 L1881.38 43.2069 Q1881.38 38.0623 1877.97 35.2672 Q1874.61 32.4315 1868.5 32.4315 Q1864.61 32.4315 1860.92 33.3632 Q1857.23 34.295 1853.83 36.1584 L1853.83 29.2718 Q1857.92 27.692 1861.77 26.9223 Q1865.62 26.1121 1869.27 26.1121 Q1879.11 26.1121 1883.97 31.2163 Q1888.83 36.3204 1888.83 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1936.83 28.9478 L1936.83 35.9153 Q1933.67 34.1734 1930.47 33.3227 Q1927.31 32.4315 1924.07 32.4315 Q1916.82 32.4315 1912.81 37.0496 Q1908.8 41.6271 1908.8 49.9314 Q1908.8 58.2358 1912.81 62.8538 Q1916.82 67.4314 1924.07 67.4314 Q1927.31 67.4314 1930.47 66.5807 Q1933.67 65.6895 1936.83 63.9476 L1936.83 70.8341 Q1933.72 72.2924 1930.35 73.0216 Q1927.03 73.7508 1923.26 73.7508 Q1913.01 73.7508 1906.98 67.3098 Q1900.94 60.8689 1900.94 49.9314 Q1900.94 38.832 1907.02 32.472 Q1913.14 26.1121 1923.75 26.1121 Q1927.19 26.1121 1930.47 26.8413 Q1933.76 27.5299 1936.83 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M1957.17 14.324 L1957.17 27.2059 L1972.52 27.2059 L1972.52 32.9987 L1957.17 32.9987 L1957.17 57.6282 Q1957.17 63.1779 1958.67 64.7578 Q1960.21 66.3376 1964.87 66.3376 L1972.52 66.3376 L1972.52 72.576 L1964.87 72.576 Q1956.24 72.576 1952.96 69.3758 Q1949.68 66.1351 1949.68 57.6282 L1949.68 32.9987 L1944.21 32.9987 L1944.21 27.2059 L1949.68 27.2059 L1949.68 14.324 L1957.17 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2021.13 48.0275 L2021.13 51.6733 L1986.86 51.6733 Q1987.35 59.3701 1991.48 63.421 Q1995.65 67.4314 2003.07 67.4314 Q2007.36 67.4314 2011.37 66.3781 Q2015.42 65.3249 2019.39 63.2184 L2019.39 70.267 Q2015.38 71.9684 2011.17 72.8596 Q2006.96 73.7508 2002.62 73.7508 Q1991.76 73.7508 1985.4 67.4314 Q1979.09 61.1119 1979.09 50.3365 Q1979.09 39.1965 1985.08 32.6746 Q1991.12 26.1121 2001.32 26.1121 Q2010.48 26.1121 2015.79 32.0264 Q2021.13 37.9003 2021.13 48.0275 M2013.68 45.84 Q2013.6 39.7232 2010.24 36.0774 Q2006.91 32.4315 2001.41 32.4315 Q1995.17 32.4315 1991.4 35.9558 Q1987.67 39.4801 1987.11 45.8805 L2013.68 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2059.66 34.1734 Q2058.4 33.4443 2056.9 33.1202 Q2055.44 32.7556 2053.66 32.7556 Q2047.34 32.7556 2043.94 36.8875 Q2040.58 40.9789 2040.58 48.6757 L2040.58 72.576 L2033.08 72.576 L2033.08 27.2059 L2040.58 27.2059 L2040.58 34.2544 Q2042.93 30.1225 2046.69 28.1376 Q2050.46 26.1121 2055.85 26.1121 Q2056.62 26.1121 2057.55 26.2337 Q2058.48 26.3147 2059.62 26.5172 L2059.66 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip250)" d="M2096.4 28.5427 L2096.4 35.5912 Q2093.24 33.9709 2089.84 33.1607 Q2086.43 32.3505 2082.79 32.3505 Q2077.24 32.3505 2074.44 34.0519 Q2071.69 35.7533 2071.69 39.156 Q2071.69 41.7486 2073.67 43.2475 Q2075.66 44.7058 2081.65 46.0426 L2084.21 46.6097 Q2092.15 48.3111 2095.47 51.4303 Q2098.83 54.509 2098.83 60.0587 Q2098.83 66.3781 2093.81 70.0644 Q2088.82 73.7508 2080.07 73.7508 Q2076.43 73.7508 2072.46 73.0216 Q2068.53 72.3329 2064.15 70.9151 L2064.15 63.2184 Q2068.29 65.3654 2072.3 66.4591 Q2076.31 67.5124 2080.24 67.5124 Q2085.5 67.5124 2088.34 65.73 Q2091.17 63.9071 2091.17 60.6258 Q2091.17 57.5877 2089.11 55.9673 Q2087.08 54.3469 2080.16 52.8481 L2077.56 52.2405 Q2070.64 50.7821 2067.56 47.7845 Q2064.48 44.7463 2064.48 39.4801 Q2064.48 33.0797 2069.02 29.5959 Q2073.55 26.1121 2081.9 26.1121 Q2086.03 26.1121 2089.67 26.7198 Q2093.32 27.3274 2096.4 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip252)" cx="1205.56" cy="608.696" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="641.076" cy="708.341" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1104.24" cy="678.014" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="698.971" cy="773.326" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1270.69" cy="587.035" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1104.24" cy="634.69" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1147.66" cy="643.355" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1075.29" cy="673.682" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="452.916" cy="1007.27" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1075.29" cy="699.676" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="235.808" cy="1137.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1147.66" cy="665.017" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1234.5" cy="587.035" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1610.83" cy="435.402" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1133.19" cy="652.02" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1118.71" cy="630.358" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="1133.19" cy="630.358" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="2291.1" cy="487.39" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip252)" cx="713.445" cy="989.944" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
 </svg>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/index.html b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/index.html
index 8851852fa46..9c08d362934 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/index.html
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_data_and_plotting/index.html
@@ -14,7 +14,7 @@
     csv_df.Height;
     xlabel = &quot;Weight&quot;,
     ylabel = &quot;Height&quot;,
-)</code></pre><img src="054004dc.svg" alt="Example block output"/><p>That doesn&#39;t look right. What happened? If you look at the dataframe above, it read <code>Weight</code> in as a <code>String</code> column because there are &quot;NA&quot; fields. Let&#39;s correct that, by telling CSV to consider &quot;NA&quot; as <code>missing</code>.</p><pre><code class="language-julia hljs">csv_df = CSV.read(
+)</code></pre><img src="76abfa9b.svg" alt="Example block output"/><p>That doesn&#39;t look right. What happened? If you look at the dataframe above, it read <code>Weight</code> in as a <code>String</code> column because there are &quot;NA&quot; fields. Let&#39;s correct that, by telling CSV to consider &quot;NA&quot; as <code>missing</code>.</p><pre><code class="language-julia hljs">csv_df = CSV.read(
     joinpath(DATA_DIR, &quot;StarWars.csv&quot;),
     DataFrames.DataFrame;
     missingstring = &quot;NA&quot;,
@@ -26,7 +26,7 @@
     ylabel = &quot;Height&quot;,
     label = false,
     ylims = (0, 3),
-)</code></pre><img src="c55aa536.svg" alt="Example block output"/><p>That looks better.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Read the <a href="https://csv.juliadata.org/stable/">CSV documentation</a> for other parsing options.</p></div></div><p>DataFrames.jl supports manipulation using functions similar to pandas. For example, split the dataframe into groups based on eye-color:</p><pre><code class="language-julia hljs">by_eyecolor = DataFrames.groupby(csv_df, :Eyecolor)</code></pre><p><b>GroupedDataFrame with 7 groups based on key: Eyecolor</b></p><div><div style = "float: left;"><span>First Group (5 rows): Eyecolor = &quot;blue&quot;</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Gender</th><th style = "text-align: left;">Height</th><th style = "text-align: left;">Weight</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">Haircolor</th><th style = "text-align: left;">Skincolor</th><th style = "text-align: left;">Homeland</th><th style = "text-align: left;">Born</th><th style = "text-align: left;">Died</th><th style = "text-align: left;">Jedi</th><th style = "text-align: left;">Species</th><th style = "text-align: left;">Weapon</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Union{Missing, Float64}" style = "text-align: left;">Float64?</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Union{Missing, InlineStrings.String7}" style = "text-align: left;">String7</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">Anakin Skywalker</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.88</td><td style = "text-align: right;">84.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">blond</td><td style = "text-align: left;">fair</td><td style = "text-align: left;">Tatooine</td><td style = "text-align: left;">41.9BBY</td><td style = "text-align: left;">4ABY</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">Luke Skywalker</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.72</td><td style = "text-align: right;">77.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">blond</td><td style = "text-align: left;">fair</td><td style = "text-align: left;">Tatooine</td><td style = "text-align: left;">19BBY</td><td style = "text-align: left;">unk_died</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">Qui-Gon Jinn</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.93</td><td style = "text-align: right;">88.5</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">brown</td><td style = "text-align: left;">light</td><td style = "text-align: left;">unk_planet</td><td style = "text-align: left;">92BBY</td><td style = "text-align: left;">32BBY</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">Sheev Palpatine</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.73</td><td style = "text-align: right;">75.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">red</td><td style = "text-align: left;">pale</td><td style = "text-align: left;">Naboo</td><td style = "text-align: left;">82BBY</td><td style = "text-align: left;">10ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">force-lightning</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">Chewbacca</td><td style = "text-align: left;">male</td><td style = "text-align: right;">2.28</td><td style = "text-align: right;">112.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">brown</td><td style = "font-style: italic; text-align: left;">missing</td><td style = "text-align: left;">Kashyyyk</td><td style = "text-align: left;">200BBY</td><td style = "text-align: left;">25ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">wookiee</td><td style = "text-align: left;">bowcaster</td></tr></tbody></table></div><p>&vellip;</p><div><div style = "float: left;"><span>Last Group (1 row): Eyecolor = &quot;black&quot;</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Gender</th><th style = "text-align: left;">Height</th><th style = "text-align: left;">Weight</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">Haircolor</th><th style = "text-align: left;">Skincolor</th><th style = "text-align: left;">Homeland</th><th style = "text-align: left;">Born</th><th style = "text-align: left;">Died</th><th style = "text-align: left;">Jedi</th><th style = "text-align: left;">Species</th><th style = "text-align: left;">Weapon</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Union{Missing, Float64}" style = "text-align: left;">Float64?</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Union{Missing, InlineStrings.String7}" style = "text-align: left;">String7</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">Chief Chirpa</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">50.0</td><td style = "text-align: left;">black</td><td style = "text-align: left;">gray</td><td style = "text-align: left;">brown</td><td style = "text-align: left;">Endor</td><td style = "text-align: left;">unk_born</td><td style = "text-align: left;">4ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">ewok</td><td style = "text-align: left;">spear</td></tr></tbody></table></div><p>Then recombine into a single dataframe based on a function operating over the split dataframes:</p><pre><code class="language-julia hljs">eyecolor_count = DataFrames.combine(by_eyecolor) do df
+)</code></pre><img src="c9b9ec62.svg" alt="Example block output"/><p>That looks better.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>Read the <a href="https://csv.juliadata.org/stable/">CSV documentation</a> for other parsing options.</p></div></div><p>DataFrames.jl supports manipulation using functions similar to pandas. For example, split the dataframe into groups based on eye-color:</p><pre><code class="language-julia hljs">by_eyecolor = DataFrames.groupby(csv_df, :Eyecolor)</code></pre><p><b>GroupedDataFrame with 7 groups based on key: Eyecolor</b></p><div><div style = "float: left;"><span>First Group (5 rows): Eyecolor = &quot;blue&quot;</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Gender</th><th style = "text-align: left;">Height</th><th style = "text-align: left;">Weight</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">Haircolor</th><th style = "text-align: left;">Skincolor</th><th style = "text-align: left;">Homeland</th><th style = "text-align: left;">Born</th><th style = "text-align: left;">Died</th><th style = "text-align: left;">Jedi</th><th style = "text-align: left;">Species</th><th style = "text-align: left;">Weapon</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Union{Missing, Float64}" style = "text-align: left;">Float64?</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Union{Missing, InlineStrings.String7}" style = "text-align: left;">String7</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">Anakin Skywalker</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.88</td><td style = "text-align: right;">84.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">blond</td><td style = "text-align: left;">fair</td><td style = "text-align: left;">Tatooine</td><td style = "text-align: left;">41.9BBY</td><td style = "text-align: left;">4ABY</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">Luke Skywalker</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.72</td><td style = "text-align: right;">77.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">blond</td><td style = "text-align: left;">fair</td><td style = "text-align: left;">Tatooine</td><td style = "text-align: left;">19BBY</td><td style = "text-align: left;">unk_died</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">Qui-Gon Jinn</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.93</td><td style = "text-align: right;">88.5</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">brown</td><td style = "text-align: left;">light</td><td style = "text-align: left;">unk_planet</td><td style = "text-align: left;">92BBY</td><td style = "text-align: left;">32BBY</td><td style = "text-align: left;">jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">lightsaber</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">Sheev Palpatine</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.73</td><td style = "text-align: right;">75.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">red</td><td style = "text-align: left;">pale</td><td style = "text-align: left;">Naboo</td><td style = "text-align: left;">82BBY</td><td style = "text-align: left;">10ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">human</td><td style = "text-align: left;">force-lightning</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">Chewbacca</td><td style = "text-align: left;">male</td><td style = "text-align: right;">2.28</td><td style = "text-align: right;">112.0</td><td style = "text-align: left;">blue</td><td style = "text-align: left;">brown</td><td style = "font-style: italic; text-align: left;">missing</td><td style = "text-align: left;">Kashyyyk</td><td style = "text-align: left;">200BBY</td><td style = "text-align: left;">25ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">wookiee</td><td style = "text-align: left;">bowcaster</td></tr></tbody></table></div><p>&vellip;</p><div><div style = "float: left;"><span>Last Group (1 row): Eyecolor = &quot;black&quot;</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Gender</th><th style = "text-align: left;">Height</th><th style = "text-align: left;">Weight</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">Haircolor</th><th style = "text-align: left;">Skincolor</th><th style = "text-align: left;">Homeland</th><th style = "text-align: left;">Born</th><th style = "text-align: left;">Died</th><th style = "text-align: left;">Jedi</th><th style = "text-align: left;">Species</th><th style = "text-align: left;">Weapon</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Union{Missing, Float64}" style = "text-align: left;">Float64?</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Union{Missing, InlineStrings.String7}" style = "text-align: left;">String7</th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String7" style = "text-align: left;">String7</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">Chief Chirpa</td><td style = "text-align: left;">male</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">50.0</td><td style = "text-align: left;">black</td><td style = "text-align: left;">gray</td><td style = "text-align: left;">brown</td><td style = "text-align: left;">Endor</td><td style = "text-align: left;">unk_born</td><td style = "text-align: left;">4ABY</td><td style = "text-align: left;">no_jedi</td><td style = "text-align: left;">ewok</td><td style = "text-align: left;">spear</td></tr></tbody></table></div><p>Then recombine into a single dataframe based on a function operating over the split dataframes:</p><pre><code class="language-julia hljs">eyecolor_count = DataFrames.combine(by_eyecolor) do df
     return DataFrames.nrow(df)
 end</code></pre><div><div style = "float: left;"><span>7×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">x1</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Int64" style = "text-align: left;">Int64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">blue</td><td style = "text-align: right;">5</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">brown</td><td style = "text-align: right;">8</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">bluegray</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "font-style: italic; text-align: left;">missing</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">yellow</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">gold</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">black</td><td style = "text-align: right;">1</td></tr></tbody></table></div><p>We can rename columns:</p><pre><code class="language-julia hljs">DataFrames.rename!(eyecolor_count, :x1 =&gt; :count)</code></pre><div><div style = "float: left;"><span>7×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">count</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "Union{Missing, InlineStrings.String15}" style = "text-align: left;">String15</th><th title = "Int64" style = "text-align: left;">Int64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">blue</td><td style = "text-align: right;">5</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">brown</td><td style = "text-align: right;">8</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">bluegray</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "font-style: italic; text-align: left;">missing</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">yellow</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">gold</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">black</td><td style = "text-align: right;">1</td></tr></tbody></table></div><p>Drop some missing rows:</p><pre><code class="language-julia hljs">DataFrames.dropmissing!(eyecolor_count, :Eyecolor)</code></pre><div><div style = "float: left;"><span>6×2 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Eyecolor</th><th style = "text-align: left;">count</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String15" style = "text-align: left;">String15</th><th title = "Int64" style = "text-align: left;">Int64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">blue</td><td style = "text-align: right;">5</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">brown</td><td style = "text-align: right;">8</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">bluegray</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">yellow</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">gold</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">black</td><td style = "text-align: right;">1</td></tr></tbody></table></div><p>Then we can visualize the data:</p><pre><code class="language-julia hljs">sort!(eyecolor_count, :count; rev = true)
 Plots.bar(
@@ -35,7 +35,7 @@
     xlabel = &quot;Eye color&quot;,
     ylabel = &quot;Number of characters&quot;,
     label = false,
-)</code></pre><img src="1e380636.svg" alt="Example block output"/><h3 id="Other-Delimited-Files"><a class="docs-heading-anchor" href="#Other-Delimited-Files">Other Delimited Files</a><a id="Other-Delimited-Files-1"></a><a class="docs-heading-anchor-permalink" href="#Other-Delimited-Files" title="Permalink"></a></h3><p>We can also use the <code>CSV.jl</code> package to read any other delimited text file format.</p><p>By default, CSV.File will try to detect a file&#39;s delimiter from the first 10 lines of the file.</p><p>Candidate delimiters include <code>&#39;,&#39;</code>, <code>&#39;\t&#39;</code>, <code>&#39; &#39;</code>, <code>&#39;|&#39;</code>, <code>&#39;;&#39;</code>, and <code>&#39;:&#39;</code>. If it can&#39;t auto-detect the delimiter, it will assume <code>&#39;,&#39;</code>.</p><p>Let&#39;s take the example of space separated data.</p><pre><code class="language-julia hljs">ss_df = CSV.read(joinpath(DATA_DIR, &quot;Cereal.txt&quot;), DataFrames.DataFrame)</code></pre><div><div style = "float: left;"><span>23×10 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Cups</th><th style = "text-align: left;">Calories</th><th style = "text-align: left;">Carbs</th><th style = "text-align: left;">Fat</th><th style = "text-align: left;">Fiber</th><th style = "text-align: left;">Potassium</th><th style = "text-align: left;">Protein</th><th style = "text-align: left;">Sodium</th><th style = "text-align: left;">Sugars</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">CapnCrunch</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">120</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">35</td><td style = "text-align: right;">1</td><td style = "text-align: right;">220</td><td style = "text-align: right;">12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">CocoaPuffs</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">55</td><td style = "text-align: right;">1</td><td style = "text-align: right;">180</td><td style = "text-align: right;">13</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">Trix</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">13.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">25</td><td style = "text-align: right;">1</td><td style = "text-align: right;">140</td><td style = "text-align: right;">12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">AppleJacks</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">11.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">30</td><td style = "text-align: right;">2</td><td style = "text-align: right;">125</td><td style = "text-align: right;">14</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">CornChex</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">22.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">25</td><td style = "text-align: right;">2</td><td style = "text-align: right;">280</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">CornFlakes</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">21.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">35</td><td style = "text-align: right;">2</td><td style = "text-align: right;">290</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">Nut&amp;Honey</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">120</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">40</td><td style = "text-align: right;">2</td><td style = "text-align: right;">190</td><td style = "text-align: right;">9</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: left;">Smacks</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">110</td><td style = "text-align: right;">9.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">40</td><td style = "text-align: right;">2</td><td style = "text-align: right;">70</td><td style = "text-align: right;">15</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: left;">MultiGrain</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">90</td><td style = "text-align: right;">2</td><td style = "text-align: right;">220</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: left;">CracklinOat</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">110</td><td style = "text-align: right;">10.0</td><td style = "text-align: right;">3</td><td style = "text-align: right;">4.0</td><td style = "text-align: right;">160</td><td style = "text-align: right;">3</td><td style = "text-align: right;">140</td><td style = "text-align: right;">7</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: left;">GrapeNuts</td><td style = "text-align: right;">0.25</td><td style = "text-align: right;">110</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">90</td><td style = "text-align: right;">3</td><td style = "text-align: right;">179</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: left;">HoneyNutCheerios</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">110</td><td style = "text-align: right;">11.5</td><td style = "text-align: right;">1</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">90</td><td style = "text-align: right;">3</td><td style = "text-align: right;">250</td><td style = "text-align: right;">10</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: left;">NutriGrain</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">140</td><td style = "text-align: right;">21.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">130</td><td style = "text-align: right;">3</td><td style = "text-align: right;">220</td><td style = "text-align: right;">7</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: left;">Product19</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">45</td><td style = "text-align: right;">3</td><td style = "text-align: right;">320</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: left;">TotalRaisinBran</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">140</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">4.0</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td><td style = "text-align: right;">190</td><td style = "text-align: right;">14</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: left;">WheatChex</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">100</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">115</td><td style = "text-align: right;">3</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: left;">Oatmeal</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">130</td><td style = "text-align: right;">13.5</td><td style = "text-align: right;">2</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">120</td><td style = "text-align: right;">3</td><td style = "text-align: right;">170</td><td style = "text-align: right;">10</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: left;">Life</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">100</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">95</td><td style = "text-align: right;">4</td><td style = "text-align: right;">150</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: left;">Maypo</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: left;">QuakerOats</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">100</td><td style = "text-align: right;">14.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">4</td><td style = "text-align: right;">135</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: left;">Muesli</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">150</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">3</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">170</td><td style = "text-align: right;">4</td><td style = "text-align: right;">150</td><td style = "text-align: right;">11</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: left;">Cheerios</td><td style = "text-align: right;">1.25</td><td style = "text-align: right;">110</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">105</td><td style = "text-align: right;">6</td><td style = "text-align: right;">290</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: left;">SpecialK</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">55</td><td style = "text-align: right;">6</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td></tr></tbody></table></div><p>We can also specify the delimiter as follows:</p><pre><code class="language-julia hljs">delim_df = CSV.read(
+)</code></pre><img src="6cdf6131.svg" alt="Example block output"/><h3 id="Other-Delimited-Files"><a class="docs-heading-anchor" href="#Other-Delimited-Files">Other Delimited Files</a><a id="Other-Delimited-Files-1"></a><a class="docs-heading-anchor-permalink" href="#Other-Delimited-Files" title="Permalink"></a></h3><p>We can also use the <code>CSV.jl</code> package to read any other delimited text file format.</p><p>By default, CSV.File will try to detect a file&#39;s delimiter from the first 10 lines of the file.</p><p>Candidate delimiters include <code>&#39;,&#39;</code>, <code>&#39;\t&#39;</code>, <code>&#39; &#39;</code>, <code>&#39;|&#39;</code>, <code>&#39;;&#39;</code>, and <code>&#39;:&#39;</code>. If it can&#39;t auto-detect the delimiter, it will assume <code>&#39;,&#39;</code>.</p><p>Let&#39;s take the example of space separated data.</p><pre><code class="language-julia hljs">ss_df = CSV.read(joinpath(DATA_DIR, &quot;Cereal.txt&quot;), DataFrames.DataFrame)</code></pre><div><div style = "float: left;"><span>23×10 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">Name</th><th style = "text-align: left;">Cups</th><th style = "text-align: left;">Calories</th><th style = "text-align: left;">Carbs</th><th style = "text-align: left;">Fat</th><th style = "text-align: left;">Fiber</th><th style = "text-align: left;">Potassium</th><th style = "text-align: left;">Protein</th><th style = "text-align: left;">Sodium</th><th style = "text-align: left;">Sugars</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "InlineStrings.String31" style = "text-align: left;">String31</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th><th title = "Int64" style = "text-align: left;">Int64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">CapnCrunch</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">120</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">35</td><td style = "text-align: right;">1</td><td style = "text-align: right;">220</td><td style = "text-align: right;">12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">CocoaPuffs</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">55</td><td style = "text-align: right;">1</td><td style = "text-align: right;">180</td><td style = "text-align: right;">13</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">Trix</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">13.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">25</td><td style = "text-align: right;">1</td><td style = "text-align: right;">140</td><td style = "text-align: right;">12</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">AppleJacks</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">11.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">30</td><td style = "text-align: right;">2</td><td style = "text-align: right;">125</td><td style = "text-align: right;">14</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">CornChex</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">22.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">25</td><td style = "text-align: right;">2</td><td style = "text-align: right;">280</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">CornFlakes</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">21.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">35</td><td style = "text-align: right;">2</td><td style = "text-align: right;">290</td><td style = "text-align: right;">2</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">Nut&amp;Honey</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">120</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">40</td><td style = "text-align: right;">2</td><td style = "text-align: right;">190</td><td style = "text-align: right;">9</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: left;">Smacks</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">110</td><td style = "text-align: right;">9.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">40</td><td style = "text-align: right;">2</td><td style = "text-align: right;">70</td><td style = "text-align: right;">15</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: left;">MultiGrain</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">90</td><td style = "text-align: right;">2</td><td style = "text-align: right;">220</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">10</td><td style = "text-align: left;">CracklinOat</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">110</td><td style = "text-align: right;">10.0</td><td style = "text-align: right;">3</td><td style = "text-align: right;">4.0</td><td style = "text-align: right;">160</td><td style = "text-align: right;">3</td><td style = "text-align: right;">140</td><td style = "text-align: right;">7</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">11</td><td style = "text-align: left;">GrapeNuts</td><td style = "text-align: right;">0.25</td><td style = "text-align: right;">110</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">90</td><td style = "text-align: right;">3</td><td style = "text-align: right;">179</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">12</td><td style = "text-align: left;">HoneyNutCheerios</td><td style = "text-align: right;">0.75</td><td style = "text-align: right;">110</td><td style = "text-align: right;">11.5</td><td style = "text-align: right;">1</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">90</td><td style = "text-align: right;">3</td><td style = "text-align: right;">250</td><td style = "text-align: right;">10</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">13</td><td style = "text-align: left;">NutriGrain</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">140</td><td style = "text-align: right;">21.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">130</td><td style = "text-align: right;">3</td><td style = "text-align: right;">220</td><td style = "text-align: right;">7</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">14</td><td style = "text-align: left;">Product19</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">45</td><td style = "text-align: right;">3</td><td style = "text-align: right;">320</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">15</td><td style = "text-align: left;">TotalRaisinBran</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">140</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">4.0</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td><td style = "text-align: right;">190</td><td style = "text-align: right;">14</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">16</td><td style = "text-align: left;">WheatChex</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">100</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">115</td><td style = "text-align: right;">3</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">17</td><td style = "text-align: left;">Oatmeal</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">130</td><td style = "text-align: right;">13.5</td><td style = "text-align: right;">2</td><td style = "text-align: right;">1.5</td><td style = "text-align: right;">120</td><td style = "text-align: right;">3</td><td style = "text-align: right;">170</td><td style = "text-align: right;">10</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">18</td><td style = "text-align: left;">Life</td><td style = "text-align: right;">0.67</td><td style = "text-align: right;">100</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">95</td><td style = "text-align: right;">4</td><td style = "text-align: right;">150</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">19</td><td style = "text-align: left;">Maypo</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">100</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">95</td><td style = "text-align: right;">4</td><td style = "text-align: right;">0</td><td style = "text-align: right;">3</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">20</td><td style = "text-align: left;">QuakerOats</td><td style = "text-align: right;">0.5</td><td style = "text-align: right;">100</td><td style = "text-align: right;">14.0</td><td style = "text-align: right;">1</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">4</td><td style = "text-align: right;">135</td><td style = "text-align: right;">6</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">21</td><td style = "text-align: left;">Muesli</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">150</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">3</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">170</td><td style = "text-align: right;">4</td><td style = "text-align: right;">150</td><td style = "text-align: right;">11</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">22</td><td style = "text-align: left;">Cheerios</td><td style = "text-align: right;">1.25</td><td style = "text-align: right;">110</td><td style = "text-align: right;">17.0</td><td style = "text-align: right;">2</td><td style = "text-align: right;">2.0</td><td style = "text-align: right;">105</td><td style = "text-align: right;">6</td><td style = "text-align: right;">290</td><td style = "text-align: right;">1</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">23</td><td style = "text-align: left;">SpecialK</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">110</td><td style = "text-align: right;">16.0</td><td style = "text-align: right;">0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">55</td><td style = "text-align: right;">6</td><td style = "text-align: right;">230</td><td style = "text-align: right;">3</td></tr></tbody></table></div><p>We can also specify the delimiter as follows:</p><pre><code class="language-julia hljs">delim_df = CSV.read(
     joinpath(DATA_DIR, &quot;Soccer.txt&quot;),
     DataFrames.DataFrame;
     delim = &quot;::&quot;,
@@ -203,7 +203,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 4.16134e-01
+  Solve time (sec)   : 8.87251e-03
   Simplex iterations : 26
   Barrier iterations : -1
   Node count         : 1
@@ -235,4 +235,4 @@
  * Uganda
  * United Arab Emirates
  * United States
- * Zimbabwe</code></pre><p>We need some passports, like New Zealand and the United States, which have widespread access to a large number of countries. However, we also need passports like North Korea which only have visa-free access to a very limited number of countries.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>We use <code>value(x[c]) &gt; 0.5</code> rather than <code>value(x[c]) == 1</code> to avoid excluding solutions like <code>x[c] = 0.99999</code> that are &quot;1&quot; to some tolerance.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_sets_and_indexing/">« Getting started with sets and indexing</a><a class="docs-footer-nextpage" href="../debugging/">Debugging »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ * Zimbabwe</code></pre><p>We need some passports, like New Zealand and the United States, which have widespread access to a large number of countries. However, we also need passports like North Korea which only have visa-free access to a very limited number of countries.</p><div class="admonition is-info"><header class="admonition-header">Note</header><div class="admonition-body"><p>We use <code>value(x[c]) &gt; 0.5</code> rather than <code>value(x[c]) == 1</code> to avoid excluding solutions like <code>x[c] = 0.99999</code> that are &quot;1&quot; to some tolerance.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_sets_and_indexing/">« Getting started with sets and indexing</a><a class="docs-footer-nextpage" href="../debugging/">Debugging »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_julia/index.html b/previews/PR3562/tutorials/getting_started/getting_started_with_julia/index.html
index 5a46febfe26..d54a79fdc24 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_julia/index.html
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_julia/index.html
@@ -180,4 +180,4 @@
  0.5103924401614957
  0.9296414851080324</code></pre><p>The Package Manager is used to install packages that are not part of Julia&#39;s standard library.</p><p>For example the following can be used to install JuMP,</p><pre><code class="language-julia hljs">using Pkg
 Pkg.add(&quot;JuMP&quot;)</code></pre><p>For a complete list of registered Julia packages see the package listing at <a href="https://juliahub.com">JuliaHub</a>.</p><p>From time to you may wish to use a Julia package that is not registered.  In this case a git repository URL can be used to install the package.</p><pre><code class="language-julia hljs">using Pkg
-Pkg.add(&quot;https://github.com/user-name/MyPackage.jl.git&quot;)</code></pre><h3 id="Package-environments"><a class="docs-heading-anchor" href="#Package-environments">Package environments</a><a id="Package-environments-1"></a><a class="docs-heading-anchor-permalink" href="#Package-environments" title="Permalink"></a></h3><p>By default, <code>Pkg.add</code> will add packages to Julia&#39;s global environment. However, Julia also has built-in support for virtual environments.</p><p>Activate a virtual environment with:</p><pre><code class="language-julia hljs">import Pkg; Pkg.activate(&quot;/path/to/environment&quot;)</code></pre><p>You can see what packages are installed in the current environment with <code>Pkg.status()</code>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>We <em>strongly</em> recommend you create a Pkg environment for each project that you create in Julia, and add only the packages that you need, instead of adding lots of packages to the global environment. The <a href="https://julialang.github.io/Pkg.jl/v1/environments/">Pkg manager documentation</a> has more information on this topic.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../getting_started_with_JuMP/">Getting started with JuMP »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Pkg.add(&quot;https://github.com/user-name/MyPackage.jl.git&quot;)</code></pre><h3 id="Package-environments"><a class="docs-heading-anchor" href="#Package-environments">Package environments</a><a id="Package-environments-1"></a><a class="docs-heading-anchor-permalink" href="#Package-environments" title="Permalink"></a></h3><p>By default, <code>Pkg.add</code> will add packages to Julia&#39;s global environment. However, Julia also has built-in support for virtual environments.</p><p>Activate a virtual environment with:</p><pre><code class="language-julia hljs">import Pkg; Pkg.activate(&quot;/path/to/environment&quot;)</code></pre><p>You can see what packages are installed in the current environment with <code>Pkg.status()</code>.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>We <em>strongly</em> recommend you create a Pkg environment for each project that you create in Julia, and add only the packages that you need, instead of adding lots of packages to the global environment. The <a href="https://julialang.github.io/Pkg.jl/v1/environments/">Pkg manager documentation</a> has more information on this topic.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../getting_started_with_JuMP/">Getting started with JuMP »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/getting_started_with_sets_and_indexing/index.html b/previews/PR3562/tutorials/getting_started/getting_started_with_sets_and_indexing/index.html
index 1af9e87215d..06fdc0dc800 100644
--- a/previews/PR3562/tutorials/getting_started/getting_started_with_sets_and_indexing/index.html
+++ b/previews/PR3562/tutorials/getting_started/getting_started_with_sets_and_indexing/index.html
@@ -219,4 +219,4 @@
   (&quot;Dunedin&quot;, &quot;Auckland&quot;)        =&gt; 1426
   (&quot;Auckland&quot;, &quot;Christchurch&quot;)   =&gt; 1071</code></pre><p>Then, we can create our model like so:</p><pre><code class="language-julia hljs">model = Model()
 @variable(model, x[keys(routes)])
-@objective(model, Min, sum(v * x[k] for (k, v) in routes))</code></pre><p class="math-container">\[ 643 x_{(&quot;Auckland&quot;, &quot;Wellington&quot;)} + 436 x_{(&quot;Wellington&quot;, &quot;Christchurch&quot;)} + 790 x_{(&quot;Wellington&quot;, &quot;Dunedin&quot;)} + 360 x_{(&quot;Christchurch&quot;, &quot;Dunedin&quot;)} + 1426 x_{(&quot;Auckland&quot;, &quot;Dunedin&quot;)} + 1426 x_{(&quot;Dunedin&quot;, &quot;Auckland&quot;)} + 1071 x_{(&quot;Auckland&quot;, &quot;Christchurch&quot;)} \]</p><p>This has a number of benefits over the other approaches, including a compacter algebraic model and variables that are named in a more meaningful way.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>If you&#39;re struggling to formulate a problem using the available syntax in JuMP, it&#39;s probably a sign that you should convert your data into a different form.</p></div></div><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>The purpose of this tutorial was to show how JuMP does not have specialized syntax for set creation and manipulation. Instead, you should use the tools provided by Julia itself.</p><p>This is both an opportunity and a challenge, because you are free to pick the syntax and data structures that best suit your problem, but for new users it can be daunting to decide which structure to use.</p><p>Read through some of the other JuMP tutorials to get inspiration and ideas for how you can use Julia&#39;s syntax and data structures to your advantage.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_JuMP/">« Getting started with JuMP</a><a class="docs-footer-nextpage" href="../getting_started_with_data_and_plotting/">Getting started with data and plotting »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+@objective(model, Min, sum(v * x[k] for (k, v) in routes))</code></pre><p class="math-container">\[ 643 x_{(&quot;Auckland&quot;, &quot;Wellington&quot;)} + 436 x_{(&quot;Wellington&quot;, &quot;Christchurch&quot;)} + 790 x_{(&quot;Wellington&quot;, &quot;Dunedin&quot;)} + 360 x_{(&quot;Christchurch&quot;, &quot;Dunedin&quot;)} + 1426 x_{(&quot;Auckland&quot;, &quot;Dunedin&quot;)} + 1426 x_{(&quot;Dunedin&quot;, &quot;Auckland&quot;)} + 1071 x_{(&quot;Auckland&quot;, &quot;Christchurch&quot;)} \]</p><p>This has a number of benefits over the other approaches, including a compacter algebraic model and variables that are named in a more meaningful way.</p><div class="admonition is-success"><header class="admonition-header">Tip</header><div class="admonition-body"><p>If you&#39;re struggling to formulate a problem using the available syntax in JuMP, it&#39;s probably a sign that you should convert your data into a different form.</p></div></div><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>The purpose of this tutorial was to show how JuMP does not have specialized syntax for set creation and manipulation. Instead, you should use the tools provided by Julia itself.</p><p>This is both an opportunity and a challenge, because you are free to pick the syntax and data structures that best suit your problem, but for new users it can be daunting to decide which structure to use.</p><p>Read through some of the other JuMP tutorials to get inspiration and ideas for how you can use Julia&#39;s syntax and data structures to your advantage.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../getting_started_with_JuMP/">« Getting started with JuMP</a><a class="docs-footer-nextpage" href="../getting_started_with_data_and_plotting/">Getting started with data and plotting »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/introduction/index.html b/previews/PR3562/tutorials/getting_started/introduction/index.html
index 017d0c31b91..5d9890b94b1 100644
--- a/previews/PR3562/tutorials/getting_started/introduction/index.html
+++ b/previews/PR3562/tutorials/getting_started/introduction/index.html
@@ -3,4 +3,4 @@
   function gtag(){dataLayer.push(arguments);}
   gtag('js', new Date());
   gtag('config', 'UA-44252521-1', {'page_path': location.pathname + location.search + location.hash});
-</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox" checked/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox" checked/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#How-these-tutorials-are-structured"><span>How these tutorials are structured</span></a></li></ul></li><li><a class="tocitem" href="../getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../debugging/">Debugging</a></li><li><a class="tocitem" href="../design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/">Introduction</a></li><li><a class="tocitem" href="../../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Tutorials</a></li><li><a class="is-disabled">Getting started</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/tutorials/getting_started/introduction.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>The purpose of these &quot;Getting started&quot; tutorials is to teach new users the basics of Julia and JuMP.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The &quot;Getting started with&quot; tutorials are basic introductions to different aspects of JuMP and Julia. If you are new to JuMP and Julia, start by reading them in the following order:<ul><li><a href="../getting_started_with_julia/#Getting-started-with-Julia">Getting started with Julia</a></li><li><a href="../getting_started_with_JuMP/#Getting-started-with-JuMP">Getting started with JuMP</a></li><li><a href="../getting_started_with_sets_and_indexing/#Getting-started-with-sets-and-indexing">Getting started with sets and indexing</a></li><li><a href="../getting_started_with_data_and_plotting/#Getting-started-with-data-and-plotting">Getting started with data and plotting</a></li></ul></li><li>Julia has a reputation for being &quot;fast.&quot; Unfortunately, it is also easy to write <em>slow</em> Julia code. <a href="../performance_tips/#Performance-tips">Performance tips</a> contains a number of important tips on how to improve the performance of models you write in JuMP.</li><li><a href="../design_patterns_for_larger_models/#Design-patterns-for-larger-models">Design patterns for larger models</a> is a more advanced tutorial that is aimed at users writing large JuMP models. It&#39;s in the &quot;Getting started&quot; section to give you an early preview of how JuMP makes it easy to structure larger models. If you are new to JuMP you may want to skip or briefly skim this tutorial, and come back to it once you have written a few JuMP models.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../../installation/">« Installation Guide</a><a class="docs-footer-nextpage" href="../getting_started_with_julia/">Getting started with Julia »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</script><script data-outdated-warner src="../../../assets/warner.js"></script><link href="https://cdnjs.cloudflare.com/ajax/libs/lato-font/3.0.0/css/lato-font.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/juliamono/0.050/juliamono.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/fontawesome.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/solid.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/brands.min.css" rel="stylesheet" type="text/css"/><link href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.16.8/katex.min.css" rel="stylesheet" type="text/css"/><script>documenterBaseURL="../../.."</script><script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js" data-main="../../../assets/documenter.js"></script><script src="../../../search_index.js"></script><script src="../../../siteinfo.js"></script><script src="../../../../versions.js"></script><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-dark.css" data-theme-name="documenter-dark" data-theme-primary-dark/><link class="docs-theme-link" rel="stylesheet" type="text/css" href="../../../assets/themes/documenter-light.css" data-theme-name="documenter-light" data-theme-primary/><script src="../../../assets/themeswap.js"></script><link href="../../../assets/extra_styles.css" rel="stylesheet" type="text/css"/></head><body><div id="documenter"><nav class="docs-sidebar"><a class="docs-logo" href="../../../"><img class="docs-light-only" src="../../../assets/logo.svg" alt="JuMP logo"/><img class="docs-dark-only" src="../../../assets/logo-dark.svg" alt="JuMP logo"/></a><button class="docs-search-query input is-rounded is-small is-clickable my-2 mx-auto py-1 px-2" id="documenter-search-query">Search docs (Ctrl + /)</button><ul class="docs-menu"><li><input class="collapse-toggle" id="menuitem-1" type="checkbox"/><label class="tocitem" for="menuitem-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../">Introduction</a></li><li><a class="tocitem" href="../../../should_i_use/">Should you use JuMP?</a></li><li><a class="tocitem" href="../../../installation/">Installation Guide</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2" type="checkbox" checked/><label class="tocitem" for="menuitem-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-2-1" type="checkbox" checked/><label class="tocitem" for="menuitem-2-1"><span class="docs-label">Getting started</span><i class="docs-chevron"></i></label><ul class="collapsed"><li class="is-active"><a class="tocitem" href>Introduction</a><ul class="internal"><li><a class="tocitem" href="#How-these-tutorials-are-structured"><span>How these tutorials are structured</span></a></li></ul></li><li><a class="tocitem" href="../getting_started_with_julia/">Getting started with Julia</a></li><li><a class="tocitem" href="../getting_started_with_JuMP/">Getting started with JuMP</a></li><li><a class="tocitem" href="../getting_started_with_sets_and_indexing/">Getting started with sets and indexing</a></li><li><a class="tocitem" href="../getting_started_with_data_and_plotting/">Getting started with data and plotting</a></li><li><a class="tocitem" href="../debugging/">Debugging</a></li><li><a class="tocitem" href="../design_patterns_for_larger_models/">Design patterns for larger models</a></li><li><a class="tocitem" href="../performance_tips/">Performance tips</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-2" type="checkbox"/><label class="tocitem" for="menuitem-2-2"><span class="docs-label">Linear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../linear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../linear/knapsack/">The knapsack problem example</a></li><li><a class="tocitem" href="../../linear/diet/">The diet problem</a></li><li><a class="tocitem" href="../../linear/cannery/">The cannery problem</a></li><li><a class="tocitem" href="../../linear/factory_schedule/">The factory schedule example</a></li><li><a class="tocitem" href="../../linear/multi/">The multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/multi_commodity_network/">The network multi-commodity flow problem</a></li><li><a class="tocitem" href="../../linear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../linear/facility_location/">The facility location problem</a></li><li><a class="tocitem" href="../../linear/finance/">Financial modeling problems</a></li><li><a class="tocitem" href="../../linear/geographic_clustering/">Geographical clustering</a></li><li><a class="tocitem" href="../../linear/network_flows/">Network flow problems</a></li><li><a class="tocitem" href="../../linear/transp/">The transportation problem</a></li><li><a class="tocitem" href="../../linear/multi_objective_knapsack/">Multi-objective knapsack</a></li><li><a class="tocitem" href="../../linear/multi_objective_examples/">Simple multi-objective examples</a></li><li><a class="tocitem" href="../../linear/sudoku/">Sudoku</a></li><li><a class="tocitem" href="../../linear/n-queens/">N-Queens</a></li><li><a class="tocitem" href="../../linear/constraint_programming/">Constraint programming</a></li><li><a class="tocitem" href="../../linear/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../linear/lp_sensitivity/">Sensitivity analysis of a linear program</a></li><li><a class="tocitem" href="../../linear/mip_duality/">Computing the duals of a mixed-integer program</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-3" type="checkbox"/><label class="tocitem" for="menuitem-2-3"><span class="docs-label">Nonlinear programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../nonlinear/introduction/">Introduction</a></li><li><a class="tocitem" href="../../nonlinear/simple_examples/">Simple examples</a></li><li><a class="tocitem" href="../../nonlinear/portfolio/">Portfolio optimization</a></li><li><a class="tocitem" href="../../nonlinear/rocket_control/">Rocket Control</a></li><li><a class="tocitem" href="../../nonlinear/space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a class="tocitem" href="../../nonlinear/tips_and_tricks/">Tips and tricks</a></li><li><a class="tocitem" href="../../nonlinear/user_defined_hessians/">User-defined Hessians</a></li><li><a class="tocitem" href="../../nonlinear/nested_problems/">Nested optimization problems</a></li><li><a class="tocitem" href="../../nonlinear/querying_hessians/">Computing Hessians</a></li><li><a class="tocitem" href="../../nonlinear/complementarity/">Mixed complementarity problems</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-4" type="checkbox"/><label class="tocitem" for="menuitem-2-4"><span class="docs-label">Conic programs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../conic/introduction/">Introduction</a></li><li><a class="tocitem" href="../../conic/start_values/">Primal and dual warm-starts</a></li><li><a class="tocitem" href="../../conic/tips_and_tricks/">Tips and Tricks</a></li><li><a class="tocitem" href="../../conic/simple_examples/">Simple semidefinite programming examples</a></li><li><a class="tocitem" href="../../conic/dualization/">Dualization</a></li><li><a class="tocitem" href="../../conic/arbitrary_precision/">Arbitrary precision arithmetic</a></li><li><a class="tocitem" href="../../conic/logistic_regression/">Logistic regression</a></li><li><a class="tocitem" href="../../conic/experiment_design/">Experiment design</a></li><li><a class="tocitem" href="../../conic/min_ellipse/">Minimal ellipses</a></li><li><a class="tocitem" href="../../conic/ellipse_approx/">Ellipsoid approximation</a></li><li><a class="tocitem" href="../../conic/quantum_discrimination/">Quantum state discrimination</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-5" type="checkbox"/><label class="tocitem" for="menuitem-2-5"><span class="docs-label">Algorithms</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../algorithms/benders_decomposition/">Benders decomposition</a></li><li><a class="tocitem" href="../../algorithms/cutting_stock_column_generation/">Column generation</a></li><li><a class="tocitem" href="../../algorithms/tsp_lazy_constraints/">Traveling Salesperson Problem</a></li><li><a class="tocitem" href="../../algorithms/parallelism/">Parallelism</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-2-6" type="checkbox"/><label class="tocitem" for="menuitem-2-6"><span class="docs-label">Applications</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../applications/power_systems/">Power Systems</a></li><li><a class="tocitem" href="../../applications/optimal_power_flow/">Optimal power flow</a></li><li><a class="tocitem" href="../../applications/web_app/">Serving web apps</a></li><li><a class="tocitem" href="../../applications/two_stage_stochastic/">Two-stage stochastic programs</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-3" type="checkbox"/><label class="tocitem" for="menuitem-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../manual/models/">Models</a></li><li><a class="tocitem" href="../../../manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../manual/expressions/">Expressions</a></li><li><a class="tocitem" href="../../../manual/objective/">Objectives</a></li><li><a class="tocitem" href="../../../manual/containers/">Containers</a></li><li><a class="tocitem" href="../../../manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../manual/callbacks/">Solver-independent Callbacks</a></li><li><a class="tocitem" href="../../../manual/complex/">Complex number support</a></li><li><a class="tocitem" href="../../../manual/nonlinear/">Nonlinear Modeling</a></li><li><a class="tocitem" href="../../../manual/nlp/">Nonlinear Modeling (Legacy)</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-4" type="checkbox"/><label class="tocitem" for="menuitem-4"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../api/JuMP/">JuMP</a></li><li><a class="tocitem" href="../../../api/JuMP.Containers/">JuMP.Containers</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-5" type="checkbox"/><label class="tocitem" for="menuitem-5"><span class="docs-label">Background Information</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../background/algebraic_modeling_languages/">Algebraic modeling languages</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-6" type="checkbox"/><label class="tocitem" for="menuitem-6"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../developers/contributing/">Contributing</a></li><li><a class="tocitem" href="../../../developers/extensions/">Extensions</a></li><li><a class="tocitem" href="../../../developers/custom_solver_binaries/">Custom binaries</a></li><li><a class="tocitem" href="../../../developers/style/">Style Guide</a></li><li><a class="tocitem" href="../../../developers/roadmap/">Roadmap</a></li><li><a class="tocitem" href="../../../developers/checklists/">Checklists</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-7" type="checkbox"/><label class="tocitem" for="menuitem-7"><span class="docs-label">Solvers</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../packages/solvers/">Introduction</a></li><li><a class="tocitem" href="../../../packages/AmplNLWriter/">jump-dev/AmplNLWriter.jl</a></li><li><a class="tocitem" href="../../../packages/BARON/">jump-dev/BARON.jl</a></li><li><a class="tocitem" href="../../../packages/CPLEX/">jump-dev/CPLEX.jl</a></li><li><a class="tocitem" href="../../../packages/CSDP/">jump-dev/CSDP.jl</a></li><li><a class="tocitem" href="../../../packages/Cbc/">jump-dev/Cbc.jl</a></li><li><a class="tocitem" href="../../../packages/Clp/">jump-dev/Clp.jl</a></li><li><a class="tocitem" href="../../../packages/Dualization/">jump-dev/Dualization.jl</a></li><li><a class="tocitem" href="../../../packages/ECOS/">jump-dev/ECOS.jl</a></li><li><a class="tocitem" href="../../../packages/GLPK/">jump-dev/GLPK.jl</a></li><li><a class="tocitem" href="../../../packages/Gurobi/">jump-dev/Gurobi.jl</a></li><li><a class="tocitem" href="../../../packages/HiGHS/">jump-dev/HiGHS.jl</a></li><li><a class="tocitem" href="../../../packages/Ipopt/">jump-dev/Ipopt.jl</a></li><li><a class="tocitem" href="../../../packages/KNITRO/">jump-dev/KNITRO.jl</a></li><li><a class="tocitem" href="../../../packages/MiniZinc/">jump-dev/MiniZinc.jl</a></li><li><a class="tocitem" href="../../../packages/MosekTools/">jump-dev/MosekTools.jl</a></li><li><a class="tocitem" href="../../../packages/MultiObjectiveAlgorithms/">jump-dev/MultiObjectiveAlgorithms.jl</a></li><li><a class="tocitem" href="../../../packages/Pajarito/">jump-dev/Pajarito.jl</a></li><li><a class="tocitem" href="../../../packages/Pavito/">jump-dev/Pavito.jl</a></li><li><a class="tocitem" href="../../../packages/SCS/">jump-dev/SCS.jl</a></li><li><a class="tocitem" href="../../../packages/SDPA/">jump-dev/SDPA.jl</a></li><li><a class="tocitem" href="../../../packages/SDPNAL/">jump-dev/SDPNAL.jl</a></li><li><a class="tocitem" href="../../../packages/SDPT3/">jump-dev/SDPT3.jl</a></li><li><a class="tocitem" href="../../../packages/SeDuMi/">jump-dev/SeDuMi.jl</a></li><li><a class="tocitem" href="../../../packages/Xpress/">jump-dev/Xpress.jl</a></li><li><a class="tocitem" href="../../../packages/COPT/">COPT-Public/COPT.jl</a></li><li><a class="tocitem" href="../../../packages/GAMS/">GAMS-dev/GAMS.jl</a></li><li><a class="tocitem" href="../../../packages/NLopt/">JuliaOpt/NLopt.jl</a></li><li><a class="tocitem" href="../../../packages/CDDLib/">JuliaPolyhedra/CDDLib.jl</a></li><li><a class="tocitem" href="../../../packages/MadNLP/">MadNLP/MadNLP.jl</a></li><li><a class="tocitem" href="../../../packages/PATHSolver/">chkwon/PATHSolver.jl</a></li><li><a class="tocitem" href="../../../packages/Hypatia/">chriscoey/Hypatia.jl</a></li><li><a class="tocitem" href="../../../packages/DAQP/">darnstrom/DAQP.jl</a></li><li><a class="tocitem" href="../../../packages/Tulip/">ds4dm/Tulip.jl</a></li><li><a class="tocitem" href="../../../packages/Loraine/">kocvara/Loraine.jl</a></li><li><a class="tocitem" href="../../../packages/Alpine/">lanl-ansi/Alpine.jl</a></li><li><a class="tocitem" href="../../../packages/Juniper/">lanl-ansi/Juniper.jl</a></li><li><a class="tocitem" href="../../../packages/ProxSDP/">mariohsouto/ProxSDP.jl</a></li><li><a class="tocitem" href="../../../packages/NEOSServer/">odow/NEOSServer.jl</a></li><li><a class="tocitem" href="../../../packages/OSQP/">osqp/OSQP.jl</a></li><li><a class="tocitem" href="../../../packages/CDCS/">oxfordcontrol/CDCS.jl</a></li><li><a class="tocitem" href="../../../packages/COSMO/">oxfordcontrol/COSMO.jl</a></li><li><a class="tocitem" href="../../../packages/Clarabel/">oxfordcontrol/Clarabel.jl</a></li><li><a class="tocitem" href="../../../packages/SCIP/">scipopt/SCIP.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-8" type="checkbox"/><label class="tocitem" for="menuitem-8"><span class="docs-label">Extensions</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../extensions/introduction/">Introduction</a></li><li><a class="tocitem" href="../../../packages/DiffOpt/">jump-dev/DiffOpt.jl</a></li><li><a class="tocitem" href="../../../packages/ParametricOptInterface/">jump-dev/ParametricOptInterface.jl</a></li><li><a class="tocitem" href="../../../packages/PolyJuMP/">jump-dev/PolyJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SumOfSquares/">jump-dev/SumOfSquares.jl</a></li><li><a class="tocitem" href="../../../packages/InfiniteOpt/">infiniteopt/InfiniteOpt.jl</a></li><li><a class="tocitem" href="../../../packages/BilevelJuMP/">joaquimg/BilevelJuMP.jl</a></li><li><a class="tocitem" href="../../../packages/SDDP/">odow/SDDP.jl</a></li><li><a class="tocitem" href="../../../packages/Plasmo/">plasmo-dev/Plasmo.jl</a></li><li><a class="tocitem" href="../../../extensions/DimensionalData/">rafaqz/DimensionalData.jl</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9" type="checkbox"/><label class="tocitem" for="menuitem-9"><span class="docs-label">MathOptInterface</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-1" type="checkbox"/><label class="tocitem" for="menuitem-9-1"><span class="docs-label">Introduction</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/">Introduction</a></li><li><a class="tocitem" href="../../../moi/background/motivation/">Motivation</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-2" type="checkbox"/><label class="tocitem" for="menuitem-9-2"><span class="docs-label">Tutorials</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/tutorials/example/">Solving a problem using MathOptInterface</a></li><li><a class="tocitem" href="../../../moi/tutorials/implementing/">Implementing a solver interface</a></li><li><a class="tocitem" href="../../../moi/tutorials/mathprogbase/">Transitioning from MathProgBase</a></li><li><a class="tocitem" href="../../../moi/tutorials/bridging_constraint/">Implementing a constraint bridge</a></li><li><a class="tocitem" href="../../../moi/tutorials/manipulating_expressions/">Manipulating expressions</a></li><li><a class="tocitem" href="../../../moi/tutorials/latency/">Latency</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-3" type="checkbox"/><label class="tocitem" for="menuitem-9-3"><span class="docs-label">Manual</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/manual/standard_form/">Standard form problem</a></li><li><a class="tocitem" href="../../../moi/manual/models/">Models</a></li><li><a class="tocitem" href="../../../moi/manual/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/manual/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/manual/solutions/">Solutions</a></li><li><a class="tocitem" href="../../../moi/manual/modification/">Problem modification</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-4" type="checkbox"/><label class="tocitem" for="menuitem-9-4"><span class="docs-label">Background</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/background/duality/">Duality</a></li><li><a class="tocitem" href="../../../moi/background/infeasibility_certificates/">Infeasibility certificates</a></li><li><a class="tocitem" href="../../../moi/background/naming_conventions/">Naming conventions</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-5" type="checkbox"/><label class="tocitem" for="menuitem-9-5"><span class="docs-label">API Reference</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/reference/standard_form/">Standard form</a></li><li><a class="tocitem" href="../../../moi/reference/models/">Models</a></li><li><a class="tocitem" href="../../../moi/reference/variables/">Variables</a></li><li><a class="tocitem" href="../../../moi/reference/constraints/">Constraints</a></li><li><a class="tocitem" href="../../../moi/reference/modification/">Modifications</a></li><li><a class="tocitem" href="../../../moi/reference/nonlinear/">Nonlinear programming</a></li><li><a class="tocitem" href="../../../moi/reference/callbacks/">Callbacks</a></li><li><a class="tocitem" href="../../../moi/reference/errors/">Errors</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6"><span class="docs-label">Submodules</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><input class="collapse-toggle" id="menuitem-9-6-1" type="checkbox"/><label class="tocitem" for="menuitem-9-6-1"><span class="docs-label">Benchmarks</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Benchmarks/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-2" type="checkbox"/><label class="tocitem" for="menuitem-9-6-2"><span class="docs-label">Bridges</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Bridges/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/list_of_bridges/">List of bridges</a></li><li><a class="tocitem" href="../../../moi/submodules/Bridges/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-3" type="checkbox"/><label class="tocitem" for="menuitem-9-6-3"><span class="docs-label">FileFormats</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/FileFormats/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/FileFormats/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-4" type="checkbox"/><label class="tocitem" for="menuitem-9-6-4"><span class="docs-label">Nonlinear</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Nonlinear/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-5" type="checkbox"/><label class="tocitem" for="menuitem-9-6-5"><span class="docs-label">Utilities</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Utilities/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Utilities/reference/">API Reference</a></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-6-6" type="checkbox"/><label class="tocitem" for="menuitem-9-6-6"><span class="docs-label">Test</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/submodules/Test/overview/">Overview</a></li><li><a class="tocitem" href="../../../moi/submodules/Test/reference/">API Reference</a></li></ul></li></ul></li><li><input class="collapse-toggle" id="menuitem-9-7" type="checkbox"/><label class="tocitem" for="menuitem-9-7"><span class="docs-label">Developer Docs</span><i class="docs-chevron"></i></label><ul class="collapsed"><li><a class="tocitem" href="../../../moi/developer/checklists/">Checklists</a></li></ul></li><li><a class="tocitem" href="../../../moi/release_notes/">Release notes</a></li></ul></li><li><a class="tocitem" href="../../../release_notes/">Release notes</a></li></ul><div class="docs-version-selector field has-addons"><div class="control"><span class="docs-label button is-static is-size-7">Version</span></div><div class="docs-selector control is-expanded"><div class="select is-fullwidth is-size-7"><select id="documenter-version-selector"></select></div></div></div></nav><div class="docs-main"><header class="docs-navbar"><a class="docs-sidebar-button docs-navbar-link fa-solid fa-bars is-hidden-desktop" id="documenter-sidebar-button" href="#"></a><nav class="breadcrumb"><ul class="is-hidden-mobile"><li><a class="is-disabled">Tutorials</a></li><li><a class="is-disabled">Getting started</a></li><li class="is-active"><a href>Introduction</a></li></ul><ul class="is-hidden-tablet"><li class="is-active"><a href>Introduction</a></li></ul></nav><div class="docs-right"><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl" title="View the repository on GitHub"><span class="docs-icon fa-brands"></span><span class="docs-label is-hidden-touch">GitHub</span></a><a class="docs-navbar-link" href="https://github.com/jump-dev/JuMP.jl/blob/master/docs/src/tutorials/getting_started/introduction.md" title="Edit source on GitHub"><span class="docs-icon fa-solid"></span></a><a class="docs-settings-button docs-navbar-link fa-solid fa-gear" id="documenter-settings-button" href="#" title="Settings"></a><a class="docs-article-toggle-button fa-solid fa-chevron-up" id="documenter-article-toggle-button" href="javascript:;" title="Collapse all docstrings"></a></div></header><article class="content" id="documenter-page"><h1 id="Introduction"><a class="docs-heading-anchor" href="#Introduction">Introduction</a><a id="Introduction-1"></a><a class="docs-heading-anchor-permalink" href="#Introduction" title="Permalink"></a></h1><p>The purpose of these &quot;Getting started&quot; tutorials is to teach new users the basics of Julia and JuMP.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The &quot;Getting started with&quot; tutorials are basic introductions to different aspects of JuMP and Julia. If you are new to JuMP and Julia, start by reading them in the following order:<ul><li><a href="../getting_started_with_julia/#Getting-started-with-Julia">Getting started with Julia</a></li><li><a href="../getting_started_with_JuMP/#Getting-started-with-JuMP">Getting started with JuMP</a></li><li><a href="../getting_started_with_sets_and_indexing/#Getting-started-with-sets-and-indexing">Getting started with sets and indexing</a></li><li><a href="../getting_started_with_data_and_plotting/#Getting-started-with-data-and-plotting">Getting started with data and plotting</a></li></ul></li><li>Julia has a reputation for being &quot;fast.&quot; Unfortunately, it is also easy to write <em>slow</em> Julia code. <a href="../performance_tips/#Performance-tips">Performance tips</a> contains a number of important tips on how to improve the performance of models you write in JuMP.</li><li><a href="../design_patterns_for_larger_models/#Design-patterns-for-larger-models">Design patterns for larger models</a> is a more advanced tutorial that is aimed at users writing large JuMP models. It&#39;s in the &quot;Getting started&quot; section to give you an early preview of how JuMP makes it easy to structure larger models. If you are new to JuMP you may want to skip or briefly skim this tutorial, and come back to it once you have written a few JuMP models.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../../installation/">« Installation Guide</a><a class="docs-footer-nextpage" href="../getting_started_with_julia/">Getting started with Julia »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/getting_started/performance_tips/index.html b/previews/PR3562/tutorials/getting_started/performance_tips/index.html
index 66d3fdb7920..9eaab4c44f6 100644
--- a/previews/PR3562/tutorials/getting_started/performance_tips/index.html
+++ b/previews/PR3562/tutorials/getting_started/performance_tips/index.html
@@ -37,4 +37,4 @@
 Solver name: No optimizer attached.</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x[1:3])</code><code class="nohighlight hljs ansi" style="display:block;">3-element Vector{VariableRef}:
  x[1]
  x[2]
- x[3]</code></pre><p>Here&#39;s what happens if we construct the expression outside the macro:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; @allocated x[1] + x[2] + x[3]</code><code class="nohighlight hljs ansi" style="display:block;">1424</code></pre><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The <code>@allocated</code> measures how many bytes were allocated during the evaluation of an expression. Fewer is better.</p></div></div><p>If we use the <a href="../../../api/JuMP/#JuMP.@expression"><code>@expression</code></a> macro, we get many fewer allocations:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; @allocated @expression(model, x[1] + x[2] + x[3])</code><code class="nohighlight hljs ansi" style="display:block;">800</code></pre><h2 id="Disable-string-names"><a class="docs-heading-anchor" href="#Disable-string-names">Disable string names</a><a id="Disable-string-names-1"></a><a class="docs-heading-anchor-permalink" href="#Disable-string-names" title="Permalink"></a></h2><p>By default, JuMP creates <code>String</code> names for variables and constraints and passes these to the solver. The benefit of passing names is that it improves the readability of log messages from the solver (for example, &quot;variable x has invalid bounds&quot; instead of &quot;variable v1203 has invalid bounds&quot;), but for larger models the overhead of passing names can be non-trivial.</p><p>Disable the creation of <code>String</code> names by setting <code>set_string_name = false</code> in the <a href="../../../api/JuMP/#JuMP.@variable"><code>@variable</code></a> and <a href="../../../api/JuMP/#JuMP.@constraint"><code>@constraint</code></a> macros, or by calling <a href="../../../api/JuMP/#JuMP.set_string_names_on_creation"><code>set_string_names_on_creation</code></a> to disable all names for a particular model:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_string_names_on_creation(model, false)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x)</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, c, 2x &lt;= 1)</code><code class="nohighlight hljs ansi" style="display:block;">2 _[1] ≤ 1</code></pre><p>Note that this doesn&#39;t change how symbolic names and bindings are stored:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; x</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; model[:x]</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; x === model[:x]</code><code class="nohighlight hljs ansi" style="display:block;">true</code></pre><p>But you can no longer look up the variable by the string name:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; variable_by_name(model, &quot;x&quot;) === nothing</code><code class="nohighlight hljs ansi" style="display:block;">true</code></pre><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>For more information on the difference between string names, symbolic names, and bindings, see <a href="../../../manual/variables/#variable_names_and_bindings">String names, symbolic names, and bindings</a>.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../design_patterns_for_larger_models/">« Design patterns for larger models</a><a class="docs-footer-nextpage" href="../../linear/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ x[3]</code></pre><p>Here&#39;s what happens if we construct the expression outside the macro:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; @allocated x[1] + x[2] + x[3]</code><code class="nohighlight hljs ansi" style="display:block;">1424</code></pre><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The <code>@allocated</code> measures how many bytes were allocated during the evaluation of an expression. Fewer is better.</p></div></div><p>If we use the <a href="../../../api/JuMP/#JuMP.@expression"><code>@expression</code></a> macro, we get many fewer allocations:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; @allocated @expression(model, x[1] + x[2] + x[3])</code><code class="nohighlight hljs ansi" style="display:block;">800</code></pre><h2 id="Disable-string-names"><a class="docs-heading-anchor" href="#Disable-string-names">Disable string names</a><a id="Disable-string-names-1"></a><a class="docs-heading-anchor-permalink" href="#Disable-string-names" title="Permalink"></a></h2><p>By default, JuMP creates <code>String</code> names for variables and constraints and passes these to the solver. The benefit of passing names is that it improves the readability of log messages from the solver (for example, &quot;variable x has invalid bounds&quot; instead of &quot;variable v1203 has invalid bounds&quot;), but for larger models the overhead of passing names can be non-trivial.</p><p>Disable the creation of <code>String</code> names by setting <code>set_string_name = false</code> in the <a href="../../../api/JuMP/#JuMP.@variable"><code>@variable</code></a> and <a href="../../../api/JuMP/#JuMP.@constraint"><code>@constraint</code></a> macros, or by calling <a href="../../../api/JuMP/#JuMP.set_string_names_on_creation"><code>set_string_names_on_creation</code></a> to disable all names for a particular model:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; set_string_names_on_creation(model, false)</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x)</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, c, 2x &lt;= 1)</code><code class="nohighlight hljs ansi" style="display:block;">2 _[1] ≤ 1</code></pre><p>Note that this doesn&#39;t change how symbolic names and bindings are stored:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; x</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; model[:x]</code><code class="nohighlight hljs ansi" style="display:block;">_[1]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; x === model[:x]</code><code class="nohighlight hljs ansi" style="display:block;">true</code></pre><p>But you can no longer look up the variable by the string name:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; variable_by_name(model, &quot;x&quot;) === nothing</code><code class="nohighlight hljs ansi" style="display:block;">true</code></pre><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>For more information on the difference between string names, symbolic names, and bindings, see <a href="../../../manual/variables/#variable_names_and_bindings">String names, symbolic names, and bindings</a>.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../design_patterns_for_larger_models/">« Design patterns for larger models</a><a class="docs-footer-nextpage" href="../../linear/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/callbacks/index.html b/previews/PR3562/tutorials/linear/callbacks/index.html
index 607484e793b..164e2604163 100644
--- a/previews/PR3562/tutorials/linear/callbacks/index.html
+++ b/previews/PR3562/tutorials/linear/callbacks/index.html
@@ -149,4 +149,4 @@
 Adding ScalarConstraint{AffExpr, MathOptInterface.LessThan{Float64}}(y - x, MathOptInterface.LessThan{Float64}(1.0))
 Called from reason = 7
 Called from reason = 1
-Called from reason = 2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraint_programming/">« Constraint programming</a><a class="docs-footer-nextpage" href="../lp_sensitivity/">Sensitivity analysis of a linear program »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Called from reason = 2</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../constraint_programming/">« Constraint programming</a><a class="docs-footer-nextpage" href="../lp_sensitivity/">Sensitivity analysis of a linear program »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/cannery/index.html b/previews/PR3562/tutorials/linear/cannery/index.html
index 350d06ead01..1fb3d9c6b57 100644
--- a/previews/PR3562/tutorials/linear/cannery/index.html
+++ b/previews/PR3562/tutorials/linear/cannery/index.html
@@ -75,7 +75,7 @@
   Dual objective value : 1.68000e+03
 
 * Work counters
-  Solve time (sec)   : 2.12193e-04
+  Solve time (sec)   : 2.66790e-04
   Simplex iterations : 3
   Barrier iterations : 0
   Node count         : -1
@@ -86,4 +86,4 @@
 Seattle =&gt; New-York: 0.0
 San-Diego =&gt; Chicago: 0.0
 San-Diego =&gt; Topeka: 300.0
-San-Diego =&gt; New-York: 300.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../diet/">« The diet problem</a><a class="docs-footer-nextpage" href="../factory_schedule/">The factory schedule example »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+San-Diego =&gt; New-York: 300.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../diet/">« The diet problem</a><a class="docs-footer-nextpage" href="../factory_schedule/">The factory schedule example »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/constraint_programming/index.html b/previews/PR3562/tutorials/linear/constraint_programming/index.html
index d1f165164e4..6a4163c6340 100644
--- a/previews/PR3562/tutorials/linear/constraint_programming/index.html
+++ b/previews/PR3562/tutorials/linear/constraint_programming/index.html
@@ -89,4 +89,4 @@
 value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">3-element Vector{Float64}:
  1.0
  1.0
- 0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../n-queens/">« N-Queens</a><a class="docs-footer-nextpage" href="../callbacks/">Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../n-queens/">« N-Queens</a><a class="docs-footer-nextpage" href="../callbacks/">Callbacks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/diet/index.html b/previews/PR3562/tutorials/linear/diet/index.html
index 01271930c60..bf3625b15e9 100644
--- a/previews/PR3562/tutorials/linear/diet/index.html
+++ b/previews/PR3562/tutorials/linear/diet/index.html
@@ -10,7 +10,7 @@
 \min &amp; \sum\limits_{f \in F} c_f x_f \\
 \text{s.t.}\ \ &amp; l_m \le \sum\limits_{f \in F} a_{m,f} x_f \le u_m, &amp;&amp; \forall m \in M \\
 &amp; x_f \ge 0, &amp;&amp; \forall f \in F.
-\end{aligned}\]</p><p>In the rest of this tutorial, we will create and solve this problem in JuMP, and learn what we should cook for dinner.</p><h2 id="Data"><a class="docs-heading-anchor" href="#Data">Data</a><a id="Data-1"></a><a class="docs-heading-anchor-permalink" href="#Data" title="Permalink"></a></h2><p>First, we need some data for the problem. For this tutorial, we&#39;ll write CSV files to a temporary directory from Julia. If you have existing files, you could change the filenames to point to them instead.</p><pre><code class="language-julia hljs">dir = mktempdir()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">&quot;/tmp/jl_D3mWEy&quot;</code></pre><p>The first file is a list of foods with their macro-nutrient profile:</p><pre><code class="language-julia hljs">food_csv_filename = joinpath(dir, &quot;diet_foods.csv&quot;)
+\end{aligned}\]</p><p>In the rest of this tutorial, we will create and solve this problem in JuMP, and learn what we should cook for dinner.</p><h2 id="Data"><a class="docs-heading-anchor" href="#Data">Data</a><a id="Data-1"></a><a class="docs-heading-anchor-permalink" href="#Data" title="Permalink"></a></h2><p>First, we need some data for the problem. For this tutorial, we&#39;ll write CSV files to a temporary directory from Julia. If you have existing files, you could change the filenames to point to them instead.</p><pre><code class="language-julia hljs">dir = mktempdir()</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">&quot;/tmp/jl_0q7TEa&quot;</code></pre><p>The first file is a list of foods with their macro-nutrient profile:</p><pre><code class="language-julia hljs">food_csv_filename = joinpath(dir, &quot;diet_foods.csv&quot;)
 open(food_csv_filename, &quot;w&quot;) do io
     write(
         io,
@@ -101,7 +101,7 @@
   Dual objective value : 1.18289e+01
 
 * Work counters
-  Solve time (sec)   : 2.17199e-04
+  Solve time (sec)   : 2.07186e-04
   Simplex iterations : 6
   Barrier iterations : 0
   Node count         : -1
@@ -137,8 +137,8 @@
   Dual objective value : 3.56146e+00
 
 * Work counters
-  Solve time (sec)   : 1.41144e-04
+  Solve time (sec)   : 1.95980e-04
   Simplex iterations : 0
   Barrier iterations : 0
   Node count         : -1
-</code></pre><p>There exists no feasible solution to our problem. Looks like we&#39;re stuck eating ice cream for dinner.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>You can delete a constraint using <code>delete(model, dairy_constraint)</code>. Can you add a different constraint to provide a diet with less dairy?</li><li>Some food items (like hamburgers) are discrete. You can use <code>set_integer</code> to force a variable to take integer values. What happens to the solution if you do?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../knapsack/">« The knapsack problem example</a><a class="docs-footer-nextpage" href="../cannery/">The cannery problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</code></pre><p>There exists no feasible solution to our problem. Looks like we&#39;re stuck eating ice cream for dinner.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>You can delete a constraint using <code>delete(model, dairy_constraint)</code>. Can you add a different constraint to provide a diet with less dairy?</li><li>Some food items (like hamburgers) are discrete. You can use <code>set_integer</code> to force a variable to take integer values. What happens to the solution if you do?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../knapsack/">« The knapsack problem example</a><a class="docs-footer-nextpage" href="../cannery/">The cannery problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/facility_location/528bd41e.svg b/previews/PR3562/tutorials/linear/facility_location/492bb32b.svg
similarity index 79%
rename from previews/PR3562/tutorials/linear/facility_location/528bd41e.svg
rename to previews/PR3562/tutorials/linear/facility_location/492bb32b.svg
index 1096a928f5c..9176a7e3d41 100644
--- a/previews/PR3562/tutorials/linear/facility_location/528bd41e.svg
+++ b/previews/PR3562/tutorials/linear/facility_location/492bb32b.svg
@@ -1,69 +1,69 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip610">
+  <clipPath id="clip430">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip610)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip430)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip611">
+  <clipPath id="clip431">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip610)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip430)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip612">
+  <clipPath id="clip432">
     <rect x="156" y="47" width="2197" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
-<path clip-path="url(#clip610)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
-<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
-<path clip-path="url(#clip610)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip612)" cx="900.754" cy="117.213" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="1337.66" cy="998.336" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="2288.96" cy="1030.31" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="1840.64" cy="87.9763" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="218.754" cy="1410.69" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="768.03" cy="1026.75" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="2017.02" cy="967.646" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="2283.71" cy="1087.25" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="1955.74" cy="1445.72" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="257.146" cy="888.562" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="1413.3" cy="296.531" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip612)" cx="611.082" cy="118.991" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip612)" d="M614.682 166.046 L614.682 214.046 L662.682 214.046 L662.682 166.046 L614.682 166.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip612)" d="M1958.4 760.137 L1958.4 808.137 L2006.4 808.137 L2006.4 760.137 L1958.4 760.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip612)" d="M994.777 719.237 L994.777 767.237 L1042.78 767.237 L1042.78 719.237 L994.777 719.237 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
-<path clip-path="url(#clip612)" d="M806.323 1254.75 L806.323 1302.75 L854.323 1302.75 L854.323 1254.75 L806.323 1254.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip612)" d="M2266.6 876.347 L2266.6 924.347 L2314.6 924.347 L2314.6 876.347 L2266.6 876.347 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="900.754,117.213 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1337.66,998.336 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2288.96,1030.31 2290.6,900.347 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.64,87.9763 2290.6,900.347 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="218.754,1410.69 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="768.03,1026.75 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2017.02,967.646 2290.6,900.347 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2283.71,1087.25 2290.6,900.347 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1955.74,1445.72 2290.6,900.347 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="257.146,888.562 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.3,296.531 1018.78,743.237 "/>
-<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="611.082,118.991 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
+<path clip-path="url(#clip430)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
+<polyline clip-path="url(#clip430)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
+<path clip-path="url(#clip430)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip430)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip432)" cx="900.754" cy="117.213" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="1337.66" cy="998.336" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="2288.96" cy="1030.31" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="1840.64" cy="87.9763" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="218.754" cy="1410.69" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="768.03" cy="1026.75" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="2017.02" cy="967.646" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="2283.71" cy="1087.25" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="1955.74" cy="1445.72" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="257.146" cy="888.562" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="1413.3" cy="296.531" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip432)" cx="611.082" cy="118.991" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip432)" d="M614.682 166.046 L614.682 214.046 L662.682 214.046 L662.682 166.046 L614.682 166.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip432)" d="M1958.4 760.137 L1958.4 808.137 L2006.4 808.137 L2006.4 760.137 L1958.4 760.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip432)" d="M994.777 719.237 L994.777 767.237 L1042.78 767.237 L1042.78 719.237 L994.777 719.237 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
+<path clip-path="url(#clip432)" d="M806.323 1254.75 L806.323 1302.75 L854.323 1302.75 L854.323 1254.75 L806.323 1254.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip432)" d="M2266.6 876.347 L2266.6 924.347 L2314.6 924.347 L2314.6 876.347 L2266.6 876.347 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="900.754,117.213 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1337.66,998.336 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2288.96,1030.31 2290.6,900.347 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.64,87.9763 2290.6,900.347 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="218.754,1410.69 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="768.03,1026.75 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2017.02,967.646 2290.6,900.347 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2283.71,1087.25 2290.6,900.347 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1955.74,1445.72 2290.6,900.347 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="257.146,888.562 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.3,296.531 1018.78,743.237 "/>
+<polyline clip-path="url(#clip432)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="611.082,118.991 1018.78,743.237 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/linear/facility_location/0d89ccc9.svg b/previews/PR3562/tutorials/linear/facility_location/96afa50f.svg
similarity index 79%
rename from previews/PR3562/tutorials/linear/facility_location/0d89ccc9.svg
rename to previews/PR3562/tutorials/linear/facility_location/96afa50f.svg
index 64f59f3b506..63117b03f16 100644
--- a/previews/PR3562/tutorials/linear/facility_location/0d89ccc9.svg
+++ b/previews/PR3562/tutorials/linear/facility_location/96afa50f.svg
@@ -1,69 +1,69 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip730">
+  <clipPath id="clip550">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip730)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip550)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip731">
+  <clipPath id="clip551">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip730)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip550)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip732">
+  <clipPath id="clip552">
     <rect x="156" y="47" width="2197" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
-<path clip-path="url(#clip730)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
-<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
-<path clip-path="url(#clip730)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip732)" cx="900.754" cy="117.213" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="1337.66" cy="998.336" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="2288.96" cy="1030.31" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="1840.64" cy="87.9763" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="218.754" cy="1410.69" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="768.03" cy="1026.75" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="2017.02" cy="967.646" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="2283.71" cy="1087.25" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="1955.74" cy="1445.72" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="257.146" cy="888.562" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="1413.3" cy="296.531" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip732)" cx="611.082" cy="118.991" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip732)" d="M602.682 154.046 L602.682 226.046 L674.682 226.046 L674.682 154.046 L602.682 154.046 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
-<path clip-path="url(#clip732)" d="M1946.4 748.137 L1946.4 820.137 L2018.4 820.137 L2018.4 748.137 L1946.4 748.137 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
-<path clip-path="url(#clip732)" d="M998.777 723.237 L998.777 763.237 L1038.78 763.237 L1038.78 723.237 L998.777 723.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip732)" d="M798.323 1246.75 L798.323 1310.75 L862.323 1310.75 L862.323 1246.75 L798.323 1246.75 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
-<path clip-path="url(#clip732)" d="M2262.6 872.347 L2262.6 928.347 L2318.6 928.347 L2318.6 872.347 L2262.6 872.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="900.754,117.213 638.682,190.046 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1337.66,998.336 830.323,1278.75 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2288.96,1030.31 1982.4,784.137 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.64,87.9763 638.682,190.046 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="218.754,1410.69 830.323,1278.75 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="768.03,1026.75 830.323,1278.75 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2017.02,967.646 1982.4,784.137 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2283.71,1087.25 1982.4,784.137 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1955.74,1445.72 1982.4,784.137 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="257.146,888.562 638.682,190.046 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.3,296.531 638.682,190.046 "/>
-<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="611.082,118.991 638.682,190.046 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
+<path clip-path="url(#clip550)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
+<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
+<path clip-path="url(#clip550)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip552)" cx="900.754" cy="117.213" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="1337.66" cy="998.336" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="2288.96" cy="1030.31" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="1840.64" cy="87.9763" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="218.754" cy="1410.69" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="768.03" cy="1026.75" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="2017.02" cy="967.646" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="2283.71" cy="1087.25" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="1955.74" cy="1445.72" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="257.146" cy="888.562" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="1413.3" cy="296.531" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip552)" cx="611.082" cy="118.991" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip552)" d="M602.682 154.046 L602.682 226.046 L674.682 226.046 L674.682 154.046 L602.682 154.046 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
+<path clip-path="url(#clip552)" d="M1946.4 748.137 L1946.4 820.137 L2018.4 820.137 L2018.4 748.137 L1946.4 748.137 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
+<path clip-path="url(#clip552)" d="M998.777 723.237 L998.777 763.237 L1038.78 763.237 L1038.78 723.237 L998.777 723.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip552)" d="M798.323 1246.75 L798.323 1310.75 L862.323 1310.75 L862.323 1246.75 L798.323 1246.75 Z" fill="#ff0000" fill-rule="evenodd" fill-opacity="1" stroke="none"/>
+<path clip-path="url(#clip552)" d="M2262.6 872.347 L2262.6 928.347 L2318.6 928.347 L2318.6 872.347 L2262.6 872.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="900.754,117.213 638.682,190.046 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1337.66,998.336 830.323,1278.75 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2288.96,1030.31 1982.4,784.137 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.64,87.9763 638.682,190.046 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="218.754,1410.69 830.323,1278.75 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="768.03,1026.75 830.323,1278.75 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2017.02,967.646 1982.4,784.137 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2283.71,1087.25 1982.4,784.137 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1955.74,1445.72 1982.4,784.137 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="257.146,888.562 638.682,190.046 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1413.3,296.531 638.682,190.046 "/>
+<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="611.082,118.991 638.682,190.046 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/linear/facility_location/d825a2fa.svg b/previews/PR3562/tutorials/linear/facility_location/ca63c86d.svg
similarity index 81%
rename from previews/PR3562/tutorials/linear/facility_location/d825a2fa.svg
rename to previews/PR3562/tutorials/linear/facility_location/ca63c86d.svg
index 7a1d266bd89..62100ee5bf7 100644
--- a/previews/PR3562/tutorials/linear/facility_location/d825a2fa.svg
+++ b/previews/PR3562/tutorials/linear/facility_location/ca63c86d.svg
@@ -1,61 +1,61 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip550">
+  <clipPath id="clip370">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip550)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip370)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip551">
+  <clipPath id="clip371">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip550)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip370)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip552">
+  <clipPath id="clip372">
     <rect x="156" y="47" width="2197" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
-<path clip-path="url(#clip550)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
-<polyline clip-path="url(#clip552)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
-<path clip-path="url(#clip550)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip552)" cx="900.754" cy="117.213" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="1337.66" cy="998.336" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="2288.96" cy="1030.31" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="1840.64" cy="87.9763" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="218.754" cy="1410.69" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="768.03" cy="1026.75" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="2017.02" cy="967.646" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="2283.71" cy="1087.25" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="1955.74" cy="1445.72" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="257.146" cy="888.562" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="1413.3" cy="296.531" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip552)" cx="611.082" cy="118.991" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip552)" d="M614.682 166.046 L614.682 214.046 L662.682 214.046 L662.682 166.046 L614.682 166.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip552)" d="M1958.4 760.137 L1958.4 808.137 L2006.4 808.137 L2006.4 760.137 L1958.4 760.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip552)" d="M994.777 719.237 L994.777 767.237 L1042.78 767.237 L1042.78 719.237 L994.777 719.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip552)" d="M806.323 1254.75 L806.323 1302.75 L854.323 1302.75 L854.323 1254.75 L806.323 1254.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip552)" d="M2266.6 876.347 L2266.6 924.347 L2314.6 924.347 L2314.6 876.347 L2266.6 876.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip550)" d="M1901.81 250.738 L2279.55 250.738 L2279.55 95.2176 L1901.81 95.2176  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip550)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1901.81,250.738 2279.55,250.738 2279.55,95.2176 1901.81,95.2176 1901.81,250.738 "/>
-<circle clip-path="url(#clip550)" cx="1999.42" cy="147.058" r="20.48" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
-<path clip-path="url(#clip550)" d="M2124.89 132.44 L2124.89 137.37 Q2122.53 135.171 2119.85 134.083 Q2117.19 132.995 2114.18 132.995 Q2108.25 132.995 2105.1 136.629 Q2101.96 140.24 2101.96 147.092 Q2101.96 153.921 2105.1 157.555 Q2108.25 161.166 2114.18 161.166 Q2117.19 161.166 2119.85 160.078 Q2122.53 158.99 2124.89 156.791 L2124.89 161.676 Q2122.44 163.342 2119.69 164.176 Q2116.96 165.009 2113.9 165.009 Q2106.05 165.009 2101.54 160.217 Q2097.02 155.402 2097.02 147.092 Q2097.02 138.759 2101.54 133.967 Q2106.05 129.153 2113.9 129.153 Q2117 129.153 2119.73 129.986 Q2122.49 130.796 2124.89 132.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2131.93 128.319 L2136.19 128.319 L2136.19 164.338 L2131.93 164.338 L2131.93 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2145.1 138.412 L2149.36 138.412 L2149.36 164.338 L2145.1 164.338 L2145.1 138.412 M2145.1 128.319 L2149.36 128.319 L2149.36 133.713 L2145.1 133.713 L2145.1 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2180.45 150.31 L2180.45 152.393 L2160.87 152.393 Q2161.14 156.791 2163.51 159.106 Q2165.89 161.398 2170.13 161.398 Q2172.58 161.398 2174.87 160.796 Q2177.19 160.194 2179.45 158.99 L2179.45 163.018 Q2177.16 163.99 2174.76 164.5 Q2172.35 165.009 2169.87 165.009 Q2163.67 165.009 2160.03 161.398 Q2156.42 157.787 2156.42 151.629 Q2156.42 145.264 2159.85 141.537 Q2163.3 137.787 2169.13 137.787 Q2174.36 137.787 2177.39 141.166 Q2180.45 144.523 2180.45 150.31 M2176.19 149.06 Q2176.14 145.565 2174.22 143.481 Q2172.33 141.398 2169.18 141.398 Q2165.61 141.398 2163.46 143.412 Q2161.33 145.426 2161.01 149.083 L2176.19 149.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2208.99 148.689 L2208.99 164.338 L2204.73 164.338 L2204.73 148.828 Q2204.73 145.148 2203.3 143.319 Q2201.86 141.49 2198.99 141.49 Q2195.54 141.49 2193.55 143.69 Q2191.56 145.889 2191.56 149.685 L2191.56 164.338 L2187.28 164.338 L2187.28 138.412 L2191.56 138.412 L2191.56 142.44 Q2193.09 140.102 2195.15 138.944 Q2197.23 137.787 2199.94 137.787 Q2204.41 137.787 2206.7 140.565 Q2208.99 143.319 2208.99 148.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2221.7 131.051 L2221.7 138.412 L2230.47 138.412 L2230.47 141.722 L2221.7 141.722 L2221.7 155.796 Q2221.7 158.967 2222.56 159.87 Q2223.44 160.773 2226.1 160.773 L2230.47 160.773 L2230.47 164.338 L2226.1 164.338 Q2221.17 164.338 2219.29 162.509 Q2217.42 160.657 2217.42 155.796 L2217.42 141.722 L2214.29 141.722 L2214.29 138.412 L2217.42 138.412 L2217.42 131.051 L2221.7 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2252.6 139.176 L2252.6 143.203 Q2250.8 142.277 2248.85 141.815 Q2246.91 141.352 2244.82 141.352 Q2241.65 141.352 2240.06 142.324 Q2238.48 143.296 2238.48 145.24 Q2238.48 146.722 2239.62 147.578 Q2240.75 148.412 2244.18 149.176 L2245.64 149.5 Q2250.17 150.472 2252.07 152.254 Q2253.99 154.014 2253.99 157.185 Q2253.99 160.796 2251.12 162.902 Q2248.27 165.009 2243.27 165.009 Q2241.19 165.009 2238.92 164.592 Q2236.68 164.199 2234.18 163.388 L2234.18 158.99 Q2236.54 160.217 2238.83 160.842 Q2241.12 161.444 2243.37 161.444 Q2246.38 161.444 2248 160.426 Q2249.62 159.384 2249.62 157.509 Q2249.62 155.773 2248.44 154.847 Q2247.28 153.921 2243.32 153.064 L2241.84 152.717 Q2237.88 151.884 2236.12 150.171 Q2234.36 148.435 2234.36 145.426 Q2234.36 141.768 2236.95 139.778 Q2239.55 137.787 2244.32 137.787 Q2246.68 137.787 2248.76 138.134 Q2250.84 138.481 2252.6 139.176 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M1977.47 176.955 L1977.47 220.84 L2021.36 220.84 L2021.36 176.955 L1977.47 176.955 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="5.85143"/>
-<path clip-path="url(#clip550)" d="M2097.02 181.618 L2116.89 181.618 L2116.89 185.553 L2101.7 185.553 L2101.7 195.738 L2115.4 195.738 L2115.4 199.673 L2101.7 199.673 L2101.7 216.178 L2097.02 216.178 L2097.02 181.618 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2131.54 203.145 Q2126.38 203.145 2124.39 204.326 Q2122.39 205.506 2122.39 208.354 Q2122.39 210.622 2123.88 211.965 Q2125.38 213.284 2127.95 213.284 Q2131.49 213.284 2133.62 210.784 Q2135.77 208.261 2135.77 204.094 L2135.77 203.145 L2131.54 203.145 M2140.03 201.386 L2140.03 216.178 L2135.77 216.178 L2135.77 212.242 Q2134.32 214.603 2132.14 215.738 Q2129.96 216.849 2126.82 216.849 Q2122.83 216.849 2120.47 214.627 Q2118.14 212.381 2118.14 208.631 Q2118.14 204.256 2121.05 202.034 Q2123.99 199.812 2129.8 199.812 L2135.77 199.812 L2135.77 199.395 Q2135.77 196.455 2133.83 194.858 Q2131.91 193.238 2128.41 193.238 Q2126.19 193.238 2124.08 193.77 Q2121.98 194.303 2120.03 195.367 L2120.03 191.432 Q2122.37 190.53 2124.57 190.09 Q2126.77 189.627 2128.85 189.627 Q2134.48 189.627 2137.26 192.543 Q2140.03 195.46 2140.03 201.386 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2167.46 191.247 L2167.46 195.229 Q2165.66 194.233 2163.83 193.747 Q2162.02 193.238 2160.17 193.238 Q2156.03 193.238 2153.74 195.877 Q2151.45 198.492 2151.45 203.238 Q2151.45 207.983 2153.74 210.622 Q2156.03 213.238 2160.17 213.238 Q2162.02 213.238 2163.83 212.752 Q2165.66 212.242 2167.46 211.247 L2167.46 215.182 Q2165.68 216.016 2163.76 216.432 Q2161.86 216.849 2159.71 216.849 Q2153.85 216.849 2150.4 213.168 Q2146.96 209.488 2146.96 203.238 Q2146.96 196.895 2150.43 193.261 Q2153.92 189.627 2159.99 189.627 Q2161.95 189.627 2163.83 190.043 Q2165.7 190.437 2167.46 191.247 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2174.87 190.252 L2179.13 190.252 L2179.13 216.178 L2174.87 216.178 L2174.87 190.252 M2174.87 180.159 L2179.13 180.159 L2179.13 185.553 L2174.87 185.553 L2174.87 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2188.04 180.159 L2192.3 180.159 L2192.3 216.178 L2188.04 216.178 L2188.04 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2201.21 190.252 L2205.47 190.252 L2205.47 216.178 L2201.21 216.178 L2201.21 190.252 M2201.21 180.159 L2205.47 180.159 L2205.47 185.553 L2201.21 185.553 L2201.21 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2218.6 182.891 L2218.6 190.252 L2227.37 190.252 L2227.37 193.562 L2218.6 193.562 L2218.6 207.636 Q2218.6 210.807 2219.45 211.71 Q2220.33 212.613 2223 212.613 L2227.37 212.613 L2227.37 216.178 L2223 216.178 Q2218.07 216.178 2216.19 214.349 Q2214.32 212.497 2214.32 207.636 L2214.32 193.562 L2211.19 193.562 L2211.19 190.252 L2214.32 190.252 L2214.32 182.891 L2218.6 182.891 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip550)" d="M2243.76 218.585 Q2241.95 223.215 2240.24 224.627 Q2238.53 226.039 2235.66 226.039 L2232.26 226.039 L2232.26 222.474 L2234.76 222.474 Q2236.51 222.474 2237.49 221.64 Q2238.46 220.807 2239.64 217.705 L2240.4 215.761 L2229.92 190.252 L2234.43 190.252 L2242.53 210.529 L2250.64 190.252 L2255.15 190.252 L2243.76 218.585 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
+<path clip-path="url(#clip370)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
+<polyline clip-path="url(#clip372)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
+<path clip-path="url(#clip370)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip372)" cx="900.754" cy="117.213" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="1337.66" cy="998.336" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="2288.96" cy="1030.31" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="1840.64" cy="87.9763" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="218.754" cy="1410.69" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="768.03" cy="1026.75" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="2017.02" cy="967.646" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="2283.71" cy="1087.25" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="1955.74" cy="1445.72" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="257.146" cy="888.562" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="1413.3" cy="296.531" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip372)" cx="611.082" cy="118.991" r="14.4" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip372)" d="M614.682 166.046 L614.682 214.046 L662.682 214.046 L662.682 166.046 L614.682 166.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip372)" d="M1958.4 760.137 L1958.4 808.137 L2006.4 808.137 L2006.4 760.137 L1958.4 760.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip372)" d="M994.777 719.237 L994.777 767.237 L1042.78 767.237 L1042.78 719.237 L994.777 719.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip372)" d="M806.323 1254.75 L806.323 1302.75 L854.323 1302.75 L854.323 1254.75 L806.323 1254.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip372)" d="M2266.6 876.347 L2266.6 924.347 L2314.6 924.347 L2314.6 876.347 L2266.6 876.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip370)" d="M1901.81 250.738 L2279.55 250.738 L2279.55 95.2176 L1901.81 95.2176  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip370)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1901.81,250.738 2279.55,250.738 2279.55,95.2176 1901.81,95.2176 1901.81,250.738 "/>
+<circle clip-path="url(#clip370)" cx="1999.42" cy="147.058" r="20.48" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
+<path clip-path="url(#clip370)" d="M2124.89 132.44 L2124.89 137.37 Q2122.53 135.171 2119.85 134.083 Q2117.19 132.995 2114.18 132.995 Q2108.25 132.995 2105.1 136.629 Q2101.96 140.24 2101.96 147.092 Q2101.96 153.921 2105.1 157.555 Q2108.25 161.166 2114.18 161.166 Q2117.19 161.166 2119.85 160.078 Q2122.53 158.99 2124.89 156.791 L2124.89 161.676 Q2122.44 163.342 2119.69 164.176 Q2116.96 165.009 2113.9 165.009 Q2106.05 165.009 2101.54 160.217 Q2097.02 155.402 2097.02 147.092 Q2097.02 138.759 2101.54 133.967 Q2106.05 129.153 2113.9 129.153 Q2117 129.153 2119.73 129.986 Q2122.49 130.796 2124.89 132.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2131.93 128.319 L2136.19 128.319 L2136.19 164.338 L2131.93 164.338 L2131.93 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2145.1 138.412 L2149.36 138.412 L2149.36 164.338 L2145.1 164.338 L2145.1 138.412 M2145.1 128.319 L2149.36 128.319 L2149.36 133.713 L2145.1 133.713 L2145.1 128.319 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2180.45 150.31 L2180.45 152.393 L2160.87 152.393 Q2161.14 156.791 2163.51 159.106 Q2165.89 161.398 2170.13 161.398 Q2172.58 161.398 2174.87 160.796 Q2177.19 160.194 2179.45 158.99 L2179.45 163.018 Q2177.16 163.99 2174.76 164.5 Q2172.35 165.009 2169.87 165.009 Q2163.67 165.009 2160.03 161.398 Q2156.42 157.787 2156.42 151.629 Q2156.42 145.264 2159.85 141.537 Q2163.3 137.787 2169.13 137.787 Q2174.36 137.787 2177.39 141.166 Q2180.45 144.523 2180.45 150.31 M2176.19 149.06 Q2176.14 145.565 2174.22 143.481 Q2172.33 141.398 2169.18 141.398 Q2165.61 141.398 2163.46 143.412 Q2161.33 145.426 2161.01 149.083 L2176.19 149.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2208.99 148.689 L2208.99 164.338 L2204.73 164.338 L2204.73 148.828 Q2204.73 145.148 2203.3 143.319 Q2201.86 141.49 2198.99 141.49 Q2195.54 141.49 2193.55 143.69 Q2191.56 145.889 2191.56 149.685 L2191.56 164.338 L2187.28 164.338 L2187.28 138.412 L2191.56 138.412 L2191.56 142.44 Q2193.09 140.102 2195.15 138.944 Q2197.23 137.787 2199.94 137.787 Q2204.41 137.787 2206.7 140.565 Q2208.99 143.319 2208.99 148.689 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2221.7 131.051 L2221.7 138.412 L2230.47 138.412 L2230.47 141.722 L2221.7 141.722 L2221.7 155.796 Q2221.7 158.967 2222.56 159.87 Q2223.44 160.773 2226.1 160.773 L2230.47 160.773 L2230.47 164.338 L2226.1 164.338 Q2221.17 164.338 2219.29 162.509 Q2217.42 160.657 2217.42 155.796 L2217.42 141.722 L2214.29 141.722 L2214.29 138.412 L2217.42 138.412 L2217.42 131.051 L2221.7 131.051 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2252.6 139.176 L2252.6 143.203 Q2250.8 142.277 2248.85 141.815 Q2246.91 141.352 2244.82 141.352 Q2241.65 141.352 2240.06 142.324 Q2238.48 143.296 2238.48 145.24 Q2238.48 146.722 2239.62 147.578 Q2240.75 148.412 2244.18 149.176 L2245.64 149.5 Q2250.17 150.472 2252.07 152.254 Q2253.99 154.014 2253.99 157.185 Q2253.99 160.796 2251.12 162.902 Q2248.27 165.009 2243.27 165.009 Q2241.19 165.009 2238.92 164.592 Q2236.68 164.199 2234.18 163.388 L2234.18 158.99 Q2236.54 160.217 2238.83 160.842 Q2241.12 161.444 2243.37 161.444 Q2246.38 161.444 2248 160.426 Q2249.62 159.384 2249.62 157.509 Q2249.62 155.773 2248.44 154.847 Q2247.28 153.921 2243.32 153.064 L2241.84 152.717 Q2237.88 151.884 2236.12 150.171 Q2234.36 148.435 2234.36 145.426 Q2234.36 141.768 2236.95 139.778 Q2239.55 137.787 2244.32 137.787 Q2246.68 137.787 2248.76 138.134 Q2250.84 138.481 2252.6 139.176 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M1977.47 176.955 L1977.47 220.84 L2021.36 220.84 L2021.36 176.955 L1977.47 176.955 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="5.85143"/>
+<path clip-path="url(#clip370)" d="M2097.02 181.618 L2116.89 181.618 L2116.89 185.553 L2101.7 185.553 L2101.7 195.738 L2115.4 195.738 L2115.4 199.673 L2101.7 199.673 L2101.7 216.178 L2097.02 216.178 L2097.02 181.618 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2131.54 203.145 Q2126.38 203.145 2124.39 204.326 Q2122.39 205.506 2122.39 208.354 Q2122.39 210.622 2123.88 211.965 Q2125.38 213.284 2127.95 213.284 Q2131.49 213.284 2133.62 210.784 Q2135.77 208.261 2135.77 204.094 L2135.77 203.145 L2131.54 203.145 M2140.03 201.386 L2140.03 216.178 L2135.77 216.178 L2135.77 212.242 Q2134.32 214.603 2132.14 215.738 Q2129.96 216.849 2126.82 216.849 Q2122.83 216.849 2120.47 214.627 Q2118.14 212.381 2118.14 208.631 Q2118.14 204.256 2121.05 202.034 Q2123.99 199.812 2129.8 199.812 L2135.77 199.812 L2135.77 199.395 Q2135.77 196.455 2133.83 194.858 Q2131.91 193.238 2128.41 193.238 Q2126.19 193.238 2124.08 193.77 Q2121.98 194.303 2120.03 195.367 L2120.03 191.432 Q2122.37 190.53 2124.57 190.09 Q2126.77 189.627 2128.85 189.627 Q2134.48 189.627 2137.26 192.543 Q2140.03 195.46 2140.03 201.386 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2167.46 191.247 L2167.46 195.229 Q2165.66 194.233 2163.83 193.747 Q2162.02 193.238 2160.17 193.238 Q2156.03 193.238 2153.74 195.877 Q2151.45 198.492 2151.45 203.238 Q2151.45 207.983 2153.74 210.622 Q2156.03 213.238 2160.17 213.238 Q2162.02 213.238 2163.83 212.752 Q2165.66 212.242 2167.46 211.247 L2167.46 215.182 Q2165.68 216.016 2163.76 216.432 Q2161.86 216.849 2159.71 216.849 Q2153.85 216.849 2150.4 213.168 Q2146.96 209.488 2146.96 203.238 Q2146.96 196.895 2150.43 193.261 Q2153.92 189.627 2159.99 189.627 Q2161.95 189.627 2163.83 190.043 Q2165.7 190.437 2167.46 191.247 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2174.87 190.252 L2179.13 190.252 L2179.13 216.178 L2174.87 216.178 L2174.87 190.252 M2174.87 180.159 L2179.13 180.159 L2179.13 185.553 L2174.87 185.553 L2174.87 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2188.04 180.159 L2192.3 180.159 L2192.3 216.178 L2188.04 216.178 L2188.04 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2201.21 190.252 L2205.47 190.252 L2205.47 216.178 L2201.21 216.178 L2201.21 190.252 M2201.21 180.159 L2205.47 180.159 L2205.47 185.553 L2201.21 185.553 L2201.21 180.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2218.6 182.891 L2218.6 190.252 L2227.37 190.252 L2227.37 193.562 L2218.6 193.562 L2218.6 207.636 Q2218.6 210.807 2219.45 211.71 Q2220.33 212.613 2223 212.613 L2227.37 212.613 L2227.37 216.178 L2223 216.178 Q2218.07 216.178 2216.19 214.349 Q2214.32 212.497 2214.32 207.636 L2214.32 193.562 L2211.19 193.562 L2211.19 190.252 L2214.32 190.252 L2214.32 182.891 L2218.6 182.891 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip370)" d="M2243.76 218.585 Q2241.95 223.215 2240.24 224.627 Q2238.53 226.039 2235.66 226.039 L2232.26 226.039 L2232.26 222.474 L2234.76 222.474 Q2236.51 222.474 2237.49 221.64 Q2238.46 220.807 2239.64 217.705 L2240.4 215.761 L2229.92 190.252 L2234.43 190.252 L2242.53 210.529 L2250.64 190.252 L2255.15 190.252 L2243.76 218.585 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/linear/facility_location/7f2d6f44.svg b/previews/PR3562/tutorials/linear/facility_location/de2fb0ef.svg
similarity index 80%
rename from previews/PR3562/tutorials/linear/facility_location/7f2d6f44.svg
rename to previews/PR3562/tutorials/linear/facility_location/de2fb0ef.svg
index daddaac5822..74d43956eb3 100644
--- a/previews/PR3562/tutorials/linear/facility_location/7f2d6f44.svg
+++ b/previews/PR3562/tutorials/linear/facility_location/de2fb0ef.svg
@@ -1,57 +1,57 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip670">
+  <clipPath id="clip490">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip670)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip490)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip671">
+  <clipPath id="clip491">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip670)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip490)" d="M156.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L156.598 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip672">
+  <clipPath id="clip492">
     <rect x="156" y="47" width="2197" height="1440"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
-<path clip-path="url(#clip670)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
-<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
-<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
-<path clip-path="url(#clip670)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip672)" cx="900.754" cy="117.213" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="1337.66" cy="998.336" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="2288.96" cy="1030.31" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="1840.64" cy="87.9763" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="218.754" cy="1410.69" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="768.03" cy="1026.75" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="2017.02" cy="967.646" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="2283.71" cy="1087.25" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="1955.74" cy="1445.72" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="257.146" cy="888.562" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="1413.3" cy="296.531" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip672)" cx="611.082" cy="118.991" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip672)" d="M602.682 154.046 L602.682 226.046 L674.682 226.046 L674.682 154.046 L602.682 154.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip672)" d="M1946.4 748.137 L1946.4 820.137 L2018.4 820.137 L2018.4 748.137 L1946.4 748.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip672)" d="M998.777 723.237 L998.777 763.237 L1038.78 763.237 L1038.78 723.237 L998.777 723.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip672)" d="M798.323 1246.75 L798.323 1310.75 L862.323 1310.75 L862.323 1246.75 L798.323 1246.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
-<path clip-path="url(#clip672)" d="M2262.6 872.347 L2262.6 928.347 L2318.6 928.347 L2318.6 872.347 L2262.6 872.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="496.242,1486.45 496.242,47.2441 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="953.502,1486.45 953.502,47.2441 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1410.76,1486.45 1410.76,47.2441 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1868.02,1486.45 1868.02,47.2441 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2325.28,1486.45 2325.28,47.2441 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 2352.76,1486.45 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="496.242,1486.45 496.242,1467.55 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.502,1486.45 953.502,1467.55 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1410.76,1486.45 1410.76,1467.55 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1868.02,1486.45 1868.02,1467.55 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2325.28,1486.45 2325.28,1467.55 "/>
+<path clip-path="url(#clip490)" d="M474.425 1517.37 Q470.814 1517.37 468.985 1520.93 Q467.18 1524.47 467.18 1531.6 Q467.18 1538.71 468.985 1542.27 Q470.814 1545.82 474.425 1545.82 Q478.059 1545.82 479.865 1542.27 Q481.694 1538.71 481.694 1531.6 Q481.694 1524.47 479.865 1520.93 Q478.059 1517.37 474.425 1517.37 M474.425 1513.66 Q480.235 1513.66 483.291 1518.27 Q486.37 1522.85 486.37 1531.6 Q486.37 1540.33 483.291 1544.94 Q480.235 1549.52 474.425 1549.52 Q468.615 1549.52 465.536 1544.94 Q462.481 1540.33 462.481 1531.6 Q462.481 1522.85 465.536 1518.27 Q468.615 1513.66 474.425 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M494.587 1542.97 L499.471 1542.97 L499.471 1548.85 L494.587 1548.85 L494.587 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M513.684 1544.91 L530.004 1544.91 L530.004 1548.85 L508.059 1548.85 L508.059 1544.91 Q510.721 1542.16 515.305 1537.53 Q519.911 1532.88 521.092 1531.53 Q523.337 1529.01 524.217 1527.27 Q525.119 1525.51 525.119 1523.82 Q525.119 1521.07 523.175 1519.33 Q521.254 1517.6 518.152 1517.6 Q515.953 1517.6 513.499 1518.36 Q511.069 1519.13 508.291 1520.68 L508.291 1515.95 Q511.115 1514.82 513.569 1514.24 Q516.022 1513.66 518.059 1513.66 Q523.43 1513.66 526.624 1516.35 Q529.818 1519.03 529.818 1523.52 Q529.818 1525.65 529.008 1527.57 Q528.221 1529.47 526.115 1532.07 Q525.536 1532.74 522.434 1535.95 Q519.332 1539.15 513.684 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M930.644 1517.37 Q927.032 1517.37 925.204 1520.93 Q923.398 1524.47 923.398 1531.6 Q923.398 1538.71 925.204 1542.27 Q927.032 1545.82 930.644 1545.82 Q934.278 1545.82 936.083 1542.27 Q937.912 1538.71 937.912 1531.6 Q937.912 1524.47 936.083 1520.93 Q934.278 1517.37 930.644 1517.37 M930.644 1513.66 Q936.454 1513.66 939.509 1518.27 Q942.588 1522.85 942.588 1531.6 Q942.588 1540.33 939.509 1544.94 Q936.454 1549.52 930.644 1549.52 Q924.833 1549.52 921.755 1544.94 Q918.699 1540.33 918.699 1531.6 Q918.699 1522.85 921.755 1518.27 Q924.833 1513.66 930.644 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M950.805 1542.97 L955.69 1542.97 L955.69 1548.85 L950.805 1548.85 L950.805 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M978.722 1518.36 L966.917 1536.81 L978.722 1536.81 L978.722 1518.36 M977.495 1514.29 L983.375 1514.29 L983.375 1536.81 L988.305 1536.81 L988.305 1540.7 L983.375 1540.7 L983.375 1548.85 L978.722 1548.85 L978.722 1540.7 L963.12 1540.7 L963.12 1536.19 L977.495 1514.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1388.07 1517.37 Q1384.45 1517.37 1382.63 1520.93 Q1380.82 1524.47 1380.82 1531.6 Q1380.82 1538.71 1382.63 1542.27 Q1384.45 1545.82 1388.07 1545.82 Q1391.7 1545.82 1393.51 1542.27 Q1395.33 1538.71 1395.33 1531.6 Q1395.33 1524.47 1393.51 1520.93 Q1391.7 1517.37 1388.07 1517.37 M1388.07 1513.66 Q1393.88 1513.66 1396.93 1518.27 Q1400.01 1522.85 1400.01 1531.6 Q1400.01 1540.33 1396.93 1544.94 Q1393.88 1549.52 1388.07 1549.52 Q1382.26 1549.52 1379.18 1544.94 Q1376.12 1540.33 1376.12 1531.6 Q1376.12 1522.85 1379.18 1518.27 Q1382.26 1513.66 1388.07 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1408.23 1542.97 L1413.11 1542.97 L1413.11 1548.85 L1408.23 1548.85 L1408.23 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1433.88 1529.7 Q1430.73 1529.7 1428.88 1531.86 Q1427.05 1534.01 1427.05 1537.76 Q1427.05 1541.49 1428.88 1543.66 Q1430.73 1545.82 1433.88 1545.82 Q1437.02 1545.82 1438.85 1543.66 Q1440.7 1541.49 1440.7 1537.76 Q1440.7 1534.01 1438.85 1531.86 Q1437.02 1529.7 1433.88 1529.7 M1443.16 1515.05 L1443.16 1519.31 Q1441.4 1518.48 1439.59 1518.04 Q1437.81 1517.6 1436.05 1517.6 Q1431.42 1517.6 1428.97 1520.72 Q1426.54 1523.85 1426.19 1530.17 Q1427.56 1528.15 1429.62 1527.09 Q1431.68 1526 1434.15 1526 Q1439.36 1526 1442.37 1529.17 Q1445.4 1532.32 1445.4 1537.76 Q1445.4 1543.08 1442.26 1546.3 Q1439.11 1549.52 1433.88 1549.52 Q1427.88 1549.52 1424.71 1544.94 Q1421.54 1540.33 1421.54 1531.6 Q1421.54 1523.41 1425.43 1518.55 Q1429.32 1513.66 1435.87 1513.66 Q1437.63 1513.66 1439.41 1514.01 Q1441.21 1514.36 1443.16 1515.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1845.45 1517.37 Q1841.84 1517.37 1840.01 1520.93 Q1838.21 1524.47 1838.21 1531.6 Q1838.21 1538.71 1840.01 1542.27 Q1841.84 1545.82 1845.45 1545.82 Q1849.09 1545.82 1850.89 1542.27 Q1852.72 1538.71 1852.72 1531.6 Q1852.72 1524.47 1850.89 1520.93 Q1849.09 1517.37 1845.45 1517.37 M1845.45 1513.66 Q1851.26 1513.66 1854.32 1518.27 Q1857.4 1522.85 1857.4 1531.6 Q1857.4 1540.33 1854.32 1544.94 Q1851.26 1549.52 1845.45 1549.52 Q1839.64 1549.52 1836.56 1544.94 Q1833.51 1540.33 1833.51 1531.6 Q1833.51 1522.85 1836.56 1518.27 Q1839.64 1513.66 1845.45 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1865.61 1542.97 L1870.5 1542.97 L1870.5 1548.85 L1865.61 1548.85 L1865.61 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M1890.68 1532.44 Q1887.35 1532.44 1885.43 1534.22 Q1883.53 1536 1883.53 1539.13 Q1883.53 1542.25 1885.43 1544.03 Q1887.35 1545.82 1890.68 1545.82 Q1894.02 1545.82 1895.94 1544.03 Q1897.86 1542.23 1897.86 1539.13 Q1897.86 1536 1895.94 1534.22 Q1894.04 1532.44 1890.68 1532.44 M1886.01 1530.45 Q1883 1529.7 1881.31 1527.64 Q1879.64 1525.58 1879.64 1522.62 Q1879.64 1518.48 1882.58 1516.07 Q1885.55 1513.66 1890.68 1513.66 Q1895.85 1513.66 1898.79 1516.07 Q1901.73 1518.48 1901.73 1522.62 Q1901.73 1525.58 1900.04 1527.64 Q1898.37 1529.7 1895.38 1530.45 Q1898.76 1531.23 1900.64 1533.52 Q1902.54 1535.82 1902.54 1539.13 Q1902.54 1544.15 1899.46 1546.83 Q1896.4 1549.52 1890.68 1549.52 Q1884.97 1549.52 1881.89 1546.83 Q1878.83 1544.15 1878.83 1539.13 Q1878.83 1535.82 1880.73 1533.52 Q1882.63 1531.23 1886.01 1530.45 M1884.3 1523.06 Q1884.3 1525.75 1885.96 1527.25 Q1887.65 1528.76 1890.68 1528.76 Q1893.69 1528.76 1895.38 1527.25 Q1897.1 1525.75 1897.1 1523.06 Q1897.1 1520.38 1895.38 1518.87 Q1893.69 1517.37 1890.68 1517.37 Q1887.65 1517.37 1885.96 1518.87 Q1884.3 1520.38 1884.3 1523.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2292.44 1544.91 L2300.07 1544.91 L2300.07 1518.55 L2291.76 1520.21 L2291.76 1515.95 L2300.03 1514.29 L2304.7 1514.29 L2304.7 1544.91 L2312.34 1544.91 L2312.34 1548.85 L2292.44 1548.85 L2292.44 1544.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2321.79 1542.97 L2326.67 1542.97 L2326.67 1548.85 L2321.79 1548.85 L2321.79 1542.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M2346.86 1517.37 Q2343.25 1517.37 2341.42 1520.93 Q2339.61 1524.47 2339.61 1531.6 Q2339.61 1538.71 2341.42 1542.27 Q2343.25 1545.82 2346.86 1545.82 Q2350.49 1545.82 2352.3 1542.27 Q2354.12 1538.71 2354.12 1531.6 Q2354.12 1524.47 2352.3 1520.93 Q2350.49 1517.37 2346.86 1517.37 M2346.86 1513.66 Q2352.67 1513.66 2355.72 1518.27 Q2358.8 1522.85 2358.8 1531.6 Q2358.8 1540.33 2355.72 1544.94 Q2352.67 1549.52 2346.86 1549.52 Q2341.05 1549.52 2337.97 1544.94 Q2334.91 1540.33 2334.91 1531.6 Q2334.91 1522.85 2337.97 1518.27 Q2341.05 1513.66 2346.86 1513.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,1203.47 2352.76,1203.47 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,883.412 2352.76,883.412 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,563.356 2352.76,563.356 "/>
+<polyline clip-path="url(#clip492)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="156.598,243.3 2352.76,243.3 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1486.45 156.598,47.2441 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,1203.47 175.496,1203.47 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,883.412 175.496,883.412 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,563.356 175.496,563.356 "/>
+<polyline clip-path="url(#clip490)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="156.598,243.3 175.496,243.3 "/>
+<path clip-path="url(#clip490)" d="M65.0198 1189.27 Q61.4087 1189.27 59.58 1192.83 Q57.7745 1196.37 57.7745 1203.5 Q57.7745 1210.61 59.58 1214.17 Q61.4087 1217.72 65.0198 1217.72 Q68.6541 1217.72 70.4596 1214.17 Q72.2883 1210.61 72.2883 1203.5 Q72.2883 1196.37 70.4596 1192.83 Q68.6541 1189.27 65.0198 1189.27 M65.0198 1185.56 Q70.83 1185.56 73.8855 1190.17 Q76.9642 1194.75 76.9642 1203.5 Q76.9642 1212.23 73.8855 1216.84 Q70.83 1221.42 65.0198 1221.42 Q59.2097 1221.42 56.131 1216.84 Q53.0754 1212.23 53.0754 1203.5 Q53.0754 1194.75 56.131 1190.17 Q59.2097 1185.56 65.0198 1185.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M85.1818 1214.87 L90.066 1214.87 L90.066 1220.75 L85.1818 1220.75 L85.1818 1214.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M104.279 1216.81 L120.598 1216.81 L120.598 1220.75 L98.6539 1220.75 L98.6539 1216.81 Q101.316 1214.06 105.899 1209.43 Q110.506 1204.78 111.686 1203.43 Q113.932 1200.91 114.811 1199.17 Q115.714 1197.41 115.714 1195.73 Q115.714 1192.97 113.77 1191.23 Q111.848 1189.5 108.746 1189.5 Q106.547 1189.5 104.094 1190.26 Q101.663 1191.03 98.8854 1192.58 L98.8854 1187.85 Q101.709 1186.72 104.163 1186.14 Q106.617 1185.56 108.654 1185.56 Q114.024 1185.56 117.219 1188.25 Q120.413 1190.93 120.413 1195.42 Q120.413 1197.55 119.603 1199.48 Q118.816 1201.37 116.709 1203.97 Q116.131 1204.64 113.029 1207.85 Q109.927 1211.05 104.279 1216.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M62.9365 869.211 Q59.3254 869.211 57.4967 872.775 Q55.6912 876.317 55.6912 883.447 Q55.6912 890.553 57.4967 894.118 Q59.3254 897.66 62.9365 897.66 Q66.5707 897.66 68.3763 894.118 Q70.205 890.553 70.205 883.447 Q70.205 876.317 68.3763 872.775 Q66.5707 869.211 62.9365 869.211 M62.9365 865.507 Q68.7467 865.507 71.8022 870.113 Q74.8809 874.697 74.8809 883.447 Q74.8809 892.174 71.8022 896.78 Q68.7467 901.363 62.9365 901.363 Q57.1264 901.363 54.0477 896.78 Q50.9921 892.174 50.9921 883.447 Q50.9921 874.697 54.0477 870.113 Q57.1264 865.507 62.9365 865.507 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M83.0984 894.812 L87.9827 894.812 L87.9827 900.692 L83.0984 900.692 L83.0984 894.812 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M111.015 870.206 L99.2095 888.655 L111.015 888.655 L111.015 870.206 M109.788 866.132 L115.668 866.132 L115.668 888.655 L120.598 888.655 L120.598 892.544 L115.668 892.544 L115.668 900.692 L111.015 900.692 L111.015 892.544 L95.4132 892.544 L95.4132 888.03 L109.788 866.132 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M63.2606 549.155 Q59.6495 549.155 57.8208 552.719 Q56.0152 556.261 56.0152 563.391 Q56.0152 570.497 57.8208 574.062 Q59.6495 577.604 63.2606 577.604 Q66.8948 577.604 68.7004 574.062 Q70.5291 570.497 70.5291 563.391 Q70.5291 556.261 68.7004 552.719 Q66.8948 549.155 63.2606 549.155 M63.2606 545.451 Q69.0707 545.451 72.1263 550.057 Q75.205 554.641 75.205 563.391 Q75.205 572.117 72.1263 576.724 Q69.0707 581.307 63.2606 581.307 Q57.4504 581.307 54.3717 576.724 Q51.3162 572.117 51.3162 563.391 Q51.3162 554.641 54.3717 550.057 Q57.4504 545.451 63.2606 545.451 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M83.4225 574.756 L88.3067 574.756 L88.3067 580.636 L83.4225 580.636 L83.4225 574.756 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M109.071 561.493 Q105.922 561.493 104.071 563.645 Q102.242 565.798 102.242 569.548 Q102.242 573.275 104.071 575.451 Q105.922 577.604 109.071 577.604 Q112.219 577.604 114.047 575.451 Q115.899 573.275 115.899 569.548 Q115.899 565.798 114.047 563.645 Q112.219 561.493 109.071 561.493 M118.353 546.84 L118.353 551.099 Q116.594 550.266 114.788 549.826 Q113.006 549.386 111.246 549.386 Q106.617 549.386 104.163 552.511 Q101.733 555.636 101.385 561.955 Q102.751 559.942 104.811 558.877 Q106.871 557.789 109.348 557.789 Q114.557 557.789 117.566 560.96 Q120.598 564.108 120.598 569.548 Q120.598 574.872 117.45 578.09 Q114.302 581.307 109.071 581.307 Q103.075 581.307 99.9039 576.724 Q96.7326 572.117 96.7326 563.391 Q96.7326 555.196 100.621 550.335 Q104.51 545.451 111.061 545.451 Q112.82 545.451 114.603 545.798 Q116.408 546.145 118.353 546.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M63.5152 229.099 Q59.9041 229.099 58.0754 232.663 Q56.2699 236.205 56.2699 243.335 Q56.2699 250.441 58.0754 254.006 Q59.9041 257.547 63.5152 257.547 Q67.1494 257.547 68.955 254.006 Q70.7837 250.441 70.7837 243.335 Q70.7837 236.205 68.955 232.663 Q67.1494 229.099 63.5152 229.099 M63.5152 225.395 Q69.3254 225.395 72.3809 230.001 Q75.4596 234.585 75.4596 243.335 Q75.4596 252.061 72.3809 256.668 Q69.3254 261.251 63.5152 261.251 Q57.7051 261.251 54.6264 256.668 Q51.5708 252.061 51.5708 243.335 Q51.5708 234.585 54.6264 230.001 Q57.7051 225.395 63.5152 225.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M83.6771 254.7 L88.5614 254.7 L88.5614 260.58 L83.6771 260.58 L83.6771 254.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip490)" d="M108.746 244.168 Q105.413 244.168 103.492 245.95 Q101.594 247.733 101.594 250.858 Q101.594 253.983 103.492 255.765 Q105.413 257.547 108.746 257.547 Q112.08 257.547 114.001 255.765 Q115.922 253.96 115.922 250.858 Q115.922 247.733 114.001 245.95 Q112.103 244.168 108.746 244.168 M104.071 242.177 Q101.061 241.436 99.3715 239.376 Q97.7048 237.316 97.7048 234.353 Q97.7048 230.21 100.645 227.802 Q103.608 225.395 108.746 225.395 Q113.908 225.395 116.848 227.802 Q119.788 230.21 119.788 234.353 Q119.788 237.316 118.098 239.376 Q116.432 241.436 113.445 242.177 Q116.825 242.964 118.7 245.256 Q120.598 247.548 120.598 250.858 Q120.598 255.881 117.52 258.566 Q114.464 261.251 108.746 261.251 Q103.029 261.251 99.9502 258.566 Q96.8947 255.881 96.8947 250.858 Q96.8947 247.548 98.7928 245.256 Q100.691 242.964 104.071 242.177 M102.358 234.793 Q102.358 237.478 104.024 238.983 Q105.714 240.487 108.746 240.487 Q111.756 240.487 113.445 238.983 Q115.158 237.478 115.158 234.793 Q115.158 232.108 113.445 230.603 Q111.756 229.099 108.746 229.099 Q105.714 229.099 104.024 230.603 Q102.358 232.108 102.358 234.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip492)" cx="900.754" cy="117.213" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="1337.66" cy="998.336" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="2288.96" cy="1030.31" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="1840.64" cy="87.9763" r="21.6" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="218.754" cy="1410.69" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="768.03" cy="1026.75" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="2017.02" cy="967.646" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="2283.71" cy="1087.25" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="1955.74" cy="1445.72" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="257.146" cy="888.562" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="1413.3" cy="296.531" r="36" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip492)" cx="611.082" cy="118.991" r="28.8" fill="#0000ff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip492)" d="M602.682 154.046 L602.682 226.046 L674.682 226.046 L674.682 154.046 L602.682 154.046 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip492)" d="M1946.4 748.137 L1946.4 820.137 L2018.4 820.137 L2018.4 748.137 L1946.4 748.137 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip492)" d="M998.777 723.237 L998.777 763.237 L1038.78 763.237 L1038.78 723.237 L998.777 723.237 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip492)" d="M798.323 1246.75 L798.323 1310.75 L862.323 1310.75 L862.323 1246.75 L798.323 1246.75 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
+<path clip-path="url(#clip492)" d="M2262.6 872.347 L2262.6 928.347 L2318.6 928.347 L2318.6 872.347 L2262.6 872.347 Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#ff0000" stroke-opacity="1" stroke-width="6.4"/>
 </svg>
diff --git a/previews/PR3562/tutorials/linear/facility_location/index.html b/previews/PR3562/tutorials/linear/facility_location/index.html
index c37f1aeac03..9ab80a32ebd 100644
--- a/previews/PR3562/tutorials/linear/facility_location/index.html
+++ b/previews/PR3562/tutorials/linear/facility_location/index.html
@@ -51,7 +51,7 @@
     markersize = 6,
     markerstrokecolor = :red,
     markerstrokewidth = 2,
-)</code></pre><img src="d825a2fa.svg" alt="Example block output"/><h3 id="JuMP-implementation"><a class="docs-heading-anchor" href="#JuMP-implementation">JuMP implementation</a><a id="JuMP-implementation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-implementation" title="Permalink"></a></h3><p>Create a JuMP model</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
+)</code></pre><img src="ca63c86d.svg" alt="Example block output"/><h3 id="JuMP-implementation"><a class="docs-heading-anchor" href="#JuMP-implementation">JuMP implementation</a><a id="JuMP-implementation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-implementation" title="Permalink"></a></h3><p>Create a JuMP model</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
 set_silent(model)
 @variable(model, y[1:n], Bin);
 @variable(model, x[1:m, 1:n], Bin);
@@ -93,7 +93,7 @@
     end
 end
 
-p</code></pre><img src="528bd41e.svg" alt="Example block output"/><h2 id="Capacitated-facility-location"><a class="docs-heading-anchor" href="#Capacitated-facility-location">Capacitated facility location</a><a id="Capacitated-facility-location-1"></a><a class="docs-heading-anchor-permalink" href="#Capacitated-facility-location" title="Permalink"></a></h2><h3 id="Problem-formulation"><a class="docs-heading-anchor" href="#Problem-formulation">Problem formulation</a><a id="Problem-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#Problem-formulation" title="Permalink"></a></h3><p>The capacitated variant introduces a capacity constraint on each facility, that is, clients have a certain level of demand to be served, while each facility only has finite capacity which cannot be exceeded.</p><p>Specifically,</p><ul><li>The demand of client <span>$i$</span> is denoted by <span>$a_{i} \geq 0$</span></li><li>The capacity of facility <span>$j$</span> is denoted by <span>$q_{j} \geq 0$</span></li></ul><p>The capacity constraints then write</p><p class="math-container">\[\begin{aligned}
+p</code></pre><img src="492bb32b.svg" alt="Example block output"/><h2 id="Capacitated-facility-location"><a class="docs-heading-anchor" href="#Capacitated-facility-location">Capacitated facility location</a><a id="Capacitated-facility-location-1"></a><a class="docs-heading-anchor-permalink" href="#Capacitated-facility-location" title="Permalink"></a></h2><h3 id="Problem-formulation"><a class="docs-heading-anchor" href="#Problem-formulation">Problem formulation</a><a id="Problem-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#Problem-formulation" title="Permalink"></a></h3><p>The capacitated variant introduces a capacity constraint on each facility, that is, clients have a certain level of demand to be served, while each facility only has finite capacity which cannot be exceeded.</p><p>Specifically,</p><ul><li>The demand of client <span>$i$</span> is denoted by <span>$a_{i} \geq 0$</span></li><li>The capacity of facility <span>$j$</span> is denoted by <span>$q_{j} \geq 0$</span></li></ul><p>The capacity constraints then write</p><p class="math-container">\[\begin{aligned}
 \sum_{i} a_{i} x_{i, j} &amp;\leq q_{j} y_{j} &amp;&amp; \forall j \in N
 \end{aligned}\]</p><p>Note that, if <span>$y_{j}$</span> is set to <span>$0$</span>, the capacity constraint above automatically forces <span>$x_{i, j}$</span> to <span>$0$</span>.</p><p>Thus, the capacitated facility location can be formulated as follows</p><p class="math-container">\[\begin{aligned}
 \min_{x, y} \ \ \ &amp;
@@ -125,7 +125,7 @@
     markersize = q,
     markerstrokecolor = :red,
     markerstrokewidth = 2,
-)</code></pre><img src="7f2d6f44.svg" alt="Example block output"/><h3 id="JuMP-implementation-2"><a class="docs-heading-anchor" href="#JuMP-implementation-2">JuMP implementation</a><a class="docs-heading-anchor-permalink" href="#JuMP-implementation-2" title="Permalink"></a></h3><p>Create a JuMP model</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
+)</code></pre><img src="de2fb0ef.svg" alt="Example block output"/><h3 id="JuMP-implementation-2"><a class="docs-heading-anchor" href="#JuMP-implementation-2">JuMP implementation</a><a class="docs-heading-anchor-permalink" href="#JuMP-implementation-2" title="Permalink"></a></h3><p>Create a JuMP model</p><pre><code class="language-julia hljs">model = Model(HiGHS.Optimizer)
 set_silent(model)
 @variable(model, y[1:n], Bin);
 @variable(model, x[1:m, 1:n], Bin);
@@ -167,4 +167,4 @@
     end
 end
 
-p</code></pre><img src="0d89ccc9.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and tricks</a><a class="docs-footer-nextpage" href="../finance/">Financial modeling problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+p</code></pre><img src="96afa50f.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and tricks</a><a class="docs-footer-nextpage" href="../finance/">Financial modeling problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/factory_schedule/9faf22a9.svg b/previews/PR3562/tutorials/linear/factory_schedule/22aa7ce9.svg
similarity index 79%
rename from previews/PR3562/tutorials/linear/factory_schedule/9faf22a9.svg
rename to previews/PR3562/tutorials/linear/factory_schedule/22aa7ce9.svg
index 506ef07135a..c8f3ec8db50 100644
--- a/previews/PR3562/tutorials/linear/factory_schedule/9faf22a9.svg
+++ b/previews/PR3562/tutorials/linear/factory_schedule/22aa7ce9.svg
@@ -1,162 +1,162 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip790">
+  <clipPath id="clip610">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip790)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip610)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip791">
+  <clipPath id="clip611">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip790)" d="M369.518 1423.18 L2352.76 1423.18 L2352.76 47.2441 L369.518 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip610)" d="M369.518 1423.18 L2352.76 1423.18 L2352.76 47.2441 L369.518 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip792">
+  <clipPath id="clip612">
     <rect x="369" y="47" width="1984" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="388.85,1423.18 388.85,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="688.015,1423.18 688.015,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="987.181,1423.18 987.181,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1286.35,1423.18 1286.35,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1585.51,1423.18 1585.51,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1884.68,1423.18 1884.68,47.2441 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2183.84,1423.18 2183.84,47.2441 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="388.85,1423.18 388.85,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="688.015,1423.18 688.015,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="987.181,1423.18 987.181,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.35,1423.18 1286.35,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1585.51,1423.18 1585.51,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1884.68,1423.18 1884.68,1404.28 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2183.84,1423.18 2183.84,1404.28 "/>
-<path clip-path="url(#clip790)" d="M388.85 1454.1 Q385.239 1454.1 383.411 1457.66 Q381.605 1461.2 381.605 1468.33 Q381.605 1475.44 383.411 1479.01 Q385.239 1482.55 388.85 1482.55 Q392.485 1482.55 394.29 1479.01 Q396.119 1475.44 396.119 1468.33 Q396.119 1461.2 394.29 1457.66 Q392.485 1454.1 388.85 1454.1 M388.85 1450.39 Q394.66 1450.39 397.716 1455 Q400.795 1459.58 400.795 1468.33 Q400.795 1477.06 397.716 1481.67 Q394.66 1486.25 388.85 1486.25 Q383.04 1486.25 379.961 1481.67 Q376.906 1477.06 376.906 1468.33 Q376.906 1459.58 379.961 1455 Q383.04 1450.39 388.85 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M682.668 1481.64 L698.988 1481.64 L698.988 1485.58 L677.043 1485.58 L677.043 1481.64 Q679.705 1478.89 684.289 1474.26 Q688.895 1469.61 690.076 1468.27 Q692.321 1465.74 693.201 1464.01 Q694.103 1462.25 694.103 1460.56 Q694.103 1457.8 692.159 1456.07 Q690.238 1454.33 687.136 1454.33 Q684.937 1454.33 682.483 1455.09 Q680.053 1455.86 677.275 1457.41 L677.275 1452.69 Q680.099 1451.55 682.553 1450.97 Q685.006 1450.39 687.043 1450.39 Q692.414 1450.39 695.608 1453.08 Q698.802 1455.77 698.802 1460.26 Q698.802 1462.39 697.992 1464.31 Q697.205 1466.2 695.099 1468.8 Q694.52 1469.47 691.418 1472.69 Q688.316 1475.88 682.668 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M990.19 1455.09 L978.384 1473.54 L990.19 1473.54 L990.19 1455.09 M988.963 1451.02 L994.843 1451.02 L994.843 1473.54 L999.773 1473.54 L999.773 1477.43 L994.843 1477.43 L994.843 1485.58 L990.19 1485.58 L990.19 1477.43 L974.588 1477.43 L974.588 1472.92 L988.963 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1286.75 1466.44 Q1283.6 1466.44 1281.75 1468.59 Q1279.92 1470.74 1279.92 1474.49 Q1279.92 1478.22 1281.75 1480.39 Q1283.6 1482.55 1286.75 1482.55 Q1289.9 1482.55 1291.73 1480.39 Q1293.58 1478.22 1293.58 1474.49 Q1293.58 1470.74 1291.73 1468.59 Q1289.9 1466.44 1286.75 1466.44 M1296.03 1451.78 L1296.03 1456.04 Q1294.27 1455.21 1292.47 1454.77 Q1290.69 1454.33 1288.93 1454.33 Q1284.3 1454.33 1281.84 1457.45 Q1279.41 1460.58 1279.07 1466.9 Q1280.43 1464.89 1282.49 1463.82 Q1284.55 1462.73 1287.03 1462.73 Q1292.24 1462.73 1295.25 1465.9 Q1298.28 1469.05 1298.28 1474.49 Q1298.28 1479.82 1295.13 1483.03 Q1291.98 1486.25 1286.75 1486.25 Q1280.76 1486.25 1277.58 1481.67 Q1274.41 1477.06 1274.41 1468.33 Q1274.41 1460.14 1278.3 1455.28 Q1282.19 1450.39 1288.74 1450.39 Q1290.5 1450.39 1292.28 1450.74 Q1294.09 1451.09 1296.03 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1585.51 1469.17 Q1582.18 1469.17 1580.26 1470.95 Q1578.36 1472.73 1578.36 1475.86 Q1578.36 1478.98 1580.26 1480.77 Q1582.18 1482.55 1585.51 1482.55 Q1588.84 1482.55 1590.77 1480.77 Q1592.69 1478.96 1592.69 1475.86 Q1592.69 1472.73 1590.77 1470.95 Q1588.87 1469.17 1585.51 1469.17 M1580.84 1467.18 Q1577.83 1466.44 1576.14 1464.38 Q1574.47 1462.32 1574.47 1459.35 Q1574.47 1455.21 1577.41 1452.8 Q1580.37 1450.39 1585.51 1450.39 Q1590.67 1450.39 1593.61 1452.8 Q1596.55 1455.21 1596.55 1459.35 Q1596.55 1462.32 1594.86 1464.38 Q1593.2 1466.44 1590.21 1467.18 Q1593.59 1467.96 1595.46 1470.26 Q1597.36 1472.55 1597.36 1475.86 Q1597.36 1480.88 1594.28 1483.57 Q1591.23 1486.25 1585.51 1486.25 Q1579.79 1486.25 1576.71 1483.57 Q1573.66 1480.88 1573.66 1475.86 Q1573.66 1472.55 1575.56 1470.26 Q1577.46 1467.96 1580.84 1467.18 M1579.12 1459.79 Q1579.12 1462.48 1580.79 1463.98 Q1582.48 1465.49 1585.51 1465.49 Q1588.52 1465.49 1590.21 1463.98 Q1591.92 1462.48 1591.92 1459.79 Q1591.92 1457.11 1590.21 1455.6 Q1588.52 1454.1 1585.51 1454.1 Q1582.48 1454.1 1580.79 1455.6 Q1579.12 1457.11 1579.12 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1859.36 1481.64 L1867 1481.64 L1867 1455.28 L1858.69 1456.95 L1858.69 1452.69 L1866.96 1451.02 L1871.63 1451.02 L1871.63 1481.64 L1879.27 1481.64 L1879.27 1485.58 L1859.36 1485.58 L1859.36 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1898.72 1454.1 Q1895.1 1454.1 1893.28 1457.66 Q1891.47 1461.2 1891.47 1468.33 Q1891.47 1475.44 1893.28 1479.01 Q1895.1 1482.55 1898.72 1482.55 Q1902.35 1482.55 1904.16 1479.01 Q1905.98 1475.44 1905.98 1468.33 Q1905.98 1461.2 1904.16 1457.66 Q1902.35 1454.1 1898.72 1454.1 M1898.72 1450.39 Q1904.53 1450.39 1907.58 1455 Q1910.66 1459.58 1910.66 1468.33 Q1910.66 1477.06 1907.58 1481.67 Q1904.53 1486.25 1898.72 1486.25 Q1892.91 1486.25 1889.83 1481.67 Q1886.77 1477.06 1886.77 1468.33 Q1886.77 1459.58 1889.83 1455 Q1892.91 1450.39 1898.72 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M2159.33 1481.64 L2166.97 1481.64 L2166.97 1455.28 L2158.66 1456.95 L2158.66 1452.69 L2166.92 1451.02 L2171.6 1451.02 L2171.6 1481.64 L2179.23 1481.64 L2179.23 1485.58 L2159.33 1485.58 L2159.33 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M2192.71 1481.64 L2209.03 1481.64 L2209.03 1485.58 L2187.08 1485.58 L2187.08 1481.64 Q2189.74 1478.89 2194.33 1474.26 Q2198.93 1469.61 2200.11 1468.27 Q2202.36 1465.74 2203.24 1464.01 Q2204.14 1462.25 2204.14 1460.56 Q2204.14 1457.8 2202.2 1456.07 Q2200.28 1454.33 2197.17 1454.33 Q2194.98 1454.33 2192.52 1455.09 Q2190.09 1455.86 2187.31 1457.41 L2187.31 1452.69 Q2190.14 1451.55 2192.59 1450.97 Q2195.04 1450.39 2197.08 1450.39 Q2202.45 1450.39 2205.65 1453.08 Q2208.84 1455.77 2208.84 1460.26 Q2208.84 1462.39 2208.03 1464.31 Q2207.24 1466.2 2205.14 1468.8 Q2204.56 1469.47 2201.46 1472.69 Q2198.36 1475.88 2192.71 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1264.95 1520.52 L1274.53 1520.52 L1286.66 1552.86 L1298.85 1520.52 L1308.43 1520.52 L1308.43 1568.04 L1302.16 1568.04 L1302.16 1526.32 L1289.9 1558.91 L1283.44 1558.91 L1271.19 1526.32 L1271.19 1568.04 L1264.95 1568.04 L1264.95 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1334.75 1536.5 Q1330.04 1536.5 1327.3 1540.19 Q1324.57 1543.85 1324.57 1550.25 Q1324.57 1556.65 1327.27 1560.34 Q1330.01 1564 1334.75 1564 Q1339.43 1564 1342.17 1560.31 Q1344.9 1556.62 1344.9 1550.25 Q1344.9 1543.92 1342.17 1540.23 Q1339.43 1536.5 1334.75 1536.5 M1334.75 1531.54 Q1342.39 1531.54 1346.75 1536.5 Q1351.11 1541.47 1351.11 1550.25 Q1351.11 1559 1346.75 1564 Q1342.39 1568.97 1334.75 1568.97 Q1327.08 1568.97 1322.72 1564 Q1318.39 1559 1318.39 1550.25 Q1318.39 1541.47 1322.72 1536.5 Q1327.08 1531.54 1334.75 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1390.45 1546.53 L1390.45 1568.04 L1384.59 1568.04 L1384.59 1546.72 Q1384.59 1541.66 1382.62 1539.14 Q1380.65 1536.63 1376.7 1536.63 Q1371.96 1536.63 1369.22 1539.65 Q1366.48 1542.68 1366.48 1547.9 L1366.48 1568.04 L1360.6 1568.04 L1360.6 1532.4 L1366.48 1532.4 L1366.48 1537.93 Q1368.58 1534.72 1371.42 1533.13 Q1374.28 1531.54 1378.01 1531.54 Q1384.15 1531.54 1387.3 1535.36 Q1390.45 1539.14 1390.45 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1407.93 1522.27 L1407.93 1532.4 L1419.99 1532.4 L1419.99 1536.95 L1407.93 1536.95 L1407.93 1556.3 Q1407.93 1560.66 1409.1 1561.9 Q1410.31 1563.14 1413.97 1563.14 L1419.99 1563.14 L1419.99 1568.04 L1413.97 1568.04 Q1407.19 1568.04 1404.61 1565.53 Q1402.04 1562.98 1402.04 1556.3 L1402.04 1536.95 L1397.74 1536.95 L1397.74 1532.4 L1402.04 1532.4 L1402.04 1522.27 L1407.93 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1457.32 1546.53 L1457.32 1568.04 L1451.47 1568.04 L1451.47 1546.72 Q1451.47 1541.66 1449.49 1539.14 Q1447.52 1536.63 1443.57 1536.63 Q1438.83 1536.63 1436.09 1539.65 Q1433.36 1542.68 1433.36 1547.9 L1433.36 1568.04 L1427.47 1568.04 L1427.47 1518.52 L1433.36 1518.52 L1433.36 1537.93 Q1435.46 1534.72 1438.29 1533.13 Q1441.15 1531.54 1444.88 1531.54 Q1451.02 1531.54 1454.17 1535.36 Q1457.32 1539.14 1457.32 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,1079.2 2352.76,1079.2 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,735.212 2352.76,735.212 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,391.228 2352.76,391.228 "/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,47.2441 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 369.518,47.2441 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 388.416,1423.18 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1079.2 388.416,1079.2 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,735.212 388.416,735.212 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,391.228 388.416,391.228 "/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,47.2441 388.416,47.2441 "/>
-<path clip-path="url(#clip790)" d="M321.574 1408.98 Q317.963 1408.98 316.134 1412.54 Q314.329 1416.08 314.329 1423.21 Q314.329 1430.32 316.134 1433.89 Q317.963 1437.43 321.574 1437.43 Q325.208 1437.43 327.014 1433.89 Q328.842 1430.32 328.842 1423.21 Q328.842 1416.08 327.014 1412.54 Q325.208 1408.98 321.574 1408.98 M321.574 1405.27 Q327.384 1405.27 330.44 1409.88 Q333.518 1414.46 333.518 1423.21 Q333.518 1431.94 330.44 1436.55 Q327.384 1441.13 321.574 1441.13 Q315.764 1441.13 312.685 1436.55 Q309.629 1431.94 309.629 1423.21 Q309.629 1414.46 312.685 1409.88 Q315.764 1405.27 321.574 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M115.718 1068.36 L134.075 1068.36 L134.075 1072.3 L120.001 1072.3 L120.001 1080.77 Q121.019 1080.42 122.038 1080.26 Q123.056 1080.08 124.075 1080.08 Q129.862 1080.08 133.242 1083.25 Q136.621 1086.42 136.621 1091.84 Q136.621 1097.41 133.149 1100.52 Q129.677 1103.59 123.357 1103.59 Q121.181 1103.59 118.913 1103.22 Q116.668 1102.85 114.26 1102.11 L114.26 1097.41 Q116.343 1098.55 118.566 1099.1 Q120.788 1099.66 123.265 1099.66 Q127.269 1099.66 129.607 1097.55 Q131.945 1095.45 131.945 1091.84 Q131.945 1088.22 129.607 1086.12 Q127.269 1084.01 123.265 1084.01 Q121.39 1084.01 119.515 1084.43 Q117.663 1084.84 115.718 1085.72 L115.718 1068.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M145.834 1097.04 L150.718 1097.04 L150.718 1102.92 L145.834 1102.92 L145.834 1097.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M170.903 1071.44 Q167.292 1071.44 165.464 1075.01 Q163.658 1078.55 163.658 1085.68 Q163.658 1092.78 165.464 1096.35 Q167.292 1099.89 170.903 1099.89 Q174.538 1099.89 176.343 1096.35 Q178.172 1092.78 178.172 1085.68 Q178.172 1078.55 176.343 1075.01 Q174.538 1071.44 170.903 1071.44 M170.903 1067.74 Q176.714 1067.74 179.769 1072.34 Q182.848 1076.93 182.848 1085.68 Q182.848 1094.41 179.769 1099.01 Q176.714 1103.59 170.903 1103.59 Q165.093 1103.59 162.015 1099.01 Q158.959 1094.41 158.959 1085.68 Q158.959 1076.93 162.015 1072.34 Q165.093 1067.74 170.903 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M201.065 1071.44 Q197.454 1071.44 195.625 1075.01 Q193.82 1078.55 193.82 1085.68 Q193.82 1092.78 195.625 1096.35 Q197.454 1099.89 201.065 1099.89 Q204.699 1099.89 206.505 1096.35 Q208.334 1092.78 208.334 1085.68 Q208.334 1078.55 206.505 1075.01 Q204.699 1071.44 201.065 1071.44 M201.065 1067.74 Q206.875 1067.74 209.931 1072.34 Q213.01 1076.93 213.01 1085.68 Q213.01 1094.41 209.931 1099.01 Q206.875 1103.59 201.065 1103.59 Q195.255 1103.59 192.176 1099.01 Q189.121 1094.41 189.121 1085.68 Q189.121 1076.93 192.176 1072.34 Q195.255 1067.74 201.065 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M249.398 1077.46 L238.82 1088.09 L249.398 1098.66 L246.644 1101.47 L236.019 1090.84 L225.394 1101.47 L222.662 1098.66 L233.218 1088.09 L222.662 1077.46 L225.394 1074.66 L236.019 1085.28 L246.644 1074.66 L249.398 1077.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M261.759 1098.99 L269.398 1098.99 L269.398 1072.62 L261.088 1074.29 L261.088 1070.03 L269.352 1068.36 L274.028 1068.36 L274.028 1098.99 L281.667 1098.99 L281.667 1102.92 L261.759 1102.92 L261.759 1098.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M301.111 1071.44 Q297.5 1071.44 295.671 1075.01 Q293.866 1078.55 293.866 1085.68 Q293.866 1092.78 295.671 1096.35 Q297.5 1099.89 301.111 1099.89 Q304.745 1099.89 306.551 1096.35 Q308.379 1092.78 308.379 1085.68 Q308.379 1078.55 306.551 1075.01 Q304.745 1071.44 301.111 1071.44 M301.111 1067.74 Q306.921 1067.74 309.977 1072.34 Q313.055 1076.93 313.055 1085.68 Q313.055 1094.41 309.977 1099.01 Q306.921 1103.59 301.111 1103.59 Q295.301 1103.59 292.222 1099.01 Q289.167 1094.41 289.167 1085.68 Q289.167 1076.93 292.222 1072.34 Q295.301 1067.74 301.111 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M325.732 1050.74 L316.14 1065.73 L325.732 1065.73 L325.732 1050.74 M324.735 1047.43 L329.512 1047.43 L329.512 1065.73 L333.518 1065.73 L333.518 1068.89 L329.512 1068.89 L329.512 1075.51 L325.732 1075.51 L325.732 1068.89 L313.055 1068.89 L313.055 1065.23 L324.735 1047.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M118.777 755.004 L126.416 755.004 L126.416 728.639 L118.106 730.305 L118.106 726.046 L126.369 724.38 L131.045 724.38 L131.045 755.004 L138.684 755.004 L138.684 758.94 L118.777 758.94 L118.777 755.004 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M148.129 753.06 L153.013 753.06 L153.013 758.94 L148.129 758.94 L148.129 753.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M173.198 727.458 Q169.587 727.458 167.758 731.023 Q165.953 734.565 165.953 741.694 Q165.953 748.801 167.758 752.366 Q169.587 755.907 173.198 755.907 Q176.832 755.907 178.638 752.366 Q180.466 748.801 180.466 741.694 Q180.466 734.565 178.638 731.023 Q176.832 727.458 173.198 727.458 M173.198 723.755 Q179.008 723.755 182.064 728.361 Q185.142 732.944 185.142 741.694 Q185.142 750.421 182.064 755.028 Q179.008 759.611 173.198 759.611 Q167.388 759.611 164.309 755.028 Q161.254 750.421 161.254 741.694 Q161.254 732.944 164.309 728.361 Q167.388 723.755 173.198 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M203.36 727.458 Q199.749 727.458 197.92 731.023 Q196.114 734.565 196.114 741.694 Q196.114 748.801 197.92 752.366 Q199.749 755.907 203.36 755.907 Q206.994 755.907 208.8 752.366 Q210.628 748.801 210.628 741.694 Q210.628 734.565 208.8 731.023 Q206.994 727.458 203.36 727.458 M203.36 723.755 Q209.17 723.755 212.225 728.361 Q215.304 732.944 215.304 741.694 Q215.304 750.421 212.225 755.028 Q209.17 759.611 203.36 759.611 Q197.55 759.611 194.471 755.028 Q191.415 750.421 191.415 741.694 Q191.415 732.944 194.471 728.361 Q197.55 723.755 203.36 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M251.693 733.477 L241.114 744.102 L251.693 754.68 L248.938 757.481 L238.313 746.856 L227.688 757.481 L224.957 754.68 L235.512 744.102 L224.957 733.477 L227.688 730.676 L238.313 741.301 L248.938 730.676 L251.693 733.477 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M264.054 755.004 L271.693 755.004 L271.693 728.639 L263.383 730.305 L263.383 726.046 L271.646 724.38 L276.322 724.38 L276.322 755.004 L283.961 755.004 L283.961 758.94 L264.054 758.94 L264.054 755.004 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M303.406 727.458 Q299.794 727.458 297.966 731.023 Q296.16 734.565 296.16 741.694 Q296.16 748.801 297.966 752.366 Q299.794 755.907 303.406 755.907 Q307.04 755.907 308.845 752.366 Q310.674 748.801 310.674 741.694 Q310.674 734.565 308.845 731.023 Q307.04 727.458 303.406 727.458 M303.406 723.755 Q309.216 723.755 312.271 728.361 Q315.35 732.944 315.35 741.694 Q315.35 750.421 312.271 755.028 Q309.216 759.611 303.406 759.611 Q297.595 759.611 294.517 755.028 Q291.461 750.421 291.461 741.694 Q291.461 732.944 294.517 728.361 Q297.595 723.755 303.406 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M316.535 703.449 L331.449 703.449 L331.449 706.646 L320.014 706.646 L320.014 713.53 Q320.842 713.248 321.669 713.116 Q322.497 712.966 323.324 712.966 Q328.026 712.966 330.772 715.543 Q333.518 718.119 333.518 722.52 Q333.518 727.053 330.697 729.573 Q327.876 732.075 322.741 732.075 Q320.973 732.075 319.13 731.774 Q317.306 731.473 315.35 730.871 L315.35 727.053 Q317.043 727.975 318.848 728.426 Q320.654 728.877 322.666 728.877 Q325.92 728.877 327.819 727.166 Q329.719 725.454 329.719 722.52 Q329.719 719.586 327.819 717.875 Q325.92 716.163 322.666 716.163 Q321.143 716.163 319.619 716.502 Q318.115 716.84 316.535 717.555 L316.535 703.449 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M118.777 411.02 L126.416 411.02 L126.416 384.655 L118.106 386.322 L118.106 382.062 L126.369 380.396 L131.045 380.396 L131.045 411.02 L138.684 411.02 L138.684 414.956 L118.777 414.956 L118.777 411.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M148.129 409.076 L153.013 409.076 L153.013 414.956 L148.129 414.956 L148.129 409.076 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M163.244 380.396 L181.601 380.396 L181.601 384.331 L167.527 384.331 L167.527 392.803 Q168.545 392.456 169.564 392.294 Q170.582 392.109 171.601 392.109 Q177.388 392.109 180.767 395.28 Q184.147 398.451 184.147 403.868 Q184.147 409.446 180.675 412.548 Q177.203 415.627 170.883 415.627 Q168.707 415.627 166.439 415.257 Q164.193 414.886 161.786 414.145 L161.786 409.446 Q163.869 410.581 166.091 411.136 Q168.314 411.692 170.791 411.692 Q174.795 411.692 177.133 409.585 Q179.471 407.479 179.471 403.868 Q179.471 400.257 177.133 398.15 Q174.795 396.044 170.791 396.044 Q168.916 396.044 167.041 396.46 Q165.189 396.877 163.244 397.757 L163.244 380.396 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M203.36 383.474 Q199.749 383.474 197.92 387.039 Q196.114 390.581 196.114 397.71 Q196.114 404.817 197.92 408.382 Q199.749 411.923 203.36 411.923 Q206.994 411.923 208.8 408.382 Q210.628 404.817 210.628 397.71 Q210.628 390.581 208.8 387.039 Q206.994 383.474 203.36 383.474 M203.36 379.771 Q209.17 379.771 212.225 384.377 Q215.304 388.96 215.304 397.71 Q215.304 406.437 212.225 411.044 Q209.17 415.627 203.36 415.627 Q197.55 415.627 194.471 411.044 Q191.415 406.437 191.415 397.71 Q191.415 388.96 194.471 384.377 Q197.55 379.771 203.36 379.771 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M251.693 389.493 L241.114 400.118 L251.693 410.696 L248.938 413.497 L238.313 402.872 L227.688 413.497 L224.957 410.696 L235.512 400.118 L224.957 389.493 L227.688 386.692 L238.313 397.317 L248.938 386.692 L251.693 389.493 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M264.054 411.02 L271.693 411.02 L271.693 384.655 L263.383 386.322 L263.383 382.062 L271.646 380.396 L276.322 380.396 L276.322 411.02 L283.961 411.02 L283.961 414.956 L264.054 414.956 L264.054 411.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M303.406 383.474 Q299.794 383.474 297.966 387.039 Q296.16 390.581 296.16 397.71 Q296.16 404.817 297.966 408.382 Q299.794 411.923 303.406 411.923 Q307.04 411.923 308.845 408.382 Q310.674 404.817 310.674 397.71 Q310.674 390.581 308.845 387.039 Q307.04 383.474 303.406 383.474 M303.406 379.771 Q309.216 379.771 312.271 384.377 Q315.35 388.96 315.35 397.71 Q315.35 406.437 312.271 411.044 Q309.216 415.627 303.406 415.627 Q297.595 415.627 294.517 411.044 Q291.461 406.437 291.461 397.71 Q291.461 388.96 294.517 384.377 Q297.595 379.771 303.406 379.771 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M316.535 359.465 L331.449 359.465 L331.449 362.663 L320.014 362.663 L320.014 369.546 Q320.842 369.264 321.669 369.132 Q322.497 368.982 323.324 368.982 Q328.026 368.982 330.772 371.559 Q333.518 374.135 333.518 378.536 Q333.518 383.069 330.697 385.589 Q327.876 388.091 322.741 388.091 Q320.973 388.091 319.13 387.79 Q317.306 387.489 315.35 386.887 L315.35 383.069 Q317.043 383.991 318.848 384.442 Q320.654 384.893 322.666 384.893 Q325.92 384.893 327.819 383.182 Q329.719 381.47 329.719 378.536 Q329.719 375.602 327.819 373.891 Q325.92 372.179 322.666 372.179 Q321.143 372.179 319.619 372.518 Q318.115 372.856 316.535 373.571 L316.535 359.465 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M121.994 67.0365 L138.314 67.0365 L138.314 70.9717 L116.37 70.9717 L116.37 67.0365 Q119.032 64.2819 123.615 59.6523 Q128.221 54.9996 129.402 53.657 Q131.647 51.1338 132.527 49.3977 Q133.43 47.6385 133.43 45.9487 Q133.43 43.1941 131.485 41.458 Q129.564 39.7219 126.462 39.7219 Q124.263 39.7219 121.809 40.4857 Q119.379 41.2496 116.601 42.8005 L116.601 38.0784 Q119.425 36.9441 121.879 36.3654 Q124.332 35.7867 126.369 35.7867 Q131.74 35.7867 134.934 38.4719 Q138.129 41.157 138.129 45.6478 Q138.129 47.7774 137.318 49.6987 Q136.531 51.5968 134.425 54.1894 Q133.846 54.8607 130.744 58.0782 Q127.643 61.2727 121.994 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M148.129 65.0921 L153.013 65.0921 L153.013 70.9717 L148.129 70.9717 L148.129 65.0921 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M173.198 39.4904 Q169.587 39.4904 167.758 43.0552 Q165.953 46.5968 165.953 53.7264 Q165.953 60.8329 167.758 64.3977 Q169.587 67.9393 173.198 67.9393 Q176.832 67.9393 178.638 64.3977 Q180.466 60.8329 180.466 53.7264 Q180.466 46.5968 178.638 43.0552 Q176.832 39.4904 173.198 39.4904 M173.198 35.7867 Q179.008 35.7867 182.064 40.3932 Q185.142 44.9765 185.142 53.7264 Q185.142 62.4532 182.064 67.0597 Q179.008 71.643 173.198 71.643 Q167.388 71.643 164.309 67.0597 Q161.254 62.4532 161.254 53.7264 Q161.254 44.9765 164.309 40.3932 Q167.388 35.7867 173.198 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M203.36 39.4904 Q199.749 39.4904 197.92 43.0552 Q196.114 46.5968 196.114 53.7264 Q196.114 60.8329 197.92 64.3977 Q199.749 67.9393 203.36 67.9393 Q206.994 67.9393 208.8 64.3977 Q210.628 60.8329 210.628 53.7264 Q210.628 46.5968 208.8 43.0552 Q206.994 39.4904 203.36 39.4904 M203.36 35.7867 Q209.17 35.7867 212.225 40.3932 Q215.304 44.9765 215.304 53.7264 Q215.304 62.4532 212.225 67.0597 Q209.17 71.643 203.36 71.643 Q197.55 71.643 194.471 67.0597 Q191.415 62.4532 191.415 53.7264 Q191.415 44.9765 194.471 40.3932 Q197.55 35.7867 203.36 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M251.693 45.5089 L241.114 56.1338 L251.693 66.7125 L248.938 69.5134 L238.313 58.8884 L227.688 69.5134 L224.957 66.7125 L235.512 56.1338 L224.957 45.5089 L227.688 42.708 L238.313 53.3329 L248.938 42.708 L251.693 45.5089 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M264.054 67.0365 L271.693 67.0365 L271.693 40.6709 L263.383 42.3376 L263.383 38.0784 L271.646 36.4117 L276.322 36.4117 L276.322 67.0365 L283.961 67.0365 L283.961 70.9717 L264.054 70.9717 L264.054 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M303.406 39.4904 Q299.794 39.4904 297.966 43.0552 Q296.16 46.5968 296.16 53.7264 Q296.16 60.8329 297.966 64.3977 Q299.794 67.9393 303.406 67.9393 Q307.04 67.9393 308.845 64.3977 Q310.674 60.8329 310.674 53.7264 Q310.674 46.5968 308.845 43.0552 Q307.04 39.4904 303.406 39.4904 M303.406 35.7867 Q309.216 35.7867 312.271 40.3932 Q315.35 44.9765 315.35 53.7264 Q315.35 62.4532 312.271 67.0597 Q309.216 71.643 303.406 71.643 Q297.595 71.643 294.517 67.0597 Q291.461 62.4532 291.461 53.7264 Q291.461 44.9765 294.517 40.3932 Q297.595 35.7867 303.406 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M316.535 15.4813 L331.449 15.4813 L331.449 18.6786 L320.014 18.6786 L320.014 25.5623 Q320.842 25.2801 321.669 25.1485 Q322.497 24.998 323.324 24.998 Q328.026 24.998 330.772 27.5747 Q333.518 30.1514 333.518 34.5524 Q333.518 39.085 330.697 41.6053 Q327.876 44.1067 322.741 44.1067 Q320.973 44.1067 319.13 43.8058 Q317.306 43.5049 315.35 42.903 L315.35 39.085 Q317.043 40.0066 318.848 40.458 Q320.654 40.9094 322.666 40.9094 Q325.92 40.9094 327.819 39.1979 Q329.719 37.4864 329.719 34.5524 Q329.719 31.6184 327.819 29.9069 Q325.92 28.1953 322.666 28.1953 Q321.143 28.1953 319.619 28.5339 Q318.115 28.8724 316.535 29.5871 L316.535 15.4813 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M21.7677 896.217 L39.6235 896.217 L39.6235 888.132 Q39.6235 883.644 37.3 881.194 Q34.9765 878.743 30.6797 878.743 Q26.4147 878.743 24.0912 881.194 Q21.7677 883.644 21.7677 888.132 L21.7677 896.217 M16.4842 902.646 L16.4842 888.132 Q16.4842 880.143 20.1126 876.069 Q23.7092 871.963 30.6797 871.963 Q37.7138 871.963 41.3104 876.069 Q44.907 880.143 44.907 888.132 L44.907 896.217 L64.0042 896.217 L64.0042 902.646 L16.4842 902.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M33.8307 844.081 Q33.2578 845.068 33.0032 846.246 Q32.7167 847.392 32.7167 848.792 Q32.7167 853.757 35.9632 856.431 Q39.1779 859.073 45.2253 859.073 L64.0042 859.073 L64.0042 864.961 L28.3562 864.961 L28.3562 859.073 L33.8944 859.073 Q30.6479 857.227 29.0883 854.267 Q27.4968 851.307 27.4968 847.073 Q27.4968 846.469 27.5923 845.737 Q27.656 845.004 27.8151 844.113 L33.8307 844.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M32.4621 825.557 Q32.4621 830.268 36.1542 833.005 Q39.8145 835.742 46.212 835.742 Q52.6095 835.742 56.3017 833.037 Q59.9619 830.3 59.9619 825.557 Q59.9619 820.878 56.2698 818.141 Q52.5777 815.404 46.212 815.404 Q39.8781 815.404 36.186 818.141 Q32.4621 820.878 32.4621 825.557 M27.4968 825.557 Q27.4968 817.918 32.4621 813.558 Q37.4273 809.197 46.212 809.197 Q54.9649 809.197 59.9619 813.558 Q64.9272 817.918 64.9272 825.557 Q64.9272 833.228 59.9619 837.588 Q54.9649 841.917 46.212 841.917 Q37.4273 841.917 32.4621 837.588 Q27.4968 833.228 27.4968 825.557 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M33.7671 776.032 L14.479 776.032 L14.479 770.176 L64.0042 770.176 L64.0042 776.032 L58.657 776.032 Q61.8398 777.878 63.3994 780.711 Q64.9272 783.512 64.9272 787.459 Q64.9272 793.92 59.771 797.994 Q54.6147 802.036 46.212 802.036 Q37.8093 802.036 32.6531 797.994 Q27.4968 793.92 27.4968 787.459 Q27.4968 783.512 29.0564 780.711 Q30.5842 777.878 33.7671 776.032 M46.212 795.989 Q52.6732 795.989 56.3653 793.347 Q60.0256 790.673 60.0256 786.026 Q60.0256 781.379 56.3653 778.706 Q52.6732 776.032 46.212 776.032 Q39.7508 776.032 36.0905 778.706 Q32.3984 781.379 32.3984 786.026 Q32.3984 790.673 36.0905 793.347 Q39.7508 795.989 46.212 795.989 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M49.9359 758.717 L28.3562 758.717 L28.3562 752.861 L49.7131 752.861 Q54.7739 752.861 57.3202 750.888 Q59.8346 748.914 59.8346 744.967 Q59.8346 740.225 56.8109 737.488 Q53.7872 734.719 48.5673 734.719 L28.3562 734.719 L28.3562 728.862 L64.0042 728.862 L64.0042 734.719 L58.5296 734.719 Q61.7762 736.851 63.3676 739.684 Q64.9272 742.485 64.9272 746.209 Q64.9272 752.352 61.1078 755.534 Q57.2883 758.717 49.9359 758.717 M27.4968 743.981 L27.4968 743.981 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M29.7248 691.145 L35.1993 691.145 Q33.8307 693.628 33.1623 696.142 Q32.4621 698.625 32.4621 701.171 Q32.4621 706.869 36.0905 710.02 Q39.6872 713.171 46.212 713.171 Q52.7369 713.171 56.3653 710.02 Q59.9619 706.869 59.9619 701.171 Q59.9619 698.625 59.2935 696.142 Q58.5933 693.628 57.2247 691.145 L62.6355 691.145 Q63.7814 693.596 64.3543 696.238 Q64.9272 698.848 64.9272 701.808 Q64.9272 709.861 59.8664 714.603 Q54.8057 719.345 46.212 719.345 Q37.491 719.345 32.4939 714.571 Q27.4968 709.765 27.4968 701.426 Q27.4968 698.721 28.0697 696.142 Q28.6108 693.564 29.7248 691.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M18.2347 675.167 L28.3562 675.167 L28.3562 663.104 L32.9077 663.104 L32.9077 675.167 L52.2594 675.167 Q56.6199 675.167 57.8613 673.99 Q59.1026 672.78 59.1026 669.12 L59.1026 663.104 L64.0042 663.104 L64.0042 669.12 Q64.0042 675.899 61.4897 678.478 Q58.9434 681.056 52.2594 681.056 L32.9077 681.056 L32.9077 685.353 L28.3562 685.353 L28.3562 681.056 L18.2347 681.056 L18.2347 675.167 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M28.3562 655.402 L28.3562 649.545 L64.0042 649.545 L64.0042 655.402 L28.3562 655.402 M14.479 655.402 L14.479 649.545 L21.895 649.545 L21.895 655.402 L14.479 655.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M32.4621 623.478 Q32.4621 628.189 36.1542 630.926 Q39.8145 633.663 46.212 633.663 Q52.6095 633.663 56.3017 630.958 Q59.9619 628.22 59.9619 623.478 Q59.9619 618.799 56.2698 616.062 Q52.5777 613.325 46.212 613.325 Q39.8781 613.325 36.186 616.062 Q32.4621 618.799 32.4621 623.478 M27.4968 623.478 Q27.4968 615.839 32.4621 611.479 Q37.4273 607.118 46.212 607.118 Q54.9649 607.118 59.9619 611.479 Q64.9272 615.839 64.9272 623.478 Q64.9272 631.149 59.9619 635.509 Q54.9649 639.838 46.212 639.838 Q37.4273 639.838 32.4621 635.509 Q27.4968 631.149 27.4968 623.478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M42.4881 567.778 L64.0042 567.778 L64.0042 573.634 L42.679 573.634 Q37.6183 573.634 35.1038 575.608 Q32.5894 577.581 32.5894 581.528 Q32.5894 586.27 35.6131 589.008 Q38.6368 591.745 43.8567 591.745 L64.0042 591.745 L64.0042 597.633 L28.3562 597.633 L28.3562 591.745 L33.8944 591.745 Q30.6797 589.644 29.0883 586.811 Q27.4968 583.947 27.4968 580.223 Q27.4968 574.08 31.3163 570.929 Q35.1038 567.778 42.4881 567.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip792)" d="M478.6 597.618 L478.6 597.618 L598.266 597.618 L598.266 597.618 L478.6 597.618 L478.6 597.618  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,597.618 478.6,597.618 598.266,597.618 478.6,597.618 "/>
-<path clip-path="url(#clip792)" d="M628.182 735.212 L628.182 735.212 L747.848 735.212 L747.848 735.212 L628.182 735.212 L628.182 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,735.212 628.182,735.212 747.848,735.212 628.182,735.212 "/>
-<path clip-path="url(#clip792)" d="M777.765 528.822 L777.765 528.822 L897.431 528.822 L897.431 528.822 L777.765 528.822 L777.765 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,528.822 777.765,528.822 897.431,528.822 777.765,528.822 "/>
-<path clip-path="url(#clip792)" d="M927.348 528.822 L927.348 528.822 L1047.01 528.822 L1047.01 528.822 L927.348 528.822 L927.348 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,528.822 927.348,528.822 1047.01,528.822 927.348,528.822 "/>
-<path clip-path="url(#clip792)" d="M1076.93 460.025 L1076.93 460.025 L1196.6 460.025 L1196.6 460.025 L1076.93 460.025 L1076.93 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,460.025 1076.93,460.025 1196.6,460.025 1076.93,460.025 "/>
-<path clip-path="url(#clip792)" d="M1226.51 528.822 L1226.51 528.822 L1346.18 528.822 L1346.18 528.822 L1226.51 528.822 L1226.51 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,528.822 1226.51,528.822 1346.18,528.822 1226.51,528.822 "/>
-<path clip-path="url(#clip792)" d="M1376.1 391.228 L1376.1 391.228 L1495.76 391.228 L1495.76 391.228 L1376.1 391.228 L1376.1 391.228  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,391.228 1376.1,391.228 1495.76,391.228 1376.1,391.228 "/>
-<path clip-path="url(#clip792)" d="M1525.68 253.634 L1525.68 253.634 L1645.34 253.634 L1645.34 253.634 L1525.68 253.634 L1525.68 253.634  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,253.634 1525.68,253.634 1645.34,253.634 1525.68,253.634 "/>
-<path clip-path="url(#clip792)" d="M1675.26 47.2441 L1675.26 47.2441 L1794.93 47.2441 L1794.93 47.2441 L1675.26 47.2441 L1675.26 47.2441  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,47.2441 1675.26,47.2441 1794.93,47.2441 1675.26,47.2441 "/>
-<path clip-path="url(#clip792)" d="M1824.84 116.041 L1824.84 116.041 L1944.51 116.041 L1944.51 116.041 L1824.84 116.041 L1824.84 116.041  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,116.041 1824.84,116.041 1944.51,116.041 1824.84,116.041 "/>
-<path clip-path="url(#clip792)" d="M1974.43 460.025 L1974.43 460.025 L2094.09 460.025 L2094.09 460.025 L1974.43 460.025 L1974.43 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,460.025 1974.43,460.025 2094.09,460.025 1974.43,460.025 "/>
-<path clip-path="url(#clip792)" d="M2124.01 735.212 L2124.01 735.212 L2243.67 735.212 L2243.67 735.212 L2124.01 735.212 L2124.01 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,735.212 2124.01,735.212 2243.67,735.212 2124.01,735.212 "/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="538.433" cy="597.618" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="688.015" cy="735.212" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="837.598" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="987.181" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1136.76" cy="460.025" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1286.35" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1435.93" cy="391.228" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1585.51" cy="253.634" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1735.09" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1884.68" cy="116.041" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2034.26" cy="460.025" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2183.84" cy="735.212" r="2"/>
-<path clip-path="url(#clip792)" d="M478.6 597.618 L478.6 1079.2 L598.266 1079.2 L598.266 597.618 L478.6 597.618 L478.6 597.618  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,597.618 478.6,1079.2 598.266,1079.2 598.266,597.618 478.6,597.618 "/>
-<path clip-path="url(#clip792)" d="M628.182 735.212 L628.182 1044.8 L747.848 1044.8 L747.848 735.212 L628.182 735.212 L628.182 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,735.212 628.182,1044.8 747.848,1044.8 747.848,735.212 628.182,735.212 "/>
-<path clip-path="url(#clip792)" d="M777.765 528.822 L777.765 1010.4 L897.431 1010.4 L897.431 528.822 L777.765 528.822 L777.765 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,528.822 777.765,1010.4 897.431,1010.4 897.431,528.822 777.765,528.822 "/>
-<path clip-path="url(#clip792)" d="M927.348 528.822 L927.348 735.212 L1047.01 735.212 L1047.01 528.822 L927.348 528.822 L927.348 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,528.822 927.348,735.212 1047.01,735.212 1047.01,528.822 927.348,528.822 "/>
-<path clip-path="url(#clip792)" d="M1076.93 460.025 L1076.93 1423.18 L1196.6 1423.18 L1196.6 460.025 L1076.93 460.025 L1076.93 460.025  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,460.025 1076.93,1423.18 1196.6,1423.18 1196.6,460.025 1076.93,460.025 "/>
-<path clip-path="url(#clip792)" d="M1226.51 528.822 L1226.51 941.602 L1346.18 941.602 L1346.18 528.822 L1226.51 528.822 L1226.51 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,528.822 1226.51,941.602 1346.18,941.602 1346.18,528.822 1226.51,528.822 "/>
-<path clip-path="url(#clip792)" d="M1376.1 391.228 L1376.1 1010.4 L1495.76 1010.4 L1495.76 391.228 L1376.1 391.228 L1376.1 391.228  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,391.228 1376.1,1010.4 1495.76,1010.4 1495.76,391.228 1376.1,391.228 "/>
-<path clip-path="url(#clip792)" d="M1525.68 253.634 L1525.68 735.212 L1645.34 735.212 L1645.34 253.634 L1525.68 253.634 L1525.68 253.634  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,253.634 1525.68,735.212 1645.34,735.212 1645.34,253.634 1525.68,253.634 "/>
-<path clip-path="url(#clip792)" d="M1675.26 47.2441 L1675.26 735.212 L1794.93 735.212 L1794.93 47.2441 L1675.26 47.2441 L1675.26 47.2441  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,47.2441 1675.26,735.212 1794.93,735.212 1794.93,47.2441 1675.26,47.2441 "/>
-<path clip-path="url(#clip792)" d="M1824.84 116.041 L1824.84 1423.18 L1944.51 1423.18 L1944.51 116.041 L1824.84 116.041 L1824.84 116.041  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,116.041 1824.84,1423.18 1944.51,1423.18 1944.51,116.041 1824.84,116.041 "/>
-<path clip-path="url(#clip792)" d="M1974.43 460.025 L1974.43 1010.4 L2094.09 1010.4 L2094.09 460.025 L1974.43 460.025 L1974.43 460.025  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,460.025 1974.43,1010.4 2094.09,1010.4 2094.09,460.025 1974.43,460.025 "/>
-<path clip-path="url(#clip792)" d="M2124.01 735.212 L2124.01 1423.18 L2243.67 1423.18 L2243.67 735.212 L2124.01 735.212 L2124.01 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,735.212 2124.01,1423.18 2243.67,1423.18 2243.67,735.212 2124.01,735.212 "/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="538.433" cy="597.618" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="688.015" cy="735.212" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="837.598" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="987.181" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1136.76" cy="460.025" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1286.35" cy="528.822" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1435.93" cy="391.228" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1585.51" cy="253.634" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1735.09" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1884.68" cy="116.041" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2034.26" cy="460.025" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2183.84" cy="735.212" r="2"/>
-<path clip-path="url(#clip792)" d="M478.6 1079.2 L478.6 1423.18 L598.266 1423.18 L598.266 1079.2 L478.6 1079.2 L478.6 1079.2  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,1079.2 478.6,1423.18 598.266,1423.18 598.266,1079.2 478.6,1079.2 "/>
-<path clip-path="url(#clip792)" d="M628.182 1044.8 L628.182 1423.18 L747.848 1423.18 L747.848 1044.8 L628.182 1044.8 L628.182 1044.8  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,1044.8 628.182,1423.18 747.848,1423.18 747.848,1044.8 628.182,1044.8 "/>
-<path clip-path="url(#clip792)" d="M777.765 1010.4 L777.765 1423.18 L897.431 1423.18 L897.431 1010.4 L777.765 1010.4 L777.765 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,1010.4 777.765,1423.18 897.431,1423.18 897.431,1010.4 777.765,1010.4 "/>
-<path clip-path="url(#clip792)" d="M927.348 735.212 L927.348 1423.18 L1047.01 1423.18 L1047.01 735.212 L927.348 735.212 L927.348 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,735.212 927.348,1423.18 1047.01,1423.18 1047.01,735.212 927.348,735.212 "/>
-<path clip-path="url(#clip792)" d="M1076.93 1423.18 L1076.93 1423.18 L1196.6 1423.18 L1196.6 1423.18 L1076.93 1423.18 L1076.93 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,1423.18 1076.93,1423.18 1196.6,1423.18 1076.93,1423.18 "/>
-<path clip-path="url(#clip792)" d="M1226.51 941.602 L1226.51 1423.18 L1346.18 1423.18 L1346.18 941.602 L1226.51 941.602 L1226.51 941.602  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,941.602 1226.51,1423.18 1346.18,1423.18 1346.18,941.602 1226.51,941.602 "/>
-<path clip-path="url(#clip792)" d="M1376.1 1010.4 L1376.1 1423.18 L1495.76 1423.18 L1495.76 1010.4 L1376.1 1010.4 L1376.1 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,1010.4 1376.1,1423.18 1495.76,1423.18 1495.76,1010.4 1376.1,1010.4 "/>
-<path clip-path="url(#clip792)" d="M1525.68 735.212 L1525.68 1423.18 L1645.34 1423.18 L1645.34 735.212 L1525.68 735.212 L1525.68 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,735.212 1525.68,1423.18 1645.34,1423.18 1645.34,735.212 1525.68,735.212 "/>
-<path clip-path="url(#clip792)" d="M1675.26 735.212 L1675.26 1423.18 L1794.93 1423.18 L1794.93 735.212 L1675.26 735.212 L1675.26 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,735.212 1675.26,1423.18 1794.93,1423.18 1794.93,735.212 1675.26,735.212 "/>
-<path clip-path="url(#clip792)" d="M1824.84 1423.18 L1824.84 1423.18 L1944.51 1423.18 L1944.51 1423.18 L1824.84 1423.18 L1824.84 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,1423.18 1824.84,1423.18 1944.51,1423.18 1824.84,1423.18 "/>
-<path clip-path="url(#clip792)" d="M1974.43 1010.4 L1974.43 1423.18 L2094.09 1423.18 L2094.09 1010.4 L1974.43 1010.4 L1974.43 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,1010.4 1974.43,1423.18 2094.09,1423.18 2094.09,1010.4 1974.43,1010.4 "/>
-<path clip-path="url(#clip792)" d="M2124.01 1423.18 L2124.01 1423.18 L2243.67 1423.18 L2243.67 1423.18 L2124.01 1423.18 L2124.01 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,1423.18 2124.01,1423.18 2243.67,1423.18 2124.01,1423.18 "/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="538.433" cy="1079.2" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="688.015" cy="1044.8" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="837.598" cy="1010.4" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="987.181" cy="735.212" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1136.76" cy="1423.18" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1286.35" cy="941.602" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1435.93" cy="1010.4" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1585.51" cy="735.212" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1735.09" cy="735.212" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1884.68" cy="1423.18" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2034.26" cy="1010.4" r="2"/>
-<circle clip-path="url(#clip792)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2183.84" cy="1423.18" r="2"/>
-<path clip-path="url(#clip790)" d="M435.626 300.469 L989.342 300.469 L989.342 93.1086 L435.626 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="435.626,300.469 989.342,300.469 989.342,93.1086 435.626,93.1086 435.626,300.469 "/>
-<path clip-path="url(#clip790)" d="M457.662 165.685 L589.878 165.685 L589.878 124.213 L457.662 124.213 L457.662 165.685  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,165.685 589.878,165.685 589.878,124.213 457.662,124.213 457.662,165.685 "/>
-<path clip-path="url(#clip790)" d="M611.914 151.997 L611.914 136.303 L616.173 136.303 L616.173 151.835 Q616.173 155.516 617.608 157.368 Q619.044 159.196 621.914 159.196 Q625.363 159.196 627.354 156.997 Q629.368 154.798 629.368 151.002 L629.368 136.303 L633.627 136.303 L633.627 162.229 L629.368 162.229 L629.368 158.247 Q627.817 160.608 625.756 161.766 Q623.719 162.9 621.011 162.9 Q616.544 162.9 614.229 160.122 Q611.914 157.344 611.914 151.997 M622.631 135.678 L622.631 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M663.951 146.581 L663.951 162.229 L659.691 162.229 L659.691 146.719 Q659.691 143.039 658.256 141.21 Q656.821 139.382 653.951 139.382 Q650.502 139.382 648.511 141.581 Q646.52 143.78 646.52 147.576 L646.52 162.229 L642.238 162.229 L642.238 136.303 L646.52 136.303 L646.52 140.331 Q648.048 137.993 650.108 136.835 Q652.192 135.678 654.9 135.678 Q659.367 135.678 661.659 138.456 Q663.951 141.21 663.951 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M692.631 141.28 Q694.228 138.409 696.451 137.044 Q698.673 135.678 701.682 135.678 Q705.733 135.678 707.932 138.525 Q710.131 141.349 710.131 146.581 L710.131 162.229 L705.849 162.229 L705.849 146.719 Q705.849 142.993 704.529 141.187 Q703.21 139.382 700.501 139.382 Q697.191 139.382 695.27 141.581 Q693.349 143.78 693.349 147.576 L693.349 162.229 L689.066 162.229 L689.066 146.719 Q689.066 142.969 687.747 141.187 Q686.427 139.382 683.673 139.382 Q680.409 139.382 678.488 141.604 Q676.566 143.803 676.566 147.576 L676.566 162.229 L672.284 162.229 L672.284 136.303 L676.566 136.303 L676.566 140.331 Q678.025 137.946 680.062 136.812 Q682.099 135.678 684.9 135.678 Q687.724 135.678 689.691 137.113 Q691.682 138.548 692.631 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M740.802 148.201 L740.802 150.284 L721.219 150.284 Q721.497 154.682 723.858 156.997 Q726.242 159.289 730.478 159.289 Q732.932 159.289 735.223 158.687 Q737.538 158.085 739.807 156.881 L739.807 160.909 Q737.515 161.881 735.108 162.391 Q732.7 162.9 730.224 162.9 Q724.02 162.9 720.386 159.289 Q716.775 155.678 716.775 149.52 Q716.775 143.155 720.2 139.428 Q723.649 135.678 729.483 135.678 Q734.714 135.678 737.747 139.057 Q740.802 142.414 740.802 148.201 M736.543 146.951 Q736.497 143.456 734.575 141.372 Q732.677 139.289 729.529 139.289 Q725.964 139.289 723.812 141.303 Q721.682 143.317 721.358 146.974 L736.543 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M752.006 128.942 L752.006 136.303 L760.779 136.303 L760.779 139.613 L752.006 139.613 L752.006 153.687 Q752.006 156.858 752.862 157.761 Q753.742 158.664 756.404 158.664 L760.779 158.664 L760.779 162.229 L756.404 162.229 Q751.473 162.229 749.598 160.4 Q747.723 158.548 747.723 153.687 L747.723 139.613 L744.598 139.613 L744.598 136.303 L747.723 136.303 L747.723 128.942 L752.006 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M798.51 140.238 L798.51 126.21 L802.769 126.21 L802.769 162.229 L798.51 162.229 L798.51 158.34 Q797.168 160.655 795.107 161.789 Q793.07 162.9 790.2 162.9 Q785.501 162.9 782.538 159.15 Q779.598 155.4 779.598 149.289 Q779.598 143.178 782.538 139.428 Q785.501 135.678 790.2 135.678 Q793.07 135.678 795.107 136.812 Q797.168 137.923 798.51 140.238 M783.996 149.289 Q783.996 153.988 785.918 156.673 Q787.862 159.335 791.242 159.335 Q794.621 159.335 796.566 156.673 Q798.51 153.988 798.51 149.289 Q798.51 144.59 796.566 141.928 Q794.621 139.243 791.242 139.243 Q787.862 139.243 785.918 141.928 Q783.996 144.59 783.996 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M833.718 148.201 L833.718 150.284 L814.135 150.284 Q814.413 154.682 816.774 156.997 Q819.158 159.289 823.394 159.289 Q825.848 159.289 828.14 158.687 Q830.454 158.085 832.723 156.881 L832.723 160.909 Q830.431 161.881 828.024 162.391 Q825.617 162.9 823.14 162.9 Q816.936 162.9 813.302 159.289 Q809.691 155.678 809.691 149.52 Q809.691 143.155 813.117 139.428 Q816.566 135.678 822.399 135.678 Q827.63 135.678 830.663 139.057 Q833.718 142.414 833.718 148.201 M829.459 146.951 Q829.413 143.456 827.492 141.372 Q825.593 139.289 822.445 139.289 Q818.88 139.289 816.728 141.303 Q814.598 143.317 814.274 146.974 L829.459 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M860.894 141.28 Q862.491 138.409 864.714 137.044 Q866.936 135.678 869.945 135.678 Q873.996 135.678 876.195 138.525 Q878.394 141.349 878.394 146.581 L878.394 162.229 L874.112 162.229 L874.112 146.719 Q874.112 142.993 872.792 141.187 Q871.473 139.382 868.764 139.382 Q865.454 139.382 863.533 141.581 Q861.612 143.78 861.612 147.576 L861.612 162.229 L857.329 162.229 L857.329 146.719 Q857.329 142.969 856.01 141.187 Q854.69 139.382 851.936 139.382 Q848.672 139.382 846.751 141.604 Q844.829 143.803 844.829 147.576 L844.829 162.229 L840.547 162.229 L840.547 136.303 L844.829 136.303 L844.829 140.331 Q846.288 137.946 848.325 136.812 Q850.362 135.678 853.163 135.678 Q855.987 135.678 857.954 137.113 Q859.945 138.548 860.894 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M898.672 149.196 Q893.51 149.196 891.519 150.377 Q889.528 151.557 889.528 154.405 Q889.528 156.673 891.01 158.016 Q892.514 159.335 895.084 159.335 Q898.625 159.335 900.755 156.835 Q902.908 154.312 902.908 150.145 L902.908 149.196 L898.672 149.196 M907.167 147.437 L907.167 162.229 L902.908 162.229 L902.908 158.293 Q901.449 160.655 899.274 161.789 Q897.098 162.9 893.949 162.9 Q889.968 162.9 887.607 160.678 Q885.269 158.432 885.269 154.682 Q885.269 150.307 888.186 148.085 Q891.125 145.863 896.936 145.863 L902.908 145.863 L902.908 145.446 Q902.908 142.507 900.963 140.909 Q899.042 139.289 895.547 139.289 Q893.324 139.289 891.218 139.821 Q889.112 140.354 887.167 141.419 L887.167 137.483 Q889.505 136.581 891.704 136.141 Q893.903 135.678 895.987 135.678 Q901.611 135.678 904.389 138.594 Q907.167 141.511 907.167 147.437 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M937.491 146.581 L937.491 162.229 L933.232 162.229 L933.232 146.719 Q933.232 143.039 931.797 141.21 Q930.361 139.382 927.491 139.382 Q924.042 139.382 922.051 141.581 Q920.06 143.78 920.06 147.576 L920.06 162.229 L915.778 162.229 L915.778 136.303 L920.06 136.303 L920.06 140.331 Q921.588 137.993 923.648 136.835 Q925.732 135.678 928.44 135.678 Q932.908 135.678 935.199 138.456 Q937.491 141.21 937.491 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M963.046 140.238 L963.046 126.21 L967.306 126.21 L967.306 162.229 L963.046 162.229 L963.046 158.34 Q961.704 160.655 959.644 161.789 Q957.607 162.9 954.736 162.9 Q950.037 162.9 947.074 159.15 Q944.134 155.4 944.134 149.289 Q944.134 143.178 947.074 139.428 Q950.037 135.678 954.736 135.678 Q957.607 135.678 959.644 136.812 Q961.704 137.923 963.046 140.238 M948.533 149.289 Q948.533 153.988 950.454 156.673 Q952.398 159.335 955.778 159.335 Q959.157 159.335 961.102 156.673 Q963.046 153.988 963.046 149.289 Q963.046 144.59 961.102 141.928 Q959.157 139.243 955.778 139.243 Q952.398 139.243 950.454 141.928 Q948.533 144.59 948.533 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M457.662 217.525 L589.878 217.525 L589.878 176.053 L457.662 176.053 L457.662 217.525  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,217.525 589.878,217.525 589.878,176.053 457.662,176.053 457.662,217.525 "/>
-<path clip-path="url(#clip790)" d="M627.747 184.115 L621.405 201.314 L634.113 201.314 L627.747 184.115 M625.108 179.509 L630.409 179.509 L643.58 214.069 L638.719 214.069 L635.571 205.203 L619.993 205.203 L616.844 214.069 L611.914 214.069 L625.108 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M457.662 269.365 L589.878 269.365 L589.878 227.893 L457.662 227.893 L457.662 269.365  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,269.365 589.878,269.365 589.878,227.893 457.662,227.893 457.662,269.365 "/>
-<path clip-path="url(#clip790)" d="M616.59 249.404 L616.59 262.066 L624.09 262.066 Q627.863 262.066 629.668 260.515 Q631.497 258.941 631.497 255.723 Q631.497 252.483 629.668 250.955 Q627.863 249.404 624.09 249.404 L616.59 249.404 M616.59 235.191 L616.59 245.608 L623.511 245.608 Q626.937 245.608 628.604 244.335 Q630.293 243.038 630.293 240.399 Q630.293 237.784 628.604 236.487 Q626.937 235.191 623.511 235.191 L616.59 235.191 M611.914 231.349 L623.858 231.349 Q629.206 231.349 632.099 233.571 Q634.993 235.793 634.993 239.89 Q634.993 243.062 633.511 244.937 Q632.03 246.811 629.159 247.274 Q632.608 248.015 634.506 250.376 Q636.428 252.714 636.428 256.233 Q636.428 260.862 633.28 263.385 Q630.131 265.909 624.321 265.909 L611.914 265.909 L611.914 231.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="388.85,1423.18 388.85,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="688.015,1423.18 688.015,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="987.181,1423.18 987.181,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1286.35,1423.18 1286.35,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1585.51,1423.18 1585.51,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1884.68,1423.18 1884.68,47.2441 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2183.84,1423.18 2183.84,47.2441 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="388.85,1423.18 388.85,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="688.015,1423.18 688.015,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="987.181,1423.18 987.181,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.35,1423.18 1286.35,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1585.51,1423.18 1585.51,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1884.68,1423.18 1884.68,1404.28 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2183.84,1423.18 2183.84,1404.28 "/>
+<path clip-path="url(#clip610)" d="M388.85 1454.1 Q385.239 1454.1 383.411 1457.66 Q381.605 1461.2 381.605 1468.33 Q381.605 1475.44 383.411 1479.01 Q385.239 1482.55 388.85 1482.55 Q392.485 1482.55 394.29 1479.01 Q396.119 1475.44 396.119 1468.33 Q396.119 1461.2 394.29 1457.66 Q392.485 1454.1 388.85 1454.1 M388.85 1450.39 Q394.66 1450.39 397.716 1455 Q400.795 1459.58 400.795 1468.33 Q400.795 1477.06 397.716 1481.67 Q394.66 1486.25 388.85 1486.25 Q383.04 1486.25 379.961 1481.67 Q376.906 1477.06 376.906 1468.33 Q376.906 1459.58 379.961 1455 Q383.04 1450.39 388.85 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M682.668 1481.64 L698.988 1481.64 L698.988 1485.58 L677.043 1485.58 L677.043 1481.64 Q679.705 1478.89 684.289 1474.26 Q688.895 1469.61 690.076 1468.27 Q692.321 1465.74 693.201 1464.01 Q694.103 1462.25 694.103 1460.56 Q694.103 1457.8 692.159 1456.07 Q690.238 1454.33 687.136 1454.33 Q684.937 1454.33 682.483 1455.09 Q680.053 1455.86 677.275 1457.41 L677.275 1452.69 Q680.099 1451.55 682.553 1450.97 Q685.006 1450.39 687.043 1450.39 Q692.414 1450.39 695.608 1453.08 Q698.802 1455.77 698.802 1460.26 Q698.802 1462.39 697.992 1464.31 Q697.205 1466.2 695.099 1468.8 Q694.52 1469.47 691.418 1472.69 Q688.316 1475.88 682.668 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M990.19 1455.09 L978.384 1473.54 L990.19 1473.54 L990.19 1455.09 M988.963 1451.02 L994.843 1451.02 L994.843 1473.54 L999.773 1473.54 L999.773 1477.43 L994.843 1477.43 L994.843 1485.58 L990.19 1485.58 L990.19 1477.43 L974.588 1477.43 L974.588 1472.92 L988.963 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1286.75 1466.44 Q1283.6 1466.44 1281.75 1468.59 Q1279.92 1470.74 1279.92 1474.49 Q1279.92 1478.22 1281.75 1480.39 Q1283.6 1482.55 1286.75 1482.55 Q1289.9 1482.55 1291.73 1480.39 Q1293.58 1478.22 1293.58 1474.49 Q1293.58 1470.74 1291.73 1468.59 Q1289.9 1466.44 1286.75 1466.44 M1296.03 1451.78 L1296.03 1456.04 Q1294.27 1455.21 1292.47 1454.77 Q1290.69 1454.33 1288.93 1454.33 Q1284.3 1454.33 1281.84 1457.45 Q1279.41 1460.58 1279.07 1466.9 Q1280.43 1464.89 1282.49 1463.82 Q1284.55 1462.73 1287.03 1462.73 Q1292.24 1462.73 1295.25 1465.9 Q1298.28 1469.05 1298.28 1474.49 Q1298.28 1479.82 1295.13 1483.03 Q1291.98 1486.25 1286.75 1486.25 Q1280.76 1486.25 1277.58 1481.67 Q1274.41 1477.06 1274.41 1468.33 Q1274.41 1460.14 1278.3 1455.28 Q1282.19 1450.39 1288.74 1450.39 Q1290.5 1450.39 1292.28 1450.74 Q1294.09 1451.09 1296.03 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1585.51 1469.17 Q1582.18 1469.17 1580.26 1470.95 Q1578.36 1472.73 1578.36 1475.86 Q1578.36 1478.98 1580.26 1480.77 Q1582.18 1482.55 1585.51 1482.55 Q1588.84 1482.55 1590.77 1480.77 Q1592.69 1478.96 1592.69 1475.86 Q1592.69 1472.73 1590.77 1470.95 Q1588.87 1469.17 1585.51 1469.17 M1580.84 1467.18 Q1577.83 1466.44 1576.14 1464.38 Q1574.47 1462.32 1574.47 1459.35 Q1574.47 1455.21 1577.41 1452.8 Q1580.37 1450.39 1585.51 1450.39 Q1590.67 1450.39 1593.61 1452.8 Q1596.55 1455.21 1596.55 1459.35 Q1596.55 1462.32 1594.86 1464.38 Q1593.2 1466.44 1590.21 1467.18 Q1593.59 1467.96 1595.46 1470.26 Q1597.36 1472.55 1597.36 1475.86 Q1597.36 1480.88 1594.28 1483.57 Q1591.23 1486.25 1585.51 1486.25 Q1579.79 1486.25 1576.71 1483.57 Q1573.66 1480.88 1573.66 1475.86 Q1573.66 1472.55 1575.56 1470.26 Q1577.46 1467.96 1580.84 1467.18 M1579.12 1459.79 Q1579.12 1462.48 1580.79 1463.98 Q1582.48 1465.49 1585.51 1465.49 Q1588.52 1465.49 1590.21 1463.98 Q1591.92 1462.48 1591.92 1459.79 Q1591.92 1457.11 1590.21 1455.6 Q1588.52 1454.1 1585.51 1454.1 Q1582.48 1454.1 1580.79 1455.6 Q1579.12 1457.11 1579.12 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1859.36 1481.64 L1867 1481.64 L1867 1455.28 L1858.69 1456.95 L1858.69 1452.69 L1866.96 1451.02 L1871.63 1451.02 L1871.63 1481.64 L1879.27 1481.64 L1879.27 1485.58 L1859.36 1485.58 L1859.36 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1898.72 1454.1 Q1895.1 1454.1 1893.28 1457.66 Q1891.47 1461.2 1891.47 1468.33 Q1891.47 1475.44 1893.28 1479.01 Q1895.1 1482.55 1898.72 1482.55 Q1902.35 1482.55 1904.16 1479.01 Q1905.98 1475.44 1905.98 1468.33 Q1905.98 1461.2 1904.16 1457.66 Q1902.35 1454.1 1898.72 1454.1 M1898.72 1450.39 Q1904.53 1450.39 1907.58 1455 Q1910.66 1459.58 1910.66 1468.33 Q1910.66 1477.06 1907.58 1481.67 Q1904.53 1486.25 1898.72 1486.25 Q1892.91 1486.25 1889.83 1481.67 Q1886.77 1477.06 1886.77 1468.33 Q1886.77 1459.58 1889.83 1455 Q1892.91 1450.39 1898.72 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M2159.33 1481.64 L2166.97 1481.64 L2166.97 1455.28 L2158.66 1456.95 L2158.66 1452.69 L2166.92 1451.02 L2171.6 1451.02 L2171.6 1481.64 L2179.23 1481.64 L2179.23 1485.58 L2159.33 1485.58 L2159.33 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M2192.71 1481.64 L2209.03 1481.64 L2209.03 1485.58 L2187.08 1485.58 L2187.08 1481.64 Q2189.74 1478.89 2194.33 1474.26 Q2198.93 1469.61 2200.11 1468.27 Q2202.36 1465.74 2203.24 1464.01 Q2204.14 1462.25 2204.14 1460.56 Q2204.14 1457.8 2202.2 1456.07 Q2200.28 1454.33 2197.17 1454.33 Q2194.98 1454.33 2192.52 1455.09 Q2190.09 1455.86 2187.31 1457.41 L2187.31 1452.69 Q2190.14 1451.55 2192.59 1450.97 Q2195.04 1450.39 2197.08 1450.39 Q2202.45 1450.39 2205.65 1453.08 Q2208.84 1455.77 2208.84 1460.26 Q2208.84 1462.39 2208.03 1464.31 Q2207.24 1466.2 2205.14 1468.8 Q2204.56 1469.47 2201.46 1472.69 Q2198.36 1475.88 2192.71 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1264.95 1520.52 L1274.53 1520.52 L1286.66 1552.86 L1298.85 1520.52 L1308.43 1520.52 L1308.43 1568.04 L1302.16 1568.04 L1302.16 1526.32 L1289.9 1558.91 L1283.44 1558.91 L1271.19 1526.32 L1271.19 1568.04 L1264.95 1568.04 L1264.95 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1334.75 1536.5 Q1330.04 1536.5 1327.3 1540.19 Q1324.57 1543.85 1324.57 1550.25 Q1324.57 1556.65 1327.27 1560.34 Q1330.01 1564 1334.75 1564 Q1339.43 1564 1342.17 1560.31 Q1344.9 1556.62 1344.9 1550.25 Q1344.9 1543.92 1342.17 1540.23 Q1339.43 1536.5 1334.75 1536.5 M1334.75 1531.54 Q1342.39 1531.54 1346.75 1536.5 Q1351.11 1541.47 1351.11 1550.25 Q1351.11 1559 1346.75 1564 Q1342.39 1568.97 1334.75 1568.97 Q1327.08 1568.97 1322.72 1564 Q1318.39 1559 1318.39 1550.25 Q1318.39 1541.47 1322.72 1536.5 Q1327.08 1531.54 1334.75 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1390.45 1546.53 L1390.45 1568.04 L1384.59 1568.04 L1384.59 1546.72 Q1384.59 1541.66 1382.62 1539.14 Q1380.65 1536.63 1376.7 1536.63 Q1371.96 1536.63 1369.22 1539.65 Q1366.48 1542.68 1366.48 1547.9 L1366.48 1568.04 L1360.6 1568.04 L1360.6 1532.4 L1366.48 1532.4 L1366.48 1537.93 Q1368.58 1534.72 1371.42 1533.13 Q1374.28 1531.54 1378.01 1531.54 Q1384.15 1531.54 1387.3 1535.36 Q1390.45 1539.14 1390.45 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1407.93 1522.27 L1407.93 1532.4 L1419.99 1532.4 L1419.99 1536.95 L1407.93 1536.95 L1407.93 1556.3 Q1407.93 1560.66 1409.1 1561.9 Q1410.31 1563.14 1413.97 1563.14 L1419.99 1563.14 L1419.99 1568.04 L1413.97 1568.04 Q1407.19 1568.04 1404.61 1565.53 Q1402.04 1562.98 1402.04 1556.3 L1402.04 1536.95 L1397.74 1536.95 L1397.74 1532.4 L1402.04 1532.4 L1402.04 1522.27 L1407.93 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M1457.32 1546.53 L1457.32 1568.04 L1451.47 1568.04 L1451.47 1546.72 Q1451.47 1541.66 1449.49 1539.14 Q1447.52 1536.63 1443.57 1536.63 Q1438.83 1536.63 1436.09 1539.65 Q1433.36 1542.68 1433.36 1547.9 L1433.36 1568.04 L1427.47 1568.04 L1427.47 1518.52 L1433.36 1518.52 L1433.36 1537.93 Q1435.46 1534.72 1438.29 1533.13 Q1441.15 1531.54 1444.88 1531.54 Q1451.02 1531.54 1454.17 1535.36 Q1457.32 1539.14 1457.32 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,1079.2 2352.76,1079.2 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,735.212 2352.76,735.212 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,391.228 2352.76,391.228 "/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="369.518,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 369.518,47.2441 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1423.18 388.416,1423.18 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,1079.2 388.416,1079.2 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,735.212 388.416,735.212 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,391.228 388.416,391.228 "/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="369.518,47.2441 388.416,47.2441 "/>
+<path clip-path="url(#clip610)" d="M321.574 1408.98 Q317.963 1408.98 316.134 1412.54 Q314.329 1416.08 314.329 1423.21 Q314.329 1430.32 316.134 1433.89 Q317.963 1437.43 321.574 1437.43 Q325.208 1437.43 327.014 1433.89 Q328.842 1430.32 328.842 1423.21 Q328.842 1416.08 327.014 1412.54 Q325.208 1408.98 321.574 1408.98 M321.574 1405.27 Q327.384 1405.27 330.44 1409.88 Q333.518 1414.46 333.518 1423.21 Q333.518 1431.94 330.44 1436.55 Q327.384 1441.13 321.574 1441.13 Q315.764 1441.13 312.685 1436.55 Q309.629 1431.94 309.629 1423.21 Q309.629 1414.46 312.685 1409.88 Q315.764 1405.27 321.574 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M115.718 1068.36 L134.075 1068.36 L134.075 1072.3 L120.001 1072.3 L120.001 1080.77 Q121.019 1080.42 122.038 1080.26 Q123.056 1080.08 124.075 1080.08 Q129.862 1080.08 133.242 1083.25 Q136.621 1086.42 136.621 1091.84 Q136.621 1097.41 133.149 1100.52 Q129.677 1103.59 123.357 1103.59 Q121.181 1103.59 118.913 1103.22 Q116.668 1102.85 114.26 1102.11 L114.26 1097.41 Q116.343 1098.55 118.566 1099.1 Q120.788 1099.66 123.265 1099.66 Q127.269 1099.66 129.607 1097.55 Q131.945 1095.45 131.945 1091.84 Q131.945 1088.22 129.607 1086.12 Q127.269 1084.01 123.265 1084.01 Q121.39 1084.01 119.515 1084.43 Q117.663 1084.84 115.718 1085.72 L115.718 1068.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M145.834 1097.04 L150.718 1097.04 L150.718 1102.92 L145.834 1102.92 L145.834 1097.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M170.903 1071.44 Q167.292 1071.44 165.464 1075.01 Q163.658 1078.55 163.658 1085.68 Q163.658 1092.78 165.464 1096.35 Q167.292 1099.89 170.903 1099.89 Q174.538 1099.89 176.343 1096.35 Q178.172 1092.78 178.172 1085.68 Q178.172 1078.55 176.343 1075.01 Q174.538 1071.44 170.903 1071.44 M170.903 1067.74 Q176.714 1067.74 179.769 1072.34 Q182.848 1076.93 182.848 1085.68 Q182.848 1094.41 179.769 1099.01 Q176.714 1103.59 170.903 1103.59 Q165.093 1103.59 162.015 1099.01 Q158.959 1094.41 158.959 1085.68 Q158.959 1076.93 162.015 1072.34 Q165.093 1067.74 170.903 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M201.065 1071.44 Q197.454 1071.44 195.625 1075.01 Q193.82 1078.55 193.82 1085.68 Q193.82 1092.78 195.625 1096.35 Q197.454 1099.89 201.065 1099.89 Q204.699 1099.89 206.505 1096.35 Q208.334 1092.78 208.334 1085.68 Q208.334 1078.55 206.505 1075.01 Q204.699 1071.44 201.065 1071.44 M201.065 1067.74 Q206.875 1067.74 209.931 1072.34 Q213.01 1076.93 213.01 1085.68 Q213.01 1094.41 209.931 1099.01 Q206.875 1103.59 201.065 1103.59 Q195.255 1103.59 192.176 1099.01 Q189.121 1094.41 189.121 1085.68 Q189.121 1076.93 192.176 1072.34 Q195.255 1067.74 201.065 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M249.398 1077.46 L238.82 1088.09 L249.398 1098.66 L246.644 1101.47 L236.019 1090.84 L225.394 1101.47 L222.662 1098.66 L233.218 1088.09 L222.662 1077.46 L225.394 1074.66 L236.019 1085.28 L246.644 1074.66 L249.398 1077.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M261.759 1098.99 L269.398 1098.99 L269.398 1072.62 L261.088 1074.29 L261.088 1070.03 L269.352 1068.36 L274.028 1068.36 L274.028 1098.99 L281.667 1098.99 L281.667 1102.92 L261.759 1102.92 L261.759 1098.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M301.111 1071.44 Q297.5 1071.44 295.671 1075.01 Q293.866 1078.55 293.866 1085.68 Q293.866 1092.78 295.671 1096.35 Q297.5 1099.89 301.111 1099.89 Q304.745 1099.89 306.551 1096.35 Q308.379 1092.78 308.379 1085.68 Q308.379 1078.55 306.551 1075.01 Q304.745 1071.44 301.111 1071.44 M301.111 1067.74 Q306.921 1067.74 309.977 1072.34 Q313.055 1076.93 313.055 1085.68 Q313.055 1094.41 309.977 1099.01 Q306.921 1103.59 301.111 1103.59 Q295.301 1103.59 292.222 1099.01 Q289.167 1094.41 289.167 1085.68 Q289.167 1076.93 292.222 1072.34 Q295.301 1067.74 301.111 1067.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M325.732 1050.74 L316.14 1065.73 L325.732 1065.73 L325.732 1050.74 M324.735 1047.43 L329.512 1047.43 L329.512 1065.73 L333.518 1065.73 L333.518 1068.89 L329.512 1068.89 L329.512 1075.51 L325.732 1075.51 L325.732 1068.89 L313.055 1068.89 L313.055 1065.23 L324.735 1047.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M118.777 755.004 L126.416 755.004 L126.416 728.639 L118.106 730.305 L118.106 726.046 L126.369 724.38 L131.045 724.38 L131.045 755.004 L138.684 755.004 L138.684 758.94 L118.777 758.94 L118.777 755.004 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M148.129 753.06 L153.013 753.06 L153.013 758.94 L148.129 758.94 L148.129 753.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M173.198 727.458 Q169.587 727.458 167.758 731.023 Q165.953 734.565 165.953 741.694 Q165.953 748.801 167.758 752.366 Q169.587 755.907 173.198 755.907 Q176.832 755.907 178.638 752.366 Q180.466 748.801 180.466 741.694 Q180.466 734.565 178.638 731.023 Q176.832 727.458 173.198 727.458 M173.198 723.755 Q179.008 723.755 182.064 728.361 Q185.142 732.944 185.142 741.694 Q185.142 750.421 182.064 755.028 Q179.008 759.611 173.198 759.611 Q167.388 759.611 164.309 755.028 Q161.254 750.421 161.254 741.694 Q161.254 732.944 164.309 728.361 Q167.388 723.755 173.198 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M203.36 727.458 Q199.749 727.458 197.92 731.023 Q196.114 734.565 196.114 741.694 Q196.114 748.801 197.92 752.366 Q199.749 755.907 203.36 755.907 Q206.994 755.907 208.8 752.366 Q210.628 748.801 210.628 741.694 Q210.628 734.565 208.8 731.023 Q206.994 727.458 203.36 727.458 M203.36 723.755 Q209.17 723.755 212.225 728.361 Q215.304 732.944 215.304 741.694 Q215.304 750.421 212.225 755.028 Q209.17 759.611 203.36 759.611 Q197.55 759.611 194.471 755.028 Q191.415 750.421 191.415 741.694 Q191.415 732.944 194.471 728.361 Q197.55 723.755 203.36 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M251.693 733.477 L241.114 744.102 L251.693 754.68 L248.938 757.481 L238.313 746.856 L227.688 757.481 L224.957 754.68 L235.512 744.102 L224.957 733.477 L227.688 730.676 L238.313 741.301 L248.938 730.676 L251.693 733.477 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M264.054 755.004 L271.693 755.004 L271.693 728.639 L263.383 730.305 L263.383 726.046 L271.646 724.38 L276.322 724.38 L276.322 755.004 L283.961 755.004 L283.961 758.94 L264.054 758.94 L264.054 755.004 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M303.406 727.458 Q299.794 727.458 297.966 731.023 Q296.16 734.565 296.16 741.694 Q296.16 748.801 297.966 752.366 Q299.794 755.907 303.406 755.907 Q307.04 755.907 308.845 752.366 Q310.674 748.801 310.674 741.694 Q310.674 734.565 308.845 731.023 Q307.04 727.458 303.406 727.458 M303.406 723.755 Q309.216 723.755 312.271 728.361 Q315.35 732.944 315.35 741.694 Q315.35 750.421 312.271 755.028 Q309.216 759.611 303.406 759.611 Q297.595 759.611 294.517 755.028 Q291.461 750.421 291.461 741.694 Q291.461 732.944 294.517 728.361 Q297.595 723.755 303.406 723.755 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M316.535 703.449 L331.449 703.449 L331.449 706.646 L320.014 706.646 L320.014 713.53 Q320.842 713.248 321.669 713.116 Q322.497 712.966 323.324 712.966 Q328.026 712.966 330.772 715.543 Q333.518 718.119 333.518 722.52 Q333.518 727.053 330.697 729.573 Q327.876 732.075 322.741 732.075 Q320.973 732.075 319.13 731.774 Q317.306 731.473 315.35 730.871 L315.35 727.053 Q317.043 727.975 318.848 728.426 Q320.654 728.877 322.666 728.877 Q325.92 728.877 327.819 727.166 Q329.719 725.454 329.719 722.52 Q329.719 719.586 327.819 717.875 Q325.92 716.163 322.666 716.163 Q321.143 716.163 319.619 716.502 Q318.115 716.84 316.535 717.555 L316.535 703.449 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M118.777 411.02 L126.416 411.02 L126.416 384.655 L118.106 386.322 L118.106 382.062 L126.369 380.396 L131.045 380.396 L131.045 411.02 L138.684 411.02 L138.684 414.956 L118.777 414.956 L118.777 411.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M148.129 409.076 L153.013 409.076 L153.013 414.956 L148.129 414.956 L148.129 409.076 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M163.244 380.396 L181.601 380.396 L181.601 384.331 L167.527 384.331 L167.527 392.803 Q168.545 392.456 169.564 392.294 Q170.582 392.109 171.601 392.109 Q177.388 392.109 180.767 395.28 Q184.147 398.451 184.147 403.868 Q184.147 409.446 180.675 412.548 Q177.203 415.627 170.883 415.627 Q168.707 415.627 166.439 415.257 Q164.193 414.886 161.786 414.145 L161.786 409.446 Q163.869 410.581 166.091 411.136 Q168.314 411.692 170.791 411.692 Q174.795 411.692 177.133 409.585 Q179.471 407.479 179.471 403.868 Q179.471 400.257 177.133 398.15 Q174.795 396.044 170.791 396.044 Q168.916 396.044 167.041 396.46 Q165.189 396.877 163.244 397.757 L163.244 380.396 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M203.36 383.474 Q199.749 383.474 197.92 387.039 Q196.114 390.581 196.114 397.71 Q196.114 404.817 197.92 408.382 Q199.749 411.923 203.36 411.923 Q206.994 411.923 208.8 408.382 Q210.628 404.817 210.628 397.71 Q210.628 390.581 208.8 387.039 Q206.994 383.474 203.36 383.474 M203.36 379.771 Q209.17 379.771 212.225 384.377 Q215.304 388.96 215.304 397.71 Q215.304 406.437 212.225 411.044 Q209.17 415.627 203.36 415.627 Q197.55 415.627 194.471 411.044 Q191.415 406.437 191.415 397.71 Q191.415 388.96 194.471 384.377 Q197.55 379.771 203.36 379.771 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M251.693 389.493 L241.114 400.118 L251.693 410.696 L248.938 413.497 L238.313 402.872 L227.688 413.497 L224.957 410.696 L235.512 400.118 L224.957 389.493 L227.688 386.692 L238.313 397.317 L248.938 386.692 L251.693 389.493 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M264.054 411.02 L271.693 411.02 L271.693 384.655 L263.383 386.322 L263.383 382.062 L271.646 380.396 L276.322 380.396 L276.322 411.02 L283.961 411.02 L283.961 414.956 L264.054 414.956 L264.054 411.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M303.406 383.474 Q299.794 383.474 297.966 387.039 Q296.16 390.581 296.16 397.71 Q296.16 404.817 297.966 408.382 Q299.794 411.923 303.406 411.923 Q307.04 411.923 308.845 408.382 Q310.674 404.817 310.674 397.71 Q310.674 390.581 308.845 387.039 Q307.04 383.474 303.406 383.474 M303.406 379.771 Q309.216 379.771 312.271 384.377 Q315.35 388.96 315.35 397.71 Q315.35 406.437 312.271 411.044 Q309.216 415.627 303.406 415.627 Q297.595 415.627 294.517 411.044 Q291.461 406.437 291.461 397.71 Q291.461 388.96 294.517 384.377 Q297.595 379.771 303.406 379.771 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M316.535 359.465 L331.449 359.465 L331.449 362.663 L320.014 362.663 L320.014 369.546 Q320.842 369.264 321.669 369.132 Q322.497 368.982 323.324 368.982 Q328.026 368.982 330.772 371.559 Q333.518 374.135 333.518 378.536 Q333.518 383.069 330.697 385.589 Q327.876 388.091 322.741 388.091 Q320.973 388.091 319.13 387.79 Q317.306 387.489 315.35 386.887 L315.35 383.069 Q317.043 383.991 318.848 384.442 Q320.654 384.893 322.666 384.893 Q325.92 384.893 327.819 383.182 Q329.719 381.47 329.719 378.536 Q329.719 375.602 327.819 373.891 Q325.92 372.179 322.666 372.179 Q321.143 372.179 319.619 372.518 Q318.115 372.856 316.535 373.571 L316.535 359.465 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M121.994 67.0365 L138.314 67.0365 L138.314 70.9717 L116.37 70.9717 L116.37 67.0365 Q119.032 64.2819 123.615 59.6523 Q128.221 54.9996 129.402 53.657 Q131.647 51.1338 132.527 49.3977 Q133.43 47.6385 133.43 45.9487 Q133.43 43.1941 131.485 41.458 Q129.564 39.7219 126.462 39.7219 Q124.263 39.7219 121.809 40.4857 Q119.379 41.2496 116.601 42.8005 L116.601 38.0784 Q119.425 36.9441 121.879 36.3654 Q124.332 35.7867 126.369 35.7867 Q131.74 35.7867 134.934 38.4719 Q138.129 41.157 138.129 45.6478 Q138.129 47.7774 137.318 49.6987 Q136.531 51.5968 134.425 54.1894 Q133.846 54.8607 130.744 58.0782 Q127.643 61.2727 121.994 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M148.129 65.0921 L153.013 65.0921 L153.013 70.9717 L148.129 70.9717 L148.129 65.0921 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M173.198 39.4904 Q169.587 39.4904 167.758 43.0552 Q165.953 46.5968 165.953 53.7264 Q165.953 60.8329 167.758 64.3977 Q169.587 67.9393 173.198 67.9393 Q176.832 67.9393 178.638 64.3977 Q180.466 60.8329 180.466 53.7264 Q180.466 46.5968 178.638 43.0552 Q176.832 39.4904 173.198 39.4904 M173.198 35.7867 Q179.008 35.7867 182.064 40.3932 Q185.142 44.9765 185.142 53.7264 Q185.142 62.4532 182.064 67.0597 Q179.008 71.643 173.198 71.643 Q167.388 71.643 164.309 67.0597 Q161.254 62.4532 161.254 53.7264 Q161.254 44.9765 164.309 40.3932 Q167.388 35.7867 173.198 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M203.36 39.4904 Q199.749 39.4904 197.92 43.0552 Q196.114 46.5968 196.114 53.7264 Q196.114 60.8329 197.92 64.3977 Q199.749 67.9393 203.36 67.9393 Q206.994 67.9393 208.8 64.3977 Q210.628 60.8329 210.628 53.7264 Q210.628 46.5968 208.8 43.0552 Q206.994 39.4904 203.36 39.4904 M203.36 35.7867 Q209.17 35.7867 212.225 40.3932 Q215.304 44.9765 215.304 53.7264 Q215.304 62.4532 212.225 67.0597 Q209.17 71.643 203.36 71.643 Q197.55 71.643 194.471 67.0597 Q191.415 62.4532 191.415 53.7264 Q191.415 44.9765 194.471 40.3932 Q197.55 35.7867 203.36 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M251.693 45.5089 L241.114 56.1338 L251.693 66.7125 L248.938 69.5134 L238.313 58.8884 L227.688 69.5134 L224.957 66.7125 L235.512 56.1338 L224.957 45.5089 L227.688 42.708 L238.313 53.3329 L248.938 42.708 L251.693 45.5089 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M264.054 67.0365 L271.693 67.0365 L271.693 40.6709 L263.383 42.3376 L263.383 38.0784 L271.646 36.4117 L276.322 36.4117 L276.322 67.0365 L283.961 67.0365 L283.961 70.9717 L264.054 70.9717 L264.054 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M303.406 39.4904 Q299.794 39.4904 297.966 43.0552 Q296.16 46.5968 296.16 53.7264 Q296.16 60.8329 297.966 64.3977 Q299.794 67.9393 303.406 67.9393 Q307.04 67.9393 308.845 64.3977 Q310.674 60.8329 310.674 53.7264 Q310.674 46.5968 308.845 43.0552 Q307.04 39.4904 303.406 39.4904 M303.406 35.7867 Q309.216 35.7867 312.271 40.3932 Q315.35 44.9765 315.35 53.7264 Q315.35 62.4532 312.271 67.0597 Q309.216 71.643 303.406 71.643 Q297.595 71.643 294.517 67.0597 Q291.461 62.4532 291.461 53.7264 Q291.461 44.9765 294.517 40.3932 Q297.595 35.7867 303.406 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M316.535 15.4813 L331.449 15.4813 L331.449 18.6786 L320.014 18.6786 L320.014 25.5623 Q320.842 25.2801 321.669 25.1485 Q322.497 24.998 323.324 24.998 Q328.026 24.998 330.772 27.5747 Q333.518 30.1514 333.518 34.5524 Q333.518 39.085 330.697 41.6053 Q327.876 44.1067 322.741 44.1067 Q320.973 44.1067 319.13 43.8058 Q317.306 43.5049 315.35 42.903 L315.35 39.085 Q317.043 40.0066 318.848 40.458 Q320.654 40.9094 322.666 40.9094 Q325.92 40.9094 327.819 39.1979 Q329.719 37.4864 329.719 34.5524 Q329.719 31.6184 327.819 29.9069 Q325.92 28.1953 322.666 28.1953 Q321.143 28.1953 319.619 28.5339 Q318.115 28.8724 316.535 29.5871 L316.535 15.4813 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M21.7677 896.217 L39.6235 896.217 L39.6235 888.132 Q39.6235 883.644 37.3 881.194 Q34.9765 878.743 30.6797 878.743 Q26.4147 878.743 24.0912 881.194 Q21.7677 883.644 21.7677 888.132 L21.7677 896.217 M16.4842 902.646 L16.4842 888.132 Q16.4842 880.143 20.1126 876.069 Q23.7092 871.963 30.6797 871.963 Q37.7138 871.963 41.3104 876.069 Q44.907 880.143 44.907 888.132 L44.907 896.217 L64.0042 896.217 L64.0042 902.646 L16.4842 902.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M33.8307 844.081 Q33.2578 845.068 33.0032 846.246 Q32.7167 847.392 32.7167 848.792 Q32.7167 853.757 35.9632 856.431 Q39.1779 859.073 45.2253 859.073 L64.0042 859.073 L64.0042 864.961 L28.3562 864.961 L28.3562 859.073 L33.8944 859.073 Q30.6479 857.227 29.0883 854.267 Q27.4968 851.307 27.4968 847.073 Q27.4968 846.469 27.5923 845.737 Q27.656 845.004 27.8151 844.113 L33.8307 844.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M32.4621 825.557 Q32.4621 830.268 36.1542 833.005 Q39.8145 835.742 46.212 835.742 Q52.6095 835.742 56.3017 833.037 Q59.9619 830.3 59.9619 825.557 Q59.9619 820.878 56.2698 818.141 Q52.5777 815.404 46.212 815.404 Q39.8781 815.404 36.186 818.141 Q32.4621 820.878 32.4621 825.557 M27.4968 825.557 Q27.4968 817.918 32.4621 813.558 Q37.4273 809.197 46.212 809.197 Q54.9649 809.197 59.9619 813.558 Q64.9272 817.918 64.9272 825.557 Q64.9272 833.228 59.9619 837.588 Q54.9649 841.917 46.212 841.917 Q37.4273 841.917 32.4621 837.588 Q27.4968 833.228 27.4968 825.557 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M33.7671 776.032 L14.479 776.032 L14.479 770.176 L64.0042 770.176 L64.0042 776.032 L58.657 776.032 Q61.8398 777.878 63.3994 780.711 Q64.9272 783.512 64.9272 787.459 Q64.9272 793.92 59.771 797.994 Q54.6147 802.036 46.212 802.036 Q37.8093 802.036 32.6531 797.994 Q27.4968 793.92 27.4968 787.459 Q27.4968 783.512 29.0564 780.711 Q30.5842 777.878 33.7671 776.032 M46.212 795.989 Q52.6732 795.989 56.3653 793.347 Q60.0256 790.673 60.0256 786.026 Q60.0256 781.379 56.3653 778.706 Q52.6732 776.032 46.212 776.032 Q39.7508 776.032 36.0905 778.706 Q32.3984 781.379 32.3984 786.026 Q32.3984 790.673 36.0905 793.347 Q39.7508 795.989 46.212 795.989 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M49.9359 758.717 L28.3562 758.717 L28.3562 752.861 L49.7131 752.861 Q54.7739 752.861 57.3202 750.888 Q59.8346 748.914 59.8346 744.967 Q59.8346 740.225 56.8109 737.488 Q53.7872 734.719 48.5673 734.719 L28.3562 734.719 L28.3562 728.862 L64.0042 728.862 L64.0042 734.719 L58.5296 734.719 Q61.7762 736.851 63.3676 739.684 Q64.9272 742.485 64.9272 746.209 Q64.9272 752.352 61.1078 755.534 Q57.2883 758.717 49.9359 758.717 M27.4968 743.981 L27.4968 743.981 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M29.7248 691.145 L35.1993 691.145 Q33.8307 693.628 33.1623 696.142 Q32.4621 698.625 32.4621 701.171 Q32.4621 706.869 36.0905 710.02 Q39.6872 713.171 46.212 713.171 Q52.7369 713.171 56.3653 710.02 Q59.9619 706.869 59.9619 701.171 Q59.9619 698.625 59.2935 696.142 Q58.5933 693.628 57.2247 691.145 L62.6355 691.145 Q63.7814 693.596 64.3543 696.238 Q64.9272 698.848 64.9272 701.808 Q64.9272 709.861 59.8664 714.603 Q54.8057 719.345 46.212 719.345 Q37.491 719.345 32.4939 714.571 Q27.4968 709.765 27.4968 701.426 Q27.4968 698.721 28.0697 696.142 Q28.6108 693.564 29.7248 691.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M18.2347 675.167 L28.3562 675.167 L28.3562 663.104 L32.9077 663.104 L32.9077 675.167 L52.2594 675.167 Q56.6199 675.167 57.8613 673.99 Q59.1026 672.78 59.1026 669.12 L59.1026 663.104 L64.0042 663.104 L64.0042 669.12 Q64.0042 675.899 61.4897 678.478 Q58.9434 681.056 52.2594 681.056 L32.9077 681.056 L32.9077 685.353 L28.3562 685.353 L28.3562 681.056 L18.2347 681.056 L18.2347 675.167 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M28.3562 655.402 L28.3562 649.545 L64.0042 649.545 L64.0042 655.402 L28.3562 655.402 M14.479 655.402 L14.479 649.545 L21.895 649.545 L21.895 655.402 L14.479 655.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M32.4621 623.478 Q32.4621 628.189 36.1542 630.926 Q39.8145 633.663 46.212 633.663 Q52.6095 633.663 56.3017 630.958 Q59.9619 628.22 59.9619 623.478 Q59.9619 618.799 56.2698 616.062 Q52.5777 613.325 46.212 613.325 Q39.8781 613.325 36.186 616.062 Q32.4621 618.799 32.4621 623.478 M27.4968 623.478 Q27.4968 615.839 32.4621 611.479 Q37.4273 607.118 46.212 607.118 Q54.9649 607.118 59.9619 611.479 Q64.9272 615.839 64.9272 623.478 Q64.9272 631.149 59.9619 635.509 Q54.9649 639.838 46.212 639.838 Q37.4273 639.838 32.4621 635.509 Q27.4968 631.149 27.4968 623.478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M42.4881 567.778 L64.0042 567.778 L64.0042 573.634 L42.679 573.634 Q37.6183 573.634 35.1038 575.608 Q32.5894 577.581 32.5894 581.528 Q32.5894 586.27 35.6131 589.008 Q38.6368 591.745 43.8567 591.745 L64.0042 591.745 L64.0042 597.633 L28.3562 597.633 L28.3562 591.745 L33.8944 591.745 Q30.6797 589.644 29.0883 586.811 Q27.4968 583.947 27.4968 580.223 Q27.4968 574.08 31.3163 570.929 Q35.1038 567.778 42.4881 567.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip612)" d="M478.6 597.618 L478.6 597.618 L598.266 597.618 L598.266 597.618 L478.6 597.618 L478.6 597.618  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,597.618 478.6,597.618 598.266,597.618 478.6,597.618 "/>
+<path clip-path="url(#clip612)" d="M628.182 735.212 L628.182 735.212 L747.848 735.212 L747.848 735.212 L628.182 735.212 L628.182 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,735.212 628.182,735.212 747.848,735.212 628.182,735.212 "/>
+<path clip-path="url(#clip612)" d="M777.765 528.822 L777.765 528.822 L897.431 528.822 L897.431 528.822 L777.765 528.822 L777.765 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,528.822 777.765,528.822 897.431,528.822 777.765,528.822 "/>
+<path clip-path="url(#clip612)" d="M927.348 528.822 L927.348 528.822 L1047.01 528.822 L1047.01 528.822 L927.348 528.822 L927.348 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,528.822 927.348,528.822 1047.01,528.822 927.348,528.822 "/>
+<path clip-path="url(#clip612)" d="M1076.93 460.025 L1076.93 460.025 L1196.6 460.025 L1196.6 460.025 L1076.93 460.025 L1076.93 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,460.025 1076.93,460.025 1196.6,460.025 1076.93,460.025 "/>
+<path clip-path="url(#clip612)" d="M1226.51 528.822 L1226.51 528.822 L1346.18 528.822 L1346.18 528.822 L1226.51 528.822 L1226.51 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,528.822 1226.51,528.822 1346.18,528.822 1226.51,528.822 "/>
+<path clip-path="url(#clip612)" d="M1376.1 391.228 L1376.1 391.228 L1495.76 391.228 L1495.76 391.228 L1376.1 391.228 L1376.1 391.228  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,391.228 1376.1,391.228 1495.76,391.228 1376.1,391.228 "/>
+<path clip-path="url(#clip612)" d="M1525.68 253.634 L1525.68 253.634 L1645.34 253.634 L1645.34 253.634 L1525.68 253.634 L1525.68 253.634  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,253.634 1525.68,253.634 1645.34,253.634 1525.68,253.634 "/>
+<path clip-path="url(#clip612)" d="M1675.26 47.2441 L1675.26 47.2441 L1794.93 47.2441 L1794.93 47.2441 L1675.26 47.2441 L1675.26 47.2441  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,47.2441 1675.26,47.2441 1794.93,47.2441 1675.26,47.2441 "/>
+<path clip-path="url(#clip612)" d="M1824.84 116.041 L1824.84 116.041 L1944.51 116.041 L1944.51 116.041 L1824.84 116.041 L1824.84 116.041  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,116.041 1824.84,116.041 1944.51,116.041 1824.84,116.041 "/>
+<path clip-path="url(#clip612)" d="M1974.43 460.025 L1974.43 460.025 L2094.09 460.025 L2094.09 460.025 L1974.43 460.025 L1974.43 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,460.025 1974.43,460.025 2094.09,460.025 1974.43,460.025 "/>
+<path clip-path="url(#clip612)" d="M2124.01 735.212 L2124.01 735.212 L2243.67 735.212 L2243.67 735.212 L2124.01 735.212 L2124.01 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,735.212 2124.01,735.212 2243.67,735.212 2124.01,735.212 "/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="538.433" cy="597.618" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="688.015" cy="735.212" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="837.598" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="987.181" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1136.76" cy="460.025" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1286.35" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1435.93" cy="391.228" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1585.51" cy="253.634" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1735.09" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1884.68" cy="116.041" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2034.26" cy="460.025" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2183.84" cy="735.212" r="2"/>
+<path clip-path="url(#clip612)" d="M478.6 597.618 L478.6 1079.2 L598.266 1079.2 L598.266 597.618 L478.6 597.618 L478.6 597.618  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,597.618 478.6,1079.2 598.266,1079.2 598.266,597.618 478.6,597.618 "/>
+<path clip-path="url(#clip612)" d="M628.182 735.212 L628.182 1044.8 L747.848 1044.8 L747.848 735.212 L628.182 735.212 L628.182 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,735.212 628.182,1044.8 747.848,1044.8 747.848,735.212 628.182,735.212 "/>
+<path clip-path="url(#clip612)" d="M777.765 528.822 L777.765 1010.4 L897.431 1010.4 L897.431 528.822 L777.765 528.822 L777.765 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,528.822 777.765,1010.4 897.431,1010.4 897.431,528.822 777.765,528.822 "/>
+<path clip-path="url(#clip612)" d="M927.348 528.822 L927.348 735.212 L1047.01 735.212 L1047.01 528.822 L927.348 528.822 L927.348 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,528.822 927.348,735.212 1047.01,735.212 1047.01,528.822 927.348,528.822 "/>
+<path clip-path="url(#clip612)" d="M1076.93 460.025 L1076.93 1423.18 L1196.6 1423.18 L1196.6 460.025 L1076.93 460.025 L1076.93 460.025  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,460.025 1076.93,1423.18 1196.6,1423.18 1196.6,460.025 1076.93,460.025 "/>
+<path clip-path="url(#clip612)" d="M1226.51 528.822 L1226.51 941.602 L1346.18 941.602 L1346.18 528.822 L1226.51 528.822 L1226.51 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,528.822 1226.51,941.602 1346.18,941.602 1346.18,528.822 1226.51,528.822 "/>
+<path clip-path="url(#clip612)" d="M1376.1 391.228 L1376.1 1010.4 L1495.76 1010.4 L1495.76 391.228 L1376.1 391.228 L1376.1 391.228  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,391.228 1376.1,1010.4 1495.76,1010.4 1495.76,391.228 1376.1,391.228 "/>
+<path clip-path="url(#clip612)" d="M1525.68 253.634 L1525.68 735.212 L1645.34 735.212 L1645.34 253.634 L1525.68 253.634 L1525.68 253.634  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,253.634 1525.68,735.212 1645.34,735.212 1645.34,253.634 1525.68,253.634 "/>
+<path clip-path="url(#clip612)" d="M1675.26 47.2441 L1675.26 735.212 L1794.93 735.212 L1794.93 47.2441 L1675.26 47.2441 L1675.26 47.2441  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,47.2441 1675.26,735.212 1794.93,735.212 1794.93,47.2441 1675.26,47.2441 "/>
+<path clip-path="url(#clip612)" d="M1824.84 116.041 L1824.84 1423.18 L1944.51 1423.18 L1944.51 116.041 L1824.84 116.041 L1824.84 116.041  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,116.041 1824.84,1423.18 1944.51,1423.18 1944.51,116.041 1824.84,116.041 "/>
+<path clip-path="url(#clip612)" d="M1974.43 460.025 L1974.43 1010.4 L2094.09 1010.4 L2094.09 460.025 L1974.43 460.025 L1974.43 460.025  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,460.025 1974.43,1010.4 2094.09,1010.4 2094.09,460.025 1974.43,460.025 "/>
+<path clip-path="url(#clip612)" d="M2124.01 735.212 L2124.01 1423.18 L2243.67 1423.18 L2243.67 735.212 L2124.01 735.212 L2124.01 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,735.212 2124.01,1423.18 2243.67,1423.18 2243.67,735.212 2124.01,735.212 "/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="538.433" cy="597.618" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="688.015" cy="735.212" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="837.598" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="987.181" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1136.76" cy="460.025" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1286.35" cy="528.822" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1435.93" cy="391.228" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1585.51" cy="253.634" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1735.09" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1884.68" cy="116.041" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2034.26" cy="460.025" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2183.84" cy="735.212" r="2"/>
+<path clip-path="url(#clip612)" d="M478.6 1079.2 L478.6 1423.18 L598.266 1423.18 L598.266 1079.2 L478.6 1079.2 L478.6 1079.2  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="478.6,1079.2 478.6,1423.18 598.266,1423.18 598.266,1079.2 478.6,1079.2 "/>
+<path clip-path="url(#clip612)" d="M628.182 1044.8 L628.182 1423.18 L747.848 1423.18 L747.848 1044.8 L628.182 1044.8 L628.182 1044.8  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="628.182,1044.8 628.182,1423.18 747.848,1423.18 747.848,1044.8 628.182,1044.8 "/>
+<path clip-path="url(#clip612)" d="M777.765 1010.4 L777.765 1423.18 L897.431 1423.18 L897.431 1010.4 L777.765 1010.4 L777.765 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="777.765,1010.4 777.765,1423.18 897.431,1423.18 897.431,1010.4 777.765,1010.4 "/>
+<path clip-path="url(#clip612)" d="M927.348 735.212 L927.348 1423.18 L1047.01 1423.18 L1047.01 735.212 L927.348 735.212 L927.348 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="927.348,735.212 927.348,1423.18 1047.01,1423.18 1047.01,735.212 927.348,735.212 "/>
+<path clip-path="url(#clip612)" d="M1076.93 1423.18 L1076.93 1423.18 L1196.6 1423.18 L1196.6 1423.18 L1076.93 1423.18 L1076.93 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1076.93,1423.18 1076.93,1423.18 1196.6,1423.18 1076.93,1423.18 "/>
+<path clip-path="url(#clip612)" d="M1226.51 941.602 L1226.51 1423.18 L1346.18 1423.18 L1346.18 941.602 L1226.51 941.602 L1226.51 941.602  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1226.51,941.602 1226.51,1423.18 1346.18,1423.18 1346.18,941.602 1226.51,941.602 "/>
+<path clip-path="url(#clip612)" d="M1376.1 1010.4 L1376.1 1423.18 L1495.76 1423.18 L1495.76 1010.4 L1376.1 1010.4 L1376.1 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1376.1,1010.4 1376.1,1423.18 1495.76,1423.18 1495.76,1010.4 1376.1,1010.4 "/>
+<path clip-path="url(#clip612)" d="M1525.68 735.212 L1525.68 1423.18 L1645.34 1423.18 L1645.34 735.212 L1525.68 735.212 L1525.68 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1525.68,735.212 1525.68,1423.18 1645.34,1423.18 1645.34,735.212 1525.68,735.212 "/>
+<path clip-path="url(#clip612)" d="M1675.26 735.212 L1675.26 1423.18 L1794.93 1423.18 L1794.93 735.212 L1675.26 735.212 L1675.26 735.212  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1675.26,735.212 1675.26,1423.18 1794.93,1423.18 1794.93,735.212 1675.26,735.212 "/>
+<path clip-path="url(#clip612)" d="M1824.84 1423.18 L1824.84 1423.18 L1944.51 1423.18 L1944.51 1423.18 L1824.84 1423.18 L1824.84 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1824.84,1423.18 1824.84,1423.18 1944.51,1423.18 1824.84,1423.18 "/>
+<path clip-path="url(#clip612)" d="M1974.43 1010.4 L1974.43 1423.18 L2094.09 1423.18 L2094.09 1010.4 L1974.43 1010.4 L1974.43 1010.4  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1974.43,1010.4 1974.43,1423.18 2094.09,1423.18 2094.09,1010.4 1974.43,1010.4 "/>
+<path clip-path="url(#clip612)" d="M2124.01 1423.18 L2124.01 1423.18 L2243.67 1423.18 L2243.67 1423.18 L2124.01 1423.18 L2124.01 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip612)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2124.01,1423.18 2124.01,1423.18 2243.67,1423.18 2124.01,1423.18 "/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="538.433" cy="1079.2" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="688.015" cy="1044.8" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="837.598" cy="1010.4" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="987.181" cy="735.212" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1136.76" cy="1423.18" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1286.35" cy="941.602" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1435.93" cy="1010.4" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1585.51" cy="735.212" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1735.09" cy="735.212" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1884.68" cy="1423.18" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2034.26" cy="1010.4" r="2"/>
+<circle clip-path="url(#clip612)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2183.84" cy="1423.18" r="2"/>
+<path clip-path="url(#clip610)" d="M435.626 300.469 L989.342 300.469 L989.342 93.1086 L435.626 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="435.626,300.469 989.342,300.469 989.342,93.1086 435.626,93.1086 435.626,300.469 "/>
+<path clip-path="url(#clip610)" d="M457.662 165.685 L589.878 165.685 L589.878 124.213 L457.662 124.213 L457.662 165.685  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,165.685 589.878,165.685 589.878,124.213 457.662,124.213 457.662,165.685 "/>
+<path clip-path="url(#clip610)" d="M611.914 151.997 L611.914 136.303 L616.173 136.303 L616.173 151.835 Q616.173 155.516 617.608 157.368 Q619.044 159.196 621.914 159.196 Q625.363 159.196 627.354 156.997 Q629.368 154.798 629.368 151.002 L629.368 136.303 L633.627 136.303 L633.627 162.229 L629.368 162.229 L629.368 158.247 Q627.817 160.608 625.756 161.766 Q623.719 162.9 621.011 162.9 Q616.544 162.9 614.229 160.122 Q611.914 157.344 611.914 151.997 M622.631 135.678 L622.631 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M663.951 146.581 L663.951 162.229 L659.691 162.229 L659.691 146.719 Q659.691 143.039 658.256 141.21 Q656.821 139.382 653.951 139.382 Q650.502 139.382 648.511 141.581 Q646.52 143.78 646.52 147.576 L646.52 162.229 L642.238 162.229 L642.238 136.303 L646.52 136.303 L646.52 140.331 Q648.048 137.993 650.108 136.835 Q652.192 135.678 654.9 135.678 Q659.367 135.678 661.659 138.456 Q663.951 141.21 663.951 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M692.631 141.28 Q694.228 138.409 696.451 137.044 Q698.673 135.678 701.682 135.678 Q705.733 135.678 707.932 138.525 Q710.131 141.349 710.131 146.581 L710.131 162.229 L705.849 162.229 L705.849 146.719 Q705.849 142.993 704.529 141.187 Q703.21 139.382 700.501 139.382 Q697.191 139.382 695.27 141.581 Q693.349 143.78 693.349 147.576 L693.349 162.229 L689.066 162.229 L689.066 146.719 Q689.066 142.969 687.747 141.187 Q686.427 139.382 683.673 139.382 Q680.409 139.382 678.488 141.604 Q676.566 143.803 676.566 147.576 L676.566 162.229 L672.284 162.229 L672.284 136.303 L676.566 136.303 L676.566 140.331 Q678.025 137.946 680.062 136.812 Q682.099 135.678 684.9 135.678 Q687.724 135.678 689.691 137.113 Q691.682 138.548 692.631 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M740.802 148.201 L740.802 150.284 L721.219 150.284 Q721.497 154.682 723.858 156.997 Q726.242 159.289 730.478 159.289 Q732.932 159.289 735.223 158.687 Q737.538 158.085 739.807 156.881 L739.807 160.909 Q737.515 161.881 735.108 162.391 Q732.7 162.9 730.224 162.9 Q724.02 162.9 720.386 159.289 Q716.775 155.678 716.775 149.52 Q716.775 143.155 720.2 139.428 Q723.649 135.678 729.483 135.678 Q734.714 135.678 737.747 139.057 Q740.802 142.414 740.802 148.201 M736.543 146.951 Q736.497 143.456 734.575 141.372 Q732.677 139.289 729.529 139.289 Q725.964 139.289 723.812 141.303 Q721.682 143.317 721.358 146.974 L736.543 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M752.006 128.942 L752.006 136.303 L760.779 136.303 L760.779 139.613 L752.006 139.613 L752.006 153.687 Q752.006 156.858 752.862 157.761 Q753.742 158.664 756.404 158.664 L760.779 158.664 L760.779 162.229 L756.404 162.229 Q751.473 162.229 749.598 160.4 Q747.723 158.548 747.723 153.687 L747.723 139.613 L744.598 139.613 L744.598 136.303 L747.723 136.303 L747.723 128.942 L752.006 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M798.51 140.238 L798.51 126.21 L802.769 126.21 L802.769 162.229 L798.51 162.229 L798.51 158.34 Q797.168 160.655 795.107 161.789 Q793.07 162.9 790.2 162.9 Q785.501 162.9 782.538 159.15 Q779.598 155.4 779.598 149.289 Q779.598 143.178 782.538 139.428 Q785.501 135.678 790.2 135.678 Q793.07 135.678 795.107 136.812 Q797.168 137.923 798.51 140.238 M783.996 149.289 Q783.996 153.988 785.918 156.673 Q787.862 159.335 791.242 159.335 Q794.621 159.335 796.566 156.673 Q798.51 153.988 798.51 149.289 Q798.51 144.59 796.566 141.928 Q794.621 139.243 791.242 139.243 Q787.862 139.243 785.918 141.928 Q783.996 144.59 783.996 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M833.718 148.201 L833.718 150.284 L814.135 150.284 Q814.413 154.682 816.774 156.997 Q819.158 159.289 823.394 159.289 Q825.848 159.289 828.14 158.687 Q830.454 158.085 832.723 156.881 L832.723 160.909 Q830.431 161.881 828.024 162.391 Q825.617 162.9 823.14 162.9 Q816.936 162.9 813.302 159.289 Q809.691 155.678 809.691 149.52 Q809.691 143.155 813.117 139.428 Q816.566 135.678 822.399 135.678 Q827.63 135.678 830.663 139.057 Q833.718 142.414 833.718 148.201 M829.459 146.951 Q829.413 143.456 827.492 141.372 Q825.593 139.289 822.445 139.289 Q818.88 139.289 816.728 141.303 Q814.598 143.317 814.274 146.974 L829.459 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M860.894 141.28 Q862.491 138.409 864.714 137.044 Q866.936 135.678 869.945 135.678 Q873.996 135.678 876.195 138.525 Q878.394 141.349 878.394 146.581 L878.394 162.229 L874.112 162.229 L874.112 146.719 Q874.112 142.993 872.792 141.187 Q871.473 139.382 868.764 139.382 Q865.454 139.382 863.533 141.581 Q861.612 143.78 861.612 147.576 L861.612 162.229 L857.329 162.229 L857.329 146.719 Q857.329 142.969 856.01 141.187 Q854.69 139.382 851.936 139.382 Q848.672 139.382 846.751 141.604 Q844.829 143.803 844.829 147.576 L844.829 162.229 L840.547 162.229 L840.547 136.303 L844.829 136.303 L844.829 140.331 Q846.288 137.946 848.325 136.812 Q850.362 135.678 853.163 135.678 Q855.987 135.678 857.954 137.113 Q859.945 138.548 860.894 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M898.672 149.196 Q893.51 149.196 891.519 150.377 Q889.528 151.557 889.528 154.405 Q889.528 156.673 891.01 158.016 Q892.514 159.335 895.084 159.335 Q898.625 159.335 900.755 156.835 Q902.908 154.312 902.908 150.145 L902.908 149.196 L898.672 149.196 M907.167 147.437 L907.167 162.229 L902.908 162.229 L902.908 158.293 Q901.449 160.655 899.274 161.789 Q897.098 162.9 893.949 162.9 Q889.968 162.9 887.607 160.678 Q885.269 158.432 885.269 154.682 Q885.269 150.307 888.186 148.085 Q891.125 145.863 896.936 145.863 L902.908 145.863 L902.908 145.446 Q902.908 142.507 900.963 140.909 Q899.042 139.289 895.547 139.289 Q893.324 139.289 891.218 139.821 Q889.112 140.354 887.167 141.419 L887.167 137.483 Q889.505 136.581 891.704 136.141 Q893.903 135.678 895.987 135.678 Q901.611 135.678 904.389 138.594 Q907.167 141.511 907.167 147.437 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M937.491 146.581 L937.491 162.229 L933.232 162.229 L933.232 146.719 Q933.232 143.039 931.797 141.21 Q930.361 139.382 927.491 139.382 Q924.042 139.382 922.051 141.581 Q920.06 143.78 920.06 147.576 L920.06 162.229 L915.778 162.229 L915.778 136.303 L920.06 136.303 L920.06 140.331 Q921.588 137.993 923.648 136.835 Q925.732 135.678 928.44 135.678 Q932.908 135.678 935.199 138.456 Q937.491 141.21 937.491 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M963.046 140.238 L963.046 126.21 L967.306 126.21 L967.306 162.229 L963.046 162.229 L963.046 158.34 Q961.704 160.655 959.644 161.789 Q957.607 162.9 954.736 162.9 Q950.037 162.9 947.074 159.15 Q944.134 155.4 944.134 149.289 Q944.134 143.178 947.074 139.428 Q950.037 135.678 954.736 135.678 Q957.607 135.678 959.644 136.812 Q961.704 137.923 963.046 140.238 M948.533 149.289 Q948.533 153.988 950.454 156.673 Q952.398 159.335 955.778 159.335 Q959.157 159.335 961.102 156.673 Q963.046 153.988 963.046 149.289 Q963.046 144.59 961.102 141.928 Q959.157 139.243 955.778 139.243 Q952.398 139.243 950.454 141.928 Q948.533 144.59 948.533 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M457.662 217.525 L589.878 217.525 L589.878 176.053 L457.662 176.053 L457.662 217.525  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,217.525 589.878,217.525 589.878,176.053 457.662,176.053 457.662,217.525 "/>
+<path clip-path="url(#clip610)" d="M627.747 184.115 L621.405 201.314 L634.113 201.314 L627.747 184.115 M625.108 179.509 L630.409 179.509 L643.58 214.069 L638.719 214.069 L635.571 205.203 L619.993 205.203 L616.844 214.069 L611.914 214.069 L625.108 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip610)" d="M457.662 269.365 L589.878 269.365 L589.878 227.893 L457.662 227.893 L457.662 269.365  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip610)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="457.662,269.365 589.878,269.365 589.878,227.893 457.662,227.893 457.662,269.365 "/>
+<path clip-path="url(#clip610)" d="M616.59 249.404 L616.59 262.066 L624.09 262.066 Q627.863 262.066 629.668 260.515 Q631.497 258.941 631.497 255.723 Q631.497 252.483 629.668 250.955 Q627.863 249.404 624.09 249.404 L616.59 249.404 M616.59 235.191 L616.59 245.608 L623.511 245.608 Q626.937 245.608 628.604 244.335 Q630.293 243.038 630.293 240.399 Q630.293 237.784 628.604 236.487 Q626.937 235.191 623.511 235.191 L616.59 235.191 M611.914 231.349 L623.858 231.349 Q629.206 231.349 632.099 233.571 Q634.993 235.793 634.993 239.89 Q634.993 243.062 633.511 244.937 Q632.03 246.811 629.159 247.274 Q632.608 248.015 634.506 250.376 Q636.428 252.714 636.428 256.233 Q636.428 260.862 633.28 263.385 Q630.131 265.909 624.321 265.909 L611.914 265.909 L611.914 231.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/linear/factory_schedule/79a4ee80.svg b/previews/PR3562/tutorials/linear/factory_schedule/57fc90bf.svg
similarity index 78%
rename from previews/PR3562/tutorials/linear/factory_schedule/79a4ee80.svg
rename to previews/PR3562/tutorials/linear/factory_schedule/57fc90bf.svg
index c95161d8816..32d8bb1833d 100644
--- a/previews/PR3562/tutorials/linear/factory_schedule/79a4ee80.svg
+++ b/previews/PR3562/tutorials/linear/factory_schedule/57fc90bf.svg
@@ -1,160 +1,160 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip850">
+  <clipPath id="clip670">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip850)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip670)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip851">
+  <clipPath id="clip671">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip850)" d="M337.247 1423.18 L2352.76 1423.18 L2352.76 47.2441 L337.247 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip670)" d="M337.247 1423.18 L2352.76 1423.18 L2352.76 47.2441 L337.247 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip852">
+  <clipPath id="clip672">
     <rect x="337" y="47" width="2017" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="356.894,1423.18 356.894,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="660.927,1423.18 660.927,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="964.96,1423.18 964.96,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1268.99,1423.18 1268.99,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1573.03,1423.18 1573.03,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1877.06,1423.18 1877.06,47.2441 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2181.09,1423.18 2181.09,47.2441 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="356.894,1423.18 356.894,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="660.927,1423.18 660.927,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="964.96,1423.18 964.96,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1268.99,1423.18 1268.99,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1573.03,1423.18 1573.03,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.06,1423.18 1877.06,1404.28 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2181.09,1423.18 2181.09,1404.28 "/>
-<path clip-path="url(#clip850)" d="M356.894 1454.1 Q353.283 1454.1 351.454 1457.66 Q349.648 1461.2 349.648 1468.33 Q349.648 1475.44 351.454 1479.01 Q353.283 1482.55 356.894 1482.55 Q360.528 1482.55 362.333 1479.01 Q364.162 1475.44 364.162 1468.33 Q364.162 1461.2 362.333 1457.66 Q360.528 1454.1 356.894 1454.1 M356.894 1450.39 Q362.704 1450.39 365.759 1455 Q368.838 1459.58 368.838 1468.33 Q368.838 1477.06 365.759 1481.67 Q362.704 1486.25 356.894 1486.25 Q351.083 1486.25 348.005 1481.67 Q344.949 1477.06 344.949 1468.33 Q344.949 1459.58 348.005 1455 Q351.083 1450.39 356.894 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M655.58 1481.64 L671.899 1481.64 L671.899 1485.58 L649.955 1485.58 L649.955 1481.64 Q652.617 1478.89 657.2 1474.26 Q661.806 1469.61 662.987 1468.27 Q665.232 1465.74 666.112 1464.01 Q667.015 1462.25 667.015 1460.56 Q667.015 1457.8 665.07 1456.07 Q663.149 1454.33 660.047 1454.33 Q657.848 1454.33 655.394 1455.09 Q652.964 1455.86 650.186 1457.41 L650.186 1452.69 Q653.01 1451.55 655.464 1450.97 Q657.918 1450.39 659.955 1450.39 Q665.325 1450.39 668.519 1453.08 Q671.714 1455.77 671.714 1460.26 Q671.714 1462.39 670.904 1464.31 Q670.117 1466.2 668.01 1468.8 Q667.431 1469.47 664.33 1472.69 Q661.228 1475.88 655.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M967.969 1455.09 L956.164 1473.54 L967.969 1473.54 L967.969 1455.09 M966.742 1451.02 L972.622 1451.02 L972.622 1473.54 L977.553 1473.54 L977.553 1477.43 L972.622 1477.43 L972.622 1485.58 L967.969 1485.58 L967.969 1477.43 L952.367 1477.43 L952.367 1472.92 L966.742 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1269.4 1466.44 Q1266.25 1466.44 1264.4 1468.59 Q1262.57 1470.74 1262.57 1474.49 Q1262.57 1478.22 1264.4 1480.39 Q1266.25 1482.55 1269.4 1482.55 Q1272.55 1482.55 1274.38 1480.39 Q1276.23 1478.22 1276.23 1474.49 Q1276.23 1470.74 1274.38 1468.59 Q1272.55 1466.44 1269.4 1466.44 M1278.68 1451.78 L1278.68 1456.04 Q1276.92 1455.21 1275.12 1454.77 Q1273.33 1454.33 1271.57 1454.33 Q1266.94 1454.33 1264.49 1457.45 Q1262.06 1460.58 1261.71 1466.9 Q1263.08 1464.89 1265.14 1463.82 Q1267.2 1462.73 1269.68 1462.73 Q1274.88 1462.73 1277.89 1465.9 Q1280.93 1469.05 1280.93 1474.49 Q1280.93 1479.82 1277.78 1483.03 Q1274.63 1486.25 1269.4 1486.25 Q1263.4 1486.25 1260.23 1481.67 Q1257.06 1477.06 1257.06 1468.33 Q1257.06 1460.14 1260.95 1455.28 Q1264.84 1450.39 1271.39 1450.39 Q1273.15 1450.39 1274.93 1450.74 Q1276.74 1451.09 1278.68 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1573.03 1469.17 Q1569.69 1469.17 1567.77 1470.95 Q1565.87 1472.73 1565.87 1475.86 Q1565.87 1478.98 1567.77 1480.77 Q1569.69 1482.55 1573.03 1482.55 Q1576.36 1482.55 1578.28 1480.77 Q1580.2 1478.96 1580.2 1475.86 Q1580.2 1472.73 1578.28 1470.95 Q1576.38 1469.17 1573.03 1469.17 M1568.35 1467.18 Q1565.34 1466.44 1563.65 1464.38 Q1561.98 1462.32 1561.98 1459.35 Q1561.98 1455.21 1564.92 1452.8 Q1567.89 1450.39 1573.03 1450.39 Q1578.19 1450.39 1581.13 1452.8 Q1584.07 1455.21 1584.07 1459.35 Q1584.07 1462.32 1582.38 1464.38 Q1580.71 1466.44 1577.73 1467.18 Q1581.1 1467.96 1582.98 1470.26 Q1584.88 1472.55 1584.88 1475.86 Q1584.88 1480.88 1581.8 1483.57 Q1578.74 1486.25 1573.03 1486.25 Q1567.31 1486.25 1564.23 1483.57 Q1561.17 1480.88 1561.17 1475.86 Q1561.17 1472.55 1563.07 1470.26 Q1564.97 1467.96 1568.35 1467.18 M1566.64 1459.79 Q1566.64 1462.48 1568.3 1463.98 Q1569.99 1465.49 1573.03 1465.49 Q1576.04 1465.49 1577.73 1463.98 Q1579.44 1462.48 1579.44 1459.79 Q1579.44 1457.11 1577.73 1455.6 Q1576.04 1454.1 1573.03 1454.1 Q1569.99 1454.1 1568.3 1455.6 Q1566.64 1457.11 1566.64 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1851.75 1481.64 L1859.39 1481.64 L1859.39 1455.28 L1851.08 1456.95 L1851.08 1452.69 L1859.34 1451.02 L1864.02 1451.02 L1864.02 1481.64 L1871.65 1481.64 L1871.65 1485.58 L1851.75 1485.58 L1851.75 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1891.1 1454.1 Q1887.49 1454.1 1885.66 1457.66 Q1883.85 1461.2 1883.85 1468.33 Q1883.85 1475.44 1885.66 1479.01 Q1887.49 1482.55 1891.1 1482.55 Q1894.73 1482.55 1896.54 1479.01 Q1898.37 1475.44 1898.37 1468.33 Q1898.37 1461.2 1896.54 1457.66 Q1894.73 1454.1 1891.1 1454.1 M1891.1 1450.39 Q1896.91 1450.39 1899.96 1455 Q1903.04 1459.58 1903.04 1468.33 Q1903.04 1477.06 1899.96 1481.67 Q1896.91 1486.25 1891.1 1486.25 Q1885.29 1486.25 1882.21 1481.67 Q1879.15 1477.06 1879.15 1468.33 Q1879.15 1459.58 1882.21 1455 Q1885.29 1450.39 1891.1 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2156.58 1481.64 L2164.22 1481.64 L2164.22 1455.28 L2155.91 1456.95 L2155.91 1452.69 L2164.17 1451.02 L2168.85 1451.02 L2168.85 1481.64 L2176.49 1481.64 L2176.49 1485.58 L2156.58 1485.58 L2156.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2189.96 1481.64 L2206.28 1481.64 L2206.28 1485.58 L2184.33 1485.58 L2184.33 1481.64 Q2187 1478.89 2191.58 1474.26 Q2196.19 1469.61 2197.37 1468.27 Q2199.61 1465.74 2200.49 1464.01 Q2201.39 1462.25 2201.39 1460.56 Q2201.39 1457.8 2199.45 1456.07 Q2197.53 1454.33 2194.43 1454.33 Q2192.23 1454.33 2189.77 1455.09 Q2187.34 1455.86 2184.56 1457.41 L2184.56 1452.69 Q2187.39 1451.55 2189.84 1450.97 Q2192.3 1450.39 2194.33 1450.39 Q2199.7 1450.39 2202.9 1453.08 Q2206.09 1455.77 2206.09 1460.26 Q2206.09 1462.39 2205.28 1464.31 Q2204.5 1466.2 2202.39 1468.8 Q2201.81 1469.47 2198.71 1472.69 Q2195.61 1475.88 2189.96 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1248.82 1520.52 L1258.4 1520.52 L1270.52 1552.86 L1282.71 1520.52 L1292.29 1520.52 L1292.29 1568.04 L1286.02 1568.04 L1286.02 1526.32 L1273.77 1558.91 L1267.31 1558.91 L1255.05 1526.32 L1255.05 1568.04 L1248.82 1568.04 L1248.82 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1318.62 1536.5 Q1313.9 1536.5 1311.17 1540.19 Q1308.43 1543.85 1308.43 1550.25 Q1308.43 1556.65 1311.14 1560.34 Q1313.87 1564 1318.62 1564 Q1323.29 1564 1326.03 1560.31 Q1328.77 1556.62 1328.77 1550.25 Q1328.77 1543.92 1326.03 1540.23 Q1323.29 1536.5 1318.62 1536.5 M1318.62 1531.54 Q1326.25 1531.54 1330.61 1536.5 Q1334.98 1541.47 1334.98 1550.25 Q1334.98 1559 1330.61 1564 Q1326.25 1568.97 1318.62 1568.97 Q1310.94 1568.97 1306.58 1564 Q1302.26 1559 1302.26 1550.25 Q1302.26 1541.47 1306.58 1536.5 Q1310.94 1531.54 1318.62 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1374.32 1546.53 L1374.32 1568.04 L1368.46 1568.04 L1368.46 1546.72 Q1368.46 1541.66 1366.49 1539.14 Q1364.51 1536.63 1360.57 1536.63 Q1355.82 1536.63 1353.09 1539.65 Q1350.35 1542.68 1350.35 1547.9 L1350.35 1568.04 L1344.46 1568.04 L1344.46 1532.4 L1350.35 1532.4 L1350.35 1537.93 Q1352.45 1534.72 1355.28 1533.13 Q1358.15 1531.54 1361.87 1531.54 Q1368.01 1531.54 1371.16 1535.36 Q1374.32 1539.14 1374.32 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1391.79 1522.27 L1391.79 1532.4 L1403.85 1532.4 L1403.85 1536.95 L1391.79 1536.95 L1391.79 1556.3 Q1391.79 1560.66 1392.97 1561.9 Q1394.18 1563.14 1397.84 1563.14 L1403.85 1563.14 L1403.85 1568.04 L1397.84 1568.04 Q1391.06 1568.04 1388.48 1565.53 Q1385.9 1562.98 1385.9 1556.3 L1385.9 1536.95 L1381.6 1536.95 L1381.6 1532.4 L1385.9 1532.4 L1385.9 1522.27 L1391.79 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1441.19 1546.53 L1441.19 1568.04 L1435.33 1568.04 L1435.33 1546.72 Q1435.33 1541.66 1433.36 1539.14 Q1431.38 1536.63 1427.44 1536.63 Q1422.69 1536.63 1419.96 1539.65 Q1417.22 1542.68 1417.22 1547.9 L1417.22 1568.04 L1411.33 1568.04 L1411.33 1518.52 L1417.22 1518.52 L1417.22 1537.93 Q1419.32 1534.72 1422.15 1533.13 Q1425.02 1531.54 1428.74 1531.54 Q1434.89 1531.54 1438.04 1535.36 Q1441.19 1539.14 1441.19 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,964.535 2352.76,964.535 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,505.889 2352.76,505.889 "/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,47.2441 2352.76,47.2441 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 337.247,47.2441 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 356.145,1423.18 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,964.535 356.145,964.535 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,505.889 356.145,505.889 "/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,47.2441 356.145,47.2441 "/>
-<path clip-path="url(#clip850)" d="M289.303 1408.98 Q285.692 1408.98 283.863 1412.54 Q282.057 1416.08 282.057 1423.21 Q282.057 1430.32 283.863 1433.89 Q285.692 1437.43 289.303 1437.43 Q292.937 1437.43 294.742 1433.89 Q296.571 1430.32 296.571 1423.21 Q296.571 1416.08 294.742 1412.54 Q292.937 1408.98 289.303 1408.98 M289.303 1405.27 Q295.113 1405.27 298.168 1409.88 Q301.247 1414.46 301.247 1423.21 Q301.247 1431.94 298.168 1436.55 Q295.113 1441.13 289.303 1441.13 Q283.492 1441.13 280.414 1436.55 Q277.358 1431.94 277.358 1423.21 Q277.358 1414.46 280.414 1409.88 Q283.492 1405.27 289.303 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M116.668 984.327 L124.306 984.327 L124.306 957.961 L115.996 959.628 L115.996 955.369 L124.26 953.702 L128.936 953.702 L128.936 984.327 L136.575 984.327 L136.575 988.262 L116.668 988.262 L116.668 984.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M146.019 982.383 L150.903 982.383 L150.903 988.262 L146.019 988.262 L146.019 982.383 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M171.089 956.781 Q167.477 956.781 165.649 960.346 Q163.843 963.887 163.843 971.017 Q163.843 978.123 165.649 981.688 Q167.477 985.23 171.089 985.23 Q174.723 985.23 176.528 981.688 Q178.357 978.123 178.357 971.017 Q178.357 963.887 176.528 960.346 Q174.723 956.781 171.089 956.781 M171.089 953.077 Q176.899 953.077 179.954 957.684 Q183.033 962.267 183.033 971.017 Q183.033 979.744 179.954 984.35 Q176.899 988.933 171.089 988.933 Q165.278 988.933 162.2 984.35 Q159.144 979.744 159.144 971.017 Q159.144 962.267 162.2 957.684 Q165.278 953.077 171.089 953.077 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M219.422 962.799 L208.843 973.424 L219.422 984.003 L216.667 986.804 L206.042 976.179 L195.417 986.804 L192.686 984.003 L203.241 973.424 L192.686 962.799 L195.417 959.998 L206.042 970.623 L216.667 959.998 L219.422 962.799 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M231.783 984.327 L239.422 984.327 L239.422 957.961 L231.111 959.628 L231.111 955.369 L239.375 953.702 L244.051 953.702 L244.051 984.327 L251.69 984.327 L251.69 988.262 L231.783 988.262 L231.783 984.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M271.134 956.781 Q267.523 956.781 265.695 960.346 Q263.889 963.887 263.889 971.017 Q263.889 978.123 265.695 981.688 Q267.523 985.23 271.134 985.23 Q274.769 985.23 276.574 981.688 Q278.403 978.123 278.403 971.017 Q278.403 963.887 276.574 960.346 Q274.769 956.781 271.134 956.781 M271.134 953.077 Q276.944 953.077 280 957.684 Q283.079 962.267 283.079 971.017 Q283.079 979.744 280 984.35 Q276.944 988.933 271.134 988.933 Q265.324 988.933 262.245 984.35 Q259.19 979.744 259.19 971.017 Q259.19 962.267 262.245 957.684 Q265.324 953.077 271.134 953.077 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M284.264 932.772 L299.178 932.772 L299.178 935.969 L287.743 935.969 L287.743 942.853 Q288.571 942.571 289.398 942.439 Q290.226 942.289 291.053 942.289 Q295.755 942.289 298.501 944.865 Q301.247 947.442 301.247 951.843 Q301.247 956.376 298.426 958.896 Q295.605 961.397 290.47 961.397 Q288.702 961.397 286.859 961.096 Q285.035 960.795 283.079 960.194 L283.079 956.376 Q284.771 957.297 286.577 957.749 Q288.382 958.2 290.395 958.2 Q293.649 958.2 295.548 956.488 Q297.448 954.777 297.448 951.843 Q297.448 948.909 295.548 947.197 Q293.649 945.486 290.395 945.486 Q288.871 945.486 287.348 945.824 Q285.843 946.163 284.264 946.878 L284.264 932.772 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M119.885 525.682 L136.204 525.682 L136.204 529.617 L114.26 529.617 L114.26 525.682 Q116.922 522.927 121.505 518.298 Q126.112 513.645 127.292 512.302 Q129.538 509.779 130.417 508.043 Q131.32 506.284 131.32 504.594 Q131.32 501.839 129.376 500.103 Q127.455 498.367 124.353 498.367 Q122.154 498.367 119.7 499.131 Q117.269 499.895 114.492 501.446 L114.492 496.724 Q117.316 495.589 119.769 495.011 Q122.223 494.432 124.26 494.432 Q129.63 494.432 132.825 497.117 Q136.019 499.802 136.019 504.293 Q136.019 506.423 135.209 508.344 Q134.422 510.242 132.316 512.835 Q131.737 513.506 128.635 516.723 Q125.533 519.918 119.885 525.682 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M146.019 523.737 L150.903 523.737 L150.903 529.617 L146.019 529.617 L146.019 523.737 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M171.089 498.136 Q167.477 498.136 165.649 501.7 Q163.843 505.242 163.843 512.372 Q163.843 519.478 165.649 523.043 Q167.477 526.585 171.089 526.585 Q174.723 526.585 176.528 523.043 Q178.357 519.478 178.357 512.372 Q178.357 505.242 176.528 501.7 Q174.723 498.136 171.089 498.136 M171.089 494.432 Q176.899 494.432 179.954 499.038 Q183.033 503.622 183.033 512.372 Q183.033 521.098 179.954 525.705 Q176.899 530.288 171.089 530.288 Q165.278 530.288 162.2 525.705 Q159.144 521.098 159.144 512.372 Q159.144 503.622 162.2 499.038 Q165.278 494.432 171.089 494.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M219.422 504.154 L208.843 514.779 L219.422 525.358 L216.667 528.159 L206.042 517.534 L195.417 528.159 L192.686 525.358 L203.241 514.779 L192.686 504.154 L195.417 501.353 L206.042 511.978 L216.667 501.353 L219.422 504.154 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M231.783 525.682 L239.422 525.682 L239.422 499.316 L231.111 500.983 L231.111 496.724 L239.375 495.057 L244.051 495.057 L244.051 525.682 L251.69 525.682 L251.69 529.617 L231.783 529.617 L231.783 525.682 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M271.134 498.136 Q267.523 498.136 265.695 501.7 Q263.889 505.242 263.889 512.372 Q263.889 519.478 265.695 523.043 Q267.523 526.585 271.134 526.585 Q274.769 526.585 276.574 523.043 Q278.403 519.478 278.403 512.372 Q278.403 505.242 276.574 501.7 Q274.769 498.136 271.134 498.136 M271.134 494.432 Q276.944 494.432 280 499.038 Q283.079 503.622 283.079 512.372 Q283.079 521.098 280 525.705 Q276.944 530.288 271.134 530.288 Q265.324 530.288 262.245 525.705 Q259.19 521.098 259.19 512.372 Q259.19 503.622 262.245 499.038 Q265.324 494.432 271.134 494.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M284.264 474.127 L299.178 474.127 L299.178 477.324 L287.743 477.324 L287.743 484.208 Q288.571 483.925 289.398 483.794 Q290.226 483.643 291.053 483.643 Q295.755 483.643 298.501 486.22 Q301.247 488.797 301.247 493.198 Q301.247 497.73 298.426 500.251 Q295.605 502.752 290.47 502.752 Q288.702 502.752 286.859 502.451 Q285.035 502.15 283.079 501.548 L283.079 497.73 Q284.771 498.652 286.577 499.103 Q288.382 499.555 290.395 499.555 Q293.649 499.555 295.548 497.843 Q297.448 496.132 297.448 493.198 Q297.448 490.264 295.548 488.552 Q293.649 486.841 290.395 486.841 Q288.871 486.841 287.348 487.179 Q285.843 487.518 284.264 488.232 L284.264 474.127 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M130.024 52.3375 Q133.38 53.0551 135.255 55.3236 Q137.154 57.5921 137.154 60.9255 Q137.154 66.0412 133.635 68.8421 Q130.117 71.643 123.635 71.643 Q121.459 71.643 119.144 71.2032 Q116.853 70.7865 114.399 69.93 L114.399 65.4162 Q116.343 66.5504 118.658 67.1291 Q120.973 67.7078 123.496 67.7078 Q127.894 67.7078 130.186 65.9717 Q132.501 64.2356 132.501 60.9255 Q132.501 57.8699 130.348 56.157 Q128.218 54.4209 124.399 54.4209 L120.371 54.4209 L120.371 50.5783 L124.584 50.5783 Q128.033 50.5783 129.862 49.2126 Q131.691 47.8237 131.691 45.2311 Q131.691 42.5691 129.792 41.157 Q127.917 39.7219 124.399 39.7219 Q122.478 39.7219 120.279 40.1385 Q118.08 40.5552 115.441 41.4348 L115.441 37.2682 Q118.103 36.5274 120.418 36.1571 Q122.755 35.7867 124.816 35.7867 Q130.14 35.7867 133.242 38.2172 Q136.343 40.6246 136.343 44.745 Q136.343 47.6153 134.7 49.6061 Q133.056 51.5737 130.024 52.3375 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M146.019 65.0921 L150.903 65.0921 L150.903 70.9717 L146.019 70.9717 L146.019 65.0921 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M171.089 39.4904 Q167.477 39.4904 165.649 43.0552 Q163.843 46.5968 163.843 53.7264 Q163.843 60.8329 165.649 64.3977 Q167.477 67.9393 171.089 67.9393 Q174.723 67.9393 176.528 64.3977 Q178.357 60.8329 178.357 53.7264 Q178.357 46.5968 176.528 43.0552 Q174.723 39.4904 171.089 39.4904 M171.089 35.7867 Q176.899 35.7867 179.954 40.3932 Q183.033 44.9765 183.033 53.7264 Q183.033 62.4532 179.954 67.0597 Q176.899 71.643 171.089 71.643 Q165.278 71.643 162.2 67.0597 Q159.144 62.4532 159.144 53.7264 Q159.144 44.9765 162.2 40.3932 Q165.278 35.7867 171.089 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M219.422 45.5089 L208.843 56.1338 L219.422 66.7125 L216.667 69.5134 L206.042 58.8884 L195.417 69.5134 L192.686 66.7125 L203.241 56.1338 L192.686 45.5089 L195.417 42.708 L206.042 53.3329 L216.667 42.708 L219.422 45.5089 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M231.783 67.0365 L239.422 67.0365 L239.422 40.6709 L231.111 42.3376 L231.111 38.0784 L239.375 36.4117 L244.051 36.4117 L244.051 67.0365 L251.69 67.0365 L251.69 70.9717 L231.783 70.9717 L231.783 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M271.134 39.4904 Q267.523 39.4904 265.695 43.0552 Q263.889 46.5968 263.889 53.7264 Q263.889 60.8329 265.695 64.3977 Q267.523 67.9393 271.134 67.9393 Q274.769 67.9393 276.574 64.3977 Q278.403 60.8329 278.403 53.7264 Q278.403 46.5968 276.574 43.0552 Q274.769 39.4904 271.134 39.4904 M271.134 35.7867 Q276.944 35.7867 280 40.3932 Q283.079 44.9765 283.079 53.7264 Q283.079 62.4532 280 67.0597 Q276.944 71.643 271.134 71.643 Q265.324 71.643 262.245 67.0597 Q259.19 62.4532 259.19 53.7264 Q259.19 44.9765 262.245 40.3932 Q265.324 35.7867 271.134 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M284.264 15.4813 L299.178 15.4813 L299.178 18.6786 L287.743 18.6786 L287.743 25.5623 Q288.571 25.2801 289.398 25.1485 Q290.226 24.998 291.053 24.998 Q295.755 24.998 298.501 27.5747 Q301.247 30.1514 301.247 34.5524 Q301.247 39.085 298.426 41.6053 Q295.605 44.1067 290.47 44.1067 Q288.702 44.1067 286.859 43.8058 Q285.035 43.5049 283.079 42.903 L283.079 39.085 Q284.771 40.0066 286.577 40.458 Q288.382 40.9094 290.395 40.9094 Q293.649 40.9094 295.548 39.1979 Q297.448 37.4864 297.448 34.5524 Q297.448 31.6184 295.548 29.9069 Q293.649 28.1953 290.395 28.1953 Q288.871 28.1953 287.348 28.5339 Q285.843 28.8724 284.264 29.5871 L284.264 15.4813 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M21.7677 896.217 L39.6235 896.217 L39.6235 888.132 Q39.6235 883.644 37.3 881.194 Q34.9765 878.743 30.6797 878.743 Q26.4147 878.743 24.0912 881.194 Q21.7677 883.644 21.7677 888.132 L21.7677 896.217 M16.4842 902.646 L16.4842 888.132 Q16.4842 880.143 20.1126 876.069 Q23.7092 871.963 30.6797 871.963 Q37.7138 871.963 41.3104 876.069 Q44.907 880.143 44.907 888.132 L44.907 896.217 L64.0042 896.217 L64.0042 902.646 L16.4842 902.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M33.8307 844.081 Q33.2578 845.068 33.0032 846.246 Q32.7167 847.392 32.7167 848.792 Q32.7167 853.757 35.9632 856.431 Q39.1779 859.073 45.2253 859.073 L64.0042 859.073 L64.0042 864.961 L28.3562 864.961 L28.3562 859.073 L33.8944 859.073 Q30.6479 857.227 29.0883 854.267 Q27.4968 851.307 27.4968 847.073 Q27.4968 846.469 27.5923 845.737 Q27.656 845.004 27.8151 844.113 L33.8307 844.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M32.4621 825.557 Q32.4621 830.268 36.1542 833.005 Q39.8145 835.742 46.212 835.742 Q52.6095 835.742 56.3017 833.037 Q59.9619 830.3 59.9619 825.557 Q59.9619 820.878 56.2698 818.141 Q52.5777 815.404 46.212 815.404 Q39.8781 815.404 36.186 818.141 Q32.4621 820.878 32.4621 825.557 M27.4968 825.557 Q27.4968 817.918 32.4621 813.558 Q37.4273 809.197 46.212 809.197 Q54.9649 809.197 59.9619 813.558 Q64.9272 817.918 64.9272 825.557 Q64.9272 833.228 59.9619 837.588 Q54.9649 841.917 46.212 841.917 Q37.4273 841.917 32.4621 837.588 Q27.4968 833.228 27.4968 825.557 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M33.7671 776.032 L14.479 776.032 L14.479 770.176 L64.0042 770.176 L64.0042 776.032 L58.657 776.032 Q61.8398 777.878 63.3994 780.711 Q64.9272 783.512 64.9272 787.459 Q64.9272 793.92 59.771 797.994 Q54.6147 802.036 46.212 802.036 Q37.8093 802.036 32.6531 797.994 Q27.4968 793.92 27.4968 787.459 Q27.4968 783.512 29.0564 780.711 Q30.5842 777.878 33.7671 776.032 M46.212 795.989 Q52.6732 795.989 56.3653 793.347 Q60.0256 790.673 60.0256 786.026 Q60.0256 781.379 56.3653 778.706 Q52.6732 776.032 46.212 776.032 Q39.7508 776.032 36.0905 778.706 Q32.3984 781.379 32.3984 786.026 Q32.3984 790.673 36.0905 793.347 Q39.7508 795.989 46.212 795.989 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M49.9359 758.717 L28.3562 758.717 L28.3562 752.861 L49.7131 752.861 Q54.7739 752.861 57.3202 750.888 Q59.8346 748.914 59.8346 744.967 Q59.8346 740.225 56.8109 737.488 Q53.7872 734.719 48.5673 734.719 L28.3562 734.719 L28.3562 728.862 L64.0042 728.862 L64.0042 734.719 L58.5296 734.719 Q61.7762 736.851 63.3676 739.684 Q64.9272 742.485 64.9272 746.209 Q64.9272 752.352 61.1078 755.534 Q57.2883 758.717 49.9359 758.717 M27.4968 743.981 L27.4968 743.981 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M29.7248 691.145 L35.1993 691.145 Q33.8307 693.628 33.1623 696.142 Q32.4621 698.625 32.4621 701.171 Q32.4621 706.869 36.0905 710.02 Q39.6872 713.171 46.212 713.171 Q52.7369 713.171 56.3653 710.02 Q59.9619 706.869 59.9619 701.171 Q59.9619 698.625 59.2935 696.142 Q58.5933 693.628 57.2247 691.145 L62.6355 691.145 Q63.7814 693.596 64.3543 696.238 Q64.9272 698.848 64.9272 701.808 Q64.9272 709.861 59.8664 714.603 Q54.8057 719.345 46.212 719.345 Q37.491 719.345 32.4939 714.571 Q27.4968 709.765 27.4968 701.426 Q27.4968 698.721 28.0697 696.142 Q28.6108 693.564 29.7248 691.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M18.2347 675.167 L28.3562 675.167 L28.3562 663.104 L32.9077 663.104 L32.9077 675.167 L52.2594 675.167 Q56.6199 675.167 57.8613 673.99 Q59.1026 672.78 59.1026 669.12 L59.1026 663.104 L64.0042 663.104 L64.0042 669.12 Q64.0042 675.899 61.4897 678.478 Q58.9434 681.056 52.2594 681.056 L32.9077 681.056 L32.9077 685.353 L28.3562 685.353 L28.3562 681.056 L18.2347 681.056 L18.2347 675.167 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M28.3562 655.402 L28.3562 649.545 L64.0042 649.545 L64.0042 655.402 L28.3562 655.402 M14.479 655.402 L14.479 649.545 L21.895 649.545 L21.895 655.402 L14.479 655.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M32.4621 623.478 Q32.4621 628.189 36.1542 630.926 Q39.8145 633.663 46.212 633.663 Q52.6095 633.663 56.3017 630.958 Q59.9619 628.22 59.9619 623.478 Q59.9619 618.799 56.2698 616.062 Q52.5777 613.325 46.212 613.325 Q39.8781 613.325 36.186 616.062 Q32.4621 618.799 32.4621 623.478 M27.4968 623.478 Q27.4968 615.839 32.4621 611.479 Q37.4273 607.118 46.212 607.118 Q54.9649 607.118 59.9619 611.479 Q64.9272 615.839 64.9272 623.478 Q64.9272 631.149 59.9619 635.509 Q54.9649 639.838 46.212 639.838 Q37.4273 639.838 32.4621 635.509 Q27.4968 631.149 27.4968 623.478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M42.4881 567.778 L64.0042 567.778 L64.0042 573.634 L42.679 573.634 Q37.6183 573.634 35.1038 575.608 Q32.5894 577.581 32.5894 581.528 Q32.5894 586.27 35.6131 589.008 Q38.6368 591.745 43.8567 591.745 L64.0042 591.745 L64.0042 597.633 L28.3562 597.633 L28.3562 591.745 L33.8944 591.745 Q30.6797 589.644 29.0883 586.811 Q27.4968 583.947 27.4968 580.223 Q27.4968 574.08 31.3163 570.929 Q35.1038 567.778 42.4881 567.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip852)" d="M448.104 597.618 L448.104 735.212 L569.717 735.212 L569.717 597.618 L448.104 597.618 L448.104 597.618  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,597.618 448.104,735.212 569.717,735.212 569.717,597.618 448.104,597.618 "/>
-<path clip-path="url(#clip852)" d="M600.12 735.212 L600.12 735.212 L721.733 735.212 L721.733 735.212 L600.12 735.212 L600.12 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,735.212 600.12,735.212 721.733,735.212 600.12,735.212 "/>
-<path clip-path="url(#clip852)" d="M752.137 528.822 L752.137 597.618 L873.75 597.618 L873.75 528.822 L752.137 528.822 L752.137 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,528.822 752.137,597.618 873.75,597.618 873.75,528.822 752.137,528.822 "/>
-<path clip-path="url(#clip852)" d="M904.153 528.822 L904.153 528.822 L1025.77 528.822 L1025.77 528.822 L904.153 528.822 L904.153 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,528.822 904.153,528.822 1025.77,528.822 904.153,528.822 "/>
-<path clip-path="url(#clip852)" d="M1056.17 460.025 L1056.17 689.347 L1177.78 689.347 L1177.78 460.025 L1056.17 460.025 L1056.17 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,460.025 1056.17,689.347 1177.78,689.347 1177.78,460.025 1056.17,460.025 "/>
-<path clip-path="url(#clip852)" d="M1208.19 528.822 L1208.19 528.822 L1329.8 528.822 L1329.8 528.822 L1208.19 528.822 L1208.19 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,528.822 1208.19,528.822 1329.8,528.822 1208.19,528.822 "/>
-<path clip-path="url(#clip852)" d="M1360.2 391.228 L1360.2 437.093 L1481.82 437.093 L1481.82 391.228 L1360.2 391.228 L1360.2 391.228  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,391.228 1360.2,437.093 1481.82,437.093 1481.82,391.228 1360.2,391.228 "/>
-<path clip-path="url(#clip852)" d="M1512.22 253.634 L1512.22 253.634 L1633.83 253.634 L1633.83 253.634 L1512.22 253.634 L1512.22 253.634  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,253.634 1512.22,253.634 1633.83,253.634 1512.22,253.634 "/>
-<path clip-path="url(#clip852)" d="M1664.24 47.2441 L1664.24 47.2441 L1785.85 47.2441 L1785.85 47.2441 L1664.24 47.2441 L1664.24 47.2441  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,47.2441 1664.24,47.2441 1785.85,47.2441 1664.24,47.2441 "/>
-<path clip-path="url(#clip852)" d="M1816.25 116.041 L1816.25 116.041 L1937.87 116.041 L1937.87 116.041 L1816.25 116.041 L1816.25 116.041  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,116.041 1816.25,116.041 1937.87,116.041 1816.25,116.041 "/>
-<path clip-path="url(#clip852)" d="M1968.27 460.025 L1968.27 505.889 L2089.88 505.889 L2089.88 460.025 L1968.27 460.025 L1968.27 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,460.025 1968.27,505.889 2089.88,505.889 2089.88,460.025 1968.27,460.025 "/>
-<path clip-path="url(#clip852)" d="M2120.29 735.212 L2120.29 735.212 L2241.9 735.212 L2241.9 735.212 L2120.29 735.212 L2120.29 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,735.212 2120.29,735.212 2241.9,735.212 2120.29,735.212 "/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="508.91" cy="597.618" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="660.927" cy="735.212" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="812.943" cy="528.822" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="964.96" cy="528.822" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1116.98" cy="460.025" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1268.99" cy="528.822" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1421.01" cy="391.228" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1573.03" cy="253.634" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1725.04" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1877.06" cy="116.041" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2029.08" cy="460.025" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2181.09" cy="735.212" r="2"/>
-<path clip-path="url(#clip852)" d="M448.104 735.212 L448.104 1193.86 L569.717 1193.86 L569.717 735.212 L448.104 735.212 L448.104 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,735.212 448.104,1193.86 569.717,1193.86 569.717,735.212 448.104,735.212 "/>
-<path clip-path="url(#clip852)" d="M600.12 735.212 L600.12 1170.92 L721.733 1170.92 L721.733 735.212 L600.12 735.212 L600.12 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,735.212 600.12,1170.92 721.733,1170.92 721.733,735.212 600.12,735.212 "/>
-<path clip-path="url(#clip852)" d="M752.137 597.618 L752.137 1147.99 L873.75 1147.99 L873.75 597.618 L752.137 597.618 L752.137 597.618  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,597.618 752.137,1147.99 873.75,1147.99 873.75,597.618 752.137,597.618 "/>
-<path clip-path="url(#clip852)" d="M904.153 528.822 L904.153 964.535 L1025.77 964.535 L1025.77 528.822 L904.153 528.822 L904.153 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,528.822 904.153,964.535 1025.77,964.535 1025.77,528.822 904.153,528.822 "/>
-<path clip-path="url(#clip852)" d="M1056.17 689.347 L1056.17 1423.18 L1177.78 1423.18 L1177.78 689.347 L1056.17 689.347 L1056.17 689.347  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,689.347 1056.17,1423.18 1177.78,1423.18 1177.78,689.347 1056.17,689.347 "/>
-<path clip-path="url(#clip852)" d="M1208.19 528.822 L1208.19 1102.13 L1329.8 1102.13 L1329.8 528.822 L1208.19 528.822 L1208.19 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,528.822 1208.19,1102.13 1329.8,1102.13 1329.8,528.822 1208.19,528.822 "/>
-<path clip-path="url(#clip852)" d="M1360.2 437.093 L1360.2 1147.99 L1481.82 1147.99 L1481.82 437.093 L1360.2 437.093 L1360.2 437.093  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,437.093 1360.2,1147.99 1481.82,1147.99 1481.82,437.093 1360.2,437.093 "/>
-<path clip-path="url(#clip852)" d="M1512.22 253.634 L1512.22 964.535 L1633.83 964.535 L1633.83 253.634 L1512.22 253.634 L1512.22 253.634  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,253.634 1512.22,964.535 1633.83,964.535 1633.83,253.634 1512.22,253.634 "/>
-<path clip-path="url(#clip852)" d="M1664.24 47.2441 L1664.24 964.535 L1785.85 964.535 L1785.85 47.2441 L1664.24 47.2441 L1664.24 47.2441  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,47.2441 1664.24,964.535 1785.85,964.535 1785.85,47.2441 1664.24,47.2441 "/>
-<path clip-path="url(#clip852)" d="M1816.25 116.041 L1816.25 1019.57 L1937.87 1019.57 L1937.87 116.041 L1816.25 116.041 L1816.25 116.041  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,116.041 1816.25,1019.57 1937.87,1019.57 1937.87,116.041 1816.25,116.041 "/>
-<path clip-path="url(#clip852)" d="M1968.27 505.889 L1968.27 872.806 L2089.88 872.806 L2089.88 505.889 L1968.27 505.889 L1968.27 505.889  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,505.889 1968.27,872.806 2089.88,872.806 2089.88,505.889 1968.27,505.889 "/>
-<path clip-path="url(#clip852)" d="M2120.29 735.212 L2120.29 1423.18 L2241.9 1423.18 L2241.9 735.212 L2120.29 735.212 L2120.29 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,735.212 2120.29,1423.18 2241.9,1423.18 2241.9,735.212 2120.29,735.212 "/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="508.91" cy="735.212" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="660.927" cy="735.212" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="812.943" cy="597.618" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="964.96" cy="528.822" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1116.98" cy="689.347" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1268.99" cy="528.822" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1421.01" cy="437.093" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1573.03" cy="253.634" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1725.04" cy="47.2441" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1877.06" cy="116.041" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2029.08" cy="505.889" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2181.09" cy="735.212" r="2"/>
-<path clip-path="url(#clip852)" d="M448.104 1193.86 L448.104 1423.18 L569.717 1423.18 L569.717 1193.86 L448.104 1193.86 L448.104 1193.86  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,1193.86 448.104,1423.18 569.717,1423.18 569.717,1193.86 448.104,1193.86 "/>
-<path clip-path="url(#clip852)" d="M600.12 1170.92 L600.12 1423.18 L721.733 1423.18 L721.733 1170.92 L600.12 1170.92 L600.12 1170.92  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,1170.92 600.12,1423.18 721.733,1423.18 721.733,1170.92 600.12,1170.92 "/>
-<path clip-path="url(#clip852)" d="M752.137 1147.99 L752.137 1423.18 L873.75 1423.18 L873.75 1147.99 L752.137 1147.99 L752.137 1147.99  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,1147.99 752.137,1423.18 873.75,1423.18 873.75,1147.99 752.137,1147.99 "/>
-<path clip-path="url(#clip852)" d="M904.153 964.535 L904.153 1423.18 L1025.77 1423.18 L1025.77 964.535 L904.153 964.535 L904.153 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,964.535 904.153,1423.18 1025.77,1423.18 1025.77,964.535 904.153,964.535 "/>
-<path clip-path="url(#clip852)" d="M1056.17 1423.18 L1056.17 1423.18 L1177.78 1423.18 L1177.78 1423.18 L1056.17 1423.18 L1056.17 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,1423.18 1056.17,1423.18 1177.78,1423.18 1056.17,1423.18 "/>
-<path clip-path="url(#clip852)" d="M1208.19 1102.13 L1208.19 1423.18 L1329.8 1423.18 L1329.8 1102.13 L1208.19 1102.13 L1208.19 1102.13  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,1102.13 1208.19,1423.18 1329.8,1423.18 1329.8,1102.13 1208.19,1102.13 "/>
-<path clip-path="url(#clip852)" d="M1360.2 1147.99 L1360.2 1423.18 L1481.82 1423.18 L1481.82 1147.99 L1360.2 1147.99 L1360.2 1147.99  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,1147.99 1360.2,1423.18 1481.82,1423.18 1481.82,1147.99 1360.2,1147.99 "/>
-<path clip-path="url(#clip852)" d="M1512.22 964.535 L1512.22 1423.18 L1633.83 1423.18 L1633.83 964.535 L1512.22 964.535 L1512.22 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,964.535 1512.22,1423.18 1633.83,1423.18 1633.83,964.535 1512.22,964.535 "/>
-<path clip-path="url(#clip852)" d="M1664.24 964.535 L1664.24 1423.18 L1785.85 1423.18 L1785.85 964.535 L1664.24 964.535 L1664.24 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,964.535 1664.24,1423.18 1785.85,1423.18 1785.85,964.535 1664.24,964.535 "/>
-<path clip-path="url(#clip852)" d="M1816.25 1019.57 L1816.25 1423.18 L1937.87 1423.18 L1937.87 1019.57 L1816.25 1019.57 L1816.25 1019.57  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,1019.57 1816.25,1423.18 1937.87,1423.18 1937.87,1019.57 1816.25,1019.57 "/>
-<path clip-path="url(#clip852)" d="M1968.27 872.806 L1968.27 1423.18 L2089.88 1423.18 L2089.88 872.806 L1968.27 872.806 L1968.27 872.806  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,872.806 1968.27,1423.18 2089.88,1423.18 2089.88,872.806 1968.27,872.806 "/>
-<path clip-path="url(#clip852)" d="M2120.29 1423.18 L2120.29 1423.18 L2241.9 1423.18 L2241.9 1423.18 L2120.29 1423.18 L2120.29 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,1423.18 2120.29,1423.18 2241.9,1423.18 2120.29,1423.18 "/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="508.91" cy="1193.86" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="660.927" cy="1170.92" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="812.943" cy="1147.99" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="964.96" cy="964.535" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1116.98" cy="1423.18" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1268.99" cy="1102.13" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1421.01" cy="1147.99" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1573.03" cy="964.535" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1725.04" cy="964.535" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1877.06" cy="1019.57" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2029.08" cy="872.806" r="2"/>
-<circle clip-path="url(#clip852)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2181.09" cy="1423.18" r="2"/>
-<path clip-path="url(#clip850)" d="M404.431 300.469 L961.373 300.469 L961.373 93.1086 L404.431 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="404.431,300.469 961.373,300.469 961.373,93.1086 404.431,93.1086 404.431,300.469 "/>
-<path clip-path="url(#clip850)" d="M426.825 165.685 L561.192 165.685 L561.192 124.213 L426.825 124.213 L426.825 165.685  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,165.685 561.192,165.685 561.192,124.213 426.825,124.213 426.825,165.685 "/>
-<path clip-path="url(#clip850)" d="M583.587 151.997 L583.587 136.303 L587.846 136.303 L587.846 151.835 Q587.846 155.516 589.281 157.368 Q590.717 159.196 593.587 159.196 Q597.036 159.196 599.027 156.997 Q601.041 154.798 601.041 151.002 L601.041 136.303 L605.3 136.303 L605.3 162.229 L601.041 162.229 L601.041 158.247 Q599.49 160.608 597.429 161.766 Q595.392 162.9 592.684 162.9 Q588.217 162.9 585.902 160.122 Q583.587 157.344 583.587 151.997 M594.305 135.678 L594.305 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M635.624 146.581 L635.624 162.229 L631.365 162.229 L631.365 146.719 Q631.365 143.039 629.929 141.21 Q628.494 139.382 625.624 139.382 Q622.175 139.382 620.184 141.581 Q618.193 143.78 618.193 147.576 L618.193 162.229 L613.911 162.229 L613.911 136.303 L618.193 136.303 L618.193 140.331 Q619.721 137.993 621.781 136.835 Q623.865 135.678 626.573 135.678 Q631.04 135.678 633.332 138.456 Q635.624 141.21 635.624 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M664.304 141.28 Q665.901 138.409 668.124 137.044 Q670.346 135.678 673.355 135.678 Q677.406 135.678 679.605 138.525 Q681.804 141.349 681.804 146.581 L681.804 162.229 L677.522 162.229 L677.522 146.719 Q677.522 142.993 676.202 141.187 Q674.883 139.382 672.174 139.382 Q668.864 139.382 666.943 141.581 Q665.022 143.78 665.022 147.576 L665.022 162.229 L660.739 162.229 L660.739 146.719 Q660.739 142.969 659.42 141.187 Q658.1 139.382 655.346 139.382 Q652.082 139.382 650.161 141.604 Q648.239 143.803 648.239 147.576 L648.239 162.229 L643.957 162.229 L643.957 136.303 L648.239 136.303 L648.239 140.331 Q649.698 137.946 651.735 136.812 Q653.772 135.678 656.573 135.678 Q659.397 135.678 661.364 137.113 Q663.355 138.548 664.304 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M712.475 148.201 L712.475 150.284 L692.892 150.284 Q693.17 154.682 695.531 156.997 Q697.915 159.289 702.151 159.289 Q704.605 159.289 706.897 158.687 Q709.211 158.085 711.48 156.881 L711.48 160.909 Q709.188 161.881 706.781 162.391 Q704.373 162.9 701.897 162.9 Q695.693 162.9 692.059 159.289 Q688.448 155.678 688.448 149.52 Q688.448 143.155 691.873 139.428 Q695.322 135.678 701.156 135.678 Q706.387 135.678 709.42 139.057 Q712.475 142.414 712.475 148.201 M708.216 146.951 Q708.17 143.456 706.248 141.372 Q704.35 139.289 701.202 139.289 Q697.637 139.289 695.485 141.303 Q693.355 143.317 693.031 146.974 L708.216 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M723.679 128.942 L723.679 136.303 L732.452 136.303 L732.452 139.613 L723.679 139.613 L723.679 153.687 Q723.679 156.858 724.535 157.761 Q725.415 158.664 728.077 158.664 L732.452 158.664 L732.452 162.229 L728.077 162.229 Q723.146 162.229 721.271 160.4 Q719.396 158.548 719.396 153.687 L719.396 139.613 L716.271 139.613 L716.271 136.303 L719.396 136.303 L719.396 128.942 L723.679 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M770.183 140.238 L770.183 126.21 L774.442 126.21 L774.442 162.229 L770.183 162.229 L770.183 158.34 Q768.841 160.655 766.78 161.789 Q764.743 162.9 761.873 162.9 Q757.174 162.9 754.211 159.15 Q751.271 155.4 751.271 149.289 Q751.271 143.178 754.211 139.428 Q757.174 135.678 761.873 135.678 Q764.743 135.678 766.78 136.812 Q768.841 137.923 770.183 140.238 M755.669 149.289 Q755.669 153.988 757.591 156.673 Q759.535 159.335 762.915 159.335 Q766.294 159.335 768.239 156.673 Q770.183 153.988 770.183 149.289 Q770.183 144.59 768.239 141.928 Q766.294 139.243 762.915 139.243 Q759.535 139.243 757.591 141.928 Q755.669 144.59 755.669 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M805.391 148.201 L805.391 150.284 L785.808 150.284 Q786.086 154.682 788.447 156.997 Q790.831 159.289 795.067 159.289 Q797.521 159.289 799.813 158.687 Q802.127 158.085 804.396 156.881 L804.396 160.909 Q802.104 161.881 799.697 162.391 Q797.29 162.9 794.813 162.9 Q788.609 162.9 784.975 159.289 Q781.364 155.678 781.364 149.52 Q781.364 143.155 784.79 139.428 Q788.239 135.678 794.072 135.678 Q799.303 135.678 802.336 139.057 Q805.391 142.414 805.391 148.201 M801.132 146.951 Q801.086 143.456 799.165 141.372 Q797.266 139.289 794.118 139.289 Q790.553 139.289 788.401 141.303 Q786.271 143.317 785.947 146.974 L801.132 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M832.567 141.28 Q834.164 138.409 836.387 137.044 Q838.609 135.678 841.618 135.678 Q845.669 135.678 847.868 138.525 Q850.067 141.349 850.067 146.581 L850.067 162.229 L845.785 162.229 L845.785 146.719 Q845.785 142.993 844.465 141.187 Q843.146 139.382 840.437 139.382 Q837.127 139.382 835.206 141.581 Q833.285 143.78 833.285 147.576 L833.285 162.229 L829.002 162.229 L829.002 146.719 Q829.002 142.969 827.683 141.187 Q826.363 139.382 823.609 139.382 Q820.345 139.382 818.424 141.604 Q816.502 143.803 816.502 147.576 L816.502 162.229 L812.22 162.229 L812.22 136.303 L816.502 136.303 L816.502 140.331 Q817.961 137.946 819.998 136.812 Q822.035 135.678 824.836 135.678 Q827.66 135.678 829.627 137.113 Q831.618 138.548 832.567 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M870.345 149.196 Q865.183 149.196 863.192 150.377 Q861.201 151.557 861.201 154.405 Q861.201 156.673 862.683 158.016 Q864.187 159.335 866.757 159.335 Q870.298 159.335 872.428 156.835 Q874.581 154.312 874.581 150.145 L874.581 149.196 L870.345 149.196 M878.84 147.437 L878.84 162.229 L874.581 162.229 L874.581 158.293 Q873.122 160.655 870.947 161.789 Q868.771 162.9 865.623 162.9 Q861.641 162.9 859.28 160.678 Q856.942 158.432 856.942 154.682 Q856.942 150.307 859.859 148.085 Q862.798 145.863 868.609 145.863 L874.581 145.863 L874.581 145.446 Q874.581 142.507 872.636 140.909 Q870.715 139.289 867.22 139.289 Q864.998 139.289 862.891 139.821 Q860.785 140.354 858.84 141.419 L858.84 137.483 Q861.178 136.581 863.377 136.141 Q865.576 135.678 867.66 135.678 Q873.285 135.678 876.062 138.594 Q878.84 141.511 878.84 147.437 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M909.164 146.581 L909.164 162.229 L904.905 162.229 L904.905 146.719 Q904.905 143.039 903.47 141.21 Q902.034 139.382 899.164 139.382 Q895.715 139.382 893.724 141.581 Q891.733 143.78 891.733 147.576 L891.733 162.229 L887.451 162.229 L887.451 136.303 L891.733 136.303 L891.733 140.331 Q893.261 137.993 895.321 136.835 Q897.405 135.678 900.113 135.678 Q904.581 135.678 906.872 138.456 Q909.164 141.21 909.164 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M934.719 140.238 L934.719 126.21 L938.979 126.21 L938.979 162.229 L934.719 162.229 L934.719 158.34 Q933.377 160.655 931.317 161.789 Q929.28 162.9 926.409 162.9 Q921.71 162.9 918.747 159.15 Q915.807 155.4 915.807 149.289 Q915.807 143.178 918.747 139.428 Q921.71 135.678 926.409 135.678 Q929.28 135.678 931.317 136.812 Q933.377 137.923 934.719 140.238 M920.206 149.289 Q920.206 153.988 922.127 156.673 Q924.071 159.335 927.451 159.335 Q930.83 159.335 932.775 156.673 Q934.719 153.988 934.719 149.289 Q934.719 144.59 932.775 141.928 Q930.83 139.243 927.451 139.243 Q924.071 139.243 922.127 141.928 Q920.206 144.59 920.206 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M426.825 217.525 L561.192 217.525 L561.192 176.053 L426.825 176.053 L426.825 217.525  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,217.525 561.192,217.525 561.192,176.053 426.825,176.053 426.825,217.525 "/>
-<path clip-path="url(#clip850)" d="M599.42 184.115 L593.078 201.314 L605.786 201.314 L599.42 184.115 M596.781 179.509 L602.082 179.509 L615.253 214.069 L610.392 214.069 L607.244 205.203 L591.666 205.203 L588.518 214.069 L583.587 214.069 L596.781 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M426.825 269.365 L561.192 269.365 L561.192 227.893 L426.825 227.893 L426.825 269.365  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,269.365 561.192,269.365 561.192,227.893 426.825,227.893 426.825,269.365 "/>
-<path clip-path="url(#clip850)" d="M588.263 249.404 L588.263 262.066 L595.763 262.066 Q599.536 262.066 601.342 260.515 Q603.17 258.941 603.17 255.723 Q603.17 252.483 601.342 250.955 Q599.536 249.404 595.763 249.404 L588.263 249.404 M588.263 235.191 L588.263 245.608 L595.184 245.608 Q598.61 245.608 600.277 244.335 Q601.967 243.038 601.967 240.399 Q601.967 237.784 600.277 236.487 Q598.61 235.191 595.184 235.191 L588.263 235.191 M583.587 231.349 L595.531 231.349 Q600.879 231.349 603.772 233.571 Q606.666 235.793 606.666 239.89 Q606.666 243.062 605.184 244.937 Q603.703 246.811 600.832 247.274 Q604.281 248.015 606.179 250.376 Q608.101 252.714 608.101 256.233 Q608.101 260.862 604.953 263.385 Q601.804 265.909 595.994 265.909 L583.587 265.909 L583.587 231.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="356.894,1423.18 356.894,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="660.927,1423.18 660.927,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="964.96,1423.18 964.96,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1268.99,1423.18 1268.99,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1573.03,1423.18 1573.03,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1877.06,1423.18 1877.06,47.2441 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2181.09,1423.18 2181.09,47.2441 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="356.894,1423.18 356.894,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="660.927,1423.18 660.927,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="964.96,1423.18 964.96,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1268.99,1423.18 1268.99,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1573.03,1423.18 1573.03,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.06,1423.18 1877.06,1404.28 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2181.09,1423.18 2181.09,1404.28 "/>
+<path clip-path="url(#clip670)" d="M356.894 1454.1 Q353.283 1454.1 351.454 1457.66 Q349.648 1461.2 349.648 1468.33 Q349.648 1475.44 351.454 1479.01 Q353.283 1482.55 356.894 1482.55 Q360.528 1482.55 362.333 1479.01 Q364.162 1475.44 364.162 1468.33 Q364.162 1461.2 362.333 1457.66 Q360.528 1454.1 356.894 1454.1 M356.894 1450.39 Q362.704 1450.39 365.759 1455 Q368.838 1459.58 368.838 1468.33 Q368.838 1477.06 365.759 1481.67 Q362.704 1486.25 356.894 1486.25 Q351.083 1486.25 348.005 1481.67 Q344.949 1477.06 344.949 1468.33 Q344.949 1459.58 348.005 1455 Q351.083 1450.39 356.894 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M655.58 1481.64 L671.899 1481.64 L671.899 1485.58 L649.955 1485.58 L649.955 1481.64 Q652.617 1478.89 657.2 1474.26 Q661.806 1469.61 662.987 1468.27 Q665.232 1465.74 666.112 1464.01 Q667.015 1462.25 667.015 1460.56 Q667.015 1457.8 665.07 1456.07 Q663.149 1454.33 660.047 1454.33 Q657.848 1454.33 655.394 1455.09 Q652.964 1455.86 650.186 1457.41 L650.186 1452.69 Q653.01 1451.55 655.464 1450.97 Q657.918 1450.39 659.955 1450.39 Q665.325 1450.39 668.519 1453.08 Q671.714 1455.77 671.714 1460.26 Q671.714 1462.39 670.904 1464.31 Q670.117 1466.2 668.01 1468.8 Q667.431 1469.47 664.33 1472.69 Q661.228 1475.88 655.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M967.969 1455.09 L956.164 1473.54 L967.969 1473.54 L967.969 1455.09 M966.742 1451.02 L972.622 1451.02 L972.622 1473.54 L977.553 1473.54 L977.553 1477.43 L972.622 1477.43 L972.622 1485.58 L967.969 1485.58 L967.969 1477.43 L952.367 1477.43 L952.367 1472.92 L966.742 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1269.4 1466.44 Q1266.25 1466.44 1264.4 1468.59 Q1262.57 1470.74 1262.57 1474.49 Q1262.57 1478.22 1264.4 1480.39 Q1266.25 1482.55 1269.4 1482.55 Q1272.55 1482.55 1274.38 1480.39 Q1276.23 1478.22 1276.23 1474.49 Q1276.23 1470.74 1274.38 1468.59 Q1272.55 1466.44 1269.4 1466.44 M1278.68 1451.78 L1278.68 1456.04 Q1276.92 1455.21 1275.12 1454.77 Q1273.33 1454.33 1271.57 1454.33 Q1266.94 1454.33 1264.49 1457.45 Q1262.06 1460.58 1261.71 1466.9 Q1263.08 1464.89 1265.14 1463.82 Q1267.2 1462.73 1269.68 1462.73 Q1274.88 1462.73 1277.89 1465.9 Q1280.93 1469.05 1280.93 1474.49 Q1280.93 1479.82 1277.78 1483.03 Q1274.63 1486.25 1269.4 1486.25 Q1263.4 1486.25 1260.23 1481.67 Q1257.06 1477.06 1257.06 1468.33 Q1257.06 1460.14 1260.95 1455.28 Q1264.84 1450.39 1271.39 1450.39 Q1273.15 1450.39 1274.93 1450.74 Q1276.74 1451.09 1278.68 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1573.03 1469.17 Q1569.69 1469.17 1567.77 1470.95 Q1565.87 1472.73 1565.87 1475.86 Q1565.87 1478.98 1567.77 1480.77 Q1569.69 1482.55 1573.03 1482.55 Q1576.36 1482.55 1578.28 1480.77 Q1580.2 1478.96 1580.2 1475.86 Q1580.2 1472.73 1578.28 1470.95 Q1576.38 1469.17 1573.03 1469.17 M1568.35 1467.18 Q1565.34 1466.44 1563.65 1464.38 Q1561.98 1462.32 1561.98 1459.35 Q1561.98 1455.21 1564.92 1452.8 Q1567.89 1450.39 1573.03 1450.39 Q1578.19 1450.39 1581.13 1452.8 Q1584.07 1455.21 1584.07 1459.35 Q1584.07 1462.32 1582.38 1464.38 Q1580.71 1466.44 1577.73 1467.18 Q1581.1 1467.96 1582.98 1470.26 Q1584.88 1472.55 1584.88 1475.86 Q1584.88 1480.88 1581.8 1483.57 Q1578.74 1486.25 1573.03 1486.25 Q1567.31 1486.25 1564.23 1483.57 Q1561.17 1480.88 1561.17 1475.86 Q1561.17 1472.55 1563.07 1470.26 Q1564.97 1467.96 1568.35 1467.18 M1566.64 1459.79 Q1566.64 1462.48 1568.3 1463.98 Q1569.99 1465.49 1573.03 1465.49 Q1576.04 1465.49 1577.73 1463.98 Q1579.44 1462.48 1579.44 1459.79 Q1579.44 1457.11 1577.73 1455.6 Q1576.04 1454.1 1573.03 1454.1 Q1569.99 1454.1 1568.3 1455.6 Q1566.64 1457.11 1566.64 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1851.75 1481.64 L1859.39 1481.64 L1859.39 1455.28 L1851.08 1456.95 L1851.08 1452.69 L1859.34 1451.02 L1864.02 1451.02 L1864.02 1481.64 L1871.65 1481.64 L1871.65 1485.58 L1851.75 1485.58 L1851.75 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1891.1 1454.1 Q1887.49 1454.1 1885.66 1457.66 Q1883.85 1461.2 1883.85 1468.33 Q1883.85 1475.44 1885.66 1479.01 Q1887.49 1482.55 1891.1 1482.55 Q1894.73 1482.55 1896.54 1479.01 Q1898.37 1475.44 1898.37 1468.33 Q1898.37 1461.2 1896.54 1457.66 Q1894.73 1454.1 1891.1 1454.1 M1891.1 1450.39 Q1896.91 1450.39 1899.96 1455 Q1903.04 1459.58 1903.04 1468.33 Q1903.04 1477.06 1899.96 1481.67 Q1896.91 1486.25 1891.1 1486.25 Q1885.29 1486.25 1882.21 1481.67 Q1879.15 1477.06 1879.15 1468.33 Q1879.15 1459.58 1882.21 1455 Q1885.29 1450.39 1891.1 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M2156.58 1481.64 L2164.22 1481.64 L2164.22 1455.28 L2155.91 1456.95 L2155.91 1452.69 L2164.17 1451.02 L2168.85 1451.02 L2168.85 1481.64 L2176.49 1481.64 L2176.49 1485.58 L2156.58 1485.58 L2156.58 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M2189.96 1481.64 L2206.28 1481.64 L2206.28 1485.58 L2184.33 1485.58 L2184.33 1481.64 Q2187 1478.89 2191.58 1474.26 Q2196.19 1469.61 2197.37 1468.27 Q2199.61 1465.74 2200.49 1464.01 Q2201.39 1462.25 2201.39 1460.56 Q2201.39 1457.8 2199.45 1456.07 Q2197.53 1454.33 2194.43 1454.33 Q2192.23 1454.33 2189.77 1455.09 Q2187.34 1455.86 2184.56 1457.41 L2184.56 1452.69 Q2187.39 1451.55 2189.84 1450.97 Q2192.3 1450.39 2194.33 1450.39 Q2199.7 1450.39 2202.9 1453.08 Q2206.09 1455.77 2206.09 1460.26 Q2206.09 1462.39 2205.28 1464.31 Q2204.5 1466.2 2202.39 1468.8 Q2201.81 1469.47 2198.71 1472.69 Q2195.61 1475.88 2189.96 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1248.82 1520.52 L1258.4 1520.52 L1270.52 1552.86 L1282.71 1520.52 L1292.29 1520.52 L1292.29 1568.04 L1286.02 1568.04 L1286.02 1526.32 L1273.77 1558.91 L1267.31 1558.91 L1255.05 1526.32 L1255.05 1568.04 L1248.82 1568.04 L1248.82 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1318.62 1536.5 Q1313.9 1536.5 1311.17 1540.19 Q1308.43 1543.85 1308.43 1550.25 Q1308.43 1556.65 1311.14 1560.34 Q1313.87 1564 1318.62 1564 Q1323.29 1564 1326.03 1560.31 Q1328.77 1556.62 1328.77 1550.25 Q1328.77 1543.92 1326.03 1540.23 Q1323.29 1536.5 1318.62 1536.5 M1318.62 1531.54 Q1326.25 1531.54 1330.61 1536.5 Q1334.98 1541.47 1334.98 1550.25 Q1334.98 1559 1330.61 1564 Q1326.25 1568.97 1318.62 1568.97 Q1310.94 1568.97 1306.58 1564 Q1302.26 1559 1302.26 1550.25 Q1302.26 1541.47 1306.58 1536.5 Q1310.94 1531.54 1318.62 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1374.32 1546.53 L1374.32 1568.04 L1368.46 1568.04 L1368.46 1546.72 Q1368.46 1541.66 1366.49 1539.14 Q1364.51 1536.63 1360.57 1536.63 Q1355.82 1536.63 1353.09 1539.65 Q1350.35 1542.68 1350.35 1547.9 L1350.35 1568.04 L1344.46 1568.04 L1344.46 1532.4 L1350.35 1532.4 L1350.35 1537.93 Q1352.45 1534.72 1355.28 1533.13 Q1358.15 1531.54 1361.87 1531.54 Q1368.01 1531.54 1371.16 1535.36 Q1374.32 1539.14 1374.32 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1391.79 1522.27 L1391.79 1532.4 L1403.85 1532.4 L1403.85 1536.95 L1391.79 1536.95 L1391.79 1556.3 Q1391.79 1560.66 1392.97 1561.9 Q1394.18 1563.14 1397.84 1563.14 L1403.85 1563.14 L1403.85 1568.04 L1397.84 1568.04 Q1391.06 1568.04 1388.48 1565.53 Q1385.9 1562.98 1385.9 1556.3 L1385.9 1536.95 L1381.6 1536.95 L1381.6 1532.4 L1385.9 1532.4 L1385.9 1522.27 L1391.79 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M1441.19 1546.53 L1441.19 1568.04 L1435.33 1568.04 L1435.33 1546.72 Q1435.33 1541.66 1433.36 1539.14 Q1431.38 1536.63 1427.44 1536.63 Q1422.69 1536.63 1419.96 1539.65 Q1417.22 1542.68 1417.22 1547.9 L1417.22 1568.04 L1411.33 1568.04 L1411.33 1518.52 L1417.22 1518.52 L1417.22 1537.93 Q1419.32 1534.72 1422.15 1533.13 Q1425.02 1531.54 1428.74 1531.54 Q1434.89 1531.54 1438.04 1535.36 Q1441.19 1539.14 1441.19 1546.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,964.535 2352.76,964.535 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,505.889 2352.76,505.889 "/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="337.247,47.2441 2352.76,47.2441 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 337.247,47.2441 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,1423.18 356.145,1423.18 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,964.535 356.145,964.535 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,505.889 356.145,505.889 "/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="337.247,47.2441 356.145,47.2441 "/>
+<path clip-path="url(#clip670)" d="M289.303 1408.98 Q285.692 1408.98 283.863 1412.54 Q282.057 1416.08 282.057 1423.21 Q282.057 1430.32 283.863 1433.89 Q285.692 1437.43 289.303 1437.43 Q292.937 1437.43 294.742 1433.89 Q296.571 1430.32 296.571 1423.21 Q296.571 1416.08 294.742 1412.54 Q292.937 1408.98 289.303 1408.98 M289.303 1405.27 Q295.113 1405.27 298.168 1409.88 Q301.247 1414.46 301.247 1423.21 Q301.247 1431.94 298.168 1436.55 Q295.113 1441.13 289.303 1441.13 Q283.492 1441.13 280.414 1436.55 Q277.358 1431.94 277.358 1423.21 Q277.358 1414.46 280.414 1409.88 Q283.492 1405.27 289.303 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M116.668 984.327 L124.306 984.327 L124.306 957.961 L115.996 959.628 L115.996 955.369 L124.26 953.702 L128.936 953.702 L128.936 984.327 L136.575 984.327 L136.575 988.262 L116.668 988.262 L116.668 984.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M146.019 982.383 L150.903 982.383 L150.903 988.262 L146.019 988.262 L146.019 982.383 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M171.089 956.781 Q167.477 956.781 165.649 960.346 Q163.843 963.887 163.843 971.017 Q163.843 978.123 165.649 981.688 Q167.477 985.23 171.089 985.23 Q174.723 985.23 176.528 981.688 Q178.357 978.123 178.357 971.017 Q178.357 963.887 176.528 960.346 Q174.723 956.781 171.089 956.781 M171.089 953.077 Q176.899 953.077 179.954 957.684 Q183.033 962.267 183.033 971.017 Q183.033 979.744 179.954 984.35 Q176.899 988.933 171.089 988.933 Q165.278 988.933 162.2 984.35 Q159.144 979.744 159.144 971.017 Q159.144 962.267 162.2 957.684 Q165.278 953.077 171.089 953.077 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M219.422 962.799 L208.843 973.424 L219.422 984.003 L216.667 986.804 L206.042 976.179 L195.417 986.804 L192.686 984.003 L203.241 973.424 L192.686 962.799 L195.417 959.998 L206.042 970.623 L216.667 959.998 L219.422 962.799 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M231.783 984.327 L239.422 984.327 L239.422 957.961 L231.111 959.628 L231.111 955.369 L239.375 953.702 L244.051 953.702 L244.051 984.327 L251.69 984.327 L251.69 988.262 L231.783 988.262 L231.783 984.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M271.134 956.781 Q267.523 956.781 265.695 960.346 Q263.889 963.887 263.889 971.017 Q263.889 978.123 265.695 981.688 Q267.523 985.23 271.134 985.23 Q274.769 985.23 276.574 981.688 Q278.403 978.123 278.403 971.017 Q278.403 963.887 276.574 960.346 Q274.769 956.781 271.134 956.781 M271.134 953.077 Q276.944 953.077 280 957.684 Q283.079 962.267 283.079 971.017 Q283.079 979.744 280 984.35 Q276.944 988.933 271.134 988.933 Q265.324 988.933 262.245 984.35 Q259.19 979.744 259.19 971.017 Q259.19 962.267 262.245 957.684 Q265.324 953.077 271.134 953.077 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M284.264 932.772 L299.178 932.772 L299.178 935.969 L287.743 935.969 L287.743 942.853 Q288.571 942.571 289.398 942.439 Q290.226 942.289 291.053 942.289 Q295.755 942.289 298.501 944.865 Q301.247 947.442 301.247 951.843 Q301.247 956.376 298.426 958.896 Q295.605 961.397 290.47 961.397 Q288.702 961.397 286.859 961.096 Q285.035 960.795 283.079 960.194 L283.079 956.376 Q284.771 957.297 286.577 957.749 Q288.382 958.2 290.395 958.2 Q293.649 958.2 295.548 956.488 Q297.448 954.777 297.448 951.843 Q297.448 948.909 295.548 947.197 Q293.649 945.486 290.395 945.486 Q288.871 945.486 287.348 945.824 Q285.843 946.163 284.264 946.878 L284.264 932.772 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M119.885 525.682 L136.204 525.682 L136.204 529.617 L114.26 529.617 L114.26 525.682 Q116.922 522.927 121.505 518.298 Q126.112 513.645 127.292 512.302 Q129.538 509.779 130.417 508.043 Q131.32 506.284 131.32 504.594 Q131.32 501.839 129.376 500.103 Q127.455 498.367 124.353 498.367 Q122.154 498.367 119.7 499.131 Q117.269 499.895 114.492 501.446 L114.492 496.724 Q117.316 495.589 119.769 495.011 Q122.223 494.432 124.26 494.432 Q129.63 494.432 132.825 497.117 Q136.019 499.802 136.019 504.293 Q136.019 506.423 135.209 508.344 Q134.422 510.242 132.316 512.835 Q131.737 513.506 128.635 516.723 Q125.533 519.918 119.885 525.682 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M146.019 523.737 L150.903 523.737 L150.903 529.617 L146.019 529.617 L146.019 523.737 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M171.089 498.136 Q167.477 498.136 165.649 501.7 Q163.843 505.242 163.843 512.372 Q163.843 519.478 165.649 523.043 Q167.477 526.585 171.089 526.585 Q174.723 526.585 176.528 523.043 Q178.357 519.478 178.357 512.372 Q178.357 505.242 176.528 501.7 Q174.723 498.136 171.089 498.136 M171.089 494.432 Q176.899 494.432 179.954 499.038 Q183.033 503.622 183.033 512.372 Q183.033 521.098 179.954 525.705 Q176.899 530.288 171.089 530.288 Q165.278 530.288 162.2 525.705 Q159.144 521.098 159.144 512.372 Q159.144 503.622 162.2 499.038 Q165.278 494.432 171.089 494.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M219.422 504.154 L208.843 514.779 L219.422 525.358 L216.667 528.159 L206.042 517.534 L195.417 528.159 L192.686 525.358 L203.241 514.779 L192.686 504.154 L195.417 501.353 L206.042 511.978 L216.667 501.353 L219.422 504.154 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M231.783 525.682 L239.422 525.682 L239.422 499.316 L231.111 500.983 L231.111 496.724 L239.375 495.057 L244.051 495.057 L244.051 525.682 L251.69 525.682 L251.69 529.617 L231.783 529.617 L231.783 525.682 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M271.134 498.136 Q267.523 498.136 265.695 501.7 Q263.889 505.242 263.889 512.372 Q263.889 519.478 265.695 523.043 Q267.523 526.585 271.134 526.585 Q274.769 526.585 276.574 523.043 Q278.403 519.478 278.403 512.372 Q278.403 505.242 276.574 501.7 Q274.769 498.136 271.134 498.136 M271.134 494.432 Q276.944 494.432 280 499.038 Q283.079 503.622 283.079 512.372 Q283.079 521.098 280 525.705 Q276.944 530.288 271.134 530.288 Q265.324 530.288 262.245 525.705 Q259.19 521.098 259.19 512.372 Q259.19 503.622 262.245 499.038 Q265.324 494.432 271.134 494.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M284.264 474.127 L299.178 474.127 L299.178 477.324 L287.743 477.324 L287.743 484.208 Q288.571 483.925 289.398 483.794 Q290.226 483.643 291.053 483.643 Q295.755 483.643 298.501 486.22 Q301.247 488.797 301.247 493.198 Q301.247 497.73 298.426 500.251 Q295.605 502.752 290.47 502.752 Q288.702 502.752 286.859 502.451 Q285.035 502.15 283.079 501.548 L283.079 497.73 Q284.771 498.652 286.577 499.103 Q288.382 499.555 290.395 499.555 Q293.649 499.555 295.548 497.843 Q297.448 496.132 297.448 493.198 Q297.448 490.264 295.548 488.552 Q293.649 486.841 290.395 486.841 Q288.871 486.841 287.348 487.179 Q285.843 487.518 284.264 488.232 L284.264 474.127 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M130.024 52.3375 Q133.38 53.0551 135.255 55.3236 Q137.154 57.5921 137.154 60.9255 Q137.154 66.0412 133.635 68.8421 Q130.117 71.643 123.635 71.643 Q121.459 71.643 119.144 71.2032 Q116.853 70.7865 114.399 69.93 L114.399 65.4162 Q116.343 66.5504 118.658 67.1291 Q120.973 67.7078 123.496 67.7078 Q127.894 67.7078 130.186 65.9717 Q132.501 64.2356 132.501 60.9255 Q132.501 57.8699 130.348 56.157 Q128.218 54.4209 124.399 54.4209 L120.371 54.4209 L120.371 50.5783 L124.584 50.5783 Q128.033 50.5783 129.862 49.2126 Q131.691 47.8237 131.691 45.2311 Q131.691 42.5691 129.792 41.157 Q127.917 39.7219 124.399 39.7219 Q122.478 39.7219 120.279 40.1385 Q118.08 40.5552 115.441 41.4348 L115.441 37.2682 Q118.103 36.5274 120.418 36.1571 Q122.755 35.7867 124.816 35.7867 Q130.14 35.7867 133.242 38.2172 Q136.343 40.6246 136.343 44.745 Q136.343 47.6153 134.7 49.6061 Q133.056 51.5737 130.024 52.3375 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M146.019 65.0921 L150.903 65.0921 L150.903 70.9717 L146.019 70.9717 L146.019 65.0921 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M171.089 39.4904 Q167.477 39.4904 165.649 43.0552 Q163.843 46.5968 163.843 53.7264 Q163.843 60.8329 165.649 64.3977 Q167.477 67.9393 171.089 67.9393 Q174.723 67.9393 176.528 64.3977 Q178.357 60.8329 178.357 53.7264 Q178.357 46.5968 176.528 43.0552 Q174.723 39.4904 171.089 39.4904 M171.089 35.7867 Q176.899 35.7867 179.954 40.3932 Q183.033 44.9765 183.033 53.7264 Q183.033 62.4532 179.954 67.0597 Q176.899 71.643 171.089 71.643 Q165.278 71.643 162.2 67.0597 Q159.144 62.4532 159.144 53.7264 Q159.144 44.9765 162.2 40.3932 Q165.278 35.7867 171.089 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M219.422 45.5089 L208.843 56.1338 L219.422 66.7125 L216.667 69.5134 L206.042 58.8884 L195.417 69.5134 L192.686 66.7125 L203.241 56.1338 L192.686 45.5089 L195.417 42.708 L206.042 53.3329 L216.667 42.708 L219.422 45.5089 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M231.783 67.0365 L239.422 67.0365 L239.422 40.6709 L231.111 42.3376 L231.111 38.0784 L239.375 36.4117 L244.051 36.4117 L244.051 67.0365 L251.69 67.0365 L251.69 70.9717 L231.783 70.9717 L231.783 67.0365 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M271.134 39.4904 Q267.523 39.4904 265.695 43.0552 Q263.889 46.5968 263.889 53.7264 Q263.889 60.8329 265.695 64.3977 Q267.523 67.9393 271.134 67.9393 Q274.769 67.9393 276.574 64.3977 Q278.403 60.8329 278.403 53.7264 Q278.403 46.5968 276.574 43.0552 Q274.769 39.4904 271.134 39.4904 M271.134 35.7867 Q276.944 35.7867 280 40.3932 Q283.079 44.9765 283.079 53.7264 Q283.079 62.4532 280 67.0597 Q276.944 71.643 271.134 71.643 Q265.324 71.643 262.245 67.0597 Q259.19 62.4532 259.19 53.7264 Q259.19 44.9765 262.245 40.3932 Q265.324 35.7867 271.134 35.7867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M284.264 15.4813 L299.178 15.4813 L299.178 18.6786 L287.743 18.6786 L287.743 25.5623 Q288.571 25.2801 289.398 25.1485 Q290.226 24.998 291.053 24.998 Q295.755 24.998 298.501 27.5747 Q301.247 30.1514 301.247 34.5524 Q301.247 39.085 298.426 41.6053 Q295.605 44.1067 290.47 44.1067 Q288.702 44.1067 286.859 43.8058 Q285.035 43.5049 283.079 42.903 L283.079 39.085 Q284.771 40.0066 286.577 40.458 Q288.382 40.9094 290.395 40.9094 Q293.649 40.9094 295.548 39.1979 Q297.448 37.4864 297.448 34.5524 Q297.448 31.6184 295.548 29.9069 Q293.649 28.1953 290.395 28.1953 Q288.871 28.1953 287.348 28.5339 Q285.843 28.8724 284.264 29.5871 L284.264 15.4813 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M21.7677 896.217 L39.6235 896.217 L39.6235 888.132 Q39.6235 883.644 37.3 881.194 Q34.9765 878.743 30.6797 878.743 Q26.4147 878.743 24.0912 881.194 Q21.7677 883.644 21.7677 888.132 L21.7677 896.217 M16.4842 902.646 L16.4842 888.132 Q16.4842 880.143 20.1126 876.069 Q23.7092 871.963 30.6797 871.963 Q37.7138 871.963 41.3104 876.069 Q44.907 880.143 44.907 888.132 L44.907 896.217 L64.0042 896.217 L64.0042 902.646 L16.4842 902.646 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M33.8307 844.081 Q33.2578 845.068 33.0032 846.246 Q32.7167 847.392 32.7167 848.792 Q32.7167 853.757 35.9632 856.431 Q39.1779 859.073 45.2253 859.073 L64.0042 859.073 L64.0042 864.961 L28.3562 864.961 L28.3562 859.073 L33.8944 859.073 Q30.6479 857.227 29.0883 854.267 Q27.4968 851.307 27.4968 847.073 Q27.4968 846.469 27.5923 845.737 Q27.656 845.004 27.8151 844.113 L33.8307 844.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M32.4621 825.557 Q32.4621 830.268 36.1542 833.005 Q39.8145 835.742 46.212 835.742 Q52.6095 835.742 56.3017 833.037 Q59.9619 830.3 59.9619 825.557 Q59.9619 820.878 56.2698 818.141 Q52.5777 815.404 46.212 815.404 Q39.8781 815.404 36.186 818.141 Q32.4621 820.878 32.4621 825.557 M27.4968 825.557 Q27.4968 817.918 32.4621 813.558 Q37.4273 809.197 46.212 809.197 Q54.9649 809.197 59.9619 813.558 Q64.9272 817.918 64.9272 825.557 Q64.9272 833.228 59.9619 837.588 Q54.9649 841.917 46.212 841.917 Q37.4273 841.917 32.4621 837.588 Q27.4968 833.228 27.4968 825.557 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M33.7671 776.032 L14.479 776.032 L14.479 770.176 L64.0042 770.176 L64.0042 776.032 L58.657 776.032 Q61.8398 777.878 63.3994 780.711 Q64.9272 783.512 64.9272 787.459 Q64.9272 793.92 59.771 797.994 Q54.6147 802.036 46.212 802.036 Q37.8093 802.036 32.6531 797.994 Q27.4968 793.92 27.4968 787.459 Q27.4968 783.512 29.0564 780.711 Q30.5842 777.878 33.7671 776.032 M46.212 795.989 Q52.6732 795.989 56.3653 793.347 Q60.0256 790.673 60.0256 786.026 Q60.0256 781.379 56.3653 778.706 Q52.6732 776.032 46.212 776.032 Q39.7508 776.032 36.0905 778.706 Q32.3984 781.379 32.3984 786.026 Q32.3984 790.673 36.0905 793.347 Q39.7508 795.989 46.212 795.989 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M49.9359 758.717 L28.3562 758.717 L28.3562 752.861 L49.7131 752.861 Q54.7739 752.861 57.3202 750.888 Q59.8346 748.914 59.8346 744.967 Q59.8346 740.225 56.8109 737.488 Q53.7872 734.719 48.5673 734.719 L28.3562 734.719 L28.3562 728.862 L64.0042 728.862 L64.0042 734.719 L58.5296 734.719 Q61.7762 736.851 63.3676 739.684 Q64.9272 742.485 64.9272 746.209 Q64.9272 752.352 61.1078 755.534 Q57.2883 758.717 49.9359 758.717 M27.4968 743.981 L27.4968 743.981 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M29.7248 691.145 L35.1993 691.145 Q33.8307 693.628 33.1623 696.142 Q32.4621 698.625 32.4621 701.171 Q32.4621 706.869 36.0905 710.02 Q39.6872 713.171 46.212 713.171 Q52.7369 713.171 56.3653 710.02 Q59.9619 706.869 59.9619 701.171 Q59.9619 698.625 59.2935 696.142 Q58.5933 693.628 57.2247 691.145 L62.6355 691.145 Q63.7814 693.596 64.3543 696.238 Q64.9272 698.848 64.9272 701.808 Q64.9272 709.861 59.8664 714.603 Q54.8057 719.345 46.212 719.345 Q37.491 719.345 32.4939 714.571 Q27.4968 709.765 27.4968 701.426 Q27.4968 698.721 28.0697 696.142 Q28.6108 693.564 29.7248 691.145 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M18.2347 675.167 L28.3562 675.167 L28.3562 663.104 L32.9077 663.104 L32.9077 675.167 L52.2594 675.167 Q56.6199 675.167 57.8613 673.99 Q59.1026 672.78 59.1026 669.12 L59.1026 663.104 L64.0042 663.104 L64.0042 669.12 Q64.0042 675.899 61.4897 678.478 Q58.9434 681.056 52.2594 681.056 L32.9077 681.056 L32.9077 685.353 L28.3562 685.353 L28.3562 681.056 L18.2347 681.056 L18.2347 675.167 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M28.3562 655.402 L28.3562 649.545 L64.0042 649.545 L64.0042 655.402 L28.3562 655.402 M14.479 655.402 L14.479 649.545 L21.895 649.545 L21.895 655.402 L14.479 655.402 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M32.4621 623.478 Q32.4621 628.189 36.1542 630.926 Q39.8145 633.663 46.212 633.663 Q52.6095 633.663 56.3017 630.958 Q59.9619 628.22 59.9619 623.478 Q59.9619 618.799 56.2698 616.062 Q52.5777 613.325 46.212 613.325 Q39.8781 613.325 36.186 616.062 Q32.4621 618.799 32.4621 623.478 M27.4968 623.478 Q27.4968 615.839 32.4621 611.479 Q37.4273 607.118 46.212 607.118 Q54.9649 607.118 59.9619 611.479 Q64.9272 615.839 64.9272 623.478 Q64.9272 631.149 59.9619 635.509 Q54.9649 639.838 46.212 639.838 Q37.4273 639.838 32.4621 635.509 Q27.4968 631.149 27.4968 623.478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M42.4881 567.778 L64.0042 567.778 L64.0042 573.634 L42.679 573.634 Q37.6183 573.634 35.1038 575.608 Q32.5894 577.581 32.5894 581.528 Q32.5894 586.27 35.6131 589.008 Q38.6368 591.745 43.8567 591.745 L64.0042 591.745 L64.0042 597.633 L28.3562 597.633 L28.3562 591.745 L33.8944 591.745 Q30.6797 589.644 29.0883 586.811 Q27.4968 583.947 27.4968 580.223 Q27.4968 574.08 31.3163 570.929 Q35.1038 567.778 42.4881 567.778 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip672)" d="M448.104 597.618 L448.104 735.212 L569.717 735.212 L569.717 597.618 L448.104 597.618 L448.104 597.618  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,597.618 448.104,735.212 569.717,735.212 569.717,597.618 448.104,597.618 "/>
+<path clip-path="url(#clip672)" d="M600.12 735.212 L600.12 735.212 L721.733 735.212 L721.733 735.212 L600.12 735.212 L600.12 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,735.212 600.12,735.212 721.733,735.212 600.12,735.212 "/>
+<path clip-path="url(#clip672)" d="M752.137 528.822 L752.137 597.618 L873.75 597.618 L873.75 528.822 L752.137 528.822 L752.137 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,528.822 752.137,597.618 873.75,597.618 873.75,528.822 752.137,528.822 "/>
+<path clip-path="url(#clip672)" d="M904.153 528.822 L904.153 528.822 L1025.77 528.822 L1025.77 528.822 L904.153 528.822 L904.153 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,528.822 904.153,528.822 1025.77,528.822 904.153,528.822 "/>
+<path clip-path="url(#clip672)" d="M1056.17 460.025 L1056.17 689.347 L1177.78 689.347 L1177.78 460.025 L1056.17 460.025 L1056.17 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,460.025 1056.17,689.347 1177.78,689.347 1177.78,460.025 1056.17,460.025 "/>
+<path clip-path="url(#clip672)" d="M1208.19 528.822 L1208.19 528.822 L1329.8 528.822 L1329.8 528.822 L1208.19 528.822 L1208.19 528.822  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,528.822 1208.19,528.822 1329.8,528.822 1208.19,528.822 "/>
+<path clip-path="url(#clip672)" d="M1360.2 391.228 L1360.2 437.093 L1481.82 437.093 L1481.82 391.228 L1360.2 391.228 L1360.2 391.228  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,391.228 1360.2,437.093 1481.82,437.093 1481.82,391.228 1360.2,391.228 "/>
+<path clip-path="url(#clip672)" d="M1512.22 253.634 L1512.22 253.634 L1633.83 253.634 L1633.83 253.634 L1512.22 253.634 L1512.22 253.634  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,253.634 1512.22,253.634 1633.83,253.634 1512.22,253.634 "/>
+<path clip-path="url(#clip672)" d="M1664.24 47.2441 L1664.24 47.2441 L1785.85 47.2441 L1785.85 47.2441 L1664.24 47.2441 L1664.24 47.2441  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,47.2441 1664.24,47.2441 1785.85,47.2441 1664.24,47.2441 "/>
+<path clip-path="url(#clip672)" d="M1816.25 116.041 L1816.25 116.041 L1937.87 116.041 L1937.87 116.041 L1816.25 116.041 L1816.25 116.041  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,116.041 1816.25,116.041 1937.87,116.041 1816.25,116.041 "/>
+<path clip-path="url(#clip672)" d="M1968.27 460.025 L1968.27 505.889 L2089.88 505.889 L2089.88 460.025 L1968.27 460.025 L1968.27 460.025  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,460.025 1968.27,505.889 2089.88,505.889 2089.88,460.025 1968.27,460.025 "/>
+<path clip-path="url(#clip672)" d="M2120.29 735.212 L2120.29 735.212 L2241.9 735.212 L2241.9 735.212 L2120.29 735.212 L2120.29 735.212  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,735.212 2120.29,735.212 2241.9,735.212 2120.29,735.212 "/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="508.91" cy="597.618" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="660.927" cy="735.212" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="812.943" cy="528.822" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="964.96" cy="528.822" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1116.98" cy="460.025" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1268.99" cy="528.822" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1421.01" cy="391.228" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1573.03" cy="253.634" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1725.04" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="1877.06" cy="116.041" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2029.08" cy="460.025" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#20326c; stroke:none; fill-opacity:0" cx="2181.09" cy="735.212" r="2"/>
+<path clip-path="url(#clip672)" d="M448.104 735.212 L448.104 1193.86 L569.717 1193.86 L569.717 735.212 L448.104 735.212 L448.104 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,735.212 448.104,1193.86 569.717,1193.86 569.717,735.212 448.104,735.212 "/>
+<path clip-path="url(#clip672)" d="M600.12 735.212 L600.12 1170.92 L721.733 1170.92 L721.733 735.212 L600.12 735.212 L600.12 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,735.212 600.12,1170.92 721.733,1170.92 721.733,735.212 600.12,735.212 "/>
+<path clip-path="url(#clip672)" d="M752.137 597.618 L752.137 1147.99 L873.75 1147.99 L873.75 597.618 L752.137 597.618 L752.137 597.618  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,597.618 752.137,1147.99 873.75,1147.99 873.75,597.618 752.137,597.618 "/>
+<path clip-path="url(#clip672)" d="M904.153 528.822 L904.153 964.535 L1025.77 964.535 L1025.77 528.822 L904.153 528.822 L904.153 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,528.822 904.153,964.535 1025.77,964.535 1025.77,528.822 904.153,528.822 "/>
+<path clip-path="url(#clip672)" d="M1056.17 689.347 L1056.17 1423.18 L1177.78 1423.18 L1177.78 689.347 L1056.17 689.347 L1056.17 689.347  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,689.347 1056.17,1423.18 1177.78,1423.18 1177.78,689.347 1056.17,689.347 "/>
+<path clip-path="url(#clip672)" d="M1208.19 528.822 L1208.19 1102.13 L1329.8 1102.13 L1329.8 528.822 L1208.19 528.822 L1208.19 528.822  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,528.822 1208.19,1102.13 1329.8,1102.13 1329.8,528.822 1208.19,528.822 "/>
+<path clip-path="url(#clip672)" d="M1360.2 437.093 L1360.2 1147.99 L1481.82 1147.99 L1481.82 437.093 L1360.2 437.093 L1360.2 437.093  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,437.093 1360.2,1147.99 1481.82,1147.99 1481.82,437.093 1360.2,437.093 "/>
+<path clip-path="url(#clip672)" d="M1512.22 253.634 L1512.22 964.535 L1633.83 964.535 L1633.83 253.634 L1512.22 253.634 L1512.22 253.634  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,253.634 1512.22,964.535 1633.83,964.535 1633.83,253.634 1512.22,253.634 "/>
+<path clip-path="url(#clip672)" d="M1664.24 47.2441 L1664.24 964.535 L1785.85 964.535 L1785.85 47.2441 L1664.24 47.2441 L1664.24 47.2441  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,47.2441 1664.24,964.535 1785.85,964.535 1785.85,47.2441 1664.24,47.2441 "/>
+<path clip-path="url(#clip672)" d="M1816.25 116.041 L1816.25 1019.57 L1937.87 1019.57 L1937.87 116.041 L1816.25 116.041 L1816.25 116.041  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,116.041 1816.25,1019.57 1937.87,1019.57 1937.87,116.041 1816.25,116.041 "/>
+<path clip-path="url(#clip672)" d="M1968.27 505.889 L1968.27 872.806 L2089.88 872.806 L2089.88 505.889 L1968.27 505.889 L1968.27 505.889  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,505.889 1968.27,872.806 2089.88,872.806 2089.88,505.889 1968.27,505.889 "/>
+<path clip-path="url(#clip672)" d="M2120.29 735.212 L2120.29 1423.18 L2241.9 1423.18 L2241.9 735.212 L2120.29 735.212 L2120.29 735.212  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,735.212 2120.29,1423.18 2241.9,1423.18 2241.9,735.212 2120.29,735.212 "/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="508.91" cy="735.212" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="660.927" cy="735.212" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="812.943" cy="597.618" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="964.96" cy="528.822" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1116.98" cy="689.347" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1268.99" cy="528.822" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1421.01" cy="437.093" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1573.03" cy="253.634" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1725.04" cy="47.2441" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="1877.06" cy="116.041" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2029.08" cy="505.889" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#4063d8; stroke:none; fill-opacity:0" cx="2181.09" cy="735.212" r="2"/>
+<path clip-path="url(#clip672)" d="M448.104 1193.86 L448.104 1423.18 L569.717 1423.18 L569.717 1193.86 L448.104 1193.86 L448.104 1193.86  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="448.104,1193.86 448.104,1423.18 569.717,1423.18 569.717,1193.86 448.104,1193.86 "/>
+<path clip-path="url(#clip672)" d="M600.12 1170.92 L600.12 1423.18 L721.733 1423.18 L721.733 1170.92 L600.12 1170.92 L600.12 1170.92  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="600.12,1170.92 600.12,1423.18 721.733,1423.18 721.733,1170.92 600.12,1170.92 "/>
+<path clip-path="url(#clip672)" d="M752.137 1147.99 L752.137 1423.18 L873.75 1423.18 L873.75 1147.99 L752.137 1147.99 L752.137 1147.99  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="752.137,1147.99 752.137,1423.18 873.75,1423.18 873.75,1147.99 752.137,1147.99 "/>
+<path clip-path="url(#clip672)" d="M904.153 964.535 L904.153 1423.18 L1025.77 1423.18 L1025.77 964.535 L904.153 964.535 L904.153 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="904.153,964.535 904.153,1423.18 1025.77,1423.18 1025.77,964.535 904.153,964.535 "/>
+<path clip-path="url(#clip672)" d="M1056.17 1423.18 L1056.17 1423.18 L1177.78 1423.18 L1177.78 1423.18 L1056.17 1423.18 L1056.17 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1056.17,1423.18 1056.17,1423.18 1177.78,1423.18 1056.17,1423.18 "/>
+<path clip-path="url(#clip672)" d="M1208.19 1102.13 L1208.19 1423.18 L1329.8 1423.18 L1329.8 1102.13 L1208.19 1102.13 L1208.19 1102.13  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1208.19,1102.13 1208.19,1423.18 1329.8,1423.18 1329.8,1102.13 1208.19,1102.13 "/>
+<path clip-path="url(#clip672)" d="M1360.2 1147.99 L1360.2 1423.18 L1481.82 1423.18 L1481.82 1147.99 L1360.2 1147.99 L1360.2 1147.99  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.2,1147.99 1360.2,1423.18 1481.82,1423.18 1481.82,1147.99 1360.2,1147.99 "/>
+<path clip-path="url(#clip672)" d="M1512.22 964.535 L1512.22 1423.18 L1633.83 1423.18 L1633.83 964.535 L1512.22 964.535 L1512.22 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1512.22,964.535 1512.22,1423.18 1633.83,1423.18 1633.83,964.535 1512.22,964.535 "/>
+<path clip-path="url(#clip672)" d="M1664.24 964.535 L1664.24 1423.18 L1785.85 1423.18 L1785.85 964.535 L1664.24 964.535 L1664.24 964.535  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1664.24,964.535 1664.24,1423.18 1785.85,1423.18 1785.85,964.535 1664.24,964.535 "/>
+<path clip-path="url(#clip672)" d="M1816.25 1019.57 L1816.25 1423.18 L1937.87 1423.18 L1937.87 1019.57 L1816.25 1019.57 L1816.25 1019.57  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1816.25,1019.57 1816.25,1423.18 1937.87,1423.18 1937.87,1019.57 1816.25,1019.57 "/>
+<path clip-path="url(#clip672)" d="M1968.27 872.806 L1968.27 1423.18 L2089.88 1423.18 L2089.88 872.806 L1968.27 872.806 L1968.27 872.806  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1968.27,872.806 1968.27,1423.18 2089.88,1423.18 2089.88,872.806 1968.27,872.806 "/>
+<path clip-path="url(#clip672)" d="M2120.29 1423.18 L2120.29 1423.18 L2241.9 1423.18 L2241.9 1423.18 L2120.29 1423.18 L2120.29 1423.18  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip672)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2120.29,1423.18 2120.29,1423.18 2241.9,1423.18 2120.29,1423.18 "/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="508.91" cy="1193.86" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="660.927" cy="1170.92" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="812.943" cy="1147.99" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="964.96" cy="964.535" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1116.98" cy="1423.18" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1268.99" cy="1102.13" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1421.01" cy="1147.99" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1573.03" cy="964.535" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1725.04" cy="964.535" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="1877.06" cy="1019.57" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2029.08" cy="872.806" r="2"/>
+<circle clip-path="url(#clip672)" style="fill:#a0b1ec; stroke:none; fill-opacity:0" cx="2181.09" cy="1423.18" r="2"/>
+<path clip-path="url(#clip670)" d="M404.431 300.469 L961.373 300.469 L961.373 93.1086 L404.431 93.1086  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="404.431,300.469 961.373,300.469 961.373,93.1086 404.431,93.1086 404.431,300.469 "/>
+<path clip-path="url(#clip670)" d="M426.825 165.685 L561.192 165.685 L561.192 124.213 L426.825 124.213 L426.825 165.685  Z" fill="#20326c" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,165.685 561.192,165.685 561.192,124.213 426.825,124.213 426.825,165.685 "/>
+<path clip-path="url(#clip670)" d="M583.587 151.997 L583.587 136.303 L587.846 136.303 L587.846 151.835 Q587.846 155.516 589.281 157.368 Q590.717 159.196 593.587 159.196 Q597.036 159.196 599.027 156.997 Q601.041 154.798 601.041 151.002 L601.041 136.303 L605.3 136.303 L605.3 162.229 L601.041 162.229 L601.041 158.247 Q599.49 160.608 597.429 161.766 Q595.392 162.9 592.684 162.9 Q588.217 162.9 585.902 160.122 Q583.587 157.344 583.587 151.997 M594.305 135.678 L594.305 135.678 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M635.624 146.581 L635.624 162.229 L631.365 162.229 L631.365 146.719 Q631.365 143.039 629.929 141.21 Q628.494 139.382 625.624 139.382 Q622.175 139.382 620.184 141.581 Q618.193 143.78 618.193 147.576 L618.193 162.229 L613.911 162.229 L613.911 136.303 L618.193 136.303 L618.193 140.331 Q619.721 137.993 621.781 136.835 Q623.865 135.678 626.573 135.678 Q631.04 135.678 633.332 138.456 Q635.624 141.21 635.624 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M664.304 141.28 Q665.901 138.409 668.124 137.044 Q670.346 135.678 673.355 135.678 Q677.406 135.678 679.605 138.525 Q681.804 141.349 681.804 146.581 L681.804 162.229 L677.522 162.229 L677.522 146.719 Q677.522 142.993 676.202 141.187 Q674.883 139.382 672.174 139.382 Q668.864 139.382 666.943 141.581 Q665.022 143.78 665.022 147.576 L665.022 162.229 L660.739 162.229 L660.739 146.719 Q660.739 142.969 659.42 141.187 Q658.1 139.382 655.346 139.382 Q652.082 139.382 650.161 141.604 Q648.239 143.803 648.239 147.576 L648.239 162.229 L643.957 162.229 L643.957 136.303 L648.239 136.303 L648.239 140.331 Q649.698 137.946 651.735 136.812 Q653.772 135.678 656.573 135.678 Q659.397 135.678 661.364 137.113 Q663.355 138.548 664.304 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M712.475 148.201 L712.475 150.284 L692.892 150.284 Q693.17 154.682 695.531 156.997 Q697.915 159.289 702.151 159.289 Q704.605 159.289 706.897 158.687 Q709.211 158.085 711.48 156.881 L711.48 160.909 Q709.188 161.881 706.781 162.391 Q704.373 162.9 701.897 162.9 Q695.693 162.9 692.059 159.289 Q688.448 155.678 688.448 149.52 Q688.448 143.155 691.873 139.428 Q695.322 135.678 701.156 135.678 Q706.387 135.678 709.42 139.057 Q712.475 142.414 712.475 148.201 M708.216 146.951 Q708.17 143.456 706.248 141.372 Q704.35 139.289 701.202 139.289 Q697.637 139.289 695.485 141.303 Q693.355 143.317 693.031 146.974 L708.216 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M723.679 128.942 L723.679 136.303 L732.452 136.303 L732.452 139.613 L723.679 139.613 L723.679 153.687 Q723.679 156.858 724.535 157.761 Q725.415 158.664 728.077 158.664 L732.452 158.664 L732.452 162.229 L728.077 162.229 Q723.146 162.229 721.271 160.4 Q719.396 158.548 719.396 153.687 L719.396 139.613 L716.271 139.613 L716.271 136.303 L719.396 136.303 L719.396 128.942 L723.679 128.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M770.183 140.238 L770.183 126.21 L774.442 126.21 L774.442 162.229 L770.183 162.229 L770.183 158.34 Q768.841 160.655 766.78 161.789 Q764.743 162.9 761.873 162.9 Q757.174 162.9 754.211 159.15 Q751.271 155.4 751.271 149.289 Q751.271 143.178 754.211 139.428 Q757.174 135.678 761.873 135.678 Q764.743 135.678 766.78 136.812 Q768.841 137.923 770.183 140.238 M755.669 149.289 Q755.669 153.988 757.591 156.673 Q759.535 159.335 762.915 159.335 Q766.294 159.335 768.239 156.673 Q770.183 153.988 770.183 149.289 Q770.183 144.59 768.239 141.928 Q766.294 139.243 762.915 139.243 Q759.535 139.243 757.591 141.928 Q755.669 144.59 755.669 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M805.391 148.201 L805.391 150.284 L785.808 150.284 Q786.086 154.682 788.447 156.997 Q790.831 159.289 795.067 159.289 Q797.521 159.289 799.813 158.687 Q802.127 158.085 804.396 156.881 L804.396 160.909 Q802.104 161.881 799.697 162.391 Q797.29 162.9 794.813 162.9 Q788.609 162.9 784.975 159.289 Q781.364 155.678 781.364 149.52 Q781.364 143.155 784.79 139.428 Q788.239 135.678 794.072 135.678 Q799.303 135.678 802.336 139.057 Q805.391 142.414 805.391 148.201 M801.132 146.951 Q801.086 143.456 799.165 141.372 Q797.266 139.289 794.118 139.289 Q790.553 139.289 788.401 141.303 Q786.271 143.317 785.947 146.974 L801.132 146.951 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M832.567 141.28 Q834.164 138.409 836.387 137.044 Q838.609 135.678 841.618 135.678 Q845.669 135.678 847.868 138.525 Q850.067 141.349 850.067 146.581 L850.067 162.229 L845.785 162.229 L845.785 146.719 Q845.785 142.993 844.465 141.187 Q843.146 139.382 840.437 139.382 Q837.127 139.382 835.206 141.581 Q833.285 143.78 833.285 147.576 L833.285 162.229 L829.002 162.229 L829.002 146.719 Q829.002 142.969 827.683 141.187 Q826.363 139.382 823.609 139.382 Q820.345 139.382 818.424 141.604 Q816.502 143.803 816.502 147.576 L816.502 162.229 L812.22 162.229 L812.22 136.303 L816.502 136.303 L816.502 140.331 Q817.961 137.946 819.998 136.812 Q822.035 135.678 824.836 135.678 Q827.66 135.678 829.627 137.113 Q831.618 138.548 832.567 141.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M870.345 149.196 Q865.183 149.196 863.192 150.377 Q861.201 151.557 861.201 154.405 Q861.201 156.673 862.683 158.016 Q864.187 159.335 866.757 159.335 Q870.298 159.335 872.428 156.835 Q874.581 154.312 874.581 150.145 L874.581 149.196 L870.345 149.196 M878.84 147.437 L878.84 162.229 L874.581 162.229 L874.581 158.293 Q873.122 160.655 870.947 161.789 Q868.771 162.9 865.623 162.9 Q861.641 162.9 859.28 160.678 Q856.942 158.432 856.942 154.682 Q856.942 150.307 859.859 148.085 Q862.798 145.863 868.609 145.863 L874.581 145.863 L874.581 145.446 Q874.581 142.507 872.636 140.909 Q870.715 139.289 867.22 139.289 Q864.998 139.289 862.891 139.821 Q860.785 140.354 858.84 141.419 L858.84 137.483 Q861.178 136.581 863.377 136.141 Q865.576 135.678 867.66 135.678 Q873.285 135.678 876.062 138.594 Q878.84 141.511 878.84 147.437 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M909.164 146.581 L909.164 162.229 L904.905 162.229 L904.905 146.719 Q904.905 143.039 903.47 141.21 Q902.034 139.382 899.164 139.382 Q895.715 139.382 893.724 141.581 Q891.733 143.78 891.733 147.576 L891.733 162.229 L887.451 162.229 L887.451 136.303 L891.733 136.303 L891.733 140.331 Q893.261 137.993 895.321 136.835 Q897.405 135.678 900.113 135.678 Q904.581 135.678 906.872 138.456 Q909.164 141.21 909.164 146.581 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M934.719 140.238 L934.719 126.21 L938.979 126.21 L938.979 162.229 L934.719 162.229 L934.719 158.34 Q933.377 160.655 931.317 161.789 Q929.28 162.9 926.409 162.9 Q921.71 162.9 918.747 159.15 Q915.807 155.4 915.807 149.289 Q915.807 143.178 918.747 139.428 Q921.71 135.678 926.409 135.678 Q929.28 135.678 931.317 136.812 Q933.377 137.923 934.719 140.238 M920.206 149.289 Q920.206 153.988 922.127 156.673 Q924.071 159.335 927.451 159.335 Q930.83 159.335 932.775 156.673 Q934.719 153.988 934.719 149.289 Q934.719 144.59 932.775 141.928 Q930.83 139.243 927.451 139.243 Q924.071 139.243 922.127 141.928 Q920.206 144.59 920.206 149.289 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M426.825 217.525 L561.192 217.525 L561.192 176.053 L426.825 176.053 L426.825 217.525  Z" fill="#4063d8" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,217.525 561.192,217.525 561.192,176.053 426.825,176.053 426.825,217.525 "/>
+<path clip-path="url(#clip670)" d="M599.42 184.115 L593.078 201.314 L605.786 201.314 L599.42 184.115 M596.781 179.509 L602.082 179.509 L615.253 214.069 L610.392 214.069 L607.244 205.203 L591.666 205.203 L588.518 214.069 L583.587 214.069 L596.781 179.509 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip670)" d="M426.825 269.365 L561.192 269.365 L561.192 227.893 L426.825 227.893 L426.825 269.365  Z" fill="#a0b1ec" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip670)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="426.825,269.365 561.192,269.365 561.192,227.893 426.825,227.893 426.825,269.365 "/>
+<path clip-path="url(#clip670)" d="M588.263 249.404 L588.263 262.066 L595.763 262.066 Q599.536 262.066 601.342 260.515 Q603.17 258.941 603.17 255.723 Q603.17 252.483 601.342 250.955 Q599.536 249.404 595.763 249.404 L588.263 249.404 M588.263 235.191 L588.263 245.608 L595.184 245.608 Q598.61 245.608 600.277 244.335 Q601.967 243.038 601.967 240.399 Q601.967 237.784 600.277 236.487 Q598.61 235.191 595.184 235.191 L588.263 235.191 M583.587 231.349 L595.531 231.349 Q600.879 231.349 603.772 233.571 Q606.666 235.793 606.666 239.89 Q606.666 243.062 605.184 244.937 Q603.703 246.811 600.832 247.274 Q604.281 248.015 606.179 250.376 Q608.101 252.714 608.101 256.233 Q608.101 260.862 604.953 263.385 Q601.804 265.909 595.994 265.909 L583.587 265.909 L583.587 231.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/linear/factory_schedule/cc63b3c1.svg b/previews/PR3562/tutorials/linear/factory_schedule/f8d71086.svg
similarity index 86%
rename from previews/PR3562/tutorials/linear/factory_schedule/cc63b3c1.svg
rename to previews/PR3562/tutorials/linear/factory_schedule/f8d71086.svg
index 0e7d685b302..380caa3f216 100644
--- a/previews/PR3562/tutorials/linear/factory_schedule/cc63b3c1.svg
+++ b/previews/PR3562/tutorials/linear/factory_schedule/f8d71086.svg
@@ -1,74 +1,74 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip910">
+  <clipPath id="clip730">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip910)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip730)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip911">
+  <clipPath id="clip731">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip910)" d="M219.38 1423.18 L2112.76 1423.18 L2112.76 47.2441 L219.38 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip730)" d="M219.38 1423.18 L2112.76 1423.18 L2112.76 47.2441 L219.38 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip912">
+  <clipPath id="clip732">
     <rect x="219" y="47" width="1894" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,1423.18 219.38,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="850.505,1423.18 850.505,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1481.63,1423.18 1481.63,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2112.76,1423.18 2112.76,47.2441 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 2112.76,1423.18 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 219.38,1404.28 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="850.505,1423.18 850.505,1404.28 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1481.63,1423.18 1481.63,1404.28 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,1423.18 2112.76,1404.28 "/>
-<path clip-path="url(#clip910)" d="M196.765 1454.1 Q193.153 1454.1 191.325 1457.66 Q189.519 1461.2 189.519 1468.33 Q189.519 1475.44 191.325 1479.01 Q193.153 1482.55 196.765 1482.55 Q200.399 1482.55 202.204 1479.01 Q204.033 1475.44 204.033 1468.33 Q204.033 1461.2 202.204 1457.66 Q200.399 1454.1 196.765 1454.1 M196.765 1450.39 Q202.575 1450.39 205.63 1455 Q208.709 1459.58 208.709 1468.33 Q208.709 1477.06 205.63 1481.67 Q202.575 1486.25 196.765 1486.25 Q190.954 1486.25 187.876 1481.67 Q184.82 1477.06 184.82 1468.33 Q184.82 1459.58 187.876 1455 Q190.954 1450.39 196.765 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M216.926 1479.7 L221.811 1479.7 L221.811 1485.58 L216.926 1485.58 L216.926 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M241.996 1454.1 Q238.385 1454.1 236.556 1457.66 Q234.75 1461.2 234.75 1468.33 Q234.75 1475.44 236.556 1479.01 Q238.385 1482.55 241.996 1482.55 Q245.63 1482.55 247.436 1479.01 Q249.264 1475.44 249.264 1468.33 Q249.264 1461.2 247.436 1457.66 Q245.63 1454.1 241.996 1454.1 M241.996 1450.39 Q247.806 1450.39 250.861 1455 Q253.94 1459.58 253.94 1468.33 Q253.94 1477.06 250.861 1481.67 Q247.806 1486.25 241.996 1486.25 Q236.186 1486.25 233.107 1481.67 Q230.051 1477.06 230.051 1468.33 Q230.051 1459.58 233.107 1455 Q236.186 1450.39 241.996 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M828.387 1454.1 Q824.776 1454.1 822.948 1457.66 Q821.142 1461.2 821.142 1468.33 Q821.142 1475.44 822.948 1479.01 Q824.776 1482.55 828.387 1482.55 Q832.022 1482.55 833.827 1479.01 Q835.656 1475.44 835.656 1468.33 Q835.656 1461.2 833.827 1457.66 Q832.022 1454.1 828.387 1454.1 M828.387 1450.39 Q834.198 1450.39 837.253 1455 Q840.332 1459.58 840.332 1468.33 Q840.332 1477.06 837.253 1481.67 Q834.198 1486.25 828.387 1486.25 Q822.577 1486.25 819.499 1481.67 Q816.443 1477.06 816.443 1468.33 Q816.443 1459.58 819.499 1455 Q822.577 1450.39 828.387 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M848.549 1479.7 L853.434 1479.7 L853.434 1485.58 L848.549 1485.58 L848.549 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M863.665 1451.02 L882.021 1451.02 L882.021 1454.96 L867.947 1454.96 L867.947 1463.43 Q868.966 1463.08 869.984 1462.92 Q871.003 1462.73 872.021 1462.73 Q877.809 1462.73 881.188 1465.9 Q884.568 1469.08 884.568 1474.49 Q884.568 1480.07 881.096 1483.17 Q877.623 1486.25 871.304 1486.25 Q869.128 1486.25 866.859 1485.88 Q864.614 1485.51 862.207 1484.77 L862.207 1480.07 Q864.29 1481.2 866.512 1481.76 Q868.734 1482.32 871.211 1482.32 Q875.216 1482.32 877.554 1480.21 Q879.892 1478.1 879.892 1474.49 Q879.892 1470.88 877.554 1468.77 Q875.216 1466.67 871.211 1466.67 Q869.336 1466.67 867.461 1467.08 Q865.609 1467.5 863.665 1468.38 L863.665 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1448.78 1481.64 L1456.42 1481.64 L1456.42 1455.28 L1448.11 1456.95 L1448.11 1452.69 L1456.38 1451.02 L1461.05 1451.02 L1461.05 1481.64 L1468.69 1481.64 L1468.69 1485.58 L1448.78 1485.58 L1448.78 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1478.14 1479.7 L1483.02 1479.7 L1483.02 1485.58 L1478.14 1485.58 L1478.14 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1503.2 1454.1 Q1499.59 1454.1 1497.76 1457.66 Q1495.96 1461.2 1495.96 1468.33 Q1495.96 1475.44 1497.76 1479.01 Q1499.59 1482.55 1503.2 1482.55 Q1506.84 1482.55 1508.64 1479.01 Q1510.47 1475.44 1510.47 1468.33 Q1510.47 1461.2 1508.64 1457.66 Q1506.84 1454.1 1503.2 1454.1 M1503.2 1450.39 Q1509.01 1450.39 1512.07 1455 Q1515.15 1459.58 1515.15 1468.33 Q1515.15 1477.06 1512.07 1481.67 Q1509.01 1486.25 1503.2 1486.25 Q1497.39 1486.25 1494.32 1481.67 Q1491.26 1477.06 1491.26 1468.33 Q1491.26 1459.58 1494.32 1455 Q1497.39 1450.39 1503.2 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2080.41 1481.64 L2088.05 1481.64 L2088.05 1455.28 L2079.74 1456.95 L2079.74 1452.69 L2088 1451.02 L2092.67 1451.02 L2092.67 1481.64 L2100.31 1481.64 L2100.31 1485.58 L2080.41 1485.58 L2080.41 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2109.76 1479.7 L2114.64 1479.7 L2114.64 1485.58 L2109.76 1485.58 L2109.76 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2124.87 1451.02 L2143.23 1451.02 L2143.23 1454.96 L2129.16 1454.96 L2129.16 1463.43 Q2130.17 1463.08 2131.19 1462.92 Q2132.21 1462.73 2133.23 1462.73 Q2139.02 1462.73 2142.4 1465.9 Q2145.78 1469.08 2145.78 1474.49 Q2145.78 1480.07 2142.3 1483.17 Q2138.83 1486.25 2132.51 1486.25 Q2130.34 1486.25 2128.07 1485.88 Q2125.82 1485.51 2123.42 1484.77 L2123.42 1480.07 Q2125.5 1481.2 2127.72 1481.76 Q2129.94 1482.32 2132.42 1482.32 Q2136.42 1482.32 2138.76 1480.21 Q2141.1 1478.1 2141.1 1474.49 Q2141.1 1470.88 2138.76 1468.77 Q2136.42 1466.67 2132.42 1466.67 Q2130.55 1466.67 2128.67 1467.08 Q2126.82 1467.5 2124.87 1468.38 L2124.87 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M911.822 1522.08 L911.822 1528.35 Q908.161 1526.6 904.915 1525.74 Q901.668 1524.88 898.645 1524.88 Q893.393 1524.88 890.528 1526.92 Q887.696 1528.96 887.696 1532.71 Q887.696 1535.87 889.574 1537.49 Q891.483 1539.08 896.767 1540.07 L900.65 1540.86 Q907.843 1542.23 911.249 1545.7 Q914.686 1549.14 914.686 1554.93 Q914.686 1561.84 910.039 1565.4 Q905.424 1568.97 896.48 1568.97 Q893.107 1568.97 889.287 1568.2 Q885.5 1567.44 881.425 1565.94 L881.425 1559.32 Q885.34 1561.52 889.096 1562.63 Q892.852 1563.75 896.48 1563.75 Q901.987 1563.75 904.979 1561.58 Q907.97 1559.42 907.97 1555.41 Q907.97 1551.91 905.806 1549.93 Q903.674 1547.96 898.772 1546.97 L894.857 1546.21 Q887.664 1544.78 884.449 1541.72 Q881.234 1538.67 881.234 1533.22 Q881.234 1526.92 885.659 1523.29 Q890.115 1519.66 897.913 1519.66 Q901.255 1519.66 904.724 1520.27 Q908.193 1520.87 911.822 1522.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M950.111 1533.76 L950.111 1539.24 Q947.629 1537.87 945.114 1537.2 Q942.632 1536.5 940.085 1536.5 Q934.388 1536.5 931.237 1540.13 Q928.086 1543.73 928.086 1550.25 Q928.086 1556.78 931.237 1560.4 Q934.388 1564 940.085 1564 Q942.632 1564 945.114 1563.33 Q947.629 1562.63 950.111 1561.26 L950.111 1566.68 Q947.661 1567.82 945.019 1568.39 Q942.409 1568.97 939.449 1568.97 Q931.396 1568.97 926.654 1563.91 Q921.911 1558.85 921.911 1550.25 Q921.911 1541.53 926.686 1536.53 Q931.492 1531.54 939.831 1531.54 Q942.536 1531.54 945.114 1532.11 Q947.692 1532.65 950.111 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M976.497 1550.12 Q969.4 1550.12 966.662 1551.75 Q963.925 1553.37 963.925 1557.29 Q963.925 1560.4 965.962 1562.25 Q968.031 1564.07 971.564 1564.07 Q976.434 1564.07 979.362 1560.63 Q982.322 1557.16 982.322 1551.43 L982.322 1550.12 L976.497 1550.12 M988.178 1547.71 L988.178 1568.04 L982.322 1568.04 L982.322 1562.63 Q980.317 1565.88 977.325 1567.44 Q974.333 1568.97 970.004 1568.97 Q964.53 1568.97 961.283 1565.91 Q958.069 1562.82 958.069 1557.67 Q958.069 1551.65 962.079 1548.6 Q966.121 1545.54 974.11 1545.54 L982.322 1545.54 L982.322 1544.97 Q982.322 1540.93 979.648 1538.73 Q977.007 1536.5 972.2 1536.5 Q969.145 1536.5 966.249 1537.23 Q963.352 1537.97 960.679 1539.43 L960.679 1534.02 Q963.893 1532.78 966.917 1532.17 Q969.941 1531.54 972.805 1531.54 Q980.54 1531.54 984.359 1535.55 Q988.178 1539.56 988.178 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1000.24 1518.52 L1006.1 1518.52 L1006.1 1568.04 L1000.24 1568.04 L1000.24 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1048.84 1548.76 L1048.84 1551.62 L1021.92 1551.62 Q1022.3 1557.67 1025.55 1560.85 Q1028.82 1564 1034.65 1564 Q1038.02 1564 1041.17 1563.17 Q1044.36 1562.35 1047.47 1560.69 L1047.47 1566.23 Q1044.32 1567.57 1041.01 1568.27 Q1037.7 1568.97 1034.3 1568.97 Q1025.77 1568.97 1020.77 1564 Q1015.81 1559.04 1015.81 1550.57 Q1015.81 1541.82 1020.52 1536.69 Q1025.26 1531.54 1033.28 1531.54 Q1040.47 1531.54 1044.64 1536.18 Q1048.84 1540.8 1048.84 1548.76 M1042.99 1547.04 Q1042.92 1542.23 1040.28 1539.37 Q1037.67 1536.5 1033.34 1536.5 Q1028.44 1536.5 1025.48 1539.27 Q1022.55 1542.04 1022.11 1547.07 L1042.99 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1092.99 1536.5 Q1088.28 1536.5 1085.54 1540.19 Q1082.8 1543.85 1082.8 1550.25 Q1082.8 1556.65 1085.51 1560.34 Q1088.25 1564 1092.99 1564 Q1097.67 1564 1100.41 1560.31 Q1103.14 1556.62 1103.14 1550.25 Q1103.14 1543.92 1100.41 1540.23 Q1097.67 1536.5 1092.99 1536.5 M1092.99 1531.54 Q1100.63 1531.54 1104.99 1536.5 Q1109.35 1541.47 1109.35 1550.25 Q1109.35 1559 1104.99 1564 Q1100.63 1568.97 1092.99 1568.97 Q1085.32 1568.97 1080.96 1564 Q1076.63 1559 1076.63 1550.25 Q1076.63 1541.47 1080.96 1536.5 Q1085.32 1531.54 1092.99 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1137.1 1518.52 L1137.1 1523.39 L1131.5 1523.39 Q1128.35 1523.39 1127.11 1524.66 Q1125.9 1525.93 1125.9 1529.24 L1125.9 1532.4 L1135.54 1532.4 L1135.54 1536.95 L1125.9 1536.95 L1125.9 1568.04 L1120.01 1568.04 L1120.01 1536.95 L1114.41 1536.95 L1114.41 1532.4 L1120.01 1532.4 L1120.01 1529.91 Q1120.01 1523.96 1122.78 1521.26 Q1125.55 1518.52 1131.57 1518.52 L1137.1 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1180.77 1518.52 L1180.77 1523.39 L1175.17 1523.39 Q1172.02 1523.39 1170.78 1524.66 Q1169.57 1525.93 1169.57 1529.24 L1169.57 1532.4 L1179.21 1532.4 L1179.21 1536.95 L1169.57 1536.95 L1169.57 1568.04 L1163.68 1568.04 L1163.68 1536.95 L1158.08 1536.95 L1158.08 1532.4 L1163.68 1532.4 L1163.68 1529.91 Q1163.68 1523.96 1166.45 1521.26 Q1169.22 1518.52 1175.23 1518.52 L1180.77 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1201.88 1550.12 Q1194.78 1550.12 1192.04 1551.75 Q1189.3 1553.37 1189.3 1557.29 Q1189.3 1560.4 1191.34 1562.25 Q1193.41 1564.07 1196.94 1564.07 Q1201.81 1564.07 1204.74 1560.63 Q1207.7 1557.16 1207.7 1551.43 L1207.7 1550.12 L1201.88 1550.12 M1213.56 1547.71 L1213.56 1568.04 L1207.7 1568.04 L1207.7 1562.63 Q1205.69 1565.88 1202.7 1567.44 Q1199.71 1568.97 1195.38 1568.97 Q1189.91 1568.97 1186.66 1565.91 Q1183.45 1562.82 1183.45 1557.67 Q1183.45 1551.65 1187.46 1548.6 Q1191.5 1545.54 1199.49 1545.54 L1207.7 1545.54 L1207.7 1544.97 Q1207.7 1540.93 1205.03 1538.73 Q1202.38 1536.5 1197.58 1536.5 Q1194.52 1536.5 1191.63 1537.23 Q1188.73 1537.97 1186.06 1539.43 L1186.06 1534.02 Q1189.27 1532.78 1192.29 1532.17 Q1195.32 1531.54 1198.18 1531.54 Q1205.92 1531.54 1209.74 1535.55 Q1213.56 1539.56 1213.56 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1251.27 1533.76 L1251.27 1539.24 Q1248.79 1537.87 1246.28 1537.2 Q1243.79 1536.5 1241.25 1536.5 Q1235.55 1536.5 1232.4 1540.13 Q1229.25 1543.73 1229.25 1550.25 Q1229.25 1556.78 1232.4 1560.4 Q1235.55 1564 1241.25 1564 Q1243.79 1564 1246.28 1563.33 Q1248.79 1562.63 1251.27 1561.26 L1251.27 1566.68 Q1248.82 1567.82 1246.18 1568.39 Q1243.57 1568.97 1240.61 1568.97 Q1232.56 1568.97 1227.82 1563.91 Q1223.07 1558.85 1223.07 1550.25 Q1223.07 1541.53 1227.85 1536.53 Q1232.65 1531.54 1240.99 1531.54 Q1243.7 1531.54 1246.28 1532.11 Q1248.85 1532.65 1251.27 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1267.25 1522.27 L1267.25 1532.4 L1279.31 1532.4 L1279.31 1536.95 L1267.25 1536.95 L1267.25 1556.3 Q1267.25 1560.66 1268.43 1561.9 Q1269.64 1563.14 1273.3 1563.14 L1279.31 1563.14 L1279.31 1568.04 L1273.3 1568.04 Q1266.52 1568.04 1263.94 1565.53 Q1261.36 1562.98 1261.36 1556.3 L1261.36 1536.95 L1257.07 1536.95 L1257.07 1532.4 L1261.36 1532.4 L1261.36 1522.27 L1267.25 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1300.83 1536.5 Q1296.12 1536.5 1293.38 1540.19 Q1290.64 1543.85 1290.64 1550.25 Q1290.64 1556.65 1293.35 1560.34 Q1296.09 1564 1300.83 1564 Q1305.51 1564 1308.25 1560.31 Q1310.98 1556.62 1310.98 1550.25 Q1310.98 1543.92 1308.25 1540.23 Q1305.51 1536.5 1300.83 1536.5 M1300.83 1531.54 Q1308.47 1531.54 1312.83 1536.5 Q1317.19 1541.47 1317.19 1550.25 Q1317.19 1559 1312.83 1564 Q1308.47 1568.97 1300.83 1568.97 Q1293.16 1568.97 1288.8 1564 Q1284.47 1559 1284.47 1550.25 Q1284.47 1541.47 1288.8 1536.5 Q1293.16 1531.54 1300.83 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1347.55 1537.87 Q1346.57 1537.3 1345.39 1537.04 Q1344.24 1536.76 1342.84 1536.76 Q1337.88 1536.76 1335.2 1540 Q1332.56 1543.22 1332.56 1549.27 L1332.56 1568.04 L1326.67 1568.04 L1326.67 1532.4 L1332.56 1532.4 L1332.56 1537.93 Q1334.41 1534.69 1337.37 1533.13 Q1340.33 1531.54 1344.56 1531.54 Q1345.17 1531.54 1345.9 1531.63 Q1346.63 1531.7 1347.52 1531.85 L1347.55 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1368.53 1571.35 Q1366.05 1577.72 1363.69 1579.66 Q1361.34 1581.6 1357.39 1581.6 L1352.71 1581.6 L1352.71 1576.7 L1356.15 1576.7 Q1358.57 1576.7 1359.9 1575.56 Q1361.24 1574.41 1362.86 1570.14 L1363.91 1567.47 L1349.5 1532.4 L1355.7 1532.4 L1366.84 1560.28 L1377.98 1532.4 L1384.19 1532.4 L1368.53 1571.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1429.13 1526.86 L1420.41 1550.51 L1437.88 1550.51 L1429.13 1526.86 M1425.5 1520.52 L1432.79 1520.52 L1450.9 1568.04 L1444.22 1568.04 L1439.89 1555.85 L1418.47 1555.85 L1414.14 1568.04 L1407.36 1568.04 L1425.5 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,1423.18 2112.76,1423.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,964.535 2112.76,964.535 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,505.889 2112.76,505.889 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,47.2441 2112.76,47.2441 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 219.38,47.2441 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 236.152,1423.18 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,964.535 236.152,964.535 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,505.889 236.152,505.889 "/>
-<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,47.2441 236.152,47.2441 "/>
-<path clip-path="url(#clip910)" d="M126.205 1408.98 Q122.593 1408.98 120.765 1412.54 Q118.959 1416.08 118.959 1423.21 Q118.959 1430.32 120.765 1433.89 Q122.593 1437.43 126.205 1437.43 Q129.839 1437.43 131.644 1433.89 Q133.473 1430.32 133.473 1423.21 Q133.473 1416.08 131.644 1412.54 Q129.839 1408.98 126.205 1408.98 M126.205 1405.27 Q132.015 1405.27 135.07 1409.88 Q138.149 1414.46 138.149 1423.21 Q138.149 1431.94 135.07 1436.55 Q132.015 1441.13 126.205 1441.13 Q120.394 1441.13 117.316 1436.55 Q114.26 1431.94 114.26 1423.21 Q114.26 1414.46 117.316 1409.88 Q120.394 1405.27 126.205 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M146.366 1434.58 L151.251 1434.58 L151.251 1440.46 L146.366 1440.46 L146.366 1434.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M171.436 1408.98 Q167.825 1408.98 165.996 1412.54 Q164.19 1416.08 164.19 1423.21 Q164.19 1430.32 165.996 1433.89 Q167.825 1437.43 171.436 1437.43 Q175.07 1437.43 176.876 1433.89 Q178.704 1430.32 178.704 1423.21 Q178.704 1416.08 176.876 1412.54 Q175.07 1408.98 171.436 1408.98 M171.436 1405.27 Q177.246 1405.27 180.301 1409.88 Q183.38 1414.46 183.38 1423.21 Q183.38 1431.94 180.301 1436.55 Q177.246 1441.13 171.436 1441.13 Q165.626 1441.13 162.547 1436.55 Q159.491 1431.94 159.491 1423.21 Q159.491 1414.46 162.547 1409.88 Q165.626 1405.27 171.436 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M127.2 950.333 Q123.589 950.333 121.76 953.898 Q119.955 957.44 119.955 964.569 Q119.955 971.676 121.76 975.241 Q123.589 978.782 127.2 978.782 Q130.834 978.782 132.64 975.241 Q134.468 971.676 134.468 964.569 Q134.468 957.44 132.64 953.898 Q130.834 950.333 127.2 950.333 M127.2 946.63 Q133.01 946.63 136.066 951.236 Q139.144 955.819 139.144 964.569 Q139.144 973.296 136.066 977.903 Q133.01 982.486 127.2 982.486 Q121.39 982.486 118.311 977.903 Q115.256 973.296 115.256 964.569 Q115.256 955.819 118.311 951.236 Q121.39 946.63 127.2 946.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M147.362 975.935 L152.246 975.935 L152.246 981.815 L147.362 981.815 L147.362 975.935 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M162.477 947.255 L180.834 947.255 L180.834 951.19 L166.76 951.19 L166.76 959.662 Q167.778 959.315 168.797 959.153 Q169.815 958.967 170.834 958.967 Q176.621 958.967 180.001 962.139 Q183.38 965.31 183.38 970.727 Q183.38 976.305 179.908 979.407 Q176.436 982.486 170.116 982.486 Q167.94 982.486 165.672 982.116 Q163.427 981.745 161.019 981.004 L161.019 976.305 Q163.102 977.44 165.325 977.995 Q167.547 978.551 170.024 978.551 Q174.028 978.551 176.366 976.444 Q178.704 974.338 178.704 970.727 Q178.704 967.116 176.366 965.009 Q174.028 962.903 170.024 962.903 Q168.149 962.903 166.274 963.319 Q164.422 963.736 162.477 964.616 L162.477 947.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M117.015 519.234 L124.654 519.234 L124.654 492.869 L116.343 494.535 L116.343 490.276 L124.607 488.609 L129.283 488.609 L129.283 519.234 L136.922 519.234 L136.922 523.169 L117.015 523.169 L117.015 519.234 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M146.366 517.29 L151.251 517.29 L151.251 523.169 L146.366 523.169 L146.366 517.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M171.436 491.688 Q167.825 491.688 165.996 495.253 Q164.19 498.794 164.19 505.924 Q164.19 513.031 165.996 516.595 Q167.825 520.137 171.436 520.137 Q175.07 520.137 176.876 516.595 Q178.704 513.031 178.704 505.924 Q178.704 498.794 176.876 495.253 Q175.07 491.688 171.436 491.688 M171.436 487.984 Q177.246 487.984 180.301 492.591 Q183.38 497.174 183.38 505.924 Q183.38 514.651 180.301 519.257 Q177.246 523.841 171.436 523.841 Q165.626 523.841 162.547 519.257 Q159.491 514.651 159.491 505.924 Q159.491 497.174 162.547 492.591 Q165.626 487.984 171.436 487.984 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M118.01 60.5889 L125.649 60.5889 L125.649 34.2233 L117.339 35.89 L117.339 31.6308 L125.603 29.9641 L130.279 29.9641 L130.279 60.5889 L137.917 60.5889 L137.917 64.5241 L118.01 64.5241 L118.01 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M147.362 58.6445 L152.246 58.6445 L152.246 64.5241 L147.362 64.5241 L147.362 58.6445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M162.477 29.9641 L180.834 29.9641 L180.834 33.8993 L166.76 33.8993 L166.76 42.3714 Q167.778 42.0242 168.797 41.8622 Q169.815 41.677 170.834 41.677 Q176.621 41.677 180.001 44.8483 Q183.38 48.0196 183.38 53.4362 Q183.38 59.0149 179.908 62.1167 Q176.436 65.1954 170.116 65.1954 Q167.94 65.1954 165.672 64.825 Q163.427 64.4547 161.019 63.7139 L161.019 59.0149 Q163.102 60.1491 165.325 60.7047 Q167.547 61.2602 170.024 61.2602 Q174.028 61.2602 176.366 59.1538 Q178.704 57.0473 178.704 53.4362 Q178.704 49.8251 176.366 47.7186 Q174.028 45.6122 170.024 45.6122 Q168.149 45.6122 166.274 46.0288 Q164.422 46.4455 162.477 47.3251 L162.477 29.9641 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M18.0438 987.485 L24.314 987.485 Q22.5634 991.145 21.704 994.392 Q20.8447 997.638 20.8447 1000.66 Q20.8447 1005.91 22.8817 1008.78 Q24.9187 1011.61 28.6745 1011.61 Q31.8255 1011.61 33.4488 1009.73 Q35.0402 1007.82 36.0269 1002.54 L36.8226 998.657 Q38.1912 991.463 41.6605 988.058 Q45.098 984.62 50.8908 984.62 Q57.7976 984.62 61.3624 989.267 Q64.9272 993.882 64.9272 1002.83 Q64.9272 1006.2 64.1633 1010.02 Q63.3994 1013.81 61.9035 1017.88 L55.2831 1017.88 Q57.4793 1013.97 58.5933 1010.21 Q59.7073 1006.45 59.7073 1002.83 Q59.7073 997.32 57.543 994.328 Q55.3786 991.336 51.3682 991.336 Q47.8671 991.336 45.8937 993.501 Q43.9204 995.633 42.9337 1000.53 L42.1698 1004.45 Q40.7375 1011.64 37.682 1014.86 Q34.6264 1018.07 29.1837 1018.07 Q22.8817 1018.07 19.2532 1013.65 Q15.6248 1009.19 15.6248 1001.39 Q15.6248 998.052 16.2295 994.583 Q16.8343 991.113 18.0438 987.485 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M29.7248 949.195 L35.1993 949.195 Q33.8307 951.678 33.1623 954.192 Q32.4621 956.675 32.4621 959.221 Q32.4621 964.918 36.0905 968.07 Q39.6872 971.221 46.212 971.221 Q52.7369 971.221 56.3653 968.07 Q59.9619 964.918 59.9619 959.221 Q59.9619 956.675 59.2935 954.192 Q58.5933 951.678 57.2247 949.195 L62.6355 949.195 Q63.7814 951.646 64.3543 954.288 Q64.9272 956.898 64.9272 959.858 Q64.9272 967.91 59.8664 972.653 Q54.8057 977.395 46.212 977.395 Q37.491 977.395 32.4939 972.621 Q27.4968 967.815 27.4968 959.476 Q27.4968 956.77 28.0697 954.192 Q28.6108 951.614 29.7248 949.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M46.0847 922.809 Q46.0847 929.907 47.7079 932.644 Q49.3312 935.382 53.2461 935.382 Q56.3653 935.382 58.2114 933.345 Q60.0256 931.276 60.0256 927.743 Q60.0256 922.873 56.5881 919.945 Q53.1188 916.985 47.3897 916.985 L46.0847 916.985 L46.0847 922.809 M43.6657 911.128 L64.0042 911.128 L64.0042 916.985 L58.5933 916.985 Q61.8398 918.99 63.3994 921.982 Q64.9272 924.974 64.9272 929.302 Q64.9272 934.777 61.8716 938.023 Q58.7843 941.238 53.6281 941.238 Q47.6125 941.238 44.5569 937.228 Q41.5014 933.185 41.5014 925.196 L41.5014 916.985 L40.9285 916.985 Q36.8862 916.985 34.6901 919.658 Q32.4621 922.3 32.4621 927.106 Q32.4621 930.162 33.1941 933.058 Q33.9262 935.955 35.3903 938.628 L29.9795 938.628 Q28.7381 935.413 28.1334 932.39 Q27.4968 929.366 27.4968 926.501 Q27.4968 918.767 31.5072 914.948 Q35.5176 911.128 43.6657 911.128 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M14.479 899.065 L14.479 893.209 L64.0042 893.209 L64.0042 899.065 L14.479 899.065 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M44.7161 850.463 L47.5806 850.463 L47.5806 877.39 Q53.6281 877.008 56.8109 873.762 Q59.9619 870.483 59.9619 864.659 Q59.9619 861.285 59.1344 858.134 Q58.3069 854.951 56.6518 851.832 L62.1899 851.832 Q63.5267 854.983 64.227 858.293 Q64.9272 861.603 64.9272 865.009 Q64.9272 873.539 59.9619 878.536 Q54.9967 883.501 46.5303 883.501 Q37.7774 883.501 32.6531 878.79 Q27.4968 874.048 27.4968 866.027 Q27.4968 858.834 32.1438 854.664 Q36.7589 850.463 44.7161 850.463 M42.9973 856.32 Q38.1912 856.383 35.3266 859.025 Q32.4621 861.635 32.4621 865.964 Q32.4621 870.865 35.2312 873.825 Q38.0002 876.753 43.0292 877.199 L42.9973 856.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M32.4621 806.317 Q32.4621 811.028 36.1542 813.765 Q39.8145 816.502 46.212 816.502 Q52.6095 816.502 56.3017 813.797 Q59.9619 811.059 59.9619 806.317 Q59.9619 801.638 56.2698 798.901 Q52.5777 796.164 46.212 796.164 Q39.8781 796.164 36.186 798.901 Q32.4621 801.638 32.4621 806.317 M27.4968 806.317 Q27.4968 798.678 32.4621 794.318 Q37.4273 789.957 46.212 789.957 Q54.9649 789.957 59.9619 794.318 Q64.9272 798.678 64.9272 806.317 Q64.9272 813.988 59.9619 818.348 Q54.9649 822.677 46.212 822.677 Q37.4273 822.677 32.4621 818.348 Q27.4968 813.988 27.4968 806.317 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M14.479 762.203 L19.3487 762.203 L19.3487 767.804 Q19.3487 770.955 20.6219 772.197 Q21.895 773.406 25.2052 773.406 L28.3562 773.406 L28.3562 763.762 L32.9077 763.762 L32.9077 773.406 L64.0042 773.406 L64.0042 779.294 L32.9077 779.294 L32.9077 784.896 L28.3562 784.896 L28.3562 779.294 L25.8736 779.294 Q19.9216 779.294 17.2162 776.525 Q14.479 773.756 14.479 767.741 L14.479 762.203 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M14.479 718.534 L19.3487 718.534 L19.3487 724.136 Q19.3487 727.287 20.6219 728.528 Q21.895 729.737 25.2052 729.737 L28.3562 729.737 L28.3562 720.093 L32.9077 720.093 L32.9077 729.737 L64.0042 729.737 L64.0042 735.626 L32.9077 735.626 L32.9077 741.228 L28.3562 741.228 L28.3562 735.626 L25.8736 735.626 Q19.9216 735.626 17.2162 732.857 Q14.479 730.088 14.479 724.072 L14.479 718.534 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M46.0847 697.431 Q46.0847 704.529 47.7079 707.267 Q49.3312 710.004 53.2461 710.004 Q56.3653 710.004 58.2114 707.967 Q60.0256 705.898 60.0256 702.365 Q60.0256 697.495 56.5881 694.567 Q53.1188 691.607 47.3897 691.607 L46.0847 691.607 L46.0847 697.431 M43.6657 685.75 L64.0042 685.75 L64.0042 691.607 L58.5933 691.607 Q61.8398 693.612 63.3994 696.604 Q64.9272 699.596 64.9272 703.925 Q64.9272 709.399 61.8716 712.646 Q58.7843 715.86 53.6281 715.86 Q47.6125 715.86 44.5569 711.85 Q41.5014 707.808 41.5014 699.819 L41.5014 691.607 L40.9285 691.607 Q36.8862 691.607 34.6901 694.28 Q32.4621 696.922 32.4621 701.728 Q32.4621 704.784 33.1941 707.68 Q33.9262 710.577 35.3903 713.25 L29.9795 713.25 Q28.7381 710.036 28.1334 707.012 Q27.4968 703.988 27.4968 701.124 Q27.4968 693.389 31.5072 689.57 Q35.5176 685.75 43.6657 685.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M29.7248 648.034 L35.1993 648.034 Q33.8307 650.516 33.1623 653.031 Q32.4621 655.513 32.4621 658.06 Q32.4621 663.757 36.0905 666.908 Q39.6872 670.059 46.212 670.059 Q52.7369 670.059 56.3653 666.908 Q59.9619 663.757 59.9619 658.06 Q59.9619 655.513 59.2935 653.031 Q58.5933 650.516 57.2247 648.034 L62.6355 648.034 Q63.7814 650.484 64.3543 653.126 Q64.9272 655.736 64.9272 658.696 Q64.9272 666.749 59.8664 671.491 Q54.8057 676.234 46.212 676.234 Q37.491 676.234 32.4939 671.459 Q27.4968 666.653 27.4968 658.314 Q27.4968 655.609 28.0697 653.031 Q28.6108 650.453 29.7248 648.034 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M18.2347 632.056 L28.3562 632.056 L28.3562 619.993 L32.9077 619.993 L32.9077 632.056 L52.2594 632.056 Q56.6199 632.056 57.8613 630.878 Q59.1026 629.669 59.1026 626.008 L59.1026 619.993 L64.0042 619.993 L64.0042 626.008 Q64.0042 632.788 61.4897 635.366 Q58.9434 637.944 52.2594 637.944 L32.9077 637.944 L32.9077 642.241 L28.3562 642.241 L28.3562 637.944 L18.2347 637.944 L18.2347 632.056 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M32.4621 598.477 Q32.4621 603.187 36.1542 605.924 Q39.8145 608.662 46.212 608.662 Q52.6095 608.662 56.3017 605.956 Q59.9619 603.219 59.9619 598.477 Q59.9619 593.798 56.2698 591.061 Q52.5777 588.323 46.212 588.323 Q39.8781 588.323 36.186 591.061 Q32.4621 593.798 32.4621 598.477 M27.4968 598.477 Q27.4968 590.838 32.4621 586.477 Q37.4273 582.117 46.212 582.117 Q54.9649 582.117 59.9619 586.477 Q64.9272 590.838 64.9272 598.477 Q64.9272 606.147 59.9619 610.508 Q54.9649 614.836 46.212 614.836 Q37.4273 614.836 32.4621 610.508 Q27.4968 606.147 27.4968 598.477 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M33.8307 551.752 Q33.2578 552.739 33.0032 553.917 Q32.7167 555.062 32.7167 556.463 Q32.7167 561.428 35.9632 564.102 Q39.1779 566.744 45.2253 566.744 L64.0042 566.744 L64.0042 572.632 L28.3562 572.632 L28.3562 566.744 L33.8944 566.744 Q30.6479 564.897 29.0883 561.937 Q27.4968 558.977 27.4968 554.744 Q27.4968 554.139 27.5923 553.407 Q27.656 552.675 27.8151 551.784 L33.8307 551.752 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M67.3143 530.777 Q73.68 533.26 75.6216 535.615 Q77.5631 537.971 77.5631 541.917 L77.5631 546.596 L72.6615 546.596 L72.6615 543.159 Q72.6615 540.74 71.5157 539.403 Q70.3699 538.066 66.1048 536.443 L63.4312 535.392 L28.3562 549.811 L28.3562 543.604 L56.238 532.464 L28.3562 521.324 L28.3562 515.118 L67.3143 530.777 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M41.3104 479.629 L58.7206 479.629 L58.7206 469.316 Q58.7206 464.128 56.5881 461.646 Q54.4238 459.131 49.9996 459.131 Q45.5436 459.131 43.4429 461.646 Q41.3104 464.128 41.3104 469.316 L41.3104 479.629 M21.7677 479.629 L36.0905 479.629 L36.0905 470.112 Q36.0905 465.401 34.34 463.11 Q32.5576 460.786 28.9291 460.786 Q25.3325 460.786 23.5501 463.11 Q21.7677 465.401 21.7677 470.112 L21.7677 479.629 M16.4842 486.058 L16.4842 469.635 Q16.4842 462.282 19.5397 458.304 Q22.5952 454.325 28.2289 454.325 Q32.5894 454.325 35.1675 456.362 Q37.7456 458.399 38.3822 462.346 Q39.4007 457.604 42.6472 454.994 Q45.8619 452.352 50.6998 452.352 Q57.0655 452.352 60.5348 456.68 Q64.0042 461.009 64.0042 468.998 L64.0042 486.058 L16.4842 486.058 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip912)" style="stroke:#090621; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.715,1423.18 345.605,1342.53 338.299,1331.45 311.239,1239.72 304.275,1147.99 299.052,1056.26 292.088,964.535 285.124,872.806 278.16,781.076 272.937,689.347 265.972,597.618 259.008,505.889 252.044,414.16 245.08,322.431 239.857,230.702 232.893,138.973 225.929,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#1f0c47; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="331.716,47.2441 338.68,138.973 345.605,230.186 345.645,230.702 350.941,322.431 358.003,414.16 365.064,505.889 372.126,597.618 379.187,689.347 384.483,781.076 391.545,872.806 398.606,964.535 407.396,1056.26 415.584,1147.99 436.704,1239.72 471.83,1278.77 556.579,1331.45 598.055,1363.92 667.279,1423.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#3b0964; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="901.944,1423.18 850.505,1377.88 788.712,1331.45 724.28,1293.11 645.537,1239.72 598.055,1189.57 565.364,1147.99 535.903,1056.26 520.698,964.535 506.109,872.806 495.971,781.076 489.553,689.347 480.694,597.618 472.417,505.889 471.83,499.382 465.27,414.16 458.208,322.431 452.912,230.702 445.85,138.973 438.789,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#550f6d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="556.07,47.2441 564.085,138.973 572.099,230.702 578.11,322.431 589.871,414.16 598.055,504.857 598.136,505.889 605.298,597.618 616.762,689.347 622.349,781.076 638.767,872.806 660.364,964.535 693.661,1056.26 724.28,1113.22 749.472,1147.99 850.505,1218.45 880.799,1239.72 976.73,1306.63 1009.15,1331.45 1102.96,1422.15 1104.48,1423.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#6e186e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.74,1423.18 1355.41,1415.12 1246.72,1331.45 1229.18,1320.76 1102.96,1253.66 1084.76,1239.72 976.73,1156.44 964.697,1147.99 883.397,1056.26 850.505,1016.38 810.477,964.535 770.999,872.806 738.909,781.076 731.626,689.347 724.28,657.513 714.086,597.618 706.924,505.889 699.763,414.16 689.02,322.431 683.649,230.702 676.487,138.973 669.325,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#88216a; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="788.607,47.2441 797.165,138.973 809.658,230.702 816.387,322.431 835.267,414.16 844.699,505.889 850.505,546.959 856.866,597.618 875.188,689.347 883.361,781.076 916.204,872.806 971.914,964.535 976.73,969.796 1075.15,1056.26 1102.96,1085.92 1186.81,1147.99 1229.18,1170.67 1341.68,1239.72 1355.41,1248.09 1464.27,1331.45 1481.63,1348.23 1559.14,1423.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#a12b61; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1800.62,1423.18 1734.08,1378.52 1672.94,1331.45 1607.86,1291.77 1539.88,1239.72 1481.63,1194.81 1420.89,1147.99 1355.41,1078.16 1329.61,1056.26 1229.18,995.436 1170.67,964.535 1102.96,922.59 1061.09,872.806 1018,781.076 1001.58,689.347 983.534,597.618 976.73,566.79 963.283,505.889 953.185,414.16 936.639,322.431 930.966,230.702 918.228,138.973 910.667,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#ba3655; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1020.02,47.2441 1026.73,138.973 1039.3,230.702 1044.44,322.431 1059.41,414.16 1075.48,505.889 1101.02,597.618 1102.96,605.445 1144.84,689.347 1189.73,781.076 1229.18,816.376 1285.35,872.806 1355.41,922.693 1409.67,964.535 1481.63,1019.27 1530.27,1056.26 1607.86,1138.99 1618.46,1147.99 1734.08,1218.96 1772.87,1239.72 1860.31,1286.2 1919.4,1331.45 1986.53,1396.36 2016.98,1423.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#d04544; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,1321.27 1995.84,1239.72 1986.53,1233.19 1876.02,1147.99 1860.31,1131.24 1759.03,1056.26 1734.08,1043.09 1607.86,966.647 1605.08,964.535 1486.09,872.806 1481.63,869.092 1387.9,781.076 1355.41,717.633 1340.92,689.347 1293.27,597.618 1238.25,505.889 1229.18,474.953 1208.41,414.16 1182.28,322.431 1173.33,230.702 1143.89,138.973 1132.27,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#e35832; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1280.18,47.2441 1298.13,138.973 1329.81,230.702 1336.99,322.431 1355.41,403.127 1357.54,414.16 1391.63,505.889 1449.48,597.618 1481.63,653.862 1505.6,689.347 1571.35,781.076 1607.86,811.481 1700.34,872.806 1734.08,893.513 1860.31,961.07 1864.8,964.535 1985.4,1056.26 1986.53,1057.47 2079.73,1147.99 2112.76,1171.18 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#f0701e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,995.946 2067.42,964.535 1986.53,907.74 1941.21,872.806 1860.31,805.425 1818.19,781.076 1734.08,724.603 1694.61,689.347 1608.43,597.618 1607.86,597.1 1538.27,505.889 1481.63,416.062 1480.63,414.16 1458.01,322.431 1447.32,230.702 1423,138.973 1403.75,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#f98c09; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1515.28,47.2441 1537.37,138.973 1566.86,230.702 1588.55,322.431 1607.86,370.518 1630.55,414.16 1710.54,505.889 1734.08,526.887 1848.38,597.618 1860.31,607.162 1931.61,689.347 1986.53,742.457 2030.38,781.076 2112.76,843.559 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#fbaa0e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,664.893 2035.73,597.618 1986.53,557.67 1935,505.889 1860.31,423.06 1847.29,414.16 1764.28,322.431 1734.08,260.722 1723.38,230.702 1670.15,138.973 1634.33,47.2441 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#f8c931; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1799.26,47.2441 1860.31,125.534 1868.77,138.973 1920.04,230.702 1957.02,322.431 1986.53,369.044 2019.67,414.16 2112.76,500.307 "/>
-<polyline clip-path="url(#clip912)" style="stroke:#f1e864; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,289.305 2082.36,230.702 2017.57,138.973 1986.53,93.1516 1957.56,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,1423.18 219.38,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="850.505,1423.18 850.505,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1481.63,1423.18 1481.63,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2112.76,1423.18 2112.76,47.2441 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 2112.76,1423.18 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 219.38,1404.28 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="850.505,1423.18 850.505,1404.28 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1481.63,1423.18 1481.63,1404.28 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,1423.18 2112.76,1404.28 "/>
+<path clip-path="url(#clip730)" d="M196.765 1454.1 Q193.153 1454.1 191.325 1457.66 Q189.519 1461.2 189.519 1468.33 Q189.519 1475.44 191.325 1479.01 Q193.153 1482.55 196.765 1482.55 Q200.399 1482.55 202.204 1479.01 Q204.033 1475.44 204.033 1468.33 Q204.033 1461.2 202.204 1457.66 Q200.399 1454.1 196.765 1454.1 M196.765 1450.39 Q202.575 1450.39 205.63 1455 Q208.709 1459.58 208.709 1468.33 Q208.709 1477.06 205.63 1481.67 Q202.575 1486.25 196.765 1486.25 Q190.954 1486.25 187.876 1481.67 Q184.82 1477.06 184.82 1468.33 Q184.82 1459.58 187.876 1455 Q190.954 1450.39 196.765 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M216.926 1479.7 L221.811 1479.7 L221.811 1485.58 L216.926 1485.58 L216.926 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M241.996 1454.1 Q238.385 1454.1 236.556 1457.66 Q234.75 1461.2 234.75 1468.33 Q234.75 1475.44 236.556 1479.01 Q238.385 1482.55 241.996 1482.55 Q245.63 1482.55 247.436 1479.01 Q249.264 1475.44 249.264 1468.33 Q249.264 1461.2 247.436 1457.66 Q245.63 1454.1 241.996 1454.1 M241.996 1450.39 Q247.806 1450.39 250.861 1455 Q253.94 1459.58 253.94 1468.33 Q253.94 1477.06 250.861 1481.67 Q247.806 1486.25 241.996 1486.25 Q236.186 1486.25 233.107 1481.67 Q230.051 1477.06 230.051 1468.33 Q230.051 1459.58 233.107 1455 Q236.186 1450.39 241.996 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M828.387 1454.1 Q824.776 1454.1 822.948 1457.66 Q821.142 1461.2 821.142 1468.33 Q821.142 1475.44 822.948 1479.01 Q824.776 1482.55 828.387 1482.55 Q832.022 1482.55 833.827 1479.01 Q835.656 1475.44 835.656 1468.33 Q835.656 1461.2 833.827 1457.66 Q832.022 1454.1 828.387 1454.1 M828.387 1450.39 Q834.198 1450.39 837.253 1455 Q840.332 1459.58 840.332 1468.33 Q840.332 1477.06 837.253 1481.67 Q834.198 1486.25 828.387 1486.25 Q822.577 1486.25 819.499 1481.67 Q816.443 1477.06 816.443 1468.33 Q816.443 1459.58 819.499 1455 Q822.577 1450.39 828.387 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M848.549 1479.7 L853.434 1479.7 L853.434 1485.58 L848.549 1485.58 L848.549 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M863.665 1451.02 L882.021 1451.02 L882.021 1454.96 L867.947 1454.96 L867.947 1463.43 Q868.966 1463.08 869.984 1462.92 Q871.003 1462.73 872.021 1462.73 Q877.809 1462.73 881.188 1465.9 Q884.568 1469.08 884.568 1474.49 Q884.568 1480.07 881.096 1483.17 Q877.623 1486.25 871.304 1486.25 Q869.128 1486.25 866.859 1485.88 Q864.614 1485.51 862.207 1484.77 L862.207 1480.07 Q864.29 1481.2 866.512 1481.76 Q868.734 1482.32 871.211 1482.32 Q875.216 1482.32 877.554 1480.21 Q879.892 1478.1 879.892 1474.49 Q879.892 1470.88 877.554 1468.77 Q875.216 1466.67 871.211 1466.67 Q869.336 1466.67 867.461 1467.08 Q865.609 1467.5 863.665 1468.38 L863.665 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1448.78 1481.64 L1456.42 1481.64 L1456.42 1455.28 L1448.11 1456.95 L1448.11 1452.69 L1456.38 1451.02 L1461.05 1451.02 L1461.05 1481.64 L1468.69 1481.64 L1468.69 1485.58 L1448.78 1485.58 L1448.78 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1478.14 1479.7 L1483.02 1479.7 L1483.02 1485.58 L1478.14 1485.58 L1478.14 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1503.2 1454.1 Q1499.59 1454.1 1497.76 1457.66 Q1495.96 1461.2 1495.96 1468.33 Q1495.96 1475.44 1497.76 1479.01 Q1499.59 1482.55 1503.2 1482.55 Q1506.84 1482.55 1508.64 1479.01 Q1510.47 1475.44 1510.47 1468.33 Q1510.47 1461.2 1508.64 1457.66 Q1506.84 1454.1 1503.2 1454.1 M1503.2 1450.39 Q1509.01 1450.39 1512.07 1455 Q1515.15 1459.58 1515.15 1468.33 Q1515.15 1477.06 1512.07 1481.67 Q1509.01 1486.25 1503.2 1486.25 Q1497.39 1486.25 1494.32 1481.67 Q1491.26 1477.06 1491.26 1468.33 Q1491.26 1459.58 1494.32 1455 Q1497.39 1450.39 1503.2 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2080.41 1481.64 L2088.05 1481.64 L2088.05 1455.28 L2079.74 1456.95 L2079.74 1452.69 L2088 1451.02 L2092.67 1451.02 L2092.67 1481.64 L2100.31 1481.64 L2100.31 1485.58 L2080.41 1485.58 L2080.41 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2109.76 1479.7 L2114.64 1479.7 L2114.64 1485.58 L2109.76 1485.58 L2109.76 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2124.87 1451.02 L2143.23 1451.02 L2143.23 1454.96 L2129.16 1454.96 L2129.16 1463.43 Q2130.17 1463.08 2131.19 1462.92 Q2132.21 1462.73 2133.23 1462.73 Q2139.02 1462.73 2142.4 1465.9 Q2145.78 1469.08 2145.78 1474.49 Q2145.78 1480.07 2142.3 1483.17 Q2138.83 1486.25 2132.51 1486.25 Q2130.34 1486.25 2128.07 1485.88 Q2125.82 1485.51 2123.42 1484.77 L2123.42 1480.07 Q2125.5 1481.2 2127.72 1481.76 Q2129.94 1482.32 2132.42 1482.32 Q2136.42 1482.32 2138.76 1480.21 Q2141.1 1478.1 2141.1 1474.49 Q2141.1 1470.88 2138.76 1468.77 Q2136.42 1466.67 2132.42 1466.67 Q2130.55 1466.67 2128.67 1467.08 Q2126.82 1467.5 2124.87 1468.38 L2124.87 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M911.822 1522.08 L911.822 1528.35 Q908.161 1526.6 904.915 1525.74 Q901.668 1524.88 898.645 1524.88 Q893.393 1524.88 890.528 1526.92 Q887.696 1528.96 887.696 1532.71 Q887.696 1535.87 889.574 1537.49 Q891.483 1539.08 896.767 1540.07 L900.65 1540.86 Q907.843 1542.23 911.249 1545.7 Q914.686 1549.14 914.686 1554.93 Q914.686 1561.84 910.039 1565.4 Q905.424 1568.97 896.48 1568.97 Q893.107 1568.97 889.287 1568.2 Q885.5 1567.44 881.425 1565.94 L881.425 1559.32 Q885.34 1561.52 889.096 1562.63 Q892.852 1563.75 896.48 1563.75 Q901.987 1563.75 904.979 1561.58 Q907.97 1559.42 907.97 1555.41 Q907.97 1551.91 905.806 1549.93 Q903.674 1547.96 898.772 1546.97 L894.857 1546.21 Q887.664 1544.78 884.449 1541.72 Q881.234 1538.67 881.234 1533.22 Q881.234 1526.92 885.659 1523.29 Q890.115 1519.66 897.913 1519.66 Q901.255 1519.66 904.724 1520.27 Q908.193 1520.87 911.822 1522.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M950.111 1533.76 L950.111 1539.24 Q947.629 1537.87 945.114 1537.2 Q942.632 1536.5 940.085 1536.5 Q934.388 1536.5 931.237 1540.13 Q928.086 1543.73 928.086 1550.25 Q928.086 1556.78 931.237 1560.4 Q934.388 1564 940.085 1564 Q942.632 1564 945.114 1563.33 Q947.629 1562.63 950.111 1561.26 L950.111 1566.68 Q947.661 1567.82 945.019 1568.39 Q942.409 1568.97 939.449 1568.97 Q931.396 1568.97 926.654 1563.91 Q921.911 1558.85 921.911 1550.25 Q921.911 1541.53 926.686 1536.53 Q931.492 1531.54 939.831 1531.54 Q942.536 1531.54 945.114 1532.11 Q947.692 1532.65 950.111 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M976.497 1550.12 Q969.4 1550.12 966.662 1551.75 Q963.925 1553.37 963.925 1557.29 Q963.925 1560.4 965.962 1562.25 Q968.031 1564.07 971.564 1564.07 Q976.434 1564.07 979.362 1560.63 Q982.322 1557.16 982.322 1551.43 L982.322 1550.12 L976.497 1550.12 M988.178 1547.71 L988.178 1568.04 L982.322 1568.04 L982.322 1562.63 Q980.317 1565.88 977.325 1567.44 Q974.333 1568.97 970.004 1568.97 Q964.53 1568.97 961.283 1565.91 Q958.069 1562.82 958.069 1557.67 Q958.069 1551.65 962.079 1548.6 Q966.121 1545.54 974.11 1545.54 L982.322 1545.54 L982.322 1544.97 Q982.322 1540.93 979.648 1538.73 Q977.007 1536.5 972.2 1536.5 Q969.145 1536.5 966.249 1537.23 Q963.352 1537.97 960.679 1539.43 L960.679 1534.02 Q963.893 1532.78 966.917 1532.17 Q969.941 1531.54 972.805 1531.54 Q980.54 1531.54 984.359 1535.55 Q988.178 1539.56 988.178 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1000.24 1518.52 L1006.1 1518.52 L1006.1 1568.04 L1000.24 1568.04 L1000.24 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1048.84 1548.76 L1048.84 1551.62 L1021.92 1551.62 Q1022.3 1557.67 1025.55 1560.85 Q1028.82 1564 1034.65 1564 Q1038.02 1564 1041.17 1563.17 Q1044.36 1562.35 1047.47 1560.69 L1047.47 1566.23 Q1044.32 1567.57 1041.01 1568.27 Q1037.7 1568.97 1034.3 1568.97 Q1025.77 1568.97 1020.77 1564 Q1015.81 1559.04 1015.81 1550.57 Q1015.81 1541.82 1020.52 1536.69 Q1025.26 1531.54 1033.28 1531.54 Q1040.47 1531.54 1044.64 1536.18 Q1048.84 1540.8 1048.84 1548.76 M1042.99 1547.04 Q1042.92 1542.23 1040.28 1539.37 Q1037.67 1536.5 1033.34 1536.5 Q1028.44 1536.5 1025.48 1539.27 Q1022.55 1542.04 1022.11 1547.07 L1042.99 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1092.99 1536.5 Q1088.28 1536.5 1085.54 1540.19 Q1082.8 1543.85 1082.8 1550.25 Q1082.8 1556.65 1085.51 1560.34 Q1088.25 1564 1092.99 1564 Q1097.67 1564 1100.41 1560.31 Q1103.14 1556.62 1103.14 1550.25 Q1103.14 1543.92 1100.41 1540.23 Q1097.67 1536.5 1092.99 1536.5 M1092.99 1531.54 Q1100.63 1531.54 1104.99 1536.5 Q1109.35 1541.47 1109.35 1550.25 Q1109.35 1559 1104.99 1564 Q1100.63 1568.97 1092.99 1568.97 Q1085.32 1568.97 1080.96 1564 Q1076.63 1559 1076.63 1550.25 Q1076.63 1541.47 1080.96 1536.5 Q1085.32 1531.54 1092.99 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1137.1 1518.52 L1137.1 1523.39 L1131.5 1523.39 Q1128.35 1523.39 1127.11 1524.66 Q1125.9 1525.93 1125.9 1529.24 L1125.9 1532.4 L1135.54 1532.4 L1135.54 1536.95 L1125.9 1536.95 L1125.9 1568.04 L1120.01 1568.04 L1120.01 1536.95 L1114.41 1536.95 L1114.41 1532.4 L1120.01 1532.4 L1120.01 1529.91 Q1120.01 1523.96 1122.78 1521.26 Q1125.55 1518.52 1131.57 1518.52 L1137.1 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1180.77 1518.52 L1180.77 1523.39 L1175.17 1523.39 Q1172.02 1523.39 1170.78 1524.66 Q1169.57 1525.93 1169.57 1529.24 L1169.57 1532.4 L1179.21 1532.4 L1179.21 1536.95 L1169.57 1536.95 L1169.57 1568.04 L1163.68 1568.04 L1163.68 1536.95 L1158.08 1536.95 L1158.08 1532.4 L1163.68 1532.4 L1163.68 1529.91 Q1163.68 1523.96 1166.45 1521.26 Q1169.22 1518.52 1175.23 1518.52 L1180.77 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1201.88 1550.12 Q1194.78 1550.12 1192.04 1551.75 Q1189.3 1553.37 1189.3 1557.29 Q1189.3 1560.4 1191.34 1562.25 Q1193.41 1564.07 1196.94 1564.07 Q1201.81 1564.07 1204.74 1560.63 Q1207.7 1557.16 1207.7 1551.43 L1207.7 1550.12 L1201.88 1550.12 M1213.56 1547.71 L1213.56 1568.04 L1207.7 1568.04 L1207.7 1562.63 Q1205.69 1565.88 1202.7 1567.44 Q1199.71 1568.97 1195.38 1568.97 Q1189.91 1568.97 1186.66 1565.91 Q1183.45 1562.82 1183.45 1557.67 Q1183.45 1551.65 1187.46 1548.6 Q1191.5 1545.54 1199.49 1545.54 L1207.7 1545.54 L1207.7 1544.97 Q1207.7 1540.93 1205.03 1538.73 Q1202.38 1536.5 1197.58 1536.5 Q1194.52 1536.5 1191.63 1537.23 Q1188.73 1537.97 1186.06 1539.43 L1186.06 1534.02 Q1189.27 1532.78 1192.29 1532.17 Q1195.32 1531.54 1198.18 1531.54 Q1205.92 1531.54 1209.74 1535.55 Q1213.56 1539.56 1213.56 1547.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1251.27 1533.76 L1251.27 1539.24 Q1248.79 1537.87 1246.28 1537.2 Q1243.79 1536.5 1241.25 1536.5 Q1235.55 1536.5 1232.4 1540.13 Q1229.25 1543.73 1229.25 1550.25 Q1229.25 1556.78 1232.4 1560.4 Q1235.55 1564 1241.25 1564 Q1243.79 1564 1246.28 1563.33 Q1248.79 1562.63 1251.27 1561.26 L1251.27 1566.68 Q1248.82 1567.82 1246.18 1568.39 Q1243.57 1568.97 1240.61 1568.97 Q1232.56 1568.97 1227.82 1563.91 Q1223.07 1558.85 1223.07 1550.25 Q1223.07 1541.53 1227.85 1536.53 Q1232.65 1531.54 1240.99 1531.54 Q1243.7 1531.54 1246.28 1532.11 Q1248.85 1532.65 1251.27 1533.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1267.25 1522.27 L1267.25 1532.4 L1279.31 1532.4 L1279.31 1536.95 L1267.25 1536.95 L1267.25 1556.3 Q1267.25 1560.66 1268.43 1561.9 Q1269.64 1563.14 1273.3 1563.14 L1279.31 1563.14 L1279.31 1568.04 L1273.3 1568.04 Q1266.52 1568.04 1263.94 1565.53 Q1261.36 1562.98 1261.36 1556.3 L1261.36 1536.95 L1257.07 1536.95 L1257.07 1532.4 L1261.36 1532.4 L1261.36 1522.27 L1267.25 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1300.83 1536.5 Q1296.12 1536.5 1293.38 1540.19 Q1290.64 1543.85 1290.64 1550.25 Q1290.64 1556.65 1293.35 1560.34 Q1296.09 1564 1300.83 1564 Q1305.51 1564 1308.25 1560.31 Q1310.98 1556.62 1310.98 1550.25 Q1310.98 1543.92 1308.25 1540.23 Q1305.51 1536.5 1300.83 1536.5 M1300.83 1531.54 Q1308.47 1531.54 1312.83 1536.5 Q1317.19 1541.47 1317.19 1550.25 Q1317.19 1559 1312.83 1564 Q1308.47 1568.97 1300.83 1568.97 Q1293.16 1568.97 1288.8 1564 Q1284.47 1559 1284.47 1550.25 Q1284.47 1541.47 1288.8 1536.5 Q1293.16 1531.54 1300.83 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1347.55 1537.87 Q1346.57 1537.3 1345.39 1537.04 Q1344.24 1536.76 1342.84 1536.76 Q1337.88 1536.76 1335.2 1540 Q1332.56 1543.22 1332.56 1549.27 L1332.56 1568.04 L1326.67 1568.04 L1326.67 1532.4 L1332.56 1532.4 L1332.56 1537.93 Q1334.41 1534.69 1337.37 1533.13 Q1340.33 1531.54 1344.56 1531.54 Q1345.17 1531.54 1345.9 1531.63 Q1346.63 1531.7 1347.52 1531.85 L1347.55 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1368.53 1571.35 Q1366.05 1577.72 1363.69 1579.66 Q1361.34 1581.6 1357.39 1581.6 L1352.71 1581.6 L1352.71 1576.7 L1356.15 1576.7 Q1358.57 1576.7 1359.9 1575.56 Q1361.24 1574.41 1362.86 1570.14 L1363.91 1567.47 L1349.5 1532.4 L1355.7 1532.4 L1366.84 1560.28 L1377.98 1532.4 L1384.19 1532.4 L1368.53 1571.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M1429.13 1526.86 L1420.41 1550.51 L1437.88 1550.51 L1429.13 1526.86 M1425.5 1520.52 L1432.79 1520.52 L1450.9 1568.04 L1444.22 1568.04 L1439.89 1555.85 L1418.47 1555.85 L1414.14 1568.04 L1407.36 1568.04 L1425.5 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,1423.18 2112.76,1423.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,964.535 2112.76,964.535 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,505.889 2112.76,505.889 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="219.38,47.2441 2112.76,47.2441 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 219.38,47.2441 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,1423.18 236.152,1423.18 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,964.535 236.152,964.535 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,505.889 236.152,505.889 "/>
+<polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="219.38,47.2441 236.152,47.2441 "/>
+<path clip-path="url(#clip730)" d="M126.205 1408.98 Q122.593 1408.98 120.765 1412.54 Q118.959 1416.08 118.959 1423.21 Q118.959 1430.32 120.765 1433.89 Q122.593 1437.43 126.205 1437.43 Q129.839 1437.43 131.644 1433.89 Q133.473 1430.32 133.473 1423.21 Q133.473 1416.08 131.644 1412.54 Q129.839 1408.98 126.205 1408.98 M126.205 1405.27 Q132.015 1405.27 135.07 1409.88 Q138.149 1414.46 138.149 1423.21 Q138.149 1431.94 135.07 1436.55 Q132.015 1441.13 126.205 1441.13 Q120.394 1441.13 117.316 1436.55 Q114.26 1431.94 114.26 1423.21 Q114.26 1414.46 117.316 1409.88 Q120.394 1405.27 126.205 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M146.366 1434.58 L151.251 1434.58 L151.251 1440.46 L146.366 1440.46 L146.366 1434.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M171.436 1408.98 Q167.825 1408.98 165.996 1412.54 Q164.19 1416.08 164.19 1423.21 Q164.19 1430.32 165.996 1433.89 Q167.825 1437.43 171.436 1437.43 Q175.07 1437.43 176.876 1433.89 Q178.704 1430.32 178.704 1423.21 Q178.704 1416.08 176.876 1412.54 Q175.07 1408.98 171.436 1408.98 M171.436 1405.27 Q177.246 1405.27 180.301 1409.88 Q183.38 1414.46 183.38 1423.21 Q183.38 1431.94 180.301 1436.55 Q177.246 1441.13 171.436 1441.13 Q165.626 1441.13 162.547 1436.55 Q159.491 1431.94 159.491 1423.21 Q159.491 1414.46 162.547 1409.88 Q165.626 1405.27 171.436 1405.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M127.2 950.333 Q123.589 950.333 121.76 953.898 Q119.955 957.44 119.955 964.569 Q119.955 971.676 121.76 975.241 Q123.589 978.782 127.2 978.782 Q130.834 978.782 132.64 975.241 Q134.468 971.676 134.468 964.569 Q134.468 957.44 132.64 953.898 Q130.834 950.333 127.2 950.333 M127.2 946.63 Q133.01 946.63 136.066 951.236 Q139.144 955.819 139.144 964.569 Q139.144 973.296 136.066 977.903 Q133.01 982.486 127.2 982.486 Q121.39 982.486 118.311 977.903 Q115.256 973.296 115.256 964.569 Q115.256 955.819 118.311 951.236 Q121.39 946.63 127.2 946.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M147.362 975.935 L152.246 975.935 L152.246 981.815 L147.362 981.815 L147.362 975.935 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M162.477 947.255 L180.834 947.255 L180.834 951.19 L166.76 951.19 L166.76 959.662 Q167.778 959.315 168.797 959.153 Q169.815 958.967 170.834 958.967 Q176.621 958.967 180.001 962.139 Q183.38 965.31 183.38 970.727 Q183.38 976.305 179.908 979.407 Q176.436 982.486 170.116 982.486 Q167.94 982.486 165.672 982.116 Q163.427 981.745 161.019 981.004 L161.019 976.305 Q163.102 977.44 165.325 977.995 Q167.547 978.551 170.024 978.551 Q174.028 978.551 176.366 976.444 Q178.704 974.338 178.704 970.727 Q178.704 967.116 176.366 965.009 Q174.028 962.903 170.024 962.903 Q168.149 962.903 166.274 963.319 Q164.422 963.736 162.477 964.616 L162.477 947.255 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M117.015 519.234 L124.654 519.234 L124.654 492.869 L116.343 494.535 L116.343 490.276 L124.607 488.609 L129.283 488.609 L129.283 519.234 L136.922 519.234 L136.922 523.169 L117.015 523.169 L117.015 519.234 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M146.366 517.29 L151.251 517.29 L151.251 523.169 L146.366 523.169 L146.366 517.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M171.436 491.688 Q167.825 491.688 165.996 495.253 Q164.19 498.794 164.19 505.924 Q164.19 513.031 165.996 516.595 Q167.825 520.137 171.436 520.137 Q175.07 520.137 176.876 516.595 Q178.704 513.031 178.704 505.924 Q178.704 498.794 176.876 495.253 Q175.07 491.688 171.436 491.688 M171.436 487.984 Q177.246 487.984 180.301 492.591 Q183.38 497.174 183.38 505.924 Q183.38 514.651 180.301 519.257 Q177.246 523.841 171.436 523.841 Q165.626 523.841 162.547 519.257 Q159.491 514.651 159.491 505.924 Q159.491 497.174 162.547 492.591 Q165.626 487.984 171.436 487.984 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M118.01 60.5889 L125.649 60.5889 L125.649 34.2233 L117.339 35.89 L117.339 31.6308 L125.603 29.9641 L130.279 29.9641 L130.279 60.5889 L137.917 60.5889 L137.917 64.5241 L118.01 64.5241 L118.01 60.5889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M147.362 58.6445 L152.246 58.6445 L152.246 64.5241 L147.362 64.5241 L147.362 58.6445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M162.477 29.9641 L180.834 29.9641 L180.834 33.8993 L166.76 33.8993 L166.76 42.3714 Q167.778 42.0242 168.797 41.8622 Q169.815 41.677 170.834 41.677 Q176.621 41.677 180.001 44.8483 Q183.38 48.0196 183.38 53.4362 Q183.38 59.0149 179.908 62.1167 Q176.436 65.1954 170.116 65.1954 Q167.94 65.1954 165.672 64.825 Q163.427 64.4547 161.019 63.7139 L161.019 59.0149 Q163.102 60.1491 165.325 60.7047 Q167.547 61.2602 170.024 61.2602 Q174.028 61.2602 176.366 59.1538 Q178.704 57.0473 178.704 53.4362 Q178.704 49.8251 176.366 47.7186 Q174.028 45.6122 170.024 45.6122 Q168.149 45.6122 166.274 46.0288 Q164.422 46.4455 162.477 47.3251 L162.477 29.9641 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M18.0438 987.485 L24.314 987.485 Q22.5634 991.145 21.704 994.392 Q20.8447 997.638 20.8447 1000.66 Q20.8447 1005.91 22.8817 1008.78 Q24.9187 1011.61 28.6745 1011.61 Q31.8255 1011.61 33.4488 1009.73 Q35.0402 1007.82 36.0269 1002.54 L36.8226 998.657 Q38.1912 991.463 41.6605 988.058 Q45.098 984.62 50.8908 984.62 Q57.7976 984.62 61.3624 989.267 Q64.9272 993.882 64.9272 1002.83 Q64.9272 1006.2 64.1633 1010.02 Q63.3994 1013.81 61.9035 1017.88 L55.2831 1017.88 Q57.4793 1013.97 58.5933 1010.21 Q59.7073 1006.45 59.7073 1002.83 Q59.7073 997.32 57.543 994.328 Q55.3786 991.336 51.3682 991.336 Q47.8671 991.336 45.8937 993.501 Q43.9204 995.633 42.9337 1000.53 L42.1698 1004.45 Q40.7375 1011.64 37.682 1014.86 Q34.6264 1018.07 29.1837 1018.07 Q22.8817 1018.07 19.2532 1013.65 Q15.6248 1009.19 15.6248 1001.39 Q15.6248 998.052 16.2295 994.583 Q16.8343 991.113 18.0438 987.485 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M29.7248 949.195 L35.1993 949.195 Q33.8307 951.678 33.1623 954.192 Q32.4621 956.675 32.4621 959.221 Q32.4621 964.918 36.0905 968.07 Q39.6872 971.221 46.212 971.221 Q52.7369 971.221 56.3653 968.07 Q59.9619 964.918 59.9619 959.221 Q59.9619 956.675 59.2935 954.192 Q58.5933 951.678 57.2247 949.195 L62.6355 949.195 Q63.7814 951.646 64.3543 954.288 Q64.9272 956.898 64.9272 959.858 Q64.9272 967.91 59.8664 972.653 Q54.8057 977.395 46.212 977.395 Q37.491 977.395 32.4939 972.621 Q27.4968 967.815 27.4968 959.476 Q27.4968 956.77 28.0697 954.192 Q28.6108 951.614 29.7248 949.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M46.0847 922.809 Q46.0847 929.907 47.7079 932.644 Q49.3312 935.382 53.2461 935.382 Q56.3653 935.382 58.2114 933.345 Q60.0256 931.276 60.0256 927.743 Q60.0256 922.873 56.5881 919.945 Q53.1188 916.985 47.3897 916.985 L46.0847 916.985 L46.0847 922.809 M43.6657 911.128 L64.0042 911.128 L64.0042 916.985 L58.5933 916.985 Q61.8398 918.99 63.3994 921.982 Q64.9272 924.974 64.9272 929.302 Q64.9272 934.777 61.8716 938.023 Q58.7843 941.238 53.6281 941.238 Q47.6125 941.238 44.5569 937.228 Q41.5014 933.185 41.5014 925.196 L41.5014 916.985 L40.9285 916.985 Q36.8862 916.985 34.6901 919.658 Q32.4621 922.3 32.4621 927.106 Q32.4621 930.162 33.1941 933.058 Q33.9262 935.955 35.3903 938.628 L29.9795 938.628 Q28.7381 935.413 28.1334 932.39 Q27.4968 929.366 27.4968 926.501 Q27.4968 918.767 31.5072 914.948 Q35.5176 911.128 43.6657 911.128 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M14.479 899.065 L14.479 893.209 L64.0042 893.209 L64.0042 899.065 L14.479 899.065 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M44.7161 850.463 L47.5806 850.463 L47.5806 877.39 Q53.6281 877.008 56.8109 873.762 Q59.9619 870.483 59.9619 864.659 Q59.9619 861.285 59.1344 858.134 Q58.3069 854.951 56.6518 851.832 L62.1899 851.832 Q63.5267 854.983 64.227 858.293 Q64.9272 861.603 64.9272 865.009 Q64.9272 873.539 59.9619 878.536 Q54.9967 883.501 46.5303 883.501 Q37.7774 883.501 32.6531 878.79 Q27.4968 874.048 27.4968 866.027 Q27.4968 858.834 32.1438 854.664 Q36.7589 850.463 44.7161 850.463 M42.9973 856.32 Q38.1912 856.383 35.3266 859.025 Q32.4621 861.635 32.4621 865.964 Q32.4621 870.865 35.2312 873.825 Q38.0002 876.753 43.0292 877.199 L42.9973 856.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M32.4621 806.317 Q32.4621 811.028 36.1542 813.765 Q39.8145 816.502 46.212 816.502 Q52.6095 816.502 56.3017 813.797 Q59.9619 811.059 59.9619 806.317 Q59.9619 801.638 56.2698 798.901 Q52.5777 796.164 46.212 796.164 Q39.8781 796.164 36.186 798.901 Q32.4621 801.638 32.4621 806.317 M27.4968 806.317 Q27.4968 798.678 32.4621 794.318 Q37.4273 789.957 46.212 789.957 Q54.9649 789.957 59.9619 794.318 Q64.9272 798.678 64.9272 806.317 Q64.9272 813.988 59.9619 818.348 Q54.9649 822.677 46.212 822.677 Q37.4273 822.677 32.4621 818.348 Q27.4968 813.988 27.4968 806.317 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M14.479 762.203 L19.3487 762.203 L19.3487 767.804 Q19.3487 770.955 20.6219 772.197 Q21.895 773.406 25.2052 773.406 L28.3562 773.406 L28.3562 763.762 L32.9077 763.762 L32.9077 773.406 L64.0042 773.406 L64.0042 779.294 L32.9077 779.294 L32.9077 784.896 L28.3562 784.896 L28.3562 779.294 L25.8736 779.294 Q19.9216 779.294 17.2162 776.525 Q14.479 773.756 14.479 767.741 L14.479 762.203 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M14.479 718.534 L19.3487 718.534 L19.3487 724.136 Q19.3487 727.287 20.6219 728.528 Q21.895 729.737 25.2052 729.737 L28.3562 729.737 L28.3562 720.093 L32.9077 720.093 L32.9077 729.737 L64.0042 729.737 L64.0042 735.626 L32.9077 735.626 L32.9077 741.228 L28.3562 741.228 L28.3562 735.626 L25.8736 735.626 Q19.9216 735.626 17.2162 732.857 Q14.479 730.088 14.479 724.072 L14.479 718.534 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M46.0847 697.431 Q46.0847 704.529 47.7079 707.267 Q49.3312 710.004 53.2461 710.004 Q56.3653 710.004 58.2114 707.967 Q60.0256 705.898 60.0256 702.365 Q60.0256 697.495 56.5881 694.567 Q53.1188 691.607 47.3897 691.607 L46.0847 691.607 L46.0847 697.431 M43.6657 685.75 L64.0042 685.75 L64.0042 691.607 L58.5933 691.607 Q61.8398 693.612 63.3994 696.604 Q64.9272 699.596 64.9272 703.925 Q64.9272 709.399 61.8716 712.646 Q58.7843 715.86 53.6281 715.86 Q47.6125 715.86 44.5569 711.85 Q41.5014 707.808 41.5014 699.819 L41.5014 691.607 L40.9285 691.607 Q36.8862 691.607 34.6901 694.28 Q32.4621 696.922 32.4621 701.728 Q32.4621 704.784 33.1941 707.68 Q33.9262 710.577 35.3903 713.25 L29.9795 713.25 Q28.7381 710.036 28.1334 707.012 Q27.4968 703.988 27.4968 701.124 Q27.4968 693.389 31.5072 689.57 Q35.5176 685.75 43.6657 685.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M29.7248 648.034 L35.1993 648.034 Q33.8307 650.516 33.1623 653.031 Q32.4621 655.513 32.4621 658.06 Q32.4621 663.757 36.0905 666.908 Q39.6872 670.059 46.212 670.059 Q52.7369 670.059 56.3653 666.908 Q59.9619 663.757 59.9619 658.06 Q59.9619 655.513 59.2935 653.031 Q58.5933 650.516 57.2247 648.034 L62.6355 648.034 Q63.7814 650.484 64.3543 653.126 Q64.9272 655.736 64.9272 658.696 Q64.9272 666.749 59.8664 671.491 Q54.8057 676.234 46.212 676.234 Q37.491 676.234 32.4939 671.459 Q27.4968 666.653 27.4968 658.314 Q27.4968 655.609 28.0697 653.031 Q28.6108 650.453 29.7248 648.034 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M18.2347 632.056 L28.3562 632.056 L28.3562 619.993 L32.9077 619.993 L32.9077 632.056 L52.2594 632.056 Q56.6199 632.056 57.8613 630.878 Q59.1026 629.669 59.1026 626.008 L59.1026 619.993 L64.0042 619.993 L64.0042 626.008 Q64.0042 632.788 61.4897 635.366 Q58.9434 637.944 52.2594 637.944 L32.9077 637.944 L32.9077 642.241 L28.3562 642.241 L28.3562 637.944 L18.2347 637.944 L18.2347 632.056 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M32.4621 598.477 Q32.4621 603.187 36.1542 605.924 Q39.8145 608.662 46.212 608.662 Q52.6095 608.662 56.3017 605.956 Q59.9619 603.219 59.9619 598.477 Q59.9619 593.798 56.2698 591.061 Q52.5777 588.323 46.212 588.323 Q39.8781 588.323 36.186 591.061 Q32.4621 593.798 32.4621 598.477 M27.4968 598.477 Q27.4968 590.838 32.4621 586.477 Q37.4273 582.117 46.212 582.117 Q54.9649 582.117 59.9619 586.477 Q64.9272 590.838 64.9272 598.477 Q64.9272 606.147 59.9619 610.508 Q54.9649 614.836 46.212 614.836 Q37.4273 614.836 32.4621 610.508 Q27.4968 606.147 27.4968 598.477 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M33.8307 551.752 Q33.2578 552.739 33.0032 553.917 Q32.7167 555.062 32.7167 556.463 Q32.7167 561.428 35.9632 564.102 Q39.1779 566.744 45.2253 566.744 L64.0042 566.744 L64.0042 572.632 L28.3562 572.632 L28.3562 566.744 L33.8944 566.744 Q30.6479 564.897 29.0883 561.937 Q27.4968 558.977 27.4968 554.744 Q27.4968 554.139 27.5923 553.407 Q27.656 552.675 27.8151 551.784 L33.8307 551.752 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M67.3143 530.777 Q73.68 533.26 75.6216 535.615 Q77.5631 537.971 77.5631 541.917 L77.5631 546.596 L72.6615 546.596 L72.6615 543.159 Q72.6615 540.74 71.5157 539.403 Q70.3699 538.066 66.1048 536.443 L63.4312 535.392 L28.3562 549.811 L28.3562 543.604 L56.238 532.464 L28.3562 521.324 L28.3562 515.118 L67.3143 530.777 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M41.3104 479.629 L58.7206 479.629 L58.7206 469.316 Q58.7206 464.128 56.5881 461.646 Q54.4238 459.131 49.9996 459.131 Q45.5436 459.131 43.4429 461.646 Q41.3104 464.128 41.3104 469.316 L41.3104 479.629 M21.7677 479.629 L36.0905 479.629 L36.0905 470.112 Q36.0905 465.401 34.34 463.11 Q32.5576 460.786 28.9291 460.786 Q25.3325 460.786 23.5501 463.11 Q21.7677 465.401 21.7677 470.112 L21.7677 479.629 M16.4842 486.058 L16.4842 469.635 Q16.4842 462.282 19.5397 458.304 Q22.5952 454.325 28.2289 454.325 Q32.5894 454.325 35.1675 456.362 Q37.7456 458.399 38.3822 462.346 Q39.4007 457.604 42.6472 454.994 Q45.8619 452.352 50.6998 452.352 Q57.0655 452.352 60.5348 456.68 Q64.0042 461.009 64.0042 468.998 L64.0042 486.058 L16.4842 486.058 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip732)" style="stroke:#090621; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.715,1423.18 345.605,1342.53 338.299,1331.45 311.239,1239.72 304.275,1147.99 299.052,1056.26 292.088,964.535 285.124,872.806 278.16,781.076 272.937,689.347 265.972,597.618 259.008,505.889 252.044,414.16 245.08,322.431 239.857,230.702 232.893,138.973 225.929,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#1f0c47; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="331.716,47.2441 338.68,138.973 345.605,230.186 345.645,230.702 350.941,322.431 358.003,414.16 365.064,505.889 372.126,597.618 379.187,689.347 384.483,781.076 391.545,872.806 398.606,964.535 407.396,1056.26 415.584,1147.99 436.704,1239.72 471.83,1278.77 556.579,1331.45 598.055,1363.92 667.279,1423.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#3b0964; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="901.944,1423.18 850.505,1377.88 788.712,1331.45 724.28,1293.11 645.537,1239.72 598.055,1189.57 565.364,1147.99 535.903,1056.26 520.698,964.535 506.109,872.806 495.971,781.076 489.553,689.347 480.694,597.618 472.417,505.889 471.83,499.382 465.27,414.16 458.208,322.431 452.912,230.702 445.85,138.973 438.789,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#550f6d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="556.07,47.2441 564.085,138.973 572.099,230.702 578.11,322.431 589.871,414.16 598.055,504.857 598.136,505.889 605.298,597.618 616.762,689.347 622.349,781.076 638.767,872.806 660.364,964.535 693.661,1056.26 724.28,1113.22 749.472,1147.99 850.505,1218.45 880.799,1239.72 976.73,1306.63 1009.15,1331.45 1102.96,1422.15 1104.48,1423.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#6e186e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1363.74,1423.18 1355.41,1415.12 1246.72,1331.45 1229.18,1320.76 1102.96,1253.66 1084.76,1239.72 976.73,1156.44 964.697,1147.99 883.397,1056.26 850.505,1016.38 810.477,964.535 770.999,872.806 738.909,781.076 731.626,689.347 724.28,657.513 714.086,597.618 706.924,505.889 699.763,414.16 689.02,322.431 683.649,230.702 676.487,138.973 669.325,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#88216a; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="788.607,47.2441 797.165,138.973 809.658,230.702 816.387,322.431 835.267,414.16 844.699,505.889 850.505,546.959 856.866,597.618 875.188,689.347 883.361,781.076 916.204,872.806 971.914,964.535 976.73,969.796 1075.15,1056.26 1102.96,1085.92 1186.81,1147.99 1229.18,1170.67 1341.68,1239.72 1355.41,1248.09 1464.27,1331.45 1481.63,1348.23 1559.14,1423.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#a12b61; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1800.62,1423.18 1734.08,1378.52 1672.94,1331.45 1607.86,1291.77 1539.88,1239.72 1481.63,1194.81 1420.89,1147.99 1355.41,1078.16 1329.61,1056.26 1229.18,995.436 1170.67,964.535 1102.96,922.59 1061.09,872.806 1018,781.076 1001.58,689.347 983.534,597.618 976.73,566.79 963.283,505.889 953.185,414.16 936.639,322.431 930.966,230.702 918.228,138.973 910.667,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#ba3655; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1020.02,47.2441 1026.73,138.973 1039.3,230.702 1044.44,322.431 1059.41,414.16 1075.48,505.889 1101.02,597.618 1102.96,605.445 1144.84,689.347 1189.73,781.076 1229.18,816.376 1285.35,872.806 1355.41,922.693 1409.67,964.535 1481.63,1019.27 1530.27,1056.26 1607.86,1138.99 1618.46,1147.99 1734.08,1218.96 1772.87,1239.72 1860.31,1286.2 1919.4,1331.45 1986.53,1396.36 2016.98,1423.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#d04544; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,1321.27 1995.84,1239.72 1986.53,1233.19 1876.02,1147.99 1860.31,1131.24 1759.03,1056.26 1734.08,1043.09 1607.86,966.647 1605.08,964.535 1486.09,872.806 1481.63,869.092 1387.9,781.076 1355.41,717.633 1340.92,689.347 1293.27,597.618 1238.25,505.889 1229.18,474.953 1208.41,414.16 1182.28,322.431 1173.33,230.702 1143.89,138.973 1132.27,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#e35832; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1280.18,47.2441 1298.13,138.973 1329.81,230.702 1336.99,322.431 1355.41,403.127 1357.54,414.16 1391.63,505.889 1449.48,597.618 1481.63,653.862 1505.6,689.347 1571.35,781.076 1607.86,811.481 1700.34,872.806 1734.08,893.513 1860.31,961.07 1864.8,964.535 1985.4,1056.26 1986.53,1057.47 2079.73,1147.99 2112.76,1171.18 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#f0701e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,995.946 2067.42,964.535 1986.53,907.74 1941.21,872.806 1860.31,805.425 1818.19,781.076 1734.08,724.603 1694.61,689.347 1608.43,597.618 1607.86,597.1 1538.27,505.889 1481.63,416.062 1480.63,414.16 1458.01,322.431 1447.32,230.702 1423,138.973 1403.75,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#f98c09; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1515.28,47.2441 1537.37,138.973 1566.86,230.702 1588.55,322.431 1607.86,370.518 1630.55,414.16 1710.54,505.889 1734.08,526.887 1848.38,597.618 1860.31,607.162 1931.61,689.347 1986.53,742.457 2030.38,781.076 2112.76,843.559 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#fbaa0e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,664.893 2035.73,597.618 1986.53,557.67 1935,505.889 1860.31,423.06 1847.29,414.16 1764.28,322.431 1734.08,260.722 1723.38,230.702 1670.15,138.973 1634.33,47.2441 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#f8c931; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1799.26,47.2441 1860.31,125.534 1868.77,138.973 1920.04,230.702 1957.02,322.431 1986.53,369.044 2019.67,414.16 2112.76,500.307 "/>
+<polyline clip-path="url(#clip732)" style="stroke:#f1e864; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2112.76,289.305 2082.36,230.702 2017.57,138.973 1986.53,93.1516 1957.56,47.2441 "/>
 <defs>
-  <clipPath id="clip913">
+  <clipPath id="clip733">
     <rect x="2160" y="47" width="73" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip913)" style="stroke:#0a0723; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1337.18 2232.76,1337.18 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#200c4a; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1251.19 2232.76,1251.19 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#3c0965; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1165.19 2232.76,1165.19 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#570f6d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1079.2 2232.76,1079.2 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#70196e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,993.2 2232.76,993.2 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#892269; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,907.204 2232.76,907.204 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#a32b61; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,821.208 2232.76,821.208 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#bb3754; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,735.212 2232.76,735.212 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#d04544; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,649.216 2232.76,649.216 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#e35832; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,563.22 2232.76,563.22 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#f0701e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,477.224 2232.76,477.224 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#f98c09; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,391.228 2232.76,391.228 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#fbaa0e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,305.232 2232.76,305.232 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#f8c931; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,219.236 2232.76,219.236 "/>
-<polyline clip-path="url(#clip913)" style="stroke:#f1e864; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,133.24 2232.76,133.24 "/>
-<path clip-path="url(#clip910)" d="M2274.38 1256.53 L2290.7 1256.53 L2290.7 1260.47 L2268.76 1260.47 L2268.76 1256.53 Q2271.42 1253.78 2276 1249.15 Q2280.61 1244.5 2281.79 1243.15 Q2284.03 1240.63 2284.91 1238.9 Q2285.82 1237.14 2285.82 1235.45 Q2285.82 1232.69 2283.87 1230.96 Q2281.95 1229.22 2278.85 1229.22 Q2276.65 1229.22 2274.2 1229.98 Q2271.77 1230.75 2268.99 1232.3 L2268.99 1227.58 Q2271.81 1226.44 2274.27 1225.86 Q2276.72 1225.28 2278.76 1225.28 Q2284.13 1225.28 2287.32 1227.97 Q2290.52 1230.65 2290.52 1235.15 Q2290.52 1237.27 2289.7 1239.2 Q2288.92 1241.09 2286.81 1243.69 Q2286.23 1244.36 2283.13 1247.58 Q2280.03 1250.77 2274.38 1256.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2300.56 1225.91 L2318.92 1225.91 L2318.92 1229.84 L2304.84 1229.84 L2304.84 1238.32 Q2305.86 1237.97 2306.88 1237.81 Q2307.9 1237.62 2308.92 1237.62 Q2314.7 1237.62 2318.08 1240.79 Q2321.46 1243.96 2321.46 1249.38 Q2321.46 1254.96 2317.99 1258.06 Q2314.52 1261.14 2308.2 1261.14 Q2306.02 1261.14 2303.76 1260.77 Q2301.51 1260.4 2299.1 1259.66 L2299.1 1254.96 Q2301.19 1256.09 2303.41 1256.65 Q2305.63 1257.21 2308.11 1257.21 Q2312.11 1257.21 2314.45 1255.1 Q2316.79 1252.99 2316.79 1249.38 Q2316.79 1245.77 2314.45 1243.66 Q2312.11 1241.56 2308.11 1241.56 Q2306.23 1241.56 2304.36 1241.97 Q2302.51 1242.39 2300.56 1243.27 L2300.56 1225.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2340.68 1228.99 Q2337.07 1228.99 2335.24 1232.55 Q2333.43 1236.09 2333.43 1243.22 Q2333.43 1250.33 2335.24 1253.9 Q2337.07 1257.44 2340.68 1257.44 Q2344.31 1257.44 2346.12 1253.9 Q2347.95 1250.33 2347.95 1243.22 Q2347.95 1236.09 2346.12 1232.55 Q2344.31 1228.99 2340.68 1228.99 M2340.68 1225.28 Q2346.49 1225.28 2349.54 1229.89 Q2352.62 1234.47 2352.62 1243.22 Q2352.62 1251.95 2349.54 1256.56 Q2346.49 1261.14 2340.68 1261.14 Q2334.87 1261.14 2331.79 1256.56 Q2328.73 1251.95 2328.73 1243.22 Q2328.73 1234.47 2331.79 1229.89 Q2334.87 1225.28 2340.68 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2370.84 1228.99 Q2367.23 1228.99 2365.4 1232.55 Q2363.59 1236.09 2363.59 1243.22 Q2363.59 1250.33 2365.4 1253.9 Q2367.23 1257.44 2370.84 1257.44 Q2374.47 1257.44 2376.28 1253.9 Q2378.11 1250.33 2378.11 1243.22 Q2378.11 1236.09 2376.28 1232.55 Q2374.47 1228.99 2370.84 1228.99 M2370.84 1225.28 Q2376.65 1225.28 2379.7 1229.89 Q2382.78 1234.47 2382.78 1243.22 Q2382.78 1251.95 2379.7 1256.56 Q2376.65 1261.14 2370.84 1261.14 Q2365.03 1261.14 2361.95 1256.56 Q2358.89 1251.95 2358.89 1243.22 Q2358.89 1234.47 2361.95 1229.89 Q2365.03 1225.28 2370.84 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2401 1228.99 Q2397.39 1228.99 2395.56 1232.55 Q2393.76 1236.09 2393.76 1243.22 Q2393.76 1250.33 2395.56 1253.9 Q2397.39 1257.44 2401 1257.44 Q2404.63 1257.44 2406.44 1253.9 Q2408.27 1250.33 2408.27 1243.22 Q2408.27 1236.09 2406.44 1232.55 Q2404.63 1228.99 2401 1228.99 M2401 1225.28 Q2406.81 1225.28 2409.87 1229.89 Q2412.94 1234.47 2412.94 1243.22 Q2412.94 1251.95 2409.87 1256.56 Q2406.81 1261.14 2401 1261.14 Q2395.19 1261.14 2392.11 1256.56 Q2389.06 1251.95 2389.06 1243.22 Q2389.06 1234.47 2392.11 1229.89 Q2395.19 1225.28 2401 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2431.16 1228.99 Q2427.55 1228.99 2425.72 1232.55 Q2423.92 1236.09 2423.92 1243.22 Q2423.92 1250.33 2425.72 1253.9 Q2427.55 1257.44 2431.16 1257.44 Q2434.8 1257.44 2436.6 1253.9 Q2438.43 1250.33 2438.43 1243.22 Q2438.43 1236.09 2436.6 1232.55 Q2434.8 1228.99 2431.16 1228.99 M2431.16 1225.28 Q2436.97 1225.28 2440.03 1229.89 Q2443.11 1234.47 2443.11 1243.22 Q2443.11 1251.95 2440.03 1256.56 Q2436.97 1261.14 2431.16 1261.14 Q2425.35 1261.14 2422.27 1256.56 Q2419.22 1251.95 2419.22 1243.22 Q2419.22 1234.47 2422.27 1229.89 Q2425.35 1225.28 2431.16 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2461.32 1228.99 Q2457.71 1228.99 2455.88 1232.55 Q2454.08 1236.09 2454.08 1243.22 Q2454.08 1250.33 2455.88 1253.9 Q2457.71 1257.44 2461.32 1257.44 Q2464.96 1257.44 2466.76 1253.9 Q2468.59 1250.33 2468.59 1243.22 Q2468.59 1236.09 2466.76 1232.55 Q2464.96 1228.99 2461.32 1228.99 M2461.32 1225.28 Q2467.13 1225.28 2470.19 1229.89 Q2473.27 1234.47 2473.27 1243.22 Q2473.27 1251.95 2470.19 1256.56 Q2467.13 1261.14 2461.32 1261.14 Q2455.51 1261.14 2452.44 1256.56 Q2449.38 1251.95 2449.38 1243.22 Q2449.38 1234.47 2452.44 1229.89 Q2455.51 1225.28 2461.32 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2270.21 1049 L2288.57 1049 L2288.57 1052.93 L2274.5 1052.93 L2274.5 1061.4 Q2275.52 1061.06 2276.53 1060.89 Q2277.55 1060.71 2278.57 1060.71 Q2284.36 1060.71 2287.74 1063.88 Q2291.12 1067.05 2291.12 1072.47 Q2291.12 1078.05 2287.64 1081.15 Q2284.17 1084.23 2277.85 1084.23 Q2275.68 1084.23 2273.41 1083.86 Q2271.16 1083.49 2268.76 1082.75 L2268.76 1078.05 Q2270.84 1079.18 2273.06 1079.74 Q2275.28 1080.29 2277.76 1080.29 Q2281.77 1080.29 2284.1 1078.19 Q2286.44 1076.08 2286.44 1072.47 Q2286.44 1068.86 2284.1 1066.75 Q2281.77 1064.64 2277.76 1064.64 Q2275.89 1064.64 2274.01 1065.06 Q2272.16 1065.48 2270.21 1066.36 L2270.21 1049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2310.33 1052.07 Q2306.72 1052.07 2304.89 1055.64 Q2303.08 1059.18 2303.08 1066.31 Q2303.08 1073.42 2304.89 1076.98 Q2306.72 1080.52 2310.33 1080.52 Q2313.96 1080.52 2315.77 1076.98 Q2317.6 1073.42 2317.6 1066.31 Q2317.6 1059.18 2315.77 1055.64 Q2313.96 1052.07 2310.33 1052.07 M2310.33 1048.37 Q2316.14 1048.37 2319.2 1052.98 Q2322.27 1057.56 2322.27 1066.31 Q2322.27 1075.04 2319.2 1079.64 Q2316.14 1084.23 2310.33 1084.23 Q2304.52 1084.23 2301.44 1079.64 Q2298.39 1075.04 2298.39 1066.31 Q2298.39 1057.56 2301.44 1052.98 Q2304.52 1048.37 2310.33 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2340.49 1052.07 Q2336.88 1052.07 2335.05 1055.64 Q2333.25 1059.18 2333.25 1066.31 Q2333.25 1073.42 2335.05 1076.98 Q2336.88 1080.52 2340.49 1080.52 Q2344.13 1080.52 2345.93 1076.98 Q2347.76 1073.42 2347.76 1066.31 Q2347.76 1059.18 2345.93 1055.64 Q2344.13 1052.07 2340.49 1052.07 M2340.49 1048.37 Q2346.3 1048.37 2349.36 1052.98 Q2352.44 1057.56 2352.44 1066.31 Q2352.44 1075.04 2349.36 1079.64 Q2346.3 1084.23 2340.49 1084.23 Q2334.68 1084.23 2331.6 1079.64 Q2328.55 1075.04 2328.55 1066.31 Q2328.55 1057.56 2331.6 1052.98 Q2334.68 1048.37 2340.49 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2370.65 1052.07 Q2367.04 1052.07 2365.21 1055.64 Q2363.41 1059.18 2363.41 1066.31 Q2363.41 1073.42 2365.21 1076.98 Q2367.04 1080.52 2370.65 1080.52 Q2374.29 1080.52 2376.09 1076.98 Q2377.92 1073.42 2377.92 1066.31 Q2377.92 1059.18 2376.09 1055.64 Q2374.29 1052.07 2370.65 1052.07 M2370.65 1048.37 Q2376.46 1048.37 2379.52 1052.98 Q2382.6 1057.56 2382.6 1066.31 Q2382.6 1075.04 2379.52 1079.64 Q2376.46 1084.23 2370.65 1084.23 Q2364.84 1084.23 2361.76 1079.64 Q2358.71 1075.04 2358.71 1066.31 Q2358.71 1057.56 2361.76 1052.98 Q2364.84 1048.37 2370.65 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2400.82 1052.07 Q2397.2 1052.07 2395.38 1055.64 Q2393.57 1059.18 2393.57 1066.31 Q2393.57 1073.42 2395.38 1076.98 Q2397.2 1080.52 2400.82 1080.52 Q2404.45 1080.52 2406.26 1076.98 Q2408.08 1073.42 2408.08 1066.31 Q2408.08 1059.18 2406.26 1055.64 Q2404.45 1052.07 2400.82 1052.07 M2400.82 1048.37 Q2406.63 1048.37 2409.68 1052.98 Q2412.76 1057.56 2412.76 1066.31 Q2412.76 1075.04 2409.68 1079.64 Q2406.63 1084.23 2400.82 1084.23 Q2395.01 1084.23 2391.93 1079.64 Q2388.87 1075.04 2388.87 1066.31 Q2388.87 1057.56 2391.93 1052.98 Q2395.01 1048.37 2400.82 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2430.98 1052.07 Q2427.37 1052.07 2425.54 1055.64 Q2423.73 1059.18 2423.73 1066.31 Q2423.73 1073.42 2425.54 1076.98 Q2427.37 1080.52 2430.98 1080.52 Q2434.61 1080.52 2436.42 1076.98 Q2438.25 1073.42 2438.25 1066.31 Q2438.25 1059.18 2436.42 1055.64 Q2434.61 1052.07 2430.98 1052.07 M2430.98 1048.37 Q2436.79 1048.37 2439.84 1052.98 Q2442.92 1057.56 2442.92 1066.31 Q2442.92 1075.04 2439.84 1079.64 Q2436.79 1084.23 2430.98 1084.23 Q2425.17 1084.23 2422.09 1079.64 Q2419.03 1075.04 2419.03 1066.31 Q2419.03 1057.56 2422.09 1052.98 Q2425.17 1048.37 2430.98 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2461.14 1052.07 Q2457.53 1052.07 2455.7 1055.64 Q2453.89 1059.18 2453.89 1066.31 Q2453.89 1073.42 2455.7 1076.98 Q2457.53 1080.52 2461.14 1080.52 Q2464.77 1080.52 2466.58 1076.98 Q2468.41 1073.42 2468.41 1066.31 Q2468.41 1059.18 2466.58 1055.64 Q2464.77 1052.07 2461.14 1052.07 M2461.14 1048.37 Q2466.95 1048.37 2470 1052.98 Q2473.08 1057.56 2473.08 1066.31 Q2473.08 1075.04 2470 1079.64 Q2466.95 1084.23 2461.14 1084.23 Q2455.33 1084.23 2452.25 1079.64 Q2449.19 1075.04 2449.19 1066.31 Q2449.19 1057.56 2452.25 1052.98 Q2455.33 1048.37 2461.14 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2268.76 872.082 L2290.98 872.082 L2290.98 874.073 L2278.43 906.642 L2273.55 906.642 L2285.35 876.017 L2268.76 876.017 L2268.76 872.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2300.14 872.082 L2318.5 872.082 L2318.5 876.017 L2304.43 876.017 L2304.43 884.489 Q2305.45 884.142 2306.46 883.98 Q2307.48 883.795 2308.5 883.795 Q2314.29 883.795 2317.67 886.966 Q2321.05 890.137 2321.05 895.554 Q2321.05 901.133 2317.58 904.234 Q2314.1 907.313 2307.78 907.313 Q2305.61 907.313 2303.34 906.943 Q2301.09 906.572 2298.69 905.832 L2298.69 901.133 Q2300.77 902.267 2302.99 902.822 Q2305.21 903.378 2307.69 903.378 Q2311.7 903.378 2314.03 901.272 Q2316.37 899.165 2316.37 895.554 Q2316.37 891.943 2314.03 889.836 Q2311.7 887.73 2307.69 887.73 Q2305.82 887.73 2303.94 888.147 Q2302.09 888.563 2300.14 889.443 L2300.14 872.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2340.26 875.161 Q2336.65 875.161 2334.82 878.725 Q2333.01 882.267 2333.01 889.397 Q2333.01 896.503 2334.82 900.068 Q2336.65 903.609 2340.26 903.609 Q2343.89 903.609 2345.7 900.068 Q2347.53 896.503 2347.53 889.397 Q2347.53 882.267 2345.7 878.725 Q2343.89 875.161 2340.26 875.161 M2340.26 871.457 Q2346.07 871.457 2349.13 876.063 Q2352.2 880.647 2352.2 889.397 Q2352.2 898.123 2349.13 902.73 Q2346.07 907.313 2340.26 907.313 Q2334.45 907.313 2331.37 902.73 Q2328.32 898.123 2328.32 889.397 Q2328.32 880.647 2331.37 876.063 Q2334.45 871.457 2340.26 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2370.42 875.161 Q2366.81 875.161 2364.98 878.725 Q2363.18 882.267 2363.18 889.397 Q2363.18 896.503 2364.98 900.068 Q2366.81 903.609 2370.42 903.609 Q2374.06 903.609 2375.86 900.068 Q2377.69 896.503 2377.69 889.397 Q2377.69 882.267 2375.86 878.725 Q2374.06 875.161 2370.42 875.161 M2370.42 871.457 Q2376.23 871.457 2379.29 876.063 Q2382.37 880.647 2382.37 889.397 Q2382.37 898.123 2379.29 902.73 Q2376.23 907.313 2370.42 907.313 Q2364.61 907.313 2361.53 902.73 Q2358.48 898.123 2358.48 889.397 Q2358.48 880.647 2361.53 876.063 Q2364.61 871.457 2370.42 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2400.58 875.161 Q2396.97 875.161 2395.14 878.725 Q2393.34 882.267 2393.34 889.397 Q2393.34 896.503 2395.14 900.068 Q2396.97 903.609 2400.58 903.609 Q2404.22 903.609 2406.02 900.068 Q2407.85 896.503 2407.85 889.397 Q2407.85 882.267 2406.02 878.725 Q2404.22 875.161 2400.58 875.161 M2400.58 871.457 Q2406.39 871.457 2409.45 876.063 Q2412.53 880.647 2412.53 889.397 Q2412.53 898.123 2409.45 902.73 Q2406.39 907.313 2400.58 907.313 Q2394.77 907.313 2391.7 902.73 Q2388.64 898.123 2388.64 889.397 Q2388.64 880.647 2391.7 876.063 Q2394.77 871.457 2400.58 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2430.75 875.161 Q2427.13 875.161 2425.31 878.725 Q2423.5 882.267 2423.5 889.397 Q2423.5 896.503 2425.31 900.068 Q2427.13 903.609 2430.75 903.609 Q2434.38 903.609 2436.19 900.068 Q2438.01 896.503 2438.01 889.397 Q2438.01 882.267 2436.19 878.725 Q2434.38 875.161 2430.75 875.161 M2430.75 871.457 Q2436.56 871.457 2439.61 876.063 Q2442.69 880.647 2442.69 889.397 Q2442.69 898.123 2439.61 902.73 Q2436.56 907.313 2430.75 907.313 Q2424.94 907.313 2421.86 902.73 Q2418.8 898.123 2418.8 889.397 Q2418.8 880.647 2421.86 876.063 Q2424.94 871.457 2430.75 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2460.91 875.161 Q2457.3 875.161 2455.47 878.725 Q2453.66 882.267 2453.66 889.397 Q2453.66 896.503 2455.47 900.068 Q2457.3 903.609 2460.91 903.609 Q2464.54 903.609 2466.35 900.068 Q2468.18 896.503 2468.18 889.397 Q2468.18 882.267 2466.35 878.725 Q2464.54 875.161 2460.91 875.161 M2460.91 871.457 Q2466.72 871.457 2469.77 876.063 Q2472.85 880.647 2472.85 889.397 Q2472.85 898.123 2469.77 902.73 Q2466.72 907.313 2460.91 907.313 Q2455.1 907.313 2452.02 902.73 Q2448.96 898.123 2448.96 889.397 Q2448.96 880.647 2452.02 876.063 Q2455.1 871.457 2460.91 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2269.43 725.793 L2277.07 725.793 L2277.07 699.427 L2268.76 701.094 L2268.76 696.835 L2277.02 695.168 L2281.7 695.168 L2281.7 725.793 L2289.33 725.793 L2289.33 729.728 L2269.43 729.728 L2269.43 725.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2308.78 698.247 Q2305.17 698.247 2303.34 701.812 Q2301.53 705.353 2301.53 712.483 Q2301.53 719.589 2303.34 723.154 Q2305.17 726.696 2308.78 726.696 Q2312.41 726.696 2314.22 723.154 Q2316.05 719.589 2316.05 712.483 Q2316.05 705.353 2314.22 701.812 Q2312.41 698.247 2308.78 698.247 M2308.78 694.543 Q2314.59 694.543 2317.64 699.15 Q2320.72 703.733 2320.72 712.483 Q2320.72 721.21 2317.64 725.816 Q2314.59 730.399 2308.78 730.399 Q2302.97 730.399 2299.89 725.816 Q2296.83 721.21 2296.83 712.483 Q2296.83 703.733 2299.89 699.15 Q2302.97 694.543 2308.78 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2338.94 698.247 Q2335.33 698.247 2333.5 701.812 Q2331.7 705.353 2331.7 712.483 Q2331.7 719.589 2333.5 723.154 Q2335.33 726.696 2338.94 726.696 Q2342.57 726.696 2344.38 723.154 Q2346.21 719.589 2346.21 712.483 Q2346.21 705.353 2344.38 701.812 Q2342.57 698.247 2338.94 698.247 M2338.94 694.543 Q2344.75 694.543 2347.81 699.15 Q2350.89 703.733 2350.89 712.483 Q2350.89 721.21 2347.81 725.816 Q2344.75 730.399 2338.94 730.399 Q2333.13 730.399 2330.05 725.816 Q2327 721.21 2327 712.483 Q2327 703.733 2330.05 699.15 Q2333.13 694.543 2338.94 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2369.1 698.247 Q2365.49 698.247 2363.66 701.812 Q2361.86 705.353 2361.86 712.483 Q2361.86 719.589 2363.66 723.154 Q2365.49 726.696 2369.1 726.696 Q2372.74 726.696 2374.54 723.154 Q2376.37 719.589 2376.37 712.483 Q2376.37 705.353 2374.54 701.812 Q2372.74 698.247 2369.1 698.247 M2369.1 694.543 Q2374.91 694.543 2377.97 699.15 Q2381.05 703.733 2381.05 712.483 Q2381.05 721.21 2377.97 725.816 Q2374.91 730.399 2369.1 730.399 Q2363.29 730.399 2360.21 725.816 Q2357.16 721.21 2357.16 712.483 Q2357.16 703.733 2360.21 699.15 Q2363.29 694.543 2369.1 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2399.26 698.247 Q2395.65 698.247 2393.82 701.812 Q2392.02 705.353 2392.02 712.483 Q2392.02 719.589 2393.82 723.154 Q2395.65 726.696 2399.26 726.696 Q2402.9 726.696 2404.7 723.154 Q2406.53 719.589 2406.53 712.483 Q2406.53 705.353 2404.7 701.812 Q2402.9 698.247 2399.26 698.247 M2399.26 694.543 Q2405.07 694.543 2408.13 699.15 Q2411.21 703.733 2411.21 712.483 Q2411.21 721.21 2408.13 725.816 Q2405.07 730.399 2399.26 730.399 Q2393.45 730.399 2390.38 725.816 Q2387.32 721.21 2387.32 712.483 Q2387.32 703.733 2390.38 699.15 Q2393.45 694.543 2399.26 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2429.43 698.247 Q2425.82 698.247 2423.99 701.812 Q2422.18 705.353 2422.18 712.483 Q2422.18 719.589 2423.99 723.154 Q2425.82 726.696 2429.43 726.696 Q2433.06 726.696 2434.87 723.154 Q2436.69 719.589 2436.69 712.483 Q2436.69 705.353 2434.87 701.812 Q2433.06 698.247 2429.43 698.247 M2429.43 694.543 Q2435.24 694.543 2438.29 699.15 Q2441.37 703.733 2441.37 712.483 Q2441.37 721.21 2438.29 725.816 Q2435.24 730.399 2429.43 730.399 Q2423.62 730.399 2420.54 725.816 Q2417.48 721.21 2417.48 712.483 Q2417.48 703.733 2420.54 699.15 Q2423.62 694.543 2429.43 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2459.59 698.247 Q2455.98 698.247 2454.15 701.812 Q2452.34 705.353 2452.34 712.483 Q2452.34 719.589 2454.15 723.154 Q2455.98 726.696 2459.59 726.696 Q2463.22 726.696 2465.03 723.154 Q2466.86 719.589 2466.86 712.483 Q2466.86 705.353 2465.03 701.812 Q2463.22 698.247 2459.59 698.247 M2459.59 694.543 Q2465.4 694.543 2468.45 699.15 Q2471.53 703.733 2471.53 712.483 Q2471.53 721.21 2468.45 725.816 Q2465.4 730.399 2459.59 730.399 Q2453.78 730.399 2450.7 725.816 Q2447.64 721.21 2447.64 712.483 Q2447.64 703.733 2450.7 699.15 Q2453.78 694.543 2459.59 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2489.75 698.247 Q2486.14 698.247 2484.31 701.812 Q2482.5 705.353 2482.5 712.483 Q2482.5 719.589 2484.31 723.154 Q2486.14 726.696 2489.75 726.696 Q2493.38 726.696 2495.19 723.154 Q2497.02 719.589 2497.02 712.483 Q2497.02 705.353 2495.19 701.812 Q2493.38 698.247 2489.75 698.247 M2489.75 694.543 Q2495.56 694.543 2498.62 699.15 Q2501.69 703.733 2501.69 712.483 Q2501.69 721.21 2498.62 725.816 Q2495.56 730.399 2489.75 730.399 Q2483.94 730.399 2480.86 725.816 Q2477.81 721.21 2477.81 712.483 Q2477.81 703.733 2480.86 699.15 Q2483.94 694.543 2489.75 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2269.43 548.879 L2277.07 548.879 L2277.07 522.514 L2268.76 524.18 L2268.76 519.921 L2277.02 518.254 L2281.7 518.254 L2281.7 548.879 L2289.33 548.879 L2289.33 552.814 L2269.43 552.814 L2269.43 548.879 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2302.81 548.879 L2319.13 548.879 L2319.13 552.814 L2297.18 552.814 L2297.18 548.879 Q2299.84 546.125 2304.43 541.495 Q2309.03 536.842 2310.21 535.5 Q2312.46 532.977 2313.34 531.241 Q2314.24 529.481 2314.24 527.791 Q2314.24 525.037 2312.3 523.301 Q2310.38 521.565 2307.27 521.565 Q2305.08 521.565 2302.62 522.329 Q2300.19 523.092 2297.41 524.643 L2297.41 519.921 Q2300.24 518.787 2302.69 518.208 Q2305.14 517.629 2307.18 517.629 Q2312.55 517.629 2315.75 520.315 Q2318.94 523 2318.94 527.491 Q2318.94 529.62 2318.13 531.541 Q2317.34 533.44 2315.24 536.032 Q2314.66 536.703 2311.56 539.921 Q2308.45 543.115 2302.81 548.879 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2328.99 518.254 L2347.34 518.254 L2347.34 522.19 L2333.27 522.19 L2333.27 530.662 Q2334.29 530.315 2335.31 530.153 Q2336.32 529.967 2337.34 529.967 Q2343.13 529.967 2346.51 533.139 Q2349.89 536.31 2349.89 541.727 Q2349.89 547.305 2346.42 550.407 Q2342.95 553.486 2336.63 553.486 Q2334.45 553.486 2332.18 553.115 Q2329.94 552.745 2327.53 552.004 L2327.53 547.305 Q2329.61 548.439 2331.83 548.995 Q2334.06 549.551 2336.53 549.551 Q2340.54 549.551 2342.88 547.444 Q2345.21 545.338 2345.21 541.727 Q2345.21 538.115 2342.88 536.009 Q2340.54 533.903 2336.53 533.903 Q2334.66 533.903 2332.78 534.319 Q2330.93 534.736 2328.99 535.615 L2328.99 518.254 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2369.1 521.333 Q2365.49 521.333 2363.66 524.898 Q2361.86 528.44 2361.86 535.569 Q2361.86 542.676 2363.66 546.24 Q2365.49 549.782 2369.1 549.782 Q2372.74 549.782 2374.54 546.24 Q2376.37 542.676 2376.37 535.569 Q2376.37 528.44 2374.54 524.898 Q2372.74 521.333 2369.1 521.333 M2369.1 517.629 Q2374.91 517.629 2377.97 522.236 Q2381.05 526.819 2381.05 535.569 Q2381.05 544.296 2377.97 548.902 Q2374.91 553.486 2369.1 553.486 Q2363.29 553.486 2360.21 548.902 Q2357.16 544.296 2357.16 535.569 Q2357.16 526.819 2360.21 522.236 Q2363.29 517.629 2369.1 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2399.26 521.333 Q2395.65 521.333 2393.82 524.898 Q2392.02 528.44 2392.02 535.569 Q2392.02 542.676 2393.82 546.24 Q2395.65 549.782 2399.26 549.782 Q2402.9 549.782 2404.7 546.24 Q2406.53 542.676 2406.53 535.569 Q2406.53 528.44 2404.7 524.898 Q2402.9 521.333 2399.26 521.333 M2399.26 517.629 Q2405.07 517.629 2408.13 522.236 Q2411.21 526.819 2411.21 535.569 Q2411.21 544.296 2408.13 548.902 Q2405.07 553.486 2399.26 553.486 Q2393.45 553.486 2390.38 548.902 Q2387.32 544.296 2387.32 535.569 Q2387.32 526.819 2390.38 522.236 Q2393.45 517.629 2399.26 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2429.43 521.333 Q2425.82 521.333 2423.99 524.898 Q2422.18 528.44 2422.18 535.569 Q2422.18 542.676 2423.99 546.24 Q2425.82 549.782 2429.43 549.782 Q2433.06 549.782 2434.87 546.24 Q2436.69 542.676 2436.69 535.569 Q2436.69 528.44 2434.87 524.898 Q2433.06 521.333 2429.43 521.333 M2429.43 517.629 Q2435.24 517.629 2438.29 522.236 Q2441.37 526.819 2441.37 535.569 Q2441.37 544.296 2438.29 548.902 Q2435.24 553.486 2429.43 553.486 Q2423.62 553.486 2420.54 548.902 Q2417.48 544.296 2417.48 535.569 Q2417.48 526.819 2420.54 522.236 Q2423.62 517.629 2429.43 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2459.59 521.333 Q2455.98 521.333 2454.15 524.898 Q2452.34 528.44 2452.34 535.569 Q2452.34 542.676 2454.15 546.24 Q2455.98 549.782 2459.59 549.782 Q2463.22 549.782 2465.03 546.24 Q2466.86 542.676 2466.86 535.569 Q2466.86 528.44 2465.03 524.898 Q2463.22 521.333 2459.59 521.333 M2459.59 517.629 Q2465.4 517.629 2468.45 522.236 Q2471.53 526.819 2471.53 535.569 Q2471.53 544.296 2468.45 548.902 Q2465.4 553.486 2459.59 553.486 Q2453.78 553.486 2450.7 548.902 Q2447.64 544.296 2447.64 535.569 Q2447.64 526.819 2450.7 522.236 Q2453.78 517.629 2459.59 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2489.75 521.333 Q2486.14 521.333 2484.31 524.898 Q2482.5 528.44 2482.5 535.569 Q2482.5 542.676 2484.31 546.24 Q2486.14 549.782 2489.75 549.782 Q2493.38 549.782 2495.19 546.24 Q2497.02 542.676 2497.02 535.569 Q2497.02 528.44 2495.19 524.898 Q2493.38 521.333 2489.75 521.333 M2489.75 517.629 Q2495.56 517.629 2498.62 522.236 Q2501.69 526.819 2501.69 535.569 Q2501.69 544.296 2498.62 548.902 Q2495.56 553.486 2489.75 553.486 Q2483.94 553.486 2480.86 548.902 Q2477.81 544.296 2477.81 535.569 Q2477.81 526.819 2480.86 522.236 Q2483.94 517.629 2489.75 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2269.43 371.966 L2277.07 371.966 L2277.07 345.6 L2268.76 347.267 L2268.76 343.007 L2277.02 341.341 L2281.7 341.341 L2281.7 371.966 L2289.33 371.966 L2289.33 375.901 L2269.43 375.901 L2269.43 371.966 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2298.83 341.341 L2317.18 341.341 L2317.18 345.276 L2303.11 345.276 L2303.11 353.748 Q2304.13 353.401 2305.14 353.239 Q2306.16 353.054 2307.18 353.054 Q2312.97 353.054 2316.35 356.225 Q2319.73 359.396 2319.73 364.813 Q2319.73 370.392 2316.26 373.493 Q2312.78 376.572 2306.46 376.572 Q2304.29 376.572 2302.02 376.202 Q2299.77 375.831 2297.37 375.091 L2297.37 370.392 Q2299.45 371.526 2301.67 372.081 Q2303.89 372.637 2306.37 372.637 Q2310.38 372.637 2312.71 370.53 Q2315.05 368.424 2315.05 364.813 Q2315.05 361.202 2312.71 359.095 Q2310.38 356.989 2306.37 356.989 Q2304.5 356.989 2302.62 357.405 Q2300.77 357.822 2298.83 358.702 L2298.83 341.341 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2338.94 344.419 Q2335.33 344.419 2333.5 347.984 Q2331.7 351.526 2331.7 358.655 Q2331.7 365.762 2333.5 369.327 Q2335.33 372.868 2338.94 372.868 Q2342.57 372.868 2344.38 369.327 Q2346.21 365.762 2346.21 358.655 Q2346.21 351.526 2344.38 347.984 Q2342.57 344.419 2338.94 344.419 M2338.94 340.716 Q2344.75 340.716 2347.81 345.322 Q2350.89 349.906 2350.89 358.655 Q2350.89 367.382 2347.81 371.989 Q2344.75 376.572 2338.94 376.572 Q2333.13 376.572 2330.05 371.989 Q2327 367.382 2327 358.655 Q2327 349.906 2330.05 345.322 Q2333.13 340.716 2338.94 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2369.1 344.419 Q2365.49 344.419 2363.66 347.984 Q2361.86 351.526 2361.86 358.655 Q2361.86 365.762 2363.66 369.327 Q2365.49 372.868 2369.1 372.868 Q2372.74 372.868 2374.54 369.327 Q2376.37 365.762 2376.37 358.655 Q2376.37 351.526 2374.54 347.984 Q2372.74 344.419 2369.1 344.419 M2369.1 340.716 Q2374.91 340.716 2377.97 345.322 Q2381.05 349.906 2381.05 358.655 Q2381.05 367.382 2377.97 371.989 Q2374.91 376.572 2369.1 376.572 Q2363.29 376.572 2360.21 371.989 Q2357.16 367.382 2357.16 358.655 Q2357.16 349.906 2360.21 345.322 Q2363.29 340.716 2369.1 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2399.26 344.419 Q2395.65 344.419 2393.82 347.984 Q2392.02 351.526 2392.02 358.655 Q2392.02 365.762 2393.82 369.327 Q2395.65 372.868 2399.26 372.868 Q2402.9 372.868 2404.7 369.327 Q2406.53 365.762 2406.53 358.655 Q2406.53 351.526 2404.7 347.984 Q2402.9 344.419 2399.26 344.419 M2399.26 340.716 Q2405.07 340.716 2408.13 345.322 Q2411.21 349.906 2411.21 358.655 Q2411.21 367.382 2408.13 371.989 Q2405.07 376.572 2399.26 376.572 Q2393.45 376.572 2390.38 371.989 Q2387.32 367.382 2387.32 358.655 Q2387.32 349.906 2390.38 345.322 Q2393.45 340.716 2399.26 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2429.43 344.419 Q2425.82 344.419 2423.99 347.984 Q2422.18 351.526 2422.18 358.655 Q2422.18 365.762 2423.99 369.327 Q2425.82 372.868 2429.43 372.868 Q2433.06 372.868 2434.87 369.327 Q2436.69 365.762 2436.69 358.655 Q2436.69 351.526 2434.87 347.984 Q2433.06 344.419 2429.43 344.419 M2429.43 340.716 Q2435.24 340.716 2438.29 345.322 Q2441.37 349.906 2441.37 358.655 Q2441.37 367.382 2438.29 371.989 Q2435.24 376.572 2429.43 376.572 Q2423.62 376.572 2420.54 371.989 Q2417.48 367.382 2417.48 358.655 Q2417.48 349.906 2420.54 345.322 Q2423.62 340.716 2429.43 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2459.59 344.419 Q2455.98 344.419 2454.15 347.984 Q2452.34 351.526 2452.34 358.655 Q2452.34 365.762 2454.15 369.327 Q2455.98 372.868 2459.59 372.868 Q2463.22 372.868 2465.03 369.327 Q2466.86 365.762 2466.86 358.655 Q2466.86 351.526 2465.03 347.984 Q2463.22 344.419 2459.59 344.419 M2459.59 340.716 Q2465.4 340.716 2468.45 345.322 Q2471.53 349.906 2471.53 358.655 Q2471.53 367.382 2468.45 371.989 Q2465.4 376.572 2459.59 376.572 Q2453.78 376.572 2450.7 371.989 Q2447.64 367.382 2447.64 358.655 Q2447.64 349.906 2450.7 345.322 Q2453.78 340.716 2459.59 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2489.75 344.419 Q2486.14 344.419 2484.31 347.984 Q2482.5 351.526 2482.5 358.655 Q2482.5 365.762 2484.31 369.327 Q2486.14 372.868 2489.75 372.868 Q2493.38 372.868 2495.19 369.327 Q2497.02 365.762 2497.02 358.655 Q2497.02 351.526 2495.19 347.984 Q2493.38 344.419 2489.75 344.419 M2489.75 340.716 Q2495.56 340.716 2498.62 345.322 Q2501.69 349.906 2501.69 358.655 Q2501.69 367.382 2498.62 371.989 Q2495.56 376.572 2489.75 376.572 Q2483.94 376.572 2480.86 371.989 Q2477.81 367.382 2477.81 358.655 Q2477.81 349.906 2480.86 345.322 Q2483.94 340.716 2489.75 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2269.43 195.052 L2277.07 195.052 L2277.07 168.686 L2268.76 170.353 L2268.76 166.094 L2277.02 164.427 L2281.7 164.427 L2281.7 195.052 L2289.33 195.052 L2289.33 198.987 L2269.43 198.987 L2269.43 195.052 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2297.6 164.427 L2319.82 164.427 L2319.82 166.418 L2307.27 198.987 L2302.39 198.987 L2314.2 168.362 L2297.6 168.362 L2297.6 164.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2328.99 164.427 L2347.34 164.427 L2347.34 168.362 L2333.27 168.362 L2333.27 176.834 Q2334.29 176.487 2335.31 176.325 Q2336.32 176.14 2337.34 176.14 Q2343.13 176.14 2346.51 179.311 Q2349.89 182.483 2349.89 187.899 Q2349.89 193.478 2346.42 196.58 Q2342.95 199.658 2336.63 199.658 Q2334.45 199.658 2332.18 199.288 Q2329.94 198.918 2327.53 198.177 L2327.53 193.478 Q2329.61 194.612 2331.83 195.168 Q2334.06 195.723 2336.53 195.723 Q2340.54 195.723 2342.88 193.617 Q2345.21 191.51 2345.21 187.899 Q2345.21 184.288 2342.88 182.182 Q2340.54 180.075 2336.53 180.075 Q2334.66 180.075 2332.78 180.492 Q2330.93 180.908 2328.99 181.788 L2328.99 164.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2369.1 167.506 Q2365.49 167.506 2363.66 171.071 Q2361.86 174.612 2361.86 181.742 Q2361.86 188.848 2363.66 192.413 Q2365.49 195.955 2369.1 195.955 Q2372.74 195.955 2374.54 192.413 Q2376.37 188.848 2376.37 181.742 Q2376.37 174.612 2374.54 171.071 Q2372.74 167.506 2369.1 167.506 M2369.1 163.802 Q2374.91 163.802 2377.97 168.409 Q2381.05 172.992 2381.05 181.742 Q2381.05 190.469 2377.97 195.075 Q2374.91 199.658 2369.1 199.658 Q2363.29 199.658 2360.21 195.075 Q2357.16 190.469 2357.16 181.742 Q2357.16 172.992 2360.21 168.409 Q2363.29 163.802 2369.1 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2399.26 167.506 Q2395.65 167.506 2393.82 171.071 Q2392.02 174.612 2392.02 181.742 Q2392.02 188.848 2393.82 192.413 Q2395.65 195.955 2399.26 195.955 Q2402.9 195.955 2404.7 192.413 Q2406.53 188.848 2406.53 181.742 Q2406.53 174.612 2404.7 171.071 Q2402.9 167.506 2399.26 167.506 M2399.26 163.802 Q2405.07 163.802 2408.13 168.409 Q2411.21 172.992 2411.21 181.742 Q2411.21 190.469 2408.13 195.075 Q2405.07 199.658 2399.26 199.658 Q2393.45 199.658 2390.38 195.075 Q2387.32 190.469 2387.32 181.742 Q2387.32 172.992 2390.38 168.409 Q2393.45 163.802 2399.26 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2429.43 167.506 Q2425.82 167.506 2423.99 171.071 Q2422.18 174.612 2422.18 181.742 Q2422.18 188.848 2423.99 192.413 Q2425.82 195.955 2429.43 195.955 Q2433.06 195.955 2434.87 192.413 Q2436.69 188.848 2436.69 181.742 Q2436.69 174.612 2434.87 171.071 Q2433.06 167.506 2429.43 167.506 M2429.43 163.802 Q2435.24 163.802 2438.29 168.409 Q2441.37 172.992 2441.37 181.742 Q2441.37 190.469 2438.29 195.075 Q2435.24 199.658 2429.43 199.658 Q2423.62 199.658 2420.54 195.075 Q2417.48 190.469 2417.48 181.742 Q2417.48 172.992 2420.54 168.409 Q2423.62 163.802 2429.43 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2459.59 167.506 Q2455.98 167.506 2454.15 171.071 Q2452.34 174.612 2452.34 181.742 Q2452.34 188.848 2454.15 192.413 Q2455.98 195.955 2459.59 195.955 Q2463.22 195.955 2465.03 192.413 Q2466.86 188.848 2466.86 181.742 Q2466.86 174.612 2465.03 171.071 Q2463.22 167.506 2459.59 167.506 M2459.59 163.802 Q2465.4 163.802 2468.45 168.409 Q2471.53 172.992 2471.53 181.742 Q2471.53 190.469 2468.45 195.075 Q2465.4 199.658 2459.59 199.658 Q2453.78 199.658 2450.7 195.075 Q2447.64 190.469 2447.64 181.742 Q2447.64 172.992 2450.7 168.409 Q2453.78 163.802 2459.59 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2489.75 167.506 Q2486.14 167.506 2484.31 171.071 Q2482.5 174.612 2482.5 181.742 Q2482.5 188.848 2484.31 192.413 Q2486.14 195.955 2489.75 195.955 Q2493.38 195.955 2495.19 192.413 Q2497.02 188.848 2497.02 181.742 Q2497.02 174.612 2495.19 171.071 Q2493.38 167.506 2489.75 167.506 M2489.75 163.802 Q2495.56 163.802 2498.62 168.409 Q2501.69 172.992 2501.69 181.742 Q2501.69 190.469 2498.62 195.075 Q2495.56 199.658 2489.75 199.658 Q2483.94 199.658 2480.86 195.075 Q2477.81 190.469 2477.81 181.742 Q2477.81 172.992 2480.86 168.409 Q2483.94 163.802 2489.75 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2232.76,1423.18 2232.76,1246.82 2256.76,1246.82 2232.76,1246.82 2232.76,1069.9 2256.76,1069.9 2232.76,1069.9 2232.76,892.991 2256.76,892.991 2232.76,892.991 2232.76,716.077 2256.76,716.077 2232.76,716.077 2232.76,539.163 2256.76,539.163 2232.76,539.163 2232.76,362.25 2256.76,362.25 2232.76,362.25 2232.76,185.336 2256.76,185.336 2232.76,185.336 2232.76,47.2441 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#0a0723; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1337.18 2232.76,1337.18 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#200c4a; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1251.19 2232.76,1251.19 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#3c0965; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1165.19 2232.76,1165.19 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#570f6d; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,1079.2 2232.76,1079.2 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#70196e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,993.2 2232.76,993.2 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#892269; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,907.204 2232.76,907.204 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#a32b61; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,821.208 2232.76,821.208 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#bb3754; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,735.212 2232.76,735.212 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#d04544; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,649.216 2232.76,649.216 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#e35832; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,563.22 2232.76,563.22 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#f0701e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,477.224 2232.76,477.224 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#f98c09; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,391.228 2232.76,391.228 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#fbaa0e; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,305.232 2232.76,305.232 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#f8c931; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,219.236 2232.76,219.236 "/>
+<polyline clip-path="url(#clip733)" style="stroke:#f1e864; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2160.76,133.24 2232.76,133.24 "/>
+<path clip-path="url(#clip730)" d="M2274.38 1256.53 L2290.7 1256.53 L2290.7 1260.47 L2268.76 1260.47 L2268.76 1256.53 Q2271.42 1253.78 2276 1249.15 Q2280.61 1244.5 2281.79 1243.15 Q2284.03 1240.63 2284.91 1238.9 Q2285.82 1237.14 2285.82 1235.45 Q2285.82 1232.69 2283.87 1230.96 Q2281.95 1229.22 2278.85 1229.22 Q2276.65 1229.22 2274.2 1229.98 Q2271.77 1230.75 2268.99 1232.3 L2268.99 1227.58 Q2271.81 1226.44 2274.27 1225.86 Q2276.72 1225.28 2278.76 1225.28 Q2284.13 1225.28 2287.32 1227.97 Q2290.52 1230.65 2290.52 1235.15 Q2290.52 1237.27 2289.7 1239.2 Q2288.92 1241.09 2286.81 1243.69 Q2286.23 1244.36 2283.13 1247.58 Q2280.03 1250.77 2274.38 1256.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2300.56 1225.91 L2318.92 1225.91 L2318.92 1229.84 L2304.84 1229.84 L2304.84 1238.32 Q2305.86 1237.97 2306.88 1237.81 Q2307.9 1237.62 2308.92 1237.62 Q2314.7 1237.62 2318.08 1240.79 Q2321.46 1243.96 2321.46 1249.38 Q2321.46 1254.96 2317.99 1258.06 Q2314.52 1261.14 2308.2 1261.14 Q2306.02 1261.14 2303.76 1260.77 Q2301.51 1260.4 2299.1 1259.66 L2299.1 1254.96 Q2301.19 1256.09 2303.41 1256.65 Q2305.63 1257.21 2308.11 1257.21 Q2312.11 1257.21 2314.45 1255.1 Q2316.79 1252.99 2316.79 1249.38 Q2316.79 1245.77 2314.45 1243.66 Q2312.11 1241.56 2308.11 1241.56 Q2306.23 1241.56 2304.36 1241.97 Q2302.51 1242.39 2300.56 1243.27 L2300.56 1225.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2340.68 1228.99 Q2337.07 1228.99 2335.24 1232.55 Q2333.43 1236.09 2333.43 1243.22 Q2333.43 1250.33 2335.24 1253.9 Q2337.07 1257.44 2340.68 1257.44 Q2344.31 1257.44 2346.12 1253.9 Q2347.95 1250.33 2347.95 1243.22 Q2347.95 1236.09 2346.12 1232.55 Q2344.31 1228.99 2340.68 1228.99 M2340.68 1225.28 Q2346.49 1225.28 2349.54 1229.89 Q2352.62 1234.47 2352.62 1243.22 Q2352.62 1251.95 2349.54 1256.56 Q2346.49 1261.14 2340.68 1261.14 Q2334.87 1261.14 2331.79 1256.56 Q2328.73 1251.95 2328.73 1243.22 Q2328.73 1234.47 2331.79 1229.89 Q2334.87 1225.28 2340.68 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2370.84 1228.99 Q2367.23 1228.99 2365.4 1232.55 Q2363.59 1236.09 2363.59 1243.22 Q2363.59 1250.33 2365.4 1253.9 Q2367.23 1257.44 2370.84 1257.44 Q2374.47 1257.44 2376.28 1253.9 Q2378.11 1250.33 2378.11 1243.22 Q2378.11 1236.09 2376.28 1232.55 Q2374.47 1228.99 2370.84 1228.99 M2370.84 1225.28 Q2376.65 1225.28 2379.7 1229.89 Q2382.78 1234.47 2382.78 1243.22 Q2382.78 1251.95 2379.7 1256.56 Q2376.65 1261.14 2370.84 1261.14 Q2365.03 1261.14 2361.95 1256.56 Q2358.89 1251.95 2358.89 1243.22 Q2358.89 1234.47 2361.95 1229.89 Q2365.03 1225.28 2370.84 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2401 1228.99 Q2397.39 1228.99 2395.56 1232.55 Q2393.76 1236.09 2393.76 1243.22 Q2393.76 1250.33 2395.56 1253.9 Q2397.39 1257.44 2401 1257.44 Q2404.63 1257.44 2406.44 1253.9 Q2408.27 1250.33 2408.27 1243.22 Q2408.27 1236.09 2406.44 1232.55 Q2404.63 1228.99 2401 1228.99 M2401 1225.28 Q2406.81 1225.28 2409.87 1229.89 Q2412.94 1234.47 2412.94 1243.22 Q2412.94 1251.95 2409.87 1256.56 Q2406.81 1261.14 2401 1261.14 Q2395.19 1261.14 2392.11 1256.56 Q2389.06 1251.95 2389.06 1243.22 Q2389.06 1234.47 2392.11 1229.89 Q2395.19 1225.28 2401 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2431.16 1228.99 Q2427.55 1228.99 2425.72 1232.55 Q2423.92 1236.09 2423.92 1243.22 Q2423.92 1250.33 2425.72 1253.9 Q2427.55 1257.44 2431.16 1257.44 Q2434.8 1257.44 2436.6 1253.9 Q2438.43 1250.33 2438.43 1243.22 Q2438.43 1236.09 2436.6 1232.55 Q2434.8 1228.99 2431.16 1228.99 M2431.16 1225.28 Q2436.97 1225.28 2440.03 1229.89 Q2443.11 1234.47 2443.11 1243.22 Q2443.11 1251.95 2440.03 1256.56 Q2436.97 1261.14 2431.16 1261.14 Q2425.35 1261.14 2422.27 1256.56 Q2419.22 1251.95 2419.22 1243.22 Q2419.22 1234.47 2422.27 1229.89 Q2425.35 1225.28 2431.16 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2461.32 1228.99 Q2457.71 1228.99 2455.88 1232.55 Q2454.08 1236.09 2454.08 1243.22 Q2454.08 1250.33 2455.88 1253.9 Q2457.71 1257.44 2461.32 1257.44 Q2464.96 1257.44 2466.76 1253.9 Q2468.59 1250.33 2468.59 1243.22 Q2468.59 1236.09 2466.76 1232.55 Q2464.96 1228.99 2461.32 1228.99 M2461.32 1225.28 Q2467.13 1225.28 2470.19 1229.89 Q2473.27 1234.47 2473.27 1243.22 Q2473.27 1251.95 2470.19 1256.56 Q2467.13 1261.14 2461.32 1261.14 Q2455.51 1261.14 2452.44 1256.56 Q2449.38 1251.95 2449.38 1243.22 Q2449.38 1234.47 2452.44 1229.89 Q2455.51 1225.28 2461.32 1225.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2270.21 1049 L2288.57 1049 L2288.57 1052.93 L2274.5 1052.93 L2274.5 1061.4 Q2275.52 1061.06 2276.53 1060.89 Q2277.55 1060.71 2278.57 1060.71 Q2284.36 1060.71 2287.74 1063.88 Q2291.12 1067.05 2291.12 1072.47 Q2291.12 1078.05 2287.64 1081.15 Q2284.17 1084.23 2277.85 1084.23 Q2275.68 1084.23 2273.41 1083.86 Q2271.16 1083.49 2268.76 1082.75 L2268.76 1078.05 Q2270.84 1079.18 2273.06 1079.74 Q2275.28 1080.29 2277.76 1080.29 Q2281.77 1080.29 2284.1 1078.19 Q2286.44 1076.08 2286.44 1072.47 Q2286.44 1068.86 2284.1 1066.75 Q2281.77 1064.64 2277.76 1064.64 Q2275.89 1064.64 2274.01 1065.06 Q2272.16 1065.48 2270.21 1066.36 L2270.21 1049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2310.33 1052.07 Q2306.72 1052.07 2304.89 1055.64 Q2303.08 1059.18 2303.08 1066.31 Q2303.08 1073.42 2304.89 1076.98 Q2306.72 1080.52 2310.33 1080.52 Q2313.96 1080.52 2315.77 1076.98 Q2317.6 1073.42 2317.6 1066.31 Q2317.6 1059.18 2315.77 1055.64 Q2313.96 1052.07 2310.33 1052.07 M2310.33 1048.37 Q2316.14 1048.37 2319.2 1052.98 Q2322.27 1057.56 2322.27 1066.31 Q2322.27 1075.04 2319.2 1079.64 Q2316.14 1084.23 2310.33 1084.23 Q2304.52 1084.23 2301.44 1079.64 Q2298.39 1075.04 2298.39 1066.31 Q2298.39 1057.56 2301.44 1052.98 Q2304.52 1048.37 2310.33 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2340.49 1052.07 Q2336.88 1052.07 2335.05 1055.64 Q2333.25 1059.18 2333.25 1066.31 Q2333.25 1073.42 2335.05 1076.98 Q2336.88 1080.52 2340.49 1080.52 Q2344.13 1080.52 2345.93 1076.98 Q2347.76 1073.42 2347.76 1066.31 Q2347.76 1059.18 2345.93 1055.64 Q2344.13 1052.07 2340.49 1052.07 M2340.49 1048.37 Q2346.3 1048.37 2349.36 1052.98 Q2352.44 1057.56 2352.44 1066.31 Q2352.44 1075.04 2349.36 1079.64 Q2346.3 1084.23 2340.49 1084.23 Q2334.68 1084.23 2331.6 1079.64 Q2328.55 1075.04 2328.55 1066.31 Q2328.55 1057.56 2331.6 1052.98 Q2334.68 1048.37 2340.49 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2370.65 1052.07 Q2367.04 1052.07 2365.21 1055.64 Q2363.41 1059.18 2363.41 1066.31 Q2363.41 1073.42 2365.21 1076.98 Q2367.04 1080.52 2370.65 1080.52 Q2374.29 1080.52 2376.09 1076.98 Q2377.92 1073.42 2377.92 1066.31 Q2377.92 1059.18 2376.09 1055.64 Q2374.29 1052.07 2370.65 1052.07 M2370.65 1048.37 Q2376.46 1048.37 2379.52 1052.98 Q2382.6 1057.56 2382.6 1066.31 Q2382.6 1075.04 2379.52 1079.64 Q2376.46 1084.23 2370.65 1084.23 Q2364.84 1084.23 2361.76 1079.64 Q2358.71 1075.04 2358.71 1066.31 Q2358.71 1057.56 2361.76 1052.98 Q2364.84 1048.37 2370.65 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2400.82 1052.07 Q2397.2 1052.07 2395.38 1055.64 Q2393.57 1059.18 2393.57 1066.31 Q2393.57 1073.42 2395.38 1076.98 Q2397.2 1080.52 2400.82 1080.52 Q2404.45 1080.52 2406.26 1076.98 Q2408.08 1073.42 2408.08 1066.31 Q2408.08 1059.18 2406.26 1055.64 Q2404.45 1052.07 2400.82 1052.07 M2400.82 1048.37 Q2406.63 1048.37 2409.68 1052.98 Q2412.76 1057.56 2412.76 1066.31 Q2412.76 1075.04 2409.68 1079.64 Q2406.63 1084.23 2400.82 1084.23 Q2395.01 1084.23 2391.93 1079.64 Q2388.87 1075.04 2388.87 1066.31 Q2388.87 1057.56 2391.93 1052.98 Q2395.01 1048.37 2400.82 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2430.98 1052.07 Q2427.37 1052.07 2425.54 1055.64 Q2423.73 1059.18 2423.73 1066.31 Q2423.73 1073.42 2425.54 1076.98 Q2427.37 1080.52 2430.98 1080.52 Q2434.61 1080.52 2436.42 1076.98 Q2438.25 1073.42 2438.25 1066.31 Q2438.25 1059.18 2436.42 1055.64 Q2434.61 1052.07 2430.98 1052.07 M2430.98 1048.37 Q2436.79 1048.37 2439.84 1052.98 Q2442.92 1057.56 2442.92 1066.31 Q2442.92 1075.04 2439.84 1079.64 Q2436.79 1084.23 2430.98 1084.23 Q2425.17 1084.23 2422.09 1079.64 Q2419.03 1075.04 2419.03 1066.31 Q2419.03 1057.56 2422.09 1052.98 Q2425.17 1048.37 2430.98 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2461.14 1052.07 Q2457.53 1052.07 2455.7 1055.64 Q2453.89 1059.18 2453.89 1066.31 Q2453.89 1073.42 2455.7 1076.98 Q2457.53 1080.52 2461.14 1080.52 Q2464.77 1080.52 2466.58 1076.98 Q2468.41 1073.42 2468.41 1066.31 Q2468.41 1059.18 2466.58 1055.64 Q2464.77 1052.07 2461.14 1052.07 M2461.14 1048.37 Q2466.95 1048.37 2470 1052.98 Q2473.08 1057.56 2473.08 1066.31 Q2473.08 1075.04 2470 1079.64 Q2466.95 1084.23 2461.14 1084.23 Q2455.33 1084.23 2452.25 1079.64 Q2449.19 1075.04 2449.19 1066.31 Q2449.19 1057.56 2452.25 1052.98 Q2455.33 1048.37 2461.14 1048.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2268.76 872.082 L2290.98 872.082 L2290.98 874.073 L2278.43 906.642 L2273.55 906.642 L2285.35 876.017 L2268.76 876.017 L2268.76 872.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2300.14 872.082 L2318.5 872.082 L2318.5 876.017 L2304.43 876.017 L2304.43 884.489 Q2305.45 884.142 2306.46 883.98 Q2307.48 883.795 2308.5 883.795 Q2314.29 883.795 2317.67 886.966 Q2321.05 890.137 2321.05 895.554 Q2321.05 901.133 2317.58 904.234 Q2314.1 907.313 2307.78 907.313 Q2305.61 907.313 2303.34 906.943 Q2301.09 906.572 2298.69 905.832 L2298.69 901.133 Q2300.77 902.267 2302.99 902.822 Q2305.21 903.378 2307.69 903.378 Q2311.7 903.378 2314.03 901.272 Q2316.37 899.165 2316.37 895.554 Q2316.37 891.943 2314.03 889.836 Q2311.7 887.73 2307.69 887.73 Q2305.82 887.73 2303.94 888.147 Q2302.09 888.563 2300.14 889.443 L2300.14 872.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2340.26 875.161 Q2336.65 875.161 2334.82 878.725 Q2333.01 882.267 2333.01 889.397 Q2333.01 896.503 2334.82 900.068 Q2336.65 903.609 2340.26 903.609 Q2343.89 903.609 2345.7 900.068 Q2347.53 896.503 2347.53 889.397 Q2347.53 882.267 2345.7 878.725 Q2343.89 875.161 2340.26 875.161 M2340.26 871.457 Q2346.07 871.457 2349.13 876.063 Q2352.2 880.647 2352.2 889.397 Q2352.2 898.123 2349.13 902.73 Q2346.07 907.313 2340.26 907.313 Q2334.45 907.313 2331.37 902.73 Q2328.32 898.123 2328.32 889.397 Q2328.32 880.647 2331.37 876.063 Q2334.45 871.457 2340.26 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2370.42 875.161 Q2366.81 875.161 2364.98 878.725 Q2363.18 882.267 2363.18 889.397 Q2363.18 896.503 2364.98 900.068 Q2366.81 903.609 2370.42 903.609 Q2374.06 903.609 2375.86 900.068 Q2377.69 896.503 2377.69 889.397 Q2377.69 882.267 2375.86 878.725 Q2374.06 875.161 2370.42 875.161 M2370.42 871.457 Q2376.23 871.457 2379.29 876.063 Q2382.37 880.647 2382.37 889.397 Q2382.37 898.123 2379.29 902.73 Q2376.23 907.313 2370.42 907.313 Q2364.61 907.313 2361.53 902.73 Q2358.48 898.123 2358.48 889.397 Q2358.48 880.647 2361.53 876.063 Q2364.61 871.457 2370.42 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2400.58 875.161 Q2396.97 875.161 2395.14 878.725 Q2393.34 882.267 2393.34 889.397 Q2393.34 896.503 2395.14 900.068 Q2396.97 903.609 2400.58 903.609 Q2404.22 903.609 2406.02 900.068 Q2407.85 896.503 2407.85 889.397 Q2407.85 882.267 2406.02 878.725 Q2404.22 875.161 2400.58 875.161 M2400.58 871.457 Q2406.39 871.457 2409.45 876.063 Q2412.53 880.647 2412.53 889.397 Q2412.53 898.123 2409.45 902.73 Q2406.39 907.313 2400.58 907.313 Q2394.77 907.313 2391.7 902.73 Q2388.64 898.123 2388.64 889.397 Q2388.64 880.647 2391.7 876.063 Q2394.77 871.457 2400.58 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2430.75 875.161 Q2427.13 875.161 2425.31 878.725 Q2423.5 882.267 2423.5 889.397 Q2423.5 896.503 2425.31 900.068 Q2427.13 903.609 2430.75 903.609 Q2434.38 903.609 2436.19 900.068 Q2438.01 896.503 2438.01 889.397 Q2438.01 882.267 2436.19 878.725 Q2434.38 875.161 2430.75 875.161 M2430.75 871.457 Q2436.56 871.457 2439.61 876.063 Q2442.69 880.647 2442.69 889.397 Q2442.69 898.123 2439.61 902.73 Q2436.56 907.313 2430.75 907.313 Q2424.94 907.313 2421.86 902.73 Q2418.8 898.123 2418.8 889.397 Q2418.8 880.647 2421.86 876.063 Q2424.94 871.457 2430.75 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2460.91 875.161 Q2457.3 875.161 2455.47 878.725 Q2453.66 882.267 2453.66 889.397 Q2453.66 896.503 2455.47 900.068 Q2457.3 903.609 2460.91 903.609 Q2464.54 903.609 2466.35 900.068 Q2468.18 896.503 2468.18 889.397 Q2468.18 882.267 2466.35 878.725 Q2464.54 875.161 2460.91 875.161 M2460.91 871.457 Q2466.72 871.457 2469.77 876.063 Q2472.85 880.647 2472.85 889.397 Q2472.85 898.123 2469.77 902.73 Q2466.72 907.313 2460.91 907.313 Q2455.1 907.313 2452.02 902.73 Q2448.96 898.123 2448.96 889.397 Q2448.96 880.647 2452.02 876.063 Q2455.1 871.457 2460.91 871.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2269.43 725.793 L2277.07 725.793 L2277.07 699.427 L2268.76 701.094 L2268.76 696.835 L2277.02 695.168 L2281.7 695.168 L2281.7 725.793 L2289.33 725.793 L2289.33 729.728 L2269.43 729.728 L2269.43 725.793 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2308.78 698.247 Q2305.17 698.247 2303.34 701.812 Q2301.53 705.353 2301.53 712.483 Q2301.53 719.589 2303.34 723.154 Q2305.17 726.696 2308.78 726.696 Q2312.41 726.696 2314.22 723.154 Q2316.05 719.589 2316.05 712.483 Q2316.05 705.353 2314.22 701.812 Q2312.41 698.247 2308.78 698.247 M2308.78 694.543 Q2314.59 694.543 2317.64 699.15 Q2320.72 703.733 2320.72 712.483 Q2320.72 721.21 2317.64 725.816 Q2314.59 730.399 2308.78 730.399 Q2302.97 730.399 2299.89 725.816 Q2296.83 721.21 2296.83 712.483 Q2296.83 703.733 2299.89 699.15 Q2302.97 694.543 2308.78 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2338.94 698.247 Q2335.33 698.247 2333.5 701.812 Q2331.7 705.353 2331.7 712.483 Q2331.7 719.589 2333.5 723.154 Q2335.33 726.696 2338.94 726.696 Q2342.57 726.696 2344.38 723.154 Q2346.21 719.589 2346.21 712.483 Q2346.21 705.353 2344.38 701.812 Q2342.57 698.247 2338.94 698.247 M2338.94 694.543 Q2344.75 694.543 2347.81 699.15 Q2350.89 703.733 2350.89 712.483 Q2350.89 721.21 2347.81 725.816 Q2344.75 730.399 2338.94 730.399 Q2333.13 730.399 2330.05 725.816 Q2327 721.21 2327 712.483 Q2327 703.733 2330.05 699.15 Q2333.13 694.543 2338.94 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2369.1 698.247 Q2365.49 698.247 2363.66 701.812 Q2361.86 705.353 2361.86 712.483 Q2361.86 719.589 2363.66 723.154 Q2365.49 726.696 2369.1 726.696 Q2372.74 726.696 2374.54 723.154 Q2376.37 719.589 2376.37 712.483 Q2376.37 705.353 2374.54 701.812 Q2372.74 698.247 2369.1 698.247 M2369.1 694.543 Q2374.91 694.543 2377.97 699.15 Q2381.05 703.733 2381.05 712.483 Q2381.05 721.21 2377.97 725.816 Q2374.91 730.399 2369.1 730.399 Q2363.29 730.399 2360.21 725.816 Q2357.16 721.21 2357.16 712.483 Q2357.16 703.733 2360.21 699.15 Q2363.29 694.543 2369.1 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2399.26 698.247 Q2395.65 698.247 2393.82 701.812 Q2392.02 705.353 2392.02 712.483 Q2392.02 719.589 2393.82 723.154 Q2395.65 726.696 2399.26 726.696 Q2402.9 726.696 2404.7 723.154 Q2406.53 719.589 2406.53 712.483 Q2406.53 705.353 2404.7 701.812 Q2402.9 698.247 2399.26 698.247 M2399.26 694.543 Q2405.07 694.543 2408.13 699.15 Q2411.21 703.733 2411.21 712.483 Q2411.21 721.21 2408.13 725.816 Q2405.07 730.399 2399.26 730.399 Q2393.45 730.399 2390.38 725.816 Q2387.32 721.21 2387.32 712.483 Q2387.32 703.733 2390.38 699.15 Q2393.45 694.543 2399.26 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2429.43 698.247 Q2425.82 698.247 2423.99 701.812 Q2422.18 705.353 2422.18 712.483 Q2422.18 719.589 2423.99 723.154 Q2425.82 726.696 2429.43 726.696 Q2433.06 726.696 2434.87 723.154 Q2436.69 719.589 2436.69 712.483 Q2436.69 705.353 2434.87 701.812 Q2433.06 698.247 2429.43 698.247 M2429.43 694.543 Q2435.24 694.543 2438.29 699.15 Q2441.37 703.733 2441.37 712.483 Q2441.37 721.21 2438.29 725.816 Q2435.24 730.399 2429.43 730.399 Q2423.62 730.399 2420.54 725.816 Q2417.48 721.21 2417.48 712.483 Q2417.48 703.733 2420.54 699.15 Q2423.62 694.543 2429.43 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2459.59 698.247 Q2455.98 698.247 2454.15 701.812 Q2452.34 705.353 2452.34 712.483 Q2452.34 719.589 2454.15 723.154 Q2455.98 726.696 2459.59 726.696 Q2463.22 726.696 2465.03 723.154 Q2466.86 719.589 2466.86 712.483 Q2466.86 705.353 2465.03 701.812 Q2463.22 698.247 2459.59 698.247 M2459.59 694.543 Q2465.4 694.543 2468.45 699.15 Q2471.53 703.733 2471.53 712.483 Q2471.53 721.21 2468.45 725.816 Q2465.4 730.399 2459.59 730.399 Q2453.78 730.399 2450.7 725.816 Q2447.64 721.21 2447.64 712.483 Q2447.64 703.733 2450.7 699.15 Q2453.78 694.543 2459.59 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2489.75 698.247 Q2486.14 698.247 2484.31 701.812 Q2482.5 705.353 2482.5 712.483 Q2482.5 719.589 2484.31 723.154 Q2486.14 726.696 2489.75 726.696 Q2493.38 726.696 2495.19 723.154 Q2497.02 719.589 2497.02 712.483 Q2497.02 705.353 2495.19 701.812 Q2493.38 698.247 2489.75 698.247 M2489.75 694.543 Q2495.56 694.543 2498.62 699.15 Q2501.69 703.733 2501.69 712.483 Q2501.69 721.21 2498.62 725.816 Q2495.56 730.399 2489.75 730.399 Q2483.94 730.399 2480.86 725.816 Q2477.81 721.21 2477.81 712.483 Q2477.81 703.733 2480.86 699.15 Q2483.94 694.543 2489.75 694.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2269.43 548.879 L2277.07 548.879 L2277.07 522.514 L2268.76 524.18 L2268.76 519.921 L2277.02 518.254 L2281.7 518.254 L2281.7 548.879 L2289.33 548.879 L2289.33 552.814 L2269.43 552.814 L2269.43 548.879 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2302.81 548.879 L2319.13 548.879 L2319.13 552.814 L2297.18 552.814 L2297.18 548.879 Q2299.84 546.125 2304.43 541.495 Q2309.03 536.842 2310.21 535.5 Q2312.46 532.977 2313.34 531.241 Q2314.24 529.481 2314.24 527.791 Q2314.24 525.037 2312.3 523.301 Q2310.38 521.565 2307.27 521.565 Q2305.08 521.565 2302.62 522.329 Q2300.19 523.092 2297.41 524.643 L2297.41 519.921 Q2300.24 518.787 2302.69 518.208 Q2305.14 517.629 2307.18 517.629 Q2312.55 517.629 2315.75 520.315 Q2318.94 523 2318.94 527.491 Q2318.94 529.62 2318.13 531.541 Q2317.34 533.44 2315.24 536.032 Q2314.66 536.703 2311.56 539.921 Q2308.45 543.115 2302.81 548.879 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2328.99 518.254 L2347.34 518.254 L2347.34 522.19 L2333.27 522.19 L2333.27 530.662 Q2334.29 530.315 2335.31 530.153 Q2336.32 529.967 2337.34 529.967 Q2343.13 529.967 2346.51 533.139 Q2349.89 536.31 2349.89 541.727 Q2349.89 547.305 2346.42 550.407 Q2342.95 553.486 2336.63 553.486 Q2334.45 553.486 2332.18 553.115 Q2329.94 552.745 2327.53 552.004 L2327.53 547.305 Q2329.61 548.439 2331.83 548.995 Q2334.06 549.551 2336.53 549.551 Q2340.54 549.551 2342.88 547.444 Q2345.21 545.338 2345.21 541.727 Q2345.21 538.115 2342.88 536.009 Q2340.54 533.903 2336.53 533.903 Q2334.66 533.903 2332.78 534.319 Q2330.93 534.736 2328.99 535.615 L2328.99 518.254 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2369.1 521.333 Q2365.49 521.333 2363.66 524.898 Q2361.86 528.44 2361.86 535.569 Q2361.86 542.676 2363.66 546.24 Q2365.49 549.782 2369.1 549.782 Q2372.74 549.782 2374.54 546.24 Q2376.37 542.676 2376.37 535.569 Q2376.37 528.44 2374.54 524.898 Q2372.74 521.333 2369.1 521.333 M2369.1 517.629 Q2374.91 517.629 2377.97 522.236 Q2381.05 526.819 2381.05 535.569 Q2381.05 544.296 2377.97 548.902 Q2374.91 553.486 2369.1 553.486 Q2363.29 553.486 2360.21 548.902 Q2357.16 544.296 2357.16 535.569 Q2357.16 526.819 2360.21 522.236 Q2363.29 517.629 2369.1 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2399.26 521.333 Q2395.65 521.333 2393.82 524.898 Q2392.02 528.44 2392.02 535.569 Q2392.02 542.676 2393.82 546.24 Q2395.65 549.782 2399.26 549.782 Q2402.9 549.782 2404.7 546.24 Q2406.53 542.676 2406.53 535.569 Q2406.53 528.44 2404.7 524.898 Q2402.9 521.333 2399.26 521.333 M2399.26 517.629 Q2405.07 517.629 2408.13 522.236 Q2411.21 526.819 2411.21 535.569 Q2411.21 544.296 2408.13 548.902 Q2405.07 553.486 2399.26 553.486 Q2393.45 553.486 2390.38 548.902 Q2387.32 544.296 2387.32 535.569 Q2387.32 526.819 2390.38 522.236 Q2393.45 517.629 2399.26 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2429.43 521.333 Q2425.82 521.333 2423.99 524.898 Q2422.18 528.44 2422.18 535.569 Q2422.18 542.676 2423.99 546.24 Q2425.82 549.782 2429.43 549.782 Q2433.06 549.782 2434.87 546.24 Q2436.69 542.676 2436.69 535.569 Q2436.69 528.44 2434.87 524.898 Q2433.06 521.333 2429.43 521.333 M2429.43 517.629 Q2435.24 517.629 2438.29 522.236 Q2441.37 526.819 2441.37 535.569 Q2441.37 544.296 2438.29 548.902 Q2435.24 553.486 2429.43 553.486 Q2423.62 553.486 2420.54 548.902 Q2417.48 544.296 2417.48 535.569 Q2417.48 526.819 2420.54 522.236 Q2423.62 517.629 2429.43 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2459.59 521.333 Q2455.98 521.333 2454.15 524.898 Q2452.34 528.44 2452.34 535.569 Q2452.34 542.676 2454.15 546.24 Q2455.98 549.782 2459.59 549.782 Q2463.22 549.782 2465.03 546.24 Q2466.86 542.676 2466.86 535.569 Q2466.86 528.44 2465.03 524.898 Q2463.22 521.333 2459.59 521.333 M2459.59 517.629 Q2465.4 517.629 2468.45 522.236 Q2471.53 526.819 2471.53 535.569 Q2471.53 544.296 2468.45 548.902 Q2465.4 553.486 2459.59 553.486 Q2453.78 553.486 2450.7 548.902 Q2447.64 544.296 2447.64 535.569 Q2447.64 526.819 2450.7 522.236 Q2453.78 517.629 2459.59 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2489.75 521.333 Q2486.14 521.333 2484.31 524.898 Q2482.5 528.44 2482.5 535.569 Q2482.5 542.676 2484.31 546.24 Q2486.14 549.782 2489.75 549.782 Q2493.38 549.782 2495.19 546.24 Q2497.02 542.676 2497.02 535.569 Q2497.02 528.44 2495.19 524.898 Q2493.38 521.333 2489.75 521.333 M2489.75 517.629 Q2495.56 517.629 2498.62 522.236 Q2501.69 526.819 2501.69 535.569 Q2501.69 544.296 2498.62 548.902 Q2495.56 553.486 2489.75 553.486 Q2483.94 553.486 2480.86 548.902 Q2477.81 544.296 2477.81 535.569 Q2477.81 526.819 2480.86 522.236 Q2483.94 517.629 2489.75 517.629 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2269.43 371.966 L2277.07 371.966 L2277.07 345.6 L2268.76 347.267 L2268.76 343.007 L2277.02 341.341 L2281.7 341.341 L2281.7 371.966 L2289.33 371.966 L2289.33 375.901 L2269.43 375.901 L2269.43 371.966 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2298.83 341.341 L2317.18 341.341 L2317.18 345.276 L2303.11 345.276 L2303.11 353.748 Q2304.13 353.401 2305.14 353.239 Q2306.16 353.054 2307.18 353.054 Q2312.97 353.054 2316.35 356.225 Q2319.73 359.396 2319.73 364.813 Q2319.73 370.392 2316.26 373.493 Q2312.78 376.572 2306.46 376.572 Q2304.29 376.572 2302.02 376.202 Q2299.77 375.831 2297.37 375.091 L2297.37 370.392 Q2299.45 371.526 2301.67 372.081 Q2303.89 372.637 2306.37 372.637 Q2310.38 372.637 2312.71 370.53 Q2315.05 368.424 2315.05 364.813 Q2315.05 361.202 2312.71 359.095 Q2310.38 356.989 2306.37 356.989 Q2304.5 356.989 2302.62 357.405 Q2300.77 357.822 2298.83 358.702 L2298.83 341.341 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2338.94 344.419 Q2335.33 344.419 2333.5 347.984 Q2331.7 351.526 2331.7 358.655 Q2331.7 365.762 2333.5 369.327 Q2335.33 372.868 2338.94 372.868 Q2342.57 372.868 2344.38 369.327 Q2346.21 365.762 2346.21 358.655 Q2346.21 351.526 2344.38 347.984 Q2342.57 344.419 2338.94 344.419 M2338.94 340.716 Q2344.75 340.716 2347.81 345.322 Q2350.89 349.906 2350.89 358.655 Q2350.89 367.382 2347.81 371.989 Q2344.75 376.572 2338.94 376.572 Q2333.13 376.572 2330.05 371.989 Q2327 367.382 2327 358.655 Q2327 349.906 2330.05 345.322 Q2333.13 340.716 2338.94 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2369.1 344.419 Q2365.49 344.419 2363.66 347.984 Q2361.86 351.526 2361.86 358.655 Q2361.86 365.762 2363.66 369.327 Q2365.49 372.868 2369.1 372.868 Q2372.74 372.868 2374.54 369.327 Q2376.37 365.762 2376.37 358.655 Q2376.37 351.526 2374.54 347.984 Q2372.74 344.419 2369.1 344.419 M2369.1 340.716 Q2374.91 340.716 2377.97 345.322 Q2381.05 349.906 2381.05 358.655 Q2381.05 367.382 2377.97 371.989 Q2374.91 376.572 2369.1 376.572 Q2363.29 376.572 2360.21 371.989 Q2357.16 367.382 2357.16 358.655 Q2357.16 349.906 2360.21 345.322 Q2363.29 340.716 2369.1 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2399.26 344.419 Q2395.65 344.419 2393.82 347.984 Q2392.02 351.526 2392.02 358.655 Q2392.02 365.762 2393.82 369.327 Q2395.65 372.868 2399.26 372.868 Q2402.9 372.868 2404.7 369.327 Q2406.53 365.762 2406.53 358.655 Q2406.53 351.526 2404.7 347.984 Q2402.9 344.419 2399.26 344.419 M2399.26 340.716 Q2405.07 340.716 2408.13 345.322 Q2411.21 349.906 2411.21 358.655 Q2411.21 367.382 2408.13 371.989 Q2405.07 376.572 2399.26 376.572 Q2393.45 376.572 2390.38 371.989 Q2387.32 367.382 2387.32 358.655 Q2387.32 349.906 2390.38 345.322 Q2393.45 340.716 2399.26 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2429.43 344.419 Q2425.82 344.419 2423.99 347.984 Q2422.18 351.526 2422.18 358.655 Q2422.18 365.762 2423.99 369.327 Q2425.82 372.868 2429.43 372.868 Q2433.06 372.868 2434.87 369.327 Q2436.69 365.762 2436.69 358.655 Q2436.69 351.526 2434.87 347.984 Q2433.06 344.419 2429.43 344.419 M2429.43 340.716 Q2435.24 340.716 2438.29 345.322 Q2441.37 349.906 2441.37 358.655 Q2441.37 367.382 2438.29 371.989 Q2435.24 376.572 2429.43 376.572 Q2423.62 376.572 2420.54 371.989 Q2417.48 367.382 2417.48 358.655 Q2417.48 349.906 2420.54 345.322 Q2423.62 340.716 2429.43 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2459.59 344.419 Q2455.98 344.419 2454.15 347.984 Q2452.34 351.526 2452.34 358.655 Q2452.34 365.762 2454.15 369.327 Q2455.98 372.868 2459.59 372.868 Q2463.22 372.868 2465.03 369.327 Q2466.86 365.762 2466.86 358.655 Q2466.86 351.526 2465.03 347.984 Q2463.22 344.419 2459.59 344.419 M2459.59 340.716 Q2465.4 340.716 2468.45 345.322 Q2471.53 349.906 2471.53 358.655 Q2471.53 367.382 2468.45 371.989 Q2465.4 376.572 2459.59 376.572 Q2453.78 376.572 2450.7 371.989 Q2447.64 367.382 2447.64 358.655 Q2447.64 349.906 2450.7 345.322 Q2453.78 340.716 2459.59 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2489.75 344.419 Q2486.14 344.419 2484.31 347.984 Q2482.5 351.526 2482.5 358.655 Q2482.5 365.762 2484.31 369.327 Q2486.14 372.868 2489.75 372.868 Q2493.38 372.868 2495.19 369.327 Q2497.02 365.762 2497.02 358.655 Q2497.02 351.526 2495.19 347.984 Q2493.38 344.419 2489.75 344.419 M2489.75 340.716 Q2495.56 340.716 2498.62 345.322 Q2501.69 349.906 2501.69 358.655 Q2501.69 367.382 2498.62 371.989 Q2495.56 376.572 2489.75 376.572 Q2483.94 376.572 2480.86 371.989 Q2477.81 367.382 2477.81 358.655 Q2477.81 349.906 2480.86 345.322 Q2483.94 340.716 2489.75 340.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2269.43 195.052 L2277.07 195.052 L2277.07 168.686 L2268.76 170.353 L2268.76 166.094 L2277.02 164.427 L2281.7 164.427 L2281.7 195.052 L2289.33 195.052 L2289.33 198.987 L2269.43 198.987 L2269.43 195.052 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2297.6 164.427 L2319.82 164.427 L2319.82 166.418 L2307.27 198.987 L2302.39 198.987 L2314.2 168.362 L2297.6 168.362 L2297.6 164.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2328.99 164.427 L2347.34 164.427 L2347.34 168.362 L2333.27 168.362 L2333.27 176.834 Q2334.29 176.487 2335.31 176.325 Q2336.32 176.14 2337.34 176.14 Q2343.13 176.14 2346.51 179.311 Q2349.89 182.483 2349.89 187.899 Q2349.89 193.478 2346.42 196.58 Q2342.95 199.658 2336.63 199.658 Q2334.45 199.658 2332.18 199.288 Q2329.94 198.918 2327.53 198.177 L2327.53 193.478 Q2329.61 194.612 2331.83 195.168 Q2334.06 195.723 2336.53 195.723 Q2340.54 195.723 2342.88 193.617 Q2345.21 191.51 2345.21 187.899 Q2345.21 184.288 2342.88 182.182 Q2340.54 180.075 2336.53 180.075 Q2334.66 180.075 2332.78 180.492 Q2330.93 180.908 2328.99 181.788 L2328.99 164.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2369.1 167.506 Q2365.49 167.506 2363.66 171.071 Q2361.86 174.612 2361.86 181.742 Q2361.86 188.848 2363.66 192.413 Q2365.49 195.955 2369.1 195.955 Q2372.74 195.955 2374.54 192.413 Q2376.37 188.848 2376.37 181.742 Q2376.37 174.612 2374.54 171.071 Q2372.74 167.506 2369.1 167.506 M2369.1 163.802 Q2374.91 163.802 2377.97 168.409 Q2381.05 172.992 2381.05 181.742 Q2381.05 190.469 2377.97 195.075 Q2374.91 199.658 2369.1 199.658 Q2363.29 199.658 2360.21 195.075 Q2357.16 190.469 2357.16 181.742 Q2357.16 172.992 2360.21 168.409 Q2363.29 163.802 2369.1 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2399.26 167.506 Q2395.65 167.506 2393.82 171.071 Q2392.02 174.612 2392.02 181.742 Q2392.02 188.848 2393.82 192.413 Q2395.65 195.955 2399.26 195.955 Q2402.9 195.955 2404.7 192.413 Q2406.53 188.848 2406.53 181.742 Q2406.53 174.612 2404.7 171.071 Q2402.9 167.506 2399.26 167.506 M2399.26 163.802 Q2405.07 163.802 2408.13 168.409 Q2411.21 172.992 2411.21 181.742 Q2411.21 190.469 2408.13 195.075 Q2405.07 199.658 2399.26 199.658 Q2393.45 199.658 2390.38 195.075 Q2387.32 190.469 2387.32 181.742 Q2387.32 172.992 2390.38 168.409 Q2393.45 163.802 2399.26 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2429.43 167.506 Q2425.82 167.506 2423.99 171.071 Q2422.18 174.612 2422.18 181.742 Q2422.18 188.848 2423.99 192.413 Q2425.82 195.955 2429.43 195.955 Q2433.06 195.955 2434.87 192.413 Q2436.69 188.848 2436.69 181.742 Q2436.69 174.612 2434.87 171.071 Q2433.06 167.506 2429.43 167.506 M2429.43 163.802 Q2435.24 163.802 2438.29 168.409 Q2441.37 172.992 2441.37 181.742 Q2441.37 190.469 2438.29 195.075 Q2435.24 199.658 2429.43 199.658 Q2423.62 199.658 2420.54 195.075 Q2417.48 190.469 2417.48 181.742 Q2417.48 172.992 2420.54 168.409 Q2423.62 163.802 2429.43 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2459.59 167.506 Q2455.98 167.506 2454.15 171.071 Q2452.34 174.612 2452.34 181.742 Q2452.34 188.848 2454.15 192.413 Q2455.98 195.955 2459.59 195.955 Q2463.22 195.955 2465.03 192.413 Q2466.86 188.848 2466.86 181.742 Q2466.86 174.612 2465.03 171.071 Q2463.22 167.506 2459.59 167.506 M2459.59 163.802 Q2465.4 163.802 2468.45 168.409 Q2471.53 172.992 2471.53 181.742 Q2471.53 190.469 2468.45 195.075 Q2465.4 199.658 2459.59 199.658 Q2453.78 199.658 2450.7 195.075 Q2447.64 190.469 2447.64 181.742 Q2447.64 172.992 2450.7 168.409 Q2453.78 163.802 2459.59 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip730)" d="M2489.75 167.506 Q2486.14 167.506 2484.31 171.071 Q2482.5 174.612 2482.5 181.742 Q2482.5 188.848 2484.31 192.413 Q2486.14 195.955 2489.75 195.955 Q2493.38 195.955 2495.19 192.413 Q2497.02 188.848 2497.02 181.742 Q2497.02 174.612 2495.19 171.071 Q2493.38 167.506 2489.75 167.506 M2489.75 163.802 Q2495.56 163.802 2498.62 168.409 Q2501.69 172.992 2501.69 181.742 Q2501.69 190.469 2498.62 195.075 Q2495.56 199.658 2489.75 199.658 Q2483.94 199.658 2480.86 195.075 Q2477.81 190.469 2477.81 181.742 Q2477.81 172.992 2480.86 168.409 Q2483.94 163.802 2489.75 163.802 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip730)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2232.76,1423.18 2232.76,1246.82 2256.76,1246.82 2232.76,1246.82 2232.76,1069.9 2256.76,1069.9 2232.76,1069.9 2232.76,892.991 2256.76,892.991 2232.76,892.991 2232.76,716.077 2256.76,716.077 2232.76,716.077 2232.76,539.163 2256.76,539.163 2232.76,539.163 2232.76,362.25 2256.76,362.25 2232.76,362.25 2232.76,185.336 2256.76,185.336 2232.76,185.336 2232.76,47.2441 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/linear/factory_schedule/index.html b/previews/PR3562/tutorials/linear/factory_schedule/index.html
index f6661693e91..04dcdb541ff 100644
--- a/previews/PR3562/tutorials/linear/factory_schedule/index.html
+++ b/previews/PR3562/tutorials/linear/factory_schedule/index.html
@@ -125,7 +125,7 @@
     ylabel = &quot;Production&quot;,
     legend = :topleft,
     color = [&quot;#20326c&quot; &quot;#4063d8&quot; &quot;#a0b1ec&quot;],
-)</code></pre><img src="9faf22a9.svg" alt="Example block output"/><p>Note that we don&#39;t have any unmet demand.</p><h2 id="What-happens-if-demand-increases?"><a class="docs-heading-anchor" href="#What-happens-if-demand-increases?">What happens if demand increases?</a><a id="What-happens-if-demand-increases?-1"></a><a class="docs-heading-anchor-permalink" href="#What-happens-if-demand-increases?" title="Permalink"></a></h2><p>Let&#39;s run an experiment by increasing the demand by 50% in all time periods:</p><pre><code class="language-julia hljs">demand_df.demand .*= 1.5</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">12-element Vector{Float64}:
+)</code></pre><img src="22aa7ce9.svg" alt="Example block output"/><p>Note that we don&#39;t have any unmet demand.</p><h2 id="What-happens-if-demand-increases?"><a class="docs-heading-anchor" href="#What-happens-if-demand-increases?">What happens if demand increases?</a><a id="What-happens-if-demand-increases?-1"></a><a class="docs-heading-anchor-permalink" href="#What-happens-if-demand-increases?" title="Permalink"></a></h2><p>Let&#39;s run an experiment by increasing the demand by 50% in all time periods:</p><pre><code class="language-julia hljs">demand_df.demand .*= 1.5</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">12-element Vector{Float64}:
  180000.0
  150000.0
  195000.0
@@ -145,7 +145,7 @@
     ylabel = &quot;Production&quot;,
     legend = :topleft,
     color = [&quot;#20326c&quot; &quot;#4063d8&quot; &quot;#a0b1ec&quot;],
-)</code></pre><img src="79a4ee80.svg" alt="Example block output"/><p>Uh oh, we can&#39;t satisfy all of the demand.</p><h2 id="How-sensitive-is-the-solution-to-changes-in-variable-cost?"><a class="docs-heading-anchor" href="#How-sensitive-is-the-solution-to-changes-in-variable-cost?">How sensitive is the solution to changes in variable cost?</a><a id="How-sensitive-is-the-solution-to-changes-in-variable-cost?-1"></a><a class="docs-heading-anchor-permalink" href="#How-sensitive-is-the-solution-to-changes-in-variable-cost?" title="Permalink"></a></h2><p>Let&#39;s run another experiment, this time seeing how the optimal objective value changes as we vary the variable costs of each factory.</p><p>First though, let&#39;s reset the demand to it&#39;s original level:</p><pre><code class="language-julia hljs">demand_df.demand ./= 1.5;</code></pre><p>For our experiment, we&#39;re going to scale the variable costs of both factories by a set of values from <code>0.0</code> to <code>1.5</code>:</p><pre><code class="language-julia hljs">scale_factors = 0:0.1:1.5</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.0:0.1:1.5</code></pre><p>At a high level, we&#39;re going to loop over the scale factors for A, then the scale factors for B, rescale the input data, call our <code>solve_factory_scheduling</code> example, and then store the optimal objective value in the following <code>cost</code> matrix:</p><pre><code class="language-julia hljs">cost = zeros(length(scale_factors), length(scale_factors));</code></pre><p>Because we&#39;re modifying <code>factory_df</code> in-place, we need to store the original variable costs in a new column:</p><pre><code class="language-julia hljs">factory_df[!, :old_variable_cost] = copy(factory_df.variable_cost);</code></pre><p>Then, we need a function to scale the <code>:variable_cost</code> column for a particular factory by a value <code>scale</code>:</p><pre><code class="language-julia hljs">function scale_variable_cost(df, factory, scale)
+)</code></pre><img src="57fc90bf.svg" alt="Example block output"/><p>Uh oh, we can&#39;t satisfy all of the demand.</p><h2 id="How-sensitive-is-the-solution-to-changes-in-variable-cost?"><a class="docs-heading-anchor" href="#How-sensitive-is-the-solution-to-changes-in-variable-cost?">How sensitive is the solution to changes in variable cost?</a><a id="How-sensitive-is-the-solution-to-changes-in-variable-cost?-1"></a><a class="docs-heading-anchor-permalink" href="#How-sensitive-is-the-solution-to-changes-in-variable-cost?" title="Permalink"></a></h2><p>Let&#39;s run another experiment, this time seeing how the optimal objective value changes as we vary the variable costs of each factory.</p><p>First though, let&#39;s reset the demand to it&#39;s original level:</p><pre><code class="language-julia hljs">demand_df.demand ./= 1.5;</code></pre><p>For our experiment, we&#39;re going to scale the variable costs of both factories by a set of values from <code>0.0</code> to <code>1.5</code>:</p><pre><code class="language-julia hljs">scale_factors = 0:0.1:1.5</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">0.0:0.1:1.5</code></pre><p>At a high level, we&#39;re going to loop over the scale factors for A, then the scale factors for B, rescale the input data, call our <code>solve_factory_scheduling</code> example, and then store the optimal objective value in the following <code>cost</code> matrix:</p><pre><code class="language-julia hljs">cost = zeros(length(scale_factors), length(scale_factors));</code></pre><p>Because we&#39;re modifying <code>factory_df</code> in-place, we need to store the original variable costs in a new column:</p><pre><code class="language-julia hljs">factory_df[!, :old_variable_cost] = copy(factory_df.variable_cost);</code></pre><p>Then, we need a function to scale the <code>:variable_cost</code> column for a particular factory by a value <code>scale</code>:</p><pre><code class="language-julia hljs">function scale_variable_cost(df, factory, scale)
     rows = df.factory .== factory
     df[rows, :variable_cost] .=
         round.(Int, df[rows, :old_variable_cost] .* scale)
@@ -162,4 +162,4 @@
     cost;
     xlabel = &quot;Scale of factory A&quot;,
     ylabel = &quot;Scale of factory B&quot;,
-)</code></pre><img src="cc63b3c1.svg" alt="Example block output"/><p>What can you infer from the solution?</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The <a href="../../applications/power_systems/#Power-Systems">Power Systems</a> tutorial explains a number of other ways you can structure a problem to perform a parametric analysis of the solution. In particular, you can use in-place modification to reduce the time it takes to build and solve the resulting models.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../cannery/">« The cannery problem</a><a class="docs-footer-nextpage" href="../multi/">The multi-commodity flow problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><img src="f8d71086.svg" alt="Example block output"/><p>What can you infer from the solution?</p><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>The <a href="../../applications/power_systems/#Power-Systems">Power Systems</a> tutorial explains a number of other ways you can structure a problem to perform a parametric analysis of the solution. In particular, you can use in-place modification to reduce the time it takes to build and solve the resulting models.</p></div></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../cannery/">« The cannery problem</a><a class="docs-footer-nextpage" href="../multi/">The multi-commodity flow problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/finance/index.html b/previews/PR3562/tutorials/linear/finance/index.html
index 6d6ca18888d..25a792b320d 100644
--- a/previews/PR3562/tutorials/linear/finance/index.html
+++ b/previews/PR3562/tutorials/linear/finance/index.html
@@ -51,4 +51,4 @@
   1.0
  -0.0
  -0.0
-  0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../facility_location/">« The facility location problem</a><a class="docs-footer-nextpage" href="../geographic_clustering/">Geographical clustering »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../facility_location/">« The facility location problem</a><a class="docs-footer-nextpage" href="../geographic_clustering/">Geographical clustering »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/geographic_clustering/index.html b/previews/PR3562/tutorials/linear/geographic_clustering/index.html
index 655c98d4abd..273891e48c5 100644
--- a/previews/PR3562/tutorials/linear/geographic_clustering/index.html
+++ b/previews/PR3562/tutorials/linear/geographic_clustering/index.html
@@ -134,4 +134,4 @@
    5 │ San Francisco, CA  0.837       37.7749  -122.419      3.0
    6 │ El Paso, TX        0.674       31.7775  -106.442      3.0
 
-sum(group.population) = 9.261000000000001</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../finance/">« Financial modeling problems</a><a class="docs-footer-nextpage" href="../network_flows/">Network flow problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+sum(group.population) = 9.261000000000001</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../finance/">« Financial modeling problems</a><a class="docs-footer-nextpage" href="../network_flows/">Network flow problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/introduction/index.html b/previews/PR3562/tutorials/linear/introduction/index.html
index 9cf09ab5632..844507052b5 100644
--- a/previews/PR3562/tutorials/linear/introduction/index.html
+++ b/previews/PR3562/tutorials/linear/introduction/index.html
@@ -7,4 +7,4 @@
     \min_{x \in \mathbb{R}^n} &amp; \sum\limits_{i=1}^n c_i x_i \\
     \;\;\text{s.t.} &amp; l_j \le \sum\limits_{i=1}^n a_{ij} x_i \le u_j &amp; j = 1 \ldots m \\
     &amp; l_i \le x_i \le u_i &amp; i = 1 \ldots n.
-\end{align}\]</p><p>The most important thing to note is that all terms are of the form <code>coefficient * variable</code>, and that there are no nonlinear terms or multiplications between variables.</p><p>Mixed-integer linear programs (MILPs) are extensions of linear programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>Almost all solvers support linear programs; look for &quot;LP&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>. However, fewer solvers support mixed-integer linear programs. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../diet/#The-diet-problem">The diet problem</a></li><li><a href="../cannery/#The-cannery-problem">The cannery problem</a></li><li><a href="../facility_location/#The-facility-location-problem">The facility location problem</a></li><li><a href="../finance/#Financial-modeling-problems">Financial modeling problems</a></li><li><a href="../network_flows/#Network-flow-problems">Network flow problems</a></li><li><a href="../n-queens/#N-Queens">N-Queens</a></li><li><a href="../sudoku/#Sudoku">Sudoku</a></li></ul></li><li>The <a href="../tips_and_tricks/#linear_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling linear programs. Look here if you are stuck trying to formulate a problem as a linear program.</li><li>The <a href="../lp_sensitivity/#Sensitivity-analysis-of-a-linear-program">Sensitivity analysis of a linear program</a> tutorial explains how to create sensitivity reports like those produced by the Excel Solver.</li><li>The <a href="../callbacks/#callbacks_tutorial">Callbacks</a> tutorial explains how to write a variety of solver-independent callbacks. Look here if you want to write a callback.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../getting_started/performance_tips/">« Performance tips</a><a class="docs-footer-nextpage" href="../knapsack/">The knapsack problem example »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>The most important thing to note is that all terms are of the form <code>coefficient * variable</code>, and that there are no nonlinear terms or multiplications between variables.</p><p>Mixed-integer linear programs (MILPs) are extensions of linear programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>Almost all solvers support linear programs; look for &quot;LP&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>. However, fewer solvers support mixed-integer linear programs. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../diet/#The-diet-problem">The diet problem</a></li><li><a href="../cannery/#The-cannery-problem">The cannery problem</a></li><li><a href="../facility_location/#The-facility-location-problem">The facility location problem</a></li><li><a href="../finance/#Financial-modeling-problems">Financial modeling problems</a></li><li><a href="../network_flows/#Network-flow-problems">Network flow problems</a></li><li><a href="../n-queens/#N-Queens">N-Queens</a></li><li><a href="../sudoku/#Sudoku">Sudoku</a></li></ul></li><li>The <a href="../tips_and_tricks/#linear_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling linear programs. Look here if you are stuck trying to formulate a problem as a linear program.</li><li>The <a href="../lp_sensitivity/#Sensitivity-analysis-of-a-linear-program">Sensitivity analysis of a linear program</a> tutorial explains how to create sensitivity reports like those produced by the Excel Solver.</li><li>The <a href="../callbacks/#callbacks_tutorial">Callbacks</a> tutorial explains how to write a variety of solver-independent callbacks. Look here if you want to write a callback.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../getting_started/performance_tips/">« Performance tips</a><a class="docs-footer-nextpage" href="../knapsack/">The knapsack problem example »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/knapsack/index.html b/previews/PR3562/tutorials/linear/knapsack/index.html
index d87fa381787..e7603a95b03 100644
--- a/previews/PR3562/tutorials/linear/knapsack/index.html
+++ b/previews/PR3562/tutorials/linear/knapsack/index.html
@@ -47,7 +47,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 1.55643e-01
+  Solve time (sec)   : 5.89848e-04
   Simplex iterations : 1
   Barrier iterations : -1
   Node count         : 1
@@ -83,4 +83,4 @@
 solve_knapsack_problem(; profit = profit, weight = weight, capacity = capacity)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">3-element Vector{Int64}:
  1
  4
- 5</code></pre><p>We observe that the chosen items (1, 4, and 5) have the best profit to weight ratio in this particular example.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>Here are some things to try next:</p><ul><li>Call the function with different data. What happens as the capacity increases?</li><li>What happens if the profit and weight vectors are different lengths?</li><li>Instead of creating a binary variable with <code>Bin</code>, we could have written <code>@variable(model, 0 &lt;= x[1:n] &lt;= 1, Int)</code>. Verify that this formulation finds the same solution. What happens if we are allowed to take more than one of each item?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../diet/">The diet problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 5</code></pre><p>We observe that the chosen items (1, 4, and 5) have the best profit to weight ratio in this particular example.</p><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p>Here are some things to try next:</p><ul><li>Call the function with different data. What happens as the capacity increases?</li><li>What happens if the profit and weight vectors are different lengths?</li><li>Instead of creating a binary variable with <code>Bin</code>, we could have written <code>@variable(model, 0 &lt;= x[1:n] &lt;= 1, Int)</code>. Verify that this formulation finds the same solution. What happens if we are allowed to take more than one of each item?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../diet/">The diet problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/lp_sensitivity/index.html b/previews/PR3562/tutorials/linear/lp_sensitivity/index.html
index eef8cf4b9d4..0d58a6e0110 100644
--- a/previews/PR3562/tutorials/linear/lp_sensitivity/index.html
+++ b/previews/PR3562/tutorials/linear/lp_sensitivity/index.html
@@ -39,11 +39,11 @@
     c3 : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 2.84433e-04
+  Solve time (sec)   : 4.01497e-04
   Simplex iterations : 2
   Barrier iterations : 0
   Node count         : -1
-</code></pre><p>Can you identify:</p><ul><li>The objective coefficient of each variable?</li><li>The right-hand side of each constraint?</li><li>The optimal primal and dual solutions?</li></ul><h2 id="Sensitivity-reports"><a class="docs-heading-anchor" href="#Sensitivity-reports">Sensitivity reports</a><a id="Sensitivity-reports-1"></a><a class="docs-heading-anchor-permalink" href="#Sensitivity-reports" title="Permalink"></a></h2><p>Now let&#39;s call <a href="../../../api/JuMP/#JuMP.lp_sensitivity_report"><code>lp_sensitivity_report</code></a>:</p><pre><code class="language-julia hljs">report = lp_sensitivity_report(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">SensitivityReport{Float64}(Dict{ConstraintRef, Tuple{Float64, Float64}}(x ≥ 0 =&gt; (-Inf, 15.0), y ≤ 3 =&gt; (-1.75, Inf), z ≤ 1 =&gt; (-Inf, Inf), c3 : x + y ≤ 20 =&gt; (-3.75, Inf), y ≥ 0 =&gt; (-Inf, 1.25), c1 : 6 x + 8 y ≥ 100 =&gt; (-4.0, 2.857142857142857), c2 : 7 x + 12 y ≥ 120 =&gt; (-3.3333333333333335, 4.666666666666667)), Dict{VariableRef, Tuple{Float64, Float64}}(x =&gt; (-0.3333333333333333, 3.0), z =&gt; (-Inf, 1.0), y =&gt; (-4.0, 0.5714285714285714)))</code></pre><p>It returns a <a href="../../../api/JuMP/#JuMP.SensitivityReport"><code>SensitivityReport</code></a> object, which maps:</p><ul><li>Every variable reference to a tuple <code>(d_lo, d_hi)::Tuple{Float64,Float64}</code>, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.</li><li>Every constraint reference to a tuple <code>(d_lo, d_hi)::Tuple{Float64,Float64}</code>, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.</li></ul><p>Both tuples are relative, rather than absolute. So, given an objective coefficient of <code>1.0</code> and a tuple <code>(-0.5, 0.5)</code>, the objective coefficient can range between <code>1.0 - 0.5</code> an <code>1.0 + 0.5</code>.</p><p>For example:</p><pre><code class="language-julia hljs">report[x]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(-0.3333333333333333, 3.0)</code></pre><p>indicates that the objective coefficient on <code>x</code>, that is, <code>12</code>, can decrease by <code>-0.333</code> or increase by <code>3.0</code> and the primal solution <code>(15, 1.25)</code> will remain optimal. In addition:</p><pre><code class="language-julia hljs">report[c1]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(-4.0, 2.857142857142857)</code></pre><p>means that the right-hand side of the <code>c1</code> constraint (100), can decrease by 4 units, or increase by 2.85 units, and the primal solution <code>(15, 1.25)</code> will remain optimal.</p><h2 id="Variable-sensitivity"><a class="docs-heading-anchor" href="#Variable-sensitivity">Variable sensitivity</a><a id="Variable-sensitivity-1"></a><a class="docs-heading-anchor-permalink" href="#Variable-sensitivity" title="Permalink"></a></h2><p>By themselves, the tuples aren&#39;t informative. Let&#39;s put them in context by collating a range of other information about a variable:</p><pre><code class="language-julia hljs">function variable_report(xi)
+</code></pre><p>Can you identify:</p><ul><li>The objective coefficient of each variable?</li><li>The right-hand side of each constraint?</li><li>The optimal primal and dual solutions?</li></ul><h2 id="Sensitivity-reports"><a class="docs-heading-anchor" href="#Sensitivity-reports">Sensitivity reports</a><a id="Sensitivity-reports-1"></a><a class="docs-heading-anchor-permalink" href="#Sensitivity-reports" title="Permalink"></a></h2><p>Now let&#39;s call <a href="../../../api/JuMP/#JuMP.lp_sensitivity_report"><code>lp_sensitivity_report</code></a>:</p><pre><code class="language-julia hljs">report = lp_sensitivity_report(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">SensitivityReport{Float64}(Dict{ConstraintRef, Tuple{Float64, Float64}}(z ≤ 1 =&gt; (-Inf, Inf), y ≥ 0 =&gt; (-Inf, 1.25), c2 : 7 x + 12 y ≥ 120 =&gt; (-3.3333333333333335, 4.666666666666667), c3 : x + y ≤ 20 =&gt; (-3.75, Inf), y ≤ 3 =&gt; (-1.75, Inf), c1 : 6 x + 8 y ≥ 100 =&gt; (-4.0, 2.857142857142857), x ≥ 0 =&gt; (-Inf, 15.0)), Dict{VariableRef, Tuple{Float64, Float64}}(y =&gt; (-4.0, 0.5714285714285714), x =&gt; (-0.3333333333333333, 3.0), z =&gt; (-Inf, 1.0)))</code></pre><p>It returns a <a href="../../../api/JuMP/#JuMP.SensitivityReport"><code>SensitivityReport</code></a> object, which maps:</p><ul><li>Every variable reference to a tuple <code>(d_lo, d_hi)::Tuple{Float64,Float64}</code>, explaining how much the objective coefficient of the corresponding variable can change by, such that the original basis remains optimal.</li><li>Every constraint reference to a tuple <code>(d_lo, d_hi)::Tuple{Float64,Float64}</code>, explaining how much the right-hand side of the corresponding constraint can change by, such that the basis remains optimal.</li></ul><p>Both tuples are relative, rather than absolute. So, given an objective coefficient of <code>1.0</code> and a tuple <code>(-0.5, 0.5)</code>, the objective coefficient can range between <code>1.0 - 0.5</code> an <code>1.0 + 0.5</code>.</p><p>For example:</p><pre><code class="language-julia hljs">report[x]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(-0.3333333333333333, 3.0)</code></pre><p>indicates that the objective coefficient on <code>x</code>, that is, <code>12</code>, can decrease by <code>-0.333</code> or increase by <code>3.0</code> and the primal solution <code>(15, 1.25)</code> will remain optimal. In addition:</p><pre><code class="language-julia hljs">report[c1]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(-4.0, 2.857142857142857)</code></pre><p>means that the right-hand side of the <code>c1</code> constraint (100), can decrease by 4 units, or increase by 2.85 units, and the primal solution <code>(15, 1.25)</code> will remain optimal.</p><h2 id="Variable-sensitivity"><a class="docs-heading-anchor" href="#Variable-sensitivity">Variable sensitivity</a><a id="Variable-sensitivity-1"></a><a class="docs-heading-anchor-permalink" href="#Variable-sensitivity" title="Permalink"></a></h2><p>By themselves, the tuples aren&#39;t informative. Let&#39;s put them in context by collating a range of other information about a variable:</p><pre><code class="language-julia hljs">function variable_report(xi)
     return (
         name = name(xi),
         lower_bound = has_lower_bound(xi) ? lower_bound(xi) : -Inf,
@@ -70,4 +70,4 @@
 c1_report = constraint_report(c1)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">(name = &quot;c1&quot;, value = 100.0, rhs = 100.0, slack = 0.0, shadow_price = -0.25, allowed_decrease = -4.0, allowed_increase = 2.857142857142857)</code></pre><p>That&#39;s a bit hard to read, so let&#39;s call this on every variable in the model and put things into a DataFrame:</p><pre><code class="language-julia hljs">constraint_df = DataFrames.DataFrame(
     constraint_report(ci) for (F, S) in list_of_constraint_types(model) for
     ci in all_constraints(model, F, S) if F == AffExpr
-)</code></pre><div><div style = "float: left;"><span>3×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">c3</td><td style = "text-align: right;">16.25</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">3.75</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-3.75</td><td style = "text-align: right;">Inf</td></tr></tbody></table></div><h2 id="Analysis-questions"><a class="docs-heading-anchor" href="#Analysis-questions">Analysis questions</a><a id="Analysis-questions-1"></a><a class="docs-heading-anchor-permalink" href="#Analysis-questions" title="Permalink"></a></h2><p>Now we can use these dataframes to ask questions of the solution.</p><p>For example, we can find basic variables by looking for variables with a reduced cost of 0:</p><pre><code class="language-julia hljs">basic = filter(row -&gt; iszero(row.reduced_cost), variable_df)</code></pre><div><div style = "float: left;"><span>2×8 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">lower_bound</th><th style = "text-align: left;">value</th><th style = "text-align: left;">upper_bound</th><th style = "text-align: left;">reduced_cost</th><th style = "text-align: left;">obj_coefficient</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">x</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">Inf</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">-0.333333</td><td style = "text-align: right;">3.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">y</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">1.25</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">0.571429</td></tr></tbody></table></div><p>and non-basic variables by looking for non-zero reduced costs:</p><pre><code class="language-julia hljs">non_basic = filter(row -&gt; !iszero(row.reduced_cost), variable_df)</code></pre><div><div style = "float: left;"><span>1×8 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">lower_bound</th><th style = "text-align: left;">value</th><th style = "text-align: left;">upper_bound</th><th style = "text-align: left;">reduced_cost</th><th style = "text-align: left;">obj_coefficient</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">z</td><td style = "text-align: right;">-Inf</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">-1.0</td><td style = "text-align: right;">-1.0</td><td style = "text-align: right;">-Inf</td><td style = "text-align: right;">1.0</td></tr></tbody></table></div><p>we can also find constraints that are binding by looking for zero slacks:</p><pre><code class="language-julia hljs">binding = filter(row -&gt; iszero(row.slack), constraint_df)</code></pre><div><div style = "float: left;"><span>2×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr></tbody></table></div><p>or non-zero shadow prices:</p><pre><code class="language-julia hljs">binding2 = filter(row -&gt; !iszero(row.shadow_price), constraint_df)</code></pre><div><div style = "float: left;"><span>2×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr></tbody></table></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Callbacks</a><a class="docs-footer-nextpage" href="../mip_duality/">Computing the duals of a mixed-integer program »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><div><div style = "float: left;"><span>3×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">c3</td><td style = "text-align: right;">16.25</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">3.75</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-3.75</td><td style = "text-align: right;">Inf</td></tr></tbody></table></div><h2 id="Analysis-questions"><a class="docs-heading-anchor" href="#Analysis-questions">Analysis questions</a><a id="Analysis-questions-1"></a><a class="docs-heading-anchor-permalink" href="#Analysis-questions" title="Permalink"></a></h2><p>Now we can use these dataframes to ask questions of the solution.</p><p>For example, we can find basic variables by looking for variables with a reduced cost of 0:</p><pre><code class="language-julia hljs">basic = filter(row -&gt; iszero(row.reduced_cost), variable_df)</code></pre><div><div style = "float: left;"><span>2×8 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">lower_bound</th><th style = "text-align: left;">value</th><th style = "text-align: left;">upper_bound</th><th style = "text-align: left;">reduced_cost</th><th style = "text-align: left;">obj_coefficient</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">x</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">15.0</td><td style = "text-align: right;">Inf</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">12.0</td><td style = "text-align: right;">-0.333333</td><td style = "text-align: right;">3.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">y</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">1.25</td><td style = "text-align: right;">3.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">20.0</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">0.571429</td></tr></tbody></table></div><p>and non-basic variables by looking for non-zero reduced costs:</p><pre><code class="language-julia hljs">non_basic = filter(row -&gt; !iszero(row.reduced_cost), variable_df)</code></pre><div><div style = "float: left;"><span>1×8 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">lower_bound</th><th style = "text-align: left;">value</th><th style = "text-align: left;">upper_bound</th><th style = "text-align: left;">reduced_cost</th><th style = "text-align: left;">obj_coefficient</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">z</td><td style = "text-align: right;">-Inf</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">1.0</td><td style = "text-align: right;">-1.0</td><td style = "text-align: right;">-1.0</td><td style = "text-align: right;">-Inf</td><td style = "text-align: right;">1.0</td></tr></tbody></table></div><p>we can also find constraints that are binding by looking for zero slacks:</p><pre><code class="language-julia hljs">binding = filter(row -&gt; iszero(row.slack), constraint_df)</code></pre><div><div style = "float: left;"><span>2×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr></tbody></table></div><p>or non-zero shadow prices:</p><pre><code class="language-julia hljs">binding2 = filter(row -&gt; !iszero(row.shadow_price), constraint_df)</code></pre><div><div style = "float: left;"><span>2×7 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">name</th><th style = "text-align: left;">value</th><th style = "text-align: left;">rhs</th><th style = "text-align: left;">slack</th><th style = "text-align: left;">shadow_price</th><th style = "text-align: left;">allowed_decrease</th><th style = "text-align: left;">allowed_increase</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">c1</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">100.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-0.25</td><td style = "text-align: right;">-4.0</td><td style = "text-align: right;">2.85714</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">c2</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">120.0</td><td style = "text-align: right;">0.0</td><td style = "text-align: right;">-1.5</td><td style = "text-align: right;">-3.33333</td><td style = "text-align: right;">4.66667</td></tr></tbody></table></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../callbacks/">« Callbacks</a><a class="docs-footer-nextpage" href="../mip_duality/">Computing the duals of a mixed-integer program »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/mip_duality/index.html b/previews/PR3562/tutorials/linear/mip_duality/index.html
index 816ed56b1a1..62fa7fe4443 100644
--- a/previews/PR3562/tutorials/linear/mip_duality/index.html
+++ b/previews/PR3562/tutorials/linear/mip_duality/index.html
@@ -109,4 +109,4 @@
  g[2] ≤ 1000
  w ≤ 200
  dispatch[1] binary
- dispatch[2] binary</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../lp_sensitivity/">« Sensitivity analysis of a linear program</a><a class="docs-footer-nextpage" href="../../nonlinear/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ dispatch[2] binary</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../lp_sensitivity/">« Sensitivity analysis of a linear program</a><a class="docs-footer-nextpage" href="../../nonlinear/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/multi/index.html b/previews/PR3562/tutorials/linear/multi/index.html
index dbbca7e8477..9b3f0e5bdde 100644
--- a/previews/PR3562/tutorials/linear/multi/index.html
+++ b/previews/PR3562/tutorials/linear/multi/index.html
@@ -117,7 +117,7 @@
     INNER JOIN locations b
     ON a.type = &#39;origin&#39; AND b.type = &#39;destination&#39;
     &quot;&quot;&quot;,
-)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">SQLite.Query{false}(SQLite.Stmt(SQLite.DB(&quot;/home/runner/work/JuMP.jl/JuMP.jl/docs/build/tutorials/linear/multi.sqlite&quot;), Base.RefValue{Ptr{SQLite.C.sqlite3_stmt}}(Ptr{SQLite.C.sqlite3_stmt} @0x0000000040bf0028), Dict{Int64, Any}()), Base.RefValue{Int32}(100), [:origin, :destination], Type[Union{Missing, String}, Union{Missing, String}], Dict(:origin =&gt; 1, :destination =&gt; 2), Base.RefValue{Int64}(0))</code></pre><p>With a constraint that we cannot send more than 625 units between each pair:</p><pre><code class="language-julia hljs">for r in Tables.rows(od_pairs)
+)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">SQLite.Query{false}(SQLite.Stmt(SQLite.DB(&quot;/home/runner/work/JuMP.jl/JuMP.jl/docs/build/tutorials/linear/multi.sqlite&quot;), Base.RefValue{Ptr{SQLite.C.sqlite3_stmt}}(Ptr{SQLite.C.sqlite3_stmt} @0x0000000038198d28), Dict{Int64, Any}()), Base.RefValue{Int32}(100), [:origin, :destination], Type[Union{Missing, String}, Union{Missing, String}], Dict(:origin =&gt; 1, :destination =&gt; 2), Base.RefValue{Int64}(0))</code></pre><p>With a constraint that we cannot send more than 625 units between each pair:</p><pre><code class="language-julia hljs">for r in Tables.rows(od_pairs)
     @constraint(model, sum(x[r.origin, r.destination, :]) &lt;= 625)
 end</code></pre><h2 id="Solution"><a class="docs-heading-anchor" href="#Solution">Solution</a><a id="Solution-1"></a><a class="docs-heading-anchor-permalink" href="#Solution" title="Permalink"></a></h2><p>Finally, we can optimize the model:</p><pre><code class="language-julia hljs">optimize!(model)
 solution_summary(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">* Solver : HiGHS
@@ -137,7 +137,7 @@
   Dual objective value : 2.25700e+05
 
 * Work counters
-  Solve time (sec)   : 6.59466e-04
+  Solve time (sec)   : 9.35793e-04
   Simplex iterations : 54
   Barrier iterations : 0
   Node count         : -1
@@ -168,4 +168,4 @@
 PITT WIN    75   250    .
 PITT STL   400    25   200
 PITT FRE    .    450   100
-PITT LAF   250   125    .</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../factory_schedule/">« The factory schedule example</a><a class="docs-footer-nextpage" href="../multi_commodity_network/">The network multi-commodity flow problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+PITT LAF   250   125    .</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../factory_schedule/">« The factory schedule example</a><a class="docs-footer-nextpage" href="../multi_commodity_network/">The network multi-commodity flow problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/multi_commodity_network/index.html b/previews/PR3562/tutorials/linear/multi_commodity_network/index.html
index 4e1ef55a278..7eeaa56a62d 100644
--- a/previews/PR3562/tutorials/linear/multi_commodity_network/index.html
+++ b/previews/PR3562/tutorials/linear/multi_commodity_network/index.html
@@ -85,7 +85,7 @@
   Dual objective value : 1.43228e+02
 
 * Work counters
-  Solve time (sec)   : 3.07560e-04
+  Solve time (sec)   : 3.99113e-04
   Simplex iterations : 8
   Barrier iterations : 0
   Node count         : -1
@@ -93,4 +93,4 @@
 df_supply.x_supply = value.(df_supply.x_supply);</code></pre><p>and display the optimal solution for flows:</p><pre><code class="language-julia hljs">DataFrames.select(
     filter!(row -&gt; row.x_flow &gt; 0.0, df_shipping),
     [:origin, :destination, :product, :x_flow],
-)</code></pre><div><div style = "float: left;"><span>9×4 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">origin</th><th style = "text-align: left;">destination</th><th style = "text-align: left;">product</th><th style = "text-align: left;">x_flow</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "String" style = "text-align: left;">String</th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">10.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">wellington</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">christchurch</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">4.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">christchurch</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">4.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">20.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">wellington</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">22.0</td></tr></tbody></table></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi/">« The multi-commodity flow problem</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><div><div style = "float: left;"><span>9×4 DataFrame</span></div><div style = "clear: both;"></div></div><div class = "data-frame" style = "overflow-x: scroll;"><table class = "data-frame" style = "margin-bottom: 6px;"><thead><tr class = "header"><th class = "rowNumber" style = "font-weight: bold; text-align: right;">Row</th><th style = "text-align: left;">origin</th><th style = "text-align: left;">destination</th><th style = "text-align: left;">product</th><th style = "text-align: left;">x_flow</th></tr><tr class = "subheader headerLastRow"><th class = "rowNumber" style = "font-weight: bold; text-align: right;"></th><th title = "String" style = "text-align: left;">String</th><th title = "String" style = "text-align: left;">String</th><th title = "String" style = "text-align: left;">String</th><th title = "Float64" style = "text-align: left;">Float64</th></tr></thead><tbody><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">1</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">10.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">2</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">wellington</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">3</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">4</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">5</td><td style = "text-align: left;">christchurch</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">milk</td><td style = "text-align: right;">4.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">6</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">christchurch</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">4.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">7</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">auckland</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">20.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">8</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">wellington</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">2.0</td></tr><tr><td class = "rowNumber" style = "font-weight: bold; text-align: right;">9</td><td style = "text-align: left;">tauranga</td><td style = "text-align: left;">waikato</td><td style = "text-align: left;">kiwifruit</td><td style = "text-align: right;">22.0</td></tr></tbody></table></div></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi/">« The multi-commodity flow problem</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/multi_objective_examples/index.html b/previews/PR3562/tutorials/linear/multi_objective_examples/index.html
index d0c0b2e7a76..086760c1a47 100644
--- a/previews/PR3562/tutorials/linear/multi_objective_examples/index.html
+++ b/previews/PR3562/tutorials/linear/multi_objective_examples/index.html
@@ -31,7 +31,7 @@
   Dual objective value : -9.00000e+00
 
 * Work counters
-  Solve time (sec)   : 9.59873e-04
+  Solve time (sec)   : 1.19400e-03
   Simplex iterations : 1
   Barrier iterations : 0
   Node count         : -1
@@ -67,7 +67,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 2.94809e-01
+  Solve time (sec)   : 8.98790e-03
   Simplex iterations : 5
   Barrier iterations : -1
   Node count         : 1
@@ -123,7 +123,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 2.68528e-02
+  Solve time (sec)   : 5.01800e-03
   Simplex iterations : 4
   Barrier iterations : -1
   Node count         : 1
@@ -139,4 +139,4 @@
 end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">1: z = [8, 9] | Path: 1-&gt;2 2-&gt;4 4-&gt;6
 2: z = [10, 7] | Path: 1-&gt;2 2-&gt;5 5-&gt;6
 3: z = [11, 5] | Path: 1-&gt;2 2-&gt;6
-4: z = [13, 4] | Path: 1-&gt;3 3-&gt;4 4-&gt;6</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_objective_knapsack/">« Multi-objective knapsack</a><a class="docs-footer-nextpage" href="../sudoku/">Sudoku »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+4: z = [13, 4] | Path: 1-&gt;3 3-&gt;4 4-&gt;6</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_objective_knapsack/">« Multi-objective knapsack</a><a class="docs-footer-nextpage" href="../sudoku/">Sudoku »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/multi_objective_knapsack/1f70cf5d.svg b/previews/PR3562/tutorials/linear/multi_objective_knapsack/ba5f521f.svg
similarity index 87%
rename from previews/PR3562/tutorials/linear/multi_objective_knapsack/1f70cf5d.svg
rename to previews/PR3562/tutorials/linear/multi_objective_knapsack/ba5f521f.svg
index 7380a16b1ca..62519465f0d 100644
--- a/previews/PR3562/tutorials/linear/multi_objective_knapsack/1f70cf5d.svg
+++ b/previews/PR3562/tutorials/linear/multi_objective_knapsack/ba5f521f.svg
@@ -1,53 +1,53 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip030">
+  <clipPath id="clip850">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip030)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip850)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip031">
+  <clipPath id="clip851">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip030)" d="M234.612 1423.18 L2352.76 1423.18 L2352.76 123.472 L234.612 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip850)" d="M234.612 1423.18 L2352.76 1423.18 L2352.76 123.472 L234.612 123.472  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip032">
+  <clipPath id="clip852">
     <rect x="234" y="123" width="2119" height="1301"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="469.961,1423.18 469.961,123.472 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="940.66,1423.18 940.66,123.472 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1411.36,1423.18 1411.36,123.472 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1882.06,1423.18 1882.06,123.472 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2352.76,1423.18 2352.76,123.472 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="469.961,1423.18 469.961,1404.28 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="940.66,1423.18 940.66,1404.28 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1411.36,1423.18 1411.36,1404.28 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1882.06,1423.18 1882.06,1404.28 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2352.76,1423.18 2352.76,1404.28 "/>
-<path clip-path="url(#clip030)" d="M430.007 1484.86 L430.007 1480.6 Q431.767 1481.44 433.572 1481.88 Q435.378 1482.32 437.114 1482.32 Q441.743 1482.32 444.174 1479.21 Q446.628 1476.09 446.975 1469.75 Q445.632 1471.74 443.572 1472.8 Q441.512 1473.87 439.012 1473.87 Q433.827 1473.87 430.794 1470.74 Q427.785 1467.59 427.785 1462.15 Q427.785 1456.83 430.933 1453.61 Q434.081 1450.39 439.313 1450.39 Q445.308 1450.39 448.456 1455 Q451.628 1459.58 451.628 1468.33 Q451.628 1476.51 447.739 1481.39 Q443.873 1486.25 437.322 1486.25 Q435.563 1486.25 433.757 1485.9 Q431.952 1485.56 430.007 1484.86 M439.313 1470.21 Q442.461 1470.21 444.29 1468.06 Q446.142 1465.9 446.142 1462.15 Q446.142 1458.43 444.29 1456.27 Q442.461 1454.1 439.313 1454.1 Q436.165 1454.1 434.313 1456.27 Q432.484 1458.43 432.484 1462.15 Q432.484 1465.9 434.313 1468.06 Q436.165 1470.21 439.313 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M464.058 1481.64 L480.378 1481.64 L480.378 1485.58 L458.433 1485.58 L458.433 1481.64 Q461.095 1478.89 465.679 1474.26 Q470.285 1469.61 471.466 1468.27 Q473.711 1465.74 474.591 1464.01 Q475.493 1462.25 475.493 1460.56 Q475.493 1457.8 473.549 1456.07 Q471.628 1454.33 468.526 1454.33 Q466.327 1454.33 463.873 1455.09 Q461.442 1455.86 458.665 1457.41 L458.665 1452.69 Q461.489 1451.55 463.942 1450.97 Q466.396 1450.39 468.433 1450.39 Q473.804 1450.39 476.998 1453.08 Q480.192 1455.77 480.192 1460.26 Q480.192 1462.39 479.382 1464.31 Q478.595 1466.2 476.489 1468.8 Q475.91 1469.47 472.808 1472.69 Q469.706 1475.88 464.058 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M500.192 1454.1 Q496.581 1454.1 494.752 1457.66 Q492.947 1461.2 492.947 1468.33 Q492.947 1475.44 494.752 1479.01 Q496.581 1482.55 500.192 1482.55 Q503.826 1482.55 505.632 1479.01 Q507.461 1475.44 507.461 1468.33 Q507.461 1461.2 505.632 1457.66 Q503.826 1454.1 500.192 1454.1 M500.192 1450.39 Q506.002 1450.39 509.058 1455 Q512.137 1459.58 512.137 1468.33 Q512.137 1477.06 509.058 1481.67 Q506.002 1486.25 500.192 1486.25 Q494.382 1486.25 491.303 1481.67 Q488.248 1477.06 488.248 1468.33 Q488.248 1459.58 491.303 1455 Q494.382 1450.39 500.192 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M900.706 1484.86 L900.706 1480.6 Q902.465 1481.44 904.271 1481.88 Q906.077 1482.32 907.813 1482.32 Q912.442 1482.32 914.873 1479.21 Q917.326 1476.09 917.674 1469.75 Q916.331 1471.74 914.271 1472.8 Q912.211 1473.87 909.711 1473.87 Q904.526 1473.87 901.493 1470.74 Q898.484 1467.59 898.484 1462.15 Q898.484 1456.83 901.632 1453.61 Q904.78 1450.39 910.012 1450.39 Q916.007 1450.39 919.155 1455 Q922.326 1459.58 922.326 1468.33 Q922.326 1476.51 918.438 1481.39 Q914.572 1486.25 908.021 1486.25 Q906.262 1486.25 904.456 1485.9 Q902.651 1485.56 900.706 1484.86 M910.012 1470.21 Q913.16 1470.21 914.989 1468.06 Q916.84 1465.9 916.84 1462.15 Q916.84 1458.43 914.989 1456.27 Q913.16 1454.1 910.012 1454.1 Q906.864 1454.1 905.012 1456.27 Q903.183 1458.43 903.183 1462.15 Q903.183 1465.9 905.012 1468.06 Q906.864 1470.21 910.012 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M944.896 1466.95 Q948.252 1467.66 950.127 1469.93 Q952.025 1472.2 952.025 1475.53 Q952.025 1480.65 948.507 1483.45 Q944.988 1486.25 938.507 1486.25 Q936.331 1486.25 934.016 1485.81 Q931.725 1485.39 929.271 1484.54 L929.271 1480.02 Q931.215 1481.16 933.53 1481.74 Q935.845 1482.32 938.368 1482.32 Q942.766 1482.32 945.058 1480.58 Q947.373 1478.84 947.373 1475.53 Q947.373 1472.48 945.22 1470.77 Q943.09 1469.03 939.271 1469.03 L935.243 1469.03 L935.243 1465.19 L939.456 1465.19 Q942.905 1465.19 944.734 1463.82 Q946.562 1462.43 946.562 1459.84 Q946.562 1457.18 944.664 1455.77 Q942.789 1454.33 939.271 1454.33 Q937.35 1454.33 935.15 1454.75 Q932.951 1455.16 930.313 1456.04 L930.313 1451.88 Q932.975 1451.14 935.289 1450.77 Q937.627 1450.39 939.687 1450.39 Q945.012 1450.39 948.113 1452.83 Q951.215 1455.23 951.215 1459.35 Q951.215 1462.22 949.572 1464.21 Q947.928 1466.18 944.896 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M970.891 1454.1 Q967.28 1454.1 965.451 1457.66 Q963.646 1461.2 963.646 1468.33 Q963.646 1475.44 965.451 1479.01 Q967.28 1482.55 970.891 1482.55 Q974.525 1482.55 976.331 1479.01 Q978.159 1475.44 978.159 1468.33 Q978.159 1461.2 976.331 1457.66 Q974.525 1454.1 970.891 1454.1 M970.891 1450.39 Q976.701 1450.39 979.757 1455 Q982.835 1459.58 982.835 1468.33 Q982.835 1477.06 979.757 1481.67 Q976.701 1486.25 970.891 1486.25 Q965.081 1486.25 962.002 1481.67 Q958.947 1477.06 958.947 1468.33 Q958.947 1459.58 962.002 1455 Q965.081 1450.39 970.891 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1371.4 1484.86 L1371.4 1480.6 Q1373.16 1481.44 1374.97 1481.88 Q1376.78 1482.32 1378.51 1482.32 Q1383.14 1482.32 1385.57 1479.21 Q1388.03 1476.09 1388.37 1469.75 Q1387.03 1471.74 1384.97 1472.8 Q1382.91 1473.87 1380.41 1473.87 Q1375.22 1473.87 1372.19 1470.74 Q1369.18 1467.59 1369.18 1462.15 Q1369.18 1456.83 1372.33 1453.61 Q1375.48 1450.39 1380.71 1450.39 Q1386.71 1450.39 1389.85 1455 Q1393.03 1459.58 1393.03 1468.33 Q1393.03 1476.51 1389.14 1481.39 Q1385.27 1486.25 1378.72 1486.25 Q1376.96 1486.25 1375.15 1485.9 Q1373.35 1485.56 1371.4 1484.86 M1380.71 1470.21 Q1383.86 1470.21 1385.69 1468.06 Q1387.54 1465.9 1387.54 1462.15 Q1387.54 1458.43 1385.69 1456.27 Q1383.86 1454.1 1380.71 1454.1 Q1377.56 1454.1 1375.71 1456.27 Q1373.88 1458.43 1373.88 1462.15 Q1373.88 1465.9 1375.71 1468.06 Q1377.56 1470.21 1380.71 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1414.28 1455.09 L1402.47 1473.54 L1414.28 1473.54 L1414.28 1455.09 M1413.05 1451.02 L1418.93 1451.02 L1418.93 1473.54 L1423.86 1473.54 L1423.86 1477.43 L1418.93 1477.43 L1418.93 1485.58 L1414.28 1485.58 L1414.28 1477.43 L1398.67 1477.43 L1398.67 1472.92 L1413.05 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1441.59 1454.1 Q1437.98 1454.1 1436.15 1457.66 Q1434.34 1461.2 1434.34 1468.33 Q1434.34 1475.44 1436.15 1479.01 Q1437.98 1482.55 1441.59 1482.55 Q1445.22 1482.55 1447.03 1479.01 Q1448.86 1475.44 1448.86 1468.33 Q1448.86 1461.2 1447.03 1457.66 Q1445.22 1454.1 1441.59 1454.1 M1441.59 1450.39 Q1447.4 1450.39 1450.46 1455 Q1453.53 1459.58 1453.53 1468.33 Q1453.53 1477.06 1450.46 1481.67 Q1447.4 1486.25 1441.59 1486.25 Q1435.78 1486.25 1432.7 1481.67 Q1429.65 1477.06 1429.65 1468.33 Q1429.65 1459.58 1432.7 1455 Q1435.78 1450.39 1441.59 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1842.1 1484.86 L1842.1 1480.6 Q1843.86 1481.44 1845.67 1481.88 Q1847.47 1482.32 1849.21 1482.32 Q1853.84 1482.32 1856.27 1479.21 Q1858.72 1476.09 1859.07 1469.75 Q1857.73 1471.74 1855.67 1472.8 Q1853.61 1473.87 1851.11 1473.87 Q1845.92 1473.87 1842.89 1470.74 Q1839.88 1467.59 1839.88 1462.15 Q1839.88 1456.83 1843.03 1453.61 Q1846.18 1450.39 1851.41 1450.39 Q1857.4 1450.39 1860.55 1455 Q1863.72 1459.58 1863.72 1468.33 Q1863.72 1476.51 1859.84 1481.39 Q1855.97 1486.25 1849.42 1486.25 Q1847.66 1486.25 1845.85 1485.9 Q1844.05 1485.56 1842.1 1484.86 M1851.41 1470.21 Q1854.56 1470.21 1856.39 1468.06 Q1858.24 1465.9 1858.24 1462.15 Q1858.24 1458.43 1856.39 1456.27 Q1854.56 1454.1 1851.41 1454.1 Q1848.26 1454.1 1846.41 1456.27 Q1844.58 1458.43 1844.58 1462.15 Q1844.58 1465.9 1846.41 1468.06 Q1848.26 1470.21 1851.41 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1872.17 1451.02 L1890.53 1451.02 L1890.53 1454.96 L1876.46 1454.96 L1876.46 1463.43 Q1877.47 1463.08 1878.49 1462.92 Q1879.51 1462.73 1880.53 1462.73 Q1886.32 1462.73 1889.7 1465.9 Q1893.08 1469.08 1893.08 1474.49 Q1893.08 1480.07 1889.6 1483.17 Q1886.13 1486.25 1879.81 1486.25 Q1877.64 1486.25 1875.37 1485.88 Q1873.12 1485.51 1870.71 1484.77 L1870.71 1480.07 Q1872.8 1481.2 1875.02 1481.76 Q1877.24 1482.32 1879.72 1482.32 Q1883.72 1482.32 1886.06 1480.21 Q1888.4 1478.1 1888.4 1474.49 Q1888.4 1470.88 1886.06 1468.77 Q1883.72 1466.67 1879.72 1466.67 Q1877.84 1466.67 1875.97 1467.08 Q1874.12 1467.5 1872.17 1468.38 L1872.17 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1912.29 1454.1 Q1908.68 1454.1 1906.85 1457.66 Q1905.04 1461.2 1905.04 1468.33 Q1905.04 1475.44 1906.85 1479.01 Q1908.68 1482.55 1912.29 1482.55 Q1915.92 1482.55 1917.73 1479.01 Q1919.56 1475.44 1919.56 1468.33 Q1919.56 1461.2 1917.73 1457.66 Q1915.92 1454.1 1912.29 1454.1 M1912.29 1450.39 Q1918.1 1450.39 1921.15 1455 Q1924.23 1459.58 1924.23 1468.33 Q1924.23 1477.06 1921.15 1481.67 Q1918.1 1486.25 1912.29 1486.25 Q1906.48 1486.25 1903.4 1481.67 Q1900.34 1477.06 1900.34 1468.33 Q1900.34 1459.58 1903.4 1455 Q1906.48 1450.39 1912.29 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2312.8 1484.86 L2312.8 1480.6 Q2314.56 1481.44 2316.37 1481.88 Q2318.17 1482.32 2319.91 1482.32 Q2324.54 1482.32 2326.97 1479.21 Q2329.42 1476.09 2329.77 1469.75 Q2328.43 1471.74 2326.37 1472.8 Q2324.31 1473.87 2321.81 1473.87 Q2316.62 1473.87 2313.59 1470.74 Q2310.58 1467.59 2310.58 1462.15 Q2310.58 1456.83 2313.73 1453.61 Q2316.88 1450.39 2322.11 1450.39 Q2328.1 1450.39 2331.25 1455 Q2334.42 1459.58 2334.42 1468.33 Q2334.42 1476.51 2330.53 1481.39 Q2326.67 1486.25 2320.12 1486.25 Q2318.36 1486.25 2316.55 1485.9 Q2314.75 1485.56 2312.8 1484.86 M2322.11 1470.21 Q2325.26 1470.21 2327.08 1468.06 Q2328.94 1465.9 2328.94 1462.15 Q2328.94 1458.43 2327.08 1456.27 Q2325.26 1454.1 2322.11 1454.1 Q2318.96 1454.1 2317.11 1456.27 Q2315.28 1458.43 2315.28 1462.15 Q2315.28 1465.9 2317.11 1468.06 Q2318.96 1470.21 2322.11 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2353.4 1466.44 Q2350.26 1466.44 2348.4 1468.59 Q2346.58 1470.74 2346.58 1474.49 Q2346.58 1478.22 2348.4 1480.39 Q2350.26 1482.55 2353.4 1482.55 Q2356.55 1482.55 2358.38 1480.39 Q2360.23 1478.22 2360.23 1474.49 Q2360.23 1470.74 2358.38 1468.59 Q2356.55 1466.44 2353.4 1466.44 M2362.69 1451.78 L2362.69 1456.04 Q2360.93 1455.21 2359.12 1454.77 Q2357.34 1454.33 2355.58 1454.33 Q2350.95 1454.33 2348.5 1457.45 Q2346.07 1460.58 2345.72 1466.9 Q2347.08 1464.89 2349.14 1463.82 Q2351.2 1462.73 2353.68 1462.73 Q2358.89 1462.73 2361.9 1465.9 Q2364.93 1469.05 2364.93 1474.49 Q2364.93 1479.82 2361.78 1483.03 Q2358.64 1486.25 2353.4 1486.25 Q2347.41 1486.25 2344.24 1481.67 Q2341.07 1477.06 2341.07 1468.33 Q2341.07 1460.14 2344.96 1455.28 Q2348.84 1450.39 2355.39 1450.39 Q2357.15 1450.39 2358.94 1450.74 Q2360.74 1451.09 2362.69 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2382.99 1454.1 Q2379.38 1454.1 2377.55 1457.66 Q2375.74 1461.2 2375.74 1468.33 Q2375.74 1475.44 2377.55 1479.01 Q2379.38 1482.55 2382.99 1482.55 Q2386.62 1482.55 2388.43 1479.01 Q2390.26 1475.44 2390.26 1468.33 Q2390.26 1461.2 2388.43 1457.66 Q2386.62 1454.1 2382.99 1454.1 M2382.99 1450.39 Q2388.8 1450.39 2391.85 1455 Q2394.93 1459.58 2394.93 1468.33 Q2394.93 1477.06 2391.85 1481.67 Q2388.8 1486.25 2382.99 1486.25 Q2377.18 1486.25 2374.1 1481.67 Q2371.04 1477.06 2371.04 1468.33 Q2371.04 1459.58 2374.1 1455 Q2377.18 1450.39 2382.99 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1219.08 1525.81 L1219.08 1543.66 L1227.16 1543.66 Q1231.65 1543.66 1234.1 1541.34 Q1236.55 1539.02 1236.55 1534.72 Q1236.55 1530.45 1234.1 1528.13 Q1231.65 1525.81 1227.16 1525.81 L1219.08 1525.81 M1212.65 1520.52 L1227.16 1520.52 Q1235.15 1520.52 1239.23 1524.15 Q1243.33 1527.75 1243.33 1534.72 Q1243.33 1541.75 1239.23 1545.35 Q1235.15 1548.95 1227.16 1548.95 L1219.08 1548.95 L1219.08 1568.04 L1212.65 1568.04 L1212.65 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1271.21 1537.87 Q1270.23 1537.3 1269.05 1537.04 Q1267.9 1536.76 1266.5 1536.76 Q1261.54 1536.76 1258.86 1540 Q1256.22 1543.22 1256.22 1549.27 L1256.22 1568.04 L1250.33 1568.04 L1250.33 1532.4 L1256.22 1532.4 L1256.22 1537.93 Q1258.07 1534.69 1261.03 1533.13 Q1263.99 1531.54 1268.22 1531.54 Q1268.83 1531.54 1269.56 1531.63 Q1270.29 1531.7 1271.18 1531.85 L1271.21 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1289.74 1536.5 Q1285.03 1536.5 1282.29 1540.19 Q1279.55 1543.85 1279.55 1550.25 Q1279.55 1556.65 1282.26 1560.34 Q1284.99 1564 1289.74 1564 Q1294.42 1564 1297.15 1560.31 Q1299.89 1556.62 1299.89 1550.25 Q1299.89 1543.92 1297.15 1540.23 Q1294.42 1536.5 1289.74 1536.5 M1289.74 1531.54 Q1297.38 1531.54 1301.74 1536.5 Q1306.1 1541.47 1306.1 1550.25 Q1306.1 1559 1301.74 1564 Q1297.38 1568.97 1289.74 1568.97 Q1282.07 1568.97 1277.71 1564 Q1273.38 1559 1273.38 1550.25 Q1273.38 1541.47 1277.71 1536.5 Q1282.07 1531.54 1289.74 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1333.85 1518.52 L1333.85 1523.39 L1328.25 1523.39 Q1325.1 1523.39 1323.86 1524.66 Q1322.65 1525.93 1322.65 1529.24 L1322.65 1532.4 L1332.29 1532.4 L1332.29 1536.95 L1322.65 1536.95 L1322.65 1568.04 L1316.76 1568.04 L1316.76 1536.95 L1311.16 1536.95 L1311.16 1532.4 L1316.76 1532.4 L1316.76 1529.91 Q1316.76 1523.96 1319.53 1521.26 Q1322.3 1518.52 1328.31 1518.52 L1333.85 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1338.75 1532.4 L1344.61 1532.4 L1344.61 1568.04 L1338.75 1568.04 L1338.75 1532.4 M1338.75 1518.52 L1344.61 1518.52 L1344.61 1525.93 L1338.75 1525.93 L1338.75 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1362.66 1522.27 L1362.66 1532.4 L1374.72 1532.4 L1374.72 1536.95 L1362.66 1536.95 L1362.66 1556.3 Q1362.66 1560.66 1363.83 1561.9 Q1365.04 1563.14 1368.7 1563.14 L1374.72 1563.14 L1374.72 1568.04 L1368.7 1568.04 Q1361.92 1568.04 1359.35 1565.53 Q1356.77 1562.98 1356.77 1556.3 L1356.77 1536.95 L1352.47 1536.95 L1352.47 1532.4 L1356.77 1532.4 L1356.77 1522.27 L1362.66 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,1163.46 2352.76,1163.46 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,844.983 2352.76,844.983 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,526.506 2352.76,526.506 "/>
-<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,208.028 2352.76,208.028 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1423.18 234.612,123.472 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1163.46 253.509,1163.46 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,844.983 253.509,844.983 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,526.506 253.509,526.506 "/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,208.028 253.509,208.028 "/>
-<path clip-path="url(#clip030)" d="M116.482 1180.02 L116.482 1175.76 Q118.242 1176.6 120.047 1177.04 Q121.853 1177.48 123.589 1177.48 Q128.218 1177.48 130.649 1174.38 Q133.103 1171.25 133.45 1164.91 Q132.107 1166.9 130.047 1167.96 Q127.987 1169.03 125.487 1169.03 Q120.302 1169.03 117.269 1165.9 Q114.26 1162.76 114.26 1157.32 Q114.26 1151.99 117.408 1148.77 Q120.556 1145.56 125.788 1145.56 Q131.783 1145.56 134.931 1150.16 Q138.103 1154.75 138.103 1163.5 Q138.103 1171.67 134.214 1176.55 Q130.348 1181.41 123.797 1181.41 Q122.038 1181.41 120.232 1181.07 Q118.427 1180.72 116.482 1180.02 M125.788 1165.37 Q128.936 1165.37 130.765 1163.22 Q132.617 1161.07 132.617 1157.32 Q132.617 1153.59 130.765 1151.44 Q128.936 1149.26 125.788 1149.26 Q122.64 1149.26 120.788 1151.44 Q118.959 1153.59 118.959 1157.32 Q118.959 1161.07 120.788 1163.22 Q122.64 1165.37 125.788 1165.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M150.533 1176.81 L166.852 1176.81 L166.852 1180.74 L144.908 1180.74 L144.908 1176.81 Q147.57 1174.05 152.153 1169.42 Q156.76 1164.77 157.94 1163.43 Q160.186 1160.9 161.065 1159.17 Q161.968 1157.41 161.968 1155.72 Q161.968 1152.96 160.024 1151.23 Q158.103 1149.49 155.001 1149.49 Q152.802 1149.49 150.348 1150.26 Q147.917 1151.02 145.14 1152.57 L145.14 1147.85 Q147.964 1146.71 150.417 1146.13 Q152.871 1145.56 154.908 1145.56 Q160.278 1145.56 163.473 1148.24 Q166.667 1150.93 166.667 1155.42 Q166.667 1157.55 165.857 1159.47 Q165.07 1161.37 162.964 1163.96 Q162.385 1164.63 159.283 1167.85 Q156.181 1171.04 150.533 1176.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M186.667 1149.26 Q183.056 1149.26 181.227 1152.82 Q179.422 1156.37 179.422 1163.5 Q179.422 1170.6 181.227 1174.17 Q183.056 1177.71 186.667 1177.71 Q190.301 1177.71 192.107 1174.17 Q193.936 1170.6 193.936 1163.5 Q193.936 1156.37 192.107 1152.82 Q190.301 1149.26 186.667 1149.26 M186.667 1145.56 Q192.477 1145.56 195.533 1150.16 Q198.612 1154.75 198.612 1163.5 Q198.612 1172.22 195.533 1176.83 Q192.477 1181.41 186.667 1181.41 Q180.857 1181.41 177.778 1176.83 Q174.723 1172.22 174.723 1163.5 Q174.723 1154.75 177.778 1150.16 Q180.857 1145.56 186.667 1145.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M116.482 861.546 L116.482 857.287 Q118.242 858.12 120.047 858.56 Q121.853 859 123.589 859 Q128.218 859 130.649 855.898 Q133.103 852.773 133.45 846.43 Q132.107 848.421 130.047 849.486 Q127.987 850.551 125.487 850.551 Q120.302 850.551 117.269 847.426 Q114.26 844.277 114.26 838.838 Q114.26 833.514 117.408 830.296 Q120.556 827.078 125.788 827.078 Q131.783 827.078 134.931 831.685 Q138.103 836.268 138.103 845.018 Q138.103 853.189 134.214 858.074 Q130.348 862.935 123.797 862.935 Q122.038 862.935 120.232 862.588 Q118.427 862.24 116.482 861.546 M125.788 846.893 Q128.936 846.893 130.765 844.74 Q132.617 842.588 132.617 838.838 Q132.617 835.111 130.765 832.958 Q128.936 830.782 125.788 830.782 Q122.64 830.782 120.788 832.958 Q118.959 835.111 118.959 838.838 Q118.959 842.588 120.788 844.74 Q122.64 846.893 125.788 846.893 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M159.353 831.778 L147.547 850.227 L159.353 850.227 L159.353 831.778 M158.126 827.703 L164.005 827.703 L164.005 850.227 L168.936 850.227 L168.936 854.115 L164.005 854.115 L164.005 862.263 L159.353 862.263 L159.353 854.115 L143.751 854.115 L143.751 849.602 L158.126 827.703 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M186.667 830.782 Q183.056 830.782 181.227 834.347 Q179.422 837.889 179.422 845.018 Q179.422 852.125 181.227 855.689 Q183.056 859.231 186.667 859.231 Q190.301 859.231 192.107 855.689 Q193.936 852.125 193.936 845.018 Q193.936 837.889 192.107 834.347 Q190.301 830.782 186.667 830.782 M186.667 827.078 Q192.477 827.078 195.533 831.685 Q198.612 836.268 198.612 845.018 Q198.612 853.745 195.533 858.351 Q192.477 862.935 186.667 862.935 Q180.857 862.935 177.778 858.351 Q174.723 853.745 174.723 845.018 Q174.723 836.268 177.778 831.685 Q180.857 827.078 186.667 827.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M116.482 543.068 L116.482 538.809 Q118.242 539.642 120.047 540.082 Q121.853 540.522 123.589 540.522 Q128.218 540.522 130.649 537.42 Q133.103 534.295 133.45 527.952 Q132.107 529.943 130.047 531.008 Q127.987 532.073 125.487 532.073 Q120.302 532.073 117.269 528.948 Q114.26 525.8 114.26 520.36 Q114.26 515.036 117.408 511.818 Q120.556 508.601 125.788 508.601 Q131.783 508.601 134.931 513.207 Q138.103 517.79 138.103 526.54 Q138.103 534.712 134.214 539.596 Q130.348 544.457 123.797 544.457 Q122.038 544.457 120.232 544.11 Q118.427 543.763 116.482 543.068 M125.788 528.415 Q128.936 528.415 130.765 526.263 Q132.617 524.11 132.617 520.36 Q132.617 516.633 130.765 514.48 Q128.936 512.304 125.788 512.304 Q122.64 512.304 120.788 514.48 Q118.959 516.633 118.959 520.36 Q118.959 524.11 120.788 526.263 Q122.64 528.415 125.788 528.415 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M157.084 524.642 Q153.936 524.642 152.084 526.795 Q150.255 528.948 150.255 532.698 Q150.255 536.425 152.084 538.601 Q153.936 540.753 157.084 540.753 Q160.232 540.753 162.061 538.601 Q163.913 536.425 163.913 532.698 Q163.913 528.948 162.061 526.795 Q160.232 524.642 157.084 524.642 M166.366 509.99 L166.366 514.249 Q164.607 513.416 162.802 512.976 Q161.019 512.536 159.26 512.536 Q154.63 512.536 152.177 515.661 Q149.746 518.786 149.399 525.105 Q150.765 523.091 152.825 522.027 Q154.885 520.939 157.362 520.939 Q162.57 520.939 165.579 524.11 Q168.612 527.258 168.612 532.698 Q168.612 538.022 165.464 541.239 Q162.315 544.457 157.084 544.457 Q151.089 544.457 147.917 539.874 Q144.746 535.267 144.746 526.54 Q144.746 518.346 148.635 513.485 Q152.524 508.601 159.075 508.601 Q160.834 508.601 162.616 508.948 Q164.422 509.295 166.366 509.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M186.667 512.304 Q183.056 512.304 181.227 515.869 Q179.422 519.411 179.422 526.54 Q179.422 533.647 181.227 537.212 Q183.056 540.753 186.667 540.753 Q190.301 540.753 192.107 537.212 Q193.936 533.647 193.936 526.54 Q193.936 519.411 192.107 515.869 Q190.301 512.304 186.667 512.304 M186.667 508.601 Q192.477 508.601 195.533 513.207 Q198.612 517.79 198.612 526.54 Q198.612 535.267 195.533 539.874 Q192.477 544.457 186.667 544.457 Q180.857 544.457 177.778 539.874 Q174.723 535.267 174.723 526.54 Q174.723 517.79 177.778 513.207 Q180.857 508.601 186.667 508.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M116.482 224.59 L116.482 220.331 Q118.242 221.164 120.047 221.604 Q121.853 222.044 123.589 222.044 Q128.218 222.044 130.649 218.942 Q133.103 215.817 133.45 209.475 Q132.107 211.465 130.047 212.53 Q127.987 213.595 125.487 213.595 Q120.302 213.595 117.269 210.47 Q114.26 207.322 114.26 201.882 Q114.26 196.558 117.408 193.341 Q120.556 190.123 125.788 190.123 Q131.783 190.123 134.931 194.729 Q138.103 199.313 138.103 208.063 Q138.103 216.234 134.214 221.118 Q130.348 225.979 123.797 225.979 Q122.038 225.979 120.232 225.632 Q118.427 225.285 116.482 224.59 M125.788 209.938 Q128.936 209.938 130.765 207.785 Q132.617 205.632 132.617 201.882 Q132.617 198.155 130.765 196.003 Q128.936 193.827 125.788 193.827 Q122.64 193.827 120.788 196.003 Q118.959 198.155 118.959 201.882 Q118.959 205.632 120.788 207.785 Q122.64 209.938 125.788 209.938 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M156.505 208.896 Q153.172 208.896 151.251 210.678 Q149.353 212.461 149.353 215.586 Q149.353 218.711 151.251 220.493 Q153.172 222.276 156.505 222.276 Q159.839 222.276 161.76 220.493 Q163.681 218.688 163.681 215.586 Q163.681 212.461 161.76 210.678 Q159.862 208.896 156.505 208.896 M151.829 206.905 Q148.82 206.165 147.13 204.104 Q145.464 202.044 145.464 199.081 Q145.464 194.938 148.403 192.53 Q151.366 190.123 156.505 190.123 Q161.667 190.123 164.607 192.53 Q167.547 194.938 167.547 199.081 Q167.547 202.044 165.857 204.104 Q164.19 206.165 161.204 206.905 Q164.584 207.692 166.459 209.984 Q168.357 212.276 168.357 215.586 Q168.357 220.609 165.278 223.294 Q162.223 225.979 156.505 225.979 Q150.788 225.979 147.709 223.294 Q144.654 220.609 144.654 215.586 Q144.654 212.276 146.552 209.984 Q148.45 207.692 151.829 206.905 M150.116 199.521 Q150.116 202.206 151.783 203.711 Q153.473 205.215 156.505 205.215 Q159.515 205.215 161.204 203.711 Q162.917 202.206 162.917 199.521 Q162.917 196.836 161.204 195.331 Q159.515 193.827 156.505 193.827 Q153.473 193.827 151.783 195.331 Q150.116 196.836 150.116 199.521 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M186.667 193.827 Q183.056 193.827 181.227 197.391 Q179.422 200.933 179.422 208.063 Q179.422 215.169 181.227 218.734 Q183.056 222.276 186.667 222.276 Q190.301 222.276 192.107 218.734 Q193.936 215.169 193.936 208.063 Q193.936 200.933 192.107 197.391 Q190.301 193.827 186.667 193.827 M186.667 190.123 Q192.477 190.123 195.533 194.729 Q198.612 199.313 198.612 208.063 Q198.612 216.789 195.533 221.396 Q192.477 225.979 186.667 225.979 Q180.857 225.979 177.778 221.396 Q174.723 216.789 174.723 208.063 Q174.723 199.313 177.778 194.729 Q180.857 190.123 186.667 190.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M21.7677 865.883 L58.7206 865.883 L58.7206 858.117 Q58.7206 848.282 54.2646 843.731 Q49.8086 839.147 40.1964 839.147 Q30.6479 839.147 26.2237 843.731 Q21.7677 848.282 21.7677 858.117 L21.7677 865.883 M16.4842 872.313 L16.4842 859.104 Q16.4842 845.29 22.2451 838.829 Q27.9743 832.368 40.1964 832.368 Q52.4822 832.368 58.2432 838.861 Q64.0042 845.354 64.0042 859.104 L64.0042 872.313 L16.4842 872.313 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M44.7161 791.882 L47.5806 791.882 L47.5806 818.809 Q53.6281 818.427 56.8109 815.181 Q59.9619 811.902 59.9619 806.078 Q59.9619 802.704 59.1344 799.553 Q58.3069 796.37 56.6518 793.251 L62.1899 793.251 Q63.5267 796.402 64.227 799.712 Q64.9272 803.022 64.9272 806.428 Q64.9272 814.958 59.9619 819.955 Q54.9967 824.92 46.5303 824.92 Q37.7774 824.92 32.6531 820.209 Q27.4968 815.467 27.4968 807.446 Q27.4968 800.253 32.1438 796.083 Q36.7589 791.882 44.7161 791.882 M42.9973 797.738 Q38.1912 797.802 35.3266 800.444 Q32.4621 803.054 32.4621 807.383 Q32.4621 812.284 35.2312 815.244 Q38.0002 818.172 43.0292 818.618 L42.9973 797.738 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M29.4065 759.544 L34.9447 759.544 Q33.6716 762.027 33.035 764.7 Q32.3984 767.374 32.3984 770.239 Q32.3984 774.599 33.7352 776.795 Q35.072 778.96 37.7456 778.96 Q39.7826 778.96 40.9603 777.4 Q42.1061 775.84 43.1565 771.13 L43.6021 769.125 Q44.9389 762.886 47.3897 760.276 Q49.8086 757.635 54.1691 757.635 Q59.1344 757.635 62.0308 761.581 Q64.9272 765.496 64.9272 772.371 Q64.9272 775.236 64.3543 778.355 Q63.8132 781.442 62.6992 784.88 L56.6518 784.88 Q58.3387 781.633 59.198 778.482 Q60.0256 775.331 60.0256 772.244 Q60.0256 768.106 58.6251 765.878 Q57.1929 763.65 54.6147 763.65 Q52.2276 763.65 50.9545 765.273 Q49.6813 766.865 48.5037 772.307 L48.0262 774.344 Q46.8804 779.787 44.5251 782.206 Q42.138 784.625 38.0002 784.625 Q32.9713 784.625 30.2341 781.06 Q27.4968 777.496 27.4968 770.939 Q27.4968 767.692 27.9743 764.828 Q28.4517 761.963 29.4065 759.544 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M28.3562 748.309 L28.3562 742.452 L64.0042 742.452 L64.0042 748.309 L28.3562 748.309 M14.479 748.309 L14.479 742.452 L21.895 742.452 L21.895 748.309 L14.479 748.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M33.8307 709.542 Q33.2578 710.528 33.0032 711.706 Q32.7167 712.852 32.7167 714.252 Q32.7167 719.217 35.9632 721.891 Q39.1779 724.533 45.2253 724.533 L64.0042 724.533 L64.0042 730.421 L28.3562 730.421 L28.3562 724.533 L33.8944 724.533 Q30.6479 722.687 29.0883 719.727 Q27.4968 716.767 27.4968 712.533 Q27.4968 711.929 27.5923 711.197 Q27.656 710.465 27.8151 709.573 L33.8307 709.542 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M44.7161 674.339 L47.5806 674.339 L47.5806 701.266 Q53.6281 700.884 56.8109 697.638 Q59.9619 694.359 59.9619 688.535 Q59.9619 685.161 59.1344 682.01 Q58.3069 678.827 56.6518 675.708 L62.1899 675.708 Q63.5267 678.859 64.227 682.169 Q64.9272 685.479 64.9272 688.885 Q64.9272 697.415 59.9619 702.412 Q54.9967 707.377 46.5303 707.377 Q37.7774 707.377 32.6531 702.667 Q27.4968 697.924 27.4968 689.903 Q27.4968 682.71 32.1438 678.541 Q36.7589 674.339 44.7161 674.339 M42.9973 680.196 Q38.1912 680.259 35.3266 682.901 Q32.4621 685.511 32.4621 689.84 Q32.4621 694.741 35.2312 697.701 Q38.0002 700.63 43.0292 701.075 L42.9973 680.196 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M994.686 17.6457 Q985.775 17.6457 980.508 24.2892 Q975.283 30.9327 975.283 42.3968 Q975.283 53.8203 980.508 60.4638 Q985.775 67.1073 994.686 67.1073 Q1003.6 67.1073 1008.78 60.4638 Q1014.01 53.8203 1014.01 42.3968 Q1014.01 30.9327 1008.78 24.2892 Q1003.6 17.6457 994.686 17.6457 M994.686 11.0023 Q1007.41 11.0023 1015.02 19.5497 Q1022.64 28.0566 1022.64 42.3968 Q1022.64 56.6965 1015.02 65.2439 Q1007.41 73.7508 994.686 73.7508 Q981.926 73.7508 974.27 65.2439 Q966.654 56.737 966.654 42.3968 Q966.654 28.0566 974.27 19.5497 Q981.926 11.0023 994.686 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1067.68 49.9314 Q1067.68 41.7081 1064.28 37.0496 Q1060.92 32.3505 1055 32.3505 Q1049.09 32.3505 1045.69 37.0496 Q1042.33 41.7081 1042.33 49.9314 Q1042.33 58.1548 1045.69 62.8538 Q1049.09 67.5124 1055 67.5124 Q1060.92 67.5124 1064.28 62.8538 Q1067.68 58.1548 1067.68 49.9314 M1042.33 34.0924 Q1044.67 30.0415 1048.24 28.0971 Q1051.84 26.1121 1056.83 26.1121 Q1065.09 26.1121 1070.24 32.6746 Q1075.42 39.2371 1075.42 49.9314 Q1075.42 60.6258 1070.24 67.1883 Q1065.09 73.7508 1056.83 73.7508 Q1051.84 73.7508 1048.24 71.8063 Q1044.67 69.8214 1042.33 65.7705 L1042.33 72.576 L1034.83 72.576 L1034.83 9.54393 L1042.33 9.54393 L1042.33 34.0924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1087.78 27.2059 L1095.23 27.2059 L1095.23 73.3862 Q1095.23 82.0551 1091.91 85.944 Q1088.63 89.8329 1081.29 89.8329 L1078.46 89.8329 L1078.46 83.5134 L1080.44 83.5134 Q1084.7 83.5134 1086.24 81.5285 Q1087.78 79.5841 1087.78 73.3862 L1087.78 27.2059 M1087.78 9.54393 L1095.23 9.54393 L1095.23 18.9825 L1087.78 18.9825 L1087.78 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1149.63 48.0275 L1149.63 51.6733 L1115.36 51.6733 Q1115.85 59.3701 1119.98 63.421 Q1124.15 67.4314 1131.57 67.4314 Q1135.86 67.4314 1139.87 66.3781 Q1143.92 65.3249 1147.89 63.2184 L1147.89 70.267 Q1143.88 71.9684 1139.67 72.8596 Q1135.46 73.7508 1131.12 73.7508 Q1120.26 73.7508 1113.9 67.4314 Q1107.59 61.1119 1107.59 50.3365 Q1107.59 39.1965 1113.58 32.6746 Q1119.62 26.1121 1129.82 26.1121 Q1138.98 26.1121 1144.29 32.0264 Q1149.63 37.9003 1149.63 48.0275 M1142.18 45.84 Q1142.1 39.7232 1138.74 36.0774 Q1135.41 32.4315 1129.91 32.4315 Q1123.67 32.4315 1119.9 35.9558 Q1116.17 39.4801 1115.61 45.8805 L1142.18 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1194.52 28.9478 L1194.52 35.9153 Q1191.36 34.1734 1188.16 33.3227 Q1185 32.4315 1181.76 32.4315 Q1174.51 32.4315 1170.5 37.0496 Q1166.49 41.6271 1166.49 49.9314 Q1166.49 58.2358 1170.5 62.8538 Q1174.51 67.4314 1181.76 67.4314 Q1185 67.4314 1188.16 66.5807 Q1191.36 65.6895 1194.52 63.9476 L1194.52 70.8341 Q1191.4 72.2924 1188.04 73.0216 Q1184.71 73.7508 1180.95 73.7508 Q1170.7 73.7508 1164.66 67.3098 Q1158.63 60.8689 1158.63 49.9314 Q1158.63 38.832 1164.7 32.472 Q1170.82 26.1121 1181.43 26.1121 Q1184.88 26.1121 1188.16 26.8413 Q1191.44 27.5299 1194.52 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1214.85 14.324 L1214.85 27.2059 L1230.21 27.2059 L1230.21 32.9987 L1214.85 32.9987 L1214.85 57.6282 Q1214.85 63.1779 1216.35 64.7578 Q1217.89 66.3376 1222.55 66.3376 L1230.21 66.3376 L1230.21 72.576 L1222.55 72.576 Q1213.92 72.576 1210.64 69.3758 Q1207.36 66.1351 1207.36 57.6282 L1207.36 32.9987 L1201.89 32.9987 L1201.89 27.2059 L1207.36 27.2059 L1207.36 14.324 L1214.85 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1240.01 27.2059 L1247.46 27.2059 L1247.46 72.576 L1240.01 72.576 L1240.01 27.2059 M1240.01 9.54393 L1247.46 9.54393 L1247.46 18.9825 L1240.01 18.9825 L1240.01 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1257.71 27.2059 L1265.61 27.2059 L1279.79 65.2844 L1293.97 27.2059 L1301.87 27.2059 L1284.85 72.576 L1274.73 72.576 L1257.71 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1350.96 48.0275 L1350.96 51.6733 L1316.69 51.6733 Q1317.18 59.3701 1321.31 63.421 Q1325.48 67.4314 1332.9 67.4314 Q1337.19 67.4314 1341.2 66.3781 Q1345.25 65.3249 1349.22 63.2184 L1349.22 70.267 Q1345.21 71.9684 1341 72.8596 Q1336.79 73.7508 1332.45 73.7508 Q1321.59 73.7508 1315.23 67.4314 Q1308.92 61.1119 1308.92 50.3365 Q1308.92 39.1965 1314.91 32.6746 Q1320.95 26.1121 1331.15 26.1121 Q1340.31 26.1121 1345.62 32.0264 Q1350.96 37.9003 1350.96 48.0275 M1343.51 45.84 Q1343.43 39.7232 1340.07 36.0774 Q1336.74 32.4315 1331.24 32.4315 Q1325 32.4315 1321.23 35.9558 Q1317.5 39.4801 1316.94 45.8805 L1343.51 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1418.49 28.5427 L1418.49 35.5912 Q1415.33 33.9709 1411.93 33.1607 Q1408.53 32.3505 1404.88 32.3505 Q1399.33 32.3505 1396.54 34.0519 Q1393.78 35.7533 1393.78 39.156 Q1393.78 41.7486 1395.77 43.2475 Q1397.75 44.7058 1403.75 46.0426 L1406.3 46.6097 Q1414.24 48.3111 1417.56 51.4303 Q1420.92 54.509 1420.92 60.0587 Q1420.92 66.3781 1415.9 70.0644 Q1410.92 73.7508 1402.17 73.7508 Q1398.52 73.7508 1394.55 73.0216 Q1390.62 72.3329 1386.25 70.9151 L1386.25 63.2184 Q1390.38 65.3654 1394.39 66.4591 Q1398.4 67.5124 1402.33 67.5124 Q1407.6 67.5124 1410.43 65.73 Q1413.27 63.9071 1413.27 60.6258 Q1413.27 57.5877 1411.2 55.9673 Q1409.18 54.3469 1402.25 52.8481 L1399.66 52.2405 Q1392.73 50.7821 1389.65 47.7845 Q1386.57 44.7463 1386.57 39.4801 Q1386.57 33.0797 1391.11 29.5959 Q1395.64 26.1121 1403.99 26.1121 Q1408.12 26.1121 1411.77 26.7198 Q1415.41 27.3274 1418.49 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1440 65.7705 L1440 89.8329 L1432.51 89.8329 L1432.51 27.2059 L1440 27.2059 L1440 34.0924 Q1442.35 30.0415 1445.92 28.0971 Q1449.52 26.1121 1454.5 26.1121 Q1462.77 26.1121 1467.91 32.6746 Q1473.1 39.2371 1473.1 49.9314 Q1473.1 60.6258 1467.91 67.1883 Q1462.77 73.7508 1454.5 73.7508 Q1449.52 73.7508 1445.92 71.8063 Q1442.35 69.8214 1440 65.7705 M1465.36 49.9314 Q1465.36 41.7081 1461.96 37.0496 Q1458.6 32.3505 1452.68 32.3505 Q1446.77 32.3505 1443.36 37.0496 Q1440 41.7081 1440 49.9314 Q1440 58.1548 1443.36 62.8538 Q1446.77 67.5124 1452.68 67.5124 Q1458.6 67.5124 1461.96 62.8538 Q1465.36 58.1548 1465.36 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1506.07 49.7694 Q1497.04 49.7694 1493.56 51.8354 Q1490.07 53.9013 1490.07 58.8839 Q1490.07 62.8538 1492.66 65.2034 Q1495.3 67.5124 1499.79 67.5124 Q1505.99 67.5124 1509.72 63.1374 Q1513.49 58.7219 1513.49 51.4303 L1513.49 49.7694 L1506.07 49.7694 M1520.94 46.6907 L1520.94 72.576 L1513.49 72.576 L1513.49 65.6895 Q1510.93 69.8214 1507.13 71.8063 Q1503.32 73.7508 1497.81 73.7508 Q1490.84 73.7508 1486.71 69.8619 Q1482.62 65.9325 1482.62 59.3701 Q1482.62 51.7138 1487.72 47.825 Q1492.87 43.9361 1503.03 43.9361 L1513.49 43.9361 L1513.49 43.2069 Q1513.49 38.0623 1510.08 35.2672 Q1506.72 32.4315 1500.6 32.4315 Q1496.72 32.4315 1493.03 33.3632 Q1489.34 34.295 1485.94 36.1584 L1485.94 29.2718 Q1490.03 27.692 1493.88 26.9223 Q1497.73 26.1121 1501.37 26.1121 Q1511.22 26.1121 1516.08 31.2163 Q1520.94 36.3204 1520.94 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1568.94 28.9478 L1568.94 35.9153 Q1565.78 34.1734 1562.58 33.3227 Q1559.42 32.4315 1556.18 32.4315 Q1548.93 32.4315 1544.92 37.0496 Q1540.91 41.6271 1540.91 49.9314 Q1540.91 58.2358 1544.92 62.8538 Q1548.93 67.4314 1556.18 67.4314 Q1559.42 67.4314 1562.58 66.5807 Q1565.78 65.6895 1568.94 63.9476 L1568.94 70.8341 Q1565.82 72.2924 1562.46 73.0216 Q1559.14 73.7508 1555.37 73.7508 Q1545.12 73.7508 1539.09 67.3098 Q1533.05 60.8689 1533.05 49.9314 Q1533.05 38.832 1539.13 32.472 Q1545.24 26.1121 1555.86 26.1121 Q1559.3 26.1121 1562.58 26.8413 Q1565.86 27.5299 1568.94 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1620.71 48.0275 L1620.71 51.6733 L1586.44 51.6733 Q1586.93 59.3701 1591.06 63.421 Q1595.23 67.4314 1602.65 67.4314 Q1606.94 67.4314 1610.95 66.3781 Q1615 65.3249 1618.97 63.2184 L1618.97 70.267 Q1614.96 71.9684 1610.75 72.8596 Q1606.54 73.7508 1602.2 73.7508 Q1591.34 73.7508 1584.98 67.4314 Q1578.66 61.1119 1578.66 50.3365 Q1578.66 39.1965 1584.66 32.6746 Q1590.7 26.1121 1600.9 26.1121 Q1610.06 26.1121 1615.37 32.0264 Q1620.71 37.9003 1620.71 48.0275 M1613.26 45.84 Q1613.18 39.7232 1609.82 36.0774 Q1606.49 32.4315 1600.99 32.4315 Q1594.75 32.4315 1590.98 35.9558 Q1587.25 39.4801 1586.69 45.8805 L1613.26 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip032)" cx="375.821" cy="160.256" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="799.45" cy="335.419" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1128.94" cy="351.343" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1176.01" cy="733.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1223.08" cy="813.136" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1552.57" cy="844.983" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1787.92" cy="860.907" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="1834.99" cy="1243.08" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="2117.41" cy="1386.4" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip032)" cx="2117.41" cy="160.256" r="14.4" fill="#e26f46" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<path clip-path="url(#clip030)" d="M305.216 1379.86 L762.678 1379.86 L762.678 1276.18 L305.216 1276.18  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="305.216,1379.86 762.678,1379.86 762.678,1276.18 305.216,1276.18 305.216,1379.86 "/>
-<circle clip-path="url(#clip030)" cx="399.356" cy="1328.02" r="20.48" fill="#e26f46" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
-<path clip-path="url(#clip030)" d="M493.496 1310.74 L498.172 1310.74 L498.172 1345.3 L493.496 1345.3 L493.496 1310.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M524.352 1323.31 L524.352 1309.28 L528.611 1309.28 L528.611 1345.3 L524.352 1345.3 L524.352 1341.41 Q523.01 1343.72 520.949 1344.86 Q518.912 1345.97 516.042 1345.97 Q511.343 1345.97 508.38 1342.22 Q505.44 1338.47 505.44 1332.36 Q505.44 1326.25 508.38 1322.5 Q511.343 1318.75 516.042 1318.75 Q518.912 1318.75 520.949 1319.88 Q523.01 1320.99 524.352 1323.31 M509.838 1332.36 Q509.838 1337.06 511.76 1339.74 Q513.704 1342.4 517.084 1342.4 Q520.463 1342.4 522.408 1339.74 Q524.352 1337.06 524.352 1332.36 Q524.352 1327.66 522.408 1325 Q520.463 1322.31 517.084 1322.31 Q513.704 1322.31 511.76 1325 Q509.838 1327.66 509.838 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M559.56 1331.27 L559.56 1333.35 L539.977 1333.35 Q540.255 1337.75 542.616 1340.06 Q545 1342.36 549.236 1342.36 Q551.69 1342.36 553.982 1341.75 Q556.296 1341.15 558.565 1339.95 L558.565 1343.98 Q556.273 1344.95 553.866 1345.46 Q551.459 1345.97 548.982 1345.97 Q542.778 1345.97 539.144 1342.36 Q535.533 1338.75 535.533 1332.59 Q535.533 1326.22 538.959 1322.5 Q542.408 1318.75 548.241 1318.75 Q553.472 1318.75 556.505 1322.13 Q559.56 1325.48 559.56 1331.27 M555.301 1330.02 Q555.255 1326.52 553.334 1324.44 Q551.435 1322.36 548.287 1322.36 Q544.722 1322.36 542.57 1324.37 Q540.44 1326.38 540.116 1330.04 L555.301 1330.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M578.333 1332.26 Q573.171 1332.26 571.181 1333.44 Q569.19 1334.63 569.19 1337.47 Q569.19 1339.74 570.671 1341.08 Q572.176 1342.4 574.745 1342.4 Q578.287 1342.4 580.417 1339.9 Q582.569 1337.38 582.569 1333.21 L582.569 1332.26 L578.333 1332.26 M586.829 1330.5 L586.829 1345.3 L582.569 1345.3 L582.569 1341.36 Q581.111 1343.72 578.935 1344.86 Q576.759 1345.97 573.611 1345.97 Q569.63 1345.97 567.269 1343.75 Q564.931 1341.5 564.931 1337.75 Q564.931 1333.38 567.847 1331.15 Q570.787 1328.93 576.597 1328.93 L582.569 1328.93 L582.569 1328.51 Q582.569 1325.57 580.625 1323.98 Q578.704 1322.36 575.208 1322.36 Q572.986 1322.36 570.88 1322.89 Q568.773 1323.42 566.829 1324.49 L566.829 1320.55 Q569.167 1319.65 571.366 1319.21 Q573.565 1318.75 575.648 1318.75 Q581.273 1318.75 584.051 1321.66 Q586.829 1324.58 586.829 1330.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M595.602 1309.28 L599.861 1309.28 L599.861 1345.3 L595.602 1345.3 L595.602 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M627.963 1341.41 L627.963 1355.16 L623.68 1355.16 L623.68 1319.37 L627.963 1319.37 L627.963 1323.31 Q629.305 1320.99 631.342 1319.88 Q633.403 1318.75 636.25 1318.75 Q640.972 1318.75 643.912 1322.5 Q646.875 1326.25 646.875 1332.36 Q646.875 1338.47 643.912 1342.22 Q640.972 1345.97 636.25 1345.97 Q633.403 1345.97 631.342 1344.86 Q629.305 1343.72 627.963 1341.41 M642.453 1332.36 Q642.453 1327.66 640.509 1325 Q638.588 1322.31 635.208 1322.31 Q631.828 1322.31 629.884 1325 Q627.963 1327.66 627.963 1332.36 Q627.963 1337.06 629.884 1339.74 Q631.828 1342.4 635.208 1342.4 Q638.588 1342.4 640.509 1339.74 Q642.453 1337.06 642.453 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M663.981 1322.36 Q660.555 1322.36 658.564 1325.04 Q656.574 1327.7 656.574 1332.36 Q656.574 1337.01 658.541 1339.69 Q660.532 1342.36 663.981 1342.36 Q667.384 1342.36 669.375 1339.67 Q671.365 1336.99 671.365 1332.36 Q671.365 1327.75 669.375 1325.06 Q667.384 1322.36 663.981 1322.36 M663.981 1318.75 Q669.537 1318.75 672.708 1322.36 Q675.879 1325.97 675.879 1332.36 Q675.879 1338.72 672.708 1342.36 Q669.537 1345.97 663.981 1345.97 Q658.402 1345.97 655.231 1342.36 Q652.083 1338.72 652.083 1332.36 Q652.083 1325.97 655.231 1322.36 Q658.402 1318.75 663.981 1318.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M682.939 1319.37 L687.199 1319.37 L687.199 1345.3 L682.939 1345.3 L682.939 1319.37 M682.939 1309.28 L687.199 1309.28 L687.199 1314.67 L682.939 1314.67 L682.939 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M717.661 1329.65 L717.661 1345.3 L713.402 1345.3 L713.402 1329.79 Q713.402 1326.11 711.967 1324.28 Q710.532 1322.45 707.661 1322.45 Q704.212 1322.45 702.222 1324.65 Q700.231 1326.85 700.231 1330.64 L700.231 1345.3 L695.948 1345.3 L695.948 1319.37 L700.231 1319.37 L700.231 1323.4 Q701.759 1321.06 703.819 1319.9 Q705.902 1318.75 708.61 1318.75 Q713.078 1318.75 715.37 1321.52 Q717.661 1324.28 717.661 1329.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M730.37 1312.01 L730.37 1319.37 L739.143 1319.37 L739.143 1322.68 L730.37 1322.68 L730.37 1336.75 Q730.37 1339.93 731.226 1340.83 Q732.106 1341.73 734.768 1341.73 L739.143 1341.73 L739.143 1345.3 L734.768 1345.3 Q729.837 1345.3 727.962 1343.47 Q726.087 1341.62 726.087 1336.75 L726.087 1322.68 L722.962 1322.68 L722.962 1319.37 L726.087 1319.37 L726.087 1312.01 L730.37 1312.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M316.729 176.937 L326.277 176.937 L326.277 143.98 L315.89 146.064 L315.89 140.74 L326.219 138.656 L332.064 138.656 L332.064 176.937 L341.613 176.937 L341.613 181.856 L316.729 181.856 L316.729 176.937 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M745.696 352.1 L766.095 352.1 L766.095 357.019 L738.665 357.019 L738.665 352.1 Q741.992 348.657 747.722 342.87 Q753.48 337.054 754.955 335.376 Q757.762 332.222 758.862 330.052 Q759.99 327.853 759.99 325.74 Q759.99 322.297 757.56 320.127 Q755.158 317.957 751.281 317.957 Q748.532 317.957 745.465 318.912 Q742.427 319.866 738.954 321.805 L738.954 315.902 Q742.484 314.485 745.552 313.761 Q748.619 313.038 751.165 313.038 Q757.878 313.038 761.871 316.394 Q765.864 319.751 765.864 325.364 Q765.864 328.026 764.851 330.428 Q763.867 332.8 761.234 336.041 Q760.511 336.88 756.634 340.902 Q752.756 344.895 745.696 352.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1087.18 349.65 Q1091.37 350.547 1093.72 353.383 Q1096.09 356.219 1096.09 360.385 Q1096.09 366.78 1091.69 370.281 Q1087.29 373.782 1079.19 373.782 Q1076.47 373.782 1073.58 373.232 Q1070.71 372.711 1067.65 371.641 L1067.65 365.999 Q1070.08 367.416 1072.97 368.14 Q1075.87 368.863 1079.02 368.863 Q1084.52 368.863 1087.38 366.693 Q1090.27 364.523 1090.27 360.385 Q1090.27 356.566 1087.58 354.425 Q1084.92 352.254 1080.15 352.254 L1075.11 352.254 L1075.11 347.451 L1080.38 347.451 Q1084.69 347.451 1086.98 345.744 Q1089.26 344.008 1089.26 340.767 Q1089.26 337.44 1086.89 335.675 Q1084.55 333.881 1080.15 333.881 Q1077.75 333.881 1075 334.401 Q1072.25 334.922 1068.95 336.022 L1068.95 330.814 Q1072.28 329.888 1075.17 329.425 Q1078.09 328.962 1080.67 328.962 Q1087.32 328.962 1091.2 332 Q1095.08 335.009 1095.08 340.16 Q1095.08 343.748 1093.02 346.236 Q1090.97 348.695 1087.18 349.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1132.7 717.009 L1117.94 740.07 L1132.7 740.07 L1132.7 717.009 M1131.17 711.916 L1138.52 711.916 L1138.52 740.07 L1144.68 740.07 L1144.68 744.931 L1138.52 744.931 L1138.52 755.116 L1132.7 755.116 L1132.7 744.931 L1113.2 744.931 L1113.2 739.289 L1131.17 711.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1163.86 791.536 L1186.8 791.536 L1186.8 796.455 L1169.21 796.455 L1169.21 807.045 Q1170.48 806.611 1171.76 806.408 Q1173.03 806.177 1174.3 806.177 Q1181.54 806.177 1185.76 810.141 Q1189.98 814.105 1189.98 820.876 Q1189.98 827.849 1185.64 831.726 Q1181.3 835.575 1173.4 835.575 Q1170.68 835.575 1167.85 835.112 Q1165.04 834.649 1162.03 833.723 L1162.03 827.849 Q1164.64 829.267 1167.42 829.961 Q1170.19 830.656 1173.29 830.656 Q1178.29 830.656 1181.22 828.023 Q1184.14 825.39 1184.14 820.876 Q1184.14 816.362 1181.22 813.729 Q1178.29 811.096 1173.29 811.096 Q1170.95 811.096 1168.6 811.617 Q1166.29 812.137 1163.86 813.237 L1163.86 791.536 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1506 842.654 Q1502.07 842.654 1499.75 845.345 Q1497.47 848.036 1497.47 852.724 Q1497.47 857.382 1499.75 860.102 Q1502.07 862.793 1506 862.793 Q1509.94 862.793 1512.23 860.102 Q1514.54 857.382 1514.54 852.724 Q1514.54 848.036 1512.23 845.345 Q1509.94 842.654 1506 842.654 M1517.61 824.338 L1517.61 829.662 Q1515.41 828.621 1513.15 828.071 Q1510.92 827.521 1508.72 827.521 Q1502.94 827.521 1499.87 831.427 Q1496.83 835.334 1496.4 843.233 Q1498.11 840.716 1500.68 839.385 Q1503.26 838.025 1506.35 838.025 Q1512.86 838.025 1516.62 841.989 Q1520.41 845.924 1520.41 852.724 Q1520.41 859.379 1516.48 863.401 Q1512.54 867.423 1506 867.423 Q1498.51 867.423 1494.55 861.693 Q1490.58 855.935 1490.58 845.027 Q1490.58 834.784 1495.44 828.708 Q1500.3 822.602 1508.49 822.602 Q1510.69 822.602 1512.92 823.036 Q1515.18 823.47 1517.61 824.338 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1726.96 839.307 L1754.74 839.307 L1754.74 841.796 L1739.05 882.507 L1732.95 882.507 L1747.71 844.226 L1726.96 844.226 L1726.96 839.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1787.92 1244.17 Q1783.75 1244.17 1781.35 1246.39 Q1778.98 1248.62 1778.98 1252.53 Q1778.98 1256.43 1781.35 1258.66 Q1783.75 1260.89 1787.92 1260.89 Q1792.08 1260.89 1794.49 1258.66 Q1796.89 1256.41 1796.89 1252.53 Q1796.89 1248.62 1794.49 1246.39 Q1792.11 1244.17 1787.92 1244.17 M1782.07 1241.68 Q1778.31 1240.75 1776.2 1238.18 Q1774.12 1235.6 1774.12 1231.9 Q1774.12 1226.72 1777.79 1223.71 Q1781.49 1220.7 1787.92 1220.7 Q1794.37 1220.7 1798.04 1223.71 Q1801.72 1226.72 1801.72 1231.9 Q1801.72 1235.6 1799.61 1238.18 Q1797.52 1240.75 1793.79 1241.68 Q1798.02 1242.66 1800.36 1245.53 Q1802.73 1248.39 1802.73 1252.53 Q1802.73 1258.81 1798.88 1262.16 Q1795.06 1265.52 1787.92 1265.52 Q1780.77 1265.52 1776.92 1262.16 Q1773.1 1258.81 1773.1 1252.53 Q1773.1 1248.39 1775.48 1245.53 Q1777.85 1242.66 1782.07 1241.68 M1779.93 1232.45 Q1779.93 1235.8 1782.01 1237.68 Q1784.13 1239.57 1787.92 1239.57 Q1791.68 1239.57 1793.79 1237.68 Q1795.93 1235.8 1795.93 1232.45 Q1795.93 1229.09 1793.79 1227.21 Q1791.68 1225.33 1787.92 1225.33 Q1784.13 1225.33 1782.01 1227.21 Q1779.93 1229.09 1779.93 1232.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2058.21 1407.1 L2058.21 1401.77 Q2060.41 1402.82 2062.67 1403.37 Q2064.93 1403.92 2067.1 1403.92 Q2072.88 1403.92 2075.92 1400.04 Q2078.99 1396.13 2079.42 1388.2 Q2077.74 1390.69 2075.17 1392.02 Q2072.59 1393.35 2069.47 1393.35 Q2062.99 1393.35 2059.2 1389.45 Q2055.44 1385.51 2055.44 1378.71 Q2055.44 1372.06 2059.37 1368.04 Q2063.31 1364.01 2069.84 1364.01 Q2077.34 1364.01 2081.27 1369.77 Q2085.24 1375.5 2085.24 1386.44 Q2085.24 1396.65 2080.38 1402.76 Q2075.54 1408.83 2067.36 1408.83 Q2065.16 1408.83 2062.9 1408.4 Q2060.64 1407.97 2058.21 1407.1 M2069.84 1388.78 Q2073.78 1388.78 2076.07 1386.09 Q2078.38 1383.4 2078.38 1378.71 Q2078.38 1374.05 2076.07 1371.36 Q2073.78 1368.64 2069.84 1368.64 Q2065.91 1368.64 2063.59 1371.36 Q2061.31 1374.05 2061.31 1378.71 Q2061.31 1383.4 2063.59 1386.09 Q2065.91 1388.78 2069.84 1388.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="469.961,1423.18 469.961,123.472 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="940.66,1423.18 940.66,123.472 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1411.36,1423.18 1411.36,123.472 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1882.06,1423.18 1882.06,123.472 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2352.76,1423.18 2352.76,123.472 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="469.961,1423.18 469.961,1404.28 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="940.66,1423.18 940.66,1404.28 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1411.36,1423.18 1411.36,1404.28 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1882.06,1423.18 1882.06,1404.28 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2352.76,1423.18 2352.76,1404.28 "/>
+<path clip-path="url(#clip850)" d="M430.007 1484.86 L430.007 1480.6 Q431.767 1481.44 433.572 1481.88 Q435.378 1482.32 437.114 1482.32 Q441.743 1482.32 444.174 1479.21 Q446.628 1476.09 446.975 1469.75 Q445.632 1471.74 443.572 1472.8 Q441.512 1473.87 439.012 1473.87 Q433.827 1473.87 430.794 1470.74 Q427.785 1467.59 427.785 1462.15 Q427.785 1456.83 430.933 1453.61 Q434.081 1450.39 439.313 1450.39 Q445.308 1450.39 448.456 1455 Q451.628 1459.58 451.628 1468.33 Q451.628 1476.51 447.739 1481.39 Q443.873 1486.25 437.322 1486.25 Q435.563 1486.25 433.757 1485.9 Q431.952 1485.56 430.007 1484.86 M439.313 1470.21 Q442.461 1470.21 444.29 1468.06 Q446.142 1465.9 446.142 1462.15 Q446.142 1458.43 444.29 1456.27 Q442.461 1454.1 439.313 1454.1 Q436.165 1454.1 434.313 1456.27 Q432.484 1458.43 432.484 1462.15 Q432.484 1465.9 434.313 1468.06 Q436.165 1470.21 439.313 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M464.058 1481.64 L480.378 1481.64 L480.378 1485.58 L458.433 1485.58 L458.433 1481.64 Q461.095 1478.89 465.679 1474.26 Q470.285 1469.61 471.466 1468.27 Q473.711 1465.74 474.591 1464.01 Q475.493 1462.25 475.493 1460.56 Q475.493 1457.8 473.549 1456.07 Q471.628 1454.33 468.526 1454.33 Q466.327 1454.33 463.873 1455.09 Q461.442 1455.86 458.665 1457.41 L458.665 1452.69 Q461.489 1451.55 463.942 1450.97 Q466.396 1450.39 468.433 1450.39 Q473.804 1450.39 476.998 1453.08 Q480.192 1455.77 480.192 1460.26 Q480.192 1462.39 479.382 1464.31 Q478.595 1466.2 476.489 1468.8 Q475.91 1469.47 472.808 1472.69 Q469.706 1475.88 464.058 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M500.192 1454.1 Q496.581 1454.1 494.752 1457.66 Q492.947 1461.2 492.947 1468.33 Q492.947 1475.44 494.752 1479.01 Q496.581 1482.55 500.192 1482.55 Q503.826 1482.55 505.632 1479.01 Q507.461 1475.44 507.461 1468.33 Q507.461 1461.2 505.632 1457.66 Q503.826 1454.1 500.192 1454.1 M500.192 1450.39 Q506.002 1450.39 509.058 1455 Q512.137 1459.58 512.137 1468.33 Q512.137 1477.06 509.058 1481.67 Q506.002 1486.25 500.192 1486.25 Q494.382 1486.25 491.303 1481.67 Q488.248 1477.06 488.248 1468.33 Q488.248 1459.58 491.303 1455 Q494.382 1450.39 500.192 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M900.706 1484.86 L900.706 1480.6 Q902.465 1481.44 904.271 1481.88 Q906.077 1482.32 907.813 1482.32 Q912.442 1482.32 914.873 1479.21 Q917.326 1476.09 917.674 1469.75 Q916.331 1471.74 914.271 1472.8 Q912.211 1473.87 909.711 1473.87 Q904.526 1473.87 901.493 1470.74 Q898.484 1467.59 898.484 1462.15 Q898.484 1456.83 901.632 1453.61 Q904.78 1450.39 910.012 1450.39 Q916.007 1450.39 919.155 1455 Q922.326 1459.58 922.326 1468.33 Q922.326 1476.51 918.438 1481.39 Q914.572 1486.25 908.021 1486.25 Q906.262 1486.25 904.456 1485.9 Q902.651 1485.56 900.706 1484.86 M910.012 1470.21 Q913.16 1470.21 914.989 1468.06 Q916.84 1465.9 916.84 1462.15 Q916.84 1458.43 914.989 1456.27 Q913.16 1454.1 910.012 1454.1 Q906.864 1454.1 905.012 1456.27 Q903.183 1458.43 903.183 1462.15 Q903.183 1465.9 905.012 1468.06 Q906.864 1470.21 910.012 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M944.896 1466.95 Q948.252 1467.66 950.127 1469.93 Q952.025 1472.2 952.025 1475.53 Q952.025 1480.65 948.507 1483.45 Q944.988 1486.25 938.507 1486.25 Q936.331 1486.25 934.016 1485.81 Q931.725 1485.39 929.271 1484.54 L929.271 1480.02 Q931.215 1481.16 933.53 1481.74 Q935.845 1482.32 938.368 1482.32 Q942.766 1482.32 945.058 1480.58 Q947.373 1478.84 947.373 1475.53 Q947.373 1472.48 945.22 1470.77 Q943.09 1469.03 939.271 1469.03 L935.243 1469.03 L935.243 1465.19 L939.456 1465.19 Q942.905 1465.19 944.734 1463.82 Q946.562 1462.43 946.562 1459.84 Q946.562 1457.18 944.664 1455.77 Q942.789 1454.33 939.271 1454.33 Q937.35 1454.33 935.15 1454.75 Q932.951 1455.16 930.313 1456.04 L930.313 1451.88 Q932.975 1451.14 935.289 1450.77 Q937.627 1450.39 939.687 1450.39 Q945.012 1450.39 948.113 1452.83 Q951.215 1455.23 951.215 1459.35 Q951.215 1462.22 949.572 1464.21 Q947.928 1466.18 944.896 1466.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M970.891 1454.1 Q967.28 1454.1 965.451 1457.66 Q963.646 1461.2 963.646 1468.33 Q963.646 1475.44 965.451 1479.01 Q967.28 1482.55 970.891 1482.55 Q974.525 1482.55 976.331 1479.01 Q978.159 1475.44 978.159 1468.33 Q978.159 1461.2 976.331 1457.66 Q974.525 1454.1 970.891 1454.1 M970.891 1450.39 Q976.701 1450.39 979.757 1455 Q982.835 1459.58 982.835 1468.33 Q982.835 1477.06 979.757 1481.67 Q976.701 1486.25 970.891 1486.25 Q965.081 1486.25 962.002 1481.67 Q958.947 1477.06 958.947 1468.33 Q958.947 1459.58 962.002 1455 Q965.081 1450.39 970.891 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1371.4 1484.86 L1371.4 1480.6 Q1373.16 1481.44 1374.97 1481.88 Q1376.78 1482.32 1378.51 1482.32 Q1383.14 1482.32 1385.57 1479.21 Q1388.03 1476.09 1388.37 1469.75 Q1387.03 1471.74 1384.97 1472.8 Q1382.91 1473.87 1380.41 1473.87 Q1375.22 1473.87 1372.19 1470.74 Q1369.18 1467.59 1369.18 1462.15 Q1369.18 1456.83 1372.33 1453.61 Q1375.48 1450.39 1380.71 1450.39 Q1386.71 1450.39 1389.85 1455 Q1393.03 1459.58 1393.03 1468.33 Q1393.03 1476.51 1389.14 1481.39 Q1385.27 1486.25 1378.72 1486.25 Q1376.96 1486.25 1375.15 1485.9 Q1373.35 1485.56 1371.4 1484.86 M1380.71 1470.21 Q1383.86 1470.21 1385.69 1468.06 Q1387.54 1465.9 1387.54 1462.15 Q1387.54 1458.43 1385.69 1456.27 Q1383.86 1454.1 1380.71 1454.1 Q1377.56 1454.1 1375.71 1456.27 Q1373.88 1458.43 1373.88 1462.15 Q1373.88 1465.9 1375.71 1468.06 Q1377.56 1470.21 1380.71 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1414.28 1455.09 L1402.47 1473.54 L1414.28 1473.54 L1414.28 1455.09 M1413.05 1451.02 L1418.93 1451.02 L1418.93 1473.54 L1423.86 1473.54 L1423.86 1477.43 L1418.93 1477.43 L1418.93 1485.58 L1414.28 1485.58 L1414.28 1477.43 L1398.67 1477.43 L1398.67 1472.92 L1413.05 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1441.59 1454.1 Q1437.98 1454.1 1436.15 1457.66 Q1434.34 1461.2 1434.34 1468.33 Q1434.34 1475.44 1436.15 1479.01 Q1437.98 1482.55 1441.59 1482.55 Q1445.22 1482.55 1447.03 1479.01 Q1448.86 1475.44 1448.86 1468.33 Q1448.86 1461.2 1447.03 1457.66 Q1445.22 1454.1 1441.59 1454.1 M1441.59 1450.39 Q1447.4 1450.39 1450.46 1455 Q1453.53 1459.58 1453.53 1468.33 Q1453.53 1477.06 1450.46 1481.67 Q1447.4 1486.25 1441.59 1486.25 Q1435.78 1486.25 1432.7 1481.67 Q1429.65 1477.06 1429.65 1468.33 Q1429.65 1459.58 1432.7 1455 Q1435.78 1450.39 1441.59 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1842.1 1484.86 L1842.1 1480.6 Q1843.86 1481.44 1845.67 1481.88 Q1847.47 1482.32 1849.21 1482.32 Q1853.84 1482.32 1856.27 1479.21 Q1858.72 1476.09 1859.07 1469.75 Q1857.73 1471.74 1855.67 1472.8 Q1853.61 1473.87 1851.11 1473.87 Q1845.92 1473.87 1842.89 1470.74 Q1839.88 1467.59 1839.88 1462.15 Q1839.88 1456.83 1843.03 1453.61 Q1846.18 1450.39 1851.41 1450.39 Q1857.4 1450.39 1860.55 1455 Q1863.72 1459.58 1863.72 1468.33 Q1863.72 1476.51 1859.84 1481.39 Q1855.97 1486.25 1849.42 1486.25 Q1847.66 1486.25 1845.85 1485.9 Q1844.05 1485.56 1842.1 1484.86 M1851.41 1470.21 Q1854.56 1470.21 1856.39 1468.06 Q1858.24 1465.9 1858.24 1462.15 Q1858.24 1458.43 1856.39 1456.27 Q1854.56 1454.1 1851.41 1454.1 Q1848.26 1454.1 1846.41 1456.27 Q1844.58 1458.43 1844.58 1462.15 Q1844.58 1465.9 1846.41 1468.06 Q1848.26 1470.21 1851.41 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1872.17 1451.02 L1890.53 1451.02 L1890.53 1454.96 L1876.46 1454.96 L1876.46 1463.43 Q1877.47 1463.08 1878.49 1462.92 Q1879.51 1462.73 1880.53 1462.73 Q1886.32 1462.73 1889.7 1465.9 Q1893.08 1469.08 1893.08 1474.49 Q1893.08 1480.07 1889.6 1483.17 Q1886.13 1486.25 1879.81 1486.25 Q1877.64 1486.25 1875.37 1485.88 Q1873.12 1485.51 1870.71 1484.77 L1870.71 1480.07 Q1872.8 1481.2 1875.02 1481.76 Q1877.24 1482.32 1879.72 1482.32 Q1883.72 1482.32 1886.06 1480.21 Q1888.4 1478.1 1888.4 1474.49 Q1888.4 1470.88 1886.06 1468.77 Q1883.72 1466.67 1879.72 1466.67 Q1877.84 1466.67 1875.97 1467.08 Q1874.12 1467.5 1872.17 1468.38 L1872.17 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1912.29 1454.1 Q1908.68 1454.1 1906.85 1457.66 Q1905.04 1461.2 1905.04 1468.33 Q1905.04 1475.44 1906.85 1479.01 Q1908.68 1482.55 1912.29 1482.55 Q1915.92 1482.55 1917.73 1479.01 Q1919.56 1475.44 1919.56 1468.33 Q1919.56 1461.2 1917.73 1457.66 Q1915.92 1454.1 1912.29 1454.1 M1912.29 1450.39 Q1918.1 1450.39 1921.15 1455 Q1924.23 1459.58 1924.23 1468.33 Q1924.23 1477.06 1921.15 1481.67 Q1918.1 1486.25 1912.29 1486.25 Q1906.48 1486.25 1903.4 1481.67 Q1900.34 1477.06 1900.34 1468.33 Q1900.34 1459.58 1903.4 1455 Q1906.48 1450.39 1912.29 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2312.8 1484.86 L2312.8 1480.6 Q2314.56 1481.44 2316.37 1481.88 Q2318.17 1482.32 2319.91 1482.32 Q2324.54 1482.32 2326.97 1479.21 Q2329.42 1476.09 2329.77 1469.75 Q2328.43 1471.74 2326.37 1472.8 Q2324.31 1473.87 2321.81 1473.87 Q2316.62 1473.87 2313.59 1470.74 Q2310.58 1467.59 2310.58 1462.15 Q2310.58 1456.83 2313.73 1453.61 Q2316.88 1450.39 2322.11 1450.39 Q2328.1 1450.39 2331.25 1455 Q2334.42 1459.58 2334.42 1468.33 Q2334.42 1476.51 2330.53 1481.39 Q2326.67 1486.25 2320.12 1486.25 Q2318.36 1486.25 2316.55 1485.9 Q2314.75 1485.56 2312.8 1484.86 M2322.11 1470.21 Q2325.26 1470.21 2327.08 1468.06 Q2328.94 1465.9 2328.94 1462.15 Q2328.94 1458.43 2327.08 1456.27 Q2325.26 1454.1 2322.11 1454.1 Q2318.96 1454.1 2317.11 1456.27 Q2315.28 1458.43 2315.28 1462.15 Q2315.28 1465.9 2317.11 1468.06 Q2318.96 1470.21 2322.11 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2353.4 1466.44 Q2350.26 1466.44 2348.4 1468.59 Q2346.58 1470.74 2346.58 1474.49 Q2346.58 1478.22 2348.4 1480.39 Q2350.26 1482.55 2353.4 1482.55 Q2356.55 1482.55 2358.38 1480.39 Q2360.23 1478.22 2360.23 1474.49 Q2360.23 1470.74 2358.38 1468.59 Q2356.55 1466.44 2353.4 1466.44 M2362.69 1451.78 L2362.69 1456.04 Q2360.93 1455.21 2359.12 1454.77 Q2357.34 1454.33 2355.58 1454.33 Q2350.95 1454.33 2348.5 1457.45 Q2346.07 1460.58 2345.72 1466.9 Q2347.08 1464.89 2349.14 1463.82 Q2351.2 1462.73 2353.68 1462.73 Q2358.89 1462.73 2361.9 1465.9 Q2364.93 1469.05 2364.93 1474.49 Q2364.93 1479.82 2361.78 1483.03 Q2358.64 1486.25 2353.4 1486.25 Q2347.41 1486.25 2344.24 1481.67 Q2341.07 1477.06 2341.07 1468.33 Q2341.07 1460.14 2344.96 1455.28 Q2348.84 1450.39 2355.39 1450.39 Q2357.15 1450.39 2358.94 1450.74 Q2360.74 1451.09 2362.69 1451.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2382.99 1454.1 Q2379.38 1454.1 2377.55 1457.66 Q2375.74 1461.2 2375.74 1468.33 Q2375.74 1475.44 2377.55 1479.01 Q2379.38 1482.55 2382.99 1482.55 Q2386.62 1482.55 2388.43 1479.01 Q2390.26 1475.44 2390.26 1468.33 Q2390.26 1461.2 2388.43 1457.66 Q2386.62 1454.1 2382.99 1454.1 M2382.99 1450.39 Q2388.8 1450.39 2391.85 1455 Q2394.93 1459.58 2394.93 1468.33 Q2394.93 1477.06 2391.85 1481.67 Q2388.8 1486.25 2382.99 1486.25 Q2377.18 1486.25 2374.1 1481.67 Q2371.04 1477.06 2371.04 1468.33 Q2371.04 1459.58 2374.1 1455 Q2377.18 1450.39 2382.99 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1219.08 1525.81 L1219.08 1543.66 L1227.16 1543.66 Q1231.65 1543.66 1234.1 1541.34 Q1236.55 1539.02 1236.55 1534.72 Q1236.55 1530.45 1234.1 1528.13 Q1231.65 1525.81 1227.16 1525.81 L1219.08 1525.81 M1212.65 1520.52 L1227.16 1520.52 Q1235.15 1520.52 1239.23 1524.15 Q1243.33 1527.75 1243.33 1534.72 Q1243.33 1541.75 1239.23 1545.35 Q1235.15 1548.95 1227.16 1548.95 L1219.08 1548.95 L1219.08 1568.04 L1212.65 1568.04 L1212.65 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1271.21 1537.87 Q1270.23 1537.3 1269.05 1537.04 Q1267.9 1536.76 1266.5 1536.76 Q1261.54 1536.76 1258.86 1540 Q1256.22 1543.22 1256.22 1549.27 L1256.22 1568.04 L1250.33 1568.04 L1250.33 1532.4 L1256.22 1532.4 L1256.22 1537.93 Q1258.07 1534.69 1261.03 1533.13 Q1263.99 1531.54 1268.22 1531.54 Q1268.83 1531.54 1269.56 1531.63 Q1270.29 1531.7 1271.18 1531.85 L1271.21 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1289.74 1536.5 Q1285.03 1536.5 1282.29 1540.19 Q1279.55 1543.85 1279.55 1550.25 Q1279.55 1556.65 1282.26 1560.34 Q1284.99 1564 1289.74 1564 Q1294.42 1564 1297.15 1560.31 Q1299.89 1556.62 1299.89 1550.25 Q1299.89 1543.92 1297.15 1540.23 Q1294.42 1536.5 1289.74 1536.5 M1289.74 1531.54 Q1297.38 1531.54 1301.74 1536.5 Q1306.1 1541.47 1306.1 1550.25 Q1306.1 1559 1301.74 1564 Q1297.38 1568.97 1289.74 1568.97 Q1282.07 1568.97 1277.71 1564 Q1273.38 1559 1273.38 1550.25 Q1273.38 1541.47 1277.71 1536.5 Q1282.07 1531.54 1289.74 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1333.85 1518.52 L1333.85 1523.39 L1328.25 1523.39 Q1325.1 1523.39 1323.86 1524.66 Q1322.65 1525.93 1322.65 1529.24 L1322.65 1532.4 L1332.29 1532.4 L1332.29 1536.95 L1322.65 1536.95 L1322.65 1568.04 L1316.76 1568.04 L1316.76 1536.95 L1311.16 1536.95 L1311.16 1532.4 L1316.76 1532.4 L1316.76 1529.91 Q1316.76 1523.96 1319.53 1521.26 Q1322.3 1518.52 1328.31 1518.52 L1333.85 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1338.75 1532.4 L1344.61 1532.4 L1344.61 1568.04 L1338.75 1568.04 L1338.75 1532.4 M1338.75 1518.52 L1344.61 1518.52 L1344.61 1525.93 L1338.75 1525.93 L1338.75 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1362.66 1522.27 L1362.66 1532.4 L1374.72 1532.4 L1374.72 1536.95 L1362.66 1536.95 L1362.66 1556.3 Q1362.66 1560.66 1363.83 1561.9 Q1365.04 1563.14 1368.7 1563.14 L1374.72 1563.14 L1374.72 1568.04 L1368.7 1568.04 Q1361.92 1568.04 1359.35 1565.53 Q1356.77 1562.98 1356.77 1556.3 L1356.77 1536.95 L1352.47 1536.95 L1352.47 1532.4 L1356.77 1532.4 L1356.77 1522.27 L1362.66 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,1163.46 2352.76,1163.46 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,844.983 2352.76,844.983 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,526.506 2352.76,526.506 "/>
+<polyline clip-path="url(#clip852)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="234.612,208.028 2352.76,208.028 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1423.18 234.612,123.472 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,1163.46 253.509,1163.46 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,844.983 253.509,844.983 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,526.506 253.509,526.506 "/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="234.612,208.028 253.509,208.028 "/>
+<path clip-path="url(#clip850)" d="M116.482 1180.02 L116.482 1175.76 Q118.242 1176.6 120.047 1177.04 Q121.853 1177.48 123.589 1177.48 Q128.218 1177.48 130.649 1174.38 Q133.103 1171.25 133.45 1164.91 Q132.107 1166.9 130.047 1167.96 Q127.987 1169.03 125.487 1169.03 Q120.302 1169.03 117.269 1165.9 Q114.26 1162.76 114.26 1157.32 Q114.26 1151.99 117.408 1148.77 Q120.556 1145.56 125.788 1145.56 Q131.783 1145.56 134.931 1150.16 Q138.103 1154.75 138.103 1163.5 Q138.103 1171.67 134.214 1176.55 Q130.348 1181.41 123.797 1181.41 Q122.038 1181.41 120.232 1181.07 Q118.427 1180.72 116.482 1180.02 M125.788 1165.37 Q128.936 1165.37 130.765 1163.22 Q132.617 1161.07 132.617 1157.32 Q132.617 1153.59 130.765 1151.44 Q128.936 1149.26 125.788 1149.26 Q122.64 1149.26 120.788 1151.44 Q118.959 1153.59 118.959 1157.32 Q118.959 1161.07 120.788 1163.22 Q122.64 1165.37 125.788 1165.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M150.533 1176.81 L166.852 1176.81 L166.852 1180.74 L144.908 1180.74 L144.908 1176.81 Q147.57 1174.05 152.153 1169.42 Q156.76 1164.77 157.94 1163.43 Q160.186 1160.9 161.065 1159.17 Q161.968 1157.41 161.968 1155.72 Q161.968 1152.96 160.024 1151.23 Q158.103 1149.49 155.001 1149.49 Q152.802 1149.49 150.348 1150.26 Q147.917 1151.02 145.14 1152.57 L145.14 1147.85 Q147.964 1146.71 150.417 1146.13 Q152.871 1145.56 154.908 1145.56 Q160.278 1145.56 163.473 1148.24 Q166.667 1150.93 166.667 1155.42 Q166.667 1157.55 165.857 1159.47 Q165.07 1161.37 162.964 1163.96 Q162.385 1164.63 159.283 1167.85 Q156.181 1171.04 150.533 1176.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M186.667 1149.26 Q183.056 1149.26 181.227 1152.82 Q179.422 1156.37 179.422 1163.5 Q179.422 1170.6 181.227 1174.17 Q183.056 1177.71 186.667 1177.71 Q190.301 1177.71 192.107 1174.17 Q193.936 1170.6 193.936 1163.5 Q193.936 1156.37 192.107 1152.82 Q190.301 1149.26 186.667 1149.26 M186.667 1145.56 Q192.477 1145.56 195.533 1150.16 Q198.612 1154.75 198.612 1163.5 Q198.612 1172.22 195.533 1176.83 Q192.477 1181.41 186.667 1181.41 Q180.857 1181.41 177.778 1176.83 Q174.723 1172.22 174.723 1163.5 Q174.723 1154.75 177.778 1150.16 Q180.857 1145.56 186.667 1145.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M116.482 861.546 L116.482 857.287 Q118.242 858.12 120.047 858.56 Q121.853 859 123.589 859 Q128.218 859 130.649 855.898 Q133.103 852.773 133.45 846.43 Q132.107 848.421 130.047 849.486 Q127.987 850.551 125.487 850.551 Q120.302 850.551 117.269 847.426 Q114.26 844.277 114.26 838.838 Q114.26 833.514 117.408 830.296 Q120.556 827.078 125.788 827.078 Q131.783 827.078 134.931 831.685 Q138.103 836.268 138.103 845.018 Q138.103 853.189 134.214 858.074 Q130.348 862.935 123.797 862.935 Q122.038 862.935 120.232 862.588 Q118.427 862.24 116.482 861.546 M125.788 846.893 Q128.936 846.893 130.765 844.74 Q132.617 842.588 132.617 838.838 Q132.617 835.111 130.765 832.958 Q128.936 830.782 125.788 830.782 Q122.64 830.782 120.788 832.958 Q118.959 835.111 118.959 838.838 Q118.959 842.588 120.788 844.74 Q122.64 846.893 125.788 846.893 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M159.353 831.778 L147.547 850.227 L159.353 850.227 L159.353 831.778 M158.126 827.703 L164.005 827.703 L164.005 850.227 L168.936 850.227 L168.936 854.115 L164.005 854.115 L164.005 862.263 L159.353 862.263 L159.353 854.115 L143.751 854.115 L143.751 849.602 L158.126 827.703 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M186.667 830.782 Q183.056 830.782 181.227 834.347 Q179.422 837.889 179.422 845.018 Q179.422 852.125 181.227 855.689 Q183.056 859.231 186.667 859.231 Q190.301 859.231 192.107 855.689 Q193.936 852.125 193.936 845.018 Q193.936 837.889 192.107 834.347 Q190.301 830.782 186.667 830.782 M186.667 827.078 Q192.477 827.078 195.533 831.685 Q198.612 836.268 198.612 845.018 Q198.612 853.745 195.533 858.351 Q192.477 862.935 186.667 862.935 Q180.857 862.935 177.778 858.351 Q174.723 853.745 174.723 845.018 Q174.723 836.268 177.778 831.685 Q180.857 827.078 186.667 827.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M116.482 543.068 L116.482 538.809 Q118.242 539.642 120.047 540.082 Q121.853 540.522 123.589 540.522 Q128.218 540.522 130.649 537.42 Q133.103 534.295 133.45 527.952 Q132.107 529.943 130.047 531.008 Q127.987 532.073 125.487 532.073 Q120.302 532.073 117.269 528.948 Q114.26 525.8 114.26 520.36 Q114.26 515.036 117.408 511.818 Q120.556 508.601 125.788 508.601 Q131.783 508.601 134.931 513.207 Q138.103 517.79 138.103 526.54 Q138.103 534.712 134.214 539.596 Q130.348 544.457 123.797 544.457 Q122.038 544.457 120.232 544.11 Q118.427 543.763 116.482 543.068 M125.788 528.415 Q128.936 528.415 130.765 526.263 Q132.617 524.11 132.617 520.36 Q132.617 516.633 130.765 514.48 Q128.936 512.304 125.788 512.304 Q122.64 512.304 120.788 514.48 Q118.959 516.633 118.959 520.36 Q118.959 524.11 120.788 526.263 Q122.64 528.415 125.788 528.415 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M157.084 524.642 Q153.936 524.642 152.084 526.795 Q150.255 528.948 150.255 532.698 Q150.255 536.425 152.084 538.601 Q153.936 540.753 157.084 540.753 Q160.232 540.753 162.061 538.601 Q163.913 536.425 163.913 532.698 Q163.913 528.948 162.061 526.795 Q160.232 524.642 157.084 524.642 M166.366 509.99 L166.366 514.249 Q164.607 513.416 162.802 512.976 Q161.019 512.536 159.26 512.536 Q154.63 512.536 152.177 515.661 Q149.746 518.786 149.399 525.105 Q150.765 523.091 152.825 522.027 Q154.885 520.939 157.362 520.939 Q162.57 520.939 165.579 524.11 Q168.612 527.258 168.612 532.698 Q168.612 538.022 165.464 541.239 Q162.315 544.457 157.084 544.457 Q151.089 544.457 147.917 539.874 Q144.746 535.267 144.746 526.54 Q144.746 518.346 148.635 513.485 Q152.524 508.601 159.075 508.601 Q160.834 508.601 162.616 508.948 Q164.422 509.295 166.366 509.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M186.667 512.304 Q183.056 512.304 181.227 515.869 Q179.422 519.411 179.422 526.54 Q179.422 533.647 181.227 537.212 Q183.056 540.753 186.667 540.753 Q190.301 540.753 192.107 537.212 Q193.936 533.647 193.936 526.54 Q193.936 519.411 192.107 515.869 Q190.301 512.304 186.667 512.304 M186.667 508.601 Q192.477 508.601 195.533 513.207 Q198.612 517.79 198.612 526.54 Q198.612 535.267 195.533 539.874 Q192.477 544.457 186.667 544.457 Q180.857 544.457 177.778 539.874 Q174.723 535.267 174.723 526.54 Q174.723 517.79 177.778 513.207 Q180.857 508.601 186.667 508.601 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M116.482 224.59 L116.482 220.331 Q118.242 221.164 120.047 221.604 Q121.853 222.044 123.589 222.044 Q128.218 222.044 130.649 218.942 Q133.103 215.817 133.45 209.475 Q132.107 211.465 130.047 212.53 Q127.987 213.595 125.487 213.595 Q120.302 213.595 117.269 210.47 Q114.26 207.322 114.26 201.882 Q114.26 196.558 117.408 193.341 Q120.556 190.123 125.788 190.123 Q131.783 190.123 134.931 194.729 Q138.103 199.313 138.103 208.063 Q138.103 216.234 134.214 221.118 Q130.348 225.979 123.797 225.979 Q122.038 225.979 120.232 225.632 Q118.427 225.285 116.482 224.59 M125.788 209.938 Q128.936 209.938 130.765 207.785 Q132.617 205.632 132.617 201.882 Q132.617 198.155 130.765 196.003 Q128.936 193.827 125.788 193.827 Q122.64 193.827 120.788 196.003 Q118.959 198.155 118.959 201.882 Q118.959 205.632 120.788 207.785 Q122.64 209.938 125.788 209.938 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M156.505 208.896 Q153.172 208.896 151.251 210.678 Q149.353 212.461 149.353 215.586 Q149.353 218.711 151.251 220.493 Q153.172 222.276 156.505 222.276 Q159.839 222.276 161.76 220.493 Q163.681 218.688 163.681 215.586 Q163.681 212.461 161.76 210.678 Q159.862 208.896 156.505 208.896 M151.829 206.905 Q148.82 206.165 147.13 204.104 Q145.464 202.044 145.464 199.081 Q145.464 194.938 148.403 192.53 Q151.366 190.123 156.505 190.123 Q161.667 190.123 164.607 192.53 Q167.547 194.938 167.547 199.081 Q167.547 202.044 165.857 204.104 Q164.19 206.165 161.204 206.905 Q164.584 207.692 166.459 209.984 Q168.357 212.276 168.357 215.586 Q168.357 220.609 165.278 223.294 Q162.223 225.979 156.505 225.979 Q150.788 225.979 147.709 223.294 Q144.654 220.609 144.654 215.586 Q144.654 212.276 146.552 209.984 Q148.45 207.692 151.829 206.905 M150.116 199.521 Q150.116 202.206 151.783 203.711 Q153.473 205.215 156.505 205.215 Q159.515 205.215 161.204 203.711 Q162.917 202.206 162.917 199.521 Q162.917 196.836 161.204 195.331 Q159.515 193.827 156.505 193.827 Q153.473 193.827 151.783 195.331 Q150.116 196.836 150.116 199.521 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M186.667 193.827 Q183.056 193.827 181.227 197.391 Q179.422 200.933 179.422 208.063 Q179.422 215.169 181.227 218.734 Q183.056 222.276 186.667 222.276 Q190.301 222.276 192.107 218.734 Q193.936 215.169 193.936 208.063 Q193.936 200.933 192.107 197.391 Q190.301 193.827 186.667 193.827 M186.667 190.123 Q192.477 190.123 195.533 194.729 Q198.612 199.313 198.612 208.063 Q198.612 216.789 195.533 221.396 Q192.477 225.979 186.667 225.979 Q180.857 225.979 177.778 221.396 Q174.723 216.789 174.723 208.063 Q174.723 199.313 177.778 194.729 Q180.857 190.123 186.667 190.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M21.7677 865.883 L58.7206 865.883 L58.7206 858.117 Q58.7206 848.282 54.2646 843.731 Q49.8086 839.147 40.1964 839.147 Q30.6479 839.147 26.2237 843.731 Q21.7677 848.282 21.7677 858.117 L21.7677 865.883 M16.4842 872.313 L16.4842 859.104 Q16.4842 845.29 22.2451 838.829 Q27.9743 832.368 40.1964 832.368 Q52.4822 832.368 58.2432 838.861 Q64.0042 845.354 64.0042 859.104 L64.0042 872.313 L16.4842 872.313 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M44.7161 791.882 L47.5806 791.882 L47.5806 818.809 Q53.6281 818.427 56.8109 815.181 Q59.9619 811.902 59.9619 806.078 Q59.9619 802.704 59.1344 799.553 Q58.3069 796.37 56.6518 793.251 L62.1899 793.251 Q63.5267 796.402 64.227 799.712 Q64.9272 803.022 64.9272 806.428 Q64.9272 814.958 59.9619 819.955 Q54.9967 824.92 46.5303 824.92 Q37.7774 824.92 32.6531 820.209 Q27.4968 815.467 27.4968 807.446 Q27.4968 800.253 32.1438 796.083 Q36.7589 791.882 44.7161 791.882 M42.9973 797.738 Q38.1912 797.802 35.3266 800.444 Q32.4621 803.054 32.4621 807.383 Q32.4621 812.284 35.2312 815.244 Q38.0002 818.172 43.0292 818.618 L42.9973 797.738 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M29.4065 759.544 L34.9447 759.544 Q33.6716 762.027 33.035 764.7 Q32.3984 767.374 32.3984 770.239 Q32.3984 774.599 33.7352 776.795 Q35.072 778.96 37.7456 778.96 Q39.7826 778.96 40.9603 777.4 Q42.1061 775.84 43.1565 771.13 L43.6021 769.125 Q44.9389 762.886 47.3897 760.276 Q49.8086 757.635 54.1691 757.635 Q59.1344 757.635 62.0308 761.581 Q64.9272 765.496 64.9272 772.371 Q64.9272 775.236 64.3543 778.355 Q63.8132 781.442 62.6992 784.88 L56.6518 784.88 Q58.3387 781.633 59.198 778.482 Q60.0256 775.331 60.0256 772.244 Q60.0256 768.106 58.6251 765.878 Q57.1929 763.65 54.6147 763.65 Q52.2276 763.65 50.9545 765.273 Q49.6813 766.865 48.5037 772.307 L48.0262 774.344 Q46.8804 779.787 44.5251 782.206 Q42.138 784.625 38.0002 784.625 Q32.9713 784.625 30.2341 781.06 Q27.4968 777.496 27.4968 770.939 Q27.4968 767.692 27.9743 764.828 Q28.4517 761.963 29.4065 759.544 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M28.3562 748.309 L28.3562 742.452 L64.0042 742.452 L64.0042 748.309 L28.3562 748.309 M14.479 748.309 L14.479 742.452 L21.895 742.452 L21.895 748.309 L14.479 748.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M33.8307 709.542 Q33.2578 710.528 33.0032 711.706 Q32.7167 712.852 32.7167 714.252 Q32.7167 719.217 35.9632 721.891 Q39.1779 724.533 45.2253 724.533 L64.0042 724.533 L64.0042 730.421 L28.3562 730.421 L28.3562 724.533 L33.8944 724.533 Q30.6479 722.687 29.0883 719.727 Q27.4968 716.767 27.4968 712.533 Q27.4968 711.929 27.5923 711.197 Q27.656 710.465 27.8151 709.573 L33.8307 709.542 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M44.7161 674.339 L47.5806 674.339 L47.5806 701.266 Q53.6281 700.884 56.8109 697.638 Q59.9619 694.359 59.9619 688.535 Q59.9619 685.161 59.1344 682.01 Q58.3069 678.827 56.6518 675.708 L62.1899 675.708 Q63.5267 678.859 64.227 682.169 Q64.9272 685.479 64.9272 688.885 Q64.9272 697.415 59.9619 702.412 Q54.9967 707.377 46.5303 707.377 Q37.7774 707.377 32.6531 702.667 Q27.4968 697.924 27.4968 689.903 Q27.4968 682.71 32.1438 678.541 Q36.7589 674.339 44.7161 674.339 M42.9973 680.196 Q38.1912 680.259 35.3266 682.901 Q32.4621 685.511 32.4621 689.84 Q32.4621 694.741 35.2312 697.701 Q38.0002 700.63 43.0292 701.075 L42.9973 680.196 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M994.686 17.6457 Q985.775 17.6457 980.508 24.2892 Q975.283 30.9327 975.283 42.3968 Q975.283 53.8203 980.508 60.4638 Q985.775 67.1073 994.686 67.1073 Q1003.6 67.1073 1008.78 60.4638 Q1014.01 53.8203 1014.01 42.3968 Q1014.01 30.9327 1008.78 24.2892 Q1003.6 17.6457 994.686 17.6457 M994.686 11.0023 Q1007.41 11.0023 1015.02 19.5497 Q1022.64 28.0566 1022.64 42.3968 Q1022.64 56.6965 1015.02 65.2439 Q1007.41 73.7508 994.686 73.7508 Q981.926 73.7508 974.27 65.2439 Q966.654 56.737 966.654 42.3968 Q966.654 28.0566 974.27 19.5497 Q981.926 11.0023 994.686 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1067.68 49.9314 Q1067.68 41.7081 1064.28 37.0496 Q1060.92 32.3505 1055 32.3505 Q1049.09 32.3505 1045.69 37.0496 Q1042.33 41.7081 1042.33 49.9314 Q1042.33 58.1548 1045.69 62.8538 Q1049.09 67.5124 1055 67.5124 Q1060.92 67.5124 1064.28 62.8538 Q1067.68 58.1548 1067.68 49.9314 M1042.33 34.0924 Q1044.67 30.0415 1048.24 28.0971 Q1051.84 26.1121 1056.83 26.1121 Q1065.09 26.1121 1070.24 32.6746 Q1075.42 39.2371 1075.42 49.9314 Q1075.42 60.6258 1070.24 67.1883 Q1065.09 73.7508 1056.83 73.7508 Q1051.84 73.7508 1048.24 71.8063 Q1044.67 69.8214 1042.33 65.7705 L1042.33 72.576 L1034.83 72.576 L1034.83 9.54393 L1042.33 9.54393 L1042.33 34.0924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1087.78 27.2059 L1095.23 27.2059 L1095.23 73.3862 Q1095.23 82.0551 1091.91 85.944 Q1088.63 89.8329 1081.29 89.8329 L1078.46 89.8329 L1078.46 83.5134 L1080.44 83.5134 Q1084.7 83.5134 1086.24 81.5285 Q1087.78 79.5841 1087.78 73.3862 L1087.78 27.2059 M1087.78 9.54393 L1095.23 9.54393 L1095.23 18.9825 L1087.78 18.9825 L1087.78 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1149.63 48.0275 L1149.63 51.6733 L1115.36 51.6733 Q1115.85 59.3701 1119.98 63.421 Q1124.15 67.4314 1131.57 67.4314 Q1135.86 67.4314 1139.87 66.3781 Q1143.92 65.3249 1147.89 63.2184 L1147.89 70.267 Q1143.88 71.9684 1139.67 72.8596 Q1135.46 73.7508 1131.12 73.7508 Q1120.26 73.7508 1113.9 67.4314 Q1107.59 61.1119 1107.59 50.3365 Q1107.59 39.1965 1113.58 32.6746 Q1119.62 26.1121 1129.82 26.1121 Q1138.98 26.1121 1144.29 32.0264 Q1149.63 37.9003 1149.63 48.0275 M1142.18 45.84 Q1142.1 39.7232 1138.74 36.0774 Q1135.41 32.4315 1129.91 32.4315 Q1123.67 32.4315 1119.9 35.9558 Q1116.17 39.4801 1115.61 45.8805 L1142.18 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1194.52 28.9478 L1194.52 35.9153 Q1191.36 34.1734 1188.16 33.3227 Q1185 32.4315 1181.76 32.4315 Q1174.51 32.4315 1170.5 37.0496 Q1166.49 41.6271 1166.49 49.9314 Q1166.49 58.2358 1170.5 62.8538 Q1174.51 67.4314 1181.76 67.4314 Q1185 67.4314 1188.16 66.5807 Q1191.36 65.6895 1194.52 63.9476 L1194.52 70.8341 Q1191.4 72.2924 1188.04 73.0216 Q1184.71 73.7508 1180.95 73.7508 Q1170.7 73.7508 1164.66 67.3098 Q1158.63 60.8689 1158.63 49.9314 Q1158.63 38.832 1164.7 32.472 Q1170.82 26.1121 1181.43 26.1121 Q1184.88 26.1121 1188.16 26.8413 Q1191.44 27.5299 1194.52 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1214.85 14.324 L1214.85 27.2059 L1230.21 27.2059 L1230.21 32.9987 L1214.85 32.9987 L1214.85 57.6282 Q1214.85 63.1779 1216.35 64.7578 Q1217.89 66.3376 1222.55 66.3376 L1230.21 66.3376 L1230.21 72.576 L1222.55 72.576 Q1213.92 72.576 1210.64 69.3758 Q1207.36 66.1351 1207.36 57.6282 L1207.36 32.9987 L1201.89 32.9987 L1201.89 27.2059 L1207.36 27.2059 L1207.36 14.324 L1214.85 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1240.01 27.2059 L1247.46 27.2059 L1247.46 72.576 L1240.01 72.576 L1240.01 27.2059 M1240.01 9.54393 L1247.46 9.54393 L1247.46 18.9825 L1240.01 18.9825 L1240.01 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1257.71 27.2059 L1265.61 27.2059 L1279.79 65.2844 L1293.97 27.2059 L1301.87 27.2059 L1284.85 72.576 L1274.73 72.576 L1257.71 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1350.96 48.0275 L1350.96 51.6733 L1316.69 51.6733 Q1317.18 59.3701 1321.31 63.421 Q1325.48 67.4314 1332.9 67.4314 Q1337.19 67.4314 1341.2 66.3781 Q1345.25 65.3249 1349.22 63.2184 L1349.22 70.267 Q1345.21 71.9684 1341 72.8596 Q1336.79 73.7508 1332.45 73.7508 Q1321.59 73.7508 1315.23 67.4314 Q1308.92 61.1119 1308.92 50.3365 Q1308.92 39.1965 1314.91 32.6746 Q1320.95 26.1121 1331.15 26.1121 Q1340.31 26.1121 1345.62 32.0264 Q1350.96 37.9003 1350.96 48.0275 M1343.51 45.84 Q1343.43 39.7232 1340.07 36.0774 Q1336.74 32.4315 1331.24 32.4315 Q1325 32.4315 1321.23 35.9558 Q1317.5 39.4801 1316.94 45.8805 L1343.51 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1418.49 28.5427 L1418.49 35.5912 Q1415.33 33.9709 1411.93 33.1607 Q1408.53 32.3505 1404.88 32.3505 Q1399.33 32.3505 1396.54 34.0519 Q1393.78 35.7533 1393.78 39.156 Q1393.78 41.7486 1395.77 43.2475 Q1397.75 44.7058 1403.75 46.0426 L1406.3 46.6097 Q1414.24 48.3111 1417.56 51.4303 Q1420.92 54.509 1420.92 60.0587 Q1420.92 66.3781 1415.9 70.0644 Q1410.92 73.7508 1402.17 73.7508 Q1398.52 73.7508 1394.55 73.0216 Q1390.62 72.3329 1386.25 70.9151 L1386.25 63.2184 Q1390.38 65.3654 1394.39 66.4591 Q1398.4 67.5124 1402.33 67.5124 Q1407.6 67.5124 1410.43 65.73 Q1413.27 63.9071 1413.27 60.6258 Q1413.27 57.5877 1411.2 55.9673 Q1409.18 54.3469 1402.25 52.8481 L1399.66 52.2405 Q1392.73 50.7821 1389.65 47.7845 Q1386.57 44.7463 1386.57 39.4801 Q1386.57 33.0797 1391.11 29.5959 Q1395.64 26.1121 1403.99 26.1121 Q1408.12 26.1121 1411.77 26.7198 Q1415.41 27.3274 1418.49 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1440 65.7705 L1440 89.8329 L1432.51 89.8329 L1432.51 27.2059 L1440 27.2059 L1440 34.0924 Q1442.35 30.0415 1445.92 28.0971 Q1449.52 26.1121 1454.5 26.1121 Q1462.77 26.1121 1467.91 32.6746 Q1473.1 39.2371 1473.1 49.9314 Q1473.1 60.6258 1467.91 67.1883 Q1462.77 73.7508 1454.5 73.7508 Q1449.52 73.7508 1445.92 71.8063 Q1442.35 69.8214 1440 65.7705 M1465.36 49.9314 Q1465.36 41.7081 1461.96 37.0496 Q1458.6 32.3505 1452.68 32.3505 Q1446.77 32.3505 1443.36 37.0496 Q1440 41.7081 1440 49.9314 Q1440 58.1548 1443.36 62.8538 Q1446.77 67.5124 1452.68 67.5124 Q1458.6 67.5124 1461.96 62.8538 Q1465.36 58.1548 1465.36 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1506.07 49.7694 Q1497.04 49.7694 1493.56 51.8354 Q1490.07 53.9013 1490.07 58.8839 Q1490.07 62.8538 1492.66 65.2034 Q1495.3 67.5124 1499.79 67.5124 Q1505.99 67.5124 1509.72 63.1374 Q1513.49 58.7219 1513.49 51.4303 L1513.49 49.7694 L1506.07 49.7694 M1520.94 46.6907 L1520.94 72.576 L1513.49 72.576 L1513.49 65.6895 Q1510.93 69.8214 1507.13 71.8063 Q1503.32 73.7508 1497.81 73.7508 Q1490.84 73.7508 1486.71 69.8619 Q1482.62 65.9325 1482.62 59.3701 Q1482.62 51.7138 1487.72 47.825 Q1492.87 43.9361 1503.03 43.9361 L1513.49 43.9361 L1513.49 43.2069 Q1513.49 38.0623 1510.08 35.2672 Q1506.72 32.4315 1500.6 32.4315 Q1496.72 32.4315 1493.03 33.3632 Q1489.34 34.295 1485.94 36.1584 L1485.94 29.2718 Q1490.03 27.692 1493.88 26.9223 Q1497.73 26.1121 1501.37 26.1121 Q1511.22 26.1121 1516.08 31.2163 Q1520.94 36.3204 1520.94 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1568.94 28.9478 L1568.94 35.9153 Q1565.78 34.1734 1562.58 33.3227 Q1559.42 32.4315 1556.18 32.4315 Q1548.93 32.4315 1544.92 37.0496 Q1540.91 41.6271 1540.91 49.9314 Q1540.91 58.2358 1544.92 62.8538 Q1548.93 67.4314 1556.18 67.4314 Q1559.42 67.4314 1562.58 66.5807 Q1565.78 65.6895 1568.94 63.9476 L1568.94 70.8341 Q1565.82 72.2924 1562.46 73.0216 Q1559.14 73.7508 1555.37 73.7508 Q1545.12 73.7508 1539.09 67.3098 Q1533.05 60.8689 1533.05 49.9314 Q1533.05 38.832 1539.13 32.472 Q1545.24 26.1121 1555.86 26.1121 Q1559.3 26.1121 1562.58 26.8413 Q1565.86 27.5299 1568.94 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1620.71 48.0275 L1620.71 51.6733 L1586.44 51.6733 Q1586.93 59.3701 1591.06 63.421 Q1595.23 67.4314 1602.65 67.4314 Q1606.94 67.4314 1610.95 66.3781 Q1615 65.3249 1618.97 63.2184 L1618.97 70.267 Q1614.96 71.9684 1610.75 72.8596 Q1606.54 73.7508 1602.2 73.7508 Q1591.34 73.7508 1584.98 67.4314 Q1578.66 61.1119 1578.66 50.3365 Q1578.66 39.1965 1584.66 32.6746 Q1590.7 26.1121 1600.9 26.1121 Q1610.06 26.1121 1615.37 32.0264 Q1620.71 37.9003 1620.71 48.0275 M1613.26 45.84 Q1613.18 39.7232 1609.82 36.0774 Q1606.49 32.4315 1600.99 32.4315 Q1594.75 32.4315 1590.98 35.9558 Q1587.25 39.4801 1586.69 45.8805 L1613.26 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip852)" cx="375.821" cy="160.256" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="799.45" cy="335.419" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1128.94" cy="351.343" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1176.01" cy="733.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1223.08" cy="813.136" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1552.57" cy="844.983" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1787.92" cy="860.907" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="1834.99" cy="1243.08" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="2117.41" cy="1386.4" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip852)" cx="2117.41" cy="160.256" r="14.4" fill="#e26f46" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<path clip-path="url(#clip850)" d="M305.216 1379.86 L762.678 1379.86 L762.678 1276.18 L305.216 1276.18  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip850)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="305.216,1379.86 762.678,1379.86 762.678,1276.18 305.216,1276.18 305.216,1379.86 "/>
+<circle clip-path="url(#clip850)" cx="399.356" cy="1328.02" r="20.48" fill="#e26f46" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.55111"/>
+<path clip-path="url(#clip850)" d="M493.496 1310.74 L498.172 1310.74 L498.172 1345.3 L493.496 1345.3 L493.496 1310.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M524.352 1323.31 L524.352 1309.28 L528.611 1309.28 L528.611 1345.3 L524.352 1345.3 L524.352 1341.41 Q523.01 1343.72 520.949 1344.86 Q518.912 1345.97 516.042 1345.97 Q511.343 1345.97 508.38 1342.22 Q505.44 1338.47 505.44 1332.36 Q505.44 1326.25 508.38 1322.5 Q511.343 1318.75 516.042 1318.75 Q518.912 1318.75 520.949 1319.88 Q523.01 1320.99 524.352 1323.31 M509.838 1332.36 Q509.838 1337.06 511.76 1339.74 Q513.704 1342.4 517.084 1342.4 Q520.463 1342.4 522.408 1339.74 Q524.352 1337.06 524.352 1332.36 Q524.352 1327.66 522.408 1325 Q520.463 1322.31 517.084 1322.31 Q513.704 1322.31 511.76 1325 Q509.838 1327.66 509.838 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M559.56 1331.27 L559.56 1333.35 L539.977 1333.35 Q540.255 1337.75 542.616 1340.06 Q545 1342.36 549.236 1342.36 Q551.69 1342.36 553.982 1341.75 Q556.296 1341.15 558.565 1339.95 L558.565 1343.98 Q556.273 1344.95 553.866 1345.46 Q551.459 1345.97 548.982 1345.97 Q542.778 1345.97 539.144 1342.36 Q535.533 1338.75 535.533 1332.59 Q535.533 1326.22 538.959 1322.5 Q542.408 1318.75 548.241 1318.75 Q553.472 1318.75 556.505 1322.13 Q559.56 1325.48 559.56 1331.27 M555.301 1330.02 Q555.255 1326.52 553.334 1324.44 Q551.435 1322.36 548.287 1322.36 Q544.722 1322.36 542.57 1324.37 Q540.44 1326.38 540.116 1330.04 L555.301 1330.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M578.333 1332.26 Q573.171 1332.26 571.181 1333.44 Q569.19 1334.63 569.19 1337.47 Q569.19 1339.74 570.671 1341.08 Q572.176 1342.4 574.745 1342.4 Q578.287 1342.4 580.417 1339.9 Q582.569 1337.38 582.569 1333.21 L582.569 1332.26 L578.333 1332.26 M586.829 1330.5 L586.829 1345.3 L582.569 1345.3 L582.569 1341.36 Q581.111 1343.72 578.935 1344.86 Q576.759 1345.97 573.611 1345.97 Q569.63 1345.97 567.269 1343.75 Q564.931 1341.5 564.931 1337.75 Q564.931 1333.38 567.847 1331.15 Q570.787 1328.93 576.597 1328.93 L582.569 1328.93 L582.569 1328.51 Q582.569 1325.57 580.625 1323.98 Q578.704 1322.36 575.208 1322.36 Q572.986 1322.36 570.88 1322.89 Q568.773 1323.42 566.829 1324.49 L566.829 1320.55 Q569.167 1319.65 571.366 1319.21 Q573.565 1318.75 575.648 1318.75 Q581.273 1318.75 584.051 1321.66 Q586.829 1324.58 586.829 1330.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M595.602 1309.28 L599.861 1309.28 L599.861 1345.3 L595.602 1345.3 L595.602 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M627.963 1341.41 L627.963 1355.16 L623.68 1355.16 L623.68 1319.37 L627.963 1319.37 L627.963 1323.31 Q629.305 1320.99 631.342 1319.88 Q633.403 1318.75 636.25 1318.75 Q640.972 1318.75 643.912 1322.5 Q646.875 1326.25 646.875 1332.36 Q646.875 1338.47 643.912 1342.22 Q640.972 1345.97 636.25 1345.97 Q633.403 1345.97 631.342 1344.86 Q629.305 1343.72 627.963 1341.41 M642.453 1332.36 Q642.453 1327.66 640.509 1325 Q638.588 1322.31 635.208 1322.31 Q631.828 1322.31 629.884 1325 Q627.963 1327.66 627.963 1332.36 Q627.963 1337.06 629.884 1339.74 Q631.828 1342.4 635.208 1342.4 Q638.588 1342.4 640.509 1339.74 Q642.453 1337.06 642.453 1332.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M663.981 1322.36 Q660.555 1322.36 658.564 1325.04 Q656.574 1327.7 656.574 1332.36 Q656.574 1337.01 658.541 1339.69 Q660.532 1342.36 663.981 1342.36 Q667.384 1342.36 669.375 1339.67 Q671.365 1336.99 671.365 1332.36 Q671.365 1327.75 669.375 1325.06 Q667.384 1322.36 663.981 1322.36 M663.981 1318.75 Q669.537 1318.75 672.708 1322.36 Q675.879 1325.97 675.879 1332.36 Q675.879 1338.72 672.708 1342.36 Q669.537 1345.97 663.981 1345.97 Q658.402 1345.97 655.231 1342.36 Q652.083 1338.72 652.083 1332.36 Q652.083 1325.97 655.231 1322.36 Q658.402 1318.75 663.981 1318.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M682.939 1319.37 L687.199 1319.37 L687.199 1345.3 L682.939 1345.3 L682.939 1319.37 M682.939 1309.28 L687.199 1309.28 L687.199 1314.67 L682.939 1314.67 L682.939 1309.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M717.661 1329.65 L717.661 1345.3 L713.402 1345.3 L713.402 1329.79 Q713.402 1326.11 711.967 1324.28 Q710.532 1322.45 707.661 1322.45 Q704.212 1322.45 702.222 1324.65 Q700.231 1326.85 700.231 1330.64 L700.231 1345.3 L695.948 1345.3 L695.948 1319.37 L700.231 1319.37 L700.231 1323.4 Q701.759 1321.06 703.819 1319.9 Q705.902 1318.75 708.61 1318.75 Q713.078 1318.75 715.37 1321.52 Q717.661 1324.28 717.661 1329.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M730.37 1312.01 L730.37 1319.37 L739.143 1319.37 L739.143 1322.68 L730.37 1322.68 L730.37 1336.75 Q730.37 1339.93 731.226 1340.83 Q732.106 1341.73 734.768 1341.73 L739.143 1341.73 L739.143 1345.3 L734.768 1345.3 Q729.837 1345.3 727.962 1343.47 Q726.087 1341.62 726.087 1336.75 L726.087 1322.68 L722.962 1322.68 L722.962 1319.37 L726.087 1319.37 L726.087 1312.01 L730.37 1312.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M316.729 176.937 L326.277 176.937 L326.277 143.98 L315.89 146.064 L315.89 140.74 L326.219 138.656 L332.064 138.656 L332.064 176.937 L341.613 176.937 L341.613 181.856 L316.729 181.856 L316.729 176.937 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M745.696 352.1 L766.095 352.1 L766.095 357.019 L738.665 357.019 L738.665 352.1 Q741.992 348.657 747.722 342.87 Q753.48 337.054 754.955 335.376 Q757.762 332.222 758.862 330.052 Q759.99 327.853 759.99 325.74 Q759.99 322.297 757.56 320.127 Q755.158 317.957 751.281 317.957 Q748.532 317.957 745.465 318.912 Q742.427 319.866 738.954 321.805 L738.954 315.902 Q742.484 314.485 745.552 313.761 Q748.619 313.038 751.165 313.038 Q757.878 313.038 761.871 316.394 Q765.864 319.751 765.864 325.364 Q765.864 328.026 764.851 330.428 Q763.867 332.8 761.234 336.041 Q760.511 336.88 756.634 340.902 Q752.756 344.895 745.696 352.1 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1087.18 349.65 Q1091.37 350.547 1093.72 353.383 Q1096.09 356.219 1096.09 360.385 Q1096.09 366.78 1091.69 370.281 Q1087.29 373.782 1079.19 373.782 Q1076.47 373.782 1073.58 373.232 Q1070.71 372.711 1067.65 371.641 L1067.65 365.999 Q1070.08 367.416 1072.97 368.14 Q1075.87 368.863 1079.02 368.863 Q1084.52 368.863 1087.38 366.693 Q1090.27 364.523 1090.27 360.385 Q1090.27 356.566 1087.58 354.425 Q1084.92 352.254 1080.15 352.254 L1075.11 352.254 L1075.11 347.451 L1080.38 347.451 Q1084.69 347.451 1086.98 345.744 Q1089.26 344.008 1089.26 340.767 Q1089.26 337.44 1086.89 335.675 Q1084.55 333.881 1080.15 333.881 Q1077.75 333.881 1075 334.401 Q1072.25 334.922 1068.95 336.022 L1068.95 330.814 Q1072.28 329.888 1075.17 329.425 Q1078.09 328.962 1080.67 328.962 Q1087.32 328.962 1091.2 332 Q1095.08 335.009 1095.08 340.16 Q1095.08 343.748 1093.02 346.236 Q1090.97 348.695 1087.18 349.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1132.7 717.009 L1117.94 740.07 L1132.7 740.07 L1132.7 717.009 M1131.17 711.916 L1138.52 711.916 L1138.52 740.07 L1144.68 740.07 L1144.68 744.931 L1138.52 744.931 L1138.52 755.116 L1132.7 755.116 L1132.7 744.931 L1113.2 744.931 L1113.2 739.289 L1131.17 711.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1163.86 791.536 L1186.8 791.536 L1186.8 796.455 L1169.21 796.455 L1169.21 807.045 Q1170.48 806.611 1171.76 806.408 Q1173.03 806.177 1174.3 806.177 Q1181.54 806.177 1185.76 810.141 Q1189.98 814.105 1189.98 820.876 Q1189.98 827.849 1185.64 831.726 Q1181.3 835.575 1173.4 835.575 Q1170.68 835.575 1167.85 835.112 Q1165.04 834.649 1162.03 833.723 L1162.03 827.849 Q1164.64 829.267 1167.42 829.961 Q1170.19 830.656 1173.29 830.656 Q1178.29 830.656 1181.22 828.023 Q1184.14 825.39 1184.14 820.876 Q1184.14 816.362 1181.22 813.729 Q1178.29 811.096 1173.29 811.096 Q1170.95 811.096 1168.6 811.617 Q1166.29 812.137 1163.86 813.237 L1163.86 791.536 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1506 842.654 Q1502.07 842.654 1499.75 845.345 Q1497.47 848.036 1497.47 852.724 Q1497.47 857.382 1499.75 860.102 Q1502.07 862.793 1506 862.793 Q1509.94 862.793 1512.23 860.102 Q1514.54 857.382 1514.54 852.724 Q1514.54 848.036 1512.23 845.345 Q1509.94 842.654 1506 842.654 M1517.61 824.338 L1517.61 829.662 Q1515.41 828.621 1513.15 828.071 Q1510.92 827.521 1508.72 827.521 Q1502.94 827.521 1499.87 831.427 Q1496.83 835.334 1496.4 843.233 Q1498.11 840.716 1500.68 839.385 Q1503.26 838.025 1506.35 838.025 Q1512.86 838.025 1516.62 841.989 Q1520.41 845.924 1520.41 852.724 Q1520.41 859.379 1516.48 863.401 Q1512.54 867.423 1506 867.423 Q1498.51 867.423 1494.55 861.693 Q1490.58 855.935 1490.58 845.027 Q1490.58 834.784 1495.44 828.708 Q1500.3 822.602 1508.49 822.602 Q1510.69 822.602 1512.92 823.036 Q1515.18 823.47 1517.61 824.338 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1726.96 839.307 L1754.74 839.307 L1754.74 841.796 L1739.05 882.507 L1732.95 882.507 L1747.71 844.226 L1726.96 844.226 L1726.96 839.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M1787.92 1244.17 Q1783.75 1244.17 1781.35 1246.39 Q1778.98 1248.62 1778.98 1252.53 Q1778.98 1256.43 1781.35 1258.66 Q1783.75 1260.89 1787.92 1260.89 Q1792.08 1260.89 1794.49 1258.66 Q1796.89 1256.41 1796.89 1252.53 Q1796.89 1248.62 1794.49 1246.39 Q1792.11 1244.17 1787.92 1244.17 M1782.07 1241.68 Q1778.31 1240.75 1776.2 1238.18 Q1774.12 1235.6 1774.12 1231.9 Q1774.12 1226.72 1777.79 1223.71 Q1781.49 1220.7 1787.92 1220.7 Q1794.37 1220.7 1798.04 1223.71 Q1801.72 1226.72 1801.72 1231.9 Q1801.72 1235.6 1799.61 1238.18 Q1797.52 1240.75 1793.79 1241.68 Q1798.02 1242.66 1800.36 1245.53 Q1802.73 1248.39 1802.73 1252.53 Q1802.73 1258.81 1798.88 1262.16 Q1795.06 1265.52 1787.92 1265.52 Q1780.77 1265.52 1776.92 1262.16 Q1773.1 1258.81 1773.1 1252.53 Q1773.1 1248.39 1775.48 1245.53 Q1777.85 1242.66 1782.07 1241.68 M1779.93 1232.45 Q1779.93 1235.8 1782.01 1237.68 Q1784.13 1239.57 1787.92 1239.57 Q1791.68 1239.57 1793.79 1237.68 Q1795.93 1235.8 1795.93 1232.45 Q1795.93 1229.09 1793.79 1227.21 Q1791.68 1225.33 1787.92 1225.33 Q1784.13 1225.33 1782.01 1227.21 Q1779.93 1229.09 1779.93 1232.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip850)" d="M2058.21 1407.1 L2058.21 1401.77 Q2060.41 1402.82 2062.67 1403.37 Q2064.93 1403.92 2067.1 1403.92 Q2072.88 1403.92 2075.92 1400.04 Q2078.99 1396.13 2079.42 1388.2 Q2077.74 1390.69 2075.17 1392.02 Q2072.59 1393.35 2069.47 1393.35 Q2062.99 1393.35 2059.2 1389.45 Q2055.44 1385.51 2055.44 1378.71 Q2055.44 1372.06 2059.37 1368.04 Q2063.31 1364.01 2069.84 1364.01 Q2077.34 1364.01 2081.27 1369.77 Q2085.24 1375.5 2085.24 1386.44 Q2085.24 1396.65 2080.38 1402.76 Q2075.54 1408.83 2067.36 1408.83 Q2065.16 1408.83 2062.9 1408.4 Q2060.64 1407.97 2058.21 1407.1 M2069.84 1388.78 Q2073.78 1388.78 2076.07 1386.09 Q2078.38 1383.4 2078.38 1378.71 Q2078.38 1374.05 2076.07 1371.36 Q2073.78 1368.64 2069.84 1368.64 Q2065.91 1368.64 2063.59 1371.36 Q2061.31 1374.05 2061.31 1378.71 Q2061.31 1383.4 2063.59 1386.09 Q2065.91 1388.78 2069.84 1388.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/linear/multi_objective_knapsack/717b7729.svg b/previews/PR3562/tutorials/linear/multi_objective_knapsack/c5b0bcc4.svg
similarity index 83%
rename from previews/PR3562/tutorials/linear/multi_objective_knapsack/717b7729.svg
rename to previews/PR3562/tutorials/linear/multi_objective_knapsack/c5b0bcc4.svg
index f08bcfd6989..04a35bcbd2b 100644
--- a/previews/PR3562/tutorials/linear/multi_objective_knapsack/717b7729.svg
+++ b/previews/PR3562/tutorials/linear/multi_objective_knapsack/c5b0bcc4.svg
@@ -1,55 +1,55 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip970">
+  <clipPath id="clip790">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip970)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip790)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip971">
+  <clipPath id="clip791">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip970)" d="M204.45 1423.18 L2352.76 1423.18 L2352.76 47.2441 L204.45 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip790)" d="M204.45 1423.18 L2352.76 1423.18 L2352.76 47.2441 L204.45 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip972">
+  <clipPath id="clip792">
     <rect x="204" y="47" width="2149" height="1377"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="659.332,1423.18 659.332,47.2441 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1222.31,1423.18 1222.31,47.2441 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1785.28,1423.18 1785.28,47.2441 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2348.25,1423.18 2348.25,47.2441 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="659.332,1423.18 659.332,1404.28 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1222.31,1423.18 1222.31,1404.28 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1785.28,1423.18 1785.28,1404.28 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2348.25,1423.18 2348.25,1404.28 "/>
-<path clip-path="url(#clip970)" d="M632.689 1451.02 L654.911 1451.02 L654.911 1453.01 L642.365 1485.58 L637.48 1485.58 L649.286 1454.96 L632.689 1454.96 L632.689 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M674.031 1454.1 Q670.42 1454.1 668.591 1457.66 Q666.786 1461.2 666.786 1468.33 Q666.786 1475.44 668.591 1479.01 Q670.42 1482.55 674.031 1482.55 Q677.665 1482.55 679.471 1479.01 Q681.3 1475.44 681.3 1468.33 Q681.3 1461.2 679.471 1457.66 Q677.665 1454.1 674.031 1454.1 M674.031 1450.39 Q679.841 1450.39 682.897 1455 Q685.976 1459.58 685.976 1468.33 Q685.976 1477.06 682.897 1481.67 Q679.841 1486.25 674.031 1486.25 Q668.221 1486.25 665.142 1481.67 Q662.087 1477.06 662.087 1468.33 Q662.087 1459.58 665.142 1455 Q668.221 1450.39 674.031 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1207.18 1469.17 Q1203.84 1469.17 1201.92 1470.95 Q1200.03 1472.73 1200.03 1475.86 Q1200.03 1478.98 1201.92 1480.77 Q1203.84 1482.55 1207.18 1482.55 Q1210.51 1482.55 1212.43 1480.77 Q1214.35 1478.96 1214.35 1475.86 Q1214.35 1472.73 1212.43 1470.95 Q1210.53 1469.17 1207.18 1469.17 M1202.5 1467.18 Q1199.49 1466.44 1197.8 1464.38 Q1196.14 1462.32 1196.14 1459.35 Q1196.14 1455.21 1199.08 1452.8 Q1202.04 1450.39 1207.18 1450.39 Q1212.34 1450.39 1215.28 1452.8 Q1218.22 1455.21 1218.22 1459.35 Q1218.22 1462.32 1216.53 1464.38 Q1214.86 1466.44 1211.88 1467.18 Q1215.26 1467.96 1217.13 1470.26 Q1219.03 1472.55 1219.03 1475.86 Q1219.03 1480.88 1215.95 1483.57 Q1212.9 1486.25 1207.18 1486.25 Q1201.46 1486.25 1198.38 1483.57 Q1195.33 1480.88 1195.33 1475.86 Q1195.33 1472.55 1197.22 1470.26 Q1199.12 1467.96 1202.5 1467.18 M1200.79 1459.79 Q1200.79 1462.48 1202.46 1463.98 Q1204.15 1465.49 1207.18 1465.49 Q1210.19 1465.49 1211.88 1463.98 Q1213.59 1462.48 1213.59 1459.79 Q1213.59 1457.11 1211.88 1455.6 Q1210.19 1454.1 1207.18 1454.1 Q1204.15 1454.1 1202.46 1455.6 Q1200.79 1457.11 1200.79 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1237.34 1454.1 Q1233.73 1454.1 1231.9 1457.66 Q1230.09 1461.2 1230.09 1468.33 Q1230.09 1475.44 1231.9 1479.01 Q1233.73 1482.55 1237.34 1482.55 Q1240.97 1482.55 1242.78 1479.01 Q1244.61 1475.44 1244.61 1468.33 Q1244.61 1461.2 1242.78 1457.66 Q1240.97 1454.1 1237.34 1454.1 M1237.34 1450.39 Q1243.15 1450.39 1246.21 1455 Q1249.28 1459.58 1249.28 1468.33 Q1249.28 1477.06 1246.21 1481.67 Q1243.15 1486.25 1237.34 1486.25 Q1231.53 1486.25 1228.45 1481.67 Q1225.4 1477.06 1225.4 1468.33 Q1225.4 1459.58 1228.45 1455 Q1231.53 1450.39 1237.34 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1760.41 1484.86 L1760.41 1480.6 Q1762.17 1481.44 1763.97 1481.88 Q1765.78 1482.32 1767.51 1482.32 Q1772.14 1482.32 1774.57 1479.21 Q1777.03 1476.09 1777.37 1469.75 Q1776.03 1471.74 1773.97 1472.8 Q1771.91 1473.87 1769.41 1473.87 Q1764.23 1473.87 1761.19 1470.74 Q1758.18 1467.59 1758.18 1462.15 Q1758.18 1456.83 1761.33 1453.61 Q1764.48 1450.39 1769.71 1450.39 Q1775.71 1450.39 1778.86 1455 Q1782.03 1459.58 1782.03 1468.33 Q1782.03 1476.51 1778.14 1481.39 Q1774.27 1486.25 1767.72 1486.25 Q1765.96 1486.25 1764.16 1485.9 Q1762.35 1485.56 1760.41 1484.86 M1769.71 1470.21 Q1772.86 1470.21 1774.69 1468.06 Q1776.54 1465.9 1776.54 1462.15 Q1776.54 1458.43 1774.69 1456.27 Q1772.86 1454.1 1769.71 1454.1 Q1766.56 1454.1 1764.71 1456.27 Q1762.88 1458.43 1762.88 1462.15 Q1762.88 1465.9 1764.71 1468.06 Q1766.56 1470.21 1769.71 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1800.43 1454.1 Q1796.82 1454.1 1794.99 1457.66 Q1793.18 1461.2 1793.18 1468.33 Q1793.18 1475.44 1794.99 1479.01 Q1796.82 1482.55 1800.43 1482.55 Q1804.06 1482.55 1805.87 1479.01 Q1807.7 1475.44 1807.7 1468.33 Q1807.7 1461.2 1805.87 1457.66 Q1804.06 1454.1 1800.43 1454.1 M1800.43 1450.39 Q1806.24 1450.39 1809.29 1455 Q1812.37 1459.58 1812.37 1468.33 Q1812.37 1477.06 1809.29 1481.67 Q1806.24 1486.25 1800.43 1486.25 Q1794.62 1486.25 1791.54 1481.67 Q1788.48 1477.06 1788.48 1468.33 Q1788.48 1459.58 1791.54 1455 Q1794.62 1450.39 1800.43 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2307.86 1481.64 L2315.5 1481.64 L2315.5 1455.28 L2307.19 1456.95 L2307.19 1452.69 L2315.45 1451.02 L2320.13 1451.02 L2320.13 1481.64 L2327.77 1481.64 L2327.77 1485.58 L2307.86 1485.58 L2307.86 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2347.21 1454.1 Q2343.6 1454.1 2341.77 1457.66 Q2339.97 1461.2 2339.97 1468.33 Q2339.97 1475.44 2341.77 1479.01 Q2343.6 1482.55 2347.21 1482.55 Q2350.84 1482.55 2352.65 1479.01 Q2354.48 1475.44 2354.48 1468.33 Q2354.48 1461.2 2352.65 1457.66 Q2350.84 1454.1 2347.21 1454.1 M2347.21 1450.39 Q2353.02 1450.39 2356.08 1455 Q2359.15 1459.58 2359.15 1468.33 Q2359.15 1477.06 2356.08 1481.67 Q2353.02 1486.25 2347.21 1486.25 Q2341.4 1486.25 2338.32 1481.67 Q2335.27 1477.06 2335.27 1468.33 Q2335.27 1459.58 2338.32 1455 Q2341.4 1450.39 2347.21 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2377.37 1454.1 Q2373.76 1454.1 2371.93 1457.66 Q2370.13 1461.2 2370.13 1468.33 Q2370.13 1475.44 2371.93 1479.01 Q2373.76 1482.55 2377.37 1482.55 Q2381.01 1482.55 2382.81 1479.01 Q2384.64 1475.44 2384.64 1468.33 Q2384.64 1461.2 2382.81 1457.66 Q2381.01 1454.1 2377.37 1454.1 M2377.37 1450.39 Q2383.18 1450.39 2386.24 1455 Q2389.32 1459.58 2389.32 1468.33 Q2389.32 1477.06 2386.24 1481.67 Q2383.18 1486.25 2377.37 1486.25 Q2371.56 1486.25 2368.48 1481.67 Q2365.43 1477.06 2365.43 1468.33 Q2365.43 1459.58 2368.48 1455 Q2371.56 1450.39 2377.37 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1204 1525.81 L1204 1543.66 L1212.08 1543.66 Q1216.57 1543.66 1219.02 1541.34 Q1221.47 1539.02 1221.47 1534.72 Q1221.47 1530.45 1219.02 1528.13 Q1216.57 1525.81 1212.08 1525.81 L1204 1525.81 M1197.57 1520.52 L1212.08 1520.52 Q1220.07 1520.52 1224.14 1524.15 Q1228.25 1527.75 1228.25 1534.72 Q1228.25 1541.75 1224.14 1545.35 Q1220.07 1548.95 1212.08 1548.95 L1204 1548.95 L1204 1568.04 L1197.57 1568.04 L1197.57 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1256.13 1537.87 Q1255.15 1537.3 1253.97 1537.04 Q1252.82 1536.76 1251.42 1536.76 Q1246.46 1536.76 1243.78 1540 Q1241.14 1543.22 1241.14 1549.27 L1241.14 1568.04 L1235.25 1568.04 L1235.25 1532.4 L1241.14 1532.4 L1241.14 1537.93 Q1242.99 1534.69 1245.95 1533.13 Q1248.91 1531.54 1253.14 1531.54 Q1253.74 1531.54 1254.48 1531.63 Q1255.21 1531.7 1256.1 1531.85 L1256.13 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1274.66 1536.5 Q1269.95 1536.5 1267.21 1540.19 Q1264.47 1543.85 1264.47 1550.25 Q1264.47 1556.65 1267.18 1560.34 Q1269.91 1564 1274.66 1564 Q1279.33 1564 1282.07 1560.31 Q1284.81 1556.62 1284.81 1550.25 Q1284.81 1543.92 1282.07 1540.23 Q1279.33 1536.5 1274.66 1536.5 M1274.66 1531.54 Q1282.29 1531.54 1286.66 1536.5 Q1291.02 1541.47 1291.02 1550.25 Q1291.02 1559 1286.66 1564 Q1282.29 1568.97 1274.66 1568.97 Q1266.99 1568.97 1262.62 1564 Q1258.3 1559 1258.3 1550.25 Q1258.3 1541.47 1262.62 1536.5 Q1266.99 1531.54 1274.66 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1318.77 1518.52 L1318.77 1523.39 L1313.17 1523.39 Q1310.02 1523.39 1308.78 1524.66 Q1307.57 1525.93 1307.57 1529.24 L1307.57 1532.4 L1317.21 1532.4 L1317.21 1536.95 L1307.57 1536.95 L1307.57 1568.04 L1301.68 1568.04 L1301.68 1536.95 L1296.08 1536.95 L1296.08 1532.4 L1301.68 1532.4 L1301.68 1529.91 Q1301.68 1523.96 1304.45 1521.26 Q1307.22 1518.52 1313.23 1518.52 L1318.77 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1323.67 1532.4 L1329.53 1532.4 L1329.53 1568.04 L1323.67 1568.04 L1323.67 1532.4 M1323.67 1518.52 L1329.53 1518.52 L1329.53 1525.93 L1323.67 1525.93 L1323.67 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1347.58 1522.27 L1347.58 1532.4 L1359.64 1532.4 L1359.64 1536.95 L1347.58 1536.95 L1347.58 1556.3 Q1347.58 1560.66 1348.75 1561.9 Q1349.96 1563.14 1353.62 1563.14 L1359.64 1563.14 L1359.64 1568.04 L1353.62 1568.04 Q1346.84 1568.04 1344.27 1565.53 Q1341.69 1562.98 1341.69 1556.3 L1341.69 1536.95 L1337.39 1536.95 L1337.39 1532.4 L1341.69 1532.4 L1341.69 1522.27 L1347.58 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,1384.24 2352.76,1384.24 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,1033.41 2352.76,1033.41 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,682.588 2352.76,682.588 "/>
-<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,331.763 2352.76,331.763 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1423.18 204.45,47.2441 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1384.24 223.347,1384.24 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1033.41 223.347,1033.41 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,682.588 223.347,682.588 "/>
-<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,331.763 223.347,331.763 "/>
-<path clip-path="url(#clip970)" d="M126.922 1382.37 Q123.774 1382.37 121.922 1384.53 Q120.093 1386.68 120.093 1390.43 Q120.093 1394.16 121.922 1396.33 Q123.774 1398.49 126.922 1398.49 Q130.07 1398.49 131.899 1396.33 Q133.751 1394.16 133.751 1390.43 Q133.751 1386.68 131.899 1384.53 Q130.07 1382.37 126.922 1382.37 M136.204 1367.72 L136.204 1371.98 Q134.445 1371.15 132.64 1370.71 Q130.857 1370.27 129.098 1370.27 Q124.468 1370.27 122.015 1373.39 Q119.584 1376.52 119.237 1382.84 Q120.603 1380.82 122.663 1379.76 Q124.723 1378.67 127.2 1378.67 Q132.408 1378.67 135.417 1381.84 Q138.45 1384.99 138.45 1390.43 Q138.45 1395.75 135.302 1398.97 Q132.154 1402.19 126.922 1402.19 Q120.927 1402.19 117.756 1397.61 Q114.584 1393 114.584 1384.27 Q114.584 1376.08 118.473 1371.22 Q122.362 1366.33 128.913 1366.33 Q130.672 1366.33 132.455 1366.68 Q134.26 1367.03 136.204 1367.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M156.505 1370.04 Q152.894 1370.04 151.066 1373.6 Q149.26 1377.14 149.26 1384.27 Q149.26 1391.38 151.066 1394.94 Q152.894 1398.49 156.505 1398.49 Q160.14 1398.49 161.945 1394.94 Q163.774 1391.38 163.774 1384.27 Q163.774 1377.14 161.945 1373.6 Q160.14 1370.04 156.505 1370.04 M156.505 1366.33 Q162.315 1366.33 165.371 1370.94 Q168.45 1375.52 168.45 1384.27 Q168.45 1393 165.371 1397.61 Q162.315 1402.19 156.505 1402.19 Q150.695 1402.19 147.616 1397.61 Q144.561 1393 144.561 1384.27 Q144.561 1375.52 147.616 1370.94 Q150.695 1366.33 156.505 1366.33 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M115.163 1016.13 L137.385 1016.13 L137.385 1018.12 L124.839 1050.69 L119.955 1050.69 L131.76 1020.07 L115.163 1020.07 L115.163 1016.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M156.505 1019.21 Q152.894 1019.21 151.066 1022.78 Q149.26 1026.32 149.26 1033.45 Q149.26 1040.55 151.066 1044.12 Q152.894 1047.66 156.505 1047.66 Q160.14 1047.66 161.945 1044.12 Q163.774 1040.55 163.774 1033.45 Q163.774 1026.32 161.945 1022.78 Q160.14 1019.21 156.505 1019.21 M156.505 1015.51 Q162.315 1015.51 165.371 1020.11 Q168.45 1024.7 168.45 1033.45 Q168.45 1042.17 165.371 1046.78 Q162.315 1051.36 156.505 1051.36 Q150.695 1051.36 147.616 1046.78 Q144.561 1042.17 144.561 1033.45 Q144.561 1024.7 147.616 1020.11 Q150.695 1015.51 156.505 1015.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M126.343 683.456 Q123.01 683.456 121.089 685.239 Q119.191 687.021 119.191 690.146 Q119.191 693.271 121.089 695.053 Q123.01 696.836 126.343 696.836 Q129.677 696.836 131.598 695.053 Q133.519 693.248 133.519 690.146 Q133.519 687.021 131.598 685.239 Q129.7 683.456 126.343 683.456 M121.668 681.466 Q118.658 680.725 116.968 678.665 Q115.302 676.604 115.302 673.642 Q115.302 669.498 118.242 667.091 Q121.205 664.683 126.343 664.683 Q131.505 664.683 134.445 667.091 Q137.385 669.498 137.385 673.642 Q137.385 676.604 135.695 678.665 Q134.029 680.725 131.042 681.466 Q134.422 682.253 136.297 684.544 Q138.195 686.836 138.195 690.146 Q138.195 695.169 135.117 697.854 Q132.061 700.54 126.343 700.54 Q120.626 700.54 117.547 697.854 Q114.492 695.169 114.492 690.146 Q114.492 686.836 116.39 684.544 Q118.288 682.253 121.668 681.466 M119.955 674.081 Q119.955 676.766 121.621 678.271 Q123.311 679.776 126.343 679.776 Q129.353 679.776 131.042 678.271 Q132.755 676.766 132.755 674.081 Q132.755 671.396 131.042 669.892 Q129.353 668.387 126.343 668.387 Q123.311 668.387 121.621 669.892 Q119.955 671.396 119.955 674.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M156.505 668.387 Q152.894 668.387 151.066 671.952 Q149.26 675.493 149.26 682.623 Q149.26 689.729 151.066 693.294 Q152.894 696.836 156.505 696.836 Q160.14 696.836 161.945 693.294 Q163.774 689.729 163.774 682.623 Q163.774 675.493 161.945 671.952 Q160.14 668.387 156.505 668.387 M156.505 664.683 Q162.315 664.683 165.371 669.29 Q168.45 673.873 168.45 682.623 Q168.45 691.35 165.371 695.956 Q162.315 700.54 156.505 700.54 Q150.695 700.54 147.616 695.956 Q144.561 691.35 144.561 682.623 Q144.561 673.873 147.616 669.29 Q150.695 664.683 156.505 664.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M116.482 348.326 L116.482 344.066 Q118.242 344.9 120.047 345.34 Q121.853 345.779 123.589 345.779 Q128.218 345.779 130.649 342.677 Q133.103 339.552 133.45 333.21 Q132.107 335.201 130.047 336.265 Q127.987 337.33 125.487 337.33 Q120.302 337.33 117.269 334.205 Q114.26 331.057 114.26 325.617 Q114.26 320.293 117.408 317.076 Q120.556 313.858 125.788 313.858 Q131.783 313.858 134.931 318.465 Q138.103 323.048 138.103 331.798 Q138.103 339.969 134.214 344.853 Q130.348 349.714 123.797 349.714 Q122.038 349.714 120.232 349.367 Q118.427 349.02 116.482 348.326 M125.788 333.673 Q128.936 333.673 130.765 331.52 Q132.617 329.367 132.617 325.617 Q132.617 321.891 130.765 319.738 Q128.936 317.562 125.788 317.562 Q122.64 317.562 120.788 319.738 Q118.959 321.891 118.959 325.617 Q118.959 329.367 120.788 331.52 Q122.64 333.673 125.788 333.673 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M156.505 317.562 Q152.894 317.562 151.066 321.127 Q149.26 324.668 149.26 331.798 Q149.26 338.904 151.066 342.469 Q152.894 346.011 156.505 346.011 Q160.14 346.011 161.945 342.469 Q163.774 338.904 163.774 331.798 Q163.774 324.668 161.945 321.127 Q160.14 317.562 156.505 317.562 M156.505 313.858 Q162.315 313.858 165.371 318.465 Q168.45 323.048 168.45 331.798 Q168.45 340.525 165.371 345.131 Q162.315 349.714 156.505 349.714 Q150.695 349.714 147.616 345.131 Q144.561 340.525 144.561 331.798 Q144.561 323.048 147.616 318.465 Q150.695 313.858 156.505 313.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M21.7677 827.769 L58.7206 827.769 L58.7206 820.003 Q58.7206 810.168 54.2646 805.617 Q49.8086 801.033 40.1964 801.033 Q30.6479 801.033 26.2237 805.617 Q21.7677 810.168 21.7677 820.003 L21.7677 827.769 M16.4842 834.199 L16.4842 820.99 Q16.4842 807.176 22.2451 800.715 Q27.9743 794.254 40.1964 794.254 Q52.4822 794.254 58.2432 800.747 Q64.0042 807.24 64.0042 820.99 L64.0042 834.199 L16.4842 834.199 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M44.7161 753.768 L47.5806 753.768 L47.5806 780.695 Q53.6281 780.313 56.8109 777.066 Q59.9619 773.788 59.9619 767.964 Q59.9619 764.59 59.1344 761.439 Q58.3069 758.256 56.6518 755.137 L62.1899 755.137 Q63.5267 758.288 64.227 761.598 Q64.9272 764.908 64.9272 768.314 Q64.9272 776.844 59.9619 781.841 Q54.9967 786.806 46.5303 786.806 Q37.7774 786.806 32.6531 782.095 Q27.4968 777.353 27.4968 769.332 Q27.4968 762.139 32.1438 757.969 Q36.7589 753.768 44.7161 753.768 M42.9973 759.624 Q38.1912 759.688 35.3266 762.33 Q32.4621 764.94 32.4621 769.268 Q32.4621 774.17 35.2312 777.13 Q38.0002 780.058 43.0292 780.504 L42.9973 759.624 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M29.4065 721.43 L34.9447 721.43 Q33.6716 723.913 33.035 726.586 Q32.3984 729.26 32.3984 732.125 Q32.3984 736.485 33.7352 738.681 Q35.072 740.846 37.7456 740.846 Q39.7826 740.846 40.9603 739.286 Q42.1061 737.726 43.1565 733.016 L43.6021 731.011 Q44.9389 724.772 47.3897 722.162 Q49.8086 719.521 54.1691 719.521 Q59.1344 719.521 62.0308 723.467 Q64.9272 727.382 64.9272 734.257 Q64.9272 737.122 64.3543 740.241 Q63.8132 743.328 62.6992 746.766 L56.6518 746.766 Q58.3387 743.519 59.198 740.368 Q60.0256 737.217 60.0256 734.13 Q60.0256 729.992 58.6251 727.764 Q57.1929 725.536 54.6147 725.536 Q52.2276 725.536 50.9545 727.159 Q49.6813 728.751 48.5037 734.193 L48.0262 736.23 Q46.8804 741.673 44.5251 744.092 Q42.138 746.511 38.0002 746.511 Q32.9713 746.511 30.2341 742.946 Q27.4968 739.382 27.4968 732.825 Q27.4968 729.578 27.9743 726.714 Q28.4517 723.849 29.4065 721.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M28.3562 710.195 L28.3562 704.338 L64.0042 704.338 L64.0042 710.195 L28.3562 710.195 M14.479 710.195 L14.479 704.338 L21.895 704.338 L21.895 710.195 L14.479 710.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M33.8307 671.428 Q33.2578 672.414 33.0032 673.592 Q32.7167 674.738 32.7167 676.138 Q32.7167 681.103 35.9632 683.777 Q39.1779 686.419 45.2253 686.419 L64.0042 686.419 L64.0042 692.307 L28.3562 692.307 L28.3562 686.419 L33.8944 686.419 Q30.6479 684.573 29.0883 681.613 Q27.4968 678.653 27.4968 674.419 Q27.4968 673.815 27.5923 673.083 Q27.656 672.351 27.8151 671.459 L33.8307 671.428 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M44.7161 636.225 L47.5806 636.225 L47.5806 663.152 Q53.6281 662.77 56.8109 659.524 Q59.9619 656.245 59.9619 650.421 Q59.9619 647.047 59.1344 643.896 Q58.3069 640.713 56.6518 637.594 L62.1899 637.594 Q63.5267 640.745 64.227 644.055 Q64.9272 647.365 64.9272 650.771 Q64.9272 659.301 59.9619 664.298 Q54.9967 669.263 46.5303 669.263 Q37.7774 669.263 32.6531 664.553 Q27.4968 659.81 27.4968 651.789 Q27.4968 644.596 32.1438 640.427 Q36.7589 636.225 44.7161 636.225 M42.9973 642.082 Q38.1912 642.145 35.3266 644.787 Q32.4621 647.397 32.4621 651.726 Q32.4621 656.627 35.2312 659.587 Q38.0002 662.516 43.0292 662.961 L42.9973 642.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip972)" cx="1053.41" cy="1208.83" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="2010.47" cy="331.763" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="715.629" cy="331.763" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="265.251" cy="787.836" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="2123.06" cy="156.351" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1334.9" cy="542.258" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1503.79" cy="1033.41" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="940.819" cy="191.433" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="771.927" cy="1173.74" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1841.58" cy="261.598" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="2291.95" cy="893.083" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="265.251" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1447.49" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1616.39" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1166.01" cy="1208.83" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="2010.47" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
-<circle clip-path="url(#clip972)" cx="1785.28" cy="226.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="659.332,1423.18 659.332,47.2441 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1222.31,1423.18 1222.31,47.2441 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1785.28,1423.18 1785.28,47.2441 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2348.25,1423.18 2348.25,47.2441 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="659.332,1423.18 659.332,1404.28 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1222.31,1423.18 1222.31,1404.28 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1785.28,1423.18 1785.28,1404.28 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2348.25,1423.18 2348.25,1404.28 "/>
+<path clip-path="url(#clip790)" d="M632.689 1451.02 L654.911 1451.02 L654.911 1453.01 L642.365 1485.58 L637.48 1485.58 L649.286 1454.96 L632.689 1454.96 L632.689 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M674.031 1454.1 Q670.42 1454.1 668.591 1457.66 Q666.786 1461.2 666.786 1468.33 Q666.786 1475.44 668.591 1479.01 Q670.42 1482.55 674.031 1482.55 Q677.665 1482.55 679.471 1479.01 Q681.3 1475.44 681.3 1468.33 Q681.3 1461.2 679.471 1457.66 Q677.665 1454.1 674.031 1454.1 M674.031 1450.39 Q679.841 1450.39 682.897 1455 Q685.976 1459.58 685.976 1468.33 Q685.976 1477.06 682.897 1481.67 Q679.841 1486.25 674.031 1486.25 Q668.221 1486.25 665.142 1481.67 Q662.087 1477.06 662.087 1468.33 Q662.087 1459.58 665.142 1455 Q668.221 1450.39 674.031 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1207.18 1469.17 Q1203.84 1469.17 1201.92 1470.95 Q1200.03 1472.73 1200.03 1475.86 Q1200.03 1478.98 1201.92 1480.77 Q1203.84 1482.55 1207.18 1482.55 Q1210.51 1482.55 1212.43 1480.77 Q1214.35 1478.96 1214.35 1475.86 Q1214.35 1472.73 1212.43 1470.95 Q1210.53 1469.17 1207.18 1469.17 M1202.5 1467.18 Q1199.49 1466.44 1197.8 1464.38 Q1196.14 1462.32 1196.14 1459.35 Q1196.14 1455.21 1199.08 1452.8 Q1202.04 1450.39 1207.18 1450.39 Q1212.34 1450.39 1215.28 1452.8 Q1218.22 1455.21 1218.22 1459.35 Q1218.22 1462.32 1216.53 1464.38 Q1214.86 1466.44 1211.88 1467.18 Q1215.26 1467.96 1217.13 1470.26 Q1219.03 1472.55 1219.03 1475.86 Q1219.03 1480.88 1215.95 1483.57 Q1212.9 1486.25 1207.18 1486.25 Q1201.46 1486.25 1198.38 1483.57 Q1195.33 1480.88 1195.33 1475.86 Q1195.33 1472.55 1197.22 1470.26 Q1199.12 1467.96 1202.5 1467.18 M1200.79 1459.79 Q1200.79 1462.48 1202.46 1463.98 Q1204.15 1465.49 1207.18 1465.49 Q1210.19 1465.49 1211.88 1463.98 Q1213.59 1462.48 1213.59 1459.79 Q1213.59 1457.11 1211.88 1455.6 Q1210.19 1454.1 1207.18 1454.1 Q1204.15 1454.1 1202.46 1455.6 Q1200.79 1457.11 1200.79 1459.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1237.34 1454.1 Q1233.73 1454.1 1231.9 1457.66 Q1230.09 1461.2 1230.09 1468.33 Q1230.09 1475.44 1231.9 1479.01 Q1233.73 1482.55 1237.34 1482.55 Q1240.97 1482.55 1242.78 1479.01 Q1244.61 1475.44 1244.61 1468.33 Q1244.61 1461.2 1242.78 1457.66 Q1240.97 1454.1 1237.34 1454.1 M1237.34 1450.39 Q1243.15 1450.39 1246.21 1455 Q1249.28 1459.58 1249.28 1468.33 Q1249.28 1477.06 1246.21 1481.67 Q1243.15 1486.25 1237.34 1486.25 Q1231.53 1486.25 1228.45 1481.67 Q1225.4 1477.06 1225.4 1468.33 Q1225.4 1459.58 1228.45 1455 Q1231.53 1450.39 1237.34 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1760.41 1484.86 L1760.41 1480.6 Q1762.17 1481.44 1763.97 1481.88 Q1765.78 1482.32 1767.51 1482.32 Q1772.14 1482.32 1774.57 1479.21 Q1777.03 1476.09 1777.37 1469.75 Q1776.03 1471.74 1773.97 1472.8 Q1771.91 1473.87 1769.41 1473.87 Q1764.23 1473.87 1761.19 1470.74 Q1758.18 1467.59 1758.18 1462.15 Q1758.18 1456.83 1761.33 1453.61 Q1764.48 1450.39 1769.71 1450.39 Q1775.71 1450.39 1778.86 1455 Q1782.03 1459.58 1782.03 1468.33 Q1782.03 1476.51 1778.14 1481.39 Q1774.27 1486.25 1767.72 1486.25 Q1765.96 1486.25 1764.16 1485.9 Q1762.35 1485.56 1760.41 1484.86 M1769.71 1470.21 Q1772.86 1470.21 1774.69 1468.06 Q1776.54 1465.9 1776.54 1462.15 Q1776.54 1458.43 1774.69 1456.27 Q1772.86 1454.1 1769.71 1454.1 Q1766.56 1454.1 1764.71 1456.27 Q1762.88 1458.43 1762.88 1462.15 Q1762.88 1465.9 1764.71 1468.06 Q1766.56 1470.21 1769.71 1470.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1800.43 1454.1 Q1796.82 1454.1 1794.99 1457.66 Q1793.18 1461.2 1793.18 1468.33 Q1793.18 1475.44 1794.99 1479.01 Q1796.82 1482.55 1800.43 1482.55 Q1804.06 1482.55 1805.87 1479.01 Q1807.7 1475.44 1807.7 1468.33 Q1807.7 1461.2 1805.87 1457.66 Q1804.06 1454.1 1800.43 1454.1 M1800.43 1450.39 Q1806.24 1450.39 1809.29 1455 Q1812.37 1459.58 1812.37 1468.33 Q1812.37 1477.06 1809.29 1481.67 Q1806.24 1486.25 1800.43 1486.25 Q1794.62 1486.25 1791.54 1481.67 Q1788.48 1477.06 1788.48 1468.33 Q1788.48 1459.58 1791.54 1455 Q1794.62 1450.39 1800.43 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M2307.86 1481.64 L2315.5 1481.64 L2315.5 1455.28 L2307.19 1456.95 L2307.19 1452.69 L2315.45 1451.02 L2320.13 1451.02 L2320.13 1481.64 L2327.77 1481.64 L2327.77 1485.58 L2307.86 1485.58 L2307.86 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M2347.21 1454.1 Q2343.6 1454.1 2341.77 1457.66 Q2339.97 1461.2 2339.97 1468.33 Q2339.97 1475.44 2341.77 1479.01 Q2343.6 1482.55 2347.21 1482.55 Q2350.84 1482.55 2352.65 1479.01 Q2354.48 1475.44 2354.48 1468.33 Q2354.48 1461.2 2352.65 1457.66 Q2350.84 1454.1 2347.21 1454.1 M2347.21 1450.39 Q2353.02 1450.39 2356.08 1455 Q2359.15 1459.58 2359.15 1468.33 Q2359.15 1477.06 2356.08 1481.67 Q2353.02 1486.25 2347.21 1486.25 Q2341.4 1486.25 2338.32 1481.67 Q2335.27 1477.06 2335.27 1468.33 Q2335.27 1459.58 2338.32 1455 Q2341.4 1450.39 2347.21 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M2377.37 1454.1 Q2373.76 1454.1 2371.93 1457.66 Q2370.13 1461.2 2370.13 1468.33 Q2370.13 1475.44 2371.93 1479.01 Q2373.76 1482.55 2377.37 1482.55 Q2381.01 1482.55 2382.81 1479.01 Q2384.64 1475.44 2384.64 1468.33 Q2384.64 1461.2 2382.81 1457.66 Q2381.01 1454.1 2377.37 1454.1 M2377.37 1450.39 Q2383.18 1450.39 2386.24 1455 Q2389.32 1459.58 2389.32 1468.33 Q2389.32 1477.06 2386.24 1481.67 Q2383.18 1486.25 2377.37 1486.25 Q2371.56 1486.25 2368.48 1481.67 Q2365.43 1477.06 2365.43 1468.33 Q2365.43 1459.58 2368.48 1455 Q2371.56 1450.39 2377.37 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1204 1525.81 L1204 1543.66 L1212.08 1543.66 Q1216.57 1543.66 1219.02 1541.34 Q1221.47 1539.02 1221.47 1534.72 Q1221.47 1530.45 1219.02 1528.13 Q1216.57 1525.81 1212.08 1525.81 L1204 1525.81 M1197.57 1520.52 L1212.08 1520.52 Q1220.07 1520.52 1224.14 1524.15 Q1228.25 1527.75 1228.25 1534.72 Q1228.25 1541.75 1224.14 1545.35 Q1220.07 1548.95 1212.08 1548.95 L1204 1548.95 L1204 1568.04 L1197.57 1568.04 L1197.57 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1256.13 1537.87 Q1255.15 1537.3 1253.97 1537.04 Q1252.82 1536.76 1251.42 1536.76 Q1246.46 1536.76 1243.78 1540 Q1241.14 1543.22 1241.14 1549.27 L1241.14 1568.04 L1235.25 1568.04 L1235.25 1532.4 L1241.14 1532.4 L1241.14 1537.93 Q1242.99 1534.69 1245.95 1533.13 Q1248.91 1531.54 1253.14 1531.54 Q1253.74 1531.54 1254.48 1531.63 Q1255.21 1531.7 1256.1 1531.85 L1256.13 1537.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1274.66 1536.5 Q1269.95 1536.5 1267.21 1540.19 Q1264.47 1543.85 1264.47 1550.25 Q1264.47 1556.65 1267.18 1560.34 Q1269.91 1564 1274.66 1564 Q1279.33 1564 1282.07 1560.31 Q1284.81 1556.62 1284.81 1550.25 Q1284.81 1543.92 1282.07 1540.23 Q1279.33 1536.5 1274.66 1536.5 M1274.66 1531.54 Q1282.29 1531.54 1286.66 1536.5 Q1291.02 1541.47 1291.02 1550.25 Q1291.02 1559 1286.66 1564 Q1282.29 1568.97 1274.66 1568.97 Q1266.99 1568.97 1262.62 1564 Q1258.3 1559 1258.3 1550.25 Q1258.3 1541.47 1262.62 1536.5 Q1266.99 1531.54 1274.66 1531.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1318.77 1518.52 L1318.77 1523.39 L1313.17 1523.39 Q1310.02 1523.39 1308.78 1524.66 Q1307.57 1525.93 1307.57 1529.24 L1307.57 1532.4 L1317.21 1532.4 L1317.21 1536.95 L1307.57 1536.95 L1307.57 1568.04 L1301.68 1568.04 L1301.68 1536.95 L1296.08 1536.95 L1296.08 1532.4 L1301.68 1532.4 L1301.68 1529.91 Q1301.68 1523.96 1304.45 1521.26 Q1307.22 1518.52 1313.23 1518.52 L1318.77 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1323.67 1532.4 L1329.53 1532.4 L1329.53 1568.04 L1323.67 1568.04 L1323.67 1532.4 M1323.67 1518.52 L1329.53 1518.52 L1329.53 1525.93 L1323.67 1525.93 L1323.67 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M1347.58 1522.27 L1347.58 1532.4 L1359.64 1532.4 L1359.64 1536.95 L1347.58 1536.95 L1347.58 1556.3 Q1347.58 1560.66 1348.75 1561.9 Q1349.96 1563.14 1353.62 1563.14 L1359.64 1563.14 L1359.64 1568.04 L1353.62 1568.04 Q1346.84 1568.04 1344.27 1565.53 Q1341.69 1562.98 1341.69 1556.3 L1341.69 1536.95 L1337.39 1536.95 L1337.39 1532.4 L1341.69 1532.4 L1341.69 1522.27 L1347.58 1522.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,1384.24 2352.76,1384.24 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,1033.41 2352.76,1033.41 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,682.588 2352.76,682.588 "/>
+<polyline clip-path="url(#clip792)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="204.45,331.763 2352.76,331.763 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1423.18 204.45,47.2441 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1384.24 223.347,1384.24 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,1033.41 223.347,1033.41 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,682.588 223.347,682.588 "/>
+<polyline clip-path="url(#clip790)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="204.45,331.763 223.347,331.763 "/>
+<path clip-path="url(#clip790)" d="M126.922 1382.37 Q123.774 1382.37 121.922 1384.53 Q120.093 1386.68 120.093 1390.43 Q120.093 1394.16 121.922 1396.33 Q123.774 1398.49 126.922 1398.49 Q130.07 1398.49 131.899 1396.33 Q133.751 1394.16 133.751 1390.43 Q133.751 1386.68 131.899 1384.53 Q130.07 1382.37 126.922 1382.37 M136.204 1367.72 L136.204 1371.98 Q134.445 1371.15 132.64 1370.71 Q130.857 1370.27 129.098 1370.27 Q124.468 1370.27 122.015 1373.39 Q119.584 1376.52 119.237 1382.84 Q120.603 1380.82 122.663 1379.76 Q124.723 1378.67 127.2 1378.67 Q132.408 1378.67 135.417 1381.84 Q138.45 1384.99 138.45 1390.43 Q138.45 1395.75 135.302 1398.97 Q132.154 1402.19 126.922 1402.19 Q120.927 1402.19 117.756 1397.61 Q114.584 1393 114.584 1384.27 Q114.584 1376.08 118.473 1371.22 Q122.362 1366.33 128.913 1366.33 Q130.672 1366.33 132.455 1366.68 Q134.26 1367.03 136.204 1367.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M156.505 1370.04 Q152.894 1370.04 151.066 1373.6 Q149.26 1377.14 149.26 1384.27 Q149.26 1391.38 151.066 1394.94 Q152.894 1398.49 156.505 1398.49 Q160.14 1398.49 161.945 1394.94 Q163.774 1391.38 163.774 1384.27 Q163.774 1377.14 161.945 1373.6 Q160.14 1370.04 156.505 1370.04 M156.505 1366.33 Q162.315 1366.33 165.371 1370.94 Q168.45 1375.52 168.45 1384.27 Q168.45 1393 165.371 1397.61 Q162.315 1402.19 156.505 1402.19 Q150.695 1402.19 147.616 1397.61 Q144.561 1393 144.561 1384.27 Q144.561 1375.52 147.616 1370.94 Q150.695 1366.33 156.505 1366.33 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M115.163 1016.13 L137.385 1016.13 L137.385 1018.12 L124.839 1050.69 L119.955 1050.69 L131.76 1020.07 L115.163 1020.07 L115.163 1016.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M156.505 1019.21 Q152.894 1019.21 151.066 1022.78 Q149.26 1026.32 149.26 1033.45 Q149.26 1040.55 151.066 1044.12 Q152.894 1047.66 156.505 1047.66 Q160.14 1047.66 161.945 1044.12 Q163.774 1040.55 163.774 1033.45 Q163.774 1026.32 161.945 1022.78 Q160.14 1019.21 156.505 1019.21 M156.505 1015.51 Q162.315 1015.51 165.371 1020.11 Q168.45 1024.7 168.45 1033.45 Q168.45 1042.17 165.371 1046.78 Q162.315 1051.36 156.505 1051.36 Q150.695 1051.36 147.616 1046.78 Q144.561 1042.17 144.561 1033.45 Q144.561 1024.7 147.616 1020.11 Q150.695 1015.51 156.505 1015.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M126.343 683.456 Q123.01 683.456 121.089 685.239 Q119.191 687.021 119.191 690.146 Q119.191 693.271 121.089 695.053 Q123.01 696.836 126.343 696.836 Q129.677 696.836 131.598 695.053 Q133.519 693.248 133.519 690.146 Q133.519 687.021 131.598 685.239 Q129.7 683.456 126.343 683.456 M121.668 681.466 Q118.658 680.725 116.968 678.665 Q115.302 676.604 115.302 673.642 Q115.302 669.498 118.242 667.091 Q121.205 664.683 126.343 664.683 Q131.505 664.683 134.445 667.091 Q137.385 669.498 137.385 673.642 Q137.385 676.604 135.695 678.665 Q134.029 680.725 131.042 681.466 Q134.422 682.253 136.297 684.544 Q138.195 686.836 138.195 690.146 Q138.195 695.169 135.117 697.854 Q132.061 700.54 126.343 700.54 Q120.626 700.54 117.547 697.854 Q114.492 695.169 114.492 690.146 Q114.492 686.836 116.39 684.544 Q118.288 682.253 121.668 681.466 M119.955 674.081 Q119.955 676.766 121.621 678.271 Q123.311 679.776 126.343 679.776 Q129.353 679.776 131.042 678.271 Q132.755 676.766 132.755 674.081 Q132.755 671.396 131.042 669.892 Q129.353 668.387 126.343 668.387 Q123.311 668.387 121.621 669.892 Q119.955 671.396 119.955 674.081 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M156.505 668.387 Q152.894 668.387 151.066 671.952 Q149.26 675.493 149.26 682.623 Q149.26 689.729 151.066 693.294 Q152.894 696.836 156.505 696.836 Q160.14 696.836 161.945 693.294 Q163.774 689.729 163.774 682.623 Q163.774 675.493 161.945 671.952 Q160.14 668.387 156.505 668.387 M156.505 664.683 Q162.315 664.683 165.371 669.29 Q168.45 673.873 168.45 682.623 Q168.45 691.35 165.371 695.956 Q162.315 700.54 156.505 700.54 Q150.695 700.54 147.616 695.956 Q144.561 691.35 144.561 682.623 Q144.561 673.873 147.616 669.29 Q150.695 664.683 156.505 664.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M116.482 348.326 L116.482 344.066 Q118.242 344.9 120.047 345.34 Q121.853 345.779 123.589 345.779 Q128.218 345.779 130.649 342.677 Q133.103 339.552 133.45 333.21 Q132.107 335.201 130.047 336.265 Q127.987 337.33 125.487 337.33 Q120.302 337.33 117.269 334.205 Q114.26 331.057 114.26 325.617 Q114.26 320.293 117.408 317.076 Q120.556 313.858 125.788 313.858 Q131.783 313.858 134.931 318.465 Q138.103 323.048 138.103 331.798 Q138.103 339.969 134.214 344.853 Q130.348 349.714 123.797 349.714 Q122.038 349.714 120.232 349.367 Q118.427 349.02 116.482 348.326 M125.788 333.673 Q128.936 333.673 130.765 331.52 Q132.617 329.367 132.617 325.617 Q132.617 321.891 130.765 319.738 Q128.936 317.562 125.788 317.562 Q122.64 317.562 120.788 319.738 Q118.959 321.891 118.959 325.617 Q118.959 329.367 120.788 331.52 Q122.64 333.673 125.788 333.673 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M156.505 317.562 Q152.894 317.562 151.066 321.127 Q149.26 324.668 149.26 331.798 Q149.26 338.904 151.066 342.469 Q152.894 346.011 156.505 346.011 Q160.14 346.011 161.945 342.469 Q163.774 338.904 163.774 331.798 Q163.774 324.668 161.945 321.127 Q160.14 317.562 156.505 317.562 M156.505 313.858 Q162.315 313.858 165.371 318.465 Q168.45 323.048 168.45 331.798 Q168.45 340.525 165.371 345.131 Q162.315 349.714 156.505 349.714 Q150.695 349.714 147.616 345.131 Q144.561 340.525 144.561 331.798 Q144.561 323.048 147.616 318.465 Q150.695 313.858 156.505 313.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M21.7677 827.769 L58.7206 827.769 L58.7206 820.003 Q58.7206 810.168 54.2646 805.617 Q49.8086 801.033 40.1964 801.033 Q30.6479 801.033 26.2237 805.617 Q21.7677 810.168 21.7677 820.003 L21.7677 827.769 M16.4842 834.199 L16.4842 820.99 Q16.4842 807.176 22.2451 800.715 Q27.9743 794.254 40.1964 794.254 Q52.4822 794.254 58.2432 800.747 Q64.0042 807.24 64.0042 820.99 L64.0042 834.199 L16.4842 834.199 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M44.7161 753.768 L47.5806 753.768 L47.5806 780.695 Q53.6281 780.313 56.8109 777.066 Q59.9619 773.788 59.9619 767.964 Q59.9619 764.59 59.1344 761.439 Q58.3069 758.256 56.6518 755.137 L62.1899 755.137 Q63.5267 758.288 64.227 761.598 Q64.9272 764.908 64.9272 768.314 Q64.9272 776.844 59.9619 781.841 Q54.9967 786.806 46.5303 786.806 Q37.7774 786.806 32.6531 782.095 Q27.4968 777.353 27.4968 769.332 Q27.4968 762.139 32.1438 757.969 Q36.7589 753.768 44.7161 753.768 M42.9973 759.624 Q38.1912 759.688 35.3266 762.33 Q32.4621 764.94 32.4621 769.268 Q32.4621 774.17 35.2312 777.13 Q38.0002 780.058 43.0292 780.504 L42.9973 759.624 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M29.4065 721.43 L34.9447 721.43 Q33.6716 723.913 33.035 726.586 Q32.3984 729.26 32.3984 732.125 Q32.3984 736.485 33.7352 738.681 Q35.072 740.846 37.7456 740.846 Q39.7826 740.846 40.9603 739.286 Q42.1061 737.726 43.1565 733.016 L43.6021 731.011 Q44.9389 724.772 47.3897 722.162 Q49.8086 719.521 54.1691 719.521 Q59.1344 719.521 62.0308 723.467 Q64.9272 727.382 64.9272 734.257 Q64.9272 737.122 64.3543 740.241 Q63.8132 743.328 62.6992 746.766 L56.6518 746.766 Q58.3387 743.519 59.198 740.368 Q60.0256 737.217 60.0256 734.13 Q60.0256 729.992 58.6251 727.764 Q57.1929 725.536 54.6147 725.536 Q52.2276 725.536 50.9545 727.159 Q49.6813 728.751 48.5037 734.193 L48.0262 736.23 Q46.8804 741.673 44.5251 744.092 Q42.138 746.511 38.0002 746.511 Q32.9713 746.511 30.2341 742.946 Q27.4968 739.382 27.4968 732.825 Q27.4968 729.578 27.9743 726.714 Q28.4517 723.849 29.4065 721.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M28.3562 710.195 L28.3562 704.338 L64.0042 704.338 L64.0042 710.195 L28.3562 710.195 M14.479 710.195 L14.479 704.338 L21.895 704.338 L21.895 710.195 L14.479 710.195 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M33.8307 671.428 Q33.2578 672.414 33.0032 673.592 Q32.7167 674.738 32.7167 676.138 Q32.7167 681.103 35.9632 683.777 Q39.1779 686.419 45.2253 686.419 L64.0042 686.419 L64.0042 692.307 L28.3562 692.307 L28.3562 686.419 L33.8944 686.419 Q30.6479 684.573 29.0883 681.613 Q27.4968 678.653 27.4968 674.419 Q27.4968 673.815 27.5923 673.083 Q27.656 672.351 27.8151 671.459 L33.8307 671.428 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip790)" d="M44.7161 636.225 L47.5806 636.225 L47.5806 663.152 Q53.6281 662.77 56.8109 659.524 Q59.9619 656.245 59.9619 650.421 Q59.9619 647.047 59.1344 643.896 Q58.3069 640.713 56.6518 637.594 L62.1899 637.594 Q63.5267 640.745 64.227 644.055 Q64.9272 647.365 64.9272 650.771 Q64.9272 659.301 59.9619 664.298 Q54.9967 669.263 46.5303 669.263 Q37.7774 669.263 32.6531 664.553 Q27.4968 659.81 27.4968 651.789 Q27.4968 644.596 32.1438 640.427 Q36.7589 636.225 44.7161 636.225 M42.9973 642.082 Q38.1912 642.145 35.3266 644.787 Q32.4621 647.397 32.4621 651.726 Q32.4621 656.627 35.2312 659.587 Q38.0002 662.516 43.0292 662.961 L42.9973 642.082 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><circle clip-path="url(#clip792)" cx="1053.41" cy="1208.83" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="2010.47" cy="331.763" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="715.629" cy="331.763" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="265.251" cy="787.836" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="2123.06" cy="156.351" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1334.9" cy="542.258" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1503.79" cy="1033.41" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="940.819" cy="191.433" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="771.927" cy="1173.74" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1841.58" cy="261.598" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="2291.95" cy="893.083" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="265.251" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1447.49" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1616.39" cy="1384.24" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1166.01" cy="1208.83" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="2010.47" cy="86.1857" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
+<circle clip-path="url(#clip792)" cx="1785.28" cy="226.516" r="14.4" fill="#009af9" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="3.2"/>
 </svg>
diff --git a/previews/PR3562/tutorials/linear/multi_objective_knapsack/index.html b/previews/PR3562/tutorials/linear/multi_objective_knapsack/index.html
index d9b6a13c793..47368f8ddd8 100644
--- a/previews/PR3562/tutorials/linear/multi_objective_knapsack/index.html
+++ b/previews/PR3562/tutorials/linear/multi_objective_knapsack/index.html
@@ -25,7 +25,7 @@
     xlabel = &quot;Profit&quot;,
     ylabel = &quot;Desire&quot;,
     legend = false,
-)</code></pre><img src="717b7729.svg" alt="Example block output"/><p>The goal of the bi-objective knapsack problem is to choose a subset which maximizes both objectives.</p><h2 id="JuMP-formulation"><a class="docs-heading-anchor" href="#JuMP-formulation">JuMP formulation</a><a id="JuMP-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-formulation" title="Permalink"></a></h2><p>Our JuMP formulation is a direct translation of the mathematical formulation:</p><pre><code class="language-julia hljs">model = Model()
+)</code></pre><img src="c5b0bcc4.svg" alt="Example block output"/><p>The goal of the bi-objective knapsack problem is to choose a subset which maximizes both objectives.</p><h2 id="JuMP-formulation"><a class="docs-heading-anchor" href="#JuMP-formulation">JuMP formulation</a><a id="JuMP-formulation-1"></a><a class="docs-heading-anchor-permalink" href="#JuMP-formulation" title="Permalink"></a></h2><p>Our JuMP formulation is a direct translation of the mathematical formulation:</p><pre><code class="language-julia hljs">model = Model()
 @variable(model, x[1:N], Bin)
 @constraint(model, sum(weight[i] * x[i] for i in 1:N) &lt;= capacity)
 @expression(model, profit_expr, sum(profit[i] * x[i] for i in 1:N))
@@ -50,7 +50,7 @@
   Relative gap       : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 1.09786e-01
+  Solve time (sec)   : 1.32877e-01
   Simplex iterations : 6
   Barrier iterations : -1
   Node count         : 1
@@ -80,7 +80,7 @@
     Plots.annotate!(y[1] - 1, y[2], (i, 10))
 end
 ideal_point = objective_bound(model)
-Plots.scatter!([ideal_point[1]], [ideal_point[2]]; label = &quot;Ideal point&quot;)</code></pre><img src="1f70cf5d.svg" alt="Example block output"/><p>Visualizing the objective space lets the decision maker choose a solution that suits their personal preferences. For example, result <code>#7</code> is close to the maximum value of profit, but offers significantly higher desirability compared with solutions <code>#8</code> and <code>#9</code>.</p><p>The set of items that are chosen in solution <code>#7</code> are:</p><pre><code class="language-julia hljs">items_chosen = [i for i in 1:N if value(x[i]; result = 7) &gt; 0.9]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">11-element Vector{Int64}:
+Plots.scatter!([ideal_point[1]], [ideal_point[2]]; label = &quot;Ideal point&quot;)</code></pre><img src="ba5f521f.svg" alt="Example block output"/><p>Visualizing the objective space lets the decision maker choose a solution that suits their personal preferences. For example, result <code>#7</code> is close to the maximum value of profit, but offers significantly higher desirability compared with solutions <code>#8</code> and <code>#9</code>.</p><p>The set of items that are chosen in solution <code>#7</code> are:</p><pre><code class="language-julia hljs">items_chosen = [i for i in 1:N if value(x[i]; result = 7) &gt; 0.9]</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">11-element Vector{Int64}:
   1
   2
   3
@@ -91,4 +91,4 @@
  11
  15
  16
- 17</code></pre><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p><a href="../../../packages/MultiObjectiveAlgorithms/#MultiObjectiveAlgorithms.jl">MultiObjectiveAlgorithms.jl</a> implements a number of different algorithms. Try solving the same problem using <code>MOA.Dichotomy()</code>. Does it find the same solution?</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../transp/">« The transportation problem</a><a class="docs-footer-nextpage" href="../multi_objective_examples/">Simple multi-objective examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 17</code></pre><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><p><a href="../../../packages/MultiObjectiveAlgorithms/#MultiObjectiveAlgorithms.jl">MultiObjectiveAlgorithms.jl</a> implements a number of different algorithms. Try solving the same problem using <code>MOA.Dichotomy()</code>. Does it find the same solution?</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../transp/">« The transportation problem</a><a class="docs-footer-nextpage" href="../multi_objective_examples/">Simple multi-objective examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/n-queens/index.html b/previews/PR3562/tutorials/linear/n-queens/index.html
index 5ef0f8083b4..78e9cb26c76 100644
--- a/previews/PR3562/tutorials/linear/n-queens/index.html
+++ b/previews/PR3562/tutorials/linear/n-queens/index.html
@@ -22,4 +22,4 @@
  0  0  0  0  0  0  1  0
  1  0  0  0  0  0  0  0
  0  0  0  0  0  1  0  0
- 0  1  0  0  0  0  0  0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../sudoku/">« Sudoku</a><a class="docs-footer-nextpage" href="../constraint_programming/">Constraint programming »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 0  1  0  0  0  0  0  0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../sudoku/">« Sudoku</a><a class="docs-footer-nextpage" href="../constraint_programming/">Constraint programming »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/network_flows/index.html b/previews/PR3562/tutorials/linear/network_flows/index.html
index e69f9be94da..b0d007240e7 100644
--- a/previews/PR3562/tutorials/linear/network_flows/index.html
+++ b/previews/PR3562/tutorials/linear/network_flows/index.html
@@ -88,4 +88,4 @@
   0.0   0.0   0.0  0.0   0.0   0.0   0.0   3.0
   0.0   0.0   0.0  0.0   0.0   0.0   0.0   2.0
   0.0   0.0   0.0  0.0   0.0   0.0   0.0   1.0
-  0.0   0.0   0.0  0.0   0.0   0.0   0.0   0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../geographic_clustering/">« Geographical clustering</a><a class="docs-footer-nextpage" href="../transp/">The transportation problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  0.0   0.0   0.0  0.0   0.0   0.0   0.0   0.0</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../geographic_clustering/">« Geographical clustering</a><a class="docs-footer-nextpage" href="../transp/">The transportation problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/sudoku/index.html b/previews/PR3562/tutorials/linear/sudoku/index.html
index 71b73282d29..d35fdfe90ad 100644
--- a/previews/PR3562/tutorials/linear/sudoku/index.html
+++ b/previews/PR3562/tutorials/linear/sudoku/index.html
@@ -91,4 +91,4 @@
  7  1  3  9  2  4  8  5  6
  9  6  1  5  3  7  2  8  4
  2  8  7  4  1  9  6  3  5
- 3  4  5  2  8  6  1  7  9</code></pre><p>Which is the same as we found before:</p><pre><code class="language-julia hljs">sol == csp_sol</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">true</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_objective_examples/">« Simple multi-objective examples</a><a class="docs-footer-nextpage" href="../n-queens/">N-Queens »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 3  4  5  2  8  6  1  7  9</code></pre><p>Which is the same as we found before:</p><pre><code class="language-julia hljs">sol == csp_sol</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">true</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_objective_examples/">« Simple multi-objective examples</a><a class="docs-footer-nextpage" href="../n-queens/">N-Queens »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/tips_and_tricks.jl b/previews/PR3562/tutorials/linear/tips_and_tricks.jl
index b339521c36e..ab2b5a7e9d2 100644
--- a/previews/PR3562/tutorials/linear/tips_and_tricks.jl
+++ b/previews/PR3562/tutorials/linear/tips_and_tricks.jl
@@ -237,6 +237,15 @@ M = 100
 
 # ## Semi-continuous variables
 
+# !!! info
+#     This section uses sets from [MathOptInterface](@ref moi_documentation).
+#     By default, JuMP exports the `MOI` symbol as an alias for the
+#     MathOptInterface.jl package. We recommend making this more explicit in
+#     your code by adding the following lines:
+#     ```julia
+#     import MathOptInterface as MOI
+#     ```
+
 # A semi-continuous variable is a continuous variable between bounds $[l,u]$
 # that also can assume the value zero, that is:
 # $$x \in \{0\} \cup [l,u].$$
@@ -246,17 +255,6 @@ M = 100
 model = Model();
 @variable(model, x in Semicontinuous(1.0, 2.0))
 
-# You can also represent a semi-continuous variable using the reformulation:
-
-model = Model();
-@variable(model, x)
-@variable(model, z, Bin)
-@constraint(model, x <= 2 * z)
-@constraint(model, x >= 1 * z)
-
-# When `z = 0` the two constraints are equivalent to `0 <= x <= 0`. When `z = 1`,
-# the two constraints are equivalennt to `1 <= x <= 2`.
-
 # ## Semi-integer variables
 
 # A semi-integer variable is a variable which assumes integer values between
@@ -266,17 +264,6 @@ model = Model();
 model = Model();
 @variable(model, x in Semiinteger(5.0, 10.0))
 
-# You can also represent a semi-integer variable using the reformulation:
-
-model = Model();
-@variable(model, x, Int)
-@variable(model, z, Bin)
-@constraint(model, x <= 10 * z)
-@constraint(model, x >= 5 * z)
-
-# When `z = 0` the two constraints are equivalent to `0 <= x <= 0`. When `z = 1`,
-# the two constraints are equivalennt to `5 <= x <= 10`.
-
 # ## Special Ordered Sets of Type 1
 
 # A Special Ordered Set of Type 1 is a set of variables, at most one of which
diff --git a/previews/PR3562/tutorials/linear/tips_and_tricks/index.html b/previews/PR3562/tutorials/linear/tips_and_tricks/index.html
index b8560022873..22d2ac2e619 100644
--- a/previews/PR3562/tutorials/linear/tips_and_tricks/index.html
+++ b/previews/PR3562/tutorials/linear/tips_and_tricks/index.html
@@ -39,7 +39,7 @@
  x[1]
  x[2]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, z, Bin)</code><code class="nohighlight hljs ansi" style="display:block;">z</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; M = 100</code><code class="nohighlight hljs ansi" style="display:block;">100</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, sum(x) &lt;= 1 + M * (1 - z))</code><code class="nohighlight hljs ansi" style="display:block;">x[1] + x[2] + 100 z ≤ 101</code></pre><p><strong>Example</strong> <span>$x_1 + x_2 \leq 1$</span> if <span>$z = 0$</span>.</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x[1:2])</code><code class="nohighlight hljs ansi" style="display:block;">2-element Vector{VariableRef}:
  x[1]
- x[2]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, z, Bin)</code><code class="nohighlight hljs ansi" style="display:block;">z</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; M = 100</code><code class="nohighlight hljs ansi" style="display:block;">100</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, sum(x) &lt;= 1 + M * z)</code><code class="nohighlight hljs ansi" style="display:block;">x[1] + x[2] - 100 z ≤ 1</code></pre><h2 id="Semi-continuous-variables"><a class="docs-heading-anchor" href="#Semi-continuous-variables">Semi-continuous variables</a><a id="Semi-continuous-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Semi-continuous-variables" title="Permalink"></a></h2><p>A semi-continuous variable is a continuous variable between bounds <span>$[l,u]$</span> that also can assume the value zero, that is: <span>$x \in \{0\} \cup [l,u].$</span></p><p><strong>Example</strong> <span>$x \in \{0\}\cup [1, 2]$</span></p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x in Semicontinuous(1.0, 2.0))</code><code class="nohighlight hljs ansi" style="display:block;">x</code></pre><p>You can also represent a semi-continuous variable using the reformulation:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x)</code><code class="nohighlight hljs ansi" style="display:block;">x</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, z, Bin)</code><code class="nohighlight hljs ansi" style="display:block;">z</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x &lt;= 2 * z)</code><code class="nohighlight hljs ansi" style="display:block;">x - 2 z ≤ 0</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x &gt;= 1 * z)</code><code class="nohighlight hljs ansi" style="display:block;">x - z ≥ 0</code></pre><p>When <code>z = 0</code> the two constraints are equivalent to <code>0 &lt;= x &lt;= 0</code>. When <code>z = 1</code>, the two constraints are equivalennt to <code>1 &lt;= x &lt;= 2</code>.</p><h2 id="Semi-integer-variables"><a class="docs-heading-anchor" href="#Semi-integer-variables">Semi-integer variables</a><a id="Semi-integer-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Semi-integer-variables" title="Permalink"></a></h2><p>A semi-integer variable is a variable which assumes integer values between bounds <span>$[l,u]$</span> and can also assume the value zero: <span>$x \in \{0\} \cup [l, u] \cap \mathbb{Z}.$</span></p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x in Semiinteger(5.0, 10.0))</code><code class="nohighlight hljs ansi" style="display:block;">x</code></pre><p>You can also represent a semi-integer variable using the reformulation:</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x, Int)</code><code class="nohighlight hljs ansi" style="display:block;">x</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, z, Bin)</code><code class="nohighlight hljs ansi" style="display:block;">z</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x &lt;= 10 * z)</code><code class="nohighlight hljs ansi" style="display:block;">x - 10 z ≤ 0</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x &gt;= 5 * z)</code><code class="nohighlight hljs ansi" style="display:block;">x - 5 z ≥ 0</code></pre><p>When <code>z = 0</code> the two constraints are equivalent to <code>0 &lt;= x &lt;= 0</code>. When <code>z = 1</code>, the two constraints are equivalennt to <code>5 &lt;= x &lt;= 10</code>.</p><h2 id="Special-Ordered-Sets-of-Type-1"><a class="docs-heading-anchor" href="#Special-Ordered-Sets-of-Type-1">Special Ordered Sets of Type 1</a><a id="Special-Ordered-Sets-of-Type-1-1"></a><a class="docs-heading-anchor-permalink" href="#Special-Ordered-Sets-of-Type-1" title="Permalink"></a></h2><p>A Special Ordered Set of Type 1 is a set of variables, at most one of which can take a non-zero value, all others being at 0.</p><p>They most frequently apply where a set of variables are actually binary variables. In other words, we have to choose at most one from a set of possibilities.</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x[1:3], Bin)</code><code class="nohighlight hljs ansi" style="display:block;">3-element Vector{VariableRef}:
+ x[2]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, z, Bin)</code><code class="nohighlight hljs ansi" style="display:block;">z</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; M = 100</code><code class="nohighlight hljs ansi" style="display:block;">100</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, sum(x) &lt;= 1 + M * z)</code><code class="nohighlight hljs ansi" style="display:block;">x[1] + x[2] - 100 z ≤ 1</code></pre><h2 id="Semi-continuous-variables"><a class="docs-heading-anchor" href="#Semi-continuous-variables">Semi-continuous variables</a><a id="Semi-continuous-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Semi-continuous-variables" title="Permalink"></a></h2><div class="admonition is-info"><header class="admonition-header">Info</header><div class="admonition-body"><p>This section uses sets from <a href="../../../moi/#moi_documentation">MathOptInterface</a>. By default, JuMP exports the <code>MOI</code> symbol as an alias for the MathOptInterface.jl package. We recommend making this more explicit in your code by adding the following lines:</p><pre><code class="language-julia hljs">import MathOptInterface as MOI</code></pre></div></div><p>A semi-continuous variable is a continuous variable between bounds <span>$[l,u]$</span> that also can assume the value zero, that is: <span>$x \in \{0\} \cup [l,u].$</span></p><p><strong>Example</strong> <span>$x \in \{0\}\cup [1, 2]$</span></p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x in Semicontinuous(1.0, 2.0))</code><code class="nohighlight hljs ansi" style="display:block;">x</code></pre><h2 id="Semi-integer-variables"><a class="docs-heading-anchor" href="#Semi-integer-variables">Semi-integer variables</a><a id="Semi-integer-variables-1"></a><a class="docs-heading-anchor-permalink" href="#Semi-integer-variables" title="Permalink"></a></h2><p>A semi-integer variable is a variable which assumes integer values between bounds <span>$[l,u]$</span> and can also assume the value zero: <span>$x \in \{0\} \cup [l, u] \cap \mathbb{Z}.$</span></p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x in Semiinteger(5.0, 10.0))</code><code class="nohighlight hljs ansi" style="display:block;">x</code></pre><h2 id="Special-Ordered-Sets-of-Type-1"><a class="docs-heading-anchor" href="#Special-Ordered-Sets-of-Type-1">Special Ordered Sets of Type 1</a><a id="Special-Ordered-Sets-of-Type-1-1"></a><a class="docs-heading-anchor-permalink" href="#Special-Ordered-Sets-of-Type-1" title="Permalink"></a></h2><p>A Special Ordered Set of Type 1 is a set of variables, at most one of which can take a non-zero value, all others being at 0.</p><p>They most frequently apply where a set of variables are actually binary variables. In other words, we have to choose at most one from a set of possibilities.</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x[1:3], Bin)</code><code class="nohighlight hljs ansi" style="display:block;">3-element Vector{VariableRef}:
  x[1]
  x[2]
  x[3]</code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x in SOS1())</code><code class="nohighlight hljs ansi" style="display:block;">[x[1], x[2], x[3]] ∈ MathOptInterface.SOS1{Float64}([1.0, 2.0, 3.0])</code></pre><p>You can optionally pass <code>SOS1</code> a weight vector like</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; @constraint(model, x in SOS1([0.2, 0.5, 0.3]))</code><code class="nohighlight hljs ansi" style="display:block;">[x[1], x[2], x[3]] ∈ MathOptInterface.SOS1{Float64}([0.2, 0.5, 0.3])</code></pre><p>If the decision variables are related and have a physical ordering, then the weight vector, although not used directly in the constraint, can help the solver make a better decision in the solution process.</p><h2 id="tip_sos2"><a class="docs-heading-anchor" href="#tip_sos2">Special Ordered Sets of Type 2</a><a id="tip_sos2-1"></a><a class="docs-heading-anchor-permalink" href="#tip_sos2" title="Permalink"></a></h2><p>A Special Ordered Set of type 2 is a set of non-negative variables, of which at most two can be non-zero, and if two are non-zero these must be consecutive in their ordering.</p><pre><code class="language-julia-repl hljs" style="display:block;">julia&gt; model = Model();</code><code class="nohighlight hljs ansi" style="display:block;"></code><br/><code class="language-julia-repl hljs" style="display:block;">julia&gt; @variable(model, x[1:3])</code><code class="nohighlight hljs ansi" style="display:block;">3-element Vector{VariableRef}:
@@ -64,4 +64,4 @@
            y == sum(ŷ[i] * λ[i] for i in 1:N)
            sum(λ) == 1
            λ in SOS2()
-       end)</code><code class="nohighlight hljs ansi" style="display:block;">(x + λ[1] + 0.5 λ[2] - 0.5 λ[4] - λ[5] - 1.5 λ[6] - 2 λ[7] = 0, y - λ[1] - 0.25 λ[2] - 0.25 λ[4] - λ[5] - 2.25 λ[6] - 4 λ[7] = 0, λ[1] + λ[2] + λ[3] + λ[4] + λ[5] + λ[6] + λ[7] = 1, [λ[1], λ[2], λ[3], λ[4], λ[5], λ[6], λ[7]] ∈ MathOptInterface.SOS2{Float64}([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0]))</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_commodity_network/">« The network multi-commodity flow problem</a><a class="docs-footer-nextpage" href="../facility_location/">The facility location problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+       end)</code><code class="nohighlight hljs ansi" style="display:block;">(x + λ[1] + 0.5 λ[2] - 0.5 λ[4] - λ[5] - 1.5 λ[6] - 2 λ[7] = 0, y - λ[1] - 0.25 λ[2] - 0.25 λ[4] - λ[5] - 2.25 λ[6] - 4 λ[7] = 0, λ[1] + λ[2] + λ[3] + λ[4] + λ[5] + λ[6] + λ[7] = 1, [λ[1], λ[2], λ[3], λ[4], λ[5], λ[6], λ[7]] ∈ MathOptInterface.SOS2{Float64}([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0]))</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../multi_commodity_network/">« The network multi-commodity flow problem</a><a class="docs-footer-nextpage" href="../facility_location/">The facility location problem »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/linear/transp/index.html b/previews/PR3562/tutorials/linear/transp/index.html
index 35cd611f610..d402f43d1aa 100644
--- a/previews/PR3562/tutorials/linear/transp/index.html
+++ b/previews/PR3562/tutorials/linear/transp/index.html
@@ -73,4 +73,4 @@
 end</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">solve_transportation_problem (generic function with 1 method)</code></pre><h2 id="Solution"><a class="docs-heading-anchor" href="#Solution">Solution</a><a id="Solution-1"></a><a class="docs-heading-anchor-permalink" href="#Solution" title="Permalink"></a></h2><p>Let&#39;s solve and view the solution:</p><pre><code class="language-julia hljs">solve_transportation_problem(data)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">        FRA    DET    LAN    WIN    STL    FRE    LAF
 GARY      .      .      .      .  300.0 1100.0      .
 CLEV      .      .  600.0      . 1000.0      . 1000.0
-PITT  900.0 1200.0      .  400.0  400.0      .      .</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../network_flows/">« Network flow problems</a><a class="docs-footer-nextpage" href="../multi_objective_knapsack/">Multi-objective knapsack »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+PITT  900.0 1200.0      .  400.0  400.0      .      .</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../network_flows/">« Network flow problems</a><a class="docs-footer-nextpage" href="../multi_objective_knapsack/">Multi-objective knapsack »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/complementarity/index.html b/previews/PR3562/tutorials/nonlinear/complementarity/index.html
index 93e21c27062..715607ed8e6 100644
--- a/previews/PR3562/tutorials/nonlinear/complementarity/index.html
+++ b/previews/PR3562/tutorials/nonlinear/complementarity/index.html
@@ -126,7 +126,7 @@
   Objective value    : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 1.51000e-04
+  Solve time (sec)   : 2.05000e-04
 </code></pre><p>An equilibrium solution is to build 389 MW:</p><pre><code class="language-julia hljs">value(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">389.31506849315065</code></pre><p>The production in each scenario is:</p><pre><code class="language-julia hljs">value.(Q)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">5-element Vector{Float64}:
  240.00000000000014
  289.99999999999994
@@ -137,4 +137,4 @@
   60.0
   59.99999999999999
   60.68493150684928
- 110.68493150684935</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../querying_hessians/">« Computing Hessians</a><a class="docs-footer-nextpage" href="../../conic/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 110.68493150684935</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../querying_hessians/">« Computing Hessians</a><a class="docs-footer-nextpage" href="../../conic/introduction/">Introduction »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/introduction/index.html b/previews/PR3562/tutorials/nonlinear/introduction/index.html
index 6c4081a2b24..459ac4f265f 100644
--- a/previews/PR3562/tutorials/nonlinear/introduction/index.html
+++ b/previews/PR3562/tutorials/nonlinear/introduction/index.html
@@ -7,4 +7,4 @@
     \min_{x \in \mathbb{R}^n} &amp; f_0(x) \\
     \;\;\text{s.t.} &amp; l_j \le f_j(x) \le u_j &amp; j = 1 \ldots m \\
     &amp; l_i \le x_i \le u_i &amp; i = 1 \ldots n.
-\end{align}\]</p><p>Mixed-integer nonlinear linear programs (MINLPs) are extensions of nonlinear programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>JuMP supports a range of nonlinear solvers; look for &quot;NLP&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>. However, very few solvers support mixed-integer nonlinear linear programs. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><p>If the only nonlinearities in your model are quadratic terms (that is, multiplication between two decision variables), you can also use second-order cone solvers, which are indicated by &quot;SOCP.&quot; In most cases, these solvers are restricted to convex quadratic problems and will error if you pass a nonconvex quadratic function; however, Gurobi has the ability to solve nonconvex quadratic terms.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../rocket_control/#Rocket-Control">Rocket Control</a></li><li><a href="../space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a href="../portfolio/#Portfolio-optimization">Portfolio optimization</a></li></ul></li><li>The <a href="../tips_and_tricks/#nonlinear_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling nonlinear programs. Look here if you are stuck trying to formulate a problem as a nonlinear program.</li><li>The <a href="../querying_hessians/#Computing-Hessians">Computing Hessians</a> is an advanced tutorial which explains how to compute the Hessian of the Lagrangian of a nonlinear program. This is useful only in particular cases.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../linear/mip_duality/">« Computing the duals of a mixed-integer program</a><a class="docs-footer-nextpage" href="../simple_examples/">Simple examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+\end{align}\]</p><p>Mixed-integer nonlinear linear programs (MINLPs) are extensions of nonlinear programs in which some (or all) of the decision variables take discrete values.</p><h2 id="How-to-choose-a-solver"><a class="docs-heading-anchor" href="#How-to-choose-a-solver">How to choose a solver</a><a id="How-to-choose-a-solver-1"></a><a class="docs-heading-anchor-permalink" href="#How-to-choose-a-solver" title="Permalink"></a></h2><p>JuMP supports a range of nonlinear solvers; look for &quot;NLP&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>. However, very few solvers support mixed-integer nonlinear linear programs. Solvers supporting discrete variables start with &quot;(MI)&quot; in the list of <a href="../../../installation/#Supported-solvers">Supported solvers</a>.</p><p>If the only nonlinearities in your model are quadratic terms (that is, multiplication between two decision variables), you can also use second-order cone solvers, which are indicated by &quot;SOCP.&quot; In most cases, these solvers are restricted to convex quadratic problems and will error if you pass a nonconvex quadratic function; however, Gurobi has the ability to solve nonconvex quadratic terms.</p><h2 id="How-these-tutorials-are-structured"><a class="docs-heading-anchor" href="#How-these-tutorials-are-structured">How these tutorials are structured</a><a id="How-these-tutorials-are-structured-1"></a><a class="docs-heading-anchor-permalink" href="#How-these-tutorials-are-structured" title="Permalink"></a></h2><p>Having a high-level overview of how this part of the documentation is structured will help you know where to look for certain things.</p><ul><li>The following tutorials are worked examples that present a problem in words, then formulate it in mathematics, and then solve it in JuMP. This usually involves some sort of visualization of the solution. Start here if you are new to JuMP.<ul><li><a href="../rocket_control/#Rocket-Control">Rocket Control</a></li><li><a href="../space_shuttle_reentry_trajectory/#Optimal-control-for-a-Space-Shuttle-reentry-trajectory">Optimal control for a Space Shuttle reentry trajectory</a></li><li><a href="../portfolio/#Portfolio-optimization">Portfolio optimization</a></li></ul></li><li>The <a href="../tips_and_tricks/#nonlinear_tips_and_tricks">Tips and tricks</a> tutorial contains a number of helpful reformulations and tricks you can use when modeling nonlinear programs. Look here if you are stuck trying to formulate a problem as a nonlinear program.</li><li>The <a href="../querying_hessians/#Computing-Hessians">Computing Hessians</a> is an advanced tutorial which explains how to compute the Hessian of the Lagrangian of a nonlinear program. This is useful only in particular cases.</li><li>The remaining tutorials are less verbose and styled in the form of short code examples. These tutorials have less explanation, but may contain useful code snippets, particularly if they are similar to a problem you are trying to solve.</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../../linear/mip_duality/">« Computing the duals of a mixed-integer program</a><a class="docs-footer-nextpage" href="../simple_examples/">Simple examples »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/nested_problems/index.html b/previews/PR3562/tutorials/nonlinear/nested_problems/index.html
index 7c5ca1fbfbb..5623c980fa8 100644
--- a/previews/PR3562/tutorials/nonlinear/nested_problems/index.html
+++ b/previews/PR3562/tutorials/nonlinear/nested_problems/index.html
@@ -63,14 +63,14 @@
   Dual objective value : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 6.93423e-01
-  Barrier iterations : 32
-</code></pre><p>The optimal objective value is:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">-418983.48680640775</code></pre><p>and the optimal upper-level decision variables <span>$x$</span> are:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 154.97862337234338
- 180.0096143098799</code></pre><p>To compute the optimal lower-level decision variables, we need to call <code>solve_lower_level</code> with the optimal upper-level decision variables:</p><pre><code class="language-julia hljs">_, y = solve_lower_level(value.(x)...)
+  Solve time (sec)   : 9.46160e-01
+  Barrier iterations : 34
+</code></pre><p>The optimal objective value is:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">-418983.4868064087</code></pre><p>and the optimal upper-level decision variables <span>$x$</span> are:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
+ 154.97862339279123
+ 180.00961428934832</code></pre><p>To compute the optimal lower-level decision variables, we need to call <code>solve_lower_level</code> with the optimal upper-level decision variables:</p><pre><code class="language-julia hljs">_, y = solve_lower_level(value.(x)...)
 y</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 7.072593961143734
- 5.94656989283847</code></pre><h2 id="Improving-performance"><a class="docs-heading-anchor" href="#Improving-performance">Improving performance</a><a id="Improving-performance-1"></a><a class="docs-heading-anchor-permalink" href="#Improving-performance" title="Permalink"></a></h2><p>Our solution approach works, but it has a performance problem: every time we need to compute the value, gradient, or Hessian of <span>$V$</span>, we have to re-solve the lower-level optimization problem. This is wasteful, because we will often call the gradient and Hessian at the same point, and so solving the problem twice with the same input repeats work unnecessarily.</p><p>We can work around this by using a cache:</p><pre><code class="language-julia hljs">mutable struct Cache
+ 7.0725939609898285
+ 5.946569892949626</code></pre><h2 id="Improving-performance"><a class="docs-heading-anchor" href="#Improving-performance">Improving performance</a><a id="Improving-performance-1"></a><a class="docs-heading-anchor-permalink" href="#Improving-performance" title="Permalink"></a></h2><p>Our solution approach works, but it has a performance problem: every time we need to compute the value, gradient, or Hessian of <span>$V$</span>, we have to re-solve the lower-level optimization problem. This is wasteful, because we will often call the gradient and Hessian at the same point, and so solving the problem twice with the same input repeats work unnecessarily.</p><p>We can work around this by using a cache:</p><pre><code class="language-julia hljs">mutable struct Cache
     x::Any
     f::Float64
     y::Vector{Float64}
@@ -125,8 +125,8 @@
   Dual objective value : 0.00000e+00
 
 * Work counters
-  Solve time (sec)   : 2.47634e-01
-  Barrier iterations : 32
-</code></pre><p>an we can check we get the same objective value:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">-418983.48680640775</code></pre><p>and upper-level decision variable <span>$x$</span>:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
- 154.97862337234338
- 180.0096143098799</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../user_defined_hessians/">« User-defined Hessians</a><a class="docs-footer-nextpage" href="../querying_hessians/">Computing Hessians »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+  Solve time (sec)   : 3.78973e-01
+  Barrier iterations : 34
+</code></pre><p>an we can check we get the same objective value:</p><pre><code class="language-julia hljs">objective_value(model)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">-418983.4868064087</code></pre><p>and upper-level decision variable <span>$x$</span>:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
+ 154.97862339279123
+ 180.00961428934832</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../user_defined_hessians/">« User-defined Hessians</a><a class="docs-footer-nextpage" href="../querying_hessians/">Computing Hessians »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/portfolio/a315d69e.svg b/previews/PR3562/tutorials/nonlinear/portfolio/e1860e4a.svg
similarity index 79%
rename from previews/PR3562/tutorials/nonlinear/portfolio/a315d69e.svg
rename to previews/PR3562/tutorials/nonlinear/portfolio/e1860e4a.svg
index 094bae7c34d..3092f359955 100644
--- a/previews/PR3562/tutorials/nonlinear/portfolio/a315d69e.svg
+++ b/previews/PR3562/tutorials/nonlinear/portfolio/e1860e4a.svg
@@ -1,581 +1,581 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="600" viewBox="0 0 2400 2400">
 <defs>
-  <clipPath id="clip090">
+  <clipPath id="clip910">
     <rect x="0" y="0" width="2400" height="2400"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip090)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip910)" d="M0 2400 L2400 2400 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip091">
+  <clipPath id="clip911">
     <rect x="480" y="240" width="1681" height="1681"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip090)" d="M286.311 964.927 L2352.76 964.927 L2352.76 153.712 L286.311 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip910)" d="M286.311 964.927 L2352.76 964.927 L2352.76 153.712 L286.311 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip092">
+  <clipPath id="clip912">
     <rect x="286" y="153" width="2067" height="812"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="344.796,964.927 344.796,153.712 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="912.471,964.927 912.471,153.712 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1480.15,964.927 1480.15,153.712 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2047.82,964.927 2047.82,153.712 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,964.927 2352.76,964.927 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="344.796,964.927 344.796,946.03 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="912.471,964.927 912.471,946.03 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1480.15,964.927 1480.15,946.03 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2047.82,964.927 2047.82,946.03 "/>
-<path clip-path="url(#clip090)" d="M344.796 995.846 Q341.184 995.846 339.356 999.411 Q337.55 1002.95 337.55 1010.08 Q337.55 1017.19 339.356 1020.75 Q341.184 1024.3 344.796 1024.3 Q348.43 1024.3 350.235 1020.75 Q352.064 1017.19 352.064 1010.08 Q352.064 1002.95 350.235 999.411 Q348.43 995.846 344.796 995.846 M344.796 992.142 Q350.606 992.142 353.661 996.749 Q356.74 1001.33 356.74 1010.08 Q356.74 1018.81 353.661 1023.42 Q350.606 1028 344.796 1028 Q338.985 1028 335.907 1023.42 Q332.851 1018.81 332.851 1010.08 Q332.851 1001.33 335.907 996.749 Q338.985 992.142 344.796 992.142 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M823.455 1039.05 L839.775 1039.05 L839.775 1042.99 L817.83 1042.99 L817.83 1039.05 Q820.492 1036.3 825.076 1031.67 Q829.682 1027.02 830.863 1025.67 Q833.108 1023.15 833.988 1021.41 Q834.89 1019.65 834.89 1017.96 Q834.89 1015.21 832.946 1013.47 Q831.025 1011.74 827.923 1011.74 Q825.724 1011.74 823.27 1012.5 Q820.84 1013.27 818.062 1014.82 L818.062 1010.09 Q820.886 1008.96 823.34 1008.38 Q825.793 1007.8 827.83 1007.8 Q833.201 1007.8 836.395 1010.49 Q839.589 1013.17 839.589 1017.66 Q839.589 1019.79 838.779 1021.71 Q837.992 1023.61 835.886 1026.21 Q835.307 1026.88 832.205 1030.09 Q829.103 1033.29 823.455 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M849.589 1037.11 L854.474 1037.11 L854.474 1042.99 L849.589 1042.99 L849.589 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M874.659 1011.51 Q871.048 1011.51 869.219 1015.07 Q867.413 1018.61 867.413 1025.74 Q867.413 1032.85 869.219 1036.41 Q871.048 1039.96 874.659 1039.96 Q878.293 1039.96 880.098 1036.41 Q881.927 1032.85 881.927 1025.74 Q881.927 1018.61 880.098 1015.07 Q878.293 1011.51 874.659 1011.51 M874.659 1007.8 Q880.469 1007.8 883.524 1012.41 Q886.603 1016.99 886.603 1025.74 Q886.603 1034.47 883.524 1039.08 Q880.469 1043.66 874.659 1043.66 Q868.849 1043.66 865.77 1039.08 Q862.714 1034.47 862.714 1025.74 Q862.714 1016.99 865.77 1012.41 Q868.849 1007.8 874.659 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M922.992 1017.52 L912.413 1028.15 L922.992 1038.73 L920.237 1041.53 L909.612 1030.9 L898.987 1041.53 L896.256 1038.73 L906.811 1028.15 L896.256 1017.52 L898.987 1014.72 L909.612 1025.35 L920.237 1014.72 L922.992 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M935.353 1039.05 L942.992 1039.05 L942.992 1012.69 L934.682 1014.35 L934.682 1010.09 L942.945 1008.43 L947.621 1008.43 L947.621 1039.05 L955.26 1039.05 L955.26 1042.99 L935.353 1042.99 L935.353 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M974.704 1011.51 Q971.093 1011.51 969.265 1015.07 Q967.459 1018.61 967.459 1025.74 Q967.459 1032.85 969.265 1036.41 Q971.093 1039.96 974.704 1039.96 Q978.339 1039.96 980.144 1036.41 Q981.973 1032.85 981.973 1025.74 Q981.973 1018.61 980.144 1015.07 Q978.339 1011.51 974.704 1011.51 M974.704 1007.8 Q980.515 1007.8 983.57 1012.41 Q986.649 1016.99 986.649 1025.74 Q986.649 1034.47 983.57 1039.08 Q980.515 1043.66 974.704 1043.66 Q968.894 1043.66 965.816 1039.08 Q962.76 1034.47 962.76 1025.74 Q962.76 1016.99 965.816 1012.41 Q968.894 1007.8 974.704 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M999.325 990.807 L989.733 1005.8 L999.325 1005.8 L999.325 990.807 M998.328 987.497 L1003.11 987.497 L1003.11 1005.8 L1007.11 1005.8 L1007.11 1008.96 L1003.11 1008.96 L1003.11 1015.58 L999.325 1015.58 L999.325 1008.96 L986.649 1008.96 L986.649 1005.29 L998.328 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1400.53 1012.5 L1388.72 1030.95 L1400.53 1030.95 L1400.53 1012.5 M1399.3 1008.43 L1405.18 1008.43 L1405.18 1030.95 L1410.11 1030.95 L1410.11 1034.84 L1405.18 1034.84 L1405.18 1042.99 L1400.53 1042.99 L1400.53 1034.84 L1384.93 1034.84 L1384.93 1030.33 L1399.3 1008.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1417.84 1037.11 L1422.73 1037.11 L1422.73 1042.99 L1417.84 1042.99 L1417.84 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1442.91 1011.51 Q1439.3 1011.51 1437.47 1015.07 Q1435.67 1018.61 1435.67 1025.74 Q1435.67 1032.85 1437.47 1036.41 Q1439.3 1039.96 1442.91 1039.96 Q1446.55 1039.96 1448.35 1036.41 Q1450.18 1032.85 1450.18 1025.74 Q1450.18 1018.61 1448.35 1015.07 Q1446.55 1011.51 1442.91 1011.51 M1442.91 1007.8 Q1448.72 1007.8 1451.78 1012.41 Q1454.86 1016.99 1454.86 1025.74 Q1454.86 1034.47 1451.78 1039.08 Q1448.72 1043.66 1442.91 1043.66 Q1437.1 1043.66 1434.02 1039.08 Q1430.97 1034.47 1430.97 1025.74 Q1430.97 1016.99 1434.02 1012.41 Q1437.1 1007.8 1442.91 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1491.25 1017.52 L1480.67 1028.15 L1491.25 1038.73 L1488.49 1041.53 L1477.87 1030.9 L1467.24 1041.53 L1464.51 1038.73 L1475.07 1028.15 L1464.51 1017.52 L1467.24 1014.72 L1477.87 1025.35 L1488.49 1014.72 L1491.25 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1503.61 1039.05 L1511.25 1039.05 L1511.25 1012.69 L1502.94 1014.35 L1502.94 1010.09 L1511.2 1008.43 L1515.88 1008.43 L1515.88 1039.05 L1523.51 1039.05 L1523.51 1042.99 L1503.61 1042.99 L1503.61 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1542.96 1011.51 Q1539.35 1011.51 1537.52 1015.07 Q1535.71 1018.61 1535.71 1025.74 Q1535.71 1032.85 1537.52 1036.41 Q1539.35 1039.96 1542.96 1039.96 Q1546.59 1039.96 1548.4 1036.41 Q1550.23 1032.85 1550.23 1025.74 Q1550.23 1018.61 1548.4 1015.07 Q1546.59 1011.51 1542.96 1011.51 M1542.96 1007.8 Q1548.77 1007.8 1551.82 1012.41 Q1554.9 1016.99 1554.9 1025.74 Q1554.9 1034.47 1551.82 1039.08 Q1548.77 1043.66 1542.96 1043.66 Q1537.15 1043.66 1534.07 1039.08 Q1531.01 1034.47 1531.01 1025.74 Q1531.01 1016.99 1534.07 1012.41 Q1537.15 1007.8 1542.96 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1567.58 990.807 L1557.99 1005.8 L1567.58 1005.8 L1567.58 990.807 M1566.58 987.497 L1571.36 987.497 L1571.36 1005.8 L1575.37 1005.8 L1575.37 1008.96 L1571.36 1008.96 L1571.36 1015.58 L1567.58 1015.58 L1567.58 1008.96 L1554.9 1008.96 L1554.9 1005.29 L1566.58 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1965.44 1023.84 Q1962.29 1023.84 1960.44 1026 Q1958.61 1028.15 1958.61 1031.9 Q1958.61 1035.63 1960.44 1037.8 Q1962.29 1039.96 1965.44 1039.96 Q1968.59 1039.96 1970.41 1037.8 Q1972.27 1035.63 1972.27 1031.9 Q1972.27 1028.15 1970.41 1026 Q1968.59 1023.84 1965.44 1023.84 M1974.72 1009.19 L1974.72 1013.45 Q1972.96 1012.62 1971.16 1012.18 Q1969.37 1011.74 1967.61 1011.74 Q1962.98 1011.74 1960.53 1014.86 Q1958.1 1017.99 1957.75 1024.31 Q1959.12 1022.29 1961.18 1021.23 Q1963.24 1020.14 1965.72 1020.14 Q1970.92 1020.14 1973.93 1023.31 Q1976.97 1026.46 1976.97 1031.9 Q1976.97 1037.22 1973.82 1040.44 Q1970.67 1043.66 1965.44 1043.66 Q1959.44 1043.66 1956.27 1039.08 Q1953.1 1034.47 1953.1 1025.74 Q1953.1 1017.55 1956.99 1012.69 Q1960.88 1007.8 1967.43 1007.8 Q1969.19 1007.8 1970.97 1008.15 Q1972.78 1008.5 1974.72 1009.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1985.02 1037.11 L1989.91 1037.11 L1989.91 1042.99 L1985.02 1042.99 L1985.02 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2010.09 1011.51 Q2006.48 1011.51 2004.65 1015.07 Q2002.85 1018.61 2002.85 1025.74 Q2002.85 1032.85 2004.65 1036.41 Q2006.48 1039.96 2010.09 1039.96 Q2013.72 1039.96 2015.53 1036.41 Q2017.36 1032.85 2017.36 1025.74 Q2017.36 1018.61 2015.53 1015.07 Q2013.72 1011.51 2010.09 1011.51 M2010.09 1007.8 Q2015.9 1007.8 2018.96 1012.41 Q2022.04 1016.99 2022.04 1025.74 Q2022.04 1034.47 2018.96 1039.08 Q2015.9 1043.66 2010.09 1043.66 Q2004.28 1043.66 2001.2 1039.08 Q1998.15 1034.47 1998.15 1025.74 Q1998.15 1016.99 2001.2 1012.41 Q2004.28 1007.8 2010.09 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2058.42 1017.52 L2047.85 1028.15 L2058.42 1038.73 L2055.67 1041.53 L2045.04 1030.9 L2034.42 1041.53 L2031.69 1038.73 L2042.24 1028.15 L2031.69 1017.52 L2034.42 1014.72 L2045.04 1025.35 L2055.67 1014.72 L2058.42 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2070.78 1039.05 L2078.42 1039.05 L2078.42 1012.69 L2070.11 1014.35 L2070.11 1010.09 L2078.38 1008.43 L2083.05 1008.43 L2083.05 1039.05 L2090.69 1039.05 L2090.69 1042.99 L2070.78 1042.99 L2070.78 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2110.14 1011.51 Q2106.53 1011.51 2104.7 1015.07 Q2102.89 1018.61 2102.89 1025.74 Q2102.89 1032.85 2104.7 1036.41 Q2106.53 1039.96 2110.14 1039.96 Q2113.77 1039.96 2115.58 1036.41 Q2117.4 1032.85 2117.4 1025.74 Q2117.4 1018.61 2115.58 1015.07 Q2113.77 1011.51 2110.14 1011.51 M2110.14 1007.8 Q2115.95 1007.8 2119 1012.41 Q2122.08 1016.99 2122.08 1025.74 Q2122.08 1034.47 2119 1039.08 Q2115.95 1043.66 2110.14 1043.66 Q2104.33 1043.66 2101.25 1039.08 Q2098.19 1034.47 2098.19 1025.74 Q2098.19 1016.99 2101.25 1012.41 Q2104.33 1007.8 2110.14 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2134.76 990.807 L2125.17 1005.8 L2134.76 1005.8 L2134.76 990.807 M2133.76 987.497 L2138.54 987.497 L2138.54 1005.8 L2142.54 1005.8 L2142.54 1008.96 L2138.54 1008.96 L2138.54 1015.58 L2134.76 1015.58 L2134.76 1008.96 L2122.08 1008.96 L2122.08 1005.29 L2133.76 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1198.87 1128.22 L1180.73 1080.7 L1187.45 1080.7 L1202.5 1120.7 L1217.59 1080.7 L1224.27 1080.7 L1206.16 1128.22 L1198.87 1128.22 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1242.09 1110.3 Q1235 1110.3 1232.26 1111.92 Q1229.52 1113.54 1229.52 1117.46 Q1229.52 1120.58 1231.56 1122.42 Q1233.63 1124.24 1237.16 1124.24 Q1242.03 1124.24 1244.96 1120.8 Q1247.92 1117.33 1247.92 1111.6 L1247.92 1110.3 L1242.09 1110.3 M1253.78 1107.88 L1253.78 1128.22 L1247.92 1128.22 L1247.92 1122.81 Q1245.91 1126.05 1242.92 1127.61 Q1239.93 1129.14 1235.6 1129.14 Q1230.13 1129.14 1226.88 1126.08 Q1223.67 1123 1223.67 1117.84 Q1223.67 1111.82 1227.68 1108.77 Q1231.72 1105.71 1239.71 1105.71 L1247.92 1105.71 L1247.92 1105.14 Q1247.92 1101.1 1245.25 1098.9 Q1242.6 1096.67 1237.8 1096.67 Q1234.74 1096.67 1231.85 1097.41 Q1228.95 1098.14 1226.28 1099.6 L1226.28 1094.19 Q1229.49 1092.95 1232.51 1092.35 Q1235.54 1091.71 1238.4 1091.71 Q1246.14 1091.71 1249.96 1095.72 Q1253.78 1099.73 1253.78 1107.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1286.5 1098.04 Q1285.51 1097.47 1284.33 1097.22 Q1283.19 1096.93 1281.78 1096.93 Q1276.82 1096.93 1274.15 1100.18 Q1271.5 1103.39 1271.5 1109.44 L1271.5 1128.22 L1265.62 1128.22 L1265.62 1092.57 L1271.5 1092.57 L1271.5 1098.11 Q1273.35 1094.86 1276.31 1093.3 Q1279.27 1091.71 1283.5 1091.71 Q1284.11 1091.71 1284.84 1091.8 Q1285.57 1091.87 1286.46 1092.03 L1286.5 1098.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1292.64 1092.57 L1298.49 1092.57 L1298.49 1128.22 L1292.64 1128.22 L1292.64 1092.57 M1292.64 1078.69 L1298.49 1078.69 L1298.49 1086.11 L1292.64 1086.11 L1292.64 1078.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1326.95 1110.3 Q1319.85 1110.3 1317.11 1111.92 Q1314.38 1113.54 1314.38 1117.46 Q1314.38 1120.58 1316.41 1122.42 Q1318.48 1124.24 1322.02 1124.24 Q1326.89 1124.24 1329.81 1120.8 Q1332.77 1117.33 1332.77 1111.6 L1332.77 1110.3 L1326.95 1110.3 M1338.63 1107.88 L1338.63 1128.22 L1332.77 1128.22 L1332.77 1122.81 Q1330.77 1126.05 1327.78 1127.61 Q1324.79 1129.14 1320.46 1129.14 Q1314.98 1129.14 1311.74 1126.08 Q1308.52 1123 1308.52 1117.84 Q1308.52 1111.82 1312.53 1108.77 Q1316.57 1105.71 1324.56 1105.71 L1332.77 1105.71 L1332.77 1105.14 Q1332.77 1101.1 1330.1 1098.9 Q1327.46 1096.67 1322.65 1096.67 Q1319.6 1096.67 1316.7 1097.41 Q1313.8 1098.14 1311.13 1099.6 L1311.13 1094.19 Q1314.35 1092.95 1317.37 1092.35 Q1320.39 1091.71 1323.26 1091.71 Q1330.99 1091.71 1334.81 1095.72 Q1338.63 1099.73 1338.63 1107.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1380.33 1106.7 L1380.33 1128.22 L1374.47 1128.22 L1374.47 1106.89 Q1374.47 1101.83 1372.5 1099.32 Q1370.52 1096.8 1366.58 1096.8 Q1361.83 1096.8 1359.1 1099.83 Q1356.36 1102.85 1356.36 1108.07 L1356.36 1128.22 L1350.47 1128.22 L1350.47 1092.57 L1356.36 1092.57 L1356.36 1098.11 Q1358.46 1094.89 1361.29 1093.3 Q1364.16 1091.71 1367.88 1091.71 Q1374.02 1091.71 1377.18 1095.53 Q1380.33 1099.32 1380.33 1106.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1417.66 1093.94 L1417.66 1099.41 Q1415.18 1098.04 1412.66 1097.37 Q1410.18 1096.67 1407.63 1096.67 Q1401.94 1096.67 1398.79 1100.3 Q1395.64 1103.9 1395.64 1110.42 Q1395.64 1116.95 1398.79 1120.58 Q1401.94 1124.17 1407.63 1124.17 Q1410.18 1124.17 1412.66 1123.51 Q1415.18 1122.81 1417.66 1121.44 L1417.66 1126.85 Q1415.21 1127.99 1412.57 1128.57 Q1409.96 1129.14 1407 1129.14 Q1398.95 1129.14 1394.2 1124.08 Q1389.46 1119.02 1389.46 1110.42 Q1389.46 1101.7 1394.24 1096.71 Q1399.04 1091.71 1407.38 1091.71 Q1410.09 1091.71 1412.66 1092.28 Q1415.24 1092.82 1417.66 1093.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1458.34 1108.93 L1458.34 1111.79 L1431.41 1111.79 Q1431.79 1117.84 1435.04 1121.02 Q1438.32 1124.17 1444.14 1124.17 Q1447.52 1124.17 1450.67 1123.35 Q1453.85 1122.52 1456.97 1120.86 L1456.97 1126.4 Q1453.82 1127.74 1450.51 1128.44 Q1447.2 1129.14 1443.79 1129.14 Q1435.26 1129.14 1430.27 1124.17 Q1425.3 1119.21 1425.3 1110.74 Q1425.3 1101.99 1430.01 1096.87 Q1434.75 1091.71 1442.77 1091.71 Q1449.97 1091.71 1454.14 1096.36 Q1458.34 1100.97 1458.34 1108.93 M1452.48 1107.21 Q1452.42 1102.4 1449.78 1099.54 Q1447.17 1096.67 1442.84 1096.67 Q1437.94 1096.67 1434.98 1099.44 Q1432.05 1102.21 1431.6 1107.24 L1452.48 1107.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,941.969 2352.76,941.969 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,734.335 2352.76,734.335 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,526.7 2352.76,526.7 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,319.066 2352.76,319.066 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,964.927 286.311,153.712 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,941.969 305.209,941.969 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,734.335 305.209,734.335 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,526.7 305.209,526.7 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,319.066 305.209,319.066 "/>
-<path clip-path="url(#clip090)" d="M238.367 927.768 Q234.756 927.768 232.927 931.333 Q231.122 934.874 231.122 942.004 Q231.122 949.11 232.927 952.675 Q234.756 956.217 238.367 956.217 Q242.001 956.217 243.807 952.675 Q245.635 949.11 245.635 942.004 Q245.635 934.874 243.807 931.333 Q242.001 927.768 238.367 927.768 M238.367 924.064 Q244.177 924.064 247.233 928.671 Q250.311 933.254 250.311 942.004 Q250.311 950.731 247.233 955.337 Q244.177 959.92 238.367 959.92 Q232.557 959.92 229.478 955.337 Q226.422 950.731 226.422 942.004 Q226.422 933.254 229.478 928.671 Q232.557 924.064 238.367 924.064 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M202.233 747.679 L218.552 747.679 L218.552 751.615 L196.608 751.615 L196.608 747.679 Q199.27 744.925 203.853 740.295 Q208.46 735.643 209.64 734.3 Q211.886 731.777 212.765 730.041 Q213.668 728.281 213.668 726.592 Q213.668 723.837 211.723 722.101 Q209.802 720.365 206.7 720.365 Q204.501 720.365 202.048 721.129 Q199.617 721.893 196.839 723.444 L196.839 718.721 Q199.663 717.587 202.117 717.008 Q204.571 716.43 206.608 716.43 Q211.978 716.43 215.173 719.115 Q218.367 721.8 218.367 726.291 Q218.367 728.42 217.557 730.342 Q216.77 732.24 214.663 734.832 Q214.085 735.504 210.983 738.721 Q207.881 741.916 202.233 747.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 720.133 Q234.756 720.133 232.927 723.698 Q231.122 727.24 231.122 734.369 Q231.122 741.476 232.927 745.041 Q234.756 748.582 238.367 748.582 Q242.001 748.582 243.807 745.041 Q245.635 741.476 245.635 734.369 Q245.635 727.24 243.807 723.698 Q242.001 720.133 238.367 720.133 M238.367 716.43 Q244.177 716.43 247.233 721.036 Q250.311 725.619 250.311 734.369 Q250.311 743.096 247.233 747.703 Q244.177 752.286 238.367 752.286 Q232.557 752.286 229.478 747.703 Q226.422 743.096 226.422 734.369 Q226.422 725.619 229.478 721.036 Q232.557 716.43 238.367 716.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M211.052 513.494 L199.247 531.943 L211.052 531.943 L211.052 513.494 M209.825 509.42 L215.705 509.42 L215.705 531.943 L220.635 531.943 L220.635 535.832 L215.705 535.832 L215.705 543.98 L211.052 543.98 L211.052 535.832 L195.45 535.832 L195.45 531.318 L209.825 509.42 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 512.499 Q234.756 512.499 232.927 516.064 Q231.122 519.605 231.122 526.735 Q231.122 533.841 232.927 537.406 Q234.756 540.948 238.367 540.948 Q242.001 540.948 243.807 537.406 Q245.635 533.841 245.635 526.735 Q245.635 519.605 243.807 516.064 Q242.001 512.499 238.367 512.499 M238.367 508.795 Q244.177 508.795 247.233 513.402 Q250.311 517.985 250.311 526.735 Q250.311 535.462 247.233 540.068 Q244.177 544.652 238.367 544.652 Q232.557 544.652 229.478 540.068 Q226.422 535.462 226.422 526.735 Q226.422 517.985 229.478 513.402 Q232.557 508.795 238.367 508.795 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M208.784 317.202 Q205.636 317.202 203.784 319.355 Q201.955 321.508 201.955 325.258 Q201.955 328.985 203.784 331.161 Q205.636 333.313 208.784 333.313 Q211.932 333.313 213.76 331.161 Q215.612 328.985 215.612 325.258 Q215.612 321.508 213.76 319.355 Q211.932 317.202 208.784 317.202 M218.066 302.55 L218.066 306.809 Q216.307 305.976 214.501 305.536 Q212.719 305.096 210.96 305.096 Q206.33 305.096 203.876 308.221 Q201.446 311.346 201.099 317.665 Q202.464 315.651 204.524 314.587 Q206.585 313.499 209.061 313.499 Q214.27 313.499 217.279 316.67 Q220.311 319.818 220.311 325.258 Q220.311 330.582 217.163 333.8 Q214.015 337.017 208.784 337.017 Q202.788 337.017 199.617 332.434 Q196.446 327.827 196.446 319.101 Q196.446 310.906 200.335 306.045 Q204.224 301.161 210.774 301.161 Q212.534 301.161 214.316 301.508 Q216.122 301.855 218.066 302.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 304.864 Q234.756 304.864 232.927 308.429 Q231.122 311.971 231.122 319.101 Q231.122 326.207 232.927 329.772 Q234.756 333.313 238.367 333.313 Q242.001 333.313 243.807 329.772 Q245.635 326.207 245.635 319.101 Q245.635 311.971 243.807 308.429 Q242.001 304.864 238.367 304.864 M238.367 301.161 Q244.177 301.161 247.233 305.767 Q250.311 310.351 250.311 319.101 Q250.311 327.827 247.233 332.434 Q244.177 337.017 238.367 337.017 Q232.557 337.017 229.478 332.434 Q226.422 327.827 226.422 319.101 Q226.422 310.351 229.478 305.767 Q232.557 301.161 238.367 301.161 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M97.6744 823.926 L97.6744 793.88 L103.085 793.88 L103.085 817.497 L117.153 817.497 L117.153 794.867 L122.564 794.867 L122.564 817.497 L139.784 817.497 L139.784 793.307 L145.194 793.307 L145.194 823.926 L97.6744 823.926 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M109.546 753.362 L126.893 766.253 L145.194 752.694 L145.194 759.601 L131.19 769.977 L145.194 780.353 L145.194 787.26 L126.543 773.414 L109.546 786.082 L109.546 779.175 L122.246 769.722 L109.546 760.269 L109.546 753.362 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M139.847 738.753 L158.753 738.753 L158.753 744.641 L109.546 744.641 L109.546 738.753 L114.957 738.753 Q111.774 736.907 110.247 734.106 Q108.687 731.273 108.687 727.359 Q108.687 720.866 113.843 716.823 Q119 712.749 127.402 712.749 Q135.805 712.749 140.961 716.823 Q146.117 720.866 146.117 727.359 Q146.117 731.273 144.59 734.106 Q143.03 736.907 139.847 738.753 M127.402 718.828 Q120.941 718.828 117.281 721.502 Q113.589 724.144 113.589 728.791 Q113.589 733.438 117.281 736.111 Q120.941 738.753 127.402 738.753 Q133.863 738.753 137.556 736.111 Q141.216 733.438 141.216 728.791 Q141.216 724.144 137.556 721.502 Q133.863 718.828 127.402 718.828 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M125.906 672.55 L128.771 672.55 L128.771 699.477 Q134.818 699.095 138.001 695.848 Q141.152 692.57 141.152 686.745 Q141.152 683.371 140.325 680.22 Q139.497 677.038 137.842 673.918 L143.38 673.918 Q144.717 677.069 145.417 680.38 Q146.117 683.69 146.117 687.095 Q146.117 695.625 141.152 700.623 Q136.187 705.588 127.721 705.588 Q118.968 705.588 113.843 700.877 Q108.687 696.135 108.687 688.114 Q108.687 680.921 113.334 676.751 Q117.949 672.55 125.906 672.55 M124.188 678.406 Q119.381 678.47 116.517 681.112 Q113.652 683.722 113.652 688.05 Q113.652 692.952 116.421 695.912 Q119.19 698.84 124.219 699.286 L124.188 678.406 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M110.915 637.284 L116.39 637.284 Q115.021 639.766 114.353 642.281 Q113.652 644.763 113.652 647.31 Q113.652 653.007 117.281 656.158 Q120.877 659.309 127.402 659.309 Q133.927 659.309 137.556 656.158 Q141.152 653.007 141.152 647.31 Q141.152 644.763 140.484 642.281 Q139.784 639.766 138.415 637.284 L143.826 637.284 Q144.972 639.735 145.545 642.376 Q146.117 644.986 146.117 647.946 Q146.117 655.999 141.057 660.741 Q135.996 665.484 127.402 665.484 Q118.681 665.484 113.684 660.71 Q108.687 655.903 108.687 647.564 Q108.687 644.859 109.26 642.281 Q109.801 639.703 110.915 637.284 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M99.425 621.306 L109.546 621.306 L109.546 609.243 L114.098 609.243 L114.098 621.306 L133.45 621.306 Q137.81 621.306 139.051 620.128 Q140.293 618.919 140.293 615.258 L140.293 609.243 L145.194 609.243 L145.194 615.258 Q145.194 622.038 142.68 624.616 Q140.134 627.194 133.45 627.194 L114.098 627.194 L114.098 631.491 L109.546 631.491 L109.546 627.194 L99.425 627.194 L99.425 621.306 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M125.906 571.049 L128.771 571.049 L128.771 597.976 Q134.818 597.594 138.001 594.347 Q141.152 591.069 141.152 585.244 Q141.152 581.87 140.325 578.719 Q139.497 575.536 137.842 572.417 L143.38 572.417 Q144.717 575.568 145.417 578.878 Q146.117 582.189 146.117 585.594 Q146.117 594.124 141.152 599.121 Q136.187 604.087 127.721 604.087 Q118.968 604.087 113.843 599.376 Q108.687 594.634 108.687 586.613 Q108.687 579.42 113.334 575.25 Q117.949 571.049 125.906 571.049 M124.188 576.905 Q119.381 576.969 116.517 579.61 Q113.652 582.22 113.652 586.549 Q113.652 591.451 116.421 594.411 Q119.19 597.339 124.219 597.785 L124.188 576.905 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M114.957 537.979 L95.6692 537.979 L95.6692 532.122 L145.194 532.122 L145.194 537.979 L139.847 537.979 Q143.03 539.825 144.59 542.658 Q146.117 545.458 146.117 549.405 Q146.117 555.866 140.961 559.94 Q135.805 563.983 127.402 563.983 Q119 563.983 113.843 559.94 Q108.687 555.866 108.687 549.405 Q108.687 545.458 110.247 542.658 Q111.774 539.825 114.957 537.979 M127.402 557.935 Q133.863 557.935 137.556 555.293 Q141.216 552.62 141.216 547.973 Q141.216 543.326 137.556 540.652 Q133.863 537.979 127.402 537.979 Q120.941 537.979 117.281 540.652 Q113.589 543.326 113.589 547.973 Q113.589 552.62 117.281 555.293 Q120.941 557.935 127.402 557.935 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M122.914 476.55 Q123.615 474.481 125.906 472.539 Q128.198 470.566 132.208 468.593 L145.194 462.068 L145.194 468.974 L133.004 475.054 Q128.23 477.409 126.67 479.637 Q125.111 481.833 125.111 485.653 L125.111 492.655 L145.194 492.655 L145.194 499.084 L97.6744 499.084 L97.6744 484.57 Q97.6744 476.422 101.08 472.412 Q104.486 468.402 111.361 468.402 Q115.848 468.402 118.809 470.502 Q121.769 472.571 122.914 476.55 M102.958 492.655 L119.827 492.655 L119.827 484.57 Q119.827 479.924 117.695 477.568 Q115.53 475.181 111.361 475.181 Q107.191 475.181 105.09 477.568 Q102.958 479.924 102.958 484.57 L102.958 492.655 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M125.906 426.483 L128.771 426.483 L128.771 453.41 Q134.818 453.028 138.001 449.782 Q141.152 446.504 141.152 440.679 Q141.152 437.305 140.325 434.154 Q139.497 430.971 137.842 427.852 L143.38 427.852 Q144.717 431.003 145.417 434.313 Q146.117 437.623 146.117 441.029 Q146.117 449.559 141.152 454.556 Q136.187 459.521 127.721 459.521 Q118.968 459.521 113.843 454.811 Q108.687 450.068 108.687 442.048 Q108.687 434.854 113.334 430.685 Q117.949 426.483 125.906 426.483 M124.188 432.34 Q119.381 432.404 116.517 435.045 Q113.652 437.655 113.652 441.984 Q113.652 446.885 116.421 449.846 Q119.19 452.774 124.219 453.219 L124.188 432.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M99.425 411.078 L109.546 411.078 L109.546 399.015 L114.098 399.015 L114.098 411.078 L133.45 411.078 Q137.81 411.078 139.051 409.901 Q140.293 408.691 140.293 405.031 L140.293 399.015 L145.194 399.015 L145.194 405.031 Q145.194 411.81 142.68 414.389 Q140.134 416.967 133.45 416.967 L114.098 416.967 L114.098 421.264 L109.546 421.264 L109.546 416.967 L99.425 416.967 L99.425 411.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M131.126 391.918 L109.546 391.918 L109.546 386.061 L130.903 386.061 Q135.964 386.061 138.51 384.088 Q141.025 382.114 141.025 378.168 Q141.025 373.425 138.001 370.688 Q134.977 367.919 129.758 367.919 L109.546 367.919 L109.546 362.062 L145.194 362.062 L145.194 367.919 L139.72 367.919 Q142.966 370.051 144.558 372.884 Q146.117 375.685 146.117 379.409 Q146.117 385.552 142.298 388.735 Q138.479 391.918 131.126 391.918 M108.687 377.181 L108.687 377.181 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M115.021 329.343 Q114.448 330.329 114.193 331.507 Q113.907 332.653 113.907 334.053 Q113.907 339.019 117.153 341.692 Q120.368 344.334 126.416 344.334 L145.194 344.334 L145.194 350.222 L109.546 350.222 L109.546 344.334 L115.085 344.334 Q111.838 342.488 110.279 339.528 Q108.687 336.568 108.687 332.335 Q108.687 331.73 108.783 330.998 Q108.846 330.266 109.005 329.375 L115.021 329.343 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M123.678 294.713 L145.194 294.713 L145.194 300.57 L123.869 300.57 Q118.809 300.57 116.294 302.543 Q113.78 304.516 113.78 308.463 Q113.78 313.206 116.803 315.943 Q119.827 318.68 125.047 318.68 L145.194 318.68 L145.194 324.568 L109.546 324.568 L109.546 318.68 L115.085 318.68 Q111.87 316.579 110.279 313.747 Q108.687 310.882 108.687 307.158 Q108.687 301.015 112.507 297.864 Q116.294 294.713 123.678 294.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1020.54 17.6457 Q1011.62 17.6457 1006.36 24.2892 Q1001.13 30.9327 1001.13 42.3968 Q1001.13 53.8203 1006.36 60.4638 Q1011.62 67.1073 1020.54 67.1073 Q1029.45 67.1073 1034.63 60.4638 Q1039.86 53.8203 1039.86 42.3968 Q1039.86 30.9327 1034.63 24.2892 Q1029.45 17.6457 1020.54 17.6457 M1020.54 11.0023 Q1033.26 11.0023 1040.87 19.5497 Q1048.49 28.0566 1048.49 42.3968 Q1048.49 56.6965 1040.87 65.2439 Q1033.26 73.7508 1020.54 73.7508 Q1007.78 73.7508 1000.12 65.2439 Q992.504 56.737 992.504 42.3968 Q992.504 28.0566 1000.12 19.5497 Q1007.78 11.0023 1020.54 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1093.53 49.9314 Q1093.53 41.7081 1090.13 37.0496 Q1086.77 32.3505 1080.85 32.3505 Q1074.94 32.3505 1071.54 37.0496 Q1068.17 41.7081 1068.17 49.9314 Q1068.17 58.1548 1071.54 62.8538 Q1074.94 67.5124 1080.85 67.5124 Q1086.77 67.5124 1090.13 62.8538 Q1093.53 58.1548 1093.53 49.9314 M1068.17 34.0924 Q1070.52 30.0415 1074.09 28.0971 Q1077.69 26.1121 1082.68 26.1121 Q1090.94 26.1121 1096.09 32.6746 Q1101.27 39.2371 1101.27 49.9314 Q1101.27 60.6258 1096.09 67.1883 Q1090.94 73.7508 1082.68 73.7508 Q1077.69 73.7508 1074.09 71.8063 Q1070.52 69.8214 1068.17 65.7705 L1068.17 72.576 L1060.68 72.576 L1060.68 9.54393 L1068.17 9.54393 L1068.17 34.0924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1113.63 27.2059 L1121.08 27.2059 L1121.08 73.3862 Q1121.08 82.0551 1117.76 85.944 Q1114.48 89.8329 1107.14 89.8329 L1104.31 89.8329 L1104.31 83.5134 L1106.29 83.5134 Q1110.55 83.5134 1112.09 81.5285 Q1113.63 79.5841 1113.63 73.3862 L1113.63 27.2059 M1113.63 9.54393 L1121.08 9.54393 L1121.08 18.9825 L1113.63 18.9825 L1113.63 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1175.48 48.0275 L1175.48 51.6733 L1141.21 51.6733 Q1141.7 59.3701 1145.83 63.421 Q1150 67.4314 1157.42 67.4314 Q1161.71 67.4314 1165.72 66.3781 Q1169.77 65.3249 1173.74 63.2184 L1173.74 70.267 Q1169.73 71.9684 1165.52 72.8596 Q1161.31 73.7508 1156.97 73.7508 Q1146.11 73.7508 1139.75 67.4314 Q1133.44 61.1119 1133.44 50.3365 Q1133.44 39.1965 1139.43 32.6746 Q1145.47 26.1121 1155.67 26.1121 Q1164.83 26.1121 1170.14 32.0264 Q1175.48 37.9003 1175.48 48.0275 M1168.03 45.84 Q1167.95 39.7232 1164.59 36.0774 Q1161.26 32.4315 1155.76 32.4315 Q1149.52 32.4315 1145.75 35.9558 Q1142.02 39.4801 1141.46 45.8805 L1168.03 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1220.37 28.9478 L1220.37 35.9153 Q1217.21 34.1734 1214.01 33.3227 Q1210.85 32.4315 1207.61 32.4315 Q1200.36 32.4315 1196.35 37.0496 Q1192.34 41.6271 1192.34 49.9314 Q1192.34 58.2358 1196.35 62.8538 Q1200.36 67.4314 1207.61 67.4314 Q1210.85 67.4314 1214.01 66.5807 Q1217.21 65.6895 1220.37 63.9476 L1220.37 70.8341 Q1217.25 72.2924 1213.89 73.0216 Q1210.56 73.7508 1206.8 73.7508 Q1196.55 73.7508 1190.51 67.3098 Q1184.48 60.8689 1184.48 49.9314 Q1184.48 38.832 1190.55 32.472 Q1196.67 26.1121 1207.28 26.1121 Q1210.73 26.1121 1214.01 26.8413 Q1217.29 27.5299 1220.37 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1240.7 14.324 L1240.7 27.2059 L1256.06 27.2059 L1256.06 32.9987 L1240.7 32.9987 L1240.7 57.6282 Q1240.7 63.1779 1242.2 64.7578 Q1243.74 66.3376 1248.4 66.3376 L1256.06 66.3376 L1256.06 72.576 L1248.4 72.576 Q1239.77 72.576 1236.49 69.3758 Q1233.21 66.1351 1233.21 57.6282 L1233.21 32.9987 L1227.74 32.9987 L1227.74 27.2059 L1233.21 27.2059 L1233.21 14.324 L1240.7 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1265.86 27.2059 L1273.31 27.2059 L1273.31 72.576 L1265.86 72.576 L1265.86 27.2059 M1265.86 9.54393 L1273.31 9.54393 L1273.31 18.9825 L1265.86 18.9825 L1265.86 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1283.56 27.2059 L1291.46 27.2059 L1305.64 65.2844 L1319.82 27.2059 L1327.72 27.2059 L1310.7 72.576 L1300.58 72.576 L1283.56 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1376.81 48.0275 L1376.81 51.6733 L1342.54 51.6733 Q1343.03 59.3701 1347.16 63.421 Q1351.33 67.4314 1358.75 67.4314 Q1363.04 67.4314 1367.05 66.3781 Q1371.1 65.3249 1375.07 63.2184 L1375.07 70.267 Q1371.06 71.9684 1366.85 72.8596 Q1362.64 73.7508 1358.3 73.7508 Q1347.44 73.7508 1341.08 67.4314 Q1334.76 61.1119 1334.76 50.3365 Q1334.76 39.1965 1340.76 32.6746 Q1346.8 26.1121 1357 26.1121 Q1366.16 26.1121 1371.47 32.0264 Q1376.81 37.9003 1376.81 48.0275 M1369.36 45.84 Q1369.28 39.7232 1365.92 36.0774 Q1362.59 32.4315 1357.09 32.4315 Q1350.85 32.4315 1347.08 35.9558 Q1343.35 39.4801 1342.79 45.8805 L1369.36 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1444.34 28.5427 L1444.34 35.5912 Q1441.18 33.9709 1437.78 33.1607 Q1434.38 32.3505 1430.73 32.3505 Q1425.18 32.3505 1422.39 34.0519 Q1419.63 35.7533 1419.63 39.156 Q1419.63 41.7486 1421.62 43.2475 Q1423.6 44.7058 1429.6 46.0426 L1432.15 46.6097 Q1440.09 48.3111 1443.41 51.4303 Q1446.77 54.509 1446.77 60.0587 Q1446.77 66.3781 1441.75 70.0644 Q1436.77 73.7508 1428.02 73.7508 Q1424.37 73.7508 1420.4 73.0216 Q1416.47 72.3329 1412.1 70.9151 L1412.1 63.2184 Q1416.23 65.3654 1420.24 66.4591 Q1424.25 67.5124 1428.18 67.5124 Q1433.44 67.5124 1436.28 65.73 Q1439.12 63.9071 1439.12 60.6258 Q1439.12 57.5877 1437.05 55.9673 Q1435.02 54.3469 1428.1 52.8481 L1425.51 52.2405 Q1418.58 50.7821 1415.5 47.7845 Q1412.42 44.7463 1412.42 39.4801 Q1412.42 33.0797 1416.96 29.5959 Q1421.49 26.1121 1429.84 26.1121 Q1433.97 26.1121 1437.62 26.7198 Q1441.26 27.3274 1444.34 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1465.85 65.7705 L1465.85 89.8329 L1458.36 89.8329 L1458.36 27.2059 L1465.85 27.2059 L1465.85 34.0924 Q1468.2 30.0415 1471.77 28.0971 Q1475.37 26.1121 1480.35 26.1121 Q1488.62 26.1121 1493.76 32.6746 Q1498.95 39.2371 1498.95 49.9314 Q1498.95 60.6258 1493.76 67.1883 Q1488.62 73.7508 1480.35 73.7508 Q1475.37 73.7508 1471.77 71.8063 Q1468.2 69.8214 1465.85 65.7705 M1491.21 49.9314 Q1491.21 41.7081 1487.81 37.0496 Q1484.45 32.3505 1478.53 32.3505 Q1472.62 32.3505 1469.21 37.0496 Q1465.85 41.7081 1465.85 49.9314 Q1465.85 58.1548 1469.21 62.8538 Q1472.62 67.5124 1478.53 67.5124 Q1484.45 67.5124 1487.81 62.8538 Q1491.21 58.1548 1491.21 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1531.92 49.7694 Q1522.89 49.7694 1519.41 51.8354 Q1515.92 53.9013 1515.92 58.8839 Q1515.92 62.8538 1518.51 65.2034 Q1521.15 67.5124 1525.64 67.5124 Q1531.84 67.5124 1535.57 63.1374 Q1539.34 58.7219 1539.34 51.4303 L1539.34 49.7694 L1531.92 49.7694 M1546.79 46.6907 L1546.79 72.576 L1539.34 72.576 L1539.34 65.6895 Q1536.78 69.8214 1532.98 71.8063 Q1529.17 73.7508 1523.66 73.7508 Q1516.69 73.7508 1512.56 69.8619 Q1508.47 65.9325 1508.47 59.3701 Q1508.47 51.7138 1513.57 47.825 Q1518.72 43.9361 1528.88 43.9361 L1539.34 43.9361 L1539.34 43.2069 Q1539.34 38.0623 1535.93 35.2672 Q1532.57 32.4315 1526.45 32.4315 Q1522.56 32.4315 1518.88 33.3632 Q1515.19 34.295 1511.79 36.1584 L1511.79 29.2718 Q1515.88 27.692 1519.73 26.9223 Q1523.58 26.1121 1527.22 26.1121 Q1537.07 26.1121 1541.93 31.2163 Q1546.79 36.3204 1546.79 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1594.79 28.9478 L1594.79 35.9153 Q1591.63 34.1734 1588.43 33.3227 Q1585.27 32.4315 1582.03 32.4315 Q1574.78 32.4315 1570.77 37.0496 Q1566.76 41.6271 1566.76 49.9314 Q1566.76 58.2358 1570.77 62.8538 Q1574.78 67.4314 1582.03 67.4314 Q1585.27 67.4314 1588.43 66.5807 Q1591.63 65.6895 1594.79 63.9476 L1594.79 70.8341 Q1591.67 72.2924 1588.31 73.0216 Q1584.99 73.7508 1581.22 73.7508 Q1570.97 73.7508 1564.94 67.3098 Q1558.9 60.8689 1558.9 49.9314 Q1558.9 38.832 1564.98 32.472 Q1571.09 26.1121 1581.71 26.1121 Q1585.15 26.1121 1588.43 26.8413 Q1591.71 27.5299 1594.79 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1646.56 48.0275 L1646.56 51.6733 L1612.29 51.6733 Q1612.78 59.3701 1616.91 63.421 Q1621.08 67.4314 1628.5 67.4314 Q1632.79 67.4314 1636.8 66.3781 Q1640.85 65.3249 1644.82 63.2184 L1644.82 70.267 Q1640.81 71.9684 1636.6 72.8596 Q1632.38 73.7508 1628.05 73.7508 Q1617.19 73.7508 1610.83 67.4314 Q1604.51 61.1119 1604.51 50.3365 Q1604.51 39.1965 1610.51 32.6746 Q1616.55 26.1121 1626.75 26.1121 Q1635.91 26.1121 1641.22 32.0264 Q1646.56 37.9003 1646.56 48.0275 M1639.11 45.84 Q1639.03 39.7232 1635.67 36.0774 Q1632.34 32.4315 1626.84 32.4315 Q1620.6 32.4315 1616.83 35.9558 Q1613.1 39.4801 1612.54 45.8805 L1639.11 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip092)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="-1780.13,422.883 4419.2,422.883 "/>
-<polyline clip-path="url(#clip092)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="987.246,1776.14 987.246,-657.503 "/>
-<circle clip-path="url(#clip092)" cx="344.796" cy="941.969" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="384.581" cy="812.114" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="424.366" cy="758.327" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="464.151" cy="717.054" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="503.936" cy="682.259" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="543.722" cy="651.605" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="583.507" cy="623.891" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="623.292" cy="598.406" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="663.077" cy="574.684" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="702.863" cy="552.405" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="742.648" cy="531.332" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="782.433" cy="511.289" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="822.218" cy="492.139" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="862.003" cy="473.771" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="901.789" cy="456.097" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="941.574" cy="439.599" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="981.359" cy="424.93" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1021.14" cy="411.608" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1060.93" cy="399.319" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1100.71" cy="387.854" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1140.5" cy="377.067" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1180.29" cy="366.85" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1220.07" cy="357.121" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1259.86" cy="347.817" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1299.64" cy="338.886" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1339.43" cy="330.286" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1379.21" cy="321.984" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1419" cy="313.95" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1458.78" cy="306.16" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1498.57" cy="298.592" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1538.35" cy="291.23" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1578.14" cy="284.057" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1617.92" cy="277.059" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1657.71" cy="270.224" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1697.49" cy="263.541" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1737.28" cy="257" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1777.06" cy="250.593" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1816.85" cy="244.312" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1856.63" cy="238.149" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1896.42" cy="232.098" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1936.2" cy="226.154" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="1975.99" cy="220.311" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2015.78" cy="214.563" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2055.56" cy="208.907" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2095.35" cy="203.338" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2135.13" cy="197.852" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2174.92" cy="192.445" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2214.7" cy="187.115" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2254.49" cy="181.858" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<circle clip-path="url(#clip092)" cx="2294.27" cy="176.671" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
-<path clip-path="url(#clip090)" d="M1501.38 937.887 L2283.87 937.887 L2283.87 834.207 L1501.38 834.207  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1501.38,937.887 2283.87,937.887 2283.87,834.207 1501.38,834.207 1501.38,937.887 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1524.34,886.047 1662.1,886.047 "/>
-<path clip-path="url(#clip090)" d="M1707.31 869.901 L1707.31 874.461 Q1704.64 873.188 1702.28 872.563 Q1699.92 871.938 1697.72 871.938 Q1693.9 871.938 1691.82 873.42 Q1689.76 874.901 1689.76 877.633 Q1689.76 879.924 1691.13 881.105 Q1692.51 882.262 1696.36 882.98 L1699.18 883.559 Q1704.41 884.554 1706.89 887.077 Q1709.39 889.577 1709.39 893.79 Q1709.39 898.813 1706.01 901.406 Q1702.65 903.998 1696.15 903.998 Q1693.69 903.998 1690.92 903.443 Q1688.16 902.887 1685.2 901.799 L1685.2 896.984 Q1688.05 898.582 1690.78 899.392 Q1693.51 900.202 1696.15 900.202 Q1700.15 900.202 1702.33 898.628 Q1704.5 897.054 1704.5 894.137 Q1704.5 891.591 1702.93 890.156 Q1701.38 888.721 1697.82 888.003 L1694.97 887.447 Q1689.74 886.406 1687.4 884.184 Q1685.06 881.961 1685.06 878.003 Q1685.06 873.42 1688.28 870.781 Q1691.52 868.142 1697.19 868.142 Q1699.62 868.142 1702.14 868.582 Q1704.67 869.022 1707.31 869.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1716.5 877.401 L1720.75 877.401 L1720.75 903.327 L1716.5 903.327 L1716.5 877.401 M1716.5 867.309 L1720.75 867.309 L1720.75 872.702 L1716.5 872.702 L1716.5 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1751.22 887.679 L1751.22 903.327 L1746.96 903.327 L1746.96 887.818 Q1746.96 884.137 1745.52 882.309 Q1744.09 880.48 1741.22 880.48 Q1737.77 880.48 1735.78 882.679 Q1733.79 884.878 1733.79 888.674 L1733.79 903.327 L1729.5 903.327 L1729.5 877.401 L1733.79 877.401 L1733.79 881.429 Q1735.31 879.091 1737.38 877.934 Q1739.46 876.776 1742.17 876.776 Q1746.63 876.776 1748.93 879.554 Q1751.22 882.309 1751.22 887.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1776.77 890.063 Q1776.77 885.434 1774.85 882.887 Q1772.95 880.341 1769.5 880.341 Q1766.08 880.341 1764.16 882.887 Q1762.26 885.434 1762.26 890.063 Q1762.26 894.67 1764.16 897.216 Q1766.08 899.762 1769.5 899.762 Q1772.95 899.762 1774.85 897.216 Q1776.77 894.67 1776.77 890.063 M1781.03 900.109 Q1781.03 906.73 1778.09 909.947 Q1775.15 913.188 1769.09 913.188 Q1766.84 913.188 1764.85 912.841 Q1762.86 912.517 1760.99 911.822 L1760.99 907.679 Q1762.86 908.697 1764.69 909.183 Q1766.52 909.669 1768.42 909.669 Q1772.61 909.669 1774.69 907.47 Q1776.77 905.295 1776.77 900.873 L1776.77 898.767 Q1775.45 901.058 1773.39 902.193 Q1771.33 903.327 1768.46 903.327 Q1763.69 903.327 1760.78 899.693 Q1757.86 896.058 1757.86 890.063 Q1757.86 884.045 1760.78 880.41 Q1763.69 876.776 1768.46 876.776 Q1771.33 876.776 1773.39 877.91 Q1775.45 879.045 1776.77 881.336 L1776.77 877.401 L1781.03 877.401 L1781.03 900.109 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1789.81 867.309 L1794.06 867.309 L1794.06 903.327 L1789.81 903.327 L1789.81 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1825.15 889.299 L1825.15 891.383 L1805.57 891.383 Q1805.85 895.781 1808.21 898.095 Q1810.59 900.387 1814.83 900.387 Q1817.28 900.387 1819.57 899.785 Q1821.89 899.183 1824.16 897.98 L1824.16 902.007 Q1821.87 902.98 1819.46 903.489 Q1817.05 903.998 1814.57 903.998 Q1808.37 903.998 1804.74 900.387 Q1801.12 896.776 1801.12 890.619 Q1801.12 884.253 1804.55 880.526 Q1808 876.776 1813.83 876.776 Q1819.06 876.776 1822.1 880.156 Q1825.15 883.512 1825.15 889.299 M1820.89 888.049 Q1820.85 884.554 1818.93 882.471 Q1817.03 880.387 1813.88 880.387 Q1810.31 880.387 1808.16 882.401 Q1806.03 884.415 1805.71 888.072 L1820.89 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1829.99 888.443 L1842.47 888.443 L1842.47 892.239 L1829.99 892.239 L1829.99 888.443 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1860.18 880.387 Q1856.75 880.387 1854.76 883.072 Q1852.77 885.734 1852.77 890.387 Q1852.77 895.04 1854.74 897.725 Q1856.73 900.387 1860.18 900.387 Q1863.58 900.387 1865.57 897.702 Q1867.56 895.017 1867.56 890.387 Q1867.56 885.781 1865.57 883.096 Q1863.58 880.387 1860.18 880.387 M1860.18 876.776 Q1865.73 876.776 1868.9 880.387 Q1872.07 883.998 1872.07 890.387 Q1872.07 896.753 1868.9 900.387 Q1865.73 903.998 1860.18 903.998 Q1854.6 903.998 1851.43 900.387 Q1848.28 896.753 1848.28 890.387 Q1848.28 883.998 1851.43 880.387 Q1854.6 876.776 1860.18 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1897.74 890.387 Q1897.74 885.688 1895.8 883.026 Q1893.88 880.341 1890.5 880.341 Q1887.12 880.341 1885.18 883.026 Q1883.25 885.688 1883.25 890.387 Q1883.25 895.086 1885.18 897.771 Q1887.12 900.433 1890.5 900.433 Q1893.88 900.433 1895.8 897.771 Q1897.74 895.086 1897.74 890.387 M1883.25 881.336 Q1884.6 879.022 1886.63 877.91 Q1888.69 876.776 1891.54 876.776 Q1896.26 876.776 1899.2 880.526 Q1902.17 884.276 1902.17 890.387 Q1902.17 896.498 1899.2 900.248 Q1896.26 903.998 1891.54 903.998 Q1888.69 903.998 1886.63 902.887 Q1884.6 901.753 1883.25 899.438 L1883.25 903.327 L1878.97 903.327 L1878.97 867.309 L1883.25 867.309 L1883.25 881.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1909.23 877.401 L1913.49 877.401 L1913.49 903.79 Q1913.49 908.744 1911.59 910.966 Q1909.71 913.188 1905.52 913.188 L1903.9 913.188 L1903.9 909.577 L1905.04 909.577 Q1907.47 909.577 1908.35 908.443 Q1909.23 907.332 1909.23 903.79 L1909.23 877.401 M1909.23 867.309 L1913.49 867.309 L1913.49 872.702 L1909.23 872.702 L1909.23 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1944.57 889.299 L1944.57 891.383 L1924.99 891.383 Q1925.27 895.781 1927.63 898.095 Q1930.01 900.387 1934.25 900.387 Q1936.7 900.387 1938.99 899.785 Q1941.31 899.183 1943.58 897.98 L1943.58 902.007 Q1941.29 902.98 1938.88 903.489 Q1936.47 903.998 1933.99 903.998 Q1927.79 903.998 1924.16 900.387 Q1920.55 896.776 1920.55 890.619 Q1920.55 884.253 1923.97 880.526 Q1927.42 876.776 1933.25 876.776 Q1938.49 876.776 1941.52 880.156 Q1944.57 883.512 1944.57 889.299 M1940.31 888.049 Q1940.27 884.554 1938.35 882.471 Q1936.45 880.387 1933.3 880.387 Q1929.74 880.387 1927.58 882.401 Q1925.45 884.415 1925.13 888.072 L1940.31 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1970.22 878.397 L1970.22 882.378 Q1968.42 881.383 1966.59 880.896 Q1964.78 880.387 1962.93 880.387 Q1958.79 880.387 1956.49 883.026 Q1954.2 885.642 1954.2 890.387 Q1954.2 895.133 1956.49 897.771 Q1958.79 900.387 1962.93 900.387 Q1964.78 900.387 1966.59 899.901 Q1968.42 899.392 1970.22 898.396 L1970.22 902.332 Q1968.44 903.165 1966.52 903.582 Q1964.62 903.998 1962.47 903.998 Q1956.61 903.998 1953.16 900.318 Q1949.71 896.637 1949.71 890.387 Q1949.71 884.045 1953.18 880.41 Q1956.68 876.776 1962.74 876.776 Q1964.71 876.776 1966.59 877.193 Q1968.46 877.586 1970.22 878.397 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1981.84 870.04 L1981.84 877.401 L1990.61 877.401 L1990.61 880.711 L1981.84 880.711 L1981.84 894.785 Q1981.84 897.957 1982.7 898.859 Q1983.58 899.762 1986.24 899.762 L1990.61 899.762 L1990.61 903.327 L1986.24 903.327 Q1981.31 903.327 1979.43 901.498 Q1977.56 899.646 1977.56 894.785 L1977.56 880.711 L1974.43 880.711 L1974.43 877.401 L1977.56 877.401 L1977.56 870.04 L1981.84 870.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1996.22 877.401 L2000.48 877.401 L2000.48 903.327 L1996.22 903.327 L1996.22 877.401 M1996.22 867.309 L2000.48 867.309 L2000.48 872.702 L1996.22 872.702 L1996.22 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2006.33 877.401 L2010.85 877.401 L2018.95 899.16 L2027.05 877.401 L2031.56 877.401 L2021.84 903.327 L2016.05 903.327 L2006.33 877.401 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2059.62 889.299 L2059.62 891.383 L2040.04 891.383 Q2040.31 895.781 2042.67 898.095 Q2045.06 900.387 2049.29 900.387 Q2051.75 900.387 2054.04 899.785 Q2056.35 899.183 2058.62 897.98 L2058.62 902.007 Q2056.33 902.98 2053.92 903.489 Q2051.52 903.998 2049.04 903.998 Q2042.84 903.998 2039.2 900.387 Q2035.59 896.776 2035.59 890.619 Q2035.59 884.253 2039.02 880.526 Q2042.47 876.776 2048.3 876.776 Q2053.53 876.776 2056.56 880.156 Q2059.62 883.512 2059.62 889.299 M2055.36 888.049 Q2055.31 884.554 2053.39 882.471 Q2051.49 880.387 2048.35 880.387 Q2044.78 880.387 2042.63 882.401 Q2040.5 884.415 2040.17 888.072 L2055.36 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2098.21 878.165 L2098.21 882.193 Q2096.4 881.267 2094.46 880.804 Q2092.51 880.341 2090.43 880.341 Q2087.26 880.341 2085.66 881.313 Q2084.09 882.285 2084.09 884.23 Q2084.09 885.711 2085.22 886.568 Q2086.35 887.401 2089.78 888.165 L2091.24 888.489 Q2095.78 889.461 2097.67 891.244 Q2099.6 893.003 2099.6 896.174 Q2099.6 899.785 2096.72 901.892 Q2093.88 903.998 2088.88 903.998 Q2086.79 903.998 2084.53 903.582 Q2082.28 903.188 2079.78 902.378 L2079.78 897.98 Q2082.14 899.207 2084.43 899.832 Q2086.73 900.433 2088.97 900.433 Q2091.98 900.433 2093.6 899.415 Q2095.22 898.373 2095.22 896.498 Q2095.22 894.762 2094.04 893.836 Q2092.88 892.91 2088.92 892.054 L2087.44 891.707 Q2083.48 890.873 2081.73 889.16 Q2079.97 887.424 2079.97 884.415 Q2079.97 880.758 2082.56 878.767 Q2085.15 876.776 2089.92 876.776 Q2092.28 876.776 2094.36 877.123 Q2096.45 877.471 2098.21 878.165 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2116.42 880.387 Q2113 880.387 2111.01 883.072 Q2109.02 885.734 2109.02 890.387 Q2109.02 895.04 2110.98 897.725 Q2112.97 900.387 2116.42 900.387 Q2119.83 900.387 2121.82 897.702 Q2123.81 895.017 2123.81 890.387 Q2123.81 885.781 2121.82 883.096 Q2119.83 880.387 2116.42 880.387 M2116.42 876.776 Q2121.98 876.776 2125.15 880.387 Q2128.32 883.998 2128.32 890.387 Q2128.32 896.753 2125.15 900.387 Q2121.98 903.998 2116.42 903.998 Q2110.85 903.998 2107.67 900.387 Q2104.53 896.753 2104.53 890.387 Q2104.53 883.998 2107.67 880.387 Q2110.85 876.776 2116.42 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2135.38 867.309 L2139.64 867.309 L2139.64 903.327 L2135.38 903.327 L2135.38 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2148.11 893.096 L2148.11 877.401 L2152.37 877.401 L2152.37 892.933 Q2152.37 896.614 2153.81 898.466 Q2155.24 900.295 2158.11 900.295 Q2161.56 900.295 2163.55 898.095 Q2165.57 895.896 2165.57 892.1 L2165.57 877.401 L2169.83 877.401 L2169.83 903.327 L2165.57 903.327 L2165.57 899.345 Q2164.02 901.707 2161.96 902.864 Q2159.92 903.998 2157.21 903.998 Q2152.74 903.998 2150.43 901.22 Q2148.11 898.443 2148.11 893.096 M2158.83 876.776 L2158.83 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2182.81 870.04 L2182.81 877.401 L2191.59 877.401 L2191.59 880.711 L2182.81 880.711 L2182.81 894.785 Q2182.81 897.957 2183.67 898.859 Q2184.55 899.762 2187.21 899.762 L2191.59 899.762 L2191.59 903.327 L2187.21 903.327 Q2182.28 903.327 2180.41 901.498 Q2178.53 899.646 2178.53 894.785 L2178.53 880.711 L2175.41 880.711 L2175.41 877.401 L2178.53 877.401 L2178.53 870.04 L2182.81 870.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2197.19 877.401 L2201.45 877.401 L2201.45 903.327 L2197.19 903.327 L2197.19 877.401 M2197.19 867.309 L2201.45 867.309 L2201.45 872.702 L2197.19 872.702 L2197.19 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2220.4 880.387 Q2216.98 880.387 2214.99 883.072 Q2213 885.734 2213 890.387 Q2213 895.04 2214.97 897.725 Q2216.96 900.387 2220.4 900.387 Q2223.81 900.387 2225.8 897.702 Q2227.79 895.017 2227.79 890.387 Q2227.79 885.781 2225.8 883.096 Q2223.81 880.387 2220.4 880.387 M2220.4 876.776 Q2225.96 876.776 2229.13 880.387 Q2232.3 883.998 2232.3 890.387 Q2232.3 896.753 2229.13 900.387 Q2225.96 903.998 2220.4 903.998 Q2214.83 903.998 2211.65 900.387 Q2208.51 896.753 2208.51 890.387 Q2208.51 883.998 2211.65 880.387 Q2214.83 876.776 2220.4 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2260.91 887.679 L2260.91 903.327 L2256.65 903.327 L2256.65 887.818 Q2256.65 884.137 2255.22 882.309 Q2253.78 880.48 2250.91 880.48 Q2247.46 880.48 2245.47 882.679 Q2243.48 884.878 2243.48 888.674 L2243.48 903.327 L2239.2 903.327 L2239.2 877.401 L2243.48 877.401 L2243.48 881.429 Q2245.01 879.091 2247.07 877.934 Q2249.15 876.776 2251.86 876.776 Q2256.33 876.776 2258.62 879.554 Q2260.91 882.309 2260.91 887.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M341.189 946.973 L344.053 946.973 L344.053 937.086 L340.937 937.711 L340.937 936.114 L344.036 935.489 L345.789 935.489 L345.789 946.973 L348.654 946.973 L348.654 948.449 L341.189 948.449 L341.189 946.973 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M382.576 817.119 L388.695 817.119 L388.695 818.594 L380.466 818.594 L380.466 817.119 Q381.464 816.086 383.183 814.349 Q384.911 812.605 385.353 812.101 Q386.195 811.155 386.525 810.504 Q386.864 809.844 386.864 809.211 Q386.864 808.178 386.135 807.527 Q385.414 806.876 384.251 806.876 Q383.426 806.876 382.506 807.162 Q381.595 807.448 380.553 808.03 L380.553 806.259 Q381.612 805.834 382.532 805.617 Q383.452 805.4 384.216 805.4 Q386.23 805.4 387.428 806.407 Q388.626 807.414 388.626 809.098 Q388.626 809.896 388.322 810.617 Q388.027 811.329 387.237 812.301 Q387.02 812.553 385.857 813.759 Q384.694 814.957 382.576 817.119 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M425.959 757.819 Q427.218 758.088 427.921 758.939 Q428.632 759.789 428.632 761.039 Q428.632 762.958 427.313 764.008 Q425.994 765.058 423.563 765.058 Q422.747 765.058 421.879 764.893 Q421.02 764.737 420.099 764.416 L420.099 762.723 Q420.829 763.149 421.697 763.366 Q422.565 763.583 423.511 763.583 Q425.16 763.583 426.02 762.932 Q426.888 762.281 426.888 761.039 Q426.888 759.893 426.08 759.251 Q425.282 758.6 423.849 758.6 L422.339 758.6 L422.339 757.159 L423.919 757.159 Q425.212 757.159 425.898 756.647 Q426.584 756.126 426.584 755.154 Q426.584 754.156 425.872 753.626 Q425.169 753.088 423.849 753.088 Q423.129 753.088 422.304 753.244 Q421.48 753.4 420.49 753.73 L420.49 752.168 Q421.488 751.89 422.356 751.751 Q423.233 751.612 424.006 751.612 Q426.002 751.612 427.165 752.524 Q428.329 753.426 428.329 754.972 Q428.329 756.048 427.712 756.795 Q427.096 757.532 425.959 757.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M465.28 712.102 L460.853 719.02 L465.28 719.02 L465.28 712.102 M464.82 710.574 L467.024 710.574 L467.024 719.02 L468.873 719.02 L468.873 720.478 L467.024 720.478 L467.024 723.534 L465.28 723.534 L465.28 720.478 L459.429 720.478 L459.429 718.786 L464.82 710.574 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M500.291 675.779 L507.174 675.779 L507.174 677.255 L501.896 677.255 L501.896 680.432 Q502.278 680.302 502.66 680.241 Q503.042 680.172 503.424 680.172 Q505.594 680.172 506.862 681.361 Q508.129 682.55 508.129 684.581 Q508.129 686.673 506.827 687.837 Q505.525 688.991 503.155 688.991 Q502.339 688.991 501.489 688.852 Q500.647 688.713 499.744 688.436 L499.744 686.673 Q500.525 687.099 501.358 687.307 Q502.192 687.515 503.12 687.515 Q504.622 687.515 505.499 686.726 Q506.376 685.936 506.376 684.581 Q506.376 683.227 505.499 682.437 Q504.622 681.647 503.12 681.647 Q502.417 681.647 501.714 681.804 Q501.02 681.96 500.291 682.29 L500.291 675.779 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M543.874 650.906 Q542.693 650.906 541.999 651.713 Q541.313 652.521 541.313 653.927 Q541.313 655.324 541.999 656.14 Q542.693 656.948 543.874 656.948 Q545.054 656.948 545.74 656.14 Q546.434 655.324 546.434 653.927 Q546.434 652.521 545.74 651.713 Q545.054 650.906 543.874 650.906 M547.354 645.411 L547.354 647.009 Q546.695 646.696 546.018 646.531 Q545.349 646.366 544.69 646.366 Q542.953 646.366 542.033 647.538 Q541.122 648.71 540.992 651.08 Q541.504 650.324 542.276 649.925 Q543.049 649.517 543.978 649.517 Q545.931 649.517 547.059 650.706 Q548.196 651.887 548.196 653.927 Q548.196 655.923 547.016 657.13 Q545.835 658.337 543.874 658.337 Q541.625 658.337 540.436 656.618 Q539.247 654.89 539.247 651.618 Q539.247 648.545 540.705 646.722 Q542.163 644.89 544.62 644.89 Q545.28 644.89 545.948 645.021 Q546.625 645.151 547.354 645.411 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M579.34 617.411 L587.674 617.411 L587.674 618.158 L582.969 630.371 L581.137 630.371 L585.564 618.887 L579.34 618.887 L579.34 617.411 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M623.292 598.731 Q622.042 598.731 621.322 599.399 Q620.61 600.068 620.61 601.24 Q620.61 602.412 621.322 603.08 Q622.042 603.748 623.292 603.748 Q624.542 603.748 625.263 603.08 Q625.983 602.403 625.983 601.24 Q625.983 600.068 625.263 599.399 Q624.551 598.731 623.292 598.731 M621.539 597.985 Q620.41 597.707 619.776 596.934 Q619.151 596.162 619.151 595.051 Q619.151 593.497 620.254 592.594 Q621.365 591.691 623.292 591.691 Q625.228 591.691 626.33 592.594 Q627.433 593.497 627.433 595.051 Q627.433 596.162 626.799 596.934 Q626.174 597.707 625.054 597.985 Q626.322 598.28 627.025 599.139 Q627.737 599.998 627.737 601.24 Q627.737 603.123 626.582 604.13 Q625.436 605.137 623.292 605.137 Q621.148 605.137 619.994 604.13 Q618.848 603.123 618.848 601.24 Q618.848 599.998 619.559 599.139 Q620.271 598.28 621.539 597.985 M620.896 595.215 Q620.896 596.222 621.521 596.787 Q622.155 597.351 623.292 597.351 Q624.421 597.351 625.054 596.787 Q625.697 596.222 625.697 595.215 Q625.697 594.208 625.054 593.644 Q624.421 593.08 623.292 593.08 Q622.155 593.08 621.521 593.644 Q620.896 594.208 620.896 595.215 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M659.44 580.895 L659.44 579.298 Q660.1 579.61 660.777 579.775 Q661.454 579.94 662.105 579.94 Q663.841 579.94 664.753 578.777 Q665.673 577.605 665.803 575.227 Q665.3 575.973 664.527 576.373 Q663.754 576.772 662.817 576.772 Q660.872 576.772 659.735 575.6 Q658.607 574.419 658.607 572.38 Q658.607 570.383 659.787 569.176 Q660.968 567.97 662.93 567.97 Q665.178 567.97 666.359 569.697 Q667.548 571.416 667.548 574.697 Q667.548 577.761 666.089 579.593 Q664.64 581.416 662.183 581.416 Q661.524 581.416 660.846 581.286 Q660.169 581.156 659.44 580.895 M662.93 575.4 Q664.11 575.4 664.796 574.593 Q665.491 573.786 665.491 572.38 Q665.491 570.982 664.796 570.175 Q664.11 569.359 662.93 569.359 Q661.749 569.359 661.055 570.175 Q660.369 570.982 660.369 572.38 Q660.369 573.786 661.055 574.593 Q661.749 575.4 662.93 575.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M693.37 557.409 L696.235 557.409 L696.235 547.522 L693.119 548.147 L693.119 546.55 L696.218 545.925 L697.971 545.925 L697.971 557.409 L700.836 557.409 L700.836 558.885 L693.37 558.885 L693.37 557.409 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M708.127 547.079 Q706.773 547.079 706.087 548.416 Q705.41 549.744 705.41 552.418 Q705.41 555.083 706.087 556.419 Q706.773 557.747 708.127 557.747 Q709.49 557.747 710.167 556.419 Q710.853 555.083 710.853 552.418 Q710.853 549.744 710.167 548.416 Q709.49 547.079 708.127 547.079 M708.127 545.69 Q710.306 545.69 711.452 547.418 Q712.606 549.136 712.606 552.418 Q712.606 555.69 711.452 557.418 Q710.306 559.136 708.127 559.136 Q705.948 559.136 704.794 557.418 Q703.648 555.69 703.648 552.418 Q703.648 549.136 704.794 547.418 Q705.948 545.69 708.127 545.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M733.386 536.336 L736.25 536.336 L736.25 526.449 L733.134 527.074 L733.134 525.477 L736.233 524.852 L737.986 524.852 L737.986 536.336 L740.851 536.336 L740.851 537.812 L733.386 537.812 L733.386 536.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M744.696 536.336 L747.561 536.336 L747.561 526.449 L744.445 527.074 L744.445 525.477 L747.544 524.852 L749.297 524.852 L749.297 536.336 L752.162 536.336 L752.162 537.812 L744.696 537.812 L744.696 536.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M773.24 516.294 L776.105 516.294 L776.105 506.407 L772.989 507.032 L772.989 505.434 L776.088 504.809 L777.841 504.809 L777.841 516.294 L780.706 516.294 L780.706 517.769 L773.24 517.769 L773.24 516.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M785.758 516.294 L791.877 516.294 L791.877 517.769 L783.648 517.769 L783.648 516.294 Q784.647 515.261 786.365 513.525 Q788.093 511.78 788.535 511.276 Q789.377 510.33 789.707 509.679 Q790.046 509.019 790.046 508.386 Q790.046 507.353 789.317 506.702 Q788.596 506.051 787.433 506.051 Q786.608 506.051 785.688 506.337 Q784.777 506.624 783.735 507.205 L783.735 505.434 Q784.794 505.009 785.714 504.792 Q786.634 504.575 787.398 504.575 Q789.412 504.575 790.61 505.582 Q791.808 506.589 791.808 508.273 Q791.808 509.071 791.504 509.792 Q791.209 510.504 790.419 511.476 Q790.202 511.728 789.039 512.934 Q787.876 514.132 785.758 516.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M812.848 497.143 L815.712 497.143 L815.712 487.256 L812.596 487.881 L812.596 486.284 L815.695 485.659 L817.448 485.659 L817.448 497.143 L820.313 497.143 L820.313 498.619 L812.848 498.619 L812.848 497.143 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M829.167 491.631 Q830.426 491.9 831.129 492.751 Q831.841 493.601 831.841 494.851 Q831.841 496.77 830.521 497.82 Q829.202 498.871 826.771 498.871 Q825.955 498.871 825.087 498.706 Q824.228 498.549 823.308 498.228 L823.308 496.535 Q824.037 496.961 824.905 497.178 Q825.773 497.395 826.719 497.395 Q828.368 497.395 829.228 496.744 Q830.096 496.093 830.096 494.851 Q830.096 493.706 829.289 493.063 Q828.49 492.412 827.058 492.412 L825.547 492.412 L825.547 490.971 L827.127 490.971 Q828.42 490.971 829.106 490.459 Q829.792 489.938 829.792 488.966 Q829.792 487.968 829.08 487.438 Q828.377 486.9 827.058 486.9 Q826.337 486.9 825.512 487.056 Q824.688 487.213 823.698 487.542 L823.698 485.98 Q824.697 485.702 825.565 485.563 Q826.441 485.424 827.214 485.424 Q829.21 485.424 830.374 486.336 Q831.537 487.239 831.537 488.784 Q831.537 489.86 830.92 490.607 Q830.304 491.345 829.167 491.631 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M852.42 478.775 L855.285 478.775 L855.285 468.888 L852.168 469.513 L852.168 467.916 L855.267 467.291 L857.021 467.291 L857.021 478.775 L859.885 478.775 L859.885 480.251 L852.42 480.251 L852.42 478.775 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M868.245 468.819 L863.818 475.737 L868.245 475.737 L868.245 468.819 M867.785 467.291 L869.99 467.291 L869.99 475.737 L871.838 475.737 L871.838 477.195 L869.99 477.195 L869.99 480.251 L868.245 480.251 L868.245 477.195 L862.394 477.195 L862.394 475.503 L867.785 467.291 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M892.483 461.101 L895.348 461.101 L895.348 451.214 L892.231 451.839 L892.231 450.242 L895.33 449.617 L897.084 449.617 L897.084 461.101 L899.948 461.101 L899.948 462.577 L892.483 462.577 L892.483 461.101 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M903.507 449.617 L910.391 449.617 L910.391 451.093 L905.113 451.093 L905.113 454.27 Q905.495 454.139 905.877 454.079 Q906.259 454.009 906.641 454.009 Q908.811 454.009 910.079 455.198 Q911.346 456.388 911.346 458.419 Q911.346 460.511 910.044 461.674 Q908.742 462.829 906.372 462.829 Q905.556 462.829 904.705 462.69 Q903.863 462.551 902.961 462.273 L902.961 460.511 Q903.742 460.936 904.575 461.145 Q905.408 461.353 906.337 461.353 Q907.839 461.353 908.716 460.563 Q909.592 459.773 909.592 458.419 Q909.592 457.065 908.716 456.275 Q907.839 455.485 906.337 455.485 Q905.634 455.485 904.931 455.641 Q904.237 455.797 903.507 456.127 L903.507 449.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M932.051 444.603 L934.916 444.603 L934.916 434.716 L931.8 435.341 L931.8 433.744 L934.899 433.119 L936.652 433.119 L936.652 444.603 L939.517 444.603 L939.517 446.079 L932.051 446.079 L932.051 444.603 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M947.025 438.9 Q945.845 438.9 945.15 439.707 Q944.465 440.514 944.465 441.921 Q944.465 443.318 945.15 444.134 Q945.845 444.941 947.025 444.941 Q948.206 444.941 948.892 444.134 Q949.586 443.318 949.586 441.921 Q949.586 440.514 948.892 439.707 Q948.206 438.9 947.025 438.9 M950.506 433.405 L950.506 435.002 Q949.846 434.69 949.169 434.525 Q948.501 434.36 947.841 434.36 Q946.105 434.36 945.185 435.532 Q944.274 436.704 944.143 439.073 Q944.655 438.318 945.428 437.919 Q946.201 437.511 947.129 437.511 Q949.083 437.511 950.211 438.7 Q951.348 439.881 951.348 441.921 Q951.348 443.917 950.168 445.124 Q948.987 446.33 947.025 446.33 Q944.777 446.33 943.588 444.612 Q942.399 442.884 942.399 439.612 Q942.399 436.539 943.857 434.716 Q945.315 432.884 947.772 432.884 Q948.432 432.884 949.1 433.014 Q949.777 433.145 950.506 433.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M972.036 429.934 L974.901 429.934 L974.901 420.047 L971.785 420.672 L971.785 419.075 L974.883 418.45 L976.637 418.45 L976.637 429.934 L979.502 429.934 L979.502 431.41 L972.036 431.41 L972.036 429.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M982.6 418.45 L990.934 418.45 L990.934 419.196 L986.229 431.41 L984.397 431.41 L988.824 419.925 L982.6 419.925 L982.6 418.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1011.67 416.612 L1014.53 416.612 L1014.53 406.725 L1011.42 407.35 L1011.42 405.753 L1014.52 405.128 L1016.27 405.128 L1016.27 416.612 L1019.13 416.612 L1019.13 418.088 L1011.67 418.088 L1011.67 416.612 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1026.43 411.934 Q1025.18 411.934 1024.46 412.602 Q1023.74 413.27 1023.74 414.442 Q1023.74 415.614 1024.46 416.283 Q1025.18 416.951 1026.43 416.951 Q1027.68 416.951 1028.4 416.283 Q1029.12 415.606 1029.12 414.442 Q1029.12 413.27 1028.4 412.602 Q1027.69 411.934 1026.43 411.934 M1024.67 411.187 Q1023.54 410.909 1022.91 410.137 Q1022.29 409.364 1022.29 408.253 Q1022.29 406.699 1023.39 405.797 Q1024.5 404.894 1026.43 404.894 Q1028.36 404.894 1029.46 405.797 Q1030.57 406.699 1030.57 408.253 Q1030.57 409.364 1029.93 410.137 Q1029.31 410.909 1028.19 411.187 Q1029.46 411.482 1030.16 412.342 Q1030.87 413.201 1030.87 414.442 Q1030.87 416.326 1029.72 417.333 Q1028.57 418.34 1026.43 418.34 Q1024.28 418.34 1023.13 417.333 Q1021.98 416.326 1021.98 414.442 Q1021.98 413.201 1022.69 412.342 Q1023.41 411.482 1024.67 411.187 M1024.03 408.418 Q1024.03 409.425 1024.66 409.989 Q1025.29 410.553 1026.43 410.553 Q1027.55 410.553 1028.19 409.989 Q1028.83 409.425 1028.83 408.418 Q1028.83 407.411 1028.19 406.847 Q1027.55 406.283 1026.43 406.283 Q1025.29 406.283 1024.66 406.847 Q1024.03 407.411 1024.03 408.418 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1051.47 404.323 L1054.34 404.323 L1054.34 394.436 L1051.22 395.061 L1051.22 393.464 L1054.32 392.839 L1056.07 392.839 L1056.07 404.323 L1058.94 404.323 L1058.94 405.799 L1051.47 405.799 L1051.47 404.323 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1062.53 405.53 L1062.53 403.933 Q1063.19 404.245 1063.87 404.41 Q1064.55 404.575 1065.2 404.575 Q1066.93 404.575 1067.84 403.412 Q1068.76 402.24 1068.89 399.862 Q1068.39 400.608 1067.62 401.007 Q1066.85 401.407 1065.91 401.407 Q1063.96 401.407 1062.83 400.235 Q1061.7 399.054 1061.7 397.014 Q1061.7 395.018 1062.88 393.811 Q1064.06 392.605 1066.02 392.605 Q1068.27 392.605 1069.45 394.332 Q1070.64 396.051 1070.64 399.332 Q1070.64 402.396 1069.18 404.228 Q1067.73 406.051 1065.27 406.051 Q1064.61 406.051 1063.94 405.921 Q1063.26 405.79 1062.53 405.53 M1066.02 400.035 Q1067.2 400.035 1067.89 399.228 Q1068.58 398.421 1068.58 397.014 Q1068.58 395.617 1067.89 394.81 Q1067.2 393.994 1066.02 393.994 Q1064.84 393.994 1064.15 394.81 Q1063.46 395.617 1063.46 397.014 Q1063.46 398.421 1064.15 399.228 Q1064.84 400.035 1066.02 400.035 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1092.75 392.859 L1098.87 392.859 L1098.87 394.334 L1090.65 394.334 L1090.65 392.859 Q1091.64 391.826 1093.36 390.089 Q1095.09 388.345 1095.53 387.841 Q1096.37 386.895 1096.7 386.244 Q1097.04 385.584 1097.04 384.951 Q1097.04 383.918 1096.31 383.267 Q1095.59 382.616 1094.43 382.616 Q1093.61 382.616 1092.69 382.902 Q1091.77 383.188 1090.73 383.77 L1090.73 381.999 Q1091.79 381.574 1092.71 381.357 Q1093.63 381.14 1094.4 381.14 Q1096.41 381.14 1097.61 382.147 Q1098.81 383.154 1098.81 384.838 Q1098.81 385.636 1098.5 386.357 Q1098.21 387.069 1097.42 388.041 Q1097.2 388.293 1096.04 389.499 Q1094.87 390.697 1092.75 392.859 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1106.31 382.529 Q1104.95 382.529 1104.27 383.866 Q1103.59 385.194 1103.59 387.867 Q1103.59 390.532 1104.27 391.869 Q1104.95 393.197 1106.31 393.197 Q1107.67 393.197 1108.34 391.869 Q1109.03 390.532 1109.03 387.867 Q1109.03 385.194 1108.34 383.866 Q1107.67 382.529 1106.31 382.529 M1106.31 381.14 Q1108.48 381.14 1109.63 382.867 Q1110.78 384.586 1110.78 387.867 Q1110.78 391.14 1109.63 392.867 Q1108.48 394.586 1106.31 394.586 Q1104.13 394.586 1102.97 392.867 Q1101.83 391.14 1101.83 387.867 Q1101.83 384.586 1102.97 382.867 Q1104.13 381.14 1106.31 381.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1132.77 382.071 L1138.89 382.071 L1138.89 383.547 L1130.66 383.547 L1130.66 382.071 Q1131.66 381.038 1133.38 379.302 Q1135.11 377.558 1135.55 377.054 Q1136.39 376.108 1136.72 375.457 Q1137.06 374.797 1137.06 374.163 Q1137.06 373.131 1136.33 372.479 Q1135.61 371.828 1134.45 371.828 Q1133.62 371.828 1132.7 372.115 Q1131.79 372.401 1130.75 372.983 L1130.75 371.212 Q1131.81 370.787 1132.73 370.57 Q1133.65 370.353 1134.41 370.353 Q1136.42 370.353 1137.62 371.36 Q1138.82 372.367 1138.82 374.051 Q1138.82 374.849 1138.52 375.57 Q1138.22 376.282 1137.43 377.254 Q1137.21 377.505 1136.05 378.712 Q1134.89 379.91 1132.77 382.071 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1142.87 382.071 L1145.74 382.071 L1145.74 372.184 L1142.62 372.809 L1142.62 371.212 L1145.72 370.587 L1147.47 370.587 L1147.47 382.071 L1150.34 382.071 L1150.34 383.547 L1142.87 383.547 L1142.87 382.071 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1172.62 371.854 L1178.74 371.854 L1178.74 373.33 L1170.52 373.33 L1170.52 371.854 Q1171.51 370.821 1173.23 369.085 Q1174.96 367.341 1175.4 366.837 Q1176.24 365.891 1176.57 365.24 Q1176.91 364.58 1176.91 363.947 Q1176.91 362.914 1176.18 362.262 Q1175.46 361.611 1174.3 361.611 Q1173.48 361.611 1172.56 361.898 Q1171.64 362.184 1170.6 362.766 L1170.6 360.995 Q1171.66 360.57 1172.58 360.353 Q1173.5 360.136 1174.27 360.136 Q1176.28 360.136 1177.48 361.143 Q1178.68 362.15 1178.68 363.834 Q1178.68 364.632 1178.37 365.353 Q1178.08 366.065 1177.29 367.037 Q1177.07 367.289 1175.91 368.495 Q1174.74 369.693 1172.62 371.854 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1183.94 371.854 L1190.06 371.854 L1190.06 373.33 L1181.83 373.33 L1181.83 371.854 Q1182.82 370.821 1184.54 369.085 Q1186.27 367.341 1186.71 366.837 Q1187.56 365.891 1187.89 365.24 Q1188.22 364.58 1188.22 363.947 Q1188.22 362.914 1187.49 362.262 Q1186.77 361.611 1185.61 361.611 Q1184.79 361.611 1183.87 361.898 Q1182.95 362.184 1181.91 362.766 L1181.91 360.995 Q1182.97 360.57 1183.89 360.353 Q1184.81 360.136 1185.58 360.136 Q1187.59 360.136 1188.79 361.143 Q1189.99 362.15 1189.99 363.834 Q1189.99 364.632 1189.68 365.353 Q1189.39 366.065 1188.6 367.037 Q1188.38 367.289 1187.22 368.495 Q1186.05 369.693 1183.94 371.854 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1212.23 362.126 L1218.35 362.126 L1218.35 363.601 L1210.12 363.601 L1210.12 362.126 Q1211.12 361.093 1212.84 359.357 Q1214.57 357.612 1215.01 357.108 Q1215.85 356.162 1216.18 355.511 Q1216.52 354.851 1216.52 354.218 Q1216.52 353.185 1215.79 352.534 Q1215.07 351.883 1213.91 351.883 Q1213.08 351.883 1212.16 352.169 Q1211.25 352.456 1210.21 353.037 L1210.21 351.266 Q1211.27 350.841 1212.19 350.624 Q1213.11 350.407 1213.87 350.407 Q1215.89 350.407 1217.08 351.414 Q1218.28 352.421 1218.28 354.105 Q1218.28 354.904 1217.98 355.624 Q1217.68 356.336 1216.89 357.308 Q1216.68 357.56 1215.51 358.766 Q1214.35 359.964 1212.23 362.126 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1227.34 356.614 Q1228.6 356.883 1229.31 357.733 Q1230.02 358.584 1230.02 359.834 Q1230.02 361.752 1228.7 362.803 Q1227.38 363.853 1224.95 363.853 Q1224.13 363.853 1223.26 363.688 Q1222.41 363.532 1221.49 363.211 L1221.49 361.518 Q1222.21 361.943 1223.08 362.16 Q1223.95 362.377 1224.9 362.377 Q1226.55 362.377 1227.41 361.726 Q1228.27 361.075 1228.27 359.834 Q1228.27 358.688 1227.47 358.046 Q1226.67 357.395 1225.24 357.395 L1223.72 357.395 L1223.72 355.954 L1225.3 355.954 Q1226.6 355.954 1227.28 355.442 Q1227.97 354.921 1227.97 353.949 Q1227.97 352.95 1227.26 352.421 Q1226.55 351.883 1225.24 351.883 Q1224.51 351.883 1223.69 352.039 Q1222.87 352.195 1221.88 352.525 L1221.88 350.963 Q1222.87 350.685 1223.74 350.546 Q1224.62 350.407 1225.39 350.407 Q1227.39 350.407 1228.55 351.319 Q1229.71 352.221 1229.71 353.766 Q1229.71 354.843 1229.1 355.589 Q1228.48 356.327 1227.34 356.614 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1251.8 352.821 L1257.92 352.821 L1257.92 354.297 L1249.7 354.297 L1249.7 352.821 Q1250.69 351.788 1252.41 350.052 Q1254.14 348.307 1254.58 347.804 Q1255.42 346.858 1255.75 346.207 Q1256.09 345.547 1256.09 344.913 Q1256.09 343.88 1255.36 343.229 Q1254.64 342.578 1253.48 342.578 Q1252.66 342.578 1251.74 342.865 Q1250.82 343.151 1249.78 343.733 L1249.78 341.962 Q1250.84 341.537 1251.76 341.32 Q1252.68 341.103 1253.45 341.103 Q1255.46 341.103 1256.66 342.11 Q1257.85 343.117 1257.85 344.801 Q1257.85 345.599 1257.55 346.32 Q1257.26 347.031 1256.47 348.004 Q1256.25 348.255 1255.09 349.462 Q1253.92 350.66 1251.8 352.821 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1266.42 342.865 L1262 349.783 L1266.42 349.783 L1266.42 342.865 M1265.96 341.337 L1268.17 341.337 L1268.17 349.783 L1270.02 349.783 L1270.02 351.241 L1268.17 351.241 L1268.17 354.297 L1266.42 354.297 L1266.42 351.241 L1260.57 351.241 L1260.57 349.549 L1265.96 341.337 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1291.87 343.89 L1297.99 343.89 L1297.99 345.366 L1289.76 345.366 L1289.76 343.89 Q1290.76 342.857 1292.48 341.121 Q1294.2 339.376 1294.65 338.873 Q1295.49 337.927 1295.82 337.276 Q1296.16 336.616 1296.16 335.982 Q1296.16 334.949 1295.43 334.298 Q1294.71 333.647 1293.54 333.647 Q1292.72 333.647 1291.8 333.934 Q1290.89 334.22 1289.84 334.802 L1289.84 333.031 Q1290.9 332.605 1291.82 332.388 Q1292.74 332.171 1293.51 332.171 Q1295.52 332.171 1296.72 333.178 Q1297.92 334.185 1297.92 335.869 Q1297.92 336.668 1297.61 337.388 Q1297.32 338.1 1296.53 339.072 Q1296.31 339.324 1295.15 340.531 Q1293.99 341.729 1291.87 343.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1301.69 332.406 L1308.57 332.406 L1308.57 333.882 L1303.29 333.882 L1303.29 337.059 Q1303.67 336.928 1304.05 336.868 Q1304.44 336.798 1304.82 336.798 Q1306.99 336.798 1308.26 337.987 Q1309.52 339.177 1309.52 341.208 Q1309.52 343.3 1308.22 344.463 Q1306.92 345.618 1304.55 345.618 Q1303.73 345.618 1302.88 345.479 Q1302.04 345.34 1301.14 345.062 L1301.14 343.3 Q1301.92 343.725 1302.75 343.934 Q1303.59 344.142 1304.52 344.142 Q1306.02 344.142 1306.89 343.352 Q1307.77 342.562 1307.77 341.208 Q1307.77 339.854 1306.89 339.064 Q1306.02 338.274 1304.52 338.274 Q1303.81 338.274 1303.11 338.43 Q1302.41 338.586 1301.69 338.916 L1301.69 332.406 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1331.44 335.291 L1337.56 335.291 L1337.56 336.766 L1329.33 336.766 L1329.33 335.291 Q1330.32 334.258 1332.04 332.521 Q1333.77 330.777 1334.21 330.273 Q1335.06 329.327 1335.39 328.676 Q1335.72 328.016 1335.72 327.383 Q1335.72 326.35 1334.99 325.699 Q1334.27 325.048 1333.11 325.048 Q1332.29 325.048 1331.37 325.334 Q1330.45 325.62 1329.41 326.202 L1329.41 324.431 Q1330.47 324.006 1331.39 323.789 Q1332.31 323.572 1333.08 323.572 Q1335.09 323.572 1336.29 324.579 Q1337.49 325.586 1337.49 327.27 Q1337.49 328.068 1337.18 328.789 Q1336.89 329.501 1336.1 330.473 Q1335.88 330.725 1334.72 331.931 Q1333.55 333.129 1331.44 335.291 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1345.2 329.587 Q1344.02 329.587 1343.33 330.395 Q1342.64 331.202 1342.64 332.608 Q1342.64 334.006 1343.33 334.822 Q1344.02 335.629 1345.2 335.629 Q1346.38 335.629 1347.07 334.822 Q1347.76 334.006 1347.76 332.608 Q1347.76 331.202 1347.07 330.395 Q1346.38 329.587 1345.2 329.587 M1348.68 324.093 L1348.68 325.69 Q1348.02 325.377 1347.35 325.212 Q1346.68 325.048 1346.02 325.048 Q1344.28 325.048 1343.36 326.219 Q1342.45 327.391 1342.32 329.761 Q1342.83 329.006 1343.61 328.607 Q1344.38 328.199 1345.31 328.199 Q1347.26 328.199 1348.39 329.388 Q1349.53 330.568 1349.53 332.608 Q1349.53 334.605 1348.35 335.811 Q1347.16 337.018 1345.2 337.018 Q1342.95 337.018 1341.77 335.299 Q1340.58 333.572 1340.58 330.299 Q1340.58 327.226 1342.03 325.403 Q1343.49 323.572 1345.95 323.572 Q1346.61 323.572 1347.28 323.702 Q1347.95 323.832 1348.68 324.093 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1371.42 326.988 L1377.54 326.988 L1377.54 328.464 L1369.31 328.464 L1369.31 326.988 Q1370.31 325.955 1372.03 324.219 Q1373.76 322.474 1374.2 321.971 Q1375.04 321.025 1375.37 320.374 Q1375.71 319.714 1375.71 319.08 Q1375.71 318.047 1374.98 317.396 Q1374.26 316.745 1373.1 316.745 Q1372.27 316.745 1371.35 317.032 Q1370.44 317.318 1369.4 317.9 L1369.4 316.129 Q1370.46 315.703 1371.38 315.486 Q1372.3 315.269 1373.06 315.269 Q1375.08 315.269 1376.27 316.276 Q1377.47 317.283 1377.47 318.967 Q1377.47 319.766 1377.17 320.486 Q1376.87 321.198 1376.08 322.17 Q1375.87 322.422 1374.7 323.629 Q1373.54 324.827 1371.42 326.988 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1380.78 315.504 L1389.11 315.504 L1389.11 316.25 L1384.41 328.464 L1382.58 328.464 L1387 316.979 L1380.78 316.979 L1380.78 315.504 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1411.05 318.954 L1417.17 318.954 L1417.17 320.43 L1408.94 320.43 L1408.94 318.954 Q1409.94 317.921 1411.66 316.185 Q1413.39 314.44 1413.83 313.937 Q1414.67 312.991 1415 312.34 Q1415.34 311.68 1415.34 311.046 Q1415.34 310.013 1414.61 309.362 Q1413.89 308.711 1412.73 308.711 Q1411.9 308.711 1410.98 308.998 Q1410.07 309.284 1409.03 309.866 L1409.03 308.095 Q1410.09 307.669 1411.01 307.452 Q1411.93 307.235 1412.69 307.235 Q1414.71 307.235 1415.91 308.242 Q1417.1 309.249 1417.1 310.933 Q1417.1 311.732 1416.8 312.452 Q1416.51 313.164 1415.72 314.136 Q1415.5 314.388 1414.34 315.595 Q1413.17 316.793 1411.05 318.954 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1424.6 314.275 Q1423.35 314.275 1422.63 314.944 Q1421.92 315.612 1421.92 316.784 Q1421.92 317.956 1422.63 318.624 Q1423.35 319.293 1424.6 319.293 Q1425.85 319.293 1426.57 318.624 Q1427.3 317.947 1427.3 316.784 Q1427.3 315.612 1426.57 314.944 Q1425.86 314.275 1424.6 314.275 M1422.85 313.529 Q1421.72 313.251 1421.09 312.478 Q1420.46 311.706 1420.46 310.595 Q1420.46 309.041 1421.57 308.138 Q1422.68 307.235 1424.6 307.235 Q1426.54 307.235 1427.64 308.138 Q1428.74 309.041 1428.74 310.595 Q1428.74 311.706 1428.11 312.478 Q1427.49 313.251 1426.37 313.529 Q1427.63 313.824 1428.34 314.683 Q1429.05 315.543 1429.05 316.784 Q1429.05 318.668 1427.89 319.675 Q1426.75 320.682 1424.6 320.682 Q1422.46 320.682 1421.31 319.675 Q1420.16 318.668 1420.16 316.784 Q1420.16 315.543 1420.87 314.683 Q1421.58 313.824 1422.85 313.529 M1422.21 310.76 Q1422.21 311.767 1422.83 312.331 Q1423.47 312.895 1424.6 312.895 Q1425.73 312.895 1426.37 312.331 Q1427.01 311.767 1427.01 310.76 Q1427.01 309.753 1426.37 309.189 Q1425.73 308.624 1424.6 308.624 Q1423.47 308.624 1422.83 309.189 Q1422.21 309.753 1422.21 310.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1450.86 311.164 L1456.98 311.164 L1456.98 312.64 L1448.75 312.64 L1448.75 311.164 Q1449.75 310.131 1451.46 308.395 Q1453.19 306.65 1453.63 306.147 Q1454.48 305.2 1454.81 304.549 Q1455.14 303.89 1455.14 303.256 Q1455.14 302.223 1454.42 301.572 Q1453.7 300.921 1452.53 300.921 Q1451.71 300.921 1450.79 301.207 Q1449.88 301.494 1448.83 302.075 L1448.83 300.305 Q1449.89 299.879 1450.81 299.662 Q1451.73 299.445 1452.5 299.445 Q1454.51 299.445 1455.71 300.452 Q1456.91 301.459 1456.91 303.143 Q1456.91 303.942 1456.6 304.662 Q1456.31 305.374 1455.52 306.346 Q1455.3 306.598 1454.14 307.805 Q1452.97 309.003 1450.86 311.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1460.71 312.371 L1460.71 310.773 Q1461.37 311.086 1462.05 311.251 Q1462.72 311.416 1463.37 311.416 Q1465.11 311.416 1466.02 310.253 Q1466.94 309.081 1467.07 306.702 Q1466.57 307.449 1465.8 307.848 Q1465.02 308.247 1464.09 308.247 Q1462.14 308.247 1461 307.075 Q1459.88 305.895 1459.88 303.855 Q1459.88 301.858 1461.06 300.652 Q1462.24 299.445 1464.2 299.445 Q1466.45 299.445 1467.63 301.173 Q1468.82 302.891 1468.82 306.173 Q1468.82 309.237 1467.36 311.068 Q1465.91 312.891 1463.45 312.891 Q1462.79 312.891 1462.12 312.761 Q1461.44 312.631 1460.71 312.371 M1464.2 306.876 Q1465.38 306.876 1466.06 306.068 Q1466.76 305.261 1466.76 303.855 Q1466.76 302.457 1466.06 301.65 Q1465.38 300.834 1464.2 300.834 Q1463.02 300.834 1462.32 301.65 Q1461.64 302.457 1461.64 303.855 Q1461.64 305.261 1462.32 306.068 Q1463.02 306.876 1464.2 306.876 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1494.38 298.085 Q1495.64 298.354 1496.34 299.204 Q1497.06 300.055 1497.06 301.305 Q1497.06 303.224 1495.74 304.274 Q1494.42 305.324 1491.99 305.324 Q1491.17 305.324 1490.3 305.159 Q1489.44 305.003 1488.52 304.682 L1488.52 302.989 Q1489.25 303.414 1490.12 303.632 Q1490.99 303.849 1491.94 303.849 Q1493.58 303.849 1494.44 303.197 Q1495.31 302.546 1495.31 301.305 Q1495.31 300.159 1494.5 299.517 Q1493.71 298.866 1492.27 298.866 L1490.76 298.866 L1490.76 297.425 L1492.34 297.425 Q1493.64 297.425 1494.32 296.913 Q1495.01 296.392 1495.01 295.42 Q1495.01 294.421 1494.3 293.892 Q1493.59 293.354 1492.27 293.354 Q1491.55 293.354 1490.73 293.51 Q1489.9 293.666 1488.91 293.996 L1488.91 292.434 Q1489.91 292.156 1490.78 292.017 Q1491.66 291.878 1492.43 291.878 Q1494.43 291.878 1495.59 292.79 Q1496.75 293.692 1496.75 295.237 Q1496.75 296.314 1496.14 297.06 Q1495.52 297.798 1494.38 298.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1504.13 293.267 Q1502.78 293.267 1502.09 294.604 Q1501.41 295.932 1501.41 298.605 Q1501.41 301.27 1502.09 302.607 Q1502.78 303.935 1504.13 303.935 Q1505.49 303.935 1506.17 302.607 Q1506.86 301.27 1506.86 298.605 Q1506.86 295.932 1506.17 294.604 Q1505.49 293.267 1504.13 293.267 M1504.13 291.878 Q1506.31 291.878 1507.46 293.606 Q1508.61 295.324 1508.61 298.605 Q1508.61 301.878 1507.46 303.605 Q1506.31 305.324 1504.13 305.324 Q1501.95 305.324 1500.8 303.605 Q1499.65 301.878 1499.65 298.605 Q1499.65 295.324 1500.8 293.606 Q1501.95 291.878 1504.13 291.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1534.4 290.722 Q1535.66 290.991 1536.36 291.842 Q1537.07 292.693 1537.07 293.943 Q1537.07 295.861 1535.75 296.912 Q1534.43 297.962 1532 297.962 Q1531.19 297.962 1530.32 297.797 Q1529.46 297.641 1528.54 297.32 L1528.54 295.627 Q1529.27 296.052 1530.14 296.269 Q1531 296.486 1531.95 296.486 Q1533.6 296.486 1534.46 295.835 Q1535.33 295.184 1535.33 293.943 Q1535.33 292.797 1534.52 292.155 Q1533.72 291.504 1532.29 291.504 L1530.78 291.504 L1530.78 290.063 L1532.36 290.063 Q1533.65 290.063 1534.34 289.55 Q1535.02 289.03 1535.02 288.057 Q1535.02 287.059 1534.31 286.53 Q1533.61 285.991 1532.29 285.991 Q1531.57 285.991 1530.74 286.148 Q1529.92 286.304 1528.93 286.634 L1528.93 285.071 Q1529.93 284.794 1530.8 284.655 Q1531.67 284.516 1532.45 284.516 Q1534.44 284.516 1535.6 285.427 Q1536.77 286.33 1536.77 287.875 Q1536.77 288.952 1536.15 289.698 Q1535.54 290.436 1534.4 290.722 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1540.7 296.234 L1543.56 296.234 L1543.56 286.347 L1540.45 286.972 L1540.45 285.375 L1543.55 284.75 L1545.3 284.75 L1545.3 296.234 L1548.17 296.234 L1548.17 297.71 L1540.7 297.71 L1540.7 296.234 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1574.25 283.549 Q1575.51 283.818 1576.21 284.669 Q1576.93 285.52 1576.93 286.77 Q1576.93 288.688 1575.61 289.738 Q1574.29 290.789 1571.86 290.789 Q1571.04 290.789 1570.17 290.624 Q1569.31 290.467 1568.39 290.146 L1568.39 288.454 Q1569.12 288.879 1569.99 289.096 Q1570.86 289.313 1571.81 289.313 Q1573.45 289.313 1574.31 288.662 Q1575.18 288.011 1575.18 286.77 Q1575.18 285.624 1574.37 284.981 Q1573.58 284.33 1572.14 284.33 L1570.63 284.33 L1570.63 282.889 L1572.21 282.889 Q1573.51 282.889 1574.19 282.377 Q1574.88 281.856 1574.88 280.884 Q1574.88 279.886 1574.17 279.356 Q1573.46 278.818 1572.14 278.818 Q1571.42 278.818 1570.6 278.974 Q1569.77 279.131 1568.78 279.461 L1568.78 277.898 Q1569.78 277.62 1570.65 277.481 Q1571.53 277.342 1572.3 277.342 Q1574.3 277.342 1575.46 278.254 Q1576.62 279.157 1576.62 280.702 Q1576.62 281.778 1576.01 282.525 Q1575.39 283.263 1574.25 283.549 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1581.76 289.061 L1587.88 289.061 L1587.88 290.537 L1579.65 290.537 L1579.65 289.061 Q1580.65 288.028 1582.37 286.292 Q1584.1 284.547 1584.54 284.044 Q1585.38 283.098 1585.71 282.447 Q1586.05 281.787 1586.05 281.153 Q1586.05 280.12 1585.32 279.469 Q1584.6 278.818 1583.44 278.818 Q1582.61 278.818 1581.69 279.105 Q1580.78 279.391 1579.74 279.973 L1579.74 278.202 Q1580.8 277.777 1581.72 277.56 Q1582.64 277.342 1583.4 277.342 Q1585.42 277.342 1586.61 278.349 Q1587.81 279.356 1587.81 281.04 Q1587.81 281.839 1587.51 282.559 Q1587.21 283.271 1586.42 284.243 Q1586.21 284.495 1585.04 285.702 Q1583.88 286.9 1581.76 289.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1613.86 276.551 Q1615.12 276.82 1615.82 277.671 Q1616.53 278.521 1616.53 279.771 Q1616.53 281.69 1615.21 282.74 Q1613.9 283.791 1611.46 283.791 Q1610.65 283.791 1609.78 283.626 Q1608.92 283.469 1608 283.148 L1608 281.455 Q1608.73 281.881 1609.6 282.098 Q1610.47 282.315 1611.41 282.315 Q1613.06 282.315 1613.92 281.664 Q1614.79 281.013 1614.79 279.771 Q1614.79 278.626 1613.98 277.983 Q1613.18 277.332 1611.75 277.332 L1610.24 277.332 L1610.24 275.891 L1611.82 275.891 Q1613.11 275.891 1613.8 275.379 Q1614.49 274.858 1614.49 273.886 Q1614.49 272.888 1613.77 272.358 Q1613.07 271.82 1611.75 271.82 Q1611.03 271.82 1610.21 271.976 Q1609.38 272.133 1608.39 272.462 L1608.39 270.9 Q1609.39 270.622 1610.26 270.483 Q1611.13 270.344 1611.91 270.344 Q1613.9 270.344 1615.07 271.256 Q1616.23 272.159 1616.23 273.704 Q1616.23 274.78 1615.61 275.527 Q1615 276.265 1613.86 276.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1625.17 276.551 Q1626.43 276.82 1627.13 277.671 Q1627.84 278.521 1627.84 279.771 Q1627.84 281.69 1626.53 282.74 Q1625.21 283.791 1622.78 283.791 Q1621.96 283.791 1621.09 283.626 Q1620.23 283.469 1619.31 283.148 L1619.31 281.455 Q1620.04 281.881 1620.91 282.098 Q1621.78 282.315 1622.72 282.315 Q1624.37 282.315 1625.23 281.664 Q1626.1 281.013 1626.1 279.771 Q1626.1 278.626 1625.29 277.983 Q1624.49 277.332 1623.06 277.332 L1621.55 277.332 L1621.55 275.891 L1623.13 275.891 Q1624.42 275.891 1625.11 275.379 Q1625.8 274.858 1625.8 273.886 Q1625.8 272.888 1625.08 272.358 Q1624.38 271.82 1623.06 271.82 Q1622.34 271.82 1621.52 271.976 Q1620.69 272.133 1619.7 272.462 L1619.7 270.9 Q1620.7 270.622 1621.57 270.483 Q1622.45 270.344 1623.22 270.344 Q1625.21 270.344 1626.38 271.256 Q1627.54 272.159 1627.54 273.704 Q1627.54 274.78 1626.92 275.527 Q1626.31 276.265 1625.17 276.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1653.43 269.716 Q1654.69 269.985 1655.39 270.836 Q1656.11 271.686 1656.11 272.936 Q1656.11 274.855 1654.79 275.905 Q1653.47 276.955 1651.04 276.955 Q1650.22 276.955 1649.35 276.79 Q1648.49 276.634 1647.57 276.313 L1647.57 274.62 Q1648.3 275.046 1649.17 275.263 Q1650.04 275.48 1650.98 275.48 Q1652.63 275.48 1653.49 274.829 Q1654.36 274.178 1654.36 272.936 Q1654.36 271.791 1653.55 271.148 Q1652.76 270.497 1651.32 270.497 L1649.81 270.497 L1649.81 269.056 L1651.39 269.056 Q1652.69 269.056 1653.37 268.544 Q1654.06 268.023 1654.06 267.051 Q1654.06 266.053 1653.35 265.523 Q1652.64 264.985 1651.32 264.985 Q1650.6 264.985 1649.78 265.141 Q1648.95 265.297 1647.96 265.627 L1647.96 264.065 Q1648.96 263.787 1649.83 263.648 Q1650.71 263.509 1651.48 263.509 Q1653.48 263.509 1654.64 264.421 Q1655.8 265.324 1655.8 266.869 Q1655.8 267.945 1655.19 268.692 Q1654.57 269.429 1653.43 269.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1664.25 265.271 L1659.82 272.19 L1664.25 272.19 L1664.25 265.271 M1663.79 263.744 L1665.99 263.744 L1665.99 272.19 L1667.84 272.19 L1667.84 273.648 L1665.99 273.648 L1665.99 276.704 L1664.25 276.704 L1664.25 273.648 L1658.4 273.648 L1658.4 271.955 L1663.79 263.744 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1693.5 263.033 Q1694.75 263.302 1695.46 264.153 Q1696.17 265.003 1696.17 266.253 Q1696.17 268.172 1694.85 269.222 Q1693.53 270.272 1691.1 270.272 Q1690.28 270.272 1689.42 270.107 Q1688.56 269.951 1687.64 269.63 L1687.64 267.937 Q1688.37 268.363 1689.23 268.58 Q1690.1 268.797 1691.05 268.797 Q1692.7 268.797 1693.56 268.146 Q1694.42 267.495 1694.42 266.253 Q1694.42 265.107 1693.62 264.465 Q1692.82 263.814 1691.39 263.814 L1689.88 263.814 L1689.88 262.373 L1691.46 262.373 Q1692.75 262.373 1693.44 261.861 Q1694.12 261.34 1694.12 260.368 Q1694.12 259.37 1693.41 258.84 Q1692.71 258.302 1691.39 258.302 Q1690.67 258.302 1689.84 258.458 Q1689.02 258.614 1688.03 258.944 L1688.03 257.382 Q1689.03 257.104 1689.89 256.965 Q1690.77 256.826 1691.54 256.826 Q1693.54 256.826 1694.7 257.738 Q1695.87 258.64 1695.87 260.186 Q1695.87 261.262 1695.25 262.009 Q1694.63 262.746 1693.5 263.033 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1699.51 257.061 L1706.4 257.061 L1706.4 258.536 L1701.12 258.536 L1701.12 261.713 Q1701.5 261.583 1701.88 261.522 Q1702.26 261.453 1702.65 261.453 Q1704.82 261.453 1706.08 262.642 Q1707.35 263.831 1707.35 265.863 Q1707.35 267.955 1706.05 269.118 Q1704.75 270.272 1702.38 270.272 Q1701.56 270.272 1700.71 270.133 Q1699.87 269.995 1698.96 269.717 L1698.96 267.955 Q1699.75 268.38 1700.58 268.588 Q1701.41 268.797 1702.34 268.797 Q1703.84 268.797 1704.72 268.007 Q1705.6 267.217 1705.6 265.863 Q1705.6 264.509 1704.72 263.719 Q1703.84 262.929 1702.34 262.929 Q1701.64 262.929 1700.94 263.085 Q1700.24 263.241 1699.51 263.571 L1699.51 257.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1733.06 256.492 Q1734.32 256.761 1735.03 257.612 Q1735.74 258.463 1735.74 259.713 Q1735.74 261.631 1734.42 262.681 Q1733.1 263.732 1730.67 263.732 Q1729.85 263.732 1728.98 263.567 Q1728.12 263.411 1727.2 263.089 L1727.2 261.397 Q1727.93 261.822 1728.8 262.039 Q1729.67 262.256 1730.62 262.256 Q1732.27 262.256 1733.12 261.605 Q1733.99 260.954 1733.99 259.713 Q1733.99 258.567 1733.19 257.924 Q1732.39 257.273 1730.95 257.273 L1729.44 257.273 L1729.44 255.832 L1731.02 255.832 Q1732.32 255.832 1733 255.32 Q1733.69 254.799 1733.69 253.827 Q1733.69 252.829 1732.98 252.299 Q1732.27 251.761 1730.95 251.761 Q1730.23 251.761 1729.41 251.918 Q1728.58 252.074 1727.6 252.404 L1727.6 250.841 Q1728.59 250.563 1729.46 250.424 Q1730.34 250.286 1731.11 250.286 Q1733.11 250.286 1734.27 251.197 Q1735.43 252.1 1735.43 253.645 Q1735.43 254.721 1734.82 255.468 Q1734.2 256.206 1733.06 256.492 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1743.03 256.301 Q1741.85 256.301 1741.15 257.108 Q1740.47 257.916 1740.47 259.322 Q1740.47 260.72 1741.15 261.536 Q1741.85 262.343 1743.03 262.343 Q1744.21 262.343 1744.9 261.536 Q1745.59 260.72 1745.59 259.322 Q1745.59 257.916 1744.9 257.108 Q1744.21 256.301 1743.03 256.301 M1746.51 250.806 L1746.51 252.404 Q1745.85 252.091 1745.17 251.926 Q1744.5 251.761 1743.85 251.761 Q1742.11 251.761 1741.19 252.933 Q1740.28 254.105 1740.15 256.475 Q1740.66 255.72 1741.43 255.32 Q1742.2 254.912 1743.13 254.912 Q1745.09 254.912 1746.22 256.102 Q1747.35 257.282 1747.35 259.322 Q1747.35 261.319 1746.17 262.525 Q1744.99 263.732 1743.03 263.732 Q1740.78 263.732 1739.59 262.013 Q1738.4 260.286 1738.4 257.013 Q1738.4 253.94 1739.86 252.117 Q1741.32 250.286 1743.78 250.286 Q1744.44 250.286 1745.1 250.416 Q1745.78 250.546 1746.51 250.806 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1773.05 250.085 Q1774.31 250.354 1775.01 251.205 Q1775.72 252.056 1775.72 253.306 Q1775.72 255.224 1774.4 256.274 Q1773.08 257.325 1770.65 257.325 Q1769.84 257.325 1768.97 257.16 Q1768.11 257.003 1767.19 256.682 L1767.19 254.99 Q1767.92 255.415 1768.79 255.632 Q1769.65 255.849 1770.6 255.849 Q1772.25 255.849 1773.11 255.198 Q1773.98 254.547 1773.98 253.306 Q1773.98 252.16 1773.17 251.517 Q1772.37 250.866 1770.94 250.866 L1769.43 250.866 L1769.43 249.425 L1771.01 249.425 Q1772.3 249.425 1772.99 248.913 Q1773.67 248.392 1773.67 247.42 Q1773.67 246.422 1772.96 245.892 Q1772.26 245.354 1770.94 245.354 Q1770.22 245.354 1769.39 245.51 Q1768.57 245.667 1767.58 245.997 L1767.58 244.434 Q1768.58 244.156 1769.45 244.017 Q1770.32 243.879 1771.1 243.879 Q1773.09 243.879 1774.26 244.79 Q1775.42 245.693 1775.42 247.238 Q1775.42 248.314 1774.8 249.061 Q1774.19 249.799 1773.05 250.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1778.6 244.113 L1786.94 244.113 L1786.94 244.859 L1782.23 257.073 L1780.4 257.073 L1784.83 245.589 L1778.6 245.589 L1778.6 244.113 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1812.68 243.804 Q1813.94 244.073 1814.64 244.924 Q1815.36 245.774 1815.36 247.024 Q1815.36 248.943 1814.04 249.993 Q1812.72 251.043 1810.29 251.043 Q1809.47 251.043 1808.6 250.878 Q1807.74 250.722 1806.82 250.401 L1806.82 248.708 Q1807.55 249.134 1808.42 249.351 Q1809.29 249.568 1810.23 249.568 Q1811.88 249.568 1812.74 248.917 Q1813.61 248.266 1813.61 247.024 Q1813.61 245.878 1812.8 245.236 Q1812.01 244.585 1810.57 244.585 L1809.06 244.585 L1809.06 243.144 L1810.64 243.144 Q1811.94 243.144 1812.62 242.632 Q1813.31 242.111 1813.31 241.139 Q1813.31 240.141 1812.6 239.611 Q1811.89 239.073 1810.57 239.073 Q1809.85 239.073 1809.03 239.229 Q1808.2 239.385 1807.21 239.715 L1807.21 238.153 Q1808.21 237.875 1809.08 237.736 Q1809.96 237.597 1810.73 237.597 Q1812.73 237.597 1813.89 238.509 Q1815.05 239.411 1815.05 240.957 Q1815.05 242.033 1814.44 242.78 Q1813.82 243.517 1812.68 243.804 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1822.43 244.637 Q1821.18 244.637 1820.46 245.306 Q1819.75 245.974 1819.75 247.146 Q1819.75 248.318 1820.46 248.986 Q1821.18 249.654 1822.43 249.654 Q1823.68 249.654 1824.4 248.986 Q1825.12 248.309 1825.12 247.146 Q1825.12 245.974 1824.4 245.306 Q1823.69 244.637 1822.43 244.637 M1820.68 243.891 Q1819.55 243.613 1818.91 242.84 Q1818.29 242.068 1818.29 240.957 Q1818.29 239.403 1819.39 238.5 Q1820.5 237.597 1822.43 237.597 Q1824.37 237.597 1825.47 238.5 Q1826.57 239.403 1826.57 240.957 Q1826.57 242.068 1825.94 242.84 Q1825.31 243.613 1824.19 243.891 Q1825.46 244.186 1826.16 245.045 Q1826.87 245.904 1826.87 247.146 Q1826.87 249.029 1825.72 250.036 Q1824.57 251.043 1822.43 251.043 Q1820.29 251.043 1819.13 250.036 Q1817.99 249.029 1817.99 247.146 Q1817.99 245.904 1818.7 245.045 Q1819.41 244.186 1820.68 243.891 M1820.03 241.122 Q1820.03 242.128 1820.66 242.693 Q1821.29 243.257 1822.43 243.257 Q1823.56 243.257 1824.19 242.693 Q1824.83 242.128 1824.83 241.122 Q1824.83 240.115 1824.19 239.55 Q1823.56 238.986 1822.43 238.986 Q1821.29 238.986 1820.66 239.55 Q1820.03 240.115 1820.03 241.122 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1852.48 237.641 Q1853.74 237.91 1854.45 238.761 Q1855.16 239.612 1855.16 240.862 Q1855.16 242.78 1853.84 243.83 Q1852.52 244.881 1850.09 244.881 Q1849.27 244.881 1848.41 244.716 Q1847.55 244.56 1846.63 244.238 L1846.63 242.546 Q1847.35 242.971 1848.22 243.188 Q1849.09 243.405 1850.04 243.405 Q1851.69 243.405 1852.55 242.754 Q1853.41 242.103 1853.41 240.862 Q1853.41 239.716 1852.61 239.073 Q1851.81 238.422 1850.38 238.422 L1848.87 238.422 L1848.87 236.981 L1850.44 236.981 Q1851.74 236.981 1852.42 236.469 Q1853.11 235.948 1853.11 234.976 Q1853.11 233.978 1852.4 233.448 Q1851.69 232.91 1850.38 232.91 Q1849.65 232.91 1848.83 233.067 Q1848.01 233.223 1847.02 233.553 L1847.02 231.99 Q1848.01 231.712 1848.88 231.573 Q1849.76 231.435 1850.53 231.435 Q1852.53 231.435 1853.69 232.346 Q1854.85 233.249 1854.85 234.794 Q1854.85 235.87 1854.24 236.617 Q1853.62 237.355 1852.48 237.641 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1858.54 244.36 L1858.54 242.763 Q1859.19 243.075 1859.87 243.24 Q1860.55 243.405 1861.2 243.405 Q1862.94 243.405 1863.85 242.242 Q1864.77 241.07 1864.9 238.691 Q1864.39 239.438 1863.62 239.837 Q1862.85 240.237 1861.91 240.237 Q1859.97 240.237 1858.83 239.065 Q1857.7 237.884 1857.7 235.844 Q1857.7 233.848 1858.88 232.641 Q1860.06 231.435 1862.02 231.435 Q1864.27 231.435 1865.45 233.162 Q1866.64 234.881 1866.64 238.162 Q1866.64 241.226 1865.18 243.058 Q1863.73 244.881 1861.28 244.881 Q1860.62 244.881 1859.94 244.75 Q1859.26 244.62 1858.54 244.36 M1862.02 238.865 Q1863.21 238.865 1863.89 238.058 Q1864.59 237.251 1864.59 235.844 Q1864.59 234.447 1863.89 233.639 Q1863.21 232.823 1862.02 232.823 Q1860.84 232.823 1860.15 233.639 Q1859.46 234.447 1859.46 235.844 Q1859.46 237.251 1860.15 238.058 Q1860.84 238.865 1862.02 238.865 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1891.98 227.146 L1887.56 234.065 L1891.98 234.065 L1891.98 227.146 M1891.52 225.618 L1893.73 225.618 L1893.73 234.065 L1895.58 234.065 L1895.58 235.523 L1893.73 235.523 L1893.73 238.578 L1891.98 238.578 L1891.98 235.523 L1886.13 235.523 L1886.13 233.83 L1891.52 225.618 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1902.23 226.773 Q1900.87 226.773 1900.19 228.11 Q1899.51 229.438 1899.51 232.111 Q1899.51 234.776 1900.19 236.113 Q1900.87 237.441 1902.23 237.441 Q1903.59 237.441 1904.27 236.113 Q1904.95 234.776 1904.95 232.111 Q1904.95 229.438 1904.27 228.11 Q1903.59 226.773 1902.23 226.773 M1902.23 225.384 Q1904.41 225.384 1905.55 227.111 Q1906.71 228.83 1906.71 232.111 Q1906.71 235.384 1905.55 237.111 Q1904.41 238.83 1902.23 238.83 Q1900.05 238.83 1898.89 237.111 Q1897.75 235.384 1897.75 232.111 Q1897.75 228.83 1898.89 227.111 Q1900.05 225.384 1902.23 225.384 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1932 221.202 L1927.57 228.12 L1932 228.12 L1932 221.202 M1931.54 219.674 L1933.74 219.674 L1933.74 228.12 L1935.59 228.12 L1935.59 229.579 L1933.74 229.579 L1933.74 232.634 L1932 232.634 L1932 229.579 L1926.15 229.579 L1926.15 227.886 L1931.54 219.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1938.8 231.159 L1941.66 231.159 L1941.66 221.271 L1938.54 221.896 L1938.54 220.299 L1941.64 219.674 L1943.4 219.674 L1943.4 231.159 L1946.26 231.159 L1946.26 232.634 L1938.8 232.634 L1938.8 231.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1971.85 215.359 L1967.43 222.277 L1971.85 222.277 L1971.85 215.359 M1971.39 213.831 L1973.6 213.831 L1973.6 222.277 L1975.45 222.277 L1975.45 223.735 L1973.6 223.735 L1973.6 226.791 L1971.85 226.791 L1971.85 223.735 L1966 223.735 L1966 222.043 L1971.39 213.831 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1979.86 225.315 L1985.98 225.315 L1985.98 226.791 L1977.75 226.791 L1977.75 225.315 Q1978.75 224.282 1980.46 222.546 Q1982.19 220.801 1982.63 220.298 Q1983.48 219.352 1983.81 218.701 Q1984.15 218.041 1984.15 217.407 Q1984.15 216.374 1983.42 215.723 Q1982.7 215.072 1981.53 215.072 Q1980.71 215.072 1979.79 215.359 Q1978.88 215.645 1977.83 216.227 L1977.83 214.456 Q1978.89 214.03 1979.81 213.813 Q1980.73 213.596 1981.5 213.596 Q1983.51 213.596 1984.71 214.603 Q1985.91 215.61 1985.91 217.294 Q1985.91 218.093 1985.6 218.813 Q1985.31 219.525 1984.52 220.497 Q1984.3 220.749 1983.14 221.956 Q1981.98 223.154 1979.86 225.315 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2011.46 209.611 L2007.03 216.529 L2011.46 216.529 L2011.46 209.611 M2011 208.083 L2013.21 208.083 L2013.21 216.529 L2015.05 216.529 L2015.05 217.988 L2013.21 217.988 L2013.21 221.043 L2011.46 221.043 L2011.46 217.988 L2005.61 217.988 L2005.61 216.295 L2011 208.083 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2023.27 214.055 Q2024.52 214.325 2025.23 215.175 Q2025.94 216.026 2025.94 217.276 Q2025.94 219.194 2024.62 220.245 Q2023.3 221.295 2020.87 221.295 Q2020.05 221.295 2019.19 221.13 Q2018.33 220.974 2017.41 220.653 L2017.41 218.96 Q2018.14 219.385 2019 219.602 Q2019.87 219.819 2020.82 219.819 Q2022.47 219.819 2023.33 219.168 Q2024.2 218.517 2024.2 217.276 Q2024.2 216.13 2023.39 215.488 Q2022.59 214.837 2021.16 214.837 L2019.65 214.837 L2019.65 213.396 L2021.23 213.396 Q2022.52 213.396 2023.21 212.884 Q2023.89 212.363 2023.89 211.39 Q2023.89 210.392 2023.18 209.863 Q2022.48 209.325 2021.16 209.325 Q2020.44 209.325 2019.61 209.481 Q2018.79 209.637 2017.8 209.967 L2017.8 208.404 Q2018.8 208.127 2019.66 207.988 Q2020.54 207.849 2021.31 207.849 Q2023.31 207.849 2024.47 208.76 Q2025.64 209.663 2025.64 211.208 Q2025.64 212.285 2025.02 213.031 Q2024.4 213.769 2023.27 214.055 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2051.03 203.955 L2046.61 210.873 L2051.03 210.873 L2051.03 203.955 M2050.57 202.427 L2052.78 202.427 L2052.78 210.873 L2054.63 210.873 L2054.63 212.331 L2052.78 212.331 L2052.78 215.387 L2051.03 215.387 L2051.03 212.331 L2045.18 212.331 L2045.18 210.639 L2050.57 202.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2062.34 203.955 L2057.92 210.873 L2062.34 210.873 L2062.34 203.955 M2061.88 202.427 L2064.09 202.427 L2064.09 210.873 L2065.94 210.873 L2065.94 212.331 L2064.09 212.331 L2064.09 215.387 L2062.34 215.387 L2062.34 212.331 L2056.49 212.331 L2056.49 210.639 L2061.88 202.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2091.1 198.386 L2086.67 205.304 L2091.1 205.304 L2091.1 198.386 M2090.64 196.858 L2092.84 196.858 L2092.84 205.304 L2094.69 205.304 L2094.69 206.762 L2092.84 206.762 L2092.84 209.818 L2091.1 209.818 L2091.1 206.762 L2085.25 206.762 L2085.25 205.07 L2090.64 196.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2097.61 196.858 L2104.49 196.858 L2104.49 198.333 L2099.21 198.333 L2099.21 201.511 Q2099.59 201.38 2099.98 201.32 Q2100.36 201.25 2100.74 201.25 Q2102.91 201.25 2104.18 202.439 Q2105.45 203.629 2105.45 205.66 Q2105.45 207.752 2104.14 208.915 Q2102.84 210.07 2100.47 210.07 Q2099.66 210.07 2098.8 209.931 Q2097.96 209.792 2097.06 209.514 L2097.06 207.752 Q2097.84 208.177 2098.67 208.386 Q2099.51 208.594 2100.44 208.594 Q2101.94 208.594 2102.82 207.804 Q2103.69 207.014 2103.69 205.66 Q2103.69 204.306 2102.82 203.516 Q2101.94 202.726 2100.44 202.726 Q2099.73 202.726 2099.03 202.882 Q2098.34 203.038 2097.61 203.368 L2097.61 196.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2130.66 192.9 L2126.24 199.818 L2130.66 199.818 L2130.66 192.9 M2130.2 191.372 L2132.41 191.372 L2132.41 199.818 L2134.26 199.818 L2134.26 201.276 L2132.41 201.276 L2132.41 204.332 L2130.66 204.332 L2130.66 201.276 L2124.81 201.276 L2124.81 199.584 L2130.2 191.372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2141.12 197.153 Q2139.94 197.153 2139.25 197.96 Q2138.56 198.768 2138.56 200.174 Q2138.56 201.571 2139.25 202.387 Q2139.94 203.195 2141.12 203.195 Q2142.31 203.195 2142.99 202.387 Q2143.69 201.571 2143.69 200.174 Q2143.69 198.768 2142.99 197.96 Q2142.31 197.153 2141.12 197.153 M2144.61 191.658 L2144.61 193.256 Q2143.95 192.943 2143.27 192.778 Q2142.6 192.613 2141.94 192.613 Q2140.2 192.613 2139.28 193.785 Q2138.37 194.957 2138.24 197.327 Q2138.75 196.571 2139.53 196.172 Q2140.3 195.764 2141.23 195.764 Q2143.18 195.764 2144.31 196.953 Q2145.45 198.134 2145.45 200.174 Q2145.45 202.17 2144.27 203.377 Q2143.09 204.584 2141.12 204.584 Q2138.88 204.584 2137.69 202.865 Q2136.5 201.137 2136.5 197.865 Q2136.5 194.792 2137.96 192.969 Q2139.41 191.137 2141.87 191.137 Q2142.53 191.137 2143.2 191.268 Q2143.88 191.398 2144.61 191.658 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2170.65 187.493 L2166.22 194.412 L2170.65 194.412 L2170.65 187.493 M2170.19 185.965 L2172.39 185.965 L2172.39 194.412 L2174.24 194.412 L2174.24 195.87 L2172.39 195.87 L2172.39 198.925 L2170.65 198.925 L2170.65 195.87 L2164.8 195.87 L2164.8 194.177 L2170.19 185.965 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2176.7 185.965 L2185.03 185.965 L2185.03 186.712 L2180.33 198.925 L2178.5 198.925 L2182.92 187.441 L2176.7 187.441 L2176.7 185.965 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2210.28 182.163 L2205.86 189.081 L2210.28 189.081 L2210.28 182.163 M2209.82 180.635 L2212.03 180.635 L2212.03 189.081 L2213.88 189.081 L2213.88 190.54 L2212.03 190.54 L2212.03 193.595 L2210.28 193.595 L2210.28 190.54 L2204.43 190.54 L2204.43 188.847 L2209.82 180.635 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2220.53 187.441 Q2219.28 187.441 2218.56 188.109 Q2217.84 188.778 2217.84 189.949 Q2217.84 191.121 2218.56 191.79 Q2219.28 192.458 2220.53 192.458 Q2221.78 192.458 2222.5 191.79 Q2223.22 191.113 2223.22 189.949 Q2223.22 188.778 2222.5 188.109 Q2221.78 187.441 2220.53 187.441 M2218.77 186.694 Q2217.64 186.416 2217.01 185.644 Q2216.39 184.871 2216.39 183.76 Q2216.39 182.206 2217.49 181.304 Q2218.6 180.401 2220.53 180.401 Q2222.46 180.401 2223.56 181.304 Q2224.67 182.206 2224.67 183.76 Q2224.67 184.871 2224.03 185.644 Q2223.41 186.416 2222.29 186.694 Q2223.56 186.989 2224.26 187.849 Q2224.97 188.708 2224.97 189.949 Q2224.97 191.833 2223.82 192.84 Q2222.67 193.847 2220.53 193.847 Q2218.38 193.847 2217.23 192.84 Q2216.08 191.833 2216.08 189.949 Q2216.08 188.708 2216.79 187.849 Q2217.5 186.989 2218.77 186.694 M2218.13 183.925 Q2218.13 184.932 2218.75 185.496 Q2219.39 186.061 2220.53 186.061 Q2221.65 186.061 2222.29 185.496 Q2222.93 184.932 2222.93 183.925 Q2222.93 182.918 2222.29 182.354 Q2221.65 181.79 2220.53 181.79 Q2219.39 181.79 2218.75 182.354 Q2218.13 182.918 2218.13 183.925 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2250.09 176.906 L2245.66 183.824 L2250.09 183.824 L2250.09 176.906 M2249.63 175.378 L2251.83 175.378 L2251.83 183.824 L2253.68 183.824 L2253.68 185.283 L2251.83 185.283 L2251.83 188.338 L2250.09 188.338 L2250.09 185.283 L2244.23 185.283 L2244.23 183.59 L2249.63 175.378 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2256.63 188.069 L2256.63 186.472 Q2257.29 186.784 2257.97 186.949 Q2258.64 187.114 2259.3 187.114 Q2261.03 187.114 2261.94 185.951 Q2262.86 184.779 2262.99 182.401 Q2262.49 183.147 2261.72 183.546 Q2260.94 183.946 2260.01 183.946 Q2258.06 183.946 2256.93 182.774 Q2255.8 181.593 2255.8 179.553 Q2255.8 177.557 2256.98 176.35 Q2258.16 175.144 2260.12 175.144 Q2262.37 175.144 2263.55 176.871 Q2264.74 178.59 2264.74 181.871 Q2264.74 184.935 2263.28 186.767 Q2261.83 188.59 2259.37 188.59 Q2258.71 188.59 2258.04 188.46 Q2257.36 188.329 2256.63 188.069 M2260.12 182.574 Q2261.3 182.574 2261.99 181.767 Q2262.68 180.96 2262.68 179.553 Q2262.68 178.156 2261.99 177.349 Q2261.3 176.533 2260.12 176.533 Q2258.94 176.533 2258.25 177.349 Q2257.56 178.156 2257.56 179.553 Q2257.56 180.96 2258.25 181.767 Q2258.94 182.574 2260.12 182.574 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2284.78 170.191 L2291.67 170.191 L2291.67 171.667 L2286.39 171.667 L2286.39 174.844 Q2286.77 174.714 2287.15 174.653 Q2287.54 174.583 2287.92 174.583 Q2290.09 174.583 2291.35 175.773 Q2292.62 176.962 2292.62 178.993 Q2292.62 181.085 2291.32 182.248 Q2290.02 183.403 2287.65 183.403 Q2286.83 183.403 2285.98 183.264 Q2285.14 183.125 2284.24 182.847 L2284.24 181.085 Q2285.02 181.51 2285.85 181.719 Q2286.68 181.927 2287.61 181.927 Q2289.12 181.927 2289.99 181.137 Q2290.87 180.347 2290.87 178.993 Q2290.87 177.639 2289.99 176.849 Q2289.12 176.059 2287.61 176.059 Q2286.91 176.059 2286.21 176.215 Q2285.51 176.372 2284.78 176.701 L2284.78 170.191 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2299.83 171.346 Q2298.47 171.346 2297.79 172.682 Q2297.11 174.01 2297.11 176.684 Q2297.11 179.349 2297.79 180.686 Q2298.47 182.014 2299.83 182.014 Q2301.19 182.014 2301.87 180.686 Q2302.55 179.349 2302.55 176.684 Q2302.55 174.01 2301.87 172.682 Q2301.19 171.346 2299.83 171.346 M2299.83 169.957 Q2302.01 169.957 2303.15 171.684 Q2304.31 173.403 2304.31 176.684 Q2304.31 179.957 2303.15 181.684 Q2302.01 183.403 2299.83 183.403 Q2297.65 183.403 2296.49 181.684 Q2295.35 179.957 2295.35 176.684 Q2295.35 173.403 2296.49 171.684 Q2297.65 169.957 2299.83 169.957 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M286.311 2174.14 L2352.76 2174.14 L2352.76 1362.92 L286.311 1362.92  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="344.796,964.927 344.796,153.712 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="912.471,964.927 912.471,153.712 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1480.15,964.927 1480.15,153.712 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2047.82,964.927 2047.82,153.712 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,964.927 2352.76,964.927 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="344.796,964.927 344.796,946.03 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="912.471,964.927 912.471,946.03 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1480.15,964.927 1480.15,946.03 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2047.82,964.927 2047.82,946.03 "/>
+<path clip-path="url(#clip910)" d="M344.796 995.846 Q341.184 995.846 339.356 999.411 Q337.55 1002.95 337.55 1010.08 Q337.55 1017.19 339.356 1020.75 Q341.184 1024.3 344.796 1024.3 Q348.43 1024.3 350.235 1020.75 Q352.064 1017.19 352.064 1010.08 Q352.064 1002.95 350.235 999.411 Q348.43 995.846 344.796 995.846 M344.796 992.142 Q350.606 992.142 353.661 996.749 Q356.74 1001.33 356.74 1010.08 Q356.74 1018.81 353.661 1023.42 Q350.606 1028 344.796 1028 Q338.985 1028 335.907 1023.42 Q332.851 1018.81 332.851 1010.08 Q332.851 1001.33 335.907 996.749 Q338.985 992.142 344.796 992.142 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M823.455 1039.05 L839.775 1039.05 L839.775 1042.99 L817.83 1042.99 L817.83 1039.05 Q820.492 1036.3 825.076 1031.67 Q829.682 1027.02 830.863 1025.67 Q833.108 1023.15 833.988 1021.41 Q834.89 1019.65 834.89 1017.96 Q834.89 1015.21 832.946 1013.47 Q831.025 1011.74 827.923 1011.74 Q825.724 1011.74 823.27 1012.5 Q820.84 1013.27 818.062 1014.82 L818.062 1010.09 Q820.886 1008.96 823.34 1008.38 Q825.793 1007.8 827.83 1007.8 Q833.201 1007.8 836.395 1010.49 Q839.589 1013.17 839.589 1017.66 Q839.589 1019.79 838.779 1021.71 Q837.992 1023.61 835.886 1026.21 Q835.307 1026.88 832.205 1030.09 Q829.103 1033.29 823.455 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M849.589 1037.11 L854.474 1037.11 L854.474 1042.99 L849.589 1042.99 L849.589 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M874.659 1011.51 Q871.048 1011.51 869.219 1015.07 Q867.413 1018.61 867.413 1025.74 Q867.413 1032.85 869.219 1036.41 Q871.048 1039.96 874.659 1039.96 Q878.293 1039.96 880.098 1036.41 Q881.927 1032.85 881.927 1025.74 Q881.927 1018.61 880.098 1015.07 Q878.293 1011.51 874.659 1011.51 M874.659 1007.8 Q880.469 1007.8 883.524 1012.41 Q886.603 1016.99 886.603 1025.74 Q886.603 1034.47 883.524 1039.08 Q880.469 1043.66 874.659 1043.66 Q868.849 1043.66 865.77 1039.08 Q862.714 1034.47 862.714 1025.74 Q862.714 1016.99 865.77 1012.41 Q868.849 1007.8 874.659 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M922.992 1017.52 L912.413 1028.15 L922.992 1038.73 L920.237 1041.53 L909.612 1030.9 L898.987 1041.53 L896.256 1038.73 L906.811 1028.15 L896.256 1017.52 L898.987 1014.72 L909.612 1025.35 L920.237 1014.72 L922.992 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M935.353 1039.05 L942.992 1039.05 L942.992 1012.69 L934.682 1014.35 L934.682 1010.09 L942.945 1008.43 L947.621 1008.43 L947.621 1039.05 L955.26 1039.05 L955.26 1042.99 L935.353 1042.99 L935.353 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M974.704 1011.51 Q971.093 1011.51 969.265 1015.07 Q967.459 1018.61 967.459 1025.74 Q967.459 1032.85 969.265 1036.41 Q971.093 1039.96 974.704 1039.96 Q978.339 1039.96 980.144 1036.41 Q981.973 1032.85 981.973 1025.74 Q981.973 1018.61 980.144 1015.07 Q978.339 1011.51 974.704 1011.51 M974.704 1007.8 Q980.515 1007.8 983.57 1012.41 Q986.649 1016.99 986.649 1025.74 Q986.649 1034.47 983.57 1039.08 Q980.515 1043.66 974.704 1043.66 Q968.894 1043.66 965.816 1039.08 Q962.76 1034.47 962.76 1025.74 Q962.76 1016.99 965.816 1012.41 Q968.894 1007.8 974.704 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M999.325 990.807 L989.733 1005.8 L999.325 1005.8 L999.325 990.807 M998.328 987.497 L1003.11 987.497 L1003.11 1005.8 L1007.11 1005.8 L1007.11 1008.96 L1003.11 1008.96 L1003.11 1015.58 L999.325 1015.58 L999.325 1008.96 L986.649 1008.96 L986.649 1005.29 L998.328 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1400.53 1012.5 L1388.72 1030.95 L1400.53 1030.95 L1400.53 1012.5 M1399.3 1008.43 L1405.18 1008.43 L1405.18 1030.95 L1410.11 1030.95 L1410.11 1034.84 L1405.18 1034.84 L1405.18 1042.99 L1400.53 1042.99 L1400.53 1034.84 L1384.93 1034.84 L1384.93 1030.33 L1399.3 1008.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1417.84 1037.11 L1422.73 1037.11 L1422.73 1042.99 L1417.84 1042.99 L1417.84 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1442.91 1011.51 Q1439.3 1011.51 1437.47 1015.07 Q1435.67 1018.61 1435.67 1025.74 Q1435.67 1032.85 1437.47 1036.41 Q1439.3 1039.96 1442.91 1039.96 Q1446.55 1039.96 1448.35 1036.41 Q1450.18 1032.85 1450.18 1025.74 Q1450.18 1018.61 1448.35 1015.07 Q1446.55 1011.51 1442.91 1011.51 M1442.91 1007.8 Q1448.72 1007.8 1451.78 1012.41 Q1454.86 1016.99 1454.86 1025.74 Q1454.86 1034.47 1451.78 1039.08 Q1448.72 1043.66 1442.91 1043.66 Q1437.1 1043.66 1434.02 1039.08 Q1430.97 1034.47 1430.97 1025.74 Q1430.97 1016.99 1434.02 1012.41 Q1437.1 1007.8 1442.91 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1491.25 1017.52 L1480.67 1028.15 L1491.25 1038.73 L1488.49 1041.53 L1477.87 1030.9 L1467.24 1041.53 L1464.51 1038.73 L1475.07 1028.15 L1464.51 1017.52 L1467.24 1014.72 L1477.87 1025.35 L1488.49 1014.72 L1491.25 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1503.61 1039.05 L1511.25 1039.05 L1511.25 1012.69 L1502.94 1014.35 L1502.94 1010.09 L1511.2 1008.43 L1515.88 1008.43 L1515.88 1039.05 L1523.51 1039.05 L1523.51 1042.99 L1503.61 1042.99 L1503.61 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1542.96 1011.51 Q1539.35 1011.51 1537.52 1015.07 Q1535.71 1018.61 1535.71 1025.74 Q1535.71 1032.85 1537.52 1036.41 Q1539.35 1039.96 1542.96 1039.96 Q1546.59 1039.96 1548.4 1036.41 Q1550.23 1032.85 1550.23 1025.74 Q1550.23 1018.61 1548.4 1015.07 Q1546.59 1011.51 1542.96 1011.51 M1542.96 1007.8 Q1548.77 1007.8 1551.82 1012.41 Q1554.9 1016.99 1554.9 1025.74 Q1554.9 1034.47 1551.82 1039.08 Q1548.77 1043.66 1542.96 1043.66 Q1537.15 1043.66 1534.07 1039.08 Q1531.01 1034.47 1531.01 1025.74 Q1531.01 1016.99 1534.07 1012.41 Q1537.15 1007.8 1542.96 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1567.58 990.807 L1557.99 1005.8 L1567.58 1005.8 L1567.58 990.807 M1566.58 987.497 L1571.36 987.497 L1571.36 1005.8 L1575.37 1005.8 L1575.37 1008.96 L1571.36 1008.96 L1571.36 1015.58 L1567.58 1015.58 L1567.58 1008.96 L1554.9 1008.96 L1554.9 1005.29 L1566.58 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1965.44 1023.84 Q1962.29 1023.84 1960.44 1026 Q1958.61 1028.15 1958.61 1031.9 Q1958.61 1035.63 1960.44 1037.8 Q1962.29 1039.96 1965.44 1039.96 Q1968.59 1039.96 1970.41 1037.8 Q1972.27 1035.63 1972.27 1031.9 Q1972.27 1028.15 1970.41 1026 Q1968.59 1023.84 1965.44 1023.84 M1974.72 1009.19 L1974.72 1013.45 Q1972.96 1012.62 1971.16 1012.18 Q1969.37 1011.74 1967.61 1011.74 Q1962.98 1011.74 1960.53 1014.86 Q1958.1 1017.99 1957.75 1024.31 Q1959.12 1022.29 1961.18 1021.23 Q1963.24 1020.14 1965.72 1020.14 Q1970.92 1020.14 1973.93 1023.31 Q1976.97 1026.46 1976.97 1031.9 Q1976.97 1037.22 1973.82 1040.44 Q1970.67 1043.66 1965.44 1043.66 Q1959.44 1043.66 1956.27 1039.08 Q1953.1 1034.47 1953.1 1025.74 Q1953.1 1017.55 1956.99 1012.69 Q1960.88 1007.8 1967.43 1007.8 Q1969.19 1007.8 1970.97 1008.15 Q1972.78 1008.5 1974.72 1009.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1985.02 1037.11 L1989.91 1037.11 L1989.91 1042.99 L1985.02 1042.99 L1985.02 1037.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2010.09 1011.51 Q2006.48 1011.51 2004.65 1015.07 Q2002.85 1018.61 2002.85 1025.74 Q2002.85 1032.85 2004.65 1036.41 Q2006.48 1039.96 2010.09 1039.96 Q2013.72 1039.96 2015.53 1036.41 Q2017.36 1032.85 2017.36 1025.74 Q2017.36 1018.61 2015.53 1015.07 Q2013.72 1011.51 2010.09 1011.51 M2010.09 1007.8 Q2015.9 1007.8 2018.96 1012.41 Q2022.04 1016.99 2022.04 1025.74 Q2022.04 1034.47 2018.96 1039.08 Q2015.9 1043.66 2010.09 1043.66 Q2004.28 1043.66 2001.2 1039.08 Q1998.15 1034.47 1998.15 1025.74 Q1998.15 1016.99 2001.2 1012.41 Q2004.28 1007.8 2010.09 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2058.42 1017.52 L2047.85 1028.15 L2058.42 1038.73 L2055.67 1041.53 L2045.04 1030.9 L2034.42 1041.53 L2031.69 1038.73 L2042.24 1028.15 L2031.69 1017.52 L2034.42 1014.72 L2045.04 1025.35 L2055.67 1014.72 L2058.42 1017.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2070.78 1039.05 L2078.42 1039.05 L2078.42 1012.69 L2070.11 1014.35 L2070.11 1010.09 L2078.38 1008.43 L2083.05 1008.43 L2083.05 1039.05 L2090.69 1039.05 L2090.69 1042.99 L2070.78 1042.99 L2070.78 1039.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2110.14 1011.51 Q2106.53 1011.51 2104.7 1015.07 Q2102.89 1018.61 2102.89 1025.74 Q2102.89 1032.85 2104.7 1036.41 Q2106.53 1039.96 2110.14 1039.96 Q2113.77 1039.96 2115.58 1036.41 Q2117.4 1032.85 2117.4 1025.74 Q2117.4 1018.61 2115.58 1015.07 Q2113.77 1011.51 2110.14 1011.51 M2110.14 1007.8 Q2115.95 1007.8 2119 1012.41 Q2122.08 1016.99 2122.08 1025.74 Q2122.08 1034.47 2119 1039.08 Q2115.95 1043.66 2110.14 1043.66 Q2104.33 1043.66 2101.25 1039.08 Q2098.19 1034.47 2098.19 1025.74 Q2098.19 1016.99 2101.25 1012.41 Q2104.33 1007.8 2110.14 1007.8 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2134.76 990.807 L2125.17 1005.8 L2134.76 1005.8 L2134.76 990.807 M2133.76 987.497 L2138.54 987.497 L2138.54 1005.8 L2142.54 1005.8 L2142.54 1008.96 L2138.54 1008.96 L2138.54 1015.58 L2134.76 1015.58 L2134.76 1008.96 L2122.08 1008.96 L2122.08 1005.29 L2133.76 987.497 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1198.87 1128.22 L1180.73 1080.7 L1187.45 1080.7 L1202.5 1120.7 L1217.59 1080.7 L1224.27 1080.7 L1206.16 1128.22 L1198.87 1128.22 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1242.09 1110.3 Q1235 1110.3 1232.26 1111.92 Q1229.52 1113.54 1229.52 1117.46 Q1229.52 1120.58 1231.56 1122.42 Q1233.63 1124.24 1237.16 1124.24 Q1242.03 1124.24 1244.96 1120.8 Q1247.92 1117.33 1247.92 1111.6 L1247.92 1110.3 L1242.09 1110.3 M1253.78 1107.88 L1253.78 1128.22 L1247.92 1128.22 L1247.92 1122.81 Q1245.91 1126.05 1242.92 1127.61 Q1239.93 1129.14 1235.6 1129.14 Q1230.13 1129.14 1226.88 1126.08 Q1223.67 1123 1223.67 1117.84 Q1223.67 1111.82 1227.68 1108.77 Q1231.72 1105.71 1239.71 1105.71 L1247.92 1105.71 L1247.92 1105.14 Q1247.92 1101.1 1245.25 1098.9 Q1242.6 1096.67 1237.8 1096.67 Q1234.74 1096.67 1231.85 1097.41 Q1228.95 1098.14 1226.28 1099.6 L1226.28 1094.19 Q1229.49 1092.95 1232.51 1092.35 Q1235.54 1091.71 1238.4 1091.71 Q1246.14 1091.71 1249.96 1095.72 Q1253.78 1099.73 1253.78 1107.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1286.5 1098.04 Q1285.51 1097.47 1284.33 1097.22 Q1283.19 1096.93 1281.78 1096.93 Q1276.82 1096.93 1274.15 1100.18 Q1271.5 1103.39 1271.5 1109.44 L1271.5 1128.22 L1265.62 1128.22 L1265.62 1092.57 L1271.5 1092.57 L1271.5 1098.11 Q1273.35 1094.86 1276.31 1093.3 Q1279.27 1091.71 1283.5 1091.71 Q1284.11 1091.71 1284.84 1091.8 Q1285.57 1091.87 1286.46 1092.03 L1286.5 1098.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1292.64 1092.57 L1298.49 1092.57 L1298.49 1128.22 L1292.64 1128.22 L1292.64 1092.57 M1292.64 1078.69 L1298.49 1078.69 L1298.49 1086.11 L1292.64 1086.11 L1292.64 1078.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1326.95 1110.3 Q1319.85 1110.3 1317.11 1111.92 Q1314.38 1113.54 1314.38 1117.46 Q1314.38 1120.58 1316.41 1122.42 Q1318.48 1124.24 1322.02 1124.24 Q1326.89 1124.24 1329.81 1120.8 Q1332.77 1117.33 1332.77 1111.6 L1332.77 1110.3 L1326.95 1110.3 M1338.63 1107.88 L1338.63 1128.22 L1332.77 1128.22 L1332.77 1122.81 Q1330.77 1126.05 1327.78 1127.61 Q1324.79 1129.14 1320.46 1129.14 Q1314.98 1129.14 1311.74 1126.08 Q1308.52 1123 1308.52 1117.84 Q1308.52 1111.82 1312.53 1108.77 Q1316.57 1105.71 1324.56 1105.71 L1332.77 1105.71 L1332.77 1105.14 Q1332.77 1101.1 1330.1 1098.9 Q1327.46 1096.67 1322.65 1096.67 Q1319.6 1096.67 1316.7 1097.41 Q1313.8 1098.14 1311.13 1099.6 L1311.13 1094.19 Q1314.35 1092.95 1317.37 1092.35 Q1320.39 1091.71 1323.26 1091.71 Q1330.99 1091.71 1334.81 1095.72 Q1338.63 1099.73 1338.63 1107.88 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1380.33 1106.7 L1380.33 1128.22 L1374.47 1128.22 L1374.47 1106.89 Q1374.47 1101.83 1372.5 1099.32 Q1370.52 1096.8 1366.58 1096.8 Q1361.83 1096.8 1359.1 1099.83 Q1356.36 1102.85 1356.36 1108.07 L1356.36 1128.22 L1350.47 1128.22 L1350.47 1092.57 L1356.36 1092.57 L1356.36 1098.11 Q1358.46 1094.89 1361.29 1093.3 Q1364.16 1091.71 1367.88 1091.71 Q1374.02 1091.71 1377.18 1095.53 Q1380.33 1099.32 1380.33 1106.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1417.66 1093.94 L1417.66 1099.41 Q1415.18 1098.04 1412.66 1097.37 Q1410.18 1096.67 1407.63 1096.67 Q1401.94 1096.67 1398.79 1100.3 Q1395.64 1103.9 1395.64 1110.42 Q1395.64 1116.95 1398.79 1120.58 Q1401.94 1124.17 1407.63 1124.17 Q1410.18 1124.17 1412.66 1123.51 Q1415.18 1122.81 1417.66 1121.44 L1417.66 1126.85 Q1415.21 1127.99 1412.57 1128.57 Q1409.96 1129.14 1407 1129.14 Q1398.95 1129.14 1394.2 1124.08 Q1389.46 1119.02 1389.46 1110.42 Q1389.46 1101.7 1394.24 1096.71 Q1399.04 1091.71 1407.38 1091.71 Q1410.09 1091.71 1412.66 1092.28 Q1415.24 1092.82 1417.66 1093.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1458.34 1108.93 L1458.34 1111.79 L1431.41 1111.79 Q1431.79 1117.84 1435.04 1121.02 Q1438.32 1124.17 1444.14 1124.17 Q1447.52 1124.17 1450.67 1123.35 Q1453.85 1122.52 1456.97 1120.86 L1456.97 1126.4 Q1453.82 1127.74 1450.51 1128.44 Q1447.2 1129.14 1443.79 1129.14 Q1435.26 1129.14 1430.27 1124.17 Q1425.3 1119.21 1425.3 1110.74 Q1425.3 1101.99 1430.01 1096.87 Q1434.75 1091.71 1442.77 1091.71 Q1449.97 1091.71 1454.14 1096.36 Q1458.34 1100.97 1458.34 1108.93 M1452.48 1107.21 Q1452.42 1102.4 1449.78 1099.54 Q1447.17 1096.67 1442.84 1096.67 Q1437.94 1096.67 1434.98 1099.44 Q1432.05 1102.21 1431.6 1107.24 L1452.48 1107.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,941.969 2352.76,941.969 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,734.335 2352.76,734.335 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,526.7 2352.76,526.7 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,319.066 2352.76,319.066 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,964.927 286.311,153.712 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,941.969 305.209,941.969 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,734.335 305.209,734.335 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,526.7 305.209,526.7 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,319.066 305.209,319.066 "/>
+<path clip-path="url(#clip910)" d="M238.367 927.768 Q234.756 927.768 232.927 931.333 Q231.122 934.874 231.122 942.004 Q231.122 949.11 232.927 952.675 Q234.756 956.217 238.367 956.217 Q242.001 956.217 243.807 952.675 Q245.635 949.11 245.635 942.004 Q245.635 934.874 243.807 931.333 Q242.001 927.768 238.367 927.768 M238.367 924.064 Q244.177 924.064 247.233 928.671 Q250.311 933.254 250.311 942.004 Q250.311 950.731 247.233 955.337 Q244.177 959.92 238.367 959.92 Q232.557 959.92 229.478 955.337 Q226.422 950.731 226.422 942.004 Q226.422 933.254 229.478 928.671 Q232.557 924.064 238.367 924.064 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M202.233 747.679 L218.552 747.679 L218.552 751.615 L196.608 751.615 L196.608 747.679 Q199.27 744.925 203.853 740.295 Q208.46 735.643 209.64 734.3 Q211.886 731.777 212.765 730.041 Q213.668 728.281 213.668 726.592 Q213.668 723.837 211.723 722.101 Q209.802 720.365 206.7 720.365 Q204.501 720.365 202.048 721.129 Q199.617 721.893 196.839 723.444 L196.839 718.721 Q199.663 717.587 202.117 717.008 Q204.571 716.43 206.608 716.43 Q211.978 716.43 215.173 719.115 Q218.367 721.8 218.367 726.291 Q218.367 728.42 217.557 730.342 Q216.77 732.24 214.663 734.832 Q214.085 735.504 210.983 738.721 Q207.881 741.916 202.233 747.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 720.133 Q234.756 720.133 232.927 723.698 Q231.122 727.24 231.122 734.369 Q231.122 741.476 232.927 745.041 Q234.756 748.582 238.367 748.582 Q242.001 748.582 243.807 745.041 Q245.635 741.476 245.635 734.369 Q245.635 727.24 243.807 723.698 Q242.001 720.133 238.367 720.133 M238.367 716.43 Q244.177 716.43 247.233 721.036 Q250.311 725.619 250.311 734.369 Q250.311 743.096 247.233 747.703 Q244.177 752.286 238.367 752.286 Q232.557 752.286 229.478 747.703 Q226.422 743.096 226.422 734.369 Q226.422 725.619 229.478 721.036 Q232.557 716.43 238.367 716.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M211.052 513.494 L199.247 531.943 L211.052 531.943 L211.052 513.494 M209.825 509.42 L215.705 509.42 L215.705 531.943 L220.635 531.943 L220.635 535.832 L215.705 535.832 L215.705 543.98 L211.052 543.98 L211.052 535.832 L195.45 535.832 L195.45 531.318 L209.825 509.42 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 512.499 Q234.756 512.499 232.927 516.064 Q231.122 519.605 231.122 526.735 Q231.122 533.841 232.927 537.406 Q234.756 540.948 238.367 540.948 Q242.001 540.948 243.807 537.406 Q245.635 533.841 245.635 526.735 Q245.635 519.605 243.807 516.064 Q242.001 512.499 238.367 512.499 M238.367 508.795 Q244.177 508.795 247.233 513.402 Q250.311 517.985 250.311 526.735 Q250.311 535.462 247.233 540.068 Q244.177 544.652 238.367 544.652 Q232.557 544.652 229.478 540.068 Q226.422 535.462 226.422 526.735 Q226.422 517.985 229.478 513.402 Q232.557 508.795 238.367 508.795 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M208.784 317.202 Q205.636 317.202 203.784 319.355 Q201.955 321.508 201.955 325.258 Q201.955 328.985 203.784 331.161 Q205.636 333.313 208.784 333.313 Q211.932 333.313 213.76 331.161 Q215.612 328.985 215.612 325.258 Q215.612 321.508 213.76 319.355 Q211.932 317.202 208.784 317.202 M218.066 302.55 L218.066 306.809 Q216.307 305.976 214.501 305.536 Q212.719 305.096 210.96 305.096 Q206.33 305.096 203.876 308.221 Q201.446 311.346 201.099 317.665 Q202.464 315.651 204.524 314.587 Q206.585 313.499 209.061 313.499 Q214.27 313.499 217.279 316.67 Q220.311 319.818 220.311 325.258 Q220.311 330.582 217.163 333.8 Q214.015 337.017 208.784 337.017 Q202.788 337.017 199.617 332.434 Q196.446 327.827 196.446 319.101 Q196.446 310.906 200.335 306.045 Q204.224 301.161 210.774 301.161 Q212.534 301.161 214.316 301.508 Q216.122 301.855 218.066 302.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 304.864 Q234.756 304.864 232.927 308.429 Q231.122 311.971 231.122 319.101 Q231.122 326.207 232.927 329.772 Q234.756 333.313 238.367 333.313 Q242.001 333.313 243.807 329.772 Q245.635 326.207 245.635 319.101 Q245.635 311.971 243.807 308.429 Q242.001 304.864 238.367 304.864 M238.367 301.161 Q244.177 301.161 247.233 305.767 Q250.311 310.351 250.311 319.101 Q250.311 327.827 247.233 332.434 Q244.177 337.017 238.367 337.017 Q232.557 337.017 229.478 332.434 Q226.422 327.827 226.422 319.101 Q226.422 310.351 229.478 305.767 Q232.557 301.161 238.367 301.161 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M97.6744 823.926 L97.6744 793.88 L103.085 793.88 L103.085 817.497 L117.153 817.497 L117.153 794.867 L122.564 794.867 L122.564 817.497 L139.784 817.497 L139.784 793.307 L145.194 793.307 L145.194 823.926 L97.6744 823.926 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M109.546 753.362 L126.893 766.253 L145.194 752.694 L145.194 759.601 L131.19 769.977 L145.194 780.353 L145.194 787.26 L126.543 773.414 L109.546 786.082 L109.546 779.175 L122.246 769.722 L109.546 760.269 L109.546 753.362 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M139.847 738.753 L158.753 738.753 L158.753 744.641 L109.546 744.641 L109.546 738.753 L114.957 738.753 Q111.774 736.907 110.247 734.106 Q108.687 731.273 108.687 727.359 Q108.687 720.866 113.843 716.823 Q119 712.749 127.402 712.749 Q135.805 712.749 140.961 716.823 Q146.117 720.866 146.117 727.359 Q146.117 731.273 144.59 734.106 Q143.03 736.907 139.847 738.753 M127.402 718.828 Q120.941 718.828 117.281 721.502 Q113.589 724.144 113.589 728.791 Q113.589 733.438 117.281 736.111 Q120.941 738.753 127.402 738.753 Q133.863 738.753 137.556 736.111 Q141.216 733.438 141.216 728.791 Q141.216 724.144 137.556 721.502 Q133.863 718.828 127.402 718.828 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M125.906 672.55 L128.771 672.55 L128.771 699.477 Q134.818 699.095 138.001 695.848 Q141.152 692.57 141.152 686.745 Q141.152 683.371 140.325 680.22 Q139.497 677.038 137.842 673.918 L143.38 673.918 Q144.717 677.069 145.417 680.38 Q146.117 683.69 146.117 687.095 Q146.117 695.625 141.152 700.623 Q136.187 705.588 127.721 705.588 Q118.968 705.588 113.843 700.877 Q108.687 696.135 108.687 688.114 Q108.687 680.921 113.334 676.751 Q117.949 672.55 125.906 672.55 M124.188 678.406 Q119.381 678.47 116.517 681.112 Q113.652 683.722 113.652 688.05 Q113.652 692.952 116.421 695.912 Q119.19 698.84 124.219 699.286 L124.188 678.406 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M110.915 637.284 L116.39 637.284 Q115.021 639.766 114.353 642.281 Q113.652 644.763 113.652 647.31 Q113.652 653.007 117.281 656.158 Q120.877 659.309 127.402 659.309 Q133.927 659.309 137.556 656.158 Q141.152 653.007 141.152 647.31 Q141.152 644.763 140.484 642.281 Q139.784 639.766 138.415 637.284 L143.826 637.284 Q144.972 639.735 145.545 642.376 Q146.117 644.986 146.117 647.946 Q146.117 655.999 141.057 660.741 Q135.996 665.484 127.402 665.484 Q118.681 665.484 113.684 660.71 Q108.687 655.903 108.687 647.564 Q108.687 644.859 109.26 642.281 Q109.801 639.703 110.915 637.284 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M99.425 621.306 L109.546 621.306 L109.546 609.243 L114.098 609.243 L114.098 621.306 L133.45 621.306 Q137.81 621.306 139.051 620.128 Q140.293 618.919 140.293 615.258 L140.293 609.243 L145.194 609.243 L145.194 615.258 Q145.194 622.038 142.68 624.616 Q140.134 627.194 133.45 627.194 L114.098 627.194 L114.098 631.491 L109.546 631.491 L109.546 627.194 L99.425 627.194 L99.425 621.306 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M125.906 571.049 L128.771 571.049 L128.771 597.976 Q134.818 597.594 138.001 594.347 Q141.152 591.069 141.152 585.244 Q141.152 581.87 140.325 578.719 Q139.497 575.536 137.842 572.417 L143.38 572.417 Q144.717 575.568 145.417 578.878 Q146.117 582.189 146.117 585.594 Q146.117 594.124 141.152 599.121 Q136.187 604.087 127.721 604.087 Q118.968 604.087 113.843 599.376 Q108.687 594.634 108.687 586.613 Q108.687 579.42 113.334 575.25 Q117.949 571.049 125.906 571.049 M124.188 576.905 Q119.381 576.969 116.517 579.61 Q113.652 582.22 113.652 586.549 Q113.652 591.451 116.421 594.411 Q119.19 597.339 124.219 597.785 L124.188 576.905 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M114.957 537.979 L95.6692 537.979 L95.6692 532.122 L145.194 532.122 L145.194 537.979 L139.847 537.979 Q143.03 539.825 144.59 542.658 Q146.117 545.458 146.117 549.405 Q146.117 555.866 140.961 559.94 Q135.805 563.983 127.402 563.983 Q119 563.983 113.843 559.94 Q108.687 555.866 108.687 549.405 Q108.687 545.458 110.247 542.658 Q111.774 539.825 114.957 537.979 M127.402 557.935 Q133.863 557.935 137.556 555.293 Q141.216 552.62 141.216 547.973 Q141.216 543.326 137.556 540.652 Q133.863 537.979 127.402 537.979 Q120.941 537.979 117.281 540.652 Q113.589 543.326 113.589 547.973 Q113.589 552.62 117.281 555.293 Q120.941 557.935 127.402 557.935 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M122.914 476.55 Q123.615 474.481 125.906 472.539 Q128.198 470.566 132.208 468.593 L145.194 462.068 L145.194 468.974 L133.004 475.054 Q128.23 477.409 126.67 479.637 Q125.111 481.833 125.111 485.653 L125.111 492.655 L145.194 492.655 L145.194 499.084 L97.6744 499.084 L97.6744 484.57 Q97.6744 476.422 101.08 472.412 Q104.486 468.402 111.361 468.402 Q115.848 468.402 118.809 470.502 Q121.769 472.571 122.914 476.55 M102.958 492.655 L119.827 492.655 L119.827 484.57 Q119.827 479.924 117.695 477.568 Q115.53 475.181 111.361 475.181 Q107.191 475.181 105.09 477.568 Q102.958 479.924 102.958 484.57 L102.958 492.655 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M125.906 426.483 L128.771 426.483 L128.771 453.41 Q134.818 453.028 138.001 449.782 Q141.152 446.504 141.152 440.679 Q141.152 437.305 140.325 434.154 Q139.497 430.971 137.842 427.852 L143.38 427.852 Q144.717 431.003 145.417 434.313 Q146.117 437.623 146.117 441.029 Q146.117 449.559 141.152 454.556 Q136.187 459.521 127.721 459.521 Q118.968 459.521 113.843 454.811 Q108.687 450.068 108.687 442.048 Q108.687 434.854 113.334 430.685 Q117.949 426.483 125.906 426.483 M124.188 432.34 Q119.381 432.404 116.517 435.045 Q113.652 437.655 113.652 441.984 Q113.652 446.885 116.421 449.846 Q119.19 452.774 124.219 453.219 L124.188 432.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M99.425 411.078 L109.546 411.078 L109.546 399.015 L114.098 399.015 L114.098 411.078 L133.45 411.078 Q137.81 411.078 139.051 409.901 Q140.293 408.691 140.293 405.031 L140.293 399.015 L145.194 399.015 L145.194 405.031 Q145.194 411.81 142.68 414.389 Q140.134 416.967 133.45 416.967 L114.098 416.967 L114.098 421.264 L109.546 421.264 L109.546 416.967 L99.425 416.967 L99.425 411.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M131.126 391.918 L109.546 391.918 L109.546 386.061 L130.903 386.061 Q135.964 386.061 138.51 384.088 Q141.025 382.114 141.025 378.168 Q141.025 373.425 138.001 370.688 Q134.977 367.919 129.758 367.919 L109.546 367.919 L109.546 362.062 L145.194 362.062 L145.194 367.919 L139.72 367.919 Q142.966 370.051 144.558 372.884 Q146.117 375.685 146.117 379.409 Q146.117 385.552 142.298 388.735 Q138.479 391.918 131.126 391.918 M108.687 377.181 L108.687 377.181 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M115.021 329.343 Q114.448 330.329 114.193 331.507 Q113.907 332.653 113.907 334.053 Q113.907 339.019 117.153 341.692 Q120.368 344.334 126.416 344.334 L145.194 344.334 L145.194 350.222 L109.546 350.222 L109.546 344.334 L115.085 344.334 Q111.838 342.488 110.279 339.528 Q108.687 336.568 108.687 332.335 Q108.687 331.73 108.783 330.998 Q108.846 330.266 109.005 329.375 L115.021 329.343 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M123.678 294.713 L145.194 294.713 L145.194 300.57 L123.869 300.57 Q118.809 300.57 116.294 302.543 Q113.78 304.516 113.78 308.463 Q113.78 313.206 116.803 315.943 Q119.827 318.68 125.047 318.68 L145.194 318.68 L145.194 324.568 L109.546 324.568 L109.546 318.68 L115.085 318.68 Q111.87 316.579 110.279 313.747 Q108.687 310.882 108.687 307.158 Q108.687 301.015 112.507 297.864 Q116.294 294.713 123.678 294.713 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1020.54 17.6457 Q1011.62 17.6457 1006.36 24.2892 Q1001.13 30.9327 1001.13 42.3968 Q1001.13 53.8203 1006.36 60.4638 Q1011.62 67.1073 1020.54 67.1073 Q1029.45 67.1073 1034.63 60.4638 Q1039.86 53.8203 1039.86 42.3968 Q1039.86 30.9327 1034.63 24.2892 Q1029.45 17.6457 1020.54 17.6457 M1020.54 11.0023 Q1033.26 11.0023 1040.87 19.5497 Q1048.49 28.0566 1048.49 42.3968 Q1048.49 56.6965 1040.87 65.2439 Q1033.26 73.7508 1020.54 73.7508 Q1007.78 73.7508 1000.12 65.2439 Q992.504 56.737 992.504 42.3968 Q992.504 28.0566 1000.12 19.5497 Q1007.78 11.0023 1020.54 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1093.53 49.9314 Q1093.53 41.7081 1090.13 37.0496 Q1086.77 32.3505 1080.85 32.3505 Q1074.94 32.3505 1071.54 37.0496 Q1068.17 41.7081 1068.17 49.9314 Q1068.17 58.1548 1071.54 62.8538 Q1074.94 67.5124 1080.85 67.5124 Q1086.77 67.5124 1090.13 62.8538 Q1093.53 58.1548 1093.53 49.9314 M1068.17 34.0924 Q1070.52 30.0415 1074.09 28.0971 Q1077.69 26.1121 1082.68 26.1121 Q1090.94 26.1121 1096.09 32.6746 Q1101.27 39.2371 1101.27 49.9314 Q1101.27 60.6258 1096.09 67.1883 Q1090.94 73.7508 1082.68 73.7508 Q1077.69 73.7508 1074.09 71.8063 Q1070.52 69.8214 1068.17 65.7705 L1068.17 72.576 L1060.68 72.576 L1060.68 9.54393 L1068.17 9.54393 L1068.17 34.0924 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1113.63 27.2059 L1121.08 27.2059 L1121.08 73.3862 Q1121.08 82.0551 1117.76 85.944 Q1114.48 89.8329 1107.14 89.8329 L1104.31 89.8329 L1104.31 83.5134 L1106.29 83.5134 Q1110.55 83.5134 1112.09 81.5285 Q1113.63 79.5841 1113.63 73.3862 L1113.63 27.2059 M1113.63 9.54393 L1121.08 9.54393 L1121.08 18.9825 L1113.63 18.9825 L1113.63 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1175.48 48.0275 L1175.48 51.6733 L1141.21 51.6733 Q1141.7 59.3701 1145.83 63.421 Q1150 67.4314 1157.42 67.4314 Q1161.71 67.4314 1165.72 66.3781 Q1169.77 65.3249 1173.74 63.2184 L1173.74 70.267 Q1169.73 71.9684 1165.52 72.8596 Q1161.31 73.7508 1156.97 73.7508 Q1146.11 73.7508 1139.75 67.4314 Q1133.44 61.1119 1133.44 50.3365 Q1133.44 39.1965 1139.43 32.6746 Q1145.47 26.1121 1155.67 26.1121 Q1164.83 26.1121 1170.14 32.0264 Q1175.48 37.9003 1175.48 48.0275 M1168.03 45.84 Q1167.95 39.7232 1164.59 36.0774 Q1161.26 32.4315 1155.76 32.4315 Q1149.52 32.4315 1145.75 35.9558 Q1142.02 39.4801 1141.46 45.8805 L1168.03 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1220.37 28.9478 L1220.37 35.9153 Q1217.21 34.1734 1214.01 33.3227 Q1210.85 32.4315 1207.61 32.4315 Q1200.36 32.4315 1196.35 37.0496 Q1192.34 41.6271 1192.34 49.9314 Q1192.34 58.2358 1196.35 62.8538 Q1200.36 67.4314 1207.61 67.4314 Q1210.85 67.4314 1214.01 66.5807 Q1217.21 65.6895 1220.37 63.9476 L1220.37 70.8341 Q1217.25 72.2924 1213.89 73.0216 Q1210.56 73.7508 1206.8 73.7508 Q1196.55 73.7508 1190.51 67.3098 Q1184.48 60.8689 1184.48 49.9314 Q1184.48 38.832 1190.55 32.472 Q1196.67 26.1121 1207.28 26.1121 Q1210.73 26.1121 1214.01 26.8413 Q1217.29 27.5299 1220.37 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1240.7 14.324 L1240.7 27.2059 L1256.06 27.2059 L1256.06 32.9987 L1240.7 32.9987 L1240.7 57.6282 Q1240.7 63.1779 1242.2 64.7578 Q1243.74 66.3376 1248.4 66.3376 L1256.06 66.3376 L1256.06 72.576 L1248.4 72.576 Q1239.77 72.576 1236.49 69.3758 Q1233.21 66.1351 1233.21 57.6282 L1233.21 32.9987 L1227.74 32.9987 L1227.74 27.2059 L1233.21 27.2059 L1233.21 14.324 L1240.7 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1265.86 27.2059 L1273.31 27.2059 L1273.31 72.576 L1265.86 72.576 L1265.86 27.2059 M1265.86 9.54393 L1273.31 9.54393 L1273.31 18.9825 L1265.86 18.9825 L1265.86 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1283.56 27.2059 L1291.46 27.2059 L1305.64 65.2844 L1319.82 27.2059 L1327.72 27.2059 L1310.7 72.576 L1300.58 72.576 L1283.56 27.2059 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1376.81 48.0275 L1376.81 51.6733 L1342.54 51.6733 Q1343.03 59.3701 1347.16 63.421 Q1351.33 67.4314 1358.75 67.4314 Q1363.04 67.4314 1367.05 66.3781 Q1371.1 65.3249 1375.07 63.2184 L1375.07 70.267 Q1371.06 71.9684 1366.85 72.8596 Q1362.64 73.7508 1358.3 73.7508 Q1347.44 73.7508 1341.08 67.4314 Q1334.76 61.1119 1334.76 50.3365 Q1334.76 39.1965 1340.76 32.6746 Q1346.8 26.1121 1357 26.1121 Q1366.16 26.1121 1371.47 32.0264 Q1376.81 37.9003 1376.81 48.0275 M1369.36 45.84 Q1369.28 39.7232 1365.92 36.0774 Q1362.59 32.4315 1357.09 32.4315 Q1350.85 32.4315 1347.08 35.9558 Q1343.35 39.4801 1342.79 45.8805 L1369.36 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1444.34 28.5427 L1444.34 35.5912 Q1441.18 33.9709 1437.78 33.1607 Q1434.38 32.3505 1430.73 32.3505 Q1425.18 32.3505 1422.39 34.0519 Q1419.63 35.7533 1419.63 39.156 Q1419.63 41.7486 1421.62 43.2475 Q1423.6 44.7058 1429.6 46.0426 L1432.15 46.6097 Q1440.09 48.3111 1443.41 51.4303 Q1446.77 54.509 1446.77 60.0587 Q1446.77 66.3781 1441.75 70.0644 Q1436.77 73.7508 1428.02 73.7508 Q1424.37 73.7508 1420.4 73.0216 Q1416.47 72.3329 1412.1 70.9151 L1412.1 63.2184 Q1416.23 65.3654 1420.24 66.4591 Q1424.25 67.5124 1428.18 67.5124 Q1433.44 67.5124 1436.28 65.73 Q1439.12 63.9071 1439.12 60.6258 Q1439.12 57.5877 1437.05 55.9673 Q1435.02 54.3469 1428.1 52.8481 L1425.51 52.2405 Q1418.58 50.7821 1415.5 47.7845 Q1412.42 44.7463 1412.42 39.4801 Q1412.42 33.0797 1416.96 29.5959 Q1421.49 26.1121 1429.84 26.1121 Q1433.97 26.1121 1437.62 26.7198 Q1441.26 27.3274 1444.34 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1465.85 65.7705 L1465.85 89.8329 L1458.36 89.8329 L1458.36 27.2059 L1465.85 27.2059 L1465.85 34.0924 Q1468.2 30.0415 1471.77 28.0971 Q1475.37 26.1121 1480.35 26.1121 Q1488.62 26.1121 1493.76 32.6746 Q1498.95 39.2371 1498.95 49.9314 Q1498.95 60.6258 1493.76 67.1883 Q1488.62 73.7508 1480.35 73.7508 Q1475.37 73.7508 1471.77 71.8063 Q1468.2 69.8214 1465.85 65.7705 M1491.21 49.9314 Q1491.21 41.7081 1487.81 37.0496 Q1484.45 32.3505 1478.53 32.3505 Q1472.62 32.3505 1469.21 37.0496 Q1465.85 41.7081 1465.85 49.9314 Q1465.85 58.1548 1469.21 62.8538 Q1472.62 67.5124 1478.53 67.5124 Q1484.45 67.5124 1487.81 62.8538 Q1491.21 58.1548 1491.21 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1531.92 49.7694 Q1522.89 49.7694 1519.41 51.8354 Q1515.92 53.9013 1515.92 58.8839 Q1515.92 62.8538 1518.51 65.2034 Q1521.15 67.5124 1525.64 67.5124 Q1531.84 67.5124 1535.57 63.1374 Q1539.34 58.7219 1539.34 51.4303 L1539.34 49.7694 L1531.92 49.7694 M1546.79 46.6907 L1546.79 72.576 L1539.34 72.576 L1539.34 65.6895 Q1536.78 69.8214 1532.98 71.8063 Q1529.17 73.7508 1523.66 73.7508 Q1516.69 73.7508 1512.56 69.8619 Q1508.47 65.9325 1508.47 59.3701 Q1508.47 51.7138 1513.57 47.825 Q1518.72 43.9361 1528.88 43.9361 L1539.34 43.9361 L1539.34 43.2069 Q1539.34 38.0623 1535.93 35.2672 Q1532.57 32.4315 1526.45 32.4315 Q1522.56 32.4315 1518.88 33.3632 Q1515.19 34.295 1511.79 36.1584 L1511.79 29.2718 Q1515.88 27.692 1519.73 26.9223 Q1523.58 26.1121 1527.22 26.1121 Q1537.07 26.1121 1541.93 31.2163 Q1546.79 36.3204 1546.79 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1594.79 28.9478 L1594.79 35.9153 Q1591.63 34.1734 1588.43 33.3227 Q1585.27 32.4315 1582.03 32.4315 Q1574.78 32.4315 1570.77 37.0496 Q1566.76 41.6271 1566.76 49.9314 Q1566.76 58.2358 1570.77 62.8538 Q1574.78 67.4314 1582.03 67.4314 Q1585.27 67.4314 1588.43 66.5807 Q1591.63 65.6895 1594.79 63.9476 L1594.79 70.8341 Q1591.67 72.2924 1588.31 73.0216 Q1584.99 73.7508 1581.22 73.7508 Q1570.97 73.7508 1564.94 67.3098 Q1558.9 60.8689 1558.9 49.9314 Q1558.9 38.832 1564.98 32.472 Q1571.09 26.1121 1581.71 26.1121 Q1585.15 26.1121 1588.43 26.8413 Q1591.71 27.5299 1594.79 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1646.56 48.0275 L1646.56 51.6733 L1612.29 51.6733 Q1612.78 59.3701 1616.91 63.421 Q1621.08 67.4314 1628.5 67.4314 Q1632.79 67.4314 1636.8 66.3781 Q1640.85 65.3249 1644.82 63.2184 L1644.82 70.267 Q1640.81 71.9684 1636.6 72.8596 Q1632.38 73.7508 1628.05 73.7508 Q1617.19 73.7508 1610.83 67.4314 Q1604.51 61.1119 1604.51 50.3365 Q1604.51 39.1965 1610.51 32.6746 Q1616.55 26.1121 1626.75 26.1121 Q1635.91 26.1121 1641.22 32.0264 Q1646.56 37.9003 1646.56 48.0275 M1639.11 45.84 Q1639.03 39.7232 1635.67 36.0774 Q1632.34 32.4315 1626.84 32.4315 Q1620.6 32.4315 1616.83 35.9558 Q1613.1 39.4801 1612.54 45.8805 L1639.11 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip912)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="-1780.13,422.883 4419.2,422.883 "/>
+<polyline clip-path="url(#clip912)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="987.246,1776.14 987.246,-657.503 "/>
+<circle clip-path="url(#clip912)" cx="344.796" cy="941.969" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="384.581" cy="812.114" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="424.366" cy="758.327" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="464.151" cy="717.054" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="503.936" cy="682.259" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="543.722" cy="651.605" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="583.507" cy="623.891" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="623.292" cy="598.406" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="663.077" cy="574.684" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="702.863" cy="552.405" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="742.648" cy="531.332" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="782.433" cy="511.289" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="822.218" cy="492.139" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="862.003" cy="473.771" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="901.789" cy="456.097" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="941.574" cy="439.599" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="981.359" cy="424.93" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1021.14" cy="411.608" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1060.93" cy="399.319" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1100.71" cy="387.854" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1140.5" cy="377.067" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1180.29" cy="366.85" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1220.07" cy="357.121" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1259.86" cy="347.817" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1299.64" cy="338.886" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1339.43" cy="330.286" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1379.21" cy="321.984" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1419" cy="313.95" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1458.78" cy="306.16" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1498.57" cy="298.592" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1538.35" cy="291.23" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1578.14" cy="284.057" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1617.92" cy="277.059" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1657.71" cy="270.224" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1697.49" cy="263.541" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1737.28" cy="257" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1777.06" cy="250.593" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1816.85" cy="244.312" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1856.63" cy="238.149" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1896.42" cy="232.098" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1936.2" cy="226.154" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="1975.99" cy="220.311" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2015.78" cy="214.563" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2055.56" cy="208.907" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2095.35" cy="203.338" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2135.13" cy="197.852" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2174.92" cy="192.445" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2214.7" cy="187.115" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2254.49" cy="181.858" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<circle clip-path="url(#clip912)" cx="2294.27" cy="176.671" r="18" fill="#ffffff" fill-rule="evenodd" fill-opacity="1" stroke="#000000" stroke-opacity="1" stroke-width="4.8"/>
+<path clip-path="url(#clip910)" d="M1501.38 937.887 L2283.87 937.887 L2283.87 834.207 L1501.38 834.207  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1501.38,937.887 2283.87,937.887 2283.87,834.207 1501.38,834.207 1501.38,937.887 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#ff0000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1524.34,886.047 1662.1,886.047 "/>
+<path clip-path="url(#clip910)" d="M1707.31 869.901 L1707.31 874.461 Q1704.64 873.188 1702.28 872.563 Q1699.92 871.938 1697.72 871.938 Q1693.9 871.938 1691.82 873.42 Q1689.76 874.901 1689.76 877.633 Q1689.76 879.924 1691.13 881.105 Q1692.51 882.262 1696.36 882.98 L1699.18 883.559 Q1704.41 884.554 1706.89 887.077 Q1709.39 889.577 1709.39 893.79 Q1709.39 898.813 1706.01 901.406 Q1702.65 903.998 1696.15 903.998 Q1693.69 903.998 1690.92 903.443 Q1688.16 902.887 1685.2 901.799 L1685.2 896.984 Q1688.05 898.582 1690.78 899.392 Q1693.51 900.202 1696.15 900.202 Q1700.15 900.202 1702.33 898.628 Q1704.5 897.054 1704.5 894.137 Q1704.5 891.591 1702.93 890.156 Q1701.38 888.721 1697.82 888.003 L1694.97 887.447 Q1689.74 886.406 1687.4 884.184 Q1685.06 881.961 1685.06 878.003 Q1685.06 873.42 1688.28 870.781 Q1691.52 868.142 1697.19 868.142 Q1699.62 868.142 1702.14 868.582 Q1704.67 869.022 1707.31 869.901 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1716.5 877.401 L1720.75 877.401 L1720.75 903.327 L1716.5 903.327 L1716.5 877.401 M1716.5 867.309 L1720.75 867.309 L1720.75 872.702 L1716.5 872.702 L1716.5 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1751.22 887.679 L1751.22 903.327 L1746.96 903.327 L1746.96 887.818 Q1746.96 884.137 1745.52 882.309 Q1744.09 880.48 1741.22 880.48 Q1737.77 880.48 1735.78 882.679 Q1733.79 884.878 1733.79 888.674 L1733.79 903.327 L1729.5 903.327 L1729.5 877.401 L1733.79 877.401 L1733.79 881.429 Q1735.31 879.091 1737.38 877.934 Q1739.46 876.776 1742.17 876.776 Q1746.63 876.776 1748.93 879.554 Q1751.22 882.309 1751.22 887.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1776.77 890.063 Q1776.77 885.434 1774.85 882.887 Q1772.95 880.341 1769.5 880.341 Q1766.08 880.341 1764.16 882.887 Q1762.26 885.434 1762.26 890.063 Q1762.26 894.67 1764.16 897.216 Q1766.08 899.762 1769.5 899.762 Q1772.95 899.762 1774.85 897.216 Q1776.77 894.67 1776.77 890.063 M1781.03 900.109 Q1781.03 906.73 1778.09 909.947 Q1775.15 913.188 1769.09 913.188 Q1766.84 913.188 1764.85 912.841 Q1762.86 912.517 1760.99 911.822 L1760.99 907.679 Q1762.86 908.697 1764.69 909.183 Q1766.52 909.669 1768.42 909.669 Q1772.61 909.669 1774.69 907.47 Q1776.77 905.295 1776.77 900.873 L1776.77 898.767 Q1775.45 901.058 1773.39 902.193 Q1771.33 903.327 1768.46 903.327 Q1763.69 903.327 1760.78 899.693 Q1757.86 896.058 1757.86 890.063 Q1757.86 884.045 1760.78 880.41 Q1763.69 876.776 1768.46 876.776 Q1771.33 876.776 1773.39 877.91 Q1775.45 879.045 1776.77 881.336 L1776.77 877.401 L1781.03 877.401 L1781.03 900.109 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1789.81 867.309 L1794.06 867.309 L1794.06 903.327 L1789.81 903.327 L1789.81 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1825.15 889.299 L1825.15 891.383 L1805.57 891.383 Q1805.85 895.781 1808.21 898.095 Q1810.59 900.387 1814.83 900.387 Q1817.28 900.387 1819.57 899.785 Q1821.89 899.183 1824.16 897.98 L1824.16 902.007 Q1821.87 902.98 1819.46 903.489 Q1817.05 903.998 1814.57 903.998 Q1808.37 903.998 1804.74 900.387 Q1801.12 896.776 1801.12 890.619 Q1801.12 884.253 1804.55 880.526 Q1808 876.776 1813.83 876.776 Q1819.06 876.776 1822.1 880.156 Q1825.15 883.512 1825.15 889.299 M1820.89 888.049 Q1820.85 884.554 1818.93 882.471 Q1817.03 880.387 1813.88 880.387 Q1810.31 880.387 1808.16 882.401 Q1806.03 884.415 1805.71 888.072 L1820.89 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1829.99 888.443 L1842.47 888.443 L1842.47 892.239 L1829.99 892.239 L1829.99 888.443 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1860.18 880.387 Q1856.75 880.387 1854.76 883.072 Q1852.77 885.734 1852.77 890.387 Q1852.77 895.04 1854.74 897.725 Q1856.73 900.387 1860.18 900.387 Q1863.58 900.387 1865.57 897.702 Q1867.56 895.017 1867.56 890.387 Q1867.56 885.781 1865.57 883.096 Q1863.58 880.387 1860.18 880.387 M1860.18 876.776 Q1865.73 876.776 1868.9 880.387 Q1872.07 883.998 1872.07 890.387 Q1872.07 896.753 1868.9 900.387 Q1865.73 903.998 1860.18 903.998 Q1854.6 903.998 1851.43 900.387 Q1848.28 896.753 1848.28 890.387 Q1848.28 883.998 1851.43 880.387 Q1854.6 876.776 1860.18 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1897.74 890.387 Q1897.74 885.688 1895.8 883.026 Q1893.88 880.341 1890.5 880.341 Q1887.12 880.341 1885.18 883.026 Q1883.25 885.688 1883.25 890.387 Q1883.25 895.086 1885.18 897.771 Q1887.12 900.433 1890.5 900.433 Q1893.88 900.433 1895.8 897.771 Q1897.74 895.086 1897.74 890.387 M1883.25 881.336 Q1884.6 879.022 1886.63 877.91 Q1888.69 876.776 1891.54 876.776 Q1896.26 876.776 1899.2 880.526 Q1902.17 884.276 1902.17 890.387 Q1902.17 896.498 1899.2 900.248 Q1896.26 903.998 1891.54 903.998 Q1888.69 903.998 1886.63 902.887 Q1884.6 901.753 1883.25 899.438 L1883.25 903.327 L1878.97 903.327 L1878.97 867.309 L1883.25 867.309 L1883.25 881.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1909.23 877.401 L1913.49 877.401 L1913.49 903.79 Q1913.49 908.744 1911.59 910.966 Q1909.71 913.188 1905.52 913.188 L1903.9 913.188 L1903.9 909.577 L1905.04 909.577 Q1907.47 909.577 1908.35 908.443 Q1909.23 907.332 1909.23 903.79 L1909.23 877.401 M1909.23 867.309 L1913.49 867.309 L1913.49 872.702 L1909.23 872.702 L1909.23 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1944.57 889.299 L1944.57 891.383 L1924.99 891.383 Q1925.27 895.781 1927.63 898.095 Q1930.01 900.387 1934.25 900.387 Q1936.7 900.387 1938.99 899.785 Q1941.31 899.183 1943.58 897.98 L1943.58 902.007 Q1941.29 902.98 1938.88 903.489 Q1936.47 903.998 1933.99 903.998 Q1927.79 903.998 1924.16 900.387 Q1920.55 896.776 1920.55 890.619 Q1920.55 884.253 1923.97 880.526 Q1927.42 876.776 1933.25 876.776 Q1938.49 876.776 1941.52 880.156 Q1944.57 883.512 1944.57 889.299 M1940.31 888.049 Q1940.27 884.554 1938.35 882.471 Q1936.45 880.387 1933.3 880.387 Q1929.74 880.387 1927.58 882.401 Q1925.45 884.415 1925.13 888.072 L1940.31 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1970.22 878.397 L1970.22 882.378 Q1968.42 881.383 1966.59 880.896 Q1964.78 880.387 1962.93 880.387 Q1958.79 880.387 1956.49 883.026 Q1954.2 885.642 1954.2 890.387 Q1954.2 895.133 1956.49 897.771 Q1958.79 900.387 1962.93 900.387 Q1964.78 900.387 1966.59 899.901 Q1968.42 899.392 1970.22 898.396 L1970.22 902.332 Q1968.44 903.165 1966.52 903.582 Q1964.62 903.998 1962.47 903.998 Q1956.61 903.998 1953.16 900.318 Q1949.71 896.637 1949.71 890.387 Q1949.71 884.045 1953.18 880.41 Q1956.68 876.776 1962.74 876.776 Q1964.71 876.776 1966.59 877.193 Q1968.46 877.586 1970.22 878.397 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1981.84 870.04 L1981.84 877.401 L1990.61 877.401 L1990.61 880.711 L1981.84 880.711 L1981.84 894.785 Q1981.84 897.957 1982.7 898.859 Q1983.58 899.762 1986.24 899.762 L1990.61 899.762 L1990.61 903.327 L1986.24 903.327 Q1981.31 903.327 1979.43 901.498 Q1977.56 899.646 1977.56 894.785 L1977.56 880.711 L1974.43 880.711 L1974.43 877.401 L1977.56 877.401 L1977.56 870.04 L1981.84 870.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1996.22 877.401 L2000.48 877.401 L2000.48 903.327 L1996.22 903.327 L1996.22 877.401 M1996.22 867.309 L2000.48 867.309 L2000.48 872.702 L1996.22 872.702 L1996.22 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2006.33 877.401 L2010.85 877.401 L2018.95 899.16 L2027.05 877.401 L2031.56 877.401 L2021.84 903.327 L2016.05 903.327 L2006.33 877.401 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2059.62 889.299 L2059.62 891.383 L2040.04 891.383 Q2040.31 895.781 2042.67 898.095 Q2045.06 900.387 2049.29 900.387 Q2051.75 900.387 2054.04 899.785 Q2056.35 899.183 2058.62 897.98 L2058.62 902.007 Q2056.33 902.98 2053.92 903.489 Q2051.52 903.998 2049.04 903.998 Q2042.84 903.998 2039.2 900.387 Q2035.59 896.776 2035.59 890.619 Q2035.59 884.253 2039.02 880.526 Q2042.47 876.776 2048.3 876.776 Q2053.53 876.776 2056.56 880.156 Q2059.62 883.512 2059.62 889.299 M2055.36 888.049 Q2055.31 884.554 2053.39 882.471 Q2051.49 880.387 2048.35 880.387 Q2044.78 880.387 2042.63 882.401 Q2040.5 884.415 2040.17 888.072 L2055.36 888.049 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2098.21 878.165 L2098.21 882.193 Q2096.4 881.267 2094.46 880.804 Q2092.51 880.341 2090.43 880.341 Q2087.26 880.341 2085.66 881.313 Q2084.09 882.285 2084.09 884.23 Q2084.09 885.711 2085.22 886.568 Q2086.35 887.401 2089.78 888.165 L2091.24 888.489 Q2095.78 889.461 2097.67 891.244 Q2099.6 893.003 2099.6 896.174 Q2099.6 899.785 2096.72 901.892 Q2093.88 903.998 2088.88 903.998 Q2086.79 903.998 2084.53 903.582 Q2082.28 903.188 2079.78 902.378 L2079.78 897.98 Q2082.14 899.207 2084.43 899.832 Q2086.73 900.433 2088.97 900.433 Q2091.98 900.433 2093.6 899.415 Q2095.22 898.373 2095.22 896.498 Q2095.22 894.762 2094.04 893.836 Q2092.88 892.91 2088.92 892.054 L2087.44 891.707 Q2083.48 890.873 2081.73 889.16 Q2079.97 887.424 2079.97 884.415 Q2079.97 880.758 2082.56 878.767 Q2085.15 876.776 2089.92 876.776 Q2092.28 876.776 2094.36 877.123 Q2096.45 877.471 2098.21 878.165 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2116.42 880.387 Q2113 880.387 2111.01 883.072 Q2109.02 885.734 2109.02 890.387 Q2109.02 895.04 2110.98 897.725 Q2112.97 900.387 2116.42 900.387 Q2119.83 900.387 2121.82 897.702 Q2123.81 895.017 2123.81 890.387 Q2123.81 885.781 2121.82 883.096 Q2119.83 880.387 2116.42 880.387 M2116.42 876.776 Q2121.98 876.776 2125.15 880.387 Q2128.32 883.998 2128.32 890.387 Q2128.32 896.753 2125.15 900.387 Q2121.98 903.998 2116.42 903.998 Q2110.85 903.998 2107.67 900.387 Q2104.53 896.753 2104.53 890.387 Q2104.53 883.998 2107.67 880.387 Q2110.85 876.776 2116.42 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2135.38 867.309 L2139.64 867.309 L2139.64 903.327 L2135.38 903.327 L2135.38 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2148.11 893.096 L2148.11 877.401 L2152.37 877.401 L2152.37 892.933 Q2152.37 896.614 2153.81 898.466 Q2155.24 900.295 2158.11 900.295 Q2161.56 900.295 2163.55 898.095 Q2165.57 895.896 2165.57 892.1 L2165.57 877.401 L2169.83 877.401 L2169.83 903.327 L2165.57 903.327 L2165.57 899.345 Q2164.02 901.707 2161.96 902.864 Q2159.92 903.998 2157.21 903.998 Q2152.74 903.998 2150.43 901.22 Q2148.11 898.443 2148.11 893.096 M2158.83 876.776 L2158.83 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2182.81 870.04 L2182.81 877.401 L2191.59 877.401 L2191.59 880.711 L2182.81 880.711 L2182.81 894.785 Q2182.81 897.957 2183.67 898.859 Q2184.55 899.762 2187.21 899.762 L2191.59 899.762 L2191.59 903.327 L2187.21 903.327 Q2182.28 903.327 2180.41 901.498 Q2178.53 899.646 2178.53 894.785 L2178.53 880.711 L2175.41 880.711 L2175.41 877.401 L2178.53 877.401 L2178.53 870.04 L2182.81 870.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2197.19 877.401 L2201.45 877.401 L2201.45 903.327 L2197.19 903.327 L2197.19 877.401 M2197.19 867.309 L2201.45 867.309 L2201.45 872.702 L2197.19 872.702 L2197.19 867.309 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2220.4 880.387 Q2216.98 880.387 2214.99 883.072 Q2213 885.734 2213 890.387 Q2213 895.04 2214.97 897.725 Q2216.96 900.387 2220.4 900.387 Q2223.81 900.387 2225.8 897.702 Q2227.79 895.017 2227.79 890.387 Q2227.79 885.781 2225.8 883.096 Q2223.81 880.387 2220.4 880.387 M2220.4 876.776 Q2225.96 876.776 2229.13 880.387 Q2232.3 883.998 2232.3 890.387 Q2232.3 896.753 2229.13 900.387 Q2225.96 903.998 2220.4 903.998 Q2214.83 903.998 2211.65 900.387 Q2208.51 896.753 2208.51 890.387 Q2208.51 883.998 2211.65 880.387 Q2214.83 876.776 2220.4 876.776 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2260.91 887.679 L2260.91 903.327 L2256.65 903.327 L2256.65 887.818 Q2256.65 884.137 2255.22 882.309 Q2253.78 880.48 2250.91 880.48 Q2247.46 880.48 2245.47 882.679 Q2243.48 884.878 2243.48 888.674 L2243.48 903.327 L2239.2 903.327 L2239.2 877.401 L2243.48 877.401 L2243.48 881.429 Q2245.01 879.091 2247.07 877.934 Q2249.15 876.776 2251.86 876.776 Q2256.33 876.776 2258.62 879.554 Q2260.91 882.309 2260.91 887.679 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M341.189 946.973 L344.053 946.973 L344.053 937.086 L340.937 937.711 L340.937 936.114 L344.036 935.489 L345.789 935.489 L345.789 946.973 L348.654 946.973 L348.654 948.449 L341.189 948.449 L341.189 946.973 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M382.576 817.119 L388.695 817.119 L388.695 818.594 L380.466 818.594 L380.466 817.119 Q381.464 816.086 383.183 814.349 Q384.911 812.605 385.353 812.101 Q386.195 811.155 386.525 810.504 Q386.864 809.844 386.864 809.211 Q386.864 808.178 386.135 807.527 Q385.414 806.876 384.251 806.876 Q383.426 806.876 382.506 807.162 Q381.595 807.448 380.553 808.03 L380.553 806.259 Q381.612 805.834 382.532 805.617 Q383.452 805.4 384.216 805.4 Q386.23 805.4 387.428 806.407 Q388.626 807.414 388.626 809.098 Q388.626 809.896 388.322 810.617 Q388.027 811.329 387.237 812.301 Q387.02 812.553 385.857 813.759 Q384.694 814.957 382.576 817.119 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M425.959 757.819 Q427.218 758.088 427.921 758.939 Q428.632 759.789 428.632 761.039 Q428.632 762.958 427.313 764.008 Q425.994 765.058 423.563 765.058 Q422.747 765.058 421.879 764.893 Q421.02 764.737 420.099 764.416 L420.099 762.723 Q420.829 763.149 421.697 763.366 Q422.565 763.583 423.511 763.583 Q425.16 763.583 426.02 762.932 Q426.888 762.281 426.888 761.039 Q426.888 759.893 426.08 759.251 Q425.282 758.6 423.849 758.6 L422.339 758.6 L422.339 757.159 L423.919 757.159 Q425.212 757.159 425.898 756.647 Q426.584 756.126 426.584 755.154 Q426.584 754.156 425.872 753.626 Q425.169 753.088 423.849 753.088 Q423.129 753.088 422.304 753.244 Q421.48 753.4 420.49 753.73 L420.49 752.168 Q421.488 751.89 422.356 751.751 Q423.233 751.612 424.006 751.612 Q426.002 751.612 427.165 752.524 Q428.329 753.426 428.329 754.972 Q428.329 756.048 427.712 756.795 Q427.096 757.532 425.959 757.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M465.28 712.102 L460.853 719.02 L465.28 719.02 L465.28 712.102 M464.82 710.574 L467.024 710.574 L467.024 719.02 L468.873 719.02 L468.873 720.478 L467.024 720.478 L467.024 723.534 L465.28 723.534 L465.28 720.478 L459.429 720.478 L459.429 718.786 L464.82 710.574 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M500.291 675.779 L507.174 675.779 L507.174 677.255 L501.896 677.255 L501.896 680.432 Q502.278 680.302 502.66 680.241 Q503.042 680.172 503.424 680.172 Q505.594 680.172 506.862 681.361 Q508.129 682.55 508.129 684.581 Q508.129 686.673 506.827 687.837 Q505.525 688.991 503.155 688.991 Q502.339 688.991 501.489 688.852 Q500.647 688.713 499.744 688.436 L499.744 686.673 Q500.525 687.099 501.358 687.307 Q502.192 687.515 503.12 687.515 Q504.622 687.515 505.499 686.726 Q506.376 685.936 506.376 684.581 Q506.376 683.227 505.499 682.437 Q504.622 681.647 503.12 681.647 Q502.417 681.647 501.714 681.804 Q501.02 681.96 500.291 682.29 L500.291 675.779 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M543.874 650.906 Q542.693 650.906 541.999 651.713 Q541.313 652.521 541.313 653.927 Q541.313 655.324 541.999 656.14 Q542.693 656.948 543.874 656.948 Q545.054 656.948 545.74 656.14 Q546.434 655.324 546.434 653.927 Q546.434 652.521 545.74 651.713 Q545.054 650.906 543.874 650.906 M547.354 645.411 L547.354 647.009 Q546.695 646.696 546.018 646.531 Q545.349 646.366 544.69 646.366 Q542.953 646.366 542.033 647.538 Q541.122 648.71 540.992 651.08 Q541.504 650.324 542.276 649.925 Q543.049 649.517 543.978 649.517 Q545.931 649.517 547.059 650.706 Q548.196 651.887 548.196 653.927 Q548.196 655.923 547.016 657.13 Q545.835 658.337 543.874 658.337 Q541.625 658.337 540.436 656.618 Q539.247 654.89 539.247 651.618 Q539.247 648.545 540.705 646.722 Q542.163 644.89 544.62 644.89 Q545.28 644.89 545.948 645.021 Q546.625 645.151 547.354 645.411 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M579.34 617.411 L587.674 617.411 L587.674 618.158 L582.969 630.371 L581.137 630.371 L585.564 618.887 L579.34 618.887 L579.34 617.411 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M623.292 598.731 Q622.042 598.731 621.322 599.399 Q620.61 600.068 620.61 601.24 Q620.61 602.412 621.322 603.08 Q622.042 603.748 623.292 603.748 Q624.542 603.748 625.263 603.08 Q625.983 602.403 625.983 601.24 Q625.983 600.068 625.263 599.399 Q624.551 598.731 623.292 598.731 M621.539 597.985 Q620.41 597.707 619.776 596.934 Q619.151 596.162 619.151 595.051 Q619.151 593.497 620.254 592.594 Q621.365 591.691 623.292 591.691 Q625.228 591.691 626.33 592.594 Q627.433 593.497 627.433 595.051 Q627.433 596.162 626.799 596.934 Q626.174 597.707 625.054 597.985 Q626.322 598.28 627.025 599.139 Q627.737 599.998 627.737 601.24 Q627.737 603.123 626.582 604.13 Q625.436 605.137 623.292 605.137 Q621.148 605.137 619.994 604.13 Q618.848 603.123 618.848 601.24 Q618.848 599.998 619.559 599.139 Q620.271 598.28 621.539 597.985 M620.896 595.215 Q620.896 596.222 621.521 596.787 Q622.155 597.351 623.292 597.351 Q624.421 597.351 625.054 596.787 Q625.697 596.222 625.697 595.215 Q625.697 594.208 625.054 593.644 Q624.421 593.08 623.292 593.08 Q622.155 593.08 621.521 593.644 Q620.896 594.208 620.896 595.215 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M659.44 580.895 L659.44 579.298 Q660.1 579.61 660.777 579.775 Q661.454 579.94 662.105 579.94 Q663.841 579.94 664.753 578.777 Q665.673 577.605 665.803 575.227 Q665.3 575.973 664.527 576.373 Q663.754 576.772 662.817 576.772 Q660.872 576.772 659.735 575.6 Q658.607 574.419 658.607 572.38 Q658.607 570.383 659.787 569.176 Q660.968 567.97 662.93 567.97 Q665.178 567.97 666.359 569.697 Q667.548 571.416 667.548 574.697 Q667.548 577.761 666.089 579.593 Q664.64 581.416 662.183 581.416 Q661.524 581.416 660.846 581.286 Q660.169 581.156 659.44 580.895 M662.93 575.4 Q664.11 575.4 664.796 574.593 Q665.491 573.786 665.491 572.38 Q665.491 570.982 664.796 570.175 Q664.11 569.359 662.93 569.359 Q661.749 569.359 661.055 570.175 Q660.369 570.982 660.369 572.38 Q660.369 573.786 661.055 574.593 Q661.749 575.4 662.93 575.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M693.37 557.409 L696.235 557.409 L696.235 547.522 L693.119 548.147 L693.119 546.55 L696.218 545.925 L697.971 545.925 L697.971 557.409 L700.836 557.409 L700.836 558.885 L693.37 558.885 L693.37 557.409 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M708.127 547.079 Q706.773 547.079 706.087 548.416 Q705.41 549.744 705.41 552.418 Q705.41 555.083 706.087 556.419 Q706.773 557.747 708.127 557.747 Q709.49 557.747 710.167 556.419 Q710.853 555.083 710.853 552.418 Q710.853 549.744 710.167 548.416 Q709.49 547.079 708.127 547.079 M708.127 545.69 Q710.306 545.69 711.452 547.418 Q712.606 549.136 712.606 552.418 Q712.606 555.69 711.452 557.418 Q710.306 559.136 708.127 559.136 Q705.948 559.136 704.794 557.418 Q703.648 555.69 703.648 552.418 Q703.648 549.136 704.794 547.418 Q705.948 545.69 708.127 545.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M733.386 536.336 L736.25 536.336 L736.25 526.449 L733.134 527.074 L733.134 525.477 L736.233 524.852 L737.986 524.852 L737.986 536.336 L740.851 536.336 L740.851 537.812 L733.386 537.812 L733.386 536.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M744.696 536.336 L747.561 536.336 L747.561 526.449 L744.445 527.074 L744.445 525.477 L747.544 524.852 L749.297 524.852 L749.297 536.336 L752.162 536.336 L752.162 537.812 L744.696 537.812 L744.696 536.336 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M773.24 516.294 L776.105 516.294 L776.105 506.407 L772.989 507.032 L772.989 505.434 L776.088 504.809 L777.841 504.809 L777.841 516.294 L780.706 516.294 L780.706 517.769 L773.24 517.769 L773.24 516.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M785.758 516.294 L791.877 516.294 L791.877 517.769 L783.648 517.769 L783.648 516.294 Q784.647 515.261 786.365 513.525 Q788.093 511.78 788.535 511.276 Q789.377 510.33 789.707 509.679 Q790.046 509.019 790.046 508.386 Q790.046 507.353 789.317 506.702 Q788.596 506.051 787.433 506.051 Q786.608 506.051 785.688 506.337 Q784.777 506.624 783.735 507.205 L783.735 505.434 Q784.794 505.009 785.714 504.792 Q786.634 504.575 787.398 504.575 Q789.412 504.575 790.61 505.582 Q791.808 506.589 791.808 508.273 Q791.808 509.071 791.504 509.792 Q791.209 510.504 790.419 511.476 Q790.202 511.728 789.039 512.934 Q787.876 514.132 785.758 516.294 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M812.848 497.143 L815.712 497.143 L815.712 487.256 L812.596 487.881 L812.596 486.284 L815.695 485.659 L817.448 485.659 L817.448 497.143 L820.313 497.143 L820.313 498.619 L812.848 498.619 L812.848 497.143 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M829.167 491.631 Q830.426 491.9 831.129 492.751 Q831.841 493.601 831.841 494.851 Q831.841 496.77 830.521 497.82 Q829.202 498.871 826.771 498.871 Q825.955 498.871 825.087 498.706 Q824.228 498.549 823.308 498.228 L823.308 496.535 Q824.037 496.961 824.905 497.178 Q825.773 497.395 826.719 497.395 Q828.368 497.395 829.228 496.744 Q830.096 496.093 830.096 494.851 Q830.096 493.706 829.289 493.063 Q828.49 492.412 827.058 492.412 L825.547 492.412 L825.547 490.971 L827.127 490.971 Q828.42 490.971 829.106 490.459 Q829.792 489.938 829.792 488.966 Q829.792 487.968 829.08 487.438 Q828.377 486.9 827.058 486.9 Q826.337 486.9 825.512 487.056 Q824.688 487.213 823.698 487.542 L823.698 485.98 Q824.697 485.702 825.565 485.563 Q826.441 485.424 827.214 485.424 Q829.21 485.424 830.374 486.336 Q831.537 487.239 831.537 488.784 Q831.537 489.86 830.92 490.607 Q830.304 491.345 829.167 491.631 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M852.42 478.775 L855.285 478.775 L855.285 468.888 L852.168 469.513 L852.168 467.916 L855.267 467.291 L857.021 467.291 L857.021 478.775 L859.885 478.775 L859.885 480.251 L852.42 480.251 L852.42 478.775 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M868.245 468.819 L863.818 475.737 L868.245 475.737 L868.245 468.819 M867.785 467.291 L869.99 467.291 L869.99 475.737 L871.838 475.737 L871.838 477.195 L869.99 477.195 L869.99 480.251 L868.245 480.251 L868.245 477.195 L862.394 477.195 L862.394 475.503 L867.785 467.291 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M892.483 461.101 L895.348 461.101 L895.348 451.214 L892.231 451.839 L892.231 450.242 L895.33 449.617 L897.084 449.617 L897.084 461.101 L899.948 461.101 L899.948 462.577 L892.483 462.577 L892.483 461.101 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M903.507 449.617 L910.391 449.617 L910.391 451.093 L905.113 451.093 L905.113 454.27 Q905.495 454.139 905.877 454.079 Q906.259 454.009 906.641 454.009 Q908.811 454.009 910.079 455.198 Q911.346 456.388 911.346 458.419 Q911.346 460.511 910.044 461.674 Q908.742 462.829 906.372 462.829 Q905.556 462.829 904.705 462.69 Q903.863 462.551 902.961 462.273 L902.961 460.511 Q903.742 460.936 904.575 461.145 Q905.408 461.353 906.337 461.353 Q907.839 461.353 908.716 460.563 Q909.592 459.773 909.592 458.419 Q909.592 457.065 908.716 456.275 Q907.839 455.485 906.337 455.485 Q905.634 455.485 904.931 455.641 Q904.237 455.797 903.507 456.127 L903.507 449.617 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M932.051 444.603 L934.916 444.603 L934.916 434.716 L931.8 435.341 L931.8 433.744 L934.899 433.119 L936.652 433.119 L936.652 444.603 L939.517 444.603 L939.517 446.079 L932.051 446.079 L932.051 444.603 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M947.025 438.9 Q945.845 438.9 945.15 439.707 Q944.465 440.514 944.465 441.921 Q944.465 443.318 945.15 444.134 Q945.845 444.941 947.025 444.941 Q948.206 444.941 948.892 444.134 Q949.586 443.318 949.586 441.921 Q949.586 440.514 948.892 439.707 Q948.206 438.9 947.025 438.9 M950.506 433.405 L950.506 435.002 Q949.846 434.69 949.169 434.525 Q948.501 434.36 947.841 434.36 Q946.105 434.36 945.185 435.532 Q944.274 436.704 944.143 439.073 Q944.655 438.318 945.428 437.919 Q946.201 437.511 947.129 437.511 Q949.083 437.511 950.211 438.7 Q951.348 439.881 951.348 441.921 Q951.348 443.917 950.168 445.124 Q948.987 446.33 947.025 446.33 Q944.777 446.33 943.588 444.612 Q942.399 442.884 942.399 439.612 Q942.399 436.539 943.857 434.716 Q945.315 432.884 947.772 432.884 Q948.432 432.884 949.1 433.014 Q949.777 433.145 950.506 433.405 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M972.036 429.934 L974.901 429.934 L974.901 420.047 L971.785 420.672 L971.785 419.075 L974.883 418.45 L976.637 418.45 L976.637 429.934 L979.502 429.934 L979.502 431.41 L972.036 431.41 L972.036 429.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M982.6 418.45 L990.934 418.45 L990.934 419.196 L986.229 431.41 L984.397 431.41 L988.824 419.925 L982.6 419.925 L982.6 418.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1011.67 416.612 L1014.53 416.612 L1014.53 406.725 L1011.42 407.35 L1011.42 405.753 L1014.52 405.128 L1016.27 405.128 L1016.27 416.612 L1019.13 416.612 L1019.13 418.088 L1011.67 418.088 L1011.67 416.612 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1026.43 411.934 Q1025.18 411.934 1024.46 412.602 Q1023.74 413.27 1023.74 414.442 Q1023.74 415.614 1024.46 416.283 Q1025.18 416.951 1026.43 416.951 Q1027.68 416.951 1028.4 416.283 Q1029.12 415.606 1029.12 414.442 Q1029.12 413.27 1028.4 412.602 Q1027.69 411.934 1026.43 411.934 M1024.67 411.187 Q1023.54 410.909 1022.91 410.137 Q1022.29 409.364 1022.29 408.253 Q1022.29 406.699 1023.39 405.797 Q1024.5 404.894 1026.43 404.894 Q1028.36 404.894 1029.46 405.797 Q1030.57 406.699 1030.57 408.253 Q1030.57 409.364 1029.93 410.137 Q1029.31 410.909 1028.19 411.187 Q1029.46 411.482 1030.16 412.342 Q1030.87 413.201 1030.87 414.442 Q1030.87 416.326 1029.72 417.333 Q1028.57 418.34 1026.43 418.34 Q1024.28 418.34 1023.13 417.333 Q1021.98 416.326 1021.98 414.442 Q1021.98 413.201 1022.69 412.342 Q1023.41 411.482 1024.67 411.187 M1024.03 408.418 Q1024.03 409.425 1024.66 409.989 Q1025.29 410.553 1026.43 410.553 Q1027.55 410.553 1028.19 409.989 Q1028.83 409.425 1028.83 408.418 Q1028.83 407.411 1028.19 406.847 Q1027.55 406.283 1026.43 406.283 Q1025.29 406.283 1024.66 406.847 Q1024.03 407.411 1024.03 408.418 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1051.47 404.323 L1054.34 404.323 L1054.34 394.436 L1051.22 395.061 L1051.22 393.464 L1054.32 392.839 L1056.07 392.839 L1056.07 404.323 L1058.94 404.323 L1058.94 405.799 L1051.47 405.799 L1051.47 404.323 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1062.53 405.53 L1062.53 403.933 Q1063.19 404.245 1063.87 404.41 Q1064.55 404.575 1065.2 404.575 Q1066.93 404.575 1067.84 403.412 Q1068.76 402.24 1068.89 399.862 Q1068.39 400.608 1067.62 401.007 Q1066.85 401.407 1065.91 401.407 Q1063.96 401.407 1062.83 400.235 Q1061.7 399.054 1061.7 397.014 Q1061.7 395.018 1062.88 393.811 Q1064.06 392.605 1066.02 392.605 Q1068.27 392.605 1069.45 394.332 Q1070.64 396.051 1070.64 399.332 Q1070.64 402.396 1069.18 404.228 Q1067.73 406.051 1065.27 406.051 Q1064.61 406.051 1063.94 405.921 Q1063.26 405.79 1062.53 405.53 M1066.02 400.035 Q1067.2 400.035 1067.89 399.228 Q1068.58 398.421 1068.58 397.014 Q1068.58 395.617 1067.89 394.81 Q1067.2 393.994 1066.02 393.994 Q1064.84 393.994 1064.15 394.81 Q1063.46 395.617 1063.46 397.014 Q1063.46 398.421 1064.15 399.228 Q1064.84 400.035 1066.02 400.035 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1092.75 392.859 L1098.87 392.859 L1098.87 394.334 L1090.65 394.334 L1090.65 392.859 Q1091.64 391.826 1093.36 390.089 Q1095.09 388.345 1095.53 387.841 Q1096.37 386.895 1096.7 386.244 Q1097.04 385.584 1097.04 384.951 Q1097.04 383.918 1096.31 383.267 Q1095.59 382.616 1094.43 382.616 Q1093.61 382.616 1092.69 382.902 Q1091.77 383.188 1090.73 383.77 L1090.73 381.999 Q1091.79 381.574 1092.71 381.357 Q1093.63 381.14 1094.4 381.14 Q1096.41 381.14 1097.61 382.147 Q1098.81 383.154 1098.81 384.838 Q1098.81 385.636 1098.5 386.357 Q1098.21 387.069 1097.42 388.041 Q1097.2 388.293 1096.04 389.499 Q1094.87 390.697 1092.75 392.859 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1106.31 382.529 Q1104.95 382.529 1104.27 383.866 Q1103.59 385.194 1103.59 387.867 Q1103.59 390.532 1104.27 391.869 Q1104.95 393.197 1106.31 393.197 Q1107.67 393.197 1108.34 391.869 Q1109.03 390.532 1109.03 387.867 Q1109.03 385.194 1108.34 383.866 Q1107.67 382.529 1106.31 382.529 M1106.31 381.14 Q1108.48 381.14 1109.63 382.867 Q1110.78 384.586 1110.78 387.867 Q1110.78 391.14 1109.63 392.867 Q1108.48 394.586 1106.31 394.586 Q1104.13 394.586 1102.97 392.867 Q1101.83 391.14 1101.83 387.867 Q1101.83 384.586 1102.97 382.867 Q1104.13 381.14 1106.31 381.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1132.77 382.071 L1138.89 382.071 L1138.89 383.547 L1130.66 383.547 L1130.66 382.071 Q1131.66 381.038 1133.38 379.302 Q1135.11 377.558 1135.55 377.054 Q1136.39 376.108 1136.72 375.457 Q1137.06 374.797 1137.06 374.163 Q1137.06 373.131 1136.33 372.479 Q1135.61 371.828 1134.45 371.828 Q1133.62 371.828 1132.7 372.115 Q1131.79 372.401 1130.75 372.983 L1130.75 371.212 Q1131.81 370.787 1132.73 370.57 Q1133.65 370.353 1134.41 370.353 Q1136.42 370.353 1137.62 371.36 Q1138.82 372.367 1138.82 374.051 Q1138.82 374.849 1138.52 375.57 Q1138.22 376.282 1137.43 377.254 Q1137.21 377.505 1136.05 378.712 Q1134.89 379.91 1132.77 382.071 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1142.87 382.071 L1145.74 382.071 L1145.74 372.184 L1142.62 372.809 L1142.62 371.212 L1145.72 370.587 L1147.47 370.587 L1147.47 382.071 L1150.34 382.071 L1150.34 383.547 L1142.87 383.547 L1142.87 382.071 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1172.62 371.854 L1178.74 371.854 L1178.74 373.33 L1170.52 373.33 L1170.52 371.854 Q1171.51 370.821 1173.23 369.085 Q1174.96 367.341 1175.4 366.837 Q1176.24 365.891 1176.57 365.24 Q1176.91 364.58 1176.91 363.947 Q1176.91 362.914 1176.18 362.262 Q1175.46 361.611 1174.3 361.611 Q1173.48 361.611 1172.56 361.898 Q1171.64 362.184 1170.6 362.766 L1170.6 360.995 Q1171.66 360.57 1172.58 360.353 Q1173.5 360.136 1174.27 360.136 Q1176.28 360.136 1177.48 361.143 Q1178.68 362.15 1178.68 363.834 Q1178.68 364.632 1178.37 365.353 Q1178.08 366.065 1177.29 367.037 Q1177.07 367.289 1175.91 368.495 Q1174.74 369.693 1172.62 371.854 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1183.94 371.854 L1190.06 371.854 L1190.06 373.33 L1181.83 373.33 L1181.83 371.854 Q1182.82 370.821 1184.54 369.085 Q1186.27 367.341 1186.71 366.837 Q1187.56 365.891 1187.89 365.24 Q1188.22 364.58 1188.22 363.947 Q1188.22 362.914 1187.49 362.262 Q1186.77 361.611 1185.61 361.611 Q1184.79 361.611 1183.87 361.898 Q1182.95 362.184 1181.91 362.766 L1181.91 360.995 Q1182.97 360.57 1183.89 360.353 Q1184.81 360.136 1185.58 360.136 Q1187.59 360.136 1188.79 361.143 Q1189.99 362.15 1189.99 363.834 Q1189.99 364.632 1189.68 365.353 Q1189.39 366.065 1188.6 367.037 Q1188.38 367.289 1187.22 368.495 Q1186.05 369.693 1183.94 371.854 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1212.23 362.126 L1218.35 362.126 L1218.35 363.601 L1210.12 363.601 L1210.12 362.126 Q1211.12 361.093 1212.84 359.357 Q1214.57 357.612 1215.01 357.108 Q1215.85 356.162 1216.18 355.511 Q1216.52 354.851 1216.52 354.218 Q1216.52 353.185 1215.79 352.534 Q1215.07 351.883 1213.91 351.883 Q1213.08 351.883 1212.16 352.169 Q1211.25 352.456 1210.21 353.037 L1210.21 351.266 Q1211.27 350.841 1212.19 350.624 Q1213.11 350.407 1213.87 350.407 Q1215.89 350.407 1217.08 351.414 Q1218.28 352.421 1218.28 354.105 Q1218.28 354.904 1217.98 355.624 Q1217.68 356.336 1216.89 357.308 Q1216.68 357.56 1215.51 358.766 Q1214.35 359.964 1212.23 362.126 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1227.34 356.614 Q1228.6 356.883 1229.31 357.733 Q1230.02 358.584 1230.02 359.834 Q1230.02 361.752 1228.7 362.803 Q1227.38 363.853 1224.95 363.853 Q1224.13 363.853 1223.26 363.688 Q1222.41 363.532 1221.49 363.211 L1221.49 361.518 Q1222.21 361.943 1223.08 362.16 Q1223.95 362.377 1224.9 362.377 Q1226.55 362.377 1227.41 361.726 Q1228.27 361.075 1228.27 359.834 Q1228.27 358.688 1227.47 358.046 Q1226.67 357.395 1225.24 357.395 L1223.72 357.395 L1223.72 355.954 L1225.3 355.954 Q1226.6 355.954 1227.28 355.442 Q1227.97 354.921 1227.97 353.949 Q1227.97 352.95 1227.26 352.421 Q1226.55 351.883 1225.24 351.883 Q1224.51 351.883 1223.69 352.039 Q1222.87 352.195 1221.88 352.525 L1221.88 350.963 Q1222.87 350.685 1223.74 350.546 Q1224.62 350.407 1225.39 350.407 Q1227.39 350.407 1228.55 351.319 Q1229.71 352.221 1229.71 353.766 Q1229.71 354.843 1229.1 355.589 Q1228.48 356.327 1227.34 356.614 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1251.8 352.821 L1257.92 352.821 L1257.92 354.297 L1249.7 354.297 L1249.7 352.821 Q1250.69 351.788 1252.41 350.052 Q1254.14 348.307 1254.58 347.804 Q1255.42 346.858 1255.75 346.207 Q1256.09 345.547 1256.09 344.913 Q1256.09 343.88 1255.36 343.229 Q1254.64 342.578 1253.48 342.578 Q1252.66 342.578 1251.74 342.865 Q1250.82 343.151 1249.78 343.733 L1249.78 341.962 Q1250.84 341.537 1251.76 341.32 Q1252.68 341.103 1253.45 341.103 Q1255.46 341.103 1256.66 342.11 Q1257.85 343.117 1257.85 344.801 Q1257.85 345.599 1257.55 346.32 Q1257.26 347.031 1256.47 348.004 Q1256.25 348.255 1255.09 349.462 Q1253.92 350.66 1251.8 352.821 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1266.42 342.865 L1262 349.783 L1266.42 349.783 L1266.42 342.865 M1265.96 341.337 L1268.17 341.337 L1268.17 349.783 L1270.02 349.783 L1270.02 351.241 L1268.17 351.241 L1268.17 354.297 L1266.42 354.297 L1266.42 351.241 L1260.57 351.241 L1260.57 349.549 L1265.96 341.337 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1291.87 343.89 L1297.99 343.89 L1297.99 345.366 L1289.76 345.366 L1289.76 343.89 Q1290.76 342.857 1292.48 341.121 Q1294.2 339.376 1294.65 338.873 Q1295.49 337.927 1295.82 337.276 Q1296.16 336.616 1296.16 335.982 Q1296.16 334.949 1295.43 334.298 Q1294.71 333.647 1293.54 333.647 Q1292.72 333.647 1291.8 333.934 Q1290.89 334.22 1289.84 334.802 L1289.84 333.031 Q1290.9 332.605 1291.82 332.388 Q1292.74 332.171 1293.51 332.171 Q1295.52 332.171 1296.72 333.178 Q1297.92 334.185 1297.92 335.869 Q1297.92 336.668 1297.61 337.388 Q1297.32 338.1 1296.53 339.072 Q1296.31 339.324 1295.15 340.531 Q1293.99 341.729 1291.87 343.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1301.69 332.406 L1308.57 332.406 L1308.57 333.882 L1303.29 333.882 L1303.29 337.059 Q1303.67 336.928 1304.05 336.868 Q1304.44 336.798 1304.82 336.798 Q1306.99 336.798 1308.26 337.987 Q1309.52 339.177 1309.52 341.208 Q1309.52 343.3 1308.22 344.463 Q1306.92 345.618 1304.55 345.618 Q1303.73 345.618 1302.88 345.479 Q1302.04 345.34 1301.14 345.062 L1301.14 343.3 Q1301.92 343.725 1302.75 343.934 Q1303.59 344.142 1304.52 344.142 Q1306.02 344.142 1306.89 343.352 Q1307.77 342.562 1307.77 341.208 Q1307.77 339.854 1306.89 339.064 Q1306.02 338.274 1304.52 338.274 Q1303.81 338.274 1303.11 338.43 Q1302.41 338.586 1301.69 338.916 L1301.69 332.406 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1331.44 335.291 L1337.56 335.291 L1337.56 336.766 L1329.33 336.766 L1329.33 335.291 Q1330.32 334.258 1332.04 332.521 Q1333.77 330.777 1334.21 330.273 Q1335.06 329.327 1335.39 328.676 Q1335.72 328.016 1335.72 327.383 Q1335.72 326.35 1334.99 325.699 Q1334.27 325.048 1333.11 325.048 Q1332.29 325.048 1331.37 325.334 Q1330.45 325.62 1329.41 326.202 L1329.41 324.431 Q1330.47 324.006 1331.39 323.789 Q1332.31 323.572 1333.08 323.572 Q1335.09 323.572 1336.29 324.579 Q1337.49 325.586 1337.49 327.27 Q1337.49 328.068 1337.18 328.789 Q1336.89 329.501 1336.1 330.473 Q1335.88 330.725 1334.72 331.931 Q1333.55 333.129 1331.44 335.291 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1345.2 329.587 Q1344.02 329.587 1343.33 330.395 Q1342.64 331.202 1342.64 332.608 Q1342.64 334.006 1343.33 334.822 Q1344.02 335.629 1345.2 335.629 Q1346.38 335.629 1347.07 334.822 Q1347.76 334.006 1347.76 332.608 Q1347.76 331.202 1347.07 330.395 Q1346.38 329.587 1345.2 329.587 M1348.68 324.093 L1348.68 325.69 Q1348.02 325.377 1347.35 325.212 Q1346.68 325.048 1346.02 325.048 Q1344.28 325.048 1343.36 326.219 Q1342.45 327.391 1342.32 329.761 Q1342.83 329.006 1343.61 328.607 Q1344.38 328.199 1345.31 328.199 Q1347.26 328.199 1348.39 329.388 Q1349.53 330.568 1349.53 332.608 Q1349.53 334.605 1348.35 335.811 Q1347.16 337.018 1345.2 337.018 Q1342.95 337.018 1341.77 335.299 Q1340.58 333.572 1340.58 330.299 Q1340.58 327.226 1342.03 325.403 Q1343.49 323.572 1345.95 323.572 Q1346.61 323.572 1347.28 323.702 Q1347.95 323.832 1348.68 324.093 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1371.42 326.988 L1377.54 326.988 L1377.54 328.464 L1369.31 328.464 L1369.31 326.988 Q1370.31 325.955 1372.03 324.219 Q1373.76 322.474 1374.2 321.971 Q1375.04 321.025 1375.37 320.374 Q1375.71 319.714 1375.71 319.08 Q1375.71 318.047 1374.98 317.396 Q1374.26 316.745 1373.1 316.745 Q1372.27 316.745 1371.35 317.032 Q1370.44 317.318 1369.4 317.9 L1369.4 316.129 Q1370.46 315.703 1371.38 315.486 Q1372.3 315.269 1373.06 315.269 Q1375.08 315.269 1376.27 316.276 Q1377.47 317.283 1377.47 318.967 Q1377.47 319.766 1377.17 320.486 Q1376.87 321.198 1376.08 322.17 Q1375.87 322.422 1374.7 323.629 Q1373.54 324.827 1371.42 326.988 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1380.78 315.504 L1389.11 315.504 L1389.11 316.25 L1384.41 328.464 L1382.58 328.464 L1387 316.979 L1380.78 316.979 L1380.78 315.504 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1411.05 318.954 L1417.17 318.954 L1417.17 320.43 L1408.94 320.43 L1408.94 318.954 Q1409.94 317.921 1411.66 316.185 Q1413.39 314.44 1413.83 313.937 Q1414.67 312.991 1415 312.34 Q1415.34 311.68 1415.34 311.046 Q1415.34 310.013 1414.61 309.362 Q1413.89 308.711 1412.73 308.711 Q1411.9 308.711 1410.98 308.998 Q1410.07 309.284 1409.03 309.866 L1409.03 308.095 Q1410.09 307.669 1411.01 307.452 Q1411.93 307.235 1412.69 307.235 Q1414.71 307.235 1415.91 308.242 Q1417.1 309.249 1417.1 310.933 Q1417.1 311.732 1416.8 312.452 Q1416.51 313.164 1415.72 314.136 Q1415.5 314.388 1414.34 315.595 Q1413.17 316.793 1411.05 318.954 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1424.6 314.275 Q1423.35 314.275 1422.63 314.944 Q1421.92 315.612 1421.92 316.784 Q1421.92 317.956 1422.63 318.624 Q1423.35 319.293 1424.6 319.293 Q1425.85 319.293 1426.57 318.624 Q1427.3 317.947 1427.3 316.784 Q1427.3 315.612 1426.57 314.944 Q1425.86 314.275 1424.6 314.275 M1422.85 313.529 Q1421.72 313.251 1421.09 312.478 Q1420.46 311.706 1420.46 310.595 Q1420.46 309.041 1421.57 308.138 Q1422.68 307.235 1424.6 307.235 Q1426.54 307.235 1427.64 308.138 Q1428.74 309.041 1428.74 310.595 Q1428.74 311.706 1428.11 312.478 Q1427.49 313.251 1426.37 313.529 Q1427.63 313.824 1428.34 314.683 Q1429.05 315.543 1429.05 316.784 Q1429.05 318.668 1427.89 319.675 Q1426.75 320.682 1424.6 320.682 Q1422.46 320.682 1421.31 319.675 Q1420.16 318.668 1420.16 316.784 Q1420.16 315.543 1420.87 314.683 Q1421.58 313.824 1422.85 313.529 M1422.21 310.76 Q1422.21 311.767 1422.83 312.331 Q1423.47 312.895 1424.6 312.895 Q1425.73 312.895 1426.37 312.331 Q1427.01 311.767 1427.01 310.76 Q1427.01 309.753 1426.37 309.189 Q1425.73 308.624 1424.6 308.624 Q1423.47 308.624 1422.83 309.189 Q1422.21 309.753 1422.21 310.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1450.86 311.164 L1456.98 311.164 L1456.98 312.64 L1448.75 312.64 L1448.75 311.164 Q1449.75 310.131 1451.46 308.395 Q1453.19 306.65 1453.63 306.147 Q1454.48 305.2 1454.81 304.549 Q1455.14 303.89 1455.14 303.256 Q1455.14 302.223 1454.42 301.572 Q1453.7 300.921 1452.53 300.921 Q1451.71 300.921 1450.79 301.207 Q1449.88 301.494 1448.83 302.075 L1448.83 300.305 Q1449.89 299.879 1450.81 299.662 Q1451.73 299.445 1452.5 299.445 Q1454.51 299.445 1455.71 300.452 Q1456.91 301.459 1456.91 303.143 Q1456.91 303.942 1456.6 304.662 Q1456.31 305.374 1455.52 306.346 Q1455.3 306.598 1454.14 307.805 Q1452.97 309.003 1450.86 311.164 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1460.71 312.371 L1460.71 310.773 Q1461.37 311.086 1462.05 311.251 Q1462.72 311.416 1463.37 311.416 Q1465.11 311.416 1466.02 310.253 Q1466.94 309.081 1467.07 306.702 Q1466.57 307.449 1465.8 307.848 Q1465.02 308.247 1464.09 308.247 Q1462.14 308.247 1461 307.075 Q1459.88 305.895 1459.88 303.855 Q1459.88 301.858 1461.06 300.652 Q1462.24 299.445 1464.2 299.445 Q1466.45 299.445 1467.63 301.173 Q1468.82 302.891 1468.82 306.173 Q1468.82 309.237 1467.36 311.068 Q1465.91 312.891 1463.45 312.891 Q1462.79 312.891 1462.12 312.761 Q1461.44 312.631 1460.71 312.371 M1464.2 306.876 Q1465.38 306.876 1466.06 306.068 Q1466.76 305.261 1466.76 303.855 Q1466.76 302.457 1466.06 301.65 Q1465.38 300.834 1464.2 300.834 Q1463.02 300.834 1462.32 301.65 Q1461.64 302.457 1461.64 303.855 Q1461.64 305.261 1462.32 306.068 Q1463.02 306.876 1464.2 306.876 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1494.38 298.085 Q1495.64 298.354 1496.34 299.204 Q1497.06 300.055 1497.06 301.305 Q1497.06 303.224 1495.74 304.274 Q1494.42 305.324 1491.99 305.324 Q1491.17 305.324 1490.3 305.159 Q1489.44 305.003 1488.52 304.682 L1488.52 302.989 Q1489.25 303.414 1490.12 303.632 Q1490.99 303.849 1491.94 303.849 Q1493.58 303.849 1494.44 303.197 Q1495.31 302.546 1495.31 301.305 Q1495.31 300.159 1494.5 299.517 Q1493.71 298.866 1492.27 298.866 L1490.76 298.866 L1490.76 297.425 L1492.34 297.425 Q1493.64 297.425 1494.32 296.913 Q1495.01 296.392 1495.01 295.42 Q1495.01 294.421 1494.3 293.892 Q1493.59 293.354 1492.27 293.354 Q1491.55 293.354 1490.73 293.51 Q1489.9 293.666 1488.91 293.996 L1488.91 292.434 Q1489.91 292.156 1490.78 292.017 Q1491.66 291.878 1492.43 291.878 Q1494.43 291.878 1495.59 292.79 Q1496.75 293.692 1496.75 295.237 Q1496.75 296.314 1496.14 297.06 Q1495.52 297.798 1494.38 298.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1504.13 293.267 Q1502.78 293.267 1502.09 294.604 Q1501.41 295.932 1501.41 298.605 Q1501.41 301.27 1502.09 302.607 Q1502.78 303.935 1504.13 303.935 Q1505.49 303.935 1506.17 302.607 Q1506.86 301.27 1506.86 298.605 Q1506.86 295.932 1506.17 294.604 Q1505.49 293.267 1504.13 293.267 M1504.13 291.878 Q1506.31 291.878 1507.46 293.606 Q1508.61 295.324 1508.61 298.605 Q1508.61 301.878 1507.46 303.605 Q1506.31 305.324 1504.13 305.324 Q1501.95 305.324 1500.8 303.605 Q1499.65 301.878 1499.65 298.605 Q1499.65 295.324 1500.8 293.606 Q1501.95 291.878 1504.13 291.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1534.4 290.722 Q1535.66 290.991 1536.36 291.842 Q1537.07 292.693 1537.07 293.943 Q1537.07 295.861 1535.75 296.912 Q1534.43 297.962 1532 297.962 Q1531.19 297.962 1530.32 297.797 Q1529.46 297.641 1528.54 297.32 L1528.54 295.627 Q1529.27 296.052 1530.14 296.269 Q1531 296.486 1531.95 296.486 Q1533.6 296.486 1534.46 295.835 Q1535.33 295.184 1535.33 293.943 Q1535.33 292.797 1534.52 292.155 Q1533.72 291.504 1532.29 291.504 L1530.78 291.504 L1530.78 290.063 L1532.36 290.063 Q1533.65 290.063 1534.34 289.55 Q1535.02 289.03 1535.02 288.057 Q1535.02 287.059 1534.31 286.53 Q1533.61 285.991 1532.29 285.991 Q1531.57 285.991 1530.74 286.148 Q1529.92 286.304 1528.93 286.634 L1528.93 285.071 Q1529.93 284.794 1530.8 284.655 Q1531.67 284.516 1532.45 284.516 Q1534.44 284.516 1535.6 285.427 Q1536.77 286.33 1536.77 287.875 Q1536.77 288.952 1536.15 289.698 Q1535.54 290.436 1534.4 290.722 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1540.7 296.234 L1543.56 296.234 L1543.56 286.347 L1540.45 286.972 L1540.45 285.375 L1543.55 284.75 L1545.3 284.75 L1545.3 296.234 L1548.17 296.234 L1548.17 297.71 L1540.7 297.71 L1540.7 296.234 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1574.25 283.549 Q1575.51 283.818 1576.21 284.669 Q1576.93 285.52 1576.93 286.77 Q1576.93 288.688 1575.61 289.738 Q1574.29 290.789 1571.86 290.789 Q1571.04 290.789 1570.17 290.624 Q1569.31 290.467 1568.39 290.146 L1568.39 288.454 Q1569.12 288.879 1569.99 289.096 Q1570.86 289.313 1571.81 289.313 Q1573.45 289.313 1574.31 288.662 Q1575.18 288.011 1575.18 286.77 Q1575.18 285.624 1574.37 284.981 Q1573.58 284.33 1572.14 284.33 L1570.63 284.33 L1570.63 282.889 L1572.21 282.889 Q1573.51 282.889 1574.19 282.377 Q1574.88 281.856 1574.88 280.884 Q1574.88 279.886 1574.17 279.356 Q1573.46 278.818 1572.14 278.818 Q1571.42 278.818 1570.6 278.974 Q1569.77 279.131 1568.78 279.461 L1568.78 277.898 Q1569.78 277.62 1570.65 277.481 Q1571.53 277.342 1572.3 277.342 Q1574.3 277.342 1575.46 278.254 Q1576.62 279.157 1576.62 280.702 Q1576.62 281.778 1576.01 282.525 Q1575.39 283.263 1574.25 283.549 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1581.76 289.061 L1587.88 289.061 L1587.88 290.537 L1579.65 290.537 L1579.65 289.061 Q1580.65 288.028 1582.37 286.292 Q1584.1 284.547 1584.54 284.044 Q1585.38 283.098 1585.71 282.447 Q1586.05 281.787 1586.05 281.153 Q1586.05 280.12 1585.32 279.469 Q1584.6 278.818 1583.44 278.818 Q1582.61 278.818 1581.69 279.105 Q1580.78 279.391 1579.74 279.973 L1579.74 278.202 Q1580.8 277.777 1581.72 277.56 Q1582.64 277.342 1583.4 277.342 Q1585.42 277.342 1586.61 278.349 Q1587.81 279.356 1587.81 281.04 Q1587.81 281.839 1587.51 282.559 Q1587.21 283.271 1586.42 284.243 Q1586.21 284.495 1585.04 285.702 Q1583.88 286.9 1581.76 289.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1613.86 276.551 Q1615.12 276.82 1615.82 277.671 Q1616.53 278.521 1616.53 279.771 Q1616.53 281.69 1615.21 282.74 Q1613.9 283.791 1611.46 283.791 Q1610.65 283.791 1609.78 283.626 Q1608.92 283.469 1608 283.148 L1608 281.455 Q1608.73 281.881 1609.6 282.098 Q1610.47 282.315 1611.41 282.315 Q1613.06 282.315 1613.92 281.664 Q1614.79 281.013 1614.79 279.771 Q1614.79 278.626 1613.98 277.983 Q1613.18 277.332 1611.75 277.332 L1610.24 277.332 L1610.24 275.891 L1611.82 275.891 Q1613.11 275.891 1613.8 275.379 Q1614.49 274.858 1614.49 273.886 Q1614.49 272.888 1613.77 272.358 Q1613.07 271.82 1611.75 271.82 Q1611.03 271.82 1610.21 271.976 Q1609.38 272.133 1608.39 272.462 L1608.39 270.9 Q1609.39 270.622 1610.26 270.483 Q1611.13 270.344 1611.91 270.344 Q1613.9 270.344 1615.07 271.256 Q1616.23 272.159 1616.23 273.704 Q1616.23 274.78 1615.61 275.527 Q1615 276.265 1613.86 276.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1625.17 276.551 Q1626.43 276.82 1627.13 277.671 Q1627.84 278.521 1627.84 279.771 Q1627.84 281.69 1626.53 282.74 Q1625.21 283.791 1622.78 283.791 Q1621.96 283.791 1621.09 283.626 Q1620.23 283.469 1619.31 283.148 L1619.31 281.455 Q1620.04 281.881 1620.91 282.098 Q1621.78 282.315 1622.72 282.315 Q1624.37 282.315 1625.23 281.664 Q1626.1 281.013 1626.1 279.771 Q1626.1 278.626 1625.29 277.983 Q1624.49 277.332 1623.06 277.332 L1621.55 277.332 L1621.55 275.891 L1623.13 275.891 Q1624.42 275.891 1625.11 275.379 Q1625.8 274.858 1625.8 273.886 Q1625.8 272.888 1625.08 272.358 Q1624.38 271.82 1623.06 271.82 Q1622.34 271.82 1621.52 271.976 Q1620.69 272.133 1619.7 272.462 L1619.7 270.9 Q1620.7 270.622 1621.57 270.483 Q1622.45 270.344 1623.22 270.344 Q1625.21 270.344 1626.38 271.256 Q1627.54 272.159 1627.54 273.704 Q1627.54 274.78 1626.92 275.527 Q1626.31 276.265 1625.17 276.551 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1653.43 269.716 Q1654.69 269.985 1655.39 270.836 Q1656.11 271.686 1656.11 272.936 Q1656.11 274.855 1654.79 275.905 Q1653.47 276.955 1651.04 276.955 Q1650.22 276.955 1649.35 276.79 Q1648.49 276.634 1647.57 276.313 L1647.57 274.62 Q1648.3 275.046 1649.17 275.263 Q1650.04 275.48 1650.98 275.48 Q1652.63 275.48 1653.49 274.829 Q1654.36 274.178 1654.36 272.936 Q1654.36 271.791 1653.55 271.148 Q1652.76 270.497 1651.32 270.497 L1649.81 270.497 L1649.81 269.056 L1651.39 269.056 Q1652.69 269.056 1653.37 268.544 Q1654.06 268.023 1654.06 267.051 Q1654.06 266.053 1653.35 265.523 Q1652.64 264.985 1651.32 264.985 Q1650.6 264.985 1649.78 265.141 Q1648.95 265.297 1647.96 265.627 L1647.96 264.065 Q1648.96 263.787 1649.83 263.648 Q1650.71 263.509 1651.48 263.509 Q1653.48 263.509 1654.64 264.421 Q1655.8 265.324 1655.8 266.869 Q1655.8 267.945 1655.19 268.692 Q1654.57 269.429 1653.43 269.716 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1664.25 265.271 L1659.82 272.19 L1664.25 272.19 L1664.25 265.271 M1663.79 263.744 L1665.99 263.744 L1665.99 272.19 L1667.84 272.19 L1667.84 273.648 L1665.99 273.648 L1665.99 276.704 L1664.25 276.704 L1664.25 273.648 L1658.4 273.648 L1658.4 271.955 L1663.79 263.744 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1693.5 263.033 Q1694.75 263.302 1695.46 264.153 Q1696.17 265.003 1696.17 266.253 Q1696.17 268.172 1694.85 269.222 Q1693.53 270.272 1691.1 270.272 Q1690.28 270.272 1689.42 270.107 Q1688.56 269.951 1687.64 269.63 L1687.64 267.937 Q1688.37 268.363 1689.23 268.58 Q1690.1 268.797 1691.05 268.797 Q1692.7 268.797 1693.56 268.146 Q1694.42 267.495 1694.42 266.253 Q1694.42 265.107 1693.62 264.465 Q1692.82 263.814 1691.39 263.814 L1689.88 263.814 L1689.88 262.373 L1691.46 262.373 Q1692.75 262.373 1693.44 261.861 Q1694.12 261.34 1694.12 260.368 Q1694.12 259.37 1693.41 258.84 Q1692.71 258.302 1691.39 258.302 Q1690.67 258.302 1689.84 258.458 Q1689.02 258.614 1688.03 258.944 L1688.03 257.382 Q1689.03 257.104 1689.89 256.965 Q1690.77 256.826 1691.54 256.826 Q1693.54 256.826 1694.7 257.738 Q1695.87 258.64 1695.87 260.186 Q1695.87 261.262 1695.25 262.009 Q1694.63 262.746 1693.5 263.033 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1699.51 257.061 L1706.4 257.061 L1706.4 258.536 L1701.12 258.536 L1701.12 261.713 Q1701.5 261.583 1701.88 261.522 Q1702.26 261.453 1702.65 261.453 Q1704.82 261.453 1706.08 262.642 Q1707.35 263.831 1707.35 265.863 Q1707.35 267.955 1706.05 269.118 Q1704.75 270.272 1702.38 270.272 Q1701.56 270.272 1700.71 270.133 Q1699.87 269.995 1698.96 269.717 L1698.96 267.955 Q1699.75 268.38 1700.58 268.588 Q1701.41 268.797 1702.34 268.797 Q1703.84 268.797 1704.72 268.007 Q1705.6 267.217 1705.6 265.863 Q1705.6 264.509 1704.72 263.719 Q1703.84 262.929 1702.34 262.929 Q1701.64 262.929 1700.94 263.085 Q1700.24 263.241 1699.51 263.571 L1699.51 257.061 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1733.06 256.492 Q1734.32 256.761 1735.03 257.612 Q1735.74 258.463 1735.74 259.713 Q1735.74 261.631 1734.42 262.681 Q1733.1 263.732 1730.67 263.732 Q1729.85 263.732 1728.98 263.567 Q1728.12 263.411 1727.2 263.089 L1727.2 261.397 Q1727.93 261.822 1728.8 262.039 Q1729.67 262.256 1730.62 262.256 Q1732.27 262.256 1733.12 261.605 Q1733.99 260.954 1733.99 259.713 Q1733.99 258.567 1733.19 257.924 Q1732.39 257.273 1730.95 257.273 L1729.44 257.273 L1729.44 255.832 L1731.02 255.832 Q1732.32 255.832 1733 255.32 Q1733.69 254.799 1733.69 253.827 Q1733.69 252.829 1732.98 252.299 Q1732.27 251.761 1730.95 251.761 Q1730.23 251.761 1729.41 251.918 Q1728.58 252.074 1727.6 252.404 L1727.6 250.841 Q1728.59 250.563 1729.46 250.424 Q1730.34 250.286 1731.11 250.286 Q1733.11 250.286 1734.27 251.197 Q1735.43 252.1 1735.43 253.645 Q1735.43 254.721 1734.82 255.468 Q1734.2 256.206 1733.06 256.492 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1743.03 256.301 Q1741.85 256.301 1741.15 257.108 Q1740.47 257.916 1740.47 259.322 Q1740.47 260.72 1741.15 261.536 Q1741.85 262.343 1743.03 262.343 Q1744.21 262.343 1744.9 261.536 Q1745.59 260.72 1745.59 259.322 Q1745.59 257.916 1744.9 257.108 Q1744.21 256.301 1743.03 256.301 M1746.51 250.806 L1746.51 252.404 Q1745.85 252.091 1745.17 251.926 Q1744.5 251.761 1743.85 251.761 Q1742.11 251.761 1741.19 252.933 Q1740.28 254.105 1740.15 256.475 Q1740.66 255.72 1741.43 255.32 Q1742.2 254.912 1743.13 254.912 Q1745.09 254.912 1746.22 256.102 Q1747.35 257.282 1747.35 259.322 Q1747.35 261.319 1746.17 262.525 Q1744.99 263.732 1743.03 263.732 Q1740.78 263.732 1739.59 262.013 Q1738.4 260.286 1738.4 257.013 Q1738.4 253.94 1739.86 252.117 Q1741.32 250.286 1743.78 250.286 Q1744.44 250.286 1745.1 250.416 Q1745.78 250.546 1746.51 250.806 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1773.05 250.085 Q1774.31 250.354 1775.01 251.205 Q1775.72 252.056 1775.72 253.306 Q1775.72 255.224 1774.4 256.274 Q1773.08 257.325 1770.65 257.325 Q1769.84 257.325 1768.97 257.16 Q1768.11 257.003 1767.19 256.682 L1767.19 254.99 Q1767.92 255.415 1768.79 255.632 Q1769.65 255.849 1770.6 255.849 Q1772.25 255.849 1773.11 255.198 Q1773.98 254.547 1773.98 253.306 Q1773.98 252.16 1773.17 251.517 Q1772.37 250.866 1770.94 250.866 L1769.43 250.866 L1769.43 249.425 L1771.01 249.425 Q1772.3 249.425 1772.99 248.913 Q1773.67 248.392 1773.67 247.42 Q1773.67 246.422 1772.96 245.892 Q1772.26 245.354 1770.94 245.354 Q1770.22 245.354 1769.39 245.51 Q1768.57 245.667 1767.58 245.997 L1767.58 244.434 Q1768.58 244.156 1769.45 244.017 Q1770.32 243.879 1771.1 243.879 Q1773.09 243.879 1774.26 244.79 Q1775.42 245.693 1775.42 247.238 Q1775.42 248.314 1774.8 249.061 Q1774.19 249.799 1773.05 250.085 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1778.6 244.113 L1786.94 244.113 L1786.94 244.859 L1782.23 257.073 L1780.4 257.073 L1784.83 245.589 L1778.6 245.589 L1778.6 244.113 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1812.68 243.804 Q1813.94 244.073 1814.64 244.924 Q1815.36 245.774 1815.36 247.024 Q1815.36 248.943 1814.04 249.993 Q1812.72 251.043 1810.29 251.043 Q1809.47 251.043 1808.6 250.878 Q1807.74 250.722 1806.82 250.401 L1806.82 248.708 Q1807.55 249.134 1808.42 249.351 Q1809.29 249.568 1810.23 249.568 Q1811.88 249.568 1812.74 248.917 Q1813.61 248.266 1813.61 247.024 Q1813.61 245.878 1812.8 245.236 Q1812.01 244.585 1810.57 244.585 L1809.06 244.585 L1809.06 243.144 L1810.64 243.144 Q1811.94 243.144 1812.62 242.632 Q1813.31 242.111 1813.31 241.139 Q1813.31 240.141 1812.6 239.611 Q1811.89 239.073 1810.57 239.073 Q1809.85 239.073 1809.03 239.229 Q1808.2 239.385 1807.21 239.715 L1807.21 238.153 Q1808.21 237.875 1809.08 237.736 Q1809.96 237.597 1810.73 237.597 Q1812.73 237.597 1813.89 238.509 Q1815.05 239.411 1815.05 240.957 Q1815.05 242.033 1814.44 242.78 Q1813.82 243.517 1812.68 243.804 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1822.43 244.637 Q1821.18 244.637 1820.46 245.306 Q1819.75 245.974 1819.75 247.146 Q1819.75 248.318 1820.46 248.986 Q1821.18 249.654 1822.43 249.654 Q1823.68 249.654 1824.4 248.986 Q1825.12 248.309 1825.12 247.146 Q1825.12 245.974 1824.4 245.306 Q1823.69 244.637 1822.43 244.637 M1820.68 243.891 Q1819.55 243.613 1818.91 242.84 Q1818.29 242.068 1818.29 240.957 Q1818.29 239.403 1819.39 238.5 Q1820.5 237.597 1822.43 237.597 Q1824.37 237.597 1825.47 238.5 Q1826.57 239.403 1826.57 240.957 Q1826.57 242.068 1825.94 242.84 Q1825.31 243.613 1824.19 243.891 Q1825.46 244.186 1826.16 245.045 Q1826.87 245.904 1826.87 247.146 Q1826.87 249.029 1825.72 250.036 Q1824.57 251.043 1822.43 251.043 Q1820.29 251.043 1819.13 250.036 Q1817.99 249.029 1817.99 247.146 Q1817.99 245.904 1818.7 245.045 Q1819.41 244.186 1820.68 243.891 M1820.03 241.122 Q1820.03 242.128 1820.66 242.693 Q1821.29 243.257 1822.43 243.257 Q1823.56 243.257 1824.19 242.693 Q1824.83 242.128 1824.83 241.122 Q1824.83 240.115 1824.19 239.55 Q1823.56 238.986 1822.43 238.986 Q1821.29 238.986 1820.66 239.55 Q1820.03 240.115 1820.03 241.122 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1852.48 237.641 Q1853.74 237.91 1854.45 238.761 Q1855.16 239.612 1855.16 240.862 Q1855.16 242.78 1853.84 243.83 Q1852.52 244.881 1850.09 244.881 Q1849.27 244.881 1848.41 244.716 Q1847.55 244.56 1846.63 244.238 L1846.63 242.546 Q1847.35 242.971 1848.22 243.188 Q1849.09 243.405 1850.04 243.405 Q1851.69 243.405 1852.55 242.754 Q1853.41 242.103 1853.41 240.862 Q1853.41 239.716 1852.61 239.073 Q1851.81 238.422 1850.38 238.422 L1848.87 238.422 L1848.87 236.981 L1850.44 236.981 Q1851.74 236.981 1852.42 236.469 Q1853.11 235.948 1853.11 234.976 Q1853.11 233.978 1852.4 233.448 Q1851.69 232.91 1850.38 232.91 Q1849.65 232.91 1848.83 233.067 Q1848.01 233.223 1847.02 233.553 L1847.02 231.99 Q1848.01 231.712 1848.88 231.573 Q1849.76 231.435 1850.53 231.435 Q1852.53 231.435 1853.69 232.346 Q1854.85 233.249 1854.85 234.794 Q1854.85 235.87 1854.24 236.617 Q1853.62 237.355 1852.48 237.641 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1858.54 244.36 L1858.54 242.763 Q1859.19 243.075 1859.87 243.24 Q1860.55 243.405 1861.2 243.405 Q1862.94 243.405 1863.85 242.242 Q1864.77 241.07 1864.9 238.691 Q1864.39 239.438 1863.62 239.837 Q1862.85 240.237 1861.91 240.237 Q1859.97 240.237 1858.83 239.065 Q1857.7 237.884 1857.7 235.844 Q1857.7 233.848 1858.88 232.641 Q1860.06 231.435 1862.02 231.435 Q1864.27 231.435 1865.45 233.162 Q1866.64 234.881 1866.64 238.162 Q1866.64 241.226 1865.18 243.058 Q1863.73 244.881 1861.28 244.881 Q1860.62 244.881 1859.94 244.75 Q1859.26 244.62 1858.54 244.36 M1862.02 238.865 Q1863.21 238.865 1863.89 238.058 Q1864.59 237.251 1864.59 235.844 Q1864.59 234.447 1863.89 233.639 Q1863.21 232.823 1862.02 232.823 Q1860.84 232.823 1860.15 233.639 Q1859.46 234.447 1859.46 235.844 Q1859.46 237.251 1860.15 238.058 Q1860.84 238.865 1862.02 238.865 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1891.98 227.146 L1887.56 234.065 L1891.98 234.065 L1891.98 227.146 M1891.52 225.618 L1893.73 225.618 L1893.73 234.065 L1895.58 234.065 L1895.58 235.523 L1893.73 235.523 L1893.73 238.578 L1891.98 238.578 L1891.98 235.523 L1886.13 235.523 L1886.13 233.83 L1891.52 225.618 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1902.23 226.773 Q1900.87 226.773 1900.19 228.11 Q1899.51 229.438 1899.51 232.111 Q1899.51 234.776 1900.19 236.113 Q1900.87 237.441 1902.23 237.441 Q1903.59 237.441 1904.27 236.113 Q1904.95 234.776 1904.95 232.111 Q1904.95 229.438 1904.27 228.11 Q1903.59 226.773 1902.23 226.773 M1902.23 225.384 Q1904.41 225.384 1905.55 227.111 Q1906.71 228.83 1906.71 232.111 Q1906.71 235.384 1905.55 237.111 Q1904.41 238.83 1902.23 238.83 Q1900.05 238.83 1898.89 237.111 Q1897.75 235.384 1897.75 232.111 Q1897.75 228.83 1898.89 227.111 Q1900.05 225.384 1902.23 225.384 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1932 221.202 L1927.57 228.12 L1932 228.12 L1932 221.202 M1931.54 219.674 L1933.74 219.674 L1933.74 228.12 L1935.59 228.12 L1935.59 229.579 L1933.74 229.579 L1933.74 232.634 L1932 232.634 L1932 229.579 L1926.15 229.579 L1926.15 227.886 L1931.54 219.674 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1938.8 231.159 L1941.66 231.159 L1941.66 221.271 L1938.54 221.896 L1938.54 220.299 L1941.64 219.674 L1943.4 219.674 L1943.4 231.159 L1946.26 231.159 L1946.26 232.634 L1938.8 232.634 L1938.8 231.159 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1971.85 215.359 L1967.43 222.277 L1971.85 222.277 L1971.85 215.359 M1971.39 213.831 L1973.6 213.831 L1973.6 222.277 L1975.45 222.277 L1975.45 223.735 L1973.6 223.735 L1973.6 226.791 L1971.85 226.791 L1971.85 223.735 L1966 223.735 L1966 222.043 L1971.39 213.831 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1979.86 225.315 L1985.98 225.315 L1985.98 226.791 L1977.75 226.791 L1977.75 225.315 Q1978.75 224.282 1980.46 222.546 Q1982.19 220.801 1982.63 220.298 Q1983.48 219.352 1983.81 218.701 Q1984.15 218.041 1984.15 217.407 Q1984.15 216.374 1983.42 215.723 Q1982.7 215.072 1981.53 215.072 Q1980.71 215.072 1979.79 215.359 Q1978.88 215.645 1977.83 216.227 L1977.83 214.456 Q1978.89 214.03 1979.81 213.813 Q1980.73 213.596 1981.5 213.596 Q1983.51 213.596 1984.71 214.603 Q1985.91 215.61 1985.91 217.294 Q1985.91 218.093 1985.6 218.813 Q1985.31 219.525 1984.52 220.497 Q1984.3 220.749 1983.14 221.956 Q1981.98 223.154 1979.86 225.315 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2011.46 209.611 L2007.03 216.529 L2011.46 216.529 L2011.46 209.611 M2011 208.083 L2013.21 208.083 L2013.21 216.529 L2015.05 216.529 L2015.05 217.988 L2013.21 217.988 L2013.21 221.043 L2011.46 221.043 L2011.46 217.988 L2005.61 217.988 L2005.61 216.295 L2011 208.083 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2023.27 214.055 Q2024.52 214.325 2025.23 215.175 Q2025.94 216.026 2025.94 217.276 Q2025.94 219.194 2024.62 220.245 Q2023.3 221.295 2020.87 221.295 Q2020.05 221.295 2019.19 221.13 Q2018.33 220.974 2017.41 220.653 L2017.41 218.96 Q2018.14 219.385 2019 219.602 Q2019.87 219.819 2020.82 219.819 Q2022.47 219.819 2023.33 219.168 Q2024.2 218.517 2024.2 217.276 Q2024.2 216.13 2023.39 215.488 Q2022.59 214.837 2021.16 214.837 L2019.65 214.837 L2019.65 213.396 L2021.23 213.396 Q2022.52 213.396 2023.21 212.884 Q2023.89 212.363 2023.89 211.39 Q2023.89 210.392 2023.18 209.863 Q2022.48 209.325 2021.16 209.325 Q2020.44 209.325 2019.61 209.481 Q2018.79 209.637 2017.8 209.967 L2017.8 208.404 Q2018.8 208.127 2019.66 207.988 Q2020.54 207.849 2021.31 207.849 Q2023.31 207.849 2024.47 208.76 Q2025.64 209.663 2025.64 211.208 Q2025.64 212.285 2025.02 213.031 Q2024.4 213.769 2023.27 214.055 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2051.03 203.955 L2046.61 210.873 L2051.03 210.873 L2051.03 203.955 M2050.57 202.427 L2052.78 202.427 L2052.78 210.873 L2054.63 210.873 L2054.63 212.331 L2052.78 212.331 L2052.78 215.387 L2051.03 215.387 L2051.03 212.331 L2045.18 212.331 L2045.18 210.639 L2050.57 202.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2062.34 203.955 L2057.92 210.873 L2062.34 210.873 L2062.34 203.955 M2061.88 202.427 L2064.09 202.427 L2064.09 210.873 L2065.94 210.873 L2065.94 212.331 L2064.09 212.331 L2064.09 215.387 L2062.34 215.387 L2062.34 212.331 L2056.49 212.331 L2056.49 210.639 L2061.88 202.427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2091.1 198.386 L2086.67 205.304 L2091.1 205.304 L2091.1 198.386 M2090.64 196.858 L2092.84 196.858 L2092.84 205.304 L2094.69 205.304 L2094.69 206.762 L2092.84 206.762 L2092.84 209.818 L2091.1 209.818 L2091.1 206.762 L2085.25 206.762 L2085.25 205.07 L2090.64 196.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2097.61 196.858 L2104.49 196.858 L2104.49 198.333 L2099.21 198.333 L2099.21 201.511 Q2099.59 201.38 2099.98 201.32 Q2100.36 201.25 2100.74 201.25 Q2102.91 201.25 2104.18 202.439 Q2105.45 203.629 2105.45 205.66 Q2105.45 207.752 2104.14 208.915 Q2102.84 210.07 2100.47 210.07 Q2099.66 210.07 2098.8 209.931 Q2097.96 209.792 2097.06 209.514 L2097.06 207.752 Q2097.84 208.177 2098.67 208.386 Q2099.51 208.594 2100.44 208.594 Q2101.94 208.594 2102.82 207.804 Q2103.69 207.014 2103.69 205.66 Q2103.69 204.306 2102.82 203.516 Q2101.94 202.726 2100.44 202.726 Q2099.73 202.726 2099.03 202.882 Q2098.34 203.038 2097.61 203.368 L2097.61 196.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2130.66 192.9 L2126.24 199.818 L2130.66 199.818 L2130.66 192.9 M2130.2 191.372 L2132.41 191.372 L2132.41 199.818 L2134.26 199.818 L2134.26 201.276 L2132.41 201.276 L2132.41 204.332 L2130.66 204.332 L2130.66 201.276 L2124.81 201.276 L2124.81 199.584 L2130.2 191.372 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2141.12 197.153 Q2139.94 197.153 2139.25 197.96 Q2138.56 198.768 2138.56 200.174 Q2138.56 201.571 2139.25 202.387 Q2139.94 203.195 2141.12 203.195 Q2142.31 203.195 2142.99 202.387 Q2143.69 201.571 2143.69 200.174 Q2143.69 198.768 2142.99 197.96 Q2142.31 197.153 2141.12 197.153 M2144.61 191.658 L2144.61 193.256 Q2143.95 192.943 2143.27 192.778 Q2142.6 192.613 2141.94 192.613 Q2140.2 192.613 2139.28 193.785 Q2138.37 194.957 2138.24 197.327 Q2138.75 196.571 2139.53 196.172 Q2140.3 195.764 2141.23 195.764 Q2143.18 195.764 2144.31 196.953 Q2145.45 198.134 2145.45 200.174 Q2145.45 202.17 2144.27 203.377 Q2143.09 204.584 2141.12 204.584 Q2138.88 204.584 2137.69 202.865 Q2136.5 201.137 2136.5 197.865 Q2136.5 194.792 2137.96 192.969 Q2139.41 191.137 2141.87 191.137 Q2142.53 191.137 2143.2 191.268 Q2143.88 191.398 2144.61 191.658 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2170.65 187.493 L2166.22 194.412 L2170.65 194.412 L2170.65 187.493 M2170.19 185.965 L2172.39 185.965 L2172.39 194.412 L2174.24 194.412 L2174.24 195.87 L2172.39 195.87 L2172.39 198.925 L2170.65 198.925 L2170.65 195.87 L2164.8 195.87 L2164.8 194.177 L2170.19 185.965 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2176.7 185.965 L2185.03 185.965 L2185.03 186.712 L2180.33 198.925 L2178.5 198.925 L2182.92 187.441 L2176.7 187.441 L2176.7 185.965 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2210.28 182.163 L2205.86 189.081 L2210.28 189.081 L2210.28 182.163 M2209.82 180.635 L2212.03 180.635 L2212.03 189.081 L2213.88 189.081 L2213.88 190.54 L2212.03 190.54 L2212.03 193.595 L2210.28 193.595 L2210.28 190.54 L2204.43 190.54 L2204.43 188.847 L2209.82 180.635 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2220.53 187.441 Q2219.28 187.441 2218.56 188.109 Q2217.84 188.778 2217.84 189.949 Q2217.84 191.121 2218.56 191.79 Q2219.28 192.458 2220.53 192.458 Q2221.78 192.458 2222.5 191.79 Q2223.22 191.113 2223.22 189.949 Q2223.22 188.778 2222.5 188.109 Q2221.78 187.441 2220.53 187.441 M2218.77 186.694 Q2217.64 186.416 2217.01 185.644 Q2216.39 184.871 2216.39 183.76 Q2216.39 182.206 2217.49 181.304 Q2218.6 180.401 2220.53 180.401 Q2222.46 180.401 2223.56 181.304 Q2224.67 182.206 2224.67 183.76 Q2224.67 184.871 2224.03 185.644 Q2223.41 186.416 2222.29 186.694 Q2223.56 186.989 2224.26 187.849 Q2224.97 188.708 2224.97 189.949 Q2224.97 191.833 2223.82 192.84 Q2222.67 193.847 2220.53 193.847 Q2218.38 193.847 2217.23 192.84 Q2216.08 191.833 2216.08 189.949 Q2216.08 188.708 2216.79 187.849 Q2217.5 186.989 2218.77 186.694 M2218.13 183.925 Q2218.13 184.932 2218.75 185.496 Q2219.39 186.061 2220.53 186.061 Q2221.65 186.061 2222.29 185.496 Q2222.93 184.932 2222.93 183.925 Q2222.93 182.918 2222.29 182.354 Q2221.65 181.79 2220.53 181.79 Q2219.39 181.79 2218.75 182.354 Q2218.13 182.918 2218.13 183.925 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2250.09 176.906 L2245.66 183.824 L2250.09 183.824 L2250.09 176.906 M2249.63 175.378 L2251.83 175.378 L2251.83 183.824 L2253.68 183.824 L2253.68 185.283 L2251.83 185.283 L2251.83 188.338 L2250.09 188.338 L2250.09 185.283 L2244.23 185.283 L2244.23 183.59 L2249.63 175.378 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2256.63 188.069 L2256.63 186.472 Q2257.29 186.784 2257.97 186.949 Q2258.64 187.114 2259.3 187.114 Q2261.03 187.114 2261.94 185.951 Q2262.86 184.779 2262.99 182.401 Q2262.49 183.147 2261.72 183.546 Q2260.94 183.946 2260.01 183.946 Q2258.06 183.946 2256.93 182.774 Q2255.8 181.593 2255.8 179.553 Q2255.8 177.557 2256.98 176.35 Q2258.16 175.144 2260.12 175.144 Q2262.37 175.144 2263.55 176.871 Q2264.74 178.59 2264.74 181.871 Q2264.74 184.935 2263.28 186.767 Q2261.83 188.59 2259.37 188.59 Q2258.71 188.59 2258.04 188.46 Q2257.36 188.329 2256.63 188.069 M2260.12 182.574 Q2261.3 182.574 2261.99 181.767 Q2262.68 180.96 2262.68 179.553 Q2262.68 178.156 2261.99 177.349 Q2261.3 176.533 2260.12 176.533 Q2258.94 176.533 2258.25 177.349 Q2257.56 178.156 2257.56 179.553 Q2257.56 180.96 2258.25 181.767 Q2258.94 182.574 2260.12 182.574 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2284.78 170.191 L2291.67 170.191 L2291.67 171.667 L2286.39 171.667 L2286.39 174.844 Q2286.77 174.714 2287.15 174.653 Q2287.54 174.583 2287.92 174.583 Q2290.09 174.583 2291.35 175.773 Q2292.62 176.962 2292.62 178.993 Q2292.62 181.085 2291.32 182.248 Q2290.02 183.403 2287.65 183.403 Q2286.83 183.403 2285.98 183.264 Q2285.14 183.125 2284.24 182.847 L2284.24 181.085 Q2285.02 181.51 2285.85 181.719 Q2286.68 181.927 2287.61 181.927 Q2289.12 181.927 2289.99 181.137 Q2290.87 180.347 2290.87 178.993 Q2290.87 177.639 2289.99 176.849 Q2289.12 176.059 2287.61 176.059 Q2286.91 176.059 2286.21 176.215 Q2285.51 176.372 2284.78 176.701 L2284.78 170.191 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2299.83 171.346 Q2298.47 171.346 2297.79 172.682 Q2297.11 174.01 2297.11 176.684 Q2297.11 179.349 2297.79 180.686 Q2298.47 182.014 2299.83 182.014 Q2301.19 182.014 2301.87 180.686 Q2302.55 179.349 2302.55 176.684 Q2302.55 174.01 2301.87 172.682 Q2301.19 171.346 2299.83 171.346 M2299.83 169.957 Q2302.01 169.957 2303.15 171.684 Q2304.31 173.403 2304.31 176.684 Q2304.31 179.957 2303.15 181.684 Q2302.01 183.403 2299.83 183.403 Q2297.65 183.403 2296.49 181.684 Q2295.35 179.957 2295.35 176.684 Q2295.35 173.403 2296.49 171.684 Q2297.65 169.957 2299.83 169.957 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M286.311 2174.14 L2352.76 2174.14 L2352.76 1362.92 L286.311 1362.92  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip093">
+  <clipPath id="clip913">
     <rect x="286" y="1362" width="2067" height="812"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="377.811,2174.14 377.811,1362.92 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="747.114,2174.14 747.114,1362.92 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1116.42,2174.14 1116.42,1362.92 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1485.72,2174.14 1485.72,1362.92 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1855.02,2174.14 1855.02,1362.92 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2224.33,2174.14 2224.33,1362.92 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 2352.76,2174.14 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="377.811,2174.14 377.811,2155.24 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.114,2174.14 747.114,2155.24 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1116.42,2174.14 1116.42,2155.24 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1485.72,2174.14 1485.72,2155.24 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1855.02,2174.14 1855.02,2155.24 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2224.33,2174.14 2224.33,2155.24 "/>
-<path clip-path="url(#clip090)" d="M377.811 2205.06 Q374.2 2205.06 372.371 2208.62 Q370.566 2212.16 370.566 2219.29 Q370.566 2226.4 372.371 2229.97 Q374.2 2233.51 377.811 2233.51 Q381.445 2233.51 383.251 2229.97 Q385.08 2226.4 385.08 2219.29 Q385.08 2212.16 383.251 2208.62 Q381.445 2205.06 377.811 2205.06 M377.811 2201.35 Q383.621 2201.35 386.677 2205.96 Q389.756 2210.54 389.756 2219.29 Q389.756 2228.02 386.677 2232.63 Q383.621 2237.21 377.811 2237.21 Q372.001 2237.21 368.922 2232.63 Q365.867 2228.02 365.867 2219.29 Q365.867 2210.54 368.922 2205.96 Q372.001 2201.35 377.811 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M721.802 2232.6 L729.441 2232.6 L729.441 2206.24 L721.13 2207.91 L721.13 2203.65 L729.394 2201.98 L734.07 2201.98 L734.07 2232.6 L741.709 2232.6 L741.709 2236.54 L721.802 2236.54 L721.802 2232.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M761.153 2205.06 Q757.542 2205.06 755.714 2208.62 Q753.908 2212.16 753.908 2219.29 Q753.908 2226.4 755.714 2229.97 Q757.542 2233.51 761.153 2233.51 Q764.788 2233.51 766.593 2229.97 Q768.422 2226.4 768.422 2219.29 Q768.422 2212.16 766.593 2208.62 Q764.788 2205.06 761.153 2205.06 M761.153 2201.35 Q766.964 2201.35 770.019 2205.96 Q773.098 2210.54 773.098 2219.29 Q773.098 2228.02 770.019 2232.63 Q766.964 2237.21 761.153 2237.21 Q755.343 2237.21 752.265 2232.63 Q749.209 2228.02 749.209 2219.29 Q749.209 2210.54 752.265 2205.96 Q755.343 2201.35 761.153 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1095.19 2232.6 L1111.51 2232.6 L1111.51 2236.54 L1089.57 2236.54 L1089.57 2232.6 Q1092.23 2229.85 1096.81 2225.22 Q1101.42 2220.57 1102.6 2219.23 Q1104.84 2216.7 1105.72 2214.97 Q1106.63 2213.21 1106.63 2211.52 Q1106.63 2208.76 1104.68 2207.03 Q1102.76 2205.29 1099.66 2205.29 Q1097.46 2205.29 1095.01 2206.05 Q1092.57 2206.82 1089.8 2208.37 L1089.8 2203.65 Q1092.62 2202.51 1095.07 2201.93 Q1097.53 2201.35 1099.57 2201.35 Q1104.94 2201.35 1108.13 2204.04 Q1111.32 2206.73 1111.32 2211.22 Q1111.32 2213.35 1110.51 2215.27 Q1109.73 2217.16 1107.62 2219.76 Q1107.04 2220.43 1103.94 2223.65 Q1100.84 2226.84 1095.19 2232.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1131.32 2205.06 Q1127.71 2205.06 1125.88 2208.62 Q1124.08 2212.16 1124.08 2219.29 Q1124.08 2226.4 1125.88 2229.97 Q1127.71 2233.51 1131.32 2233.51 Q1134.96 2233.51 1136.76 2229.97 Q1138.59 2226.4 1138.59 2219.29 Q1138.59 2212.16 1136.76 2208.62 Q1134.96 2205.06 1131.32 2205.06 M1131.32 2201.35 Q1137.13 2201.35 1140.19 2205.96 Q1143.27 2210.54 1143.27 2219.29 Q1143.27 2228.02 1140.19 2232.63 Q1137.13 2237.21 1131.32 2237.21 Q1125.51 2237.21 1122.44 2232.63 Q1119.38 2228.02 1119.38 2219.29 Q1119.38 2210.54 1122.44 2205.96 Q1125.51 2201.35 1131.32 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1474.56 2217.91 Q1477.92 2218.62 1479.79 2220.89 Q1481.69 2223.16 1481.69 2226.49 Q1481.69 2231.61 1478.17 2234.41 Q1474.66 2237.21 1468.17 2237.21 Q1466 2237.21 1463.68 2236.77 Q1461.39 2236.35 1458.94 2235.5 L1458.94 2230.98 Q1460.88 2232.12 1463.2 2232.7 Q1465.51 2233.28 1468.03 2233.28 Q1472.43 2233.28 1474.72 2231.54 Q1477.04 2229.8 1477.04 2226.49 Q1477.04 2223.44 1474.89 2221.73 Q1472.76 2219.99 1468.94 2219.99 L1464.91 2219.99 L1464.91 2216.15 L1469.12 2216.15 Q1472.57 2216.15 1474.4 2214.78 Q1476.23 2213.39 1476.23 2210.8 Q1476.23 2208.14 1474.33 2206.73 Q1472.46 2205.29 1468.94 2205.29 Q1467.02 2205.29 1464.82 2205.71 Q1462.62 2206.12 1459.98 2207 L1459.98 2202.84 Q1462.64 2202.1 1464.96 2201.73 Q1467.29 2201.35 1469.35 2201.35 Q1474.68 2201.35 1477.78 2203.79 Q1480.88 2206.19 1480.88 2210.31 Q1480.88 2213.18 1479.24 2215.17 Q1477.59 2217.14 1474.56 2217.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1500.56 2205.06 Q1496.95 2205.06 1495.12 2208.62 Q1493.31 2212.16 1493.31 2219.29 Q1493.31 2226.4 1495.12 2229.97 Q1496.95 2233.51 1500.56 2233.51 Q1504.19 2233.51 1506 2229.97 Q1507.83 2226.4 1507.83 2219.29 Q1507.83 2212.16 1506 2208.62 Q1504.19 2205.06 1500.56 2205.06 M1500.56 2201.35 Q1506.37 2201.35 1509.42 2205.96 Q1512.5 2210.54 1512.5 2219.29 Q1512.5 2228.02 1509.42 2232.63 Q1506.37 2237.21 1500.56 2237.21 Q1494.75 2237.21 1491.67 2232.63 Q1488.61 2228.02 1488.61 2219.29 Q1488.61 2210.54 1491.67 2205.96 Q1494.75 2201.35 1500.56 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1843.19 2206.05 L1831.39 2224.5 L1843.19 2224.5 L1843.19 2206.05 M1841.97 2201.98 L1847.85 2201.98 L1847.85 2224.5 L1852.78 2224.5 L1852.78 2228.39 L1847.85 2228.39 L1847.85 2236.54 L1843.19 2236.54 L1843.19 2228.39 L1827.59 2228.39 L1827.59 2223.88 L1841.97 2201.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1870.51 2205.06 Q1866.9 2205.06 1865.07 2208.62 Q1863.26 2212.16 1863.26 2219.29 Q1863.26 2226.4 1865.07 2229.97 Q1866.9 2233.51 1870.51 2233.51 Q1874.14 2233.51 1875.95 2229.97 Q1877.78 2226.4 1877.78 2219.29 Q1877.78 2212.16 1875.95 2208.62 Q1874.14 2205.06 1870.51 2205.06 M1870.51 2201.35 Q1876.32 2201.35 1879.37 2205.96 Q1882.45 2210.54 1882.45 2219.29 Q1882.45 2228.02 1879.37 2232.63 Q1876.32 2237.21 1870.51 2237.21 Q1864.7 2237.21 1861.62 2232.63 Q1858.56 2228.02 1858.56 2219.29 Q1858.56 2210.54 1861.62 2205.96 Q1864.7 2201.35 1870.51 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2199.02 2201.98 L2217.38 2201.98 L2217.38 2205.92 L2203.31 2205.92 L2203.31 2214.39 Q2204.33 2214.04 2205.34 2213.88 Q2206.36 2213.69 2207.38 2213.69 Q2213.17 2213.69 2216.55 2216.86 Q2219.93 2220.04 2219.93 2225.45 Q2219.93 2231.03 2216.46 2234.13 Q2212.98 2237.21 2206.66 2237.21 Q2204.49 2237.21 2202.22 2236.84 Q2199.97 2236.47 2197.57 2235.73 L2197.57 2231.03 Q2199.65 2232.16 2201.87 2232.72 Q2204.09 2233.28 2206.57 2233.28 Q2210.58 2233.28 2212.91 2231.17 Q2215.25 2229.06 2215.25 2225.45 Q2215.25 2221.84 2212.91 2219.73 Q2210.58 2217.63 2206.57 2217.63 Q2204.7 2217.63 2202.82 2218.04 Q2200.97 2218.46 2199.02 2219.34 L2199.02 2201.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2239.14 2205.06 Q2235.53 2205.06 2233.7 2208.62 Q2231.9 2212.16 2231.9 2219.29 Q2231.9 2226.4 2233.7 2229.97 Q2235.53 2233.51 2239.14 2233.51 Q2242.77 2233.51 2244.58 2229.97 Q2246.41 2226.4 2246.41 2219.29 Q2246.41 2212.16 2244.58 2208.62 Q2242.77 2205.06 2239.14 2205.06 M2239.14 2201.35 Q2244.95 2201.35 2248.01 2205.96 Q2251.08 2210.54 2251.08 2219.29 Q2251.08 2228.02 2248.01 2232.63 Q2244.95 2237.21 2239.14 2237.21 Q2233.33 2237.21 2230.25 2232.63 Q2227.2 2228.02 2227.2 2219.29 Q2227.2 2210.54 2230.25 2205.96 Q2233.33 2201.35 2239.14 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1184.33 2273.04 L1184.33 2279.31 Q1180.67 2277.56 1177.42 2276.7 Q1174.17 2275.84 1171.15 2275.84 Q1165.9 2275.84 1163.03 2277.88 Q1160.2 2279.92 1160.2 2283.67 Q1160.2 2286.83 1162.08 2288.45 Q1163.99 2290.04 1169.27 2291.03 L1173.15 2291.82 Q1180.35 2293.19 1183.75 2296.66 Q1187.19 2300.1 1187.19 2305.89 Q1187.19 2312.8 1182.54 2316.36 Q1177.93 2319.93 1168.98 2319.93 Q1165.61 2319.93 1161.79 2319.16 Q1158 2318.4 1153.93 2316.9 L1153.93 2310.28 Q1157.84 2312.48 1161.6 2313.59 Q1165.36 2314.71 1168.98 2314.71 Q1174.49 2314.71 1177.48 2312.54 Q1180.47 2310.38 1180.47 2306.37 Q1180.47 2302.87 1178.31 2300.89 Q1176.18 2298.92 1171.28 2297.93 L1167.36 2297.17 Q1160.17 2295.74 1156.95 2292.68 Q1153.74 2289.63 1153.74 2284.18 Q1153.74 2277.88 1158.16 2274.25 Q1162.62 2270.62 1170.42 2270.62 Q1173.76 2270.62 1177.23 2271.23 Q1180.7 2271.83 1184.33 2273.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1210.78 2287.46 Q1206.06 2287.46 1203.33 2291.15 Q1200.59 2294.81 1200.59 2301.21 Q1200.59 2307.61 1203.3 2311.3 Q1206.03 2314.96 1210.78 2314.96 Q1215.45 2314.96 1218.19 2311.27 Q1220.93 2307.58 1220.93 2301.21 Q1220.93 2294.88 1218.19 2291.19 Q1215.45 2287.46 1210.78 2287.46 M1210.78 2282.5 Q1218.41 2282.5 1222.77 2287.46 Q1227.14 2292.43 1227.14 2301.21 Q1227.14 2309.96 1222.77 2314.96 Q1218.41 2319.93 1210.78 2319.93 Q1203.1 2319.93 1198.74 2314.96 Q1194.42 2309.96 1194.42 2301.21 Q1194.42 2292.43 1198.74 2287.46 Q1203.1 2282.5 1210.78 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1236.84 2269.48 L1242.7 2269.48 L1242.7 2319 L1236.84 2319 L1236.84 2269.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1254.35 2304.94 L1254.35 2283.36 L1260.21 2283.36 L1260.21 2304.71 Q1260.21 2309.77 1262.18 2312.32 Q1264.15 2314.83 1268.1 2314.83 Q1272.84 2314.83 1275.58 2311.81 Q1278.35 2308.79 1278.35 2303.57 L1278.35 2283.36 L1284.2 2283.36 L1284.2 2319 L1278.35 2319 L1278.35 2313.53 Q1276.21 2316.78 1273.38 2318.37 Q1270.58 2319.93 1266.86 2319.93 Q1260.71 2319.93 1257.53 2316.11 Q1254.35 2312.29 1254.35 2304.94 M1269.09 2282.5 L1269.09 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1302.06 2273.23 L1302.06 2283.36 L1314.12 2283.36 L1314.12 2287.91 L1302.06 2287.91 L1302.06 2307.26 Q1302.06 2311.62 1303.24 2312.86 Q1304.45 2314.1 1308.11 2314.1 L1314.12 2314.1 L1314.12 2319 L1308.11 2319 Q1301.33 2319 1298.75 2316.49 Q1296.17 2313.94 1296.17 2307.26 L1296.17 2287.91 L1291.87 2287.91 L1291.87 2283.36 L1296.17 2283.36 L1296.17 2273.23 L1302.06 2273.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1321.83 2283.36 L1327.68 2283.36 L1327.68 2319 L1321.83 2319 L1321.83 2283.36 M1321.83 2269.48 L1327.68 2269.48 L1327.68 2276.89 L1321.83 2276.89 L1321.83 2269.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1353.75 2287.46 Q1349.04 2287.46 1346.3 2291.15 Q1343.56 2294.81 1343.56 2301.21 Q1343.56 2307.61 1346.27 2311.3 Q1349.01 2314.96 1353.75 2314.96 Q1358.43 2314.96 1361.17 2311.27 Q1363.9 2307.58 1363.9 2301.21 Q1363.9 2294.88 1361.17 2291.19 Q1358.43 2287.46 1353.75 2287.46 M1353.75 2282.5 Q1361.39 2282.5 1365.75 2287.46 Q1370.11 2292.43 1370.11 2301.21 Q1370.11 2309.96 1365.75 2314.96 Q1361.39 2319.93 1353.75 2319.93 Q1346.08 2319.93 1341.72 2314.96 Q1337.39 2309.96 1337.39 2301.21 Q1337.39 2292.43 1341.72 2287.46 Q1346.08 2282.5 1353.75 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1409.45 2297.49 L1409.45 2319 L1403.59 2319 L1403.59 2297.68 Q1403.59 2292.62 1401.62 2290.1 Q1399.65 2287.59 1395.7 2287.59 Q1390.96 2287.59 1388.22 2290.61 Q1385.48 2293.64 1385.48 2298.86 L1385.48 2319 L1379.59 2319 L1379.59 2283.36 L1385.48 2283.36 L1385.48 2288.89 Q1387.58 2285.68 1390.42 2284.09 Q1393.28 2282.5 1397 2282.5 Q1403.15 2282.5 1406.3 2286.32 Q1409.45 2290.1 1409.45 2297.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1469.03 2290.33 L1459.77 2290.33 L1457.1 2300.96 L1466.42 2300.96 L1469.03 2290.33 M1464.26 2272.22 L1460.95 2285.42 L1470.24 2285.42 L1473.58 2272.22 L1478.68 2272.22 L1475.4 2285.42 L1485.33 2285.42 L1485.33 2290.33 L1474.16 2290.33 L1471.55 2300.96 L1481.67 2300.96 L1481.67 2305.83 L1470.31 2305.83 L1467 2319 L1461.9 2319 L1465.18 2305.83 L1455.86 2305.83 L1452.58 2319 L1447.45 2319 L1450.76 2305.83 L1440.74 2305.83 L1440.74 2300.96 L1451.94 2300.96 L1454.61 2290.33 L1444.37 2290.33 L1444.37 2285.42 L1455.86 2285.42 L1459.1 2272.22 L1464.26 2272.22 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,2174.14 2352.76,2174.14 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1971.34 2352.76,1971.34 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1768.53 2352.76,1768.53 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1565.73 2352.76,1565.73 "/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1362.92 2352.76,1362.92 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 286.311,1362.92 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 305.209,2174.14 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1971.34 305.209,1971.34 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1768.53 305.209,1768.53 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1565.73 305.209,1565.73 "/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1362.92 305.209,1362.92 "/>
-<path clip-path="url(#clip090)" d="M238.367 2159.94 Q234.756 2159.94 232.927 2163.5 Q231.122 2167.04 231.122 2174.17 Q231.122 2181.28 232.927 2184.85 Q234.756 2188.39 238.367 2188.39 Q242.001 2188.39 243.807 2184.85 Q245.635 2181.28 245.635 2174.17 Q245.635 2167.04 243.807 2163.5 Q242.001 2159.94 238.367 2159.94 M238.367 2156.23 Q244.177 2156.23 247.233 2160.84 Q250.311 2165.42 250.311 2174.17 Q250.311 2182.9 247.233 2187.51 Q244.177 2192.09 238.367 2192.09 Q232.557 2192.09 229.478 2187.51 Q226.422 2182.9 226.422 2174.17 Q226.422 2165.42 229.478 2160.84 Q232.557 2156.23 238.367 2156.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M172.071 1984.68 L188.39 1984.68 L188.39 1988.62 L166.446 1988.62 L166.446 1984.68 Q169.108 1981.93 173.691 1977.3 Q178.298 1972.64 179.478 1971.3 Q181.724 1968.78 182.603 1967.04 Q183.506 1965.28 183.506 1963.59 Q183.506 1960.84 181.562 1959.1 Q179.64 1957.37 176.538 1957.37 Q174.339 1957.37 171.886 1958.13 Q169.455 1958.89 166.677 1960.44 L166.677 1955.72 Q169.501 1954.59 171.955 1954.01 Q174.409 1953.43 176.446 1953.43 Q181.816 1953.43 185.011 1956.12 Q188.205 1958.8 188.205 1963.29 Q188.205 1965.42 187.395 1967.34 Q186.608 1969.24 184.501 1971.83 Q183.923 1972.5 180.821 1975.72 Q177.719 1978.92 172.071 1984.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M198.251 1954.06 L216.608 1954.06 L216.608 1957.99 L202.534 1957.99 L202.534 1966.46 Q203.552 1966.12 204.571 1965.95 Q205.589 1965.77 206.608 1965.77 Q212.395 1965.77 215.774 1968.94 Q219.154 1972.11 219.154 1977.53 Q219.154 1983.11 215.682 1986.21 Q212.21 1989.29 205.89 1989.29 Q203.714 1989.29 201.446 1988.92 Q199.2 1988.55 196.793 1987.81 L196.793 1983.11 Q198.876 1984.24 201.099 1984.8 Q203.321 1985.35 205.798 1985.35 Q209.802 1985.35 212.14 1983.25 Q214.478 1981.14 214.478 1977.53 Q214.478 1973.92 212.14 1971.81 Q209.802 1969.7 205.798 1969.7 Q203.923 1969.7 202.048 1970.12 Q200.196 1970.54 198.251 1971.42 L198.251 1954.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 1957.13 Q234.756 1957.13 232.927 1960.7 Q231.122 1964.24 231.122 1971.37 Q231.122 1978.48 232.927 1982.04 Q234.756 1985.58 238.367 1985.58 Q242.001 1985.58 243.807 1982.04 Q245.635 1978.48 245.635 1971.37 Q245.635 1964.24 243.807 1960.7 Q242.001 1957.13 238.367 1957.13 M238.367 1953.43 Q244.177 1953.43 247.233 1958.04 Q250.311 1962.62 250.311 1971.37 Q250.311 1980.1 247.233 1984.7 Q244.177 1989.29 238.367 1989.29 Q232.557 1989.29 229.478 1984.7 Q226.422 1980.1 226.422 1971.37 Q226.422 1962.62 229.478 1958.04 Q232.557 1953.43 238.367 1953.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M168.089 1751.25 L186.446 1751.25 L186.446 1755.19 L172.372 1755.19 L172.372 1763.66 Q173.39 1763.31 174.409 1763.15 Q175.427 1762.97 176.446 1762.97 Q182.233 1762.97 185.612 1766.14 Q188.992 1769.31 188.992 1774.72 Q188.992 1780.3 185.52 1783.4 Q182.048 1786.48 175.728 1786.48 Q173.552 1786.48 171.284 1786.11 Q169.039 1785.74 166.631 1785 L166.631 1780.3 Q168.714 1781.44 170.937 1781.99 Q173.159 1782.55 175.636 1782.55 Q179.64 1782.55 181.978 1780.44 Q184.316 1778.34 184.316 1774.72 Q184.316 1771.11 181.978 1769.01 Q179.64 1766.9 175.636 1766.9 Q173.761 1766.9 171.886 1767.32 Q170.034 1767.73 168.089 1768.61 L168.089 1751.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M208.205 1754.33 Q204.594 1754.33 202.765 1757.9 Q200.96 1761.44 200.96 1768.57 Q200.96 1775.67 202.765 1779.24 Q204.594 1782.78 208.205 1782.78 Q211.839 1782.78 213.645 1779.24 Q215.473 1775.67 215.473 1768.57 Q215.473 1761.44 213.645 1757.9 Q211.839 1754.33 208.205 1754.33 M208.205 1750.63 Q214.015 1750.63 217.071 1755.23 Q220.149 1759.82 220.149 1768.57 Q220.149 1777.29 217.071 1781.9 Q214.015 1786.48 208.205 1786.48 Q202.395 1786.48 199.316 1781.9 Q196.261 1777.29 196.261 1768.57 Q196.261 1759.82 199.316 1755.23 Q202.395 1750.63 208.205 1750.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 1754.33 Q234.756 1754.33 232.927 1757.9 Q231.122 1761.44 231.122 1768.57 Q231.122 1775.67 232.927 1779.24 Q234.756 1782.78 238.367 1782.78 Q242.001 1782.78 243.807 1779.24 Q245.635 1775.67 245.635 1768.57 Q245.635 1761.44 243.807 1757.9 Q242.001 1754.33 238.367 1754.33 M238.367 1750.63 Q244.177 1750.63 247.233 1755.23 Q250.311 1759.82 250.311 1768.57 Q250.311 1777.29 247.233 1781.9 Q244.177 1786.48 238.367 1786.48 Q232.557 1786.48 229.478 1781.9 Q226.422 1777.29 226.422 1768.57 Q226.422 1759.82 229.478 1755.23 Q232.557 1750.63 238.367 1750.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M166.863 1548.45 L189.085 1548.45 L189.085 1550.44 L176.538 1583.01 L171.654 1583.01 L183.46 1552.38 L166.863 1552.38 L166.863 1548.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M198.251 1548.45 L216.608 1548.45 L216.608 1552.38 L202.534 1552.38 L202.534 1560.86 Q203.552 1560.51 204.571 1560.35 Q205.589 1560.16 206.608 1560.16 Q212.395 1560.16 215.774 1563.33 Q219.154 1566.5 219.154 1571.92 Q219.154 1577.5 215.682 1580.6 Q212.21 1583.68 205.89 1583.68 Q203.714 1583.68 201.446 1583.31 Q199.2 1582.94 196.793 1582.2 L196.793 1577.5 Q198.876 1578.63 201.099 1579.19 Q203.321 1579.74 205.798 1579.74 Q209.802 1579.74 212.14 1577.64 Q214.478 1575.53 214.478 1571.92 Q214.478 1568.31 212.14 1566.2 Q209.802 1564.1 205.798 1564.1 Q203.923 1564.1 202.048 1564.51 Q200.196 1564.93 198.251 1565.81 L198.251 1548.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 1551.53 Q234.756 1551.53 232.927 1555.09 Q231.122 1558.63 231.122 1565.76 Q231.122 1572.87 232.927 1576.43 Q234.756 1579.98 238.367 1579.98 Q242.001 1579.98 243.807 1576.43 Q245.635 1572.87 245.635 1565.76 Q245.635 1558.63 243.807 1555.09 Q242.001 1551.53 238.367 1551.53 M238.367 1547.82 Q244.177 1547.82 247.233 1552.43 Q250.311 1557.01 250.311 1565.76 Q250.311 1574.49 247.233 1579.1 Q244.177 1583.68 238.367 1583.68 Q232.557 1583.68 229.478 1579.1 Q226.422 1574.49 226.422 1565.76 Q226.422 1557.01 229.478 1552.43 Q232.557 1547.82 238.367 1547.82 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M138.691 1376.27 L146.33 1376.27 L146.33 1349.9 L138.02 1351.57 L138.02 1347.31 L146.284 1345.64 L150.96 1345.64 L150.96 1376.27 L158.599 1376.27 L158.599 1380.2 L138.691 1380.2 L138.691 1376.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M178.043 1348.72 Q174.432 1348.72 172.603 1352.29 Q170.798 1355.83 170.798 1362.96 Q170.798 1370.07 172.603 1373.63 Q174.432 1377.17 178.043 1377.17 Q181.677 1377.17 183.483 1373.63 Q185.312 1370.07 185.312 1362.96 Q185.312 1355.83 183.483 1352.29 Q181.677 1348.72 178.043 1348.72 M178.043 1345.02 Q183.853 1345.02 186.909 1349.63 Q189.987 1354.21 189.987 1362.96 Q189.987 1371.69 186.909 1376.29 Q183.853 1380.88 178.043 1380.88 Q172.233 1380.88 169.154 1376.29 Q166.099 1371.69 166.099 1362.96 Q166.099 1354.21 169.154 1349.63 Q172.233 1345.02 178.043 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M208.205 1348.72 Q204.594 1348.72 202.765 1352.29 Q200.96 1355.83 200.96 1362.96 Q200.96 1370.07 202.765 1373.63 Q204.594 1377.17 208.205 1377.17 Q211.839 1377.17 213.645 1373.63 Q215.473 1370.07 215.473 1362.96 Q215.473 1355.83 213.645 1352.29 Q211.839 1348.72 208.205 1348.72 M208.205 1345.02 Q214.015 1345.02 217.071 1349.63 Q220.149 1354.21 220.149 1362.96 Q220.149 1371.69 217.071 1376.29 Q214.015 1380.88 208.205 1380.88 Q202.395 1380.88 199.316 1376.29 Q196.261 1371.69 196.261 1362.96 Q196.261 1354.21 199.316 1349.63 Q202.395 1345.02 208.205 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M238.367 1348.72 Q234.756 1348.72 232.927 1352.29 Q231.122 1355.83 231.122 1362.96 Q231.122 1370.07 232.927 1373.63 Q234.756 1377.17 238.367 1377.17 Q242.001 1377.17 243.807 1373.63 Q245.635 1370.07 245.635 1362.96 Q245.635 1355.83 243.807 1352.29 Q242.001 1348.72 238.367 1348.72 M238.367 1345.02 Q244.177 1345.02 247.233 1349.63 Q250.311 1354.21 250.311 1362.96 Q250.311 1371.69 247.233 1376.29 Q244.177 1380.88 238.367 1380.88 Q232.557 1380.88 229.478 1376.29 Q226.422 1371.69 226.422 1362.96 Q226.422 1354.21 229.478 1349.63 Q232.557 1345.02 238.367 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M40.2442 2003.67 L40.2442 1997.24 L87.7642 1997.24 L87.7642 2003.67 L40.2442 2003.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M66.2481 1955.06 L87.7642 1955.06 L87.7642 1960.92 L66.439 1960.92 Q61.3783 1960.92 58.8638 1962.89 Q56.3494 1964.87 56.3494 1968.81 Q56.3494 1973.56 59.3731 1976.29 Q62.3968 1979.03 67.6167 1979.03 L87.7642 1979.03 L87.7642 1984.92 L52.1162 1984.92 L52.1162 1979.03 L57.6544 1979.03 Q54.4397 1976.93 52.8483 1974.1 Q51.2568 1971.23 51.2568 1967.51 Q51.2568 1961.37 55.0763 1958.21 Q58.8638 1955.06 66.2481 1955.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M52.1162 1947.58 L52.1162 1941.38 L82.035 1930.24 L52.1162 1919.1 L52.1162 1912.89 L87.7642 1926.26 L87.7642 1934.22 L52.1162 1947.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M68.4761 1874.31 L71.3406 1874.31 L71.3406 1901.24 Q77.3881 1900.86 80.5709 1897.61 Q83.7219 1894.33 83.7219 1888.51 Q83.7219 1885.14 82.8944 1881.98 Q82.0669 1878.8 80.4118 1875.68 L85.9499 1875.68 Q87.2867 1878.83 87.987 1882.14 Q88.6872 1885.45 88.6872 1888.86 Q88.6872 1897.39 83.7219 1902.39 Q78.7567 1907.35 70.2903 1907.35 Q61.5374 1907.35 56.4131 1902.64 Q51.2568 1897.9 51.2568 1889.88 Q51.2568 1882.69 55.9038 1878.52 Q60.5189 1874.31 68.4761 1874.31 M66.7573 1880.17 Q61.9512 1880.23 59.0866 1882.88 Q56.2221 1885.49 56.2221 1889.81 Q56.2221 1894.72 58.9912 1897.68 Q61.7602 1900.6 66.7892 1901.05 L66.7573 1880.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M53.1665 1841.98 L58.7047 1841.98 Q57.4316 1844.46 56.795 1847.13 Q56.1584 1849.81 56.1584 1852.67 Q56.1584 1857.03 57.4952 1859.23 Q58.832 1861.39 61.5056 1861.39 Q63.5426 1861.39 64.7203 1859.83 Q65.8661 1858.27 66.9165 1853.56 L67.3621 1851.56 Q68.6989 1845.32 71.1497 1842.71 Q73.5686 1840.07 77.9291 1840.07 Q82.8944 1840.07 85.7908 1844.01 Q88.6872 1847.93 88.6872 1854.8 Q88.6872 1857.67 88.1143 1860.79 Q87.5732 1863.87 86.4592 1867.31 L80.4118 1867.31 Q82.0987 1864.07 82.958 1860.91 Q83.7856 1857.76 83.7856 1854.68 Q83.7856 1850.54 82.3851 1848.31 Q80.9529 1846.08 78.3747 1846.08 Q75.9876 1846.08 74.7145 1847.71 Q73.4413 1849.3 72.2637 1854.74 L71.7862 1856.78 Q70.6404 1862.22 68.2851 1864.64 Q65.898 1867.06 61.7602 1867.06 Q56.7313 1867.06 53.9941 1863.49 Q51.2568 1859.93 51.2568 1853.37 Q51.2568 1850.12 51.7343 1847.26 Q52.2117 1844.4 53.1665 1841.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M41.9947 1824.95 L52.1162 1824.95 L52.1162 1812.89 L56.6677 1812.89 L56.6677 1824.95 L76.0194 1824.95 Q80.3799 1824.95 81.6213 1823.77 Q82.8626 1822.56 82.8626 1818.9 L82.8626 1812.89 L87.7642 1812.89 L87.7642 1818.9 Q87.7642 1825.68 85.2497 1828.26 Q82.7034 1830.84 76.0194 1830.84 L56.6677 1830.84 L56.6677 1835.13 L52.1162 1835.13 L52.1162 1830.84 L41.9947 1830.84 L41.9947 1824.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M58.9593 1777.43 Q55.0126 1775.23 53.1347 1772.18 Q51.2568 1769.12 51.2568 1764.98 Q51.2568 1759.41 55.1717 1756.39 Q59.0548 1753.37 66.2481 1753.37 L87.7642 1753.37 L87.7642 1759.25 L66.439 1759.25 Q61.3146 1759.25 58.832 1761.07 Q56.3494 1762.88 56.3494 1766.61 Q56.3494 1771.16 59.3731 1773.8 Q62.3968 1776.44 67.6167 1776.44 L87.7642 1776.44 L87.7642 1782.33 L66.439 1782.33 Q61.2828 1782.33 58.832 1784.14 Q56.3494 1785.96 56.3494 1789.75 Q56.3494 1794.23 59.4049 1796.88 Q62.4286 1799.52 67.6167 1799.52 L87.7642 1799.52 L87.7642 1805.41 L52.1162 1805.41 L52.1162 1799.52 L57.6544 1799.52 Q54.376 1797.51 52.8164 1794.71 Q51.2568 1791.91 51.2568 1788.06 Q51.2568 1784.18 53.2302 1781.47 Q55.2036 1778.73 58.9593 1777.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M68.4761 1711.19 L71.3406 1711.19 L71.3406 1738.12 Q77.3881 1737.74 80.5709 1734.49 Q83.7219 1731.21 83.7219 1725.39 Q83.7219 1722.01 82.8944 1718.86 Q82.0669 1715.68 80.4118 1712.56 L85.9499 1712.56 Q87.2867 1715.71 87.987 1719.02 Q88.6872 1722.33 88.6872 1725.74 Q88.6872 1734.27 83.7219 1739.27 Q78.7567 1744.23 70.2903 1744.23 Q61.5374 1744.23 56.4131 1739.52 Q51.2568 1734.78 51.2568 1726.76 Q51.2568 1719.56 55.9038 1715.39 Q60.5189 1711.19 68.4761 1711.19 M66.7573 1717.05 Q61.9512 1717.11 59.0866 1719.75 Q56.2221 1722.36 56.2221 1726.69 Q56.2221 1731.6 58.9912 1734.56 Q61.7602 1737.48 66.7892 1737.93 L66.7573 1717.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M66.2481 1671.95 L87.7642 1671.95 L87.7642 1677.8 L66.439 1677.8 Q61.3783 1677.8 58.8638 1679.78 Q56.3494 1681.75 56.3494 1685.7 Q56.3494 1690.44 59.3731 1693.18 Q62.3968 1695.92 67.6167 1695.92 L87.7642 1695.92 L87.7642 1701.8 L52.1162 1701.8 L52.1162 1695.92 L57.6544 1695.92 Q54.4397 1693.81 52.8483 1690.98 Q51.2568 1688.12 51.2568 1684.39 Q51.2568 1678.25 55.0763 1675.1 Q58.8638 1671.95 66.2481 1671.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M41.9947 1654.47 L52.1162 1654.47 L52.1162 1642.41 L56.6677 1642.41 L56.6677 1654.47 L76.0194 1654.47 Q80.3799 1654.47 81.6213 1653.3 Q82.8626 1652.09 82.8626 1648.43 L82.8626 1642.41 L87.7642 1642.41 L87.7642 1648.43 Q87.7642 1655.21 85.2497 1657.78 Q82.7034 1660.36 76.0194 1660.36 L56.6677 1660.36 L56.6677 1664.66 L52.1162 1664.66 L52.1162 1660.36 L41.9947 1660.36 L41.9947 1654.47 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M38.3026 1599.92 Q45.6232 1604.19 52.7846 1606.25 Q59.946 1608.32 67.2984 1608.32 Q74.6508 1608.32 81.8759 1606.25 Q89.0691 1604.15 96.3579 1599.92 L96.3579 1605.01 Q88.8782 1609.79 81.6531 1612.17 Q74.428 1614.53 67.2984 1614.53 Q60.2006 1614.53 53.0074 1612.17 Q45.8142 1609.82 38.3026 1605.01 L38.3026 1599.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M97.3445 1572.68 L97.3445 1575.86 L87.7642 1575.89 Q87.7005 1579.23 87.0003 1582.57 Q86.2682 1585.92 84.8359 1589.29 L79.1068 1589.29 Q81.1438 1586.04 82.1942 1582.73 Q83.2127 1579.39 83.2445 1575.86 L68.7307 1575.86 Q67.5849 1582.89 64.8476 1586.11 Q62.1104 1589.29 57.3361 1589.29 Q52.148 1589.29 49.1561 1585.82 Q46.1643 1582.35 45.7187 1575.86 L38.239 1575.86 L38.239 1572.68 L45.6232 1572.68 Q45.7505 1569.72 46.2598 1566.95 Q46.7372 1564.18 47.5965 1561.54 L53.1665 1561.54 Q51.8297 1564.18 51.0977 1566.98 Q50.3656 1569.75 50.2383 1572.68 L63.8291 1572.68 Q64.9431 1565.45 67.8077 1562.04 Q70.6722 1558.64 75.6375 1558.64 Q81.0165 1558.64 84.1357 1562.27 Q87.2231 1565.86 87.7005 1572.68 L97.3445 1572.68 M63.2562 1575.86 L50.2065 1575.86 Q50.6203 1579.55 52.3072 1581.49 Q53.9941 1583.43 56.795 1583.43 Q59.5322 1583.43 61.06 1581.65 Q62.5878 1579.84 63.2562 1575.86 M69.3673 1572.68 L83.149 1572.68 Q82.6079 1568.63 80.8574 1566.6 Q79.1068 1564.53 76.2422 1564.53 Q73.4413 1564.53 71.7862 1566.5 Q70.1312 1568.44 69.3673 1572.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M38.3026 1548.01 L38.3026 1542.92 Q45.8142 1538.14 53.0074 1535.79 Q60.2006 1533.4 67.2984 1533.4 Q74.428 1533.4 81.6531 1535.79 Q88.8782 1538.14 96.3579 1542.92 L96.3579 1548.01 Q89.0691 1543.77 81.8759 1541.71 Q74.6508 1539.61 67.2984 1539.61 Q59.946 1539.61 52.7846 1541.71 Q45.6232 1543.77 38.3026 1548.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1022.52 1228.03 L1022.52 1275.06 L1032.41 1275.06 Q1044.92 1275.06 1050.72 1269.39 Q1056.55 1263.72 1056.55 1251.49 Q1056.55 1239.33 1050.72 1233.7 Q1044.92 1228.03 1032.41 1228.03 L1022.52 1228.03 M1014.34 1221.31 L1031.15 1221.31 Q1048.73 1221.31 1056.95 1228.64 Q1065.18 1235.93 1065.18 1251.49 Q1065.18 1267.12 1056.91 1274.46 Q1048.65 1281.79 1031.15 1281.79 L1014.34 1281.79 L1014.34 1221.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1116.7 1257.24 L1116.7 1260.89 L1082.43 1260.89 Q1082.92 1268.58 1087.05 1272.63 Q1091.22 1276.64 1098.64 1276.64 Q1102.93 1276.64 1106.94 1275.59 Q1110.99 1274.54 1114.96 1272.43 L1114.96 1279.48 Q1110.95 1281.18 1106.74 1282.07 Q1102.53 1282.96 1098.19 1282.96 Q1087.34 1282.96 1080.98 1276.64 Q1074.66 1270.32 1074.66 1259.55 Q1074.66 1248.41 1080.65 1241.89 Q1086.69 1235.32 1096.9 1235.32 Q1106.05 1235.32 1111.36 1241.24 Q1116.7 1247.11 1116.7 1257.24 M1109.25 1255.05 Q1109.17 1248.94 1105.81 1245.29 Q1102.49 1241.64 1096.98 1241.64 Q1090.74 1241.64 1086.97 1245.17 Q1083.24 1248.69 1082.68 1255.09 L1109.25 1255.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1161.59 1238.16 L1161.59 1245.13 Q1158.43 1243.39 1155.23 1242.54 Q1152.07 1241.64 1148.83 1241.64 Q1141.58 1241.64 1137.57 1246.26 Q1133.56 1250.84 1133.56 1259.14 Q1133.56 1267.45 1137.57 1272.07 Q1141.58 1276.64 1148.83 1276.64 Q1152.07 1276.64 1155.23 1275.79 Q1158.43 1274.9 1161.59 1273.16 L1161.59 1280.05 Q1158.47 1281.5 1155.11 1282.23 Q1151.79 1282.96 1148.02 1282.96 Q1137.77 1282.96 1131.73 1276.52 Q1125.7 1270.08 1125.7 1259.14 Q1125.7 1248.04 1131.77 1241.68 Q1137.89 1235.32 1148.5 1235.32 Q1151.95 1235.32 1155.23 1236.05 Q1158.51 1236.74 1161.59 1238.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1174.55 1236.42 L1182.01 1236.42 L1182.01 1281.79 L1174.55 1281.79 L1174.55 1236.42 M1174.55 1218.76 L1182.01 1218.76 L1182.01 1228.19 L1174.55 1228.19 L1174.55 1218.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1226.52 1237.76 L1226.52 1244.8 Q1223.37 1243.18 1219.96 1242.37 Q1216.56 1241.56 1212.91 1241.56 Q1207.36 1241.56 1204.57 1243.26 Q1201.81 1244.97 1201.81 1248.37 Q1201.81 1250.96 1203.8 1252.46 Q1205.78 1253.92 1211.78 1255.25 L1214.33 1255.82 Q1222.27 1257.52 1225.59 1260.64 Q1228.96 1263.72 1228.96 1269.27 Q1228.96 1275.59 1223.93 1279.28 Q1218.95 1282.96 1210.2 1282.96 Q1206.55 1282.96 1202.58 1282.23 Q1198.65 1281.55 1194.28 1280.13 L1194.28 1272.43 Q1198.41 1274.58 1202.42 1275.67 Q1206.43 1276.72 1210.36 1276.72 Q1215.63 1276.72 1218.46 1274.94 Q1221.3 1273.12 1221.3 1269.84 Q1221.3 1266.8 1219.23 1265.18 Q1217.21 1263.56 1210.28 1262.06 L1207.69 1261.45 Q1200.76 1259.99 1197.68 1257 Q1194.6 1253.96 1194.6 1248.69 Q1194.6 1242.29 1199.14 1238.81 Q1203.68 1235.32 1212.02 1235.32 Q1216.15 1235.32 1219.8 1235.93 Q1223.45 1236.54 1226.52 1237.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1240.82 1236.42 L1248.28 1236.42 L1248.28 1281.79 L1240.82 1281.79 L1240.82 1236.42 M1240.82 1218.76 L1248.28 1218.76 L1248.28 1228.19 L1240.82 1228.19 L1240.82 1218.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1281.46 1241.64 Q1275.46 1241.64 1271.98 1246.34 Q1268.49 1251 1268.49 1259.14 Q1268.49 1267.29 1271.94 1271.99 Q1275.42 1276.64 1281.46 1276.64 Q1287.41 1276.64 1290.89 1271.94 Q1294.38 1267.25 1294.38 1259.14 Q1294.38 1251.08 1290.89 1246.38 Q1287.41 1241.64 1281.46 1241.64 M1281.46 1235.32 Q1291.18 1235.32 1296.73 1241.64 Q1302.28 1247.96 1302.28 1259.14 Q1302.28 1270.28 1296.73 1276.64 Q1291.18 1282.96 1281.46 1282.96 Q1271.69 1282.96 1266.14 1276.64 Q1260.63 1270.28 1260.63 1259.14 Q1260.63 1247.96 1266.14 1241.64 Q1271.69 1235.32 1281.46 1235.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1352.35 1254.4 L1352.35 1281.79 L1344.89 1281.79 L1344.89 1254.65 Q1344.89 1248.21 1342.38 1245.01 Q1339.87 1241.81 1334.85 1241.81 Q1328.81 1241.81 1325.33 1245.65 Q1321.84 1249.5 1321.84 1256.15 L1321.84 1281.79 L1314.35 1281.79 L1314.35 1236.42 L1321.84 1236.42 L1321.84 1243.47 Q1324.52 1239.38 1328.12 1237.35 Q1331.77 1235.32 1336.51 1235.32 Q1344.33 1235.32 1348.34 1240.19 Q1352.35 1245.01 1352.35 1254.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1422.51 1237.76 L1422.51 1244.8 Q1419.35 1243.18 1415.95 1242.37 Q1412.54 1241.56 1408.9 1241.56 Q1403.35 1241.56 1400.55 1243.26 Q1397.8 1244.97 1397.8 1248.37 Q1397.8 1250.96 1399.78 1252.46 Q1401.77 1253.92 1407.76 1255.25 L1410.31 1255.82 Q1418.25 1257.52 1421.58 1260.64 Q1424.94 1263.72 1424.94 1269.27 Q1424.94 1275.59 1419.91 1279.28 Q1414.93 1282.96 1406.18 1282.96 Q1402.54 1282.96 1398.57 1282.23 Q1394.64 1281.55 1390.26 1280.13 L1390.26 1272.43 Q1394.39 1274.58 1398.4 1275.67 Q1402.42 1276.72 1406.34 1276.72 Q1411.61 1276.72 1414.45 1274.94 Q1417.28 1273.12 1417.28 1269.84 Q1417.28 1266.8 1415.22 1265.18 Q1413.19 1263.56 1406.26 1262.06 L1403.67 1261.45 Q1396.74 1259.99 1393.67 1257 Q1390.59 1253.96 1390.59 1248.69 Q1390.59 1242.29 1395.12 1238.81 Q1399.66 1235.32 1408.01 1235.32 Q1412.14 1235.32 1415.78 1235.93 Q1419.43 1236.54 1422.51 1237.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1444.02 1274.98 L1444.02 1299.05 L1436.52 1299.05 L1436.52 1236.42 L1444.02 1236.42 L1444.02 1243.3 Q1446.37 1239.25 1449.93 1237.31 Q1453.54 1235.32 1458.52 1235.32 Q1466.78 1235.32 1471.93 1241.89 Q1477.11 1248.45 1477.11 1259.14 Q1477.11 1269.84 1471.93 1276.4 Q1466.78 1282.96 1458.52 1282.96 Q1453.54 1282.96 1449.93 1281.02 Q1446.37 1279.03 1444.02 1274.98 M1469.38 1259.14 Q1469.38 1250.92 1465.97 1246.26 Q1462.61 1241.56 1456.7 1241.56 Q1450.78 1241.56 1447.38 1246.26 Q1444.02 1250.92 1444.02 1259.14 Q1444.02 1267.37 1447.38 1272.07 Q1450.78 1276.72 1456.7 1276.72 Q1462.61 1276.72 1465.97 1272.07 Q1469.38 1267.37 1469.38 1259.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1510.09 1258.98 Q1501.05 1258.98 1497.57 1261.05 Q1494.09 1263.11 1494.09 1268.1 Q1494.09 1272.07 1496.68 1274.42 Q1499.31 1276.72 1503.81 1276.72 Q1510.01 1276.72 1513.73 1272.35 Q1517.5 1267.93 1517.5 1260.64 L1517.5 1258.98 L1510.09 1258.98 M1524.95 1255.9 L1524.95 1281.79 L1517.5 1281.79 L1517.5 1274.9 Q1514.95 1279.03 1511.14 1281.02 Q1507.33 1282.96 1501.82 1282.96 Q1494.86 1282.96 1490.72 1279.07 Q1486.63 1275.14 1486.63 1268.58 Q1486.63 1260.93 1491.74 1257.04 Q1496.88 1253.15 1507.05 1253.15 L1517.5 1253.15 L1517.5 1252.42 Q1517.5 1247.27 1514.1 1244.48 Q1510.74 1241.64 1504.62 1241.64 Q1500.73 1241.64 1497.04 1242.58 Q1493.36 1243.51 1489.96 1245.37 L1489.96 1238.48 Q1494.05 1236.9 1497.89 1236.13 Q1501.74 1235.32 1505.39 1235.32 Q1515.23 1235.32 1520.09 1240.43 Q1524.95 1245.53 1524.95 1255.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1572.96 1238.16 L1572.96 1245.13 Q1569.8 1243.39 1566.6 1242.54 Q1563.44 1241.64 1560.2 1241.64 Q1552.95 1241.64 1548.94 1246.26 Q1544.93 1250.84 1544.93 1259.14 Q1544.93 1267.45 1548.94 1272.07 Q1552.95 1276.64 1560.2 1276.64 Q1563.44 1276.64 1566.6 1275.79 Q1569.8 1274.9 1572.96 1273.16 L1572.96 1280.05 Q1569.84 1281.5 1566.48 1282.23 Q1563.15 1282.96 1559.39 1282.96 Q1549.14 1282.96 1543.1 1276.52 Q1537.07 1270.08 1537.07 1259.14 Q1537.07 1248.04 1543.14 1241.68 Q1549.26 1235.32 1559.87 1235.32 Q1563.32 1235.32 1566.6 1236.05 Q1569.88 1236.74 1572.96 1238.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1624.73 1257.24 L1624.73 1260.89 L1590.46 1260.89 Q1590.94 1268.58 1595.08 1272.63 Q1599.25 1276.64 1606.66 1276.64 Q1610.96 1276.64 1614.97 1275.59 Q1619.02 1274.54 1622.99 1272.43 L1622.99 1279.48 Q1618.98 1281.18 1614.76 1282.07 Q1610.55 1282.96 1606.22 1282.96 Q1595.36 1282.96 1589 1276.64 Q1582.68 1270.32 1582.68 1259.55 Q1582.68 1248.41 1588.68 1241.89 Q1594.71 1235.32 1604.92 1235.32 Q1614.07 1235.32 1619.38 1241.24 Q1624.73 1247.11 1624.73 1257.24 M1617.28 1255.05 Q1617.19 1248.94 1613.83 1245.29 Q1610.51 1241.64 1605 1241.64 Q1598.76 1241.64 1595 1245.17 Q1591.27 1248.69 1590.7 1255.09 L1617.28 1255.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip093)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
-<path clip-path="url(#clip093)" d="M436.9 1958.75 L436.9 2078.86 L466.444 2078.86 L466.444 1958.75 L436.9 1958.75 L436.9 1958.75  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,1958.75 436.9,2078.86 466.444,2078.86 466.444,1958.75 436.9,1958.75 "/>
-<path clip-path="url(#clip093)" d="M473.83 1869.53 L473.83 2039.4 L503.374 2039.4 L503.374 1869.53 L473.83 1869.53 L473.83 1869.53  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,1869.53 473.83,2039.4 503.374,2039.4 503.374,1869.53 473.83,1869.53 "/>
-<path clip-path="url(#clip093)" d="M510.76 1801.07 L510.76 2009.11 L540.304 2009.11 L540.304 1801.07 L510.76 1801.07 L510.76 1801.07  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,1801.07 510.76,2009.11 540.304,2009.11 540.304,1801.07 510.76,1801.07 "/>
-<path clip-path="url(#clip093)" d="M547.691 1743.36 L547.691 1983.58 L577.235 1983.58 L577.235 1743.36 L547.691 1743.36 L547.691 1743.36  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1743.36 547.691,1983.58 577.235,1983.58 577.235,1743.36 547.691,1743.36 "/>
-<path clip-path="url(#clip093)" d="M584.621 1692.51 L584.621 1961.09 L614.165 1961.09 L614.165 1692.51 L584.621 1692.51 L584.621 1692.51  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1692.51 584.621,1961.09 614.165,1961.09 614.165,1692.51 584.621,1692.51 "/>
-<path clip-path="url(#clip093)" d="M621.551 1646.54 L621.551 1940.76 L651.095 1940.76 L651.095 1646.54 L621.551 1646.54 L621.551 1646.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1646.54 621.551,1940.76 651.095,1940.76 651.095,1646.54 621.551,1646.54 "/>
-<path clip-path="url(#clip093)" d="M658.481 1604.27 L658.481 1922.06 L688.026 1922.06 L688.026 1604.27 L658.481 1604.27 L658.481 1604.27  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1604.27 658.481,1922.06 688.026,1922.06 688.026,1604.27 658.481,1604.27 "/>
-<path clip-path="url(#clip093)" d="M695.412 1564.93 L695.412 1904.65 L724.956 1904.65 L724.956 1564.93 L695.412 1564.93 L695.412 1564.93  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1564.93 695.412,1904.65 724.956,1904.65 724.956,1564.93 695.412,1564.93 "/>
-<path clip-path="url(#clip093)" d="M732.342 1527.97 L732.342 1888.3 L761.886 1888.3 L761.886 1527.97 L732.342 1527.97 L732.342 1527.97  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1527.97 732.342,1888.3 761.886,1888.3 761.886,1527.97 732.342,1527.97 "/>
-<path clip-path="url(#clip093)" d="M769.272 1493.02 L769.272 1872.84 L798.816 1872.84 L798.816 1493.02 L769.272 1493.02 L769.272 1493.02  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1493.02 769.272,1872.84 798.816,1872.84 798.816,1493.02 769.272,1493.02 "/>
-<path clip-path="url(#clip093)" d="M806.203 1459.77 L806.203 1858.14 L835.747 1858.14 L835.747 1459.77 L806.203 1459.77 L806.203 1459.77  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1459.77 806.203,1858.14 835.747,1858.14 835.747,1459.77 806.203,1459.77 "/>
-<path clip-path="url(#clip093)" d="M843.133 1428.01 L843.133 1844.08 L872.677 1844.08 L872.677 1428.01 L843.133 1428.01 L843.133 1428.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1428.01 843.133,1844.08 872.677,1844.08 872.677,1428.01 843.133,1428.01 "/>
-<path clip-path="url(#clip093)" d="M880.063 1397.54 L880.063 1830.61 L909.607 1830.61 L909.607 1397.54 L880.063 1397.54 L880.063 1397.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1397.54 880.063,1830.61 909.607,1830.61 909.607,1397.54 880.063,1397.54 "/>
-<path clip-path="url(#clip093)" d="M916.993 1368.23 L916.993 1817.64 L946.538 1817.64 L946.538 1368.23 L916.993 1368.23 L916.993 1368.23  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1368.23 916.993,1817.64 946.538,1817.64 946.538,1368.23 916.993,1368.23 "/>
-<path clip-path="url(#clip093)" d="M953.924 1362.92 L953.924 1793.51 L983.468 1793.51 L983.468 1362.92 L953.924 1362.92 L953.924 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1362.92 953.924,1793.51 983.468,1793.51 983.468,1362.92 953.924,1362.92 "/>
-<path clip-path="url(#clip093)" d="M990.854 1362.92 L990.854 1769.49 L1020.4 1769.49 L1020.4 1362.92 L990.854 1362.92 L990.854 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1362.92 990.854,1769.49 1020.4,1769.49 1020.4,1362.92 990.854,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1027.78 1362.92 L1027.78 1747.67 L1057.33 1747.67 L1057.33 1362.92 L1027.78 1362.92 L1027.78 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1362.92 1027.78,1747.67 1057.33,1747.67 1057.33,1362.92 1027.78,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1064.71 1362.92 L1064.71 1727.55 L1094.26 1727.55 L1094.26 1362.92 L1064.71 1362.92 L1064.71 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1362.92 1064.71,1727.55 1094.26,1727.55 1094.26,1362.92 1064.71,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1101.64 1362.92 L1101.64 1708.77 L1131.19 1708.77 L1131.19 1362.92 L1101.64 1362.92 L1101.64 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1362.92 1101.64,1708.77 1131.19,1708.77 1131.19,1362.92 1101.64,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1138.58 1362.92 L1138.58 1691.1 L1168.12 1691.1 L1168.12 1362.92 L1138.58 1362.92 L1138.58 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1362.92 1138.58,1691.1 1168.12,1691.1 1168.12,1362.92 1138.58,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1175.51 1362.92 L1175.51 1674.37 L1205.05 1674.37 L1205.05 1362.92 L1175.51 1362.92 L1175.51 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1362.92 1175.51,1674.37 1205.05,1674.37 1205.05,1362.92 1175.51,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1212.44 1362.92 L1212.44 1658.44 L1241.98 1658.44 L1241.98 1362.92 L1212.44 1362.92 L1212.44 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1362.92 1212.44,1658.44 1241.98,1658.44 1241.98,1362.92 1212.44,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1249.37 1362.92 L1249.37 1643.2 L1278.91 1643.2 L1278.91 1362.92 L1249.37 1362.92 L1249.37 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1362.92 1249.37,1643.2 1278.91,1643.2 1278.91,1362.92 1249.37,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1286.3 1362.92 L1286.3 1628.58 L1315.84 1628.58 L1315.84 1362.92 L1286.3 1362.92 L1286.3 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1362.92 1286.3,1628.58 1315.84,1628.58 1315.84,1362.92 1286.3,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1323.23 1362.92 L1323.23 1614.49 L1352.77 1614.49 L1352.77 1362.92 L1323.23 1362.92 L1323.23 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1362.92 1323.23,1614.49 1352.77,1614.49 1352.77,1362.92 1323.23,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1360.16 1362.92 L1360.16 1600.9 L1389.7 1600.9 L1389.7 1362.92 L1360.16 1362.92 L1360.16 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1362.92 1360.16,1600.9 1389.7,1600.9 1389.7,1362.92 1360.16,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1397.09 1362.92 L1397.09 1587.74 L1426.63 1587.74 L1426.63 1362.92 L1397.09 1362.92 L1397.09 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1362.92 1397.09,1587.74 1426.63,1587.74 1426.63,1362.92 1397.09,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1434.02 1362.92 L1434.02 1574.98 L1463.56 1574.98 L1463.56 1362.92 L1434.02 1362.92 L1434.02 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1362.92 1434.02,1574.98 1463.56,1574.98 1463.56,1362.92 1434.02,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1470.95 1362.92 L1470.95 1562.59 L1500.49 1562.59 L1500.49 1362.92 L1470.95 1362.92 L1470.95 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1362.92 1470.95,1562.59 1500.49,1562.59 1500.49,1362.92 1470.95,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1507.88 1362.92 L1507.88 1550.53 L1537.42 1550.53 L1537.42 1362.92 L1507.88 1362.92 L1507.88 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1362.92 1507.88,1550.53 1537.42,1550.53 1537.42,1362.92 1507.88,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1544.81 1362.92 L1544.81 1538.79 L1574.35 1538.79 L1574.35 1362.92 L1544.81 1362.92 L1544.81 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1362.92 1544.81,1538.79 1574.35,1538.79 1574.35,1362.92 1544.81,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1581.74 1362.92 L1581.74 1527.33 L1611.28 1527.33 L1611.28 1362.92 L1581.74 1362.92 L1581.74 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1362.92 1581.74,1527.33 1611.28,1527.33 1611.28,1362.92 1581.74,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1618.67 1362.92 L1618.67 1516.13 L1648.21 1516.13 L1648.21 1362.92 L1618.67 1362.92 L1618.67 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1362.92 1618.67,1516.13 1648.21,1516.13 1648.21,1362.92 1618.67,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1655.6 1362.92 L1655.6 1505.19 L1685.14 1505.19 L1685.14 1362.92 L1655.6 1362.92 L1655.6 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1362.92 1655.6,1505.19 1685.14,1505.19 1685.14,1362.92 1655.6,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1692.53 1362.92 L1692.53 1494.48 L1722.07 1494.48 L1722.07 1362.92 L1692.53 1362.92 L1692.53 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1362.92 1692.53,1494.48 1722.07,1494.48 1722.07,1362.92 1692.53,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1729.46 1362.92 L1729.46 1483.98 L1759 1483.98 L1759 1362.92 L1729.46 1362.92 L1729.46 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1362.92 1729.46,1483.98 1759,1483.98 1759,1362.92 1729.46,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1766.39 1362.92 L1766.39 1473.7 L1795.93 1473.7 L1795.93 1362.92 L1766.39 1362.92 L1766.39 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1362.92 1766.39,1473.7 1795.93,1473.7 1795.93,1362.92 1766.39,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1803.32 1362.92 L1803.32 1463.6 L1832.86 1463.6 L1832.86 1362.92 L1803.32 1362.92 L1803.32 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1362.92 1803.32,1463.6 1832.86,1463.6 1832.86,1362.92 1803.32,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1840.25 1362.92 L1840.25 1453.7 L1869.79 1453.7 L1869.79 1362.92 L1840.25 1362.92 L1840.25 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1362.92 1840.25,1453.7 1869.79,1453.7 1869.79,1362.92 1840.25,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1877.18 1362.92 L1877.18 1443.96 L1906.73 1443.96 L1906.73 1362.92 L1877.18 1362.92 L1877.18 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1362.92 1877.18,1443.96 1906.73,1443.96 1906.73,1362.92 1877.18,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1914.11 1362.92 L1914.11 1434.39 L1943.66 1434.39 L1943.66 1362.92 L1914.11 1362.92 L1914.11 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1362.92 1914.11,1434.39 1943.66,1434.39 1943.66,1362.92 1914.11,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1951.04 1362.92 L1951.04 1424.98 L1980.59 1424.98 L1980.59 1362.92 L1951.04 1362.92 L1951.04 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1362.92 1951.04,1424.98 1980.59,1424.98 1980.59,1362.92 1951.04,1362.92 "/>
-<path clip-path="url(#clip093)" d="M1987.97 1362.92 L1987.97 1415.72 L2017.52 1415.72 L2017.52 1362.92 L1987.97 1362.92 L1987.97 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1362.92 1987.97,1415.72 2017.52,1415.72 2017.52,1362.92 1987.97,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2024.9 1362.92 L2024.9 1406.6 L2054.45 1406.6 L2054.45 1362.92 L2024.9 1362.92 L2024.9 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1362.92 2024.9,1406.6 2054.45,1406.6 2054.45,1362.92 2024.9,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2061.83 1362.92 L2061.83 1397.61 L2091.38 1397.61 L2091.38 1362.92 L2061.83 1362.92 L2061.83 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1362.92 2061.83,1397.61 2091.38,1397.61 2091.38,1362.92 2061.83,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2098.76 1362.92 L2098.76 1388.76 L2128.31 1388.76 L2128.31 1362.92 L2098.76 1362.92 L2098.76 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1362.92 2098.76,1388.76 2128.31,1388.76 2128.31,1362.92 2098.76,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2135.69 1362.92 L2135.69 1380.03 L2165.24 1380.03 L2165.24 1362.92 L2135.69 1362.92 L2135.69 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1362.92 2135.69,1380.03 2165.24,1380.03 2165.24,1362.92 2135.69,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2172.62 1362.92 L2172.62 1371.42 L2202.17 1371.42 L2202.17 1362.92 L2172.62 1362.92 L2172.62 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1362.92 2172.62,1371.42 2202.17,1371.42 2202.17,1362.92 2172.62,1362.92 "/>
-<path clip-path="url(#clip093)" d="M2209.55 1362.92 L2209.55 1362.92 L2239.1 1362.92 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,1362.92 2239.1,1362.92 2239.1,1362.92 2209.55,1362.92 "/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="451.672" cy="1958.75" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="488.602" cy="1869.53" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="525.532" cy="1801.07" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="562.463" cy="1743.36" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="599.393" cy="1692.51" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="636.323" cy="1646.54" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="673.253" cy="1604.27" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="710.184" cy="1564.93" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="747.114" cy="1527.97" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="784.044" cy="1493.02" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="820.975" cy="1459.77" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="857.905" cy="1428.01" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="894.835" cy="1397.54" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="931.766" cy="1368.23" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="968.696" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1005.63" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1042.56" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1079.49" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1116.42" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1153.35" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1190.28" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1227.21" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1264.14" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1301.07" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1338" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1374.93" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1411.86" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1448.79" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1485.72" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1522.65" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1559.58" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1596.51" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1633.44" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1670.37" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1707.3" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1744.23" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1781.16" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1818.09" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1855.02" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1891.95" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1928.88" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1965.81" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2002.74" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2039.67" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2076.6" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2113.53" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2150.47" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2187.4" cy="1362.92" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
-<path clip-path="url(#clip093)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
-<path clip-path="url(#clip093)" d="M436.9 2078.86 L436.9 2078.86 L466.444 2078.86 L466.444 2078.86 L436.9 2078.86 L436.9 2078.86  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,2078.86 436.9,2078.86 466.444,2078.86 466.444,2078.86 436.9,2078.86 "/>
-<path clip-path="url(#clip093)" d="M473.83 2039.4 L473.83 2039.4 L503.374 2039.4 L503.374 2039.4 L473.83 2039.4 L473.83 2039.4  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,2039.4 473.83,2039.4 503.374,2039.4 503.374,2039.4 473.83,2039.4 "/>
-<path clip-path="url(#clip093)" d="M510.76 2009.11 L510.76 2009.11 L540.304 2009.11 L540.304 2009.11 L510.76 2009.11 L510.76 2009.11  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,2009.11 510.76,2009.11 540.304,2009.11 540.304,2009.11 510.76,2009.11 "/>
-<path clip-path="url(#clip093)" d="M547.691 1983.58 L547.691 1983.58 L577.235 1983.58 L577.235 1983.58 L547.691 1983.58 L547.691 1983.58  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1983.58 547.691,1983.58 577.235,1983.58 577.235,1983.58 547.691,1983.58 "/>
-<path clip-path="url(#clip093)" d="M584.621 1961.09 L584.621 1961.09 L614.165 1961.09 L614.165 1961.09 L584.621 1961.09 L584.621 1961.09  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1961.09 584.621,1961.09 614.165,1961.09 614.165,1961.09 584.621,1961.09 "/>
-<path clip-path="url(#clip093)" d="M621.551 1940.76 L621.551 1940.76 L651.095 1940.76 L651.095 1940.76 L621.551 1940.76 L621.551 1940.76  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1940.76 621.551,1940.76 651.095,1940.76 651.095,1940.76 621.551,1940.76 "/>
-<path clip-path="url(#clip093)" d="M658.481 1922.06 L658.481 1922.06 L688.026 1922.06 L688.026 1922.06 L658.481 1922.06 L658.481 1922.06  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1922.06 658.481,1922.06 688.026,1922.06 688.026,1922.06 658.481,1922.06 "/>
-<path clip-path="url(#clip093)" d="M695.412 1904.65 L695.412 1904.65 L724.956 1904.65 L724.956 1904.65 L695.412 1904.65 L695.412 1904.65  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1904.65 695.412,1904.65 724.956,1904.65 724.956,1904.65 695.412,1904.65 "/>
-<path clip-path="url(#clip093)" d="M732.342 1888.3 L732.342 1888.3 L761.886 1888.3 L761.886 1888.3 L732.342 1888.3 L732.342 1888.3  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1888.3 732.342,1888.3 761.886,1888.3 761.886,1888.3 732.342,1888.3 "/>
-<path clip-path="url(#clip093)" d="M769.272 1872.84 L769.272 1872.84 L798.816 1872.84 L798.816 1872.84 L769.272 1872.84 L769.272 1872.84  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1872.84 769.272,1872.84 798.816,1872.84 798.816,1872.84 769.272,1872.84 "/>
-<path clip-path="url(#clip093)" d="M806.203 1858.14 L806.203 1858.14 L835.747 1858.14 L835.747 1858.14 L806.203 1858.14 L806.203 1858.14  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1858.14 806.203,1858.14 835.747,1858.14 835.747,1858.14 806.203,1858.14 "/>
-<path clip-path="url(#clip093)" d="M843.133 1844.08 L843.133 1844.08 L872.677 1844.08 L872.677 1844.08 L843.133 1844.08 L843.133 1844.08  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1844.08 843.133,1844.08 872.677,1844.08 872.677,1844.08 843.133,1844.08 "/>
-<path clip-path="url(#clip093)" d="M880.063 1830.61 L880.063 1830.61 L909.607 1830.61 L909.607 1830.61 L880.063 1830.61 L880.063 1830.61  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1830.61 880.063,1830.61 909.607,1830.61 909.607,1830.61 880.063,1830.61 "/>
-<path clip-path="url(#clip093)" d="M916.993 1817.64 L916.993 1817.64 L946.538 1817.64 L946.538 1817.64 L916.993 1817.64 L916.993 1817.64  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1817.64 916.993,1817.64 946.538,1817.64 946.538,1817.64 916.993,1817.64 "/>
-<path clip-path="url(#clip093)" d="M953.924 1793.51 L953.924 1793.51 L983.468 1793.51 L983.468 1793.51 L953.924 1793.51 L953.924 1793.51  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1793.51 953.924,1793.51 983.468,1793.51 983.468,1793.51 953.924,1793.51 "/>
-<path clip-path="url(#clip093)" d="M990.854 1769.49 L990.854 1769.49 L1020.4 1769.49 L1020.4 1769.49 L990.854 1769.49 L990.854 1769.49  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1769.49 990.854,1769.49 1020.4,1769.49 1020.4,1769.49 990.854,1769.49 "/>
-<path clip-path="url(#clip093)" d="M1027.78 1747.67 L1027.78 1747.67 L1057.33 1747.67 L1057.33 1747.67 L1027.78 1747.67 L1027.78 1747.67  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1747.67 1027.78,1747.67 1057.33,1747.67 1057.33,1747.67 1027.78,1747.67 "/>
-<path clip-path="url(#clip093)" d="M1064.71 1727.55 L1064.71 1727.55 L1094.26 1727.55 L1094.26 1727.55 L1064.71 1727.55 L1064.71 1727.55  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1727.55 1064.71,1727.55 1094.26,1727.55 1094.26,1727.55 1064.71,1727.55 "/>
-<path clip-path="url(#clip093)" d="M1101.64 1708.77 L1101.64 1708.77 L1131.19 1708.77 L1131.19 1708.77 L1101.64 1708.77 L1101.64 1708.77  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1708.77 1101.64,1708.77 1131.19,1708.77 1131.19,1708.77 1101.64,1708.77 "/>
-<path clip-path="url(#clip093)" d="M1138.58 1691.1 L1138.58 1691.1 L1168.12 1691.1 L1168.12 1691.1 L1138.58 1691.1 L1138.58 1691.1  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1691.1 1138.58,1691.1 1168.12,1691.1 1168.12,1691.1 1138.58,1691.1 "/>
-<path clip-path="url(#clip093)" d="M1175.51 1674.37 L1175.51 1674.37 L1205.05 1674.37 L1205.05 1674.37 L1175.51 1674.37 L1175.51 1674.37  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1674.37 1175.51,1674.37 1205.05,1674.37 1205.05,1674.37 1175.51,1674.37 "/>
-<path clip-path="url(#clip093)" d="M1212.44 1658.44 L1212.44 1658.44 L1241.98 1658.44 L1241.98 1658.44 L1212.44 1658.44 L1212.44 1658.44  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1658.44 1212.44,1658.44 1241.98,1658.44 1241.98,1658.44 1212.44,1658.44 "/>
-<path clip-path="url(#clip093)" d="M1249.37 1643.2 L1249.37 1643.2 L1278.91 1643.2 L1278.91 1643.2 L1249.37 1643.2 L1249.37 1643.2  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1643.2 1249.37,1643.2 1278.91,1643.2 1278.91,1643.2 1249.37,1643.2 "/>
-<path clip-path="url(#clip093)" d="M1286.3 1628.58 L1286.3 1628.58 L1315.84 1628.58 L1315.84 1628.58 L1286.3 1628.58 L1286.3 1628.58  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1628.58 1286.3,1628.58 1315.84,1628.58 1315.84,1628.58 1286.3,1628.58 "/>
-<path clip-path="url(#clip093)" d="M1323.23 1614.49 L1323.23 1614.49 L1352.77 1614.49 L1352.77 1614.49 L1323.23 1614.49 L1323.23 1614.49  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1614.49 1323.23,1614.49 1352.77,1614.49 1352.77,1614.49 1323.23,1614.49 "/>
-<path clip-path="url(#clip093)" d="M1360.16 1600.9 L1360.16 1600.9 L1389.7 1600.9 L1389.7 1600.9 L1360.16 1600.9 L1360.16 1600.9  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1600.9 1360.16,1600.9 1389.7,1600.9 1389.7,1600.9 1360.16,1600.9 "/>
-<path clip-path="url(#clip093)" d="M1397.09 1587.74 L1397.09 1587.74 L1426.63 1587.74 L1426.63 1587.74 L1397.09 1587.74 L1397.09 1587.74  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1587.74 1397.09,1587.74 1426.63,1587.74 1426.63,1587.74 1397.09,1587.74 "/>
-<path clip-path="url(#clip093)" d="M1434.02 1574.98 L1434.02 1574.98 L1463.56 1574.98 L1463.56 1574.98 L1434.02 1574.98 L1434.02 1574.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1574.98 1434.02,1574.98 1463.56,1574.98 1463.56,1574.98 1434.02,1574.98 "/>
-<path clip-path="url(#clip093)" d="M1470.95 1562.59 L1470.95 1562.59 L1500.49 1562.59 L1500.49 1562.59 L1470.95 1562.59 L1470.95 1562.59  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1562.59 1470.95,1562.59 1500.49,1562.59 1500.49,1562.59 1470.95,1562.59 "/>
-<path clip-path="url(#clip093)" d="M1507.88 1550.53 L1507.88 1550.53 L1537.42 1550.53 L1537.42 1550.53 L1507.88 1550.53 L1507.88 1550.53  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1550.53 1507.88,1550.53 1537.42,1550.53 1537.42,1550.53 1507.88,1550.53 "/>
-<path clip-path="url(#clip093)" d="M1544.81 1538.79 L1544.81 1538.79 L1574.35 1538.79 L1574.35 1538.79 L1544.81 1538.79 L1544.81 1538.79  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1538.79 1544.81,1538.79 1574.35,1538.79 1574.35,1538.79 1544.81,1538.79 "/>
-<path clip-path="url(#clip093)" d="M1581.74 1527.33 L1581.74 1527.33 L1611.28 1527.33 L1611.28 1527.33 L1581.74 1527.33 L1581.74 1527.33  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1527.33 1581.74,1527.33 1611.28,1527.33 1611.28,1527.33 1581.74,1527.33 "/>
-<path clip-path="url(#clip093)" d="M1618.67 1516.13 L1618.67 1516.13 L1648.21 1516.13 L1648.21 1516.13 L1618.67 1516.13 L1618.67 1516.13  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1516.13 1618.67,1516.13 1648.21,1516.13 1648.21,1516.13 1618.67,1516.13 "/>
-<path clip-path="url(#clip093)" d="M1655.6 1505.19 L1655.6 1505.19 L1685.14 1505.19 L1685.14 1505.19 L1655.6 1505.19 L1655.6 1505.19  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1505.19 1655.6,1505.19 1685.14,1505.19 1685.14,1505.19 1655.6,1505.19 "/>
-<path clip-path="url(#clip093)" d="M1692.53 1494.48 L1692.53 1494.48 L1722.07 1494.48 L1722.07 1494.48 L1692.53 1494.48 L1692.53 1494.48  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1494.48 1692.53,1494.48 1722.07,1494.48 1722.07,1494.48 1692.53,1494.48 "/>
-<path clip-path="url(#clip093)" d="M1729.46 1483.98 L1729.46 1483.98 L1759 1483.98 L1759 1483.98 L1729.46 1483.98 L1729.46 1483.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1483.98 1729.46,1483.98 1759,1483.98 1759,1483.98 1729.46,1483.98 "/>
-<path clip-path="url(#clip093)" d="M1766.39 1473.7 L1766.39 1473.7 L1795.93 1473.7 L1795.93 1473.7 L1766.39 1473.7 L1766.39 1473.7  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1473.7 1766.39,1473.7 1795.93,1473.7 1795.93,1473.7 1766.39,1473.7 "/>
-<path clip-path="url(#clip093)" d="M1803.32 1463.6 L1803.32 1463.6 L1832.86 1463.6 L1832.86 1463.6 L1803.32 1463.6 L1803.32 1463.6  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1463.6 1803.32,1463.6 1832.86,1463.6 1832.86,1463.6 1803.32,1463.6 "/>
-<path clip-path="url(#clip093)" d="M1840.25 1453.7 L1840.25 1453.7 L1869.79 1453.7 L1869.79 1453.7 L1840.25 1453.7 L1840.25 1453.7  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1453.7 1840.25,1453.7 1869.79,1453.7 1869.79,1453.7 1840.25,1453.7 "/>
-<path clip-path="url(#clip093)" d="M1877.18 1443.96 L1877.18 1443.96 L1906.73 1443.96 L1906.73 1443.96 L1877.18 1443.96 L1877.18 1443.96  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1443.96 1877.18,1443.96 1906.73,1443.96 1906.73,1443.96 1877.18,1443.96 "/>
-<path clip-path="url(#clip093)" d="M1914.11 1434.39 L1914.11 1434.39 L1943.66 1434.39 L1943.66 1434.39 L1914.11 1434.39 L1914.11 1434.39  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1434.39 1914.11,1434.39 1943.66,1434.39 1943.66,1434.39 1914.11,1434.39 "/>
-<path clip-path="url(#clip093)" d="M1951.04 1424.98 L1951.04 1424.98 L1980.59 1424.98 L1980.59 1424.98 L1951.04 1424.98 L1951.04 1424.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1424.98 1951.04,1424.98 1980.59,1424.98 1980.59,1424.98 1951.04,1424.98 "/>
-<path clip-path="url(#clip093)" d="M1987.97 1415.72 L1987.97 1415.72 L2017.52 1415.72 L2017.52 1415.72 L1987.97 1415.72 L1987.97 1415.72  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1415.72 1987.97,1415.72 2017.52,1415.72 2017.52,1415.72 1987.97,1415.72 "/>
-<path clip-path="url(#clip093)" d="M2024.9 1406.6 L2024.9 1406.6 L2054.45 1406.6 L2054.45 1406.6 L2024.9 1406.6 L2024.9 1406.6  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1406.6 2024.9,1406.6 2054.45,1406.6 2054.45,1406.6 2024.9,1406.6 "/>
-<path clip-path="url(#clip093)" d="M2061.83 1397.61 L2061.83 1397.61 L2091.38 1397.61 L2091.38 1397.61 L2061.83 1397.61 L2061.83 1397.61  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1397.61 2061.83,1397.61 2091.38,1397.61 2091.38,1397.61 2061.83,1397.61 "/>
-<path clip-path="url(#clip093)" d="M2098.76 1388.76 L2098.76 1388.76 L2128.31 1388.76 L2128.31 1388.76 L2098.76 1388.76 L2098.76 1388.76  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1388.76 2098.76,1388.76 2128.31,1388.76 2128.31,1388.76 2098.76,1388.76 "/>
-<path clip-path="url(#clip093)" d="M2135.69 1380.03 L2135.69 1380.03 L2165.24 1380.03 L2165.24 1380.03 L2135.69 1380.03 L2135.69 1380.03  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1380.03 2135.69,1380.03 2165.24,1380.03 2165.24,1380.03 2135.69,1380.03 "/>
-<path clip-path="url(#clip093)" d="M2172.62 1371.42 L2172.62 1371.42 L2202.17 1371.42 L2202.17 1371.42 L2172.62 1371.42 L2172.62 1371.42  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1371.42 2172.62,1371.42 2202.17,1371.42 2202.17,1371.42 2172.62,1371.42 "/>
-<path clip-path="url(#clip093)" d="M2209.55 1362.92 L2209.55 1362.92 L2239.1 1362.92 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,1362.92 2239.1,1362.92 2239.1,1362.92 2209.55,1362.92 "/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="451.672" cy="2078.86" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="488.602" cy="2039.4" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="525.532" cy="2009.11" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="562.463" cy="1983.58" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="599.393" cy="1961.09" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="636.323" cy="1940.76" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="673.253" cy="1922.06" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="710.184" cy="1904.65" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="747.114" cy="1888.3" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="784.044" cy="1872.84" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="820.975" cy="1858.14" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="857.905" cy="1844.08" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="894.835" cy="1830.61" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="931.766" cy="1817.64" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="968.696" cy="1793.51" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1005.63" cy="1769.49" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1042.56" cy="1747.67" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1079.49" cy="1727.55" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1116.42" cy="1708.77" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1153.35" cy="1691.1" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1190.28" cy="1674.37" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1227.21" cy="1658.44" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1264.14" cy="1643.2" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1301.07" cy="1628.58" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1338" cy="1614.49" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1374.93" cy="1600.9" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1411.86" cy="1587.74" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1448.79" cy="1574.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1485.72" cy="1562.59" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1522.65" cy="1550.53" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1559.58" cy="1538.79" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1596.51" cy="1527.33" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1633.44" cy="1516.13" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1670.37" cy="1505.19" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1707.3" cy="1494.48" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1744.23" cy="1483.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1781.16" cy="1473.7" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1818.09" cy="1463.6" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1855.02" cy="1453.7" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1891.95" cy="1443.96" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1928.88" cy="1434.39" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1965.81" cy="1424.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2002.74" cy="1415.72" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2039.67" cy="1406.6" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2076.6" cy="1397.61" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2113.53" cy="1388.76" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2150.47" cy="1380.03" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2187.4" cy="1371.42" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
-<path clip-path="url(#clip093)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
-<path clip-path="url(#clip093)" d="M436.9 2078.86 L436.9 2174.14 L466.444 2174.14 L466.444 2078.86 L436.9 2078.86 L436.9 2078.86  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,2078.86 436.9,2174.14 466.444,2174.14 466.444,2078.86 436.9,2078.86 "/>
-<path clip-path="url(#clip093)" d="M473.83 2039.4 L473.83 2174.14 L503.374 2174.14 L503.374 2039.4 L473.83 2039.4 L473.83 2039.4  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,2039.4 473.83,2174.14 503.374,2174.14 503.374,2039.4 473.83,2039.4 "/>
-<path clip-path="url(#clip093)" d="M510.76 2009.11 L510.76 2174.14 L540.304 2174.14 L540.304 2009.11 L510.76 2009.11 L510.76 2009.11  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,2009.11 510.76,2174.14 540.304,2174.14 540.304,2009.11 510.76,2009.11 "/>
-<path clip-path="url(#clip093)" d="M547.691 1983.58 L547.691 2174.14 L577.235 2174.14 L577.235 1983.58 L547.691 1983.58 L547.691 1983.58  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1983.58 547.691,2174.14 577.235,2174.14 577.235,1983.58 547.691,1983.58 "/>
-<path clip-path="url(#clip093)" d="M584.621 1961.09 L584.621 2174.14 L614.165 2174.14 L614.165 1961.09 L584.621 1961.09 L584.621 1961.09  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1961.09 584.621,2174.14 614.165,2174.14 614.165,1961.09 584.621,1961.09 "/>
-<path clip-path="url(#clip093)" d="M621.551 1940.76 L621.551 2174.14 L651.095 2174.14 L651.095 1940.76 L621.551 1940.76 L621.551 1940.76  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1940.76 621.551,2174.14 651.095,2174.14 651.095,1940.76 621.551,1940.76 "/>
-<path clip-path="url(#clip093)" d="M658.481 1922.06 L658.481 2174.14 L688.026 2174.14 L688.026 1922.06 L658.481 1922.06 L658.481 1922.06  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1922.06 658.481,2174.14 688.026,2174.14 688.026,1922.06 658.481,1922.06 "/>
-<path clip-path="url(#clip093)" d="M695.412 1904.65 L695.412 2174.14 L724.956 2174.14 L724.956 1904.65 L695.412 1904.65 L695.412 1904.65  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1904.65 695.412,2174.14 724.956,2174.14 724.956,1904.65 695.412,1904.65 "/>
-<path clip-path="url(#clip093)" d="M732.342 1888.3 L732.342 2174.14 L761.886 2174.14 L761.886 1888.3 L732.342 1888.3 L732.342 1888.3  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1888.3 732.342,2174.14 761.886,2174.14 761.886,1888.3 732.342,1888.3 "/>
-<path clip-path="url(#clip093)" d="M769.272 1872.84 L769.272 2174.14 L798.816 2174.14 L798.816 1872.84 L769.272 1872.84 L769.272 1872.84  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1872.84 769.272,2174.14 798.816,2174.14 798.816,1872.84 769.272,1872.84 "/>
-<path clip-path="url(#clip093)" d="M806.203 1858.14 L806.203 2174.14 L835.747 2174.14 L835.747 1858.14 L806.203 1858.14 L806.203 1858.14  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1858.14 806.203,2174.14 835.747,2174.14 835.747,1858.14 806.203,1858.14 "/>
-<path clip-path="url(#clip093)" d="M843.133 1844.08 L843.133 2174.14 L872.677 2174.14 L872.677 1844.08 L843.133 1844.08 L843.133 1844.08  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1844.08 843.133,2174.14 872.677,2174.14 872.677,1844.08 843.133,1844.08 "/>
-<path clip-path="url(#clip093)" d="M880.063 1830.61 L880.063 2174.14 L909.607 2174.14 L909.607 1830.61 L880.063 1830.61 L880.063 1830.61  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1830.61 880.063,2174.14 909.607,2174.14 909.607,1830.61 880.063,1830.61 "/>
-<path clip-path="url(#clip093)" d="M916.993 1817.64 L916.993 2174.14 L946.538 2174.14 L946.538 1817.64 L916.993 1817.64 L916.993 1817.64  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1817.64 916.993,2174.14 946.538,2174.14 946.538,1817.64 916.993,1817.64 "/>
-<path clip-path="url(#clip093)" d="M953.924 1793.51 L953.924 2174.14 L983.468 2174.14 L983.468 1793.51 L953.924 1793.51 L953.924 1793.51  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1793.51 953.924,2174.14 983.468,2174.14 983.468,1793.51 953.924,1793.51 "/>
-<path clip-path="url(#clip093)" d="M990.854 1769.49 L990.854 2174.14 L1020.4 2174.14 L1020.4 1769.49 L990.854 1769.49 L990.854 1769.49  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1769.49 990.854,2174.14 1020.4,2174.14 1020.4,1769.49 990.854,1769.49 "/>
-<path clip-path="url(#clip093)" d="M1027.78 1747.67 L1027.78 2174.14 L1057.33 2174.14 L1057.33 1747.67 L1027.78 1747.67 L1027.78 1747.67  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1747.67 1027.78,2174.14 1057.33,2174.14 1057.33,1747.67 1027.78,1747.67 "/>
-<path clip-path="url(#clip093)" d="M1064.71 1727.55 L1064.71 2174.14 L1094.26 2174.14 L1094.26 1727.55 L1064.71 1727.55 L1064.71 1727.55  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1727.55 1064.71,2174.14 1094.26,2174.14 1094.26,1727.55 1064.71,1727.55 "/>
-<path clip-path="url(#clip093)" d="M1101.64 1708.77 L1101.64 2174.14 L1131.19 2174.14 L1131.19 1708.77 L1101.64 1708.77 L1101.64 1708.77  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1708.77 1101.64,2174.14 1131.19,2174.14 1131.19,1708.77 1101.64,1708.77 "/>
-<path clip-path="url(#clip093)" d="M1138.58 1691.1 L1138.58 2174.14 L1168.12 2174.14 L1168.12 1691.1 L1138.58 1691.1 L1138.58 1691.1  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1691.1 1138.58,2174.14 1168.12,2174.14 1168.12,1691.1 1138.58,1691.1 "/>
-<path clip-path="url(#clip093)" d="M1175.51 1674.37 L1175.51 2174.14 L1205.05 2174.14 L1205.05 1674.37 L1175.51 1674.37 L1175.51 1674.37  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1674.37 1175.51,2174.14 1205.05,2174.14 1205.05,1674.37 1175.51,1674.37 "/>
-<path clip-path="url(#clip093)" d="M1212.44 1658.44 L1212.44 2174.14 L1241.98 2174.14 L1241.98 1658.44 L1212.44 1658.44 L1212.44 1658.44  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1658.44 1212.44,2174.14 1241.98,2174.14 1241.98,1658.44 1212.44,1658.44 "/>
-<path clip-path="url(#clip093)" d="M1249.37 1643.2 L1249.37 2174.14 L1278.91 2174.14 L1278.91 1643.2 L1249.37 1643.2 L1249.37 1643.2  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1643.2 1249.37,2174.14 1278.91,2174.14 1278.91,1643.2 1249.37,1643.2 "/>
-<path clip-path="url(#clip093)" d="M1286.3 1628.58 L1286.3 2174.14 L1315.84 2174.14 L1315.84 1628.58 L1286.3 1628.58 L1286.3 1628.58  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1628.58 1286.3,2174.14 1315.84,2174.14 1315.84,1628.58 1286.3,1628.58 "/>
-<path clip-path="url(#clip093)" d="M1323.23 1614.49 L1323.23 2174.14 L1352.77 2174.14 L1352.77 1614.49 L1323.23 1614.49 L1323.23 1614.49  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1614.49 1323.23,2174.14 1352.77,2174.14 1352.77,1614.49 1323.23,1614.49 "/>
-<path clip-path="url(#clip093)" d="M1360.16 1600.9 L1360.16 2174.14 L1389.7 2174.14 L1389.7 1600.9 L1360.16 1600.9 L1360.16 1600.9  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1600.9 1360.16,2174.14 1389.7,2174.14 1389.7,1600.9 1360.16,1600.9 "/>
-<path clip-path="url(#clip093)" d="M1397.09 1587.74 L1397.09 2174.14 L1426.63 2174.14 L1426.63 1587.74 L1397.09 1587.74 L1397.09 1587.74  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1587.74 1397.09,2174.14 1426.63,2174.14 1426.63,1587.74 1397.09,1587.74 "/>
-<path clip-path="url(#clip093)" d="M1434.02 1574.98 L1434.02 2174.14 L1463.56 2174.14 L1463.56 1574.98 L1434.02 1574.98 L1434.02 1574.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1574.98 1434.02,2174.14 1463.56,2174.14 1463.56,1574.98 1434.02,1574.98 "/>
-<path clip-path="url(#clip093)" d="M1470.95 1562.59 L1470.95 2174.14 L1500.49 2174.14 L1500.49 1562.59 L1470.95 1562.59 L1470.95 1562.59  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1562.59 1470.95,2174.14 1500.49,2174.14 1500.49,1562.59 1470.95,1562.59 "/>
-<path clip-path="url(#clip093)" d="M1507.88 1550.53 L1507.88 2174.14 L1537.42 2174.14 L1537.42 1550.53 L1507.88 1550.53 L1507.88 1550.53  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1550.53 1507.88,2174.14 1537.42,2174.14 1537.42,1550.53 1507.88,1550.53 "/>
-<path clip-path="url(#clip093)" d="M1544.81 1538.79 L1544.81 2174.14 L1574.35 2174.14 L1574.35 1538.79 L1544.81 1538.79 L1544.81 1538.79  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1538.79 1544.81,2174.14 1574.35,2174.14 1574.35,1538.79 1544.81,1538.79 "/>
-<path clip-path="url(#clip093)" d="M1581.74 1527.33 L1581.74 2174.14 L1611.28 2174.14 L1611.28 1527.33 L1581.74 1527.33 L1581.74 1527.33  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1527.33 1581.74,2174.14 1611.28,2174.14 1611.28,1527.33 1581.74,1527.33 "/>
-<path clip-path="url(#clip093)" d="M1618.67 1516.13 L1618.67 2174.14 L1648.21 2174.14 L1648.21 1516.13 L1618.67 1516.13 L1618.67 1516.13  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1516.13 1618.67,2174.14 1648.21,2174.14 1648.21,1516.13 1618.67,1516.13 "/>
-<path clip-path="url(#clip093)" d="M1655.6 1505.19 L1655.6 2174.14 L1685.14 2174.14 L1685.14 1505.19 L1655.6 1505.19 L1655.6 1505.19  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1505.19 1655.6,2174.14 1685.14,2174.14 1685.14,1505.19 1655.6,1505.19 "/>
-<path clip-path="url(#clip093)" d="M1692.53 1494.48 L1692.53 2174.14 L1722.07 2174.14 L1722.07 1494.48 L1692.53 1494.48 L1692.53 1494.48  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1494.48 1692.53,2174.14 1722.07,2174.14 1722.07,1494.48 1692.53,1494.48 "/>
-<path clip-path="url(#clip093)" d="M1729.46 1483.98 L1729.46 2174.14 L1759 2174.14 L1759 1483.98 L1729.46 1483.98 L1729.46 1483.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1483.98 1729.46,2174.14 1759,2174.14 1759,1483.98 1729.46,1483.98 "/>
-<path clip-path="url(#clip093)" d="M1766.39 1473.7 L1766.39 2174.14 L1795.93 2174.14 L1795.93 1473.7 L1766.39 1473.7 L1766.39 1473.7  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1473.7 1766.39,2174.14 1795.93,2174.14 1795.93,1473.7 1766.39,1473.7 "/>
-<path clip-path="url(#clip093)" d="M1803.32 1463.6 L1803.32 2174.14 L1832.86 2174.14 L1832.86 1463.6 L1803.32 1463.6 L1803.32 1463.6  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1463.6 1803.32,2174.14 1832.86,2174.14 1832.86,1463.6 1803.32,1463.6 "/>
-<path clip-path="url(#clip093)" d="M1840.25 1453.7 L1840.25 2174.14 L1869.79 2174.14 L1869.79 1453.7 L1840.25 1453.7 L1840.25 1453.7  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1453.7 1840.25,2174.14 1869.79,2174.14 1869.79,1453.7 1840.25,1453.7 "/>
-<path clip-path="url(#clip093)" d="M1877.18 1443.96 L1877.18 2174.14 L1906.73 2174.14 L1906.73 1443.96 L1877.18 1443.96 L1877.18 1443.96  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1443.96 1877.18,2174.14 1906.73,2174.14 1906.73,1443.96 1877.18,1443.96 "/>
-<path clip-path="url(#clip093)" d="M1914.11 1434.39 L1914.11 2174.14 L1943.66 2174.14 L1943.66 1434.39 L1914.11 1434.39 L1914.11 1434.39  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1434.39 1914.11,2174.14 1943.66,2174.14 1943.66,1434.39 1914.11,1434.39 "/>
-<path clip-path="url(#clip093)" d="M1951.04 1424.98 L1951.04 2174.14 L1980.59 2174.14 L1980.59 1424.98 L1951.04 1424.98 L1951.04 1424.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1424.98 1951.04,2174.14 1980.59,2174.14 1980.59,1424.98 1951.04,1424.98 "/>
-<path clip-path="url(#clip093)" d="M1987.97 1415.72 L1987.97 2174.14 L2017.52 2174.14 L2017.52 1415.72 L1987.97 1415.72 L1987.97 1415.72  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1415.72 1987.97,2174.14 2017.52,2174.14 2017.52,1415.72 1987.97,1415.72 "/>
-<path clip-path="url(#clip093)" d="M2024.9 1406.6 L2024.9 2174.14 L2054.45 2174.14 L2054.45 1406.6 L2024.9 1406.6 L2024.9 1406.6  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1406.6 2024.9,2174.14 2054.45,2174.14 2054.45,1406.6 2024.9,1406.6 "/>
-<path clip-path="url(#clip093)" d="M2061.83 1397.61 L2061.83 2174.14 L2091.38 2174.14 L2091.38 1397.61 L2061.83 1397.61 L2061.83 1397.61  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1397.61 2061.83,2174.14 2091.38,2174.14 2091.38,1397.61 2061.83,1397.61 "/>
-<path clip-path="url(#clip093)" d="M2098.76 1388.76 L2098.76 2174.14 L2128.31 2174.14 L2128.31 1388.76 L2098.76 1388.76 L2098.76 1388.76  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1388.76 2098.76,2174.14 2128.31,2174.14 2128.31,1388.76 2098.76,1388.76 "/>
-<path clip-path="url(#clip093)" d="M2135.69 1380.03 L2135.69 2174.14 L2165.24 2174.14 L2165.24 1380.03 L2135.69 1380.03 L2135.69 1380.03  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1380.03 2135.69,2174.14 2165.24,2174.14 2165.24,1380.03 2135.69,1380.03 "/>
-<path clip-path="url(#clip093)" d="M2172.62 1371.42 L2172.62 2174.14 L2202.17 2174.14 L2202.17 1371.42 L2172.62 1371.42 L2172.62 1371.42  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1371.42 2172.62,2174.14 2202.17,2174.14 2202.17,1371.42 2172.62,1371.42 "/>
-<path clip-path="url(#clip093)" d="M2209.55 1362.92 L2209.55 2174.14 L2239.1 2174.14 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,2174.14 2239.1,2174.14 2239.1,1362.92 2209.55,1362.92 "/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="451.672" cy="2078.86" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="488.602" cy="2039.4" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="525.532" cy="2009.11" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="562.463" cy="1983.58" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="599.393" cy="1961.09" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="636.323" cy="1940.76" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="673.253" cy="1922.06" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="710.184" cy="1904.65" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="747.114" cy="1888.3" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="784.044" cy="1872.84" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="820.975" cy="1858.14" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="857.905" cy="1844.08" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="894.835" cy="1830.61" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="931.766" cy="1817.64" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="968.696" cy="1793.51" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1005.63" cy="1769.49" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1042.56" cy="1747.67" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1079.49" cy="1727.55" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1116.42" cy="1708.77" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1153.35" cy="1691.1" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1190.28" cy="1674.37" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1227.21" cy="1658.44" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1264.14" cy="1643.2" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1301.07" cy="1628.58" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1338" cy="1614.49" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1374.93" cy="1600.9" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1411.86" cy="1587.74" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1448.79" cy="1574.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1485.72" cy="1562.59" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1522.65" cy="1550.53" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1559.58" cy="1538.79" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1596.51" cy="1527.33" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1633.44" cy="1516.13" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1670.37" cy="1505.19" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1707.3" cy="1494.48" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1744.23" cy="1483.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1781.16" cy="1473.7" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1818.09" cy="1463.6" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1855.02" cy="1453.7" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1891.95" cy="1443.96" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1928.88" cy="1434.39" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1965.81" cy="1424.98" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2002.74" cy="1415.72" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2039.67" cy="1406.6" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2076.6" cy="1397.61" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2113.53" cy="1388.76" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2150.47" cy="1380.03" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2187.4" cy="1371.42" r="2"/>
-<circle clip-path="url(#clip093)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
-<path clip-path="url(#clip090)" d="M355.193 1597.33 L677.137 1597.33 L677.137 1389.97 L355.193 1389.97  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="355.193,1597.33 677.137,1597.33 677.137,1389.97 355.193,1389.97 355.193,1597.33 "/>
-<path clip-path="url(#clip090)" d="M378.153 1462.54 L515.916 1462.54 L515.916 1421.07 L378.153 1421.07 L378.153 1462.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1462.54 515.916,1462.54 515.916,1421.07 378.153,1421.07 378.153,1462.54 "/>
-<path clip-path="url(#clip090)" d="M538.877 1424.53 L543.553 1424.53 L543.553 1459.09 L538.877 1459.09 L538.877 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M557.534 1442.58 L557.534 1455.24 L565.034 1455.24 Q568.807 1455.24 570.613 1453.69 Q572.441 1452.12 572.441 1448.9 Q572.441 1445.66 570.613 1444.13 Q568.807 1442.58 565.034 1442.58 L557.534 1442.58 M557.534 1428.37 L557.534 1438.78 L564.455 1438.78 Q567.881 1438.78 569.548 1437.51 Q571.238 1436.21 571.238 1433.58 Q571.238 1430.96 569.548 1429.66 Q567.881 1428.37 564.455 1428.37 L557.534 1428.37 M552.858 1424.53 L564.802 1424.53 Q570.15 1424.53 573.043 1426.75 Q575.937 1428.97 575.937 1433.07 Q575.937 1436.24 574.455 1438.11 Q572.974 1439.99 570.103 1440.45 Q573.552 1441.19 575.451 1443.55 Q577.372 1445.89 577.372 1449.41 Q577.372 1454.04 574.224 1456.56 Q571.076 1459.09 565.265 1459.09 L552.858 1459.09 L552.858 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M585.381 1424.53 L592.349 1424.53 L601.168 1448.04 L610.034 1424.53 L617.001 1424.53 L617.001 1459.09 L612.441 1459.09 L612.441 1428.74 L603.529 1452.44 L598.83 1452.44 L589.918 1428.74 L589.918 1459.09 L585.381 1459.09 L585.381 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M378.153 1514.38 L515.916 1514.38 L515.916 1472.91 L378.153 1472.91 L378.153 1514.38  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1514.38 515.916,1514.38 515.916,1472.91 378.153,1472.91 378.153,1514.38 "/>
-<path clip-path="url(#clip090)" d="M538.877 1476.37 L543.599 1476.37 L550.867 1505.58 L558.113 1476.37 L563.367 1476.37 L570.636 1505.58 L577.881 1476.37 L582.626 1476.37 L573.946 1510.93 L568.066 1510.93 L560.775 1480.93 L553.414 1510.93 L547.534 1510.93 L538.877 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M588.83 1476.37 L595.798 1476.37 L604.617 1499.88 L613.483 1476.37 L620.45 1476.37 L620.45 1510.93 L615.89 1510.93 L615.89 1480.58 L606.978 1504.28 L602.279 1504.28 L593.367 1480.58 L593.367 1510.93 L588.83 1510.93 L588.83 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M624.941 1476.37 L654.177 1476.37 L654.177 1480.3 L641.909 1480.3 L641.909 1510.93 L637.209 1510.93 L637.209 1480.3 L624.941 1480.3 L624.941 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M378.153 1566.22 L515.916 1566.22 L515.916 1524.75 L378.153 1524.75 L378.153 1566.22  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1566.22 515.916,1566.22 515.916,1524.75 378.153,1524.75 378.153,1566.22 "/>
-<path clip-path="url(#clip090)" d="M561.122 1529.34 L561.122 1533.9 Q558.46 1532.63 556.099 1532 Q553.738 1531.38 551.539 1531.38 Q547.719 1531.38 545.636 1532.86 Q543.576 1534.34 543.576 1537.07 Q543.576 1539.36 544.941 1540.54 Q546.33 1541.7 550.173 1542.42 L552.997 1543 Q558.228 1543.99 560.705 1546.52 Q563.205 1549.02 563.205 1553.23 Q563.205 1558.25 559.826 1560.84 Q556.469 1563.44 549.965 1563.44 Q547.511 1563.44 544.733 1562.88 Q541.979 1562.33 539.016 1561.24 L539.016 1556.42 Q541.863 1558.02 544.594 1558.83 Q547.326 1559.64 549.965 1559.64 Q553.969 1559.64 556.145 1558.07 Q558.321 1556.49 558.321 1553.58 Q558.321 1551.03 556.747 1549.59 Q555.196 1548.16 551.631 1547.44 L548.784 1546.89 Q543.553 1545.84 541.215 1543.62 Q538.877 1541.4 538.877 1537.44 Q538.877 1532.86 542.094 1530.22 Q545.335 1527.58 551.006 1527.58 Q553.437 1527.58 555.96 1528.02 Q558.483 1528.46 561.122 1529.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M570.497 1528.21 L592.349 1528.21 L592.349 1532.14 L575.173 1532.14 L575.173 1542.37 L591.631 1542.37 L591.631 1546.31 L575.173 1546.31 L575.173 1558.83 L592.765 1558.83 L592.765 1562.77 L570.497 1562.77 L570.497 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M600.45 1528.21 L605.126 1528.21 L605.126 1542.37 L622.117 1542.37 L622.117 1528.21 L626.793 1528.21 L626.793 1562.77 L622.117 1562.77 L622.117 1546.31 L605.126 1546.31 L605.126 1562.77 L600.45 1562.77 L600.45 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M636.098 1528.21 L640.774 1528.21 L640.774 1562.77 L636.098 1562.77 L636.098 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="377.811,2174.14 377.811,1362.92 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="747.114,2174.14 747.114,1362.92 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1116.42,2174.14 1116.42,1362.92 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1485.72,2174.14 1485.72,1362.92 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1855.02,2174.14 1855.02,1362.92 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2224.33,2174.14 2224.33,1362.92 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 2352.76,2174.14 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="377.811,2174.14 377.811,2155.24 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="747.114,2174.14 747.114,2155.24 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1116.42,2174.14 1116.42,2155.24 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1485.72,2174.14 1485.72,2155.24 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1855.02,2174.14 1855.02,2155.24 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2224.33,2174.14 2224.33,2155.24 "/>
+<path clip-path="url(#clip910)" d="M377.811 2205.06 Q374.2 2205.06 372.371 2208.62 Q370.566 2212.16 370.566 2219.29 Q370.566 2226.4 372.371 2229.97 Q374.2 2233.51 377.811 2233.51 Q381.445 2233.51 383.251 2229.97 Q385.08 2226.4 385.08 2219.29 Q385.08 2212.16 383.251 2208.62 Q381.445 2205.06 377.811 2205.06 M377.811 2201.35 Q383.621 2201.35 386.677 2205.96 Q389.756 2210.54 389.756 2219.29 Q389.756 2228.02 386.677 2232.63 Q383.621 2237.21 377.811 2237.21 Q372.001 2237.21 368.922 2232.63 Q365.867 2228.02 365.867 2219.29 Q365.867 2210.54 368.922 2205.96 Q372.001 2201.35 377.811 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M721.802 2232.6 L729.441 2232.6 L729.441 2206.24 L721.13 2207.91 L721.13 2203.65 L729.394 2201.98 L734.07 2201.98 L734.07 2232.6 L741.709 2232.6 L741.709 2236.54 L721.802 2236.54 L721.802 2232.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M761.153 2205.06 Q757.542 2205.06 755.714 2208.62 Q753.908 2212.16 753.908 2219.29 Q753.908 2226.4 755.714 2229.97 Q757.542 2233.51 761.153 2233.51 Q764.788 2233.51 766.593 2229.97 Q768.422 2226.4 768.422 2219.29 Q768.422 2212.16 766.593 2208.62 Q764.788 2205.06 761.153 2205.06 M761.153 2201.35 Q766.964 2201.35 770.019 2205.96 Q773.098 2210.54 773.098 2219.29 Q773.098 2228.02 770.019 2232.63 Q766.964 2237.21 761.153 2237.21 Q755.343 2237.21 752.265 2232.63 Q749.209 2228.02 749.209 2219.29 Q749.209 2210.54 752.265 2205.96 Q755.343 2201.35 761.153 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1095.19 2232.6 L1111.51 2232.6 L1111.51 2236.54 L1089.57 2236.54 L1089.57 2232.6 Q1092.23 2229.85 1096.81 2225.22 Q1101.42 2220.57 1102.6 2219.23 Q1104.84 2216.7 1105.72 2214.97 Q1106.63 2213.21 1106.63 2211.52 Q1106.63 2208.76 1104.68 2207.03 Q1102.76 2205.29 1099.66 2205.29 Q1097.46 2205.29 1095.01 2206.05 Q1092.57 2206.82 1089.8 2208.37 L1089.8 2203.65 Q1092.62 2202.51 1095.07 2201.93 Q1097.53 2201.35 1099.57 2201.35 Q1104.94 2201.35 1108.13 2204.04 Q1111.32 2206.73 1111.32 2211.22 Q1111.32 2213.35 1110.51 2215.27 Q1109.73 2217.16 1107.62 2219.76 Q1107.04 2220.43 1103.94 2223.65 Q1100.84 2226.84 1095.19 2232.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1131.32 2205.06 Q1127.71 2205.06 1125.88 2208.62 Q1124.08 2212.16 1124.08 2219.29 Q1124.08 2226.4 1125.88 2229.97 Q1127.71 2233.51 1131.32 2233.51 Q1134.96 2233.51 1136.76 2229.97 Q1138.59 2226.4 1138.59 2219.29 Q1138.59 2212.16 1136.76 2208.62 Q1134.96 2205.06 1131.32 2205.06 M1131.32 2201.35 Q1137.13 2201.35 1140.19 2205.96 Q1143.27 2210.54 1143.27 2219.29 Q1143.27 2228.02 1140.19 2232.63 Q1137.13 2237.21 1131.32 2237.21 Q1125.51 2237.21 1122.44 2232.63 Q1119.38 2228.02 1119.38 2219.29 Q1119.38 2210.54 1122.44 2205.96 Q1125.51 2201.35 1131.32 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1474.56 2217.91 Q1477.92 2218.62 1479.79 2220.89 Q1481.69 2223.16 1481.69 2226.49 Q1481.69 2231.61 1478.17 2234.41 Q1474.66 2237.21 1468.17 2237.21 Q1466 2237.21 1463.68 2236.77 Q1461.39 2236.35 1458.94 2235.5 L1458.94 2230.98 Q1460.88 2232.12 1463.2 2232.7 Q1465.51 2233.28 1468.03 2233.28 Q1472.43 2233.28 1474.72 2231.54 Q1477.04 2229.8 1477.04 2226.49 Q1477.04 2223.44 1474.89 2221.73 Q1472.76 2219.99 1468.94 2219.99 L1464.91 2219.99 L1464.91 2216.15 L1469.12 2216.15 Q1472.57 2216.15 1474.4 2214.78 Q1476.23 2213.39 1476.23 2210.8 Q1476.23 2208.14 1474.33 2206.73 Q1472.46 2205.29 1468.94 2205.29 Q1467.02 2205.29 1464.82 2205.71 Q1462.62 2206.12 1459.98 2207 L1459.98 2202.84 Q1462.64 2202.1 1464.96 2201.73 Q1467.29 2201.35 1469.35 2201.35 Q1474.68 2201.35 1477.78 2203.79 Q1480.88 2206.19 1480.88 2210.31 Q1480.88 2213.18 1479.24 2215.17 Q1477.59 2217.14 1474.56 2217.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1500.56 2205.06 Q1496.95 2205.06 1495.12 2208.62 Q1493.31 2212.16 1493.31 2219.29 Q1493.31 2226.4 1495.12 2229.97 Q1496.95 2233.51 1500.56 2233.51 Q1504.19 2233.51 1506 2229.97 Q1507.83 2226.4 1507.83 2219.29 Q1507.83 2212.16 1506 2208.62 Q1504.19 2205.06 1500.56 2205.06 M1500.56 2201.35 Q1506.37 2201.35 1509.42 2205.96 Q1512.5 2210.54 1512.5 2219.29 Q1512.5 2228.02 1509.42 2232.63 Q1506.37 2237.21 1500.56 2237.21 Q1494.75 2237.21 1491.67 2232.63 Q1488.61 2228.02 1488.61 2219.29 Q1488.61 2210.54 1491.67 2205.96 Q1494.75 2201.35 1500.56 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1843.19 2206.05 L1831.39 2224.5 L1843.19 2224.5 L1843.19 2206.05 M1841.97 2201.98 L1847.85 2201.98 L1847.85 2224.5 L1852.78 2224.5 L1852.78 2228.39 L1847.85 2228.39 L1847.85 2236.54 L1843.19 2236.54 L1843.19 2228.39 L1827.59 2228.39 L1827.59 2223.88 L1841.97 2201.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1870.51 2205.06 Q1866.9 2205.06 1865.07 2208.62 Q1863.26 2212.16 1863.26 2219.29 Q1863.26 2226.4 1865.07 2229.97 Q1866.9 2233.51 1870.51 2233.51 Q1874.14 2233.51 1875.95 2229.97 Q1877.78 2226.4 1877.78 2219.29 Q1877.78 2212.16 1875.95 2208.62 Q1874.14 2205.06 1870.51 2205.06 M1870.51 2201.35 Q1876.32 2201.35 1879.37 2205.96 Q1882.45 2210.54 1882.45 2219.29 Q1882.45 2228.02 1879.37 2232.63 Q1876.32 2237.21 1870.51 2237.21 Q1864.7 2237.21 1861.62 2232.63 Q1858.56 2228.02 1858.56 2219.29 Q1858.56 2210.54 1861.62 2205.96 Q1864.7 2201.35 1870.51 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2199.02 2201.98 L2217.38 2201.98 L2217.38 2205.92 L2203.31 2205.92 L2203.31 2214.39 Q2204.33 2214.04 2205.34 2213.88 Q2206.36 2213.69 2207.38 2213.69 Q2213.17 2213.69 2216.55 2216.86 Q2219.93 2220.04 2219.93 2225.45 Q2219.93 2231.03 2216.46 2234.13 Q2212.98 2237.21 2206.66 2237.21 Q2204.49 2237.21 2202.22 2236.84 Q2199.97 2236.47 2197.57 2235.73 L2197.57 2231.03 Q2199.65 2232.16 2201.87 2232.72 Q2204.09 2233.28 2206.57 2233.28 Q2210.58 2233.28 2212.91 2231.17 Q2215.25 2229.06 2215.25 2225.45 Q2215.25 2221.84 2212.91 2219.73 Q2210.58 2217.63 2206.57 2217.63 Q2204.7 2217.63 2202.82 2218.04 Q2200.97 2218.46 2199.02 2219.34 L2199.02 2201.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M2239.14 2205.06 Q2235.53 2205.06 2233.7 2208.62 Q2231.9 2212.16 2231.9 2219.29 Q2231.9 2226.4 2233.7 2229.97 Q2235.53 2233.51 2239.14 2233.51 Q2242.77 2233.51 2244.58 2229.97 Q2246.41 2226.4 2246.41 2219.29 Q2246.41 2212.16 2244.58 2208.62 Q2242.77 2205.06 2239.14 2205.06 M2239.14 2201.35 Q2244.95 2201.35 2248.01 2205.96 Q2251.08 2210.54 2251.08 2219.29 Q2251.08 2228.02 2248.01 2232.63 Q2244.95 2237.21 2239.14 2237.21 Q2233.33 2237.21 2230.25 2232.63 Q2227.2 2228.02 2227.2 2219.29 Q2227.2 2210.54 2230.25 2205.96 Q2233.33 2201.35 2239.14 2201.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1184.33 2273.04 L1184.33 2279.31 Q1180.67 2277.56 1177.42 2276.7 Q1174.17 2275.84 1171.15 2275.84 Q1165.9 2275.84 1163.03 2277.88 Q1160.2 2279.92 1160.2 2283.67 Q1160.2 2286.83 1162.08 2288.45 Q1163.99 2290.04 1169.27 2291.03 L1173.15 2291.82 Q1180.35 2293.19 1183.75 2296.66 Q1187.19 2300.1 1187.19 2305.89 Q1187.19 2312.8 1182.54 2316.36 Q1177.93 2319.93 1168.98 2319.93 Q1165.61 2319.93 1161.79 2319.16 Q1158 2318.4 1153.93 2316.9 L1153.93 2310.28 Q1157.84 2312.48 1161.6 2313.59 Q1165.36 2314.71 1168.98 2314.71 Q1174.49 2314.71 1177.48 2312.54 Q1180.47 2310.38 1180.47 2306.37 Q1180.47 2302.87 1178.31 2300.89 Q1176.18 2298.92 1171.28 2297.93 L1167.36 2297.17 Q1160.17 2295.74 1156.95 2292.68 Q1153.74 2289.63 1153.74 2284.18 Q1153.74 2277.88 1158.16 2274.25 Q1162.62 2270.62 1170.42 2270.62 Q1173.76 2270.62 1177.23 2271.23 Q1180.7 2271.83 1184.33 2273.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1210.78 2287.46 Q1206.06 2287.46 1203.33 2291.15 Q1200.59 2294.81 1200.59 2301.21 Q1200.59 2307.61 1203.3 2311.3 Q1206.03 2314.96 1210.78 2314.96 Q1215.45 2314.96 1218.19 2311.27 Q1220.93 2307.58 1220.93 2301.21 Q1220.93 2294.88 1218.19 2291.19 Q1215.45 2287.46 1210.78 2287.46 M1210.78 2282.5 Q1218.41 2282.5 1222.77 2287.46 Q1227.14 2292.43 1227.14 2301.21 Q1227.14 2309.96 1222.77 2314.96 Q1218.41 2319.93 1210.78 2319.93 Q1203.1 2319.93 1198.74 2314.96 Q1194.42 2309.96 1194.42 2301.21 Q1194.42 2292.43 1198.74 2287.46 Q1203.1 2282.5 1210.78 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1236.84 2269.48 L1242.7 2269.48 L1242.7 2319 L1236.84 2319 L1236.84 2269.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1254.35 2304.94 L1254.35 2283.36 L1260.21 2283.36 L1260.21 2304.71 Q1260.21 2309.77 1262.18 2312.32 Q1264.15 2314.83 1268.1 2314.83 Q1272.84 2314.83 1275.58 2311.81 Q1278.35 2308.79 1278.35 2303.57 L1278.35 2283.36 L1284.2 2283.36 L1284.2 2319 L1278.35 2319 L1278.35 2313.53 Q1276.21 2316.78 1273.38 2318.37 Q1270.58 2319.93 1266.86 2319.93 Q1260.71 2319.93 1257.53 2316.11 Q1254.35 2312.29 1254.35 2304.94 M1269.09 2282.5 L1269.09 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1302.06 2273.23 L1302.06 2283.36 L1314.12 2283.36 L1314.12 2287.91 L1302.06 2287.91 L1302.06 2307.26 Q1302.06 2311.62 1303.24 2312.86 Q1304.45 2314.1 1308.11 2314.1 L1314.12 2314.1 L1314.12 2319 L1308.11 2319 Q1301.33 2319 1298.75 2316.49 Q1296.17 2313.94 1296.17 2307.26 L1296.17 2287.91 L1291.87 2287.91 L1291.87 2283.36 L1296.17 2283.36 L1296.17 2273.23 L1302.06 2273.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1321.83 2283.36 L1327.68 2283.36 L1327.68 2319 L1321.83 2319 L1321.83 2283.36 M1321.83 2269.48 L1327.68 2269.48 L1327.68 2276.89 L1321.83 2276.89 L1321.83 2269.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1353.75 2287.46 Q1349.04 2287.46 1346.3 2291.15 Q1343.56 2294.81 1343.56 2301.21 Q1343.56 2307.61 1346.27 2311.3 Q1349.01 2314.96 1353.75 2314.96 Q1358.43 2314.96 1361.17 2311.27 Q1363.9 2307.58 1363.9 2301.21 Q1363.9 2294.88 1361.17 2291.19 Q1358.43 2287.46 1353.75 2287.46 M1353.75 2282.5 Q1361.39 2282.5 1365.75 2287.46 Q1370.11 2292.43 1370.11 2301.21 Q1370.11 2309.96 1365.75 2314.96 Q1361.39 2319.93 1353.75 2319.93 Q1346.08 2319.93 1341.72 2314.96 Q1337.39 2309.96 1337.39 2301.21 Q1337.39 2292.43 1341.72 2287.46 Q1346.08 2282.5 1353.75 2282.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1409.45 2297.49 L1409.45 2319 L1403.59 2319 L1403.59 2297.68 Q1403.59 2292.62 1401.62 2290.1 Q1399.65 2287.59 1395.7 2287.59 Q1390.96 2287.59 1388.22 2290.61 Q1385.48 2293.64 1385.48 2298.86 L1385.48 2319 L1379.59 2319 L1379.59 2283.36 L1385.48 2283.36 L1385.48 2288.89 Q1387.58 2285.68 1390.42 2284.09 Q1393.28 2282.5 1397 2282.5 Q1403.15 2282.5 1406.3 2286.32 Q1409.45 2290.1 1409.45 2297.49 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1469.03 2290.33 L1459.77 2290.33 L1457.1 2300.96 L1466.42 2300.96 L1469.03 2290.33 M1464.26 2272.22 L1460.95 2285.42 L1470.24 2285.42 L1473.58 2272.22 L1478.68 2272.22 L1475.4 2285.42 L1485.33 2285.42 L1485.33 2290.33 L1474.16 2290.33 L1471.55 2300.96 L1481.67 2300.96 L1481.67 2305.83 L1470.31 2305.83 L1467 2319 L1461.9 2319 L1465.18 2305.83 L1455.86 2305.83 L1452.58 2319 L1447.45 2319 L1450.76 2305.83 L1440.74 2305.83 L1440.74 2300.96 L1451.94 2300.96 L1454.61 2290.33 L1444.37 2290.33 L1444.37 2285.42 L1455.86 2285.42 L1459.1 2272.22 L1464.26 2272.22 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,2174.14 2352.76,2174.14 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1971.34 2352.76,1971.34 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1768.53 2352.76,1768.53 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1565.73 2352.76,1565.73 "/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="286.311,1362.92 2352.76,1362.92 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 286.311,1362.92 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,2174.14 305.209,2174.14 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1971.34 305.209,1971.34 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1768.53 305.209,1768.53 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1565.73 305.209,1565.73 "/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="286.311,1362.92 305.209,1362.92 "/>
+<path clip-path="url(#clip910)" d="M238.367 2159.94 Q234.756 2159.94 232.927 2163.5 Q231.122 2167.04 231.122 2174.17 Q231.122 2181.28 232.927 2184.85 Q234.756 2188.39 238.367 2188.39 Q242.001 2188.39 243.807 2184.85 Q245.635 2181.28 245.635 2174.17 Q245.635 2167.04 243.807 2163.5 Q242.001 2159.94 238.367 2159.94 M238.367 2156.23 Q244.177 2156.23 247.233 2160.84 Q250.311 2165.42 250.311 2174.17 Q250.311 2182.9 247.233 2187.51 Q244.177 2192.09 238.367 2192.09 Q232.557 2192.09 229.478 2187.51 Q226.422 2182.9 226.422 2174.17 Q226.422 2165.42 229.478 2160.84 Q232.557 2156.23 238.367 2156.23 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M172.071 1984.68 L188.39 1984.68 L188.39 1988.62 L166.446 1988.62 L166.446 1984.68 Q169.108 1981.93 173.691 1977.3 Q178.298 1972.64 179.478 1971.3 Q181.724 1968.78 182.603 1967.04 Q183.506 1965.28 183.506 1963.59 Q183.506 1960.84 181.562 1959.1 Q179.64 1957.37 176.538 1957.37 Q174.339 1957.37 171.886 1958.13 Q169.455 1958.89 166.677 1960.44 L166.677 1955.72 Q169.501 1954.59 171.955 1954.01 Q174.409 1953.43 176.446 1953.43 Q181.816 1953.43 185.011 1956.12 Q188.205 1958.8 188.205 1963.29 Q188.205 1965.42 187.395 1967.34 Q186.608 1969.24 184.501 1971.83 Q183.923 1972.5 180.821 1975.72 Q177.719 1978.92 172.071 1984.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M198.251 1954.06 L216.608 1954.06 L216.608 1957.99 L202.534 1957.99 L202.534 1966.46 Q203.552 1966.12 204.571 1965.95 Q205.589 1965.77 206.608 1965.77 Q212.395 1965.77 215.774 1968.94 Q219.154 1972.11 219.154 1977.53 Q219.154 1983.11 215.682 1986.21 Q212.21 1989.29 205.89 1989.29 Q203.714 1989.29 201.446 1988.92 Q199.2 1988.55 196.793 1987.81 L196.793 1983.11 Q198.876 1984.24 201.099 1984.8 Q203.321 1985.35 205.798 1985.35 Q209.802 1985.35 212.14 1983.25 Q214.478 1981.14 214.478 1977.53 Q214.478 1973.92 212.14 1971.81 Q209.802 1969.7 205.798 1969.7 Q203.923 1969.7 202.048 1970.12 Q200.196 1970.54 198.251 1971.42 L198.251 1954.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 1957.13 Q234.756 1957.13 232.927 1960.7 Q231.122 1964.24 231.122 1971.37 Q231.122 1978.48 232.927 1982.04 Q234.756 1985.58 238.367 1985.58 Q242.001 1985.58 243.807 1982.04 Q245.635 1978.48 245.635 1971.37 Q245.635 1964.24 243.807 1960.7 Q242.001 1957.13 238.367 1957.13 M238.367 1953.43 Q244.177 1953.43 247.233 1958.04 Q250.311 1962.62 250.311 1971.37 Q250.311 1980.1 247.233 1984.7 Q244.177 1989.29 238.367 1989.29 Q232.557 1989.29 229.478 1984.7 Q226.422 1980.1 226.422 1971.37 Q226.422 1962.62 229.478 1958.04 Q232.557 1953.43 238.367 1953.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M168.089 1751.25 L186.446 1751.25 L186.446 1755.19 L172.372 1755.19 L172.372 1763.66 Q173.39 1763.31 174.409 1763.15 Q175.427 1762.97 176.446 1762.97 Q182.233 1762.97 185.612 1766.14 Q188.992 1769.31 188.992 1774.72 Q188.992 1780.3 185.52 1783.4 Q182.048 1786.48 175.728 1786.48 Q173.552 1786.48 171.284 1786.11 Q169.039 1785.74 166.631 1785 L166.631 1780.3 Q168.714 1781.44 170.937 1781.99 Q173.159 1782.55 175.636 1782.55 Q179.64 1782.55 181.978 1780.44 Q184.316 1778.34 184.316 1774.72 Q184.316 1771.11 181.978 1769.01 Q179.64 1766.9 175.636 1766.9 Q173.761 1766.9 171.886 1767.32 Q170.034 1767.73 168.089 1768.61 L168.089 1751.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M208.205 1754.33 Q204.594 1754.33 202.765 1757.9 Q200.96 1761.44 200.96 1768.57 Q200.96 1775.67 202.765 1779.24 Q204.594 1782.78 208.205 1782.78 Q211.839 1782.78 213.645 1779.24 Q215.473 1775.67 215.473 1768.57 Q215.473 1761.44 213.645 1757.9 Q211.839 1754.33 208.205 1754.33 M208.205 1750.63 Q214.015 1750.63 217.071 1755.23 Q220.149 1759.82 220.149 1768.57 Q220.149 1777.29 217.071 1781.9 Q214.015 1786.48 208.205 1786.48 Q202.395 1786.48 199.316 1781.9 Q196.261 1777.29 196.261 1768.57 Q196.261 1759.82 199.316 1755.23 Q202.395 1750.63 208.205 1750.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 1754.33 Q234.756 1754.33 232.927 1757.9 Q231.122 1761.44 231.122 1768.57 Q231.122 1775.67 232.927 1779.24 Q234.756 1782.78 238.367 1782.78 Q242.001 1782.78 243.807 1779.24 Q245.635 1775.67 245.635 1768.57 Q245.635 1761.44 243.807 1757.9 Q242.001 1754.33 238.367 1754.33 M238.367 1750.63 Q244.177 1750.63 247.233 1755.23 Q250.311 1759.82 250.311 1768.57 Q250.311 1777.29 247.233 1781.9 Q244.177 1786.48 238.367 1786.48 Q232.557 1786.48 229.478 1781.9 Q226.422 1777.29 226.422 1768.57 Q226.422 1759.82 229.478 1755.23 Q232.557 1750.63 238.367 1750.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M166.863 1548.45 L189.085 1548.45 L189.085 1550.44 L176.538 1583.01 L171.654 1583.01 L183.46 1552.38 L166.863 1552.38 L166.863 1548.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M198.251 1548.45 L216.608 1548.45 L216.608 1552.38 L202.534 1552.38 L202.534 1560.86 Q203.552 1560.51 204.571 1560.35 Q205.589 1560.16 206.608 1560.16 Q212.395 1560.16 215.774 1563.33 Q219.154 1566.5 219.154 1571.92 Q219.154 1577.5 215.682 1580.6 Q212.21 1583.68 205.89 1583.68 Q203.714 1583.68 201.446 1583.31 Q199.2 1582.94 196.793 1582.2 L196.793 1577.5 Q198.876 1578.63 201.099 1579.19 Q203.321 1579.74 205.798 1579.74 Q209.802 1579.74 212.14 1577.64 Q214.478 1575.53 214.478 1571.92 Q214.478 1568.31 212.14 1566.2 Q209.802 1564.1 205.798 1564.1 Q203.923 1564.1 202.048 1564.51 Q200.196 1564.93 198.251 1565.81 L198.251 1548.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 1551.53 Q234.756 1551.53 232.927 1555.09 Q231.122 1558.63 231.122 1565.76 Q231.122 1572.87 232.927 1576.43 Q234.756 1579.98 238.367 1579.98 Q242.001 1579.98 243.807 1576.43 Q245.635 1572.87 245.635 1565.76 Q245.635 1558.63 243.807 1555.09 Q242.001 1551.53 238.367 1551.53 M238.367 1547.82 Q244.177 1547.82 247.233 1552.43 Q250.311 1557.01 250.311 1565.76 Q250.311 1574.49 247.233 1579.1 Q244.177 1583.68 238.367 1583.68 Q232.557 1583.68 229.478 1579.1 Q226.422 1574.49 226.422 1565.76 Q226.422 1557.01 229.478 1552.43 Q232.557 1547.82 238.367 1547.82 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M138.691 1376.27 L146.33 1376.27 L146.33 1349.9 L138.02 1351.57 L138.02 1347.31 L146.284 1345.64 L150.96 1345.64 L150.96 1376.27 L158.599 1376.27 L158.599 1380.2 L138.691 1380.2 L138.691 1376.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M178.043 1348.72 Q174.432 1348.72 172.603 1352.29 Q170.798 1355.83 170.798 1362.96 Q170.798 1370.07 172.603 1373.63 Q174.432 1377.17 178.043 1377.17 Q181.677 1377.17 183.483 1373.63 Q185.312 1370.07 185.312 1362.96 Q185.312 1355.83 183.483 1352.29 Q181.677 1348.72 178.043 1348.72 M178.043 1345.02 Q183.853 1345.02 186.909 1349.63 Q189.987 1354.21 189.987 1362.96 Q189.987 1371.69 186.909 1376.29 Q183.853 1380.88 178.043 1380.88 Q172.233 1380.88 169.154 1376.29 Q166.099 1371.69 166.099 1362.96 Q166.099 1354.21 169.154 1349.63 Q172.233 1345.02 178.043 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M208.205 1348.72 Q204.594 1348.72 202.765 1352.29 Q200.96 1355.83 200.96 1362.96 Q200.96 1370.07 202.765 1373.63 Q204.594 1377.17 208.205 1377.17 Q211.839 1377.17 213.645 1373.63 Q215.473 1370.07 215.473 1362.96 Q215.473 1355.83 213.645 1352.29 Q211.839 1348.72 208.205 1348.72 M208.205 1345.02 Q214.015 1345.02 217.071 1349.63 Q220.149 1354.21 220.149 1362.96 Q220.149 1371.69 217.071 1376.29 Q214.015 1380.88 208.205 1380.88 Q202.395 1380.88 199.316 1376.29 Q196.261 1371.69 196.261 1362.96 Q196.261 1354.21 199.316 1349.63 Q202.395 1345.02 208.205 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M238.367 1348.72 Q234.756 1348.72 232.927 1352.29 Q231.122 1355.83 231.122 1362.96 Q231.122 1370.07 232.927 1373.63 Q234.756 1377.17 238.367 1377.17 Q242.001 1377.17 243.807 1373.63 Q245.635 1370.07 245.635 1362.96 Q245.635 1355.83 243.807 1352.29 Q242.001 1348.72 238.367 1348.72 M238.367 1345.02 Q244.177 1345.02 247.233 1349.63 Q250.311 1354.21 250.311 1362.96 Q250.311 1371.69 247.233 1376.29 Q244.177 1380.88 238.367 1380.88 Q232.557 1380.88 229.478 1376.29 Q226.422 1371.69 226.422 1362.96 Q226.422 1354.21 229.478 1349.63 Q232.557 1345.02 238.367 1345.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M40.2442 2003.67 L40.2442 1997.24 L87.7642 1997.24 L87.7642 2003.67 L40.2442 2003.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M66.2481 1955.06 L87.7642 1955.06 L87.7642 1960.92 L66.439 1960.92 Q61.3783 1960.92 58.8638 1962.89 Q56.3494 1964.87 56.3494 1968.81 Q56.3494 1973.56 59.3731 1976.29 Q62.3968 1979.03 67.6167 1979.03 L87.7642 1979.03 L87.7642 1984.92 L52.1162 1984.92 L52.1162 1979.03 L57.6544 1979.03 Q54.4397 1976.93 52.8483 1974.1 Q51.2568 1971.23 51.2568 1967.51 Q51.2568 1961.37 55.0763 1958.21 Q58.8638 1955.06 66.2481 1955.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M52.1162 1947.58 L52.1162 1941.38 L82.035 1930.24 L52.1162 1919.1 L52.1162 1912.89 L87.7642 1926.26 L87.7642 1934.22 L52.1162 1947.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M68.4761 1874.31 L71.3406 1874.31 L71.3406 1901.24 Q77.3881 1900.86 80.5709 1897.61 Q83.7219 1894.33 83.7219 1888.51 Q83.7219 1885.14 82.8944 1881.98 Q82.0669 1878.8 80.4118 1875.68 L85.9499 1875.68 Q87.2867 1878.83 87.987 1882.14 Q88.6872 1885.45 88.6872 1888.86 Q88.6872 1897.39 83.7219 1902.39 Q78.7567 1907.35 70.2903 1907.35 Q61.5374 1907.35 56.4131 1902.64 Q51.2568 1897.9 51.2568 1889.88 Q51.2568 1882.69 55.9038 1878.52 Q60.5189 1874.31 68.4761 1874.31 M66.7573 1880.17 Q61.9512 1880.23 59.0866 1882.88 Q56.2221 1885.49 56.2221 1889.81 Q56.2221 1894.72 58.9912 1897.68 Q61.7602 1900.6 66.7892 1901.05 L66.7573 1880.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M53.1665 1841.98 L58.7047 1841.98 Q57.4316 1844.46 56.795 1847.13 Q56.1584 1849.81 56.1584 1852.67 Q56.1584 1857.03 57.4952 1859.23 Q58.832 1861.39 61.5056 1861.39 Q63.5426 1861.39 64.7203 1859.83 Q65.8661 1858.27 66.9165 1853.56 L67.3621 1851.56 Q68.6989 1845.32 71.1497 1842.71 Q73.5686 1840.07 77.9291 1840.07 Q82.8944 1840.07 85.7908 1844.01 Q88.6872 1847.93 88.6872 1854.8 Q88.6872 1857.67 88.1143 1860.79 Q87.5732 1863.87 86.4592 1867.31 L80.4118 1867.31 Q82.0987 1864.07 82.958 1860.91 Q83.7856 1857.76 83.7856 1854.68 Q83.7856 1850.54 82.3851 1848.31 Q80.9529 1846.08 78.3747 1846.08 Q75.9876 1846.08 74.7145 1847.71 Q73.4413 1849.3 72.2637 1854.74 L71.7862 1856.78 Q70.6404 1862.22 68.2851 1864.64 Q65.898 1867.06 61.7602 1867.06 Q56.7313 1867.06 53.9941 1863.49 Q51.2568 1859.93 51.2568 1853.37 Q51.2568 1850.12 51.7343 1847.26 Q52.2117 1844.4 53.1665 1841.98 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M41.9947 1824.95 L52.1162 1824.95 L52.1162 1812.89 L56.6677 1812.89 L56.6677 1824.95 L76.0194 1824.95 Q80.3799 1824.95 81.6213 1823.77 Q82.8626 1822.56 82.8626 1818.9 L82.8626 1812.89 L87.7642 1812.89 L87.7642 1818.9 Q87.7642 1825.68 85.2497 1828.26 Q82.7034 1830.84 76.0194 1830.84 L56.6677 1830.84 L56.6677 1835.13 L52.1162 1835.13 L52.1162 1830.84 L41.9947 1830.84 L41.9947 1824.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M58.9593 1777.43 Q55.0126 1775.23 53.1347 1772.18 Q51.2568 1769.12 51.2568 1764.98 Q51.2568 1759.41 55.1717 1756.39 Q59.0548 1753.37 66.2481 1753.37 L87.7642 1753.37 L87.7642 1759.25 L66.439 1759.25 Q61.3146 1759.25 58.832 1761.07 Q56.3494 1762.88 56.3494 1766.61 Q56.3494 1771.16 59.3731 1773.8 Q62.3968 1776.44 67.6167 1776.44 L87.7642 1776.44 L87.7642 1782.33 L66.439 1782.33 Q61.2828 1782.33 58.832 1784.14 Q56.3494 1785.96 56.3494 1789.75 Q56.3494 1794.23 59.4049 1796.88 Q62.4286 1799.52 67.6167 1799.52 L87.7642 1799.52 L87.7642 1805.41 L52.1162 1805.41 L52.1162 1799.52 L57.6544 1799.52 Q54.376 1797.51 52.8164 1794.71 Q51.2568 1791.91 51.2568 1788.06 Q51.2568 1784.18 53.2302 1781.47 Q55.2036 1778.73 58.9593 1777.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M68.4761 1711.19 L71.3406 1711.19 L71.3406 1738.12 Q77.3881 1737.74 80.5709 1734.49 Q83.7219 1731.21 83.7219 1725.39 Q83.7219 1722.01 82.8944 1718.86 Q82.0669 1715.68 80.4118 1712.56 L85.9499 1712.56 Q87.2867 1715.71 87.987 1719.02 Q88.6872 1722.33 88.6872 1725.74 Q88.6872 1734.27 83.7219 1739.27 Q78.7567 1744.23 70.2903 1744.23 Q61.5374 1744.23 56.4131 1739.52 Q51.2568 1734.78 51.2568 1726.76 Q51.2568 1719.56 55.9038 1715.39 Q60.5189 1711.19 68.4761 1711.19 M66.7573 1717.05 Q61.9512 1717.11 59.0866 1719.75 Q56.2221 1722.36 56.2221 1726.69 Q56.2221 1731.6 58.9912 1734.56 Q61.7602 1737.48 66.7892 1737.93 L66.7573 1717.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M66.2481 1671.95 L87.7642 1671.95 L87.7642 1677.8 L66.439 1677.8 Q61.3783 1677.8 58.8638 1679.78 Q56.3494 1681.75 56.3494 1685.7 Q56.3494 1690.44 59.3731 1693.18 Q62.3968 1695.92 67.6167 1695.92 L87.7642 1695.92 L87.7642 1701.8 L52.1162 1701.8 L52.1162 1695.92 L57.6544 1695.92 Q54.4397 1693.81 52.8483 1690.98 Q51.2568 1688.12 51.2568 1684.39 Q51.2568 1678.25 55.0763 1675.1 Q58.8638 1671.95 66.2481 1671.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M41.9947 1654.47 L52.1162 1654.47 L52.1162 1642.41 L56.6677 1642.41 L56.6677 1654.47 L76.0194 1654.47 Q80.3799 1654.47 81.6213 1653.3 Q82.8626 1652.09 82.8626 1648.43 L82.8626 1642.41 L87.7642 1642.41 L87.7642 1648.43 Q87.7642 1655.21 85.2497 1657.78 Q82.7034 1660.36 76.0194 1660.36 L56.6677 1660.36 L56.6677 1664.66 L52.1162 1664.66 L52.1162 1660.36 L41.9947 1660.36 L41.9947 1654.47 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M38.3026 1599.92 Q45.6232 1604.19 52.7846 1606.25 Q59.946 1608.32 67.2984 1608.32 Q74.6508 1608.32 81.8759 1606.25 Q89.0691 1604.15 96.3579 1599.92 L96.3579 1605.01 Q88.8782 1609.79 81.6531 1612.17 Q74.428 1614.53 67.2984 1614.53 Q60.2006 1614.53 53.0074 1612.17 Q45.8142 1609.82 38.3026 1605.01 L38.3026 1599.92 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M97.3445 1572.68 L97.3445 1575.86 L87.7642 1575.89 Q87.7005 1579.23 87.0003 1582.57 Q86.2682 1585.92 84.8359 1589.29 L79.1068 1589.29 Q81.1438 1586.04 82.1942 1582.73 Q83.2127 1579.39 83.2445 1575.86 L68.7307 1575.86 Q67.5849 1582.89 64.8476 1586.11 Q62.1104 1589.29 57.3361 1589.29 Q52.148 1589.29 49.1561 1585.82 Q46.1643 1582.35 45.7187 1575.86 L38.239 1575.86 L38.239 1572.68 L45.6232 1572.68 Q45.7505 1569.72 46.2598 1566.95 Q46.7372 1564.18 47.5965 1561.54 L53.1665 1561.54 Q51.8297 1564.18 51.0977 1566.98 Q50.3656 1569.75 50.2383 1572.68 L63.8291 1572.68 Q64.9431 1565.45 67.8077 1562.04 Q70.6722 1558.64 75.6375 1558.64 Q81.0165 1558.64 84.1357 1562.27 Q87.2231 1565.86 87.7005 1572.68 L97.3445 1572.68 M63.2562 1575.86 L50.2065 1575.86 Q50.6203 1579.55 52.3072 1581.49 Q53.9941 1583.43 56.795 1583.43 Q59.5322 1583.43 61.06 1581.65 Q62.5878 1579.84 63.2562 1575.86 M69.3673 1572.68 L83.149 1572.68 Q82.6079 1568.63 80.8574 1566.6 Q79.1068 1564.53 76.2422 1564.53 Q73.4413 1564.53 71.7862 1566.5 Q70.1312 1568.44 69.3673 1572.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M38.3026 1548.01 L38.3026 1542.92 Q45.8142 1538.14 53.0074 1535.79 Q60.2006 1533.4 67.2984 1533.4 Q74.428 1533.4 81.6531 1535.79 Q88.8782 1538.14 96.3579 1542.92 L96.3579 1548.01 Q89.0691 1543.77 81.8759 1541.71 Q74.6508 1539.61 67.2984 1539.61 Q59.946 1539.61 52.7846 1541.71 Q45.6232 1543.77 38.3026 1548.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1022.52 1228.03 L1022.52 1275.06 L1032.41 1275.06 Q1044.92 1275.06 1050.72 1269.39 Q1056.55 1263.72 1056.55 1251.49 Q1056.55 1239.33 1050.72 1233.7 Q1044.92 1228.03 1032.41 1228.03 L1022.52 1228.03 M1014.34 1221.31 L1031.15 1221.31 Q1048.73 1221.31 1056.95 1228.64 Q1065.18 1235.93 1065.18 1251.49 Q1065.18 1267.12 1056.91 1274.46 Q1048.65 1281.79 1031.15 1281.79 L1014.34 1281.79 L1014.34 1221.31 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1116.7 1257.24 L1116.7 1260.89 L1082.43 1260.89 Q1082.92 1268.58 1087.05 1272.63 Q1091.22 1276.64 1098.64 1276.64 Q1102.93 1276.64 1106.94 1275.59 Q1110.99 1274.54 1114.96 1272.43 L1114.96 1279.48 Q1110.95 1281.18 1106.74 1282.07 Q1102.53 1282.96 1098.19 1282.96 Q1087.34 1282.96 1080.98 1276.64 Q1074.66 1270.32 1074.66 1259.55 Q1074.66 1248.41 1080.65 1241.89 Q1086.69 1235.32 1096.9 1235.32 Q1106.05 1235.32 1111.36 1241.24 Q1116.7 1247.11 1116.7 1257.24 M1109.25 1255.05 Q1109.17 1248.94 1105.81 1245.29 Q1102.49 1241.64 1096.98 1241.64 Q1090.74 1241.64 1086.97 1245.17 Q1083.24 1248.69 1082.68 1255.09 L1109.25 1255.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1161.59 1238.16 L1161.59 1245.13 Q1158.43 1243.39 1155.23 1242.54 Q1152.07 1241.64 1148.83 1241.64 Q1141.58 1241.64 1137.57 1246.26 Q1133.56 1250.84 1133.56 1259.14 Q1133.56 1267.45 1137.57 1272.07 Q1141.58 1276.64 1148.83 1276.64 Q1152.07 1276.64 1155.23 1275.79 Q1158.43 1274.9 1161.59 1273.16 L1161.59 1280.05 Q1158.47 1281.5 1155.11 1282.23 Q1151.79 1282.96 1148.02 1282.96 Q1137.77 1282.96 1131.73 1276.52 Q1125.7 1270.08 1125.7 1259.14 Q1125.7 1248.04 1131.77 1241.68 Q1137.89 1235.32 1148.5 1235.32 Q1151.95 1235.32 1155.23 1236.05 Q1158.51 1236.74 1161.59 1238.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1174.55 1236.42 L1182.01 1236.42 L1182.01 1281.79 L1174.55 1281.79 L1174.55 1236.42 M1174.55 1218.76 L1182.01 1218.76 L1182.01 1228.19 L1174.55 1228.19 L1174.55 1218.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1226.52 1237.76 L1226.52 1244.8 Q1223.37 1243.18 1219.96 1242.37 Q1216.56 1241.56 1212.91 1241.56 Q1207.36 1241.56 1204.57 1243.26 Q1201.81 1244.97 1201.81 1248.37 Q1201.81 1250.96 1203.8 1252.46 Q1205.78 1253.92 1211.78 1255.25 L1214.33 1255.82 Q1222.27 1257.52 1225.59 1260.64 Q1228.96 1263.72 1228.96 1269.27 Q1228.96 1275.59 1223.93 1279.28 Q1218.95 1282.96 1210.2 1282.96 Q1206.55 1282.96 1202.58 1282.23 Q1198.65 1281.55 1194.28 1280.13 L1194.28 1272.43 Q1198.41 1274.58 1202.42 1275.67 Q1206.43 1276.72 1210.36 1276.72 Q1215.63 1276.72 1218.46 1274.94 Q1221.3 1273.12 1221.3 1269.84 Q1221.3 1266.8 1219.23 1265.18 Q1217.21 1263.56 1210.28 1262.06 L1207.69 1261.45 Q1200.76 1259.99 1197.68 1257 Q1194.6 1253.96 1194.6 1248.69 Q1194.6 1242.29 1199.14 1238.81 Q1203.68 1235.32 1212.02 1235.32 Q1216.15 1235.32 1219.8 1235.93 Q1223.45 1236.54 1226.52 1237.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1240.82 1236.42 L1248.28 1236.42 L1248.28 1281.79 L1240.82 1281.79 L1240.82 1236.42 M1240.82 1218.76 L1248.28 1218.76 L1248.28 1228.19 L1240.82 1228.19 L1240.82 1218.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1281.46 1241.64 Q1275.46 1241.64 1271.98 1246.34 Q1268.49 1251 1268.49 1259.14 Q1268.49 1267.29 1271.94 1271.99 Q1275.42 1276.64 1281.46 1276.64 Q1287.41 1276.64 1290.89 1271.94 Q1294.38 1267.25 1294.38 1259.14 Q1294.38 1251.08 1290.89 1246.38 Q1287.41 1241.64 1281.46 1241.64 M1281.46 1235.32 Q1291.18 1235.32 1296.73 1241.64 Q1302.28 1247.96 1302.28 1259.14 Q1302.28 1270.28 1296.73 1276.64 Q1291.18 1282.96 1281.46 1282.96 Q1271.69 1282.96 1266.14 1276.64 Q1260.63 1270.28 1260.63 1259.14 Q1260.63 1247.96 1266.14 1241.64 Q1271.69 1235.32 1281.46 1235.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1352.35 1254.4 L1352.35 1281.79 L1344.89 1281.79 L1344.89 1254.65 Q1344.89 1248.21 1342.38 1245.01 Q1339.87 1241.81 1334.85 1241.81 Q1328.81 1241.81 1325.33 1245.65 Q1321.84 1249.5 1321.84 1256.15 L1321.84 1281.79 L1314.35 1281.79 L1314.35 1236.42 L1321.84 1236.42 L1321.84 1243.47 Q1324.52 1239.38 1328.12 1237.35 Q1331.77 1235.32 1336.51 1235.32 Q1344.33 1235.32 1348.34 1240.19 Q1352.35 1245.01 1352.35 1254.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1422.51 1237.76 L1422.51 1244.8 Q1419.35 1243.18 1415.95 1242.37 Q1412.54 1241.56 1408.9 1241.56 Q1403.35 1241.56 1400.55 1243.26 Q1397.8 1244.97 1397.8 1248.37 Q1397.8 1250.96 1399.78 1252.46 Q1401.77 1253.92 1407.76 1255.25 L1410.31 1255.82 Q1418.25 1257.52 1421.58 1260.64 Q1424.94 1263.72 1424.94 1269.27 Q1424.94 1275.59 1419.91 1279.28 Q1414.93 1282.96 1406.18 1282.96 Q1402.54 1282.96 1398.57 1282.23 Q1394.64 1281.55 1390.26 1280.13 L1390.26 1272.43 Q1394.39 1274.58 1398.4 1275.67 Q1402.42 1276.72 1406.34 1276.72 Q1411.61 1276.72 1414.45 1274.94 Q1417.28 1273.12 1417.28 1269.84 Q1417.28 1266.8 1415.22 1265.18 Q1413.19 1263.56 1406.26 1262.06 L1403.67 1261.45 Q1396.74 1259.99 1393.67 1257 Q1390.59 1253.96 1390.59 1248.69 Q1390.59 1242.29 1395.12 1238.81 Q1399.66 1235.32 1408.01 1235.32 Q1412.14 1235.32 1415.78 1235.93 Q1419.43 1236.54 1422.51 1237.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1444.02 1274.98 L1444.02 1299.05 L1436.52 1299.05 L1436.52 1236.42 L1444.02 1236.42 L1444.02 1243.3 Q1446.37 1239.25 1449.93 1237.31 Q1453.54 1235.32 1458.52 1235.32 Q1466.78 1235.32 1471.93 1241.89 Q1477.11 1248.45 1477.11 1259.14 Q1477.11 1269.84 1471.93 1276.4 Q1466.78 1282.96 1458.52 1282.96 Q1453.54 1282.96 1449.93 1281.02 Q1446.37 1279.03 1444.02 1274.98 M1469.38 1259.14 Q1469.38 1250.92 1465.97 1246.26 Q1462.61 1241.56 1456.7 1241.56 Q1450.78 1241.56 1447.38 1246.26 Q1444.02 1250.92 1444.02 1259.14 Q1444.02 1267.37 1447.38 1272.07 Q1450.78 1276.72 1456.7 1276.72 Q1462.61 1276.72 1465.97 1272.07 Q1469.38 1267.37 1469.38 1259.14 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1510.09 1258.98 Q1501.05 1258.98 1497.57 1261.05 Q1494.09 1263.11 1494.09 1268.1 Q1494.09 1272.07 1496.68 1274.42 Q1499.31 1276.72 1503.81 1276.72 Q1510.01 1276.72 1513.73 1272.35 Q1517.5 1267.93 1517.5 1260.64 L1517.5 1258.98 L1510.09 1258.98 M1524.95 1255.9 L1524.95 1281.79 L1517.5 1281.79 L1517.5 1274.9 Q1514.95 1279.03 1511.14 1281.02 Q1507.33 1282.96 1501.82 1282.96 Q1494.86 1282.96 1490.72 1279.07 Q1486.63 1275.14 1486.63 1268.58 Q1486.63 1260.93 1491.74 1257.04 Q1496.88 1253.15 1507.05 1253.15 L1517.5 1253.15 L1517.5 1252.42 Q1517.5 1247.27 1514.1 1244.48 Q1510.74 1241.64 1504.62 1241.64 Q1500.73 1241.64 1497.04 1242.58 Q1493.36 1243.51 1489.96 1245.37 L1489.96 1238.48 Q1494.05 1236.9 1497.89 1236.13 Q1501.74 1235.32 1505.39 1235.32 Q1515.23 1235.32 1520.09 1240.43 Q1524.95 1245.53 1524.95 1255.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1572.96 1238.16 L1572.96 1245.13 Q1569.8 1243.39 1566.6 1242.54 Q1563.44 1241.64 1560.2 1241.64 Q1552.95 1241.64 1548.94 1246.26 Q1544.93 1250.84 1544.93 1259.14 Q1544.93 1267.45 1548.94 1272.07 Q1552.95 1276.64 1560.2 1276.64 Q1563.44 1276.64 1566.6 1275.79 Q1569.8 1274.9 1572.96 1273.16 L1572.96 1280.05 Q1569.84 1281.5 1566.48 1282.23 Q1563.15 1282.96 1559.39 1282.96 Q1549.14 1282.96 1543.1 1276.52 Q1537.07 1270.08 1537.07 1259.14 Q1537.07 1248.04 1543.14 1241.68 Q1549.26 1235.32 1559.87 1235.32 Q1563.32 1235.32 1566.6 1236.05 Q1569.88 1236.74 1572.96 1238.16 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M1624.73 1257.24 L1624.73 1260.89 L1590.46 1260.89 Q1590.94 1268.58 1595.08 1272.63 Q1599.25 1276.64 1606.66 1276.64 Q1610.96 1276.64 1614.97 1275.59 Q1619.02 1274.54 1622.99 1272.43 L1622.99 1279.48 Q1618.98 1281.18 1614.76 1282.07 Q1610.55 1282.96 1606.22 1282.96 Q1595.36 1282.96 1589 1276.64 Q1582.68 1270.32 1582.68 1259.55 Q1582.68 1248.41 1588.68 1241.89 Q1594.71 1235.32 1604.92 1235.32 Q1614.07 1235.32 1619.38 1241.24 Q1624.73 1247.11 1624.73 1257.24 M1617.28 1255.05 Q1617.19 1248.94 1613.83 1245.29 Q1610.51 1241.64 1605 1241.64 Q1598.76 1241.64 1595 1245.17 Q1591.27 1248.69 1590.7 1255.09 L1617.28 1255.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip913)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
+<path clip-path="url(#clip913)" d="M436.9 1958.75 L436.9 2078.86 L466.444 2078.86 L466.444 1958.75 L436.9 1958.75 L436.9 1958.75  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,1958.75 436.9,2078.86 466.444,2078.86 466.444,1958.75 436.9,1958.75 "/>
+<path clip-path="url(#clip913)" d="M473.83 1869.53 L473.83 2039.4 L503.374 2039.4 L503.374 1869.53 L473.83 1869.53 L473.83 1869.53  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,1869.53 473.83,2039.4 503.374,2039.4 503.374,1869.53 473.83,1869.53 "/>
+<path clip-path="url(#clip913)" d="M510.76 1801.07 L510.76 2009.11 L540.304 2009.11 L540.304 1801.07 L510.76 1801.07 L510.76 1801.07  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,1801.07 510.76,2009.11 540.304,2009.11 540.304,1801.07 510.76,1801.07 "/>
+<path clip-path="url(#clip913)" d="M547.691 1743.36 L547.691 1983.58 L577.235 1983.58 L577.235 1743.36 L547.691 1743.36 L547.691 1743.36  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1743.36 547.691,1983.58 577.235,1983.58 577.235,1743.36 547.691,1743.36 "/>
+<path clip-path="url(#clip913)" d="M584.621 1692.51 L584.621 1961.09 L614.165 1961.09 L614.165 1692.51 L584.621 1692.51 L584.621 1692.51  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1692.51 584.621,1961.09 614.165,1961.09 614.165,1692.51 584.621,1692.51 "/>
+<path clip-path="url(#clip913)" d="M621.551 1646.54 L621.551 1940.76 L651.095 1940.76 L651.095 1646.54 L621.551 1646.54 L621.551 1646.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1646.54 621.551,1940.76 651.095,1940.76 651.095,1646.54 621.551,1646.54 "/>
+<path clip-path="url(#clip913)" d="M658.481 1604.27 L658.481 1922.06 L688.026 1922.06 L688.026 1604.27 L658.481 1604.27 L658.481 1604.27  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1604.27 658.481,1922.06 688.026,1922.06 688.026,1604.27 658.481,1604.27 "/>
+<path clip-path="url(#clip913)" d="M695.412 1564.93 L695.412 1904.65 L724.956 1904.65 L724.956 1564.93 L695.412 1564.93 L695.412 1564.93  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1564.93 695.412,1904.65 724.956,1904.65 724.956,1564.93 695.412,1564.93 "/>
+<path clip-path="url(#clip913)" d="M732.342 1527.97 L732.342 1888.3 L761.886 1888.3 L761.886 1527.97 L732.342 1527.97 L732.342 1527.97  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1527.97 732.342,1888.3 761.886,1888.3 761.886,1527.97 732.342,1527.97 "/>
+<path clip-path="url(#clip913)" d="M769.272 1493.02 L769.272 1872.84 L798.816 1872.84 L798.816 1493.02 L769.272 1493.02 L769.272 1493.02  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1493.02 769.272,1872.84 798.816,1872.84 798.816,1493.02 769.272,1493.02 "/>
+<path clip-path="url(#clip913)" d="M806.203 1459.77 L806.203 1858.14 L835.747 1858.14 L835.747 1459.77 L806.203 1459.77 L806.203 1459.77  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1459.77 806.203,1858.14 835.747,1858.14 835.747,1459.77 806.203,1459.77 "/>
+<path clip-path="url(#clip913)" d="M843.133 1428.01 L843.133 1844.08 L872.677 1844.08 L872.677 1428.01 L843.133 1428.01 L843.133 1428.01  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1428.01 843.133,1844.08 872.677,1844.08 872.677,1428.01 843.133,1428.01 "/>
+<path clip-path="url(#clip913)" d="M880.063 1397.54 L880.063 1830.61 L909.607 1830.61 L909.607 1397.54 L880.063 1397.54 L880.063 1397.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1397.54 880.063,1830.61 909.607,1830.61 909.607,1397.54 880.063,1397.54 "/>
+<path clip-path="url(#clip913)" d="M916.993 1368.23 L916.993 1817.64 L946.538 1817.64 L946.538 1368.23 L916.993 1368.23 L916.993 1368.23  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1368.23 916.993,1817.64 946.538,1817.64 946.538,1368.23 916.993,1368.23 "/>
+<path clip-path="url(#clip913)" d="M953.924 1362.92 L953.924 1793.51 L983.468 1793.51 L983.468 1362.92 L953.924 1362.92 L953.924 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1362.92 953.924,1793.51 983.468,1793.51 983.468,1362.92 953.924,1362.92 "/>
+<path clip-path="url(#clip913)" d="M990.854 1362.92 L990.854 1769.49 L1020.4 1769.49 L1020.4 1362.92 L990.854 1362.92 L990.854 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1362.92 990.854,1769.49 1020.4,1769.49 1020.4,1362.92 990.854,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1027.78 1362.92 L1027.78 1747.67 L1057.33 1747.67 L1057.33 1362.92 L1027.78 1362.92 L1027.78 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1362.92 1027.78,1747.67 1057.33,1747.67 1057.33,1362.92 1027.78,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1064.71 1362.92 L1064.71 1727.55 L1094.26 1727.55 L1094.26 1362.92 L1064.71 1362.92 L1064.71 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1362.92 1064.71,1727.55 1094.26,1727.55 1094.26,1362.92 1064.71,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1101.64 1362.92 L1101.64 1708.77 L1131.19 1708.77 L1131.19 1362.92 L1101.64 1362.92 L1101.64 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1362.92 1101.64,1708.77 1131.19,1708.77 1131.19,1362.92 1101.64,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1138.58 1362.92 L1138.58 1691.1 L1168.12 1691.1 L1168.12 1362.92 L1138.58 1362.92 L1138.58 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1362.92 1138.58,1691.1 1168.12,1691.1 1168.12,1362.92 1138.58,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1175.51 1362.92 L1175.51 1674.37 L1205.05 1674.37 L1205.05 1362.92 L1175.51 1362.92 L1175.51 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1362.92 1175.51,1674.37 1205.05,1674.37 1205.05,1362.92 1175.51,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1212.44 1362.92 L1212.44 1658.44 L1241.98 1658.44 L1241.98 1362.92 L1212.44 1362.92 L1212.44 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1362.92 1212.44,1658.44 1241.98,1658.44 1241.98,1362.92 1212.44,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1249.37 1362.92 L1249.37 1643.2 L1278.91 1643.2 L1278.91 1362.92 L1249.37 1362.92 L1249.37 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1362.92 1249.37,1643.2 1278.91,1643.2 1278.91,1362.92 1249.37,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1286.3 1362.92 L1286.3 1628.58 L1315.84 1628.58 L1315.84 1362.92 L1286.3 1362.92 L1286.3 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1362.92 1286.3,1628.58 1315.84,1628.58 1315.84,1362.92 1286.3,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1323.23 1362.92 L1323.23 1614.49 L1352.77 1614.49 L1352.77 1362.92 L1323.23 1362.92 L1323.23 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1362.92 1323.23,1614.49 1352.77,1614.49 1352.77,1362.92 1323.23,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1360.16 1362.92 L1360.16 1600.9 L1389.7 1600.9 L1389.7 1362.92 L1360.16 1362.92 L1360.16 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1362.92 1360.16,1600.9 1389.7,1600.9 1389.7,1362.92 1360.16,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1397.09 1362.92 L1397.09 1587.74 L1426.63 1587.74 L1426.63 1362.92 L1397.09 1362.92 L1397.09 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1362.92 1397.09,1587.74 1426.63,1587.74 1426.63,1362.92 1397.09,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1434.02 1362.92 L1434.02 1574.98 L1463.56 1574.98 L1463.56 1362.92 L1434.02 1362.92 L1434.02 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1362.92 1434.02,1574.98 1463.56,1574.98 1463.56,1362.92 1434.02,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1470.95 1362.92 L1470.95 1562.59 L1500.49 1562.59 L1500.49 1362.92 L1470.95 1362.92 L1470.95 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1362.92 1470.95,1562.59 1500.49,1562.59 1500.49,1362.92 1470.95,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1507.88 1362.92 L1507.88 1550.53 L1537.42 1550.53 L1537.42 1362.92 L1507.88 1362.92 L1507.88 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1362.92 1507.88,1550.53 1537.42,1550.53 1537.42,1362.92 1507.88,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1544.81 1362.92 L1544.81 1538.79 L1574.35 1538.79 L1574.35 1362.92 L1544.81 1362.92 L1544.81 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1362.92 1544.81,1538.79 1574.35,1538.79 1574.35,1362.92 1544.81,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1581.74 1362.92 L1581.74 1527.33 L1611.28 1527.33 L1611.28 1362.92 L1581.74 1362.92 L1581.74 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1362.92 1581.74,1527.33 1611.28,1527.33 1611.28,1362.92 1581.74,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1618.67 1362.92 L1618.67 1516.13 L1648.21 1516.13 L1648.21 1362.92 L1618.67 1362.92 L1618.67 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1362.92 1618.67,1516.13 1648.21,1516.13 1648.21,1362.92 1618.67,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1655.6 1362.92 L1655.6 1505.19 L1685.14 1505.19 L1685.14 1362.92 L1655.6 1362.92 L1655.6 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1362.92 1655.6,1505.19 1685.14,1505.19 1685.14,1362.92 1655.6,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1692.53 1362.92 L1692.53 1494.48 L1722.07 1494.48 L1722.07 1362.92 L1692.53 1362.92 L1692.53 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1362.92 1692.53,1494.48 1722.07,1494.48 1722.07,1362.92 1692.53,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1729.46 1362.92 L1729.46 1483.98 L1759 1483.98 L1759 1362.92 L1729.46 1362.92 L1729.46 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1362.92 1729.46,1483.98 1759,1483.98 1759,1362.92 1729.46,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1766.39 1362.92 L1766.39 1473.7 L1795.93 1473.7 L1795.93 1362.92 L1766.39 1362.92 L1766.39 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1362.92 1766.39,1473.7 1795.93,1473.7 1795.93,1362.92 1766.39,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1803.32 1362.92 L1803.32 1463.6 L1832.86 1463.6 L1832.86 1362.92 L1803.32 1362.92 L1803.32 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1362.92 1803.32,1463.6 1832.86,1463.6 1832.86,1362.92 1803.32,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1840.25 1362.92 L1840.25 1453.7 L1869.79 1453.7 L1869.79 1362.92 L1840.25 1362.92 L1840.25 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1362.92 1840.25,1453.7 1869.79,1453.7 1869.79,1362.92 1840.25,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1877.18 1362.92 L1877.18 1443.96 L1906.73 1443.96 L1906.73 1362.92 L1877.18 1362.92 L1877.18 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1362.92 1877.18,1443.96 1906.73,1443.96 1906.73,1362.92 1877.18,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1914.11 1362.92 L1914.11 1434.39 L1943.66 1434.39 L1943.66 1362.92 L1914.11 1362.92 L1914.11 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1362.92 1914.11,1434.39 1943.66,1434.39 1943.66,1362.92 1914.11,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1951.04 1362.92 L1951.04 1424.98 L1980.59 1424.98 L1980.59 1362.92 L1951.04 1362.92 L1951.04 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1362.92 1951.04,1424.98 1980.59,1424.98 1980.59,1362.92 1951.04,1362.92 "/>
+<path clip-path="url(#clip913)" d="M1987.97 1362.92 L1987.97 1415.72 L2017.52 1415.72 L2017.52 1362.92 L1987.97 1362.92 L1987.97 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1362.92 1987.97,1415.72 2017.52,1415.72 2017.52,1362.92 1987.97,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2024.9 1362.92 L2024.9 1406.6 L2054.45 1406.6 L2054.45 1362.92 L2024.9 1362.92 L2024.9 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1362.92 2024.9,1406.6 2054.45,1406.6 2054.45,1362.92 2024.9,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2061.83 1362.92 L2061.83 1397.61 L2091.38 1397.61 L2091.38 1362.92 L2061.83 1362.92 L2061.83 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1362.92 2061.83,1397.61 2091.38,1397.61 2091.38,1362.92 2061.83,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2098.76 1362.92 L2098.76 1388.76 L2128.31 1388.76 L2128.31 1362.92 L2098.76 1362.92 L2098.76 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1362.92 2098.76,1388.76 2128.31,1388.76 2128.31,1362.92 2098.76,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2135.69 1362.92 L2135.69 1380.03 L2165.24 1380.03 L2165.24 1362.92 L2135.69 1362.92 L2135.69 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1362.92 2135.69,1380.03 2165.24,1380.03 2165.24,1362.92 2135.69,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2172.62 1362.92 L2172.62 1371.42 L2202.17 1371.42 L2202.17 1362.92 L2172.62 1362.92 L2172.62 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1362.92 2172.62,1371.42 2202.17,1371.42 2202.17,1362.92 2172.62,1362.92 "/>
+<path clip-path="url(#clip913)" d="M2209.55 1362.92 L2209.55 1362.92 L2239.1 1362.92 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,1362.92 2239.1,1362.92 2239.1,1362.92 2209.55,1362.92 "/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="451.672" cy="1958.75" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="488.602" cy="1869.53" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="525.532" cy="1801.07" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="562.463" cy="1743.36" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="599.393" cy="1692.51" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="636.323" cy="1646.54" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="673.253" cy="1604.27" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="710.184" cy="1564.93" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="747.114" cy="1527.97" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="784.044" cy="1493.02" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="820.975" cy="1459.77" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="857.905" cy="1428.01" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="894.835" cy="1397.54" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="931.766" cy="1368.23" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="968.696" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1005.63" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1042.56" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1079.49" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1116.42" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1153.35" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1190.28" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1227.21" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1264.14" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1301.07" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1338" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1374.93" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1411.86" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1448.79" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1485.72" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1522.65" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1559.58" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1596.51" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1633.44" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1670.37" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1707.3" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1744.23" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1781.16" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1818.09" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1855.02" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1891.95" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1928.88" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="1965.81" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2002.74" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2039.67" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2076.6" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2113.53" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2150.47" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2187.4" cy="1362.92" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#009af9; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
+<path clip-path="url(#clip913)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
+<path clip-path="url(#clip913)" d="M436.9 2078.86 L436.9 2078.86 L466.444 2078.86 L466.444 2078.86 L436.9 2078.86 L436.9 2078.86  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,2078.86 436.9,2078.86 466.444,2078.86 466.444,2078.86 436.9,2078.86 "/>
+<path clip-path="url(#clip913)" d="M473.83 2039.4 L473.83 2039.4 L503.374 2039.4 L503.374 2039.4 L473.83 2039.4 L473.83 2039.4  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,2039.4 473.83,2039.4 503.374,2039.4 503.374,2039.4 473.83,2039.4 "/>
+<path clip-path="url(#clip913)" d="M510.76 2009.11 L510.76 2009.11 L540.304 2009.11 L540.304 2009.11 L510.76 2009.11 L510.76 2009.11  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,2009.11 510.76,2009.11 540.304,2009.11 540.304,2009.11 510.76,2009.11 "/>
+<path clip-path="url(#clip913)" d="M547.691 1983.58 L547.691 1983.58 L577.235 1983.58 L577.235 1983.58 L547.691 1983.58 L547.691 1983.58  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1983.58 547.691,1983.58 577.235,1983.58 577.235,1983.58 547.691,1983.58 "/>
+<path clip-path="url(#clip913)" d="M584.621 1961.09 L584.621 1961.09 L614.165 1961.09 L614.165 1961.09 L584.621 1961.09 L584.621 1961.09  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1961.09 584.621,1961.09 614.165,1961.09 614.165,1961.09 584.621,1961.09 "/>
+<path clip-path="url(#clip913)" d="M621.551 1940.76 L621.551 1940.76 L651.095 1940.76 L651.095 1940.76 L621.551 1940.76 L621.551 1940.76  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1940.76 621.551,1940.76 651.095,1940.76 651.095,1940.76 621.551,1940.76 "/>
+<path clip-path="url(#clip913)" d="M658.481 1922.06 L658.481 1922.06 L688.026 1922.06 L688.026 1922.06 L658.481 1922.06 L658.481 1922.06  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1922.06 658.481,1922.06 688.026,1922.06 688.026,1922.06 658.481,1922.06 "/>
+<path clip-path="url(#clip913)" d="M695.412 1904.65 L695.412 1904.65 L724.956 1904.65 L724.956 1904.65 L695.412 1904.65 L695.412 1904.65  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1904.65 695.412,1904.65 724.956,1904.65 724.956,1904.65 695.412,1904.65 "/>
+<path clip-path="url(#clip913)" d="M732.342 1888.3 L732.342 1888.3 L761.886 1888.3 L761.886 1888.3 L732.342 1888.3 L732.342 1888.3  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1888.3 732.342,1888.3 761.886,1888.3 761.886,1888.3 732.342,1888.3 "/>
+<path clip-path="url(#clip913)" d="M769.272 1872.84 L769.272 1872.84 L798.816 1872.84 L798.816 1872.84 L769.272 1872.84 L769.272 1872.84  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1872.84 769.272,1872.84 798.816,1872.84 798.816,1872.84 769.272,1872.84 "/>
+<path clip-path="url(#clip913)" d="M806.203 1858.14 L806.203 1858.14 L835.747 1858.14 L835.747 1858.14 L806.203 1858.14 L806.203 1858.14  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1858.14 806.203,1858.14 835.747,1858.14 835.747,1858.14 806.203,1858.14 "/>
+<path clip-path="url(#clip913)" d="M843.133 1844.08 L843.133 1844.08 L872.677 1844.08 L872.677 1844.08 L843.133 1844.08 L843.133 1844.08  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1844.08 843.133,1844.08 872.677,1844.08 872.677,1844.08 843.133,1844.08 "/>
+<path clip-path="url(#clip913)" d="M880.063 1830.61 L880.063 1830.61 L909.607 1830.61 L909.607 1830.61 L880.063 1830.61 L880.063 1830.61  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1830.61 880.063,1830.61 909.607,1830.61 909.607,1830.61 880.063,1830.61 "/>
+<path clip-path="url(#clip913)" d="M916.993 1817.64 L916.993 1817.64 L946.538 1817.64 L946.538 1817.64 L916.993 1817.64 L916.993 1817.64  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1817.64 916.993,1817.64 946.538,1817.64 946.538,1817.64 916.993,1817.64 "/>
+<path clip-path="url(#clip913)" d="M953.924 1793.51 L953.924 1793.51 L983.468 1793.51 L983.468 1793.51 L953.924 1793.51 L953.924 1793.51  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1793.51 953.924,1793.51 983.468,1793.51 983.468,1793.51 953.924,1793.51 "/>
+<path clip-path="url(#clip913)" d="M990.854 1769.49 L990.854 1769.49 L1020.4 1769.49 L1020.4 1769.49 L990.854 1769.49 L990.854 1769.49  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1769.49 990.854,1769.49 1020.4,1769.49 1020.4,1769.49 990.854,1769.49 "/>
+<path clip-path="url(#clip913)" d="M1027.78 1747.67 L1027.78 1747.67 L1057.33 1747.67 L1057.33 1747.67 L1027.78 1747.67 L1027.78 1747.67  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1747.67 1027.78,1747.67 1057.33,1747.67 1057.33,1747.67 1027.78,1747.67 "/>
+<path clip-path="url(#clip913)" d="M1064.71 1727.55 L1064.71 1727.55 L1094.26 1727.55 L1094.26 1727.55 L1064.71 1727.55 L1064.71 1727.55  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1727.55 1064.71,1727.55 1094.26,1727.55 1094.26,1727.55 1064.71,1727.55 "/>
+<path clip-path="url(#clip913)" d="M1101.64 1708.77 L1101.64 1708.77 L1131.19 1708.77 L1131.19 1708.77 L1101.64 1708.77 L1101.64 1708.77  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1708.77 1101.64,1708.77 1131.19,1708.77 1131.19,1708.77 1101.64,1708.77 "/>
+<path clip-path="url(#clip913)" d="M1138.58 1691.1 L1138.58 1691.1 L1168.12 1691.1 L1168.12 1691.1 L1138.58 1691.1 L1138.58 1691.1  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1691.1 1138.58,1691.1 1168.12,1691.1 1168.12,1691.1 1138.58,1691.1 "/>
+<path clip-path="url(#clip913)" d="M1175.51 1674.37 L1175.51 1674.37 L1205.05 1674.37 L1205.05 1674.37 L1175.51 1674.37 L1175.51 1674.37  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1674.37 1175.51,1674.37 1205.05,1674.37 1205.05,1674.37 1175.51,1674.37 "/>
+<path clip-path="url(#clip913)" d="M1212.44 1658.44 L1212.44 1658.44 L1241.98 1658.44 L1241.98 1658.44 L1212.44 1658.44 L1212.44 1658.44  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1658.44 1212.44,1658.44 1241.98,1658.44 1241.98,1658.44 1212.44,1658.44 "/>
+<path clip-path="url(#clip913)" d="M1249.37 1643.2 L1249.37 1643.2 L1278.91 1643.2 L1278.91 1643.2 L1249.37 1643.2 L1249.37 1643.2  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1643.2 1249.37,1643.2 1278.91,1643.2 1278.91,1643.2 1249.37,1643.2 "/>
+<path clip-path="url(#clip913)" d="M1286.3 1628.58 L1286.3 1628.58 L1315.84 1628.58 L1315.84 1628.58 L1286.3 1628.58 L1286.3 1628.58  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1628.58 1286.3,1628.58 1315.84,1628.58 1315.84,1628.58 1286.3,1628.58 "/>
+<path clip-path="url(#clip913)" d="M1323.23 1614.49 L1323.23 1614.49 L1352.77 1614.49 L1352.77 1614.49 L1323.23 1614.49 L1323.23 1614.49  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1614.49 1323.23,1614.49 1352.77,1614.49 1352.77,1614.49 1323.23,1614.49 "/>
+<path clip-path="url(#clip913)" d="M1360.16 1600.9 L1360.16 1600.9 L1389.7 1600.9 L1389.7 1600.9 L1360.16 1600.9 L1360.16 1600.9  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1600.9 1360.16,1600.9 1389.7,1600.9 1389.7,1600.9 1360.16,1600.9 "/>
+<path clip-path="url(#clip913)" d="M1397.09 1587.74 L1397.09 1587.74 L1426.63 1587.74 L1426.63 1587.74 L1397.09 1587.74 L1397.09 1587.74  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1587.74 1397.09,1587.74 1426.63,1587.74 1426.63,1587.74 1397.09,1587.74 "/>
+<path clip-path="url(#clip913)" d="M1434.02 1574.98 L1434.02 1574.98 L1463.56 1574.98 L1463.56 1574.98 L1434.02 1574.98 L1434.02 1574.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1574.98 1434.02,1574.98 1463.56,1574.98 1463.56,1574.98 1434.02,1574.98 "/>
+<path clip-path="url(#clip913)" d="M1470.95 1562.59 L1470.95 1562.59 L1500.49 1562.59 L1500.49 1562.59 L1470.95 1562.59 L1470.95 1562.59  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1562.59 1470.95,1562.59 1500.49,1562.59 1500.49,1562.59 1470.95,1562.59 "/>
+<path clip-path="url(#clip913)" d="M1507.88 1550.53 L1507.88 1550.53 L1537.42 1550.53 L1537.42 1550.53 L1507.88 1550.53 L1507.88 1550.53  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1550.53 1507.88,1550.53 1537.42,1550.53 1537.42,1550.53 1507.88,1550.53 "/>
+<path clip-path="url(#clip913)" d="M1544.81 1538.79 L1544.81 1538.79 L1574.35 1538.79 L1574.35 1538.79 L1544.81 1538.79 L1544.81 1538.79  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1538.79 1544.81,1538.79 1574.35,1538.79 1574.35,1538.79 1544.81,1538.79 "/>
+<path clip-path="url(#clip913)" d="M1581.74 1527.33 L1581.74 1527.33 L1611.28 1527.33 L1611.28 1527.33 L1581.74 1527.33 L1581.74 1527.33  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1527.33 1581.74,1527.33 1611.28,1527.33 1611.28,1527.33 1581.74,1527.33 "/>
+<path clip-path="url(#clip913)" d="M1618.67 1516.13 L1618.67 1516.13 L1648.21 1516.13 L1648.21 1516.13 L1618.67 1516.13 L1618.67 1516.13  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1516.13 1618.67,1516.13 1648.21,1516.13 1648.21,1516.13 1618.67,1516.13 "/>
+<path clip-path="url(#clip913)" d="M1655.6 1505.19 L1655.6 1505.19 L1685.14 1505.19 L1685.14 1505.19 L1655.6 1505.19 L1655.6 1505.19  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1505.19 1655.6,1505.19 1685.14,1505.19 1685.14,1505.19 1655.6,1505.19 "/>
+<path clip-path="url(#clip913)" d="M1692.53 1494.48 L1692.53 1494.48 L1722.07 1494.48 L1722.07 1494.48 L1692.53 1494.48 L1692.53 1494.48  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1494.48 1692.53,1494.48 1722.07,1494.48 1722.07,1494.48 1692.53,1494.48 "/>
+<path clip-path="url(#clip913)" d="M1729.46 1483.98 L1729.46 1483.98 L1759 1483.98 L1759 1483.98 L1729.46 1483.98 L1729.46 1483.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1483.98 1729.46,1483.98 1759,1483.98 1759,1483.98 1729.46,1483.98 "/>
+<path clip-path="url(#clip913)" d="M1766.39 1473.7 L1766.39 1473.7 L1795.93 1473.7 L1795.93 1473.7 L1766.39 1473.7 L1766.39 1473.7  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1473.7 1766.39,1473.7 1795.93,1473.7 1795.93,1473.7 1766.39,1473.7 "/>
+<path clip-path="url(#clip913)" d="M1803.32 1463.6 L1803.32 1463.6 L1832.86 1463.6 L1832.86 1463.6 L1803.32 1463.6 L1803.32 1463.6  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1463.6 1803.32,1463.6 1832.86,1463.6 1832.86,1463.6 1803.32,1463.6 "/>
+<path clip-path="url(#clip913)" d="M1840.25 1453.7 L1840.25 1453.7 L1869.79 1453.7 L1869.79 1453.7 L1840.25 1453.7 L1840.25 1453.7  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1453.7 1840.25,1453.7 1869.79,1453.7 1869.79,1453.7 1840.25,1453.7 "/>
+<path clip-path="url(#clip913)" d="M1877.18 1443.96 L1877.18 1443.96 L1906.73 1443.96 L1906.73 1443.96 L1877.18 1443.96 L1877.18 1443.96  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1443.96 1877.18,1443.96 1906.73,1443.96 1906.73,1443.96 1877.18,1443.96 "/>
+<path clip-path="url(#clip913)" d="M1914.11 1434.39 L1914.11 1434.39 L1943.66 1434.39 L1943.66 1434.39 L1914.11 1434.39 L1914.11 1434.39  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1434.39 1914.11,1434.39 1943.66,1434.39 1943.66,1434.39 1914.11,1434.39 "/>
+<path clip-path="url(#clip913)" d="M1951.04 1424.98 L1951.04 1424.98 L1980.59 1424.98 L1980.59 1424.98 L1951.04 1424.98 L1951.04 1424.98  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1424.98 1951.04,1424.98 1980.59,1424.98 1980.59,1424.98 1951.04,1424.98 "/>
+<path clip-path="url(#clip913)" d="M1987.97 1415.72 L1987.97 1415.72 L2017.52 1415.72 L2017.52 1415.72 L1987.97 1415.72 L1987.97 1415.72  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1415.72 1987.97,1415.72 2017.52,1415.72 2017.52,1415.72 1987.97,1415.72 "/>
+<path clip-path="url(#clip913)" d="M2024.9 1406.6 L2024.9 1406.6 L2054.45 1406.6 L2054.45 1406.6 L2024.9 1406.6 L2024.9 1406.6  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1406.6 2024.9,1406.6 2054.45,1406.6 2054.45,1406.6 2024.9,1406.6 "/>
+<path clip-path="url(#clip913)" d="M2061.83 1397.61 L2061.83 1397.61 L2091.38 1397.61 L2091.38 1397.61 L2061.83 1397.61 L2061.83 1397.61  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1397.61 2061.83,1397.61 2091.38,1397.61 2091.38,1397.61 2061.83,1397.61 "/>
+<path clip-path="url(#clip913)" d="M2098.76 1388.76 L2098.76 1388.76 L2128.31 1388.76 L2128.31 1388.76 L2098.76 1388.76 L2098.76 1388.76  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1388.76 2098.76,1388.76 2128.31,1388.76 2128.31,1388.76 2098.76,1388.76 "/>
+<path clip-path="url(#clip913)" d="M2135.69 1380.03 L2135.69 1380.03 L2165.24 1380.03 L2165.24 1380.03 L2135.69 1380.03 L2135.69 1380.03  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1380.03 2135.69,1380.03 2165.24,1380.03 2165.24,1380.03 2135.69,1380.03 "/>
+<path clip-path="url(#clip913)" d="M2172.62 1371.42 L2172.62 1371.42 L2202.17 1371.42 L2202.17 1371.42 L2172.62 1371.42 L2172.62 1371.42  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1371.42 2172.62,1371.42 2202.17,1371.42 2202.17,1371.42 2172.62,1371.42 "/>
+<path clip-path="url(#clip913)" d="M2209.55 1362.92 L2209.55 1362.92 L2239.1 1362.92 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,1362.92 2239.1,1362.92 2239.1,1362.92 2209.55,1362.92 "/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="451.672" cy="2078.86" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="488.602" cy="2039.4" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="525.532" cy="2009.11" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="562.463" cy="1983.58" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="599.393" cy="1961.09" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="636.323" cy="1940.76" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="673.253" cy="1922.06" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="710.184" cy="1904.65" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="747.114" cy="1888.3" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="784.044" cy="1872.84" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="820.975" cy="1858.14" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="857.905" cy="1844.08" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="894.835" cy="1830.61" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="931.766" cy="1817.64" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="968.696" cy="1793.51" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1005.63" cy="1769.49" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1042.56" cy="1747.67" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1079.49" cy="1727.55" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1116.42" cy="1708.77" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1153.35" cy="1691.1" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1190.28" cy="1674.37" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1227.21" cy="1658.44" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1264.14" cy="1643.2" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1301.07" cy="1628.58" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1338" cy="1614.49" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1374.93" cy="1600.9" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1411.86" cy="1587.74" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1448.79" cy="1574.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1485.72" cy="1562.59" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1522.65" cy="1550.53" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1559.58" cy="1538.79" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1596.51" cy="1527.33" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1633.44" cy="1516.13" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1670.37" cy="1505.19" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1707.3" cy="1494.48" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1744.23" cy="1483.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1781.16" cy="1473.7" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1818.09" cy="1463.6" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1855.02" cy="1453.7" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1891.95" cy="1443.96" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1928.88" cy="1434.39" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="1965.81" cy="1424.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2002.74" cy="1415.72" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2039.67" cy="1406.6" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2076.6" cy="1397.61" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2113.53" cy="1388.76" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2150.47" cy="1380.03" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2187.4" cy="1371.42" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#e26f46; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
+<path clip-path="url(#clip913)" d="M399.969 2174.14 L399.969 2174.14 L429.514 2174.14 L429.514 2174.14 L399.969 2174.14 L399.969 2174.14  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="399.969,2174.14 399.969,2174.14 429.514,2174.14 429.514,2174.14 399.969,2174.14 "/>
+<path clip-path="url(#clip913)" d="M436.9 2078.86 L436.9 2174.14 L466.444 2174.14 L466.444 2078.86 L436.9 2078.86 L436.9 2078.86  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="436.9,2078.86 436.9,2174.14 466.444,2174.14 466.444,2078.86 436.9,2078.86 "/>
+<path clip-path="url(#clip913)" d="M473.83 2039.4 L473.83 2174.14 L503.374 2174.14 L503.374 2039.4 L473.83 2039.4 L473.83 2039.4  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.83,2039.4 473.83,2174.14 503.374,2174.14 503.374,2039.4 473.83,2039.4 "/>
+<path clip-path="url(#clip913)" d="M510.76 2009.11 L510.76 2174.14 L540.304 2174.14 L540.304 2009.11 L510.76 2009.11 L510.76 2009.11  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="510.76,2009.11 510.76,2174.14 540.304,2174.14 540.304,2009.11 510.76,2009.11 "/>
+<path clip-path="url(#clip913)" d="M547.691 1983.58 L547.691 2174.14 L577.235 2174.14 L577.235 1983.58 L547.691 1983.58 L547.691 1983.58  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="547.691,1983.58 547.691,2174.14 577.235,2174.14 577.235,1983.58 547.691,1983.58 "/>
+<path clip-path="url(#clip913)" d="M584.621 1961.09 L584.621 2174.14 L614.165 2174.14 L614.165 1961.09 L584.621 1961.09 L584.621 1961.09  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="584.621,1961.09 584.621,2174.14 614.165,2174.14 614.165,1961.09 584.621,1961.09 "/>
+<path clip-path="url(#clip913)" d="M621.551 1940.76 L621.551 2174.14 L651.095 2174.14 L651.095 1940.76 L621.551 1940.76 L621.551 1940.76  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="621.551,1940.76 621.551,2174.14 651.095,2174.14 651.095,1940.76 621.551,1940.76 "/>
+<path clip-path="url(#clip913)" d="M658.481 1922.06 L658.481 2174.14 L688.026 2174.14 L688.026 1922.06 L658.481 1922.06 L658.481 1922.06  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="658.481,1922.06 658.481,2174.14 688.026,2174.14 688.026,1922.06 658.481,1922.06 "/>
+<path clip-path="url(#clip913)" d="M695.412 1904.65 L695.412 2174.14 L724.956 2174.14 L724.956 1904.65 L695.412 1904.65 L695.412 1904.65  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="695.412,1904.65 695.412,2174.14 724.956,2174.14 724.956,1904.65 695.412,1904.65 "/>
+<path clip-path="url(#clip913)" d="M732.342 1888.3 L732.342 2174.14 L761.886 2174.14 L761.886 1888.3 L732.342 1888.3 L732.342 1888.3  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="732.342,1888.3 732.342,2174.14 761.886,2174.14 761.886,1888.3 732.342,1888.3 "/>
+<path clip-path="url(#clip913)" d="M769.272 1872.84 L769.272 2174.14 L798.816 2174.14 L798.816 1872.84 L769.272 1872.84 L769.272 1872.84  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="769.272,1872.84 769.272,2174.14 798.816,2174.14 798.816,1872.84 769.272,1872.84 "/>
+<path clip-path="url(#clip913)" d="M806.203 1858.14 L806.203 2174.14 L835.747 2174.14 L835.747 1858.14 L806.203 1858.14 L806.203 1858.14  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="806.203,1858.14 806.203,2174.14 835.747,2174.14 835.747,1858.14 806.203,1858.14 "/>
+<path clip-path="url(#clip913)" d="M843.133 1844.08 L843.133 2174.14 L872.677 2174.14 L872.677 1844.08 L843.133 1844.08 L843.133 1844.08  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="843.133,1844.08 843.133,2174.14 872.677,2174.14 872.677,1844.08 843.133,1844.08 "/>
+<path clip-path="url(#clip913)" d="M880.063 1830.61 L880.063 2174.14 L909.607 2174.14 L909.607 1830.61 L880.063 1830.61 L880.063 1830.61  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="880.063,1830.61 880.063,2174.14 909.607,2174.14 909.607,1830.61 880.063,1830.61 "/>
+<path clip-path="url(#clip913)" d="M916.993 1817.64 L916.993 2174.14 L946.538 2174.14 L946.538 1817.64 L916.993 1817.64 L916.993 1817.64  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="916.993,1817.64 916.993,2174.14 946.538,2174.14 946.538,1817.64 916.993,1817.64 "/>
+<path clip-path="url(#clip913)" d="M953.924 1793.51 L953.924 2174.14 L983.468 2174.14 L983.468 1793.51 L953.924 1793.51 L953.924 1793.51  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="953.924,1793.51 953.924,2174.14 983.468,2174.14 983.468,1793.51 953.924,1793.51 "/>
+<path clip-path="url(#clip913)" d="M990.854 1769.49 L990.854 2174.14 L1020.4 2174.14 L1020.4 1769.49 L990.854 1769.49 L990.854 1769.49  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="990.854,1769.49 990.854,2174.14 1020.4,2174.14 1020.4,1769.49 990.854,1769.49 "/>
+<path clip-path="url(#clip913)" d="M1027.78 1747.67 L1027.78 2174.14 L1057.33 2174.14 L1057.33 1747.67 L1027.78 1747.67 L1027.78 1747.67  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1027.78,1747.67 1027.78,2174.14 1057.33,2174.14 1057.33,1747.67 1027.78,1747.67 "/>
+<path clip-path="url(#clip913)" d="M1064.71 1727.55 L1064.71 2174.14 L1094.26 2174.14 L1094.26 1727.55 L1064.71 1727.55 L1064.71 1727.55  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1064.71,1727.55 1064.71,2174.14 1094.26,2174.14 1094.26,1727.55 1064.71,1727.55 "/>
+<path clip-path="url(#clip913)" d="M1101.64 1708.77 L1101.64 2174.14 L1131.19 2174.14 L1131.19 1708.77 L1101.64 1708.77 L1101.64 1708.77  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1101.64,1708.77 1101.64,2174.14 1131.19,2174.14 1131.19,1708.77 1101.64,1708.77 "/>
+<path clip-path="url(#clip913)" d="M1138.58 1691.1 L1138.58 2174.14 L1168.12 2174.14 L1168.12 1691.1 L1138.58 1691.1 L1138.58 1691.1  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1138.58,1691.1 1138.58,2174.14 1168.12,2174.14 1168.12,1691.1 1138.58,1691.1 "/>
+<path clip-path="url(#clip913)" d="M1175.51 1674.37 L1175.51 2174.14 L1205.05 2174.14 L1205.05 1674.37 L1175.51 1674.37 L1175.51 1674.37  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1175.51,1674.37 1175.51,2174.14 1205.05,2174.14 1205.05,1674.37 1175.51,1674.37 "/>
+<path clip-path="url(#clip913)" d="M1212.44 1658.44 L1212.44 2174.14 L1241.98 2174.14 L1241.98 1658.44 L1212.44 1658.44 L1212.44 1658.44  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1212.44,1658.44 1212.44,2174.14 1241.98,2174.14 1241.98,1658.44 1212.44,1658.44 "/>
+<path clip-path="url(#clip913)" d="M1249.37 1643.2 L1249.37 2174.14 L1278.91 2174.14 L1278.91 1643.2 L1249.37 1643.2 L1249.37 1643.2  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1249.37,1643.2 1249.37,2174.14 1278.91,2174.14 1278.91,1643.2 1249.37,1643.2 "/>
+<path clip-path="url(#clip913)" d="M1286.3 1628.58 L1286.3 2174.14 L1315.84 2174.14 L1315.84 1628.58 L1286.3 1628.58 L1286.3 1628.58  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1286.3,1628.58 1286.3,2174.14 1315.84,2174.14 1315.84,1628.58 1286.3,1628.58 "/>
+<path clip-path="url(#clip913)" d="M1323.23 1614.49 L1323.23 2174.14 L1352.77 2174.14 L1352.77 1614.49 L1323.23 1614.49 L1323.23 1614.49  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1323.23,1614.49 1323.23,2174.14 1352.77,2174.14 1352.77,1614.49 1323.23,1614.49 "/>
+<path clip-path="url(#clip913)" d="M1360.16 1600.9 L1360.16 2174.14 L1389.7 2174.14 L1389.7 1600.9 L1360.16 1600.9 L1360.16 1600.9  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1360.16,1600.9 1360.16,2174.14 1389.7,2174.14 1389.7,1600.9 1360.16,1600.9 "/>
+<path clip-path="url(#clip913)" d="M1397.09 1587.74 L1397.09 2174.14 L1426.63 2174.14 L1426.63 1587.74 L1397.09 1587.74 L1397.09 1587.74  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1397.09,1587.74 1397.09,2174.14 1426.63,2174.14 1426.63,1587.74 1397.09,1587.74 "/>
+<path clip-path="url(#clip913)" d="M1434.02 1574.98 L1434.02 2174.14 L1463.56 2174.14 L1463.56 1574.98 L1434.02 1574.98 L1434.02 1574.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1434.02,1574.98 1434.02,2174.14 1463.56,2174.14 1463.56,1574.98 1434.02,1574.98 "/>
+<path clip-path="url(#clip913)" d="M1470.95 1562.59 L1470.95 2174.14 L1500.49 2174.14 L1500.49 1562.59 L1470.95 1562.59 L1470.95 1562.59  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1470.95,1562.59 1470.95,2174.14 1500.49,2174.14 1500.49,1562.59 1470.95,1562.59 "/>
+<path clip-path="url(#clip913)" d="M1507.88 1550.53 L1507.88 2174.14 L1537.42 2174.14 L1537.42 1550.53 L1507.88 1550.53 L1507.88 1550.53  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1507.88,1550.53 1507.88,2174.14 1537.42,2174.14 1537.42,1550.53 1507.88,1550.53 "/>
+<path clip-path="url(#clip913)" d="M1544.81 1538.79 L1544.81 2174.14 L1574.35 2174.14 L1574.35 1538.79 L1544.81 1538.79 L1544.81 1538.79  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1544.81,1538.79 1544.81,2174.14 1574.35,2174.14 1574.35,1538.79 1544.81,1538.79 "/>
+<path clip-path="url(#clip913)" d="M1581.74 1527.33 L1581.74 2174.14 L1611.28 2174.14 L1611.28 1527.33 L1581.74 1527.33 L1581.74 1527.33  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1581.74,1527.33 1581.74,2174.14 1611.28,2174.14 1611.28,1527.33 1581.74,1527.33 "/>
+<path clip-path="url(#clip913)" d="M1618.67 1516.13 L1618.67 2174.14 L1648.21 2174.14 L1648.21 1516.13 L1618.67 1516.13 L1618.67 1516.13  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1618.67,1516.13 1618.67,2174.14 1648.21,2174.14 1648.21,1516.13 1618.67,1516.13 "/>
+<path clip-path="url(#clip913)" d="M1655.6 1505.19 L1655.6 2174.14 L1685.14 2174.14 L1685.14 1505.19 L1655.6 1505.19 L1655.6 1505.19  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1655.6,1505.19 1655.6,2174.14 1685.14,2174.14 1685.14,1505.19 1655.6,1505.19 "/>
+<path clip-path="url(#clip913)" d="M1692.53 1494.48 L1692.53 2174.14 L1722.07 2174.14 L1722.07 1494.48 L1692.53 1494.48 L1692.53 1494.48  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1692.53,1494.48 1692.53,2174.14 1722.07,2174.14 1722.07,1494.48 1692.53,1494.48 "/>
+<path clip-path="url(#clip913)" d="M1729.46 1483.98 L1729.46 2174.14 L1759 2174.14 L1759 1483.98 L1729.46 1483.98 L1729.46 1483.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1729.46,1483.98 1729.46,2174.14 1759,2174.14 1759,1483.98 1729.46,1483.98 "/>
+<path clip-path="url(#clip913)" d="M1766.39 1473.7 L1766.39 2174.14 L1795.93 2174.14 L1795.93 1473.7 L1766.39 1473.7 L1766.39 1473.7  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1766.39,1473.7 1766.39,2174.14 1795.93,2174.14 1795.93,1473.7 1766.39,1473.7 "/>
+<path clip-path="url(#clip913)" d="M1803.32 1463.6 L1803.32 2174.14 L1832.86 2174.14 L1832.86 1463.6 L1803.32 1463.6 L1803.32 1463.6  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1803.32,1463.6 1803.32,2174.14 1832.86,2174.14 1832.86,1463.6 1803.32,1463.6 "/>
+<path clip-path="url(#clip913)" d="M1840.25 1453.7 L1840.25 2174.14 L1869.79 2174.14 L1869.79 1453.7 L1840.25 1453.7 L1840.25 1453.7  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1840.25,1453.7 1840.25,2174.14 1869.79,2174.14 1869.79,1453.7 1840.25,1453.7 "/>
+<path clip-path="url(#clip913)" d="M1877.18 1443.96 L1877.18 2174.14 L1906.73 2174.14 L1906.73 1443.96 L1877.18 1443.96 L1877.18 1443.96  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1877.18,1443.96 1877.18,2174.14 1906.73,2174.14 1906.73,1443.96 1877.18,1443.96 "/>
+<path clip-path="url(#clip913)" d="M1914.11 1434.39 L1914.11 2174.14 L1943.66 2174.14 L1943.66 1434.39 L1914.11 1434.39 L1914.11 1434.39  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1914.11,1434.39 1914.11,2174.14 1943.66,2174.14 1943.66,1434.39 1914.11,1434.39 "/>
+<path clip-path="url(#clip913)" d="M1951.04 1424.98 L1951.04 2174.14 L1980.59 2174.14 L1980.59 1424.98 L1951.04 1424.98 L1951.04 1424.98  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1951.04,1424.98 1951.04,2174.14 1980.59,2174.14 1980.59,1424.98 1951.04,1424.98 "/>
+<path clip-path="url(#clip913)" d="M1987.97 1415.72 L1987.97 2174.14 L2017.52 2174.14 L2017.52 1415.72 L1987.97 1415.72 L1987.97 1415.72  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1987.97,1415.72 1987.97,2174.14 2017.52,2174.14 2017.52,1415.72 1987.97,1415.72 "/>
+<path clip-path="url(#clip913)" d="M2024.9 1406.6 L2024.9 2174.14 L2054.45 2174.14 L2054.45 1406.6 L2024.9 1406.6 L2024.9 1406.6  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2024.9,1406.6 2024.9,2174.14 2054.45,2174.14 2054.45,1406.6 2024.9,1406.6 "/>
+<path clip-path="url(#clip913)" d="M2061.83 1397.61 L2061.83 2174.14 L2091.38 2174.14 L2091.38 1397.61 L2061.83 1397.61 L2061.83 1397.61  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2061.83,1397.61 2061.83,2174.14 2091.38,2174.14 2091.38,1397.61 2061.83,1397.61 "/>
+<path clip-path="url(#clip913)" d="M2098.76 1388.76 L2098.76 2174.14 L2128.31 2174.14 L2128.31 1388.76 L2098.76 1388.76 L2098.76 1388.76  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2098.76,1388.76 2098.76,2174.14 2128.31,2174.14 2128.31,1388.76 2098.76,1388.76 "/>
+<path clip-path="url(#clip913)" d="M2135.69 1380.03 L2135.69 2174.14 L2165.24 2174.14 L2165.24 1380.03 L2135.69 1380.03 L2135.69 1380.03  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2135.69,1380.03 2135.69,2174.14 2165.24,2174.14 2165.24,1380.03 2135.69,1380.03 "/>
+<path clip-path="url(#clip913)" d="M2172.62 1371.42 L2172.62 2174.14 L2202.17 2174.14 L2202.17 1371.42 L2172.62 1371.42 L2172.62 1371.42  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2172.62,1371.42 2172.62,2174.14 2202.17,2174.14 2202.17,1371.42 2172.62,1371.42 "/>
+<path clip-path="url(#clip913)" d="M2209.55 1362.92 L2209.55 2174.14 L2239.1 2174.14 L2239.1 1362.92 L2209.55 1362.92 L2209.55 1362.92  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip913)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2209.55,1362.92 2209.55,2174.14 2239.1,2174.14 2239.1,1362.92 2209.55,1362.92 "/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="414.741" cy="2174.14" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="451.672" cy="2078.86" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="488.602" cy="2039.4" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="525.532" cy="2009.11" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="562.463" cy="1983.58" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="599.393" cy="1961.09" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="636.323" cy="1940.76" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="673.253" cy="1922.06" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="710.184" cy="1904.65" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="747.114" cy="1888.3" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="784.044" cy="1872.84" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="820.975" cy="1858.14" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="857.905" cy="1844.08" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="894.835" cy="1830.61" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="931.766" cy="1817.64" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="968.696" cy="1793.51" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1005.63" cy="1769.49" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1042.56" cy="1747.67" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1079.49" cy="1727.55" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1116.42" cy="1708.77" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1153.35" cy="1691.1" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1190.28" cy="1674.37" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1227.21" cy="1658.44" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1264.14" cy="1643.2" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1301.07" cy="1628.58" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1338" cy="1614.49" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1374.93" cy="1600.9" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1411.86" cy="1587.74" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1448.79" cy="1574.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1485.72" cy="1562.59" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1522.65" cy="1550.53" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1559.58" cy="1538.79" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1596.51" cy="1527.33" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1633.44" cy="1516.13" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1670.37" cy="1505.19" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1707.3" cy="1494.48" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1744.23" cy="1483.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1781.16" cy="1473.7" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1818.09" cy="1463.6" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1855.02" cy="1453.7" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1891.95" cy="1443.96" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1928.88" cy="1434.39" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="1965.81" cy="1424.98" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2002.74" cy="1415.72" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2039.67" cy="1406.6" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2076.6" cy="1397.61" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2113.53" cy="1388.76" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2150.47" cy="1380.03" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2187.4" cy="1371.42" r="2"/>
+<circle clip-path="url(#clip913)" style="fill:#3da44d; stroke:none; fill-opacity:0" cx="2224.33" cy="1362.92" r="2"/>
+<path clip-path="url(#clip910)" d="M355.193 1597.33 L677.137 1597.33 L677.137 1389.97 L355.193 1389.97  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="355.193,1597.33 677.137,1597.33 677.137,1389.97 355.193,1389.97 355.193,1597.33 "/>
+<path clip-path="url(#clip910)" d="M378.153 1462.54 L515.916 1462.54 L515.916 1421.07 L378.153 1421.07 L378.153 1462.54  Z" fill="#009af9" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1462.54 515.916,1462.54 515.916,1421.07 378.153,1421.07 378.153,1462.54 "/>
+<path clip-path="url(#clip910)" d="M538.877 1424.53 L543.553 1424.53 L543.553 1459.09 L538.877 1459.09 L538.877 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M557.534 1442.58 L557.534 1455.24 L565.034 1455.24 Q568.807 1455.24 570.613 1453.69 Q572.441 1452.12 572.441 1448.9 Q572.441 1445.66 570.613 1444.13 Q568.807 1442.58 565.034 1442.58 L557.534 1442.58 M557.534 1428.37 L557.534 1438.78 L564.455 1438.78 Q567.881 1438.78 569.548 1437.51 Q571.238 1436.21 571.238 1433.58 Q571.238 1430.96 569.548 1429.66 Q567.881 1428.37 564.455 1428.37 L557.534 1428.37 M552.858 1424.53 L564.802 1424.53 Q570.15 1424.53 573.043 1426.75 Q575.937 1428.97 575.937 1433.07 Q575.937 1436.24 574.455 1438.11 Q572.974 1439.99 570.103 1440.45 Q573.552 1441.19 575.451 1443.55 Q577.372 1445.89 577.372 1449.41 Q577.372 1454.04 574.224 1456.56 Q571.076 1459.09 565.265 1459.09 L552.858 1459.09 L552.858 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M585.381 1424.53 L592.349 1424.53 L601.168 1448.04 L610.034 1424.53 L617.001 1424.53 L617.001 1459.09 L612.441 1459.09 L612.441 1428.74 L603.529 1452.44 L598.83 1452.44 L589.918 1428.74 L589.918 1459.09 L585.381 1459.09 L585.381 1424.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M378.153 1514.38 L515.916 1514.38 L515.916 1472.91 L378.153 1472.91 L378.153 1514.38  Z" fill="#e26f46" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1514.38 515.916,1514.38 515.916,1472.91 378.153,1472.91 378.153,1514.38 "/>
+<path clip-path="url(#clip910)" d="M538.877 1476.37 L543.599 1476.37 L550.867 1505.58 L558.113 1476.37 L563.367 1476.37 L570.636 1505.58 L577.881 1476.37 L582.626 1476.37 L573.946 1510.93 L568.066 1510.93 L560.775 1480.93 L553.414 1510.93 L547.534 1510.93 L538.877 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M588.83 1476.37 L595.798 1476.37 L604.617 1499.88 L613.483 1476.37 L620.45 1476.37 L620.45 1510.93 L615.89 1510.93 L615.89 1480.58 L606.978 1504.28 L602.279 1504.28 L593.367 1480.58 L593.367 1510.93 L588.83 1510.93 L588.83 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M624.941 1476.37 L654.177 1476.37 L654.177 1480.3 L641.909 1480.3 L641.909 1510.93 L637.209 1510.93 L637.209 1480.3 L624.941 1480.3 L624.941 1476.37 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M378.153 1566.22 L515.916 1566.22 L515.916 1524.75 L378.153 1524.75 L378.153 1566.22  Z" fill="#3da44d" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip910)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="378.153,1566.22 515.916,1566.22 515.916,1524.75 378.153,1524.75 378.153,1566.22 "/>
+<path clip-path="url(#clip910)" d="M561.122 1529.34 L561.122 1533.9 Q558.46 1532.63 556.099 1532 Q553.738 1531.38 551.539 1531.38 Q547.719 1531.38 545.636 1532.86 Q543.576 1534.34 543.576 1537.07 Q543.576 1539.36 544.941 1540.54 Q546.33 1541.7 550.173 1542.42 L552.997 1543 Q558.228 1543.99 560.705 1546.52 Q563.205 1549.02 563.205 1553.23 Q563.205 1558.25 559.826 1560.84 Q556.469 1563.44 549.965 1563.44 Q547.511 1563.44 544.733 1562.88 Q541.979 1562.33 539.016 1561.24 L539.016 1556.42 Q541.863 1558.02 544.594 1558.83 Q547.326 1559.64 549.965 1559.64 Q553.969 1559.64 556.145 1558.07 Q558.321 1556.49 558.321 1553.58 Q558.321 1551.03 556.747 1549.59 Q555.196 1548.16 551.631 1547.44 L548.784 1546.89 Q543.553 1545.84 541.215 1543.62 Q538.877 1541.4 538.877 1537.44 Q538.877 1532.86 542.094 1530.22 Q545.335 1527.58 551.006 1527.58 Q553.437 1527.58 555.96 1528.02 Q558.483 1528.46 561.122 1529.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M570.497 1528.21 L592.349 1528.21 L592.349 1532.14 L575.173 1532.14 L575.173 1542.37 L591.631 1542.37 L591.631 1546.31 L575.173 1546.31 L575.173 1558.83 L592.765 1558.83 L592.765 1562.77 L570.497 1562.77 L570.497 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M600.45 1528.21 L605.126 1528.21 L605.126 1542.37 L622.117 1542.37 L622.117 1528.21 L626.793 1528.21 L626.793 1562.77 L622.117 1562.77 L622.117 1546.31 L605.126 1546.31 L605.126 1562.77 L600.45 1562.77 L600.45 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip910)" d="M636.098 1528.21 L640.774 1528.21 L640.774 1562.77 L636.098 1562.77 L636.098 1528.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /></svg>
diff --git a/previews/PR3562/tutorials/nonlinear/portfolio/index.html b/previews/PR3562/tutorials/nonlinear/portfolio/index.html
index 3dd97d19425..efea9358856 100644
--- a/previews/PR3562/tutorials/nonlinear/portfolio/index.html
+++ b/previews/PR3562/tutorials/nonlinear/portfolio/index.html
@@ -71,12 +71,12 @@
   Dual objective value : 4.52688e+04
 
 * Work counters
-  Solve time (sec)   : 4.47297e-03
+  Solve time (sec)   : 6.39391e-03
   Barrier iterations : 11
 </code></pre><p>The optimal allocation of our assets is:</p><pre><code class="language-julia hljs">value.(x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">3-element Vector{Float64}:
- 497.045529849864
+ 497.04552984986407
   -9.670578501816873e-9
- 502.9544801594809</code></pre><p>So we spend <span>$</span>497 on IBM, and <span>$</span>503 on SEHI. This results in a variance of:</p><pre><code class="language-julia hljs">scalar_variance = value(x&#39; * Q * x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">22634.417849884147</code></pre><p>and an expected return of:</p><pre><code class="language-julia hljs">scalar_return = value(r&#39; * x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">49.999999500002374</code></pre><h2 id="Multi-objective-portfolio-optimization"><a class="docs-heading-anchor" href="#Multi-objective-portfolio-optimization">Multi-objective portfolio optimization</a><a id="Multi-objective-portfolio-optimization-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-objective-portfolio-optimization" title="Permalink"></a></h2><p>The previous model returned a single solution that minimized the variance, ensuring that our expected return was at least <span>$</span>50. In practice, we might be willing to accept a slightly higher variance if it meant a much increased expected return. To explore this problem space, we can instead formulate our portfolio optimization problem with two objectives:</p><ol><li>to minimize the variance</li><li>to maximize the expected return</li></ol><p>The solution to this bi-objective problem is the <a href="https://en.wikipedia.org/wiki/Efficient_frontier">efficient frontier</a> of modern portfolio theory, and each point in the solution is a point with the best return for a fixed level of risk.</p><pre><code class="language-julia hljs">model = Model(() -&gt; MOA.Optimizer(Ipopt.Optimizer))
+ 502.9544801594808</code></pre><p>So we spend <span>$</span>497 on IBM, and <span>$</span>503 on SEHI. This results in a variance of:</p><pre><code class="language-julia hljs">scalar_variance = value(x&#39; * Q * x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">22634.41784988414</code></pre><p>and an expected return of:</p><pre><code class="language-julia hljs">scalar_return = value(r&#39; * x)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">49.99999950000236</code></pre><h2 id="Multi-objective-portfolio-optimization"><a class="docs-heading-anchor" href="#Multi-objective-portfolio-optimization">Multi-objective portfolio optimization</a><a id="Multi-objective-portfolio-optimization-1"></a><a class="docs-heading-anchor-permalink" href="#Multi-objective-portfolio-optimization" title="Permalink"></a></h2><p>The previous model returned a single solution that minimized the variance, ensuring that our expected return was at least <span>$</span>50. In practice, we might be willing to accept a slightly higher variance if it meant a much increased expected return. To explore this problem space, we can instead formulate our portfolio optimization problem with two objectives:</p><ol><li>to minimize the variance</li><li>to maximize the expected return</li></ol><p>The solution to this bi-objective problem is the <a href="https://en.wikipedia.org/wiki/Efficient_frontier">efficient frontier</a> of modern portfolio theory, and each point in the solution is a point with the best return for a fixed level of risk.</p><pre><code class="language-julia hljs">model = Model(() -&gt; MOA.Optimizer(Ipopt.Optimizer))
 set_silent(model)</code></pre><p>We also need to choose a solution algorithm for <code>MOA</code>. For our problem, the efficient frontier will have an infinite number of solutions. Since we cannot find all of the solutions, we choose an approximation algorithm and limit the number of solution points that are returned:</p><pre><code class="language-julia hljs">set_optimizer_attribute(model, MOA.Algorithm(), MOA.EpsilonConstraint())
 set_optimizer_attribute(model, MOA.SolutionLimit(), 50)</code></pre><p>Now we can define the rest of the model:</p><pre><code class="language-julia hljs">@variable(model, x[1:3] &gt;= 0)
 @constraint(model, sum(x) &lt;= 1000)
@@ -102,7 +102,7 @@
   Dual objective value : -7.37159e+01
 
 * Work counters
-  Solve time (sec)   : 3.20848e-01
+  Solve time (sec)   : 4.73567e-01
   Barrier iterations : 7
 </code></pre><p>The algorithm found 50 different solutions. Let&#39;s plot them to see how they differ:</p><pre><code class="language-julia hljs">objective_space = Plots.hline(
     [scalar_return];
@@ -135,4 +135,4 @@
     ylabel = &quot;Investment (\$)&quot;,
     title = &quot;Decision space&quot;,
 )
-Plots.plot(objective_space, decision_space; layout = (2, 1), size = (600, 600))</code></pre><img src="a315d69e.svg" alt="Example block output"/><p>Perhaps our trade-off wasn&#39;t so bad after all. Our original solution corresponded to picking a solution #17. If we buy more SEHI, we can increase the return, but the variance also increases. If we buy less SEHI, such as a solution like #5 or #6, then we can achieve the corresponding return without deploying all of our capital. We should also note that at no point should we buy WMT.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../simple_examples/">« Simple examples</a><a class="docs-footer-nextpage" href="../rocket_control/">Rocket Control »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+Plots.plot(objective_space, decision_space; layout = (2, 1), size = (600, 600))</code></pre><img src="e1860e4a.svg" alt="Example block output"/><p>Perhaps our trade-off wasn&#39;t so bad after all. Our original solution corresponded to picking a solution #17. If we buy more SEHI, we can increase the return, but the variance also increases. If we buy less SEHI, such as a solution like #5 or #6, then we can achieve the corresponding return without deploying all of our capital. We should also note that at no point should we buy WMT.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../simple_examples/">« Simple examples</a><a class="docs-footer-nextpage" href="../rocket_control/">Rocket Control »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/querying_hessians/index.html b/previews/PR3562/tutorials/nonlinear/querying_hessians/index.html
index c7163805378..5a513e9aed8 100644
--- a/previews/PR3562/tutorials/nonlinear/querying_hessians/index.html
+++ b/previews/PR3562/tutorials/nonlinear/querying_hessians/index.html
@@ -136,4 +136,4 @@
   501.944  -316.258
  -316.258   199.885</code></pre><p>If we look at the eigenvalues of the Hessian:</p><pre><code class="language-julia hljs">LinearAlgebra.eigvals(H_star)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">2-element Vector{Float64}:
    0.4443995924983142
- 701.3843426037456</code></pre><p>we see that they are all positive. Therefore, the Hessian is positive definite, and so the solution found by Ipopt is a local minimizer.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nested_problems/">« Nested optimization problems</a><a class="docs-footer-nextpage" href="../complementarity/">Mixed complementarity problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+ 701.3843426037456</code></pre><p>we see that they are all positive. Therefore, the Hessian is positive definite, and so the solution found by Ipopt is a local minimizer.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nested_problems/">« Nested optimization problems</a><a class="docs-footer-nextpage" href="../complementarity/">Mixed complementarity problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/rocket_control/5e56f1a3.svg b/previews/PR3562/tutorials/nonlinear/rocket_control/cf89a871.svg
similarity index 89%
rename from previews/PR3562/tutorials/nonlinear/rocket_control/5e56f1a3.svg
rename to previews/PR3562/tutorials/nonlinear/rocket_control/cf89a871.svg
index 39cc4d37782..14fecbee2bc 100644
--- a/previews/PR3562/tutorials/nonlinear/rocket_control/5e56f1a3.svg
+++ b/previews/PR3562/tutorials/nonlinear/rocket_control/cf89a871.svg
@@ -1,130 +1,130 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip150">
+  <clipPath id="clip970">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip150)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip970)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip151">
+  <clipPath id="clip971">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip150)" d="M277.528 623.18 L1181.75 623.18 L1181.75 47.2441 L277.528 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip970)" d="M277.528 623.18 L1181.75 623.18 L1181.75 47.2441 L277.528 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip152">
+  <clipPath id="clip972">
     <rect x="277" y="47" width="905" height="577"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="302.265,623.18 302.265,47.2441 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="515.738,623.18 515.738,47.2441 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.211,623.18 729.211,47.2441 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="942.685,623.18 942.685,47.2441 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1156.16,623.18 1156.16,47.2441 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,623.18 1181.75,623.18 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="302.265,623.18 302.265,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="515.738,623.18 515.738,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.211,623.18 729.211,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="942.685,623.18 942.685,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1156.16,623.18 1156.16,604.282 "/>
-<path clip-path="url(#clip150)" d="M264.569 654.099 Q260.958 654.099 259.129 657.663 Q257.323 661.205 257.323 668.335 Q257.323 675.441 259.129 679.006 Q260.958 682.547 264.569 682.547 Q268.203 682.547 270.008 679.006 Q271.837 675.441 271.837 668.335 Q271.837 661.205 270.008 657.663 Q268.203 654.099 264.569 654.099 M264.569 650.395 Q270.379 650.395 273.434 655.001 Q276.513 659.585 276.513 668.335 Q276.513 677.061 273.434 681.668 Q270.379 686.251 264.569 686.251 Q258.759 686.251 255.68 681.668 Q252.624 677.061 252.624 668.335 Q252.624 659.585 255.68 655.001 Q258.759 650.395 264.569 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M284.731 679.7 L289.615 679.7 L289.615 685.58 L284.731 685.58 L284.731 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M309.8 654.099 Q306.189 654.099 304.36 657.663 Q302.555 661.205 302.555 668.335 Q302.555 675.441 304.36 679.006 Q306.189 682.547 309.8 682.547 Q313.434 682.547 315.24 679.006 Q317.068 675.441 317.068 668.335 Q317.068 661.205 315.24 657.663 Q313.434 654.099 309.8 654.099 M309.8 650.395 Q315.61 650.395 318.666 655.001 Q321.744 659.585 321.744 668.335 Q321.744 677.061 318.666 681.668 Q315.61 686.251 309.8 686.251 Q303.99 686.251 300.911 681.668 Q297.856 677.061 297.856 668.335 Q297.856 659.585 300.911 655.001 Q303.99 650.395 309.8 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M339.962 654.099 Q336.351 654.099 334.522 657.663 Q332.716 661.205 332.716 668.335 Q332.716 675.441 334.522 679.006 Q336.351 682.547 339.962 682.547 Q343.596 682.547 345.402 679.006 Q347.23 675.441 347.23 668.335 Q347.23 661.205 345.402 657.663 Q343.596 654.099 339.962 654.099 M339.962 650.395 Q345.772 650.395 348.828 655.001 Q351.906 659.585 351.906 668.335 Q351.906 677.061 348.828 681.668 Q345.772 686.251 339.962 686.251 Q334.152 686.251 331.073 681.668 Q328.017 677.061 328.017 668.335 Q328.017 659.585 331.073 655.001 Q334.152 650.395 339.962 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M478.539 654.099 Q474.928 654.099 473.1 657.663 Q471.294 661.205 471.294 668.335 Q471.294 675.441 473.1 679.006 Q474.928 682.547 478.539 682.547 Q482.174 682.547 483.979 679.006 Q485.808 675.441 485.808 668.335 Q485.808 661.205 483.979 657.663 Q482.174 654.099 478.539 654.099 M478.539 650.395 Q484.35 650.395 487.405 655.001 Q490.484 659.585 490.484 668.335 Q490.484 677.061 487.405 681.668 Q484.35 686.251 478.539 686.251 Q472.729 686.251 469.651 681.668 Q466.595 677.061 466.595 668.335 Q466.595 659.585 469.651 655.001 Q472.729 650.395 478.539 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M498.701 679.7 L503.586 679.7 L503.586 685.58 L498.701 685.58 L498.701 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M523.771 654.099 Q520.16 654.099 518.331 657.663 Q516.525 661.205 516.525 668.335 Q516.525 675.441 518.331 679.006 Q520.16 682.547 523.771 682.547 Q527.405 682.547 529.211 679.006 Q531.039 675.441 531.039 668.335 Q531.039 661.205 529.211 657.663 Q527.405 654.099 523.771 654.099 M523.771 650.395 Q529.581 650.395 532.636 655.001 Q535.715 659.585 535.715 668.335 Q535.715 677.061 532.636 681.668 Q529.581 686.251 523.771 686.251 Q517.961 686.251 514.882 681.668 Q511.826 677.061 511.826 668.335 Q511.826 659.585 514.882 655.001 Q517.961 650.395 523.771 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M543.979 651.02 L562.335 651.02 L562.335 654.955 L548.261 654.955 L548.261 663.427 Q549.28 663.08 550.298 662.918 Q551.317 662.733 552.335 662.733 Q558.122 662.733 561.502 665.904 Q564.882 669.075 564.882 674.492 Q564.882 680.071 561.409 683.172 Q557.937 686.251 551.618 686.251 Q549.442 686.251 547.173 685.881 Q544.928 685.51 542.521 684.77 L542.521 680.071 Q544.604 681.205 546.826 681.76 Q549.048 682.316 551.525 682.316 Q555.53 682.316 557.868 680.21 Q560.206 678.103 560.206 674.492 Q560.206 670.881 557.868 668.774 Q555.53 666.668 551.525 666.668 Q549.65 666.668 547.775 667.085 Q545.923 667.501 543.979 668.381 L543.979 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M691.515 654.099 Q687.904 654.099 686.075 657.663 Q684.27 661.205 684.27 668.335 Q684.27 675.441 686.075 679.006 Q687.904 682.547 691.515 682.547 Q695.149 682.547 696.955 679.006 Q698.783 675.441 698.783 668.335 Q698.783 661.205 696.955 657.663 Q695.149 654.099 691.515 654.099 M691.515 650.395 Q697.325 650.395 700.381 655.001 Q703.459 659.585 703.459 668.335 Q703.459 677.061 700.381 681.668 Q697.325 686.251 691.515 686.251 Q685.705 686.251 682.626 681.668 Q679.571 677.061 679.571 668.335 Q679.571 659.585 682.626 655.001 Q685.705 650.395 691.515 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M711.677 679.7 L716.561 679.7 L716.561 685.58 L711.677 685.58 L711.677 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M727.556 681.645 L735.195 681.645 L735.195 655.279 L726.885 656.946 L726.885 652.687 L735.149 651.02 L739.825 651.02 L739.825 681.645 L747.464 681.645 L747.464 685.58 L727.556 685.58 L727.556 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M766.908 654.099 Q763.297 654.099 761.468 657.663 Q759.663 661.205 759.663 668.335 Q759.663 675.441 761.468 679.006 Q763.297 682.547 766.908 682.547 Q770.542 682.547 772.348 679.006 Q774.176 675.441 774.176 668.335 Q774.176 661.205 772.348 657.663 Q770.542 654.099 766.908 654.099 M766.908 650.395 Q772.718 650.395 775.774 655.001 Q778.852 659.585 778.852 668.335 Q778.852 677.061 775.774 681.668 Q772.718 686.251 766.908 686.251 Q761.098 686.251 758.019 681.668 Q754.964 677.061 754.964 668.335 Q754.964 659.585 758.019 655.001 Q761.098 650.395 766.908 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M905.486 654.099 Q901.875 654.099 900.046 657.663 Q898.24 661.205 898.24 668.335 Q898.24 675.441 900.046 679.006 Q901.875 682.547 905.486 682.547 Q909.12 682.547 910.925 679.006 Q912.754 675.441 912.754 668.335 Q912.754 661.205 910.925 657.663 Q909.12 654.099 905.486 654.099 M905.486 650.395 Q911.296 650.395 914.351 655.001 Q917.43 659.585 917.43 668.335 Q917.43 677.061 914.351 681.668 Q911.296 686.251 905.486 686.251 Q899.676 686.251 896.597 681.668 Q893.541 677.061 893.541 668.335 Q893.541 659.585 896.597 655.001 Q899.676 650.395 905.486 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M925.648 679.7 L930.532 679.7 L930.532 685.58 L925.648 685.58 L925.648 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M941.527 681.645 L949.166 681.645 L949.166 655.279 L940.856 656.946 L940.856 652.687 L949.12 651.02 L953.796 651.02 L953.796 681.645 L961.434 681.645 L961.434 685.58 L941.527 685.58 L941.527 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M970.925 651.02 L989.282 651.02 L989.282 654.955 L975.208 654.955 L975.208 663.427 Q976.226 663.08 977.245 662.918 Q978.263 662.733 979.282 662.733 Q985.069 662.733 988.448 665.904 Q991.828 669.075 991.828 674.492 Q991.828 680.071 988.356 683.172 Q984.883 686.251 978.564 686.251 Q976.388 686.251 974.12 685.881 Q971.874 685.51 969.467 684.77 L969.467 680.071 Q971.55 681.205 973.772 681.76 Q975.995 682.316 978.471 682.316 Q982.476 682.316 984.814 680.21 Q987.152 678.103 987.152 674.492 Q987.152 670.881 984.814 668.774 Q982.476 666.668 978.471 666.668 Q976.596 666.668 974.721 667.085 Q972.87 667.501 970.925 668.381 L970.925 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1118.46 654.099 Q1114.85 654.099 1113.02 657.663 Q1111.22 661.205 1111.22 668.335 Q1111.22 675.441 1113.02 679.006 Q1114.85 682.547 1118.46 682.547 Q1122.1 682.547 1123.9 679.006 Q1125.73 675.441 1125.73 668.335 Q1125.73 661.205 1123.9 657.663 Q1122.1 654.099 1118.46 654.099 M1118.46 650.395 Q1124.27 650.395 1127.33 655.001 Q1130.41 659.585 1130.41 668.335 Q1130.41 677.061 1127.33 681.668 Q1124.27 686.251 1118.46 686.251 Q1112.65 686.251 1109.57 681.668 Q1106.52 677.061 1106.52 668.335 Q1106.52 659.585 1109.57 655.001 Q1112.65 650.395 1118.46 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1138.62 679.7 L1143.51 679.7 L1143.51 685.58 L1138.62 685.58 L1138.62 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1157.72 681.645 L1174.04 681.645 L1174.04 685.58 L1152.1 685.58 L1152.1 681.645 Q1154.76 678.89 1159.34 674.26 Q1163.95 669.608 1165.13 668.265 Q1167.37 665.742 1168.25 664.006 Q1169.16 662.247 1169.16 660.557 Q1169.16 657.802 1167.21 656.066 Q1165.29 654.33 1162.19 654.33 Q1159.99 654.33 1157.53 655.094 Q1155.1 655.858 1152.33 657.409 L1152.33 652.687 Q1155.15 651.552 1157.6 650.974 Q1160.06 650.395 1162.1 650.395 Q1167.47 650.395 1170.66 653.08 Q1173.85 655.765 1173.85 660.256 Q1173.85 662.386 1173.04 664.307 Q1172.26 666.205 1170.15 668.798 Q1169.57 669.469 1166.47 672.686 Q1163.37 675.881 1157.72 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1193.85 654.099 Q1190.24 654.099 1188.41 657.663 Q1186.61 661.205 1186.61 668.335 Q1186.61 675.441 1188.41 679.006 Q1190.24 682.547 1193.85 682.547 Q1197.49 682.547 1199.29 679.006 Q1201.12 675.441 1201.12 668.335 Q1201.12 661.205 1199.29 657.663 Q1197.49 654.099 1193.85 654.099 M1193.85 650.395 Q1199.66 650.395 1202.72 655.001 Q1205.8 659.585 1205.8 668.335 Q1205.8 677.061 1202.72 681.668 Q1199.66 686.251 1193.85 686.251 Q1188.04 686.251 1184.97 681.668 Q1181.91 677.061 1181.91 668.335 Q1181.91 659.585 1184.97 655.001 Q1188.04 650.395 1193.85 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M599.81 720.524 L640.009 720.524 L640.009 725.935 L623.14 725.935 L623.14 768.044 L616.679 768.044 L616.679 725.935 L599.81 725.935 L599.81 720.524 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M643.956 732.396 L649.812 732.396 L649.812 768.044 L643.956 768.044 L643.956 732.396 M643.956 718.519 L649.812 718.519 L649.812 725.935 L643.956 725.935 L643.956 718.519 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M689.821 739.239 Q692.017 735.292 695.073 733.414 Q698.128 731.537 702.266 731.537 Q707.836 731.537 710.86 735.451 Q713.883 739.335 713.883 746.528 L713.883 768.044 L707.995 768.044 L707.995 746.719 Q707.995 741.594 706.181 739.112 Q704.367 736.629 700.643 736.629 Q696.091 736.629 693.449 739.653 Q690.808 742.677 690.808 747.896 L690.808 768.044 L684.919 768.044 L684.919 746.719 Q684.919 741.563 683.105 739.112 Q681.291 736.629 677.503 736.629 Q673.015 736.629 670.374 739.685 Q667.732 742.708 667.732 747.896 L667.732 768.044 L661.844 768.044 L661.844 732.396 L667.732 732.396 L667.732 737.934 Q669.737 734.656 672.538 733.096 Q675.339 731.537 679.19 731.537 Q683.073 731.537 685.779 733.51 Q688.516 735.483 689.821 739.239 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M756.056 748.756 L756.056 751.62 L729.129 751.62 Q729.511 757.668 732.758 760.851 Q736.036 764.002 741.861 764.002 Q745.234 764.002 748.385 763.174 Q751.568 762.347 754.687 760.691 L754.687 766.23 Q751.536 767.566 748.226 768.267 Q744.916 768.967 741.51 768.967 Q732.98 768.967 727.983 764.002 Q723.018 759.036 723.018 750.57 Q723.018 741.817 727.729 736.693 Q732.471 731.537 740.492 731.537 Q747.685 731.537 751.855 736.183 Q756.056 740.799 756.056 748.756 M750.2 747.037 Q750.136 742.231 747.494 739.366 Q744.884 736.502 740.556 736.502 Q735.654 736.502 732.694 739.271 Q729.766 742.04 729.32 747.069 L750.2 747.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M800.457 718.582 Q796.192 725.903 794.123 733.064 Q792.054 740.226 792.054 747.578 Q792.054 754.93 794.123 762.156 Q796.224 769.349 800.457 776.638 L795.364 776.638 Q790.59 769.158 788.203 761.933 Q785.848 754.708 785.848 747.578 Q785.848 740.48 788.203 733.287 Q790.558 726.094 795.364 718.582 L800.457 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M834.545 733.446 L834.545 738.984 Q832.063 737.711 829.389 737.075 Q826.715 736.438 823.851 736.438 Q819.49 736.438 817.294 737.775 Q815.13 739.112 815.13 741.785 Q815.13 743.822 816.689 745 Q818.249 746.146 822.96 747.196 L824.965 747.642 Q831.203 748.979 833.813 751.429 Q836.455 753.848 836.455 758.209 Q836.455 763.174 832.508 766.07 Q828.593 768.967 821.718 768.967 Q818.854 768.967 815.735 768.394 Q812.647 767.853 809.21 766.739 L809.21 760.691 Q812.456 762.378 815.607 763.238 Q818.758 764.065 821.846 764.065 Q825.983 764.065 828.211 762.665 Q830.439 761.233 830.439 758.654 Q830.439 756.267 828.816 754.994 Q827.225 753.721 821.782 752.543 L819.745 752.066 Q814.302 750.92 811.883 748.565 Q809.464 746.178 809.464 742.04 Q809.464 737.011 813.029 734.274 Q816.594 731.537 823.151 731.537 Q826.397 731.537 829.262 732.014 Q832.126 732.491 834.545 733.446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M844.858 718.582 L849.95 718.582 Q854.725 726.094 857.08 733.287 Q859.467 740.48 859.467 747.578 Q859.467 754.708 857.08 761.933 Q854.725 769.158 849.95 776.638 L844.858 776.638 Q849.091 769.349 851.16 762.156 Q853.26 754.93 853.26 747.578 Q853.26 740.226 851.16 733.064 Q849.091 725.903 844.858 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,521.847 1181.75,521.847 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,394.297 1181.75,394.297 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,266.747 1181.75,266.747 "/>
-<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,139.198 1181.75,139.198 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,623.18 277.528,47.2441 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,521.847 296.426,521.847 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,394.297 296.426,394.297 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,266.747 296.426,266.747 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,139.198 296.426,139.198 "/>
-<path clip-path="url(#clip150)" d="M116.436 535.192 L124.075 535.192 L124.075 508.826 L115.765 510.493 L115.765 506.233 L124.029 504.567 L128.705 504.567 L128.705 535.192 L136.343 535.192 L136.343 539.127 L116.436 539.127 L116.436 535.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M145.788 533.247 L150.672 533.247 L150.672 539.127 L145.788 539.127 L145.788 533.247 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M170.857 507.645 Q167.246 507.645 165.417 511.21 Q163.612 514.752 163.612 521.881 Q163.612 528.988 165.417 532.553 Q167.246 536.094 170.857 536.094 Q174.491 536.094 176.297 532.553 Q178.126 528.988 178.126 521.881 Q178.126 514.752 176.297 511.21 Q174.491 507.645 170.857 507.645 M170.857 503.942 Q176.667 503.942 179.723 508.548 Q182.801 513.131 182.801 521.881 Q182.801 530.608 179.723 535.215 Q176.667 539.798 170.857 539.798 Q165.047 539.798 161.968 535.215 Q158.913 530.608 158.913 521.881 Q158.913 513.131 161.968 508.548 Q165.047 503.942 170.857 503.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M201.019 507.645 Q197.408 507.645 195.579 511.21 Q193.774 514.752 193.774 521.881 Q193.774 528.988 195.579 532.553 Q197.408 536.094 201.019 536.094 Q204.653 536.094 206.459 532.553 Q208.287 528.988 208.287 521.881 Q208.287 514.752 206.459 511.21 Q204.653 507.645 201.019 507.645 M201.019 503.942 Q206.829 503.942 209.885 508.548 Q212.963 513.131 212.963 521.881 Q212.963 530.608 209.885 535.215 Q206.829 539.798 201.019 539.798 Q195.209 539.798 192.13 535.215 Q189.075 530.608 189.075 521.881 Q189.075 513.131 192.13 508.548 Q195.209 503.942 201.019 503.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M225.209 535.192 L241.528 535.192 L241.528 539.127 L219.584 539.127 L219.584 535.192 Q222.246 532.437 226.829 527.807 Q231.435 523.155 232.616 521.812 Q234.861 519.289 235.741 517.553 Q236.644 515.793 236.644 514.104 Q236.644 511.349 234.699 509.613 Q232.778 507.877 229.676 507.877 Q227.477 507.877 225.023 508.641 Q222.593 509.405 219.815 510.956 L219.815 506.233 Q222.639 505.099 225.093 504.52 Q227.547 503.942 229.584 503.942 Q234.954 503.942 238.148 506.627 Q241.343 509.312 241.343 513.803 Q241.343 515.932 240.533 517.854 Q239.746 519.752 237.639 522.344 Q237.06 523.016 233.959 526.233 Q230.857 529.428 225.209 535.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M115.834 407.642 L123.473 407.642 L123.473 381.276 L115.163 382.943 L115.163 378.684 L123.427 377.017 L128.103 377.017 L128.103 407.642 L135.742 407.642 L135.742 411.577 L115.834 411.577 L115.834 407.642 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M145.186 405.697 L150.07 405.697 L150.07 411.577 L145.186 411.577 L145.186 405.697 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M170.255 380.096 Q166.644 380.096 164.815 383.661 Q163.01 387.202 163.01 394.332 Q163.01 401.438 164.815 405.003 Q166.644 408.545 170.255 408.545 Q173.889 408.545 175.695 405.003 Q177.524 401.438 177.524 394.332 Q177.524 387.202 175.695 383.661 Q173.889 380.096 170.255 380.096 M170.255 376.392 Q176.065 376.392 179.121 380.999 Q182.2 385.582 182.2 394.332 Q182.2 403.059 179.121 407.665 Q176.065 412.248 170.255 412.248 Q164.445 412.248 161.366 407.665 Q158.311 403.059 158.311 394.332 Q158.311 385.582 161.366 380.999 Q164.445 376.392 170.255 376.392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M200.417 380.096 Q196.806 380.096 194.977 383.661 Q193.172 387.202 193.172 394.332 Q193.172 401.438 194.977 405.003 Q196.806 408.545 200.417 408.545 Q204.051 408.545 205.857 405.003 Q207.686 401.438 207.686 394.332 Q207.686 387.202 205.857 383.661 Q204.051 380.096 200.417 380.096 M200.417 376.392 Q206.227 376.392 209.283 380.999 Q212.361 385.582 212.361 394.332 Q212.361 403.059 209.283 407.665 Q206.227 412.248 200.417 412.248 Q194.607 412.248 191.528 407.665 Q188.473 403.059 188.473 394.332 Q188.473 385.582 191.528 380.999 Q194.607 376.392 200.417 376.392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M220.625 377.017 L238.982 377.017 L238.982 380.952 L224.908 380.952 L224.908 389.424 Q225.926 389.077 226.945 388.915 Q227.963 388.73 228.982 388.73 Q234.769 388.73 238.148 391.901 Q241.528 395.073 241.528 400.489 Q241.528 406.068 238.056 409.17 Q234.584 412.248 228.264 412.248 Q226.088 412.248 223.82 411.878 Q221.574 411.508 219.167 410.767 L219.167 406.068 Q221.25 407.202 223.473 407.758 Q225.695 408.313 228.172 408.313 Q232.176 408.313 234.514 406.207 Q236.852 404.1 236.852 400.489 Q236.852 396.878 234.514 394.772 Q232.176 392.665 228.172 392.665 Q226.297 392.665 224.422 393.082 Q222.57 393.498 220.625 394.378 L220.625 377.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M114.931 280.092 L122.57 280.092 L122.57 253.727 L114.26 255.393 L114.26 251.134 L122.524 249.467 L127.2 249.467 L127.2 280.092 L134.839 280.092 L134.839 284.027 L114.931 284.027 L114.931 280.092 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M144.283 278.148 L149.167 278.148 L149.167 284.027 L144.283 284.027 L144.283 278.148 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M169.352 252.546 Q165.741 252.546 163.913 256.111 Q162.107 259.653 162.107 266.782 Q162.107 273.889 163.913 277.453 Q165.741 280.995 169.352 280.995 Q172.987 280.995 174.792 277.453 Q176.621 273.889 176.621 266.782 Q176.621 259.653 174.792 256.111 Q172.987 252.546 169.352 252.546 M169.352 248.842 Q175.163 248.842 178.218 253.449 Q181.297 258.032 181.297 266.782 Q181.297 275.509 178.218 280.115 Q175.163 284.699 169.352 284.699 Q163.542 284.699 160.464 280.115 Q157.408 275.509 157.408 266.782 Q157.408 258.032 160.464 253.449 Q163.542 248.842 169.352 248.842 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M199.514 252.546 Q195.903 252.546 194.075 256.111 Q192.269 259.653 192.269 266.782 Q192.269 273.889 194.075 277.453 Q195.903 280.995 199.514 280.995 Q203.149 280.995 204.954 277.453 Q206.783 273.889 206.783 266.782 Q206.783 259.653 204.954 256.111 Q203.149 252.546 199.514 252.546 M199.514 248.842 Q205.324 248.842 208.38 253.449 Q211.459 258.032 211.459 266.782 Q211.459 275.509 208.38 280.115 Q205.324 284.699 199.514 284.699 Q193.704 284.699 190.625 280.115 Q187.57 275.509 187.57 266.782 Q187.57 258.032 190.625 253.449 Q193.704 248.842 199.514 248.842 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M229.676 267.615 Q226.343 267.615 224.422 269.398 Q222.523 271.18 222.523 274.305 Q222.523 277.43 224.422 279.213 Q226.343 280.995 229.676 280.995 Q233.01 280.995 234.931 279.213 Q236.852 277.407 236.852 274.305 Q236.852 271.18 234.931 269.398 Q233.033 267.615 229.676 267.615 M225 265.625 Q221.991 264.884 220.301 262.824 Q218.635 260.764 218.635 257.801 Q218.635 253.657 221.574 251.25 Q224.537 248.842 229.676 248.842 Q234.838 248.842 237.778 251.25 Q240.718 253.657 240.718 257.801 Q240.718 260.764 239.028 262.824 Q237.361 264.884 234.375 265.625 Q237.755 266.412 239.63 268.703 Q241.528 270.995 241.528 274.305 Q241.528 279.328 238.449 282.014 Q235.394 284.699 229.676 284.699 Q223.959 284.699 220.88 282.014 Q217.824 279.328 217.824 274.305 Q217.824 270.995 219.723 268.703 Q221.621 266.412 225 265.625 M223.287 258.241 Q223.287 260.926 224.954 262.43 Q226.644 263.935 229.676 263.935 Q232.685 263.935 234.375 262.43 Q236.088 260.926 236.088 258.241 Q236.088 255.555 234.375 254.051 Q232.685 252.546 229.676 252.546 Q226.644 252.546 224.954 254.051 Q223.287 255.555 223.287 258.241 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M116.066 152.543 L123.705 152.543 L123.705 126.177 L115.394 127.844 L115.394 123.584 L123.658 121.918 L128.334 121.918 L128.334 152.543 L135.973 152.543 L135.973 156.478 L116.066 156.478 L116.066 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M145.417 150.598 L150.302 150.598 L150.302 156.478 L145.417 156.478 L145.417 150.598 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M170.487 124.996 Q166.876 124.996 165.047 128.561 Q163.241 132.103 163.241 139.232 Q163.241 146.339 165.047 149.904 Q166.876 153.445 170.487 153.445 Q174.121 153.445 175.926 149.904 Q177.755 146.339 177.755 139.232 Q177.755 132.103 175.926 128.561 Q174.121 124.996 170.487 124.996 M170.487 121.293 Q176.297 121.293 179.352 125.899 Q182.431 130.483 182.431 139.232 Q182.431 147.959 179.352 152.566 Q176.297 157.149 170.487 157.149 Q164.677 157.149 161.598 152.566 Q158.542 147.959 158.542 139.232 Q158.542 130.483 161.598 125.899 Q164.677 121.293 170.487 121.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M191.459 152.543 L199.098 152.543 L199.098 126.177 L190.788 127.844 L190.788 123.584 L199.051 121.918 L203.727 121.918 L203.727 152.543 L211.366 152.543 L211.366 156.478 L191.459 156.478 L191.459 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M221.621 152.543 L229.26 152.543 L229.26 126.177 L220.949 127.844 L220.949 123.584 L229.213 121.918 L233.889 121.918 L233.889 152.543 L241.528 152.543 L241.528 156.478 L221.621 156.478 L221.621 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M22.818 438.814 L46.4666 447.535 L46.4666 430.061 L22.818 438.814 M16.4842 442.442 L16.4842 435.154 L64.0042 417.043 L64.0042 423.727 L51.8138 428.056 L51.8138 449.476 L64.0042 453.805 L64.0042 460.585 L16.4842 442.442 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M14.479 410.359 L14.479 404.503 L64.0042 404.503 L64.0042 410.359 L14.479 410.359 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M18.2347 386.456 L28.3562 386.456 L28.3562 374.393 L32.9077 374.393 L32.9077 386.456 L52.2594 386.456 Q56.6199 386.456 57.8613 385.278 Q59.1026 384.069 59.1026 380.408 L59.1026 374.393 L64.0042 374.393 L64.0042 380.408 Q64.0042 387.188 61.4897 389.766 Q58.9434 392.344 52.2594 392.344 L32.9077 392.344 L32.9077 396.641 L28.3562 396.641 L28.3562 392.344 L18.2347 392.344 L18.2347 386.456 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M28.3562 366.69 L28.3562 360.834 L64.0042 360.834 L64.0042 366.69 L28.3562 366.69 M14.479 366.69 L14.479 360.834 L21.895 360.834 L21.895 366.69 L14.479 366.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M18.2347 342.787 L28.3562 342.787 L28.3562 330.724 L32.9077 330.724 L32.9077 342.787 L52.2594 342.787 Q56.6199 342.787 57.8613 341.61 Q59.1026 340.4 59.1026 336.74 L59.1026 330.724 L64.0042 330.724 L64.0042 336.74 Q64.0042 343.519 61.4897 346.097 Q58.9434 348.675 52.2594 348.675 L32.9077 348.675 L32.9077 352.972 L28.3562 352.972 L28.3562 348.675 L18.2347 348.675 L18.2347 342.787 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M49.9359 323.626 L28.3562 323.626 L28.3562 317.77 L49.7131 317.77 Q54.7739 317.77 57.3202 315.797 Q59.8346 313.823 59.8346 309.876 Q59.8346 305.134 56.8109 302.397 Q53.7872 299.628 48.5673 299.628 L28.3562 299.628 L28.3562 293.771 L64.0042 293.771 L64.0042 299.628 L58.5296 299.628 Q61.7762 301.76 63.3676 304.593 Q64.9272 307.394 64.9272 311.118 Q64.9272 317.261 61.1078 320.444 Q57.2883 323.626 49.9359 323.626 M27.4968 308.89 L27.4968 308.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M33.7671 258.251 L14.479 258.251 L14.479 252.394 L64.0042 252.394 L64.0042 258.251 L58.657 258.251 Q61.8398 260.097 63.3994 262.929 Q64.9272 265.73 64.9272 269.677 Q64.9272 276.138 59.771 280.212 Q54.6147 284.254 46.212 284.254 Q37.8093 284.254 32.6531 280.212 Q27.4968 276.138 27.4968 269.677 Q27.4968 265.73 29.0564 262.929 Q30.5842 260.097 33.7671 258.251 M46.212 278.207 Q52.6732 278.207 56.3653 275.565 Q60.0256 272.892 60.0256 268.245 Q60.0256 263.598 56.3653 260.924 Q52.6732 258.251 46.212 258.251 Q39.7508 258.251 36.0905 260.924 Q32.3984 263.598 32.3984 268.245 Q32.3984 272.892 36.0905 275.565 Q39.7508 278.207 46.212 278.207 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M44.7161 209.839 L47.5806 209.839 L47.5806 236.766 Q53.6281 236.384 56.8109 233.138 Q59.9619 229.86 59.9619 224.035 Q59.9619 220.661 59.1344 217.51 Q58.3069 214.327 56.6518 211.208 L62.1899 211.208 Q63.5267 214.359 64.227 217.669 Q64.9272 220.979 64.9272 224.385 Q64.9272 232.915 59.9619 237.912 Q54.9967 242.877 46.5303 242.877 Q37.7774 242.877 32.6531 238.167 Q27.4968 233.424 27.4968 225.404 Q27.4968 218.21 32.1438 214.041 Q36.7589 209.839 44.7161 209.839 M42.9973 215.696 Q38.1912 215.759 35.3266 218.401 Q32.4621 221.011 32.4621 225.34 Q32.4621 230.241 35.2312 233.202 Q38.0002 236.13 43.0292 236.575 L42.9973 215.696 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip152)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.119,606.88 303.973,606.88 304.827,606.876 305.681,606.867 306.535,606.854 307.389,606.837 308.243,606.816 309.096,606.79 309.95,606.76 310.804,606.726 311.658,606.688 312.512,606.645 313.366,606.598 314.22,606.546 315.074,606.49 315.928,606.43 316.781,606.365 317.635,606.297 318.489,606.223 319.343,606.146 320.197,606.064 321.051,605.978 321.905,605.887 322.759,605.792 323.613,605.692 324.466,605.589 325.32,605.481 326.174,605.368 327.028,605.251 327.882,605.13 328.736,605.004 329.59,604.874 330.444,604.74 331.298,604.601 332.151,604.457 333.005,604.31 333.859,604.158 334.713,604.001 335.567,603.841 336.421,603.676 337.275,603.506 338.129,603.332 338.983,603.154 339.837,602.971 340.69,602.784 341.544,602.593 342.398,602.397 343.252,602.197 344.106,601.992 344.96,601.783 345.814,601.57 346.668,601.352 347.522,601.13 348.375,600.904 349.229,600.673 350.083,600.438 350.937,600.199 351.791,599.956 352.645,599.708 353.499,599.455 354.353,599.199 355.207,598.938 356.06,598.673 356.914,598.403 357.768,598.129 358.622,597.851 359.476,597.569 360.33,597.282 361.184,596.992 362.038,596.696 362.892,596.397 363.746,596.093 364.599,595.786 365.453,595.473 366.307,595.157 367.161,594.837 368.015,594.512 368.869,594.183 369.723,593.85 370.577,593.512 371.431,593.171 372.284,592.825 373.138,592.476 373.992,592.122 374.846,591.763 375.7,591.401 376.554,591.035 377.408,590.664 378.262,590.29 379.116,589.911 379.969,589.528 380.823,589.141 381.677,588.75 382.531,588.355 383.385,587.956 384.239,587.553 385.093,587.146 385.947,586.734 386.801,586.319 387.654,585.9 388.508,585.477 389.362,585.049 390.216,584.618 391.07,584.183 391.924,583.744 392.778,583.301 393.632,582.854 394.486,582.403 395.34,581.948 396.193,581.49 397.047,581.028 397.901,580.562 398.755,580.093 399.609,579.622 400.463,579.148 401.317,578.671 402.171,578.193 403.025,577.713 403.878,577.231 404.732,576.747 405.586,576.262 406.44,575.776 407.294,575.288 408.148,574.799 409.002,574.309 409.856,573.818 410.71,573.325 411.563,572.832 412.417,572.337 413.271,571.84 414.125,571.343 414.979,570.845 415.833,570.345 416.687,569.844 417.541,569.342 418.395,568.839 419.249,568.335 420.102,567.83 420.956,567.323 421.81,566.815 422.664,566.306 423.518,565.796 424.372,565.285 425.226,564.772 426.08,564.258 426.934,563.743 427.787,563.227 428.641,562.71 429.495,562.191 430.349,561.671 431.203,561.15 432.057,560.627 432.911,560.104 433.765,559.579 434.619,559.053 435.472,558.525 436.326,557.996 437.18,557.466 438.034,556.935 438.888,556.402 439.742,555.869 440.596,555.333 441.45,554.797 442.304,554.259 443.158,553.72 444.011,553.179 444.865,552.637 445.719,552.094 446.573,551.549 447.427,551.003 448.281,550.456 449.135,549.907 449.989,549.357 450.843,548.805 451.696,548.253 452.55,547.698 453.404,547.142 454.258,546.585 455.112,546.027 455.966,545.467 456.82,544.905 457.674,544.342 458.528,543.778 459.381,543.212 460.235,542.644 461.089,542.076 461.943,541.505 462.797,540.933 463.651,540.36 464.505,539.785 465.359,539.209 466.213,538.631 467.066,538.051 467.92,537.47 468.774,536.888 469.628,536.303 470.482,535.718 471.336,535.13 472.19,534.541 473.044,533.951 473.898,533.359 474.752,532.765 475.605,532.17 476.459,531.573 477.313,530.974 478.167,530.374 479.021,529.772 479.875,529.168 480.729,528.563 481.583,527.956 482.437,527.347 483.29,526.736 484.144,526.124 484.998,525.51 485.852,524.895 486.706,524.277 487.56,523.658 488.414,523.037 489.268,522.414 490.122,521.79 490.975,521.164 491.829,520.535 492.683,519.905 493.537,519.274 494.391,518.64 495.245,518.005 496.099,517.367 496.953,516.728 497.807,516.087 498.661,515.444 499.514,514.799 500.368,514.152 501.222,513.503 502.076,512.852 502.93,512.2 503.784,511.545 504.638,510.888 505.492,510.229 506.346,509.569 507.199,508.906 508.053,508.241 508.907,507.575 509.761,506.906 510.615,506.235 511.469,505.562 512.323,504.887 513.177,504.21 514.031,503.53 514.884,502.849 515.738,502.165 516.592,501.479 517.446,500.791 518.3,500.101 519.154,499.409 520.008,498.714 520.862,498.018 521.716,497.319 522.569,496.617 523.423,495.914 524.277,495.208 525.131,494.5 525.985,493.789 526.839,493.077 527.693,492.361 528.547,491.644 529.401,490.924 530.255,490.202 531.108,489.477 531.962,488.75 532.816,488.02 533.67,487.288 534.524,486.554 535.378,485.817 536.232,485.077 537.086,484.335 537.94,483.591 538.793,482.843 539.647,482.094 540.501,481.341 541.355,480.586 542.209,479.829 543.063,479.069 543.917,478.306 544.771,477.54 545.625,476.772 546.478,476.001 547.332,475.227 548.186,474.451 549.04,473.671 549.894,472.889 550.748,472.104 551.602,471.317 552.456,470.526 553.31,469.732 554.164,468.936 555.017,468.137 555.871,467.334 556.725,466.529 557.579,465.721 558.433,464.91 559.287,464.095 560.141,463.278 560.995,462.458 561.849,461.634 562.702,460.808 563.556,459.978 564.41,459.145 565.264,458.309 566.118,457.469 566.972,456.627 567.826,455.781 568.68,454.932 569.534,454.079 570.387,453.223 571.241,452.364 572.095,451.502 572.949,450.636 573.803,449.766 574.657,448.893 575.511,448.017 576.365,447.137 577.219,446.254 578.072,445.367 578.926,444.476 579.78,443.582 580.634,442.684 581.488,441.782 582.342,440.877 583.196,439.968 584.05,439.055 584.904,438.138 585.758,437.218 586.611,436.293 587.465,435.365 588.319,434.433 589.173,433.497 590.027,432.557 590.881,431.613 591.735,430.665 592.589,429.713 593.443,428.757 594.296,427.796 595.15,426.832 596.004,425.863 596.858,424.89 597.712,423.913 598.566,422.932 599.42,421.947 600.274,420.957 601.128,419.963 601.981,418.965 602.835,417.962 603.689,416.955 604.543,415.944 605.397,414.928 606.251,413.909 607.105,412.884 607.959,411.856 608.813,410.823 609.667,409.786 610.52,408.745 611.374,407.699 612.228,406.649 613.082,405.596 613.936,404.538 614.79,403.476 615.644,402.41 616.498,401.341 617.352,400.268 618.205,399.192 619.059,398.112 619.913,397.029 620.767,395.944 621.621,394.856 622.475,393.765 623.329,392.673 624.183,391.578 625.037,390.483 625.89,389.387 626.744,388.29 627.598,387.194 628.452,386.098 629.306,385.004 630.16,383.911 631.014,382.821 631.868,381.733 632.722,380.647 633.576,379.563 634.429,378.482 635.283,377.404 636.137,376.328 636.991,375.255 637.845,374.184 638.699,373.116 639.553,372.05 640.407,370.987 641.261,369.926 642.114,368.868 642.968,367.812 643.822,366.759 644.676,365.708 645.53,364.66 646.384,363.614 647.238,362.57 648.092,361.529 648.946,360.49 649.799,359.454 650.653,358.42 651.507,357.388 652.361,356.359 653.215,355.332 654.069,354.308 654.923,353.285 655.777,352.265 656.631,351.248 657.484,350.233 658.338,349.22 659.192,348.209 660.046,347.2 660.9,346.194 661.754,345.19 662.608,344.188 663.462,343.189 664.316,342.192 665.17,341.196 666.023,340.204 666.877,339.213 667.731,338.224 668.585,337.238 669.439,336.254 670.293,335.272 671.147,334.292 672.001,333.315 672.855,332.339 673.708,331.366 674.562,330.394 675.416,329.425 676.27,328.458 677.124,327.493 677.978,326.53 678.832,325.57 679.686,324.611 680.54,323.655 681.393,322.7 682.247,321.748 683.101,320.797 683.955,319.849 684.809,318.903 685.663,317.959 686.517,317.016 687.371,316.076 688.225,315.138 689.079,314.202 689.932,313.268 690.786,312.336 691.64,311.406 692.494,310.478 693.348,309.552 694.202,308.628 695.056,307.706 695.91,306.786 696.764,305.868 697.617,304.952 698.471,304.038 699.325,303.125 700.179,302.215 701.033,301.307 701.887,300.401 702.741,299.496 703.595,298.594 704.449,297.693 705.302,296.795 706.156,295.898 707.01,295.003 707.864,294.111 708.718,293.22 709.572,292.331 710.426,291.444 711.28,290.559 712.134,289.676 712.987,288.794 713.841,287.915 714.695,287.037 715.549,286.162 716.403,285.288 717.257,284.416 718.111,283.546 718.965,282.678 719.819,281.812 720.673,280.948 721.526,280.085 722.38,279.225 723.234,278.366 724.088,277.509 724.942,276.654 725.796,275.801 726.65,274.95 727.504,274.1 728.358,273.253 729.211,272.407 730.065,271.563 730.919,270.721 731.773,269.881 732.627,269.042 733.481,268.206 734.335,267.371 735.189,266.538 736.043,265.707 736.896,264.878 737.75,264.051 738.604,263.225 739.458,262.401 740.312,261.579 741.166,260.759 742.02,259.941 742.874,259.124 743.728,258.31 744.582,257.497 745.435,256.686 746.289,255.876 747.143,255.069 747.997,254.263 748.851,253.459 749.705,252.657 750.559,251.857 751.413,251.058 752.267,250.262 753.12,249.467 753.974,248.674 754.828,247.882 755.682,247.093 756.536,246.305 757.39,245.519 758.244,244.735 759.098,243.952 759.952,243.171 760.805,242.393 761.659,241.615 762.513,240.84 763.367,240.066 764.221,239.295 765.075,238.525 765.929,237.756 766.783,236.99 767.637,236.225 768.491,235.462 769.344,234.701 770.198,233.941 771.052,233.184 771.906,232.428 772.76,231.673 773.614,230.921 774.468,230.17 775.322,229.421 776.176,228.674 777.029,227.929 777.883,227.185 778.737,226.443 779.591,225.703 780.445,224.964 781.299,224.228 782.153,223.493 783.007,222.759 783.861,222.028 784.714,221.298 785.568,220.57 786.422,219.844 787.276,219.119 788.13,218.396 788.984,217.675 789.838,216.956 790.692,216.238 791.546,215.523 792.399,214.808 793.253,214.096 794.107,213.385 794.961,212.676 795.815,211.969 796.669,211.264 797.523,210.56 798.377,209.858 799.231,209.158 800.085,208.459 800.938,207.762 801.792,207.067 802.646,206.374 803.5,205.682 804.354,204.992 805.208,204.304 806.062,203.617 806.916,202.932 807.77,202.249 808.623,201.568 809.477,200.888 810.331,200.21 811.185,199.534 812.039,198.859 812.893,198.187 813.747,197.515 814.601,196.846 815.455,196.178 816.308,195.512 817.162,194.848 818.016,194.186 818.87,193.525 819.724,192.866 820.578,192.208 821.432,191.552 822.286,190.898 823.14,190.246 823.994,189.596 824.847,188.947 825.701,188.299 826.555,187.654 827.409,187.01 828.263,186.368 829.117,185.728 829.971,185.089 830.825,184.452 831.679,183.817 832.532,183.183 833.386,182.551 834.24,181.921 835.094,181.293 835.948,180.666 836.802,180.041 837.656,179.418 838.51,178.796 839.364,178.176 840.217,177.558 841.071,176.941 841.925,176.326 842.779,175.713 843.633,175.102 844.487,174.492 845.341,173.884 846.195,173.277 847.049,172.672 847.902,172.069 848.756,171.468 849.61,170.868 850.464,170.27 851.318,169.674 852.172,169.08 853.026,168.487 853.88,167.896 854.734,167.306 855.588,166.718 856.441,166.132 857.295,165.548 858.149,164.965 859.003,164.384 859.857,163.804 860.711,163.227 861.565,162.651 862.419,162.076 863.273,161.504 864.126,160.933 864.98,160.364 865.834,159.796 866.688,159.23 867.542,158.666 868.396,158.104 869.25,157.543 870.104,156.984 870.958,156.426 871.811,155.87 872.665,155.316 873.519,154.764 874.373,154.213 875.227,153.664 876.081,153.117 876.935,152.571 877.789,152.027 878.643,151.485 879.497,150.944 880.35,150.405 881.204,149.868 882.058,149.332 882.912,148.798 883.766,148.266 884.62,147.736 885.474,147.207 886.328,146.679 887.182,146.154 888.035,145.63 888.889,145.108 889.743,144.587 890.597,144.069 891.451,143.552 892.305,143.036 893.159,142.522 894.013,142.01 894.867,141.5 895.72,140.991 896.574,140.484 897.428,139.979 898.282,139.475 899.136,138.973 899.99,138.473 900.844,137.974 901.698,137.477 902.552,136.981 903.406,136.488 904.259,135.996 905.113,135.506 905.967,135.017 906.821,134.53 907.675,134.045 908.529,133.561 909.383,133.079 910.237,132.599 911.091,132.12 911.944,131.643 912.798,131.168 913.652,130.695 914.506,130.223 915.36,129.752 916.214,129.284 917.068,128.817 917.922,128.352 918.776,127.888 919.629,127.426 920.483,126.966 921.337,126.508 922.191,126.051 923.045,125.596 923.899,125.142 924.753,124.69 925.607,124.24 926.461,123.792 927.314,123.345 928.168,122.9 929.022,122.456 929.876,122.014 930.73,121.574 931.584,121.136 932.438,120.699 933.292,120.264 934.146,119.83 935,119.398 935.853,118.968 936.707,118.54 937.561,118.113 938.415,117.688 939.269,117.265 940.123,116.843 940.977,116.423 941.831,116.004 942.685,115.587 943.538,115.172 944.392,114.759 945.246,114.347 946.1,113.937 946.954,113.528 947.808,113.122 948.662,112.717 949.516,112.313 950.37,111.911 951.223,111.511 952.077,111.113 952.931,110.716 953.785,110.321 954.639,109.928 955.493,109.536 956.347,109.146 957.201,108.757 958.055,108.37 958.909,107.985 959.762,107.602 960.616,107.22 961.47,106.84 962.324,106.462 963.178,106.085 964.032,105.71 964.886,105.336 965.74,104.964 966.594,104.594 967.447,104.226 968.301,103.859 969.155,103.494 970.009,103.131 970.863,102.769 971.717,102.409 972.571,102.05 973.425,101.693 974.279,101.338 975.132,100.985 975.986,100.633 976.84,100.283 977.694,99.9343 978.548,99.5875 979.402,99.2424 980.256,98.8989 981.11,98.5571 981.964,98.2169 982.817,97.8784 983.671,97.5416 984.525,97.2064 985.379,96.8729 986.233,96.5411 987.087,96.2109 987.941,95.8824 988.795,95.5556 989.649,95.2304 990.503,94.9069 991.356,94.585 992.21,94.2648 993.064,93.9463 993.918,93.6294 994.772,93.3142 995.626,93.0007 996.48,92.6888 997.334,92.3786 998.188,92.07 999.041,91.7631 999.895,91.4579 1000.75,91.1543 1001.6,90.8524 1002.46,90.5522 1003.31,90.2536 1004.16,89.9567 1005.02,89.6614 1005.87,89.3678 1006.73,89.0759 1007.58,88.7856 1008.43,88.497 1009.29,88.21 1010.14,87.9248 1011,87.6411 1011.85,87.3592 1012.7,87.0789 1013.56,86.8002 1014.41,86.5232 1015.27,86.2479 1016.12,85.9743 1016.97,85.7023 1017.83,85.4319 1018.68,85.1633 1019.53,84.8962 1020.39,84.6309 1021.24,84.3672 1022.1,84.1052 1022.95,83.8448 1023.8,83.5861 1024.66,83.329 1025.51,83.0737 1026.37,82.8199 1027.22,82.5679 1028.07,82.3175 1028.93,82.0687 1029.78,81.8216 1030.64,81.5762 1031.49,81.3324 1032.34,81.0903 1033.2,80.8499 1034.05,80.6111 1034.9,80.374 1035.76,80.1385 1036.61,79.9047 1037.47,79.6726 1038.32,79.4421 1039.17,79.2133 1040.03,78.9861 1040.88,78.7606 1041.74,78.5368 1042.59,78.3146 1043.44,78.0941 1044.3,77.8752 1045.15,77.658 1046.01,77.4425 1046.86,77.2286 1047.71,77.0164 1048.57,76.8058 1049.42,76.5969 1050.27,76.3897 1051.13,76.1841 1051.98,75.9801 1052.84,75.7779 1053.69,75.5773 1054.54,75.3783 1055.4,75.181 1056.25,74.9854 1057.11,74.7914 1057.96,74.5991 1058.81,74.4085 1059.67,74.2195 1060.52,74.0322 1061.38,73.8465 1062.23,73.6625 1063.08,73.4801 1063.94,73.2994 1064.79,73.1204 1065.65,72.943 1066.5,72.7673 1067.35,72.5932 1068.21,72.4208 1069.06,72.2501 1069.91,72.081 1070.77,71.9136 1071.62,71.7478 1072.48,71.5837 1073.33,71.4212 1074.18,71.2605 1075.04,71.1013 1075.89,70.9439 1076.75,70.788 1077.6,70.6339 1078.45,70.4814 1079.31,70.3305 1080.16,70.1814 1081.02,70.0338 1081.87,69.888 1082.72,69.7438 1083.58,69.6012 1084.43,69.4603 1085.28,69.3211 1086.14,69.1835 1086.99,69.0476 1087.85,68.9134 1088.7,68.7808 1089.55,68.6498 1090.41,68.5206 1091.26,68.3929 1092.12,68.267 1092.97,68.1427 1093.82,68.02 1094.68,67.899 1095.53,67.7797 1096.39,67.662 1097.24,67.546 1098.09,67.4317 1098.95,67.319 1099.8,67.2079 1100.65,67.0985 1101.51,66.9908 1102.36,66.8847 1103.22,66.7803 1104.07,66.6776 1104.92,66.5765 1105.78,66.4771 1106.63,66.3793 1107.49,66.2832 1108.34,66.1887 1109.19,66.0959 1110.05,66.0047 1110.9,65.9153 1111.76,65.8274 1112.61,65.7413 1113.46,65.6567 1114.32,65.5739 1115.17,65.4927 1116.02,65.4131 1116.88,65.3352 1117.73,65.259 1118.59,65.1845 1119.44,65.1115 1120.29,65.0403 1121.15,64.9707 1122,64.9028 1122.86,64.8365 1123.71,64.7718 1124.56,64.7089 1125.42,64.6476 1126.27,64.5879 1127.13,64.5299 1127.98,64.4736 1128.83,64.4189 1129.69,64.3659 1130.54,64.3145 1131.39,64.2648 1132.25,64.2168 1133.1,64.1704 1133.96,64.1256 1134.81,64.0826 1135.66,64.0411 1136.52,64.0014 1137.37,63.9633 1138.23,63.9268 1139.08,63.892 1139.93,63.8589 1140.79,63.8274 1141.64,63.7976 1142.5,63.7694 1143.35,63.7429 1144.2,63.7181 1145.06,63.6949 1145.91,63.6734 1146.76,63.6535 1147.62,63.6353 1148.47,63.6187 1149.33,63.6038 1150.18,63.5905 1151.03,63.5789 1151.89,63.569 1152.74,63.5607 1153.6,63.5541 1154.45,63.5491 1155.3,63.5458 1156.16,63.5442 "/>
-<path clip-path="url(#clip150)" d="M1448.54 623.18 L2352.76 623.18 L2352.76 47.2441 L1448.54 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="302.265,623.18 302.265,47.2441 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="515.738,623.18 515.738,47.2441 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.211,623.18 729.211,47.2441 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="942.685,623.18 942.685,47.2441 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1156.16,623.18 1156.16,47.2441 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,623.18 1181.75,623.18 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="302.265,623.18 302.265,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="515.738,623.18 515.738,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.211,623.18 729.211,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="942.685,623.18 942.685,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1156.16,623.18 1156.16,604.282 "/>
+<path clip-path="url(#clip970)" d="M264.569 654.099 Q260.958 654.099 259.129 657.663 Q257.323 661.205 257.323 668.335 Q257.323 675.441 259.129 679.006 Q260.958 682.547 264.569 682.547 Q268.203 682.547 270.008 679.006 Q271.837 675.441 271.837 668.335 Q271.837 661.205 270.008 657.663 Q268.203 654.099 264.569 654.099 M264.569 650.395 Q270.379 650.395 273.434 655.001 Q276.513 659.585 276.513 668.335 Q276.513 677.061 273.434 681.668 Q270.379 686.251 264.569 686.251 Q258.759 686.251 255.68 681.668 Q252.624 677.061 252.624 668.335 Q252.624 659.585 255.68 655.001 Q258.759 650.395 264.569 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M284.731 679.7 L289.615 679.7 L289.615 685.58 L284.731 685.58 L284.731 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M309.8 654.099 Q306.189 654.099 304.36 657.663 Q302.555 661.205 302.555 668.335 Q302.555 675.441 304.36 679.006 Q306.189 682.547 309.8 682.547 Q313.434 682.547 315.24 679.006 Q317.068 675.441 317.068 668.335 Q317.068 661.205 315.24 657.663 Q313.434 654.099 309.8 654.099 M309.8 650.395 Q315.61 650.395 318.666 655.001 Q321.744 659.585 321.744 668.335 Q321.744 677.061 318.666 681.668 Q315.61 686.251 309.8 686.251 Q303.99 686.251 300.911 681.668 Q297.856 677.061 297.856 668.335 Q297.856 659.585 300.911 655.001 Q303.99 650.395 309.8 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M339.962 654.099 Q336.351 654.099 334.522 657.663 Q332.716 661.205 332.716 668.335 Q332.716 675.441 334.522 679.006 Q336.351 682.547 339.962 682.547 Q343.596 682.547 345.402 679.006 Q347.23 675.441 347.23 668.335 Q347.23 661.205 345.402 657.663 Q343.596 654.099 339.962 654.099 M339.962 650.395 Q345.772 650.395 348.828 655.001 Q351.906 659.585 351.906 668.335 Q351.906 677.061 348.828 681.668 Q345.772 686.251 339.962 686.251 Q334.152 686.251 331.073 681.668 Q328.017 677.061 328.017 668.335 Q328.017 659.585 331.073 655.001 Q334.152 650.395 339.962 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M478.539 654.099 Q474.928 654.099 473.1 657.663 Q471.294 661.205 471.294 668.335 Q471.294 675.441 473.1 679.006 Q474.928 682.547 478.539 682.547 Q482.174 682.547 483.979 679.006 Q485.808 675.441 485.808 668.335 Q485.808 661.205 483.979 657.663 Q482.174 654.099 478.539 654.099 M478.539 650.395 Q484.35 650.395 487.405 655.001 Q490.484 659.585 490.484 668.335 Q490.484 677.061 487.405 681.668 Q484.35 686.251 478.539 686.251 Q472.729 686.251 469.651 681.668 Q466.595 677.061 466.595 668.335 Q466.595 659.585 469.651 655.001 Q472.729 650.395 478.539 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M498.701 679.7 L503.586 679.7 L503.586 685.58 L498.701 685.58 L498.701 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M523.771 654.099 Q520.16 654.099 518.331 657.663 Q516.525 661.205 516.525 668.335 Q516.525 675.441 518.331 679.006 Q520.16 682.547 523.771 682.547 Q527.405 682.547 529.211 679.006 Q531.039 675.441 531.039 668.335 Q531.039 661.205 529.211 657.663 Q527.405 654.099 523.771 654.099 M523.771 650.395 Q529.581 650.395 532.636 655.001 Q535.715 659.585 535.715 668.335 Q535.715 677.061 532.636 681.668 Q529.581 686.251 523.771 686.251 Q517.961 686.251 514.882 681.668 Q511.826 677.061 511.826 668.335 Q511.826 659.585 514.882 655.001 Q517.961 650.395 523.771 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M543.979 651.02 L562.335 651.02 L562.335 654.955 L548.261 654.955 L548.261 663.427 Q549.28 663.08 550.298 662.918 Q551.317 662.733 552.335 662.733 Q558.122 662.733 561.502 665.904 Q564.882 669.075 564.882 674.492 Q564.882 680.071 561.409 683.172 Q557.937 686.251 551.618 686.251 Q549.442 686.251 547.173 685.881 Q544.928 685.51 542.521 684.77 L542.521 680.071 Q544.604 681.205 546.826 681.76 Q549.048 682.316 551.525 682.316 Q555.53 682.316 557.868 680.21 Q560.206 678.103 560.206 674.492 Q560.206 670.881 557.868 668.774 Q555.53 666.668 551.525 666.668 Q549.65 666.668 547.775 667.085 Q545.923 667.501 543.979 668.381 L543.979 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M691.515 654.099 Q687.904 654.099 686.075 657.663 Q684.27 661.205 684.27 668.335 Q684.27 675.441 686.075 679.006 Q687.904 682.547 691.515 682.547 Q695.149 682.547 696.955 679.006 Q698.783 675.441 698.783 668.335 Q698.783 661.205 696.955 657.663 Q695.149 654.099 691.515 654.099 M691.515 650.395 Q697.325 650.395 700.381 655.001 Q703.459 659.585 703.459 668.335 Q703.459 677.061 700.381 681.668 Q697.325 686.251 691.515 686.251 Q685.705 686.251 682.626 681.668 Q679.571 677.061 679.571 668.335 Q679.571 659.585 682.626 655.001 Q685.705 650.395 691.515 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M711.677 679.7 L716.561 679.7 L716.561 685.58 L711.677 685.58 L711.677 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M727.556 681.645 L735.195 681.645 L735.195 655.279 L726.885 656.946 L726.885 652.687 L735.149 651.02 L739.825 651.02 L739.825 681.645 L747.464 681.645 L747.464 685.58 L727.556 685.58 L727.556 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M766.908 654.099 Q763.297 654.099 761.468 657.663 Q759.663 661.205 759.663 668.335 Q759.663 675.441 761.468 679.006 Q763.297 682.547 766.908 682.547 Q770.542 682.547 772.348 679.006 Q774.176 675.441 774.176 668.335 Q774.176 661.205 772.348 657.663 Q770.542 654.099 766.908 654.099 M766.908 650.395 Q772.718 650.395 775.774 655.001 Q778.852 659.585 778.852 668.335 Q778.852 677.061 775.774 681.668 Q772.718 686.251 766.908 686.251 Q761.098 686.251 758.019 681.668 Q754.964 677.061 754.964 668.335 Q754.964 659.585 758.019 655.001 Q761.098 650.395 766.908 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M905.486 654.099 Q901.875 654.099 900.046 657.663 Q898.24 661.205 898.24 668.335 Q898.24 675.441 900.046 679.006 Q901.875 682.547 905.486 682.547 Q909.12 682.547 910.925 679.006 Q912.754 675.441 912.754 668.335 Q912.754 661.205 910.925 657.663 Q909.12 654.099 905.486 654.099 M905.486 650.395 Q911.296 650.395 914.351 655.001 Q917.43 659.585 917.43 668.335 Q917.43 677.061 914.351 681.668 Q911.296 686.251 905.486 686.251 Q899.676 686.251 896.597 681.668 Q893.541 677.061 893.541 668.335 Q893.541 659.585 896.597 655.001 Q899.676 650.395 905.486 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M925.648 679.7 L930.532 679.7 L930.532 685.58 L925.648 685.58 L925.648 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M941.527 681.645 L949.166 681.645 L949.166 655.279 L940.856 656.946 L940.856 652.687 L949.12 651.02 L953.796 651.02 L953.796 681.645 L961.434 681.645 L961.434 685.58 L941.527 685.58 L941.527 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M970.925 651.02 L989.282 651.02 L989.282 654.955 L975.208 654.955 L975.208 663.427 Q976.226 663.08 977.245 662.918 Q978.263 662.733 979.282 662.733 Q985.069 662.733 988.448 665.904 Q991.828 669.075 991.828 674.492 Q991.828 680.071 988.356 683.172 Q984.883 686.251 978.564 686.251 Q976.388 686.251 974.12 685.881 Q971.874 685.51 969.467 684.77 L969.467 680.071 Q971.55 681.205 973.772 681.76 Q975.995 682.316 978.471 682.316 Q982.476 682.316 984.814 680.21 Q987.152 678.103 987.152 674.492 Q987.152 670.881 984.814 668.774 Q982.476 666.668 978.471 666.668 Q976.596 666.668 974.721 667.085 Q972.87 667.501 970.925 668.381 L970.925 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1118.46 654.099 Q1114.85 654.099 1113.02 657.663 Q1111.22 661.205 1111.22 668.335 Q1111.22 675.441 1113.02 679.006 Q1114.85 682.547 1118.46 682.547 Q1122.1 682.547 1123.9 679.006 Q1125.73 675.441 1125.73 668.335 Q1125.73 661.205 1123.9 657.663 Q1122.1 654.099 1118.46 654.099 M1118.46 650.395 Q1124.27 650.395 1127.33 655.001 Q1130.41 659.585 1130.41 668.335 Q1130.41 677.061 1127.33 681.668 Q1124.27 686.251 1118.46 686.251 Q1112.65 686.251 1109.57 681.668 Q1106.52 677.061 1106.52 668.335 Q1106.52 659.585 1109.57 655.001 Q1112.65 650.395 1118.46 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1138.62 679.7 L1143.51 679.7 L1143.51 685.58 L1138.62 685.58 L1138.62 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1157.72 681.645 L1174.04 681.645 L1174.04 685.58 L1152.1 685.58 L1152.1 681.645 Q1154.76 678.89 1159.34 674.26 Q1163.95 669.608 1165.13 668.265 Q1167.37 665.742 1168.25 664.006 Q1169.16 662.247 1169.16 660.557 Q1169.16 657.802 1167.21 656.066 Q1165.29 654.33 1162.19 654.33 Q1159.99 654.33 1157.53 655.094 Q1155.1 655.858 1152.33 657.409 L1152.33 652.687 Q1155.15 651.552 1157.6 650.974 Q1160.06 650.395 1162.1 650.395 Q1167.47 650.395 1170.66 653.08 Q1173.85 655.765 1173.85 660.256 Q1173.85 662.386 1173.04 664.307 Q1172.26 666.205 1170.15 668.798 Q1169.57 669.469 1166.47 672.686 Q1163.37 675.881 1157.72 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1193.85 654.099 Q1190.24 654.099 1188.41 657.663 Q1186.61 661.205 1186.61 668.335 Q1186.61 675.441 1188.41 679.006 Q1190.24 682.547 1193.85 682.547 Q1197.49 682.547 1199.29 679.006 Q1201.12 675.441 1201.12 668.335 Q1201.12 661.205 1199.29 657.663 Q1197.49 654.099 1193.85 654.099 M1193.85 650.395 Q1199.66 650.395 1202.72 655.001 Q1205.8 659.585 1205.8 668.335 Q1205.8 677.061 1202.72 681.668 Q1199.66 686.251 1193.85 686.251 Q1188.04 686.251 1184.97 681.668 Q1181.91 677.061 1181.91 668.335 Q1181.91 659.585 1184.97 655.001 Q1188.04 650.395 1193.85 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M599.81 720.524 L640.009 720.524 L640.009 725.935 L623.14 725.935 L623.14 768.044 L616.679 768.044 L616.679 725.935 L599.81 725.935 L599.81 720.524 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M643.956 732.396 L649.812 732.396 L649.812 768.044 L643.956 768.044 L643.956 732.396 M643.956 718.519 L649.812 718.519 L649.812 725.935 L643.956 725.935 L643.956 718.519 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M689.821 739.239 Q692.017 735.292 695.073 733.414 Q698.128 731.537 702.266 731.537 Q707.836 731.537 710.86 735.451 Q713.883 739.335 713.883 746.528 L713.883 768.044 L707.995 768.044 L707.995 746.719 Q707.995 741.594 706.181 739.112 Q704.367 736.629 700.643 736.629 Q696.091 736.629 693.449 739.653 Q690.808 742.677 690.808 747.896 L690.808 768.044 L684.919 768.044 L684.919 746.719 Q684.919 741.563 683.105 739.112 Q681.291 736.629 677.503 736.629 Q673.015 736.629 670.374 739.685 Q667.732 742.708 667.732 747.896 L667.732 768.044 L661.844 768.044 L661.844 732.396 L667.732 732.396 L667.732 737.934 Q669.737 734.656 672.538 733.096 Q675.339 731.537 679.19 731.537 Q683.073 731.537 685.779 733.51 Q688.516 735.483 689.821 739.239 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M756.056 748.756 L756.056 751.62 L729.129 751.62 Q729.511 757.668 732.758 760.851 Q736.036 764.002 741.861 764.002 Q745.234 764.002 748.385 763.174 Q751.568 762.347 754.687 760.691 L754.687 766.23 Q751.536 767.566 748.226 768.267 Q744.916 768.967 741.51 768.967 Q732.98 768.967 727.983 764.002 Q723.018 759.036 723.018 750.57 Q723.018 741.817 727.729 736.693 Q732.471 731.537 740.492 731.537 Q747.685 731.537 751.855 736.183 Q756.056 740.799 756.056 748.756 M750.2 747.037 Q750.136 742.231 747.494 739.366 Q744.884 736.502 740.556 736.502 Q735.654 736.502 732.694 739.271 Q729.766 742.04 729.32 747.069 L750.2 747.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M800.457 718.582 Q796.192 725.903 794.123 733.064 Q792.054 740.226 792.054 747.578 Q792.054 754.93 794.123 762.156 Q796.224 769.349 800.457 776.638 L795.364 776.638 Q790.59 769.158 788.203 761.933 Q785.848 754.708 785.848 747.578 Q785.848 740.48 788.203 733.287 Q790.558 726.094 795.364 718.582 L800.457 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M834.545 733.446 L834.545 738.984 Q832.063 737.711 829.389 737.075 Q826.715 736.438 823.851 736.438 Q819.49 736.438 817.294 737.775 Q815.13 739.112 815.13 741.785 Q815.13 743.822 816.689 745 Q818.249 746.146 822.96 747.196 L824.965 747.642 Q831.203 748.979 833.813 751.429 Q836.455 753.848 836.455 758.209 Q836.455 763.174 832.508 766.07 Q828.593 768.967 821.718 768.967 Q818.854 768.967 815.735 768.394 Q812.647 767.853 809.21 766.739 L809.21 760.691 Q812.456 762.378 815.607 763.238 Q818.758 764.065 821.846 764.065 Q825.983 764.065 828.211 762.665 Q830.439 761.233 830.439 758.654 Q830.439 756.267 828.816 754.994 Q827.225 753.721 821.782 752.543 L819.745 752.066 Q814.302 750.92 811.883 748.565 Q809.464 746.178 809.464 742.04 Q809.464 737.011 813.029 734.274 Q816.594 731.537 823.151 731.537 Q826.397 731.537 829.262 732.014 Q832.126 732.491 834.545 733.446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M844.858 718.582 L849.95 718.582 Q854.725 726.094 857.08 733.287 Q859.467 740.48 859.467 747.578 Q859.467 754.708 857.08 761.933 Q854.725 769.158 849.95 776.638 L844.858 776.638 Q849.091 769.349 851.16 762.156 Q853.26 754.93 853.26 747.578 Q853.26 740.226 851.16 733.064 Q849.091 725.903 844.858 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,521.847 1181.75,521.847 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,394.297 1181.75,394.297 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,266.747 1181.75,266.747 "/>
+<polyline clip-path="url(#clip972)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,139.198 1181.75,139.198 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,623.18 277.528,47.2441 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,521.847 296.426,521.847 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,394.297 296.426,394.297 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,266.747 296.426,266.747 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,139.198 296.426,139.198 "/>
+<path clip-path="url(#clip970)" d="M116.436 535.192 L124.075 535.192 L124.075 508.826 L115.765 510.493 L115.765 506.233 L124.029 504.567 L128.705 504.567 L128.705 535.192 L136.343 535.192 L136.343 539.127 L116.436 539.127 L116.436 535.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M145.788 533.247 L150.672 533.247 L150.672 539.127 L145.788 539.127 L145.788 533.247 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M170.857 507.645 Q167.246 507.645 165.417 511.21 Q163.612 514.752 163.612 521.881 Q163.612 528.988 165.417 532.553 Q167.246 536.094 170.857 536.094 Q174.491 536.094 176.297 532.553 Q178.126 528.988 178.126 521.881 Q178.126 514.752 176.297 511.21 Q174.491 507.645 170.857 507.645 M170.857 503.942 Q176.667 503.942 179.723 508.548 Q182.801 513.131 182.801 521.881 Q182.801 530.608 179.723 535.215 Q176.667 539.798 170.857 539.798 Q165.047 539.798 161.968 535.215 Q158.913 530.608 158.913 521.881 Q158.913 513.131 161.968 508.548 Q165.047 503.942 170.857 503.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M201.019 507.645 Q197.408 507.645 195.579 511.21 Q193.774 514.752 193.774 521.881 Q193.774 528.988 195.579 532.553 Q197.408 536.094 201.019 536.094 Q204.653 536.094 206.459 532.553 Q208.287 528.988 208.287 521.881 Q208.287 514.752 206.459 511.21 Q204.653 507.645 201.019 507.645 M201.019 503.942 Q206.829 503.942 209.885 508.548 Q212.963 513.131 212.963 521.881 Q212.963 530.608 209.885 535.215 Q206.829 539.798 201.019 539.798 Q195.209 539.798 192.13 535.215 Q189.075 530.608 189.075 521.881 Q189.075 513.131 192.13 508.548 Q195.209 503.942 201.019 503.942 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M225.209 535.192 L241.528 535.192 L241.528 539.127 L219.584 539.127 L219.584 535.192 Q222.246 532.437 226.829 527.807 Q231.435 523.155 232.616 521.812 Q234.861 519.289 235.741 517.553 Q236.644 515.793 236.644 514.104 Q236.644 511.349 234.699 509.613 Q232.778 507.877 229.676 507.877 Q227.477 507.877 225.023 508.641 Q222.593 509.405 219.815 510.956 L219.815 506.233 Q222.639 505.099 225.093 504.52 Q227.547 503.942 229.584 503.942 Q234.954 503.942 238.148 506.627 Q241.343 509.312 241.343 513.803 Q241.343 515.932 240.533 517.854 Q239.746 519.752 237.639 522.344 Q237.06 523.016 233.959 526.233 Q230.857 529.428 225.209 535.192 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M115.834 407.642 L123.473 407.642 L123.473 381.276 L115.163 382.943 L115.163 378.684 L123.427 377.017 L128.103 377.017 L128.103 407.642 L135.742 407.642 L135.742 411.577 L115.834 411.577 L115.834 407.642 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M145.186 405.697 L150.07 405.697 L150.07 411.577 L145.186 411.577 L145.186 405.697 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M170.255 380.096 Q166.644 380.096 164.815 383.661 Q163.01 387.202 163.01 394.332 Q163.01 401.438 164.815 405.003 Q166.644 408.545 170.255 408.545 Q173.889 408.545 175.695 405.003 Q177.524 401.438 177.524 394.332 Q177.524 387.202 175.695 383.661 Q173.889 380.096 170.255 380.096 M170.255 376.392 Q176.065 376.392 179.121 380.999 Q182.2 385.582 182.2 394.332 Q182.2 403.059 179.121 407.665 Q176.065 412.248 170.255 412.248 Q164.445 412.248 161.366 407.665 Q158.311 403.059 158.311 394.332 Q158.311 385.582 161.366 380.999 Q164.445 376.392 170.255 376.392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M200.417 380.096 Q196.806 380.096 194.977 383.661 Q193.172 387.202 193.172 394.332 Q193.172 401.438 194.977 405.003 Q196.806 408.545 200.417 408.545 Q204.051 408.545 205.857 405.003 Q207.686 401.438 207.686 394.332 Q207.686 387.202 205.857 383.661 Q204.051 380.096 200.417 380.096 M200.417 376.392 Q206.227 376.392 209.283 380.999 Q212.361 385.582 212.361 394.332 Q212.361 403.059 209.283 407.665 Q206.227 412.248 200.417 412.248 Q194.607 412.248 191.528 407.665 Q188.473 403.059 188.473 394.332 Q188.473 385.582 191.528 380.999 Q194.607 376.392 200.417 376.392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M220.625 377.017 L238.982 377.017 L238.982 380.952 L224.908 380.952 L224.908 389.424 Q225.926 389.077 226.945 388.915 Q227.963 388.73 228.982 388.73 Q234.769 388.73 238.148 391.901 Q241.528 395.073 241.528 400.489 Q241.528 406.068 238.056 409.17 Q234.584 412.248 228.264 412.248 Q226.088 412.248 223.82 411.878 Q221.574 411.508 219.167 410.767 L219.167 406.068 Q221.25 407.202 223.473 407.758 Q225.695 408.313 228.172 408.313 Q232.176 408.313 234.514 406.207 Q236.852 404.1 236.852 400.489 Q236.852 396.878 234.514 394.772 Q232.176 392.665 228.172 392.665 Q226.297 392.665 224.422 393.082 Q222.57 393.498 220.625 394.378 L220.625 377.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M114.931 280.092 L122.57 280.092 L122.57 253.727 L114.26 255.393 L114.26 251.134 L122.524 249.467 L127.2 249.467 L127.2 280.092 L134.839 280.092 L134.839 284.027 L114.931 284.027 L114.931 280.092 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M144.283 278.148 L149.167 278.148 L149.167 284.027 L144.283 284.027 L144.283 278.148 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M169.352 252.546 Q165.741 252.546 163.913 256.111 Q162.107 259.653 162.107 266.782 Q162.107 273.889 163.913 277.453 Q165.741 280.995 169.352 280.995 Q172.987 280.995 174.792 277.453 Q176.621 273.889 176.621 266.782 Q176.621 259.653 174.792 256.111 Q172.987 252.546 169.352 252.546 M169.352 248.842 Q175.163 248.842 178.218 253.449 Q181.297 258.032 181.297 266.782 Q181.297 275.509 178.218 280.115 Q175.163 284.699 169.352 284.699 Q163.542 284.699 160.464 280.115 Q157.408 275.509 157.408 266.782 Q157.408 258.032 160.464 253.449 Q163.542 248.842 169.352 248.842 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M199.514 252.546 Q195.903 252.546 194.075 256.111 Q192.269 259.653 192.269 266.782 Q192.269 273.889 194.075 277.453 Q195.903 280.995 199.514 280.995 Q203.149 280.995 204.954 277.453 Q206.783 273.889 206.783 266.782 Q206.783 259.653 204.954 256.111 Q203.149 252.546 199.514 252.546 M199.514 248.842 Q205.324 248.842 208.38 253.449 Q211.459 258.032 211.459 266.782 Q211.459 275.509 208.38 280.115 Q205.324 284.699 199.514 284.699 Q193.704 284.699 190.625 280.115 Q187.57 275.509 187.57 266.782 Q187.57 258.032 190.625 253.449 Q193.704 248.842 199.514 248.842 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M229.676 267.615 Q226.343 267.615 224.422 269.398 Q222.523 271.18 222.523 274.305 Q222.523 277.43 224.422 279.213 Q226.343 280.995 229.676 280.995 Q233.01 280.995 234.931 279.213 Q236.852 277.407 236.852 274.305 Q236.852 271.18 234.931 269.398 Q233.033 267.615 229.676 267.615 M225 265.625 Q221.991 264.884 220.301 262.824 Q218.635 260.764 218.635 257.801 Q218.635 253.657 221.574 251.25 Q224.537 248.842 229.676 248.842 Q234.838 248.842 237.778 251.25 Q240.718 253.657 240.718 257.801 Q240.718 260.764 239.028 262.824 Q237.361 264.884 234.375 265.625 Q237.755 266.412 239.63 268.703 Q241.528 270.995 241.528 274.305 Q241.528 279.328 238.449 282.014 Q235.394 284.699 229.676 284.699 Q223.959 284.699 220.88 282.014 Q217.824 279.328 217.824 274.305 Q217.824 270.995 219.723 268.703 Q221.621 266.412 225 265.625 M223.287 258.241 Q223.287 260.926 224.954 262.43 Q226.644 263.935 229.676 263.935 Q232.685 263.935 234.375 262.43 Q236.088 260.926 236.088 258.241 Q236.088 255.555 234.375 254.051 Q232.685 252.546 229.676 252.546 Q226.644 252.546 224.954 254.051 Q223.287 255.555 223.287 258.241 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M116.066 152.543 L123.705 152.543 L123.705 126.177 L115.394 127.844 L115.394 123.584 L123.658 121.918 L128.334 121.918 L128.334 152.543 L135.973 152.543 L135.973 156.478 L116.066 156.478 L116.066 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M145.417 150.598 L150.302 150.598 L150.302 156.478 L145.417 156.478 L145.417 150.598 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M170.487 124.996 Q166.876 124.996 165.047 128.561 Q163.241 132.103 163.241 139.232 Q163.241 146.339 165.047 149.904 Q166.876 153.445 170.487 153.445 Q174.121 153.445 175.926 149.904 Q177.755 146.339 177.755 139.232 Q177.755 132.103 175.926 128.561 Q174.121 124.996 170.487 124.996 M170.487 121.293 Q176.297 121.293 179.352 125.899 Q182.431 130.483 182.431 139.232 Q182.431 147.959 179.352 152.566 Q176.297 157.149 170.487 157.149 Q164.677 157.149 161.598 152.566 Q158.542 147.959 158.542 139.232 Q158.542 130.483 161.598 125.899 Q164.677 121.293 170.487 121.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M191.459 152.543 L199.098 152.543 L199.098 126.177 L190.788 127.844 L190.788 123.584 L199.051 121.918 L203.727 121.918 L203.727 152.543 L211.366 152.543 L211.366 156.478 L191.459 156.478 L191.459 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M221.621 152.543 L229.26 152.543 L229.26 126.177 L220.949 127.844 L220.949 123.584 L229.213 121.918 L233.889 121.918 L233.889 152.543 L241.528 152.543 L241.528 156.478 L221.621 156.478 L221.621 152.543 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M22.818 438.814 L46.4666 447.535 L46.4666 430.061 L22.818 438.814 M16.4842 442.442 L16.4842 435.154 L64.0042 417.043 L64.0042 423.727 L51.8138 428.056 L51.8138 449.476 L64.0042 453.805 L64.0042 460.585 L16.4842 442.442 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M14.479 410.359 L14.479 404.503 L64.0042 404.503 L64.0042 410.359 L14.479 410.359 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M18.2347 386.456 L28.3562 386.456 L28.3562 374.393 L32.9077 374.393 L32.9077 386.456 L52.2594 386.456 Q56.6199 386.456 57.8613 385.278 Q59.1026 384.069 59.1026 380.408 L59.1026 374.393 L64.0042 374.393 L64.0042 380.408 Q64.0042 387.188 61.4897 389.766 Q58.9434 392.344 52.2594 392.344 L32.9077 392.344 L32.9077 396.641 L28.3562 396.641 L28.3562 392.344 L18.2347 392.344 L18.2347 386.456 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M28.3562 366.69 L28.3562 360.834 L64.0042 360.834 L64.0042 366.69 L28.3562 366.69 M14.479 366.69 L14.479 360.834 L21.895 360.834 L21.895 366.69 L14.479 366.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M18.2347 342.787 L28.3562 342.787 L28.3562 330.724 L32.9077 330.724 L32.9077 342.787 L52.2594 342.787 Q56.6199 342.787 57.8613 341.61 Q59.1026 340.4 59.1026 336.74 L59.1026 330.724 L64.0042 330.724 L64.0042 336.74 Q64.0042 343.519 61.4897 346.097 Q58.9434 348.675 52.2594 348.675 L32.9077 348.675 L32.9077 352.972 L28.3562 352.972 L28.3562 348.675 L18.2347 348.675 L18.2347 342.787 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M49.9359 323.626 L28.3562 323.626 L28.3562 317.77 L49.7131 317.77 Q54.7739 317.77 57.3202 315.797 Q59.8346 313.823 59.8346 309.876 Q59.8346 305.134 56.8109 302.397 Q53.7872 299.628 48.5673 299.628 L28.3562 299.628 L28.3562 293.771 L64.0042 293.771 L64.0042 299.628 L58.5296 299.628 Q61.7762 301.76 63.3676 304.593 Q64.9272 307.394 64.9272 311.118 Q64.9272 317.261 61.1078 320.444 Q57.2883 323.626 49.9359 323.626 M27.4968 308.89 L27.4968 308.89 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M33.7671 258.251 L14.479 258.251 L14.479 252.394 L64.0042 252.394 L64.0042 258.251 L58.657 258.251 Q61.8398 260.097 63.3994 262.929 Q64.9272 265.73 64.9272 269.677 Q64.9272 276.138 59.771 280.212 Q54.6147 284.254 46.212 284.254 Q37.8093 284.254 32.6531 280.212 Q27.4968 276.138 27.4968 269.677 Q27.4968 265.73 29.0564 262.929 Q30.5842 260.097 33.7671 258.251 M46.212 278.207 Q52.6732 278.207 56.3653 275.565 Q60.0256 272.892 60.0256 268.245 Q60.0256 263.598 56.3653 260.924 Q52.6732 258.251 46.212 258.251 Q39.7508 258.251 36.0905 260.924 Q32.3984 263.598 32.3984 268.245 Q32.3984 272.892 36.0905 275.565 Q39.7508 278.207 46.212 278.207 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M44.7161 209.839 L47.5806 209.839 L47.5806 236.766 Q53.6281 236.384 56.8109 233.138 Q59.9619 229.86 59.9619 224.035 Q59.9619 220.661 59.1344 217.51 Q58.3069 214.327 56.6518 211.208 L62.1899 211.208 Q63.5267 214.359 64.227 217.669 Q64.9272 220.979 64.9272 224.385 Q64.9272 232.915 59.9619 237.912 Q54.9967 242.877 46.5303 242.877 Q37.7774 242.877 32.6531 238.167 Q27.4968 233.424 27.4968 225.404 Q27.4968 218.21 32.1438 214.041 Q36.7589 209.839 44.7161 209.839 M42.9973 215.696 Q38.1912 215.759 35.3266 218.401 Q32.4621 221.011 32.4621 225.34 Q32.4621 230.241 35.2312 233.202 Q38.0002 236.13 43.0292 236.575 L42.9973 215.696 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip972)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.119,606.88 303.973,606.88 304.827,606.876 305.681,606.867 306.535,606.854 307.389,606.837 308.243,606.816 309.096,606.79 309.95,606.76 310.804,606.726 311.658,606.688 312.512,606.645 313.366,606.598 314.22,606.546 315.074,606.49 315.928,606.43 316.781,606.365 317.635,606.297 318.489,606.223 319.343,606.146 320.197,606.064 321.051,605.978 321.905,605.887 322.759,605.792 323.613,605.692 324.466,605.589 325.32,605.481 326.174,605.368 327.028,605.251 327.882,605.13 328.736,605.004 329.59,604.874 330.444,604.74 331.298,604.601 332.151,604.457 333.005,604.31 333.859,604.158 334.713,604.001 335.567,603.841 336.421,603.676 337.275,603.506 338.129,603.332 338.983,603.154 339.837,602.971 340.69,602.784 341.544,602.593 342.398,602.397 343.252,602.197 344.106,601.992 344.96,601.783 345.814,601.57 346.668,601.352 347.522,601.13 348.375,600.904 349.229,600.673 350.083,600.438 350.937,600.199 351.791,599.956 352.645,599.708 353.499,599.455 354.353,599.199 355.207,598.938 356.06,598.673 356.914,598.403 357.768,598.129 358.622,597.851 359.476,597.569 360.33,597.282 361.184,596.992 362.038,596.696 362.892,596.397 363.746,596.093 364.599,595.786 365.453,595.473 366.307,595.157 367.161,594.837 368.015,594.512 368.869,594.183 369.723,593.85 370.577,593.512 371.431,593.171 372.284,592.825 373.138,592.476 373.992,592.122 374.846,591.763 375.7,591.401 376.554,591.035 377.408,590.664 378.262,590.29 379.116,589.911 379.969,589.528 380.823,589.141 381.677,588.75 382.531,588.355 383.385,587.956 384.239,587.553 385.093,587.146 385.947,586.734 386.801,586.319 387.654,585.9 388.508,585.477 389.362,585.049 390.216,584.618 391.07,584.183 391.924,583.744 392.778,583.301 393.632,582.854 394.486,582.403 395.34,581.948 396.193,581.49 397.047,581.028 397.901,580.562 398.755,580.093 399.609,579.622 400.463,579.148 401.317,578.671 402.171,578.193 403.025,577.713 403.878,577.231 404.732,576.747 405.586,576.262 406.44,575.776 407.294,575.288 408.148,574.799 409.002,574.309 409.856,573.818 410.71,573.325 411.563,572.832 412.417,572.337 413.271,571.84 414.125,571.343 414.979,570.845 415.833,570.345 416.687,569.844 417.541,569.342 418.395,568.839 419.249,568.335 420.102,567.83 420.956,567.323 421.81,566.815 422.664,566.306 423.518,565.796 424.372,565.285 425.226,564.772 426.08,564.258 426.934,563.743 427.787,563.227 428.641,562.71 429.495,562.191 430.349,561.671 431.203,561.15 432.057,560.627 432.911,560.104 433.765,559.579 434.619,559.053 435.472,558.525 436.326,557.996 437.18,557.466 438.034,556.935 438.888,556.402 439.742,555.869 440.596,555.333 441.45,554.797 442.304,554.259 443.158,553.72 444.011,553.179 444.865,552.637 445.719,552.094 446.573,551.549 447.427,551.003 448.281,550.456 449.135,549.907 449.989,549.357 450.843,548.805 451.696,548.253 452.55,547.698 453.404,547.142 454.258,546.585 455.112,546.027 455.966,545.467 456.82,544.905 457.674,544.342 458.528,543.778 459.381,543.212 460.235,542.644 461.089,542.076 461.943,541.505 462.797,540.933 463.651,540.36 464.505,539.785 465.359,539.209 466.213,538.631 467.066,538.051 467.92,537.47 468.774,536.888 469.628,536.303 470.482,535.718 471.336,535.13 472.19,534.541 473.044,533.951 473.898,533.359 474.752,532.765 475.605,532.17 476.459,531.573 477.313,530.974 478.167,530.374 479.021,529.772 479.875,529.168 480.729,528.563 481.583,527.956 482.437,527.347 483.29,526.736 484.144,526.124 484.998,525.51 485.852,524.895 486.706,524.277 487.56,523.658 488.414,523.037 489.268,522.414 490.122,521.79 490.975,521.164 491.829,520.535 492.683,519.905 493.537,519.274 494.391,518.64 495.245,518.005 496.099,517.367 496.953,516.728 497.807,516.087 498.661,515.444 499.514,514.799 500.368,514.152 501.222,513.503 502.076,512.852 502.93,512.2 503.784,511.545 504.638,510.888 505.492,510.229 506.346,509.569 507.199,508.906 508.053,508.241 508.907,507.575 509.761,506.906 510.615,506.235 511.469,505.562 512.323,504.887 513.177,504.21 514.031,503.53 514.884,502.849 515.738,502.165 516.592,501.479 517.446,500.791 518.3,500.101 519.154,499.409 520.008,498.714 520.862,498.018 521.716,497.319 522.569,496.617 523.423,495.914 524.277,495.208 525.131,494.5 525.985,493.789 526.839,493.077 527.693,492.361 528.547,491.644 529.401,490.924 530.255,490.202 531.108,489.477 531.962,488.75 532.816,488.02 533.67,487.288 534.524,486.554 535.378,485.817 536.232,485.077 537.086,484.335 537.94,483.591 538.793,482.843 539.647,482.094 540.501,481.341 541.355,480.586 542.209,479.829 543.063,479.069 543.917,478.306 544.771,477.54 545.625,476.772 546.478,476.001 547.332,475.227 548.186,474.451 549.04,473.671 549.894,472.889 550.748,472.104 551.602,471.317 552.456,470.526 553.31,469.732 554.164,468.936 555.017,468.137 555.871,467.334 556.725,466.529 557.579,465.721 558.433,464.91 559.287,464.095 560.141,463.278 560.995,462.458 561.849,461.634 562.702,460.808 563.556,459.978 564.41,459.145 565.264,458.309 566.118,457.469 566.972,456.627 567.826,455.781 568.68,454.932 569.534,454.079 570.387,453.223 571.241,452.364 572.095,451.502 572.949,450.636 573.803,449.766 574.657,448.893 575.511,448.017 576.365,447.137 577.219,446.254 578.072,445.367 578.926,444.476 579.78,443.582 580.634,442.684 581.488,441.782 582.342,440.877 583.196,439.968 584.05,439.055 584.904,438.138 585.758,437.218 586.611,436.293 587.465,435.365 588.319,434.433 589.173,433.497 590.027,432.557 590.881,431.613 591.735,430.665 592.589,429.713 593.443,428.757 594.296,427.796 595.15,426.832 596.004,425.863 596.858,424.89 597.712,423.913 598.566,422.932 599.42,421.947 600.274,420.957 601.128,419.963 601.981,418.965 602.835,417.962 603.689,416.955 604.543,415.944 605.397,414.928 606.251,413.909 607.105,412.884 607.959,411.856 608.813,410.823 609.667,409.786 610.52,408.745 611.374,407.699 612.228,406.649 613.082,405.596 613.936,404.538 614.79,403.476 615.644,402.41 616.498,401.341 617.352,400.268 618.205,399.192 619.059,398.112 619.913,397.029 620.767,395.944 621.621,394.856 622.475,393.765 623.329,392.673 624.183,391.578 625.037,390.483 625.89,389.387 626.744,388.29 627.598,387.194 628.452,386.098 629.306,385.004 630.16,383.911 631.014,382.821 631.868,381.733 632.722,380.647 633.576,379.563 634.429,378.482 635.283,377.404 636.137,376.328 636.991,375.255 637.845,374.184 638.699,373.116 639.553,372.05 640.407,370.987 641.261,369.926 642.114,368.868 642.968,367.812 643.822,366.759 644.676,365.708 645.53,364.66 646.384,363.614 647.238,362.57 648.092,361.529 648.946,360.49 649.799,359.454 650.653,358.42 651.507,357.388 652.361,356.359 653.215,355.332 654.069,354.308 654.923,353.285 655.777,352.265 656.631,351.248 657.484,350.233 658.338,349.22 659.192,348.209 660.046,347.2 660.9,346.194 661.754,345.19 662.608,344.188 663.462,343.189 664.316,342.192 665.17,341.196 666.023,340.204 666.877,339.213 667.731,338.224 668.585,337.238 669.439,336.254 670.293,335.272 671.147,334.292 672.001,333.315 672.855,332.339 673.708,331.366 674.562,330.394 675.416,329.425 676.27,328.458 677.124,327.493 677.978,326.53 678.832,325.57 679.686,324.611 680.54,323.655 681.393,322.7 682.247,321.748 683.101,320.797 683.955,319.849 684.809,318.903 685.663,317.959 686.517,317.016 687.371,316.076 688.225,315.138 689.079,314.202 689.932,313.268 690.786,312.336 691.64,311.406 692.494,310.478 693.348,309.552 694.202,308.628 695.056,307.706 695.91,306.786 696.764,305.868 697.617,304.952 698.471,304.038 699.325,303.125 700.179,302.215 701.033,301.307 701.887,300.401 702.741,299.496 703.595,298.594 704.449,297.693 705.302,296.795 706.156,295.898 707.01,295.003 707.864,294.111 708.718,293.22 709.572,292.331 710.426,291.444 711.28,290.559 712.134,289.676 712.987,288.794 713.841,287.915 714.695,287.037 715.549,286.162 716.403,285.288 717.257,284.416 718.111,283.546 718.965,282.678 719.819,281.812 720.673,280.948 721.526,280.085 722.38,279.225 723.234,278.366 724.088,277.509 724.942,276.654 725.796,275.801 726.65,274.95 727.504,274.1 728.358,273.253 729.211,272.407 730.065,271.563 730.919,270.721 731.773,269.881 732.627,269.042 733.481,268.206 734.335,267.371 735.189,266.538 736.043,265.707 736.896,264.878 737.75,264.051 738.604,263.225 739.458,262.401 740.312,261.579 741.166,260.759 742.02,259.941 742.874,259.124 743.728,258.31 744.582,257.497 745.435,256.686 746.289,255.876 747.143,255.069 747.997,254.263 748.851,253.459 749.705,252.657 750.559,251.857 751.413,251.058 752.267,250.262 753.12,249.467 753.974,248.674 754.828,247.882 755.682,247.093 756.536,246.305 757.39,245.519 758.244,244.735 759.098,243.952 759.952,243.171 760.805,242.393 761.659,241.615 762.513,240.84 763.367,240.066 764.221,239.295 765.075,238.525 765.929,237.756 766.783,236.99 767.637,236.225 768.491,235.462 769.344,234.701 770.198,233.941 771.052,233.184 771.906,232.428 772.76,231.673 773.614,230.921 774.468,230.17 775.322,229.421 776.176,228.674 777.029,227.929 777.883,227.185 778.737,226.443 779.591,225.703 780.445,224.964 781.299,224.228 782.153,223.493 783.007,222.759 783.861,222.028 784.714,221.298 785.568,220.57 786.422,219.844 787.276,219.119 788.13,218.396 788.984,217.675 789.838,216.956 790.692,216.238 791.546,215.523 792.399,214.808 793.253,214.096 794.107,213.385 794.961,212.676 795.815,211.969 796.669,211.264 797.523,210.56 798.377,209.858 799.231,209.158 800.085,208.459 800.938,207.762 801.792,207.067 802.646,206.374 803.5,205.682 804.354,204.992 805.208,204.304 806.062,203.617 806.916,202.932 807.77,202.249 808.623,201.568 809.477,200.888 810.331,200.21 811.185,199.534 812.039,198.859 812.893,198.187 813.747,197.515 814.601,196.846 815.455,196.178 816.308,195.512 817.162,194.848 818.016,194.186 818.87,193.525 819.724,192.866 820.578,192.208 821.432,191.552 822.286,190.898 823.14,190.246 823.994,189.596 824.847,188.947 825.701,188.299 826.555,187.654 827.409,187.01 828.263,186.368 829.117,185.728 829.971,185.089 830.825,184.452 831.679,183.817 832.532,183.183 833.386,182.551 834.24,181.921 835.094,181.293 835.948,180.666 836.802,180.041 837.656,179.418 838.51,178.796 839.364,178.176 840.217,177.558 841.071,176.941 841.925,176.326 842.779,175.713 843.633,175.102 844.487,174.492 845.341,173.884 846.195,173.277 847.049,172.672 847.902,172.069 848.756,171.468 849.61,170.868 850.464,170.27 851.318,169.674 852.172,169.08 853.026,168.487 853.88,167.896 854.734,167.306 855.588,166.718 856.441,166.132 857.295,165.548 858.149,164.965 859.003,164.384 859.857,163.804 860.711,163.227 861.565,162.651 862.419,162.076 863.273,161.504 864.126,160.933 864.98,160.364 865.834,159.796 866.688,159.23 867.542,158.666 868.396,158.104 869.25,157.543 870.104,156.984 870.958,156.426 871.811,155.87 872.665,155.316 873.519,154.764 874.373,154.213 875.227,153.664 876.081,153.117 876.935,152.571 877.789,152.027 878.643,151.485 879.497,150.944 880.35,150.405 881.204,149.868 882.058,149.332 882.912,148.798 883.766,148.266 884.62,147.736 885.474,147.207 886.328,146.679 887.182,146.154 888.035,145.63 888.889,145.108 889.743,144.587 890.597,144.069 891.451,143.552 892.305,143.036 893.159,142.522 894.013,142.01 894.867,141.5 895.72,140.991 896.574,140.484 897.428,139.979 898.282,139.475 899.136,138.973 899.99,138.473 900.844,137.974 901.698,137.477 902.552,136.981 903.406,136.488 904.259,135.996 905.113,135.506 905.967,135.017 906.821,134.53 907.675,134.045 908.529,133.561 909.383,133.079 910.237,132.599 911.091,132.12 911.944,131.643 912.798,131.168 913.652,130.695 914.506,130.223 915.36,129.752 916.214,129.284 917.068,128.817 917.922,128.352 918.776,127.888 919.629,127.426 920.483,126.966 921.337,126.508 922.191,126.051 923.045,125.596 923.899,125.142 924.753,124.69 925.607,124.24 926.461,123.792 927.314,123.345 928.168,122.9 929.022,122.456 929.876,122.014 930.73,121.574 931.584,121.136 932.438,120.699 933.292,120.264 934.146,119.83 935,119.398 935.853,118.968 936.707,118.54 937.561,118.113 938.415,117.688 939.269,117.265 940.123,116.843 940.977,116.423 941.831,116.004 942.685,115.587 943.538,115.172 944.392,114.759 945.246,114.347 946.1,113.937 946.954,113.528 947.808,113.122 948.662,112.717 949.516,112.313 950.37,111.911 951.223,111.511 952.077,111.113 952.931,110.716 953.785,110.321 954.639,109.928 955.493,109.536 956.347,109.146 957.201,108.757 958.055,108.37 958.909,107.985 959.762,107.602 960.616,107.22 961.47,106.84 962.324,106.462 963.178,106.085 964.032,105.71 964.886,105.336 965.74,104.964 966.594,104.594 967.447,104.226 968.301,103.859 969.155,103.494 970.009,103.131 970.863,102.769 971.717,102.409 972.571,102.05 973.425,101.693 974.279,101.338 975.132,100.985 975.986,100.633 976.84,100.283 977.694,99.9343 978.548,99.5875 979.402,99.2424 980.256,98.8989 981.11,98.5571 981.964,98.2169 982.817,97.8784 983.671,97.5416 984.525,97.2064 985.379,96.8729 986.233,96.5411 987.087,96.2109 987.941,95.8824 988.795,95.5556 989.649,95.2304 990.503,94.9069 991.356,94.585 992.21,94.2648 993.064,93.9463 993.918,93.6294 994.772,93.3142 995.626,93.0007 996.48,92.6888 997.334,92.3786 998.188,92.07 999.041,91.7631 999.895,91.4579 1000.75,91.1543 1001.6,90.8524 1002.46,90.5522 1003.31,90.2536 1004.16,89.9567 1005.02,89.6614 1005.87,89.3678 1006.73,89.0759 1007.58,88.7856 1008.43,88.497 1009.29,88.21 1010.14,87.9248 1011,87.6411 1011.85,87.3592 1012.7,87.0789 1013.56,86.8002 1014.41,86.5232 1015.27,86.2479 1016.12,85.9743 1016.97,85.7023 1017.83,85.4319 1018.68,85.1633 1019.53,84.8962 1020.39,84.6309 1021.24,84.3672 1022.1,84.1052 1022.95,83.8448 1023.8,83.5861 1024.66,83.329 1025.51,83.0737 1026.37,82.8199 1027.22,82.5679 1028.07,82.3175 1028.93,82.0687 1029.78,81.8216 1030.64,81.5762 1031.49,81.3324 1032.34,81.0903 1033.2,80.8499 1034.05,80.6111 1034.9,80.374 1035.76,80.1385 1036.61,79.9047 1037.47,79.6726 1038.32,79.4421 1039.17,79.2133 1040.03,78.9861 1040.88,78.7606 1041.74,78.5368 1042.59,78.3146 1043.44,78.0941 1044.3,77.8752 1045.15,77.658 1046.01,77.4425 1046.86,77.2286 1047.71,77.0164 1048.57,76.8058 1049.42,76.5969 1050.27,76.3897 1051.13,76.1841 1051.98,75.9801 1052.84,75.7779 1053.69,75.5773 1054.54,75.3783 1055.4,75.181 1056.25,74.9854 1057.11,74.7914 1057.96,74.5991 1058.81,74.4085 1059.67,74.2195 1060.52,74.0322 1061.38,73.8465 1062.23,73.6625 1063.08,73.4801 1063.94,73.2994 1064.79,73.1204 1065.65,72.943 1066.5,72.7673 1067.35,72.5932 1068.21,72.4208 1069.06,72.2501 1069.91,72.081 1070.77,71.9136 1071.62,71.7478 1072.48,71.5837 1073.33,71.4212 1074.18,71.2605 1075.04,71.1013 1075.89,70.9439 1076.75,70.788 1077.6,70.6339 1078.45,70.4814 1079.31,70.3305 1080.16,70.1814 1081.02,70.0338 1081.87,69.888 1082.72,69.7438 1083.58,69.6012 1084.43,69.4603 1085.28,69.3211 1086.14,69.1835 1086.99,69.0476 1087.85,68.9134 1088.7,68.7808 1089.55,68.6498 1090.41,68.5206 1091.26,68.3929 1092.12,68.267 1092.97,68.1427 1093.82,68.02 1094.68,67.899 1095.53,67.7797 1096.39,67.662 1097.24,67.546 1098.09,67.4317 1098.95,67.319 1099.8,67.2079 1100.65,67.0985 1101.51,66.9908 1102.36,66.8847 1103.22,66.7803 1104.07,66.6776 1104.92,66.5765 1105.78,66.4771 1106.63,66.3793 1107.49,66.2832 1108.34,66.1887 1109.19,66.0959 1110.05,66.0047 1110.9,65.9153 1111.76,65.8274 1112.61,65.7413 1113.46,65.6567 1114.32,65.5739 1115.17,65.4927 1116.02,65.4131 1116.88,65.3352 1117.73,65.259 1118.59,65.1845 1119.44,65.1115 1120.29,65.0403 1121.15,64.9707 1122,64.9028 1122.86,64.8365 1123.71,64.7718 1124.56,64.7089 1125.42,64.6476 1126.27,64.5879 1127.13,64.5299 1127.98,64.4736 1128.83,64.4189 1129.69,64.3659 1130.54,64.3145 1131.39,64.2648 1132.25,64.2168 1133.1,64.1704 1133.96,64.1256 1134.81,64.0826 1135.66,64.0411 1136.52,64.0014 1137.37,63.9633 1138.23,63.9268 1139.08,63.892 1139.93,63.8589 1140.79,63.8274 1141.64,63.7976 1142.5,63.7694 1143.35,63.7429 1144.2,63.7181 1145.06,63.6949 1145.91,63.6734 1146.76,63.6535 1147.62,63.6353 1148.47,63.6187 1149.33,63.6038 1150.18,63.5905 1151.03,63.5789 1151.89,63.569 1152.74,63.5607 1153.6,63.5541 1154.45,63.5491 1155.3,63.5458 1156.16,63.5442 "/>
+<path clip-path="url(#clip970)" d="M1448.54 623.18 L2352.76 623.18 L2352.76 47.2441 L1448.54 47.2441  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip153">
+  <clipPath id="clip973">
     <rect x="1448" y="47" width="905" height="577"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1473.27,623.18 1473.27,47.2441 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1686.75,623.18 1686.75,47.2441 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1900.22,623.18 1900.22,47.2441 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2113.69,623.18 2113.69,47.2441 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2327.16,623.18 2327.16,47.2441 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,623.18 2352.76,623.18 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1473.27,623.18 1473.27,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1686.75,623.18 1686.75,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1900.22,623.18 1900.22,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2113.69,623.18 2113.69,604.282 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2327.16,623.18 2327.16,604.282 "/>
-<path clip-path="url(#clip150)" d="M1435.58 654.099 Q1431.96 654.099 1430.14 657.663 Q1428.33 661.205 1428.33 668.335 Q1428.33 675.441 1430.14 679.006 Q1431.96 682.547 1435.58 682.547 Q1439.21 682.547 1441.02 679.006 Q1442.84 675.441 1442.84 668.335 Q1442.84 661.205 1441.02 657.663 Q1439.21 654.099 1435.58 654.099 M1435.58 650.395 Q1441.39 650.395 1444.44 655.001 Q1447.52 659.585 1447.52 668.335 Q1447.52 677.061 1444.44 681.668 Q1441.39 686.251 1435.58 686.251 Q1429.77 686.251 1426.69 681.668 Q1423.63 677.061 1423.63 668.335 Q1423.63 659.585 1426.69 655.001 Q1429.77 650.395 1435.58 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1455.74 679.7 L1460.62 679.7 L1460.62 685.58 L1455.74 685.58 L1455.74 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1480.81 654.099 Q1477.2 654.099 1475.37 657.663 Q1473.56 661.205 1473.56 668.335 Q1473.56 675.441 1475.37 679.006 Q1477.2 682.547 1480.81 682.547 Q1484.44 682.547 1486.25 679.006 Q1488.08 675.441 1488.08 668.335 Q1488.08 661.205 1486.25 657.663 Q1484.44 654.099 1480.81 654.099 M1480.81 650.395 Q1486.62 650.395 1489.67 655.001 Q1492.75 659.585 1492.75 668.335 Q1492.75 677.061 1489.67 681.668 Q1486.62 686.251 1480.81 686.251 Q1475 686.251 1471.92 681.668 Q1468.86 677.061 1468.86 668.335 Q1468.86 659.585 1471.92 655.001 Q1475 650.395 1480.81 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1510.97 654.099 Q1507.36 654.099 1505.53 657.663 Q1503.72 661.205 1503.72 668.335 Q1503.72 675.441 1505.53 679.006 Q1507.36 682.547 1510.97 682.547 Q1514.6 682.547 1516.41 679.006 Q1518.24 675.441 1518.24 668.335 Q1518.24 661.205 1516.41 657.663 Q1514.6 654.099 1510.97 654.099 M1510.97 650.395 Q1516.78 650.395 1519.83 655.001 Q1522.91 659.585 1522.91 668.335 Q1522.91 677.061 1519.83 681.668 Q1516.78 686.251 1510.97 686.251 Q1505.16 686.251 1502.08 681.668 Q1499.02 677.061 1499.02 668.335 Q1499.02 659.585 1502.08 655.001 Q1505.16 650.395 1510.97 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1649.55 654.099 Q1645.94 654.099 1644.11 657.663 Q1642.3 661.205 1642.3 668.335 Q1642.3 675.441 1644.11 679.006 Q1645.94 682.547 1649.55 682.547 Q1653.18 682.547 1654.99 679.006 Q1656.82 675.441 1656.82 668.335 Q1656.82 661.205 1654.99 657.663 Q1653.18 654.099 1649.55 654.099 M1649.55 650.395 Q1655.36 650.395 1658.41 655.001 Q1661.49 659.585 1661.49 668.335 Q1661.49 677.061 1658.41 681.668 Q1655.36 686.251 1649.55 686.251 Q1643.74 686.251 1640.66 681.668 Q1637.6 677.061 1637.6 668.335 Q1637.6 659.585 1640.66 655.001 Q1643.74 650.395 1649.55 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1669.71 679.7 L1674.59 679.7 L1674.59 685.58 L1669.71 685.58 L1669.71 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1694.78 654.099 Q1691.17 654.099 1689.34 657.663 Q1687.53 661.205 1687.53 668.335 Q1687.53 675.441 1689.34 679.006 Q1691.17 682.547 1694.78 682.547 Q1698.41 682.547 1700.22 679.006 Q1702.05 675.441 1702.05 668.335 Q1702.05 661.205 1700.22 657.663 Q1698.41 654.099 1694.78 654.099 M1694.78 650.395 Q1700.59 650.395 1703.64 655.001 Q1706.72 659.585 1706.72 668.335 Q1706.72 677.061 1703.64 681.668 Q1700.59 686.251 1694.78 686.251 Q1688.97 686.251 1685.89 681.668 Q1682.83 677.061 1682.83 668.335 Q1682.83 659.585 1685.89 655.001 Q1688.97 650.395 1694.78 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1714.99 651.02 L1733.34 651.02 L1733.34 654.955 L1719.27 654.955 L1719.27 663.427 Q1720.29 663.08 1721.31 662.918 Q1722.32 662.733 1723.34 662.733 Q1729.13 662.733 1732.51 665.904 Q1735.89 669.075 1735.89 674.492 Q1735.89 680.071 1732.42 683.172 Q1728.94 686.251 1722.62 686.251 Q1720.45 686.251 1718.18 685.881 Q1715.94 685.51 1713.53 684.77 L1713.53 680.071 Q1715.61 681.205 1717.83 681.76 Q1720.06 682.316 1722.53 682.316 Q1726.54 682.316 1728.87 680.21 Q1731.21 678.103 1731.21 674.492 Q1731.21 670.881 1728.87 668.774 Q1726.54 666.668 1722.53 666.668 Q1720.66 666.668 1718.78 667.085 Q1716.93 667.501 1714.99 668.381 L1714.99 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1862.52 654.099 Q1858.91 654.099 1857.08 657.663 Q1855.28 661.205 1855.28 668.335 Q1855.28 675.441 1857.08 679.006 Q1858.91 682.547 1862.52 682.547 Q1866.16 682.547 1867.96 679.006 Q1869.79 675.441 1869.79 668.335 Q1869.79 661.205 1867.96 657.663 Q1866.16 654.099 1862.52 654.099 M1862.52 650.395 Q1868.33 650.395 1871.39 655.001 Q1874.47 659.585 1874.47 668.335 Q1874.47 677.061 1871.39 681.668 Q1868.33 686.251 1862.52 686.251 Q1856.71 686.251 1853.63 681.668 Q1850.58 677.061 1850.58 668.335 Q1850.58 659.585 1853.63 655.001 Q1856.71 650.395 1862.52 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1882.68 679.7 L1887.57 679.7 L1887.57 685.58 L1882.68 685.58 L1882.68 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1898.56 681.645 L1906.2 681.645 L1906.2 655.279 L1897.89 656.946 L1897.89 652.687 L1906.16 651.02 L1910.83 651.02 L1910.83 681.645 L1918.47 681.645 L1918.47 685.58 L1898.56 685.58 L1898.56 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1937.92 654.099 Q1934.3 654.099 1932.48 657.663 Q1930.67 661.205 1930.67 668.335 Q1930.67 675.441 1932.48 679.006 Q1934.3 682.547 1937.92 682.547 Q1941.55 682.547 1943.35 679.006 Q1945.18 675.441 1945.18 668.335 Q1945.18 661.205 1943.35 657.663 Q1941.55 654.099 1937.92 654.099 M1937.92 650.395 Q1943.73 650.395 1946.78 655.001 Q1949.86 659.585 1949.86 668.335 Q1949.86 677.061 1946.78 681.668 Q1943.73 686.251 1937.92 686.251 Q1932.1 686.251 1929.03 681.668 Q1925.97 677.061 1925.97 668.335 Q1925.97 659.585 1929.03 655.001 Q1932.1 650.395 1937.92 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2076.49 654.099 Q2072.88 654.099 2071.05 657.663 Q2069.25 661.205 2069.25 668.335 Q2069.25 675.441 2071.05 679.006 Q2072.88 682.547 2076.49 682.547 Q2080.13 682.547 2081.93 679.006 Q2083.76 675.441 2083.76 668.335 Q2083.76 661.205 2081.93 657.663 Q2080.13 654.099 2076.49 654.099 M2076.49 650.395 Q2082.3 650.395 2085.36 655.001 Q2088.44 659.585 2088.44 668.335 Q2088.44 677.061 2085.36 681.668 Q2082.3 686.251 2076.49 686.251 Q2070.68 686.251 2067.6 681.668 Q2064.55 677.061 2064.55 668.335 Q2064.55 659.585 2067.6 655.001 Q2070.68 650.395 2076.49 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2096.65 679.7 L2101.54 679.7 L2101.54 685.58 L2096.65 685.58 L2096.65 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2112.53 681.645 L2120.17 681.645 L2120.17 655.279 L2111.86 656.946 L2111.86 652.687 L2120.13 651.02 L2124.8 651.02 L2124.8 681.645 L2132.44 681.645 L2132.44 685.58 L2112.53 685.58 L2112.53 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2141.93 651.02 L2160.29 651.02 L2160.29 654.955 L2146.21 654.955 L2146.21 663.427 Q2147.23 663.08 2148.25 662.918 Q2149.27 662.733 2150.29 662.733 Q2156.08 662.733 2159.46 665.904 Q2162.83 669.075 2162.83 674.492 Q2162.83 680.071 2159.36 683.172 Q2155.89 686.251 2149.57 686.251 Q2147.4 686.251 2145.13 685.881 Q2142.88 685.51 2140.47 684.77 L2140.47 680.071 Q2142.56 681.205 2144.78 681.76 Q2147 682.316 2149.48 682.316 Q2153.48 682.316 2155.82 680.21 Q2158.16 678.103 2158.16 674.492 Q2158.16 670.881 2155.82 668.774 Q2153.48 666.668 2149.48 666.668 Q2147.6 666.668 2145.73 667.085 Q2143.88 667.501 2141.93 668.381 L2141.93 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2289.47 654.099 Q2285.86 654.099 2284.03 657.663 Q2282.22 661.205 2282.22 668.335 Q2282.22 675.441 2284.03 679.006 Q2285.86 682.547 2289.47 682.547 Q2293.1 682.547 2294.91 679.006 Q2296.74 675.441 2296.74 668.335 Q2296.74 661.205 2294.91 657.663 Q2293.1 654.099 2289.47 654.099 M2289.47 650.395 Q2295.28 650.395 2298.33 655.001 Q2301.41 659.585 2301.41 668.335 Q2301.41 677.061 2298.33 681.668 Q2295.28 686.251 2289.47 686.251 Q2283.66 686.251 2280.58 681.668 Q2277.52 677.061 2277.52 668.335 Q2277.52 659.585 2280.58 655.001 Q2283.66 650.395 2289.47 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2309.63 679.7 L2314.51 679.7 L2314.51 685.58 L2309.63 685.58 L2309.63 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2328.73 681.645 L2345.05 681.645 L2345.05 685.58 L2323.1 685.58 L2323.1 681.645 Q2325.76 678.89 2330.35 674.26 Q2334.95 669.608 2336.13 668.265 Q2338.38 665.742 2339.26 664.006 Q2340.16 662.247 2340.16 660.557 Q2340.16 657.802 2338.22 656.066 Q2336.3 654.33 2333.19 654.33 Q2331 654.33 2328.54 655.094 Q2326.11 655.858 2323.33 657.409 L2323.33 652.687 Q2326.16 651.552 2328.61 650.974 Q2331.07 650.395 2333.1 650.395 Q2338.47 650.395 2341.67 653.08 Q2344.86 655.765 2344.86 660.256 Q2344.86 662.386 2344.05 664.307 Q2343.26 666.205 2341.16 668.798 Q2340.58 669.469 2337.48 672.686 Q2334.38 675.881 2328.73 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2364.86 654.099 Q2361.25 654.099 2359.42 657.663 Q2357.62 661.205 2357.62 668.335 Q2357.62 675.441 2359.42 679.006 Q2361.25 682.547 2364.86 682.547 Q2368.5 682.547 2370.3 679.006 Q2372.13 675.441 2372.13 668.335 Q2372.13 661.205 2370.3 657.663 Q2368.5 654.099 2364.86 654.099 M2364.86 650.395 Q2370.67 650.395 2373.73 655.001 Q2376.81 659.585 2376.81 668.335 Q2376.81 677.061 2373.73 681.668 Q2370.67 686.251 2364.86 686.251 Q2359.05 686.251 2355.97 681.668 Q2352.92 677.061 2352.92 668.335 Q2352.92 659.585 2355.97 655.001 Q2359.05 650.395 2364.86 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1770.82 720.524 L1811.02 720.524 L1811.02 725.935 L1794.15 725.935 L1794.15 768.044 L1787.69 768.044 L1787.69 725.935 L1770.82 725.935 L1770.82 720.524 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1814.96 732.396 L1820.82 732.396 L1820.82 768.044 L1814.96 768.044 L1814.96 732.396 M1814.96 718.519 L1820.82 718.519 L1820.82 725.935 L1814.96 725.935 L1814.96 718.519 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1860.83 739.239 Q1863.02 735.292 1866.08 733.414 Q1869.14 731.537 1873.27 731.537 Q1878.84 731.537 1881.87 735.451 Q1884.89 739.335 1884.89 746.528 L1884.89 768.044 L1879 768.044 L1879 746.719 Q1879 741.594 1877.19 739.112 Q1875.37 736.629 1871.65 736.629 Q1867.1 736.629 1864.46 739.653 Q1861.81 742.677 1861.81 747.896 L1861.81 768.044 L1855.93 768.044 L1855.93 746.719 Q1855.93 741.563 1854.11 739.112 Q1852.3 736.629 1848.51 736.629 Q1844.02 736.629 1841.38 739.685 Q1838.74 742.708 1838.74 747.896 L1838.74 768.044 L1832.85 768.044 L1832.85 732.396 L1838.74 732.396 L1838.74 737.934 Q1840.74 734.656 1843.55 733.096 Q1846.35 731.537 1850.2 731.537 Q1854.08 731.537 1856.79 733.51 Q1859.52 735.483 1860.83 739.239 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1927.06 748.756 L1927.06 751.62 L1900.14 751.62 Q1900.52 757.668 1903.76 760.851 Q1907.04 764.002 1912.87 764.002 Q1916.24 764.002 1919.39 763.174 Q1922.58 762.347 1925.69 760.691 L1925.69 766.23 Q1922.54 767.566 1919.23 768.267 Q1915.92 768.967 1912.52 768.967 Q1903.99 768.967 1898.99 764.002 Q1894.03 759.036 1894.03 750.57 Q1894.03 741.817 1898.74 736.693 Q1903.48 731.537 1911.5 731.537 Q1918.69 731.537 1922.86 736.183 Q1927.06 740.799 1927.06 748.756 M1921.21 747.037 Q1921.14 742.231 1918.5 739.366 Q1915.89 736.502 1911.56 736.502 Q1906.66 736.502 1903.7 739.271 Q1900.77 742.04 1900.33 747.069 L1921.21 747.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1971.46 718.582 Q1967.2 725.903 1965.13 733.064 Q1963.06 740.226 1963.06 747.578 Q1963.06 754.93 1965.13 762.156 Q1967.23 769.349 1971.46 776.638 L1966.37 776.638 Q1961.6 769.158 1959.21 761.933 Q1956.85 754.708 1956.85 747.578 Q1956.85 740.48 1959.21 733.287 Q1961.57 726.094 1966.37 718.582 L1971.46 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2005.55 733.446 L2005.55 738.984 Q2003.07 737.711 2000.4 737.075 Q1997.72 736.438 1994.86 736.438 Q1990.5 736.438 1988.3 737.775 Q1986.14 739.112 1986.14 741.785 Q1986.14 743.822 1987.7 745 Q1989.26 746.146 1993.97 747.196 L1995.97 747.642 Q2002.21 748.979 2004.82 751.429 Q2007.46 753.848 2007.46 758.209 Q2007.46 763.174 2003.52 766.07 Q1999.6 768.967 1992.73 768.967 Q1989.86 768.967 1986.74 768.394 Q1983.65 767.853 1980.22 766.739 L1980.22 760.691 Q1983.46 762.378 1986.61 763.238 Q1989.77 764.065 1992.85 764.065 Q1996.99 764.065 1999.22 762.665 Q2001.45 761.233 2001.45 758.654 Q2001.45 756.267 1999.82 754.994 Q1998.23 753.721 1992.79 752.543 L1990.75 752.066 Q1985.31 750.92 1982.89 748.565 Q1980.47 746.178 1980.47 742.04 Q1980.47 737.011 1984.04 734.274 Q1987.6 731.537 1994.16 731.537 Q1997.4 731.537 2000.27 732.014 Q2003.13 732.491 2005.55 733.446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2015.86 718.582 L2020.96 718.582 Q2025.73 726.094 2028.09 733.287 Q2030.47 740.48 2030.47 747.578 Q2030.47 754.708 2028.09 761.933 Q2025.73 769.158 2020.96 776.638 L2015.86 776.638 Q2020.1 769.349 2022.17 762.156 Q2024.27 754.93 2024.27 747.578 Q2024.27 740.226 2022.17 733.064 Q2020.1 725.903 2015.86 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,606.88 2352.76,606.88 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,471.046 2352.76,471.046 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,335.212 2352.76,335.212 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,199.378 2352.76,199.378 "/>
-<polyline clip-path="url(#clip153)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,63.5442 2352.76,63.5442 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,623.18 1448.54,47.2441 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,606.88 1467.43,606.88 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,471.046 1467.43,471.046 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,335.212 1467.43,335.212 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,199.378 1467.43,199.378 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,63.5442 1467.43,63.5442 "/>
-<path clip-path="url(#clip150)" d="M1355.2 592.679 Q1351.59 592.679 1349.76 596.244 Q1347.95 599.785 1347.95 606.915 Q1347.95 614.021 1349.76 617.586 Q1351.59 621.128 1355.2 621.128 Q1358.83 621.128 1360.64 617.586 Q1362.47 614.021 1362.47 606.915 Q1362.47 599.785 1360.64 596.244 Q1358.83 592.679 1355.2 592.679 M1355.2 588.975 Q1361.01 588.975 1364.06 593.582 Q1367.14 598.165 1367.14 606.915 Q1367.14 615.642 1364.06 620.248 Q1361.01 624.831 1355.2 624.831 Q1349.39 624.831 1346.31 620.248 Q1343.25 615.642 1343.25 606.915 Q1343.25 598.165 1346.31 593.582 Q1349.39 588.975 1355.2 588.975 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1375.36 618.281 L1380.24 618.281 L1380.24 624.16 L1375.36 624.16 L1375.36 618.281 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1401.01 605.017 Q1397.86 605.017 1396.01 607.169 Q1394.18 609.322 1394.18 613.072 Q1394.18 616.799 1396.01 618.975 Q1397.86 621.128 1401.01 621.128 Q1404.16 621.128 1405.98 618.975 Q1407.84 616.799 1407.84 613.072 Q1407.84 609.322 1405.98 607.169 Q1404.16 605.017 1401.01 605.017 M1410.29 590.364 L1410.29 594.623 Q1408.53 593.79 1406.72 593.35 Q1404.94 592.91 1403.18 592.91 Q1398.55 592.91 1396.1 596.035 Q1393.67 599.16 1393.32 605.48 Q1394.69 603.466 1396.75 602.401 Q1398.81 601.313 1401.29 601.313 Q1406.49 601.313 1409.5 604.484 Q1412.54 607.632 1412.54 613.072 Q1412.54 618.396 1409.39 621.614 Q1406.24 624.831 1401.01 624.831 Q1395.01 624.831 1391.84 620.248 Q1388.67 615.642 1388.67 606.915 Q1388.67 598.72 1392.56 593.859 Q1396.45 588.975 1403 588.975 Q1404.76 588.975 1406.54 589.322 Q1408.35 589.67 1410.29 590.364 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1356.26 456.845 Q1352.65 456.845 1350.82 460.41 Q1349.02 463.951 1349.02 471.081 Q1349.02 478.187 1350.82 481.752 Q1352.65 485.294 1356.26 485.294 Q1359.9 485.294 1361.7 481.752 Q1363.53 478.187 1363.53 471.081 Q1363.53 463.951 1361.7 460.41 Q1359.9 456.845 1356.26 456.845 M1356.26 453.141 Q1362.07 453.141 1365.13 457.748 Q1368.21 462.331 1368.21 471.081 Q1368.21 479.808 1365.13 484.414 Q1362.07 488.997 1356.26 488.997 Q1350.45 488.997 1347.37 484.414 Q1344.32 479.808 1344.32 471.081 Q1344.32 462.331 1347.37 457.748 Q1350.45 453.141 1356.26 453.141 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1376.42 482.447 L1381.31 482.447 L1381.31 488.326 L1376.42 488.326 L1376.42 482.447 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1390.31 453.766 L1412.54 453.766 L1412.54 455.757 L1399.99 488.326 L1395.1 488.326 L1406.91 457.701 L1390.31 457.701 L1390.31 453.766 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1355.45 321.011 Q1351.84 321.011 1350.01 324.576 Q1348.21 328.117 1348.21 335.247 Q1348.21 342.353 1350.01 345.918 Q1351.84 349.46 1355.45 349.46 Q1359.09 349.46 1360.89 345.918 Q1362.72 342.353 1362.72 335.247 Q1362.72 328.117 1360.89 324.576 Q1359.09 321.011 1355.45 321.011 M1355.45 317.307 Q1361.26 317.307 1364.32 321.914 Q1367.4 326.497 1367.4 335.247 Q1367.4 343.974 1364.32 348.58 Q1361.26 353.163 1355.45 353.163 Q1349.64 353.163 1346.56 348.58 Q1343.51 343.974 1343.51 335.247 Q1343.51 326.497 1346.56 321.914 Q1349.64 317.307 1355.45 317.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1375.61 346.613 L1380.5 346.613 L1380.5 352.492 L1375.61 352.492 L1375.61 346.613 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1400.68 336.08 Q1397.35 336.08 1395.43 337.863 Q1393.53 339.645 1393.53 342.77 Q1393.53 345.895 1395.43 347.677 Q1397.35 349.46 1400.68 349.46 Q1404.02 349.46 1405.94 347.677 Q1407.86 345.872 1407.86 342.77 Q1407.86 339.645 1405.94 337.863 Q1404.04 336.08 1400.68 336.08 M1396.01 334.089 Q1393 333.349 1391.31 331.289 Q1389.64 329.228 1389.64 326.265 Q1389.64 322.122 1392.58 319.715 Q1395.54 317.307 1400.68 317.307 Q1405.85 317.307 1408.79 319.715 Q1411.72 322.122 1411.72 326.265 Q1411.72 329.228 1410.04 331.289 Q1408.37 333.349 1405.38 334.089 Q1408.76 334.877 1410.64 337.168 Q1412.54 339.46 1412.54 342.77 Q1412.54 347.793 1409.46 350.478 Q1406.4 353.163 1400.68 353.163 Q1394.97 353.163 1391.89 350.478 Q1388.83 347.793 1388.83 342.77 Q1388.83 339.46 1390.73 337.168 Q1392.63 334.877 1396.01 334.089 M1394.29 326.705 Q1394.29 329.39 1395.96 330.895 Q1397.65 332.4 1400.68 332.4 Q1403.69 332.4 1405.38 330.895 Q1407.1 329.39 1407.1 326.705 Q1407.1 324.02 1405.38 322.515 Q1403.69 321.011 1400.68 321.011 Q1397.65 321.011 1395.96 322.515 Q1394.29 324.02 1394.29 326.705 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1355.54 185.177 Q1351.93 185.177 1350.1 188.742 Q1348.3 192.283 1348.3 199.413 Q1348.3 206.519 1350.1 210.084 Q1351.93 213.626 1355.54 213.626 Q1359.18 213.626 1360.98 210.084 Q1362.81 206.519 1362.81 199.413 Q1362.81 192.283 1360.98 188.742 Q1359.18 185.177 1355.54 185.177 M1355.54 181.473 Q1361.35 181.473 1364.41 186.08 Q1367.49 190.663 1367.49 199.413 Q1367.49 208.14 1364.41 212.746 Q1361.35 217.329 1355.54 217.329 Q1349.73 217.329 1346.66 212.746 Q1343.6 208.14 1343.6 199.413 Q1343.6 190.663 1346.66 186.08 Q1349.73 181.473 1355.54 181.473 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1375.71 210.779 L1380.59 210.779 L1380.59 216.658 L1375.71 216.658 L1375.71 210.779 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1390.91 215.941 L1390.91 211.681 Q1392.67 212.515 1394.48 212.954 Q1396.29 213.394 1398.02 213.394 Q1402.65 213.394 1405.08 210.292 Q1407.54 207.167 1407.88 200.825 Q1406.54 202.816 1404.48 203.88 Q1402.42 204.945 1399.92 204.945 Q1394.73 204.945 1391.7 201.82 Q1388.69 198.672 1388.69 193.232 Q1388.69 187.908 1391.84 184.691 Q1394.99 181.473 1400.22 181.473 Q1406.22 181.473 1409.36 186.08 Q1412.54 190.663 1412.54 199.413 Q1412.54 207.584 1408.65 212.468 Q1404.78 217.329 1398.23 217.329 Q1396.47 217.329 1394.66 216.982 Q1392.86 216.635 1390.91 215.941 M1400.22 201.288 Q1403.37 201.288 1405.2 199.135 Q1407.05 196.982 1407.05 193.232 Q1407.05 189.506 1405.2 187.353 Q1403.37 185.177 1400.22 185.177 Q1397.07 185.177 1395.22 187.353 Q1393.39 189.506 1393.39 193.232 Q1393.39 196.982 1395.22 199.135 Q1397.07 201.288 1400.22 201.288 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1346.17 76.889 L1353.81 76.889 L1353.81 50.5234 L1345.5 52.1901 L1345.5 47.9308 L1353.76 46.2642 L1358.44 46.2642 L1358.44 76.889 L1366.08 76.889 L1366.08 80.8242 L1346.17 80.8242 L1346.17 76.889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1375.52 74.9446 L1380.41 74.9446 L1380.41 80.8242 L1375.52 80.8242 L1375.52 74.9446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1400.59 49.3429 Q1396.98 49.3429 1395.15 52.9076 Q1393.35 56.4493 1393.35 63.5789 Q1393.35 70.6853 1395.15 74.2501 Q1396.98 77.7918 1400.59 77.7918 Q1404.22 77.7918 1406.03 74.2501 Q1407.86 70.6853 1407.86 63.5789 Q1407.86 56.4493 1406.03 52.9076 Q1404.22 49.3429 1400.59 49.3429 M1400.59 45.6392 Q1406.4 45.6392 1409.46 50.2456 Q1412.54 54.8289 1412.54 63.5789 Q1412.54 72.3057 1409.46 76.9121 Q1406.4 81.4955 1400.59 81.4955 Q1394.78 81.4955 1391.7 76.9121 Q1388.65 72.3057 1388.65 63.5789 Q1388.65 54.8289 1391.7 50.2456 Q1394.78 45.6392 1400.59 45.6392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1245.48 412.476 L1245.48 402.895 L1277.81 390.769 L1245.48 378.578 L1245.48 368.998 L1293 368.998 L1293 375.268 L1251.27 375.268 L1283.86 387.522 L1283.86 393.983 L1251.27 406.237 L1293 406.237 L1293 412.476 L1245.48 412.476 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1275.08 340.289 Q1275.08 347.386 1276.7 350.124 Q1278.32 352.861 1282.24 352.861 Q1285.36 352.861 1287.2 350.824 Q1289.02 348.755 1289.02 345.222 Q1289.02 340.352 1285.58 337.424 Q1282.11 334.464 1276.38 334.464 L1275.08 334.464 L1275.08 340.289 M1272.66 328.608 L1293 328.608 L1293 334.464 L1287.59 334.464 Q1290.83 336.469 1292.39 339.461 Q1293.92 342.453 1293.92 346.782 Q1293.92 352.256 1290.86 355.503 Q1287.78 358.717 1282.62 358.717 Q1276.61 358.717 1273.55 354.707 Q1270.49 350.665 1270.49 342.676 L1270.49 334.464 L1269.92 334.464 Q1265.88 334.464 1263.68 337.138 Q1261.45 339.779 1261.45 344.585 Q1261.45 347.641 1262.19 350.537 Q1262.92 353.434 1264.38 356.107 L1258.97 356.107 Q1257.73 352.893 1257.13 349.869 Q1256.49 346.845 1256.49 343.981 Q1256.49 336.246 1260.5 332.427 Q1264.51 328.608 1272.66 328.608 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1258.4 293.819 L1263.94 293.819 Q1262.66 296.302 1262.03 298.975 Q1261.39 301.649 1261.39 304.513 Q1261.39 308.874 1262.73 311.07 Q1264.06 313.234 1266.74 313.234 Q1268.78 313.234 1269.95 311.675 Q1271.1 310.115 1272.15 305.405 L1272.59 303.399 Q1273.93 297.161 1276.38 294.551 Q1278.8 291.909 1283.16 291.909 Q1288.13 291.909 1291.02 295.856 Q1293.92 299.771 1293.92 306.646 Q1293.92 309.51 1293.35 312.63 Q1292.81 315.717 1291.69 319.154 L1285.64 319.154 Q1287.33 315.908 1288.19 312.757 Q1289.02 309.606 1289.02 306.519 Q1289.02 302.381 1287.62 300.153 Q1286.19 297.925 1283.61 297.925 Q1281.22 297.925 1279.95 299.548 Q1278.67 301.14 1277.5 306.582 L1277.02 308.619 Q1275.87 314.062 1273.52 316.481 Q1271.13 318.9 1266.99 318.9 Q1261.96 318.9 1259.23 315.335 Q1256.49 311.77 1256.49 305.214 Q1256.49 301.967 1256.97 299.102 Q1257.44 296.238 1258.4 293.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1258.4 259.858 L1263.94 259.858 Q1262.66 262.341 1262.03 265.014 Q1261.39 267.688 1261.39 270.552 Q1261.39 274.913 1262.73 277.109 Q1264.06 279.273 1266.74 279.273 Q1268.78 279.273 1269.95 277.714 Q1271.1 276.154 1272.15 271.444 L1272.59 269.438 Q1273.93 263.2 1276.38 260.59 Q1278.8 257.948 1283.16 257.948 Q1288.13 257.948 1291.02 261.895 Q1293.92 265.81 1293.92 272.685 Q1293.92 275.549 1293.35 278.669 Q1292.81 281.756 1291.69 285.193 L1285.64 285.193 Q1287.33 281.947 1288.19 278.796 Q1289.02 275.645 1289.02 272.558 Q1289.02 268.42 1287.62 266.192 Q1286.19 263.964 1283.61 263.964 Q1281.22 263.964 1279.95 265.587 Q1278.67 267.178 1277.5 272.621 L1277.02 274.658 Q1275.87 280.101 1273.52 282.52 Q1271.13 284.939 1266.99 284.939 Q1261.96 284.939 1259.23 281.374 Q1256.49 277.809 1256.49 271.253 Q1256.49 268.006 1256.97 265.141 Q1257.44 262.277 1258.4 259.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip153)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1474.13,63.5442 1474.98,65.4457 1475.83,67.3472 1476.69,69.2488 1477.54,71.1503 1478.4,73.0518 1479.25,74.9533 1480.1,76.8549 1480.96,78.7564 1481.81,80.6579 1482.67,82.5594 1483.52,84.4609 1484.37,86.3624 1485.23,88.2639 1486.08,90.1654 1486.93,92.0669 1487.79,93.9684 1488.64,95.8699 1489.5,97.7713 1490.35,99.6728 1491.2,101.574 1492.06,103.476 1492.91,105.377 1493.77,107.279 1494.62,109.18 1495.47,111.082 1496.33,112.983 1497.18,114.884 1498.04,116.786 1498.89,118.687 1499.74,120.589 1500.6,122.49 1501.45,124.392 1502.3,126.293 1503.16,128.194 1504.01,130.096 1504.87,131.997 1505.72,133.898 1506.57,135.8 1507.43,137.701 1508.28,139.602 1509.14,141.504 1509.99,143.405 1510.84,145.306 1511.7,147.208 1512.55,149.109 1513.41,151.01 1514.26,152.912 1515.11,154.813 1515.97,156.714 1516.82,158.615 1517.67,160.516 1518.53,162.418 1519.38,164.319 1520.24,166.22 1521.09,168.121 1521.94,170.022 1522.8,171.923 1523.65,173.824 1524.51,175.725 1525.36,177.626 1526.21,179.527 1527.07,181.428 1527.92,183.329 1528.78,185.23 1529.63,187.131 1530.48,189.032 1531.34,190.933 1532.19,192.833 1533.04,194.734 1533.9,196.635 1534.75,198.536 1535.61,200.436 1536.46,202.337 1537.31,204.237 1538.17,206.138 1539.02,208.038 1539.88,209.938 1540.73,211.838 1541.58,213.738 1542.44,215.638 1543.29,217.538 1544.15,219.438 1545,221.338 1545.85,223.237 1546.71,225.136 1547.56,227.036 1548.41,228.935 1549.27,230.833 1550.12,232.732 1550.98,234.63 1551.83,236.528 1552.68,238.426 1553.54,240.323 1554.39,242.22 1555.25,244.116 1556.1,246.011 1556.95,247.906 1557.81,249.8 1558.66,251.692 1559.52,253.584 1560.37,255.473 1561.22,257.361 1562.08,259.246 1562.93,261.127 1563.78,263.003 1564.64,264.871 1565.49,266.729 1566.35,268.57 1567.2,270.384 1568.05,272.151 1568.91,273.849 1569.76,275.461 1570.62,276.985 1571.47,278.433 1572.32,279.816 1573.18,281.146 1574.03,282.435 1574.89,283.689 1575.74,284.918 1576.59,286.126 1577.45,287.317 1578.3,288.495 1579.16,289.663 1580.01,290.824 1580.86,291.978 1581.72,293.127 1582.57,294.273 1583.42,295.415 1584.28,296.556 1585.13,297.694 1585.99,298.832 1586.84,299.968 1587.69,301.103 1588.55,302.238 1589.4,303.373 1590.26,304.507 1591.11,305.641 1591.96,306.775 1592.82,307.909 1593.67,309.043 1594.53,310.177 1595.38,311.311 1596.23,312.445 1597.09,313.58 1597.94,314.715 1598.79,315.849 1599.65,316.985 1600.5,318.12 1601.36,319.256 1602.21,320.392 1603.06,321.528 1603.92,322.664 1604.77,323.801 1605.63,324.938 1606.48,326.076 1607.33,327.214 1608.19,328.352 1609.04,329.49 1609.9,330.629 1610.75,331.768 1611.6,332.908 1612.46,334.048 1613.31,335.188 1614.16,336.329 1615.02,337.47 1615.87,338.612 1616.73,339.754 1617.58,340.896 1618.43,342.039 1619.29,343.182 1620.14,344.326 1621,345.471 1621.85,346.615 1622.7,347.761 1623.56,348.906 1624.41,350.053 1625.27,351.199 1626.12,352.347 1626.97,353.494 1627.83,354.643 1628.68,355.792 1629.53,356.941 1630.39,358.091 1631.24,359.242 1632.1,360.393 1632.95,361.545 1633.8,362.698 1634.66,363.851 1635.51,365.005 1636.37,366.159 1637.22,367.314 1638.07,368.47 1638.93,369.626 1639.78,370.783 1640.64,371.941 1641.49,373.1 1642.34,374.259 1643.2,375.419 1644.05,376.58 1644.9,377.741 1645.76,378.903 1646.61,380.066 1647.47,381.23 1648.32,382.395 1649.17,383.56 1650.03,384.726 1650.88,385.894 1651.74,387.062 1652.59,388.23 1653.44,389.4 1654.3,390.571 1655.15,391.742 1656.01,392.915 1656.86,394.088 1657.71,395.262 1658.57,396.437 1659.42,397.614 1660.27,398.791 1661.13,399.969 1661.98,401.148 1662.84,402.329 1663.69,403.51 1664.54,404.692 1665.4,405.876 1666.25,407.06 1667.11,408.246 1667.96,409.432 1668.81,410.62 1669.67,411.809 1670.52,412.999 1671.38,414.191 1672.23,415.383 1673.08,416.577 1673.94,417.772 1674.79,418.968 1675.64,420.165 1676.5,421.364 1677.35,422.564 1678.21,423.765 1679.06,424.968 1679.91,426.171 1680.77,427.377 1681.62,428.583 1682.48,429.791 1683.33,431.001 1684.18,432.212 1685.04,433.424 1685.89,434.638 1686.75,435.853 1687.6,437.07 1688.45,438.288 1689.31,439.508 1690.16,440.729 1691.01,441.952 1691.87,443.176 1692.72,444.402 1693.58,445.63 1694.43,446.86 1695.28,448.091 1696.14,449.323 1696.99,450.558 1697.85,451.794 1698.7,453.032 1699.55,454.272 1700.41,455.514 1701.26,456.757 1702.12,458.002 1702.97,459.249 1703.82,460.498 1704.68,461.749 1705.53,463.002 1706.38,464.257 1707.24,465.514 1708.09,466.773 1708.95,468.034 1709.8,469.297 1710.65,470.562 1711.51,471.829 1712.36,473.098 1713.22,474.369 1714.07,475.643 1714.92,476.919 1715.78,478.197 1716.63,479.477 1717.49,480.76 1718.34,482.045 1719.19,483.332 1720.05,484.622 1720.9,485.914 1721.76,487.209 1722.61,488.506 1723.46,489.805 1724.32,491.108 1725.17,492.412 1726.02,493.719 1726.88,495.029 1727.73,496.342 1728.59,497.657 1729.44,498.975 1730.29,500.295 1731.15,501.619 1732,502.945 1732.86,504.274 1733.71,505.606 1734.56,506.941 1735.42,508.279 1736.27,509.619 1737.13,510.963 1737.98,512.31 1738.83,513.66 1739.69,515.012 1740.54,516.368 1741.39,517.728 1742.25,519.09 1743.1,520.456 1743.96,521.825 1744.81,523.197 1745.66,524.572 1746.52,525.951 1747.37,527.333 1748.23,528.718 1749.08,530.107 1749.93,531.499 1750.79,532.895 1751.64,534.294 1752.5,535.697 1753.35,537.103 1754.2,538.512 1755.06,539.925 1755.91,541.341 1756.76,542.761 1757.62,544.184 1758.47,545.61 1759.33,547.039 1760.18,548.472 1761.03,549.908 1761.89,551.347 1762.74,552.789 1763.6,554.233 1764.45,555.681 1765.3,557.131 1766.16,558.583 1767.01,560.037 1767.87,561.493 1768.72,562.951 1769.57,564.41 1770.43,565.87 1771.28,567.331 1772.13,568.792 1772.99,570.252 1773.84,571.711 1774.7,573.169 1775.55,574.624 1776.4,576.076 1777.26,577.524 1778.11,578.967 1778.97,580.404 1779.82,581.833 1780.67,583.253 1781.53,584.663 1782.38,586.061 1783.24,587.444 1784.09,588.81 1784.94,590.158 1785.8,591.483 1786.65,592.784 1787.5,594.056 1788.36,595.296 1789.21,596.499 1790.07,597.66 1790.92,598.775 1791.77,599.838 1792.63,600.843 1793.48,601.782 1794.34,602.649 1795.19,603.437 1796.04,604.137 1796.9,604.742 1797.75,605.246 1798.61,605.643 1799.46,605.936 1800.31,606.136 1801.17,606.264 1802.02,606.349 1802.87,606.408 1803.73,606.452 1804.58,606.486 1805.44,606.513 1806.29,606.536 1807.14,606.555 1808,606.572 1808.85,606.587 1809.71,606.599 1810.56,606.611 1811.41,606.621 1812.27,606.63 1813.12,606.639 1813.98,606.646 1814.83,606.653 1815.68,606.66 1816.54,606.666 1817.39,606.671 1818.24,606.677 1819.1,606.681 1819.95,606.686 1820.81,606.69 1821.66,606.694 1822.51,606.698 1823.37,606.702 1824.22,606.705 1825.08,606.708 1825.93,606.711 1826.78,606.714 1827.64,606.717 1828.49,606.72 1829.35,606.722 1830.2,606.725 1831.05,606.727 1831.91,606.729 1832.76,606.731 1833.61,606.734 1834.47,606.736 1835.32,606.737 1836.18,606.739 1837.03,606.741 1837.88,606.743 1838.74,606.744 1839.59,606.746 1840.45,606.748 1841.3,606.749 1842.15,606.751 1843.01,606.752 1843.86,606.753 1844.72,606.755 1845.57,606.756 1846.42,606.757 1847.28,606.759 1848.13,606.76 1848.99,606.761 1849.84,606.762 1850.69,606.763 1851.55,606.764 1852.4,606.765 1853.25,606.766 1854.11,606.767 1854.96,606.768 1855.82,606.769 1856.67,606.77 1857.52,606.771 1858.38,606.772 1859.23,606.773 1860.09,606.774 1860.94,606.774 1861.79,606.775 1862.65,606.776 1863.5,606.777 1864.36,606.778 1865.21,606.778 1866.06,606.779 1866.92,606.78 1867.77,606.781 1868.62,606.781 1869.48,606.782 1870.33,606.783 1871.19,606.783 1872.04,606.784 1872.89,606.785 1873.75,606.785 1874.6,606.786 1875.46,606.787 1876.31,606.787 1877.16,606.788 1878.02,606.788 1878.87,606.789 1879.73,606.789 1880.58,606.79 1881.43,606.791 1882.29,606.791 1883.14,606.792 1883.99,606.792 1884.85,606.793 1885.7,606.793 1886.56,606.794 1887.41,606.794 1888.26,606.795 1889.12,606.795 1889.97,606.796 1890.83,606.796 1891.68,606.797 1892.53,606.797 1893.39,606.797 1894.24,606.798 1895.1,606.798 1895.95,606.799 1896.8,606.799 1897.66,606.8 1898.51,606.8 1899.36,606.8 1900.22,606.801 1901.07,606.801 1901.93,606.802 1902.78,606.802 1903.63,606.802 1904.49,606.803 1905.34,606.803 1906.2,606.804 1907.05,606.804 1907.9,606.804 1908.76,606.805 1909.61,606.805 1910.47,606.805 1911.32,606.806 1912.17,606.806 1913.03,606.806 1913.88,606.807 1914.73,606.807 1915.59,606.807 1916.44,606.808 1917.3,606.808 1918.15,606.808 1919,606.809 1919.86,606.809 1920.71,606.809 1921.57,606.81 1922.42,606.81 1923.27,606.81 1924.13,606.811 1924.98,606.811 1925.84,606.811 1926.69,606.812 1927.54,606.812 1928.4,606.812 1929.25,606.812 1930.1,606.813 1930.96,606.813 1931.81,606.813 1932.67,606.814 1933.52,606.814 1934.37,606.814 1935.23,606.814 1936.08,606.815 1936.94,606.815 1937.79,606.815 1938.64,606.815 1939.5,606.816 1940.35,606.816 1941.21,606.816 1942.06,606.817 1942.91,606.817 1943.77,606.817 1944.62,606.817 1945.47,606.818 1946.33,606.818 1947.18,606.818 1948.04,606.818 1948.89,606.819 1949.74,606.819 1950.6,606.819 1951.45,606.819 1952.31,606.819 1953.16,606.82 1954.01,606.82 1954.87,606.82 1955.72,606.82 1956.58,606.821 1957.43,606.821 1958.28,606.821 1959.14,606.821 1959.99,606.822 1960.84,606.822 1961.7,606.822 1962.55,606.822 1963.41,606.822 1964.26,606.823 1965.11,606.823 1965.97,606.823 1966.82,606.823 1967.68,606.823 1968.53,606.824 1969.38,606.824 1970.24,606.824 1971.09,606.824 1971.95,606.824 1972.8,606.825 1973.65,606.825 1974.51,606.825 1975.36,606.825 1976.21,606.825 1977.07,606.826 1977.92,606.826 1978.78,606.826 1979.63,606.826 1980.48,606.826 1981.34,606.827 1982.19,606.827 1983.05,606.827 1983.9,606.827 1984.75,606.827 1985.61,606.828 1986.46,606.828 1987.32,606.828 1988.17,606.828 1989.02,606.828 1989.88,606.829 1990.73,606.829 1991.59,606.829 1992.44,606.829 1993.29,606.829 1994.15,606.829 1995,606.83 1995.85,606.83 1996.71,606.83 1997.56,606.83 1998.42,606.83 1999.27,606.83 2000.12,606.831 2000.98,606.831 2001.83,606.831 2002.69,606.831 2003.54,606.831 2004.39,606.831 2005.25,606.832 2006.1,606.832 2006.96,606.832 2007.81,606.832 2008.66,606.832 2009.52,606.832 2010.37,606.833 2011.22,606.833 2012.08,606.833 2012.93,606.833 2013.79,606.833 2014.64,606.833 2015.49,606.834 2016.35,606.834 2017.2,606.834 2018.06,606.834 2018.91,606.834 2019.76,606.834 2020.62,606.835 2021.47,606.835 2022.33,606.835 2023.18,606.835 2024.03,606.835 2024.89,606.835 2025.74,606.835 2026.59,606.836 2027.45,606.836 2028.3,606.836 2029.16,606.836 2030.01,606.836 2030.86,606.836 2031.72,606.836 2032.57,606.837 2033.43,606.837 2034.28,606.837 2035.13,606.837 2035.99,606.837 2036.84,606.837 2037.7,606.837 2038.55,606.838 2039.4,606.838 2040.26,606.838 2041.11,606.838 2041.96,606.838 2042.82,606.838 2043.67,606.838 2044.53,606.839 2045.38,606.839 2046.23,606.839 2047.09,606.839 2047.94,606.839 2048.8,606.839 2049.65,606.839 2050.5,606.84 2051.36,606.84 2052.21,606.84 2053.07,606.84 2053.92,606.84 2054.77,606.84 2055.63,606.84 2056.48,606.841 2057.33,606.841 2058.19,606.841 2059.04,606.841 2059.9,606.841 2060.75,606.841 2061.6,606.841 2062.46,606.841 2063.31,606.842 2064.17,606.842 2065.02,606.842 2065.87,606.842 2066.73,606.842 2067.58,606.842 2068.44,606.842 2069.29,606.842 2070.14,606.843 2071,606.843 2071.85,606.843 2072.7,606.843 2073.56,606.843 2074.41,606.843 2075.27,606.843 2076.12,606.843 2076.97,606.844 2077.83,606.844 2078.68,606.844 2079.54,606.844 2080.39,606.844 2081.24,606.844 2082.1,606.844 2082.95,606.844 2083.81,606.845 2084.66,606.845 2085.51,606.845 2086.37,606.845 2087.22,606.845 2088.07,606.845 2088.93,606.845 2089.78,606.845 2090.64,606.846 2091.49,606.846 2092.34,606.846 2093.2,606.846 2094.05,606.846 2094.91,606.846 2095.76,606.846 2096.61,606.846 2097.47,606.846 2098.32,606.847 2099.18,606.847 2100.03,606.847 2100.88,606.847 2101.74,606.847 2102.59,606.847 2103.44,606.847 2104.3,606.847 2105.15,606.848 2106.01,606.848 2106.86,606.848 2107.71,606.848 2108.57,606.848 2109.42,606.848 2110.28,606.848 2111.13,606.848 2111.98,606.848 2112.84,606.849 2113.69,606.849 2114.55,606.849 2115.4,606.849 2116.25,606.849 2117.11,606.849 2117.96,606.849 2118.82,606.849 2119.67,606.849 2120.52,606.85 2121.38,606.85 2122.23,606.85 2123.08,606.85 2123.94,606.85 2124.79,606.85 2125.65,606.85 2126.5,606.85 2127.35,606.85 2128.21,606.851 2129.06,606.851 2129.92,606.851 2130.77,606.851 2131.62,606.851 2132.48,606.851 2133.33,606.851 2134.19,606.851 2135.04,606.851 2135.89,606.852 2136.75,606.852 2137.6,606.852 2138.45,606.852 2139.31,606.852 2140.16,606.852 2141.02,606.852 2141.87,606.852 2142.72,606.852 2143.58,606.853 2144.43,606.853 2145.29,606.853 2146.14,606.853 2146.99,606.853 2147.85,606.853 2148.7,606.853 2149.56,606.853 2150.41,606.853 2151.26,606.853 2152.12,606.854 2152.97,606.854 2153.82,606.854 2154.68,606.854 2155.53,606.854 2156.39,606.854 2157.24,606.854 2158.09,606.854 2158.95,606.854 2159.8,606.855 2160.66,606.855 2161.51,606.855 2162.36,606.855 2163.22,606.855 2164.07,606.855 2164.93,606.855 2165.78,606.855 2166.63,606.855 2167.49,606.855 2168.34,606.856 2169.19,606.856 2170.05,606.856 2170.9,606.856 2171.76,606.856 2172.61,606.856 2173.46,606.856 2174.32,606.856 2175.17,606.856 2176.03,606.856 2176.88,606.857 2177.73,606.857 2178.59,606.857 2179.44,606.857 2180.3,606.857 2181.15,606.857 2182,606.857 2182.86,606.857 2183.71,606.857 2184.56,606.857 2185.42,606.858 2186.27,606.858 2187.13,606.858 2187.98,606.858 2188.83,606.858 2189.69,606.858 2190.54,606.858 2191.4,606.858 2192.25,606.858 2193.1,606.859 2193.96,606.859 2194.81,606.859 2195.67,606.859 2196.52,606.859 2197.37,606.859 2198.23,606.859 2199.08,606.859 2199.93,606.859 2200.79,606.859 2201.64,606.86 2202.5,606.86 2203.35,606.86 2204.2,606.86 2205.06,606.86 2205.91,606.86 2206.77,606.86 2207.62,606.86 2208.47,606.86 2209.33,606.86 2210.18,606.861 2211.04,606.861 2211.89,606.861 2212.74,606.861 2213.6,606.861 2214.45,606.861 2215.3,606.861 2216.16,606.861 2217.01,606.861 2217.87,606.861 2218.72,606.862 2219.57,606.862 2220.43,606.862 2221.28,606.862 2222.14,606.862 2222.99,606.862 2223.84,606.862 2224.7,606.862 2225.55,606.862 2226.41,606.862 2227.26,606.863 2228.11,606.863 2228.97,606.863 2229.82,606.863 2230.67,606.863 2231.53,606.863 2232.38,606.863 2233.24,606.863 2234.09,606.863 2234.94,606.863 2235.8,606.864 2236.65,606.864 2237.51,606.864 2238.36,606.864 2239.21,606.864 2240.07,606.864 2240.92,606.864 2241.78,606.864 2242.63,606.864 2243.48,606.865 2244.34,606.865 2245.19,606.865 2246.04,606.865 2246.9,606.865 2247.75,606.865 2248.61,606.865 2249.46,606.865 2250.31,606.865 2251.17,606.865 2252.02,606.866 2252.88,606.866 2253.73,606.866 2254.58,606.866 2255.44,606.866 2256.29,606.866 2257.15,606.866 2258,606.866 2258.85,606.866 2259.71,606.867 2260.56,606.867 2261.42,606.867 2262.27,606.867 2263.12,606.867 2263.98,606.867 2264.83,606.867 2265.68,606.867 2266.54,606.867 2267.39,606.868 2268.25,606.868 2269.1,606.868 2269.95,606.868 2270.81,606.868 2271.66,606.868 2272.52,606.868 2273.37,606.868 2274.22,606.868 2275.08,606.869 2275.93,606.869 2276.79,606.869 2277.64,606.869 2278.49,606.869 2279.35,606.869 2280.2,606.869 2281.05,606.869 2281.91,606.87 2282.76,606.87 2283.62,606.87 2284.47,606.87 2285.32,606.87 2286.18,606.87 2287.03,606.87 2287.89,606.87 2288.74,606.871 2289.59,606.871 2290.45,606.871 2291.3,606.871 2292.16,606.871 2293.01,606.871 2293.86,606.871 2294.72,606.871 2295.57,606.872 2296.42,606.872 2297.28,606.872 2298.13,606.872 2298.99,606.872 2299.84,606.872 2300.69,606.872 2301.55,606.873 2302.4,606.873 2303.26,606.873 2304.11,606.873 2304.96,606.873 2305.82,606.873 2306.67,606.873 2307.53,606.874 2308.38,606.874 2309.23,606.874 2310.09,606.874 2310.94,606.874 2311.79,606.874 2312.65,606.875 2313.5,606.875 2314.36,606.875 2315.21,606.875 2316.06,606.875 2316.92,606.876 2317.77,606.876 2318.63,606.876 2319.48,606.876 2320.33,606.876 2321.19,606.877 2322.04,606.877 2322.9,606.877 2323.75,606.878 2324.6,606.878 2325.46,606.878 2326.31,606.879 2327.16,606.88 "/>
-<path clip-path="url(#clip150)" d="M277.528 1423.18 L1181.75 1423.18 L1181.75 847.244 L277.528 847.244  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1473.27,623.18 1473.27,47.2441 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1686.75,623.18 1686.75,47.2441 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1900.22,623.18 1900.22,47.2441 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2113.69,623.18 2113.69,47.2441 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2327.16,623.18 2327.16,47.2441 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,623.18 2352.76,623.18 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1473.27,623.18 1473.27,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1686.75,623.18 1686.75,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1900.22,623.18 1900.22,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2113.69,623.18 2113.69,604.282 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2327.16,623.18 2327.16,604.282 "/>
+<path clip-path="url(#clip970)" d="M1435.58 654.099 Q1431.96 654.099 1430.14 657.663 Q1428.33 661.205 1428.33 668.335 Q1428.33 675.441 1430.14 679.006 Q1431.96 682.547 1435.58 682.547 Q1439.21 682.547 1441.02 679.006 Q1442.84 675.441 1442.84 668.335 Q1442.84 661.205 1441.02 657.663 Q1439.21 654.099 1435.58 654.099 M1435.58 650.395 Q1441.39 650.395 1444.44 655.001 Q1447.52 659.585 1447.52 668.335 Q1447.52 677.061 1444.44 681.668 Q1441.39 686.251 1435.58 686.251 Q1429.77 686.251 1426.69 681.668 Q1423.63 677.061 1423.63 668.335 Q1423.63 659.585 1426.69 655.001 Q1429.77 650.395 1435.58 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1455.74 679.7 L1460.62 679.7 L1460.62 685.58 L1455.74 685.58 L1455.74 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1480.81 654.099 Q1477.2 654.099 1475.37 657.663 Q1473.56 661.205 1473.56 668.335 Q1473.56 675.441 1475.37 679.006 Q1477.2 682.547 1480.81 682.547 Q1484.44 682.547 1486.25 679.006 Q1488.08 675.441 1488.08 668.335 Q1488.08 661.205 1486.25 657.663 Q1484.44 654.099 1480.81 654.099 M1480.81 650.395 Q1486.62 650.395 1489.67 655.001 Q1492.75 659.585 1492.75 668.335 Q1492.75 677.061 1489.67 681.668 Q1486.62 686.251 1480.81 686.251 Q1475 686.251 1471.92 681.668 Q1468.86 677.061 1468.86 668.335 Q1468.86 659.585 1471.92 655.001 Q1475 650.395 1480.81 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1510.97 654.099 Q1507.36 654.099 1505.53 657.663 Q1503.72 661.205 1503.72 668.335 Q1503.72 675.441 1505.53 679.006 Q1507.36 682.547 1510.97 682.547 Q1514.6 682.547 1516.41 679.006 Q1518.24 675.441 1518.24 668.335 Q1518.24 661.205 1516.41 657.663 Q1514.6 654.099 1510.97 654.099 M1510.97 650.395 Q1516.78 650.395 1519.83 655.001 Q1522.91 659.585 1522.91 668.335 Q1522.91 677.061 1519.83 681.668 Q1516.78 686.251 1510.97 686.251 Q1505.16 686.251 1502.08 681.668 Q1499.02 677.061 1499.02 668.335 Q1499.02 659.585 1502.08 655.001 Q1505.16 650.395 1510.97 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1649.55 654.099 Q1645.94 654.099 1644.11 657.663 Q1642.3 661.205 1642.3 668.335 Q1642.3 675.441 1644.11 679.006 Q1645.94 682.547 1649.55 682.547 Q1653.18 682.547 1654.99 679.006 Q1656.82 675.441 1656.82 668.335 Q1656.82 661.205 1654.99 657.663 Q1653.18 654.099 1649.55 654.099 M1649.55 650.395 Q1655.36 650.395 1658.41 655.001 Q1661.49 659.585 1661.49 668.335 Q1661.49 677.061 1658.41 681.668 Q1655.36 686.251 1649.55 686.251 Q1643.74 686.251 1640.66 681.668 Q1637.6 677.061 1637.6 668.335 Q1637.6 659.585 1640.66 655.001 Q1643.74 650.395 1649.55 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1669.71 679.7 L1674.59 679.7 L1674.59 685.58 L1669.71 685.58 L1669.71 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1694.78 654.099 Q1691.17 654.099 1689.34 657.663 Q1687.53 661.205 1687.53 668.335 Q1687.53 675.441 1689.34 679.006 Q1691.17 682.547 1694.78 682.547 Q1698.41 682.547 1700.22 679.006 Q1702.05 675.441 1702.05 668.335 Q1702.05 661.205 1700.22 657.663 Q1698.41 654.099 1694.78 654.099 M1694.78 650.395 Q1700.59 650.395 1703.64 655.001 Q1706.72 659.585 1706.72 668.335 Q1706.72 677.061 1703.64 681.668 Q1700.59 686.251 1694.78 686.251 Q1688.97 686.251 1685.89 681.668 Q1682.83 677.061 1682.83 668.335 Q1682.83 659.585 1685.89 655.001 Q1688.97 650.395 1694.78 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1714.99 651.02 L1733.34 651.02 L1733.34 654.955 L1719.27 654.955 L1719.27 663.427 Q1720.29 663.08 1721.31 662.918 Q1722.32 662.733 1723.34 662.733 Q1729.13 662.733 1732.51 665.904 Q1735.89 669.075 1735.89 674.492 Q1735.89 680.071 1732.42 683.172 Q1728.94 686.251 1722.62 686.251 Q1720.45 686.251 1718.18 685.881 Q1715.94 685.51 1713.53 684.77 L1713.53 680.071 Q1715.61 681.205 1717.83 681.76 Q1720.06 682.316 1722.53 682.316 Q1726.54 682.316 1728.87 680.21 Q1731.21 678.103 1731.21 674.492 Q1731.21 670.881 1728.87 668.774 Q1726.54 666.668 1722.53 666.668 Q1720.66 666.668 1718.78 667.085 Q1716.93 667.501 1714.99 668.381 L1714.99 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1862.52 654.099 Q1858.91 654.099 1857.08 657.663 Q1855.28 661.205 1855.28 668.335 Q1855.28 675.441 1857.08 679.006 Q1858.91 682.547 1862.52 682.547 Q1866.16 682.547 1867.96 679.006 Q1869.79 675.441 1869.79 668.335 Q1869.79 661.205 1867.96 657.663 Q1866.16 654.099 1862.52 654.099 M1862.52 650.395 Q1868.33 650.395 1871.39 655.001 Q1874.47 659.585 1874.47 668.335 Q1874.47 677.061 1871.39 681.668 Q1868.33 686.251 1862.52 686.251 Q1856.71 686.251 1853.63 681.668 Q1850.58 677.061 1850.58 668.335 Q1850.58 659.585 1853.63 655.001 Q1856.71 650.395 1862.52 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1882.68 679.7 L1887.57 679.7 L1887.57 685.58 L1882.68 685.58 L1882.68 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1898.56 681.645 L1906.2 681.645 L1906.2 655.279 L1897.89 656.946 L1897.89 652.687 L1906.16 651.02 L1910.83 651.02 L1910.83 681.645 L1918.47 681.645 L1918.47 685.58 L1898.56 685.58 L1898.56 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1937.92 654.099 Q1934.3 654.099 1932.48 657.663 Q1930.67 661.205 1930.67 668.335 Q1930.67 675.441 1932.48 679.006 Q1934.3 682.547 1937.92 682.547 Q1941.55 682.547 1943.35 679.006 Q1945.18 675.441 1945.18 668.335 Q1945.18 661.205 1943.35 657.663 Q1941.55 654.099 1937.92 654.099 M1937.92 650.395 Q1943.73 650.395 1946.78 655.001 Q1949.86 659.585 1949.86 668.335 Q1949.86 677.061 1946.78 681.668 Q1943.73 686.251 1937.92 686.251 Q1932.1 686.251 1929.03 681.668 Q1925.97 677.061 1925.97 668.335 Q1925.97 659.585 1929.03 655.001 Q1932.1 650.395 1937.92 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2076.49 654.099 Q2072.88 654.099 2071.05 657.663 Q2069.25 661.205 2069.25 668.335 Q2069.25 675.441 2071.05 679.006 Q2072.88 682.547 2076.49 682.547 Q2080.13 682.547 2081.93 679.006 Q2083.76 675.441 2083.76 668.335 Q2083.76 661.205 2081.93 657.663 Q2080.13 654.099 2076.49 654.099 M2076.49 650.395 Q2082.3 650.395 2085.36 655.001 Q2088.44 659.585 2088.44 668.335 Q2088.44 677.061 2085.36 681.668 Q2082.3 686.251 2076.49 686.251 Q2070.68 686.251 2067.6 681.668 Q2064.55 677.061 2064.55 668.335 Q2064.55 659.585 2067.6 655.001 Q2070.68 650.395 2076.49 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2096.65 679.7 L2101.54 679.7 L2101.54 685.58 L2096.65 685.58 L2096.65 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2112.53 681.645 L2120.17 681.645 L2120.17 655.279 L2111.86 656.946 L2111.86 652.687 L2120.13 651.02 L2124.8 651.02 L2124.8 681.645 L2132.44 681.645 L2132.44 685.58 L2112.53 685.58 L2112.53 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2141.93 651.02 L2160.29 651.02 L2160.29 654.955 L2146.21 654.955 L2146.21 663.427 Q2147.23 663.08 2148.25 662.918 Q2149.27 662.733 2150.29 662.733 Q2156.08 662.733 2159.46 665.904 Q2162.83 669.075 2162.83 674.492 Q2162.83 680.071 2159.36 683.172 Q2155.89 686.251 2149.57 686.251 Q2147.4 686.251 2145.13 685.881 Q2142.88 685.51 2140.47 684.77 L2140.47 680.071 Q2142.56 681.205 2144.78 681.76 Q2147 682.316 2149.48 682.316 Q2153.48 682.316 2155.82 680.21 Q2158.16 678.103 2158.16 674.492 Q2158.16 670.881 2155.82 668.774 Q2153.48 666.668 2149.48 666.668 Q2147.6 666.668 2145.73 667.085 Q2143.88 667.501 2141.93 668.381 L2141.93 651.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2289.47 654.099 Q2285.86 654.099 2284.03 657.663 Q2282.22 661.205 2282.22 668.335 Q2282.22 675.441 2284.03 679.006 Q2285.86 682.547 2289.47 682.547 Q2293.1 682.547 2294.91 679.006 Q2296.74 675.441 2296.74 668.335 Q2296.74 661.205 2294.91 657.663 Q2293.1 654.099 2289.47 654.099 M2289.47 650.395 Q2295.28 650.395 2298.33 655.001 Q2301.41 659.585 2301.41 668.335 Q2301.41 677.061 2298.33 681.668 Q2295.28 686.251 2289.47 686.251 Q2283.66 686.251 2280.58 681.668 Q2277.52 677.061 2277.52 668.335 Q2277.52 659.585 2280.58 655.001 Q2283.66 650.395 2289.47 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2309.63 679.7 L2314.51 679.7 L2314.51 685.58 L2309.63 685.58 L2309.63 679.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2328.73 681.645 L2345.05 681.645 L2345.05 685.58 L2323.1 685.58 L2323.1 681.645 Q2325.76 678.89 2330.35 674.26 Q2334.95 669.608 2336.13 668.265 Q2338.38 665.742 2339.26 664.006 Q2340.16 662.247 2340.16 660.557 Q2340.16 657.802 2338.22 656.066 Q2336.3 654.33 2333.19 654.33 Q2331 654.33 2328.54 655.094 Q2326.11 655.858 2323.33 657.409 L2323.33 652.687 Q2326.16 651.552 2328.61 650.974 Q2331.07 650.395 2333.1 650.395 Q2338.47 650.395 2341.67 653.08 Q2344.86 655.765 2344.86 660.256 Q2344.86 662.386 2344.05 664.307 Q2343.26 666.205 2341.16 668.798 Q2340.58 669.469 2337.48 672.686 Q2334.38 675.881 2328.73 681.645 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2364.86 654.099 Q2361.25 654.099 2359.42 657.663 Q2357.62 661.205 2357.62 668.335 Q2357.62 675.441 2359.42 679.006 Q2361.25 682.547 2364.86 682.547 Q2368.5 682.547 2370.3 679.006 Q2372.13 675.441 2372.13 668.335 Q2372.13 661.205 2370.3 657.663 Q2368.5 654.099 2364.86 654.099 M2364.86 650.395 Q2370.67 650.395 2373.73 655.001 Q2376.81 659.585 2376.81 668.335 Q2376.81 677.061 2373.73 681.668 Q2370.67 686.251 2364.86 686.251 Q2359.05 686.251 2355.97 681.668 Q2352.92 677.061 2352.92 668.335 Q2352.92 659.585 2355.97 655.001 Q2359.05 650.395 2364.86 650.395 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1770.82 720.524 L1811.02 720.524 L1811.02 725.935 L1794.15 725.935 L1794.15 768.044 L1787.69 768.044 L1787.69 725.935 L1770.82 725.935 L1770.82 720.524 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1814.96 732.396 L1820.82 732.396 L1820.82 768.044 L1814.96 768.044 L1814.96 732.396 M1814.96 718.519 L1820.82 718.519 L1820.82 725.935 L1814.96 725.935 L1814.96 718.519 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1860.83 739.239 Q1863.02 735.292 1866.08 733.414 Q1869.14 731.537 1873.27 731.537 Q1878.84 731.537 1881.87 735.451 Q1884.89 739.335 1884.89 746.528 L1884.89 768.044 L1879 768.044 L1879 746.719 Q1879 741.594 1877.19 739.112 Q1875.37 736.629 1871.65 736.629 Q1867.1 736.629 1864.46 739.653 Q1861.81 742.677 1861.81 747.896 L1861.81 768.044 L1855.93 768.044 L1855.93 746.719 Q1855.93 741.563 1854.11 739.112 Q1852.3 736.629 1848.51 736.629 Q1844.02 736.629 1841.38 739.685 Q1838.74 742.708 1838.74 747.896 L1838.74 768.044 L1832.85 768.044 L1832.85 732.396 L1838.74 732.396 L1838.74 737.934 Q1840.74 734.656 1843.55 733.096 Q1846.35 731.537 1850.2 731.537 Q1854.08 731.537 1856.79 733.51 Q1859.52 735.483 1860.83 739.239 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1927.06 748.756 L1927.06 751.62 L1900.14 751.62 Q1900.52 757.668 1903.76 760.851 Q1907.04 764.002 1912.87 764.002 Q1916.24 764.002 1919.39 763.174 Q1922.58 762.347 1925.69 760.691 L1925.69 766.23 Q1922.54 767.566 1919.23 768.267 Q1915.92 768.967 1912.52 768.967 Q1903.99 768.967 1898.99 764.002 Q1894.03 759.036 1894.03 750.57 Q1894.03 741.817 1898.74 736.693 Q1903.48 731.537 1911.5 731.537 Q1918.69 731.537 1922.86 736.183 Q1927.06 740.799 1927.06 748.756 M1921.21 747.037 Q1921.14 742.231 1918.5 739.366 Q1915.89 736.502 1911.56 736.502 Q1906.66 736.502 1903.7 739.271 Q1900.77 742.04 1900.33 747.069 L1921.21 747.037 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1971.46 718.582 Q1967.2 725.903 1965.13 733.064 Q1963.06 740.226 1963.06 747.578 Q1963.06 754.93 1965.13 762.156 Q1967.23 769.349 1971.46 776.638 L1966.37 776.638 Q1961.6 769.158 1959.21 761.933 Q1956.85 754.708 1956.85 747.578 Q1956.85 740.48 1959.21 733.287 Q1961.57 726.094 1966.37 718.582 L1971.46 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2005.55 733.446 L2005.55 738.984 Q2003.07 737.711 2000.4 737.075 Q1997.72 736.438 1994.86 736.438 Q1990.5 736.438 1988.3 737.775 Q1986.14 739.112 1986.14 741.785 Q1986.14 743.822 1987.7 745 Q1989.26 746.146 1993.97 747.196 L1995.97 747.642 Q2002.21 748.979 2004.82 751.429 Q2007.46 753.848 2007.46 758.209 Q2007.46 763.174 2003.52 766.07 Q1999.6 768.967 1992.73 768.967 Q1989.86 768.967 1986.74 768.394 Q1983.65 767.853 1980.22 766.739 L1980.22 760.691 Q1983.46 762.378 1986.61 763.238 Q1989.77 764.065 1992.85 764.065 Q1996.99 764.065 1999.22 762.665 Q2001.45 761.233 2001.45 758.654 Q2001.45 756.267 1999.82 754.994 Q1998.23 753.721 1992.79 752.543 L1990.75 752.066 Q1985.31 750.92 1982.89 748.565 Q1980.47 746.178 1980.47 742.04 Q1980.47 737.011 1984.04 734.274 Q1987.6 731.537 1994.16 731.537 Q1997.4 731.537 2000.27 732.014 Q2003.13 732.491 2005.55 733.446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2015.86 718.582 L2020.96 718.582 Q2025.73 726.094 2028.09 733.287 Q2030.47 740.48 2030.47 747.578 Q2030.47 754.708 2028.09 761.933 Q2025.73 769.158 2020.96 776.638 L2015.86 776.638 Q2020.1 769.349 2022.17 762.156 Q2024.27 754.93 2024.27 747.578 Q2024.27 740.226 2022.17 733.064 Q2020.1 725.903 2015.86 718.582 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,606.88 2352.76,606.88 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,471.046 2352.76,471.046 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,335.212 2352.76,335.212 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,199.378 2352.76,199.378 "/>
+<polyline clip-path="url(#clip973)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,63.5442 2352.76,63.5442 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,623.18 1448.54,47.2441 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,606.88 1467.43,606.88 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,471.046 1467.43,471.046 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,335.212 1467.43,335.212 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,199.378 1467.43,199.378 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,63.5442 1467.43,63.5442 "/>
+<path clip-path="url(#clip970)" d="M1355.2 592.679 Q1351.59 592.679 1349.76 596.244 Q1347.95 599.785 1347.95 606.915 Q1347.95 614.021 1349.76 617.586 Q1351.59 621.128 1355.2 621.128 Q1358.83 621.128 1360.64 617.586 Q1362.47 614.021 1362.47 606.915 Q1362.47 599.785 1360.64 596.244 Q1358.83 592.679 1355.2 592.679 M1355.2 588.975 Q1361.01 588.975 1364.06 593.582 Q1367.14 598.165 1367.14 606.915 Q1367.14 615.642 1364.06 620.248 Q1361.01 624.831 1355.2 624.831 Q1349.39 624.831 1346.31 620.248 Q1343.25 615.642 1343.25 606.915 Q1343.25 598.165 1346.31 593.582 Q1349.39 588.975 1355.2 588.975 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1375.36 618.281 L1380.24 618.281 L1380.24 624.16 L1375.36 624.16 L1375.36 618.281 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1401.01 605.017 Q1397.86 605.017 1396.01 607.169 Q1394.18 609.322 1394.18 613.072 Q1394.18 616.799 1396.01 618.975 Q1397.86 621.128 1401.01 621.128 Q1404.16 621.128 1405.98 618.975 Q1407.84 616.799 1407.84 613.072 Q1407.84 609.322 1405.98 607.169 Q1404.16 605.017 1401.01 605.017 M1410.29 590.364 L1410.29 594.623 Q1408.53 593.79 1406.72 593.35 Q1404.94 592.91 1403.18 592.91 Q1398.55 592.91 1396.1 596.035 Q1393.67 599.16 1393.32 605.48 Q1394.69 603.466 1396.75 602.401 Q1398.81 601.313 1401.29 601.313 Q1406.49 601.313 1409.5 604.484 Q1412.54 607.632 1412.54 613.072 Q1412.54 618.396 1409.39 621.614 Q1406.24 624.831 1401.01 624.831 Q1395.01 624.831 1391.84 620.248 Q1388.67 615.642 1388.67 606.915 Q1388.67 598.72 1392.56 593.859 Q1396.45 588.975 1403 588.975 Q1404.76 588.975 1406.54 589.322 Q1408.35 589.67 1410.29 590.364 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1356.26 456.845 Q1352.65 456.845 1350.82 460.41 Q1349.02 463.951 1349.02 471.081 Q1349.02 478.187 1350.82 481.752 Q1352.65 485.294 1356.26 485.294 Q1359.9 485.294 1361.7 481.752 Q1363.53 478.187 1363.53 471.081 Q1363.53 463.951 1361.7 460.41 Q1359.9 456.845 1356.26 456.845 M1356.26 453.141 Q1362.07 453.141 1365.13 457.748 Q1368.21 462.331 1368.21 471.081 Q1368.21 479.808 1365.13 484.414 Q1362.07 488.997 1356.26 488.997 Q1350.45 488.997 1347.37 484.414 Q1344.32 479.808 1344.32 471.081 Q1344.32 462.331 1347.37 457.748 Q1350.45 453.141 1356.26 453.141 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1376.42 482.447 L1381.31 482.447 L1381.31 488.326 L1376.42 488.326 L1376.42 482.447 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1390.31 453.766 L1412.54 453.766 L1412.54 455.757 L1399.99 488.326 L1395.1 488.326 L1406.91 457.701 L1390.31 457.701 L1390.31 453.766 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1355.45 321.011 Q1351.84 321.011 1350.01 324.576 Q1348.21 328.117 1348.21 335.247 Q1348.21 342.353 1350.01 345.918 Q1351.84 349.46 1355.45 349.46 Q1359.09 349.46 1360.89 345.918 Q1362.72 342.353 1362.72 335.247 Q1362.72 328.117 1360.89 324.576 Q1359.09 321.011 1355.45 321.011 M1355.45 317.307 Q1361.26 317.307 1364.32 321.914 Q1367.4 326.497 1367.4 335.247 Q1367.4 343.974 1364.32 348.58 Q1361.26 353.163 1355.45 353.163 Q1349.64 353.163 1346.56 348.58 Q1343.51 343.974 1343.51 335.247 Q1343.51 326.497 1346.56 321.914 Q1349.64 317.307 1355.45 317.307 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1375.61 346.613 L1380.5 346.613 L1380.5 352.492 L1375.61 352.492 L1375.61 346.613 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1400.68 336.08 Q1397.35 336.08 1395.43 337.863 Q1393.53 339.645 1393.53 342.77 Q1393.53 345.895 1395.43 347.677 Q1397.35 349.46 1400.68 349.46 Q1404.02 349.46 1405.94 347.677 Q1407.86 345.872 1407.86 342.77 Q1407.86 339.645 1405.94 337.863 Q1404.04 336.08 1400.68 336.08 M1396.01 334.089 Q1393 333.349 1391.31 331.289 Q1389.64 329.228 1389.64 326.265 Q1389.64 322.122 1392.58 319.715 Q1395.54 317.307 1400.68 317.307 Q1405.85 317.307 1408.79 319.715 Q1411.72 322.122 1411.72 326.265 Q1411.72 329.228 1410.04 331.289 Q1408.37 333.349 1405.38 334.089 Q1408.76 334.877 1410.64 337.168 Q1412.54 339.46 1412.54 342.77 Q1412.54 347.793 1409.46 350.478 Q1406.4 353.163 1400.68 353.163 Q1394.97 353.163 1391.89 350.478 Q1388.83 347.793 1388.83 342.77 Q1388.83 339.46 1390.73 337.168 Q1392.63 334.877 1396.01 334.089 M1394.29 326.705 Q1394.29 329.39 1395.96 330.895 Q1397.65 332.4 1400.68 332.4 Q1403.69 332.4 1405.38 330.895 Q1407.1 329.39 1407.1 326.705 Q1407.1 324.02 1405.38 322.515 Q1403.69 321.011 1400.68 321.011 Q1397.65 321.011 1395.96 322.515 Q1394.29 324.02 1394.29 326.705 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1355.54 185.177 Q1351.93 185.177 1350.1 188.742 Q1348.3 192.283 1348.3 199.413 Q1348.3 206.519 1350.1 210.084 Q1351.93 213.626 1355.54 213.626 Q1359.18 213.626 1360.98 210.084 Q1362.81 206.519 1362.81 199.413 Q1362.81 192.283 1360.98 188.742 Q1359.18 185.177 1355.54 185.177 M1355.54 181.473 Q1361.35 181.473 1364.41 186.08 Q1367.49 190.663 1367.49 199.413 Q1367.49 208.14 1364.41 212.746 Q1361.35 217.329 1355.54 217.329 Q1349.73 217.329 1346.66 212.746 Q1343.6 208.14 1343.6 199.413 Q1343.6 190.663 1346.66 186.08 Q1349.73 181.473 1355.54 181.473 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1375.71 210.779 L1380.59 210.779 L1380.59 216.658 L1375.71 216.658 L1375.71 210.779 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1390.91 215.941 L1390.91 211.681 Q1392.67 212.515 1394.48 212.954 Q1396.29 213.394 1398.02 213.394 Q1402.65 213.394 1405.08 210.292 Q1407.54 207.167 1407.88 200.825 Q1406.54 202.816 1404.48 203.88 Q1402.42 204.945 1399.92 204.945 Q1394.73 204.945 1391.7 201.82 Q1388.69 198.672 1388.69 193.232 Q1388.69 187.908 1391.84 184.691 Q1394.99 181.473 1400.22 181.473 Q1406.22 181.473 1409.36 186.08 Q1412.54 190.663 1412.54 199.413 Q1412.54 207.584 1408.65 212.468 Q1404.78 217.329 1398.23 217.329 Q1396.47 217.329 1394.66 216.982 Q1392.86 216.635 1390.91 215.941 M1400.22 201.288 Q1403.37 201.288 1405.2 199.135 Q1407.05 196.982 1407.05 193.232 Q1407.05 189.506 1405.2 187.353 Q1403.37 185.177 1400.22 185.177 Q1397.07 185.177 1395.22 187.353 Q1393.39 189.506 1393.39 193.232 Q1393.39 196.982 1395.22 199.135 Q1397.07 201.288 1400.22 201.288 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1346.17 76.889 L1353.81 76.889 L1353.81 50.5234 L1345.5 52.1901 L1345.5 47.9308 L1353.76 46.2642 L1358.44 46.2642 L1358.44 76.889 L1366.08 76.889 L1366.08 80.8242 L1346.17 80.8242 L1346.17 76.889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1375.52 74.9446 L1380.41 74.9446 L1380.41 80.8242 L1375.52 80.8242 L1375.52 74.9446 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1400.59 49.3429 Q1396.98 49.3429 1395.15 52.9076 Q1393.35 56.4493 1393.35 63.5789 Q1393.35 70.6853 1395.15 74.2501 Q1396.98 77.7918 1400.59 77.7918 Q1404.22 77.7918 1406.03 74.2501 Q1407.86 70.6853 1407.86 63.5789 Q1407.86 56.4493 1406.03 52.9076 Q1404.22 49.3429 1400.59 49.3429 M1400.59 45.6392 Q1406.4 45.6392 1409.46 50.2456 Q1412.54 54.8289 1412.54 63.5789 Q1412.54 72.3057 1409.46 76.9121 Q1406.4 81.4955 1400.59 81.4955 Q1394.78 81.4955 1391.7 76.9121 Q1388.65 72.3057 1388.65 63.5789 Q1388.65 54.8289 1391.7 50.2456 Q1394.78 45.6392 1400.59 45.6392 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1245.48 412.476 L1245.48 402.895 L1277.81 390.769 L1245.48 378.578 L1245.48 368.998 L1293 368.998 L1293 375.268 L1251.27 375.268 L1283.86 387.522 L1283.86 393.983 L1251.27 406.237 L1293 406.237 L1293 412.476 L1245.48 412.476 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1275.08 340.289 Q1275.08 347.386 1276.7 350.124 Q1278.32 352.861 1282.24 352.861 Q1285.36 352.861 1287.2 350.824 Q1289.02 348.755 1289.02 345.222 Q1289.02 340.352 1285.58 337.424 Q1282.11 334.464 1276.38 334.464 L1275.08 334.464 L1275.08 340.289 M1272.66 328.608 L1293 328.608 L1293 334.464 L1287.59 334.464 Q1290.83 336.469 1292.39 339.461 Q1293.92 342.453 1293.92 346.782 Q1293.92 352.256 1290.86 355.503 Q1287.78 358.717 1282.62 358.717 Q1276.61 358.717 1273.55 354.707 Q1270.49 350.665 1270.49 342.676 L1270.49 334.464 L1269.92 334.464 Q1265.88 334.464 1263.68 337.138 Q1261.45 339.779 1261.45 344.585 Q1261.45 347.641 1262.19 350.537 Q1262.92 353.434 1264.38 356.107 L1258.97 356.107 Q1257.73 352.893 1257.13 349.869 Q1256.49 346.845 1256.49 343.981 Q1256.49 336.246 1260.5 332.427 Q1264.51 328.608 1272.66 328.608 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1258.4 293.819 L1263.94 293.819 Q1262.66 296.302 1262.03 298.975 Q1261.39 301.649 1261.39 304.513 Q1261.39 308.874 1262.73 311.07 Q1264.06 313.234 1266.74 313.234 Q1268.78 313.234 1269.95 311.675 Q1271.1 310.115 1272.15 305.405 L1272.59 303.399 Q1273.93 297.161 1276.38 294.551 Q1278.8 291.909 1283.16 291.909 Q1288.13 291.909 1291.02 295.856 Q1293.92 299.771 1293.92 306.646 Q1293.92 309.51 1293.35 312.63 Q1292.81 315.717 1291.69 319.154 L1285.64 319.154 Q1287.33 315.908 1288.19 312.757 Q1289.02 309.606 1289.02 306.519 Q1289.02 302.381 1287.62 300.153 Q1286.19 297.925 1283.61 297.925 Q1281.22 297.925 1279.95 299.548 Q1278.67 301.14 1277.5 306.582 L1277.02 308.619 Q1275.87 314.062 1273.52 316.481 Q1271.13 318.9 1266.99 318.9 Q1261.96 318.9 1259.23 315.335 Q1256.49 311.77 1256.49 305.214 Q1256.49 301.967 1256.97 299.102 Q1257.44 296.238 1258.4 293.819 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1258.4 259.858 L1263.94 259.858 Q1262.66 262.341 1262.03 265.014 Q1261.39 267.688 1261.39 270.552 Q1261.39 274.913 1262.73 277.109 Q1264.06 279.273 1266.74 279.273 Q1268.78 279.273 1269.95 277.714 Q1271.1 276.154 1272.15 271.444 L1272.59 269.438 Q1273.93 263.2 1276.38 260.59 Q1278.8 257.948 1283.16 257.948 Q1288.13 257.948 1291.02 261.895 Q1293.92 265.81 1293.92 272.685 Q1293.92 275.549 1293.35 278.669 Q1292.81 281.756 1291.69 285.193 L1285.64 285.193 Q1287.33 281.947 1288.19 278.796 Q1289.02 275.645 1289.02 272.558 Q1289.02 268.42 1287.62 266.192 Q1286.19 263.964 1283.61 263.964 Q1281.22 263.964 1279.95 265.587 Q1278.67 267.178 1277.5 272.621 L1277.02 274.658 Q1275.87 280.101 1273.52 282.52 Q1271.13 284.939 1266.99 284.939 Q1261.96 284.939 1259.23 281.374 Q1256.49 277.809 1256.49 271.253 Q1256.49 268.006 1256.97 265.141 Q1257.44 262.277 1258.4 259.858 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip973)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1474.13,63.5442 1474.98,65.4457 1475.83,67.3472 1476.69,69.2488 1477.54,71.1503 1478.4,73.0518 1479.25,74.9533 1480.1,76.8549 1480.96,78.7564 1481.81,80.6579 1482.67,82.5594 1483.52,84.4609 1484.37,86.3624 1485.23,88.2639 1486.08,90.1654 1486.93,92.0669 1487.79,93.9684 1488.64,95.8699 1489.5,97.7713 1490.35,99.6728 1491.2,101.574 1492.06,103.476 1492.91,105.377 1493.77,107.279 1494.62,109.18 1495.47,111.082 1496.33,112.983 1497.18,114.884 1498.04,116.786 1498.89,118.687 1499.74,120.589 1500.6,122.49 1501.45,124.392 1502.3,126.293 1503.16,128.194 1504.01,130.096 1504.87,131.997 1505.72,133.898 1506.57,135.8 1507.43,137.701 1508.28,139.602 1509.14,141.504 1509.99,143.405 1510.84,145.306 1511.7,147.208 1512.55,149.109 1513.41,151.01 1514.26,152.912 1515.11,154.813 1515.97,156.714 1516.82,158.615 1517.67,160.516 1518.53,162.418 1519.38,164.319 1520.24,166.22 1521.09,168.121 1521.94,170.022 1522.8,171.923 1523.65,173.824 1524.51,175.725 1525.36,177.626 1526.21,179.527 1527.07,181.428 1527.92,183.329 1528.78,185.23 1529.63,187.131 1530.48,189.032 1531.34,190.933 1532.19,192.833 1533.04,194.734 1533.9,196.635 1534.75,198.536 1535.61,200.436 1536.46,202.337 1537.31,204.237 1538.17,206.138 1539.02,208.038 1539.88,209.938 1540.73,211.838 1541.58,213.738 1542.44,215.638 1543.29,217.538 1544.15,219.438 1545,221.338 1545.85,223.237 1546.71,225.136 1547.56,227.036 1548.41,228.935 1549.27,230.833 1550.12,232.732 1550.98,234.63 1551.83,236.528 1552.68,238.426 1553.54,240.323 1554.39,242.22 1555.25,244.116 1556.1,246.011 1556.95,247.906 1557.81,249.8 1558.66,251.692 1559.52,253.584 1560.37,255.473 1561.22,257.361 1562.08,259.246 1562.93,261.127 1563.78,263.003 1564.64,264.871 1565.49,266.729 1566.35,268.57 1567.2,270.384 1568.05,272.151 1568.91,273.849 1569.76,275.461 1570.62,276.985 1571.47,278.433 1572.32,279.816 1573.18,281.146 1574.03,282.435 1574.89,283.689 1575.74,284.918 1576.59,286.126 1577.45,287.317 1578.3,288.495 1579.16,289.663 1580.01,290.824 1580.86,291.978 1581.72,293.127 1582.57,294.273 1583.42,295.415 1584.28,296.556 1585.13,297.694 1585.99,298.832 1586.84,299.968 1587.69,301.103 1588.55,302.238 1589.4,303.373 1590.26,304.507 1591.11,305.641 1591.96,306.775 1592.82,307.909 1593.67,309.043 1594.53,310.177 1595.38,311.311 1596.23,312.445 1597.09,313.58 1597.94,314.715 1598.79,315.849 1599.65,316.985 1600.5,318.12 1601.36,319.256 1602.21,320.392 1603.06,321.528 1603.92,322.664 1604.77,323.801 1605.63,324.938 1606.48,326.076 1607.33,327.214 1608.19,328.352 1609.04,329.49 1609.9,330.629 1610.75,331.768 1611.6,332.908 1612.46,334.048 1613.31,335.188 1614.16,336.329 1615.02,337.47 1615.87,338.612 1616.73,339.754 1617.58,340.896 1618.43,342.039 1619.29,343.182 1620.14,344.326 1621,345.471 1621.85,346.615 1622.7,347.761 1623.56,348.906 1624.41,350.053 1625.27,351.199 1626.12,352.347 1626.97,353.494 1627.83,354.643 1628.68,355.792 1629.53,356.941 1630.39,358.091 1631.24,359.242 1632.1,360.393 1632.95,361.545 1633.8,362.698 1634.66,363.851 1635.51,365.005 1636.37,366.159 1637.22,367.314 1638.07,368.47 1638.93,369.626 1639.78,370.783 1640.64,371.941 1641.49,373.1 1642.34,374.259 1643.2,375.419 1644.05,376.58 1644.9,377.741 1645.76,378.903 1646.61,380.066 1647.47,381.23 1648.32,382.395 1649.17,383.56 1650.03,384.726 1650.88,385.894 1651.74,387.062 1652.59,388.23 1653.44,389.4 1654.3,390.571 1655.15,391.742 1656.01,392.915 1656.86,394.088 1657.71,395.262 1658.57,396.437 1659.42,397.614 1660.27,398.791 1661.13,399.969 1661.98,401.148 1662.84,402.329 1663.69,403.51 1664.54,404.692 1665.4,405.876 1666.25,407.06 1667.11,408.246 1667.96,409.432 1668.81,410.62 1669.67,411.809 1670.52,412.999 1671.38,414.191 1672.23,415.383 1673.08,416.577 1673.94,417.772 1674.79,418.968 1675.64,420.165 1676.5,421.364 1677.35,422.564 1678.21,423.765 1679.06,424.968 1679.91,426.171 1680.77,427.377 1681.62,428.583 1682.48,429.791 1683.33,431.001 1684.18,432.212 1685.04,433.424 1685.89,434.638 1686.75,435.853 1687.6,437.07 1688.45,438.288 1689.31,439.508 1690.16,440.729 1691.01,441.952 1691.87,443.176 1692.72,444.402 1693.58,445.63 1694.43,446.86 1695.28,448.091 1696.14,449.323 1696.99,450.558 1697.85,451.794 1698.7,453.032 1699.55,454.272 1700.41,455.514 1701.26,456.757 1702.12,458.002 1702.97,459.249 1703.82,460.498 1704.68,461.749 1705.53,463.002 1706.38,464.257 1707.24,465.514 1708.09,466.773 1708.95,468.034 1709.8,469.297 1710.65,470.562 1711.51,471.829 1712.36,473.098 1713.22,474.369 1714.07,475.643 1714.92,476.919 1715.78,478.197 1716.63,479.477 1717.49,480.76 1718.34,482.045 1719.19,483.332 1720.05,484.622 1720.9,485.914 1721.76,487.209 1722.61,488.506 1723.46,489.805 1724.32,491.108 1725.17,492.412 1726.02,493.719 1726.88,495.029 1727.73,496.342 1728.59,497.657 1729.44,498.975 1730.29,500.295 1731.15,501.619 1732,502.945 1732.86,504.274 1733.71,505.606 1734.56,506.941 1735.42,508.279 1736.27,509.619 1737.13,510.963 1737.98,512.31 1738.83,513.66 1739.69,515.012 1740.54,516.368 1741.39,517.728 1742.25,519.09 1743.1,520.456 1743.96,521.825 1744.81,523.197 1745.66,524.572 1746.52,525.951 1747.37,527.333 1748.23,528.718 1749.08,530.107 1749.93,531.499 1750.79,532.895 1751.64,534.294 1752.5,535.697 1753.35,537.103 1754.2,538.512 1755.06,539.925 1755.91,541.341 1756.76,542.761 1757.62,544.184 1758.47,545.61 1759.33,547.039 1760.18,548.472 1761.03,549.908 1761.89,551.347 1762.74,552.789 1763.6,554.233 1764.45,555.681 1765.3,557.131 1766.16,558.583 1767.01,560.037 1767.87,561.493 1768.72,562.951 1769.57,564.41 1770.43,565.87 1771.28,567.331 1772.13,568.792 1772.99,570.252 1773.84,571.711 1774.7,573.169 1775.55,574.624 1776.4,576.076 1777.26,577.524 1778.11,578.967 1778.97,580.404 1779.82,581.833 1780.67,583.253 1781.53,584.663 1782.38,586.061 1783.24,587.444 1784.09,588.81 1784.94,590.158 1785.8,591.483 1786.65,592.784 1787.5,594.056 1788.36,595.296 1789.21,596.499 1790.07,597.66 1790.92,598.775 1791.77,599.838 1792.63,600.843 1793.48,601.782 1794.34,602.649 1795.19,603.437 1796.04,604.137 1796.9,604.742 1797.75,605.246 1798.61,605.643 1799.46,605.936 1800.31,606.136 1801.17,606.264 1802.02,606.349 1802.87,606.408 1803.73,606.452 1804.58,606.486 1805.44,606.513 1806.29,606.536 1807.14,606.555 1808,606.572 1808.85,606.587 1809.71,606.599 1810.56,606.611 1811.41,606.621 1812.27,606.63 1813.12,606.639 1813.98,606.646 1814.83,606.653 1815.68,606.66 1816.54,606.666 1817.39,606.671 1818.24,606.677 1819.1,606.681 1819.95,606.686 1820.81,606.69 1821.66,606.694 1822.51,606.698 1823.37,606.702 1824.22,606.705 1825.08,606.708 1825.93,606.711 1826.78,606.714 1827.64,606.717 1828.49,606.72 1829.35,606.722 1830.2,606.725 1831.05,606.727 1831.91,606.729 1832.76,606.731 1833.61,606.734 1834.47,606.736 1835.32,606.737 1836.18,606.739 1837.03,606.741 1837.88,606.743 1838.74,606.744 1839.59,606.746 1840.45,606.748 1841.3,606.749 1842.15,606.751 1843.01,606.752 1843.86,606.753 1844.72,606.755 1845.57,606.756 1846.42,606.757 1847.28,606.759 1848.13,606.76 1848.99,606.761 1849.84,606.762 1850.69,606.763 1851.55,606.764 1852.4,606.765 1853.25,606.766 1854.11,606.767 1854.96,606.768 1855.82,606.769 1856.67,606.77 1857.52,606.771 1858.38,606.772 1859.23,606.773 1860.09,606.774 1860.94,606.774 1861.79,606.775 1862.65,606.776 1863.5,606.777 1864.36,606.778 1865.21,606.778 1866.06,606.779 1866.92,606.78 1867.77,606.781 1868.62,606.781 1869.48,606.782 1870.33,606.783 1871.19,606.783 1872.04,606.784 1872.89,606.785 1873.75,606.785 1874.6,606.786 1875.46,606.787 1876.31,606.787 1877.16,606.788 1878.02,606.788 1878.87,606.789 1879.73,606.789 1880.58,606.79 1881.43,606.791 1882.29,606.791 1883.14,606.792 1883.99,606.792 1884.85,606.793 1885.7,606.793 1886.56,606.794 1887.41,606.794 1888.26,606.795 1889.12,606.795 1889.97,606.796 1890.83,606.796 1891.68,606.797 1892.53,606.797 1893.39,606.797 1894.24,606.798 1895.1,606.798 1895.95,606.799 1896.8,606.799 1897.66,606.8 1898.51,606.8 1899.36,606.8 1900.22,606.801 1901.07,606.801 1901.93,606.802 1902.78,606.802 1903.63,606.802 1904.49,606.803 1905.34,606.803 1906.2,606.804 1907.05,606.804 1907.9,606.804 1908.76,606.805 1909.61,606.805 1910.47,606.805 1911.32,606.806 1912.17,606.806 1913.03,606.806 1913.88,606.807 1914.73,606.807 1915.59,606.807 1916.44,606.808 1917.3,606.808 1918.15,606.808 1919,606.809 1919.86,606.809 1920.71,606.809 1921.57,606.81 1922.42,606.81 1923.27,606.81 1924.13,606.811 1924.98,606.811 1925.84,606.811 1926.69,606.812 1927.54,606.812 1928.4,606.812 1929.25,606.812 1930.1,606.813 1930.96,606.813 1931.81,606.813 1932.67,606.814 1933.52,606.814 1934.37,606.814 1935.23,606.814 1936.08,606.815 1936.94,606.815 1937.79,606.815 1938.64,606.815 1939.5,606.816 1940.35,606.816 1941.21,606.816 1942.06,606.817 1942.91,606.817 1943.77,606.817 1944.62,606.817 1945.47,606.818 1946.33,606.818 1947.18,606.818 1948.04,606.818 1948.89,606.819 1949.74,606.819 1950.6,606.819 1951.45,606.819 1952.31,606.819 1953.16,606.82 1954.01,606.82 1954.87,606.82 1955.72,606.82 1956.58,606.821 1957.43,606.821 1958.28,606.821 1959.14,606.821 1959.99,606.822 1960.84,606.822 1961.7,606.822 1962.55,606.822 1963.41,606.822 1964.26,606.823 1965.11,606.823 1965.97,606.823 1966.82,606.823 1967.68,606.823 1968.53,606.824 1969.38,606.824 1970.24,606.824 1971.09,606.824 1971.95,606.824 1972.8,606.825 1973.65,606.825 1974.51,606.825 1975.36,606.825 1976.21,606.825 1977.07,606.826 1977.92,606.826 1978.78,606.826 1979.63,606.826 1980.48,606.826 1981.34,606.827 1982.19,606.827 1983.05,606.827 1983.9,606.827 1984.75,606.827 1985.61,606.828 1986.46,606.828 1987.32,606.828 1988.17,606.828 1989.02,606.828 1989.88,606.829 1990.73,606.829 1991.59,606.829 1992.44,606.829 1993.29,606.829 1994.15,606.829 1995,606.83 1995.85,606.83 1996.71,606.83 1997.56,606.83 1998.42,606.83 1999.27,606.83 2000.12,606.831 2000.98,606.831 2001.83,606.831 2002.69,606.831 2003.54,606.831 2004.39,606.831 2005.25,606.832 2006.1,606.832 2006.96,606.832 2007.81,606.832 2008.66,606.832 2009.52,606.832 2010.37,606.833 2011.22,606.833 2012.08,606.833 2012.93,606.833 2013.79,606.833 2014.64,606.833 2015.49,606.834 2016.35,606.834 2017.2,606.834 2018.06,606.834 2018.91,606.834 2019.76,606.834 2020.62,606.835 2021.47,606.835 2022.33,606.835 2023.18,606.835 2024.03,606.835 2024.89,606.835 2025.74,606.835 2026.59,606.836 2027.45,606.836 2028.3,606.836 2029.16,606.836 2030.01,606.836 2030.86,606.836 2031.72,606.836 2032.57,606.837 2033.43,606.837 2034.28,606.837 2035.13,606.837 2035.99,606.837 2036.84,606.837 2037.7,606.837 2038.55,606.838 2039.4,606.838 2040.26,606.838 2041.11,606.838 2041.96,606.838 2042.82,606.838 2043.67,606.838 2044.53,606.839 2045.38,606.839 2046.23,606.839 2047.09,606.839 2047.94,606.839 2048.8,606.839 2049.65,606.839 2050.5,606.84 2051.36,606.84 2052.21,606.84 2053.07,606.84 2053.92,606.84 2054.77,606.84 2055.63,606.84 2056.48,606.841 2057.33,606.841 2058.19,606.841 2059.04,606.841 2059.9,606.841 2060.75,606.841 2061.6,606.841 2062.46,606.841 2063.31,606.842 2064.17,606.842 2065.02,606.842 2065.87,606.842 2066.73,606.842 2067.58,606.842 2068.44,606.842 2069.29,606.842 2070.14,606.843 2071,606.843 2071.85,606.843 2072.7,606.843 2073.56,606.843 2074.41,606.843 2075.27,606.843 2076.12,606.843 2076.97,606.844 2077.83,606.844 2078.68,606.844 2079.54,606.844 2080.39,606.844 2081.24,606.844 2082.1,606.844 2082.95,606.844 2083.81,606.845 2084.66,606.845 2085.51,606.845 2086.37,606.845 2087.22,606.845 2088.07,606.845 2088.93,606.845 2089.78,606.845 2090.64,606.846 2091.49,606.846 2092.34,606.846 2093.2,606.846 2094.05,606.846 2094.91,606.846 2095.76,606.846 2096.61,606.846 2097.47,606.846 2098.32,606.847 2099.18,606.847 2100.03,606.847 2100.88,606.847 2101.74,606.847 2102.59,606.847 2103.44,606.847 2104.3,606.847 2105.15,606.848 2106.01,606.848 2106.86,606.848 2107.71,606.848 2108.57,606.848 2109.42,606.848 2110.28,606.848 2111.13,606.848 2111.98,606.848 2112.84,606.849 2113.69,606.849 2114.55,606.849 2115.4,606.849 2116.25,606.849 2117.11,606.849 2117.96,606.849 2118.82,606.849 2119.67,606.849 2120.52,606.85 2121.38,606.85 2122.23,606.85 2123.08,606.85 2123.94,606.85 2124.79,606.85 2125.65,606.85 2126.5,606.85 2127.35,606.85 2128.21,606.851 2129.06,606.851 2129.92,606.851 2130.77,606.851 2131.62,606.851 2132.48,606.851 2133.33,606.851 2134.19,606.851 2135.04,606.851 2135.89,606.852 2136.75,606.852 2137.6,606.852 2138.45,606.852 2139.31,606.852 2140.16,606.852 2141.02,606.852 2141.87,606.852 2142.72,606.852 2143.58,606.853 2144.43,606.853 2145.29,606.853 2146.14,606.853 2146.99,606.853 2147.85,606.853 2148.7,606.853 2149.56,606.853 2150.41,606.853 2151.26,606.853 2152.12,606.854 2152.97,606.854 2153.82,606.854 2154.68,606.854 2155.53,606.854 2156.39,606.854 2157.24,606.854 2158.09,606.854 2158.95,606.854 2159.8,606.855 2160.66,606.855 2161.51,606.855 2162.36,606.855 2163.22,606.855 2164.07,606.855 2164.93,606.855 2165.78,606.855 2166.63,606.855 2167.49,606.855 2168.34,606.856 2169.19,606.856 2170.05,606.856 2170.9,606.856 2171.76,606.856 2172.61,606.856 2173.46,606.856 2174.32,606.856 2175.17,606.856 2176.03,606.856 2176.88,606.857 2177.73,606.857 2178.59,606.857 2179.44,606.857 2180.3,606.857 2181.15,606.857 2182,606.857 2182.86,606.857 2183.71,606.857 2184.56,606.857 2185.42,606.858 2186.27,606.858 2187.13,606.858 2187.98,606.858 2188.83,606.858 2189.69,606.858 2190.54,606.858 2191.4,606.858 2192.25,606.858 2193.1,606.859 2193.96,606.859 2194.81,606.859 2195.67,606.859 2196.52,606.859 2197.37,606.859 2198.23,606.859 2199.08,606.859 2199.93,606.859 2200.79,606.859 2201.64,606.86 2202.5,606.86 2203.35,606.86 2204.2,606.86 2205.06,606.86 2205.91,606.86 2206.77,606.86 2207.62,606.86 2208.47,606.86 2209.33,606.86 2210.18,606.861 2211.04,606.861 2211.89,606.861 2212.74,606.861 2213.6,606.861 2214.45,606.861 2215.3,606.861 2216.16,606.861 2217.01,606.861 2217.87,606.861 2218.72,606.862 2219.57,606.862 2220.43,606.862 2221.28,606.862 2222.14,606.862 2222.99,606.862 2223.84,606.862 2224.7,606.862 2225.55,606.862 2226.41,606.862 2227.26,606.863 2228.11,606.863 2228.97,606.863 2229.82,606.863 2230.67,606.863 2231.53,606.863 2232.38,606.863 2233.24,606.863 2234.09,606.863 2234.94,606.863 2235.8,606.864 2236.65,606.864 2237.51,606.864 2238.36,606.864 2239.21,606.864 2240.07,606.864 2240.92,606.864 2241.78,606.864 2242.63,606.864 2243.48,606.865 2244.34,606.865 2245.19,606.865 2246.04,606.865 2246.9,606.865 2247.75,606.865 2248.61,606.865 2249.46,606.865 2250.31,606.865 2251.17,606.865 2252.02,606.866 2252.88,606.866 2253.73,606.866 2254.58,606.866 2255.44,606.866 2256.29,606.866 2257.15,606.866 2258,606.866 2258.85,606.866 2259.71,606.867 2260.56,606.867 2261.42,606.867 2262.27,606.867 2263.12,606.867 2263.98,606.867 2264.83,606.867 2265.68,606.867 2266.54,606.867 2267.39,606.868 2268.25,606.868 2269.1,606.868 2269.95,606.868 2270.81,606.868 2271.66,606.868 2272.52,606.868 2273.37,606.868 2274.22,606.868 2275.08,606.869 2275.93,606.869 2276.79,606.869 2277.64,606.869 2278.49,606.869 2279.35,606.869 2280.2,606.869 2281.05,606.869 2281.91,606.87 2282.76,606.87 2283.62,606.87 2284.47,606.87 2285.32,606.87 2286.18,606.87 2287.03,606.87 2287.89,606.87 2288.74,606.871 2289.59,606.871 2290.45,606.871 2291.3,606.871 2292.16,606.871 2293.01,606.871 2293.86,606.871 2294.72,606.871 2295.57,606.872 2296.42,606.872 2297.28,606.872 2298.13,606.872 2298.99,606.872 2299.84,606.872 2300.69,606.872 2301.55,606.873 2302.4,606.873 2303.26,606.873 2304.11,606.873 2304.96,606.873 2305.82,606.873 2306.67,606.873 2307.53,606.874 2308.38,606.874 2309.23,606.874 2310.09,606.874 2310.94,606.874 2311.79,606.874 2312.65,606.875 2313.5,606.875 2314.36,606.875 2315.21,606.875 2316.06,606.875 2316.92,606.876 2317.77,606.876 2318.63,606.876 2319.48,606.876 2320.33,606.876 2321.19,606.877 2322.04,606.877 2322.9,606.877 2323.75,606.878 2324.6,606.878 2325.46,606.878 2326.31,606.879 2327.16,606.88 "/>
+<path clip-path="url(#clip970)" d="M277.528 1423.18 L1181.75 1423.18 L1181.75 847.244 L277.528 847.244  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip154">
+  <clipPath id="clip974">
     <rect x="277" y="847" width="905" height="577"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="302.265,1423.18 302.265,847.244 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="515.738,1423.18 515.738,847.244 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.211,1423.18 729.211,847.244 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="942.685,1423.18 942.685,847.244 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1156.16,1423.18 1156.16,847.244 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1423.18 1181.75,1423.18 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="302.265,1423.18 302.265,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="515.738,1423.18 515.738,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.211,1423.18 729.211,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="942.685,1423.18 942.685,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1156.16,1423.18 1156.16,1404.28 "/>
-<path clip-path="url(#clip150)" d="M264.569 1454.1 Q260.958 1454.1 259.129 1457.66 Q257.323 1461.2 257.323 1468.33 Q257.323 1475.44 259.129 1479.01 Q260.958 1482.55 264.569 1482.55 Q268.203 1482.55 270.008 1479.01 Q271.837 1475.44 271.837 1468.33 Q271.837 1461.2 270.008 1457.66 Q268.203 1454.1 264.569 1454.1 M264.569 1450.39 Q270.379 1450.39 273.434 1455 Q276.513 1459.58 276.513 1468.33 Q276.513 1477.06 273.434 1481.67 Q270.379 1486.25 264.569 1486.25 Q258.759 1486.25 255.68 1481.67 Q252.624 1477.06 252.624 1468.33 Q252.624 1459.58 255.68 1455 Q258.759 1450.39 264.569 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M284.731 1479.7 L289.615 1479.7 L289.615 1485.58 L284.731 1485.58 L284.731 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M309.8 1454.1 Q306.189 1454.1 304.36 1457.66 Q302.555 1461.2 302.555 1468.33 Q302.555 1475.44 304.36 1479.01 Q306.189 1482.55 309.8 1482.55 Q313.434 1482.55 315.24 1479.01 Q317.068 1475.44 317.068 1468.33 Q317.068 1461.2 315.24 1457.66 Q313.434 1454.1 309.8 1454.1 M309.8 1450.39 Q315.61 1450.39 318.666 1455 Q321.744 1459.58 321.744 1468.33 Q321.744 1477.06 318.666 1481.67 Q315.61 1486.25 309.8 1486.25 Q303.99 1486.25 300.911 1481.67 Q297.856 1477.06 297.856 1468.33 Q297.856 1459.58 300.911 1455 Q303.99 1450.39 309.8 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M339.962 1454.1 Q336.351 1454.1 334.522 1457.66 Q332.716 1461.2 332.716 1468.33 Q332.716 1475.44 334.522 1479.01 Q336.351 1482.55 339.962 1482.55 Q343.596 1482.55 345.402 1479.01 Q347.23 1475.44 347.23 1468.33 Q347.23 1461.2 345.402 1457.66 Q343.596 1454.1 339.962 1454.1 M339.962 1450.39 Q345.772 1450.39 348.828 1455 Q351.906 1459.58 351.906 1468.33 Q351.906 1477.06 348.828 1481.67 Q345.772 1486.25 339.962 1486.25 Q334.152 1486.25 331.073 1481.67 Q328.017 1477.06 328.017 1468.33 Q328.017 1459.58 331.073 1455 Q334.152 1450.39 339.962 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M478.539 1454.1 Q474.928 1454.1 473.1 1457.66 Q471.294 1461.2 471.294 1468.33 Q471.294 1475.44 473.1 1479.01 Q474.928 1482.55 478.539 1482.55 Q482.174 1482.55 483.979 1479.01 Q485.808 1475.44 485.808 1468.33 Q485.808 1461.2 483.979 1457.66 Q482.174 1454.1 478.539 1454.1 M478.539 1450.39 Q484.35 1450.39 487.405 1455 Q490.484 1459.58 490.484 1468.33 Q490.484 1477.06 487.405 1481.67 Q484.35 1486.25 478.539 1486.25 Q472.729 1486.25 469.651 1481.67 Q466.595 1477.06 466.595 1468.33 Q466.595 1459.58 469.651 1455 Q472.729 1450.39 478.539 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M498.701 1479.7 L503.586 1479.7 L503.586 1485.58 L498.701 1485.58 L498.701 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M523.771 1454.1 Q520.16 1454.1 518.331 1457.66 Q516.525 1461.2 516.525 1468.33 Q516.525 1475.44 518.331 1479.01 Q520.16 1482.55 523.771 1482.55 Q527.405 1482.55 529.211 1479.01 Q531.039 1475.44 531.039 1468.33 Q531.039 1461.2 529.211 1457.66 Q527.405 1454.1 523.771 1454.1 M523.771 1450.39 Q529.581 1450.39 532.636 1455 Q535.715 1459.58 535.715 1468.33 Q535.715 1477.06 532.636 1481.67 Q529.581 1486.25 523.771 1486.25 Q517.961 1486.25 514.882 1481.67 Q511.826 1477.06 511.826 1468.33 Q511.826 1459.58 514.882 1455 Q517.961 1450.39 523.771 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M543.979 1451.02 L562.335 1451.02 L562.335 1454.96 L548.261 1454.96 L548.261 1463.43 Q549.28 1463.08 550.298 1462.92 Q551.317 1462.73 552.335 1462.73 Q558.122 1462.73 561.502 1465.9 Q564.882 1469.08 564.882 1474.49 Q564.882 1480.07 561.409 1483.17 Q557.937 1486.25 551.618 1486.25 Q549.442 1486.25 547.173 1485.88 Q544.928 1485.51 542.521 1484.77 L542.521 1480.07 Q544.604 1481.2 546.826 1481.76 Q549.048 1482.32 551.525 1482.32 Q555.53 1482.32 557.868 1480.21 Q560.206 1478.1 560.206 1474.49 Q560.206 1470.88 557.868 1468.77 Q555.53 1466.67 551.525 1466.67 Q549.65 1466.67 547.775 1467.08 Q545.923 1467.5 543.979 1468.38 L543.979 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M691.515 1454.1 Q687.904 1454.1 686.075 1457.66 Q684.27 1461.2 684.27 1468.33 Q684.27 1475.44 686.075 1479.01 Q687.904 1482.55 691.515 1482.55 Q695.149 1482.55 696.955 1479.01 Q698.783 1475.44 698.783 1468.33 Q698.783 1461.2 696.955 1457.66 Q695.149 1454.1 691.515 1454.1 M691.515 1450.39 Q697.325 1450.39 700.381 1455 Q703.459 1459.58 703.459 1468.33 Q703.459 1477.06 700.381 1481.67 Q697.325 1486.25 691.515 1486.25 Q685.705 1486.25 682.626 1481.67 Q679.571 1477.06 679.571 1468.33 Q679.571 1459.58 682.626 1455 Q685.705 1450.39 691.515 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M711.677 1479.7 L716.561 1479.7 L716.561 1485.58 L711.677 1485.58 L711.677 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M727.556 1481.64 L735.195 1481.64 L735.195 1455.28 L726.885 1456.95 L726.885 1452.69 L735.149 1451.02 L739.825 1451.02 L739.825 1481.64 L747.464 1481.64 L747.464 1485.58 L727.556 1485.58 L727.556 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M766.908 1454.1 Q763.297 1454.1 761.468 1457.66 Q759.663 1461.2 759.663 1468.33 Q759.663 1475.44 761.468 1479.01 Q763.297 1482.55 766.908 1482.55 Q770.542 1482.55 772.348 1479.01 Q774.176 1475.44 774.176 1468.33 Q774.176 1461.2 772.348 1457.66 Q770.542 1454.1 766.908 1454.1 M766.908 1450.39 Q772.718 1450.39 775.774 1455 Q778.852 1459.58 778.852 1468.33 Q778.852 1477.06 775.774 1481.67 Q772.718 1486.25 766.908 1486.25 Q761.098 1486.25 758.019 1481.67 Q754.964 1477.06 754.964 1468.33 Q754.964 1459.58 758.019 1455 Q761.098 1450.39 766.908 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M905.486 1454.1 Q901.875 1454.1 900.046 1457.66 Q898.24 1461.2 898.24 1468.33 Q898.24 1475.44 900.046 1479.01 Q901.875 1482.55 905.486 1482.55 Q909.12 1482.55 910.925 1479.01 Q912.754 1475.44 912.754 1468.33 Q912.754 1461.2 910.925 1457.66 Q909.12 1454.1 905.486 1454.1 M905.486 1450.39 Q911.296 1450.39 914.351 1455 Q917.43 1459.58 917.43 1468.33 Q917.43 1477.06 914.351 1481.67 Q911.296 1486.25 905.486 1486.25 Q899.676 1486.25 896.597 1481.67 Q893.541 1477.06 893.541 1468.33 Q893.541 1459.58 896.597 1455 Q899.676 1450.39 905.486 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M925.648 1479.7 L930.532 1479.7 L930.532 1485.58 L925.648 1485.58 L925.648 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M941.527 1481.64 L949.166 1481.64 L949.166 1455.28 L940.856 1456.95 L940.856 1452.69 L949.12 1451.02 L953.796 1451.02 L953.796 1481.64 L961.434 1481.64 L961.434 1485.58 L941.527 1485.58 L941.527 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M970.925 1451.02 L989.282 1451.02 L989.282 1454.96 L975.208 1454.96 L975.208 1463.43 Q976.226 1463.08 977.245 1462.92 Q978.263 1462.73 979.282 1462.73 Q985.069 1462.73 988.448 1465.9 Q991.828 1469.08 991.828 1474.49 Q991.828 1480.07 988.356 1483.17 Q984.883 1486.25 978.564 1486.25 Q976.388 1486.25 974.12 1485.88 Q971.874 1485.51 969.467 1484.77 L969.467 1480.07 Q971.55 1481.2 973.772 1481.76 Q975.995 1482.32 978.471 1482.32 Q982.476 1482.32 984.814 1480.21 Q987.152 1478.1 987.152 1474.49 Q987.152 1470.88 984.814 1468.77 Q982.476 1466.67 978.471 1466.67 Q976.596 1466.67 974.721 1467.08 Q972.87 1467.5 970.925 1468.38 L970.925 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1118.46 1454.1 Q1114.85 1454.1 1113.02 1457.66 Q1111.22 1461.2 1111.22 1468.33 Q1111.22 1475.44 1113.02 1479.01 Q1114.85 1482.55 1118.46 1482.55 Q1122.1 1482.55 1123.9 1479.01 Q1125.73 1475.44 1125.73 1468.33 Q1125.73 1461.2 1123.9 1457.66 Q1122.1 1454.1 1118.46 1454.1 M1118.46 1450.39 Q1124.27 1450.39 1127.33 1455 Q1130.41 1459.58 1130.41 1468.33 Q1130.41 1477.06 1127.33 1481.67 Q1124.27 1486.25 1118.46 1486.25 Q1112.65 1486.25 1109.57 1481.67 Q1106.52 1477.06 1106.52 1468.33 Q1106.52 1459.58 1109.57 1455 Q1112.65 1450.39 1118.46 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1138.62 1479.7 L1143.51 1479.7 L1143.51 1485.58 L1138.62 1485.58 L1138.62 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1157.72 1481.64 L1174.04 1481.64 L1174.04 1485.58 L1152.1 1485.58 L1152.1 1481.64 Q1154.76 1478.89 1159.34 1474.26 Q1163.95 1469.61 1165.13 1468.27 Q1167.37 1465.74 1168.25 1464.01 Q1169.16 1462.25 1169.16 1460.56 Q1169.16 1457.8 1167.21 1456.07 Q1165.29 1454.33 1162.19 1454.33 Q1159.99 1454.33 1157.53 1455.09 Q1155.1 1455.86 1152.33 1457.41 L1152.33 1452.69 Q1155.15 1451.55 1157.6 1450.97 Q1160.06 1450.39 1162.1 1450.39 Q1167.47 1450.39 1170.66 1453.08 Q1173.85 1455.77 1173.85 1460.26 Q1173.85 1462.39 1173.04 1464.31 Q1172.26 1466.2 1170.15 1468.8 Q1169.57 1469.47 1166.47 1472.69 Q1163.37 1475.88 1157.72 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1193.85 1454.1 Q1190.24 1454.1 1188.41 1457.66 Q1186.61 1461.2 1186.61 1468.33 Q1186.61 1475.44 1188.41 1479.01 Q1190.24 1482.55 1193.85 1482.55 Q1197.49 1482.55 1199.29 1479.01 Q1201.12 1475.44 1201.12 1468.33 Q1201.12 1461.2 1199.29 1457.66 Q1197.49 1454.1 1193.85 1454.1 M1193.85 1450.39 Q1199.66 1450.39 1202.72 1455 Q1205.8 1459.58 1205.8 1468.33 Q1205.8 1477.06 1202.72 1481.67 Q1199.66 1486.25 1193.85 1486.25 Q1188.04 1486.25 1184.97 1481.67 Q1181.91 1477.06 1181.91 1468.33 Q1181.91 1459.58 1184.97 1455 Q1188.04 1450.39 1193.85 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M599.81 1520.52 L640.009 1520.52 L640.009 1525.93 L623.14 1525.93 L623.14 1568.04 L616.679 1568.04 L616.679 1525.93 L599.81 1525.93 L599.81 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M643.956 1532.4 L649.812 1532.4 L649.812 1568.04 L643.956 1568.04 L643.956 1532.4 M643.956 1518.52 L649.812 1518.52 L649.812 1525.93 L643.956 1525.93 L643.956 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M689.821 1539.24 Q692.017 1535.29 695.073 1533.41 Q698.128 1531.54 702.266 1531.54 Q707.836 1531.54 710.86 1535.45 Q713.883 1539.33 713.883 1546.53 L713.883 1568.04 L707.995 1568.04 L707.995 1546.72 Q707.995 1541.59 706.181 1539.11 Q704.367 1536.63 700.643 1536.63 Q696.091 1536.63 693.449 1539.65 Q690.808 1542.68 690.808 1547.9 L690.808 1568.04 L684.919 1568.04 L684.919 1546.72 Q684.919 1541.56 683.105 1539.11 Q681.291 1536.63 677.503 1536.63 Q673.015 1536.63 670.374 1539.68 Q667.732 1542.71 667.732 1547.9 L667.732 1568.04 L661.844 1568.04 L661.844 1532.4 L667.732 1532.4 L667.732 1537.93 Q669.737 1534.66 672.538 1533.1 Q675.339 1531.54 679.19 1531.54 Q683.073 1531.54 685.779 1533.51 Q688.516 1535.48 689.821 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M756.056 1548.76 L756.056 1551.62 L729.129 1551.62 Q729.511 1557.67 732.758 1560.85 Q736.036 1564 741.861 1564 Q745.234 1564 748.385 1563.17 Q751.568 1562.35 754.687 1560.69 L754.687 1566.23 Q751.536 1567.57 748.226 1568.27 Q744.916 1568.97 741.51 1568.97 Q732.98 1568.97 727.983 1564 Q723.018 1559.04 723.018 1550.57 Q723.018 1541.82 727.729 1536.69 Q732.471 1531.54 740.492 1531.54 Q747.685 1531.54 751.855 1536.18 Q756.056 1540.8 756.056 1548.76 M750.2 1547.04 Q750.136 1542.23 747.494 1539.37 Q744.884 1536.5 740.556 1536.5 Q735.654 1536.5 732.694 1539.27 Q729.766 1542.04 729.32 1547.07 L750.2 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M800.457 1518.58 Q796.192 1525.9 794.123 1533.06 Q792.054 1540.23 792.054 1547.58 Q792.054 1554.93 794.123 1562.16 Q796.224 1569.35 800.457 1576.64 L795.364 1576.64 Q790.59 1569.16 788.203 1561.93 Q785.848 1554.71 785.848 1547.58 Q785.848 1540.48 788.203 1533.29 Q790.558 1526.09 795.364 1518.58 L800.457 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M834.545 1533.45 L834.545 1538.98 Q832.063 1537.71 829.389 1537.07 Q826.715 1536.44 823.851 1536.44 Q819.49 1536.44 817.294 1537.77 Q815.13 1539.11 815.13 1541.79 Q815.13 1543.82 816.689 1545 Q818.249 1546.15 822.96 1547.2 L824.965 1547.64 Q831.203 1548.98 833.813 1551.43 Q836.455 1553.85 836.455 1558.21 Q836.455 1563.17 832.508 1566.07 Q828.593 1568.97 821.718 1568.97 Q818.854 1568.97 815.735 1568.39 Q812.647 1567.85 809.21 1566.74 L809.21 1560.69 Q812.456 1562.38 815.607 1563.24 Q818.758 1564.07 821.846 1564.07 Q825.983 1564.07 828.211 1562.66 Q830.439 1561.23 830.439 1558.65 Q830.439 1556.27 828.816 1554.99 Q827.225 1553.72 821.782 1552.54 L819.745 1552.07 Q814.302 1550.92 811.883 1548.56 Q809.464 1546.18 809.464 1542.04 Q809.464 1537.01 813.029 1534.27 Q816.594 1531.54 823.151 1531.54 Q826.397 1531.54 829.262 1532.01 Q832.126 1532.49 834.545 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M844.858 1518.58 L849.95 1518.58 Q854.725 1526.09 857.08 1533.29 Q859.467 1540.48 859.467 1547.58 Q859.467 1554.71 857.08 1561.93 Q854.725 1569.16 849.95 1576.64 L844.858 1576.64 Q849.091 1569.35 851.16 1562.16 Q853.26 1554.93 853.26 1547.58 Q853.26 1540.23 851.16 1533.06 Q849.091 1525.9 844.858 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1406.88 1181.75,1406.88 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1322.61 1181.75,1322.61 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1238.34 1181.75,1238.34 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1154.07 1181.75,1154.07 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1069.8 1181.75,1069.8 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,985.533 1181.75,985.533 "/>
-<polyline clip-path="url(#clip154)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,901.264 1181.75,901.264 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1423.18 277.528,847.244 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1406.88 296.426,1406.88 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1322.61 296.426,1322.61 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1238.34 296.426,1238.34 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1154.07 296.426,1154.07 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1069.8 296.426,1069.8 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,985.533 296.426,985.533 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,901.264 296.426,901.264 "/>
-<path clip-path="url(#clip150)" d="M154.19 1392.68 Q150.579 1392.68 148.751 1396.24 Q146.945 1399.78 146.945 1406.91 Q146.945 1414.02 148.751 1417.59 Q150.579 1421.13 154.19 1421.13 Q157.825 1421.13 159.63 1417.59 Q161.459 1414.02 161.459 1406.91 Q161.459 1399.78 159.63 1396.24 Q157.825 1392.68 154.19 1392.68 M154.19 1388.97 Q160.001 1388.97 163.056 1393.58 Q166.135 1398.16 166.135 1406.91 Q166.135 1415.64 163.056 1420.25 Q160.001 1424.83 154.19 1424.83 Q148.38 1424.83 145.302 1420.25 Q142.246 1415.64 142.246 1406.91 Q142.246 1398.16 145.302 1393.58 Q148.38 1388.97 154.19 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M174.352 1418.28 L179.237 1418.28 L179.237 1424.16 L174.352 1424.16 L174.352 1418.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M199.422 1392.68 Q195.811 1392.68 193.982 1396.24 Q192.176 1399.78 192.176 1406.91 Q192.176 1414.02 193.982 1417.59 Q195.811 1421.13 199.422 1421.13 Q203.056 1421.13 204.862 1417.59 Q206.69 1414.02 206.69 1406.91 Q206.69 1399.78 204.862 1396.24 Q203.056 1392.68 199.422 1392.68 M199.422 1388.97 Q205.232 1388.97 208.287 1393.58 Q211.366 1398.16 211.366 1406.91 Q211.366 1415.64 208.287 1420.25 Q205.232 1424.83 199.422 1424.83 Q193.612 1424.83 190.533 1420.25 Q187.477 1415.64 187.477 1406.91 Q187.477 1398.16 190.533 1393.58 Q193.612 1388.97 199.422 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M229.584 1392.68 Q225.973 1392.68 224.144 1396.24 Q222.338 1399.78 222.338 1406.91 Q222.338 1414.02 224.144 1417.59 Q225.973 1421.13 229.584 1421.13 Q233.218 1421.13 235.023 1417.59 Q236.852 1414.02 236.852 1406.91 Q236.852 1399.78 235.023 1396.24 Q233.218 1392.68 229.584 1392.68 M229.584 1388.97 Q235.394 1388.97 238.449 1393.58 Q241.528 1398.16 241.528 1406.91 Q241.528 1415.64 238.449 1420.25 Q235.394 1424.83 229.584 1424.83 Q223.773 1424.83 220.695 1420.25 Q217.639 1415.64 217.639 1406.91 Q217.639 1398.16 220.695 1393.58 Q223.773 1388.97 229.584 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M155.788 1308.41 Q152.177 1308.41 150.348 1311.97 Q148.542 1315.52 148.542 1322.65 Q148.542 1329.75 150.348 1333.32 Q152.177 1336.86 155.788 1336.86 Q159.422 1336.86 161.227 1333.32 Q163.056 1329.75 163.056 1322.65 Q163.056 1315.52 161.227 1311.97 Q159.422 1308.41 155.788 1308.41 M155.788 1304.71 Q161.598 1304.71 164.653 1309.31 Q167.732 1313.9 167.732 1322.65 Q167.732 1331.37 164.653 1335.98 Q161.598 1340.56 155.788 1340.56 Q149.978 1340.56 146.899 1335.98 Q143.843 1331.37 143.843 1322.65 Q143.843 1313.9 146.899 1309.31 Q149.978 1304.71 155.788 1304.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M175.95 1334.01 L180.834 1334.01 L180.834 1339.89 L175.95 1339.89 L175.95 1334.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M201.019 1308.41 Q197.408 1308.41 195.579 1311.97 Q193.774 1315.52 193.774 1322.65 Q193.774 1329.75 195.579 1333.32 Q197.408 1336.86 201.019 1336.86 Q204.653 1336.86 206.459 1333.32 Q208.287 1329.75 208.287 1322.65 Q208.287 1315.52 206.459 1311.97 Q204.653 1308.41 201.019 1308.41 M201.019 1304.71 Q206.829 1304.71 209.885 1309.31 Q212.963 1313.9 212.963 1322.65 Q212.963 1331.37 209.885 1335.98 Q206.829 1340.56 201.019 1340.56 Q195.209 1340.56 192.13 1335.98 Q189.075 1331.37 189.075 1322.65 Q189.075 1313.9 192.13 1309.31 Q195.209 1304.71 201.019 1304.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M225.209 1335.96 L241.528 1335.96 L241.528 1339.89 L219.584 1339.89 L219.584 1335.96 Q222.246 1333.2 226.829 1328.57 Q231.435 1323.92 232.616 1322.58 Q234.861 1320.05 235.741 1318.32 Q236.644 1316.56 236.644 1314.87 Q236.644 1312.11 234.699 1310.38 Q232.778 1308.64 229.676 1308.64 Q227.477 1308.64 225.023 1309.4 Q222.593 1310.17 219.815 1311.72 L219.815 1307 Q222.639 1305.86 225.093 1305.28 Q227.547 1304.71 229.584 1304.71 Q234.954 1304.71 238.148 1307.39 Q241.343 1310.08 241.343 1314.57 Q241.343 1316.7 240.533 1318.62 Q239.746 1320.52 237.639 1323.11 Q237.06 1323.78 233.959 1327 Q230.857 1330.19 225.209 1335.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M153.704 1224.14 Q150.093 1224.14 148.265 1227.7 Q146.459 1231.25 146.459 1238.38 Q146.459 1245.48 148.265 1249.05 Q150.093 1252.59 153.704 1252.59 Q157.339 1252.59 159.144 1249.05 Q160.973 1245.48 160.973 1238.38 Q160.973 1231.25 159.144 1227.7 Q157.339 1224.14 153.704 1224.14 M153.704 1220.44 Q159.515 1220.44 162.57 1225.04 Q165.649 1229.63 165.649 1238.38 Q165.649 1247.1 162.57 1251.71 Q159.515 1256.29 153.704 1256.29 Q147.894 1256.29 144.816 1251.71 Q141.76 1247.1 141.76 1238.38 Q141.76 1229.63 144.816 1225.04 Q147.894 1220.44 153.704 1220.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M173.866 1249.74 L178.751 1249.74 L178.751 1255.62 L173.866 1255.62 L173.866 1249.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M198.936 1224.14 Q195.325 1224.14 193.496 1227.7 Q191.69 1231.25 191.69 1238.38 Q191.69 1245.48 193.496 1249.05 Q195.325 1252.59 198.936 1252.59 Q202.57 1252.59 204.375 1249.05 Q206.204 1245.48 206.204 1238.38 Q206.204 1231.25 204.375 1227.7 Q202.57 1224.14 198.936 1224.14 M198.936 1220.44 Q204.746 1220.44 207.801 1225.04 Q210.88 1229.63 210.88 1238.38 Q210.88 1247.1 207.801 1251.71 Q204.746 1256.29 198.936 1256.29 Q193.125 1256.29 190.047 1251.71 Q186.991 1247.1 186.991 1238.38 Q186.991 1229.63 190.047 1225.04 Q193.125 1220.44 198.936 1220.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M231.945 1225.14 L220.139 1243.58 L231.945 1243.58 L231.945 1225.14 M230.718 1221.06 L236.597 1221.06 L236.597 1243.58 L241.528 1243.58 L241.528 1247.47 L236.597 1247.47 L236.597 1255.62 L231.945 1255.62 L231.945 1247.47 L216.343 1247.47 L216.343 1242.96 L230.718 1221.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M154.028 1139.87 Q150.417 1139.87 148.589 1143.44 Q146.783 1146.98 146.783 1154.11 Q146.783 1161.21 148.589 1164.78 Q150.417 1168.32 154.028 1168.32 Q157.663 1168.32 159.468 1164.78 Q161.297 1161.21 161.297 1154.11 Q161.297 1146.98 159.468 1143.44 Q157.663 1139.87 154.028 1139.87 M154.028 1136.17 Q159.839 1136.17 162.894 1140.77 Q165.973 1145.36 165.973 1154.11 Q165.973 1162.83 162.894 1167.44 Q159.839 1172.02 154.028 1172.02 Q148.218 1172.02 145.14 1167.44 Q142.084 1162.83 142.084 1154.11 Q142.084 1145.36 145.14 1140.77 Q148.218 1136.17 154.028 1136.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M174.19 1165.47 L179.075 1165.47 L179.075 1171.35 L174.19 1171.35 L174.19 1165.47 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M199.26 1139.87 Q195.649 1139.87 193.82 1143.44 Q192.014 1146.98 192.014 1154.11 Q192.014 1161.21 193.82 1164.78 Q195.649 1168.32 199.26 1168.32 Q202.894 1168.32 204.699 1164.78 Q206.528 1161.21 206.528 1154.11 Q206.528 1146.98 204.699 1143.44 Q202.894 1139.87 199.26 1139.87 M199.26 1136.17 Q205.07 1136.17 208.125 1140.77 Q211.204 1145.36 211.204 1154.11 Q211.204 1162.83 208.125 1167.44 Q205.07 1172.02 199.26 1172.02 Q193.45 1172.02 190.371 1167.44 Q187.315 1162.83 187.315 1154.11 Q187.315 1145.36 190.371 1140.77 Q193.45 1136.17 199.26 1136.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M230 1152.21 Q226.852 1152.21 225 1154.36 Q223.172 1156.51 223.172 1160.26 Q223.172 1163.99 225 1166.17 Q226.852 1168.32 230 1168.32 Q233.148 1168.32 234.977 1166.17 Q236.829 1163.99 236.829 1160.26 Q236.829 1156.51 234.977 1154.36 Q233.148 1152.21 230 1152.21 M239.283 1137.56 L239.283 1141.82 Q237.523 1140.98 235.718 1140.54 Q233.935 1140.1 232.176 1140.1 Q227.547 1140.1 225.093 1143.23 Q222.662 1146.35 222.315 1152.67 Q223.681 1150.66 225.741 1149.59 Q227.801 1148.5 230.278 1148.5 Q235.486 1148.5 238.496 1151.68 Q241.528 1154.82 241.528 1160.26 Q241.528 1165.59 238.38 1168.81 Q235.232 1172.02 230 1172.02 Q224.005 1172.02 220.834 1167.44 Q217.662 1162.83 217.662 1154.11 Q217.662 1145.91 221.551 1141.05 Q225.44 1136.17 231.991 1136.17 Q233.75 1136.17 235.533 1136.51 Q237.338 1136.86 239.283 1137.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M154.283 1055.6 Q150.672 1055.6 148.843 1059.17 Q147.038 1062.71 147.038 1069.84 Q147.038 1076.94 148.843 1080.51 Q150.672 1084.05 154.283 1084.05 Q157.917 1084.05 159.723 1080.51 Q161.552 1076.94 161.552 1069.84 Q161.552 1062.71 159.723 1059.17 Q157.917 1055.6 154.283 1055.6 M154.283 1051.9 Q160.093 1051.9 163.149 1056.5 Q166.227 1061.09 166.227 1069.84 Q166.227 1078.56 163.149 1083.17 Q160.093 1087.75 154.283 1087.75 Q148.473 1087.75 145.394 1083.17 Q142.339 1078.56 142.339 1069.84 Q142.339 1061.09 145.394 1056.5 Q148.473 1051.9 154.283 1051.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M174.445 1081.2 L179.329 1081.2 L179.329 1087.08 L174.445 1087.08 L174.445 1081.2 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M199.514 1055.6 Q195.903 1055.6 194.075 1059.17 Q192.269 1062.71 192.269 1069.84 Q192.269 1076.94 194.075 1080.51 Q195.903 1084.05 199.514 1084.05 Q203.149 1084.05 204.954 1080.51 Q206.783 1076.94 206.783 1069.84 Q206.783 1062.71 204.954 1059.17 Q203.149 1055.6 199.514 1055.6 M199.514 1051.9 Q205.324 1051.9 208.38 1056.5 Q211.459 1061.09 211.459 1069.84 Q211.459 1078.56 208.38 1083.17 Q205.324 1087.75 199.514 1087.75 Q193.704 1087.75 190.625 1083.17 Q187.57 1078.56 187.57 1069.84 Q187.57 1061.09 190.625 1056.5 Q193.704 1051.9 199.514 1051.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M229.676 1070.67 Q226.343 1070.67 224.422 1072.45 Q222.523 1074.24 222.523 1077.36 Q222.523 1080.49 224.422 1082.27 Q226.343 1084.05 229.676 1084.05 Q233.01 1084.05 234.931 1082.27 Q236.852 1080.46 236.852 1077.36 Q236.852 1074.24 234.931 1072.45 Q233.033 1070.67 229.676 1070.67 M225 1068.68 Q221.991 1067.94 220.301 1065.88 Q218.635 1063.82 218.635 1060.86 Q218.635 1056.71 221.574 1054.31 Q224.537 1051.9 229.676 1051.9 Q234.838 1051.9 237.778 1054.31 Q240.718 1056.71 240.718 1060.86 Q240.718 1063.82 239.028 1065.88 Q237.361 1067.94 234.375 1068.68 Q237.755 1069.47 239.63 1071.76 Q241.528 1074.05 241.528 1077.36 Q241.528 1082.38 238.449 1085.07 Q235.394 1087.75 229.676 1087.75 Q223.959 1087.75 220.88 1085.07 Q217.824 1082.38 217.824 1077.36 Q217.824 1074.05 219.723 1071.76 Q221.621 1069.47 225 1068.68 M223.287 1061.3 Q223.287 1063.98 224.954 1065.49 Q226.644 1066.99 229.676 1066.99 Q232.685 1066.99 234.375 1065.49 Q236.088 1063.98 236.088 1061.3 Q236.088 1058.61 234.375 1057.11 Q232.685 1055.6 229.676 1055.6 Q226.644 1055.6 224.954 1057.11 Q223.287 1058.61 223.287 1061.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M154.19 971.332 Q150.579 971.332 148.751 974.897 Q146.945 978.439 146.945 985.568 Q146.945 992.675 148.751 996.239 Q150.579 999.781 154.19 999.781 Q157.825 999.781 159.63 996.239 Q161.459 992.675 161.459 985.568 Q161.459 978.439 159.63 974.897 Q157.825 971.332 154.19 971.332 M154.19 967.628 Q160.001 967.628 163.056 972.235 Q166.135 976.818 166.135 985.568 Q166.135 994.295 163.056 998.901 Q160.001 1003.48 154.19 1003.48 Q148.38 1003.48 145.302 998.901 Q142.246 994.295 142.246 985.568 Q142.246 976.818 145.302 972.235 Q148.38 967.628 154.19 967.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M174.352 996.934 L179.237 996.934 L179.237 1002.81 L174.352 1002.81 L174.352 996.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M190.232 998.878 L197.871 998.878 L197.871 972.513 L189.561 974.179 L189.561 969.92 L197.825 968.253 L202.5 968.253 L202.5 998.878 L210.139 998.878 L210.139 1002.81 L190.232 1002.81 L190.232 998.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M229.584 971.332 Q225.973 971.332 224.144 974.897 Q222.338 978.439 222.338 985.568 Q222.338 992.675 224.144 996.239 Q225.973 999.781 229.584 999.781 Q233.218 999.781 235.023 996.239 Q236.852 992.675 236.852 985.568 Q236.852 978.439 235.023 974.897 Q233.218 971.332 229.584 971.332 M229.584 967.628 Q235.394 967.628 238.449 972.235 Q241.528 976.818 241.528 985.568 Q241.528 994.295 238.449 998.901 Q235.394 1003.48 229.584 1003.48 Q223.773 1003.48 220.695 998.901 Q217.639 994.295 217.639 985.568 Q217.639 976.818 220.695 972.235 Q223.773 967.628 229.584 967.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M155.788 887.063 Q152.177 887.063 150.348 890.628 Q148.542 894.169 148.542 901.299 Q148.542 908.405 150.348 911.97 Q152.177 915.512 155.788 915.512 Q159.422 915.512 161.227 911.97 Q163.056 908.405 163.056 901.299 Q163.056 894.169 161.227 890.628 Q159.422 887.063 155.788 887.063 M155.788 883.359 Q161.598 883.359 164.653 887.966 Q167.732 892.549 167.732 901.299 Q167.732 910.026 164.653 914.632 Q161.598 919.215 155.788 919.215 Q149.978 919.215 146.899 914.632 Q143.843 910.026 143.843 901.299 Q143.843 892.549 146.899 887.966 Q149.978 883.359 155.788 883.359 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M175.95 912.665 L180.834 912.665 L180.834 918.544 L175.95 918.544 L175.95 912.665 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M191.829 914.609 L199.468 914.609 L199.468 888.243 L191.158 889.91 L191.158 885.651 L199.422 883.984 L204.098 883.984 L204.098 914.609 L211.736 914.609 L211.736 918.544 L191.829 918.544 L191.829 914.609 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M225.209 914.609 L241.528 914.609 L241.528 918.544 L219.584 918.544 L219.584 914.609 Q222.246 911.854 226.829 907.225 Q231.435 902.572 232.616 901.229 Q234.861 898.706 235.741 896.97 Q236.644 895.211 236.644 893.521 Q236.644 890.767 234.699 889.03 Q232.778 887.294 229.676 887.294 Q227.477 887.294 225.023 888.058 Q222.593 888.822 219.815 890.373 L219.815 885.651 Q222.639 884.517 225.093 883.938 Q227.547 883.359 229.584 883.359 Q234.954 883.359 238.148 886.044 Q241.343 888.73 241.343 893.22 Q241.343 895.35 240.533 897.271 Q239.746 899.169 237.639 901.762 Q237.06 902.433 233.959 905.651 Q230.857 908.845 225.209 914.609 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M91.504 1243.7 L43.984 1261.84 L43.984 1255.13 L83.9925 1240.07 L43.984 1224.98 L43.984 1218.3 L91.504 1236.41 L91.504 1243.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M72.2159 1186.19 L75.0805 1186.19 L75.0805 1213.11 Q81.1279 1212.73 84.3108 1209.48 Q87.4618 1206.21 87.4618 1200.38 Q87.4618 1197.01 86.6342 1193.86 Q85.8067 1190.67 84.1516 1187.55 L89.6898 1187.55 Q91.0266 1190.71 91.7268 1194.02 Q92.427 1197.33 92.427 1200.73 Q92.427 1209.26 87.4618 1214.26 Q82.4965 1219.22 74.0301 1219.22 Q65.2773 1219.22 60.1529 1214.51 Q54.9967 1209.77 54.9967 1201.75 Q54.9967 1194.56 59.6436 1190.39 Q64.2588 1186.19 72.2159 1186.19 M70.4972 1192.04 Q65.6911 1192.11 62.8265 1194.75 Q59.9619 1197.36 59.9619 1201.69 Q59.9619 1206.59 62.731 1209.55 Q65.5001 1212.48 70.529 1212.92 L70.4972 1192.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M41.9788 1176.57 L41.9788 1170.72 L91.504 1170.72 L91.504 1176.57 L41.9788 1176.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M59.9619 1144.65 Q59.9619 1149.36 63.654 1152.1 Q67.3143 1154.83 73.7119 1154.83 Q80.1094 1154.83 83.8015 1152.13 Q87.4618 1149.39 87.4618 1144.65 Q87.4618 1139.97 83.7697 1137.23 Q80.0776 1134.5 73.7119 1134.5 Q67.378 1134.5 63.6859 1137.23 Q59.9619 1139.97 59.9619 1144.65 M54.9967 1144.65 Q54.9967 1137.01 59.9619 1132.65 Q64.9272 1128.29 73.7119 1128.29 Q82.4647 1128.29 87.4618 1132.65 Q92.427 1137.01 92.427 1144.65 Q92.427 1152.32 87.4618 1156.68 Q82.4647 1161.01 73.7119 1161.01 Q64.9272 1161.01 59.9619 1156.68 Q54.9967 1152.32 54.9967 1144.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M57.2247 1092.93 L62.6992 1092.93 Q61.3306 1095.41 60.6622 1097.92 Q59.9619 1100.41 59.9619 1102.95 Q59.9619 1108.65 63.5904 1111.8 Q67.187 1114.95 73.7119 1114.95 Q80.2367 1114.95 83.8652 1111.8 Q87.4618 1108.65 87.4618 1102.95 Q87.4618 1100.41 86.7934 1097.92 Q86.0932 1095.41 84.7245 1092.93 L90.1354 1092.93 Q91.2812 1095.38 91.8541 1098.02 Q92.427 1100.63 92.427 1103.59 Q92.427 1111.64 87.3663 1116.39 Q82.3056 1121.13 73.7119 1121.13 Q64.9908 1121.13 59.9938 1116.35 Q54.9967 1111.55 54.9967 1103.21 Q54.9967 1100.5 55.5696 1097.92 Q56.1107 1095.35 57.2247 1092.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M55.8561 1082.74 L55.8561 1076.89 L91.504 1076.89 L91.504 1082.74 L55.8561 1082.74 M41.9788 1082.74 L41.9788 1076.89 L49.3949 1076.89 L49.3949 1082.74 L41.9788 1082.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M45.7346 1058.84 L55.8561 1058.84 L55.8561 1046.78 L60.4075 1046.78 L60.4075 1058.84 L79.7593 1058.84 Q84.1198 1058.84 85.3611 1057.66 Q86.6024 1056.45 86.6024 1052.79 L86.6024 1046.78 L91.504 1046.78 L91.504 1052.79 Q91.504 1059.57 88.9896 1062.15 Q86.4433 1064.73 79.7593 1064.73 L60.4075 1064.73 L60.4075 1069.02 L55.8561 1069.02 L55.8561 1064.73 L45.7346 1064.73 L45.7346 1058.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M94.8142 1024.24 Q101.18 1026.72 103.121 1029.08 Q105.063 1031.44 105.063 1035.38 L105.063 1040.06 L100.161 1040.06 L100.161 1036.62 Q100.161 1034.2 99.0155 1032.87 Q97.8697 1031.53 93.6047 1029.91 L90.9311 1028.86 L55.8561 1043.28 L55.8561 1037.07 L83.7378 1025.93 L55.8561 1014.79 L55.8561 1008.58 L94.8142 1024.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip154)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.119,1406.88 303.973,1404.77 304.827,1402.66 305.681,1400.55 306.535,1398.43 307.389,1396.31 308.243,1394.18 309.096,1392.05 309.95,1389.92 310.804,1387.79 311.658,1385.65 312.512,1383.5 313.366,1381.36 314.22,1379.21 315.074,1377.06 315.928,1374.91 316.781,1372.76 317.635,1370.6 318.489,1368.44 319.343,1366.28 320.197,1364.12 321.051,1361.95 321.905,1359.79 322.759,1357.62 323.613,1355.45 324.466,1353.29 325.32,1351.11 326.174,1348.94 327.028,1346.77 327.882,1344.6 328.736,1342.43 329.59,1340.25 330.444,1338.08 331.298,1335.91 332.151,1333.73 333.005,1331.56 333.859,1329.39 334.713,1327.21 335.567,1325.04 336.421,1322.87 337.275,1320.7 338.129,1318.53 338.983,1316.36 339.837,1314.19 340.69,1312.03 341.544,1309.86 342.398,1307.7 343.252,1305.53 344.106,1303.37 344.96,1301.21 345.814,1299.06 346.668,1296.9 347.522,1294.75 348.375,1292.6 349.229,1290.45 350.083,1288.3 350.937,1286.16 351.791,1284.02 352.645,1281.88 353.499,1279.74 354.353,1277.61 355.207,1275.47 356.06,1273.35 356.914,1271.22 357.768,1269.1 358.622,1266.98 359.476,1264.86 360.33,1262.75 361.184,1260.64 362.038,1258.54 362.892,1256.43 363.746,1254.33 364.599,1252.24 365.453,1250.15 366.307,1248.06 367.161,1245.97 368.015,1243.89 368.869,1241.81 369.723,1239.74 370.577,1237.67 371.431,1235.6 372.284,1233.54 373.138,1231.48 373.992,1229.43 374.846,1227.38 375.7,1225.33 376.554,1223.29 377.408,1221.25 378.262,1219.22 379.116,1217.19 379.969,1215.16 380.823,1213.14 381.677,1211.12 382.531,1209.11 383.385,1207.1 384.239,1205.1 385.093,1203.1 385.947,1201.11 386.801,1199.12 387.654,1197.14 388.508,1195.16 389.362,1193.19 390.216,1191.23 391.07,1189.28 391.924,1187.33 392.778,1185.4 393.632,1183.49 394.486,1181.6 395.34,1179.74 396.193,1177.94 397.047,1176.22 397.901,1174.63 398.755,1173.21 399.609,1171.94 400.463,1170.82 401.317,1169.81 402.171,1168.91 403.025,1168.08 403.878,1167.31 404.732,1166.59 405.586,1165.9 406.44,1165.25 407.294,1164.62 408.148,1164 409.002,1163.4 409.856,1162.8 410.71,1162.21 411.563,1161.63 412.417,1161.05 413.271,1160.48 414.125,1159.9 414.979,1159.32 415.833,1158.75 416.687,1158.17 417.541,1157.59 418.395,1157.01 419.249,1156.43 420.102,1155.85 420.956,1155.26 421.81,1154.67 422.664,1154.08 423.518,1153.49 424.372,1152.89 425.226,1152.29 426.08,1151.69 426.934,1151.09 427.787,1150.48 428.641,1149.87 429.495,1149.26 430.349,1148.64 431.203,1148.02 432.057,1147.4 432.911,1146.77 433.765,1146.14 434.619,1145.51 435.472,1144.88 436.326,1144.24 437.18,1143.6 438.034,1142.96 438.888,1142.31 439.742,1141.66 440.596,1141.01 441.45,1140.35 442.304,1139.69 443.158,1139.02 444.011,1138.36 444.865,1137.69 445.719,1137.01 446.573,1136.34 447.427,1135.65 448.281,1134.97 449.135,1134.28 449.989,1133.59 450.843,1132.9 451.696,1132.2 452.55,1131.49 453.404,1130.79 454.258,1130.08 455.112,1129.37 455.966,1128.65 456.82,1127.93 457.674,1127.2 458.528,1126.47 459.381,1125.74 460.235,1125.01 461.089,1124.27 461.943,1123.52 462.797,1122.77 463.651,1122.02 464.505,1121.27 465.359,1120.51 466.213,1119.74 467.066,1118.97 467.92,1118.2 468.774,1117.42 469.628,1116.64 470.482,1115.86 471.336,1115.07 472.19,1114.27 473.044,1113.47 473.898,1112.67 474.752,1111.86 475.605,1111.05 476.459,1110.24 477.313,1109.41 478.167,1108.59 479.021,1107.76 479.875,1106.92 480.729,1106.09 481.583,1105.24 482.437,1104.39 483.29,1103.54 484.144,1102.68 484.998,1101.82 485.852,1100.95 486.706,1100.07 487.56,1099.2 488.414,1098.31 489.268,1097.42 490.122,1096.53 490.975,1095.63 491.829,1094.73 492.683,1093.82 493.537,1092.9 494.391,1091.98 495.245,1091.05 496.099,1090.12 496.953,1089.19 497.807,1088.24 498.661,1087.3 499.514,1086.34 500.368,1085.38 501.222,1084.42 502.076,1083.45 502.93,1082.47 503.784,1081.49 504.638,1080.5 505.492,1079.5 506.346,1078.5 507.199,1077.5 508.053,1076.48 508.907,1075.47 509.761,1074.44 510.615,1073.41 511.469,1072.37 512.323,1071.32 513.177,1070.27 514.031,1069.22 514.884,1068.15 515.738,1067.08 516.592,1066 517.446,1064.92 518.3,1063.83 519.154,1062.73 520.008,1061.62 520.862,1060.51 521.716,1059.39 522.569,1058.26 523.423,1057.13 524.277,1055.99 525.131,1054.84 525.985,1053.68 526.839,1052.52 527.693,1051.34 528.547,1050.16 529.401,1048.98 530.255,1047.78 531.108,1046.58 531.962,1045.37 532.816,1044.15 533.67,1042.92 534.524,1041.69 535.378,1040.44 536.232,1039.19 537.086,1037.93 537.94,1036.66 538.793,1035.38 539.647,1034.1 540.501,1032.8 541.355,1031.5 542.209,1030.18 543.063,1028.86 543.917,1027.53 544.771,1026.19 545.625,1024.84 546.478,1023.48 547.332,1022.11 548.186,1020.73 549.04,1019.34 549.894,1017.95 550.748,1016.54 551.602,1015.12 552.456,1013.69 553.31,1012.25 554.164,1010.8 555.017,1009.34 555.871,1007.87 556.725,1006.39 557.579,1004.9 558.433,1003.4 559.287,1001.88 560.141,1000.36 560.995,998.823 561.849,997.276 562.702,995.717 563.556,994.147 564.41,992.565 565.264,990.971 566.118,989.365 566.972,987.747 567.826,986.117 568.68,984.475 569.534,982.82 570.387,981.152 571.241,979.472 572.095,977.779 572.949,976.074 573.803,974.355 574.657,972.623 575.511,970.877 576.365,969.119 577.219,967.346 578.072,965.561 578.926,963.761 579.78,961.948 580.634,960.121 581.488,958.28 582.342,956.425 583.196,954.556 584.05,952.673 584.904,950.776 585.758,948.865 586.611,946.939 587.465,945 588.319,943.046 589.173,941.079 590.027,939.098 590.881,937.103 591.735,935.094 592.589,933.072 593.443,931.038 594.296,928.99 595.15,926.931 596.004,924.859 596.858,922.776 597.712,920.682 598.566,918.579 599.42,916.466 600.274,914.345 601.128,912.217 601.981,910.082 602.835,907.943 603.689,905.801 604.543,903.657 605.397,901.514 606.251,899.374 607.105,897.239 607.959,895.112 608.813,892.997 609.667,890.896 610.52,888.814 611.374,886.756 612.228,884.726 613.082,882.731 613.936,880.776 614.79,878.869 615.644,877.018 616.498,875.232 617.352,873.52 618.205,871.894 619.059,870.367 619.913,868.95 620.767,867.66 621.621,866.511 622.475,865.522 623.329,864.712 624.183,864.098 625.037,863.703 625.89,863.544 626.744,863.64 627.598,864 628.452,864.622 629.306,865.479 630.16,866.509 631.014,867.645 631.868,868.837 632.722,870.06 633.576,871.3 634.429,872.548 635.283,873.799 636.137,875.052 636.991,876.303 637.845,877.552 638.699,878.798 639.553,880.039 640.407,881.276 641.261,882.509 642.114,883.736 642.968,884.958 643.822,886.174 644.676,887.385 645.53,888.591 646.384,889.791 647.238,890.986 648.092,892.176 648.946,893.36 649.799,894.538 650.653,895.711 651.507,896.879 652.361,898.042 653.215,899.2 654.069,900.352 654.923,901.5 655.777,902.642 656.631,903.78 657.484,904.913 658.338,906.041 659.192,907.164 660.046,908.283 660.9,909.397 661.754,910.507 662.608,911.613 663.462,912.714 664.316,913.811 665.17,914.903 666.023,915.992 666.877,917.076 667.731,918.157 668.585,919.233 669.439,920.306 670.293,921.375 671.147,922.44 672.001,923.502 672.855,924.56 673.708,925.614 674.562,926.665 675.416,927.713 676.27,928.757 677.124,929.798 677.978,930.835 678.832,931.87 679.686,932.901 680.54,933.929 681.393,934.954 682.247,935.976 683.101,936.995 683.955,938.011 684.809,939.025 685.663,940.035 686.517,941.043 687.371,942.048 688.225,943.05 689.079,944.05 689.932,945.047 690.786,946.042 691.64,947.034 692.494,948.023 693.348,949.011 694.202,949.996 695.056,950.978 695.91,951.958 696.764,952.936 697.617,953.912 698.471,954.885 699.325,955.857 700.179,956.826 701.033,957.793 701.887,958.758 702.741,959.721 703.595,960.682 704.449,961.641 705.302,962.598 706.156,963.553 707.01,964.507 707.864,965.458 708.718,966.408 709.572,967.356 710.426,968.302 711.28,969.247 712.134,970.189 712.987,971.13 713.841,972.07 714.695,973.008 715.549,973.944 716.403,974.878 717.257,975.811 718.111,976.743 718.965,977.673 719.819,978.602 720.673,979.529 721.526,980.454 722.38,981.379 723.234,982.302 724.088,983.223 724.942,984.143 725.796,985.062 726.65,985.98 727.504,986.896 728.358,987.811 729.211,988.724 730.065,989.637 730.919,990.548 731.773,991.458 732.627,992.367 733.481,993.275 734.335,994.182 735.189,995.087 736.043,995.991 736.896,996.895 737.75,997.797 738.604,998.698 739.458,999.598 740.312,1000.5 741.166,1001.4 742.02,1002.29 742.874,1003.19 743.728,1004.08 744.582,1004.98 745.435,1005.87 746.289,1006.76 747.143,1007.65 747.997,1008.55 748.851,1009.43 749.705,1010.32 750.559,1011.21 751.413,1012.1 752.267,1012.98 753.12,1013.87 753.974,1014.75 754.828,1015.64 755.682,1016.52 756.536,1017.4 757.39,1018.28 758.244,1019.17 759.098,1020.05 759.952,1020.92 760.805,1021.8 761.659,1022.68 762.513,1023.56 763.367,1024.43 764.221,1025.31 765.075,1026.18 765.929,1027.06 766.783,1027.93 767.637,1028.81 768.491,1029.68 769.344,1030.55 770.198,1031.42 771.052,1032.29 771.906,1033.16 772.76,1034.03 773.614,1034.9 774.468,1035.77 775.322,1036.63 776.176,1037.5 777.029,1038.37 777.883,1039.23 778.737,1040.1 779.591,1040.96 780.445,1041.83 781.299,1042.69 782.153,1043.56 783.007,1044.42 783.861,1045.28 784.714,1046.14 785.568,1047 786.422,1047.86 787.276,1048.72 788.13,1049.58 788.984,1050.44 789.838,1051.3 790.692,1052.16 791.546,1053.02 792.399,1053.88 793.253,1054.73 794.107,1055.59 794.961,1056.45 795.815,1057.3 796.669,1058.16 797.523,1059.01 798.377,1059.87 799.231,1060.72 800.085,1061.58 800.938,1062.43 801.792,1063.28 802.646,1064.14 803.5,1064.99 804.354,1065.84 805.208,1066.69 806.062,1067.54 806.916,1068.39 807.77,1069.25 808.623,1070.1 809.477,1070.95 810.331,1071.8 811.185,1072.65 812.039,1073.49 812.893,1074.34 813.747,1075.19 814.601,1076.04 815.455,1076.89 816.308,1077.73 817.162,1078.58 818.016,1079.43 818.87,1080.28 819.724,1081.12 820.578,1081.97 821.432,1082.81 822.286,1083.66 823.14,1084.5 823.994,1085.35 824.847,1086.19 825.701,1087.04 826.555,1087.88 827.409,1088.73 828.263,1089.57 829.117,1090.41 829.971,1091.26 830.825,1092.1 831.679,1092.94 832.532,1093.78 833.386,1094.63 834.24,1095.47 835.094,1096.31 835.948,1097.15 836.802,1097.99 837.656,1098.83 838.51,1099.68 839.364,1100.52 840.217,1101.36 841.071,1102.2 841.925,1103.04 842.779,1103.88 843.633,1104.72 844.487,1105.56 845.341,1106.39 846.195,1107.23 847.049,1108.07 847.902,1108.91 848.756,1109.75 849.61,1110.59 850.464,1111.43 851.318,1112.26 852.172,1113.1 853.026,1113.94 853.88,1114.78 854.734,1115.61 855.588,1116.45 856.441,1117.29 857.295,1118.12 858.149,1118.96 859.003,1119.8 859.857,1120.63 860.711,1121.47 861.565,1122.3 862.419,1123.14 863.273,1123.98 864.126,1124.81 864.98,1125.65 865.834,1126.48 866.688,1127.32 867.542,1128.15 868.396,1128.99 869.25,1129.82 870.104,1130.65 870.958,1131.49 871.811,1132.32 872.665,1133.16 873.519,1133.99 874.373,1134.82 875.227,1135.66 876.081,1136.49 876.935,1137.32 877.789,1138.16 878.643,1138.99 879.497,1139.82 880.35,1140.66 881.204,1141.49 882.058,1142.32 882.912,1143.15 883.766,1143.99 884.62,1144.82 885.474,1145.65 886.328,1146.48 887.182,1147.31 888.035,1148.15 888.889,1148.98 889.743,1149.81 890.597,1150.64 891.451,1151.47 892.305,1152.3 893.159,1153.13 894.013,1153.96 894.867,1154.79 895.72,1155.63 896.574,1156.46 897.428,1157.29 898.282,1158.12 899.136,1158.95 899.99,1159.78 900.844,1160.61 901.698,1161.44 902.552,1162.27 903.406,1163.1 904.259,1163.93 905.113,1164.76 905.967,1165.59 906.821,1166.42 907.675,1167.25 908.529,1168.08 909.383,1168.9 910.237,1169.73 911.091,1170.56 911.944,1171.39 912.798,1172.22 913.652,1173.05 914.506,1173.88 915.36,1174.71 916.214,1175.54 917.068,1176.36 917.922,1177.19 918.776,1178.02 919.629,1178.85 920.483,1179.68 921.337,1180.51 922.191,1181.33 923.045,1182.16 923.899,1182.99 924.753,1183.82 925.607,1184.65 926.461,1185.47 927.314,1186.3 928.168,1187.13 929.022,1187.96 929.876,1188.78 930.73,1189.61 931.584,1190.44 932.438,1191.27 933.292,1192.09 934.146,1192.92 935,1193.75 935.853,1194.57 936.707,1195.4 937.561,1196.23 938.415,1197.05 939.269,1197.88 940.123,1198.71 940.977,1199.53 941.831,1200.36 942.685,1201.19 943.538,1202.01 944.392,1202.84 945.246,1203.67 946.1,1204.49 946.954,1205.32 947.808,1206.15 948.662,1206.97 949.516,1207.8 950.37,1208.62 951.223,1209.45 952.077,1210.28 952.931,1211.1 953.785,1211.93 954.639,1212.75 955.493,1213.58 956.347,1214.4 957.201,1215.23 958.055,1216.06 958.909,1216.88 959.762,1217.71 960.616,1218.53 961.47,1219.36 962.324,1220.18 963.178,1221.01 964.032,1221.83 964.886,1222.66 965.74,1223.48 966.594,1224.31 967.447,1225.13 968.301,1225.96 969.155,1226.78 970.009,1227.61 970.863,1228.43 971.717,1229.26 972.571,1230.08 973.425,1230.91 974.279,1231.73 975.132,1232.56 975.986,1233.38 976.84,1234.21 977.694,1235.03 978.548,1235.86 979.402,1236.68 980.256,1237.51 981.11,1238.33 981.964,1239.15 982.817,1239.98 983.671,1240.8 984.525,1241.63 985.379,1242.45 986.233,1243.28 987.087,1244.1 987.941,1244.92 988.795,1245.75 989.649,1246.57 990.503,1247.4 991.356,1248.22 992.21,1249.04 993.064,1249.87 993.918,1250.69 994.772,1251.52 995.626,1252.34 996.48,1253.16 997.334,1253.99 998.188,1254.81 999.041,1255.63 999.895,1256.46 1000.75,1257.28 1001.6,1258.11 1002.46,1258.93 1003.31,1259.75 1004.16,1260.58 1005.02,1261.4 1005.87,1262.22 1006.73,1263.05 1007.58,1263.87 1008.43,1264.69 1009.29,1265.52 1010.14,1266.34 1011,1267.16 1011.85,1267.99 1012.7,1268.81 1013.56,1269.63 1014.41,1270.46 1015.27,1271.28 1016.12,1272.1 1016.97,1272.93 1017.83,1273.75 1018.68,1274.57 1019.53,1275.4 1020.39,1276.22 1021.24,1277.04 1022.1,1277.86 1022.95,1278.69 1023.8,1279.51 1024.66,1280.33 1025.51,1281.16 1026.37,1281.98 1027.22,1282.8 1028.07,1283.62 1028.93,1284.45 1029.78,1285.27 1030.64,1286.09 1031.49,1286.92 1032.34,1287.74 1033.2,1288.56 1034.05,1289.38 1034.9,1290.21 1035.76,1291.03 1036.61,1291.85 1037.47,1292.67 1038.32,1293.5 1039.17,1294.32 1040.03,1295.14 1040.88,1295.96 1041.74,1296.79 1042.59,1297.61 1043.44,1298.43 1044.3,1299.25 1045.15,1300.08 1046.01,1300.9 1046.86,1301.72 1047.71,1302.54 1048.57,1303.37 1049.42,1304.19 1050.27,1305.01 1051.13,1305.83 1051.98,1306.66 1052.84,1307.48 1053.69,1308.3 1054.54,1309.12 1055.4,1309.94 1056.25,1310.77 1057.11,1311.59 1057.96,1312.41 1058.81,1313.23 1059.67,1314.05 1060.52,1314.88 1061.38,1315.7 1062.23,1316.52 1063.08,1317.34 1063.94,1318.17 1064.79,1318.99 1065.65,1319.81 1066.5,1320.63 1067.35,1321.45 1068.21,1322.28 1069.06,1323.1 1069.91,1323.92 1070.77,1324.74 1071.62,1325.56 1072.48,1326.38 1073.33,1327.21 1074.18,1328.03 1075.04,1328.85 1075.89,1329.67 1076.75,1330.49 1077.6,1331.32 1078.45,1332.14 1079.31,1332.96 1080.16,1333.78 1081.02,1334.6 1081.87,1335.43 1082.72,1336.25 1083.58,1337.07 1084.43,1337.89 1085.28,1338.71 1086.14,1339.53 1086.99,1340.36 1087.85,1341.18 1088.7,1342 1089.55,1342.82 1090.41,1343.64 1091.26,1344.46 1092.12,1345.29 1092.97,1346.11 1093.82,1346.93 1094.68,1347.75 1095.53,1348.57 1096.39,1349.39 1097.24,1350.21 1098.09,1351.04 1098.95,1351.86 1099.8,1352.68 1100.65,1353.5 1101.51,1354.32 1102.36,1355.14 1103.22,1355.97 1104.07,1356.79 1104.92,1357.61 1105.78,1358.43 1106.63,1359.25 1107.49,1360.07 1108.34,1360.89 1109.19,1361.72 1110.05,1362.54 1110.9,1363.36 1111.76,1364.18 1112.61,1365 1113.46,1365.82 1114.32,1366.64 1115.17,1367.47 1116.02,1368.29 1116.88,1369.11 1117.73,1369.93 1118.59,1370.75 1119.44,1371.57 1120.29,1372.39 1121.15,1373.22 1122,1374.04 1122.86,1374.86 1123.71,1375.68 1124.56,1376.5 1125.42,1377.32 1126.27,1378.14 1127.13,1378.96 1127.98,1379.79 1128.83,1380.61 1129.69,1381.43 1130.54,1382.25 1131.39,1383.07 1132.25,1383.89 1133.1,1384.71 1133.96,1385.53 1134.81,1386.36 1135.66,1387.18 1136.52,1388 1137.37,1388.82 1138.23,1389.64 1139.08,1390.46 1139.93,1391.28 1140.79,1392.1 1141.64,1392.92 1142.5,1393.75 1143.35,1394.57 1144.2,1395.39 1145.06,1396.21 1145.91,1397.03 1146.76,1397.85 1147.62,1398.67 1148.47,1399.49 1149.33,1400.31 1150.18,1401.14 1151.03,1401.96 1151.89,1402.78 1152.74,1403.6 1153.6,1404.42 1154.45,1405.24 1155.3,1406.06 1156.16,1406.88 "/>
-<path clip-path="url(#clip150)" d="M1448.54 1423.18 L2352.76 1423.18 L2352.76 847.244 L1448.54 847.244  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="302.265,1423.18 302.265,847.244 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="515.738,1423.18 515.738,847.244 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="729.211,1423.18 729.211,847.244 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="942.685,1423.18 942.685,847.244 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1156.16,1423.18 1156.16,847.244 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1423.18 1181.75,1423.18 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="302.265,1423.18 302.265,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="515.738,1423.18 515.738,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="729.211,1423.18 729.211,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="942.685,1423.18 942.685,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1156.16,1423.18 1156.16,1404.28 "/>
+<path clip-path="url(#clip970)" d="M264.569 1454.1 Q260.958 1454.1 259.129 1457.66 Q257.323 1461.2 257.323 1468.33 Q257.323 1475.44 259.129 1479.01 Q260.958 1482.55 264.569 1482.55 Q268.203 1482.55 270.008 1479.01 Q271.837 1475.44 271.837 1468.33 Q271.837 1461.2 270.008 1457.66 Q268.203 1454.1 264.569 1454.1 M264.569 1450.39 Q270.379 1450.39 273.434 1455 Q276.513 1459.58 276.513 1468.33 Q276.513 1477.06 273.434 1481.67 Q270.379 1486.25 264.569 1486.25 Q258.759 1486.25 255.68 1481.67 Q252.624 1477.06 252.624 1468.33 Q252.624 1459.58 255.68 1455 Q258.759 1450.39 264.569 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M284.731 1479.7 L289.615 1479.7 L289.615 1485.58 L284.731 1485.58 L284.731 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M309.8 1454.1 Q306.189 1454.1 304.36 1457.66 Q302.555 1461.2 302.555 1468.33 Q302.555 1475.44 304.36 1479.01 Q306.189 1482.55 309.8 1482.55 Q313.434 1482.55 315.24 1479.01 Q317.068 1475.44 317.068 1468.33 Q317.068 1461.2 315.24 1457.66 Q313.434 1454.1 309.8 1454.1 M309.8 1450.39 Q315.61 1450.39 318.666 1455 Q321.744 1459.58 321.744 1468.33 Q321.744 1477.06 318.666 1481.67 Q315.61 1486.25 309.8 1486.25 Q303.99 1486.25 300.911 1481.67 Q297.856 1477.06 297.856 1468.33 Q297.856 1459.58 300.911 1455 Q303.99 1450.39 309.8 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M339.962 1454.1 Q336.351 1454.1 334.522 1457.66 Q332.716 1461.2 332.716 1468.33 Q332.716 1475.44 334.522 1479.01 Q336.351 1482.55 339.962 1482.55 Q343.596 1482.55 345.402 1479.01 Q347.23 1475.44 347.23 1468.33 Q347.23 1461.2 345.402 1457.66 Q343.596 1454.1 339.962 1454.1 M339.962 1450.39 Q345.772 1450.39 348.828 1455 Q351.906 1459.58 351.906 1468.33 Q351.906 1477.06 348.828 1481.67 Q345.772 1486.25 339.962 1486.25 Q334.152 1486.25 331.073 1481.67 Q328.017 1477.06 328.017 1468.33 Q328.017 1459.58 331.073 1455 Q334.152 1450.39 339.962 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M478.539 1454.1 Q474.928 1454.1 473.1 1457.66 Q471.294 1461.2 471.294 1468.33 Q471.294 1475.44 473.1 1479.01 Q474.928 1482.55 478.539 1482.55 Q482.174 1482.55 483.979 1479.01 Q485.808 1475.44 485.808 1468.33 Q485.808 1461.2 483.979 1457.66 Q482.174 1454.1 478.539 1454.1 M478.539 1450.39 Q484.35 1450.39 487.405 1455 Q490.484 1459.58 490.484 1468.33 Q490.484 1477.06 487.405 1481.67 Q484.35 1486.25 478.539 1486.25 Q472.729 1486.25 469.651 1481.67 Q466.595 1477.06 466.595 1468.33 Q466.595 1459.58 469.651 1455 Q472.729 1450.39 478.539 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M498.701 1479.7 L503.586 1479.7 L503.586 1485.58 L498.701 1485.58 L498.701 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M523.771 1454.1 Q520.16 1454.1 518.331 1457.66 Q516.525 1461.2 516.525 1468.33 Q516.525 1475.44 518.331 1479.01 Q520.16 1482.55 523.771 1482.55 Q527.405 1482.55 529.211 1479.01 Q531.039 1475.44 531.039 1468.33 Q531.039 1461.2 529.211 1457.66 Q527.405 1454.1 523.771 1454.1 M523.771 1450.39 Q529.581 1450.39 532.636 1455 Q535.715 1459.58 535.715 1468.33 Q535.715 1477.06 532.636 1481.67 Q529.581 1486.25 523.771 1486.25 Q517.961 1486.25 514.882 1481.67 Q511.826 1477.06 511.826 1468.33 Q511.826 1459.58 514.882 1455 Q517.961 1450.39 523.771 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M543.979 1451.02 L562.335 1451.02 L562.335 1454.96 L548.261 1454.96 L548.261 1463.43 Q549.28 1463.08 550.298 1462.92 Q551.317 1462.73 552.335 1462.73 Q558.122 1462.73 561.502 1465.9 Q564.882 1469.08 564.882 1474.49 Q564.882 1480.07 561.409 1483.17 Q557.937 1486.25 551.618 1486.25 Q549.442 1486.25 547.173 1485.88 Q544.928 1485.51 542.521 1484.77 L542.521 1480.07 Q544.604 1481.2 546.826 1481.76 Q549.048 1482.32 551.525 1482.32 Q555.53 1482.32 557.868 1480.21 Q560.206 1478.1 560.206 1474.49 Q560.206 1470.88 557.868 1468.77 Q555.53 1466.67 551.525 1466.67 Q549.65 1466.67 547.775 1467.08 Q545.923 1467.5 543.979 1468.38 L543.979 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M691.515 1454.1 Q687.904 1454.1 686.075 1457.66 Q684.27 1461.2 684.27 1468.33 Q684.27 1475.44 686.075 1479.01 Q687.904 1482.55 691.515 1482.55 Q695.149 1482.55 696.955 1479.01 Q698.783 1475.44 698.783 1468.33 Q698.783 1461.2 696.955 1457.66 Q695.149 1454.1 691.515 1454.1 M691.515 1450.39 Q697.325 1450.39 700.381 1455 Q703.459 1459.58 703.459 1468.33 Q703.459 1477.06 700.381 1481.67 Q697.325 1486.25 691.515 1486.25 Q685.705 1486.25 682.626 1481.67 Q679.571 1477.06 679.571 1468.33 Q679.571 1459.58 682.626 1455 Q685.705 1450.39 691.515 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M711.677 1479.7 L716.561 1479.7 L716.561 1485.58 L711.677 1485.58 L711.677 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M727.556 1481.64 L735.195 1481.64 L735.195 1455.28 L726.885 1456.95 L726.885 1452.69 L735.149 1451.02 L739.825 1451.02 L739.825 1481.64 L747.464 1481.64 L747.464 1485.58 L727.556 1485.58 L727.556 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M766.908 1454.1 Q763.297 1454.1 761.468 1457.66 Q759.663 1461.2 759.663 1468.33 Q759.663 1475.44 761.468 1479.01 Q763.297 1482.55 766.908 1482.55 Q770.542 1482.55 772.348 1479.01 Q774.176 1475.44 774.176 1468.33 Q774.176 1461.2 772.348 1457.66 Q770.542 1454.1 766.908 1454.1 M766.908 1450.39 Q772.718 1450.39 775.774 1455 Q778.852 1459.58 778.852 1468.33 Q778.852 1477.06 775.774 1481.67 Q772.718 1486.25 766.908 1486.25 Q761.098 1486.25 758.019 1481.67 Q754.964 1477.06 754.964 1468.33 Q754.964 1459.58 758.019 1455 Q761.098 1450.39 766.908 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M905.486 1454.1 Q901.875 1454.1 900.046 1457.66 Q898.24 1461.2 898.24 1468.33 Q898.24 1475.44 900.046 1479.01 Q901.875 1482.55 905.486 1482.55 Q909.12 1482.55 910.925 1479.01 Q912.754 1475.44 912.754 1468.33 Q912.754 1461.2 910.925 1457.66 Q909.12 1454.1 905.486 1454.1 M905.486 1450.39 Q911.296 1450.39 914.351 1455 Q917.43 1459.58 917.43 1468.33 Q917.43 1477.06 914.351 1481.67 Q911.296 1486.25 905.486 1486.25 Q899.676 1486.25 896.597 1481.67 Q893.541 1477.06 893.541 1468.33 Q893.541 1459.58 896.597 1455 Q899.676 1450.39 905.486 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M925.648 1479.7 L930.532 1479.7 L930.532 1485.58 L925.648 1485.58 L925.648 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M941.527 1481.64 L949.166 1481.64 L949.166 1455.28 L940.856 1456.95 L940.856 1452.69 L949.12 1451.02 L953.796 1451.02 L953.796 1481.64 L961.434 1481.64 L961.434 1485.58 L941.527 1485.58 L941.527 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M970.925 1451.02 L989.282 1451.02 L989.282 1454.96 L975.208 1454.96 L975.208 1463.43 Q976.226 1463.08 977.245 1462.92 Q978.263 1462.73 979.282 1462.73 Q985.069 1462.73 988.448 1465.9 Q991.828 1469.08 991.828 1474.49 Q991.828 1480.07 988.356 1483.17 Q984.883 1486.25 978.564 1486.25 Q976.388 1486.25 974.12 1485.88 Q971.874 1485.51 969.467 1484.77 L969.467 1480.07 Q971.55 1481.2 973.772 1481.76 Q975.995 1482.32 978.471 1482.32 Q982.476 1482.32 984.814 1480.21 Q987.152 1478.1 987.152 1474.49 Q987.152 1470.88 984.814 1468.77 Q982.476 1466.67 978.471 1466.67 Q976.596 1466.67 974.721 1467.08 Q972.87 1467.5 970.925 1468.38 L970.925 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1118.46 1454.1 Q1114.85 1454.1 1113.02 1457.66 Q1111.22 1461.2 1111.22 1468.33 Q1111.22 1475.44 1113.02 1479.01 Q1114.85 1482.55 1118.46 1482.55 Q1122.1 1482.55 1123.9 1479.01 Q1125.73 1475.44 1125.73 1468.33 Q1125.73 1461.2 1123.9 1457.66 Q1122.1 1454.1 1118.46 1454.1 M1118.46 1450.39 Q1124.27 1450.39 1127.33 1455 Q1130.41 1459.58 1130.41 1468.33 Q1130.41 1477.06 1127.33 1481.67 Q1124.27 1486.25 1118.46 1486.25 Q1112.65 1486.25 1109.57 1481.67 Q1106.52 1477.06 1106.52 1468.33 Q1106.52 1459.58 1109.57 1455 Q1112.65 1450.39 1118.46 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1138.62 1479.7 L1143.51 1479.7 L1143.51 1485.58 L1138.62 1485.58 L1138.62 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1157.72 1481.64 L1174.04 1481.64 L1174.04 1485.58 L1152.1 1485.58 L1152.1 1481.64 Q1154.76 1478.89 1159.34 1474.26 Q1163.95 1469.61 1165.13 1468.27 Q1167.37 1465.74 1168.25 1464.01 Q1169.16 1462.25 1169.16 1460.56 Q1169.16 1457.8 1167.21 1456.07 Q1165.29 1454.33 1162.19 1454.33 Q1159.99 1454.33 1157.53 1455.09 Q1155.1 1455.86 1152.33 1457.41 L1152.33 1452.69 Q1155.15 1451.55 1157.6 1450.97 Q1160.06 1450.39 1162.1 1450.39 Q1167.47 1450.39 1170.66 1453.08 Q1173.85 1455.77 1173.85 1460.26 Q1173.85 1462.39 1173.04 1464.31 Q1172.26 1466.2 1170.15 1468.8 Q1169.57 1469.47 1166.47 1472.69 Q1163.37 1475.88 1157.72 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1193.85 1454.1 Q1190.24 1454.1 1188.41 1457.66 Q1186.61 1461.2 1186.61 1468.33 Q1186.61 1475.44 1188.41 1479.01 Q1190.24 1482.55 1193.85 1482.55 Q1197.49 1482.55 1199.29 1479.01 Q1201.12 1475.44 1201.12 1468.33 Q1201.12 1461.2 1199.29 1457.66 Q1197.49 1454.1 1193.85 1454.1 M1193.85 1450.39 Q1199.66 1450.39 1202.72 1455 Q1205.8 1459.58 1205.8 1468.33 Q1205.8 1477.06 1202.72 1481.67 Q1199.66 1486.25 1193.85 1486.25 Q1188.04 1486.25 1184.97 1481.67 Q1181.91 1477.06 1181.91 1468.33 Q1181.91 1459.58 1184.97 1455 Q1188.04 1450.39 1193.85 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M599.81 1520.52 L640.009 1520.52 L640.009 1525.93 L623.14 1525.93 L623.14 1568.04 L616.679 1568.04 L616.679 1525.93 L599.81 1525.93 L599.81 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M643.956 1532.4 L649.812 1532.4 L649.812 1568.04 L643.956 1568.04 L643.956 1532.4 M643.956 1518.52 L649.812 1518.52 L649.812 1525.93 L643.956 1525.93 L643.956 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M689.821 1539.24 Q692.017 1535.29 695.073 1533.41 Q698.128 1531.54 702.266 1531.54 Q707.836 1531.54 710.86 1535.45 Q713.883 1539.33 713.883 1546.53 L713.883 1568.04 L707.995 1568.04 L707.995 1546.72 Q707.995 1541.59 706.181 1539.11 Q704.367 1536.63 700.643 1536.63 Q696.091 1536.63 693.449 1539.65 Q690.808 1542.68 690.808 1547.9 L690.808 1568.04 L684.919 1568.04 L684.919 1546.72 Q684.919 1541.56 683.105 1539.11 Q681.291 1536.63 677.503 1536.63 Q673.015 1536.63 670.374 1539.68 Q667.732 1542.71 667.732 1547.9 L667.732 1568.04 L661.844 1568.04 L661.844 1532.4 L667.732 1532.4 L667.732 1537.93 Q669.737 1534.66 672.538 1533.1 Q675.339 1531.54 679.19 1531.54 Q683.073 1531.54 685.779 1533.51 Q688.516 1535.48 689.821 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M756.056 1548.76 L756.056 1551.62 L729.129 1551.62 Q729.511 1557.67 732.758 1560.85 Q736.036 1564 741.861 1564 Q745.234 1564 748.385 1563.17 Q751.568 1562.35 754.687 1560.69 L754.687 1566.23 Q751.536 1567.57 748.226 1568.27 Q744.916 1568.97 741.51 1568.97 Q732.98 1568.97 727.983 1564 Q723.018 1559.04 723.018 1550.57 Q723.018 1541.82 727.729 1536.69 Q732.471 1531.54 740.492 1531.54 Q747.685 1531.54 751.855 1536.18 Q756.056 1540.8 756.056 1548.76 M750.2 1547.04 Q750.136 1542.23 747.494 1539.37 Q744.884 1536.5 740.556 1536.5 Q735.654 1536.5 732.694 1539.27 Q729.766 1542.04 729.32 1547.07 L750.2 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M800.457 1518.58 Q796.192 1525.9 794.123 1533.06 Q792.054 1540.23 792.054 1547.58 Q792.054 1554.93 794.123 1562.16 Q796.224 1569.35 800.457 1576.64 L795.364 1576.64 Q790.59 1569.16 788.203 1561.93 Q785.848 1554.71 785.848 1547.58 Q785.848 1540.48 788.203 1533.29 Q790.558 1526.09 795.364 1518.58 L800.457 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M834.545 1533.45 L834.545 1538.98 Q832.063 1537.71 829.389 1537.07 Q826.715 1536.44 823.851 1536.44 Q819.49 1536.44 817.294 1537.77 Q815.13 1539.11 815.13 1541.79 Q815.13 1543.82 816.689 1545 Q818.249 1546.15 822.96 1547.2 L824.965 1547.64 Q831.203 1548.98 833.813 1551.43 Q836.455 1553.85 836.455 1558.21 Q836.455 1563.17 832.508 1566.07 Q828.593 1568.97 821.718 1568.97 Q818.854 1568.97 815.735 1568.39 Q812.647 1567.85 809.21 1566.74 L809.21 1560.69 Q812.456 1562.38 815.607 1563.24 Q818.758 1564.07 821.846 1564.07 Q825.983 1564.07 828.211 1562.66 Q830.439 1561.23 830.439 1558.65 Q830.439 1556.27 828.816 1554.99 Q827.225 1553.72 821.782 1552.54 L819.745 1552.07 Q814.302 1550.92 811.883 1548.56 Q809.464 1546.18 809.464 1542.04 Q809.464 1537.01 813.029 1534.27 Q816.594 1531.54 823.151 1531.54 Q826.397 1531.54 829.262 1532.01 Q832.126 1532.49 834.545 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M844.858 1518.58 L849.95 1518.58 Q854.725 1526.09 857.08 1533.29 Q859.467 1540.48 859.467 1547.58 Q859.467 1554.71 857.08 1561.93 Q854.725 1569.16 849.95 1576.64 L844.858 1576.64 Q849.091 1569.35 851.16 1562.16 Q853.26 1554.93 853.26 1547.58 Q853.26 1540.23 851.16 1533.06 Q849.091 1525.9 844.858 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1406.88 1181.75,1406.88 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1322.61 1181.75,1322.61 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1238.34 1181.75,1238.34 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1154.07 1181.75,1154.07 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,1069.8 1181.75,1069.8 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,985.533 1181.75,985.533 "/>
+<polyline clip-path="url(#clip974)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="277.528,901.264 1181.75,901.264 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1423.18 277.528,847.244 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1406.88 296.426,1406.88 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1322.61 296.426,1322.61 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1238.34 296.426,1238.34 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1154.07 296.426,1154.07 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,1069.8 296.426,1069.8 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,985.533 296.426,985.533 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="277.528,901.264 296.426,901.264 "/>
+<path clip-path="url(#clip970)" d="M154.19 1392.68 Q150.579 1392.68 148.751 1396.24 Q146.945 1399.78 146.945 1406.91 Q146.945 1414.02 148.751 1417.59 Q150.579 1421.13 154.19 1421.13 Q157.825 1421.13 159.63 1417.59 Q161.459 1414.02 161.459 1406.91 Q161.459 1399.78 159.63 1396.24 Q157.825 1392.68 154.19 1392.68 M154.19 1388.97 Q160.001 1388.97 163.056 1393.58 Q166.135 1398.16 166.135 1406.91 Q166.135 1415.64 163.056 1420.25 Q160.001 1424.83 154.19 1424.83 Q148.38 1424.83 145.302 1420.25 Q142.246 1415.64 142.246 1406.91 Q142.246 1398.16 145.302 1393.58 Q148.38 1388.97 154.19 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M174.352 1418.28 L179.237 1418.28 L179.237 1424.16 L174.352 1424.16 L174.352 1418.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M199.422 1392.68 Q195.811 1392.68 193.982 1396.24 Q192.176 1399.78 192.176 1406.91 Q192.176 1414.02 193.982 1417.59 Q195.811 1421.13 199.422 1421.13 Q203.056 1421.13 204.862 1417.59 Q206.69 1414.02 206.69 1406.91 Q206.69 1399.78 204.862 1396.24 Q203.056 1392.68 199.422 1392.68 M199.422 1388.97 Q205.232 1388.97 208.287 1393.58 Q211.366 1398.16 211.366 1406.91 Q211.366 1415.64 208.287 1420.25 Q205.232 1424.83 199.422 1424.83 Q193.612 1424.83 190.533 1420.25 Q187.477 1415.64 187.477 1406.91 Q187.477 1398.16 190.533 1393.58 Q193.612 1388.97 199.422 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M229.584 1392.68 Q225.973 1392.68 224.144 1396.24 Q222.338 1399.78 222.338 1406.91 Q222.338 1414.02 224.144 1417.59 Q225.973 1421.13 229.584 1421.13 Q233.218 1421.13 235.023 1417.59 Q236.852 1414.02 236.852 1406.91 Q236.852 1399.78 235.023 1396.24 Q233.218 1392.68 229.584 1392.68 M229.584 1388.97 Q235.394 1388.97 238.449 1393.58 Q241.528 1398.16 241.528 1406.91 Q241.528 1415.64 238.449 1420.25 Q235.394 1424.83 229.584 1424.83 Q223.773 1424.83 220.695 1420.25 Q217.639 1415.64 217.639 1406.91 Q217.639 1398.16 220.695 1393.58 Q223.773 1388.97 229.584 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M155.788 1308.41 Q152.177 1308.41 150.348 1311.97 Q148.542 1315.52 148.542 1322.65 Q148.542 1329.75 150.348 1333.32 Q152.177 1336.86 155.788 1336.86 Q159.422 1336.86 161.227 1333.32 Q163.056 1329.75 163.056 1322.65 Q163.056 1315.52 161.227 1311.97 Q159.422 1308.41 155.788 1308.41 M155.788 1304.71 Q161.598 1304.71 164.653 1309.31 Q167.732 1313.9 167.732 1322.65 Q167.732 1331.37 164.653 1335.98 Q161.598 1340.56 155.788 1340.56 Q149.978 1340.56 146.899 1335.98 Q143.843 1331.37 143.843 1322.65 Q143.843 1313.9 146.899 1309.31 Q149.978 1304.71 155.788 1304.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M175.95 1334.01 L180.834 1334.01 L180.834 1339.89 L175.95 1339.89 L175.95 1334.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M201.019 1308.41 Q197.408 1308.41 195.579 1311.97 Q193.774 1315.52 193.774 1322.65 Q193.774 1329.75 195.579 1333.32 Q197.408 1336.86 201.019 1336.86 Q204.653 1336.86 206.459 1333.32 Q208.287 1329.75 208.287 1322.65 Q208.287 1315.52 206.459 1311.97 Q204.653 1308.41 201.019 1308.41 M201.019 1304.71 Q206.829 1304.71 209.885 1309.31 Q212.963 1313.9 212.963 1322.65 Q212.963 1331.37 209.885 1335.98 Q206.829 1340.56 201.019 1340.56 Q195.209 1340.56 192.13 1335.98 Q189.075 1331.37 189.075 1322.65 Q189.075 1313.9 192.13 1309.31 Q195.209 1304.71 201.019 1304.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M225.209 1335.96 L241.528 1335.96 L241.528 1339.89 L219.584 1339.89 L219.584 1335.96 Q222.246 1333.2 226.829 1328.57 Q231.435 1323.92 232.616 1322.58 Q234.861 1320.05 235.741 1318.32 Q236.644 1316.56 236.644 1314.87 Q236.644 1312.11 234.699 1310.38 Q232.778 1308.64 229.676 1308.64 Q227.477 1308.64 225.023 1309.4 Q222.593 1310.17 219.815 1311.72 L219.815 1307 Q222.639 1305.86 225.093 1305.28 Q227.547 1304.71 229.584 1304.71 Q234.954 1304.71 238.148 1307.39 Q241.343 1310.08 241.343 1314.57 Q241.343 1316.7 240.533 1318.62 Q239.746 1320.52 237.639 1323.11 Q237.06 1323.78 233.959 1327 Q230.857 1330.19 225.209 1335.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M153.704 1224.14 Q150.093 1224.14 148.265 1227.7 Q146.459 1231.25 146.459 1238.38 Q146.459 1245.48 148.265 1249.05 Q150.093 1252.59 153.704 1252.59 Q157.339 1252.59 159.144 1249.05 Q160.973 1245.48 160.973 1238.38 Q160.973 1231.25 159.144 1227.7 Q157.339 1224.14 153.704 1224.14 M153.704 1220.44 Q159.515 1220.44 162.57 1225.04 Q165.649 1229.63 165.649 1238.38 Q165.649 1247.1 162.57 1251.71 Q159.515 1256.29 153.704 1256.29 Q147.894 1256.29 144.816 1251.71 Q141.76 1247.1 141.76 1238.38 Q141.76 1229.63 144.816 1225.04 Q147.894 1220.44 153.704 1220.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M173.866 1249.74 L178.751 1249.74 L178.751 1255.62 L173.866 1255.62 L173.866 1249.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M198.936 1224.14 Q195.325 1224.14 193.496 1227.7 Q191.69 1231.25 191.69 1238.38 Q191.69 1245.48 193.496 1249.05 Q195.325 1252.59 198.936 1252.59 Q202.57 1252.59 204.375 1249.05 Q206.204 1245.48 206.204 1238.38 Q206.204 1231.25 204.375 1227.7 Q202.57 1224.14 198.936 1224.14 M198.936 1220.44 Q204.746 1220.44 207.801 1225.04 Q210.88 1229.63 210.88 1238.38 Q210.88 1247.1 207.801 1251.71 Q204.746 1256.29 198.936 1256.29 Q193.125 1256.29 190.047 1251.71 Q186.991 1247.1 186.991 1238.38 Q186.991 1229.63 190.047 1225.04 Q193.125 1220.44 198.936 1220.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M231.945 1225.14 L220.139 1243.58 L231.945 1243.58 L231.945 1225.14 M230.718 1221.06 L236.597 1221.06 L236.597 1243.58 L241.528 1243.58 L241.528 1247.47 L236.597 1247.47 L236.597 1255.62 L231.945 1255.62 L231.945 1247.47 L216.343 1247.47 L216.343 1242.96 L230.718 1221.06 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M154.028 1139.87 Q150.417 1139.87 148.589 1143.44 Q146.783 1146.98 146.783 1154.11 Q146.783 1161.21 148.589 1164.78 Q150.417 1168.32 154.028 1168.32 Q157.663 1168.32 159.468 1164.78 Q161.297 1161.21 161.297 1154.11 Q161.297 1146.98 159.468 1143.44 Q157.663 1139.87 154.028 1139.87 M154.028 1136.17 Q159.839 1136.17 162.894 1140.77 Q165.973 1145.36 165.973 1154.11 Q165.973 1162.83 162.894 1167.44 Q159.839 1172.02 154.028 1172.02 Q148.218 1172.02 145.14 1167.44 Q142.084 1162.83 142.084 1154.11 Q142.084 1145.36 145.14 1140.77 Q148.218 1136.17 154.028 1136.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M174.19 1165.47 L179.075 1165.47 L179.075 1171.35 L174.19 1171.35 L174.19 1165.47 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M199.26 1139.87 Q195.649 1139.87 193.82 1143.44 Q192.014 1146.98 192.014 1154.11 Q192.014 1161.21 193.82 1164.78 Q195.649 1168.32 199.26 1168.32 Q202.894 1168.32 204.699 1164.78 Q206.528 1161.21 206.528 1154.11 Q206.528 1146.98 204.699 1143.44 Q202.894 1139.87 199.26 1139.87 M199.26 1136.17 Q205.07 1136.17 208.125 1140.77 Q211.204 1145.36 211.204 1154.11 Q211.204 1162.83 208.125 1167.44 Q205.07 1172.02 199.26 1172.02 Q193.45 1172.02 190.371 1167.44 Q187.315 1162.83 187.315 1154.11 Q187.315 1145.36 190.371 1140.77 Q193.45 1136.17 199.26 1136.17 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M230 1152.21 Q226.852 1152.21 225 1154.36 Q223.172 1156.51 223.172 1160.26 Q223.172 1163.99 225 1166.17 Q226.852 1168.32 230 1168.32 Q233.148 1168.32 234.977 1166.17 Q236.829 1163.99 236.829 1160.26 Q236.829 1156.51 234.977 1154.36 Q233.148 1152.21 230 1152.21 M239.283 1137.56 L239.283 1141.82 Q237.523 1140.98 235.718 1140.54 Q233.935 1140.1 232.176 1140.1 Q227.547 1140.1 225.093 1143.23 Q222.662 1146.35 222.315 1152.67 Q223.681 1150.66 225.741 1149.59 Q227.801 1148.5 230.278 1148.5 Q235.486 1148.5 238.496 1151.68 Q241.528 1154.82 241.528 1160.26 Q241.528 1165.59 238.38 1168.81 Q235.232 1172.02 230 1172.02 Q224.005 1172.02 220.834 1167.44 Q217.662 1162.83 217.662 1154.11 Q217.662 1145.91 221.551 1141.05 Q225.44 1136.17 231.991 1136.17 Q233.75 1136.17 235.533 1136.51 Q237.338 1136.86 239.283 1137.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M154.283 1055.6 Q150.672 1055.6 148.843 1059.17 Q147.038 1062.71 147.038 1069.84 Q147.038 1076.94 148.843 1080.51 Q150.672 1084.05 154.283 1084.05 Q157.917 1084.05 159.723 1080.51 Q161.552 1076.94 161.552 1069.84 Q161.552 1062.71 159.723 1059.17 Q157.917 1055.6 154.283 1055.6 M154.283 1051.9 Q160.093 1051.9 163.149 1056.5 Q166.227 1061.09 166.227 1069.84 Q166.227 1078.56 163.149 1083.17 Q160.093 1087.75 154.283 1087.75 Q148.473 1087.75 145.394 1083.17 Q142.339 1078.56 142.339 1069.84 Q142.339 1061.09 145.394 1056.5 Q148.473 1051.9 154.283 1051.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M174.445 1081.2 L179.329 1081.2 L179.329 1087.08 L174.445 1087.08 L174.445 1081.2 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M199.514 1055.6 Q195.903 1055.6 194.075 1059.17 Q192.269 1062.71 192.269 1069.84 Q192.269 1076.94 194.075 1080.51 Q195.903 1084.05 199.514 1084.05 Q203.149 1084.05 204.954 1080.51 Q206.783 1076.94 206.783 1069.84 Q206.783 1062.71 204.954 1059.17 Q203.149 1055.6 199.514 1055.6 M199.514 1051.9 Q205.324 1051.9 208.38 1056.5 Q211.459 1061.09 211.459 1069.84 Q211.459 1078.56 208.38 1083.17 Q205.324 1087.75 199.514 1087.75 Q193.704 1087.75 190.625 1083.17 Q187.57 1078.56 187.57 1069.84 Q187.57 1061.09 190.625 1056.5 Q193.704 1051.9 199.514 1051.9 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M229.676 1070.67 Q226.343 1070.67 224.422 1072.45 Q222.523 1074.24 222.523 1077.36 Q222.523 1080.49 224.422 1082.27 Q226.343 1084.05 229.676 1084.05 Q233.01 1084.05 234.931 1082.27 Q236.852 1080.46 236.852 1077.36 Q236.852 1074.24 234.931 1072.45 Q233.033 1070.67 229.676 1070.67 M225 1068.68 Q221.991 1067.94 220.301 1065.88 Q218.635 1063.82 218.635 1060.86 Q218.635 1056.71 221.574 1054.31 Q224.537 1051.9 229.676 1051.9 Q234.838 1051.9 237.778 1054.31 Q240.718 1056.71 240.718 1060.86 Q240.718 1063.82 239.028 1065.88 Q237.361 1067.94 234.375 1068.68 Q237.755 1069.47 239.63 1071.76 Q241.528 1074.05 241.528 1077.36 Q241.528 1082.38 238.449 1085.07 Q235.394 1087.75 229.676 1087.75 Q223.959 1087.75 220.88 1085.07 Q217.824 1082.38 217.824 1077.36 Q217.824 1074.05 219.723 1071.76 Q221.621 1069.47 225 1068.68 M223.287 1061.3 Q223.287 1063.98 224.954 1065.49 Q226.644 1066.99 229.676 1066.99 Q232.685 1066.99 234.375 1065.49 Q236.088 1063.98 236.088 1061.3 Q236.088 1058.61 234.375 1057.11 Q232.685 1055.6 229.676 1055.6 Q226.644 1055.6 224.954 1057.11 Q223.287 1058.61 223.287 1061.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M154.19 971.332 Q150.579 971.332 148.751 974.897 Q146.945 978.439 146.945 985.568 Q146.945 992.675 148.751 996.239 Q150.579 999.781 154.19 999.781 Q157.825 999.781 159.63 996.239 Q161.459 992.675 161.459 985.568 Q161.459 978.439 159.63 974.897 Q157.825 971.332 154.19 971.332 M154.19 967.628 Q160.001 967.628 163.056 972.235 Q166.135 976.818 166.135 985.568 Q166.135 994.295 163.056 998.901 Q160.001 1003.48 154.19 1003.48 Q148.38 1003.48 145.302 998.901 Q142.246 994.295 142.246 985.568 Q142.246 976.818 145.302 972.235 Q148.38 967.628 154.19 967.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M174.352 996.934 L179.237 996.934 L179.237 1002.81 L174.352 1002.81 L174.352 996.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M190.232 998.878 L197.871 998.878 L197.871 972.513 L189.561 974.179 L189.561 969.92 L197.825 968.253 L202.5 968.253 L202.5 998.878 L210.139 998.878 L210.139 1002.81 L190.232 1002.81 L190.232 998.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M229.584 971.332 Q225.973 971.332 224.144 974.897 Q222.338 978.439 222.338 985.568 Q222.338 992.675 224.144 996.239 Q225.973 999.781 229.584 999.781 Q233.218 999.781 235.023 996.239 Q236.852 992.675 236.852 985.568 Q236.852 978.439 235.023 974.897 Q233.218 971.332 229.584 971.332 M229.584 967.628 Q235.394 967.628 238.449 972.235 Q241.528 976.818 241.528 985.568 Q241.528 994.295 238.449 998.901 Q235.394 1003.48 229.584 1003.48 Q223.773 1003.48 220.695 998.901 Q217.639 994.295 217.639 985.568 Q217.639 976.818 220.695 972.235 Q223.773 967.628 229.584 967.628 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M155.788 887.063 Q152.177 887.063 150.348 890.628 Q148.542 894.169 148.542 901.299 Q148.542 908.405 150.348 911.97 Q152.177 915.512 155.788 915.512 Q159.422 915.512 161.227 911.97 Q163.056 908.405 163.056 901.299 Q163.056 894.169 161.227 890.628 Q159.422 887.063 155.788 887.063 M155.788 883.359 Q161.598 883.359 164.653 887.966 Q167.732 892.549 167.732 901.299 Q167.732 910.026 164.653 914.632 Q161.598 919.215 155.788 919.215 Q149.978 919.215 146.899 914.632 Q143.843 910.026 143.843 901.299 Q143.843 892.549 146.899 887.966 Q149.978 883.359 155.788 883.359 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M175.95 912.665 L180.834 912.665 L180.834 918.544 L175.95 918.544 L175.95 912.665 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M191.829 914.609 L199.468 914.609 L199.468 888.243 L191.158 889.91 L191.158 885.651 L199.422 883.984 L204.098 883.984 L204.098 914.609 L211.736 914.609 L211.736 918.544 L191.829 918.544 L191.829 914.609 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M225.209 914.609 L241.528 914.609 L241.528 918.544 L219.584 918.544 L219.584 914.609 Q222.246 911.854 226.829 907.225 Q231.435 902.572 232.616 901.229 Q234.861 898.706 235.741 896.97 Q236.644 895.211 236.644 893.521 Q236.644 890.767 234.699 889.03 Q232.778 887.294 229.676 887.294 Q227.477 887.294 225.023 888.058 Q222.593 888.822 219.815 890.373 L219.815 885.651 Q222.639 884.517 225.093 883.938 Q227.547 883.359 229.584 883.359 Q234.954 883.359 238.148 886.044 Q241.343 888.73 241.343 893.22 Q241.343 895.35 240.533 897.271 Q239.746 899.169 237.639 901.762 Q237.06 902.433 233.959 905.651 Q230.857 908.845 225.209 914.609 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M91.504 1243.7 L43.984 1261.84 L43.984 1255.13 L83.9925 1240.07 L43.984 1224.98 L43.984 1218.3 L91.504 1236.41 L91.504 1243.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M72.2159 1186.19 L75.0805 1186.19 L75.0805 1213.11 Q81.1279 1212.73 84.3108 1209.48 Q87.4618 1206.21 87.4618 1200.38 Q87.4618 1197.01 86.6342 1193.86 Q85.8067 1190.67 84.1516 1187.55 L89.6898 1187.55 Q91.0266 1190.71 91.7268 1194.02 Q92.427 1197.33 92.427 1200.73 Q92.427 1209.26 87.4618 1214.26 Q82.4965 1219.22 74.0301 1219.22 Q65.2773 1219.22 60.1529 1214.51 Q54.9967 1209.77 54.9967 1201.75 Q54.9967 1194.56 59.6436 1190.39 Q64.2588 1186.19 72.2159 1186.19 M70.4972 1192.04 Q65.6911 1192.11 62.8265 1194.75 Q59.9619 1197.36 59.9619 1201.69 Q59.9619 1206.59 62.731 1209.55 Q65.5001 1212.48 70.529 1212.92 L70.4972 1192.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M41.9788 1176.57 L41.9788 1170.72 L91.504 1170.72 L91.504 1176.57 L41.9788 1176.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M59.9619 1144.65 Q59.9619 1149.36 63.654 1152.1 Q67.3143 1154.83 73.7119 1154.83 Q80.1094 1154.83 83.8015 1152.13 Q87.4618 1149.39 87.4618 1144.65 Q87.4618 1139.97 83.7697 1137.23 Q80.0776 1134.5 73.7119 1134.5 Q67.378 1134.5 63.6859 1137.23 Q59.9619 1139.97 59.9619 1144.65 M54.9967 1144.65 Q54.9967 1137.01 59.9619 1132.65 Q64.9272 1128.29 73.7119 1128.29 Q82.4647 1128.29 87.4618 1132.65 Q92.427 1137.01 92.427 1144.65 Q92.427 1152.32 87.4618 1156.68 Q82.4647 1161.01 73.7119 1161.01 Q64.9272 1161.01 59.9619 1156.68 Q54.9967 1152.32 54.9967 1144.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M57.2247 1092.93 L62.6992 1092.93 Q61.3306 1095.41 60.6622 1097.92 Q59.9619 1100.41 59.9619 1102.95 Q59.9619 1108.65 63.5904 1111.8 Q67.187 1114.95 73.7119 1114.95 Q80.2367 1114.95 83.8652 1111.8 Q87.4618 1108.65 87.4618 1102.95 Q87.4618 1100.41 86.7934 1097.92 Q86.0932 1095.41 84.7245 1092.93 L90.1354 1092.93 Q91.2812 1095.38 91.8541 1098.02 Q92.427 1100.63 92.427 1103.59 Q92.427 1111.64 87.3663 1116.39 Q82.3056 1121.13 73.7119 1121.13 Q64.9908 1121.13 59.9938 1116.35 Q54.9967 1111.55 54.9967 1103.21 Q54.9967 1100.5 55.5696 1097.92 Q56.1107 1095.35 57.2247 1092.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M55.8561 1082.74 L55.8561 1076.89 L91.504 1076.89 L91.504 1082.74 L55.8561 1082.74 M41.9788 1082.74 L41.9788 1076.89 L49.3949 1076.89 L49.3949 1082.74 L41.9788 1082.74 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M45.7346 1058.84 L55.8561 1058.84 L55.8561 1046.78 L60.4075 1046.78 L60.4075 1058.84 L79.7593 1058.84 Q84.1198 1058.84 85.3611 1057.66 Q86.6024 1056.45 86.6024 1052.79 L86.6024 1046.78 L91.504 1046.78 L91.504 1052.79 Q91.504 1059.57 88.9896 1062.15 Q86.4433 1064.73 79.7593 1064.73 L60.4075 1064.73 L60.4075 1069.02 L55.8561 1069.02 L55.8561 1064.73 L45.7346 1064.73 L45.7346 1058.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M94.8142 1024.24 Q101.18 1026.72 103.121 1029.08 Q105.063 1031.44 105.063 1035.38 L105.063 1040.06 L100.161 1040.06 L100.161 1036.62 Q100.161 1034.2 99.0155 1032.87 Q97.8697 1031.53 93.6047 1029.91 L90.9311 1028.86 L55.8561 1043.28 L55.8561 1037.07 L83.7378 1025.93 L55.8561 1014.79 L55.8561 1008.58 L94.8142 1024.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip974)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="303.119,1406.88 303.973,1404.77 304.827,1402.66 305.681,1400.55 306.535,1398.43 307.389,1396.31 308.243,1394.18 309.096,1392.05 309.95,1389.92 310.804,1387.79 311.658,1385.65 312.512,1383.5 313.366,1381.36 314.22,1379.21 315.074,1377.06 315.928,1374.91 316.781,1372.76 317.635,1370.6 318.489,1368.44 319.343,1366.28 320.197,1364.12 321.051,1361.95 321.905,1359.79 322.759,1357.62 323.613,1355.45 324.466,1353.29 325.32,1351.11 326.174,1348.94 327.028,1346.77 327.882,1344.6 328.736,1342.43 329.59,1340.25 330.444,1338.08 331.298,1335.91 332.151,1333.73 333.005,1331.56 333.859,1329.39 334.713,1327.21 335.567,1325.04 336.421,1322.87 337.275,1320.7 338.129,1318.53 338.983,1316.36 339.837,1314.19 340.69,1312.03 341.544,1309.86 342.398,1307.7 343.252,1305.53 344.106,1303.37 344.96,1301.21 345.814,1299.06 346.668,1296.9 347.522,1294.75 348.375,1292.6 349.229,1290.45 350.083,1288.3 350.937,1286.16 351.791,1284.02 352.645,1281.88 353.499,1279.74 354.353,1277.61 355.207,1275.47 356.06,1273.35 356.914,1271.22 357.768,1269.1 358.622,1266.98 359.476,1264.86 360.33,1262.75 361.184,1260.64 362.038,1258.54 362.892,1256.43 363.746,1254.33 364.599,1252.24 365.453,1250.15 366.307,1248.06 367.161,1245.97 368.015,1243.89 368.869,1241.81 369.723,1239.74 370.577,1237.67 371.431,1235.6 372.284,1233.54 373.138,1231.48 373.992,1229.43 374.846,1227.38 375.7,1225.33 376.554,1223.29 377.408,1221.25 378.262,1219.22 379.116,1217.19 379.969,1215.16 380.823,1213.14 381.677,1211.12 382.531,1209.11 383.385,1207.1 384.239,1205.1 385.093,1203.1 385.947,1201.11 386.801,1199.12 387.654,1197.14 388.508,1195.16 389.362,1193.19 390.216,1191.23 391.07,1189.28 391.924,1187.33 392.778,1185.4 393.632,1183.49 394.486,1181.6 395.34,1179.74 396.193,1177.94 397.047,1176.22 397.901,1174.63 398.755,1173.21 399.609,1171.94 400.463,1170.82 401.317,1169.81 402.171,1168.91 403.025,1168.08 403.878,1167.31 404.732,1166.59 405.586,1165.9 406.44,1165.25 407.294,1164.62 408.148,1164 409.002,1163.4 409.856,1162.8 410.71,1162.21 411.563,1161.63 412.417,1161.05 413.271,1160.48 414.125,1159.9 414.979,1159.32 415.833,1158.75 416.687,1158.17 417.541,1157.59 418.395,1157.01 419.249,1156.43 420.102,1155.85 420.956,1155.26 421.81,1154.67 422.664,1154.08 423.518,1153.49 424.372,1152.89 425.226,1152.29 426.08,1151.69 426.934,1151.09 427.787,1150.48 428.641,1149.87 429.495,1149.26 430.349,1148.64 431.203,1148.02 432.057,1147.4 432.911,1146.77 433.765,1146.14 434.619,1145.51 435.472,1144.88 436.326,1144.24 437.18,1143.6 438.034,1142.96 438.888,1142.31 439.742,1141.66 440.596,1141.01 441.45,1140.35 442.304,1139.69 443.158,1139.02 444.011,1138.36 444.865,1137.69 445.719,1137.01 446.573,1136.34 447.427,1135.65 448.281,1134.97 449.135,1134.28 449.989,1133.59 450.843,1132.9 451.696,1132.2 452.55,1131.49 453.404,1130.79 454.258,1130.08 455.112,1129.37 455.966,1128.65 456.82,1127.93 457.674,1127.2 458.528,1126.47 459.381,1125.74 460.235,1125.01 461.089,1124.27 461.943,1123.52 462.797,1122.77 463.651,1122.02 464.505,1121.27 465.359,1120.51 466.213,1119.74 467.066,1118.97 467.92,1118.2 468.774,1117.42 469.628,1116.64 470.482,1115.86 471.336,1115.07 472.19,1114.27 473.044,1113.47 473.898,1112.67 474.752,1111.86 475.605,1111.05 476.459,1110.24 477.313,1109.41 478.167,1108.59 479.021,1107.76 479.875,1106.92 480.729,1106.09 481.583,1105.24 482.437,1104.39 483.29,1103.54 484.144,1102.68 484.998,1101.82 485.852,1100.95 486.706,1100.07 487.56,1099.2 488.414,1098.31 489.268,1097.42 490.122,1096.53 490.975,1095.63 491.829,1094.73 492.683,1093.82 493.537,1092.9 494.391,1091.98 495.245,1091.05 496.099,1090.12 496.953,1089.19 497.807,1088.24 498.661,1087.3 499.514,1086.34 500.368,1085.38 501.222,1084.42 502.076,1083.45 502.93,1082.47 503.784,1081.49 504.638,1080.5 505.492,1079.5 506.346,1078.5 507.199,1077.5 508.053,1076.48 508.907,1075.47 509.761,1074.44 510.615,1073.41 511.469,1072.37 512.323,1071.32 513.177,1070.27 514.031,1069.22 514.884,1068.15 515.738,1067.08 516.592,1066 517.446,1064.92 518.3,1063.83 519.154,1062.73 520.008,1061.62 520.862,1060.51 521.716,1059.39 522.569,1058.26 523.423,1057.13 524.277,1055.99 525.131,1054.84 525.985,1053.68 526.839,1052.52 527.693,1051.34 528.547,1050.16 529.401,1048.98 530.255,1047.78 531.108,1046.58 531.962,1045.37 532.816,1044.15 533.67,1042.92 534.524,1041.69 535.378,1040.44 536.232,1039.19 537.086,1037.93 537.94,1036.66 538.793,1035.38 539.647,1034.1 540.501,1032.8 541.355,1031.5 542.209,1030.18 543.063,1028.86 543.917,1027.53 544.771,1026.19 545.625,1024.84 546.478,1023.48 547.332,1022.11 548.186,1020.73 549.04,1019.34 549.894,1017.95 550.748,1016.54 551.602,1015.12 552.456,1013.69 553.31,1012.25 554.164,1010.8 555.017,1009.34 555.871,1007.87 556.725,1006.39 557.579,1004.9 558.433,1003.4 559.287,1001.88 560.141,1000.36 560.995,998.823 561.849,997.276 562.702,995.717 563.556,994.147 564.41,992.565 565.264,990.971 566.118,989.365 566.972,987.747 567.826,986.117 568.68,984.475 569.534,982.82 570.387,981.152 571.241,979.472 572.095,977.779 572.949,976.074 573.803,974.355 574.657,972.623 575.511,970.877 576.365,969.119 577.219,967.346 578.072,965.561 578.926,963.761 579.78,961.948 580.634,960.121 581.488,958.28 582.342,956.425 583.196,954.556 584.05,952.673 584.904,950.776 585.758,948.865 586.611,946.939 587.465,945 588.319,943.046 589.173,941.079 590.027,939.098 590.881,937.103 591.735,935.094 592.589,933.072 593.443,931.038 594.296,928.99 595.15,926.931 596.004,924.859 596.858,922.776 597.712,920.682 598.566,918.579 599.42,916.466 600.274,914.345 601.128,912.217 601.981,910.082 602.835,907.943 603.689,905.801 604.543,903.657 605.397,901.514 606.251,899.374 607.105,897.239 607.959,895.112 608.813,892.997 609.667,890.896 610.52,888.814 611.374,886.756 612.228,884.726 613.082,882.731 613.936,880.776 614.79,878.869 615.644,877.018 616.498,875.232 617.352,873.52 618.205,871.894 619.059,870.367 619.913,868.95 620.767,867.66 621.621,866.511 622.475,865.522 623.329,864.712 624.183,864.098 625.037,863.703 625.89,863.544 626.744,863.64 627.598,864 628.452,864.622 629.306,865.479 630.16,866.509 631.014,867.645 631.868,868.837 632.722,870.06 633.576,871.3 634.429,872.548 635.283,873.799 636.137,875.052 636.991,876.303 637.845,877.552 638.699,878.798 639.553,880.039 640.407,881.276 641.261,882.509 642.114,883.736 642.968,884.958 643.822,886.174 644.676,887.385 645.53,888.591 646.384,889.791 647.238,890.986 648.092,892.176 648.946,893.36 649.799,894.538 650.653,895.711 651.507,896.879 652.361,898.042 653.215,899.2 654.069,900.352 654.923,901.5 655.777,902.642 656.631,903.78 657.484,904.913 658.338,906.041 659.192,907.164 660.046,908.283 660.9,909.397 661.754,910.507 662.608,911.613 663.462,912.714 664.316,913.811 665.17,914.903 666.023,915.992 666.877,917.076 667.731,918.157 668.585,919.233 669.439,920.306 670.293,921.375 671.147,922.44 672.001,923.502 672.855,924.56 673.708,925.614 674.562,926.665 675.416,927.713 676.27,928.757 677.124,929.798 677.978,930.835 678.832,931.87 679.686,932.901 680.54,933.929 681.393,934.954 682.247,935.976 683.101,936.995 683.955,938.011 684.809,939.025 685.663,940.035 686.517,941.043 687.371,942.048 688.225,943.05 689.079,944.05 689.932,945.047 690.786,946.042 691.64,947.034 692.494,948.023 693.348,949.011 694.202,949.996 695.056,950.978 695.91,951.958 696.764,952.936 697.617,953.912 698.471,954.885 699.325,955.857 700.179,956.826 701.033,957.793 701.887,958.758 702.741,959.721 703.595,960.682 704.449,961.641 705.302,962.598 706.156,963.553 707.01,964.507 707.864,965.458 708.718,966.408 709.572,967.356 710.426,968.302 711.28,969.247 712.134,970.189 712.987,971.13 713.841,972.07 714.695,973.008 715.549,973.944 716.403,974.878 717.257,975.811 718.111,976.743 718.965,977.673 719.819,978.602 720.673,979.529 721.526,980.454 722.38,981.379 723.234,982.302 724.088,983.223 724.942,984.143 725.796,985.062 726.65,985.98 727.504,986.896 728.358,987.811 729.211,988.724 730.065,989.637 730.919,990.548 731.773,991.458 732.627,992.367 733.481,993.275 734.335,994.182 735.189,995.087 736.043,995.991 736.896,996.895 737.75,997.797 738.604,998.698 739.458,999.598 740.312,1000.5 741.166,1001.4 742.02,1002.29 742.874,1003.19 743.728,1004.08 744.582,1004.98 745.435,1005.87 746.289,1006.76 747.143,1007.65 747.997,1008.55 748.851,1009.43 749.705,1010.32 750.559,1011.21 751.413,1012.1 752.267,1012.98 753.12,1013.87 753.974,1014.75 754.828,1015.64 755.682,1016.52 756.536,1017.4 757.39,1018.28 758.244,1019.17 759.098,1020.05 759.952,1020.92 760.805,1021.8 761.659,1022.68 762.513,1023.56 763.367,1024.43 764.221,1025.31 765.075,1026.18 765.929,1027.06 766.783,1027.93 767.637,1028.81 768.491,1029.68 769.344,1030.55 770.198,1031.42 771.052,1032.29 771.906,1033.16 772.76,1034.03 773.614,1034.9 774.468,1035.77 775.322,1036.63 776.176,1037.5 777.029,1038.37 777.883,1039.23 778.737,1040.1 779.591,1040.96 780.445,1041.83 781.299,1042.69 782.153,1043.56 783.007,1044.42 783.861,1045.28 784.714,1046.14 785.568,1047 786.422,1047.86 787.276,1048.72 788.13,1049.58 788.984,1050.44 789.838,1051.3 790.692,1052.16 791.546,1053.02 792.399,1053.88 793.253,1054.73 794.107,1055.59 794.961,1056.45 795.815,1057.3 796.669,1058.16 797.523,1059.01 798.377,1059.87 799.231,1060.72 800.085,1061.58 800.938,1062.43 801.792,1063.28 802.646,1064.14 803.5,1064.99 804.354,1065.84 805.208,1066.69 806.062,1067.54 806.916,1068.39 807.77,1069.25 808.623,1070.1 809.477,1070.95 810.331,1071.8 811.185,1072.65 812.039,1073.49 812.893,1074.34 813.747,1075.19 814.601,1076.04 815.455,1076.89 816.308,1077.73 817.162,1078.58 818.016,1079.43 818.87,1080.28 819.724,1081.12 820.578,1081.97 821.432,1082.81 822.286,1083.66 823.14,1084.5 823.994,1085.35 824.847,1086.19 825.701,1087.04 826.555,1087.88 827.409,1088.73 828.263,1089.57 829.117,1090.41 829.971,1091.26 830.825,1092.1 831.679,1092.94 832.532,1093.78 833.386,1094.63 834.24,1095.47 835.094,1096.31 835.948,1097.15 836.802,1097.99 837.656,1098.83 838.51,1099.68 839.364,1100.52 840.217,1101.36 841.071,1102.2 841.925,1103.04 842.779,1103.88 843.633,1104.72 844.487,1105.56 845.341,1106.39 846.195,1107.23 847.049,1108.07 847.902,1108.91 848.756,1109.75 849.61,1110.59 850.464,1111.43 851.318,1112.26 852.172,1113.1 853.026,1113.94 853.88,1114.78 854.734,1115.61 855.588,1116.45 856.441,1117.29 857.295,1118.12 858.149,1118.96 859.003,1119.8 859.857,1120.63 860.711,1121.47 861.565,1122.3 862.419,1123.14 863.273,1123.98 864.126,1124.81 864.98,1125.65 865.834,1126.48 866.688,1127.32 867.542,1128.15 868.396,1128.99 869.25,1129.82 870.104,1130.65 870.958,1131.49 871.811,1132.32 872.665,1133.16 873.519,1133.99 874.373,1134.82 875.227,1135.66 876.081,1136.49 876.935,1137.32 877.789,1138.16 878.643,1138.99 879.497,1139.82 880.35,1140.66 881.204,1141.49 882.058,1142.32 882.912,1143.15 883.766,1143.99 884.62,1144.82 885.474,1145.65 886.328,1146.48 887.182,1147.31 888.035,1148.15 888.889,1148.98 889.743,1149.81 890.597,1150.64 891.451,1151.47 892.305,1152.3 893.159,1153.13 894.013,1153.96 894.867,1154.79 895.72,1155.63 896.574,1156.46 897.428,1157.29 898.282,1158.12 899.136,1158.95 899.99,1159.78 900.844,1160.61 901.698,1161.44 902.552,1162.27 903.406,1163.1 904.259,1163.93 905.113,1164.76 905.967,1165.59 906.821,1166.42 907.675,1167.25 908.529,1168.08 909.383,1168.9 910.237,1169.73 911.091,1170.56 911.944,1171.39 912.798,1172.22 913.652,1173.05 914.506,1173.88 915.36,1174.71 916.214,1175.54 917.068,1176.36 917.922,1177.19 918.776,1178.02 919.629,1178.85 920.483,1179.68 921.337,1180.51 922.191,1181.33 923.045,1182.16 923.899,1182.99 924.753,1183.82 925.607,1184.65 926.461,1185.47 927.314,1186.3 928.168,1187.13 929.022,1187.96 929.876,1188.78 930.73,1189.61 931.584,1190.44 932.438,1191.27 933.292,1192.09 934.146,1192.92 935,1193.75 935.853,1194.57 936.707,1195.4 937.561,1196.23 938.415,1197.05 939.269,1197.88 940.123,1198.71 940.977,1199.53 941.831,1200.36 942.685,1201.19 943.538,1202.01 944.392,1202.84 945.246,1203.67 946.1,1204.49 946.954,1205.32 947.808,1206.15 948.662,1206.97 949.516,1207.8 950.37,1208.62 951.223,1209.45 952.077,1210.28 952.931,1211.1 953.785,1211.93 954.639,1212.75 955.493,1213.58 956.347,1214.4 957.201,1215.23 958.055,1216.06 958.909,1216.88 959.762,1217.71 960.616,1218.53 961.47,1219.36 962.324,1220.18 963.178,1221.01 964.032,1221.83 964.886,1222.66 965.74,1223.48 966.594,1224.31 967.447,1225.13 968.301,1225.96 969.155,1226.78 970.009,1227.61 970.863,1228.43 971.717,1229.26 972.571,1230.08 973.425,1230.91 974.279,1231.73 975.132,1232.56 975.986,1233.38 976.84,1234.21 977.694,1235.03 978.548,1235.86 979.402,1236.68 980.256,1237.51 981.11,1238.33 981.964,1239.15 982.817,1239.98 983.671,1240.8 984.525,1241.63 985.379,1242.45 986.233,1243.28 987.087,1244.1 987.941,1244.92 988.795,1245.75 989.649,1246.57 990.503,1247.4 991.356,1248.22 992.21,1249.04 993.064,1249.87 993.918,1250.69 994.772,1251.52 995.626,1252.34 996.48,1253.16 997.334,1253.99 998.188,1254.81 999.041,1255.63 999.895,1256.46 1000.75,1257.28 1001.6,1258.11 1002.46,1258.93 1003.31,1259.75 1004.16,1260.58 1005.02,1261.4 1005.87,1262.22 1006.73,1263.05 1007.58,1263.87 1008.43,1264.69 1009.29,1265.52 1010.14,1266.34 1011,1267.16 1011.85,1267.99 1012.7,1268.81 1013.56,1269.63 1014.41,1270.46 1015.27,1271.28 1016.12,1272.1 1016.97,1272.93 1017.83,1273.75 1018.68,1274.57 1019.53,1275.4 1020.39,1276.22 1021.24,1277.04 1022.1,1277.86 1022.95,1278.69 1023.8,1279.51 1024.66,1280.33 1025.51,1281.16 1026.37,1281.98 1027.22,1282.8 1028.07,1283.62 1028.93,1284.45 1029.78,1285.27 1030.64,1286.09 1031.49,1286.92 1032.34,1287.74 1033.2,1288.56 1034.05,1289.38 1034.9,1290.21 1035.76,1291.03 1036.61,1291.85 1037.47,1292.67 1038.32,1293.5 1039.17,1294.32 1040.03,1295.14 1040.88,1295.96 1041.74,1296.79 1042.59,1297.61 1043.44,1298.43 1044.3,1299.25 1045.15,1300.08 1046.01,1300.9 1046.86,1301.72 1047.71,1302.54 1048.57,1303.37 1049.42,1304.19 1050.27,1305.01 1051.13,1305.83 1051.98,1306.66 1052.84,1307.48 1053.69,1308.3 1054.54,1309.12 1055.4,1309.94 1056.25,1310.77 1057.11,1311.59 1057.96,1312.41 1058.81,1313.23 1059.67,1314.05 1060.52,1314.88 1061.38,1315.7 1062.23,1316.52 1063.08,1317.34 1063.94,1318.17 1064.79,1318.99 1065.65,1319.81 1066.5,1320.63 1067.35,1321.45 1068.21,1322.28 1069.06,1323.1 1069.91,1323.92 1070.77,1324.74 1071.62,1325.56 1072.48,1326.38 1073.33,1327.21 1074.18,1328.03 1075.04,1328.85 1075.89,1329.67 1076.75,1330.49 1077.6,1331.32 1078.45,1332.14 1079.31,1332.96 1080.16,1333.78 1081.02,1334.6 1081.87,1335.43 1082.72,1336.25 1083.58,1337.07 1084.43,1337.89 1085.28,1338.71 1086.14,1339.53 1086.99,1340.36 1087.85,1341.18 1088.7,1342 1089.55,1342.82 1090.41,1343.64 1091.26,1344.46 1092.12,1345.29 1092.97,1346.11 1093.82,1346.93 1094.68,1347.75 1095.53,1348.57 1096.39,1349.39 1097.24,1350.21 1098.09,1351.04 1098.95,1351.86 1099.8,1352.68 1100.65,1353.5 1101.51,1354.32 1102.36,1355.14 1103.22,1355.97 1104.07,1356.79 1104.92,1357.61 1105.78,1358.43 1106.63,1359.25 1107.49,1360.07 1108.34,1360.89 1109.19,1361.72 1110.05,1362.54 1110.9,1363.36 1111.76,1364.18 1112.61,1365 1113.46,1365.82 1114.32,1366.64 1115.17,1367.47 1116.02,1368.29 1116.88,1369.11 1117.73,1369.93 1118.59,1370.75 1119.44,1371.57 1120.29,1372.39 1121.15,1373.22 1122,1374.04 1122.86,1374.86 1123.71,1375.68 1124.56,1376.5 1125.42,1377.32 1126.27,1378.14 1127.13,1378.96 1127.98,1379.79 1128.83,1380.61 1129.69,1381.43 1130.54,1382.25 1131.39,1383.07 1132.25,1383.89 1133.1,1384.71 1133.96,1385.53 1134.81,1386.36 1135.66,1387.18 1136.52,1388 1137.37,1388.82 1138.23,1389.64 1139.08,1390.46 1139.93,1391.28 1140.79,1392.1 1141.64,1392.92 1142.5,1393.75 1143.35,1394.57 1144.2,1395.39 1145.06,1396.21 1145.91,1397.03 1146.76,1397.85 1147.62,1398.67 1148.47,1399.49 1149.33,1400.31 1150.18,1401.14 1151.03,1401.96 1151.89,1402.78 1152.74,1403.6 1153.6,1404.42 1154.45,1405.24 1155.3,1406.06 1156.16,1406.88 "/>
+<path clip-path="url(#clip970)" d="M1448.54 1423.18 L2352.76 1423.18 L2352.76 847.244 L1448.54 847.244  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip155">
+  <clipPath id="clip975">
     <rect x="1448" y="847" width="905" height="577"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1473.27,1423.18 1473.27,847.244 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1686.75,1423.18 1686.75,847.244 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1900.22,1423.18 1900.22,847.244 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2113.69,1423.18 2113.69,847.244 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2327.16,1423.18 2327.16,847.244 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1423.18 2352.76,1423.18 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1473.27,1423.18 1473.27,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1686.75,1423.18 1686.75,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1900.22,1423.18 1900.22,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2113.69,1423.18 2113.69,1404.28 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2327.16,1423.18 2327.16,1404.28 "/>
-<path clip-path="url(#clip150)" d="M1435.58 1454.1 Q1431.96 1454.1 1430.14 1457.66 Q1428.33 1461.2 1428.33 1468.33 Q1428.33 1475.44 1430.14 1479.01 Q1431.96 1482.55 1435.58 1482.55 Q1439.21 1482.55 1441.02 1479.01 Q1442.84 1475.44 1442.84 1468.33 Q1442.84 1461.2 1441.02 1457.66 Q1439.21 1454.1 1435.58 1454.1 M1435.58 1450.39 Q1441.39 1450.39 1444.44 1455 Q1447.52 1459.58 1447.52 1468.33 Q1447.52 1477.06 1444.44 1481.67 Q1441.39 1486.25 1435.58 1486.25 Q1429.77 1486.25 1426.69 1481.67 Q1423.63 1477.06 1423.63 1468.33 Q1423.63 1459.58 1426.69 1455 Q1429.77 1450.39 1435.58 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1455.74 1479.7 L1460.62 1479.7 L1460.62 1485.58 L1455.74 1485.58 L1455.74 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1480.81 1454.1 Q1477.2 1454.1 1475.37 1457.66 Q1473.56 1461.2 1473.56 1468.33 Q1473.56 1475.44 1475.37 1479.01 Q1477.2 1482.55 1480.81 1482.55 Q1484.44 1482.55 1486.25 1479.01 Q1488.08 1475.44 1488.08 1468.33 Q1488.08 1461.2 1486.25 1457.66 Q1484.44 1454.1 1480.81 1454.1 M1480.81 1450.39 Q1486.62 1450.39 1489.67 1455 Q1492.75 1459.58 1492.75 1468.33 Q1492.75 1477.06 1489.67 1481.67 Q1486.62 1486.25 1480.81 1486.25 Q1475 1486.25 1471.92 1481.67 Q1468.86 1477.06 1468.86 1468.33 Q1468.86 1459.58 1471.92 1455 Q1475 1450.39 1480.81 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1510.97 1454.1 Q1507.36 1454.1 1505.53 1457.66 Q1503.72 1461.2 1503.72 1468.33 Q1503.72 1475.44 1505.53 1479.01 Q1507.36 1482.55 1510.97 1482.55 Q1514.6 1482.55 1516.41 1479.01 Q1518.24 1475.44 1518.24 1468.33 Q1518.24 1461.2 1516.41 1457.66 Q1514.6 1454.1 1510.97 1454.1 M1510.97 1450.39 Q1516.78 1450.39 1519.83 1455 Q1522.91 1459.58 1522.91 1468.33 Q1522.91 1477.06 1519.83 1481.67 Q1516.78 1486.25 1510.97 1486.25 Q1505.16 1486.25 1502.08 1481.67 Q1499.02 1477.06 1499.02 1468.33 Q1499.02 1459.58 1502.08 1455 Q1505.16 1450.39 1510.97 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1649.55 1454.1 Q1645.94 1454.1 1644.11 1457.66 Q1642.3 1461.2 1642.3 1468.33 Q1642.3 1475.44 1644.11 1479.01 Q1645.94 1482.55 1649.55 1482.55 Q1653.18 1482.55 1654.99 1479.01 Q1656.82 1475.44 1656.82 1468.33 Q1656.82 1461.2 1654.99 1457.66 Q1653.18 1454.1 1649.55 1454.1 M1649.55 1450.39 Q1655.36 1450.39 1658.41 1455 Q1661.49 1459.58 1661.49 1468.33 Q1661.49 1477.06 1658.41 1481.67 Q1655.36 1486.25 1649.55 1486.25 Q1643.74 1486.25 1640.66 1481.67 Q1637.6 1477.06 1637.6 1468.33 Q1637.6 1459.58 1640.66 1455 Q1643.74 1450.39 1649.55 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1669.71 1479.7 L1674.59 1479.7 L1674.59 1485.58 L1669.71 1485.58 L1669.71 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1694.78 1454.1 Q1691.17 1454.1 1689.34 1457.66 Q1687.53 1461.2 1687.53 1468.33 Q1687.53 1475.44 1689.34 1479.01 Q1691.17 1482.55 1694.78 1482.55 Q1698.41 1482.55 1700.22 1479.01 Q1702.05 1475.44 1702.05 1468.33 Q1702.05 1461.2 1700.22 1457.66 Q1698.41 1454.1 1694.78 1454.1 M1694.78 1450.39 Q1700.59 1450.39 1703.64 1455 Q1706.72 1459.58 1706.72 1468.33 Q1706.72 1477.06 1703.64 1481.67 Q1700.59 1486.25 1694.78 1486.25 Q1688.97 1486.25 1685.89 1481.67 Q1682.83 1477.06 1682.83 1468.33 Q1682.83 1459.58 1685.89 1455 Q1688.97 1450.39 1694.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1714.99 1451.02 L1733.34 1451.02 L1733.34 1454.96 L1719.27 1454.96 L1719.27 1463.43 Q1720.29 1463.08 1721.31 1462.92 Q1722.32 1462.73 1723.34 1462.73 Q1729.13 1462.73 1732.51 1465.9 Q1735.89 1469.08 1735.89 1474.49 Q1735.89 1480.07 1732.42 1483.17 Q1728.94 1486.25 1722.62 1486.25 Q1720.45 1486.25 1718.18 1485.88 Q1715.94 1485.51 1713.53 1484.77 L1713.53 1480.07 Q1715.61 1481.2 1717.83 1481.76 Q1720.06 1482.32 1722.53 1482.32 Q1726.54 1482.32 1728.87 1480.21 Q1731.21 1478.1 1731.21 1474.49 Q1731.21 1470.88 1728.87 1468.77 Q1726.54 1466.67 1722.53 1466.67 Q1720.66 1466.67 1718.78 1467.08 Q1716.93 1467.5 1714.99 1468.38 L1714.99 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1862.52 1454.1 Q1858.91 1454.1 1857.08 1457.66 Q1855.28 1461.2 1855.28 1468.33 Q1855.28 1475.44 1857.08 1479.01 Q1858.91 1482.55 1862.52 1482.55 Q1866.16 1482.55 1867.96 1479.01 Q1869.79 1475.44 1869.79 1468.33 Q1869.79 1461.2 1867.96 1457.66 Q1866.16 1454.1 1862.52 1454.1 M1862.52 1450.39 Q1868.33 1450.39 1871.39 1455 Q1874.47 1459.58 1874.47 1468.33 Q1874.47 1477.06 1871.39 1481.67 Q1868.33 1486.25 1862.52 1486.25 Q1856.71 1486.25 1853.63 1481.67 Q1850.58 1477.06 1850.58 1468.33 Q1850.58 1459.58 1853.63 1455 Q1856.71 1450.39 1862.52 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1882.68 1479.7 L1887.57 1479.7 L1887.57 1485.58 L1882.68 1485.58 L1882.68 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1898.56 1481.64 L1906.2 1481.64 L1906.2 1455.28 L1897.89 1456.95 L1897.89 1452.69 L1906.16 1451.02 L1910.83 1451.02 L1910.83 1481.64 L1918.47 1481.64 L1918.47 1485.58 L1898.56 1485.58 L1898.56 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1937.92 1454.1 Q1934.3 1454.1 1932.48 1457.66 Q1930.67 1461.2 1930.67 1468.33 Q1930.67 1475.44 1932.48 1479.01 Q1934.3 1482.55 1937.92 1482.55 Q1941.55 1482.55 1943.35 1479.01 Q1945.18 1475.44 1945.18 1468.33 Q1945.18 1461.2 1943.35 1457.66 Q1941.55 1454.1 1937.92 1454.1 M1937.92 1450.39 Q1943.73 1450.39 1946.78 1455 Q1949.86 1459.58 1949.86 1468.33 Q1949.86 1477.06 1946.78 1481.67 Q1943.73 1486.25 1937.92 1486.25 Q1932.1 1486.25 1929.03 1481.67 Q1925.97 1477.06 1925.97 1468.33 Q1925.97 1459.58 1929.03 1455 Q1932.1 1450.39 1937.92 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2076.49 1454.1 Q2072.88 1454.1 2071.05 1457.66 Q2069.25 1461.2 2069.25 1468.33 Q2069.25 1475.44 2071.05 1479.01 Q2072.88 1482.55 2076.49 1482.55 Q2080.13 1482.55 2081.93 1479.01 Q2083.76 1475.44 2083.76 1468.33 Q2083.76 1461.2 2081.93 1457.66 Q2080.13 1454.1 2076.49 1454.1 M2076.49 1450.39 Q2082.3 1450.39 2085.36 1455 Q2088.44 1459.58 2088.44 1468.33 Q2088.44 1477.06 2085.36 1481.67 Q2082.3 1486.25 2076.49 1486.25 Q2070.68 1486.25 2067.6 1481.67 Q2064.55 1477.06 2064.55 1468.33 Q2064.55 1459.58 2067.6 1455 Q2070.68 1450.39 2076.49 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2096.65 1479.7 L2101.54 1479.7 L2101.54 1485.58 L2096.65 1485.58 L2096.65 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2112.53 1481.64 L2120.17 1481.64 L2120.17 1455.28 L2111.86 1456.95 L2111.86 1452.69 L2120.13 1451.02 L2124.8 1451.02 L2124.8 1481.64 L2132.44 1481.64 L2132.44 1485.58 L2112.53 1485.58 L2112.53 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2141.93 1451.02 L2160.29 1451.02 L2160.29 1454.96 L2146.21 1454.96 L2146.21 1463.43 Q2147.23 1463.08 2148.25 1462.92 Q2149.27 1462.73 2150.29 1462.73 Q2156.08 1462.73 2159.46 1465.9 Q2162.83 1469.08 2162.83 1474.49 Q2162.83 1480.07 2159.36 1483.17 Q2155.89 1486.25 2149.57 1486.25 Q2147.4 1486.25 2145.13 1485.88 Q2142.88 1485.51 2140.47 1484.77 L2140.47 1480.07 Q2142.56 1481.2 2144.78 1481.76 Q2147 1482.32 2149.48 1482.32 Q2153.48 1482.32 2155.82 1480.21 Q2158.16 1478.1 2158.16 1474.49 Q2158.16 1470.88 2155.82 1468.77 Q2153.48 1466.67 2149.48 1466.67 Q2147.6 1466.67 2145.73 1467.08 Q2143.88 1467.5 2141.93 1468.38 L2141.93 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2289.47 1454.1 Q2285.86 1454.1 2284.03 1457.66 Q2282.22 1461.2 2282.22 1468.33 Q2282.22 1475.44 2284.03 1479.01 Q2285.86 1482.55 2289.47 1482.55 Q2293.1 1482.55 2294.91 1479.01 Q2296.74 1475.44 2296.74 1468.33 Q2296.74 1461.2 2294.91 1457.66 Q2293.1 1454.1 2289.47 1454.1 M2289.47 1450.39 Q2295.28 1450.39 2298.33 1455 Q2301.41 1459.58 2301.41 1468.33 Q2301.41 1477.06 2298.33 1481.67 Q2295.28 1486.25 2289.47 1486.25 Q2283.66 1486.25 2280.58 1481.67 Q2277.52 1477.06 2277.52 1468.33 Q2277.52 1459.58 2280.58 1455 Q2283.66 1450.39 2289.47 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2309.63 1479.7 L2314.51 1479.7 L2314.51 1485.58 L2309.63 1485.58 L2309.63 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2328.73 1481.64 L2345.05 1481.64 L2345.05 1485.58 L2323.1 1485.58 L2323.1 1481.64 Q2325.76 1478.89 2330.35 1474.26 Q2334.95 1469.61 2336.13 1468.27 Q2338.38 1465.74 2339.26 1464.01 Q2340.16 1462.25 2340.16 1460.56 Q2340.16 1457.8 2338.22 1456.07 Q2336.3 1454.33 2333.19 1454.33 Q2331 1454.33 2328.54 1455.09 Q2326.11 1455.86 2323.33 1457.41 L2323.33 1452.69 Q2326.16 1451.55 2328.61 1450.97 Q2331.07 1450.39 2333.1 1450.39 Q2338.47 1450.39 2341.67 1453.08 Q2344.86 1455.77 2344.86 1460.26 Q2344.86 1462.39 2344.05 1464.31 Q2343.26 1466.2 2341.16 1468.8 Q2340.58 1469.47 2337.48 1472.69 Q2334.38 1475.88 2328.73 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2364.86 1454.1 Q2361.25 1454.1 2359.42 1457.66 Q2357.62 1461.2 2357.62 1468.33 Q2357.62 1475.44 2359.42 1479.01 Q2361.25 1482.55 2364.86 1482.55 Q2368.5 1482.55 2370.3 1479.01 Q2372.13 1475.44 2372.13 1468.33 Q2372.13 1461.2 2370.3 1457.66 Q2368.5 1454.1 2364.86 1454.1 M2364.86 1450.39 Q2370.67 1450.39 2373.73 1455 Q2376.81 1459.58 2376.81 1468.33 Q2376.81 1477.06 2373.73 1481.67 Q2370.67 1486.25 2364.86 1486.25 Q2359.05 1486.25 2355.97 1481.67 Q2352.92 1477.06 2352.92 1468.33 Q2352.92 1459.58 2355.97 1455 Q2359.05 1450.39 2364.86 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1770.82 1520.52 L1811.02 1520.52 L1811.02 1525.93 L1794.15 1525.93 L1794.15 1568.04 L1787.69 1568.04 L1787.69 1525.93 L1770.82 1525.93 L1770.82 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1814.96 1532.4 L1820.82 1532.4 L1820.82 1568.04 L1814.96 1568.04 L1814.96 1532.4 M1814.96 1518.52 L1820.82 1518.52 L1820.82 1525.93 L1814.96 1525.93 L1814.96 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1860.83 1539.24 Q1863.02 1535.29 1866.08 1533.41 Q1869.14 1531.54 1873.27 1531.54 Q1878.84 1531.54 1881.87 1535.45 Q1884.89 1539.33 1884.89 1546.53 L1884.89 1568.04 L1879 1568.04 L1879 1546.72 Q1879 1541.59 1877.19 1539.11 Q1875.37 1536.63 1871.65 1536.63 Q1867.1 1536.63 1864.46 1539.65 Q1861.81 1542.68 1861.81 1547.9 L1861.81 1568.04 L1855.93 1568.04 L1855.93 1546.72 Q1855.93 1541.56 1854.11 1539.11 Q1852.3 1536.63 1848.51 1536.63 Q1844.02 1536.63 1841.38 1539.68 Q1838.74 1542.71 1838.74 1547.9 L1838.74 1568.04 L1832.85 1568.04 L1832.85 1532.4 L1838.74 1532.4 L1838.74 1537.93 Q1840.74 1534.66 1843.55 1533.1 Q1846.35 1531.54 1850.2 1531.54 Q1854.08 1531.54 1856.79 1533.51 Q1859.52 1535.48 1860.83 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1927.06 1548.76 L1927.06 1551.62 L1900.14 1551.62 Q1900.52 1557.67 1903.76 1560.85 Q1907.04 1564 1912.87 1564 Q1916.24 1564 1919.39 1563.17 Q1922.58 1562.35 1925.69 1560.69 L1925.69 1566.23 Q1922.54 1567.57 1919.23 1568.27 Q1915.92 1568.97 1912.52 1568.97 Q1903.99 1568.97 1898.99 1564 Q1894.03 1559.04 1894.03 1550.57 Q1894.03 1541.82 1898.74 1536.69 Q1903.48 1531.54 1911.5 1531.54 Q1918.69 1531.54 1922.86 1536.18 Q1927.06 1540.8 1927.06 1548.76 M1921.21 1547.04 Q1921.14 1542.23 1918.5 1539.37 Q1915.89 1536.5 1911.56 1536.5 Q1906.66 1536.5 1903.7 1539.27 Q1900.77 1542.04 1900.33 1547.07 L1921.21 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1971.46 1518.58 Q1967.2 1525.9 1965.13 1533.06 Q1963.06 1540.23 1963.06 1547.58 Q1963.06 1554.93 1965.13 1562.16 Q1967.23 1569.35 1971.46 1576.64 L1966.37 1576.64 Q1961.6 1569.16 1959.21 1561.93 Q1956.85 1554.71 1956.85 1547.58 Q1956.85 1540.48 1959.21 1533.29 Q1961.57 1526.09 1966.37 1518.58 L1971.46 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2005.55 1533.45 L2005.55 1538.98 Q2003.07 1537.71 2000.4 1537.07 Q1997.72 1536.44 1994.86 1536.44 Q1990.5 1536.44 1988.3 1537.77 Q1986.14 1539.11 1986.14 1541.79 Q1986.14 1543.82 1987.7 1545 Q1989.26 1546.15 1993.97 1547.2 L1995.97 1547.64 Q2002.21 1548.98 2004.82 1551.43 Q2007.46 1553.85 2007.46 1558.21 Q2007.46 1563.17 2003.52 1566.07 Q1999.6 1568.97 1992.73 1568.97 Q1989.86 1568.97 1986.74 1568.39 Q1983.65 1567.85 1980.22 1566.74 L1980.22 1560.69 Q1983.46 1562.38 1986.61 1563.24 Q1989.77 1564.07 1992.85 1564.07 Q1996.99 1564.07 1999.22 1562.66 Q2001.45 1561.23 2001.45 1558.65 Q2001.45 1556.27 1999.82 1554.99 Q1998.23 1553.72 1992.79 1552.54 L1990.75 1552.07 Q1985.31 1550.92 1982.89 1548.56 Q1980.47 1546.18 1980.47 1542.04 Q1980.47 1537.01 1984.04 1534.27 Q1987.6 1531.54 1994.16 1531.54 Q1997.4 1531.54 2000.27 1532.01 Q2003.13 1532.49 2005.55 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M2015.86 1518.58 L2020.96 1518.58 Q2025.73 1526.09 2028.09 1533.29 Q2030.47 1540.48 2030.47 1547.58 Q2030.47 1554.71 2028.09 1561.93 Q2025.73 1569.16 2020.96 1576.64 L2015.86 1576.64 Q2020.1 1569.35 2022.17 1562.16 Q2024.27 1554.93 2024.27 1547.58 Q2024.27 1540.23 2022.17 1533.06 Q2020.1 1525.9 2015.86 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1406.88 2352.76,1406.88 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1251.63 2352.76,1251.63 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1096.38 2352.76,1096.38 "/>
-<polyline clip-path="url(#clip155)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,941.129 2352.76,941.129 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1423.18 1448.54,847.244 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1406.88 1467.43,1406.88 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1251.63 1467.43,1251.63 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1096.38 1467.43,1096.38 "/>
-<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,941.129 1467.43,941.129 "/>
-<path clip-path="url(#clip150)" d="M1400.59 1392.68 Q1396.98 1392.68 1395.15 1396.24 Q1393.35 1399.78 1393.35 1406.91 Q1393.35 1414.02 1395.15 1417.59 Q1396.98 1421.13 1400.59 1421.13 Q1404.22 1421.13 1406.03 1417.59 Q1407.86 1414.02 1407.86 1406.91 Q1407.86 1399.78 1406.03 1396.24 Q1404.22 1392.68 1400.59 1392.68 M1400.59 1388.97 Q1406.4 1388.97 1409.46 1393.58 Q1412.54 1398.16 1412.54 1406.91 Q1412.54 1415.64 1409.46 1420.25 Q1406.4 1424.83 1400.59 1424.83 Q1394.78 1424.83 1391.7 1420.25 Q1388.65 1415.64 1388.65 1406.91 Q1388.65 1398.16 1391.7 1393.58 Q1394.78 1388.97 1400.59 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1392.63 1264.97 L1400.27 1264.97 L1400.27 1238.61 L1391.96 1240.28 L1391.96 1236.02 L1400.22 1234.35 L1404.9 1234.35 L1404.9 1264.97 L1412.54 1264.97 L1412.54 1268.91 L1392.63 1268.91 L1392.63 1264.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1396.22 1109.72 L1412.54 1109.72 L1412.54 1113.66 L1390.59 1113.66 L1390.59 1109.72 Q1393.25 1106.97 1397.84 1102.34 Q1402.44 1097.69 1403.62 1096.34 Q1405.87 1093.82 1406.75 1092.09 Q1407.65 1090.33 1407.65 1088.64 Q1407.65 1085.88 1405.71 1084.15 Q1403.79 1082.41 1400.68 1082.41 Q1398.48 1082.41 1396.03 1083.17 Q1393.6 1083.94 1390.82 1085.49 L1390.82 1080.77 Q1393.65 1079.63 1396.1 1079.05 Q1398.55 1078.47 1400.59 1078.47 Q1405.96 1078.47 1409.16 1081.16 Q1412.35 1083.84 1412.35 1088.34 Q1412.35 1090.47 1411.54 1092.39 Q1410.75 1094.28 1408.65 1096.88 Q1408.07 1097.55 1404.97 1100.77 Q1401.86 1103.96 1396.22 1109.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1405.41 939.775 Q1408.76 940.493 1410.64 942.761 Q1412.54 945.03 1412.54 948.363 Q1412.54 953.479 1409.02 956.28 Q1405.5 959.081 1399.02 959.081 Q1396.84 959.081 1394.53 958.641 Q1392.23 958.224 1389.78 957.368 L1389.78 952.854 Q1391.73 953.988 1394.04 954.567 Q1396.35 955.146 1398.88 955.146 Q1403.28 955.146 1405.57 953.41 Q1407.88 951.673 1407.88 948.363 Q1407.88 945.308 1405.73 943.595 Q1403.6 941.859 1399.78 941.859 L1395.75 941.859 L1395.75 938.016 L1399.97 938.016 Q1403.41 938.016 1405.24 936.65 Q1407.07 935.261 1407.07 932.669 Q1407.07 930.007 1405.17 928.595 Q1403.3 927.16 1399.78 927.16 Q1397.86 927.16 1395.66 927.576 Q1393.46 927.993 1390.82 928.873 L1390.82 924.706 Q1393.48 923.965 1395.8 923.595 Q1398.14 923.225 1400.2 923.225 Q1405.52 923.225 1408.62 925.655 Q1411.72 928.062 1411.72 932.183 Q1411.72 935.053 1410.08 937.044 Q1408.44 939.011 1405.41 939.775 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1290.87 1238.91 L1290.87 1198.71 L1296.28 1198.71 L1296.28 1215.58 L1338.39 1215.58 L1338.39 1222.04 L1296.28 1222.04 L1296.28 1238.91 L1290.87 1238.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1316.87 1163.13 L1338.39 1163.13 L1338.39 1168.98 L1317.07 1168.98 Q1312 1168.98 1309.49 1170.96 Q1306.98 1172.93 1306.98 1176.88 Q1306.98 1181.62 1310 1184.36 Q1313.02 1187.09 1318.24 1187.09 L1338.39 1187.09 L1338.39 1192.98 L1288.87 1192.98 L1288.87 1187.09 L1308.28 1187.09 Q1305.07 1184.99 1303.47 1182.16 Q1301.88 1179.29 1301.88 1175.57 Q1301.88 1169.43 1305.7 1166.28 Q1309.49 1163.13 1316.87 1163.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1308.22 1130.79 Q1307.64 1131.77 1307.39 1132.95 Q1307.1 1134.1 1307.1 1135.5 Q1307.1 1140.46 1310.35 1143.14 Q1313.56 1145.78 1319.61 1145.78 L1338.39 1145.78 L1338.39 1151.67 L1302.74 1151.67 L1302.74 1145.78 L1308.28 1145.78 Q1305.03 1143.93 1303.47 1140.97 Q1301.88 1138.01 1301.88 1133.78 Q1301.88 1133.17 1301.98 1132.44 Q1302.04 1131.71 1302.2 1130.82 L1308.22 1130.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1324.32 1125.25 L1302.74 1125.25 L1302.74 1119.39 L1324.1 1119.39 Q1329.16 1119.39 1331.71 1117.42 Q1334.22 1115.45 1334.22 1111.5 Q1334.22 1106.76 1331.2 1104.02 Q1328.17 1101.25 1322.95 1101.25 L1302.74 1101.25 L1302.74 1095.39 L1338.39 1095.39 L1338.39 1101.25 L1332.92 1101.25 Q1336.16 1103.38 1337.75 1106.22 Q1339.31 1109.02 1339.31 1112.74 Q1339.31 1118.88 1335.49 1122.07 Q1331.67 1125.25 1324.32 1125.25 M1301.88 1110.51 L1301.88 1110.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1303.79 1060.61 L1309.33 1060.61 Q1308.06 1063.09 1307.42 1065.76 Q1306.78 1068.44 1306.78 1071.3 Q1306.78 1075.66 1308.12 1077.86 Q1309.46 1080.02 1312.13 1080.02 Q1314.17 1080.02 1315.35 1078.46 Q1316.49 1076.9 1317.54 1072.19 L1317.99 1070.19 Q1319.33 1063.95 1321.78 1061.34 Q1324.19 1058.7 1328.56 1058.7 Q1333.52 1058.7 1336.42 1062.64 Q1339.31 1066.56 1339.31 1073.43 Q1339.31 1076.3 1338.74 1079.42 Q1338.2 1082.5 1337.09 1085.94 L1331.04 1085.94 Q1332.72 1082.69 1333.58 1079.54 Q1334.41 1076.39 1334.41 1073.31 Q1334.41 1069.17 1333.01 1066.94 Q1331.58 1064.71 1329 1064.71 Q1326.61 1064.71 1325.34 1066.34 Q1324.07 1067.93 1322.89 1073.37 L1322.41 1075.41 Q1321.27 1080.85 1318.91 1083.27 Q1316.52 1085.69 1312.39 1085.69 Q1307.36 1085.69 1304.62 1082.12 Q1301.88 1078.56 1301.88 1072 Q1301.88 1068.75 1302.36 1065.89 Q1302.84 1063.02 1303.79 1060.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1292.62 1043.58 L1302.74 1043.58 L1302.74 1031.51 L1307.29 1031.51 L1307.29 1043.58 L1326.65 1043.58 Q1331.01 1043.58 1332.25 1042.4 Q1333.49 1041.19 1333.49 1037.53 L1333.49 1031.51 L1338.39 1031.51 L1338.39 1037.53 Q1338.39 1044.31 1335.88 1046.89 Q1333.33 1049.47 1326.65 1049.47 L1307.29 1049.47 L1307.29 1053.76 L1302.74 1053.76 L1302.74 1049.47 L1292.62 1049.47 L1292.62 1043.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip155)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1474.13,863.544 1474.98,863.545 1475.83,863.546 1476.69,863.547 1477.54,863.547 1478.4,863.548 1479.25,863.549 1480.1,863.55 1480.96,863.551 1481.81,863.552 1482.67,863.553 1483.52,863.554 1484.37,863.555 1485.23,863.556 1486.08,863.557 1486.93,863.558 1487.79,863.559 1488.64,863.561 1489.5,863.562 1490.35,863.563 1491.2,863.565 1492.06,863.566 1492.91,863.567 1493.77,863.569 1494.62,863.57 1495.47,863.572 1496.33,863.574 1497.18,863.575 1498.04,863.577 1498.89,863.579 1499.74,863.581 1500.6,863.583 1501.45,863.585 1502.3,863.587 1503.16,863.589 1504.01,863.591 1504.87,863.594 1505.72,863.596 1506.57,863.599 1507.43,863.602 1508.28,863.604 1509.14,863.607 1509.99,863.61 1510.84,863.613 1511.7,863.617 1512.55,863.62 1513.41,863.624 1514.26,863.628 1515.11,863.632 1515.97,863.636 1516.82,863.64 1517.67,863.645 1518.53,863.65 1519.38,863.655 1520.24,863.66 1521.09,863.666 1521.94,863.671 1522.8,863.678 1523.65,863.684 1524.51,863.691 1525.36,863.698 1526.21,863.706 1527.07,863.714 1527.92,863.723 1528.78,863.732 1529.63,863.742 1530.48,863.753 1531.34,863.764 1532.19,863.776 1533.04,863.789 1533.9,863.803 1534.75,863.817 1535.61,863.833 1536.46,863.85 1537.31,863.869 1538.17,863.888 1539.02,863.91 1539.88,863.933 1540.73,863.959 1541.58,863.986 1542.44,864.017 1543.29,864.05 1544.15,864.086 1545,864.127 1545.85,864.171 1546.71,864.221 1547.56,864.276 1548.41,864.337 1549.27,864.407 1550.12,864.485 1550.98,864.574 1551.83,864.675 1552.68,864.792 1553.54,864.926 1554.39,865.083 1555.25,865.267 1556.1,865.486 1556.95,865.748 1557.81,866.065 1558.66,866.454 1559.52,866.939 1560.37,867.556 1561.22,868.354 1562.08,869.416 1562.93,870.871 1563.78,872.937 1564.64,876.006 1565.49,880.805 1566.35,888.692 1567.2,901.827 1568.05,921.785 1568.91,946.426 1569.76,971.216 1570.62,993.263 1571.47,1011.72 1572.32,1026.73 1573.18,1038.76 1574.03,1048.31 1574.89,1055.86 1575.74,1061.82 1576.59,1066.5 1577.45,1070.18 1578.3,1073.06 1579.16,1075.32 1580.01,1077.1 1580.86,1078.48 1581.72,1079.56 1582.57,1080.4 1583.42,1081.05 1584.28,1081.55 1585.13,1081.94 1585.99,1082.23 1586.84,1082.45 1587.69,1082.61 1588.55,1082.72 1589.4,1082.8 1590.26,1082.84 1591.11,1082.87 1591.96,1082.87 1592.82,1082.86 1593.67,1082.84 1594.53,1082.8 1595.38,1082.76 1596.23,1082.71 1597.09,1082.65 1597.94,1082.59 1598.79,1082.52 1599.65,1082.45 1600.5,1082.38 1601.36,1082.3 1602.21,1082.22 1603.06,1082.14 1603.92,1082.05 1604.77,1081.96 1605.63,1081.87 1606.48,1081.77 1607.33,1081.68 1608.19,1081.58 1609.04,1081.47 1609.9,1081.37 1610.75,1081.26 1611.6,1081.15 1612.46,1081.04 1613.31,1080.92 1614.16,1080.81 1615.02,1080.69 1615.87,1080.57 1616.73,1080.44 1617.58,1080.31 1618.43,1080.18 1619.29,1080.05 1620.14,1079.92 1621,1079.78 1621.85,1079.64 1622.7,1079.5 1623.56,1079.35 1624.41,1079.2 1625.27,1079.05 1626.12,1078.9 1626.97,1078.74 1627.83,1078.59 1628.68,1078.42 1629.53,1078.26 1630.39,1078.09 1631.24,1077.92 1632.1,1077.75 1632.95,1077.57 1633.8,1077.39 1634.66,1077.21 1635.51,1077.03 1636.37,1076.84 1637.22,1076.65 1638.07,1076.45 1638.93,1076.26 1639.78,1076.06 1640.64,1075.85 1641.49,1075.65 1642.34,1075.44 1643.2,1075.22 1644.05,1075.01 1644.9,1074.79 1645.76,1074.56 1646.61,1074.34 1647.47,1074.11 1648.32,1073.87 1649.17,1073.64 1650.03,1073.4 1650.88,1073.15 1651.74,1072.9 1652.59,1072.65 1653.44,1072.4 1654.3,1072.14 1655.15,1071.88 1656.01,1071.61 1656.86,1071.34 1657.71,1071.07 1658.57,1070.79 1659.42,1070.51 1660.27,1070.23 1661.13,1069.94 1661.98,1069.64 1662.84,1069.35 1663.69,1069.05 1664.54,1068.74 1665.4,1068.43 1666.25,1068.12 1667.11,1067.8 1667.96,1067.48 1668.81,1067.15 1669.67,1066.82 1670.52,1066.49 1671.38,1066.15 1672.23,1065.8 1673.08,1065.45 1673.94,1065.1 1674.79,1064.74 1675.64,1064.38 1676.5,1064.01 1677.35,1063.64 1678.21,1063.26 1679.06,1062.88 1679.91,1062.5 1680.77,1062.1 1681.62,1061.71 1682.48,1061.31 1683.33,1060.9 1684.18,1060.49 1685.04,1060.07 1685.89,1059.65 1686.75,1059.22 1687.6,1058.79 1688.45,1058.35 1689.31,1057.9 1690.16,1057.45 1691.01,1057 1691.87,1056.54 1692.72,1056.07 1693.58,1055.6 1694.43,1055.12 1695.28,1054.63 1696.14,1054.14 1696.99,1053.64 1697.85,1053.14 1698.7,1052.63 1699.55,1052.12 1700.41,1051.6 1701.26,1051.07 1702.12,1050.53 1702.97,1049.99 1703.82,1049.44 1704.68,1048.89 1705.53,1048.33 1706.38,1047.76 1707.24,1047.18 1708.09,1046.6 1708.95,1046.01 1709.8,1045.42 1710.65,1044.81 1711.51,1044.2 1712.36,1043.58 1713.22,1042.96 1714.07,1042.32 1714.92,1041.68 1715.78,1041.03 1716.63,1040.38 1717.49,1039.71 1718.34,1039.04 1719.19,1038.36 1720.05,1037.67 1720.9,1036.97 1721.76,1036.27 1722.61,1035.55 1723.46,1034.83 1724.32,1034.1 1725.17,1033.36 1726.02,1032.61 1726.88,1031.86 1727.73,1031.09 1728.59,1030.31 1729.44,1029.53 1730.29,1028.74 1731.15,1027.94 1732,1027.12 1732.86,1026.3 1733.71,1025.47 1734.56,1024.64 1735.42,1023.79 1736.27,1022.93 1737.13,1022.06 1737.98,1021.19 1738.83,1020.3 1739.69,1019.41 1740.54,1018.51 1741.39,1017.6 1742.25,1016.68 1743.1,1015.75 1743.96,1014.82 1744.81,1013.87 1745.66,1012.93 1746.52,1011.97 1747.37,1011.01 1748.23,1010.04 1749.08,1009.07 1749.93,1008.09 1750.79,1007.11 1751.64,1006.13 1752.5,1005.15 1753.35,1004.17 1754.2,1003.19 1755.06,1002.22 1755.91,1001.25 1756.76,1000.29 1757.62,999.347 1758.47,998.414 1759.33,997.499 1760.18,996.604 1761.03,995.734 1761.89,994.894 1762.74,994.088 1763.6,993.322 1764.45,992.602 1765.3,991.935 1766.16,991.329 1767.01,990.792 1767.87,990.333 1768.72,989.964 1769.57,989.696 1770.43,989.542 1771.28,989.517 1772.13,989.636 1772.99,989.919 1773.84,990.384 1774.7,991.055 1775.55,991.957 1776.4,993.118 1777.26,994.57 1778.11,996.347 1778.97,998.489 1779.82,1001.04 1780.67,1004.05 1781.53,1007.57 1782.38,1011.67 1783.24,1016.41 1784.09,1021.87 1784.94,1028.13 1785.8,1035.28 1786.65,1043.42 1787.5,1052.67 1788.36,1063.14 1789.21,1074.95 1790.07,1088.25 1790.92,1103.17 1791.77,1119.85 1792.63,1138.45 1793.48,1159.09 1794.34,1181.88 1795.19,1206.86 1796.04,1233.99 1796.9,1263 1797.75,1293.2 1798.61,1323.12 1799.46,1349.95 1800.31,1370.16 1801.17,1382.76 1802.02,1390.02 1802.87,1394.36 1803.73,1397.12 1804.58,1399 1805.44,1400.35 1806.29,1401.35 1807.14,1402.12 1808,1402.73 1808.85,1403.22 1809.71,1403.62 1810.56,1403.96 1811.41,1404.24 1812.27,1404.48 1813.12,1404.69 1813.98,1404.87 1814.83,1405.03 1815.68,1405.16 1816.54,1405.29 1817.39,1405.39 1818.24,1405.49 1819.1,1405.58 1819.95,1405.66 1820.81,1405.73 1821.66,1405.79 1822.51,1405.85 1823.37,1405.91 1824.22,1405.96 1825.08,1406 1825.93,1406.04 1826.78,1406.08 1827.64,1406.12 1828.49,1406.15 1829.35,1406.18 1830.2,1406.21 1831.05,1406.24 1831.91,1406.27 1832.76,1406.29 1833.61,1406.31 1834.47,1406.33 1835.32,1406.35 1836.18,1406.37 1837.03,1406.39 1837.88,1406.4 1838.74,1406.42 1839.59,1406.43 1840.45,1406.45 1841.3,1406.46 1842.15,1406.47 1843.01,1406.49 1843.86,1406.5 1844.72,1406.51 1845.57,1406.52 1846.42,1406.53 1847.28,1406.54 1848.13,1406.55 1848.99,1406.56 1849.84,1406.56 1850.69,1406.57 1851.55,1406.58 1852.4,1406.59 1853.25,1406.59 1854.11,1406.6 1854.96,1406.61 1855.82,1406.61 1856.67,1406.62 1857.52,1406.62 1858.38,1406.63 1859.23,1406.64 1860.09,1406.64 1860.94,1406.65 1861.79,1406.65 1862.65,1406.65 1863.5,1406.66 1864.36,1406.66 1865.21,1406.67 1866.06,1406.67 1866.92,1406.68 1867.77,1406.68 1868.62,1406.68 1869.48,1406.69 1870.33,1406.69 1871.19,1406.69 1872.04,1406.7 1872.89,1406.7 1873.75,1406.7 1874.6,1406.71 1875.46,1406.71 1876.31,1406.71 1877.16,1406.71 1878.02,1406.72 1878.87,1406.72 1879.73,1406.72 1880.58,1406.72 1881.43,1406.73 1882.29,1406.73 1883.14,1406.73 1883.99,1406.73 1884.85,1406.73 1885.7,1406.74 1886.56,1406.74 1887.41,1406.74 1888.26,1406.74 1889.12,1406.74 1889.97,1406.75 1890.83,1406.75 1891.68,1406.75 1892.53,1406.75 1893.39,1406.75 1894.24,1406.75 1895.1,1406.76 1895.95,1406.76 1896.8,1406.76 1897.66,1406.76 1898.51,1406.76 1899.36,1406.76 1900.22,1406.76 1901.07,1406.77 1901.93,1406.77 1902.78,1406.77 1903.63,1406.77 1904.49,1406.77 1905.34,1406.77 1906.2,1406.77 1907.05,1406.77 1907.9,1406.78 1908.76,1406.78 1909.61,1406.78 1910.47,1406.78 1911.32,1406.78 1912.17,1406.78 1913.03,1406.78 1913.88,1406.78 1914.73,1406.78 1915.59,1406.78 1916.44,1406.79 1917.3,1406.79 1918.15,1406.79 1919,1406.79 1919.86,1406.79 1920.71,1406.79 1921.57,1406.79 1922.42,1406.79 1923.27,1406.79 1924.13,1406.79 1924.98,1406.79 1925.84,1406.79 1926.69,1406.8 1927.54,1406.8 1928.4,1406.8 1929.25,1406.8 1930.1,1406.8 1930.96,1406.8 1931.81,1406.8 1932.67,1406.8 1933.52,1406.8 1934.37,1406.8 1935.23,1406.8 1936.08,1406.8 1936.94,1406.8 1937.79,1406.8 1938.64,1406.8 1939.5,1406.81 1940.35,1406.81 1941.21,1406.81 1942.06,1406.81 1942.91,1406.81 1943.77,1406.81 1944.62,1406.81 1945.47,1406.81 1946.33,1406.81 1947.18,1406.81 1948.04,1406.81 1948.89,1406.81 1949.74,1406.81 1950.6,1406.81 1951.45,1406.81 1952.31,1406.81 1953.16,1406.81 1954.01,1406.81 1954.87,1406.81 1955.72,1406.81 1956.58,1406.82 1957.43,1406.82 1958.28,1406.82 1959.14,1406.82 1959.99,1406.82 1960.84,1406.82 1961.7,1406.82 1962.55,1406.82 1963.41,1406.82 1964.26,1406.82 1965.11,1406.82 1965.97,1406.82 1966.82,1406.82 1967.68,1406.82 1968.53,1406.82 1969.38,1406.82 1970.24,1406.82 1971.09,1406.82 1971.95,1406.82 1972.8,1406.82 1973.65,1406.82 1974.51,1406.82 1975.36,1406.82 1976.21,1406.82 1977.07,1406.82 1977.92,1406.82 1978.78,1406.82 1979.63,1406.82 1980.48,1406.83 1981.34,1406.83 1982.19,1406.83 1983.05,1406.83 1983.9,1406.83 1984.75,1406.83 1985.61,1406.83 1986.46,1406.83 1987.32,1406.83 1988.17,1406.83 1989.02,1406.83 1989.88,1406.83 1990.73,1406.83 1991.59,1406.83 1992.44,1406.83 1993.29,1406.83 1994.15,1406.83 1995,1406.83 1995.85,1406.83 1996.71,1406.83 1997.56,1406.83 1998.42,1406.83 1999.27,1406.83 2000.12,1406.83 2000.98,1406.83 2001.83,1406.83 2002.69,1406.83 2003.54,1406.83 2004.39,1406.83 2005.25,1406.83 2006.1,1406.83 2006.96,1406.83 2007.81,1406.83 2008.66,1406.83 2009.52,1406.83 2010.37,1406.83 2011.22,1406.83 2012.08,1406.83 2012.93,1406.83 2013.79,1406.83 2014.64,1406.83 2015.49,1406.83 2016.35,1406.83 2017.2,1406.84 2018.06,1406.84 2018.91,1406.84 2019.76,1406.84 2020.62,1406.84 2021.47,1406.84 2022.33,1406.84 2023.18,1406.84 2024.03,1406.84 2024.89,1406.84 2025.74,1406.84 2026.59,1406.84 2027.45,1406.84 2028.3,1406.84 2029.16,1406.84 2030.01,1406.84 2030.86,1406.84 2031.72,1406.84 2032.57,1406.84 2033.43,1406.84 2034.28,1406.84 2035.13,1406.84 2035.99,1406.84 2036.84,1406.84 2037.7,1406.84 2038.55,1406.84 2039.4,1406.84 2040.26,1406.84 2041.11,1406.84 2041.96,1406.84 2042.82,1406.84 2043.67,1406.84 2044.53,1406.84 2045.38,1406.84 2046.23,1406.84 2047.09,1406.84 2047.94,1406.84 2048.8,1406.84 2049.65,1406.84 2050.5,1406.84 2051.36,1406.84 2052.21,1406.84 2053.07,1406.84 2053.92,1406.84 2054.77,1406.84 2055.63,1406.84 2056.48,1406.84 2057.33,1406.84 2058.19,1406.84 2059.04,1406.84 2059.9,1406.84 2060.75,1406.84 2061.6,1406.84 2062.46,1406.84 2063.31,1406.84 2064.17,1406.84 2065.02,1406.84 2065.87,1406.84 2066.73,1406.84 2067.58,1406.84 2068.44,1406.84 2069.29,1406.84 2070.14,1406.84 2071,1406.84 2071.85,1406.84 2072.7,1406.84 2073.56,1406.84 2074.41,1406.84 2075.27,1406.84 2076.12,1406.84 2076.97,1406.84 2077.83,1406.84 2078.68,1406.84 2079.54,1406.84 2080.39,1406.84 2081.24,1406.84 2082.1,1406.84 2082.95,1406.84 2083.81,1406.85 2084.66,1406.85 2085.51,1406.85 2086.37,1406.85 2087.22,1406.85 2088.07,1406.85 2088.93,1406.85 2089.78,1406.85 2090.64,1406.85 2091.49,1406.85 2092.34,1406.85 2093.2,1406.85 2094.05,1406.85 2094.91,1406.85 2095.76,1406.85 2096.61,1406.85 2097.47,1406.85 2098.32,1406.85 2099.18,1406.85 2100.03,1406.85 2100.88,1406.85 2101.74,1406.85 2102.59,1406.85 2103.44,1406.85 2104.3,1406.85 2105.15,1406.85 2106.01,1406.85 2106.86,1406.85 2107.71,1406.85 2108.57,1406.85 2109.42,1406.85 2110.28,1406.85 2111.13,1406.85 2111.98,1406.85 2112.84,1406.85 2113.69,1406.85 2114.55,1406.85 2115.4,1406.85 2116.25,1406.85 2117.11,1406.85 2117.96,1406.85 2118.82,1406.85 2119.67,1406.85 2120.52,1406.85 2121.38,1406.85 2122.23,1406.85 2123.08,1406.85 2123.94,1406.85 2124.79,1406.85 2125.65,1406.85 2126.5,1406.85 2127.35,1406.85 2128.21,1406.85 2129.06,1406.85 2129.92,1406.85 2130.77,1406.85 2131.62,1406.85 2132.48,1406.85 2133.33,1406.85 2134.19,1406.85 2135.04,1406.85 2135.89,1406.85 2136.75,1406.85 2137.6,1406.85 2138.45,1406.85 2139.31,1406.85 2140.16,1406.85 2141.02,1406.85 2141.87,1406.85 2142.72,1406.85 2143.58,1406.85 2144.43,1406.85 2145.29,1406.85 2146.14,1406.85 2146.99,1406.85 2147.85,1406.85 2148.7,1406.85 2149.56,1406.85 2150.41,1406.85 2151.26,1406.85 2152.12,1406.85 2152.97,1406.85 2153.82,1406.85 2154.68,1406.85 2155.53,1406.85 2156.39,1406.85 2157.24,1406.85 2158.09,1406.85 2158.95,1406.85 2159.8,1406.85 2160.66,1406.85 2161.51,1406.85 2162.36,1406.85 2163.22,1406.85 2164.07,1406.85 2164.93,1406.85 2165.78,1406.85 2166.63,1406.85 2167.49,1406.85 2168.34,1406.85 2169.19,1406.85 2170.05,1406.85 2170.9,1406.85 2171.76,1406.85 2172.61,1406.85 2173.46,1406.85 2174.32,1406.85 2175.17,1406.85 2176.03,1406.85 2176.88,1406.85 2177.73,1406.85 2178.59,1406.85 2179.44,1406.85 2180.3,1406.85 2181.15,1406.85 2182,1406.85 2182.86,1406.85 2183.71,1406.85 2184.56,1406.85 2185.42,1406.85 2186.27,1406.85 2187.13,1406.85 2187.98,1406.85 2188.83,1406.85 2189.69,1406.85 2190.54,1406.85 2191.4,1406.85 2192.25,1406.85 2193.1,1406.85 2193.96,1406.85 2194.81,1406.85 2195.67,1406.85 2196.52,1406.85 2197.37,1406.85 2198.23,1406.85 2199.08,1406.85 2199.93,1406.85 2200.79,1406.85 2201.64,1406.85 2202.5,1406.85 2203.35,1406.85 2204.2,1406.85 2205.06,1406.85 2205.91,1406.85 2206.77,1406.85 2207.62,1406.85 2208.47,1406.85 2209.33,1406.85 2210.18,1406.85 2211.04,1406.85 2211.89,1406.85 2212.74,1406.85 2213.6,1406.85 2214.45,1406.85 2215.3,1406.85 2216.16,1406.85 2217.01,1406.85 2217.87,1406.85 2218.72,1406.85 2219.57,1406.85 2220.43,1406.85 2221.28,1406.85 2222.14,1406.85 2222.99,1406.85 2223.84,1406.85 2224.7,1406.85 2225.55,1406.85 2226.41,1406.85 2227.26,1406.85 2228.11,1406.85 2228.97,1406.85 2229.82,1406.85 2230.67,1406.85 2231.53,1406.85 2232.38,1406.85 2233.24,1406.85 2234.09,1406.85 2234.94,1406.85 2235.8,1406.85 2236.65,1406.85 2237.51,1406.85 2238.36,1406.85 2239.21,1406.85 2240.07,1406.85 2240.92,1406.85 2241.78,1406.85 2242.63,1406.85 2243.48,1406.85 2244.34,1406.85 2245.19,1406.85 2246.04,1406.85 2246.9,1406.85 2247.75,1406.85 2248.61,1406.85 2249.46,1406.85 2250.31,1406.85 2251.17,1406.85 2252.02,1406.85 2252.88,1406.85 2253.73,1406.85 2254.58,1406.85 2255.44,1406.85 2256.29,1406.85 2257.15,1406.85 2258,1406.85 2258.85,1406.85 2259.71,1406.85 2260.56,1406.85 2261.42,1406.85 2262.27,1406.85 2263.12,1406.85 2263.98,1406.85 2264.83,1406.85 2265.68,1406.85 2266.54,1406.85 2267.39,1406.85 2268.25,1406.85 2269.1,1406.85 2269.95,1406.85 2270.81,1406.85 2271.66,1406.85 2272.52,1406.85 2273.37,1406.85 2274.22,1406.85 2275.08,1406.85 2275.93,1406.85 2276.79,1406.85 2277.64,1406.85 2278.49,1406.85 2279.35,1406.85 2280.2,1406.85 2281.05,1406.85 2281.91,1406.85 2282.76,1406.85 2283.62,1406.84 2284.47,1406.84 2285.32,1406.84 2286.18,1406.84 2287.03,1406.84 2287.89,1406.84 2288.74,1406.84 2289.59,1406.84 2290.45,1406.84 2291.3,1406.84 2292.16,1406.84 2293.01,1406.84 2293.86,1406.84 2294.72,1406.84 2295.57,1406.84 2296.42,1406.84 2297.28,1406.84 2298.13,1406.84 2298.99,1406.84 2299.84,1406.84 2300.69,1406.84 2301.55,1406.84 2302.4,1406.84 2303.26,1406.84 2304.11,1406.84 2304.96,1406.84 2305.82,1406.84 2306.67,1406.83 2307.53,1406.83 2308.38,1406.83 2309.23,1406.83 2310.09,1406.83 2310.94,1406.83 2311.79,1406.83 2312.65,1406.83 2313.5,1406.83 2314.36,1406.83 2315.21,1406.82 2316.06,1406.82 2316.92,1406.82 2317.77,1406.82 2318.63,1406.81 2319.48,1406.81 2320.33,1406.81 2321.19,1406.8 2322.04,1406.8 2322.9,1406.79 2323.75,1406.78 2324.6,1406.76 2325.46,1406.73 2326.31,1406.63 2327.16,1406.88 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1473.27,1423.18 1473.27,847.244 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1686.75,1423.18 1686.75,847.244 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1900.22,1423.18 1900.22,847.244 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2113.69,1423.18 2113.69,847.244 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2327.16,1423.18 2327.16,847.244 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1423.18 2352.76,1423.18 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1473.27,1423.18 1473.27,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1686.75,1423.18 1686.75,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1900.22,1423.18 1900.22,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2113.69,1423.18 2113.69,1404.28 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2327.16,1423.18 2327.16,1404.28 "/>
+<path clip-path="url(#clip970)" d="M1435.58 1454.1 Q1431.96 1454.1 1430.14 1457.66 Q1428.33 1461.2 1428.33 1468.33 Q1428.33 1475.44 1430.14 1479.01 Q1431.96 1482.55 1435.58 1482.55 Q1439.21 1482.55 1441.02 1479.01 Q1442.84 1475.44 1442.84 1468.33 Q1442.84 1461.2 1441.02 1457.66 Q1439.21 1454.1 1435.58 1454.1 M1435.58 1450.39 Q1441.39 1450.39 1444.44 1455 Q1447.52 1459.58 1447.52 1468.33 Q1447.52 1477.06 1444.44 1481.67 Q1441.39 1486.25 1435.58 1486.25 Q1429.77 1486.25 1426.69 1481.67 Q1423.63 1477.06 1423.63 1468.33 Q1423.63 1459.58 1426.69 1455 Q1429.77 1450.39 1435.58 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1455.74 1479.7 L1460.62 1479.7 L1460.62 1485.58 L1455.74 1485.58 L1455.74 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1480.81 1454.1 Q1477.2 1454.1 1475.37 1457.66 Q1473.56 1461.2 1473.56 1468.33 Q1473.56 1475.44 1475.37 1479.01 Q1477.2 1482.55 1480.81 1482.55 Q1484.44 1482.55 1486.25 1479.01 Q1488.08 1475.44 1488.08 1468.33 Q1488.08 1461.2 1486.25 1457.66 Q1484.44 1454.1 1480.81 1454.1 M1480.81 1450.39 Q1486.62 1450.39 1489.67 1455 Q1492.75 1459.58 1492.75 1468.33 Q1492.75 1477.06 1489.67 1481.67 Q1486.62 1486.25 1480.81 1486.25 Q1475 1486.25 1471.92 1481.67 Q1468.86 1477.06 1468.86 1468.33 Q1468.86 1459.58 1471.92 1455 Q1475 1450.39 1480.81 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1510.97 1454.1 Q1507.36 1454.1 1505.53 1457.66 Q1503.72 1461.2 1503.72 1468.33 Q1503.72 1475.44 1505.53 1479.01 Q1507.36 1482.55 1510.97 1482.55 Q1514.6 1482.55 1516.41 1479.01 Q1518.24 1475.44 1518.24 1468.33 Q1518.24 1461.2 1516.41 1457.66 Q1514.6 1454.1 1510.97 1454.1 M1510.97 1450.39 Q1516.78 1450.39 1519.83 1455 Q1522.91 1459.58 1522.91 1468.33 Q1522.91 1477.06 1519.83 1481.67 Q1516.78 1486.25 1510.97 1486.25 Q1505.16 1486.25 1502.08 1481.67 Q1499.02 1477.06 1499.02 1468.33 Q1499.02 1459.58 1502.08 1455 Q1505.16 1450.39 1510.97 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1649.55 1454.1 Q1645.94 1454.1 1644.11 1457.66 Q1642.3 1461.2 1642.3 1468.33 Q1642.3 1475.44 1644.11 1479.01 Q1645.94 1482.55 1649.55 1482.55 Q1653.18 1482.55 1654.99 1479.01 Q1656.82 1475.44 1656.82 1468.33 Q1656.82 1461.2 1654.99 1457.66 Q1653.18 1454.1 1649.55 1454.1 M1649.55 1450.39 Q1655.36 1450.39 1658.41 1455 Q1661.49 1459.58 1661.49 1468.33 Q1661.49 1477.06 1658.41 1481.67 Q1655.36 1486.25 1649.55 1486.25 Q1643.74 1486.25 1640.66 1481.67 Q1637.6 1477.06 1637.6 1468.33 Q1637.6 1459.58 1640.66 1455 Q1643.74 1450.39 1649.55 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1669.71 1479.7 L1674.59 1479.7 L1674.59 1485.58 L1669.71 1485.58 L1669.71 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1694.78 1454.1 Q1691.17 1454.1 1689.34 1457.66 Q1687.53 1461.2 1687.53 1468.33 Q1687.53 1475.44 1689.34 1479.01 Q1691.17 1482.55 1694.78 1482.55 Q1698.41 1482.55 1700.22 1479.01 Q1702.05 1475.44 1702.05 1468.33 Q1702.05 1461.2 1700.22 1457.66 Q1698.41 1454.1 1694.78 1454.1 M1694.78 1450.39 Q1700.59 1450.39 1703.64 1455 Q1706.72 1459.58 1706.72 1468.33 Q1706.72 1477.06 1703.64 1481.67 Q1700.59 1486.25 1694.78 1486.25 Q1688.97 1486.25 1685.89 1481.67 Q1682.83 1477.06 1682.83 1468.33 Q1682.83 1459.58 1685.89 1455 Q1688.97 1450.39 1694.78 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1714.99 1451.02 L1733.34 1451.02 L1733.34 1454.96 L1719.27 1454.96 L1719.27 1463.43 Q1720.29 1463.08 1721.31 1462.92 Q1722.32 1462.73 1723.34 1462.73 Q1729.13 1462.73 1732.51 1465.9 Q1735.89 1469.08 1735.89 1474.49 Q1735.89 1480.07 1732.42 1483.17 Q1728.94 1486.25 1722.62 1486.25 Q1720.45 1486.25 1718.18 1485.88 Q1715.94 1485.51 1713.53 1484.77 L1713.53 1480.07 Q1715.61 1481.2 1717.83 1481.76 Q1720.06 1482.32 1722.53 1482.32 Q1726.54 1482.32 1728.87 1480.21 Q1731.21 1478.1 1731.21 1474.49 Q1731.21 1470.88 1728.87 1468.77 Q1726.54 1466.67 1722.53 1466.67 Q1720.66 1466.67 1718.78 1467.08 Q1716.93 1467.5 1714.99 1468.38 L1714.99 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1862.52 1454.1 Q1858.91 1454.1 1857.08 1457.66 Q1855.28 1461.2 1855.28 1468.33 Q1855.28 1475.44 1857.08 1479.01 Q1858.91 1482.55 1862.52 1482.55 Q1866.16 1482.55 1867.96 1479.01 Q1869.79 1475.44 1869.79 1468.33 Q1869.79 1461.2 1867.96 1457.66 Q1866.16 1454.1 1862.52 1454.1 M1862.52 1450.39 Q1868.33 1450.39 1871.39 1455 Q1874.47 1459.58 1874.47 1468.33 Q1874.47 1477.06 1871.39 1481.67 Q1868.33 1486.25 1862.52 1486.25 Q1856.71 1486.25 1853.63 1481.67 Q1850.58 1477.06 1850.58 1468.33 Q1850.58 1459.58 1853.63 1455 Q1856.71 1450.39 1862.52 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1882.68 1479.7 L1887.57 1479.7 L1887.57 1485.58 L1882.68 1485.58 L1882.68 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1898.56 1481.64 L1906.2 1481.64 L1906.2 1455.28 L1897.89 1456.95 L1897.89 1452.69 L1906.16 1451.02 L1910.83 1451.02 L1910.83 1481.64 L1918.47 1481.64 L1918.47 1485.58 L1898.56 1485.58 L1898.56 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1937.92 1454.1 Q1934.3 1454.1 1932.48 1457.66 Q1930.67 1461.2 1930.67 1468.33 Q1930.67 1475.44 1932.48 1479.01 Q1934.3 1482.55 1937.92 1482.55 Q1941.55 1482.55 1943.35 1479.01 Q1945.18 1475.44 1945.18 1468.33 Q1945.18 1461.2 1943.35 1457.66 Q1941.55 1454.1 1937.92 1454.1 M1937.92 1450.39 Q1943.73 1450.39 1946.78 1455 Q1949.86 1459.58 1949.86 1468.33 Q1949.86 1477.06 1946.78 1481.67 Q1943.73 1486.25 1937.92 1486.25 Q1932.1 1486.25 1929.03 1481.67 Q1925.97 1477.06 1925.97 1468.33 Q1925.97 1459.58 1929.03 1455 Q1932.1 1450.39 1937.92 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2076.49 1454.1 Q2072.88 1454.1 2071.05 1457.66 Q2069.25 1461.2 2069.25 1468.33 Q2069.25 1475.44 2071.05 1479.01 Q2072.88 1482.55 2076.49 1482.55 Q2080.13 1482.55 2081.93 1479.01 Q2083.76 1475.44 2083.76 1468.33 Q2083.76 1461.2 2081.93 1457.66 Q2080.13 1454.1 2076.49 1454.1 M2076.49 1450.39 Q2082.3 1450.39 2085.36 1455 Q2088.44 1459.58 2088.44 1468.33 Q2088.44 1477.06 2085.36 1481.67 Q2082.3 1486.25 2076.49 1486.25 Q2070.68 1486.25 2067.6 1481.67 Q2064.55 1477.06 2064.55 1468.33 Q2064.55 1459.58 2067.6 1455 Q2070.68 1450.39 2076.49 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2096.65 1479.7 L2101.54 1479.7 L2101.54 1485.58 L2096.65 1485.58 L2096.65 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2112.53 1481.64 L2120.17 1481.64 L2120.17 1455.28 L2111.86 1456.95 L2111.86 1452.69 L2120.13 1451.02 L2124.8 1451.02 L2124.8 1481.64 L2132.44 1481.64 L2132.44 1485.58 L2112.53 1485.58 L2112.53 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2141.93 1451.02 L2160.29 1451.02 L2160.29 1454.96 L2146.21 1454.96 L2146.21 1463.43 Q2147.23 1463.08 2148.25 1462.92 Q2149.27 1462.73 2150.29 1462.73 Q2156.08 1462.73 2159.46 1465.9 Q2162.83 1469.08 2162.83 1474.49 Q2162.83 1480.07 2159.36 1483.17 Q2155.89 1486.25 2149.57 1486.25 Q2147.4 1486.25 2145.13 1485.88 Q2142.88 1485.51 2140.47 1484.77 L2140.47 1480.07 Q2142.56 1481.2 2144.78 1481.76 Q2147 1482.32 2149.48 1482.32 Q2153.48 1482.32 2155.82 1480.21 Q2158.16 1478.1 2158.16 1474.49 Q2158.16 1470.88 2155.82 1468.77 Q2153.48 1466.67 2149.48 1466.67 Q2147.6 1466.67 2145.73 1467.08 Q2143.88 1467.5 2141.93 1468.38 L2141.93 1451.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2289.47 1454.1 Q2285.86 1454.1 2284.03 1457.66 Q2282.22 1461.2 2282.22 1468.33 Q2282.22 1475.44 2284.03 1479.01 Q2285.86 1482.55 2289.47 1482.55 Q2293.1 1482.55 2294.91 1479.01 Q2296.74 1475.44 2296.74 1468.33 Q2296.74 1461.2 2294.91 1457.66 Q2293.1 1454.1 2289.47 1454.1 M2289.47 1450.39 Q2295.28 1450.39 2298.33 1455 Q2301.41 1459.58 2301.41 1468.33 Q2301.41 1477.06 2298.33 1481.67 Q2295.28 1486.25 2289.47 1486.25 Q2283.66 1486.25 2280.58 1481.67 Q2277.52 1477.06 2277.52 1468.33 Q2277.52 1459.58 2280.58 1455 Q2283.66 1450.39 2289.47 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2309.63 1479.7 L2314.51 1479.7 L2314.51 1485.58 L2309.63 1485.58 L2309.63 1479.7 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2328.73 1481.64 L2345.05 1481.64 L2345.05 1485.58 L2323.1 1485.58 L2323.1 1481.64 Q2325.76 1478.89 2330.35 1474.26 Q2334.95 1469.61 2336.13 1468.27 Q2338.38 1465.74 2339.26 1464.01 Q2340.16 1462.25 2340.16 1460.56 Q2340.16 1457.8 2338.22 1456.07 Q2336.3 1454.33 2333.19 1454.33 Q2331 1454.33 2328.54 1455.09 Q2326.11 1455.86 2323.33 1457.41 L2323.33 1452.69 Q2326.16 1451.55 2328.61 1450.97 Q2331.07 1450.39 2333.1 1450.39 Q2338.47 1450.39 2341.67 1453.08 Q2344.86 1455.77 2344.86 1460.26 Q2344.86 1462.39 2344.05 1464.31 Q2343.26 1466.2 2341.16 1468.8 Q2340.58 1469.47 2337.48 1472.69 Q2334.38 1475.88 2328.73 1481.64 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2364.86 1454.1 Q2361.25 1454.1 2359.42 1457.66 Q2357.62 1461.2 2357.62 1468.33 Q2357.62 1475.44 2359.42 1479.01 Q2361.25 1482.55 2364.86 1482.55 Q2368.5 1482.55 2370.3 1479.01 Q2372.13 1475.44 2372.13 1468.33 Q2372.13 1461.2 2370.3 1457.66 Q2368.5 1454.1 2364.86 1454.1 M2364.86 1450.39 Q2370.67 1450.39 2373.73 1455 Q2376.81 1459.58 2376.81 1468.33 Q2376.81 1477.06 2373.73 1481.67 Q2370.67 1486.25 2364.86 1486.25 Q2359.05 1486.25 2355.97 1481.67 Q2352.92 1477.06 2352.92 1468.33 Q2352.92 1459.58 2355.97 1455 Q2359.05 1450.39 2364.86 1450.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1770.82 1520.52 L1811.02 1520.52 L1811.02 1525.93 L1794.15 1525.93 L1794.15 1568.04 L1787.69 1568.04 L1787.69 1525.93 L1770.82 1525.93 L1770.82 1520.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1814.96 1532.4 L1820.82 1532.4 L1820.82 1568.04 L1814.96 1568.04 L1814.96 1532.4 M1814.96 1518.52 L1820.82 1518.52 L1820.82 1525.93 L1814.96 1525.93 L1814.96 1518.52 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1860.83 1539.24 Q1863.02 1535.29 1866.08 1533.41 Q1869.14 1531.54 1873.27 1531.54 Q1878.84 1531.54 1881.87 1535.45 Q1884.89 1539.33 1884.89 1546.53 L1884.89 1568.04 L1879 1568.04 L1879 1546.72 Q1879 1541.59 1877.19 1539.11 Q1875.37 1536.63 1871.65 1536.63 Q1867.1 1536.63 1864.46 1539.65 Q1861.81 1542.68 1861.81 1547.9 L1861.81 1568.04 L1855.93 1568.04 L1855.93 1546.72 Q1855.93 1541.56 1854.11 1539.11 Q1852.3 1536.63 1848.51 1536.63 Q1844.02 1536.63 1841.38 1539.68 Q1838.74 1542.71 1838.74 1547.9 L1838.74 1568.04 L1832.85 1568.04 L1832.85 1532.4 L1838.74 1532.4 L1838.74 1537.93 Q1840.74 1534.66 1843.55 1533.1 Q1846.35 1531.54 1850.2 1531.54 Q1854.08 1531.54 1856.79 1533.51 Q1859.52 1535.48 1860.83 1539.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1927.06 1548.76 L1927.06 1551.62 L1900.14 1551.62 Q1900.52 1557.67 1903.76 1560.85 Q1907.04 1564 1912.87 1564 Q1916.24 1564 1919.39 1563.17 Q1922.58 1562.35 1925.69 1560.69 L1925.69 1566.23 Q1922.54 1567.57 1919.23 1568.27 Q1915.92 1568.97 1912.52 1568.97 Q1903.99 1568.97 1898.99 1564 Q1894.03 1559.04 1894.03 1550.57 Q1894.03 1541.82 1898.74 1536.69 Q1903.48 1531.54 1911.5 1531.54 Q1918.69 1531.54 1922.86 1536.18 Q1927.06 1540.8 1927.06 1548.76 M1921.21 1547.04 Q1921.14 1542.23 1918.5 1539.37 Q1915.89 1536.5 1911.56 1536.5 Q1906.66 1536.5 1903.7 1539.27 Q1900.77 1542.04 1900.33 1547.07 L1921.21 1547.04 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1971.46 1518.58 Q1967.2 1525.9 1965.13 1533.06 Q1963.06 1540.23 1963.06 1547.58 Q1963.06 1554.93 1965.13 1562.16 Q1967.23 1569.35 1971.46 1576.64 L1966.37 1576.64 Q1961.6 1569.16 1959.21 1561.93 Q1956.85 1554.71 1956.85 1547.58 Q1956.85 1540.48 1959.21 1533.29 Q1961.57 1526.09 1966.37 1518.58 L1971.46 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2005.55 1533.45 L2005.55 1538.98 Q2003.07 1537.71 2000.4 1537.07 Q1997.72 1536.44 1994.86 1536.44 Q1990.5 1536.44 1988.3 1537.77 Q1986.14 1539.11 1986.14 1541.79 Q1986.14 1543.82 1987.7 1545 Q1989.26 1546.15 1993.97 1547.2 L1995.97 1547.64 Q2002.21 1548.98 2004.82 1551.43 Q2007.46 1553.85 2007.46 1558.21 Q2007.46 1563.17 2003.52 1566.07 Q1999.6 1568.97 1992.73 1568.97 Q1989.86 1568.97 1986.74 1568.39 Q1983.65 1567.85 1980.22 1566.74 L1980.22 1560.69 Q1983.46 1562.38 1986.61 1563.24 Q1989.77 1564.07 1992.85 1564.07 Q1996.99 1564.07 1999.22 1562.66 Q2001.45 1561.23 2001.45 1558.65 Q2001.45 1556.27 1999.82 1554.99 Q1998.23 1553.72 1992.79 1552.54 L1990.75 1552.07 Q1985.31 1550.92 1982.89 1548.56 Q1980.47 1546.18 1980.47 1542.04 Q1980.47 1537.01 1984.04 1534.27 Q1987.6 1531.54 1994.16 1531.54 Q1997.4 1531.54 2000.27 1532.01 Q2003.13 1532.49 2005.55 1533.45 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M2015.86 1518.58 L2020.96 1518.58 Q2025.73 1526.09 2028.09 1533.29 Q2030.47 1540.48 2030.47 1547.58 Q2030.47 1554.71 2028.09 1561.93 Q2025.73 1569.16 2020.96 1576.64 L2015.86 1576.64 Q2020.1 1569.35 2022.17 1562.16 Q2024.27 1554.93 2024.27 1547.58 Q2024.27 1540.23 2022.17 1533.06 Q2020.1 1525.9 2015.86 1518.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1406.88 2352.76,1406.88 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1251.63 2352.76,1251.63 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,1096.38 2352.76,1096.38 "/>
+<polyline clip-path="url(#clip975)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1448.54,941.129 2352.76,941.129 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1423.18 1448.54,847.244 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1406.88 1467.43,1406.88 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1251.63 1467.43,1251.63 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,1096.38 1467.43,1096.38 "/>
+<polyline clip-path="url(#clip970)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1448.54,941.129 1467.43,941.129 "/>
+<path clip-path="url(#clip970)" d="M1400.59 1392.68 Q1396.98 1392.68 1395.15 1396.24 Q1393.35 1399.78 1393.35 1406.91 Q1393.35 1414.02 1395.15 1417.59 Q1396.98 1421.13 1400.59 1421.13 Q1404.22 1421.13 1406.03 1417.59 Q1407.86 1414.02 1407.86 1406.91 Q1407.86 1399.78 1406.03 1396.24 Q1404.22 1392.68 1400.59 1392.68 M1400.59 1388.97 Q1406.4 1388.97 1409.46 1393.58 Q1412.54 1398.16 1412.54 1406.91 Q1412.54 1415.64 1409.46 1420.25 Q1406.4 1424.83 1400.59 1424.83 Q1394.78 1424.83 1391.7 1420.25 Q1388.65 1415.64 1388.65 1406.91 Q1388.65 1398.16 1391.7 1393.58 Q1394.78 1388.97 1400.59 1388.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1392.63 1264.97 L1400.27 1264.97 L1400.27 1238.61 L1391.96 1240.28 L1391.96 1236.02 L1400.22 1234.35 L1404.9 1234.35 L1404.9 1264.97 L1412.54 1264.97 L1412.54 1268.91 L1392.63 1268.91 L1392.63 1264.97 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1396.22 1109.72 L1412.54 1109.72 L1412.54 1113.66 L1390.59 1113.66 L1390.59 1109.72 Q1393.25 1106.97 1397.84 1102.34 Q1402.44 1097.69 1403.62 1096.34 Q1405.87 1093.82 1406.75 1092.09 Q1407.65 1090.33 1407.65 1088.64 Q1407.65 1085.88 1405.71 1084.15 Q1403.79 1082.41 1400.68 1082.41 Q1398.48 1082.41 1396.03 1083.17 Q1393.6 1083.94 1390.82 1085.49 L1390.82 1080.77 Q1393.65 1079.63 1396.1 1079.05 Q1398.55 1078.47 1400.59 1078.47 Q1405.96 1078.47 1409.16 1081.16 Q1412.35 1083.84 1412.35 1088.34 Q1412.35 1090.47 1411.54 1092.39 Q1410.75 1094.28 1408.65 1096.88 Q1408.07 1097.55 1404.97 1100.77 Q1401.86 1103.96 1396.22 1109.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1405.41 939.775 Q1408.76 940.493 1410.64 942.761 Q1412.54 945.03 1412.54 948.363 Q1412.54 953.479 1409.02 956.28 Q1405.5 959.081 1399.02 959.081 Q1396.84 959.081 1394.53 958.641 Q1392.23 958.224 1389.78 957.368 L1389.78 952.854 Q1391.73 953.988 1394.04 954.567 Q1396.35 955.146 1398.88 955.146 Q1403.28 955.146 1405.57 953.41 Q1407.88 951.673 1407.88 948.363 Q1407.88 945.308 1405.73 943.595 Q1403.6 941.859 1399.78 941.859 L1395.75 941.859 L1395.75 938.016 L1399.97 938.016 Q1403.41 938.016 1405.24 936.65 Q1407.07 935.261 1407.07 932.669 Q1407.07 930.007 1405.17 928.595 Q1403.3 927.16 1399.78 927.16 Q1397.86 927.16 1395.66 927.576 Q1393.46 927.993 1390.82 928.873 L1390.82 924.706 Q1393.48 923.965 1395.8 923.595 Q1398.14 923.225 1400.2 923.225 Q1405.52 923.225 1408.62 925.655 Q1411.72 928.062 1411.72 932.183 Q1411.72 935.053 1410.08 937.044 Q1408.44 939.011 1405.41 939.775 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1290.87 1238.91 L1290.87 1198.71 L1296.28 1198.71 L1296.28 1215.58 L1338.39 1215.58 L1338.39 1222.04 L1296.28 1222.04 L1296.28 1238.91 L1290.87 1238.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1316.87 1163.13 L1338.39 1163.13 L1338.39 1168.98 L1317.07 1168.98 Q1312 1168.98 1309.49 1170.96 Q1306.98 1172.93 1306.98 1176.88 Q1306.98 1181.62 1310 1184.36 Q1313.02 1187.09 1318.24 1187.09 L1338.39 1187.09 L1338.39 1192.98 L1288.87 1192.98 L1288.87 1187.09 L1308.28 1187.09 Q1305.07 1184.99 1303.47 1182.16 Q1301.88 1179.29 1301.88 1175.57 Q1301.88 1169.43 1305.7 1166.28 Q1309.49 1163.13 1316.87 1163.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1308.22 1130.79 Q1307.64 1131.77 1307.39 1132.95 Q1307.1 1134.1 1307.1 1135.5 Q1307.1 1140.46 1310.35 1143.14 Q1313.56 1145.78 1319.61 1145.78 L1338.39 1145.78 L1338.39 1151.67 L1302.74 1151.67 L1302.74 1145.78 L1308.28 1145.78 Q1305.03 1143.93 1303.47 1140.97 Q1301.88 1138.01 1301.88 1133.78 Q1301.88 1133.17 1301.98 1132.44 Q1302.04 1131.71 1302.2 1130.82 L1308.22 1130.79 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1324.32 1125.25 L1302.74 1125.25 L1302.74 1119.39 L1324.1 1119.39 Q1329.16 1119.39 1331.71 1117.42 Q1334.22 1115.45 1334.22 1111.5 Q1334.22 1106.76 1331.2 1104.02 Q1328.17 1101.25 1322.95 1101.25 L1302.74 1101.25 L1302.74 1095.39 L1338.39 1095.39 L1338.39 1101.25 L1332.92 1101.25 Q1336.16 1103.38 1337.75 1106.22 Q1339.31 1109.02 1339.31 1112.74 Q1339.31 1118.88 1335.49 1122.07 Q1331.67 1125.25 1324.32 1125.25 M1301.88 1110.51 L1301.88 1110.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1303.79 1060.61 L1309.33 1060.61 Q1308.06 1063.09 1307.42 1065.76 Q1306.78 1068.44 1306.78 1071.3 Q1306.78 1075.66 1308.12 1077.86 Q1309.46 1080.02 1312.13 1080.02 Q1314.17 1080.02 1315.35 1078.46 Q1316.49 1076.9 1317.54 1072.19 L1317.99 1070.19 Q1319.33 1063.95 1321.78 1061.34 Q1324.19 1058.7 1328.56 1058.7 Q1333.52 1058.7 1336.42 1062.64 Q1339.31 1066.56 1339.31 1073.43 Q1339.31 1076.3 1338.74 1079.42 Q1338.2 1082.5 1337.09 1085.94 L1331.04 1085.94 Q1332.72 1082.69 1333.58 1079.54 Q1334.41 1076.39 1334.41 1073.31 Q1334.41 1069.17 1333.01 1066.94 Q1331.58 1064.71 1329 1064.71 Q1326.61 1064.71 1325.34 1066.34 Q1324.07 1067.93 1322.89 1073.37 L1322.41 1075.41 Q1321.27 1080.85 1318.91 1083.27 Q1316.52 1085.69 1312.39 1085.69 Q1307.36 1085.69 1304.62 1082.12 Q1301.88 1078.56 1301.88 1072 Q1301.88 1068.75 1302.36 1065.89 Q1302.84 1063.02 1303.79 1060.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip970)" d="M1292.62 1043.58 L1302.74 1043.58 L1302.74 1031.51 L1307.29 1031.51 L1307.29 1043.58 L1326.65 1043.58 Q1331.01 1043.58 1332.25 1042.4 Q1333.49 1041.19 1333.49 1037.53 L1333.49 1031.51 L1338.39 1031.51 L1338.39 1037.53 Q1338.39 1044.31 1335.88 1046.89 Q1333.33 1049.47 1326.65 1049.47 L1307.29 1049.47 L1307.29 1053.76 L1302.74 1053.76 L1302.74 1049.47 L1292.62 1049.47 L1292.62 1043.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip975)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1474.13,863.544 1474.98,863.545 1475.83,863.546 1476.69,863.547 1477.54,863.547 1478.4,863.548 1479.25,863.549 1480.1,863.55 1480.96,863.551 1481.81,863.552 1482.67,863.553 1483.52,863.554 1484.37,863.555 1485.23,863.556 1486.08,863.557 1486.93,863.558 1487.79,863.559 1488.64,863.561 1489.5,863.562 1490.35,863.563 1491.2,863.565 1492.06,863.566 1492.91,863.567 1493.77,863.569 1494.62,863.57 1495.47,863.572 1496.33,863.574 1497.18,863.575 1498.04,863.577 1498.89,863.579 1499.74,863.581 1500.6,863.583 1501.45,863.585 1502.3,863.587 1503.16,863.589 1504.01,863.591 1504.87,863.594 1505.72,863.596 1506.57,863.599 1507.43,863.602 1508.28,863.604 1509.14,863.607 1509.99,863.61 1510.84,863.613 1511.7,863.617 1512.55,863.62 1513.41,863.624 1514.26,863.628 1515.11,863.632 1515.97,863.636 1516.82,863.64 1517.67,863.645 1518.53,863.65 1519.38,863.655 1520.24,863.66 1521.09,863.666 1521.94,863.671 1522.8,863.678 1523.65,863.684 1524.51,863.691 1525.36,863.698 1526.21,863.706 1527.07,863.714 1527.92,863.723 1528.78,863.732 1529.63,863.742 1530.48,863.753 1531.34,863.764 1532.19,863.776 1533.04,863.789 1533.9,863.803 1534.75,863.817 1535.61,863.833 1536.46,863.85 1537.31,863.869 1538.17,863.888 1539.02,863.91 1539.88,863.933 1540.73,863.959 1541.58,863.986 1542.44,864.017 1543.29,864.05 1544.15,864.086 1545,864.127 1545.85,864.171 1546.71,864.221 1547.56,864.276 1548.41,864.337 1549.27,864.407 1550.12,864.485 1550.98,864.574 1551.83,864.675 1552.68,864.792 1553.54,864.926 1554.39,865.083 1555.25,865.267 1556.1,865.486 1556.95,865.748 1557.81,866.065 1558.66,866.454 1559.52,866.939 1560.37,867.556 1561.22,868.354 1562.08,869.416 1562.93,870.871 1563.78,872.937 1564.64,876.006 1565.49,880.805 1566.35,888.692 1567.2,901.827 1568.05,921.785 1568.91,946.426 1569.76,971.216 1570.62,993.263 1571.47,1011.72 1572.32,1026.73 1573.18,1038.76 1574.03,1048.31 1574.89,1055.86 1575.74,1061.82 1576.59,1066.5 1577.45,1070.18 1578.3,1073.06 1579.16,1075.32 1580.01,1077.1 1580.86,1078.48 1581.72,1079.56 1582.57,1080.4 1583.42,1081.05 1584.28,1081.55 1585.13,1081.94 1585.99,1082.23 1586.84,1082.45 1587.69,1082.61 1588.55,1082.72 1589.4,1082.8 1590.26,1082.84 1591.11,1082.87 1591.96,1082.87 1592.82,1082.86 1593.67,1082.84 1594.53,1082.8 1595.38,1082.76 1596.23,1082.71 1597.09,1082.65 1597.94,1082.59 1598.79,1082.52 1599.65,1082.45 1600.5,1082.38 1601.36,1082.3 1602.21,1082.22 1603.06,1082.14 1603.92,1082.05 1604.77,1081.96 1605.63,1081.87 1606.48,1081.77 1607.33,1081.68 1608.19,1081.58 1609.04,1081.47 1609.9,1081.37 1610.75,1081.26 1611.6,1081.15 1612.46,1081.04 1613.31,1080.92 1614.16,1080.81 1615.02,1080.69 1615.87,1080.57 1616.73,1080.44 1617.58,1080.31 1618.43,1080.18 1619.29,1080.05 1620.14,1079.92 1621,1079.78 1621.85,1079.64 1622.7,1079.5 1623.56,1079.35 1624.41,1079.2 1625.27,1079.05 1626.12,1078.9 1626.97,1078.74 1627.83,1078.59 1628.68,1078.42 1629.53,1078.26 1630.39,1078.09 1631.24,1077.92 1632.1,1077.75 1632.95,1077.57 1633.8,1077.39 1634.66,1077.21 1635.51,1077.03 1636.37,1076.84 1637.22,1076.65 1638.07,1076.45 1638.93,1076.26 1639.78,1076.06 1640.64,1075.85 1641.49,1075.65 1642.34,1075.44 1643.2,1075.22 1644.05,1075.01 1644.9,1074.79 1645.76,1074.56 1646.61,1074.34 1647.47,1074.11 1648.32,1073.87 1649.17,1073.64 1650.03,1073.4 1650.88,1073.15 1651.74,1072.9 1652.59,1072.65 1653.44,1072.4 1654.3,1072.14 1655.15,1071.88 1656.01,1071.61 1656.86,1071.34 1657.71,1071.07 1658.57,1070.79 1659.42,1070.51 1660.27,1070.23 1661.13,1069.94 1661.98,1069.64 1662.84,1069.35 1663.69,1069.05 1664.54,1068.74 1665.4,1068.43 1666.25,1068.12 1667.11,1067.8 1667.96,1067.48 1668.81,1067.15 1669.67,1066.82 1670.52,1066.49 1671.38,1066.15 1672.23,1065.8 1673.08,1065.45 1673.94,1065.1 1674.79,1064.74 1675.64,1064.38 1676.5,1064.01 1677.35,1063.64 1678.21,1063.26 1679.06,1062.88 1679.91,1062.5 1680.77,1062.1 1681.62,1061.71 1682.48,1061.31 1683.33,1060.9 1684.18,1060.49 1685.04,1060.07 1685.89,1059.65 1686.75,1059.22 1687.6,1058.79 1688.45,1058.35 1689.31,1057.9 1690.16,1057.45 1691.01,1057 1691.87,1056.54 1692.72,1056.07 1693.58,1055.6 1694.43,1055.12 1695.28,1054.63 1696.14,1054.14 1696.99,1053.64 1697.85,1053.14 1698.7,1052.63 1699.55,1052.12 1700.41,1051.6 1701.26,1051.07 1702.12,1050.53 1702.97,1049.99 1703.82,1049.44 1704.68,1048.89 1705.53,1048.33 1706.38,1047.76 1707.24,1047.18 1708.09,1046.6 1708.95,1046.01 1709.8,1045.42 1710.65,1044.81 1711.51,1044.2 1712.36,1043.58 1713.22,1042.96 1714.07,1042.32 1714.92,1041.68 1715.78,1041.03 1716.63,1040.38 1717.49,1039.71 1718.34,1039.04 1719.19,1038.36 1720.05,1037.67 1720.9,1036.97 1721.76,1036.27 1722.61,1035.55 1723.46,1034.83 1724.32,1034.1 1725.17,1033.36 1726.02,1032.61 1726.88,1031.86 1727.73,1031.09 1728.59,1030.31 1729.44,1029.53 1730.29,1028.74 1731.15,1027.94 1732,1027.12 1732.86,1026.3 1733.71,1025.47 1734.56,1024.64 1735.42,1023.79 1736.27,1022.93 1737.13,1022.06 1737.98,1021.19 1738.83,1020.3 1739.69,1019.41 1740.54,1018.51 1741.39,1017.6 1742.25,1016.68 1743.1,1015.75 1743.96,1014.82 1744.81,1013.87 1745.66,1012.93 1746.52,1011.97 1747.37,1011.01 1748.23,1010.04 1749.08,1009.07 1749.93,1008.09 1750.79,1007.11 1751.64,1006.13 1752.5,1005.15 1753.35,1004.17 1754.2,1003.19 1755.06,1002.22 1755.91,1001.25 1756.76,1000.29 1757.62,999.347 1758.47,998.414 1759.33,997.499 1760.18,996.604 1761.03,995.734 1761.89,994.894 1762.74,994.088 1763.6,993.322 1764.45,992.602 1765.3,991.935 1766.16,991.329 1767.01,990.792 1767.87,990.333 1768.72,989.964 1769.57,989.696 1770.43,989.542 1771.28,989.517 1772.13,989.636 1772.99,989.919 1773.84,990.384 1774.7,991.055 1775.55,991.957 1776.4,993.118 1777.26,994.57 1778.11,996.347 1778.97,998.489 1779.82,1001.04 1780.67,1004.05 1781.53,1007.57 1782.38,1011.67 1783.24,1016.41 1784.09,1021.87 1784.94,1028.13 1785.8,1035.28 1786.65,1043.42 1787.5,1052.67 1788.36,1063.14 1789.21,1074.95 1790.07,1088.25 1790.92,1103.17 1791.77,1119.85 1792.63,1138.45 1793.48,1159.09 1794.34,1181.88 1795.19,1206.86 1796.04,1233.99 1796.9,1263 1797.75,1293.2 1798.61,1323.12 1799.46,1349.95 1800.31,1370.16 1801.17,1382.76 1802.02,1390.02 1802.87,1394.36 1803.73,1397.12 1804.58,1399 1805.44,1400.35 1806.29,1401.35 1807.14,1402.12 1808,1402.73 1808.85,1403.22 1809.71,1403.62 1810.56,1403.96 1811.41,1404.24 1812.27,1404.48 1813.12,1404.69 1813.98,1404.87 1814.83,1405.03 1815.68,1405.16 1816.54,1405.29 1817.39,1405.39 1818.24,1405.49 1819.1,1405.58 1819.95,1405.66 1820.81,1405.73 1821.66,1405.79 1822.51,1405.85 1823.37,1405.91 1824.22,1405.96 1825.08,1406 1825.93,1406.04 1826.78,1406.08 1827.64,1406.12 1828.49,1406.15 1829.35,1406.18 1830.2,1406.21 1831.05,1406.24 1831.91,1406.27 1832.76,1406.29 1833.61,1406.31 1834.47,1406.33 1835.32,1406.35 1836.18,1406.37 1837.03,1406.39 1837.88,1406.4 1838.74,1406.42 1839.59,1406.43 1840.45,1406.45 1841.3,1406.46 1842.15,1406.47 1843.01,1406.49 1843.86,1406.5 1844.72,1406.51 1845.57,1406.52 1846.42,1406.53 1847.28,1406.54 1848.13,1406.55 1848.99,1406.56 1849.84,1406.56 1850.69,1406.57 1851.55,1406.58 1852.4,1406.59 1853.25,1406.59 1854.11,1406.6 1854.96,1406.61 1855.82,1406.61 1856.67,1406.62 1857.52,1406.62 1858.38,1406.63 1859.23,1406.64 1860.09,1406.64 1860.94,1406.65 1861.79,1406.65 1862.65,1406.65 1863.5,1406.66 1864.36,1406.66 1865.21,1406.67 1866.06,1406.67 1866.92,1406.68 1867.77,1406.68 1868.62,1406.68 1869.48,1406.69 1870.33,1406.69 1871.19,1406.69 1872.04,1406.7 1872.89,1406.7 1873.75,1406.7 1874.6,1406.71 1875.46,1406.71 1876.31,1406.71 1877.16,1406.71 1878.02,1406.72 1878.87,1406.72 1879.73,1406.72 1880.58,1406.72 1881.43,1406.73 1882.29,1406.73 1883.14,1406.73 1883.99,1406.73 1884.85,1406.73 1885.7,1406.74 1886.56,1406.74 1887.41,1406.74 1888.26,1406.74 1889.12,1406.74 1889.97,1406.75 1890.83,1406.75 1891.68,1406.75 1892.53,1406.75 1893.39,1406.75 1894.24,1406.75 1895.1,1406.76 1895.95,1406.76 1896.8,1406.76 1897.66,1406.76 1898.51,1406.76 1899.36,1406.76 1900.22,1406.76 1901.07,1406.77 1901.93,1406.77 1902.78,1406.77 1903.63,1406.77 1904.49,1406.77 1905.34,1406.77 1906.2,1406.77 1907.05,1406.77 1907.9,1406.78 1908.76,1406.78 1909.61,1406.78 1910.47,1406.78 1911.32,1406.78 1912.17,1406.78 1913.03,1406.78 1913.88,1406.78 1914.73,1406.78 1915.59,1406.78 1916.44,1406.79 1917.3,1406.79 1918.15,1406.79 1919,1406.79 1919.86,1406.79 1920.71,1406.79 1921.57,1406.79 1922.42,1406.79 1923.27,1406.79 1924.13,1406.79 1924.98,1406.79 1925.84,1406.79 1926.69,1406.8 1927.54,1406.8 1928.4,1406.8 1929.25,1406.8 1930.1,1406.8 1930.96,1406.8 1931.81,1406.8 1932.67,1406.8 1933.52,1406.8 1934.37,1406.8 1935.23,1406.8 1936.08,1406.8 1936.94,1406.8 1937.79,1406.8 1938.64,1406.8 1939.5,1406.81 1940.35,1406.81 1941.21,1406.81 1942.06,1406.81 1942.91,1406.81 1943.77,1406.81 1944.62,1406.81 1945.47,1406.81 1946.33,1406.81 1947.18,1406.81 1948.04,1406.81 1948.89,1406.81 1949.74,1406.81 1950.6,1406.81 1951.45,1406.81 1952.31,1406.81 1953.16,1406.81 1954.01,1406.81 1954.87,1406.81 1955.72,1406.81 1956.58,1406.82 1957.43,1406.82 1958.28,1406.82 1959.14,1406.82 1959.99,1406.82 1960.84,1406.82 1961.7,1406.82 1962.55,1406.82 1963.41,1406.82 1964.26,1406.82 1965.11,1406.82 1965.97,1406.82 1966.82,1406.82 1967.68,1406.82 1968.53,1406.82 1969.38,1406.82 1970.24,1406.82 1971.09,1406.82 1971.95,1406.82 1972.8,1406.82 1973.65,1406.82 1974.51,1406.82 1975.36,1406.82 1976.21,1406.82 1977.07,1406.82 1977.92,1406.82 1978.78,1406.82 1979.63,1406.82 1980.48,1406.83 1981.34,1406.83 1982.19,1406.83 1983.05,1406.83 1983.9,1406.83 1984.75,1406.83 1985.61,1406.83 1986.46,1406.83 1987.32,1406.83 1988.17,1406.83 1989.02,1406.83 1989.88,1406.83 1990.73,1406.83 1991.59,1406.83 1992.44,1406.83 1993.29,1406.83 1994.15,1406.83 1995,1406.83 1995.85,1406.83 1996.71,1406.83 1997.56,1406.83 1998.42,1406.83 1999.27,1406.83 2000.12,1406.83 2000.98,1406.83 2001.83,1406.83 2002.69,1406.83 2003.54,1406.83 2004.39,1406.83 2005.25,1406.83 2006.1,1406.83 2006.96,1406.83 2007.81,1406.83 2008.66,1406.83 2009.52,1406.83 2010.37,1406.83 2011.22,1406.83 2012.08,1406.83 2012.93,1406.83 2013.79,1406.83 2014.64,1406.83 2015.49,1406.83 2016.35,1406.83 2017.2,1406.84 2018.06,1406.84 2018.91,1406.84 2019.76,1406.84 2020.62,1406.84 2021.47,1406.84 2022.33,1406.84 2023.18,1406.84 2024.03,1406.84 2024.89,1406.84 2025.74,1406.84 2026.59,1406.84 2027.45,1406.84 2028.3,1406.84 2029.16,1406.84 2030.01,1406.84 2030.86,1406.84 2031.72,1406.84 2032.57,1406.84 2033.43,1406.84 2034.28,1406.84 2035.13,1406.84 2035.99,1406.84 2036.84,1406.84 2037.7,1406.84 2038.55,1406.84 2039.4,1406.84 2040.26,1406.84 2041.11,1406.84 2041.96,1406.84 2042.82,1406.84 2043.67,1406.84 2044.53,1406.84 2045.38,1406.84 2046.23,1406.84 2047.09,1406.84 2047.94,1406.84 2048.8,1406.84 2049.65,1406.84 2050.5,1406.84 2051.36,1406.84 2052.21,1406.84 2053.07,1406.84 2053.92,1406.84 2054.77,1406.84 2055.63,1406.84 2056.48,1406.84 2057.33,1406.84 2058.19,1406.84 2059.04,1406.84 2059.9,1406.84 2060.75,1406.84 2061.6,1406.84 2062.46,1406.84 2063.31,1406.84 2064.17,1406.84 2065.02,1406.84 2065.87,1406.84 2066.73,1406.84 2067.58,1406.84 2068.44,1406.84 2069.29,1406.84 2070.14,1406.84 2071,1406.84 2071.85,1406.84 2072.7,1406.84 2073.56,1406.84 2074.41,1406.84 2075.27,1406.84 2076.12,1406.84 2076.97,1406.84 2077.83,1406.84 2078.68,1406.84 2079.54,1406.84 2080.39,1406.84 2081.24,1406.84 2082.1,1406.84 2082.95,1406.84 2083.81,1406.85 2084.66,1406.85 2085.51,1406.85 2086.37,1406.85 2087.22,1406.85 2088.07,1406.85 2088.93,1406.85 2089.78,1406.85 2090.64,1406.85 2091.49,1406.85 2092.34,1406.85 2093.2,1406.85 2094.05,1406.85 2094.91,1406.85 2095.76,1406.85 2096.61,1406.85 2097.47,1406.85 2098.32,1406.85 2099.18,1406.85 2100.03,1406.85 2100.88,1406.85 2101.74,1406.85 2102.59,1406.85 2103.44,1406.85 2104.3,1406.85 2105.15,1406.85 2106.01,1406.85 2106.86,1406.85 2107.71,1406.85 2108.57,1406.85 2109.42,1406.85 2110.28,1406.85 2111.13,1406.85 2111.98,1406.85 2112.84,1406.85 2113.69,1406.85 2114.55,1406.85 2115.4,1406.85 2116.25,1406.85 2117.11,1406.85 2117.96,1406.85 2118.82,1406.85 2119.67,1406.85 2120.52,1406.85 2121.38,1406.85 2122.23,1406.85 2123.08,1406.85 2123.94,1406.85 2124.79,1406.85 2125.65,1406.85 2126.5,1406.85 2127.35,1406.85 2128.21,1406.85 2129.06,1406.85 2129.92,1406.85 2130.77,1406.85 2131.62,1406.85 2132.48,1406.85 2133.33,1406.85 2134.19,1406.85 2135.04,1406.85 2135.89,1406.85 2136.75,1406.85 2137.6,1406.85 2138.45,1406.85 2139.31,1406.85 2140.16,1406.85 2141.02,1406.85 2141.87,1406.85 2142.72,1406.85 2143.58,1406.85 2144.43,1406.85 2145.29,1406.85 2146.14,1406.85 2146.99,1406.85 2147.85,1406.85 2148.7,1406.85 2149.56,1406.85 2150.41,1406.85 2151.26,1406.85 2152.12,1406.85 2152.97,1406.85 2153.82,1406.85 2154.68,1406.85 2155.53,1406.85 2156.39,1406.85 2157.24,1406.85 2158.09,1406.85 2158.95,1406.85 2159.8,1406.85 2160.66,1406.85 2161.51,1406.85 2162.36,1406.85 2163.22,1406.85 2164.07,1406.85 2164.93,1406.85 2165.78,1406.85 2166.63,1406.85 2167.49,1406.85 2168.34,1406.85 2169.19,1406.85 2170.05,1406.85 2170.9,1406.85 2171.76,1406.85 2172.61,1406.85 2173.46,1406.85 2174.32,1406.85 2175.17,1406.85 2176.03,1406.85 2176.88,1406.85 2177.73,1406.85 2178.59,1406.85 2179.44,1406.85 2180.3,1406.85 2181.15,1406.85 2182,1406.85 2182.86,1406.85 2183.71,1406.85 2184.56,1406.85 2185.42,1406.85 2186.27,1406.85 2187.13,1406.85 2187.98,1406.85 2188.83,1406.85 2189.69,1406.85 2190.54,1406.85 2191.4,1406.85 2192.25,1406.85 2193.1,1406.85 2193.96,1406.85 2194.81,1406.85 2195.67,1406.85 2196.52,1406.85 2197.37,1406.85 2198.23,1406.85 2199.08,1406.85 2199.93,1406.85 2200.79,1406.85 2201.64,1406.85 2202.5,1406.85 2203.35,1406.85 2204.2,1406.85 2205.06,1406.85 2205.91,1406.85 2206.77,1406.85 2207.62,1406.85 2208.47,1406.85 2209.33,1406.85 2210.18,1406.85 2211.04,1406.85 2211.89,1406.85 2212.74,1406.85 2213.6,1406.85 2214.45,1406.85 2215.3,1406.85 2216.16,1406.85 2217.01,1406.85 2217.87,1406.85 2218.72,1406.85 2219.57,1406.85 2220.43,1406.85 2221.28,1406.85 2222.14,1406.85 2222.99,1406.85 2223.84,1406.85 2224.7,1406.85 2225.55,1406.85 2226.41,1406.85 2227.26,1406.85 2228.11,1406.85 2228.97,1406.85 2229.82,1406.85 2230.67,1406.85 2231.53,1406.85 2232.38,1406.85 2233.24,1406.85 2234.09,1406.85 2234.94,1406.85 2235.8,1406.85 2236.65,1406.85 2237.51,1406.85 2238.36,1406.85 2239.21,1406.85 2240.07,1406.85 2240.92,1406.85 2241.78,1406.85 2242.63,1406.85 2243.48,1406.85 2244.34,1406.85 2245.19,1406.85 2246.04,1406.85 2246.9,1406.85 2247.75,1406.85 2248.61,1406.85 2249.46,1406.85 2250.31,1406.85 2251.17,1406.85 2252.02,1406.85 2252.88,1406.85 2253.73,1406.85 2254.58,1406.85 2255.44,1406.85 2256.29,1406.85 2257.15,1406.85 2258,1406.85 2258.85,1406.85 2259.71,1406.85 2260.56,1406.85 2261.42,1406.85 2262.27,1406.85 2263.12,1406.85 2263.98,1406.85 2264.83,1406.85 2265.68,1406.85 2266.54,1406.85 2267.39,1406.85 2268.25,1406.85 2269.1,1406.85 2269.95,1406.85 2270.81,1406.85 2271.66,1406.85 2272.52,1406.85 2273.37,1406.85 2274.22,1406.85 2275.08,1406.85 2275.93,1406.85 2276.79,1406.85 2277.64,1406.85 2278.49,1406.85 2279.35,1406.85 2280.2,1406.85 2281.05,1406.85 2281.91,1406.85 2282.76,1406.85 2283.62,1406.84 2284.47,1406.84 2285.32,1406.84 2286.18,1406.84 2287.03,1406.84 2287.89,1406.84 2288.74,1406.84 2289.59,1406.84 2290.45,1406.84 2291.3,1406.84 2292.16,1406.84 2293.01,1406.84 2293.86,1406.84 2294.72,1406.84 2295.57,1406.84 2296.42,1406.84 2297.28,1406.84 2298.13,1406.84 2298.99,1406.84 2299.84,1406.84 2300.69,1406.84 2301.55,1406.84 2302.4,1406.84 2303.26,1406.84 2304.11,1406.84 2304.96,1406.84 2305.82,1406.84 2306.67,1406.83 2307.53,1406.83 2308.38,1406.83 2309.23,1406.83 2310.09,1406.83 2310.94,1406.83 2311.79,1406.83 2312.65,1406.83 2313.5,1406.83 2314.36,1406.83 2315.21,1406.82 2316.06,1406.82 2316.92,1406.82 2317.77,1406.82 2318.63,1406.81 2319.48,1406.81 2320.33,1406.81 2321.19,1406.8 2322.04,1406.8 2322.9,1406.79 2323.75,1406.78 2324.6,1406.76 2325.46,1406.73 2326.31,1406.63 2327.16,1406.88 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/nonlinear/rocket_control/index.html b/previews/PR3562/tutorials/nonlinear/rocket_control/index.html
index c8d3338ced2..1a4856a54c4 100644
--- a/previews/PR3562/tutorials/nonlinear/rocket_control/index.html
+++ b/previews/PR3562/tutorials/nonlinear/rocket_control/index.html
@@ -47,7 +47,7 @@
   Dual objective value : 4.66547e+00
 
 * Work counters
-  Solve time (sec)   : 1.75263e-01
+  Solve time (sec)   : 2.95382e-01
   Barrier iterations : 24
 </code></pre><p>Finally, we plot the solution:</p><pre><code class="language-julia hljs">function plot_trajectory(y; kwargs...)
     return Plots.plot(
@@ -65,4 +65,4 @@
     plot_trajectory(x_v; ylabel = &quot;Velocity&quot;),
     plot_trajectory(u_t; ylabel = &quot;Thrust&quot;);
     layout = (2, 2),
-)</code></pre><img src="5e56f1a3.svg" alt="Example block output"/><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Experiment with different values for the constants. How does the solution change? In particular, what happens if you change <code>T_max</code>?</li><li>The dynamical equations use rectangular integration for the right-hand side terms. Modify the equations to use the <a href="https://en.wikipedia.org/wiki/Trapezoidal_rule_(differential_equations)">Trapezoidal rule</a> instead. (As an example, <code>x_v[t-1]</code> would become <code>0.5 * (x_v[t-1] + x_v[t])</code>.) Is there a difference?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../portfolio/">« Portfolio optimization</a><a class="docs-footer-nextpage" href="../space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><img src="cf89a871.svg" alt="Example block output"/><h2 id="Next-steps"><a class="docs-heading-anchor" href="#Next-steps">Next steps</a><a id="Next-steps-1"></a><a class="docs-heading-anchor-permalink" href="#Next-steps" title="Permalink"></a></h2><ul><li>Experiment with different values for the constants. How does the solution change? In particular, what happens if you change <code>T_max</code>?</li><li>The dynamical equations use rectangular integration for the right-hand side terms. Modify the equations to use the <a href="https://en.wikipedia.org/wiki/Trapezoidal_rule_(differential_equations)">Trapezoidal rule</a> instead. (As an example, <code>x_v[t-1]</code> would become <code>0.5 * (x_v[t-1] + x_v[t])</code>.) Is there a difference?</li></ul></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../portfolio/">« Portfolio optimization</a><a class="docs-footer-nextpage" href="../space_shuttle_reentry_trajectory/">Optimal control for a Space Shuttle reentry trajectory »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/simple_examples/index.html b/previews/PR3562/tutorials/nonlinear/simple_examples/index.html
index c64ea100183..9a0221c8fb3 100644
--- a/previews/PR3562/tutorials/nonlinear/simple_examples/index.html
+++ b/previews/PR3562/tutorials/nonlinear/simple_examples/index.html
@@ -100,7 +100,7 @@
 Number of equality constraint Jacobian evaluations   = 5
 Number of inequality constraint Jacobian evaluations = 0
 Number of Lagrangian Hessian evaluations             = 4
-Total seconds in IPOPT                               = 0.031
+Total seconds in IPOPT                               = 0.048
 
 EXIT: Optimal Solution Found.
 termination_status = LOCALLY_SOLVED
@@ -180,4 +180,4 @@
  z ≥ 0
 Objective value: 0.32699283491387243
 x = 0.32699283491387243
-y = 0.2570658388068964</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../portfolio/">Portfolio optimization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+y = 0.2570658388068964</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../introduction/">« Introduction</a><a class="docs-footer-nextpage" href="../portfolio/">Portfolio optimization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/97a644df.svg b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b724747b.svg
similarity index 88%
rename from previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/97a644df.svg
rename to previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b724747b.svg
index eb116422ae2..83f9eb7435d 100644
--- a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/97a644df.svg
+++ b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b724747b.svg
@@ -1,97 +1,97 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="700" height="700" viewBox="0 0 2800 2800">
 <defs>
-  <clipPath id="clip270">
+  <clipPath id="clip090">
     <rect x="0" y="0" width="2800" height="2800"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip270)" d="M0 2800 L2800 2800 L2800 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip090)" d="M0 2800 L2800 2800 L2800 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip271">
+  <clipPath id="clip091">
     <rect x="560" y="280" width="1961" height="1961"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip270)" d="M188.677 790.501 L2752.76 790.501 L2752.76 153.712 L188.677 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip090)" d="M188.677 790.501 L2752.76 790.501 L2752.76 153.712 L188.677 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip272">
+  <clipPath id="clip092">
     <rect x="188" y="153" width="2565" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,790.501 261.245,153.712 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="864.774,790.501 864.774,153.712 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1468.3,790.501 1468.3,153.712 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2071.83,790.501 2071.83,153.712 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2675.36,790.501 2675.36,153.712 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,790.501 2752.76,790.501 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,790.501 261.245,771.604 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="864.774,790.501 864.774,771.604 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1468.3,790.501 1468.3,771.604 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2071.83,790.501 2071.83,771.604 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2675.36,790.501 2675.36,771.604 "/>
-<path clip-path="url(#clip270)" d="M261.245 824.62 Q257.634 824.62 255.805 828.185 Q254 831.726 254 838.856 Q254 845.962 255.805 849.527 Q257.634 853.069 261.245 853.069 Q264.879 853.069 266.685 849.527 Q268.514 845.962 268.514 838.856 Q268.514 831.726 266.685 828.185 Q264.879 824.62 261.245 824.62 M261.245 820.916 Q267.055 820.916 270.111 825.523 Q273.189 830.106 273.189 838.856 Q273.189 847.583 270.111 852.189 Q267.055 856.773 261.245 856.773 Q255.435 856.773 252.356 852.189 Q249.301 847.583 249.301 838.856 Q249.301 830.106 252.356 825.523 Q255.435 820.916 261.245 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M824.392 821.541 L842.748 821.541 L842.748 825.476 L828.674 825.476 L828.674 833.949 Q829.693 833.601 830.711 833.439 Q831.73 833.254 832.748 833.254 Q838.535 833.254 841.915 836.425 Q845.295 839.597 845.295 845.013 Q845.295 850.592 841.822 853.694 Q838.35 856.773 832.031 856.773 Q829.855 856.773 827.586 856.402 Q825.341 856.032 822.934 855.291 L822.934 850.592 Q825.017 851.726 827.239 852.282 Q829.461 852.837 831.938 852.837 Q835.943 852.837 838.281 850.731 Q840.619 848.624 840.619 845.013 Q840.619 841.402 838.281 839.296 Q835.943 837.189 831.938 837.189 Q830.063 837.189 828.188 837.606 Q826.336 838.023 824.392 838.902 L824.392 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M864.507 824.62 Q860.896 824.62 859.068 828.185 Q857.262 831.726 857.262 838.856 Q857.262 845.962 859.068 849.527 Q860.896 853.069 864.507 853.069 Q868.142 853.069 869.947 849.527 Q871.776 845.962 871.776 838.856 Q871.776 831.726 869.947 828.185 Q868.142 824.62 864.507 824.62 M864.507 820.916 Q870.318 820.916 873.373 825.523 Q876.452 830.106 876.452 838.856 Q876.452 847.583 873.373 852.189 Q870.318 856.773 864.507 856.773 Q858.697 856.773 855.619 852.189 Q852.563 847.583 852.563 838.856 Q852.563 830.106 855.619 825.523 Q858.697 820.916 864.507 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M894.669 824.62 Q891.058 824.62 889.23 828.185 Q887.424 831.726 887.424 838.856 Q887.424 845.962 889.23 849.527 Q891.058 853.069 894.669 853.069 Q898.304 853.069 900.109 849.527 Q901.938 845.962 901.938 838.856 Q901.938 831.726 900.109 828.185 Q898.304 824.62 894.669 824.62 M894.669 820.916 Q900.479 820.916 903.535 825.523 Q906.614 830.106 906.614 838.856 Q906.614 847.583 903.535 852.189 Q900.479 856.773 894.669 856.773 Q888.859 856.773 885.78 852.189 Q882.725 847.583 882.725 838.856 Q882.725 830.106 885.78 825.523 Q888.859 820.916 894.669 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1412.83 852.166 L1420.47 852.166 L1420.47 825.8 L1412.16 827.467 L1412.16 823.208 L1420.42 821.541 L1425.1 821.541 L1425.1 852.166 L1432.74 852.166 L1432.74 856.101 L1412.83 856.101 L1412.83 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1452.18 824.62 Q1448.57 824.62 1446.74 828.185 Q1444.93 831.726 1444.93 838.856 Q1444.93 845.962 1446.74 849.527 Q1448.57 853.069 1452.18 853.069 Q1455.81 853.069 1457.62 849.527 Q1459.45 845.962 1459.45 838.856 Q1459.45 831.726 1457.62 828.185 Q1455.81 824.62 1452.18 824.62 M1452.18 820.916 Q1457.99 820.916 1461.05 825.523 Q1464.12 830.106 1464.12 838.856 Q1464.12 847.583 1461.05 852.189 Q1457.99 856.773 1452.18 856.773 Q1446.37 856.773 1443.29 852.189 Q1440.24 847.583 1440.24 838.856 Q1440.24 830.106 1443.29 825.523 Q1446.37 820.916 1452.18 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1482.34 824.62 Q1478.73 824.62 1476.9 828.185 Q1475.1 831.726 1475.1 838.856 Q1475.1 845.962 1476.9 849.527 Q1478.73 853.069 1482.34 853.069 Q1485.98 853.069 1487.78 849.527 Q1489.61 845.962 1489.61 838.856 Q1489.61 831.726 1487.78 828.185 Q1485.98 824.62 1482.34 824.62 M1482.34 820.916 Q1488.15 820.916 1491.21 825.523 Q1494.29 830.106 1494.29 838.856 Q1494.29 847.583 1491.21 852.189 Q1488.15 856.773 1482.34 856.773 Q1476.53 856.773 1473.45 852.189 Q1470.4 847.583 1470.4 838.856 Q1470.4 830.106 1473.45 825.523 Q1476.53 820.916 1482.34 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1512.5 824.62 Q1508.89 824.62 1507.06 828.185 Q1505.26 831.726 1505.26 838.856 Q1505.26 845.962 1507.06 849.527 Q1508.89 853.069 1512.5 853.069 Q1516.14 853.069 1517.94 849.527 Q1519.77 845.962 1519.77 838.856 Q1519.77 831.726 1517.94 828.185 Q1516.14 824.62 1512.5 824.62 M1512.5 820.916 Q1518.31 820.916 1521.37 825.523 Q1524.45 830.106 1524.45 838.856 Q1524.45 847.583 1521.37 852.189 Q1518.31 856.773 1512.5 856.773 Q1506.69 856.773 1503.61 852.189 Q1500.56 847.583 1500.56 838.856 Q1500.56 830.106 1503.61 825.523 Q1506.69 820.916 1512.5 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2016.36 852.166 L2024 852.166 L2024 825.8 L2015.69 827.467 L2015.69 823.208 L2023.95 821.541 L2028.63 821.541 L2028.63 852.166 L2036.26 852.166 L2036.26 856.101 L2016.36 856.101 L2016.36 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2045.75 821.541 L2064.11 821.541 L2064.11 825.476 L2050.04 825.476 L2050.04 833.949 Q2051.06 833.601 2052.07 833.439 Q2053.09 833.254 2054.11 833.254 Q2059.9 833.254 2063.28 836.425 Q2066.66 839.597 2066.66 845.013 Q2066.66 850.592 2063.19 853.694 Q2059.71 856.773 2053.39 856.773 Q2051.22 856.773 2048.95 856.402 Q2046.7 856.032 2044.3 855.291 L2044.3 850.592 Q2046.38 851.726 2048.6 852.282 Q2050.82 852.837 2053.3 852.837 Q2057.31 852.837 2059.64 850.731 Q2061.98 848.624 2061.98 845.013 Q2061.98 841.402 2059.64 839.296 Q2057.31 837.189 2053.3 837.189 Q2051.43 837.189 2049.55 837.606 Q2047.7 838.023 2045.75 838.902 L2045.75 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2085.87 824.62 Q2082.26 824.62 2080.43 828.185 Q2078.62 831.726 2078.62 838.856 Q2078.62 845.962 2080.43 849.527 Q2082.26 853.069 2085.87 853.069 Q2089.5 853.069 2091.31 849.527 Q2093.14 845.962 2093.14 838.856 Q2093.14 831.726 2091.31 828.185 Q2089.5 824.62 2085.87 824.62 M2085.87 820.916 Q2091.68 820.916 2094.74 825.523 Q2097.81 830.106 2097.81 838.856 Q2097.81 847.583 2094.74 852.189 Q2091.68 856.773 2085.87 856.773 Q2080.06 856.773 2076.98 852.189 Q2073.93 847.583 2073.93 838.856 Q2073.93 830.106 2076.98 825.523 Q2080.06 820.916 2085.87 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2116.03 824.62 Q2112.42 824.62 2110.59 828.185 Q2108.79 831.726 2108.79 838.856 Q2108.79 845.962 2110.59 849.527 Q2112.42 853.069 2116.03 853.069 Q2119.67 853.069 2121.47 849.527 Q2123.3 845.962 2123.3 838.856 Q2123.3 831.726 2121.47 828.185 Q2119.67 824.62 2116.03 824.62 M2116.03 820.916 Q2121.84 820.916 2124.9 825.523 Q2127.98 830.106 2127.98 838.856 Q2127.98 847.583 2124.9 852.189 Q2121.84 856.773 2116.03 856.773 Q2110.22 856.773 2107.14 852.189 Q2104.09 847.583 2104.09 838.856 Q2104.09 830.106 2107.14 825.523 Q2110.22 820.916 2116.03 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2623.97 852.166 L2640.29 852.166 L2640.29 856.101 L2618.35 856.101 L2618.35 852.166 Q2621.01 849.411 2625.59 844.782 Q2630.2 840.129 2631.38 838.786 Q2633.62 836.263 2634.5 834.527 Q2635.41 832.768 2635.41 831.078 Q2635.41 828.324 2633.46 826.587 Q2631.54 824.851 2628.44 824.851 Q2626.24 824.851 2623.79 825.615 Q2621.36 826.379 2618.58 827.93 L2618.58 823.208 Q2621.4 822.074 2623.86 821.495 Q2626.31 820.916 2628.35 820.916 Q2633.72 820.916 2636.91 823.601 Q2640.1 826.287 2640.1 830.777 Q2640.1 832.907 2639.29 834.828 Q2638.51 836.726 2636.4 839.319 Q2635.82 839.99 2632.72 843.208 Q2629.62 846.402 2623.97 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2660.1 824.62 Q2656.49 824.62 2654.67 828.185 Q2652.86 831.726 2652.86 838.856 Q2652.86 845.962 2654.67 849.527 Q2656.49 853.069 2660.1 853.069 Q2663.74 853.069 2665.54 849.527 Q2667.37 845.962 2667.37 838.856 Q2667.37 831.726 2665.54 828.185 Q2663.74 824.62 2660.1 824.62 M2660.1 820.916 Q2665.92 820.916 2668.97 825.523 Q2672.05 830.106 2672.05 838.856 Q2672.05 847.583 2668.97 852.189 Q2665.92 856.773 2660.1 856.773 Q2654.29 856.773 2651.22 852.189 Q2648.16 847.583 2648.16 838.856 Q2648.16 830.106 2651.22 825.523 Q2654.29 820.916 2660.1 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2690.27 824.62 Q2686.66 824.62 2684.83 828.185 Q2683.02 831.726 2683.02 838.856 Q2683.02 845.962 2684.83 849.527 Q2686.66 853.069 2690.27 853.069 Q2693.9 853.069 2695.71 849.527 Q2697.54 845.962 2697.54 838.856 Q2697.54 831.726 2695.71 828.185 Q2693.9 824.62 2690.27 824.62 M2690.27 820.916 Q2696.08 820.916 2699.13 825.523 Q2702.21 830.106 2702.21 838.856 Q2702.21 847.583 2699.13 852.189 Q2696.08 856.773 2690.27 856.773 Q2684.46 856.773 2681.38 852.189 Q2678.32 847.583 2678.32 838.856 Q2678.32 830.106 2681.38 825.523 Q2684.46 820.916 2690.27 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2720.43 824.62 Q2716.82 824.62 2714.99 828.185 Q2713.18 831.726 2713.18 838.856 Q2713.18 845.962 2714.99 849.527 Q2716.82 853.069 2720.43 853.069 Q2724.06 853.069 2725.87 849.527 Q2727.7 845.962 2727.7 838.856 Q2727.7 831.726 2725.87 828.185 Q2724.06 824.62 2720.43 824.62 M2720.43 820.916 Q2726.24 820.916 2729.29 825.523 Q2732.37 830.106 2732.37 838.856 Q2732.37 847.583 2729.29 852.189 Q2726.24 856.773 2720.43 856.773 Q2714.62 856.773 2711.54 852.189 Q2708.48 847.583 2708.48 838.856 Q2708.48 830.106 2711.54 825.523 Q2714.62 820.916 2720.43 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,786.533 2752.76,786.533 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,659.948 2752.76,659.948 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,533.363 2752.76,533.363 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,406.778 2752.76,406.778 "/>
-<polyline clip-path="url(#clip272)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,280.193 2752.76,280.193 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,790.501 188.677,153.712 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,786.533 207.574,786.533 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,659.948 207.574,659.948 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,533.363 207.574,533.363 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,406.778 207.574,406.778 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,280.193 207.574,280.193 "/>
-<path clip-path="url(#clip270)" d="M49.9875 799.878 L57.6263 799.878 L57.6263 773.513 L49.3162 775.179 L49.3162 770.92 L57.58 769.253 L62.2559 769.253 L62.2559 799.878 L69.8948 799.878 L69.8948 803.813 L49.9875 803.813 L49.9875 799.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M89.9178 784.67 Q86.7697 784.67 84.9179 786.823 Q83.0892 788.976 83.0892 792.726 Q83.0892 796.452 84.9179 798.628 Q86.7697 800.781 89.9178 800.781 Q93.066 800.781 94.8947 798.628 Q96.7465 796.452 96.7465 792.726 Q96.7465 788.976 94.8947 786.823 Q93.066 784.67 89.9178 784.67 M99.2002 770.017 L99.2002 774.277 Q97.4409 773.443 95.6354 773.003 Q93.853 772.564 92.0937 772.564 Q87.4641 772.564 85.0105 775.689 Q82.5799 778.814 82.2327 785.133 Q83.5984 783.119 85.6586 782.054 Q87.7188 780.966 90.1956 780.966 Q95.4039 780.966 98.4132 784.138 Q101.446 787.286 101.446 792.726 Q101.446 798.05 98.2974 801.267 Q95.1493 804.485 89.9178 804.485 Q83.9225 804.485 80.7512 799.901 Q77.5799 795.295 77.5799 786.568 Q77.5799 778.374 81.4688 773.513 Q85.3577 768.628 91.9086 768.628 Q93.6678 768.628 95.4502 768.976 Q97.2558 769.323 99.2002 770.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M109.501 797.934 L114.385 797.934 L114.385 803.813 L109.501 803.813 L109.501 797.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M135.149 784.67 Q132.001 784.67 130.149 786.823 Q128.32 788.976 128.32 792.726 Q128.32 796.452 130.149 798.628 Q132.001 800.781 135.149 800.781 Q138.297 800.781 140.126 798.628 Q141.978 796.452 141.978 792.726 Q141.978 788.976 140.126 786.823 Q138.297 784.67 135.149 784.67 M144.431 770.017 L144.431 774.277 Q142.672 773.443 140.867 773.003 Q139.084 772.564 137.325 772.564 Q132.695 772.564 130.242 775.689 Q127.811 778.814 127.464 785.133 Q128.83 783.119 130.89 782.054 Q132.95 780.966 135.427 780.966 Q140.635 780.966 143.644 784.138 Q146.677 787.286 146.677 792.726 Q146.677 798.05 143.529 801.267 Q140.381 804.485 135.149 804.485 Q129.154 804.485 125.982 799.901 Q122.811 795.295 122.811 786.568 Q122.811 778.374 126.7 773.513 Q130.589 768.628 137.14 768.628 Q138.899 768.628 140.681 768.976 Q142.487 769.323 144.431 770.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M50.2421 673.293 L57.881 673.293 L57.881 646.928 L49.5708 648.594 L49.5708 644.335 L57.8347 642.668 L62.5106 642.668 L62.5106 673.293 L70.1494 673.293 L70.1494 677.228 L50.2421 677.228 L50.2421 673.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M90.1725 658.085 Q87.0243 658.085 85.1725 660.238 Q83.3438 662.39 83.3438 666.14 Q83.3438 669.867 85.1725 672.043 Q87.0243 674.196 90.1725 674.196 Q93.3206 674.196 95.1493 672.043 Q97.0011 669.867 97.0011 666.14 Q97.0011 662.39 95.1493 660.238 Q93.3206 658.085 90.1725 658.085 M99.4548 643.432 L99.4548 647.691 Q97.6956 646.858 95.89 646.418 Q94.1076 645.978 92.3484 645.978 Q87.7188 645.978 85.2651 649.103 Q82.8345 652.228 82.4873 658.548 Q83.8531 656.534 85.9132 655.469 Q87.9734 654.381 90.4502 654.381 Q95.6585 654.381 98.6678 657.553 Q101.7 660.701 101.7 666.14 Q101.7 671.464 98.552 674.682 Q95.4039 677.9 90.1725 677.9 Q84.1771 677.9 81.0058 673.316 Q77.8346 668.71 77.8346 659.983 Q77.8346 651.789 81.7234 646.928 Q85.6123 642.043 92.1632 642.043 Q93.9224 642.043 95.7048 642.391 Q97.5104 642.738 99.4548 643.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M109.756 671.349 L114.64 671.349 L114.64 677.228 L109.756 677.228 L109.756 671.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.825 660.816 Q131.492 660.816 129.57 662.599 Q127.672 664.381 127.672 667.506 Q127.672 670.631 129.57 672.414 Q131.492 674.196 134.825 674.196 Q138.158 674.196 140.08 672.414 Q142.001 670.608 142.001 667.506 Q142.001 664.381 140.08 662.599 Q138.181 660.816 134.825 660.816 M130.149 658.826 Q127.14 658.085 125.45 656.025 Q123.783 653.965 123.783 651.002 Q123.783 646.858 126.723 644.451 Q129.686 642.043 134.825 642.043 Q139.987 642.043 142.927 644.451 Q145.867 646.858 145.867 651.002 Q145.867 653.965 144.177 656.025 Q142.51 658.085 139.524 658.826 Q142.904 659.613 144.779 661.904 Q146.677 664.196 146.677 667.506 Q146.677 672.529 143.598 675.214 Q140.543 677.9 134.825 677.9 Q129.107 677.9 126.029 675.214 Q122.973 672.529 122.973 667.506 Q122.973 664.196 124.871 661.904 Q126.769 659.613 130.149 658.826 M128.436 651.441 Q128.436 654.127 130.103 655.631 Q131.793 657.136 134.825 657.136 Q137.834 657.136 139.524 655.631 Q141.237 654.127 141.237 651.441 Q141.237 648.756 139.524 647.252 Q137.834 645.747 134.825 645.747 Q131.793 645.747 130.103 647.252 Q128.436 648.756 128.436 651.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M50.1495 546.708 L57.7884 546.708 L57.7884 520.342 L49.4782 522.009 L49.4782 517.75 L57.7421 516.083 L62.418 516.083 L62.418 546.708 L70.0568 546.708 L70.0568 550.643 L50.1495 550.643 L50.1495 546.708 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M78.3207 516.083 L100.543 516.083 L100.543 518.074 L87.9965 550.643 L83.1123 550.643 L94.9178 520.018 L78.3207 520.018 L78.3207 516.083 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M109.663 544.764 L114.547 544.764 L114.547 550.643 L109.663 550.643 L109.663 544.764 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 519.162 Q131.121 519.162 129.293 522.727 Q127.487 526.268 127.487 533.398 Q127.487 540.504 129.293 544.069 Q131.121 547.611 134.732 547.611 Q138.367 547.611 140.172 544.069 Q142.001 540.504 142.001 533.398 Q142.001 526.268 140.172 522.727 Q138.367 519.162 134.732 519.162 M134.732 515.458 Q140.543 515.458 143.598 520.065 Q146.677 524.648 146.677 533.398 Q146.677 542.125 143.598 546.731 Q140.543 551.315 134.732 551.315 Q128.922 551.315 125.844 546.731 Q122.788 542.125 122.788 533.398 Q122.788 524.648 125.844 520.065 Q128.922 515.458 134.732 515.458 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M51.7467 420.123 L59.3856 420.123 L59.3856 393.757 L51.0754 395.424 L51.0754 391.165 L59.3393 389.498 L64.0152 389.498 L64.0152 420.123 L71.654 420.123 L71.654 424.058 L51.7467 424.058 L51.7467 420.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M79.9179 389.498 L102.14 389.498 L102.14 391.489 L89.5938 424.058 L84.7095 424.058 L96.515 393.433 L79.9179 393.433 L79.9179 389.498 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M111.26 418.179 L116.145 418.179 L116.145 424.058 L111.26 424.058 L111.26 418.179 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M130.357 420.123 L146.677 420.123 L146.677 424.058 L124.732 424.058 L124.732 420.123 Q127.394 417.368 131.978 412.739 Q136.584 408.086 137.765 406.743 Q140.01 404.22 140.89 402.484 Q141.793 400.725 141.793 399.035 Q141.793 396.28 139.848 394.544 Q137.927 392.808 134.825 392.808 Q132.626 392.808 130.172 393.572 Q127.742 394.336 124.964 395.887 L124.964 391.165 Q127.788 390.031 130.242 389.452 Q132.695 388.873 134.732 388.873 Q140.103 388.873 143.297 391.558 Q146.492 394.243 146.492 398.734 Q146.492 400.864 145.681 402.785 Q144.894 404.683 142.788 407.276 Q142.209 407.947 139.107 411.165 Q136.006 414.359 130.357 420.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M49.6634 293.538 L57.3023 293.538 L57.3023 267.172 L48.9921 268.839 L48.9921 264.58 L57.256 262.913 L61.9319 262.913 L61.9319 293.538 L69.5707 293.538 L69.5707 297.473 L49.6634 297.473 L49.6634 293.538 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M77.8346 262.913 L100.057 262.913 L100.057 264.904 L87.5104 297.473 L82.6262 297.473 L94.4317 266.848 L77.8346 266.848 L77.8346 262.913 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M109.177 291.593 L114.061 291.593 L114.061 297.473 L109.177 297.473 L109.177 291.593 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M137.094 266.987 L125.288 285.436 L137.094 285.436 L137.094 266.987 M135.867 262.913 L141.746 262.913 L141.746 285.436 L146.677 285.436 L146.677 289.325 L141.746 289.325 L141.746 297.473 L137.094 297.473 L137.094 289.325 L121.492 289.325 L121.492 284.811 L135.867 262.913 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1061.76 20.1573 L1050.66 50.2555 L1072.9 50.2555 L1061.76 20.1573 M1057.14 12.096 L1066.42 12.096 L1089.47 72.576 L1080.96 72.576 L1075.45 57.061 L1048.19 57.061 L1042.68 72.576 L1034.05 72.576 L1057.14 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1135.69 45.1919 L1135.69 72.576 L1128.23 72.576 L1128.23 45.4349 Q1128.23 38.994 1125.72 35.7938 Q1123.21 32.5936 1118.19 32.5936 Q1112.15 32.5936 1108.67 36.4419 Q1105.18 40.2903 1105.18 46.9338 L1105.18 72.576 L1097.69 72.576 L1097.69 27.2059 L1105.18 27.2059 L1105.18 34.2544 Q1107.86 30.163 1111.46 28.1376 Q1115.11 26.1121 1119.85 26.1121 Q1127.67 26.1121 1131.68 30.9732 Q1135.69 35.7938 1135.69 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1180.41 49.3643 Q1180.41 41.2625 1177.05 36.8065 Q1173.72 32.3505 1167.69 32.3505 Q1161.69 32.3505 1158.33 36.8065 Q1155.01 41.2625 1155.01 49.3643 Q1155.01 57.4256 1158.33 61.8816 Q1161.69 66.3376 1167.69 66.3376 Q1173.72 66.3376 1177.05 61.8816 Q1180.41 57.4256 1180.41 49.3643 M1187.86 66.9452 Q1187.86 78.5308 1182.72 84.1616 Q1177.57 89.8329 1166.96 89.8329 Q1163.03 89.8329 1159.55 89.2252 Q1156.06 88.6581 1152.78 87.4428 L1152.78 80.1917 Q1156.06 81.9741 1159.26 82.8248 Q1162.46 83.6755 1165.78 83.6755 Q1173.12 83.6755 1176.76 79.8271 Q1180.41 76.0193 1180.41 68.282 L1180.41 64.5957 Q1178.1 68.6061 1174.49 70.5911 Q1170.89 72.576 1165.87 72.576 Q1157.52 72.576 1152.42 66.2161 Q1147.31 59.8562 1147.31 49.3643 Q1147.31 38.832 1152.42 32.472 Q1157.52 26.1121 1165.87 26.1121 Q1170.89 26.1121 1174.49 28.0971 Q1178.1 30.082 1180.41 34.0924 L1180.41 27.2059 L1187.86 27.2059 L1187.86 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1203.21 9.54393 L1210.67 9.54393 L1210.67 72.576 L1203.21 72.576 L1203.21 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1265.07 48.0275 L1265.07 51.6733 L1230.8 51.6733 Q1231.29 59.3701 1235.42 63.421 Q1239.59 67.4314 1247.01 67.4314 Q1251.3 67.4314 1255.31 66.3781 Q1259.36 65.3249 1263.33 63.2184 L1263.33 70.267 Q1259.32 71.9684 1255.11 72.8596 Q1250.89 73.7508 1246.56 73.7508 Q1235.7 73.7508 1229.34 67.4314 Q1223.02 61.1119 1223.02 50.3365 Q1223.02 39.1965 1229.02 32.6746 Q1235.05 26.1121 1245.26 26.1121 Q1254.42 26.1121 1259.73 32.0264 Q1265.07 37.9003 1265.07 48.0275 M1257.62 45.84 Q1257.54 39.7232 1254.18 36.0774 Q1250.85 32.4315 1245.34 32.4315 Q1239.11 32.4315 1235.34 35.9558 Q1231.61 39.4801 1231.04 45.8805 L1257.62 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1321.26 32.4315 Q1315.26 32.4315 1311.78 37.1306 Q1308.3 41.7891 1308.3 49.9314 Q1308.3 58.0738 1311.74 62.7728 Q1315.22 67.4314 1321.26 67.4314 Q1327.21 67.4314 1330.7 62.7323 Q1334.18 58.0333 1334.18 49.9314 Q1334.18 41.8701 1330.7 37.1711 Q1327.21 32.4315 1321.26 32.4315 M1321.26 26.1121 Q1330.98 26.1121 1336.53 32.4315 Q1342.08 38.7509 1342.08 49.9314 Q1342.08 61.0714 1336.53 67.4314 Q1330.98 73.7508 1321.26 73.7508 Q1311.5 73.7508 1305.95 67.4314 Q1300.44 61.0714 1300.44 49.9314 Q1300.44 38.7509 1305.95 32.4315 Q1311.5 26.1121 1321.26 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1377.4 9.54393 L1377.4 15.7418 L1370.27 15.7418 Q1366.26 15.7418 1364.68 17.3622 Q1363.14 18.9825 1363.14 23.1955 L1363.14 27.2059 L1375.42 27.2059 L1375.42 32.9987 L1363.14 32.9987 L1363.14 72.576 L1355.65 72.576 L1355.65 32.9987 L1348.52 32.9987 L1348.52 27.2059 L1355.65 27.2059 L1355.65 24.0462 Q1355.65 16.471 1359.17 13.0277 Q1362.7 9.54393 1370.36 9.54393 L1377.4 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1430.55 20.1573 L1419.45 50.2555 L1441.69 50.2555 L1430.55 20.1573 M1425.93 12.096 L1435.21 12.096 L1458.26 72.576 L1449.75 72.576 L1444.24 57.061 L1416.98 57.061 L1411.47 72.576 L1402.84 72.576 L1425.93 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1472.68 14.324 L1472.68 27.2059 L1488.03 27.2059 L1488.03 32.9987 L1472.68 32.9987 L1472.68 57.6282 Q1472.68 63.1779 1474.18 64.7578 Q1475.72 66.3376 1480.38 66.3376 L1488.03 66.3376 L1488.03 72.576 L1480.38 72.576 Q1471.75 72.576 1468.47 69.3758 Q1465.19 66.1351 1465.19 57.6282 L1465.19 32.9987 L1459.72 32.9987 L1459.72 27.2059 L1465.19 27.2059 L1465.19 14.324 L1472.68 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1505.21 14.324 L1505.21 27.2059 L1520.56 27.2059 L1520.56 32.9987 L1505.21 32.9987 L1505.21 57.6282 Q1505.21 63.1779 1506.71 64.7578 Q1508.25 66.3376 1512.91 66.3376 L1520.56 66.3376 L1520.56 72.576 L1512.91 72.576 Q1504.28 72.576 1501 69.3758 Q1497.72 66.1351 1497.72 57.6282 L1497.72 32.9987 L1492.25 32.9987 L1492.25 27.2059 L1497.72 27.2059 L1497.72 14.324 L1505.21 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1550.99 49.7694 Q1541.95 49.7694 1538.47 51.8354 Q1534.98 53.9013 1534.98 58.8839 Q1534.98 62.8538 1537.58 65.2034 Q1540.21 67.5124 1544.71 67.5124 Q1550.9 67.5124 1554.63 63.1374 Q1558.4 58.7219 1558.4 51.4303 L1558.4 49.7694 L1550.99 49.7694 M1565.85 46.6907 L1565.85 72.576 L1558.4 72.576 L1558.4 65.6895 Q1555.85 69.8214 1552.04 71.8063 Q1548.23 73.7508 1542.72 73.7508 Q1535.75 73.7508 1531.62 69.8619 Q1527.53 65.9325 1527.53 59.3701 Q1527.53 51.7138 1532.63 47.825 Q1537.78 43.9361 1547.95 43.9361 L1558.4 43.9361 L1558.4 43.2069 Q1558.4 38.0623 1555 35.2672 Q1551.63 32.4315 1545.52 32.4315 Q1541.63 32.4315 1537.94 33.3632 Q1534.25 34.295 1530.85 36.1584 L1530.85 29.2718 Q1534.94 27.692 1538.79 26.9223 Q1542.64 26.1121 1546.29 26.1121 Q1556.13 26.1121 1560.99 31.2163 Q1565.85 36.3204 1565.85 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1613.86 28.9478 L1613.86 35.9153 Q1610.7 34.1734 1607.5 33.3227 Q1604.34 32.4315 1601.09 32.4315 Q1593.84 32.4315 1589.83 37.0496 Q1585.82 41.6271 1585.82 49.9314 Q1585.82 58.2358 1589.83 62.8538 Q1593.84 67.4314 1601.09 67.4314 Q1604.34 67.4314 1607.5 66.5807 Q1610.7 65.6895 1613.86 63.9476 L1613.86 70.8341 Q1610.74 72.2924 1607.37 73.0216 Q1604.05 73.7508 1600.28 73.7508 Q1590.04 73.7508 1584 67.3098 Q1577.96 60.8689 1577.96 49.9314 Q1577.96 38.832 1584.04 32.472 Q1590.16 26.1121 1600.77 26.1121 Q1604.21 26.1121 1607.5 26.8413 Q1610.78 27.5299 1613.86 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1626.53 9.54393 L1634.03 9.54393 L1634.03 46.7717 L1656.27 27.2059 L1665.79 27.2059 L1641.73 48.4326 L1666.8 72.576 L1657.08 72.576 L1634.03 50.4176 L1634.03 72.576 L1626.53 72.576 L1626.53 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1719.14 9.62495 Q1713.71 18.942 1711.08 28.0566 Q1708.44 37.1711 1708.44 46.5287 Q1708.44 55.8863 1711.08 65.0818 Q1713.75 74.2369 1719.14 83.5134 L1712.66 83.5134 Q1706.58 73.9938 1703.54 64.7983 Q1700.54 55.6027 1700.54 46.5287 Q1700.54 37.4952 1703.54 28.3401 Q1706.54 19.1851 1712.66 9.62495 L1719.14 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1763.45 34.0924 L1763.45 9.54393 L1770.91 9.54393 L1770.91 72.576 L1763.45 72.576 L1763.45 65.7705 Q1761.11 69.8214 1757.5 71.8063 Q1753.94 73.7508 1748.91 73.7508 Q1740.69 73.7508 1735.5 67.1883 Q1730.36 60.6258 1730.36 49.9314 Q1730.36 39.2371 1735.5 32.6746 Q1740.69 26.1121 1748.91 26.1121 Q1753.94 26.1121 1757.5 28.0971 Q1761.11 30.0415 1763.45 34.0924 M1738.06 49.9314 Q1738.06 58.1548 1741.42 62.8538 Q1744.82 67.5124 1750.74 67.5124 Q1756.65 67.5124 1760.05 62.8538 Q1763.45 58.1548 1763.45 49.9314 Q1763.45 41.7081 1760.05 37.0496 Q1756.65 32.3505 1750.74 32.3505 Q1744.82 32.3505 1741.42 37.0496 Q1738.06 41.7081 1738.06 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1825.07 48.0275 L1825.07 51.6733 L1790.8 51.6733 Q1791.28 59.3701 1795.42 63.421 Q1799.59 67.4314 1807 67.4314 Q1811.3 67.4314 1815.31 66.3781 Q1819.36 65.3249 1823.33 63.2184 L1823.33 70.267 Q1819.32 71.9684 1815.1 72.8596 Q1810.89 73.7508 1806.56 73.7508 Q1795.7 73.7508 1789.34 67.4314 Q1783.02 61.1119 1783.02 50.3365 Q1783.02 39.1965 1789.02 32.6746 Q1795.05 26.1121 1805.26 26.1121 Q1814.42 26.1121 1819.72 32.0264 Q1825.07 37.9003 1825.07 48.0275 M1817.62 45.84 Q1817.53 39.7232 1814.17 36.0774 Q1810.85 32.4315 1805.34 32.4315 Q1799.1 32.4315 1795.34 35.9558 Q1791.61 39.4801 1791.04 45.8805 L1817.62 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1867.16 49.3643 Q1867.16 41.2625 1863.8 36.8065 Q1860.47 32.3505 1854.44 32.3505 Q1848.44 32.3505 1845.08 36.8065 Q1841.76 41.2625 1841.76 49.3643 Q1841.76 57.4256 1845.08 61.8816 Q1848.44 66.3376 1854.44 66.3376 Q1860.47 66.3376 1863.8 61.8816 Q1867.16 57.4256 1867.16 49.3643 M1874.61 66.9452 Q1874.61 78.5308 1869.47 84.1616 Q1864.32 89.8329 1853.71 89.8329 Q1849.78 89.8329 1846.3 89.2252 Q1842.81 88.6581 1839.53 87.4428 L1839.53 80.1917 Q1842.81 81.9741 1846.01 82.8248 Q1849.21 83.6755 1852.53 83.6755 Q1859.87 83.6755 1863.51 79.8271 Q1867.16 76.0193 1867.16 68.282 L1867.16 64.5957 Q1864.85 68.6061 1861.24 70.5911 Q1857.64 72.576 1852.62 72.576 Q1844.27 72.576 1839.17 66.2161 Q1834.06 59.8562 1834.06 49.3643 Q1834.06 38.832 1839.17 32.472 Q1844.27 26.1121 1852.62 26.1121 Q1857.64 26.1121 1861.24 28.0971 Q1864.85 30.082 1867.16 34.0924 L1867.16 27.2059 L1874.61 27.2059 L1874.61 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1888.79 9.62495 L1895.27 9.62495 Q1901.35 19.1851 1904.35 28.3401 Q1907.38 37.4952 1907.38 46.5287 Q1907.38 55.6027 1904.35 64.7983 Q1901.35 73.9938 1895.27 83.5134 L1888.79 83.5134 Q1894.18 74.2369 1896.81 65.0818 Q1899.48 55.8863 1899.48 46.5287 Q1899.48 37.1711 1896.81 28.0566 Q1894.18 18.942 1888.79 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip272)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,237.832 266.073,234.59 270.902,231.464 275.73,228.457 280.558,225.572 285.386,222.813 290.214,220.184 295.043,217.689 299.871,215.334 304.699,213.124 309.527,211.063 314.356,209.159 319.184,207.418 324.012,205.846 328.84,204.451 333.668,203.24 338.497,202.221 343.325,201.402 348.153,200.792 352.981,200.398 357.81,200.23 362.638,200.295 367.466,200.603 372.294,201.16 377.123,201.973 381.951,203.05 386.779,204.396 391.607,206.015 396.435,207.91 401.264,210.082 406.092,212.531 410.92,215.255 415.748,218.248 420.577,221.504 425.405,225.012 430.233,228.762 435.061,232.738 439.89,236.924 444.718,241.298 449.546,245.841 454.374,250.529 459.202,255.335 464.031,260.234 468.859,265.197 473.687,270.197 478.515,275.206 483.344,280.195 488.172,285.138 493,290.007 497.828,294.777 502.656,299.426 507.485,303.93 512.313,308.271 517.141,312.43 521.969,316.391 526.798,320.14 531.626,323.665 536.454,326.957 541.282,330.006 546.111,332.806 550.939,335.352 555.767,337.641 560.595,339.669 565.423,341.435 570.252,342.94 575.08,344.183 579.908,345.166 584.736,345.89 589.565,346.358 594.393,346.573 599.221,346.537 604.049,346.255 608.878,345.73 613.706,344.967 618.534,343.968 623.362,342.74 628.19,341.286 633.019,339.612 637.847,337.721 642.675,335.619 647.503,333.311 652.332,330.802 657.16,328.098 661.988,325.204 666.816,322.126 671.645,318.87 676.473,315.442 681.301,311.849 686.129,308.096 690.957,304.192 695.786,300.143 700.614,295.958 705.442,291.644 710.27,287.21 715.099,282.665 719.927,278.018 724.755,273.281 729.583,268.464 734.411,263.578 739.24,258.636 744.068,253.651 748.896,248.637 753.724,243.608 758.553,238.581 763.381,233.572 768.209,228.599 773.037,223.682 777.866,218.839 782.694,214.093 787.522,209.465 792.35,204.978 797.178,200.658 802.007,196.53 806.835,192.62 811.663,188.956 816.491,185.566 821.32,182.478 826.148,179.721 830.976,177.325 835.804,175.319 840.633,173.73 845.461,172.586 850.289,171.913 855.117,171.734 859.945,172.072 864.774,172.944 869.602,174.365 874.43,176.346 879.258,178.891 884.087,182.002 888.915,185.672 893.743,189.889 898.571,194.636 903.399,199.888 908.228,205.612 913.056,211.771 917.884,218.319 922.712,225.208 927.541,232.381 932.369,239.779 937.197,247.339 942.025,254.996 946.854,262.682 951.682,270.331 956.51,277.875 961.338,285.25 966.166,292.394 970.995,299.247 975.823,305.754 980.651,311.865 985.479,317.535 990.308,322.723 995.136,327.395 999.964,331.522 1004.79,335.08 1009.62,338.05 1014.45,340.42 1019.28,342.181 1024.11,343.329 1028.93,343.865 1033.76,343.795 1038.59,343.125 1043.42,341.868 1048.25,340.039 1053.07,337.656 1057.9,334.739 1062.73,331.312 1067.56,327.399 1072.39,323.028 1077.22,318.23 1082.04,313.035 1086.87,307.476 1091.7,301.59 1096.53,295.414 1101.36,288.986 1106.19,282.346 1111.01,275.537 1115.84,268.604 1120.67,261.591 1125.5,254.545 1130.33,247.515 1135.15,240.552 1139.98,233.706 1144.81,227.029 1149.64,220.575 1154.47,214.398 1159.3,208.55 1164.12,203.087 1168.95,198.06 1173.78,193.523 1178.61,189.524 1183.44,186.112 1188.26,183.331 1193.09,181.221 1197.92,179.818 1202.75,179.152 1207.58,179.245 1212.41,180.113 1217.23,181.763 1222.06,184.194 1226.89,187.395 1231.72,191.343 1236.55,196.007 1241.38,201.346 1246.2,207.308 1251.03,213.831 1255.86,220.845 1260.69,228.271 1265.52,236.026 1270.34,244.018 1275.17,252.154 1280,260.337 1284.83,268.471 1289.66,276.461 1294.49,284.212 1299.31,291.637 1304.14,298.652 1308.97,305.18 1313.8,311.151 1318.63,316.505 1323.46,321.189 1328.28,325.159 1333.11,328.382 1337.94,330.833 1342.77,332.496 1347.6,333.364 1352.42,333.44 1357.25,332.731 1362.08,331.257 1366.91,329.042 1371.74,326.116 1376.57,322.517 1381.39,318.289 1386.22,313.48 1391.05,308.142 1395.88,302.334 1400.71,296.118 1405.54,289.557 1410.36,282.72 1415.19,275.678 1420.02,268.504 1424.85,261.273 1429.68,254.061 1434.5,246.946 1439.33,240.003 1444.16,233.311 1448.99,226.944 1453.82,220.978 1458.65,215.482 1463.47,210.524 1468.3,206.169 1473.13,202.473 1477.96,199.488 1482.79,197.258 1487.62,195.818 1492.44,195.193 1497.27,195.401 1502.1,196.444 1506.93,198.316 1511.76,200.999 1516.58,204.46 1521.41,208.657 1526.24,213.534 1531.07,219.026 1535.9,225.055 1540.73,231.537 1545.55,238.379 1550.38,245.481 1555.21,252.741 1560.04,260.053 1564.87,267.312 1569.7,274.414 1574.52,281.258 1579.35,287.748 1584.18,293.797 1589.01,299.323 1593.84,304.256 1598.66,308.532 1603.49,312.103 1608.32,314.929 1613.15,316.981 1617.98,318.244 1622.81,318.711 1627.63,318.388 1632.46,317.291 1637.29,315.447 1642.12,312.89 1646.95,309.665 1651.77,305.824 1656.6,301.424 1661.43,296.533 1666.26,291.221 1671.09,285.563 1675.92,279.639 1680.74,273.531 1685.57,267.325 1690.4,261.106 1695.23,254.96 1700.06,248.973 1704.89,243.227 1709.71,237.804 1714.54,232.78 1719.37,228.227 1724.2,224.21 1729.03,220.79 1733.85,218.015 1738.68,215.928 1743.51,214.559 1748.34,213.929 1753.17,214.048 1758,214.911 1762.82,216.505 1767.65,218.803 1772.48,221.765 1777.31,225.341 1782.14,229.471 1786.97,234.084 1791.79,239.102 1796.62,244.439 1801.45,250.005 1806.28,255.705 1811.11,261.445 1815.93,267.129 1820.76,272.664 1825.59,277.96 1830.42,282.935 1835.25,287.511 1840.08,291.621 1844.9,295.203 1849.73,298.211 1854.56,300.605 1859.39,302.357 1864.22,303.452 1869.05,303.885 1873.87,303.661 1878.7,302.796 1883.53,301.318 1888.36,299.262 1893.19,296.671 1898.01,293.598 1902.84,290.101 1907.67,286.244 1912.5,282.096 1917.33,277.73 1922.16,273.221 1926.98,268.645 1931.81,264.08 1936.64,259.599 1941.47,255.279 1946.3,251.188 1951.13,247.394 1955.95,243.958 1960.78,240.933 1965.61,238.369 1970.44,236.303 1975.27,234.767 1980.09,233.781 1984.92,233.357 1989.75,233.496 1994.58,234.19 1999.41,235.418 2004.24,237.152 2009.06,239.356 2013.89,241.982 2018.72,244.977 2023.55,248.282 2028.38,251.831 2033.2,255.558 2038.03,259.391 2042.86,263.259 2047.69,267.093 2052.52,270.824 2057.35,274.389 2062.17,277.727 2067,280.785 2071.83,283.517 2076.66,285.884 2081.49,287.854 2086.32,289.405 2091.14,290.523 2095.97,291.203 2100.8,291.449 2105.63,291.272 2110.46,290.69 2115.28,289.73 2120.11,288.425 2124.94,286.811 2129.77,284.932 2134.6,282.832 2139.43,280.562 2144.25,278.172 2149.08,275.711 2153.91,273.233 2158.74,270.785 2163.57,268.415 2168.4,266.169 2173.22,264.086 2178.05,262.202 2182.88,260.549 2187.71,259.151 2192.54,258.028 2197.36,257.193 2202.19,256.652 2207.02,256.405 2211.85,256.446 2216.68,256.763 2221.51,257.339 2226.33,258.152 2231.16,259.175 2235.99,260.379 2240.82,261.732 2245.65,263.201 2250.48,264.752 2255.3,266.354 2260.13,267.973 2264.96,269.581 2269.79,271.152 2274.62,272.665 2279.44,274.1 2284.27,275.446 2289.1,276.694 2293.93,277.841 2298.76,278.89 2303.59,279.848 2308.41,280.725 2313.24,281.537 2318.07,282.302 2322.9,283.041 2327.73,283.775 2332.56,284.528 2337.38,285.321 2342.21,286.177 2347.04,287.113 2351.87,288.145 2356.7,289.285 2361.52,290.538 2366.35,291.907 2371.18,293.386 2376.01,294.964 2380.84,296.621 2385.67,298.333 2390.49,300.069 2395.32,301.789 2400.15,303.452 2404.98,305.009 2409.81,306.409 2414.64,307.597 2419.46,308.52 2424.29,309.124 2429.12,309.359 2433.95,309.177 2438.78,308.542 2443.6,307.422 2448.43,305.799 2453.26,303.666 2458.09,301.032 2462.92,297.923 2467.75,294.38 2472.57,290.467 2477.4,286.264 2482.23,281.873 2487.06,277.417 2491.89,273.036 2496.71,268.893 2501.54,265.164 2506.37,262.042 2511.2,259.735 2516.03,258.459 2520.86,258.437 2525.68,259.896 2530.51,263.059 2535.34,268.144 2540.17,275.355 2545,284.877 2549.83,296.873 2554.65,311.468 2559.48,328.752 2564.31,348.763 2569.14,371.485 2573.97,396.837 2578.79,424.666 2583.62,454.736 2588.45,486.725 2593.28,520.214 2598.11,554.686 2602.94,589.516 2607.76,623.975 2612.59,657.227 2617.42,688.332 2622.25,716.25 2627.08,739.855 2631.91,757.944 2636.73,769.253 2641.56,772.479 2646.39,766.3 2651.22,749.403 2656.05,720.513 2660.87,678.419 2665.7,622.005 2670.53,550.282 2675.36,462.415 2680.19,357.751 "/>
-<path clip-path="url(#clip270)" d="M188.677 1723.83 L2752.76 1723.83 L2752.76 1087.05 L188.677 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,790.501 261.245,153.712 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="864.774,790.501 864.774,153.712 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1468.3,790.501 1468.3,153.712 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2071.83,790.501 2071.83,153.712 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2675.36,790.501 2675.36,153.712 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,790.501 2752.76,790.501 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,790.501 261.245,771.604 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="864.774,790.501 864.774,771.604 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1468.3,790.501 1468.3,771.604 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2071.83,790.501 2071.83,771.604 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2675.36,790.501 2675.36,771.604 "/>
+<path clip-path="url(#clip090)" d="M261.245 824.62 Q257.634 824.62 255.805 828.185 Q254 831.726 254 838.856 Q254 845.962 255.805 849.527 Q257.634 853.069 261.245 853.069 Q264.879 853.069 266.685 849.527 Q268.514 845.962 268.514 838.856 Q268.514 831.726 266.685 828.185 Q264.879 824.62 261.245 824.62 M261.245 820.916 Q267.055 820.916 270.111 825.523 Q273.189 830.106 273.189 838.856 Q273.189 847.583 270.111 852.189 Q267.055 856.773 261.245 856.773 Q255.435 856.773 252.356 852.189 Q249.301 847.583 249.301 838.856 Q249.301 830.106 252.356 825.523 Q255.435 820.916 261.245 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M824.392 821.541 L842.748 821.541 L842.748 825.476 L828.674 825.476 L828.674 833.949 Q829.693 833.601 830.711 833.439 Q831.73 833.254 832.748 833.254 Q838.535 833.254 841.915 836.425 Q845.295 839.597 845.295 845.013 Q845.295 850.592 841.822 853.694 Q838.35 856.773 832.031 856.773 Q829.855 856.773 827.586 856.402 Q825.341 856.032 822.934 855.291 L822.934 850.592 Q825.017 851.726 827.239 852.282 Q829.461 852.837 831.938 852.837 Q835.943 852.837 838.281 850.731 Q840.619 848.624 840.619 845.013 Q840.619 841.402 838.281 839.296 Q835.943 837.189 831.938 837.189 Q830.063 837.189 828.188 837.606 Q826.336 838.023 824.392 838.902 L824.392 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M864.507 824.62 Q860.896 824.62 859.068 828.185 Q857.262 831.726 857.262 838.856 Q857.262 845.962 859.068 849.527 Q860.896 853.069 864.507 853.069 Q868.142 853.069 869.947 849.527 Q871.776 845.962 871.776 838.856 Q871.776 831.726 869.947 828.185 Q868.142 824.62 864.507 824.62 M864.507 820.916 Q870.318 820.916 873.373 825.523 Q876.452 830.106 876.452 838.856 Q876.452 847.583 873.373 852.189 Q870.318 856.773 864.507 856.773 Q858.697 856.773 855.619 852.189 Q852.563 847.583 852.563 838.856 Q852.563 830.106 855.619 825.523 Q858.697 820.916 864.507 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M894.669 824.62 Q891.058 824.62 889.23 828.185 Q887.424 831.726 887.424 838.856 Q887.424 845.962 889.23 849.527 Q891.058 853.069 894.669 853.069 Q898.304 853.069 900.109 849.527 Q901.938 845.962 901.938 838.856 Q901.938 831.726 900.109 828.185 Q898.304 824.62 894.669 824.62 M894.669 820.916 Q900.479 820.916 903.535 825.523 Q906.614 830.106 906.614 838.856 Q906.614 847.583 903.535 852.189 Q900.479 856.773 894.669 856.773 Q888.859 856.773 885.78 852.189 Q882.725 847.583 882.725 838.856 Q882.725 830.106 885.78 825.523 Q888.859 820.916 894.669 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1412.83 852.166 L1420.47 852.166 L1420.47 825.8 L1412.16 827.467 L1412.16 823.208 L1420.42 821.541 L1425.1 821.541 L1425.1 852.166 L1432.74 852.166 L1432.74 856.101 L1412.83 856.101 L1412.83 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1452.18 824.62 Q1448.57 824.62 1446.74 828.185 Q1444.93 831.726 1444.93 838.856 Q1444.93 845.962 1446.74 849.527 Q1448.57 853.069 1452.18 853.069 Q1455.81 853.069 1457.62 849.527 Q1459.45 845.962 1459.45 838.856 Q1459.45 831.726 1457.62 828.185 Q1455.81 824.62 1452.18 824.62 M1452.18 820.916 Q1457.99 820.916 1461.05 825.523 Q1464.12 830.106 1464.12 838.856 Q1464.12 847.583 1461.05 852.189 Q1457.99 856.773 1452.18 856.773 Q1446.37 856.773 1443.29 852.189 Q1440.24 847.583 1440.24 838.856 Q1440.24 830.106 1443.29 825.523 Q1446.37 820.916 1452.18 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1482.34 824.62 Q1478.73 824.62 1476.9 828.185 Q1475.1 831.726 1475.1 838.856 Q1475.1 845.962 1476.9 849.527 Q1478.73 853.069 1482.34 853.069 Q1485.98 853.069 1487.78 849.527 Q1489.61 845.962 1489.61 838.856 Q1489.61 831.726 1487.78 828.185 Q1485.98 824.62 1482.34 824.62 M1482.34 820.916 Q1488.15 820.916 1491.21 825.523 Q1494.29 830.106 1494.29 838.856 Q1494.29 847.583 1491.21 852.189 Q1488.15 856.773 1482.34 856.773 Q1476.53 856.773 1473.45 852.189 Q1470.4 847.583 1470.4 838.856 Q1470.4 830.106 1473.45 825.523 Q1476.53 820.916 1482.34 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1512.5 824.62 Q1508.89 824.62 1507.06 828.185 Q1505.26 831.726 1505.26 838.856 Q1505.26 845.962 1507.06 849.527 Q1508.89 853.069 1512.5 853.069 Q1516.14 853.069 1517.94 849.527 Q1519.77 845.962 1519.77 838.856 Q1519.77 831.726 1517.94 828.185 Q1516.14 824.62 1512.5 824.62 M1512.5 820.916 Q1518.31 820.916 1521.37 825.523 Q1524.45 830.106 1524.45 838.856 Q1524.45 847.583 1521.37 852.189 Q1518.31 856.773 1512.5 856.773 Q1506.69 856.773 1503.61 852.189 Q1500.56 847.583 1500.56 838.856 Q1500.56 830.106 1503.61 825.523 Q1506.69 820.916 1512.5 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2016.36 852.166 L2024 852.166 L2024 825.8 L2015.69 827.467 L2015.69 823.208 L2023.95 821.541 L2028.63 821.541 L2028.63 852.166 L2036.26 852.166 L2036.26 856.101 L2016.36 856.101 L2016.36 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2045.75 821.541 L2064.11 821.541 L2064.11 825.476 L2050.04 825.476 L2050.04 833.949 Q2051.06 833.601 2052.07 833.439 Q2053.09 833.254 2054.11 833.254 Q2059.9 833.254 2063.28 836.425 Q2066.66 839.597 2066.66 845.013 Q2066.66 850.592 2063.19 853.694 Q2059.71 856.773 2053.39 856.773 Q2051.22 856.773 2048.95 856.402 Q2046.7 856.032 2044.3 855.291 L2044.3 850.592 Q2046.38 851.726 2048.6 852.282 Q2050.82 852.837 2053.3 852.837 Q2057.31 852.837 2059.64 850.731 Q2061.98 848.624 2061.98 845.013 Q2061.98 841.402 2059.64 839.296 Q2057.31 837.189 2053.3 837.189 Q2051.43 837.189 2049.55 837.606 Q2047.7 838.023 2045.75 838.902 L2045.75 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2085.87 824.62 Q2082.26 824.62 2080.43 828.185 Q2078.62 831.726 2078.62 838.856 Q2078.62 845.962 2080.43 849.527 Q2082.26 853.069 2085.87 853.069 Q2089.5 853.069 2091.31 849.527 Q2093.14 845.962 2093.14 838.856 Q2093.14 831.726 2091.31 828.185 Q2089.5 824.62 2085.87 824.62 M2085.87 820.916 Q2091.68 820.916 2094.74 825.523 Q2097.81 830.106 2097.81 838.856 Q2097.81 847.583 2094.74 852.189 Q2091.68 856.773 2085.87 856.773 Q2080.06 856.773 2076.98 852.189 Q2073.93 847.583 2073.93 838.856 Q2073.93 830.106 2076.98 825.523 Q2080.06 820.916 2085.87 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2116.03 824.62 Q2112.42 824.62 2110.59 828.185 Q2108.79 831.726 2108.79 838.856 Q2108.79 845.962 2110.59 849.527 Q2112.42 853.069 2116.03 853.069 Q2119.67 853.069 2121.47 849.527 Q2123.3 845.962 2123.3 838.856 Q2123.3 831.726 2121.47 828.185 Q2119.67 824.62 2116.03 824.62 M2116.03 820.916 Q2121.84 820.916 2124.9 825.523 Q2127.98 830.106 2127.98 838.856 Q2127.98 847.583 2124.9 852.189 Q2121.84 856.773 2116.03 856.773 Q2110.22 856.773 2107.14 852.189 Q2104.09 847.583 2104.09 838.856 Q2104.09 830.106 2107.14 825.523 Q2110.22 820.916 2116.03 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2623.97 852.166 L2640.29 852.166 L2640.29 856.101 L2618.35 856.101 L2618.35 852.166 Q2621.01 849.411 2625.59 844.782 Q2630.2 840.129 2631.38 838.786 Q2633.62 836.263 2634.5 834.527 Q2635.41 832.768 2635.41 831.078 Q2635.41 828.324 2633.46 826.587 Q2631.54 824.851 2628.44 824.851 Q2626.24 824.851 2623.79 825.615 Q2621.36 826.379 2618.58 827.93 L2618.58 823.208 Q2621.4 822.074 2623.86 821.495 Q2626.31 820.916 2628.35 820.916 Q2633.72 820.916 2636.91 823.601 Q2640.1 826.287 2640.1 830.777 Q2640.1 832.907 2639.29 834.828 Q2638.51 836.726 2636.4 839.319 Q2635.82 839.99 2632.72 843.208 Q2629.62 846.402 2623.97 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2660.1 824.62 Q2656.49 824.62 2654.67 828.185 Q2652.86 831.726 2652.86 838.856 Q2652.86 845.962 2654.67 849.527 Q2656.49 853.069 2660.1 853.069 Q2663.74 853.069 2665.54 849.527 Q2667.37 845.962 2667.37 838.856 Q2667.37 831.726 2665.54 828.185 Q2663.74 824.62 2660.1 824.62 M2660.1 820.916 Q2665.92 820.916 2668.97 825.523 Q2672.05 830.106 2672.05 838.856 Q2672.05 847.583 2668.97 852.189 Q2665.92 856.773 2660.1 856.773 Q2654.29 856.773 2651.22 852.189 Q2648.16 847.583 2648.16 838.856 Q2648.16 830.106 2651.22 825.523 Q2654.29 820.916 2660.1 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2690.27 824.62 Q2686.66 824.62 2684.83 828.185 Q2683.02 831.726 2683.02 838.856 Q2683.02 845.962 2684.83 849.527 Q2686.66 853.069 2690.27 853.069 Q2693.9 853.069 2695.71 849.527 Q2697.54 845.962 2697.54 838.856 Q2697.54 831.726 2695.71 828.185 Q2693.9 824.62 2690.27 824.62 M2690.27 820.916 Q2696.08 820.916 2699.13 825.523 Q2702.21 830.106 2702.21 838.856 Q2702.21 847.583 2699.13 852.189 Q2696.08 856.773 2690.27 856.773 Q2684.46 856.773 2681.38 852.189 Q2678.32 847.583 2678.32 838.856 Q2678.32 830.106 2681.38 825.523 Q2684.46 820.916 2690.27 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2720.43 824.62 Q2716.82 824.62 2714.99 828.185 Q2713.18 831.726 2713.18 838.856 Q2713.18 845.962 2714.99 849.527 Q2716.82 853.069 2720.43 853.069 Q2724.06 853.069 2725.87 849.527 Q2727.7 845.962 2727.7 838.856 Q2727.7 831.726 2725.87 828.185 Q2724.06 824.62 2720.43 824.62 M2720.43 820.916 Q2726.24 820.916 2729.29 825.523 Q2732.37 830.106 2732.37 838.856 Q2732.37 847.583 2729.29 852.189 Q2726.24 856.773 2720.43 856.773 Q2714.62 856.773 2711.54 852.189 Q2708.48 847.583 2708.48 838.856 Q2708.48 830.106 2711.54 825.523 Q2714.62 820.916 2720.43 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,786.533 2752.76,786.533 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,659.948 2752.76,659.948 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,533.363 2752.76,533.363 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,406.778 2752.76,406.778 "/>
+<polyline clip-path="url(#clip092)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,280.193 2752.76,280.193 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,790.501 188.677,153.712 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,786.533 207.574,786.533 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,659.948 207.574,659.948 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,533.363 207.574,533.363 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,406.778 207.574,406.778 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,280.193 207.574,280.193 "/>
+<path clip-path="url(#clip090)" d="M49.9875 799.878 L57.6263 799.878 L57.6263 773.513 L49.3162 775.179 L49.3162 770.92 L57.58 769.253 L62.2559 769.253 L62.2559 799.878 L69.8948 799.878 L69.8948 803.813 L49.9875 803.813 L49.9875 799.878 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M89.9178 784.67 Q86.7697 784.67 84.9179 786.823 Q83.0892 788.976 83.0892 792.726 Q83.0892 796.452 84.9179 798.628 Q86.7697 800.781 89.9178 800.781 Q93.066 800.781 94.8947 798.628 Q96.7465 796.452 96.7465 792.726 Q96.7465 788.976 94.8947 786.823 Q93.066 784.67 89.9178 784.67 M99.2002 770.017 L99.2002 774.277 Q97.4409 773.443 95.6354 773.003 Q93.853 772.564 92.0937 772.564 Q87.4641 772.564 85.0105 775.689 Q82.5799 778.814 82.2327 785.133 Q83.5984 783.119 85.6586 782.054 Q87.7188 780.966 90.1956 780.966 Q95.4039 780.966 98.4132 784.138 Q101.446 787.286 101.446 792.726 Q101.446 798.05 98.2974 801.267 Q95.1493 804.485 89.9178 804.485 Q83.9225 804.485 80.7512 799.901 Q77.5799 795.295 77.5799 786.568 Q77.5799 778.374 81.4688 773.513 Q85.3577 768.628 91.9086 768.628 Q93.6678 768.628 95.4502 768.976 Q97.2558 769.323 99.2002 770.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M109.501 797.934 L114.385 797.934 L114.385 803.813 L109.501 803.813 L109.501 797.934 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M135.149 784.67 Q132.001 784.67 130.149 786.823 Q128.32 788.976 128.32 792.726 Q128.32 796.452 130.149 798.628 Q132.001 800.781 135.149 800.781 Q138.297 800.781 140.126 798.628 Q141.978 796.452 141.978 792.726 Q141.978 788.976 140.126 786.823 Q138.297 784.67 135.149 784.67 M144.431 770.017 L144.431 774.277 Q142.672 773.443 140.867 773.003 Q139.084 772.564 137.325 772.564 Q132.695 772.564 130.242 775.689 Q127.811 778.814 127.464 785.133 Q128.83 783.119 130.89 782.054 Q132.95 780.966 135.427 780.966 Q140.635 780.966 143.644 784.138 Q146.677 787.286 146.677 792.726 Q146.677 798.05 143.529 801.267 Q140.381 804.485 135.149 804.485 Q129.154 804.485 125.982 799.901 Q122.811 795.295 122.811 786.568 Q122.811 778.374 126.7 773.513 Q130.589 768.628 137.14 768.628 Q138.899 768.628 140.681 768.976 Q142.487 769.323 144.431 770.017 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M50.2421 673.293 L57.881 673.293 L57.881 646.928 L49.5708 648.594 L49.5708 644.335 L57.8347 642.668 L62.5106 642.668 L62.5106 673.293 L70.1494 673.293 L70.1494 677.228 L50.2421 677.228 L50.2421 673.293 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M90.1725 658.085 Q87.0243 658.085 85.1725 660.238 Q83.3438 662.39 83.3438 666.14 Q83.3438 669.867 85.1725 672.043 Q87.0243 674.196 90.1725 674.196 Q93.3206 674.196 95.1493 672.043 Q97.0011 669.867 97.0011 666.14 Q97.0011 662.39 95.1493 660.238 Q93.3206 658.085 90.1725 658.085 M99.4548 643.432 L99.4548 647.691 Q97.6956 646.858 95.89 646.418 Q94.1076 645.978 92.3484 645.978 Q87.7188 645.978 85.2651 649.103 Q82.8345 652.228 82.4873 658.548 Q83.8531 656.534 85.9132 655.469 Q87.9734 654.381 90.4502 654.381 Q95.6585 654.381 98.6678 657.553 Q101.7 660.701 101.7 666.14 Q101.7 671.464 98.552 674.682 Q95.4039 677.9 90.1725 677.9 Q84.1771 677.9 81.0058 673.316 Q77.8346 668.71 77.8346 659.983 Q77.8346 651.789 81.7234 646.928 Q85.6123 642.043 92.1632 642.043 Q93.9224 642.043 95.7048 642.391 Q97.5104 642.738 99.4548 643.432 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M109.756 671.349 L114.64 671.349 L114.64 677.228 L109.756 677.228 L109.756 671.349 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.825 660.816 Q131.492 660.816 129.57 662.599 Q127.672 664.381 127.672 667.506 Q127.672 670.631 129.57 672.414 Q131.492 674.196 134.825 674.196 Q138.158 674.196 140.08 672.414 Q142.001 670.608 142.001 667.506 Q142.001 664.381 140.08 662.599 Q138.181 660.816 134.825 660.816 M130.149 658.826 Q127.14 658.085 125.45 656.025 Q123.783 653.965 123.783 651.002 Q123.783 646.858 126.723 644.451 Q129.686 642.043 134.825 642.043 Q139.987 642.043 142.927 644.451 Q145.867 646.858 145.867 651.002 Q145.867 653.965 144.177 656.025 Q142.51 658.085 139.524 658.826 Q142.904 659.613 144.779 661.904 Q146.677 664.196 146.677 667.506 Q146.677 672.529 143.598 675.214 Q140.543 677.9 134.825 677.9 Q129.107 677.9 126.029 675.214 Q122.973 672.529 122.973 667.506 Q122.973 664.196 124.871 661.904 Q126.769 659.613 130.149 658.826 M128.436 651.441 Q128.436 654.127 130.103 655.631 Q131.793 657.136 134.825 657.136 Q137.834 657.136 139.524 655.631 Q141.237 654.127 141.237 651.441 Q141.237 648.756 139.524 647.252 Q137.834 645.747 134.825 645.747 Q131.793 645.747 130.103 647.252 Q128.436 648.756 128.436 651.441 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M50.1495 546.708 L57.7884 546.708 L57.7884 520.342 L49.4782 522.009 L49.4782 517.75 L57.7421 516.083 L62.418 516.083 L62.418 546.708 L70.0568 546.708 L70.0568 550.643 L50.1495 550.643 L50.1495 546.708 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M78.3207 516.083 L100.543 516.083 L100.543 518.074 L87.9965 550.643 L83.1123 550.643 L94.9178 520.018 L78.3207 520.018 L78.3207 516.083 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M109.663 544.764 L114.547 544.764 L114.547 550.643 L109.663 550.643 L109.663 544.764 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 519.162 Q131.121 519.162 129.293 522.727 Q127.487 526.268 127.487 533.398 Q127.487 540.504 129.293 544.069 Q131.121 547.611 134.732 547.611 Q138.367 547.611 140.172 544.069 Q142.001 540.504 142.001 533.398 Q142.001 526.268 140.172 522.727 Q138.367 519.162 134.732 519.162 M134.732 515.458 Q140.543 515.458 143.598 520.065 Q146.677 524.648 146.677 533.398 Q146.677 542.125 143.598 546.731 Q140.543 551.315 134.732 551.315 Q128.922 551.315 125.844 546.731 Q122.788 542.125 122.788 533.398 Q122.788 524.648 125.844 520.065 Q128.922 515.458 134.732 515.458 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M51.7467 420.123 L59.3856 420.123 L59.3856 393.757 L51.0754 395.424 L51.0754 391.165 L59.3393 389.498 L64.0152 389.498 L64.0152 420.123 L71.654 420.123 L71.654 424.058 L51.7467 424.058 L51.7467 420.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M79.9179 389.498 L102.14 389.498 L102.14 391.489 L89.5938 424.058 L84.7095 424.058 L96.515 393.433 L79.9179 393.433 L79.9179 389.498 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M111.26 418.179 L116.145 418.179 L116.145 424.058 L111.26 424.058 L111.26 418.179 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M130.357 420.123 L146.677 420.123 L146.677 424.058 L124.732 424.058 L124.732 420.123 Q127.394 417.368 131.978 412.739 Q136.584 408.086 137.765 406.743 Q140.01 404.22 140.89 402.484 Q141.793 400.725 141.793 399.035 Q141.793 396.28 139.848 394.544 Q137.927 392.808 134.825 392.808 Q132.626 392.808 130.172 393.572 Q127.742 394.336 124.964 395.887 L124.964 391.165 Q127.788 390.031 130.242 389.452 Q132.695 388.873 134.732 388.873 Q140.103 388.873 143.297 391.558 Q146.492 394.243 146.492 398.734 Q146.492 400.864 145.681 402.785 Q144.894 404.683 142.788 407.276 Q142.209 407.947 139.107 411.165 Q136.006 414.359 130.357 420.123 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M49.6634 293.538 L57.3023 293.538 L57.3023 267.172 L48.9921 268.839 L48.9921 264.58 L57.256 262.913 L61.9319 262.913 L61.9319 293.538 L69.5707 293.538 L69.5707 297.473 L49.6634 297.473 L49.6634 293.538 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M77.8346 262.913 L100.057 262.913 L100.057 264.904 L87.5104 297.473 L82.6262 297.473 L94.4317 266.848 L77.8346 266.848 L77.8346 262.913 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M109.177 291.593 L114.061 291.593 L114.061 297.473 L109.177 297.473 L109.177 291.593 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M137.094 266.987 L125.288 285.436 L137.094 285.436 L137.094 266.987 M135.867 262.913 L141.746 262.913 L141.746 285.436 L146.677 285.436 L146.677 289.325 L141.746 289.325 L141.746 297.473 L137.094 297.473 L137.094 289.325 L121.492 289.325 L121.492 284.811 L135.867 262.913 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1061.76 20.1573 L1050.66 50.2555 L1072.9 50.2555 L1061.76 20.1573 M1057.14 12.096 L1066.42 12.096 L1089.47 72.576 L1080.96 72.576 L1075.45 57.061 L1048.19 57.061 L1042.68 72.576 L1034.05 72.576 L1057.14 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1135.69 45.1919 L1135.69 72.576 L1128.23 72.576 L1128.23 45.4349 Q1128.23 38.994 1125.72 35.7938 Q1123.21 32.5936 1118.19 32.5936 Q1112.15 32.5936 1108.67 36.4419 Q1105.18 40.2903 1105.18 46.9338 L1105.18 72.576 L1097.69 72.576 L1097.69 27.2059 L1105.18 27.2059 L1105.18 34.2544 Q1107.86 30.163 1111.46 28.1376 Q1115.11 26.1121 1119.85 26.1121 Q1127.67 26.1121 1131.68 30.9732 Q1135.69 35.7938 1135.69 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1180.41 49.3643 Q1180.41 41.2625 1177.05 36.8065 Q1173.72 32.3505 1167.69 32.3505 Q1161.69 32.3505 1158.33 36.8065 Q1155.01 41.2625 1155.01 49.3643 Q1155.01 57.4256 1158.33 61.8816 Q1161.69 66.3376 1167.69 66.3376 Q1173.72 66.3376 1177.05 61.8816 Q1180.41 57.4256 1180.41 49.3643 M1187.86 66.9452 Q1187.86 78.5308 1182.72 84.1616 Q1177.57 89.8329 1166.96 89.8329 Q1163.03 89.8329 1159.55 89.2252 Q1156.06 88.6581 1152.78 87.4428 L1152.78 80.1917 Q1156.06 81.9741 1159.26 82.8248 Q1162.46 83.6755 1165.78 83.6755 Q1173.12 83.6755 1176.76 79.8271 Q1180.41 76.0193 1180.41 68.282 L1180.41 64.5957 Q1178.1 68.6061 1174.49 70.5911 Q1170.89 72.576 1165.87 72.576 Q1157.52 72.576 1152.42 66.2161 Q1147.31 59.8562 1147.31 49.3643 Q1147.31 38.832 1152.42 32.472 Q1157.52 26.1121 1165.87 26.1121 Q1170.89 26.1121 1174.49 28.0971 Q1178.1 30.082 1180.41 34.0924 L1180.41 27.2059 L1187.86 27.2059 L1187.86 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1203.21 9.54393 L1210.67 9.54393 L1210.67 72.576 L1203.21 72.576 L1203.21 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1265.07 48.0275 L1265.07 51.6733 L1230.8 51.6733 Q1231.29 59.3701 1235.42 63.421 Q1239.59 67.4314 1247.01 67.4314 Q1251.3 67.4314 1255.31 66.3781 Q1259.36 65.3249 1263.33 63.2184 L1263.33 70.267 Q1259.32 71.9684 1255.11 72.8596 Q1250.89 73.7508 1246.56 73.7508 Q1235.7 73.7508 1229.34 67.4314 Q1223.02 61.1119 1223.02 50.3365 Q1223.02 39.1965 1229.02 32.6746 Q1235.05 26.1121 1245.26 26.1121 Q1254.42 26.1121 1259.73 32.0264 Q1265.07 37.9003 1265.07 48.0275 M1257.62 45.84 Q1257.54 39.7232 1254.18 36.0774 Q1250.85 32.4315 1245.34 32.4315 Q1239.11 32.4315 1235.34 35.9558 Q1231.61 39.4801 1231.04 45.8805 L1257.62 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1321.26 32.4315 Q1315.26 32.4315 1311.78 37.1306 Q1308.3 41.7891 1308.3 49.9314 Q1308.3 58.0738 1311.74 62.7728 Q1315.22 67.4314 1321.26 67.4314 Q1327.21 67.4314 1330.7 62.7323 Q1334.18 58.0333 1334.18 49.9314 Q1334.18 41.8701 1330.7 37.1711 Q1327.21 32.4315 1321.26 32.4315 M1321.26 26.1121 Q1330.98 26.1121 1336.53 32.4315 Q1342.08 38.7509 1342.08 49.9314 Q1342.08 61.0714 1336.53 67.4314 Q1330.98 73.7508 1321.26 73.7508 Q1311.5 73.7508 1305.95 67.4314 Q1300.44 61.0714 1300.44 49.9314 Q1300.44 38.7509 1305.95 32.4315 Q1311.5 26.1121 1321.26 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1377.4 9.54393 L1377.4 15.7418 L1370.27 15.7418 Q1366.26 15.7418 1364.68 17.3622 Q1363.14 18.9825 1363.14 23.1955 L1363.14 27.2059 L1375.42 27.2059 L1375.42 32.9987 L1363.14 32.9987 L1363.14 72.576 L1355.65 72.576 L1355.65 32.9987 L1348.52 32.9987 L1348.52 27.2059 L1355.65 27.2059 L1355.65 24.0462 Q1355.65 16.471 1359.17 13.0277 Q1362.7 9.54393 1370.36 9.54393 L1377.4 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1430.55 20.1573 L1419.45 50.2555 L1441.69 50.2555 L1430.55 20.1573 M1425.93 12.096 L1435.21 12.096 L1458.26 72.576 L1449.75 72.576 L1444.24 57.061 L1416.98 57.061 L1411.47 72.576 L1402.84 72.576 L1425.93 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1472.68 14.324 L1472.68 27.2059 L1488.03 27.2059 L1488.03 32.9987 L1472.68 32.9987 L1472.68 57.6282 Q1472.68 63.1779 1474.18 64.7578 Q1475.72 66.3376 1480.38 66.3376 L1488.03 66.3376 L1488.03 72.576 L1480.38 72.576 Q1471.75 72.576 1468.47 69.3758 Q1465.19 66.1351 1465.19 57.6282 L1465.19 32.9987 L1459.72 32.9987 L1459.72 27.2059 L1465.19 27.2059 L1465.19 14.324 L1472.68 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1505.21 14.324 L1505.21 27.2059 L1520.56 27.2059 L1520.56 32.9987 L1505.21 32.9987 L1505.21 57.6282 Q1505.21 63.1779 1506.71 64.7578 Q1508.25 66.3376 1512.91 66.3376 L1520.56 66.3376 L1520.56 72.576 L1512.91 72.576 Q1504.28 72.576 1501 69.3758 Q1497.72 66.1351 1497.72 57.6282 L1497.72 32.9987 L1492.25 32.9987 L1492.25 27.2059 L1497.72 27.2059 L1497.72 14.324 L1505.21 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1550.99 49.7694 Q1541.95 49.7694 1538.47 51.8354 Q1534.98 53.9013 1534.98 58.8839 Q1534.98 62.8538 1537.58 65.2034 Q1540.21 67.5124 1544.71 67.5124 Q1550.9 67.5124 1554.63 63.1374 Q1558.4 58.7219 1558.4 51.4303 L1558.4 49.7694 L1550.99 49.7694 M1565.85 46.6907 L1565.85 72.576 L1558.4 72.576 L1558.4 65.6895 Q1555.85 69.8214 1552.04 71.8063 Q1548.23 73.7508 1542.72 73.7508 Q1535.75 73.7508 1531.62 69.8619 Q1527.53 65.9325 1527.53 59.3701 Q1527.53 51.7138 1532.63 47.825 Q1537.78 43.9361 1547.95 43.9361 L1558.4 43.9361 L1558.4 43.2069 Q1558.4 38.0623 1555 35.2672 Q1551.63 32.4315 1545.52 32.4315 Q1541.63 32.4315 1537.94 33.3632 Q1534.25 34.295 1530.85 36.1584 L1530.85 29.2718 Q1534.94 27.692 1538.79 26.9223 Q1542.64 26.1121 1546.29 26.1121 Q1556.13 26.1121 1560.99 31.2163 Q1565.85 36.3204 1565.85 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1613.86 28.9478 L1613.86 35.9153 Q1610.7 34.1734 1607.5 33.3227 Q1604.34 32.4315 1601.09 32.4315 Q1593.84 32.4315 1589.83 37.0496 Q1585.82 41.6271 1585.82 49.9314 Q1585.82 58.2358 1589.83 62.8538 Q1593.84 67.4314 1601.09 67.4314 Q1604.34 67.4314 1607.5 66.5807 Q1610.7 65.6895 1613.86 63.9476 L1613.86 70.8341 Q1610.74 72.2924 1607.37 73.0216 Q1604.05 73.7508 1600.28 73.7508 Q1590.04 73.7508 1584 67.3098 Q1577.96 60.8689 1577.96 49.9314 Q1577.96 38.832 1584.04 32.472 Q1590.16 26.1121 1600.77 26.1121 Q1604.21 26.1121 1607.5 26.8413 Q1610.78 27.5299 1613.86 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1626.53 9.54393 L1634.03 9.54393 L1634.03 46.7717 L1656.27 27.2059 L1665.79 27.2059 L1641.73 48.4326 L1666.8 72.576 L1657.08 72.576 L1634.03 50.4176 L1634.03 72.576 L1626.53 72.576 L1626.53 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1719.14 9.62495 Q1713.71 18.942 1711.08 28.0566 Q1708.44 37.1711 1708.44 46.5287 Q1708.44 55.8863 1711.08 65.0818 Q1713.75 74.2369 1719.14 83.5134 L1712.66 83.5134 Q1706.58 73.9938 1703.54 64.7983 Q1700.54 55.6027 1700.54 46.5287 Q1700.54 37.4952 1703.54 28.3401 Q1706.54 19.1851 1712.66 9.62495 L1719.14 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1763.45 34.0924 L1763.45 9.54393 L1770.91 9.54393 L1770.91 72.576 L1763.45 72.576 L1763.45 65.7705 Q1761.11 69.8214 1757.5 71.8063 Q1753.94 73.7508 1748.91 73.7508 Q1740.69 73.7508 1735.5 67.1883 Q1730.36 60.6258 1730.36 49.9314 Q1730.36 39.2371 1735.5 32.6746 Q1740.69 26.1121 1748.91 26.1121 Q1753.94 26.1121 1757.5 28.0971 Q1761.11 30.0415 1763.45 34.0924 M1738.06 49.9314 Q1738.06 58.1548 1741.42 62.8538 Q1744.82 67.5124 1750.74 67.5124 Q1756.65 67.5124 1760.05 62.8538 Q1763.45 58.1548 1763.45 49.9314 Q1763.45 41.7081 1760.05 37.0496 Q1756.65 32.3505 1750.74 32.3505 Q1744.82 32.3505 1741.42 37.0496 Q1738.06 41.7081 1738.06 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1825.07 48.0275 L1825.07 51.6733 L1790.8 51.6733 Q1791.28 59.3701 1795.42 63.421 Q1799.59 67.4314 1807 67.4314 Q1811.3 67.4314 1815.31 66.3781 Q1819.36 65.3249 1823.33 63.2184 L1823.33 70.267 Q1819.32 71.9684 1815.1 72.8596 Q1810.89 73.7508 1806.56 73.7508 Q1795.7 73.7508 1789.34 67.4314 Q1783.02 61.1119 1783.02 50.3365 Q1783.02 39.1965 1789.02 32.6746 Q1795.05 26.1121 1805.26 26.1121 Q1814.42 26.1121 1819.72 32.0264 Q1825.07 37.9003 1825.07 48.0275 M1817.62 45.84 Q1817.53 39.7232 1814.17 36.0774 Q1810.85 32.4315 1805.34 32.4315 Q1799.1 32.4315 1795.34 35.9558 Q1791.61 39.4801 1791.04 45.8805 L1817.62 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1867.16 49.3643 Q1867.16 41.2625 1863.8 36.8065 Q1860.47 32.3505 1854.44 32.3505 Q1848.44 32.3505 1845.08 36.8065 Q1841.76 41.2625 1841.76 49.3643 Q1841.76 57.4256 1845.08 61.8816 Q1848.44 66.3376 1854.44 66.3376 Q1860.47 66.3376 1863.8 61.8816 Q1867.16 57.4256 1867.16 49.3643 M1874.61 66.9452 Q1874.61 78.5308 1869.47 84.1616 Q1864.32 89.8329 1853.71 89.8329 Q1849.78 89.8329 1846.3 89.2252 Q1842.81 88.6581 1839.53 87.4428 L1839.53 80.1917 Q1842.81 81.9741 1846.01 82.8248 Q1849.21 83.6755 1852.53 83.6755 Q1859.87 83.6755 1863.51 79.8271 Q1867.16 76.0193 1867.16 68.282 L1867.16 64.5957 Q1864.85 68.6061 1861.24 70.5911 Q1857.64 72.576 1852.62 72.576 Q1844.27 72.576 1839.17 66.2161 Q1834.06 59.8562 1834.06 49.3643 Q1834.06 38.832 1839.17 32.472 Q1844.27 26.1121 1852.62 26.1121 Q1857.64 26.1121 1861.24 28.0971 Q1864.85 30.082 1867.16 34.0924 L1867.16 27.2059 L1874.61 27.2059 L1874.61 66.9452 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1888.79 9.62495 L1895.27 9.62495 Q1901.35 19.1851 1904.35 28.3401 Q1907.38 37.4952 1907.38 46.5287 Q1907.38 55.6027 1904.35 64.7983 Q1901.35 73.9938 1895.27 83.5134 L1888.79 83.5134 Q1894.18 74.2369 1896.81 65.0818 Q1899.48 55.8863 1899.48 46.5287 Q1899.48 37.1711 1896.81 28.0566 Q1894.18 18.942 1888.79 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip092)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,237.832 266.073,234.59 270.902,231.464 275.73,228.457 280.558,225.572 285.386,222.813 290.214,220.184 295.043,217.689 299.871,215.334 304.699,213.124 309.527,211.063 314.356,209.159 319.184,207.418 324.012,205.846 328.84,204.451 333.668,203.24 338.497,202.221 343.325,201.402 348.153,200.792 352.981,200.398 357.81,200.23 362.638,200.295 367.466,200.603 372.294,201.16 377.123,201.973 381.951,203.05 386.779,204.396 391.607,206.015 396.435,207.91 401.264,210.082 406.092,212.531 410.92,215.255 415.748,218.248 420.577,221.504 425.405,225.012 430.233,228.762 435.061,232.738 439.89,236.924 444.718,241.298 449.546,245.841 454.374,250.529 459.202,255.335 464.031,260.234 468.859,265.197 473.687,270.197 478.515,275.206 483.344,280.195 488.172,285.138 493,290.007 497.828,294.777 502.656,299.426 507.485,303.93 512.313,308.271 517.141,312.43 521.969,316.391 526.798,320.14 531.626,323.665 536.454,326.957 541.282,330.006 546.111,332.806 550.939,335.352 555.767,337.641 560.595,339.669 565.423,341.435 570.252,342.94 575.08,344.183 579.908,345.166 584.736,345.89 589.565,346.358 594.393,346.573 599.221,346.537 604.049,346.255 608.878,345.73 613.706,344.967 618.534,343.968 623.362,342.74 628.19,341.286 633.019,339.612 637.847,337.721 642.675,335.619 647.503,333.311 652.332,330.802 657.16,328.098 661.988,325.204 666.816,322.126 671.645,318.87 676.473,315.442 681.301,311.849 686.129,308.096 690.957,304.192 695.786,300.143 700.614,295.958 705.442,291.644 710.27,287.21 715.099,282.665 719.927,278.018 724.755,273.281 729.583,268.464 734.411,263.578 739.24,258.636 744.068,253.651 748.896,248.637 753.724,243.608 758.553,238.581 763.381,233.572 768.209,228.599 773.037,223.682 777.866,218.839 782.694,214.093 787.522,209.465 792.35,204.978 797.178,200.658 802.007,196.53 806.835,192.62 811.663,188.956 816.491,185.566 821.32,182.478 826.148,179.721 830.976,177.325 835.804,175.319 840.633,173.73 845.461,172.586 850.289,171.913 855.117,171.734 859.945,172.072 864.774,172.944 869.602,174.365 874.43,176.346 879.258,178.891 884.087,182.002 888.915,185.672 893.743,189.889 898.571,194.636 903.399,199.888 908.228,205.612 913.056,211.771 917.884,218.319 922.712,225.208 927.541,232.381 932.369,239.779 937.197,247.339 942.025,254.996 946.854,262.682 951.682,270.331 956.51,277.875 961.338,285.25 966.166,292.394 970.995,299.247 975.823,305.754 980.651,311.865 985.479,317.535 990.308,322.723 995.136,327.395 999.964,331.522 1004.79,335.08 1009.62,338.05 1014.45,340.42 1019.28,342.181 1024.11,343.329 1028.93,343.865 1033.76,343.795 1038.59,343.125 1043.42,341.868 1048.25,340.039 1053.07,337.656 1057.9,334.739 1062.73,331.312 1067.56,327.399 1072.39,323.028 1077.22,318.23 1082.04,313.035 1086.87,307.476 1091.7,301.59 1096.53,295.414 1101.36,288.986 1106.19,282.346 1111.01,275.537 1115.84,268.604 1120.67,261.591 1125.5,254.545 1130.33,247.515 1135.15,240.552 1139.98,233.706 1144.81,227.029 1149.64,220.575 1154.47,214.398 1159.3,208.55 1164.12,203.087 1168.95,198.06 1173.78,193.523 1178.61,189.524 1183.44,186.112 1188.26,183.331 1193.09,181.221 1197.92,179.818 1202.75,179.152 1207.58,179.245 1212.41,180.113 1217.23,181.763 1222.06,184.194 1226.89,187.395 1231.72,191.343 1236.55,196.007 1241.38,201.346 1246.2,207.308 1251.03,213.831 1255.86,220.845 1260.69,228.271 1265.52,236.026 1270.34,244.018 1275.17,252.154 1280,260.337 1284.83,268.471 1289.66,276.461 1294.49,284.212 1299.31,291.637 1304.14,298.652 1308.97,305.18 1313.8,311.151 1318.63,316.505 1323.46,321.189 1328.28,325.159 1333.11,328.382 1337.94,330.833 1342.77,332.496 1347.6,333.364 1352.42,333.44 1357.25,332.731 1362.08,331.257 1366.91,329.042 1371.74,326.116 1376.57,322.517 1381.39,318.289 1386.22,313.48 1391.05,308.142 1395.88,302.334 1400.71,296.118 1405.54,289.557 1410.36,282.72 1415.19,275.678 1420.02,268.504 1424.85,261.273 1429.68,254.061 1434.5,246.946 1439.33,240.003 1444.16,233.311 1448.99,226.944 1453.82,220.978 1458.65,215.482 1463.47,210.524 1468.3,206.169 1473.13,202.473 1477.96,199.488 1482.79,197.258 1487.62,195.818 1492.44,195.193 1497.27,195.401 1502.1,196.444 1506.93,198.316 1511.76,200.999 1516.58,204.46 1521.41,208.657 1526.24,213.534 1531.07,219.026 1535.9,225.055 1540.73,231.537 1545.55,238.379 1550.38,245.481 1555.21,252.741 1560.04,260.053 1564.87,267.312 1569.7,274.414 1574.52,281.258 1579.35,287.748 1584.18,293.797 1589.01,299.323 1593.84,304.256 1598.66,308.532 1603.49,312.103 1608.32,314.929 1613.15,316.981 1617.98,318.244 1622.81,318.711 1627.63,318.388 1632.46,317.291 1637.29,315.447 1642.12,312.89 1646.95,309.665 1651.77,305.824 1656.6,301.424 1661.43,296.533 1666.26,291.221 1671.09,285.563 1675.92,279.639 1680.74,273.531 1685.57,267.325 1690.4,261.106 1695.23,254.96 1700.06,248.973 1704.89,243.227 1709.71,237.804 1714.54,232.78 1719.37,228.227 1724.2,224.21 1729.03,220.79 1733.85,218.015 1738.68,215.928 1743.51,214.559 1748.34,213.929 1753.17,214.048 1758,214.911 1762.82,216.505 1767.65,218.803 1772.48,221.765 1777.31,225.341 1782.14,229.471 1786.97,234.084 1791.79,239.102 1796.62,244.439 1801.45,250.005 1806.28,255.705 1811.11,261.445 1815.93,267.129 1820.76,272.664 1825.59,277.96 1830.42,282.935 1835.25,287.511 1840.08,291.621 1844.9,295.203 1849.73,298.211 1854.56,300.605 1859.39,302.357 1864.22,303.452 1869.05,303.885 1873.87,303.661 1878.7,302.796 1883.53,301.318 1888.36,299.262 1893.19,296.671 1898.01,293.598 1902.84,290.101 1907.67,286.244 1912.5,282.096 1917.33,277.73 1922.16,273.221 1926.98,268.645 1931.81,264.08 1936.64,259.599 1941.47,255.279 1946.3,251.188 1951.13,247.394 1955.95,243.958 1960.78,240.933 1965.61,238.369 1970.44,236.303 1975.27,234.767 1980.09,233.781 1984.92,233.357 1989.75,233.496 1994.58,234.19 1999.41,235.418 2004.24,237.152 2009.06,239.356 2013.89,241.982 2018.72,244.977 2023.55,248.282 2028.38,251.831 2033.2,255.558 2038.03,259.391 2042.86,263.259 2047.69,267.093 2052.52,270.824 2057.35,274.389 2062.17,277.727 2067,280.785 2071.83,283.517 2076.66,285.884 2081.49,287.854 2086.32,289.405 2091.14,290.523 2095.97,291.203 2100.8,291.449 2105.63,291.272 2110.46,290.69 2115.28,289.73 2120.11,288.425 2124.94,286.811 2129.77,284.932 2134.6,282.832 2139.43,280.562 2144.25,278.172 2149.08,275.711 2153.91,273.233 2158.74,270.785 2163.57,268.415 2168.4,266.169 2173.22,264.086 2178.05,262.202 2182.88,260.549 2187.71,259.151 2192.54,258.028 2197.36,257.193 2202.19,256.652 2207.02,256.405 2211.85,256.446 2216.68,256.763 2221.51,257.339 2226.33,258.152 2231.16,259.175 2235.99,260.379 2240.82,261.732 2245.65,263.201 2250.48,264.752 2255.3,266.354 2260.13,267.973 2264.96,269.581 2269.79,271.152 2274.62,272.665 2279.44,274.1 2284.27,275.446 2289.1,276.694 2293.93,277.841 2298.76,278.89 2303.59,279.848 2308.41,280.725 2313.24,281.537 2318.07,282.302 2322.9,283.041 2327.73,283.775 2332.56,284.528 2337.38,285.321 2342.21,286.177 2347.04,287.113 2351.87,288.145 2356.7,289.285 2361.52,290.538 2366.35,291.907 2371.18,293.386 2376.01,294.964 2380.84,296.621 2385.67,298.333 2390.49,300.069 2395.32,301.789 2400.15,303.452 2404.98,305.009 2409.81,306.409 2414.64,307.597 2419.46,308.52 2424.29,309.124 2429.12,309.359 2433.95,309.177 2438.78,308.542 2443.6,307.422 2448.43,305.799 2453.26,303.666 2458.09,301.032 2462.92,297.923 2467.75,294.38 2472.57,290.467 2477.4,286.264 2482.23,281.873 2487.06,277.417 2491.89,273.036 2496.71,268.893 2501.54,265.164 2506.37,262.042 2511.2,259.735 2516.03,258.459 2520.86,258.437 2525.68,259.896 2530.51,263.059 2535.34,268.144 2540.17,275.355 2545,284.877 2549.83,296.873 2554.65,311.468 2559.48,328.752 2564.31,348.763 2569.14,371.485 2573.97,396.837 2578.79,424.666 2583.62,454.736 2588.45,486.725 2593.28,520.214 2598.11,554.686 2602.94,589.516 2607.76,623.975 2612.59,657.227 2617.42,688.332 2622.25,716.25 2627.08,739.855 2631.91,757.944 2636.73,769.253 2641.56,772.479 2646.39,766.3 2651.22,749.403 2656.05,720.513 2660.87,678.419 2665.7,622.005 2670.53,550.282 2675.36,462.415 2680.19,357.751 "/>
+<path clip-path="url(#clip090)" d="M188.677 1723.83 L2752.76 1723.83 L2752.76 1087.05 L188.677 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip273">
+  <clipPath id="clip093">
     <rect x="188" y="1087" width="2565" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,1723.83 261.245,1087.05 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="864.774,1723.83 864.774,1087.05 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1468.3,1723.83 1468.3,1087.05 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2071.83,1723.83 2071.83,1087.05 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2675.36,1723.83 2675.36,1087.05 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1723.83 2752.76,1723.83 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,1723.83 261.245,1704.94 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="864.774,1723.83 864.774,1704.94 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1468.3,1723.83 1468.3,1704.94 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2071.83,1723.83 2071.83,1704.94 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2675.36,1723.83 2675.36,1704.94 "/>
-<path clip-path="url(#clip270)" d="M261.245 1757.95 Q257.634 1757.95 255.805 1761.52 Q254 1765.06 254 1772.19 Q254 1779.3 255.805 1782.86 Q257.634 1786.4 261.245 1786.4 Q264.879 1786.4 266.685 1782.86 Q268.514 1779.3 268.514 1772.19 Q268.514 1765.06 266.685 1761.52 Q264.879 1757.95 261.245 1757.95 M261.245 1754.25 Q267.055 1754.25 270.111 1758.86 Q273.189 1763.44 273.189 1772.19 Q273.189 1780.92 270.111 1785.52 Q267.055 1790.11 261.245 1790.11 Q255.435 1790.11 252.356 1785.52 Q249.301 1780.92 249.301 1772.19 Q249.301 1763.44 252.356 1758.86 Q255.435 1754.25 261.245 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M824.392 1754.87 L842.748 1754.87 L842.748 1758.81 L828.674 1758.81 L828.674 1767.28 Q829.693 1766.93 830.711 1766.77 Q831.73 1766.59 832.748 1766.59 Q838.535 1766.59 841.915 1769.76 Q845.295 1772.93 845.295 1778.35 Q845.295 1783.93 841.822 1787.03 Q838.35 1790.11 832.031 1790.11 Q829.855 1790.11 827.586 1789.74 Q825.341 1789.37 822.934 1788.62 L822.934 1783.93 Q825.017 1785.06 827.239 1785.62 Q829.461 1786.17 831.938 1786.17 Q835.943 1786.17 838.281 1784.06 Q840.619 1781.96 840.619 1778.35 Q840.619 1774.74 838.281 1772.63 Q835.943 1770.52 831.938 1770.52 Q830.063 1770.52 828.188 1770.94 Q826.336 1771.36 824.392 1772.24 L824.392 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M864.507 1757.95 Q860.896 1757.95 859.068 1761.52 Q857.262 1765.06 857.262 1772.19 Q857.262 1779.3 859.068 1782.86 Q860.896 1786.4 864.507 1786.4 Q868.142 1786.4 869.947 1782.86 Q871.776 1779.3 871.776 1772.19 Q871.776 1765.06 869.947 1761.52 Q868.142 1757.95 864.507 1757.95 M864.507 1754.25 Q870.318 1754.25 873.373 1758.86 Q876.452 1763.44 876.452 1772.19 Q876.452 1780.92 873.373 1785.52 Q870.318 1790.11 864.507 1790.11 Q858.697 1790.11 855.619 1785.52 Q852.563 1780.92 852.563 1772.19 Q852.563 1763.44 855.619 1758.86 Q858.697 1754.25 864.507 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M894.669 1757.95 Q891.058 1757.95 889.23 1761.52 Q887.424 1765.06 887.424 1772.19 Q887.424 1779.3 889.23 1782.86 Q891.058 1786.4 894.669 1786.4 Q898.304 1786.4 900.109 1782.86 Q901.938 1779.3 901.938 1772.19 Q901.938 1765.06 900.109 1761.52 Q898.304 1757.95 894.669 1757.95 M894.669 1754.25 Q900.479 1754.25 903.535 1758.86 Q906.614 1763.44 906.614 1772.19 Q906.614 1780.92 903.535 1785.52 Q900.479 1790.11 894.669 1790.11 Q888.859 1790.11 885.78 1785.52 Q882.725 1780.92 882.725 1772.19 Q882.725 1763.44 885.78 1758.86 Q888.859 1754.25 894.669 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1412.83 1785.5 L1420.47 1785.5 L1420.47 1759.13 L1412.16 1760.8 L1412.16 1756.54 L1420.42 1754.87 L1425.1 1754.87 L1425.1 1785.5 L1432.74 1785.5 L1432.74 1789.43 L1412.83 1789.43 L1412.83 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1452.18 1757.95 Q1448.57 1757.95 1446.74 1761.52 Q1444.93 1765.06 1444.93 1772.19 Q1444.93 1779.3 1446.74 1782.86 Q1448.57 1786.4 1452.18 1786.4 Q1455.81 1786.4 1457.62 1782.86 Q1459.45 1779.3 1459.45 1772.19 Q1459.45 1765.06 1457.62 1761.52 Q1455.81 1757.95 1452.18 1757.95 M1452.18 1754.25 Q1457.99 1754.25 1461.05 1758.86 Q1464.12 1763.44 1464.12 1772.19 Q1464.12 1780.92 1461.05 1785.52 Q1457.99 1790.11 1452.18 1790.11 Q1446.37 1790.11 1443.29 1785.52 Q1440.24 1780.92 1440.24 1772.19 Q1440.24 1763.44 1443.29 1758.86 Q1446.37 1754.25 1452.18 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1482.34 1757.95 Q1478.73 1757.95 1476.9 1761.52 Q1475.1 1765.06 1475.1 1772.19 Q1475.1 1779.3 1476.9 1782.86 Q1478.73 1786.4 1482.34 1786.4 Q1485.98 1786.4 1487.78 1782.86 Q1489.61 1779.3 1489.61 1772.19 Q1489.61 1765.06 1487.78 1761.52 Q1485.98 1757.95 1482.34 1757.95 M1482.34 1754.25 Q1488.15 1754.25 1491.21 1758.86 Q1494.29 1763.44 1494.29 1772.19 Q1494.29 1780.92 1491.21 1785.52 Q1488.15 1790.11 1482.34 1790.11 Q1476.53 1790.11 1473.45 1785.52 Q1470.4 1780.92 1470.4 1772.19 Q1470.4 1763.44 1473.45 1758.86 Q1476.53 1754.25 1482.34 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1512.5 1757.95 Q1508.89 1757.95 1507.06 1761.52 Q1505.26 1765.06 1505.26 1772.19 Q1505.26 1779.3 1507.06 1782.86 Q1508.89 1786.4 1512.5 1786.4 Q1516.14 1786.4 1517.94 1782.86 Q1519.77 1779.3 1519.77 1772.19 Q1519.77 1765.06 1517.94 1761.52 Q1516.14 1757.95 1512.5 1757.95 M1512.5 1754.25 Q1518.31 1754.25 1521.37 1758.86 Q1524.45 1763.44 1524.45 1772.19 Q1524.45 1780.92 1521.37 1785.52 Q1518.31 1790.11 1512.5 1790.11 Q1506.69 1790.11 1503.61 1785.52 Q1500.56 1780.92 1500.56 1772.19 Q1500.56 1763.44 1503.61 1758.86 Q1506.69 1754.25 1512.5 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2016.36 1785.5 L2024 1785.5 L2024 1759.13 L2015.69 1760.8 L2015.69 1756.54 L2023.95 1754.87 L2028.63 1754.87 L2028.63 1785.5 L2036.26 1785.5 L2036.26 1789.43 L2016.36 1789.43 L2016.36 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2045.75 1754.87 L2064.11 1754.87 L2064.11 1758.81 L2050.04 1758.81 L2050.04 1767.28 Q2051.06 1766.93 2052.07 1766.77 Q2053.09 1766.59 2054.11 1766.59 Q2059.9 1766.59 2063.28 1769.76 Q2066.66 1772.93 2066.66 1778.35 Q2066.66 1783.93 2063.19 1787.03 Q2059.71 1790.11 2053.39 1790.11 Q2051.22 1790.11 2048.95 1789.74 Q2046.7 1789.37 2044.3 1788.62 L2044.3 1783.93 Q2046.38 1785.06 2048.6 1785.62 Q2050.82 1786.17 2053.3 1786.17 Q2057.31 1786.17 2059.64 1784.06 Q2061.98 1781.96 2061.98 1778.35 Q2061.98 1774.74 2059.64 1772.63 Q2057.31 1770.52 2053.3 1770.52 Q2051.43 1770.52 2049.55 1770.94 Q2047.7 1771.36 2045.75 1772.24 L2045.75 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2085.87 1757.95 Q2082.26 1757.95 2080.43 1761.52 Q2078.62 1765.06 2078.62 1772.19 Q2078.62 1779.3 2080.43 1782.86 Q2082.26 1786.4 2085.87 1786.4 Q2089.5 1786.4 2091.31 1782.86 Q2093.14 1779.3 2093.14 1772.19 Q2093.14 1765.06 2091.31 1761.52 Q2089.5 1757.95 2085.87 1757.95 M2085.87 1754.25 Q2091.68 1754.25 2094.74 1758.86 Q2097.81 1763.44 2097.81 1772.19 Q2097.81 1780.92 2094.74 1785.52 Q2091.68 1790.11 2085.87 1790.11 Q2080.06 1790.11 2076.98 1785.52 Q2073.93 1780.92 2073.93 1772.19 Q2073.93 1763.44 2076.98 1758.86 Q2080.06 1754.25 2085.87 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2116.03 1757.95 Q2112.42 1757.95 2110.59 1761.52 Q2108.79 1765.06 2108.79 1772.19 Q2108.79 1779.3 2110.59 1782.86 Q2112.42 1786.4 2116.03 1786.4 Q2119.67 1786.4 2121.47 1782.86 Q2123.3 1779.3 2123.3 1772.19 Q2123.3 1765.06 2121.47 1761.52 Q2119.67 1757.95 2116.03 1757.95 M2116.03 1754.25 Q2121.84 1754.25 2124.9 1758.86 Q2127.98 1763.44 2127.98 1772.19 Q2127.98 1780.92 2124.9 1785.52 Q2121.84 1790.11 2116.03 1790.11 Q2110.22 1790.11 2107.14 1785.52 Q2104.09 1780.92 2104.09 1772.19 Q2104.09 1763.44 2107.14 1758.86 Q2110.22 1754.25 2116.03 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2623.97 1785.5 L2640.29 1785.5 L2640.29 1789.43 L2618.35 1789.43 L2618.35 1785.5 Q2621.01 1782.74 2625.59 1778.12 Q2630.2 1773.46 2631.38 1772.12 Q2633.62 1769.6 2634.5 1767.86 Q2635.41 1766.1 2635.41 1764.41 Q2635.41 1761.66 2633.46 1759.92 Q2631.54 1758.18 2628.44 1758.18 Q2626.24 1758.18 2623.79 1758.95 Q2621.36 1759.71 2618.58 1761.26 L2618.58 1756.54 Q2621.4 1755.41 2623.86 1754.83 Q2626.31 1754.25 2628.35 1754.25 Q2633.72 1754.25 2636.91 1756.93 Q2640.1 1759.62 2640.1 1764.11 Q2640.1 1766.24 2639.29 1768.16 Q2638.51 1770.06 2636.4 1772.65 Q2635.82 1773.32 2632.72 1776.54 Q2629.62 1779.74 2623.97 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2660.1 1757.95 Q2656.49 1757.95 2654.67 1761.52 Q2652.86 1765.06 2652.86 1772.19 Q2652.86 1779.3 2654.67 1782.86 Q2656.49 1786.4 2660.1 1786.4 Q2663.74 1786.4 2665.54 1782.86 Q2667.37 1779.3 2667.37 1772.19 Q2667.37 1765.06 2665.54 1761.52 Q2663.74 1757.95 2660.1 1757.95 M2660.1 1754.25 Q2665.92 1754.25 2668.97 1758.86 Q2672.05 1763.44 2672.05 1772.19 Q2672.05 1780.92 2668.97 1785.52 Q2665.92 1790.11 2660.1 1790.11 Q2654.29 1790.11 2651.22 1785.52 Q2648.16 1780.92 2648.16 1772.19 Q2648.16 1763.44 2651.22 1758.86 Q2654.29 1754.25 2660.1 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2690.27 1757.95 Q2686.66 1757.95 2684.83 1761.52 Q2683.02 1765.06 2683.02 1772.19 Q2683.02 1779.3 2684.83 1782.86 Q2686.66 1786.4 2690.27 1786.4 Q2693.9 1786.4 2695.71 1782.86 Q2697.54 1779.3 2697.54 1772.19 Q2697.54 1765.06 2695.71 1761.52 Q2693.9 1757.95 2690.27 1757.95 M2690.27 1754.25 Q2696.08 1754.25 2699.13 1758.86 Q2702.21 1763.44 2702.21 1772.19 Q2702.21 1780.92 2699.13 1785.52 Q2696.08 1790.11 2690.27 1790.11 Q2684.46 1790.11 2681.38 1785.52 Q2678.32 1780.92 2678.32 1772.19 Q2678.32 1763.44 2681.38 1758.86 Q2684.46 1754.25 2690.27 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2720.43 1757.95 Q2716.82 1757.95 2714.99 1761.52 Q2713.18 1765.06 2713.18 1772.19 Q2713.18 1779.3 2714.99 1782.86 Q2716.82 1786.4 2720.43 1786.4 Q2724.06 1786.4 2725.87 1782.86 Q2727.7 1779.3 2727.7 1772.19 Q2727.7 1765.06 2725.87 1761.52 Q2724.06 1757.95 2720.43 1757.95 M2720.43 1754.25 Q2726.24 1754.25 2729.29 1758.86 Q2732.37 1763.44 2732.37 1772.19 Q2732.37 1780.92 2729.29 1785.52 Q2726.24 1790.11 2720.43 1790.11 Q2714.62 1790.11 2711.54 1785.52 Q2708.48 1780.92 2708.48 1772.19 Q2708.48 1763.44 2711.54 1758.86 Q2714.62 1754.25 2720.43 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1590.75 2752.76,1590.75 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1428.81 2752.76,1428.81 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1266.87 2752.76,1266.87 "/>
-<polyline clip-path="url(#clip273)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1104.93 2752.76,1104.93 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1723.83 188.677,1087.05 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1590.75 207.574,1590.75 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1428.81 207.574,1428.81 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1266.87 207.574,1266.87 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1104.93 207.574,1104.93 "/>
-<path clip-path="url(#clip270)" d="M54.8023 1591.21 L84.478 1591.21 L84.478 1595.14 L54.8023 1595.14 L54.8023 1591.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M105.149 1588.89 Q102.001 1588.89 100.149 1591.04 Q98.3206 1593.2 98.3206 1596.95 Q98.3206 1600.67 100.149 1602.85 Q102.001 1605 105.149 1605 Q108.297 1605 110.126 1602.85 Q111.978 1600.67 111.978 1596.95 Q111.978 1593.2 110.126 1591.04 Q108.297 1588.89 105.149 1588.89 M114.432 1574.24 L114.432 1578.5 Q112.672 1577.66 110.867 1577.22 Q109.084 1576.78 107.325 1576.78 Q102.696 1576.78 100.242 1579.91 Q97.8113 1583.03 97.4641 1589.35 Q98.8298 1587.34 100.89 1586.27 Q102.95 1585.19 105.427 1585.19 Q110.635 1585.19 113.645 1588.36 Q116.677 1591.51 116.677 1596.95 Q116.677 1602.27 113.529 1605.49 Q110.381 1608.71 105.149 1608.71 Q99.1539 1608.71 95.9826 1604.12 Q92.8113 1599.52 92.8113 1590.79 Q92.8113 1582.59 96.7002 1577.73 Q100.589 1572.85 107.14 1572.85 Q108.899 1572.85 110.682 1573.2 Q112.487 1573.54 114.432 1574.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 1576.55 Q131.121 1576.55 129.293 1580.12 Q127.487 1583.66 127.487 1590.79 Q127.487 1597.89 129.293 1601.46 Q131.121 1605 134.732 1605 Q138.367 1605 140.172 1601.46 Q142.001 1597.89 142.001 1590.79 Q142.001 1583.66 140.172 1580.12 Q138.367 1576.55 134.732 1576.55 M134.732 1572.85 Q140.543 1572.85 143.598 1577.46 Q146.677 1582.04 146.677 1590.79 Q146.677 1599.52 143.598 1604.12 Q140.543 1608.71 134.732 1608.71 Q128.922 1608.71 125.844 1604.12 Q122.788 1599.52 122.788 1590.79 Q122.788 1582.04 125.844 1577.46 Q128.922 1572.85 134.732 1572.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M54.8023 1429.26 L84.478 1429.26 L84.478 1433.2 L54.8023 1433.2 L54.8023 1429.26 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M107.418 1415.61 L95.6122 1434.05 L107.418 1434.05 L107.418 1415.61 M106.191 1411.53 L112.07 1411.53 L112.07 1434.05 L117.001 1434.05 L117.001 1437.94 L112.07 1437.94 L112.07 1446.09 L107.418 1446.09 L107.418 1437.94 L91.816 1437.94 L91.816 1433.43 L106.191 1411.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 1414.61 Q131.121 1414.61 129.293 1418.18 Q127.487 1421.72 127.487 1428.85 Q127.487 1435.95 129.293 1439.52 Q131.121 1443.06 134.732 1443.06 Q138.367 1443.06 140.172 1439.52 Q142.001 1435.95 142.001 1428.85 Q142.001 1421.72 140.172 1418.18 Q138.367 1414.61 134.732 1414.61 M134.732 1410.91 Q140.543 1410.91 143.598 1415.51 Q146.677 1420.1 146.677 1428.85 Q146.677 1437.57 143.598 1442.18 Q140.543 1446.76 134.732 1446.76 Q128.922 1446.76 125.844 1442.18 Q122.788 1437.57 122.788 1428.85 Q122.788 1420.1 125.844 1415.51 Q128.922 1410.91 134.732 1410.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M54.8023 1267.32 L84.478 1267.32 L84.478 1271.26 L54.8023 1271.26 L54.8023 1267.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M98.5983 1280.21 L114.918 1280.21 L114.918 1284.15 L92.9734 1284.15 L92.9734 1280.21 Q95.6354 1277.46 100.219 1272.83 Q104.825 1268.18 106.006 1266.83 Q108.251 1264.31 109.131 1262.58 Q110.033 1260.82 110.033 1259.13 Q110.033 1256.37 108.089 1254.64 Q106.168 1252.9 103.066 1252.9 Q100.867 1252.9 98.4132 1253.66 Q95.9826 1254.43 93.2049 1255.98 L93.2049 1251.26 Q96.0289 1250.12 98.4826 1249.54 Q100.936 1248.96 102.973 1248.96 Q108.344 1248.96 111.538 1251.65 Q114.733 1254.33 114.733 1258.83 Q114.733 1260.96 113.922 1262.88 Q113.135 1264.77 111.029 1267.37 Q110.45 1268.04 107.348 1271.26 Q104.246 1274.45 98.5983 1280.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 1252.67 Q131.121 1252.67 129.293 1256.23 Q127.487 1259.77 127.487 1266.9 Q127.487 1274.01 129.293 1277.58 Q131.121 1281.12 134.732 1281.12 Q138.367 1281.12 140.172 1277.58 Q142.001 1274.01 142.001 1266.9 Q142.001 1259.77 140.172 1256.23 Q138.367 1252.67 134.732 1252.67 M134.732 1248.96 Q140.543 1248.96 143.598 1253.57 Q146.677 1258.15 146.677 1266.9 Q146.677 1275.63 143.598 1280.24 Q140.543 1284.82 134.732 1284.82 Q128.922 1284.82 125.844 1280.24 Q122.788 1275.63 122.788 1266.9 Q122.788 1258.15 125.844 1253.57 Q128.922 1248.96 134.732 1248.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 1090.73 Q131.121 1090.73 129.293 1094.29 Q127.487 1097.83 127.487 1104.96 Q127.487 1112.07 129.293 1115.63 Q131.121 1119.17 134.732 1119.17 Q138.367 1119.17 140.172 1115.63 Q142.001 1112.07 142.001 1104.96 Q142.001 1097.83 140.172 1094.29 Q138.367 1090.73 134.732 1090.73 M134.732 1087.02 Q140.543 1087.02 143.598 1091.63 Q146.677 1096.21 146.677 1104.96 Q146.677 1113.69 143.598 1118.3 Q140.543 1122.88 134.732 1122.88 Q128.922 1122.88 125.844 1118.3 Q122.788 1113.69 122.788 1104.96 Q122.788 1096.21 125.844 1091.63 Q128.922 1087.02 134.732 1087.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1127.38 977.026 L1127.38 999.185 L1140.51 999.185 Q1147.11 999.185 1150.27 996.471 Q1153.47 993.716 1153.47 988.085 Q1153.47 982.414 1150.27 979.74 Q1147.11 977.026 1140.51 977.026 L1127.38 977.026 M1127.38 952.154 L1127.38 970.383 L1139.49 970.383 Q1145.49 970.383 1148.41 968.155 Q1151.36 965.886 1151.36 961.268 Q1151.36 956.691 1148.41 954.422 Q1145.49 952.154 1139.49 952.154 L1127.38 952.154 M1119.2 945.429 L1140.1 945.429 Q1149.46 945.429 1154.52 949.318 Q1159.59 953.207 1159.59 960.377 Q1159.59 965.927 1156.99 969.208 Q1154.4 972.489 1149.38 973.3 Q1155.41 974.596 1158.74 978.728 Q1162.1 982.819 1162.1 988.977 Q1162.1 997.078 1156.59 1001.49 Q1151.08 1005.91 1140.91 1005.91 L1119.2 1005.91 L1119.2 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1196.41 983.103 Q1187.38 983.103 1183.89 985.169 Q1180.41 987.235 1180.41 992.217 Q1180.41 996.187 1183 998.537 Q1185.63 1000.85 1190.13 1000.85 Q1196.33 1000.85 1200.06 996.471 Q1203.82 992.055 1203.82 984.764 L1203.82 983.103 L1196.41 983.103 M1211.28 980.024 L1211.28 1005.91 L1203.82 1005.91 L1203.82 999.023 Q1201.27 1003.15 1197.46 1005.14 Q1193.65 1007.08 1188.15 1007.08 Q1181.18 1007.08 1177.05 1003.2 Q1172.95 999.266 1172.95 992.703 Q1172.95 985.047 1178.06 981.158 Q1183.2 977.269 1193.37 977.269 L1203.82 977.269 L1203.82 976.54 Q1203.82 971.396 1200.42 968.601 Q1197.06 965.765 1190.94 965.765 Q1187.05 965.765 1183.37 966.697 Q1179.68 967.628 1176.28 969.492 L1176.28 962.605 Q1180.37 961.025 1184.22 960.256 Q1188.06 959.445 1191.71 959.445 Q1201.55 959.445 1206.42 964.55 Q1211.28 969.654 1211.28 980.024 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1264.34 978.525 L1264.34 1005.91 L1256.89 1005.91 L1256.89 978.768 Q1256.89 972.327 1254.38 969.127 Q1251.87 965.927 1246.84 965.927 Q1240.81 965.927 1237.32 969.775 Q1233.84 973.624 1233.84 980.267 L1233.84 1005.91 L1226.35 1005.91 L1226.35 960.539 L1233.84 960.539 L1233.84 967.588 Q1236.51 963.496 1240.12 961.471 Q1243.76 959.445 1248.5 959.445 Q1256.32 959.445 1260.33 964.307 Q1264.34 969.127 1264.34 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1278.93 942.877 L1286.42 942.877 L1286.42 980.105 L1308.66 960.539 L1318.18 960.539 L1294.12 981.766 L1319.19 1005.91 L1309.47 1005.91 L1286.42 983.751 L1286.42 1005.91 L1278.93 1005.91 L1278.93 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1374.16 953.491 L1363.06 983.589 L1385.3 983.589 L1374.16 953.491 M1369.55 945.429 L1378.82 945.429 L1401.87 1005.91 L1393.36 1005.91 L1387.86 990.394 L1360.59 990.394 L1355.08 1005.91 L1346.45 1005.91 L1369.55 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1448.09 978.525 L1448.09 1005.91 L1440.64 1005.91 L1440.64 978.768 Q1440.64 972.327 1438.13 969.127 Q1435.62 965.927 1430.59 965.927 Q1424.56 965.927 1421.07 969.775 Q1417.59 973.624 1417.59 980.267 L1417.59 1005.91 L1410.09 1005.91 L1410.09 960.539 L1417.59 960.539 L1417.59 967.588 Q1420.26 963.496 1423.87 961.471 Q1427.51 959.445 1432.25 959.445 Q1440.07 959.445 1444.08 964.307 Q1448.09 969.127 1448.09 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1492.81 982.698 Q1492.81 974.596 1489.45 970.14 Q1486.13 965.684 1480.09 965.684 Q1474.1 965.684 1470.74 970.14 Q1467.41 974.596 1467.41 982.698 Q1467.41 990.759 1470.74 995.215 Q1474.1 999.671 1480.09 999.671 Q1486.13 999.671 1489.45 995.215 Q1492.81 990.759 1492.81 982.698 M1500.27 1000.28 Q1500.27 1011.86 1495.12 1017.49 Q1489.98 1023.17 1479.37 1023.17 Q1475.44 1023.17 1471.95 1022.56 Q1468.47 1021.99 1465.19 1020.78 L1465.19 1013.53 Q1468.47 1015.31 1471.67 1016.16 Q1474.87 1017.01 1478.19 1017.01 Q1485.52 1017.01 1489.17 1013.16 Q1492.81 1009.35 1492.81 1001.62 L1492.81 997.929 Q1490.51 1001.94 1486.9 1003.92 Q1483.29 1005.91 1478.27 1005.91 Q1469.93 1005.91 1464.82 999.549 Q1459.72 993.189 1459.72 982.698 Q1459.72 972.165 1464.82 965.805 Q1469.93 959.445 1478.27 959.445 Q1483.29 959.445 1486.9 961.43 Q1490.51 963.415 1492.81 967.426 L1492.81 960.539 L1500.27 960.539 L1500.27 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1515.62 942.877 L1523.07 942.877 L1523.07 1005.91 L1515.62 1005.91 L1515.62 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1577.48 981.361 L1577.48 985.007 L1543.21 985.007 Q1543.69 992.703 1547.83 996.754 Q1552 1000.76 1559.41 1000.76 Q1563.7 1000.76 1567.72 999.711 Q1571.77 998.658 1575.74 996.552 L1575.74 1003.6 Q1571.73 1005.3 1567.51 1006.19 Q1563.3 1007.08 1558.97 1007.08 Q1548.11 1007.08 1541.75 1000.76 Q1535.43 994.445 1535.43 983.67 Q1535.43 972.53 1541.42 966.008 Q1547.46 959.445 1557.67 959.445 Q1566.82 959.445 1572.13 965.36 Q1577.48 971.234 1577.48 981.361 M1570.02 979.173 Q1569.94 973.057 1566.58 969.411 Q1563.26 965.765 1557.75 965.765 Q1551.51 965.765 1547.74 969.289 Q1544.02 972.813 1543.45 979.214 L1570.02 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1633.99 942.958 Q1628.56 952.275 1625.93 961.39 Q1623.29 970.504 1623.29 979.862 Q1623.29 989.22 1625.93 998.415 Q1628.6 1007.57 1633.99 1016.85 L1627.51 1016.85 Q1621.43 1007.33 1618.39 998.132 Q1615.39 988.936 1615.39 979.862 Q1615.39 970.829 1618.39 961.673 Q1621.39 952.518 1627.51 942.958 L1633.99 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1678.3 967.426 L1678.3 942.877 L1685.76 942.877 L1685.76 1005.91 L1678.3 1005.91 L1678.3 999.104 Q1675.96 1003.15 1672.35 1005.14 Q1668.79 1007.08 1663.76 1007.08 Q1655.54 1007.08 1650.35 1000.52 Q1645.21 993.959 1645.21 983.265 Q1645.21 972.57 1650.35 966.008 Q1655.54 959.445 1663.76 959.445 Q1668.79 959.445 1672.35 961.43 Q1675.96 963.375 1678.3 967.426 M1652.91 983.265 Q1652.91 991.488 1656.27 996.187 Q1659.67 1000.85 1665.59 1000.85 Q1671.5 1000.85 1674.9 996.187 Q1678.3 991.488 1678.3 983.265 Q1678.3 975.041 1674.9 970.383 Q1671.5 965.684 1665.59 965.684 Q1659.67 965.684 1656.27 970.383 Q1652.91 975.041 1652.91 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1739.92 981.361 L1739.92 985.007 L1705.65 985.007 Q1706.13 992.703 1710.27 996.754 Q1714.44 1000.76 1721.85 1000.76 Q1726.15 1000.76 1730.16 999.711 Q1734.21 998.658 1738.18 996.552 L1738.18 1003.6 Q1734.17 1005.3 1729.95 1006.19 Q1725.74 1007.08 1721.41 1007.08 Q1710.55 1007.08 1704.19 1000.76 Q1697.87 994.445 1697.87 983.67 Q1697.87 972.53 1703.87 966.008 Q1709.9 959.445 1720.11 959.445 Q1729.27 959.445 1734.57 965.36 Q1739.92 971.234 1739.92 981.361 M1732.47 979.173 Q1732.38 973.057 1729.02 969.411 Q1725.7 965.765 1720.19 965.765 Q1713.95 965.765 1710.19 969.289 Q1706.46 972.813 1705.89 979.214 L1732.47 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1782.01 982.698 Q1782.01 974.596 1778.65 970.14 Q1775.32 965.684 1769.29 965.684 Q1763.29 965.684 1759.93 970.14 Q1756.61 974.596 1756.61 982.698 Q1756.61 990.759 1759.93 995.215 Q1763.29 999.671 1769.29 999.671 Q1775.32 999.671 1778.65 995.215 Q1782.01 990.759 1782.01 982.698 M1789.46 1000.28 Q1789.46 1011.86 1784.32 1017.49 Q1779.17 1023.17 1768.56 1023.17 Q1764.63 1023.17 1761.15 1022.56 Q1757.66 1021.99 1754.38 1020.78 L1754.38 1013.53 Q1757.66 1015.31 1760.86 1016.16 Q1764.06 1017.01 1767.38 1017.01 Q1774.72 1017.01 1778.36 1013.16 Q1782.01 1009.35 1782.01 1001.62 L1782.01 997.929 Q1779.7 1001.94 1776.09 1003.92 Q1772.49 1005.91 1767.47 1005.91 Q1759.12 1005.91 1754.02 999.549 Q1748.91 993.189 1748.91 982.698 Q1748.91 972.165 1754.02 965.805 Q1759.12 959.445 1767.47 959.445 Q1772.49 959.445 1776.09 961.43 Q1779.7 963.415 1782.01 967.426 L1782.01 960.539 L1789.46 960.539 L1789.46 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1803.64 942.958 L1810.12 942.958 Q1816.2 952.518 1819.2 961.673 Q1822.23 970.829 1822.23 979.862 Q1822.23 988.936 1819.2 998.132 Q1816.2 1007.33 1810.12 1016.85 L1803.64 1016.85 Q1809.03 1007.57 1811.66 998.415 Q1814.33 989.22 1814.33 979.862 Q1814.33 970.504 1811.66 961.39 Q1809.03 952.275 1803.64 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip273)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,1705.81 266.073,1705.04 270.902,1704.26 275.73,1703.49 280.558,1702.72 285.386,1701.96 290.214,1701.2 295.043,1700.44 299.871,1699.68 304.699,1698.94 309.527,1698.19 314.356,1697.46 319.184,1696.74 324.012,1696.02 328.84,1695.31 333.668,1694.62 338.497,1693.94 343.325,1693.27 348.153,1692.62 352.981,1691.98 357.81,1691.37 362.638,1690.77 367.466,1690.19 372.294,1689.63 377.123,1689.1 381.951,1688.59 386.779,1688.11 391.607,1687.65 396.435,1687.22 401.264,1686.81 406.092,1686.44 410.92,1686.09 415.748,1685.77 420.577,1685.47 425.405,1685.21 430.233,1684.96 435.061,1684.74 439.89,1684.54 444.718,1684.36 449.546,1684.2 454.374,1684.05 459.202,1683.91 464.031,1683.78 468.859,1683.65 473.687,1683.51 478.515,1683.37 483.344,1683.22 488.172,1683.06 493,1682.87 497.828,1682.67 502.656,1682.43 507.485,1682.17 512.313,1681.87 517.141,1681.54 521.969,1681.16 526.798,1680.75 531.626,1680.28 536.454,1679.77 541.282,1679.22 546.111,1678.61 550.939,1677.95 555.767,1677.25 560.595,1676.49 565.423,1675.68 570.252,1674.82 575.08,1673.92 579.908,1672.96 584.736,1671.96 589.565,1670.9 594.393,1669.81 599.221,1668.67 604.049,1667.48 608.878,1666.26 613.706,1664.99 618.534,1663.68 623.362,1662.34 628.19,1660.97 633.019,1659.55 637.847,1658.11 642.675,1656.64 647.503,1655.14 652.332,1653.61 657.16,1652.05 661.988,1650.47 666.816,1648.87 671.645,1647.24 676.473,1645.6 681.301,1643.94 686.129,1642.26 690.957,1640.56 695.786,1638.85 700.614,1637.13 705.442,1635.4 710.27,1633.66 715.099,1631.91 719.927,1630.15 724.755,1628.39 729.583,1626.62 734.411,1624.86 739.24,1623.09 744.068,1621.32 748.896,1619.55 753.724,1617.79 758.553,1616.03 763.381,1614.28 768.209,1612.54 773.037,1610.81 777.866,1609.1 782.694,1607.4 787.522,1605.71 792.35,1604.04 797.178,1602.39 802.007,1600.77 806.835,1599.17 811.663,1597.6 816.491,1596.05 821.32,1594.54 826.148,1593.06 830.976,1591.62 835.804,1590.21 840.633,1588.85 845.461,1587.52 850.289,1586.24 855.117,1585 859.945,1583.81 864.774,1582.67 869.602,1581.58 874.43,1580.54 879.258,1579.54 884.087,1578.59 888.915,1577.7 893.743,1576.85 898.571,1576.04 903.399,1575.28 908.228,1574.56 913.056,1573.87 917.884,1573.22 922.712,1572.6 927.541,1572 932.369,1571.42 937.197,1570.85 942.025,1570.29 946.854,1569.74 951.682,1569.18 956.51,1568.61 961.338,1568.02 966.166,1567.41 970.995,1566.78 975.823,1566.11 980.651,1565.41 985.479,1564.66 990.308,1563.88 995.136,1563.05 999.964,1562.16 1004.79,1561.23 1009.62,1560.24 1014.45,1559.2 1019.28,1558.11 1024.11,1556.96 1028.93,1555.76 1033.76,1554.51 1038.59,1553.2 1043.42,1551.85 1048.25,1550.45 1053.07,1549 1057.9,1547.51 1062.73,1545.98 1067.56,1544.42 1072.39,1542.82 1077.22,1541.18 1082.04,1539.52 1086.87,1537.84 1091.7,1536.13 1096.53,1534.4 1101.36,1532.66 1106.19,1530.91 1111.01,1529.15 1115.84,1527.38 1120.67,1525.62 1125.5,1523.85 1130.33,1522.1 1135.15,1520.35 1139.98,1518.61 1144.81,1516.89 1149.64,1515.19 1154.47,1513.51 1159.3,1511.86 1164.12,1510.23 1168.95,1508.64 1173.78,1507.09 1178.61,1505.57 1183.44,1504.09 1188.26,1502.65 1193.09,1501.26 1197.92,1499.91 1202.75,1498.61 1207.58,1497.36 1212.41,1496.16 1217.23,1495.01 1222.06,1493.91 1226.89,1492.85 1231.72,1491.84 1236.55,1490.87 1241.38,1489.94 1246.2,1489.05 1251.03,1488.2 1255.86,1487.37 1260.69,1486.57 1265.52,1485.79 1270.34,1485.03 1275.17,1484.27 1280,1483.51 1284.83,1482.75 1289.66,1481.99 1294.49,1481.2 1299.31,1480.4 1304.14,1479.57 1308.97,1478.72 1313.8,1477.83 1318.63,1476.9 1323.46,1475.93 1328.28,1474.92 1333.11,1473.86 1337.94,1472.76 1342.77,1471.61 1347.6,1470.42 1352.42,1469.17 1357.25,1467.89 1362.08,1466.55 1366.91,1465.18 1371.74,1463.77 1376.57,1462.32 1381.39,1460.83 1386.22,1459.32 1391.05,1457.78 1395.88,1456.21 1400.71,1454.62 1405.54,1453.02 1410.36,1451.4 1415.19,1449.78 1420.02,1448.15 1424.85,1446.52 1429.68,1444.9 1434.5,1443.28 1439.33,1441.68 1444.16,1440.08 1448.99,1438.51 1453.82,1436.96 1458.65,1435.44 1463.47,1433.94 1468.3,1432.47 1473.13,1431.04 1477.96,1429.64 1482.79,1428.27 1487.62,1426.95 1492.44,1425.66 1497.27,1424.41 1502.1,1423.2 1506.93,1422.03 1511.76,1420.9 1516.58,1419.8 1521.41,1418.74 1526.24,1417.7 1531.07,1416.7 1535.9,1415.72 1540.73,1414.75 1545.55,1413.81 1550.38,1412.87 1555.21,1411.95 1560.04,1411.02 1564.87,1410.1 1569.7,1409.16 1574.52,1408.22 1579.35,1407.26 1584.18,1406.28 1589.01,1405.28 1593.84,1404.25 1598.66,1403.2 1603.49,1402.11 1608.32,1400.99 1613.15,1399.84 1617.98,1398.66 1622.81,1397.44 1627.63,1396.19 1632.46,1394.9 1637.29,1393.59 1642.12,1392.25 1646.95,1390.88 1651.77,1389.48 1656.6,1388.06 1661.43,1386.63 1666.26,1385.18 1671.09,1383.71 1675.92,1382.24 1680.74,1380.76 1685.57,1379.28 1690.4,1377.8 1695.23,1376.32 1700.06,1374.85 1704.89,1373.39 1709.71,1371.94 1714.54,1370.51 1719.37,1369.1 1724.2,1367.7 1729.03,1366.33 1733.85,1364.98 1738.68,1363.66 1743.51,1362.36 1748.34,1361.09 1753.17,1359.84 1758,1358.62 1762.82,1357.43 1767.65,1356.26 1772.48,1355.11 1777.31,1353.98 1782.14,1352.88 1786.97,1351.78 1791.79,1350.71 1796.62,1349.64 1801.45,1348.58 1806.28,1347.53 1811.11,1346.48 1815.93,1345.43 1820.76,1344.37 1825.59,1343.31 1830.42,1342.23 1835.25,1341.15 1840.08,1340.05 1844.9,1338.93 1849.73,1337.8 1854.56,1336.65 1859.39,1335.47 1864.22,1334.28 1869.05,1333.07 1873.87,1331.84 1878.7,1330.58 1883.53,1329.32 1888.36,1328.03 1893.19,1326.73 1898.01,1325.41 1902.84,1324.08 1907.67,1322.74 1912.5,1321.39 1917.33,1320.04 1922.16,1318.68 1926.98,1317.33 1931.81,1315.97 1936.64,1314.61 1941.47,1313.26 1946.3,1311.92 1951.13,1310.58 1955.95,1309.25 1960.78,1307.94 1965.61,1306.63 1970.44,1305.35 1975.27,1304.07 1980.09,1302.81 1984.92,1301.56 1989.75,1300.33 1994.58,1299.11 1999.41,1297.91 2004.24,1296.72 2009.06,1295.54 2013.89,1294.37 2018.72,1293.21 2023.55,1292.06 2028.38,1290.92 2033.2,1289.78 2038.03,1288.65 2042.86,1287.51 2047.69,1286.38 2052.52,1285.25 2057.35,1284.11 2062.17,1282.97 2067,1281.83 2071.83,1280.68 2076.66,1279.52 2081.49,1278.35 2086.32,1277.18 2091.14,1275.99 2095.97,1274.8 2100.8,1273.6 2105.63,1272.39 2110.46,1271.17 2115.28,1269.94 2120.11,1268.7 2124.94,1267.46 2129.77,1266.22 2134.6,1264.97 2139.43,1263.71 2144.25,1262.45 2149.08,1261.19 2153.91,1259.93 2158.74,1258.67 2163.57,1257.41 2168.4,1256.16 2173.22,1254.9 2178.05,1253.65 2182.88,1252.4 2187.71,1251.16 2192.54,1249.92 2197.36,1248.69 2202.19,1247.46 2207.02,1246.24 2211.85,1245.02 2216.68,1243.81 2221.51,1242.6 2226.33,1241.4 2231.16,1240.21 2235.99,1239.01 2240.82,1237.82 2245.65,1236.64 2250.48,1235.46 2255.3,1234.28 2260.13,1233.1 2264.96,1231.92 2269.79,1230.75 2274.62,1229.58 2279.44,1228.4 2284.27,1227.23 2289.1,1226.06 2293.93,1224.89 2298.76,1223.71 2303.59,1222.54 2308.41,1221.37 2313.24,1220.2 2318.07,1219.02 2322.9,1217.85 2327.73,1216.67 2332.56,1215.5 2337.38,1214.32 2342.21,1213.15 2347.04,1211.97 2351.87,1210.79 2356.7,1209.61 2361.52,1208.43 2366.35,1207.25 2371.18,1206.06 2376.01,1204.87 2380.84,1203.68 2385.67,1202.49 2390.49,1201.29 2395.32,1200.09 2400.15,1198.89 2404.98,1197.68 2409.81,1196.46 2414.64,1195.24 2419.46,1194.02 2424.29,1192.79 2429.12,1191.55 2433.95,1190.31 2438.78,1189.06 2443.6,1187.81 2448.43,1186.55 2453.26,1185.28 2458.09,1184.02 2462.92,1182.75 2467.75,1181.47 2472.57,1180.2 2477.4,1178.93 2482.23,1177.65 2487.06,1176.38 2491.89,1175.1 2496.71,1173.83 2501.54,1172.57 2506.37,1171.3 2511.2,1170.04 2516.03,1168.79 2520.86,1167.53 2525.68,1166.28 2530.51,1165.03 2535.34,1163.79 2540.17,1162.54 2545,1161.28 2549.83,1160.02 2554.65,1158.76 2559.48,1157.48 2564.31,1156.18 2569.14,1154.86 2573.97,1153.52 2578.79,1152.15 2583.62,1150.74 2588.45,1149.28 2593.28,1147.77 2598.11,1146.2 2602.94,1144.57 2607.76,1142.87 2612.59,1141.08 2617.42,1139.2 2622.25,1137.23 2627.08,1135.16 2631.91,1132.98 2636.73,1130.69 2641.56,1128.29 2646.39,1125.77 2651.22,1123.14 2656.05,1120.39 2660.87,1117.54 2665.7,1114.57 2670.53,1111.5 2675.36,1108.33 2680.19,1105.07 "/>
-<path clip-path="url(#clip270)" d="M188.677 2657.17 L2752.76 2657.17 L2752.76 2020.38 L188.677 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,1723.83 261.245,1087.05 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="864.774,1723.83 864.774,1087.05 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1468.3,1723.83 1468.3,1087.05 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2071.83,1723.83 2071.83,1087.05 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2675.36,1723.83 2675.36,1087.05 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1723.83 2752.76,1723.83 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,1723.83 261.245,1704.94 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="864.774,1723.83 864.774,1704.94 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1468.3,1723.83 1468.3,1704.94 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2071.83,1723.83 2071.83,1704.94 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2675.36,1723.83 2675.36,1704.94 "/>
+<path clip-path="url(#clip090)" d="M261.245 1757.95 Q257.634 1757.95 255.805 1761.52 Q254 1765.06 254 1772.19 Q254 1779.3 255.805 1782.86 Q257.634 1786.4 261.245 1786.4 Q264.879 1786.4 266.685 1782.86 Q268.514 1779.3 268.514 1772.19 Q268.514 1765.06 266.685 1761.52 Q264.879 1757.95 261.245 1757.95 M261.245 1754.25 Q267.055 1754.25 270.111 1758.86 Q273.189 1763.44 273.189 1772.19 Q273.189 1780.92 270.111 1785.52 Q267.055 1790.11 261.245 1790.11 Q255.435 1790.11 252.356 1785.52 Q249.301 1780.92 249.301 1772.19 Q249.301 1763.44 252.356 1758.86 Q255.435 1754.25 261.245 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M824.392 1754.87 L842.748 1754.87 L842.748 1758.81 L828.674 1758.81 L828.674 1767.28 Q829.693 1766.93 830.711 1766.77 Q831.73 1766.59 832.748 1766.59 Q838.535 1766.59 841.915 1769.76 Q845.295 1772.93 845.295 1778.35 Q845.295 1783.93 841.822 1787.03 Q838.35 1790.11 832.031 1790.11 Q829.855 1790.11 827.586 1789.74 Q825.341 1789.37 822.934 1788.62 L822.934 1783.93 Q825.017 1785.06 827.239 1785.62 Q829.461 1786.17 831.938 1786.17 Q835.943 1786.17 838.281 1784.06 Q840.619 1781.96 840.619 1778.35 Q840.619 1774.74 838.281 1772.63 Q835.943 1770.52 831.938 1770.52 Q830.063 1770.52 828.188 1770.94 Q826.336 1771.36 824.392 1772.24 L824.392 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M864.507 1757.95 Q860.896 1757.95 859.068 1761.52 Q857.262 1765.06 857.262 1772.19 Q857.262 1779.3 859.068 1782.86 Q860.896 1786.4 864.507 1786.4 Q868.142 1786.4 869.947 1782.86 Q871.776 1779.3 871.776 1772.19 Q871.776 1765.06 869.947 1761.52 Q868.142 1757.95 864.507 1757.95 M864.507 1754.25 Q870.318 1754.25 873.373 1758.86 Q876.452 1763.44 876.452 1772.19 Q876.452 1780.92 873.373 1785.52 Q870.318 1790.11 864.507 1790.11 Q858.697 1790.11 855.619 1785.52 Q852.563 1780.92 852.563 1772.19 Q852.563 1763.44 855.619 1758.86 Q858.697 1754.25 864.507 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M894.669 1757.95 Q891.058 1757.95 889.23 1761.52 Q887.424 1765.06 887.424 1772.19 Q887.424 1779.3 889.23 1782.86 Q891.058 1786.4 894.669 1786.4 Q898.304 1786.4 900.109 1782.86 Q901.938 1779.3 901.938 1772.19 Q901.938 1765.06 900.109 1761.52 Q898.304 1757.95 894.669 1757.95 M894.669 1754.25 Q900.479 1754.25 903.535 1758.86 Q906.614 1763.44 906.614 1772.19 Q906.614 1780.92 903.535 1785.52 Q900.479 1790.11 894.669 1790.11 Q888.859 1790.11 885.78 1785.52 Q882.725 1780.92 882.725 1772.19 Q882.725 1763.44 885.78 1758.86 Q888.859 1754.25 894.669 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1412.83 1785.5 L1420.47 1785.5 L1420.47 1759.13 L1412.16 1760.8 L1412.16 1756.54 L1420.42 1754.87 L1425.1 1754.87 L1425.1 1785.5 L1432.74 1785.5 L1432.74 1789.43 L1412.83 1789.43 L1412.83 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1452.18 1757.95 Q1448.57 1757.95 1446.74 1761.52 Q1444.93 1765.06 1444.93 1772.19 Q1444.93 1779.3 1446.74 1782.86 Q1448.57 1786.4 1452.18 1786.4 Q1455.81 1786.4 1457.62 1782.86 Q1459.45 1779.3 1459.45 1772.19 Q1459.45 1765.06 1457.62 1761.52 Q1455.81 1757.95 1452.18 1757.95 M1452.18 1754.25 Q1457.99 1754.25 1461.05 1758.86 Q1464.12 1763.44 1464.12 1772.19 Q1464.12 1780.92 1461.05 1785.52 Q1457.99 1790.11 1452.18 1790.11 Q1446.37 1790.11 1443.29 1785.52 Q1440.24 1780.92 1440.24 1772.19 Q1440.24 1763.44 1443.29 1758.86 Q1446.37 1754.25 1452.18 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1482.34 1757.95 Q1478.73 1757.95 1476.9 1761.52 Q1475.1 1765.06 1475.1 1772.19 Q1475.1 1779.3 1476.9 1782.86 Q1478.73 1786.4 1482.34 1786.4 Q1485.98 1786.4 1487.78 1782.86 Q1489.61 1779.3 1489.61 1772.19 Q1489.61 1765.06 1487.78 1761.52 Q1485.98 1757.95 1482.34 1757.95 M1482.34 1754.25 Q1488.15 1754.25 1491.21 1758.86 Q1494.29 1763.44 1494.29 1772.19 Q1494.29 1780.92 1491.21 1785.52 Q1488.15 1790.11 1482.34 1790.11 Q1476.53 1790.11 1473.45 1785.52 Q1470.4 1780.92 1470.4 1772.19 Q1470.4 1763.44 1473.45 1758.86 Q1476.53 1754.25 1482.34 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1512.5 1757.95 Q1508.89 1757.95 1507.06 1761.52 Q1505.26 1765.06 1505.26 1772.19 Q1505.26 1779.3 1507.06 1782.86 Q1508.89 1786.4 1512.5 1786.4 Q1516.14 1786.4 1517.94 1782.86 Q1519.77 1779.3 1519.77 1772.19 Q1519.77 1765.06 1517.94 1761.52 Q1516.14 1757.95 1512.5 1757.95 M1512.5 1754.25 Q1518.31 1754.25 1521.37 1758.86 Q1524.45 1763.44 1524.45 1772.19 Q1524.45 1780.92 1521.37 1785.52 Q1518.31 1790.11 1512.5 1790.11 Q1506.69 1790.11 1503.61 1785.52 Q1500.56 1780.92 1500.56 1772.19 Q1500.56 1763.44 1503.61 1758.86 Q1506.69 1754.25 1512.5 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2016.36 1785.5 L2024 1785.5 L2024 1759.13 L2015.69 1760.8 L2015.69 1756.54 L2023.95 1754.87 L2028.63 1754.87 L2028.63 1785.5 L2036.26 1785.5 L2036.26 1789.43 L2016.36 1789.43 L2016.36 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2045.75 1754.87 L2064.11 1754.87 L2064.11 1758.81 L2050.04 1758.81 L2050.04 1767.28 Q2051.06 1766.93 2052.07 1766.77 Q2053.09 1766.59 2054.11 1766.59 Q2059.9 1766.59 2063.28 1769.76 Q2066.66 1772.93 2066.66 1778.35 Q2066.66 1783.93 2063.19 1787.03 Q2059.71 1790.11 2053.39 1790.11 Q2051.22 1790.11 2048.95 1789.74 Q2046.7 1789.37 2044.3 1788.62 L2044.3 1783.93 Q2046.38 1785.06 2048.6 1785.62 Q2050.82 1786.17 2053.3 1786.17 Q2057.31 1786.17 2059.64 1784.06 Q2061.98 1781.96 2061.98 1778.35 Q2061.98 1774.74 2059.64 1772.63 Q2057.31 1770.52 2053.3 1770.52 Q2051.43 1770.52 2049.55 1770.94 Q2047.7 1771.36 2045.75 1772.24 L2045.75 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2085.87 1757.95 Q2082.26 1757.95 2080.43 1761.52 Q2078.62 1765.06 2078.62 1772.19 Q2078.62 1779.3 2080.43 1782.86 Q2082.26 1786.4 2085.87 1786.4 Q2089.5 1786.4 2091.31 1782.86 Q2093.14 1779.3 2093.14 1772.19 Q2093.14 1765.06 2091.31 1761.52 Q2089.5 1757.95 2085.87 1757.95 M2085.87 1754.25 Q2091.68 1754.25 2094.74 1758.86 Q2097.81 1763.44 2097.81 1772.19 Q2097.81 1780.92 2094.74 1785.52 Q2091.68 1790.11 2085.87 1790.11 Q2080.06 1790.11 2076.98 1785.52 Q2073.93 1780.92 2073.93 1772.19 Q2073.93 1763.44 2076.98 1758.86 Q2080.06 1754.25 2085.87 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2116.03 1757.95 Q2112.42 1757.95 2110.59 1761.52 Q2108.79 1765.06 2108.79 1772.19 Q2108.79 1779.3 2110.59 1782.86 Q2112.42 1786.4 2116.03 1786.4 Q2119.67 1786.4 2121.47 1782.86 Q2123.3 1779.3 2123.3 1772.19 Q2123.3 1765.06 2121.47 1761.52 Q2119.67 1757.95 2116.03 1757.95 M2116.03 1754.25 Q2121.84 1754.25 2124.9 1758.86 Q2127.98 1763.44 2127.98 1772.19 Q2127.98 1780.92 2124.9 1785.52 Q2121.84 1790.11 2116.03 1790.11 Q2110.22 1790.11 2107.14 1785.52 Q2104.09 1780.92 2104.09 1772.19 Q2104.09 1763.44 2107.14 1758.86 Q2110.22 1754.25 2116.03 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2623.97 1785.5 L2640.29 1785.5 L2640.29 1789.43 L2618.35 1789.43 L2618.35 1785.5 Q2621.01 1782.74 2625.59 1778.12 Q2630.2 1773.46 2631.38 1772.12 Q2633.62 1769.6 2634.5 1767.86 Q2635.41 1766.1 2635.41 1764.41 Q2635.41 1761.66 2633.46 1759.92 Q2631.54 1758.18 2628.44 1758.18 Q2626.24 1758.18 2623.79 1758.95 Q2621.36 1759.71 2618.58 1761.26 L2618.58 1756.54 Q2621.4 1755.41 2623.86 1754.83 Q2626.31 1754.25 2628.35 1754.25 Q2633.72 1754.25 2636.91 1756.93 Q2640.1 1759.62 2640.1 1764.11 Q2640.1 1766.24 2639.29 1768.16 Q2638.51 1770.06 2636.4 1772.65 Q2635.82 1773.32 2632.72 1776.54 Q2629.62 1779.74 2623.97 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2660.1 1757.95 Q2656.49 1757.95 2654.67 1761.52 Q2652.86 1765.06 2652.86 1772.19 Q2652.86 1779.3 2654.67 1782.86 Q2656.49 1786.4 2660.1 1786.4 Q2663.74 1786.4 2665.54 1782.86 Q2667.37 1779.3 2667.37 1772.19 Q2667.37 1765.06 2665.54 1761.52 Q2663.74 1757.95 2660.1 1757.95 M2660.1 1754.25 Q2665.92 1754.25 2668.97 1758.86 Q2672.05 1763.44 2672.05 1772.19 Q2672.05 1780.92 2668.97 1785.52 Q2665.92 1790.11 2660.1 1790.11 Q2654.29 1790.11 2651.22 1785.52 Q2648.16 1780.92 2648.16 1772.19 Q2648.16 1763.44 2651.22 1758.86 Q2654.29 1754.25 2660.1 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2690.27 1757.95 Q2686.66 1757.95 2684.83 1761.52 Q2683.02 1765.06 2683.02 1772.19 Q2683.02 1779.3 2684.83 1782.86 Q2686.66 1786.4 2690.27 1786.4 Q2693.9 1786.4 2695.71 1782.86 Q2697.54 1779.3 2697.54 1772.19 Q2697.54 1765.06 2695.71 1761.52 Q2693.9 1757.95 2690.27 1757.95 M2690.27 1754.25 Q2696.08 1754.25 2699.13 1758.86 Q2702.21 1763.44 2702.21 1772.19 Q2702.21 1780.92 2699.13 1785.52 Q2696.08 1790.11 2690.27 1790.11 Q2684.46 1790.11 2681.38 1785.52 Q2678.32 1780.92 2678.32 1772.19 Q2678.32 1763.44 2681.38 1758.86 Q2684.46 1754.25 2690.27 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2720.43 1757.95 Q2716.82 1757.95 2714.99 1761.52 Q2713.18 1765.06 2713.18 1772.19 Q2713.18 1779.3 2714.99 1782.86 Q2716.82 1786.4 2720.43 1786.4 Q2724.06 1786.4 2725.87 1782.86 Q2727.7 1779.3 2727.7 1772.19 Q2727.7 1765.06 2725.87 1761.52 Q2724.06 1757.95 2720.43 1757.95 M2720.43 1754.25 Q2726.24 1754.25 2729.29 1758.86 Q2732.37 1763.44 2732.37 1772.19 Q2732.37 1780.92 2729.29 1785.52 Q2726.24 1790.11 2720.43 1790.11 Q2714.62 1790.11 2711.54 1785.52 Q2708.48 1780.92 2708.48 1772.19 Q2708.48 1763.44 2711.54 1758.86 Q2714.62 1754.25 2720.43 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1590.75 2752.76,1590.75 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1428.81 2752.76,1428.81 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1266.87 2752.76,1266.87 "/>
+<polyline clip-path="url(#clip093)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,1104.93 2752.76,1104.93 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1723.83 188.677,1087.05 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1590.75 207.574,1590.75 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1428.81 207.574,1428.81 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1266.87 207.574,1266.87 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,1104.93 207.574,1104.93 "/>
+<path clip-path="url(#clip090)" d="M54.8023 1591.21 L84.478 1591.21 L84.478 1595.14 L54.8023 1595.14 L54.8023 1591.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M105.149 1588.89 Q102.001 1588.89 100.149 1591.04 Q98.3206 1593.2 98.3206 1596.95 Q98.3206 1600.67 100.149 1602.85 Q102.001 1605 105.149 1605 Q108.297 1605 110.126 1602.85 Q111.978 1600.67 111.978 1596.95 Q111.978 1593.2 110.126 1591.04 Q108.297 1588.89 105.149 1588.89 M114.432 1574.24 L114.432 1578.5 Q112.672 1577.66 110.867 1577.22 Q109.084 1576.78 107.325 1576.78 Q102.696 1576.78 100.242 1579.91 Q97.8113 1583.03 97.4641 1589.35 Q98.8298 1587.34 100.89 1586.27 Q102.95 1585.19 105.427 1585.19 Q110.635 1585.19 113.645 1588.36 Q116.677 1591.51 116.677 1596.95 Q116.677 1602.27 113.529 1605.49 Q110.381 1608.71 105.149 1608.71 Q99.1539 1608.71 95.9826 1604.12 Q92.8113 1599.52 92.8113 1590.79 Q92.8113 1582.59 96.7002 1577.73 Q100.589 1572.85 107.14 1572.85 Q108.899 1572.85 110.682 1573.2 Q112.487 1573.54 114.432 1574.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 1576.55 Q131.121 1576.55 129.293 1580.12 Q127.487 1583.66 127.487 1590.79 Q127.487 1597.89 129.293 1601.46 Q131.121 1605 134.732 1605 Q138.367 1605 140.172 1601.46 Q142.001 1597.89 142.001 1590.79 Q142.001 1583.66 140.172 1580.12 Q138.367 1576.55 134.732 1576.55 M134.732 1572.85 Q140.543 1572.85 143.598 1577.46 Q146.677 1582.04 146.677 1590.79 Q146.677 1599.52 143.598 1604.12 Q140.543 1608.71 134.732 1608.71 Q128.922 1608.71 125.844 1604.12 Q122.788 1599.52 122.788 1590.79 Q122.788 1582.04 125.844 1577.46 Q128.922 1572.85 134.732 1572.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M54.8023 1429.26 L84.478 1429.26 L84.478 1433.2 L54.8023 1433.2 L54.8023 1429.26 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M107.418 1415.61 L95.6122 1434.05 L107.418 1434.05 L107.418 1415.61 M106.191 1411.53 L112.07 1411.53 L112.07 1434.05 L117.001 1434.05 L117.001 1437.94 L112.07 1437.94 L112.07 1446.09 L107.418 1446.09 L107.418 1437.94 L91.816 1437.94 L91.816 1433.43 L106.191 1411.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 1414.61 Q131.121 1414.61 129.293 1418.18 Q127.487 1421.72 127.487 1428.85 Q127.487 1435.95 129.293 1439.52 Q131.121 1443.06 134.732 1443.06 Q138.367 1443.06 140.172 1439.52 Q142.001 1435.95 142.001 1428.85 Q142.001 1421.72 140.172 1418.18 Q138.367 1414.61 134.732 1414.61 M134.732 1410.91 Q140.543 1410.91 143.598 1415.51 Q146.677 1420.1 146.677 1428.85 Q146.677 1437.57 143.598 1442.18 Q140.543 1446.76 134.732 1446.76 Q128.922 1446.76 125.844 1442.18 Q122.788 1437.57 122.788 1428.85 Q122.788 1420.1 125.844 1415.51 Q128.922 1410.91 134.732 1410.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M54.8023 1267.32 L84.478 1267.32 L84.478 1271.26 L54.8023 1271.26 L54.8023 1267.32 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M98.5983 1280.21 L114.918 1280.21 L114.918 1284.15 L92.9734 1284.15 L92.9734 1280.21 Q95.6354 1277.46 100.219 1272.83 Q104.825 1268.18 106.006 1266.83 Q108.251 1264.31 109.131 1262.58 Q110.033 1260.82 110.033 1259.13 Q110.033 1256.37 108.089 1254.64 Q106.168 1252.9 103.066 1252.9 Q100.867 1252.9 98.4132 1253.66 Q95.9826 1254.43 93.2049 1255.98 L93.2049 1251.26 Q96.0289 1250.12 98.4826 1249.54 Q100.936 1248.96 102.973 1248.96 Q108.344 1248.96 111.538 1251.65 Q114.733 1254.33 114.733 1258.83 Q114.733 1260.96 113.922 1262.88 Q113.135 1264.77 111.029 1267.37 Q110.45 1268.04 107.348 1271.26 Q104.246 1274.45 98.5983 1280.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 1252.67 Q131.121 1252.67 129.293 1256.23 Q127.487 1259.77 127.487 1266.9 Q127.487 1274.01 129.293 1277.58 Q131.121 1281.12 134.732 1281.12 Q138.367 1281.12 140.172 1277.58 Q142.001 1274.01 142.001 1266.9 Q142.001 1259.77 140.172 1256.23 Q138.367 1252.67 134.732 1252.67 M134.732 1248.96 Q140.543 1248.96 143.598 1253.57 Q146.677 1258.15 146.677 1266.9 Q146.677 1275.63 143.598 1280.24 Q140.543 1284.82 134.732 1284.82 Q128.922 1284.82 125.844 1280.24 Q122.788 1275.63 122.788 1266.9 Q122.788 1258.15 125.844 1253.57 Q128.922 1248.96 134.732 1248.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 1090.73 Q131.121 1090.73 129.293 1094.29 Q127.487 1097.83 127.487 1104.96 Q127.487 1112.07 129.293 1115.63 Q131.121 1119.17 134.732 1119.17 Q138.367 1119.17 140.172 1115.63 Q142.001 1112.07 142.001 1104.96 Q142.001 1097.83 140.172 1094.29 Q138.367 1090.73 134.732 1090.73 M134.732 1087.02 Q140.543 1087.02 143.598 1091.63 Q146.677 1096.21 146.677 1104.96 Q146.677 1113.69 143.598 1118.3 Q140.543 1122.88 134.732 1122.88 Q128.922 1122.88 125.844 1118.3 Q122.788 1113.69 122.788 1104.96 Q122.788 1096.21 125.844 1091.63 Q128.922 1087.02 134.732 1087.02 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1127.38 977.026 L1127.38 999.185 L1140.51 999.185 Q1147.11 999.185 1150.27 996.471 Q1153.47 993.716 1153.47 988.085 Q1153.47 982.414 1150.27 979.74 Q1147.11 977.026 1140.51 977.026 L1127.38 977.026 M1127.38 952.154 L1127.38 970.383 L1139.49 970.383 Q1145.49 970.383 1148.41 968.155 Q1151.36 965.886 1151.36 961.268 Q1151.36 956.691 1148.41 954.422 Q1145.49 952.154 1139.49 952.154 L1127.38 952.154 M1119.2 945.429 L1140.1 945.429 Q1149.46 945.429 1154.52 949.318 Q1159.59 953.207 1159.59 960.377 Q1159.59 965.927 1156.99 969.208 Q1154.4 972.489 1149.38 973.3 Q1155.41 974.596 1158.74 978.728 Q1162.1 982.819 1162.1 988.977 Q1162.1 997.078 1156.59 1001.49 Q1151.08 1005.91 1140.91 1005.91 L1119.2 1005.91 L1119.2 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1196.41 983.103 Q1187.38 983.103 1183.89 985.169 Q1180.41 987.235 1180.41 992.217 Q1180.41 996.187 1183 998.537 Q1185.63 1000.85 1190.13 1000.85 Q1196.33 1000.85 1200.06 996.471 Q1203.82 992.055 1203.82 984.764 L1203.82 983.103 L1196.41 983.103 M1211.28 980.024 L1211.28 1005.91 L1203.82 1005.91 L1203.82 999.023 Q1201.27 1003.15 1197.46 1005.14 Q1193.65 1007.08 1188.15 1007.08 Q1181.18 1007.08 1177.05 1003.2 Q1172.95 999.266 1172.95 992.703 Q1172.95 985.047 1178.06 981.158 Q1183.2 977.269 1193.37 977.269 L1203.82 977.269 L1203.82 976.54 Q1203.82 971.396 1200.42 968.601 Q1197.06 965.765 1190.94 965.765 Q1187.05 965.765 1183.37 966.697 Q1179.68 967.628 1176.28 969.492 L1176.28 962.605 Q1180.37 961.025 1184.22 960.256 Q1188.06 959.445 1191.71 959.445 Q1201.55 959.445 1206.42 964.55 Q1211.28 969.654 1211.28 980.024 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1264.34 978.525 L1264.34 1005.91 L1256.89 1005.91 L1256.89 978.768 Q1256.89 972.327 1254.38 969.127 Q1251.87 965.927 1246.84 965.927 Q1240.81 965.927 1237.32 969.775 Q1233.84 973.624 1233.84 980.267 L1233.84 1005.91 L1226.35 1005.91 L1226.35 960.539 L1233.84 960.539 L1233.84 967.588 Q1236.51 963.496 1240.12 961.471 Q1243.76 959.445 1248.5 959.445 Q1256.32 959.445 1260.33 964.307 Q1264.34 969.127 1264.34 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1278.93 942.877 L1286.42 942.877 L1286.42 980.105 L1308.66 960.539 L1318.18 960.539 L1294.12 981.766 L1319.19 1005.91 L1309.47 1005.91 L1286.42 983.751 L1286.42 1005.91 L1278.93 1005.91 L1278.93 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1374.16 953.491 L1363.06 983.589 L1385.3 983.589 L1374.16 953.491 M1369.55 945.429 L1378.82 945.429 L1401.87 1005.91 L1393.36 1005.91 L1387.86 990.394 L1360.59 990.394 L1355.08 1005.91 L1346.45 1005.91 L1369.55 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1448.09 978.525 L1448.09 1005.91 L1440.64 1005.91 L1440.64 978.768 Q1440.64 972.327 1438.13 969.127 Q1435.62 965.927 1430.59 965.927 Q1424.56 965.927 1421.07 969.775 Q1417.59 973.624 1417.59 980.267 L1417.59 1005.91 L1410.09 1005.91 L1410.09 960.539 L1417.59 960.539 L1417.59 967.588 Q1420.26 963.496 1423.87 961.471 Q1427.51 959.445 1432.25 959.445 Q1440.07 959.445 1444.08 964.307 Q1448.09 969.127 1448.09 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1492.81 982.698 Q1492.81 974.596 1489.45 970.14 Q1486.13 965.684 1480.09 965.684 Q1474.1 965.684 1470.74 970.14 Q1467.41 974.596 1467.41 982.698 Q1467.41 990.759 1470.74 995.215 Q1474.1 999.671 1480.09 999.671 Q1486.13 999.671 1489.45 995.215 Q1492.81 990.759 1492.81 982.698 M1500.27 1000.28 Q1500.27 1011.86 1495.12 1017.49 Q1489.98 1023.17 1479.37 1023.17 Q1475.44 1023.17 1471.95 1022.56 Q1468.47 1021.99 1465.19 1020.78 L1465.19 1013.53 Q1468.47 1015.31 1471.67 1016.16 Q1474.87 1017.01 1478.19 1017.01 Q1485.52 1017.01 1489.17 1013.16 Q1492.81 1009.35 1492.81 1001.62 L1492.81 997.929 Q1490.51 1001.94 1486.9 1003.92 Q1483.29 1005.91 1478.27 1005.91 Q1469.93 1005.91 1464.82 999.549 Q1459.72 993.189 1459.72 982.698 Q1459.72 972.165 1464.82 965.805 Q1469.93 959.445 1478.27 959.445 Q1483.29 959.445 1486.9 961.43 Q1490.51 963.415 1492.81 967.426 L1492.81 960.539 L1500.27 960.539 L1500.27 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1515.62 942.877 L1523.07 942.877 L1523.07 1005.91 L1515.62 1005.91 L1515.62 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1577.48 981.361 L1577.48 985.007 L1543.21 985.007 Q1543.69 992.703 1547.83 996.754 Q1552 1000.76 1559.41 1000.76 Q1563.7 1000.76 1567.72 999.711 Q1571.77 998.658 1575.74 996.552 L1575.74 1003.6 Q1571.73 1005.3 1567.51 1006.19 Q1563.3 1007.08 1558.97 1007.08 Q1548.11 1007.08 1541.75 1000.76 Q1535.43 994.445 1535.43 983.67 Q1535.43 972.53 1541.42 966.008 Q1547.46 959.445 1557.67 959.445 Q1566.82 959.445 1572.13 965.36 Q1577.48 971.234 1577.48 981.361 M1570.02 979.173 Q1569.94 973.057 1566.58 969.411 Q1563.26 965.765 1557.75 965.765 Q1551.51 965.765 1547.74 969.289 Q1544.02 972.813 1543.45 979.214 L1570.02 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1633.99 942.958 Q1628.56 952.275 1625.93 961.39 Q1623.29 970.504 1623.29 979.862 Q1623.29 989.22 1625.93 998.415 Q1628.6 1007.57 1633.99 1016.85 L1627.51 1016.85 Q1621.43 1007.33 1618.39 998.132 Q1615.39 988.936 1615.39 979.862 Q1615.39 970.829 1618.39 961.673 Q1621.39 952.518 1627.51 942.958 L1633.99 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1678.3 967.426 L1678.3 942.877 L1685.76 942.877 L1685.76 1005.91 L1678.3 1005.91 L1678.3 999.104 Q1675.96 1003.15 1672.35 1005.14 Q1668.79 1007.08 1663.76 1007.08 Q1655.54 1007.08 1650.35 1000.52 Q1645.21 993.959 1645.21 983.265 Q1645.21 972.57 1650.35 966.008 Q1655.54 959.445 1663.76 959.445 Q1668.79 959.445 1672.35 961.43 Q1675.96 963.375 1678.3 967.426 M1652.91 983.265 Q1652.91 991.488 1656.27 996.187 Q1659.67 1000.85 1665.59 1000.85 Q1671.5 1000.85 1674.9 996.187 Q1678.3 991.488 1678.3 983.265 Q1678.3 975.041 1674.9 970.383 Q1671.5 965.684 1665.59 965.684 Q1659.67 965.684 1656.27 970.383 Q1652.91 975.041 1652.91 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1739.92 981.361 L1739.92 985.007 L1705.65 985.007 Q1706.13 992.703 1710.27 996.754 Q1714.44 1000.76 1721.85 1000.76 Q1726.15 1000.76 1730.16 999.711 Q1734.21 998.658 1738.18 996.552 L1738.18 1003.6 Q1734.17 1005.3 1729.95 1006.19 Q1725.74 1007.08 1721.41 1007.08 Q1710.55 1007.08 1704.19 1000.76 Q1697.87 994.445 1697.87 983.67 Q1697.87 972.53 1703.87 966.008 Q1709.9 959.445 1720.11 959.445 Q1729.27 959.445 1734.57 965.36 Q1739.92 971.234 1739.92 981.361 M1732.47 979.173 Q1732.38 973.057 1729.02 969.411 Q1725.7 965.765 1720.19 965.765 Q1713.95 965.765 1710.19 969.289 Q1706.46 972.813 1705.89 979.214 L1732.47 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1782.01 982.698 Q1782.01 974.596 1778.65 970.14 Q1775.32 965.684 1769.29 965.684 Q1763.29 965.684 1759.93 970.14 Q1756.61 974.596 1756.61 982.698 Q1756.61 990.759 1759.93 995.215 Q1763.29 999.671 1769.29 999.671 Q1775.32 999.671 1778.65 995.215 Q1782.01 990.759 1782.01 982.698 M1789.46 1000.28 Q1789.46 1011.86 1784.32 1017.49 Q1779.17 1023.17 1768.56 1023.17 Q1764.63 1023.17 1761.15 1022.56 Q1757.66 1021.99 1754.38 1020.78 L1754.38 1013.53 Q1757.66 1015.31 1760.86 1016.16 Q1764.06 1017.01 1767.38 1017.01 Q1774.72 1017.01 1778.36 1013.16 Q1782.01 1009.35 1782.01 1001.62 L1782.01 997.929 Q1779.7 1001.94 1776.09 1003.92 Q1772.49 1005.91 1767.47 1005.91 Q1759.12 1005.91 1754.02 999.549 Q1748.91 993.189 1748.91 982.698 Q1748.91 972.165 1754.02 965.805 Q1759.12 959.445 1767.47 959.445 Q1772.49 959.445 1776.09 961.43 Q1779.7 963.415 1782.01 967.426 L1782.01 960.539 L1789.46 960.539 L1789.46 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1803.64 942.958 L1810.12 942.958 Q1816.2 952.518 1819.2 961.673 Q1822.23 970.829 1822.23 979.862 Q1822.23 988.936 1819.2 998.132 Q1816.2 1007.33 1810.12 1016.85 L1803.64 1016.85 Q1809.03 1007.57 1811.66 998.415 Q1814.33 989.22 1814.33 979.862 Q1814.33 970.504 1811.66 961.39 Q1809.03 952.275 1803.64 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip093)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,1705.81 266.073,1705.04 270.902,1704.26 275.73,1703.49 280.558,1702.72 285.386,1701.96 290.214,1701.2 295.043,1700.44 299.871,1699.68 304.699,1698.94 309.527,1698.19 314.356,1697.46 319.184,1696.74 324.012,1696.02 328.84,1695.31 333.668,1694.62 338.497,1693.94 343.325,1693.27 348.153,1692.62 352.981,1691.98 357.81,1691.37 362.638,1690.77 367.466,1690.19 372.294,1689.63 377.123,1689.1 381.951,1688.59 386.779,1688.11 391.607,1687.65 396.435,1687.22 401.264,1686.81 406.092,1686.44 410.92,1686.09 415.748,1685.77 420.577,1685.47 425.405,1685.21 430.233,1684.96 435.061,1684.74 439.89,1684.54 444.718,1684.36 449.546,1684.2 454.374,1684.05 459.202,1683.91 464.031,1683.78 468.859,1683.65 473.687,1683.51 478.515,1683.37 483.344,1683.22 488.172,1683.06 493,1682.87 497.828,1682.67 502.656,1682.43 507.485,1682.17 512.313,1681.87 517.141,1681.54 521.969,1681.16 526.798,1680.75 531.626,1680.28 536.454,1679.77 541.282,1679.22 546.111,1678.61 550.939,1677.95 555.767,1677.25 560.595,1676.49 565.423,1675.68 570.252,1674.82 575.08,1673.92 579.908,1672.96 584.736,1671.96 589.565,1670.9 594.393,1669.81 599.221,1668.67 604.049,1667.48 608.878,1666.26 613.706,1664.99 618.534,1663.68 623.362,1662.34 628.19,1660.97 633.019,1659.55 637.847,1658.11 642.675,1656.64 647.503,1655.14 652.332,1653.61 657.16,1652.05 661.988,1650.47 666.816,1648.87 671.645,1647.24 676.473,1645.6 681.301,1643.94 686.129,1642.26 690.957,1640.56 695.786,1638.85 700.614,1637.13 705.442,1635.4 710.27,1633.66 715.099,1631.91 719.927,1630.15 724.755,1628.39 729.583,1626.62 734.411,1624.86 739.24,1623.09 744.068,1621.32 748.896,1619.55 753.724,1617.79 758.553,1616.03 763.381,1614.28 768.209,1612.54 773.037,1610.81 777.866,1609.1 782.694,1607.4 787.522,1605.71 792.35,1604.04 797.178,1602.39 802.007,1600.77 806.835,1599.17 811.663,1597.6 816.491,1596.05 821.32,1594.54 826.148,1593.06 830.976,1591.62 835.804,1590.21 840.633,1588.85 845.461,1587.52 850.289,1586.24 855.117,1585 859.945,1583.81 864.774,1582.67 869.602,1581.58 874.43,1580.54 879.258,1579.54 884.087,1578.59 888.915,1577.7 893.743,1576.85 898.571,1576.04 903.399,1575.28 908.228,1574.56 913.056,1573.87 917.884,1573.22 922.712,1572.6 927.541,1572 932.369,1571.42 937.197,1570.85 942.025,1570.29 946.854,1569.74 951.682,1569.18 956.51,1568.61 961.338,1568.02 966.166,1567.41 970.995,1566.78 975.823,1566.11 980.651,1565.41 985.479,1564.66 990.308,1563.88 995.136,1563.05 999.964,1562.16 1004.79,1561.23 1009.62,1560.24 1014.45,1559.2 1019.28,1558.11 1024.11,1556.96 1028.93,1555.76 1033.76,1554.51 1038.59,1553.2 1043.42,1551.85 1048.25,1550.45 1053.07,1549 1057.9,1547.51 1062.73,1545.98 1067.56,1544.42 1072.39,1542.82 1077.22,1541.18 1082.04,1539.52 1086.87,1537.84 1091.7,1536.13 1096.53,1534.4 1101.36,1532.66 1106.19,1530.91 1111.01,1529.15 1115.84,1527.38 1120.67,1525.62 1125.5,1523.85 1130.33,1522.1 1135.15,1520.35 1139.98,1518.61 1144.81,1516.89 1149.64,1515.19 1154.47,1513.51 1159.3,1511.86 1164.12,1510.23 1168.95,1508.64 1173.78,1507.09 1178.61,1505.57 1183.44,1504.09 1188.26,1502.65 1193.09,1501.26 1197.92,1499.91 1202.75,1498.61 1207.58,1497.36 1212.41,1496.16 1217.23,1495.01 1222.06,1493.91 1226.89,1492.85 1231.72,1491.84 1236.55,1490.87 1241.38,1489.94 1246.2,1489.05 1251.03,1488.2 1255.86,1487.37 1260.69,1486.57 1265.52,1485.79 1270.34,1485.03 1275.17,1484.27 1280,1483.51 1284.83,1482.75 1289.66,1481.99 1294.49,1481.2 1299.31,1480.4 1304.14,1479.57 1308.97,1478.72 1313.8,1477.83 1318.63,1476.9 1323.46,1475.93 1328.28,1474.92 1333.11,1473.86 1337.94,1472.76 1342.77,1471.61 1347.6,1470.42 1352.42,1469.17 1357.25,1467.89 1362.08,1466.55 1366.91,1465.18 1371.74,1463.77 1376.57,1462.32 1381.39,1460.83 1386.22,1459.32 1391.05,1457.78 1395.88,1456.21 1400.71,1454.62 1405.54,1453.02 1410.36,1451.4 1415.19,1449.78 1420.02,1448.15 1424.85,1446.52 1429.68,1444.9 1434.5,1443.28 1439.33,1441.68 1444.16,1440.08 1448.99,1438.51 1453.82,1436.96 1458.65,1435.44 1463.47,1433.94 1468.3,1432.47 1473.13,1431.04 1477.96,1429.64 1482.79,1428.27 1487.62,1426.95 1492.44,1425.66 1497.27,1424.41 1502.1,1423.2 1506.93,1422.03 1511.76,1420.9 1516.58,1419.8 1521.41,1418.74 1526.24,1417.7 1531.07,1416.7 1535.9,1415.72 1540.73,1414.75 1545.55,1413.81 1550.38,1412.87 1555.21,1411.95 1560.04,1411.02 1564.87,1410.1 1569.7,1409.16 1574.52,1408.22 1579.35,1407.26 1584.18,1406.28 1589.01,1405.28 1593.84,1404.25 1598.66,1403.2 1603.49,1402.11 1608.32,1400.99 1613.15,1399.84 1617.98,1398.66 1622.81,1397.44 1627.63,1396.19 1632.46,1394.9 1637.29,1393.59 1642.12,1392.25 1646.95,1390.88 1651.77,1389.48 1656.6,1388.06 1661.43,1386.63 1666.26,1385.18 1671.09,1383.71 1675.92,1382.24 1680.74,1380.76 1685.57,1379.28 1690.4,1377.8 1695.23,1376.32 1700.06,1374.85 1704.89,1373.39 1709.71,1371.94 1714.54,1370.51 1719.37,1369.1 1724.2,1367.7 1729.03,1366.33 1733.85,1364.98 1738.68,1363.66 1743.51,1362.36 1748.34,1361.09 1753.17,1359.84 1758,1358.62 1762.82,1357.43 1767.65,1356.26 1772.48,1355.11 1777.31,1353.98 1782.14,1352.88 1786.97,1351.78 1791.79,1350.71 1796.62,1349.64 1801.45,1348.58 1806.28,1347.53 1811.11,1346.48 1815.93,1345.43 1820.76,1344.37 1825.59,1343.31 1830.42,1342.23 1835.25,1341.15 1840.08,1340.05 1844.9,1338.93 1849.73,1337.8 1854.56,1336.65 1859.39,1335.47 1864.22,1334.28 1869.05,1333.07 1873.87,1331.84 1878.7,1330.58 1883.53,1329.32 1888.36,1328.03 1893.19,1326.73 1898.01,1325.41 1902.84,1324.08 1907.67,1322.74 1912.5,1321.39 1917.33,1320.04 1922.16,1318.68 1926.98,1317.33 1931.81,1315.97 1936.64,1314.61 1941.47,1313.26 1946.3,1311.92 1951.13,1310.58 1955.95,1309.25 1960.78,1307.94 1965.61,1306.63 1970.44,1305.35 1975.27,1304.07 1980.09,1302.81 1984.92,1301.56 1989.75,1300.33 1994.58,1299.11 1999.41,1297.91 2004.24,1296.72 2009.06,1295.54 2013.89,1294.37 2018.72,1293.21 2023.55,1292.06 2028.38,1290.92 2033.2,1289.78 2038.03,1288.65 2042.86,1287.51 2047.69,1286.38 2052.52,1285.25 2057.35,1284.11 2062.17,1282.97 2067,1281.83 2071.83,1280.68 2076.66,1279.52 2081.49,1278.35 2086.32,1277.18 2091.14,1275.99 2095.97,1274.8 2100.8,1273.6 2105.63,1272.39 2110.46,1271.17 2115.28,1269.94 2120.11,1268.7 2124.94,1267.46 2129.77,1266.22 2134.6,1264.97 2139.43,1263.71 2144.25,1262.45 2149.08,1261.19 2153.91,1259.93 2158.74,1258.67 2163.57,1257.41 2168.4,1256.16 2173.22,1254.9 2178.05,1253.65 2182.88,1252.4 2187.71,1251.16 2192.54,1249.92 2197.36,1248.69 2202.19,1247.46 2207.02,1246.24 2211.85,1245.02 2216.68,1243.81 2221.51,1242.6 2226.33,1241.4 2231.16,1240.21 2235.99,1239.01 2240.82,1237.82 2245.65,1236.64 2250.48,1235.46 2255.3,1234.28 2260.13,1233.1 2264.96,1231.92 2269.79,1230.75 2274.62,1229.58 2279.44,1228.4 2284.27,1227.23 2289.1,1226.06 2293.93,1224.89 2298.76,1223.71 2303.59,1222.54 2308.41,1221.37 2313.24,1220.2 2318.07,1219.02 2322.9,1217.85 2327.73,1216.67 2332.56,1215.5 2337.38,1214.32 2342.21,1213.15 2347.04,1211.97 2351.87,1210.79 2356.7,1209.61 2361.52,1208.43 2366.35,1207.25 2371.18,1206.06 2376.01,1204.87 2380.84,1203.68 2385.67,1202.49 2390.49,1201.29 2395.32,1200.09 2400.15,1198.89 2404.98,1197.68 2409.81,1196.46 2414.64,1195.24 2419.46,1194.02 2424.29,1192.79 2429.12,1191.55 2433.95,1190.31 2438.78,1189.06 2443.6,1187.81 2448.43,1186.55 2453.26,1185.28 2458.09,1184.02 2462.92,1182.75 2467.75,1181.47 2472.57,1180.2 2477.4,1178.93 2482.23,1177.65 2487.06,1176.38 2491.89,1175.1 2496.71,1173.83 2501.54,1172.57 2506.37,1171.3 2511.2,1170.04 2516.03,1168.79 2520.86,1167.53 2525.68,1166.28 2530.51,1165.03 2535.34,1163.79 2540.17,1162.54 2545,1161.28 2549.83,1160.02 2554.65,1158.76 2559.48,1157.48 2564.31,1156.18 2569.14,1154.86 2573.97,1153.52 2578.79,1152.15 2583.62,1150.74 2588.45,1149.28 2593.28,1147.77 2598.11,1146.2 2602.94,1144.57 2607.76,1142.87 2612.59,1141.08 2617.42,1139.2 2622.25,1137.23 2627.08,1135.16 2631.91,1132.98 2636.73,1130.69 2641.56,1128.29 2646.39,1125.77 2651.22,1123.14 2656.05,1120.39 2660.87,1117.54 2665.7,1114.57 2670.53,1111.5 2675.36,1108.33 2680.19,1105.07 "/>
+<path clip-path="url(#clip090)" d="M188.677 2657.17 L2752.76 2657.17 L2752.76 2020.38 L188.677 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip274">
+  <clipPath id="clip094">
     <rect x="188" y="2020" width="2565" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,2657.17 261.245,2020.38 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="863.571,2657.17 863.571,2020.38 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1465.9,2657.17 1465.9,2020.38 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2068.22,2657.17 2068.22,2020.38 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2670.55,2657.17 2670.55,2020.38 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2657.17 2752.76,2657.17 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,2657.17 261.245,2638.27 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.571,2657.17 863.571,2638.27 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1465.9,2657.17 1465.9,2638.27 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2068.22,2657.17 2068.22,2638.27 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2670.55,2657.17 2670.55,2638.27 "/>
-<path clip-path="url(#clip270)" d="M261.245 2691.29 Q257.634 2691.29 255.805 2694.85 Q254 2698.39 254 2705.52 Q254 2712.63 255.805 2716.19 Q257.634 2719.74 261.245 2719.74 Q264.879 2719.74 266.685 2716.19 Q268.514 2712.63 268.514 2705.52 Q268.514 2698.39 266.685 2694.85 Q264.879 2691.29 261.245 2691.29 M261.245 2687.58 Q267.055 2687.58 270.111 2692.19 Q273.189 2696.77 273.189 2705.52 Q273.189 2714.25 270.111 2718.86 Q267.055 2723.44 261.245 2723.44 Q255.435 2723.44 252.356 2718.86 Q249.301 2714.25 249.301 2705.52 Q249.301 2696.77 252.356 2692.19 Q255.435 2687.58 261.245 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M823.19 2688.21 L841.546 2688.21 L841.546 2692.14 L827.472 2692.14 L827.472 2700.62 Q828.491 2700.27 829.509 2700.11 Q830.528 2699.92 831.546 2699.92 Q837.333 2699.92 840.713 2703.09 Q844.092 2706.26 844.092 2711.68 Q844.092 2717.26 840.62 2720.36 Q837.148 2723.44 830.829 2723.44 Q828.653 2723.44 826.384 2723.07 Q824.139 2722.7 821.731 2721.96 L821.731 2717.26 Q823.815 2718.39 826.037 2718.95 Q828.259 2719.5 830.736 2719.5 Q834.741 2719.5 837.078 2717.4 Q839.416 2715.29 839.416 2711.68 Q839.416 2708.07 837.078 2705.96 Q834.741 2703.86 830.736 2703.86 Q828.861 2703.86 826.986 2704.27 Q825.134 2704.69 823.19 2705.57 L823.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M863.305 2691.29 Q859.694 2691.29 857.865 2694.85 Q856.06 2698.39 856.06 2705.52 Q856.06 2712.63 857.865 2716.19 Q859.694 2719.74 863.305 2719.74 Q866.939 2719.74 868.745 2716.19 Q870.574 2712.63 870.574 2705.52 Q870.574 2698.39 868.745 2694.85 Q866.939 2691.29 863.305 2691.29 M863.305 2687.58 Q869.115 2687.58 872.171 2692.19 Q875.25 2696.77 875.25 2705.52 Q875.25 2714.25 872.171 2718.86 Q869.115 2723.44 863.305 2723.44 Q857.495 2723.44 854.416 2718.86 Q851.361 2714.25 851.361 2705.52 Q851.361 2696.77 854.416 2692.19 Q857.495 2687.58 863.305 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M893.467 2691.29 Q889.856 2691.29 888.027 2694.85 Q886.222 2698.39 886.222 2705.52 Q886.222 2712.63 888.027 2716.19 Q889.856 2719.74 893.467 2719.74 Q897.101 2719.74 898.907 2716.19 Q900.736 2712.63 900.736 2705.52 Q900.736 2698.39 898.907 2694.85 Q897.101 2691.29 893.467 2691.29 M893.467 2687.58 Q899.277 2687.58 902.333 2692.19 Q905.411 2696.77 905.411 2705.52 Q905.411 2714.25 902.333 2718.86 Q899.277 2723.44 893.467 2723.44 Q887.657 2723.44 884.578 2718.86 Q881.523 2714.25 881.523 2705.52 Q881.523 2696.77 884.578 2692.19 Q887.657 2687.58 893.467 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1410.42 2718.83 L1418.06 2718.83 L1418.06 2692.47 L1409.75 2694.13 L1409.75 2689.87 L1418.02 2688.21 L1422.69 2688.21 L1422.69 2718.83 L1430.33 2718.83 L1430.33 2722.77 L1410.42 2722.77 L1410.42 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1449.78 2691.29 Q1446.16 2691.29 1444.34 2694.85 Q1442.53 2698.39 1442.53 2705.52 Q1442.53 2712.63 1444.34 2716.19 Q1446.16 2719.74 1449.78 2719.74 Q1453.41 2719.74 1455.21 2716.19 Q1457.04 2712.63 1457.04 2705.52 Q1457.04 2698.39 1455.21 2694.85 Q1453.41 2691.29 1449.78 2691.29 M1449.78 2687.58 Q1455.59 2687.58 1458.64 2692.19 Q1461.72 2696.77 1461.72 2705.52 Q1461.72 2714.25 1458.64 2718.86 Q1455.59 2723.44 1449.78 2723.44 Q1443.96 2723.44 1440.89 2718.86 Q1437.83 2714.25 1437.83 2705.52 Q1437.83 2696.77 1440.89 2692.19 Q1443.96 2687.58 1449.78 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1479.94 2691.29 Q1476.33 2691.29 1474.5 2694.85 Q1472.69 2698.39 1472.69 2705.52 Q1472.69 2712.63 1474.5 2716.19 Q1476.33 2719.74 1479.94 2719.74 Q1483.57 2719.74 1485.38 2716.19 Q1487.21 2712.63 1487.21 2705.52 Q1487.21 2698.39 1485.38 2694.85 Q1483.57 2691.29 1479.94 2691.29 M1479.94 2687.58 Q1485.75 2687.58 1488.8 2692.19 Q1491.88 2696.77 1491.88 2705.52 Q1491.88 2714.25 1488.8 2718.86 Q1485.75 2723.44 1479.94 2723.44 Q1474.13 2723.44 1471.05 2718.86 Q1467.99 2714.25 1467.99 2705.52 Q1467.99 2696.77 1471.05 2692.19 Q1474.13 2687.58 1479.94 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1510.1 2691.29 Q1506.49 2691.29 1504.66 2694.85 Q1502.85 2698.39 1502.85 2705.52 Q1502.85 2712.63 1504.66 2716.19 Q1506.49 2719.74 1510.1 2719.74 Q1513.73 2719.74 1515.54 2716.19 Q1517.37 2712.63 1517.37 2705.52 Q1517.37 2698.39 1515.54 2694.85 Q1513.73 2691.29 1510.1 2691.29 M1510.1 2687.58 Q1515.91 2687.58 1518.96 2692.19 Q1522.04 2696.77 1522.04 2705.52 Q1522.04 2714.25 1518.96 2718.86 Q1515.91 2723.44 1510.1 2723.44 Q1504.29 2723.44 1501.21 2718.86 Q1498.15 2714.25 1498.15 2705.52 Q1498.15 2696.77 1501.21 2692.19 Q1504.29 2687.58 1510.1 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2012.75 2718.83 L2020.39 2718.83 L2020.39 2692.47 L2012.08 2694.13 L2012.08 2689.87 L2020.34 2688.21 L2025.02 2688.21 L2025.02 2718.83 L2032.66 2718.83 L2032.66 2722.77 L2012.75 2722.77 L2012.75 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2042.15 2688.21 L2060.5 2688.21 L2060.5 2692.14 L2046.43 2692.14 L2046.43 2700.62 Q2047.45 2700.27 2048.47 2700.11 Q2049.49 2699.92 2050.5 2699.92 Q2056.29 2699.92 2059.67 2703.09 Q2063.05 2706.26 2063.05 2711.68 Q2063.05 2717.26 2059.58 2720.36 Q2056.11 2723.44 2049.79 2723.44 Q2047.61 2723.44 2045.34 2723.07 Q2043.1 2722.7 2040.69 2721.96 L2040.69 2717.26 Q2042.77 2718.39 2045 2718.95 Q2047.22 2719.5 2049.69 2719.5 Q2053.7 2719.5 2056.04 2717.4 Q2058.37 2715.29 2058.37 2711.68 Q2058.37 2708.07 2056.04 2705.96 Q2053.7 2703.86 2049.69 2703.86 Q2047.82 2703.86 2045.94 2704.27 Q2044.09 2704.69 2042.15 2705.57 L2042.15 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2082.26 2691.29 Q2078.65 2691.29 2076.82 2694.85 Q2075.02 2698.39 2075.02 2705.52 Q2075.02 2712.63 2076.82 2716.19 Q2078.65 2719.74 2082.26 2719.74 Q2085.9 2719.74 2087.7 2716.19 Q2089.53 2712.63 2089.53 2705.52 Q2089.53 2698.39 2087.7 2694.85 Q2085.9 2691.29 2082.26 2691.29 M2082.26 2687.58 Q2088.07 2687.58 2091.13 2692.19 Q2094.21 2696.77 2094.21 2705.52 Q2094.21 2714.25 2091.13 2718.86 Q2088.07 2723.44 2082.26 2723.44 Q2076.45 2723.44 2073.37 2718.86 Q2070.32 2714.25 2070.32 2705.52 Q2070.32 2696.77 2073.37 2692.19 Q2076.45 2687.58 2082.26 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2112.43 2691.29 Q2108.81 2691.29 2106.99 2694.85 Q2105.18 2698.39 2105.18 2705.52 Q2105.18 2712.63 2106.99 2716.19 Q2108.81 2719.74 2112.43 2719.74 Q2116.06 2719.74 2117.87 2716.19 Q2119.69 2712.63 2119.69 2705.52 Q2119.69 2698.39 2117.87 2694.85 Q2116.06 2691.29 2112.43 2691.29 M2112.43 2687.58 Q2118.24 2687.58 2121.29 2692.19 Q2124.37 2696.77 2124.37 2705.52 Q2124.37 2714.25 2121.29 2718.86 Q2118.24 2723.44 2112.43 2723.44 Q2106.62 2723.44 2103.54 2718.86 Q2100.48 2714.25 2100.48 2705.52 Q2100.48 2696.77 2103.54 2692.19 Q2106.62 2687.58 2112.43 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2619.16 2718.83 L2635.48 2718.83 L2635.48 2722.77 L2613.54 2722.77 L2613.54 2718.83 Q2616.2 2716.08 2620.78 2711.45 Q2625.39 2706.8 2626.57 2705.45 Q2628.81 2702.93 2629.69 2701.19 Q2630.6 2699.43 2630.6 2697.74 Q2630.6 2694.99 2628.65 2693.25 Q2626.73 2691.52 2623.63 2691.52 Q2621.43 2691.52 2618.98 2692.28 Q2616.55 2693.05 2613.77 2694.6 L2613.77 2689.87 Q2616.59 2688.74 2619.05 2688.16 Q2621.5 2687.58 2623.54 2687.58 Q2628.91 2687.58 2632.1 2690.27 Q2635.3 2692.95 2635.3 2697.44 Q2635.3 2699.57 2634.49 2701.49 Q2633.7 2703.39 2631.59 2705.99 Q2631.01 2706.66 2627.91 2709.87 Q2624.81 2713.07 2619.16 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2655.3 2691.29 Q2651.68 2691.29 2649.86 2694.85 Q2648.05 2698.39 2648.05 2705.52 Q2648.05 2712.63 2649.86 2716.19 Q2651.68 2719.74 2655.3 2719.74 Q2658.93 2719.74 2660.74 2716.19 Q2662.56 2712.63 2662.56 2705.52 Q2662.56 2698.39 2660.74 2694.85 Q2658.93 2691.29 2655.3 2691.29 M2655.3 2687.58 Q2661.11 2687.58 2664.16 2692.19 Q2667.24 2696.77 2667.24 2705.52 Q2667.24 2714.25 2664.16 2718.86 Q2661.11 2723.44 2655.3 2723.44 Q2649.49 2723.44 2646.41 2718.86 Q2643.35 2714.25 2643.35 2705.52 Q2643.35 2696.77 2646.41 2692.19 Q2649.49 2687.58 2655.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2685.46 2691.29 Q2681.85 2691.29 2680.02 2694.85 Q2678.21 2698.39 2678.21 2705.52 Q2678.21 2712.63 2680.02 2716.19 Q2681.85 2719.74 2685.46 2719.74 Q2689.09 2719.74 2690.9 2716.19 Q2692.73 2712.63 2692.73 2705.52 Q2692.73 2698.39 2690.9 2694.85 Q2689.09 2691.29 2685.46 2691.29 M2685.46 2687.58 Q2691.27 2687.58 2694.32 2692.19 Q2697.4 2696.77 2697.4 2705.52 Q2697.4 2714.25 2694.32 2718.86 Q2691.27 2723.44 2685.46 2723.44 Q2679.65 2723.44 2676.57 2718.86 Q2673.51 2714.25 2673.51 2705.52 Q2673.51 2696.77 2676.57 2692.19 Q2679.65 2687.58 2685.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M2715.62 2691.29 Q2712.01 2691.29 2710.18 2694.85 Q2708.37 2698.39 2708.37 2705.52 Q2708.37 2712.63 2710.18 2716.19 Q2712.01 2719.74 2715.62 2719.74 Q2719.25 2719.74 2721.06 2716.19 Q2722.89 2712.63 2722.89 2705.52 Q2722.89 2698.39 2721.06 2694.85 Q2719.25 2691.29 2715.62 2691.29 M2715.62 2687.58 Q2721.43 2687.58 2724.49 2692.19 Q2727.56 2696.77 2727.56 2705.52 Q2727.56 2714.25 2724.49 2718.86 Q2721.43 2723.44 2715.62 2723.44 Q2709.81 2723.44 2706.73 2718.86 Q2703.68 2714.25 2703.68 2705.52 Q2703.68 2696.77 2706.73 2692.19 Q2709.81 2687.58 2715.62 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2645.99 2752.76,2645.99 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2466.41 2752.76,2466.41 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2286.84 2752.76,2286.84 "/>
-<polyline clip-path="url(#clip274)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2107.27 2752.76,2107.27 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2657.17 188.677,2020.38 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2645.99 207.574,2645.99 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2466.41 207.574,2466.41 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2286.84 207.574,2286.84 "/>
-<polyline clip-path="url(#clip270)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2107.27 207.574,2107.27 "/>
-<path clip-path="url(#clip270)" d="M134.732 2631.79 Q131.121 2631.79 129.293 2635.35 Q127.487 2638.89 127.487 2646.02 Q127.487 2653.13 129.293 2656.69 Q131.121 2660.23 134.732 2660.23 Q138.367 2660.23 140.172 2656.69 Q142.001 2653.13 142.001 2646.02 Q142.001 2638.89 140.172 2635.35 Q138.367 2631.79 134.732 2631.79 M134.732 2628.08 Q140.543 2628.08 143.598 2632.69 Q146.677 2637.27 146.677 2646.02 Q146.677 2654.75 143.598 2659.35 Q140.543 2663.94 134.732 2663.94 Q128.922 2663.94 125.844 2659.35 Q122.788 2654.75 122.788 2646.02 Q122.788 2637.27 125.844 2632.69 Q128.922 2628.08 134.732 2628.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M94.6169 2449.13 L112.973 2449.13 L112.973 2453.07 L98.8993 2453.07 L98.8993 2461.54 Q99.9178 2461.19 100.936 2461.03 Q101.955 2460.85 102.973 2460.85 Q108.76 2460.85 112.14 2464.02 Q115.52 2467.19 115.52 2472.61 Q115.52 2478.18 112.047 2481.29 Q108.575 2484.36 102.256 2484.36 Q100.08 2484.36 97.8113 2483.99 Q95.566 2483.62 93.1586 2482.88 L93.1586 2478.18 Q95.2419 2479.32 97.4641 2479.87 Q99.6863 2480.43 102.163 2480.43 Q106.168 2480.43 108.506 2478.32 Q110.844 2476.22 110.844 2472.61 Q110.844 2468.99 108.506 2466.89 Q106.168 2464.78 102.163 2464.78 Q100.288 2464.78 98.4132 2465.2 Q96.5613 2465.62 94.6169 2466.49 L94.6169 2449.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 2452.21 Q131.121 2452.21 129.293 2455.78 Q127.487 2459.32 127.487 2466.45 Q127.487 2473.55 129.293 2477.12 Q131.121 2480.66 134.732 2480.66 Q138.367 2480.66 140.172 2477.12 Q142.001 2473.55 142.001 2466.45 Q142.001 2459.32 140.172 2455.78 Q138.367 2452.21 134.732 2452.21 M134.732 2448.51 Q140.543 2448.51 143.598 2453.12 Q146.677 2457.7 146.677 2466.45 Q146.677 2475.18 143.598 2479.78 Q140.543 2484.36 134.732 2484.36 Q128.922 2484.36 125.844 2479.78 Q122.788 2475.18 122.788 2466.45 Q122.788 2457.7 125.844 2453.12 Q128.922 2448.51 134.732 2448.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M65.2189 2300.19 L72.8577 2300.19 L72.8577 2273.82 L64.5476 2275.49 L64.5476 2271.23 L72.8114 2269.56 L77.4873 2269.56 L77.4873 2300.19 L85.1262 2300.19 L85.1262 2304.12 L65.2189 2304.12 L65.2189 2300.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M104.571 2272.64 Q100.959 2272.64 99.1307 2276.2 Q97.3252 2279.75 97.3252 2286.88 Q97.3252 2293.98 99.1307 2297.55 Q100.959 2301.09 104.571 2301.09 Q108.205 2301.09 110.01 2297.55 Q111.839 2293.98 111.839 2286.88 Q111.839 2279.75 110.01 2276.2 Q108.205 2272.64 104.571 2272.64 M104.571 2268.94 Q110.381 2268.94 113.436 2273.54 Q116.515 2278.13 116.515 2286.88 Q116.515 2295.6 113.436 2300.21 Q110.381 2304.79 104.571 2304.79 Q98.7604 2304.79 95.6817 2300.21 Q92.6262 2295.6 92.6262 2286.88 Q92.6262 2278.13 95.6817 2273.54 Q98.7604 2268.94 104.571 2268.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 2272.64 Q131.121 2272.64 129.293 2276.2 Q127.487 2279.75 127.487 2286.88 Q127.487 2293.98 129.293 2297.55 Q131.121 2301.09 134.732 2301.09 Q138.367 2301.09 140.172 2297.55 Q142.001 2293.98 142.001 2286.88 Q142.001 2279.75 140.172 2276.2 Q138.367 2272.64 134.732 2272.64 M134.732 2268.94 Q140.543 2268.94 143.598 2273.54 Q146.677 2278.13 146.677 2286.88 Q146.677 2295.6 143.598 2300.21 Q140.543 2304.79 134.732 2304.79 Q128.922 2304.79 125.844 2300.21 Q122.788 2295.6 122.788 2286.88 Q122.788 2278.13 125.844 2273.54 Q128.922 2268.94 134.732 2268.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M65.2189 2120.61 L72.8577 2120.61 L72.8577 2094.25 L64.5476 2095.91 L64.5476 2091.65 L72.8114 2089.99 L77.4873 2089.99 L77.4873 2120.61 L85.1262 2120.61 L85.1262 2124.55 L65.2189 2124.55 L65.2189 2120.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M94.6169 2089.99 L112.973 2089.99 L112.973 2093.92 L98.8993 2093.92 L98.8993 2102.39 Q99.9178 2102.05 100.936 2101.89 Q101.955 2101.7 102.973 2101.7 Q108.76 2101.7 112.14 2104.87 Q115.52 2108.04 115.52 2113.46 Q115.52 2119.04 112.047 2122.14 Q108.575 2125.22 102.256 2125.22 Q100.08 2125.22 97.8113 2124.85 Q95.566 2124.48 93.1586 2123.74 L93.1586 2119.04 Q95.2419 2120.17 97.4641 2120.73 Q99.6863 2121.28 102.163 2121.28 Q106.168 2121.28 108.506 2119.18 Q110.844 2117.07 110.844 2113.46 Q110.844 2109.85 108.506 2107.74 Q106.168 2105.64 102.163 2105.64 Q100.288 2105.64 98.4132 2106.05 Q96.5613 2106.47 94.6169 2107.35 L94.6169 2089.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M134.732 2093.07 Q131.121 2093.07 129.293 2096.63 Q127.487 2100.17 127.487 2107.3 Q127.487 2114.41 129.293 2117.97 Q131.121 2121.52 134.732 2121.52 Q138.367 2121.52 140.172 2117.97 Q142.001 2114.41 142.001 2107.3 Q142.001 2100.17 140.172 2096.63 Q138.367 2093.07 134.732 2093.07 M134.732 2089.36 Q140.543 2089.36 143.598 2093.97 Q146.677 2098.55 146.677 2107.3 Q146.677 2116.03 143.598 2120.64 Q140.543 2125.22 134.732 2125.22 Q128.922 2125.22 125.844 2120.64 Q122.788 2116.03 122.788 2107.3 Q122.788 2098.55 125.844 2093.97 Q128.922 2089.36 134.732 2089.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1013.96 1878.76 L1022.14 1878.76 L1022.14 1903.55 L1051.87 1903.55 L1051.87 1878.76 L1060.06 1878.76 L1060.06 1939.24 L1051.87 1939.24 L1051.87 1910.44 L1022.14 1910.44 L1022.14 1939.24 L1013.96 1939.24 L1013.96 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1114.82 1914.69 L1114.82 1918.34 L1080.55 1918.34 Q1081.04 1926.04 1085.17 1930.09 Q1089.34 1934.1 1096.76 1934.1 Q1101.05 1934.1 1105.06 1933.04 Q1109.11 1931.99 1113.08 1929.89 L1113.08 1936.93 Q1109.07 1938.64 1104.86 1939.53 Q1100.65 1940.42 1096.31 1940.42 Q1085.46 1940.42 1079.1 1934.1 Q1072.78 1927.78 1072.78 1917 Q1072.78 1905.86 1078.77 1899.34 Q1084.81 1892.78 1095.02 1892.78 Q1104.17 1892.78 1109.48 1898.69 Q1114.82 1904.57 1114.82 1914.69 M1107.37 1912.51 Q1107.29 1906.39 1103.93 1902.74 Q1100.61 1899.1 1095.1 1899.1 Q1088.86 1899.1 1085.09 1902.62 Q1081.36 1906.15 1080.8 1912.55 L1107.37 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1147.68 1916.44 Q1138.64 1916.44 1135.16 1918.5 Q1131.68 1920.57 1131.68 1925.55 Q1131.68 1929.52 1134.27 1931.87 Q1136.9 1934.18 1141.4 1934.18 Q1147.6 1934.18 1151.32 1929.8 Q1155.09 1925.39 1155.09 1918.1 L1155.09 1916.44 L1147.68 1916.44 M1162.54 1913.36 L1162.54 1939.24 L1155.09 1939.24 L1155.09 1932.36 Q1152.54 1936.49 1148.73 1938.47 Q1144.92 1940.42 1139.41 1940.42 Q1132.45 1940.42 1128.31 1936.53 Q1124.22 1932.6 1124.22 1926.04 Q1124.22 1918.38 1129.33 1914.49 Q1134.47 1910.6 1144.64 1910.6 L1155.09 1910.6 L1155.09 1909.87 Q1155.09 1904.73 1151.69 1901.93 Q1148.33 1899.1 1142.21 1899.1 Q1138.32 1899.1 1134.63 1900.03 Q1130.95 1900.96 1127.54 1902.83 L1127.54 1895.94 Q1131.64 1894.36 1135.48 1893.59 Q1139.33 1892.78 1142.98 1892.78 Q1152.82 1892.78 1157.68 1897.88 Q1162.54 1902.99 1162.54 1913.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1185.27 1880.99 L1185.27 1893.87 L1200.62 1893.87 L1200.62 1899.67 L1185.27 1899.67 L1185.27 1924.29 Q1185.27 1929.84 1186.77 1931.42 Q1188.31 1933 1192.97 1933 L1200.62 1933 L1200.62 1939.24 L1192.97 1939.24 Q1184.34 1939.24 1181.06 1936.04 Q1177.78 1932.8 1177.78 1924.29 L1177.78 1899.67 L1172.31 1899.67 L1172.31 1893.87 L1177.78 1893.87 L1177.78 1880.99 L1185.27 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1210.43 1893.87 L1217.88 1893.87 L1217.88 1939.24 L1210.43 1939.24 L1210.43 1893.87 M1210.43 1876.21 L1217.88 1876.21 L1217.88 1885.65 L1210.43 1885.65 L1210.43 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1271.19 1911.86 L1271.19 1939.24 L1263.74 1939.24 L1263.74 1912.1 Q1263.74 1905.66 1261.22 1902.46 Q1258.71 1899.26 1253.69 1899.26 Q1247.65 1899.26 1244.17 1903.11 Q1240.69 1906.96 1240.69 1913.6 L1240.69 1939.24 L1233.19 1939.24 L1233.19 1893.87 L1240.69 1893.87 L1240.69 1900.92 Q1243.36 1896.83 1246.96 1894.8 Q1250.61 1892.78 1255.35 1892.78 Q1263.17 1892.78 1267.18 1897.64 Q1271.19 1902.46 1271.19 1911.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1315.91 1916.03 Q1315.91 1907.93 1312.55 1903.47 Q1309.23 1899.02 1303.19 1899.02 Q1297.2 1899.02 1293.83 1903.47 Q1290.51 1907.93 1290.51 1916.03 Q1290.51 1924.09 1293.83 1928.55 Q1297.2 1933 1303.19 1933 Q1309.23 1933 1312.55 1928.55 Q1315.91 1924.09 1315.91 1916.03 M1323.36 1933.61 Q1323.36 1945.2 1318.22 1950.83 Q1313.08 1956.5 1302.46 1956.5 Q1298.53 1956.5 1295.05 1955.89 Q1291.57 1955.32 1288.28 1954.11 L1288.28 1946.86 Q1291.57 1948.64 1294.77 1949.49 Q1297.97 1950.34 1301.29 1950.34 Q1308.62 1950.34 1312.27 1946.49 Q1315.91 1942.69 1315.91 1934.95 L1315.91 1931.26 Q1313.6 1935.27 1310 1937.26 Q1306.39 1939.24 1301.37 1939.24 Q1293.02 1939.24 1287.92 1932.88 Q1282.82 1926.52 1282.82 1916.03 Q1282.82 1905.5 1287.92 1899.14 Q1293.02 1892.78 1301.37 1892.78 Q1306.39 1892.78 1310 1894.76 Q1313.6 1896.75 1315.91 1900.76 L1315.91 1893.87 L1323.36 1893.87 L1323.36 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1382.99 1876.29 Q1377.57 1885.61 1374.93 1894.72 Q1372.3 1903.84 1372.3 1913.2 Q1372.3 1922.55 1374.93 1931.75 Q1377.61 1940.9 1382.99 1950.18 L1376.51 1950.18 Q1370.44 1940.66 1367.4 1931.46 Q1364.4 1922.27 1364.4 1913.2 Q1364.4 1904.16 1367.4 1895.01 Q1370.4 1885.85 1376.51 1876.29 L1382.99 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1405.96 1910.36 L1405.96 1932.52 L1419.09 1932.52 Q1425.69 1932.52 1428.85 1929.8 Q1432.05 1927.05 1432.05 1921.42 Q1432.05 1915.75 1428.85 1913.07 Q1425.69 1910.36 1419.09 1910.36 L1405.96 1910.36 M1405.96 1885.49 L1405.96 1903.72 L1418.07 1903.72 Q1424.07 1903.72 1426.99 1901.49 Q1429.94 1899.22 1429.94 1894.6 Q1429.94 1890.02 1426.99 1887.76 Q1424.07 1885.49 1418.07 1885.49 L1405.96 1885.49 M1397.78 1878.76 L1418.68 1878.76 Q1428.04 1878.76 1433.1 1882.65 Q1438.17 1886.54 1438.17 1893.71 Q1438.17 1899.26 1435.57 1902.54 Q1432.98 1905.82 1427.96 1906.63 Q1433.99 1907.93 1437.32 1912.06 Q1440.68 1916.15 1440.68 1922.31 Q1440.68 1930.41 1435.17 1934.83 Q1429.66 1939.24 1419.49 1939.24 L1397.78 1939.24 L1397.78 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1446.31 1878.76 L1497.47 1878.76 L1497.47 1885.65 L1476 1885.65 L1476 1939.24 L1467.78 1939.24 L1467.78 1885.65 L1446.31 1885.65 L1446.31 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1504.44 1878.76 L1512.66 1878.76 L1512.66 1915.5 Q1512.66 1925.23 1516.19 1929.52 Q1519.71 1933.77 1527.61 1933.77 Q1535.47 1933.77 1538.99 1929.52 Q1542.52 1925.23 1542.52 1915.5 L1542.52 1878.76 L1550.74 1878.76 L1550.74 1916.52 Q1550.74 1928.35 1544.87 1934.38 Q1539.03 1940.42 1527.61 1940.42 Q1516.15 1940.42 1510.27 1934.38 Q1504.44 1928.35 1504.44 1916.52 L1504.44 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1579.02 1878.76 L1585.9 1878.76 L1564.84 1946.94 L1557.95 1946.94 L1579.02 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1616.69 1876.21 L1616.69 1882.41 L1609.56 1882.41 Q1605.55 1882.41 1603.97 1884.03 Q1602.43 1885.65 1602.43 1889.86 L1602.43 1893.87 L1614.71 1893.87 L1614.71 1899.67 L1602.43 1899.67 L1602.43 1939.24 L1594.94 1939.24 L1594.94 1899.67 L1587.81 1899.67 L1587.81 1893.87 L1594.94 1893.87 L1594.94 1890.71 Q1594.94 1883.14 1598.46 1879.69 Q1601.99 1876.21 1609.64 1876.21 L1616.69 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1628.84 1880.99 L1628.84 1893.87 L1644.2 1893.87 L1644.2 1899.67 L1628.84 1899.67 L1628.84 1924.29 Q1628.84 1929.84 1630.34 1931.42 Q1631.88 1933 1636.54 1933 L1644.2 1933 L1644.2 1939.24 L1636.54 1939.24 Q1627.91 1939.24 1624.63 1936.04 Q1621.35 1932.8 1621.35 1924.29 L1621.35 1899.67 L1615.88 1899.67 L1615.88 1893.87 L1621.35 1893.87 L1621.35 1880.99 L1628.84 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1667.25 1878.76 L1674.13 1878.76 L1653.07 1946.94 L1646.18 1946.94 L1667.25 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1704.92 1876.21 L1704.92 1882.41 L1697.79 1882.41 Q1693.78 1882.41 1692.2 1884.03 Q1690.66 1885.65 1690.66 1889.86 L1690.66 1893.87 L1702.93 1893.87 L1702.93 1899.67 L1690.66 1899.67 L1690.66 1939.24 L1683.17 1939.24 L1683.17 1899.67 L1676.04 1899.67 L1676.04 1893.87 L1683.17 1893.87 L1683.17 1890.71 Q1683.17 1883.14 1686.69 1879.69 Q1690.21 1876.21 1697.87 1876.21 L1704.92 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1717.07 1880.99 L1717.07 1893.87 L1732.43 1893.87 L1732.43 1899.67 L1717.07 1899.67 L1717.07 1924.29 Q1717.07 1929.84 1718.57 1931.42 Q1720.11 1933 1724.77 1933 L1732.43 1933 L1732.43 1939.24 L1724.77 1939.24 Q1716.14 1939.24 1712.86 1936.04 Q1709.58 1932.8 1709.58 1924.29 L1709.58 1899.67 L1704.11 1899.67 L1704.11 1893.87 L1709.58 1893.87 L1709.58 1880.99 L1717.07 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1755.47 1878.76 L1762.36 1878.76 L1741.3 1946.94 L1734.41 1946.94 L1755.47 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1799.1 1895.21 L1799.1 1902.26 Q1795.94 1900.64 1792.54 1899.83 Q1789.14 1899.02 1785.49 1899.02 Q1779.94 1899.02 1777.15 1900.72 Q1774.39 1902.42 1774.39 1905.82 Q1774.39 1908.42 1776.38 1909.91 Q1778.36 1911.37 1784.36 1912.71 L1786.91 1913.28 Q1794.85 1914.98 1798.17 1918.1 Q1801.53 1921.18 1801.53 1926.73 Q1801.53 1933.04 1796.51 1936.73 Q1791.53 1940.42 1782.78 1940.42 Q1779.13 1940.42 1775.16 1939.69 Q1771.23 1939 1766.86 1937.58 L1766.86 1929.89 Q1770.99 1932.03 1775 1933.13 Q1779.01 1934.18 1782.94 1934.18 Q1788.21 1934.18 1791.04 1932.4 Q1793.88 1930.57 1793.88 1927.29 Q1793.88 1924.25 1791.81 1922.63 Q1789.79 1921.01 1782.86 1919.51 L1780.27 1918.91 Q1773.34 1917.45 1770.26 1914.45 Q1767.18 1911.41 1767.18 1906.15 Q1767.18 1899.75 1771.72 1896.26 Q1776.26 1892.78 1784.6 1892.78 Q1788.73 1892.78 1792.38 1893.39 Q1796.02 1893.99 1799.1 1895.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1852.21 1914.69 L1852.21 1918.34 L1817.94 1918.34 Q1818.43 1926.04 1822.56 1930.09 Q1826.73 1934.1 1834.14 1934.1 Q1838.44 1934.1 1842.45 1933.04 Q1846.5 1931.99 1850.47 1929.89 L1850.47 1936.93 Q1846.46 1938.64 1842.25 1939.53 Q1838.03 1940.42 1833.7 1940.42 Q1822.84 1940.42 1816.48 1934.1 Q1810.16 1927.78 1810.16 1917 Q1810.16 1905.86 1816.16 1899.34 Q1822.19 1892.78 1832.4 1892.78 Q1841.56 1892.78 1846.86 1898.69 Q1852.21 1904.57 1852.21 1914.69 M1844.76 1912.51 Q1844.68 1906.39 1841.31 1902.74 Q1837.99 1899.1 1832.48 1899.1 Q1826.24 1899.1 1822.48 1902.62 Q1818.75 1906.15 1818.18 1912.55 L1844.76 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1897.09 1895.61 L1897.09 1902.58 Q1893.93 1900.84 1890.73 1899.99 Q1887.57 1899.1 1884.33 1899.1 Q1877.08 1899.1 1873.07 1903.72 Q1869.06 1908.29 1869.06 1916.6 Q1869.06 1924.9 1873.07 1929.52 Q1877.08 1934.1 1884.33 1934.1 Q1887.57 1934.1 1890.73 1933.25 Q1893.93 1932.36 1897.09 1930.61 L1897.09 1937.5 Q1893.98 1938.96 1890.61 1939.69 Q1887.29 1940.42 1883.52 1940.42 Q1873.27 1940.42 1867.24 1933.98 Q1861.2 1927.54 1861.2 1916.6 Q1861.2 1905.5 1867.28 1899.14 Q1873.4 1892.78 1884.01 1892.78 Q1887.45 1892.78 1890.73 1893.51 Q1894.02 1894.2 1897.09 1895.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip270)" d="M1908.88 1876.29 L1915.36 1876.29 Q1921.44 1885.85 1924.44 1895.01 Q1927.48 1904.16 1927.48 1913.2 Q1927.48 1922.27 1924.44 1931.46 Q1921.44 1940.66 1915.36 1950.18 L1908.88 1950.18 Q1914.27 1940.9 1916.9 1931.75 Q1919.58 1922.55 1919.58 1913.2 Q1919.58 1903.84 1916.9 1894.72 Q1914.27 1885.61 1908.88 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip274)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,2459.47 266.064,2452.39 270.882,2445.05 275.701,2437.44 280.52,2429.56 285.338,2421.41 290.157,2412.97 294.975,2404.26 299.794,2395.25 304.613,2385.95 309.431,2376.37 314.25,2366.5 319.068,2356.34 323.887,2345.9 328.706,2335.18 333.524,2324.19 338.343,2312.95 343.161,2301.47 347.98,2289.76 352.799,2277.84 357.617,2265.74 362.436,2253.48 367.254,2241.1 372.073,2228.62 376.892,2216.09 381.71,2203.55 386.529,2191.05 391.348,2178.64 396.166,2166.37 400.985,2154.3 405.803,2142.5 410.622,2131.01 415.441,2119.93 420.259,2109.29 425.078,2099.19 429.896,2089.67 434.715,2080.8 439.534,2072.65 444.352,2065.26 449.171,2058.7 453.989,2053 458.808,2048.2 463.627,2044.33 468.445,2041.41 473.264,2039.45 478.083,2038.45 482.901,2038.4 487.72,2039.29 492.538,2041.08 497.357,2043.75 502.176,2047.25 506.994,2051.53 511.813,2056.54 516.631,2062.22 521.45,2068.51 526.269,2075.34 531.087,2082.66 535.906,2090.39 540.724,2098.47 545.543,2106.83 550.362,2115.43 555.18,2124.2 559.999,2133.08 564.818,2142.03 569.636,2150.99 574.455,2159.92 579.273,2168.79 584.092,2177.55 588.911,2186.18 593.729,2194.63 598.548,2202.9 603.366,2210.96 608.185,2218.78 613.004,2226.35 617.822,2233.65 622.641,2240.68 627.459,2247.42 632.278,2253.87 637.097,2260.02 641.915,2265.86 646.734,2271.4 651.553,2276.62 656.371,2281.53 661.19,2286.13 666.008,2290.41 670.827,2294.38 675.646,2298.04 680.464,2301.38 685.283,2304.42 690.101,2307.15 694.92,2309.57 699.739,2311.68 704.557,2313.5 709.376,2315.01 714.194,2316.22 719.013,2317.13 723.832,2317.75 728.65,2318.07 733.469,2318.09 738.288,2317.83 743.106,2317.27 747.925,2316.43 752.743,2315.3 757.562,2313.88 762.381,2312.18 767.199,2310.2 772.018,2307.93 776.836,2305.4 781.655,2302.59 786.474,2299.51 791.292,2296.17 796.111,2292.57 800.929,2288.73 805.748,2284.64 810.567,2280.33 815.385,2275.79 820.204,2271.06 825.023,2266.13 829.841,2261.03 834.66,2255.78 839.478,2250.41 844.297,2244.93 849.116,2239.38 853.934,2233.78 858.753,2228.18 863.571,2222.59 868.39,2217.07 873.209,2211.66 878.027,2206.38 882.846,2201.3 887.664,2196.44 892.483,2191.87 897.302,2187.61 902.12,2183.72 906.939,2180.22 911.758,2177.17 916.576,2174.59 921.395,2172.51 926.213,2170.96 931.032,2169.96 935.851,2169.51 940.669,2169.63 945.488,2170.3 950.306,2171.53 955.125,2173.3 959.944,2175.59 964.762,2178.37 969.581,2181.62 974.399,2185.28 979.218,2189.34 984.037,2193.73 988.855,2198.42 993.674,2203.37 998.492,2208.52 1003.31,2213.84 1008.13,2219.27 1012.95,2224.77 1017.77,2230.31 1022.59,2235.83 1027.4,2241.32 1032.22,2246.72 1037.04,2252.02 1041.86,2257.17 1046.68,2262.17 1051.5,2266.98 1056.32,2271.58 1061.13,2275.95 1065.95,2280.09 1070.77,2283.98 1075.59,2287.6 1080.41,2290.95 1085.23,2294.03 1090.05,2296.82 1094.86,2299.33 1099.68,2301.54 1104.5,2303.47 1109.32,2305.1 1114.14,2306.45 1118.96,2307.51 1123.78,2308.29 1128.59,2308.8 1133.41,2309.03 1138.23,2309.01 1143.05,2308.73 1147.87,2308.2 1152.69,2307.45 1157.51,2306.48 1162.33,2305.3 1167.14,2303.93 1171.96,2302.4 1176.78,2300.72 1181.6,2298.9 1186.42,2296.98 1191.24,2294.98 1196.06,2292.93 1200.87,2290.84 1205.69,2288.76 1210.51,2286.71 1215.33,2284.73 1220.15,2282.83 1224.97,2281.07 1229.79,2279.46 1234.6,2278.04 1239.42,2276.83 1244.24,2275.87 1249.06,2275.17 1253.88,2274.75 1258.7,2274.64 1263.52,2274.85 1268.33,2275.38 1273.15,2276.24 1277.97,2277.43 1282.79,2278.94 1287.61,2280.76 1292.43,2282.88 1297.25,2285.29 1302.06,2287.95 1306.88,2290.85 1311.7,2293.96 1316.52,2297.24 1321.34,2300.68 1326.16,2304.24 1330.98,2307.89 1335.8,2311.59 1340.61,2315.32 1345.43,2319.06 1350.25,2322.76 1355.07,2326.42 1359.89,2330 1364.71,2333.48 1369.53,2336.85 1374.34,2340.08 1379.16,2343.17 1383.98,2346.1 1388.8,2348.85 1393.62,2351.43 1398.44,2353.82 1403.26,2356.02 1408.07,2358.03 1412.89,2359.84 1417.71,2361.46 1422.53,2362.89 1427.35,2364.13 1432.17,2365.19 1436.99,2366.08 1441.8,2366.8 1446.62,2367.36 1451.44,2367.78 1456.26,2368.07 1461.08,2368.24 1465.9,2368.31 1470.72,2368.29 1475.53,2368.2 1480.35,2368.06 1485.17,2367.9 1489.99,2367.72 1494.81,2367.55 1499.63,2367.42 1504.45,2367.33 1509.27,2367.31 1514.08,2367.38 1518.9,2367.56 1523.72,2367.86 1528.54,2368.3 1533.36,2368.89 1538.18,2369.64 1543,2370.55 1547.81,2371.64 1552.63,2372.91 1557.45,2374.34 1562.27,2375.95 1567.09,2377.72 1571.91,2379.65 1576.73,2381.72 1581.54,2383.92 1586.36,2386.23 1591.18,2388.64 1596,2391.14 1600.82,2393.7 1605.64,2396.31 1610.46,2398.94 1615.27,2401.59 1620.09,2404.23 1624.91,2406.85 1629.73,2409.43 1634.55,2411.96 1639.37,2414.43 1644.19,2416.83 1649,2419.14 1653.82,2421.36 1658.64,2423.48 1663.46,2425.49 1668.28,2427.4 1673.1,2429.21 1677.92,2430.9 1682.74,2432.48 1687.55,2433.96 1692.37,2435.34 1697.19,2436.62 1702.01,2437.8 1706.83,2438.9 1711.65,2439.92 1716.47,2440.87 1721.28,2441.76 1726.1,2442.6 1730.92,2443.41 1735.74,2444.18 1740.56,2444.94 1745.38,2445.69 1750.2,2446.44 1755.01,2447.22 1759.83,2448.02 1764.65,2448.86 1769.47,2449.75 1774.29,2450.7 1779.11,2451.7 1783.93,2452.78 1788.74,2453.93 1793.56,2455.16 1798.38,2456.47 1803.2,2457.86 1808.02,2459.33 1812.84,2460.88 1817.66,2462.49 1822.47,2464.18 1827.29,2465.92 1832.11,2467.72 1836.93,2469.56 1841.75,2471.45 1846.57,2473.35 1851.39,2475.28 1856.21,2477.22 1861.02,2479.16 1865.84,2481.09 1870.66,2483 1875.48,2484.89 1880.3,2486.76 1885.12,2488.58 1889.94,2490.37 1894.75,2492.11 1899.57,2493.8 1904.39,2495.44 1909.21,2497.02 1914.03,2498.56 1918.85,2500.03 1923.67,2501.46 1928.48,2502.83 1933.3,2504.15 1938.12,2505.43 1942.94,2506.67 1947.76,2507.86 1952.58,2509.02 1957.4,2510.15 1962.21,2511.26 1967.03,2512.34 1971.85,2513.42 1976.67,2514.48 1981.49,2515.55 1986.31,2516.61 1991.13,2517.68 1995.94,2518.77 2000.76,2519.87 2005.58,2520.99 2010.4,2522.13 2015.22,2523.3 2020.04,2524.5 2024.86,2525.72 2029.68,2526.97 2034.49,2528.25 2039.31,2529.56 2044.13,2530.9 2048.95,2532.25 2053.77,2533.63 2058.59,2535.03 2063.41,2536.44 2068.22,2537.86 2073.04,2539.29 2077.86,2540.72 2082.68,2542.15 2087.5,2543.58 2092.32,2545 2097.14,2546.41 2101.95,2547.8 2106.77,2549.17 2111.59,2550.53 2116.41,2551.86 2121.23,2553.17 2126.05,2554.45 2130.87,2555.71 2135.68,2556.94 2140.5,2558.15 2145.32,2559.33 2150.14,2560.48 2154.96,2561.61 2159.78,2562.72 2164.6,2563.81 2169.41,2564.88 2174.23,2565.93 2179.05,2566.96 2183.87,2567.99 2188.69,2569 2193.51,2570 2198.33,2571 2203.15,2571.99 2207.96,2572.98 2212.78,2573.97 2217.6,2574.96 2222.42,2575.96 2227.24,2576.96 2232.06,2577.96 2236.88,2578.96 2241.69,2579.97 2246.51,2580.99 2251.33,2582.01 2256.15,2583.04 2260.97,2584.06 2265.79,2585.09 2270.61,2586.13 2275.42,2587.16 2280.24,2588.19 2285.06,2589.21 2289.88,2590.23 2294.7,2591.25 2299.52,2592.25 2304.34,2593.25 2309.15,2594.24 2313.97,2595.21 2318.79,2596.17 2323.61,2597.12 2328.43,2598.05 2333.25,2598.97 2338.07,2599.87 2342.88,2600.76 2347.7,2601.63 2352.52,2602.48 2357.34,2603.32 2362.16,2604.14 2366.98,2604.95 2371.8,2605.74 2376.62,2606.52 2381.43,2607.29 2386.25,2608.04 2391.07,2608.79 2395.89,2609.52 2400.71,2610.25 2405.53,2610.97 2410.35,2611.68 2415.16,2612.38 2419.98,2613.09 2424.8,2613.78 2429.62,2614.48 2434.44,2615.17 2439.26,2615.85 2444.08,2616.54 2448.89,2617.22 2453.71,2617.91 2458.53,2618.58 2463.35,2619.26 2468.17,2619.93 2472.99,2620.6 2477.81,2621.27 2482.62,2621.92 2487.44,2622.58 2492.26,2623.22 2497.08,2623.86 2501.9,2624.49 2506.72,2625.11 2511.54,2625.71 2516.35,2626.31 2521.17,2626.89 2525.99,2627.46 2530.81,2628.01 2535.63,2628.55 2540.45,2629.07 2545.27,2629.57 2550.09,2630.06 2554.9,2630.53 2559.72,2630.99 2564.54,2631.43 2569.36,2631.85 2574.18,2632.26 2579,2632.65 2583.82,2633.03 2588.63,2633.39 2593.45,2633.75 2598.27,2634.09 2603.09,2634.41 2607.91,2634.73 2612.73,2635.05 2617.55,2635.35 2622.36,2635.65 2627.18,2635.95 2632,2636.25 2636.82,2636.55 2641.64,2636.85 2646.46,2637.15 2651.28,2637.46 2656.09,2637.78 2660.91,2638.11 2665.73,2638.44 2670.55,2638.79 2675.37,2639.15 2680.19,2637.25 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="261.245,2657.17 261.245,2020.38 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="863.571,2657.17 863.571,2020.38 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1465.9,2657.17 1465.9,2020.38 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2068.22,2657.17 2068.22,2020.38 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2670.55,2657.17 2670.55,2020.38 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2657.17 2752.76,2657.17 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="261.245,2657.17 261.245,2638.27 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="863.571,2657.17 863.571,2638.27 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1465.9,2657.17 1465.9,2638.27 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2068.22,2657.17 2068.22,2638.27 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2670.55,2657.17 2670.55,2638.27 "/>
+<path clip-path="url(#clip090)" d="M261.245 2691.29 Q257.634 2691.29 255.805 2694.85 Q254 2698.39 254 2705.52 Q254 2712.63 255.805 2716.19 Q257.634 2719.74 261.245 2719.74 Q264.879 2719.74 266.685 2716.19 Q268.514 2712.63 268.514 2705.52 Q268.514 2698.39 266.685 2694.85 Q264.879 2691.29 261.245 2691.29 M261.245 2687.58 Q267.055 2687.58 270.111 2692.19 Q273.189 2696.77 273.189 2705.52 Q273.189 2714.25 270.111 2718.86 Q267.055 2723.44 261.245 2723.44 Q255.435 2723.44 252.356 2718.86 Q249.301 2714.25 249.301 2705.52 Q249.301 2696.77 252.356 2692.19 Q255.435 2687.58 261.245 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M823.19 2688.21 L841.546 2688.21 L841.546 2692.14 L827.472 2692.14 L827.472 2700.62 Q828.491 2700.27 829.509 2700.11 Q830.528 2699.92 831.546 2699.92 Q837.333 2699.92 840.713 2703.09 Q844.092 2706.26 844.092 2711.68 Q844.092 2717.26 840.62 2720.36 Q837.148 2723.44 830.829 2723.44 Q828.653 2723.44 826.384 2723.07 Q824.139 2722.7 821.731 2721.96 L821.731 2717.26 Q823.815 2718.39 826.037 2718.95 Q828.259 2719.5 830.736 2719.5 Q834.741 2719.5 837.078 2717.4 Q839.416 2715.29 839.416 2711.68 Q839.416 2708.07 837.078 2705.96 Q834.741 2703.86 830.736 2703.86 Q828.861 2703.86 826.986 2704.27 Q825.134 2704.69 823.19 2705.57 L823.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M863.305 2691.29 Q859.694 2691.29 857.865 2694.85 Q856.06 2698.39 856.06 2705.52 Q856.06 2712.63 857.865 2716.19 Q859.694 2719.74 863.305 2719.74 Q866.939 2719.74 868.745 2716.19 Q870.574 2712.63 870.574 2705.52 Q870.574 2698.39 868.745 2694.85 Q866.939 2691.29 863.305 2691.29 M863.305 2687.58 Q869.115 2687.58 872.171 2692.19 Q875.25 2696.77 875.25 2705.52 Q875.25 2714.25 872.171 2718.86 Q869.115 2723.44 863.305 2723.44 Q857.495 2723.44 854.416 2718.86 Q851.361 2714.25 851.361 2705.52 Q851.361 2696.77 854.416 2692.19 Q857.495 2687.58 863.305 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M893.467 2691.29 Q889.856 2691.29 888.027 2694.85 Q886.222 2698.39 886.222 2705.52 Q886.222 2712.63 888.027 2716.19 Q889.856 2719.74 893.467 2719.74 Q897.101 2719.74 898.907 2716.19 Q900.736 2712.63 900.736 2705.52 Q900.736 2698.39 898.907 2694.85 Q897.101 2691.29 893.467 2691.29 M893.467 2687.58 Q899.277 2687.58 902.333 2692.19 Q905.411 2696.77 905.411 2705.52 Q905.411 2714.25 902.333 2718.86 Q899.277 2723.44 893.467 2723.44 Q887.657 2723.44 884.578 2718.86 Q881.523 2714.25 881.523 2705.52 Q881.523 2696.77 884.578 2692.19 Q887.657 2687.58 893.467 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1410.42 2718.83 L1418.06 2718.83 L1418.06 2692.47 L1409.75 2694.13 L1409.75 2689.87 L1418.02 2688.21 L1422.69 2688.21 L1422.69 2718.83 L1430.33 2718.83 L1430.33 2722.77 L1410.42 2722.77 L1410.42 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1449.78 2691.29 Q1446.16 2691.29 1444.34 2694.85 Q1442.53 2698.39 1442.53 2705.52 Q1442.53 2712.63 1444.34 2716.19 Q1446.16 2719.74 1449.78 2719.74 Q1453.41 2719.74 1455.21 2716.19 Q1457.04 2712.63 1457.04 2705.52 Q1457.04 2698.39 1455.21 2694.85 Q1453.41 2691.29 1449.78 2691.29 M1449.78 2687.58 Q1455.59 2687.58 1458.64 2692.19 Q1461.72 2696.77 1461.72 2705.52 Q1461.72 2714.25 1458.64 2718.86 Q1455.59 2723.44 1449.78 2723.44 Q1443.96 2723.44 1440.89 2718.86 Q1437.83 2714.25 1437.83 2705.52 Q1437.83 2696.77 1440.89 2692.19 Q1443.96 2687.58 1449.78 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1479.94 2691.29 Q1476.33 2691.29 1474.5 2694.85 Q1472.69 2698.39 1472.69 2705.52 Q1472.69 2712.63 1474.5 2716.19 Q1476.33 2719.74 1479.94 2719.74 Q1483.57 2719.74 1485.38 2716.19 Q1487.21 2712.63 1487.21 2705.52 Q1487.21 2698.39 1485.38 2694.85 Q1483.57 2691.29 1479.94 2691.29 M1479.94 2687.58 Q1485.75 2687.58 1488.8 2692.19 Q1491.88 2696.77 1491.88 2705.52 Q1491.88 2714.25 1488.8 2718.86 Q1485.75 2723.44 1479.94 2723.44 Q1474.13 2723.44 1471.05 2718.86 Q1467.99 2714.25 1467.99 2705.52 Q1467.99 2696.77 1471.05 2692.19 Q1474.13 2687.58 1479.94 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1510.1 2691.29 Q1506.49 2691.29 1504.66 2694.85 Q1502.85 2698.39 1502.85 2705.52 Q1502.85 2712.63 1504.66 2716.19 Q1506.49 2719.74 1510.1 2719.74 Q1513.73 2719.74 1515.54 2716.19 Q1517.37 2712.63 1517.37 2705.52 Q1517.37 2698.39 1515.54 2694.85 Q1513.73 2691.29 1510.1 2691.29 M1510.1 2687.58 Q1515.91 2687.58 1518.96 2692.19 Q1522.04 2696.77 1522.04 2705.52 Q1522.04 2714.25 1518.96 2718.86 Q1515.91 2723.44 1510.1 2723.44 Q1504.29 2723.44 1501.21 2718.86 Q1498.15 2714.25 1498.15 2705.52 Q1498.15 2696.77 1501.21 2692.19 Q1504.29 2687.58 1510.1 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2012.75 2718.83 L2020.39 2718.83 L2020.39 2692.47 L2012.08 2694.13 L2012.08 2689.87 L2020.34 2688.21 L2025.02 2688.21 L2025.02 2718.83 L2032.66 2718.83 L2032.66 2722.77 L2012.75 2722.77 L2012.75 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2042.15 2688.21 L2060.5 2688.21 L2060.5 2692.14 L2046.43 2692.14 L2046.43 2700.62 Q2047.45 2700.27 2048.47 2700.11 Q2049.49 2699.92 2050.5 2699.92 Q2056.29 2699.92 2059.67 2703.09 Q2063.05 2706.26 2063.05 2711.68 Q2063.05 2717.26 2059.58 2720.36 Q2056.11 2723.44 2049.79 2723.44 Q2047.61 2723.44 2045.34 2723.07 Q2043.1 2722.7 2040.69 2721.96 L2040.69 2717.26 Q2042.77 2718.39 2045 2718.95 Q2047.22 2719.5 2049.69 2719.5 Q2053.7 2719.5 2056.04 2717.4 Q2058.37 2715.29 2058.37 2711.68 Q2058.37 2708.07 2056.04 2705.96 Q2053.7 2703.86 2049.69 2703.86 Q2047.82 2703.86 2045.94 2704.27 Q2044.09 2704.69 2042.15 2705.57 L2042.15 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2082.26 2691.29 Q2078.65 2691.29 2076.82 2694.85 Q2075.02 2698.39 2075.02 2705.52 Q2075.02 2712.63 2076.82 2716.19 Q2078.65 2719.74 2082.26 2719.74 Q2085.9 2719.74 2087.7 2716.19 Q2089.53 2712.63 2089.53 2705.52 Q2089.53 2698.39 2087.7 2694.85 Q2085.9 2691.29 2082.26 2691.29 M2082.26 2687.58 Q2088.07 2687.58 2091.13 2692.19 Q2094.21 2696.77 2094.21 2705.52 Q2094.21 2714.25 2091.13 2718.86 Q2088.07 2723.44 2082.26 2723.44 Q2076.45 2723.44 2073.37 2718.86 Q2070.32 2714.25 2070.32 2705.52 Q2070.32 2696.77 2073.37 2692.19 Q2076.45 2687.58 2082.26 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2112.43 2691.29 Q2108.81 2691.29 2106.99 2694.85 Q2105.18 2698.39 2105.18 2705.52 Q2105.18 2712.63 2106.99 2716.19 Q2108.81 2719.74 2112.43 2719.74 Q2116.06 2719.74 2117.87 2716.19 Q2119.69 2712.63 2119.69 2705.52 Q2119.69 2698.39 2117.87 2694.85 Q2116.06 2691.29 2112.43 2691.29 M2112.43 2687.58 Q2118.24 2687.58 2121.29 2692.19 Q2124.37 2696.77 2124.37 2705.52 Q2124.37 2714.25 2121.29 2718.86 Q2118.24 2723.44 2112.43 2723.44 Q2106.62 2723.44 2103.54 2718.86 Q2100.48 2714.25 2100.48 2705.52 Q2100.48 2696.77 2103.54 2692.19 Q2106.62 2687.58 2112.43 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2619.16 2718.83 L2635.48 2718.83 L2635.48 2722.77 L2613.54 2722.77 L2613.54 2718.83 Q2616.2 2716.08 2620.78 2711.45 Q2625.39 2706.8 2626.57 2705.45 Q2628.81 2702.93 2629.69 2701.19 Q2630.6 2699.43 2630.6 2697.74 Q2630.6 2694.99 2628.65 2693.25 Q2626.73 2691.52 2623.63 2691.52 Q2621.43 2691.52 2618.98 2692.28 Q2616.55 2693.05 2613.77 2694.6 L2613.77 2689.87 Q2616.59 2688.74 2619.05 2688.16 Q2621.5 2687.58 2623.54 2687.58 Q2628.91 2687.58 2632.1 2690.27 Q2635.3 2692.95 2635.3 2697.44 Q2635.3 2699.57 2634.49 2701.49 Q2633.7 2703.39 2631.59 2705.99 Q2631.01 2706.66 2627.91 2709.87 Q2624.81 2713.07 2619.16 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2655.3 2691.29 Q2651.68 2691.29 2649.86 2694.85 Q2648.05 2698.39 2648.05 2705.52 Q2648.05 2712.63 2649.86 2716.19 Q2651.68 2719.74 2655.3 2719.74 Q2658.93 2719.74 2660.74 2716.19 Q2662.56 2712.63 2662.56 2705.52 Q2662.56 2698.39 2660.74 2694.85 Q2658.93 2691.29 2655.3 2691.29 M2655.3 2687.58 Q2661.11 2687.58 2664.16 2692.19 Q2667.24 2696.77 2667.24 2705.52 Q2667.24 2714.25 2664.16 2718.86 Q2661.11 2723.44 2655.3 2723.44 Q2649.49 2723.44 2646.41 2718.86 Q2643.35 2714.25 2643.35 2705.52 Q2643.35 2696.77 2646.41 2692.19 Q2649.49 2687.58 2655.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2685.46 2691.29 Q2681.85 2691.29 2680.02 2694.85 Q2678.21 2698.39 2678.21 2705.52 Q2678.21 2712.63 2680.02 2716.19 Q2681.85 2719.74 2685.46 2719.74 Q2689.09 2719.74 2690.9 2716.19 Q2692.73 2712.63 2692.73 2705.52 Q2692.73 2698.39 2690.9 2694.85 Q2689.09 2691.29 2685.46 2691.29 M2685.46 2687.58 Q2691.27 2687.58 2694.32 2692.19 Q2697.4 2696.77 2697.4 2705.52 Q2697.4 2714.25 2694.32 2718.86 Q2691.27 2723.44 2685.46 2723.44 Q2679.65 2723.44 2676.57 2718.86 Q2673.51 2714.25 2673.51 2705.52 Q2673.51 2696.77 2676.57 2692.19 Q2679.65 2687.58 2685.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M2715.62 2691.29 Q2712.01 2691.29 2710.18 2694.85 Q2708.37 2698.39 2708.37 2705.52 Q2708.37 2712.63 2710.18 2716.19 Q2712.01 2719.74 2715.62 2719.74 Q2719.25 2719.74 2721.06 2716.19 Q2722.89 2712.63 2722.89 2705.52 Q2722.89 2698.39 2721.06 2694.85 Q2719.25 2691.29 2715.62 2691.29 M2715.62 2687.58 Q2721.43 2687.58 2724.49 2692.19 Q2727.56 2696.77 2727.56 2705.52 Q2727.56 2714.25 2724.49 2718.86 Q2721.43 2723.44 2715.62 2723.44 Q2709.81 2723.44 2706.73 2718.86 Q2703.68 2714.25 2703.68 2705.52 Q2703.68 2696.77 2706.73 2692.19 Q2709.81 2687.58 2715.62 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2645.99 2752.76,2645.99 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2466.41 2752.76,2466.41 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2286.84 2752.76,2286.84 "/>
+<polyline clip-path="url(#clip094)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="188.677,2107.27 2752.76,2107.27 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2657.17 188.677,2020.38 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2645.99 207.574,2645.99 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2466.41 207.574,2466.41 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2286.84 207.574,2286.84 "/>
+<polyline clip-path="url(#clip090)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="188.677,2107.27 207.574,2107.27 "/>
+<path clip-path="url(#clip090)" d="M134.732 2631.79 Q131.121 2631.79 129.293 2635.35 Q127.487 2638.89 127.487 2646.02 Q127.487 2653.13 129.293 2656.69 Q131.121 2660.23 134.732 2660.23 Q138.367 2660.23 140.172 2656.69 Q142.001 2653.13 142.001 2646.02 Q142.001 2638.89 140.172 2635.35 Q138.367 2631.79 134.732 2631.79 M134.732 2628.08 Q140.543 2628.08 143.598 2632.69 Q146.677 2637.27 146.677 2646.02 Q146.677 2654.75 143.598 2659.35 Q140.543 2663.94 134.732 2663.94 Q128.922 2663.94 125.844 2659.35 Q122.788 2654.75 122.788 2646.02 Q122.788 2637.27 125.844 2632.69 Q128.922 2628.08 134.732 2628.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M94.6169 2449.13 L112.973 2449.13 L112.973 2453.07 L98.8993 2453.07 L98.8993 2461.54 Q99.9178 2461.19 100.936 2461.03 Q101.955 2460.85 102.973 2460.85 Q108.76 2460.85 112.14 2464.02 Q115.52 2467.19 115.52 2472.61 Q115.52 2478.18 112.047 2481.29 Q108.575 2484.36 102.256 2484.36 Q100.08 2484.36 97.8113 2483.99 Q95.566 2483.62 93.1586 2482.88 L93.1586 2478.18 Q95.2419 2479.32 97.4641 2479.87 Q99.6863 2480.43 102.163 2480.43 Q106.168 2480.43 108.506 2478.32 Q110.844 2476.22 110.844 2472.61 Q110.844 2468.99 108.506 2466.89 Q106.168 2464.78 102.163 2464.78 Q100.288 2464.78 98.4132 2465.2 Q96.5613 2465.62 94.6169 2466.49 L94.6169 2449.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 2452.21 Q131.121 2452.21 129.293 2455.78 Q127.487 2459.32 127.487 2466.45 Q127.487 2473.55 129.293 2477.12 Q131.121 2480.66 134.732 2480.66 Q138.367 2480.66 140.172 2477.12 Q142.001 2473.55 142.001 2466.45 Q142.001 2459.32 140.172 2455.78 Q138.367 2452.21 134.732 2452.21 M134.732 2448.51 Q140.543 2448.51 143.598 2453.12 Q146.677 2457.7 146.677 2466.45 Q146.677 2475.18 143.598 2479.78 Q140.543 2484.36 134.732 2484.36 Q128.922 2484.36 125.844 2479.78 Q122.788 2475.18 122.788 2466.45 Q122.788 2457.7 125.844 2453.12 Q128.922 2448.51 134.732 2448.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M65.2189 2300.19 L72.8577 2300.19 L72.8577 2273.82 L64.5476 2275.49 L64.5476 2271.23 L72.8114 2269.56 L77.4873 2269.56 L77.4873 2300.19 L85.1262 2300.19 L85.1262 2304.12 L65.2189 2304.12 L65.2189 2300.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M104.571 2272.64 Q100.959 2272.64 99.1307 2276.2 Q97.3252 2279.75 97.3252 2286.88 Q97.3252 2293.98 99.1307 2297.55 Q100.959 2301.09 104.571 2301.09 Q108.205 2301.09 110.01 2297.55 Q111.839 2293.98 111.839 2286.88 Q111.839 2279.75 110.01 2276.2 Q108.205 2272.64 104.571 2272.64 M104.571 2268.94 Q110.381 2268.94 113.436 2273.54 Q116.515 2278.13 116.515 2286.88 Q116.515 2295.6 113.436 2300.21 Q110.381 2304.79 104.571 2304.79 Q98.7604 2304.79 95.6817 2300.21 Q92.6262 2295.6 92.6262 2286.88 Q92.6262 2278.13 95.6817 2273.54 Q98.7604 2268.94 104.571 2268.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 2272.64 Q131.121 2272.64 129.293 2276.2 Q127.487 2279.75 127.487 2286.88 Q127.487 2293.98 129.293 2297.55 Q131.121 2301.09 134.732 2301.09 Q138.367 2301.09 140.172 2297.55 Q142.001 2293.98 142.001 2286.88 Q142.001 2279.75 140.172 2276.2 Q138.367 2272.64 134.732 2272.64 M134.732 2268.94 Q140.543 2268.94 143.598 2273.54 Q146.677 2278.13 146.677 2286.88 Q146.677 2295.6 143.598 2300.21 Q140.543 2304.79 134.732 2304.79 Q128.922 2304.79 125.844 2300.21 Q122.788 2295.6 122.788 2286.88 Q122.788 2278.13 125.844 2273.54 Q128.922 2268.94 134.732 2268.94 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M65.2189 2120.61 L72.8577 2120.61 L72.8577 2094.25 L64.5476 2095.91 L64.5476 2091.65 L72.8114 2089.99 L77.4873 2089.99 L77.4873 2120.61 L85.1262 2120.61 L85.1262 2124.55 L65.2189 2124.55 L65.2189 2120.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M94.6169 2089.99 L112.973 2089.99 L112.973 2093.92 L98.8993 2093.92 L98.8993 2102.39 Q99.9178 2102.05 100.936 2101.89 Q101.955 2101.7 102.973 2101.7 Q108.76 2101.7 112.14 2104.87 Q115.52 2108.04 115.52 2113.46 Q115.52 2119.04 112.047 2122.14 Q108.575 2125.22 102.256 2125.22 Q100.08 2125.22 97.8113 2124.85 Q95.566 2124.48 93.1586 2123.74 L93.1586 2119.04 Q95.2419 2120.17 97.4641 2120.73 Q99.6863 2121.28 102.163 2121.28 Q106.168 2121.28 108.506 2119.18 Q110.844 2117.07 110.844 2113.46 Q110.844 2109.85 108.506 2107.74 Q106.168 2105.64 102.163 2105.64 Q100.288 2105.64 98.4132 2106.05 Q96.5613 2106.47 94.6169 2107.35 L94.6169 2089.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M134.732 2093.07 Q131.121 2093.07 129.293 2096.63 Q127.487 2100.17 127.487 2107.3 Q127.487 2114.41 129.293 2117.97 Q131.121 2121.52 134.732 2121.52 Q138.367 2121.52 140.172 2117.97 Q142.001 2114.41 142.001 2107.3 Q142.001 2100.17 140.172 2096.63 Q138.367 2093.07 134.732 2093.07 M134.732 2089.36 Q140.543 2089.36 143.598 2093.97 Q146.677 2098.55 146.677 2107.3 Q146.677 2116.03 143.598 2120.64 Q140.543 2125.22 134.732 2125.22 Q128.922 2125.22 125.844 2120.64 Q122.788 2116.03 122.788 2107.3 Q122.788 2098.55 125.844 2093.97 Q128.922 2089.36 134.732 2089.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1013.96 1878.76 L1022.14 1878.76 L1022.14 1903.55 L1051.87 1903.55 L1051.87 1878.76 L1060.06 1878.76 L1060.06 1939.24 L1051.87 1939.24 L1051.87 1910.44 L1022.14 1910.44 L1022.14 1939.24 L1013.96 1939.24 L1013.96 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1114.82 1914.69 L1114.82 1918.34 L1080.55 1918.34 Q1081.04 1926.04 1085.17 1930.09 Q1089.34 1934.1 1096.76 1934.1 Q1101.05 1934.1 1105.06 1933.04 Q1109.11 1931.99 1113.08 1929.89 L1113.08 1936.93 Q1109.07 1938.64 1104.86 1939.53 Q1100.65 1940.42 1096.31 1940.42 Q1085.46 1940.42 1079.1 1934.1 Q1072.78 1927.78 1072.78 1917 Q1072.78 1905.86 1078.77 1899.34 Q1084.81 1892.78 1095.02 1892.78 Q1104.17 1892.78 1109.48 1898.69 Q1114.82 1904.57 1114.82 1914.69 M1107.37 1912.51 Q1107.29 1906.39 1103.93 1902.74 Q1100.61 1899.1 1095.1 1899.1 Q1088.86 1899.1 1085.09 1902.62 Q1081.36 1906.15 1080.8 1912.55 L1107.37 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1147.68 1916.44 Q1138.64 1916.44 1135.16 1918.5 Q1131.68 1920.57 1131.68 1925.55 Q1131.68 1929.52 1134.27 1931.87 Q1136.9 1934.18 1141.4 1934.18 Q1147.6 1934.18 1151.32 1929.8 Q1155.09 1925.39 1155.09 1918.1 L1155.09 1916.44 L1147.68 1916.44 M1162.54 1913.36 L1162.54 1939.24 L1155.09 1939.24 L1155.09 1932.36 Q1152.54 1936.49 1148.73 1938.47 Q1144.92 1940.42 1139.41 1940.42 Q1132.45 1940.42 1128.31 1936.53 Q1124.22 1932.6 1124.22 1926.04 Q1124.22 1918.38 1129.33 1914.49 Q1134.47 1910.6 1144.64 1910.6 L1155.09 1910.6 L1155.09 1909.87 Q1155.09 1904.73 1151.69 1901.93 Q1148.33 1899.1 1142.21 1899.1 Q1138.32 1899.1 1134.63 1900.03 Q1130.95 1900.96 1127.54 1902.83 L1127.54 1895.94 Q1131.64 1894.36 1135.48 1893.59 Q1139.33 1892.78 1142.98 1892.78 Q1152.82 1892.78 1157.68 1897.88 Q1162.54 1902.99 1162.54 1913.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1185.27 1880.99 L1185.27 1893.87 L1200.62 1893.87 L1200.62 1899.67 L1185.27 1899.67 L1185.27 1924.29 Q1185.27 1929.84 1186.77 1931.42 Q1188.31 1933 1192.97 1933 L1200.62 1933 L1200.62 1939.24 L1192.97 1939.24 Q1184.34 1939.24 1181.06 1936.04 Q1177.78 1932.8 1177.78 1924.29 L1177.78 1899.67 L1172.31 1899.67 L1172.31 1893.87 L1177.78 1893.87 L1177.78 1880.99 L1185.27 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1210.43 1893.87 L1217.88 1893.87 L1217.88 1939.24 L1210.43 1939.24 L1210.43 1893.87 M1210.43 1876.21 L1217.88 1876.21 L1217.88 1885.65 L1210.43 1885.65 L1210.43 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1271.19 1911.86 L1271.19 1939.24 L1263.74 1939.24 L1263.74 1912.1 Q1263.74 1905.66 1261.22 1902.46 Q1258.71 1899.26 1253.69 1899.26 Q1247.65 1899.26 1244.17 1903.11 Q1240.69 1906.96 1240.69 1913.6 L1240.69 1939.24 L1233.19 1939.24 L1233.19 1893.87 L1240.69 1893.87 L1240.69 1900.92 Q1243.36 1896.83 1246.96 1894.8 Q1250.61 1892.78 1255.35 1892.78 Q1263.17 1892.78 1267.18 1897.64 Q1271.19 1902.46 1271.19 1911.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1315.91 1916.03 Q1315.91 1907.93 1312.55 1903.47 Q1309.23 1899.02 1303.19 1899.02 Q1297.2 1899.02 1293.83 1903.47 Q1290.51 1907.93 1290.51 1916.03 Q1290.51 1924.09 1293.83 1928.55 Q1297.2 1933 1303.19 1933 Q1309.23 1933 1312.55 1928.55 Q1315.91 1924.09 1315.91 1916.03 M1323.36 1933.61 Q1323.36 1945.2 1318.22 1950.83 Q1313.08 1956.5 1302.46 1956.5 Q1298.53 1956.5 1295.05 1955.89 Q1291.57 1955.32 1288.28 1954.11 L1288.28 1946.86 Q1291.57 1948.64 1294.77 1949.49 Q1297.97 1950.34 1301.29 1950.34 Q1308.62 1950.34 1312.27 1946.49 Q1315.91 1942.69 1315.91 1934.95 L1315.91 1931.26 Q1313.6 1935.27 1310 1937.26 Q1306.39 1939.24 1301.37 1939.24 Q1293.02 1939.24 1287.92 1932.88 Q1282.82 1926.52 1282.82 1916.03 Q1282.82 1905.5 1287.92 1899.14 Q1293.02 1892.78 1301.37 1892.78 Q1306.39 1892.78 1310 1894.76 Q1313.6 1896.75 1315.91 1900.76 L1315.91 1893.87 L1323.36 1893.87 L1323.36 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1382.99 1876.29 Q1377.57 1885.61 1374.93 1894.72 Q1372.3 1903.84 1372.3 1913.2 Q1372.3 1922.55 1374.93 1931.75 Q1377.61 1940.9 1382.99 1950.18 L1376.51 1950.18 Q1370.44 1940.66 1367.4 1931.46 Q1364.4 1922.27 1364.4 1913.2 Q1364.4 1904.16 1367.4 1895.01 Q1370.4 1885.85 1376.51 1876.29 L1382.99 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1405.96 1910.36 L1405.96 1932.52 L1419.09 1932.52 Q1425.69 1932.52 1428.85 1929.8 Q1432.05 1927.05 1432.05 1921.42 Q1432.05 1915.75 1428.85 1913.07 Q1425.69 1910.36 1419.09 1910.36 L1405.96 1910.36 M1405.96 1885.49 L1405.96 1903.72 L1418.07 1903.72 Q1424.07 1903.72 1426.99 1901.49 Q1429.94 1899.22 1429.94 1894.6 Q1429.94 1890.02 1426.99 1887.76 Q1424.07 1885.49 1418.07 1885.49 L1405.96 1885.49 M1397.78 1878.76 L1418.68 1878.76 Q1428.04 1878.76 1433.1 1882.65 Q1438.17 1886.54 1438.17 1893.71 Q1438.17 1899.26 1435.57 1902.54 Q1432.98 1905.82 1427.96 1906.63 Q1433.99 1907.93 1437.32 1912.06 Q1440.68 1916.15 1440.68 1922.31 Q1440.68 1930.41 1435.17 1934.83 Q1429.66 1939.24 1419.49 1939.24 L1397.78 1939.24 L1397.78 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1446.31 1878.76 L1497.47 1878.76 L1497.47 1885.65 L1476 1885.65 L1476 1939.24 L1467.78 1939.24 L1467.78 1885.65 L1446.31 1885.65 L1446.31 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1504.44 1878.76 L1512.66 1878.76 L1512.66 1915.5 Q1512.66 1925.23 1516.19 1929.52 Q1519.71 1933.77 1527.61 1933.77 Q1535.47 1933.77 1538.99 1929.52 Q1542.52 1925.23 1542.52 1915.5 L1542.52 1878.76 L1550.74 1878.76 L1550.74 1916.52 Q1550.74 1928.35 1544.87 1934.38 Q1539.03 1940.42 1527.61 1940.42 Q1516.15 1940.42 1510.27 1934.38 Q1504.44 1928.35 1504.44 1916.52 L1504.44 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1579.02 1878.76 L1585.9 1878.76 L1564.84 1946.94 L1557.95 1946.94 L1579.02 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1616.69 1876.21 L1616.69 1882.41 L1609.56 1882.41 Q1605.55 1882.41 1603.97 1884.03 Q1602.43 1885.65 1602.43 1889.86 L1602.43 1893.87 L1614.71 1893.87 L1614.71 1899.67 L1602.43 1899.67 L1602.43 1939.24 L1594.94 1939.24 L1594.94 1899.67 L1587.81 1899.67 L1587.81 1893.87 L1594.94 1893.87 L1594.94 1890.71 Q1594.94 1883.14 1598.46 1879.69 Q1601.99 1876.21 1609.64 1876.21 L1616.69 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1628.84 1880.99 L1628.84 1893.87 L1644.2 1893.87 L1644.2 1899.67 L1628.84 1899.67 L1628.84 1924.29 Q1628.84 1929.84 1630.34 1931.42 Q1631.88 1933 1636.54 1933 L1644.2 1933 L1644.2 1939.24 L1636.54 1939.24 Q1627.91 1939.24 1624.63 1936.04 Q1621.35 1932.8 1621.35 1924.29 L1621.35 1899.67 L1615.88 1899.67 L1615.88 1893.87 L1621.35 1893.87 L1621.35 1880.99 L1628.84 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1667.25 1878.76 L1674.13 1878.76 L1653.07 1946.94 L1646.18 1946.94 L1667.25 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1704.92 1876.21 L1704.92 1882.41 L1697.79 1882.41 Q1693.78 1882.41 1692.2 1884.03 Q1690.66 1885.65 1690.66 1889.86 L1690.66 1893.87 L1702.93 1893.87 L1702.93 1899.67 L1690.66 1899.67 L1690.66 1939.24 L1683.17 1939.24 L1683.17 1899.67 L1676.04 1899.67 L1676.04 1893.87 L1683.17 1893.87 L1683.17 1890.71 Q1683.17 1883.14 1686.69 1879.69 Q1690.21 1876.21 1697.87 1876.21 L1704.92 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1717.07 1880.99 L1717.07 1893.87 L1732.43 1893.87 L1732.43 1899.67 L1717.07 1899.67 L1717.07 1924.29 Q1717.07 1929.84 1718.57 1931.42 Q1720.11 1933 1724.77 1933 L1732.43 1933 L1732.43 1939.24 L1724.77 1939.24 Q1716.14 1939.24 1712.86 1936.04 Q1709.58 1932.8 1709.58 1924.29 L1709.58 1899.67 L1704.11 1899.67 L1704.11 1893.87 L1709.58 1893.87 L1709.58 1880.99 L1717.07 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1755.47 1878.76 L1762.36 1878.76 L1741.3 1946.94 L1734.41 1946.94 L1755.47 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1799.1 1895.21 L1799.1 1902.26 Q1795.94 1900.64 1792.54 1899.83 Q1789.14 1899.02 1785.49 1899.02 Q1779.94 1899.02 1777.15 1900.72 Q1774.39 1902.42 1774.39 1905.82 Q1774.39 1908.42 1776.38 1909.91 Q1778.36 1911.37 1784.36 1912.71 L1786.91 1913.28 Q1794.85 1914.98 1798.17 1918.1 Q1801.53 1921.18 1801.53 1926.73 Q1801.53 1933.04 1796.51 1936.73 Q1791.53 1940.42 1782.78 1940.42 Q1779.13 1940.42 1775.16 1939.69 Q1771.23 1939 1766.86 1937.58 L1766.86 1929.89 Q1770.99 1932.03 1775 1933.13 Q1779.01 1934.18 1782.94 1934.18 Q1788.21 1934.18 1791.04 1932.4 Q1793.88 1930.57 1793.88 1927.29 Q1793.88 1924.25 1791.81 1922.63 Q1789.79 1921.01 1782.86 1919.51 L1780.27 1918.91 Q1773.34 1917.45 1770.26 1914.45 Q1767.18 1911.41 1767.18 1906.15 Q1767.18 1899.75 1771.72 1896.26 Q1776.26 1892.78 1784.6 1892.78 Q1788.73 1892.78 1792.38 1893.39 Q1796.02 1893.99 1799.1 1895.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1852.21 1914.69 L1852.21 1918.34 L1817.94 1918.34 Q1818.43 1926.04 1822.56 1930.09 Q1826.73 1934.1 1834.14 1934.1 Q1838.44 1934.1 1842.45 1933.04 Q1846.5 1931.99 1850.47 1929.89 L1850.47 1936.93 Q1846.46 1938.64 1842.25 1939.53 Q1838.03 1940.42 1833.7 1940.42 Q1822.84 1940.42 1816.48 1934.1 Q1810.16 1927.78 1810.16 1917 Q1810.16 1905.86 1816.16 1899.34 Q1822.19 1892.78 1832.4 1892.78 Q1841.56 1892.78 1846.86 1898.69 Q1852.21 1904.57 1852.21 1914.69 M1844.76 1912.51 Q1844.68 1906.39 1841.31 1902.74 Q1837.99 1899.1 1832.48 1899.1 Q1826.24 1899.1 1822.48 1902.62 Q1818.75 1906.15 1818.18 1912.55 L1844.76 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1897.09 1895.61 L1897.09 1902.58 Q1893.93 1900.84 1890.73 1899.99 Q1887.57 1899.1 1884.33 1899.1 Q1877.08 1899.1 1873.07 1903.72 Q1869.06 1908.29 1869.06 1916.6 Q1869.06 1924.9 1873.07 1929.52 Q1877.08 1934.1 1884.33 1934.1 Q1887.57 1934.1 1890.73 1933.25 Q1893.93 1932.36 1897.09 1930.61 L1897.09 1937.5 Q1893.98 1938.96 1890.61 1939.69 Q1887.29 1940.42 1883.52 1940.42 Q1873.27 1940.42 1867.24 1933.98 Q1861.2 1927.54 1861.2 1916.6 Q1861.2 1905.5 1867.28 1899.14 Q1873.4 1892.78 1884.01 1892.78 Q1887.45 1892.78 1890.73 1893.51 Q1894.02 1894.2 1897.09 1895.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip090)" d="M1908.88 1876.29 L1915.36 1876.29 Q1921.44 1885.85 1924.44 1895.01 Q1927.48 1904.16 1927.48 1913.2 Q1927.48 1922.27 1924.44 1931.46 Q1921.44 1940.66 1915.36 1950.18 L1908.88 1950.18 Q1914.27 1940.9 1916.9 1931.75 Q1919.58 1922.55 1919.58 1913.2 Q1919.58 1903.84 1916.9 1894.72 Q1914.27 1885.61 1908.88 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip094)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="261.245,2459.47 266.064,2452.39 270.882,2445.05 275.701,2437.44 280.52,2429.56 285.338,2421.41 290.157,2412.97 294.975,2404.26 299.794,2395.25 304.613,2385.95 309.431,2376.37 314.25,2366.5 319.068,2356.34 323.887,2345.9 328.706,2335.18 333.524,2324.19 338.343,2312.95 343.161,2301.47 347.98,2289.76 352.799,2277.84 357.617,2265.74 362.436,2253.48 367.254,2241.1 372.073,2228.62 376.892,2216.09 381.71,2203.55 386.529,2191.05 391.348,2178.64 396.166,2166.37 400.985,2154.3 405.803,2142.5 410.622,2131.01 415.441,2119.93 420.259,2109.29 425.078,2099.19 429.896,2089.67 434.715,2080.8 439.534,2072.65 444.352,2065.26 449.171,2058.7 453.989,2053 458.808,2048.2 463.627,2044.33 468.445,2041.41 473.264,2039.45 478.083,2038.45 482.901,2038.4 487.72,2039.29 492.538,2041.08 497.357,2043.75 502.176,2047.25 506.994,2051.53 511.813,2056.54 516.631,2062.22 521.45,2068.51 526.269,2075.34 531.087,2082.66 535.906,2090.39 540.724,2098.47 545.543,2106.83 550.362,2115.43 555.18,2124.2 559.999,2133.08 564.818,2142.03 569.636,2150.99 574.455,2159.92 579.273,2168.79 584.092,2177.55 588.911,2186.18 593.729,2194.63 598.548,2202.9 603.366,2210.96 608.185,2218.78 613.004,2226.35 617.822,2233.65 622.641,2240.68 627.459,2247.42 632.278,2253.87 637.097,2260.02 641.915,2265.86 646.734,2271.4 651.553,2276.62 656.371,2281.53 661.19,2286.13 666.008,2290.41 670.827,2294.38 675.646,2298.04 680.464,2301.38 685.283,2304.42 690.101,2307.15 694.92,2309.57 699.739,2311.68 704.557,2313.5 709.376,2315.01 714.194,2316.22 719.013,2317.13 723.832,2317.75 728.65,2318.07 733.469,2318.09 738.288,2317.83 743.106,2317.27 747.925,2316.43 752.743,2315.3 757.562,2313.88 762.381,2312.18 767.199,2310.2 772.018,2307.93 776.836,2305.4 781.655,2302.59 786.474,2299.51 791.292,2296.17 796.111,2292.57 800.929,2288.73 805.748,2284.64 810.567,2280.33 815.385,2275.79 820.204,2271.06 825.023,2266.13 829.841,2261.03 834.66,2255.78 839.478,2250.41 844.297,2244.93 849.116,2239.38 853.934,2233.78 858.753,2228.18 863.571,2222.59 868.39,2217.07 873.209,2211.66 878.027,2206.38 882.846,2201.3 887.664,2196.44 892.483,2191.87 897.302,2187.61 902.12,2183.72 906.939,2180.22 911.758,2177.17 916.576,2174.59 921.395,2172.51 926.213,2170.96 931.032,2169.96 935.851,2169.51 940.669,2169.63 945.488,2170.3 950.306,2171.53 955.125,2173.3 959.944,2175.59 964.762,2178.37 969.581,2181.62 974.399,2185.28 979.218,2189.34 984.037,2193.73 988.855,2198.42 993.674,2203.37 998.492,2208.52 1003.31,2213.84 1008.13,2219.27 1012.95,2224.77 1017.77,2230.31 1022.59,2235.83 1027.4,2241.32 1032.22,2246.72 1037.04,2252.02 1041.86,2257.17 1046.68,2262.17 1051.5,2266.98 1056.32,2271.58 1061.13,2275.95 1065.95,2280.09 1070.77,2283.98 1075.59,2287.6 1080.41,2290.95 1085.23,2294.03 1090.05,2296.82 1094.86,2299.33 1099.68,2301.54 1104.5,2303.47 1109.32,2305.1 1114.14,2306.45 1118.96,2307.51 1123.78,2308.29 1128.59,2308.8 1133.41,2309.03 1138.23,2309.01 1143.05,2308.73 1147.87,2308.2 1152.69,2307.45 1157.51,2306.48 1162.33,2305.3 1167.14,2303.93 1171.96,2302.4 1176.78,2300.72 1181.6,2298.9 1186.42,2296.98 1191.24,2294.98 1196.06,2292.93 1200.87,2290.84 1205.69,2288.76 1210.51,2286.71 1215.33,2284.73 1220.15,2282.83 1224.97,2281.07 1229.79,2279.46 1234.6,2278.04 1239.42,2276.83 1244.24,2275.87 1249.06,2275.17 1253.88,2274.75 1258.7,2274.64 1263.52,2274.85 1268.33,2275.38 1273.15,2276.24 1277.97,2277.43 1282.79,2278.94 1287.61,2280.76 1292.43,2282.88 1297.25,2285.29 1302.06,2287.95 1306.88,2290.85 1311.7,2293.96 1316.52,2297.24 1321.34,2300.68 1326.16,2304.24 1330.98,2307.89 1335.8,2311.59 1340.61,2315.32 1345.43,2319.06 1350.25,2322.76 1355.07,2326.42 1359.89,2330 1364.71,2333.48 1369.53,2336.85 1374.34,2340.08 1379.16,2343.17 1383.98,2346.1 1388.8,2348.85 1393.62,2351.43 1398.44,2353.82 1403.26,2356.02 1408.07,2358.03 1412.89,2359.84 1417.71,2361.46 1422.53,2362.89 1427.35,2364.13 1432.17,2365.19 1436.99,2366.08 1441.8,2366.8 1446.62,2367.36 1451.44,2367.78 1456.26,2368.07 1461.08,2368.24 1465.9,2368.31 1470.72,2368.29 1475.53,2368.2 1480.35,2368.06 1485.17,2367.9 1489.99,2367.72 1494.81,2367.55 1499.63,2367.42 1504.45,2367.33 1509.27,2367.31 1514.08,2367.38 1518.9,2367.56 1523.72,2367.86 1528.54,2368.3 1533.36,2368.89 1538.18,2369.64 1543,2370.55 1547.81,2371.64 1552.63,2372.91 1557.45,2374.34 1562.27,2375.95 1567.09,2377.72 1571.91,2379.65 1576.73,2381.72 1581.54,2383.92 1586.36,2386.23 1591.18,2388.64 1596,2391.14 1600.82,2393.7 1605.64,2396.31 1610.46,2398.94 1615.27,2401.59 1620.09,2404.23 1624.91,2406.85 1629.73,2409.43 1634.55,2411.96 1639.37,2414.43 1644.19,2416.83 1649,2419.14 1653.82,2421.36 1658.64,2423.48 1663.46,2425.49 1668.28,2427.4 1673.1,2429.21 1677.92,2430.9 1682.74,2432.48 1687.55,2433.96 1692.37,2435.34 1697.19,2436.62 1702.01,2437.8 1706.83,2438.9 1711.65,2439.92 1716.47,2440.87 1721.28,2441.76 1726.1,2442.6 1730.92,2443.41 1735.74,2444.18 1740.56,2444.94 1745.38,2445.69 1750.2,2446.44 1755.01,2447.22 1759.83,2448.02 1764.65,2448.86 1769.47,2449.75 1774.29,2450.7 1779.11,2451.7 1783.93,2452.78 1788.74,2453.93 1793.56,2455.16 1798.38,2456.47 1803.2,2457.86 1808.02,2459.33 1812.84,2460.88 1817.66,2462.49 1822.47,2464.18 1827.29,2465.92 1832.11,2467.72 1836.93,2469.56 1841.75,2471.45 1846.57,2473.35 1851.39,2475.28 1856.21,2477.22 1861.02,2479.16 1865.84,2481.09 1870.66,2483 1875.48,2484.89 1880.3,2486.76 1885.12,2488.58 1889.94,2490.37 1894.75,2492.11 1899.57,2493.8 1904.39,2495.44 1909.21,2497.02 1914.03,2498.56 1918.85,2500.03 1923.67,2501.46 1928.48,2502.83 1933.3,2504.15 1938.12,2505.43 1942.94,2506.67 1947.76,2507.86 1952.58,2509.02 1957.4,2510.15 1962.21,2511.26 1967.03,2512.34 1971.85,2513.42 1976.67,2514.48 1981.49,2515.55 1986.31,2516.61 1991.13,2517.68 1995.94,2518.77 2000.76,2519.87 2005.58,2520.99 2010.4,2522.13 2015.22,2523.3 2020.04,2524.5 2024.86,2525.72 2029.68,2526.97 2034.49,2528.25 2039.31,2529.56 2044.13,2530.9 2048.95,2532.25 2053.77,2533.63 2058.59,2535.03 2063.41,2536.44 2068.22,2537.86 2073.04,2539.29 2077.86,2540.72 2082.68,2542.15 2087.5,2543.58 2092.32,2545 2097.14,2546.41 2101.95,2547.8 2106.77,2549.17 2111.59,2550.53 2116.41,2551.86 2121.23,2553.17 2126.05,2554.45 2130.87,2555.71 2135.68,2556.94 2140.5,2558.15 2145.32,2559.33 2150.14,2560.48 2154.96,2561.61 2159.78,2562.72 2164.6,2563.81 2169.41,2564.88 2174.23,2565.93 2179.05,2566.96 2183.87,2567.99 2188.69,2569 2193.51,2570 2198.33,2571 2203.15,2571.99 2207.96,2572.98 2212.78,2573.97 2217.6,2574.96 2222.42,2575.96 2227.24,2576.96 2232.06,2577.96 2236.88,2578.96 2241.69,2579.97 2246.51,2580.99 2251.33,2582.01 2256.15,2583.04 2260.97,2584.06 2265.79,2585.09 2270.61,2586.13 2275.42,2587.16 2280.24,2588.19 2285.06,2589.21 2289.88,2590.23 2294.7,2591.25 2299.52,2592.25 2304.34,2593.25 2309.15,2594.24 2313.97,2595.21 2318.79,2596.17 2323.61,2597.12 2328.43,2598.05 2333.25,2598.97 2338.07,2599.87 2342.88,2600.76 2347.7,2601.63 2352.52,2602.48 2357.34,2603.32 2362.16,2604.14 2366.98,2604.95 2371.8,2605.74 2376.62,2606.52 2381.43,2607.29 2386.25,2608.04 2391.07,2608.79 2395.89,2609.52 2400.71,2610.25 2405.53,2610.97 2410.35,2611.68 2415.16,2612.38 2419.98,2613.09 2424.8,2613.78 2429.62,2614.48 2434.44,2615.17 2439.26,2615.85 2444.08,2616.54 2448.89,2617.22 2453.71,2617.91 2458.53,2618.58 2463.35,2619.26 2468.17,2619.93 2472.99,2620.6 2477.81,2621.27 2482.62,2621.92 2487.44,2622.58 2492.26,2623.22 2497.08,2623.86 2501.9,2624.49 2506.72,2625.11 2511.54,2625.71 2516.35,2626.31 2521.17,2626.89 2525.99,2627.46 2530.81,2628.01 2535.63,2628.55 2540.45,2629.07 2545.27,2629.57 2550.09,2630.06 2554.9,2630.53 2559.72,2630.99 2564.54,2631.43 2569.36,2631.85 2574.18,2632.26 2579,2632.65 2583.82,2633.03 2588.63,2633.39 2593.45,2633.75 2598.27,2634.09 2603.09,2634.41 2607.91,2634.73 2612.73,2635.05 2617.55,2635.35 2622.36,2635.65 2627.18,2635.95 2632,2636.25 2636.82,2636.55 2641.64,2636.85 2646.46,2637.15 2651.28,2637.46 2656.09,2637.78 2660.91,2638.11 2665.73,2638.44 2670.55,2638.79 2675.37,2639.15 2680.19,2637.25 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/07f114d3.svg b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b9835bf8.svg
similarity index 87%
rename from previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/07f114d3.svg
rename to previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b9835bf8.svg
index a5ece47efcf..fbec90a2223 100644
--- a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/07f114d3.svg
+++ b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/b9835bf8.svg
@@ -1,48 +1,48 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="600" height="400" viewBox="0 0 2400 1600">
 <defs>
-  <clipPath id="clip330">
+  <clipPath id="clip150">
     <rect x="0" y="0" width="2400" height="1600"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip330)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip150)" d="M0 1600 L2400 1600 L2400 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip331">
+  <clipPath id="clip151">
     <rect x="480" y="0" width="1681" height="1600"/>
   </clipPath>
 </defs>
 <defs>
-  <clipPath id="clip332">
+  <clipPath id="clip152">
     <rect x="219" y="123" width="2135" height="1301"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip332)" d="M266.588 1408.12 L266.588 1372.25 L275.983 1354.31 L292.255 1364.67 L292.255 1400.54 L282.86 1418.48 L266.588 1408.12  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="791.763,1172.23 1166.96,847.3 1166.96,197.447 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="954.483,1219.2 1329.68,894.274 1329.68,244.42 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1117.2,1266.17 1492.4,941.247 1492.4,291.393 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1279.92,1313.15 1655.12,988.22 1655.12,338.367 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1166.92 1423.22,1354.51 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="791.763,1172.23 796.265,1168.33 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="954.483,1219.2 958.986,1215.3 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1117.2,1266.17 1121.71,1262.27 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1279.92,1313.15 1284.43,1309.25 "/>
-<path clip-path="url(#clip330)" d="M764.355 1182.32 Q760.744 1182.32 758.916 1185.89 Q757.11 1189.43 757.11 1196.56 Q757.11 1203.67 758.916 1207.23 Q760.744 1210.77 764.355 1210.77 Q767.99 1210.77 769.795 1207.23 Q771.624 1203.67 771.624 1196.56 Q771.624 1189.43 769.795 1185.89 Q767.99 1182.32 764.355 1182.32 M764.355 1178.62 Q770.166 1178.62 773.221 1183.23 Q776.3 1187.81 776.3 1196.56 Q776.3 1205.29 773.221 1209.89 Q770.166 1214.48 764.355 1214.48 Q758.545 1214.48 755.467 1209.89 Q752.411 1205.29 752.411 1196.56 Q752.411 1187.81 755.467 1183.23 Q758.545 1178.62 764.355 1178.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M890.942 1256.84 L907.261 1256.84 L907.261 1260.78 L885.317 1260.78 L885.317 1256.84 Q887.979 1254.09 892.562 1249.46 Q897.169 1244.81 898.349 1243.46 Q900.594 1240.94 901.474 1239.2 Q902.377 1237.44 902.377 1235.75 Q902.377 1233 900.432 1231.26 Q898.511 1229.53 895.409 1229.53 Q893.21 1229.53 890.757 1230.29 Q888.326 1231.06 885.548 1232.61 L885.548 1227.88 Q888.372 1226.75 890.826 1226.17 Q893.28 1225.59 895.317 1225.59 Q900.687 1225.59 903.881 1228.28 Q907.076 1230.96 907.076 1235.45 Q907.076 1237.58 906.266 1239.5 Q905.479 1241.4 903.372 1244 Q902.793 1244.67 899.692 1247.88 Q896.59 1251.08 890.942 1256.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M927.076 1229.3 Q923.465 1229.3 921.636 1232.86 Q919.83 1236.4 919.83 1243.53 Q919.83 1250.64 921.636 1254.2 Q923.465 1257.75 927.076 1257.75 Q930.71 1257.75 932.516 1254.2 Q934.344 1250.64 934.344 1243.53 Q934.344 1236.4 932.516 1232.86 Q930.71 1229.3 927.076 1229.3 M927.076 1225.59 Q932.886 1225.59 935.941 1230.2 Q939.02 1234.78 939.02 1243.53 Q939.02 1252.26 935.941 1256.87 Q932.886 1261.45 927.076 1261.45 Q921.266 1261.45 918.187 1256.87 Q915.131 1252.26 915.131 1243.53 Q915.131 1234.78 918.187 1230.2 Q921.266 1225.59 927.076 1225.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1062.48 1277.26 L1050.68 1295.71 L1062.48 1295.71 L1062.48 1277.26 M1061.25 1273.19 L1067.13 1273.19 L1067.13 1295.71 L1072.06 1295.71 L1072.06 1299.6 L1067.13 1299.6 L1067.13 1307.75 L1062.48 1307.75 L1062.48 1299.6 L1046.88 1299.6 L1046.88 1295.09 L1061.25 1273.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1089.8 1276.27 Q1086.18 1276.27 1084.36 1279.83 Q1082.55 1283.38 1082.55 1290.51 Q1082.55 1297.61 1084.36 1301.18 Q1086.18 1304.72 1089.8 1304.72 Q1093.43 1304.72 1095.24 1301.18 Q1097.06 1297.61 1097.06 1290.51 Q1097.06 1283.38 1095.24 1279.83 Q1093.43 1276.27 1089.8 1276.27 M1089.8 1272.57 Q1095.61 1272.57 1098.66 1277.17 Q1101.74 1281.76 1101.74 1290.51 Q1101.74 1299.23 1098.66 1303.84 Q1095.61 1308.42 1089.8 1308.42 Q1083.99 1308.42 1080.91 1303.84 Q1077.85 1299.23 1077.85 1290.51 Q1077.85 1281.76 1080.91 1277.17 Q1083.99 1272.57 1089.8 1272.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1222.93 1335.58 Q1219.79 1335.58 1217.93 1337.73 Q1216.1 1339.89 1216.1 1343.64 Q1216.1 1347.36 1217.93 1349.54 Q1219.79 1351.69 1222.93 1351.69 Q1226.08 1351.69 1227.91 1349.54 Q1229.76 1347.36 1229.76 1343.64 Q1229.76 1339.89 1227.91 1337.73 Q1226.08 1335.58 1222.93 1335.58 M1232.22 1320.93 L1232.22 1325.19 Q1230.46 1324.35 1228.65 1323.91 Q1226.87 1323.47 1225.11 1323.47 Q1220.48 1323.47 1218.03 1326.6 Q1215.6 1329.72 1215.25 1336.04 Q1216.61 1334.03 1218.67 1332.96 Q1220.73 1331.88 1223.21 1331.88 Q1228.42 1331.88 1231.43 1335.05 Q1234.46 1338.2 1234.46 1343.64 Q1234.46 1348.96 1231.31 1352.18 Q1228.16 1355.4 1222.93 1355.4 Q1216.94 1355.4 1213.77 1350.81 Q1210.6 1346.21 1210.6 1337.48 Q1210.6 1329.28 1214.48 1324.42 Q1218.37 1319.54 1224.92 1319.54 Q1226.68 1319.54 1228.47 1319.89 Q1230.27 1320.23 1232.22 1320.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1252.52 1323.24 Q1248.91 1323.24 1247.08 1326.81 Q1245.27 1330.35 1245.27 1337.48 Q1245.27 1344.59 1247.08 1348.15 Q1248.91 1351.69 1252.52 1351.69 Q1256.15 1351.69 1257.96 1348.15 Q1259.78 1344.59 1259.78 1337.48 Q1259.78 1330.35 1257.96 1326.81 Q1256.15 1323.24 1252.52 1323.24 M1252.52 1319.54 Q1258.33 1319.54 1261.38 1324.15 Q1264.46 1328.73 1264.46 1337.48 Q1264.46 1346.21 1261.38 1350.81 Q1258.33 1355.4 1252.52 1355.4 Q1246.71 1355.4 1243.63 1350.81 Q1240.57 1346.21 1240.57 1337.48 Q1240.57 1328.73 1243.63 1324.15 Q1246.71 1319.54 1252.52 1319.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M721.854 1393.34 L728.284 1393.34 L728.284 1435.45 L751.423 1435.45 L751.423 1440.86 L721.854 1440.86 L721.854 1393.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M770.584 1409.32 Q765.873 1409.32 763.136 1413.01 Q760.399 1416.67 760.399 1423.07 Q760.399 1429.47 763.104 1433.16 Q765.842 1436.82 770.584 1436.82 Q775.263 1436.82 778 1433.13 Q780.737 1429.43 780.737 1423.07 Q780.737 1416.73 778 1413.04 Q775.263 1409.32 770.584 1409.32 M770.584 1404.35 Q778.223 1404.35 782.583 1409.32 Q786.944 1414.28 786.944 1423.07 Q786.944 1431.82 782.583 1436.82 Q778.223 1441.78 770.584 1441.78 Q762.913 1441.78 758.553 1436.82 Q754.224 1431.82 754.224 1423.07 Q754.224 1414.28 758.553 1409.32 Q762.913 1404.35 770.584 1404.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M826.284 1419.34 L826.284 1440.86 L820.427 1440.86 L820.427 1419.54 Q820.427 1414.47 818.454 1411.96 Q816.481 1409.45 812.534 1409.45 Q807.792 1409.45 805.054 1412.47 Q802.317 1415.49 802.317 1420.71 L802.317 1440.86 L796.429 1440.86 L796.429 1405.21 L802.317 1405.21 L802.317 1410.75 Q804.418 1407.54 807.25 1405.94 Q810.115 1404.35 813.839 1404.35 Q819.982 1404.35 823.133 1408.17 Q826.284 1411.96 826.284 1419.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M861.423 1422.62 Q861.423 1416.26 858.781 1412.76 Q856.171 1409.25 851.428 1409.25 Q846.718 1409.25 844.076 1412.76 Q841.466 1416.26 841.466 1422.62 Q841.466 1428.96 844.076 1432.46 Q846.718 1435.96 851.428 1435.96 Q856.171 1435.96 858.781 1432.46 Q861.423 1428.96 861.423 1422.62 M867.279 1436.44 Q867.279 1445.54 863.237 1449.96 Q859.195 1454.42 850.856 1454.42 Q847.768 1454.42 845.031 1453.94 Q842.294 1453.5 839.716 1452.54 L839.716 1446.84 Q842.294 1448.24 844.808 1448.91 Q847.323 1449.58 849.932 1449.58 Q855.693 1449.58 858.558 1446.56 Q861.423 1443.57 861.423 1437.49 L861.423 1434.59 Q859.608 1437.74 856.776 1439.3 Q853.943 1440.86 849.996 1440.86 Q843.439 1440.86 839.429 1435.86 Q835.419 1430.87 835.419 1422.62 Q835.419 1414.35 839.429 1409.35 Q843.439 1404.35 849.996 1404.35 Q853.943 1404.35 856.776 1405.91 Q859.608 1407.47 861.423 1410.62 L861.423 1405.21 L867.279 1405.21 L867.279 1436.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M879.342 1405.21 L885.199 1405.21 L885.199 1440.86 L879.342 1440.86 L879.342 1405.21 M879.342 1391.34 L885.199 1391.34 L885.199 1398.75 L879.342 1398.75 L879.342 1391.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M903.245 1395.09 L903.245 1405.21 L915.308 1405.21 L915.308 1409.76 L903.245 1409.76 L903.245 1429.12 Q903.245 1433.48 904.423 1434.72 Q905.632 1435.96 909.293 1435.96 L915.308 1435.96 L915.308 1440.86 L909.293 1440.86 Q902.513 1440.86 899.935 1438.35 Q897.357 1435.8 897.357 1429.12 L897.357 1409.76 L893.06 1409.76 L893.06 1405.21 L897.357 1405.21 L897.357 1395.09 L903.245 1395.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M922.406 1426.79 L922.406 1405.21 L928.263 1405.21 L928.263 1426.57 Q928.263 1431.63 930.236 1434.18 Q932.209 1436.69 936.156 1436.69 Q940.898 1436.69 943.636 1433.67 Q946.405 1430.64 946.405 1425.42 L946.405 1405.21 L952.261 1405.21 L952.261 1440.86 L946.405 1440.86 L946.405 1435.39 Q944.272 1438.63 941.44 1440.22 Q938.639 1441.78 934.915 1441.78 Q928.772 1441.78 925.589 1437.96 Q922.406 1434.14 922.406 1426.79 M937.143 1404.35 L937.143 1404.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M987.782 1410.62 L987.782 1391.34 L993.638 1391.34 L993.638 1440.86 L987.782 1440.86 L987.782 1435.51 Q985.936 1438.7 983.103 1440.26 Q980.302 1441.78 976.355 1441.78 Q969.894 1441.78 965.82 1436.63 Q961.778 1431.47 961.778 1423.07 Q961.778 1414.67 965.82 1409.51 Q969.894 1404.35 976.355 1404.35 Q980.302 1404.35 983.103 1405.91 Q985.936 1407.44 987.782 1410.62 M967.825 1423.07 Q967.825 1429.53 970.467 1433.22 Q973.141 1436.88 977.788 1436.88 Q982.435 1436.88 985.108 1433.22 Q987.782 1429.53 987.782 1423.07 Q987.782 1416.61 985.108 1412.95 Q982.435 1409.25 977.788 1409.25 Q973.141 1409.25 970.467 1412.95 Q967.825 1416.61 967.825 1423.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1036.19 1421.57 L1036.19 1424.44 L1009.27 1424.44 Q1009.65 1430.48 1012.89 1433.67 Q1016.17 1436.82 1022 1436.82 Q1025.37 1436.82 1028.52 1435.99 Q1031.71 1435.16 1034.82 1433.51 L1034.82 1439.05 Q1031.67 1440.38 1028.36 1441.08 Q1025.05 1441.78 1021.65 1441.78 Q1013.12 1441.78 1008.12 1436.82 Q1003.16 1431.85 1003.16 1423.39 Q1003.16 1414.63 1007.87 1409.51 Q1012.61 1404.35 1020.63 1404.35 Q1027.82 1404.35 1031.99 1409 Q1036.19 1413.62 1036.19 1421.57 M1030.34 1419.85 Q1030.27 1415.05 1027.63 1412.18 Q1025.02 1409.32 1020.69 1409.32 Q1015.79 1409.32 1012.83 1412.09 Q1009.9 1414.86 1009.46 1419.89 L1030.34 1419.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1080.59 1391.4 Q1076.33 1398.72 1074.26 1405.88 Q1072.19 1413.04 1072.19 1420.39 Q1072.19 1427.75 1074.26 1434.97 Q1076.36 1442.17 1080.59 1449.45 L1075.5 1449.45 Q1070.73 1441.97 1068.34 1434.75 Q1065.98 1427.52 1065.98 1420.39 Q1065.98 1413.3 1068.34 1406.1 Q1070.7 1398.91 1075.5 1391.4 L1080.59 1391.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1115.41 1410.62 L1115.41 1391.34 L1121.27 1391.34 L1121.27 1440.86 L1115.41 1440.86 L1115.41 1435.51 Q1113.57 1438.7 1110.74 1440.26 Q1107.93 1441.78 1103.99 1441.78 Q1097.53 1441.78 1093.45 1436.63 Q1089.41 1431.47 1089.41 1423.07 Q1089.41 1414.67 1093.45 1409.51 Q1097.53 1404.35 1103.99 1404.35 Q1107.93 1404.35 1110.74 1405.91 Q1113.57 1407.44 1115.41 1410.62 M1095.46 1423.07 Q1095.46 1429.53 1098.1 1433.22 Q1100.77 1436.88 1105.42 1436.88 Q1110.07 1436.88 1112.74 1433.22 Q1115.41 1429.53 1115.41 1423.07 Q1115.41 1416.61 1112.74 1412.95 Q1110.07 1409.25 1105.42 1409.25 Q1100.77 1409.25 1098.1 1412.95 Q1095.46 1416.61 1095.46 1423.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1163.83 1421.57 L1163.83 1424.44 L1136.9 1424.44 Q1137.28 1430.48 1140.53 1433.67 Q1143.81 1436.82 1149.63 1436.82 Q1153 1436.82 1156.15 1435.99 Q1159.34 1435.16 1162.46 1433.51 L1162.46 1439.05 Q1159.31 1440.38 1156 1441.08 Q1152.69 1441.78 1149.28 1441.78 Q1140.75 1441.78 1135.75 1436.82 Q1130.79 1431.85 1130.79 1423.39 Q1130.79 1414.63 1135.5 1409.51 Q1140.24 1404.35 1148.26 1404.35 Q1155.45 1404.35 1159.62 1409 Q1163.83 1413.62 1163.83 1421.57 M1157.97 1419.85 Q1157.91 1415.05 1155.26 1412.18 Q1152.65 1409.32 1148.32 1409.32 Q1143.42 1409.32 1140.46 1412.09 Q1137.54 1414.86 1137.09 1419.89 L1157.97 1419.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1196.9 1422.62 Q1196.9 1416.26 1194.25 1412.76 Q1191.64 1409.25 1186.9 1409.25 Q1182.19 1409.25 1179.55 1412.76 Q1176.94 1416.26 1176.94 1422.62 Q1176.94 1428.96 1179.55 1432.46 Q1182.19 1435.96 1186.9 1435.96 Q1191.64 1435.96 1194.25 1432.46 Q1196.9 1428.96 1196.9 1422.62 M1202.75 1436.44 Q1202.75 1445.54 1198.71 1449.96 Q1194.67 1454.42 1186.33 1454.42 Q1183.24 1454.42 1180.5 1453.94 Q1177.77 1453.5 1175.19 1452.54 L1175.19 1446.84 Q1177.77 1448.24 1180.28 1448.91 Q1182.8 1449.58 1185.41 1449.58 Q1191.17 1449.58 1194.03 1446.56 Q1196.9 1443.57 1196.9 1437.49 L1196.9 1434.59 Q1195.08 1437.74 1192.25 1439.3 Q1189.42 1440.86 1185.47 1440.86 Q1178.91 1440.86 1174.9 1435.86 Q1170.89 1430.87 1170.89 1422.62 Q1170.89 1414.35 1174.9 1409.35 Q1178.91 1404.35 1185.47 1404.35 Q1189.42 1404.35 1192.25 1405.91 Q1195.08 1407.47 1196.9 1410.62 L1196.9 1405.21 L1202.75 1405.21 L1202.75 1436.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1213.89 1391.4 L1218.98 1391.4 Q1223.76 1398.91 1226.11 1406.1 Q1228.5 1413.3 1228.5 1420.39 Q1228.5 1427.52 1226.11 1434.75 Q1223.76 1441.97 1218.98 1449.45 L1213.89 1449.45 Q1218.12 1442.17 1220.19 1434.97 Q1222.29 1427.75 1222.29 1420.39 Q1222.29 1413.04 1220.19 1405.88 Q1218.12 1398.72 1213.89 1391.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1433.84,1345.32 783.99,1157.72 783.99,507.868 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1537.4,1255.63 887.549,1068.04 887.549,418.183 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1640.96,1165.95 991.108,978.352 991.108,328.498 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1744.52,1076.26 1094.67,888.667 1094.67,238.813 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1423.22,1354.51 1798.42,1029.59 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1433.84,1345.32 1426.05,1343.07 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1537.4,1255.63 1529.6,1253.38 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1640.96,1165.95 1633.16,1163.7 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1744.52,1076.26 1736.72,1074.01 "/>
-<path clip-path="url(#clip330)" d="M1471.3 1344.48 Q1467.69 1344.48 1465.86 1348.04 Q1464.05 1351.59 1464.05 1358.72 Q1464.05 1365.82 1465.86 1369.39 Q1467.69 1372.93 1471.3 1372.93 Q1474.93 1372.93 1476.74 1369.39 Q1478.57 1365.82 1478.57 1358.72 Q1478.57 1351.59 1476.74 1348.04 Q1474.93 1344.48 1471.3 1344.48 M1471.3 1340.78 Q1477.11 1340.78 1480.17 1345.38 Q1483.24 1349.97 1483.24 1358.72 Q1483.24 1367.44 1480.17 1372.05 Q1477.11 1376.63 1471.3 1376.63 Q1465.49 1376.63 1462.41 1372.05 Q1459.36 1367.44 1459.36 1358.72 Q1459.36 1349.97 1462.41 1345.38 Q1465.49 1340.78 1471.3 1340.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1563.59 1282.34 L1571.23 1282.34 L1571.23 1255.98 L1562.92 1257.64 L1562.92 1253.38 L1571.18 1251.72 L1575.85 1251.72 L1575.85 1282.34 L1583.49 1282.34 L1583.49 1286.28 L1563.59 1286.28 L1563.59 1282.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1602.94 1254.79 Q1599.33 1254.79 1597.5 1258.36 Q1595.69 1261.9 1595.69 1269.03 Q1595.69 1276.14 1597.5 1279.7 Q1599.33 1283.24 1602.94 1283.24 Q1606.57 1283.24 1608.38 1279.7 Q1610.21 1276.14 1610.21 1269.03 Q1610.21 1261.9 1608.38 1258.36 Q1606.57 1254.79 1602.94 1254.79 M1602.94 1251.09 Q1608.75 1251.09 1611.8 1255.7 Q1614.88 1260.28 1614.88 1269.03 Q1614.88 1277.76 1611.8 1282.36 Q1608.75 1286.95 1602.94 1286.95 Q1597.13 1286.95 1594.05 1282.36 Q1590.99 1277.76 1590.99 1269.03 Q1590.99 1260.28 1594.05 1255.7 Q1597.13 1251.09 1602.94 1251.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1672.1 1192.66 L1688.42 1192.66 L1688.42 1196.59 L1666.47 1196.59 L1666.47 1192.66 Q1669.14 1189.9 1673.72 1185.27 Q1678.33 1180.62 1679.51 1179.28 Q1681.75 1176.75 1682.63 1175.02 Q1683.53 1173.26 1683.53 1171.57 Q1683.53 1168.81 1681.59 1167.08 Q1679.67 1165.34 1676.57 1165.34 Q1674.37 1165.34 1671.91 1166.11 Q1669.48 1166.87 1666.71 1168.42 L1666.71 1163.7 Q1669.53 1162.56 1671.98 1161.98 Q1674.44 1161.41 1676.47 1161.41 Q1681.84 1161.41 1685.04 1164.09 Q1688.23 1166.78 1688.23 1171.27 Q1688.23 1173.4 1687.42 1175.32 Q1686.64 1177.22 1684.53 1179.81 Q1683.95 1180.48 1680.85 1183.7 Q1677.75 1186.89 1672.1 1192.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1708.23 1165.11 Q1704.62 1165.11 1702.79 1168.67 Q1700.99 1172.22 1700.99 1179.35 Q1700.99 1186.45 1702.79 1190.02 Q1704.62 1193.56 1708.23 1193.56 Q1711.87 1193.56 1713.67 1190.02 Q1715.5 1186.45 1715.5 1179.35 Q1715.5 1172.22 1713.67 1168.67 Q1711.87 1165.11 1708.23 1165.11 M1708.23 1161.41 Q1714.04 1161.41 1717.1 1166.01 Q1720.18 1170.6 1720.18 1179.35 Q1720.18 1188.07 1717.1 1192.68 Q1714.04 1197.26 1708.23 1197.26 Q1702.42 1197.26 1699.34 1192.68 Q1696.29 1188.07 1696.29 1179.35 Q1696.29 1170.6 1699.34 1166.01 Q1702.42 1161.41 1708.23 1161.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1785.66 1088.27 Q1789.02 1088.99 1790.89 1091.26 Q1792.79 1093.53 1792.79 1096.86 Q1792.79 1101.98 1789.27 1104.78 Q1785.75 1107.58 1779.27 1107.58 Q1777.09 1107.58 1774.78 1107.14 Q1772.49 1106.72 1770.03 1105.86 L1770.03 1101.35 Q1771.98 1102.48 1774.29 1103.06 Q1776.61 1103.64 1779.13 1103.64 Q1783.53 1103.64 1785.82 1101.91 Q1788.14 1100.17 1788.14 1096.86 Q1788.14 1093.8 1785.98 1092.09 Q1783.85 1090.36 1780.03 1090.36 L1776.01 1090.36 L1776.01 1086.51 L1780.22 1086.51 Q1783.67 1086.51 1785.5 1085.15 Q1787.33 1083.76 1787.33 1081.17 Q1787.33 1078.5 1785.43 1077.09 Q1783.55 1075.66 1780.03 1075.66 Q1778.11 1075.66 1775.91 1076.07 Q1773.71 1076.49 1771.08 1077.37 L1771.08 1073.2 Q1773.74 1072.46 1776.05 1072.09 Q1778.39 1071.72 1780.45 1071.72 Q1785.77 1071.72 1788.88 1074.15 Q1791.98 1076.56 1791.98 1080.68 Q1791.98 1083.55 1790.33 1085.54 Q1788.69 1087.51 1785.66 1088.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1811.65 1075.42 Q1808.04 1075.42 1806.21 1078.99 Q1804.41 1082.53 1804.41 1089.66 Q1804.41 1096.77 1806.21 1100.33 Q1808.04 1103.87 1811.65 1103.87 Q1815.29 1103.87 1817.09 1100.33 Q1818.92 1096.77 1818.92 1089.66 Q1818.92 1082.53 1817.09 1078.99 Q1815.29 1075.42 1811.65 1075.42 M1811.65 1071.72 Q1817.46 1071.72 1820.52 1076.33 Q1823.6 1080.91 1823.6 1089.66 Q1823.6 1098.39 1820.52 1102.99 Q1817.46 1107.58 1811.65 1107.58 Q1805.84 1107.58 1802.76 1102.99 Q1799.71 1098.39 1799.71 1089.66 Q1799.71 1080.91 1802.76 1076.33 Q1805.84 1071.72 1811.65 1071.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1508.58 1324.68 L1515.01 1324.68 L1515.01 1366.78 L1538.15 1366.78 L1538.15 1372.2 L1508.58 1372.2 L1508.58 1324.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1560.84 1354.28 Q1553.74 1354.28 1551.01 1355.9 Q1548.27 1357.52 1548.27 1361.44 Q1548.27 1364.56 1550.31 1366.4 Q1552.38 1368.22 1555.91 1368.22 Q1560.78 1368.22 1563.71 1364.78 Q1566.67 1361.31 1566.67 1355.58 L1566.67 1354.28 L1560.84 1354.28 M1572.52 1351.86 L1572.52 1372.2 L1566.67 1372.2 L1566.67 1366.78 Q1564.66 1370.03 1561.67 1371.59 Q1558.68 1373.12 1554.35 1373.12 Q1548.87 1373.12 1545.63 1370.06 Q1542.41 1366.98 1542.41 1361.82 Q1542.41 1355.8 1546.42 1352.75 Q1550.47 1349.69 1558.45 1349.69 L1566.67 1349.69 L1566.67 1349.12 Q1566.67 1345.08 1563.99 1342.88 Q1561.35 1340.65 1556.55 1340.65 Q1553.49 1340.65 1550.59 1341.39 Q1547.7 1342.12 1545.02 1343.58 L1545.02 1338.17 Q1548.24 1336.93 1551.26 1336.32 Q1554.29 1335.69 1557.15 1335.69 Q1564.88 1335.69 1568.7 1339.7 Q1572.52 1343.71 1572.52 1351.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1590.38 1326.43 L1590.38 1336.55 L1602.44 1336.55 L1602.44 1341.1 L1590.38 1341.1 L1590.38 1360.45 Q1590.38 1364.81 1591.56 1366.05 Q1592.77 1367.29 1596.43 1367.29 L1602.44 1367.29 L1602.44 1372.2 L1596.43 1372.2 Q1589.65 1372.2 1587.07 1369.68 Q1584.49 1367.13 1584.49 1360.45 L1584.49 1341.1 L1580.19 1341.1 L1580.19 1336.55 L1584.49 1336.55 L1584.49 1326.43 L1590.38 1326.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1610.14 1336.55 L1616 1336.55 L1616 1372.2 L1610.14 1372.2 L1610.14 1336.55 M1610.14 1322.67 L1616 1322.67 L1616 1330.09 L1610.14 1330.09 L1610.14 1322.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1634.05 1326.43 L1634.05 1336.55 L1646.11 1336.55 L1646.11 1341.1 L1634.05 1341.1 L1634.05 1360.45 Q1634.05 1364.81 1635.23 1366.05 Q1636.43 1367.29 1640.09 1367.29 L1646.11 1367.29 L1646.11 1372.2 L1640.09 1372.2 Q1633.32 1372.2 1630.74 1369.68 Q1628.16 1367.13 1628.16 1360.45 L1628.16 1341.1 L1623.86 1341.1 L1623.86 1336.55 L1628.16 1336.55 L1628.16 1326.43 L1634.05 1326.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1653.21 1358.13 L1653.21 1336.55 L1659.06 1336.55 L1659.06 1357.9 Q1659.06 1362.96 1661.04 1365.51 Q1663.01 1368.03 1666.96 1368.03 Q1671.7 1368.03 1674.44 1365 Q1677.21 1361.98 1677.21 1356.76 L1677.21 1336.55 L1683.06 1336.55 L1683.06 1372.2 L1677.21 1372.2 L1677.21 1366.72 Q1675.07 1369.97 1672.24 1371.56 Q1669.44 1373.12 1665.72 1373.12 Q1659.57 1373.12 1656.39 1369.3 Q1653.21 1365.48 1653.21 1358.13 M1667.94 1335.69 L1667.94 1335.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1718.58 1341.96 L1718.58 1322.67 L1724.44 1322.67 L1724.44 1372.2 L1718.58 1372.2 L1718.58 1366.85 Q1716.74 1370.03 1713.91 1371.59 Q1711.1 1373.12 1707.16 1373.12 Q1700.7 1373.12 1696.62 1367.96 Q1692.58 1362.81 1692.58 1354.4 Q1692.58 1346 1696.62 1340.84 Q1700.7 1335.69 1707.16 1335.69 Q1711.1 1335.69 1713.91 1337.25 Q1716.74 1338.78 1718.58 1341.96 M1698.63 1354.4 Q1698.63 1360.86 1701.27 1364.56 Q1703.94 1368.22 1708.59 1368.22 Q1713.24 1368.22 1715.91 1364.56 Q1718.58 1360.86 1718.58 1354.4 Q1718.58 1347.94 1715.91 1344.28 Q1713.24 1340.59 1708.59 1340.59 Q1703.94 1340.59 1701.27 1344.28 Q1698.63 1347.94 1698.63 1354.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1767 1352.91 L1767 1355.77 L1740.07 1355.77 Q1740.45 1361.82 1743.7 1365 Q1746.98 1368.15 1752.8 1368.15 Q1756.17 1368.15 1759.32 1367.33 Q1762.51 1366.5 1765.63 1364.84 L1765.63 1370.38 Q1762.48 1371.72 1759.17 1372.42 Q1755.86 1373.12 1752.45 1373.12 Q1743.92 1373.12 1738.92 1368.15 Q1733.96 1363.19 1733.96 1354.72 Q1733.96 1345.97 1738.67 1340.84 Q1743.41 1335.69 1751.43 1335.69 Q1758.62 1335.69 1762.79 1340.33 Q1767 1344.95 1767 1352.91 M1761.14 1351.19 Q1761.08 1346.38 1758.43 1343.52 Q1755.82 1340.65 1751.49 1340.65 Q1746.59 1340.65 1743.63 1343.42 Q1740.7 1346.19 1740.26 1351.22 L1761.14 1351.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1811.4 1322.73 Q1807.13 1330.05 1805.06 1337.22 Q1802.99 1344.38 1802.99 1351.73 Q1802.99 1359.08 1805.06 1366.31 Q1807.16 1373.5 1811.4 1380.79 L1806.3 1380.79 Q1801.53 1373.31 1799.14 1366.08 Q1796.79 1358.86 1796.79 1351.73 Q1796.79 1344.63 1799.14 1337.44 Q1801.5 1330.25 1806.3 1322.73 L1811.4 1322.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1846.22 1341.96 L1846.22 1322.67 L1852.07 1322.67 L1852.07 1372.2 L1846.22 1372.2 L1846.22 1366.85 Q1844.37 1370.03 1841.54 1371.59 Q1838.74 1373.12 1834.79 1373.12 Q1828.33 1373.12 1824.25 1367.96 Q1820.21 1362.81 1820.21 1354.4 Q1820.21 1346 1824.25 1340.84 Q1828.33 1335.69 1834.79 1335.69 Q1838.74 1335.69 1841.54 1337.25 Q1844.37 1338.78 1846.22 1341.96 M1826.26 1354.4 Q1826.26 1360.86 1828.9 1364.56 Q1831.58 1368.22 1836.22 1368.22 Q1840.87 1368.22 1843.54 1364.56 Q1846.22 1360.86 1846.22 1354.4 Q1846.22 1347.94 1843.54 1344.28 Q1840.87 1340.59 1836.22 1340.59 Q1831.58 1340.59 1828.9 1344.28 Q1826.26 1347.94 1826.26 1354.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1894.63 1352.91 L1894.63 1355.77 L1867.7 1355.77 Q1868.08 1361.82 1871.33 1365 Q1874.61 1368.15 1880.43 1368.15 Q1883.81 1368.15 1886.96 1367.33 Q1890.14 1366.5 1893.26 1364.84 L1893.26 1370.38 Q1890.11 1371.72 1886.8 1372.42 Q1883.49 1373.12 1880.08 1373.12 Q1871.55 1373.12 1866.55 1368.15 Q1861.59 1363.19 1861.59 1354.72 Q1861.59 1345.97 1866.3 1340.84 Q1871.04 1335.69 1879.06 1335.69 Q1886.26 1335.69 1890.43 1340.33 Q1894.63 1344.95 1894.63 1352.91 M1888.77 1351.19 Q1888.71 1346.38 1886.07 1343.52 Q1883.46 1340.65 1879.13 1340.65 Q1874.23 1340.65 1871.27 1343.42 Q1868.34 1346.19 1867.89 1351.22 L1888.77 1351.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1927.7 1353.96 Q1927.7 1347.59 1925.06 1344.09 Q1922.45 1340.59 1917.7 1340.59 Q1912.99 1340.59 1910.35 1344.09 Q1907.74 1347.59 1907.74 1353.96 Q1907.74 1360.29 1910.35 1363.79 Q1912.99 1367.29 1917.7 1367.29 Q1922.45 1367.29 1925.06 1363.79 Q1927.7 1360.29 1927.7 1353.96 M1933.55 1367.77 Q1933.55 1376.87 1929.51 1381.3 Q1925.47 1385.75 1917.13 1385.75 Q1914.04 1385.75 1911.31 1385.28 Q1908.57 1384.83 1905.99 1383.88 L1905.99 1378.18 Q1908.57 1379.58 1911.08 1380.25 Q1913.6 1380.92 1916.21 1380.92 Q1921.97 1380.92 1924.83 1377.89 Q1927.7 1374.9 1927.7 1368.82 L1927.7 1365.93 Q1925.88 1369.08 1923.05 1370.64 Q1920.22 1372.2 1916.27 1372.2 Q1909.71 1372.2 1905.7 1367.2 Q1901.69 1362.2 1901.69 1353.96 Q1901.69 1345.68 1905.7 1340.68 Q1909.71 1335.69 1916.27 1335.69 Q1920.22 1335.69 1923.05 1337.25 Q1925.88 1338.81 1927.7 1341.96 L1927.7 1336.55 L1933.55 1336.55 L1933.55 1367.77 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1944.69 1322.73 L1949.79 1322.73 Q1954.56 1330.25 1956.92 1337.44 Q1959.3 1344.63 1959.3 1351.73 Q1959.3 1358.86 1956.92 1366.08 Q1954.56 1373.31 1949.79 1380.79 L1944.69 1380.79 Q1948.93 1373.5 1951 1366.31 Q1953.1 1359.08 1953.1 1351.73 Q1953.1 1344.38 1951 1337.22 Q1948.93 1330.05 1944.69 1322.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,1080.41 1148.56,755.48 1798.42,943.077 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,910.11 1148.56,585.183 1798.42,772.78 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,739.813 1148.56,414.886 1798.42,602.483 "/>
-<polyline clip-path="url(#clip332)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,569.516 1148.56,244.589 1798.42,432.185 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1166.92 773.371,517.064 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1080.41 777.873,1076.51 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,910.11 777.873,906.211 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,739.813 777.873,735.914 "/>
-<polyline clip-path="url(#clip330)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,569.516 777.873,565.617 "/>
-<path clip-path="url(#clip330)" d="M678.205 1093.75 L685.844 1093.75 L685.844 1067.39 L677.534 1069.05 L677.534 1064.79 L685.798 1063.13 L690.474 1063.13 L690.474 1093.75 L698.113 1093.75 L698.113 1097.69 L678.205 1097.69 L678.205 1093.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M707.557 1091.81 L712.441 1091.81 L712.441 1097.69 L707.557 1097.69 L707.557 1091.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M732.626 1066.21 Q729.015 1066.21 727.187 1069.77 Q725.381 1073.31 725.381 1080.44 Q725.381 1087.55 727.187 1091.11 Q729.015 1094.65 732.626 1094.65 Q736.261 1094.65 738.066 1091.11 Q739.895 1087.55 739.895 1080.44 Q739.895 1073.31 738.066 1069.77 Q736.261 1066.21 732.626 1066.21 M732.626 1062.5 Q738.437 1062.5 741.492 1067.11 Q744.571 1071.69 744.571 1080.44 Q744.571 1089.17 741.492 1093.78 Q738.437 1098.36 732.626 1098.36 Q726.816 1098.36 723.738 1093.78 Q720.682 1089.17 720.682 1080.44 Q720.682 1071.69 723.738 1067.11 Q726.816 1062.5 732.626 1062.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M679.201 923.455 L686.84 923.455 L686.84 897.089 L678.53 898.756 L678.53 894.497 L686.793 892.83 L691.469 892.83 L691.469 923.455 L699.108 923.455 L699.108 927.39 L679.201 927.39 L679.201 923.455 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M708.553 921.51 L713.437 921.51 L713.437 927.39 L708.553 927.39 L708.553 921.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M723.668 892.83 L742.025 892.83 L742.025 896.765 L727.951 896.765 L727.951 905.237 Q728.969 904.89 729.988 904.728 Q731.006 904.543 732.025 904.543 Q737.812 904.543 741.191 907.714 Q744.571 910.885 744.571 916.302 Q744.571 921.881 741.099 924.983 Q737.626 928.061 731.307 928.061 Q729.131 928.061 726.863 927.691 Q724.617 927.321 722.21 926.58 L722.21 921.881 Q724.293 923.015 726.515 923.571 Q728.738 924.126 731.214 924.126 Q735.219 924.126 737.557 922.02 Q739.895 919.913 739.895 916.302 Q739.895 912.691 737.557 910.585 Q735.219 908.478 731.214 908.478 Q729.339 908.478 727.464 908.895 Q725.613 909.311 723.668 910.191 L723.668 892.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M681.423 753.158 L697.742 753.158 L697.742 757.093 L675.798 757.093 L675.798 753.158 Q678.46 750.403 683.043 745.773 Q687.65 741.121 688.83 739.778 Q691.076 737.255 691.955 735.519 Q692.858 733.76 692.858 732.07 Q692.858 729.315 690.914 727.579 Q688.992 725.843 685.891 725.843 Q683.692 725.843 681.238 726.607 Q678.807 727.371 676.03 728.922 L676.03 724.2 Q678.854 723.065 681.307 722.487 Q683.761 721.908 685.798 721.908 Q691.168 721.908 694.363 724.593 Q697.557 727.278 697.557 731.769 Q697.557 733.899 696.747 735.82 Q695.96 737.718 693.854 740.311 Q693.275 740.982 690.173 744.199 Q687.071 747.394 681.423 753.158 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M707.557 751.213 L712.441 751.213 L712.441 757.093 L707.557 757.093 L707.557 751.213 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M732.626 725.612 Q729.015 725.612 727.187 729.176 Q725.381 732.718 725.381 739.848 Q725.381 746.954 727.187 750.519 Q729.015 754.06 732.626 754.06 Q736.261 754.06 738.066 750.519 Q739.895 746.954 739.895 739.848 Q739.895 732.718 738.066 729.176 Q736.261 725.612 732.626 725.612 M732.626 721.908 Q738.437 721.908 741.492 726.514 Q744.571 731.098 744.571 739.848 Q744.571 748.574 741.492 753.181 Q738.437 757.764 732.626 757.764 Q726.816 757.764 723.738 753.181 Q720.682 748.574 720.682 739.848 Q720.682 731.098 723.738 726.514 Q726.816 721.908 732.626 721.908 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M682.418 582.861 L698.738 582.861 L698.738 586.796 L676.793 586.796 L676.793 582.861 Q679.455 580.106 684.039 575.476 Q688.645 570.824 689.826 569.481 Q692.071 566.958 692.951 565.222 Q693.854 563.463 693.854 561.773 Q693.854 559.018 691.909 557.282 Q689.988 555.546 686.886 555.546 Q684.687 555.546 682.233 556.31 Q679.803 557.074 677.025 558.625 L677.025 553.902 Q679.849 552.768 682.303 552.189 Q684.756 551.611 686.793 551.611 Q692.164 551.611 695.358 554.296 Q698.553 556.981 698.553 561.472 Q698.553 563.601 697.742 565.523 Q696.955 567.421 694.849 570.013 Q694.27 570.685 691.168 573.902 Q688.067 577.097 682.418 582.861 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M708.553 580.916 L713.437 580.916 L713.437 586.796 L708.553 586.796 L708.553 580.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M723.668 552.236 L742.025 552.236 L742.025 556.171 L727.951 556.171 L727.951 564.643 Q728.969 564.296 729.988 564.134 Q731.006 563.949 732.025 563.949 Q737.812 563.949 741.191 567.12 Q744.571 570.291 744.571 575.708 Q744.571 581.286 741.099 584.388 Q737.626 587.467 731.307 587.467 Q729.131 587.467 726.863 587.097 Q724.617 586.726 722.21 585.986 L722.21 581.286 Q724.293 582.421 726.515 582.976 Q728.738 583.532 731.214 583.532 Q735.219 583.532 737.557 581.425 Q739.895 579.319 739.895 575.708 Q739.895 572.097 737.557 569.99 Q735.219 567.884 731.214 567.884 Q729.339 567.884 727.464 568.3 Q725.613 568.717 723.668 569.597 L723.668 552.236 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M577.156 1149.14 L600.805 1157.86 L600.805 1140.38 L577.156 1149.14 M570.822 1152.76 L570.822 1145.48 L618.342 1127.37 L618.342 1134.05 L606.152 1138.38 L606.152 1159.8 L618.342 1164.13 L618.342 1170.91 L570.822 1152.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M568.817 1120.68 L568.817 1114.83 L618.342 1114.83 L618.342 1120.68 L568.817 1120.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M572.573 1096.78 L582.694 1096.78 L582.694 1084.72 L587.246 1084.72 L587.246 1096.78 L606.597 1096.78 Q610.958 1096.78 612.199 1095.6 Q613.44 1094.39 613.44 1090.73 L613.44 1084.72 L618.342 1084.72 L618.342 1090.73 Q618.342 1097.51 615.828 1100.09 Q613.281 1102.67 606.597 1102.67 L587.246 1102.67 L587.246 1106.96 L582.694 1106.96 L582.694 1102.67 L572.573 1102.67 L572.573 1096.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M582.694 1077.01 L582.694 1071.16 L618.342 1071.16 L618.342 1077.01 L582.694 1077.01 M568.817 1077.01 L568.817 1071.16 L576.233 1071.16 L576.233 1077.01 L568.817 1077.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M572.573 1053.11 L582.694 1053.11 L582.694 1041.05 L587.246 1041.05 L587.246 1053.11 L606.597 1053.11 Q610.958 1053.11 612.199 1051.93 Q613.44 1050.72 613.44 1047.06 L613.44 1041.05 L618.342 1041.05 L618.342 1047.06 Q618.342 1053.84 615.828 1056.42 Q613.281 1059 606.597 1059 L587.246 1059 L587.246 1063.29 L582.694 1063.29 L582.694 1059 L572.573 1059 L572.573 1053.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M604.274 1033.95 L582.694 1033.95 L582.694 1028.09 L604.051 1028.09 Q609.112 1028.09 611.658 1026.12 Q614.173 1024.15 614.173 1020.2 Q614.173 1015.46 611.149 1012.72 Q608.125 1009.95 602.905 1009.95 L582.694 1009.95 L582.694 1004.09 L618.342 1004.09 L618.342 1009.95 L612.868 1009.95 Q616.114 1012.08 617.705 1014.92 Q619.265 1017.72 619.265 1021.44 Q619.265 1027.58 615.446 1030.77 Q611.626 1033.95 604.274 1033.95 M581.835 1019.21 L581.835 1019.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M588.105 968.573 L568.817 968.573 L568.817 962.717 L618.342 962.717 L618.342 968.573 L612.995 968.573 Q616.178 970.419 617.737 973.252 Q619.265 976.053 619.265 980 Q619.265 986.461 614.109 990.535 Q608.953 994.577 600.55 994.577 Q592.147 994.577 586.991 990.535 Q581.835 986.461 581.835 980 Q581.835 976.053 583.394 973.252 Q584.922 970.419 588.105 968.573 M600.55 988.53 Q607.011 988.53 610.703 985.888 Q614.363 983.214 614.363 978.567 Q614.363 973.92 610.703 971.247 Q607.011 968.573 600.55 968.573 Q594.089 968.573 590.428 971.247 Q586.736 973.92 586.736 978.567 Q586.736 983.214 590.428 985.888 Q594.089 988.53 600.55 988.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M599.054 920.162 L601.919 920.162 L601.919 947.089 Q607.966 946.707 611.149 943.46 Q614.3 940.182 614.3 934.358 Q614.3 930.984 613.472 927.833 Q612.645 924.65 610.99 921.531 L616.528 921.531 Q617.865 924.682 618.565 927.992 Q619.265 931.302 619.265 934.708 Q619.265 943.238 614.3 948.235 Q609.335 953.2 600.868 953.2 Q592.115 953.2 586.991 948.489 Q581.835 943.747 581.835 935.726 Q581.835 928.533 586.482 924.363 Q591.097 920.162 599.054 920.162 M597.335 926.018 Q592.529 926.082 589.665 928.724 Q586.8 931.334 586.8 935.662 Q586.8 940.564 589.569 943.524 Q592.338 946.452 597.367 946.898 L597.335 926.018 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M568.881 875.761 Q576.201 880.026 583.363 882.095 Q590.524 884.164 597.876 884.164 Q605.229 884.164 612.454 882.095 Q619.647 879.994 626.936 875.761 L626.936 880.854 Q619.456 885.628 612.231 888.015 Q605.006 890.37 597.876 890.37 Q590.779 890.37 583.585 888.015 Q576.392 885.66 568.881 880.854 L568.881 875.761 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M612.931 862.457 L612.931 851.953 L576.679 851.953 L578.97 863.38 L573.114 863.38 L570.822 852.017 L570.822 845.588 L612.931 845.588 L612.931 835.084 L618.342 835.084 L618.342 862.457 L612.931 862.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M575.055 808.348 Q575.055 813.314 579.957 815.828 Q584.827 818.311 594.63 818.311 Q604.401 818.311 609.303 815.828 Q614.173 813.314 614.173 808.348 Q614.173 803.351 609.303 800.869 Q604.401 798.354 594.63 798.354 Q584.827 798.354 579.957 800.869 Q575.055 803.351 575.055 808.348 M569.963 808.348 Q569.963 800.359 576.297 796.158 Q582.599 791.925 594.63 791.925 Q606.629 791.925 612.963 796.158 Q619.265 800.359 619.265 808.348 Q619.265 816.337 612.963 820.571 Q606.629 824.772 594.63 824.772 Q582.599 824.772 576.297 820.571 Q569.963 816.337 569.963 808.348 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M575.055 766.876 Q575.055 771.841 579.957 774.355 Q584.827 776.838 594.63 776.838 Q604.401 776.838 609.303 774.355 Q614.173 771.841 614.173 766.876 Q614.173 761.879 609.303 759.396 Q604.401 756.882 594.63 756.882 Q584.827 756.882 579.957 759.396 Q575.055 761.879 575.055 766.876 M569.963 766.876 Q569.963 758.887 576.297 754.685 Q582.599 750.452 594.63 750.452 Q606.629 750.452 612.963 754.685 Q619.265 758.887 619.265 766.876 Q619.265 774.865 612.963 779.098 Q606.629 783.299 594.63 783.299 Q582.599 783.299 576.297 779.098 Q569.963 774.865 569.963 766.876 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M610.258 738.485 L610.258 731.769 L615.732 731.769 L625.917 736.989 L625.917 741.095 L615.732 738.485 L610.258 738.485 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M575.055 704.683 Q575.055 709.648 579.957 712.163 Q584.827 714.645 594.63 714.645 Q604.401 714.645 609.303 712.163 Q614.173 709.648 614.173 704.683 Q614.173 699.686 609.303 697.203 Q604.401 694.689 594.63 694.689 Q584.827 694.689 579.957 697.203 Q575.055 699.686 575.055 704.683 M569.963 704.683 Q569.963 696.694 576.297 692.493 Q582.599 688.259 594.63 688.259 Q606.629 688.259 612.963 692.493 Q619.265 696.694 619.265 704.683 Q619.265 712.672 612.963 716.905 Q606.629 721.106 594.63 721.106 Q582.599 721.106 576.297 716.905 Q569.963 712.672 569.963 704.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M575.055 663.21 Q575.055 668.176 579.957 670.69 Q584.827 673.173 594.63 673.173 Q604.401 673.173 609.303 670.69 Q614.173 668.176 614.173 663.21 Q614.173 658.213 609.303 655.731 Q604.401 653.216 594.63 653.216 Q584.827 653.216 579.957 655.731 Q575.055 658.213 575.055 663.21 M569.963 663.21 Q569.963 655.221 576.297 651.02 Q582.599 646.787 594.63 646.787 Q606.629 646.787 612.963 651.02 Q619.265 655.221 619.265 663.21 Q619.265 671.199 612.963 675.432 Q606.629 679.634 594.63 679.634 Q582.599 679.634 576.297 675.432 Q569.963 671.199 569.963 663.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M575.055 621.738 Q575.055 626.703 579.957 629.217 Q584.827 631.7 594.63 631.7 Q604.401 631.7 609.303 629.217 Q614.173 626.703 614.173 621.738 Q614.173 616.741 609.303 614.258 Q604.401 611.744 594.63 611.744 Q584.827 611.744 579.957 614.258 Q575.055 616.741 575.055 621.738 M569.963 621.738 Q569.963 613.749 576.297 609.547 Q582.599 605.314 594.63 605.314 Q606.629 605.314 612.963 609.547 Q619.265 613.749 619.265 621.738 Q619.265 629.727 612.963 633.96 Q606.629 638.161 594.63 638.161 Q582.599 638.161 576.297 633.96 Q569.963 629.727 569.963 621.738 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M568.817 556.075 L573.687 556.075 L573.687 561.677 Q573.687 564.828 574.96 566.07 Q576.233 567.279 579.543 567.279 L582.694 567.279 L582.694 557.635 L587.246 557.635 L587.246 567.279 L618.342 567.279 L618.342 573.167 L587.246 573.167 L587.246 578.769 L582.694 578.769 L582.694 573.167 L580.211 573.167 Q574.26 573.167 571.554 570.398 Q568.817 567.629 568.817 561.614 L568.817 556.075 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M572.573 546.527 L582.694 546.527 L582.694 534.464 L587.246 534.464 L587.246 546.527 L606.597 546.527 Q610.958 546.527 612.199 545.349 Q613.44 544.14 613.44 540.479 L613.44 534.464 L618.342 534.464 L618.342 540.479 Q618.342 547.259 615.828 549.837 Q613.281 552.415 606.597 552.415 L587.246 552.415 L587.246 556.712 L582.694 556.712 L582.694 552.415 L572.573 552.415 L572.573 546.527 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M568.881 527.684 L568.881 522.592 Q576.392 517.818 583.585 515.462 Q590.779 513.075 597.876 513.075 Q605.006 513.075 612.231 515.462 Q619.456 517.818 626.936 522.592 L626.936 527.684 Q619.647 523.451 612.454 521.382 Q605.229 519.282 597.876 519.282 Q590.524 519.282 583.363 521.382 Q576.201 523.451 568.881 527.684 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M647.371 14.0809 L647.371 22.0612 Q642.712 19.8332 638.58 18.7395 Q634.448 17.6457 630.6 17.6457 Q623.916 17.6457 620.27 20.2383 Q616.665 22.8309 616.665 27.611 Q616.665 31.6214 619.055 33.6873 Q621.485 35.7128 628.21 36.9686 L633.152 37.9813 Q642.307 39.7232 646.641 44.1387 Q651.016 48.5136 651.016 55.8863 Q651.016 64.6767 645.102 69.2137 Q639.228 73.7508 627.845 73.7508 Q623.551 73.7508 618.69 72.7785 Q613.87 71.8063 608.685 69.9024 L608.685 61.4765 Q613.667 64.2716 618.447 65.6895 Q623.227 67.1073 627.845 67.1073 Q634.853 67.1073 638.661 64.3527 Q642.469 61.598 642.469 56.4939 Q642.469 52.0379 639.714 49.5264 Q637 47.0148 630.762 45.759 L625.779 44.7868 Q616.624 42.9639 612.533 39.075 Q608.441 35.1862 608.441 28.2591 Q608.441 20.2383 614.072 15.6203 Q619.743 11.0023 629.668 11.0023 Q633.922 11.0023 638.337 11.7719 Q642.753 12.5416 647.371 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M670.663 65.7705 L670.663 89.8329 L663.169 89.8329 L663.169 27.2059 L670.663 27.2059 L670.663 34.0924 Q673.013 30.0415 676.578 28.0971 Q680.183 26.1121 685.166 26.1121 Q693.429 26.1121 698.574 32.6746 Q703.759 39.2371 703.759 49.9314 Q703.759 60.6258 698.574 67.1883 Q693.429 73.7508 685.166 73.7508 Q680.183 73.7508 676.578 71.8063 Q673.013 69.8214 670.663 65.7705 M696.022 49.9314 Q696.022 41.7081 692.619 37.0496 Q689.257 32.3505 683.343 32.3505 Q677.428 32.3505 674.026 37.0496 Q670.663 41.7081 670.663 49.9314 Q670.663 58.1548 674.026 62.8538 Q677.428 67.5124 683.343 67.5124 Q689.257 67.5124 692.619 62.8538 Q696.022 58.1548 696.022 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M736.734 49.7694 Q727.7 49.7694 724.216 51.8354 Q720.733 53.9013 720.733 58.8839 Q720.733 62.8538 723.325 65.2034 Q725.958 67.5124 730.455 67.5124 Q736.653 67.5124 740.379 63.1374 Q744.147 58.7219 744.147 51.4303 L744.147 49.7694 L736.734 49.7694 M751.6 46.6907 L751.6 72.576 L744.147 72.576 L744.147 65.6895 Q741.595 69.8214 737.787 71.8063 Q733.979 73.7508 728.47 73.7508 Q721.502 73.7508 717.37 69.8619 Q713.279 65.9325 713.279 59.3701 Q713.279 51.7138 718.383 47.825 Q723.528 43.9361 733.695 43.9361 L744.147 43.9361 L744.147 43.2069 Q744.147 38.0623 740.744 35.2672 Q737.382 32.4315 731.265 32.4315 Q727.376 32.4315 723.69 33.3632 Q720.003 34.295 716.601 36.1584 L716.601 29.2718 Q720.692 27.692 724.54 26.9223 Q728.389 26.1121 732.035 26.1121 Q741.878 26.1121 746.739 31.2163 Q751.6 36.3204 751.6 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M799.604 28.9478 L799.604 35.9153 Q796.444 34.1734 793.244 33.3227 Q790.084 32.4315 786.843 32.4315 Q779.592 32.4315 775.582 37.0496 Q771.571 41.6271 771.571 49.9314 Q771.571 58.2358 775.582 62.8538 Q779.592 67.4314 786.843 67.4314 Q790.084 67.4314 793.244 66.5807 Q796.444 65.6895 799.604 63.9476 L799.604 70.8341 Q796.484 72.2924 793.122 73.0216 Q789.8 73.7508 786.033 73.7508 Q775.784 73.7508 769.748 67.3098 Q763.713 60.8689 763.713 49.9314 Q763.713 38.832 769.789 32.472 Q775.906 26.1121 786.519 26.1121 Q789.962 26.1121 793.244 26.8413 Q796.525 27.5299 799.604 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M851.374 48.0275 L851.374 51.6733 L817.104 51.6733 Q817.59 59.3701 821.722 63.421 Q825.894 67.4314 833.307 67.4314 Q837.601 67.4314 841.612 66.3781 Q845.662 65.3249 849.632 63.2184 L849.632 70.267 Q845.622 71.9684 841.409 72.8596 Q837.196 73.7508 832.862 73.7508 Q822.005 73.7508 815.645 67.4314 Q809.326 61.1119 809.326 50.3365 Q809.326 39.1965 815.321 32.6746 Q821.357 26.1121 831.565 26.1121 Q840.72 26.1121 846.027 32.0264 Q851.374 37.9003 851.374 48.0275 M843.921 45.84 Q843.839 39.7232 840.477 36.0774 Q837.156 32.4315 831.646 32.4315 Q825.408 32.4315 821.641 35.9558 Q817.914 39.4801 817.347 45.8805 L843.921 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M926.559 14.0809 L926.559 22.0612 Q921.9 19.8332 917.769 18.7395 Q913.637 17.6457 909.788 17.6457 Q903.104 17.6457 899.458 20.2383 Q895.853 22.8309 895.853 27.611 Q895.853 31.6214 898.243 33.6873 Q900.674 35.7128 907.398 36.9686 L912.34 37.9813 Q921.495 39.7232 925.83 44.1387 Q930.205 48.5136 930.205 55.8863 Q930.205 64.6767 924.29 69.2137 Q918.417 73.7508 907.034 73.7508 Q902.74 73.7508 897.879 72.7785 Q893.058 71.8063 887.873 69.9024 L887.873 61.4765 Q892.855 64.2716 897.636 65.6895 Q902.416 67.1073 907.034 67.1073 Q914.042 67.1073 917.85 64.3527 Q921.657 61.598 921.657 56.4939 Q921.657 52.0379 918.903 49.5264 Q916.189 47.0148 909.95 45.759 L904.968 44.7868 Q895.813 42.9639 891.721 39.075 Q887.63 35.1862 887.63 28.2591 Q887.63 20.2383 893.261 15.6203 Q898.932 11.0023 908.857 11.0023 Q913.11 11.0023 917.525 11.7719 Q921.941 12.5416 926.559 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M980.355 45.1919 L980.355 72.576 L972.901 72.576 L972.901 45.4349 Q972.901 38.994 970.39 35.7938 Q967.878 32.5936 962.855 32.5936 Q956.819 32.5936 953.335 36.4419 Q949.852 40.2903 949.852 46.9338 L949.852 72.576 L942.357 72.576 L942.357 9.54393 L949.852 9.54393 L949.852 34.2544 Q952.525 30.163 956.131 28.1376 Q959.776 26.1121 964.516 26.1121 Q972.334 26.1121 976.345 30.9732 Q980.355 35.7938 980.355 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M994.452 54.671 L994.452 27.2059 L1001.91 27.2059 L1001.91 54.3874 Q1001.91 60.8284 1004.42 64.0691 Q1006.93 67.2693 1011.95 67.2693 Q1017.99 67.2693 1021.47 63.421 Q1025 59.5726 1025 52.9291 L1025 27.2059 L1032.45 27.2059 L1032.45 72.576 L1025 72.576 L1025 65.6084 Q1022.28 69.7404 1018.68 71.7658 Q1015.11 73.7508 1010.37 73.7508 Q1002.55 73.7508 998.503 68.8897 Q994.452 64.0286 994.452 54.671 M1013.21 26.1121 L1013.21 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1055.18 14.324 L1055.18 27.2059 L1070.53 27.2059 L1070.53 32.9987 L1055.18 32.9987 L1055.18 57.6282 Q1055.18 63.1779 1056.67 64.7578 Q1058.21 66.3376 1062.87 66.3376 L1070.53 66.3376 L1070.53 72.576 L1062.87 72.576 Q1054.24 72.576 1050.96 69.3758 Q1047.68 66.1351 1047.68 57.6282 L1047.68 32.9987 L1042.21 32.9987 L1042.21 27.2059 L1047.68 27.2059 L1047.68 14.324 L1055.18 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1087.7 14.324 L1087.7 27.2059 L1103.06 27.2059 L1103.06 32.9987 L1087.7 32.9987 L1087.7 57.6282 Q1087.7 63.1779 1089.2 64.7578 Q1090.74 66.3376 1095.4 66.3376 L1103.06 66.3376 L1103.06 72.576 L1095.4 72.576 Q1086.77 72.576 1083.49 69.3758 Q1080.21 66.1351 1080.21 57.6282 L1080.21 32.9987 L1074.74 32.9987 L1074.74 27.2059 L1080.21 27.2059 L1080.21 14.324 L1087.7 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1112.86 9.54393 L1120.31 9.54393 L1120.31 72.576 L1112.86 72.576 L1112.86 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1174.72 48.0275 L1174.72 51.6733 L1140.45 51.6733 Q1140.93 59.3701 1145.06 63.421 Q1149.24 67.4314 1156.65 67.4314 Q1160.94 67.4314 1164.95 66.3781 Q1169.01 65.3249 1172.98 63.2184 L1172.98 70.267 Q1168.97 71.9684 1164.75 72.8596 Q1160.54 73.7508 1156.2 73.7508 Q1145.35 73.7508 1138.99 67.4314 Q1132.67 61.1119 1132.67 50.3365 Q1132.67 39.1965 1138.66 32.6746 Q1144.7 26.1121 1154.91 26.1121 Q1164.06 26.1121 1169.37 32.0264 Q1174.72 37.9003 1174.72 48.0275 M1167.26 45.84 Q1167.18 39.7232 1163.82 36.0774 Q1160.5 32.4315 1154.99 32.4315 Q1148.75 32.4315 1144.98 35.9558 Q1141.26 39.4801 1140.69 45.8805 L1167.26 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1242.33 44.2197 Q1244.96 45.1109 1247.43 48.0275 Q1249.94 50.9442 1252.45 56.0483 L1260.76 72.576 L1251.97 72.576 L1244.23 57.061 Q1241.23 50.9847 1238.4 48.9997 Q1235.6 47.0148 1230.74 47.0148 L1221.83 47.0148 L1221.83 72.576 L1213.65 72.576 L1213.65 12.096 L1232.12 12.096 Q1242.49 12.096 1247.59 16.4305 Q1252.7 20.7649 1252.7 29.5149 Q1252.7 35.2267 1250.02 38.994 Q1247.39 42.7613 1242.33 44.2197 M1221.83 18.8205 L1221.83 40.2903 L1232.12 40.2903 Q1238.03 40.2903 1241.03 37.5762 Q1244.07 34.8216 1244.07 29.5149 Q1244.07 24.2082 1241.03 21.5346 Q1238.03 18.8205 1232.12 18.8205 L1221.83 18.8205 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1306.05 48.0275 L1306.05 51.6733 L1271.78 51.6733 Q1272.26 59.3701 1276.4 63.421 Q1280.57 67.4314 1287.98 67.4314 Q1292.27 67.4314 1296.28 66.3781 Q1300.34 65.3249 1304.31 63.2184 L1304.31 70.267 Q1300.3 71.9684 1296.08 72.8596 Q1291.87 73.7508 1287.54 73.7508 Q1276.68 73.7508 1270.32 67.4314 Q1264 61.1119 1264 50.3365 Q1264 39.1965 1269.99 32.6746 Q1276.03 26.1121 1286.24 26.1121 Q1295.39 26.1121 1300.7 32.0264 Q1306.05 37.9003 1306.05 48.0275 M1298.59 45.84 Q1298.51 39.7232 1295.15 36.0774 Q1291.83 32.4315 1286.32 32.4315 Q1280.08 32.4315 1276.31 35.9558 Q1272.59 39.4801 1272.02 45.8805 L1298.59 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1357.09 48.0275 L1357.09 51.6733 L1322.82 51.6733 Q1323.3 59.3701 1327.44 63.421 Q1331.61 67.4314 1339.02 67.4314 Q1343.32 67.4314 1347.33 66.3781 Q1351.38 65.3249 1355.35 63.2184 L1355.35 70.267 Q1351.34 71.9684 1347.12 72.8596 Q1342.91 73.7508 1338.58 73.7508 Q1327.72 73.7508 1321.36 67.4314 Q1315.04 61.1119 1315.04 50.3365 Q1315.04 39.1965 1321.04 32.6746 Q1327.07 26.1121 1337.28 26.1121 Q1346.44 26.1121 1351.74 32.0264 Q1357.09 37.9003 1357.09 48.0275 M1349.64 45.84 Q1349.55 39.7232 1346.19 36.0774 Q1342.87 32.4315 1337.36 32.4315 Q1331.12 32.4315 1327.36 35.9558 Q1323.63 39.4801 1323.06 45.8805 L1349.64 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1407.04 45.1919 L1407.04 72.576 L1399.58 72.576 L1399.58 45.4349 Q1399.58 38.994 1397.07 35.7938 Q1394.56 32.5936 1389.54 32.5936 Q1383.5 32.5936 1380.02 36.4419 Q1376.53 40.2903 1376.53 46.9338 L1376.53 72.576 L1369.04 72.576 L1369.04 27.2059 L1376.53 27.2059 L1376.53 34.2544 Q1379.21 30.163 1382.81 28.1376 Q1386.46 26.1121 1391.2 26.1121 Q1399.02 26.1121 1403.03 30.9732 Q1407.04 35.7938 1407.04 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1429.28 14.324 L1429.28 27.2059 L1444.63 27.2059 L1444.63 32.9987 L1429.28 32.9987 L1429.28 57.6282 Q1429.28 63.1779 1430.78 64.7578 Q1432.31 66.3376 1436.97 66.3376 L1444.63 66.3376 L1444.63 72.576 L1436.97 72.576 Q1428.34 72.576 1425.06 69.3758 Q1421.78 66.1351 1421.78 57.6282 L1421.78 32.9987 L1416.31 32.9987 L1416.31 27.2059 L1421.78 27.2059 L1421.78 14.324 L1429.28 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1480.72 34.1734 Q1479.47 33.4443 1477.97 33.1202 Q1476.51 32.7556 1474.73 32.7556 Q1468.41 32.7556 1465.01 36.8875 Q1461.64 40.9789 1461.64 48.6757 L1461.64 72.576 L1454.15 72.576 L1454.15 27.2059 L1461.64 27.2059 L1461.64 34.2544 Q1463.99 30.1225 1467.76 28.1376 Q1471.53 26.1121 1476.91 26.1121 Q1477.68 26.1121 1478.62 26.2337 Q1479.55 26.3147 1480.68 26.5172 L1480.72 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1507.42 76.7889 Q1504.26 84.8907 1501.26 87.3618 Q1498.26 89.8329 1493.24 89.8329 L1487.29 89.8329 L1487.29 83.5945 L1491.66 83.5945 Q1494.74 83.5945 1496.44 82.1361 Q1498.14 80.6778 1500.21 75.2496 L1501.54 71.8468 L1483.19 27.2059 L1491.09 27.2059 L1505.27 62.6918 L1519.45 27.2059 L1527.35 27.2059 L1507.42 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1555.95 12.096 L1607.11 12.096 L1607.11 18.9825 L1585.64 18.9825 L1585.64 72.576 L1577.42 72.576 L1577.42 18.9825 L1555.95 18.9825 L1555.95 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1628.78 34.1734 Q1627.53 33.4443 1626.03 33.1202 Q1624.57 32.7556 1622.79 32.7556 Q1616.47 32.7556 1613.07 36.8875 Q1609.7 40.9789 1609.7 48.6757 L1609.7 72.576 L1602.21 72.576 L1602.21 27.2059 L1609.7 27.2059 L1609.7 34.2544 Q1612.05 30.1225 1615.82 28.1376 Q1619.59 26.1121 1624.98 26.1121 Q1625.75 26.1121 1626.68 26.2337 Q1627.61 26.3147 1628.74 26.5172 L1628.78 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1657.22 49.7694 Q1648.19 49.7694 1644.7 51.8354 Q1641.22 53.9013 1641.22 58.8839 Q1641.22 62.8538 1643.81 65.2034 Q1646.45 67.5124 1650.94 67.5124 Q1657.14 67.5124 1660.87 63.1374 Q1664.63 58.7219 1664.63 51.4303 L1664.63 49.7694 L1657.22 49.7694 M1672.09 46.6907 L1672.09 72.576 L1664.63 72.576 L1664.63 65.6895 Q1662.08 69.8214 1658.27 71.8063 Q1654.47 73.7508 1648.96 73.7508 Q1641.99 73.7508 1637.86 69.8619 Q1633.77 65.9325 1633.77 59.3701 Q1633.77 51.7138 1638.87 47.825 Q1644.01 43.9361 1654.18 43.9361 L1664.63 43.9361 L1664.63 43.2069 Q1664.63 38.0623 1661.23 35.2672 Q1657.87 32.4315 1651.75 32.4315 Q1647.86 32.4315 1644.18 33.3632 Q1640.49 34.295 1637.09 36.1584 L1637.09 29.2718 Q1641.18 27.692 1645.03 26.9223 Q1648.88 26.1121 1652.52 26.1121 Q1662.37 26.1121 1667.23 31.2163 Q1672.09 36.3204 1672.09 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1687.44 27.2059 L1694.89 27.2059 L1694.89 73.3862 Q1694.89 82.0551 1691.57 85.944 Q1688.29 89.8329 1680.96 89.8329 L1678.12 89.8329 L1678.12 83.5134 L1680.11 83.5134 Q1684.36 83.5134 1685.9 81.5285 Q1687.44 79.5841 1687.44 73.3862 L1687.44 27.2059 M1687.44 9.54393 L1694.89 9.54393 L1694.89 18.9825 L1687.44 18.9825 L1687.44 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1749.3 48.0275 L1749.3 51.6733 L1715.03 51.6733 Q1715.51 59.3701 1719.64 63.421 Q1723.82 67.4314 1731.23 67.4314 Q1735.52 67.4314 1739.53 66.3781 Q1743.59 65.3249 1747.56 63.2184 L1747.56 70.267 Q1743.55 71.9684 1739.33 72.8596 Q1735.12 73.7508 1730.78 73.7508 Q1719.93 73.7508 1713.57 67.4314 Q1707.25 61.1119 1707.25 50.3365 Q1707.25 39.1965 1713.24 32.6746 Q1719.28 26.1121 1729.49 26.1121 Q1738.64 26.1121 1743.95 32.0264 Q1749.3 37.9003 1749.3 48.0275 M1741.84 45.84 Q1741.76 39.7232 1738.4 36.0774 Q1735.08 32.4315 1729.57 32.4315 Q1723.33 32.4315 1719.56 35.9558 Q1715.84 39.4801 1715.27 45.8805 L1741.84 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1794.18 28.9478 L1794.18 35.9153 Q1791.02 34.1734 1787.82 33.3227 Q1784.66 32.4315 1781.42 32.4315 Q1774.17 32.4315 1770.16 37.0496 Q1766.15 41.6271 1766.15 49.9314 Q1766.15 58.2358 1770.16 62.8538 Q1774.17 67.4314 1781.42 67.4314 Q1784.66 67.4314 1787.82 66.5807 Q1791.02 65.6895 1794.18 63.9476 L1794.18 70.8341 Q1791.06 72.2924 1787.7 73.0216 Q1784.38 73.7508 1780.61 73.7508 Q1770.36 73.7508 1764.33 67.3098 Q1758.29 60.8689 1758.29 49.9314 Q1758.29 38.832 1764.37 32.472 Q1770.48 26.1121 1781.1 26.1121 Q1784.54 26.1121 1787.82 26.8413 Q1791.1 27.5299 1794.18 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1814.52 14.324 L1814.52 27.2059 L1829.87 27.2059 L1829.87 32.9987 L1814.52 32.9987 L1814.52 57.6282 Q1814.52 63.1779 1816.02 64.7578 Q1817.56 66.3376 1822.21 66.3376 L1829.87 66.3376 L1829.87 72.576 L1822.21 72.576 Q1813.59 72.576 1810.3 69.3758 Q1807.02 66.1351 1807.02 57.6282 L1807.02 32.9987 L1801.55 32.9987 L1801.55 27.2059 L1807.02 27.2059 L1807.02 14.324 L1814.52 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1857.25 32.4315 Q1851.26 32.4315 1847.78 37.1306 Q1844.29 41.7891 1844.29 49.9314 Q1844.29 58.0738 1847.73 62.7728 Q1851.22 67.4314 1857.25 67.4314 Q1863.21 67.4314 1866.69 62.7323 Q1870.18 58.0333 1870.18 49.9314 Q1870.18 41.8701 1866.69 37.1711 Q1863.21 32.4315 1857.25 32.4315 M1857.25 26.1121 Q1866.98 26.1121 1872.53 32.4315 Q1878.08 38.7509 1878.08 49.9314 Q1878.08 61.0714 1872.53 67.4314 Q1866.98 73.7508 1857.25 73.7508 Q1847.49 73.7508 1841.94 67.4314 Q1836.43 61.0714 1836.43 49.9314 Q1836.43 38.7509 1841.94 32.4315 Q1847.49 26.1121 1857.25 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1916.72 34.1734 Q1915.47 33.4443 1913.97 33.1202 Q1912.51 32.7556 1910.73 32.7556 Q1904.41 32.7556 1901 36.8875 Q1897.64 40.9789 1897.64 48.6757 L1897.64 72.576 L1890.15 72.576 L1890.15 27.2059 L1897.64 27.2059 L1897.64 34.2544 Q1899.99 30.1225 1903.76 28.1376 Q1907.53 26.1121 1912.91 26.1121 Q1913.68 26.1121 1914.62 26.2337 Q1915.55 26.3147 1916.68 26.5172 L1916.72 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip330)" d="M1943.42 76.7889 Q1940.26 84.8907 1937.26 87.3618 Q1934.26 89.8329 1929.24 89.8329 L1923.28 89.8329 L1923.28 83.5945 L1927.66 83.5945 Q1930.74 83.5945 1932.44 82.1361 Q1934.14 80.6778 1936.21 75.2496 L1937.54 71.8468 L1919.19 27.2059 L1927.09 27.2059 L1941.27 62.6918 L1955.45 27.2059 L1963.35 27.2059 L1943.42 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip332)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="802.382,531.57 804.637,538.307 806.893,545.043 809.15,551.773 811.408,558.496 813.666,565.21 815.926,571.911 818.187,578.598 820.449,585.267 822.712,591.915 824.976,598.539 827.242,605.135 829.508,611.698 831.776,618.226 834.046,624.712 836.317,631.152 838.59,637.542 840.864,643.875 843.14,650.146 845.418,656.349 847.698,662.476 849.98,668.521 852.264,674.477 854.551,680.336 856.84,686.09 859.131,691.731 861.425,697.251 863.722,702.639 866.022,707.889 868.324,712.989 870.63,717.931 872.939,722.706 875.252,727.305 877.568,731.717 879.888,735.935 882.211,739.949 884.539,743.752 886.87,747.335 889.206,750.691 891.546,753.815 893.89,756.7 896.238,759.342 898.591,761.737 900.948,763.883 903.31,765.777 905.676,767.421 908.046,768.814 910.421,769.959 912.8,770.859 915.184,771.519 917.571,771.943 919.963,772.14 922.359,772.116 924.759,771.879 927.163,771.441 929.571,770.81 931.983,769.997 934.398,769.014 936.816,767.873 939.238,766.586 941.663,765.166 944.09,763.624 946.521,761.974 948.955,760.229 951.391,758.401 953.83,756.503 956.271,754.547 958.714,752.545 961.159,750.509 963.607,748.451 966.056,746.382 968.508,744.313 970.961,742.254 973.416,740.215 975.872,738.207 978.33,736.238 980.789,734.317 983.25,732.452 985.712,730.652 988.175,728.925 990.639,727.276 993.105,725.714 995.571,724.245 998.039,722.873 1000.51,721.606 1002.98,720.448 1005.45,719.404 1007.92,718.478 1010.39,717.674 1012.86,716.997 1015.34,716.449 1017.81,716.032 1020.29,715.751 1022.76,715.607 1025.24,715.601 1027.72,715.735 1030.2,716.011 1032.68,716.429 1035.16,716.989 1037.64,717.692 1040.12,718.537 1042.6,719.522 1045.08,720.648 1047.56,721.913 1050.05,723.313 1052.53,724.848 1055.02,726.513 1057.51,728.307 1059.99,730.224 1062.48,732.261 1064.97,734.413 1067.46,736.675 1069.95,739.04 1072.44,741.502 1074.93,744.055 1077.42,746.692 1079.92,749.403 1082.41,752.181 1084.91,755.016 1087.4,757.899 1089.9,760.819 1092.4,763.767 1094.9,766.73 1097.4,769.697 1099.9,772.656 1102.4,775.594 1104.9,778.499 1107.4,781.358 1109.91,784.158 1112.42,786.885 1114.92,789.527 1117.43,792.071 1119.94,794.505 1122.45,796.816 1124.96,798.993 1127.47,801.026 1129.98,802.905 1132.5,804.621 1135.01,806.167 1137.53,807.536 1140.04,808.725 1142.56,809.729 1145.07,810.547 1147.59,811.178 1150.11,811.623 1152.63,811.887 1155.15,811.972 1157.67,811.884 1160.19,811.631 1162.71,811.222 1165.22,810.665 1167.74,809.971 1170.26,809.153 1172.78,808.221 1175.3,807.19 1177.82,806.072 1180.34,804.882 1182.86,803.632 1185.38,802.338 1187.9,801.013 1190.42,799.671 1192.93,798.325 1195.45,796.989 1197.97,795.675 1200.48,794.397 1203,793.164 1205.51,791.989 1208.03,790.883 1210.54,789.854 1213.05,788.912 1215.56,788.064 1218.07,787.32 1220.58,786.684 1223.09,786.163 1225.6,785.762 1228.11,785.485 1230.62,785.335 1233.12,785.314 1235.63,785.424 1238.13,785.665 1240.64,786.038 1243.14,786.54 1245.64,787.17 1248.14,787.925 1250.64,788.8 1253.14,789.791 1255.64,790.891 1258.14,792.096 1260.63,793.396 1263.13,794.783 1265.62,796.249 1268.12,797.784 1270.61,799.377 1273.1,801.017 1275.59,802.692 1278.07,804.39 1280.56,806.097 1283.04,807.803 1285.53,809.492 1288.01,811.152 1290.49,812.769 1292.97,814.332 1295.44,815.827 1297.92,817.242 1300.39,818.567 1302.86,819.79 1305.33,820.903 1307.8,821.897 1310.26,822.766 1312.72,823.503 1315.18,824.105 1317.64,824.57 1320.09,824.896 1322.54,825.083 1324.99,825.136 1327.44,825.056 1329.88,824.849 1332.32,824.523 1334.76,824.084 1337.19,823.543 1339.63,822.908 1342.05,822.191 1344.48,821.404 1346.9,820.559 1349.32,819.668 1351.74,818.743 1354.15,817.799 1356.56,816.847 1358.97,815.899 1361.37,814.969 1363.78,814.068 1366.17,813.206 1368.57,812.395 1370.96,811.643 1373.35,810.961 1375.74,810.355 1378.12,809.833 1380.5,809.401 1382.87,809.063 1385.25,808.825 1387.62,808.688 1389.98,808.655 1392.35,808.725 1394.71,808.9 1397.07,809.177 1399.42,809.553 1401.77,810.025 1404.12,810.588 1406.46,811.236 1408.8,811.963 1411.14,812.762 1413.47,813.623 1415.8,814.538 1418.13,815.496 1420.45,816.488 1422.77,817.503 1425.08,818.529 1427.39,819.556 1429.7,820.571 1432,821.565 1434.3,822.525 1436.59,823.441 1438.88,824.304 1441.16,825.103 1443.44,825.832 1445.72,826.481 1447.99,827.045 1450.25,827.52 1452.51,827.9 1454.77,828.185 1457.02,828.372 1459.26,828.462 1461.5,828.457 1463.73,828.361 1465.96,828.177 1468.18,827.911 1470.4,827.57 1472.61,827.161 1474.81,826.693 1477.01,826.175 1479.21,825.617 1481.4,825.029 1483.58,824.42 1485.76,823.802 1487.93,823.184 1490.09,822.577 1492.26,821.99 1494.41,821.432 1496.56,820.913 1498.7,820.439 1500.84,820.018 1502.98,819.656 1505.1,819.359 1507.23,819.132 1509.34,818.977 1511.45,818.897 1513.56,818.894 1515.66,818.967 1517.75,819.117 1519.84,819.341 1521.92,819.637 1524,820 1526.07,820.428 1528.13,820.913 1530.19,821.449 1532.24,822.031 1534.29,822.65 1536.33,823.299 1538.36,823.969 1540.39,824.651 1542.41,825.338 1544.43,826.019 1546.44,826.688 1548.44,827.335 1550.43,827.954 1552.42,828.536 1554.4,829.076 1556.37,829.567 1558.34,830.004 1560.29,830.385 1562.25,830.706 1564.19,830.965 1566.13,831.162 1568.05,831.297 1569.98,831.372 1571.89,831.388 1573.79,831.351 1575.69,831.263 1577.58,831.131 1579.46,830.959 1581.34,830.755 1583.2,830.526 1585.06,830.278 1586.91,830.02 1588.76,829.758 1590.59,829.5 1592.42,829.253 1594.24,829.024 1596.05,828.821 1597.86,828.648 1599.65,828.512 1601.44,828.418 1603.22,828.368 1605,828.368 1606.76,828.42 1608.52,828.524 1610.27,828.684 1612.01,828.897 1613.74,829.164 1615.47,829.484 1617.19,829.853 1618.9,830.269 1620.6,830.728 1622.29,831.227 1623.98,831.759 1625.66,832.321 1627.32,832.906 1628.98,833.509 1630.64,834.123 1632.28,834.744 1633.91,835.364 1635.54,835.979 1637.16,836.582 1638.77,837.17 1640.37,837.737 1641.96,838.279 1643.54,838.794 1645.11,839.278 1646.67,839.73 1648.23,840.148 1649.77,840.532 1651.31,840.883 1652.83,841.202 1654.35,841.489 1655.86,841.749 1657.35,841.984 1658.84,842.198 1660.32,842.394 1661.79,842.578 1663.25,842.753 1664.7,842.926 1666.14,843.101 1667.57,843.282 1669,843.476 1670.41,843.686 1671.81,843.917 1673.21,844.174 1674.59,844.459 1675.97,844.776 1677.33,845.128 1678.69,845.518 1680.03,845.946 1681.37,846.413 1682.7,846.921 1684.02,847.469 1685.33,848.056 1686.63,848.681 1687.92,849.343 1689.2,850.038 1690.47,850.765 1691.73,851.52 1692.98,852.3 1694.22,853.102 1695.46,853.921 1696.68,854.755 1697.89,855.598 1699.09,856.449 1700.29,857.302 1701.47,858.155 1702.64,859.005 1703.8,859.849 1704.96,860.685 1706.1,861.511 1707.23,862.325 1708.36,863.127 1709.47,863.917 1710.57,864.695 1711.66,865.461 1712.74,866.217 1713.82,866.965 1714.88,867.706 1715.93,868.443 1716.97,869.179 1718,869.917 1719.02,870.66 1720.04,871.413 1721.04,872.177 1722.03,872.958 1723.01,873.759 1723.98,874.582 1724.94,875.431 1725.89,876.309 1726.83,877.22 1727.76,878.164 1728.69,879.145 1729.6,880.164 1730.5,881.221 1731.39,882.319 1732.27,883.456 1733.14,884.634 1734.01,885.851 1734.86,887.106 1735.7,888.399 1736.54,889.727 1737.36,891.088 1738.17,892.48 1738.98,893.9 1739.77,895.345 1740.56,896.813 1741.33,898.299 1742.1,899.802 1742.85,901.317 1743.6,902.843 1744.33,904.376 1745.06,905.913 1745.77,907.453 1746.48,908.993 1747.18,910.532 1747.87,912.068 1748.54,913.602 1749.21,915.134 1749.87,916.663 1750.52,918.191 1751.16,919.719 1751.79,921.25 1752.41,922.787 1753.02,924.333 1753.62,925.892 1754.21,927.468 1754.8,929.066 1755.37,930.692 1755.94,932.35 1756.49,934.047 1757.04,935.789 1757.58,937.581 1758.11,939.429 1758.63,941.339 1759.14,943.317 1759.64,945.368 1760.14,947.497 1760.63,949.709 1761.11,952.007 1761.58,954.394 1762.04,956.873 1762.49,959.445 1762.94,962.11 1763.38,964.868 1763.81,967.716 1764.24,970.652 1764.65,973.67 1765.06,976.763 1765.46,979.925 1765.86,983.144 1766.25,986.408 1766.63,989.706 1767,993.02 1767.36,996.334 1767.72,999.629 1768.07,1002.88 1768.42,1006.08 1768.75,1009.19 1769.08,1012.2 1769.41,1015.08 "/>
+<path clip-path="url(#clip152)" d="M266.588 1408.12 L266.588 1372.25 L275.983 1354.31 L292.255 1364.67 L292.255 1400.54 L282.86 1418.48 L266.588 1408.12  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="791.763,1172.23 1166.96,847.3 1166.96,197.447 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="954.483,1219.2 1329.68,894.274 1329.68,244.42 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1117.2,1266.17 1492.4,941.247 1492.4,291.393 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1279.92,1313.15 1655.12,988.22 1655.12,338.367 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1166.92 1423.22,1354.51 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="791.763,1172.23 796.265,1168.33 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="954.483,1219.2 958.986,1215.3 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1117.2,1266.17 1121.71,1262.27 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1279.92,1313.15 1284.43,1309.25 "/>
+<path clip-path="url(#clip150)" d="M764.355 1182.32 Q760.744 1182.32 758.916 1185.89 Q757.11 1189.43 757.11 1196.56 Q757.11 1203.67 758.916 1207.23 Q760.744 1210.77 764.355 1210.77 Q767.99 1210.77 769.795 1207.23 Q771.624 1203.67 771.624 1196.56 Q771.624 1189.43 769.795 1185.89 Q767.99 1182.32 764.355 1182.32 M764.355 1178.62 Q770.166 1178.62 773.221 1183.23 Q776.3 1187.81 776.3 1196.56 Q776.3 1205.29 773.221 1209.89 Q770.166 1214.48 764.355 1214.48 Q758.545 1214.48 755.467 1209.89 Q752.411 1205.29 752.411 1196.56 Q752.411 1187.81 755.467 1183.23 Q758.545 1178.62 764.355 1178.62 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M890.942 1256.84 L907.261 1256.84 L907.261 1260.78 L885.317 1260.78 L885.317 1256.84 Q887.979 1254.09 892.562 1249.46 Q897.169 1244.81 898.349 1243.46 Q900.594 1240.94 901.474 1239.2 Q902.377 1237.44 902.377 1235.75 Q902.377 1233 900.432 1231.26 Q898.511 1229.53 895.409 1229.53 Q893.21 1229.53 890.757 1230.29 Q888.326 1231.06 885.548 1232.61 L885.548 1227.88 Q888.372 1226.75 890.826 1226.17 Q893.28 1225.59 895.317 1225.59 Q900.687 1225.59 903.881 1228.28 Q907.076 1230.96 907.076 1235.45 Q907.076 1237.58 906.266 1239.5 Q905.479 1241.4 903.372 1244 Q902.793 1244.67 899.692 1247.88 Q896.59 1251.08 890.942 1256.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M927.076 1229.3 Q923.465 1229.3 921.636 1232.86 Q919.83 1236.4 919.83 1243.53 Q919.83 1250.64 921.636 1254.2 Q923.465 1257.75 927.076 1257.75 Q930.71 1257.75 932.516 1254.2 Q934.344 1250.64 934.344 1243.53 Q934.344 1236.4 932.516 1232.86 Q930.71 1229.3 927.076 1229.3 M927.076 1225.59 Q932.886 1225.59 935.941 1230.2 Q939.02 1234.78 939.02 1243.53 Q939.02 1252.26 935.941 1256.87 Q932.886 1261.45 927.076 1261.45 Q921.266 1261.45 918.187 1256.87 Q915.131 1252.26 915.131 1243.53 Q915.131 1234.78 918.187 1230.2 Q921.266 1225.59 927.076 1225.59 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1062.48 1277.26 L1050.68 1295.71 L1062.48 1295.71 L1062.48 1277.26 M1061.25 1273.19 L1067.13 1273.19 L1067.13 1295.71 L1072.06 1295.71 L1072.06 1299.6 L1067.13 1299.6 L1067.13 1307.75 L1062.48 1307.75 L1062.48 1299.6 L1046.88 1299.6 L1046.88 1295.09 L1061.25 1273.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1089.8 1276.27 Q1086.18 1276.27 1084.36 1279.83 Q1082.55 1283.38 1082.55 1290.51 Q1082.55 1297.61 1084.36 1301.18 Q1086.18 1304.72 1089.8 1304.72 Q1093.43 1304.72 1095.24 1301.18 Q1097.06 1297.61 1097.06 1290.51 Q1097.06 1283.38 1095.24 1279.83 Q1093.43 1276.27 1089.8 1276.27 M1089.8 1272.57 Q1095.61 1272.57 1098.66 1277.17 Q1101.74 1281.76 1101.74 1290.51 Q1101.74 1299.23 1098.66 1303.84 Q1095.61 1308.42 1089.8 1308.42 Q1083.99 1308.42 1080.91 1303.84 Q1077.85 1299.23 1077.85 1290.51 Q1077.85 1281.76 1080.91 1277.17 Q1083.99 1272.57 1089.8 1272.57 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1222.93 1335.58 Q1219.79 1335.58 1217.93 1337.73 Q1216.1 1339.89 1216.1 1343.64 Q1216.1 1347.36 1217.93 1349.54 Q1219.79 1351.69 1222.93 1351.69 Q1226.08 1351.69 1227.91 1349.54 Q1229.76 1347.36 1229.76 1343.64 Q1229.76 1339.89 1227.91 1337.73 Q1226.08 1335.58 1222.93 1335.58 M1232.22 1320.93 L1232.22 1325.19 Q1230.46 1324.35 1228.65 1323.91 Q1226.87 1323.47 1225.11 1323.47 Q1220.48 1323.47 1218.03 1326.6 Q1215.6 1329.72 1215.25 1336.04 Q1216.61 1334.03 1218.67 1332.96 Q1220.73 1331.88 1223.21 1331.88 Q1228.42 1331.88 1231.43 1335.05 Q1234.46 1338.2 1234.46 1343.64 Q1234.46 1348.96 1231.31 1352.18 Q1228.16 1355.4 1222.93 1355.4 Q1216.94 1355.4 1213.77 1350.81 Q1210.6 1346.21 1210.6 1337.48 Q1210.6 1329.28 1214.48 1324.42 Q1218.37 1319.54 1224.92 1319.54 Q1226.68 1319.54 1228.47 1319.89 Q1230.27 1320.23 1232.22 1320.93 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1252.52 1323.24 Q1248.91 1323.24 1247.08 1326.81 Q1245.27 1330.35 1245.27 1337.48 Q1245.27 1344.59 1247.08 1348.15 Q1248.91 1351.69 1252.52 1351.69 Q1256.15 1351.69 1257.96 1348.15 Q1259.78 1344.59 1259.78 1337.48 Q1259.78 1330.35 1257.96 1326.81 Q1256.15 1323.24 1252.52 1323.24 M1252.52 1319.54 Q1258.33 1319.54 1261.38 1324.15 Q1264.46 1328.73 1264.46 1337.48 Q1264.46 1346.21 1261.38 1350.81 Q1258.33 1355.4 1252.52 1355.4 Q1246.71 1355.4 1243.63 1350.81 Q1240.57 1346.21 1240.57 1337.48 Q1240.57 1328.73 1243.63 1324.15 Q1246.71 1319.54 1252.52 1319.54 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M721.854 1393.34 L728.284 1393.34 L728.284 1435.45 L751.423 1435.45 L751.423 1440.86 L721.854 1440.86 L721.854 1393.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M770.584 1409.32 Q765.873 1409.32 763.136 1413.01 Q760.399 1416.67 760.399 1423.07 Q760.399 1429.47 763.104 1433.16 Q765.842 1436.82 770.584 1436.82 Q775.263 1436.82 778 1433.13 Q780.737 1429.43 780.737 1423.07 Q780.737 1416.73 778 1413.04 Q775.263 1409.32 770.584 1409.32 M770.584 1404.35 Q778.223 1404.35 782.583 1409.32 Q786.944 1414.28 786.944 1423.07 Q786.944 1431.82 782.583 1436.82 Q778.223 1441.78 770.584 1441.78 Q762.913 1441.78 758.553 1436.82 Q754.224 1431.82 754.224 1423.07 Q754.224 1414.28 758.553 1409.32 Q762.913 1404.35 770.584 1404.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M826.284 1419.34 L826.284 1440.86 L820.427 1440.86 L820.427 1419.54 Q820.427 1414.47 818.454 1411.96 Q816.481 1409.45 812.534 1409.45 Q807.792 1409.45 805.054 1412.47 Q802.317 1415.49 802.317 1420.71 L802.317 1440.86 L796.429 1440.86 L796.429 1405.21 L802.317 1405.21 L802.317 1410.75 Q804.418 1407.54 807.25 1405.94 Q810.115 1404.35 813.839 1404.35 Q819.982 1404.35 823.133 1408.17 Q826.284 1411.96 826.284 1419.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M861.423 1422.62 Q861.423 1416.26 858.781 1412.76 Q856.171 1409.25 851.428 1409.25 Q846.718 1409.25 844.076 1412.76 Q841.466 1416.26 841.466 1422.62 Q841.466 1428.96 844.076 1432.46 Q846.718 1435.96 851.428 1435.96 Q856.171 1435.96 858.781 1432.46 Q861.423 1428.96 861.423 1422.62 M867.279 1436.44 Q867.279 1445.54 863.237 1449.96 Q859.195 1454.42 850.856 1454.42 Q847.768 1454.42 845.031 1453.94 Q842.294 1453.5 839.716 1452.54 L839.716 1446.84 Q842.294 1448.24 844.808 1448.91 Q847.323 1449.58 849.932 1449.58 Q855.693 1449.58 858.558 1446.56 Q861.423 1443.57 861.423 1437.49 L861.423 1434.59 Q859.608 1437.74 856.776 1439.3 Q853.943 1440.86 849.996 1440.86 Q843.439 1440.86 839.429 1435.86 Q835.419 1430.87 835.419 1422.62 Q835.419 1414.35 839.429 1409.35 Q843.439 1404.35 849.996 1404.35 Q853.943 1404.35 856.776 1405.91 Q859.608 1407.47 861.423 1410.62 L861.423 1405.21 L867.279 1405.21 L867.279 1436.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M879.342 1405.21 L885.199 1405.21 L885.199 1440.86 L879.342 1440.86 L879.342 1405.21 M879.342 1391.34 L885.199 1391.34 L885.199 1398.75 L879.342 1398.75 L879.342 1391.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M903.245 1395.09 L903.245 1405.21 L915.308 1405.21 L915.308 1409.76 L903.245 1409.76 L903.245 1429.12 Q903.245 1433.48 904.423 1434.72 Q905.632 1435.96 909.293 1435.96 L915.308 1435.96 L915.308 1440.86 L909.293 1440.86 Q902.513 1440.86 899.935 1438.35 Q897.357 1435.8 897.357 1429.12 L897.357 1409.76 L893.06 1409.76 L893.06 1405.21 L897.357 1405.21 L897.357 1395.09 L903.245 1395.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M922.406 1426.79 L922.406 1405.21 L928.263 1405.21 L928.263 1426.57 Q928.263 1431.63 930.236 1434.18 Q932.209 1436.69 936.156 1436.69 Q940.898 1436.69 943.636 1433.67 Q946.405 1430.64 946.405 1425.42 L946.405 1405.21 L952.261 1405.21 L952.261 1440.86 L946.405 1440.86 L946.405 1435.39 Q944.272 1438.63 941.44 1440.22 Q938.639 1441.78 934.915 1441.78 Q928.772 1441.78 925.589 1437.96 Q922.406 1434.14 922.406 1426.79 M937.143 1404.35 L937.143 1404.35 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M987.782 1410.62 L987.782 1391.34 L993.638 1391.34 L993.638 1440.86 L987.782 1440.86 L987.782 1435.51 Q985.936 1438.7 983.103 1440.26 Q980.302 1441.78 976.355 1441.78 Q969.894 1441.78 965.82 1436.63 Q961.778 1431.47 961.778 1423.07 Q961.778 1414.67 965.82 1409.51 Q969.894 1404.35 976.355 1404.35 Q980.302 1404.35 983.103 1405.91 Q985.936 1407.44 987.782 1410.62 M967.825 1423.07 Q967.825 1429.53 970.467 1433.22 Q973.141 1436.88 977.788 1436.88 Q982.435 1436.88 985.108 1433.22 Q987.782 1429.53 987.782 1423.07 Q987.782 1416.61 985.108 1412.95 Q982.435 1409.25 977.788 1409.25 Q973.141 1409.25 970.467 1412.95 Q967.825 1416.61 967.825 1423.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1036.19 1421.57 L1036.19 1424.44 L1009.27 1424.44 Q1009.65 1430.48 1012.89 1433.67 Q1016.17 1436.82 1022 1436.82 Q1025.37 1436.82 1028.52 1435.99 Q1031.71 1435.16 1034.82 1433.51 L1034.82 1439.05 Q1031.67 1440.38 1028.36 1441.08 Q1025.05 1441.78 1021.65 1441.78 Q1013.12 1441.78 1008.12 1436.82 Q1003.16 1431.85 1003.16 1423.39 Q1003.16 1414.63 1007.87 1409.51 Q1012.61 1404.35 1020.63 1404.35 Q1027.82 1404.35 1031.99 1409 Q1036.19 1413.62 1036.19 1421.57 M1030.34 1419.85 Q1030.27 1415.05 1027.63 1412.18 Q1025.02 1409.32 1020.69 1409.32 Q1015.79 1409.32 1012.83 1412.09 Q1009.9 1414.86 1009.46 1419.89 L1030.34 1419.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1080.59 1391.4 Q1076.33 1398.72 1074.26 1405.88 Q1072.19 1413.04 1072.19 1420.39 Q1072.19 1427.75 1074.26 1434.97 Q1076.36 1442.17 1080.59 1449.45 L1075.5 1449.45 Q1070.73 1441.97 1068.34 1434.75 Q1065.98 1427.52 1065.98 1420.39 Q1065.98 1413.3 1068.34 1406.1 Q1070.7 1398.91 1075.5 1391.4 L1080.59 1391.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1115.41 1410.62 L1115.41 1391.34 L1121.27 1391.34 L1121.27 1440.86 L1115.41 1440.86 L1115.41 1435.51 Q1113.57 1438.7 1110.74 1440.26 Q1107.93 1441.78 1103.99 1441.78 Q1097.53 1441.78 1093.45 1436.63 Q1089.41 1431.47 1089.41 1423.07 Q1089.41 1414.67 1093.45 1409.51 Q1097.53 1404.35 1103.99 1404.35 Q1107.93 1404.35 1110.74 1405.91 Q1113.57 1407.44 1115.41 1410.62 M1095.46 1423.07 Q1095.46 1429.53 1098.1 1433.22 Q1100.77 1436.88 1105.42 1436.88 Q1110.07 1436.88 1112.74 1433.22 Q1115.41 1429.53 1115.41 1423.07 Q1115.41 1416.61 1112.74 1412.95 Q1110.07 1409.25 1105.42 1409.25 Q1100.77 1409.25 1098.1 1412.95 Q1095.46 1416.61 1095.46 1423.07 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1163.83 1421.57 L1163.83 1424.44 L1136.9 1424.44 Q1137.28 1430.48 1140.53 1433.67 Q1143.81 1436.82 1149.63 1436.82 Q1153 1436.82 1156.15 1435.99 Q1159.34 1435.16 1162.46 1433.51 L1162.46 1439.05 Q1159.31 1440.38 1156 1441.08 Q1152.69 1441.78 1149.28 1441.78 Q1140.75 1441.78 1135.75 1436.82 Q1130.79 1431.85 1130.79 1423.39 Q1130.79 1414.63 1135.5 1409.51 Q1140.24 1404.35 1148.26 1404.35 Q1155.45 1404.35 1159.62 1409 Q1163.83 1413.62 1163.83 1421.57 M1157.97 1419.85 Q1157.91 1415.05 1155.26 1412.18 Q1152.65 1409.32 1148.32 1409.32 Q1143.42 1409.32 1140.46 1412.09 Q1137.54 1414.86 1137.09 1419.89 L1157.97 1419.85 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1196.9 1422.62 Q1196.9 1416.26 1194.25 1412.76 Q1191.64 1409.25 1186.9 1409.25 Q1182.19 1409.25 1179.55 1412.76 Q1176.94 1416.26 1176.94 1422.62 Q1176.94 1428.96 1179.55 1432.46 Q1182.19 1435.96 1186.9 1435.96 Q1191.64 1435.96 1194.25 1432.46 Q1196.9 1428.96 1196.9 1422.62 M1202.75 1436.44 Q1202.75 1445.54 1198.71 1449.96 Q1194.67 1454.42 1186.33 1454.42 Q1183.24 1454.42 1180.5 1453.94 Q1177.77 1453.5 1175.19 1452.54 L1175.19 1446.84 Q1177.77 1448.24 1180.28 1448.91 Q1182.8 1449.58 1185.41 1449.58 Q1191.17 1449.58 1194.03 1446.56 Q1196.9 1443.57 1196.9 1437.49 L1196.9 1434.59 Q1195.08 1437.74 1192.25 1439.3 Q1189.42 1440.86 1185.47 1440.86 Q1178.91 1440.86 1174.9 1435.86 Q1170.89 1430.87 1170.89 1422.62 Q1170.89 1414.35 1174.9 1409.35 Q1178.91 1404.35 1185.47 1404.35 Q1189.42 1404.35 1192.25 1405.91 Q1195.08 1407.47 1196.9 1410.62 L1196.9 1405.21 L1202.75 1405.21 L1202.75 1436.44 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1213.89 1391.4 L1218.98 1391.4 Q1223.76 1398.91 1226.11 1406.1 Q1228.5 1413.3 1228.5 1420.39 Q1228.5 1427.52 1226.11 1434.75 Q1223.76 1441.97 1218.98 1449.45 L1213.89 1449.45 Q1218.12 1442.17 1220.19 1434.97 Q1222.29 1427.75 1222.29 1420.39 Q1222.29 1413.04 1220.19 1405.88 Q1218.12 1398.72 1213.89 1391.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1433.84,1345.32 783.99,1157.72 783.99,507.868 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1537.4,1255.63 887.549,1068.04 887.549,418.183 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1640.96,1165.95 991.108,978.352 991.108,328.498 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1744.52,1076.26 1094.67,888.667 1094.67,238.813 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1423.22,1354.51 1798.42,1029.59 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1433.84,1345.32 1426.05,1343.07 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1537.4,1255.63 1529.6,1253.38 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1640.96,1165.95 1633.16,1163.7 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1744.52,1076.26 1736.72,1074.01 "/>
+<path clip-path="url(#clip150)" d="M1471.3 1344.48 Q1467.69 1344.48 1465.86 1348.04 Q1464.05 1351.59 1464.05 1358.72 Q1464.05 1365.82 1465.86 1369.39 Q1467.69 1372.93 1471.3 1372.93 Q1474.93 1372.93 1476.74 1369.39 Q1478.57 1365.82 1478.57 1358.72 Q1478.57 1351.59 1476.74 1348.04 Q1474.93 1344.48 1471.3 1344.48 M1471.3 1340.78 Q1477.11 1340.78 1480.17 1345.38 Q1483.24 1349.97 1483.24 1358.72 Q1483.24 1367.44 1480.17 1372.05 Q1477.11 1376.63 1471.3 1376.63 Q1465.49 1376.63 1462.41 1372.05 Q1459.36 1367.44 1459.36 1358.72 Q1459.36 1349.97 1462.41 1345.38 Q1465.49 1340.78 1471.3 1340.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1563.59 1282.34 L1571.23 1282.34 L1571.23 1255.98 L1562.92 1257.64 L1562.92 1253.38 L1571.18 1251.72 L1575.85 1251.72 L1575.85 1282.34 L1583.49 1282.34 L1583.49 1286.28 L1563.59 1286.28 L1563.59 1282.34 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1602.94 1254.79 Q1599.33 1254.79 1597.5 1258.36 Q1595.69 1261.9 1595.69 1269.03 Q1595.69 1276.14 1597.5 1279.7 Q1599.33 1283.24 1602.94 1283.24 Q1606.57 1283.24 1608.38 1279.7 Q1610.21 1276.14 1610.21 1269.03 Q1610.21 1261.9 1608.38 1258.36 Q1606.57 1254.79 1602.94 1254.79 M1602.94 1251.09 Q1608.75 1251.09 1611.8 1255.7 Q1614.88 1260.28 1614.88 1269.03 Q1614.88 1277.76 1611.8 1282.36 Q1608.75 1286.95 1602.94 1286.95 Q1597.13 1286.95 1594.05 1282.36 Q1590.99 1277.76 1590.99 1269.03 Q1590.99 1260.28 1594.05 1255.7 Q1597.13 1251.09 1602.94 1251.09 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1672.1 1192.66 L1688.42 1192.66 L1688.42 1196.59 L1666.47 1196.59 L1666.47 1192.66 Q1669.14 1189.9 1673.72 1185.27 Q1678.33 1180.62 1679.51 1179.28 Q1681.75 1176.75 1682.63 1175.02 Q1683.53 1173.26 1683.53 1171.57 Q1683.53 1168.81 1681.59 1167.08 Q1679.67 1165.34 1676.57 1165.34 Q1674.37 1165.34 1671.91 1166.11 Q1669.48 1166.87 1666.71 1168.42 L1666.71 1163.7 Q1669.53 1162.56 1671.98 1161.98 Q1674.44 1161.41 1676.47 1161.41 Q1681.84 1161.41 1685.04 1164.09 Q1688.23 1166.78 1688.23 1171.27 Q1688.23 1173.4 1687.42 1175.32 Q1686.64 1177.22 1684.53 1179.81 Q1683.95 1180.48 1680.85 1183.7 Q1677.75 1186.89 1672.1 1192.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1708.23 1165.11 Q1704.62 1165.11 1702.79 1168.67 Q1700.99 1172.22 1700.99 1179.35 Q1700.99 1186.45 1702.79 1190.02 Q1704.62 1193.56 1708.23 1193.56 Q1711.87 1193.56 1713.67 1190.02 Q1715.5 1186.45 1715.5 1179.35 Q1715.5 1172.22 1713.67 1168.67 Q1711.87 1165.11 1708.23 1165.11 M1708.23 1161.41 Q1714.04 1161.41 1717.1 1166.01 Q1720.18 1170.6 1720.18 1179.35 Q1720.18 1188.07 1717.1 1192.68 Q1714.04 1197.26 1708.23 1197.26 Q1702.42 1197.26 1699.34 1192.68 Q1696.29 1188.07 1696.29 1179.35 Q1696.29 1170.6 1699.34 1166.01 Q1702.42 1161.41 1708.23 1161.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1785.66 1088.27 Q1789.02 1088.99 1790.89 1091.26 Q1792.79 1093.53 1792.79 1096.86 Q1792.79 1101.98 1789.27 1104.78 Q1785.75 1107.58 1779.27 1107.58 Q1777.09 1107.58 1774.78 1107.14 Q1772.49 1106.72 1770.03 1105.86 L1770.03 1101.35 Q1771.98 1102.48 1774.29 1103.06 Q1776.61 1103.64 1779.13 1103.64 Q1783.53 1103.64 1785.82 1101.91 Q1788.14 1100.17 1788.14 1096.86 Q1788.14 1093.8 1785.98 1092.09 Q1783.85 1090.36 1780.03 1090.36 L1776.01 1090.36 L1776.01 1086.51 L1780.22 1086.51 Q1783.67 1086.51 1785.5 1085.15 Q1787.33 1083.76 1787.33 1081.17 Q1787.33 1078.5 1785.43 1077.09 Q1783.55 1075.66 1780.03 1075.66 Q1778.11 1075.66 1775.91 1076.07 Q1773.71 1076.49 1771.08 1077.37 L1771.08 1073.2 Q1773.74 1072.46 1776.05 1072.09 Q1778.39 1071.72 1780.45 1071.72 Q1785.77 1071.72 1788.88 1074.15 Q1791.98 1076.56 1791.98 1080.68 Q1791.98 1083.55 1790.33 1085.54 Q1788.69 1087.51 1785.66 1088.27 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1811.65 1075.42 Q1808.04 1075.42 1806.21 1078.99 Q1804.41 1082.53 1804.41 1089.66 Q1804.41 1096.77 1806.21 1100.33 Q1808.04 1103.87 1811.65 1103.87 Q1815.29 1103.87 1817.09 1100.33 Q1818.92 1096.77 1818.92 1089.66 Q1818.92 1082.53 1817.09 1078.99 Q1815.29 1075.42 1811.65 1075.42 M1811.65 1071.72 Q1817.46 1071.72 1820.52 1076.33 Q1823.6 1080.91 1823.6 1089.66 Q1823.6 1098.39 1820.52 1102.99 Q1817.46 1107.58 1811.65 1107.58 Q1805.84 1107.58 1802.76 1102.99 Q1799.71 1098.39 1799.71 1089.66 Q1799.71 1080.91 1802.76 1076.33 Q1805.84 1071.72 1811.65 1071.72 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1508.58 1324.68 L1515.01 1324.68 L1515.01 1366.78 L1538.15 1366.78 L1538.15 1372.2 L1508.58 1372.2 L1508.58 1324.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1560.84 1354.28 Q1553.74 1354.28 1551.01 1355.9 Q1548.27 1357.52 1548.27 1361.44 Q1548.27 1364.56 1550.31 1366.4 Q1552.38 1368.22 1555.91 1368.22 Q1560.78 1368.22 1563.71 1364.78 Q1566.67 1361.31 1566.67 1355.58 L1566.67 1354.28 L1560.84 1354.28 M1572.52 1351.86 L1572.52 1372.2 L1566.67 1372.2 L1566.67 1366.78 Q1564.66 1370.03 1561.67 1371.59 Q1558.68 1373.12 1554.35 1373.12 Q1548.87 1373.12 1545.63 1370.06 Q1542.41 1366.98 1542.41 1361.82 Q1542.41 1355.8 1546.42 1352.75 Q1550.47 1349.69 1558.45 1349.69 L1566.67 1349.69 L1566.67 1349.12 Q1566.67 1345.08 1563.99 1342.88 Q1561.35 1340.65 1556.55 1340.65 Q1553.49 1340.65 1550.59 1341.39 Q1547.7 1342.12 1545.02 1343.58 L1545.02 1338.17 Q1548.24 1336.93 1551.26 1336.32 Q1554.29 1335.69 1557.15 1335.69 Q1564.88 1335.69 1568.7 1339.7 Q1572.52 1343.71 1572.52 1351.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1590.38 1326.43 L1590.38 1336.55 L1602.44 1336.55 L1602.44 1341.1 L1590.38 1341.1 L1590.38 1360.45 Q1590.38 1364.81 1591.56 1366.05 Q1592.77 1367.29 1596.43 1367.29 L1602.44 1367.29 L1602.44 1372.2 L1596.43 1372.2 Q1589.65 1372.2 1587.07 1369.68 Q1584.49 1367.13 1584.49 1360.45 L1584.49 1341.1 L1580.19 1341.1 L1580.19 1336.55 L1584.49 1336.55 L1584.49 1326.43 L1590.38 1326.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1610.14 1336.55 L1616 1336.55 L1616 1372.2 L1610.14 1372.2 L1610.14 1336.55 M1610.14 1322.67 L1616 1322.67 L1616 1330.09 L1610.14 1330.09 L1610.14 1322.67 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1634.05 1326.43 L1634.05 1336.55 L1646.11 1336.55 L1646.11 1341.1 L1634.05 1341.1 L1634.05 1360.45 Q1634.05 1364.81 1635.23 1366.05 Q1636.43 1367.29 1640.09 1367.29 L1646.11 1367.29 L1646.11 1372.2 L1640.09 1372.2 Q1633.32 1372.2 1630.74 1369.68 Q1628.16 1367.13 1628.16 1360.45 L1628.16 1341.1 L1623.86 1341.1 L1623.86 1336.55 L1628.16 1336.55 L1628.16 1326.43 L1634.05 1326.43 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1653.21 1358.13 L1653.21 1336.55 L1659.06 1336.55 L1659.06 1357.9 Q1659.06 1362.96 1661.04 1365.51 Q1663.01 1368.03 1666.96 1368.03 Q1671.7 1368.03 1674.44 1365 Q1677.21 1361.98 1677.21 1356.76 L1677.21 1336.55 L1683.06 1336.55 L1683.06 1372.2 L1677.21 1372.2 L1677.21 1366.72 Q1675.07 1369.97 1672.24 1371.56 Q1669.44 1373.12 1665.72 1373.12 Q1659.57 1373.12 1656.39 1369.3 Q1653.21 1365.48 1653.21 1358.13 M1667.94 1335.69 L1667.94 1335.69 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1718.58 1341.96 L1718.58 1322.67 L1724.44 1322.67 L1724.44 1372.2 L1718.58 1372.2 L1718.58 1366.85 Q1716.74 1370.03 1713.91 1371.59 Q1711.1 1373.12 1707.16 1373.12 Q1700.7 1373.12 1696.62 1367.96 Q1692.58 1362.81 1692.58 1354.4 Q1692.58 1346 1696.62 1340.84 Q1700.7 1335.69 1707.16 1335.69 Q1711.1 1335.69 1713.91 1337.25 Q1716.74 1338.78 1718.58 1341.96 M1698.63 1354.4 Q1698.63 1360.86 1701.27 1364.56 Q1703.94 1368.22 1708.59 1368.22 Q1713.24 1368.22 1715.91 1364.56 Q1718.58 1360.86 1718.58 1354.4 Q1718.58 1347.94 1715.91 1344.28 Q1713.24 1340.59 1708.59 1340.59 Q1703.94 1340.59 1701.27 1344.28 Q1698.63 1347.94 1698.63 1354.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1767 1352.91 L1767 1355.77 L1740.07 1355.77 Q1740.45 1361.82 1743.7 1365 Q1746.98 1368.15 1752.8 1368.15 Q1756.17 1368.15 1759.32 1367.33 Q1762.51 1366.5 1765.63 1364.84 L1765.63 1370.38 Q1762.48 1371.72 1759.17 1372.42 Q1755.86 1373.12 1752.45 1373.12 Q1743.92 1373.12 1738.92 1368.15 Q1733.96 1363.19 1733.96 1354.72 Q1733.96 1345.97 1738.67 1340.84 Q1743.41 1335.69 1751.43 1335.69 Q1758.62 1335.69 1762.79 1340.33 Q1767 1344.95 1767 1352.91 M1761.14 1351.19 Q1761.08 1346.38 1758.43 1343.52 Q1755.82 1340.65 1751.49 1340.65 Q1746.59 1340.65 1743.63 1343.42 Q1740.7 1346.19 1740.26 1351.22 L1761.14 1351.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1811.4 1322.73 Q1807.13 1330.05 1805.06 1337.22 Q1802.99 1344.38 1802.99 1351.73 Q1802.99 1359.08 1805.06 1366.31 Q1807.16 1373.5 1811.4 1380.79 L1806.3 1380.79 Q1801.53 1373.31 1799.14 1366.08 Q1796.79 1358.86 1796.79 1351.73 Q1796.79 1344.63 1799.14 1337.44 Q1801.5 1330.25 1806.3 1322.73 L1811.4 1322.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1846.22 1341.96 L1846.22 1322.67 L1852.07 1322.67 L1852.07 1372.2 L1846.22 1372.2 L1846.22 1366.85 Q1844.37 1370.03 1841.54 1371.59 Q1838.74 1373.12 1834.79 1373.12 Q1828.33 1373.12 1824.25 1367.96 Q1820.21 1362.81 1820.21 1354.4 Q1820.21 1346 1824.25 1340.84 Q1828.33 1335.69 1834.79 1335.69 Q1838.74 1335.69 1841.54 1337.25 Q1844.37 1338.78 1846.22 1341.96 M1826.26 1354.4 Q1826.26 1360.86 1828.9 1364.56 Q1831.58 1368.22 1836.22 1368.22 Q1840.87 1368.22 1843.54 1364.56 Q1846.22 1360.86 1846.22 1354.4 Q1846.22 1347.94 1843.54 1344.28 Q1840.87 1340.59 1836.22 1340.59 Q1831.58 1340.59 1828.9 1344.28 Q1826.26 1347.94 1826.26 1354.4 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1894.63 1352.91 L1894.63 1355.77 L1867.7 1355.77 Q1868.08 1361.82 1871.33 1365 Q1874.61 1368.15 1880.43 1368.15 Q1883.81 1368.15 1886.96 1367.33 Q1890.14 1366.5 1893.26 1364.84 L1893.26 1370.38 Q1890.11 1371.72 1886.8 1372.42 Q1883.49 1373.12 1880.08 1373.12 Q1871.55 1373.12 1866.55 1368.15 Q1861.59 1363.19 1861.59 1354.72 Q1861.59 1345.97 1866.3 1340.84 Q1871.04 1335.69 1879.06 1335.69 Q1886.26 1335.69 1890.43 1340.33 Q1894.63 1344.95 1894.63 1352.91 M1888.77 1351.19 Q1888.71 1346.38 1886.07 1343.52 Q1883.46 1340.65 1879.13 1340.65 Q1874.23 1340.65 1871.27 1343.42 Q1868.34 1346.19 1867.89 1351.22 L1888.77 1351.19 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1927.7 1353.96 Q1927.7 1347.59 1925.06 1344.09 Q1922.45 1340.59 1917.7 1340.59 Q1912.99 1340.59 1910.35 1344.09 Q1907.74 1347.59 1907.74 1353.96 Q1907.74 1360.29 1910.35 1363.79 Q1912.99 1367.29 1917.7 1367.29 Q1922.45 1367.29 1925.06 1363.79 Q1927.7 1360.29 1927.7 1353.96 M1933.55 1367.77 Q1933.55 1376.87 1929.51 1381.3 Q1925.47 1385.75 1917.13 1385.75 Q1914.04 1385.75 1911.31 1385.28 Q1908.57 1384.83 1905.99 1383.88 L1905.99 1378.18 Q1908.57 1379.58 1911.08 1380.25 Q1913.6 1380.92 1916.21 1380.92 Q1921.97 1380.92 1924.83 1377.89 Q1927.7 1374.9 1927.7 1368.82 L1927.7 1365.93 Q1925.88 1369.08 1923.05 1370.64 Q1920.22 1372.2 1916.27 1372.2 Q1909.71 1372.2 1905.7 1367.2 Q1901.69 1362.2 1901.69 1353.96 Q1901.69 1345.68 1905.7 1340.68 Q1909.71 1335.69 1916.27 1335.69 Q1920.22 1335.69 1923.05 1337.25 Q1925.88 1338.81 1927.7 1341.96 L1927.7 1336.55 L1933.55 1336.55 L1933.55 1367.77 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1944.69 1322.73 L1949.79 1322.73 Q1954.56 1330.25 1956.92 1337.44 Q1959.3 1344.63 1959.3 1351.73 Q1959.3 1358.86 1956.92 1366.08 Q1954.56 1373.31 1949.79 1380.79 L1944.69 1380.79 Q1948.93 1373.5 1951 1366.31 Q1953.1 1359.08 1953.1 1351.73 Q1953.1 1344.38 1951 1337.22 Q1948.93 1330.05 1944.69 1322.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,1080.41 1148.56,755.48 1798.42,943.077 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,910.11 1148.56,585.183 1798.42,772.78 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,739.813 1148.56,414.886 1798.42,602.483 "/>
+<polyline clip-path="url(#clip152)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="773.371,569.516 1148.56,244.589 1798.42,432.185 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1166.92 773.371,517.064 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,1080.41 777.873,1076.51 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,910.11 777.873,906.211 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,739.813 777.873,735.914 "/>
+<polyline clip-path="url(#clip150)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="773.371,569.516 777.873,565.617 "/>
+<path clip-path="url(#clip150)" d="M678.205 1093.75 L685.844 1093.75 L685.844 1067.39 L677.534 1069.05 L677.534 1064.79 L685.798 1063.13 L690.474 1063.13 L690.474 1093.75 L698.113 1093.75 L698.113 1097.69 L678.205 1097.69 L678.205 1093.75 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M707.557 1091.81 L712.441 1091.81 L712.441 1097.69 L707.557 1097.69 L707.557 1091.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M732.626 1066.21 Q729.015 1066.21 727.187 1069.77 Q725.381 1073.31 725.381 1080.44 Q725.381 1087.55 727.187 1091.11 Q729.015 1094.65 732.626 1094.65 Q736.261 1094.65 738.066 1091.11 Q739.895 1087.55 739.895 1080.44 Q739.895 1073.31 738.066 1069.77 Q736.261 1066.21 732.626 1066.21 M732.626 1062.5 Q738.437 1062.5 741.492 1067.11 Q744.571 1071.69 744.571 1080.44 Q744.571 1089.17 741.492 1093.78 Q738.437 1098.36 732.626 1098.36 Q726.816 1098.36 723.738 1093.78 Q720.682 1089.17 720.682 1080.44 Q720.682 1071.69 723.738 1067.11 Q726.816 1062.5 732.626 1062.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M679.201 923.455 L686.84 923.455 L686.84 897.089 L678.53 898.756 L678.53 894.497 L686.793 892.83 L691.469 892.83 L691.469 923.455 L699.108 923.455 L699.108 927.39 L679.201 927.39 L679.201 923.455 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M708.553 921.51 L713.437 921.51 L713.437 927.39 L708.553 927.39 L708.553 921.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M723.668 892.83 L742.025 892.83 L742.025 896.765 L727.951 896.765 L727.951 905.237 Q728.969 904.89 729.988 904.728 Q731.006 904.543 732.025 904.543 Q737.812 904.543 741.191 907.714 Q744.571 910.885 744.571 916.302 Q744.571 921.881 741.099 924.983 Q737.626 928.061 731.307 928.061 Q729.131 928.061 726.863 927.691 Q724.617 927.321 722.21 926.58 L722.21 921.881 Q724.293 923.015 726.515 923.571 Q728.738 924.126 731.214 924.126 Q735.219 924.126 737.557 922.02 Q739.895 919.913 739.895 916.302 Q739.895 912.691 737.557 910.585 Q735.219 908.478 731.214 908.478 Q729.339 908.478 727.464 908.895 Q725.613 909.311 723.668 910.191 L723.668 892.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M681.423 753.158 L697.742 753.158 L697.742 757.093 L675.798 757.093 L675.798 753.158 Q678.46 750.403 683.043 745.773 Q687.65 741.121 688.83 739.778 Q691.076 737.255 691.955 735.519 Q692.858 733.76 692.858 732.07 Q692.858 729.315 690.914 727.579 Q688.992 725.843 685.891 725.843 Q683.692 725.843 681.238 726.607 Q678.807 727.371 676.03 728.922 L676.03 724.2 Q678.854 723.065 681.307 722.487 Q683.761 721.908 685.798 721.908 Q691.168 721.908 694.363 724.593 Q697.557 727.278 697.557 731.769 Q697.557 733.899 696.747 735.82 Q695.96 737.718 693.854 740.311 Q693.275 740.982 690.173 744.199 Q687.071 747.394 681.423 753.158 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M707.557 751.213 L712.441 751.213 L712.441 757.093 L707.557 757.093 L707.557 751.213 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M732.626 725.612 Q729.015 725.612 727.187 729.176 Q725.381 732.718 725.381 739.848 Q725.381 746.954 727.187 750.519 Q729.015 754.06 732.626 754.06 Q736.261 754.06 738.066 750.519 Q739.895 746.954 739.895 739.848 Q739.895 732.718 738.066 729.176 Q736.261 725.612 732.626 725.612 M732.626 721.908 Q738.437 721.908 741.492 726.514 Q744.571 731.098 744.571 739.848 Q744.571 748.574 741.492 753.181 Q738.437 757.764 732.626 757.764 Q726.816 757.764 723.738 753.181 Q720.682 748.574 720.682 739.848 Q720.682 731.098 723.738 726.514 Q726.816 721.908 732.626 721.908 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M682.418 582.861 L698.738 582.861 L698.738 586.796 L676.793 586.796 L676.793 582.861 Q679.455 580.106 684.039 575.476 Q688.645 570.824 689.826 569.481 Q692.071 566.958 692.951 565.222 Q693.854 563.463 693.854 561.773 Q693.854 559.018 691.909 557.282 Q689.988 555.546 686.886 555.546 Q684.687 555.546 682.233 556.31 Q679.803 557.074 677.025 558.625 L677.025 553.902 Q679.849 552.768 682.303 552.189 Q684.756 551.611 686.793 551.611 Q692.164 551.611 695.358 554.296 Q698.553 556.981 698.553 561.472 Q698.553 563.601 697.742 565.523 Q696.955 567.421 694.849 570.013 Q694.27 570.685 691.168 573.902 Q688.067 577.097 682.418 582.861 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M708.553 580.916 L713.437 580.916 L713.437 586.796 L708.553 586.796 L708.553 580.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M723.668 552.236 L742.025 552.236 L742.025 556.171 L727.951 556.171 L727.951 564.643 Q728.969 564.296 729.988 564.134 Q731.006 563.949 732.025 563.949 Q737.812 563.949 741.191 567.12 Q744.571 570.291 744.571 575.708 Q744.571 581.286 741.099 584.388 Q737.626 587.467 731.307 587.467 Q729.131 587.467 726.863 587.097 Q724.617 586.726 722.21 585.986 L722.21 581.286 Q724.293 582.421 726.515 582.976 Q728.738 583.532 731.214 583.532 Q735.219 583.532 737.557 581.425 Q739.895 579.319 739.895 575.708 Q739.895 572.097 737.557 569.99 Q735.219 567.884 731.214 567.884 Q729.339 567.884 727.464 568.3 Q725.613 568.717 723.668 569.597 L723.668 552.236 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M577.156 1149.14 L600.805 1157.86 L600.805 1140.38 L577.156 1149.14 M570.822 1152.76 L570.822 1145.48 L618.342 1127.37 L618.342 1134.05 L606.152 1138.38 L606.152 1159.8 L618.342 1164.13 L618.342 1170.91 L570.822 1152.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M568.817 1120.68 L568.817 1114.83 L618.342 1114.83 L618.342 1120.68 L568.817 1120.68 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M572.573 1096.78 L582.694 1096.78 L582.694 1084.72 L587.246 1084.72 L587.246 1096.78 L606.597 1096.78 Q610.958 1096.78 612.199 1095.6 Q613.44 1094.39 613.44 1090.73 L613.44 1084.72 L618.342 1084.72 L618.342 1090.73 Q618.342 1097.51 615.828 1100.09 Q613.281 1102.67 606.597 1102.67 L587.246 1102.67 L587.246 1106.96 L582.694 1106.96 L582.694 1102.67 L572.573 1102.67 L572.573 1096.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M582.694 1077.01 L582.694 1071.16 L618.342 1071.16 L618.342 1077.01 L582.694 1077.01 M568.817 1077.01 L568.817 1071.16 L576.233 1071.16 L576.233 1077.01 L568.817 1077.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M572.573 1053.11 L582.694 1053.11 L582.694 1041.05 L587.246 1041.05 L587.246 1053.11 L606.597 1053.11 Q610.958 1053.11 612.199 1051.93 Q613.44 1050.72 613.44 1047.06 L613.44 1041.05 L618.342 1041.05 L618.342 1047.06 Q618.342 1053.84 615.828 1056.42 Q613.281 1059 606.597 1059 L587.246 1059 L587.246 1063.29 L582.694 1063.29 L582.694 1059 L572.573 1059 L572.573 1053.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M604.274 1033.95 L582.694 1033.95 L582.694 1028.09 L604.051 1028.09 Q609.112 1028.09 611.658 1026.12 Q614.173 1024.15 614.173 1020.2 Q614.173 1015.46 611.149 1012.72 Q608.125 1009.95 602.905 1009.95 L582.694 1009.95 L582.694 1004.09 L618.342 1004.09 L618.342 1009.95 L612.868 1009.95 Q616.114 1012.08 617.705 1014.92 Q619.265 1017.72 619.265 1021.44 Q619.265 1027.58 615.446 1030.77 Q611.626 1033.95 604.274 1033.95 M581.835 1019.21 L581.835 1019.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M588.105 968.573 L568.817 968.573 L568.817 962.717 L618.342 962.717 L618.342 968.573 L612.995 968.573 Q616.178 970.419 617.737 973.252 Q619.265 976.053 619.265 980 Q619.265 986.461 614.109 990.535 Q608.953 994.577 600.55 994.577 Q592.147 994.577 586.991 990.535 Q581.835 986.461 581.835 980 Q581.835 976.053 583.394 973.252 Q584.922 970.419 588.105 968.573 M600.55 988.53 Q607.011 988.53 610.703 985.888 Q614.363 983.214 614.363 978.567 Q614.363 973.92 610.703 971.247 Q607.011 968.573 600.55 968.573 Q594.089 968.573 590.428 971.247 Q586.736 973.92 586.736 978.567 Q586.736 983.214 590.428 985.888 Q594.089 988.53 600.55 988.53 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M599.054 920.162 L601.919 920.162 L601.919 947.089 Q607.966 946.707 611.149 943.46 Q614.3 940.182 614.3 934.358 Q614.3 930.984 613.472 927.833 Q612.645 924.65 610.99 921.531 L616.528 921.531 Q617.865 924.682 618.565 927.992 Q619.265 931.302 619.265 934.708 Q619.265 943.238 614.3 948.235 Q609.335 953.2 600.868 953.2 Q592.115 953.2 586.991 948.489 Q581.835 943.747 581.835 935.726 Q581.835 928.533 586.482 924.363 Q591.097 920.162 599.054 920.162 M597.335 926.018 Q592.529 926.082 589.665 928.724 Q586.8 931.334 586.8 935.662 Q586.8 940.564 589.569 943.524 Q592.338 946.452 597.367 946.898 L597.335 926.018 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M568.881 875.761 Q576.201 880.026 583.363 882.095 Q590.524 884.164 597.876 884.164 Q605.229 884.164 612.454 882.095 Q619.647 879.994 626.936 875.761 L626.936 880.854 Q619.456 885.628 612.231 888.015 Q605.006 890.37 597.876 890.37 Q590.779 890.37 583.585 888.015 Q576.392 885.66 568.881 880.854 L568.881 875.761 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M612.931 862.457 L612.931 851.953 L576.679 851.953 L578.97 863.38 L573.114 863.38 L570.822 852.017 L570.822 845.588 L612.931 845.588 L612.931 835.084 L618.342 835.084 L618.342 862.457 L612.931 862.457 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M575.055 808.348 Q575.055 813.314 579.957 815.828 Q584.827 818.311 594.63 818.311 Q604.401 818.311 609.303 815.828 Q614.173 813.314 614.173 808.348 Q614.173 803.351 609.303 800.869 Q604.401 798.354 594.63 798.354 Q584.827 798.354 579.957 800.869 Q575.055 803.351 575.055 808.348 M569.963 808.348 Q569.963 800.359 576.297 796.158 Q582.599 791.925 594.63 791.925 Q606.629 791.925 612.963 796.158 Q619.265 800.359 619.265 808.348 Q619.265 816.337 612.963 820.571 Q606.629 824.772 594.63 824.772 Q582.599 824.772 576.297 820.571 Q569.963 816.337 569.963 808.348 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M575.055 766.876 Q575.055 771.841 579.957 774.355 Q584.827 776.838 594.63 776.838 Q604.401 776.838 609.303 774.355 Q614.173 771.841 614.173 766.876 Q614.173 761.879 609.303 759.396 Q604.401 756.882 594.63 756.882 Q584.827 756.882 579.957 759.396 Q575.055 761.879 575.055 766.876 M569.963 766.876 Q569.963 758.887 576.297 754.685 Q582.599 750.452 594.63 750.452 Q606.629 750.452 612.963 754.685 Q619.265 758.887 619.265 766.876 Q619.265 774.865 612.963 779.098 Q606.629 783.299 594.63 783.299 Q582.599 783.299 576.297 779.098 Q569.963 774.865 569.963 766.876 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M610.258 738.485 L610.258 731.769 L615.732 731.769 L625.917 736.989 L625.917 741.095 L615.732 738.485 L610.258 738.485 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M575.055 704.683 Q575.055 709.648 579.957 712.163 Q584.827 714.645 594.63 714.645 Q604.401 714.645 609.303 712.163 Q614.173 709.648 614.173 704.683 Q614.173 699.686 609.303 697.203 Q604.401 694.689 594.63 694.689 Q584.827 694.689 579.957 697.203 Q575.055 699.686 575.055 704.683 M569.963 704.683 Q569.963 696.694 576.297 692.493 Q582.599 688.259 594.63 688.259 Q606.629 688.259 612.963 692.493 Q619.265 696.694 619.265 704.683 Q619.265 712.672 612.963 716.905 Q606.629 721.106 594.63 721.106 Q582.599 721.106 576.297 716.905 Q569.963 712.672 569.963 704.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M575.055 663.21 Q575.055 668.176 579.957 670.69 Q584.827 673.173 594.63 673.173 Q604.401 673.173 609.303 670.69 Q614.173 668.176 614.173 663.21 Q614.173 658.213 609.303 655.731 Q604.401 653.216 594.63 653.216 Q584.827 653.216 579.957 655.731 Q575.055 658.213 575.055 663.21 M569.963 663.21 Q569.963 655.221 576.297 651.02 Q582.599 646.787 594.63 646.787 Q606.629 646.787 612.963 651.02 Q619.265 655.221 619.265 663.21 Q619.265 671.199 612.963 675.432 Q606.629 679.634 594.63 679.634 Q582.599 679.634 576.297 675.432 Q569.963 671.199 569.963 663.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M575.055 621.738 Q575.055 626.703 579.957 629.217 Q584.827 631.7 594.63 631.7 Q604.401 631.7 609.303 629.217 Q614.173 626.703 614.173 621.738 Q614.173 616.741 609.303 614.258 Q604.401 611.744 594.63 611.744 Q584.827 611.744 579.957 614.258 Q575.055 616.741 575.055 621.738 M569.963 621.738 Q569.963 613.749 576.297 609.547 Q582.599 605.314 594.63 605.314 Q606.629 605.314 612.963 609.547 Q619.265 613.749 619.265 621.738 Q619.265 629.727 612.963 633.96 Q606.629 638.161 594.63 638.161 Q582.599 638.161 576.297 633.96 Q569.963 629.727 569.963 621.738 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M568.817 556.075 L573.687 556.075 L573.687 561.677 Q573.687 564.828 574.96 566.07 Q576.233 567.279 579.543 567.279 L582.694 567.279 L582.694 557.635 L587.246 557.635 L587.246 567.279 L618.342 567.279 L618.342 573.167 L587.246 573.167 L587.246 578.769 L582.694 578.769 L582.694 573.167 L580.211 573.167 Q574.26 573.167 571.554 570.398 Q568.817 567.629 568.817 561.614 L568.817 556.075 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M572.573 546.527 L582.694 546.527 L582.694 534.464 L587.246 534.464 L587.246 546.527 L606.597 546.527 Q610.958 546.527 612.199 545.349 Q613.44 544.14 613.44 540.479 L613.44 534.464 L618.342 534.464 L618.342 540.479 Q618.342 547.259 615.828 549.837 Q613.281 552.415 606.597 552.415 L587.246 552.415 L587.246 556.712 L582.694 556.712 L582.694 552.415 L572.573 552.415 L572.573 546.527 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M568.881 527.684 L568.881 522.592 Q576.392 517.818 583.585 515.462 Q590.779 513.075 597.876 513.075 Q605.006 513.075 612.231 515.462 Q619.456 517.818 626.936 522.592 L626.936 527.684 Q619.647 523.451 612.454 521.382 Q605.229 519.282 597.876 519.282 Q590.524 519.282 583.363 521.382 Q576.201 523.451 568.881 527.684 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M647.371 14.0809 L647.371 22.0612 Q642.712 19.8332 638.58 18.7395 Q634.448 17.6457 630.6 17.6457 Q623.916 17.6457 620.27 20.2383 Q616.665 22.8309 616.665 27.611 Q616.665 31.6214 619.055 33.6873 Q621.485 35.7128 628.21 36.9686 L633.152 37.9813 Q642.307 39.7232 646.641 44.1387 Q651.016 48.5136 651.016 55.8863 Q651.016 64.6767 645.102 69.2137 Q639.228 73.7508 627.845 73.7508 Q623.551 73.7508 618.69 72.7785 Q613.87 71.8063 608.685 69.9024 L608.685 61.4765 Q613.667 64.2716 618.447 65.6895 Q623.227 67.1073 627.845 67.1073 Q634.853 67.1073 638.661 64.3527 Q642.469 61.598 642.469 56.4939 Q642.469 52.0379 639.714 49.5264 Q637 47.0148 630.762 45.759 L625.779 44.7868 Q616.624 42.9639 612.533 39.075 Q608.441 35.1862 608.441 28.2591 Q608.441 20.2383 614.072 15.6203 Q619.743 11.0023 629.668 11.0023 Q633.922 11.0023 638.337 11.7719 Q642.753 12.5416 647.371 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M670.663 65.7705 L670.663 89.8329 L663.169 89.8329 L663.169 27.2059 L670.663 27.2059 L670.663 34.0924 Q673.013 30.0415 676.578 28.0971 Q680.183 26.1121 685.166 26.1121 Q693.429 26.1121 698.574 32.6746 Q703.759 39.2371 703.759 49.9314 Q703.759 60.6258 698.574 67.1883 Q693.429 73.7508 685.166 73.7508 Q680.183 73.7508 676.578 71.8063 Q673.013 69.8214 670.663 65.7705 M696.022 49.9314 Q696.022 41.7081 692.619 37.0496 Q689.257 32.3505 683.343 32.3505 Q677.428 32.3505 674.026 37.0496 Q670.663 41.7081 670.663 49.9314 Q670.663 58.1548 674.026 62.8538 Q677.428 67.5124 683.343 67.5124 Q689.257 67.5124 692.619 62.8538 Q696.022 58.1548 696.022 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M736.734 49.7694 Q727.7 49.7694 724.216 51.8354 Q720.733 53.9013 720.733 58.8839 Q720.733 62.8538 723.325 65.2034 Q725.958 67.5124 730.455 67.5124 Q736.653 67.5124 740.379 63.1374 Q744.147 58.7219 744.147 51.4303 L744.147 49.7694 L736.734 49.7694 M751.6 46.6907 L751.6 72.576 L744.147 72.576 L744.147 65.6895 Q741.595 69.8214 737.787 71.8063 Q733.979 73.7508 728.47 73.7508 Q721.502 73.7508 717.37 69.8619 Q713.279 65.9325 713.279 59.3701 Q713.279 51.7138 718.383 47.825 Q723.528 43.9361 733.695 43.9361 L744.147 43.9361 L744.147 43.2069 Q744.147 38.0623 740.744 35.2672 Q737.382 32.4315 731.265 32.4315 Q727.376 32.4315 723.69 33.3632 Q720.003 34.295 716.601 36.1584 L716.601 29.2718 Q720.692 27.692 724.54 26.9223 Q728.389 26.1121 732.035 26.1121 Q741.878 26.1121 746.739 31.2163 Q751.6 36.3204 751.6 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M799.604 28.9478 L799.604 35.9153 Q796.444 34.1734 793.244 33.3227 Q790.084 32.4315 786.843 32.4315 Q779.592 32.4315 775.582 37.0496 Q771.571 41.6271 771.571 49.9314 Q771.571 58.2358 775.582 62.8538 Q779.592 67.4314 786.843 67.4314 Q790.084 67.4314 793.244 66.5807 Q796.444 65.6895 799.604 63.9476 L799.604 70.8341 Q796.484 72.2924 793.122 73.0216 Q789.8 73.7508 786.033 73.7508 Q775.784 73.7508 769.748 67.3098 Q763.713 60.8689 763.713 49.9314 Q763.713 38.832 769.789 32.472 Q775.906 26.1121 786.519 26.1121 Q789.962 26.1121 793.244 26.8413 Q796.525 27.5299 799.604 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M851.374 48.0275 L851.374 51.6733 L817.104 51.6733 Q817.59 59.3701 821.722 63.421 Q825.894 67.4314 833.307 67.4314 Q837.601 67.4314 841.612 66.3781 Q845.662 65.3249 849.632 63.2184 L849.632 70.267 Q845.622 71.9684 841.409 72.8596 Q837.196 73.7508 832.862 73.7508 Q822.005 73.7508 815.645 67.4314 Q809.326 61.1119 809.326 50.3365 Q809.326 39.1965 815.321 32.6746 Q821.357 26.1121 831.565 26.1121 Q840.72 26.1121 846.027 32.0264 Q851.374 37.9003 851.374 48.0275 M843.921 45.84 Q843.839 39.7232 840.477 36.0774 Q837.156 32.4315 831.646 32.4315 Q825.408 32.4315 821.641 35.9558 Q817.914 39.4801 817.347 45.8805 L843.921 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M926.559 14.0809 L926.559 22.0612 Q921.9 19.8332 917.769 18.7395 Q913.637 17.6457 909.788 17.6457 Q903.104 17.6457 899.458 20.2383 Q895.853 22.8309 895.853 27.611 Q895.853 31.6214 898.243 33.6873 Q900.674 35.7128 907.398 36.9686 L912.34 37.9813 Q921.495 39.7232 925.83 44.1387 Q930.205 48.5136 930.205 55.8863 Q930.205 64.6767 924.29 69.2137 Q918.417 73.7508 907.034 73.7508 Q902.74 73.7508 897.879 72.7785 Q893.058 71.8063 887.873 69.9024 L887.873 61.4765 Q892.855 64.2716 897.636 65.6895 Q902.416 67.1073 907.034 67.1073 Q914.042 67.1073 917.85 64.3527 Q921.657 61.598 921.657 56.4939 Q921.657 52.0379 918.903 49.5264 Q916.189 47.0148 909.95 45.759 L904.968 44.7868 Q895.813 42.9639 891.721 39.075 Q887.63 35.1862 887.63 28.2591 Q887.63 20.2383 893.261 15.6203 Q898.932 11.0023 908.857 11.0023 Q913.11 11.0023 917.525 11.7719 Q921.941 12.5416 926.559 14.0809 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M980.355 45.1919 L980.355 72.576 L972.901 72.576 L972.901 45.4349 Q972.901 38.994 970.39 35.7938 Q967.878 32.5936 962.855 32.5936 Q956.819 32.5936 953.335 36.4419 Q949.852 40.2903 949.852 46.9338 L949.852 72.576 L942.357 72.576 L942.357 9.54393 L949.852 9.54393 L949.852 34.2544 Q952.525 30.163 956.131 28.1376 Q959.776 26.1121 964.516 26.1121 Q972.334 26.1121 976.345 30.9732 Q980.355 35.7938 980.355 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M994.452 54.671 L994.452 27.2059 L1001.91 27.2059 L1001.91 54.3874 Q1001.91 60.8284 1004.42 64.0691 Q1006.93 67.2693 1011.95 67.2693 Q1017.99 67.2693 1021.47 63.421 Q1025 59.5726 1025 52.9291 L1025 27.2059 L1032.45 27.2059 L1032.45 72.576 L1025 72.576 L1025 65.6084 Q1022.28 69.7404 1018.68 71.7658 Q1015.11 73.7508 1010.37 73.7508 Q1002.55 73.7508 998.503 68.8897 Q994.452 64.0286 994.452 54.671 M1013.21 26.1121 L1013.21 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1055.18 14.324 L1055.18 27.2059 L1070.53 27.2059 L1070.53 32.9987 L1055.18 32.9987 L1055.18 57.6282 Q1055.18 63.1779 1056.67 64.7578 Q1058.21 66.3376 1062.87 66.3376 L1070.53 66.3376 L1070.53 72.576 L1062.87 72.576 Q1054.24 72.576 1050.96 69.3758 Q1047.68 66.1351 1047.68 57.6282 L1047.68 32.9987 L1042.21 32.9987 L1042.21 27.2059 L1047.68 27.2059 L1047.68 14.324 L1055.18 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1087.7 14.324 L1087.7 27.2059 L1103.06 27.2059 L1103.06 32.9987 L1087.7 32.9987 L1087.7 57.6282 Q1087.7 63.1779 1089.2 64.7578 Q1090.74 66.3376 1095.4 66.3376 L1103.06 66.3376 L1103.06 72.576 L1095.4 72.576 Q1086.77 72.576 1083.49 69.3758 Q1080.21 66.1351 1080.21 57.6282 L1080.21 32.9987 L1074.74 32.9987 L1074.74 27.2059 L1080.21 27.2059 L1080.21 14.324 L1087.7 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1112.86 9.54393 L1120.31 9.54393 L1120.31 72.576 L1112.86 72.576 L1112.86 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1174.72 48.0275 L1174.72 51.6733 L1140.45 51.6733 Q1140.93 59.3701 1145.06 63.421 Q1149.24 67.4314 1156.65 67.4314 Q1160.94 67.4314 1164.95 66.3781 Q1169.01 65.3249 1172.98 63.2184 L1172.98 70.267 Q1168.97 71.9684 1164.75 72.8596 Q1160.54 73.7508 1156.2 73.7508 Q1145.35 73.7508 1138.99 67.4314 Q1132.67 61.1119 1132.67 50.3365 Q1132.67 39.1965 1138.66 32.6746 Q1144.7 26.1121 1154.91 26.1121 Q1164.06 26.1121 1169.37 32.0264 Q1174.72 37.9003 1174.72 48.0275 M1167.26 45.84 Q1167.18 39.7232 1163.82 36.0774 Q1160.5 32.4315 1154.99 32.4315 Q1148.75 32.4315 1144.98 35.9558 Q1141.26 39.4801 1140.69 45.8805 L1167.26 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1242.33 44.2197 Q1244.96 45.1109 1247.43 48.0275 Q1249.94 50.9442 1252.45 56.0483 L1260.76 72.576 L1251.97 72.576 L1244.23 57.061 Q1241.23 50.9847 1238.4 48.9997 Q1235.6 47.0148 1230.74 47.0148 L1221.83 47.0148 L1221.83 72.576 L1213.65 72.576 L1213.65 12.096 L1232.12 12.096 Q1242.49 12.096 1247.59 16.4305 Q1252.7 20.7649 1252.7 29.5149 Q1252.7 35.2267 1250.02 38.994 Q1247.39 42.7613 1242.33 44.2197 M1221.83 18.8205 L1221.83 40.2903 L1232.12 40.2903 Q1238.03 40.2903 1241.03 37.5762 Q1244.07 34.8216 1244.07 29.5149 Q1244.07 24.2082 1241.03 21.5346 Q1238.03 18.8205 1232.12 18.8205 L1221.83 18.8205 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1306.05 48.0275 L1306.05 51.6733 L1271.78 51.6733 Q1272.26 59.3701 1276.4 63.421 Q1280.57 67.4314 1287.98 67.4314 Q1292.27 67.4314 1296.28 66.3781 Q1300.34 65.3249 1304.31 63.2184 L1304.31 70.267 Q1300.3 71.9684 1296.08 72.8596 Q1291.87 73.7508 1287.54 73.7508 Q1276.68 73.7508 1270.32 67.4314 Q1264 61.1119 1264 50.3365 Q1264 39.1965 1269.99 32.6746 Q1276.03 26.1121 1286.24 26.1121 Q1295.39 26.1121 1300.7 32.0264 Q1306.05 37.9003 1306.05 48.0275 M1298.59 45.84 Q1298.51 39.7232 1295.15 36.0774 Q1291.83 32.4315 1286.32 32.4315 Q1280.08 32.4315 1276.31 35.9558 Q1272.59 39.4801 1272.02 45.8805 L1298.59 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1357.09 48.0275 L1357.09 51.6733 L1322.82 51.6733 Q1323.3 59.3701 1327.44 63.421 Q1331.61 67.4314 1339.02 67.4314 Q1343.32 67.4314 1347.33 66.3781 Q1351.38 65.3249 1355.35 63.2184 L1355.35 70.267 Q1351.34 71.9684 1347.12 72.8596 Q1342.91 73.7508 1338.58 73.7508 Q1327.72 73.7508 1321.36 67.4314 Q1315.04 61.1119 1315.04 50.3365 Q1315.04 39.1965 1321.04 32.6746 Q1327.07 26.1121 1337.28 26.1121 Q1346.44 26.1121 1351.74 32.0264 Q1357.09 37.9003 1357.09 48.0275 M1349.64 45.84 Q1349.55 39.7232 1346.19 36.0774 Q1342.87 32.4315 1337.36 32.4315 Q1331.12 32.4315 1327.36 35.9558 Q1323.63 39.4801 1323.06 45.8805 L1349.64 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1407.04 45.1919 L1407.04 72.576 L1399.58 72.576 L1399.58 45.4349 Q1399.58 38.994 1397.07 35.7938 Q1394.56 32.5936 1389.54 32.5936 Q1383.5 32.5936 1380.02 36.4419 Q1376.53 40.2903 1376.53 46.9338 L1376.53 72.576 L1369.04 72.576 L1369.04 27.2059 L1376.53 27.2059 L1376.53 34.2544 Q1379.21 30.163 1382.81 28.1376 Q1386.46 26.1121 1391.2 26.1121 Q1399.02 26.1121 1403.03 30.9732 Q1407.04 35.7938 1407.04 45.1919 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1429.28 14.324 L1429.28 27.2059 L1444.63 27.2059 L1444.63 32.9987 L1429.28 32.9987 L1429.28 57.6282 Q1429.28 63.1779 1430.78 64.7578 Q1432.31 66.3376 1436.97 66.3376 L1444.63 66.3376 L1444.63 72.576 L1436.97 72.576 Q1428.34 72.576 1425.06 69.3758 Q1421.78 66.1351 1421.78 57.6282 L1421.78 32.9987 L1416.31 32.9987 L1416.31 27.2059 L1421.78 27.2059 L1421.78 14.324 L1429.28 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1480.72 34.1734 Q1479.47 33.4443 1477.97 33.1202 Q1476.51 32.7556 1474.73 32.7556 Q1468.41 32.7556 1465.01 36.8875 Q1461.64 40.9789 1461.64 48.6757 L1461.64 72.576 L1454.15 72.576 L1454.15 27.2059 L1461.64 27.2059 L1461.64 34.2544 Q1463.99 30.1225 1467.76 28.1376 Q1471.53 26.1121 1476.91 26.1121 Q1477.68 26.1121 1478.62 26.2337 Q1479.55 26.3147 1480.68 26.5172 L1480.72 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1507.42 76.7889 Q1504.26 84.8907 1501.26 87.3618 Q1498.26 89.8329 1493.24 89.8329 L1487.29 89.8329 L1487.29 83.5945 L1491.66 83.5945 Q1494.74 83.5945 1496.44 82.1361 Q1498.14 80.6778 1500.21 75.2496 L1501.54 71.8468 L1483.19 27.2059 L1491.09 27.2059 L1505.27 62.6918 L1519.45 27.2059 L1527.35 27.2059 L1507.42 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1555.95 12.096 L1607.11 12.096 L1607.11 18.9825 L1585.64 18.9825 L1585.64 72.576 L1577.42 72.576 L1577.42 18.9825 L1555.95 18.9825 L1555.95 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1628.78 34.1734 Q1627.53 33.4443 1626.03 33.1202 Q1624.57 32.7556 1622.79 32.7556 Q1616.47 32.7556 1613.07 36.8875 Q1609.7 40.9789 1609.7 48.6757 L1609.7 72.576 L1602.21 72.576 L1602.21 27.2059 L1609.7 27.2059 L1609.7 34.2544 Q1612.05 30.1225 1615.82 28.1376 Q1619.59 26.1121 1624.98 26.1121 Q1625.75 26.1121 1626.68 26.2337 Q1627.61 26.3147 1628.74 26.5172 L1628.78 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1657.22 49.7694 Q1648.19 49.7694 1644.7 51.8354 Q1641.22 53.9013 1641.22 58.8839 Q1641.22 62.8538 1643.81 65.2034 Q1646.45 67.5124 1650.94 67.5124 Q1657.14 67.5124 1660.87 63.1374 Q1664.63 58.7219 1664.63 51.4303 L1664.63 49.7694 L1657.22 49.7694 M1672.09 46.6907 L1672.09 72.576 L1664.63 72.576 L1664.63 65.6895 Q1662.08 69.8214 1658.27 71.8063 Q1654.47 73.7508 1648.96 73.7508 Q1641.99 73.7508 1637.86 69.8619 Q1633.77 65.9325 1633.77 59.3701 Q1633.77 51.7138 1638.87 47.825 Q1644.01 43.9361 1654.18 43.9361 L1664.63 43.9361 L1664.63 43.2069 Q1664.63 38.0623 1661.23 35.2672 Q1657.87 32.4315 1651.75 32.4315 Q1647.86 32.4315 1644.18 33.3632 Q1640.49 34.295 1637.09 36.1584 L1637.09 29.2718 Q1641.18 27.692 1645.03 26.9223 Q1648.88 26.1121 1652.52 26.1121 Q1662.37 26.1121 1667.23 31.2163 Q1672.09 36.3204 1672.09 46.6907 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1687.44 27.2059 L1694.89 27.2059 L1694.89 73.3862 Q1694.89 82.0551 1691.57 85.944 Q1688.29 89.8329 1680.96 89.8329 L1678.12 89.8329 L1678.12 83.5134 L1680.11 83.5134 Q1684.36 83.5134 1685.9 81.5285 Q1687.44 79.5841 1687.44 73.3862 L1687.44 27.2059 M1687.44 9.54393 L1694.89 9.54393 L1694.89 18.9825 L1687.44 18.9825 L1687.44 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1749.3 48.0275 L1749.3 51.6733 L1715.03 51.6733 Q1715.51 59.3701 1719.64 63.421 Q1723.82 67.4314 1731.23 67.4314 Q1735.52 67.4314 1739.53 66.3781 Q1743.59 65.3249 1747.56 63.2184 L1747.56 70.267 Q1743.55 71.9684 1739.33 72.8596 Q1735.12 73.7508 1730.78 73.7508 Q1719.93 73.7508 1713.57 67.4314 Q1707.25 61.1119 1707.25 50.3365 Q1707.25 39.1965 1713.24 32.6746 Q1719.28 26.1121 1729.49 26.1121 Q1738.64 26.1121 1743.95 32.0264 Q1749.3 37.9003 1749.3 48.0275 M1741.84 45.84 Q1741.76 39.7232 1738.4 36.0774 Q1735.08 32.4315 1729.57 32.4315 Q1723.33 32.4315 1719.56 35.9558 Q1715.84 39.4801 1715.27 45.8805 L1741.84 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1794.18 28.9478 L1794.18 35.9153 Q1791.02 34.1734 1787.82 33.3227 Q1784.66 32.4315 1781.42 32.4315 Q1774.17 32.4315 1770.16 37.0496 Q1766.15 41.6271 1766.15 49.9314 Q1766.15 58.2358 1770.16 62.8538 Q1774.17 67.4314 1781.42 67.4314 Q1784.66 67.4314 1787.82 66.5807 Q1791.02 65.6895 1794.18 63.9476 L1794.18 70.8341 Q1791.06 72.2924 1787.7 73.0216 Q1784.38 73.7508 1780.61 73.7508 Q1770.36 73.7508 1764.33 67.3098 Q1758.29 60.8689 1758.29 49.9314 Q1758.29 38.832 1764.37 32.472 Q1770.48 26.1121 1781.1 26.1121 Q1784.54 26.1121 1787.82 26.8413 Q1791.1 27.5299 1794.18 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1814.52 14.324 L1814.52 27.2059 L1829.87 27.2059 L1829.87 32.9987 L1814.52 32.9987 L1814.52 57.6282 Q1814.52 63.1779 1816.02 64.7578 Q1817.56 66.3376 1822.21 66.3376 L1829.87 66.3376 L1829.87 72.576 L1822.21 72.576 Q1813.59 72.576 1810.3 69.3758 Q1807.02 66.1351 1807.02 57.6282 L1807.02 32.9987 L1801.55 32.9987 L1801.55 27.2059 L1807.02 27.2059 L1807.02 14.324 L1814.52 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1857.25 32.4315 Q1851.26 32.4315 1847.78 37.1306 Q1844.29 41.7891 1844.29 49.9314 Q1844.29 58.0738 1847.73 62.7728 Q1851.22 67.4314 1857.25 67.4314 Q1863.21 67.4314 1866.69 62.7323 Q1870.18 58.0333 1870.18 49.9314 Q1870.18 41.8701 1866.69 37.1711 Q1863.21 32.4315 1857.25 32.4315 M1857.25 26.1121 Q1866.98 26.1121 1872.53 32.4315 Q1878.08 38.7509 1878.08 49.9314 Q1878.08 61.0714 1872.53 67.4314 Q1866.98 73.7508 1857.25 73.7508 Q1847.49 73.7508 1841.94 67.4314 Q1836.43 61.0714 1836.43 49.9314 Q1836.43 38.7509 1841.94 32.4315 Q1847.49 26.1121 1857.25 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1916.72 34.1734 Q1915.47 33.4443 1913.97 33.1202 Q1912.51 32.7556 1910.73 32.7556 Q1904.41 32.7556 1901 36.8875 Q1897.64 40.9789 1897.64 48.6757 L1897.64 72.576 L1890.15 72.576 L1890.15 27.2059 L1897.64 27.2059 L1897.64 34.2544 Q1899.99 30.1225 1903.76 28.1376 Q1907.53 26.1121 1912.91 26.1121 Q1913.68 26.1121 1914.62 26.2337 Q1915.55 26.3147 1916.68 26.5172 L1916.72 34.1734 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip150)" d="M1943.42 76.7889 Q1940.26 84.8907 1937.26 87.3618 Q1934.26 89.8329 1929.24 89.8329 L1923.28 89.8329 L1923.28 83.5945 L1927.66 83.5945 Q1930.74 83.5945 1932.44 82.1361 Q1934.14 80.6778 1936.21 75.2496 L1937.54 71.8468 L1919.19 27.2059 L1927.09 27.2059 L1941.27 62.6918 L1955.45 27.2059 L1963.35 27.2059 L1943.42 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip152)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="802.382,531.57 804.637,538.307 806.893,545.043 809.15,551.773 811.408,558.496 813.666,565.21 815.926,571.911 818.187,578.598 820.449,585.267 822.712,591.915 824.976,598.539 827.242,605.135 829.508,611.698 831.776,618.226 834.046,624.712 836.317,631.152 838.59,637.542 840.864,643.875 843.14,650.146 845.418,656.349 847.698,662.476 849.98,668.521 852.264,674.477 854.551,680.336 856.84,686.09 859.131,691.731 861.425,697.251 863.722,702.639 866.022,707.889 868.324,712.989 870.63,717.931 872.939,722.706 875.252,727.305 877.568,731.717 879.888,735.935 882.211,739.949 884.539,743.752 886.87,747.335 889.206,750.691 891.546,753.815 893.89,756.7 896.238,759.342 898.591,761.737 900.948,763.883 903.31,765.777 905.676,767.421 908.046,768.814 910.421,769.959 912.8,770.859 915.184,771.519 917.571,771.943 919.963,772.14 922.359,772.116 924.759,771.879 927.163,771.441 929.571,770.81 931.983,769.997 934.398,769.014 936.816,767.873 939.238,766.586 941.663,765.166 944.09,763.624 946.521,761.974 948.955,760.229 951.391,758.401 953.83,756.503 956.271,754.547 958.714,752.545 961.159,750.509 963.607,748.451 966.056,746.382 968.508,744.313 970.961,742.254 973.416,740.215 975.872,738.207 978.33,736.238 980.789,734.317 983.25,732.452 985.712,730.652 988.175,728.925 990.639,727.276 993.105,725.714 995.571,724.245 998.039,722.873 1000.51,721.606 1002.98,720.448 1005.45,719.404 1007.92,718.478 1010.39,717.674 1012.86,716.997 1015.34,716.449 1017.81,716.032 1020.29,715.751 1022.76,715.607 1025.24,715.601 1027.72,715.735 1030.2,716.011 1032.68,716.429 1035.16,716.989 1037.64,717.692 1040.12,718.537 1042.6,719.522 1045.08,720.648 1047.56,721.913 1050.05,723.313 1052.53,724.848 1055.02,726.513 1057.51,728.307 1059.99,730.224 1062.48,732.261 1064.97,734.413 1067.46,736.675 1069.95,739.04 1072.44,741.502 1074.93,744.055 1077.42,746.692 1079.92,749.403 1082.41,752.181 1084.91,755.016 1087.4,757.899 1089.9,760.819 1092.4,763.767 1094.9,766.73 1097.4,769.697 1099.9,772.656 1102.4,775.594 1104.9,778.499 1107.4,781.358 1109.91,784.158 1112.42,786.885 1114.92,789.527 1117.43,792.071 1119.94,794.505 1122.45,796.816 1124.96,798.993 1127.47,801.026 1129.98,802.905 1132.5,804.621 1135.01,806.167 1137.53,807.536 1140.04,808.725 1142.56,809.729 1145.07,810.547 1147.59,811.178 1150.11,811.623 1152.63,811.887 1155.15,811.972 1157.67,811.884 1160.19,811.631 1162.71,811.222 1165.22,810.665 1167.74,809.971 1170.26,809.153 1172.78,808.221 1175.3,807.19 1177.82,806.072 1180.34,804.882 1182.86,803.632 1185.38,802.338 1187.9,801.013 1190.42,799.671 1192.93,798.325 1195.45,796.989 1197.97,795.675 1200.48,794.397 1203,793.164 1205.51,791.989 1208.03,790.883 1210.54,789.854 1213.05,788.912 1215.56,788.064 1218.07,787.32 1220.58,786.684 1223.09,786.163 1225.6,785.762 1228.11,785.485 1230.62,785.335 1233.12,785.314 1235.63,785.424 1238.13,785.665 1240.64,786.038 1243.14,786.54 1245.64,787.17 1248.14,787.925 1250.64,788.8 1253.14,789.791 1255.64,790.891 1258.14,792.096 1260.63,793.396 1263.13,794.783 1265.62,796.249 1268.12,797.784 1270.61,799.377 1273.1,801.017 1275.59,802.692 1278.07,804.39 1280.56,806.097 1283.04,807.803 1285.53,809.492 1288.01,811.152 1290.49,812.769 1292.97,814.332 1295.44,815.827 1297.92,817.242 1300.39,818.567 1302.86,819.79 1305.33,820.903 1307.8,821.897 1310.26,822.766 1312.72,823.503 1315.18,824.105 1317.64,824.57 1320.09,824.896 1322.54,825.083 1324.99,825.136 1327.44,825.056 1329.88,824.849 1332.32,824.523 1334.76,824.084 1337.19,823.543 1339.63,822.908 1342.05,822.191 1344.48,821.404 1346.9,820.559 1349.32,819.668 1351.74,818.743 1354.15,817.799 1356.56,816.847 1358.97,815.899 1361.37,814.969 1363.78,814.068 1366.17,813.206 1368.57,812.395 1370.96,811.643 1373.35,810.961 1375.74,810.355 1378.12,809.833 1380.5,809.401 1382.87,809.063 1385.25,808.825 1387.62,808.688 1389.98,808.655 1392.35,808.725 1394.71,808.9 1397.07,809.177 1399.42,809.553 1401.77,810.025 1404.12,810.588 1406.46,811.236 1408.8,811.963 1411.14,812.762 1413.47,813.623 1415.8,814.538 1418.13,815.496 1420.45,816.488 1422.77,817.503 1425.08,818.529 1427.39,819.556 1429.7,820.571 1432,821.565 1434.3,822.525 1436.59,823.441 1438.88,824.304 1441.16,825.103 1443.44,825.832 1445.72,826.481 1447.99,827.045 1450.25,827.52 1452.51,827.9 1454.77,828.185 1457.02,828.372 1459.26,828.462 1461.5,828.457 1463.73,828.361 1465.96,828.177 1468.18,827.911 1470.4,827.57 1472.61,827.161 1474.81,826.693 1477.01,826.175 1479.21,825.617 1481.4,825.029 1483.58,824.42 1485.76,823.802 1487.93,823.184 1490.09,822.577 1492.26,821.99 1494.41,821.432 1496.56,820.913 1498.7,820.439 1500.84,820.018 1502.98,819.656 1505.1,819.359 1507.23,819.132 1509.34,818.977 1511.45,818.897 1513.56,818.894 1515.66,818.967 1517.75,819.117 1519.84,819.341 1521.92,819.637 1524,820 1526.07,820.428 1528.13,820.913 1530.19,821.449 1532.24,822.031 1534.29,822.65 1536.33,823.299 1538.36,823.969 1540.39,824.651 1542.41,825.338 1544.43,826.019 1546.44,826.688 1548.44,827.335 1550.43,827.954 1552.42,828.536 1554.4,829.076 1556.37,829.567 1558.34,830.004 1560.29,830.385 1562.25,830.706 1564.19,830.965 1566.13,831.162 1568.05,831.297 1569.98,831.372 1571.89,831.388 1573.79,831.351 1575.69,831.263 1577.58,831.131 1579.46,830.959 1581.34,830.755 1583.2,830.526 1585.06,830.278 1586.91,830.02 1588.76,829.758 1590.59,829.5 1592.42,829.253 1594.24,829.024 1596.05,828.821 1597.86,828.648 1599.65,828.512 1601.44,828.418 1603.22,828.368 1605,828.368 1606.76,828.42 1608.52,828.524 1610.27,828.684 1612.01,828.897 1613.74,829.164 1615.47,829.484 1617.19,829.853 1618.9,830.269 1620.6,830.728 1622.29,831.227 1623.98,831.759 1625.66,832.321 1627.32,832.906 1628.98,833.509 1630.64,834.123 1632.28,834.744 1633.91,835.364 1635.54,835.979 1637.16,836.582 1638.77,837.17 1640.37,837.737 1641.96,838.279 1643.54,838.794 1645.11,839.278 1646.67,839.73 1648.23,840.148 1649.77,840.532 1651.31,840.883 1652.83,841.202 1654.35,841.489 1655.86,841.749 1657.35,841.984 1658.84,842.198 1660.32,842.394 1661.79,842.578 1663.25,842.753 1664.7,842.926 1666.14,843.101 1667.57,843.282 1669,843.476 1670.41,843.686 1671.81,843.917 1673.21,844.174 1674.59,844.459 1675.97,844.776 1677.33,845.128 1678.69,845.518 1680.03,845.946 1681.37,846.413 1682.7,846.921 1684.02,847.469 1685.33,848.056 1686.63,848.681 1687.92,849.343 1689.2,850.038 1690.47,850.765 1691.73,851.52 1692.98,852.3 1694.22,853.102 1695.46,853.921 1696.68,854.755 1697.89,855.598 1699.09,856.449 1700.29,857.302 1701.47,858.155 1702.64,859.005 1703.8,859.849 1704.96,860.685 1706.1,861.511 1707.23,862.325 1708.36,863.127 1709.47,863.917 1710.57,864.695 1711.66,865.461 1712.74,866.217 1713.82,866.965 1714.88,867.706 1715.93,868.443 1716.97,869.179 1718,869.917 1719.02,870.66 1720.04,871.413 1721.04,872.177 1722.03,872.958 1723.01,873.759 1723.98,874.582 1724.94,875.431 1725.89,876.309 1726.83,877.22 1727.76,878.164 1728.69,879.145 1729.6,880.164 1730.5,881.221 1731.39,882.319 1732.27,883.456 1733.14,884.634 1734.01,885.851 1734.86,887.106 1735.7,888.399 1736.54,889.727 1737.36,891.088 1738.17,892.48 1738.98,893.9 1739.77,895.345 1740.56,896.813 1741.33,898.299 1742.1,899.802 1742.85,901.317 1743.6,902.843 1744.33,904.376 1745.06,905.913 1745.77,907.453 1746.48,908.993 1747.18,910.532 1747.87,912.068 1748.54,913.602 1749.21,915.134 1749.87,916.663 1750.52,918.191 1751.16,919.719 1751.79,921.25 1752.41,922.787 1753.02,924.333 1753.62,925.892 1754.21,927.468 1754.8,929.066 1755.37,930.692 1755.94,932.35 1756.49,934.047 1757.04,935.789 1757.58,937.581 1758.11,939.429 1758.63,941.339 1759.14,943.317 1759.64,945.368 1760.14,947.497 1760.63,949.709 1761.11,952.007 1761.58,954.394 1762.04,956.873 1762.49,959.445 1762.94,962.11 1763.38,964.868 1763.81,967.716 1764.24,970.652 1764.65,973.67 1765.06,976.763 1765.46,979.925 1765.86,983.144 1766.25,986.408 1766.63,989.706 1767,993.02 1767.36,996.334 1767.72,999.629 1768.07,1002.88 1768.42,1006.08 1768.75,1009.19 1769.08,1012.2 1769.41,1015.08 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/d9b9da98.svg b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/e697ac28.svg
similarity index 88%
rename from previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/d9b9da98.svg
rename to previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/e697ac28.svg
index 574462a6264..4f0aca269b6 100644
--- a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/d9b9da98.svg
+++ b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/e697ac28.svg
@@ -1,182 +1,182 @@
 <?xml version="1.0" encoding="utf-8"?>
 <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="700" height="700" viewBox="0 0 2800 2800">
 <defs>
-  <clipPath id="clip210">
+  <clipPath id="clip030">
     <rect x="0" y="0" width="2800" height="2800"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip210)" d="M0 2800 L2800 2800 L2800 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip030)" d="M0 2800 L2800 2800 L2800 0 L0 0  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip211">
+  <clipPath id="clip031">
     <rect x="560" y="280" width="1961" height="1961"/>
   </clipPath>
 </defs>
-<path clip-path="url(#clip210)" d="M159.765 790.501 L1352.76 790.501 L1352.76 153.712 L159.765 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<path clip-path="url(#clip030)" d="M159.765 790.501 L1352.76 790.501 L1352.76 153.712 L159.765 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip212">
+  <clipPath id="clip032">
     <rect x="159" y="153" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,790.501 193.529,153.712 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,790.501 473.774,153.712 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,790.501 754.018,153.712 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,790.501 1034.26,153.712 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,790.501 1314.51,153.712 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,790.501 1352.76,790.501 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,790.501 193.529,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,790.501 473.774,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,790.501 754.018,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,790.501 1034.26,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,790.501 1314.51,771.604 "/>
-<path clip-path="url(#clip210)" d="M193.529 824.62 Q189.918 824.62 188.089 828.185 Q186.283 831.726 186.283 838.856 Q186.283 845.962 188.089 849.527 Q189.918 853.069 193.529 853.069 Q197.163 853.069 198.969 849.527 Q200.797 845.962 200.797 838.856 Q200.797 831.726 198.969 828.185 Q197.163 824.62 193.529 824.62 M193.529 820.916 Q199.339 820.916 202.394 825.523 Q205.473 830.106 205.473 838.856 Q205.473 847.583 202.394 852.189 Q199.339 856.773 193.529 856.773 Q187.719 856.773 184.64 852.189 Q181.584 847.583 181.584 838.856 Q181.584 830.106 184.64 825.523 Q187.719 820.916 193.529 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M433.392 821.541 L451.748 821.541 L451.748 825.476 L437.674 825.476 L437.674 833.949 Q438.693 833.601 439.711 833.439 Q440.73 833.254 441.748 833.254 Q447.535 833.254 450.915 836.425 Q454.295 839.597 454.295 845.013 Q454.295 850.592 450.822 853.694 Q447.35 856.773 441.031 856.773 Q438.855 856.773 436.586 856.402 Q434.341 856.032 431.934 855.291 L431.934 850.592 Q434.017 851.726 436.239 852.282 Q438.461 852.837 440.938 852.837 Q444.943 852.837 447.281 850.731 Q449.619 848.624 449.619 845.013 Q449.619 841.402 447.281 839.296 Q444.943 837.189 440.938 837.189 Q439.063 837.189 437.188 837.606 Q435.336 838.023 433.392 838.902 L433.392 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M473.507 824.62 Q469.896 824.62 468.068 828.185 Q466.262 831.726 466.262 838.856 Q466.262 845.962 468.068 849.527 Q469.896 853.069 473.507 853.069 Q477.142 853.069 478.947 849.527 Q480.776 845.962 480.776 838.856 Q480.776 831.726 478.947 828.185 Q477.142 824.62 473.507 824.62 M473.507 820.916 Q479.318 820.916 482.373 825.523 Q485.452 830.106 485.452 838.856 Q485.452 847.583 482.373 852.189 Q479.318 856.773 473.507 856.773 Q467.697 856.773 464.619 852.189 Q461.563 847.583 461.563 838.856 Q461.563 830.106 464.619 825.523 Q467.697 820.916 473.507 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M503.669 824.62 Q500.058 824.62 498.23 828.185 Q496.424 831.726 496.424 838.856 Q496.424 845.962 498.23 849.527 Q500.058 853.069 503.669 853.069 Q507.304 853.069 509.109 849.527 Q510.938 845.962 510.938 838.856 Q510.938 831.726 509.109 828.185 Q507.304 824.62 503.669 824.62 M503.669 820.916 Q509.479 820.916 512.535 825.523 Q515.614 830.106 515.614 838.856 Q515.614 847.583 512.535 852.189 Q509.479 856.773 503.669 856.773 Q497.859 856.773 494.78 852.189 Q491.725 847.583 491.725 838.856 Q491.725 830.106 494.78 825.523 Q497.859 820.916 503.669 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M698.544 852.166 L706.183 852.166 L706.183 825.8 L697.873 827.467 L697.873 823.208 L706.137 821.541 L710.813 821.541 L710.813 852.166 L718.452 852.166 L718.452 856.101 L698.544 856.101 L698.544 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M737.896 824.62 Q734.285 824.62 732.456 828.185 Q730.651 831.726 730.651 838.856 Q730.651 845.962 732.456 849.527 Q734.285 853.069 737.896 853.069 Q741.53 853.069 743.336 849.527 Q745.164 845.962 745.164 838.856 Q745.164 831.726 743.336 828.185 Q741.53 824.62 737.896 824.62 M737.896 820.916 Q743.706 820.916 746.762 825.523 Q749.84 830.106 749.84 838.856 Q749.84 847.583 746.762 852.189 Q743.706 856.773 737.896 856.773 Q732.086 856.773 729.007 852.189 Q725.951 847.583 725.951 838.856 Q725.951 830.106 729.007 825.523 Q732.086 820.916 737.896 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M768.058 824.62 Q764.447 824.62 762.618 828.185 Q760.812 831.726 760.812 838.856 Q760.812 845.962 762.618 849.527 Q764.447 853.069 768.058 853.069 Q771.692 853.069 773.498 849.527 Q775.326 845.962 775.326 838.856 Q775.326 831.726 773.498 828.185 Q771.692 824.62 768.058 824.62 M768.058 820.916 Q773.868 820.916 776.923 825.523 Q780.002 830.106 780.002 838.856 Q780.002 847.583 776.923 852.189 Q773.868 856.773 768.058 856.773 Q762.248 856.773 759.169 852.189 Q756.113 847.583 756.113 838.856 Q756.113 830.106 759.169 825.523 Q762.248 820.916 768.058 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M798.22 824.62 Q794.609 824.62 792.78 828.185 Q790.974 831.726 790.974 838.856 Q790.974 845.962 792.78 849.527 Q794.609 853.069 798.22 853.069 Q801.854 853.069 803.659 849.527 Q805.488 845.962 805.488 838.856 Q805.488 831.726 803.659 828.185 Q801.854 824.62 798.22 824.62 M798.22 820.916 Q804.03 820.916 807.085 825.523 Q810.164 830.106 810.164 838.856 Q810.164 847.583 807.085 852.189 Q804.03 856.773 798.22 856.773 Q792.409 856.773 789.331 852.189 Q786.275 847.583 786.275 838.856 Q786.275 830.106 789.331 825.523 Q792.409 820.916 798.22 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M978.789 852.166 L986.428 852.166 L986.428 825.8 L978.118 827.467 L978.118 823.208 L986.382 821.541 L991.057 821.541 L991.057 852.166 L998.696 852.166 L998.696 856.101 L978.789 856.101 L978.789 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1008.19 821.541 L1026.54 821.541 L1026.54 825.476 L1012.47 825.476 L1012.47 833.949 Q1013.49 833.601 1014.51 833.439 Q1015.52 833.254 1016.54 833.254 Q1022.33 833.254 1025.71 836.425 Q1029.09 839.597 1029.09 845.013 Q1029.09 850.592 1025.62 853.694 Q1022.15 856.773 1015.83 856.773 Q1013.65 856.773 1011.38 856.402 Q1009.14 856.032 1006.73 855.291 L1006.73 850.592 Q1008.81 851.726 1011.03 852.282 Q1013.26 852.837 1015.73 852.837 Q1019.74 852.837 1022.08 850.731 Q1024.41 848.624 1024.41 845.013 Q1024.41 841.402 1022.08 839.296 Q1019.74 837.189 1015.73 837.189 Q1013.86 837.189 1011.98 837.606 Q1010.13 838.023 1008.19 838.902 L1008.19 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1048.3 824.62 Q1044.69 824.62 1042.86 828.185 Q1041.06 831.726 1041.06 838.856 Q1041.06 845.962 1042.86 849.527 Q1044.69 853.069 1048.3 853.069 Q1051.94 853.069 1053.74 849.527 Q1055.57 845.962 1055.57 838.856 Q1055.57 831.726 1053.74 828.185 Q1051.94 824.62 1048.3 824.62 M1048.3 820.916 Q1054.11 820.916 1057.17 825.523 Q1060.25 830.106 1060.25 838.856 Q1060.25 847.583 1057.17 852.189 Q1054.11 856.773 1048.3 856.773 Q1042.49 856.773 1039.41 852.189 Q1036.36 847.583 1036.36 838.856 Q1036.36 830.106 1039.41 825.523 Q1042.49 820.916 1048.3 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1078.46 824.62 Q1074.85 824.62 1073.02 828.185 Q1071.22 831.726 1071.22 838.856 Q1071.22 845.962 1073.02 849.527 Q1074.85 853.069 1078.46 853.069 Q1082.1 853.069 1083.9 849.527 Q1085.73 845.962 1085.73 838.856 Q1085.73 831.726 1083.9 828.185 Q1082.1 824.62 1078.46 824.62 M1078.46 820.916 Q1084.27 820.916 1087.33 825.523 Q1090.41 830.106 1090.41 838.856 Q1090.41 847.583 1087.33 852.189 Q1084.27 856.773 1078.46 856.773 Q1072.65 856.773 1069.58 852.189 Q1066.52 847.583 1066.52 838.856 Q1066.52 830.106 1069.58 825.523 Q1072.65 820.916 1078.46 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1263.12 852.166 L1279.44 852.166 L1279.44 856.101 L1257.49 856.101 L1257.49 852.166 Q1260.16 849.411 1264.74 844.782 Q1269.35 840.129 1270.53 838.786 Q1272.77 836.263 1273.65 834.527 Q1274.55 832.768 1274.55 831.078 Q1274.55 828.324 1272.61 826.587 Q1270.69 824.851 1267.59 824.851 Q1265.39 824.851 1262.93 825.615 Q1260.5 826.379 1257.73 827.93 L1257.73 823.208 Q1260.55 822.074 1263 821.495 Q1265.46 820.916 1267.49 820.916 Q1272.86 820.916 1276.06 823.601 Q1279.25 826.287 1279.25 830.777 Q1279.25 832.907 1278.44 834.828 Q1277.66 836.726 1275.55 839.319 Q1274.97 839.99 1271.87 843.208 Q1268.77 846.402 1263.12 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1299.25 824.62 Q1295.64 824.62 1293.81 828.185 Q1292.01 831.726 1292.01 838.856 Q1292.01 845.962 1293.81 849.527 Q1295.64 853.069 1299.25 853.069 Q1302.89 853.069 1304.69 849.527 Q1306.52 845.962 1306.52 838.856 Q1306.52 831.726 1304.69 828.185 Q1302.89 824.62 1299.25 824.62 M1299.25 820.916 Q1305.06 820.916 1308.12 825.523 Q1311.2 830.106 1311.2 838.856 Q1311.2 847.583 1308.12 852.189 Q1305.06 856.773 1299.25 856.773 Q1293.44 856.773 1290.36 852.189 Q1287.31 847.583 1287.31 838.856 Q1287.31 830.106 1290.36 825.523 Q1293.44 820.916 1299.25 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1329.42 824.62 Q1325.8 824.62 1323.98 828.185 Q1322.17 831.726 1322.17 838.856 Q1322.17 845.962 1323.98 849.527 Q1325.8 853.069 1329.42 853.069 Q1333.05 853.069 1334.86 849.527 Q1336.68 845.962 1336.68 838.856 Q1336.68 831.726 1334.86 828.185 Q1333.05 824.62 1329.42 824.62 M1329.42 820.916 Q1335.23 820.916 1338.28 825.523 Q1341.36 830.106 1341.36 838.856 Q1341.36 847.583 1338.28 852.189 Q1335.23 856.773 1329.42 856.773 Q1323.61 856.773 1320.53 852.189 Q1317.47 847.583 1317.47 838.856 Q1317.47 830.106 1320.53 825.523 Q1323.61 820.916 1329.42 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1359.58 824.62 Q1355.97 824.62 1354.14 828.185 Q1352.33 831.726 1352.33 838.856 Q1352.33 845.962 1354.14 849.527 Q1355.97 853.069 1359.58 853.069 Q1363.21 853.069 1365.02 849.527 Q1366.85 845.962 1366.85 838.856 Q1366.85 831.726 1365.02 828.185 Q1363.21 824.62 1359.58 824.62 M1359.58 820.916 Q1365.39 820.916 1368.44 825.523 Q1371.52 830.106 1371.52 838.856 Q1371.52 847.583 1368.44 852.189 Q1365.39 856.773 1359.58 856.773 Q1353.77 856.773 1350.69 852.189 Q1347.63 847.583 1347.63 838.856 Q1347.63 830.106 1350.69 825.523 Q1353.77 820.916 1359.58 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,705.729 1352.76,705.729 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,538.856 1352.76,538.856 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,371.983 1352.76,371.983 "/>
-<polyline clip-path="url(#clip212)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,205.109 1352.76,205.109 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,790.501 159.765,153.712 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,705.729 178.663,705.729 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,538.856 178.663,538.856 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,371.983 178.663,371.983 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,205.109 178.663,205.109 "/>
-<path clip-path="url(#clip210)" d="M51.3995 719.074 L59.0384 719.074 L59.0384 692.709 L50.7282 694.375 L50.7282 690.116 L58.9921 688.449 L63.668 688.449 L63.668 719.074 L71.3068 719.074 L71.3068 723.009 L51.3995 723.009 L51.3995 719.074 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M80.7512 717.13 L85.6354 717.13 L85.6354 723.009 L80.7512 723.009 L80.7512 717.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 691.528 Q102.209 691.528 100.381 695.093 Q98.5752 698.635 98.5752 705.764 Q98.5752 712.871 100.381 716.435 Q102.209 719.977 105.821 719.977 Q109.455 719.977 111.26 716.435 Q113.089 712.871 113.089 705.764 Q113.089 698.635 111.26 695.093 Q109.455 691.528 105.821 691.528 M105.821 687.824 Q111.631 687.824 114.686 692.431 Q117.765 697.014 117.765 705.764 Q117.765 714.491 114.686 719.097 Q111.631 723.681 105.821 723.681 Q100.01 723.681 96.9317 719.097 Q93.8761 714.491 93.8761 705.764 Q93.8761 697.014 96.9317 692.431 Q100.01 687.824 105.821 687.824 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M52.3949 552.201 L60.0337 552.201 L60.0337 525.835 L51.7236 527.502 L51.7236 523.243 L59.9874 521.576 L64.6633 521.576 L64.6633 552.201 L72.3022 552.201 L72.3022 556.136 L52.3949 556.136 L52.3949 552.201 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M81.7466 550.256 L86.6308 550.256 L86.6308 556.136 L81.7466 556.136 L81.7466 550.256 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M96.8622 521.576 L115.219 521.576 L115.219 525.511 L101.145 525.511 L101.145 533.983 Q102.163 533.636 103.182 533.474 Q104.2 533.289 105.219 533.289 Q111.006 533.289 114.385 536.46 Q117.765 539.632 117.765 545.048 Q117.765 550.627 114.293 553.729 Q110.82 556.807 104.501 556.807 Q102.325 556.807 100.057 556.437 Q97.8113 556.067 95.4039 555.326 L95.4039 550.627 Q97.4872 551.761 99.7094 552.317 Q101.932 552.872 104.408 552.872 Q108.413 552.872 110.751 550.766 Q113.089 548.659 113.089 545.048 Q113.089 541.437 110.751 539.331 Q108.413 537.224 104.408 537.224 Q102.534 537.224 100.659 537.641 Q98.8067 538.057 96.8622 538.937 L96.8622 521.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M54.6171 385.327 L70.9365 385.327 L70.9365 389.263 L48.9921 389.263 L48.9921 385.327 Q51.6541 382.573 56.2375 377.943 Q60.8439 373.29 62.0245 371.948 Q64.2698 369.425 65.1494 367.689 Q66.0522 365.929 66.0522 364.24 Q66.0522 361.485 64.1078 359.749 Q62.1865 358.013 59.0847 358.013 Q56.8856 358.013 54.4319 358.777 Q52.0014 359.541 49.2236 361.091 L49.2236 356.369 Q52.0477 355.235 54.5014 354.656 Q56.955 354.078 58.9921 354.078 Q64.3624 354.078 67.5568 356.763 Q70.7513 359.448 70.7513 363.939 Q70.7513 366.068 69.9411 367.99 Q69.1541 369.888 67.0476 372.48 Q66.4689 373.152 63.367 376.369 Q60.2652 379.564 54.6171 385.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M80.7512 383.383 L85.6354 383.383 L85.6354 389.263 L80.7512 389.263 L80.7512 383.383 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 357.781 Q102.209 357.781 100.381 361.346 Q98.5752 364.888 98.5752 372.017 Q98.5752 379.124 100.381 382.689 Q102.209 386.23 105.821 386.23 Q109.455 386.23 111.26 382.689 Q113.089 379.124 113.089 372.017 Q113.089 364.888 111.26 361.346 Q109.455 357.781 105.821 357.781 M105.821 354.078 Q111.631 354.078 114.686 358.684 Q117.765 363.267 117.765 372.017 Q117.765 380.744 114.686 385.351 Q111.631 389.934 105.821 389.934 Q100.01 389.934 96.9317 385.351 Q93.8761 380.744 93.8761 372.017 Q93.8761 363.267 96.9317 358.684 Q100.01 354.078 105.821 354.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M55.6125 218.454 L71.9318 218.454 L71.9318 222.389 L49.9875 222.389 L49.9875 218.454 Q52.6495 215.699 57.2328 211.07 Q61.8393 206.417 63.0198 205.074 Q65.2652 202.551 66.1448 200.815 Q67.0476 199.056 67.0476 197.366 Q67.0476 194.612 65.1032 192.875 Q63.1819 191.139 60.08 191.139 Q57.881 191.139 55.4273 191.903 Q52.9967 192.667 50.219 194.218 L50.219 189.496 Q53.043 188.362 55.4967 187.783 Q57.9504 187.204 59.9874 187.204 Q65.3578 187.204 68.5522 189.889 Q71.7466 192.574 71.7466 197.065 Q71.7466 199.195 70.9365 201.116 Q70.1494 203.014 68.0429 205.607 Q67.4642 206.278 64.3624 209.496 Q61.2606 212.69 55.6125 218.454 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M81.7466 216.51 L86.6308 216.51 L86.6308 222.389 L81.7466 222.389 L81.7466 216.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M96.8622 187.829 L115.219 187.829 L115.219 191.764 L101.145 191.764 L101.145 200.236 Q102.163 199.889 103.182 199.727 Q104.2 199.542 105.219 199.542 Q111.006 199.542 114.385 202.713 Q117.765 205.885 117.765 211.301 Q117.765 216.88 114.293 219.982 Q110.82 223.06 104.501 223.06 Q102.325 223.06 100.057 222.69 Q97.8113 222.32 95.4039 221.579 L95.4039 216.88 Q97.4872 218.014 99.7094 218.57 Q101.932 219.125 104.408 219.125 Q108.413 219.125 110.751 217.019 Q113.089 214.912 113.089 211.301 Q113.089 207.69 110.751 205.584 Q108.413 203.477 104.408 203.477 Q102.534 203.477 100.659 203.894 Q98.8067 204.311 96.8622 205.19 L96.8622 187.829 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M365.348 20.1573 L354.249 50.2555 L376.488 50.2555 L365.348 20.1573 M360.73 12.096 L370.007 12.096 L393.056 72.576 L384.549 72.576 L379.04 57.061 L351.778 57.061 L346.268 72.576 L337.64 72.576 L360.73 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M401.563 9.54393 L409.017 9.54393 L409.017 72.576 L401.563 72.576 L401.563 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M431.986 14.324 L431.986 27.2059 L447.338 27.2059 L447.338 32.9987 L431.986 32.9987 L431.986 57.6282 Q431.986 63.1779 433.484 64.7578 Q435.024 66.3376 439.682 66.3376 L447.338 66.3376 L447.338 72.576 L439.682 72.576 Q431.054 72.576 427.773 69.3758 Q424.491 66.1351 424.491 57.6282 L424.491 32.9987 L419.023 32.9987 L419.023 27.2059 L424.491 27.2059 L424.491 14.324 L431.986 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M457.142 27.2059 L464.595 27.2059 L464.595 72.576 L457.142 72.576 L457.142 27.2059 M457.142 9.54393 L464.595 9.54393 L464.595 18.9825 L457.142 18.9825 L457.142 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M487.564 14.324 L487.564 27.2059 L502.917 27.2059 L502.917 32.9987 L487.564 32.9987 L487.564 57.6282 Q487.564 63.1779 489.063 64.7578 Q490.602 66.3376 495.261 66.3376 L502.917 66.3376 L502.917 72.576 L495.261 72.576 Q486.632 72.576 483.351 69.3758 Q480.07 66.1351 480.07 57.6282 L480.07 32.9987 L474.601 32.9987 L474.601 27.2059 L480.07 27.2059 L480.07 14.324 L487.564 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M511.95 54.671 L511.95 27.2059 L519.404 27.2059 L519.404 54.3874 Q519.404 60.8284 521.916 64.0691 Q524.427 67.2693 529.45 67.2693 Q535.486 67.2693 538.97 63.421 Q542.494 59.5726 542.494 52.9291 L542.494 27.2059 L549.948 27.2059 L549.948 72.576 L542.494 72.576 L542.494 65.6084 Q539.78 69.7404 536.175 71.7658 Q532.61 73.7508 527.87 73.7508 Q520.052 73.7508 516.001 68.8897 Q511.95 64.0286 511.95 54.671 M530.706 26.1121 L530.706 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M595.156 34.0924 L595.156 9.54393 L602.61 9.54393 L602.61 72.576 L595.156 72.576 L595.156 65.7705 Q592.806 69.8214 589.201 71.8063 Q585.636 73.7508 580.613 73.7508 Q572.39 73.7508 567.205 67.1883 Q562.06 60.6258 562.06 49.9314 Q562.06 39.2371 567.205 32.6746 Q572.39 26.1121 580.613 26.1121 Q585.636 26.1121 589.201 28.0971 Q592.806 30.0415 595.156 34.0924 M569.757 49.9314 Q569.757 58.1548 573.119 62.8538 Q576.522 67.5124 582.436 67.5124 Q588.35 67.5124 591.753 62.8538 Q595.156 58.1548 595.156 49.9314 Q595.156 41.7081 591.753 37.0496 Q588.35 32.3505 582.436 32.3505 Q576.522 32.3505 573.119 37.0496 Q569.757 41.7081 569.757 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M656.77 48.0275 L656.77 51.6733 L622.5 51.6733 Q622.986 59.3701 627.118 63.421 Q631.29 67.4314 638.703 67.4314 Q642.997 67.4314 647.008 66.3781 Q651.058 65.3249 655.028 63.2184 L655.028 70.267 Q651.018 71.9684 646.805 72.8596 Q642.592 73.7508 638.258 73.7508 Q627.401 73.7508 621.041 67.4314 Q614.722 61.1119 614.722 50.3365 Q614.722 39.1965 620.717 32.6746 Q626.753 26.1121 636.961 26.1121 Q646.116 26.1121 651.423 32.0264 Q656.77 37.9003 656.77 48.0275 M649.317 45.84 Q649.236 39.7232 645.873 36.0774 Q642.552 32.4315 637.042 32.4315 Q630.804 32.4315 627.037 35.9558 Q623.31 39.4801 622.743 45.8805 L649.317 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M713.28 9.62495 Q707.852 18.942 705.219 28.0566 Q702.586 37.1711 702.586 46.5287 Q702.586 55.8863 705.219 65.0818 Q707.893 74.2369 713.28 83.5134 L706.799 83.5134 Q700.723 73.9938 697.684 64.7983 Q694.687 55.6027 694.687 46.5287 Q694.687 37.4952 697.684 28.3401 Q700.682 19.1851 706.799 9.62495 L713.28 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M730.213 65.6895 L743.581 65.6895 L743.581 19.5497 L729.038 22.4663 L729.038 15.0127 L743.5 12.096 L751.683 12.096 L751.683 65.6895 L765.051 65.6895 L765.051 72.576 L730.213 72.576 L730.213 65.6895 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M799.078 17.4837 Q792.759 17.4837 789.559 23.7221 Q786.399 29.92 786.399 42.3968 Q786.399 54.833 789.559 61.0714 Q792.759 67.2693 799.078 67.2693 Q805.438 67.2693 808.598 61.0714 Q811.798 54.833 811.798 42.3968 Q811.798 29.92 808.598 23.7221 Q805.438 17.4837 799.078 17.4837 M799.078 11.0023 Q809.246 11.0023 814.593 19.0636 Q819.981 27.0843 819.981 42.3968 Q819.981 57.6687 814.593 65.73 Q809.246 73.7508 799.078 73.7508 Q788.911 73.7508 783.523 65.73 Q778.176 57.6687 778.176 42.3968 Q778.176 27.0843 783.523 19.0636 Q788.911 11.0023 799.078 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M851.862 17.4837 Q845.542 17.4837 842.342 23.7221 Q839.182 29.92 839.182 42.3968 Q839.182 54.833 842.342 61.0714 Q845.542 67.2693 851.862 67.2693 Q858.222 67.2693 861.381 61.0714 Q864.582 54.833 864.582 42.3968 Q864.582 29.92 861.381 23.7221 Q858.222 17.4837 851.862 17.4837 M851.862 11.0023 Q862.03 11.0023 867.377 19.0636 Q872.764 27.0843 872.764 42.3968 Q872.764 57.6687 867.377 65.73 Q862.03 73.7508 851.862 73.7508 Q841.694 73.7508 836.306 65.73 Q830.959 57.6687 830.959 42.3968 Q830.959 27.0843 836.306 19.0636 Q841.694 11.0023 851.862 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M887.996 62.2867 L896.543 62.2867 L896.543 69.2543 L889.9 82.2172 L884.674 82.2172 L887.996 69.2543 L887.996 62.2867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M931.016 17.4837 Q924.697 17.4837 921.497 23.7221 Q918.337 29.92 918.337 42.3968 Q918.337 54.833 921.497 61.0714 Q924.697 67.2693 931.016 67.2693 Q937.376 67.2693 940.536 61.0714 Q943.736 54.833 943.736 42.3968 Q943.736 29.92 940.536 23.7221 Q937.376 17.4837 931.016 17.4837 M931.016 11.0023 Q941.184 11.0023 946.531 19.0636 Q951.919 27.0843 951.919 42.3968 Q951.919 57.6687 946.531 65.73 Q941.184 73.7508 931.016 73.7508 Q920.849 73.7508 915.461 65.73 Q910.114 57.6687 910.114 42.3968 Q910.114 27.0843 915.461 19.0636 Q920.849 11.0023 931.016 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M983.8 17.4837 Q977.48 17.4837 974.28 23.7221 Q971.12 29.92 971.12 42.3968 Q971.12 54.833 974.28 61.0714 Q977.48 67.2693 983.8 67.2693 Q990.16 67.2693 993.319 61.0714 Q996.52 54.833 996.52 42.3968 Q996.52 29.92 993.319 23.7221 Q990.16 17.4837 983.8 17.4837 M983.8 11.0023 Q993.967 11.0023 999.315 19.0636 Q1004.7 27.0843 1004.7 42.3968 Q1004.7 57.6687 999.315 65.73 Q993.967 73.7508 983.8 73.7508 Q973.632 73.7508 968.244 65.73 Q962.897 57.6687 962.897 42.3968 Q962.897 27.0843 968.244 19.0636 Q973.632 11.0023 983.8 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1036.58 17.4837 Q1030.26 17.4837 1027.06 23.7221 Q1023.9 29.92 1023.9 42.3968 Q1023.9 54.833 1027.06 61.0714 Q1030.26 67.2693 1036.58 67.2693 Q1042.94 67.2693 1046.1 61.0714 Q1049.3 54.833 1049.3 42.3968 Q1049.3 29.92 1046.1 23.7221 Q1042.94 17.4837 1036.58 17.4837 M1036.58 11.0023 Q1046.75 11.0023 1052.1 19.0636 Q1057.49 27.0843 1057.49 42.3968 Q1057.49 57.6687 1052.1 65.73 Q1046.75 73.7508 1036.58 73.7508 Q1026.42 73.7508 1021.03 65.73 Q1015.68 57.6687 1015.68 42.3968 Q1015.68 27.0843 1021.03 19.0636 Q1026.42 11.0023 1036.58 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1120.15 9.54393 L1120.15 15.7418 L1113.02 15.7418 Q1109.01 15.7418 1107.43 17.3622 Q1105.89 18.9825 1105.89 23.1955 L1105.89 27.2059 L1118.17 27.2059 L1118.17 32.9987 L1105.89 32.9987 L1105.89 72.576 L1098.4 72.576 L1098.4 32.9987 L1091.27 32.9987 L1091.27 27.2059 L1098.4 27.2059 L1098.4 24.0462 Q1098.4 16.471 1101.92 13.0277 Q1105.45 9.54393 1113.1 9.54393 L1120.15 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1132.31 14.324 L1132.31 27.2059 L1147.66 27.2059 L1147.66 32.9987 L1132.31 32.9987 L1132.31 57.6282 Q1132.31 63.1779 1133.8 64.7578 Q1135.34 66.3376 1140 66.3376 L1147.66 66.3376 L1147.66 72.576 L1140 72.576 Q1131.37 72.576 1128.09 69.3758 Q1124.81 66.1351 1124.81 57.6282 L1124.81 32.9987 L1119.34 32.9987 L1119.34 27.2059 L1124.81 27.2059 L1124.81 14.324 L1132.31 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1156.29 9.62495 L1162.77 9.62495 Q1168.85 19.1851 1171.84 28.3401 Q1174.88 37.4952 1174.88 46.5287 Q1174.88 55.6027 1171.84 64.7983 Q1168.85 73.9938 1162.77 83.5134 L1156.29 83.5134 Q1161.67 74.2369 1164.31 65.0818 Q1166.98 55.8863 1166.98 46.5287 Q1166.98 37.1711 1164.31 28.0566 Q1161.67 18.942 1156.29 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip212)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,171.734 195.771,177.699 198.013,183.661 200.255,189.62 202.497,195.572 204.739,201.516 206.981,207.449 209.223,213.368 211.464,219.271 213.706,225.155 215.948,231.016 218.19,236.851 220.432,242.656 222.674,248.427 224.916,254.159 227.158,259.849 229.4,265.491 231.642,271.08 233.884,276.61 236.126,282.075 238.368,287.47 240.61,292.787 242.852,298.02 245.094,303.161 247.336,308.203 249.578,313.137 251.82,317.957 254.062,322.653 256.304,327.217 258.546,331.64 260.788,335.913 263.03,340.027 265.271,343.974 267.513,347.745 269.755,351.332 271.997,354.725 274.239,357.918 276.481,360.904 278.723,363.674 280.965,366.224 283.207,368.548 285.449,370.642 287.691,372.502 289.933,374.127 292.175,375.514 294.417,376.663 296.659,377.577 298.901,378.256 301.143,378.704 303.385,378.925 305.627,378.925 307.869,378.71 310.111,378.289 312.353,377.668 314.595,376.858 316.837,375.867 319.078,374.708 321.32,373.39 323.562,371.925 325.804,370.325 328.046,368.602 330.288,366.768 332.53,364.835 334.772,362.816 337.014,360.723 339.256,358.567 341.498,356.361 343.74,354.117 345.982,351.846 348.224,349.558 350.466,347.266 352.708,344.979 354.95,342.707 357.192,340.46 359.434,338.248 361.676,336.078 363.918,333.961 366.16,331.903 368.402,329.913 370.644,327.998 372.885,326.165 375.127,324.42 377.369,322.769 379.611,321.219 381.853,319.774 384.095,318.439 386.337,317.219 388.579,316.119 390.821,315.142 393.063,314.291 395.305,313.57 397.547,312.981 399.789,312.528 402.031,312.212 404.273,312.034 406.515,311.997 408.757,312.101 410.999,312.347 413.241,312.736 415.483,313.267 417.725,313.94 419.967,314.754 422.209,315.708 424.451,316.8 426.692,318.029 428.934,319.392 431.176,320.886 433.418,322.509 435.66,324.256 437.902,326.123 440.144,328.106 442.386,330.2 444.628,332.398 446.87,334.695 449.112,337.084 451.354,339.558 453.596,342.109 455.838,344.729 458.08,347.41 460.322,350.14 462.564,352.912 464.806,355.714 467.048,358.536 469.29,361.366 471.532,364.193 473.774,367.004 476.016,369.788 478.258,372.532 480.5,375.222 482.741,377.848 484.983,380.395 487.225,382.851 489.467,385.206 491.709,387.446 493.951,389.561 496.193,391.541 498.435,393.377 500.677,395.06 502.919,396.582 505.161,397.939 507.403,399.125 509.645,400.137 511.887,400.973 514.129,401.633 516.371,402.119 518.613,402.433 520.855,402.579 523.097,402.562 525.339,402.391 527.581,402.073 529.823,401.617 532.065,401.033 534.307,400.334 536.548,399.53 538.79,398.634 541.032,397.66 543.274,396.621 545.516,395.529 547.758,394.399 550,393.244 552.242,392.078 554.484,390.914 556.726,389.764 558.968,388.641 561.21,387.557 563.452,386.523 565.694,385.55 567.936,384.648 570.178,383.826 572.42,383.094 574.662,382.459 576.904,381.928 579.146,381.508 581.388,381.204 583.63,381.022 585.872,380.964 588.114,381.035 590.355,381.236 592.597,381.569 594.839,382.034 597.081,382.632 599.323,383.36 601.565,384.216 603.807,385.198 606.049,386.302 608.291,387.523 610.533,388.855 612.775,390.292 615.017,391.827 617.259,393.451 619.501,395.156 621.743,396.932 623.985,398.769 626.227,400.655 628.469,402.58 630.711,404.531 632.953,406.497 635.195,408.464 637.437,410.419 639.679,412.351 641.921,414.245 644.162,416.09 646.404,417.873 648.646,419.583 650.888,421.208 653.13,422.739 655.372,424.167 657.614,425.483 659.856,426.681 662.098,427.756 664.34,428.703 666.582,429.52 668.824,430.205 671.066,430.761 673.308,431.189 675.55,431.492 677.792,431.676 680.034,431.747 682.276,431.713 684.518,431.582 686.76,431.365 689.002,431.072 691.244,430.714 693.486,430.304 695.728,429.853 697.969,429.373 700.211,428.878 702.453,428.379 704.695,427.888 706.937,427.418 709.179,426.98 711.421,426.585 713.663,426.242 715.905,425.961 718.147,425.751 720.389,425.619 722.631,425.573 724.873,425.618 727.115,425.759 729.357,425.999 731.599,426.343 733.841,426.79 736.083,427.343 738.325,428 740.567,428.76 742.809,429.62 745.051,430.577 747.293,431.626 749.535,432.762 751.776,433.977 754.018,435.265 756.26,436.617 758.502,438.025 760.744,439.479 762.986,440.968 765.228,442.482 767.47,444.01 769.712,445.542 771.954,447.066 774.196,448.571 776.438,450.046 778.68,451.482 780.922,452.868 783.164,454.195 785.406,455.455 787.648,456.641 789.89,457.747 792.132,458.767 794.374,459.699 796.616,460.539 798.858,461.286 801.1,461.942 803.342,462.508 805.583,462.986 807.825,463.382 810.067,463.7 812.309,463.947 814.551,464.131 816.793,464.259 819.035,464.341 821.277,464.385 823.519,464.403 825.761,464.403 828.003,464.395 830.245,464.391 832.487,464.398 834.729,464.428 836.971,464.489 839.213,464.588 841.455,464.735 843.697,464.935 845.939,465.195 848.181,465.521 850.423,465.916 852.665,466.384 854.907,466.927 857.149,467.547 859.391,468.243 861.632,469.016 863.874,469.863 866.116,470.782 868.358,471.769 870.6,472.819 872.842,473.928 875.084,475.089 877.326,476.295 879.568,477.539 881.81,478.814 884.052,480.11 886.294,481.421 888.536,482.736 890.778,484.048 893.02,485.349 895.262,486.63 897.504,487.884 899.746,489.104 901.988,490.282 904.23,491.415 906.472,492.497 908.714,493.523 910.956,494.492 913.198,495.402 915.439,496.252 917.681,497.041 919.923,497.773 922.165,498.449 924.407,499.072 926.649,499.647 928.891,500.179 931.133,500.673 933.375,501.137 935.617,501.575 937.859,501.997 940.101,502.408 942.343,502.816 944.585,503.229 946.827,503.653 949.069,504.096 951.311,504.563 953.553,505.061 955.795,505.594 958.037,506.169 960.279,506.788 962.521,507.455 964.763,508.173 967.005,508.943 969.246,509.767 971.488,510.643 973.73,511.572 975.972,512.552 978.214,513.581 980.456,514.656 982.698,515.773 984.94,516.928 987.182,518.116 989.424,519.332 991.666,520.57 993.908,521.826 996.15,523.093 998.392,524.365 1000.63,525.636 1002.88,526.902 1005.12,528.156 1007.36,529.393 1009.6,530.61 1011.84,531.802 1014.09,532.966 1016.33,534.099 1018.57,535.2 1020.81,536.267 1023.05,537.3 1025.3,538.299 1027.54,539.265 1029.78,540.2 1032.02,541.107 1034.26,541.988 1036.51,542.848 1038.75,543.689 1040.99,544.516 1043.23,545.335 1045.47,546.149 1047.72,546.963 1049.96,547.783 1052.2,548.614 1054.44,549.459 1056.68,550.323 1058.92,551.21 1061.17,552.125 1063.41,553.068 1065.65,554.045 1067.89,555.056 1070.13,556.103 1072.38,557.188 1074.62,558.31 1076.86,559.469 1079.1,560.666 1081.34,561.897 1083.59,563.162 1085.83,564.459 1088.07,565.784 1090.31,567.135 1092.55,568.508 1094.8,569.9 1097.04,571.308 1099.28,572.727 1101.52,574.154 1103.76,575.585 1106.01,577.016 1108.25,578.445 1110.49,579.869 1112.73,581.284 1114.97,582.688 1117.22,584.08 1119.46,585.459 1121.7,586.823 1123.94,588.172 1126.18,589.506 1128.43,590.827 1130.67,592.134 1132.91,593.43 1135.15,594.718 1137.39,595.998 1139.64,597.274 1141.88,598.55 1144.12,599.827 1146.36,601.111 1148.6,602.403 1150.85,603.708 1153.09,605.029 1155.33,606.369 1157.57,607.731 1159.81,609.119 1162.05,610.535 1164.3,611.981 1166.54,613.459 1168.78,614.971 1171.02,616.518 1173.26,618.1 1175.51,619.719 1177.75,621.373 1179.99,623.062 1182.23,624.785 1184.47,626.541 1186.72,628.329 1188.96,630.145 1191.2,631.987 1193.44,633.854 1195.68,635.741 1197.93,637.646 1200.17,639.567 1202.41,641.499 1204.65,643.44 1206.89,645.387 1209.14,647.337 1211.38,649.287 1213.62,651.236 1215.86,653.181 1218.1,655.121 1220.35,657.054 1222.59,658.981 1224.83,660.9 1227.07,662.814 1229.31,664.721 1231.56,666.626 1233.8,668.528 1236.04,670.432 1238.28,672.34 1240.52,674.256 1242.77,676.185 1245.01,678.131 1247.25,680.1 1249.49,682.096 1251.73,684.126 1253.98,686.194 1256.22,688.308 1258.46,690.472 1260.7,692.693 1262.94,694.976 1265.19,697.326 1267.43,699.748 1269.67,702.247 1271.91,704.825 1274.15,707.487 1276.39,710.234 1278.64,713.068 1280.88,715.989 1283.12,718.997 1285.36,722.089 1287.6,725.262 1289.85,728.511 1292.09,731.831 1294.33,735.212 1296.57,738.645 1298.81,742.119 1301.06,745.62 1303.3,749.133 1305.54,752.642 1307.78,756.127 1310.02,759.568 1312.27,762.946 1314.51,766.238 1316.75,769.422 1318.99,772.479 "/>
-<path clip-path="url(#clip210)" d="M1559.76 790.501 L2752.76 790.501 L2752.76 153.712 L1559.76 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,790.501 193.529,153.712 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,790.501 473.774,153.712 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,790.501 754.018,153.712 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,790.501 1034.26,153.712 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,790.501 1314.51,153.712 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,790.501 1352.76,790.501 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,790.501 193.529,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,790.501 473.774,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,790.501 754.018,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,790.501 1034.26,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,790.501 1314.51,771.604 "/>
+<path clip-path="url(#clip030)" d="M193.529 824.62 Q189.918 824.62 188.089 828.185 Q186.283 831.726 186.283 838.856 Q186.283 845.962 188.089 849.527 Q189.918 853.069 193.529 853.069 Q197.163 853.069 198.969 849.527 Q200.797 845.962 200.797 838.856 Q200.797 831.726 198.969 828.185 Q197.163 824.62 193.529 824.62 M193.529 820.916 Q199.339 820.916 202.394 825.523 Q205.473 830.106 205.473 838.856 Q205.473 847.583 202.394 852.189 Q199.339 856.773 193.529 856.773 Q187.719 856.773 184.64 852.189 Q181.584 847.583 181.584 838.856 Q181.584 830.106 184.64 825.523 Q187.719 820.916 193.529 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M433.392 821.541 L451.748 821.541 L451.748 825.476 L437.674 825.476 L437.674 833.949 Q438.693 833.601 439.711 833.439 Q440.73 833.254 441.748 833.254 Q447.535 833.254 450.915 836.425 Q454.295 839.597 454.295 845.013 Q454.295 850.592 450.822 853.694 Q447.35 856.773 441.031 856.773 Q438.855 856.773 436.586 856.402 Q434.341 856.032 431.934 855.291 L431.934 850.592 Q434.017 851.726 436.239 852.282 Q438.461 852.837 440.938 852.837 Q444.943 852.837 447.281 850.731 Q449.619 848.624 449.619 845.013 Q449.619 841.402 447.281 839.296 Q444.943 837.189 440.938 837.189 Q439.063 837.189 437.188 837.606 Q435.336 838.023 433.392 838.902 L433.392 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M473.507 824.62 Q469.896 824.62 468.068 828.185 Q466.262 831.726 466.262 838.856 Q466.262 845.962 468.068 849.527 Q469.896 853.069 473.507 853.069 Q477.142 853.069 478.947 849.527 Q480.776 845.962 480.776 838.856 Q480.776 831.726 478.947 828.185 Q477.142 824.62 473.507 824.62 M473.507 820.916 Q479.318 820.916 482.373 825.523 Q485.452 830.106 485.452 838.856 Q485.452 847.583 482.373 852.189 Q479.318 856.773 473.507 856.773 Q467.697 856.773 464.619 852.189 Q461.563 847.583 461.563 838.856 Q461.563 830.106 464.619 825.523 Q467.697 820.916 473.507 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M503.669 824.62 Q500.058 824.62 498.23 828.185 Q496.424 831.726 496.424 838.856 Q496.424 845.962 498.23 849.527 Q500.058 853.069 503.669 853.069 Q507.304 853.069 509.109 849.527 Q510.938 845.962 510.938 838.856 Q510.938 831.726 509.109 828.185 Q507.304 824.62 503.669 824.62 M503.669 820.916 Q509.479 820.916 512.535 825.523 Q515.614 830.106 515.614 838.856 Q515.614 847.583 512.535 852.189 Q509.479 856.773 503.669 856.773 Q497.859 856.773 494.78 852.189 Q491.725 847.583 491.725 838.856 Q491.725 830.106 494.78 825.523 Q497.859 820.916 503.669 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M698.544 852.166 L706.183 852.166 L706.183 825.8 L697.873 827.467 L697.873 823.208 L706.137 821.541 L710.813 821.541 L710.813 852.166 L718.452 852.166 L718.452 856.101 L698.544 856.101 L698.544 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M737.896 824.62 Q734.285 824.62 732.456 828.185 Q730.651 831.726 730.651 838.856 Q730.651 845.962 732.456 849.527 Q734.285 853.069 737.896 853.069 Q741.53 853.069 743.336 849.527 Q745.164 845.962 745.164 838.856 Q745.164 831.726 743.336 828.185 Q741.53 824.62 737.896 824.62 M737.896 820.916 Q743.706 820.916 746.762 825.523 Q749.84 830.106 749.84 838.856 Q749.84 847.583 746.762 852.189 Q743.706 856.773 737.896 856.773 Q732.086 856.773 729.007 852.189 Q725.951 847.583 725.951 838.856 Q725.951 830.106 729.007 825.523 Q732.086 820.916 737.896 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M768.058 824.62 Q764.447 824.62 762.618 828.185 Q760.812 831.726 760.812 838.856 Q760.812 845.962 762.618 849.527 Q764.447 853.069 768.058 853.069 Q771.692 853.069 773.498 849.527 Q775.326 845.962 775.326 838.856 Q775.326 831.726 773.498 828.185 Q771.692 824.62 768.058 824.62 M768.058 820.916 Q773.868 820.916 776.923 825.523 Q780.002 830.106 780.002 838.856 Q780.002 847.583 776.923 852.189 Q773.868 856.773 768.058 856.773 Q762.248 856.773 759.169 852.189 Q756.113 847.583 756.113 838.856 Q756.113 830.106 759.169 825.523 Q762.248 820.916 768.058 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M798.22 824.62 Q794.609 824.62 792.78 828.185 Q790.974 831.726 790.974 838.856 Q790.974 845.962 792.78 849.527 Q794.609 853.069 798.22 853.069 Q801.854 853.069 803.659 849.527 Q805.488 845.962 805.488 838.856 Q805.488 831.726 803.659 828.185 Q801.854 824.62 798.22 824.62 M798.22 820.916 Q804.03 820.916 807.085 825.523 Q810.164 830.106 810.164 838.856 Q810.164 847.583 807.085 852.189 Q804.03 856.773 798.22 856.773 Q792.409 856.773 789.331 852.189 Q786.275 847.583 786.275 838.856 Q786.275 830.106 789.331 825.523 Q792.409 820.916 798.22 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M978.789 852.166 L986.428 852.166 L986.428 825.8 L978.118 827.467 L978.118 823.208 L986.382 821.541 L991.057 821.541 L991.057 852.166 L998.696 852.166 L998.696 856.101 L978.789 856.101 L978.789 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1008.19 821.541 L1026.54 821.541 L1026.54 825.476 L1012.47 825.476 L1012.47 833.949 Q1013.49 833.601 1014.51 833.439 Q1015.52 833.254 1016.54 833.254 Q1022.33 833.254 1025.71 836.425 Q1029.09 839.597 1029.09 845.013 Q1029.09 850.592 1025.62 853.694 Q1022.15 856.773 1015.83 856.773 Q1013.65 856.773 1011.38 856.402 Q1009.14 856.032 1006.73 855.291 L1006.73 850.592 Q1008.81 851.726 1011.03 852.282 Q1013.26 852.837 1015.73 852.837 Q1019.74 852.837 1022.08 850.731 Q1024.41 848.624 1024.41 845.013 Q1024.41 841.402 1022.08 839.296 Q1019.74 837.189 1015.73 837.189 Q1013.86 837.189 1011.98 837.606 Q1010.13 838.023 1008.19 838.902 L1008.19 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1048.3 824.62 Q1044.69 824.62 1042.86 828.185 Q1041.06 831.726 1041.06 838.856 Q1041.06 845.962 1042.86 849.527 Q1044.69 853.069 1048.3 853.069 Q1051.94 853.069 1053.74 849.527 Q1055.57 845.962 1055.57 838.856 Q1055.57 831.726 1053.74 828.185 Q1051.94 824.62 1048.3 824.62 M1048.3 820.916 Q1054.11 820.916 1057.17 825.523 Q1060.25 830.106 1060.25 838.856 Q1060.25 847.583 1057.17 852.189 Q1054.11 856.773 1048.3 856.773 Q1042.49 856.773 1039.41 852.189 Q1036.36 847.583 1036.36 838.856 Q1036.36 830.106 1039.41 825.523 Q1042.49 820.916 1048.3 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1078.46 824.62 Q1074.85 824.62 1073.02 828.185 Q1071.22 831.726 1071.22 838.856 Q1071.22 845.962 1073.02 849.527 Q1074.85 853.069 1078.46 853.069 Q1082.1 853.069 1083.9 849.527 Q1085.73 845.962 1085.73 838.856 Q1085.73 831.726 1083.9 828.185 Q1082.1 824.62 1078.46 824.62 M1078.46 820.916 Q1084.27 820.916 1087.33 825.523 Q1090.41 830.106 1090.41 838.856 Q1090.41 847.583 1087.33 852.189 Q1084.27 856.773 1078.46 856.773 Q1072.65 856.773 1069.58 852.189 Q1066.52 847.583 1066.52 838.856 Q1066.52 830.106 1069.58 825.523 Q1072.65 820.916 1078.46 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1263.12 852.166 L1279.44 852.166 L1279.44 856.101 L1257.49 856.101 L1257.49 852.166 Q1260.16 849.411 1264.74 844.782 Q1269.35 840.129 1270.53 838.786 Q1272.77 836.263 1273.65 834.527 Q1274.55 832.768 1274.55 831.078 Q1274.55 828.324 1272.61 826.587 Q1270.69 824.851 1267.59 824.851 Q1265.39 824.851 1262.93 825.615 Q1260.5 826.379 1257.73 827.93 L1257.73 823.208 Q1260.55 822.074 1263 821.495 Q1265.46 820.916 1267.49 820.916 Q1272.86 820.916 1276.06 823.601 Q1279.25 826.287 1279.25 830.777 Q1279.25 832.907 1278.44 834.828 Q1277.66 836.726 1275.55 839.319 Q1274.97 839.99 1271.87 843.208 Q1268.77 846.402 1263.12 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1299.25 824.62 Q1295.64 824.62 1293.81 828.185 Q1292.01 831.726 1292.01 838.856 Q1292.01 845.962 1293.81 849.527 Q1295.64 853.069 1299.25 853.069 Q1302.89 853.069 1304.69 849.527 Q1306.52 845.962 1306.52 838.856 Q1306.52 831.726 1304.69 828.185 Q1302.89 824.62 1299.25 824.62 M1299.25 820.916 Q1305.06 820.916 1308.12 825.523 Q1311.2 830.106 1311.2 838.856 Q1311.2 847.583 1308.12 852.189 Q1305.06 856.773 1299.25 856.773 Q1293.44 856.773 1290.36 852.189 Q1287.31 847.583 1287.31 838.856 Q1287.31 830.106 1290.36 825.523 Q1293.44 820.916 1299.25 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1329.42 824.62 Q1325.8 824.62 1323.98 828.185 Q1322.17 831.726 1322.17 838.856 Q1322.17 845.962 1323.98 849.527 Q1325.8 853.069 1329.42 853.069 Q1333.05 853.069 1334.86 849.527 Q1336.68 845.962 1336.68 838.856 Q1336.68 831.726 1334.86 828.185 Q1333.05 824.62 1329.42 824.62 M1329.42 820.916 Q1335.23 820.916 1338.28 825.523 Q1341.36 830.106 1341.36 838.856 Q1341.36 847.583 1338.28 852.189 Q1335.23 856.773 1329.42 856.773 Q1323.61 856.773 1320.53 852.189 Q1317.47 847.583 1317.47 838.856 Q1317.47 830.106 1320.53 825.523 Q1323.61 820.916 1329.42 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1359.58 824.62 Q1355.97 824.62 1354.14 828.185 Q1352.33 831.726 1352.33 838.856 Q1352.33 845.962 1354.14 849.527 Q1355.97 853.069 1359.58 853.069 Q1363.21 853.069 1365.02 849.527 Q1366.85 845.962 1366.85 838.856 Q1366.85 831.726 1365.02 828.185 Q1363.21 824.62 1359.58 824.62 M1359.58 820.916 Q1365.39 820.916 1368.44 825.523 Q1371.52 830.106 1371.52 838.856 Q1371.52 847.583 1368.44 852.189 Q1365.39 856.773 1359.58 856.773 Q1353.77 856.773 1350.69 852.189 Q1347.63 847.583 1347.63 838.856 Q1347.63 830.106 1350.69 825.523 Q1353.77 820.916 1359.58 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,705.729 1352.76,705.729 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,538.856 1352.76,538.856 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,371.983 1352.76,371.983 "/>
+<polyline clip-path="url(#clip032)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,205.109 1352.76,205.109 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,790.501 159.765,153.712 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,705.729 178.663,705.729 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,538.856 178.663,538.856 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,371.983 178.663,371.983 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,205.109 178.663,205.109 "/>
+<path clip-path="url(#clip030)" d="M51.3995 719.074 L59.0384 719.074 L59.0384 692.709 L50.7282 694.375 L50.7282 690.116 L58.9921 688.449 L63.668 688.449 L63.668 719.074 L71.3068 719.074 L71.3068 723.009 L51.3995 723.009 L51.3995 719.074 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M80.7512 717.13 L85.6354 717.13 L85.6354 723.009 L80.7512 723.009 L80.7512 717.13 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 691.528 Q102.209 691.528 100.381 695.093 Q98.5752 698.635 98.5752 705.764 Q98.5752 712.871 100.381 716.435 Q102.209 719.977 105.821 719.977 Q109.455 719.977 111.26 716.435 Q113.089 712.871 113.089 705.764 Q113.089 698.635 111.26 695.093 Q109.455 691.528 105.821 691.528 M105.821 687.824 Q111.631 687.824 114.686 692.431 Q117.765 697.014 117.765 705.764 Q117.765 714.491 114.686 719.097 Q111.631 723.681 105.821 723.681 Q100.01 723.681 96.9317 719.097 Q93.8761 714.491 93.8761 705.764 Q93.8761 697.014 96.9317 692.431 Q100.01 687.824 105.821 687.824 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M52.3949 552.201 L60.0337 552.201 L60.0337 525.835 L51.7236 527.502 L51.7236 523.243 L59.9874 521.576 L64.6633 521.576 L64.6633 552.201 L72.3022 552.201 L72.3022 556.136 L52.3949 556.136 L52.3949 552.201 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M81.7466 550.256 L86.6308 550.256 L86.6308 556.136 L81.7466 556.136 L81.7466 550.256 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M96.8622 521.576 L115.219 521.576 L115.219 525.511 L101.145 525.511 L101.145 533.983 Q102.163 533.636 103.182 533.474 Q104.2 533.289 105.219 533.289 Q111.006 533.289 114.385 536.46 Q117.765 539.632 117.765 545.048 Q117.765 550.627 114.293 553.729 Q110.82 556.807 104.501 556.807 Q102.325 556.807 100.057 556.437 Q97.8113 556.067 95.4039 555.326 L95.4039 550.627 Q97.4872 551.761 99.7094 552.317 Q101.932 552.872 104.408 552.872 Q108.413 552.872 110.751 550.766 Q113.089 548.659 113.089 545.048 Q113.089 541.437 110.751 539.331 Q108.413 537.224 104.408 537.224 Q102.534 537.224 100.659 537.641 Q98.8067 538.057 96.8622 538.937 L96.8622 521.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M54.6171 385.327 L70.9365 385.327 L70.9365 389.263 L48.9921 389.263 L48.9921 385.327 Q51.6541 382.573 56.2375 377.943 Q60.8439 373.29 62.0245 371.948 Q64.2698 369.425 65.1494 367.689 Q66.0522 365.929 66.0522 364.24 Q66.0522 361.485 64.1078 359.749 Q62.1865 358.013 59.0847 358.013 Q56.8856 358.013 54.4319 358.777 Q52.0014 359.541 49.2236 361.091 L49.2236 356.369 Q52.0477 355.235 54.5014 354.656 Q56.955 354.078 58.9921 354.078 Q64.3624 354.078 67.5568 356.763 Q70.7513 359.448 70.7513 363.939 Q70.7513 366.068 69.9411 367.99 Q69.1541 369.888 67.0476 372.48 Q66.4689 373.152 63.367 376.369 Q60.2652 379.564 54.6171 385.327 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M80.7512 383.383 L85.6354 383.383 L85.6354 389.263 L80.7512 389.263 L80.7512 383.383 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 357.781 Q102.209 357.781 100.381 361.346 Q98.5752 364.888 98.5752 372.017 Q98.5752 379.124 100.381 382.689 Q102.209 386.23 105.821 386.23 Q109.455 386.23 111.26 382.689 Q113.089 379.124 113.089 372.017 Q113.089 364.888 111.26 361.346 Q109.455 357.781 105.821 357.781 M105.821 354.078 Q111.631 354.078 114.686 358.684 Q117.765 363.267 117.765 372.017 Q117.765 380.744 114.686 385.351 Q111.631 389.934 105.821 389.934 Q100.01 389.934 96.9317 385.351 Q93.8761 380.744 93.8761 372.017 Q93.8761 363.267 96.9317 358.684 Q100.01 354.078 105.821 354.078 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M55.6125 218.454 L71.9318 218.454 L71.9318 222.389 L49.9875 222.389 L49.9875 218.454 Q52.6495 215.699 57.2328 211.07 Q61.8393 206.417 63.0198 205.074 Q65.2652 202.551 66.1448 200.815 Q67.0476 199.056 67.0476 197.366 Q67.0476 194.612 65.1032 192.875 Q63.1819 191.139 60.08 191.139 Q57.881 191.139 55.4273 191.903 Q52.9967 192.667 50.219 194.218 L50.219 189.496 Q53.043 188.362 55.4967 187.783 Q57.9504 187.204 59.9874 187.204 Q65.3578 187.204 68.5522 189.889 Q71.7466 192.574 71.7466 197.065 Q71.7466 199.195 70.9365 201.116 Q70.1494 203.014 68.0429 205.607 Q67.4642 206.278 64.3624 209.496 Q61.2606 212.69 55.6125 218.454 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M81.7466 216.51 L86.6308 216.51 L86.6308 222.389 L81.7466 222.389 L81.7466 216.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M96.8622 187.829 L115.219 187.829 L115.219 191.764 L101.145 191.764 L101.145 200.236 Q102.163 199.889 103.182 199.727 Q104.2 199.542 105.219 199.542 Q111.006 199.542 114.385 202.713 Q117.765 205.885 117.765 211.301 Q117.765 216.88 114.293 219.982 Q110.82 223.06 104.501 223.06 Q102.325 223.06 100.057 222.69 Q97.8113 222.32 95.4039 221.579 L95.4039 216.88 Q97.4872 218.014 99.7094 218.57 Q101.932 219.125 104.408 219.125 Q108.413 219.125 110.751 217.019 Q113.089 214.912 113.089 211.301 Q113.089 207.69 110.751 205.584 Q108.413 203.477 104.408 203.477 Q102.534 203.477 100.659 203.894 Q98.8067 204.311 96.8622 205.19 L96.8622 187.829 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M365.348 20.1573 L354.249 50.2555 L376.488 50.2555 L365.348 20.1573 M360.73 12.096 L370.007 12.096 L393.056 72.576 L384.549 72.576 L379.04 57.061 L351.778 57.061 L346.268 72.576 L337.64 72.576 L360.73 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M401.563 9.54393 L409.017 9.54393 L409.017 72.576 L401.563 72.576 L401.563 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M431.986 14.324 L431.986 27.2059 L447.338 27.2059 L447.338 32.9987 L431.986 32.9987 L431.986 57.6282 Q431.986 63.1779 433.484 64.7578 Q435.024 66.3376 439.682 66.3376 L447.338 66.3376 L447.338 72.576 L439.682 72.576 Q431.054 72.576 427.773 69.3758 Q424.491 66.1351 424.491 57.6282 L424.491 32.9987 L419.023 32.9987 L419.023 27.2059 L424.491 27.2059 L424.491 14.324 L431.986 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M457.142 27.2059 L464.595 27.2059 L464.595 72.576 L457.142 72.576 L457.142 27.2059 M457.142 9.54393 L464.595 9.54393 L464.595 18.9825 L457.142 18.9825 L457.142 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M487.564 14.324 L487.564 27.2059 L502.917 27.2059 L502.917 32.9987 L487.564 32.9987 L487.564 57.6282 Q487.564 63.1779 489.063 64.7578 Q490.602 66.3376 495.261 66.3376 L502.917 66.3376 L502.917 72.576 L495.261 72.576 Q486.632 72.576 483.351 69.3758 Q480.07 66.1351 480.07 57.6282 L480.07 32.9987 L474.601 32.9987 L474.601 27.2059 L480.07 27.2059 L480.07 14.324 L487.564 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M511.95 54.671 L511.95 27.2059 L519.404 27.2059 L519.404 54.3874 Q519.404 60.8284 521.916 64.0691 Q524.427 67.2693 529.45 67.2693 Q535.486 67.2693 538.97 63.421 Q542.494 59.5726 542.494 52.9291 L542.494 27.2059 L549.948 27.2059 L549.948 72.576 L542.494 72.576 L542.494 65.6084 Q539.78 69.7404 536.175 71.7658 Q532.61 73.7508 527.87 73.7508 Q520.052 73.7508 516.001 68.8897 Q511.95 64.0286 511.95 54.671 M530.706 26.1121 L530.706 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M595.156 34.0924 L595.156 9.54393 L602.61 9.54393 L602.61 72.576 L595.156 72.576 L595.156 65.7705 Q592.806 69.8214 589.201 71.8063 Q585.636 73.7508 580.613 73.7508 Q572.39 73.7508 567.205 67.1883 Q562.06 60.6258 562.06 49.9314 Q562.06 39.2371 567.205 32.6746 Q572.39 26.1121 580.613 26.1121 Q585.636 26.1121 589.201 28.0971 Q592.806 30.0415 595.156 34.0924 M569.757 49.9314 Q569.757 58.1548 573.119 62.8538 Q576.522 67.5124 582.436 67.5124 Q588.35 67.5124 591.753 62.8538 Q595.156 58.1548 595.156 49.9314 Q595.156 41.7081 591.753 37.0496 Q588.35 32.3505 582.436 32.3505 Q576.522 32.3505 573.119 37.0496 Q569.757 41.7081 569.757 49.9314 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M656.77 48.0275 L656.77 51.6733 L622.5 51.6733 Q622.986 59.3701 627.118 63.421 Q631.29 67.4314 638.703 67.4314 Q642.997 67.4314 647.008 66.3781 Q651.058 65.3249 655.028 63.2184 L655.028 70.267 Q651.018 71.9684 646.805 72.8596 Q642.592 73.7508 638.258 73.7508 Q627.401 73.7508 621.041 67.4314 Q614.722 61.1119 614.722 50.3365 Q614.722 39.1965 620.717 32.6746 Q626.753 26.1121 636.961 26.1121 Q646.116 26.1121 651.423 32.0264 Q656.77 37.9003 656.77 48.0275 M649.317 45.84 Q649.236 39.7232 645.873 36.0774 Q642.552 32.4315 637.042 32.4315 Q630.804 32.4315 627.037 35.9558 Q623.31 39.4801 622.743 45.8805 L649.317 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M713.28 9.62495 Q707.852 18.942 705.219 28.0566 Q702.586 37.1711 702.586 46.5287 Q702.586 55.8863 705.219 65.0818 Q707.893 74.2369 713.28 83.5134 L706.799 83.5134 Q700.723 73.9938 697.684 64.7983 Q694.687 55.6027 694.687 46.5287 Q694.687 37.4952 697.684 28.3401 Q700.682 19.1851 706.799 9.62495 L713.28 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M730.213 65.6895 L743.581 65.6895 L743.581 19.5497 L729.038 22.4663 L729.038 15.0127 L743.5 12.096 L751.683 12.096 L751.683 65.6895 L765.051 65.6895 L765.051 72.576 L730.213 72.576 L730.213 65.6895 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M799.078 17.4837 Q792.759 17.4837 789.559 23.7221 Q786.399 29.92 786.399 42.3968 Q786.399 54.833 789.559 61.0714 Q792.759 67.2693 799.078 67.2693 Q805.438 67.2693 808.598 61.0714 Q811.798 54.833 811.798 42.3968 Q811.798 29.92 808.598 23.7221 Q805.438 17.4837 799.078 17.4837 M799.078 11.0023 Q809.246 11.0023 814.593 19.0636 Q819.981 27.0843 819.981 42.3968 Q819.981 57.6687 814.593 65.73 Q809.246 73.7508 799.078 73.7508 Q788.911 73.7508 783.523 65.73 Q778.176 57.6687 778.176 42.3968 Q778.176 27.0843 783.523 19.0636 Q788.911 11.0023 799.078 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M851.862 17.4837 Q845.542 17.4837 842.342 23.7221 Q839.182 29.92 839.182 42.3968 Q839.182 54.833 842.342 61.0714 Q845.542 67.2693 851.862 67.2693 Q858.222 67.2693 861.381 61.0714 Q864.582 54.833 864.582 42.3968 Q864.582 29.92 861.381 23.7221 Q858.222 17.4837 851.862 17.4837 M851.862 11.0023 Q862.03 11.0023 867.377 19.0636 Q872.764 27.0843 872.764 42.3968 Q872.764 57.6687 867.377 65.73 Q862.03 73.7508 851.862 73.7508 Q841.694 73.7508 836.306 65.73 Q830.959 57.6687 830.959 42.3968 Q830.959 27.0843 836.306 19.0636 Q841.694 11.0023 851.862 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M887.996 62.2867 L896.543 62.2867 L896.543 69.2543 L889.9 82.2172 L884.674 82.2172 L887.996 69.2543 L887.996 62.2867 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M931.016 17.4837 Q924.697 17.4837 921.497 23.7221 Q918.337 29.92 918.337 42.3968 Q918.337 54.833 921.497 61.0714 Q924.697 67.2693 931.016 67.2693 Q937.376 67.2693 940.536 61.0714 Q943.736 54.833 943.736 42.3968 Q943.736 29.92 940.536 23.7221 Q937.376 17.4837 931.016 17.4837 M931.016 11.0023 Q941.184 11.0023 946.531 19.0636 Q951.919 27.0843 951.919 42.3968 Q951.919 57.6687 946.531 65.73 Q941.184 73.7508 931.016 73.7508 Q920.849 73.7508 915.461 65.73 Q910.114 57.6687 910.114 42.3968 Q910.114 27.0843 915.461 19.0636 Q920.849 11.0023 931.016 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M983.8 17.4837 Q977.48 17.4837 974.28 23.7221 Q971.12 29.92 971.12 42.3968 Q971.12 54.833 974.28 61.0714 Q977.48 67.2693 983.8 67.2693 Q990.16 67.2693 993.319 61.0714 Q996.52 54.833 996.52 42.3968 Q996.52 29.92 993.319 23.7221 Q990.16 17.4837 983.8 17.4837 M983.8 11.0023 Q993.967 11.0023 999.315 19.0636 Q1004.7 27.0843 1004.7 42.3968 Q1004.7 57.6687 999.315 65.73 Q993.967 73.7508 983.8 73.7508 Q973.632 73.7508 968.244 65.73 Q962.897 57.6687 962.897 42.3968 Q962.897 27.0843 968.244 19.0636 Q973.632 11.0023 983.8 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1036.58 17.4837 Q1030.26 17.4837 1027.06 23.7221 Q1023.9 29.92 1023.9 42.3968 Q1023.9 54.833 1027.06 61.0714 Q1030.26 67.2693 1036.58 67.2693 Q1042.94 67.2693 1046.1 61.0714 Q1049.3 54.833 1049.3 42.3968 Q1049.3 29.92 1046.1 23.7221 Q1042.94 17.4837 1036.58 17.4837 M1036.58 11.0023 Q1046.75 11.0023 1052.1 19.0636 Q1057.49 27.0843 1057.49 42.3968 Q1057.49 57.6687 1052.1 65.73 Q1046.75 73.7508 1036.58 73.7508 Q1026.42 73.7508 1021.03 65.73 Q1015.68 57.6687 1015.68 42.3968 Q1015.68 27.0843 1021.03 19.0636 Q1026.42 11.0023 1036.58 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1120.15 9.54393 L1120.15 15.7418 L1113.02 15.7418 Q1109.01 15.7418 1107.43 17.3622 Q1105.89 18.9825 1105.89 23.1955 L1105.89 27.2059 L1118.17 27.2059 L1118.17 32.9987 L1105.89 32.9987 L1105.89 72.576 L1098.4 72.576 L1098.4 32.9987 L1091.27 32.9987 L1091.27 27.2059 L1098.4 27.2059 L1098.4 24.0462 Q1098.4 16.471 1101.92 13.0277 Q1105.45 9.54393 1113.1 9.54393 L1120.15 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1132.31 14.324 L1132.31 27.2059 L1147.66 27.2059 L1147.66 32.9987 L1132.31 32.9987 L1132.31 57.6282 Q1132.31 63.1779 1133.8 64.7578 Q1135.34 66.3376 1140 66.3376 L1147.66 66.3376 L1147.66 72.576 L1140 72.576 Q1131.37 72.576 1128.09 69.3758 Q1124.81 66.1351 1124.81 57.6282 L1124.81 32.9987 L1119.34 32.9987 L1119.34 27.2059 L1124.81 27.2059 L1124.81 14.324 L1132.31 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1156.29 9.62495 L1162.77 9.62495 Q1168.85 19.1851 1171.84 28.3401 Q1174.88 37.4952 1174.88 46.5287 Q1174.88 55.6027 1171.84 64.7983 Q1168.85 73.9938 1162.77 83.5134 L1156.29 83.5134 Q1161.67 74.2369 1164.31 65.0818 Q1166.98 55.8863 1166.98 46.5287 Q1166.98 37.1711 1164.31 28.0566 Q1161.67 18.942 1156.29 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip032)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,171.734 195.771,177.699 198.013,183.661 200.255,189.62 202.497,195.572 204.739,201.516 206.981,207.449 209.223,213.368 211.464,219.271 213.706,225.155 215.948,231.016 218.19,236.851 220.432,242.656 222.674,248.427 224.916,254.159 227.158,259.849 229.4,265.491 231.642,271.08 233.884,276.61 236.126,282.075 238.368,287.47 240.61,292.787 242.852,298.02 245.094,303.161 247.336,308.203 249.578,313.137 251.82,317.957 254.062,322.653 256.304,327.217 258.546,331.64 260.788,335.913 263.03,340.027 265.271,343.974 267.513,347.745 269.755,351.332 271.997,354.725 274.239,357.918 276.481,360.904 278.723,363.674 280.965,366.224 283.207,368.548 285.449,370.642 287.691,372.502 289.933,374.127 292.175,375.514 294.417,376.663 296.659,377.577 298.901,378.256 301.143,378.704 303.385,378.925 305.627,378.925 307.869,378.71 310.111,378.289 312.353,377.668 314.595,376.858 316.837,375.867 319.078,374.708 321.32,373.39 323.562,371.925 325.804,370.325 328.046,368.602 330.288,366.768 332.53,364.835 334.772,362.816 337.014,360.723 339.256,358.567 341.498,356.361 343.74,354.117 345.982,351.846 348.224,349.558 350.466,347.266 352.708,344.979 354.95,342.707 357.192,340.46 359.434,338.248 361.676,336.078 363.918,333.961 366.16,331.903 368.402,329.913 370.644,327.998 372.885,326.165 375.127,324.42 377.369,322.769 379.611,321.219 381.853,319.774 384.095,318.439 386.337,317.219 388.579,316.119 390.821,315.142 393.063,314.291 395.305,313.57 397.547,312.981 399.789,312.528 402.031,312.212 404.273,312.034 406.515,311.997 408.757,312.101 410.999,312.347 413.241,312.736 415.483,313.267 417.725,313.94 419.967,314.754 422.209,315.708 424.451,316.8 426.692,318.029 428.934,319.392 431.176,320.886 433.418,322.509 435.66,324.256 437.902,326.123 440.144,328.106 442.386,330.2 444.628,332.398 446.87,334.695 449.112,337.084 451.354,339.558 453.596,342.109 455.838,344.729 458.08,347.41 460.322,350.14 462.564,352.912 464.806,355.714 467.048,358.536 469.29,361.366 471.532,364.193 473.774,367.004 476.016,369.788 478.258,372.532 480.5,375.222 482.741,377.848 484.983,380.395 487.225,382.851 489.467,385.206 491.709,387.446 493.951,389.561 496.193,391.541 498.435,393.377 500.677,395.06 502.919,396.582 505.161,397.939 507.403,399.125 509.645,400.137 511.887,400.973 514.129,401.633 516.371,402.119 518.613,402.433 520.855,402.579 523.097,402.562 525.339,402.391 527.581,402.073 529.823,401.617 532.065,401.033 534.307,400.334 536.548,399.53 538.79,398.634 541.032,397.66 543.274,396.621 545.516,395.529 547.758,394.399 550,393.244 552.242,392.078 554.484,390.914 556.726,389.764 558.968,388.641 561.21,387.557 563.452,386.523 565.694,385.55 567.936,384.648 570.178,383.826 572.42,383.094 574.662,382.459 576.904,381.928 579.146,381.508 581.388,381.204 583.63,381.022 585.872,380.964 588.114,381.035 590.355,381.236 592.597,381.569 594.839,382.034 597.081,382.632 599.323,383.36 601.565,384.216 603.807,385.198 606.049,386.302 608.291,387.523 610.533,388.855 612.775,390.292 615.017,391.827 617.259,393.451 619.501,395.156 621.743,396.932 623.985,398.769 626.227,400.655 628.469,402.58 630.711,404.531 632.953,406.497 635.195,408.464 637.437,410.419 639.679,412.351 641.921,414.245 644.162,416.09 646.404,417.873 648.646,419.583 650.888,421.208 653.13,422.739 655.372,424.167 657.614,425.483 659.856,426.681 662.098,427.756 664.34,428.703 666.582,429.52 668.824,430.205 671.066,430.761 673.308,431.189 675.55,431.492 677.792,431.676 680.034,431.747 682.276,431.713 684.518,431.582 686.76,431.365 689.002,431.072 691.244,430.714 693.486,430.304 695.728,429.853 697.969,429.373 700.211,428.878 702.453,428.379 704.695,427.888 706.937,427.418 709.179,426.98 711.421,426.585 713.663,426.242 715.905,425.961 718.147,425.751 720.389,425.619 722.631,425.573 724.873,425.618 727.115,425.759 729.357,425.999 731.599,426.343 733.841,426.79 736.083,427.343 738.325,428 740.567,428.76 742.809,429.62 745.051,430.577 747.293,431.626 749.535,432.762 751.776,433.977 754.018,435.265 756.26,436.617 758.502,438.025 760.744,439.479 762.986,440.968 765.228,442.482 767.47,444.01 769.712,445.542 771.954,447.066 774.196,448.571 776.438,450.046 778.68,451.482 780.922,452.868 783.164,454.195 785.406,455.455 787.648,456.641 789.89,457.747 792.132,458.767 794.374,459.699 796.616,460.539 798.858,461.286 801.1,461.942 803.342,462.508 805.583,462.986 807.825,463.382 810.067,463.7 812.309,463.947 814.551,464.131 816.793,464.259 819.035,464.341 821.277,464.385 823.519,464.403 825.761,464.403 828.003,464.395 830.245,464.391 832.487,464.398 834.729,464.428 836.971,464.489 839.213,464.588 841.455,464.735 843.697,464.935 845.939,465.195 848.181,465.521 850.423,465.916 852.665,466.384 854.907,466.927 857.149,467.547 859.391,468.243 861.632,469.016 863.874,469.863 866.116,470.782 868.358,471.769 870.6,472.819 872.842,473.928 875.084,475.089 877.326,476.295 879.568,477.539 881.81,478.814 884.052,480.11 886.294,481.421 888.536,482.736 890.778,484.048 893.02,485.349 895.262,486.63 897.504,487.884 899.746,489.104 901.988,490.282 904.23,491.415 906.472,492.497 908.714,493.523 910.956,494.492 913.198,495.402 915.439,496.252 917.681,497.041 919.923,497.773 922.165,498.449 924.407,499.072 926.649,499.647 928.891,500.179 931.133,500.673 933.375,501.137 935.617,501.575 937.859,501.997 940.101,502.408 942.343,502.816 944.585,503.229 946.827,503.653 949.069,504.096 951.311,504.563 953.553,505.061 955.795,505.594 958.037,506.169 960.279,506.788 962.521,507.455 964.763,508.173 967.005,508.943 969.246,509.767 971.488,510.643 973.73,511.572 975.972,512.552 978.214,513.581 980.456,514.656 982.698,515.773 984.94,516.928 987.182,518.116 989.424,519.332 991.666,520.57 993.908,521.826 996.15,523.093 998.392,524.365 1000.63,525.636 1002.88,526.902 1005.12,528.156 1007.36,529.393 1009.6,530.61 1011.84,531.802 1014.09,532.966 1016.33,534.099 1018.57,535.2 1020.81,536.267 1023.05,537.3 1025.3,538.299 1027.54,539.265 1029.78,540.2 1032.02,541.107 1034.26,541.988 1036.51,542.848 1038.75,543.689 1040.99,544.516 1043.23,545.335 1045.47,546.149 1047.72,546.963 1049.96,547.783 1052.2,548.614 1054.44,549.459 1056.68,550.323 1058.92,551.21 1061.17,552.125 1063.41,553.068 1065.65,554.045 1067.89,555.056 1070.13,556.103 1072.38,557.188 1074.62,558.31 1076.86,559.469 1079.1,560.666 1081.34,561.897 1083.59,563.162 1085.83,564.459 1088.07,565.784 1090.31,567.135 1092.55,568.508 1094.8,569.9 1097.04,571.308 1099.28,572.727 1101.52,574.154 1103.76,575.585 1106.01,577.016 1108.25,578.445 1110.49,579.869 1112.73,581.284 1114.97,582.688 1117.22,584.08 1119.46,585.459 1121.7,586.823 1123.94,588.172 1126.18,589.506 1128.43,590.827 1130.67,592.134 1132.91,593.43 1135.15,594.718 1137.39,595.998 1139.64,597.274 1141.88,598.55 1144.12,599.827 1146.36,601.111 1148.6,602.403 1150.85,603.708 1153.09,605.029 1155.33,606.369 1157.57,607.731 1159.81,609.119 1162.05,610.535 1164.3,611.981 1166.54,613.459 1168.78,614.971 1171.02,616.518 1173.26,618.1 1175.51,619.719 1177.75,621.373 1179.99,623.062 1182.23,624.785 1184.47,626.541 1186.72,628.329 1188.96,630.145 1191.2,631.987 1193.44,633.854 1195.68,635.741 1197.93,637.646 1200.17,639.567 1202.41,641.499 1204.65,643.44 1206.89,645.387 1209.14,647.337 1211.38,649.287 1213.62,651.236 1215.86,653.181 1218.1,655.121 1220.35,657.054 1222.59,658.981 1224.83,660.9 1227.07,662.814 1229.31,664.721 1231.56,666.626 1233.8,668.528 1236.04,670.432 1238.28,672.34 1240.52,674.256 1242.77,676.185 1245.01,678.131 1247.25,680.1 1249.49,682.096 1251.73,684.126 1253.98,686.194 1256.22,688.308 1258.46,690.472 1260.7,692.693 1262.94,694.976 1265.19,697.326 1267.43,699.748 1269.67,702.247 1271.91,704.825 1274.15,707.487 1276.39,710.234 1278.64,713.068 1280.88,715.989 1283.12,718.997 1285.36,722.089 1287.6,725.262 1289.85,728.511 1292.09,731.831 1294.33,735.212 1296.57,738.645 1298.81,742.119 1301.06,745.62 1303.3,749.133 1305.54,752.642 1307.78,756.127 1310.02,759.568 1312.27,762.946 1314.51,766.238 1316.75,769.422 1318.99,772.479 "/>
+<path clip-path="url(#clip030)" d="M1559.76 790.501 L2752.76 790.501 L2752.76 153.712 L1559.76 153.712  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip213">
+  <clipPath id="clip033">
     <rect x="1559" y="153" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,790.501 1593.53,153.712 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,790.501 1873.77,153.712 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,790.501 2154.02,153.712 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,790.501 2434.26,153.712 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,790.501 2714.51,153.712 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,790.501 2752.76,790.501 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,790.501 1593.53,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,790.501 1873.77,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,790.501 2154.02,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,790.501 2434.26,771.604 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,790.501 2714.51,771.604 "/>
-<path clip-path="url(#clip210)" d="M1593.53 824.62 Q1589.92 824.62 1588.09 828.185 Q1586.28 831.726 1586.28 838.856 Q1586.28 845.962 1588.09 849.527 Q1589.92 853.069 1593.53 853.069 Q1597.16 853.069 1598.97 849.527 Q1600.8 845.962 1600.8 838.856 Q1600.8 831.726 1598.97 828.185 Q1597.16 824.62 1593.53 824.62 M1593.53 820.916 Q1599.34 820.916 1602.39 825.523 Q1605.47 830.106 1605.47 838.856 Q1605.47 847.583 1602.39 852.189 Q1599.34 856.773 1593.53 856.773 Q1587.72 856.773 1584.64 852.189 Q1581.58 847.583 1581.58 838.856 Q1581.58 830.106 1584.64 825.523 Q1587.72 820.916 1593.53 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1833.39 821.541 L1851.75 821.541 L1851.75 825.476 L1837.67 825.476 L1837.67 833.949 Q1838.69 833.601 1839.71 833.439 Q1840.73 833.254 1841.75 833.254 Q1847.54 833.254 1850.91 836.425 Q1854.29 839.597 1854.29 845.013 Q1854.29 850.592 1850.82 853.694 Q1847.35 856.773 1841.03 856.773 Q1838.85 856.773 1836.59 856.402 Q1834.34 856.032 1831.93 855.291 L1831.93 850.592 Q1834.02 851.726 1836.24 852.282 Q1838.46 852.837 1840.94 852.837 Q1844.94 852.837 1847.28 850.731 Q1849.62 848.624 1849.62 845.013 Q1849.62 841.402 1847.28 839.296 Q1844.94 837.189 1840.94 837.189 Q1839.06 837.189 1837.19 837.606 Q1835.34 838.023 1833.39 838.902 L1833.39 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1873.51 824.62 Q1869.9 824.62 1868.07 828.185 Q1866.26 831.726 1866.26 838.856 Q1866.26 845.962 1868.07 849.527 Q1869.9 853.069 1873.51 853.069 Q1877.14 853.069 1878.95 849.527 Q1880.78 845.962 1880.78 838.856 Q1880.78 831.726 1878.95 828.185 Q1877.14 824.62 1873.51 824.62 M1873.51 820.916 Q1879.32 820.916 1882.37 825.523 Q1885.45 830.106 1885.45 838.856 Q1885.45 847.583 1882.37 852.189 Q1879.32 856.773 1873.51 856.773 Q1867.7 856.773 1864.62 852.189 Q1861.56 847.583 1861.56 838.856 Q1861.56 830.106 1864.62 825.523 Q1867.7 820.916 1873.51 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1903.67 824.62 Q1900.06 824.62 1898.23 828.185 Q1896.42 831.726 1896.42 838.856 Q1896.42 845.962 1898.23 849.527 Q1900.06 853.069 1903.67 853.069 Q1907.3 853.069 1909.11 849.527 Q1910.94 845.962 1910.94 838.856 Q1910.94 831.726 1909.11 828.185 Q1907.3 824.62 1903.67 824.62 M1903.67 820.916 Q1909.48 820.916 1912.53 825.523 Q1915.61 830.106 1915.61 838.856 Q1915.61 847.583 1912.53 852.189 Q1909.48 856.773 1903.67 856.773 Q1897.86 856.773 1894.78 852.189 Q1891.72 847.583 1891.72 838.856 Q1891.72 830.106 1894.78 825.523 Q1897.86 820.916 1903.67 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2098.54 852.166 L2106.18 852.166 L2106.18 825.8 L2097.87 827.467 L2097.87 823.208 L2106.14 821.541 L2110.81 821.541 L2110.81 852.166 L2118.45 852.166 L2118.45 856.101 L2098.54 856.101 L2098.54 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2137.9 824.62 Q2134.28 824.62 2132.46 828.185 Q2130.65 831.726 2130.65 838.856 Q2130.65 845.962 2132.46 849.527 Q2134.28 853.069 2137.9 853.069 Q2141.53 853.069 2143.34 849.527 Q2145.16 845.962 2145.16 838.856 Q2145.16 831.726 2143.34 828.185 Q2141.53 824.62 2137.9 824.62 M2137.9 820.916 Q2143.71 820.916 2146.76 825.523 Q2149.84 830.106 2149.84 838.856 Q2149.84 847.583 2146.76 852.189 Q2143.71 856.773 2137.9 856.773 Q2132.09 856.773 2129.01 852.189 Q2125.95 847.583 2125.95 838.856 Q2125.95 830.106 2129.01 825.523 Q2132.09 820.916 2137.9 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2168.06 824.62 Q2164.45 824.62 2162.62 828.185 Q2160.81 831.726 2160.81 838.856 Q2160.81 845.962 2162.62 849.527 Q2164.45 853.069 2168.06 853.069 Q2171.69 853.069 2173.5 849.527 Q2175.33 845.962 2175.33 838.856 Q2175.33 831.726 2173.5 828.185 Q2171.69 824.62 2168.06 824.62 M2168.06 820.916 Q2173.87 820.916 2176.92 825.523 Q2180 830.106 2180 838.856 Q2180 847.583 2176.92 852.189 Q2173.87 856.773 2168.06 856.773 Q2162.25 856.773 2159.17 852.189 Q2156.11 847.583 2156.11 838.856 Q2156.11 830.106 2159.17 825.523 Q2162.25 820.916 2168.06 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2198.22 824.62 Q2194.61 824.62 2192.78 828.185 Q2190.97 831.726 2190.97 838.856 Q2190.97 845.962 2192.78 849.527 Q2194.61 853.069 2198.22 853.069 Q2201.85 853.069 2203.66 849.527 Q2205.49 845.962 2205.49 838.856 Q2205.49 831.726 2203.66 828.185 Q2201.85 824.62 2198.22 824.62 M2198.22 820.916 Q2204.03 820.916 2207.09 825.523 Q2210.16 830.106 2210.16 838.856 Q2210.16 847.583 2207.09 852.189 Q2204.03 856.773 2198.22 856.773 Q2192.41 856.773 2189.33 852.189 Q2186.28 847.583 2186.28 838.856 Q2186.28 830.106 2189.33 825.523 Q2192.41 820.916 2198.22 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2378.79 852.166 L2386.43 852.166 L2386.43 825.8 L2378.12 827.467 L2378.12 823.208 L2386.38 821.541 L2391.06 821.541 L2391.06 852.166 L2398.7 852.166 L2398.7 856.101 L2378.79 856.101 L2378.79 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2408.19 821.541 L2426.54 821.541 L2426.54 825.476 L2412.47 825.476 L2412.47 833.949 Q2413.49 833.601 2414.51 833.439 Q2415.52 833.254 2416.54 833.254 Q2422.33 833.254 2425.71 836.425 Q2429.09 839.597 2429.09 845.013 Q2429.09 850.592 2425.62 853.694 Q2422.15 856.773 2415.83 856.773 Q2413.65 856.773 2411.38 856.402 Q2409.14 856.032 2406.73 855.291 L2406.73 850.592 Q2408.81 851.726 2411.03 852.282 Q2413.26 852.837 2415.73 852.837 Q2419.74 852.837 2422.08 850.731 Q2424.41 848.624 2424.41 845.013 Q2424.41 841.402 2422.08 839.296 Q2419.74 837.189 2415.73 837.189 Q2413.86 837.189 2411.98 837.606 Q2410.13 838.023 2408.19 838.902 L2408.19 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2448.3 824.62 Q2444.69 824.62 2442.86 828.185 Q2441.06 831.726 2441.06 838.856 Q2441.06 845.962 2442.86 849.527 Q2444.69 853.069 2448.3 853.069 Q2451.94 853.069 2453.74 849.527 Q2455.57 845.962 2455.57 838.856 Q2455.57 831.726 2453.74 828.185 Q2451.94 824.62 2448.3 824.62 M2448.3 820.916 Q2454.11 820.916 2457.17 825.523 Q2460.25 830.106 2460.25 838.856 Q2460.25 847.583 2457.17 852.189 Q2454.11 856.773 2448.3 856.773 Q2442.49 856.773 2439.41 852.189 Q2436.36 847.583 2436.36 838.856 Q2436.36 830.106 2439.41 825.523 Q2442.49 820.916 2448.3 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2478.46 824.62 Q2474.85 824.62 2473.02 828.185 Q2471.22 831.726 2471.22 838.856 Q2471.22 845.962 2473.02 849.527 Q2474.85 853.069 2478.46 853.069 Q2482.1 853.069 2483.9 849.527 Q2485.73 845.962 2485.73 838.856 Q2485.73 831.726 2483.9 828.185 Q2482.1 824.62 2478.46 824.62 M2478.46 820.916 Q2484.27 820.916 2487.33 825.523 Q2490.41 830.106 2490.41 838.856 Q2490.41 847.583 2487.33 852.189 Q2484.27 856.773 2478.46 856.773 Q2472.65 856.773 2469.58 852.189 Q2466.52 847.583 2466.52 838.856 Q2466.52 830.106 2469.58 825.523 Q2472.65 820.916 2478.46 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2663.12 852.166 L2679.44 852.166 L2679.44 856.101 L2657.49 856.101 L2657.49 852.166 Q2660.16 849.411 2664.74 844.782 Q2669.35 840.129 2670.53 838.786 Q2672.77 836.263 2673.65 834.527 Q2674.55 832.768 2674.55 831.078 Q2674.55 828.324 2672.61 826.587 Q2670.69 824.851 2667.59 824.851 Q2665.39 824.851 2662.93 825.615 Q2660.5 826.379 2657.73 827.93 L2657.73 823.208 Q2660.55 822.074 2663 821.495 Q2665.46 820.916 2667.49 820.916 Q2672.86 820.916 2676.06 823.601 Q2679.25 826.287 2679.25 830.777 Q2679.25 832.907 2678.44 834.828 Q2677.66 836.726 2675.55 839.319 Q2674.97 839.99 2671.87 843.208 Q2668.77 846.402 2663.12 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2699.25 824.62 Q2695.64 824.62 2693.81 828.185 Q2692.01 831.726 2692.01 838.856 Q2692.01 845.962 2693.81 849.527 Q2695.64 853.069 2699.25 853.069 Q2702.89 853.069 2704.69 849.527 Q2706.52 845.962 2706.52 838.856 Q2706.52 831.726 2704.69 828.185 Q2702.89 824.62 2699.25 824.62 M2699.25 820.916 Q2705.06 820.916 2708.12 825.523 Q2711.2 830.106 2711.2 838.856 Q2711.2 847.583 2708.12 852.189 Q2705.06 856.773 2699.25 856.773 Q2693.44 856.773 2690.36 852.189 Q2687.31 847.583 2687.31 838.856 Q2687.31 830.106 2690.36 825.523 Q2693.44 820.916 2699.25 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2729.42 824.62 Q2725.8 824.62 2723.98 828.185 Q2722.17 831.726 2722.17 838.856 Q2722.17 845.962 2723.98 849.527 Q2725.8 853.069 2729.42 853.069 Q2733.05 853.069 2734.86 849.527 Q2736.68 845.962 2736.68 838.856 Q2736.68 831.726 2734.86 828.185 Q2733.05 824.62 2729.42 824.62 M2729.42 820.916 Q2735.23 820.916 2738.28 825.523 Q2741.36 830.106 2741.36 838.856 Q2741.36 847.583 2738.28 852.189 Q2735.23 856.773 2729.42 856.773 Q2723.61 856.773 2720.53 852.189 Q2717.47 847.583 2717.47 838.856 Q2717.47 830.106 2720.53 825.523 Q2723.61 820.916 2729.42 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2759.58 824.62 Q2755.97 824.62 2754.14 828.185 Q2752.33 831.726 2752.33 838.856 Q2752.33 845.962 2754.14 849.527 Q2755.97 853.069 2759.58 853.069 Q2763.21 853.069 2765.02 849.527 Q2766.85 845.962 2766.85 838.856 Q2766.85 831.726 2765.02 828.185 Q2763.21 824.62 2759.58 824.62 M2759.58 820.916 Q2765.39 820.916 2768.44 825.523 Q2771.52 830.106 2771.52 838.856 Q2771.52 847.583 2768.44 852.189 Q2765.39 856.773 2759.58 856.773 Q2753.77 856.773 2750.69 852.189 Q2747.63 847.583 2747.63 838.856 Q2747.63 830.106 2750.69 825.523 Q2753.77 820.916 2759.58 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,707.463 2752.76,707.463 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,577.432 2752.76,577.432 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,447.401 2752.76,447.401 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,317.369 2752.76,317.369 "/>
-<polyline clip-path="url(#clip213)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,187.338 2752.76,187.338 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,790.501 1559.76,153.712 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,707.463 1578.66,707.463 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,577.432 1578.66,577.432 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,447.401 1578.66,447.401 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,317.369 1578.66,317.369 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,187.338 1578.66,187.338 "/>
-<path clip-path="url(#clip210)" d="M1461.58 693.262 Q1457.97 693.262 1456.14 696.827 Q1454.34 700.368 1454.34 707.498 Q1454.34 714.604 1456.14 718.169 Q1457.97 721.711 1461.58 721.711 Q1465.22 721.711 1467.02 718.169 Q1468.85 714.604 1468.85 707.498 Q1468.85 700.368 1467.02 696.827 Q1465.22 693.262 1461.58 693.262 M1461.58 689.558 Q1467.39 689.558 1470.45 694.165 Q1473.53 698.748 1473.53 707.498 Q1473.53 716.225 1470.45 720.831 Q1467.39 725.415 1461.58 725.415 Q1455.77 725.415 1452.7 720.831 Q1449.64 716.225 1449.64 707.498 Q1449.64 698.748 1452.7 694.165 Q1455.77 689.558 1461.58 689.558 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1481.75 718.864 L1486.63 718.864 L1486.63 724.743 L1481.75 724.743 L1481.75 718.864 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1496.86 690.183 L1515.22 690.183 L1515.22 694.118 L1501.14 694.118 L1501.14 702.591 Q1502.16 702.243 1503.18 702.081 Q1504.2 701.896 1505.22 701.896 Q1511.01 701.896 1514.39 705.067 Q1517.76 708.239 1517.76 713.655 Q1517.76 719.234 1514.29 722.336 Q1510.82 725.415 1504.5 725.415 Q1502.33 725.415 1500.06 725.044 Q1497.81 724.674 1495.4 723.933 L1495.4 719.234 Q1497.49 720.368 1499.71 720.924 Q1501.93 721.479 1504.41 721.479 Q1508.41 721.479 1510.75 719.373 Q1513.09 717.266 1513.09 713.655 Q1513.09 710.044 1510.75 707.938 Q1508.41 705.831 1504.41 705.831 Q1502.53 705.831 1500.66 706.248 Q1498.81 706.665 1496.86 707.544 L1496.86 690.183 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1451.4 590.777 L1459.04 590.777 L1459.04 564.411 L1450.73 566.078 L1450.73 561.819 L1458.99 560.152 L1463.67 560.152 L1463.67 590.777 L1471.31 590.777 L1471.31 594.712 L1451.4 594.712 L1451.4 590.777 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1480.75 588.832 L1485.64 588.832 L1485.64 594.712 L1480.75 594.712 L1480.75 588.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 563.231 Q1502.21 563.231 1500.38 566.795 Q1498.58 570.337 1498.58 577.467 Q1498.58 584.573 1500.38 588.138 Q1502.21 591.68 1505.82 591.68 Q1509.45 591.68 1511.26 588.138 Q1513.09 584.573 1513.09 577.467 Q1513.09 570.337 1511.26 566.795 Q1509.45 563.231 1505.82 563.231 M1505.82 559.527 Q1511.63 559.527 1514.69 564.133 Q1517.76 568.717 1517.76 577.467 Q1517.76 586.194 1514.69 590.8 Q1511.63 595.383 1505.82 595.383 Q1500.01 595.383 1496.93 590.8 Q1493.88 586.194 1493.88 577.467 Q1493.88 568.717 1496.93 564.133 Q1500.01 559.527 1505.82 559.527 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1452.39 460.746 L1460.03 460.746 L1460.03 434.38 L1451.72 436.047 L1451.72 431.787 L1459.99 430.121 L1464.66 430.121 L1464.66 460.746 L1472.3 460.746 L1472.3 464.681 L1452.39 464.681 L1452.39 460.746 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1481.75 458.801 L1486.63 458.801 L1486.63 464.681 L1481.75 464.681 L1481.75 458.801 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1496.86 430.121 L1515.22 430.121 L1515.22 434.056 L1501.14 434.056 L1501.14 442.528 Q1502.16 442.181 1503.18 442.019 Q1504.2 441.834 1505.22 441.834 Q1511.01 441.834 1514.39 445.005 Q1517.76 448.176 1517.76 453.593 Q1517.76 459.171 1514.29 462.273 Q1510.82 465.352 1504.5 465.352 Q1502.33 465.352 1500.06 464.982 Q1497.81 464.611 1495.4 463.871 L1495.4 459.171 Q1497.49 460.306 1499.71 460.861 Q1501.93 461.417 1504.41 461.417 Q1508.41 461.417 1510.75 459.31 Q1513.09 457.204 1513.09 453.593 Q1513.09 449.982 1510.75 447.875 Q1508.41 445.769 1504.41 445.769 Q1502.53 445.769 1500.66 446.185 Q1498.81 446.602 1496.86 447.482 L1496.86 430.121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1454.62 330.714 L1470.94 330.714 L1470.94 334.649 L1448.99 334.649 L1448.99 330.714 Q1451.65 327.96 1456.24 323.33 Q1460.84 318.677 1462.02 317.335 Q1464.27 314.812 1465.15 313.076 Q1466.05 311.316 1466.05 309.626 Q1466.05 306.872 1464.11 305.136 Q1462.19 303.4 1459.08 303.4 Q1456.89 303.4 1454.43 304.164 Q1452 304.927 1449.22 306.478 L1449.22 301.756 Q1452.05 300.622 1454.5 300.043 Q1456.96 299.464 1458.99 299.464 Q1464.36 299.464 1467.56 302.15 Q1470.75 304.835 1470.75 309.326 Q1470.75 311.455 1469.94 313.376 Q1469.15 315.275 1467.05 317.867 Q1466.47 318.538 1463.37 321.756 Q1460.27 324.95 1454.62 330.714 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1480.75 328.77 L1485.64 328.77 L1485.64 334.649 L1480.75 334.649 L1480.75 328.77 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 303.168 Q1502.21 303.168 1500.38 306.733 Q1498.58 310.275 1498.58 317.404 Q1498.58 324.511 1500.38 328.075 Q1502.21 331.617 1505.82 331.617 Q1509.45 331.617 1511.26 328.075 Q1513.09 324.511 1513.09 317.404 Q1513.09 310.275 1511.26 306.733 Q1509.45 303.168 1505.82 303.168 M1505.82 299.464 Q1511.63 299.464 1514.69 304.071 Q1517.76 308.654 1517.76 317.404 Q1517.76 326.131 1514.69 330.737 Q1511.63 335.321 1505.82 335.321 Q1500.01 335.321 1496.93 330.737 Q1493.88 326.131 1493.88 317.404 Q1493.88 308.654 1496.93 304.071 Q1500.01 299.464 1505.82 299.464 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1455.61 200.683 L1471.93 200.683 L1471.93 204.618 L1449.99 204.618 L1449.99 200.683 Q1452.65 197.928 1457.23 193.299 Q1461.84 188.646 1463.02 187.303 Q1465.27 184.78 1466.14 183.044 Q1467.05 181.285 1467.05 179.595 Q1467.05 176.841 1465.1 175.104 Q1463.18 173.368 1460.08 173.368 Q1457.88 173.368 1455.43 174.132 Q1453 174.896 1450.22 176.447 L1450.22 171.725 Q1453.04 170.591 1455.5 170.012 Q1457.95 169.433 1459.99 169.433 Q1465.36 169.433 1468.55 172.118 Q1471.75 174.804 1471.75 179.294 Q1471.75 181.424 1470.94 183.345 Q1470.15 185.243 1468.04 187.836 Q1467.46 188.507 1464.36 191.725 Q1461.26 194.919 1455.61 200.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1481.75 198.739 L1486.63 198.739 L1486.63 204.618 L1481.75 204.618 L1481.75 198.739 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1496.86 170.058 L1515.22 170.058 L1515.22 173.993 L1501.14 173.993 L1501.14 182.466 Q1502.16 182.118 1503.18 181.956 Q1504.2 181.771 1505.22 181.771 Q1511.01 181.771 1514.39 184.942 Q1517.76 188.114 1517.76 193.53 Q1517.76 199.109 1514.29 202.211 Q1510.82 205.29 1504.5 205.29 Q1502.33 205.29 1500.06 204.919 Q1497.81 204.549 1495.4 203.808 L1495.4 199.109 Q1497.49 200.243 1499.71 200.799 Q1501.93 201.354 1504.41 201.354 Q1508.41 201.354 1510.75 199.248 Q1513.09 197.141 1513.09 193.53 Q1513.09 189.919 1510.75 187.813 Q1508.41 185.706 1504.41 185.706 Q1502.53 185.706 1500.66 186.123 Q1498.81 186.54 1496.86 187.419 L1496.86 170.058 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1742.16 72.576 L1719.07 12.096 L1727.61 12.096 L1746.77 63.0159 L1765.98 12.096 L1774.48 12.096 L1751.43 72.576 L1742.16 72.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1815.36 48.0275 L1815.36 51.6733 L1781.09 51.6733 Q1781.57 59.3701 1785.7 63.421 Q1789.88 67.4314 1797.29 67.4314 Q1801.58 67.4314 1805.59 66.3781 Q1809.64 65.3249 1813.61 63.2184 L1813.61 70.267 Q1809.6 71.9684 1805.39 72.8596 Q1801.18 73.7508 1796.84 73.7508 Q1785.99 73.7508 1779.63 67.4314 Q1773.31 61.1119 1773.31 50.3365 Q1773.31 39.1965 1779.3 32.6746 Q1785.34 26.1121 1795.55 26.1121 Q1804.7 26.1121 1810.01 32.0264 Q1815.36 37.9003 1815.36 48.0275 M1807.9 45.84 Q1807.82 39.7232 1804.46 36.0774 Q1801.14 32.4315 1795.63 32.4315 Q1789.39 32.4315 1785.62 35.9558 Q1781.9 39.4801 1781.33 45.8805 L1807.9 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1827.59 9.54393 L1835.04 9.54393 L1835.04 72.576 L1827.59 72.576 L1827.59 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1868.22 32.4315 Q1862.23 32.4315 1858.74 37.1306 Q1855.26 41.7891 1855.26 49.9314 Q1855.26 58.0738 1858.7 62.7728 Q1862.19 67.4314 1868.22 67.4314 Q1874.18 67.4314 1877.66 62.7323 Q1881.14 58.0333 1881.14 49.9314 Q1881.14 41.8701 1877.66 37.1711 Q1874.18 32.4315 1868.22 32.4315 M1868.22 26.1121 Q1877.94 26.1121 1883.49 32.4315 Q1889.04 38.7509 1889.04 49.9314 Q1889.04 61.0714 1883.49 67.4314 Q1877.94 73.7508 1868.22 73.7508 Q1858.46 73.7508 1852.91 67.4314 Q1847.4 61.0714 1847.4 49.9314 Q1847.4 38.7509 1852.91 32.4315 Q1858.46 26.1121 1868.22 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1934.05 28.9478 L1934.05 35.9153 Q1930.89 34.1734 1927.69 33.3227 Q1924.53 32.4315 1921.29 32.4315 Q1914.04 32.4315 1910.03 37.0496 Q1906.02 41.6271 1906.02 49.9314 Q1906.02 58.2358 1910.03 62.8538 Q1914.04 67.4314 1921.29 67.4314 Q1924.53 67.4314 1927.69 66.5807 Q1930.89 65.6895 1934.05 63.9476 L1934.05 70.8341 Q1930.93 72.2924 1927.57 73.0216 Q1924.24 73.7508 1920.48 73.7508 Q1910.23 73.7508 1904.19 67.3098 Q1898.16 60.8689 1898.16 49.9314 Q1898.16 38.832 1904.23 32.472 Q1910.35 26.1121 1920.96 26.1121 Q1924.41 26.1121 1927.69 26.8413 Q1930.97 27.5299 1934.05 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1947.01 27.2059 L1954.46 27.2059 L1954.46 72.576 L1947.01 72.576 L1947.01 27.2059 M1947.01 9.54393 L1954.46 9.54393 L1954.46 18.9825 L1947.01 18.9825 L1947.01 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1977.43 14.324 L1977.43 27.2059 L1992.79 27.2059 L1992.79 32.9987 L1977.43 32.9987 L1977.43 57.6282 Q1977.43 63.1779 1978.93 64.7578 Q1980.47 66.3376 1985.13 66.3376 L1992.79 66.3376 L1992.79 72.576 L1985.13 72.576 Q1976.5 72.576 1973.22 69.3758 Q1969.94 66.1351 1969.94 57.6282 L1969.94 32.9987 L1964.47 32.9987 L1964.47 27.2059 L1969.94 27.2059 L1969.94 14.324 L1977.43 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2021.47 76.7889 Q2018.31 84.8907 2015.31 87.3618 Q2012.31 89.8329 2007.29 89.8329 L2001.33 89.8329 L2001.33 83.5945 L2005.71 83.5945 Q2008.79 83.5945 2010.49 82.1361 Q2012.19 80.6778 2014.26 75.2496 L2015.59 71.8468 L1997.24 27.2059 L2005.14 27.2059 L2019.32 62.6918 L2033.5 27.2059 L2041.4 27.2059 L2021.47 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2095.96 9.62495 Q2090.53 18.942 2087.9 28.0566 Q2085.27 37.1711 2085.27 46.5287 Q2085.27 55.8863 2087.9 65.0818 Q2090.57 74.2369 2095.96 83.5134 L2089.48 83.5134 Q2083.4 73.9938 2080.37 64.7983 Q2077.37 55.6027 2077.37 46.5287 Q2077.37 37.4952 2080.37 28.3401 Q2083.36 19.1851 2089.48 9.62495 L2095.96 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2112.9 65.6895 L2126.26 65.6895 L2126.26 19.5497 L2111.72 22.4663 L2111.72 15.0127 L2126.18 12.096 L2134.37 12.096 L2134.37 65.6895 L2147.73 65.6895 L2147.73 72.576 L2112.9 72.576 L2112.9 65.6895 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2181.76 17.4837 Q2175.44 17.4837 2172.24 23.7221 Q2169.08 29.92 2169.08 42.3968 Q2169.08 54.833 2172.24 61.0714 Q2175.44 67.2693 2181.76 67.2693 Q2188.12 67.2693 2191.28 61.0714 Q2194.48 54.833 2194.48 42.3968 Q2194.48 29.92 2191.28 23.7221 Q2188.12 17.4837 2181.76 17.4837 M2181.76 11.0023 Q2191.93 11.0023 2197.28 19.0636 Q2202.66 27.0843 2202.66 42.3968 Q2202.66 57.6687 2197.28 65.73 Q2191.93 73.7508 2181.76 73.7508 Q2171.59 73.7508 2166.21 65.73 Q2160.86 57.6687 2160.86 42.3968 Q2160.86 27.0843 2166.21 19.0636 Q2171.59 11.0023 2181.76 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2234.54 17.4837 Q2228.22 17.4837 2225.02 23.7221 Q2221.86 29.92 2221.86 42.3968 Q2221.86 54.833 2225.02 61.0714 Q2228.22 67.2693 2234.54 67.2693 Q2240.9 67.2693 2244.06 61.0714 Q2247.26 54.833 2247.26 42.3968 Q2247.26 29.92 2244.06 23.7221 Q2240.9 17.4837 2234.54 17.4837 M2234.54 11.0023 Q2244.71 11.0023 2250.06 19.0636 Q2255.45 27.0843 2255.45 42.3968 Q2255.45 57.6687 2250.06 65.73 Q2244.71 73.7508 2234.54 73.7508 Q2224.38 73.7508 2218.99 65.73 Q2213.64 57.6687 2213.64 42.3968 Q2213.64 27.0843 2218.99 19.0636 Q2224.38 11.0023 2234.54 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2287.33 17.4837 Q2281.01 17.4837 2277.81 23.7221 Q2274.65 29.92 2274.65 42.3968 Q2274.65 54.833 2277.81 61.0714 Q2281.01 67.2693 2287.33 67.2693 Q2293.69 67.2693 2296.85 61.0714 Q2300.05 54.833 2300.05 42.3968 Q2300.05 29.92 2296.85 23.7221 Q2293.69 17.4837 2287.33 17.4837 M2287.33 11.0023 Q2297.5 11.0023 2302.84 19.0636 Q2308.23 27.0843 2308.23 42.3968 Q2308.23 57.6687 2302.84 65.73 Q2297.5 73.7508 2287.33 73.7508 Q2277.16 73.7508 2271.77 65.73 Q2266.42 57.6687 2266.42 42.3968 Q2266.42 27.0843 2271.77 19.0636 Q2277.16 11.0023 2287.33 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2370.9 9.54393 L2370.9 15.7418 L2363.77 15.7418 Q2359.76 15.7418 2358.18 17.3622 Q2356.64 18.9825 2356.64 23.1955 L2356.64 27.2059 L2368.91 27.2059 L2368.91 32.9987 L2356.64 32.9987 L2356.64 72.576 L2349.14 72.576 L2349.14 32.9987 L2342.01 32.9987 L2342.01 27.2059 L2349.14 27.2059 L2349.14 24.0462 Q2349.14 16.471 2352.67 13.0277 Q2356.19 9.54393 2363.85 9.54393 L2370.9 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2383.05 14.324 L2383.05 27.2059 L2398.4 27.2059 L2398.4 32.9987 L2383.05 32.9987 L2383.05 57.6282 Q2383.05 63.1779 2384.55 64.7578 Q2386.09 66.3376 2390.75 66.3376 L2398.4 66.3376 L2398.4 72.576 L2390.75 72.576 Q2382.12 72.576 2378.84 69.3758 Q2375.56 66.1351 2375.56 57.6282 L2375.56 32.9987 L2370.09 32.9987 L2370.09 27.2059 L2375.56 27.2059 L2375.56 14.324 L2383.05 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2421.45 12.096 L2428.34 12.096 L2407.27 80.2727 L2400.39 80.2727 L2421.45 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2465.08 28.5427 L2465.08 35.5912 Q2461.92 33.9709 2458.52 33.1607 Q2455.12 32.3505 2451.47 32.3505 Q2445.92 32.3505 2443.13 34.0519 Q2440.37 35.7533 2440.37 39.156 Q2440.37 41.7486 2442.36 43.2475 Q2444.34 44.7058 2450.34 46.0426 L2452.89 46.6097 Q2460.83 48.3111 2464.15 51.4303 Q2467.51 54.509 2467.51 60.0587 Q2467.51 66.3781 2462.49 70.0644 Q2457.51 73.7508 2448.76 73.7508 Q2445.11 73.7508 2441.14 73.0216 Q2437.21 72.3329 2432.84 70.9151 L2432.84 63.2184 Q2436.97 65.3654 2440.98 66.4591 Q2444.99 67.5124 2448.92 67.5124 Q2454.18 67.5124 2457.02 65.73 Q2459.86 63.9071 2459.86 60.6258 Q2459.86 57.5877 2457.79 55.9673 Q2455.76 54.3469 2448.84 52.8481 L2446.24 52.2405 Q2439.32 50.7821 2436.24 47.7845 Q2433.16 44.7463 2433.16 39.4801 Q2433.16 33.0797 2437.7 29.5959 Q2442.23 26.1121 2450.58 26.1121 Q2454.71 26.1121 2458.36 26.7198 Q2462 27.3274 2465.08 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2518.19 48.0275 L2518.19 51.6733 L2483.92 51.6733 Q2484.4 59.3701 2488.54 63.421 Q2492.71 67.4314 2500.12 67.4314 Q2504.42 67.4314 2508.43 66.3781 Q2512.48 65.3249 2516.45 63.2184 L2516.45 70.267 Q2512.44 71.9684 2508.22 72.8596 Q2504.01 73.7508 2499.68 73.7508 Q2488.82 73.7508 2482.46 67.4314 Q2476.14 61.1119 2476.14 50.3365 Q2476.14 39.1965 2482.14 32.6746 Q2488.17 26.1121 2498.38 26.1121 Q2507.53 26.1121 2512.84 32.0264 Q2518.19 37.9003 2518.19 48.0275 M2510.73 45.84 Q2510.65 39.7232 2507.29 36.0774 Q2503.97 32.4315 2498.46 32.4315 Q2492.22 32.4315 2488.45 35.9558 Q2484.73 39.4801 2484.16 45.8805 L2510.73 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2563.07 28.9478 L2563.07 35.9153 Q2559.91 34.1734 2556.71 33.3227 Q2553.55 32.4315 2550.31 32.4315 Q2543.06 32.4315 2539.05 37.0496 Q2535.04 41.6271 2535.04 49.9314 Q2535.04 58.2358 2539.05 62.8538 Q2543.06 67.4314 2550.31 67.4314 Q2553.55 67.4314 2556.71 66.5807 Q2559.91 65.6895 2563.07 63.9476 L2563.07 70.8341 Q2559.95 72.2924 2556.59 73.0216 Q2553.27 73.7508 2549.5 73.7508 Q2539.25 73.7508 2533.22 67.3098 Q2527.18 60.8689 2527.18 49.9314 Q2527.18 38.832 2533.26 32.472 Q2539.37 26.1121 2549.99 26.1121 Q2553.43 26.1121 2556.71 26.8413 Q2559.99 27.5299 2563.07 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2574.86 9.62495 L2581.34 9.62495 Q2587.42 19.1851 2590.42 28.3401 Q2593.45 37.4952 2593.45 46.5287 Q2593.45 55.6027 2590.42 64.7983 Q2587.42 73.9938 2581.34 83.5134 L2574.86 83.5134 Q2580.25 74.2369 2582.88 65.0818 Q2585.55 55.8863 2585.55 46.5287 Q2585.55 37.1711 2582.88 28.0566 Q2580.25 18.942 2574.86 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip213)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,171.734 1595.77,171.777 1598.01,171.828 1600.25,171.887 1602.5,171.956 1604.74,172.034 1606.98,172.123 1609.22,172.223 1611.46,172.335 1613.71,172.461 1615.95,172.601 1618.19,172.756 1620.43,172.927 1622.67,173.116 1624.92,173.324 1627.16,173.551 1629.4,173.8 1631.64,174.072 1633.88,174.367 1636.13,174.688 1638.37,175.035 1640.61,175.411 1642.85,175.817 1645.09,176.254 1647.34,176.724 1649.58,177.228 1651.82,177.768 1654.06,178.344 1656.3,178.958 1658.55,179.61 1660.79,180.303 1663.03,181.035 1665.27,181.808 1667.51,182.621 1669.76,183.475 1672,184.369 1674.24,185.302 1676.48,186.273 1678.72,187.281 1680.97,188.323 1683.21,189.399 1685.45,190.504 1687.69,191.637 1689.93,192.795 1692.17,193.974 1694.42,195.171 1696.66,196.383 1698.9,197.605 1701.14,198.835 1703.38,200.068 1705.63,201.301 1707.87,202.531 1710.11,203.754 1712.35,204.967 1714.59,206.167 1716.84,207.352 1719.08,208.519 1721.32,209.666 1723.56,210.791 1725.8,211.893 1728.05,212.97 1730.29,214.022 1732.53,215.048 1734.77,216.047 1737.01,217.018 1739.26,217.963 1741.5,218.88 1743.74,219.771 1745.98,220.635 1748.22,221.474 1750.47,222.287 1752.71,223.076 1754.95,223.841 1757.19,224.583 1759.43,225.304 1761.68,226.003 1763.92,226.682 1766.16,227.342 1768.4,227.983 1770.64,228.608 1772.89,229.216 1775.13,229.809 1777.37,230.387 1779.61,230.953 1781.85,231.505 1784.1,232.046 1786.34,232.577 1788.58,233.098 1790.82,233.61 1793.06,234.113 1795.31,234.61 1797.55,235.1 1799.79,235.585 1802.03,236.065 1804.27,236.542 1806.51,237.015 1808.76,237.487 1811,237.957 1813.24,238.426 1815.48,238.896 1817.72,239.368 1819.97,239.841 1822.21,240.318 1824.45,240.798 1826.69,241.283 1828.93,241.775 1831.18,242.273 1833.42,242.779 1835.66,243.294 1837.9,243.819 1840.14,244.355 1842.39,244.903 1844.63,245.465 1846.87,246.042 1849.11,246.634 1851.35,247.243 1853.6,247.87 1855.84,248.517 1858.08,249.185 1860.32,249.874 1862.56,250.587 1864.81,251.324 1867.05,252.086 1869.29,252.875 1871.53,253.692 1873.77,254.536 1876.02,255.41 1878.26,256.314 1880.5,257.247 1882.74,258.211 1884.98,259.205 1887.23,260.229 1889.47,261.283 1891.71,262.365 1893.95,263.476 1896.19,264.612 1898.44,265.774 1900.68,266.96 1902.92,268.166 1905.16,269.391 1907.4,270.633 1909.64,271.888 1911.89,273.154 1914.13,274.429 1916.37,275.709 1918.61,276.991 1920.85,278.273 1923.1,279.552 1925.34,280.826 1927.58,282.091 1929.82,283.346 1932.06,284.589 1934.31,285.817 1936.55,287.03 1938.79,288.226 1941.03,289.403 1943.27,290.562 1945.52,291.701 1947.76,292.821 1950,293.92 1952.24,295 1954.48,296.06 1956.73,297.1 1958.97,298.122 1961.21,299.126 1963.45,300.113 1965.69,301.083 1967.94,302.038 1970.18,302.978 1972.42,303.905 1974.66,304.82 1976.9,305.724 1979.15,306.619 1981.39,307.505 1983.63,308.385 1985.87,309.259 1988.11,310.128 1990.36,310.995 1992.6,311.861 1994.84,312.726 1997.08,313.594 1999.32,314.464 2001.57,315.339 2003.81,316.219 2006.05,317.108 2008.29,318.005 2010.53,318.912 2012.78,319.832 2015.02,320.764 2017.26,321.712 2019.5,322.675 2021.74,323.656 2023.98,324.655 2026.23,325.674 2028.47,326.713 2030.71,327.774 2032.95,328.857 2035.19,329.963 2037.44,331.093 2039.68,332.246 2041.92,333.422 2044.16,334.621 2046.4,335.843 2048.65,337.088 2050.89,338.353 2053.13,339.638 2055.37,340.942 2057.61,342.263 2059.86,343.599 2062.1,344.948 2064.34,346.307 2066.58,347.676 2068.82,349.051 2071.07,350.43 2073.31,351.811 2075.55,353.192 2077.79,354.57 2080.03,355.943 2082.28,357.31 2084.52,358.668 2086.76,360.017 2089,361.355 2091.24,362.68 2093.49,363.992 2095.73,365.29 2097.97,366.574 2100.21,367.844 2102.45,369.099 2104.7,370.339 2106.94,371.566 2109.18,372.779 2111.42,373.979 2113.66,375.166 2115.91,376.343 2118.15,377.51 2120.39,378.667 2122.63,379.816 2124.87,380.959 2127.11,382.097 2129.36,383.231 2131.6,384.362 2133.84,385.492 2136.08,386.623 2138.32,387.756 2140.57,388.892 2142.81,390.033 2145.05,391.181 2147.29,392.336 2149.53,393.501 2151.78,394.676 2154.02,395.863 2156.26,397.063 2158.5,398.277 2160.74,399.506 2162.99,400.75 2165.23,402.012 2167.47,403.29 2169.71,404.586 2171.95,405.899 2174.2,407.229 2176.44,408.577 2178.68,409.942 2180.92,411.322 2183.16,412.718 2185.41,414.128 2187.65,415.551 2189.89,416.986 2192.13,418.43 2194.37,419.883 2196.62,421.343 2198.86,422.808 2201.1,424.276 2203.34,425.746 2205.58,427.215 2207.83,428.683 2210.07,430.147 2212.31,431.606 2214.55,433.059 2216.79,434.505 2219.04,435.943 2221.28,437.372 2223.52,438.792 2225.76,440.202 2228,441.602 2230.25,442.993 2232.49,444.373 2234.73,445.744 2236.97,447.106 2239.21,448.459 2241.45,449.805 2243.7,451.144 2245.94,452.477 2248.18,453.806 2250.42,455.13 2252.66,456.452 2254.91,457.772 2257.15,459.092 2259.39,460.414 2261.63,461.737 2263.87,463.064 2266.12,464.396 2268.36,465.734 2270.6,467.078 2272.84,468.431 2275.08,469.793 2277.33,471.164 2279.57,472.546 2281.81,473.94 2284.05,475.345 2286.29,476.763 2288.54,478.192 2290.78,479.634 2293.02,481.088 2295.26,482.555 2297.5,484.032 2299.75,485.521 2301.99,487.02 2304.23,488.528 2306.47,490.044 2308.71,491.568 2310.96,493.097 2313.2,494.632 2315.44,496.17 2317.68,497.71 2319.92,499.252 2322.17,500.793 2324.41,502.334 2326.65,503.872 2328.89,505.406 2331.13,506.937 2333.38,508.463 2335.62,509.984 2337.86,511.498 2340.1,513.007 2342.34,514.509 2344.58,516.005 2346.83,517.495 2349.07,518.979 2351.31,520.457 2353.55,521.93 2355.79,523.399 2358.04,524.863 2360.28,526.324 2362.52,527.783 2364.76,529.24 2367,530.697 2369.25,532.154 2371.49,533.611 2373.73,535.071 2375.97,536.534 2378.21,538 2380.46,539.471 2382.7,540.947 2384.94,542.429 2387.18,543.918 2389.42,545.413 2391.67,546.917 2393.91,548.428 2396.15,549.947 2398.39,551.474 2400.63,553.009 2402.88,554.552 2405.12,556.103 2407.36,557.661 2409.6,559.226 2411.84,560.797 2414.09,562.373 2416.33,563.955 2418.57,565.54 2420.81,567.129 2423.05,568.721 2425.3,570.314 2427.54,571.908 2429.78,573.501 2432.02,575.095 2434.26,576.686 2436.51,578.276 2438.75,579.864 2440.99,581.448 2443.23,583.029 2445.47,584.606 2447.72,586.18 2449.96,587.749 2452.2,589.315 2454.44,590.877 2456.68,592.435 2458.92,593.991 2461.17,595.543 2463.41,597.093 2465.65,598.64 2467.89,600.187 2470.13,601.732 2472.38,603.276 2474.62,604.821 2476.86,606.367 2479.1,607.914 2481.34,609.463 2483.59,611.014 2485.83,612.569 2488.07,614.126 2490.31,615.688 2492.55,617.253 2494.8,618.823 2497.04,620.398 2499.28,621.977 2501.52,623.561 2503.76,625.15 2506.01,626.743 2508.25,628.341 2510.49,629.942 2512.73,631.548 2514.97,633.157 2517.22,634.769 2519.46,636.384 2521.7,638 2523.94,639.618 2526.18,641.237 2528.43,642.856 2530.67,644.474 2532.91,646.093 2535.15,647.709 2537.39,649.324 2539.64,650.937 2541.88,652.548 2544.12,654.155 2546.36,655.76 2548.6,657.361 2550.85,658.959 2553.09,660.553 2555.33,662.144 2557.57,663.732 2559.81,665.316 2562.05,666.897 2564.3,668.475 2566.54,670.051 2568.78,671.624 2571.02,673.195 2573.26,674.765 2575.51,676.334 2577.75,677.901 2579.99,679.468 2582.23,681.036 2584.47,682.603 2586.72,684.172 2588.96,685.741 2591.2,687.312 2593.44,688.885 2595.68,690.459 2597.93,692.036 2600.17,693.615 2602.41,695.197 2604.65,696.781 2606.89,698.367 2609.14,699.956 2611.38,701.546 2613.62,703.138 2615.86,704.732 2618.1,706.326 2620.35,707.921 2622.59,709.516 2624.83,711.111 2627.07,712.704 2629.31,714.295 2631.56,715.883 2633.8,717.468 2636.04,719.049 2638.28,720.624 2640.52,722.194 2642.77,723.757 2645.01,725.312 2647.25,726.859 2649.49,728.397 2651.73,729.926 2653.98,731.444 2656.22,732.952 2658.46,734.448 2660.7,735.932 2662.94,737.404 2665.19,738.865 2667.43,740.313 2669.67,741.748 2671.91,743.172 2674.15,744.584 2676.39,745.984 2678.64,747.374 2680.88,748.755 2683.12,750.126 2685.36,751.49 2687.6,752.848 2689.85,754.202 2692.09,755.553 2694.33,756.904 2696.57,758.256 2698.81,759.613 2701.06,760.977 2703.3,762.351 2705.54,763.738 2707.78,765.14 2710.02,766.561 2712.27,768.004 2714.51,769.469 2716.75,770.961 2718.99,772.479 "/>
-<path clip-path="url(#clip210)" d="M159.765 1723.83 L1352.76 1723.83 L1352.76 1087.05 L159.765 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,790.501 1593.53,153.712 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,790.501 1873.77,153.712 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,790.501 2154.02,153.712 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,790.501 2434.26,153.712 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,790.501 2714.51,153.712 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,790.501 2752.76,790.501 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,790.501 1593.53,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,790.501 1873.77,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,790.501 2154.02,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,790.501 2434.26,771.604 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,790.501 2714.51,771.604 "/>
+<path clip-path="url(#clip030)" d="M1593.53 824.62 Q1589.92 824.62 1588.09 828.185 Q1586.28 831.726 1586.28 838.856 Q1586.28 845.962 1588.09 849.527 Q1589.92 853.069 1593.53 853.069 Q1597.16 853.069 1598.97 849.527 Q1600.8 845.962 1600.8 838.856 Q1600.8 831.726 1598.97 828.185 Q1597.16 824.62 1593.53 824.62 M1593.53 820.916 Q1599.34 820.916 1602.39 825.523 Q1605.47 830.106 1605.47 838.856 Q1605.47 847.583 1602.39 852.189 Q1599.34 856.773 1593.53 856.773 Q1587.72 856.773 1584.64 852.189 Q1581.58 847.583 1581.58 838.856 Q1581.58 830.106 1584.64 825.523 Q1587.72 820.916 1593.53 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1833.39 821.541 L1851.75 821.541 L1851.75 825.476 L1837.67 825.476 L1837.67 833.949 Q1838.69 833.601 1839.71 833.439 Q1840.73 833.254 1841.75 833.254 Q1847.54 833.254 1850.91 836.425 Q1854.29 839.597 1854.29 845.013 Q1854.29 850.592 1850.82 853.694 Q1847.35 856.773 1841.03 856.773 Q1838.85 856.773 1836.59 856.402 Q1834.34 856.032 1831.93 855.291 L1831.93 850.592 Q1834.02 851.726 1836.24 852.282 Q1838.46 852.837 1840.94 852.837 Q1844.94 852.837 1847.28 850.731 Q1849.62 848.624 1849.62 845.013 Q1849.62 841.402 1847.28 839.296 Q1844.94 837.189 1840.94 837.189 Q1839.06 837.189 1837.19 837.606 Q1835.34 838.023 1833.39 838.902 L1833.39 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1873.51 824.62 Q1869.9 824.62 1868.07 828.185 Q1866.26 831.726 1866.26 838.856 Q1866.26 845.962 1868.07 849.527 Q1869.9 853.069 1873.51 853.069 Q1877.14 853.069 1878.95 849.527 Q1880.78 845.962 1880.78 838.856 Q1880.78 831.726 1878.95 828.185 Q1877.14 824.62 1873.51 824.62 M1873.51 820.916 Q1879.32 820.916 1882.37 825.523 Q1885.45 830.106 1885.45 838.856 Q1885.45 847.583 1882.37 852.189 Q1879.32 856.773 1873.51 856.773 Q1867.7 856.773 1864.62 852.189 Q1861.56 847.583 1861.56 838.856 Q1861.56 830.106 1864.62 825.523 Q1867.7 820.916 1873.51 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1903.67 824.62 Q1900.06 824.62 1898.23 828.185 Q1896.42 831.726 1896.42 838.856 Q1896.42 845.962 1898.23 849.527 Q1900.06 853.069 1903.67 853.069 Q1907.3 853.069 1909.11 849.527 Q1910.94 845.962 1910.94 838.856 Q1910.94 831.726 1909.11 828.185 Q1907.3 824.62 1903.67 824.62 M1903.67 820.916 Q1909.48 820.916 1912.53 825.523 Q1915.61 830.106 1915.61 838.856 Q1915.61 847.583 1912.53 852.189 Q1909.48 856.773 1903.67 856.773 Q1897.86 856.773 1894.78 852.189 Q1891.72 847.583 1891.72 838.856 Q1891.72 830.106 1894.78 825.523 Q1897.86 820.916 1903.67 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2098.54 852.166 L2106.18 852.166 L2106.18 825.8 L2097.87 827.467 L2097.87 823.208 L2106.14 821.541 L2110.81 821.541 L2110.81 852.166 L2118.45 852.166 L2118.45 856.101 L2098.54 856.101 L2098.54 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2137.9 824.62 Q2134.28 824.62 2132.46 828.185 Q2130.65 831.726 2130.65 838.856 Q2130.65 845.962 2132.46 849.527 Q2134.28 853.069 2137.9 853.069 Q2141.53 853.069 2143.34 849.527 Q2145.16 845.962 2145.16 838.856 Q2145.16 831.726 2143.34 828.185 Q2141.53 824.62 2137.9 824.62 M2137.9 820.916 Q2143.71 820.916 2146.76 825.523 Q2149.84 830.106 2149.84 838.856 Q2149.84 847.583 2146.76 852.189 Q2143.71 856.773 2137.9 856.773 Q2132.09 856.773 2129.01 852.189 Q2125.95 847.583 2125.95 838.856 Q2125.95 830.106 2129.01 825.523 Q2132.09 820.916 2137.9 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2168.06 824.62 Q2164.45 824.62 2162.62 828.185 Q2160.81 831.726 2160.81 838.856 Q2160.81 845.962 2162.62 849.527 Q2164.45 853.069 2168.06 853.069 Q2171.69 853.069 2173.5 849.527 Q2175.33 845.962 2175.33 838.856 Q2175.33 831.726 2173.5 828.185 Q2171.69 824.62 2168.06 824.62 M2168.06 820.916 Q2173.87 820.916 2176.92 825.523 Q2180 830.106 2180 838.856 Q2180 847.583 2176.92 852.189 Q2173.87 856.773 2168.06 856.773 Q2162.25 856.773 2159.17 852.189 Q2156.11 847.583 2156.11 838.856 Q2156.11 830.106 2159.17 825.523 Q2162.25 820.916 2168.06 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2198.22 824.62 Q2194.61 824.62 2192.78 828.185 Q2190.97 831.726 2190.97 838.856 Q2190.97 845.962 2192.78 849.527 Q2194.61 853.069 2198.22 853.069 Q2201.85 853.069 2203.66 849.527 Q2205.49 845.962 2205.49 838.856 Q2205.49 831.726 2203.66 828.185 Q2201.85 824.62 2198.22 824.62 M2198.22 820.916 Q2204.03 820.916 2207.09 825.523 Q2210.16 830.106 2210.16 838.856 Q2210.16 847.583 2207.09 852.189 Q2204.03 856.773 2198.22 856.773 Q2192.41 856.773 2189.33 852.189 Q2186.28 847.583 2186.28 838.856 Q2186.28 830.106 2189.33 825.523 Q2192.41 820.916 2198.22 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2378.79 852.166 L2386.43 852.166 L2386.43 825.8 L2378.12 827.467 L2378.12 823.208 L2386.38 821.541 L2391.06 821.541 L2391.06 852.166 L2398.7 852.166 L2398.7 856.101 L2378.79 856.101 L2378.79 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2408.19 821.541 L2426.54 821.541 L2426.54 825.476 L2412.47 825.476 L2412.47 833.949 Q2413.49 833.601 2414.51 833.439 Q2415.52 833.254 2416.54 833.254 Q2422.33 833.254 2425.71 836.425 Q2429.09 839.597 2429.09 845.013 Q2429.09 850.592 2425.62 853.694 Q2422.15 856.773 2415.83 856.773 Q2413.65 856.773 2411.38 856.402 Q2409.14 856.032 2406.73 855.291 L2406.73 850.592 Q2408.81 851.726 2411.03 852.282 Q2413.26 852.837 2415.73 852.837 Q2419.74 852.837 2422.08 850.731 Q2424.41 848.624 2424.41 845.013 Q2424.41 841.402 2422.08 839.296 Q2419.74 837.189 2415.73 837.189 Q2413.86 837.189 2411.98 837.606 Q2410.13 838.023 2408.19 838.902 L2408.19 821.541 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2448.3 824.62 Q2444.69 824.62 2442.86 828.185 Q2441.06 831.726 2441.06 838.856 Q2441.06 845.962 2442.86 849.527 Q2444.69 853.069 2448.3 853.069 Q2451.94 853.069 2453.74 849.527 Q2455.57 845.962 2455.57 838.856 Q2455.57 831.726 2453.74 828.185 Q2451.94 824.62 2448.3 824.62 M2448.3 820.916 Q2454.11 820.916 2457.17 825.523 Q2460.25 830.106 2460.25 838.856 Q2460.25 847.583 2457.17 852.189 Q2454.11 856.773 2448.3 856.773 Q2442.49 856.773 2439.41 852.189 Q2436.36 847.583 2436.36 838.856 Q2436.36 830.106 2439.41 825.523 Q2442.49 820.916 2448.3 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2478.46 824.62 Q2474.85 824.62 2473.02 828.185 Q2471.22 831.726 2471.22 838.856 Q2471.22 845.962 2473.02 849.527 Q2474.85 853.069 2478.46 853.069 Q2482.1 853.069 2483.9 849.527 Q2485.73 845.962 2485.73 838.856 Q2485.73 831.726 2483.9 828.185 Q2482.1 824.62 2478.46 824.62 M2478.46 820.916 Q2484.27 820.916 2487.33 825.523 Q2490.41 830.106 2490.41 838.856 Q2490.41 847.583 2487.33 852.189 Q2484.27 856.773 2478.46 856.773 Q2472.65 856.773 2469.58 852.189 Q2466.52 847.583 2466.52 838.856 Q2466.52 830.106 2469.58 825.523 Q2472.65 820.916 2478.46 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2663.12 852.166 L2679.44 852.166 L2679.44 856.101 L2657.49 856.101 L2657.49 852.166 Q2660.16 849.411 2664.74 844.782 Q2669.35 840.129 2670.53 838.786 Q2672.77 836.263 2673.65 834.527 Q2674.55 832.768 2674.55 831.078 Q2674.55 828.324 2672.61 826.587 Q2670.69 824.851 2667.59 824.851 Q2665.39 824.851 2662.93 825.615 Q2660.5 826.379 2657.73 827.93 L2657.73 823.208 Q2660.55 822.074 2663 821.495 Q2665.46 820.916 2667.49 820.916 Q2672.86 820.916 2676.06 823.601 Q2679.25 826.287 2679.25 830.777 Q2679.25 832.907 2678.44 834.828 Q2677.66 836.726 2675.55 839.319 Q2674.97 839.99 2671.87 843.208 Q2668.77 846.402 2663.12 852.166 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2699.25 824.62 Q2695.64 824.62 2693.81 828.185 Q2692.01 831.726 2692.01 838.856 Q2692.01 845.962 2693.81 849.527 Q2695.64 853.069 2699.25 853.069 Q2702.89 853.069 2704.69 849.527 Q2706.52 845.962 2706.52 838.856 Q2706.52 831.726 2704.69 828.185 Q2702.89 824.62 2699.25 824.62 M2699.25 820.916 Q2705.06 820.916 2708.12 825.523 Q2711.2 830.106 2711.2 838.856 Q2711.2 847.583 2708.12 852.189 Q2705.06 856.773 2699.25 856.773 Q2693.44 856.773 2690.36 852.189 Q2687.31 847.583 2687.31 838.856 Q2687.31 830.106 2690.36 825.523 Q2693.44 820.916 2699.25 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2729.42 824.62 Q2725.8 824.62 2723.98 828.185 Q2722.17 831.726 2722.17 838.856 Q2722.17 845.962 2723.98 849.527 Q2725.8 853.069 2729.42 853.069 Q2733.05 853.069 2734.86 849.527 Q2736.68 845.962 2736.68 838.856 Q2736.68 831.726 2734.86 828.185 Q2733.05 824.62 2729.42 824.62 M2729.42 820.916 Q2735.23 820.916 2738.28 825.523 Q2741.36 830.106 2741.36 838.856 Q2741.36 847.583 2738.28 852.189 Q2735.23 856.773 2729.42 856.773 Q2723.61 856.773 2720.53 852.189 Q2717.47 847.583 2717.47 838.856 Q2717.47 830.106 2720.53 825.523 Q2723.61 820.916 2729.42 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2759.58 824.62 Q2755.97 824.62 2754.14 828.185 Q2752.33 831.726 2752.33 838.856 Q2752.33 845.962 2754.14 849.527 Q2755.97 853.069 2759.58 853.069 Q2763.21 853.069 2765.02 849.527 Q2766.85 845.962 2766.85 838.856 Q2766.85 831.726 2765.02 828.185 Q2763.21 824.62 2759.58 824.62 M2759.58 820.916 Q2765.39 820.916 2768.44 825.523 Q2771.52 830.106 2771.52 838.856 Q2771.52 847.583 2768.44 852.189 Q2765.39 856.773 2759.58 856.773 Q2753.77 856.773 2750.69 852.189 Q2747.63 847.583 2747.63 838.856 Q2747.63 830.106 2750.69 825.523 Q2753.77 820.916 2759.58 820.916 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,707.463 2752.76,707.463 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,577.432 2752.76,577.432 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,447.401 2752.76,447.401 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,317.369 2752.76,317.369 "/>
+<polyline clip-path="url(#clip033)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,187.338 2752.76,187.338 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,790.501 1559.76,153.712 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,707.463 1578.66,707.463 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,577.432 1578.66,577.432 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,447.401 1578.66,447.401 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,317.369 1578.66,317.369 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,187.338 1578.66,187.338 "/>
+<path clip-path="url(#clip030)" d="M1461.58 693.262 Q1457.97 693.262 1456.14 696.827 Q1454.34 700.368 1454.34 707.498 Q1454.34 714.604 1456.14 718.169 Q1457.97 721.711 1461.58 721.711 Q1465.22 721.711 1467.02 718.169 Q1468.85 714.604 1468.85 707.498 Q1468.85 700.368 1467.02 696.827 Q1465.22 693.262 1461.58 693.262 M1461.58 689.558 Q1467.39 689.558 1470.45 694.165 Q1473.53 698.748 1473.53 707.498 Q1473.53 716.225 1470.45 720.831 Q1467.39 725.415 1461.58 725.415 Q1455.77 725.415 1452.7 720.831 Q1449.64 716.225 1449.64 707.498 Q1449.64 698.748 1452.7 694.165 Q1455.77 689.558 1461.58 689.558 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1481.75 718.864 L1486.63 718.864 L1486.63 724.743 L1481.75 724.743 L1481.75 718.864 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1496.86 690.183 L1515.22 690.183 L1515.22 694.118 L1501.14 694.118 L1501.14 702.591 Q1502.16 702.243 1503.18 702.081 Q1504.2 701.896 1505.22 701.896 Q1511.01 701.896 1514.39 705.067 Q1517.76 708.239 1517.76 713.655 Q1517.76 719.234 1514.29 722.336 Q1510.82 725.415 1504.5 725.415 Q1502.33 725.415 1500.06 725.044 Q1497.81 724.674 1495.4 723.933 L1495.4 719.234 Q1497.49 720.368 1499.71 720.924 Q1501.93 721.479 1504.41 721.479 Q1508.41 721.479 1510.75 719.373 Q1513.09 717.266 1513.09 713.655 Q1513.09 710.044 1510.75 707.938 Q1508.41 705.831 1504.41 705.831 Q1502.53 705.831 1500.66 706.248 Q1498.81 706.665 1496.86 707.544 L1496.86 690.183 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1451.4 590.777 L1459.04 590.777 L1459.04 564.411 L1450.73 566.078 L1450.73 561.819 L1458.99 560.152 L1463.67 560.152 L1463.67 590.777 L1471.31 590.777 L1471.31 594.712 L1451.4 594.712 L1451.4 590.777 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1480.75 588.832 L1485.64 588.832 L1485.64 594.712 L1480.75 594.712 L1480.75 588.832 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 563.231 Q1502.21 563.231 1500.38 566.795 Q1498.58 570.337 1498.58 577.467 Q1498.58 584.573 1500.38 588.138 Q1502.21 591.68 1505.82 591.68 Q1509.45 591.68 1511.26 588.138 Q1513.09 584.573 1513.09 577.467 Q1513.09 570.337 1511.26 566.795 Q1509.45 563.231 1505.82 563.231 M1505.82 559.527 Q1511.63 559.527 1514.69 564.133 Q1517.76 568.717 1517.76 577.467 Q1517.76 586.194 1514.69 590.8 Q1511.63 595.383 1505.82 595.383 Q1500.01 595.383 1496.93 590.8 Q1493.88 586.194 1493.88 577.467 Q1493.88 568.717 1496.93 564.133 Q1500.01 559.527 1505.82 559.527 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1452.39 460.746 L1460.03 460.746 L1460.03 434.38 L1451.72 436.047 L1451.72 431.787 L1459.99 430.121 L1464.66 430.121 L1464.66 460.746 L1472.3 460.746 L1472.3 464.681 L1452.39 464.681 L1452.39 460.746 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1481.75 458.801 L1486.63 458.801 L1486.63 464.681 L1481.75 464.681 L1481.75 458.801 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1496.86 430.121 L1515.22 430.121 L1515.22 434.056 L1501.14 434.056 L1501.14 442.528 Q1502.16 442.181 1503.18 442.019 Q1504.2 441.834 1505.22 441.834 Q1511.01 441.834 1514.39 445.005 Q1517.76 448.176 1517.76 453.593 Q1517.76 459.171 1514.29 462.273 Q1510.82 465.352 1504.5 465.352 Q1502.33 465.352 1500.06 464.982 Q1497.81 464.611 1495.4 463.871 L1495.4 459.171 Q1497.49 460.306 1499.71 460.861 Q1501.93 461.417 1504.41 461.417 Q1508.41 461.417 1510.75 459.31 Q1513.09 457.204 1513.09 453.593 Q1513.09 449.982 1510.75 447.875 Q1508.41 445.769 1504.41 445.769 Q1502.53 445.769 1500.66 446.185 Q1498.81 446.602 1496.86 447.482 L1496.86 430.121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1454.62 330.714 L1470.94 330.714 L1470.94 334.649 L1448.99 334.649 L1448.99 330.714 Q1451.65 327.96 1456.24 323.33 Q1460.84 318.677 1462.02 317.335 Q1464.27 314.812 1465.15 313.076 Q1466.05 311.316 1466.05 309.626 Q1466.05 306.872 1464.11 305.136 Q1462.19 303.4 1459.08 303.4 Q1456.89 303.4 1454.43 304.164 Q1452 304.927 1449.22 306.478 L1449.22 301.756 Q1452.05 300.622 1454.5 300.043 Q1456.96 299.464 1458.99 299.464 Q1464.36 299.464 1467.56 302.15 Q1470.75 304.835 1470.75 309.326 Q1470.75 311.455 1469.94 313.376 Q1469.15 315.275 1467.05 317.867 Q1466.47 318.538 1463.37 321.756 Q1460.27 324.95 1454.62 330.714 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1480.75 328.77 L1485.64 328.77 L1485.64 334.649 L1480.75 334.649 L1480.75 328.77 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 303.168 Q1502.21 303.168 1500.38 306.733 Q1498.58 310.275 1498.58 317.404 Q1498.58 324.511 1500.38 328.075 Q1502.21 331.617 1505.82 331.617 Q1509.45 331.617 1511.26 328.075 Q1513.09 324.511 1513.09 317.404 Q1513.09 310.275 1511.26 306.733 Q1509.45 303.168 1505.82 303.168 M1505.82 299.464 Q1511.63 299.464 1514.69 304.071 Q1517.76 308.654 1517.76 317.404 Q1517.76 326.131 1514.69 330.737 Q1511.63 335.321 1505.82 335.321 Q1500.01 335.321 1496.93 330.737 Q1493.88 326.131 1493.88 317.404 Q1493.88 308.654 1496.93 304.071 Q1500.01 299.464 1505.82 299.464 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1455.61 200.683 L1471.93 200.683 L1471.93 204.618 L1449.99 204.618 L1449.99 200.683 Q1452.65 197.928 1457.23 193.299 Q1461.84 188.646 1463.02 187.303 Q1465.27 184.78 1466.14 183.044 Q1467.05 181.285 1467.05 179.595 Q1467.05 176.841 1465.1 175.104 Q1463.18 173.368 1460.08 173.368 Q1457.88 173.368 1455.43 174.132 Q1453 174.896 1450.22 176.447 L1450.22 171.725 Q1453.04 170.591 1455.5 170.012 Q1457.95 169.433 1459.99 169.433 Q1465.36 169.433 1468.55 172.118 Q1471.75 174.804 1471.75 179.294 Q1471.75 181.424 1470.94 183.345 Q1470.15 185.243 1468.04 187.836 Q1467.46 188.507 1464.36 191.725 Q1461.26 194.919 1455.61 200.683 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1481.75 198.739 L1486.63 198.739 L1486.63 204.618 L1481.75 204.618 L1481.75 198.739 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1496.86 170.058 L1515.22 170.058 L1515.22 173.993 L1501.14 173.993 L1501.14 182.466 Q1502.16 182.118 1503.18 181.956 Q1504.2 181.771 1505.22 181.771 Q1511.01 181.771 1514.39 184.942 Q1517.76 188.114 1517.76 193.53 Q1517.76 199.109 1514.29 202.211 Q1510.82 205.29 1504.5 205.29 Q1502.33 205.29 1500.06 204.919 Q1497.81 204.549 1495.4 203.808 L1495.4 199.109 Q1497.49 200.243 1499.71 200.799 Q1501.93 201.354 1504.41 201.354 Q1508.41 201.354 1510.75 199.248 Q1513.09 197.141 1513.09 193.53 Q1513.09 189.919 1510.75 187.813 Q1508.41 185.706 1504.41 185.706 Q1502.53 185.706 1500.66 186.123 Q1498.81 186.54 1496.86 187.419 L1496.86 170.058 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1742.16 72.576 L1719.07 12.096 L1727.61 12.096 L1746.77 63.0159 L1765.98 12.096 L1774.48 12.096 L1751.43 72.576 L1742.16 72.576 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1815.36 48.0275 L1815.36 51.6733 L1781.09 51.6733 Q1781.57 59.3701 1785.7 63.421 Q1789.88 67.4314 1797.29 67.4314 Q1801.58 67.4314 1805.59 66.3781 Q1809.64 65.3249 1813.61 63.2184 L1813.61 70.267 Q1809.6 71.9684 1805.39 72.8596 Q1801.18 73.7508 1796.84 73.7508 Q1785.99 73.7508 1779.63 67.4314 Q1773.31 61.1119 1773.31 50.3365 Q1773.31 39.1965 1779.3 32.6746 Q1785.34 26.1121 1795.55 26.1121 Q1804.7 26.1121 1810.01 32.0264 Q1815.36 37.9003 1815.36 48.0275 M1807.9 45.84 Q1807.82 39.7232 1804.46 36.0774 Q1801.14 32.4315 1795.63 32.4315 Q1789.39 32.4315 1785.62 35.9558 Q1781.9 39.4801 1781.33 45.8805 L1807.9 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1827.59 9.54393 L1835.04 9.54393 L1835.04 72.576 L1827.59 72.576 L1827.59 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1868.22 32.4315 Q1862.23 32.4315 1858.74 37.1306 Q1855.26 41.7891 1855.26 49.9314 Q1855.26 58.0738 1858.7 62.7728 Q1862.19 67.4314 1868.22 67.4314 Q1874.18 67.4314 1877.66 62.7323 Q1881.14 58.0333 1881.14 49.9314 Q1881.14 41.8701 1877.66 37.1711 Q1874.18 32.4315 1868.22 32.4315 M1868.22 26.1121 Q1877.94 26.1121 1883.49 32.4315 Q1889.04 38.7509 1889.04 49.9314 Q1889.04 61.0714 1883.49 67.4314 Q1877.94 73.7508 1868.22 73.7508 Q1858.46 73.7508 1852.91 67.4314 Q1847.4 61.0714 1847.4 49.9314 Q1847.4 38.7509 1852.91 32.4315 Q1858.46 26.1121 1868.22 26.1121 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1934.05 28.9478 L1934.05 35.9153 Q1930.89 34.1734 1927.69 33.3227 Q1924.53 32.4315 1921.29 32.4315 Q1914.04 32.4315 1910.03 37.0496 Q1906.02 41.6271 1906.02 49.9314 Q1906.02 58.2358 1910.03 62.8538 Q1914.04 67.4314 1921.29 67.4314 Q1924.53 67.4314 1927.69 66.5807 Q1930.89 65.6895 1934.05 63.9476 L1934.05 70.8341 Q1930.93 72.2924 1927.57 73.0216 Q1924.24 73.7508 1920.48 73.7508 Q1910.23 73.7508 1904.19 67.3098 Q1898.16 60.8689 1898.16 49.9314 Q1898.16 38.832 1904.23 32.472 Q1910.35 26.1121 1920.96 26.1121 Q1924.41 26.1121 1927.69 26.8413 Q1930.97 27.5299 1934.05 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1947.01 27.2059 L1954.46 27.2059 L1954.46 72.576 L1947.01 72.576 L1947.01 27.2059 M1947.01 9.54393 L1954.46 9.54393 L1954.46 18.9825 L1947.01 18.9825 L1947.01 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1977.43 14.324 L1977.43 27.2059 L1992.79 27.2059 L1992.79 32.9987 L1977.43 32.9987 L1977.43 57.6282 Q1977.43 63.1779 1978.93 64.7578 Q1980.47 66.3376 1985.13 66.3376 L1992.79 66.3376 L1992.79 72.576 L1985.13 72.576 Q1976.5 72.576 1973.22 69.3758 Q1969.94 66.1351 1969.94 57.6282 L1969.94 32.9987 L1964.47 32.9987 L1964.47 27.2059 L1969.94 27.2059 L1969.94 14.324 L1977.43 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2021.47 76.7889 Q2018.31 84.8907 2015.31 87.3618 Q2012.31 89.8329 2007.29 89.8329 L2001.33 89.8329 L2001.33 83.5945 L2005.71 83.5945 Q2008.79 83.5945 2010.49 82.1361 Q2012.19 80.6778 2014.26 75.2496 L2015.59 71.8468 L1997.24 27.2059 L2005.14 27.2059 L2019.32 62.6918 L2033.5 27.2059 L2041.4 27.2059 L2021.47 76.7889 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2095.96 9.62495 Q2090.53 18.942 2087.9 28.0566 Q2085.27 37.1711 2085.27 46.5287 Q2085.27 55.8863 2087.9 65.0818 Q2090.57 74.2369 2095.96 83.5134 L2089.48 83.5134 Q2083.4 73.9938 2080.37 64.7983 Q2077.37 55.6027 2077.37 46.5287 Q2077.37 37.4952 2080.37 28.3401 Q2083.36 19.1851 2089.48 9.62495 L2095.96 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2112.9 65.6895 L2126.26 65.6895 L2126.26 19.5497 L2111.72 22.4663 L2111.72 15.0127 L2126.18 12.096 L2134.37 12.096 L2134.37 65.6895 L2147.73 65.6895 L2147.73 72.576 L2112.9 72.576 L2112.9 65.6895 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2181.76 17.4837 Q2175.44 17.4837 2172.24 23.7221 Q2169.08 29.92 2169.08 42.3968 Q2169.08 54.833 2172.24 61.0714 Q2175.44 67.2693 2181.76 67.2693 Q2188.12 67.2693 2191.28 61.0714 Q2194.48 54.833 2194.48 42.3968 Q2194.48 29.92 2191.28 23.7221 Q2188.12 17.4837 2181.76 17.4837 M2181.76 11.0023 Q2191.93 11.0023 2197.28 19.0636 Q2202.66 27.0843 2202.66 42.3968 Q2202.66 57.6687 2197.28 65.73 Q2191.93 73.7508 2181.76 73.7508 Q2171.59 73.7508 2166.21 65.73 Q2160.86 57.6687 2160.86 42.3968 Q2160.86 27.0843 2166.21 19.0636 Q2171.59 11.0023 2181.76 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2234.54 17.4837 Q2228.22 17.4837 2225.02 23.7221 Q2221.86 29.92 2221.86 42.3968 Q2221.86 54.833 2225.02 61.0714 Q2228.22 67.2693 2234.54 67.2693 Q2240.9 67.2693 2244.06 61.0714 Q2247.26 54.833 2247.26 42.3968 Q2247.26 29.92 2244.06 23.7221 Q2240.9 17.4837 2234.54 17.4837 M2234.54 11.0023 Q2244.71 11.0023 2250.06 19.0636 Q2255.45 27.0843 2255.45 42.3968 Q2255.45 57.6687 2250.06 65.73 Q2244.71 73.7508 2234.54 73.7508 Q2224.38 73.7508 2218.99 65.73 Q2213.64 57.6687 2213.64 42.3968 Q2213.64 27.0843 2218.99 19.0636 Q2224.38 11.0023 2234.54 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2287.33 17.4837 Q2281.01 17.4837 2277.81 23.7221 Q2274.65 29.92 2274.65 42.3968 Q2274.65 54.833 2277.81 61.0714 Q2281.01 67.2693 2287.33 67.2693 Q2293.69 67.2693 2296.85 61.0714 Q2300.05 54.833 2300.05 42.3968 Q2300.05 29.92 2296.85 23.7221 Q2293.69 17.4837 2287.33 17.4837 M2287.33 11.0023 Q2297.5 11.0023 2302.84 19.0636 Q2308.23 27.0843 2308.23 42.3968 Q2308.23 57.6687 2302.84 65.73 Q2297.5 73.7508 2287.33 73.7508 Q2277.16 73.7508 2271.77 65.73 Q2266.42 57.6687 2266.42 42.3968 Q2266.42 27.0843 2271.77 19.0636 Q2277.16 11.0023 2287.33 11.0023 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2370.9 9.54393 L2370.9 15.7418 L2363.77 15.7418 Q2359.76 15.7418 2358.18 17.3622 Q2356.64 18.9825 2356.64 23.1955 L2356.64 27.2059 L2368.91 27.2059 L2368.91 32.9987 L2356.64 32.9987 L2356.64 72.576 L2349.14 72.576 L2349.14 32.9987 L2342.01 32.9987 L2342.01 27.2059 L2349.14 27.2059 L2349.14 24.0462 Q2349.14 16.471 2352.67 13.0277 Q2356.19 9.54393 2363.85 9.54393 L2370.9 9.54393 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2383.05 14.324 L2383.05 27.2059 L2398.4 27.2059 L2398.4 32.9987 L2383.05 32.9987 L2383.05 57.6282 Q2383.05 63.1779 2384.55 64.7578 Q2386.09 66.3376 2390.75 66.3376 L2398.4 66.3376 L2398.4 72.576 L2390.75 72.576 Q2382.12 72.576 2378.84 69.3758 Q2375.56 66.1351 2375.56 57.6282 L2375.56 32.9987 L2370.09 32.9987 L2370.09 27.2059 L2375.56 27.2059 L2375.56 14.324 L2383.05 14.324 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2421.45 12.096 L2428.34 12.096 L2407.27 80.2727 L2400.39 80.2727 L2421.45 12.096 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2465.08 28.5427 L2465.08 35.5912 Q2461.92 33.9709 2458.52 33.1607 Q2455.12 32.3505 2451.47 32.3505 Q2445.92 32.3505 2443.13 34.0519 Q2440.37 35.7533 2440.37 39.156 Q2440.37 41.7486 2442.36 43.2475 Q2444.34 44.7058 2450.34 46.0426 L2452.89 46.6097 Q2460.83 48.3111 2464.15 51.4303 Q2467.51 54.509 2467.51 60.0587 Q2467.51 66.3781 2462.49 70.0644 Q2457.51 73.7508 2448.76 73.7508 Q2445.11 73.7508 2441.14 73.0216 Q2437.21 72.3329 2432.84 70.9151 L2432.84 63.2184 Q2436.97 65.3654 2440.98 66.4591 Q2444.99 67.5124 2448.92 67.5124 Q2454.18 67.5124 2457.02 65.73 Q2459.86 63.9071 2459.86 60.6258 Q2459.86 57.5877 2457.79 55.9673 Q2455.76 54.3469 2448.84 52.8481 L2446.24 52.2405 Q2439.32 50.7821 2436.24 47.7845 Q2433.16 44.7463 2433.16 39.4801 Q2433.16 33.0797 2437.7 29.5959 Q2442.23 26.1121 2450.58 26.1121 Q2454.71 26.1121 2458.36 26.7198 Q2462 27.3274 2465.08 28.5427 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2518.19 48.0275 L2518.19 51.6733 L2483.92 51.6733 Q2484.4 59.3701 2488.54 63.421 Q2492.71 67.4314 2500.12 67.4314 Q2504.42 67.4314 2508.43 66.3781 Q2512.48 65.3249 2516.45 63.2184 L2516.45 70.267 Q2512.44 71.9684 2508.22 72.8596 Q2504.01 73.7508 2499.68 73.7508 Q2488.82 73.7508 2482.46 67.4314 Q2476.14 61.1119 2476.14 50.3365 Q2476.14 39.1965 2482.14 32.6746 Q2488.17 26.1121 2498.38 26.1121 Q2507.53 26.1121 2512.84 32.0264 Q2518.19 37.9003 2518.19 48.0275 M2510.73 45.84 Q2510.65 39.7232 2507.29 36.0774 Q2503.97 32.4315 2498.46 32.4315 Q2492.22 32.4315 2488.45 35.9558 Q2484.73 39.4801 2484.16 45.8805 L2510.73 45.84 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2563.07 28.9478 L2563.07 35.9153 Q2559.91 34.1734 2556.71 33.3227 Q2553.55 32.4315 2550.31 32.4315 Q2543.06 32.4315 2539.05 37.0496 Q2535.04 41.6271 2535.04 49.9314 Q2535.04 58.2358 2539.05 62.8538 Q2543.06 67.4314 2550.31 67.4314 Q2553.55 67.4314 2556.71 66.5807 Q2559.91 65.6895 2563.07 63.9476 L2563.07 70.8341 Q2559.95 72.2924 2556.59 73.0216 Q2553.27 73.7508 2549.5 73.7508 Q2539.25 73.7508 2533.22 67.3098 Q2527.18 60.8689 2527.18 49.9314 Q2527.18 38.832 2533.26 32.472 Q2539.37 26.1121 2549.99 26.1121 Q2553.43 26.1121 2556.71 26.8413 Q2559.99 27.5299 2563.07 28.9478 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2574.86 9.62495 L2581.34 9.62495 Q2587.42 19.1851 2590.42 28.3401 Q2593.45 37.4952 2593.45 46.5287 Q2593.45 55.6027 2590.42 64.7983 Q2587.42 73.9938 2581.34 83.5134 L2574.86 83.5134 Q2580.25 74.2369 2582.88 65.0818 Q2585.55 55.8863 2585.55 46.5287 Q2585.55 37.1711 2582.88 28.0566 Q2580.25 18.942 2574.86 9.62495 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip033)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,171.734 1595.77,171.777 1598.01,171.828 1600.25,171.887 1602.5,171.956 1604.74,172.034 1606.98,172.123 1609.22,172.223 1611.46,172.335 1613.71,172.461 1615.95,172.601 1618.19,172.756 1620.43,172.927 1622.67,173.116 1624.92,173.324 1627.16,173.551 1629.4,173.8 1631.64,174.072 1633.88,174.367 1636.13,174.688 1638.37,175.035 1640.61,175.411 1642.85,175.817 1645.09,176.254 1647.34,176.724 1649.58,177.228 1651.82,177.768 1654.06,178.344 1656.3,178.958 1658.55,179.61 1660.79,180.303 1663.03,181.035 1665.27,181.808 1667.51,182.621 1669.76,183.475 1672,184.369 1674.24,185.302 1676.48,186.273 1678.72,187.281 1680.97,188.323 1683.21,189.399 1685.45,190.504 1687.69,191.637 1689.93,192.795 1692.17,193.974 1694.42,195.171 1696.66,196.383 1698.9,197.605 1701.14,198.835 1703.38,200.068 1705.63,201.301 1707.87,202.531 1710.11,203.754 1712.35,204.967 1714.59,206.167 1716.84,207.352 1719.08,208.519 1721.32,209.666 1723.56,210.791 1725.8,211.893 1728.05,212.97 1730.29,214.022 1732.53,215.048 1734.77,216.047 1737.01,217.018 1739.26,217.963 1741.5,218.88 1743.74,219.771 1745.98,220.635 1748.22,221.474 1750.47,222.287 1752.71,223.076 1754.95,223.841 1757.19,224.583 1759.43,225.304 1761.68,226.003 1763.92,226.682 1766.16,227.342 1768.4,227.983 1770.64,228.608 1772.89,229.216 1775.13,229.809 1777.37,230.387 1779.61,230.953 1781.85,231.505 1784.1,232.046 1786.34,232.577 1788.58,233.098 1790.82,233.61 1793.06,234.113 1795.31,234.61 1797.55,235.1 1799.79,235.585 1802.03,236.065 1804.27,236.542 1806.51,237.015 1808.76,237.487 1811,237.957 1813.24,238.426 1815.48,238.896 1817.72,239.368 1819.97,239.841 1822.21,240.318 1824.45,240.798 1826.69,241.283 1828.93,241.775 1831.18,242.273 1833.42,242.779 1835.66,243.294 1837.9,243.819 1840.14,244.355 1842.39,244.903 1844.63,245.465 1846.87,246.042 1849.11,246.634 1851.35,247.243 1853.6,247.87 1855.84,248.517 1858.08,249.185 1860.32,249.874 1862.56,250.587 1864.81,251.324 1867.05,252.086 1869.29,252.875 1871.53,253.692 1873.77,254.536 1876.02,255.41 1878.26,256.314 1880.5,257.247 1882.74,258.211 1884.98,259.205 1887.23,260.229 1889.47,261.283 1891.71,262.365 1893.95,263.476 1896.19,264.612 1898.44,265.774 1900.68,266.96 1902.92,268.166 1905.16,269.391 1907.4,270.633 1909.64,271.888 1911.89,273.154 1914.13,274.429 1916.37,275.709 1918.61,276.991 1920.85,278.273 1923.1,279.552 1925.34,280.826 1927.58,282.091 1929.82,283.346 1932.06,284.589 1934.31,285.817 1936.55,287.03 1938.79,288.226 1941.03,289.403 1943.27,290.562 1945.52,291.701 1947.76,292.821 1950,293.92 1952.24,295 1954.48,296.06 1956.73,297.1 1958.97,298.122 1961.21,299.126 1963.45,300.113 1965.69,301.083 1967.94,302.038 1970.18,302.978 1972.42,303.905 1974.66,304.82 1976.9,305.724 1979.15,306.619 1981.39,307.505 1983.63,308.385 1985.87,309.259 1988.11,310.128 1990.36,310.995 1992.6,311.861 1994.84,312.726 1997.08,313.594 1999.32,314.464 2001.57,315.339 2003.81,316.219 2006.05,317.108 2008.29,318.005 2010.53,318.912 2012.78,319.832 2015.02,320.764 2017.26,321.712 2019.5,322.675 2021.74,323.656 2023.98,324.655 2026.23,325.674 2028.47,326.713 2030.71,327.774 2032.95,328.857 2035.19,329.963 2037.44,331.093 2039.68,332.246 2041.92,333.422 2044.16,334.621 2046.4,335.843 2048.65,337.088 2050.89,338.353 2053.13,339.638 2055.37,340.942 2057.61,342.263 2059.86,343.599 2062.1,344.948 2064.34,346.307 2066.58,347.676 2068.82,349.051 2071.07,350.43 2073.31,351.811 2075.55,353.192 2077.79,354.57 2080.03,355.943 2082.28,357.31 2084.52,358.668 2086.76,360.017 2089,361.355 2091.24,362.68 2093.49,363.992 2095.73,365.29 2097.97,366.574 2100.21,367.844 2102.45,369.099 2104.7,370.339 2106.94,371.566 2109.18,372.779 2111.42,373.979 2113.66,375.166 2115.91,376.343 2118.15,377.51 2120.39,378.667 2122.63,379.816 2124.87,380.959 2127.11,382.097 2129.36,383.231 2131.6,384.362 2133.84,385.492 2136.08,386.623 2138.32,387.756 2140.57,388.892 2142.81,390.033 2145.05,391.181 2147.29,392.336 2149.53,393.501 2151.78,394.676 2154.02,395.863 2156.26,397.063 2158.5,398.277 2160.74,399.506 2162.99,400.75 2165.23,402.012 2167.47,403.29 2169.71,404.586 2171.95,405.899 2174.2,407.229 2176.44,408.577 2178.68,409.942 2180.92,411.322 2183.16,412.718 2185.41,414.128 2187.65,415.551 2189.89,416.986 2192.13,418.43 2194.37,419.883 2196.62,421.343 2198.86,422.808 2201.1,424.276 2203.34,425.746 2205.58,427.215 2207.83,428.683 2210.07,430.147 2212.31,431.606 2214.55,433.059 2216.79,434.505 2219.04,435.943 2221.28,437.372 2223.52,438.792 2225.76,440.202 2228,441.602 2230.25,442.993 2232.49,444.373 2234.73,445.744 2236.97,447.106 2239.21,448.459 2241.45,449.805 2243.7,451.144 2245.94,452.477 2248.18,453.806 2250.42,455.13 2252.66,456.452 2254.91,457.772 2257.15,459.092 2259.39,460.414 2261.63,461.737 2263.87,463.064 2266.12,464.396 2268.36,465.734 2270.6,467.078 2272.84,468.431 2275.08,469.793 2277.33,471.164 2279.57,472.546 2281.81,473.94 2284.05,475.345 2286.29,476.763 2288.54,478.192 2290.78,479.634 2293.02,481.088 2295.26,482.555 2297.5,484.032 2299.75,485.521 2301.99,487.02 2304.23,488.528 2306.47,490.044 2308.71,491.568 2310.96,493.097 2313.2,494.632 2315.44,496.17 2317.68,497.71 2319.92,499.252 2322.17,500.793 2324.41,502.334 2326.65,503.872 2328.89,505.406 2331.13,506.937 2333.38,508.463 2335.62,509.984 2337.86,511.498 2340.1,513.007 2342.34,514.509 2344.58,516.005 2346.83,517.495 2349.07,518.979 2351.31,520.457 2353.55,521.93 2355.79,523.399 2358.04,524.863 2360.28,526.324 2362.52,527.783 2364.76,529.24 2367,530.697 2369.25,532.154 2371.49,533.611 2373.73,535.071 2375.97,536.534 2378.21,538 2380.46,539.471 2382.7,540.947 2384.94,542.429 2387.18,543.918 2389.42,545.413 2391.67,546.917 2393.91,548.428 2396.15,549.947 2398.39,551.474 2400.63,553.009 2402.88,554.552 2405.12,556.103 2407.36,557.661 2409.6,559.226 2411.84,560.797 2414.09,562.373 2416.33,563.955 2418.57,565.54 2420.81,567.129 2423.05,568.721 2425.3,570.314 2427.54,571.908 2429.78,573.501 2432.02,575.095 2434.26,576.686 2436.51,578.276 2438.75,579.864 2440.99,581.448 2443.23,583.029 2445.47,584.606 2447.72,586.18 2449.96,587.749 2452.2,589.315 2454.44,590.877 2456.68,592.435 2458.92,593.991 2461.17,595.543 2463.41,597.093 2465.65,598.64 2467.89,600.187 2470.13,601.732 2472.38,603.276 2474.62,604.821 2476.86,606.367 2479.1,607.914 2481.34,609.463 2483.59,611.014 2485.83,612.569 2488.07,614.126 2490.31,615.688 2492.55,617.253 2494.8,618.823 2497.04,620.398 2499.28,621.977 2501.52,623.561 2503.76,625.15 2506.01,626.743 2508.25,628.341 2510.49,629.942 2512.73,631.548 2514.97,633.157 2517.22,634.769 2519.46,636.384 2521.7,638 2523.94,639.618 2526.18,641.237 2528.43,642.856 2530.67,644.474 2532.91,646.093 2535.15,647.709 2537.39,649.324 2539.64,650.937 2541.88,652.548 2544.12,654.155 2546.36,655.76 2548.6,657.361 2550.85,658.959 2553.09,660.553 2555.33,662.144 2557.57,663.732 2559.81,665.316 2562.05,666.897 2564.3,668.475 2566.54,670.051 2568.78,671.624 2571.02,673.195 2573.26,674.765 2575.51,676.334 2577.75,677.901 2579.99,679.468 2582.23,681.036 2584.47,682.603 2586.72,684.172 2588.96,685.741 2591.2,687.312 2593.44,688.885 2595.68,690.459 2597.93,692.036 2600.17,693.615 2602.41,695.197 2604.65,696.781 2606.89,698.367 2609.14,699.956 2611.38,701.546 2613.62,703.138 2615.86,704.732 2618.1,706.326 2620.35,707.921 2622.59,709.516 2624.83,711.111 2627.07,712.704 2629.31,714.295 2631.56,715.883 2633.8,717.468 2636.04,719.049 2638.28,720.624 2640.52,722.194 2642.77,723.757 2645.01,725.312 2647.25,726.859 2649.49,728.397 2651.73,729.926 2653.98,731.444 2656.22,732.952 2658.46,734.448 2660.7,735.932 2662.94,737.404 2665.19,738.865 2667.43,740.313 2669.67,741.748 2671.91,743.172 2674.15,744.584 2676.39,745.984 2678.64,747.374 2680.88,748.755 2683.12,750.126 2685.36,751.49 2687.6,752.848 2689.85,754.202 2692.09,755.553 2694.33,756.904 2696.57,758.256 2698.81,759.613 2701.06,760.977 2703.3,762.351 2705.54,763.738 2707.78,765.14 2710.02,766.561 2712.27,768.004 2714.51,769.469 2716.75,770.961 2718.99,772.479 "/>
+<path clip-path="url(#clip030)" d="M159.765 1723.83 L1352.76 1723.83 L1352.76 1087.05 L159.765 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip214">
+  <clipPath id="clip034">
     <rect x="159" y="1087" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,1723.83 193.529,1087.05 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,1723.83 473.774,1087.05 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,1723.83 754.018,1087.05 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,1723.83 1034.26,1087.05 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,1723.83 1314.51,1087.05 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1723.83 1352.76,1723.83 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,1723.83 193.529,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,1723.83 473.774,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,1723.83 754.018,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,1723.83 1034.26,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,1723.83 1314.51,1704.94 "/>
-<path clip-path="url(#clip210)" d="M193.529 1757.95 Q189.918 1757.95 188.089 1761.52 Q186.283 1765.06 186.283 1772.19 Q186.283 1779.3 188.089 1782.86 Q189.918 1786.4 193.529 1786.4 Q197.163 1786.4 198.969 1782.86 Q200.797 1779.3 200.797 1772.19 Q200.797 1765.06 198.969 1761.52 Q197.163 1757.95 193.529 1757.95 M193.529 1754.25 Q199.339 1754.25 202.394 1758.86 Q205.473 1763.44 205.473 1772.19 Q205.473 1780.92 202.394 1785.52 Q199.339 1790.11 193.529 1790.11 Q187.719 1790.11 184.64 1785.52 Q181.584 1780.92 181.584 1772.19 Q181.584 1763.44 184.64 1758.86 Q187.719 1754.25 193.529 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M433.392 1754.87 L451.748 1754.87 L451.748 1758.81 L437.674 1758.81 L437.674 1767.28 Q438.693 1766.93 439.711 1766.77 Q440.73 1766.59 441.748 1766.59 Q447.535 1766.59 450.915 1769.76 Q454.295 1772.93 454.295 1778.35 Q454.295 1783.93 450.822 1787.03 Q447.35 1790.11 441.031 1790.11 Q438.855 1790.11 436.586 1789.74 Q434.341 1789.37 431.934 1788.62 L431.934 1783.93 Q434.017 1785.06 436.239 1785.62 Q438.461 1786.17 440.938 1786.17 Q444.943 1786.17 447.281 1784.06 Q449.619 1781.96 449.619 1778.35 Q449.619 1774.74 447.281 1772.63 Q444.943 1770.52 440.938 1770.52 Q439.063 1770.52 437.188 1770.94 Q435.336 1771.36 433.392 1772.24 L433.392 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M473.507 1757.95 Q469.896 1757.95 468.068 1761.52 Q466.262 1765.06 466.262 1772.19 Q466.262 1779.3 468.068 1782.86 Q469.896 1786.4 473.507 1786.4 Q477.142 1786.4 478.947 1782.86 Q480.776 1779.3 480.776 1772.19 Q480.776 1765.06 478.947 1761.52 Q477.142 1757.95 473.507 1757.95 M473.507 1754.25 Q479.318 1754.25 482.373 1758.86 Q485.452 1763.44 485.452 1772.19 Q485.452 1780.92 482.373 1785.52 Q479.318 1790.11 473.507 1790.11 Q467.697 1790.11 464.619 1785.52 Q461.563 1780.92 461.563 1772.19 Q461.563 1763.44 464.619 1758.86 Q467.697 1754.25 473.507 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M503.669 1757.95 Q500.058 1757.95 498.23 1761.52 Q496.424 1765.06 496.424 1772.19 Q496.424 1779.3 498.23 1782.86 Q500.058 1786.4 503.669 1786.4 Q507.304 1786.4 509.109 1782.86 Q510.938 1779.3 510.938 1772.19 Q510.938 1765.06 509.109 1761.52 Q507.304 1757.95 503.669 1757.95 M503.669 1754.25 Q509.479 1754.25 512.535 1758.86 Q515.614 1763.44 515.614 1772.19 Q515.614 1780.92 512.535 1785.52 Q509.479 1790.11 503.669 1790.11 Q497.859 1790.11 494.78 1785.52 Q491.725 1780.92 491.725 1772.19 Q491.725 1763.44 494.78 1758.86 Q497.859 1754.25 503.669 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M698.544 1785.5 L706.183 1785.5 L706.183 1759.13 L697.873 1760.8 L697.873 1756.54 L706.137 1754.87 L710.813 1754.87 L710.813 1785.5 L718.452 1785.5 L718.452 1789.43 L698.544 1789.43 L698.544 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M737.896 1757.95 Q734.285 1757.95 732.456 1761.52 Q730.651 1765.06 730.651 1772.19 Q730.651 1779.3 732.456 1782.86 Q734.285 1786.4 737.896 1786.4 Q741.53 1786.4 743.336 1782.86 Q745.164 1779.3 745.164 1772.19 Q745.164 1765.06 743.336 1761.52 Q741.53 1757.95 737.896 1757.95 M737.896 1754.25 Q743.706 1754.25 746.762 1758.86 Q749.84 1763.44 749.84 1772.19 Q749.84 1780.92 746.762 1785.52 Q743.706 1790.11 737.896 1790.11 Q732.086 1790.11 729.007 1785.52 Q725.951 1780.92 725.951 1772.19 Q725.951 1763.44 729.007 1758.86 Q732.086 1754.25 737.896 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M768.058 1757.95 Q764.447 1757.95 762.618 1761.52 Q760.812 1765.06 760.812 1772.19 Q760.812 1779.3 762.618 1782.86 Q764.447 1786.4 768.058 1786.4 Q771.692 1786.4 773.498 1782.86 Q775.326 1779.3 775.326 1772.19 Q775.326 1765.06 773.498 1761.52 Q771.692 1757.95 768.058 1757.95 M768.058 1754.25 Q773.868 1754.25 776.923 1758.86 Q780.002 1763.44 780.002 1772.19 Q780.002 1780.92 776.923 1785.52 Q773.868 1790.11 768.058 1790.11 Q762.248 1790.11 759.169 1785.52 Q756.113 1780.92 756.113 1772.19 Q756.113 1763.44 759.169 1758.86 Q762.248 1754.25 768.058 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M798.22 1757.95 Q794.609 1757.95 792.78 1761.52 Q790.974 1765.06 790.974 1772.19 Q790.974 1779.3 792.78 1782.86 Q794.609 1786.4 798.22 1786.4 Q801.854 1786.4 803.659 1782.86 Q805.488 1779.3 805.488 1772.19 Q805.488 1765.06 803.659 1761.52 Q801.854 1757.95 798.22 1757.95 M798.22 1754.25 Q804.03 1754.25 807.085 1758.86 Q810.164 1763.44 810.164 1772.19 Q810.164 1780.92 807.085 1785.52 Q804.03 1790.11 798.22 1790.11 Q792.409 1790.11 789.331 1785.52 Q786.275 1780.92 786.275 1772.19 Q786.275 1763.44 789.331 1758.86 Q792.409 1754.25 798.22 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M978.789 1785.5 L986.428 1785.5 L986.428 1759.13 L978.118 1760.8 L978.118 1756.54 L986.382 1754.87 L991.057 1754.87 L991.057 1785.5 L998.696 1785.5 L998.696 1789.43 L978.789 1789.43 L978.789 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1008.19 1754.87 L1026.54 1754.87 L1026.54 1758.81 L1012.47 1758.81 L1012.47 1767.28 Q1013.49 1766.93 1014.51 1766.77 Q1015.52 1766.59 1016.54 1766.59 Q1022.33 1766.59 1025.71 1769.76 Q1029.09 1772.93 1029.09 1778.35 Q1029.09 1783.93 1025.62 1787.03 Q1022.15 1790.11 1015.83 1790.11 Q1013.65 1790.11 1011.38 1789.74 Q1009.14 1789.37 1006.73 1788.62 L1006.73 1783.93 Q1008.81 1785.06 1011.03 1785.62 Q1013.26 1786.17 1015.73 1786.17 Q1019.74 1786.17 1022.08 1784.06 Q1024.41 1781.96 1024.41 1778.35 Q1024.41 1774.74 1022.08 1772.63 Q1019.74 1770.52 1015.73 1770.52 Q1013.86 1770.52 1011.98 1770.94 Q1010.13 1771.36 1008.19 1772.24 L1008.19 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1048.3 1757.95 Q1044.69 1757.95 1042.86 1761.52 Q1041.06 1765.06 1041.06 1772.19 Q1041.06 1779.3 1042.86 1782.86 Q1044.69 1786.4 1048.3 1786.4 Q1051.94 1786.4 1053.74 1782.86 Q1055.57 1779.3 1055.57 1772.19 Q1055.57 1765.06 1053.74 1761.52 Q1051.94 1757.95 1048.3 1757.95 M1048.3 1754.25 Q1054.11 1754.25 1057.17 1758.86 Q1060.25 1763.44 1060.25 1772.19 Q1060.25 1780.92 1057.17 1785.52 Q1054.11 1790.11 1048.3 1790.11 Q1042.49 1790.11 1039.41 1785.52 Q1036.36 1780.92 1036.36 1772.19 Q1036.36 1763.44 1039.41 1758.86 Q1042.49 1754.25 1048.3 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1078.46 1757.95 Q1074.85 1757.95 1073.02 1761.52 Q1071.22 1765.06 1071.22 1772.19 Q1071.22 1779.3 1073.02 1782.86 Q1074.85 1786.4 1078.46 1786.4 Q1082.1 1786.4 1083.9 1782.86 Q1085.73 1779.3 1085.73 1772.19 Q1085.73 1765.06 1083.9 1761.52 Q1082.1 1757.95 1078.46 1757.95 M1078.46 1754.25 Q1084.27 1754.25 1087.33 1758.86 Q1090.41 1763.44 1090.41 1772.19 Q1090.41 1780.92 1087.33 1785.52 Q1084.27 1790.11 1078.46 1790.11 Q1072.65 1790.11 1069.58 1785.52 Q1066.52 1780.92 1066.52 1772.19 Q1066.52 1763.44 1069.58 1758.86 Q1072.65 1754.25 1078.46 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1263.12 1785.5 L1279.44 1785.5 L1279.44 1789.43 L1257.49 1789.43 L1257.49 1785.5 Q1260.16 1782.74 1264.74 1778.12 Q1269.35 1773.46 1270.53 1772.12 Q1272.77 1769.6 1273.65 1767.86 Q1274.55 1766.1 1274.55 1764.41 Q1274.55 1761.66 1272.61 1759.92 Q1270.69 1758.18 1267.59 1758.18 Q1265.39 1758.18 1262.93 1758.95 Q1260.5 1759.71 1257.73 1761.26 L1257.73 1756.54 Q1260.55 1755.41 1263 1754.83 Q1265.46 1754.25 1267.49 1754.25 Q1272.86 1754.25 1276.06 1756.93 Q1279.25 1759.62 1279.25 1764.11 Q1279.25 1766.24 1278.44 1768.16 Q1277.66 1770.06 1275.55 1772.65 Q1274.97 1773.32 1271.87 1776.54 Q1268.77 1779.74 1263.12 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1299.25 1757.95 Q1295.64 1757.95 1293.81 1761.52 Q1292.01 1765.06 1292.01 1772.19 Q1292.01 1779.3 1293.81 1782.86 Q1295.64 1786.4 1299.25 1786.4 Q1302.89 1786.4 1304.69 1782.86 Q1306.52 1779.3 1306.52 1772.19 Q1306.52 1765.06 1304.69 1761.52 Q1302.89 1757.95 1299.25 1757.95 M1299.25 1754.25 Q1305.06 1754.25 1308.12 1758.86 Q1311.2 1763.44 1311.2 1772.19 Q1311.2 1780.92 1308.12 1785.52 Q1305.06 1790.11 1299.25 1790.11 Q1293.44 1790.11 1290.36 1785.52 Q1287.31 1780.92 1287.31 1772.19 Q1287.31 1763.44 1290.36 1758.86 Q1293.44 1754.25 1299.25 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1329.42 1757.95 Q1325.8 1757.95 1323.98 1761.52 Q1322.17 1765.06 1322.17 1772.19 Q1322.17 1779.3 1323.98 1782.86 Q1325.8 1786.4 1329.42 1786.4 Q1333.05 1786.4 1334.86 1782.86 Q1336.68 1779.3 1336.68 1772.19 Q1336.68 1765.06 1334.86 1761.52 Q1333.05 1757.95 1329.42 1757.95 M1329.42 1754.25 Q1335.23 1754.25 1338.28 1758.86 Q1341.36 1763.44 1341.36 1772.19 Q1341.36 1780.92 1338.28 1785.52 Q1335.23 1790.11 1329.42 1790.11 Q1323.61 1790.11 1320.53 1785.52 Q1317.47 1780.92 1317.47 1772.19 Q1317.47 1763.44 1320.53 1758.86 Q1323.61 1754.25 1329.42 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1359.58 1757.95 Q1355.97 1757.95 1354.14 1761.52 Q1352.33 1765.06 1352.33 1772.19 Q1352.33 1779.3 1354.14 1782.86 Q1355.97 1786.4 1359.58 1786.4 Q1363.21 1786.4 1365.02 1782.86 Q1366.85 1779.3 1366.85 1772.19 Q1366.85 1765.06 1365.02 1761.52 Q1363.21 1757.95 1359.58 1757.95 M1359.58 1754.25 Q1365.39 1754.25 1368.44 1758.86 Q1371.52 1763.44 1371.52 1772.19 Q1371.52 1780.92 1368.44 1785.52 Q1365.39 1790.11 1359.58 1790.11 Q1353.77 1790.11 1350.69 1785.52 Q1347.63 1780.92 1347.63 1772.19 Q1347.63 1763.44 1350.69 1758.86 Q1353.77 1754.25 1359.58 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1705.81 1352.76,1705.81 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1546.36 1352.76,1546.36 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1386.91 1352.76,1386.91 "/>
-<polyline clip-path="url(#clip214)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1227.47 1352.76,1227.47 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1723.83 159.765,1087.05 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1705.81 178.663,1705.81 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1546.36 178.663,1546.36 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1386.91 178.663,1386.91 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1227.47 178.663,1227.47 "/>
-<path clip-path="url(#clip210)" d="M105.821 1691.61 Q102.209 1691.61 100.381 1695.18 Q98.5752 1698.72 98.5752 1705.85 Q98.5752 1712.95 100.381 1716.52 Q102.209 1720.06 105.821 1720.06 Q109.455 1720.06 111.26 1716.52 Q113.089 1712.95 113.089 1705.85 Q113.089 1698.72 111.26 1695.18 Q109.455 1691.61 105.821 1691.61 M105.821 1687.91 Q111.631 1687.91 114.686 1692.51 Q117.765 1697.1 117.765 1705.85 Q117.765 1714.57 114.686 1719.18 Q111.631 1723.76 105.821 1723.76 Q100.01 1723.76 96.9317 1719.18 Q93.8761 1714.57 93.8761 1705.85 Q93.8761 1697.1 96.9317 1692.51 Q100.01 1687.91 105.821 1687.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M69.6865 1559.71 L86.0058 1559.71 L86.0058 1563.64 L64.0615 1563.64 L64.0615 1559.71 Q66.7235 1556.95 71.3068 1552.32 Q75.9133 1547.67 77.0938 1546.33 Q79.3392 1543.81 80.2188 1542.07 Q81.1216 1540.31 81.1216 1538.62 Q81.1216 1535.87 79.1771 1534.13 Q77.2559 1532.39 74.154 1532.39 Q71.955 1532.39 69.5013 1533.16 Q67.0707 1533.92 64.293 1535.47 L64.293 1530.75 Q67.117 1529.62 69.5707 1529.04 Q72.0244 1528.46 74.0614 1528.46 Q79.4318 1528.46 82.6262 1531.14 Q85.8206 1533.83 85.8206 1538.32 Q85.8206 1540.45 85.0105 1542.37 Q84.2234 1544.27 82.1169 1546.86 Q81.5382 1547.53 78.4364 1550.75 Q75.3346 1553.94 69.6865 1559.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 1532.16 Q102.209 1532.16 100.381 1535.73 Q98.5752 1539.27 98.5752 1546.4 Q98.5752 1553.5 100.381 1557.07 Q102.209 1560.61 105.821 1560.61 Q109.455 1560.61 111.26 1557.07 Q113.089 1553.5 113.089 1546.4 Q113.089 1539.27 111.26 1535.73 Q109.455 1532.16 105.821 1532.16 M105.821 1528.46 Q111.631 1528.46 114.686 1533.06 Q117.765 1537.65 117.765 1546.4 Q117.765 1555.12 114.686 1559.73 Q111.631 1564.31 105.821 1564.31 Q100.01 1564.31 96.9317 1559.73 Q93.8761 1555.12 93.8761 1546.4 Q93.8761 1537.65 96.9317 1533.06 Q100.01 1528.46 105.821 1528.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M78.5059 1373.71 L66.7004 1392.16 L78.5059 1392.16 L78.5059 1373.71 M77.279 1369.63 L83.1586 1369.63 L83.1586 1392.16 L88.0891 1392.16 L88.0891 1396.05 L83.1586 1396.05 L83.1586 1404.19 L78.5059 1404.19 L78.5059 1396.05 L62.9041 1396.05 L62.9041 1391.53 L77.279 1369.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 1372.71 Q102.209 1372.71 100.381 1376.28 Q98.5752 1379.82 98.5752 1386.95 Q98.5752 1394.06 100.381 1397.62 Q102.209 1401.16 105.821 1401.16 Q109.455 1401.16 111.26 1397.62 Q113.089 1394.06 113.089 1386.95 Q113.089 1379.82 111.26 1376.28 Q109.455 1372.71 105.821 1372.71 M105.821 1369.01 Q111.631 1369.01 114.686 1373.62 Q117.765 1378.2 117.765 1386.95 Q117.765 1395.68 114.686 1400.28 Q111.631 1404.87 105.821 1404.87 Q100.01 1404.87 96.9317 1400.28 Q93.8761 1395.68 93.8761 1386.95 Q93.8761 1378.2 96.9317 1373.62 Q100.01 1369.01 105.821 1369.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M76.2374 1225.6 Q73.0892 1225.6 71.2374 1227.75 Q69.4087 1229.91 69.4087 1233.66 Q69.4087 1237.38 71.2374 1239.56 Q73.0892 1241.71 76.2374 1241.71 Q79.3855 1241.71 81.2142 1239.56 Q83.066 1237.38 83.066 1233.66 Q83.066 1229.91 81.2142 1227.75 Q79.3855 1225.6 76.2374 1225.6 M85.5197 1210.95 L85.5197 1215.21 Q83.7605 1214.38 81.9549 1213.94 Q80.1725 1213.5 78.4133 1213.5 Q73.7837 1213.5 71.33 1216.62 Q68.8994 1219.75 68.5522 1226.06 Q69.9179 1224.05 71.9781 1222.99 Q74.0383 1221.9 76.5151 1221.9 Q81.7234 1221.9 84.7327 1225.07 Q87.7651 1228.22 87.7651 1233.66 Q87.7651 1238.98 84.6169 1242.2 Q81.4688 1245.42 76.2374 1245.42 Q70.242 1245.42 67.0707 1240.83 Q63.8995 1236.23 63.8995 1227.5 Q63.8995 1219.31 67.7883 1214.44 Q71.6772 1209.56 78.2281 1209.56 Q79.9873 1209.56 81.7697 1209.91 Q83.5753 1210.25 85.5197 1210.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 1213.26 Q102.209 1213.26 100.381 1216.83 Q98.5752 1220.37 98.5752 1227.5 Q98.5752 1234.61 100.381 1238.17 Q102.209 1241.71 105.821 1241.71 Q109.455 1241.71 111.26 1238.17 Q113.089 1234.61 113.089 1227.5 Q113.089 1220.37 111.26 1216.83 Q109.455 1213.26 105.821 1213.26 M105.821 1209.56 Q111.631 1209.56 114.686 1214.17 Q117.765 1218.75 117.765 1227.5 Q117.765 1236.23 114.686 1240.83 Q111.631 1245.42 105.821 1245.42 Q100.01 1245.42 96.9317 1240.83 Q93.8761 1236.23 93.8761 1227.5 Q93.8761 1218.75 96.9317 1214.17 Q100.01 1209.56 105.821 1209.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M433.849 945.429 L442.032 945.429 L442.032 999.023 L471.482 999.023 L471.482 1005.91 L433.849 1005.91 L433.849 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M495.868 965.765 Q489.873 965.765 486.389 970.464 Q482.905 975.122 482.905 983.265 Q482.905 991.407 486.349 996.106 Q489.832 1000.76 495.868 1000.76 Q501.823 1000.76 505.307 996.066 Q508.791 991.367 508.791 983.265 Q508.791 975.203 505.307 970.504 Q501.823 965.765 495.868 965.765 M495.868 959.445 Q505.59 959.445 511.14 965.765 Q516.69 972.084 516.69 983.265 Q516.69 994.405 511.14 1000.76 Q505.59 1007.08 495.868 1007.08 Q486.106 1007.08 480.556 1000.76 Q475.047 994.405 475.047 983.265 Q475.047 972.084 480.556 965.765 Q486.106 959.445 495.868 959.445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M566.759 978.525 L566.759 1005.91 L559.305 1005.91 L559.305 978.768 Q559.305 972.327 556.794 969.127 Q554.282 965.927 549.259 965.927 Q543.223 965.927 539.74 969.775 Q536.256 973.624 536.256 980.267 L536.256 1005.91 L528.762 1005.91 L528.762 960.539 L536.256 960.539 L536.256 967.588 Q538.929 963.496 542.535 961.471 Q546.181 959.445 550.92 959.445 Q558.738 959.445 562.749 964.307 Q566.759 969.127 566.759 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M611.481 982.698 Q611.481 974.596 608.119 970.14 Q604.797 965.684 598.761 965.684 Q592.766 965.684 589.404 970.14 Q586.082 974.596 586.082 982.698 Q586.082 990.759 589.404 995.215 Q592.766 999.671 598.761 999.671 Q604.797 999.671 608.119 995.215 Q611.481 990.759 611.481 982.698 M618.935 1000.28 Q618.935 1011.86 613.79 1017.49 Q608.645 1023.17 598.032 1023.17 Q594.103 1023.17 590.619 1022.56 Q587.135 1021.99 583.854 1020.78 L583.854 1013.53 Q587.135 1015.31 590.335 1016.16 Q593.536 1017.01 596.857 1017.01 Q604.189 1017.01 607.835 1013.16 Q611.481 1009.35 611.481 1001.62 L611.481 997.929 Q609.172 1001.94 605.567 1003.92 Q601.961 1005.91 596.938 1005.91 Q588.593 1005.91 583.489 999.549 Q578.385 993.189 578.385 982.698 Q578.385 972.165 583.489 965.805 Q588.593 959.445 596.938 959.445 Q601.961 959.445 605.567 961.43 Q609.172 963.415 611.481 967.426 L611.481 960.539 L618.935 960.539 L618.935 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M634.288 960.539 L641.741 960.539 L641.741 1005.91 L634.288 1005.91 L634.288 960.539 M634.288 942.877 L641.741 942.877 L641.741 952.316 L634.288 952.316 L634.288 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M664.71 947.657 L664.71 960.539 L680.063 960.539 L680.063 966.332 L664.71 966.332 L664.71 990.961 Q664.71 996.511 666.209 998.091 Q667.748 999.671 672.407 999.671 L680.063 999.671 L680.063 1005.91 L672.407 1005.91 Q663.778 1005.91 660.497 1002.71 Q657.216 999.468 657.216 990.961 L657.216 966.332 L651.747 966.332 L651.747 960.539 L657.216 960.539 L657.216 947.657 L664.71 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M689.096 988.004 L689.096 960.539 L696.55 960.539 L696.55 987.721 Q696.55 994.162 699.062 997.402 Q701.573 1000.6 706.596 1000.6 Q712.632 1000.6 716.116 996.754 Q719.64 992.906 719.64 986.262 L719.64 960.539 L727.094 960.539 L727.094 1005.91 L719.64 1005.91 L719.64 998.942 Q716.926 1003.07 713.321 1005.1 Q709.756 1007.08 705.016 1007.08 Q697.198 1007.08 693.147 1002.22 Q689.096 997.362 689.096 988.004 M707.852 959.445 L707.852 959.445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M772.302 967.426 L772.302 942.877 L779.756 942.877 L779.756 1005.91 L772.302 1005.91 L772.302 999.104 Q769.952 1003.15 766.347 1005.14 Q762.782 1007.08 757.759 1007.08 Q749.536 1007.08 744.351 1000.52 Q739.206 993.959 739.206 983.265 Q739.206 972.57 744.351 966.008 Q749.536 959.445 757.759 959.445 Q762.782 959.445 766.347 961.43 Q769.952 963.375 772.302 967.426 M746.903 983.265 Q746.903 991.488 750.265 996.187 Q753.668 1000.85 759.582 1000.85 Q765.496 1000.85 768.899 996.187 Q772.302 991.488 772.302 983.265 Q772.302 975.041 768.899 970.383 Q765.496 965.684 759.582 965.684 Q753.668 965.684 750.265 970.383 Q746.903 975.041 746.903 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M833.916 981.361 L833.916 985.007 L799.646 985.007 Q800.132 992.703 804.264 996.754 Q808.436 1000.76 815.849 1000.76 Q820.143 1000.76 824.154 999.711 Q828.204 998.658 832.174 996.552 L832.174 1003.6 Q828.164 1005.3 823.951 1006.19 Q819.738 1007.08 815.404 1007.08 Q804.547 1007.08 798.187 1000.76 Q791.868 994.445 791.868 983.67 Q791.868 972.53 797.863 966.008 Q803.899 959.445 814.107 959.445 Q823.262 959.445 828.569 965.36 Q833.916 971.234 833.916 981.361 M826.463 979.173 Q826.382 973.057 823.019 969.411 Q819.698 965.765 814.188 965.765 Q807.95 965.765 804.183 969.289 Q800.456 972.813 799.889 979.214 L826.463 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M890.426 942.958 Q884.998 952.275 882.365 961.39 Q879.732 970.504 879.732 979.862 Q879.732 989.22 882.365 998.415 Q885.039 1007.57 890.426 1016.85 L883.945 1016.85 Q877.869 1007.33 874.83 998.132 Q871.833 988.936 871.833 979.862 Q871.833 970.829 874.83 961.673 Q877.828 952.518 883.945 942.958 L890.426 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M934.743 967.426 L934.743 942.877 L942.197 942.877 L942.197 1005.91 L934.743 1005.91 L934.743 999.104 Q932.394 1003.15 928.788 1005.14 Q925.224 1007.08 920.2 1007.08 Q911.977 1007.08 906.792 1000.52 Q901.647 993.959 901.647 983.265 Q901.647 972.57 906.792 966.008 Q911.977 959.445 920.2 959.445 Q925.224 959.445 928.788 961.43 Q932.394 963.375 934.743 967.426 M909.344 983.265 Q909.344 991.488 912.706 996.187 Q916.109 1000.85 922.023 1000.85 Q927.938 1000.85 931.34 996.187 Q934.743 991.488 934.743 983.265 Q934.743 975.041 931.34 970.383 Q927.938 965.684 922.023 965.684 Q916.109 965.684 912.706 970.383 Q909.344 975.041 909.344 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M996.357 981.361 L996.357 985.007 L962.087 985.007 Q962.573 992.703 966.705 996.754 Q970.877 1000.76 978.29 1000.76 Q982.584 1000.76 986.595 999.711 Q990.646 998.658 994.616 996.552 L994.616 1003.6 Q990.605 1005.3 986.392 1006.19 Q982.179 1007.08 977.845 1007.08 Q966.988 1007.08 960.629 1000.76 Q954.309 994.445 954.309 983.67 Q954.309 972.53 960.304 966.008 Q966.34 959.445 976.549 959.445 Q985.704 959.445 991.01 965.36 Q996.357 971.234 996.357 981.361 M988.904 979.173 Q988.823 973.057 985.461 969.411 Q982.139 965.765 976.63 965.765 Q970.391 965.765 966.624 969.289 Q962.897 972.813 962.33 979.214 L988.904 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1038.45 982.698 Q1038.45 974.596 1035.08 970.14 Q1031.76 965.684 1025.73 965.684 Q1019.73 965.684 1016.37 970.14 Q1013.05 974.596 1013.05 982.698 Q1013.05 990.759 1016.37 995.215 Q1019.73 999.671 1025.73 999.671 Q1031.76 999.671 1035.08 995.215 Q1038.45 990.759 1038.45 982.698 M1045.9 1000.28 Q1045.9 1011.86 1040.76 1017.49 Q1035.61 1023.17 1025 1023.17 Q1021.07 1023.17 1017.58 1022.56 Q1014.1 1021.99 1010.82 1020.78 L1010.82 1013.53 Q1014.1 1015.31 1017.3 1016.16 Q1020.5 1017.01 1023.82 1017.01 Q1031.15 1017.01 1034.8 1013.16 Q1038.45 1009.35 1038.45 1001.62 L1038.45 997.929 Q1036.14 1001.94 1032.53 1003.92 Q1028.93 1005.91 1023.9 1005.91 Q1015.56 1005.91 1010.45 999.549 Q1005.35 993.189 1005.35 982.698 Q1005.35 972.165 1010.45 965.805 Q1015.56 959.445 1023.9 959.445 Q1028.93 959.445 1032.53 961.43 Q1036.14 963.415 1038.45 967.426 L1038.45 960.539 L1045.9 960.539 L1045.9 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1060.08 942.958 L1066.56 942.958 Q1072.64 952.518 1075.63 961.673 Q1078.67 970.829 1078.67 979.862 Q1078.67 988.936 1075.63 998.132 Q1072.64 1007.33 1066.56 1016.85 L1060.08 1016.85 Q1065.47 1007.57 1068.1 998.415 Q1070.77 989.22 1070.77 979.862 Q1070.77 970.504 1068.1 961.39 Q1065.47 952.275 1060.08 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip214)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,1705.81 195.771,1703.6 198.013,1701.39 200.255,1699.18 202.497,1696.97 204.739,1694.76 206.981,1692.55 209.223,1690.34 211.464,1688.13 213.706,1685.92 215.948,1683.71 218.19,1681.51 220.432,1679.3 222.674,1677.09 224.916,1674.88 227.158,1672.67 229.4,1670.47 231.642,1668.26 233.884,1666.06 236.126,1663.85 238.368,1661.65 240.61,1659.45 242.852,1657.24 245.094,1655.04 247.336,1652.85 249.578,1650.65 251.82,1648.45 254.062,1646.26 256.304,1644.07 258.546,1641.88 260.788,1639.69 263.03,1637.5 265.271,1635.32 267.513,1633.14 269.755,1630.96 271.997,1628.78 274.239,1626.61 276.481,1624.44 278.723,1622.28 280.965,1620.12 283.207,1617.96 285.449,1615.8 287.691,1613.65 289.933,1611.51 292.175,1609.36 294.417,1607.23 296.659,1605.09 298.901,1602.96 301.143,1600.84 303.385,1598.72 305.627,1596.6 307.869,1594.49 310.111,1592.39 312.353,1590.29 314.595,1588.19 316.837,1586.1 319.078,1584.01 321.32,1581.93 323.562,1579.85 325.804,1577.78 328.046,1575.71 330.288,1573.65 332.53,1571.59 334.772,1569.53 337.014,1567.48 339.256,1565.43 341.498,1563.39 343.74,1561.35 345.982,1559.31 348.224,1557.28 350.466,1555.25 352.708,1553.22 354.95,1551.2 357.192,1549.18 359.434,1547.16 361.676,1545.15 363.918,1543.13 366.16,1541.13 368.402,1539.12 370.644,1537.12 372.885,1535.12 375.127,1533.12 377.369,1531.13 379.611,1529.13 381.853,1527.14 384.095,1525.16 386.337,1523.17 388.579,1521.19 390.821,1519.2 393.063,1517.23 395.305,1515.25 397.547,1513.27 399.789,1511.3 402.031,1509.33 404.273,1507.36 406.515,1505.39 408.757,1503.43 410.999,1501.46 413.241,1499.5 415.483,1497.54 417.725,1495.59 419.967,1493.63 422.209,1491.68 424.451,1489.72 426.692,1487.77 428.934,1485.83 431.176,1483.88 433.418,1481.94 435.66,1479.99 437.902,1478.05 440.144,1476.12 442.386,1474.18 444.628,1472.25 446.87,1470.32 449.112,1468.39 451.354,1466.46 453.596,1464.54 455.838,1462.62 458.08,1460.7 460.322,1458.78 462.564,1456.87 464.806,1454.96 467.048,1453.05 469.29,1451.15 471.532,1449.25 473.774,1447.35 476.016,1445.46 478.258,1443.57 480.5,1441.68 482.741,1439.8 484.983,1437.92 487.225,1436.04 489.467,1434.17 491.709,1432.31 493.951,1430.45 496.193,1428.59 498.435,1426.74 500.677,1424.89 502.919,1423.05 505.161,1421.22 507.403,1419.39 509.645,1417.56 511.887,1415.74 514.129,1413.93 516.371,1412.12 518.613,1410.32 520.855,1408.52 523.097,1406.73 525.339,1404.94 527.581,1403.16 529.823,1401.39 532.065,1399.62 534.307,1397.86 536.548,1396.1 538.79,1394.35 541.032,1392.6 543.274,1390.86 545.516,1389.12 547.758,1387.39 550,1385.67 552.242,1383.95 554.484,1382.23 556.726,1380.52 558.968,1378.81 561.21,1377.11 563.452,1375.41 565.694,1373.72 567.936,1372.03 570.178,1370.34 572.42,1368.66 574.662,1366.98 576.904,1365.31 579.146,1363.64 581.388,1361.98 583.63,1360.31 585.872,1358.66 588.114,1357 590.355,1355.35 592.597,1353.71 594.839,1352.06 597.081,1350.43 599.323,1348.79 601.565,1347.16 603.807,1345.53 606.049,1343.91 608.291,1342.29 610.533,1340.67 612.775,1339.06 615.017,1337.45 617.259,1335.84 619.501,1334.24 621.743,1332.65 623.985,1331.05 626.227,1329.47 628.469,1327.88 630.711,1326.3 632.953,1324.73 635.195,1323.16 637.437,1321.6 639.679,1320.04 641.921,1318.48 644.162,1316.93 646.404,1315.39 648.646,1313.85 650.888,1312.32 653.13,1310.79 655.372,1309.27 657.614,1307.75 659.856,1306.24 662.098,1304.73 664.34,1303.24 666.582,1301.74 668.824,1300.26 671.066,1298.78 673.308,1297.3 675.55,1295.84 677.792,1294.38 680.034,1292.92 682.276,1291.47 684.518,1290.03 686.76,1288.59 689.002,1287.16 691.244,1285.74 693.486,1284.32 695.728,1282.91 697.969,1281.5 700.211,1280.1 702.453,1278.7 704.695,1277.31 706.937,1275.93 709.179,1274.55 711.421,1273.17 713.663,1271.81 715.905,1270.44 718.147,1269.08 720.389,1267.73 722.631,1266.38 724.873,1265.04 727.115,1263.7 729.357,1262.37 731.599,1261.04 733.841,1259.72 736.083,1258.4 738.325,1257.08 740.567,1255.78 742.809,1254.47 745.051,1253.17 747.293,1251.88 749.535,1250.59 751.776,1249.3 754.018,1248.02 756.26,1246.75 758.502,1245.48 760.744,1244.22 762.986,1242.96 765.228,1241.7 767.47,1240.46 769.712,1239.21 771.954,1237.98 774.196,1236.74 776.438,1235.52 778.68,1234.3 780.922,1233.09 783.164,1231.88 785.406,1230.68 787.648,1229.48 789.89,1228.29 792.132,1227.11 794.374,1225.93 796.616,1224.76 798.858,1223.6 801.1,1222.44 803.342,1221.29 805.583,1220.14 807.825,1219 810.067,1217.87 812.309,1216.75 814.551,1215.63 816.793,1214.52 819.035,1213.41 821.277,1212.31 823.519,1211.21 825.761,1210.13 828.003,1209.04 830.245,1207.97 832.487,1206.9 834.729,1205.83 836.971,1204.78 839.213,1203.72 841.455,1202.68 843.697,1201.64 845.939,1200.6 848.181,1199.57 850.423,1198.55 852.665,1197.53 854.907,1196.51 857.149,1195.51 859.391,1194.5 861.632,1193.51 863.874,1192.52 866.116,1191.53 868.358,1190.55 870.6,1189.58 872.842,1188.61 875.084,1187.65 877.326,1186.69 879.568,1185.74 881.81,1184.79 884.052,1183.85 886.294,1182.92 888.536,1181.99 890.778,1181.06 893.02,1180.15 895.262,1179.24 897.504,1178.33 899.746,1177.43 901.988,1176.54 904.23,1175.66 906.472,1174.78 908.714,1173.9 910.956,1173.04 913.198,1172.18 915.439,1171.32 917.681,1170.47 919.923,1169.63 922.165,1168.8 924.407,1167.97 926.649,1167.15 928.891,1166.33 931.133,1165.52 933.375,1164.72 935.617,1163.92 937.859,1163.13 940.101,1162.34 942.343,1161.57 944.585,1160.79 946.827,1160.03 949.069,1159.27 951.311,1158.51 953.553,1157.76 955.795,1157.02 958.037,1156.28 960.279,1155.55 962.521,1154.83 964.763,1154.11 967.005,1153.4 969.246,1152.69 971.488,1151.99 973.73,1151.29 975.972,1150.6 978.214,1149.91 980.456,1149.23 982.698,1148.56 984.94,1147.89 987.182,1147.23 989.424,1146.58 991.666,1145.93 993.908,1145.28 996.15,1144.64 998.392,1144.01 1000.63,1143.38 1002.88,1142.76 1005.12,1142.15 1007.36,1141.54 1009.6,1140.94 1011.84,1140.34 1014.09,1139.75 1016.33,1139.16 1018.57,1138.59 1020.81,1138.01 1023.05,1137.45 1025.3,1136.88 1027.54,1136.33 1029.78,1135.78 1032.02,1135.24 1034.26,1134.7 1036.51,1134.17 1038.75,1133.65 1040.99,1133.13 1043.23,1132.61 1045.47,1132.11 1047.72,1131.6 1049.96,1131.11 1052.2,1130.62 1054.44,1130.13 1056.68,1129.65 1058.92,1129.18 1061.17,1128.71 1063.41,1128.25 1065.65,1127.79 1067.89,1127.34 1070.13,1126.9 1072.38,1126.45 1074.62,1126.02 1076.86,1125.59 1079.1,1125.17 1081.34,1124.75 1083.59,1124.33 1085.83,1123.92 1088.07,1123.52 1090.31,1123.12 1092.55,1122.73 1094.8,1122.34 1097.04,1121.96 1099.28,1121.58 1101.52,1121.21 1103.76,1120.85 1106.01,1120.49 1108.25,1120.13 1110.49,1119.78 1112.73,1119.43 1114.97,1119.09 1117.22,1118.76 1119.46,1118.43 1121.7,1118.1 1123.94,1117.78 1126.18,1117.47 1128.43,1117.16 1130.67,1116.86 1132.91,1116.56 1135.15,1116.26 1137.39,1115.97 1139.64,1115.69 1141.88,1115.41 1144.12,1115.13 1146.36,1114.86 1148.6,1114.6 1150.85,1114.34 1153.09,1114.08 1155.33,1113.83 1157.57,1113.58 1159.81,1113.34 1162.05,1113.1 1164.3,1112.87 1166.54,1112.64 1168.78,1112.41 1171.02,1112.19 1173.26,1111.98 1175.51,1111.76 1177.75,1111.56 1179.99,1111.35 1182.23,1111.15 1184.47,1110.96 1186.72,1110.77 1188.96,1110.58 1191.2,1110.4 1193.44,1110.22 1195.68,1110.04 1197.93,1109.87 1200.17,1109.7 1202.41,1109.54 1204.65,1109.38 1206.89,1109.22 1209.14,1109.07 1211.38,1108.92 1213.62,1108.77 1215.86,1108.63 1218.1,1108.49 1220.35,1108.36 1222.59,1108.23 1224.83,1108.1 1227.07,1107.97 1229.31,1107.85 1231.56,1107.73 1233.8,1107.62 1236.04,1107.51 1238.28,1107.4 1240.52,1107.29 1242.77,1107.19 1245.01,1107.09 1247.25,1107 1249.49,1106.9 1251.73,1106.81 1253.98,1106.72 1256.22,1106.64 1258.46,1106.55 1260.7,1106.47 1262.94,1106.4 1265.19,1106.32 1267.43,1106.25 1269.67,1106.17 1271.91,1106.11 1274.15,1106.04 1276.39,1105.97 1278.64,1105.91 1280.88,1105.85 1283.12,1105.79 1285.36,1105.74 1287.6,1105.68 1289.85,1105.63 1292.09,1105.58 1294.33,1105.53 1296.57,1105.48 1298.81,1105.43 1301.06,1105.39 1303.3,1105.34 1305.54,1105.3 1307.78,1105.26 1310.02,1105.22 1312.27,1105.18 1314.51,1105.14 1316.75,1105.1 1318.99,1105.07 "/>
-<path clip-path="url(#clip210)" d="M1559.76 1723.83 L2752.76 1723.83 L2752.76 1087.05 L1559.76 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,1723.83 193.529,1087.05 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,1723.83 473.774,1087.05 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,1723.83 754.018,1087.05 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,1723.83 1034.26,1087.05 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,1723.83 1314.51,1087.05 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1723.83 1352.76,1723.83 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,1723.83 193.529,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,1723.83 473.774,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,1723.83 754.018,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,1723.83 1034.26,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,1723.83 1314.51,1704.94 "/>
+<path clip-path="url(#clip030)" d="M193.529 1757.95 Q189.918 1757.95 188.089 1761.52 Q186.283 1765.06 186.283 1772.19 Q186.283 1779.3 188.089 1782.86 Q189.918 1786.4 193.529 1786.4 Q197.163 1786.4 198.969 1782.86 Q200.797 1779.3 200.797 1772.19 Q200.797 1765.06 198.969 1761.52 Q197.163 1757.95 193.529 1757.95 M193.529 1754.25 Q199.339 1754.25 202.394 1758.86 Q205.473 1763.44 205.473 1772.19 Q205.473 1780.92 202.394 1785.52 Q199.339 1790.11 193.529 1790.11 Q187.719 1790.11 184.64 1785.52 Q181.584 1780.92 181.584 1772.19 Q181.584 1763.44 184.64 1758.86 Q187.719 1754.25 193.529 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M433.392 1754.87 L451.748 1754.87 L451.748 1758.81 L437.674 1758.81 L437.674 1767.28 Q438.693 1766.93 439.711 1766.77 Q440.73 1766.59 441.748 1766.59 Q447.535 1766.59 450.915 1769.76 Q454.295 1772.93 454.295 1778.35 Q454.295 1783.93 450.822 1787.03 Q447.35 1790.11 441.031 1790.11 Q438.855 1790.11 436.586 1789.74 Q434.341 1789.37 431.934 1788.62 L431.934 1783.93 Q434.017 1785.06 436.239 1785.62 Q438.461 1786.17 440.938 1786.17 Q444.943 1786.17 447.281 1784.06 Q449.619 1781.96 449.619 1778.35 Q449.619 1774.74 447.281 1772.63 Q444.943 1770.52 440.938 1770.52 Q439.063 1770.52 437.188 1770.94 Q435.336 1771.36 433.392 1772.24 L433.392 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M473.507 1757.95 Q469.896 1757.95 468.068 1761.52 Q466.262 1765.06 466.262 1772.19 Q466.262 1779.3 468.068 1782.86 Q469.896 1786.4 473.507 1786.4 Q477.142 1786.4 478.947 1782.86 Q480.776 1779.3 480.776 1772.19 Q480.776 1765.06 478.947 1761.52 Q477.142 1757.95 473.507 1757.95 M473.507 1754.25 Q479.318 1754.25 482.373 1758.86 Q485.452 1763.44 485.452 1772.19 Q485.452 1780.92 482.373 1785.52 Q479.318 1790.11 473.507 1790.11 Q467.697 1790.11 464.619 1785.52 Q461.563 1780.92 461.563 1772.19 Q461.563 1763.44 464.619 1758.86 Q467.697 1754.25 473.507 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M503.669 1757.95 Q500.058 1757.95 498.23 1761.52 Q496.424 1765.06 496.424 1772.19 Q496.424 1779.3 498.23 1782.86 Q500.058 1786.4 503.669 1786.4 Q507.304 1786.4 509.109 1782.86 Q510.938 1779.3 510.938 1772.19 Q510.938 1765.06 509.109 1761.52 Q507.304 1757.95 503.669 1757.95 M503.669 1754.25 Q509.479 1754.25 512.535 1758.86 Q515.614 1763.44 515.614 1772.19 Q515.614 1780.92 512.535 1785.52 Q509.479 1790.11 503.669 1790.11 Q497.859 1790.11 494.78 1785.52 Q491.725 1780.92 491.725 1772.19 Q491.725 1763.44 494.78 1758.86 Q497.859 1754.25 503.669 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M698.544 1785.5 L706.183 1785.5 L706.183 1759.13 L697.873 1760.8 L697.873 1756.54 L706.137 1754.87 L710.813 1754.87 L710.813 1785.5 L718.452 1785.5 L718.452 1789.43 L698.544 1789.43 L698.544 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M737.896 1757.95 Q734.285 1757.95 732.456 1761.52 Q730.651 1765.06 730.651 1772.19 Q730.651 1779.3 732.456 1782.86 Q734.285 1786.4 737.896 1786.4 Q741.53 1786.4 743.336 1782.86 Q745.164 1779.3 745.164 1772.19 Q745.164 1765.06 743.336 1761.52 Q741.53 1757.95 737.896 1757.95 M737.896 1754.25 Q743.706 1754.25 746.762 1758.86 Q749.84 1763.44 749.84 1772.19 Q749.84 1780.92 746.762 1785.52 Q743.706 1790.11 737.896 1790.11 Q732.086 1790.11 729.007 1785.52 Q725.951 1780.92 725.951 1772.19 Q725.951 1763.44 729.007 1758.86 Q732.086 1754.25 737.896 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M768.058 1757.95 Q764.447 1757.95 762.618 1761.52 Q760.812 1765.06 760.812 1772.19 Q760.812 1779.3 762.618 1782.86 Q764.447 1786.4 768.058 1786.4 Q771.692 1786.4 773.498 1782.86 Q775.326 1779.3 775.326 1772.19 Q775.326 1765.06 773.498 1761.52 Q771.692 1757.95 768.058 1757.95 M768.058 1754.25 Q773.868 1754.25 776.923 1758.86 Q780.002 1763.44 780.002 1772.19 Q780.002 1780.92 776.923 1785.52 Q773.868 1790.11 768.058 1790.11 Q762.248 1790.11 759.169 1785.52 Q756.113 1780.92 756.113 1772.19 Q756.113 1763.44 759.169 1758.86 Q762.248 1754.25 768.058 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M798.22 1757.95 Q794.609 1757.95 792.78 1761.52 Q790.974 1765.06 790.974 1772.19 Q790.974 1779.3 792.78 1782.86 Q794.609 1786.4 798.22 1786.4 Q801.854 1786.4 803.659 1782.86 Q805.488 1779.3 805.488 1772.19 Q805.488 1765.06 803.659 1761.52 Q801.854 1757.95 798.22 1757.95 M798.22 1754.25 Q804.03 1754.25 807.085 1758.86 Q810.164 1763.44 810.164 1772.19 Q810.164 1780.92 807.085 1785.52 Q804.03 1790.11 798.22 1790.11 Q792.409 1790.11 789.331 1785.52 Q786.275 1780.92 786.275 1772.19 Q786.275 1763.44 789.331 1758.86 Q792.409 1754.25 798.22 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M978.789 1785.5 L986.428 1785.5 L986.428 1759.13 L978.118 1760.8 L978.118 1756.54 L986.382 1754.87 L991.057 1754.87 L991.057 1785.5 L998.696 1785.5 L998.696 1789.43 L978.789 1789.43 L978.789 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1008.19 1754.87 L1026.54 1754.87 L1026.54 1758.81 L1012.47 1758.81 L1012.47 1767.28 Q1013.49 1766.93 1014.51 1766.77 Q1015.52 1766.59 1016.54 1766.59 Q1022.33 1766.59 1025.71 1769.76 Q1029.09 1772.93 1029.09 1778.35 Q1029.09 1783.93 1025.62 1787.03 Q1022.15 1790.11 1015.83 1790.11 Q1013.65 1790.11 1011.38 1789.74 Q1009.14 1789.37 1006.73 1788.62 L1006.73 1783.93 Q1008.81 1785.06 1011.03 1785.62 Q1013.26 1786.17 1015.73 1786.17 Q1019.74 1786.17 1022.08 1784.06 Q1024.41 1781.96 1024.41 1778.35 Q1024.41 1774.74 1022.08 1772.63 Q1019.74 1770.52 1015.73 1770.52 Q1013.86 1770.52 1011.98 1770.94 Q1010.13 1771.36 1008.19 1772.24 L1008.19 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1048.3 1757.95 Q1044.69 1757.95 1042.86 1761.52 Q1041.06 1765.06 1041.06 1772.19 Q1041.06 1779.3 1042.86 1782.86 Q1044.69 1786.4 1048.3 1786.4 Q1051.94 1786.4 1053.74 1782.86 Q1055.57 1779.3 1055.57 1772.19 Q1055.57 1765.06 1053.74 1761.52 Q1051.94 1757.95 1048.3 1757.95 M1048.3 1754.25 Q1054.11 1754.25 1057.17 1758.86 Q1060.25 1763.44 1060.25 1772.19 Q1060.25 1780.92 1057.17 1785.52 Q1054.11 1790.11 1048.3 1790.11 Q1042.49 1790.11 1039.41 1785.52 Q1036.36 1780.92 1036.36 1772.19 Q1036.36 1763.44 1039.41 1758.86 Q1042.49 1754.25 1048.3 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1078.46 1757.95 Q1074.85 1757.95 1073.02 1761.52 Q1071.22 1765.06 1071.22 1772.19 Q1071.22 1779.3 1073.02 1782.86 Q1074.85 1786.4 1078.46 1786.4 Q1082.1 1786.4 1083.9 1782.86 Q1085.73 1779.3 1085.73 1772.19 Q1085.73 1765.06 1083.9 1761.52 Q1082.1 1757.95 1078.46 1757.95 M1078.46 1754.25 Q1084.27 1754.25 1087.33 1758.86 Q1090.41 1763.44 1090.41 1772.19 Q1090.41 1780.92 1087.33 1785.52 Q1084.27 1790.11 1078.46 1790.11 Q1072.65 1790.11 1069.58 1785.52 Q1066.52 1780.92 1066.52 1772.19 Q1066.52 1763.44 1069.58 1758.86 Q1072.65 1754.25 1078.46 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1263.12 1785.5 L1279.44 1785.5 L1279.44 1789.43 L1257.49 1789.43 L1257.49 1785.5 Q1260.16 1782.74 1264.74 1778.12 Q1269.35 1773.46 1270.53 1772.12 Q1272.77 1769.6 1273.65 1767.86 Q1274.55 1766.1 1274.55 1764.41 Q1274.55 1761.66 1272.61 1759.92 Q1270.69 1758.18 1267.59 1758.18 Q1265.39 1758.18 1262.93 1758.95 Q1260.5 1759.71 1257.73 1761.26 L1257.73 1756.54 Q1260.55 1755.41 1263 1754.83 Q1265.46 1754.25 1267.49 1754.25 Q1272.86 1754.25 1276.06 1756.93 Q1279.25 1759.62 1279.25 1764.11 Q1279.25 1766.24 1278.44 1768.16 Q1277.66 1770.06 1275.55 1772.65 Q1274.97 1773.32 1271.87 1776.54 Q1268.77 1779.74 1263.12 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1299.25 1757.95 Q1295.64 1757.95 1293.81 1761.52 Q1292.01 1765.06 1292.01 1772.19 Q1292.01 1779.3 1293.81 1782.86 Q1295.64 1786.4 1299.25 1786.4 Q1302.89 1786.4 1304.69 1782.86 Q1306.52 1779.3 1306.52 1772.19 Q1306.52 1765.06 1304.69 1761.52 Q1302.89 1757.95 1299.25 1757.95 M1299.25 1754.25 Q1305.06 1754.25 1308.12 1758.86 Q1311.2 1763.44 1311.2 1772.19 Q1311.2 1780.92 1308.12 1785.52 Q1305.06 1790.11 1299.25 1790.11 Q1293.44 1790.11 1290.36 1785.52 Q1287.31 1780.92 1287.31 1772.19 Q1287.31 1763.44 1290.36 1758.86 Q1293.44 1754.25 1299.25 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1329.42 1757.95 Q1325.8 1757.95 1323.98 1761.52 Q1322.17 1765.06 1322.17 1772.19 Q1322.17 1779.3 1323.98 1782.86 Q1325.8 1786.4 1329.42 1786.4 Q1333.05 1786.4 1334.86 1782.86 Q1336.68 1779.3 1336.68 1772.19 Q1336.68 1765.06 1334.86 1761.52 Q1333.05 1757.95 1329.42 1757.95 M1329.42 1754.25 Q1335.23 1754.25 1338.28 1758.86 Q1341.36 1763.44 1341.36 1772.19 Q1341.36 1780.92 1338.28 1785.52 Q1335.23 1790.11 1329.42 1790.11 Q1323.61 1790.11 1320.53 1785.52 Q1317.47 1780.92 1317.47 1772.19 Q1317.47 1763.44 1320.53 1758.86 Q1323.61 1754.25 1329.42 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1359.58 1757.95 Q1355.97 1757.95 1354.14 1761.52 Q1352.33 1765.06 1352.33 1772.19 Q1352.33 1779.3 1354.14 1782.86 Q1355.97 1786.4 1359.58 1786.4 Q1363.21 1786.4 1365.02 1782.86 Q1366.85 1779.3 1366.85 1772.19 Q1366.85 1765.06 1365.02 1761.52 Q1363.21 1757.95 1359.58 1757.95 M1359.58 1754.25 Q1365.39 1754.25 1368.44 1758.86 Q1371.52 1763.44 1371.52 1772.19 Q1371.52 1780.92 1368.44 1785.52 Q1365.39 1790.11 1359.58 1790.11 Q1353.77 1790.11 1350.69 1785.52 Q1347.63 1780.92 1347.63 1772.19 Q1347.63 1763.44 1350.69 1758.86 Q1353.77 1754.25 1359.58 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1705.81 1352.76,1705.81 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1546.36 1352.76,1546.36 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1386.91 1352.76,1386.91 "/>
+<polyline clip-path="url(#clip034)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,1227.47 1352.76,1227.47 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1723.83 159.765,1087.05 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1705.81 178.663,1705.81 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1546.36 178.663,1546.36 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1386.91 178.663,1386.91 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,1227.47 178.663,1227.47 "/>
+<path clip-path="url(#clip030)" d="M105.821 1691.61 Q102.209 1691.61 100.381 1695.18 Q98.5752 1698.72 98.5752 1705.85 Q98.5752 1712.95 100.381 1716.52 Q102.209 1720.06 105.821 1720.06 Q109.455 1720.06 111.26 1716.52 Q113.089 1712.95 113.089 1705.85 Q113.089 1698.72 111.26 1695.18 Q109.455 1691.61 105.821 1691.61 M105.821 1687.91 Q111.631 1687.91 114.686 1692.51 Q117.765 1697.1 117.765 1705.85 Q117.765 1714.57 114.686 1719.18 Q111.631 1723.76 105.821 1723.76 Q100.01 1723.76 96.9317 1719.18 Q93.8761 1714.57 93.8761 1705.85 Q93.8761 1697.1 96.9317 1692.51 Q100.01 1687.91 105.821 1687.91 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M69.6865 1559.71 L86.0058 1559.71 L86.0058 1563.64 L64.0615 1563.64 L64.0615 1559.71 Q66.7235 1556.95 71.3068 1552.32 Q75.9133 1547.67 77.0938 1546.33 Q79.3392 1543.81 80.2188 1542.07 Q81.1216 1540.31 81.1216 1538.62 Q81.1216 1535.87 79.1771 1534.13 Q77.2559 1532.39 74.154 1532.39 Q71.955 1532.39 69.5013 1533.16 Q67.0707 1533.92 64.293 1535.47 L64.293 1530.75 Q67.117 1529.62 69.5707 1529.04 Q72.0244 1528.46 74.0614 1528.46 Q79.4318 1528.46 82.6262 1531.14 Q85.8206 1533.83 85.8206 1538.32 Q85.8206 1540.45 85.0105 1542.37 Q84.2234 1544.27 82.1169 1546.86 Q81.5382 1547.53 78.4364 1550.75 Q75.3346 1553.94 69.6865 1559.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 1532.16 Q102.209 1532.16 100.381 1535.73 Q98.5752 1539.27 98.5752 1546.4 Q98.5752 1553.5 100.381 1557.07 Q102.209 1560.61 105.821 1560.61 Q109.455 1560.61 111.26 1557.07 Q113.089 1553.5 113.089 1546.4 Q113.089 1539.27 111.26 1535.73 Q109.455 1532.16 105.821 1532.16 M105.821 1528.46 Q111.631 1528.46 114.686 1533.06 Q117.765 1537.65 117.765 1546.4 Q117.765 1555.12 114.686 1559.73 Q111.631 1564.31 105.821 1564.31 Q100.01 1564.31 96.9317 1559.73 Q93.8761 1555.12 93.8761 1546.4 Q93.8761 1537.65 96.9317 1533.06 Q100.01 1528.46 105.821 1528.46 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M78.5059 1373.71 L66.7004 1392.16 L78.5059 1392.16 L78.5059 1373.71 M77.279 1369.63 L83.1586 1369.63 L83.1586 1392.16 L88.0891 1392.16 L88.0891 1396.05 L83.1586 1396.05 L83.1586 1404.19 L78.5059 1404.19 L78.5059 1396.05 L62.9041 1396.05 L62.9041 1391.53 L77.279 1369.63 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 1372.71 Q102.209 1372.71 100.381 1376.28 Q98.5752 1379.82 98.5752 1386.95 Q98.5752 1394.06 100.381 1397.62 Q102.209 1401.16 105.821 1401.16 Q109.455 1401.16 111.26 1397.62 Q113.089 1394.06 113.089 1386.95 Q113.089 1379.82 111.26 1376.28 Q109.455 1372.71 105.821 1372.71 M105.821 1369.01 Q111.631 1369.01 114.686 1373.62 Q117.765 1378.2 117.765 1386.95 Q117.765 1395.68 114.686 1400.28 Q111.631 1404.87 105.821 1404.87 Q100.01 1404.87 96.9317 1400.28 Q93.8761 1395.68 93.8761 1386.95 Q93.8761 1378.2 96.9317 1373.62 Q100.01 1369.01 105.821 1369.01 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M76.2374 1225.6 Q73.0892 1225.6 71.2374 1227.75 Q69.4087 1229.91 69.4087 1233.66 Q69.4087 1237.38 71.2374 1239.56 Q73.0892 1241.71 76.2374 1241.71 Q79.3855 1241.71 81.2142 1239.56 Q83.066 1237.38 83.066 1233.66 Q83.066 1229.91 81.2142 1227.75 Q79.3855 1225.6 76.2374 1225.6 M85.5197 1210.95 L85.5197 1215.21 Q83.7605 1214.38 81.9549 1213.94 Q80.1725 1213.5 78.4133 1213.5 Q73.7837 1213.5 71.33 1216.62 Q68.8994 1219.75 68.5522 1226.06 Q69.9179 1224.05 71.9781 1222.99 Q74.0383 1221.9 76.5151 1221.9 Q81.7234 1221.9 84.7327 1225.07 Q87.7651 1228.22 87.7651 1233.66 Q87.7651 1238.98 84.6169 1242.2 Q81.4688 1245.42 76.2374 1245.42 Q70.242 1245.42 67.0707 1240.83 Q63.8995 1236.23 63.8995 1227.5 Q63.8995 1219.31 67.7883 1214.44 Q71.6772 1209.56 78.2281 1209.56 Q79.9873 1209.56 81.7697 1209.91 Q83.5753 1210.25 85.5197 1210.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 1213.26 Q102.209 1213.26 100.381 1216.83 Q98.5752 1220.37 98.5752 1227.5 Q98.5752 1234.61 100.381 1238.17 Q102.209 1241.71 105.821 1241.71 Q109.455 1241.71 111.26 1238.17 Q113.089 1234.61 113.089 1227.5 Q113.089 1220.37 111.26 1216.83 Q109.455 1213.26 105.821 1213.26 M105.821 1209.56 Q111.631 1209.56 114.686 1214.17 Q117.765 1218.75 117.765 1227.5 Q117.765 1236.23 114.686 1240.83 Q111.631 1245.42 105.821 1245.42 Q100.01 1245.42 96.9317 1240.83 Q93.8761 1236.23 93.8761 1227.5 Q93.8761 1218.75 96.9317 1214.17 Q100.01 1209.56 105.821 1209.56 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M433.849 945.429 L442.032 945.429 L442.032 999.023 L471.482 999.023 L471.482 1005.91 L433.849 1005.91 L433.849 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M495.868 965.765 Q489.873 965.765 486.389 970.464 Q482.905 975.122 482.905 983.265 Q482.905 991.407 486.349 996.106 Q489.832 1000.76 495.868 1000.76 Q501.823 1000.76 505.307 996.066 Q508.791 991.367 508.791 983.265 Q508.791 975.203 505.307 970.504 Q501.823 965.765 495.868 965.765 M495.868 959.445 Q505.59 959.445 511.14 965.765 Q516.69 972.084 516.69 983.265 Q516.69 994.405 511.14 1000.76 Q505.59 1007.08 495.868 1007.08 Q486.106 1007.08 480.556 1000.76 Q475.047 994.405 475.047 983.265 Q475.047 972.084 480.556 965.765 Q486.106 959.445 495.868 959.445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M566.759 978.525 L566.759 1005.91 L559.305 1005.91 L559.305 978.768 Q559.305 972.327 556.794 969.127 Q554.282 965.927 549.259 965.927 Q543.223 965.927 539.74 969.775 Q536.256 973.624 536.256 980.267 L536.256 1005.91 L528.762 1005.91 L528.762 960.539 L536.256 960.539 L536.256 967.588 Q538.929 963.496 542.535 961.471 Q546.181 959.445 550.92 959.445 Q558.738 959.445 562.749 964.307 Q566.759 969.127 566.759 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M611.481 982.698 Q611.481 974.596 608.119 970.14 Q604.797 965.684 598.761 965.684 Q592.766 965.684 589.404 970.14 Q586.082 974.596 586.082 982.698 Q586.082 990.759 589.404 995.215 Q592.766 999.671 598.761 999.671 Q604.797 999.671 608.119 995.215 Q611.481 990.759 611.481 982.698 M618.935 1000.28 Q618.935 1011.86 613.79 1017.49 Q608.645 1023.17 598.032 1023.17 Q594.103 1023.17 590.619 1022.56 Q587.135 1021.99 583.854 1020.78 L583.854 1013.53 Q587.135 1015.31 590.335 1016.16 Q593.536 1017.01 596.857 1017.01 Q604.189 1017.01 607.835 1013.16 Q611.481 1009.35 611.481 1001.62 L611.481 997.929 Q609.172 1001.94 605.567 1003.92 Q601.961 1005.91 596.938 1005.91 Q588.593 1005.91 583.489 999.549 Q578.385 993.189 578.385 982.698 Q578.385 972.165 583.489 965.805 Q588.593 959.445 596.938 959.445 Q601.961 959.445 605.567 961.43 Q609.172 963.415 611.481 967.426 L611.481 960.539 L618.935 960.539 L618.935 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M634.288 960.539 L641.741 960.539 L641.741 1005.91 L634.288 1005.91 L634.288 960.539 M634.288 942.877 L641.741 942.877 L641.741 952.316 L634.288 952.316 L634.288 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M664.71 947.657 L664.71 960.539 L680.063 960.539 L680.063 966.332 L664.71 966.332 L664.71 990.961 Q664.71 996.511 666.209 998.091 Q667.748 999.671 672.407 999.671 L680.063 999.671 L680.063 1005.91 L672.407 1005.91 Q663.778 1005.91 660.497 1002.71 Q657.216 999.468 657.216 990.961 L657.216 966.332 L651.747 966.332 L651.747 960.539 L657.216 960.539 L657.216 947.657 L664.71 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M689.096 988.004 L689.096 960.539 L696.55 960.539 L696.55 987.721 Q696.55 994.162 699.062 997.402 Q701.573 1000.6 706.596 1000.6 Q712.632 1000.6 716.116 996.754 Q719.64 992.906 719.64 986.262 L719.64 960.539 L727.094 960.539 L727.094 1005.91 L719.64 1005.91 L719.64 998.942 Q716.926 1003.07 713.321 1005.1 Q709.756 1007.08 705.016 1007.08 Q697.198 1007.08 693.147 1002.22 Q689.096 997.362 689.096 988.004 M707.852 959.445 L707.852 959.445 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M772.302 967.426 L772.302 942.877 L779.756 942.877 L779.756 1005.91 L772.302 1005.91 L772.302 999.104 Q769.952 1003.15 766.347 1005.14 Q762.782 1007.08 757.759 1007.08 Q749.536 1007.08 744.351 1000.52 Q739.206 993.959 739.206 983.265 Q739.206 972.57 744.351 966.008 Q749.536 959.445 757.759 959.445 Q762.782 959.445 766.347 961.43 Q769.952 963.375 772.302 967.426 M746.903 983.265 Q746.903 991.488 750.265 996.187 Q753.668 1000.85 759.582 1000.85 Q765.496 1000.85 768.899 996.187 Q772.302 991.488 772.302 983.265 Q772.302 975.041 768.899 970.383 Q765.496 965.684 759.582 965.684 Q753.668 965.684 750.265 970.383 Q746.903 975.041 746.903 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M833.916 981.361 L833.916 985.007 L799.646 985.007 Q800.132 992.703 804.264 996.754 Q808.436 1000.76 815.849 1000.76 Q820.143 1000.76 824.154 999.711 Q828.204 998.658 832.174 996.552 L832.174 1003.6 Q828.164 1005.3 823.951 1006.19 Q819.738 1007.08 815.404 1007.08 Q804.547 1007.08 798.187 1000.76 Q791.868 994.445 791.868 983.67 Q791.868 972.53 797.863 966.008 Q803.899 959.445 814.107 959.445 Q823.262 959.445 828.569 965.36 Q833.916 971.234 833.916 981.361 M826.463 979.173 Q826.382 973.057 823.019 969.411 Q819.698 965.765 814.188 965.765 Q807.95 965.765 804.183 969.289 Q800.456 972.813 799.889 979.214 L826.463 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M890.426 942.958 Q884.998 952.275 882.365 961.39 Q879.732 970.504 879.732 979.862 Q879.732 989.22 882.365 998.415 Q885.039 1007.57 890.426 1016.85 L883.945 1016.85 Q877.869 1007.33 874.83 998.132 Q871.833 988.936 871.833 979.862 Q871.833 970.829 874.83 961.673 Q877.828 952.518 883.945 942.958 L890.426 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M934.743 967.426 L934.743 942.877 L942.197 942.877 L942.197 1005.91 L934.743 1005.91 L934.743 999.104 Q932.394 1003.15 928.788 1005.14 Q925.224 1007.08 920.2 1007.08 Q911.977 1007.08 906.792 1000.52 Q901.647 993.959 901.647 983.265 Q901.647 972.57 906.792 966.008 Q911.977 959.445 920.2 959.445 Q925.224 959.445 928.788 961.43 Q932.394 963.375 934.743 967.426 M909.344 983.265 Q909.344 991.488 912.706 996.187 Q916.109 1000.85 922.023 1000.85 Q927.938 1000.85 931.34 996.187 Q934.743 991.488 934.743 983.265 Q934.743 975.041 931.34 970.383 Q927.938 965.684 922.023 965.684 Q916.109 965.684 912.706 970.383 Q909.344 975.041 909.344 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M996.357 981.361 L996.357 985.007 L962.087 985.007 Q962.573 992.703 966.705 996.754 Q970.877 1000.76 978.29 1000.76 Q982.584 1000.76 986.595 999.711 Q990.646 998.658 994.616 996.552 L994.616 1003.6 Q990.605 1005.3 986.392 1006.19 Q982.179 1007.08 977.845 1007.08 Q966.988 1007.08 960.629 1000.76 Q954.309 994.445 954.309 983.67 Q954.309 972.53 960.304 966.008 Q966.34 959.445 976.549 959.445 Q985.704 959.445 991.01 965.36 Q996.357 971.234 996.357 981.361 M988.904 979.173 Q988.823 973.057 985.461 969.411 Q982.139 965.765 976.63 965.765 Q970.391 965.765 966.624 969.289 Q962.897 972.813 962.33 979.214 L988.904 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1038.45 982.698 Q1038.45 974.596 1035.08 970.14 Q1031.76 965.684 1025.73 965.684 Q1019.73 965.684 1016.37 970.14 Q1013.05 974.596 1013.05 982.698 Q1013.05 990.759 1016.37 995.215 Q1019.73 999.671 1025.73 999.671 Q1031.76 999.671 1035.08 995.215 Q1038.45 990.759 1038.45 982.698 M1045.9 1000.28 Q1045.9 1011.86 1040.76 1017.49 Q1035.61 1023.17 1025 1023.17 Q1021.07 1023.17 1017.58 1022.56 Q1014.1 1021.99 1010.82 1020.78 L1010.82 1013.53 Q1014.1 1015.31 1017.3 1016.16 Q1020.5 1017.01 1023.82 1017.01 Q1031.15 1017.01 1034.8 1013.16 Q1038.45 1009.35 1038.45 1001.62 L1038.45 997.929 Q1036.14 1001.94 1032.53 1003.92 Q1028.93 1005.91 1023.9 1005.91 Q1015.56 1005.91 1010.45 999.549 Q1005.35 993.189 1005.35 982.698 Q1005.35 972.165 1010.45 965.805 Q1015.56 959.445 1023.9 959.445 Q1028.93 959.445 1032.53 961.43 Q1036.14 963.415 1038.45 967.426 L1038.45 960.539 L1045.9 960.539 L1045.9 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1060.08 942.958 L1066.56 942.958 Q1072.64 952.518 1075.63 961.673 Q1078.67 970.829 1078.67 979.862 Q1078.67 988.936 1075.63 998.132 Q1072.64 1007.33 1066.56 1016.85 L1060.08 1016.85 Q1065.47 1007.57 1068.1 998.415 Q1070.77 989.22 1070.77 979.862 Q1070.77 970.504 1068.1 961.39 Q1065.47 952.275 1060.08 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip034)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,1705.81 195.771,1703.6 198.013,1701.39 200.255,1699.18 202.497,1696.97 204.739,1694.76 206.981,1692.55 209.223,1690.34 211.464,1688.13 213.706,1685.92 215.948,1683.71 218.19,1681.51 220.432,1679.3 222.674,1677.09 224.916,1674.88 227.158,1672.67 229.4,1670.47 231.642,1668.26 233.884,1666.06 236.126,1663.85 238.368,1661.65 240.61,1659.45 242.852,1657.24 245.094,1655.04 247.336,1652.85 249.578,1650.65 251.82,1648.45 254.062,1646.26 256.304,1644.07 258.546,1641.88 260.788,1639.69 263.03,1637.5 265.271,1635.32 267.513,1633.14 269.755,1630.96 271.997,1628.78 274.239,1626.61 276.481,1624.44 278.723,1622.28 280.965,1620.12 283.207,1617.96 285.449,1615.8 287.691,1613.65 289.933,1611.51 292.175,1609.36 294.417,1607.23 296.659,1605.09 298.901,1602.96 301.143,1600.84 303.385,1598.72 305.627,1596.6 307.869,1594.49 310.111,1592.39 312.353,1590.29 314.595,1588.19 316.837,1586.1 319.078,1584.01 321.32,1581.93 323.562,1579.85 325.804,1577.78 328.046,1575.71 330.288,1573.65 332.53,1571.59 334.772,1569.53 337.014,1567.48 339.256,1565.43 341.498,1563.39 343.74,1561.35 345.982,1559.31 348.224,1557.28 350.466,1555.25 352.708,1553.22 354.95,1551.2 357.192,1549.18 359.434,1547.16 361.676,1545.15 363.918,1543.13 366.16,1541.13 368.402,1539.12 370.644,1537.12 372.885,1535.12 375.127,1533.12 377.369,1531.13 379.611,1529.13 381.853,1527.14 384.095,1525.16 386.337,1523.17 388.579,1521.19 390.821,1519.2 393.063,1517.23 395.305,1515.25 397.547,1513.27 399.789,1511.3 402.031,1509.33 404.273,1507.36 406.515,1505.39 408.757,1503.43 410.999,1501.46 413.241,1499.5 415.483,1497.54 417.725,1495.59 419.967,1493.63 422.209,1491.68 424.451,1489.72 426.692,1487.77 428.934,1485.83 431.176,1483.88 433.418,1481.94 435.66,1479.99 437.902,1478.05 440.144,1476.12 442.386,1474.18 444.628,1472.25 446.87,1470.32 449.112,1468.39 451.354,1466.46 453.596,1464.54 455.838,1462.62 458.08,1460.7 460.322,1458.78 462.564,1456.87 464.806,1454.96 467.048,1453.05 469.29,1451.15 471.532,1449.25 473.774,1447.35 476.016,1445.46 478.258,1443.57 480.5,1441.68 482.741,1439.8 484.983,1437.92 487.225,1436.04 489.467,1434.17 491.709,1432.31 493.951,1430.45 496.193,1428.59 498.435,1426.74 500.677,1424.89 502.919,1423.05 505.161,1421.22 507.403,1419.39 509.645,1417.56 511.887,1415.74 514.129,1413.93 516.371,1412.12 518.613,1410.32 520.855,1408.52 523.097,1406.73 525.339,1404.94 527.581,1403.16 529.823,1401.39 532.065,1399.62 534.307,1397.86 536.548,1396.1 538.79,1394.35 541.032,1392.6 543.274,1390.86 545.516,1389.12 547.758,1387.39 550,1385.67 552.242,1383.95 554.484,1382.23 556.726,1380.52 558.968,1378.81 561.21,1377.11 563.452,1375.41 565.694,1373.72 567.936,1372.03 570.178,1370.34 572.42,1368.66 574.662,1366.98 576.904,1365.31 579.146,1363.64 581.388,1361.98 583.63,1360.31 585.872,1358.66 588.114,1357 590.355,1355.35 592.597,1353.71 594.839,1352.06 597.081,1350.43 599.323,1348.79 601.565,1347.16 603.807,1345.53 606.049,1343.91 608.291,1342.29 610.533,1340.67 612.775,1339.06 615.017,1337.45 617.259,1335.84 619.501,1334.24 621.743,1332.65 623.985,1331.05 626.227,1329.47 628.469,1327.88 630.711,1326.3 632.953,1324.73 635.195,1323.16 637.437,1321.6 639.679,1320.04 641.921,1318.48 644.162,1316.93 646.404,1315.39 648.646,1313.85 650.888,1312.32 653.13,1310.79 655.372,1309.27 657.614,1307.75 659.856,1306.24 662.098,1304.73 664.34,1303.24 666.582,1301.74 668.824,1300.26 671.066,1298.78 673.308,1297.3 675.55,1295.84 677.792,1294.38 680.034,1292.92 682.276,1291.47 684.518,1290.03 686.76,1288.59 689.002,1287.16 691.244,1285.74 693.486,1284.32 695.728,1282.91 697.969,1281.5 700.211,1280.1 702.453,1278.7 704.695,1277.31 706.937,1275.93 709.179,1274.55 711.421,1273.17 713.663,1271.81 715.905,1270.44 718.147,1269.08 720.389,1267.73 722.631,1266.38 724.873,1265.04 727.115,1263.7 729.357,1262.37 731.599,1261.04 733.841,1259.72 736.083,1258.4 738.325,1257.08 740.567,1255.78 742.809,1254.47 745.051,1253.17 747.293,1251.88 749.535,1250.59 751.776,1249.3 754.018,1248.02 756.26,1246.75 758.502,1245.48 760.744,1244.22 762.986,1242.96 765.228,1241.7 767.47,1240.46 769.712,1239.21 771.954,1237.98 774.196,1236.74 776.438,1235.52 778.68,1234.3 780.922,1233.09 783.164,1231.88 785.406,1230.68 787.648,1229.48 789.89,1228.29 792.132,1227.11 794.374,1225.93 796.616,1224.76 798.858,1223.6 801.1,1222.44 803.342,1221.29 805.583,1220.14 807.825,1219 810.067,1217.87 812.309,1216.75 814.551,1215.63 816.793,1214.52 819.035,1213.41 821.277,1212.31 823.519,1211.21 825.761,1210.13 828.003,1209.04 830.245,1207.97 832.487,1206.9 834.729,1205.83 836.971,1204.78 839.213,1203.72 841.455,1202.68 843.697,1201.64 845.939,1200.6 848.181,1199.57 850.423,1198.55 852.665,1197.53 854.907,1196.51 857.149,1195.51 859.391,1194.5 861.632,1193.51 863.874,1192.52 866.116,1191.53 868.358,1190.55 870.6,1189.58 872.842,1188.61 875.084,1187.65 877.326,1186.69 879.568,1185.74 881.81,1184.79 884.052,1183.85 886.294,1182.92 888.536,1181.99 890.778,1181.06 893.02,1180.15 895.262,1179.24 897.504,1178.33 899.746,1177.43 901.988,1176.54 904.23,1175.66 906.472,1174.78 908.714,1173.9 910.956,1173.04 913.198,1172.18 915.439,1171.32 917.681,1170.47 919.923,1169.63 922.165,1168.8 924.407,1167.97 926.649,1167.15 928.891,1166.33 931.133,1165.52 933.375,1164.72 935.617,1163.92 937.859,1163.13 940.101,1162.34 942.343,1161.57 944.585,1160.79 946.827,1160.03 949.069,1159.27 951.311,1158.51 953.553,1157.76 955.795,1157.02 958.037,1156.28 960.279,1155.55 962.521,1154.83 964.763,1154.11 967.005,1153.4 969.246,1152.69 971.488,1151.99 973.73,1151.29 975.972,1150.6 978.214,1149.91 980.456,1149.23 982.698,1148.56 984.94,1147.89 987.182,1147.23 989.424,1146.58 991.666,1145.93 993.908,1145.28 996.15,1144.64 998.392,1144.01 1000.63,1143.38 1002.88,1142.76 1005.12,1142.15 1007.36,1141.54 1009.6,1140.94 1011.84,1140.34 1014.09,1139.75 1016.33,1139.16 1018.57,1138.59 1020.81,1138.01 1023.05,1137.45 1025.3,1136.88 1027.54,1136.33 1029.78,1135.78 1032.02,1135.24 1034.26,1134.7 1036.51,1134.17 1038.75,1133.65 1040.99,1133.13 1043.23,1132.61 1045.47,1132.11 1047.72,1131.6 1049.96,1131.11 1052.2,1130.62 1054.44,1130.13 1056.68,1129.65 1058.92,1129.18 1061.17,1128.71 1063.41,1128.25 1065.65,1127.79 1067.89,1127.34 1070.13,1126.9 1072.38,1126.45 1074.62,1126.02 1076.86,1125.59 1079.1,1125.17 1081.34,1124.75 1083.59,1124.33 1085.83,1123.92 1088.07,1123.52 1090.31,1123.12 1092.55,1122.73 1094.8,1122.34 1097.04,1121.96 1099.28,1121.58 1101.52,1121.21 1103.76,1120.85 1106.01,1120.49 1108.25,1120.13 1110.49,1119.78 1112.73,1119.43 1114.97,1119.09 1117.22,1118.76 1119.46,1118.43 1121.7,1118.1 1123.94,1117.78 1126.18,1117.47 1128.43,1117.16 1130.67,1116.86 1132.91,1116.56 1135.15,1116.26 1137.39,1115.97 1139.64,1115.69 1141.88,1115.41 1144.12,1115.13 1146.36,1114.86 1148.6,1114.6 1150.85,1114.34 1153.09,1114.08 1155.33,1113.83 1157.57,1113.58 1159.81,1113.34 1162.05,1113.1 1164.3,1112.87 1166.54,1112.64 1168.78,1112.41 1171.02,1112.19 1173.26,1111.98 1175.51,1111.76 1177.75,1111.56 1179.99,1111.35 1182.23,1111.15 1184.47,1110.96 1186.72,1110.77 1188.96,1110.58 1191.2,1110.4 1193.44,1110.22 1195.68,1110.04 1197.93,1109.87 1200.17,1109.7 1202.41,1109.54 1204.65,1109.38 1206.89,1109.22 1209.14,1109.07 1211.38,1108.92 1213.62,1108.77 1215.86,1108.63 1218.1,1108.49 1220.35,1108.36 1222.59,1108.23 1224.83,1108.1 1227.07,1107.97 1229.31,1107.85 1231.56,1107.73 1233.8,1107.62 1236.04,1107.51 1238.28,1107.4 1240.52,1107.29 1242.77,1107.19 1245.01,1107.09 1247.25,1107 1249.49,1106.9 1251.73,1106.81 1253.98,1106.72 1256.22,1106.64 1258.46,1106.55 1260.7,1106.47 1262.94,1106.4 1265.19,1106.32 1267.43,1106.25 1269.67,1106.17 1271.91,1106.11 1274.15,1106.04 1276.39,1105.97 1278.64,1105.91 1280.88,1105.85 1283.12,1105.79 1285.36,1105.74 1287.6,1105.68 1289.85,1105.63 1292.09,1105.58 1294.33,1105.53 1296.57,1105.48 1298.81,1105.43 1301.06,1105.39 1303.3,1105.34 1305.54,1105.3 1307.78,1105.26 1310.02,1105.22 1312.27,1105.18 1314.51,1105.14 1316.75,1105.1 1318.99,1105.07 "/>
+<path clip-path="url(#clip030)" d="M1559.76 1723.83 L2752.76 1723.83 L2752.76 1087.05 L1559.76 1087.05  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip215">
+  <clipPath id="clip035">
     <rect x="1559" y="1087" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,1723.83 1593.53,1087.05 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,1723.83 1873.77,1087.05 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,1723.83 2154.02,1087.05 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,1723.83 2434.26,1087.05 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,1723.83 2714.51,1087.05 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1723.83 2752.76,1723.83 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,1723.83 1593.53,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,1723.83 1873.77,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,1723.83 2154.02,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,1723.83 2434.26,1704.94 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,1723.83 2714.51,1704.94 "/>
-<path clip-path="url(#clip210)" d="M1593.53 1757.95 Q1589.92 1757.95 1588.09 1761.52 Q1586.28 1765.06 1586.28 1772.19 Q1586.28 1779.3 1588.09 1782.86 Q1589.92 1786.4 1593.53 1786.4 Q1597.16 1786.4 1598.97 1782.86 Q1600.8 1779.3 1600.8 1772.19 Q1600.8 1765.06 1598.97 1761.52 Q1597.16 1757.95 1593.53 1757.95 M1593.53 1754.25 Q1599.34 1754.25 1602.39 1758.86 Q1605.47 1763.44 1605.47 1772.19 Q1605.47 1780.92 1602.39 1785.52 Q1599.34 1790.11 1593.53 1790.11 Q1587.72 1790.11 1584.64 1785.52 Q1581.58 1780.92 1581.58 1772.19 Q1581.58 1763.44 1584.64 1758.86 Q1587.72 1754.25 1593.53 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1833.39 1754.87 L1851.75 1754.87 L1851.75 1758.81 L1837.67 1758.81 L1837.67 1767.28 Q1838.69 1766.93 1839.71 1766.77 Q1840.73 1766.59 1841.75 1766.59 Q1847.54 1766.59 1850.91 1769.76 Q1854.29 1772.93 1854.29 1778.35 Q1854.29 1783.93 1850.82 1787.03 Q1847.35 1790.11 1841.03 1790.11 Q1838.85 1790.11 1836.59 1789.74 Q1834.34 1789.37 1831.93 1788.62 L1831.93 1783.93 Q1834.02 1785.06 1836.24 1785.62 Q1838.46 1786.17 1840.94 1786.17 Q1844.94 1786.17 1847.28 1784.06 Q1849.62 1781.96 1849.62 1778.35 Q1849.62 1774.74 1847.28 1772.63 Q1844.94 1770.52 1840.94 1770.52 Q1839.06 1770.52 1837.19 1770.94 Q1835.34 1771.36 1833.39 1772.24 L1833.39 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1873.51 1757.95 Q1869.9 1757.95 1868.07 1761.52 Q1866.26 1765.06 1866.26 1772.19 Q1866.26 1779.3 1868.07 1782.86 Q1869.9 1786.4 1873.51 1786.4 Q1877.14 1786.4 1878.95 1782.86 Q1880.78 1779.3 1880.78 1772.19 Q1880.78 1765.06 1878.95 1761.52 Q1877.14 1757.95 1873.51 1757.95 M1873.51 1754.25 Q1879.32 1754.25 1882.37 1758.86 Q1885.45 1763.44 1885.45 1772.19 Q1885.45 1780.92 1882.37 1785.52 Q1879.32 1790.11 1873.51 1790.11 Q1867.7 1790.11 1864.62 1785.52 Q1861.56 1780.92 1861.56 1772.19 Q1861.56 1763.44 1864.62 1758.86 Q1867.7 1754.25 1873.51 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1903.67 1757.95 Q1900.06 1757.95 1898.23 1761.52 Q1896.42 1765.06 1896.42 1772.19 Q1896.42 1779.3 1898.23 1782.86 Q1900.06 1786.4 1903.67 1786.4 Q1907.3 1786.4 1909.11 1782.86 Q1910.94 1779.3 1910.94 1772.19 Q1910.94 1765.06 1909.11 1761.52 Q1907.3 1757.95 1903.67 1757.95 M1903.67 1754.25 Q1909.48 1754.25 1912.53 1758.86 Q1915.61 1763.44 1915.61 1772.19 Q1915.61 1780.92 1912.53 1785.52 Q1909.48 1790.11 1903.67 1790.11 Q1897.86 1790.11 1894.78 1785.52 Q1891.72 1780.92 1891.72 1772.19 Q1891.72 1763.44 1894.78 1758.86 Q1897.86 1754.25 1903.67 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2098.54 1785.5 L2106.18 1785.5 L2106.18 1759.13 L2097.87 1760.8 L2097.87 1756.54 L2106.14 1754.87 L2110.81 1754.87 L2110.81 1785.5 L2118.45 1785.5 L2118.45 1789.43 L2098.54 1789.43 L2098.54 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2137.9 1757.95 Q2134.28 1757.95 2132.46 1761.52 Q2130.65 1765.06 2130.65 1772.19 Q2130.65 1779.3 2132.46 1782.86 Q2134.28 1786.4 2137.9 1786.4 Q2141.53 1786.4 2143.34 1782.86 Q2145.16 1779.3 2145.16 1772.19 Q2145.16 1765.06 2143.34 1761.52 Q2141.53 1757.95 2137.9 1757.95 M2137.9 1754.25 Q2143.71 1754.25 2146.76 1758.86 Q2149.84 1763.44 2149.84 1772.19 Q2149.84 1780.92 2146.76 1785.52 Q2143.71 1790.11 2137.9 1790.11 Q2132.09 1790.11 2129.01 1785.52 Q2125.95 1780.92 2125.95 1772.19 Q2125.95 1763.44 2129.01 1758.86 Q2132.09 1754.25 2137.9 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2168.06 1757.95 Q2164.45 1757.95 2162.62 1761.52 Q2160.81 1765.06 2160.81 1772.19 Q2160.81 1779.3 2162.62 1782.86 Q2164.45 1786.4 2168.06 1786.4 Q2171.69 1786.4 2173.5 1782.86 Q2175.33 1779.3 2175.33 1772.19 Q2175.33 1765.06 2173.5 1761.52 Q2171.69 1757.95 2168.06 1757.95 M2168.06 1754.25 Q2173.87 1754.25 2176.92 1758.86 Q2180 1763.44 2180 1772.19 Q2180 1780.92 2176.92 1785.52 Q2173.87 1790.11 2168.06 1790.11 Q2162.25 1790.11 2159.17 1785.52 Q2156.11 1780.92 2156.11 1772.19 Q2156.11 1763.44 2159.17 1758.86 Q2162.25 1754.25 2168.06 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2198.22 1757.95 Q2194.61 1757.95 2192.78 1761.52 Q2190.97 1765.06 2190.97 1772.19 Q2190.97 1779.3 2192.78 1782.86 Q2194.61 1786.4 2198.22 1786.4 Q2201.85 1786.4 2203.66 1782.86 Q2205.49 1779.3 2205.49 1772.19 Q2205.49 1765.06 2203.66 1761.52 Q2201.85 1757.95 2198.22 1757.95 M2198.22 1754.25 Q2204.03 1754.25 2207.09 1758.86 Q2210.16 1763.44 2210.16 1772.19 Q2210.16 1780.92 2207.09 1785.52 Q2204.03 1790.11 2198.22 1790.11 Q2192.41 1790.11 2189.33 1785.52 Q2186.28 1780.92 2186.28 1772.19 Q2186.28 1763.44 2189.33 1758.86 Q2192.41 1754.25 2198.22 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2378.79 1785.5 L2386.43 1785.5 L2386.43 1759.13 L2378.12 1760.8 L2378.12 1756.54 L2386.38 1754.87 L2391.06 1754.87 L2391.06 1785.5 L2398.7 1785.5 L2398.7 1789.43 L2378.79 1789.43 L2378.79 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2408.19 1754.87 L2426.54 1754.87 L2426.54 1758.81 L2412.47 1758.81 L2412.47 1767.28 Q2413.49 1766.93 2414.51 1766.77 Q2415.52 1766.59 2416.54 1766.59 Q2422.33 1766.59 2425.71 1769.76 Q2429.09 1772.93 2429.09 1778.35 Q2429.09 1783.93 2425.62 1787.03 Q2422.15 1790.11 2415.83 1790.11 Q2413.65 1790.11 2411.38 1789.74 Q2409.14 1789.37 2406.73 1788.62 L2406.73 1783.93 Q2408.81 1785.06 2411.03 1785.62 Q2413.26 1786.17 2415.73 1786.17 Q2419.74 1786.17 2422.08 1784.06 Q2424.41 1781.96 2424.41 1778.35 Q2424.41 1774.74 2422.08 1772.63 Q2419.74 1770.52 2415.73 1770.52 Q2413.86 1770.52 2411.98 1770.94 Q2410.13 1771.36 2408.19 1772.24 L2408.19 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2448.3 1757.95 Q2444.69 1757.95 2442.86 1761.52 Q2441.06 1765.06 2441.06 1772.19 Q2441.06 1779.3 2442.86 1782.86 Q2444.69 1786.4 2448.3 1786.4 Q2451.94 1786.4 2453.74 1782.86 Q2455.57 1779.3 2455.57 1772.19 Q2455.57 1765.06 2453.74 1761.52 Q2451.94 1757.95 2448.3 1757.95 M2448.3 1754.25 Q2454.11 1754.25 2457.17 1758.86 Q2460.25 1763.44 2460.25 1772.19 Q2460.25 1780.92 2457.17 1785.52 Q2454.11 1790.11 2448.3 1790.11 Q2442.49 1790.11 2439.41 1785.52 Q2436.36 1780.92 2436.36 1772.19 Q2436.36 1763.44 2439.41 1758.86 Q2442.49 1754.25 2448.3 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2478.46 1757.95 Q2474.85 1757.95 2473.02 1761.52 Q2471.22 1765.06 2471.22 1772.19 Q2471.22 1779.3 2473.02 1782.86 Q2474.85 1786.4 2478.46 1786.4 Q2482.1 1786.4 2483.9 1782.86 Q2485.73 1779.3 2485.73 1772.19 Q2485.73 1765.06 2483.9 1761.52 Q2482.1 1757.95 2478.46 1757.95 M2478.46 1754.25 Q2484.27 1754.25 2487.33 1758.86 Q2490.41 1763.44 2490.41 1772.19 Q2490.41 1780.92 2487.33 1785.52 Q2484.27 1790.11 2478.46 1790.11 Q2472.65 1790.11 2469.58 1785.52 Q2466.52 1780.92 2466.52 1772.19 Q2466.52 1763.44 2469.58 1758.86 Q2472.65 1754.25 2478.46 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2663.12 1785.5 L2679.44 1785.5 L2679.44 1789.43 L2657.49 1789.43 L2657.49 1785.5 Q2660.16 1782.74 2664.74 1778.12 Q2669.35 1773.46 2670.53 1772.12 Q2672.77 1769.6 2673.65 1767.86 Q2674.55 1766.1 2674.55 1764.41 Q2674.55 1761.66 2672.61 1759.92 Q2670.69 1758.18 2667.59 1758.18 Q2665.39 1758.18 2662.93 1758.95 Q2660.5 1759.71 2657.73 1761.26 L2657.73 1756.54 Q2660.55 1755.41 2663 1754.83 Q2665.46 1754.25 2667.49 1754.25 Q2672.86 1754.25 2676.06 1756.93 Q2679.25 1759.62 2679.25 1764.11 Q2679.25 1766.24 2678.44 1768.16 Q2677.66 1770.06 2675.55 1772.65 Q2674.97 1773.32 2671.87 1776.54 Q2668.77 1779.74 2663.12 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2699.25 1757.95 Q2695.64 1757.95 2693.81 1761.52 Q2692.01 1765.06 2692.01 1772.19 Q2692.01 1779.3 2693.81 1782.86 Q2695.64 1786.4 2699.25 1786.4 Q2702.89 1786.4 2704.69 1782.86 Q2706.52 1779.3 2706.52 1772.19 Q2706.52 1765.06 2704.69 1761.52 Q2702.89 1757.95 2699.25 1757.95 M2699.25 1754.25 Q2705.06 1754.25 2708.12 1758.86 Q2711.2 1763.44 2711.2 1772.19 Q2711.2 1780.92 2708.12 1785.52 Q2705.06 1790.11 2699.25 1790.11 Q2693.44 1790.11 2690.36 1785.52 Q2687.31 1780.92 2687.31 1772.19 Q2687.31 1763.44 2690.36 1758.86 Q2693.44 1754.25 2699.25 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2729.42 1757.95 Q2725.8 1757.95 2723.98 1761.52 Q2722.17 1765.06 2722.17 1772.19 Q2722.17 1779.3 2723.98 1782.86 Q2725.8 1786.4 2729.42 1786.4 Q2733.05 1786.4 2734.86 1782.86 Q2736.68 1779.3 2736.68 1772.19 Q2736.68 1765.06 2734.86 1761.52 Q2733.05 1757.95 2729.42 1757.95 M2729.42 1754.25 Q2735.23 1754.25 2738.28 1758.86 Q2741.36 1763.44 2741.36 1772.19 Q2741.36 1780.92 2738.28 1785.52 Q2735.23 1790.11 2729.42 1790.11 Q2723.61 1790.11 2720.53 1785.52 Q2717.47 1780.92 2717.47 1772.19 Q2717.47 1763.44 2720.53 1758.86 Q2723.61 1754.25 2729.42 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2759.58 1757.95 Q2755.97 1757.95 2754.14 1761.52 Q2752.33 1765.06 2752.33 1772.19 Q2752.33 1779.3 2754.14 1782.86 Q2755.97 1786.4 2759.58 1786.4 Q2763.21 1786.4 2765.02 1782.86 Q2766.85 1779.3 2766.85 1772.19 Q2766.85 1765.06 2765.02 1761.52 Q2763.21 1757.95 2759.58 1757.95 M2759.58 1754.25 Q2765.39 1754.25 2768.44 1758.86 Q2771.52 1763.44 2771.52 1772.19 Q2771.52 1780.92 2768.44 1785.52 Q2765.39 1790.11 2759.58 1790.11 Q2753.77 1790.11 2750.69 1785.52 Q2747.63 1780.92 2747.63 1772.19 Q2747.63 1763.44 2750.69 1758.86 Q2753.77 1754.25 2759.58 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1672.06 2752.76,1672.06 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1567.36 2752.76,1567.36 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1462.66 2752.76,1462.66 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1357.96 2752.76,1357.96 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1253.26 2752.76,1253.26 "/>
-<polyline clip-path="url(#clip215)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1148.56 2752.76,1148.56 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1723.83 1559.76,1087.05 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1672.06 1578.66,1672.06 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1567.36 1578.66,1567.36 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1462.66 1578.66,1462.66 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1357.96 1578.66,1357.96 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1253.26 1578.66,1253.26 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1148.56 1578.66,1148.56 "/>
-<path clip-path="url(#clip210)" d="M1457.05 1672.51 L1486.72 1672.51 L1486.72 1676.44 L1457.05 1676.44 L1457.05 1672.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1496.86 1654.78 L1515.22 1654.78 L1515.22 1658.71 L1501.14 1658.71 L1501.14 1667.18 Q1502.16 1666.84 1503.18 1666.68 Q1504.2 1666.49 1505.22 1666.49 Q1511.01 1666.49 1514.39 1669.66 Q1517.76 1672.83 1517.76 1678.25 Q1517.76 1683.83 1514.29 1686.93 Q1510.82 1690.01 1504.5 1690.01 Q1502.33 1690.01 1500.06 1689.64 Q1497.81 1689.27 1495.4 1688.53 L1495.4 1683.83 Q1497.49 1684.96 1499.71 1685.52 Q1501.93 1686.07 1504.41 1686.07 Q1508.41 1686.07 1510.75 1683.97 Q1513.09 1681.86 1513.09 1678.25 Q1513.09 1674.64 1510.75 1672.53 Q1508.41 1670.43 1504.41 1670.43 Q1502.53 1670.43 1500.66 1670.84 Q1498.81 1671.26 1496.86 1672.14 L1496.86 1654.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1455.57 1567.81 L1485.24 1567.81 L1485.24 1571.74 L1455.57 1571.74 L1455.57 1567.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1508.18 1554.15 L1496.38 1572.6 L1508.18 1572.6 L1508.18 1554.15 M1506.95 1550.08 L1512.83 1550.08 L1512.83 1572.6 L1517.76 1572.6 L1517.76 1576.49 L1512.83 1576.49 L1512.83 1584.64 L1508.18 1584.64 L1508.18 1576.49 L1492.58 1576.49 L1492.58 1571.98 L1506.95 1550.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1456.7 1463.11 L1486.38 1463.11 L1486.38 1467.04 L1456.7 1467.04 L1456.7 1463.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1510.64 1461.3 Q1513.99 1462.02 1515.87 1464.29 Q1517.76 1466.56 1517.76 1469.89 Q1517.76 1475.01 1514.25 1477.81 Q1510.73 1480.61 1504.25 1480.61 Q1502.07 1480.61 1499.76 1480.17 Q1497.46 1479.75 1495.01 1478.9 L1495.01 1474.38 Q1496.95 1475.52 1499.27 1476.09 Q1501.58 1476.67 1504.11 1476.67 Q1508.51 1476.67 1510.8 1474.94 Q1513.11 1473.2 1513.11 1469.89 Q1513.11 1466.84 1510.96 1465.12 Q1508.83 1463.39 1505.01 1463.39 L1500.98 1463.39 L1500.98 1459.54 L1505.2 1459.54 Q1508.64 1459.54 1510.47 1458.18 Q1512.3 1456.79 1512.3 1454.2 Q1512.3 1451.53 1510.4 1450.12 Q1508.53 1448.69 1505.01 1448.69 Q1503.09 1448.69 1500.89 1449.1 Q1498.69 1449.52 1496.05 1450.4 L1496.05 1446.23 Q1498.71 1445.49 1501.03 1445.12 Q1503.37 1444.75 1505.43 1444.75 Q1510.75 1444.75 1513.85 1447.18 Q1516.95 1449.59 1516.95 1453.71 Q1516.95 1456.58 1515.31 1458.57 Q1513.67 1460.54 1510.64 1461.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1457.65 1358.41 L1487.33 1358.41 L1487.33 1362.34 L1457.65 1362.34 L1457.65 1358.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1501.45 1371.3 L1517.76 1371.3 L1517.76 1375.24 L1495.82 1375.24 L1495.82 1371.3 Q1498.48 1368.55 1503.07 1363.92 Q1507.67 1359.27 1508.85 1357.92 Q1511.1 1355.4 1511.98 1353.66 Q1512.88 1351.9 1512.88 1350.21 Q1512.88 1347.46 1510.94 1345.72 Q1509.01 1343.99 1505.91 1343.99 Q1503.71 1343.99 1501.26 1344.75 Q1498.83 1345.52 1496.05 1347.07 L1496.05 1342.34 Q1498.88 1341.21 1501.33 1340.63 Q1503.78 1340.05 1505.82 1340.05 Q1511.19 1340.05 1514.39 1342.74 Q1517.58 1345.42 1517.58 1349.91 Q1517.58 1352.04 1516.77 1353.96 Q1515.98 1355.86 1513.88 1358.46 Q1513.3 1359.13 1510.2 1362.34 Q1507.09 1365.54 1501.45 1371.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1457.28 1253.71 L1486.95 1253.71 L1486.95 1257.64 L1457.28 1257.64 L1457.28 1253.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1497.86 1266.6 L1505.5 1266.6 L1505.5 1240.24 L1497.19 1241.9 L1497.19 1237.64 L1505.45 1235.98 L1510.13 1235.98 L1510.13 1266.6 L1517.76 1266.6 L1517.76 1270.54 L1497.86 1270.54 L1497.86 1266.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 1134.36 Q1502.21 1134.36 1500.38 1137.92 Q1498.58 1141.46 1498.58 1148.59 Q1498.58 1155.7 1500.38 1159.26 Q1502.21 1162.81 1505.82 1162.81 Q1509.45 1162.81 1511.26 1159.26 Q1513.09 1155.7 1513.09 1148.59 Q1513.09 1141.46 1511.26 1137.92 Q1509.45 1134.36 1505.82 1134.36 M1505.82 1130.65 Q1511.63 1130.65 1514.69 1135.26 Q1517.76 1139.84 1517.76 1148.59 Q1517.76 1157.32 1514.69 1161.93 Q1511.63 1166.51 1505.82 1166.51 Q1500.01 1166.51 1496.93 1161.93 Q1493.88 1157.32 1493.88 1148.59 Q1493.88 1139.84 1496.93 1135.26 Q1500.01 1130.65 1505.82 1130.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1820.06 945.429 L1854.81 945.429 L1854.81 952.316 L1828.24 952.316 L1828.24 970.14 L1852.22 970.14 L1852.22 977.026 L1828.24 977.026 L1828.24 1005.91 L1820.06 1005.91 L1820.06 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1867.45 942.877 L1874.9 942.877 L1874.9 1005.91 L1867.45 1005.91 L1867.45 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1890.5 960.539 L1897.95 960.539 L1897.95 1005.91 L1890.5 1005.91 L1890.5 960.539 M1890.5 942.877 L1897.95 942.877 L1897.95 952.316 L1890.5 952.316 L1890.5 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1943.41 982.698 Q1943.41 974.596 1940.04 970.14 Q1936.72 965.684 1930.69 965.684 Q1924.69 965.684 1921.33 970.14 Q1918.01 974.596 1918.01 982.698 Q1918.01 990.759 1921.33 995.215 Q1924.69 999.671 1930.69 999.671 Q1936.72 999.671 1940.04 995.215 Q1943.41 990.759 1943.41 982.698 M1950.86 1000.28 Q1950.86 1011.86 1945.71 1017.49 Q1940.57 1023.17 1929.96 1023.17 Q1926.03 1023.17 1922.54 1022.56 Q1919.06 1021.99 1915.78 1020.78 L1915.78 1013.53 Q1919.06 1015.31 1922.26 1016.16 Q1925.46 1017.01 1928.78 1017.01 Q1936.11 1017.01 1939.76 1013.16 Q1943.41 1009.35 1943.41 1001.62 L1943.41 997.929 Q1941.1 1001.94 1937.49 1003.92 Q1933.89 1005.91 1928.86 1005.91 Q1920.52 1005.91 1915.41 999.549 Q1910.31 993.189 1910.31 982.698 Q1910.31 972.165 1915.41 965.805 Q1920.52 959.445 1928.86 959.445 Q1933.89 959.445 1937.49 961.43 Q1941.1 963.415 1943.41 967.426 L1943.41 960.539 L1950.86 960.539 L1950.86 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2003.93 978.525 L2003.93 1005.91 L1996.47 1005.91 L1996.47 978.768 Q1996.47 972.327 1993.96 969.127 Q1991.45 965.927 1986.43 965.927 Q1980.39 965.927 1976.91 969.775 Q1973.42 973.624 1973.42 980.267 L1973.42 1005.91 L1965.93 1005.91 L1965.93 942.877 L1973.42 942.877 L1973.42 967.588 Q1976.1 963.496 1979.7 961.471 Q1983.35 959.445 1988.09 959.445 Q1995.91 959.445 1999.92 964.307 Q2003.93 969.127 2003.93 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2026.17 947.657 L2026.17 960.539 L2041.52 960.539 L2041.52 966.332 L2026.17 966.332 L2026.17 990.961 Q2026.17 996.511 2027.66 998.091 Q2029.2 999.671 2033.86 999.671 L2041.52 999.671 L2041.52 1005.91 L2033.86 1005.91 Q2025.23 1005.91 2021.95 1002.71 Q2018.67 999.468 2018.67 990.961 L2018.67 966.332 L2013.2 966.332 L2013.2 960.539 L2018.67 960.539 L2018.67 947.657 L2026.17 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2086.2 952.154 L2086.2 974.879 L2096.49 974.879 Q2102.2 974.879 2105.32 971.922 Q2108.44 968.965 2108.44 963.496 Q2108.44 958.068 2105.32 955.111 Q2102.2 952.154 2096.49 952.154 L2086.2 952.154 M2078.02 945.429 L2096.49 945.429 Q2106.66 945.429 2111.84 950.047 Q2117.07 954.625 2117.07 963.496 Q2117.07 972.449 2111.84 977.026 Q2106.66 981.604 2096.49 981.604 L2086.2 981.604 L2086.2 1005.91 L2078.02 1005.91 L2078.02 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2144.61 983.103 Q2135.58 983.103 2132.1 985.169 Q2128.61 987.235 2128.61 992.217 Q2128.61 996.187 2131.21 998.537 Q2133.84 1000.85 2138.34 1000.85 Q2144.53 1000.85 2148.26 996.471 Q2152.03 992.055 2152.03 984.764 L2152.03 983.103 L2144.61 983.103 M2159.48 980.024 L2159.48 1005.91 L2152.03 1005.91 L2152.03 999.023 Q2149.48 1003.15 2145.67 1005.14 Q2141.86 1007.08 2136.35 1007.08 Q2129.38 1007.08 2125.25 1003.2 Q2121.16 999.266 2121.16 992.703 Q2121.16 985.047 2126.26 981.158 Q2131.41 977.269 2141.58 977.269 L2152.03 977.269 L2152.03 976.54 Q2152.03 971.396 2148.62 968.601 Q2145.26 965.765 2139.15 965.765 Q2135.26 965.765 2131.57 966.697 Q2127.88 967.628 2124.48 969.492 L2124.48 962.605 Q2128.57 961.025 2132.42 960.256 Q2136.27 959.445 2139.92 959.445 Q2149.76 959.445 2154.62 964.55 Q2159.48 969.654 2159.48 980.024 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2182.21 947.657 L2182.21 960.539 L2197.56 960.539 L2197.56 966.332 L2182.21 966.332 L2182.21 990.961 Q2182.21 996.511 2183.71 998.091 Q2185.24 999.671 2189.9 999.671 L2197.56 999.671 L2197.56 1005.91 L2189.9 1005.91 Q2181.27 1005.91 2177.99 1002.71 Q2174.71 999.468 2174.71 990.961 L2174.71 966.332 L2169.24 966.332 L2169.24 960.539 L2174.71 960.539 L2174.71 947.657 L2182.21 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2245.08 978.525 L2245.08 1005.91 L2237.62 1005.91 L2237.62 978.768 Q2237.62 972.327 2235.11 969.127 Q2232.6 965.927 2227.58 965.927 Q2221.54 965.927 2218.06 969.775 Q2214.57 973.624 2214.57 980.267 L2214.57 1005.91 L2207.08 1005.91 L2207.08 942.877 L2214.57 942.877 L2214.57 967.588 Q2217.25 963.496 2220.85 961.471 Q2224.5 959.445 2229.24 959.445 Q2237.06 959.445 2241.07 964.307 Q2245.08 969.127 2245.08 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2304.22 942.958 Q2298.79 952.275 2296.16 961.39 Q2293.53 970.504 2293.53 979.862 Q2293.53 989.22 2296.16 998.415 Q2298.83 1007.57 2304.22 1016.85 L2297.74 1016.85 Q2291.66 1007.33 2288.62 998.132 Q2285.63 988.936 2285.63 979.862 Q2285.63 970.829 2288.62 961.673 Q2291.62 952.518 2297.74 942.958 L2304.22 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2348.54 967.426 L2348.54 942.877 L2355.99 942.877 L2355.99 1005.91 L2348.54 1005.91 L2348.54 999.104 Q2346.19 1003.15 2342.58 1005.14 Q2339.02 1007.08 2333.99 1007.08 Q2325.77 1007.08 2320.59 1000.52 Q2315.44 993.959 2315.44 983.265 Q2315.44 972.57 2320.59 966.008 Q2325.77 959.445 2333.99 959.445 Q2339.02 959.445 2342.58 961.43 Q2346.19 963.375 2348.54 967.426 M2323.14 983.265 Q2323.14 991.488 2326.5 996.187 Q2329.9 1000.85 2335.82 1000.85 Q2341.73 1000.85 2345.13 996.187 Q2348.54 991.488 2348.54 983.265 Q2348.54 975.041 2345.13 970.383 Q2341.73 965.684 2335.82 965.684 Q2329.9 965.684 2326.5 970.383 Q2323.14 975.041 2323.14 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2410.15 981.361 L2410.15 985.007 L2375.88 985.007 Q2376.37 992.703 2380.5 996.754 Q2384.67 1000.76 2392.08 1000.76 Q2396.38 1000.76 2400.39 999.711 Q2404.44 998.658 2408.41 996.552 L2408.41 1003.6 Q2404.4 1005.3 2400.19 1006.19 Q2395.97 1007.08 2391.64 1007.08 Q2380.78 1007.08 2374.42 1000.76 Q2368.1 994.445 2368.1 983.67 Q2368.1 972.53 2374.1 966.008 Q2380.13 959.445 2390.34 959.445 Q2399.5 959.445 2404.8 965.36 Q2410.15 971.234 2410.15 981.361 M2402.7 979.173 Q2402.62 973.057 2399.25 969.411 Q2395.93 965.765 2390.42 965.765 Q2384.18 965.765 2380.42 969.289 Q2376.69 972.813 2376.12 979.214 L2402.7 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2452.24 982.698 Q2452.24 974.596 2448.88 970.14 Q2445.56 965.684 2439.52 965.684 Q2433.52 965.684 2430.16 970.14 Q2426.84 974.596 2426.84 982.698 Q2426.84 990.759 2430.16 995.215 Q2433.52 999.671 2439.52 999.671 Q2445.56 999.671 2448.88 995.215 Q2452.24 990.759 2452.24 982.698 M2459.69 1000.28 Q2459.69 1011.86 2454.55 1017.49 Q2449.4 1023.17 2438.79 1023.17 Q2434.86 1023.17 2431.38 1022.56 Q2427.89 1021.99 2424.61 1020.78 L2424.61 1013.53 Q2427.89 1015.31 2431.09 1016.16 Q2434.29 1017.01 2437.62 1017.01 Q2444.95 1017.01 2448.59 1013.16 Q2452.24 1009.35 2452.24 1001.62 L2452.24 997.929 Q2449.93 1001.94 2446.33 1003.92 Q2442.72 1005.91 2437.7 1005.91 Q2429.35 1005.91 2424.25 999.549 Q2419.14 993.189 2419.14 982.698 Q2419.14 972.165 2424.25 965.805 Q2429.35 959.445 2437.7 959.445 Q2442.72 959.445 2446.33 961.43 Q2449.93 963.415 2452.24 967.426 L2452.24 960.539 L2459.69 960.539 L2459.69 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2473.87 942.958 L2480.35 942.958 Q2486.43 952.518 2489.43 961.673 Q2492.47 970.829 2492.47 979.862 Q2492.47 988.936 2489.43 998.132 Q2486.43 1007.33 2480.35 1016.85 L2473.87 1016.85 Q2479.26 1007.57 2481.89 998.415 Q2484.57 989.22 2484.57 979.862 Q2484.57 970.504 2481.89 961.39 Q2479.26 952.275 2473.87 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip215)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,1253.26 1595.77,1253.23 1598.01,1253.16 1600.25,1253.07 1602.5,1252.93 1604.74,1252.75 1606.98,1252.53 1609.22,1252.26 1611.46,1251.93 1613.71,1251.55 1615.95,1251.12 1618.19,1250.61 1620.43,1250.04 1622.67,1249.4 1624.92,1248.67 1627.16,1247.86 1629.4,1246.97 1631.64,1245.97 1633.88,1244.88 1636.13,1243.67 1638.37,1242.36 1640.61,1240.92 1642.85,1239.36 1645.09,1237.67 1647.34,1235.84 1649.58,1233.86 1651.82,1231.74 1654.06,1229.47 1656.3,1227.05 1658.55,1224.46 1660.79,1221.72 1663.03,1218.82 1665.27,1215.77 1667.51,1212.56 1669.76,1209.2 1672,1205.69 1674.24,1202.05 1676.48,1198.27 1678.72,1194.39 1680.97,1190.39 1683.21,1186.31 1685.45,1182.16 1687.69,1177.95 1689.93,1173.7 1692.17,1169.44 1694.42,1165.17 1696.66,1160.93 1698.9,1156.74 1701.14,1152.61 1703.38,1148.56 1705.63,1144.62 1707.87,1140.79 1710.11,1137.11 1712.35,1133.59 1714.59,1130.23 1716.84,1127.05 1719.08,1124.07 1721.32,1121.29 1723.56,1118.72 1725.8,1116.37 1728.05,1114.24 1730.29,1112.33 1732.53,1110.65 1734.77,1109.19 1737.01,1107.96 1739.26,1106.95 1741.5,1106.16 1743.74,1105.59 1745.98,1105.23 1748.22,1105.07 1750.47,1105.11 1752.71,1105.35 1754.95,1105.77 1757.19,1106.37 1759.43,1107.15 1761.68,1108.09 1763.92,1109.19 1766.16,1110.44 1768.4,1111.84 1770.64,1113.37 1772.89,1115.03 1775.13,1116.82 1777.37,1118.71 1779.61,1120.72 1781.85,1122.82 1784.1,1125.02 1786.34,1127.3 1788.58,1129.66 1790.82,1132.09 1793.06,1134.59 1795.31,1137.15 1797.55,1139.76 1799.79,1142.42 1802.03,1145.11 1804.27,1147.83 1806.51,1150.58 1808.76,1153.35 1811,1156.13 1813.24,1158.91 1815.48,1161.69 1817.72,1164.46 1819.97,1167.21 1822.21,1169.94 1824.45,1172.63 1826.69,1175.28 1828.93,1177.87 1831.18,1180.41 1833.42,1182.89 1835.66,1185.28 1837.9,1187.59 1840.14,1189.8 1842.39,1191.91 1844.63,1193.9 1846.87,1195.76 1849.11,1197.49 1851.35,1199.07 1853.6,1200.49 1855.84,1201.74 1858.08,1202.8 1860.32,1203.68 1862.56,1204.35 1864.81,1204.82 1867.05,1205.05 1869.29,1205.06 1871.53,1204.84 1873.77,1204.36 1876.02,1203.64 1878.26,1202.66 1880.5,1201.43 1882.74,1199.94 1884.98,1198.2 1887.23,1196.22 1889.47,1193.99 1891.71,1191.54 1893.95,1188.87 1896.19,1186 1898.44,1182.95 1900.68,1179.75 1902.92,1176.4 1905.16,1172.95 1907.4,1169.42 1909.64,1165.84 1911.89,1162.23 1914.13,1158.64 1916.37,1155.08 1918.61,1151.6 1920.85,1148.22 1923.1,1144.97 1925.34,1141.87 1927.58,1138.96 1929.82,1136.26 1932.06,1133.77 1934.31,1131.53 1936.55,1129.55 1938.79,1127.83 1941.03,1126.39 1943.27,1125.24 1945.52,1124.37 1947.76,1123.78 1950,1123.49 1952.24,1123.47 1954.48,1123.74 1956.73,1124.27 1958.97,1125.07 1961.21,1126.11 1963.45,1127.4 1965.69,1128.91 1967.94,1130.63 1970.18,1132.55 1972.42,1134.65 1974.66,1136.91 1976.9,1139.33 1979.15,1141.87 1981.39,1144.53 1983.63,1147.29 1985.87,1150.12 1988.11,1153.02 1990.36,1155.95 1992.6,1158.9 1994.84,1161.86 1997.08,1164.8 1999.32,1167.7 2001.57,1170.54 2003.81,1173.3 2006.05,1175.97 2008.29,1178.53 2010.53,1180.94 2012.78,1183.2 2015.02,1185.29 2017.26,1187.18 2019.5,1188.87 2021.74,1190.33 2023.98,1191.55 2026.23,1192.51 2028.47,1193.2 2030.71,1193.62 2032.95,1193.75 2035.19,1193.59 2037.44,1193.13 2039.68,1192.37 2041.92,1191.33 2044.16,1190 2046.4,1188.39 2048.65,1186.52 2050.89,1184.41 2053.13,1182.07 2055.37,1179.53 2057.61,1176.83 2059.86,1173.98 2062.1,1171.02 2064.34,1167.99 2066.58,1164.92 2068.82,1161.85 2071.07,1158.82 2073.31,1155.85 2075.55,1152.99 2077.79,1150.28 2080.03,1147.73 2082.28,1145.38 2084.52,1143.26 2086.76,1141.39 2089,1139.78 2091.24,1138.45 2093.49,1137.42 2095.73,1136.69 2097.97,1136.26 2100.21,1136.14 2102.45,1136.32 2104.7,1136.8 2106.94,1137.57 2109.18,1138.61 2111.42,1139.91 2113.66,1141.46 2115.91,1143.23 2118.15,1145.21 2120.39,1147.38 2122.63,1149.7 2124.87,1152.16 2127.11,1154.73 2129.36,1157.39 2131.6,1160.1 2133.84,1162.84 2136.08,1165.59 2138.32,1168.3 2140.57,1170.97 2142.81,1173.55 2145.05,1176.03 2147.29,1178.37 2149.53,1180.55 2151.78,1182.55 2154.02,1184.34 2156.26,1185.91 2158.5,1187.23 2160.74,1188.29 2162.99,1189.08 2165.23,1189.58 2167.47,1189.78 2169.71,1189.69 2171.95,1189.31 2174.2,1188.63 2176.44,1187.67 2178.68,1186.44 2180.92,1184.95 2183.16,1183.23 2185.41,1181.29 2187.65,1179.18 2189.89,1176.92 2192.13,1174.53 2194.37,1172.07 2196.62,1169.56 2198.86,1167.04 2201.1,1164.56 2203.34,1162.14 2205.58,1159.82 2207.83,1157.65 2210.07,1155.65 2212.31,1153.84 2214.55,1152.26 2216.79,1150.93 2219.04,1149.86 2221.28,1149.06 2223.52,1148.56 2225.76,1148.34 2228,1148.42 2230.25,1148.79 2232.49,1149.44 2234.73,1150.36 2236.97,1151.54 2239.21,1152.96 2241.45,1154.6 2243.7,1156.43 2245.94,1158.44 2248.18,1160.59 2250.42,1162.86 2252.66,1165.22 2254.91,1167.63 2257.15,1170.07 2259.39,1172.51 2261.63,1174.9 2263.87,1177.24 2266.12,1179.47 2268.36,1181.58 2270.6,1183.54 2272.84,1185.32 2275.08,1186.89 2277.33,1188.25 2279.57,1189.37 2281.81,1190.24 2284.05,1190.84 2286.29,1191.17 2288.54,1191.24 2290.78,1191.03 2293.02,1190.56 2295.26,1189.84 2297.5,1188.88 2299.75,1187.7 2301.99,1186.32 2304.23,1184.78 2306.47,1183.09 2308.71,1181.29 2310.96,1179.42 2313.2,1177.51 2315.44,1175.6 2317.68,1173.72 2319.92,1171.9 2322.17,1170.19 2324.41,1168.61 2326.65,1167.19 2328.89,1165.96 2331.13,1164.94 2333.38,1164.14 2335.62,1163.59 2337.86,1163.3 2340.1,1163.26 2342.34,1163.49 2344.58,1163.98 2346.83,1164.72 2349.07,1165.7 2351.31,1166.9 2353.55,1168.32 2355.79,1169.93 2358.04,1171.7 2360.28,1173.62 2362.52,1175.64 2364.76,1177.75 2367,1179.92 2369.25,1182.11 2371.49,1184.29 2373.73,1186.43 2375.97,1188.51 2378.21,1190.49 2380.46,1192.35 2382.7,1194.06 2384.94,1195.61 2387.18,1196.97 2389.42,1198.12 2391.67,1199.06 2393.91,1199.77 2396.15,1200.25 2398.39,1200.51 2400.63,1200.54 2402.88,1200.35 2405.12,1199.95 2407.36,1199.37 2409.6,1198.62 2411.84,1197.72 2414.09,1196.7 2416.33,1195.58 2418.57,1194.41 2420.81,1193.2 2423.05,1192 2425.3,1190.82 2427.54,1189.71 2429.78,1188.7 2432.02,1187.8 2434.26,1187.06 2436.51,1186.48 2438.75,1186.09 2440.99,1185.91 2443.23,1185.94 2445.47,1186.2 2447.72,1186.69 2449.96,1187.41 2452.2,1188.35 2454.44,1189.51 2456.68,1190.87 2458.92,1192.43 2461.17,1194.15 2463.41,1196.03 2465.65,1198.03 2467.89,1200.13 2470.13,1202.31 2472.38,1204.54 2474.62,1206.79 2476.86,1209.04 2479.1,1211.24 2481.34,1213.39 2483.59,1215.46 2485.83,1217.41 2488.07,1219.24 2490.31,1220.91 2492.55,1222.43 2494.8,1223.78 2497.04,1224.95 2499.28,1225.93 2501.52,1226.73 2503.76,1227.35 2506.01,1227.8 2508.25,1228.09 2510.49,1228.24 2512.73,1228.26 2514.97,1228.18 2517.22,1228.02 2519.46,1227.82 2521.7,1227.59 2523.94,1227.36 2526.18,1227.18 2528.43,1227.06 2530.67,1227.04 2532.91,1227.14 2535.15,1227.4 2537.39,1227.83 2539.64,1228.45 2541.88,1229.29 2544.12,1230.36 2546.36,1231.66 2548.6,1233.22 2550.85,1235.02 2553.09,1237.07 2555.33,1239.37 2557.57,1241.9 2559.81,1244.65 2562.05,1247.61 2564.3,1250.76 2566.54,1254.08 2568.78,1257.54 2571.02,1261.12 2573.26,1264.78 2575.51,1268.51 2577.75,1272.26 2579.99,1276.01 2582.23,1279.74 2584.47,1283.4 2586.72,1286.99 2588.96,1290.46 2591.2,1293.8 2593.44,1296.99 2595.68,1300.01 2597.93,1302.86 2600.17,1305.51 2602.41,1307.98 2604.65,1310.26 2606.89,1312.36 2609.14,1314.3 2611.38,1316.09 2613.62,1317.75 2615.86,1319.32 2618.1,1320.84 2620.35,1322.34 2622.59,1323.88 2624.83,1325.49 2627.07,1327.24 2629.31,1329.19 2631.56,1331.4 2633.8,1333.93 2636.04,1336.84 2638.28,1340.21 2640.52,1344.1 2642.77,1348.56 2645.01,1353.66 2647.25,1359.46 2649.49,1366.01 2651.73,1373.35 2653.98,1381.53 2656.22,1390.56 2658.46,1400.49 2660.7,1411.31 2662.94,1423.03 2665.19,1435.64 2667.43,1449.12 2669.67,1463.43 2671.91,1478.5 2674.15,1494.29 2676.39,1510.69 2678.64,1527.61 2680.88,1544.92 2683.12,1562.46 2685.36,1580.08 2687.6,1597.57 2689.85,1614.72 2692.09,1631.27 2694.33,1646.97 2696.57,1661.51 2698.81,1674.57 2701.06,1685.83 2703.3,1694.93 2705.54,1701.51 2707.78,1705.25 2710.02,1705.81 2712.27,1702.94 2714.51,1696.41 2716.75,1686.11 2718.99,1672.06 "/>
-<path clip-path="url(#clip210)" d="M159.765 2657.17 L1352.76 2657.17 L1352.76 2020.38 L159.765 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,1723.83 1593.53,1087.05 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,1723.83 1873.77,1087.05 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,1723.83 2154.02,1087.05 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,1723.83 2434.26,1087.05 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,1723.83 2714.51,1087.05 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1723.83 2752.76,1723.83 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,1723.83 1593.53,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,1723.83 1873.77,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,1723.83 2154.02,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,1723.83 2434.26,1704.94 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,1723.83 2714.51,1704.94 "/>
+<path clip-path="url(#clip030)" d="M1593.53 1757.95 Q1589.92 1757.95 1588.09 1761.52 Q1586.28 1765.06 1586.28 1772.19 Q1586.28 1779.3 1588.09 1782.86 Q1589.92 1786.4 1593.53 1786.4 Q1597.16 1786.4 1598.97 1782.86 Q1600.8 1779.3 1600.8 1772.19 Q1600.8 1765.06 1598.97 1761.52 Q1597.16 1757.95 1593.53 1757.95 M1593.53 1754.25 Q1599.34 1754.25 1602.39 1758.86 Q1605.47 1763.44 1605.47 1772.19 Q1605.47 1780.92 1602.39 1785.52 Q1599.34 1790.11 1593.53 1790.11 Q1587.72 1790.11 1584.64 1785.52 Q1581.58 1780.92 1581.58 1772.19 Q1581.58 1763.44 1584.64 1758.86 Q1587.72 1754.25 1593.53 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1833.39 1754.87 L1851.75 1754.87 L1851.75 1758.81 L1837.67 1758.81 L1837.67 1767.28 Q1838.69 1766.93 1839.71 1766.77 Q1840.73 1766.59 1841.75 1766.59 Q1847.54 1766.59 1850.91 1769.76 Q1854.29 1772.93 1854.29 1778.35 Q1854.29 1783.93 1850.82 1787.03 Q1847.35 1790.11 1841.03 1790.11 Q1838.85 1790.11 1836.59 1789.74 Q1834.34 1789.37 1831.93 1788.62 L1831.93 1783.93 Q1834.02 1785.06 1836.24 1785.62 Q1838.46 1786.17 1840.94 1786.17 Q1844.94 1786.17 1847.28 1784.06 Q1849.62 1781.96 1849.62 1778.35 Q1849.62 1774.74 1847.28 1772.63 Q1844.94 1770.52 1840.94 1770.52 Q1839.06 1770.52 1837.19 1770.94 Q1835.34 1771.36 1833.39 1772.24 L1833.39 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1873.51 1757.95 Q1869.9 1757.95 1868.07 1761.52 Q1866.26 1765.06 1866.26 1772.19 Q1866.26 1779.3 1868.07 1782.86 Q1869.9 1786.4 1873.51 1786.4 Q1877.14 1786.4 1878.95 1782.86 Q1880.78 1779.3 1880.78 1772.19 Q1880.78 1765.06 1878.95 1761.52 Q1877.14 1757.95 1873.51 1757.95 M1873.51 1754.25 Q1879.32 1754.25 1882.37 1758.86 Q1885.45 1763.44 1885.45 1772.19 Q1885.45 1780.92 1882.37 1785.52 Q1879.32 1790.11 1873.51 1790.11 Q1867.7 1790.11 1864.62 1785.52 Q1861.56 1780.92 1861.56 1772.19 Q1861.56 1763.44 1864.62 1758.86 Q1867.7 1754.25 1873.51 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1903.67 1757.95 Q1900.06 1757.95 1898.23 1761.52 Q1896.42 1765.06 1896.42 1772.19 Q1896.42 1779.3 1898.23 1782.86 Q1900.06 1786.4 1903.67 1786.4 Q1907.3 1786.4 1909.11 1782.86 Q1910.94 1779.3 1910.94 1772.19 Q1910.94 1765.06 1909.11 1761.52 Q1907.3 1757.95 1903.67 1757.95 M1903.67 1754.25 Q1909.48 1754.25 1912.53 1758.86 Q1915.61 1763.44 1915.61 1772.19 Q1915.61 1780.92 1912.53 1785.52 Q1909.48 1790.11 1903.67 1790.11 Q1897.86 1790.11 1894.78 1785.52 Q1891.72 1780.92 1891.72 1772.19 Q1891.72 1763.44 1894.78 1758.86 Q1897.86 1754.25 1903.67 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2098.54 1785.5 L2106.18 1785.5 L2106.18 1759.13 L2097.87 1760.8 L2097.87 1756.54 L2106.14 1754.87 L2110.81 1754.87 L2110.81 1785.5 L2118.45 1785.5 L2118.45 1789.43 L2098.54 1789.43 L2098.54 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2137.9 1757.95 Q2134.28 1757.95 2132.46 1761.52 Q2130.65 1765.06 2130.65 1772.19 Q2130.65 1779.3 2132.46 1782.86 Q2134.28 1786.4 2137.9 1786.4 Q2141.53 1786.4 2143.34 1782.86 Q2145.16 1779.3 2145.16 1772.19 Q2145.16 1765.06 2143.34 1761.52 Q2141.53 1757.95 2137.9 1757.95 M2137.9 1754.25 Q2143.71 1754.25 2146.76 1758.86 Q2149.84 1763.44 2149.84 1772.19 Q2149.84 1780.92 2146.76 1785.52 Q2143.71 1790.11 2137.9 1790.11 Q2132.09 1790.11 2129.01 1785.52 Q2125.95 1780.92 2125.95 1772.19 Q2125.95 1763.44 2129.01 1758.86 Q2132.09 1754.25 2137.9 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2168.06 1757.95 Q2164.45 1757.95 2162.62 1761.52 Q2160.81 1765.06 2160.81 1772.19 Q2160.81 1779.3 2162.62 1782.86 Q2164.45 1786.4 2168.06 1786.4 Q2171.69 1786.4 2173.5 1782.86 Q2175.33 1779.3 2175.33 1772.19 Q2175.33 1765.06 2173.5 1761.52 Q2171.69 1757.95 2168.06 1757.95 M2168.06 1754.25 Q2173.87 1754.25 2176.92 1758.86 Q2180 1763.44 2180 1772.19 Q2180 1780.92 2176.92 1785.52 Q2173.87 1790.11 2168.06 1790.11 Q2162.25 1790.11 2159.17 1785.52 Q2156.11 1780.92 2156.11 1772.19 Q2156.11 1763.44 2159.17 1758.86 Q2162.25 1754.25 2168.06 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2198.22 1757.95 Q2194.61 1757.95 2192.78 1761.52 Q2190.97 1765.06 2190.97 1772.19 Q2190.97 1779.3 2192.78 1782.86 Q2194.61 1786.4 2198.22 1786.4 Q2201.85 1786.4 2203.66 1782.86 Q2205.49 1779.3 2205.49 1772.19 Q2205.49 1765.06 2203.66 1761.52 Q2201.85 1757.95 2198.22 1757.95 M2198.22 1754.25 Q2204.03 1754.25 2207.09 1758.86 Q2210.16 1763.44 2210.16 1772.19 Q2210.16 1780.92 2207.09 1785.52 Q2204.03 1790.11 2198.22 1790.11 Q2192.41 1790.11 2189.33 1785.52 Q2186.28 1780.92 2186.28 1772.19 Q2186.28 1763.44 2189.33 1758.86 Q2192.41 1754.25 2198.22 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2378.79 1785.5 L2386.43 1785.5 L2386.43 1759.13 L2378.12 1760.8 L2378.12 1756.54 L2386.38 1754.87 L2391.06 1754.87 L2391.06 1785.5 L2398.7 1785.5 L2398.7 1789.43 L2378.79 1789.43 L2378.79 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2408.19 1754.87 L2426.54 1754.87 L2426.54 1758.81 L2412.47 1758.81 L2412.47 1767.28 Q2413.49 1766.93 2414.51 1766.77 Q2415.52 1766.59 2416.54 1766.59 Q2422.33 1766.59 2425.71 1769.76 Q2429.09 1772.93 2429.09 1778.35 Q2429.09 1783.93 2425.62 1787.03 Q2422.15 1790.11 2415.83 1790.11 Q2413.65 1790.11 2411.38 1789.74 Q2409.14 1789.37 2406.73 1788.62 L2406.73 1783.93 Q2408.81 1785.06 2411.03 1785.62 Q2413.26 1786.17 2415.73 1786.17 Q2419.74 1786.17 2422.08 1784.06 Q2424.41 1781.96 2424.41 1778.35 Q2424.41 1774.74 2422.08 1772.63 Q2419.74 1770.52 2415.73 1770.52 Q2413.86 1770.52 2411.98 1770.94 Q2410.13 1771.36 2408.19 1772.24 L2408.19 1754.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2448.3 1757.95 Q2444.69 1757.95 2442.86 1761.52 Q2441.06 1765.06 2441.06 1772.19 Q2441.06 1779.3 2442.86 1782.86 Q2444.69 1786.4 2448.3 1786.4 Q2451.94 1786.4 2453.74 1782.86 Q2455.57 1779.3 2455.57 1772.19 Q2455.57 1765.06 2453.74 1761.52 Q2451.94 1757.95 2448.3 1757.95 M2448.3 1754.25 Q2454.11 1754.25 2457.17 1758.86 Q2460.25 1763.44 2460.25 1772.19 Q2460.25 1780.92 2457.17 1785.52 Q2454.11 1790.11 2448.3 1790.11 Q2442.49 1790.11 2439.41 1785.52 Q2436.36 1780.92 2436.36 1772.19 Q2436.36 1763.44 2439.41 1758.86 Q2442.49 1754.25 2448.3 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2478.46 1757.95 Q2474.85 1757.95 2473.02 1761.52 Q2471.22 1765.06 2471.22 1772.19 Q2471.22 1779.3 2473.02 1782.86 Q2474.85 1786.4 2478.46 1786.4 Q2482.1 1786.4 2483.9 1782.86 Q2485.73 1779.3 2485.73 1772.19 Q2485.73 1765.06 2483.9 1761.52 Q2482.1 1757.95 2478.46 1757.95 M2478.46 1754.25 Q2484.27 1754.25 2487.33 1758.86 Q2490.41 1763.44 2490.41 1772.19 Q2490.41 1780.92 2487.33 1785.52 Q2484.27 1790.11 2478.46 1790.11 Q2472.65 1790.11 2469.58 1785.52 Q2466.52 1780.92 2466.52 1772.19 Q2466.52 1763.44 2469.58 1758.86 Q2472.65 1754.25 2478.46 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2663.12 1785.5 L2679.44 1785.5 L2679.44 1789.43 L2657.49 1789.43 L2657.49 1785.5 Q2660.16 1782.74 2664.74 1778.12 Q2669.35 1773.46 2670.53 1772.12 Q2672.77 1769.6 2673.65 1767.86 Q2674.55 1766.1 2674.55 1764.41 Q2674.55 1761.66 2672.61 1759.92 Q2670.69 1758.18 2667.59 1758.18 Q2665.39 1758.18 2662.93 1758.95 Q2660.5 1759.71 2657.73 1761.26 L2657.73 1756.54 Q2660.55 1755.41 2663 1754.83 Q2665.46 1754.25 2667.49 1754.25 Q2672.86 1754.25 2676.06 1756.93 Q2679.25 1759.62 2679.25 1764.11 Q2679.25 1766.24 2678.44 1768.16 Q2677.66 1770.06 2675.55 1772.65 Q2674.97 1773.32 2671.87 1776.54 Q2668.77 1779.74 2663.12 1785.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2699.25 1757.95 Q2695.64 1757.95 2693.81 1761.52 Q2692.01 1765.06 2692.01 1772.19 Q2692.01 1779.3 2693.81 1782.86 Q2695.64 1786.4 2699.25 1786.4 Q2702.89 1786.4 2704.69 1782.86 Q2706.52 1779.3 2706.52 1772.19 Q2706.52 1765.06 2704.69 1761.52 Q2702.89 1757.95 2699.25 1757.95 M2699.25 1754.25 Q2705.06 1754.25 2708.12 1758.86 Q2711.2 1763.44 2711.2 1772.19 Q2711.2 1780.92 2708.12 1785.52 Q2705.06 1790.11 2699.25 1790.11 Q2693.44 1790.11 2690.36 1785.52 Q2687.31 1780.92 2687.31 1772.19 Q2687.31 1763.44 2690.36 1758.86 Q2693.44 1754.25 2699.25 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2729.42 1757.95 Q2725.8 1757.95 2723.98 1761.52 Q2722.17 1765.06 2722.17 1772.19 Q2722.17 1779.3 2723.98 1782.86 Q2725.8 1786.4 2729.42 1786.4 Q2733.05 1786.4 2734.86 1782.86 Q2736.68 1779.3 2736.68 1772.19 Q2736.68 1765.06 2734.86 1761.52 Q2733.05 1757.95 2729.42 1757.95 M2729.42 1754.25 Q2735.23 1754.25 2738.28 1758.86 Q2741.36 1763.44 2741.36 1772.19 Q2741.36 1780.92 2738.28 1785.52 Q2735.23 1790.11 2729.42 1790.11 Q2723.61 1790.11 2720.53 1785.52 Q2717.47 1780.92 2717.47 1772.19 Q2717.47 1763.44 2720.53 1758.86 Q2723.61 1754.25 2729.42 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2759.58 1757.95 Q2755.97 1757.95 2754.14 1761.52 Q2752.33 1765.06 2752.33 1772.19 Q2752.33 1779.3 2754.14 1782.86 Q2755.97 1786.4 2759.58 1786.4 Q2763.21 1786.4 2765.02 1782.86 Q2766.85 1779.3 2766.85 1772.19 Q2766.85 1765.06 2765.02 1761.52 Q2763.21 1757.95 2759.58 1757.95 M2759.58 1754.25 Q2765.39 1754.25 2768.44 1758.86 Q2771.52 1763.44 2771.52 1772.19 Q2771.52 1780.92 2768.44 1785.52 Q2765.39 1790.11 2759.58 1790.11 Q2753.77 1790.11 2750.69 1785.52 Q2747.63 1780.92 2747.63 1772.19 Q2747.63 1763.44 2750.69 1758.86 Q2753.77 1754.25 2759.58 1754.25 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1672.06 2752.76,1672.06 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1567.36 2752.76,1567.36 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1462.66 2752.76,1462.66 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1357.96 2752.76,1357.96 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1253.26 2752.76,1253.26 "/>
+<polyline clip-path="url(#clip035)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,1148.56 2752.76,1148.56 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1723.83 1559.76,1087.05 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1672.06 1578.66,1672.06 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1567.36 1578.66,1567.36 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1462.66 1578.66,1462.66 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1357.96 1578.66,1357.96 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1253.26 1578.66,1253.26 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,1148.56 1578.66,1148.56 "/>
+<path clip-path="url(#clip030)" d="M1457.05 1672.51 L1486.72 1672.51 L1486.72 1676.44 L1457.05 1676.44 L1457.05 1672.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1496.86 1654.78 L1515.22 1654.78 L1515.22 1658.71 L1501.14 1658.71 L1501.14 1667.18 Q1502.16 1666.84 1503.18 1666.68 Q1504.2 1666.49 1505.22 1666.49 Q1511.01 1666.49 1514.39 1669.66 Q1517.76 1672.83 1517.76 1678.25 Q1517.76 1683.83 1514.29 1686.93 Q1510.82 1690.01 1504.5 1690.01 Q1502.33 1690.01 1500.06 1689.64 Q1497.81 1689.27 1495.4 1688.53 L1495.4 1683.83 Q1497.49 1684.96 1499.71 1685.52 Q1501.93 1686.07 1504.41 1686.07 Q1508.41 1686.07 1510.75 1683.97 Q1513.09 1681.86 1513.09 1678.25 Q1513.09 1674.64 1510.75 1672.53 Q1508.41 1670.43 1504.41 1670.43 Q1502.53 1670.43 1500.66 1670.84 Q1498.81 1671.26 1496.86 1672.14 L1496.86 1654.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1455.57 1567.81 L1485.24 1567.81 L1485.24 1571.74 L1455.57 1571.74 L1455.57 1567.81 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1508.18 1554.15 L1496.38 1572.6 L1508.18 1572.6 L1508.18 1554.15 M1506.95 1550.08 L1512.83 1550.08 L1512.83 1572.6 L1517.76 1572.6 L1517.76 1576.49 L1512.83 1576.49 L1512.83 1584.64 L1508.18 1584.64 L1508.18 1576.49 L1492.58 1576.49 L1492.58 1571.98 L1506.95 1550.08 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1456.7 1463.11 L1486.38 1463.11 L1486.38 1467.04 L1456.7 1467.04 L1456.7 1463.11 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1510.64 1461.3 Q1513.99 1462.02 1515.87 1464.29 Q1517.76 1466.56 1517.76 1469.89 Q1517.76 1475.01 1514.25 1477.81 Q1510.73 1480.61 1504.25 1480.61 Q1502.07 1480.61 1499.76 1480.17 Q1497.46 1479.75 1495.01 1478.9 L1495.01 1474.38 Q1496.95 1475.52 1499.27 1476.09 Q1501.58 1476.67 1504.11 1476.67 Q1508.51 1476.67 1510.8 1474.94 Q1513.11 1473.2 1513.11 1469.89 Q1513.11 1466.84 1510.96 1465.12 Q1508.83 1463.39 1505.01 1463.39 L1500.98 1463.39 L1500.98 1459.54 L1505.2 1459.54 Q1508.64 1459.54 1510.47 1458.18 Q1512.3 1456.79 1512.3 1454.2 Q1512.3 1451.53 1510.4 1450.12 Q1508.53 1448.69 1505.01 1448.69 Q1503.09 1448.69 1500.89 1449.1 Q1498.69 1449.52 1496.05 1450.4 L1496.05 1446.23 Q1498.71 1445.49 1501.03 1445.12 Q1503.37 1444.75 1505.43 1444.75 Q1510.75 1444.75 1513.85 1447.18 Q1516.95 1449.59 1516.95 1453.71 Q1516.95 1456.58 1515.31 1458.57 Q1513.67 1460.54 1510.64 1461.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1457.65 1358.41 L1487.33 1358.41 L1487.33 1362.34 L1457.65 1362.34 L1457.65 1358.41 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1501.45 1371.3 L1517.76 1371.3 L1517.76 1375.24 L1495.82 1375.24 L1495.82 1371.3 Q1498.48 1368.55 1503.07 1363.92 Q1507.67 1359.27 1508.85 1357.92 Q1511.1 1355.4 1511.98 1353.66 Q1512.88 1351.9 1512.88 1350.21 Q1512.88 1347.46 1510.94 1345.72 Q1509.01 1343.99 1505.91 1343.99 Q1503.71 1343.99 1501.26 1344.75 Q1498.83 1345.52 1496.05 1347.07 L1496.05 1342.34 Q1498.88 1341.21 1501.33 1340.63 Q1503.78 1340.05 1505.82 1340.05 Q1511.19 1340.05 1514.39 1342.74 Q1517.58 1345.42 1517.58 1349.91 Q1517.58 1352.04 1516.77 1353.96 Q1515.98 1355.86 1513.88 1358.46 Q1513.3 1359.13 1510.2 1362.34 Q1507.09 1365.54 1501.45 1371.3 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1457.28 1253.71 L1486.95 1253.71 L1486.95 1257.64 L1457.28 1257.64 L1457.28 1253.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1497.86 1266.6 L1505.5 1266.6 L1505.5 1240.24 L1497.19 1241.9 L1497.19 1237.64 L1505.45 1235.98 L1510.13 1235.98 L1510.13 1266.6 L1517.76 1266.6 L1517.76 1270.54 L1497.86 1270.54 L1497.86 1266.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 1134.36 Q1502.21 1134.36 1500.38 1137.92 Q1498.58 1141.46 1498.58 1148.59 Q1498.58 1155.7 1500.38 1159.26 Q1502.21 1162.81 1505.82 1162.81 Q1509.45 1162.81 1511.26 1159.26 Q1513.09 1155.7 1513.09 1148.59 Q1513.09 1141.46 1511.26 1137.92 Q1509.45 1134.36 1505.82 1134.36 M1505.82 1130.65 Q1511.63 1130.65 1514.69 1135.26 Q1517.76 1139.84 1517.76 1148.59 Q1517.76 1157.32 1514.69 1161.93 Q1511.63 1166.51 1505.82 1166.51 Q1500.01 1166.51 1496.93 1161.93 Q1493.88 1157.32 1493.88 1148.59 Q1493.88 1139.84 1496.93 1135.26 Q1500.01 1130.65 1505.82 1130.65 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1820.06 945.429 L1854.81 945.429 L1854.81 952.316 L1828.24 952.316 L1828.24 970.14 L1852.22 970.14 L1852.22 977.026 L1828.24 977.026 L1828.24 1005.91 L1820.06 1005.91 L1820.06 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1867.45 942.877 L1874.9 942.877 L1874.9 1005.91 L1867.45 1005.91 L1867.45 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1890.5 960.539 L1897.95 960.539 L1897.95 1005.91 L1890.5 1005.91 L1890.5 960.539 M1890.5 942.877 L1897.95 942.877 L1897.95 952.316 L1890.5 952.316 L1890.5 942.877 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1943.41 982.698 Q1943.41 974.596 1940.04 970.14 Q1936.72 965.684 1930.69 965.684 Q1924.69 965.684 1921.33 970.14 Q1918.01 974.596 1918.01 982.698 Q1918.01 990.759 1921.33 995.215 Q1924.69 999.671 1930.69 999.671 Q1936.72 999.671 1940.04 995.215 Q1943.41 990.759 1943.41 982.698 M1950.86 1000.28 Q1950.86 1011.86 1945.71 1017.49 Q1940.57 1023.17 1929.96 1023.17 Q1926.03 1023.17 1922.54 1022.56 Q1919.06 1021.99 1915.78 1020.78 L1915.78 1013.53 Q1919.06 1015.31 1922.26 1016.16 Q1925.46 1017.01 1928.78 1017.01 Q1936.11 1017.01 1939.76 1013.16 Q1943.41 1009.35 1943.41 1001.62 L1943.41 997.929 Q1941.1 1001.94 1937.49 1003.92 Q1933.89 1005.91 1928.86 1005.91 Q1920.52 1005.91 1915.41 999.549 Q1910.31 993.189 1910.31 982.698 Q1910.31 972.165 1915.41 965.805 Q1920.52 959.445 1928.86 959.445 Q1933.89 959.445 1937.49 961.43 Q1941.1 963.415 1943.41 967.426 L1943.41 960.539 L1950.86 960.539 L1950.86 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2003.93 978.525 L2003.93 1005.91 L1996.47 1005.91 L1996.47 978.768 Q1996.47 972.327 1993.96 969.127 Q1991.45 965.927 1986.43 965.927 Q1980.39 965.927 1976.91 969.775 Q1973.42 973.624 1973.42 980.267 L1973.42 1005.91 L1965.93 1005.91 L1965.93 942.877 L1973.42 942.877 L1973.42 967.588 Q1976.1 963.496 1979.7 961.471 Q1983.35 959.445 1988.09 959.445 Q1995.91 959.445 1999.92 964.307 Q2003.93 969.127 2003.93 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2026.17 947.657 L2026.17 960.539 L2041.52 960.539 L2041.52 966.332 L2026.17 966.332 L2026.17 990.961 Q2026.17 996.511 2027.66 998.091 Q2029.2 999.671 2033.86 999.671 L2041.52 999.671 L2041.52 1005.91 L2033.86 1005.91 Q2025.23 1005.91 2021.95 1002.71 Q2018.67 999.468 2018.67 990.961 L2018.67 966.332 L2013.2 966.332 L2013.2 960.539 L2018.67 960.539 L2018.67 947.657 L2026.17 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2086.2 952.154 L2086.2 974.879 L2096.49 974.879 Q2102.2 974.879 2105.32 971.922 Q2108.44 968.965 2108.44 963.496 Q2108.44 958.068 2105.32 955.111 Q2102.2 952.154 2096.49 952.154 L2086.2 952.154 M2078.02 945.429 L2096.49 945.429 Q2106.66 945.429 2111.84 950.047 Q2117.07 954.625 2117.07 963.496 Q2117.07 972.449 2111.84 977.026 Q2106.66 981.604 2096.49 981.604 L2086.2 981.604 L2086.2 1005.91 L2078.02 1005.91 L2078.02 945.429 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2144.61 983.103 Q2135.58 983.103 2132.1 985.169 Q2128.61 987.235 2128.61 992.217 Q2128.61 996.187 2131.21 998.537 Q2133.84 1000.85 2138.34 1000.85 Q2144.53 1000.85 2148.26 996.471 Q2152.03 992.055 2152.03 984.764 L2152.03 983.103 L2144.61 983.103 M2159.48 980.024 L2159.48 1005.91 L2152.03 1005.91 L2152.03 999.023 Q2149.48 1003.15 2145.67 1005.14 Q2141.86 1007.08 2136.35 1007.08 Q2129.38 1007.08 2125.25 1003.2 Q2121.16 999.266 2121.16 992.703 Q2121.16 985.047 2126.26 981.158 Q2131.41 977.269 2141.58 977.269 L2152.03 977.269 L2152.03 976.54 Q2152.03 971.396 2148.62 968.601 Q2145.26 965.765 2139.15 965.765 Q2135.26 965.765 2131.57 966.697 Q2127.88 967.628 2124.48 969.492 L2124.48 962.605 Q2128.57 961.025 2132.42 960.256 Q2136.27 959.445 2139.92 959.445 Q2149.76 959.445 2154.62 964.55 Q2159.48 969.654 2159.48 980.024 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2182.21 947.657 L2182.21 960.539 L2197.56 960.539 L2197.56 966.332 L2182.21 966.332 L2182.21 990.961 Q2182.21 996.511 2183.71 998.091 Q2185.24 999.671 2189.9 999.671 L2197.56 999.671 L2197.56 1005.91 L2189.9 1005.91 Q2181.27 1005.91 2177.99 1002.71 Q2174.71 999.468 2174.71 990.961 L2174.71 966.332 L2169.24 966.332 L2169.24 960.539 L2174.71 960.539 L2174.71 947.657 L2182.21 947.657 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2245.08 978.525 L2245.08 1005.91 L2237.62 1005.91 L2237.62 978.768 Q2237.62 972.327 2235.11 969.127 Q2232.6 965.927 2227.58 965.927 Q2221.54 965.927 2218.06 969.775 Q2214.57 973.624 2214.57 980.267 L2214.57 1005.91 L2207.08 1005.91 L2207.08 942.877 L2214.57 942.877 L2214.57 967.588 Q2217.25 963.496 2220.85 961.471 Q2224.5 959.445 2229.24 959.445 Q2237.06 959.445 2241.07 964.307 Q2245.08 969.127 2245.08 978.525 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2304.22 942.958 Q2298.79 952.275 2296.16 961.39 Q2293.53 970.504 2293.53 979.862 Q2293.53 989.22 2296.16 998.415 Q2298.83 1007.57 2304.22 1016.85 L2297.74 1016.85 Q2291.66 1007.33 2288.62 998.132 Q2285.63 988.936 2285.63 979.862 Q2285.63 970.829 2288.62 961.673 Q2291.62 952.518 2297.74 942.958 L2304.22 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2348.54 967.426 L2348.54 942.877 L2355.99 942.877 L2355.99 1005.91 L2348.54 1005.91 L2348.54 999.104 Q2346.19 1003.15 2342.58 1005.14 Q2339.02 1007.08 2333.99 1007.08 Q2325.77 1007.08 2320.59 1000.52 Q2315.44 993.959 2315.44 983.265 Q2315.44 972.57 2320.59 966.008 Q2325.77 959.445 2333.99 959.445 Q2339.02 959.445 2342.58 961.43 Q2346.19 963.375 2348.54 967.426 M2323.14 983.265 Q2323.14 991.488 2326.5 996.187 Q2329.9 1000.85 2335.82 1000.85 Q2341.73 1000.85 2345.13 996.187 Q2348.54 991.488 2348.54 983.265 Q2348.54 975.041 2345.13 970.383 Q2341.73 965.684 2335.82 965.684 Q2329.9 965.684 2326.5 970.383 Q2323.14 975.041 2323.14 983.265 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2410.15 981.361 L2410.15 985.007 L2375.88 985.007 Q2376.37 992.703 2380.5 996.754 Q2384.67 1000.76 2392.08 1000.76 Q2396.38 1000.76 2400.39 999.711 Q2404.44 998.658 2408.41 996.552 L2408.41 1003.6 Q2404.4 1005.3 2400.19 1006.19 Q2395.97 1007.08 2391.64 1007.08 Q2380.78 1007.08 2374.42 1000.76 Q2368.1 994.445 2368.1 983.67 Q2368.1 972.53 2374.1 966.008 Q2380.13 959.445 2390.34 959.445 Q2399.5 959.445 2404.8 965.36 Q2410.15 971.234 2410.15 981.361 M2402.7 979.173 Q2402.62 973.057 2399.25 969.411 Q2395.93 965.765 2390.42 965.765 Q2384.18 965.765 2380.42 969.289 Q2376.69 972.813 2376.12 979.214 L2402.7 979.173 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2452.24 982.698 Q2452.24 974.596 2448.88 970.14 Q2445.56 965.684 2439.52 965.684 Q2433.52 965.684 2430.16 970.14 Q2426.84 974.596 2426.84 982.698 Q2426.84 990.759 2430.16 995.215 Q2433.52 999.671 2439.52 999.671 Q2445.56 999.671 2448.88 995.215 Q2452.24 990.759 2452.24 982.698 M2459.69 1000.28 Q2459.69 1011.86 2454.55 1017.49 Q2449.4 1023.17 2438.79 1023.17 Q2434.86 1023.17 2431.38 1022.56 Q2427.89 1021.99 2424.61 1020.78 L2424.61 1013.53 Q2427.89 1015.31 2431.09 1016.16 Q2434.29 1017.01 2437.62 1017.01 Q2444.95 1017.01 2448.59 1013.16 Q2452.24 1009.35 2452.24 1001.62 L2452.24 997.929 Q2449.93 1001.94 2446.33 1003.92 Q2442.72 1005.91 2437.7 1005.91 Q2429.35 1005.91 2424.25 999.549 Q2419.14 993.189 2419.14 982.698 Q2419.14 972.165 2424.25 965.805 Q2429.35 959.445 2437.7 959.445 Q2442.72 959.445 2446.33 961.43 Q2449.93 963.415 2452.24 967.426 L2452.24 960.539 L2459.69 960.539 L2459.69 1000.28 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2473.87 942.958 L2480.35 942.958 Q2486.43 952.518 2489.43 961.673 Q2492.47 970.829 2492.47 979.862 Q2492.47 988.936 2489.43 998.132 Q2486.43 1007.33 2480.35 1016.85 L2473.87 1016.85 Q2479.26 1007.57 2481.89 998.415 Q2484.57 989.22 2484.57 979.862 Q2484.57 970.504 2481.89 961.39 Q2479.26 952.275 2473.87 942.958 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip035)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,1253.26 1595.77,1253.23 1598.01,1253.16 1600.25,1253.07 1602.5,1252.93 1604.74,1252.75 1606.98,1252.53 1609.22,1252.26 1611.46,1251.93 1613.71,1251.55 1615.95,1251.12 1618.19,1250.61 1620.43,1250.04 1622.67,1249.4 1624.92,1248.67 1627.16,1247.86 1629.4,1246.97 1631.64,1245.97 1633.88,1244.88 1636.13,1243.67 1638.37,1242.36 1640.61,1240.92 1642.85,1239.36 1645.09,1237.67 1647.34,1235.84 1649.58,1233.86 1651.82,1231.74 1654.06,1229.47 1656.3,1227.05 1658.55,1224.46 1660.79,1221.72 1663.03,1218.82 1665.27,1215.77 1667.51,1212.56 1669.76,1209.2 1672,1205.69 1674.24,1202.05 1676.48,1198.27 1678.72,1194.39 1680.97,1190.39 1683.21,1186.31 1685.45,1182.16 1687.69,1177.95 1689.93,1173.7 1692.17,1169.44 1694.42,1165.17 1696.66,1160.93 1698.9,1156.74 1701.14,1152.61 1703.38,1148.56 1705.63,1144.62 1707.87,1140.79 1710.11,1137.11 1712.35,1133.59 1714.59,1130.23 1716.84,1127.05 1719.08,1124.07 1721.32,1121.29 1723.56,1118.72 1725.8,1116.37 1728.05,1114.24 1730.29,1112.33 1732.53,1110.65 1734.77,1109.19 1737.01,1107.96 1739.26,1106.95 1741.5,1106.16 1743.74,1105.59 1745.98,1105.23 1748.22,1105.07 1750.47,1105.11 1752.71,1105.35 1754.95,1105.77 1757.19,1106.37 1759.43,1107.15 1761.68,1108.09 1763.92,1109.19 1766.16,1110.44 1768.4,1111.84 1770.64,1113.37 1772.89,1115.03 1775.13,1116.82 1777.37,1118.71 1779.61,1120.72 1781.85,1122.82 1784.1,1125.02 1786.34,1127.3 1788.58,1129.66 1790.82,1132.09 1793.06,1134.59 1795.31,1137.15 1797.55,1139.76 1799.79,1142.42 1802.03,1145.11 1804.27,1147.83 1806.51,1150.58 1808.76,1153.35 1811,1156.13 1813.24,1158.91 1815.48,1161.69 1817.72,1164.46 1819.97,1167.21 1822.21,1169.94 1824.45,1172.63 1826.69,1175.28 1828.93,1177.87 1831.18,1180.41 1833.42,1182.89 1835.66,1185.28 1837.9,1187.59 1840.14,1189.8 1842.39,1191.91 1844.63,1193.9 1846.87,1195.76 1849.11,1197.49 1851.35,1199.07 1853.6,1200.49 1855.84,1201.74 1858.08,1202.8 1860.32,1203.68 1862.56,1204.35 1864.81,1204.82 1867.05,1205.05 1869.29,1205.06 1871.53,1204.84 1873.77,1204.36 1876.02,1203.64 1878.26,1202.66 1880.5,1201.43 1882.74,1199.94 1884.98,1198.2 1887.23,1196.22 1889.47,1193.99 1891.71,1191.54 1893.95,1188.87 1896.19,1186 1898.44,1182.95 1900.68,1179.75 1902.92,1176.4 1905.16,1172.95 1907.4,1169.42 1909.64,1165.84 1911.89,1162.23 1914.13,1158.64 1916.37,1155.08 1918.61,1151.6 1920.85,1148.22 1923.1,1144.97 1925.34,1141.87 1927.58,1138.96 1929.82,1136.26 1932.06,1133.77 1934.31,1131.53 1936.55,1129.55 1938.79,1127.83 1941.03,1126.39 1943.27,1125.24 1945.52,1124.37 1947.76,1123.78 1950,1123.49 1952.24,1123.47 1954.48,1123.74 1956.73,1124.27 1958.97,1125.07 1961.21,1126.11 1963.45,1127.4 1965.69,1128.91 1967.94,1130.63 1970.18,1132.55 1972.42,1134.65 1974.66,1136.91 1976.9,1139.33 1979.15,1141.87 1981.39,1144.53 1983.63,1147.29 1985.87,1150.12 1988.11,1153.02 1990.36,1155.95 1992.6,1158.9 1994.84,1161.86 1997.08,1164.8 1999.32,1167.7 2001.57,1170.54 2003.81,1173.3 2006.05,1175.97 2008.29,1178.53 2010.53,1180.94 2012.78,1183.2 2015.02,1185.29 2017.26,1187.18 2019.5,1188.87 2021.74,1190.33 2023.98,1191.55 2026.23,1192.51 2028.47,1193.2 2030.71,1193.62 2032.95,1193.75 2035.19,1193.59 2037.44,1193.13 2039.68,1192.37 2041.92,1191.33 2044.16,1190 2046.4,1188.39 2048.65,1186.52 2050.89,1184.41 2053.13,1182.07 2055.37,1179.53 2057.61,1176.83 2059.86,1173.98 2062.1,1171.02 2064.34,1167.99 2066.58,1164.92 2068.82,1161.85 2071.07,1158.82 2073.31,1155.85 2075.55,1152.99 2077.79,1150.28 2080.03,1147.73 2082.28,1145.38 2084.52,1143.26 2086.76,1141.39 2089,1139.78 2091.24,1138.45 2093.49,1137.42 2095.73,1136.69 2097.97,1136.26 2100.21,1136.14 2102.45,1136.32 2104.7,1136.8 2106.94,1137.57 2109.18,1138.61 2111.42,1139.91 2113.66,1141.46 2115.91,1143.23 2118.15,1145.21 2120.39,1147.38 2122.63,1149.7 2124.87,1152.16 2127.11,1154.73 2129.36,1157.39 2131.6,1160.1 2133.84,1162.84 2136.08,1165.59 2138.32,1168.3 2140.57,1170.97 2142.81,1173.55 2145.05,1176.03 2147.29,1178.37 2149.53,1180.55 2151.78,1182.55 2154.02,1184.34 2156.26,1185.91 2158.5,1187.23 2160.74,1188.29 2162.99,1189.08 2165.23,1189.58 2167.47,1189.78 2169.71,1189.69 2171.95,1189.31 2174.2,1188.63 2176.44,1187.67 2178.68,1186.44 2180.92,1184.95 2183.16,1183.23 2185.41,1181.29 2187.65,1179.18 2189.89,1176.92 2192.13,1174.53 2194.37,1172.07 2196.62,1169.56 2198.86,1167.04 2201.1,1164.56 2203.34,1162.14 2205.58,1159.82 2207.83,1157.65 2210.07,1155.65 2212.31,1153.84 2214.55,1152.26 2216.79,1150.93 2219.04,1149.86 2221.28,1149.06 2223.52,1148.56 2225.76,1148.34 2228,1148.42 2230.25,1148.79 2232.49,1149.44 2234.73,1150.36 2236.97,1151.54 2239.21,1152.96 2241.45,1154.6 2243.7,1156.43 2245.94,1158.44 2248.18,1160.59 2250.42,1162.86 2252.66,1165.22 2254.91,1167.63 2257.15,1170.07 2259.39,1172.51 2261.63,1174.9 2263.87,1177.24 2266.12,1179.47 2268.36,1181.58 2270.6,1183.54 2272.84,1185.32 2275.08,1186.89 2277.33,1188.25 2279.57,1189.37 2281.81,1190.24 2284.05,1190.84 2286.29,1191.17 2288.54,1191.24 2290.78,1191.03 2293.02,1190.56 2295.26,1189.84 2297.5,1188.88 2299.75,1187.7 2301.99,1186.32 2304.23,1184.78 2306.47,1183.09 2308.71,1181.29 2310.96,1179.42 2313.2,1177.51 2315.44,1175.6 2317.68,1173.72 2319.92,1171.9 2322.17,1170.19 2324.41,1168.61 2326.65,1167.19 2328.89,1165.96 2331.13,1164.94 2333.38,1164.14 2335.62,1163.59 2337.86,1163.3 2340.1,1163.26 2342.34,1163.49 2344.58,1163.98 2346.83,1164.72 2349.07,1165.7 2351.31,1166.9 2353.55,1168.32 2355.79,1169.93 2358.04,1171.7 2360.28,1173.62 2362.52,1175.64 2364.76,1177.75 2367,1179.92 2369.25,1182.11 2371.49,1184.29 2373.73,1186.43 2375.97,1188.51 2378.21,1190.49 2380.46,1192.35 2382.7,1194.06 2384.94,1195.61 2387.18,1196.97 2389.42,1198.12 2391.67,1199.06 2393.91,1199.77 2396.15,1200.25 2398.39,1200.51 2400.63,1200.54 2402.88,1200.35 2405.12,1199.95 2407.36,1199.37 2409.6,1198.62 2411.84,1197.72 2414.09,1196.7 2416.33,1195.58 2418.57,1194.41 2420.81,1193.2 2423.05,1192 2425.3,1190.82 2427.54,1189.71 2429.78,1188.7 2432.02,1187.8 2434.26,1187.06 2436.51,1186.48 2438.75,1186.09 2440.99,1185.91 2443.23,1185.94 2445.47,1186.2 2447.72,1186.69 2449.96,1187.41 2452.2,1188.35 2454.44,1189.51 2456.68,1190.87 2458.92,1192.43 2461.17,1194.15 2463.41,1196.03 2465.65,1198.03 2467.89,1200.13 2470.13,1202.31 2472.38,1204.54 2474.62,1206.79 2476.86,1209.04 2479.1,1211.24 2481.34,1213.39 2483.59,1215.46 2485.83,1217.41 2488.07,1219.24 2490.31,1220.91 2492.55,1222.43 2494.8,1223.78 2497.04,1224.95 2499.28,1225.93 2501.52,1226.73 2503.76,1227.35 2506.01,1227.8 2508.25,1228.09 2510.49,1228.24 2512.73,1228.26 2514.97,1228.18 2517.22,1228.02 2519.46,1227.82 2521.7,1227.59 2523.94,1227.36 2526.18,1227.18 2528.43,1227.06 2530.67,1227.04 2532.91,1227.14 2535.15,1227.4 2537.39,1227.83 2539.64,1228.45 2541.88,1229.29 2544.12,1230.36 2546.36,1231.66 2548.6,1233.22 2550.85,1235.02 2553.09,1237.07 2555.33,1239.37 2557.57,1241.9 2559.81,1244.65 2562.05,1247.61 2564.3,1250.76 2566.54,1254.08 2568.78,1257.54 2571.02,1261.12 2573.26,1264.78 2575.51,1268.51 2577.75,1272.26 2579.99,1276.01 2582.23,1279.74 2584.47,1283.4 2586.72,1286.99 2588.96,1290.46 2591.2,1293.8 2593.44,1296.99 2595.68,1300.01 2597.93,1302.86 2600.17,1305.51 2602.41,1307.98 2604.65,1310.26 2606.89,1312.36 2609.14,1314.3 2611.38,1316.09 2613.62,1317.75 2615.86,1319.32 2618.1,1320.84 2620.35,1322.34 2622.59,1323.88 2624.83,1325.49 2627.07,1327.24 2629.31,1329.19 2631.56,1331.4 2633.8,1333.93 2636.04,1336.84 2638.28,1340.21 2640.52,1344.1 2642.77,1348.56 2645.01,1353.66 2647.25,1359.46 2649.49,1366.01 2651.73,1373.35 2653.98,1381.53 2656.22,1390.56 2658.46,1400.49 2660.7,1411.31 2662.94,1423.03 2665.19,1435.64 2667.43,1449.12 2669.67,1463.43 2671.91,1478.5 2674.15,1494.29 2676.39,1510.69 2678.64,1527.61 2680.88,1544.92 2683.12,1562.46 2685.36,1580.08 2687.6,1597.57 2689.85,1614.72 2692.09,1631.27 2694.33,1646.97 2696.57,1661.51 2698.81,1674.57 2701.06,1685.83 2703.3,1694.93 2705.54,1701.51 2707.78,1705.25 2710.02,1705.81 2712.27,1702.94 2714.51,1696.41 2716.75,1686.11 2718.99,1672.06 "/>
+<path clip-path="url(#clip030)" d="M159.765 2657.17 L1352.76 2657.17 L1352.76 2020.38 L159.765 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip216">
+  <clipPath id="clip036">
     <rect x="159" y="2020" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,2657.17 193.529,2020.38 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,2657.17 473.774,2020.38 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,2657.17 754.018,2020.38 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,2657.17 1034.26,2020.38 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,2657.17 1314.51,2020.38 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2657.17 1352.76,2657.17 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,2657.17 193.529,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,2657.17 473.774,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,2657.17 754.018,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,2657.17 1034.26,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,2657.17 1314.51,2638.27 "/>
-<path clip-path="url(#clip210)" d="M193.529 2691.29 Q189.918 2691.29 188.089 2694.85 Q186.283 2698.39 186.283 2705.52 Q186.283 2712.63 188.089 2716.19 Q189.918 2719.74 193.529 2719.74 Q197.163 2719.74 198.969 2716.19 Q200.797 2712.63 200.797 2705.52 Q200.797 2698.39 198.969 2694.85 Q197.163 2691.29 193.529 2691.29 M193.529 2687.58 Q199.339 2687.58 202.394 2692.19 Q205.473 2696.77 205.473 2705.52 Q205.473 2714.25 202.394 2718.86 Q199.339 2723.44 193.529 2723.44 Q187.719 2723.44 184.64 2718.86 Q181.584 2714.25 181.584 2705.52 Q181.584 2696.77 184.64 2692.19 Q187.719 2687.58 193.529 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M433.392 2688.21 L451.748 2688.21 L451.748 2692.14 L437.674 2692.14 L437.674 2700.62 Q438.693 2700.27 439.711 2700.11 Q440.73 2699.92 441.748 2699.92 Q447.535 2699.92 450.915 2703.09 Q454.295 2706.26 454.295 2711.68 Q454.295 2717.26 450.822 2720.36 Q447.35 2723.44 441.031 2723.44 Q438.855 2723.44 436.586 2723.07 Q434.341 2722.7 431.934 2721.96 L431.934 2717.26 Q434.017 2718.39 436.239 2718.95 Q438.461 2719.5 440.938 2719.5 Q444.943 2719.5 447.281 2717.4 Q449.619 2715.29 449.619 2711.68 Q449.619 2708.07 447.281 2705.96 Q444.943 2703.86 440.938 2703.86 Q439.063 2703.86 437.188 2704.27 Q435.336 2704.69 433.392 2705.57 L433.392 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M473.507 2691.29 Q469.896 2691.29 468.068 2694.85 Q466.262 2698.39 466.262 2705.52 Q466.262 2712.63 468.068 2716.19 Q469.896 2719.74 473.507 2719.74 Q477.142 2719.74 478.947 2716.19 Q480.776 2712.63 480.776 2705.52 Q480.776 2698.39 478.947 2694.85 Q477.142 2691.29 473.507 2691.29 M473.507 2687.58 Q479.318 2687.58 482.373 2692.19 Q485.452 2696.77 485.452 2705.52 Q485.452 2714.25 482.373 2718.86 Q479.318 2723.44 473.507 2723.44 Q467.697 2723.44 464.619 2718.86 Q461.563 2714.25 461.563 2705.52 Q461.563 2696.77 464.619 2692.19 Q467.697 2687.58 473.507 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M503.669 2691.29 Q500.058 2691.29 498.23 2694.85 Q496.424 2698.39 496.424 2705.52 Q496.424 2712.63 498.23 2716.19 Q500.058 2719.74 503.669 2719.74 Q507.304 2719.74 509.109 2716.19 Q510.938 2712.63 510.938 2705.52 Q510.938 2698.39 509.109 2694.85 Q507.304 2691.29 503.669 2691.29 M503.669 2687.58 Q509.479 2687.58 512.535 2692.19 Q515.614 2696.77 515.614 2705.52 Q515.614 2714.25 512.535 2718.86 Q509.479 2723.44 503.669 2723.44 Q497.859 2723.44 494.78 2718.86 Q491.725 2714.25 491.725 2705.52 Q491.725 2696.77 494.78 2692.19 Q497.859 2687.58 503.669 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M698.544 2718.83 L706.183 2718.83 L706.183 2692.47 L697.873 2694.13 L697.873 2689.87 L706.137 2688.21 L710.813 2688.21 L710.813 2718.83 L718.452 2718.83 L718.452 2722.77 L698.544 2722.77 L698.544 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M737.896 2691.29 Q734.285 2691.29 732.456 2694.85 Q730.651 2698.39 730.651 2705.52 Q730.651 2712.63 732.456 2716.19 Q734.285 2719.74 737.896 2719.74 Q741.53 2719.74 743.336 2716.19 Q745.164 2712.63 745.164 2705.52 Q745.164 2698.39 743.336 2694.85 Q741.53 2691.29 737.896 2691.29 M737.896 2687.58 Q743.706 2687.58 746.762 2692.19 Q749.84 2696.77 749.84 2705.52 Q749.84 2714.25 746.762 2718.86 Q743.706 2723.44 737.896 2723.44 Q732.086 2723.44 729.007 2718.86 Q725.951 2714.25 725.951 2705.52 Q725.951 2696.77 729.007 2692.19 Q732.086 2687.58 737.896 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M768.058 2691.29 Q764.447 2691.29 762.618 2694.85 Q760.812 2698.39 760.812 2705.52 Q760.812 2712.63 762.618 2716.19 Q764.447 2719.74 768.058 2719.74 Q771.692 2719.74 773.498 2716.19 Q775.326 2712.63 775.326 2705.52 Q775.326 2698.39 773.498 2694.85 Q771.692 2691.29 768.058 2691.29 M768.058 2687.58 Q773.868 2687.58 776.923 2692.19 Q780.002 2696.77 780.002 2705.52 Q780.002 2714.25 776.923 2718.86 Q773.868 2723.44 768.058 2723.44 Q762.248 2723.44 759.169 2718.86 Q756.113 2714.25 756.113 2705.52 Q756.113 2696.77 759.169 2692.19 Q762.248 2687.58 768.058 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M798.22 2691.29 Q794.609 2691.29 792.78 2694.85 Q790.974 2698.39 790.974 2705.52 Q790.974 2712.63 792.78 2716.19 Q794.609 2719.74 798.22 2719.74 Q801.854 2719.74 803.659 2716.19 Q805.488 2712.63 805.488 2705.52 Q805.488 2698.39 803.659 2694.85 Q801.854 2691.29 798.22 2691.29 M798.22 2687.58 Q804.03 2687.58 807.085 2692.19 Q810.164 2696.77 810.164 2705.52 Q810.164 2714.25 807.085 2718.86 Q804.03 2723.44 798.22 2723.44 Q792.409 2723.44 789.331 2718.86 Q786.275 2714.25 786.275 2705.52 Q786.275 2696.77 789.331 2692.19 Q792.409 2687.58 798.22 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M978.789 2718.83 L986.428 2718.83 L986.428 2692.47 L978.118 2694.13 L978.118 2689.87 L986.382 2688.21 L991.057 2688.21 L991.057 2718.83 L998.696 2718.83 L998.696 2722.77 L978.789 2722.77 L978.789 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1008.19 2688.21 L1026.54 2688.21 L1026.54 2692.14 L1012.47 2692.14 L1012.47 2700.62 Q1013.49 2700.27 1014.51 2700.11 Q1015.52 2699.92 1016.54 2699.92 Q1022.33 2699.92 1025.71 2703.09 Q1029.09 2706.26 1029.09 2711.68 Q1029.09 2717.26 1025.62 2720.36 Q1022.15 2723.44 1015.83 2723.44 Q1013.65 2723.44 1011.38 2723.07 Q1009.14 2722.7 1006.73 2721.96 L1006.73 2717.26 Q1008.81 2718.39 1011.03 2718.95 Q1013.26 2719.5 1015.73 2719.5 Q1019.74 2719.5 1022.08 2717.4 Q1024.41 2715.29 1024.41 2711.68 Q1024.41 2708.07 1022.08 2705.96 Q1019.74 2703.86 1015.73 2703.86 Q1013.86 2703.86 1011.98 2704.27 Q1010.13 2704.69 1008.19 2705.57 L1008.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1048.3 2691.29 Q1044.69 2691.29 1042.86 2694.85 Q1041.06 2698.39 1041.06 2705.52 Q1041.06 2712.63 1042.86 2716.19 Q1044.69 2719.74 1048.3 2719.74 Q1051.94 2719.74 1053.74 2716.19 Q1055.57 2712.63 1055.57 2705.52 Q1055.57 2698.39 1053.74 2694.85 Q1051.94 2691.29 1048.3 2691.29 M1048.3 2687.58 Q1054.11 2687.58 1057.17 2692.19 Q1060.25 2696.77 1060.25 2705.52 Q1060.25 2714.25 1057.17 2718.86 Q1054.11 2723.44 1048.3 2723.44 Q1042.49 2723.44 1039.41 2718.86 Q1036.36 2714.25 1036.36 2705.52 Q1036.36 2696.77 1039.41 2692.19 Q1042.49 2687.58 1048.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1078.46 2691.29 Q1074.85 2691.29 1073.02 2694.85 Q1071.22 2698.39 1071.22 2705.52 Q1071.22 2712.63 1073.02 2716.19 Q1074.85 2719.74 1078.46 2719.74 Q1082.1 2719.74 1083.9 2716.19 Q1085.73 2712.63 1085.73 2705.52 Q1085.73 2698.39 1083.9 2694.85 Q1082.1 2691.29 1078.46 2691.29 M1078.46 2687.58 Q1084.27 2687.58 1087.33 2692.19 Q1090.41 2696.77 1090.41 2705.52 Q1090.41 2714.25 1087.33 2718.86 Q1084.27 2723.44 1078.46 2723.44 Q1072.65 2723.44 1069.58 2718.86 Q1066.52 2714.25 1066.52 2705.52 Q1066.52 2696.77 1069.58 2692.19 Q1072.65 2687.58 1078.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1263.12 2718.83 L1279.44 2718.83 L1279.44 2722.77 L1257.49 2722.77 L1257.49 2718.83 Q1260.16 2716.08 1264.74 2711.45 Q1269.35 2706.8 1270.53 2705.45 Q1272.77 2702.93 1273.65 2701.19 Q1274.55 2699.43 1274.55 2697.74 Q1274.55 2694.99 1272.61 2693.25 Q1270.69 2691.52 1267.59 2691.52 Q1265.39 2691.52 1262.93 2692.28 Q1260.5 2693.05 1257.73 2694.6 L1257.73 2689.87 Q1260.55 2688.74 1263 2688.16 Q1265.46 2687.58 1267.49 2687.58 Q1272.86 2687.58 1276.06 2690.27 Q1279.25 2692.95 1279.25 2697.44 Q1279.25 2699.57 1278.44 2701.49 Q1277.66 2703.39 1275.55 2705.99 Q1274.97 2706.66 1271.87 2709.87 Q1268.77 2713.07 1263.12 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1299.25 2691.29 Q1295.64 2691.29 1293.81 2694.85 Q1292.01 2698.39 1292.01 2705.52 Q1292.01 2712.63 1293.81 2716.19 Q1295.64 2719.74 1299.25 2719.74 Q1302.89 2719.74 1304.69 2716.19 Q1306.52 2712.63 1306.52 2705.52 Q1306.52 2698.39 1304.69 2694.85 Q1302.89 2691.29 1299.25 2691.29 M1299.25 2687.58 Q1305.06 2687.58 1308.12 2692.19 Q1311.2 2696.77 1311.2 2705.52 Q1311.2 2714.25 1308.12 2718.86 Q1305.06 2723.44 1299.25 2723.44 Q1293.44 2723.44 1290.36 2718.86 Q1287.31 2714.25 1287.31 2705.52 Q1287.31 2696.77 1290.36 2692.19 Q1293.44 2687.58 1299.25 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1329.42 2691.29 Q1325.8 2691.29 1323.98 2694.85 Q1322.17 2698.39 1322.17 2705.52 Q1322.17 2712.63 1323.98 2716.19 Q1325.8 2719.74 1329.42 2719.74 Q1333.05 2719.74 1334.86 2716.19 Q1336.68 2712.63 1336.68 2705.52 Q1336.68 2698.39 1334.86 2694.85 Q1333.05 2691.29 1329.42 2691.29 M1329.42 2687.58 Q1335.23 2687.58 1338.28 2692.19 Q1341.36 2696.77 1341.36 2705.52 Q1341.36 2714.25 1338.28 2718.86 Q1335.23 2723.44 1329.42 2723.44 Q1323.61 2723.44 1320.53 2718.86 Q1317.47 2714.25 1317.47 2705.52 Q1317.47 2696.77 1320.53 2692.19 Q1323.61 2687.58 1329.42 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1359.58 2691.29 Q1355.97 2691.29 1354.14 2694.85 Q1352.33 2698.39 1352.33 2705.52 Q1352.33 2712.63 1354.14 2716.19 Q1355.97 2719.74 1359.58 2719.74 Q1363.21 2719.74 1365.02 2716.19 Q1366.85 2712.63 1366.85 2705.52 Q1366.85 2698.39 1365.02 2694.85 Q1363.21 2691.29 1359.58 2691.29 M1359.58 2687.58 Q1365.39 2687.58 1368.44 2692.19 Q1371.52 2696.77 1371.52 2705.52 Q1371.52 2714.25 1368.44 2718.86 Q1365.39 2723.44 1359.58 2723.44 Q1353.77 2723.44 1350.69 2718.86 Q1347.63 2714.25 1347.63 2705.52 Q1347.63 2696.77 1350.69 2692.19 Q1353.77 2687.58 1359.58 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2639.15 1352.76,2639.15 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2463.38 1352.76,2463.38 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2287.62 1352.76,2287.62 "/>
-<polyline clip-path="url(#clip216)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2111.85 1352.76,2111.85 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2657.17 159.765,2020.38 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2639.15 178.663,2639.15 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2463.38 178.663,2463.38 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2287.62 178.663,2287.62 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2111.85 178.663,2111.85 "/>
-<path clip-path="url(#clip210)" d="M105.821 2624.94 Q102.209 2624.94 100.381 2628.51 Q98.5752 2632.05 98.5752 2639.18 Q98.5752 2646.29 100.381 2649.85 Q102.209 2653.39 105.821 2653.39 Q109.455 2653.39 111.26 2649.85 Q113.089 2646.29 113.089 2639.18 Q113.089 2632.05 111.26 2628.51 Q109.455 2624.94 105.821 2624.94 M105.821 2621.24 Q111.631 2621.24 114.686 2625.85 Q117.765 2630.43 117.765 2639.18 Q117.765 2647.91 114.686 2652.51 Q111.631 2657.1 105.821 2657.1 Q100.01 2657.1 96.9317 2652.51 Q93.8761 2647.91 93.8761 2639.18 Q93.8761 2630.43 96.9317 2625.85 Q100.01 2621.24 105.821 2621.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M66.4689 2476.73 L74.1077 2476.73 L74.1077 2450.36 L65.7976 2452.03 L65.7976 2447.77 L74.0614 2446.1 L78.7373 2446.1 L78.7373 2476.73 L86.3762 2476.73 L86.3762 2480.66 L66.4689 2480.66 L66.4689 2476.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 2449.18 Q102.209 2449.18 100.381 2452.75 Q98.5752 2456.29 98.5752 2463.42 Q98.5752 2470.52 100.381 2474.09 Q102.209 2477.63 105.821 2477.63 Q109.455 2477.63 111.26 2474.09 Q113.089 2470.52 113.089 2463.42 Q113.089 2456.29 111.26 2452.75 Q109.455 2449.18 105.821 2449.18 M105.821 2445.48 Q111.631 2445.48 114.686 2450.08 Q117.765 2454.67 117.765 2463.42 Q117.765 2472.14 114.686 2476.75 Q111.631 2481.33 105.821 2481.33 Q100.01 2481.33 96.9317 2476.75 Q93.8761 2472.14 93.8761 2463.42 Q93.8761 2454.67 96.9317 2450.08 Q100.01 2445.48 105.821 2445.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M69.6865 2300.96 L86.0058 2300.96 L86.0058 2304.9 L64.0615 2304.9 L64.0615 2300.96 Q66.7235 2298.21 71.3068 2293.58 Q75.9133 2288.93 77.0938 2287.58 Q79.3392 2285.06 80.2188 2283.32 Q81.1216 2281.56 81.1216 2279.87 Q81.1216 2277.12 79.1771 2275.38 Q77.2559 2273.65 74.154 2273.65 Q71.955 2273.65 69.5013 2274.41 Q67.0707 2275.18 64.293 2276.73 L64.293 2272 Q67.117 2270.87 69.5707 2270.29 Q72.0244 2269.71 74.0614 2269.71 Q79.4318 2269.71 82.6262 2272.4 Q85.8206 2275.08 85.8206 2279.57 Q85.8206 2281.7 85.0105 2283.62 Q84.2234 2285.52 82.1169 2288.12 Q81.5382 2288.79 78.4364 2292 Q75.3346 2295.2 69.6865 2300.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 2273.42 Q102.209 2273.42 100.381 2276.98 Q98.5752 2280.52 98.5752 2287.65 Q98.5752 2294.76 100.381 2298.32 Q102.209 2301.87 105.821 2301.87 Q109.455 2301.87 111.26 2298.32 Q113.089 2294.76 113.089 2287.65 Q113.089 2280.52 111.26 2276.98 Q109.455 2273.42 105.821 2273.42 M105.821 2269.71 Q111.631 2269.71 114.686 2274.32 Q117.765 2278.9 117.765 2287.65 Q117.765 2296.38 114.686 2300.99 Q111.631 2305.57 105.821 2305.57 Q100.01 2305.57 96.9317 2300.99 Q93.8761 2296.38 93.8761 2287.65 Q93.8761 2278.9 96.9317 2274.32 Q100.01 2269.71 105.821 2269.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M79.8253 2110.5 Q83.1818 2111.22 85.0567 2113.49 Q86.9549 2115.75 86.9549 2119.09 Q86.9549 2124.2 83.4364 2127 Q79.9179 2129.81 73.4364 2129.81 Q71.2605 2129.81 68.9457 2129.37 Q66.6541 2128.95 64.2004 2128.09 L64.2004 2123.58 Q66.1448 2124.71 68.4596 2125.29 Q70.7744 2125.87 73.2976 2125.87 Q77.6957 2125.87 79.9873 2124.13 Q82.3021 2122.4 82.3021 2119.09 Q82.3021 2116.03 80.1494 2114.32 Q78.0197 2112.58 74.2003 2112.58 L70.1726 2112.58 L70.1726 2108.74 L74.3855 2108.74 Q77.8346 2108.74 79.6633 2107.37 Q81.492 2105.99 81.492 2103.39 Q81.492 2100.73 79.5938 2099.32 Q77.7188 2097.88 74.2003 2097.88 Q72.279 2097.88 70.08 2098.3 Q67.8809 2098.72 65.242 2099.6 L65.242 2095.43 Q67.9041 2094.69 70.2189 2094.32 Q72.5568 2093.95 74.617 2093.95 Q79.941 2093.95 83.0429 2096.38 Q86.1447 2098.79 86.1447 2102.91 Q86.1447 2105.78 84.5012 2107.77 Q82.8577 2109.74 79.8253 2110.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M105.821 2097.65 Q102.209 2097.65 100.381 2101.22 Q98.5752 2104.76 98.5752 2111.89 Q98.5752 2119 100.381 2122.56 Q102.209 2126.1 105.821 2126.1 Q109.455 2126.1 111.26 2122.56 Q113.089 2119 113.089 2111.89 Q113.089 2104.76 111.26 2101.22 Q109.455 2097.65 105.821 2097.65 M105.821 2093.95 Q111.631 2093.95 114.686 2098.56 Q117.765 2103.14 117.765 2111.89 Q117.765 2120.62 114.686 2125.22 Q111.631 2129.81 105.821 2129.81 Q100.01 2129.81 96.9317 2125.22 Q93.8761 2120.62 93.8761 2111.89 Q93.8761 2103.14 96.9317 2098.56 Q100.01 2093.95 105.821 2093.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M469.436 1878.76 L477.619 1878.76 L477.619 1932.36 L507.069 1932.36 L507.069 1939.24 L469.436 1939.24 L469.436 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M535.952 1916.44 Q526.918 1916.44 523.435 1918.5 Q519.951 1920.57 519.951 1925.55 Q519.951 1929.52 522.543 1931.87 Q525.177 1934.18 529.673 1934.18 Q535.871 1934.18 539.598 1929.8 Q543.365 1925.39 543.365 1918.1 L543.365 1916.44 L535.952 1916.44 M550.819 1913.36 L550.819 1939.24 L543.365 1939.24 L543.365 1932.36 Q540.813 1936.49 537.005 1938.47 Q533.197 1940.42 527.688 1940.42 Q520.721 1940.42 516.589 1936.53 Q512.497 1932.6 512.497 1926.04 Q512.497 1918.38 517.601 1914.49 Q522.746 1910.6 532.914 1910.6 L543.365 1910.6 L543.365 1909.87 Q543.365 1904.73 539.962 1901.93 Q536.6 1899.1 530.483 1899.1 Q526.594 1899.1 522.908 1900.03 Q519.222 1900.96 515.819 1902.83 L515.819 1895.94 Q519.91 1894.36 523.759 1893.59 Q527.607 1892.78 531.253 1892.78 Q541.097 1892.78 545.958 1897.88 Q550.819 1902.99 550.819 1913.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M573.544 1880.99 L573.544 1893.87 L588.897 1893.87 L588.897 1899.67 L573.544 1899.67 L573.544 1924.29 Q573.544 1929.84 575.043 1931.42 Q576.583 1933 581.241 1933 L588.897 1933 L588.897 1939.24 L581.241 1939.24 Q572.613 1939.24 569.331 1936.04 Q566.05 1932.8 566.05 1924.29 L566.05 1899.67 L560.581 1899.67 L560.581 1893.87 L566.05 1893.87 L566.05 1880.99 L573.544 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M598.7 1893.87 L606.154 1893.87 L606.154 1939.24 L598.7 1939.24 L598.7 1893.87 M598.7 1876.21 L606.154 1876.21 L606.154 1885.65 L598.7 1885.65 L598.7 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M629.123 1880.99 L629.123 1893.87 L644.476 1893.87 L644.476 1899.67 L629.123 1899.67 L629.123 1924.29 Q629.123 1929.84 630.622 1931.42 Q632.161 1933 636.819 1933 L644.476 1933 L644.476 1939.24 L636.819 1939.24 Q628.191 1939.24 624.91 1936.04 Q621.629 1932.8 621.629 1924.29 L621.629 1899.67 L616.16 1899.67 L616.16 1893.87 L621.629 1893.87 L621.629 1880.99 L629.123 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M653.509 1921.34 L653.509 1893.87 L660.963 1893.87 L660.963 1921.05 Q660.963 1927.5 663.474 1930.74 Q665.986 1933.94 671.009 1933.94 Q677.045 1933.94 680.529 1930.09 Q684.053 1926.24 684.053 1919.6 L684.053 1893.87 L691.507 1893.87 L691.507 1939.24 L684.053 1939.24 L684.053 1932.28 Q681.339 1936.41 677.734 1938.43 Q674.169 1940.42 669.429 1940.42 Q661.611 1940.42 657.56 1935.56 Q653.509 1930.7 653.509 1921.34 M672.265 1892.78 L672.265 1892.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M736.715 1900.76 L736.715 1876.21 L744.168 1876.21 L744.168 1939.24 L736.715 1939.24 L736.715 1932.44 Q734.365 1936.49 730.76 1938.47 Q727.195 1940.42 722.172 1940.42 Q713.949 1940.42 708.764 1933.85 Q703.619 1927.29 703.619 1916.6 Q703.619 1905.9 708.764 1899.34 Q713.949 1892.78 722.172 1892.78 Q727.195 1892.78 730.76 1894.76 Q734.365 1896.71 736.715 1900.76 M711.316 1916.6 Q711.316 1924.82 714.678 1929.52 Q718.081 1934.18 723.995 1934.18 Q729.909 1934.18 733.312 1929.52 Q736.715 1924.82 736.715 1916.6 Q736.715 1908.37 733.312 1903.72 Q729.909 1899.02 723.995 1899.02 Q718.081 1899.02 714.678 1903.72 Q711.316 1908.37 711.316 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M798.329 1914.69 L798.329 1918.34 L764.058 1918.34 Q764.545 1926.04 768.676 1930.09 Q772.849 1934.1 780.262 1934.1 Q784.556 1934.1 788.566 1933.04 Q792.617 1931.99 796.587 1929.89 L796.587 1936.93 Q792.577 1938.64 788.364 1939.53 Q784.151 1940.42 779.816 1940.42 Q768.96 1940.42 762.6 1934.1 Q756.281 1927.78 756.281 1917 Q756.281 1905.86 762.276 1899.34 Q768.312 1892.78 778.52 1892.78 Q787.675 1892.78 792.982 1898.69 Q798.329 1904.57 798.329 1914.69 M790.875 1912.51 Q790.794 1906.39 787.432 1902.74 Q784.11 1899.1 778.601 1899.1 Q772.363 1899.1 768.595 1902.62 Q764.869 1906.15 764.301 1912.55 L790.875 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M854.839 1876.29 Q849.411 1885.61 846.778 1894.72 Q844.145 1903.84 844.145 1913.2 Q844.145 1922.55 846.778 1931.75 Q849.451 1940.9 854.839 1950.18 L848.358 1950.18 Q842.281 1940.66 839.243 1931.46 Q836.246 1922.27 836.246 1913.2 Q836.246 1904.16 839.243 1895.01 Q842.241 1885.85 848.358 1876.29 L854.839 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M899.156 1900.76 L899.156 1876.21 L906.61 1876.21 L906.61 1939.24 L899.156 1939.24 L899.156 1932.44 Q896.807 1936.49 893.201 1938.47 Q889.636 1940.42 884.613 1940.42 Q876.39 1940.42 871.205 1933.85 Q866.06 1927.29 866.06 1916.6 Q866.06 1905.9 871.205 1899.34 Q876.39 1892.78 884.613 1892.78 Q889.636 1892.78 893.201 1894.76 Q896.807 1896.71 899.156 1900.76 M873.757 1916.6 Q873.757 1924.82 877.119 1929.52 Q880.522 1934.18 886.436 1934.18 Q892.351 1934.18 895.753 1929.52 Q899.156 1924.82 899.156 1916.6 Q899.156 1908.37 895.753 1903.72 Q892.351 1899.02 886.436 1899.02 Q880.522 1899.02 877.119 1903.72 Q873.757 1908.37 873.757 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M960.77 1914.69 L960.77 1918.34 L926.5 1918.34 Q926.986 1926.04 931.118 1930.09 Q935.29 1934.1 942.703 1934.1 Q946.997 1934.1 951.008 1933.04 Q955.059 1931.99 959.028 1929.89 L959.028 1936.93 Q955.018 1938.64 950.805 1939.53 Q946.592 1940.42 942.258 1940.42 Q931.401 1940.42 925.041 1934.1 Q918.722 1927.78 918.722 1917 Q918.722 1905.86 924.717 1899.34 Q930.753 1892.78 940.961 1892.78 Q950.116 1892.78 955.423 1898.69 Q960.77 1904.57 960.77 1914.69 M953.317 1912.51 Q953.236 1906.39 949.873 1902.74 Q946.552 1899.1 941.042 1899.1 Q934.804 1899.1 931.037 1902.62 Q927.31 1906.15 926.743 1912.55 L953.317 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1002.86 1916.03 Q1002.86 1907.93 999.497 1903.47 Q996.175 1899.02 990.139 1899.02 Q984.144 1899.02 980.782 1903.47 Q977.46 1907.93 977.46 1916.03 Q977.46 1924.09 980.782 1928.55 Q984.144 1933 990.139 1933 Q996.175 1933 999.497 1928.55 Q1002.86 1924.09 1002.86 1916.03 M1010.31 1933.61 Q1010.31 1945.2 1005.17 1950.83 Q1000.02 1956.5 989.41 1956.5 Q985.481 1956.5 981.997 1955.89 Q978.513 1955.32 975.232 1954.11 L975.232 1946.86 Q978.513 1948.64 981.713 1949.49 Q984.914 1950.34 988.235 1950.34 Q995.568 1950.34 999.213 1946.49 Q1002.86 1942.69 1002.86 1934.95 L1002.86 1931.26 Q1000.55 1935.27 996.945 1937.26 Q993.34 1939.24 988.316 1939.24 Q979.972 1939.24 974.867 1932.88 Q969.763 1926.52 969.763 1916.03 Q969.763 1905.5 974.867 1899.14 Q979.972 1892.78 988.316 1892.78 Q993.34 1892.78 996.945 1894.76 Q1000.55 1896.75 1002.86 1900.76 L1002.86 1893.87 L1010.31 1893.87 L1010.31 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1024.49 1876.29 L1030.97 1876.29 Q1037.05 1885.85 1040.05 1895.01 Q1043.08 1904.16 1043.08 1913.2 Q1043.08 1922.27 1040.05 1931.46 Q1037.05 1940.66 1030.97 1950.18 L1024.49 1950.18 Q1029.88 1940.9 1032.51 1931.75 Q1035.19 1922.55 1035.19 1913.2 Q1035.19 1903.84 1032.51 1894.72 Q1029.88 1885.61 1024.49 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip216)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,2639.15 195.771,2639.15 198.013,2639.14 200.255,2639.14 202.497,2639.14 204.739,2639.13 206.981,2639.12 209.223,2639.11 211.464,2639.1 213.706,2639.09 215.948,2639.07 218.19,2639.05 220.432,2639.03 222.674,2639.01 224.916,2638.98 227.158,2638.95 229.4,2638.91 231.642,2638.87 233.884,2638.83 236.126,2638.78 238.368,2638.73 240.61,2638.67 242.852,2638.6 245.094,2638.53 247.336,2638.46 249.578,2638.37 251.82,2638.28 254.062,2638.19 256.304,2638.08 258.546,2637.97 260.788,2637.84 263.03,2637.71 265.271,2637.57 267.513,2637.41 269.755,2637.25 271.997,2637.07 274.239,2636.88 276.481,2636.68 278.723,2636.47 280.965,2636.24 283.207,2636 285.449,2635.75 287.691,2635.48 289.933,2635.2 292.175,2634.9 294.417,2634.59 296.659,2634.26 298.901,2633.92 301.143,2633.56 303.385,2633.19 305.627,2632.8 307.869,2632.39 310.111,2631.97 312.353,2631.54 314.595,2631.09 316.837,2630.62 319.078,2630.15 321.32,2629.65 323.562,2629.15 325.804,2628.63 328.046,2628.1 330.288,2627.55 332.53,2626.99 334.772,2626.43 337.014,2625.84 339.256,2625.25 341.498,2624.65 343.74,2624.04 345.982,2623.42 348.224,2622.78 350.466,2622.14 352.708,2621.49 354.95,2620.83 357.192,2620.16 359.434,2619.49 361.676,2618.8 363.918,2618.11 366.16,2617.41 368.402,2616.71 370.644,2616 372.885,2615.28 375.127,2614.55 377.369,2613.82 379.611,2613.09 381.853,2612.34 384.095,2611.6 386.337,2610.84 388.579,2610.08 390.821,2609.32 393.063,2608.55 395.305,2607.78 397.547,2607 399.789,2606.21 402.031,2605.43 404.273,2604.63 406.515,2603.83 408.757,2603.03 410.999,2602.23 413.241,2601.41 415.483,2600.6 417.725,2599.78 419.967,2598.95 422.209,2598.12 424.451,2597.29 426.692,2596.45 428.934,2595.61 431.176,2594.76 433.418,2593.91 435.66,2593.05 437.902,2592.19 440.144,2591.32 442.386,2590.45 444.628,2589.57 446.87,2588.69 449.112,2587.8 451.354,2586.9 453.596,2586 455.838,2585.1 458.08,2584.19 460.322,2583.27 462.564,2582.34 464.806,2581.41 467.048,2580.48 469.29,2579.53 471.532,2578.58 473.774,2577.62 476.016,2576.65 478.258,2575.68 480.5,2574.69 482.741,2573.7 484.983,2572.7 487.225,2571.69 489.467,2570.67 491.709,2569.64 493.951,2568.6 496.193,2567.56 498.435,2566.5 500.677,2565.43 502.919,2564.35 505.161,2563.26 507.403,2562.16 509.645,2561.05 511.887,2559.93 514.129,2558.8 516.371,2557.66 518.613,2556.51 520.855,2555.35 523.097,2554.17 525.339,2552.99 527.581,2551.8 529.823,2550.59 532.065,2549.38 534.307,2548.16 536.548,2546.92 538.79,2545.68 541.032,2544.43 543.274,2543.17 545.516,2541.9 547.758,2540.63 550,2539.34 552.242,2538.05 554.484,2536.75 556.726,2535.44 558.968,2534.13 561.21,2532.81 563.452,2531.48 565.694,2530.15 567.936,2528.81 570.178,2527.46 572.42,2526.11 574.662,2524.75 576.904,2523.39 579.146,2522.02 581.388,2520.64 583.63,2519.26 585.872,2517.88 588.114,2516.49 590.355,2515.09 592.597,2513.69 594.839,2512.28 597.081,2510.87 599.323,2509.46 601.565,2508.04 603.807,2506.61 606.049,2505.18 608.291,2503.75 610.533,2502.31 612.775,2500.86 615.017,2499.41 617.259,2497.96 619.501,2496.49 621.743,2495.03 623.985,2493.56 626.227,2492.08 628.469,2490.6 630.711,2489.11 632.953,2487.62 635.195,2486.12 637.437,2484.62 639.679,2483.11 641.921,2481.59 644.162,2480.07 646.404,2478.54 648.646,2477.01 650.888,2475.47 653.13,2473.92 655.372,2472.37 657.614,2470.81 659.856,2469.24 662.098,2467.67 664.34,2466.09 666.582,2464.51 668.824,2462.91 671.066,2461.32 673.308,2459.71 675.55,2458.1 677.792,2456.49 680.034,2454.87 682.276,2453.24 684.518,2451.6 686.76,2449.97 689.002,2448.32 691.244,2446.67 693.486,2445.02 695.728,2443.36 697.969,2441.69 700.211,2440.02 702.453,2438.35 704.695,2436.67 706.937,2434.99 709.179,2433.3 711.421,2431.61 713.663,2429.92 715.905,2428.22 718.147,2426.52 720.389,2424.81 722.631,2423.1 724.873,2421.39 727.115,2419.67 729.357,2417.96 731.599,2416.23 733.841,2414.51 736.083,2412.78 738.325,2411.05 740.567,2409.32 742.809,2407.58 745.051,2405.84 747.293,2404.1 749.535,2402.36 751.776,2400.61 754.018,2398.86 756.26,2397.11 758.502,2395.35 760.744,2393.59 762.986,2391.83 765.228,2390.07 767.47,2388.3 769.712,2386.53 771.954,2384.76 774.196,2382.99 776.438,2381.21 778.68,2379.43 780.922,2377.65 783.164,2375.86 785.406,2374.08 787.648,2372.29 789.89,2370.5 792.132,2368.7 794.374,2366.9 796.616,2365.11 798.858,2363.3 801.1,2361.5 803.342,2359.7 805.583,2357.89 807.825,2356.08 810.067,2354.27 812.309,2352.46 814.551,2350.64 816.793,2348.83 819.035,2347.01 821.277,2345.19 823.519,2343.37 825.761,2341.55 828.003,2339.73 830.245,2337.91 832.487,2336.08 834.729,2334.26 836.971,2332.43 839.213,2330.61 841.455,2328.78 843.697,2326.96 845.939,2325.13 848.181,2323.3 850.423,2321.47 852.665,2319.65 854.907,2317.82 857.149,2315.99 859.391,2314.16 861.632,2312.34 863.874,2310.51 866.116,2308.68 868.358,2306.86 870.6,2305.03 872.842,2303.2 875.084,2301.38 877.326,2299.55 879.568,2297.73 881.81,2295.9 884.052,2294.08 886.294,2292.26 888.536,2290.44 890.778,2288.61 893.02,2286.79 895.262,2284.98 897.504,2283.16 899.746,2281.34 901.988,2279.52 904.23,2277.71 906.472,2275.9 908.714,2274.09 910.956,2272.28 913.198,2270.47 915.439,2268.66 917.681,2266.86 919.923,2265.05 922.165,2263.25 924.407,2261.46 926.649,2259.66 928.891,2257.86 931.133,2256.07 933.375,2254.28 935.617,2252.49 937.859,2250.71 940.101,2248.93 942.343,2247.15 944.585,2245.37 946.827,2243.6 949.069,2241.82 951.311,2240.06 953.553,2238.29 955.795,2236.53 958.037,2234.77 960.279,2233.01 962.521,2231.26 964.763,2229.51 967.005,2227.76 969.246,2226.02 971.488,2224.28 973.73,2222.54 975.972,2220.81 978.214,2219.08 980.456,2217.35 982.698,2215.63 984.94,2213.91 987.182,2212.2 989.424,2210.49 991.666,2208.78 993.908,2207.08 996.15,2205.38 998.392,2203.69 1000.63,2202 1002.88,2200.31 1005.12,2198.63 1007.36,2196.96 1009.6,2195.29 1011.84,2193.62 1014.09,2191.96 1016.33,2190.31 1018.57,2188.66 1020.81,2187.01 1023.05,2185.37 1025.3,2183.74 1027.54,2182.11 1029.78,2180.48 1032.02,2178.87 1034.26,2177.25 1036.51,2175.65 1038.75,2174.05 1040.99,2172.45 1043.23,2170.86 1045.47,2169.28 1047.72,2167.7 1049.96,2166.13 1052.2,2164.57 1054.44,2163.01 1056.68,2161.46 1058.92,2159.91 1061.17,2158.37 1063.41,2156.84 1065.65,2155.31 1067.89,2153.79 1070.13,2152.28 1072.38,2150.77 1074.62,2149.27 1076.86,2147.78 1079.1,2146.29 1081.34,2144.81 1083.59,2143.34 1085.83,2141.88 1088.07,2140.42 1090.31,2138.97 1092.55,2137.52 1094.8,2136.08 1097.04,2134.65 1099.28,2133.23 1101.52,2131.82 1103.76,2130.41 1106.01,2129.01 1108.25,2127.62 1110.49,2126.24 1112.73,2124.86 1114.97,2123.49 1117.22,2122.13 1119.46,2120.78 1121.7,2119.44 1123.94,2118.11 1126.18,2116.78 1128.43,2115.46 1130.67,2114.15 1132.91,2112.85 1135.15,2111.56 1137.39,2110.28 1139.64,2109 1141.88,2107.74 1144.12,2106.48 1146.36,2105.23 1148.6,2103.99 1150.85,2102.76 1153.09,2101.54 1155.33,2100.33 1157.57,2099.12 1159.81,2097.93 1162.05,2096.75 1164.3,2095.57 1166.54,2094.4 1168.78,2093.25 1171.02,2092.1 1173.26,2090.96 1175.51,2089.83 1177.75,2088.71 1179.99,2087.6 1182.23,2086.5 1184.47,2085.4 1186.72,2084.32 1188.96,2083.25 1191.2,2082.18 1193.44,2081.13 1195.68,2080.09 1197.93,2079.05 1200.17,2078.03 1202.41,2077.02 1204.65,2076.01 1206.89,2075.02 1209.14,2074.03 1211.38,2073.06 1213.62,2072.09 1215.86,2071.14 1218.1,2070.2 1220.35,2069.26 1222.59,2068.34 1224.83,2067.43 1227.07,2066.53 1229.31,2065.64 1231.56,2064.76 1233.8,2063.88 1236.04,2063.02 1238.28,2062.18 1240.52,2061.34 1242.77,2060.51 1245.01,2059.69 1247.25,2058.88 1249.49,2058.09 1251.73,2057.3 1253.98,2056.53 1256.22,2055.76 1258.46,2055.01 1260.7,2054.26 1262.94,2053.53 1265.19,2052.8 1267.43,2052.09 1269.67,2051.39 1271.91,2050.69 1274.15,2050.01 1276.39,2049.34 1278.64,2048.67 1280.88,2048.02 1283.12,2047.37 1285.36,2046.74 1287.6,2046.11 1289.85,2045.5 1292.09,2044.89 1294.33,2044.3 1296.57,2043.71 1298.81,2043.14 1301.06,2042.57 1303.3,2042.01 1305.54,2041.47 1307.78,2040.93 1310.02,2040.4 1312.27,2039.89 1314.51,2039.38 1316.75,2038.89 1318.99,2038.4 "/>
-<path clip-path="url(#clip210)" d="M1559.76 2657.17 L2752.76 2657.17 L2752.76 2020.38 L1559.76 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="193.529,2657.17 193.529,2020.38 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="473.774,2657.17 473.774,2020.38 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="754.018,2657.17 754.018,2020.38 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1034.26,2657.17 1034.26,2020.38 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1314.51,2657.17 1314.51,2020.38 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2657.17 1352.76,2657.17 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="193.529,2657.17 193.529,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="473.774,2657.17 473.774,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="754.018,2657.17 754.018,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1034.26,2657.17 1034.26,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1314.51,2657.17 1314.51,2638.27 "/>
+<path clip-path="url(#clip030)" d="M193.529 2691.29 Q189.918 2691.29 188.089 2694.85 Q186.283 2698.39 186.283 2705.52 Q186.283 2712.63 188.089 2716.19 Q189.918 2719.74 193.529 2719.74 Q197.163 2719.74 198.969 2716.19 Q200.797 2712.63 200.797 2705.52 Q200.797 2698.39 198.969 2694.85 Q197.163 2691.29 193.529 2691.29 M193.529 2687.58 Q199.339 2687.58 202.394 2692.19 Q205.473 2696.77 205.473 2705.52 Q205.473 2714.25 202.394 2718.86 Q199.339 2723.44 193.529 2723.44 Q187.719 2723.44 184.64 2718.86 Q181.584 2714.25 181.584 2705.52 Q181.584 2696.77 184.64 2692.19 Q187.719 2687.58 193.529 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M433.392 2688.21 L451.748 2688.21 L451.748 2692.14 L437.674 2692.14 L437.674 2700.62 Q438.693 2700.27 439.711 2700.11 Q440.73 2699.92 441.748 2699.92 Q447.535 2699.92 450.915 2703.09 Q454.295 2706.26 454.295 2711.68 Q454.295 2717.26 450.822 2720.36 Q447.35 2723.44 441.031 2723.44 Q438.855 2723.44 436.586 2723.07 Q434.341 2722.7 431.934 2721.96 L431.934 2717.26 Q434.017 2718.39 436.239 2718.95 Q438.461 2719.5 440.938 2719.5 Q444.943 2719.5 447.281 2717.4 Q449.619 2715.29 449.619 2711.68 Q449.619 2708.07 447.281 2705.96 Q444.943 2703.86 440.938 2703.86 Q439.063 2703.86 437.188 2704.27 Q435.336 2704.69 433.392 2705.57 L433.392 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M473.507 2691.29 Q469.896 2691.29 468.068 2694.85 Q466.262 2698.39 466.262 2705.52 Q466.262 2712.63 468.068 2716.19 Q469.896 2719.74 473.507 2719.74 Q477.142 2719.74 478.947 2716.19 Q480.776 2712.63 480.776 2705.52 Q480.776 2698.39 478.947 2694.85 Q477.142 2691.29 473.507 2691.29 M473.507 2687.58 Q479.318 2687.58 482.373 2692.19 Q485.452 2696.77 485.452 2705.52 Q485.452 2714.25 482.373 2718.86 Q479.318 2723.44 473.507 2723.44 Q467.697 2723.44 464.619 2718.86 Q461.563 2714.25 461.563 2705.52 Q461.563 2696.77 464.619 2692.19 Q467.697 2687.58 473.507 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M503.669 2691.29 Q500.058 2691.29 498.23 2694.85 Q496.424 2698.39 496.424 2705.52 Q496.424 2712.63 498.23 2716.19 Q500.058 2719.74 503.669 2719.74 Q507.304 2719.74 509.109 2716.19 Q510.938 2712.63 510.938 2705.52 Q510.938 2698.39 509.109 2694.85 Q507.304 2691.29 503.669 2691.29 M503.669 2687.58 Q509.479 2687.58 512.535 2692.19 Q515.614 2696.77 515.614 2705.52 Q515.614 2714.25 512.535 2718.86 Q509.479 2723.44 503.669 2723.44 Q497.859 2723.44 494.78 2718.86 Q491.725 2714.25 491.725 2705.52 Q491.725 2696.77 494.78 2692.19 Q497.859 2687.58 503.669 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M698.544 2718.83 L706.183 2718.83 L706.183 2692.47 L697.873 2694.13 L697.873 2689.87 L706.137 2688.21 L710.813 2688.21 L710.813 2718.83 L718.452 2718.83 L718.452 2722.77 L698.544 2722.77 L698.544 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M737.896 2691.29 Q734.285 2691.29 732.456 2694.85 Q730.651 2698.39 730.651 2705.52 Q730.651 2712.63 732.456 2716.19 Q734.285 2719.74 737.896 2719.74 Q741.53 2719.74 743.336 2716.19 Q745.164 2712.63 745.164 2705.52 Q745.164 2698.39 743.336 2694.85 Q741.53 2691.29 737.896 2691.29 M737.896 2687.58 Q743.706 2687.58 746.762 2692.19 Q749.84 2696.77 749.84 2705.52 Q749.84 2714.25 746.762 2718.86 Q743.706 2723.44 737.896 2723.44 Q732.086 2723.44 729.007 2718.86 Q725.951 2714.25 725.951 2705.52 Q725.951 2696.77 729.007 2692.19 Q732.086 2687.58 737.896 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M768.058 2691.29 Q764.447 2691.29 762.618 2694.85 Q760.812 2698.39 760.812 2705.52 Q760.812 2712.63 762.618 2716.19 Q764.447 2719.74 768.058 2719.74 Q771.692 2719.74 773.498 2716.19 Q775.326 2712.63 775.326 2705.52 Q775.326 2698.39 773.498 2694.85 Q771.692 2691.29 768.058 2691.29 M768.058 2687.58 Q773.868 2687.58 776.923 2692.19 Q780.002 2696.77 780.002 2705.52 Q780.002 2714.25 776.923 2718.86 Q773.868 2723.44 768.058 2723.44 Q762.248 2723.44 759.169 2718.86 Q756.113 2714.25 756.113 2705.52 Q756.113 2696.77 759.169 2692.19 Q762.248 2687.58 768.058 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M798.22 2691.29 Q794.609 2691.29 792.78 2694.85 Q790.974 2698.39 790.974 2705.52 Q790.974 2712.63 792.78 2716.19 Q794.609 2719.74 798.22 2719.74 Q801.854 2719.74 803.659 2716.19 Q805.488 2712.63 805.488 2705.52 Q805.488 2698.39 803.659 2694.85 Q801.854 2691.29 798.22 2691.29 M798.22 2687.58 Q804.03 2687.58 807.085 2692.19 Q810.164 2696.77 810.164 2705.52 Q810.164 2714.25 807.085 2718.86 Q804.03 2723.44 798.22 2723.44 Q792.409 2723.44 789.331 2718.86 Q786.275 2714.25 786.275 2705.52 Q786.275 2696.77 789.331 2692.19 Q792.409 2687.58 798.22 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M978.789 2718.83 L986.428 2718.83 L986.428 2692.47 L978.118 2694.13 L978.118 2689.87 L986.382 2688.21 L991.057 2688.21 L991.057 2718.83 L998.696 2718.83 L998.696 2722.77 L978.789 2722.77 L978.789 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1008.19 2688.21 L1026.54 2688.21 L1026.54 2692.14 L1012.47 2692.14 L1012.47 2700.62 Q1013.49 2700.27 1014.51 2700.11 Q1015.52 2699.92 1016.54 2699.92 Q1022.33 2699.92 1025.71 2703.09 Q1029.09 2706.26 1029.09 2711.68 Q1029.09 2717.26 1025.62 2720.36 Q1022.15 2723.44 1015.83 2723.44 Q1013.65 2723.44 1011.38 2723.07 Q1009.14 2722.7 1006.73 2721.96 L1006.73 2717.26 Q1008.81 2718.39 1011.03 2718.95 Q1013.26 2719.5 1015.73 2719.5 Q1019.74 2719.5 1022.08 2717.4 Q1024.41 2715.29 1024.41 2711.68 Q1024.41 2708.07 1022.08 2705.96 Q1019.74 2703.86 1015.73 2703.86 Q1013.86 2703.86 1011.98 2704.27 Q1010.13 2704.69 1008.19 2705.57 L1008.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1048.3 2691.29 Q1044.69 2691.29 1042.86 2694.85 Q1041.06 2698.39 1041.06 2705.52 Q1041.06 2712.63 1042.86 2716.19 Q1044.69 2719.74 1048.3 2719.74 Q1051.94 2719.74 1053.74 2716.19 Q1055.57 2712.63 1055.57 2705.52 Q1055.57 2698.39 1053.74 2694.85 Q1051.94 2691.29 1048.3 2691.29 M1048.3 2687.58 Q1054.11 2687.58 1057.17 2692.19 Q1060.25 2696.77 1060.25 2705.52 Q1060.25 2714.25 1057.17 2718.86 Q1054.11 2723.44 1048.3 2723.44 Q1042.49 2723.44 1039.41 2718.86 Q1036.36 2714.25 1036.36 2705.52 Q1036.36 2696.77 1039.41 2692.19 Q1042.49 2687.58 1048.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1078.46 2691.29 Q1074.85 2691.29 1073.02 2694.85 Q1071.22 2698.39 1071.22 2705.52 Q1071.22 2712.63 1073.02 2716.19 Q1074.85 2719.74 1078.46 2719.74 Q1082.1 2719.74 1083.9 2716.19 Q1085.73 2712.63 1085.73 2705.52 Q1085.73 2698.39 1083.9 2694.85 Q1082.1 2691.29 1078.46 2691.29 M1078.46 2687.58 Q1084.27 2687.58 1087.33 2692.19 Q1090.41 2696.77 1090.41 2705.52 Q1090.41 2714.25 1087.33 2718.86 Q1084.27 2723.44 1078.46 2723.44 Q1072.65 2723.44 1069.58 2718.86 Q1066.52 2714.25 1066.52 2705.52 Q1066.52 2696.77 1069.58 2692.19 Q1072.65 2687.58 1078.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1263.12 2718.83 L1279.44 2718.83 L1279.44 2722.77 L1257.49 2722.77 L1257.49 2718.83 Q1260.16 2716.08 1264.74 2711.45 Q1269.35 2706.8 1270.53 2705.45 Q1272.77 2702.93 1273.65 2701.19 Q1274.55 2699.43 1274.55 2697.74 Q1274.55 2694.99 1272.61 2693.25 Q1270.69 2691.52 1267.59 2691.52 Q1265.39 2691.52 1262.93 2692.28 Q1260.5 2693.05 1257.73 2694.6 L1257.73 2689.87 Q1260.55 2688.74 1263 2688.16 Q1265.46 2687.58 1267.49 2687.58 Q1272.86 2687.58 1276.06 2690.27 Q1279.25 2692.95 1279.25 2697.44 Q1279.25 2699.57 1278.44 2701.49 Q1277.66 2703.39 1275.55 2705.99 Q1274.97 2706.66 1271.87 2709.87 Q1268.77 2713.07 1263.12 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1299.25 2691.29 Q1295.64 2691.29 1293.81 2694.85 Q1292.01 2698.39 1292.01 2705.52 Q1292.01 2712.63 1293.81 2716.19 Q1295.64 2719.74 1299.25 2719.74 Q1302.89 2719.74 1304.69 2716.19 Q1306.52 2712.63 1306.52 2705.52 Q1306.52 2698.39 1304.69 2694.85 Q1302.89 2691.29 1299.25 2691.29 M1299.25 2687.58 Q1305.06 2687.58 1308.12 2692.19 Q1311.2 2696.77 1311.2 2705.52 Q1311.2 2714.25 1308.12 2718.86 Q1305.06 2723.44 1299.25 2723.44 Q1293.44 2723.44 1290.36 2718.86 Q1287.31 2714.25 1287.31 2705.52 Q1287.31 2696.77 1290.36 2692.19 Q1293.44 2687.58 1299.25 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1329.42 2691.29 Q1325.8 2691.29 1323.98 2694.85 Q1322.17 2698.39 1322.17 2705.52 Q1322.17 2712.63 1323.98 2716.19 Q1325.8 2719.74 1329.42 2719.74 Q1333.05 2719.74 1334.86 2716.19 Q1336.68 2712.63 1336.68 2705.52 Q1336.68 2698.39 1334.86 2694.85 Q1333.05 2691.29 1329.42 2691.29 M1329.42 2687.58 Q1335.23 2687.58 1338.28 2692.19 Q1341.36 2696.77 1341.36 2705.52 Q1341.36 2714.25 1338.28 2718.86 Q1335.23 2723.44 1329.42 2723.44 Q1323.61 2723.44 1320.53 2718.86 Q1317.47 2714.25 1317.47 2705.52 Q1317.47 2696.77 1320.53 2692.19 Q1323.61 2687.58 1329.42 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1359.58 2691.29 Q1355.97 2691.29 1354.14 2694.85 Q1352.33 2698.39 1352.33 2705.52 Q1352.33 2712.63 1354.14 2716.19 Q1355.97 2719.74 1359.58 2719.74 Q1363.21 2719.74 1365.02 2716.19 Q1366.85 2712.63 1366.85 2705.52 Q1366.85 2698.39 1365.02 2694.85 Q1363.21 2691.29 1359.58 2691.29 M1359.58 2687.58 Q1365.39 2687.58 1368.44 2692.19 Q1371.52 2696.77 1371.52 2705.52 Q1371.52 2714.25 1368.44 2718.86 Q1365.39 2723.44 1359.58 2723.44 Q1353.77 2723.44 1350.69 2718.86 Q1347.63 2714.25 1347.63 2705.52 Q1347.63 2696.77 1350.69 2692.19 Q1353.77 2687.58 1359.58 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2639.15 1352.76,2639.15 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2463.38 1352.76,2463.38 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2287.62 1352.76,2287.62 "/>
+<polyline clip-path="url(#clip036)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="159.765,2111.85 1352.76,2111.85 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2657.17 159.765,2020.38 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2639.15 178.663,2639.15 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2463.38 178.663,2463.38 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2287.62 178.663,2287.62 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="159.765,2111.85 178.663,2111.85 "/>
+<path clip-path="url(#clip030)" d="M105.821 2624.94 Q102.209 2624.94 100.381 2628.51 Q98.5752 2632.05 98.5752 2639.18 Q98.5752 2646.29 100.381 2649.85 Q102.209 2653.39 105.821 2653.39 Q109.455 2653.39 111.26 2649.85 Q113.089 2646.29 113.089 2639.18 Q113.089 2632.05 111.26 2628.51 Q109.455 2624.94 105.821 2624.94 M105.821 2621.24 Q111.631 2621.24 114.686 2625.85 Q117.765 2630.43 117.765 2639.18 Q117.765 2647.91 114.686 2652.51 Q111.631 2657.1 105.821 2657.1 Q100.01 2657.1 96.9317 2652.51 Q93.8761 2647.91 93.8761 2639.18 Q93.8761 2630.43 96.9317 2625.85 Q100.01 2621.24 105.821 2621.24 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M66.4689 2476.73 L74.1077 2476.73 L74.1077 2450.36 L65.7976 2452.03 L65.7976 2447.77 L74.0614 2446.1 L78.7373 2446.1 L78.7373 2476.73 L86.3762 2476.73 L86.3762 2480.66 L66.4689 2480.66 L66.4689 2476.73 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 2449.18 Q102.209 2449.18 100.381 2452.75 Q98.5752 2456.29 98.5752 2463.42 Q98.5752 2470.52 100.381 2474.09 Q102.209 2477.63 105.821 2477.63 Q109.455 2477.63 111.26 2474.09 Q113.089 2470.52 113.089 2463.42 Q113.089 2456.29 111.26 2452.75 Q109.455 2449.18 105.821 2449.18 M105.821 2445.48 Q111.631 2445.48 114.686 2450.08 Q117.765 2454.67 117.765 2463.42 Q117.765 2472.14 114.686 2476.75 Q111.631 2481.33 105.821 2481.33 Q100.01 2481.33 96.9317 2476.75 Q93.8761 2472.14 93.8761 2463.42 Q93.8761 2454.67 96.9317 2450.08 Q100.01 2445.48 105.821 2445.48 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M69.6865 2300.96 L86.0058 2300.96 L86.0058 2304.9 L64.0615 2304.9 L64.0615 2300.96 Q66.7235 2298.21 71.3068 2293.58 Q75.9133 2288.93 77.0938 2287.58 Q79.3392 2285.06 80.2188 2283.32 Q81.1216 2281.56 81.1216 2279.87 Q81.1216 2277.12 79.1771 2275.38 Q77.2559 2273.65 74.154 2273.65 Q71.955 2273.65 69.5013 2274.41 Q67.0707 2275.18 64.293 2276.73 L64.293 2272 Q67.117 2270.87 69.5707 2270.29 Q72.0244 2269.71 74.0614 2269.71 Q79.4318 2269.71 82.6262 2272.4 Q85.8206 2275.08 85.8206 2279.57 Q85.8206 2281.7 85.0105 2283.62 Q84.2234 2285.52 82.1169 2288.12 Q81.5382 2288.79 78.4364 2292 Q75.3346 2295.2 69.6865 2300.96 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 2273.42 Q102.209 2273.42 100.381 2276.98 Q98.5752 2280.52 98.5752 2287.65 Q98.5752 2294.76 100.381 2298.32 Q102.209 2301.87 105.821 2301.87 Q109.455 2301.87 111.26 2298.32 Q113.089 2294.76 113.089 2287.65 Q113.089 2280.52 111.26 2276.98 Q109.455 2273.42 105.821 2273.42 M105.821 2269.71 Q111.631 2269.71 114.686 2274.32 Q117.765 2278.9 117.765 2287.65 Q117.765 2296.38 114.686 2300.99 Q111.631 2305.57 105.821 2305.57 Q100.01 2305.57 96.9317 2300.99 Q93.8761 2296.38 93.8761 2287.65 Q93.8761 2278.9 96.9317 2274.32 Q100.01 2269.71 105.821 2269.71 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M79.8253 2110.5 Q83.1818 2111.22 85.0567 2113.49 Q86.9549 2115.75 86.9549 2119.09 Q86.9549 2124.2 83.4364 2127 Q79.9179 2129.81 73.4364 2129.81 Q71.2605 2129.81 68.9457 2129.37 Q66.6541 2128.95 64.2004 2128.09 L64.2004 2123.58 Q66.1448 2124.71 68.4596 2125.29 Q70.7744 2125.87 73.2976 2125.87 Q77.6957 2125.87 79.9873 2124.13 Q82.3021 2122.4 82.3021 2119.09 Q82.3021 2116.03 80.1494 2114.32 Q78.0197 2112.58 74.2003 2112.58 L70.1726 2112.58 L70.1726 2108.74 L74.3855 2108.74 Q77.8346 2108.74 79.6633 2107.37 Q81.492 2105.99 81.492 2103.39 Q81.492 2100.73 79.5938 2099.32 Q77.7188 2097.88 74.2003 2097.88 Q72.279 2097.88 70.08 2098.3 Q67.8809 2098.72 65.242 2099.6 L65.242 2095.43 Q67.9041 2094.69 70.2189 2094.32 Q72.5568 2093.95 74.617 2093.95 Q79.941 2093.95 83.0429 2096.38 Q86.1447 2098.79 86.1447 2102.91 Q86.1447 2105.78 84.5012 2107.77 Q82.8577 2109.74 79.8253 2110.5 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M105.821 2097.65 Q102.209 2097.65 100.381 2101.22 Q98.5752 2104.76 98.5752 2111.89 Q98.5752 2119 100.381 2122.56 Q102.209 2126.1 105.821 2126.1 Q109.455 2126.1 111.26 2122.56 Q113.089 2119 113.089 2111.89 Q113.089 2104.76 111.26 2101.22 Q109.455 2097.65 105.821 2097.65 M105.821 2093.95 Q111.631 2093.95 114.686 2098.56 Q117.765 2103.14 117.765 2111.89 Q117.765 2120.62 114.686 2125.22 Q111.631 2129.81 105.821 2129.81 Q100.01 2129.81 96.9317 2125.22 Q93.8761 2120.62 93.8761 2111.89 Q93.8761 2103.14 96.9317 2098.56 Q100.01 2093.95 105.821 2093.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M469.436 1878.76 L477.619 1878.76 L477.619 1932.36 L507.069 1932.36 L507.069 1939.24 L469.436 1939.24 L469.436 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M535.952 1916.44 Q526.918 1916.44 523.435 1918.5 Q519.951 1920.57 519.951 1925.55 Q519.951 1929.52 522.543 1931.87 Q525.177 1934.18 529.673 1934.18 Q535.871 1934.18 539.598 1929.8 Q543.365 1925.39 543.365 1918.1 L543.365 1916.44 L535.952 1916.44 M550.819 1913.36 L550.819 1939.24 L543.365 1939.24 L543.365 1932.36 Q540.813 1936.49 537.005 1938.47 Q533.197 1940.42 527.688 1940.42 Q520.721 1940.42 516.589 1936.53 Q512.497 1932.6 512.497 1926.04 Q512.497 1918.38 517.601 1914.49 Q522.746 1910.6 532.914 1910.6 L543.365 1910.6 L543.365 1909.87 Q543.365 1904.73 539.962 1901.93 Q536.6 1899.1 530.483 1899.1 Q526.594 1899.1 522.908 1900.03 Q519.222 1900.96 515.819 1902.83 L515.819 1895.94 Q519.91 1894.36 523.759 1893.59 Q527.607 1892.78 531.253 1892.78 Q541.097 1892.78 545.958 1897.88 Q550.819 1902.99 550.819 1913.36 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M573.544 1880.99 L573.544 1893.87 L588.897 1893.87 L588.897 1899.67 L573.544 1899.67 L573.544 1924.29 Q573.544 1929.84 575.043 1931.42 Q576.583 1933 581.241 1933 L588.897 1933 L588.897 1939.24 L581.241 1939.24 Q572.613 1939.24 569.331 1936.04 Q566.05 1932.8 566.05 1924.29 L566.05 1899.67 L560.581 1899.67 L560.581 1893.87 L566.05 1893.87 L566.05 1880.99 L573.544 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M598.7 1893.87 L606.154 1893.87 L606.154 1939.24 L598.7 1939.24 L598.7 1893.87 M598.7 1876.21 L606.154 1876.21 L606.154 1885.65 L598.7 1885.65 L598.7 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M629.123 1880.99 L629.123 1893.87 L644.476 1893.87 L644.476 1899.67 L629.123 1899.67 L629.123 1924.29 Q629.123 1929.84 630.622 1931.42 Q632.161 1933 636.819 1933 L644.476 1933 L644.476 1939.24 L636.819 1939.24 Q628.191 1939.24 624.91 1936.04 Q621.629 1932.8 621.629 1924.29 L621.629 1899.67 L616.16 1899.67 L616.16 1893.87 L621.629 1893.87 L621.629 1880.99 L629.123 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M653.509 1921.34 L653.509 1893.87 L660.963 1893.87 L660.963 1921.05 Q660.963 1927.5 663.474 1930.74 Q665.986 1933.94 671.009 1933.94 Q677.045 1933.94 680.529 1930.09 Q684.053 1926.24 684.053 1919.6 L684.053 1893.87 L691.507 1893.87 L691.507 1939.24 L684.053 1939.24 L684.053 1932.28 Q681.339 1936.41 677.734 1938.43 Q674.169 1940.42 669.429 1940.42 Q661.611 1940.42 657.56 1935.56 Q653.509 1930.7 653.509 1921.34 M672.265 1892.78 L672.265 1892.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M736.715 1900.76 L736.715 1876.21 L744.168 1876.21 L744.168 1939.24 L736.715 1939.24 L736.715 1932.44 Q734.365 1936.49 730.76 1938.47 Q727.195 1940.42 722.172 1940.42 Q713.949 1940.42 708.764 1933.85 Q703.619 1927.29 703.619 1916.6 Q703.619 1905.9 708.764 1899.34 Q713.949 1892.78 722.172 1892.78 Q727.195 1892.78 730.76 1894.76 Q734.365 1896.71 736.715 1900.76 M711.316 1916.6 Q711.316 1924.82 714.678 1929.52 Q718.081 1934.18 723.995 1934.18 Q729.909 1934.18 733.312 1929.52 Q736.715 1924.82 736.715 1916.6 Q736.715 1908.37 733.312 1903.72 Q729.909 1899.02 723.995 1899.02 Q718.081 1899.02 714.678 1903.72 Q711.316 1908.37 711.316 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M798.329 1914.69 L798.329 1918.34 L764.058 1918.34 Q764.545 1926.04 768.676 1930.09 Q772.849 1934.1 780.262 1934.1 Q784.556 1934.1 788.566 1933.04 Q792.617 1931.99 796.587 1929.89 L796.587 1936.93 Q792.577 1938.64 788.364 1939.53 Q784.151 1940.42 779.816 1940.42 Q768.96 1940.42 762.6 1934.1 Q756.281 1927.78 756.281 1917 Q756.281 1905.86 762.276 1899.34 Q768.312 1892.78 778.52 1892.78 Q787.675 1892.78 792.982 1898.69 Q798.329 1904.57 798.329 1914.69 M790.875 1912.51 Q790.794 1906.39 787.432 1902.74 Q784.11 1899.1 778.601 1899.1 Q772.363 1899.1 768.595 1902.62 Q764.869 1906.15 764.301 1912.55 L790.875 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M854.839 1876.29 Q849.411 1885.61 846.778 1894.72 Q844.145 1903.84 844.145 1913.2 Q844.145 1922.55 846.778 1931.75 Q849.451 1940.9 854.839 1950.18 L848.358 1950.18 Q842.281 1940.66 839.243 1931.46 Q836.246 1922.27 836.246 1913.2 Q836.246 1904.16 839.243 1895.01 Q842.241 1885.85 848.358 1876.29 L854.839 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M899.156 1900.76 L899.156 1876.21 L906.61 1876.21 L906.61 1939.24 L899.156 1939.24 L899.156 1932.44 Q896.807 1936.49 893.201 1938.47 Q889.636 1940.42 884.613 1940.42 Q876.39 1940.42 871.205 1933.85 Q866.06 1927.29 866.06 1916.6 Q866.06 1905.9 871.205 1899.34 Q876.39 1892.78 884.613 1892.78 Q889.636 1892.78 893.201 1894.76 Q896.807 1896.71 899.156 1900.76 M873.757 1916.6 Q873.757 1924.82 877.119 1929.52 Q880.522 1934.18 886.436 1934.18 Q892.351 1934.18 895.753 1929.52 Q899.156 1924.82 899.156 1916.6 Q899.156 1908.37 895.753 1903.72 Q892.351 1899.02 886.436 1899.02 Q880.522 1899.02 877.119 1903.72 Q873.757 1908.37 873.757 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M960.77 1914.69 L960.77 1918.34 L926.5 1918.34 Q926.986 1926.04 931.118 1930.09 Q935.29 1934.1 942.703 1934.1 Q946.997 1934.1 951.008 1933.04 Q955.059 1931.99 959.028 1929.89 L959.028 1936.93 Q955.018 1938.64 950.805 1939.53 Q946.592 1940.42 942.258 1940.42 Q931.401 1940.42 925.041 1934.1 Q918.722 1927.78 918.722 1917 Q918.722 1905.86 924.717 1899.34 Q930.753 1892.78 940.961 1892.78 Q950.116 1892.78 955.423 1898.69 Q960.77 1904.57 960.77 1914.69 M953.317 1912.51 Q953.236 1906.39 949.873 1902.74 Q946.552 1899.1 941.042 1899.1 Q934.804 1899.1 931.037 1902.62 Q927.31 1906.15 926.743 1912.55 L953.317 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1002.86 1916.03 Q1002.86 1907.93 999.497 1903.47 Q996.175 1899.02 990.139 1899.02 Q984.144 1899.02 980.782 1903.47 Q977.46 1907.93 977.46 1916.03 Q977.46 1924.09 980.782 1928.55 Q984.144 1933 990.139 1933 Q996.175 1933 999.497 1928.55 Q1002.86 1924.09 1002.86 1916.03 M1010.31 1933.61 Q1010.31 1945.2 1005.17 1950.83 Q1000.02 1956.5 989.41 1956.5 Q985.481 1956.5 981.997 1955.89 Q978.513 1955.32 975.232 1954.11 L975.232 1946.86 Q978.513 1948.64 981.713 1949.49 Q984.914 1950.34 988.235 1950.34 Q995.568 1950.34 999.213 1946.49 Q1002.86 1942.69 1002.86 1934.95 L1002.86 1931.26 Q1000.55 1935.27 996.945 1937.26 Q993.34 1939.24 988.316 1939.24 Q979.972 1939.24 974.867 1932.88 Q969.763 1926.52 969.763 1916.03 Q969.763 1905.5 974.867 1899.14 Q979.972 1892.78 988.316 1892.78 Q993.34 1892.78 996.945 1894.76 Q1000.55 1896.75 1002.86 1900.76 L1002.86 1893.87 L1010.31 1893.87 L1010.31 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1024.49 1876.29 L1030.97 1876.29 Q1037.05 1885.85 1040.05 1895.01 Q1043.08 1904.16 1043.08 1913.2 Q1043.08 1922.27 1040.05 1931.46 Q1037.05 1940.66 1030.97 1950.18 L1024.49 1950.18 Q1029.88 1940.9 1032.51 1931.75 Q1035.19 1922.55 1035.19 1913.2 Q1035.19 1903.84 1032.51 1894.72 Q1029.88 1885.61 1024.49 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip036)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="193.529,2639.15 195.771,2639.15 198.013,2639.14 200.255,2639.14 202.497,2639.14 204.739,2639.13 206.981,2639.12 209.223,2639.11 211.464,2639.1 213.706,2639.09 215.948,2639.07 218.19,2639.05 220.432,2639.03 222.674,2639.01 224.916,2638.98 227.158,2638.95 229.4,2638.91 231.642,2638.87 233.884,2638.83 236.126,2638.78 238.368,2638.73 240.61,2638.67 242.852,2638.6 245.094,2638.53 247.336,2638.46 249.578,2638.37 251.82,2638.28 254.062,2638.19 256.304,2638.08 258.546,2637.97 260.788,2637.84 263.03,2637.71 265.271,2637.57 267.513,2637.41 269.755,2637.25 271.997,2637.07 274.239,2636.88 276.481,2636.68 278.723,2636.47 280.965,2636.24 283.207,2636 285.449,2635.75 287.691,2635.48 289.933,2635.2 292.175,2634.9 294.417,2634.59 296.659,2634.26 298.901,2633.92 301.143,2633.56 303.385,2633.19 305.627,2632.8 307.869,2632.39 310.111,2631.97 312.353,2631.54 314.595,2631.09 316.837,2630.62 319.078,2630.15 321.32,2629.65 323.562,2629.15 325.804,2628.63 328.046,2628.1 330.288,2627.55 332.53,2626.99 334.772,2626.43 337.014,2625.84 339.256,2625.25 341.498,2624.65 343.74,2624.04 345.982,2623.42 348.224,2622.78 350.466,2622.14 352.708,2621.49 354.95,2620.83 357.192,2620.16 359.434,2619.49 361.676,2618.8 363.918,2618.11 366.16,2617.41 368.402,2616.71 370.644,2616 372.885,2615.28 375.127,2614.55 377.369,2613.82 379.611,2613.09 381.853,2612.34 384.095,2611.6 386.337,2610.84 388.579,2610.08 390.821,2609.32 393.063,2608.55 395.305,2607.78 397.547,2607 399.789,2606.21 402.031,2605.43 404.273,2604.63 406.515,2603.83 408.757,2603.03 410.999,2602.23 413.241,2601.41 415.483,2600.6 417.725,2599.78 419.967,2598.95 422.209,2598.12 424.451,2597.29 426.692,2596.45 428.934,2595.61 431.176,2594.76 433.418,2593.91 435.66,2593.05 437.902,2592.19 440.144,2591.32 442.386,2590.45 444.628,2589.57 446.87,2588.69 449.112,2587.8 451.354,2586.9 453.596,2586 455.838,2585.1 458.08,2584.19 460.322,2583.27 462.564,2582.34 464.806,2581.41 467.048,2580.48 469.29,2579.53 471.532,2578.58 473.774,2577.62 476.016,2576.65 478.258,2575.68 480.5,2574.69 482.741,2573.7 484.983,2572.7 487.225,2571.69 489.467,2570.67 491.709,2569.64 493.951,2568.6 496.193,2567.56 498.435,2566.5 500.677,2565.43 502.919,2564.35 505.161,2563.26 507.403,2562.16 509.645,2561.05 511.887,2559.93 514.129,2558.8 516.371,2557.66 518.613,2556.51 520.855,2555.35 523.097,2554.17 525.339,2552.99 527.581,2551.8 529.823,2550.59 532.065,2549.38 534.307,2548.16 536.548,2546.92 538.79,2545.68 541.032,2544.43 543.274,2543.17 545.516,2541.9 547.758,2540.63 550,2539.34 552.242,2538.05 554.484,2536.75 556.726,2535.44 558.968,2534.13 561.21,2532.81 563.452,2531.48 565.694,2530.15 567.936,2528.81 570.178,2527.46 572.42,2526.11 574.662,2524.75 576.904,2523.39 579.146,2522.02 581.388,2520.64 583.63,2519.26 585.872,2517.88 588.114,2516.49 590.355,2515.09 592.597,2513.69 594.839,2512.28 597.081,2510.87 599.323,2509.46 601.565,2508.04 603.807,2506.61 606.049,2505.18 608.291,2503.75 610.533,2502.31 612.775,2500.86 615.017,2499.41 617.259,2497.96 619.501,2496.49 621.743,2495.03 623.985,2493.56 626.227,2492.08 628.469,2490.6 630.711,2489.11 632.953,2487.62 635.195,2486.12 637.437,2484.62 639.679,2483.11 641.921,2481.59 644.162,2480.07 646.404,2478.54 648.646,2477.01 650.888,2475.47 653.13,2473.92 655.372,2472.37 657.614,2470.81 659.856,2469.24 662.098,2467.67 664.34,2466.09 666.582,2464.51 668.824,2462.91 671.066,2461.32 673.308,2459.71 675.55,2458.1 677.792,2456.49 680.034,2454.87 682.276,2453.24 684.518,2451.6 686.76,2449.97 689.002,2448.32 691.244,2446.67 693.486,2445.02 695.728,2443.36 697.969,2441.69 700.211,2440.02 702.453,2438.35 704.695,2436.67 706.937,2434.99 709.179,2433.3 711.421,2431.61 713.663,2429.92 715.905,2428.22 718.147,2426.52 720.389,2424.81 722.631,2423.1 724.873,2421.39 727.115,2419.67 729.357,2417.96 731.599,2416.23 733.841,2414.51 736.083,2412.78 738.325,2411.05 740.567,2409.32 742.809,2407.58 745.051,2405.84 747.293,2404.1 749.535,2402.36 751.776,2400.61 754.018,2398.86 756.26,2397.11 758.502,2395.35 760.744,2393.59 762.986,2391.83 765.228,2390.07 767.47,2388.3 769.712,2386.53 771.954,2384.76 774.196,2382.99 776.438,2381.21 778.68,2379.43 780.922,2377.65 783.164,2375.86 785.406,2374.08 787.648,2372.29 789.89,2370.5 792.132,2368.7 794.374,2366.9 796.616,2365.11 798.858,2363.3 801.1,2361.5 803.342,2359.7 805.583,2357.89 807.825,2356.08 810.067,2354.27 812.309,2352.46 814.551,2350.64 816.793,2348.83 819.035,2347.01 821.277,2345.19 823.519,2343.37 825.761,2341.55 828.003,2339.73 830.245,2337.91 832.487,2336.08 834.729,2334.26 836.971,2332.43 839.213,2330.61 841.455,2328.78 843.697,2326.96 845.939,2325.13 848.181,2323.3 850.423,2321.47 852.665,2319.65 854.907,2317.82 857.149,2315.99 859.391,2314.16 861.632,2312.34 863.874,2310.51 866.116,2308.68 868.358,2306.86 870.6,2305.03 872.842,2303.2 875.084,2301.38 877.326,2299.55 879.568,2297.73 881.81,2295.9 884.052,2294.08 886.294,2292.26 888.536,2290.44 890.778,2288.61 893.02,2286.79 895.262,2284.98 897.504,2283.16 899.746,2281.34 901.988,2279.52 904.23,2277.71 906.472,2275.9 908.714,2274.09 910.956,2272.28 913.198,2270.47 915.439,2268.66 917.681,2266.86 919.923,2265.05 922.165,2263.25 924.407,2261.46 926.649,2259.66 928.891,2257.86 931.133,2256.07 933.375,2254.28 935.617,2252.49 937.859,2250.71 940.101,2248.93 942.343,2247.15 944.585,2245.37 946.827,2243.6 949.069,2241.82 951.311,2240.06 953.553,2238.29 955.795,2236.53 958.037,2234.77 960.279,2233.01 962.521,2231.26 964.763,2229.51 967.005,2227.76 969.246,2226.02 971.488,2224.28 973.73,2222.54 975.972,2220.81 978.214,2219.08 980.456,2217.35 982.698,2215.63 984.94,2213.91 987.182,2212.2 989.424,2210.49 991.666,2208.78 993.908,2207.08 996.15,2205.38 998.392,2203.69 1000.63,2202 1002.88,2200.31 1005.12,2198.63 1007.36,2196.96 1009.6,2195.29 1011.84,2193.62 1014.09,2191.96 1016.33,2190.31 1018.57,2188.66 1020.81,2187.01 1023.05,2185.37 1025.3,2183.74 1027.54,2182.11 1029.78,2180.48 1032.02,2178.87 1034.26,2177.25 1036.51,2175.65 1038.75,2174.05 1040.99,2172.45 1043.23,2170.86 1045.47,2169.28 1047.72,2167.7 1049.96,2166.13 1052.2,2164.57 1054.44,2163.01 1056.68,2161.46 1058.92,2159.91 1061.17,2158.37 1063.41,2156.84 1065.65,2155.31 1067.89,2153.79 1070.13,2152.28 1072.38,2150.77 1074.62,2149.27 1076.86,2147.78 1079.1,2146.29 1081.34,2144.81 1083.59,2143.34 1085.83,2141.88 1088.07,2140.42 1090.31,2138.97 1092.55,2137.52 1094.8,2136.08 1097.04,2134.65 1099.28,2133.23 1101.52,2131.82 1103.76,2130.41 1106.01,2129.01 1108.25,2127.62 1110.49,2126.24 1112.73,2124.86 1114.97,2123.49 1117.22,2122.13 1119.46,2120.78 1121.7,2119.44 1123.94,2118.11 1126.18,2116.78 1128.43,2115.46 1130.67,2114.15 1132.91,2112.85 1135.15,2111.56 1137.39,2110.28 1139.64,2109 1141.88,2107.74 1144.12,2106.48 1146.36,2105.23 1148.6,2103.99 1150.85,2102.76 1153.09,2101.54 1155.33,2100.33 1157.57,2099.12 1159.81,2097.93 1162.05,2096.75 1164.3,2095.57 1166.54,2094.4 1168.78,2093.25 1171.02,2092.1 1173.26,2090.96 1175.51,2089.83 1177.75,2088.71 1179.99,2087.6 1182.23,2086.5 1184.47,2085.4 1186.72,2084.32 1188.96,2083.25 1191.2,2082.18 1193.44,2081.13 1195.68,2080.09 1197.93,2079.05 1200.17,2078.03 1202.41,2077.02 1204.65,2076.01 1206.89,2075.02 1209.14,2074.03 1211.38,2073.06 1213.62,2072.09 1215.86,2071.14 1218.1,2070.2 1220.35,2069.26 1222.59,2068.34 1224.83,2067.43 1227.07,2066.53 1229.31,2065.64 1231.56,2064.76 1233.8,2063.88 1236.04,2063.02 1238.28,2062.18 1240.52,2061.34 1242.77,2060.51 1245.01,2059.69 1247.25,2058.88 1249.49,2058.09 1251.73,2057.3 1253.98,2056.53 1256.22,2055.76 1258.46,2055.01 1260.7,2054.26 1262.94,2053.53 1265.19,2052.8 1267.43,2052.09 1269.67,2051.39 1271.91,2050.69 1274.15,2050.01 1276.39,2049.34 1278.64,2048.67 1280.88,2048.02 1283.12,2047.37 1285.36,2046.74 1287.6,2046.11 1289.85,2045.5 1292.09,2044.89 1294.33,2044.3 1296.57,2043.71 1298.81,2043.14 1301.06,2042.57 1303.3,2042.01 1305.54,2041.47 1307.78,2040.93 1310.02,2040.4 1312.27,2039.89 1314.51,2039.38 1316.75,2038.89 1318.99,2038.4 "/>
+<path clip-path="url(#clip030)" d="M1559.76 2657.17 L2752.76 2657.17 L2752.76 2020.38 L1559.76 2020.38  Z" fill="#ffffff" fill-rule="evenodd" fill-opacity="1"/>
 <defs>
-  <clipPath id="clip217">
+  <clipPath id="clip037">
     <rect x="1559" y="2020" width="1194" height="638"/>
   </clipPath>
 </defs>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,2657.17 1593.53,2020.38 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,2657.17 1873.77,2020.38 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,2657.17 2154.02,2020.38 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,2657.17 2434.26,2020.38 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,2657.17 2714.51,2020.38 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2657.17 2752.76,2657.17 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,2657.17 1593.53,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,2657.17 1873.77,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,2657.17 2154.02,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,2657.17 2434.26,2638.27 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,2657.17 2714.51,2638.27 "/>
-<path clip-path="url(#clip210)" d="M1593.53 2691.29 Q1589.92 2691.29 1588.09 2694.85 Q1586.28 2698.39 1586.28 2705.52 Q1586.28 2712.63 1588.09 2716.19 Q1589.92 2719.74 1593.53 2719.74 Q1597.16 2719.74 1598.97 2716.19 Q1600.8 2712.63 1600.8 2705.52 Q1600.8 2698.39 1598.97 2694.85 Q1597.16 2691.29 1593.53 2691.29 M1593.53 2687.58 Q1599.34 2687.58 1602.39 2692.19 Q1605.47 2696.77 1605.47 2705.52 Q1605.47 2714.25 1602.39 2718.86 Q1599.34 2723.44 1593.53 2723.44 Q1587.72 2723.44 1584.64 2718.86 Q1581.58 2714.25 1581.58 2705.52 Q1581.58 2696.77 1584.64 2692.19 Q1587.72 2687.58 1593.53 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1833.39 2688.21 L1851.75 2688.21 L1851.75 2692.14 L1837.67 2692.14 L1837.67 2700.62 Q1838.69 2700.27 1839.71 2700.11 Q1840.73 2699.92 1841.75 2699.92 Q1847.54 2699.92 1850.91 2703.09 Q1854.29 2706.26 1854.29 2711.68 Q1854.29 2717.26 1850.82 2720.36 Q1847.35 2723.44 1841.03 2723.44 Q1838.85 2723.44 1836.59 2723.07 Q1834.34 2722.7 1831.93 2721.96 L1831.93 2717.26 Q1834.02 2718.39 1836.24 2718.95 Q1838.46 2719.5 1840.94 2719.5 Q1844.94 2719.5 1847.28 2717.4 Q1849.62 2715.29 1849.62 2711.68 Q1849.62 2708.07 1847.28 2705.96 Q1844.94 2703.86 1840.94 2703.86 Q1839.06 2703.86 1837.19 2704.27 Q1835.34 2704.69 1833.39 2705.57 L1833.39 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1873.51 2691.29 Q1869.9 2691.29 1868.07 2694.85 Q1866.26 2698.39 1866.26 2705.52 Q1866.26 2712.63 1868.07 2716.19 Q1869.9 2719.74 1873.51 2719.74 Q1877.14 2719.74 1878.95 2716.19 Q1880.78 2712.63 1880.78 2705.52 Q1880.78 2698.39 1878.95 2694.85 Q1877.14 2691.29 1873.51 2691.29 M1873.51 2687.58 Q1879.32 2687.58 1882.37 2692.19 Q1885.45 2696.77 1885.45 2705.52 Q1885.45 2714.25 1882.37 2718.86 Q1879.32 2723.44 1873.51 2723.44 Q1867.7 2723.44 1864.62 2718.86 Q1861.56 2714.25 1861.56 2705.52 Q1861.56 2696.77 1864.62 2692.19 Q1867.7 2687.58 1873.51 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1903.67 2691.29 Q1900.06 2691.29 1898.23 2694.85 Q1896.42 2698.39 1896.42 2705.52 Q1896.42 2712.63 1898.23 2716.19 Q1900.06 2719.74 1903.67 2719.74 Q1907.3 2719.74 1909.11 2716.19 Q1910.94 2712.63 1910.94 2705.52 Q1910.94 2698.39 1909.11 2694.85 Q1907.3 2691.29 1903.67 2691.29 M1903.67 2687.58 Q1909.48 2687.58 1912.53 2692.19 Q1915.61 2696.77 1915.61 2705.52 Q1915.61 2714.25 1912.53 2718.86 Q1909.48 2723.44 1903.67 2723.44 Q1897.86 2723.44 1894.78 2718.86 Q1891.72 2714.25 1891.72 2705.52 Q1891.72 2696.77 1894.78 2692.19 Q1897.86 2687.58 1903.67 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2098.54 2718.83 L2106.18 2718.83 L2106.18 2692.47 L2097.87 2694.13 L2097.87 2689.87 L2106.14 2688.21 L2110.81 2688.21 L2110.81 2718.83 L2118.45 2718.83 L2118.45 2722.77 L2098.54 2722.77 L2098.54 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2137.9 2691.29 Q2134.28 2691.29 2132.46 2694.85 Q2130.65 2698.39 2130.65 2705.52 Q2130.65 2712.63 2132.46 2716.19 Q2134.28 2719.74 2137.9 2719.74 Q2141.53 2719.74 2143.34 2716.19 Q2145.16 2712.63 2145.16 2705.52 Q2145.16 2698.39 2143.34 2694.85 Q2141.53 2691.29 2137.9 2691.29 M2137.9 2687.58 Q2143.71 2687.58 2146.76 2692.19 Q2149.84 2696.77 2149.84 2705.52 Q2149.84 2714.25 2146.76 2718.86 Q2143.71 2723.44 2137.9 2723.44 Q2132.09 2723.44 2129.01 2718.86 Q2125.95 2714.25 2125.95 2705.52 Q2125.95 2696.77 2129.01 2692.19 Q2132.09 2687.58 2137.9 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2168.06 2691.29 Q2164.45 2691.29 2162.62 2694.85 Q2160.81 2698.39 2160.81 2705.52 Q2160.81 2712.63 2162.62 2716.19 Q2164.45 2719.74 2168.06 2719.74 Q2171.69 2719.74 2173.5 2716.19 Q2175.33 2712.63 2175.33 2705.52 Q2175.33 2698.39 2173.5 2694.85 Q2171.69 2691.29 2168.06 2691.29 M2168.06 2687.58 Q2173.87 2687.58 2176.92 2692.19 Q2180 2696.77 2180 2705.52 Q2180 2714.25 2176.92 2718.86 Q2173.87 2723.44 2168.06 2723.44 Q2162.25 2723.44 2159.17 2718.86 Q2156.11 2714.25 2156.11 2705.52 Q2156.11 2696.77 2159.17 2692.19 Q2162.25 2687.58 2168.06 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2198.22 2691.29 Q2194.61 2691.29 2192.78 2694.85 Q2190.97 2698.39 2190.97 2705.52 Q2190.97 2712.63 2192.78 2716.19 Q2194.61 2719.74 2198.22 2719.74 Q2201.85 2719.74 2203.66 2716.19 Q2205.49 2712.63 2205.49 2705.52 Q2205.49 2698.39 2203.66 2694.85 Q2201.85 2691.29 2198.22 2691.29 M2198.22 2687.58 Q2204.03 2687.58 2207.09 2692.19 Q2210.16 2696.77 2210.16 2705.52 Q2210.16 2714.25 2207.09 2718.86 Q2204.03 2723.44 2198.22 2723.44 Q2192.41 2723.44 2189.33 2718.86 Q2186.28 2714.25 2186.28 2705.52 Q2186.28 2696.77 2189.33 2692.19 Q2192.41 2687.58 2198.22 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2378.79 2718.83 L2386.43 2718.83 L2386.43 2692.47 L2378.12 2694.13 L2378.12 2689.87 L2386.38 2688.21 L2391.06 2688.21 L2391.06 2718.83 L2398.7 2718.83 L2398.7 2722.77 L2378.79 2722.77 L2378.79 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2408.19 2688.21 L2426.54 2688.21 L2426.54 2692.14 L2412.47 2692.14 L2412.47 2700.62 Q2413.49 2700.27 2414.51 2700.11 Q2415.52 2699.92 2416.54 2699.92 Q2422.33 2699.92 2425.71 2703.09 Q2429.09 2706.26 2429.09 2711.68 Q2429.09 2717.26 2425.62 2720.36 Q2422.15 2723.44 2415.83 2723.44 Q2413.65 2723.44 2411.38 2723.07 Q2409.14 2722.7 2406.73 2721.96 L2406.73 2717.26 Q2408.81 2718.39 2411.03 2718.95 Q2413.26 2719.5 2415.73 2719.5 Q2419.74 2719.5 2422.08 2717.4 Q2424.41 2715.29 2424.41 2711.68 Q2424.41 2708.07 2422.08 2705.96 Q2419.74 2703.86 2415.73 2703.86 Q2413.86 2703.86 2411.98 2704.27 Q2410.13 2704.69 2408.19 2705.57 L2408.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2448.3 2691.29 Q2444.69 2691.29 2442.86 2694.85 Q2441.06 2698.39 2441.06 2705.52 Q2441.06 2712.63 2442.86 2716.19 Q2444.69 2719.74 2448.3 2719.74 Q2451.94 2719.74 2453.74 2716.19 Q2455.57 2712.63 2455.57 2705.52 Q2455.57 2698.39 2453.74 2694.85 Q2451.94 2691.29 2448.3 2691.29 M2448.3 2687.58 Q2454.11 2687.58 2457.17 2692.19 Q2460.25 2696.77 2460.25 2705.52 Q2460.25 2714.25 2457.17 2718.86 Q2454.11 2723.44 2448.3 2723.44 Q2442.49 2723.44 2439.41 2718.86 Q2436.36 2714.25 2436.36 2705.52 Q2436.36 2696.77 2439.41 2692.19 Q2442.49 2687.58 2448.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2478.46 2691.29 Q2474.85 2691.29 2473.02 2694.85 Q2471.22 2698.39 2471.22 2705.52 Q2471.22 2712.63 2473.02 2716.19 Q2474.85 2719.74 2478.46 2719.74 Q2482.1 2719.74 2483.9 2716.19 Q2485.73 2712.63 2485.73 2705.52 Q2485.73 2698.39 2483.9 2694.85 Q2482.1 2691.29 2478.46 2691.29 M2478.46 2687.58 Q2484.27 2687.58 2487.33 2692.19 Q2490.41 2696.77 2490.41 2705.52 Q2490.41 2714.25 2487.33 2718.86 Q2484.27 2723.44 2478.46 2723.44 Q2472.65 2723.44 2469.58 2718.86 Q2466.52 2714.25 2466.52 2705.52 Q2466.52 2696.77 2469.58 2692.19 Q2472.65 2687.58 2478.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2663.12 2718.83 L2679.44 2718.83 L2679.44 2722.77 L2657.49 2722.77 L2657.49 2718.83 Q2660.16 2716.08 2664.74 2711.45 Q2669.35 2706.8 2670.53 2705.45 Q2672.77 2702.93 2673.65 2701.19 Q2674.55 2699.43 2674.55 2697.74 Q2674.55 2694.99 2672.61 2693.25 Q2670.69 2691.52 2667.59 2691.52 Q2665.39 2691.52 2662.93 2692.28 Q2660.5 2693.05 2657.73 2694.6 L2657.73 2689.87 Q2660.55 2688.74 2663 2688.16 Q2665.46 2687.58 2667.49 2687.58 Q2672.86 2687.58 2676.06 2690.27 Q2679.25 2692.95 2679.25 2697.44 Q2679.25 2699.57 2678.44 2701.49 Q2677.66 2703.39 2675.55 2705.99 Q2674.97 2706.66 2671.87 2709.87 Q2668.77 2713.07 2663.12 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2699.25 2691.29 Q2695.64 2691.29 2693.81 2694.85 Q2692.01 2698.39 2692.01 2705.52 Q2692.01 2712.63 2693.81 2716.19 Q2695.64 2719.74 2699.25 2719.74 Q2702.89 2719.74 2704.69 2716.19 Q2706.52 2712.63 2706.52 2705.52 Q2706.52 2698.39 2704.69 2694.85 Q2702.89 2691.29 2699.25 2691.29 M2699.25 2687.58 Q2705.06 2687.58 2708.12 2692.19 Q2711.2 2696.77 2711.2 2705.52 Q2711.2 2714.25 2708.12 2718.86 Q2705.06 2723.44 2699.25 2723.44 Q2693.44 2723.44 2690.36 2718.86 Q2687.31 2714.25 2687.31 2705.52 Q2687.31 2696.77 2690.36 2692.19 Q2693.44 2687.58 2699.25 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2729.42 2691.29 Q2725.8 2691.29 2723.98 2694.85 Q2722.17 2698.39 2722.17 2705.52 Q2722.17 2712.63 2723.98 2716.19 Q2725.8 2719.74 2729.42 2719.74 Q2733.05 2719.74 2734.86 2716.19 Q2736.68 2712.63 2736.68 2705.52 Q2736.68 2698.39 2734.86 2694.85 Q2733.05 2691.29 2729.42 2691.29 M2729.42 2687.58 Q2735.23 2687.58 2738.28 2692.19 Q2741.36 2696.77 2741.36 2705.52 Q2741.36 2714.25 2738.28 2718.86 Q2735.23 2723.44 2729.42 2723.44 Q2723.61 2723.44 2720.53 2718.86 Q2717.47 2714.25 2717.47 2705.52 Q2717.47 2696.77 2720.53 2692.19 Q2723.61 2687.58 2729.42 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2759.58 2691.29 Q2755.97 2691.29 2754.14 2694.85 Q2752.33 2698.39 2752.33 2705.52 Q2752.33 2712.63 2754.14 2716.19 Q2755.97 2719.74 2759.58 2719.74 Q2763.21 2719.74 2765.02 2716.19 Q2766.85 2712.63 2766.85 2705.52 Q2766.85 2698.39 2765.02 2694.85 Q2763.21 2691.29 2759.58 2691.29 M2759.58 2687.58 Q2765.39 2687.58 2768.44 2692.19 Q2771.52 2696.77 2771.52 2705.52 Q2771.52 2714.25 2768.44 2718.86 Q2765.39 2723.44 2759.58 2723.44 Q2753.77 2723.44 2750.69 2718.86 Q2747.63 2714.25 2747.63 2705.52 Q2747.63 2696.77 2750.69 2692.19 Q2753.77 2687.58 2759.58 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2549.77 2752.76,2549.77 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2403.67 2752.76,2403.67 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2257.56 2752.76,2257.56 "/>
-<polyline clip-path="url(#clip217)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2111.45 2752.76,2111.45 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2657.17 1559.76,2020.38 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2549.77 1578.66,2549.77 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2403.67 1578.66,2403.67 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2257.56 1578.66,2257.56 "/>
-<polyline clip-path="url(#clip210)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2111.45 1578.66,2111.45 "/>
-<path clip-path="url(#clip210)" d="M1469.69 2563.12 L1486.01 2563.12 L1486.01 2567.05 L1464.06 2567.05 L1464.06 2563.12 Q1466.72 2560.36 1471.31 2555.74 Q1475.91 2551.08 1477.09 2549.74 Q1479.34 2547.22 1480.22 2545.48 Q1481.12 2543.72 1481.12 2542.03 Q1481.12 2539.28 1479.18 2537.54 Q1477.26 2535.8 1474.15 2535.8 Q1471.95 2535.8 1469.5 2536.57 Q1467.07 2537.33 1464.29 2538.88 L1464.29 2534.16 Q1467.12 2533.03 1469.57 2532.45 Q1472.02 2531.87 1474.06 2531.87 Q1479.43 2531.87 1482.63 2534.55 Q1485.82 2537.24 1485.82 2541.73 Q1485.82 2543.86 1485.01 2545.78 Q1484.22 2547.68 1482.12 2550.27 Q1481.54 2550.94 1478.44 2554.16 Q1475.33 2557.36 1469.69 2563.12 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 2535.57 Q1502.21 2535.57 1500.38 2539.14 Q1498.58 2542.68 1498.58 2549.81 Q1498.58 2556.92 1500.38 2560.48 Q1502.21 2564.02 1505.82 2564.02 Q1509.45 2564.02 1511.26 2560.48 Q1513.09 2556.92 1513.09 2549.81 Q1513.09 2542.68 1511.26 2539.14 Q1509.45 2535.57 1505.82 2535.57 M1505.82 2531.87 Q1511.63 2531.87 1514.69 2536.48 Q1517.76 2541.06 1517.76 2549.81 Q1517.76 2558.54 1514.69 2563.14 Q1511.63 2567.73 1505.82 2567.73 Q1500.01 2567.73 1496.93 2563.14 Q1493.88 2558.54 1493.88 2549.81 Q1493.88 2541.06 1496.93 2536.48 Q1500.01 2531.87 1505.82 2531.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1478.51 2390.46 L1466.7 2408.91 L1478.51 2408.91 L1478.51 2390.46 M1477.28 2386.39 L1483.16 2386.39 L1483.16 2408.91 L1488.09 2408.91 L1488.09 2412.8 L1483.16 2412.8 L1483.16 2420.95 L1478.51 2420.95 L1478.51 2412.8 L1462.9 2412.8 L1462.9 2408.29 L1477.28 2386.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 2389.47 Q1502.21 2389.47 1500.38 2393.03 Q1498.58 2396.57 1498.58 2403.7 Q1498.58 2410.81 1500.38 2414.37 Q1502.21 2417.92 1505.82 2417.92 Q1509.45 2417.92 1511.26 2414.37 Q1513.09 2410.81 1513.09 2403.7 Q1513.09 2396.57 1511.26 2393.03 Q1509.45 2389.47 1505.82 2389.47 M1505.82 2385.76 Q1511.63 2385.76 1514.69 2390.37 Q1517.76 2394.95 1517.76 2403.7 Q1517.76 2412.43 1514.69 2417.04 Q1511.63 2421.62 1505.82 2421.62 Q1500.01 2421.62 1496.93 2417.04 Q1493.88 2412.43 1493.88 2403.7 Q1493.88 2394.95 1496.93 2390.37 Q1500.01 2385.76 1505.82 2385.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1476.24 2255.7 Q1473.09 2255.7 1471.24 2257.85 Q1469.41 2260 1469.41 2263.75 Q1469.41 2267.48 1471.24 2269.66 Q1473.09 2271.81 1476.24 2271.81 Q1479.39 2271.81 1481.21 2269.66 Q1483.07 2267.48 1483.07 2263.75 Q1483.07 2260 1481.21 2257.85 Q1479.39 2255.7 1476.24 2255.7 M1485.52 2241.05 L1485.52 2245.3 Q1483.76 2244.47 1481.95 2244.03 Q1480.17 2243.59 1478.41 2243.59 Q1473.78 2243.59 1471.33 2246.72 Q1468.9 2249.84 1468.55 2256.16 Q1469.92 2254.15 1471.98 2253.08 Q1474.04 2251.99 1476.52 2251.99 Q1481.72 2251.99 1484.73 2255.17 Q1487.77 2258.31 1487.77 2263.75 Q1487.77 2269.08 1484.62 2272.29 Q1481.47 2275.51 1476.24 2275.51 Q1470.24 2275.51 1467.07 2270.93 Q1463.9 2266.32 1463.9 2257.6 Q1463.9 2249.4 1467.79 2244.54 Q1471.68 2239.66 1478.23 2239.66 Q1479.99 2239.66 1481.77 2240 Q1483.58 2240.35 1485.52 2241.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 2243.36 Q1502.21 2243.36 1500.38 2246.92 Q1498.58 2250.47 1498.58 2257.6 Q1498.58 2264.7 1500.38 2268.27 Q1502.21 2271.81 1505.82 2271.81 Q1509.45 2271.81 1511.26 2268.27 Q1513.09 2264.7 1513.09 2257.6 Q1513.09 2250.47 1511.26 2246.92 Q1509.45 2243.36 1505.82 2243.36 M1505.82 2239.66 Q1511.63 2239.66 1514.69 2244.26 Q1517.76 2248.85 1517.76 2257.6 Q1517.76 2266.32 1514.69 2270.93 Q1511.63 2275.51 1505.82 2275.51 Q1500.01 2275.51 1496.93 2270.93 Q1493.88 2266.32 1493.88 2257.6 Q1493.88 2248.85 1496.93 2244.26 Q1500.01 2239.66 1505.82 2239.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1475.66 2112.32 Q1472.33 2112.32 1470.4 2114.1 Q1468.51 2115.89 1468.51 2119.01 Q1468.51 2122.14 1470.4 2123.92 Q1472.33 2125.7 1475.66 2125.7 Q1478.99 2125.7 1480.91 2123.92 Q1482.83 2122.11 1482.83 2119.01 Q1482.83 2115.89 1480.91 2114.1 Q1479.02 2112.32 1475.66 2112.32 M1470.98 2110.33 Q1467.97 2109.59 1466.28 2107.53 Q1464.62 2105.47 1464.62 2102.51 Q1464.62 2098.36 1467.56 2095.96 Q1470.52 2093.55 1475.66 2093.55 Q1480.82 2093.55 1483.76 2095.96 Q1486.7 2098.36 1486.7 2102.51 Q1486.7 2105.47 1485.01 2107.53 Q1483.34 2109.59 1480.36 2110.33 Q1483.74 2111.12 1485.61 2113.41 Q1487.51 2115.7 1487.51 2119.01 Q1487.51 2124.04 1484.43 2126.72 Q1481.38 2129.41 1475.66 2129.41 Q1469.94 2129.41 1466.86 2126.72 Q1463.81 2124.04 1463.81 2119.01 Q1463.81 2115.7 1465.7 2113.41 Q1467.6 2111.12 1470.98 2110.33 M1469.27 2102.95 Q1469.27 2105.63 1470.94 2107.14 Q1472.63 2108.64 1475.66 2108.64 Q1478.67 2108.64 1480.36 2107.14 Q1482.07 2105.63 1482.07 2102.95 Q1482.07 2100.26 1480.36 2098.76 Q1478.67 2097.25 1475.66 2097.25 Q1472.63 2097.25 1470.94 2098.76 Q1469.27 2100.26 1469.27 2102.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1505.82 2097.25 Q1502.21 2097.25 1500.38 2100.82 Q1498.58 2104.36 1498.58 2111.49 Q1498.58 2118.6 1500.38 2122.16 Q1502.21 2125.7 1505.82 2125.7 Q1509.45 2125.7 1511.26 2122.16 Q1513.09 2118.6 1513.09 2111.49 Q1513.09 2104.36 1511.26 2100.82 Q1509.45 2097.25 1505.82 2097.25 M1505.82 2093.55 Q1511.63 2093.55 1514.69 2098.16 Q1517.76 2102.74 1517.76 2111.49 Q1517.76 2120.22 1514.69 2124.82 Q1511.63 2129.41 1505.82 2129.41 Q1500.01 2129.41 1496.93 2124.82 Q1493.88 2120.22 1493.88 2111.49 Q1493.88 2102.74 1496.93 2098.16 Q1500.01 2093.55 1505.82 2093.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1893.19 1886.82 L1882.1 1916.92 L1904.33 1916.92 L1893.19 1886.82 M1888.58 1878.76 L1897.85 1878.76 L1920.9 1939.24 L1912.4 1939.24 L1906.89 1923.73 L1879.62 1923.73 L1874.11 1939.24 L1865.49 1939.24 L1888.58 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1926.17 1893.87 L1961.57 1893.87 L1961.57 1900.68 L1933.54 1933.29 L1961.57 1933.29 L1961.57 1939.24 L1925.16 1939.24 L1925.16 1932.44 L1953.19 1899.83 L1926.17 1899.83 L1926.17 1893.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M1972.96 1893.87 L1980.41 1893.87 L1980.41 1939.24 L1972.96 1939.24 L1972.96 1893.87 M1972.96 1876.21 L1980.41 1876.21 L1980.41 1885.65 L1972.96 1885.65 L1972.96 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2031.33 1902.58 Q2034.13 1897.56 2038.01 1895.17 Q2041.9 1892.78 2047.17 1892.78 Q2054.26 1892.78 2058.11 1897.76 Q2061.96 1902.7 2061.96 1911.86 L2061.96 1939.24 L2054.46 1939.24 L2054.46 1912.1 Q2054.46 1905.58 2052.15 1902.42 Q2049.84 1899.26 2045.1 1899.26 Q2039.31 1899.26 2035.95 1903.11 Q2032.59 1906.96 2032.59 1913.6 L2032.59 1939.24 L2025.09 1939.24 L2025.09 1912.1 Q2025.09 1905.54 2022.78 1902.42 Q2020.47 1899.26 2015.65 1899.26 Q2009.94 1899.26 2006.58 1903.15 Q2003.22 1907 2003.22 1913.6 L2003.22 1939.24 L1995.72 1939.24 L1995.72 1893.87 L2003.22 1893.87 L2003.22 1900.92 Q2005.77 1896.75 2009.33 1894.76 Q2012.9 1892.78 2017.8 1892.78 Q2022.74 1892.78 2026.19 1895.29 Q2029.67 1897.8 2031.33 1902.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2076.05 1921.34 L2076.05 1893.87 L2083.51 1893.87 L2083.51 1921.05 Q2083.51 1927.5 2086.02 1930.74 Q2088.53 1933.94 2093.55 1933.94 Q2099.59 1933.94 2103.07 1930.09 Q2106.6 1926.24 2106.6 1919.6 L2106.6 1893.87 L2114.05 1893.87 L2114.05 1939.24 L2106.6 1939.24 L2106.6 1932.28 Q2103.88 1936.41 2100.28 1938.43 Q2096.71 1940.42 2091.97 1940.42 Q2084.15 1940.42 2080.1 1935.56 Q2076.05 1930.7 2076.05 1921.34 M2094.81 1892.78 L2094.81 1892.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2136.78 1880.99 L2136.78 1893.87 L2152.13 1893.87 L2152.13 1899.67 L2136.78 1899.67 L2136.78 1924.29 Q2136.78 1929.84 2138.27 1931.42 Q2139.81 1933 2144.47 1933 L2152.13 1933 L2152.13 1939.24 L2144.47 1939.24 Q2135.84 1939.24 2132.56 1936.04 Q2129.28 1932.8 2129.28 1924.29 L2129.28 1899.67 L2123.81 1899.67 L2123.81 1893.87 L2129.28 1893.87 L2129.28 1880.99 L2136.78 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2199.65 1911.86 L2199.65 1939.24 L2192.19 1939.24 L2192.19 1912.1 Q2192.19 1905.66 2189.68 1902.46 Q2187.17 1899.26 2182.15 1899.26 Q2176.11 1899.26 2172.63 1903.11 Q2169.14 1906.96 2169.14 1913.6 L2169.14 1939.24 L2161.65 1939.24 L2161.65 1876.21 L2169.14 1876.21 L2169.14 1900.92 Q2171.82 1896.83 2175.42 1894.8 Q2179.07 1892.78 2183.81 1892.78 Q2191.62 1892.78 2195.64 1897.64 Q2199.65 1902.46 2199.65 1911.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2258.79 1876.29 Q2253.36 1885.61 2250.73 1894.72 Q2248.09 1903.84 2248.09 1913.2 Q2248.09 1922.55 2250.73 1931.75 Q2253.4 1940.9 2258.79 1950.18 L2252.31 1950.18 Q2246.23 1940.66 2243.19 1931.46 Q2240.2 1922.27 2240.2 1913.2 Q2240.2 1904.16 2243.19 1895.01 Q2246.19 1885.85 2252.31 1876.29 L2258.79 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2303.11 1900.76 L2303.11 1876.21 L2310.56 1876.21 L2310.56 1939.24 L2303.11 1939.24 L2303.11 1932.44 Q2300.76 1936.49 2297.15 1938.47 Q2293.59 1940.42 2288.56 1940.42 Q2280.34 1940.42 2275.15 1933.85 Q2270.01 1927.29 2270.01 1916.6 Q2270.01 1905.9 2275.15 1899.34 Q2280.34 1892.78 2288.56 1892.78 Q2293.59 1892.78 2297.15 1894.76 Q2300.76 1896.71 2303.11 1900.76 M2277.71 1916.6 Q2277.71 1924.82 2281.07 1929.52 Q2284.47 1934.18 2290.39 1934.18 Q2296.3 1934.18 2299.7 1929.52 Q2303.11 1924.82 2303.11 1916.6 Q2303.11 1908.37 2299.7 1903.72 Q2296.3 1899.02 2290.39 1899.02 Q2284.47 1899.02 2281.07 1903.72 Q2277.71 1908.37 2277.71 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2364.72 1914.69 L2364.72 1918.34 L2330.45 1918.34 Q2330.94 1926.04 2335.07 1930.09 Q2339.24 1934.1 2346.65 1934.1 Q2350.95 1934.1 2354.96 1933.04 Q2359.01 1931.99 2362.98 1929.89 L2362.98 1936.93 Q2358.97 1938.64 2354.75 1939.53 Q2350.54 1940.42 2346.21 1940.42 Q2335.35 1940.42 2328.99 1934.1 Q2322.67 1927.78 2322.67 1917 Q2322.67 1905.86 2328.67 1899.34 Q2334.7 1892.78 2344.91 1892.78 Q2354.07 1892.78 2359.37 1898.69 Q2364.72 1904.57 2364.72 1914.69 M2357.27 1912.51 Q2357.19 1906.39 2353.82 1902.74 Q2350.5 1899.1 2344.99 1899.1 Q2338.75 1899.1 2334.99 1902.62 Q2331.26 1906.15 2330.69 1912.55 L2357.27 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2406.81 1916.03 Q2406.81 1907.93 2403.45 1903.47 Q2400.12 1899.02 2394.09 1899.02 Q2388.09 1899.02 2384.73 1903.47 Q2381.41 1907.93 2381.41 1916.03 Q2381.41 1924.09 2384.73 1928.55 Q2388.09 1933 2394.09 1933 Q2400.12 1933 2403.45 1928.55 Q2406.81 1924.09 2406.81 1916.03 M2414.26 1933.61 Q2414.26 1945.2 2409.12 1950.83 Q2403.97 1956.5 2393.36 1956.5 Q2389.43 1956.5 2385.95 1955.89 Q2382.46 1955.32 2379.18 1954.11 L2379.18 1946.86 Q2382.46 1948.64 2385.66 1949.49 Q2388.86 1950.34 2392.19 1950.34 Q2399.52 1950.34 2403.16 1946.49 Q2406.81 1942.69 2406.81 1934.95 L2406.81 1931.26 Q2404.5 1935.27 2400.89 1937.26 Q2397.29 1939.24 2392.27 1939.24 Q2383.92 1939.24 2378.82 1932.88 Q2373.71 1926.52 2373.71 1916.03 Q2373.71 1905.5 2378.82 1899.14 Q2383.92 1892.78 2392.27 1892.78 Q2397.29 1892.78 2400.89 1894.76 Q2404.5 1896.75 2406.81 1900.76 L2406.81 1893.87 L2414.26 1893.87 L2414.26 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip210)" d="M2428.44 1876.29 L2434.92 1876.29 Q2441 1885.85 2444 1895.01 Q2447.03 1904.16 2447.03 1913.2 Q2447.03 1922.27 2444 1931.46 Q2441 1940.66 2434.92 1950.18 L2428.44 1950.18 Q2433.83 1940.9 2436.46 1931.75 Q2439.14 1922.55 2439.14 1913.2 Q2439.14 1903.84 2436.46 1894.72 Q2433.83 1885.61 2428.44 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip217)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,2038.4 1595.77,2038.52 1598.01,2038.64 1600.25,2038.77 1602.5,2038.92 1604.74,2039.07 1606.98,2039.24 1609.22,2039.42 1611.46,2039.61 1613.71,2039.82 1615.95,2040.04 1618.19,2040.28 1620.43,2040.54 1622.67,2040.82 1624.92,2041.12 1627.16,2041.44 1629.4,2041.78 1631.64,2042.15 1633.88,2042.55 1636.13,2042.97 1638.37,2043.43 1640.61,2043.92 1642.85,2044.43 1645.09,2044.99 1647.34,2045.58 1649.58,2046.21 1651.82,2046.87 1654.06,2047.58 1656.3,2048.33 1658.55,2049.13 1660.79,2049.96 1663.03,2050.84 1665.27,2051.77 1667.51,2052.74 1669.76,2053.76 1672,2054.82 1674.24,2055.93 1676.48,2057.08 1678.72,2058.27 1680.97,2059.49 1683.21,2060.76 1685.45,2062.06 1687.69,2063.39 1689.93,2064.75 1692.17,2066.13 1694.42,2067.53 1696.66,2068.95 1698.9,2070.38 1701.14,2071.82 1703.38,2073.26 1705.63,2074.7 1707.87,2076.13 1710.11,2077.56 1712.35,2078.97 1714.59,2080.37 1716.84,2081.75 1719.08,2083.11 1721.32,2084.44 1723.56,2085.75 1725.8,2087.03 1728.05,2088.27 1730.29,2089.49 1732.53,2090.67 1734.77,2091.82 1737.01,2092.94 1739.26,2094.03 1741.5,2095.08 1743.74,2096.09 1745.98,2097.08 1748.22,2098.03 1750.47,2098.95 1752.71,2099.85 1754.95,2100.71 1757.19,2101.54 1759.43,2102.35 1761.68,2103.13 1763.92,2103.89 1766.16,2104.62 1768.4,2105.33 1770.64,2106.02 1772.89,2106.69 1775.13,2107.34 1777.37,2107.97 1779.61,2108.59 1781.85,2109.19 1784.1,2109.77 1786.34,2110.35 1788.58,2110.91 1790.82,2111.46 1793.06,2112 1795.31,2112.53 1797.55,2113.06 1799.79,2113.57 1802.03,2114.09 1804.27,2114.59 1806.51,2115.1 1808.76,2115.6 1811,2116.1 1813.24,2116.6 1815.48,2117.09 1817.72,2117.59 1819.97,2118.1 1822.21,2118.6 1824.45,2119.11 1826.69,2119.63 1828.93,2120.15 1831.18,2120.68 1833.42,2121.22 1835.66,2121.77 1837.9,2122.33 1840.14,2122.9 1842.39,2123.49 1844.63,2124.09 1846.87,2124.71 1849.11,2125.35 1851.35,2126 1853.6,2126.67 1855.84,2127.37 1858.08,2128.09 1860.32,2128.83 1862.56,2129.6 1864.81,2130.4 1867.05,2131.22 1869.29,2132.08 1871.53,2132.96 1873.77,2133.88 1876.02,2134.82 1878.26,2135.8 1880.5,2136.82 1882.74,2137.86 1884.98,2138.94 1887.23,2140.05 1889.47,2141.2 1891.71,2142.38 1893.95,2143.59 1896.19,2144.82 1898.44,2146.09 1900.68,2147.38 1902.92,2148.69 1905.16,2150.02 1907.4,2151.37 1909.64,2152.74 1911.89,2154.12 1914.13,2155.5 1916.37,2156.89 1918.61,2158.28 1920.85,2159.67 1923.1,2161.06 1925.34,2162.43 1927.58,2163.8 1929.82,2165.15 1932.06,2166.49 1934.31,2167.8 1936.55,2169.1 1938.79,2170.37 1941.03,2171.63 1943.27,2172.85 1945.52,2174.06 1947.76,2175.23 1950,2176.38 1952.24,2177.51 1954.48,2178.61 1956.73,2179.69 1958.97,2180.74 1961.21,2181.77 1963.45,2182.78 1965.69,2183.77 1967.94,2184.73 1970.18,2185.68 1972.42,2186.61 1974.66,2187.53 1976.9,2188.43 1979.15,2189.32 1981.39,2190.2 1983.63,2191.07 1985.87,2191.93 1988.11,2192.78 1990.36,2193.63 1992.6,2194.48 1994.84,2195.33 1997.08,2196.18 1999.32,2197.03 2001.57,2197.88 2003.81,2198.74 2006.05,2199.61 2008.29,2200.48 2010.53,2201.37 2012.78,2202.27 2015.02,2203.19 2017.26,2204.12 2019.5,2205.06 2021.74,2206.03 2023.98,2207.02 2026.23,2208.02 2028.47,2209.05 2030.71,2210.11 2032.95,2211.18 2035.19,2212.29 2037.44,2213.41 2039.68,2214.57 2041.92,2215.75 2044.16,2216.95 2046.4,2218.18 2048.65,2219.43 2050.89,2220.71 2053.13,2222.01 2055.37,2223.33 2057.61,2224.66 2059.86,2226.02 2062.1,2227.39 2064.34,2228.77 2066.58,2230.15 2068.82,2231.55 2071.07,2232.95 2073.31,2234.35 2075.55,2235.75 2077.79,2237.14 2080.03,2238.53 2082.28,2239.91 2084.52,2241.28 2086.76,2242.63 2089,2243.97 2091.24,2245.3 2093.49,2246.61 2095.73,2247.9 2097.97,2249.17 2100.21,2250.43 2102.45,2251.67 2104.7,2252.88 2106.94,2254.09 2109.18,2255.27 2111.42,2256.44 2113.66,2257.59 2115.91,2258.72 2118.15,2259.85 2120.39,2260.96 2122.63,2262.06 2124.87,2263.15 2127.11,2264.24 2129.36,2265.32 2131.6,2266.39 2133.84,2267.46 2136.08,2268.54 2138.32,2269.61 2140.57,2270.69 2142.81,2271.77 2145.05,2272.86 2147.29,2273.95 2149.53,2275.06 2151.78,2276.17 2154.02,2277.3 2156.26,2278.44 2158.5,2279.6 2160.74,2280.78 2162.99,2281.97 2165.23,2283.18 2167.47,2284.41 2169.71,2285.65 2171.95,2286.92 2174.2,2288.21 2176.44,2289.51 2178.68,2290.83 2180.92,2292.18 2183.16,2293.53 2185.41,2294.91 2187.65,2296.3 2189.89,2297.7 2192.13,2299.11 2194.37,2300.53 2196.62,2301.96 2198.86,2303.4 2201.1,2304.84 2203.34,2306.28 2205.58,2307.72 2207.83,2309.16 2210.07,2310.59 2212.31,2312.02 2214.55,2313.44 2216.79,2314.85 2219.04,2316.25 2221.28,2317.64 2223.52,2319.02 2225.76,2320.39 2228,2321.75 2230.25,2323.09 2232.49,2324.43 2234.73,2325.75 2236.97,2327.06 2239.21,2328.36 2241.45,2329.65 2243.7,2330.93 2245.94,2332.21 2248.18,2333.47 2250.42,2334.74 2252.66,2336 2254.91,2337.26 2257.15,2338.51 2259.39,2339.77 2261.63,2341.03 2263.87,2342.3 2266.12,2343.57 2268.36,2344.84 2270.6,2346.13 2272.84,2347.42 2275.08,2348.72 2277.33,2350.03 2279.57,2351.36 2281.81,2352.7 2284.05,2354.05 2286.29,2355.41 2288.54,2356.79 2290.78,2358.18 2293.02,2359.59 2295.26,2361.01 2297.5,2362.44 2299.75,2363.89 2301.99,2365.35 2304.23,2366.82 2306.47,2368.3 2308.71,2369.78 2310.96,2371.28 2313.2,2372.78 2315.44,2374.29 2317.68,2375.79 2319.92,2377.3 2322.17,2378.82 2324.41,2380.33 2326.65,2381.83 2328.89,2383.34 2331.13,2384.84 2333.38,2386.34 2335.62,2387.83 2337.86,2389.31 2340.1,2390.79 2342.34,2392.26 2344.58,2393.72 2346.83,2395.18 2349.07,2396.63 2351.31,2398.08 2353.55,2399.52 2355.79,2400.95 2358.04,2402.38 2360.28,2403.81 2362.52,2405.24 2364.76,2406.66 2367,2408.08 2369.25,2409.51 2371.49,2410.93 2373.73,2412.36 2375.97,2413.8 2378.21,2415.23 2380.46,2416.67 2382.7,2418.12 2384.94,2419.58 2387.18,2421.04 2389.42,2422.51 2391.67,2424 2393.91,2425.49 2396.15,2426.99 2398.39,2428.5 2400.63,2430.02 2402.88,2431.54 2405.12,2433.08 2407.36,2434.63 2409.6,2436.18 2411.84,2437.75 2414.09,2439.32 2416.33,2440.89 2418.57,2442.48 2420.81,2444.06 2423.05,2445.65 2425.3,2447.24 2427.54,2448.84 2429.78,2450.43 2432.02,2452.03 2434.26,2453.62 2436.51,2455.22 2438.75,2456.81 2440.99,2458.4 2443.23,2459.99 2445.47,2461.57 2447.72,2463.15 2449.96,2464.73 2452.2,2466.3 2454.44,2467.87 2456.68,2469.44 2458.92,2471.01 2461.17,2472.57 2463.41,2474.13 2465.65,2475.69 2467.89,2477.25 2470.13,2478.81 2472.38,2480.37 2474.62,2481.93 2476.86,2483.5 2479.1,2485.06 2481.34,2486.63 2483.59,2488.2 2485.83,2489.78 2488.07,2491.36 2490.31,2492.95 2492.55,2494.54 2494.8,2496.14 2497.04,2497.74 2499.28,2499.35 2501.52,2500.96 2503.76,2502.58 2506.01,2504.21 2508.25,2505.84 2510.49,2507.47 2512.73,2509.11 2514.97,2510.76 2517.22,2512.4 2519.46,2514.05 2521.7,2515.71 2523.94,2517.36 2526.18,2519.02 2528.43,2520.67 2530.67,2522.33 2532.91,2523.99 2535.15,2525.64 2537.39,2527.3 2539.64,2528.95 2541.88,2530.6 2544.12,2532.25 2546.36,2533.89 2548.6,2535.53 2550.85,2537.17 2553.09,2538.81 2555.33,2540.44 2557.57,2542.07 2559.81,2543.7 2562.05,2545.33 2564.3,2546.95 2566.54,2548.57 2568.78,2550.2 2571.02,2551.81 2573.26,2553.43 2575.51,2555.05 2577.75,2556.67 2579.99,2558.29 2582.23,2559.91 2584.47,2561.52 2586.72,2563.14 2588.96,2564.76 2591.2,2566.39 2593.44,2568.01 2595.68,2569.63 2597.93,2571.26 2600.17,2572.88 2602.41,2574.51 2604.65,2576.13 2606.89,2577.76 2609.14,2579.38 2611.38,2581.01 2613.62,2582.63 2615.86,2584.25 2618.1,2585.86 2620.35,2587.47 2622.59,2589.08 2624.83,2590.68 2627.07,2592.27 2629.31,2593.86 2631.56,2595.44 2633.8,2597 2636.04,2598.56 2638.28,2600.11 2640.52,2601.65 2642.77,2603.17 2645.01,2604.68 2647.25,2606.18 2649.49,2607.67 2651.73,2609.14 2653.98,2610.59 2656.22,2612.03 2658.46,2613.45 2660.7,2614.86 2662.94,2616.25 2665.19,2617.62 2667.43,2618.97 2669.67,2620.3 2671.91,2621.61 2674.15,2622.9 2676.39,2624.16 2678.64,2625.4 2680.88,2626.61 2683.12,2627.79 2685.36,2628.95 2687.6,2630.06 2689.85,2631.15 2692.09,2632.18 2694.33,2633.18 2696.57,2634.12 2698.81,2635.01 2701.06,2635.83 2703.3,2636.59 2705.54,2637.27 2707.78,2637.86 2710.02,2638.36 2712.27,2638.74 2714.51,2639.01 2716.75,2639.15 2718.99,2639.13 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1593.53,2657.17 1593.53,2020.38 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1873.77,2657.17 1873.77,2020.38 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2154.02,2657.17 2154.02,2020.38 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2434.26,2657.17 2434.26,2020.38 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="2714.51,2657.17 2714.51,2020.38 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2657.17 2752.76,2657.17 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1593.53,2657.17 1593.53,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1873.77,2657.17 1873.77,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2154.02,2657.17 2154.02,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2434.26,2657.17 2434.26,2638.27 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="2714.51,2657.17 2714.51,2638.27 "/>
+<path clip-path="url(#clip030)" d="M1593.53 2691.29 Q1589.92 2691.29 1588.09 2694.85 Q1586.28 2698.39 1586.28 2705.52 Q1586.28 2712.63 1588.09 2716.19 Q1589.92 2719.74 1593.53 2719.74 Q1597.16 2719.74 1598.97 2716.19 Q1600.8 2712.63 1600.8 2705.52 Q1600.8 2698.39 1598.97 2694.85 Q1597.16 2691.29 1593.53 2691.29 M1593.53 2687.58 Q1599.34 2687.58 1602.39 2692.19 Q1605.47 2696.77 1605.47 2705.52 Q1605.47 2714.25 1602.39 2718.86 Q1599.34 2723.44 1593.53 2723.44 Q1587.72 2723.44 1584.64 2718.86 Q1581.58 2714.25 1581.58 2705.52 Q1581.58 2696.77 1584.64 2692.19 Q1587.72 2687.58 1593.53 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1833.39 2688.21 L1851.75 2688.21 L1851.75 2692.14 L1837.67 2692.14 L1837.67 2700.62 Q1838.69 2700.27 1839.71 2700.11 Q1840.73 2699.92 1841.75 2699.92 Q1847.54 2699.92 1850.91 2703.09 Q1854.29 2706.26 1854.29 2711.68 Q1854.29 2717.26 1850.82 2720.36 Q1847.35 2723.44 1841.03 2723.44 Q1838.85 2723.44 1836.59 2723.07 Q1834.34 2722.7 1831.93 2721.96 L1831.93 2717.26 Q1834.02 2718.39 1836.24 2718.95 Q1838.46 2719.5 1840.94 2719.5 Q1844.94 2719.5 1847.28 2717.4 Q1849.62 2715.29 1849.62 2711.68 Q1849.62 2708.07 1847.28 2705.96 Q1844.94 2703.86 1840.94 2703.86 Q1839.06 2703.86 1837.19 2704.27 Q1835.34 2704.69 1833.39 2705.57 L1833.39 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1873.51 2691.29 Q1869.9 2691.29 1868.07 2694.85 Q1866.26 2698.39 1866.26 2705.52 Q1866.26 2712.63 1868.07 2716.19 Q1869.9 2719.74 1873.51 2719.74 Q1877.14 2719.74 1878.95 2716.19 Q1880.78 2712.63 1880.78 2705.52 Q1880.78 2698.39 1878.95 2694.85 Q1877.14 2691.29 1873.51 2691.29 M1873.51 2687.58 Q1879.32 2687.58 1882.37 2692.19 Q1885.45 2696.77 1885.45 2705.52 Q1885.45 2714.25 1882.37 2718.86 Q1879.32 2723.44 1873.51 2723.44 Q1867.7 2723.44 1864.62 2718.86 Q1861.56 2714.25 1861.56 2705.52 Q1861.56 2696.77 1864.62 2692.19 Q1867.7 2687.58 1873.51 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1903.67 2691.29 Q1900.06 2691.29 1898.23 2694.85 Q1896.42 2698.39 1896.42 2705.52 Q1896.42 2712.63 1898.23 2716.19 Q1900.06 2719.74 1903.67 2719.74 Q1907.3 2719.74 1909.11 2716.19 Q1910.94 2712.63 1910.94 2705.52 Q1910.94 2698.39 1909.11 2694.85 Q1907.3 2691.29 1903.67 2691.29 M1903.67 2687.58 Q1909.48 2687.58 1912.53 2692.19 Q1915.61 2696.77 1915.61 2705.52 Q1915.61 2714.25 1912.53 2718.86 Q1909.48 2723.44 1903.67 2723.44 Q1897.86 2723.44 1894.78 2718.86 Q1891.72 2714.25 1891.72 2705.52 Q1891.72 2696.77 1894.78 2692.19 Q1897.86 2687.58 1903.67 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2098.54 2718.83 L2106.18 2718.83 L2106.18 2692.47 L2097.87 2694.13 L2097.87 2689.87 L2106.14 2688.21 L2110.81 2688.21 L2110.81 2718.83 L2118.45 2718.83 L2118.45 2722.77 L2098.54 2722.77 L2098.54 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2137.9 2691.29 Q2134.28 2691.29 2132.46 2694.85 Q2130.65 2698.39 2130.65 2705.52 Q2130.65 2712.63 2132.46 2716.19 Q2134.28 2719.74 2137.9 2719.74 Q2141.53 2719.74 2143.34 2716.19 Q2145.16 2712.63 2145.16 2705.52 Q2145.16 2698.39 2143.34 2694.85 Q2141.53 2691.29 2137.9 2691.29 M2137.9 2687.58 Q2143.71 2687.58 2146.76 2692.19 Q2149.84 2696.77 2149.84 2705.52 Q2149.84 2714.25 2146.76 2718.86 Q2143.71 2723.44 2137.9 2723.44 Q2132.09 2723.44 2129.01 2718.86 Q2125.95 2714.25 2125.95 2705.52 Q2125.95 2696.77 2129.01 2692.19 Q2132.09 2687.58 2137.9 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2168.06 2691.29 Q2164.45 2691.29 2162.62 2694.85 Q2160.81 2698.39 2160.81 2705.52 Q2160.81 2712.63 2162.62 2716.19 Q2164.45 2719.74 2168.06 2719.74 Q2171.69 2719.74 2173.5 2716.19 Q2175.33 2712.63 2175.33 2705.52 Q2175.33 2698.39 2173.5 2694.85 Q2171.69 2691.29 2168.06 2691.29 M2168.06 2687.58 Q2173.87 2687.58 2176.92 2692.19 Q2180 2696.77 2180 2705.52 Q2180 2714.25 2176.92 2718.86 Q2173.87 2723.44 2168.06 2723.44 Q2162.25 2723.44 2159.17 2718.86 Q2156.11 2714.25 2156.11 2705.52 Q2156.11 2696.77 2159.17 2692.19 Q2162.25 2687.58 2168.06 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2198.22 2691.29 Q2194.61 2691.29 2192.78 2694.85 Q2190.97 2698.39 2190.97 2705.52 Q2190.97 2712.63 2192.78 2716.19 Q2194.61 2719.74 2198.22 2719.74 Q2201.85 2719.74 2203.66 2716.19 Q2205.49 2712.63 2205.49 2705.52 Q2205.49 2698.39 2203.66 2694.85 Q2201.85 2691.29 2198.22 2691.29 M2198.22 2687.58 Q2204.03 2687.58 2207.09 2692.19 Q2210.16 2696.77 2210.16 2705.52 Q2210.16 2714.25 2207.09 2718.86 Q2204.03 2723.44 2198.22 2723.44 Q2192.41 2723.44 2189.33 2718.86 Q2186.28 2714.25 2186.28 2705.52 Q2186.28 2696.77 2189.33 2692.19 Q2192.41 2687.58 2198.22 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2378.79 2718.83 L2386.43 2718.83 L2386.43 2692.47 L2378.12 2694.13 L2378.12 2689.87 L2386.38 2688.21 L2391.06 2688.21 L2391.06 2718.83 L2398.7 2718.83 L2398.7 2722.77 L2378.79 2722.77 L2378.79 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2408.19 2688.21 L2426.54 2688.21 L2426.54 2692.14 L2412.47 2692.14 L2412.47 2700.62 Q2413.49 2700.27 2414.51 2700.11 Q2415.52 2699.92 2416.54 2699.92 Q2422.33 2699.92 2425.71 2703.09 Q2429.09 2706.26 2429.09 2711.68 Q2429.09 2717.26 2425.62 2720.36 Q2422.15 2723.44 2415.83 2723.44 Q2413.65 2723.44 2411.38 2723.07 Q2409.14 2722.7 2406.73 2721.96 L2406.73 2717.26 Q2408.81 2718.39 2411.03 2718.95 Q2413.26 2719.5 2415.73 2719.5 Q2419.74 2719.5 2422.08 2717.4 Q2424.41 2715.29 2424.41 2711.68 Q2424.41 2708.07 2422.08 2705.96 Q2419.74 2703.86 2415.73 2703.86 Q2413.86 2703.86 2411.98 2704.27 Q2410.13 2704.69 2408.19 2705.57 L2408.19 2688.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2448.3 2691.29 Q2444.69 2691.29 2442.86 2694.85 Q2441.06 2698.39 2441.06 2705.52 Q2441.06 2712.63 2442.86 2716.19 Q2444.69 2719.74 2448.3 2719.74 Q2451.94 2719.74 2453.74 2716.19 Q2455.57 2712.63 2455.57 2705.52 Q2455.57 2698.39 2453.74 2694.85 Q2451.94 2691.29 2448.3 2691.29 M2448.3 2687.58 Q2454.11 2687.58 2457.17 2692.19 Q2460.25 2696.77 2460.25 2705.52 Q2460.25 2714.25 2457.17 2718.86 Q2454.11 2723.44 2448.3 2723.44 Q2442.49 2723.44 2439.41 2718.86 Q2436.36 2714.25 2436.36 2705.52 Q2436.36 2696.77 2439.41 2692.19 Q2442.49 2687.58 2448.3 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2478.46 2691.29 Q2474.85 2691.29 2473.02 2694.85 Q2471.22 2698.39 2471.22 2705.52 Q2471.22 2712.63 2473.02 2716.19 Q2474.85 2719.74 2478.46 2719.74 Q2482.1 2719.74 2483.9 2716.19 Q2485.73 2712.63 2485.73 2705.52 Q2485.73 2698.39 2483.9 2694.85 Q2482.1 2691.29 2478.46 2691.29 M2478.46 2687.58 Q2484.27 2687.58 2487.33 2692.19 Q2490.41 2696.77 2490.41 2705.52 Q2490.41 2714.25 2487.33 2718.86 Q2484.27 2723.44 2478.46 2723.44 Q2472.65 2723.44 2469.58 2718.86 Q2466.52 2714.25 2466.52 2705.52 Q2466.52 2696.77 2469.58 2692.19 Q2472.65 2687.58 2478.46 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2663.12 2718.83 L2679.44 2718.83 L2679.44 2722.77 L2657.49 2722.77 L2657.49 2718.83 Q2660.16 2716.08 2664.74 2711.45 Q2669.35 2706.8 2670.53 2705.45 Q2672.77 2702.93 2673.65 2701.19 Q2674.55 2699.43 2674.55 2697.74 Q2674.55 2694.99 2672.61 2693.25 Q2670.69 2691.52 2667.59 2691.52 Q2665.39 2691.52 2662.93 2692.28 Q2660.5 2693.05 2657.73 2694.6 L2657.73 2689.87 Q2660.55 2688.74 2663 2688.16 Q2665.46 2687.58 2667.49 2687.58 Q2672.86 2687.58 2676.06 2690.27 Q2679.25 2692.95 2679.25 2697.44 Q2679.25 2699.57 2678.44 2701.49 Q2677.66 2703.39 2675.55 2705.99 Q2674.97 2706.66 2671.87 2709.87 Q2668.77 2713.07 2663.12 2718.83 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2699.25 2691.29 Q2695.64 2691.29 2693.81 2694.85 Q2692.01 2698.39 2692.01 2705.52 Q2692.01 2712.63 2693.81 2716.19 Q2695.64 2719.74 2699.25 2719.74 Q2702.89 2719.74 2704.69 2716.19 Q2706.52 2712.63 2706.52 2705.52 Q2706.52 2698.39 2704.69 2694.85 Q2702.89 2691.29 2699.25 2691.29 M2699.25 2687.58 Q2705.06 2687.58 2708.12 2692.19 Q2711.2 2696.77 2711.2 2705.52 Q2711.2 2714.25 2708.12 2718.86 Q2705.06 2723.44 2699.25 2723.44 Q2693.44 2723.44 2690.36 2718.86 Q2687.31 2714.25 2687.31 2705.52 Q2687.31 2696.77 2690.36 2692.19 Q2693.44 2687.58 2699.25 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2729.42 2691.29 Q2725.8 2691.29 2723.98 2694.85 Q2722.17 2698.39 2722.17 2705.52 Q2722.17 2712.63 2723.98 2716.19 Q2725.8 2719.74 2729.42 2719.74 Q2733.05 2719.74 2734.86 2716.19 Q2736.68 2712.63 2736.68 2705.52 Q2736.68 2698.39 2734.86 2694.85 Q2733.05 2691.29 2729.42 2691.29 M2729.42 2687.58 Q2735.23 2687.58 2738.28 2692.19 Q2741.36 2696.77 2741.36 2705.52 Q2741.36 2714.25 2738.28 2718.86 Q2735.23 2723.44 2729.42 2723.44 Q2723.61 2723.44 2720.53 2718.86 Q2717.47 2714.25 2717.47 2705.52 Q2717.47 2696.77 2720.53 2692.19 Q2723.61 2687.58 2729.42 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2759.58 2691.29 Q2755.97 2691.29 2754.14 2694.85 Q2752.33 2698.39 2752.33 2705.52 Q2752.33 2712.63 2754.14 2716.19 Q2755.97 2719.74 2759.58 2719.74 Q2763.21 2719.74 2765.02 2716.19 Q2766.85 2712.63 2766.85 2705.52 Q2766.85 2698.39 2765.02 2694.85 Q2763.21 2691.29 2759.58 2691.29 M2759.58 2687.58 Q2765.39 2687.58 2768.44 2692.19 Q2771.52 2696.77 2771.52 2705.52 Q2771.52 2714.25 2768.44 2718.86 Q2765.39 2723.44 2759.58 2723.44 Q2753.77 2723.44 2750.69 2718.86 Q2747.63 2714.25 2747.63 2705.52 Q2747.63 2696.77 2750.69 2692.19 Q2753.77 2687.58 2759.58 2687.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2549.77 2752.76,2549.77 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2403.67 2752.76,2403.67 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2257.56 2752.76,2257.56 "/>
+<polyline clip-path="url(#clip037)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:2; stroke-opacity:0.1; fill:none" points="1559.76,2111.45 2752.76,2111.45 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2657.17 1559.76,2020.38 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2549.77 1578.66,2549.77 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2403.67 1578.66,2403.67 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2257.56 1578.66,2257.56 "/>
+<polyline clip-path="url(#clip030)" style="stroke:#000000; stroke-linecap:round; stroke-linejoin:round; stroke-width:4; stroke-opacity:1; fill:none" points="1559.76,2111.45 1578.66,2111.45 "/>
+<path clip-path="url(#clip030)" d="M1469.69 2563.12 L1486.01 2563.12 L1486.01 2567.05 L1464.06 2567.05 L1464.06 2563.12 Q1466.72 2560.36 1471.31 2555.74 Q1475.91 2551.08 1477.09 2549.74 Q1479.34 2547.22 1480.22 2545.48 Q1481.12 2543.72 1481.12 2542.03 Q1481.12 2539.28 1479.18 2537.54 Q1477.26 2535.8 1474.15 2535.8 Q1471.95 2535.8 1469.5 2536.57 Q1467.07 2537.33 1464.29 2538.88 L1464.29 2534.16 Q1467.12 2533.03 1469.57 2532.45 Q1472.02 2531.87 1474.06 2531.87 Q1479.43 2531.87 1482.63 2534.55 Q1485.82 2537.24 1485.82 2541.73 Q1485.82 2543.86 1485.01 2545.78 Q1484.22 2547.68 1482.12 2550.27 Q1481.54 2550.94 1478.44 2554.16 Q1475.33 2557.36 1469.69 2563.12 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 2535.57 Q1502.21 2535.57 1500.38 2539.14 Q1498.58 2542.68 1498.58 2549.81 Q1498.58 2556.92 1500.38 2560.48 Q1502.21 2564.02 1505.82 2564.02 Q1509.45 2564.02 1511.26 2560.48 Q1513.09 2556.92 1513.09 2549.81 Q1513.09 2542.68 1511.26 2539.14 Q1509.45 2535.57 1505.82 2535.57 M1505.82 2531.87 Q1511.63 2531.87 1514.69 2536.48 Q1517.76 2541.06 1517.76 2549.81 Q1517.76 2558.54 1514.69 2563.14 Q1511.63 2567.73 1505.82 2567.73 Q1500.01 2567.73 1496.93 2563.14 Q1493.88 2558.54 1493.88 2549.81 Q1493.88 2541.06 1496.93 2536.48 Q1500.01 2531.87 1505.82 2531.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1478.51 2390.46 L1466.7 2408.91 L1478.51 2408.91 L1478.51 2390.46 M1477.28 2386.39 L1483.16 2386.39 L1483.16 2408.91 L1488.09 2408.91 L1488.09 2412.8 L1483.16 2412.8 L1483.16 2420.95 L1478.51 2420.95 L1478.51 2412.8 L1462.9 2412.8 L1462.9 2408.29 L1477.28 2386.39 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 2389.47 Q1502.21 2389.47 1500.38 2393.03 Q1498.58 2396.57 1498.58 2403.7 Q1498.58 2410.81 1500.38 2414.37 Q1502.21 2417.92 1505.82 2417.92 Q1509.45 2417.92 1511.26 2414.37 Q1513.09 2410.81 1513.09 2403.7 Q1513.09 2396.57 1511.26 2393.03 Q1509.45 2389.47 1505.82 2389.47 M1505.82 2385.76 Q1511.63 2385.76 1514.69 2390.37 Q1517.76 2394.95 1517.76 2403.7 Q1517.76 2412.43 1514.69 2417.04 Q1511.63 2421.62 1505.82 2421.62 Q1500.01 2421.62 1496.93 2417.04 Q1493.88 2412.43 1493.88 2403.7 Q1493.88 2394.95 1496.93 2390.37 Q1500.01 2385.76 1505.82 2385.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1476.24 2255.7 Q1473.09 2255.7 1471.24 2257.85 Q1469.41 2260 1469.41 2263.75 Q1469.41 2267.48 1471.24 2269.66 Q1473.09 2271.81 1476.24 2271.81 Q1479.39 2271.81 1481.21 2269.66 Q1483.07 2267.48 1483.07 2263.75 Q1483.07 2260 1481.21 2257.85 Q1479.39 2255.7 1476.24 2255.7 M1485.52 2241.05 L1485.52 2245.3 Q1483.76 2244.47 1481.95 2244.03 Q1480.17 2243.59 1478.41 2243.59 Q1473.78 2243.59 1471.33 2246.72 Q1468.9 2249.84 1468.55 2256.16 Q1469.92 2254.15 1471.98 2253.08 Q1474.04 2251.99 1476.52 2251.99 Q1481.72 2251.99 1484.73 2255.17 Q1487.77 2258.31 1487.77 2263.75 Q1487.77 2269.08 1484.62 2272.29 Q1481.47 2275.51 1476.24 2275.51 Q1470.24 2275.51 1467.07 2270.93 Q1463.9 2266.32 1463.9 2257.6 Q1463.9 2249.4 1467.79 2244.54 Q1471.68 2239.66 1478.23 2239.66 Q1479.99 2239.66 1481.77 2240 Q1483.58 2240.35 1485.52 2241.05 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 2243.36 Q1502.21 2243.36 1500.38 2246.92 Q1498.58 2250.47 1498.58 2257.6 Q1498.58 2264.7 1500.38 2268.27 Q1502.21 2271.81 1505.82 2271.81 Q1509.45 2271.81 1511.26 2268.27 Q1513.09 2264.7 1513.09 2257.6 Q1513.09 2250.47 1511.26 2246.92 Q1509.45 2243.36 1505.82 2243.36 M1505.82 2239.66 Q1511.63 2239.66 1514.69 2244.26 Q1517.76 2248.85 1517.76 2257.6 Q1517.76 2266.32 1514.69 2270.93 Q1511.63 2275.51 1505.82 2275.51 Q1500.01 2275.51 1496.93 2270.93 Q1493.88 2266.32 1493.88 2257.6 Q1493.88 2248.85 1496.93 2244.26 Q1500.01 2239.66 1505.82 2239.66 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1475.66 2112.32 Q1472.33 2112.32 1470.4 2114.1 Q1468.51 2115.89 1468.51 2119.01 Q1468.51 2122.14 1470.4 2123.92 Q1472.33 2125.7 1475.66 2125.7 Q1478.99 2125.7 1480.91 2123.92 Q1482.83 2122.11 1482.83 2119.01 Q1482.83 2115.89 1480.91 2114.1 Q1479.02 2112.32 1475.66 2112.32 M1470.98 2110.33 Q1467.97 2109.59 1466.28 2107.53 Q1464.62 2105.47 1464.62 2102.51 Q1464.62 2098.36 1467.56 2095.96 Q1470.52 2093.55 1475.66 2093.55 Q1480.82 2093.55 1483.76 2095.96 Q1486.7 2098.36 1486.7 2102.51 Q1486.7 2105.47 1485.01 2107.53 Q1483.34 2109.59 1480.36 2110.33 Q1483.74 2111.12 1485.61 2113.41 Q1487.51 2115.7 1487.51 2119.01 Q1487.51 2124.04 1484.43 2126.72 Q1481.38 2129.41 1475.66 2129.41 Q1469.94 2129.41 1466.86 2126.72 Q1463.81 2124.04 1463.81 2119.01 Q1463.81 2115.7 1465.7 2113.41 Q1467.6 2111.12 1470.98 2110.33 M1469.27 2102.95 Q1469.27 2105.63 1470.94 2107.14 Q1472.63 2108.64 1475.66 2108.64 Q1478.67 2108.64 1480.36 2107.14 Q1482.07 2105.63 1482.07 2102.95 Q1482.07 2100.26 1480.36 2098.76 Q1478.67 2097.25 1475.66 2097.25 Q1472.63 2097.25 1470.94 2098.76 Q1469.27 2100.26 1469.27 2102.95 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1505.82 2097.25 Q1502.21 2097.25 1500.38 2100.82 Q1498.58 2104.36 1498.58 2111.49 Q1498.58 2118.6 1500.38 2122.16 Q1502.21 2125.7 1505.82 2125.7 Q1509.45 2125.7 1511.26 2122.16 Q1513.09 2118.6 1513.09 2111.49 Q1513.09 2104.36 1511.26 2100.82 Q1509.45 2097.25 1505.82 2097.25 M1505.82 2093.55 Q1511.63 2093.55 1514.69 2098.16 Q1517.76 2102.74 1517.76 2111.49 Q1517.76 2120.22 1514.69 2124.82 Q1511.63 2129.41 1505.82 2129.41 Q1500.01 2129.41 1496.93 2124.82 Q1493.88 2120.22 1493.88 2111.49 Q1493.88 2102.74 1496.93 2098.16 Q1500.01 2093.55 1505.82 2093.55 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1893.19 1886.82 L1882.1 1916.92 L1904.33 1916.92 L1893.19 1886.82 M1888.58 1878.76 L1897.85 1878.76 L1920.9 1939.24 L1912.4 1939.24 L1906.89 1923.73 L1879.62 1923.73 L1874.11 1939.24 L1865.49 1939.24 L1888.58 1878.76 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1926.17 1893.87 L1961.57 1893.87 L1961.57 1900.68 L1933.54 1933.29 L1961.57 1933.29 L1961.57 1939.24 L1925.16 1939.24 L1925.16 1932.44 L1953.19 1899.83 L1926.17 1899.83 L1926.17 1893.87 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M1972.96 1893.87 L1980.41 1893.87 L1980.41 1939.24 L1972.96 1939.24 L1972.96 1893.87 M1972.96 1876.21 L1980.41 1876.21 L1980.41 1885.65 L1972.96 1885.65 L1972.96 1876.21 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2031.33 1902.58 Q2034.13 1897.56 2038.01 1895.17 Q2041.9 1892.78 2047.17 1892.78 Q2054.26 1892.78 2058.11 1897.76 Q2061.96 1902.7 2061.96 1911.86 L2061.96 1939.24 L2054.46 1939.24 L2054.46 1912.1 Q2054.46 1905.58 2052.15 1902.42 Q2049.84 1899.26 2045.1 1899.26 Q2039.31 1899.26 2035.95 1903.11 Q2032.59 1906.96 2032.59 1913.6 L2032.59 1939.24 L2025.09 1939.24 L2025.09 1912.1 Q2025.09 1905.54 2022.78 1902.42 Q2020.47 1899.26 2015.65 1899.26 Q2009.94 1899.26 2006.58 1903.15 Q2003.22 1907 2003.22 1913.6 L2003.22 1939.24 L1995.72 1939.24 L1995.72 1893.87 L2003.22 1893.87 L2003.22 1900.92 Q2005.77 1896.75 2009.33 1894.76 Q2012.9 1892.78 2017.8 1892.78 Q2022.74 1892.78 2026.19 1895.29 Q2029.67 1897.8 2031.33 1902.58 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2076.05 1921.34 L2076.05 1893.87 L2083.51 1893.87 L2083.51 1921.05 Q2083.51 1927.5 2086.02 1930.74 Q2088.53 1933.94 2093.55 1933.94 Q2099.59 1933.94 2103.07 1930.09 Q2106.6 1926.24 2106.6 1919.6 L2106.6 1893.87 L2114.05 1893.87 L2114.05 1939.24 L2106.6 1939.24 L2106.6 1932.28 Q2103.88 1936.41 2100.28 1938.43 Q2096.71 1940.42 2091.97 1940.42 Q2084.15 1940.42 2080.1 1935.56 Q2076.05 1930.7 2076.05 1921.34 M2094.81 1892.78 L2094.81 1892.78 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2136.78 1880.99 L2136.78 1893.87 L2152.13 1893.87 L2152.13 1899.67 L2136.78 1899.67 L2136.78 1924.29 Q2136.78 1929.84 2138.27 1931.42 Q2139.81 1933 2144.47 1933 L2152.13 1933 L2152.13 1939.24 L2144.47 1939.24 Q2135.84 1939.24 2132.56 1936.04 Q2129.28 1932.8 2129.28 1924.29 L2129.28 1899.67 L2123.81 1899.67 L2123.81 1893.87 L2129.28 1893.87 L2129.28 1880.99 L2136.78 1880.99 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2199.65 1911.86 L2199.65 1939.24 L2192.19 1939.24 L2192.19 1912.1 Q2192.19 1905.66 2189.68 1902.46 Q2187.17 1899.26 2182.15 1899.26 Q2176.11 1899.26 2172.63 1903.11 Q2169.14 1906.96 2169.14 1913.6 L2169.14 1939.24 L2161.65 1939.24 L2161.65 1876.21 L2169.14 1876.21 L2169.14 1900.92 Q2171.82 1896.83 2175.42 1894.8 Q2179.07 1892.78 2183.81 1892.78 Q2191.62 1892.78 2195.64 1897.64 Q2199.65 1902.46 2199.65 1911.86 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2258.79 1876.29 Q2253.36 1885.61 2250.73 1894.72 Q2248.09 1903.84 2248.09 1913.2 Q2248.09 1922.55 2250.73 1931.75 Q2253.4 1940.9 2258.79 1950.18 L2252.31 1950.18 Q2246.23 1940.66 2243.19 1931.46 Q2240.2 1922.27 2240.2 1913.2 Q2240.2 1904.16 2243.19 1895.01 Q2246.19 1885.85 2252.31 1876.29 L2258.79 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2303.11 1900.76 L2303.11 1876.21 L2310.56 1876.21 L2310.56 1939.24 L2303.11 1939.24 L2303.11 1932.44 Q2300.76 1936.49 2297.15 1938.47 Q2293.59 1940.42 2288.56 1940.42 Q2280.34 1940.42 2275.15 1933.85 Q2270.01 1927.29 2270.01 1916.6 Q2270.01 1905.9 2275.15 1899.34 Q2280.34 1892.78 2288.56 1892.78 Q2293.59 1892.78 2297.15 1894.76 Q2300.76 1896.71 2303.11 1900.76 M2277.71 1916.6 Q2277.71 1924.82 2281.07 1929.52 Q2284.47 1934.18 2290.39 1934.18 Q2296.3 1934.18 2299.7 1929.52 Q2303.11 1924.82 2303.11 1916.6 Q2303.11 1908.37 2299.7 1903.72 Q2296.3 1899.02 2290.39 1899.02 Q2284.47 1899.02 2281.07 1903.72 Q2277.71 1908.37 2277.71 1916.6 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2364.72 1914.69 L2364.72 1918.34 L2330.45 1918.34 Q2330.94 1926.04 2335.07 1930.09 Q2339.24 1934.1 2346.65 1934.1 Q2350.95 1934.1 2354.96 1933.04 Q2359.01 1931.99 2362.98 1929.89 L2362.98 1936.93 Q2358.97 1938.64 2354.75 1939.53 Q2350.54 1940.42 2346.21 1940.42 Q2335.35 1940.42 2328.99 1934.1 Q2322.67 1927.78 2322.67 1917 Q2322.67 1905.86 2328.67 1899.34 Q2334.7 1892.78 2344.91 1892.78 Q2354.07 1892.78 2359.37 1898.69 Q2364.72 1904.57 2364.72 1914.69 M2357.27 1912.51 Q2357.19 1906.39 2353.82 1902.74 Q2350.5 1899.1 2344.99 1899.1 Q2338.75 1899.1 2334.99 1902.62 Q2331.26 1906.15 2330.69 1912.55 L2357.27 1912.51 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2406.81 1916.03 Q2406.81 1907.93 2403.45 1903.47 Q2400.12 1899.02 2394.09 1899.02 Q2388.09 1899.02 2384.73 1903.47 Q2381.41 1907.93 2381.41 1916.03 Q2381.41 1924.09 2384.73 1928.55 Q2388.09 1933 2394.09 1933 Q2400.12 1933 2403.45 1928.55 Q2406.81 1924.09 2406.81 1916.03 M2414.26 1933.61 Q2414.26 1945.2 2409.12 1950.83 Q2403.97 1956.5 2393.36 1956.5 Q2389.43 1956.5 2385.95 1955.89 Q2382.46 1955.32 2379.18 1954.11 L2379.18 1946.86 Q2382.46 1948.64 2385.66 1949.49 Q2388.86 1950.34 2392.19 1950.34 Q2399.52 1950.34 2403.16 1946.49 Q2406.81 1942.69 2406.81 1934.95 L2406.81 1931.26 Q2404.5 1935.27 2400.89 1937.26 Q2397.29 1939.24 2392.27 1939.24 Q2383.92 1939.24 2378.82 1932.88 Q2373.71 1926.52 2373.71 1916.03 Q2373.71 1905.5 2378.82 1899.14 Q2383.92 1892.78 2392.27 1892.78 Q2397.29 1892.78 2400.89 1894.76 Q2404.5 1896.75 2406.81 1900.76 L2406.81 1893.87 L2414.26 1893.87 L2414.26 1933.61 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><path clip-path="url(#clip030)" d="M2428.44 1876.29 L2434.92 1876.29 Q2441 1885.85 2444 1895.01 Q2447.03 1904.16 2447.03 1913.2 Q2447.03 1922.27 2444 1931.46 Q2441 1940.66 2434.92 1950.18 L2428.44 1950.18 Q2433.83 1940.9 2436.46 1931.75 Q2439.14 1922.55 2439.14 1913.2 Q2439.14 1903.84 2436.46 1894.72 Q2433.83 1885.61 2428.44 1876.29 Z" fill="#000000" fill-rule="nonzero" fill-opacity="1" /><polyline clip-path="url(#clip037)" style="stroke:#009af9; stroke-linecap:round; stroke-linejoin:round; stroke-width:8; stroke-opacity:1; fill:none" points="1593.53,2038.4 1595.77,2038.52 1598.01,2038.64 1600.25,2038.77 1602.5,2038.92 1604.74,2039.07 1606.98,2039.24 1609.22,2039.42 1611.46,2039.61 1613.71,2039.82 1615.95,2040.04 1618.19,2040.28 1620.43,2040.54 1622.67,2040.82 1624.92,2041.12 1627.16,2041.44 1629.4,2041.78 1631.64,2042.15 1633.88,2042.55 1636.13,2042.97 1638.37,2043.43 1640.61,2043.92 1642.85,2044.43 1645.09,2044.99 1647.34,2045.58 1649.58,2046.21 1651.82,2046.87 1654.06,2047.58 1656.3,2048.33 1658.55,2049.13 1660.79,2049.96 1663.03,2050.84 1665.27,2051.77 1667.51,2052.74 1669.76,2053.76 1672,2054.82 1674.24,2055.93 1676.48,2057.08 1678.72,2058.27 1680.97,2059.49 1683.21,2060.76 1685.45,2062.06 1687.69,2063.39 1689.93,2064.75 1692.17,2066.13 1694.42,2067.53 1696.66,2068.95 1698.9,2070.38 1701.14,2071.82 1703.38,2073.26 1705.63,2074.7 1707.87,2076.13 1710.11,2077.56 1712.35,2078.97 1714.59,2080.37 1716.84,2081.75 1719.08,2083.11 1721.32,2084.44 1723.56,2085.75 1725.8,2087.03 1728.05,2088.27 1730.29,2089.49 1732.53,2090.67 1734.77,2091.82 1737.01,2092.94 1739.26,2094.03 1741.5,2095.08 1743.74,2096.09 1745.98,2097.08 1748.22,2098.03 1750.47,2098.95 1752.71,2099.85 1754.95,2100.71 1757.19,2101.54 1759.43,2102.35 1761.68,2103.13 1763.92,2103.89 1766.16,2104.62 1768.4,2105.33 1770.64,2106.02 1772.89,2106.69 1775.13,2107.34 1777.37,2107.97 1779.61,2108.59 1781.85,2109.19 1784.1,2109.77 1786.34,2110.35 1788.58,2110.91 1790.82,2111.46 1793.06,2112 1795.31,2112.53 1797.55,2113.06 1799.79,2113.57 1802.03,2114.09 1804.27,2114.59 1806.51,2115.1 1808.76,2115.6 1811,2116.1 1813.24,2116.6 1815.48,2117.09 1817.72,2117.59 1819.97,2118.1 1822.21,2118.6 1824.45,2119.11 1826.69,2119.63 1828.93,2120.15 1831.18,2120.68 1833.42,2121.22 1835.66,2121.77 1837.9,2122.33 1840.14,2122.9 1842.39,2123.49 1844.63,2124.09 1846.87,2124.71 1849.11,2125.35 1851.35,2126 1853.6,2126.67 1855.84,2127.37 1858.08,2128.09 1860.32,2128.83 1862.56,2129.6 1864.81,2130.4 1867.05,2131.22 1869.29,2132.08 1871.53,2132.96 1873.77,2133.88 1876.02,2134.82 1878.26,2135.8 1880.5,2136.82 1882.74,2137.86 1884.98,2138.94 1887.23,2140.05 1889.47,2141.2 1891.71,2142.38 1893.95,2143.59 1896.19,2144.82 1898.44,2146.09 1900.68,2147.38 1902.92,2148.69 1905.16,2150.02 1907.4,2151.37 1909.64,2152.74 1911.89,2154.12 1914.13,2155.5 1916.37,2156.89 1918.61,2158.28 1920.85,2159.67 1923.1,2161.06 1925.34,2162.43 1927.58,2163.8 1929.82,2165.15 1932.06,2166.49 1934.31,2167.8 1936.55,2169.1 1938.79,2170.37 1941.03,2171.63 1943.27,2172.85 1945.52,2174.06 1947.76,2175.23 1950,2176.38 1952.24,2177.51 1954.48,2178.61 1956.73,2179.69 1958.97,2180.74 1961.21,2181.77 1963.45,2182.78 1965.69,2183.77 1967.94,2184.73 1970.18,2185.68 1972.42,2186.61 1974.66,2187.53 1976.9,2188.43 1979.15,2189.32 1981.39,2190.2 1983.63,2191.07 1985.87,2191.93 1988.11,2192.78 1990.36,2193.63 1992.6,2194.48 1994.84,2195.33 1997.08,2196.18 1999.32,2197.03 2001.57,2197.88 2003.81,2198.74 2006.05,2199.61 2008.29,2200.48 2010.53,2201.37 2012.78,2202.27 2015.02,2203.19 2017.26,2204.12 2019.5,2205.06 2021.74,2206.03 2023.98,2207.02 2026.23,2208.02 2028.47,2209.05 2030.71,2210.11 2032.95,2211.18 2035.19,2212.29 2037.44,2213.41 2039.68,2214.57 2041.92,2215.75 2044.16,2216.95 2046.4,2218.18 2048.65,2219.43 2050.89,2220.71 2053.13,2222.01 2055.37,2223.33 2057.61,2224.66 2059.86,2226.02 2062.1,2227.39 2064.34,2228.77 2066.58,2230.15 2068.82,2231.55 2071.07,2232.95 2073.31,2234.35 2075.55,2235.75 2077.79,2237.14 2080.03,2238.53 2082.28,2239.91 2084.52,2241.28 2086.76,2242.63 2089,2243.97 2091.24,2245.3 2093.49,2246.61 2095.73,2247.9 2097.97,2249.17 2100.21,2250.43 2102.45,2251.67 2104.7,2252.88 2106.94,2254.09 2109.18,2255.27 2111.42,2256.44 2113.66,2257.59 2115.91,2258.72 2118.15,2259.85 2120.39,2260.96 2122.63,2262.06 2124.87,2263.15 2127.11,2264.24 2129.36,2265.32 2131.6,2266.39 2133.84,2267.46 2136.08,2268.54 2138.32,2269.61 2140.57,2270.69 2142.81,2271.77 2145.05,2272.86 2147.29,2273.95 2149.53,2275.06 2151.78,2276.17 2154.02,2277.3 2156.26,2278.44 2158.5,2279.6 2160.74,2280.78 2162.99,2281.97 2165.23,2283.18 2167.47,2284.41 2169.71,2285.65 2171.95,2286.92 2174.2,2288.21 2176.44,2289.51 2178.68,2290.83 2180.92,2292.18 2183.16,2293.53 2185.41,2294.91 2187.65,2296.3 2189.89,2297.7 2192.13,2299.11 2194.37,2300.53 2196.62,2301.96 2198.86,2303.4 2201.1,2304.84 2203.34,2306.28 2205.58,2307.72 2207.83,2309.16 2210.07,2310.59 2212.31,2312.02 2214.55,2313.44 2216.79,2314.85 2219.04,2316.25 2221.28,2317.64 2223.52,2319.02 2225.76,2320.39 2228,2321.75 2230.25,2323.09 2232.49,2324.43 2234.73,2325.75 2236.97,2327.06 2239.21,2328.36 2241.45,2329.65 2243.7,2330.93 2245.94,2332.21 2248.18,2333.47 2250.42,2334.74 2252.66,2336 2254.91,2337.26 2257.15,2338.51 2259.39,2339.77 2261.63,2341.03 2263.87,2342.3 2266.12,2343.57 2268.36,2344.84 2270.6,2346.13 2272.84,2347.42 2275.08,2348.72 2277.33,2350.03 2279.57,2351.36 2281.81,2352.7 2284.05,2354.05 2286.29,2355.41 2288.54,2356.79 2290.78,2358.18 2293.02,2359.59 2295.26,2361.01 2297.5,2362.44 2299.75,2363.89 2301.99,2365.35 2304.23,2366.82 2306.47,2368.3 2308.71,2369.78 2310.96,2371.28 2313.2,2372.78 2315.44,2374.29 2317.68,2375.79 2319.92,2377.3 2322.17,2378.82 2324.41,2380.33 2326.65,2381.83 2328.89,2383.34 2331.13,2384.84 2333.38,2386.34 2335.62,2387.83 2337.86,2389.31 2340.1,2390.79 2342.34,2392.26 2344.58,2393.72 2346.83,2395.18 2349.07,2396.63 2351.31,2398.08 2353.55,2399.52 2355.79,2400.95 2358.04,2402.38 2360.28,2403.81 2362.52,2405.24 2364.76,2406.66 2367,2408.08 2369.25,2409.51 2371.49,2410.93 2373.73,2412.36 2375.97,2413.8 2378.21,2415.23 2380.46,2416.67 2382.7,2418.12 2384.94,2419.58 2387.18,2421.04 2389.42,2422.51 2391.67,2424 2393.91,2425.49 2396.15,2426.99 2398.39,2428.5 2400.63,2430.02 2402.88,2431.54 2405.12,2433.08 2407.36,2434.63 2409.6,2436.18 2411.84,2437.75 2414.09,2439.32 2416.33,2440.89 2418.57,2442.48 2420.81,2444.06 2423.05,2445.65 2425.3,2447.24 2427.54,2448.84 2429.78,2450.43 2432.02,2452.03 2434.26,2453.62 2436.51,2455.22 2438.75,2456.81 2440.99,2458.4 2443.23,2459.99 2445.47,2461.57 2447.72,2463.15 2449.96,2464.73 2452.2,2466.3 2454.44,2467.87 2456.68,2469.44 2458.92,2471.01 2461.17,2472.57 2463.41,2474.13 2465.65,2475.69 2467.89,2477.25 2470.13,2478.81 2472.38,2480.37 2474.62,2481.93 2476.86,2483.5 2479.1,2485.06 2481.34,2486.63 2483.59,2488.2 2485.83,2489.78 2488.07,2491.36 2490.31,2492.95 2492.55,2494.54 2494.8,2496.14 2497.04,2497.74 2499.28,2499.35 2501.52,2500.96 2503.76,2502.58 2506.01,2504.21 2508.25,2505.84 2510.49,2507.47 2512.73,2509.11 2514.97,2510.76 2517.22,2512.4 2519.46,2514.05 2521.7,2515.71 2523.94,2517.36 2526.18,2519.02 2528.43,2520.67 2530.67,2522.33 2532.91,2523.99 2535.15,2525.64 2537.39,2527.3 2539.64,2528.95 2541.88,2530.6 2544.12,2532.25 2546.36,2533.89 2548.6,2535.53 2550.85,2537.17 2553.09,2538.81 2555.33,2540.44 2557.57,2542.07 2559.81,2543.7 2562.05,2545.33 2564.3,2546.95 2566.54,2548.57 2568.78,2550.2 2571.02,2551.81 2573.26,2553.43 2575.51,2555.05 2577.75,2556.67 2579.99,2558.29 2582.23,2559.91 2584.47,2561.52 2586.72,2563.14 2588.96,2564.76 2591.2,2566.39 2593.44,2568.01 2595.68,2569.63 2597.93,2571.26 2600.17,2572.88 2602.41,2574.51 2604.65,2576.13 2606.89,2577.76 2609.14,2579.38 2611.38,2581.01 2613.62,2582.63 2615.86,2584.25 2618.1,2585.86 2620.35,2587.47 2622.59,2589.08 2624.83,2590.68 2627.07,2592.27 2629.31,2593.86 2631.56,2595.44 2633.8,2597 2636.04,2598.56 2638.28,2600.11 2640.52,2601.65 2642.77,2603.17 2645.01,2604.68 2647.25,2606.18 2649.49,2607.67 2651.73,2609.14 2653.98,2610.59 2656.22,2612.03 2658.46,2613.45 2660.7,2614.86 2662.94,2616.25 2665.19,2617.62 2667.43,2618.97 2669.67,2620.3 2671.91,2621.61 2674.15,2622.9 2676.39,2624.16 2678.64,2625.4 2680.88,2626.61 2683.12,2627.79 2685.36,2628.95 2687.6,2630.06 2689.85,2631.15 2692.09,2632.18 2694.33,2633.18 2696.57,2634.12 2698.81,2635.01 2701.06,2635.83 2703.3,2636.59 2705.54,2637.27 2707.78,2637.86 2710.02,2638.36 2712.27,2638.74 2714.51,2639.01 2716.75,2639.15 2718.99,2639.13 "/>
 </svg>
diff --git a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/index.html b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/index.html
index 93ce0689afb..7ddcfaca9d1 100644
--- a/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/index.html
+++ b/previews/PR3562/tutorials/nonlinear/space_shuttle_reentry_trajectory/index.html
@@ -223,7 +223,7 @@
     layout = grid(3, 2),
     linewidth = 2,
     size = (700, 700),
-)</code></pre><img src="d9b9da98.svg" alt="Example block output"/><pre><code class="language-julia hljs">function q(h, v, a)
+)</code></pre><img src="e697ac28.svg" alt="Example block output"/><pre><code class="language-julia hljs">function q(h, v, a)
     ρ(h) = ρ₀ * exp(-h / hᵣ)
     qᵣ(h, v) = 17700 * √ρ(h) * (0.0001 * v)^3.07
     qₐ(a) = c₀ + c₁ * rad2deg(a) + c₂ * rad2deg(a)^2 + c₃ * rad2deg(a)^3
@@ -257,7 +257,7 @@
     layout = grid(3, 1),
     linewidth = 2,
     size = (700, 700),
-)</code></pre><img src="97a644df.svg" alt="Example block output"/><pre><code class="language-julia hljs">plot(
+)</code></pre><img src="b724747b.svg" alt="Example block output"/><pre><code class="language-julia hljs">plot(
     rad2deg.(value.(ϕ)),
     rad2deg.(value.(θ)),
     value.(scaled_h);
@@ -267,4 +267,4 @@
     xlabel = &quot;Longitude (deg)&quot;,
     ylabel = &quot;Latitude (deg)&quot;,
     zlabel = &quot;Altitude (100,000 ft)&quot;,
-)</code></pre><img src="07f114d3.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../rocket_control/">« Rocket Control</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre><img src="b9835bf8.svg" alt="Example block output"/></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../rocket_control/">« Rocket Control</a><a class="docs-footer-nextpage" href="../tips_and_tricks/">Tips and tricks »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/tips_and_tricks/index.html b/previews/PR3562/tutorials/nonlinear/tips_and_tricks/index.html
index 611756e98e8..0c05038e0ea 100644
--- a/previews/PR3562/tutorials/nonlinear/tips_and_tricks/index.html
+++ b/previews/PR3562/tutorials/nonlinear/tips_and_tricks/index.html
@@ -67,4 +67,4 @@
 optimize!(model)
 Test.@test objective_value(model) ≈ √3 atol = 1e-4
 Test.@test value.(x) ≈ [1.0, 1.0] atol = 1e-4
-println(&quot;Memoized approach: function_calls = $(function_calls)&quot;)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Memoized approach: function_calls = 22</code></pre><p>Compared to the naive approach, the memoized approach requires half as many function evaluations.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../space_shuttle_reentry_trajectory/">« Optimal control for a Space Shuttle reentry trajectory</a><a class="docs-footer-nextpage" href="../user_defined_hessians/">User-defined Hessians »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+println(&quot;Memoized approach: function_calls = $(function_calls)&quot;)</code></pre><pre class="documenter-example-output"><code class="nohighlight hljs ansi">Memoized approach: function_calls = 22</code></pre><p>Compared to the naive approach, the memoized approach requires half as many function evaluations.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../space_shuttle_reentry_trajectory/">« Optimal control for a Space Shuttle reentry trajectory</a><a class="docs-footer-nextpage" href="../user_defined_hessians/">User-defined Hessians »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
diff --git a/previews/PR3562/tutorials/nonlinear/user_defined_hessians/index.html b/previews/PR3562/tutorials/nonlinear/user_defined_hessians/index.html
index 2b8846daf22..5f5a75e3a2d 100644
--- a/previews/PR3562/tutorials/nonlinear/user_defined_hessians/index.html
+++ b/previews/PR3562/tutorials/nonlinear/user_defined_hessians/index.html
@@ -30,7 +30,7 @@
 * Candidate solution (result #1)
   Primal status      : FEASIBLE_POINT
   Dual status        : FEASIBLE_POINT
-  Objective value    : 1.21190e-27
+  Objective value    : 8.14943e-28
   Dual objective value : 0.00000e+00
   Primal solution :
     x[1] : 1.00000e+00
@@ -38,6 +38,6 @@
   Dual solution :
 
 * Work counters
-  Solve time (sec)   : 4.11732e-02
+  Solve time (sec)   : 6.11861e-02
   Barrier iterations : 14
-</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and tricks</a><a class="docs-footer-nextpage" href="../nested_problems/">Nested optimization problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:11">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../tips_and_tricks/">« Tips and tricks</a><a class="docs-footer-nextpage" href="../nested_problems/">Nested optimization problems »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.1.1 on <span class="colophon-date" title="Monday 6 November 2023 21:16">Monday 6 November 2023</span>. Using Julia version 1.9.3.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>